--- /dev/null
+diff -urNp linux-2.6/arch/i386/boot-xen/Makefile new/arch/i386/boot-xen/Makefile
+--- linux-2.6/arch/i386/boot-xen/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/boot-xen/Makefile 2006-05-09 12:32:33.000000000 +0200
+@@ -0,0 +1,21 @@
++
++OBJCOPYFLAGS := -g --strip-unneeded
++
++vmlinuz: vmlinux-stripped FORCE
++ $(call if_changed,gzip)
++
++vmlinux-stripped: vmlinux FORCE
++ $(call if_changed,objcopy)
++
++INSTALL_ROOT := $(patsubst %/boot,%,$(INSTALL_PATH))
++
++XINSTALL_NAME ?= $(KERNELRELEASE)
++install:
++ mkdir -p $(INSTALL_ROOT)/boot
++ ln -f -s vmlinuz-$(XINSTALL_NAME)$(INSTALL_SUFFIX) $(INSTALL_ROOT)/boot/vmlinuz-$(VERSION).$(PATCHLEVEL).$(SUBLEVEL)$(XENGUEST)$(INSTALL_SUFFIX)
++ rm -f $(INSTALL_ROOT)/boot/vmlinuz-$(XINSTALL_NAME)$(INSTALL_SUFFIX)
++ install -m0644 vmlinuz $(INSTALL_ROOT)/boot/vmlinuz-$(XINSTALL_NAME)$(INSTALL_SUFFIX)
++ install -m0644 vmlinux $(INSTALL_ROOT)/boot/vmlinux-syms-$(XINSTALL_NAME)$(INSTALL_SUFFIX)
++ install -m0664 .config $(INSTALL_ROOT)/boot/config-$(XINSTALL_NAME)$(INSTALL_SUFFIX)
++ install -m0664 System.map $(INSTALL_ROOT)/boot/System.map-$(XINSTALL_NAME)$(INSTALL_SUFFIX)
++ ln -f -s vmlinuz-$(XINSTALL_NAME)$(INSTALL_SUFFIX) $(INSTALL_ROOT)/boot/vmlinuz-$(VERSION).$(PATCHLEVEL)$(XENGUEST)$(INSTALL_SUFFIX)
+diff -urNp linux-2.6/arch/i386/Kconfig new/arch/i386/Kconfig
+--- linux-2.6/arch/i386/Kconfig 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/Kconfig 2006-05-23 18:37:09.000000000 +0200
+@@ -91,6 +91,15 @@ config X86_PC
+ help
+ Choose this option if your computer is a standard PC or compatible.
+
++config X86_XEN
++ bool "Xen-compatible"
++ select X86_UP_APIC if !SMP && XEN_PRIVILEGED_GUEST
++ select X86_UP_IOAPIC if !SMP && XEN_PRIVILEGED_GUEST
++ select SWIOTLB
++ help
++ Choose this option if you plan to run this kernel on top of the
++ Xen Hypervisor.
++
+ config X86_ELAN
+ bool "AMD Elan"
+ help
+@@ -193,6 +202,7 @@ source "arch/i386/Kconfig.cpu"
+
+ config HPET_TIMER
+ bool "HPET Timer Support"
++ depends on !X86_XEN
+ help
+ This enables the use of the HPET for the kernel's internal timer.
+ HPET is the next generation timer replacing legacy 8254s.
+@@ -223,7 +233,7 @@ config NR_CPUS
+
+ config SCHED_SMT
+ bool "SMT (Hyperthreading) scheduler support"
+- depends on SMP
++ depends on SMP && !X86_XEN
+ default off
+ help
+ SMT scheduler support improves the CPU scheduler's decision making
+@@ -233,7 +243,7 @@ config SCHED_SMT
+
+ config SCHED_MC
+ bool "Multi-core scheduler support"
+- depends on SMP
++ depends on SMP && !X86_XEN
+ default y
+ help
+ Multi-core scheduler support improves the CPU scheduler's decision
+@@ -244,7 +254,7 @@ source "kernel/Kconfig.preempt"
+
+ config X86_UP_APIC
+ bool "Local APIC support on uniprocessors"
+- depends on !SMP && !(X86_VISWS || X86_VOYAGER)
++ depends on !SMP && !(X86_VISWS || X86_VOYAGER || XEN_UNPRIVILEGED_GUEST)
+ help
+ A local APIC (Advanced Programmable Interrupt Controller) is an
+ integrated interrupt controller in the CPU. If you have a single-CPU
+@@ -269,12 +279,12 @@ config X86_UP_IOAPIC
+
+ config X86_LOCAL_APIC
+ bool
+- depends on X86_UP_APIC || ((X86_VISWS || SMP) && !X86_VOYAGER)
++ depends on X86_UP_APIC || ((X86_VISWS || SMP) && !(X86_VOYAGER || XEN_UNPRIVILEGED_GUEST))
+ default y
+
+ config X86_IO_APIC
+ bool
+- depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER))
++ depends on X86_UP_IOAPIC || (SMP && !(X86_VISWS || X86_VOYAGER || XEN_UNPRIVILEGED_GUEST))
+ default y
+
+ config X86_VISWS_APIC
+@@ -282,9 +292,14 @@ config X86_VISWS_APIC
+ depends on X86_VISWS
+ default y
+
++config X86_TSC
++ bool
++ depends on (MWINCHIP3D || MWINCHIP2 || MCRUSOE || MEFFICEON || MCYRIXIII || MK7 || MK6 || MPENTIUM4 || MPENTIUMM || MPENTIUMIII || MPENTIUMII || M686 || M586MMX || M586TSC || MK8 || MVIAC3_2 || MGEODEGX1) && !X86_NUMAQ && !X86_XEN
++ default y
++
+ config X86_MCE
+ bool "Machine Check Exception"
+- depends on !X86_VOYAGER
++ depends on !(X86_VOYAGER || X86_XEN)
+ ---help---
+ Machine Check Exception support allows the processor to notify the
+ kernel if it detects a problem (e.g. overheating, component failure).
+@@ -374,6 +389,7 @@ config X86_REBOOTFIXUPS
+
+ config MICROCODE
+ tristate "/dev/cpu/microcode - Intel IA32 CPU microcode support"
++ depends on !XEN_UNPRIVILEGED_GUEST
+ ---help---
+ If you say Y here and also to "/dev file system support" in the
+ 'File systems' section, you will be able to update the microcode on
+@@ -391,6 +407,7 @@ config MICROCODE
+
+ config X86_MSR
+ tristate "/dev/cpu/*/msr - Model-specific register support"
++ depends on !X86_XEN
+ help
+ This device gives privileged processes access to the x86
+ Model-Specific Registers (MSRs). It is a character device with
+@@ -406,6 +423,10 @@ config X86_CPUID
+ with major 203 and minors 0 to 31 for /dev/cpu/0/cpuid to
+ /dev/cpu/31/cpuid.
+
++config SWIOTLB
++ bool
++ default n
++
+ source "drivers/firmware/Kconfig"
+
+ choice
+@@ -578,7 +599,7 @@ config HAVE_ARCH_EARLY_PFN_TO_NID
+
+ config HIGHPTE
+ bool "Allocate 3rd-level pagetables from highmem"
+- depends on HIGHMEM4G || HIGHMEM64G
++ depends on (HIGHMEM4G || HIGHMEM64G) && !X86_XEN
+ help
+ The VM uses one page table entry for each page of physical memory.
+ For systems with a lot of RAM, this can be wasteful of precious
+@@ -587,6 +608,7 @@ config HIGHPTE
+
+ config MATH_EMULATION
+ bool "Math emulation"
++ depends on !X86_XEN
+ ---help---
+ Linux can emulate a math coprocessor (used for floating point
+ operations) if you don't have one. 486DX and Pentium processors have
+@@ -612,6 +634,8 @@ config MATH_EMULATION
+
+ config MTRR
+ bool "MTRR (Memory Type Range Register) support"
++ depends on !XEN_UNPRIVILEGED_GUEST
++ default y if X86_XEN
+ ---help---
+ On Intel P6 family processors (Pentium Pro, Pentium II and later)
+ the Memory Type Range Registers (MTRRs) may be used to control
+@@ -646,7 +670,7 @@ config MTRR
+
+ config EFI
+ bool "Boot from EFI support (EXPERIMENTAL)"
+- depends on ACPI
++ depends on ACPI && !X86_XEN
+ default n
+ ---help---
+ This enables the the kernel to boot on EFI platforms using
+@@ -664,7 +688,7 @@ config EFI
+
+ config IRQBALANCE
+ bool "Enable kernel irq balancing"
+- depends on SMP && X86_IO_APIC
++ depends on SMP && X86_IO_APIC && !X86_XEN
+ default y
+ help
+ The default yes will allow the kernel to do irq load balancing.
+@@ -712,7 +736,7 @@ source kernel/Kconfig.hz
+
+ config KEXEC
+ bool "kexec system call (EXPERIMENTAL)"
+- depends on EXPERIMENTAL
++ depends on EXPERIMENTAL && !X86_XEN
+ help
+ kexec is a system call that implements the ability to shutdown your
+ current kernel, and to start another kernel. It is like a reboot
+@@ -767,18 +791,20 @@ endmenu
+
+
+ menu "Power management options (ACPI, APM)"
+- depends on !X86_VOYAGER
++ depends on !(X86_VOYAGER || XEN_UNPRIVILEGED_GUEST)
+
++if !X86_XEN
+ source kernel/power/Kconfig
++endif
+
+ source "drivers/acpi/Kconfig"
+
+ menu "APM (Advanced Power Management) BIOS Support"
+-depends on PM && !X86_VISWS
++depends on PM && !(X86_VISWS || X86_XEN)
+
+ config APM
+ tristate "APM (Advanced Power Management) BIOS support"
+- depends on PM
++ depends on PM && PM_LEGACY
+ ---help---
+ APM is a BIOS specification for saving power using several different
+ techniques. This is mostly useful for battery powered laptops with
+@@ -963,6 +989,7 @@ choice
+
+ config PCI_GOBIOS
+ bool "BIOS"
++ depends on !X86_XEN
+
+ config PCI_GOMMCONFIG
+ bool "MMConfig"
+@@ -970,6 +997,13 @@ config PCI_GOMMCONFIG
+ config PCI_GODIRECT
+ bool "Direct"
+
++config PCI_GOXEN_FE
++ bool "Xen PCI Frontend"
++ depends on X86_XEN
++ help
++ The PCI device frontend driver allows the kernel to import arbitrary
++ PCI devices from a PCI backend to support PCI driver domains.
++
+ config PCI_GOANY
+ bool "Any"
+
+@@ -977,7 +1011,7 @@ endchoice
+
+ config PCI_BIOS
+ bool
+- depends on !X86_VISWS && PCI && (PCI_GOBIOS || PCI_GOANY)
++ depends on !(X86_VISWS || X86_XEN) && PCI && (PCI_GOBIOS || PCI_GOANY)
+ default y
+
+ config PCI_DIRECT
+@@ -990,6 +1024,18 @@ config PCI_MMCONFIG
+ depends on PCI && ACPI && (PCI_GOMMCONFIG || PCI_GOANY)
+ default y
+
++config XEN_PCIDEV_FRONTEND
++ bool
++ depends on PCI && X86_XEN && (PCI_GOXEN_FE || PCI_GOANY)
++ default y
++
++config XEN_PCIDEV_FE_DEBUG
++ bool "Xen PCI Frontend Debugging"
++ depends on XEN_PCIDEV_FRONTEND
++ default n
++ help
++ Enables some debug statements within the PCI Frontend.
++
+ source "drivers/pci/pcie/Kconfig"
+
+ source "drivers/pci/Kconfig"
+@@ -1000,7 +1046,7 @@ config ISA_DMA_API
+
+ config ISA
+ bool "ISA support"
+- depends on !(X86_VOYAGER || X86_VISWS)
++ depends on !(X86_VOYAGER || X86_VISWS || X86_XEN)
+ help
+ Find out whether you have ISA slots on your motherboard. ISA is the
+ name of a bus system, i.e. the way the CPU talks to the other stuff
+@@ -1027,7 +1073,7 @@ config EISA
+ source "drivers/eisa/Kconfig"
+
+ config MCA
+- bool "MCA support" if !(X86_VISWS || X86_VOYAGER)
++ bool "MCA support" if !(X86_VISWS || X86_VOYAGER || X86_XEN)
+ default y if X86_VOYAGER
+ help
+ MicroChannel Architecture is found in some IBM PS/2 machines and
+@@ -1089,6 +1135,8 @@ source "security/Kconfig"
+
+ source "crypto/Kconfig"
+
++source "drivers/xen/Kconfig"
++
+ source "lib/Kconfig"
+
+ #
+@@ -1114,7 +1162,7 @@ config X86_SMP
+
+ config X86_HT
+ bool
+- depends on SMP && !(X86_VISWS || X86_VOYAGER)
++ depends on SMP && !(X86_VISWS || X86_VOYAGER || X86_XEN)
+ default y
+
+ config X86_BIOS_REBOOT
+@@ -1127,6 +1175,16 @@ config X86_TRAMPOLINE
+ depends on X86_SMP || (X86_VOYAGER && SMP)
+ default y
+
++config X86_NO_TSS
++ bool
++ depends on X86_XEN
++ default y
++
++config X86_NO_IDT
++ bool
++ depends on X86_XEN
++ default y
++
+ config KTIME_SCALAR
+ bool
+ default y
+diff -urNp linux-2.6/arch/i386/Kconfig.cpu new/arch/i386/Kconfig.cpu
+--- linux-2.6/arch/i386/Kconfig.cpu 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/Kconfig.cpu 2006-05-09 12:32:33.000000000 +0200
+@@ -251,7 +251,7 @@ config X86_PPRO_FENCE
+
+ config X86_F00F_BUG
+ bool
+- depends on M586MMX || M586TSC || M586 || M486 || M386
++ depends on (M586MMX || M586TSC || M586 || M486 || M386) && !X86_NO_IDT
+ default y
+
+ config X86_WP_WORKS_OK
+diff -urNp linux-2.6/arch/i386/Kconfig.debug new/arch/i386/Kconfig.debug
+--- linux-2.6/arch/i386/Kconfig.debug 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/Kconfig.debug 2006-05-09 12:32:33.000000000 +0200
+@@ -84,6 +84,7 @@ config X86_MPPARSE
+ config DOUBLEFAULT
+ default y
+ bool "Enable doublefault exception handler" if EMBEDDED
++ depends on !X86_NO_TSS
+ help
+ This option allows trapping of rare doublefault exceptions that
+ would otherwise cause a system to silently reboot. Disabling this
+diff -urNp linux-2.6/arch/i386/kernel/acpi/boot-xen.c new/arch/i386/kernel/acpi/boot-xen.c
+--- linux-2.6/arch/i386/kernel/acpi/boot-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/acpi/boot-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,1167 @@
++/*
++ * boot.c - Architecture-Specific Low-Level ACPI Boot Support
++ *
++ * Copyright (C) 2001, 2002 Paul Diefenbaugh <paul.s.diefenbaugh@intel.com>
++ * Copyright (C) 2001 Jun Nakajima <jun.nakajima@intel.com>
++ *
++ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++ *
++ * ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
++ */
++
++#include <linux/init.h>
++#include <linux/config.h>
++#include <linux/acpi.h>
++#include <linux/efi.h>
++#include <linux/module.h>
++#include <linux/dmi.h>
++#include <linux/irq.h>
++
++#include <asm/pgtable.h>
++#include <asm/io_apic.h>
++#include <asm/apic.h>
++#include <asm/io.h>
++#include <asm/mpspec.h>
++
++#ifdef CONFIG_X86_64
++
++extern void __init clustered_apic_check(void);
++
++extern int gsi_irq_sharing(int gsi);
++#include <asm/proto.h>
++
++static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id) { return 0; }
++
++
++#else /* X86 */
++
++#ifdef CONFIG_X86_LOCAL_APIC
++#include <mach_apic.h>
++#include <mach_mpparse.h>
++#endif /* CONFIG_X86_LOCAL_APIC */
++
++static inline int gsi_irq_sharing(int gsi) { return gsi; }
++
++#endif /* X86 */
++
++#define BAD_MADT_ENTRY(entry, end) ( \
++ (!entry) || (unsigned long)entry + sizeof(*entry) > end || \
++ ((acpi_table_entry_header *)entry)->length != sizeof(*entry))
++
++#define PREFIX "ACPI: "
++
++int acpi_noirq __initdata; /* skip ACPI IRQ initialization */
++int acpi_pci_disabled __initdata; /* skip ACPI PCI scan and IRQ initialization */
++int acpi_ht __initdata = 1; /* enable HT */
++
++int acpi_lapic;
++int acpi_ioapic;
++int acpi_strict;
++EXPORT_SYMBOL(acpi_strict);
++
++acpi_interrupt_flags acpi_sci_flags __initdata;
++int acpi_sci_override_gsi __initdata;
++int acpi_skip_timer_override __initdata;
++
++#ifdef CONFIG_X86_LOCAL_APIC
++static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE;
++#endif
++
++#ifndef __HAVE_ARCH_CMPXCHG
++#warning ACPI uses CMPXCHG, i486 and later hardware
++#endif
++
++#define MAX_MADT_ENTRIES 256
++u8 x86_acpiid_to_apicid[MAX_MADT_ENTRIES] =
++ {[0 ... MAX_MADT_ENTRIES - 1] = 0xff };
++EXPORT_SYMBOL(x86_acpiid_to_apicid);
++
++/* --------------------------------------------------------------------------
++ Boot-time Configuration
++ -------------------------------------------------------------------------- */
++
++/*
++ * The default interrupt routing model is PIC (8259). This gets
++ * overriden if IOAPICs are enumerated (below).
++ */
++enum acpi_irq_model_id acpi_irq_model = ACPI_IRQ_MODEL_PIC;
++
++#if defined(CONFIG_X86_64) && !defined(CONFIG_XEN)
++
++/* rely on all ACPI tables being in the direct mapping */
++char *__acpi_map_table(unsigned long phys_addr, unsigned long size)
++{
++ if (!phys_addr || !size)
++ return NULL;
++
++ if (phys_addr+size <= (end_pfn_map << PAGE_SHIFT) + PAGE_SIZE)
++ return __va(phys_addr);
++
++ return NULL;
++}
++
++#else
++
++/*
++ * Temporarily use the virtual area starting from FIX_IO_APIC_BASE_END,
++ * to map the target physical address. The problem is that set_fixmap()
++ * provides a single page, and it is possible that the page is not
++ * sufficient.
++ * By using this area, we can map up to MAX_IO_APICS pages temporarily,
++ * i.e. until the next __va_range() call.
++ *
++ * Important Safety Note: The fixed I/O APIC page numbers are *subtracted*
++ * from the fixed base. That's why we start at FIX_IO_APIC_BASE_END and
++ * count idx down while incrementing the phys address.
++ */
++char *__acpi_map_table(unsigned long phys, unsigned long size)
++{
++ unsigned long base, offset, mapped_size;
++ int idx;
++
++#ifndef CONFIG_XEN
++ if (phys + size < 8 * 1024 * 1024)
++ return __va(phys);
++#endif
++
++ offset = phys & (PAGE_SIZE - 1);
++ mapped_size = PAGE_SIZE - offset;
++ set_fixmap(FIX_ACPI_END, phys);
++ base = fix_to_virt(FIX_ACPI_END);
++
++ /*
++ * Most cases can be covered by the below.
++ */
++ idx = FIX_ACPI_END;
++ while (mapped_size < size) {
++ if (--idx < FIX_ACPI_BEGIN)
++ return NULL; /* cannot handle this */
++ phys += PAGE_SIZE;
++ set_fixmap(idx, phys);
++ mapped_size += PAGE_SIZE;
++ }
++
++ return ((unsigned char *)base + offset);
++}
++#endif
++
++#ifdef CONFIG_PCI_MMCONFIG
++/* The physical address of the MMCONFIG aperture. Set from ACPI tables. */
++struct acpi_table_mcfg_config *pci_mmcfg_config;
++int pci_mmcfg_config_num;
++
++int __init acpi_parse_mcfg(unsigned long phys_addr, unsigned long size)
++{
++ struct acpi_table_mcfg *mcfg;
++ unsigned long i;
++ int config_size;
++
++ if (!phys_addr || !size)
++ return -EINVAL;
++
++ mcfg = (struct acpi_table_mcfg *)__acpi_map_table(phys_addr, size);
++ if (!mcfg) {
++ printk(KERN_WARNING PREFIX "Unable to map MCFG\n");
++ return -ENODEV;
++ }
++
++ /* how many config structures do we have */
++ pci_mmcfg_config_num = 0;
++ i = size - sizeof(struct acpi_table_mcfg);
++ while (i >= sizeof(struct acpi_table_mcfg_config)) {
++ ++pci_mmcfg_config_num;
++ i -= sizeof(struct acpi_table_mcfg_config);
++ };
++ if (pci_mmcfg_config_num == 0) {
++ printk(KERN_ERR PREFIX "MMCONFIG has no entries\n");
++ return -ENODEV;
++ }
++
++ config_size = pci_mmcfg_config_num * sizeof(*pci_mmcfg_config);
++ pci_mmcfg_config = kmalloc(config_size, GFP_KERNEL);
++ if (!pci_mmcfg_config) {
++ printk(KERN_WARNING PREFIX
++ "No memory for MCFG config tables\n");
++ return -ENOMEM;
++ }
++
++ memcpy(pci_mmcfg_config, &mcfg->config, config_size);
++ for (i = 0; i < pci_mmcfg_config_num; ++i) {
++ if (mcfg->config[i].base_reserved) {
++ printk(KERN_ERR PREFIX
++ "MMCONFIG not in low 4GB of memory\n");
++ return -ENODEV;
++ }
++ }
++
++ return 0;
++}
++#endif /* CONFIG_PCI_MMCONFIG */
++
++#ifdef CONFIG_X86_LOCAL_APIC
++static int __init acpi_parse_madt(unsigned long phys_addr, unsigned long size)
++{
++ struct acpi_table_madt *madt = NULL;
++
++ if (!phys_addr || !size)
++ return -EINVAL;
++
++ madt = (struct acpi_table_madt *)__acpi_map_table(phys_addr, size);
++ if (!madt) {
++ printk(KERN_WARNING PREFIX "Unable to map MADT\n");
++ return -ENODEV;
++ }
++
++ if (madt->lapic_address) {
++ acpi_lapic_addr = (u64) madt->lapic_address;
++
++ printk(KERN_DEBUG PREFIX "Local APIC address 0x%08x\n",
++ madt->lapic_address);
++ }
++
++ acpi_madt_oem_check(madt->header.oem_id, madt->header.oem_table_id);
++
++ return 0;
++}
++
++static int __init
++acpi_parse_lapic(acpi_table_entry_header * header, const unsigned long end)
++{
++ struct acpi_table_lapic *processor = NULL;
++
++ processor = (struct acpi_table_lapic *)header;
++
++ if (BAD_MADT_ENTRY(processor, end))
++ return -EINVAL;
++
++ acpi_table_print_madt_entry(header);
++
++ /* Record local apic id only when enabled */
++ if (processor->flags.enabled)
++ x86_acpiid_to_apicid[processor->acpi_id] = processor->id;
++
++ /*
++ * We need to register disabled CPU as well to permit
++ * counting disabled CPUs. This allows us to size
++ * cpus_possible_map more accurately, to permit
++ * to not preallocating memory for all NR_CPUS
++ * when we use CPU hotplug.
++ */
++ mp_register_lapic(processor->id, /* APIC ID */
++ processor->flags.enabled); /* Enabled? */
++
++ return 0;
++}
++
++static int __init
++acpi_parse_lapic_addr_ovr(acpi_table_entry_header * header,
++ const unsigned long end)
++{
++ struct acpi_table_lapic_addr_ovr *lapic_addr_ovr = NULL;
++
++ lapic_addr_ovr = (struct acpi_table_lapic_addr_ovr *)header;
++
++ if (BAD_MADT_ENTRY(lapic_addr_ovr, end))
++ return -EINVAL;
++
++ acpi_lapic_addr = lapic_addr_ovr->address;
++
++ return 0;
++}
++
++static int __init
++acpi_parse_lapic_nmi(acpi_table_entry_header * header, const unsigned long end)
++{
++ struct acpi_table_lapic_nmi *lapic_nmi = NULL;
++
++ lapic_nmi = (struct acpi_table_lapic_nmi *)header;
++
++ if (BAD_MADT_ENTRY(lapic_nmi, end))
++ return -EINVAL;
++
++ acpi_table_print_madt_entry(header);
++
++ if (lapic_nmi->lint != 1)
++ printk(KERN_WARNING PREFIX "NMI not connected to LINT 1!\n");
++
++ return 0;
++}
++
++#endif /*CONFIG_X86_LOCAL_APIC */
++
++#ifdef CONFIG_X86_IO_APIC
++
++static int __init
++acpi_parse_ioapic(acpi_table_entry_header * header, const unsigned long end)
++{
++ struct acpi_table_ioapic *ioapic = NULL;
++
++ ioapic = (struct acpi_table_ioapic *)header;
++
++ if (BAD_MADT_ENTRY(ioapic, end))
++ return -EINVAL;
++
++ acpi_table_print_madt_entry(header);
++
++ mp_register_ioapic(ioapic->id,
++ ioapic->address, ioapic->global_irq_base);
++
++ return 0;
++}
++
++/*
++ * Parse Interrupt Source Override for the ACPI SCI
++ */
++static void acpi_sci_ioapic_setup(u32 gsi, u16 polarity, u16 trigger)
++{
++ if (trigger == 0) /* compatible SCI trigger is level */
++ trigger = 3;
++
++ if (polarity == 0) /* compatible SCI polarity is low */
++ polarity = 3;
++
++ /* Command-line over-ride via acpi_sci= */
++ if (acpi_sci_flags.trigger)
++ trigger = acpi_sci_flags.trigger;
++
++ if (acpi_sci_flags.polarity)
++ polarity = acpi_sci_flags.polarity;
++
++ /*
++ * mp_config_acpi_legacy_irqs() already setup IRQs < 16
++ * If GSI is < 16, this will update its flags,
++ * else it will create a new mp_irqs[] entry.
++ */
++ mp_override_legacy_irq(gsi, polarity, trigger, gsi);
++
++ /*
++ * stash over-ride to indicate we've been here
++ * and for later update of acpi_fadt
++ */
++ acpi_sci_override_gsi = gsi;
++ return;
++}
++
++static int __init
++acpi_parse_int_src_ovr(acpi_table_entry_header * header,
++ const unsigned long end)
++{
++ struct acpi_table_int_src_ovr *intsrc = NULL;
++
++ intsrc = (struct acpi_table_int_src_ovr *)header;
++
++ if (BAD_MADT_ENTRY(intsrc, end))
++ return -EINVAL;
++
++ acpi_table_print_madt_entry(header);
++
++ if (intsrc->bus_irq == acpi_fadt.sci_int) {
++ acpi_sci_ioapic_setup(intsrc->global_irq,
++ intsrc->flags.polarity,
++ intsrc->flags.trigger);
++ return 0;
++ }
++
++ if (acpi_skip_timer_override &&
++ intsrc->bus_irq == 0 && intsrc->global_irq == 2) {
++ printk(PREFIX "BIOS IRQ0 pin2 override ignored.\n");
++ return 0;
++ }
++
++ mp_override_legacy_irq(intsrc->bus_irq,
++ intsrc->flags.polarity,
++ intsrc->flags.trigger, intsrc->global_irq);
++
++ return 0;
++}
++
++static int __init
++acpi_parse_nmi_src(acpi_table_entry_header * header, const unsigned long end)
++{
++ struct acpi_table_nmi_src *nmi_src = NULL;
++
++ nmi_src = (struct acpi_table_nmi_src *)header;
++
++ if (BAD_MADT_ENTRY(nmi_src, end))
++ return -EINVAL;
++
++ acpi_table_print_madt_entry(header);
++
++ /* TBD: Support nimsrc entries? */
++
++ return 0;
++}
++
++#endif /* CONFIG_X86_IO_APIC */
++
++/*
++ * acpi_pic_sci_set_trigger()
++ *
++ * use ELCR to set PIC-mode trigger type for SCI
++ *
++ * If a PIC-mode SCI is not recognized or gives spurious IRQ7's
++ * it may require Edge Trigger -- use "acpi_sci=edge"
++ *
++ * Port 0x4d0-4d1 are ECLR1 and ECLR2, the Edge/Level Control Registers
++ * for the 8259 PIC. bit[n] = 1 means irq[n] is Level, otherwise Edge.
++ * ECLR1 is IRQ's 0-7 (IRQ 0, 1, 2 must be 0)
++ * ECLR2 is IRQ's 8-15 (IRQ 8, 13 must be 0)
++ */
++
++void __init acpi_pic_sci_set_trigger(unsigned int irq, u16 trigger)
++{
++ unsigned int mask = 1 << irq;
++ unsigned int old, new;
++
++ /* Real old ELCR mask */
++ old = inb(0x4d0) | (inb(0x4d1) << 8);
++
++ /*
++ * If we use ACPI to set PCI irq's, then we should clear ELCR
++ * since we will set it correctly as we enable the PCI irq
++ * routing.
++ */
++ new = acpi_noirq ? old : 0;
++
++ /*
++ * Update SCI information in the ELCR, it isn't in the PCI
++ * routing tables..
++ */
++ switch (trigger) {
++ case 1: /* Edge - clear */
++ new &= ~mask;
++ break;
++ case 3: /* Level - set */
++ new |= mask;
++ break;
++ }
++
++ if (old == new)
++ return;
++
++ printk(PREFIX "setting ELCR to %04x (from %04x)\n", new, old);
++ outb(new, 0x4d0);
++ outb(new >> 8, 0x4d1);
++}
++
++int acpi_gsi_to_irq(u32 gsi, unsigned int *irq)
++{
++#ifdef CONFIG_X86_IO_APIC
++ if (use_pci_vector() && !platform_legacy_irq(gsi))
++ *irq = IO_APIC_VECTOR(gsi);
++ else
++#endif
++ *irq = gsi_irq_sharing(gsi);
++ return 0;
++}
++
++/*
++ * success: return IRQ number (>=0)
++ * failure: return < 0
++ */
++int acpi_register_gsi(u32 gsi, int triggering, int polarity)
++{
++ unsigned int irq;
++ unsigned int plat_gsi = gsi;
++
++#ifdef CONFIG_PCI
++ /*
++ * Make sure all (legacy) PCI IRQs are set as level-triggered.
++ */
++ if (acpi_irq_model == ACPI_IRQ_MODEL_PIC) {
++ extern void eisa_set_level_irq(unsigned int irq);
++
++ if (triggering == ACPI_LEVEL_SENSITIVE)
++ eisa_set_level_irq(gsi);
++ }
++#endif
++
++#ifdef CONFIG_X86_IO_APIC
++ if (acpi_irq_model == ACPI_IRQ_MODEL_IOAPIC) {
++ plat_gsi = mp_register_gsi(gsi, triggering, polarity);
++ }
++#endif
++ acpi_gsi_to_irq(plat_gsi, &irq);
++ return irq;
++}
++
++EXPORT_SYMBOL(acpi_register_gsi);
++
++/*
++ * ACPI based hotplug support for CPU
++ */
++#ifdef CONFIG_ACPI_HOTPLUG_CPU
++int acpi_map_lsapic(acpi_handle handle, int *pcpu)
++{
++ /* TBD */
++ return -EINVAL;
++}
++
++EXPORT_SYMBOL(acpi_map_lsapic);
++
++int acpi_unmap_lsapic(int cpu)
++{
++ /* TBD */
++ return -EINVAL;
++}
++
++EXPORT_SYMBOL(acpi_unmap_lsapic);
++#endif /* CONFIG_ACPI_HOTPLUG_CPU */
++
++int acpi_register_ioapic(acpi_handle handle, u64 phys_addr, u32 gsi_base)
++{
++ /* TBD */
++ return -EINVAL;
++}
++
++EXPORT_SYMBOL(acpi_register_ioapic);
++
++int acpi_unregister_ioapic(acpi_handle handle, u32 gsi_base)
++{
++ /* TBD */
++ return -EINVAL;
++}
++
++EXPORT_SYMBOL(acpi_unregister_ioapic);
++
++static unsigned long __init
++acpi_scan_rsdp(unsigned long start, unsigned long length)
++{
++ unsigned long offset = 0;
++ unsigned long sig_len = sizeof("RSD PTR ") - 1;
++ unsigned long vstart = (unsigned long)isa_bus_to_virt(start);
++
++ /*
++ * Scan all 16-byte boundaries of the physical memory region for the
++ * RSDP signature.
++ */
++ for (offset = 0; offset < length; offset += 16) {
++ if (strncmp((char *)(vstart + offset), "RSD PTR ", sig_len))
++ continue;
++ return (start + offset);
++ }
++
++ return 0;
++}
++
++static int __init acpi_parse_sbf(unsigned long phys_addr, unsigned long size)
++{
++ struct acpi_table_sbf *sb;
++
++ if (!phys_addr || !size)
++ return -EINVAL;
++
++ sb = (struct acpi_table_sbf *)__acpi_map_table(phys_addr, size);
++ if (!sb) {
++ printk(KERN_WARNING PREFIX "Unable to map SBF\n");
++ return -ENODEV;
++ }
++
++ sbf_port = sb->sbf_cmos; /* Save CMOS port */
++
++ return 0;
++}
++
++#ifdef CONFIG_HPET_TIMER
++
++static int __init acpi_parse_hpet(unsigned long phys, unsigned long size)
++{
++ struct acpi_table_hpet *hpet_tbl;
++
++ if (!phys || !size)
++ return -EINVAL;
++
++ hpet_tbl = (struct acpi_table_hpet *)__acpi_map_table(phys, size);
++ if (!hpet_tbl) {
++ printk(KERN_WARNING PREFIX "Unable to map HPET\n");
++ return -ENODEV;
++ }
++
++ if (hpet_tbl->addr.space_id != ACPI_SPACE_MEM) {
++ printk(KERN_WARNING PREFIX "HPET timers must be located in "
++ "memory.\n");
++ return -1;
++ }
++#ifdef CONFIG_X86_64
++ vxtime.hpet_address = hpet_tbl->addr.addrl |
++ ((long)hpet_tbl->addr.addrh << 32);
++
++ printk(KERN_INFO PREFIX "HPET id: %#x base: %#lx\n",
++ hpet_tbl->id, vxtime.hpet_address);
++#else /* X86 */
++ {
++ extern unsigned long hpet_address;
++
++ hpet_address = hpet_tbl->addr.addrl;
++ printk(KERN_INFO PREFIX "HPET id: %#x base: %#lx\n",
++ hpet_tbl->id, hpet_address);
++ }
++#endif /* X86 */
++
++ return 0;
++}
++#else
++#define acpi_parse_hpet NULL
++#endif
++
++#ifdef CONFIG_X86_PM_TIMER
++extern u32 pmtmr_ioport;
++#endif
++
++static int __init acpi_parse_fadt(unsigned long phys, unsigned long size)
++{
++ struct fadt_descriptor_rev2 *fadt = NULL;
++
++ fadt = (struct fadt_descriptor_rev2 *)__acpi_map_table(phys, size);
++ if (!fadt) {
++ printk(KERN_WARNING PREFIX "Unable to map FADT\n");
++ return 0;
++ }
++ /* initialize sci_int early for INT_SRC_OVR MADT parsing */
++ acpi_fadt.sci_int = fadt->sci_int;
++
++ /* initialize rev and apic_phys_dest_mode for x86_64 genapic */
++ acpi_fadt.revision = fadt->revision;
++ acpi_fadt.force_apic_physical_destination_mode =
++ fadt->force_apic_physical_destination_mode;
++
++#if defined(CONFIG_X86_PM_TIMER) && !defined(CONFIG_XEN)
++ /* detect the location of the ACPI PM Timer */
++ if (fadt->revision >= FADT2_REVISION_ID) {
++ /* FADT rev. 2 */
++ if (fadt->xpm_tmr_blk.address_space_id !=
++ ACPI_ADR_SPACE_SYSTEM_IO)
++ return 0;
++
++ pmtmr_ioport = fadt->xpm_tmr_blk.address;
++ /*
++ * "X" fields are optional extensions to the original V1.0
++ * fields, so we must selectively expand V1.0 fields if the
++ * corresponding X field is zero.
++ */
++ if (!pmtmr_ioport)
++ pmtmr_ioport = fadt->V1_pm_tmr_blk;
++ } else {
++ /* FADT rev. 1 */
++ pmtmr_ioport = fadt->V1_pm_tmr_blk;
++ }
++ if (pmtmr_ioport)
++ printk(KERN_INFO PREFIX "PM-Timer IO Port: %#x\n",
++ pmtmr_ioport);
++#endif
++ return 0;
++}
++
++unsigned long __init acpi_find_rsdp(void)
++{
++ unsigned long rsdp_phys = 0;
++
++ if (efi_enabled) {
++ if (efi.acpi20 != EFI_INVALID_TABLE_ADDR)
++ return efi.acpi20;
++ else if (efi.acpi != EFI_INVALID_TABLE_ADDR)
++ return efi.acpi;
++ }
++ /*
++ * Scan memory looking for the RSDP signature. First search EBDA (low
++ * memory) paragraphs and then search upper memory (E0000-FFFFF).
++ */
++ rsdp_phys = acpi_scan_rsdp(0, 0x400);
++ if (!rsdp_phys)
++ rsdp_phys = acpi_scan_rsdp(0xE0000, 0x20000);
++
++ return rsdp_phys;
++}
++
++#ifdef CONFIG_X86_LOCAL_APIC
++/*
++ * Parse LAPIC entries in MADT
++ * returns 0 on success, < 0 on error
++ */
++static int __init acpi_parse_madt_lapic_entries(void)
++{
++ int count;
++
++ if (!cpu_has_apic)
++ return -ENODEV;
++
++ /*
++ * Note that the LAPIC address is obtained from the MADT (32-bit value)
++ * and (optionally) overriden by a LAPIC_ADDR_OVR entry (64-bit value).
++ */
++
++ count =
++ acpi_table_parse_madt(ACPI_MADT_LAPIC_ADDR_OVR,
++ acpi_parse_lapic_addr_ovr, 0);
++ if (count < 0) {
++ printk(KERN_ERR PREFIX
++ "Error parsing LAPIC address override entry\n");
++ return count;
++ }
++
++ mp_register_lapic_address(acpi_lapic_addr);
++
++ count = acpi_table_parse_madt(ACPI_MADT_LAPIC, acpi_parse_lapic,
++ MAX_APICS);
++ if (!count) {
++ printk(KERN_ERR PREFIX "No LAPIC entries present\n");
++ /* TBD: Cleanup to allow fallback to MPS */
++ return -ENODEV;
++ } else if (count < 0) {
++ printk(KERN_ERR PREFIX "Error parsing LAPIC entry\n");
++ /* TBD: Cleanup to allow fallback to MPS */
++ return count;
++ }
++
++ count =
++ acpi_table_parse_madt(ACPI_MADT_LAPIC_NMI, acpi_parse_lapic_nmi, 0);
++ if (count < 0) {
++ printk(KERN_ERR PREFIX "Error parsing LAPIC NMI entry\n");
++ /* TBD: Cleanup to allow fallback to MPS */
++ return count;
++ }
++ return 0;
++}
++#endif /* CONFIG_X86_LOCAL_APIC */
++
++#ifdef CONFIG_X86_IO_APIC
++/*
++ * Parse IOAPIC related entries in MADT
++ * returns 0 on success, < 0 on error
++ */
++static int __init acpi_parse_madt_ioapic_entries(void)
++{
++ int count;
++
++ /*
++ * ACPI interpreter is required to complete interrupt setup,
++ * so if it is off, don't enumerate the io-apics with ACPI.
++ * If MPS is present, it will handle them,
++ * otherwise the system will stay in PIC mode
++ */
++ if (acpi_disabled || acpi_noirq) {
++ return -ENODEV;
++ }
++
++ if (!cpu_has_apic)
++ return -ENODEV;
++
++ /*
++ * if "noapic" boot option, don't look for IO-APICs
++ */
++ if (skip_ioapic_setup) {
++ printk(KERN_INFO PREFIX "Skipping IOAPIC probe "
++ "due to 'noapic' option.\n");
++ return -ENODEV;
++ }
++
++ count =
++ acpi_table_parse_madt(ACPI_MADT_IOAPIC, acpi_parse_ioapic,
++ MAX_IO_APICS);
++ if (!count) {
++ printk(KERN_ERR PREFIX "No IOAPIC entries present\n");
++ return -ENODEV;
++ } else if (count < 0) {
++ printk(KERN_ERR PREFIX "Error parsing IOAPIC entry\n");
++ return count;
++ }
++
++ count =
++ acpi_table_parse_madt(ACPI_MADT_INT_SRC_OVR, acpi_parse_int_src_ovr,
++ NR_IRQ_VECTORS);
++ if (count < 0) {
++ printk(KERN_ERR PREFIX
++ "Error parsing interrupt source overrides entry\n");
++ /* TBD: Cleanup to allow fallback to MPS */
++ return count;
++ }
++
++ /*
++ * If BIOS did not supply an INT_SRC_OVR for the SCI
++ * pretend we got one so we can set the SCI flags.
++ */
++ if (!acpi_sci_override_gsi)
++ acpi_sci_ioapic_setup(acpi_fadt.sci_int, 0, 0);
++
++ /* Fill in identity legacy mapings where no override */
++ mp_config_acpi_legacy_irqs();
++
++ count =
++ acpi_table_parse_madt(ACPI_MADT_NMI_SRC, acpi_parse_nmi_src,
++ NR_IRQ_VECTORS);
++ if (count < 0) {
++ printk(KERN_ERR PREFIX "Error parsing NMI SRC entry\n");
++ /* TBD: Cleanup to allow fallback to MPS */
++ return count;
++ }
++
++ return 0;
++}
++#else
++static inline int acpi_parse_madt_ioapic_entries(void)
++{
++ return -1;
++}
++#endif /* !CONFIG_X86_IO_APIC */
++
++static void __init acpi_process_madt(void)
++{
++#ifdef CONFIG_X86_LOCAL_APIC
++ int count, error;
++
++ count = acpi_table_parse(ACPI_APIC, acpi_parse_madt);
++ if (count >= 1) {
++
++ /*
++ * Parse MADT LAPIC entries
++ */
++ error = acpi_parse_madt_lapic_entries();
++ if (!error) {
++ acpi_lapic = 1;
++
++#ifdef CONFIG_X86_GENERICARCH
++ generic_bigsmp_probe();
++#endif
++ /*
++ * Parse MADT IO-APIC entries
++ */
++ error = acpi_parse_madt_ioapic_entries();
++ if (!error) {
++ acpi_irq_model = ACPI_IRQ_MODEL_IOAPIC;
++ acpi_irq_balance_set(NULL);
++ acpi_ioapic = 1;
++
++ smp_found_config = 1;
++ clustered_apic_check();
++ }
++ }
++ if (error == -EINVAL) {
++ /*
++ * Dell Precision Workstation 410, 610 come here.
++ */
++ printk(KERN_ERR PREFIX
++ "Invalid BIOS MADT, disabling ACPI\n");
++ disable_acpi();
++ }
++ }
++#endif
++ return;
++}
++
++extern int acpi_force;
++
++#ifdef __i386__
++
++static int __init disable_acpi_irq(struct dmi_system_id *d)
++{
++ if (!acpi_force) {
++ printk(KERN_NOTICE "%s detected: force use of acpi=noirq\n",
++ d->ident);
++ acpi_noirq_set();
++ }
++ return 0;
++}
++
++static int __init disable_acpi_pci(struct dmi_system_id *d)
++{
++ if (!acpi_force) {
++ printk(KERN_NOTICE "%s detected: force use of pci=noacpi\n",
++ d->ident);
++ acpi_disable_pci();
++ }
++ return 0;
++}
++
++static int __init dmi_disable_acpi(struct dmi_system_id *d)
++{
++ if (!acpi_force) {
++ printk(KERN_NOTICE "%s detected: acpi off\n", d->ident);
++ disable_acpi();
++ } else {
++ printk(KERN_NOTICE
++ "Warning: DMI blacklist says broken, but acpi forced\n");
++ }
++ return 0;
++}
++
++/*
++ * Limit ACPI to CPU enumeration for HT
++ */
++static int __init force_acpi_ht(struct dmi_system_id *d)
++{
++ if (!acpi_force) {
++ printk(KERN_NOTICE "%s detected: force use of acpi=ht\n",
++ d->ident);
++ disable_acpi();
++ acpi_ht = 1;
++ } else {
++ printk(KERN_NOTICE
++ "Warning: acpi=force overrules DMI blacklist: acpi=ht\n");
++ }
++ return 0;
++}
++
++/*
++ * If your system is blacklisted here, but you find that acpi=force
++ * works for you, please contact acpi-devel@sourceforge.net
++ */
++static struct dmi_system_id __initdata acpi_dmi_table[] = {
++ /*
++ * Boxes that need ACPI disabled
++ */
++ {
++ .callback = dmi_disable_acpi,
++ .ident = "IBM Thinkpad",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
++ DMI_MATCH(DMI_BOARD_NAME, "2629H1G"),
++ },
++ },
++
++ /*
++ * Boxes that need acpi=ht
++ */
++ {
++ .callback = force_acpi_ht,
++ .ident = "FSC Primergy T850",
++ .matches = {
++ DMI_MATCH(DMI_SYS_VENDOR, "FUJITSU SIEMENS"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "PRIMERGY T850"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "DELL GX240",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "Dell Computer Corporation"),
++ DMI_MATCH(DMI_BOARD_NAME, "OptiPlex GX240"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "HP VISUALIZE NT Workstation",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "Hewlett-Packard"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "HP VISUALIZE NT Workstation"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "Compaq Workstation W8000",
++ .matches = {
++ DMI_MATCH(DMI_SYS_VENDOR, "Compaq"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "Workstation W8000"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "ASUS P4B266",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
++ DMI_MATCH(DMI_BOARD_NAME, "P4B266"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "ASUS P2B-DS",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
++ DMI_MATCH(DMI_BOARD_NAME, "P2B-DS"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "ASUS CUR-DLS",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
++ DMI_MATCH(DMI_BOARD_NAME, "CUR-DLS"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "ABIT i440BX-W83977",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ABIT <http://www.abit.com>"),
++ DMI_MATCH(DMI_BOARD_NAME, "i440BX-W83977 (BP6)"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "IBM Bladecenter",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
++ DMI_MATCH(DMI_BOARD_NAME, "IBM eServer BladeCenter HS20"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "IBM eServer xSeries 360",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
++ DMI_MATCH(DMI_BOARD_NAME, "eServer xSeries 360"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "IBM eserver xSeries 330",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
++ DMI_MATCH(DMI_BOARD_NAME, "eserver xSeries 330"),
++ },
++ },
++ {
++ .callback = force_acpi_ht,
++ .ident = "IBM eserver xSeries 440",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "IBM"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "eserver xSeries 440"),
++ },
++ },
++
++ /*
++ * Boxes that need ACPI PCI IRQ routing disabled
++ */
++ {
++ .callback = disable_acpi_irq,
++ .ident = "ASUS A7V",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC"),
++ DMI_MATCH(DMI_BOARD_NAME, "<A7V>"),
++ /* newer BIOS, Revision 1011, does work */
++ DMI_MATCH(DMI_BIOS_VERSION,
++ "ASUS A7V ACPI BIOS Revision 1007"),
++ },
++ },
++
++ /*
++ * Boxes that need ACPI PCI IRQ routing and PCI scan disabled
++ */
++ { /* _BBN 0 bug */
++ .callback = disable_acpi_pci,
++ .ident = "ASUS PR-DLS",
++ .matches = {
++ DMI_MATCH(DMI_BOARD_VENDOR, "ASUSTeK Computer INC."),
++ DMI_MATCH(DMI_BOARD_NAME, "PR-DLS"),
++ DMI_MATCH(DMI_BIOS_VERSION,
++ "ASUS PR-DLS ACPI BIOS Revision 1010"),
++ DMI_MATCH(DMI_BIOS_DATE, "03/21/2003")
++ },
++ },
++ {
++ .callback = disable_acpi_pci,
++ .ident = "Acer TravelMate 36x Laptop",
++ .matches = {
++ DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"),
++ },
++ },
++ {}
++};
++
++#endif /* __i386__ */
++
++/*
++ * acpi_boot_table_init() and acpi_boot_init()
++ * called from setup_arch(), always.
++ * 1. checksums all tables
++ * 2. enumerates lapics
++ * 3. enumerates io-apics
++ *
++ * acpi_table_init() is separate to allow reading SRAT without
++ * other side effects.
++ *
++ * side effects of acpi_boot_init:
++ * acpi_lapic = 1 if LAPIC found
++ * acpi_ioapic = 1 if IOAPIC found
++ * if (acpi_lapic && acpi_ioapic) smp_found_config = 1;
++ * if acpi_blacklisted() acpi_disabled = 1;
++ * acpi_irq_model=...
++ * ...
++ *
++ * return value: (currently ignored)
++ * 0: success
++ * !0: failure
++ */
++
++int __init acpi_boot_table_init(void)
++{
++ int error;
++
++#ifdef __i386__
++ dmi_check_system(acpi_dmi_table);
++#endif
++
++ /*
++ * If acpi_disabled, bail out
++ * One exception: acpi=ht continues far enough to enumerate LAPICs
++ */
++ if (acpi_disabled && !acpi_ht)
++ return 1;
++
++ /*
++ * Initialize the ACPI boot-time table parser.
++ */
++ error = acpi_table_init();
++ if (error) {
++ disable_acpi();
++ return error;
++ }
++
++ acpi_table_parse(ACPI_BOOT, acpi_parse_sbf);
++
++ /*
++ * blacklist may disable ACPI entirely
++ */
++ error = acpi_blacklisted();
++ if (error) {
++ if (acpi_force) {
++ printk(KERN_WARNING PREFIX "acpi=force override\n");
++ } else {
++ printk(KERN_WARNING PREFIX "Disabling ACPI support\n");
++ disable_acpi();
++ return error;
++ }
++ }
++
++ return 0;
++}
++
++int __init acpi_boot_init(void)
++{
++ /*
++ * If acpi_disabled, bail out
++ * One exception: acpi=ht continues far enough to enumerate LAPICs
++ */
++ if (acpi_disabled && !acpi_ht)
++ return 1;
++
++ acpi_table_parse(ACPI_BOOT, acpi_parse_sbf);
++
++ /*
++ * set sci_int and PM timer address
++ */
++ acpi_table_parse(ACPI_FADT, acpi_parse_fadt);
++
++ /*
++ * Process the Multiple APIC Description Table (MADT), if present
++ */
++ acpi_process_madt();
++
++ acpi_table_parse(ACPI_HPET, acpi_parse_hpet);
++
++ return 0;
++}
+diff -urNp linux-2.6/arch/i386/kernel/acpi/Makefile new/arch/i386/kernel/acpi/Makefile
+--- linux-2.6/arch/i386/kernel/acpi/Makefile 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/acpi/Makefile 2006-05-09 12:32:33.000000000 +0200
+@@ -6,3 +6,7 @@ ifneq ($(CONFIG_ACPI_PROCESSOR),)
+ obj-y += cstate.o processor.o
+ endif
+
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++obj-y := $(call cherrypickxen, $(obj-y), $(src))
++endif
+diff -urNp linux-2.6/arch/i386/kernel/apic-xen.c new/arch/i386/kernel/apic-xen.c
+--- linux-2.6/arch/i386/kernel/apic-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/apic-xen.c 2006-05-09 12:32:33.000000000 +0200
+@@ -0,0 +1,160 @@
++/*
++ * Local APIC handling, local APIC timers
++ *
++ * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
++ * thanks to Eric Gilmore
++ * and Rolf G. Tews
++ * for testing these extensively.
++ * Maciej W. Rozycki : Various updates and fixes.
++ * Mikael Pettersson : Power Management for UP-APIC.
++ * Pavel Machek and
++ * Mikael Pettersson : PM converted to driver model.
++ */
++
++#include <linux/config.h>
++#include <linux/init.h>
++
++#include <linux/mm.h>
++#include <linux/delay.h>
++#include <linux/bootmem.h>
++#include <linux/smp_lock.h>
++#include <linux/interrupt.h>
++#include <linux/mc146818rtc.h>
++#include <linux/kernel_stat.h>
++#include <linux/sysdev.h>
++#include <linux/cpu.h>
++#include <linux/module.h>
++
++#include <asm/atomic.h>
++#include <asm/smp.h>
++#include <asm/mtrr.h>
++#include <asm/mpspec.h>
++#include <asm/desc.h>
++#include <asm/arch_hooks.h>
++#include <asm/hpet.h>
++#include <asm/i8253.h>
++
++#include <mach_apic.h>
++#include <mach_apicdef.h>
++#include <mach_ipi.h>
++
++#include "io_ports.h"
++
++#ifndef CONFIG_XEN
++/*
++ * cpu_mask that denotes the CPUs that needs timer interrupt coming in as
++ * IPIs in place of local APIC timers
++ */
++static cpumask_t timer_bcast_ipi;
++#endif
++
++/*
++ * Knob to control our willingness to enable the local APIC.
++ */
++int enable_local_apic __initdata = 0; /* -1=force-disable, +1=force-enable */
++
++/*
++ * Debug level
++ */
++int apic_verbosity;
++
++int modern_apic(void)
++{
++#ifndef CONFIG_XEN
++ unsigned int lvr, version;
++ /* AMD systems use old APIC versions, so check the CPU */
++ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
++ boot_cpu_data.x86 >= 0xf)
++ return 1;
++ lvr = apic_read(APIC_LVR);
++ version = GET_APIC_VERSION(lvr);
++ return version >= 0x14;
++#else
++ return 1;
++#endif
++}
++
++/*
++ * 'what should we do if we get a hw irq event on an illegal vector'.
++ * each architecture has to answer this themselves.
++ */
++void ack_bad_irq(unsigned int irq)
++{
++ printk("unexpected IRQ trap at vector %02x\n", irq);
++ /*
++ * Currently unexpected vectors happen only on SMP and APIC.
++ * We _must_ ack these because every local APIC has only N
++ * irq slots per priority level, and a 'hanging, unacked' IRQ
++ * holds up an irq slot - in excessive cases (when multiple
++ * unexpected vectors occur) that might lock up the APIC
++ * completely.
++ * But only ack when the APIC is enabled -AK
++ */
++ if (cpu_has_apic)
++ ack_APIC_irq();
++}
++
++int get_physical_broadcast(void)
++{
++ if (modern_apic())
++ return 0xff;
++ else
++ return 0xf;
++}
++
++#ifndef CONFIG_XEN
++#ifndef CONFIG_SMP
++static void up_apic_timer_interrupt_call(struct pt_regs *regs)
++{
++ int cpu = smp_processor_id();
++
++ /*
++ * the NMI deadlock-detector uses this.
++ */
++ per_cpu(irq_stat, cpu).apic_timer_irqs++;
++
++ smp_local_timer_interrupt(regs);
++}
++#endif
++
++void smp_send_timer_broadcast_ipi(struct pt_regs *regs)
++{
++ cpumask_t mask;
++
++ cpus_and(mask, cpu_online_map, timer_bcast_ipi);
++ if (!cpus_empty(mask)) {
++#ifdef CONFIG_SMP
++ send_IPI_mask(mask, LOCAL_TIMER_VECTOR);
++#else
++ /*
++ * We can directly call the apic timer interrupt handler
++ * in UP case. Minus all irq related functions
++ */
++ up_apic_timer_interrupt_call(regs);
++#endif
++ }
++}
++#endif
++
++int setup_profiling_timer(unsigned int multiplier)
++{
++ return -EINVAL;
++}
++
++/*
++ * This initializes the IO-APIC and APIC hardware if this is
++ * a UP kernel.
++ */
++int __init APIC_init_uniprocessor (void)
++{
++#ifdef CONFIG_X86_IO_APIC
++ if (smp_found_config)
++ if (!skip_ioapic_setup && nr_ioapics)
++ setup_IO_APIC();
++#endif
++
++ return 0;
++}
+diff -urNp linux-2.6/arch/i386/kernel/asm-offsets.c new/arch/i386/kernel/asm-offsets.c
+--- linux-2.6/arch/i386/kernel/asm-offsets.c 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/asm-offsets.c 2006-05-09 12:32:34.000000000 +0200
+@@ -13,6 +13,7 @@
+ #include <asm/fixmap.h>
+ #include <asm/processor.h>
+ #include <asm/thread_info.h>
++#include <asm/elf.h>
+
+ #define DEFINE(sym, val) \
+ asm volatile("\n->" #sym " %0 " #val : : "i" (val))
+@@ -63,10 +64,15 @@ void foo(void)
+ OFFSET(pbe_orig_address, pbe, orig_address);
+ OFFSET(pbe_next, pbe, next);
+
++#ifndef CONFIG_X86_NO_TSS
+ /* Offset from the sysenter stack to tss.esp0 */
+- DEFINE(TSS_sysenter_esp0, offsetof(struct tss_struct, esp0) -
++ DEFINE(SYSENTER_stack_esp0, offsetof(struct tss_struct, esp0) -
+ sizeof(struct tss_struct));
++#else
++ /* sysenter stack points directly to esp0 */
++ DEFINE(SYSENTER_stack_esp0, 0);
++#endif
+
+ DEFINE(PAGE_SIZE_asm, PAGE_SIZE);
+- DEFINE(VSYSCALL_BASE, __fix_to_virt(FIX_VSYSCALL));
++ DEFINE(VSYSCALL_BASE, VSYSCALL_BASE);
+ }
+diff -urNp linux-2.6/arch/i386/kernel/cpu/common-xen.c new/arch/i386/kernel/cpu/common-xen.c
+--- linux-2.6/arch/i386/kernel/cpu/common-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/cpu/common-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,732 @@
++#include <linux/init.h>
++#include <linux/string.h>
++#include <linux/delay.h>
++#include <linux/smp.h>
++#include <linux/module.h>
++#include <linux/percpu.h>
++#include <linux/bootmem.h>
++#include <asm/semaphore.h>
++#include <asm/processor.h>
++#include <asm/i387.h>
++#include <asm/msr.h>
++#include <asm/io.h>
++#include <asm/mmu_context.h>
++#ifdef CONFIG_X86_LOCAL_APIC
++#include <asm/mpspec.h>
++#include <asm/apic.h>
++#include <mach_apic.h>
++#endif
++#include <asm/hypervisor.h>
++
++#include "cpu.h"
++
++DEFINE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
++EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
++
++#ifndef CONFIG_XEN
++DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
++EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
++#endif
++
++static int cachesize_override __cpuinitdata = -1;
++static int disable_x86_fxsr __cpuinitdata;
++static int disable_x86_serial_nr __cpuinitdata = 1;
++static int disable_x86_sep __cpuinitdata;
++
++struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
++
++extern int disable_pse;
++
++static void default_init(struct cpuinfo_x86 * c)
++{
++ /* Not much we can do here... */
++ /* Check if at least it has cpuid */
++ if (c->cpuid_level == -1) {
++ /* No cpuid. It must be an ancient CPU */
++ if (c->x86 == 4)
++ strcpy(c->x86_model_id, "486");
++ else if (c->x86 == 3)
++ strcpy(c->x86_model_id, "386");
++ }
++}
++
++static struct cpu_dev default_cpu = {
++ .c_init = default_init,
++ .c_vendor = "Unknown",
++};
++static struct cpu_dev * this_cpu = &default_cpu;
++
++static int __init cachesize_setup(char *str)
++{
++ get_option (&str, &cachesize_override);
++ return 1;
++}
++__setup("cachesize=", cachesize_setup);
++
++int __cpuinit get_model_name(struct cpuinfo_x86 *c)
++{
++ unsigned int *v;
++ char *p, *q;
++
++ if (cpuid_eax(0x80000000) < 0x80000004)
++ return 0;
++
++ v = (unsigned int *) c->x86_model_id;
++ cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
++ cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
++ cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
++ c->x86_model_id[48] = 0;
++
++ /* Intel chips right-justify this string for some dumb reason;
++ undo that brain damage */
++ p = q = &c->x86_model_id[0];
++ while ( *p == ' ' )
++ p++;
++ if ( p != q ) {
++ while ( *p )
++ *q++ = *p++;
++ while ( q <= &c->x86_model_id[48] )
++ *q++ = '\0'; /* Zero-pad the rest */
++ }
++
++ return 1;
++}
++
++
++void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
++{
++ unsigned int n, dummy, ecx, edx, l2size;
++
++ n = cpuid_eax(0x80000000);
++
++ if (n >= 0x80000005) {
++ cpuid(0x80000005, &dummy, &dummy, &ecx, &edx);
++ printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
++ edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
++ c->x86_cache_size=(ecx>>24)+(edx>>24);
++ }
++
++ if (n < 0x80000006) /* Some chips just has a large L1. */
++ return;
++
++ ecx = cpuid_ecx(0x80000006);
++ l2size = ecx >> 16;
++
++ /* do processor-specific cache resizing */
++ if (this_cpu->c_size_cache)
++ l2size = this_cpu->c_size_cache(c,l2size);
++
++ /* Allow user to override all this if necessary. */
++ if (cachesize_override != -1)
++ l2size = cachesize_override;
++
++ if ( l2size == 0 )
++ return; /* Again, no L2 cache is possible */
++
++ c->x86_cache_size = l2size;
++
++ printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
++ l2size, ecx & 0xFF);
++}
++
++/* Naming convention should be: <Name> [(<Codename>)] */
++/* This table only is used unless init_<vendor>() below doesn't set it; */
++/* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
++
++/* Look up CPU names by table lookup. */
++static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
++{
++ struct cpu_model_info *info;
++
++ if ( c->x86_model >= 16 )
++ return NULL; /* Range check */
++
++ if (!this_cpu)
++ return NULL;
++
++ info = this_cpu->c_models;
++
++ while (info && info->family) {
++ if (info->family == c->x86)
++ return info->model_names[c->x86_model];
++ info++;
++ }
++ return NULL; /* Not found */
++}
++
++
++static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
++{
++ char *v = c->x86_vendor_id;
++ int i;
++ static int printed;
++
++ for (i = 0; i < X86_VENDOR_NUM; i++) {
++ if (cpu_devs[i]) {
++ if (!strcmp(v,cpu_devs[i]->c_ident[0]) ||
++ (cpu_devs[i]->c_ident[1] &&
++ !strcmp(v,cpu_devs[i]->c_ident[1]))) {
++ c->x86_vendor = i;
++ if (!early)
++ this_cpu = cpu_devs[i];
++ return;
++ }
++ }
++ }
++ if (!printed) {
++ printed++;
++ printk(KERN_ERR "CPU: Vendor unknown, using generic init.\n");
++ printk(KERN_ERR "CPU: Your system may be unstable.\n");
++ }
++ c->x86_vendor = X86_VENDOR_UNKNOWN;
++ this_cpu = &default_cpu;
++}
++
++
++static int __init x86_fxsr_setup(char * s)
++{
++ disable_x86_fxsr = 1;
++ return 1;
++}
++__setup("nofxsr", x86_fxsr_setup);
++
++
++static int __init x86_sep_setup(char * s)
++{
++ disable_x86_sep = 1;
++ return 1;
++}
++__setup("nosep", x86_sep_setup);
++
++
++/* Standard macro to see if a specific flag is changeable */
++static inline int flag_is_changeable_p(u32 flag)
++{
++ u32 f1, f2;
++
++ asm("pushfl\n\t"
++ "pushfl\n\t"
++ "popl %0\n\t"
++ "movl %0,%1\n\t"
++ "xorl %2,%0\n\t"
++ "pushl %0\n\t"
++ "popfl\n\t"
++ "pushfl\n\t"
++ "popl %0\n\t"
++ "popfl\n\t"
++ : "=&r" (f1), "=&r" (f2)
++ : "ir" (flag));
++
++ return ((f1^f2) & flag) != 0;
++}
++
++
++/* Probe for the CPUID instruction */
++static int __cpuinit have_cpuid_p(void)
++{
++ return flag_is_changeable_p(X86_EFLAGS_ID);
++}
++
++/* Do minimum CPU detection early.
++ Fields really needed: vendor, cpuid_level, family, model, mask, cache alignment.
++ The others are not touched to avoid unwanted side effects.
++
++ WARNING: this function is only called on the BP. Don't add code here
++ that is supposed to run on all CPUs. */
++static void __init early_cpu_detect(void)
++{
++ struct cpuinfo_x86 *c = &boot_cpu_data;
++
++ c->x86_cache_alignment = 32;
++
++ if (!have_cpuid_p())
++ return;
++
++ /* Get vendor name */
++ cpuid(0x00000000, &c->cpuid_level,
++ (int *)&c->x86_vendor_id[0],
++ (int *)&c->x86_vendor_id[8],
++ (int *)&c->x86_vendor_id[4]);
++
++ get_cpu_vendor(c, 1);
++
++ c->x86 = 4;
++ if (c->cpuid_level >= 0x00000001) {
++ u32 junk, tfms, cap0, misc;
++ cpuid(0x00000001, &tfms, &misc, &junk, &cap0);
++ c->x86 = (tfms >> 8) & 15;
++ c->x86_model = (tfms >> 4) & 15;
++ if (c->x86 == 0xf)
++ c->x86 += (tfms >> 20) & 0xff;
++ if (c->x86 >= 0x6)
++ c->x86_model += ((tfms >> 16) & 0xF) << 4;
++ c->x86_mask = tfms & 15;
++ if (cap0 & (1<<19))
++ c->x86_cache_alignment = ((misc >> 8) & 0xff) * 8;
++ }
++}
++
++void __cpuinit generic_identify(struct cpuinfo_x86 * c)
++{
++ u32 tfms, xlvl;
++ int ebx;
++
++ if (have_cpuid_p()) {
++ /* Get vendor name */
++ cpuid(0x00000000, &c->cpuid_level,
++ (int *)&c->x86_vendor_id[0],
++ (int *)&c->x86_vendor_id[8],
++ (int *)&c->x86_vendor_id[4]);
++
++ get_cpu_vendor(c, 0);
++ /* Initialize the standard set of capabilities */
++ /* Note that the vendor-specific code below might override */
++
++ /* Intel-defined flags: level 0x00000001 */
++ if ( c->cpuid_level >= 0x00000001 ) {
++ u32 capability, excap;
++ cpuid(0x00000001, &tfms, &ebx, &excap, &capability);
++ c->x86_capability[0] = capability;
++ c->x86_capability[4] = excap;
++ c->x86 = (tfms >> 8) & 15;
++ c->x86_model = (tfms >> 4) & 15;
++ if (c->x86 == 0xf)
++ c->x86 += (tfms >> 20) & 0xff;
++ if (c->x86 >= 0x6)
++ c->x86_model += ((tfms >> 16) & 0xF) << 4;
++ c->x86_mask = tfms & 15;
++#if defined(CONFIG_SMP) && defined(CONFIG_X86_LOCAL_APIC)
++ c->apicid = phys_pkg_id((ebx >> 24) & 0xFF, 0);
++#else
++ c->apicid = (ebx >> 24) & 0xFF;
++#endif
++ } else {
++ /* Have CPUID level 0 only - unheard of */
++ c->x86 = 4;
++ }
++
++ /* AMD-defined flags: level 0x80000001 */
++ xlvl = cpuid_eax(0x80000000);
++ if ( (xlvl & 0xffff0000) == 0x80000000 ) {
++ if ( xlvl >= 0x80000001 ) {
++ c->x86_capability[1] = cpuid_edx(0x80000001);
++ c->x86_capability[6] = cpuid_ecx(0x80000001);
++ }
++ if ( xlvl >= 0x80000004 )
++ get_model_name(c); /* Default name */
++ }
++ }
++
++ early_intel_workaround(c);
++
++#ifdef CONFIG_X86_HT
++ phys_proc_id[smp_processor_id()] = (cpuid_ebx(1) >> 24) & 0xff;
++#endif
++}
++
++static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
++{
++ if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
++ /* Disable processor serial number */
++ unsigned long lo,hi;
++ rdmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
++ lo |= 0x200000;
++ wrmsr(MSR_IA32_BBL_CR_CTL,lo,hi);
++ printk(KERN_NOTICE "CPU serial number disabled.\n");
++ clear_bit(X86_FEATURE_PN, c->x86_capability);
++
++ /* Disabling the serial number may affect the cpuid level */
++ c->cpuid_level = cpuid_eax(0);
++ }
++}
++
++static int __init x86_serial_nr_setup(char *s)
++{
++ disable_x86_serial_nr = 0;
++ return 1;
++}
++__setup("serialnumber", x86_serial_nr_setup);
++
++
++
++/*
++ * This does the hard work of actually picking apart the CPU stuff...
++ */
++void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
++{
++ int i;
++
++ c->loops_per_jiffy = loops_per_jiffy;
++ c->x86_cache_size = -1;
++ c->x86_vendor = X86_VENDOR_UNKNOWN;
++ c->cpuid_level = -1; /* CPUID not detected */
++ c->x86_model = c->x86_mask = 0; /* So far unknown... */
++ c->x86_vendor_id[0] = '\0'; /* Unset */
++ c->x86_model_id[0] = '\0'; /* Unset */
++ c->x86_max_cores = 1;
++ memset(&c->x86_capability, 0, sizeof c->x86_capability);
++
++ if (!have_cpuid_p()) {
++ /* First of all, decide if this is a 486 or higher */
++ /* It's a 486 if we can modify the AC flag */
++ if ( flag_is_changeable_p(X86_EFLAGS_AC) )
++ c->x86 = 4;
++ else
++ c->x86 = 3;
++ }
++
++ generic_identify(c);
++
++ printk(KERN_DEBUG "CPU: After generic identify, caps:");
++ for (i = 0; i < NCAPINTS; i++)
++ printk(" %08lx", c->x86_capability[i]);
++ printk("\n");
++
++ if (this_cpu->c_identify) {
++ this_cpu->c_identify(c);
++
++ printk(KERN_DEBUG "CPU: After vendor identify, caps:");
++ for (i = 0; i < NCAPINTS; i++)
++ printk(" %08lx", c->x86_capability[i]);
++ printk("\n");
++ }
++
++ /*
++ * Vendor-specific initialization. In this section we
++ * canonicalize the feature flags, meaning if there are
++ * features a certain CPU supports which CPUID doesn't
++ * tell us, CPUID claiming incorrect flags, or other bugs,
++ * we handle them here.
++ *
++ * At the end of this section, c->x86_capability better
++ * indicate the features this CPU genuinely supports!
++ */
++ if (this_cpu->c_init)
++ this_cpu->c_init(c);
++
++ /* Disable the PN if appropriate */
++ squash_the_stupid_serial_number(c);
++
++ /*
++ * The vendor-specific functions might have changed features. Now
++ * we do "generic changes."
++ */
++
++ /* TSC disabled? */
++ if ( tsc_disable )
++ clear_bit(X86_FEATURE_TSC, c->x86_capability);
++
++ /* FXSR disabled? */
++ if (disable_x86_fxsr) {
++ clear_bit(X86_FEATURE_FXSR, c->x86_capability);
++ clear_bit(X86_FEATURE_XMM, c->x86_capability);
++ }
++
++ /* SEP disabled? */
++ if (disable_x86_sep)
++ clear_bit(X86_FEATURE_SEP, c->x86_capability);
++
++ if (disable_pse)
++ clear_bit(X86_FEATURE_PSE, c->x86_capability);
++
++ /* If the model name is still unset, do table lookup. */
++ if ( !c->x86_model_id[0] ) {
++ char *p;
++ p = table_lookup_model(c);
++ if ( p )
++ strcpy(c->x86_model_id, p);
++ else
++ /* Last resort... */
++ sprintf(c->x86_model_id, "%02x/%02x",
++ c->x86, c->x86_model);
++ }
++
++ /* Now the feature flags better reflect actual CPU features! */
++
++ printk(KERN_DEBUG "CPU: After all inits, caps:");
++ for (i = 0; i < NCAPINTS; i++)
++ printk(" %08lx", c->x86_capability[i]);
++ printk("\n");
++
++ /*
++ * On SMP, boot_cpu_data holds the common feature set between
++ * all CPUs; so make sure that we indicate which features are
++ * common between the CPUs. The first time this routine gets
++ * executed, c == &boot_cpu_data.
++ */
++ if ( c != &boot_cpu_data ) {
++ /* AND the already accumulated flags with these */
++ for ( i = 0 ; i < NCAPINTS ; i++ )
++ boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
++ }
++
++ /* Init Machine Check Exception if available. */
++ mcheck_init(c);
++
++ if (c == &boot_cpu_data)
++ sysenter_setup();
++ enable_sep_cpu();
++
++ if (c == &boot_cpu_data)
++ mtrr_bp_init();
++ else
++ mtrr_ap_init();
++}
++
++#ifdef CONFIG_X86_HT
++void __cpuinit detect_ht(struct cpuinfo_x86 *c)
++{
++ u32 eax, ebx, ecx, edx;
++ int index_msb, core_bits;
++ int cpu = smp_processor_id();
++
++ cpuid(1, &eax, &ebx, &ecx, &edx);
++
++
++ if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
++ return;
++
++ smp_num_siblings = (ebx & 0xff0000) >> 16;
++
++ if (smp_num_siblings == 1) {
++ printk(KERN_INFO "CPU: Hyper-Threading is disabled\n");
++ } else if (smp_num_siblings > 1 ) {
++
++ if (smp_num_siblings > NR_CPUS) {
++ printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings);
++ smp_num_siblings = 1;
++ return;
++ }
++
++ index_msb = get_count_order(smp_num_siblings);
++ phys_proc_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb);
++
++ printk(KERN_INFO "CPU: Physical Processor ID: %d\n",
++ phys_proc_id[cpu]);
++
++ smp_num_siblings = smp_num_siblings / c->x86_max_cores;
++
++ index_msb = get_count_order(smp_num_siblings) ;
++
++ core_bits = get_count_order(c->x86_max_cores);
++
++ cpu_core_id[cpu] = phys_pkg_id((ebx >> 24) & 0xFF, index_msb) &
++ ((1 << core_bits) - 1);
++
++ if (c->x86_max_cores > 1)
++ printk(KERN_INFO "CPU: Processor Core ID: %d\n",
++ cpu_core_id[cpu]);
++ }
++}
++#endif
++
++void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
++{
++ char *vendor = NULL;
++
++ if (c->x86_vendor < X86_VENDOR_NUM)
++ vendor = this_cpu->c_vendor;
++ else if (c->cpuid_level >= 0)
++ vendor = c->x86_vendor_id;
++
++ if (vendor && strncmp(c->x86_model_id, vendor, strlen(vendor)))
++ printk("%s ", vendor);
++
++ if (!c->x86_model_id[0])
++ printk("%d86", c->x86);
++ else
++ printk("%s", c->x86_model_id);
++
++ if (c->x86_mask || c->cpuid_level >= 0)
++ printk(" stepping %02x\n", c->x86_mask);
++ else
++ printk("\n");
++}
++
++cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
++
++/* This is hacky. :)
++ * We're emulating future behavior.
++ * In the future, the cpu-specific init functions will be called implicitly
++ * via the magic of initcalls.
++ * They will insert themselves into the cpu_devs structure.
++ * Then, when cpu_init() is called, we can just iterate over that array.
++ */
++
++extern int intel_cpu_init(void);
++extern int cyrix_init_cpu(void);
++extern int nsc_init_cpu(void);
++extern int amd_init_cpu(void);
++extern int centaur_init_cpu(void);
++extern int transmeta_init_cpu(void);
++extern int rise_init_cpu(void);
++extern int nexgen_init_cpu(void);
++extern int umc_init_cpu(void);
++
++void __init early_cpu_init(void)
++{
++ intel_cpu_init();
++ cyrix_init_cpu();
++ nsc_init_cpu();
++ amd_init_cpu();
++ centaur_init_cpu();
++ transmeta_init_cpu();
++ rise_init_cpu();
++ nexgen_init_cpu();
++ umc_init_cpu();
++ early_cpu_detect();
++
++#ifdef CONFIG_DEBUG_PAGEALLOC
++ /* pse is not compatible with on-the-fly unmapping,
++ * disable it even if the cpus claim to support it.
++ */
++ clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
++ disable_pse = 1;
++#endif
++}
++
++void __cpuinit cpu_gdt_init(struct Xgt_desc_struct *gdt_descr)
++{
++ unsigned long frames[16];
++ unsigned long va;
++ int f;
++
++ for (va = gdt_descr->address, f = 0;
++ va < gdt_descr->address + gdt_descr->size;
++ va += PAGE_SIZE, f++) {
++ frames[f] = virt_to_mfn(va);
++ make_lowmem_page_readonly(
++ (void *)va, XENFEAT_writable_descriptor_tables);
++ }
++ if (HYPERVISOR_set_gdt(frames, gdt_descr->size / 8))
++ BUG();
++}
++
++/*
++ * cpu_init() initializes state that is per-CPU. Some data is already
++ * initialized (naturally) in the bootstrap process, such as the GDT
++ * and IDT. We reload them nevertheless, this function acts as a
++ * 'CPU state barrier', nothing should get across.
++ */
++void __cpuinit cpu_init(void)
++{
++ int cpu = smp_processor_id();
++#ifndef CONFIG_X86_NO_TSS
++ struct tss_struct * t = &per_cpu(init_tss, cpu);
++#endif
++ struct thread_struct *thread = ¤t->thread;
++ struct desc_struct *gdt;
++ struct Xgt_desc_struct *cpu_gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
++
++ if (cpu_test_and_set(cpu, cpu_initialized)) {
++ printk(KERN_WARNING "CPU#%d already initialized!\n", cpu);
++ for (;;) local_irq_enable();
++ }
++ printk(KERN_INFO "Initializing CPU#%d\n", cpu);
++
++ if (cpu_has_vme || cpu_has_de)
++ clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
++ if (tsc_disable && cpu_has_tsc) {
++ printk(KERN_NOTICE "Disabling TSC...\n");
++ /**** FIX-HPA: DOES THIS REALLY BELONG HERE? ****/
++ clear_bit(X86_FEATURE_TSC, boot_cpu_data.x86_capability);
++ set_in_cr4(X86_CR4_TSD);
++ }
++
++#ifndef CONFIG_XEN
++ /*
++ * This is a horrible hack to allocate the GDT. The problem
++ * is that cpu_init() is called really early for the boot CPU
++ * (and hence needs bootmem) but much later for the secondary
++ * CPUs, when bootmem will have gone away
++ */
++ if (NODE_DATA(0)->bdata->node_bootmem_map) {
++ gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE);
++ /* alloc_bootmem_pages panics on failure, so no check */
++ memset(gdt, 0, PAGE_SIZE);
++ } else {
++ gdt = (struct desc_struct *)get_zeroed_page(GFP_KERNEL);
++ if (unlikely(!gdt)) {
++ printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu);
++ for (;;)
++ local_irq_enable();
++ }
++ }
++
++ /*
++ * Initialize the per-CPU GDT with the boot GDT,
++ * and set up the GDT descriptor:
++ */
++ memcpy(gdt, cpu_gdt_table, GDT_SIZE);
++
++ /* Set up GDT entry for 16bit stack */
++ *(__u64 *)(&gdt[GDT_ENTRY_ESPFIX_SS]) |=
++ ((((__u64)stk16_off) << 16) & 0x000000ffffff0000ULL) |
++ ((((__u64)stk16_off) << 32) & 0xff00000000000000ULL) |
++ (CPU_16BIT_STACK_SIZE - 1);
++
++ cpu_gdt_descr->size = GDT_SIZE - 1;
++ cpu_gdt_descr->address = (unsigned long)gdt;
++#else
++ if (cpu == 0 && cpu_gdt_descr->address == 0) {
++ gdt = (struct desc_struct *)alloc_bootmem_pages(PAGE_SIZE);
++ /* alloc_bootmem_pages panics on failure, so no check */
++ memset(gdt, 0, PAGE_SIZE);
++
++ memcpy(gdt, cpu_gdt_table, GDT_SIZE);
++
++ cpu_gdt_descr->size = GDT_SIZE;
++ cpu_gdt_descr->address = (unsigned long)gdt;
++ }
++#endif
++
++ cpu_gdt_init(cpu_gdt_descr);
++
++ /*
++ * Set up and load the per-CPU TSS and LDT
++ */
++ atomic_inc(&init_mm.mm_count);
++ current->active_mm = &init_mm;
++ if (current->mm)
++ BUG();
++ enter_lazy_tlb(&init_mm, current);
++
++ load_esp0(t, thread);
++
++ load_LDT(&init_mm.context);
++
++#ifdef CONFIG_DOUBLEFAULT
++ /* Set up doublefault TSS pointer in the GDT */
++ __set_tss_desc(cpu, GDT_ENTRY_DOUBLEFAULT_TSS, &doublefault_tss);
++#endif
++
++ /* Clear %fs and %gs. */
++ asm volatile ("xorl %eax, %eax; movl %eax, %fs; movl %eax, %gs");
++
++ /* Clear all 6 debug registers: */
++ set_debugreg(0, 0);
++ set_debugreg(0, 1);
++ set_debugreg(0, 2);
++ set_debugreg(0, 3);
++ set_debugreg(0, 6);
++ set_debugreg(0, 7);
++
++ /*
++ * Force FPU initialization:
++ */
++ current_thread_info()->status = 0;
++ clear_used_math();
++ mxcsr_feature_mask_init();
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++void __cpuinit cpu_uninit(void)
++{
++ int cpu = raw_smp_processor_id();
++ cpu_clear(cpu, cpu_initialized);
++
++ /* lazy TLB state */
++ per_cpu(cpu_tlbstate, cpu).state = 0;
++ per_cpu(cpu_tlbstate, cpu).active_mm = &init_mm;
++}
++#endif
+diff -urNp linux-2.6/arch/i386/kernel/cpu/Makefile new/arch/i386/kernel/cpu/Makefile
+--- linux-2.6/arch/i386/kernel/cpu/Makefile 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/cpu/Makefile 2006-05-09 12:32:34.000000000 +0200
+@@ -17,3 +17,8 @@ obj-$(CONFIG_X86_MCE) += mcheck/
+
+ obj-$(CONFIG_MTRR) += mtrr/
+ obj-$(CONFIG_CPU_FREQ) += cpufreq/
++
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++obj-y := $(call cherrypickxen, $(obj-y), $(src))
++endif
+diff -urNp linux-2.6/arch/i386/kernel/cpu/mtrr/main-xen.c new/arch/i386/kernel/cpu/mtrr/main-xen.c
+--- linux-2.6/arch/i386/kernel/cpu/mtrr/main-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/cpu/mtrr/main-xen.c 2006-05-09 12:32:34.000000000 +0200
+@@ -0,0 +1,197 @@
++#include <linux/init.h>
++#include <linux/proc_fs.h>
++#include <linux/ctype.h>
++#include <linux/module.h>
++#include <linux/seq_file.h>
++#include <linux/mutex.h>
++#include <asm/uaccess.h>
++
++#include <asm/mtrr.h>
++#include "mtrr.h"
++
++static DEFINE_MUTEX(mtrr_mutex);
++
++void generic_get_mtrr(unsigned int reg, unsigned long *base,
++ unsigned int *size, mtrr_type * type)
++{
++ dom0_op_t op;
++
++ op.cmd = DOM0_READ_MEMTYPE;
++ op.u.read_memtype.reg = reg;
++ (void)HYPERVISOR_dom0_op(&op);
++
++ *size = op.u.read_memtype.nr_mfns;
++ *base = op.u.read_memtype.mfn;
++ *type = op.u.read_memtype.type;
++}
++
++struct mtrr_ops generic_mtrr_ops = {
++ .use_intel_if = 1,
++ .get = generic_get_mtrr,
++};
++
++struct mtrr_ops *mtrr_if = &generic_mtrr_ops;
++unsigned int num_var_ranges;
++unsigned int *usage_table;
++
++static void __init set_num_var_ranges(void)
++{
++ dom0_op_t op;
++
++ for (num_var_ranges = 0; ; num_var_ranges++) {
++ op.cmd = DOM0_READ_MEMTYPE;
++ op.u.read_memtype.reg = num_var_ranges;
++ if (HYPERVISOR_dom0_op(&op) != 0)
++ break;
++ }
++}
++
++static void __init init_table(void)
++{
++ int i, max;
++
++ max = num_var_ranges;
++ if ((usage_table = kmalloc(max * sizeof *usage_table, GFP_KERNEL))
++ == NULL) {
++ printk(KERN_ERR "mtrr: could not allocate\n");
++ return;
++ }
++ for (i = 0; i < max; i++)
++ usage_table[i] = 0;
++}
++
++int mtrr_add_page(unsigned long base, unsigned long size,
++ unsigned int type, char increment)
++{
++ int error;
++ dom0_op_t op;
++
++ mutex_lock(&mtrr_mutex);
++
++ op.cmd = DOM0_ADD_MEMTYPE;
++ op.u.add_memtype.mfn = base;
++ op.u.add_memtype.nr_mfns = size;
++ op.u.add_memtype.type = type;
++ error = HYPERVISOR_dom0_op(&op);
++ if (error) {
++ mutex_unlock(&mtrr_mutex);
++ BUG_ON(error > 0);
++ return error;
++ }
++
++ if (increment)
++ ++usage_table[op.u.add_memtype.reg];
++
++ mutex_unlock(&mtrr_mutex);
++
++ return op.u.add_memtype.reg;
++}
++
++static int mtrr_check(unsigned long base, unsigned long size)
++{
++ if ((base & (PAGE_SIZE - 1)) || (size & (PAGE_SIZE - 1))) {
++ printk(KERN_WARNING
++ "mtrr: size and base must be multiples of 4 kiB\n");
++ printk(KERN_DEBUG
++ "mtrr: size: 0x%lx base: 0x%lx\n", size, base);
++ dump_stack();
++ return -1;
++ }
++ return 0;
++}
++
++int
++mtrr_add(unsigned long base, unsigned long size, unsigned int type,
++ char increment)
++{
++ if (mtrr_check(base, size))
++ return -EINVAL;
++ return mtrr_add_page(base >> PAGE_SHIFT, size >> PAGE_SHIFT, type,
++ increment);
++}
++
++int mtrr_del_page(int reg, unsigned long base, unsigned long size)
++{
++ unsigned i;
++ mtrr_type ltype;
++ unsigned long lbase;
++ unsigned int lsize;
++ int error = -EINVAL;
++ dom0_op_t op;
++
++ mutex_lock(&mtrr_mutex);
++
++ if (reg < 0) {
++ /* Search for existing MTRR */
++ for (i = 0; i < num_var_ranges; ++i) {
++ mtrr_if->get(i, &lbase, &lsize, <ype);
++ if (lbase == base && lsize == size) {
++ reg = i;
++ break;
++ }
++ }
++ if (reg < 0) {
++ printk(KERN_DEBUG "mtrr: no MTRR for %lx000,%lx000 found\n", base,
++ size);
++ goto out;
++ }
++ }
++ if (usage_table[reg] < 1) {
++ printk(KERN_WARNING "mtrr: reg: %d has count=0\n", reg);
++ goto out;
++ }
++ if (--usage_table[reg] < 1) {
++ op.cmd = DOM0_DEL_MEMTYPE;
++ op.u.del_memtype.handle = 0;
++ op.u.del_memtype.reg = reg;
++ error = HYPERVISOR_dom0_op(&op);
++ if (error) {
++ BUG_ON(error > 0);
++ goto out;
++ }
++ }
++ error = reg;
++ out:
++ mutex_unlock(&mtrr_mutex);
++ return error;
++}
++
++int
++mtrr_del(int reg, unsigned long base, unsigned long size)
++{
++ if (mtrr_check(base, size))
++ return -EINVAL;
++ return mtrr_del_page(reg, base >> PAGE_SHIFT, size >> PAGE_SHIFT);
++}
++
++EXPORT_SYMBOL(mtrr_add);
++EXPORT_SYMBOL(mtrr_del);
++
++void __init mtrr_bp_init(void)
++{
++}
++
++void mtrr_ap_init(void)
++{
++}
++
++static int __init mtrr_init(void)
++{
++ struct cpuinfo_x86 *c = &boot_cpu_data;
++
++ if (!(xen_start_info->flags & SIF_PRIVILEGED))
++ return -ENODEV;
++
++ if ((!cpu_has(c, X86_FEATURE_MTRR)) &&
++ (!cpu_has(c, X86_FEATURE_K6_MTRR)) &&
++ (!cpu_has(c, X86_FEATURE_CYRIX_ARR)) &&
++ (!cpu_has(c, X86_FEATURE_CENTAUR_MCR)))
++ return -ENODEV;
++
++ set_num_var_ranges();
++ init_table();
++
++ return 0;
++}
++
++subsys_initcall(mtrr_init);
+diff -urNp linux-2.6/arch/i386/kernel/cpu/mtrr/Makefile new/arch/i386/kernel/cpu/mtrr/Makefile
+--- linux-2.6/arch/i386/kernel/cpu/mtrr/Makefile 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/cpu/mtrr/Makefile 2006-05-09 12:32:34.000000000 +0200
+@@ -3,3 +3,10 @@ obj-y += amd.o
+ obj-y += cyrix.o
+ obj-y += centaur.o
+
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++n-obj-xen := generic.o state.o amd.o cyrix.o centaur.o
++
++obj-y := $(call filterxen, $(obj-y), $(n-obj-xen))
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/arch/i386/kernel/early_printk-xen.c new/arch/i386/kernel/early_printk-xen.c
+--- linux-2.6/arch/i386/kernel/early_printk-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/early_printk-xen.c 2006-05-09 12:32:34.000000000 +0200
+@@ -0,0 +1,2 @@
++
++#include "../../x86_64/kernel/early_printk-xen.c"
+diff -urNp linux-2.6/arch/i386/kernel/entry.S new/arch/i386/kernel/entry.S
+--- linux-2.6/arch/i386/kernel/entry.S 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/entry.S 2006-05-09 12:32:34.000000000 +0200
+@@ -177,7 +177,7 @@ need_resched:
+
+ # sysenter call handler stub
+ ENTRY(sysenter_entry)
+- movl TSS_sysenter_esp0(%esp),%esp
++ movl SYSENTER_stack_esp0(%esp),%esp
+ sysenter_past_esp:
+ sti
+ pushl $(__USER_DS)
+@@ -410,7 +410,7 @@ vector=0
+ ENTRY(irq_entries_start)
+ .rept NR_IRQS
+ ALIGN
+-1: pushl $vector-256
++1: pushl $~(vector)
+ jmp common_interrupt
+ .data
+ .long 1b
+@@ -427,7 +427,7 @@ common_interrupt:
+
+ #define BUILD_INTERRUPT(name, nr) \
+ ENTRY(name) \
+- pushl $nr-256; \
++ pushl $~(nr); \
+ SAVE_ALL \
+ movl %esp,%eax; \
+ call smp_/**/name; \
+@@ -496,7 +496,7 @@ device_not_available_emulate:
+ * that sets up the real kernel stack. Check here, since we can't
+ * allow the wrong stack to be used.
+ *
+- * "TSS_sysenter_esp0+12" is because the NMI/debug handler will have
++ * "SYSENTER_stack_esp0+12" is because the NMI/debug handler will have
+ * already pushed 3 words if it hits on the sysenter instruction:
+ * eflags, cs and eip.
+ *
+@@ -508,7 +508,7 @@ device_not_available_emulate:
+ cmpw $__KERNEL_CS,4(%esp); \
+ jne ok; \
+ label: \
+- movl TSS_sysenter_esp0+offset(%esp),%esp; \
++ movl SYSENTER_stack_esp0+offset(%esp),%esp; \
+ pushfl; \
+ pushl $__KERNEL_CS; \
+ pushl $sysenter_past_esp
+diff -urNp linux-2.6/arch/i386/kernel/entry-xen.S new/arch/i386/kernel/entry-xen.S
+--- linux-2.6/arch/i386/kernel/entry-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/entry-xen.S 2006-05-09 12:32:34.000000000 +0200
+@@ -0,0 +1,903 @@
++/*
++ * linux/arch/i386/entry.S
++ *
++ * Copyright (C) 1991, 1992 Linus Torvalds
++ */
++
++/*
++ * entry.S contains the system-call and fault low-level handling routines.
++ * This also contains the timer-interrupt handler, as well as all interrupts
++ * and faults that can result in a task-switch.
++ *
++ * NOTE: This code handles signal-recognition, which happens every time
++ * after a timer-interrupt and after each system call.
++ *
++ * I changed all the .align's to 4 (16 byte alignment), as that's faster
++ * on a 486.
++ *
++ * Stack layout in 'ret_from_system_call':
++ * ptrace needs to have all regs on the stack.
++ * if the order here is changed, it needs to be
++ * updated in fork.c:copy_process, signal.c:do_signal,
++ * ptrace.c and ptrace.h
++ *
++ * 0(%esp) - %ebx
++ * 4(%esp) - %ecx
++ * 8(%esp) - %edx
++ * C(%esp) - %esi
++ * 10(%esp) - %edi
++ * 14(%esp) - %ebp
++ * 18(%esp) - %eax
++ * 1C(%esp) - %ds
++ * 20(%esp) - %es
++ * 24(%esp) - orig_eax
++ * 28(%esp) - %eip
++ * 2C(%esp) - %cs
++ * 30(%esp) - %eflags
++ * 34(%esp) - %oldesp
++ * 38(%esp) - %oldss
++ *
++ * "current" is in register %ebx during any slow entries.
++ */
++
++#include <linux/config.h>
++#include <linux/linkage.h>
++#include <asm/thread_info.h>
++#include <asm/errno.h>
++#include <asm/segment.h>
++#include <asm/smp.h>
++#include <asm/page.h>
++#include <asm/desc.h>
++#include "irq_vectors.h"
++#include <xen/interface/xen.h>
++
++#define nr_syscalls ((syscall_table_size)/4)
++
++EBX = 0x00
++ECX = 0x04
++EDX = 0x08
++ESI = 0x0C
++EDI = 0x10
++EBP = 0x14
++EAX = 0x18
++DS = 0x1C
++ES = 0x20
++ORIG_EAX = 0x24
++EIP = 0x28
++CS = 0x2C
++EFLAGS = 0x30
++OLDESP = 0x34
++OLDSS = 0x38
++
++CF_MASK = 0x00000001
++TF_MASK = 0x00000100
++IF_MASK = 0x00000200
++DF_MASK = 0x00000400
++NT_MASK = 0x00004000
++VM_MASK = 0x00020000
++/* Pseudo-eflags. */
++NMI_MASK = 0x80000000
++
++#ifndef CONFIG_XEN
++#define DISABLE_INTERRUPTS cli
++#define ENABLE_INTERRUPTS sti
++#else
++/* Offsets into shared_info_t. */
++#define evtchn_upcall_pending /* 0 */
++#define evtchn_upcall_mask 1
++
++#define sizeof_vcpu_shift 6
++
++#ifdef CONFIG_SMP
++#define GET_VCPU_INFO movl TI_cpu(%ebp),%esi ; \
++ shl $sizeof_vcpu_shift,%esi ; \
++ addl HYPERVISOR_shared_info,%esi
++#else
++#define GET_VCPU_INFO movl HYPERVISOR_shared_info,%esi
++#endif
++
++#define __DISABLE_INTERRUPTS movb $1,evtchn_upcall_mask(%esi)
++#define __ENABLE_INTERRUPTS movb $0,evtchn_upcall_mask(%esi)
++#define DISABLE_INTERRUPTS GET_VCPU_INFO ; \
++ __DISABLE_INTERRUPTS
++#define ENABLE_INTERRUPTS GET_VCPU_INFO ; \
++ __ENABLE_INTERRUPTS
++#define __TEST_PENDING testb $0xFF,evtchn_upcall_pending(%esi)
++#endif
++
++#ifdef CONFIG_PREEMPT
++#define preempt_stop cli
++#else
++#define preempt_stop
++#define resume_kernel restore_nocheck
++#endif
++
++#define SAVE_ALL \
++ cld; \
++ pushl %es; \
++ pushl %ds; \
++ pushl %eax; \
++ pushl %ebp; \
++ pushl %edi; \
++ pushl %esi; \
++ pushl %edx; \
++ pushl %ecx; \
++ pushl %ebx; \
++ movl $(__USER_DS), %edx; \
++ movl %edx, %ds; \
++ movl %edx, %es;
++
++#define RESTORE_INT_REGS \
++ popl %ebx; \
++ popl %ecx; \
++ popl %edx; \
++ popl %esi; \
++ popl %edi; \
++ popl %ebp; \
++ popl %eax
++
++#define RESTORE_REGS \
++ RESTORE_INT_REGS; \
++1: popl %ds; \
++2: popl %es; \
++.section .fixup,"ax"; \
++3: movl $0,(%esp); \
++ jmp 1b; \
++4: movl $0,(%esp); \
++ jmp 2b; \
++.previous; \
++.section __ex_table,"a";\
++ .align 4; \
++ .long 1b,3b; \
++ .long 2b,4b; \
++.previous
++
++
++ENTRY(ret_from_fork)
++ pushl %eax
++ call schedule_tail
++ GET_THREAD_INFO(%ebp)
++ popl %eax
++ jmp syscall_exit
++
++/*
++ * Return to user mode is not as complex as all this looks,
++ * but we want the default path for a system call return to
++ * go as quickly as possible which is why some of this is
++ * less clear than it otherwise should be.
++ */
++
++ # userspace resumption stub bypassing syscall exit tracing
++ ALIGN
++ret_from_exception:
++ preempt_stop
++ret_from_intr:
++ GET_THREAD_INFO(%ebp)
++ movl EFLAGS(%esp), %eax # mix EFLAGS and CS
++ movb CS(%esp), %al
++ testl $(VM_MASK | 2), %eax
++ jz resume_kernel
++ENTRY(resume_userspace)
++ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
++ # setting need_resched or sigpending
++ # between sampling and the iret
++ movl TI_flags(%ebp), %ecx
++ andl $_TIF_WORK_MASK, %ecx # is there any work to be done on
++ # int/exception return?
++ jne work_pending
++ jmp restore_all
++
++#ifdef CONFIG_PREEMPT
++ENTRY(resume_kernel)
++ cli
++ cmpl $0,TI_preempt_count(%ebp) # non-zero preempt_count ?
++ jnz restore_nocheck
++need_resched:
++ movl TI_flags(%ebp), %ecx # need_resched set ?
++ testb $_TIF_NEED_RESCHED, %cl
++ jz restore_all
++ testl $IF_MASK,EFLAGS(%esp) # interrupts off (exception path) ?
++ jz restore_all
++ call preempt_schedule_irq
++ jmp need_resched
++#endif
++
++/* SYSENTER_RETURN points to after the "sysenter" instruction in
++ the vsyscall page. See vsyscall-sysentry.S, which defines the symbol. */
++
++ # sysenter call handler stub
++ENTRY(sysenter_entry)
++ movl SYSENTER_stack_esp0(%esp),%esp
++sysenter_past_esp:
++ sti
++ pushl $(__USER_DS)
++ pushl %ebp
++ pushfl
++ pushl $(__USER_CS)
++ pushl $SYSENTER_RETURN
++
++/*
++ * Load the potential sixth argument from user stack.
++ * Careful about security.
++ */
++ cmpl $__PAGE_OFFSET-3,%ebp
++ jae syscall_fault
++1: movl (%ebp),%ebp
++.section __ex_table,"a"
++ .align 4
++ .long 1b,syscall_fault
++.previous
++
++ pushl %eax
++ SAVE_ALL
++ GET_THREAD_INFO(%ebp)
++
++ /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
++ testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
++ jnz syscall_trace_entry
++ cmpl $(nr_syscalls), %eax
++ jae syscall_badsys
++ call *sys_call_table(,%eax,4)
++ movl %eax,EAX(%esp)
++ DISABLE_INTERRUPTS
++ movl TI_flags(%ebp), %ecx
++ testw $_TIF_ALLWORK_MASK, %cx
++ jne syscall_exit_work
++/* if something modifies registers it must also disable sysexit */
++ movl EIP(%esp), %edx
++ movl OLDESP(%esp), %ecx
++ xorl %ebp,%ebp
++#ifdef CONFIG_XEN
++ __ENABLE_INTERRUPTS
++sysexit_scrit: /**** START OF SYSEXIT CRITICAL REGION ****/
++ __TEST_PENDING
++ jnz 14f # process more events if necessary...
++ movl ESI(%esp), %esi
++ sysexit
++14: __DISABLE_INTERRUPTS
++sysexit_ecrit: /**** END OF SYSEXIT CRITICAL REGION ****/
++ push %esp
++ call evtchn_do_upcall
++ add $4,%esp
++ jmp ret_from_intr
++#else
++ sti
++ sysexit
++#endif /* !CONFIG_XEN */
++
++
++ # system call handler stub
++ENTRY(system_call)
++ pushl %eax # save orig_eax
++ SAVE_ALL
++ GET_THREAD_INFO(%ebp)
++ testl $TF_MASK,EFLAGS(%esp)
++ jz no_singlestep
++ orl $_TIF_SINGLESTEP,TI_flags(%ebp)
++no_singlestep:
++ # system call tracing in operation / emulation
++ /* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
++ testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
++ jnz syscall_trace_entry
++ cmpl $(nr_syscalls), %eax
++ jae syscall_badsys
++syscall_call:
++ call *sys_call_table(,%eax,4)
++ movl %eax,EAX(%esp) # store the return value
++syscall_exit:
++ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
++ # setting need_resched or sigpending
++ # between sampling and the iret
++ movl TI_flags(%ebp), %ecx
++ testw $_TIF_ALLWORK_MASK, %cx # current->work
++ jne syscall_exit_work
++
++restore_all:
++#ifndef CONFIG_XEN
++ movl EFLAGS(%esp), %eax # mix EFLAGS, SS and CS
++ # Warning: OLDSS(%esp) contains the wrong/random values if we
++ # are returning to the kernel.
++ # See comments in process.c:copy_thread() for details.
++ movb OLDSS(%esp), %ah
++ movb CS(%esp), %al
++ andl $(VM_MASK | (4 << 8) | 3), %eax
++ cmpl $((4 << 8) | 3), %eax
++ je ldt_ss # returning to user-space with LDT SS
++restore_nocheck:
++#else
++restore_nocheck:
++ movl EFLAGS(%esp), %eax
++ testl $(VM_MASK|NMI_MASK), %eax
++ jnz hypervisor_iret
++ shr $9, %eax # EAX[0] == IRET_EFLAGS.IF
++ GET_VCPU_INFO
++ andb evtchn_upcall_mask(%esi),%al
++ andb $1,%al # EAX[0] == IRET_EFLAGS.IF & event_mask
++ jnz restore_all_enable_events # != 0 => enable event delivery
++#endif
++ RESTORE_REGS
++ addl $4, %esp
++1: iret
++.section .fixup,"ax"
++iret_exc:
++#ifndef CONFIG_XEN
++ sti
++#endif
++ pushl $0 # no error code
++ pushl $do_iret_error
++ jmp error_code
++.previous
++.section __ex_table,"a"
++ .align 4
++ .long 1b,iret_exc
++.previous
++
++#ifndef CONFIG_XEN
++ldt_ss:
++ larl OLDSS(%esp), %eax
++ jnz restore_nocheck
++ testl $0x00400000, %eax # returning to 32bit stack?
++ jnz restore_nocheck # allright, normal return
++ /* If returning to userspace with 16bit stack,
++ * try to fix the higher word of ESP, as the CPU
++ * won't restore it.
++ * This is an "official" bug of all the x86-compatible
++ * CPUs, which we can try to work around to make
++ * dosemu and wine happy. */
++ subl $8, %esp # reserve space for switch16 pointer
++ cli
++ movl %esp, %eax
++ /* Set up the 16bit stack frame with switch32 pointer on top,
++ * and a switch16 pointer on top of the current frame. */
++ call setup_x86_bogus_stack
++ RESTORE_REGS
++ lss 20+4(%esp), %esp # switch to 16bit stack
++1: iret
++.section __ex_table,"a"
++ .align 4
++ .long 1b,iret_exc
++.previous
++#else
++hypervisor_iret:
++ andl $~NMI_MASK, EFLAGS(%esp)
++ RESTORE_REGS
++ addl $4, %esp
++ jmp hypercall_page + (__HYPERVISOR_iret * 32)
++#endif
++
++ # perform work that needs to be done immediately before resumption
++ ALIGN
++work_pending:
++ testb $_TIF_NEED_RESCHED, %cl
++ jz work_notifysig
++work_resched:
++ call schedule
++ DISABLE_INTERRUPTS # make sure we don't miss an interrupt
++ # setting need_resched or sigpending
++ # between sampling and the iret
++ movl TI_flags(%ebp), %ecx
++ andl $_TIF_WORK_MASK, %ecx # is there any work to be done other
++ # than syscall tracing?
++ jz restore_all
++ testb $_TIF_NEED_RESCHED, %cl
++ jnz work_resched
++
++work_notifysig: # deal with pending signals and
++ # notify-resume requests
++ testl $VM_MASK, EFLAGS(%esp)
++ movl %esp, %eax
++ jne work_notifysig_v86 # returning to kernel-space or
++ # vm86-space
++ xorl %edx, %edx
++ call do_notify_resume
++ jmp resume_userspace
++
++ ALIGN
++work_notifysig_v86:
++#ifdef CONFIG_VM86
++ pushl %ecx # save ti_flags for do_notify_resume
++ call save_v86_state # %eax contains pt_regs pointer
++ popl %ecx
++ movl %eax, %esp
++ xorl %edx, %edx
++ call do_notify_resume
++ jmp resume_userspace
++#endif
++
++ # perform syscall exit tracing
++ ALIGN
++syscall_trace_entry:
++ movl $-ENOSYS,EAX(%esp)
++ movl %esp, %eax
++ xorl %edx,%edx
++ call do_syscall_trace
++ cmpl $0, %eax
++ jne resume_userspace # ret != 0 -> running under PTRACE_SYSEMU,
++ # so must skip actual syscall
++ movl ORIG_EAX(%esp), %eax
++ cmpl $(nr_syscalls), %eax
++ jnae syscall_call
++ jmp syscall_exit
++
++ # perform syscall exit tracing
++ ALIGN
++syscall_exit_work:
++ testb $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP), %cl
++ jz work_pending
++ ENABLE_INTERRUPTS # could let do_syscall_trace() call
++ # schedule() instead
++ movl %esp, %eax
++ movl $1, %edx
++ call do_syscall_trace
++ jmp resume_userspace
++
++ ALIGN
++syscall_fault:
++ pushl %eax # save orig_eax
++ SAVE_ALL
++ GET_THREAD_INFO(%ebp)
++ movl $-EFAULT,EAX(%esp)
++ jmp resume_userspace
++
++ ALIGN
++syscall_badsys:
++ movl $-ENOSYS,EAX(%esp)
++ jmp resume_userspace
++
++#ifndef CONFIG_XEN
++#define FIXUP_ESPFIX_STACK \
++ movl %esp, %eax; \
++ /* switch to 32bit stack using the pointer on top of 16bit stack */ \
++ lss %ss:CPU_16BIT_STACK_SIZE-8, %esp; \
++ /* copy data from 16bit stack to 32bit stack */ \
++ call fixup_x86_bogus_stack; \
++ /* put ESP to the proper location */ \
++ movl %eax, %esp;
++#define UNWIND_ESPFIX_STACK \
++ pushl %eax; \
++ movl %ss, %eax; \
++ /* see if on 16bit stack */ \
++ cmpw $__ESPFIX_SS, %ax; \
++ jne 28f; \
++ movl $__KERNEL_DS, %edx; \
++ movl %edx, %ds; \
++ movl %edx, %es; \
++ /* switch to 32bit stack */ \
++ FIXUP_ESPFIX_STACK \
++28: popl %eax;
++
++/*
++ * Build the entry stubs and pointer table with
++ * some assembler magic.
++ */
++.data
++ENTRY(interrupt)
++.text
++
++vector=0
++ENTRY(irq_entries_start)
++.rept NR_IRQS
++ ALIGN
++1: pushl $~(vector)
++ jmp common_interrupt
++.data
++ .long 1b
++.text
++vector=vector+1
++.endr
++
++ ALIGN
++common_interrupt:
++ SAVE_ALL
++ movl %esp,%eax
++ call do_IRQ
++ jmp ret_from_intr
++
++#define BUILD_INTERRUPT(name, nr) \
++ENTRY(name) \
++ pushl $~(nr); \
++ SAVE_ALL \
++ movl %esp,%eax; \
++ call smp_/**/name; \
++ jmp ret_from_intr;
++
++/* The include is where all of the SMP etc. interrupts come from */
++#include "entry_arch.h"
++#else
++#define UNWIND_ESPFIX_STACK
++#endif
++
++ENTRY(divide_error)
++ pushl $0 # no error code
++ pushl $do_divide_error
++ ALIGN
++error_code:
++ pushl %ds
++ pushl %eax
++ xorl %eax, %eax
++ pushl %ebp
++ pushl %edi
++ pushl %esi
++ pushl %edx
++ decl %eax # eax = -1
++ pushl %ecx
++ pushl %ebx
++ cld
++ pushl %es
++ UNWIND_ESPFIX_STACK
++ popl %ecx
++ movl ES(%esp), %edi # get the function address
++ movl ORIG_EAX(%esp), %edx # get the error code
++ movl %eax, ORIG_EAX(%esp)
++ movl %ecx, ES(%esp)
++ movl $(__USER_DS), %ecx
++ movl %ecx, %ds
++ movl %ecx, %es
++ movl %esp,%eax # pt_regs pointer
++ call *%edi
++ jmp ret_from_exception
++
++#ifdef CONFIG_XEN
++# A note on the "critical region" in our callback handler.
++# We want to avoid stacking callback handlers due to events occurring
++# during handling of the last event. To do this, we keep events disabled
++# until we've done all processing. HOWEVER, we must enable events before
++# popping the stack frame (can't be done atomically) and so it would still
++# be possible to get enough handler activations to overflow the stack.
++# Although unlikely, bugs of that kind are hard to track down, so we'd
++# like to avoid the possibility.
++# So, on entry to the handler we detect whether we interrupted an
++# existing activation in its critical region -- if so, we pop the current
++# activation and restart the handler using the previous one.
++#
++# The sysexit critical region is slightly different. sysexit
++# atomically removes the entire stack frame. If we interrupt in the
++# critical region we know that the entire frame is present and correct
++# so we can simply throw away the new one.
++ENTRY(hypervisor_callback)
++ pushl %eax
++ SAVE_ALL
++ movl EIP(%esp),%eax
++ cmpl $scrit,%eax
++ jb 11f
++ cmpl $ecrit,%eax
++ jb critical_region_fixup
++ cmpl $sysexit_scrit,%eax
++ jb 11f
++ cmpl $sysexit_ecrit,%eax
++ ja 11f
++ addl $0x34,%esp # Remove cs...ebx from stack frame.
++11: push %esp
++ call evtchn_do_upcall
++ add $4,%esp
++ jmp ret_from_intr
++
++ ALIGN
++restore_all_enable_events:
++ __ENABLE_INTERRUPTS
++scrit: /**** START OF CRITICAL REGION ****/
++ __TEST_PENDING
++ jnz 14f # process more events if necessary...
++ RESTORE_REGS
++ addl $4, %esp
++1: iret
++.section __ex_table,"a"
++ .align 4
++ .long 1b,iret_exc
++.previous
++14: __DISABLE_INTERRUPTS
++ jmp 11b
++ecrit: /**** END OF CRITICAL REGION ****/
++# [How we do the fixup]. We want to merge the current stack frame with the
++# just-interrupted frame. How we do this depends on where in the critical
++# region the interrupted handler was executing, and so how many saved
++# registers are in each frame. We do this quickly using the lookup table
++# 'critical_fixup_table'. For each byte offset in the critical region, it
++# provides the number of bytes which have already been popped from the
++# interrupted stack frame.
++critical_region_fixup:
++ addl $critical_fixup_table-scrit,%eax
++ movzbl (%eax),%eax # %eax contains num bytes popped
++ cmpb $0xff,%al # 0xff => vcpu_info critical region
++ jne 15f
++ GET_THREAD_INFO(%ebp)
++ xorl %eax,%eax
++15: mov %esp,%esi
++ add %eax,%esi # %esi points at end of src region
++ mov %esp,%edi
++ add $0x34,%edi # %edi points at end of dst region
++ mov %eax,%ecx
++ shr $2,%ecx # convert words to bytes
++ je 17f # skip loop if nothing to copy
++16: subl $4,%esi # pre-decrementing copy loop
++ subl $4,%edi
++ movl (%esi),%eax
++ movl %eax,(%edi)
++ loop 16b
++17: movl %edi,%esp # final %edi is top of merged stack
++ jmp 11b
++
++critical_fixup_table:
++ .byte 0xff,0xff,0xff # testb $0xff,(%esi) = __TEST_PENDING
++ .byte 0xff,0xff # jnz 14f
++ .byte 0x00 # pop %ebx
++ .byte 0x04 # pop %ecx
++ .byte 0x08 # pop %edx
++ .byte 0x0c # pop %esi
++ .byte 0x10 # pop %edi
++ .byte 0x14 # pop %ebp
++ .byte 0x18 # pop %eax
++ .byte 0x1c # pop %ds
++ .byte 0x20 # pop %es
++ .byte 0x24,0x24,0x24 # add $4,%esp
++ .byte 0x28 # iret
++ .byte 0xff,0xff,0xff,0xff # movb $1,1(%esi)
++ .byte 0x00,0x00 # jmp 11b
++
++# Hypervisor uses this for application faults while it executes.
++# We get here for two reasons:
++# 1. Fault while reloading DS, ES, FS or GS
++# 2. Fault while executing IRET
++# Category 1 we fix up by reattempting the load, and zeroing the segment
++# register if the load fails.
++# Category 2 we fix up by jumping to do_iret_error. We cannot use the
++# normal Linux return path in this case because if we use the IRET hypercall
++# to pop the stack frame we end up in an infinite loop of failsafe callbacks.
++# We distinguish between categories by maintaining a status value in EAX.
++ENTRY(failsafe_callback)
++ pushl %eax
++ movl $1,%eax
++1: mov 4(%esp),%ds
++2: mov 8(%esp),%es
++3: mov 12(%esp),%fs
++4: mov 16(%esp),%gs
++ testl %eax,%eax
++ popl %eax
++ jz 5f
++ addl $16,%esp # EAX != 0 => Category 2 (Bad IRET)
++ jmp iret_exc
++5: addl $16,%esp # EAX == 0 => Category 1 (Bad segment)
++ pushl $0
++ SAVE_ALL
++ jmp ret_from_exception
++.section .fixup,"ax"; \
++6: xorl %eax,%eax; \
++ movl %eax,4(%esp); \
++ jmp 1b; \
++7: xorl %eax,%eax; \
++ movl %eax,8(%esp); \
++ jmp 2b; \
++8: xorl %eax,%eax; \
++ movl %eax,12(%esp); \
++ jmp 3b; \
++9: xorl %eax,%eax; \
++ movl %eax,16(%esp); \
++ jmp 4b; \
++.previous; \
++.section __ex_table,"a"; \
++ .align 4; \
++ .long 1b,6b; \
++ .long 2b,7b; \
++ .long 3b,8b; \
++ .long 4b,9b; \
++.previous
++#endif
++
++ENTRY(coprocessor_error)
++ pushl $0
++ pushl $do_coprocessor_error
++ jmp error_code
++
++ENTRY(simd_coprocessor_error)
++ pushl $0
++ pushl $do_simd_coprocessor_error
++ jmp error_code
++
++ENTRY(device_not_available)
++ pushl $-1 # mark this as an int
++ SAVE_ALL
++#ifndef CONFIG_XEN
++ movl %cr0, %eax
++ testl $0x4, %eax # EM (math emulation bit)
++ je device_available_emulate
++ pushl $0 # temporary storage for ORIG_EIP
++ call math_emulate
++ addl $4, %esp
++ jmp ret_from_exception
++device_available_emulate:
++#endif
++ preempt_stop
++ call math_state_restore
++ jmp ret_from_exception
++
++#ifndef CONFIG_XEN
++/*
++ * Debug traps and NMI can happen at the one SYSENTER instruction
++ * that sets up the real kernel stack. Check here, since we can't
++ * allow the wrong stack to be used.
++ *
++ * "SYSENTER_stack_esp0+12" is because the NMI/debug handler will have
++ * already pushed 3 words if it hits on the sysenter instruction:
++ * eflags, cs and eip.
++ *
++ * We just load the right stack, and push the three (known) values
++ * by hand onto the new stack - while updating the return eip past
++ * the instruction that would have done it for sysenter.
++ */
++#define FIX_STACK(offset, ok, label) \
++ cmpw $__KERNEL_CS,4(%esp); \
++ jne ok; \
++label: \
++ movl SYSENTER_stack_esp0+offset(%esp),%esp; \
++ pushfl; \
++ pushl $__KERNEL_CS; \
++ pushl $sysenter_past_esp
++#endif /* CONFIG_XEN */
++
++KPROBE_ENTRY(debug)
++#ifndef CONFIG_XEN
++ cmpl $sysenter_entry,(%esp)
++ jne debug_stack_correct
++ FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
++debug_stack_correct:
++#endif /* !CONFIG_XEN */
++ pushl $-1 # mark this as an int
++ SAVE_ALL
++ xorl %edx,%edx # error code 0
++ movl %esp,%eax # pt_regs pointer
++ call do_debug
++ jmp ret_from_exception
++ .previous .text
++
++#ifndef CONFIG_XEN
++/*
++ * NMI is doubly nasty. It can happen _while_ we're handling
++ * a debug fault, and the debug fault hasn't yet been able to
++ * clear up the stack. So we first check whether we got an
++ * NMI on the sysenter entry path, but after that we need to
++ * check whether we got an NMI on the debug path where the debug
++ * fault happened on the sysenter path.
++ */
++ENTRY(nmi)
++ pushl %eax
++ movl %ss, %eax
++ cmpw $__ESPFIX_SS, %ax
++ popl %eax
++ je nmi_16bit_stack
++ cmpl $sysenter_entry,(%esp)
++ je nmi_stack_fixup
++ pushl %eax
++ movl %esp,%eax
++ /* Do not access memory above the end of our stack page,
++ * it might not exist.
++ */
++ andl $(THREAD_SIZE-1),%eax
++ cmpl $(THREAD_SIZE-20),%eax
++ popl %eax
++ jae nmi_stack_correct
++ cmpl $sysenter_entry,12(%esp)
++ je nmi_debug_stack_check
++nmi_stack_correct:
++ pushl %eax
++ SAVE_ALL
++ xorl %edx,%edx # zero error code
++ movl %esp,%eax # pt_regs pointer
++ call do_nmi
++ jmp restore_all
++
++nmi_stack_fixup:
++ FIX_STACK(12,nmi_stack_correct, 1)
++ jmp nmi_stack_correct
++nmi_debug_stack_check:
++ cmpw $__KERNEL_CS,16(%esp)
++ jne nmi_stack_correct
++ cmpl $debug,(%esp)
++ jb nmi_stack_correct
++ cmpl $debug_esp_fix_insn,(%esp)
++ ja nmi_stack_correct
++ FIX_STACK(24,nmi_stack_correct, 1)
++ jmp nmi_stack_correct
++
++nmi_16bit_stack:
++ /* create the pointer to lss back */
++ pushl %ss
++ pushl %esp
++ movzwl %sp, %esp
++ addw $4, (%esp)
++ /* copy the iret frame of 12 bytes */
++ .rept 3
++ pushl 16(%esp)
++ .endr
++ pushl %eax
++ SAVE_ALL
++ FIXUP_ESPFIX_STACK # %eax == %esp
++ xorl %edx,%edx # zero error code
++ call do_nmi
++ RESTORE_REGS
++ lss 12+4(%esp), %esp # back to 16bit stack
++1: iret
++.section __ex_table,"a"
++ .align 4
++ .long 1b,iret_exc
++.previous
++#else
++ENTRY(nmi)
++ pushl %eax
++ SAVE_ALL
++ xorl %edx,%edx # zero error code
++ movl %esp,%eax # pt_regs pointer
++ call do_nmi
++ orl $NMI_MASK, EFLAGS(%esp)
++ jmp restore_all
++#endif
++
++KPROBE_ENTRY(int3)
++ pushl $-1 # mark this as an int
++ SAVE_ALL
++ xorl %edx,%edx # zero error code
++ movl %esp,%eax # pt_regs pointer
++ call do_int3
++ jmp ret_from_exception
++ .previous .text
++
++ENTRY(overflow)
++ pushl $0
++ pushl $do_overflow
++ jmp error_code
++
++ENTRY(bounds)
++ pushl $0
++ pushl $do_bounds
++ jmp error_code
++
++ENTRY(invalid_op)
++ pushl $0
++ pushl $do_invalid_op
++ jmp error_code
++
++ENTRY(coprocessor_segment_overrun)
++ pushl $0
++ pushl $do_coprocessor_segment_overrun
++ jmp error_code
++
++ENTRY(invalid_TSS)
++ pushl $do_invalid_TSS
++ jmp error_code
++
++ENTRY(segment_not_present)
++ pushl $do_segment_not_present
++ jmp error_code
++
++ENTRY(stack_segment)
++ pushl $do_stack_segment
++ jmp error_code
++
++KPROBE_ENTRY(general_protection)
++ pushl $do_general_protection
++ jmp error_code
++ .previous .text
++
++ENTRY(alignment_check)
++ pushl $do_alignment_check
++ jmp error_code
++
++KPROBE_ENTRY(page_fault)
++ pushl $do_page_fault
++ jmp error_code
++ .previous .text
++
++#ifdef CONFIG_X86_MCE
++ENTRY(machine_check)
++ pushl $0
++ pushl machine_check_vector
++ jmp error_code
++#endif
++
++ENTRY(fixup_4gb_segment)
++ pushl $do_fixup_4gb_segment
++ jmp error_code
++
++.section .rodata,"a"
++#include "syscall_table.S"
++
++syscall_table_size=(.-sys_call_table)
+diff -urNp linux-2.6/arch/i386/kernel/fixup.c new/arch/i386/kernel/fixup.c
+--- linux-2.6/arch/i386/kernel/fixup.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/fixup.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,86 @@
++/******************************************************************************
++ * fixup.c
++ *
++ * Binary-rewriting of certain IA32 instructions, on notification by Xen.
++ * Used to avoid repeated slow emulation of common instructions used by the
++ * user-space TLS (Thread-Local Storage) libraries.
++ *
++ * **** NOTE ****
++ * Issues with the binary rewriting have caused it to be removed. Instead
++ * we rely on Xen's emulator to boot the kernel, and then print a banner
++ * message recommending that the user disables /lib/tls.
++ *
++ * Copyright (c) 2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++ */
++
++#include <linux/config.h>
++#include <linux/init.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/kernel.h>
++#include <linux/delay.h>
++#include <linux/version.h>
++
++#define DP(_f, _args...) printk(KERN_ALERT " " _f "\n" , ## _args )
++
++fastcall void do_fixup_4gb_segment(struct pt_regs *regs, long error_code)
++{
++ static unsigned long printed = 0;
++ char info[100];
++ int i;
++
++ if (test_and_set_bit(0, &printed))
++ return;
++
++ HYPERVISOR_vm_assist(
++ VMASST_CMD_disable, VMASST_TYPE_4gb_segments_notify);
++
++ sprintf(info, "%s (pid=%d)", current->comm, current->tgid);
++
++
++ DP("");
++ DP("***************************************************************");
++ DP("***************************************************************");
++ DP("** WARNING: Currently emulating unsupported memory accesses **");
++ DP("** in /lib/tls glibc libraries. The emulation is **");
++ DP("** slow. To ensure full performance you should **");
++ DP("** install a 'xen-friendly' (nosegneg) version of **");
++ DP("** the library, or disable tls support by executing **");
++ DP("** the following as root: **");
++ DP("** mv /lib/tls /lib/tls.disabled **");
++ DP("** Offending process: %-38.38s **", info);
++ DP("***************************************************************");
++ DP("***************************************************************");
++ DP("");
++
++ for (i = 5; i > 0; i--) {
++ touch_softlockup_watchdog();
++ printk("Pausing... %d", i);
++ mdelay(1000);
++ printk("\b\b\b\b\b\b\b\b\b\b\b\b");
++ }
++
++ printk("Continuing...\n\n");
++}
++
++static int __init fixup_init(void)
++{
++ HYPERVISOR_vm_assist(
++ VMASST_CMD_enable, VMASST_TYPE_4gb_segments_notify);
++ return 0;
++}
++__initcall(fixup_init);
+diff -urNp linux-2.6/arch/i386/kernel/head-xen.S new/arch/i386/kernel/head-xen.S
+--- linux-2.6/arch/i386/kernel/head-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/head-xen.S 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,181 @@
++
++
++.text
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <linux/linkage.h>
++#include <asm/segment.h>
++#include <asm/page.h>
++#include <asm/thread_info.h>
++#include <asm/asm-offsets.h>
++#include <xen/interface/arch-x86_32.h>
++
++/*
++ * References to members of the new_cpu_data structure.
++ */
++
++#define X86 new_cpu_data+CPUINFO_x86
++#define X86_VENDOR new_cpu_data+CPUINFO_x86_vendor
++#define X86_MODEL new_cpu_data+CPUINFO_x86_model
++#define X86_MASK new_cpu_data+CPUINFO_x86_mask
++#define X86_HARD_MATH new_cpu_data+CPUINFO_hard_math
++#define X86_CPUID new_cpu_data+CPUINFO_cpuid_level
++#define X86_CAPABILITY new_cpu_data+CPUINFO_x86_capability
++#define X86_VENDOR_ID new_cpu_data+CPUINFO_x86_vendor_id
++
++#define VIRT_ENTRY_OFFSET 0x0
++.org VIRT_ENTRY_OFFSET
++ENTRY(startup_32)
++ movl %esi,xen_start_info
++ cld
++
++ /* Set up the stack pointer */
++ movl $(init_thread_union+THREAD_SIZE),%esp
++
++ /* get vendor info */
++ xorl %eax,%eax # call CPUID with 0 -> return vendor ID
++ XEN_CPUID
++ movl %eax,X86_CPUID # save CPUID level
++ movl %ebx,X86_VENDOR_ID # lo 4 chars
++ movl %edx,X86_VENDOR_ID+4 # next 4 chars
++ movl %ecx,X86_VENDOR_ID+8 # last 4 chars
++
++ movl $1,%eax # Use the CPUID instruction to get CPU type
++ XEN_CPUID
++ movb %al,%cl # save reg for future use
++ andb $0x0f,%ah # mask processor family
++ movb %ah,X86
++ andb $0xf0,%al # mask model
++ shrb $4,%al
++ movb %al,X86_MODEL
++ andb $0x0f,%cl # mask mask revision
++ movb %cl,X86_MASK
++ movl %edx,X86_CAPABILITY
++
++ movb $1,X86_HARD_MATH
++
++ xorl %eax,%eax # Clear FS/GS and LDT
++ movl %eax,%fs
++ movl %eax,%gs
++ cld # gcc2 wants the direction flag cleared at all times
++
++ call start_kernel
++L6:
++ jmp L6 # main should never return here, but
++ # just in case, we know what happens.
++
++#define HYPERCALL_PAGE_OFFSET 0x1000
++.org HYPERCALL_PAGE_OFFSET
++ENTRY(hypercall_page)
++.skip 0x1000
++
++/*
++ * Real beginning of normal "text" segment
++ */
++ENTRY(stext)
++ENTRY(_stext)
++
++/*
++ * BSS section
++ */
++.section ".bss.page_aligned","w"
++ENTRY(empty_zero_page)
++ .fill 4096,1,0
++
++/*
++ * This starts the data section.
++ */
++.data
++
++/*
++ * The Global Descriptor Table contains 28 quadwords, per-CPU.
++ */
++ENTRY(cpu_gdt_table)
++ .quad 0x0000000000000000 /* NULL descriptor */
++ .quad 0x0000000000000000 /* 0x0b reserved */
++ .quad 0x0000000000000000 /* 0x13 reserved */
++ .quad 0x0000000000000000 /* 0x1b reserved */
++ .quad 0x0000000000000000 /* 0x20 unused */
++ .quad 0x0000000000000000 /* 0x28 unused */
++ .quad 0x0000000000000000 /* 0x33 TLS entry 1 */
++ .quad 0x0000000000000000 /* 0x3b TLS entry 2 */
++ .quad 0x0000000000000000 /* 0x43 TLS entry 3 */
++ .quad 0x0000000000000000 /* 0x4b reserved */
++ .quad 0x0000000000000000 /* 0x53 reserved */
++ .quad 0x0000000000000000 /* 0x5b reserved */
++
++ .quad 0x00cf9a000000ffff /* 0x60 kernel 4GB code at 0x00000000 */
++ .quad 0x00cf92000000ffff /* 0x68 kernel 4GB data at 0x00000000 */
++ .quad 0x00cffa000000ffff /* 0x73 user 4GB code at 0x00000000 */
++ .quad 0x00cff2000000ffff /* 0x7b user 4GB data at 0x00000000 */
++
++ .quad 0x0000000000000000 /* 0x80 TSS descriptor */
++ .quad 0x0000000000000000 /* 0x88 LDT descriptor */
++
++ /*
++ * Segments used for calling PnP BIOS have byte granularity.
++ * They code segments and data segments have fixed 64k limits,
++ * the transfer segment sizes are set at run time.
++ */
++ .quad 0x0000000000000000 /* 0x90 32-bit code */
++ .quad 0x0000000000000000 /* 0x98 16-bit code */
++ .quad 0x0000000000000000 /* 0xa0 16-bit data */
++ .quad 0x0000000000000000 /* 0xa8 16-bit data */
++ .quad 0x0000000000000000 /* 0xb0 16-bit data */
++
++ /*
++ * The APM segments have byte granularity and their bases
++ * are set at run time. All have 64k limits.
++ */
++ .quad 0x0000000000000000 /* 0xb8 APM CS code */
++ .quad 0x0000000000000000 /* 0xc0 APM CS 16 code (16 bit) */
++ .quad 0x0000000000000000 /* 0xc8 APM DS data */
++
++ .quad 0x0000000000000000 /* 0xd0 - ESPFIX 16-bit SS */
++ .quad 0x0000000000000000 /* 0xd8 - unused */
++ .quad 0x0000000000000000 /* 0xe0 - unused */
++ .quad 0x0000000000000000 /* 0xe8 - unused */
++ .quad 0x0000000000000000 /* 0xf0 - unused */
++ .quad 0x0000000000000000 /* 0xf8 - GDT entry 31: double-fault TSS */
++
++/*
++ * __xen_guest information
++ */
++.macro utoa value
++ .if (\value) < 0 || (\value) >= 0x10
++ utoa (((\value)>>4)&0x0fffffff)
++ .endif
++ .if ((\value) & 0xf) < 10
++ .byte '0' + ((\value) & 0xf)
++ .else
++ .byte 'A' + ((\value) & 0xf) - 10
++ .endif
++.endm
++
++.section __xen_guest
++ .ascii "GUEST_OS=linux,GUEST_VER=2.6"
++ .ascii ",XEN_VER=xen-3.0"
++ .ascii ",VIRT_BASE=0x"
++ utoa __PAGE_OFFSET
++#ifdef CONFIG_XEN_COMPAT_030002
++ .ascii ",ELF_PADDR_OFFSET=0x"
++ utoa __PAGE_OFFSET
++#else
++ .ascii ",ELF_PADDR_OFFSET=0x0"
++#endif /* !CONFIG_XEN_COMPAT_030002 */
++ .ascii ",VIRT_ENTRY=0x"
++ utoa (__PAGE_OFFSET + __PHYSICAL_START + VIRT_ENTRY_OFFSET)
++ .ascii ",HYPERCALL_PAGE=0x"
++ utoa ((__PHYSICAL_START+HYPERCALL_PAGE_OFFSET)>>PAGE_SHIFT)
++ .ascii ",FEATURES=writable_page_tables"
++ .ascii "|writable_descriptor_tables"
++ .ascii "|auto_translated_physmap"
++ .ascii "|pae_pgdir_above_4gb"
++ .ascii "|supervisor_mode_kernel"
++#ifdef CONFIG_X86_PAE
++ .ascii ",PAE=yes[extended-cr3]"
++#else
++ .ascii ",PAE=no"
++#endif
++ .ascii ",LOADER=generic"
++ .byte 0
+diff -urNp linux-2.6/arch/i386/kernel/init_task-xen.c new/arch/i386/kernel/init_task-xen.c
+--- linux-2.6/arch/i386/kernel/init_task-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/init_task-xen.c 2006-05-09 12:32:35.000000000 +0200
+@@ -0,0 +1,51 @@
++#include <linux/mm.h>
++#include <linux/module.h>
++#include <linux/sched.h>
++#include <linux/init.h>
++#include <linux/init_task.h>
++#include <linux/fs.h>
++#include <linux/mqueue.h>
++
++#include <asm/uaccess.h>
++#include <asm/pgtable.h>
++#include <asm/desc.h>
++
++static struct fs_struct init_fs = INIT_FS;
++static struct files_struct init_files = INIT_FILES;
++static struct signal_struct init_signals = INIT_SIGNALS(init_signals);
++static struct sighand_struct init_sighand = INIT_SIGHAND(init_sighand);
++
++#define swapper_pg_dir ((pgd_t *)NULL)
++struct mm_struct init_mm = INIT_MM(init_mm);
++#undef swapper_pg_dir
++
++EXPORT_SYMBOL(init_mm);
++
++/*
++ * Initial thread structure.
++ *
++ * We need to make sure that this is THREAD_SIZE aligned due to the
++ * way process stacks are handled. This is done by having a special
++ * "init_task" linker map entry..
++ */
++union thread_union init_thread_union
++ __attribute__((__section__(".data.init_task"))) =
++ { INIT_THREAD_INFO(init_task) };
++
++/*
++ * Initial task structure.
++ *
++ * All other task structs will be allocated on slabs in fork.c
++ */
++struct task_struct init_task = INIT_TASK(init_task);
++
++EXPORT_SYMBOL(init_task);
++
++#ifndef CONFIG_X86_NO_TSS
++/*
++ * per-CPU TSS segments. Threads are completely 'soft' on Linux,
++ * no more per-task TSS's.
++ */
++DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS;
++#endif
++
+diff -urNp linux-2.6/arch/i386/kernel/io_apic-xen.c new/arch/i386/kernel/io_apic-xen.c
+--- linux-2.6/arch/i386/kernel/io_apic-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/io_apic-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,2751 @@
++/*
++ * Intel IO-APIC support for multi-Pentium hosts.
++ *
++ * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo
++ *
++ * Many thanks to Stig Venaas for trying out countless experimental
++ * patches and reporting/debugging problems patiently!
++ *
++ * (c) 1999, Multiple IO-APIC support, developed by
++ * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and
++ * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>,
++ * further tested and cleaned up by Zach Brown <zab@redhat.com>
++ * and Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
++ * thanks to Eric Gilmore
++ * and Rolf G. Tews
++ * for testing these extensively
++ * Paul Diefenbaugh : Added full ACPI support
++ */
++
++#include <linux/mm.h>
++#include <linux/interrupt.h>
++#include <linux/init.h>
++#include <linux/delay.h>
++#include <linux/sched.h>
++#include <linux/config.h>
++#include <linux/smp_lock.h>
++#include <linux/mc146818rtc.h>
++#include <linux/compiler.h>
++#include <linux/acpi.h>
++#include <linux/module.h>
++#include <linux/sysdev.h>
++
++#include <asm/io.h>
++#include <asm/smp.h>
++#include <asm/desc.h>
++#include <asm/timer.h>
++#include <asm/i8259.h>
++
++#include <mach_apic.h>
++
++#include "io_ports.h"
++
++#ifdef CONFIG_XEN
++
++#include <xen/interface/xen.h>
++#include <xen/interface/physdev.h>
++
++/* Fake i8259 */
++#define make_8259A_irq(_irq) (io_apic_irqs &= ~(1UL<<(_irq)))
++#define disable_8259A_irq(_irq) ((void)0)
++#define i8259A_irq_pending(_irq) (0)
++
++unsigned long io_apic_irqs;
++
++static inline unsigned int xen_io_apic_read(unsigned int apic, unsigned int reg)
++{
++ struct physdev_apic apic_op;
++ int ret;
++
++ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
++ apic_op.reg = reg;
++ ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
++ if (ret)
++ return ret;
++ return apic_op.value;
++}
++
++static inline void xen_io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
++{
++ struct physdev_apic apic_op;
++
++ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
++ apic_op.reg = reg;
++ apic_op.value = value;
++ HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op);
++}
++
++#define io_apic_read(a,r) xen_io_apic_read(a,r)
++#define io_apic_write(a,r,v) xen_io_apic_write(a,r,v)
++
++#endif /* CONFIG_XEN */
++
++int (*ioapic_renumber_irq)(int ioapic, int irq);
++atomic_t irq_mis_count;
++
++/* Where if anywhere is the i8259 connect in external int mode */
++static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
++
++static DEFINE_SPINLOCK(ioapic_lock);
++
++int timer_over_8254 __initdata = 1;
++
++/*
++ * Is the SiS APIC rmw bug present ?
++ * -1 = don't know, 0 = no, 1 = yes
++ */
++int sis_apic_bug = -1;
++
++/*
++ * # of IRQ routing registers
++ */
++int nr_ioapic_registers[MAX_IO_APICS];
++
++int disable_timer_pin_1 __initdata;
++
++/*
++ * Rough estimation of how many shared IRQs there are, can
++ * be changed anytime.
++ */
++#define MAX_PLUS_SHARED_IRQS NR_IRQS
++#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
++
++/*
++ * This is performance-critical, we want to do it O(1)
++ *
++ * the indexing order of this array favors 1:1 mappings
++ * between pins and IRQs.
++ */
++
++static struct irq_pin_list {
++ int apic, pin, next;
++} irq_2_pin[PIN_MAP_SIZE];
++
++int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
++#ifdef CONFIG_PCI_MSI
++#define vector_to_irq(vector) \
++ (platform_legacy_irq(vector) ? vector : vector_irq[vector])
++#else
++#define vector_to_irq(vector) (vector)
++#endif
++
++/*
++ * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
++ * shared ISA-space IRQs, so we have to support them. We are super
++ * fast in the common case, and fast for shared ISA-space IRQs.
++ */
++static void add_pin_to_irq(unsigned int irq, int apic, int pin)
++{
++ static int first_free_entry = NR_IRQS;
++ struct irq_pin_list *entry = irq_2_pin + irq;
++
++ while (entry->next)
++ entry = irq_2_pin + entry->next;
++
++ if (entry->pin != -1) {
++ entry->next = first_free_entry;
++ entry = irq_2_pin + entry->next;
++ if (++first_free_entry >= PIN_MAP_SIZE)
++ panic("io_apic.c: whoops");
++ }
++ entry->apic = apic;
++ entry->pin = pin;
++}
++
++#ifdef CONFIG_XEN
++#define clear_IO_APIC() ((void)0)
++#else
++/*
++ * Reroute an IRQ to a different pin.
++ */
++static void __init replace_pin_at_irq(unsigned int irq,
++ int oldapic, int oldpin,
++ int newapic, int newpin)
++{
++ struct irq_pin_list *entry = irq_2_pin + irq;
++
++ while (1) {
++ if (entry->apic == oldapic && entry->pin == oldpin) {
++ entry->apic = newapic;
++ entry->pin = newpin;
++ }
++ if (!entry->next)
++ break;
++ entry = irq_2_pin + entry->next;
++ }
++}
++
++static void __modify_IO_APIC_irq (unsigned int irq, unsigned long enable, unsigned long disable)
++{
++ struct irq_pin_list *entry = irq_2_pin + irq;
++ unsigned int pin, reg;
++
++ for (;;) {
++ pin = entry->pin;
++ if (pin == -1)
++ break;
++ reg = io_apic_read(entry->apic, 0x10 + pin*2);
++ reg &= ~disable;
++ reg |= enable;
++ io_apic_modify(entry->apic, 0x10 + pin*2, reg);
++ if (!entry->next)
++ break;
++ entry = irq_2_pin + entry->next;
++ }
++}
++
++/* mask = 1 */
++static void __mask_IO_APIC_irq (unsigned int irq)
++{
++ __modify_IO_APIC_irq(irq, 0x00010000, 0);
++}
++
++/* mask = 0 */
++static void __unmask_IO_APIC_irq (unsigned int irq)
++{
++ __modify_IO_APIC_irq(irq, 0, 0x00010000);
++}
++
++/* mask = 1, trigger = 0 */
++static void __mask_and_edge_IO_APIC_irq (unsigned int irq)
++{
++ __modify_IO_APIC_irq(irq, 0x00010000, 0x00008000);
++}
++
++/* mask = 0, trigger = 1 */
++static void __unmask_and_level_IO_APIC_irq (unsigned int irq)
++{
++ __modify_IO_APIC_irq(irq, 0x00008000, 0x00010000);
++}
++
++static void mask_IO_APIC_irq (unsigned int irq)
++{
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ __mask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void unmask_IO_APIC_irq (unsigned int irq)
++{
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ __unmask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ /* Check delivery_mode to be sure we're not clearing an SMI pin */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ *(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ if (entry.delivery_mode == dest_SMI)
++ return;
++
++ /*
++ * Disable it in the IO-APIC irq-routing table:
++ */
++ memset(&entry, 0, sizeof(entry));
++ entry.mask = 1;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void clear_IO_APIC (void)
++{
++ int apic, pin;
++
++ for (apic = 0; apic < nr_ioapics; apic++)
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
++ clear_IO_APIC_pin(apic, pin);
++}
++
++#ifdef CONFIG_SMP
++static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t cpumask)
++{
++ unsigned long flags;
++ int pin;
++ struct irq_pin_list *entry = irq_2_pin + irq;
++ unsigned int apicid_value;
++ cpumask_t tmp;
++
++ cpus_and(tmp, cpumask, cpu_online_map);
++ if (cpus_empty(tmp))
++ tmp = TARGET_CPUS;
++
++ cpus_and(cpumask, tmp, CPU_MASK_ALL);
++
++ apicid_value = cpu_mask_to_apicid(cpumask);
++ /* Prepare to do the io_apic_write */
++ apicid_value = apicid_value << 24;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ for (;;) {
++ pin = entry->pin;
++ if (pin == -1)
++ break;
++ io_apic_write(entry->apic, 0x10 + 1 + pin*2, apicid_value);
++ if (!entry->next)
++ break;
++ entry = irq_2_pin + entry->next;
++ }
++ set_irq_info(irq, cpumask);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++#if defined(CONFIG_IRQBALANCE)
++# include <asm/processor.h> /* kernel_thread() */
++# include <linux/kernel_stat.h> /* kstat */
++# include <linux/slab.h> /* kmalloc() */
++# include <linux/timer.h> /* time_after() */
++
++# ifdef CONFIG_BALANCED_IRQ_DEBUG
++# define TDprintk(x...) do { printk("<%ld:%s:%d>: ", jiffies, __FILE__, __LINE__); printk(x); } while (0)
++# define Dprintk(x...) do { TDprintk(x); } while (0)
++# else
++# define TDprintk(x...)
++# define Dprintk(x...)
++# endif
++
++
++#define IRQBALANCE_CHECK_ARCH -999
++static int irqbalance_disabled = IRQBALANCE_CHECK_ARCH;
++static int physical_balance = 0;
++
++static struct irq_cpu_info {
++ unsigned long * last_irq;
++ unsigned long * irq_delta;
++ unsigned long irq;
++} irq_cpu_data[NR_CPUS];
++
++#define CPU_IRQ(cpu) (irq_cpu_data[cpu].irq)
++#define LAST_CPU_IRQ(cpu,irq) (irq_cpu_data[cpu].last_irq[irq])
++#define IRQ_DELTA(cpu,irq) (irq_cpu_data[cpu].irq_delta[irq])
++
++#define IDLE_ENOUGH(cpu,now) \
++ (idle_cpu(cpu) && ((now) - per_cpu(irq_stat, (cpu)).idle_timestamp > 1))
++
++#define IRQ_ALLOWED(cpu, allowed_mask) cpu_isset(cpu, allowed_mask)
++
++#define CPU_TO_PACKAGEINDEX(i) (first_cpu(cpu_sibling_map[i]))
++
++#define MAX_BALANCED_IRQ_INTERVAL (5*HZ)
++#define MIN_BALANCED_IRQ_INTERVAL (HZ/2)
++#define BALANCED_IRQ_MORE_DELTA (HZ/10)
++#define BALANCED_IRQ_LESS_DELTA (HZ)
++
++static long balanced_irq_interval = MAX_BALANCED_IRQ_INTERVAL;
++
++static unsigned long move(int curr_cpu, cpumask_t allowed_mask,
++ unsigned long now, int direction)
++{
++ int search_idle = 1;
++ int cpu = curr_cpu;
++
++ goto inside;
++
++ do {
++ if (unlikely(cpu == curr_cpu))
++ search_idle = 0;
++inside:
++ if (direction == 1) {
++ cpu++;
++ if (cpu >= NR_CPUS)
++ cpu = 0;
++ } else {
++ cpu--;
++ if (cpu == -1)
++ cpu = NR_CPUS-1;
++ }
++ } while (!cpu_online(cpu) || !IRQ_ALLOWED(cpu,allowed_mask) ||
++ (search_idle && !IDLE_ENOUGH(cpu,now)));
++
++ return cpu;
++}
++
++static inline void balance_irq(int cpu, int irq)
++{
++ unsigned long now = jiffies;
++ cpumask_t allowed_mask;
++ unsigned int new_cpu;
++
++ if (irqbalance_disabled)
++ return;
++
++ cpus_and(allowed_mask, cpu_online_map, irq_affinity[irq]);
++ new_cpu = move(cpu, allowed_mask, now, 1);
++ if (cpu != new_cpu) {
++ set_pending_irq(irq, cpumask_of_cpu(new_cpu));
++ }
++}
++
++static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
++{
++ int i, j;
++ Dprintk("Rotating IRQs among CPUs.\n");
++ for_each_online_cpu(i) {
++ for (j = 0; j < NR_IRQS; j++) {
++ if (!irq_desc[j].action)
++ continue;
++ /* Is it a significant load ? */
++ if (IRQ_DELTA(CPU_TO_PACKAGEINDEX(i),j) <
++ useful_load_threshold)
++ continue;
++ balance_irq(i, j);
++ }
++ }
++ balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
++ balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
++ return;
++}
++
++static void do_irq_balance(void)
++{
++ int i, j;
++ unsigned long max_cpu_irq = 0, min_cpu_irq = (~0);
++ unsigned long move_this_load = 0;
++ int max_loaded = 0, min_loaded = 0;
++ int load;
++ unsigned long useful_load_threshold = balanced_irq_interval + 10;
++ int selected_irq;
++ int tmp_loaded, first_attempt = 1;
++ unsigned long tmp_cpu_irq;
++ unsigned long imbalance = 0;
++ cpumask_t allowed_mask, target_cpu_mask, tmp;
++
++ for_each_possible_cpu(i) {
++ int package_index;
++ CPU_IRQ(i) = 0;
++ if (!cpu_online(i))
++ continue;
++ package_index = CPU_TO_PACKAGEINDEX(i);
++ for (j = 0; j < NR_IRQS; j++) {
++ unsigned long value_now, delta;
++ /* Is this an active IRQ? */
++ if (!irq_desc[j].action)
++ continue;
++ if ( package_index == i )
++ IRQ_DELTA(package_index,j) = 0;
++ /* Determine the total count per processor per IRQ */
++ value_now = (unsigned long) kstat_cpu(i).irqs[j];
++
++ /* Determine the activity per processor per IRQ */
++ delta = value_now - LAST_CPU_IRQ(i,j);
++
++ /* Update last_cpu_irq[][] for the next time */
++ LAST_CPU_IRQ(i,j) = value_now;
++
++ /* Ignore IRQs whose rate is less than the clock */
++ if (delta < useful_load_threshold)
++ continue;
++ /* update the load for the processor or package total */
++ IRQ_DELTA(package_index,j) += delta;
++
++ /* Keep track of the higher numbered sibling as well */
++ if (i != package_index)
++ CPU_IRQ(i) += delta;
++ /*
++ * We have sibling A and sibling B in the package
++ *
++ * cpu_irq[A] = load for cpu A + load for cpu B
++ * cpu_irq[B] = load for cpu B
++ */
++ CPU_IRQ(package_index) += delta;
++ }
++ }
++ /* Find the least loaded processor package */
++ for_each_online_cpu(i) {
++ if (i != CPU_TO_PACKAGEINDEX(i))
++ continue;
++ if (min_cpu_irq > CPU_IRQ(i)) {
++ min_cpu_irq = CPU_IRQ(i);
++ min_loaded = i;
++ }
++ }
++ max_cpu_irq = ULONG_MAX;
++
++tryanothercpu:
++ /* Look for heaviest loaded processor.
++ * We may come back to get the next heaviest loaded processor.
++ * Skip processors with trivial loads.
++ */
++ tmp_cpu_irq = 0;
++ tmp_loaded = -1;
++ for_each_online_cpu(i) {
++ if (i != CPU_TO_PACKAGEINDEX(i))
++ continue;
++ if (max_cpu_irq <= CPU_IRQ(i))
++ continue;
++ if (tmp_cpu_irq < CPU_IRQ(i)) {
++ tmp_cpu_irq = CPU_IRQ(i);
++ tmp_loaded = i;
++ }
++ }
++
++ if (tmp_loaded == -1) {
++ /* In the case of small number of heavy interrupt sources,
++ * loading some of the cpus too much. We use Ingo's original
++ * approach to rotate them around.
++ */
++ if (!first_attempt && imbalance >= useful_load_threshold) {
++ rotate_irqs_among_cpus(useful_load_threshold);
++ return;
++ }
++ goto not_worth_the_effort;
++ }
++
++ first_attempt = 0; /* heaviest search */
++ max_cpu_irq = tmp_cpu_irq; /* load */
++ max_loaded = tmp_loaded; /* processor */
++ imbalance = (max_cpu_irq - min_cpu_irq) / 2;
++
++ Dprintk("max_loaded cpu = %d\n", max_loaded);
++ Dprintk("min_loaded cpu = %d\n", min_loaded);
++ Dprintk("max_cpu_irq load = %ld\n", max_cpu_irq);
++ Dprintk("min_cpu_irq load = %ld\n", min_cpu_irq);
++ Dprintk("load imbalance = %lu\n", imbalance);
++
++ /* if imbalance is less than approx 10% of max load, then
++ * observe diminishing returns action. - quit
++ */
++ if (imbalance < (max_cpu_irq >> 3)) {
++ Dprintk("Imbalance too trivial\n");
++ goto not_worth_the_effort;
++ }
++
++tryanotherirq:
++ /* if we select an IRQ to move that can't go where we want, then
++ * see if there is another one to try.
++ */
++ move_this_load = 0;
++ selected_irq = -1;
++ for (j = 0; j < NR_IRQS; j++) {
++ /* Is this an active IRQ? */
++ if (!irq_desc[j].action)
++ continue;
++ if (imbalance <= IRQ_DELTA(max_loaded,j))
++ continue;
++ /* Try to find the IRQ that is closest to the imbalance
++ * without going over.
++ */
++ if (move_this_load < IRQ_DELTA(max_loaded,j)) {
++ move_this_load = IRQ_DELTA(max_loaded,j);
++ selected_irq = j;
++ }
++ }
++ if (selected_irq == -1) {
++ goto tryanothercpu;
++ }
++
++ imbalance = move_this_load;
++
++ /* For physical_balance case, we accumlated both load
++ * values in the one of the siblings cpu_irq[],
++ * to use the same code for physical and logical processors
++ * as much as possible.
++ *
++ * NOTE: the cpu_irq[] array holds the sum of the load for
++ * sibling A and sibling B in the slot for the lowest numbered
++ * sibling (A), _AND_ the load for sibling B in the slot for
++ * the higher numbered sibling.
++ *
++ * We seek the least loaded sibling by making the comparison
++ * (A+B)/2 vs B
++ */
++ load = CPU_IRQ(min_loaded) >> 1;
++ for_each_cpu_mask(j, cpu_sibling_map[min_loaded]) {
++ if (load > CPU_IRQ(j)) {
++ /* This won't change cpu_sibling_map[min_loaded] */
++ load = CPU_IRQ(j);
++ min_loaded = j;
++ }
++ }
++
++ cpus_and(allowed_mask, cpu_online_map, irq_affinity[selected_irq]);
++ target_cpu_mask = cpumask_of_cpu(min_loaded);
++ cpus_and(tmp, target_cpu_mask, allowed_mask);
++
++ if (!cpus_empty(tmp)) {
++
++ Dprintk("irq = %d moved to cpu = %d\n",
++ selected_irq, min_loaded);
++ /* mark for change destination */
++ set_pending_irq(selected_irq, cpumask_of_cpu(min_loaded));
++
++ /* Since we made a change, come back sooner to
++ * check for more variation.
++ */
++ balanced_irq_interval = max((long)MIN_BALANCED_IRQ_INTERVAL,
++ balanced_irq_interval - BALANCED_IRQ_LESS_DELTA);
++ return;
++ }
++ goto tryanotherirq;
++
++not_worth_the_effort:
++ /*
++ * if we did not find an IRQ to move, then adjust the time interval
++ * upward
++ */
++ balanced_irq_interval = min((long)MAX_BALANCED_IRQ_INTERVAL,
++ balanced_irq_interval + BALANCED_IRQ_MORE_DELTA);
++ Dprintk("IRQ worth rotating not found\n");
++ return;
++}
++
++static int balanced_irq(void *unused)
++{
++ int i;
++ unsigned long prev_balance_time = jiffies;
++ long time_remaining = balanced_irq_interval;
++
++ daemonize("kirqd");
++
++ /* push everything to CPU 0 to give us a starting point. */
++ for (i = 0 ; i < NR_IRQS ; i++) {
++ pending_irq_cpumask[i] = cpumask_of_cpu(0);
++ set_pending_irq(i, cpumask_of_cpu(0));
++ }
++
++ for ( ; ; ) {
++ time_remaining = schedule_timeout_interruptible(time_remaining);
++ try_to_freeze();
++ if (time_after(jiffies,
++ prev_balance_time+balanced_irq_interval)) {
++ preempt_disable();
++ do_irq_balance();
++ prev_balance_time = jiffies;
++ time_remaining = balanced_irq_interval;
++ preempt_enable();
++ }
++ }
++ return 0;
++}
++
++static int __init balanced_irq_init(void)
++{
++ int i;
++ struct cpuinfo_x86 *c;
++ cpumask_t tmp;
++
++ cpus_shift_right(tmp, cpu_online_map, 2);
++ c = &boot_cpu_data;
++ /* When not overwritten by the command line ask subarchitecture. */
++ if (irqbalance_disabled == IRQBALANCE_CHECK_ARCH)
++ irqbalance_disabled = NO_BALANCE_IRQ;
++ if (irqbalance_disabled)
++ return 0;
++
++ /* disable irqbalance completely if there is only one processor online */
++ if (num_online_cpus() < 2) {
++ irqbalance_disabled = 1;
++ return 0;
++ }
++ /*
++ * Enable physical balance only if more than 1 physical processor
++ * is present
++ */
++ if (smp_num_siblings > 1 && !cpus_empty(tmp))
++ physical_balance = 1;
++
++ for_each_online_cpu(i) {
++ irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
++ irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
++ if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) {
++ printk(KERN_ERR "balanced_irq_init: out of memory");
++ goto failed;
++ }
++ memset(irq_cpu_data[i].irq_delta,0,sizeof(unsigned long) * NR_IRQS);
++ memset(irq_cpu_data[i].last_irq,0,sizeof(unsigned long) * NR_IRQS);
++ }
++
++ printk(KERN_INFO "Starting balanced_irq\n");
++ if (kernel_thread(balanced_irq, NULL, CLONE_KERNEL) >= 0)
++ return 0;
++ else
++ printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq");
++failed:
++ for_each_possible_cpu(i) {
++ kfree(irq_cpu_data[i].irq_delta);
++ irq_cpu_data[i].irq_delta = NULL;
++ kfree(irq_cpu_data[i].last_irq);
++ irq_cpu_data[i].last_irq = NULL;
++ }
++ return 0;
++}
++
++int __init irqbalance_disable(char *str)
++{
++ irqbalance_disabled = 1;
++ return 1;
++}
++
++__setup("noirqbalance", irqbalance_disable);
++
++late_initcall(balanced_irq_init);
++#endif /* CONFIG_IRQBALANCE */
++#endif /* CONFIG_SMP */
++#endif
++
++#ifndef CONFIG_SMP
++void fastcall send_IPI_self(int vector)
++{
++#ifndef CONFIG_XEN
++ unsigned int cfg;
++
++ /*
++ * Wait for idle.
++ */
++ apic_wait_icr_idle();
++ cfg = APIC_DM_FIXED | APIC_DEST_SELF | vector | APIC_DEST_LOGICAL;
++ /*
++ * Send the IPI. The write to APIC_ICR fires this off.
++ */
++ apic_write_around(APIC_ICR, cfg);
++#endif
++}
++#endif /* !CONFIG_SMP */
++
++
++/*
++ * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to
++ * specific CPU-side IRQs.
++ */
++
++#define MAX_PIRQS 8
++static int pirq_entries [MAX_PIRQS];
++static int pirqs_enabled;
++int skip_ioapic_setup;
++
++static int __init ioapic_setup(char *str)
++{
++ skip_ioapic_setup = 1;
++ return 1;
++}
++
++__setup("noapic", ioapic_setup);
++
++static int __init ioapic_pirq_setup(char *str)
++{
++ int i, max;
++ int ints[MAX_PIRQS+1];
++
++ get_options(str, ARRAY_SIZE(ints), ints);
++
++ for (i = 0; i < MAX_PIRQS; i++)
++ pirq_entries[i] = -1;
++
++ pirqs_enabled = 1;
++ apic_printk(APIC_VERBOSE, KERN_INFO
++ "PIRQ redirection, working around broken MP-BIOS.\n");
++ max = MAX_PIRQS;
++ if (ints[0] < MAX_PIRQS)
++ max = ints[0];
++
++ for (i = 0; i < max; i++) {
++ apic_printk(APIC_VERBOSE, KERN_DEBUG
++ "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
++ /*
++ * PIRQs are mapped upside down, usually.
++ */
++ pirq_entries[MAX_PIRQS-i-1] = ints[i+1];
++ }
++ return 1;
++}
++
++__setup("pirq=", ioapic_pirq_setup);
++
++/*
++ * Find the IRQ entry number of a certain pin.
++ */
++static int find_irq_entry(int apic, int pin, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++)
++ if (mp_irqs[i].mpc_irqtype == type &&
++ (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
++ mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
++ mp_irqs[i].mpc_dstirq == pin)
++ return i;
++
++ return -1;
++}
++
++/*
++ * Find the pin to which IRQ[irq] (ISA) is connected
++ */
++static int __init find_isa_irq_pin(int irq, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_MCA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_NEC98
++ ) &&
++ (mp_irqs[i].mpc_irqtype == type) &&
++ (mp_irqs[i].mpc_srcbusirq == irq))
++
++ return mp_irqs[i].mpc_dstirq;
++ }
++ return -1;
++}
++
++static int __init find_isa_irq_apic(int irq, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_MCA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_NEC98
++ ) &&
++ (mp_irqs[i].mpc_irqtype == type) &&
++ (mp_irqs[i].mpc_srcbusirq == irq))
++ break;
++ }
++ if (i < mp_irq_entries) {
++ int apic;
++ for(apic = 0; apic < nr_ioapics; apic++) {
++ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
++ return apic;
++ }
++ }
++
++ return -1;
++}
++
++/*
++ * Find a specific PCI IRQ entry.
++ * Not an __init, possibly needed by modules
++ */
++static int pin_2_irq(int idx, int apic, int pin);
++
++int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
++{
++ int apic, i, best_guess = -1;
++
++ apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, "
++ "slot:%d, pin:%d.\n", bus, slot, pin);
++ if (mp_bus_id_to_pci_bus[bus] == -1) {
++ printk(KERN_WARNING "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
++ return -1;
++ }
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ for (apic = 0; apic < nr_ioapics; apic++)
++ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
++ mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
++ break;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
++ !mp_irqs[i].mpc_irqtype &&
++ (bus == lbus) &&
++ (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
++ int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
++
++ if (!(apic || IO_APIC_IRQ(irq)))
++ continue;
++
++ if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
++ return irq;
++ /*
++ * Use the first all-but-pin matching entry as a
++ * best-guess fuzzy result for broken mptables.
++ */
++ if (best_guess < 0)
++ best_guess = irq;
++ }
++ }
++ return best_guess;
++}
++EXPORT_SYMBOL(IO_APIC_get_PCI_irq_vector);
++
++/*
++ * This function currently is only a helper for the i386 smp boot process where
++ * we need to reprogram the ioredtbls to cater for the cpus which have come online
++ * so mask in all cases should simply be TARGET_CPUS
++ */
++#ifdef CONFIG_SMP
++#ifndef CONFIG_XEN
++void __init setup_ioapic_dest(void)
++{
++ int pin, ioapic, irq, irq_entry;
++
++ if (skip_ioapic_setup == 1)
++ return;
++
++ for (ioapic = 0; ioapic < nr_ioapics; ioapic++) {
++ for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) {
++ irq_entry = find_irq_entry(ioapic, pin, mp_INT);
++ if (irq_entry == -1)
++ continue;
++ irq = pin_2_irq(irq_entry, ioapic, pin);
++ set_ioapic_affinity_irq(irq, TARGET_CPUS);
++ }
++
++ }
++}
++#endif /* !CONFIG_XEN */
++#endif
++
++/*
++ * EISA Edge/Level control register, ELCR
++ */
++static int EISA_ELCR(unsigned int irq)
++{
++ if (irq < 16) {
++ unsigned int port = 0x4d0 + (irq >> 3);
++ return (inb(port) >> (irq & 7)) & 1;
++ }
++ apic_printk(APIC_VERBOSE, KERN_INFO
++ "Broken MPtable reports ISA irq %d\n", irq);
++ return 0;
++}
++
++/* EISA interrupts are always polarity zero and can be edge or level
++ * trigger depending on the ELCR value. If an interrupt is listed as
++ * EISA conforming in the MP table, that means its trigger type must
++ * be read in from the ELCR */
++
++#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
++#define default_EISA_polarity(idx) (0)
++
++/* ISA interrupts are always polarity zero edge triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_ISA_trigger(idx) (0)
++#define default_ISA_polarity(idx) (0)
++
++/* PCI interrupts are always polarity one level triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_PCI_trigger(idx) (1)
++#define default_PCI_polarity(idx) (1)
++
++/* MCA interrupts are always polarity zero level triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_MCA_trigger(idx) (1)
++#define default_MCA_polarity(idx) (0)
++
++/* NEC98 interrupts are always polarity zero edge triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_NEC98_trigger(idx) (0)
++#define default_NEC98_polarity(idx) (0)
++
++static int __init MPBIOS_polarity(int idx)
++{
++ int bus = mp_irqs[idx].mpc_srcbus;
++ int polarity;
++
++ /*
++ * Determine IRQ line polarity (high active or low active):
++ */
++ switch (mp_irqs[idx].mpc_irqflag & 3)
++ {
++ case 0: /* conforms, ie. bus-type dependent polarity */
++ {
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ {
++ polarity = default_ISA_polarity(idx);
++ break;
++ }
++ case MP_BUS_EISA: /* EISA pin */
++ {
++ polarity = default_EISA_polarity(idx);
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ polarity = default_PCI_polarity(idx);
++ break;
++ }
++ case MP_BUS_MCA: /* MCA pin */
++ {
++ polarity = default_MCA_polarity(idx);
++ break;
++ }
++ case MP_BUS_NEC98: /* NEC 98 pin */
++ {
++ polarity = default_NEC98_polarity(idx);
++ break;
++ }
++ default:
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ }
++ break;
++ }
++ case 1: /* high active */
++ {
++ polarity = 0;
++ break;
++ }
++ case 2: /* reserved */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ case 3: /* low active */
++ {
++ polarity = 1;
++ break;
++ }
++ default: /* invalid */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ }
++ return polarity;
++}
++
++static int MPBIOS_trigger(int idx)
++{
++ int bus = mp_irqs[idx].mpc_srcbus;
++ int trigger;
++
++ /*
++ * Determine IRQ trigger mode (edge or level sensitive):
++ */
++ switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
++ {
++ case 0: /* conforms, ie. bus-type dependent */
++ {
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ {
++ trigger = default_ISA_trigger(idx);
++ break;
++ }
++ case MP_BUS_EISA: /* EISA pin */
++ {
++ trigger = default_EISA_trigger(idx);
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ trigger = default_PCI_trigger(idx);
++ break;
++ }
++ case MP_BUS_MCA: /* MCA pin */
++ {
++ trigger = default_MCA_trigger(idx);
++ break;
++ }
++ case MP_BUS_NEC98: /* NEC 98 pin */
++ {
++ trigger = default_NEC98_trigger(idx);
++ break;
++ }
++ default:
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 1;
++ break;
++ }
++ }
++ break;
++ }
++ case 1: /* edge */
++ {
++ trigger = 0;
++ break;
++ }
++ case 2: /* reserved */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 1;
++ break;
++ }
++ case 3: /* level */
++ {
++ trigger = 1;
++ break;
++ }
++ default: /* invalid */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 0;
++ break;
++ }
++ }
++ return trigger;
++}
++
++static inline int irq_polarity(int idx)
++{
++ return MPBIOS_polarity(idx);
++}
++
++static inline int irq_trigger(int idx)
++{
++ return MPBIOS_trigger(idx);
++}
++
++static int pin_2_irq(int idx, int apic, int pin)
++{
++ int irq, i;
++ int bus = mp_irqs[idx].mpc_srcbus;
++
++ /*
++ * Debugging check, we are in big trouble if this message pops up!
++ */
++ if (mp_irqs[idx].mpc_dstirq != pin)
++ printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
++
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ case MP_BUS_EISA:
++ case MP_BUS_MCA:
++ case MP_BUS_NEC98:
++ {
++ irq = mp_irqs[idx].mpc_srcbusirq;
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ /*
++ * PCI IRQs are mapped in order
++ */
++ i = irq = 0;
++ while (i < apic)
++ irq += nr_ioapic_registers[i++];
++ irq += pin;
++
++ /*
++ * For MPS mode, so far only needed by ES7000 platform
++ */
++ if (ioapic_renumber_irq)
++ irq = ioapic_renumber_irq(apic, irq);
++
++ break;
++ }
++ default:
++ {
++ printk(KERN_ERR "unknown bus type %d.\n",bus);
++ irq = 0;
++ break;
++ }
++ }
++
++ /*
++ * PCI IRQ command line redirection. Yes, limits are hardcoded.
++ */
++ if ((pin >= 16) && (pin <= 23)) {
++ if (pirq_entries[pin-16] != -1) {
++ if (!pirq_entries[pin-16]) {
++ apic_printk(APIC_VERBOSE, KERN_DEBUG
++ "disabling PIRQ%d\n", pin-16);
++ } else {
++ irq = pirq_entries[pin-16];
++ apic_printk(APIC_VERBOSE, KERN_DEBUG
++ "using PIRQ%d -> IRQ %d\n",
++ pin-16, irq);
++ }
++ }
++ }
++ return irq;
++}
++
++static inline int IO_APIC_irq_trigger(int irq)
++{
++ int apic, idx, pin;
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++ idx = find_irq_entry(apic,pin,mp_INT);
++ if ((idx != -1) && (irq == pin_2_irq(idx,apic,pin)))
++ return irq_trigger(idx);
++ }
++ }
++ /*
++ * nonexistent IRQs are edge default
++ */
++ return 0;
++}
++
++/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */
++u8 irq_vector[NR_IRQ_VECTORS] __read_mostly; /* = { FIRST_DEVICE_VECTOR , 0 }; */
++
++int assign_irq_vector(int irq)
++{
++ struct physdev_irq irq_op;
++
++ BUG_ON(irq >= NR_IRQ_VECTORS);
++ if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0)
++ return IO_APIC_VECTOR(irq);
++
++ irq_op.irq = irq;
++ if (HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op))
++ return -ENOSPC;
++
++ vector_irq[irq_op.vector] = irq;
++ if (irq != AUTO_ASSIGN)
++ IO_APIC_VECTOR(irq) = irq_op.vector;
++
++ return irq_op.vector;
++}
++
++#ifndef CONFIG_XEN
++static struct hw_interrupt_type ioapic_level_type;
++static struct hw_interrupt_type ioapic_edge_type;
++
++#define IOAPIC_AUTO -1
++#define IOAPIC_EDGE 0
++#define IOAPIC_LEVEL 1
++
++static inline void ioapic_register_intr(int irq, int vector, unsigned long trigger)
++{
++ if (use_pci_vector() && !platform_legacy_irq(irq)) {
++ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
++ trigger == IOAPIC_LEVEL)
++ irq_desc[vector].handler = &ioapic_level_type;
++ else
++ irq_desc[vector].handler = &ioapic_edge_type;
++ set_intr_gate(vector, interrupt[vector]);
++ } else {
++ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
++ trigger == IOAPIC_LEVEL)
++ irq_desc[irq].handler = &ioapic_level_type;
++ else
++ irq_desc[irq].handler = &ioapic_edge_type;
++ set_intr_gate(vector, interrupt[irq]);
++ }
++}
++#else
++#define ioapic_register_intr(_irq,_vector,_trigger) ((void)0)
++#endif
++
++static void __init setup_IO_APIC_irqs(void)
++{
++ struct IO_APIC_route_entry entry;
++ int apic, pin, idx, irq, first_notcon = 1, vector;
++ unsigned long flags;
++
++ apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++
++ /*
++ * add it to the IO-APIC irq-routing table:
++ */
++ memset(&entry,0,sizeof(entry));
++
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.dest_mode = INT_DEST_MODE;
++ entry.mask = 0; /* enable IRQ */
++ entry.dest.logical.logical_dest =
++ cpu_mask_to_apicid(TARGET_CPUS);
++
++ idx = find_irq_entry(apic,pin,mp_INT);
++ if (idx == -1) {
++ if (first_notcon) {
++ apic_printk(APIC_VERBOSE, KERN_DEBUG
++ " IO-APIC (apicid-pin) %d-%d",
++ mp_ioapics[apic].mpc_apicid,
++ pin);
++ first_notcon = 0;
++ } else
++ apic_printk(APIC_VERBOSE, ", %d-%d",
++ mp_ioapics[apic].mpc_apicid, pin);
++ continue;
++ }
++
++ entry.trigger = irq_trigger(idx);
++ entry.polarity = irq_polarity(idx);
++
++ if (irq_trigger(idx)) {
++ entry.trigger = 1;
++ entry.mask = 1;
++ }
++
++ irq = pin_2_irq(idx, apic, pin);
++ /*
++ * skip adding the timer int on secondary nodes, which causes
++ * a small but painful rift in the time-space continuum
++ */
++ if (multi_timer_check(apic, irq))
++ continue;
++ else
++ add_pin_to_irq(irq, apic, pin);
++
++ if (/*!apic &&*/ !IO_APIC_IRQ(irq))
++ continue;
++
++ if (IO_APIC_IRQ(irq)) {
++ vector = assign_irq_vector(irq);
++ entry.vector = vector;
++ ioapic_register_intr(irq, vector, IOAPIC_AUTO);
++
++ if (!apic && (irq < 16))
++ disable_8259A_irq(irq);
++ }
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
++ set_native_irq_info(irq, TARGET_CPUS);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ }
++ }
++
++ if (!first_notcon)
++ apic_printk(APIC_VERBOSE, " not connected.\n");
++}
++
++/*
++ * Set up the 8259A-master output pin:
++ */
++#ifndef CONFIG_XEN
++static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ memset(&entry,0,sizeof(entry));
++
++ disable_8259A_irq(0);
++
++ /* mask LVT0 */
++ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
++
++ /*
++ * We use logical delivery to get the timer IRQ
++ * to the first CPU.
++ */
++ entry.dest_mode = INT_DEST_MODE;
++ entry.mask = 0; /* unmask IRQ now */
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.polarity = 0;
++ entry.trigger = 0;
++ entry.vector = vector;
++
++ /*
++ * The timer IRQ doesn't have to know that behind the
++ * scene we have a 8259A-master in AEOI mode ...
++ */
++ irq_desc[0].handler = &ioapic_edge_type;
++
++ /*
++ * Add it to the IO-APIC irq-routing table:
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ enable_8259A_irq(0);
++}
++
++static inline void UNEXPECTED_IO_APIC(void)
++{
++}
++
++void __init print_IO_APIC(void)
++{
++ int apic, i;
++ union IO_APIC_reg_00 reg_00;
++ union IO_APIC_reg_01 reg_01;
++ union IO_APIC_reg_02 reg_02;
++ union IO_APIC_reg_03 reg_03;
++ unsigned long flags;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
++ for (i = 0; i < nr_ioapics; i++)
++ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
++ mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
++
++ /*
++ * We are a bit conservative about what we expect. We have to
++ * know about every hardware change ASAP.
++ */
++ printk(KERN_INFO "testing the IO APIC.......................\n");
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ reg_01.raw = io_apic_read(apic, 1);
++ if (reg_01.bits.version >= 0x10)
++ reg_02.raw = io_apic_read(apic, 2);
++ if (reg_01.bits.version >= 0x20)
++ reg_03.raw = io_apic_read(apic, 3);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
++ printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
++ printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID);
++ printk(KERN_DEBUG "....... : Delivery Type: %X\n", reg_00.bits.delivery_type);
++ printk(KERN_DEBUG "....... : LTS : %X\n", reg_00.bits.LTS);
++ if (reg_00.bits.ID >= get_physical_broadcast())
++ UNEXPECTED_IO_APIC();
++ if (reg_00.bits.__reserved_1 || reg_00.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++
++ printk(KERN_DEBUG ".... register #01: %08X\n", reg_01.raw);
++ printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries);
++ if ( (reg_01.bits.entries != 0x0f) && /* older (Neptune) boards */
++ (reg_01.bits.entries != 0x17) && /* typical ISA+PCI boards */
++ (reg_01.bits.entries != 0x1b) && /* Compaq Proliant boards */
++ (reg_01.bits.entries != 0x1f) && /* dual Xeon boards */
++ (reg_01.bits.entries != 0x22) && /* bigger Xeon boards */
++ (reg_01.bits.entries != 0x2E) &&
++ (reg_01.bits.entries != 0x3F)
++ )
++ UNEXPECTED_IO_APIC();
++
++ printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ);
++ printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version);
++ if ( (reg_01.bits.version != 0x01) && /* 82489DX IO-APICs */
++ (reg_01.bits.version != 0x10) && /* oldest IO-APICs */
++ (reg_01.bits.version != 0x11) && /* Pentium/Pro IO-APICs */
++ (reg_01.bits.version != 0x13) && /* Xeon IO-APICs */
++ (reg_01.bits.version != 0x20) /* Intel P64H (82806 AA) */
++ )
++ UNEXPECTED_IO_APIC();
++ if (reg_01.bits.__reserved_1 || reg_01.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++
++ /*
++ * Some Intel chipsets with IO APIC VERSION of 0x1? don't have reg_02,
++ * but the value of reg_02 is read as the previous read register
++ * value, so ignore it if reg_02 == reg_01.
++ */
++ if (reg_01.bits.version >= 0x10 && reg_02.raw != reg_01.raw) {
++ printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw);
++ printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration);
++ if (reg_02.bits.__reserved_1 || reg_02.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++ }
++
++ /*
++ * Some Intel chipsets with IO APIC VERSION of 0x2? don't have reg_02
++ * or reg_03, but the value of reg_0[23] is read as the previous read
++ * register value, so ignore it if reg_03 == reg_0[12].
++ */
++ if (reg_01.bits.version >= 0x20 && reg_03.raw != reg_02.raw &&
++ reg_03.raw != reg_01.raw) {
++ printk(KERN_DEBUG ".... register #03: %08X\n", reg_03.raw);
++ printk(KERN_DEBUG "....... : Boot DT : %X\n", reg_03.bits.boot_DT);
++ if (reg_03.bits.__reserved_1)
++ UNEXPECTED_IO_APIC();
++ }
++
++ printk(KERN_DEBUG ".... IRQ redirection table:\n");
++
++ printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol"
++ " Stat Dest Deli Vect: \n");
++
++ for (i = 0; i <= reg_01.bits.entries; i++) {
++ struct IO_APIC_route_entry entry;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
++ *(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ printk(KERN_DEBUG " %02x %03X %02X ",
++ i,
++ entry.dest.logical.logical_dest,
++ entry.dest.physical.physical_dest
++ );
++
++ printk("%1d %1d %1d %1d %1d %1d %1d %02X\n",
++ entry.mask,
++ entry.trigger,
++ entry.irr,
++ entry.polarity,
++ entry.delivery_status,
++ entry.dest_mode,
++ entry.delivery_mode,
++ entry.vector
++ );
++ }
++ }
++ if (use_pci_vector())
++ printk(KERN_INFO "Using vector-based indexing\n");
++ printk(KERN_DEBUG "IRQ to pin mappings:\n");
++ for (i = 0; i < NR_IRQS; i++) {
++ struct irq_pin_list *entry = irq_2_pin + i;
++ if (entry->pin < 0)
++ continue;
++ if (use_pci_vector() && !platform_legacy_irq(i))
++ printk(KERN_DEBUG "IRQ%d ", IO_APIC_VECTOR(i));
++ else
++ printk(KERN_DEBUG "IRQ%d ", i);
++ for (;;) {
++ printk("-> %d:%d", entry->apic, entry->pin);
++ if (!entry->next)
++ break;
++ entry = irq_2_pin + entry->next;
++ }
++ printk("\n");
++ }
++
++ printk(KERN_INFO ".................................... done.\n");
++
++ return;
++}
++
++#if 0
++
++static void print_APIC_bitfield (int base)
++{
++ unsigned int v;
++ int i, j;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG);
++ for (i = 0; i < 8; i++) {
++ v = apic_read(base + i*0x10);
++ for (j = 0; j < 32; j++) {
++ if (v & (1<<j))
++ printk("1");
++ else
++ printk("0");
++ }
++ printk("\n");
++ }
++}
++
++void /*__init*/ print_local_APIC(void * dummy)
++{
++ unsigned int v, ver, maxlvt;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
++ smp_processor_id(), hard_smp_processor_id());
++ v = apic_read(APIC_ID);
++ printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v));
++ v = apic_read(APIC_LVR);
++ printk(KERN_INFO "... APIC VERSION: %08x\n", v);
++ ver = GET_APIC_VERSION(v);
++ maxlvt = get_maxlvt();
++
++ v = apic_read(APIC_TASKPRI);
++ printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
++
++ if (APIC_INTEGRATED(ver)) { /* !82489DX */
++ v = apic_read(APIC_ARBPRI);
++ printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v,
++ v & APIC_ARBPRI_MASK);
++ v = apic_read(APIC_PROCPRI);
++ printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v);
++ }
++
++ v = apic_read(APIC_EOI);
++ printk(KERN_DEBUG "... APIC EOI: %08x\n", v);
++ v = apic_read(APIC_RRR);
++ printk(KERN_DEBUG "... APIC RRR: %08x\n", v);
++ v = apic_read(APIC_LDR);
++ printk(KERN_DEBUG "... APIC LDR: %08x\n", v);
++ v = apic_read(APIC_DFR);
++ printk(KERN_DEBUG "... APIC DFR: %08x\n", v);
++ v = apic_read(APIC_SPIV);
++ printk(KERN_DEBUG "... APIC SPIV: %08x\n", v);
++
++ printk(KERN_DEBUG "... APIC ISR field:\n");
++ print_APIC_bitfield(APIC_ISR);
++ printk(KERN_DEBUG "... APIC TMR field:\n");
++ print_APIC_bitfield(APIC_TMR);
++ printk(KERN_DEBUG "... APIC IRR field:\n");
++ print_APIC_bitfield(APIC_IRR);
++
++ if (APIC_INTEGRATED(ver)) { /* !82489DX */
++ if (maxlvt > 3) /* Due to the Pentium erratum 3AP. */
++ apic_write(APIC_ESR, 0);
++ v = apic_read(APIC_ESR);
++ printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
++ }
++
++ v = apic_read(APIC_ICR);
++ printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
++ v = apic_read(APIC_ICR2);
++ printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
++
++ v = apic_read(APIC_LVTT);
++ printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
++
++ if (maxlvt > 3) { /* PC is LVT#4. */
++ v = apic_read(APIC_LVTPC);
++ printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v);
++ }
++ v = apic_read(APIC_LVT0);
++ printk(KERN_DEBUG "... APIC LVT0: %08x\n", v);
++ v = apic_read(APIC_LVT1);
++ printk(KERN_DEBUG "... APIC LVT1: %08x\n", v);
++
++ if (maxlvt > 2) { /* ERR is LVT#3. */
++ v = apic_read(APIC_LVTERR);
++ printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v);
++ }
++
++ v = apic_read(APIC_TMICT);
++ printk(KERN_DEBUG "... APIC TMICT: %08x\n", v);
++ v = apic_read(APIC_TMCCT);
++ printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v);
++ v = apic_read(APIC_TDCR);
++ printk(KERN_DEBUG "... APIC TDCR: %08x\n", v);
++ printk("\n");
++}
++
++void print_all_local_APICs (void)
++{
++ on_each_cpu(print_local_APIC, NULL, 1, 1);
++}
++
++void /*__init*/ print_PIC(void)
++{
++ unsigned int v;
++ unsigned long flags;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "\nprinting PIC contents\n");
++
++ spin_lock_irqsave(&i8259A_lock, flags);
++
++ v = inb(0xa1) << 8 | inb(0x21);
++ printk(KERN_DEBUG "... PIC IMR: %04x\n", v);
++
++ v = inb(0xa0) << 8 | inb(0x20);
++ printk(KERN_DEBUG "... PIC IRR: %04x\n", v);
++
++ outb(0x0b,0xa0);
++ outb(0x0b,0x20);
++ v = inb(0xa0) << 8 | inb(0x20);
++ outb(0x0a,0xa0);
++ outb(0x0a,0x20);
++
++ spin_unlock_irqrestore(&i8259A_lock, flags);
++
++ printk(KERN_DEBUG "... PIC ISR: %04x\n", v);
++
++ v = inb(0x4d1) << 8 | inb(0x4d0);
++ printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
++}
++
++#endif /* 0 */
++
++#else
++void __init print_IO_APIC(void) { }
++#endif /* !CONFIG_XEN */
++
++static void __init enable_IO_APIC(void)
++{
++ union IO_APIC_reg_01 reg_01;
++ int i8259_apic, i8259_pin;
++ int i, apic;
++ unsigned long flags;
++
++ for (i = 0; i < PIN_MAP_SIZE; i++) {
++ irq_2_pin[i].pin = -1;
++ irq_2_pin[i].next = 0;
++ }
++ if (!pirqs_enabled)
++ for (i = 0; i < MAX_PIRQS; i++)
++ pirq_entries[i] = -1;
++
++ /*
++ * The number of IO-APIC IRQ registers (== #pins):
++ */
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(apic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ nr_ioapic_registers[apic] = reg_01.bits.entries+1;
++ }
++ for(apic = 0; apic < nr_ioapics; apic++) {
++ int pin;
++ /* See if any of the pins is in ExtINT mode */
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++ struct IO_APIC_route_entry entry;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ *(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++
++ /* If the interrupt line is enabled and in ExtInt mode
++ * I have found the pin where the i8259 is connected.
++ */
++ if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) {
++ ioapic_i8259.apic = apic;
++ ioapic_i8259.pin = pin;
++ goto found_i8259;
++ }
++ }
++ }
++ found_i8259:
++ /* Look to see what if the MP table has reported the ExtINT */
++ /* If we could not find the appropriate pin by looking at the ioapic
++ * the i8259 probably is not connected the ioapic but give the
++ * mptable a chance anyway.
++ */
++ i8259_pin = find_isa_irq_pin(0, mp_ExtINT);
++ i8259_apic = find_isa_irq_apic(0, mp_ExtINT);
++ /* Trust the MP table if nothing is setup in the hardware */
++ if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) {
++ printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n");
++ ioapic_i8259.pin = i8259_pin;
++ ioapic_i8259.apic = i8259_apic;
++ }
++ /* Complain if the MP table and the hardware disagree */
++ if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) &&
++ (i8259_pin >= 0) && (ioapic_i8259.pin >= 0))
++ {
++ printk(KERN_WARNING "ExtINT in hardware and MP table differ\n");
++ }
++
++ /*
++ * Do not trust the IO-APIC being empty at bootup
++ */
++ clear_IO_APIC();
++}
++
++/*
++ * Not an __init, needed by the reboot code
++ */
++void disable_IO_APIC(void)
++{
++ /*
++ * Clear the IO-APIC before rebooting:
++ */
++ clear_IO_APIC();
++
++#ifndef CONFIG_XEN
++ /*
++ * If the i8259 is routed through an IOAPIC
++ * Put that IOAPIC in virtual wire mode
++ * so legacy interrupts can be delivered.
++ */
++ if (ioapic_i8259.pin != -1) {
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ memset(&entry, 0, sizeof(entry));
++ entry.mask = 0; /* Enabled */
++ entry.trigger = 0; /* Edge */
++ entry.irr = 0;
++ entry.polarity = 0; /* High */
++ entry.delivery_status = 0;
++ entry.dest_mode = 0; /* Physical */
++ entry.delivery_mode = dest_ExtINT; /* ExtInt */
++ entry.vector = 0;
++ entry.dest.physical.physical_dest =
++ GET_APIC_ID(apic_read(APIC_ID));
++
++ /*
++ * Add it to the IO-APIC irq-routing table:
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
++ *(((int *)&entry)+1));
++ io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
++ *(((int *)&entry)+0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ }
++ disconnect_bsp_APIC(ioapic_i8259.pin != -1);
++#endif
++}
++
++/*
++ * function to set the IO-APIC physical IDs based on the
++ * values stored in the MPC table.
++ *
++ * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999
++ */
++
++#if !defined(CONFIG_XEN) && !defined(CONFIG_X86_NUMAQ)
++static void __init setup_ioapic_ids_from_mpc(void)
++{
++ union IO_APIC_reg_00 reg_00;
++ physid_mask_t phys_id_present_map;
++ int apic;
++ int i;
++ unsigned char old_id;
++ unsigned long flags;
++
++ /*
++ * Don't check I/O APIC IDs for xAPIC systems. They have
++ * no meaning without the serial APIC bus.
++ */
++ if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
++ || APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
++ return;
++ /*
++ * This is broken; anything with a real cpu count has to
++ * circumvent this idiocy regardless.
++ */
++ phys_id_present_map = ioapic_phys_id_map(phys_cpu_present_map);
++
++ /*
++ * Set the IOAPIC ID to the value stored in the MPC table.
++ */
++ for (apic = 0; apic < nr_ioapics; apic++) {
++
++ /* Read the register 0 value */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ old_id = mp_ioapics[apic].mpc_apicid;
++
++ if (mp_ioapics[apic].mpc_apicid >= get_physical_broadcast()) {
++ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID is %d in the MPC table!...\n",
++ apic, mp_ioapics[apic].mpc_apicid);
++ printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
++ reg_00.bits.ID);
++ mp_ioapics[apic].mpc_apicid = reg_00.bits.ID;
++ }
++
++ /*
++ * Sanity check, is the ID really free? Every APIC in a
++ * system must have a unique ID or we get lots of nice
++ * 'stuck on smp_invalidate_needed IPI wait' messages.
++ */
++ if (check_apicid_used(phys_id_present_map,
++ mp_ioapics[apic].mpc_apicid)) {
++ printk(KERN_ERR "BIOS bug, IO-APIC#%d ID %d is already used!...\n",
++ apic, mp_ioapics[apic].mpc_apicid);
++ for (i = 0; i < get_physical_broadcast(); i++)
++ if (!physid_isset(i, phys_id_present_map))
++ break;
++ if (i >= get_physical_broadcast())
++ panic("Max APIC ID exceeded!\n");
++ printk(KERN_ERR "... fixing up to %d. (tell your hw vendor)\n",
++ i);
++ physid_set(i, phys_id_present_map);
++ mp_ioapics[apic].mpc_apicid = i;
++ } else {
++ physid_mask_t tmp;
++ tmp = apicid_to_cpu_present(mp_ioapics[apic].mpc_apicid);
++ apic_printk(APIC_VERBOSE, "Setting %d in the "
++ "phys_id_present_map\n",
++ mp_ioapics[apic].mpc_apicid);
++ physids_or(phys_id_present_map, phys_id_present_map, tmp);
++ }
++
++
++ /*
++ * We need to adjust the IRQ routing table
++ * if the ID changed.
++ */
++ if (old_id != mp_ioapics[apic].mpc_apicid)
++ for (i = 0; i < mp_irq_entries; i++)
++ if (mp_irqs[i].mpc_dstapic == old_id)
++ mp_irqs[i].mpc_dstapic
++ = mp_ioapics[apic].mpc_apicid;
++
++ /*
++ * Read the right value from the MPC table and
++ * write it into the ID register.
++ */
++ apic_printk(APIC_VERBOSE, KERN_INFO
++ "...changing IO-APIC physical APIC ID to %d ...",
++ mp_ioapics[apic].mpc_apicid);
++
++ reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0, reg_00.raw);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ /*
++ * Sanity check
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
++ printk("could not set ID!\n");
++ else
++ apic_printk(APIC_VERBOSE, " ok.\n");
++ }
++}
++#else
++static void __init setup_ioapic_ids_from_mpc(void) { }
++#endif
++
++#ifndef CONFIG_XEN
++/*
++ * There is a nasty bug in some older SMP boards, their mptable lies
++ * about the timer IRQ. We do the following to work around the situation:
++ *
++ * - timer IRQ defaults to IO-APIC IRQ
++ * - if this function detects that timer IRQs are defunct, then we fall
++ * back to ISA timer IRQs
++ */
++static int __init timer_irq_works(void)
++{
++ unsigned long t1 = jiffies;
++
++ local_irq_enable();
++ /* Let ten ticks pass... */
++ mdelay((10 * 1000) / HZ);
++
++ /*
++ * Expect a few ticks at least, to be sure some possible
++ * glue logic does not lock up after one or two first
++ * ticks in a non-ExtINT mode. Also the local APIC
++ * might have cached one ExtINT interrupt. Finally, at
++ * least one tick may be lost due to delays.
++ */
++ if (jiffies - t1 > 4)
++ return 1;
++
++ return 0;
++}
++
++/*
++ * In the SMP+IOAPIC case it might happen that there are an unspecified
++ * number of pending IRQ events unhandled. These cases are very rare,
++ * so we 'resend' these IRQs via IPIs, to the same CPU. It's much
++ * better to do it this way as thus we do not have to be aware of
++ * 'pending' interrupts in the IRQ path, except at this point.
++ */
++/*
++ * Edge triggered needs to resend any interrupt
++ * that was delayed but this is now handled in the device
++ * independent code.
++ */
++
++/*
++ * Starting up a edge-triggered IO-APIC interrupt is
++ * nasty - we need to make sure that we get the edge.
++ * If it is already asserted for some reason, we need
++ * return 1 to indicate that is was pending.
++ *
++ * This is not complete - we should be able to fake
++ * an edge even if it isn't on the 8259A...
++ */
++static unsigned int startup_edge_ioapic_irq(unsigned int irq)
++{
++ int was_pending = 0;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ if (irq < 16) {
++ disable_8259A_irq(irq);
++ if (i8259A_irq_pending(irq))
++ was_pending = 1;
++ }
++ __unmask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return was_pending;
++}
++
++/*
++ * Once we have recorded IRQ_PENDING already, we can mask the
++ * interrupt for real. This prevents IRQ storms from unhandled
++ * devices.
++ */
++static void ack_edge_ioapic_irq(unsigned int irq)
++{
++ move_irq(irq);
++ if ((irq_desc[irq].status & (IRQ_PENDING | IRQ_DISABLED))
++ == (IRQ_PENDING | IRQ_DISABLED))
++ mask_IO_APIC_irq(irq);
++ ack_APIC_irq();
++}
++
++/*
++ * Level triggered interrupts can just be masked,
++ * and shutting down and starting up the interrupt
++ * is the same as enabling and disabling them -- except
++ * with a startup need to return a "was pending" value.
++ *
++ * Level triggered interrupts are special because we
++ * do not touch any IO-APIC register while handling
++ * them. We ack the APIC in the end-IRQ handler, not
++ * in the start-IRQ-handler. Protection against reentrance
++ * from the same interrupt is still provided, both by the
++ * generic IRQ layer and by the fact that an unacked local
++ * APIC does not accept IRQs.
++ */
++static unsigned int startup_level_ioapic_irq (unsigned int irq)
++{
++ unmask_IO_APIC_irq(irq);
++
++ return 0; /* don't check for pending */
++}
++
++static void end_level_ioapic_irq (unsigned int irq)
++{
++ unsigned long v;
++ int i;
++
++ move_irq(irq);
++/*
++ * It appears there is an erratum which affects at least version 0x11
++ * of I/O APIC (that's the 82093AA and cores integrated into various
++ * chipsets). Under certain conditions a level-triggered interrupt is
++ * erroneously delivered as edge-triggered one but the respective IRR
++ * bit gets set nevertheless. As a result the I/O unit expects an EOI
++ * message but it will never arrive and further interrupts are blocked
++ * from the source. The exact reason is so far unknown, but the
++ * phenomenon was observed when two consecutive interrupt requests
++ * from a given source get delivered to the same CPU and the source is
++ * temporarily disabled in between.
++ *
++ * A workaround is to simulate an EOI message manually. We achieve it
++ * by setting the trigger mode to edge and then to level when the edge
++ * trigger mode gets detected in the TMR of a local APIC for a
++ * level-triggered interrupt. We mask the source for the time of the
++ * operation to prevent an edge-triggered interrupt escaping meanwhile.
++ * The idea is from Manfred Spraul. --macro
++ */
++ i = IO_APIC_VECTOR(irq);
++
++ v = apic_read(APIC_TMR + ((i & ~0x1f) >> 1));
++
++ ack_APIC_irq();
++
++ if (!(v & (1 << (i & 0x1f)))) {
++ atomic_inc(&irq_mis_count);
++ spin_lock(&ioapic_lock);
++ __mask_and_edge_IO_APIC_irq(irq);
++ __unmask_and_level_IO_APIC_irq(irq);
++ spin_unlock(&ioapic_lock);
++ }
++}
++
++#ifdef CONFIG_PCI_MSI
++static unsigned int startup_edge_ioapic_vector(unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ return startup_edge_ioapic_irq(irq);
++}
++
++static void ack_edge_ioapic_vector(unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ move_native_irq(vector);
++ ack_edge_ioapic_irq(irq);
++}
++
++static unsigned int startup_level_ioapic_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ return startup_level_ioapic_irq (irq);
++}
++
++static void end_level_ioapic_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ move_native_irq(vector);
++ end_level_ioapic_irq(irq);
++}
++
++static void mask_IO_APIC_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ mask_IO_APIC_irq(irq);
++}
++
++static void unmask_IO_APIC_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ unmask_IO_APIC_irq(irq);
++}
++
++#ifdef CONFIG_SMP
++static void set_ioapic_affinity_vector (unsigned int vector,
++ cpumask_t cpu_mask)
++{
++ int irq = vector_to_irq(vector);
++
++ set_native_irq_info(vector, cpu_mask);
++ set_ioapic_affinity_irq(irq, cpu_mask);
++}
++#endif
++#endif
++
++/*
++ * Level and edge triggered IO-APIC interrupts need different handling,
++ * so we use two separate IRQ descriptors. Edge triggered IRQs can be
++ * handled with the level-triggered descriptor, but that one has slightly
++ * more overhead. Level-triggered interrupts cannot be handled with the
++ * edge-triggered handler, without risking IRQ storms and other ugly
++ * races.
++ */
++static struct hw_interrupt_type ioapic_edge_type __read_mostly = {
++ .typename = "IO-APIC-edge",
++ .startup = startup_edge_ioapic,
++ .shutdown = shutdown_edge_ioapic,
++ .enable = enable_edge_ioapic,
++ .disable = disable_edge_ioapic,
++ .ack = ack_edge_ioapic,
++ .end = end_edge_ioapic,
++#ifdef CONFIG_SMP
++ .set_affinity = set_ioapic_affinity,
++#endif
++};
++
++static struct hw_interrupt_type ioapic_level_type __read_mostly = {
++ .typename = "IO-APIC-level",
++ .startup = startup_level_ioapic,
++ .shutdown = shutdown_level_ioapic,
++ .enable = enable_level_ioapic,
++ .disable = disable_level_ioapic,
++ .ack = mask_and_ack_level_ioapic,
++ .end = end_level_ioapic,
++#ifdef CONFIG_SMP
++ .set_affinity = set_ioapic_affinity,
++#endif
++};
++#endif /* !CONFIG_XEN */
++
++static inline void init_IO_APIC_traps(void)
++{
++ int irq;
++
++ /*
++ * NOTE! The local APIC isn't very good at handling
++ * multiple interrupts at the same interrupt level.
++ * As the interrupt level is determined by taking the
++ * vector number and shifting that right by 4, we
++ * want to spread these out a bit so that they don't
++ * all fall in the same interrupt level.
++ *
++ * Also, we've got to be careful not to trash gate
++ * 0x80, because int 0x80 is hm, kind of importantish. ;)
++ */
++ for (irq = 0; irq < NR_IRQS ; irq++) {
++ int tmp = irq;
++ if (use_pci_vector()) {
++ if (!platform_legacy_irq(tmp))
++ if ((tmp = vector_to_irq(tmp)) == -1)
++ continue;
++ }
++ if (IO_APIC_IRQ(tmp) && !IO_APIC_VECTOR(tmp)) {
++ /*
++ * Hmm.. We don't have an entry for this,
++ * so default to an old-fashioned 8259
++ * interrupt if we can..
++ */
++ if (irq < 16)
++ make_8259A_irq(irq);
++#ifndef CONFIG_XEN
++ else
++ /* Strange. Oh, well.. */
++ irq_desc[irq].handler = &no_irq_type;
++#endif
++ }
++ }
++}
++
++#ifndef CONFIG_XEN
++static void enable_lapic_irq (unsigned int irq)
++{
++ unsigned long v;
++
++ v = apic_read(APIC_LVT0);
++ apic_write_around(APIC_LVT0, v & ~APIC_LVT_MASKED);
++}
++
++static void disable_lapic_irq (unsigned int irq)
++{
++ unsigned long v;
++
++ v = apic_read(APIC_LVT0);
++ apic_write_around(APIC_LVT0, v | APIC_LVT_MASKED);
++}
++
++static void ack_lapic_irq (unsigned int irq)
++{
++ ack_APIC_irq();
++}
++
++static void end_lapic_irq (unsigned int i) { /* nothing */ }
++
++static struct hw_interrupt_type lapic_irq_type __read_mostly = {
++ .typename = "local-APIC-edge",
++ .startup = NULL, /* startup_irq() not used for IRQ0 */
++ .shutdown = NULL, /* shutdown_irq() not used for IRQ0 */
++ .enable = enable_lapic_irq,
++ .disable = disable_lapic_irq,
++ .ack = ack_lapic_irq,
++ .end = end_lapic_irq
++};
++
++static void setup_nmi (void)
++{
++ /*
++ * Dirty trick to enable the NMI watchdog ...
++ * We put the 8259A master into AEOI mode and
++ * unmask on all local APICs LVT0 as NMI.
++ *
++ * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
++ * is from Maciej W. Rozycki - so we do not have to EOI from
++ * the NMI handler or the timer interrupt.
++ */
++ apic_printk(APIC_VERBOSE, KERN_INFO "activating NMI Watchdog ...");
++
++ on_each_cpu(enable_NMI_through_LVT0, NULL, 1, 1);
++
++ apic_printk(APIC_VERBOSE, " done.\n");
++}
++
++/*
++ * This looks a bit hackish but it's about the only one way of sending
++ * a few INTA cycles to 8259As and any associated glue logic. ICR does
++ * not support the ExtINT mode, unfortunately. We need to send these
++ * cycles as some i82489DX-based boards have glue logic that keeps the
++ * 8259A interrupt line asserted until INTA. --macro
++ */
++static inline void unlock_ExtINT_logic(void)
++{
++ int apic, pin, i;
++ struct IO_APIC_route_entry entry0, entry1;
++ unsigned char save_control, save_freq_select;
++ unsigned long flags;
++
++ pin = find_isa_irq_pin(8, mp_INT);
++ apic = find_isa_irq_apic(8, mp_INT);
++ if (pin == -1)
++ return;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ *(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ clear_IO_APIC_pin(apic, pin);
++
++ memset(&entry1, 0, sizeof(entry1));
++
++ entry1.dest_mode = 0; /* physical delivery */
++ entry1.mask = 0; /* unmask IRQ now */
++ entry1.dest.physical.physical_dest = hard_smp_processor_id();
++ entry1.delivery_mode = dest_ExtINT;
++ entry1.polarity = entry0.polarity;
++ entry1.trigger = 0;
++ entry1.vector = 0;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ save_control = CMOS_READ(RTC_CONTROL);
++ save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
++ CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6,
++ RTC_FREQ_SELECT);
++ CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL);
++
++ i = 100;
++ while (i-- > 0) {
++ mdelay(10);
++ if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF)
++ i -= 10;
++ }
++
++ CMOS_WRITE(save_control, RTC_CONTROL);
++ CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
++ clear_IO_APIC_pin(apic, pin);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++int timer_uses_ioapic_pin_0;
++
++/*
++ * This code may look a bit paranoid, but it's supposed to cooperate with
++ * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ
++ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast
++ * fanatically on his truly buggy board.
++ */
++static inline void check_timer(void)
++{
++ int apic1, pin1, apic2, pin2;
++ int vector;
++
++ /*
++ * get/set the timer IRQ vector:
++ */
++ disable_8259A_irq(0);
++ vector = assign_irq_vector(0);
++ set_intr_gate(vector, interrupt[0]);
++
++ /*
++ * Subtle, code in do_timer_interrupt() expects an AEOI
++ * mode for the 8259A whenever interrupts are routed
++ * through I/O APICs. Also IRQ0 has to be enabled in
++ * the 8259A which implies the virtual wire has to be
++ * disabled in the local APIC.
++ */
++ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
++ init_8259A(1);
++ timer_ack = 1;
++ if (timer_over_8254 > 0)
++ enable_8259A_irq(0);
++
++ pin1 = find_isa_irq_pin(0, mp_INT);
++ apic1 = find_isa_irq_apic(0, mp_INT);
++ pin2 = ioapic_i8259.pin;
++ apic2 = ioapic_i8259.apic;
++
++ if (pin1 == 0)
++ timer_uses_ioapic_pin_0 = 1;
++
++ printk(KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
++ vector, apic1, pin1, apic2, pin2);
++
++ if (pin1 != -1) {
++ /*
++ * Ok, does IRQ0 through the IOAPIC work?
++ */
++ unmask_IO_APIC_irq(0);
++ if (timer_irq_works()) {
++ if (nmi_watchdog == NMI_IO_APIC) {
++ disable_8259A_irq(0);
++ setup_nmi();
++ enable_8259A_irq(0);
++ }
++ if (disable_timer_pin_1 > 0)
++ clear_IO_APIC_pin(0, pin1);
++ return;
++ }
++ clear_IO_APIC_pin(apic1, pin1);
++ printk(KERN_ERR "..MP-BIOS bug: 8254 timer not connected to "
++ "IO-APIC\n");
++ }
++
++ printk(KERN_INFO "...trying to set up timer (IRQ0) through the 8259A ... ");
++ if (pin2 != -1) {
++ printk("\n..... (found pin %d) ...", pin2);
++ /*
++ * legacy devices should be connected to IO APIC #0
++ */
++ setup_ExtINT_IRQ0_pin(apic2, pin2, vector);
++ if (timer_irq_works()) {
++ printk("works.\n");
++ if (pin1 != -1)
++ replace_pin_at_irq(0, apic1, pin1, apic2, pin2);
++ else
++ add_pin_to_irq(0, apic2, pin2);
++ if (nmi_watchdog == NMI_IO_APIC) {
++ setup_nmi();
++ }
++ return;
++ }
++ /*
++ * Cleanup, just in case ...
++ */
++ clear_IO_APIC_pin(apic2, pin2);
++ }
++ printk(" failed.\n");
++
++ if (nmi_watchdog == NMI_IO_APIC) {
++ printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
++ nmi_watchdog = 0;
++ }
++
++ printk(KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
++
++ disable_8259A_irq(0);
++ irq_desc[0].handler = &lapic_irq_type;
++ apic_write_around(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */
++ enable_8259A_irq(0);
++
++ if (timer_irq_works()) {
++ printk(" works.\n");
++ return;
++ }
++ apic_write_around(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
++ printk(" failed.\n");
++
++ printk(KERN_INFO "...trying to set up timer as ExtINT IRQ...");
++
++ timer_ack = 0;
++ init_8259A(0);
++ make_8259A_irq(0);
++ apic_write_around(APIC_LVT0, APIC_DM_EXTINT);
++
++ unlock_ExtINT_logic();
++
++ if (timer_irq_works()) {
++ printk(" works.\n");
++ return;
++ }
++ printk(" failed :(.\n");
++ panic("IO-APIC + timer doesn't work! Boot with apic=debug and send a "
++ "report. Then try booting with the 'noapic' option");
++}
++#else
++int timer_uses_ioapic_pin_0;
++#define check_timer() ((void)0)
++#endif
++
++/*
++ *
++ * IRQ's that are handled by the PIC in the MPS IOAPIC case.
++ * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
++ * Linux doesn't really care, as it's not actually used
++ * for any interrupt handling anyway.
++ */
++#define PIC_IRQS (1 << PIC_CASCADE_IR)
++
++void __init setup_IO_APIC(void)
++{
++ enable_IO_APIC();
++
++ if (acpi_ioapic)
++ io_apic_irqs = ~0; /* all IRQs go through IOAPIC */
++ else
++ io_apic_irqs = ~PIC_IRQS;
++
++ printk("ENABLING IO-APIC IRQs\n");
++
++ /*
++ * Set up IO-APIC IRQ routing.
++ */
++ if (!acpi_ioapic)
++ setup_ioapic_ids_from_mpc();
++#ifndef CONFIG_XEN
++ sync_Arb_IDs();
++#endif
++ setup_IO_APIC_irqs();
++ init_IO_APIC_traps();
++ check_timer();
++ if (!acpi_ioapic)
++ print_IO_APIC();
++}
++
++static int __init setup_disable_8254_timer(char *s)
++{
++ timer_over_8254 = -1;
++ return 1;
++}
++static int __init setup_enable_8254_timer(char *s)
++{
++ timer_over_8254 = 2;
++ return 1;
++}
++
++__setup("disable_8254_timer", setup_disable_8254_timer);
++__setup("enable_8254_timer", setup_enable_8254_timer);
++
++/*
++ * Called after all the initialization is done. If we didnt find any
++ * APIC bugs then we can allow the modify fast path
++ */
++
++static int __init io_apic_bug_finalize(void)
++{
++ if(sis_apic_bug == -1)
++ sis_apic_bug = 0;
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ dom0_op_t op = { .cmd = DOM0_PLATFORM_QUIRK };
++ op.u.platform_quirk.quirk_id = sis_apic_bug ?
++ QUIRK_IOAPIC_BAD_REGSEL : QUIRK_IOAPIC_GOOD_REGSEL;
++ HYPERVISOR_dom0_op(&op);
++ }
++ return 0;
++}
++
++late_initcall(io_apic_bug_finalize);
++
++struct sysfs_ioapic_data {
++ struct sys_device dev;
++ struct IO_APIC_route_entry entry[0];
++};
++static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];
++
++static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
++{
++ struct IO_APIC_route_entry *entry;
++ struct sysfs_ioapic_data *data;
++ unsigned long flags;
++ int i;
++
++ data = container_of(dev, struct sysfs_ioapic_data, dev);
++ entry = data->entry;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
++ *(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
++ *(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
++ }
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++static int ioapic_resume(struct sys_device *dev)
++{
++ struct IO_APIC_route_entry *entry;
++ struct sysfs_ioapic_data *data;
++ unsigned long flags;
++ union IO_APIC_reg_00 reg_00;
++ int i;
++
++ data = container_of(dev, struct sysfs_ioapic_data, dev);
++ entry = data->entry;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(dev->id, 0);
++ if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
++ reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
++ io_apic_write(dev->id, 0, reg_00.raw);
++ }
++ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
++ io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
++ io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
++ }
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++static struct sysdev_class ioapic_sysdev_class = {
++ set_kset_name("ioapic"),
++ .suspend = ioapic_suspend,
++ .resume = ioapic_resume,
++};
++
++static int __init ioapic_init_sysfs(void)
++{
++ struct sys_device * dev;
++ int i, size, error = 0;
++
++ error = sysdev_class_register(&ioapic_sysdev_class);
++ if (error)
++ return error;
++
++ for (i = 0; i < nr_ioapics; i++ ) {
++ size = sizeof(struct sys_device) + nr_ioapic_registers[i]
++ * sizeof(struct IO_APIC_route_entry);
++ mp_ioapic_data[i] = kmalloc(size, GFP_KERNEL);
++ if (!mp_ioapic_data[i]) {
++ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
++ continue;
++ }
++ memset(mp_ioapic_data[i], 0, size);
++ dev = &mp_ioapic_data[i]->dev;
++ dev->id = i;
++ dev->cls = &ioapic_sysdev_class;
++ error = sysdev_register(dev);
++ if (error) {
++ kfree(mp_ioapic_data[i]);
++ mp_ioapic_data[i] = NULL;
++ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
++ continue;
++ }
++ }
++
++ return 0;
++}
++
++device_initcall(ioapic_init_sysfs);
++
++/* --------------------------------------------------------------------------
++ ACPI-based IOAPIC Configuration
++ -------------------------------------------------------------------------- */
++
++#ifdef CONFIG_ACPI
++
++int __init io_apic_get_unique_id (int ioapic, int apic_id)
++{
++#ifndef CONFIG_XEN
++ union IO_APIC_reg_00 reg_00;
++ static physid_mask_t apic_id_map = PHYSID_MASK_NONE;
++ physid_mask_t tmp;
++ unsigned long flags;
++ int i = 0;
++
++ /*
++ * The P4 platform supports up to 256 APIC IDs on two separate APIC
++ * buses (one for LAPICs, one for IOAPICs), where predecessors only
++ * supports up to 16 on one shared APIC bus.
++ *
++ * TBD: Expand LAPIC/IOAPIC support on P4-class systems to take full
++ * advantage of new APIC bus architecture.
++ */
++
++ if (physids_empty(apic_id_map))
++ apic_id_map = ioapic_phys_id_map(phys_cpu_present_map);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(ioapic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ if (apic_id >= get_physical_broadcast()) {
++ printk(KERN_WARNING "IOAPIC[%d]: Invalid apic_id %d, trying "
++ "%d\n", ioapic, apic_id, reg_00.bits.ID);
++ apic_id = reg_00.bits.ID;
++ }
++
++ /*
++ * Every APIC in a system must have a unique ID or we get lots of nice
++ * 'stuck on smp_invalidate_needed IPI wait' messages.
++ */
++ if (check_apicid_used(apic_id_map, apic_id)) {
++
++ for (i = 0; i < get_physical_broadcast(); i++) {
++ if (!check_apicid_used(apic_id_map, i))
++ break;
++ }
++
++ if (i == get_physical_broadcast())
++ panic("Max apic_id exceeded!\n");
++
++ printk(KERN_WARNING "IOAPIC[%d]: apic_id %d already used, "
++ "trying %d\n", ioapic, apic_id, i);
++
++ apic_id = i;
++ }
++
++ tmp = apicid_to_cpu_present(apic_id);
++ physids_or(apic_id_map, apic_id_map, tmp);
++
++ if (reg_00.bits.ID != apic_id) {
++ reg_00.bits.ID = apic_id;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(ioapic, 0, reg_00.raw);
++ reg_00.raw = io_apic_read(ioapic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ /* Sanity check */
++ if (reg_00.bits.ID != apic_id) {
++ printk("IOAPIC[%d]: Unable to change apic_id!\n", ioapic);
++ return -1;
++ }
++ }
++
++ apic_printk(APIC_VERBOSE, KERN_INFO
++ "IOAPIC[%d]: Assigned apic_id %d\n", ioapic, apic_id);
++#endif /* !CONFIG_XEN */
++
++ return apic_id;
++}
++
++
++int __init io_apic_get_version (int ioapic)
++{
++ union IO_APIC_reg_01 reg_01;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(ioapic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return reg_01.bits.version;
++}
++
++
++int __init io_apic_get_redir_entries (int ioapic)
++{
++ union IO_APIC_reg_01 reg_01;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(ioapic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return reg_01.bits.entries;
++}
++
++
++int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ if (!IO_APIC_IRQ(irq)) {
++ printk(KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
++ ioapic);
++ return -EINVAL;
++ }
++
++ /*
++ * Generate a PCI IRQ routing entry and program the IOAPIC accordingly.
++ * Note that we mask (disable) IRQs now -- these get enabled when the
++ * corresponding device driver registers for this IRQ.
++ */
++
++ memset(&entry,0,sizeof(entry));
++
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.dest_mode = INT_DEST_MODE;
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++ entry.trigger = edge_level;
++ entry.polarity = active_high_low;
++ entry.mask = 1;
++
++ /*
++ * IRQs < 16 are already in the irq_2_pin[] map
++ */
++ if (irq >= 16)
++ add_pin_to_irq(irq, ioapic, pin);
++
++ entry.vector = assign_irq_vector(irq);
++
++ apic_printk(APIC_DEBUG, KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry "
++ "(%d-%d -> 0x%x -> IRQ %d Mode:%i Active:%i)\n", ioapic,
++ mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
++ edge_level, active_high_low);
++
++ ioapic_register_intr(irq, entry.vector, edge_level);
++
++ if (!ioapic && (irq < 16))
++ disable_8259A_irq(irq);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
++ set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++#endif /* CONFIG_ACPI */
+diff -urNp linux-2.6/arch/i386/kernel/ioport-xen.c new/arch/i386/kernel/ioport-xen.c
+--- linux-2.6/arch/i386/kernel/ioport-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/ioport-xen.c 2006-05-09 12:32:35.000000000 +0200
+@@ -0,0 +1,121 @@
++/*
++ * linux/arch/i386/kernel/ioport.c
++ *
++ * This contains the io-permission bitmap code - written by obz, with changes
++ * by Linus.
++ */
++
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/capability.h>
++#include <linux/errno.h>
++#include <linux/types.h>
++#include <linux/ioport.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/stddef.h>
++#include <linux/slab.h>
++#include <linux/thread_info.h>
++#include <xen/interface/physdev.h>
++
++/* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */
++static void set_bitmap(unsigned long *bitmap, unsigned int base, unsigned int extent, int new_value)
++{
++ unsigned long mask;
++ unsigned long *bitmap_base = bitmap + (base / BITS_PER_LONG);
++ unsigned int low_index = base & (BITS_PER_LONG-1);
++ int length = low_index + extent;
++
++ if (low_index != 0) {
++ mask = (~0UL << low_index);
++ if (length < BITS_PER_LONG)
++ mask &= ~(~0UL << length);
++ if (new_value)
++ *bitmap_base++ |= mask;
++ else
++ *bitmap_base++ &= ~mask;
++ length -= BITS_PER_LONG;
++ }
++
++ mask = (new_value ? ~0UL : 0UL);
++ while (length >= BITS_PER_LONG) {
++ *bitmap_base++ = mask;
++ length -= BITS_PER_LONG;
++ }
++
++ if (length > 0) {
++ mask = ~(~0UL << length);
++ if (new_value)
++ *bitmap_base++ |= mask;
++ else
++ *bitmap_base++ &= ~mask;
++ }
++}
++
++
++/*
++ * this changes the io permissions bitmap in the current task.
++ */
++asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
++{
++ struct thread_struct * t = ¤t->thread;
++ unsigned long *bitmap;
++ struct physdev_set_iobitmap set_iobitmap;
++
++ if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
++ return -EINVAL;
++ if (turn_on && !capable(CAP_SYS_RAWIO))
++ return -EPERM;
++
++ /*
++ * If it's the first ioperm() call in this thread's lifetime, set the
++ * IO bitmap up. ioperm() is much less timing critical than clone(),
++ * this is why we delay this operation until now:
++ */
++ if (!t->io_bitmap_ptr) {
++ bitmap = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
++ if (!bitmap)
++ return -ENOMEM;
++
++ memset(bitmap, 0xff, IO_BITMAP_BYTES);
++ t->io_bitmap_ptr = bitmap;
++
++ set_iobitmap.bitmap = (char *)bitmap;
++ set_iobitmap.nr_ports = IO_BITMAP_BITS;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap, &set_iobitmap);
++ }
++
++ set_bitmap(t->io_bitmap_ptr, from, num, !turn_on);
++
++ return 0;
++}
++
++/*
++ * sys_iopl has to be used when you want to access the IO ports
++ * beyond the 0x3ff range: to get the full 65536 ports bitmapped
++ * you'd need 8kB of bitmaps/process, which is a bit excessive.
++ *
++ * Here we just change the eflags value on the stack: we allow
++ * only the super-user to do it. This depends on the stack-layout
++ * on system-call entry - see also fork() and the signal handling
++ * code.
++ */
++
++asmlinkage long sys_iopl(unsigned long unused)
++{
++ volatile struct pt_regs * regs = (struct pt_regs *) &unused;
++ unsigned int level = regs->ebx;
++ struct thread_struct *t = ¤t->thread;
++ unsigned int old = (t->iopl >> 12) & 3;
++
++ if (level > 3)
++ return -EINVAL;
++ /* Trying to gain more privileges? */
++ if (level > old) {
++ if (!capable(CAP_SYS_RAWIO))
++ return -EPERM;
++ }
++ t->iopl = level << 12;
++ set_iopl_mask(t->iopl);
++ return 0;
++}
+diff -urNp linux-2.6/arch/i386/kernel/irq.c new/arch/i386/kernel/irq.c
+--- linux-2.6/arch/i386/kernel/irq.c 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/irq.c 2006-05-09 12:32:35.000000000 +0200
+@@ -53,8 +53,8 @@ static union irq_ctx *softirq_ctx[NR_CPU
+ */
+ fastcall unsigned int do_IRQ(struct pt_regs *regs)
+ {
+- /* high bits used in ret_from_ code */
+- int irq = regs->orig_eax & 0xff;
++ /* high bit used in ret_from_ code */
++ int irq = ~regs->orig_eax;
+ #ifdef CONFIG_4KSTACKS
+ union irq_ctx *curctx, *irqctx;
+ u32 *isp;
+diff -urNp linux-2.6/arch/i386/kernel/irq-xen.c new/arch/i386/kernel/irq-xen.c
+--- linux-2.6/arch/i386/kernel/irq-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/irq-xen.c 2006-05-09 12:32:35.000000000 +0200
+@@ -0,0 +1,306 @@
++/*
++ * linux/arch/i386/kernel/irq.c
++ *
++ * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
++ *
++ * This file contains the lowest level x86-specific interrupt
++ * entry, irq-stacks and irq statistics code. All the remaining
++ * irq logic is done by the generic kernel/irq/ code and
++ * by the x86-specific irq controller code. (e.g. i8259.c and
++ * io_apic.c.)
++ */
++
++#include <asm/uaccess.h>
++#include <linux/module.h>
++#include <linux/seq_file.h>
++#include <linux/interrupt.h>
++#include <linux/kernel_stat.h>
++#include <linux/notifier.h>
++#include <linux/cpu.h>
++#include <linux/delay.h>
++
++DEFINE_PER_CPU(irq_cpustat_t, irq_stat) ____cacheline_internodealigned_in_smp;
++EXPORT_PER_CPU_SYMBOL(irq_stat);
++
++#ifndef CONFIG_X86_LOCAL_APIC
++/*
++ * 'what should we do if we get a hw irq event on an illegal vector'.
++ * each architecture has to answer this themselves.
++ */
++void ack_bad_irq(unsigned int irq)
++{
++ printk("unexpected IRQ trap at vector %02x\n", irq);
++}
++#endif
++
++#ifdef CONFIG_4KSTACKS
++/*
++ * per-CPU IRQ handling contexts (thread information and stack)
++ */
++union irq_ctx {
++ struct thread_info tinfo;
++ u32 stack[THREAD_SIZE/sizeof(u32)];
++};
++
++static union irq_ctx *hardirq_ctx[NR_CPUS];
++static union irq_ctx *softirq_ctx[NR_CPUS];
++#endif
++
++/*
++ * do_IRQ handles all normal device IRQ's (the special
++ * SMP cross-CPU interrupts have their own specific
++ * handlers).
++ */
++fastcall unsigned int do_IRQ(struct pt_regs *regs)
++{
++ /* high bit used in ret_from_ code */
++ int irq = ~regs->orig_eax;
++#ifdef CONFIG_4KSTACKS
++ union irq_ctx *curctx, *irqctx;
++ u32 *isp;
++#endif
++
++ irq_enter();
++#ifdef CONFIG_DEBUG_STACKOVERFLOW
++ /* Debugging check for stack overflow: is there less than 1KB free? */
++ {
++ long esp;
++
++ __asm__ __volatile__("andl %%esp,%0" :
++ "=r" (esp) : "0" (THREAD_SIZE - 1));
++ if (unlikely(esp < (sizeof(struct thread_info) + STACK_WARN))) {
++ printk("do_IRQ: stack overflow: %ld\n",
++ esp - sizeof(struct thread_info));
++ dump_stack();
++ }
++ }
++#endif
++
++#ifdef CONFIG_4KSTACKS
++
++ curctx = (union irq_ctx *) current_thread_info();
++ irqctx = hardirq_ctx[smp_processor_id()];
++
++ /*
++ * this is where we switch to the IRQ stack. However, if we are
++ * already using the IRQ stack (because we interrupted a hardirq
++ * handler) we can't do that and just have to keep using the
++ * current stack (which is the irq stack already after all)
++ */
++ if (curctx != irqctx) {
++ int arg1, arg2, ebx;
++
++ /* build the stack frame on the IRQ stack */
++ isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
++ irqctx->tinfo.task = curctx->tinfo.task;
++ irqctx->tinfo.previous_esp = current_stack_pointer;
++
++ asm volatile(
++ " xchgl %%ebx,%%esp \n"
++ " call __do_IRQ \n"
++ " movl %%ebx,%%esp \n"
++ : "=a" (arg1), "=d" (arg2), "=b" (ebx)
++ : "0" (irq), "1" (regs), "2" (isp)
++ : "memory", "cc", "ecx"
++ );
++ } else
++#endif
++ __do_IRQ(irq, regs);
++
++ irq_exit();
++
++ return 1;
++}
++
++#ifdef CONFIG_4KSTACKS
++
++/*
++ * These should really be __section__(".bss.page_aligned") as well, but
++ * gcc's 3.0 and earlier don't handle that correctly.
++ */
++static char softirq_stack[NR_CPUS * THREAD_SIZE]
++ __attribute__((__aligned__(THREAD_SIZE)));
++
++static char hardirq_stack[NR_CPUS * THREAD_SIZE]
++ __attribute__((__aligned__(THREAD_SIZE)));
++
++/*
++ * allocate per-cpu stacks for hardirq and for softirq processing
++ */
++void irq_ctx_init(int cpu)
++{
++ union irq_ctx *irqctx;
++
++ if (hardirq_ctx[cpu])
++ return;
++
++ irqctx = (union irq_ctx*) &hardirq_stack[cpu*THREAD_SIZE];
++ irqctx->tinfo.task = NULL;
++ irqctx->tinfo.exec_domain = NULL;
++ irqctx->tinfo.cpu = cpu;
++ irqctx->tinfo.preempt_count = HARDIRQ_OFFSET;
++ irqctx->tinfo.addr_limit = MAKE_MM_SEG(0);
++
++ hardirq_ctx[cpu] = irqctx;
++
++ irqctx = (union irq_ctx*) &softirq_stack[cpu*THREAD_SIZE];
++ irqctx->tinfo.task = NULL;
++ irqctx->tinfo.exec_domain = NULL;
++ irqctx->tinfo.cpu = cpu;
++ irqctx->tinfo.preempt_count = SOFTIRQ_OFFSET;
++ irqctx->tinfo.addr_limit = MAKE_MM_SEG(0);
++
++ softirq_ctx[cpu] = irqctx;
++
++ printk("CPU %u irqstacks, hard=%p soft=%p\n",
++ cpu,hardirq_ctx[cpu],softirq_ctx[cpu]);
++}
++
++void irq_ctx_exit(int cpu)
++{
++ hardirq_ctx[cpu] = NULL;
++}
++
++extern asmlinkage void __do_softirq(void);
++
++asmlinkage void do_softirq(void)
++{
++ unsigned long flags;
++ struct thread_info *curctx;
++ union irq_ctx *irqctx;
++ u32 *isp;
++
++ if (in_interrupt())
++ return;
++
++ local_irq_save(flags);
++
++ if (local_softirq_pending()) {
++ curctx = current_thread_info();
++ irqctx = softirq_ctx[smp_processor_id()];
++ irqctx->tinfo.task = curctx->task;
++ irqctx->tinfo.previous_esp = current_stack_pointer;
++
++ /* build the stack frame on the softirq stack */
++ isp = (u32*) ((char*)irqctx + sizeof(*irqctx));
++
++ asm volatile(
++ " xchgl %%ebx,%%esp \n"
++ " call __do_softirq \n"
++ " movl %%ebx,%%esp \n"
++ : "=b"(isp)
++ : "0"(isp)
++ : "memory", "cc", "edx", "ecx", "eax"
++ );
++ }
++
++ local_irq_restore(flags);
++}
++
++EXPORT_SYMBOL(do_softirq);
++#endif
++
++/*
++ * Interrupt statistics:
++ */
++
++atomic_t irq_err_count;
++
++/*
++ * /proc/interrupts printing:
++ */
++
++int show_interrupts(struct seq_file *p, void *v)
++{
++ int i = *(loff_t *) v, j;
++ struct irqaction * action;
++ unsigned long flags;
++
++ if (i == 0) {
++ seq_printf(p, " ");
++ for_each_online_cpu(j)
++ seq_printf(p, "CPU%d ",j);
++ seq_putc(p, '\n');
++ }
++
++ if (i < NR_IRQS) {
++ spin_lock_irqsave(&irq_desc[i].lock, flags);
++ action = irq_desc[i].action;
++ if (!action)
++ goto skip;
++ seq_printf(p, "%3d: ",i);
++#ifndef CONFIG_SMP
++ seq_printf(p, "%10u ", kstat_irqs(i));
++#else
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
++#endif
++ seq_printf(p, " %14s", irq_desc[i].handler->typename);
++ seq_printf(p, " %s", action->name);
++
++ for (action=action->next; action; action = action->next)
++ seq_printf(p, ", %s", action->name);
++
++ seq_putc(p, '\n');
++skip:
++ spin_unlock_irqrestore(&irq_desc[i].lock, flags);
++ } else if (i == NR_IRQS) {
++ seq_printf(p, "NMI: ");
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ", nmi_count(j));
++ seq_putc(p, '\n');
++#ifdef CONFIG_X86_LOCAL_APIC
++ seq_printf(p, "LOC: ");
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ",
++ per_cpu(irq_stat,j).apic_timer_irqs);
++ seq_putc(p, '\n');
++#endif
++ seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
++#if defined(CONFIG_X86_IO_APIC)
++ seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
++#endif
++ }
++ return 0;
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++
++void fixup_irqs(cpumask_t map)
++{
++ unsigned int irq;
++ static int warned;
++
++ for (irq = 0; irq < NR_IRQS; irq++) {
++ cpumask_t mask;
++ if (irq == 2)
++ continue;
++
++ cpus_and(mask, irq_affinity[irq], map);
++ if (any_online_cpu(mask) == NR_CPUS) {
++ /*printk("Breaking affinity for irq %i\n", irq);*/
++ mask = map;
++ }
++ if (irq_desc[irq].handler->set_affinity)
++ irq_desc[irq].handler->set_affinity(irq, mask);
++ else if (irq_desc[irq].action && !(warned++))
++ printk("Cannot set affinity for irq %i\n", irq);
++ }
++
++#if 0
++ barrier();
++ /* Ingo Molnar says: "after the IO-APIC masks have been redirected
++ [note the nop - the interrupt-enable boundary on x86 is two
++ instructions from sti] - to flush out pending hardirqs and
++ IPIs. After this point nothing is supposed to reach this CPU." */
++ __asm__ __volatile__("sti; nop; cli");
++ barrier();
++#else
++ /* That doesn't seem sufficient. Give it 1ms. */
++ local_irq_enable();
++ mdelay(1);
++ local_irq_disable();
++#endif
++}
++#endif
++
+diff -urNp linux-2.6/arch/i386/kernel/ldt-xen.c new/arch/i386/kernel/ldt-xen.c
+--- linux-2.6/arch/i386/kernel/ldt-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/ldt-xen.c 2006-05-09 12:32:35.000000000 +0200
+@@ -0,0 +1,269 @@
++/*
++ * linux/kernel/ldt.c
++ *
++ * Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds
++ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
++ */
++
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/string.h>
++#include <linux/mm.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/vmalloc.h>
++#include <linux/slab.h>
++
++#include <asm/uaccess.h>
++#include <asm/system.h>
++#include <asm/ldt.h>
++#include <asm/desc.h>
++#include <asm/mmu_context.h>
++
++#ifdef CONFIG_SMP /* avoids "defined but not used" warnig */
++static void flush_ldt(void *null)
++{
++ if (current->active_mm)
++ load_LDT(¤t->active_mm->context);
++}
++#endif
++
++static int alloc_ldt(mm_context_t *pc, int mincount, int reload)
++{
++ void *oldldt;
++ void *newldt;
++ int oldsize;
++
++ if (mincount <= pc->size)
++ return 0;
++ oldsize = pc->size;
++ mincount = (mincount+511)&(~511);
++ if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
++ newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
++ else
++ newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
++
++ if (!newldt)
++ return -ENOMEM;
++
++ if (oldsize)
++ memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE);
++ oldldt = pc->ldt;
++ memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE);
++ pc->ldt = newldt;
++ wmb();
++ pc->size = mincount;
++ wmb();
++
++ if (reload) {
++#ifdef CONFIG_SMP
++ cpumask_t mask;
++ preempt_disable();
++#endif
++ make_pages_readonly(
++ pc->ldt,
++ (pc->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ load_LDT(pc);
++#ifdef CONFIG_SMP
++ mask = cpumask_of_cpu(smp_processor_id());
++ if (!cpus_equal(current->mm->cpu_vm_mask, mask))
++ smp_call_function(flush_ldt, NULL, 1, 1);
++ preempt_enable();
++#endif
++ }
++ if (oldsize) {
++ make_pages_writable(
++ oldldt,
++ (oldsize * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE)
++ vfree(oldldt);
++ else
++ kfree(oldldt);
++ }
++ return 0;
++}
++
++static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
++{
++ int err = alloc_ldt(new, old->size, 0);
++ if (err < 0)
++ return err;
++ memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE);
++ make_pages_readonly(
++ new->ldt,
++ (new->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ return 0;
++}
++
++/*
++ * we do not have to muck with descriptors here, that is
++ * done in switch_mm() as needed.
++ */
++int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
++{
++ struct mm_struct * old_mm;
++ int retval = 0;
++
++ init_MUTEX(&mm->context.sem);
++ mm->context.size = 0;
++ old_mm = current->mm;
++ if (old_mm && old_mm->context.size > 0) {
++ down(&old_mm->context.sem);
++ retval = copy_ldt(&mm->context, &old_mm->context);
++ up(&old_mm->context.sem);
++ }
++ return retval;
++}
++
++/*
++ * No need to lock the MM as we are the last user
++ */
++void destroy_context(struct mm_struct *mm)
++{
++ if (mm->context.size) {
++ if (mm == current->active_mm)
++ clear_LDT();
++ make_pages_writable(
++ mm->context.ldt,
++ (mm->context.size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE)
++ vfree(mm->context.ldt);
++ else
++ kfree(mm->context.ldt);
++ mm->context.size = 0;
++ }
++}
++
++static int read_ldt(void __user * ptr, unsigned long bytecount)
++{
++ int err;
++ unsigned long size;
++ struct mm_struct * mm = current->mm;
++
++ if (!mm->context.size)
++ return 0;
++ if (bytecount > LDT_ENTRY_SIZE*LDT_ENTRIES)
++ bytecount = LDT_ENTRY_SIZE*LDT_ENTRIES;
++
++ down(&mm->context.sem);
++ size = mm->context.size*LDT_ENTRY_SIZE;
++ if (size > bytecount)
++ size = bytecount;
++
++ err = 0;
++ if (copy_to_user(ptr, mm->context.ldt, size))
++ err = -EFAULT;
++ up(&mm->context.sem);
++ if (err < 0)
++ goto error_return;
++ if (size != bytecount) {
++ /* zero-fill the rest */
++ if (clear_user(ptr+size, bytecount-size) != 0) {
++ err = -EFAULT;
++ goto error_return;
++ }
++ }
++ return bytecount;
++error_return:
++ return err;
++}
++
++static int read_default_ldt(void __user * ptr, unsigned long bytecount)
++{
++ int err;
++ unsigned long size;
++ void *address;
++
++ err = 0;
++ address = &default_ldt[0];
++ size = 5*sizeof(struct desc_struct);
++ if (size > bytecount)
++ size = bytecount;
++
++ err = size;
++ if (copy_to_user(ptr, address, size))
++ err = -EFAULT;
++
++ return err;
++}
++
++static int write_ldt(void __user * ptr, unsigned long bytecount, int oldmode)
++{
++ struct mm_struct * mm = current->mm;
++ __u32 entry_1, entry_2;
++ int error;
++ struct user_desc ldt_info;
++
++ error = -EINVAL;
++ if (bytecount != sizeof(ldt_info))
++ goto out;
++ error = -EFAULT;
++ if (copy_from_user(&ldt_info, ptr, sizeof(ldt_info)))
++ goto out;
++
++ error = -EINVAL;
++ if (ldt_info.entry_number >= LDT_ENTRIES)
++ goto out;
++ if (ldt_info.contents == 3) {
++ if (oldmode)
++ goto out;
++ if (ldt_info.seg_not_present == 0)
++ goto out;
++ }
++
++ down(&mm->context.sem);
++ if (ldt_info.entry_number >= mm->context.size) {
++ error = alloc_ldt(¤t->mm->context, ldt_info.entry_number+1, 1);
++ if (error < 0)
++ goto out_unlock;
++ }
++
++ /* Allow LDTs to be cleared by the user. */
++ if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
++ if (oldmode || LDT_empty(&ldt_info)) {
++ entry_1 = 0;
++ entry_2 = 0;
++ goto install;
++ }
++ }
++
++ entry_1 = LDT_entry_a(&ldt_info);
++ entry_2 = LDT_entry_b(&ldt_info);
++ if (oldmode)
++ entry_2 &= ~(1 << 20);
++
++ /* Install the new entry ... */
++install:
++ error = write_ldt_entry(mm->context.ldt, ldt_info.entry_number,
++ entry_1, entry_2);
++
++out_unlock:
++ up(&mm->context.sem);
++out:
++ return error;
++}
++
++asmlinkage int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount)
++{
++ int ret = -ENOSYS;
++
++ switch (func) {
++ case 0:
++ ret = read_ldt(ptr, bytecount);
++ break;
++ case 1:
++ ret = write_ldt(ptr, bytecount, 1);
++ break;
++ case 2:
++ ret = read_default_ldt(ptr, bytecount);
++ break;
++ case 0x11:
++ ret = write_ldt(ptr, bytecount, 0);
++ break;
++ }
++ return ret;
++}
+diff -urNp linux-2.6/arch/i386/kernel/Makefile new/arch/i386/kernel/Makefile
+--- linux-2.6/arch/i386/kernel/Makefile 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/Makefile 2006-05-09 12:32:33.000000000 +0200
+@@ -42,6 +42,12 @@ EXTRA_AFLAGS := -traditional
+
+ obj-$(CONFIG_SCx200) += scx200.o
+
++ifdef CONFIG_XEN
++vsyscall_note := vsyscall-note-xen.o
++else
++vsyscall_note := vsyscall-note.o
++endif
++
+ # vsyscall.o contains the vsyscall DSO images as __initdata.
+ # We must build both images before we can assemble it.
+ # Note: kbuild does not track this dependency due to usage of .incbin
+@@ -62,7 +68,7 @@ SYSCFLAGS_vsyscall-int80.so = $(vsyscall
+
+ $(obj)/vsyscall-int80.so $(obj)/vsyscall-sysenter.so: \
+ $(obj)/vsyscall-%.so: $(src)/vsyscall.lds \
+- $(obj)/vsyscall-%.o $(obj)/vsyscall-note.o FORCE
++ $(obj)/vsyscall-%.o $(obj)/$(vsyscall_note) FORCE
+ $(call if_changed,syscall)
+
+ # We also create a special relocatable object that should mirror the symbol
+@@ -74,5 +80,17 @@ $(obj)/built-in.o: ld_flags += -R $(obj)
+
+ SYSCFLAGS_vsyscall-syms.o = -r
+ $(obj)/vsyscall-syms.o: $(src)/vsyscall.lds \
+- $(obj)/vsyscall-sysenter.o $(obj)/vsyscall-note.o FORCE
++ $(obj)/vsyscall-sysenter.o $(obj)/$(vsyscall_note) FORCE
+ $(call if_changed,syscall)
++
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++
++obj-y += fixup.o
++microcode-$(subst m,y,$(CONFIG_MICROCODE)) := microcode-xen.o
++n-obj-xen := i8259.o timers/ reboot.o smpboot.o trampoline.o
++
++obj-y := $(call filterxen, $(obj-y), $(n-obj-xen))
++obj-y := $(call cherrypickxen, $(obj-y))
++extra-y := $(call cherrypickxen, $(extra-y))
++endif
+diff -urNp linux-2.6/arch/i386/kernel/microcode-xen.c new/arch/i386/kernel/microcode-xen.c
+--- linux-2.6/arch/i386/kernel/microcode-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/microcode-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,148 @@
++/*
++ * Intel CPU Microcode Update Driver for Linux
++ *
++ * Copyright (C) 2000-2004 Tigran Aivazian
++ *
++ * This driver allows to upgrade microcode on Intel processors
++ * belonging to IA-32 family - PentiumPro, Pentium II,
++ * Pentium III, Xeon, Pentium 4, etc.
++ *
++ * Reference: Section 8.10 of Volume III, Intel Pentium 4 Manual,
++ * Order Number 245472 or free download from:
++ *
++ * http://developer.intel.com/design/pentium4/manuals/245472.htm
++ *
++ * For more information, go to http://www.urbanmyth.org/microcode
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License
++ * as published by the Free Software Foundation; either version
++ * 2 of the License, or (at your option) any later version.
++ */
++
++//#define DEBUG /* pr_debug */
++#include <linux/capability.h>
++#include <linux/kernel.h>
++#include <linux/init.h>
++#include <linux/sched.h>
++#include <linux/cpumask.h>
++#include <linux/module.h>
++#include <linux/slab.h>
++#include <linux/vmalloc.h>
++#include <linux/miscdevice.h>
++#include <linux/spinlock.h>
++#include <linux/mm.h>
++#include <linux/mutex.h>
++#include <linux/syscalls.h>
++
++#include <asm/msr.h>
++#include <asm/uaccess.h>
++#include <asm/processor.h>
++
++MODULE_DESCRIPTION("Intel CPU (IA-32) Microcode Update Driver");
++MODULE_AUTHOR("Tigran Aivazian <tigran@veritas.com>");
++MODULE_LICENSE("GPL");
++
++#define MICROCODE_VERSION "1.14-xen"
++
++#define DEFAULT_UCODE_DATASIZE (2000) /* 2000 bytes */
++#define MC_HEADER_SIZE (sizeof (microcode_header_t)) /* 48 bytes */
++#define DEFAULT_UCODE_TOTALSIZE (DEFAULT_UCODE_DATASIZE + MC_HEADER_SIZE) /* 2048 bytes */
++
++/* no concurrent ->write()s are allowed on /dev/cpu/microcode */
++static DEFINE_MUTEX(microcode_mutex);
++
++static void __user *user_buffer; /* user area microcode data buffer */
++static unsigned int user_buffer_size; /* it's size */
++
++static int microcode_open (struct inode *unused1, struct file *unused2)
++{
++ return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
++}
++
++
++static int do_microcode_update (void)
++{
++ int err;
++ dom0_op_t op;
++
++ err = sys_mlock((unsigned long)user_buffer, user_buffer_size);
++ if (err != 0)
++ return err;
++
++ op.cmd = DOM0_MICROCODE;
++ set_xen_guest_handle(op.u.microcode.data, user_buffer);
++ op.u.microcode.length = user_buffer_size;
++ err = HYPERVISOR_dom0_op(&op);
++
++ (void)sys_munlock((unsigned long)user_buffer, user_buffer_size);
++
++ return err;
++}
++
++static ssize_t microcode_write (struct file *file, const char __user *buf, size_t len, loff_t *ppos)
++{
++ ssize_t ret;
++
++ if (len < DEFAULT_UCODE_TOTALSIZE) {
++ printk(KERN_ERR "microcode: not enough data\n");
++ return -EINVAL;
++ }
++
++ if ((len >> PAGE_SHIFT) > num_physpages) {
++ printk(KERN_ERR "microcode: too much data (max %ld pages)\n", num_physpages);
++ return -EINVAL;
++ }
++
++ mutex_lock(µcode_mutex);
++
++ user_buffer = (void __user *) buf;
++ user_buffer_size = (int) len;
++
++ ret = do_microcode_update();
++ if (!ret)
++ ret = (ssize_t)len;
++
++ mutex_unlock(µcode_mutex);
++
++ return ret;
++}
++
++static struct file_operations microcode_fops = {
++ .owner = THIS_MODULE,
++ .write = microcode_write,
++ .open = microcode_open,
++};
++
++static struct miscdevice microcode_dev = {
++ .minor = MICROCODE_MINOR,
++ .name = "microcode",
++ .devfs_name = "cpu/microcode",
++ .fops = µcode_fops,
++};
++
++static int __init microcode_init (void)
++{
++ int error;
++
++ error = misc_register(µcode_dev);
++ if (error) {
++ printk(KERN_ERR
++ "microcode: can't misc_register on minor=%d\n",
++ MICROCODE_MINOR);
++ return error;
++ }
++
++ printk(KERN_INFO
++ "IA-32 Microcode Update Driver: v" MICROCODE_VERSION " <tigran@veritas.com>\n");
++ return 0;
++}
++
++static void __exit microcode_exit (void)
++{
++ misc_deregister(µcode_dev);
++}
++
++module_init(microcode_init)
++module_exit(microcode_exit)
++MODULE_ALIAS_MISCDEV(MICROCODE_MINOR);
+diff -urNp linux-2.6/arch/i386/kernel/mpparse-xen.c new/arch/i386/kernel/mpparse-xen.c
+--- linux-2.6/arch/i386/kernel/mpparse-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/mpparse-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,1186 @@
++/*
++ * Intel Multiprocessor Specification 1.1 and 1.4
++ * compliant MP-table parsing routines.
++ *
++ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
++ * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Erich Boleyn : MP v1.4 and additional changes.
++ * Alan Cox : Added EBDA scanning
++ * Ingo Molnar : various cleanups and rewrites
++ * Maciej W. Rozycki: Bits for default MP configurations
++ * Paul Diefenbaugh: Added full ACPI support
++ */
++
++#include <linux/mm.h>
++#include <linux/init.h>
++#include <linux/acpi.h>
++#include <linux/delay.h>
++#include <linux/config.h>
++#include <linux/bootmem.h>
++#include <linux/smp_lock.h>
++#include <linux/kernel_stat.h>
++#include <linux/mc146818rtc.h>
++#include <linux/bitops.h>
++
++#include <asm/smp.h>
++#include <asm/acpi.h>
++#include <asm/mtrr.h>
++#include <asm/mpspec.h>
++#include <asm/io_apic.h>
++
++#include <mach_apic.h>
++#include <mach_mpparse.h>
++#include <bios_ebda.h>
++
++/* Have we found an MP table */
++int smp_found_config;
++unsigned int __initdata maxcpus = NR_CPUS;
++
++/*
++ * Various Linux-internal data structures created from the
++ * MP-table.
++ */
++int apic_version [MAX_APICS];
++int mp_bus_id_to_type [MAX_MP_BUSSES];
++int mp_bus_id_to_node [MAX_MP_BUSSES];
++int mp_bus_id_to_local [MAX_MP_BUSSES];
++int quad_local_to_mp_bus_id [NR_CPUS/4][4];
++int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
++static int mp_current_pci_id;
++
++/* I/O APIC entries */
++struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
++
++/* # of MP IRQ source entries */
++struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
++
++/* MP IRQ source entries */
++int mp_irq_entries;
++
++int nr_ioapics;
++
++int pic_mode;
++unsigned long mp_lapic_addr;
++
++unsigned int def_to_bigsmp = 0;
++
++/* Processor that is doing the boot up */
++unsigned int boot_cpu_physical_apicid = -1U;
++/* Internal processor count */
++static unsigned int __devinitdata num_processors;
++
++/* Bitmask of physically existing CPUs */
++physid_mask_t phys_cpu_present_map;
++
++u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
++
++/*
++ * Intel MP BIOS table parsing routines:
++ */
++
++
++/*
++ * Checksum an MP configuration block.
++ */
++
++static int __init mpf_checksum(unsigned char *mp, int len)
++{
++ int sum = 0;
++
++ while (len--)
++ sum += *mp++;
++
++ return sum & 0xFF;
++}
++
++/*
++ * Have to match translation table entries to main table entries by counter
++ * hence the mpc_record variable .... can't see a less disgusting way of
++ * doing this ....
++ */
++
++static int mpc_record;
++static struct mpc_config_translation *translation_table[MAX_MPC_ENTRY] __initdata;
++
++#ifndef CONFIG_XEN
++static void __devinit MP_processor_info (struct mpc_config_processor *m)
++{
++ int ver, apicid;
++ physid_mask_t phys_cpu;
++
++ if (!(m->mpc_cpuflag & CPU_ENABLED))
++ return;
++
++ apicid = mpc_apic_id(m, translation_table[mpc_record]);
++
++ if (m->mpc_featureflag&(1<<0))
++ Dprintk(" Floating point unit present.\n");
++ if (m->mpc_featureflag&(1<<7))
++ Dprintk(" Machine Exception supported.\n");
++ if (m->mpc_featureflag&(1<<8))
++ Dprintk(" 64 bit compare & exchange supported.\n");
++ if (m->mpc_featureflag&(1<<9))
++ Dprintk(" Internal APIC present.\n");
++ if (m->mpc_featureflag&(1<<11))
++ Dprintk(" SEP present.\n");
++ if (m->mpc_featureflag&(1<<12))
++ Dprintk(" MTRR present.\n");
++ if (m->mpc_featureflag&(1<<13))
++ Dprintk(" PGE present.\n");
++ if (m->mpc_featureflag&(1<<14))
++ Dprintk(" MCA present.\n");
++ if (m->mpc_featureflag&(1<<15))
++ Dprintk(" CMOV present.\n");
++ if (m->mpc_featureflag&(1<<16))
++ Dprintk(" PAT present.\n");
++ if (m->mpc_featureflag&(1<<17))
++ Dprintk(" PSE present.\n");
++ if (m->mpc_featureflag&(1<<18))
++ Dprintk(" PSN present.\n");
++ if (m->mpc_featureflag&(1<<19))
++ Dprintk(" Cache Line Flush Instruction present.\n");
++ /* 20 Reserved */
++ if (m->mpc_featureflag&(1<<21))
++ Dprintk(" Debug Trace and EMON Store present.\n");
++ if (m->mpc_featureflag&(1<<22))
++ Dprintk(" ACPI Thermal Throttle Registers present.\n");
++ if (m->mpc_featureflag&(1<<23))
++ Dprintk(" MMX present.\n");
++ if (m->mpc_featureflag&(1<<24))
++ Dprintk(" FXSR present.\n");
++ if (m->mpc_featureflag&(1<<25))
++ Dprintk(" XMM present.\n");
++ if (m->mpc_featureflag&(1<<26))
++ Dprintk(" Willamette New Instructions present.\n");
++ if (m->mpc_featureflag&(1<<27))
++ Dprintk(" Self Snoop present.\n");
++ if (m->mpc_featureflag&(1<<28))
++ Dprintk(" HT present.\n");
++ if (m->mpc_featureflag&(1<<29))
++ Dprintk(" Thermal Monitor present.\n");
++ /* 30, 31 Reserved */
++
++
++ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
++ Dprintk(" Bootup CPU\n");
++ boot_cpu_physical_apicid = m->mpc_apicid;
++ }
++
++ ver = m->mpc_apicver;
++
++ /*
++ * Validate version
++ */
++ if (ver == 0x0) {
++ printk(KERN_WARNING "BIOS bug, APIC version is 0 for CPU#%d! "
++ "fixing up to 0x10. (tell your hw vendor)\n",
++ m->mpc_apicid);
++ ver = 0x10;
++ }
++ apic_version[m->mpc_apicid] = ver;
++
++ phys_cpu = apicid_to_cpu_present(apicid);
++ physids_or(phys_cpu_present_map, phys_cpu_present_map, phys_cpu);
++
++ if (num_processors >= NR_CPUS) {
++ printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
++ " Processor ignored.\n", NR_CPUS);
++ return;
++ }
++
++ if (num_processors >= maxcpus) {
++ printk(KERN_WARNING "WARNING: maxcpus limit of %i reached."
++ " Processor ignored.\n", maxcpus);
++ return;
++ }
++
++ cpu_set(num_processors, cpu_possible_map);
++ num_processors++;
++
++ /*
++ * Would be preferable to switch to bigsmp when CONFIG_HOTPLUG_CPU=y
++ * but we need to work other dependencies like SMP_SUSPEND etc
++ * before this can be done without some confusion.
++ * if (CPU_HOTPLUG_ENABLED || num_processors > 8)
++ * - Ashok Raj <ashok.raj@intel.com>
++ */
++ if (num_processors > 8) {
++ switch (boot_cpu_data.x86_vendor) {
++ case X86_VENDOR_INTEL:
++ if (!APIC_XAPIC(ver)) {
++ def_to_bigsmp = 0;
++ break;
++ }
++ /* If P4 and above fall through */
++ case X86_VENDOR_AMD:
++ def_to_bigsmp = 1;
++ }
++ }
++ bios_cpu_apicid[num_processors - 1] = m->mpc_apicid;
++}
++#else
++void __init MP_processor_info (struct mpc_config_processor *m)
++{
++ num_processors++;
++}
++#endif /* CONFIG_XEN */
++
++static void __init MP_bus_info (struct mpc_config_bus *m)
++{
++ char str[7];
++
++ memcpy(str, m->mpc_bustype, 6);
++ str[6] = 0;
++
++ mpc_oem_bus_info(m, str, translation_table[mpc_record]);
++
++ if (m->mpc_busid >= MAX_MP_BUSSES) {
++ printk(KERN_WARNING "MP table busid value (%d) for bustype %s "
++ " is too large, max. supported is %d\n",
++ m->mpc_busid, str, MAX_MP_BUSSES - 1);
++ return;
++ }
++
++ if (strncmp(str, BUSTYPE_ISA, sizeof(BUSTYPE_ISA)-1) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
++ } else if (strncmp(str, BUSTYPE_EISA, sizeof(BUSTYPE_EISA)-1) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
++ } else if (strncmp(str, BUSTYPE_PCI, sizeof(BUSTYPE_PCI)-1) == 0) {
++ mpc_oem_pci_bus(m, translation_table[mpc_record]);
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
++ mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
++ mp_current_pci_id++;
++ } else if (strncmp(str, BUSTYPE_MCA, sizeof(BUSTYPE_MCA)-1) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
++ } else if (strncmp(str, BUSTYPE_NEC98, sizeof(BUSTYPE_NEC98)-1) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_NEC98;
++ } else {
++ printk(KERN_WARNING "Unknown bustype %s - ignoring\n", str);
++ }
++}
++
++static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
++{
++ if (!(m->mpc_flags & MPC_APIC_USABLE))
++ return;
++
++ printk(KERN_INFO "I/O APIC #%d Version %d at 0x%lX.\n",
++ m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
++ if (nr_ioapics >= MAX_IO_APICS) {
++ printk(KERN_CRIT "Max # of I/O APICs (%d) exceeded (found %d).\n",
++ MAX_IO_APICS, nr_ioapics);
++ panic("Recompile kernel with bigger MAX_IO_APICS!.\n");
++ }
++ if (!m->mpc_apicaddr) {
++ printk(KERN_ERR "WARNING: bogus zero I/O APIC address"
++ " found in MP table, skipping!\n");
++ return;
++ }
++ mp_ioapics[nr_ioapics] = *m;
++ nr_ioapics++;
++}
++
++static void __init MP_intsrc_info (struct mpc_config_intsrc *m)
++{
++ mp_irqs [mp_irq_entries] = *m;
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d,"
++ " IRQ %02x, APIC ID %x, APIC INT %02x\n",
++ m->mpc_irqtype, m->mpc_irqflag & 3,
++ (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
++ m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq);
++ if (++mp_irq_entries == MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!!\n");
++}
++
++static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
++{
++ Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
++ " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
++ m->mpc_irqtype, m->mpc_irqflag & 3,
++ (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
++ m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
++ /*
++ * Well it seems all SMP boards in existence
++ * use ExtINT/LVT1 == LINT0 and
++ * NMI/LVT2 == LINT1 - the following check
++ * will show us if this assumptions is false.
++ * Until then we do not have to add baggage.
++ */
++ if ((m->mpc_irqtype == mp_ExtINT) &&
++ (m->mpc_destapiclint != 0))
++ BUG();
++ if ((m->mpc_irqtype == mp_NMI) &&
++ (m->mpc_destapiclint != 1))
++ BUG();
++}
++
++#ifdef CONFIG_X86_NUMAQ
++static void __init MP_translation_info (struct mpc_config_translation *m)
++{
++ printk(KERN_INFO "Translation: record %d, type %d, quad %d, global %d, local %d\n", mpc_record, m->trans_type, m->trans_quad, m->trans_global, m->trans_local);
++
++ if (mpc_record >= MAX_MPC_ENTRY)
++ printk(KERN_ERR "MAX_MPC_ENTRY exceeded!\n");
++ else
++ translation_table[mpc_record] = m; /* stash this for later */
++ if (m->trans_quad < MAX_NUMNODES && !node_online(m->trans_quad))
++ node_set_online(m->trans_quad);
++}
++
++/*
++ * Read/parse the MPC oem tables
++ */
++
++static void __init smp_read_mpc_oem(struct mp_config_oemtable *oemtable, \
++ unsigned short oemsize)
++{
++ int count = sizeof (*oemtable); /* the header size */
++ unsigned char *oemptr = ((unsigned char *)oemtable)+count;
++
++ mpc_record = 0;
++ printk(KERN_INFO "Found an OEM MPC table at %8p - parsing it ... \n", oemtable);
++ if (memcmp(oemtable->oem_signature,MPC_OEM_SIGNATURE,4))
++ {
++ printk(KERN_WARNING "SMP mpc oemtable: bad signature [%c%c%c%c]!\n",
++ oemtable->oem_signature[0],
++ oemtable->oem_signature[1],
++ oemtable->oem_signature[2],
++ oemtable->oem_signature[3]);
++ return;
++ }
++ if (mpf_checksum((unsigned char *)oemtable,oemtable->oem_length))
++ {
++ printk(KERN_WARNING "SMP oem mptable: checksum error!\n");
++ return;
++ }
++ while (count < oemtable->oem_length) {
++ switch (*oemptr) {
++ case MP_TRANSLATION:
++ {
++ struct mpc_config_translation *m=
++ (struct mpc_config_translation *)oemptr;
++ MP_translation_info(m);
++ oemptr += sizeof(*m);
++ count += sizeof(*m);
++ ++mpc_record;
++ break;
++ }
++ default:
++ {
++ printk(KERN_WARNING "Unrecognised OEM table entry type! - %d\n", (int) *oemptr);
++ return;
++ }
++ }
++ }
++}
++
++static inline void mps_oem_check(struct mp_config_table *mpc, char *oem,
++ char *productid)
++{
++ if (strncmp(oem, "IBM NUMA", 8))
++ printk("Warning! May not be a NUMA-Q system!\n");
++ if (mpc->mpc_oemptr)
++ smp_read_mpc_oem((struct mp_config_oemtable *) mpc->mpc_oemptr,
++ mpc->mpc_oemsize);
++}
++#endif /* CONFIG_X86_NUMAQ */
++
++/*
++ * Read/parse the MPC
++ */
++
++static int __init smp_read_mpc(struct mp_config_table *mpc)
++{
++ char str[16];
++ char oem[10];
++ int count=sizeof(*mpc);
++ unsigned char *mpt=((unsigned char *)mpc)+count;
++
++ if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) {
++ printk(KERN_ERR "SMP mptable: bad signature [0x%x]!\n",
++ *(u32 *)mpc->mpc_signature);
++ return 0;
++ }
++ if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) {
++ printk(KERN_ERR "SMP mptable: checksum error!\n");
++ return 0;
++ }
++ if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) {
++ printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n",
++ mpc->mpc_spec);
++ return 0;
++ }
++ if (!mpc->mpc_lapic) {
++ printk(KERN_ERR "SMP mptable: null local APIC address!\n");
++ return 0;
++ }
++ memcpy(oem,mpc->mpc_oem,8);
++ oem[8]=0;
++ printk(KERN_INFO "OEM ID: %s ",oem);
++
++ memcpy(str,mpc->mpc_productid,12);
++ str[12]=0;
++ printk("Product ID: %s ",str);
++
++ mps_oem_check(mpc, oem, str);
++
++ printk("APIC at: 0x%lX\n",mpc->mpc_lapic);
++
++ /*
++ * Save the local APIC address (it might be non-default) -- but only
++ * if we're not using ACPI.
++ */
++ if (!acpi_lapic)
++ mp_lapic_addr = mpc->mpc_lapic;
++
++ /*
++ * Now process the configuration blocks.
++ */
++ mpc_record = 0;
++ while (count < mpc->mpc_length) {
++ switch(*mpt) {
++ case MP_PROCESSOR:
++ {
++ struct mpc_config_processor *m=
++ (struct mpc_config_processor *)mpt;
++ /* ACPI may have already provided this data */
++ if (!acpi_lapic)
++ MP_processor_info(m);
++ mpt += sizeof(*m);
++ count += sizeof(*m);
++ break;
++ }
++ case MP_BUS:
++ {
++ struct mpc_config_bus *m=
++ (struct mpc_config_bus *)mpt;
++ MP_bus_info(m);
++ mpt += sizeof(*m);
++ count += sizeof(*m);
++ break;
++ }
++ case MP_IOAPIC:
++ {
++ struct mpc_config_ioapic *m=
++ (struct mpc_config_ioapic *)mpt;
++ MP_ioapic_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ case MP_INTSRC:
++ {
++ struct mpc_config_intsrc *m=
++ (struct mpc_config_intsrc *)mpt;
++
++ MP_intsrc_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ case MP_LINTSRC:
++ {
++ struct mpc_config_lintsrc *m=
++ (struct mpc_config_lintsrc *)mpt;
++ MP_lintsrc_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ default:
++ {
++ count = mpc->mpc_length;
++ break;
++ }
++ }
++ ++mpc_record;
++ }
++ clustered_apic_check();
++ if (!num_processors)
++ printk(KERN_ERR "SMP mptable: no processors registered!\n");
++ return num_processors;
++}
++
++static int __init ELCR_trigger(unsigned int irq)
++{
++ unsigned int port;
++
++ port = 0x4d0 + (irq >> 3);
++ return (inb(port) >> (irq & 7)) & 1;
++}
++
++static void __init construct_default_ioirq_mptable(int mpc_default_type)
++{
++ struct mpc_config_intsrc intsrc;
++ int i;
++ int ELCR_fallback = 0;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqflag = 0; /* conforming */
++ intsrc.mpc_srcbus = 0;
++ intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid;
++
++ intsrc.mpc_irqtype = mp_INT;
++
++ /*
++ * If true, we have an ISA/PCI system with no IRQ entries
++ * in the MP table. To prevent the PCI interrupts from being set up
++ * incorrectly, we try to use the ELCR. The sanity check to see if
++ * there is good ELCR data is very simple - IRQ0, 1, 2 and 13 can
++ * never be level sensitive, so we simply see if the ELCR agrees.
++ * If it does, we assume it's valid.
++ */
++ if (mpc_default_type == 5) {
++ printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n");
++
++ if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13))
++ printk(KERN_WARNING "ELCR contains invalid data... not using ELCR\n");
++ else {
++ printk(KERN_INFO "Using ELCR to identify PCI interrupts\n");
++ ELCR_fallback = 1;
++ }
++ }
++
++ for (i = 0; i < 16; i++) {
++ switch (mpc_default_type) {
++ case 2:
++ if (i == 0 || i == 13)
++ continue; /* IRQ0 & IRQ13 not connected */
++ /* fall through */
++ default:
++ if (i == 2)
++ continue; /* IRQ2 is never connected */
++ }
++
++ if (ELCR_fallback) {
++ /*
++ * If the ELCR indicates a level-sensitive interrupt, we
++ * copy that information over to the MP table in the
++ * irqflag field (level sensitive, active high polarity).
++ */
++ if (ELCR_trigger(i))
++ intsrc.mpc_irqflag = 13;
++ else
++ intsrc.mpc_irqflag = 0;
++ }
++
++ intsrc.mpc_srcbusirq = i;
++ intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */
++ MP_intsrc_info(&intsrc);
++ }
++
++ intsrc.mpc_irqtype = mp_ExtINT;
++ intsrc.mpc_srcbusirq = 0;
++ intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */
++ MP_intsrc_info(&intsrc);
++}
++
++static inline void __init construct_default_ISA_mptable(int mpc_default_type)
++{
++ struct mpc_config_processor processor;
++ struct mpc_config_bus bus;
++ struct mpc_config_ioapic ioapic;
++ struct mpc_config_lintsrc lintsrc;
++ int linttypes[2] = { mp_ExtINT, mp_NMI };
++ int i;
++
++ /*
++ * local APIC has default address
++ */
++ mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
++
++ /*
++ * 2 CPUs, numbered 0 & 1.
++ */
++ processor.mpc_type = MP_PROCESSOR;
++ /* Either an integrated APIC or a discrete 82489DX. */
++ processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
++ processor.mpc_cpuflag = CPU_ENABLED;
++ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
++ (boot_cpu_data.x86_model << 4) |
++ boot_cpu_data.x86_mask;
++ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
++ processor.mpc_reserved[0] = 0;
++ processor.mpc_reserved[1] = 0;
++ for (i = 0; i < 2; i++) {
++ processor.mpc_apicid = i;
++ MP_processor_info(&processor);
++ }
++
++ bus.mpc_type = MP_BUS;
++ bus.mpc_busid = 0;
++ switch (mpc_default_type) {
++ default:
++ printk("???\n");
++ printk(KERN_ERR "Unknown standard configuration %d\n",
++ mpc_default_type);
++ /* fall through */
++ case 1:
++ case 5:
++ memcpy(bus.mpc_bustype, "ISA ", 6);
++ break;
++ case 2:
++ case 6:
++ case 3:
++ memcpy(bus.mpc_bustype, "EISA ", 6);
++ break;
++ case 4:
++ case 7:
++ memcpy(bus.mpc_bustype, "MCA ", 6);
++ }
++ MP_bus_info(&bus);
++ if (mpc_default_type > 4) {
++ bus.mpc_busid = 1;
++ memcpy(bus.mpc_bustype, "PCI ", 6);
++ MP_bus_info(&bus);
++ }
++
++ ioapic.mpc_type = MP_IOAPIC;
++ ioapic.mpc_apicid = 2;
++ ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
++ ioapic.mpc_flags = MPC_APIC_USABLE;
++ ioapic.mpc_apicaddr = 0xFEC00000;
++ MP_ioapic_info(&ioapic);
++
++ /*
++ * We set up most of the low 16 IO-APIC pins according to MPS rules.
++ */
++ construct_default_ioirq_mptable(mpc_default_type);
++
++ lintsrc.mpc_type = MP_LINTSRC;
++ lintsrc.mpc_irqflag = 0; /* conforming */
++ lintsrc.mpc_srcbusid = 0;
++ lintsrc.mpc_srcbusirq = 0;
++ lintsrc.mpc_destapic = MP_APIC_ALL;
++ for (i = 0; i < 2; i++) {
++ lintsrc.mpc_irqtype = linttypes[i];
++ lintsrc.mpc_destapiclint = i;
++ MP_lintsrc_info(&lintsrc);
++ }
++}
++
++static struct intel_mp_floating *mpf_found;
++
++/*
++ * Scan the memory blocks for an SMP configuration block.
++ */
++void __init get_smp_config (void)
++{
++ struct intel_mp_floating *mpf = mpf_found;
++
++ /*
++ * ACPI supports both logical (e.g. Hyper-Threading) and physical
++ * processors, where MPS only supports physical.
++ */
++ if (acpi_lapic && acpi_ioapic) {
++ printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n");
++ return;
++ }
++ else if (acpi_lapic)
++ printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
++
++ printk(KERN_INFO "Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
++ if (mpf->mpf_feature2 & (1<<7)) {
++ printk(KERN_INFO " IMCR and PIC compatibility mode.\n");
++ pic_mode = 1;
++ } else {
++ printk(KERN_INFO " Virtual Wire compatibility mode.\n");
++ pic_mode = 0;
++ }
++
++ /*
++ * Now see if we need to read further.
++ */
++ if (mpf->mpf_feature1 != 0) {
++
++ printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1);
++ construct_default_ISA_mptable(mpf->mpf_feature1);
++
++ } else if (mpf->mpf_physptr) {
++
++ /*
++ * Read the physical hardware table. Anything here will
++ * override the defaults.
++ */
++ if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr))) {
++ smp_found_config = 0;
++ printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
++ printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
++ return;
++ }
++ /*
++ * If there are no explicit MP IRQ entries, then we are
++ * broken. We set up most of the low 16 IO-APIC pins to
++ * ISA defaults and hope it will work.
++ */
++ if (!mp_irq_entries) {
++ struct mpc_config_bus bus;
++
++ printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n");
++
++ bus.mpc_type = MP_BUS;
++ bus.mpc_busid = 0;
++ memcpy(bus.mpc_bustype, "ISA ", 6);
++ MP_bus_info(&bus);
++
++ construct_default_ioirq_mptable(0);
++ }
++
++ } else
++ BUG();
++
++ printk(KERN_INFO "Processors: %d\n", num_processors);
++ /*
++ * Only use the first configuration found.
++ */
++}
++
++static int __init smp_scan_config (unsigned long base, unsigned long length)
++{
++ unsigned long *bp = isa_bus_to_virt(base);
++ struct intel_mp_floating *mpf;
++
++ Dprintk("Scan SMP from %p for %ld bytes.\n", bp,length);
++ if (sizeof(*mpf) != 16)
++ printk("Error: MPF size\n");
++
++ while (length > 0) {
++ mpf = (struct intel_mp_floating *)bp;
++ if ((*bp == SMP_MAGIC_IDENT) &&
++ (mpf->mpf_length == 1) &&
++ !mpf_checksum((unsigned char *)bp, 16) &&
++ ((mpf->mpf_specification == 1)
++ || (mpf->mpf_specification == 4)) ) {
++
++ smp_found_config = 1;
++#ifndef CONFIG_XEN
++ printk(KERN_INFO "found SMP MP-table at %08lx\n",
++ virt_to_phys(mpf));
++ reserve_bootmem(virt_to_phys(mpf), PAGE_SIZE);
++ if (mpf->mpf_physptr) {
++ /*
++ * We cannot access to MPC table to compute
++ * table size yet, as only few megabytes from
++ * the bottom is mapped now.
++ * PC-9800's MPC table places on the very last
++ * of physical memory; so that simply reserving
++ * PAGE_SIZE from mpg->mpf_physptr yields BUG()
++ * in reserve_bootmem.
++ */
++ unsigned long size = PAGE_SIZE;
++ unsigned long end = max_low_pfn * PAGE_SIZE;
++ if (mpf->mpf_physptr + size > end)
++ size = end - mpf->mpf_physptr;
++ reserve_bootmem(mpf->mpf_physptr, size);
++ }
++#else
++ printk(KERN_INFO "found SMP MP-table at %08lx\n",
++ ((unsigned long)bp - (unsigned long)isa_bus_to_virt(base)) + base);
++#endif
++
++ mpf_found = mpf;
++ return 1;
++ }
++ bp += 4;
++ length -= 16;
++ }
++ return 0;
++}
++
++void __init find_smp_config (void)
++{
++#ifndef CONFIG_XEN
++ unsigned int address;
++#endif
++
++ /*
++ * FIXME: Linux assumes you have 640K of base ram..
++ * this continues the error...
++ *
++ * 1) Scan the bottom 1K for a signature
++ * 2) Scan the top 1K of base RAM
++ * 3) Scan the 64K of bios
++ */
++ if (smp_scan_config(0x0,0x400) ||
++ smp_scan_config(639*0x400,0x400) ||
++ smp_scan_config(0xF0000,0x10000))
++ return;
++ /*
++ * If it is an SMP machine we should know now, unless the
++ * configuration is in an EISA/MCA bus machine with an
++ * extended bios data area.
++ *
++ * there is a real-mode segmented pointer pointing to the
++ * 4K EBDA area at 0x40E, calculate and scan it here.
++ *
++ * NOTE! There are Linux loaders that will corrupt the EBDA
++ * area, and as such this kind of SMP config may be less
++ * trustworthy, simply because the SMP table may have been
++ * stomped on during early boot. These loaders are buggy and
++ * should be fixed.
++ *
++ * MP1.4 SPEC states to only scan first 1K of 4K EBDA.
++ */
++
++#ifndef CONFIG_XEN
++ address = get_bios_ebda();
++ if (address)
++ smp_scan_config(address, 0x400);
++#endif
++}
++
++int es7000_plat;
++
++/* --------------------------------------------------------------------------
++ ACPI-based MP Configuration
++ -------------------------------------------------------------------------- */
++
++#ifdef CONFIG_ACPI
++
++void __init mp_register_lapic_address (
++ u64 address)
++{
++#ifndef CONFIG_XEN
++ mp_lapic_addr = (unsigned long) address;
++
++ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
++
++ if (boot_cpu_physical_apicid == -1U)
++ boot_cpu_physical_apicid = GET_APIC_ID(apic_read(APIC_ID));
++
++ Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
++#endif
++}
++
++
++void __devinit mp_register_lapic (
++ u8 id,
++ u8 enabled)
++{
++ struct mpc_config_processor processor;
++ int boot_cpu = 0;
++
++ if (MAX_APICS - id <= 0) {
++ printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
++ id, MAX_APICS);
++ return;
++ }
++
++ if (id == boot_cpu_physical_apicid)
++ boot_cpu = 1;
++
++#ifndef CONFIG_XEN
++ processor.mpc_type = MP_PROCESSOR;
++ processor.mpc_apicid = id;
++ processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR));
++ processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0);
++ processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0);
++ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
++ (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
++ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
++ processor.mpc_reserved[0] = 0;
++ processor.mpc_reserved[1] = 0;
++#endif
++
++ MP_processor_info(&processor);
++}
++
++#ifdef CONFIG_X86_IO_APIC
++
++#define MP_ISA_BUS 0
++#define MP_MAX_IOAPIC_PIN 127
++
++static struct mp_ioapic_routing {
++ int apic_id;
++ int gsi_base;
++ int gsi_end;
++ u32 pin_programmed[4];
++} mp_ioapic_routing[MAX_IO_APICS];
++
++
++static int mp_find_ioapic (
++ int gsi)
++{
++ int i = 0;
++
++ /* Find the IOAPIC that manages this GSI. */
++ for (i = 0; i < nr_ioapics; i++) {
++ if ((gsi >= mp_ioapic_routing[i].gsi_base)
++ && (gsi <= mp_ioapic_routing[i].gsi_end))
++ return i;
++ }
++
++ printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
++
++ return -1;
++}
++
++
++void __init mp_register_ioapic (
++ u8 id,
++ u32 address,
++ u32 gsi_base)
++{
++ int idx = 0;
++ int tmpid;
++
++ if (nr_ioapics >= MAX_IO_APICS) {
++ printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
++ "(found %d)\n", MAX_IO_APICS, nr_ioapics);
++ panic("Recompile kernel with bigger MAX_IO_APICS!\n");
++ }
++ if (!address) {
++ printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
++ " found in MADT table, skipping!\n");
++ return;
++ }
++
++ idx = nr_ioapics++;
++
++ mp_ioapics[idx].mpc_type = MP_IOAPIC;
++ mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
++ mp_ioapics[idx].mpc_apicaddr = address;
++
++#ifndef CONFIG_XEN
++ set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
++#endif
++ if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
++ && !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
++ tmpid = io_apic_get_unique_id(idx, id);
++ else
++ tmpid = id;
++ if (tmpid == -1) {
++ nr_ioapics--;
++ return;
++ }
++ mp_ioapics[idx].mpc_apicid = tmpid;
++ mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
++
++ /*
++ * Build basic GSI lookup table to facilitate gsi->io_apic lookups
++ * and to prevent reprogramming of IOAPIC pins (PCI GSIs).
++ */
++ mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
++ mp_ioapic_routing[idx].gsi_base = gsi_base;
++ mp_ioapic_routing[idx].gsi_end = gsi_base +
++ io_apic_get_redir_entries(idx);
++
++ printk("IOAPIC[%d]: apic_id %d, version %d, address 0x%lx, "
++ "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
++ mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
++ mp_ioapic_routing[idx].gsi_base,
++ mp_ioapic_routing[idx].gsi_end);
++
++ return;
++}
++
++
++void __init mp_override_legacy_irq (
++ u8 bus_irq,
++ u8 polarity,
++ u8 trigger,
++ u32 gsi)
++{
++ struct mpc_config_intsrc intsrc;
++ int ioapic = -1;
++ int pin = -1;
++
++ /*
++ * Convert 'gsi' to 'ioapic.pin'.
++ */
++ ioapic = mp_find_ioapic(gsi);
++ if (ioapic < 0)
++ return;
++ pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
++
++ /*
++ * TBD: This check is for faulty timer entries, where the override
++ * erroneously sets the trigger to level, resulting in a HUGE
++ * increase of timer interrupts!
++ */
++ if ((bus_irq == 0) && (trigger == 3))
++ trigger = 1;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqtype = mp_INT;
++ intsrc.mpc_irqflag = (trigger << 2) | polarity;
++ intsrc.mpc_srcbus = MP_ISA_BUS;
++ intsrc.mpc_srcbusirq = bus_irq; /* IRQ */
++ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */
++ intsrc.mpc_dstirq = pin; /* INTIN# */
++
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n",
++ intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
++ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
++ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq);
++
++ mp_irqs[mp_irq_entries] = intsrc;
++ if (++mp_irq_entries == MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!\n");
++
++ return;
++}
++
++void __init mp_config_acpi_legacy_irqs (void)
++{
++ struct mpc_config_intsrc intsrc;
++ int i = 0;
++ int ioapic = -1;
++
++ /*
++ * Fabricate the legacy ISA bus (bus #31).
++ */
++ mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
++ Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
++
++ /*
++ * Older generations of ES7000 have no legacy identity mappings
++ */
++ if (es7000_plat == 1)
++ return;
++
++ /*
++ * Locate the IOAPIC that manages the ISA IRQs (0-15).
++ */
++ ioapic = mp_find_ioapic(0);
++ if (ioapic < 0)
++ return;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqflag = 0; /* Conforming */
++ intsrc.mpc_srcbus = MP_ISA_BUS;
++ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
++
++ /*
++ * Use the default configuration for the IRQs 0-15. Unless
++ * overriden by (MADT) interrupt source override entries.
++ */
++ for (i = 0; i < 16; i++) {
++ int idx;
++
++ for (idx = 0; idx < mp_irq_entries; idx++) {
++ struct mpc_config_intsrc *irq = mp_irqs + idx;
++
++ /* Do we already have a mapping for this ISA IRQ? */
++ if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i)
++ break;
++
++ /* Do we already have a mapping for this IOAPIC pin */
++ if ((irq->mpc_dstapic == intsrc.mpc_dstapic) &&
++ (irq->mpc_dstirq == i))
++ break;
++ }
++
++ if (idx != mp_irq_entries) {
++ printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i);
++ continue; /* IRQ already used */
++ }
++
++ intsrc.mpc_irqtype = mp_INT;
++ intsrc.mpc_srcbusirq = i; /* Identity mapped */
++ intsrc.mpc_dstirq = i;
++
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, "
++ "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
++ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
++ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic,
++ intsrc.mpc_dstirq);
++
++ mp_irqs[mp_irq_entries] = intsrc;
++ if (++mp_irq_entries == MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!\n");
++ }
++}
++
++#define MAX_GSI_NUM 4096
++
++int mp_register_gsi (u32 gsi, int triggering, int polarity)
++{
++ int ioapic = -1;
++ int ioapic_pin = 0;
++ int idx, bit = 0;
++ static int pci_irq = 16;
++ /*
++ * Mapping between Global System Interrups, which
++ * represent all possible interrupts, and IRQs
++ * assigned to actual devices.
++ */
++ static int gsi_to_irq[MAX_GSI_NUM];
++
++ /* Don't set up the ACPI SCI because it's already set up */
++ if (acpi_fadt.sci_int == gsi)
++ return gsi;
++
++ ioapic = mp_find_ioapic(gsi);
++ if (ioapic < 0) {
++ printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi);
++ return gsi;
++ }
++
++ ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_base;
++
++ if (ioapic_renumber_irq)
++ gsi = ioapic_renumber_irq(ioapic, gsi);
++
++ /*
++ * Avoid pin reprogramming. PRTs typically include entries
++ * with redundant pin->gsi mappings (but unique PCI devices);
++ * we only program the IOAPIC on the first.
++ */
++ bit = ioapic_pin % 32;
++ idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32);
++ if (idx > 3) {
++ printk(KERN_ERR "Invalid reference to IOAPIC pin "
++ "%d-%d\n", mp_ioapic_routing[ioapic].apic_id,
++ ioapic_pin);
++ return gsi;
++ }
++ if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
++ Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
++ mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
++ return gsi_to_irq[gsi];
++ }
++
++ mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
++
++ if (triggering == ACPI_LEVEL_SENSITIVE) {
++ /*
++ * For PCI devices assign IRQs in order, avoiding gaps
++ * due to unused I/O APIC pins.
++ */
++ int irq = gsi;
++ if (gsi < MAX_GSI_NUM) {
++ /*
++ * Retain the VIA chipset work-around (gsi > 15), but
++ * avoid a problem where the 8254 timer (IRQ0) is setup
++ * via an override (so it's not on pin 0 of the ioapic),
++ * and at the same time, the pin 0 interrupt is a PCI
++ * type. The gsi > 15 test could cause these two pins
++ * to be shared as IRQ0, and they are not shareable.
++ * So test for this condition, and if necessary, avoid
++ * the pin collision.
++ */
++ if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
++ gsi = pci_irq++;
++ /*
++ * Don't assign IRQ used by ACPI SCI
++ */
++ if (gsi == acpi_fadt.sci_int)
++ gsi = pci_irq++;
++ gsi_to_irq[irq] = gsi;
++ } else {
++ printk(KERN_ERR "GSI %u is too high\n", gsi);
++ return gsi;
++ }
++ }
++
++ io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
++ triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
++ polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
++ return gsi;
++}
++
++#endif /* CONFIG_X86_IO_APIC */
++#endif /* CONFIG_ACPI */
+diff -urNp linux-2.6/arch/i386/kernel/pci-dma-xen.c new/arch/i386/kernel/pci-dma-xen.c
+--- linux-2.6/arch/i386/kernel/pci-dma-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/pci-dma-xen.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,344 @@
++/*
++ * Dynamic DMA mapping support.
++ *
++ * On i386 there is no hardware dynamic DMA address translation,
++ * so consistent alloc/free are merely page allocation/freeing.
++ * The rest of the dynamic DMA mapping interface is implemented
++ * in asm/pci.h.
++ */
++
++#include <linux/types.h>
++#include <linux/mm.h>
++#include <linux/string.h>
++#include <linux/pci.h>
++#include <linux/module.h>
++#include <linux/version.h>
++#include <asm/io.h>
++#include <xen/balloon.h>
++#include <asm/tlbflush.h>
++#include <asm-i386/mach-xen/asm/swiotlb.h>
++#include <asm/bug.h>
++
++#ifdef __x86_64__
++int iommu_merge __read_mostly = 0;
++EXPORT_SYMBOL(iommu_merge);
++
++dma_addr_t bad_dma_address __read_mostly;
++EXPORT_SYMBOL(bad_dma_address);
++
++/* This tells the BIO block layer to assume merging. Default to off
++ because we cannot guarantee merging later. */
++int iommu_bio_merge __read_mostly = 0;
++EXPORT_SYMBOL(iommu_bio_merge);
++
++__init int iommu_setup(char *p)
++{
++ return 1;
++}
++#endif
++
++struct dma_coherent_mem {
++ void *virt_base;
++ u32 device_base;
++ int size;
++ int flags;
++ unsigned long *bitmap;
++};
++
++#define IOMMU_BUG_ON(test) \
++do { \
++ if (unlikely(test)) { \
++ printk(KERN_ALERT "Fatal DMA error! " \
++ "Please use 'swiotlb=force'\n"); \
++ BUG(); \
++ } \
++} while (0)
++
++int
++dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents,
++ enum dma_data_direction direction)
++{
++ int i, rc;
++
++ if (direction == DMA_NONE)
++ BUG();
++ WARN_ON(nents == 0 || sg[0].length == 0);
++
++ if (swiotlb) {
++ rc = swiotlb_map_sg(hwdev, sg, nents, direction);
++ } else {
++ for (i = 0; i < nents; i++ ) {
++ sg[i].dma_address =
++ page_to_bus(sg[i].page) + sg[i].offset;
++ sg[i].dma_length = sg[i].length;
++ BUG_ON(!sg[i].page);
++ IOMMU_BUG_ON(address_needs_mapping(
++ hwdev, sg[i].dma_address));
++ }
++ rc = nents;
++ }
++
++ flush_write_buffers();
++ return rc;
++}
++EXPORT_SYMBOL(dma_map_sg);
++
++void
++dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
++ enum dma_data_direction direction)
++{
++ BUG_ON(direction == DMA_NONE);
++ if (swiotlb)
++ swiotlb_unmap_sg(hwdev, sg, nents, direction);
++}
++EXPORT_SYMBOL(dma_unmap_sg);
++
++/*
++ * XXX This file is also used by xenLinux/ia64.
++ * "defined(__i386__) || defined (__x86_64__)" means "!defined(__ia64__)".
++ * This #if work around should be removed once this file is merbed back into
++ * i386' pci-dma or is moved to drivers/xen/core.
++ */
++#if defined(__i386__) || defined(__x86_64__)
++dma_addr_t
++dma_map_page(struct device *dev, struct page *page, unsigned long offset,
++ size_t size, enum dma_data_direction direction)
++{
++ dma_addr_t dma_addr;
++
++ BUG_ON(direction == DMA_NONE);
++
++ if (swiotlb) {
++ dma_addr = swiotlb_map_page(
++ dev, page, offset, size, direction);
++ } else {
++ dma_addr = page_to_bus(page) + offset;
++ IOMMU_BUG_ON(address_needs_mapping(dev, dma_addr));
++ }
++
++ return dma_addr;
++}
++EXPORT_SYMBOL(dma_map_page);
++
++void
++dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
++ enum dma_data_direction direction)
++{
++ BUG_ON(direction == DMA_NONE);
++ if (swiotlb)
++ swiotlb_unmap_page(dev, dma_address, size, direction);
++}
++EXPORT_SYMBOL(dma_unmap_page);
++#endif /* defined(__i386__) || defined(__x86_64__) */
++
++int
++dma_mapping_error(dma_addr_t dma_addr)
++{
++ if (swiotlb)
++ return swiotlb_dma_mapping_error(dma_addr);
++ return 0;
++}
++EXPORT_SYMBOL(dma_mapping_error);
++
++int
++dma_supported(struct device *dev, u64 mask)
++{
++ if (swiotlb)
++ return swiotlb_dma_supported(dev, mask);
++ /*
++ * By default we'll BUG when an infeasible DMA is requested, and
++ * request swiotlb=force (see IOMMU_BUG_ON).
++ */
++ return 1;
++}
++EXPORT_SYMBOL(dma_supported);
++
++void *dma_alloc_coherent(struct device *dev, size_t size,
++ dma_addr_t *dma_handle, gfp_t gfp)
++{
++ void *ret;
++ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
++ unsigned int order = get_order(size);
++ unsigned long vstart;
++ /* ignore region specifiers */
++ gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
++
++ if (mem) {
++ int page = bitmap_find_free_region(mem->bitmap, mem->size,
++ order);
++ if (page >= 0) {
++ *dma_handle = mem->device_base + (page << PAGE_SHIFT);
++ ret = mem->virt_base + (page << PAGE_SHIFT);
++ memset(ret, 0, size);
++ return ret;
++ }
++ if (mem->flags & DMA_MEMORY_EXCLUSIVE)
++ return NULL;
++ }
++
++ if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
++ gfp |= GFP_DMA;
++
++ vstart = __get_free_pages(gfp, order);
++ ret = (void *)vstart;
++
++ if (ret != NULL) {
++ /* NB. Hardcode 31 address bits for now: aacraid limitation. */
++ if (xen_create_contiguous_region(vstart, order, 31) != 0) {
++ free_pages(vstart, order);
++ return NULL;
++ }
++ memset(ret, 0, size);
++ *dma_handle = virt_to_bus(ret);
++ }
++ return ret;
++}
++EXPORT_SYMBOL(dma_alloc_coherent);
++
++void dma_free_coherent(struct device *dev, size_t size,
++ void *vaddr, dma_addr_t dma_handle)
++{
++ struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
++ int order = get_order(size);
++
++ if (mem && vaddr >= mem->virt_base && vaddr < (mem->virt_base + (mem->size << PAGE_SHIFT))) {
++ int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
++
++ bitmap_release_region(mem->bitmap, page, order);
++ } else {
++ xen_destroy_contiguous_region((unsigned long)vaddr, order);
++ free_pages((unsigned long)vaddr, order);
++ }
++}
++EXPORT_SYMBOL(dma_free_coherent);
++
++#ifdef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
++int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
++ dma_addr_t device_addr, size_t size, int flags)
++{
++ void __iomem *mem_base;
++ int pages = size >> PAGE_SHIFT;
++ int bitmap_size = (pages + 31)/32;
++
++ if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
++ goto out;
++ if (!size)
++ goto out;
++ if (dev->dma_mem)
++ goto out;
++
++ /* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
++
++ mem_base = ioremap(bus_addr, size);
++ if (!mem_base)
++ goto out;
++
++ dev->dma_mem = kmalloc(sizeof(struct dma_coherent_mem), GFP_KERNEL);
++ if (!dev->dma_mem)
++ goto out;
++ memset(dev->dma_mem, 0, sizeof(struct dma_coherent_mem));
++ dev->dma_mem->bitmap = kmalloc(bitmap_size, GFP_KERNEL);
++ if (!dev->dma_mem->bitmap)
++ goto free1_out;
++ memset(dev->dma_mem->bitmap, 0, bitmap_size);
++
++ dev->dma_mem->virt_base = mem_base;
++ dev->dma_mem->device_base = device_addr;
++ dev->dma_mem->size = pages;
++ dev->dma_mem->flags = flags;
++
++ if (flags & DMA_MEMORY_MAP)
++ return DMA_MEMORY_MAP;
++
++ return DMA_MEMORY_IO;
++
++ free1_out:
++ kfree(dev->dma_mem->bitmap);
++ out:
++ return 0;
++}
++EXPORT_SYMBOL(dma_declare_coherent_memory);
++
++void dma_release_declared_memory(struct device *dev)
++{
++ struct dma_coherent_mem *mem = dev->dma_mem;
++
++ if(!mem)
++ return;
++ dev->dma_mem = NULL;
++ iounmap(mem->virt_base);
++ kfree(mem->bitmap);
++ kfree(mem);
++}
++EXPORT_SYMBOL(dma_release_declared_memory);
++
++void *dma_mark_declared_memory_occupied(struct device *dev,
++ dma_addr_t device_addr, size_t size)
++{
++ struct dma_coherent_mem *mem = dev->dma_mem;
++ int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1) >> PAGE_SHIFT;
++ int pos, err;
++
++ if (!mem)
++ return ERR_PTR(-EINVAL);
++
++ pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
++ err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
++ if (err != 0)
++ return ERR_PTR(err);
++ return mem->virt_base + (pos << PAGE_SHIFT);
++}
++EXPORT_SYMBOL(dma_mark_declared_memory_occupied);
++#endif /* ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY */
++
++dma_addr_t
++dma_map_single(struct device *dev, void *ptr, size_t size,
++ enum dma_data_direction direction)
++{
++ dma_addr_t dma;
++
++ if (direction == DMA_NONE)
++ BUG();
++ WARN_ON(size == 0);
++
++ if (swiotlb) {
++ dma = swiotlb_map_single(dev, ptr, size, direction);
++ } else {
++ dma = virt_to_bus(ptr);
++ IOMMU_BUG_ON(range_straddles_page_boundary(ptr, size));
++ IOMMU_BUG_ON(address_needs_mapping(dev, dma));
++ }
++
++ flush_write_buffers();
++ return dma;
++}
++EXPORT_SYMBOL(dma_map_single);
++
++void
++dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
++ enum dma_data_direction direction)
++{
++ if (direction == DMA_NONE)
++ BUG();
++ if (swiotlb)
++ swiotlb_unmap_single(dev, dma_addr, size, direction);
++}
++EXPORT_SYMBOL(dma_unmap_single);
++
++void
++dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_single_for_cpu(dev, dma_handle, size, direction);
++}
++EXPORT_SYMBOL(dma_sync_single_for_cpu);
++
++void
++dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_single_for_device(dev, dma_handle, size, direction);
++}
++EXPORT_SYMBOL(dma_sync_single_for_device);
+diff -urNp linux-2.6/arch/i386/kernel/process-xen.c new/arch/i386/kernel/process-xen.c
+--- linux-2.6/arch/i386/kernel/process-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/process-xen.c 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,812 @@
++/*
++ * linux/arch/i386/kernel/process.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ *
++ * Pentium III FXSR, SSE support
++ * Gareth Hughes <gareth@valinux.com>, May 2000
++ */
++
++/*
++ * This file handles the architecture-dependent parts of process handling..
++ */
++
++#include <stdarg.h>
++
++#include <linux/cpu.h>
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/fs.h>
++#include <linux/kernel.h>
++#include <linux/mm.h>
++#include <linux/elfcore.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/stddef.h>
++#include <linux/slab.h>
++#include <linux/vmalloc.h>
++#include <linux/user.h>
++#include <linux/a.out.h>
++#include <linux/interrupt.h>
++#include <linux/config.h>
++#include <linux/utsname.h>
++#include <linux/delay.h>
++#include <linux/reboot.h>
++#include <linux/init.h>
++#include <linux/mc146818rtc.h>
++#include <linux/module.h>
++#include <linux/kallsyms.h>
++#include <linux/ptrace.h>
++#include <linux/random.h>
++
++#include <asm/uaccess.h>
++#include <asm/pgtable.h>
++#include <asm/system.h>
++#include <asm/io.h>
++#include <asm/ldt.h>
++#include <asm/processor.h>
++#include <asm/i387.h>
++#include <asm/desc.h>
++#include <asm/vm86.h>
++#ifdef CONFIG_MATH_EMULATION
++#include <asm/math_emu.h>
++#endif
++
++#include <xen/interface/physdev.h>
++#include <xen/interface/vcpu.h>
++#include <xen/cpu_hotplug.h>
++
++#include <linux/err.h>
++
++#include <asm/tlbflush.h>
++#include <asm/cpu.h>
++
++#include <asm/tlbflush.h>
++#include <asm/cpu.h>
++
++asmlinkage void ret_from_fork(void) __asm__("ret_from_fork");
++
++static int hlt_counter;
++
++unsigned long boot_option_idle_override = 0;
++EXPORT_SYMBOL(boot_option_idle_override);
++
++/*
++ * Return saved PC of a blocked thread.
++ */
++unsigned long thread_saved_pc(struct task_struct *tsk)
++{
++ return ((unsigned long *)tsk->thread.esp)[3];
++}
++
++/*
++ * Powermanagement idle function, if any..
++ */
++void (*pm_idle)(void);
++EXPORT_SYMBOL(pm_idle);
++static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
++
++void disable_hlt(void)
++{
++ hlt_counter++;
++}
++
++EXPORT_SYMBOL(disable_hlt);
++
++void enable_hlt(void)
++{
++ hlt_counter--;
++}
++
++EXPORT_SYMBOL(enable_hlt);
++
++/* XXX XEN doesn't use default_idle(), poll_idle(). Use xen_idle() instead. */
++void xen_idle(void)
++{
++ local_irq_disable();
++
++ if (need_resched())
++ local_irq_enable();
++ else {
++ clear_thread_flag(TIF_POLLING_NRFLAG);
++ smp_mb__after_clear_bit();
++ safe_halt();
++ set_thread_flag(TIF_POLLING_NRFLAG);
++ }
++}
++#ifdef CONFIG_APM_MODULE
++EXPORT_SYMBOL(default_idle);
++#endif
++
++#ifdef CONFIG_HOTPLUG_CPU
++extern cpumask_t cpu_initialized;
++static inline void play_dead(void)
++{
++ idle_task_exit();
++ local_irq_disable();
++ cpu_clear(smp_processor_id(), cpu_initialized);
++ preempt_enable_no_resched();
++ HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
++ cpu_bringup();
++}
++#else
++static inline void play_dead(void)
++{
++ BUG();
++}
++#endif /* CONFIG_HOTPLUG_CPU */
++
++/*
++ * The idle thread. There's no useful work to be
++ * done, so just try to conserve power and have a
++ * low exit latency (ie sit in a loop waiting for
++ * somebody to say that they'd like to reschedule)
++ */
++void cpu_idle(void)
++{
++ int cpu = smp_processor_id();
++
++ set_thread_flag(TIF_POLLING_NRFLAG);
++
++ /* endless idle loop with no priority at all */
++ while (1) {
++ while (!need_resched()) {
++
++ if (__get_cpu_var(cpu_idle_state))
++ __get_cpu_var(cpu_idle_state) = 0;
++
++ rmb();
++
++ if (cpu_is_offline(cpu))
++ play_dead();
++
++ __get_cpu_var(irq_stat).idle_timestamp = jiffies;
++ xen_idle();
++ }
++ preempt_enable_no_resched();
++ schedule();
++ preempt_disable();
++ }
++}
++
++void cpu_idle_wait(void)
++{
++ unsigned int cpu, this_cpu = get_cpu();
++ cpumask_t map;
++
++ set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
++ put_cpu();
++
++ cpus_clear(map);
++ for_each_online_cpu(cpu) {
++ per_cpu(cpu_idle_state, cpu) = 1;
++ cpu_set(cpu, map);
++ }
++
++ __get_cpu_var(cpu_idle_state) = 0;
++
++ wmb();
++ do {
++ ssleep(1);
++ for_each_online_cpu(cpu) {
++ if (cpu_isset(cpu, map) && !per_cpu(cpu_idle_state, cpu))
++ cpu_clear(cpu, map);
++ }
++ cpus_and(map, map, cpu_online_map);
++ } while (!cpus_empty(map));
++}
++EXPORT_SYMBOL_GPL(cpu_idle_wait);
++
++/* XXX XEN doesn't use mwait_idle(), select_idle_routine(), idle_setup(). */
++/* Always use xen_idle() instead. */
++void __devinit select_idle_routine(const struct cpuinfo_x86 *c) {}
++
++void show_regs(struct pt_regs * regs)
++{
++ unsigned long cr0 = 0L, cr2 = 0L, cr3 = 0L, cr4 = 0L;
++
++ printk("\n");
++ printk("Pid: %d, comm: %20s\n", current->pid, current->comm);
++ printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
++ print_symbol("EIP is at %s\n", regs->eip);
++
++ if (user_mode_vm(regs))
++ printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp);
++ printk(" EFLAGS: %08lx %s (%s %.*s)\n",
++ regs->eflags, print_tainted(), system_utsname.release,
++ (int)strcspn(system_utsname.version, " "),
++ system_utsname.version);
++ printk("EAX: %08lx EBX: %08lx ECX: %08lx EDX: %08lx\n",
++ regs->eax,regs->ebx,regs->ecx,regs->edx);
++ printk("ESI: %08lx EDI: %08lx EBP: %08lx",
++ regs->esi, regs->edi, regs->ebp);
++ printk(" DS: %04x ES: %04x\n",
++ 0xffff & regs->xds,0xffff & regs->xes);
++
++ cr0 = read_cr0();
++ cr2 = read_cr2();
++ cr3 = read_cr3();
++ cr4 = read_cr4_safe();
++ printk("CR0: %08lx CR2: %08lx CR3: %08lx CR4: %08lx\n", cr0, cr2, cr3, cr4);
++ show_trace(NULL, ®s->esp);
++}
++
++/*
++ * This gets run with %ebx containing the
++ * function to call, and %edx containing
++ * the "args".
++ */
++extern void kernel_thread_helper(void);
++__asm__(".section .text\n"
++ ".align 4\n"
++ "kernel_thread_helper:\n\t"
++ "movl %edx,%eax\n\t"
++ "pushl %edx\n\t"
++ "call *%ebx\n\t"
++ "pushl %eax\n\t"
++ "call do_exit\n"
++ ".previous");
++
++/*
++ * Create a kernel thread
++ */
++int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
++{
++ struct pt_regs regs;
++
++ memset(®s, 0, sizeof(regs));
++
++ regs.ebx = (unsigned long) fn;
++ regs.edx = (unsigned long) arg;
++
++ regs.xds = __USER_DS;
++ regs.xes = __USER_DS;
++ regs.orig_eax = -1;
++ regs.eip = (unsigned long) kernel_thread_helper;
++ regs.xcs = GET_KERNEL_CS();
++ regs.eflags = X86_EFLAGS_IF | X86_EFLAGS_SF | X86_EFLAGS_PF | 0x2;
++
++ /* Ok, create the new process.. */
++ return do_fork(flags | CLONE_VM | CLONE_UNTRACED, 0, ®s, 0, NULL, NULL);
++}
++EXPORT_SYMBOL(kernel_thread);
++
++/*
++ * Free current thread data structures etc..
++ */
++void exit_thread(void)
++{
++ struct task_struct *tsk = current;
++ struct thread_struct *t = &tsk->thread;
++
++ /* The process may have allocated an io port bitmap... nuke it. */
++ if (unlikely(NULL != t->io_bitmap_ptr)) {
++ struct physdev_set_iobitmap set_iobitmap = { 0 };
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap, &set_iobitmap);
++ kfree(t->io_bitmap_ptr);
++ t->io_bitmap_ptr = NULL;
++ }
++}
++
++void flush_thread(void)
++{
++ struct task_struct *tsk = current;
++
++ memset(tsk->thread.debugreg, 0, sizeof(unsigned long)*8);
++ memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
++ /*
++ * Forget coprocessor state..
++ */
++ clear_fpu(tsk);
++ clear_used_math();
++}
++
++void release_thread(struct task_struct *dead_task)
++{
++ BUG_ON(dead_task->mm);
++ release_vm86_irqs(dead_task);
++}
++
++/*
++ * This gets called before we allocate a new thread and copy
++ * the current task into it.
++ */
++void prepare_to_copy(struct task_struct *tsk)
++{
++ unlazy_fpu(tsk);
++}
++
++int copy_thread(int nr, unsigned long clone_flags, unsigned long esp,
++ unsigned long unused,
++ struct task_struct * p, struct pt_regs * regs)
++{
++ struct pt_regs * childregs;
++ struct task_struct *tsk;
++ int err;
++
++ childregs = task_pt_regs(p);
++ *childregs = *regs;
++ childregs->eax = 0;
++ childregs->esp = esp;
++
++ p->thread.esp = (unsigned long) childregs;
++ p->thread.esp0 = (unsigned long) (childregs+1);
++
++ p->thread.eip = (unsigned long) ret_from_fork;
++
++ savesegment(fs,p->thread.fs);
++ savesegment(gs,p->thread.gs);
++
++ tsk = current;
++ if (unlikely(NULL != tsk->thread.io_bitmap_ptr)) {
++ p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
++ if (!p->thread.io_bitmap_ptr) {
++ p->thread.io_bitmap_max = 0;
++ return -ENOMEM;
++ }
++ memcpy(p->thread.io_bitmap_ptr, tsk->thread.io_bitmap_ptr,
++ IO_BITMAP_BYTES);
++ }
++
++ /*
++ * Set a new TLS for the child thread?
++ */
++ if (clone_flags & CLONE_SETTLS) {
++ struct desc_struct *desc;
++ struct user_desc info;
++ int idx;
++
++ err = -EFAULT;
++ if (copy_from_user(&info, (void __user *)childregs->esi, sizeof(info)))
++ goto out;
++ err = -EINVAL;
++ if (LDT_empty(&info))
++ goto out;
++
++ idx = info.entry_number;
++ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
++ goto out;
++
++ desc = p->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
++ desc->a = LDT_entry_a(&info);
++ desc->b = LDT_entry_b(&info);
++ }
++
++ p->thread.iopl = current->thread.iopl;
++
++ err = 0;
++ out:
++ if (err && p->thread.io_bitmap_ptr) {
++ kfree(p->thread.io_bitmap_ptr);
++ p->thread.io_bitmap_max = 0;
++ }
++ return err;
++}
++
++/*
++ * fill in the user structure for a core dump..
++ */
++void dump_thread(struct pt_regs * regs, struct user * dump)
++{
++ int i;
++
++/* changed the size calculations - should hopefully work better. lbt */
++ dump->magic = CMAGIC;
++ dump->start_code = 0;
++ dump->start_stack = regs->esp & ~(PAGE_SIZE - 1);
++ dump->u_tsize = ((unsigned long) current->mm->end_code) >> PAGE_SHIFT;
++ dump->u_dsize = ((unsigned long) (current->mm->brk + (PAGE_SIZE-1))) >> PAGE_SHIFT;
++ dump->u_dsize -= dump->u_tsize;
++ dump->u_ssize = 0;
++ for (i = 0; i < 8; i++)
++ dump->u_debugreg[i] = current->thread.debugreg[i];
++
++ if (dump->start_stack < TASK_SIZE)
++ dump->u_ssize = ((unsigned long) (TASK_SIZE - dump->start_stack)) >> PAGE_SHIFT;
++
++ dump->regs.ebx = regs->ebx;
++ dump->regs.ecx = regs->ecx;
++ dump->regs.edx = regs->edx;
++ dump->regs.esi = regs->esi;
++ dump->regs.edi = regs->edi;
++ dump->regs.ebp = regs->ebp;
++ dump->regs.eax = regs->eax;
++ dump->regs.ds = regs->xds;
++ dump->regs.es = regs->xes;
++ savesegment(fs,dump->regs.fs);
++ savesegment(gs,dump->regs.gs);
++ dump->regs.orig_eax = regs->orig_eax;
++ dump->regs.eip = regs->eip;
++ dump->regs.cs = regs->xcs;
++ dump->regs.eflags = regs->eflags;
++ dump->regs.esp = regs->esp;
++ dump->regs.ss = regs->xss;
++
++ dump->u_fpvalid = dump_fpu (regs, &dump->i387);
++}
++EXPORT_SYMBOL(dump_thread);
++
++/*
++ * Capture the user space registers if the task is not running (in user space)
++ */
++int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
++{
++ struct pt_regs ptregs = *task_pt_regs(tsk);
++ ptregs.xcs &= 0xffff;
++ ptregs.xds &= 0xffff;
++ ptregs.xes &= 0xffff;
++ ptregs.xss &= 0xffff;
++
++ elf_core_copy_regs(regs, &ptregs);
++
++ return 1;
++}
++
++/*
++ * This function selects if the context switch from prev to next
++ * has to tweak the TSC disable bit in the cr4.
++ */
++static inline void disable_tsc(struct task_struct *prev_p,
++ struct task_struct *next_p)
++{
++ struct thread_info *prev, *next;
++
++ /*
++ * gcc should eliminate the ->thread_info dereference if
++ * has_secure_computing returns 0 at compile time (SECCOMP=n).
++ */
++ prev = task_thread_info(prev_p);
++ next = task_thread_info(next_p);
++
++ if (has_secure_computing(prev) || has_secure_computing(next)) {
++ /* slow path here */
++ if (has_secure_computing(prev) &&
++ !has_secure_computing(next)) {
++ write_cr4(read_cr4() & ~X86_CR4_TSD);
++ } else if (!has_secure_computing(prev) &&
++ has_secure_computing(next))
++ write_cr4(read_cr4() | X86_CR4_TSD);
++ }
++}
++
++/*
++ * switch_to(x,yn) should switch tasks from x to y.
++ *
++ * We fsave/fwait so that an exception goes off at the right time
++ * (as a call from the fsave or fwait in effect) rather than to
++ * the wrong process. Lazy FP saving no longer makes any sense
++ * with modern CPU's, and this simplifies a lot of things (SMP
++ * and UP become the same).
++ *
++ * NOTE! We used to use the x86 hardware context switching. The
++ * reason for not using it any more becomes apparent when you
++ * try to recover gracefully from saved state that is no longer
++ * valid (stale segment register values in particular). With the
++ * hardware task-switch, there is no way to fix up bad state in
++ * a reasonable manner.
++ *
++ * The fact that Intel documents the hardware task-switching to
++ * be slow is a fairly red herring - this code is not noticeably
++ * faster. However, there _is_ some room for improvement here,
++ * so the performance issues may eventually be a valid point.
++ * More important, however, is the fact that this allows us much
++ * more flexibility.
++ *
++ * The return value (in %eax) will be the "prev" task after
++ * the task-switch, and shows up in ret_from_fork in entry.S,
++ * for example.
++ */
++struct task_struct fastcall * __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
++{
++ struct thread_struct *prev = &prev_p->thread,
++ *next = &next_p->thread;
++ int cpu = smp_processor_id();
++#ifndef CONFIG_X86_NO_TSS
++ struct tss_struct *tss = &per_cpu(init_tss, cpu);
++#endif
++ struct physdev_set_iopl iopl_op;
++ struct physdev_set_iobitmap iobmp_op;
++ multicall_entry_t _mcl[8], *mcl = _mcl;
++
++ /* XEN NOTE: FS/GS saved in switch_mm(), not here. */
++
++ /*
++ * This is basically '__unlazy_fpu', except that we queue a
++ * multicall to indicate FPU task switch, rather than
++ * synchronously trapping to Xen.
++ */
++ if (prev_p->thread_info->status & TS_USEDFPU) {
++ __save_init_fpu(prev_p); /* _not_ save_init_fpu() */
++ mcl->op = __HYPERVISOR_fpu_taskswitch;
++ mcl->args[0] = 1;
++ mcl++;
++ }
++#if 0 /* lazy fpu sanity check */
++ else BUG_ON(!(read_cr0() & 8));
++#endif
++
++ /*
++ * Reload esp0.
++ * This is load_esp0(tss, next) with a multicall.
++ */
++ mcl->op = __HYPERVISOR_stack_switch;
++ mcl->args[0] = __KERNEL_DS;
++ mcl->args[1] = next->esp0;
++ mcl++;
++
++ /*
++ * Load the per-thread Thread-Local Storage descriptor.
++ * This is load_TLS(next, cpu) with multicalls.
++ */
++#define C(i) do { \
++ if (unlikely(next->tls_array[i].a != prev->tls_array[i].a || \
++ next->tls_array[i].b != prev->tls_array[i].b)) { \
++ mcl->op = __HYPERVISOR_update_descriptor; \
++ *(u64 *)&mcl->args[0] = virt_to_machine( \
++ &get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i]);\
++ *(u64 *)&mcl->args[2] = *(u64 *)&next->tls_array[i]; \
++ mcl++; \
++ } \
++} while (0)
++ C(0); C(1); C(2);
++#undef C
++
++ if (unlikely(prev->iopl != next->iopl)) {
++ iopl_op.iopl = (next->iopl == 0) ? 1 : (next->iopl >> 12) & 3;
++ mcl->op = __HYPERVISOR_physdev_op;
++ mcl->args[0] = PHYSDEVOP_set_iopl;
++ mcl->args[1] = (unsigned long)&iopl_op;
++ mcl++;
++ }
++
++ if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
++ iobmp_op.bitmap = (char *)next->io_bitmap_ptr;
++ iobmp_op.nr_ports = next->io_bitmap_ptr ? IO_BITMAP_BITS : 0;
++ mcl->op = __HYPERVISOR_physdev_op;
++ mcl->args[0] = PHYSDEVOP_set_iobitmap;
++ mcl->args[1] = (unsigned long)&iobmp_op;
++ mcl++;
++ }
++
++ (void)HYPERVISOR_multicall(_mcl, mcl - _mcl);
++
++ /*
++ * Restore %fs and %gs if needed.
++ *
++ * Glibc normally makes %fs be zero, and %gs is one of
++ * the TLS segments.
++ */
++ if (unlikely(next->fs))
++ loadsegment(fs, next->fs);
++
++ if (next->gs)
++ loadsegment(gs, next->gs);
++
++ /*
++ * Now maybe reload the debug registers
++ */
++ if (unlikely(next->debugreg[7])) {
++ set_debugreg(next->debugreg[0], 0);
++ set_debugreg(next->debugreg[1], 1);
++ set_debugreg(next->debugreg[2], 2);
++ set_debugreg(next->debugreg[3], 3);
++ /* no 4 and 5 */
++ set_debugreg(next->debugreg[6], 6);
++ set_debugreg(next->debugreg[7], 7);
++ }
++
++ disable_tsc(prev_p, next_p);
++
++ return prev_p;
++}
++
++asmlinkage int sys_fork(struct pt_regs regs)
++{
++ return do_fork(SIGCHLD, regs.esp, ®s, 0, NULL, NULL);
++}
++
++asmlinkage int sys_clone(struct pt_regs regs)
++{
++ unsigned long clone_flags;
++ unsigned long newsp;
++ int __user *parent_tidptr, *child_tidptr;
++
++ clone_flags = regs.ebx;
++ newsp = regs.ecx;
++ parent_tidptr = (int __user *)regs.edx;
++ child_tidptr = (int __user *)regs.edi;
++ if (!newsp)
++ newsp = regs.esp;
++ return do_fork(clone_flags, newsp, ®s, 0, parent_tidptr, child_tidptr);
++}
++
++/*
++ * This is trivial, and on the face of it looks like it
++ * could equally well be done in user mode.
++ *
++ * Not so, for quite unobvious reasons - register pressure.
++ * In user mode vfork() cannot have a stack frame, and if
++ * done by calling the "clone()" system call directly, you
++ * do not have enough call-clobbered registers to hold all
++ * the information you need.
++ */
++asmlinkage int sys_vfork(struct pt_regs regs)
++{
++ return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs.esp, ®s, 0, NULL, NULL);
++}
++
++/*
++ * sys_execve() executes a new program.
++ */
++asmlinkage int sys_execve(struct pt_regs regs)
++{
++ int error;
++ char * filename;
++
++ filename = getname((char __user *) regs.ebx);
++ error = PTR_ERR(filename);
++ if (IS_ERR(filename))
++ goto out;
++ error = do_execve(filename,
++ (char __user * __user *) regs.ecx,
++ (char __user * __user *) regs.edx,
++ ®s);
++ if (error == 0) {
++ task_lock(current);
++ current->ptrace &= ~PT_DTRACE;
++ task_unlock(current);
++ /* Make sure we don't return using sysenter.. */
++ set_thread_flag(TIF_IRET);
++ }
++ putname(filename);
++out:
++ return error;
++}
++
++#define top_esp (THREAD_SIZE - sizeof(unsigned long))
++#define top_ebp (THREAD_SIZE - 2*sizeof(unsigned long))
++
++unsigned long get_wchan(struct task_struct *p)
++{
++ unsigned long ebp, esp, eip;
++ unsigned long stack_page;
++ int count = 0;
++ if (!p || p == current || p->state == TASK_RUNNING)
++ return 0;
++ stack_page = (unsigned long)task_stack_page(p);
++ esp = p->thread.esp;
++ if (!stack_page || esp < stack_page || esp > top_esp+stack_page)
++ return 0;
++ /* include/asm-i386/system.h:switch_to() pushes ebp last. */
++ ebp = *(unsigned long *) esp;
++ do {
++ if (ebp < stack_page || ebp > top_ebp+stack_page)
++ return 0;
++ eip = *(unsigned long *) (ebp+4);
++ if (!in_sched_functions(eip))
++ return eip;
++ ebp = *(unsigned long *) ebp;
++ } while (count++ < 16);
++ return 0;
++}
++
++/*
++ * sys_alloc_thread_area: get a yet unused TLS descriptor index.
++ */
++static int get_free_idx(void)
++{
++ struct thread_struct *t = ¤t->thread;
++ int idx;
++
++ for (idx = 0; idx < GDT_ENTRY_TLS_ENTRIES; idx++)
++ if (desc_empty(t->tls_array + idx))
++ return idx + GDT_ENTRY_TLS_MIN;
++ return -ESRCH;
++}
++
++/*
++ * Set a given TLS descriptor:
++ */
++asmlinkage int sys_set_thread_area(struct user_desc __user *u_info)
++{
++ struct thread_struct *t = ¤t->thread;
++ struct user_desc info;
++ struct desc_struct *desc;
++ int cpu, idx;
++
++ if (copy_from_user(&info, u_info, sizeof(info)))
++ return -EFAULT;
++ idx = info.entry_number;
++
++ /*
++ * index -1 means the kernel should try to find and
++ * allocate an empty descriptor:
++ */
++ if (idx == -1) {
++ idx = get_free_idx();
++ if (idx < 0)
++ return idx;
++ if (put_user(idx, &u_info->entry_number))
++ return -EFAULT;
++ }
++
++ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
++ return -EINVAL;
++
++ desc = t->tls_array + idx - GDT_ENTRY_TLS_MIN;
++
++ /*
++ * We must not get preempted while modifying the TLS.
++ */
++ cpu = get_cpu();
++
++ if (LDT_empty(&info)) {
++ desc->a = 0;
++ desc->b = 0;
++ } else {
++ desc->a = LDT_entry_a(&info);
++ desc->b = LDT_entry_b(&info);
++ }
++ load_TLS(t, cpu);
++
++ put_cpu();
++
++ return 0;
++}
++
++/*
++ * Get the current Thread-Local Storage area:
++ */
++
++#define GET_BASE(desc) ( \
++ (((desc)->a >> 16) & 0x0000ffff) | \
++ (((desc)->b << 16) & 0x00ff0000) | \
++ ( (desc)->b & 0xff000000) )
++
++#define GET_LIMIT(desc) ( \
++ ((desc)->a & 0x0ffff) | \
++ ((desc)->b & 0xf0000) )
++
++#define GET_32BIT(desc) (((desc)->b >> 22) & 1)
++#define GET_CONTENTS(desc) (((desc)->b >> 10) & 3)
++#define GET_WRITABLE(desc) (((desc)->b >> 9) & 1)
++#define GET_LIMIT_PAGES(desc) (((desc)->b >> 23) & 1)
++#define GET_PRESENT(desc) (((desc)->b >> 15) & 1)
++#define GET_USEABLE(desc) (((desc)->b >> 20) & 1)
++
++asmlinkage int sys_get_thread_area(struct user_desc __user *u_info)
++{
++ struct user_desc info;
++ struct desc_struct *desc;
++ int idx;
++
++ if (get_user(idx, &u_info->entry_number))
++ return -EFAULT;
++ if (idx < GDT_ENTRY_TLS_MIN || idx > GDT_ENTRY_TLS_MAX)
++ return -EINVAL;
++
++ memset(&info, 0, sizeof(info));
++
++ desc = current->thread.tls_array + idx - GDT_ENTRY_TLS_MIN;
++
++ info.entry_number = idx;
++ info.base_addr = GET_BASE(desc);
++ info.limit = GET_LIMIT(desc);
++ info.seg_32bit = GET_32BIT(desc);
++ info.contents = GET_CONTENTS(desc);
++ info.read_exec_only = !GET_WRITABLE(desc);
++ info.limit_in_pages = GET_LIMIT_PAGES(desc);
++ info.seg_not_present = !GET_PRESENT(desc);
++ info.useable = GET_USEABLE(desc);
++
++ if (copy_to_user(u_info, &info, sizeof(info)))
++ return -EFAULT;
++ return 0;
++}
++
++unsigned long arch_align_stack(unsigned long sp)
++{
++ if (randomize_va_space)
++ sp -= get_random_int() % 8192;
++ return sp & ~0xf;
++}
+diff -urNp linux-2.6/arch/i386/kernel/quirks-xen.c new/arch/i386/kernel/quirks-xen.c
+--- linux-2.6/arch/i386/kernel/quirks-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/quirks-xen.c 2006-05-09 12:32:35.000000000 +0200
+@@ -0,0 +1,48 @@
++/*
++ * This file contains work-arounds for x86 and x86_64 platform bugs.
++ */
++#include <linux/config.h>
++#include <linux/pci.h>
++#include <linux/irq.h>
++
++#if defined(CONFIG_X86_IO_APIC) && (defined(CONFIG_SMP) || defined(CONFIG_XEN)) && defined(CONFIG_PCI)
++
++static void __devinit quirk_intel_irqbalance(struct pci_dev *dev)
++{
++ u8 config, rev;
++ u32 word;
++
++ /* BIOS may enable hardware IRQ balancing for
++ * E7520/E7320/E7525(revision ID 0x9 and below)
++ * based platforms.
++ * Disable SW irqbalance/affinity on those platforms.
++ */
++ pci_read_config_byte(dev, PCI_CLASS_REVISION, &rev);
++ if (rev > 0x9)
++ return;
++
++ printk(KERN_INFO "Intel E7520/7320/7525 detected.");
++
++ /* enable access to config space*/
++ pci_read_config_byte(dev, 0xf4, &config);
++ pci_write_config_byte(dev, 0xf4, config|0x2);
++
++ /* read xTPR register */
++ raw_pci_ops->read(0, 0, 0x40, 0x4c, 2, &word);
++
++ if (!(word & (1 << 13))) {
++ dom0_op_t op;
++ printk(KERN_INFO "Disabling irq balancing and affinity\n");
++ op.cmd = DOM0_PLATFORM_QUIRK;
++ op.u.platform_quirk.quirk_id = QUIRK_NOIRQBALANCING;
++ (void)HYPERVISOR_dom0_op(&op);
++ }
++
++ /* put back the original value for config space*/
++ if (!(config & 0x2))
++ pci_write_config_byte(dev, 0xf4, config);
++}
++DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7320_MCH, quirk_intel_irqbalance);
++DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7525_MCH, quirk_intel_irqbalance);
++DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_E7520_MCH, quirk_intel_irqbalance);
++#endif
+diff -urNp linux-2.6/arch/i386/kernel/setup-xen.c new/arch/i386/kernel/setup-xen.c
+--- linux-2.6/arch/i386/kernel/setup-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/setup-xen.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,1827 @@
++/*
++ * linux/arch/i386/kernel/setup.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ *
++ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
++ *
++ * Memory region support
++ * David Parsons <orc@pell.chi.il.us>, July-August 1999
++ *
++ * Added E820 sanitization routine (removes overlapping memory regions);
++ * Brian Moyle <bmoyle@mvista.com>, February 2001
++ *
++ * Moved CPU detection code to cpu/${cpu}.c
++ * Patrick Mochel <mochel@osdl.org>, March 2002
++ *
++ * Provisions for empty E820 memory regions (reported by certain BIOSes).
++ * Alex Achenbach <xela@slit.de>, December 2002.
++ *
++ */
++
++/*
++ * This file handles the architecture-dependent parts of initialization
++ */
++
++#include <linux/config.h>
++#include <linux/sched.h>
++#include <linux/mm.h>
++#include <linux/mmzone.h>
++#include <linux/tty.h>
++#include <linux/ioport.h>
++#include <linux/acpi.h>
++#include <linux/apm_bios.h>
++#include <linux/initrd.h>
++#include <linux/bootmem.h>
++#include <linux/seq_file.h>
++#include <linux/platform_device.h>
++#include <linux/console.h>
++#include <linux/mca.h>
++#include <linux/root_dev.h>
++#include <linux/highmem.h>
++#include <linux/module.h>
++#include <linux/efi.h>
++#include <linux/init.h>
++#include <linux/edd.h>
++#include <linux/nodemask.h>
++#include <linux/kernel.h>
++#include <linux/percpu.h>
++#include <linux/notifier.h>
++#include <linux/kexec.h>
++#include <linux/crash_dump.h>
++#include <linux/dmi.h>
++#include <linux/pfn.h>
++
++#include <video/edid.h>
++
++#include <asm/apic.h>
++#include <asm/e820.h>
++#include <asm/mpspec.h>
++#include <asm/setup.h>
++#include <asm/arch_hooks.h>
++#include <asm/sections.h>
++#include <asm/io_apic.h>
++#include <asm/ist.h>
++#include <asm/io.h>
++#include <asm/hypervisor.h>
++#include <xen/interface/physdev.h>
++#include <xen/interface/memory.h>
++#include <xen/features.h>
++#include "setup_arch_pre.h"
++#include <bios_ebda.h>
++
++/* Forward Declaration. */
++void __init find_max_pfn(void);
++
++static int xen_panic_event(struct notifier_block *, unsigned long, void *);
++static struct notifier_block xen_panic_block = {
++ xen_panic_event, NULL, 0 /* try to go last */
++};
++
++extern char hypercall_page[PAGE_SIZE];
++EXPORT_SYMBOL(hypercall_page);
++
++int disable_pse __devinitdata = 0;
++
++/*
++ * Machine setup..
++ */
++
++#ifdef CONFIG_EFI
++int efi_enabled = 0;
++EXPORT_SYMBOL(efi_enabled);
++#endif
++
++/* cpu data as detected by the assembly code in head.S */
++struct cpuinfo_x86 new_cpu_data __initdata = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
++/* common cpu data for all cpus */
++struct cpuinfo_x86 boot_cpu_data __read_mostly = { 0, 0, 0, 0, -1, 1, 0, 0, -1 };
++EXPORT_SYMBOL(boot_cpu_data);
++
++unsigned long mmu_cr4_features;
++
++#ifdef CONFIG_ACPI
++ int acpi_disabled = 0;
++#else
++ int acpi_disabled = 1;
++#endif
++EXPORT_SYMBOL(acpi_disabled);
++
++#ifdef CONFIG_ACPI
++int __initdata acpi_force = 0;
++extern acpi_interrupt_flags acpi_sci_flags;
++#endif
++
++/* for MCA, but anyone else can use it if they want */
++unsigned int machine_id;
++#ifdef CONFIG_MCA
++EXPORT_SYMBOL(machine_id);
++#endif
++unsigned int machine_submodel_id;
++unsigned int BIOS_revision;
++unsigned int mca_pentium_flag;
++
++/* For PCI or other memory-mapped resources */
++unsigned long pci_mem_start = 0x10000000;
++#ifdef CONFIG_PCI
++EXPORT_SYMBOL(pci_mem_start);
++#endif
++
++/* Boot loader ID as an integer, for the benefit of proc_dointvec */
++int bootloader_type;
++
++/* user-defined highmem size */
++static unsigned int highmem_pages = -1;
++
++/*
++ * Setup options
++ */
++struct drive_info_struct { char dummy[32]; } drive_info;
++#if defined(CONFIG_BLK_DEV_IDE) || defined(CONFIG_BLK_DEV_HD) || \
++ defined(CONFIG_BLK_DEV_IDE_MODULE) || defined(CONFIG_BLK_DEV_HD_MODULE)
++EXPORT_SYMBOL(drive_info);
++#endif
++struct screen_info screen_info;
++EXPORT_SYMBOL(screen_info);
++struct apm_info apm_info;
++EXPORT_SYMBOL(apm_info);
++struct sys_desc_table_struct {
++ unsigned short length;
++ unsigned char table[0];
++};
++struct edid_info edid_info;
++EXPORT_SYMBOL_GPL(edid_info);
++struct ist_info ist_info;
++#if defined(CONFIG_X86_SPEEDSTEP_SMI) || \
++ defined(CONFIG_X86_SPEEDSTEP_SMI_MODULE)
++EXPORT_SYMBOL(ist_info);
++#endif
++struct e820map e820;
++
++extern void early_cpu_init(void);
++extern void generic_apic_probe(char *);
++extern int root_mountflags;
++
++unsigned long saved_videomode;
++
++#define RAMDISK_IMAGE_START_MASK 0x07FF
++#define RAMDISK_PROMPT_FLAG 0x8000
++#define RAMDISK_LOAD_FLAG 0x4000
++
++static char command_line[COMMAND_LINE_SIZE];
++
++unsigned char __initdata boot_params[PARAM_SIZE];
++
++static struct resource data_resource = {
++ .name = "Kernel data",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
++};
++
++static struct resource code_resource = {
++ .name = "Kernel code",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
++};
++
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++static struct resource system_rom_resource = {
++ .name = "System ROM",
++ .start = 0xf0000,
++ .end = 0xfffff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++};
++
++static struct resource extension_rom_resource = {
++ .name = "Extension ROM",
++ .start = 0xe0000,
++ .end = 0xeffff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++};
++
++static struct resource adapter_rom_resources[] = { {
++ .name = "Adapter ROM",
++ .start = 0xc8000,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++}, {
++ .name = "Adapter ROM",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++}, {
++ .name = "Adapter ROM",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++}, {
++ .name = "Adapter ROM",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++}, {
++ .name = "Adapter ROM",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++}, {
++ .name = "Adapter ROM",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++} };
++
++#define ADAPTER_ROM_RESOURCES \
++ (sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
++
++static struct resource video_rom_resource = {
++ .name = "Video ROM",
++ .start = 0xc0000,
++ .end = 0xc7fff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM
++};
++#endif
++
++static struct resource video_ram_resource = {
++ .name = "Video RAM area",
++ .start = 0xa0000,
++ .end = 0xbffff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_MEM
++};
++
++static struct resource standard_io_resources[] = { {
++ .name = "dma1",
++ .start = 0x0000,
++ .end = 0x001f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "pic1",
++ .start = 0x0020,
++ .end = 0x0021,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "timer0",
++ .start = 0x0040,
++ .end = 0x0043,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "timer1",
++ .start = 0x0050,
++ .end = 0x0053,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "keyboard",
++ .start = 0x0060,
++ .end = 0x006f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "dma page reg",
++ .start = 0x0080,
++ .end = 0x008f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "pic2",
++ .start = 0x00a0,
++ .end = 0x00a1,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "dma2",
++ .start = 0x00c0,
++ .end = 0x00df,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++}, {
++ .name = "fpu",
++ .start = 0x00f0,
++ .end = 0x00ff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO
++} };
++
++#define STANDARD_IO_RESOURCES \
++ (sizeof standard_io_resources / sizeof standard_io_resources[0])
++
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++#define romsignature(x) (*(unsigned short *)(x) == 0xaa55)
++
++static int __init romchecksum(unsigned char *rom, unsigned long length)
++{
++ unsigned char *p, sum = 0;
++
++ for (p = rom; p < rom + length; p++)
++ sum += *p;
++ return sum == 0;
++}
++
++static void __init probe_roms(void)
++{
++ unsigned long start, length, upper;
++ unsigned char *rom;
++ int i;
++
++ /* Nothing to do if not running in dom0. */
++ if (!(xen_start_info->flags & SIF_INITDOMAIN))
++ return;
++
++ /* video rom */
++ upper = adapter_rom_resources[0].start;
++ for (start = video_rom_resource.start; start < upper; start += 2048) {
++ rom = isa_bus_to_virt(start);
++ if (!romsignature(rom))
++ continue;
++
++ video_rom_resource.start = start;
++
++ /* 0 < length <= 0x7f * 512, historically */
++ length = rom[2] * 512;
++
++ /* if checksum okay, trust length byte */
++ if (length && romchecksum(rom, length))
++ video_rom_resource.end = start + length - 1;
++
++ request_resource(&iomem_resource, &video_rom_resource);
++ break;
++ }
++
++ start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
++ if (start < upper)
++ start = upper;
++
++ /* system rom */
++ request_resource(&iomem_resource, &system_rom_resource);
++ upper = system_rom_resource.start;
++
++ /* check for extension rom (ignore length byte!) */
++ rom = isa_bus_to_virt(extension_rom_resource.start);
++ if (romsignature(rom)) {
++ length = extension_rom_resource.end - extension_rom_resource.start + 1;
++ if (romchecksum(rom, length)) {
++ request_resource(&iomem_resource, &extension_rom_resource);
++ upper = extension_rom_resource.start;
++ }
++ }
++
++ /* check for adapter roms on 2k boundaries */
++ for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
++ rom = isa_bus_to_virt(start);
++ if (!romsignature(rom))
++ continue;
++
++ /* 0 < length <= 0x7f * 512, historically */
++ length = rom[2] * 512;
++
++ /* but accept any length that fits if checksum okay */
++ if (!length || start + length > upper || !romchecksum(rom, length))
++ continue;
++
++ adapter_rom_resources[i].start = start;
++ adapter_rom_resources[i].end = start + length - 1;
++ request_resource(&iomem_resource, &adapter_rom_resources[i]);
++
++ start = adapter_rom_resources[i++].end & ~2047UL;
++ }
++}
++#endif
++
++/*
++ * Point at the empty zero page to start with. We map the real shared_info
++ * page as soon as fixmap is up and running.
++ */
++shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
++EXPORT_SYMBOL(HYPERVISOR_shared_info);
++
++unsigned long *phys_to_machine_mapping;
++unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[16];
++EXPORT_SYMBOL(phys_to_machine_mapping);
++
++/* Raw start-of-day parameters from the hypervisor. */
++start_info_t *xen_start_info;
++EXPORT_SYMBOL(xen_start_info);
++
++static void __init add_memory_region(unsigned long long start,
++ unsigned long long size, int type)
++{
++ int x;
++
++ if (!efi_enabled) {
++ x = e820.nr_map;
++
++ if (x == E820MAX) {
++ printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
++ return;
++ }
++
++ e820.map[x].addr = start;
++ e820.map[x].size = size;
++ e820.map[x].type = type;
++ e820.nr_map++;
++ }
++} /* add_memory_region */
++
++static void __init limit_regions(unsigned long long size)
++{
++ unsigned long long current_addr = 0;
++ int i;
++
++ if (efi_enabled) {
++ efi_memory_desc_t *md;
++ void *p;
++
++ for (p = memmap.map, i = 0; p < memmap.map_end;
++ p += memmap.desc_size, i++) {
++ md = p;
++ current_addr = md->phys_addr + (md->num_pages << 12);
++ if (md->type == EFI_CONVENTIONAL_MEMORY) {
++ if (current_addr >= size) {
++ md->num_pages -=
++ (((current_addr-size) + PAGE_SIZE-1) >> PAGE_SHIFT);
++ memmap.nr_map = i + 1;
++ return;
++ }
++ }
++ }
++ }
++ for (i = 0; i < e820.nr_map; i++) {
++ current_addr = e820.map[i].addr + e820.map[i].size;
++ if (current_addr < size)
++ continue;
++
++ if (e820.map[i].type != E820_RAM)
++ continue;
++
++ if (e820.map[i].addr >= size) {
++ /*
++ * This region starts past the end of the
++ * requested size, skip it completely.
++ */
++ e820.nr_map = i;
++ } else {
++ e820.nr_map = i + 1;
++ e820.map[i].size -= current_addr - size;
++ }
++ return;
++ }
++#ifdef CONFIG_XEN
++ if (i==e820.nr_map && current_addr < size) {
++ /*
++ * The e820 map finished before our requested size so
++ * extend the final entry to the requested address.
++ */
++ --i;
++ if (e820.map[i].type == E820_RAM)
++ e820.map[i].size -= current_addr - size;
++ else
++ add_memory_region(current_addr, size - current_addr, E820_RAM);
++ }
++#endif
++}
++
++#define E820_DEBUG 1
++
++static void __init print_memory_map(char *who)
++{
++ int i;
++
++ for (i = 0; i < e820.nr_map; i++) {
++ printk(" %s: %016Lx - %016Lx ", who,
++ e820.map[i].addr,
++ e820.map[i].addr + e820.map[i].size);
++ switch (e820.map[i].type) {
++ case E820_RAM: printk("(usable)\n");
++ break;
++ case E820_RESERVED:
++ printk("(reserved)\n");
++ break;
++ case E820_ACPI:
++ printk("(ACPI data)\n");
++ break;
++ case E820_NVS:
++ printk("(ACPI NVS)\n");
++ break;
++ default: printk("type %lu\n", e820.map[i].type);
++ break;
++ }
++ }
++}
++
++/*
++ * Sanitize the BIOS e820 map.
++ *
++ * Some e820 responses include overlapping entries. The following
++ * replaces the original e820 map with a new one, removing overlaps.
++ *
++ */
++struct change_member {
++ struct e820entry *pbios; /* pointer to original bios entry */
++ unsigned long long addr; /* address for this change point */
++};
++static struct change_member change_point_list[2*E820MAX] __initdata;
++static struct change_member *change_point[2*E820MAX] __initdata;
++static struct e820entry *overlap_list[E820MAX] __initdata;
++static struct e820entry new_bios[E820MAX] __initdata;
++
++static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
++{
++ struct change_member *change_tmp;
++ unsigned long current_type, last_type;
++ unsigned long long last_addr;
++ int chgidx, still_changing;
++ int overlap_entries;
++ int new_bios_entry;
++ int old_nr, new_nr, chg_nr;
++ int i;
++
++ /*
++ Visually we're performing the following (1,2,3,4 = memory types)...
++
++ Sample memory map (w/overlaps):
++ ____22__________________
++ ______________________4_
++ ____1111________________
++ _44_____________________
++ 11111111________________
++ ____________________33__
++ ___________44___________
++ __________33333_________
++ ______________22________
++ ___________________2222_
++ _________111111111______
++ _____________________11_
++ _________________4______
++
++ Sanitized equivalent (no overlap):
++ 1_______________________
++ _44_____________________
++ ___1____________________
++ ____22__________________
++ ______11________________
++ _________1______________
++ __________3_____________
++ ___________44___________
++ _____________33_________
++ _______________2________
++ ________________1_______
++ _________________4______
++ ___________________2____
++ ____________________33__
++ ______________________4_
++ */
++
++ /* if there's only one memory region, don't bother */
++ if (*pnr_map < 2)
++ return -1;
++
++ old_nr = *pnr_map;
++
++ /* bail out if we find any unreasonable addresses in bios map */
++ for (i=0; i<old_nr; i++)
++ if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
++ return -1;
++
++ /* create pointers for initial change-point information (for sorting) */
++ for (i=0; i < 2*old_nr; i++)
++ change_point[i] = &change_point_list[i];
++
++ /* record all known change-points (starting and ending addresses),
++ omitting those that are for empty memory regions */
++ chgidx = 0;
++ for (i=0; i < old_nr; i++) {
++ if (biosmap[i].size != 0) {
++ change_point[chgidx]->addr = biosmap[i].addr;
++ change_point[chgidx++]->pbios = &biosmap[i];
++ change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size;
++ change_point[chgidx++]->pbios = &biosmap[i];
++ }
++ }
++ chg_nr = chgidx; /* true number of change-points */
++
++ /* sort change-point list by memory addresses (low -> high) */
++ still_changing = 1;
++ while (still_changing) {
++ still_changing = 0;
++ for (i=1; i < chg_nr; i++) {
++ /* if <current_addr> > <last_addr>, swap */
++ /* or, if current=<start_addr> & last=<end_addr>, swap */
++ if ((change_point[i]->addr < change_point[i-1]->addr) ||
++ ((change_point[i]->addr == change_point[i-1]->addr) &&
++ (change_point[i]->addr == change_point[i]->pbios->addr) &&
++ (change_point[i-1]->addr != change_point[i-1]->pbios->addr))
++ )
++ {
++ change_tmp = change_point[i];
++ change_point[i] = change_point[i-1];
++ change_point[i-1] = change_tmp;
++ still_changing=1;
++ }
++ }
++ }
++
++ /* create a new bios memory map, removing overlaps */
++ overlap_entries=0; /* number of entries in the overlap table */
++ new_bios_entry=0; /* index for creating new bios map entries */
++ last_type = 0; /* start with undefined memory type */
++ last_addr = 0; /* start with 0 as last starting address */
++ /* loop through change-points, determining affect on the new bios map */
++ for (chgidx=0; chgidx < chg_nr; chgidx++)
++ {
++ /* keep track of all overlapping bios entries */
++ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr)
++ {
++ /* add map entry to overlap list (> 1 entry implies an overlap) */
++ overlap_list[overlap_entries++]=change_point[chgidx]->pbios;
++ }
++ else
++ {
++ /* remove entry from list (order independent, so swap with last) */
++ for (i=0; i<overlap_entries; i++)
++ {
++ if (overlap_list[i] == change_point[chgidx]->pbios)
++ overlap_list[i] = overlap_list[overlap_entries-1];
++ }
++ overlap_entries--;
++ }
++ /* if there are overlapping entries, decide which "type" to use */
++ /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */
++ current_type = 0;
++ for (i=0; i<overlap_entries; i++)
++ if (overlap_list[i]->type > current_type)
++ current_type = overlap_list[i]->type;
++ /* continue building up new bios map based on this information */
++ if (current_type != last_type) {
++ if (last_type != 0) {
++ new_bios[new_bios_entry].size =
++ change_point[chgidx]->addr - last_addr;
++ /* move forward only if the new size was non-zero */
++ if (new_bios[new_bios_entry].size != 0)
++ if (++new_bios_entry >= E820MAX)
++ break; /* no more space left for new bios entries */
++ }
++ if (current_type != 0) {
++ new_bios[new_bios_entry].addr = change_point[chgidx]->addr;
++ new_bios[new_bios_entry].type = current_type;
++ last_addr=change_point[chgidx]->addr;
++ }
++ last_type = current_type;
++ }
++ }
++ new_nr = new_bios_entry; /* retain count for new bios entries */
++
++ /* copy new bios mapping into original location */
++ memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry));
++ *pnr_map = new_nr;
++
++ return 0;
++}
++
++/*
++ * Copy the BIOS e820 map into a safe place.
++ *
++ * Sanity-check it while we're at it..
++ *
++ * If we're lucky and live on a modern system, the setup code
++ * will have given us a memory map that we can use to properly
++ * set up memory. If we aren't, we'll fake a memory map.
++ *
++ * We check to see that the memory map contains at least 2 elements
++ * before we'll use it, because the detection code in setup.S may
++ * not be perfect and most every PC known to man has two memory
++ * regions: one from 0 to 640k, and one from 1mb up. (The IBM
++ * thinkpad 560x, for example, does not cooperate with the memory
++ * detection code.)
++ */
++static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
++{
++#ifndef CONFIG_XEN
++ /* Only one memory region (or negative)? Ignore it */
++ if (nr_map < 2)
++ return -1;
++#else
++ BUG_ON(nr_map < 1);
++#endif
++
++ do {
++ unsigned long long start = biosmap->addr;
++ unsigned long long size = biosmap->size;
++ unsigned long long end = start + size;
++ unsigned long type = biosmap->type;
++
++ /* Overflow in 64 bits? Ignore the memory map. */
++ if (start > end)
++ return -1;
++
++#ifndef CONFIG_XEN
++ /*
++ * Some BIOSes claim RAM in the 640k - 1M region.
++ * Not right. Fix it up.
++ */
++ if (type == E820_RAM) {
++ if (start < 0x100000ULL && end > 0xA0000ULL) {
++ if (start < 0xA0000ULL)
++ add_memory_region(start, 0xA0000ULL-start, type);
++ if (end <= 0x100000ULL)
++ continue;
++ start = 0x100000ULL;
++ size = end - start;
++ }
++ }
++#endif
++ add_memory_region(start, size, type);
++ } while (biosmap++,--nr_map);
++ return 0;
++}
++
++#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
++struct edd edd;
++#ifdef CONFIG_EDD_MODULE
++EXPORT_SYMBOL(edd);
++#endif
++/**
++ * copy_edd() - Copy the BIOS EDD information
++ * from boot_params into a safe place.
++ *
++ */
++static inline void copy_edd(void)
++{
++ memcpy(edd.mbr_signature, EDD_MBR_SIGNATURE, sizeof(edd.mbr_signature));
++ memcpy(edd.edd_info, EDD_BUF, sizeof(edd.edd_info));
++ edd.mbr_signature_nr = EDD_MBR_SIG_NR;
++ edd.edd_info_nr = EDD_NR;
++}
++#else
++static inline void copy_edd(void)
++{
++}
++#endif
++
++/*
++ * Do NOT EVER look at the BIOS memory size location.
++ * It does not work on many machines.
++ */
++#define LOWMEMSIZE() (0x9f000)
++
++static void __init parse_cmdline_early (char ** cmdline_p)
++{
++ char c = ' ', *to = command_line, *from = saved_command_line;
++ int len = 0, max_cmdline;
++ int userdef = 0;
++
++ if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
++ max_cmdline = COMMAND_LINE_SIZE;
++ memcpy(saved_command_line, xen_start_info->cmd_line, max_cmdline);
++ /* Save unparsed command line copy for /proc/cmdline */
++ saved_command_line[max_cmdline-1] = '\0';
++
++ for (;;) {
++ if (c != ' ')
++ goto next_char;
++ /*
++ * "mem=nopentium" disables the 4MB page tables.
++ * "mem=XXX[kKmM]" defines a memory region from HIGH_MEM
++ * to <mem>, overriding the bios size.
++ * "memmap=XXX[KkmM]@XXX[KkmM]" defines a memory region from
++ * <start> to <start>+<mem>, overriding the bios size.
++ *
++ * HPA tells me bootloaders need to parse mem=, so no new
++ * option should be mem= [also see Documentation/i386/boot.txt]
++ */
++ if (!memcmp(from, "mem=", 4)) {
++ if (to != command_line)
++ to--;
++ if (!memcmp(from+4, "nopentium", 9)) {
++ from += 9+4;
++ clear_bit(X86_FEATURE_PSE, boot_cpu_data.x86_capability);
++ disable_pse = 1;
++ } else {
++ /* If the user specifies memory size, we
++ * limit the BIOS-provided memory map to
++ * that size. exactmap can be used to specify
++ * the exact map. mem=number can be used to
++ * trim the existing memory map.
++ */
++ unsigned long long mem_size;
++
++ mem_size = memparse(from+4, &from);
++ limit_regions(mem_size);
++ userdef=1;
++ }
++ }
++
++ else if (!memcmp(from, "memmap=", 7)) {
++ if (to != command_line)
++ to--;
++ if (!memcmp(from+7, "exactmap", 8)) {
++#ifdef CONFIG_CRASH_DUMP
++ /* If we are doing a crash dump, we
++ * still need to know the real mem
++ * size before original memory map is
++ * reset.
++ */
++ find_max_pfn();
++ saved_max_pfn = max_pfn;
++#endif
++ from += 8+7;
++ e820.nr_map = 0;
++ userdef = 1;
++ } else {
++ /* If the user specifies memory size, we
++ * limit the BIOS-provided memory map to
++ * that size. exactmap can be used to specify
++ * the exact map. mem=number can be used to
++ * trim the existing memory map.
++ */
++ unsigned long long start_at, mem_size;
++
++ mem_size = memparse(from+7, &from);
++ if (*from == '@') {
++ start_at = memparse(from+1, &from);
++ add_memory_region(start_at, mem_size, E820_RAM);
++ } else if (*from == '#') {
++ start_at = memparse(from+1, &from);
++ add_memory_region(start_at, mem_size, E820_ACPI);
++ } else if (*from == '$') {
++ start_at = memparse(from+1, &from);
++ add_memory_region(start_at, mem_size, E820_RESERVED);
++ } else {
++ limit_regions(mem_size);
++ userdef=1;
++ }
++ }
++ }
++
++ else if (!memcmp(from, "noexec=", 7))
++ noexec_setup(from + 7);
++
++
++#ifdef CONFIG_X86_MPPARSE
++ /*
++ * If the BIOS enumerates physical processors before logical,
++ * maxcpus=N at enumeration-time can be used to disable HT.
++ */
++ else if (!memcmp(from, "maxcpus=", 8)) {
++ extern unsigned int maxcpus;
++
++ maxcpus = simple_strtoul(from + 8, NULL, 0);
++ }
++#endif
++
++#ifdef CONFIG_ACPI
++ /* "acpi=off" disables both ACPI table parsing and interpreter */
++ else if (!memcmp(from, "acpi=off", 8)) {
++ disable_acpi();
++ }
++
++ /* acpi=force to over-ride black-list */
++ else if (!memcmp(from, "acpi=force", 10)) {
++ acpi_force = 1;
++ acpi_ht = 1;
++ acpi_disabled = 0;
++ }
++
++ /* acpi=strict disables out-of-spec workarounds */
++ else if (!memcmp(from, "acpi=strict", 11)) {
++ acpi_strict = 1;
++ }
++
++ /* Limit ACPI just to boot-time to enable HT */
++ else if (!memcmp(from, "acpi=ht", 7)) {
++ if (!acpi_force)
++ disable_acpi();
++ acpi_ht = 1;
++ }
++
++ /* "pci=noacpi" disable ACPI IRQ routing and PCI scan */
++ else if (!memcmp(from, "pci=noacpi", 10)) {
++ acpi_disable_pci();
++ }
++ /* "acpi=noirq" disables ACPI interrupt routing */
++ else if (!memcmp(from, "acpi=noirq", 10)) {
++ acpi_noirq_set();
++ }
++
++ else if (!memcmp(from, "acpi_sci=edge", 13))
++ acpi_sci_flags.trigger = 1;
++
++ else if (!memcmp(from, "acpi_sci=level", 14))
++ acpi_sci_flags.trigger = 3;
++
++ else if (!memcmp(from, "acpi_sci=high", 13))
++ acpi_sci_flags.polarity = 1;
++
++ else if (!memcmp(from, "acpi_sci=low", 12))
++ acpi_sci_flags.polarity = 3;
++
++#ifdef CONFIG_X86_IO_APIC
++ else if (!memcmp(from, "acpi_skip_timer_override", 24))
++ acpi_skip_timer_override = 1;
++
++ if (!memcmp(from, "disable_timer_pin_1", 19))
++ disable_timer_pin_1 = 1;
++ if (!memcmp(from, "enable_timer_pin_1", 18))
++ disable_timer_pin_1 = -1;
++
++ /* disable IO-APIC */
++ else if (!memcmp(from, "noapic", 6))
++ disable_ioapic_setup();
++#endif /* CONFIG_X86_IO_APIC */
++#endif /* CONFIG_ACPI */
++
++#ifdef CONFIG_X86_LOCAL_APIC
++ /* enable local APIC */
++ else if (!memcmp(from, "lapic", 5))
++ lapic_enable();
++
++ /* disable local APIC */
++ else if (!memcmp(from, "nolapic", 6))
++ lapic_disable();
++#endif /* CONFIG_X86_LOCAL_APIC */
++
++#ifdef CONFIG_KEXEC
++ /* crashkernel=size@addr specifies the location to reserve for
++ * a crash kernel. By reserving this memory we guarantee
++ * that linux never set's it up as a DMA target.
++ * Useful for holding code to do something appropriate
++ * after a kernel panic.
++ */
++ else if (!memcmp(from, "crashkernel=", 12)) {
++ unsigned long size, base;
++ size = memparse(from+12, &from);
++ if (*from == '@') {
++ base = memparse(from+1, &from);
++ /* FIXME: Do I want a sanity check
++ * to validate the memory range?
++ */
++ crashk_res.start = base;
++ crashk_res.end = base + size - 1;
++ }
++ }
++#endif
++#ifdef CONFIG_PROC_VMCORE
++ /* elfcorehdr= specifies the location of elf core header
++ * stored by the crashed kernel.
++ */
++ else if (!memcmp(from, "elfcorehdr=", 11))
++ elfcorehdr_addr = memparse(from+11, &from);
++#endif
++
++ /*
++ * highmem=size forces highmem to be exactly 'size' bytes.
++ * This works even on boxes that have no highmem otherwise.
++ * This also works to reduce highmem size on bigger boxes.
++ */
++ else if (!memcmp(from, "highmem=", 8))
++ highmem_pages = memparse(from+8, &from) >> PAGE_SHIFT;
++
++ /*
++ * vmalloc=size forces the vmalloc area to be exactly 'size'
++ * bytes. This can be used to increase (or decrease) the
++ * vmalloc area - the default is 128m.
++ */
++ else if (!memcmp(from, "vmalloc=", 8))
++ __VMALLOC_RESERVE = memparse(from+8, &from);
++
++ next_char:
++ c = *(from++);
++ if (!c)
++ break;
++ if (COMMAND_LINE_SIZE <= ++len)
++ break;
++ *(to++) = c;
++ }
++ *to = '\0';
++ *cmdline_p = command_line;
++ if (userdef) {
++ printk(KERN_INFO "user-defined physical RAM map:\n");
++ print_memory_map("user");
++ }
++}
++
++/*
++ * Callback for efi_memory_walk.
++ */
++static int __init
++efi_find_max_pfn(unsigned long start, unsigned long end, void *arg)
++{
++ unsigned long *max_pfn = arg, pfn;
++
++ if (start < end) {
++ pfn = PFN_UP(end -1);
++ if (pfn > *max_pfn)
++ *max_pfn = pfn;
++ }
++ return 0;
++}
++
++static int __init
++efi_memory_present_wrapper(unsigned long start, unsigned long end, void *arg)
++{
++ memory_present(0, start, end);
++ return 0;
++}
++
++ /*
++ * This function checks if the entire range <start,end> is mapped with type.
++ *
++ * Note: this function only works correct if the e820 table is sorted and
++ * not-overlapping, which is the case
++ */
++int __init
++e820_all_mapped(unsigned long s, unsigned long e, unsigned type)
++{
++ u64 start = s;
++ u64 end = e;
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ if (type && ei->type != type)
++ continue;
++ /* is the region (part) in overlap with the current region ?*/
++ if (ei->addr >= end || ei->addr + ei->size <= start)
++ continue;
++ /* if the region is at the beginning of <start,end> we move
++ * start to the end of the region since it's ok until there
++ */
++ if (ei->addr <= start)
++ start = ei->addr + ei->size;
++ /* if start is now at or beyond end, we're done, full
++ * coverage */
++ if (start >= end)
++ return 1; /* we're done */
++ }
++ return 0;
++}
++
++/*
++ * Find the highest page frame number we have available
++ */
++void __init find_max_pfn(void)
++{
++ int i;
++
++ max_pfn = 0;
++ if (efi_enabled) {
++ efi_memmap_walk(efi_find_max_pfn, &max_pfn);
++ efi_memmap_walk(efi_memory_present_wrapper, NULL);
++ return;
++ }
++
++ for (i = 0; i < e820.nr_map; i++) {
++ unsigned long start, end;
++ /* RAM? */
++ if (e820.map[i].type != E820_RAM)
++ continue;
++ start = PFN_UP(e820.map[i].addr);
++ end = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
++ if (start >= end)
++ continue;
++ if (end > max_pfn)
++ max_pfn = end;
++ memory_present(0, start, end);
++ }
++}
++
++/*
++ * Determine low and high memory ranges:
++ */
++unsigned long __init find_max_low_pfn(void)
++{
++ unsigned long max_low_pfn;
++
++ max_low_pfn = max_pfn;
++ if (max_low_pfn > MAXMEM_PFN) {
++ if (highmem_pages == -1)
++ highmem_pages = max_pfn - MAXMEM_PFN;
++ if (highmem_pages + MAXMEM_PFN < max_pfn)
++ max_pfn = MAXMEM_PFN + highmem_pages;
++ if (highmem_pages + MAXMEM_PFN > max_pfn) {
++ printk("only %luMB highmem pages available, ignoring highmem size of %uMB.\n", pages_to_mb(max_pfn - MAXMEM_PFN), pages_to_mb(highmem_pages));
++ highmem_pages = 0;
++ }
++ max_low_pfn = MAXMEM_PFN;
++#ifndef CONFIG_HIGHMEM
++ /* Maximum memory usable is what is directly addressable */
++ printk(KERN_WARNING "Warning only %ldMB will be used.\n",
++ MAXMEM>>20);
++ if (max_pfn > MAX_NONPAE_PFN)
++ printk(KERN_WARNING "Use a PAE enabled kernel.\n");
++ else
++ printk(KERN_WARNING "Use a HIGHMEM enabled kernel.\n");
++ max_pfn = MAXMEM_PFN;
++#else /* !CONFIG_HIGHMEM */
++#ifndef CONFIG_X86_PAE
++ if (max_pfn > MAX_NONPAE_PFN) {
++ max_pfn = MAX_NONPAE_PFN;
++ printk(KERN_WARNING "Warning only 4GB will be used.\n");
++ printk(KERN_WARNING "Use a PAE enabled kernel.\n");
++ }
++#endif /* !CONFIG_X86_PAE */
++#endif /* !CONFIG_HIGHMEM */
++ } else {
++ if (highmem_pages == -1)
++ highmem_pages = 0;
++#ifdef CONFIG_HIGHMEM
++ if (highmem_pages >= max_pfn) {
++ printk(KERN_ERR "highmem size specified (%uMB) is bigger than pages available (%luMB)!.\n", pages_to_mb(highmem_pages), pages_to_mb(max_pfn));
++ highmem_pages = 0;
++ }
++ if (highmem_pages) {
++ if (max_low_pfn-highmem_pages < 64*1024*1024/PAGE_SIZE){
++ printk(KERN_ERR "highmem size %uMB results in smaller than 64MB lowmem, ignoring it.\n", pages_to_mb(highmem_pages));
++ highmem_pages = 0;
++ }
++ max_low_pfn -= highmem_pages;
++ }
++#else
++ if (highmem_pages)
++ printk(KERN_ERR "ignoring highmem size on non-highmem kernel!\n");
++#endif
++ }
++ return max_low_pfn;
++}
++
++/*
++ * Free all available memory for boot time allocation. Used
++ * as a callback function by efi_memory_walk()
++ */
++
++static int __init
++free_available_memory(unsigned long start, unsigned long end, void *arg)
++{
++ /* check max_low_pfn */
++ if (start >= ((max_low_pfn + 1) << PAGE_SHIFT))
++ return 0;
++ if (end >= ((max_low_pfn + 1) << PAGE_SHIFT))
++ end = (max_low_pfn + 1) << PAGE_SHIFT;
++ if (start < end)
++ free_bootmem(start, end - start);
++
++ return 0;
++}
++/*
++ * Register fully available low RAM pages with the bootmem allocator.
++ */
++static void __init register_bootmem_low_pages(unsigned long max_low_pfn)
++{
++ int i;
++
++ if (efi_enabled) {
++ efi_memmap_walk(free_available_memory, NULL);
++ return;
++ }
++ for (i = 0; i < e820.nr_map; i++) {
++ unsigned long curr_pfn, last_pfn, size;
++ /*
++ * Reserve usable low memory
++ */
++ if (e820.map[i].type != E820_RAM)
++ continue;
++ /*
++ * We are rounding up the start address of usable memory:
++ */
++ curr_pfn = PFN_UP(e820.map[i].addr);
++ if (curr_pfn >= max_low_pfn)
++ continue;
++ /*
++ * ... and at the end of the usable range downwards:
++ */
++ last_pfn = PFN_DOWN(e820.map[i].addr + e820.map[i].size);
++
++#ifdef CONFIG_XEN
++ /*
++ * Truncate to the number of actual pages currently
++ * present.
++ */
++ if (last_pfn > xen_start_info->nr_pages)
++ last_pfn = xen_start_info->nr_pages;
++#endif
++
++ if (last_pfn > max_low_pfn)
++ last_pfn = max_low_pfn;
++
++ /*
++ * .. finally, did all the rounding and playing
++ * around just make the area go away?
++ */
++ if (last_pfn <= curr_pfn)
++ continue;
++
++ size = last_pfn - curr_pfn;
++ free_bootmem(PFN_PHYS(curr_pfn), PFN_PHYS(size));
++ }
++}
++
++#ifndef CONFIG_XEN
++/*
++ * workaround for Dell systems that neglect to reserve EBDA
++ */
++static void __init reserve_ebda_region(void)
++{
++ unsigned int addr;
++ addr = get_bios_ebda();
++ if (addr)
++ reserve_bootmem(addr, PAGE_SIZE);
++}
++#endif
++
++#ifndef CONFIG_NEED_MULTIPLE_NODES
++void __init setup_bootmem_allocator(void);
++static unsigned long __init setup_memory(void)
++{
++ /*
++ * partially used pages are not usable - thus
++ * we are rounding upwards:
++ */
++ min_low_pfn = PFN_UP(__pa(xen_start_info->pt_base)) +
++ xen_start_info->nr_pt_frames;
++
++ find_max_pfn();
++
++ max_low_pfn = find_max_low_pfn();
++
++#ifdef CONFIG_HIGHMEM
++ highstart_pfn = highend_pfn = max_pfn;
++ if (max_pfn > max_low_pfn) {
++ highstart_pfn = max_low_pfn;
++ }
++ printk(KERN_NOTICE "%ldMB HIGHMEM available.\n",
++ pages_to_mb(highend_pfn - highstart_pfn));
++#endif
++ printk(KERN_NOTICE "%ldMB LOWMEM available.\n",
++ pages_to_mb(max_low_pfn));
++
++ setup_bootmem_allocator();
++
++ return max_low_pfn;
++}
++
++void __init zone_sizes_init(void)
++{
++ unsigned long zones_size[MAX_NR_ZONES] = {0, 0, 0};
++ unsigned int max_dma, low;
++
++ /*
++ * XEN: Our notion of "DMA memory" is fake when running over Xen.
++ * We simply put all RAM in the DMA zone so that those drivers which
++ * needlessly specify GFP_DMA do not get starved of RAM unnecessarily.
++ * Those drivers that *do* require lowmem are screwed anyway when
++ * running over Xen!
++ */
++ max_dma = max_low_pfn;
++ low = max_low_pfn;
++
++ if (low < max_dma)
++ zones_size[ZONE_DMA] = low;
++ else {
++ zones_size[ZONE_DMA] = max_dma;
++ zones_size[ZONE_NORMAL] = low - max_dma;
++#ifdef CONFIG_HIGHMEM
++ zones_size[ZONE_HIGHMEM] = highend_pfn - low;
++#endif
++ }
++ free_area_init(zones_size);
++}
++#else
++extern unsigned long __init setup_memory(void);
++extern void zone_sizes_init(void);
++#endif /* !CONFIG_NEED_MULTIPLE_NODES */
++
++void __init setup_bootmem_allocator(void)
++{
++ unsigned long bootmap_size;
++ /*
++ * Initialize the boot-time allocator (with low memory only):
++ */
++ bootmap_size = init_bootmem(min_low_pfn, max_low_pfn);
++
++ register_bootmem_low_pages(max_low_pfn);
++
++ /*
++ * Reserve the bootmem bitmap itself as well. We do this in two
++ * steps (first step was init_bootmem()) because this catches
++ * the (very unlikely) case of us accidentally initializing the
++ * bootmem allocator with an invalid RAM area.
++ */
++ reserve_bootmem(__PHYSICAL_START, (PFN_PHYS(min_low_pfn) +
++ bootmap_size + PAGE_SIZE-1) - (__PHYSICAL_START));
++
++#ifndef CONFIG_XEN
++ /*
++ * reserve physical page 0 - it's a special BIOS page on many boxes,
++ * enabling clean reboots, SMP operation, laptop functions.
++ */
++ reserve_bootmem(0, PAGE_SIZE);
++
++ /* reserve EBDA region, it's a 4K region */
++ reserve_ebda_region();
++
++ /* could be an AMD 768MPX chipset. Reserve a page before VGA to prevent
++ PCI prefetch into it (errata #56). Usually the page is reserved anyways,
++ unless you have no PS/2 mouse plugged in. */
++ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
++ boot_cpu_data.x86 == 6)
++ reserve_bootmem(0xa0000 - 4096, 4096);
++
++#ifdef CONFIG_SMP
++ /*
++ * But first pinch a few for the stack/trampoline stuff
++ * FIXME: Don't need the extra page at 4K, but need to fix
++ * trampoline before removing it. (see the GDT stuff)
++ */
++ reserve_bootmem(PAGE_SIZE, PAGE_SIZE);
++#endif
++#ifdef CONFIG_ACPI_SLEEP
++ /*
++ * Reserve low memory region for sleep support.
++ */
++ acpi_reserve_bootmem();
++#endif
++#endif /* !CONFIG_XEN */
++
++#ifdef CONFIG_BLK_DEV_INITRD
++ if (xen_start_info->mod_start) {
++ if (INITRD_START + INITRD_SIZE <= (max_low_pfn << PAGE_SHIFT)) {
++ /*reserve_bootmem(INITRD_START, INITRD_SIZE);*/
++ initrd_start = INITRD_START + PAGE_OFFSET;
++ initrd_end = initrd_start+INITRD_SIZE;
++ initrd_below_start_ok = 1;
++ }
++ else {
++ printk(KERN_ERR "initrd extends beyond end of memory "
++ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
++ INITRD_START + INITRD_SIZE,
++ max_low_pfn << PAGE_SHIFT);
++ initrd_start = 0;
++ }
++ }
++#endif
++#ifdef CONFIG_KEXEC
++ if (crashk_res.start != crashk_res.end)
++ reserve_bootmem(crashk_res.start,
++ crashk_res.end - crashk_res.start + 1);
++#endif
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap))
++ phys_to_machine_mapping =
++ (unsigned long *)xen_start_info->mfn_list;
++}
++
++/*
++ * The node 0 pgdat is initialized before all of these because
++ * it's needed for bootmem. node>0 pgdats have their virtual
++ * space allocated before the pagetables are in place to access
++ * them, so they can't be cleared then.
++ *
++ * This should all compile down to nothing when NUMA is off.
++ */
++void __init remapped_pgdat_init(void)
++{
++ int nid;
++
++ for_each_online_node(nid) {
++ if (nid != 0)
++ memset(NODE_DATA(nid), 0, sizeof(struct pglist_data));
++ }
++}
++
++/*
++ * Request address space for all standard RAM and ROM resources
++ * and also for regions reported as reserved by the e820.
++ */
++static void __init
++legacy_init_iomem_resources(struct e820entry *e820, int nr_map,
++ struct resource *code_resource,
++ struct resource *data_resource)
++{
++ int i;
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST) || !defined(CONFIG_XEN)
++ probe_roms();
++#endif
++
++ for (i = 0; i < nr_map; i++) {
++ struct resource *res;
++ if (e820[i].addr + e820[i].size > 0x100000000ULL)
++ continue;
++ res = alloc_bootmem_low(sizeof(struct resource));
++ switch (e820[i].type) {
++ case E820_RAM: res->name = "System RAM"; break;
++ case E820_ACPI: res->name = "ACPI Tables"; break;
++ case E820_NVS: res->name = "ACPI Non-volatile Storage"; break;
++ default: res->name = "reserved";
++ }
++ res->start = e820[i].addr;
++ res->end = res->start + e820[i].size - 1;
++ res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
++ request_resource(&iomem_resource, res);
++ if (e820[i].type == E820_RAM) {
++ /*
++ * We don't know which RAM region contains kernel data,
++ * so we try it repeatedly and let the resource manager
++ * test it.
++ */
++ request_resource(res, code_resource);
++ request_resource(res, data_resource);
++#ifdef CONFIG_KEXEC
++ request_resource(res, &crashk_res);
++#endif
++ }
++ }
++}
++
++/*
++ * Locate a unused range of the physical address space below 4G which
++ * can be used for PCI mappings.
++ */
++static void __init
++e820_setup_gap(struct e820entry *e820, int nr_map)
++{
++ unsigned long gapstart, gapsize, round;
++ unsigned long long last;
++ int i;
++
++ /*
++ * Search for the bigest gap in the low 32 bits of the e820
++ * memory space.
++ */
++ last = 0x100000000ull;
++ gapstart = 0x10000000;
++ gapsize = 0x400000;
++ i = nr_map;
++ while (--i >= 0) {
++ unsigned long long start = e820[i].addr;
++ unsigned long long end = start + e820[i].size;
++
++ /*
++ * Since "last" is at most 4GB, we know we'll
++ * fit in 32 bits if this condition is true
++ */
++ if (last > end) {
++ unsigned long gap = last - end;
++
++ if (gap > gapsize) {
++ gapsize = gap;
++ gapstart = end;
++ }
++ }
++ if (start < last)
++ last = start;
++ }
++
++ /*
++ * See how much we want to round up: start off with
++ * rounding to the next 1MB area.
++ */
++ round = 0x100000;
++ while ((gapsize >> 4) > round)
++ round += round;
++ /* Fun with two's complement */
++ pci_mem_start = (gapstart + round) & -round;
++
++ printk("Allocating PCI resources starting at %08lx (gap: %08lx:%08lx)\n",
++ pci_mem_start, gapstart, gapsize);
++}
++
++/*
++ * Request address space for all standard resources
++ */
++static void __init register_memory(void)
++{
++#ifdef CONFIG_XEN
++ struct e820entry *machine_e820;
++ struct xen_memory_map memmap;
++#endif
++ int i;
++
++ /* Nothing to do if not running in dom0. */
++ if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
++ legacy_init_iomem_resources(e820.map, e820.nr_map,
++ &code_resource, &data_resource);
++ return;
++ }
++
++#ifdef CONFIG_XEN
++ machine_e820 = alloc_bootmem_low_pages(PAGE_SIZE);
++
++ memmap.nr_entries = E820MAX;
++ set_xen_guest_handle(memmap.buffer, machine_e820);
++
++ BUG_ON(HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap));
++
++ legacy_init_iomem_resources(machine_e820, memmap.nr_entries,
++ &code_resource, &data_resource);
++#else
++ if (efi_enabled)
++ efi_initialize_iomem_resources(&code_resource, &data_resource);
++ else
++ legacy_init_iomem_resources(e820.map, e820.nr_map,
++ &code_resource, &data_resource);
++#endif
++
++ /* EFI systems may still have VGA */
++ request_resource(&iomem_resource, &video_ram_resource);
++
++ /* request I/O space for devices used on all i[345]86 PCs */
++ for (i = 0; i < STANDARD_IO_RESOURCES; i++)
++ request_resource(&ioport_resource, &standard_io_resources[i]);
++
++#ifdef CONFIG_XEN
++ e820_setup_gap(machine_e820, memmap.nr_entries);
++ free_bootmem(__pa(machine_e820), PAGE_SIZE);
++#else
++ e820_setup_gap(e820.map, e820.nr_map);
++#endif
++}
++
++static char * __init machine_specific_memory_setup(void);
++
++#ifdef CONFIG_MCA
++static void set_mca_bus(int x)
++{
++ MCA_bus = x;
++}
++#else
++static void set_mca_bus(int x) { }
++#endif
++
++/*
++ * Determine if we were loaded by an EFI loader. If so, then we have also been
++ * passed the efi memmap, systab, etc., so we should use these data structures
++ * for initialization. Note, the efi init code path is determined by the
++ * global efi_enabled. This allows the same kernel image to be used on existing
++ * systems (with a traditional BIOS) as well as on EFI systems.
++ */
++void __init setup_arch(char **cmdline_p)
++{
++ int i, j, k, fpp;
++ struct physdev_set_iopl set_iopl;
++ unsigned long max_low_pfn;
++
++ /* Force a quick death if the kernel panics (not domain 0). */
++ extern int panic_timeout;
++ if (!panic_timeout && !(xen_start_info->flags & SIF_INITDOMAIN))
++ panic_timeout = 1;
++
++ /* Register a call for panic conditions. */
++ atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
++
++ HYPERVISOR_vm_assist(VMASST_CMD_enable, VMASST_TYPE_4gb_segments);
++ HYPERVISOR_vm_assist(VMASST_CMD_enable,
++ VMASST_TYPE_writable_pagetables);
++
++ memcpy(&boot_cpu_data, &new_cpu_data, sizeof(new_cpu_data));
++ pre_setup_arch_hook();
++ early_cpu_init();
++
++ /*
++ * FIXME: This isn't an official loader_type right
++ * now but does currently work with elilo.
++ * If we were configured as an EFI kernel, check to make
++ * sure that we were loaded correctly from elilo and that
++ * the system table is valid. If not, then initialize normally.
++ */
++#ifdef CONFIG_EFI
++ if ((LOADER_TYPE == 0x50) && EFI_SYSTAB)
++ efi_enabled = 1;
++#endif
++
++ /* This must be initialized to UNNAMED_MAJOR for ipconfig to work
++ properly. Setting ROOT_DEV to default to /dev/ram0 breaks initrd.
++ */
++ ROOT_DEV = MKDEV(UNNAMED_MAJOR,0);
++ drive_info = DRIVE_INFO;
++ screen_info = SCREEN_INFO;
++ edid_info = EDID_INFO;
++ apm_info.bios = APM_BIOS_INFO;
++ ist_info = IST_INFO;
++ saved_videomode = VIDEO_MODE;
++ if( SYS_DESC_TABLE.length != 0 ) {
++ set_mca_bus(SYS_DESC_TABLE.table[3] & 0x2);
++ machine_id = SYS_DESC_TABLE.table[0];
++ machine_submodel_id = SYS_DESC_TABLE.table[1];
++ BIOS_revision = SYS_DESC_TABLE.table[2];
++ }
++ bootloader_type = LOADER_TYPE;
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ /* This is drawn from a dump from vgacon:startup in
++ * standard Linux. */
++ screen_info.orig_video_mode = 3;
++ screen_info.orig_video_isVGA = 1;
++ screen_info.orig_video_lines = 25;
++ screen_info.orig_video_cols = 80;
++ screen_info.orig_video_ega_bx = 3;
++ screen_info.orig_video_points = 16;
++ } else
++ screen_info.orig_video_isVGA = 0;
++
++#ifdef CONFIG_BLK_DEV_RAM
++ rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
++ rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
++ rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
++#endif
++
++ setup_xen_features();
++
++ ARCH_SETUP
++ if (efi_enabled)
++ efi_init();
++ else {
++ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
++ print_memory_map(machine_specific_memory_setup());
++ }
++
++ copy_edd();
++
++ if (!MOUNT_ROOT_RDONLY)
++ root_mountflags &= ~MS_RDONLY;
++ init_mm.start_code = (unsigned long) _text;
++ init_mm.end_code = (unsigned long) _etext;
++ init_mm.end_data = (unsigned long) _edata;
++ init_mm.brk = (PFN_UP(__pa(xen_start_info->pt_base)) +
++ xen_start_info->nr_pt_frames) << PAGE_SHIFT;
++
++ code_resource.start = virt_to_phys(_text);
++ code_resource.end = virt_to_phys(_etext)-1;
++ data_resource.start = virt_to_phys(_etext);
++ data_resource.end = virt_to_phys(_edata)-1;
++
++ parse_cmdline_early(cmdline_p);
++
++#ifdef CONFIG_EARLY_PRINTK
++ {
++ char *s = strstr(*cmdline_p, "earlyprintk=");
++ if (s) {
++ setup_early_printk(strchr(s, '=') + 1);
++ printk("early console enabled\n");
++ }
++ }
++#endif
++
++ max_low_pfn = setup_memory();
++
++ /*
++ * NOTE: before this point _nobody_ is allowed to allocate
++ * any memory using the bootmem allocator. Although the
++ * alloctor is now initialised only the first 8Mb of the kernel
++ * virtual address space has been mapped. All allocations before
++ * paging_init() has completed must use the alloc_bootmem_low_pages()
++ * variant (which allocates DMA'able memory) and care must be taken
++ * not to exceed the 8Mb limit.
++ */
++
++#ifdef CONFIG_SMP
++ smp_alloc_memory(); /* AP processor realmode stacks in low memory*/
++#endif
++ paging_init();
++ remapped_pgdat_init();
++ sparse_init();
++ zone_sizes_init();
++
++#ifdef CONFIG_X86_FIND_SMP_CONFIG
++ /*
++ * Find and reserve possible boot-time SMP configuration:
++ */
++ find_smp_config();
++#endif
++
++ /* Make sure we have a correctly sized P->M table. */
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ phys_to_machine_mapping = alloc_bootmem_low_pages(
++ max_pfn * sizeof(unsigned long));
++ memset(phys_to_machine_mapping, ~0,
++ max_pfn * sizeof(unsigned long));
++ memcpy(phys_to_machine_mapping,
++ (unsigned long *)xen_start_info->mfn_list,
++ xen_start_info->nr_pages * sizeof(unsigned long));
++ free_bootmem(
++ __pa(xen_start_info->mfn_list),
++ PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
++ sizeof(unsigned long))));
++
++ /*
++ * Initialise the list of the frames that specify the list of
++ * frames that make up the p2m table. Used by save/restore
++ */
++ pfn_to_mfn_frame_list_list = alloc_bootmem_low_pages(PAGE_SIZE);
++ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
++ virt_to_mfn(pfn_to_mfn_frame_list_list);
++
++ fpp = PAGE_SIZE/sizeof(unsigned long);
++ for (i=0, j=0, k=-1; i< max_pfn; i+=fpp, j++) {
++ if ((j % fpp) == 0) {
++ k++;
++ BUG_ON(k>=16);
++ pfn_to_mfn_frame_list[k] =
++ alloc_bootmem_low_pages(PAGE_SIZE);
++ pfn_to_mfn_frame_list_list[k] =
++ virt_to_mfn(pfn_to_mfn_frame_list[k]);
++ j=0;
++ }
++ pfn_to_mfn_frame_list[k][j] =
++ virt_to_mfn(&phys_to_machine_mapping[i]);
++ }
++ HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
++ }
++
++ /*
++ * NOTE: at this point the bootmem allocator is fully available.
++ */
++
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ dmi_scan_machine();
++
++#ifdef CONFIG_X86_GENERICARCH
++ generic_apic_probe(*cmdline_p);
++#endif
++ if (efi_enabled)
++ efi_map_memmap();
++
++ set_iopl.iopl = 1;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
++
++#ifdef CONFIG_X86_IO_APIC
++ check_acpi_pci(); /* Checks more than just ACPI actually */
++#endif
++
++#ifdef CONFIG_ACPI
++ if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
++ printk(KERN_INFO "ACPI in unprivileged domain disabled\n");
++ acpi_disabled = 1;
++ acpi_ht = 0;
++ }
++
++ /*
++ * Parse the ACPI tables for possible boot-time SMP configuration.
++ */
++ acpi_boot_table_init();
++ acpi_boot_init();
++
++#if defined(CONFIG_SMP) && defined(CONFIG_X86_PC)
++ if (def_to_bigsmp)
++ printk(KERN_WARNING "More than 8 CPUs detected and "
++ "CONFIG_X86_PC cannot handle it.\nUse "
++ "CONFIG_X86_GENERICARCH or CONFIG_X86_BIGSMP.\n");
++#endif
++#endif
++#ifdef CONFIG_X86_LOCAL_APIC
++ if (smp_found_config)
++ get_smp_config();
++#endif
++#if defined(CONFIG_XEN) && defined(CONFIG_SMP)
++ prefill_possible_map();
++#endif
++
++ register_memory();
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ if (!(xen_start_info->flags & SIF_PRIVILEGED))
++ panic("Xen granted us console access "
++ "but not privileged status");
++
++#ifdef CONFIG_VT
++#if defined(CONFIG_VGA_CONSOLE)
++ if (!efi_enabled ||
++ (efi_mem_type(0xa0000) != EFI_CONVENTIONAL_MEMORY))
++ conswitchp = &vga_con;
++#elif defined(CONFIG_DUMMY_CONSOLE)
++ conswitchp = &dummy_con;
++#endif
++#endif
++ } else {
++ extern int console_use_vt;
++ console_use_vt = 0;
++ }
++}
++
++static int
++xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
++{
++ HYPERVISOR_shutdown(SHUTDOWN_crash);
++ /* we're never actually going to get here... */
++ return NOTIFY_DONE;
++}
++
++static __init int add_pcspkr(void)
++{
++ struct platform_device *pd;
++ int ret;
++
++ pd = platform_device_alloc("pcspkr", -1);
++ if (!pd)
++ return -ENOMEM;
++
++ ret = platform_device_add(pd);
++ if (ret)
++ platform_device_put(pd);
++
++ return ret;
++}
++device_initcall(add_pcspkr);
++
++#include "setup_arch_post.h"
++/*
++ * Local Variables:
++ * mode:c
++ * c-file-style:"k&r"
++ * c-basic-offset:8
++ * End:
++ */
+diff -urNp linux-2.6/arch/i386/kernel/smp-xen.c new/arch/i386/kernel/smp-xen.c
+--- linux-2.6/arch/i386/kernel/smp-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/smp-xen.c 2006-07-07 15:10:04.000000000 +0200
+@@ -0,0 +1,614 @@
++/*
++ * Intel SMP support routines.
++ *
++ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
++ * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com>
++ *
++ * This code is released under the GNU General Public License version 2 or
++ * later.
++ */
++
++#include <linux/init.h>
++
++#include <linux/mm.h>
++#include <linux/delay.h>
++#include <linux/spinlock.h>
++#include <linux/smp_lock.h>
++#include <linux/kernel_stat.h>
++#include <linux/mc146818rtc.h>
++#include <linux/cache.h>
++#include <linux/interrupt.h>
++#include <linux/cpu.h>
++#include <linux/module.h>
++
++#include <asm/mtrr.h>
++#include <asm/tlbflush.h>
++#if 0
++#include <mach_apic.h>
++#endif
++#include <xen/evtchn.h>
++
++/*
++ * Some notes on x86 processor bugs affecting SMP operation:
++ *
++ * Pentium, Pentium Pro, II, III (and all CPUs) have bugs.
++ * The Linux implications for SMP are handled as follows:
++ *
++ * Pentium III / [Xeon]
++ * None of the E1AP-E3AP errata are visible to the user.
++ *
++ * E1AP. see PII A1AP
++ * E2AP. see PII A2AP
++ * E3AP. see PII A3AP
++ *
++ * Pentium II / [Xeon]
++ * None of the A1AP-A3AP errata are visible to the user.
++ *
++ * A1AP. see PPro 1AP
++ * A2AP. see PPro 2AP
++ * A3AP. see PPro 7AP
++ *
++ * Pentium Pro
++ * None of 1AP-9AP errata are visible to the normal user,
++ * except occasional delivery of 'spurious interrupt' as trap #15.
++ * This is very rare and a non-problem.
++ *
++ * 1AP. Linux maps APIC as non-cacheable
++ * 2AP. worked around in hardware
++ * 3AP. fixed in C0 and above steppings microcode update.
++ * Linux does not use excessive STARTUP_IPIs.
++ * 4AP. worked around in hardware
++ * 5AP. symmetric IO mode (normal Linux operation) not affected.
++ * 'noapic' mode has vector 0xf filled out properly.
++ * 6AP. 'noapic' mode might be affected - fixed in later steppings
++ * 7AP. We do not assume writes to the LVT deassering IRQs
++ * 8AP. We do not enable low power mode (deep sleep) during MP bootup
++ * 9AP. We do not use mixed mode
++ *
++ * Pentium
++ * There is a marginal case where REP MOVS on 100MHz SMP
++ * machines with B stepping processors can fail. XXX should provide
++ * an L1cache=Writethrough or L1cache=off option.
++ *
++ * B stepping CPUs may hang. There are hardware work arounds
++ * for this. We warn about it in case your board doesn't have the work
++ * arounds. Basically thats so I can tell anyone with a B stepping
++ * CPU and SMP problems "tough".
++ *
++ * Specific items [From Pentium Processor Specification Update]
++ *
++ * 1AP. Linux doesn't use remote read
++ * 2AP. Linux doesn't trust APIC errors
++ * 3AP. We work around this
++ * 4AP. Linux never generated 3 interrupts of the same priority
++ * to cause a lost local interrupt.
++ * 5AP. Remote read is never used
++ * 6AP. not affected - worked around in hardware
++ * 7AP. not affected - worked around in hardware
++ * 8AP. worked around in hardware - we get explicit CS errors if not
++ * 9AP. only 'noapic' mode affected. Might generate spurious
++ * interrupts, we log only the first one and count the
++ * rest silently.
++ * 10AP. not affected - worked around in hardware
++ * 11AP. Linux reads the APIC between writes to avoid this, as per
++ * the documentation. Make sure you preserve this as it affects
++ * the C stepping chips too.
++ * 12AP. not affected - worked around in hardware
++ * 13AP. not affected - worked around in hardware
++ * 14AP. we always deassert INIT during bootup
++ * 15AP. not affected - worked around in hardware
++ * 16AP. not affected - worked around in hardware
++ * 17AP. not affected - worked around in hardware
++ * 18AP. not affected - worked around in hardware
++ * 19AP. not affected - worked around in BIOS
++ *
++ * If this sounds worrying believe me these bugs are either ___RARE___,
++ * or are signal timing bugs worked around in hardware and there's
++ * about nothing of note with C stepping upwards.
++ */
++
++DEFINE_PER_CPU(struct tlb_state, cpu_tlbstate) ____cacheline_aligned = { &init_mm, 0, };
++
++/*
++ * the following functions deal with sending IPIs between CPUs.
++ *
++ * We use 'broadcast', CPU->CPU IPIs and self-IPIs too.
++ */
++
++static inline int __prepare_ICR (unsigned int shortcut, int vector)
++{
++ return APIC_DM_FIXED | shortcut | vector | APIC_DEST_LOGICAL;
++}
++
++static inline int __prepare_ICR2 (unsigned int mask)
++{
++ return SET_APIC_DEST_FIELD(mask);
++}
++
++DECLARE_PER_CPU(int, ipi_to_irq[NR_IPIS]);
++
++static inline void __send_IPI_one(unsigned int cpu, int vector)
++{
++ int irq = per_cpu(ipi_to_irq, cpu)[vector];
++ BUG_ON(irq < 0);
++ notify_remote_via_irq(irq);
++}
++
++void __send_IPI_shortcut(unsigned int shortcut, int vector)
++{
++ int cpu;
++
++ switch (shortcut) {
++ case APIC_DEST_SELF:
++ __send_IPI_one(smp_processor_id(), vector);
++ break;
++ case APIC_DEST_ALLBUT:
++ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
++ if (cpu == smp_processor_id())
++ continue;
++ if (cpu_isset(cpu, cpu_online_map)) {
++ __send_IPI_one(cpu, vector);
++ }
++ }
++ break;
++ default:
++ printk("XXXXXX __send_IPI_shortcut %08x vector %d\n", shortcut,
++ vector);
++ break;
++ }
++}
++
++void fastcall send_IPI_self(int vector)
++{
++ __send_IPI_shortcut(APIC_DEST_SELF, vector);
++}
++
++/*
++ * This is only used on smaller machines.
++ */
++void send_IPI_mask_bitmask(cpumask_t mask, int vector)
++{
++ unsigned long flags;
++ unsigned int cpu;
++
++ local_irq_save(flags);
++ WARN_ON(cpus_addr(mask)[0] & ~cpus_addr(cpu_online_map)[0]);
++
++ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
++ if (cpu_isset(cpu, mask)) {
++ __send_IPI_one(cpu, vector);
++ }
++ }
++
++ local_irq_restore(flags);
++}
++
++void send_IPI_mask_sequence(cpumask_t mask, int vector)
++{
++
++ send_IPI_mask_bitmask(mask, vector);
++}
++
++#include <mach_ipi.h> /* must come after the send_IPI functions above for inlining */
++
++#if 0 /* XEN */
++/*
++ * Smarter SMP flushing macros.
++ * c/o Linus Torvalds.
++ *
++ * These mean you can really definitely utterly forget about
++ * writing to user space from interrupts. (Its not allowed anyway).
++ *
++ * Optimizations Manfred Spraul <manfred@colorfullife.com>
++ */
++
++static cpumask_t flush_cpumask;
++static struct mm_struct * flush_mm;
++static unsigned long flush_va;
++static DEFINE_SPINLOCK(tlbstate_lock);
++#define FLUSH_ALL 0xffffffff
++
++/*
++ * We cannot call mmdrop() because we are in interrupt context,
++ * instead update mm->cpu_vm_mask.
++ *
++ * We need to reload %cr3 since the page tables may be going
++ * away from under us..
++ */
++static inline void leave_mm (unsigned long cpu)
++{
++ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
++ BUG();
++ cpu_clear(cpu, per_cpu(cpu_tlbstate, cpu).active_mm->cpu_vm_mask);
++ load_cr3(swapper_pg_dir);
++}
++
++/*
++ *
++ * The flush IPI assumes that a thread switch happens in this order:
++ * [cpu0: the cpu that switches]
++ * 1) switch_mm() either 1a) or 1b)
++ * 1a) thread switch to a different mm
++ * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask);
++ * Stop ipi delivery for the old mm. This is not synchronized with
++ * the other cpus, but smp_invalidate_interrupt ignore flush ipis
++ * for the wrong mm, and in the worst case we perform a superflous
++ * tlb flush.
++ * 1a2) set cpu_tlbstate to TLBSTATE_OK
++ * Now the smp_invalidate_interrupt won't call leave_mm if cpu0
++ * was in lazy tlb mode.
++ * 1a3) update cpu_tlbstate[].active_mm
++ * Now cpu0 accepts tlb flushes for the new mm.
++ * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask);
++ * Now the other cpus will send tlb flush ipis.
++ * 1a4) change cr3.
++ * 1b) thread switch without mm change
++ * cpu_tlbstate[].active_mm is correct, cpu0 already handles
++ * flush ipis.
++ * 1b1) set cpu_tlbstate to TLBSTATE_OK
++ * 1b2) test_and_set the cpu bit in cpu_vm_mask.
++ * Atomically set the bit [other cpus will start sending flush ipis],
++ * and test the bit.
++ * 1b3) if the bit was 0: leave_mm was called, flush the tlb.
++ * 2) switch %%esp, ie current
++ *
++ * The interrupt must handle 2 special cases:
++ * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm.
++ * - the cpu performs speculative tlb reads, i.e. even if the cpu only
++ * runs in kernel space, the cpu could load tlb entries for user space
++ * pages.
++ *
++ * The good news is that cpu_tlbstate is local to each cpu, no
++ * write/read ordering problems.
++ */
++
++/*
++ * TLB flush IPI:
++ *
++ * 1) Flush the tlb entries if the cpu uses the mm that's being flushed.
++ * 2) Leave the mm if we are in the lazy tlb mode.
++ */
++
++irqreturn_t smp_invalidate_interrupt(int irq, void *dev_id,
++ struct pt_regs *regs)
++{
++ unsigned long cpu;
++
++ cpu = get_cpu();
++
++ if (!cpu_isset(cpu, flush_cpumask))
++ goto out;
++ /*
++ * This was a BUG() but until someone can quote me the
++ * line from the intel manual that guarantees an IPI to
++ * multiple CPUs is retried _only_ on the erroring CPUs
++ * its staying as a return
++ *
++ * BUG();
++ */
++
++ if (flush_mm == per_cpu(cpu_tlbstate, cpu).active_mm) {
++ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK) {
++ if (flush_va == FLUSH_ALL)
++ local_flush_tlb();
++ else
++ __flush_tlb_one(flush_va);
++ } else
++ leave_mm(cpu);
++ }
++ smp_mb__before_clear_bit();
++ cpu_clear(cpu, flush_cpumask);
++ smp_mb__after_clear_bit();
++out:
++ put_cpu_no_resched();
++
++ return IRQ_HANDLED;
++}
++
++static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
++ unsigned long va)
++{
++ /*
++ * A couple of (to be removed) sanity checks:
++ *
++ * - current CPU must not be in mask
++ * - mask must exist :)
++ */
++ BUG_ON(cpus_empty(cpumask));
++ BUG_ON(cpu_isset(smp_processor_id(), cpumask));
++ BUG_ON(!mm);
++
++ /* If a CPU which we ran on has gone down, OK. */
++ cpus_and(cpumask, cpumask, cpu_online_map);
++ if (cpus_empty(cpumask))
++ return;
++
++ /*
++ * i'm not happy about this global shared spinlock in the
++ * MM hot path, but we'll see how contended it is.
++ * Temporarily this turns IRQs off, so that lockups are
++ * detected by the NMI watchdog.
++ */
++ spin_lock(&tlbstate_lock);
++
++ flush_mm = mm;
++ flush_va = va;
++#if NR_CPUS <= BITS_PER_LONG
++ atomic_set_mask(cpumask, &flush_cpumask);
++#else
++ {
++ int k;
++ unsigned long *flush_mask = (unsigned long *)&flush_cpumask;
++ unsigned long *cpu_mask = (unsigned long *)&cpumask;
++ for (k = 0; k < BITS_TO_LONGS(NR_CPUS); ++k)
++ atomic_set_mask(cpu_mask[k], &flush_mask[k]);
++ }
++#endif
++ /*
++ * We have to send the IPI only to
++ * CPUs affected.
++ */
++ send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR);
++
++ while (!cpus_empty(flush_cpumask))
++ /* nothing. lockup detection does not belong here */
++ mb();
++
++ flush_mm = NULL;
++ flush_va = 0;
++ spin_unlock(&tlbstate_lock);
++}
++
++void flush_tlb_current_task(void)
++{
++ struct mm_struct *mm = current->mm;
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ local_flush_tlb();
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
++ preempt_enable();
++}
++
++void flush_tlb_mm (struct mm_struct * mm)
++{
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ if (current->active_mm == mm) {
++ if (current->mm)
++ local_flush_tlb();
++ else
++ leave_mm(smp_processor_id());
++ }
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
++
++ preempt_enable();
++}
++
++void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
++{
++ struct mm_struct *mm = vma->vm_mm;
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ if (current->active_mm == mm) {
++ if(current->mm)
++ __flush_tlb_one(va);
++ else
++ leave_mm(smp_processor_id());
++ }
++
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, va);
++
++ preempt_enable();
++}
++EXPORT_SYMBOL(flush_tlb_page);
++
++static void do_flush_tlb_all(void* info)
++{
++ unsigned long cpu = smp_processor_id();
++
++ __flush_tlb_all();
++ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_LAZY)
++ leave_mm(cpu);
++}
++
++void flush_tlb_all(void)
++{
++ on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
++}
++
++#else
++
++irqreturn_t smp_invalidate_interrupt(int irq, void *dev_id,
++ struct pt_regs *regs)
++{ return 0; }
++void flush_tlb_current_task(void)
++{ xen_tlb_flush_mask(¤t->mm->cpu_vm_mask); }
++void flush_tlb_mm(struct mm_struct * mm)
++{ xen_tlb_flush_mask(&mm->cpu_vm_mask); }
++void flush_tlb_page(struct vm_area_struct *vma, unsigned long va)
++{ xen_invlpg_mask(&vma->vm_mm->cpu_vm_mask, va); }
++EXPORT_SYMBOL(flush_tlb_page);
++void flush_tlb_all(void)
++{ xen_tlb_flush_all(); }
++
++#endif /* XEN */
++
++/*
++ * this function sends a 'reschedule' IPI to another CPU.
++ * it goes straight through and wastes no time serializing
++ * anything. Worst case is that we lose a reschedule ...
++ */
++void smp_send_reschedule(int cpu)
++{
++ WARN_ON(cpu_is_offline(cpu));
++ send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
++}
++
++/*
++ * Structure and data for smp_call_function(). This is designed to minimise
++ * static memory requirements. It also looks cleaner.
++ */
++static DEFINE_SPINLOCK(call_lock);
++
++struct call_data_struct {
++ void (*func) (void *info);
++ void *info;
++ atomic_t started;
++ atomic_t finished;
++ int wait;
++};
++
++void lock_ipi_call_lock(void)
++{
++ spin_lock_irq(&call_lock);
++}
++
++void unlock_ipi_call_lock(void)
++{
++ spin_unlock_irq(&call_lock);
++}
++
++static struct call_data_struct *call_data;
++
++/**
++ * smp_call_function(): Run a function on all other CPUs.
++ * @func: The function to run. This must be fast and non-blocking.
++ * @info: An arbitrary pointer to pass to the function.
++ * @nonatomic: currently unused.
++ * @wait: If true, wait (atomically) until function has completed on other CPUs.
++ *
++ * Returns 0 on success, else a negative status code. Does not return until
++ * remote CPUs are nearly ready to execute <<func>> or are or have executed.
++ *
++ * You must not call this function with disabled interrupts or from a
++ * hardware interrupt handler or from a bottom half handler.
++ */
++int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
++ int wait)
++{
++ struct call_data_struct data;
++ int cpus;
++
++ /* Holding any lock stops cpus from going down. */
++ spin_lock(&call_lock);
++ cpus = num_online_cpus() - 1;
++ if (!cpus) {
++ spin_unlock(&call_lock);
++ return 0;
++ }
++
++ /* Can deadlock when called with interrupts disabled */
++ WARN_ON(irqs_disabled());
++
++ data.func = func;
++ data.info = info;
++ atomic_set(&data.started, 0);
++ data.wait = wait;
++ if (wait)
++ atomic_set(&data.finished, 0);
++
++ call_data = &data;
++ mb();
++
++ /* Send a message to all other CPUs and wait for them to respond */
++ send_IPI_allbutself(CALL_FUNCTION_VECTOR);
++
++ /* Wait for response */
++ while (atomic_read(&data.started) != cpus)
++ barrier();
++
++ if (wait)
++ while (atomic_read(&data.finished) != cpus)
++ barrier();
++ spin_unlock(&call_lock);
++
++ return 0;
++}
++EXPORT_SYMBOL(smp_call_function);
++
++static void stop_this_cpu (void * dummy)
++{
++ /*
++ * Remove this CPU:
++ */
++ cpu_clear(smp_processor_id(), cpu_online_map);
++ local_irq_disable();
++#if 0
++ disable_local_APIC();
++#endif
++ if (cpu_data[smp_processor_id()].hlt_works_ok)
++ for(;;) halt();
++ for (;;);
++}
++
++/*
++ * this function calls the 'stop' function on all other CPUs in the system.
++ */
++
++void smp_send_stop(void)
++{
++ smp_call_function(stop_this_cpu, NULL, 1, 0);
++
++ local_irq_disable();
++#if 0
++ disable_local_APIC();
++#endif
++ local_irq_enable();
++}
++
++/*
++ * Reschedule call back. Nothing to do,
++ * all the work is done automatically when
++ * we return from the interrupt.
++ */
++irqreturn_t smp_reschedule_interrupt(int irq, void *dev_id,
++ struct pt_regs *regs)
++{
++
++ return IRQ_HANDLED;
++}
++
++#include <linux/kallsyms.h>
++irqreturn_t smp_call_function_interrupt(int irq, void *dev_id,
++ struct pt_regs *regs)
++{
++ void (*func) (void *info) = call_data->func;
++ void *info = call_data->info;
++ int wait = call_data->wait;
++
++ /*
++ * Notify initiating CPU that I've grabbed the data and am
++ * about to execute the function
++ */
++ mb();
++ atomic_inc(&call_data->started);
++ /*
++ * At this point the info structure may be out of scope unless wait==1
++ */
++ irq_enter();
++ (*func)(info);
++ irq_exit();
++
++ if (wait) {
++ mb();
++ atomic_inc(&call_data->finished);
++ }
++
++ return IRQ_HANDLED;
++}
++
+diff -urNp linux-2.6/arch/i386/kernel/swiotlb.c new/arch/i386/kernel/swiotlb.c
+--- linux-2.6/arch/i386/kernel/swiotlb.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/swiotlb.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,672 @@
++/*
++ * Dynamic DMA mapping support.
++ *
++ * This implementation is a fallback for platforms that do not support
++ * I/O TLBs (aka DMA address translation hardware).
++ * Copyright (C) 2000 Asit Mallick <Asit.K.Mallick@intel.com>
++ * Copyright (C) 2000 Goutham Rao <goutham.rao@intel.com>
++ * Copyright (C) 2000, 2003 Hewlett-Packard Co
++ * David Mosberger-Tang <davidm@hpl.hp.com>
++ * Copyright (C) 2005 Keir Fraser <keir@xensource.com>
++ */
++
++#include <linux/cache.h>
++#include <linux/mm.h>
++#include <linux/module.h>
++#include <linux/pci.h>
++#include <linux/spinlock.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/ctype.h>
++#include <linux/init.h>
++#include <linux/bootmem.h>
++#include <linux/highmem.h>
++#include <asm/io.h>
++#include <asm/pci.h>
++#include <asm/dma.h>
++#include <asm/uaccess.h>
++#include <xen/interface/memory.h>
++
++int swiotlb;
++EXPORT_SYMBOL(swiotlb);
++
++#define OFFSET(val,align) ((unsigned long)((val) & ( (align) - 1)))
++
++#define SG_ENT_PHYS_ADDRESS(sg) (page_to_bus((sg)->page) + (sg)->offset)
++
++/*
++ * Maximum allowable number of contiguous slabs to map,
++ * must be a power of 2. What is the appropriate value ?
++ * The complexity of {map,unmap}_single is linearly dependent on this value.
++ */
++#define IO_TLB_SEGSIZE 128
++
++/*
++ * log of the size of each IO TLB slab. The number of slabs is command line
++ * controllable.
++ */
++#define IO_TLB_SHIFT 11
++
++/* Width of DMA addresses in the IO TLB. 31 bits is an aacraid limitation. */
++#define IO_TLB_DMA_BITS 31
++
++static int swiotlb_force;
++static char *iotlb_virt_start;
++static unsigned long iotlb_nslabs;
++
++/*
++ * Used to do a quick range check in swiotlb_unmap_single and
++ * swiotlb_sync_single_*, to see if the memory was in fact allocated by this
++ * API.
++ */
++static unsigned long iotlb_pfn_start, iotlb_pfn_end;
++
++/* Does the given dma address reside within the swiotlb aperture? */
++static inline int in_swiotlb_aperture(dma_addr_t dev_addr)
++{
++ unsigned long pfn = mfn_to_local_pfn(dev_addr >> PAGE_SHIFT);
++ return (pfn_valid(pfn)
++ && (pfn >= iotlb_pfn_start)
++ && (pfn < iotlb_pfn_end));
++}
++
++/*
++ * When the IOMMU overflows we return a fallback buffer. This sets the size.
++ */
++static unsigned long io_tlb_overflow = 32*1024;
++
++void *io_tlb_overflow_buffer;
++
++/*
++ * This is a free list describing the number of free entries available from
++ * each index
++ */
++static unsigned int *io_tlb_list;
++static unsigned int io_tlb_index;
++
++/*
++ * We need to save away the original address corresponding to a mapped entry
++ * for the sync operations.
++ */
++static struct phys_addr {
++ struct page *page;
++ unsigned int offset;
++} *io_tlb_orig_addr;
++
++/*
++ * Protect the above data structures in the map and unmap calls
++ */
++static DEFINE_SPINLOCK(io_tlb_lock);
++
++static int __init
++setup_io_tlb_npages(char *str)
++{
++ /* Unlike ia64, the size is aperture in megabytes, not 'slabs'! */
++ if (isdigit(*str)) {
++ iotlb_nslabs = simple_strtoul(str, &str, 0) <<
++ (20 - IO_TLB_SHIFT);
++ iotlb_nslabs = ALIGN(iotlb_nslabs, IO_TLB_SEGSIZE);
++ /* Round up to power of two (xen_create_contiguous_region). */
++ while (iotlb_nslabs & (iotlb_nslabs-1))
++ iotlb_nslabs += iotlb_nslabs & ~(iotlb_nslabs-1);
++ }
++ if (*str == ',')
++ ++str;
++ /*
++ * NB. 'force' enables the swiotlb, but doesn't force its use for
++ * every DMA like it does on native Linux. 'off' forcibly disables
++ * use of the swiotlb.
++ */
++ if (!strcmp(str, "force"))
++ swiotlb_force = 1;
++ else if (!strcmp(str, "off"))
++ swiotlb_force = -1;
++ return 1;
++}
++__setup("swiotlb=", setup_io_tlb_npages);
++/* make io_tlb_overflow tunable too? */
++
++/*
++ * Statically reserve bounce buffer space and initialize bounce buffer data
++ * structures for the software IO TLB used to implement the PCI DMA API.
++ */
++void
++swiotlb_init_with_default_size (size_t default_size)
++{
++ unsigned long i, bytes;
++
++ if (!iotlb_nslabs) {
++ iotlb_nslabs = (default_size >> IO_TLB_SHIFT);
++ iotlb_nslabs = ALIGN(iotlb_nslabs, IO_TLB_SEGSIZE);
++ /* Round up to power of two (xen_create_contiguous_region). */
++ while (iotlb_nslabs & (iotlb_nslabs-1))
++ iotlb_nslabs += iotlb_nslabs & ~(iotlb_nslabs-1);
++ }
++
++ bytes = iotlb_nslabs * (1UL << IO_TLB_SHIFT);
++
++ /*
++ * Get IO TLB memory from the low pages
++ */
++ iotlb_virt_start = alloc_bootmem_low_pages(bytes);
++ if (!iotlb_virt_start)
++ panic("Cannot allocate SWIOTLB buffer!\n"
++ "Use dom0_mem Xen boot parameter to reserve\n"
++ "some DMA memory (e.g., dom0_mem=-128M).\n");
++
++ for (i = 0; i < iotlb_nslabs; i += IO_TLB_SEGSIZE) {
++ int rc = xen_create_contiguous_region(
++ (unsigned long)iotlb_virt_start + (i << IO_TLB_SHIFT),
++ get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT),
++ IO_TLB_DMA_BITS);
++ BUG_ON(rc);
++ }
++
++ /*
++ * Allocate and initialize the free list array. This array is used
++ * to find contiguous free memory regions of size up to IO_TLB_SEGSIZE.
++ */
++ io_tlb_list = alloc_bootmem(iotlb_nslabs * sizeof(int));
++ for (i = 0; i < iotlb_nslabs; i++)
++ io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE);
++ io_tlb_index = 0;
++ io_tlb_orig_addr = alloc_bootmem(
++ iotlb_nslabs * sizeof(*io_tlb_orig_addr));
++
++ /*
++ * Get the overflow emergency buffer
++ */
++ io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow);
++
++ iotlb_pfn_start = __pa(iotlb_virt_start) >> PAGE_SHIFT;
++ iotlb_pfn_end = iotlb_pfn_start + (bytes >> PAGE_SHIFT);
++
++ printk(KERN_INFO "Software IO TLB enabled: \n"
++ " Aperture: %lu megabytes\n"
++ " Kernel range: 0x%016lx - 0x%016lx\n",
++ bytes >> 20,
++ (unsigned long)iotlb_virt_start,
++ (unsigned long)iotlb_virt_start + bytes);
++}
++
++void
++swiotlb_init(void)
++{
++ long ram_end;
++ size_t defsz = 64 * (1 << 20); /* 64MB default size */
++
++ if (swiotlb_force == 1) {
++ swiotlb = 1;
++ } else if ((swiotlb_force != -1) &&
++ is_running_on_xen() &&
++ (xen_start_info->flags & SIF_INITDOMAIN)) {
++ /* Domain 0 always has a swiotlb. */
++ ram_end = HYPERVISOR_memory_op(XENMEM_maximum_ram_page, NULL);
++ if (ram_end <= 0x7ffff)
++ defsz = 2 * (1 << 20); /* 2MB on <2GB on systems. */
++ swiotlb = 1;
++ }
++
++ if (swiotlb)
++ swiotlb_init_with_default_size(defsz);
++ else
++ printk(KERN_INFO "Software IO TLB disabled\n");
++}
++
++/*
++ * We use __copy_to_user_inatomic to transfer to the host buffer because the
++ * buffer may be mapped read-only (e.g, in blkback driver) but lower-level
++ * drivers map the buffer for DMA_BIDIRECTIONAL access. This causes an
++ * unnecessary copy from the aperture to the host buffer, and a page fault.
++ */
++static void
++__sync_single(struct phys_addr buffer, char *dma_addr, size_t size, int dir)
++{
++ if (PageHighMem(buffer.page)) {
++ size_t len, bytes;
++ char *dev, *host, *kmp;
++ len = size;
++ while (len != 0) {
++ if (((bytes = len) + buffer.offset) > PAGE_SIZE)
++ bytes = PAGE_SIZE - buffer.offset;
++ kmp = kmap_atomic(buffer.page, KM_SWIOTLB);
++ dev = dma_addr + size - len;
++ host = kmp + buffer.offset;
++ if (dir == DMA_FROM_DEVICE) {
++ if (__copy_to_user_inatomic(host, dev, bytes))
++ /* inaccessible */;
++ } else
++ memcpy(dev, host, bytes);
++ kunmap_atomic(kmp, KM_SWIOTLB);
++ len -= bytes;
++ buffer.page++;
++ buffer.offset = 0;
++ }
++ } else {
++ char *host = (char *)phys_to_virt(
++ page_to_pseudophys(buffer.page)) + buffer.offset;
++ if (dir == DMA_FROM_DEVICE) {
++ if (__copy_to_user_inatomic(host, dma_addr, size))
++ /* inaccessible */;
++ } else if (dir == DMA_TO_DEVICE)
++ memcpy(dma_addr, host, size);
++ }
++}
++
++/*
++ * Allocates bounce buffer and returns its kernel virtual address.
++ */
++static void *
++map_single(struct device *hwdev, struct phys_addr buffer, size_t size, int dir)
++{
++ unsigned long flags;
++ char *dma_addr;
++ unsigned int nslots, stride, index, wrap;
++ int i;
++
++ /*
++ * For mappings greater than a page, we limit the stride (and
++ * hence alignment) to a page size.
++ */
++ nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
++ if (size > PAGE_SIZE)
++ stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT));
++ else
++ stride = 1;
++
++ BUG_ON(!nslots);
++
++ /*
++ * Find suitable number of IO TLB entries size that will fit this
++ * request and allocate a buffer from that IO TLB pool.
++ */
++ spin_lock_irqsave(&io_tlb_lock, flags);
++ {
++ wrap = index = ALIGN(io_tlb_index, stride);
++
++ if (index >= iotlb_nslabs)
++ wrap = index = 0;
++
++ do {
++ /*
++ * If we find a slot that indicates we have 'nslots'
++ * number of contiguous buffers, we allocate the
++ * buffers from that slot and mark the entries as '0'
++ * indicating unavailable.
++ */
++ if (io_tlb_list[index] >= nslots) {
++ int count = 0;
++
++ for (i = index; i < (int)(index + nslots); i++)
++ io_tlb_list[i] = 0;
++ for (i = index - 1;
++ (OFFSET(i, IO_TLB_SEGSIZE) !=
++ IO_TLB_SEGSIZE -1) && io_tlb_list[i];
++ i--)
++ io_tlb_list[i] = ++count;
++ dma_addr = iotlb_virt_start +
++ (index << IO_TLB_SHIFT);
++
++ /*
++ * Update the indices to avoid searching in
++ * the next round.
++ */
++ io_tlb_index =
++ ((index + nslots) < iotlb_nslabs
++ ? (index + nslots) : 0);
++
++ goto found;
++ }
++ index += stride;
++ if (index >= iotlb_nslabs)
++ index = 0;
++ } while (index != wrap);
++
++ spin_unlock_irqrestore(&io_tlb_lock, flags);
++ return NULL;
++ }
++ found:
++ spin_unlock_irqrestore(&io_tlb_lock, flags);
++
++ /*
++ * Save away the mapping from the original address to the DMA address.
++ * This is needed when we sync the memory. Then we sync the buffer if
++ * needed.
++ */
++ io_tlb_orig_addr[index] = buffer;
++ if ((dir == DMA_TO_DEVICE) || (dir == DMA_BIDIRECTIONAL))
++ __sync_single(buffer, dma_addr, size, DMA_TO_DEVICE);
++
++ return dma_addr;
++}
++
++/*
++ * dma_addr is the kernel virtual address of the bounce buffer to unmap.
++ */
++static void
++unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
++{
++ unsigned long flags;
++ int i, count, nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
++ int index = (dma_addr - iotlb_virt_start) >> IO_TLB_SHIFT;
++ struct phys_addr buffer = io_tlb_orig_addr[index];
++
++ /*
++ * First, sync the memory before unmapping the entry
++ */
++ if ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL))
++ __sync_single(buffer, dma_addr, size, DMA_FROM_DEVICE);
++
++ /*
++ * Return the buffer to the free list by setting the corresponding
++ * entries to indicate the number of contigous entries available.
++ * While returning the entries to the free list, we merge the entries
++ * with slots below and above the pool being returned.
++ */
++ spin_lock_irqsave(&io_tlb_lock, flags);
++ {
++ count = ((index + nslots) < ALIGN(index + 1, IO_TLB_SEGSIZE) ?
++ io_tlb_list[index + nslots] : 0);
++ /*
++ * Step 1: return the slots to the free list, merging the
++ * slots with superceeding slots
++ */
++ for (i = index + nslots - 1; i >= index; i--)
++ io_tlb_list[i] = ++count;
++ /*
++ * Step 2: merge the returned slots with the preceding slots,
++ * if available (non zero)
++ */
++ for (i = index - 1;
++ (OFFSET(i, IO_TLB_SEGSIZE) !=
++ IO_TLB_SEGSIZE -1) && io_tlb_list[i];
++ i--)
++ io_tlb_list[i] = ++count;
++ }
++ spin_unlock_irqrestore(&io_tlb_lock, flags);
++}
++
++static void
++sync_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
++{
++ int index = (dma_addr - iotlb_virt_start) >> IO_TLB_SHIFT;
++ struct phys_addr buffer = io_tlb_orig_addr[index];
++ BUG_ON((dir != DMA_FROM_DEVICE) && (dir != DMA_TO_DEVICE));
++ __sync_single(buffer, dma_addr, size, dir);
++}
++
++static void
++swiotlb_full(struct device *dev, size_t size, int dir, int do_panic)
++{
++ /*
++ * Ran out of IOMMU space for this operation. This is very bad.
++ * Unfortunately the drivers cannot handle this operation properly.
++ * unless they check for pci_dma_mapping_error (most don't)
++ * When the mapping is small enough return a static buffer to limit
++ * the damage, or panic when the transfer is too big.
++ */
++ printk(KERN_ERR "PCI-DMA: Out of SW-IOMMU space for %lu bytes at "
++ "device %s\n", (unsigned long)size, dev ? dev->bus_id : "?");
++
++ if (size > io_tlb_overflow && do_panic) {
++ if (dir == PCI_DMA_FROMDEVICE || dir == PCI_DMA_BIDIRECTIONAL)
++ panic("PCI-DMA: Memory would be corrupted\n");
++ if (dir == PCI_DMA_TODEVICE || dir == PCI_DMA_BIDIRECTIONAL)
++ panic("PCI-DMA: Random memory would be DMAed\n");
++ }
++}
++
++/*
++ * Map a single buffer of the indicated size for DMA in streaming mode. The
++ * PCI address to use is returned.
++ *
++ * Once the device is given the dma address, the device owns this memory until
++ * either swiotlb_unmap_single or swiotlb_dma_sync_single is performed.
++ */
++dma_addr_t
++swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir)
++{
++ dma_addr_t dev_addr = virt_to_bus(ptr);
++ void *map;
++ struct phys_addr buffer;
++
++ BUG_ON(dir == DMA_NONE);
++
++ /*
++ * If the pointer passed in happens to be in the device's DMA window,
++ * we can safely return the device addr and not worry about bounce
++ * buffering it.
++ */
++ if (!range_straddles_page_boundary(ptr, size) &&
++ !address_needs_mapping(hwdev, dev_addr))
++ return dev_addr;
++
++ /*
++ * Oh well, have to allocate and map a bounce buffer.
++ */
++ buffer.page = virt_to_page(ptr);
++ buffer.offset = (unsigned long)ptr & ~PAGE_MASK;
++ map = map_single(hwdev, buffer, size, dir);
++ if (!map) {
++ swiotlb_full(hwdev, size, dir, 1);
++ map = io_tlb_overflow_buffer;
++ }
++
++ dev_addr = virt_to_bus(map);
++ return dev_addr;
++}
++
++/*
++ * Unmap a single streaming mode DMA translation. The dma_addr and size must
++ * match what was provided for in a previous swiotlb_map_single call. All
++ * other usages are undefined.
++ *
++ * After this call, reads by the cpu to the buffer are guaranteed to see
++ * whatever the device wrote there.
++ */
++void
++swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size,
++ int dir)
++{
++ BUG_ON(dir == DMA_NONE);
++ if (in_swiotlb_aperture(dev_addr))
++ unmap_single(hwdev, bus_to_virt(dev_addr), size, dir);
++}
++
++/*
++ * Make physical memory consistent for a single streaming mode DMA translation
++ * after a transfer.
++ *
++ * If you perform a swiotlb_map_single() but wish to interrogate the buffer
++ * using the cpu, yet do not wish to teardown the PCI dma mapping, you must
++ * call this function before doing so. At the next point you give the PCI dma
++ * address back to the card, you must first perform a
++ * swiotlb_dma_sync_for_device, and then the device again owns the buffer
++ */
++void
++swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
++ size_t size, int dir)
++{
++ BUG_ON(dir == DMA_NONE);
++ if (in_swiotlb_aperture(dev_addr))
++ sync_single(hwdev, bus_to_virt(dev_addr), size, dir);
++}
++
++void
++swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
++ size_t size, int dir)
++{
++ BUG_ON(dir == DMA_NONE);
++ if (in_swiotlb_aperture(dev_addr))
++ sync_single(hwdev, bus_to_virt(dev_addr), size, dir);
++}
++
++/*
++ * Map a set of buffers described by scatterlist in streaming mode for DMA.
++ * This is the scatter-gather version of the above swiotlb_map_single
++ * interface. Here the scatter gather list elements are each tagged with the
++ * appropriate dma address and length. They are obtained via
++ * sg_dma_{address,length}(SG).
++ *
++ * NOTE: An implementation may be able to use a smaller number of
++ * DMA address/length pairs than there are SG table elements.
++ * (for example via virtual mapping capabilities)
++ * The routine returns the number of addr/length pairs actually
++ * used, at most nents.
++ *
++ * Device ownership issues as mentioned above for swiotlb_map_single are the
++ * same here.
++ */
++int
++swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
++ int dir)
++{
++ struct phys_addr buffer;
++ dma_addr_t dev_addr;
++ char *map;
++ int i;
++
++ BUG_ON(dir == DMA_NONE);
++
++ for (i = 0; i < nelems; i++, sg++) {
++ dev_addr = SG_ENT_PHYS_ADDRESS(sg);
++ if (address_needs_mapping(hwdev, dev_addr)) {
++ buffer.page = sg->page;
++ buffer.offset = sg->offset;
++ map = map_single(hwdev, buffer, sg->length, dir);
++ if (!map) {
++ /* Don't panic here, we expect map_sg users
++ to do proper error handling. */
++ swiotlb_full(hwdev, sg->length, dir, 0);
++ swiotlb_unmap_sg(hwdev, sg - i, i, dir);
++ sg[0].dma_length = 0;
++ return 0;
++ }
++ sg->dma_address = (dma_addr_t)virt_to_bus(map);
++ } else
++ sg->dma_address = dev_addr;
++ sg->dma_length = sg->length;
++ }
++ return nelems;
++}
++
++/*
++ * Unmap a set of streaming mode DMA translations. Again, cpu read rules
++ * concerning calls here are the same as for swiotlb_unmap_single() above.
++ */
++void
++swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
++ int dir)
++{
++ int i;
++
++ BUG_ON(dir == DMA_NONE);
++
++ for (i = 0; i < nelems; i++, sg++)
++ if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
++ unmap_single(hwdev,
++ (void *)bus_to_virt(sg->dma_address),
++ sg->dma_length, dir);
++}
++
++/*
++ * Make physical memory consistent for a set of streaming mode DMA translations
++ * after a transfer.
++ *
++ * The same as swiotlb_sync_single_* but for a scatter-gather list, same rules
++ * and usage.
++ */
++void
++swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
++ int nelems, int dir)
++{
++ int i;
++
++ BUG_ON(dir == DMA_NONE);
++
++ for (i = 0; i < nelems; i++, sg++)
++ if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
++ sync_single(hwdev,
++ (void *)bus_to_virt(sg->dma_address),
++ sg->dma_length, dir);
++}
++
++void
++swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
++ int nelems, int dir)
++{
++ int i;
++
++ BUG_ON(dir == DMA_NONE);
++
++ for (i = 0; i < nelems; i++, sg++)
++ if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
++ sync_single(hwdev,
++ (void *)bus_to_virt(sg->dma_address),
++ sg->dma_length, dir);
++}
++
++dma_addr_t
++swiotlb_map_page(struct device *hwdev, struct page *page,
++ unsigned long offset, size_t size,
++ enum dma_data_direction direction)
++{
++ struct phys_addr buffer;
++ dma_addr_t dev_addr;
++ char *map;
++
++ dev_addr = page_to_bus(page) + offset;
++ if (address_needs_mapping(hwdev, dev_addr)) {
++ buffer.page = page;
++ buffer.offset = offset;
++ map = map_single(hwdev, buffer, size, direction);
++ if (!map) {
++ swiotlb_full(hwdev, size, direction, 1);
++ map = io_tlb_overflow_buffer;
++ }
++ dev_addr = (dma_addr_t)virt_to_bus(map);
++ }
++
++ return dev_addr;
++}
++
++void
++swiotlb_unmap_page(struct device *hwdev, dma_addr_t dma_address,
++ size_t size, enum dma_data_direction direction)
++{
++ BUG_ON(direction == DMA_NONE);
++ if (in_swiotlb_aperture(dma_address))
++ unmap_single(hwdev, bus_to_virt(dma_address), size, direction);
++}
++
++int
++swiotlb_dma_mapping_error(dma_addr_t dma_addr)
++{
++ return (dma_addr == virt_to_bus(io_tlb_overflow_buffer));
++}
++
++/*
++ * Return whether the given PCI device DMA address mask can be supported
++ * properly. For example, if your device can only drive the low 24-bits
++ * during PCI bus mastering, then you would pass 0x00ffffff as the mask to
++ * this function.
++ */
++int
++swiotlb_dma_supported (struct device *hwdev, u64 mask)
++{
++ return (mask >= ((1UL << IO_TLB_DMA_BITS) - 1));
++}
++
++EXPORT_SYMBOL(swiotlb_init);
++EXPORT_SYMBOL(swiotlb_map_single);
++EXPORT_SYMBOL(swiotlb_unmap_single);
++EXPORT_SYMBOL(swiotlb_map_sg);
++EXPORT_SYMBOL(swiotlb_unmap_sg);
++EXPORT_SYMBOL(swiotlb_sync_single_for_cpu);
++EXPORT_SYMBOL(swiotlb_sync_single_for_device);
++EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu);
++EXPORT_SYMBOL(swiotlb_sync_sg_for_device);
++EXPORT_SYMBOL(swiotlb_map_page);
++EXPORT_SYMBOL(swiotlb_unmap_page);
++EXPORT_SYMBOL(swiotlb_dma_mapping_error);
++EXPORT_SYMBOL(swiotlb_dma_supported);
+diff -urNp linux-2.6/arch/i386/kernel/sysenter.c new/arch/i386/kernel/sysenter.c
+--- linux-2.6/arch/i386/kernel/sysenter.c 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/sysenter.c 2006-05-09 12:32:35.000000000 +0200
+@@ -13,16 +13,22 @@
+ #include <linux/gfp.h>
+ #include <linux/string.h>
+ #include <linux/elf.h>
++#include <linux/mm.h>
+
+ #include <asm/cpufeature.h>
+ #include <asm/msr.h>
+ #include <asm/pgtable.h>
+ #include <asm/unistd.h>
+
++#ifdef CONFIG_XEN
++#include <xen/interface/callback.h>
++#endif
++
+ extern asmlinkage void sysenter_entry(void);
+
+ void enable_sep_cpu(void)
+ {
++#ifndef CONFIG_X86_NO_TSS
+ int cpu = get_cpu();
+ struct tss_struct *tss = &per_cpu(init_tss, cpu);
+
+@@ -37,6 +43,7 @@ void enable_sep_cpu(void)
+ wrmsr(MSR_IA32_SYSENTER_ESP, tss->esp1, 0);
+ wrmsr(MSR_IA32_SYSENTER_EIP, (unsigned long) sysenter_entry, 0);
+ put_cpu();
++#endif
+ }
+
+ /*
+@@ -45,23 +52,100 @@ void enable_sep_cpu(void)
+ */
+ extern const char vsyscall_int80_start, vsyscall_int80_end;
+ extern const char vsyscall_sysenter_start, vsyscall_sysenter_end;
++static void *syscall_page;
+
+ int __init sysenter_setup(void)
+ {
+- void *page = (void *)get_zeroed_page(GFP_ATOMIC);
++ syscall_page = (void *)get_zeroed_page(GFP_ATOMIC);
+
+- __set_fixmap(FIX_VSYSCALL, __pa(page), PAGE_READONLY_EXEC);
++#ifdef CONFIG_XEN
++ if (boot_cpu_has(X86_FEATURE_SEP)) {
++ struct callback_register sysenter = {
++ .type = CALLBACKTYPE_sysenter,
++ .address = { __KERNEL_CS, (unsigned long)sysenter_entry },
++ };
+
+- if (!boot_cpu_has(X86_FEATURE_SEP)) {
+- memcpy(page,
+- &vsyscall_int80_start,
+- &vsyscall_int80_end - &vsyscall_int80_start);
++ if (HYPERVISOR_callback_op(CALLBACKOP_register, &sysenter) < 0)
++ clear_bit(X86_FEATURE_SEP, boot_cpu_data.x86_capability);
++ }
++#endif
++
++ if (boot_cpu_has(X86_FEATURE_SEP)) {
++ memcpy(syscall_page,
++ &vsyscall_sysenter_start,
++ &vsyscall_sysenter_end - &vsyscall_sysenter_start);
+ return 0;
+ }
+
+- memcpy(page,
+- &vsyscall_sysenter_start,
+- &vsyscall_sysenter_end - &vsyscall_sysenter_start);
++ memcpy(syscall_page,
++ &vsyscall_int80_start,
++ &vsyscall_int80_end - &vsyscall_int80_start);
++
++ return 0;
++}
++
++static struct page*
++syscall_nopage(struct vm_area_struct *vma, unsigned long adr, int *type)
++{
++ struct page *p = virt_to_page(adr - vma->vm_start + syscall_page);
++ get_page(p);
++ return p;
++}
++
++/* Prevent VMA merging */
++static void syscall_vma_close(struct vm_area_struct *vma)
++{
++}
++
++static struct vm_operations_struct syscall_vm_ops = {
++ .close = syscall_vma_close,
++ .nopage = syscall_nopage,
++};
+
++/* Setup a VMA at program startup for the vsyscall page */
++int arch_setup_additional_pages(struct linux_binprm *bprm, int exstack)
++{
++ struct vm_area_struct *vma;
++ struct mm_struct *mm = current->mm;
++ int ret;
++
++ vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
++ if (!vma)
++ return -ENOMEM;
++
++ memset(vma, 0, sizeof(struct vm_area_struct));
++ /* Could randomize here */
++ vma->vm_start = VSYSCALL_BASE;
++ vma->vm_end = VSYSCALL_BASE + PAGE_SIZE;
++ /* MAYWRITE to allow gdb to COW and set breakpoints */
++ vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE;
++ vma->vm_flags |= mm->def_flags;
++ vma->vm_page_prot = protection_map[vma->vm_flags & 7];
++ vma->vm_ops = &syscall_vm_ops;
++ vma->vm_mm = mm;
++
++ down_write(&mm->mmap_sem);
++ if ((ret = insert_vm_struct(mm, vma))) {
++ up_write(&mm->mmap_sem);
++ kmem_cache_free(vm_area_cachep, vma);
++ return ret;
++ }
++ mm->total_vm++;
++ up_write(&mm->mmap_sem);
++ return 0;
++}
++
++struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
++{
++ return NULL;
++}
++
++int in_gate_area(struct task_struct *task, unsigned long addr)
++{
++ return 0;
++}
++
++int in_gate_area_no_task(unsigned long addr)
++{
+ return 0;
+ }
+diff -urNp linux-2.6/arch/i386/kernel/time-xen.c new/arch/i386/kernel/time-xen.c
+--- linux-2.6/arch/i386/kernel/time-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/time-xen.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,1109 @@
++/*
++ * linux/arch/i386/kernel/time.c
++ *
++ * Copyright (C) 1991, 1992, 1995 Linus Torvalds
++ *
++ * This file contains the PC-specific time handling details:
++ * reading the RTC at bootup, etc..
++ * 1994-07-02 Alan Modra
++ * fixed set_rtc_mmss, fixed time.year for >= 2000, new mktime
++ * 1995-03-26 Markus Kuhn
++ * fixed 500 ms bug at call to set_rtc_mmss, fixed DS12887
++ * precision CMOS clock update
++ * 1996-05-03 Ingo Molnar
++ * fixed time warps in do_[slow|fast]_gettimeoffset()
++ * 1997-09-10 Updated NTP code according to technical memorandum Jan '96
++ * "A Kernel Model for Precision Timekeeping" by Dave Mills
++ * 1998-09-05 (Various)
++ * More robust do_fast_gettimeoffset() algorithm implemented
++ * (works with APM, Cyrix 6x86MX and Centaur C6),
++ * monotonic gettimeofday() with fast_get_timeoffset(),
++ * drift-proof precision TSC calibration on boot
++ * (C. Scott Ananian <cananian@alumni.princeton.edu>, Andrew D.
++ * Balsa <andrebalsa@altern.org>, Philip Gladstone <philip@raptor.com>;
++ * ported from 2.0.35 Jumbo-9 by Michael Krause <m.krause@tu-harburg.de>).
++ * 1998-12-16 Andrea Arcangeli
++ * Fixed Jumbo-9 code in 2.1.131: do_gettimeofday was missing 1 jiffy
++ * because was not accounting lost_ticks.
++ * 1998-12-24 Copyright (C) 1998 Andrea Arcangeli
++ * Fixed a xtime SMP race (we need the xtime_lock rw spinlock to
++ * serialize accesses to xtime/lost_ticks).
++ */
++
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/param.h>
++#include <linux/string.h>
++#include <linux/mm.h>
++#include <linux/interrupt.h>
++#include <linux/time.h>
++#include <linux/delay.h>
++#include <linux/init.h>
++#include <linux/smp.h>
++#include <linux/module.h>
++#include <linux/sysdev.h>
++#include <linux/bcd.h>
++#include <linux/efi.h>
++#include <linux/mca.h>
++#include <linux/sysctl.h>
++#include <linux/percpu.h>
++#include <linux/kernel_stat.h>
++#include <linux/posix-timers.h>
++
++#include <asm/io.h>
++#include <asm/smp.h>
++#include <asm/irq.h>
++#include <asm/msr.h>
++#include <asm/delay.h>
++#include <asm/mpspec.h>
++#include <asm/uaccess.h>
++#include <asm/processor.h>
++#include <asm/timer.h>
++#include <asm/sections.h>
++
++#include "mach_time.h"
++
++#include <linux/timex.h>
++#include <linux/config.h>
++
++#include <asm/hpet.h>
++
++#include <asm/arch_hooks.h>
++
++#include <xen/evtchn.h>
++#include <xen/interface/vcpu.h>
++
++#if defined (__i386__)
++#include <asm/i8259.h>
++#endif
++
++int pit_latch_buggy; /* extern */
++
++#if defined(__x86_64__)
++unsigned long vxtime_hz = PIT_TICK_RATE;
++struct vxtime_data __vxtime __section_vxtime; /* for vsyscalls */
++volatile unsigned long __jiffies __section_jiffies = INITIAL_JIFFIES;
++unsigned long __wall_jiffies __section_wall_jiffies = INITIAL_JIFFIES;
++struct timespec __xtime __section_xtime;
++struct timezone __sys_tz __section_sys_tz;
++#endif
++
++unsigned int cpu_khz; /* Detected as we calibrate the TSC */
++EXPORT_SYMBOL(cpu_khz);
++
++extern unsigned long wall_jiffies;
++
++DEFINE_SPINLOCK(rtc_lock);
++EXPORT_SYMBOL(rtc_lock);
++
++#if defined (__i386__)
++#include <asm/i8253.h>
++#endif
++
++DEFINE_SPINLOCK(i8253_lock);
++EXPORT_SYMBOL(i8253_lock);
++
++extern struct init_timer_opts timer_tsc_init;
++extern struct timer_opts timer_tsc;
++#define timer_none timer_tsc
++struct timer_opts *cur_timer __read_mostly = &timer_tsc;
++
++/* These are peridically updated in shared_info, and then copied here. */
++struct shadow_time_info {
++ u64 tsc_timestamp; /* TSC at last update of time vals. */
++ u64 system_timestamp; /* Time, in nanosecs, since boot. */
++ u32 tsc_to_nsec_mul;
++ u32 tsc_to_usec_mul;
++ int tsc_shift;
++ u32 version;
++};
++static DEFINE_PER_CPU(struct shadow_time_info, shadow_time);
++static struct timespec shadow_tv;
++static u32 shadow_tv_version;
++
++/* Keep track of last time we did processing/updating of jiffies and xtime. */
++static u64 processed_system_time; /* System time (ns) at last processing. */
++static DEFINE_PER_CPU(u64, processed_system_time);
++
++/* How much CPU time was spent blocked and how much was 'stolen'? */
++static DEFINE_PER_CPU(u64, processed_stolen_time);
++static DEFINE_PER_CPU(u64, processed_blocked_time);
++
++/* Current runstate of each CPU (updated automatically by the hypervisor). */
++static DEFINE_PER_CPU(struct vcpu_runstate_info, runstate);
++
++/* Must be signed, as it's compared with s64 quantities which can be -ve. */
++#define NS_PER_TICK (1000000000LL/HZ)
++
++static inline void __normalize_time(time_t *sec, s64 *nsec)
++{
++ while (*nsec >= NSEC_PER_SEC) {
++ (*nsec) -= NSEC_PER_SEC;
++ (*sec)++;
++ }
++ while (*nsec < 0) {
++ (*nsec) += NSEC_PER_SEC;
++ (*sec)--;
++ }
++}
++
++/* Does this guest OS track Xen time, or set its wall clock independently? */
++static int independent_wallclock = 0;
++static int __init __independent_wallclock(char *str)
++{
++ independent_wallclock = 1;
++ return 1;
++}
++__setup("independent_wallclock", __independent_wallclock);
++
++/* Permitted clock jitter, in nsecs, beyond which a warning will be printed. */
++static unsigned long permitted_clock_jitter = 10000000UL; /* 10ms */
++static int __init __permitted_clock_jitter(char *str)
++{
++ permitted_clock_jitter = simple_strtoul(str, NULL, 0);
++ return 1;
++}
++__setup("permitted_clock_jitter=", __permitted_clock_jitter);
++
++int tsc_disable __devinitdata = 0;
++
++static void delay_tsc(unsigned long loops)
++{
++ unsigned long bclock, now;
++
++ rdtscl(bclock);
++ do {
++ rep_nop();
++ rdtscl(now);
++ } while ((now - bclock) < loops);
++}
++
++struct timer_opts timer_tsc = {
++ .name = "tsc",
++ .delay = delay_tsc,
++};
++
++/*
++ * Scale a 64-bit delta by scaling and multiplying by a 32-bit fraction,
++ * yielding a 64-bit result.
++ */
++static inline u64 scale_delta(u64 delta, u32 mul_frac, int shift)
++{
++ u64 product;
++#ifdef __i386__
++ u32 tmp1, tmp2;
++#endif
++
++ if (shift < 0)
++ delta >>= -shift;
++ else
++ delta <<= shift;
++
++#ifdef __i386__
++ __asm__ (
++ "mul %5 ; "
++ "mov %4,%%eax ; "
++ "mov %%edx,%4 ; "
++ "mul %5 ; "
++ "xor %5,%5 ; "
++ "add %4,%%eax ; "
++ "adc %5,%%edx ; "
++ : "=A" (product), "=r" (tmp1), "=r" (tmp2)
++ : "a" ((u32)delta), "1" ((u32)(delta >> 32)), "2" (mul_frac) );
++#else
++ __asm__ (
++ "mul %%rdx ; shrd $32,%%rdx,%%rax"
++ : "=a" (product) : "0" (delta), "d" ((u64)mul_frac) );
++#endif
++
++ return product;
++}
++
++#if defined (__i386__)
++int read_current_timer(unsigned long *timer_val)
++{
++ rdtscl(*timer_val);
++ return 0;
++}
++#endif
++
++void init_cpu_khz(void)
++{
++ u64 __cpu_khz = 1000000ULL << 32;
++ struct vcpu_time_info *info;
++ info = &HYPERVISOR_shared_info->vcpu_info[0].time;
++ do_div(__cpu_khz, info->tsc_to_system_mul);
++ if (info->tsc_shift < 0)
++ cpu_khz = __cpu_khz << -info->tsc_shift;
++ else
++ cpu_khz = __cpu_khz >> info->tsc_shift;
++}
++
++static u64 get_nsec_offset(struct shadow_time_info *shadow)
++{
++ u64 now, delta;
++ rdtscll(now);
++ delta = now - shadow->tsc_timestamp;
++ return scale_delta(delta, shadow->tsc_to_nsec_mul, shadow->tsc_shift);
++}
++
++static unsigned long get_usec_offset(struct shadow_time_info *shadow)
++{
++ u64 now, delta;
++ rdtscll(now);
++ delta = now - shadow->tsc_timestamp;
++ return scale_delta(delta, shadow->tsc_to_usec_mul, shadow->tsc_shift);
++}
++
++static void __update_wallclock(time_t sec, long nsec)
++{
++ long wtm_nsec, xtime_nsec;
++ time_t wtm_sec, xtime_sec;
++ u64 tmp, wc_nsec;
++
++ /* Adjust wall-clock time base based on wall_jiffies ticks. */
++ wc_nsec = processed_system_time;
++ wc_nsec += sec * (u64)NSEC_PER_SEC;
++ wc_nsec += nsec;
++ wc_nsec -= (jiffies - wall_jiffies) * (u64)NS_PER_TICK;
++
++ /* Split wallclock base into seconds and nanoseconds. */
++ tmp = wc_nsec;
++ xtime_nsec = do_div(tmp, 1000000000);
++ xtime_sec = (time_t)tmp;
++
++ wtm_sec = wall_to_monotonic.tv_sec + (xtime.tv_sec - xtime_sec);
++ wtm_nsec = wall_to_monotonic.tv_nsec + (xtime.tv_nsec - xtime_nsec);
++
++ set_normalized_timespec(&xtime, xtime_sec, xtime_nsec);
++ set_normalized_timespec(&wall_to_monotonic, wtm_sec, wtm_nsec);
++
++ ntp_clear();
++}
++
++static void update_wallclock(void)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++
++ do {
++ shadow_tv_version = s->wc_version;
++ rmb();
++ shadow_tv.tv_sec = s->wc_sec;
++ shadow_tv.tv_nsec = s->wc_nsec;
++ rmb();
++ } while ((s->wc_version & 1) | (shadow_tv_version ^ s->wc_version));
++
++ if (!independent_wallclock)
++ __update_wallclock(shadow_tv.tv_sec, shadow_tv.tv_nsec);
++}
++
++/*
++ * Reads a consistent set of time-base values from Xen, into a shadow data
++ * area.
++ */
++static void get_time_values_from_xen(void)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ struct vcpu_time_info *src;
++ struct shadow_time_info *dst;
++
++ src = &s->vcpu_info[smp_processor_id()].time;
++ dst = &per_cpu(shadow_time, smp_processor_id());
++
++ do {
++ dst->version = src->version;
++ rmb();
++ dst->tsc_timestamp = src->tsc_timestamp;
++ dst->system_timestamp = src->system_time;
++ dst->tsc_to_nsec_mul = src->tsc_to_system_mul;
++ dst->tsc_shift = src->tsc_shift;
++ rmb();
++ } while ((src->version & 1) | (dst->version ^ src->version));
++
++ dst->tsc_to_usec_mul = dst->tsc_to_nsec_mul / 1000;
++}
++
++static inline int time_values_up_to_date(int cpu)
++{
++ struct vcpu_time_info *src;
++ struct shadow_time_info *dst;
++
++ src = &HYPERVISOR_shared_info->vcpu_info[cpu].time;
++ dst = &per_cpu(shadow_time, cpu);
++
++ rmb();
++ return (dst->version == src->version);
++}
++
++/*
++ * This is a special lock that is owned by the CPU and holds the index
++ * register we are working with. It is required for NMI access to the
++ * CMOS/RTC registers. See include/asm-i386/mc146818rtc.h for details.
++ */
++volatile unsigned long cmos_lock = 0;
++EXPORT_SYMBOL(cmos_lock);
++
++/* Routines for accessing the CMOS RAM/RTC. */
++unsigned char rtc_cmos_read(unsigned char addr)
++{
++ unsigned char val;
++ lock_cmos_prefix(addr);
++ outb_p(addr, RTC_PORT(0));
++ val = inb_p(RTC_PORT(1));
++ lock_cmos_suffix(addr);
++ return val;
++}
++EXPORT_SYMBOL(rtc_cmos_read);
++
++void rtc_cmos_write(unsigned char val, unsigned char addr)
++{
++ lock_cmos_prefix(addr);
++ outb_p(addr, RTC_PORT(0));
++ outb_p(val, RTC_PORT(1));
++ lock_cmos_suffix(addr);
++}
++EXPORT_SYMBOL(rtc_cmos_write);
++
++/*
++ * This version of gettimeofday has microsecond resolution
++ * and better than microsecond precision on fast x86 machines with TSC.
++ */
++void do_gettimeofday(struct timeval *tv)
++{
++ unsigned long seq;
++ unsigned long usec, sec;
++ unsigned long max_ntp_tick;
++ s64 nsec;
++ unsigned int cpu;
++ struct shadow_time_info *shadow;
++ u32 local_time_version;
++
++ cpu = get_cpu();
++ shadow = &per_cpu(shadow_time, cpu);
++
++ do {
++ unsigned long lost;
++
++ local_time_version = shadow->version;
++ seq = read_seqbegin(&xtime_lock);
++
++ usec = get_usec_offset(shadow);
++ lost = jiffies - wall_jiffies;
++
++ /*
++ * If time_adjust is negative then NTP is slowing the clock
++ * so make sure not to go into next possible interval.
++ * Better to lose some accuracy than have time go backwards..
++ */
++ if (unlikely(time_adjust < 0)) {
++ max_ntp_tick = (USEC_PER_SEC / HZ) - tickadj;
++ usec = min(usec, max_ntp_tick);
++
++ if (lost)
++ usec += lost * max_ntp_tick;
++ }
++ else if (unlikely(lost))
++ usec += lost * (USEC_PER_SEC / HZ);
++
++ sec = xtime.tv_sec;
++ usec += (xtime.tv_nsec / NSEC_PER_USEC);
++
++ nsec = shadow->system_timestamp - processed_system_time;
++ __normalize_time(&sec, &nsec);
++ usec += (long)nsec / NSEC_PER_USEC;
++
++ if (unlikely(!time_values_up_to_date(cpu))) {
++ /*
++ * We may have blocked for a long time,
++ * rendering our calculations invalid
++ * (e.g. the time delta may have
++ * overflowed). Detect that and recalculate
++ * with fresh values.
++ */
++ get_time_values_from_xen();
++ continue;
++ }
++ } while (read_seqretry(&xtime_lock, seq) ||
++ (local_time_version != shadow->version));
++
++ put_cpu();
++
++ while (usec >= USEC_PER_SEC) {
++ usec -= USEC_PER_SEC;
++ sec++;
++ }
++
++ tv->tv_sec = sec;
++ tv->tv_usec = usec;
++}
++
++EXPORT_SYMBOL(do_gettimeofday);
++
++int do_settimeofday(struct timespec *tv)
++{
++ time_t sec;
++ s64 nsec;
++ unsigned int cpu;
++ struct shadow_time_info *shadow;
++ dom0_op_t op;
++
++ if ((unsigned long)tv->tv_nsec >= NSEC_PER_SEC)
++ return -EINVAL;
++
++ cpu = get_cpu();
++ shadow = &per_cpu(shadow_time, cpu);
++
++ write_seqlock_irq(&xtime_lock);
++
++ /*
++ * Ensure we don't get blocked for a long time so that our time delta
++ * overflows. If that were to happen then our shadow time values would
++ * be stale, so we can retry with fresh ones.
++ */
++ for (;;) {
++ nsec = tv->tv_nsec - get_nsec_offset(shadow);
++ if (time_values_up_to_date(cpu))
++ break;
++ get_time_values_from_xen();
++ }
++ sec = tv->tv_sec;
++ __normalize_time(&sec, &nsec);
++
++ if ((xen_start_info->flags & SIF_INITDOMAIN) &&
++ !independent_wallclock) {
++ op.cmd = DOM0_SETTIME;
++ op.u.settime.secs = sec;
++ op.u.settime.nsecs = nsec;
++ op.u.settime.system_time = shadow->system_timestamp;
++ HYPERVISOR_dom0_op(&op);
++ update_wallclock();
++ } else if (independent_wallclock) {
++ nsec -= shadow->system_timestamp;
++ __normalize_time(&sec, &nsec);
++ __update_wallclock(sec, nsec);
++ }
++
++ write_sequnlock_irq(&xtime_lock);
++
++ put_cpu();
++
++ clock_was_set();
++ return 0;
++}
++
++EXPORT_SYMBOL(do_settimeofday);
++
++static void sync_xen_wallclock(unsigned long dummy);
++static DEFINE_TIMER(sync_xen_wallclock_timer, sync_xen_wallclock, 0, 0);
++static void sync_xen_wallclock(unsigned long dummy)
++{
++ time_t sec;
++ s64 nsec;
++ dom0_op_t op;
++
++ if (!ntp_synced() || independent_wallclock ||
++ !(xen_start_info->flags & SIF_INITDOMAIN))
++ return;
++
++ write_seqlock_irq(&xtime_lock);
++
++ sec = xtime.tv_sec;
++ nsec = xtime.tv_nsec + ((jiffies - wall_jiffies) * (u64)NS_PER_TICK);
++ __normalize_time(&sec, &nsec);
++
++ op.cmd = DOM0_SETTIME;
++ op.u.settime.secs = sec;
++ op.u.settime.nsecs = nsec;
++ op.u.settime.system_time = processed_system_time;
++ HYPERVISOR_dom0_op(&op);
++
++ update_wallclock();
++
++ write_sequnlock_irq(&xtime_lock);
++
++ /* Once per minute. */
++ mod_timer(&sync_xen_wallclock_timer, jiffies + 60*HZ);
++}
++
++static int set_rtc_mmss(unsigned long nowtime)
++{
++ int retval;
++
++ WARN_ON(irqs_disabled());
++
++ if (independent_wallclock || !(xen_start_info->flags & SIF_INITDOMAIN))
++ return 0;
++
++ /* gets recalled with irq locally disabled */
++ spin_lock_irq(&rtc_lock);
++ if (efi_enabled)
++ retval = efi_set_rtc_mmss(nowtime);
++ else
++ retval = mach_set_rtc_mmss(nowtime);
++ spin_unlock_irq(&rtc_lock);
++
++ return retval;
++}
++
++/* monotonic_clock(): returns # of nanoseconds passed since time_init()
++ * Note: This function is required to return accurate
++ * time even in the absence of multiple timer ticks.
++ */
++unsigned long long monotonic_clock(void)
++{
++ int cpu = get_cpu();
++ struct shadow_time_info *shadow = &per_cpu(shadow_time, cpu);
++ u64 time;
++ u32 local_time_version;
++
++ do {
++ local_time_version = shadow->version;
++ barrier();
++ time = shadow->system_timestamp + get_nsec_offset(shadow);
++ if (!time_values_up_to_date(cpu))
++ get_time_values_from_xen();
++ barrier();
++ } while (local_time_version != shadow->version);
++
++ put_cpu();
++
++ return time;
++}
++EXPORT_SYMBOL(monotonic_clock);
++
++unsigned long long sched_clock(void)
++{
++ return monotonic_clock();
++}
++
++#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
++unsigned long profile_pc(struct pt_regs *regs)
++{
++ unsigned long pc = instruction_pointer(regs);
++
++#ifdef __x86_64__
++ /* Assume the lock function has either no stack frame or only a single word.
++ This checks if the address on the stack looks like a kernel text address.
++ There is a small window for false hits, but in that case the tick
++ is just accounted to the spinlock function.
++ Better would be to write these functions in assembler again
++ and check exactly. */
++ if (in_lock_functions(pc)) {
++ char *v = *(char **)regs->rsp;
++ if ((v >= _stext && v <= _etext) ||
++ (v >= _sinittext && v <= _einittext) ||
++ (v >= (char *)MODULES_VADDR && v <= (char *)MODULES_END))
++ return (unsigned long)v;
++ return ((unsigned long *)regs->rsp)[1];
++ }
++#else
++ if (in_lock_functions(pc))
++ return *(unsigned long *)(regs->ebp + 4);
++#endif
++
++ return pc;
++}
++EXPORT_SYMBOL(profile_pc);
++#endif
++
++irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs)
++{
++ s64 delta, delta_cpu, stolen, blocked;
++ u64 sched_time;
++ int i, cpu = smp_processor_id();
++ struct shadow_time_info *shadow = &per_cpu(shadow_time, cpu);
++ struct vcpu_runstate_info *runstate = &per_cpu(runstate, cpu);
++
++ write_seqlock(&xtime_lock);
++
++ do {
++ get_time_values_from_xen();
++
++ /* Obtain a consistent snapshot of elapsed wallclock cycles. */
++ delta = delta_cpu =
++ shadow->system_timestamp + get_nsec_offset(shadow);
++ delta -= processed_system_time;
++ delta_cpu -= per_cpu(processed_system_time, cpu);
++
++ /*
++ * Obtain a consistent snapshot of stolen/blocked cycles. We
++ * can use state_entry_time to detect if we get preempted here.
++ */
++ do {
++ sched_time = runstate->state_entry_time;
++ barrier();
++ stolen = runstate->time[RUNSTATE_runnable] +
++ runstate->time[RUNSTATE_offline] -
++ per_cpu(processed_stolen_time, cpu);
++ blocked = runstate->time[RUNSTATE_blocked] -
++ per_cpu(processed_blocked_time, cpu);
++ barrier();
++ } while (sched_time != runstate->state_entry_time);
++ } while (!time_values_up_to_date(cpu));
++
++ if ((unlikely(delta < -(s64)permitted_clock_jitter) ||
++ unlikely(delta_cpu < -(s64)permitted_clock_jitter))
++ && printk_ratelimit()) {
++ printk("Timer ISR/%d: Time went backwards: "
++ "delta=%lld delta_cpu=%lld shadow=%lld "
++ "off=%lld processed=%lld cpu_processed=%lld\n",
++ cpu, delta, delta_cpu, shadow->system_timestamp,
++ (s64)get_nsec_offset(shadow),
++ processed_system_time,
++ per_cpu(processed_system_time, cpu));
++ for (i = 0; i < num_online_cpus(); i++)
++ printk(" %d: %lld\n", i,
++ per_cpu(processed_system_time, i));
++ }
++
++ /* System-wide jiffy work. */
++ while (delta >= NS_PER_TICK) {
++ delta -= NS_PER_TICK;
++ processed_system_time += NS_PER_TICK;
++ do_timer(regs);
++ }
++
++ if (shadow_tv_version != HYPERVISOR_shared_info->wc_version) {
++ update_wallclock();
++ clock_was_set();
++ }
++
++ write_sequnlock(&xtime_lock);
++
++ /*
++ * Account stolen ticks.
++ * HACK: Passing NULL to account_steal_time()
++ * ensures that the ticks are accounted as stolen.
++ */
++ if ((stolen > 0) && (delta_cpu > 0)) {
++ delta_cpu -= stolen;
++ if (unlikely(delta_cpu < 0))
++ stolen += delta_cpu; /* clamp local-time progress */
++ do_div(stolen, NS_PER_TICK);
++ per_cpu(processed_stolen_time, cpu) += stolen * NS_PER_TICK;
++ per_cpu(processed_system_time, cpu) += stolen * NS_PER_TICK;
++ account_steal_time(NULL, (cputime_t)stolen);
++ }
++
++ /*
++ * Account blocked ticks.
++ * HACK: Passing idle_task to account_steal_time()
++ * ensures that the ticks are accounted as idle/wait.
++ */
++ if ((blocked > 0) && (delta_cpu > 0)) {
++ delta_cpu -= blocked;
++ if (unlikely(delta_cpu < 0))
++ blocked += delta_cpu; /* clamp local-time progress */
++ do_div(blocked, NS_PER_TICK);
++ per_cpu(processed_blocked_time, cpu) += blocked * NS_PER_TICK;
++ per_cpu(processed_system_time, cpu) += blocked * NS_PER_TICK;
++ account_steal_time(idle_task(cpu), (cputime_t)blocked);
++ }
++
++ /* Account user/system ticks. */
++ if (delta_cpu > 0) {
++ do_div(delta_cpu, NS_PER_TICK);
++ per_cpu(processed_system_time, cpu) += delta_cpu * NS_PER_TICK;
++ if (user_mode(regs))
++ account_user_time(current, (cputime_t)delta_cpu);
++ else
++ account_system_time(current, HARDIRQ_OFFSET,
++ (cputime_t)delta_cpu);
++ }
++
++ /* Local timer processing (see update_process_times()). */
++ run_local_timers();
++ if (rcu_pending(cpu))
++ rcu_check_callbacks(cpu, user_mode(regs));
++ scheduler_tick();
++ run_posix_cpu_timers(current);
++
++ return IRQ_HANDLED;
++}
++
++static void init_missing_ticks_accounting(int cpu)
++{
++ struct vcpu_register_runstate_memory_area area;
++ struct vcpu_runstate_info *runstate = &per_cpu(runstate, cpu);
++
++ memset(runstate, 0, sizeof(*runstate));
++
++ area.addr.v = runstate;
++ HYPERVISOR_vcpu_op(VCPUOP_register_runstate_memory_area, cpu, &area);
++
++ per_cpu(processed_blocked_time, cpu) =
++ runstate->time[RUNSTATE_blocked];
++ per_cpu(processed_stolen_time, cpu) =
++ runstate->time[RUNSTATE_runnable] +
++ runstate->time[RUNSTATE_offline];
++}
++
++/* not static: needed by APM */
++unsigned long get_cmos_time(void)
++{
++ unsigned long retval;
++
++ spin_lock(&rtc_lock);
++
++ if (efi_enabled)
++ retval = efi_get_time();
++ else
++ retval = mach_get_cmos_time();
++
++ spin_unlock(&rtc_lock);
++
++ return retval;
++}
++EXPORT_SYMBOL(get_cmos_time);
++
++static void sync_cmos_clock(unsigned long dummy);
++
++static DEFINE_TIMER(sync_cmos_timer, sync_cmos_clock, 0, 0);
++
++static void sync_cmos_clock(unsigned long dummy)
++{
++ struct timeval now, next;
++ int fail = 1;
++
++ /*
++ * If we have an externally synchronized Linux clock, then update
++ * CMOS clock accordingly every ~11 minutes. Set_rtc_mmss() has to be
++ * called as close as possible to 500 ms before the new second starts.
++ * This code is run on a timer. If the clock is set, that timer
++ * may not expire at the correct time. Thus, we adjust...
++ */
++ if (!ntp_synced())
++ /*
++ * Not synced, exit, do not restart a timer (if one is
++ * running, let it run out).
++ */
++ return;
++
++ do_gettimeofday(&now);
++ if (now.tv_usec >= USEC_AFTER - ((unsigned) TICK_SIZE) / 2 &&
++ now.tv_usec <= USEC_BEFORE + ((unsigned) TICK_SIZE) / 2)
++ fail = set_rtc_mmss(now.tv_sec);
++
++ next.tv_usec = USEC_AFTER - now.tv_usec;
++ if (next.tv_usec <= 0)
++ next.tv_usec += USEC_PER_SEC;
++
++ if (!fail)
++ next.tv_sec = 659;
++ else
++ next.tv_sec = 0;
++
++ if (next.tv_usec >= USEC_PER_SEC) {
++ next.tv_sec++;
++ next.tv_usec -= USEC_PER_SEC;
++ }
++ mod_timer(&sync_cmos_timer, jiffies + timeval_to_jiffies(&next));
++}
++
++void notify_arch_cmos_timer(void)
++{
++ mod_timer(&sync_cmos_timer, jiffies + 1);
++ mod_timer(&sync_xen_wallclock_timer, jiffies + 1);
++}
++
++static long clock_cmos_diff, sleep_start;
++
++static struct timer_opts *last_timer;
++static int timer_suspend(struct sys_device *dev, pm_message_t state)
++{
++ /*
++ * Estimate time zone so that set_time can update the clock
++ */
++ clock_cmos_diff = -get_cmos_time();
++ clock_cmos_diff += get_seconds();
++ sleep_start = get_cmos_time();
++ last_timer = cur_timer;
++ cur_timer = &timer_none;
++ if (last_timer->suspend)
++ last_timer->suspend(state);
++ return 0;
++}
++
++static int timer_resume(struct sys_device *dev)
++{
++ unsigned long flags;
++ unsigned long sec;
++ unsigned long sleep_length;
++
++#ifdef CONFIG_HPET_TIMER
++ if (is_hpet_enabled())
++ hpet_reenable();
++#endif
++ sec = get_cmos_time() + clock_cmos_diff;
++ sleep_length = (get_cmos_time() - sleep_start) * HZ;
++ write_seqlock_irqsave(&xtime_lock, flags);
++ xtime.tv_sec = sec;
++ xtime.tv_nsec = 0;
++ jiffies_64 += sleep_length;
++ wall_jiffies += sleep_length;
++ write_sequnlock_irqrestore(&xtime_lock, flags);
++ if (last_timer->resume)
++ last_timer->resume();
++ cur_timer = last_timer;
++ last_timer = NULL;
++ touch_softlockup_watchdog();
++ return 0;
++}
++
++static struct sysdev_class timer_sysclass = {
++ .resume = timer_resume,
++ .suspend = timer_suspend,
++ set_kset_name("timer"),
++};
++
++
++/* XXX this driverfs stuff should probably go elsewhere later -john */
++static struct sys_device device_timer = {
++ .id = 0,
++ .cls = &timer_sysclass,
++};
++
++static int time_init_device(void)
++{
++ int error = sysdev_class_register(&timer_sysclass);
++ if (!error)
++ error = sysdev_register(&device_timer);
++ return error;
++}
++
++device_initcall(time_init_device);
++
++#ifdef CONFIG_HPET_TIMER
++extern void (*late_time_init)(void);
++/* Duplicate of time_init() below, with hpet_enable part added */
++static void __init hpet_time_init(void)
++{
++ xtime.tv_sec = get_cmos_time();
++ xtime.tv_nsec = (INITIAL_JIFFIES % HZ) * (NSEC_PER_SEC / HZ);
++ set_normalized_timespec(&wall_to_monotonic,
++ -xtime.tv_sec, -xtime.tv_nsec);
++
++ if ((hpet_enable() >= 0) && hpet_use_timer) {
++ printk("Using HPET for base-timer\n");
++ }
++
++ cur_timer = select_timer();
++ printk(KERN_INFO "Using %s for high-res timesource\n",cur_timer->name);
++
++ time_init_hook();
++}
++#endif
++
++/* Dynamically-mapped IRQ. */
++DEFINE_PER_CPU(int, timer_irq);
++
++extern void (*late_time_init)(void);
++static void setup_cpu0_timer_irq(void)
++{
++ per_cpu(timer_irq, 0) =
++ bind_virq_to_irqhandler(
++ VIRQ_TIMER,
++ 0,
++ timer_interrupt,
++ SA_INTERRUPT,
++ "timer0",
++ NULL);
++ BUG_ON(per_cpu(timer_irq, 0) < 0);
++}
++
++void __init time_init(void)
++{
++#ifdef CONFIG_HPET_TIMER
++ if (is_hpet_capable()) {
++ /*
++ * HPET initialization needs to do memory-mapped io. So, let
++ * us do a late initialization after mem_init().
++ */
++ late_time_init = hpet_time_init;
++ return;
++ }
++#endif
++ get_time_values_from_xen();
++
++ processed_system_time = per_cpu(shadow_time, 0).system_timestamp;
++ per_cpu(processed_system_time, 0) = processed_system_time;
++ init_missing_ticks_accounting(0);
++
++ update_wallclock();
++
++ init_cpu_khz();
++ printk(KERN_INFO "Xen reported: %u.%03u MHz processor.\n",
++ cpu_khz / 1000, cpu_khz % 1000);
++
++#if defined(__x86_64__)
++ vxtime.mode = VXTIME_TSC;
++ vxtime.quot = (1000000L << 32) / vxtime_hz;
++ vxtime.tsc_quot = (1000L << 32) / cpu_khz;
++ sync_core();
++ rdtscll(vxtime.last_tsc);
++#endif
++
++ /* Cannot request_irq() until kmem is initialised. */
++ late_time_init = setup_cpu0_timer_irq;
++}
++
++/* Convert jiffies to system time. */
++u64 jiffies_to_st(unsigned long j)
++{
++ unsigned long seq;
++ long delta;
++ u64 st;
++
++ do {
++ seq = read_seqbegin(&xtime_lock);
++ delta = j - jiffies;
++ /* NB. The next check can trigger in some wrap-around cases,
++ * but that's ok: we'll just end up with a shorter timeout. */
++ if (delta < 1)
++ delta = 1;
++ st = processed_system_time + (delta * (u64)NS_PER_TICK);
++ } while (read_seqretry(&xtime_lock, seq));
++
++ return st;
++}
++EXPORT_SYMBOL(jiffies_to_st);
++
++/*
++ * stop_hz_timer / start_hz_timer - enter/exit 'tickless mode' on an idle cpu
++ * These functions are based on implementations from arch/s390/kernel/time.c
++ */
++static void stop_hz_timer(void)
++{
++ unsigned int cpu = smp_processor_id();
++ unsigned long j;
++
++ cpu_set(cpu, nohz_cpu_mask);
++
++ /* See matching smp_mb in rcu_start_batch in rcupdate.c. These mbs */
++ /* ensure that if __rcu_pending (nested in rcu_needs_cpu) fetches a */
++ /* value of rcp->cur that matches rdp->quiescbatch and allows us to */
++ /* stop the hz timer then the cpumasks created for subsequent values */
++ /* of cur in rcu_start_batch are guaranteed to pick up the updated */
++ /* nohz_cpu_mask and so will not depend on this cpu. */
++
++ smp_mb();
++
++ /* Leave ourselves in 'tick mode' if rcu or softirq or timer pending. */
++ if (rcu_needs_cpu(cpu) || local_softirq_pending() ||
++ (j = next_timer_interrupt(), time_before_eq(j, jiffies))) {
++ cpu_clear(cpu, nohz_cpu_mask);
++ j = jiffies + 1;
++ }
++
++ BUG_ON(HYPERVISOR_set_timer_op(jiffies_to_st(j)) != 0);
++}
++
++static void start_hz_timer(void)
++{
++ cpu_clear(smp_processor_id(), nohz_cpu_mask);
++}
++
++void safe_halt(void)
++{
++ stop_hz_timer();
++ /* Blocking includes an implicit local_irq_enable(). */
++ HYPERVISOR_block();
++ start_hz_timer();
++}
++EXPORT_SYMBOL(safe_halt);
++
++void halt(void)
++{
++ if (irqs_disabled())
++ HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
++}
++EXPORT_SYMBOL(halt);
++
++/* No locking required. We are only CPU running, and interrupts are off. */
++void time_resume(void)
++{
++ init_cpu_khz();
++
++ get_time_values_from_xen();
++
++ processed_system_time = per_cpu(shadow_time, 0).system_timestamp;
++ per_cpu(processed_system_time, 0) = processed_system_time;
++ init_missing_ticks_accounting(0);
++
++ update_wallclock();
++}
++
++#ifdef CONFIG_SMP
++static char timer_name[NR_CPUS][15];
++
++void local_setup_timer(unsigned int cpu)
++{
++ int seq;
++
++ BUG_ON(cpu == 0);
++
++ do {
++ seq = read_seqbegin(&xtime_lock);
++ /* Use cpu0 timestamp: cpu's shadow is not initialised yet. */
++ per_cpu(processed_system_time, cpu) =
++ per_cpu(shadow_time, 0).system_timestamp;
++ init_missing_ticks_accounting(cpu);
++ } while (read_seqretry(&xtime_lock, seq));
++
++ sprintf(timer_name[cpu], "timer%d", cpu);
++ per_cpu(timer_irq, cpu) =
++ bind_virq_to_irqhandler(
++ VIRQ_TIMER,
++ cpu,
++ timer_interrupt,
++ SA_INTERRUPT,
++ timer_name[cpu],
++ NULL);
++ BUG_ON(per_cpu(timer_irq, cpu) < 0);
++}
++
++void local_teardown_timer(unsigned int cpu)
++{
++ BUG_ON(cpu == 0);
++ unbind_from_irqhandler(per_cpu(timer_irq, cpu), NULL);
++}
++#endif
++
++/*
++ * /proc/sys/xen: This really belongs in another file. It can stay here for
++ * now however.
++ */
++static ctl_table xen_subtable[] = {
++ {
++ .ctl_name = 1,
++ .procname = "independent_wallclock",
++ .data = &independent_wallclock,
++ .maxlen = sizeof(independent_wallclock),
++ .mode = 0644,
++ .proc_handler = proc_dointvec
++ },
++ {
++ .ctl_name = 2,
++ .procname = "permitted_clock_jitter",
++ .data = &permitted_clock_jitter,
++ .maxlen = sizeof(permitted_clock_jitter),
++ .mode = 0644,
++ .proc_handler = proc_doulongvec_minmax
++ },
++ { 0 }
++};
++static ctl_table xen_table[] = {
++ {
++ .ctl_name = 123,
++ .procname = "xen",
++ .mode = 0555,
++ .child = xen_subtable},
++ { 0 }
++};
++static int __init xen_sysctl_init(void)
++{
++ (void)register_sysctl_table(xen_table, 0);
++ return 0;
++}
++__initcall(xen_sysctl_init);
+diff -urNp linux-2.6/arch/i386/kernel/traps.c new/arch/i386/kernel/traps.c
+--- linux-2.6/arch/i386/kernel/traps.c 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/traps.c 2006-05-23 18:37:09.000000000 +0200
+@@ -607,18 +607,11 @@ static void mem_parity_error(unsigned ch
+
+ static void io_check_error(unsigned char reason, struct pt_regs * regs)
+ {
+- unsigned long i;
+-
+ printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n");
+ show_registers(regs);
+
+ /* Re-enable the IOCK line, wait for a few seconds */
+- reason = (reason & 0xf) | 8;
+- outb(reason, 0x61);
+- i = 2000;
+- while (--i) udelay(1000);
+- reason &= ~8;
+- outb(reason, 0x61);
++ clear_io_check_error(reason);
+ }
+
+ static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
+diff -urNp linux-2.6/arch/i386/kernel/traps-xen.c new/arch/i386/kernel/traps-xen.c
+--- linux-2.6/arch/i386/kernel/traps-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/traps-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,1136 @@
++/*
++ * linux/arch/i386/traps.c
++ *
++ * Copyright (C) 1991, 1992 Linus Torvalds
++ *
++ * Pentium III FXSR, SSE support
++ * Gareth Hughes <gareth@valinux.com>, May 2000
++ */
++
++/*
++ * 'Traps.c' handles hardware traps and faults after we have saved some
++ * state in 'asm.s'.
++ */
++#include <linux/config.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/timer.h>
++#include <linux/mm.h>
++#include <linux/init.h>
++#include <linux/delay.h>
++#include <linux/spinlock.h>
++#include <linux/interrupt.h>
++#include <linux/highmem.h>
++#include <linux/kallsyms.h>
++#include <linux/ptrace.h>
++#include <linux/utsname.h>
++#include <linux/kprobes.h>
++#include <linux/kexec.h>
++
++#ifdef CONFIG_EISA
++#include <linux/ioport.h>
++#include <linux/eisa.h>
++#endif
++
++#ifdef CONFIG_MCA
++#include <linux/mca.h>
++#endif
++
++#include <asm/processor.h>
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/io.h>
++#include <asm/atomic.h>
++#include <asm/debugreg.h>
++#include <asm/desc.h>
++#include <asm/i387.h>
++#include <asm/nmi.h>
++
++#include <asm/smp.h>
++#include <asm/arch_hooks.h>
++#include <asm/kdebug.h>
++
++#include <linux/module.h>
++
++#include "mach_traps.h"
++
++asmlinkage int system_call(void);
++
++struct desc_struct default_ldt[] = { { 0, 0 }, { 0, 0 }, { 0, 0 },
++ { 0, 0 }, { 0, 0 } };
++
++/* Do we ignore FPU interrupts ? */
++char ignore_fpu_irq = 0;
++
++#ifndef CONFIG_X86_NO_IDT
++/*
++ * The IDT has to be page-aligned to simplify the Pentium
++ * F0 0F bug workaround.. We have a special link segment
++ * for this.
++ */
++struct desc_struct idt_table[256] __attribute__((__section__(".data.idt"))) = { {0, 0}, };
++#endif
++
++asmlinkage void divide_error(void);
++asmlinkage void debug(void);
++asmlinkage void nmi(void);
++asmlinkage void int3(void);
++asmlinkage void overflow(void);
++asmlinkage void bounds(void);
++asmlinkage void invalid_op(void);
++asmlinkage void device_not_available(void);
++asmlinkage void coprocessor_segment_overrun(void);
++asmlinkage void invalid_TSS(void);
++asmlinkage void segment_not_present(void);
++asmlinkage void stack_segment(void);
++asmlinkage void general_protection(void);
++asmlinkage void page_fault(void);
++asmlinkage void coprocessor_error(void);
++asmlinkage void simd_coprocessor_error(void);
++asmlinkage void alignment_check(void);
++#ifndef CONFIG_XEN
++asmlinkage void spurious_interrupt_bug(void);
++#else
++asmlinkage void fixup_4gb_segment(void);
++#endif
++asmlinkage void machine_check(void);
++
++static int kstack_depth_to_print = 24;
++ATOMIC_NOTIFIER_HEAD(i386die_chain);
++
++int register_die_notifier(struct notifier_block *nb)
++{
++ vmalloc_sync_all();
++ return atomic_notifier_chain_register(&i386die_chain, nb);
++}
++EXPORT_SYMBOL(register_die_notifier);
++
++int unregister_die_notifier(struct notifier_block *nb)
++{
++ return atomic_notifier_chain_unregister(&i386die_chain, nb);
++}
++EXPORT_SYMBOL(unregister_die_notifier);
++
++static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
++{
++ return p > (void *)tinfo &&
++ p < (void *)tinfo + THREAD_SIZE - 3;
++}
++
++/*
++ * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line.
++ */
++static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl,
++ int printed)
++{
++ if (!printed)
++ printk(log_lvl);
++
++#if CONFIG_STACK_BACKTRACE_COLS == 1
++ printk(" [<%08lx>] ", addr);
++#else
++ printk(" <%08lx> ", addr);
++#endif
++ print_symbol("%s", addr);
++
++ printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS;
++ if (printed)
++ printk(" ");
++ else
++ printk("\n");
++
++ return printed;
++}
++
++static inline unsigned long print_context_stack(struct thread_info *tinfo,
++ unsigned long *stack, unsigned long ebp,
++ char *log_lvl)
++{
++ unsigned long addr;
++ int printed = 0; /* nr of entries already printed on current line */
++
++#ifdef CONFIG_FRAME_POINTER
++ while (valid_stack_ptr(tinfo, (void *)ebp)) {
++ addr = *(unsigned long *)(ebp + 4);
++ printed = print_addr_and_symbol(addr, log_lvl, printed);
++ ebp = *(unsigned long *)ebp;
++ }
++#else
++ while (valid_stack_ptr(tinfo, stack)) {
++ addr = *stack++;
++ if (__kernel_text_address(addr))
++ printed = print_addr_and_symbol(addr, log_lvl, printed);
++ }
++#endif
++ if (printed)
++ printk("\n");
++
++ return ebp;
++}
++
++static void show_trace_log_lvl(struct task_struct *task,
++ unsigned long *stack, char *log_lvl)
++{
++ unsigned long ebp;
++
++ if (!task)
++ task = current;
++
++ if (task == current) {
++ /* Grab ebp right from our regs */
++ asm ("movl %%ebp, %0" : "=r" (ebp) : );
++ } else {
++ /* ebp is the last reg pushed by switch_to */
++ ebp = *(unsigned long *) task->thread.esp;
++ }
++
++ while (1) {
++ struct thread_info *context;
++ context = (struct thread_info *)
++ ((unsigned long)stack & (~(THREAD_SIZE - 1)));
++ ebp = print_context_stack(context, stack, ebp, log_lvl);
++ stack = (unsigned long*)context->previous_esp;
++ if (!stack)
++ break;
++ printk("%s =======================\n", log_lvl);
++ }
++}
++
++void show_trace(struct task_struct *task, unsigned long * stack)
++{
++ show_trace_log_lvl(task, stack, "");
++}
++
++static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp,
++ char *log_lvl)
++{
++ unsigned long *stack;
++ int i;
++
++ if (esp == NULL) {
++ if (task)
++ esp = (unsigned long*)task->thread.esp;
++ else
++ esp = (unsigned long *)&esp;
++ }
++
++ stack = esp;
++ printk(log_lvl);
++ for(i = 0; i < kstack_depth_to_print; i++) {
++ if (kstack_end(stack))
++ break;
++ if (i && ((i % 8) == 0))
++ printk("\n%s ", log_lvl);
++ printk("%08lx ", *stack++);
++ }
++ printk("\n%sCall Trace:\n", log_lvl);
++ show_trace_log_lvl(task, esp, log_lvl);
++}
++
++void show_stack(struct task_struct *task, unsigned long *esp)
++{
++ printk(" ");
++ show_stack_log_lvl(task, esp, "");
++}
++
++/*
++ * The architecture-independent dump_stack generator
++ */
++void dump_stack(void)
++{
++ unsigned long stack;
++
++ show_trace(current, &stack);
++}
++
++EXPORT_SYMBOL(dump_stack);
++
++void show_registers(struct pt_regs *regs)
++{
++ int i;
++ int in_kernel = 1;
++ unsigned long esp;
++ unsigned short ss;
++
++ esp = (unsigned long) (®s->esp);
++ savesegment(ss, ss);
++ if (user_mode_vm(regs)) {
++ in_kernel = 0;
++ esp = regs->esp;
++ ss = regs->xss & 0xffff;
++ }
++ print_modules();
++ printk(KERN_EMERG "CPU: %d\nEIP: %04x:[<%08lx>] %s VLI\n"
++ "EFLAGS: %08lx (%s %.*s) \n",
++ smp_processor_id(), 0xffff & regs->xcs, regs->eip,
++ print_tainted(), regs->eflags, system_utsname.release,
++ (int)strcspn(system_utsname.version, " "),
++ system_utsname.version);
++ print_symbol(KERN_EMERG "EIP is at %s\n", regs->eip);
++ printk(KERN_EMERG "eax: %08lx ebx: %08lx ecx: %08lx edx: %08lx\n",
++ regs->eax, regs->ebx, regs->ecx, regs->edx);
++ printk(KERN_EMERG "esi: %08lx edi: %08lx ebp: %08lx esp: %08lx\n",
++ regs->esi, regs->edi, regs->ebp, esp);
++ printk(KERN_EMERG "ds: %04x es: %04x ss: %04x\n",
++ regs->xds & 0xffff, regs->xes & 0xffff, ss);
++ printk(KERN_EMERG "Process %s (pid: %d, threadinfo=%p task=%p)",
++ current->comm, current->pid, current_thread_info(), current);
++ /*
++ * When in-kernel, we also print out the stack and code at the
++ * time of the fault..
++ */
++ if (in_kernel) {
++ u8 __user *eip;
++
++ printk("\n" KERN_EMERG "Stack: ");
++ show_stack_log_lvl(NULL, (unsigned long *)esp, KERN_EMERG);
++
++ printk(KERN_EMERG "Code: ");
++
++ eip = (u8 __user *)regs->eip - 43;
++ for (i = 0; i < 64; i++, eip++) {
++ unsigned char c;
++
++ if (eip < (u8 __user *)PAGE_OFFSET || __get_user(c, eip)) {
++ printk(" Bad EIP value.");
++ break;
++ }
++ if (eip == (u8 __user *)regs->eip)
++ printk("<%02x> ", c);
++ else
++ printk("%02x ", c);
++ }
++ }
++ printk("\n");
++}
++
++static void handle_BUG(struct pt_regs *regs)
++{
++ unsigned short ud2;
++ unsigned short line;
++ char *file;
++ char c;
++ unsigned long eip;
++
++ eip = regs->eip;
++
++ if (eip < PAGE_OFFSET)
++ goto no_bug;
++ if (__get_user(ud2, (unsigned short __user *)eip))
++ goto no_bug;
++ if (ud2 != 0x0b0f)
++ goto no_bug;
++ if (__get_user(line, (unsigned short __user *)(eip + 2)))
++ goto bug;
++ if (__get_user(file, (char * __user *)(eip + 4)) ||
++ (unsigned long)file < PAGE_OFFSET || __get_user(c, file))
++ file = "<bad filename>";
++
++ printk(KERN_EMERG "------------[ cut here ]------------\n");
++ printk(KERN_EMERG "kernel BUG at %s:%d!\n", file, line);
++
++no_bug:
++ return;
++
++ /* Here we know it was a BUG but file-n-line is unavailable */
++bug:
++ printk(KERN_EMERG "Kernel BUG\n");
++}
++
++/* This is gone through when something in the kernel
++ * has done something bad and is about to be terminated.
++*/
++void die(const char * str, struct pt_regs * regs, long err)
++{
++ static struct {
++ spinlock_t lock;
++ u32 lock_owner;
++ int lock_owner_depth;
++ } die = {
++ .lock = SPIN_LOCK_UNLOCKED,
++ .lock_owner = -1,
++ .lock_owner_depth = 0
++ };
++ static int die_counter;
++ unsigned long flags;
++
++ oops_enter();
++
++ if (die.lock_owner != raw_smp_processor_id()) {
++ console_verbose();
++ spin_lock_irqsave(&die.lock, flags);
++ die.lock_owner = smp_processor_id();
++ die.lock_owner_depth = 0;
++ bust_spinlocks(1);
++ }
++ else
++ local_save_flags(flags);
++
++ if (++die.lock_owner_depth < 3) {
++ int nl = 0;
++ unsigned long esp;
++ unsigned short ss;
++
++ handle_BUG(regs);
++ printk(KERN_EMERG "%s: %04lx [#%d]\n", str, err & 0xffff, ++die_counter);
++#ifdef CONFIG_PREEMPT
++ printk(KERN_EMERG "PREEMPT ");
++ nl = 1;
++#endif
++#ifdef CONFIG_SMP
++ if (!nl)
++ printk(KERN_EMERG);
++ printk("SMP ");
++ nl = 1;
++#endif
++#ifdef CONFIG_DEBUG_PAGEALLOC
++ if (!nl)
++ printk(KERN_EMERG);
++ printk("DEBUG_PAGEALLOC");
++ nl = 1;
++#endif
++ if (nl)
++ printk("\n");
++ if (notify_die(DIE_OOPS, str, regs, err,
++ current->thread.trap_no, SIGSEGV) !=
++ NOTIFY_STOP) {
++ show_registers(regs);
++ /* Executive summary in case the oops scrolled away */
++ esp = (unsigned long) (®s->esp);
++ savesegment(ss, ss);
++ if (user_mode(regs)) {
++ esp = regs->esp;
++ ss = regs->xss & 0xffff;
++ }
++ printk(KERN_EMERG "EIP: [<%08lx>] ", regs->eip);
++ print_symbol("%s", regs->eip);
++ printk(" SS:ESP %04x:%08lx\n", ss, esp);
++ }
++ else
++ regs = NULL;
++ } else
++ printk(KERN_EMERG "Recursive die() failure, output suppressed\n");
++
++ bust_spinlocks(0);
++ die.lock_owner = -1;
++ spin_unlock_irqrestore(&die.lock, flags);
++
++ if (!regs)
++ return;
++
++ if (kexec_should_crash(current))
++ crash_kexec(regs);
++
++ if (in_interrupt())
++ panic("Fatal exception in interrupt");
++
++ if (panic_on_oops) {
++ printk(KERN_EMERG "Fatal exception: panic in 5 seconds\n");
++ ssleep(5);
++ panic("Fatal exception");
++ }
++ oops_exit();
++ do_exit(SIGSEGV);
++}
++
++static inline void die_if_kernel(const char * str, struct pt_regs * regs, long err)
++{
++ if (!user_mode_vm(regs))
++ die(str, regs, err);
++}
++
++static void __kprobes do_trap(int trapnr, int signr, char *str, int vm86,
++ struct pt_regs * regs, long error_code,
++ siginfo_t *info)
++{
++ struct task_struct *tsk = current;
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = trapnr;
++
++ if (regs->eflags & VM_MASK) {
++ if (vm86)
++ goto vm86_trap;
++ goto trap_signal;
++ }
++
++ if (!user_mode(regs))
++ goto kernel_trap;
++
++ trap_signal: {
++ if (info)
++ force_sig_info(signr, info, tsk);
++ else
++ force_sig(signr, tsk);
++ return;
++ }
++
++ kernel_trap: {
++ if (!fixup_exception(regs))
++ die(str, regs, error_code);
++ return;
++ }
++
++ vm86_trap: {
++ int ret = handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, trapnr);
++ if (ret) goto trap_signal;
++ return;
++ }
++}
++
++#define DO_ERROR(trapnr, signr, str, name) \
++fastcall void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, 0, regs, error_code, NULL); \
++}
++
++#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
++fastcall void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ siginfo_t info; \
++ info.si_signo = signr; \
++ info.si_errno = 0; \
++ info.si_code = sicode; \
++ info.si_addr = (void __user *)siaddr; \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, 0, regs, error_code, &info); \
++}
++
++#define DO_VM86_ERROR(trapnr, signr, str, name) \
++fastcall void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, 1, regs, error_code, NULL); \
++}
++
++#define DO_VM86_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
++fastcall void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ siginfo_t info; \
++ info.si_signo = signr; \
++ info.si_errno = 0; \
++ info.si_code = sicode; \
++ info.si_addr = (void __user *)siaddr; \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, 1, regs, error_code, &info); \
++}
++
++DO_VM86_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->eip)
++#ifndef CONFIG_KPROBES
++DO_VM86_ERROR( 3, SIGTRAP, "int3", int3)
++#endif
++DO_VM86_ERROR( 4, SIGSEGV, "overflow", overflow)
++DO_VM86_ERROR( 5, SIGSEGV, "bounds", bounds)
++DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->eip)
++DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
++DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
++DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
++DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
++DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
++DO_ERROR_INFO(32, SIGSEGV, "iret exception", iret_error, ILL_BADSTK, 0)
++
++fastcall void __kprobes do_general_protection(struct pt_regs * regs,
++ long error_code)
++{
++ current->thread.error_code = error_code;
++ current->thread.trap_no = 13;
++
++ if (regs->eflags & VM_MASK)
++ goto gp_in_vm86;
++
++ if (!user_mode(regs))
++ goto gp_in_kernel;
++
++ current->thread.error_code = error_code;
++ current->thread.trap_no = 13;
++ force_sig(SIGSEGV, current);
++ return;
++
++gp_in_vm86:
++ local_irq_enable();
++ handle_vm86_fault((struct kernel_vm86_regs *) regs, error_code);
++ return;
++
++gp_in_kernel:
++ if (!fixup_exception(regs)) {
++ if (notify_die(DIE_GPF, "general protection fault", regs,
++ error_code, 13, SIGSEGV) == NOTIFY_STOP)
++ return;
++ die("general protection fault", regs, error_code);
++ }
++}
++
++static void mem_parity_error(unsigned char reason, struct pt_regs * regs)
++{
++ printk(KERN_EMERG "Uhhuh. NMI received. Dazed and confused, but trying "
++ "to continue\n");
++ printk(KERN_EMERG "You probably have a hardware problem with your RAM "
++ "chips\n");
++
++ /* Clear and disable the memory parity error line. */
++ clear_mem_error(reason);
++}
++
++static void io_check_error(unsigned char reason, struct pt_regs * regs)
++{
++ printk(KERN_EMERG "NMI: IOCK error (debug interrupt?)\n");
++ show_registers(regs);
++
++ /* Re-enable the IOCK line, wait for a few seconds */
++ clear_io_check_error(reason);
++}
++
++static void unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
++{
++#ifdef CONFIG_MCA
++ /* Might actually be able to figure out what the guilty party
++ * is. */
++ if( MCA_bus ) {
++ mca_handle_nmi();
++ return;
++ }
++#endif
++ printk("Uhhuh. NMI received for unknown reason %02x on CPU %d.\n",
++ reason, smp_processor_id());
++ printk("Dazed and confused, but trying to continue\n");
++ printk("Do you have a strange power saving mode enabled?\n");
++}
++
++static DEFINE_SPINLOCK(nmi_print_lock);
++
++void die_nmi (struct pt_regs *regs, const char *msg)
++{
++ if (notify_die(DIE_NMIWATCHDOG, msg, regs, 0, 2, SIGINT) ==
++ NOTIFY_STOP)
++ return;
++
++ spin_lock(&nmi_print_lock);
++ /*
++ * We are in trouble anyway, lets at least try
++ * to get a message out.
++ */
++ bust_spinlocks(1);
++ printk(KERN_EMERG "%s", msg);
++ printk(" on CPU%d, eip %08lx, registers:\n",
++ smp_processor_id(), regs->eip);
++ show_registers(regs);
++ printk(KERN_EMERG "console shuts up ...\n");
++ console_silent();
++ spin_unlock(&nmi_print_lock);
++ bust_spinlocks(0);
++
++ /* If we are in kernel we are probably nested up pretty bad
++ * and might aswell get out now while we still can.
++ */
++ if (!user_mode_vm(regs)) {
++ current->thread.trap_no = 2;
++ crash_kexec(regs);
++ }
++
++ do_exit(SIGSEGV);
++}
++
++static void default_do_nmi(struct pt_regs * regs)
++{
++ unsigned char reason = 0;
++
++ /* Only the BSP gets external NMIs from the system. */
++ if (!smp_processor_id())
++ reason = get_nmi_reason();
++
++ if (!(reason & 0xc0)) {
++ if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
++ == NOTIFY_STOP)
++ return;
++#ifdef CONFIG_X86_LOCAL_APIC
++ /*
++ * Ok, so this is none of the documented NMI sources,
++ * so it must be the NMI watchdog.
++ */
++ if (nmi_watchdog) {
++ nmi_watchdog_tick(regs);
++ return;
++ }
++#endif
++ unknown_nmi_error(reason, regs);
++ return;
++ }
++ if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
++ return;
++ if (reason & 0x80)
++ mem_parity_error(reason, regs);
++ if (reason & 0x40)
++ io_check_error(reason, regs);
++ /*
++ * Reassert NMI in case it became active meanwhile
++ * as it's edge-triggered.
++ */
++ reassert_nmi();
++}
++
++static int dummy_nmi_callback(struct pt_regs * regs, int cpu)
++{
++ return 0;
++}
++
++static nmi_callback_t nmi_callback = dummy_nmi_callback;
++
++fastcall void do_nmi(struct pt_regs * regs, long error_code)
++{
++ int cpu;
++
++ nmi_enter();
++
++ cpu = smp_processor_id();
++
++ ++nmi_count(cpu);
++
++ if (!rcu_dereference(nmi_callback)(regs, cpu))
++ default_do_nmi(regs);
++
++ nmi_exit();
++}
++
++void set_nmi_callback(nmi_callback_t callback)
++{
++ vmalloc_sync_all();
++ rcu_assign_pointer(nmi_callback, callback);
++}
++EXPORT_SYMBOL_GPL(set_nmi_callback);
++
++void unset_nmi_callback(void)
++{
++ nmi_callback = dummy_nmi_callback;
++}
++EXPORT_SYMBOL_GPL(unset_nmi_callback);
++
++#ifdef CONFIG_KPROBES
++fastcall void __kprobes do_int3(struct pt_regs *regs, long error_code)
++{
++ if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP)
++ == NOTIFY_STOP)
++ return;
++ /* This is an interrupt gate, because kprobes wants interrupts
++ disabled. Normal trap handlers don't. */
++ restore_interrupts(regs);
++ do_trap(3, SIGTRAP, "int3", 1, regs, error_code, NULL);
++}
++#endif
++
++/*
++ * Our handling of the processor debug registers is non-trivial.
++ * We do not clear them on entry and exit from the kernel. Therefore
++ * it is possible to get a watchpoint trap here from inside the kernel.
++ * However, the code in ./ptrace.c has ensured that the user can
++ * only set watchpoints on userspace addresses. Therefore the in-kernel
++ * watchpoint trap can only occur in code which is reading/writing
++ * from user space. Such code must not hold kernel locks (since it
++ * can equally take a page fault), therefore it is safe to call
++ * force_sig_info even though that claims and releases locks.
++ *
++ * Code in ./signal.c ensures that the debug control register
++ * is restored before we deliver any signal, and therefore that
++ * user code runs with the correct debug control register even though
++ * we clear it here.
++ *
++ * Being careful here means that we don't have to be as careful in a
++ * lot of more complicated places (task switching can be a bit lazy
++ * about restoring all the debug state, and ptrace doesn't have to
++ * find every occurrence of the TF bit that could be saved away even
++ * by user code)
++ */
++fastcall void __kprobes do_debug(struct pt_regs * regs, long error_code)
++{
++ unsigned int condition;
++ struct task_struct *tsk = current;
++
++ get_debugreg(condition, 6);
++
++ if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
++ SIGTRAP) == NOTIFY_STOP)
++ return;
++ /* It's safe to allow irq's after DR6 has been saved */
++ if (regs->eflags & X86_EFLAGS_IF)
++ local_irq_enable();
++
++ /* Mask out spurious debug traps due to lazy DR7 setting */
++ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
++ if (!tsk->thread.debugreg[7])
++ goto clear_dr7;
++ }
++
++ if (regs->eflags & VM_MASK)
++ goto debug_vm86;
++
++ /* Save debug status register where ptrace can see it */
++ tsk->thread.debugreg[6] = condition;
++
++ /*
++ * Single-stepping through TF: make sure we ignore any events in
++ * kernel space (but re-enable TF when returning to user mode).
++ */
++ if (condition & DR_STEP) {
++ /*
++ * We already checked v86 mode above, so we can
++ * check for kernel mode by just checking the CPL
++ * of CS.
++ */
++ if (!user_mode(regs))
++ goto clear_TF_reenable;
++ }
++
++ /* Ok, finally something we can handle */
++ send_sigtrap(tsk, regs, error_code);
++
++ /* Disable additional traps. They'll be re-enabled when
++ * the signal is delivered.
++ */
++clear_dr7:
++ set_debugreg(0, 7);
++ return;
++
++debug_vm86:
++ handle_vm86_trap((struct kernel_vm86_regs *) regs, error_code, 1);
++ return;
++
++clear_TF_reenable:
++ set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
++ regs->eflags &= ~TF_MASK;
++ return;
++}
++
++/*
++ * Note that we play around with the 'TS' bit in an attempt to get
++ * the correct behaviour even in the presence of the asynchronous
++ * IRQ13 behaviour
++ */
++void math_error(void __user *eip)
++{
++ struct task_struct * task;
++ siginfo_t info;
++ unsigned short cwd, swd;
++
++ /*
++ * Save the info for the exception handler and clear the error.
++ */
++ task = current;
++ save_init_fpu(task);
++ task->thread.trap_no = 16;
++ task->thread.error_code = 0;
++ info.si_signo = SIGFPE;
++ info.si_errno = 0;
++ info.si_code = __SI_FAULT;
++ info.si_addr = eip;
++ /*
++ * (~cwd & swd) will mask out exceptions that are not set to unmasked
++ * status. 0x3f is the exception bits in these regs, 0x200 is the
++ * C1 reg you need in case of a stack fault, 0x040 is the stack
++ * fault bit. We should only be taking one exception at a time,
++ * so if this combination doesn't produce any single exception,
++ * then we have a bad program that isn't syncronizing its FPU usage
++ * and it will suffer the consequences since we won't be able to
++ * fully reproduce the context of the exception
++ */
++ cwd = get_fpu_cwd(task);
++ swd = get_fpu_swd(task);
++ switch (swd & ~cwd & 0x3f) {
++ case 0x000: /* No unmasked exception */
++ return;
++ default: /* Multiple exceptions */
++ break;
++ case 0x001: /* Invalid Op */
++ /*
++ * swd & 0x240 == 0x040: Stack Underflow
++ * swd & 0x240 == 0x240: Stack Overflow
++ * User must clear the SF bit (0x40) if set
++ */
++ info.si_code = FPE_FLTINV;
++ break;
++ case 0x002: /* Denormalize */
++ case 0x010: /* Underflow */
++ info.si_code = FPE_FLTUND;
++ break;
++ case 0x004: /* Zero Divide */
++ info.si_code = FPE_FLTDIV;
++ break;
++ case 0x008: /* Overflow */
++ info.si_code = FPE_FLTOVF;
++ break;
++ case 0x020: /* Precision */
++ info.si_code = FPE_FLTRES;
++ break;
++ }
++ force_sig_info(SIGFPE, &info, task);
++}
++
++fastcall void do_coprocessor_error(struct pt_regs * regs, long error_code)
++{
++ ignore_fpu_irq = 1;
++ math_error((void __user *)regs->eip);
++}
++
++static void simd_math_error(void __user *eip)
++{
++ struct task_struct * task;
++ siginfo_t info;
++ unsigned short mxcsr;
++
++ /*
++ * Save the info for the exception handler and clear the error.
++ */
++ task = current;
++ save_init_fpu(task);
++ task->thread.trap_no = 19;
++ task->thread.error_code = 0;
++ info.si_signo = SIGFPE;
++ info.si_errno = 0;
++ info.si_code = __SI_FAULT;
++ info.si_addr = eip;
++ /*
++ * The SIMD FPU exceptions are handled a little differently, as there
++ * is only a single status/control register. Thus, to determine which
++ * unmasked exception was caught we must mask the exception mask bits
++ * at 0x1f80, and then use these to mask the exception bits at 0x3f.
++ */
++ mxcsr = get_fpu_mxcsr(task);
++ switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
++ case 0x000:
++ default:
++ break;
++ case 0x001: /* Invalid Op */
++ info.si_code = FPE_FLTINV;
++ break;
++ case 0x002: /* Denormalize */
++ case 0x010: /* Underflow */
++ info.si_code = FPE_FLTUND;
++ break;
++ case 0x004: /* Zero Divide */
++ info.si_code = FPE_FLTDIV;
++ break;
++ case 0x008: /* Overflow */
++ info.si_code = FPE_FLTOVF;
++ break;
++ case 0x020: /* Precision */
++ info.si_code = FPE_FLTRES;
++ break;
++ }
++ force_sig_info(SIGFPE, &info, task);
++}
++
++fastcall void do_simd_coprocessor_error(struct pt_regs * regs,
++ long error_code)
++{
++ if (cpu_has_xmm) {
++ /* Handle SIMD FPU exceptions on PIII+ processors. */
++ ignore_fpu_irq = 1;
++ simd_math_error((void __user *)regs->eip);
++ } else {
++ /*
++ * Handle strange cache flush from user space exception
++ * in all other cases. This is undocumented behaviour.
++ */
++ if (regs->eflags & VM_MASK) {
++ handle_vm86_fault((struct kernel_vm86_regs *)regs,
++ error_code);
++ return;
++ }
++ current->thread.trap_no = 19;
++ current->thread.error_code = error_code;
++ die_if_kernel("cache flush denied", regs, error_code);
++ force_sig(SIGSEGV, current);
++ }
++}
++
++#ifndef CONFIG_XEN
++fastcall void do_spurious_interrupt_bug(struct pt_regs * regs,
++ long error_code)
++{
++#if 0
++ /* No need to warn about this any longer. */
++ printk("Ignoring P6 Local APIC Spurious Interrupt Bug...\n");
++#endif
++}
++
++fastcall void setup_x86_bogus_stack(unsigned char * stk)
++{
++ unsigned long *switch16_ptr, *switch32_ptr;
++ struct pt_regs *regs;
++ unsigned long stack_top, stack_bot;
++ unsigned short iret_frame16_off;
++ int cpu = smp_processor_id();
++ /* reserve the space on 32bit stack for the magic switch16 pointer */
++ memmove(stk, stk + 8, sizeof(struct pt_regs));
++ switch16_ptr = (unsigned long *)(stk + sizeof(struct pt_regs));
++ regs = (struct pt_regs *)stk;
++ /* now the switch32 on 16bit stack */
++ stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
++ stack_top = stack_bot + CPU_16BIT_STACK_SIZE;
++ switch32_ptr = (unsigned long *)(stack_top - 8);
++ iret_frame16_off = CPU_16BIT_STACK_SIZE - 8 - 20;
++ /* copy iret frame on 16bit stack */
++ memcpy((void *)(stack_bot + iret_frame16_off), ®s->eip, 20);
++ /* fill in the switch pointers */
++ switch16_ptr[0] = (regs->esp & 0xffff0000) | iret_frame16_off;
++ switch16_ptr[1] = __ESPFIX_SS;
++ switch32_ptr[0] = (unsigned long)stk + sizeof(struct pt_regs) +
++ 8 - CPU_16BIT_STACK_SIZE;
++ switch32_ptr[1] = __KERNEL_DS;
++}
++
++fastcall unsigned char * fixup_x86_bogus_stack(unsigned short sp)
++{
++ unsigned long *switch32_ptr;
++ unsigned char *stack16, *stack32;
++ unsigned long stack_top, stack_bot;
++ int len;
++ int cpu = smp_processor_id();
++ stack_bot = (unsigned long)&per_cpu(cpu_16bit_stack, cpu);
++ stack_top = stack_bot + CPU_16BIT_STACK_SIZE;
++ switch32_ptr = (unsigned long *)(stack_top - 8);
++ /* copy the data from 16bit stack to 32bit stack */
++ len = CPU_16BIT_STACK_SIZE - 8 - sp;
++ stack16 = (unsigned char *)(stack_bot + sp);
++ stack32 = (unsigned char *)
++ (switch32_ptr[0] + CPU_16BIT_STACK_SIZE - 8 - len);
++ memcpy(stack32, stack16, len);
++ return stack32;
++}
++#endif
++
++/*
++ * 'math_state_restore()' saves the current math information in the
++ * old math state array, and gets the new ones from the current task
++ *
++ * Careful.. There are problems with IBM-designed IRQ13 behaviour.
++ * Don't touch unless you *really* know how it works.
++ *
++ * Must be called with kernel preemption disabled (in this case,
++ * local interrupts are disabled at the call-site in entry.S).
++ */
++asmlinkage void math_state_restore(struct pt_regs regs)
++{
++ struct thread_info *thread = current_thread_info();
++ struct task_struct *tsk = thread->task;
++
++ /* NB. 'clts' is done for us by Xen during virtual trap. */
++ if (!tsk_used_math(tsk))
++ init_fpu(tsk);
++ restore_fpu(tsk);
++ thread->status |= TS_USEDFPU; /* So we fnsave on switch_to() */
++}
++
++#ifndef CONFIG_MATH_EMULATION
++
++asmlinkage void math_emulate(long arg)
++{
++ printk(KERN_EMERG "math-emulation not enabled and no coprocessor found.\n");
++ printk(KERN_EMERG "killing %s.\n",current->comm);
++ force_sig(SIGFPE,current);
++ schedule();
++}
++
++#endif /* CONFIG_MATH_EMULATION */
++
++#ifdef CONFIG_X86_F00F_BUG
++void __init trap_init_f00f_bug(void)
++{
++ __set_fixmap(FIX_F00F_IDT, __pa(&idt_table), PAGE_KERNEL_RO);
++
++ /*
++ * Update the IDT descriptor and reload the IDT so that
++ * it uses the read-only mapped virtual address.
++ */
++ idt_descr.address = fix_to_virt(FIX_F00F_IDT);
++ load_idt(&idt_descr);
++}
++#endif
++
++
++/*
++ * NB. All these are "trap gates" (i.e. events_mask isn't set) except
++ * for those that specify <dpl>|4 in the second field.
++ */
++static trap_info_t trap_table[] = {
++ { 0, 0, __KERNEL_CS, (unsigned long)divide_error },
++ { 1, 0|4, __KERNEL_CS, (unsigned long)debug },
++ { 3, 3|4, __KERNEL_CS, (unsigned long)int3 },
++ { 4, 3, __KERNEL_CS, (unsigned long)overflow },
++ { 5, 0, __KERNEL_CS, (unsigned long)bounds },
++ { 6, 0, __KERNEL_CS, (unsigned long)invalid_op },
++ { 7, 0|4, __KERNEL_CS, (unsigned long)device_not_available },
++ { 9, 0, __KERNEL_CS, (unsigned long)coprocessor_segment_overrun },
++ { 10, 0, __KERNEL_CS, (unsigned long)invalid_TSS },
++ { 11, 0, __KERNEL_CS, (unsigned long)segment_not_present },
++ { 12, 0, __KERNEL_CS, (unsigned long)stack_segment },
++ { 13, 0, __KERNEL_CS, (unsigned long)general_protection },
++ { 14, 0|4, __KERNEL_CS, (unsigned long)page_fault },
++ { 15, 0, __KERNEL_CS, (unsigned long)fixup_4gb_segment },
++ { 16, 0, __KERNEL_CS, (unsigned long)coprocessor_error },
++ { 17, 0, __KERNEL_CS, (unsigned long)alignment_check },
++#ifdef CONFIG_X86_MCE
++ { 18, 0, __KERNEL_CS, (unsigned long)machine_check },
++#endif
++ { 19, 0, __KERNEL_CS, (unsigned long)simd_coprocessor_error },
++ { SYSCALL_VECTOR, 3, __KERNEL_CS, (unsigned long)system_call },
++ { 0, 0, 0, 0 }
++};
++
++void __init trap_init(void)
++{
++ HYPERVISOR_set_trap_table(trap_table);
++
++ if (cpu_has_fxsr) {
++ /*
++ * Verify that the FXSAVE/FXRSTOR data will be 16-byte aligned.
++ * Generates a compile-time "error: zero width for bit-field" if
++ * the alignment is wrong.
++ */
++ struct fxsrAlignAssert {
++ int _:!(offsetof(struct task_struct,
++ thread.i387.fxsave) & 15);
++ };
++
++ printk(KERN_INFO "Enabling fast FPU save and restore... ");
++ set_in_cr4(X86_CR4_OSFXSR);
++ printk("done.\n");
++ }
++ if (cpu_has_xmm) {
++ printk(KERN_INFO "Enabling unmasked SIMD FPU exception "
++ "support... ");
++ set_in_cr4(X86_CR4_OSXMMEXCPT);
++ printk("done.\n");
++ }
++
++ /*
++ * Should be a barrier for any external CPU state.
++ */
++ cpu_init();
++}
++
++void smp_trap_init(trap_info_t *trap_ctxt)
++{
++ trap_info_t *t = trap_table;
++
++ for (t = trap_table; t->address; t++) {
++ trap_ctxt[t->vector].flags = t->flags;
++ trap_ctxt[t->vector].cs = t->cs;
++ trap_ctxt[t->vector].address = t->address;
++ }
++}
++
++static int __init kstack_setup(char *s)
++{
++ kstack_depth_to_print = simple_strtoul(s, NULL, 0);
++ return 1;
++}
++__setup("kstack=", kstack_setup);
+diff -urNp linux-2.6/arch/i386/kernel/vm86.c new/arch/i386/kernel/vm86.c
+--- linux-2.6/arch/i386/kernel/vm86.c 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/kernel/vm86.c 2006-06-07 13:29:36.000000000 +0200
+@@ -98,7 +98,9 @@
+ struct pt_regs * FASTCALL(save_v86_state(struct kernel_vm86_regs * regs));
+ struct pt_regs * fastcall save_v86_state(struct kernel_vm86_regs * regs)
+ {
++#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *tss;
++#endif
+ struct pt_regs *ret;
+ unsigned long tmp;
+
+@@ -123,12 +125,16 @@ struct pt_regs * fastcall save_v86_state
+ do_exit(SIGSEGV);
+ }
+
++#ifndef CONFIG_X86_NO_TSS
+ tss = &per_cpu(init_tss, get_cpu());
++#endif
+ current->thread.esp0 = current->thread.saved_esp0;
+ current->thread.sysenter_cs = __KERNEL_CS;
+ load_esp0(tss, ¤t->thread);
+ current->thread.saved_esp0 = 0;
++#ifndef CONFIG_X86_NO_TSS
+ put_cpu();
++#endif
+
+ loadsegment(fs, current->thread.saved_fs);
+ loadsegment(gs, current->thread.saved_gs);
+@@ -252,7 +258,9 @@ out:
+
+ static void do_sys_vm86(struct kernel_vm86_struct *info, struct task_struct *tsk)
+ {
++#ifndef CONFIG_X86_NO_TSS
+ struct tss_struct *tss;
++#endif
+ long eax;
+ /*
+ * make sure the vm86() system call doesn't try to do anything silly
+@@ -297,12 +305,16 @@ static void do_sys_vm86(struct kernel_vm
+ savesegment(fs, tsk->thread.saved_fs);
+ savesegment(gs, tsk->thread.saved_gs);
+
++#ifndef CONFIG_X86_NO_TSS
+ tss = &per_cpu(init_tss, get_cpu());
++#endif
+ tsk->thread.esp0 = (unsigned long) &info->VM86_TSS_ESP0;
+ if (cpu_has_sep)
+ tsk->thread.sysenter_cs = 0;
+ load_esp0(tss, &tsk->thread);
++#ifndef CONFIG_X86_NO_TSS
+ put_cpu();
++#endif
+
+ tsk->thread.screen_bitmap = info->screen_bitmap;
+ if (info->flags & VM86_SCREEN_BITMAP)
+diff -urNp linux-2.6/arch/i386/kernel/vsyscall-note-xen.S new/arch/i386/kernel/vsyscall-note-xen.S
+--- linux-2.6/arch/i386/kernel/vsyscall-note-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/kernel/vsyscall-note-xen.S 2006-05-09 12:32:36.000000000 +0200
+@@ -0,0 +1,32 @@
++/*
++ * This supplies .note.* sections to go into the PT_NOTE inside the vDSO text.
++ * Here we can supply some information useful to userland.
++ * First we get the vanilla i386 note that supplies the kernel version info.
++ */
++
++#include "vsyscall-note.S"
++
++/*
++ * Now we add a special note telling glibc's dynamic linker a fake hardware
++ * flavor that it will use to choose the search path for libraries in the
++ * same way it uses real hardware capabilities like "mmx".
++ * We supply "nosegneg" as the fake capability, to indicate that we
++ * do not like negative offsets in instructions using segment overrides,
++ * since we implement those inefficiently. This makes it possible to
++ * install libraries optimized to avoid those access patterns in someplace
++ * like /lib/i686/tls/nosegneg. Note that an /etc/ld.so.conf.d/file
++ * corresponding to the bits here is needed to make ldconfig work right.
++ * It should contain:
++ * hwcap 0 nosegneg
++ * to match the mapping of bit to name that we give here.
++ */
++#define NOTE_KERNELCAP_BEGIN(ncaps, mask) \
++ ASM_ELF_NOTE_BEGIN(".note.kernelcap", "a", "GNU", 2) \
++ .long ncaps, mask
++#define NOTE_KERNELCAP(bit, name) \
++ .byte bit; .asciz name
++#define NOTE_KERNELCAP_END ASM_ELF_NOTE_END
++
++NOTE_KERNELCAP_BEGIN(1, 1)
++NOTE_KERNELCAP(1, "nosegneg") /* Change 1 back to 0 when glibc is fixed! */
++NOTE_KERNELCAP_END
+diff -urNp linux-2.6/arch/i386/mach-xen/Makefile new/arch/i386/mach-xen/Makefile
+--- linux-2.6/arch/i386/mach-xen/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mach-xen/Makefile 2006-05-09 12:32:36.000000000 +0200
+@@ -0,0 +1,5 @@
++#
++# Makefile for the linux kernel.
++#
++
++obj-y := setup.o
+diff -urNp linux-2.6/arch/i386/mach-xen/setup.c new/arch/i386/mach-xen/setup.c
+--- linux-2.6/arch/i386/mach-xen/setup.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mach-xen/setup.c 2006-05-09 12:32:36.000000000 +0200
+@@ -0,0 +1,49 @@
++/*
++ * Machine specific setup for generic
++ */
++
++#include <linux/config.h>
++#include <linux/smp.h>
++#include <linux/init.h>
++#include <linux/interrupt.h>
++#include <asm/acpi.h>
++#include <asm/arch_hooks.h>
++
++#ifdef CONFIG_HOTPLUG_CPU
++#define DEFAULT_SEND_IPI (1)
++#else
++#define DEFAULT_SEND_IPI (0)
++#endif
++
++int no_broadcast=DEFAULT_SEND_IPI;
++
++static __init int no_ipi_broadcast(char *str)
++{
++ get_option(&str, &no_broadcast);
++ printk ("Using %s mode\n", no_broadcast ? "No IPI Broadcast" :
++ "IPI Broadcast");
++ return 1;
++}
++
++__setup("no_ipi_broadcast", no_ipi_broadcast);
++
++static int __init print_ipi_mode(void)
++{
++ printk ("Using IPI %s mode\n", no_broadcast ? "No-Shortcut" :
++ "Shortcut");
++ return 0;
++}
++
++late_initcall(print_ipi_mode);
++
++/**
++ * pre_setup_arch_hook - hook called prior to any setup_arch() execution
++ *
++ * Description:
++ * generally used to activate any machine specific identification
++ * routines that may be needed before setup_arch() runs. On VISWS
++ * this is used to get the board revision and type.
++ **/
++void __init pre_setup_arch_hook(void)
++{
++}
+diff -urNp linux-2.6/arch/i386/Makefile new/arch/i386/Makefile
+--- linux-2.6/arch/i386/Makefile 2006-07-03 14:14:14.000000000 +0200
++++ new/arch/i386/Makefile 2006-05-09 12:32:33.000000000 +0200
+@@ -48,6 +48,11 @@ CFLAGS += $(shell if [ $(call cc-vers
+
+ CFLAGS += $(cflags-y)
+
++cppflags-$(CONFIG_XEN) += \
++ -D__XEN_INTERFACE_VERSION__=$(CONFIG_XEN_INTERFACE_VERSION)
++
++CPPFLAGS += $(cppflags-y)
++
+ # Default subarch .c files
+ mcore-y := mach-default
+
+@@ -71,6 +76,10 @@ mcore-$(CONFIG_X86_BIGSMP) := mach-defau
+ mflags-$(CONFIG_X86_SUMMIT) := -Iinclude/asm-i386/mach-summit
+ mcore-$(CONFIG_X86_SUMMIT) := mach-default
+
++# Xen subarch support
++mflags-$(CONFIG_X86_XEN) := -Iinclude/asm-i386/mach-xen
++mcore-$(CONFIG_X86_XEN) := mach-xen
++
+ # generic subarchitecture
+ mflags-$(CONFIG_X86_GENERICARCH) := -Iinclude/asm-i386/mach-generic
+ mcore-$(CONFIG_X86_GENERICARCH) := mach-default
+@@ -105,6 +114,19 @@ boot := arch/i386/boot
+ PHONY += zImage bzImage compressed zlilo bzlilo \
+ zdisk bzdisk fdimage fdimage144 fdimage288 isoimage install
+
++ifdef CONFIG_XEN
++CPPFLAGS := -Iinclude$(if $(KBUILD_SRC),2)/asm/mach-xen $(CPPFLAGS)
++head-y := arch/i386/kernel/head-xen.o arch/i386/kernel/init_task-xen.o
++boot := arch/i386/boot-xen
++.PHONY: vmlinuz
++all: vmlinuz
++
++vmlinuz: vmlinux
++ $(Q)$(MAKE) $(build)=$(boot) $@
++
++install:
++ $(Q)$(MAKE) $(build)=$(boot) XENGUEST=$(XENGUEST) $@
++else
+ all: bzImage
+
+ # KBUILD_IMAGE specify target image being built
+@@ -127,6 +149,7 @@ fdimage fdimage144 fdimage288 isoimage:
+
+ install:
+ $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(KBUILD_IMAGE) install
++endif
+
+ archclean:
+ $(Q)$(MAKE) $(clean)=arch/i386/boot
+@@ -145,3 +168,4 @@ endef
+ CLEAN_FILES += arch/$(ARCH)/boot/fdimage \
+ arch/$(ARCH)/boot/image.iso \
+ arch/$(ARCH)/boot/mtools.conf
++CLEAN_FILES += vmlinuz vmlinux-stripped
+diff -urNp linux-2.6/arch/i386/mm/fault-xen.c new/arch/i386/mm/fault-xen.c
+--- linux-2.6/arch/i386/mm/fault-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/fault-xen.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,734 @@
++/*
++ * linux/arch/i386/mm/fault.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ */
++
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/ptrace.h>
++#include <linux/mman.h>
++#include <linux/mm.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/interrupt.h>
++#include <linux/init.h>
++#include <linux/tty.h>
++#include <linux/vt_kern.h> /* For unblank_screen() */
++#include <linux/highmem.h>
++#include <linux/module.h>
++#include <linux/kprobes.h>
++
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/desc.h>
++#include <asm/kdebug.h>
++
++extern void die(const char *,struct pt_regs *,long);
++
++/*
++ * Unlock any spinlocks which will prevent us from getting the
++ * message out
++ */
++void bust_spinlocks(int yes)
++{
++ int loglevel_save = console_loglevel;
++
++ if (yes) {
++ oops_in_progress = 1;
++ return;
++ }
++#ifdef CONFIG_VT
++ unblank_screen();
++#endif
++ oops_in_progress = 0;
++ /*
++ * OK, the message is on the console. Now we call printk()
++ * without oops_in_progress set so that printk will give klogd
++ * a poke. Hold onto your hats...
++ */
++ console_loglevel = 15; /* NMI oopser may have shut the console up */
++ printk(" ");
++ console_loglevel = loglevel_save;
++}
++
++/*
++ * Return EIP plus the CS segment base. The segment limit is also
++ * adjusted, clamped to the kernel/user address space (whichever is
++ * appropriate), and returned in *eip_limit.
++ *
++ * The segment is checked, because it might have been changed by another
++ * task between the original faulting instruction and here.
++ *
++ * If CS is no longer a valid code segment, or if EIP is beyond the
++ * limit, or if it is a kernel address when CS is not a kernel segment,
++ * then the returned value will be greater than *eip_limit.
++ *
++ * This is slow, but is very rarely executed.
++ */
++static inline unsigned long get_segment_eip(struct pt_regs *regs,
++ unsigned long *eip_limit)
++{
++ unsigned long eip = regs->eip;
++ unsigned seg = regs->xcs & 0xffff;
++ u32 seg_ar, seg_limit, base, *desc;
++
++ /* The standard kernel/user address space limit. */
++ *eip_limit = (seg & 2) ? USER_DS.seg : KERNEL_DS.seg;
++
++ /* Unlikely, but must come before segment checks. */
++ if (unlikely((regs->eflags & VM_MASK) != 0))
++ return eip + (seg << 4);
++
++ /* By far the most common cases. */
++ if (likely(seg == __USER_CS || seg == GET_KERNEL_CS()))
++ return eip;
++
++ /* Check the segment exists, is within the current LDT/GDT size,
++ that kernel/user (ring 0..3) has the appropriate privilege,
++ that it's a code segment, and get the limit. */
++ __asm__ ("larl %3,%0; lsll %3,%1"
++ : "=&r" (seg_ar), "=r" (seg_limit) : "0" (0), "rm" (seg));
++ if ((~seg_ar & 0x9800) || eip > seg_limit) {
++ *eip_limit = 0;
++ return 1; /* So that returned eip > *eip_limit. */
++ }
++
++ /* Get the GDT/LDT descriptor base.
++ When you look for races in this code remember that
++ LDT and other horrors are only used in user space. */
++ if (seg & (1<<2)) {
++ /* Must lock the LDT while reading it. */
++ down(¤t->mm->context.sem);
++ desc = current->mm->context.ldt;
++ desc = (void *)desc + (seg & ~7);
++ } else {
++ /* Must disable preemption while reading the GDT. */
++ desc = (u32 *)get_cpu_gdt_table(get_cpu());
++ desc = (void *)desc + (seg & ~7);
++ }
++
++ /* Decode the code segment base from the descriptor */
++ base = get_desc_base((unsigned long *)desc);
++
++ if (seg & (1<<2)) {
++ up(¤t->mm->context.sem);
++ } else
++ put_cpu();
++
++ /* Adjust EIP and segment limit, and clamp at the kernel limit.
++ It's legitimate for segments to wrap at 0xffffffff. */
++ seg_limit += base;
++ if (seg_limit < *eip_limit && seg_limit >= base)
++ *eip_limit = seg_limit;
++ return eip + base;
++}
++
++/*
++ * Sometimes AMD Athlon/Opteron CPUs report invalid exceptions on prefetch.
++ * Check that here and ignore it.
++ */
++static int __is_prefetch(struct pt_regs *regs, unsigned long addr)
++{
++ unsigned long limit;
++ unsigned long instr = get_segment_eip (regs, &limit);
++ int scan_more = 1;
++ int prefetch = 0;
++ int i;
++
++ for (i = 0; scan_more && i < 15; i++) {
++ unsigned char opcode;
++ unsigned char instr_hi;
++ unsigned char instr_lo;
++
++ if (instr > limit)
++ break;
++ if (__get_user(opcode, (unsigned char __user *) instr))
++ break;
++
++ instr_hi = opcode & 0xf0;
++ instr_lo = opcode & 0x0f;
++ instr++;
++
++ switch (instr_hi) {
++ case 0x20:
++ case 0x30:
++ /* Values 0x26,0x2E,0x36,0x3E are valid x86 prefixes. */
++ scan_more = ((instr_lo & 7) == 0x6);
++ break;
++
++ case 0x60:
++ /* 0x64 thru 0x67 are valid prefixes in all modes. */
++ scan_more = (instr_lo & 0xC) == 0x4;
++ break;
++ case 0xF0:
++ /* 0xF0, 0xF2, and 0xF3 are valid prefixes */
++ scan_more = !instr_lo || (instr_lo>>1) == 1;
++ break;
++ case 0x00:
++ /* Prefetch instruction is 0x0F0D or 0x0F18 */
++ scan_more = 0;
++ if (instr > limit)
++ break;
++ if (__get_user(opcode, (unsigned char __user *) instr))
++ break;
++ prefetch = (instr_lo == 0xF) &&
++ (opcode == 0x0D || opcode == 0x18);
++ break;
++ default:
++ scan_more = 0;
++ break;
++ }
++ }
++ return prefetch;
++}
++
++static inline int is_prefetch(struct pt_regs *regs, unsigned long addr,
++ unsigned long error_code)
++{
++ if (unlikely(boot_cpu_data.x86_vendor == X86_VENDOR_AMD &&
++ boot_cpu_data.x86 >= 6)) {
++ /* Catch an obscure case of prefetch inside an NX page. */
++ if (nx_enabled && (error_code & 16))
++ return 0;
++ return __is_prefetch(regs, addr);
++ }
++ return 0;
++}
++
++static noinline void force_sig_info_fault(int si_signo, int si_code,
++ unsigned long address, struct task_struct *tsk)
++{
++ siginfo_t info;
++
++ info.si_signo = si_signo;
++ info.si_errno = 0;
++ info.si_code = si_code;
++ info.si_addr = (void __user *)address;
++ force_sig_info(si_signo, &info, tsk);
++}
++
++fastcall void do_invalid_op(struct pt_regs *, unsigned long);
++
++#ifdef CONFIG_X86_PAE
++static void dump_fault_path(unsigned long address)
++{
++ unsigned long *p, page;
++ unsigned long mfn;
++
++ page = read_cr3();
++ p = (unsigned long *)__va(page);
++ p += (address >> 30) * 2;
++ printk(KERN_ALERT "%08lx -> *pde = %08lx:%08lx\n", page, p[1], p[0]);
++ if (p[0] & 1) {
++ mfn = (p[0] >> PAGE_SHIFT) | ((p[1] & 0x7) << 20);
++ page = mfn_to_pfn(mfn) << PAGE_SHIFT;
++ p = (unsigned long *)__va(page);
++ address &= 0x3fffffff;
++ p += (address >> 21) * 2;
++ printk(KERN_ALERT "%08lx -> *pme = %08lx:%08lx\n",
++ page, p[1], p[0]);
++#ifndef CONFIG_HIGHPTE
++ if (p[0] & 1) {
++ mfn = (p[0] >> PAGE_SHIFT) | ((p[1] & 0x7) << 20);
++ page = mfn_to_pfn(mfn) << PAGE_SHIFT;
++ p = (unsigned long *) __va(page);
++ address &= 0x001fffff;
++ p += (address >> 12) * 2;
++ printk(KERN_ALERT "%08lx -> *pte = %08lx:%08lx\n",
++ page, p[1], p[0]);
++ }
++#endif
++ }
++}
++#else
++static void dump_fault_path(unsigned long address)
++{
++ unsigned long page;
++
++ page = read_cr3();
++ page = ((unsigned long *) __va(page))[address >> 22];
++ if (oops_may_print())
++ printk(KERN_ALERT "*pde = ma %08lx pa %08lx\n", page,
++ machine_to_phys(page));
++ /*
++ * We must not directly access the pte in the highpte
++ * case, the page table might be allocated in highmem.
++ * And lets rather not kmap-atomic the pte, just in case
++ * it's allocated already.
++ */
++#ifndef CONFIG_HIGHPTE
++ if ((page & 1) && oops_may_print()) {
++ page &= PAGE_MASK;
++ address &= 0x003ff000;
++ page = machine_to_phys(page);
++ page = ((unsigned long *) __va(page))[address >> PAGE_SHIFT];
++ printk(KERN_ALERT "*pte = ma %08lx pa %08lx\n", page,
++ machine_to_phys(page));
++ }
++#endif
++}
++#endif
++
++static int spurious_fault(struct pt_regs *regs,
++ unsigned long address,
++ unsigned long error_code)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++#ifdef CONFIG_XEN
++ /* Faults in hypervisor area are never spurious. */
++ if (address >= HYPERVISOR_VIRT_START)
++ return 0;
++#endif
++
++ /* Reserved-bit violation or user access to kernel space? */
++ if (error_code & 0x0c)
++ return 0;
++
++ pgd = init_mm.pgd + pgd_index(address);
++ if (!pgd_present(*pgd))
++ return 0;
++
++ pud = pud_offset(pgd, address);
++ if (!pud_present(*pud))
++ return 0;
++
++ pmd = pmd_offset(pud, address);
++ if (!pmd_present(*pmd))
++ return 0;
++
++ pte = pte_offset_kernel(pmd, address);
++ if (!pte_present(*pte))
++ return 0;
++ if ((error_code & 0x02) && !pte_write(*pte))
++ return 0;
++#ifdef CONFIG_X86_PAE
++ if ((error_code & 0x10) && (pte_val(*pte) & _PAGE_NX))
++ return 0;
++#endif
++
++ return 1;
++}
++
++static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
++{
++ unsigned index = pgd_index(address);
++ pgd_t *pgd_k;
++ pud_t *pud, *pud_k;
++ pmd_t *pmd, *pmd_k;
++
++ pgd += index;
++ pgd_k = init_mm.pgd + index;
++
++ if (!pgd_present(*pgd_k))
++ return NULL;
++
++ /*
++ * set_pgd(pgd, *pgd_k); here would be useless on PAE
++ * and redundant with the set_pmd() on non-PAE. As would
++ * set_pud.
++ */
++
++ pud = pud_offset(pgd, address);
++ pud_k = pud_offset(pgd_k, address);
++ if (!pud_present(*pud_k))
++ return NULL;
++
++ pmd = pmd_offset(pud, address);
++ pmd_k = pmd_offset(pud_k, address);
++ if (!pmd_present(*pmd_k))
++ return NULL;
++ if (!pmd_present(*pmd))
++#ifndef CONFIG_XEN
++ set_pmd(pmd, *pmd_k);
++#else
++ /*
++ * When running on Xen we must launder *pmd_k through
++ * pmd_val() to ensure that _PAGE_PRESENT is correctly set.
++ */
++ set_pmd(pmd, __pmd(pmd_val(*pmd_k)));
++#endif
++ else
++ BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
++ return pmd_k;
++}
++
++/*
++ * Handle a fault on the vmalloc or module mapping area
++ *
++ * This assumes no large pages in there.
++ */
++static inline int vmalloc_fault(unsigned long address)
++{
++ unsigned long pgd_paddr;
++ pmd_t *pmd_k;
++ pte_t *pte_k;
++ /*
++ * Synchronize this task's top level page-table
++ * with the 'reference' page table.
++ *
++ * Do _not_ use "current" here. We might be inside
++ * an interrupt in the middle of a task switch..
++ */
++ pgd_paddr = read_cr3();
++ pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
++ if (!pmd_k)
++ return -1;
++ pte_k = pte_offset_kernel(pmd_k, address);
++ if (!pte_present(*pte_k))
++ return -1;
++ return 0;
++}
++
++/*
++ * This routine handles page faults. It determines the address,
++ * and the problem, and then passes it off to one of the appropriate
++ * routines.
++ *
++ * error_code:
++ * bit 0 == 0 means no page found, 1 means protection fault
++ * bit 1 == 0 means read, 1 means write
++ * bit 2 == 0 means kernel, 1 means user-mode
++ * bit 3 == 1 means use of reserved bit detected
++ * bit 4 == 1 means fault was an instruction fetch
++ */
++fastcall void __kprobes do_page_fault(struct pt_regs *regs,
++ unsigned long error_code)
++{
++ struct task_struct *tsk;
++ struct mm_struct *mm;
++ struct vm_area_struct * vma;
++ unsigned long address;
++ int write, si_code;
++
++ /* get the address */
++ address = read_cr2();
++
++ /* Set the "privileged fault" bit to something sane. */
++ error_code &= ~4;
++ error_code |= (regs->xcs & 2) << 1;
++ if (regs->eflags & X86_EFLAGS_VM)
++ error_code |= 4;
++
++ tsk = current;
++
++ si_code = SEGV_MAPERR;
++
++ /*
++ * We fault-in kernel-space virtual memory on-demand. The
++ * 'reference' page table is init_mm.pgd.
++ *
++ * NOTE! We MUST NOT take any locks for this case. We may
++ * be in an interrupt or a critical region, and should
++ * only copy the information from the master page table,
++ * nothing more.
++ *
++ * This verifies that the fault happens in kernel space
++ * (error_code & 4) == 0, and that the fault was not a
++ * protection error (error_code & 9) == 0.
++ */
++ if (unlikely(address >= TASK_SIZE)) {
++#ifdef CONFIG_XEN
++ /* Faults in hypervisor area can never be patched up. */
++ if (address >= HYPERVISOR_VIRT_START)
++ goto bad_area_nosemaphore;
++#endif
++ if (!(error_code & 0x0000000d) && vmalloc_fault(address) >= 0)
++ return;
++ /* Can take a spurious fault if mapping changes R/O -> R/W. */
++ if (spurious_fault(regs, address, error_code))
++ return;
++ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
++ SIGSEGV) == NOTIFY_STOP)
++ return;
++ /*
++ * Don't take the mm semaphore here. If we fixup a prefetch
++ * fault we could otherwise deadlock.
++ */
++ goto bad_area_nosemaphore;
++ }
++
++ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
++ SIGSEGV) == NOTIFY_STOP)
++ return;
++
++ /* It's safe to allow irq's after cr2 has been saved and the vmalloc
++ fault has been handled. */
++ if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
++ local_irq_enable();
++
++ mm = tsk->mm;
++
++ /*
++ * If we're in an interrupt, have no user context or are running in an
++ * atomic region then we must not take the fault..
++ */
++ if (in_atomic() || !mm)
++ goto bad_area_nosemaphore;
++
++ /* When running in the kernel we expect faults to occur only to
++ * addresses in user space. All other faults represent errors in the
++ * kernel and should generate an OOPS. Unfortunatly, in the case of an
++ * erroneous fault occuring in a code path which already holds mmap_sem
++ * we will deadlock attempting to validate the fault against the
++ * address space. Luckily the kernel only validly references user
++ * space from well defined areas of code, which are listed in the
++ * exceptions table.
++ *
++ * As the vast majority of faults will be valid we will only perform
++ * the source reference check when there is a possibilty of a deadlock.
++ * Attempt to lock the address space, if we cannot we then validate the
++ * source. If this is invalid we can skip the address space check,
++ * thus avoiding the deadlock.
++ */
++ if (!down_read_trylock(&mm->mmap_sem)) {
++ if ((error_code & 4) == 0 &&
++ !search_exception_tables(regs->eip))
++ goto bad_area_nosemaphore;
++ down_read(&mm->mmap_sem);
++ }
++
++ vma = find_vma(mm, address);
++ if (!vma)
++ goto bad_area;
++ if (vma->vm_start <= address)
++ goto good_area;
++ if (!(vma->vm_flags & VM_GROWSDOWN))
++ goto bad_area;
++ if (error_code & 4) {
++ /*
++ * accessing the stack below %esp is always a bug.
++ * The "+ 32" is there due to some instructions (like
++ * pusha) doing post-decrement on the stack and that
++ * doesn't show up until later..
++ */
++ if (address + 32 < regs->esp)
++ goto bad_area;
++ }
++ if (expand_stack(vma, address))
++ goto bad_area;
++/*
++ * Ok, we have a good vm_area for this memory access, so
++ * we can handle it..
++ */
++good_area:
++ si_code = SEGV_ACCERR;
++ write = 0;
++ switch (error_code & 3) {
++ default: /* 3: write, present */
++#ifdef TEST_VERIFY_AREA
++ if (regs->cs == GET_KERNEL_CS())
++ printk("WP fault at %08lx\n", regs->eip);
++#endif
++ /* fall through */
++ case 2: /* write, not present */
++ if (!(vma->vm_flags & VM_WRITE))
++ goto bad_area;
++ write++;
++ break;
++ case 1: /* read, present */
++ goto bad_area;
++ case 0: /* read, not present */
++ if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
++ goto bad_area;
++ }
++
++ survive:
++ /*
++ * If for any reason at all we couldn't handle the fault,
++ * make sure we exit gracefully rather than endlessly redo
++ * the fault.
++ */
++ switch (handle_mm_fault(mm, vma, address, write)) {
++ case VM_FAULT_MINOR:
++ tsk->min_flt++;
++ break;
++ case VM_FAULT_MAJOR:
++ tsk->maj_flt++;
++ break;
++ case VM_FAULT_SIGBUS:
++ goto do_sigbus;
++ case VM_FAULT_OOM:
++ goto out_of_memory;
++ default:
++ BUG();
++ }
++
++ /*
++ * Did it hit the DOS screen memory VA from vm86 mode?
++ */
++ if (regs->eflags & VM_MASK) {
++ unsigned long bit = (address - 0xA0000) >> PAGE_SHIFT;
++ if (bit < 32)
++ tsk->thread.screen_bitmap |= 1 << bit;
++ }
++ up_read(&mm->mmap_sem);
++ return;
++
++/*
++ * Something tried to access memory that isn't in our memory map..
++ * Fix it, but check if it's kernel or user first..
++ */
++bad_area:
++ up_read(&mm->mmap_sem);
++
++bad_area_nosemaphore:
++ /* User mode accesses just cause a SIGSEGV */
++ if (error_code & 4) {
++ /*
++ * Valid to do another page fault here because this one came
++ * from user space.
++ */
++ if (is_prefetch(regs, address, error_code))
++ return;
++
++ tsk->thread.cr2 = address;
++ /* Kernel addresses are always protection faults */
++ tsk->thread.error_code = error_code | (address >= TASK_SIZE);
++ tsk->thread.trap_no = 14;
++ force_sig_info_fault(SIGSEGV, si_code, address, tsk);
++ return;
++ }
++
++#ifdef CONFIG_X86_F00F_BUG
++ /*
++ * Pentium F0 0F C7 C8 bug workaround.
++ */
++ if (boot_cpu_data.f00f_bug) {
++ unsigned long nr;
++
++ nr = (address - idt_descr.address) >> 3;
++
++ if (nr == 6) {
++ do_invalid_op(regs, 0);
++ return;
++ }
++ }
++#endif
++
++no_context:
++ /* Are we prepared to handle this kernel fault? */
++ if (fixup_exception(regs))
++ return;
++
++ /*
++ * Valid to do another page fault here, because if this fault
++ * had been triggered by is_prefetch fixup_exception would have
++ * handled it.
++ */
++ if (is_prefetch(regs, address, error_code))
++ return;
++
++/*
++ * Oops. The kernel tried to access some bad page. We'll have to
++ * terminate things with extreme prejudice.
++ */
++
++ bust_spinlocks(1);
++
++ if (oops_may_print()) {
++ #ifdef CONFIG_X86_PAE
++ if (error_code & 16) {
++ pte_t *pte = lookup_address(address);
++
++ if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
++ printk(KERN_CRIT "kernel tried to execute "
++ "NX-protected page - exploit attempt? "
++ "(uid: %d)\n", current->uid);
++ }
++ #endif
++ if (address < PAGE_SIZE)
++ printk(KERN_ALERT "BUG: unable to handle kernel NULL "
++ "pointer dereference");
++ else
++ printk(KERN_ALERT "BUG: unable to handle kernel paging"
++ " request");
++ printk(" at virtual address %08lx\n",address);
++ printk(KERN_ALERT " printing eip:\n");
++ printk("%08lx\n", regs->eip);
++ dump_fault_path(address);
++ }
++ tsk->thread.cr2 = address;
++ tsk->thread.trap_no = 14;
++ tsk->thread.error_code = error_code;
++ die("Oops", regs, error_code);
++ bust_spinlocks(0);
++ do_exit(SIGKILL);
++
++/*
++ * We ran out of memory, or some other thing happened to us that made
++ * us unable to handle the page fault gracefully.
++ */
++out_of_memory:
++ up_read(&mm->mmap_sem);
++ if (tsk->pid == 1) {
++ yield();
++ down_read(&mm->mmap_sem);
++ goto survive;
++ }
++ printk("VM: killing process %s\n", tsk->comm);
++ if (error_code & 4)
++ do_exit(SIGKILL);
++ goto no_context;
++
++do_sigbus:
++ up_read(&mm->mmap_sem);
++
++ /* Kernel mode? Handle exceptions or die */
++ if (!(error_code & 4))
++ goto no_context;
++
++ /* User space => ok to do another page fault */
++ if (is_prefetch(regs, address, error_code))
++ return;
++
++ tsk->thread.cr2 = address;
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = 14;
++ force_sig_info_fault(SIGBUS, BUS_ADRERR, address, tsk);
++}
++
++#ifndef CONFIG_X86_PAE
++void vmalloc_sync_all(void)
++{
++ /*
++ * Note that races in the updates of insync and start aren't
++ * problematic: insync can only get set bits added, and updates to
++ * start are only improving performance (without affecting correctness
++ * if undone).
++ */
++ static DECLARE_BITMAP(insync, PTRS_PER_PGD);
++ static unsigned long start = TASK_SIZE;
++ unsigned long address;
++
++ BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK);
++ for (address = start; address >= TASK_SIZE; address += PGDIR_SIZE) {
++ if (!test_bit(pgd_index(address), insync)) {
++ unsigned long flags;
++ struct page *page;
++
++ spin_lock_irqsave(&pgd_lock, flags);
++ for (page = pgd_list; page; page =
++ (struct page *)page->index)
++ if (!vmalloc_sync_one(page_address(page),
++ address)) {
++ BUG_ON(page != pgd_list);
++ break;
++ }
++ spin_unlock_irqrestore(&pgd_lock, flags);
++ if (!page)
++ set_bit(pgd_index(address), insync);
++ }
++ if (address == start && test_bit(pgd_index(address), insync))
++ start = address + PGDIR_SIZE;
++ }
++}
++#endif
+diff -urNp linux-2.6/arch/i386/mm/highmem-xen.c new/arch/i386/mm/highmem-xen.c
+--- linux-2.6/arch/i386/mm/highmem-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/highmem-xen.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,133 @@
++#include <linux/highmem.h>
++#include <linux/module.h>
++
++void *kmap(struct page *page)
++{
++ might_sleep();
++ if (!PageHighMem(page))
++ return page_address(page);
++ return kmap_high(page);
++}
++
++void kunmap(struct page *page)
++{
++ if (in_interrupt())
++ BUG();
++ if (!PageHighMem(page))
++ return;
++ kunmap_high(page);
++}
++
++/*
++ * kmap_atomic/kunmap_atomic is significantly faster than kmap/kunmap because
++ * no global lock is needed and because the kmap code must perform a global TLB
++ * invalidation when the kmap pool wraps.
++ *
++ * However when holding an atomic kmap is is not legal to sleep, so atomic
++ * kmaps are appropriate for short, tight code paths only.
++ */
++static void *__kmap_atomic(struct page *page, enum km_type type, pgprot_t prot)
++{
++ enum fixed_addresses idx;
++ unsigned long vaddr;
++
++ /* even !CONFIG_PREEMPT needs this, for in_atomic in do_page_fault */
++ inc_preempt_count();
++ if (!PageHighMem(page))
++ return page_address(page);
++
++ idx = type + KM_TYPE_NR*smp_processor_id();
++ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
++#ifdef CONFIG_DEBUG_HIGHMEM
++ if (!pte_none(*(kmap_pte-idx)))
++ BUG();
++#endif
++ set_pte_at_sync(&init_mm, vaddr, kmap_pte-idx, mk_pte(page, prot));
++
++ return (void*) vaddr;
++}
++
++void *kmap_atomic(struct page *page, enum km_type type)
++{
++ return __kmap_atomic(page, type, kmap_prot);
++}
++
++/* Same as kmap_atomic but with PAGE_KERNEL_RO page protection. */
++void *kmap_atomic_pte(struct page *page, enum km_type type)
++{
++ return __kmap_atomic(page, type, PAGE_KERNEL_RO);
++}
++
++void kunmap_atomic(void *kvaddr, enum km_type type)
++{
++#if defined(CONFIG_DEBUG_HIGHMEM) || defined(CONFIG_XEN)
++ unsigned long vaddr = (unsigned long) kvaddr & PAGE_MASK;
++ enum fixed_addresses idx = type + KM_TYPE_NR*smp_processor_id();
++
++ if (vaddr < FIXADDR_START) { // FIXME
++ dec_preempt_count();
++ preempt_check_resched();
++ return;
++ }
++#endif
++
++#if defined(CONFIG_DEBUG_HIGHMEM)
++ if (vaddr != __fix_to_virt(FIX_KMAP_BEGIN+idx))
++ BUG();
++
++ /*
++ * force other mappings to Oops if they'll try to access
++ * this pte without first remap it
++ */
++ pte_clear(&init_mm, vaddr, kmap_pte-idx);
++ __flush_tlb_one(vaddr);
++#elif defined(CONFIG_XEN)
++ /*
++ * We must ensure there are no dangling pagetable references when
++ * returning memory to Xen (decrease_reservation).
++ * XXX TODO: We could make this faster by only zapping when
++ * kmap_flush_unused is called but that is trickier and more invasive.
++ */
++ pte_clear(&init_mm, vaddr, kmap_pte-idx);
++#endif
++
++ dec_preempt_count();
++ preempt_check_resched();
++}
++
++/* This is the same as kmap_atomic() but can map memory that doesn't
++ * have a struct page associated with it.
++ */
++void *kmap_atomic_pfn(unsigned long pfn, enum km_type type)
++{
++ enum fixed_addresses idx;
++ unsigned long vaddr;
++
++ inc_preempt_count();
++
++ idx = type + KM_TYPE_NR*smp_processor_id();
++ vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
++ set_pte(kmap_pte-idx, pfn_pte(pfn, kmap_prot));
++ __flush_tlb_one(vaddr);
++
++ return (void*) vaddr;
++}
++
++struct page *kmap_atomic_to_page(void *ptr)
++{
++ unsigned long idx, vaddr = (unsigned long)ptr;
++ pte_t *pte;
++
++ if (vaddr < FIXADDR_START)
++ return virt_to_page(ptr);
++
++ idx = virt_to_fix(vaddr);
++ pte = kmap_pte - (idx - FIX_KMAP_BEGIN);
++ return pte_page(*pte);
++}
++
++EXPORT_SYMBOL(kmap);
++EXPORT_SYMBOL(kunmap);
++EXPORT_SYMBOL(kmap_atomic);
++EXPORT_SYMBOL(kunmap_atomic);
++EXPORT_SYMBOL(kmap_atomic_to_page);
+diff -urNp linux-2.6/arch/i386/mm/hypervisor.c new/arch/i386/mm/hypervisor.c
+--- linux-2.6/arch/i386/mm/hypervisor.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/hypervisor.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,453 @@
++/******************************************************************************
++ * mm/hypervisor.c
++ *
++ * Update page tables via the hypervisor.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/sched.h>
++#include <linux/mm.h>
++#include <linux/vmalloc.h>
++#include <asm/page.h>
++#include <asm/pgtable.h>
++#include <asm/hypervisor.h>
++#include <xen/balloon.h>
++#include <xen/features.h>
++#include <xen/interface/memory.h>
++#include <linux/module.h>
++#include <linux/percpu.h>
++#include <asm/tlbflush.h>
++
++#ifdef CONFIG_X86_64
++#define pmd_val_ma(v) (v).pmd
++#else
++#ifdef CONFIG_X86_PAE
++# define pmd_val_ma(v) ((v).pmd)
++# define pud_val_ma(v) ((v).pgd.pgd)
++#else
++# define pmd_val_ma(v) ((v).pud.pgd.pgd)
++#endif
++#endif
++
++void xen_l1_entry_update(pte_t *ptr, pte_t val)
++{
++ mmu_update_t u;
++ u.ptr = virt_to_machine(ptr);
++ u.val = pte_val_ma(val);
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_l2_entry_update(pmd_t *ptr, pmd_t val)
++{
++ mmu_update_t u;
++ u.ptr = virt_to_machine(ptr);
++ u.val = pmd_val_ma(val);
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++
++#ifdef CONFIG_X86_PAE
++void xen_l3_entry_update(pud_t *ptr, pud_t val)
++{
++ mmu_update_t u;
++ u.ptr = virt_to_machine(ptr);
++ u.val = pud_val_ma(val);
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++#endif
++
++#ifdef CONFIG_X86_64
++void xen_l3_entry_update(pud_t *ptr, pud_t val)
++{
++ mmu_update_t u;
++ u.ptr = virt_to_machine(ptr);
++ u.val = val.pud;
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_l4_entry_update(pgd_t *ptr, pgd_t val)
++{
++ mmu_update_t u;
++ u.ptr = virt_to_machine(ptr);
++ u.val = val.pgd;
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++#endif /* CONFIG_X86_64 */
++
++void xen_machphys_update(unsigned long mfn, unsigned long pfn)
++{
++ mmu_update_t u;
++ if (xen_feature(XENFEAT_auto_translated_physmap)) {
++ BUG_ON(pfn != mfn);
++ return;
++ }
++ u.ptr = ((unsigned long long)mfn << PAGE_SHIFT) | MMU_MACHPHYS_UPDATE;
++ u.val = pfn;
++ BUG_ON(HYPERVISOR_mmu_update(&u, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_pt_switch(unsigned long ptr)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_NEW_BASEPTR;
++ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_new_user_pt(unsigned long ptr)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_NEW_USER_BASEPTR;
++ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_tlb_flush(void)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_TLB_FLUSH_LOCAL;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++EXPORT_SYMBOL(xen_tlb_flush);
++
++void xen_invlpg(unsigned long ptr)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_INVLPG_LOCAL;
++ op.arg1.linear_addr = ptr & PAGE_MASK;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++EXPORT_SYMBOL(xen_invlpg);
++
++#ifdef CONFIG_SMP
++
++void xen_tlb_flush_all(void)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_TLB_FLUSH_ALL;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_tlb_flush_mask(cpumask_t *mask)
++{
++ struct mmuext_op op;
++ if ( cpus_empty(*mask) )
++ return;
++ op.cmd = MMUEXT_TLB_FLUSH_MULTI;
++ op.arg2.vcpumask = mask->bits;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_invlpg_all(unsigned long ptr)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_INVLPG_ALL;
++ op.arg1.linear_addr = ptr & PAGE_MASK;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_invlpg_mask(cpumask_t *mask, unsigned long ptr)
++{
++ struct mmuext_op op;
++ if ( cpus_empty(*mask) )
++ return;
++ op.cmd = MMUEXT_INVLPG_MULTI;
++ op.arg1.linear_addr = ptr & PAGE_MASK;
++ op.arg2.vcpumask = mask->bits;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++#endif /* CONFIG_SMP */
++
++void xen_pgd_pin(unsigned long ptr)
++{
++ struct mmuext_op op;
++#ifdef CONFIG_X86_64
++ op.cmd = MMUEXT_PIN_L4_TABLE;
++#elif defined(CONFIG_X86_PAE)
++ op.cmd = MMUEXT_PIN_L3_TABLE;
++#else
++ op.cmd = MMUEXT_PIN_L2_TABLE;
++#endif
++ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_pgd_unpin(unsigned long ptr)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_UNPIN_TABLE;
++ op.arg1.mfn = pfn_to_mfn(ptr >> PAGE_SHIFT);
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++void xen_set_ldt(unsigned long ptr, unsigned long len)
++{
++ struct mmuext_op op;
++ op.cmd = MMUEXT_SET_LDT;
++ op.arg1.linear_addr = ptr;
++ op.arg2.nr_ents = len;
++ BUG_ON(HYPERVISOR_mmuext_op(&op, 1, NULL, DOMID_SELF) < 0);
++}
++
++/*
++ * Bitmap is indexed by page number. If bit is set, the page is part of a
++ * xen_create_contiguous_region() area of memory.
++ */
++unsigned long *contiguous_bitmap;
++
++static void contiguous_bitmap_set(
++ unsigned long first_page, unsigned long nr_pages)
++{
++ unsigned long start_off, end_off, curr_idx, end_idx;
++
++ curr_idx = first_page / BITS_PER_LONG;
++ start_off = first_page & (BITS_PER_LONG-1);
++ end_idx = (first_page + nr_pages) / BITS_PER_LONG;
++ end_off = (first_page + nr_pages) & (BITS_PER_LONG-1);
++
++ if (curr_idx == end_idx) {
++ contiguous_bitmap[curr_idx] |=
++ ((1UL<<end_off)-1) & -(1UL<<start_off);
++ } else {
++ contiguous_bitmap[curr_idx] |= -(1UL<<start_off);
++ while ( ++curr_idx < end_idx )
++ contiguous_bitmap[curr_idx] = ~0UL;
++ contiguous_bitmap[curr_idx] |= (1UL<<end_off)-1;
++ }
++}
++
++static void contiguous_bitmap_clear(
++ unsigned long first_page, unsigned long nr_pages)
++{
++ unsigned long start_off, end_off, curr_idx, end_idx;
++
++ curr_idx = first_page / BITS_PER_LONG;
++ start_off = first_page & (BITS_PER_LONG-1);
++ end_idx = (first_page + nr_pages) / BITS_PER_LONG;
++ end_off = (first_page + nr_pages) & (BITS_PER_LONG-1);
++
++ if (curr_idx == end_idx) {
++ contiguous_bitmap[curr_idx] &=
++ -(1UL<<end_off) | ((1UL<<start_off)-1);
++ } else {
++ contiguous_bitmap[curr_idx] &= (1UL<<start_off)-1;
++ while ( ++curr_idx != end_idx )
++ contiguous_bitmap[curr_idx] = 0;
++ contiguous_bitmap[curr_idx] &= -(1UL<<end_off);
++ }
++}
++
++/* Protected by balloon_lock. */
++#define MAX_CONTIG_ORDER 9 /* 2MB */
++static unsigned long discontig_frames[1<<MAX_CONTIG_ORDER];
++
++/* Ensure multi-page extents are contiguous in machine memory. */
++int xen_create_contiguous_region(
++ unsigned long vstart, unsigned int order, unsigned int address_bits)
++{
++ unsigned long *in_frames = discontig_frames, out_frame;
++ unsigned long frame, i, flags;
++ long rc;
++ int success;
++ struct xen_memory_exchange exchange = {
++ .in = {
++ .nr_extents = 1UL << order,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ },
++ .out = {
++ .nr_extents = 1,
++ .extent_order = order,
++ .address_bits = address_bits,
++ .domid = DOMID_SELF
++ }
++ };
++
++ /*
++ * Currently an auto-translated guest will not perform I/O, nor will
++ * it require PAE page directories below 4GB. Therefore any calls to
++ * this function are redundant and can be ignored.
++ */
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return 0;
++
++ if (unlikely(order > MAX_CONTIG_ORDER))
++ return -ENOMEM;
++
++ set_xen_guest_handle(exchange.in.extent_start, in_frames);
++ set_xen_guest_handle(exchange.out.extent_start, &out_frame);
++
++ scrub_pages(vstart, 1 << order);
++
++ balloon_lock(flags);
++
++ /* 1. Zap current PTEs, remembering MFNs. */
++ for (i = 0; i < (1UL<<order); i++) {
++ in_frames[i] = pfn_to_mfn((__pa(vstart) >> PAGE_SHIFT) + i);
++ if (HYPERVISOR_update_va_mapping(vstart + (i*PAGE_SIZE),
++ __pte_ma(0), 0))
++ BUG();
++ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i,
++ INVALID_P2M_ENTRY);
++ }
++
++ /* 2. Get a new contiguous memory extent. */
++ out_frame = __pa(vstart) >> PAGE_SHIFT;
++ rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
++ success = (exchange.nr_exchanged == (1UL << order));
++ BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
++ BUG_ON(success && (rc != 0));
++ if (unlikely(rc == -ENOSYS)) {
++ /* Compatibility when XENMEM_exchange is unsupported. */
++ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
++ &exchange.in) != (1UL << order))
++ BUG();
++ success = (HYPERVISOR_memory_op(XENMEM_populate_physmap,
++ &exchange.out) == 1);
++ if (!success) {
++ /* Couldn't get special memory: fall back to normal. */
++ for (i = 0; i < (1UL<<order); i++)
++ in_frames[i] = (__pa(vstart)>>PAGE_SHIFT) + i;
++ if (HYPERVISOR_memory_op(XENMEM_populate_physmap,
++ &exchange.in) != (1UL<<order))
++ BUG();
++ }
++ }
++
++ /* 3. Map the new extent in place of old pages. */
++ for (i = 0; i < (1UL<<order); i++) {
++ frame = success ? (out_frame + i) : in_frames[i];
++ if (HYPERVISOR_update_va_mapping(vstart + (i*PAGE_SIZE),
++ pfn_pte_ma(frame,
++ PAGE_KERNEL),
++ 0))
++ BUG();
++ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i, frame);
++ }
++
++ flush_tlb_all();
++
++ if (success)
++ contiguous_bitmap_set(__pa(vstart) >> PAGE_SHIFT,
++ 1UL << order);
++
++ balloon_unlock(flags);
++
++ return success ? 0 : -ENOMEM;
++}
++
++void xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
++{
++ unsigned long *out_frames = discontig_frames, in_frame;
++ unsigned long frame, i, flags;
++ long rc;
++ int success;
++ struct xen_memory_exchange exchange = {
++ .in = {
++ .nr_extents = 1,
++ .extent_order = order,
++ .domid = DOMID_SELF
++ },
++ .out = {
++ .nr_extents = 1UL << order,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ }
++ };
++
++ if (xen_feature(XENFEAT_auto_translated_physmap) ||
++ !test_bit(__pa(vstart) >> PAGE_SHIFT, contiguous_bitmap))
++ return;
++
++ if (unlikely(order > MAX_CONTIG_ORDER))
++ return;
++
++ set_xen_guest_handle(exchange.in.extent_start, &in_frame);
++ set_xen_guest_handle(exchange.out.extent_start, out_frames);
++
++ scrub_pages(vstart, 1 << order);
++
++ balloon_lock(flags);
++
++ contiguous_bitmap_clear(__pa(vstart) >> PAGE_SHIFT, 1UL << order);
++
++ /* 1. Find start MFN of contiguous extent. */
++ in_frame = pfn_to_mfn(__pa(vstart) >> PAGE_SHIFT);
++
++ /* 2. Zap current PTEs. */
++ for (i = 0; i < (1UL<<order); i++) {
++ if (HYPERVISOR_update_va_mapping(vstart + (i*PAGE_SIZE),
++ __pte_ma(0), 0))
++ BUG();
++ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i,
++ INVALID_P2M_ENTRY);
++ out_frames[i] = (__pa(vstart) >> PAGE_SHIFT) + i;
++ }
++
++ /* 3. Do the exchange for non-contiguous MFNs. */
++ rc = HYPERVISOR_memory_op(XENMEM_exchange, &exchange);
++ success = (exchange.nr_exchanged == 1);
++ BUG_ON(!success && ((exchange.nr_exchanged != 0) || (rc == 0)));
++ BUG_ON(success && (rc != 0));
++ if (unlikely(rc == -ENOSYS)) {
++ /* Compatibility when XENMEM_exchange is unsupported. */
++ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
++ &exchange.in) != 1)
++ BUG();
++ if (HYPERVISOR_memory_op(XENMEM_populate_physmap,
++ &exchange.out) != (1UL << order))
++ BUG();
++ success = 1;
++ }
++
++ /* 4. Map new pages in place of old pages. */
++ for (i = 0; i < (1UL<<order); i++) {
++ frame = success ? out_frames[i] : (in_frame + i);
++ if (HYPERVISOR_update_va_mapping(vstart + (i*PAGE_SIZE),
++ pfn_pte_ma(frame,
++ PAGE_KERNEL),
++ 0))
++ BUG();
++ set_phys_to_machine((__pa(vstart)>>PAGE_SHIFT)+i, frame);
++ }
++
++ flush_tlb_all();
++
++ balloon_unlock(flags);
++}
++
++#ifdef __i386__
++int write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b)
++{
++ __u32 *lp = (__u32 *)((char *)ldt + entry * 8);
++ maddr_t mach_lp = arbitrary_virt_to_machine(lp);
++ return HYPERVISOR_update_descriptor(
++ mach_lp, (u64)entry_a | ((u64)entry_b<<32));
++}
++#endif
+diff -urNp linux-2.6/arch/i386/mm/init-xen.c new/arch/i386/mm/init-xen.c
+--- linux-2.6/arch/i386/mm/init-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/init-xen.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,850 @@
++/*
++ * linux/arch/i386/mm/init.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ *
++ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/ptrace.h>
++#include <linux/mman.h>
++#include <linux/mm.h>
++#include <linux/hugetlb.h>
++#include <linux/swap.h>
++#include <linux/smp.h>
++#include <linux/init.h>
++#include <linux/highmem.h>
++#include <linux/pagemap.h>
++#include <linux/bootmem.h>
++#include <linux/slab.h>
++#include <linux/proc_fs.h>
++#include <linux/efi.h>
++#include <linux/memory_hotplug.h>
++#include <linux/initrd.h>
++#include <linux/dma-mapping.h>
++#include <linux/scatterlist.h>
++
++#include <asm/processor.h>
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/pgtable.h>
++#include <asm/dma.h>
++#include <asm/fixmap.h>
++#include <asm/e820.h>
++#include <asm/apic.h>
++#include <asm/tlb.h>
++#include <asm/tlbflush.h>
++#include <asm/sections.h>
++#include <asm/hypervisor.h>
++#include <asm/swiotlb.h>
++
++extern unsigned long *contiguous_bitmap;
++
++unsigned int __VMALLOC_RESERVE = 128 << 20;
++
++DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
++unsigned long highstart_pfn, highend_pfn;
++
++static int noinline do_test_wp_bit(void);
++
++/*
++ * Creates a middle page table and puts a pointer to it in the
++ * given global directory entry. This only returns the gd entry
++ * in non-PAE compilation mode, since the middle layer is folded.
++ */
++static pmd_t * __init one_md_table_init(pgd_t *pgd)
++{
++ pud_t *pud;
++ pmd_t *pmd_table;
++
++#ifdef CONFIG_X86_PAE
++ pmd_table = (pmd_t *) alloc_bootmem_low_pages(PAGE_SIZE);
++ make_lowmem_page_readonly(pmd_table, XENFEAT_writable_page_tables);
++ set_pgd(pgd, __pgd(__pa(pmd_table) | _PAGE_PRESENT));
++ pud = pud_offset(pgd, 0);
++ if (pmd_table != pmd_offset(pud, 0))
++ BUG();
++#else
++ pud = pud_offset(pgd, 0);
++ pmd_table = pmd_offset(pud, 0);
++#endif
++
++ return pmd_table;
++}
++
++/*
++ * Create a page table and place a pointer to it in a middle page
++ * directory entry.
++ */
++static pte_t * __init one_page_table_init(pmd_t *pmd)
++{
++ if (pmd_none(*pmd)) {
++ pte_t *page_table = (pte_t *) alloc_bootmem_low_pages(PAGE_SIZE);
++ make_lowmem_page_readonly(page_table,
++ XENFEAT_writable_page_tables);
++ set_pmd(pmd, __pmd(__pa(page_table) | _PAGE_TABLE));
++ if (page_table != pte_offset_kernel(pmd, 0))
++ BUG();
++
++ return page_table;
++ }
++
++ return pte_offset_kernel(pmd, 0);
++}
++
++/*
++ * This function initializes a certain range of kernel virtual memory
++ * with new bootmem page tables, everywhere page tables are missing in
++ * the given range.
++ */
++
++/*
++ * NOTE: The pagetables are allocated contiguous on the physical space
++ * so we can cache the place of the first one and move around without
++ * checking the pgd every time.
++ */
++static void __init page_table_range_init (unsigned long start, unsigned long end, pgd_t *pgd_base)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ int pgd_idx, pmd_idx;
++ unsigned long vaddr;
++
++ vaddr = start;
++ pgd_idx = pgd_index(vaddr);
++ pmd_idx = pmd_index(vaddr);
++ pgd = pgd_base + pgd_idx;
++
++ for ( ; (pgd_idx < PTRS_PER_PGD) && (vaddr != end); pgd++, pgd_idx++) {
++ if (pgd_none(*pgd))
++ one_md_table_init(pgd);
++ pud = pud_offset(pgd, vaddr);
++ pmd = pmd_offset(pud, vaddr);
++ for (; (pmd_idx < PTRS_PER_PMD) && (vaddr != end); pmd++, pmd_idx++) {
++ if (vaddr < HYPERVISOR_VIRT_START && pmd_none(*pmd))
++ one_page_table_init(pmd);
++
++ vaddr += PMD_SIZE;
++ }
++ pmd_idx = 0;
++ }
++}
++
++static inline int is_kernel_text(unsigned long addr)
++{
++ if (addr >= PAGE_OFFSET && addr <= (unsigned long)__init_end)
++ return 1;
++ return 0;
++}
++
++/*
++ * This maps the physical memory to kernel virtual address space, a total
++ * of max_low_pfn pages, by creating page tables starting from address
++ * PAGE_OFFSET.
++ */
++static void __init kernel_physical_mapping_init(pgd_t *pgd_base)
++{
++ unsigned long pfn;
++ pgd_t *pgd;
++ pmd_t *pmd;
++ pte_t *pte;
++ int pgd_idx, pmd_idx, pte_ofs;
++
++ unsigned long max_ram_pfn = xen_start_info->nr_pages;
++ if (max_ram_pfn > max_low_pfn)
++ max_ram_pfn = max_low_pfn;
++
++ pgd_idx = pgd_index(PAGE_OFFSET);
++ pgd = pgd_base + pgd_idx;
++ pfn = 0;
++ pmd_idx = pmd_index(PAGE_OFFSET);
++ pte_ofs = pte_index(PAGE_OFFSET);
++
++ for (; pgd_idx < PTRS_PER_PGD; pgd++, pgd_idx++) {
++#ifdef CONFIG_XEN
++ /*
++ * Native linux hasn't PAE-paging enabled yet at this
++ * point. When running as xen domain we are in PAE
++ * mode already, thus we can't simply hook a empty
++ * pmd. That would kill the mappings we are currently
++ * using ...
++ */
++ pmd = pmd_offset(pud_offset(pgd, PAGE_OFFSET), PAGE_OFFSET);
++#else
++ pmd = one_md_table_init(pgd);
++#endif
++ if (pfn >= max_low_pfn)
++ continue;
++ pmd += pmd_idx;
++ for (; pmd_idx < PTRS_PER_PMD && pfn < max_low_pfn; pmd++, pmd_idx++) {
++ unsigned int address = pfn * PAGE_SIZE + PAGE_OFFSET;
++ if (address >= HYPERVISOR_VIRT_START)
++ continue;
++
++ /* Map with big pages if possible, otherwise create normal page tables. */
++ if (cpu_has_pse) {
++ unsigned int address2 = (pfn + PTRS_PER_PTE - 1) * PAGE_SIZE + PAGE_OFFSET + PAGE_SIZE-1;
++
++ if (is_kernel_text(address) || is_kernel_text(address2))
++ set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE_EXEC));
++ else
++ set_pmd(pmd, pfn_pmd(pfn, PAGE_KERNEL_LARGE));
++ pfn += PTRS_PER_PTE;
++ } else {
++ pte = one_page_table_init(pmd);
++
++ pte += pte_ofs;
++ for (; pte_ofs < PTRS_PER_PTE && pfn < max_low_pfn; pte++, pfn++, pte_ofs++) {
++ /* XEN: Only map initial RAM allocation. */
++ if ((pfn >= max_ram_pfn) || pte_present(*pte))
++ continue;
++ if (is_kernel_text(address))
++ set_pte(pte, pfn_pte(pfn, PAGE_KERNEL_EXEC));
++ else
++ set_pte(pte, pfn_pte(pfn, PAGE_KERNEL));
++ }
++ pte_ofs = 0;
++ }
++ }
++ pmd_idx = 0;
++ }
++}
++
++#ifndef CONFIG_XEN
++
++static inline int page_kills_ppro(unsigned long pagenr)
++{
++ if (pagenr >= 0x70000 && pagenr <= 0x7003F)
++ return 1;
++ return 0;
++}
++
++#else
++
++#define page_kills_ppro(p) 0
++
++#endif
++
++extern int is_available_memory(efi_memory_desc_t *);
++
++int page_is_ram(unsigned long pagenr)
++{
++ int i;
++ unsigned long addr, end;
++
++ if (efi_enabled) {
++ efi_memory_desc_t *md;
++ void *p;
++
++ for (p = memmap.map; p < memmap.map_end; p += memmap.desc_size) {
++ md = p;
++ if (!is_available_memory(md))
++ continue;
++ addr = (md->phys_addr+PAGE_SIZE-1) >> PAGE_SHIFT;
++ end = (md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT)) >> PAGE_SHIFT;
++
++ if ((pagenr >= addr) && (pagenr < end))
++ return 1;
++ }
++ return 0;
++ }
++
++ for (i = 0; i < e820.nr_map; i++) {
++
++ if (e820.map[i].type != E820_RAM) /* not usable memory */
++ continue;
++ /*
++ * !!!FIXME!!! Some BIOSen report areas as RAM that
++ * are not. Notably the 640->1Mb area. We need a sanity
++ * check here.
++ */
++ addr = (e820.map[i].addr+PAGE_SIZE-1) >> PAGE_SHIFT;
++ end = (e820.map[i].addr+e820.map[i].size) >> PAGE_SHIFT;
++ if ((pagenr >= addr) && (pagenr < end))
++ return 1;
++ }
++ return 0;
++}
++
++#ifdef CONFIG_HIGHMEM
++pte_t *kmap_pte;
++pgprot_t kmap_prot;
++
++#define kmap_get_fixmap_pte(vaddr) \
++ pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(vaddr), vaddr), (vaddr)), (vaddr))
++
++static void __init kmap_init(void)
++{
++ unsigned long kmap_vstart;
++
++ /* cache the first kmap pte */
++ kmap_vstart = __fix_to_virt(FIX_KMAP_BEGIN);
++ kmap_pte = kmap_get_fixmap_pte(kmap_vstart);
++
++ kmap_prot = PAGE_KERNEL;
++}
++
++static void __init permanent_kmaps_init(pgd_t *pgd_base)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++ unsigned long vaddr;
++
++ vaddr = PKMAP_BASE;
++ page_table_range_init(vaddr, vaddr + PAGE_SIZE*LAST_PKMAP, pgd_base);
++
++ pgd = swapper_pg_dir + pgd_index(vaddr);
++ pud = pud_offset(pgd, vaddr);
++ pmd = pmd_offset(pud, vaddr);
++ pte = pte_offset_kernel(pmd, vaddr);
++ pkmap_page_table = pte;
++}
++
++static void __meminit free_new_highpage(struct page *page, int pfn)
++{
++ init_page_count(page);
++ if (pfn < xen_start_info->nr_pages)
++ __free_page(page);
++ totalhigh_pages++;
++}
++
++void __init add_one_highpage_init(struct page *page, int pfn, int bad_ppro)
++{
++ if (page_is_ram(pfn) && !(bad_ppro && page_kills_ppro(pfn))) {
++ ClearPageReserved(page);
++ free_new_highpage(page, pfn);
++ } else
++ SetPageReserved(page);
++}
++
++static int add_one_highpage_hotplug(struct page *page, unsigned long pfn)
++{
++ free_new_highpage(page, pfn);
++ totalram_pages++;
++#ifdef CONFIG_FLATMEM
++ max_mapnr = max(pfn, max_mapnr);
++#endif
++ num_physpages++;
++ return 0;
++}
++
++/*
++ * Not currently handling the NUMA case.
++ * Assuming single node and all memory that
++ * has been added dynamically that would be
++ * onlined here is in HIGHMEM
++ */
++void online_page(struct page *page)
++{
++ ClearPageReserved(page);
++ add_one_highpage_hotplug(page, page_to_pfn(page));
++}
++
++
++#ifdef CONFIG_NUMA
++extern void set_highmem_pages_init(int);
++#else
++static void __init set_highmem_pages_init(int bad_ppro)
++{
++ int pfn;
++ for (pfn = highstart_pfn; pfn < highend_pfn; pfn++)
++ add_one_highpage_init(pfn_to_page(pfn), pfn, bad_ppro);
++ totalram_pages += totalhigh_pages;
++}
++#endif /* CONFIG_FLATMEM */
++
++#else
++#define kmap_init() do { } while (0)
++#define permanent_kmaps_init(pgd_base) do { } while (0)
++#define set_highmem_pages_init(bad_ppro) do { } while (0)
++#endif /* CONFIG_HIGHMEM */
++
++unsigned long long __PAGE_KERNEL = _PAGE_KERNEL;
++EXPORT_SYMBOL(__PAGE_KERNEL);
++unsigned long long __PAGE_KERNEL_EXEC = _PAGE_KERNEL_EXEC;
++
++#ifdef CONFIG_NUMA
++extern void __init remap_numa_kva(void);
++#else
++#define remap_numa_kva() do {} while (0)
++#endif
++
++pgd_t *swapper_pg_dir;
++
++static void __init pagetable_init (void)
++{
++ unsigned long vaddr;
++ pgd_t *pgd_base = (pgd_t *)xen_start_info->pt_base;
++
++ swapper_pg_dir = pgd_base;
++ init_mm.pgd = pgd_base;
++
++ /* Enable PSE if available */
++ if (cpu_has_pse) {
++ set_in_cr4(X86_CR4_PSE);
++ }
++
++ /* Enable PGE if available */
++ if (cpu_has_pge) {
++ set_in_cr4(X86_CR4_PGE);
++ __PAGE_KERNEL |= _PAGE_GLOBAL;
++ __PAGE_KERNEL_EXEC |= _PAGE_GLOBAL;
++ }
++
++ kernel_physical_mapping_init(pgd_base);
++ remap_numa_kva();
++
++ /*
++ * Fixed mappings, only the page table structure has to be
++ * created - mappings will be set by set_fixmap():
++ */
++ vaddr = __fix_to_virt(__end_of_fixed_addresses - 1) & PMD_MASK;
++ page_table_range_init(vaddr, 0, pgd_base);
++
++ permanent_kmaps_init(pgd_base);
++}
++
++#ifdef CONFIG_SOFTWARE_SUSPEND
++/*
++ * Swap suspend & friends need this for resume because things like the intel-agp
++ * driver might have split up a kernel 4MB mapping.
++ */
++char __nosavedata swsusp_pg_dir[PAGE_SIZE]
++ __attribute__ ((aligned (PAGE_SIZE)));
++
++static inline void save_pg_dir(void)
++{
++ memcpy(swsusp_pg_dir, swapper_pg_dir, PAGE_SIZE);
++}
++#else
++static inline void save_pg_dir(void)
++{
++}
++#endif
++
++void zap_low_mappings (void)
++{
++ int i;
++
++ save_pg_dir();
++
++ /*
++ * Zap initial low-memory mappings.
++ *
++ * Note that "pgd_clear()" doesn't do it for
++ * us, because pgd_clear() is a no-op on i386.
++ */
++ for (i = 0; i < USER_PTRS_PER_PGD; i++)
++#if defined(CONFIG_X86_PAE) && !defined(CONFIG_XEN)
++ set_pgd(swapper_pg_dir+i, __pgd(1 + __pa(empty_zero_page)));
++#else
++ set_pgd(swapper_pg_dir+i, __pgd(0));
++#endif
++ flush_tlb_all();
++}
++
++static int disable_nx __initdata = 0;
++u64 __supported_pte_mask __read_mostly = ~_PAGE_NX;
++EXPORT_SYMBOL(__supported_pte_mask);
++
++/*
++ * noexec = on|off
++ *
++ * Control non executable mappings.
++ *
++ * on Enable
++ * off Disable
++ */
++void __init noexec_setup(const char *str)
++{
++ if (!strncmp(str, "on",2) && cpu_has_nx) {
++ __supported_pte_mask |= _PAGE_NX;
++ disable_nx = 0;
++ } else if (!strncmp(str,"off",3)) {
++ disable_nx = 1;
++ __supported_pte_mask &= ~_PAGE_NX;
++ }
++}
++
++int nx_enabled = 0;
++#ifdef CONFIG_X86_PAE
++
++static void __init set_nx(void)
++{
++ unsigned int v[4], l, h;
++
++ if (cpu_has_pae && (cpuid_eax(0x80000000) > 0x80000001)) {
++ cpuid(0x80000001, &v[0], &v[1], &v[2], &v[3]);
++ if ((v[3] & (1 << 20)) && !disable_nx) {
++ rdmsr(MSR_EFER, l, h);
++ l |= EFER_NX;
++ wrmsr(MSR_EFER, l, h);
++ nx_enabled = 1;
++ __supported_pte_mask |= _PAGE_NX;
++ }
++ }
++}
++
++/*
++ * Enables/disables executability of a given kernel page and
++ * returns the previous setting.
++ */
++int __init set_kernel_exec(unsigned long vaddr, int enable)
++{
++ pte_t *pte;
++ int ret = 1;
++
++ if (!nx_enabled)
++ goto out;
++
++ pte = lookup_address(vaddr);
++ BUG_ON(!pte);
++
++ if (!pte_exec_kernel(*pte))
++ ret = 0;
++
++ if (enable)
++ pte->pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
++ else
++ pte->pte_high |= 1 << (_PAGE_BIT_NX - 32);
++ __flush_tlb_all();
++out:
++ return ret;
++}
++
++#endif
++
++/*
++ * paging_init() sets up the page tables - note that the first 8MB are
++ * already mapped by head.S.
++ *
++ * This routines also unmaps the page at virtual kernel address 0, so
++ * that we can trap those pesky NULL-reference errors in the kernel.
++ */
++void __init paging_init(void)
++{
++ int i;
++
++#ifdef CONFIG_X86_PAE
++ set_nx();
++ if (nx_enabled)
++ printk("NX (Execute Disable) protection: active\n");
++#endif
++
++ pagetable_init();
++
++#if defined(CONFIG_X86_PAE) && !defined(CONFIG_XEN)
++ /*
++ * We will bail out later - printk doesn't work right now so
++ * the user would just see a hanging kernel.
++ * when running as xen domain we are already in PAE mode at
++ * this point.
++ */
++ if (cpu_has_pae)
++ set_in_cr4(X86_CR4_PAE);
++#endif
++ __flush_tlb_all();
++
++ kmap_init();
++
++ /* Switch to the real shared_info page, and clear the
++ * dummy page. */
++ set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
++ HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
++ memset(empty_zero_page, 0, sizeof(empty_zero_page));
++
++ /* Setup mapping of lower 1st MB */
++ for (i = 0; i < NR_FIX_ISAMAPS; i++)
++ if (xen_start_info->flags & SIF_PRIVILEGED)
++ set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
++ else
++ __set_fixmap(FIX_ISAMAP_BEGIN - i,
++ virt_to_machine(empty_zero_page),
++ PAGE_KERNEL_RO);
++}
++
++/*
++ * Test if the WP bit works in supervisor mode. It isn't supported on 386's
++ * and also on some strange 486's (NexGen etc.). All 586+'s are OK. This
++ * used to involve black magic jumps to work around some nasty CPU bugs,
++ * but fortunately the switch to using exceptions got rid of all that.
++ */
++
++static void __init test_wp_bit(void)
++{
++ printk("Checking if this processor honours the WP bit even in supervisor mode... ");
++
++ /* Any page-aligned address will do, the test is non-destructive */
++ __set_fixmap(FIX_WP_TEST, __pa(&swapper_pg_dir), PAGE_READONLY);
++ boot_cpu_data.wp_works_ok = do_test_wp_bit();
++ clear_fixmap(FIX_WP_TEST);
++
++ if (!boot_cpu_data.wp_works_ok) {
++ printk("No.\n");
++#ifdef CONFIG_X86_WP_WORKS_OK
++ panic("This kernel doesn't support CPU's with broken WP. Recompile it for a 386!");
++#endif
++ } else {
++ printk("Ok.\n");
++ }
++}
++
++static void __init set_max_mapnr_init(void)
++{
++#ifdef CONFIG_HIGHMEM
++ num_physpages = highend_pfn;
++#else
++ num_physpages = max_low_pfn;
++#endif
++#ifdef CONFIG_FLATMEM
++ max_mapnr = num_physpages;
++#endif
++}
++
++static struct kcore_list kcore_mem, kcore_vmalloc;
++
++void __init mem_init(void)
++{
++ extern int ppro_with_ram_bug(void);
++ int codesize, reservedpages, datasize, initsize;
++ int tmp;
++ int bad_ppro;
++ unsigned long pfn;
++
++ contiguous_bitmap = alloc_bootmem_low_pages(
++ (max_low_pfn + 2*BITS_PER_LONG) >> 3);
++ BUG_ON(!contiguous_bitmap);
++ memset(contiguous_bitmap, 0, (max_low_pfn + 2*BITS_PER_LONG) >> 3);
++
++#if defined(CONFIG_SWIOTLB)
++ swiotlb_init();
++#endif
++
++#ifdef CONFIG_FLATMEM
++ if (!mem_map)
++ BUG();
++#endif
++
++ bad_ppro = ppro_with_ram_bug();
++
++#ifdef CONFIG_HIGHMEM
++ /* check that fixmap and pkmap do not overlap */
++ if (PKMAP_BASE+LAST_PKMAP*PAGE_SIZE >= FIXADDR_START) {
++ printk(KERN_ERR "fixmap and kmap areas overlap - this will crash\n");
++ printk(KERN_ERR "pkstart: %lxh pkend: %lxh fixstart %lxh\n",
++ PKMAP_BASE, PKMAP_BASE+LAST_PKMAP*PAGE_SIZE, FIXADDR_START);
++ BUG();
++ }
++#endif
++
++ set_max_mapnr_init();
++
++#ifdef CONFIG_HIGHMEM
++ high_memory = (void *) __va(highstart_pfn * PAGE_SIZE - 1) + 1;
++#else
++ high_memory = (void *) __va(max_low_pfn * PAGE_SIZE - 1) + 1;
++#endif
++ printk("vmalloc area: %lx-%lx, maxmem %lx\n",
++ VMALLOC_START,VMALLOC_END,MAXMEM);
++ BUG_ON(VMALLOC_START > VMALLOC_END);
++
++ /* this will put all low memory onto the freelists */
++ totalram_pages += free_all_bootmem();
++ /* XEN: init and count low-mem pages outside initial allocation. */
++ for (pfn = xen_start_info->nr_pages; pfn < max_low_pfn; pfn++) {
++ ClearPageReserved(&mem_map[pfn]);
++ init_page_count(&mem_map[pfn]);
++ totalram_pages++;
++ }
++
++ reservedpages = 0;
++ for (tmp = 0; tmp < max_low_pfn; tmp++)
++ /*
++ * Only count reserved RAM pages
++ */
++ if (page_is_ram(tmp) && PageReserved(pfn_to_page(tmp)))
++ reservedpages++;
++
++ set_highmem_pages_init(bad_ppro);
++
++ codesize = (unsigned long) &_etext - (unsigned long) &_text;
++ datasize = (unsigned long) &_edata - (unsigned long) &_etext;
++ initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
++
++ kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
++ kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
++ VMALLOC_END-VMALLOC_START);
++
++ printk(KERN_INFO "Memory: %luk/%luk available (%dk kernel code, %dk reserved, %dk data, %dk init, %ldk highmem)\n",
++ (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
++ num_physpages << (PAGE_SHIFT-10),
++ codesize >> 10,
++ reservedpages << (PAGE_SHIFT-10),
++ datasize >> 10,
++ initsize >> 10,
++ (unsigned long) (totalhigh_pages << (PAGE_SHIFT-10))
++ );
++
++#ifdef CONFIG_X86_PAE
++ if (!cpu_has_pae)
++ panic("cannot execute a PAE-enabled kernel on a PAE-less CPU!");
++#endif
++ if (boot_cpu_data.wp_works_ok < 0)
++ test_wp_bit();
++
++ /*
++ * Subtle. SMP is doing it's boot stuff late (because it has to
++ * fork idle threads) - but it also needs low mappings for the
++ * protected-mode entry to work. We zap these entries only after
++ * the WP-bit has been tested.
++ */
++#ifndef CONFIG_SMP
++ zap_low_mappings();
++#endif
++
++ set_bit(PG_pinned, &virt_to_page(init_mm.pgd)->flags);
++}
++
++/*
++ * this is for the non-NUMA, single node SMP system case.
++ * Specifically, in the case of x86, we will always add
++ * memory to the highmem for now.
++ */
++#ifdef CONFIG_MEMORY_HOTPLUG
++#ifndef CONFIG_NEED_MULTIPLE_NODES
++int add_memory(u64 start, u64 size)
++{
++ struct pglist_data *pgdata = &contig_page_data;
++ struct zone *zone = pgdata->node_zones + MAX_NR_ZONES-1;
++ unsigned long start_pfn = start >> PAGE_SHIFT;
++ unsigned long nr_pages = size >> PAGE_SHIFT;
++
++ return __add_pages(zone, start_pfn, nr_pages);
++}
++
++int remove_memory(u64 start, u64 size)
++{
++ return -EINVAL;
++}
++#endif
++#endif
++
++kmem_cache_t *pgd_cache;
++kmem_cache_t *pmd_cache;
++
++void __init pgtable_cache_init(void)
++{
++ if (PTRS_PER_PMD > 1) {
++ pmd_cache = kmem_cache_create("pmd",
++ PTRS_PER_PMD*sizeof(pmd_t),
++ PTRS_PER_PMD*sizeof(pmd_t),
++ 0,
++ pmd_ctor,
++ NULL);
++ if (!pmd_cache)
++ panic("pgtable_cache_init(): cannot create pmd cache");
++ }
++ pgd_cache = kmem_cache_create("pgd",
++#ifndef CONFIG_XEN
++ PTRS_PER_PGD*sizeof(pgd_t),
++ PTRS_PER_PGD*sizeof(pgd_t),
++#else
++ PAGE_SIZE,
++ PAGE_SIZE,
++#endif
++ 0,
++ pgd_ctor,
++ PTRS_PER_PMD == 1 ? pgd_dtor : NULL);
++ if (!pgd_cache)
++ panic("pgtable_cache_init(): Cannot create pgd cache");
++}
++
++/*
++ * This function cannot be __init, since exceptions don't work in that
++ * section. Put this after the callers, so that it cannot be inlined.
++ */
++static int noinline do_test_wp_bit(void)
++{
++ char tmp_reg;
++ int flag;
++
++ __asm__ __volatile__(
++ " movb %0,%1 \n"
++ "1: movb %1,%0 \n"
++ " xorl %2,%2 \n"
++ "2: \n"
++ ".section __ex_table,\"a\"\n"
++ " .align 4 \n"
++ " .long 1b,2b \n"
++ ".previous \n"
++ :"=m" (*(char *)fix_to_virt(FIX_WP_TEST)),
++ "=q" (tmp_reg),
++ "=r" (flag)
++ :"2" (1)
++ :"memory");
++
++ return flag;
++}
++
++#ifdef CONFIG_DEBUG_RODATA
++
++extern char __start_rodata, __end_rodata;
++void mark_rodata_ro(void)
++{
++ unsigned long addr = (unsigned long)&__start_rodata;
++
++ for (; addr < (unsigned long)&__end_rodata; addr += PAGE_SIZE)
++ change_page_attr(virt_to_page(addr), 1, PAGE_KERNEL_RO);
++
++ printk ("Write protecting the kernel read-only data: %luk\n",
++ (unsigned long)(&__end_rodata - &__start_rodata) >> 10);
++
++ /*
++ * change_page_attr() requires a global_flush_tlb() call after it.
++ * We do this after the printk so that if something went wrong in the
++ * change, the printk gets out at least to give a better debug hint
++ * of who is the culprit.
++ */
++ global_flush_tlb();
++}
++#endif
++
++void free_init_pages(char *what, unsigned long begin, unsigned long end)
++{
++ unsigned long addr;
++
++ for (addr = begin; addr < end; addr += PAGE_SIZE) {
++ ClearPageReserved(virt_to_page(addr));
++ init_page_count(virt_to_page(addr));
++ memset((void *)addr, 0xcc, PAGE_SIZE);
++ free_page(addr);
++ totalram_pages++;
++ }
++ printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
++}
++
++void free_initmem(void)
++{
++ free_init_pages("unused kernel memory",
++ (unsigned long)(&__init_begin),
++ (unsigned long)(&__init_end));
++}
++
++#ifdef CONFIG_BLK_DEV_INITRD
++void free_initrd_mem(unsigned long start, unsigned long end)
++{
++ free_init_pages("initrd memory", start, end);
++}
++#endif
++
+diff -urNp linux-2.6/arch/i386/mm/ioremap-xen.c new/arch/i386/mm/ioremap-xen.c
+--- linux-2.6/arch/i386/mm/ioremap-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/ioremap-xen.c 2006-05-23 21:31:13.000000000 +0200
+@@ -0,0 +1,476 @@
++/*
++ * arch/i386/mm/ioremap.c
++ *
++ * Re-map IO memory to kernel address space so that we can access it.
++ * This is needed for high PCI addresses that aren't mapped in the
++ * 640k-1MB IO memory area on PC's
++ *
++ * (C) Copyright 1995 1996 Linus Torvalds
++ */
++
++#include <linux/vmalloc.h>
++#include <linux/init.h>
++#include <linux/slab.h>
++#include <linux/module.h>
++#include <asm/io.h>
++#include <asm/fixmap.h>
++#include <asm/cacheflush.h>
++#include <asm/tlbflush.h>
++#include <asm/pgtable.h>
++#include <asm/pgalloc.h>
++
++#define ISA_START_ADDRESS 0x0
++#define ISA_END_ADDRESS 0x100000
++
++#if 0 /* not PAE safe */
++/* These hacky macros avoid phys->machine translations. */
++#define __direct_pte(x) ((pte_t) { (x) } )
++#define __direct_mk_pte(page_nr,pgprot) \
++ __direct_pte(((page_nr) << PAGE_SHIFT) | pgprot_val(pgprot))
++#define direct_mk_pte_phys(physpage, pgprot) \
++ __direct_mk_pte((physpage) >> PAGE_SHIFT, pgprot)
++#endif
++
++static int direct_remap_area_pte_fn(pte_t *pte,
++ struct page *pmd_page,
++ unsigned long address,
++ void *data)
++{
++ mmu_update_t **v = (mmu_update_t **)data;
++
++ (*v)->ptr = ((u64)pfn_to_mfn(page_to_pfn(pmd_page)) <<
++ PAGE_SHIFT) | ((unsigned long)pte & ~PAGE_MASK);
++ (*v)++;
++
++ return 0;
++}
++
++static int __direct_remap_pfn_range(struct mm_struct *mm,
++ unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid)
++{
++ int rc;
++ unsigned long i, start_address;
++ mmu_update_t *u, *v, *w;
++
++ u = v = w = (mmu_update_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT);
++ if (u == NULL)
++ return -ENOMEM;
++
++ start_address = address;
++
++ flush_cache_all();
++
++ for (i = 0; i < size; i += PAGE_SIZE) {
++ if ((v - u) == (PAGE_SIZE / sizeof(mmu_update_t))) {
++ /* Fill in the PTE pointers. */
++ rc = apply_to_page_range(mm, start_address,
++ address - start_address,
++ direct_remap_area_pte_fn, &w);
++ if (rc)
++ goto out;
++ w = u;
++ rc = -EFAULT;
++ if (HYPERVISOR_mmu_update(u, v - u, NULL, domid) < 0)
++ goto out;
++ v = u;
++ start_address = address;
++ }
++
++ /*
++ * Fill in the machine address: PTE ptr is done later by
++ * __direct_remap_area_pages().
++ */
++ v->val = pte_val_ma(pfn_pte_ma(mfn, prot));
++
++ mfn++;
++ address += PAGE_SIZE;
++ v++;
++ }
++
++ if (v != u) {
++ /* get the ptep's filled in */
++ rc = apply_to_page_range(mm, start_address,
++ address - start_address,
++ direct_remap_area_pte_fn, &w);
++ if (rc)
++ goto out;
++ rc = -EFAULT;
++ if (unlikely(HYPERVISOR_mmu_update(u, v - u, NULL, domid) < 0))
++ goto out;
++ }
++
++ rc = 0;
++
++ out:
++ flush_tlb_all();
++
++ free_page((unsigned long)u);
++
++ return rc;
++}
++
++int direct_remap_pfn_range(struct vm_area_struct *vma,
++ unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid)
++{
++ /* Same as remap_pfn_range(). */
++ vma->vm_flags |= VM_IO | VM_RESERVED;
++
++ if (domid == DOMID_SELF)
++ return -EINVAL;
++
++ return __direct_remap_pfn_range(
++ vma->vm_mm, address, mfn, size, prot, domid);
++}
++EXPORT_SYMBOL(direct_remap_pfn_range);
++
++int direct_kernel_remap_pfn_range(unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid)
++{
++ return __direct_remap_pfn_range(
++ &init_mm, address, mfn, size, prot, domid);
++}
++EXPORT_SYMBOL(direct_kernel_remap_pfn_range);
++
++static int lookup_pte_fn(
++ pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
++{
++ uint64_t *ptep = (uint64_t *)data;
++ if (ptep)
++ *ptep = ((uint64_t)pfn_to_mfn(page_to_pfn(pmd_page)) <<
++ PAGE_SHIFT) | ((unsigned long)pte & ~PAGE_MASK);
++ return 0;
++}
++
++int create_lookup_pte_addr(struct mm_struct *mm,
++ unsigned long address,
++ uint64_t *ptep)
++{
++ return apply_to_page_range(mm, address, PAGE_SIZE,
++ lookup_pte_fn, ptep);
++}
++
++EXPORT_SYMBOL(create_lookup_pte_addr);
++
++static int noop_fn(
++ pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
++{
++ return 0;
++}
++
++int touch_pte_range(struct mm_struct *mm,
++ unsigned long address,
++ unsigned long size)
++{
++ return apply_to_page_range(mm, address, size, noop_fn, NULL);
++}
++
++EXPORT_SYMBOL(touch_pte_range);
++
++void *vm_map_xen_pages (unsigned long maddr, int vm_size, pgprot_t prot)
++{
++ int error;
++
++ struct vm_struct *vma;
++ vma = get_vm_area (vm_size, VM_IOREMAP);
++
++ if (vma == NULL) {
++ printk ("ioremap.c,vm_map_xen_pages(): "
++ "Failed to get VMA area\n");
++ return NULL;
++ }
++
++ error = direct_kernel_remap_pfn_range((unsigned long) vma->addr,
++ maddr >> PAGE_SHIFT, vm_size,
++ prot, DOMID_SELF );
++ if (error == 0) {
++ return vma->addr;
++ } else {
++ printk ("ioremap.c,vm_map_xen_pages(): "
++ "Failed to map xen shared pages into kernel space\n");
++ return NULL;
++ }
++}
++EXPORT_SYMBOL(vm_map_xen_pages);
++
++/*
++ * Does @address reside within a non-highmem page that is local to this virtual
++ * machine (i.e., not an I/O page, nor a memory page belonging to another VM).
++ * See the comment that accompanies mfn_to_local_pfn() in page.h to understand
++ * why this works.
++ */
++static inline int is_local_lowmem(unsigned long address)
++{
++ extern unsigned long max_low_pfn;
++ return (mfn_to_local_pfn(address >> PAGE_SHIFT) < max_low_pfn);
++}
++
++/*
++ * Generic mapping function (not visible outside):
++ */
++
++/*
++ * Remap an arbitrary physical address space into the kernel virtual
++ * address space. Needed when the kernel wants to access high addresses
++ * directly.
++ *
++ * NOTE! We need to allow non-page-aligned mappings too: we will obviously
++ * have to convert them into an offset in a page-aligned mapping, but the
++ * caller shouldn't need to know that small detail.
++ */
++void __iomem * __ioremap(unsigned long phys_addr, unsigned long size, unsigned long flags)
++{
++ void __iomem * addr;
++ struct vm_struct * area;
++ unsigned long offset, last_addr;
++ domid_t domid = DOMID_IO;
++
++ /* Don't allow wraparound or zero size */
++ last_addr = phys_addr + size - 1;
++ if (!size || last_addr < phys_addr)
++ return NULL;
++
++ /*
++ * Don't remap the low PCI/ISA area, it's always mapped..
++ */
++ if (xen_start_info->flags & SIF_PRIVILEGED &&
++ phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
++ return (void __iomem *) isa_bus_to_virt(phys_addr);
++
++ /*
++ * Don't allow anybody to remap normal RAM that we're using..
++ */
++ if (is_local_lowmem(phys_addr)) {
++ char *t_addr, *t_end;
++ struct page *page;
++
++ t_addr = bus_to_virt(phys_addr);
++ t_end = t_addr + (size - 1);
++
++ for(page = virt_to_page(t_addr); page <= virt_to_page(t_end); page++)
++ if(!PageReserved(page))
++ return NULL;
++
++ domid = DOMID_SELF;
++ }
++
++ /*
++ * Mappings have to be page-aligned
++ */
++ offset = phys_addr & ~PAGE_MASK;
++ phys_addr &= PAGE_MASK;
++ size = PAGE_ALIGN(last_addr+1) - phys_addr;
++
++ /*
++ * Ok, go for it..
++ */
++ area = get_vm_area(size, VM_IOREMAP | (flags << 20));
++ if (!area)
++ return NULL;
++ area->phys_addr = phys_addr;
++ addr = (void __iomem *) area->addr;
++ flags |= _PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED;
++#ifdef __x86_64__
++ flags |= _PAGE_USER;
++#endif
++ if (__direct_remap_pfn_range(&init_mm, (unsigned long)addr,
++ phys_addr>>PAGE_SHIFT,
++ size, __pgprot(flags), domid)) {
++ vunmap((void __force *) addr);
++ return NULL;
++ }
++ return (void __iomem *) (offset + (char __iomem *)addr);
++}
++EXPORT_SYMBOL(__ioremap);
++
++/**
++ * ioremap_nocache - map bus memory into CPU space
++ * @offset: bus address of the memory
++ * @size: size of the resource to map
++ *
++ * ioremap_nocache performs a platform specific sequence of operations to
++ * make bus memory CPU accessible via the readb/readw/readl/writeb/
++ * writew/writel functions and the other mmio helpers. The returned
++ * address is not guaranteed to be usable directly as a virtual
++ * address.
++ *
++ * This version of ioremap ensures that the memory is marked uncachable
++ * on the CPU as well as honouring existing caching rules from things like
++ * the PCI bus. Note that there are other caches and buffers on many
++ * busses. In particular driver authors should read up on PCI writes
++ *
++ * It's useful if some control registers are in such an area and
++ * write combining or read caching is not desirable:
++ *
++ * Must be freed with iounmap.
++ */
++
++void __iomem *ioremap_nocache (unsigned long phys_addr, unsigned long size)
++{
++ unsigned long last_addr;
++ void __iomem *p = __ioremap(phys_addr, size, _PAGE_PCD);
++ if (!p)
++ return p;
++
++ /* Guaranteed to be > phys_addr, as per __ioremap() */
++ last_addr = phys_addr + size - 1;
++
++ if (is_local_lowmem(last_addr)) {
++ struct page *ppage = virt_to_page(bus_to_virt(phys_addr));
++ unsigned long npages;
++
++ phys_addr &= PAGE_MASK;
++
++ /* This might overflow and become zero.. */
++ last_addr = PAGE_ALIGN(last_addr);
++
++ /* .. but that's ok, because modulo-2**n arithmetic will make
++ * the page-aligned "last - first" come out right.
++ */
++ npages = (last_addr - phys_addr) >> PAGE_SHIFT;
++
++ if (change_page_attr(ppage, npages, PAGE_KERNEL_NOCACHE) < 0) {
++ iounmap(p);
++ p = NULL;
++ }
++ global_flush_tlb();
++ }
++
++ return p;
++}
++EXPORT_SYMBOL(ioremap_nocache);
++
++/**
++ * iounmap - Free a IO remapping
++ * @addr: virtual address from ioremap_*
++ *
++ * Caller must ensure there is only one unmapping for the same pointer.
++ */
++void iounmap(volatile void __iomem *addr)
++{
++ struct vm_struct *p, *o;
++
++ if ((void __force *)addr <= high_memory)
++ return;
++
++ /*
++ * __ioremap special-cases the PCI/ISA range by not instantiating a
++ * vm_area and by simply returning an address into the kernel mapping
++ * of ISA space. So handle that here.
++ */
++ if ((unsigned long) addr >= fix_to_virt(FIX_ISAMAP_BEGIN))
++ return;
++
++ addr = (volatile void __iomem *)(PAGE_MASK & (unsigned long __force)addr);
++
++ /* Use the vm area unlocked, assuming the caller
++ ensures there isn't another iounmap for the same address
++ in parallel. Reuse of the virtual address is prevented by
++ leaving it in the global lists until we're done with it.
++ cpa takes care of the direct mappings. */
++ read_lock(&vmlist_lock);
++ for (p = vmlist; p; p = p->next) {
++ if (p->addr == addr)
++ break;
++ }
++ read_unlock(&vmlist_lock);
++
++ if (!p) {
++ printk("iounmap: bad address %p\n", addr);
++ dump_stack();
++ return;
++ }
++
++ /* Reset the direct mapping. Can block */
++ if ((p->flags >> 20) && is_local_lowmem(p->phys_addr)) {
++ /* p->size includes the guard page, but cpa doesn't like that */
++ change_page_attr(virt_to_page(bus_to_virt(p->phys_addr)),
++ (p->size - PAGE_SIZE) >> PAGE_SHIFT,
++ PAGE_KERNEL);
++ global_flush_tlb();
++ }
++
++ /* Finally remove it */
++ o = remove_vm_area((void *)addr);
++ BUG_ON(p != o || o == NULL);
++ kfree(p);
++}
++EXPORT_SYMBOL(iounmap);
++
++void __init *bt_ioremap(unsigned long phys_addr, unsigned long size)
++{
++ unsigned long offset, last_addr;
++ unsigned int nrpages;
++ enum fixed_addresses idx;
++
++ /* Don't allow wraparound or zero size */
++ last_addr = phys_addr + size - 1;
++ if (!size || last_addr < phys_addr)
++ return NULL;
++
++ /*
++ * Don't remap the low PCI/ISA area, it's always mapped..
++ */
++ if (xen_start_info->flags & SIF_PRIVILEGED &&
++ phys_addr >= ISA_START_ADDRESS && last_addr < ISA_END_ADDRESS)
++ return isa_bus_to_virt(phys_addr);
++
++ /*
++ * Mappings have to be page-aligned
++ */
++ offset = phys_addr & ~PAGE_MASK;
++ phys_addr &= PAGE_MASK;
++ size = PAGE_ALIGN(last_addr) - phys_addr;
++
++ /*
++ * Mappings have to fit in the FIX_BTMAP area.
++ */
++ nrpages = size >> PAGE_SHIFT;
++ if (nrpages > NR_FIX_BTMAPS)
++ return NULL;
++
++ /*
++ * Ok, go for it..
++ */
++ idx = FIX_BTMAP_BEGIN;
++ while (nrpages > 0) {
++ set_fixmap(idx, phys_addr);
++ phys_addr += PAGE_SIZE;
++ --idx;
++ --nrpages;
++ }
++ return (void*) (offset + fix_to_virt(FIX_BTMAP_BEGIN));
++}
++
++void __init bt_iounmap(void *addr, unsigned long size)
++{
++ unsigned long virt_addr;
++ unsigned long offset;
++ unsigned int nrpages;
++ enum fixed_addresses idx;
++
++ virt_addr = (unsigned long)addr;
++ if (virt_addr < fix_to_virt(FIX_BTMAP_BEGIN))
++ return;
++ if (virt_addr >= fix_to_virt(FIX_ISAMAP_BEGIN))
++ return;
++ offset = virt_addr & ~PAGE_MASK;
++ nrpages = PAGE_ALIGN(offset + size - 1) >> PAGE_SHIFT;
++
++ idx = FIX_BTMAP_BEGIN;
++ while (nrpages > 0) {
++ clear_fixmap(idx);
++ --idx;
++ --nrpages;
++ }
++}
+diff -urNp linux-2.6/arch/i386/mm/Makefile new/arch/i386/mm/Makefile
+--- linux-2.6/arch/i386/mm/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/mm/Makefile 2006-05-09 12:32:36.000000000 +0200
+@@ -8,3 +8,11 @@ obj-$(CONFIG_NUMA) += discontig.o
+ obj-$(CONFIG_HUGETLB_PAGE) += hugetlbpage.o
+ obj-$(CONFIG_HIGHMEM) += highmem.o
+ obj-$(CONFIG_BOOT_IOREMAP) += boot_ioremap.o
++
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++
++obj-y += hypervisor.o
++
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/arch/i386/mm/pageattr.c new/arch/i386/mm/pageattr.c
+--- linux-2.6/arch/i386/mm/pageattr.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/mm/pageattr.c 2006-05-09 12:32:37.000000000 +0200
+@@ -85,7 +85,7 @@ static void set_pmd_pte(pte_t *kpte, uns
+ unsigned long flags;
+
+ set_pte_atomic(kpte, pte); /* change init_mm */
+- if (PTRS_PER_PMD > 1)
++ if (HAVE_SHARED_KERNEL_PMD)
+ return;
+
+ spin_lock_irqsave(&pgd_lock, flags);
+diff -urNp linux-2.6/arch/i386/mm/pgtable.c new/arch/i386/mm/pgtable.c
+--- linux-2.6/arch/i386/mm/pgtable.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/mm/pgtable.c 2006-05-09 12:32:37.000000000 +0200
+@@ -13,6 +13,7 @@
+ #include <linux/slab.h>
+ #include <linux/pagemap.h>
+ #include <linux/spinlock.h>
++#include <linux/module.h>
+
+ #include <asm/system.h>
+ #include <asm/pgtable.h>
+@@ -138,6 +139,10 @@ void set_pmd_pfn(unsigned long vaddr, un
+ __flush_tlb_one(vaddr);
+ }
+
++static int nr_fixmaps = 0;
++unsigned long __FIXADDR_TOP = 0xfffff000;
++EXPORT_SYMBOL(__FIXADDR_TOP);
++
+ void __set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t flags)
+ {
+ unsigned long address = __fix_to_virt(idx);
+@@ -147,6 +152,13 @@ void __set_fixmap (enum fixed_addresses
+ return;
+ }
+ set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
++ nr_fixmaps++;
++}
++
++void set_fixaddr_top(unsigned long top)
++{
++ BUG_ON(nr_fixmaps > 0);
++ __FIXADDR_TOP = top - PAGE_SIZE;
+ }
+
+ pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
+diff -urNp linux-2.6/arch/i386/mm/pgtable-xen.c new/arch/i386/mm/pgtable-xen.c
+--- linux-2.6/arch/i386/mm/pgtable-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/mm/pgtable-xen.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,694 @@
++/*
++ * linux/arch/i386/mm/pgtable.c
++ */
++
++#include <linux/config.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/mm.h>
++#include <linux/swap.h>
++#include <linux/smp.h>
++#include <linux/highmem.h>
++#include <linux/slab.h>
++#include <linux/pagemap.h>
++#include <linux/spinlock.h>
++#include <linux/module.h>
++
++#include <asm/system.h>
++#include <asm/pgtable.h>
++#include <asm/pgalloc.h>
++#include <asm/fixmap.h>
++#include <asm/e820.h>
++#include <asm/tlb.h>
++#include <asm/tlbflush.h>
++#include <asm/io.h>
++#include <asm/mmu_context.h>
++
++#include <xen/features.h>
++#include <xen/foreign_page.h>
++#include <asm/hypervisor.h>
++
++static void pgd_test_and_unpin(pgd_t *pgd);
++
++void show_mem(void)
++{
++ int total = 0, reserved = 0;
++ int shared = 0, cached = 0;
++ int highmem = 0;
++ struct page *page;
++ pg_data_t *pgdat;
++ unsigned long i;
++ struct page_state ps;
++ unsigned long flags;
++
++ printk(KERN_INFO "Mem-info:\n");
++ show_free_areas();
++ printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
++ for_each_online_pgdat(pgdat) {
++ pgdat_resize_lock(pgdat, &flags);
++ for (i = 0; i < pgdat->node_spanned_pages; ++i) {
++ page = pgdat_page_nr(pgdat, i);
++ total++;
++ if (PageHighMem(page))
++ highmem++;
++ if (PageReserved(page))
++ reserved++;
++ else if (PageSwapCache(page))
++ cached++;
++ else if (page_count(page))
++ shared += page_count(page) - 1;
++ }
++ pgdat_resize_unlock(pgdat, &flags);
++ }
++ printk(KERN_INFO "%d pages of RAM\n", total);
++ printk(KERN_INFO "%d pages of HIGHMEM\n", highmem);
++ printk(KERN_INFO "%d reserved pages\n", reserved);
++ printk(KERN_INFO "%d pages shared\n", shared);
++ printk(KERN_INFO "%d pages swap cached\n", cached);
++
++ get_page_state(&ps);
++ printk(KERN_INFO "%lu pages dirty\n", ps.nr_dirty);
++ printk(KERN_INFO "%lu pages writeback\n", ps.nr_writeback);
++ printk(KERN_INFO "%lu pages mapped\n", ps.nr_mapped);
++ printk(KERN_INFO "%lu pages slab\n", ps.nr_slab);
++ printk(KERN_INFO "%lu pages pagetables\n", ps.nr_page_table_pages);
++}
++
++/*
++ * Associate a virtual page frame with a given physical page frame
++ * and protection flags for that frame.
++ */
++static void set_pte_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++ pgd = swapper_pg_dir + pgd_index(vaddr);
++ if (pgd_none(*pgd)) {
++ BUG();
++ return;
++ }
++ pud = pud_offset(pgd, vaddr);
++ if (pud_none(*pud)) {
++ BUG();
++ return;
++ }
++ pmd = pmd_offset(pud, vaddr);
++ if (pmd_none(*pmd)) {
++ BUG();
++ return;
++ }
++ pte = pte_offset_kernel(pmd, vaddr);
++ /* <pfn,flags> stored as-is, to permit clearing entries */
++ set_pte(pte, pfn_pte(pfn, flags));
++
++ /*
++ * It's enough to flush this one mapping.
++ * (PGE mappings get flushed as well)
++ */
++ __flush_tlb_one(vaddr);
++}
++
++/*
++ * Associate a virtual page frame with a given physical page frame
++ * and protection flags for that frame.
++ */
++static void set_pte_pfn_ma(unsigned long vaddr, unsigned long pfn,
++ pgprot_t flags)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++ pgd = swapper_pg_dir + pgd_index(vaddr);
++ if (pgd_none(*pgd)) {
++ BUG();
++ return;
++ }
++ pud = pud_offset(pgd, vaddr);
++ if (pud_none(*pud)) {
++ BUG();
++ return;
++ }
++ pmd = pmd_offset(pud, vaddr);
++ if (pmd_none(*pmd)) {
++ BUG();
++ return;
++ }
++ pte = pte_offset_kernel(pmd, vaddr);
++ /* <pfn,flags> stored as-is, to permit clearing entries */
++ set_pte(pte, pfn_pte_ma(pfn, flags));
++
++ /*
++ * It's enough to flush this one mapping.
++ * (PGE mappings get flushed as well)
++ */
++ __flush_tlb_one(vaddr);
++}
++
++/*
++ * Associate a large virtual page frame with a given physical page frame
++ * and protection flags for that frame. pfn is for the base of the page,
++ * vaddr is what the page gets mapped to - both must be properly aligned.
++ * The pmd must already be instantiated. Assumes PAE mode.
++ */
++void set_pmd_pfn(unsigned long vaddr, unsigned long pfn, pgprot_t flags)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++
++ if (vaddr & (PMD_SIZE-1)) { /* vaddr is misaligned */
++ printk(KERN_WARNING "set_pmd_pfn: vaddr misaligned\n");
++ return; /* BUG(); */
++ }
++ if (pfn & (PTRS_PER_PTE-1)) { /* pfn is misaligned */
++ printk(KERN_WARNING "set_pmd_pfn: pfn misaligned\n");
++ return; /* BUG(); */
++ }
++ pgd = swapper_pg_dir + pgd_index(vaddr);
++ if (pgd_none(*pgd)) {
++ printk(KERN_WARNING "set_pmd_pfn: pgd_none\n");
++ return; /* BUG(); */
++ }
++ pud = pud_offset(pgd, vaddr);
++ pmd = pmd_offset(pud, vaddr);
++ set_pmd(pmd, pfn_pmd(pfn, flags));
++ /*
++ * It's enough to flush this one mapping.
++ * (PGE mappings get flushed as well)
++ */
++ __flush_tlb_one(vaddr);
++}
++
++static int nr_fixmaps = 0;
++unsigned long __FIXADDR_TOP = (HYPERVISOR_VIRT_START - 2 * PAGE_SIZE);
++EXPORT_SYMBOL(__FIXADDR_TOP);
++
++void __set_fixmap (enum fixed_addresses idx, maddr_t phys, pgprot_t flags)
++{
++ unsigned long address = __fix_to_virt(idx);
++
++ if (idx >= __end_of_fixed_addresses) {
++ BUG();
++ return;
++ }
++ switch (idx) {
++ case FIX_WP_TEST:
++#ifdef CONFIG_X86_F00F_BUG
++ case FIX_F00F_IDT:
++#endif
++ set_pte_pfn(address, phys >> PAGE_SHIFT, flags);
++ break;
++ default:
++ set_pte_pfn_ma(address, phys >> PAGE_SHIFT, flags);
++ break;
++ }
++ nr_fixmaps++;
++}
++
++void set_fixaddr_top(unsigned long top)
++{
++ BUG_ON(nr_fixmaps > 0);
++ __FIXADDR_TOP = top - PAGE_SIZE;
++}
++
++pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
++{
++ pte_t *pte = (pte_t *)__get_free_page(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO);
++ if (pte)
++ make_lowmem_page_readonly(pte, XENFEAT_writable_page_tables);
++ return pte;
++}
++
++struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
++{
++ struct page *pte;
++
++#ifdef CONFIG_HIGHPTE
++ pte = alloc_pages(GFP_KERNEL|__GFP_HIGHMEM|__GFP_REPEAT|__GFP_ZERO, 0);
++#else
++ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
++ if (pte) {
++ SetPageForeign(pte, pte_free);
++ init_page_count(pte);
++ }
++#endif
++ return pte;
++}
++
++void pte_free(struct page *pte)
++{
++ unsigned long va = (unsigned long)__va(page_to_pfn(pte)<<PAGE_SHIFT);
++
++ if (!pte_write(*virt_to_ptep(va)))
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ va, pfn_pte(page_to_pfn(pte), PAGE_KERNEL), 0));
++
++ ClearPageForeign(pte);
++ init_page_count(pte);
++
++ __free_page(pte);
++}
++
++void pmd_ctor(void *pmd, kmem_cache_t *cache, unsigned long flags)
++{
++ memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
++}
++
++/*
++ * List of all pgd's needed for non-PAE so it can invalidate entries
++ * in both cached and uncached pgd's; not needed for PAE since the
++ * kernel pmd is shared. If PAE were not to share the pmd a similar
++ * tactic would be needed. This is essentially codepath-based locking
++ * against pageattr.c; it is the unique case in which a valid change
++ * of kernel pagetables can't be lazily synchronized by vmalloc faults.
++ * vmalloc faults work because attached pagetables are never freed.
++ * The locking scheme was chosen on the basis of manfred's
++ * recommendations and having no core impact whatsoever.
++ * -- wli
++ */
++DEFINE_SPINLOCK(pgd_lock);
++struct page *pgd_list;
++
++static inline void pgd_list_add(pgd_t *pgd)
++{
++ struct page *page = virt_to_page(pgd);
++ page->index = (unsigned long)pgd_list;
++ if (pgd_list)
++ set_page_private(pgd_list, (unsigned long)&page->index);
++ pgd_list = page;
++ set_page_private(page, (unsigned long)&pgd_list);
++}
++
++static inline void pgd_list_del(pgd_t *pgd)
++{
++ struct page *next, **pprev, *page = virt_to_page(pgd);
++ next = (struct page *)page->index;
++ pprev = (struct page **)page_private(page);
++ *pprev = next;
++ if (next)
++ set_page_private(next, (unsigned long)pprev);
++}
++
++void pgd_ctor(void *pgd, kmem_cache_t *cache, unsigned long unused)
++{
++ unsigned long flags;
++
++ if (PTRS_PER_PMD > 1) {
++ if (HAVE_SHARED_KERNEL_PMD)
++ clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
++ swapper_pg_dir + USER_PTRS_PER_PGD,
++ KERNEL_PGD_PTRS);
++ } else {
++ spin_lock_irqsave(&pgd_lock, flags);
++ clone_pgd_range((pgd_t *)pgd + USER_PTRS_PER_PGD,
++ swapper_pg_dir + USER_PTRS_PER_PGD,
++ KERNEL_PGD_PTRS);
++ memset(pgd, 0, USER_PTRS_PER_PGD*sizeof(pgd_t));
++ pgd_list_add(pgd);
++ spin_unlock_irqrestore(&pgd_lock, flags);
++ }
++}
++
++/* never called when PTRS_PER_PMD > 1 */
++void pgd_dtor(void *pgd, kmem_cache_t *cache, unsigned long unused)
++{
++ unsigned long flags; /* can be called from interrupt context */
++
++ spin_lock_irqsave(&pgd_lock, flags);
++ pgd_list_del(pgd);
++ spin_unlock_irqrestore(&pgd_lock, flags);
++
++ pgd_test_and_unpin(pgd);
++}
++
++pgd_t *pgd_alloc(struct mm_struct *mm)
++{
++ int i;
++ pgd_t *pgd = kmem_cache_alloc(pgd_cache, GFP_KERNEL);
++ pmd_t **pmd;
++ unsigned long flags;
++
++ pgd_test_and_unpin(pgd);
++
++ if (PTRS_PER_PMD == 1 || !pgd)
++ return pgd;
++
++ if (HAVE_SHARED_KERNEL_PMD) {
++ for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
++ pmd_t *pmd = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
++ if (!pmd)
++ goto out_oom;
++ set_pgd(&pgd[i], __pgd(1 + __pa(pmd)));
++ }
++ return pgd;
++ }
++
++ /*
++ * We can race save/restore (if we sleep during a GFP_KERNEL memory
++ * allocation). We therefore store virtual addresses of pmds as they
++ * do not change across save/restore, and poke the machine addresses
++ * into the pgdir under the pgd_lock.
++ */
++ pmd = kmalloc(PTRS_PER_PGD * sizeof(pmd_t *), GFP_KERNEL);
++ if (!pmd) {
++ kmem_cache_free(pgd_cache, pgd);
++ return NULL;
++ }
++
++ /* Allocate pmds, remember virtual addresses. */
++ for (i = 0; i < PTRS_PER_PGD; ++i) {
++ pmd[i] = kmem_cache_alloc(pmd_cache, GFP_KERNEL);
++ if (!pmd[i])
++ goto out_oom;
++ }
++
++ spin_lock_irqsave(&pgd_lock, flags);
++
++ /* Protect against save/restore: move below 4GB under pgd_lock. */
++ if (!xen_feature(XENFEAT_pae_pgdir_above_4gb)) {
++ int rc = xen_create_contiguous_region(
++ (unsigned long)pgd, 0, 32);
++ if (rc) {
++ spin_unlock_irqrestore(&pgd_lock, flags);
++ goto out_oom;
++ }
++ }
++
++ /* Copy kernel pmd contents and write-protect the new pmds. */
++ for (i = USER_PTRS_PER_PGD; i < PTRS_PER_PGD; i++) {
++ unsigned long v = (unsigned long)i << PGDIR_SHIFT;
++ pgd_t *kpgd = pgd_offset_k(v);
++ pud_t *kpud = pud_offset(kpgd, v);
++ pmd_t *kpmd = pmd_offset(kpud, v);
++ memcpy(pmd[i], kpmd, PAGE_SIZE);
++ make_lowmem_page_readonly(
++ pmd[i], XENFEAT_writable_page_tables);
++ }
++
++ /* It is safe to poke machine addresses of pmds under the pmd_lock. */
++ for (i = 0; i < PTRS_PER_PGD; i++)
++ set_pgd(&pgd[i], __pgd(1 + __pa(pmd[i])));
++
++ /* Ensure this pgd gets picked up and pinned on save/restore. */
++ pgd_list_add(pgd);
++
++ spin_unlock_irqrestore(&pgd_lock, flags);
++
++ kfree(pmd);
++
++ return pgd;
++
++out_oom:
++ if (HAVE_SHARED_KERNEL_PMD) {
++ for (i--; i >= 0; i--)
++ kmem_cache_free(pmd_cache,
++ (void *)__va(pgd_val(pgd[i])-1));
++ } else {
++ for (i--; i >= 0; i--)
++ kmem_cache_free(pmd_cache, pmd[i]);
++ kfree(pmd);
++ }
++ kmem_cache_free(pgd_cache, pgd);
++ return NULL;
++}
++
++void pgd_free(pgd_t *pgd)
++{
++ int i;
++
++ /*
++ * After this the pgd should not be pinned for the duration of this
++ * function's execution. We should never sleep and thus never race:
++ * 1. User pmds will not become write-protected under our feet due
++ * to a concurrent mm_pin_all().
++ * 2. The machine addresses in PGD entries will not become invalid
++ * due to a concurrent save/restore.
++ */
++ pgd_test_and_unpin(pgd);
++
++ /* in the PAE case user pgd entries are overwritten before usage */
++ if (PTRS_PER_PMD > 1) {
++ for (i = 0; i < USER_PTRS_PER_PGD; ++i) {
++ pmd_t *pmd = (void *)__va(pgd_val(pgd[i])-1);
++ kmem_cache_free(pmd_cache, pmd);
++ }
++
++ if (!HAVE_SHARED_KERNEL_PMD) {
++ unsigned long flags;
++ spin_lock_irqsave(&pgd_lock, flags);
++ pgd_list_del(pgd);
++ spin_unlock_irqrestore(&pgd_lock, flags);
++
++ for (i = USER_PTRS_PER_PGD; i < PTRS_PER_PGD; i++) {
++ pmd_t *pmd = (void *)__va(pgd_val(pgd[i])-1);
++ make_lowmem_page_writable(
++ pmd, XENFEAT_writable_page_tables);
++ memset(pmd, 0, PTRS_PER_PMD*sizeof(pmd_t));
++ kmem_cache_free(pmd_cache, pmd);
++ }
++
++ if (!xen_feature(XENFEAT_pae_pgdir_above_4gb))
++ xen_destroy_contiguous_region(
++ (unsigned long)pgd, 0);
++ }
++ }
++
++ /* in the non-PAE case, free_pgtables() clears user pgd entries */
++ kmem_cache_free(pgd_cache, pgd);
++}
++
++void make_lowmem_page_readonly(void *va, unsigned int feature)
++{
++ pte_t *pte;
++ int rc;
++
++ if (xen_feature(feature))
++ return;
++
++ pte = virt_to_ptep(va);
++ rc = HYPERVISOR_update_va_mapping(
++ (unsigned long)va, pte_wrprotect(*pte), 0);
++ BUG_ON(rc);
++}
++
++void make_lowmem_page_writable(void *va, unsigned int feature)
++{
++ pte_t *pte;
++ int rc;
++
++ if (xen_feature(feature))
++ return;
++
++ pte = virt_to_ptep(va);
++ rc = HYPERVISOR_update_va_mapping(
++ (unsigned long)va, pte_mkwrite(*pte), 0);
++ BUG_ON(rc);
++}
++
++void make_page_readonly(void *va, unsigned int feature)
++{
++ pte_t *pte;
++ int rc;
++
++ if (xen_feature(feature))
++ return;
++
++ pte = virt_to_ptep(va);
++ rc = HYPERVISOR_update_va_mapping(
++ (unsigned long)va, pte_wrprotect(*pte), 0);
++ if (rc) /* fallback? */
++ xen_l1_entry_update(pte, pte_wrprotect(*pte));
++ if ((unsigned long)va >= (unsigned long)high_memory) {
++ unsigned long pfn = pte_pfn(*pte);
++#ifdef CONFIG_HIGHMEM
++ if (pfn >= highstart_pfn)
++ kmap_flush_unused(); /* flush stale writable kmaps */
++ else
++#endif
++ make_lowmem_page_readonly(
++ phys_to_virt(pfn << PAGE_SHIFT), feature);
++ }
++}
++
++void make_page_writable(void *va, unsigned int feature)
++{
++ pte_t *pte;
++ int rc;
++
++ if (xen_feature(feature))
++ return;
++
++ pte = virt_to_ptep(va);
++ rc = HYPERVISOR_update_va_mapping(
++ (unsigned long)va, pte_mkwrite(*pte), 0);
++ if (rc) /* fallback? */
++ xen_l1_entry_update(pte, pte_mkwrite(*pte));
++ if ((unsigned long)va >= (unsigned long)high_memory) {
++ unsigned long pfn = pte_pfn(*pte);
++#ifdef CONFIG_HIGHMEM
++ if (pfn < highstart_pfn)
++#endif
++ make_lowmem_page_writable(
++ phys_to_virt(pfn << PAGE_SHIFT), feature);
++ }
++}
++
++void make_pages_readonly(void *va, unsigned int nr, unsigned int feature)
++{
++ if (xen_feature(feature))
++ return;
++
++ while (nr-- != 0) {
++ make_page_readonly(va, feature);
++ va = (void *)((unsigned long)va + PAGE_SIZE);
++ }
++}
++
++void make_pages_writable(void *va, unsigned int nr, unsigned int feature)
++{
++ if (xen_feature(feature))
++ return;
++
++ while (nr-- != 0) {
++ make_page_writable(va, feature);
++ va = (void *)((unsigned long)va + PAGE_SIZE);
++ }
++}
++
++static inline void pgd_walk_set_prot(void *pt, pgprot_t flags)
++{
++ struct page *page = virt_to_page(pt);
++ unsigned long pfn = page_to_pfn(page);
++
++ if (PageHighMem(page))
++ return;
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__va(pfn << PAGE_SHIFT),
++ pfn_pte(pfn, flags), 0));
++}
++
++static void pgd_walk(pgd_t *pgd_base, pgprot_t flags)
++{
++ pgd_t *pgd = pgd_base;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++ int g, u, m;
++
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return;
++
++ for (g = 0; g < USER_PTRS_PER_PGD; g++, pgd++) {
++ if (pgd_none(*pgd))
++ continue;
++ pud = pud_offset(pgd, 0);
++ if (PTRS_PER_PUD > 1) /* not folded */
++ pgd_walk_set_prot(pud,flags);
++ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
++ if (pud_none(*pud))
++ continue;
++ pmd = pmd_offset(pud, 0);
++ if (PTRS_PER_PMD > 1) /* not folded */
++ pgd_walk_set_prot(pmd,flags);
++ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
++ if (pmd_none(*pmd))
++ continue;
++ pte = pte_offset_kernel(pmd,0);
++ pgd_walk_set_prot(pte,flags);
++ }
++ }
++ }
++
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pgd_base,
++ pfn_pte(virt_to_phys(pgd_base)>>PAGE_SHIFT, flags),
++ UVMF_TLB_FLUSH));
++}
++
++static void __pgd_pin(pgd_t *pgd)
++{
++ pgd_walk(pgd, PAGE_KERNEL_RO);
++ xen_pgd_pin(__pa(pgd));
++ set_bit(PG_pinned, &virt_to_page(pgd)->flags);
++}
++
++static void __pgd_unpin(pgd_t *pgd)
++{
++ xen_pgd_unpin(__pa(pgd));
++ pgd_walk(pgd, PAGE_KERNEL);
++ clear_bit(PG_pinned, &virt_to_page(pgd)->flags);
++}
++
++static void pgd_test_and_unpin(pgd_t *pgd)
++{
++ if (test_bit(PG_pinned, &virt_to_page(pgd)->flags))
++ __pgd_unpin(pgd);
++}
++
++void mm_pin(struct mm_struct *mm)
++{
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++ spin_lock(&mm->page_table_lock);
++ __pgd_pin(mm->pgd);
++ spin_unlock(&mm->page_table_lock);
++}
++
++void mm_unpin(struct mm_struct *mm)
++{
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++ spin_lock(&mm->page_table_lock);
++ __pgd_unpin(mm->pgd);
++ spin_unlock(&mm->page_table_lock);
++}
++
++void mm_pin_all(void)
++{
++ struct page *page;
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++ for (page = pgd_list; page; page = (struct page *)page->index) {
++ if (!test_bit(PG_pinned, &page->flags))
++ __pgd_pin((pgd_t *)page_address(page));
++ }
++}
++
++void _arch_dup_mmap(struct mm_struct *mm)
++{
++ if (!test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags))
++ mm_pin(mm);
++}
++
++void _arch_exit_mmap(struct mm_struct *mm)
++{
++ struct task_struct *tsk = current;
++
++ task_lock(tsk);
++
++ /*
++ * We aggressively remove defunct pgd from cr3. We execute unmap_vmas()
++ * *much* faster this way, as no tlb flushes means bigger wrpt batches.
++ */
++ if (tsk->active_mm == mm) {
++ tsk->active_mm = &init_mm;
++ atomic_inc(&init_mm.mm_count);
++
++ switch_mm(mm, &init_mm, tsk);
++
++ atomic_dec(&mm->mm_count);
++ BUG_ON(atomic_read(&mm->mm_count) == 0);
++ }
++
++ task_unlock(tsk);
++
++ if (test_bit(PG_pinned, &virt_to_page(mm->pgd)->flags) &&
++ (atomic_read(&mm->mm_count) == 1))
++ mm_unpin(mm);
++}
+diff -urNp linux-2.6/arch/i386/oprofile/Makefile new/arch/i386/oprofile/Makefile
+--- linux-2.6/arch/i386/oprofile/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/oprofile/Makefile 2006-05-09 12:32:37.000000000 +0200
+@@ -6,7 +6,11 @@ DRIVER_OBJS = $(addprefix ../../../drive
+ oprofilefs.o oprofile_stats.o \
+ timer_int.o )
+
++ifdef CONFIG_XEN
++oprofile-y := $(DRIVER_OBJS) xenoprof.o
++else
+ oprofile-y := $(DRIVER_OBJS) init.o backtrace.o
+ oprofile-$(CONFIG_X86_LOCAL_APIC) += nmi_int.o op_model_athlon.o \
+ op_model_ppro.o op_model_p4.o
+ oprofile-$(CONFIG_X86_IO_APIC) += nmi_timer_int.o
++endif
+diff -urNp linux-2.6/arch/i386/oprofile/xenoprof.c new/arch/i386/oprofile/xenoprof.c
+--- linux-2.6/arch/i386/oprofile/xenoprof.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/oprofile/xenoprof.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,542 @@
++/**
++ * @file xenoprof.c
++ *
++ * @remark Copyright 2002 OProfile authors
++ * @remark Read the file COPYING
++ *
++ * @author John Levon <levon@movementarian.org>
++ *
++ * Modified by Aravind Menon and Jose Renato Santos for Xen
++ * These modifications are:
++ * Copyright (C) 2005 Hewlett-Packard Co.
++ */
++
++#include <linux/init.h>
++#include <linux/notifier.h>
++#include <linux/smp.h>
++#include <linux/oprofile.h>
++#include <linux/sysdev.h>
++#include <linux/slab.h>
++#include <linux/interrupt.h>
++#include <linux/vmalloc.h>
++#include <asm/nmi.h>
++#include <asm/msr.h>
++#include <asm/apic.h>
++#include <asm/pgtable.h>
++#include <xen/evtchn.h>
++#include "op_counter.h"
++
++#include <xen/interface/xen.h>
++#include <xen/interface/xenoprof.h>
++#include <../../../drivers/oprofile/cpu_buffer.h>
++
++static int xenoprof_start(void);
++static void xenoprof_stop(void);
++
++void * vm_map_xen_pages(unsigned long maddr, int vm_size, pgprot_t prot);
++
++static int xenoprof_enabled = 0;
++static unsigned int num_events = 0;
++static int is_primary = 0;
++static int active_defined;
++
++/* sample buffers shared with Xen */
++xenoprof_buf_t * xenoprof_buf[MAX_VIRT_CPUS];
++/* Shared buffer area */
++char * shared_buffer;
++/* Number of buffers in shared area (one per VCPU) */
++int nbuf;
++/* Mappings of VIRQ_XENOPROF to irq number (per cpu) */
++int ovf_irq[NR_CPUS];
++/* cpu model type string - copied from Xen memory space on XENOPROF_init command */
++char cpu_type[XENOPROF_CPU_TYPE_SIZE];
++
++/* Passive sample buffers shared with Xen */
++xenoprof_buf_t *p_xenoprof_buf[MAX_OPROF_DOMAINS][MAX_VIRT_CPUS];
++/* Passive shared buffer area */
++char *p_shared_buffer[MAX_OPROF_DOMAINS];
++
++#ifdef CONFIG_PM
++
++static int xenoprof_suspend(struct sys_device * dev, pm_message_t state)
++{
++ if (xenoprof_enabled == 1)
++ xenoprof_stop();
++ return 0;
++}
++
++
++static int xenoprof_resume(struct sys_device * dev)
++{
++ if (xenoprof_enabled == 1)
++ xenoprof_start();
++ return 0;
++}
++
++
++static struct sysdev_class oprofile_sysclass = {
++ set_kset_name("oprofile"),
++ .resume = xenoprof_resume,
++ .suspend = xenoprof_suspend
++};
++
++
++static struct sys_device device_oprofile = {
++ .id = 0,
++ .cls = &oprofile_sysclass,
++};
++
++
++static int __init init_driverfs(void)
++{
++ int error;
++ if (!(error = sysdev_class_register(&oprofile_sysclass)))
++ error = sysdev_register(&device_oprofile);
++ return error;
++}
++
++
++static void __exit exit_driverfs(void)
++{
++ sysdev_unregister(&device_oprofile);
++ sysdev_class_unregister(&oprofile_sysclass);
++}
++
++#else
++#define init_driverfs() do { } while (0)
++#define exit_driverfs() do { } while (0)
++#endif /* CONFIG_PM */
++
++unsigned long long oprofile_samples = 0;
++unsigned long long p_oprofile_samples = 0;
++
++unsigned int pdomains;
++struct xenoprof_passive passive_domains[MAX_OPROF_DOMAINS];
++
++static void xenoprof_add_pc(xenoprof_buf_t *buf, int is_passive)
++{
++ int head, tail, size;
++
++ head = buf->event_head;
++ tail = buf->event_tail;
++ size = buf->event_size;
++
++ if (tail > head) {
++ while (tail < size) {
++ oprofile_add_pc(buf->event_log[tail].eip,
++ buf->event_log[tail].mode,
++ buf->event_log[tail].event);
++ if (!is_passive)
++ oprofile_samples++;
++ else
++ p_oprofile_samples++;
++ tail++;
++ }
++ tail = 0;
++ }
++ while (tail < head) {
++ oprofile_add_pc(buf->event_log[tail].eip,
++ buf->event_log[tail].mode,
++ buf->event_log[tail].event);
++ if (!is_passive)
++ oprofile_samples++;
++ else
++ p_oprofile_samples++;
++ tail++;
++ }
++
++ buf->event_tail = tail;
++}
++
++static void xenoprof_handle_passive(void)
++{
++ int i, j;
++
++ for (i = 0; i < pdomains; i++)
++ for (j = 0; j < passive_domains[i].nbuf; j++) {
++ xenoprof_buf_t *buf = p_xenoprof_buf[i][j];
++ if (buf->event_head == buf->event_tail)
++ continue;
++ oprofile_add_pc(IGNORED_PC, CPU_MODE_PASSIVE_START, passive_domains[i].domain_id);
++ xenoprof_add_pc(buf, 1);
++ oprofile_add_pc(IGNORED_PC, CPU_MODE_PASSIVE_STOP, passive_domains[i].domain_id);
++ }
++}
++
++static irqreturn_t
++xenoprof_ovf_interrupt(int irq, void * dev_id, struct pt_regs * regs)
++{
++ struct xenoprof_buf * buf;
++ int cpu;
++ static unsigned long flag;
++
++ cpu = smp_processor_id();
++ buf = xenoprof_buf[cpu];
++
++ xenoprof_add_pc(buf, 0);
++
++ if (is_primary && !test_and_set_bit(0, &flag)) {
++ xenoprof_handle_passive();
++ clear_bit(0, &flag);
++ }
++
++ return IRQ_HANDLED;
++}
++
++
++static void unbind_virq(void)
++{
++ int i;
++
++ for_each_cpu(i) {
++ if (ovf_irq[i] >= 0) {
++ unbind_from_irqhandler(ovf_irq[i], NULL);
++ ovf_irq[i] = -1;
++ }
++ }
++}
++
++
++static int bind_virq(void)
++{
++ int i, result;
++
++ for_each_cpu(i) {
++ result = bind_virq_to_irqhandler(VIRQ_XENOPROF,
++ i,
++ xenoprof_ovf_interrupt,
++ SA_INTERRUPT,
++ "xenoprof",
++ NULL);
++
++ if (result < 0) {
++ unbind_virq();
++ return result;
++ }
++
++ ovf_irq[i] = result;
++ }
++
++ return 0;
++}
++
++
++static int xenoprof_setup(void)
++{
++ int ret;
++ int i;
++
++ ret = bind_virq();
++ if (ret)
++ return ret;
++
++ if (is_primary) {
++ struct xenoprof_counter counter;
++
++ /* Define dom0 as an active domain if not done yet */
++ if (!active_defined) {
++ domid_t domid;
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_reset_active_list, NULL);
++ if (ret)
++ goto err;
++ domid = 0;
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_set_active, &domid);
++ if (ret)
++ goto err;
++ active_defined = 1;
++ }
++
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_reserve_counters, NULL);
++ if (ret)
++ goto err;
++ for (i=0; i<num_events; i++) {
++ counter.ind = i;
++ counter.count = (uint64_t)counter_config[i].count;
++ counter.enabled = (uint32_t)counter_config[i].enabled;
++ counter.event = (uint32_t)counter_config[i].event;
++ counter.kernel = (uint32_t)counter_config[i].kernel;
++ counter.user = (uint32_t)counter_config[i].user;
++ counter.unit_mask = (uint64_t)counter_config[i].unit_mask;
++ HYPERVISOR_xenoprof_op(XENOPROF_counter,
++ &counter);
++ }
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_setup_events, NULL);
++
++ if (ret)
++ goto err;
++ }
++
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_enable_virq, NULL);
++ if (ret)
++ goto err;
++
++ xenoprof_enabled = 1;
++ return 0;
++ err:
++ unbind_virq();
++ return ret;
++}
++
++
++static void xenoprof_shutdown(void)
++{
++ xenoprof_enabled = 0;
++
++ HYPERVISOR_xenoprof_op(XENOPROF_disable_virq, NULL);
++
++ if (is_primary) {
++ HYPERVISOR_xenoprof_op(XENOPROF_release_counters, NULL);
++ active_defined = 0;
++ }
++
++ unbind_virq();
++
++}
++
++
++static int xenoprof_start(void)
++{
++ int ret = 0;
++
++ if (is_primary)
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_start, NULL);
++
++ return ret;
++}
++
++
++static void xenoprof_stop(void)
++{
++ if (is_primary)
++ HYPERVISOR_xenoprof_op(XENOPROF_stop, NULL);
++}
++
++
++static int xenoprof_set_active(int * active_domains,
++ unsigned int adomains)
++{
++ int ret = 0;
++ int i;
++ int set_dom0 = 0;
++ domid_t domid;
++
++ if (!is_primary)
++ return 0;
++
++ if (adomains > MAX_OPROF_DOMAINS)
++ return -E2BIG;
++
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_reset_active_list, NULL);
++ if (ret)
++ return ret;
++
++ for (i=0; i<adomains; i++) {
++ domid = active_domains[i];
++ if (domid != active_domains[i]) {
++ ret = -EINVAL;
++ goto out;
++ }
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_set_active, &domid);
++ if (ret)
++ goto out;
++ if (active_domains[i] == 0)
++ set_dom0 = 1;
++ }
++ /* dom0 must always be active but may not be in the list */
++ if (!set_dom0) {
++ domid = 0;
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_set_active, &domid);
++ }
++
++out:
++ if (ret)
++ HYPERVISOR_xenoprof_op(XENOPROF_reset_active_list, NULL);
++ active_defined = !ret;
++ return ret;
++}
++
++static int xenoprof_set_passive(int * p_domains,
++ unsigned int pdoms)
++{
++ int ret;
++ int i, j;
++ int vm_size;
++ int npages;
++ struct xenoprof_buf *buf;
++ pgprot_t prot = __pgprot(_KERNPG_TABLE);
++
++ if (!is_primary)
++ return 0;
++
++ if (pdoms > MAX_OPROF_DOMAINS)
++ return -E2BIG;
++
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_reset_passive_list, NULL);
++ if (ret)
++ return ret;
++
++ for (i = 0; i < pdoms; i++) {
++ passive_domains[i].domain_id = p_domains[i];
++ passive_domains[i].max_samples = 2048;
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_set_passive, &passive_domains[i]);
++ if (ret)
++ return ret;
++
++ npages = (passive_domains[i].bufsize * passive_domains[i].nbuf - 1) / PAGE_SIZE + 1;
++ vm_size = npages * PAGE_SIZE;
++
++ p_shared_buffer[i] = (char *)vm_map_xen_pages(passive_domains[i].buf_maddr,
++ vm_size, prot);
++ if (!p_shared_buffer[i]) {
++ ret = -ENOMEM;
++ goto out;
++ }
++
++ for (j = 0; j < passive_domains[i].nbuf; j++) {
++ buf = (struct xenoprof_buf *)
++ &p_shared_buffer[i][j * passive_domains[i].bufsize];
++ BUG_ON(buf->vcpu_id >= MAX_VIRT_CPUS);
++ p_xenoprof_buf[i][buf->vcpu_id] = buf;
++ }
++
++ }
++
++ pdomains = pdoms;
++ return 0;
++
++out:
++ for (j = 0; j < i; j++) {
++ vunmap(p_shared_buffer[j]);
++ p_shared_buffer[j] = NULL;
++ }
++
++ return ret;
++}
++
++struct op_counter_config counter_config[OP_MAX_COUNTER];
++
++static int xenoprof_create_files(struct super_block * sb, struct dentry * root)
++{
++ unsigned int i;
++
++ for (i = 0; i < num_events; ++i) {
++ struct dentry * dir;
++ char buf[2];
++
++ snprintf(buf, 2, "%d", i);
++ dir = oprofilefs_mkdir(sb, root, buf);
++ oprofilefs_create_ulong(sb, dir, "enabled",
++ &counter_config[i].enabled);
++ oprofilefs_create_ulong(sb, dir, "event",
++ &counter_config[i].event);
++ oprofilefs_create_ulong(sb, dir, "count",
++ &counter_config[i].count);
++ oprofilefs_create_ulong(sb, dir, "unit_mask",
++ &counter_config[i].unit_mask);
++ oprofilefs_create_ulong(sb, dir, "kernel",
++ &counter_config[i].kernel);
++ oprofilefs_create_ulong(sb, dir, "user",
++ &counter_config[i].user);
++ }
++
++ return 0;
++}
++
++
++struct oprofile_operations xenoprof_ops = {
++ .create_files = xenoprof_create_files,
++ .set_active = xenoprof_set_active,
++ .set_passive = xenoprof_set_passive,
++ .setup = xenoprof_setup,
++ .shutdown = xenoprof_shutdown,
++ .start = xenoprof_start,
++ .stop = xenoprof_stop
++};
++
++
++/* in order to get driverfs right */
++static int using_xenoprof;
++
++int __init oprofile_arch_init(struct oprofile_operations * ops)
++{
++ struct xenoprof_init init;
++ struct xenoprof_buf * buf;
++ int vm_size;
++ int npages;
++ int ret;
++ int i;
++
++ init.max_samples = 16;
++ ret = HYPERVISOR_xenoprof_op(XENOPROF_init, &init);
++
++ if (!ret) {
++ pgprot_t prot = __pgprot(_KERNPG_TABLE);
++
++ num_events = init.num_events;
++ is_primary = init.is_primary;
++ nbuf = init.nbuf;
++
++ /* just in case - make sure we do not overflow event list
++ (i.e. counter_config list) */
++ if (num_events > OP_MAX_COUNTER)
++ num_events = OP_MAX_COUNTER;
++
++ npages = (init.bufsize * nbuf - 1) / PAGE_SIZE + 1;
++ vm_size = npages * PAGE_SIZE;
++
++ shared_buffer = (char *)vm_map_xen_pages(init.buf_maddr,
++ vm_size, prot);
++ if (!shared_buffer) {
++ ret = -ENOMEM;
++ goto out;
++ }
++
++ for (i=0; i< nbuf; i++) {
++ buf = (struct xenoprof_buf*)
++ &shared_buffer[i * init.bufsize];
++ BUG_ON(buf->vcpu_id >= MAX_VIRT_CPUS);
++ xenoprof_buf[buf->vcpu_id] = buf;
++ }
++
++ /* cpu_type is detected by Xen */
++ cpu_type[XENOPROF_CPU_TYPE_SIZE-1] = 0;
++ strncpy(cpu_type, init.cpu_type, XENOPROF_CPU_TYPE_SIZE - 1);
++ xenoprof_ops.cpu_type = cpu_type;
++
++ init_driverfs();
++ using_xenoprof = 1;
++ *ops = xenoprof_ops;
++
++ for (i=0; i<NR_CPUS; i++)
++ ovf_irq[i] = -1;
++
++ active_defined = 0;
++ }
++ out:
++ printk(KERN_INFO "oprofile_arch_init: ret %d, events %d, "
++ "is_primary %d\n", ret, num_events, is_primary);
++ return ret;
++}
++
++
++void __exit oprofile_arch_exit(void)
++{
++ int i;
++
++ if (using_xenoprof)
++ exit_driverfs();
++
++ if (shared_buffer) {
++ vunmap(shared_buffer);
++ shared_buffer = NULL;
++ }
++ if (is_primary) {
++ for (i = 0; i < pdomains; i++)
++ if (p_shared_buffer[i]) {
++ vunmap(p_shared_buffer[i]);
++ p_shared_buffer[i] = NULL;
++ }
++ HYPERVISOR_xenoprof_op(XENOPROF_shutdown, NULL);
++ }
++
++}
+diff -urNp linux-2.6/arch/i386/pci/irq-xen.c new/arch/i386/pci/irq-xen.c
+--- linux-2.6/arch/i386/pci/irq-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/pci/irq-xen.c 2006-05-23 18:37:09.000000000 +0200
+@@ -0,0 +1,1204 @@
++/*
++ * Low-Level PCI Support for PC -- Routing of Interrupts
++ *
++ * (c) 1999--2000 Martin Mares <mj@ucw.cz>
++ */
++
++#include <linux/config.h>
++#include <linux/types.h>
++#include <linux/kernel.h>
++#include <linux/pci.h>
++#include <linux/init.h>
++#include <linux/slab.h>
++#include <linux/interrupt.h>
++#include <linux/dmi.h>
++#include <asm/io.h>
++#include <asm/smp.h>
++#include <asm/io_apic.h>
++#include <linux/irq.h>
++#include <linux/acpi.h>
++
++#include "pci.h"
++
++#define PIRQ_SIGNATURE (('$' << 0) + ('P' << 8) + ('I' << 16) + ('R' << 24))
++#define PIRQ_VERSION 0x0100
++
++static int broken_hp_bios_irq9;
++static int acer_tm360_irqrouting;
++
++static struct irq_routing_table *pirq_table;
++
++static int pirq_enable_irq(struct pci_dev *dev);
++
++/*
++ * Never use: 0, 1, 2 (timer, keyboard, and cascade)
++ * Avoid using: 13, 14 and 15 (FP error and IDE).
++ * Penalize: 3, 4, 6, 7, 12 (known ISA uses: serial, floppy, parallel and mouse)
++ */
++unsigned int pcibios_irq_mask = 0xfff8;
++
++static int pirq_penalty[16] = {
++ 1000000, 1000000, 1000000, 1000, 1000, 0, 1000, 1000,
++ 0, 0, 0, 0, 1000, 100000, 100000, 100000
++};
++
++struct irq_router {
++ char *name;
++ u16 vendor, device;
++ int (*get)(struct pci_dev *router, struct pci_dev *dev, int pirq);
++ int (*set)(struct pci_dev *router, struct pci_dev *dev, int pirq, int new);
++};
++
++struct irq_router_handler {
++ u16 vendor;
++ int (*probe)(struct irq_router *r, struct pci_dev *router, u16 device);
++};
++
++int (*pcibios_enable_irq)(struct pci_dev *dev) = NULL;
++void (*pcibios_disable_irq)(struct pci_dev *dev) = NULL;
++
++/*
++ * Check passed address for the PCI IRQ Routing Table signature
++ * and perform checksum verification.
++ */
++
++static inline struct irq_routing_table * pirq_check_routing_table(u8 *addr)
++{
++ struct irq_routing_table *rt;
++ int i;
++ u8 sum;
++
++ rt = (struct irq_routing_table *) addr;
++ if (rt->signature != PIRQ_SIGNATURE ||
++ rt->version != PIRQ_VERSION ||
++ rt->size % 16 ||
++ rt->size < sizeof(struct irq_routing_table))
++ return NULL;
++ sum = 0;
++ for (i=0; i < rt->size; i++)
++ sum += addr[i];
++ if (!sum) {
++ DBG(KERN_DEBUG "PCI: Interrupt Routing Table found at 0x%p\n", rt);
++ return rt;
++ }
++ return NULL;
++}
++
++
++
++/*
++ * Search 0xf0000 -- 0xfffff for the PCI IRQ Routing Table.
++ */
++
++static struct irq_routing_table * __init pirq_find_routing_table(void)
++{
++ u8 *addr;
++ struct irq_routing_table *rt;
++
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++ if (pirq_table_addr) {
++ rt = pirq_check_routing_table((u8 *) isa_bus_to_virt(pirq_table_addr));
++ if (rt)
++ return rt;
++ printk(KERN_WARNING "PCI: PIRQ table NOT found at pirqaddr\n");
++ }
++ for(addr = (u8 *) isa_bus_to_virt(0xf0000); addr < (u8 *) isa_bus_to_virt(0x100000); addr += 16) {
++ rt = pirq_check_routing_table(addr);
++ if (rt)
++ return rt;
++ }
++#endif
++
++ return NULL;
++}
++
++/*
++ * If we have a IRQ routing table, use it to search for peer host
++ * bridges. It's a gross hack, but since there are no other known
++ * ways how to get a list of buses, we have to go this way.
++ */
++
++static void __init pirq_peer_trick(void)
++{
++ struct irq_routing_table *rt = pirq_table;
++ u8 busmap[256];
++ int i;
++ struct irq_info *e;
++
++ memset(busmap, 0, sizeof(busmap));
++ for(i=0; i < (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info); i++) {
++ e = &rt->slots[i];
++#ifdef DEBUG
++ {
++ int j;
++ DBG(KERN_DEBUG "%02x:%02x slot=%02x", e->bus, e->devfn/8, e->slot);
++ for(j=0; j<4; j++)
++ DBG(" %d:%02x/%04x", j, e->irq[j].link, e->irq[j].bitmap);
++ DBG("\n");
++ }
++#endif
++ busmap[e->bus] = 1;
++ }
++ for(i = 1; i < 256; i++) {
++ if (!busmap[i] || pci_find_bus(0, i))
++ continue;
++ if (pci_scan_bus(i, &pci_root_ops, NULL))
++ printk(KERN_INFO "PCI: Discovered primary peer bus %02x [IRQ]\n", i);
++ }
++ pcibios_last_bus = -1;
++}
++
++/*
++ * Code for querying and setting of IRQ routes on various interrupt routers.
++ */
++
++void eisa_set_level_irq(unsigned int irq)
++{
++ unsigned char mask = 1 << (irq & 7);
++ unsigned int port = 0x4d0 + (irq >> 3);
++ unsigned char val;
++ static u16 eisa_irq_mask;
++
++ if (irq >= 16 || (1 << irq) & eisa_irq_mask)
++ return;
++
++ eisa_irq_mask |= (1 << irq);
++ printk(KERN_DEBUG "PCI: setting IRQ %u as level-triggered\n", irq);
++ val = inb(port);
++ if (!(val & mask)) {
++ DBG(KERN_DEBUG " -> edge");
++ outb(val | mask, port);
++ }
++}
++
++/*
++ * Common IRQ routing practice: nybbles in config space,
++ * offset by some magic constant.
++ */
++static unsigned int read_config_nybble(struct pci_dev *router, unsigned offset, unsigned nr)
++{
++ u8 x;
++ unsigned reg = offset + (nr >> 1);
++
++ pci_read_config_byte(router, reg, &x);
++ return (nr & 1) ? (x >> 4) : (x & 0xf);
++}
++
++static void write_config_nybble(struct pci_dev *router, unsigned offset, unsigned nr, unsigned int val)
++{
++ u8 x;
++ unsigned reg = offset + (nr >> 1);
++
++ pci_read_config_byte(router, reg, &x);
++ x = (nr & 1) ? ((x & 0x0f) | (val << 4)) : ((x & 0xf0) | val);
++ pci_write_config_byte(router, reg, x);
++}
++
++/*
++ * ALI pirq entries are damn ugly, and completely undocumented.
++ * This has been figured out from pirq tables, and it's not a pretty
++ * picture.
++ */
++static int pirq_ali_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ static unsigned char irqmap[16] = { 0, 9, 3, 10, 4, 5, 7, 6, 1, 11, 0, 12, 0, 14, 0, 15 };
++
++ return irqmap[read_config_nybble(router, 0x48, pirq-1)];
++}
++
++static int pirq_ali_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ static unsigned char irqmap[16] = { 0, 8, 0, 2, 4, 5, 7, 6, 0, 1, 3, 9, 11, 0, 13, 15 };
++ unsigned int val = irqmap[irq];
++
++ if (val) {
++ write_config_nybble(router, 0x48, pirq-1, val);
++ return 1;
++ }
++ return 0;
++}
++
++/*
++ * The Intel PIIX4 pirq rules are fairly simple: "pirq" is
++ * just a pointer to the config space.
++ */
++static int pirq_piix_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ u8 x;
++
++ pci_read_config_byte(router, pirq, &x);
++ return (x < 16) ? x : 0;
++}
++
++static int pirq_piix_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ pci_write_config_byte(router, pirq, irq);
++ return 1;
++}
++
++/*
++ * The VIA pirq rules are nibble-based, like ALI,
++ * but without the ugly irq number munging.
++ * However, PIRQD is in the upper instead of lower 4 bits.
++ */
++static int pirq_via_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ return read_config_nybble(router, 0x55, pirq == 4 ? 5 : pirq);
++}
++
++static int pirq_via_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ write_config_nybble(router, 0x55, pirq == 4 ? 5 : pirq, irq);
++ return 1;
++}
++
++/*
++ * The VIA pirq rules are nibble-based, like ALI,
++ * but without the ugly irq number munging.
++ * However, for 82C586, nibble map is different .
++ */
++static int pirq_via586_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ static unsigned int pirqmap[4] = { 3, 2, 5, 1 };
++ return read_config_nybble(router, 0x55, pirqmap[pirq-1]);
++}
++
++static int pirq_via586_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ static unsigned int pirqmap[4] = { 3, 2, 5, 1 };
++ write_config_nybble(router, 0x55, pirqmap[pirq-1], irq);
++ return 1;
++}
++
++/*
++ * ITE 8330G pirq rules are nibble-based
++ * FIXME: pirqmap may be { 1, 0, 3, 2 },
++ * 2+3 are both mapped to irq 9 on my system
++ */
++static int pirq_ite_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ static unsigned char pirqmap[4] = { 1, 0, 2, 3 };
++ return read_config_nybble(router,0x43, pirqmap[pirq-1]);
++}
++
++static int pirq_ite_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ static unsigned char pirqmap[4] = { 1, 0, 2, 3 };
++ write_config_nybble(router, 0x43, pirqmap[pirq-1], irq);
++ return 1;
++}
++
++/*
++ * OPTI: high four bits are nibble pointer..
++ * I wonder what the low bits do?
++ */
++static int pirq_opti_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ return read_config_nybble(router, 0xb8, pirq >> 4);
++}
++
++static int pirq_opti_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ write_config_nybble(router, 0xb8, pirq >> 4, irq);
++ return 1;
++}
++
++/*
++ * Cyrix: nibble offset 0x5C
++ * 0x5C bits 7:4 is INTB bits 3:0 is INTA
++ * 0x5D bits 7:4 is INTD bits 3:0 is INTC
++ */
++static int pirq_cyrix_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ return read_config_nybble(router, 0x5C, (pirq-1)^1);
++}
++
++static int pirq_cyrix_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ write_config_nybble(router, 0x5C, (pirq-1)^1, irq);
++ return 1;
++}
++
++/*
++ * PIRQ routing for SiS 85C503 router used in several SiS chipsets.
++ * We have to deal with the following issues here:
++ * - vendors have different ideas about the meaning of link values
++ * - some onboard devices (integrated in the chipset) have special
++ * links and are thus routed differently (i.e. not via PCI INTA-INTD)
++ * - different revision of the router have a different layout for
++ * the routing registers, particularly for the onchip devices
++ *
++ * For all routing registers the common thing is we have one byte
++ * per routeable link which is defined as:
++ * bit 7 IRQ mapping enabled (0) or disabled (1)
++ * bits [6:4] reserved (sometimes used for onchip devices)
++ * bits [3:0] IRQ to map to
++ * allowed: 3-7, 9-12, 14-15
++ * reserved: 0, 1, 2, 8, 13
++ *
++ * The config-space registers located at 0x41/0x42/0x43/0x44 are
++ * always used to route the normal PCI INT A/B/C/D respectively.
++ * Apparently there are systems implementing PCI routing table using
++ * link values 0x01-0x04 and others using 0x41-0x44 for PCI INTA..D.
++ * We try our best to handle both link mappings.
++ *
++ * Currently (2003-05-21) it appears most SiS chipsets follow the
++ * definition of routing registers from the SiS-5595 southbridge.
++ * According to the SiS 5595 datasheets the revision id's of the
++ * router (ISA-bridge) should be 0x01 or 0xb0.
++ *
++ * Furthermore we've also seen lspci dumps with revision 0x00 and 0xb1.
++ * Looks like these are used in a number of SiS 5xx/6xx/7xx chipsets.
++ * They seem to work with the current routing code. However there is
++ * some concern because of the two USB-OHCI HCs (original SiS 5595
++ * had only one). YMMV.
++ *
++ * Onchip routing for router rev-id 0x01/0xb0 and probably 0x00/0xb1:
++ *
++ * 0x61: IDEIRQ:
++ * bits [6:5] must be written 01
++ * bit 4 channel-select primary (0), secondary (1)
++ *
++ * 0x62: USBIRQ:
++ * bit 6 OHCI function disabled (0), enabled (1)
++ *
++ * 0x6a: ACPI/SCI IRQ: bits 4-6 reserved
++ *
++ * 0x7e: Data Acq. Module IRQ - bits 4-6 reserved
++ *
++ * We support USBIRQ (in addition to INTA-INTD) and keep the
++ * IDE, ACPI and DAQ routing untouched as set by the BIOS.
++ *
++ * Currently the only reported exception is the new SiS 65x chipset
++ * which includes the SiS 69x southbridge. Here we have the 85C503
++ * router revision 0x04 and there are changes in the register layout
++ * mostly related to the different USB HCs with USB 2.0 support.
++ *
++ * Onchip routing for router rev-id 0x04 (try-and-error observation)
++ *
++ * 0x60/0x61/0x62/0x63: 1xEHCI and 3xOHCI (companion) USB-HCs
++ * bit 6-4 are probably unused, not like 5595
++ */
++
++#define PIRQ_SIS_IRQ_MASK 0x0f
++#define PIRQ_SIS_IRQ_DISABLE 0x80
++#define PIRQ_SIS_USB_ENABLE 0x40
++
++static int pirq_sis_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ u8 x;
++ int reg;
++
++ reg = pirq;
++ if (reg >= 0x01 && reg <= 0x04)
++ reg += 0x40;
++ pci_read_config_byte(router, reg, &x);
++ return (x & PIRQ_SIS_IRQ_DISABLE) ? 0 : (x & PIRQ_SIS_IRQ_MASK);
++}
++
++static int pirq_sis_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ u8 x;
++ int reg;
++
++ reg = pirq;
++ if (reg >= 0x01 && reg <= 0x04)
++ reg += 0x40;
++ pci_read_config_byte(router, reg, &x);
++ x &= ~(PIRQ_SIS_IRQ_MASK | PIRQ_SIS_IRQ_DISABLE);
++ x |= irq ? irq: PIRQ_SIS_IRQ_DISABLE;
++ pci_write_config_byte(router, reg, x);
++ return 1;
++}
++
++
++/*
++ * VLSI: nibble offset 0x74 - educated guess due to routing table and
++ * config space of VLSI 82C534 PCI-bridge/router (1004:0102)
++ * Tested on HP OmniBook 800 covering PIRQ 1, 2, 4, 8 for onboard
++ * devices, PIRQ 3 for non-pci(!) soundchip and (untested) PIRQ 6
++ * for the busbridge to the docking station.
++ */
++
++static int pirq_vlsi_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ if (pirq > 8) {
++ printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
++ return 0;
++ }
++ return read_config_nybble(router, 0x74, pirq-1);
++}
++
++static int pirq_vlsi_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ if (pirq > 8) {
++ printk(KERN_INFO "VLSI router pirq escape (%d)\n", pirq);
++ return 0;
++ }
++ write_config_nybble(router, 0x74, pirq-1, irq);
++ return 1;
++}
++
++/*
++ * ServerWorks: PCI interrupts mapped to system IRQ lines through Index
++ * and Redirect I/O registers (0x0c00 and 0x0c01). The Index register
++ * format is (PCIIRQ## | 0x10), e.g.: PCIIRQ10=0x1a. The Redirect
++ * register is a straight binary coding of desired PIC IRQ (low nibble).
++ *
++ * The 'link' value in the PIRQ table is already in the correct format
++ * for the Index register. There are some special index values:
++ * 0x00 for ACPI (SCI), 0x01 for USB, 0x02 for IDE0, 0x04 for IDE1,
++ * and 0x03 for SMBus.
++ */
++static int pirq_serverworks_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ outb_p(pirq, 0xc00);
++ return inb(0xc01) & 0xf;
++}
++
++static int pirq_serverworks_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ outb_p(pirq, 0xc00);
++ outb_p(irq, 0xc01);
++ return 1;
++}
++
++/* Support for AMD756 PCI IRQ Routing
++ * Jhon H. Caicedo <jhcaiced@osso.org.co>
++ * Jun/21/2001 0.2.0 Release, fixed to use "nybble" functions... (jhcaiced)
++ * Jun/19/2001 Alpha Release 0.1.0 (jhcaiced)
++ * The AMD756 pirq rules are nibble-based
++ * offset 0x56 0-3 PIRQA 4-7 PIRQB
++ * offset 0x57 0-3 PIRQC 4-7 PIRQD
++ */
++static int pirq_amd756_get(struct pci_dev *router, struct pci_dev *dev, int pirq)
++{
++ u8 irq;
++ irq = 0;
++ if (pirq <= 4)
++ {
++ irq = read_config_nybble(router, 0x56, pirq - 1);
++ }
++ printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d get irq : %2d\n",
++ dev->vendor, dev->device, pirq, irq);
++ return irq;
++}
++
++static int pirq_amd756_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ printk(KERN_INFO "AMD756: dev %04x:%04x, router pirq : %d SET irq : %2d\n",
++ dev->vendor, dev->device, pirq, irq);
++ if (pirq <= 4)
++ {
++ write_config_nybble(router, 0x56, pirq - 1, irq);
++ }
++ return 1;
++}
++
++#ifdef CONFIG_PCI_BIOS
++
++static int pirq_bios_set(struct pci_dev *router, struct pci_dev *dev, int pirq, int irq)
++{
++ struct pci_dev *bridge;
++ int pin = pci_get_interrupt_pin(dev, &bridge);
++ return pcibios_set_irq_routing(bridge, pin, irq);
++}
++
++#endif
++
++static __init int intel_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ static struct pci_device_id pirq_440gx[] = {
++ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_0) },
++ { PCI_DEVICE(PCI_VENDOR_ID_INTEL, PCI_DEVICE_ID_INTEL_82443GX_2) },
++ { },
++ };
++
++ /* 440GX has a proprietary PIRQ router -- don't use it */
++ if (pci_dev_present(pirq_440gx))
++ return 0;
++
++ switch(device)
++ {
++ case PCI_DEVICE_ID_INTEL_82371FB_0:
++ case PCI_DEVICE_ID_INTEL_82371SB_0:
++ case PCI_DEVICE_ID_INTEL_82371AB_0:
++ case PCI_DEVICE_ID_INTEL_82371MX:
++ case PCI_DEVICE_ID_INTEL_82443MX_0:
++ case PCI_DEVICE_ID_INTEL_82801AA_0:
++ case PCI_DEVICE_ID_INTEL_82801AB_0:
++ case PCI_DEVICE_ID_INTEL_82801BA_0:
++ case PCI_DEVICE_ID_INTEL_82801BA_10:
++ case PCI_DEVICE_ID_INTEL_82801CA_0:
++ case PCI_DEVICE_ID_INTEL_82801CA_12:
++ case PCI_DEVICE_ID_INTEL_82801DB_0:
++ case PCI_DEVICE_ID_INTEL_82801E_0:
++ case PCI_DEVICE_ID_INTEL_82801EB_0:
++ case PCI_DEVICE_ID_INTEL_ESB_1:
++ case PCI_DEVICE_ID_INTEL_ICH6_0:
++ case PCI_DEVICE_ID_INTEL_ICH6_1:
++ case PCI_DEVICE_ID_INTEL_ICH7_0:
++ case PCI_DEVICE_ID_INTEL_ICH7_1:
++ case PCI_DEVICE_ID_INTEL_ICH7_30:
++ case PCI_DEVICE_ID_INTEL_ICH7_31:
++ case PCI_DEVICE_ID_INTEL_ESB2_0:
++ case PCI_DEVICE_ID_INTEL_ICH8_0:
++ case PCI_DEVICE_ID_INTEL_ICH8_1:
++ case PCI_DEVICE_ID_INTEL_ICH8_2:
++ case PCI_DEVICE_ID_INTEL_ICH8_3:
++ case PCI_DEVICE_ID_INTEL_ICH8_4:
++ r->name = "PIIX/ICH";
++ r->get = pirq_piix_get;
++ r->set = pirq_piix_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int via_router_probe(struct irq_router *r,
++ struct pci_dev *router, u16 device)
++{
++ /* FIXME: We should move some of the quirk fixup stuff here */
++
++ /*
++ * work arounds for some buggy BIOSes
++ */
++ if (device == PCI_DEVICE_ID_VIA_82C586_0) {
++ switch(router->device) {
++ case PCI_DEVICE_ID_VIA_82C686:
++ /*
++ * Asus k7m bios wrongly reports 82C686A
++ * as 586-compatible
++ */
++ device = PCI_DEVICE_ID_VIA_82C686;
++ break;
++ case PCI_DEVICE_ID_VIA_8235:
++ /**
++ * Asus a7v-x bios wrongly reports 8235
++ * as 586-compatible
++ */
++ device = PCI_DEVICE_ID_VIA_8235;
++ break;
++ }
++ }
++
++ switch(device) {
++ case PCI_DEVICE_ID_VIA_82C586_0:
++ r->name = "VIA";
++ r->get = pirq_via586_get;
++ r->set = pirq_via586_set;
++ return 1;
++ case PCI_DEVICE_ID_VIA_82C596:
++ case PCI_DEVICE_ID_VIA_82C686:
++ case PCI_DEVICE_ID_VIA_8231:
++ case PCI_DEVICE_ID_VIA_8233A:
++ case PCI_DEVICE_ID_VIA_8235:
++ case PCI_DEVICE_ID_VIA_8237:
++ /* FIXME: add new ones for 8233/5 */
++ r->name = "VIA";
++ r->get = pirq_via_get;
++ r->set = pirq_via_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int vlsi_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_VLSI_82C534:
++ r->name = "VLSI 82C534";
++ r->get = pirq_vlsi_get;
++ r->set = pirq_vlsi_set;
++ return 1;
++ }
++ return 0;
++}
++
++
++static __init int serverworks_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_SERVERWORKS_OSB4:
++ case PCI_DEVICE_ID_SERVERWORKS_CSB5:
++ r->name = "ServerWorks";
++ r->get = pirq_serverworks_get;
++ r->set = pirq_serverworks_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int sis_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ if (device != PCI_DEVICE_ID_SI_503)
++ return 0;
++
++ r->name = "SIS";
++ r->get = pirq_sis_get;
++ r->set = pirq_sis_set;
++ return 1;
++}
++
++static __init int cyrix_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_CYRIX_5520:
++ r->name = "NatSemi";
++ r->get = pirq_cyrix_get;
++ r->set = pirq_cyrix_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int opti_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_OPTI_82C700:
++ r->name = "OPTI";
++ r->get = pirq_opti_get;
++ r->set = pirq_opti_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int ite_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_ITE_IT8330G_0:
++ r->name = "ITE";
++ r->get = pirq_ite_get;
++ r->set = pirq_ite_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int ali_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_AL_M1533:
++ case PCI_DEVICE_ID_AL_M1563:
++ printk(KERN_DEBUG "PCI: Using ALI IRQ Router\n");
++ r->name = "ALI";
++ r->get = pirq_ali_get;
++ r->set = pirq_ali_set;
++ return 1;
++ }
++ return 0;
++}
++
++static __init int amd_router_probe(struct irq_router *r, struct pci_dev *router, u16 device)
++{
++ switch(device)
++ {
++ case PCI_DEVICE_ID_AMD_VIPER_740B:
++ r->name = "AMD756";
++ break;
++ case PCI_DEVICE_ID_AMD_VIPER_7413:
++ r->name = "AMD766";
++ break;
++ case PCI_DEVICE_ID_AMD_VIPER_7443:
++ r->name = "AMD768";
++ break;
++ default:
++ return 0;
++ }
++ r->get = pirq_amd756_get;
++ r->set = pirq_amd756_set;
++ return 1;
++}
++
++static __initdata struct irq_router_handler pirq_routers[] = {
++ { PCI_VENDOR_ID_INTEL, intel_router_probe },
++ { PCI_VENDOR_ID_AL, ali_router_probe },
++ { PCI_VENDOR_ID_ITE, ite_router_probe },
++ { PCI_VENDOR_ID_VIA, via_router_probe },
++ { PCI_VENDOR_ID_OPTI, opti_router_probe },
++ { PCI_VENDOR_ID_SI, sis_router_probe },
++ { PCI_VENDOR_ID_CYRIX, cyrix_router_probe },
++ { PCI_VENDOR_ID_VLSI, vlsi_router_probe },
++ { PCI_VENDOR_ID_SERVERWORKS, serverworks_router_probe },
++ { PCI_VENDOR_ID_AMD, amd_router_probe },
++ /* Someone with docs needs to add the ATI Radeon IGP */
++ { 0, NULL }
++};
++static struct irq_router pirq_router;
++static struct pci_dev *pirq_router_dev;
++
++
++/*
++ * FIXME: should we have an option to say "generic for
++ * chipset" ?
++ */
++
++static void __init pirq_find_router(struct irq_router *r)
++{
++ struct irq_routing_table *rt = pirq_table;
++ struct irq_router_handler *h;
++
++#ifdef CONFIG_PCI_BIOS
++ if (!rt->signature) {
++ printk(KERN_INFO "PCI: Using BIOS for IRQ routing\n");
++ r->set = pirq_bios_set;
++ r->name = "BIOS";
++ return;
++ }
++#endif
++
++ /* Default unless a driver reloads it */
++ r->name = "default";
++ r->get = NULL;
++ r->set = NULL;
++
++ DBG(KERN_DEBUG "PCI: Attempting to find IRQ router for %04x:%04x\n",
++ rt->rtr_vendor, rt->rtr_device);
++
++ pirq_router_dev = pci_find_slot(rt->rtr_bus, rt->rtr_devfn);
++ if (!pirq_router_dev) {
++ DBG(KERN_DEBUG "PCI: Interrupt router not found at "
++ "%02x:%02x\n", rt->rtr_bus, rt->rtr_devfn);
++ return;
++ }
++
++ for( h = pirq_routers; h->vendor; h++) {
++ /* First look for a router match */
++ if (rt->rtr_vendor == h->vendor && h->probe(r, pirq_router_dev, rt->rtr_device))
++ break;
++ /* Fall back to a device match */
++ if (pirq_router_dev->vendor == h->vendor && h->probe(r, pirq_router_dev, pirq_router_dev->device))
++ break;
++ }
++ printk(KERN_INFO "PCI: Using IRQ router %s [%04x/%04x] at %s\n",
++ pirq_router.name,
++ pirq_router_dev->vendor,
++ pirq_router_dev->device,
++ pci_name(pirq_router_dev));
++}
++
++static struct irq_info *pirq_get_info(struct pci_dev *dev)
++{
++ struct irq_routing_table *rt = pirq_table;
++ int entries = (rt->size - sizeof(struct irq_routing_table)) / sizeof(struct irq_info);
++ struct irq_info *info;
++
++ for (info = rt->slots; entries--; info++)
++ if (info->bus == dev->bus->number && PCI_SLOT(info->devfn) == PCI_SLOT(dev->devfn))
++ return info;
++ return NULL;
++}
++
++static int pcibios_lookup_irq(struct pci_dev *dev, int assign)
++{
++ u8 pin;
++ struct irq_info *info;
++ int i, pirq, newirq;
++ int irq = 0;
++ u32 mask;
++ struct irq_router *r = &pirq_router;
++ struct pci_dev *dev2 = NULL;
++ char *msg = NULL;
++
++ /* Find IRQ pin */
++ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
++ if (!pin) {
++ DBG(KERN_DEBUG " -> no interrupt pin\n");
++ return 0;
++ }
++ pin = pin - 1;
++
++ /* Find IRQ routing entry */
++
++ if (!pirq_table)
++ return 0;
++
++ DBG(KERN_DEBUG "IRQ for %s[%c]", pci_name(dev), 'A' + pin);
++ info = pirq_get_info(dev);
++ if (!info) {
++ DBG(" -> not found in routing table\n" KERN_DEBUG);
++ return 0;
++ }
++ pirq = info->irq[pin].link;
++ mask = info->irq[pin].bitmap;
++ if (!pirq) {
++ DBG(" -> not routed\n" KERN_DEBUG);
++ return 0;
++ }
++ DBG(" -> PIRQ %02x, mask %04x, excl %04x", pirq, mask, pirq_table->exclusive_irqs);
++ mask &= pcibios_irq_mask;
++
++ /* Work around broken HP Pavilion Notebooks which assign USB to
++ IRQ 9 even though it is actually wired to IRQ 11 */
++
++ if (broken_hp_bios_irq9 && pirq == 0x59 && dev->irq == 9) {
++ dev->irq = 11;
++ pci_write_config_byte(dev, PCI_INTERRUPT_LINE, 11);
++ r->set(pirq_router_dev, dev, pirq, 11);
++ }
++
++ /* same for Acer Travelmate 360, but with CB and irq 11 -> 10 */
++ if (acer_tm360_irqrouting && dev->irq == 11 && dev->vendor == PCI_VENDOR_ID_O2) {
++ pirq = 0x68;
++ mask = 0x400;
++ dev->irq = r->get(pirq_router_dev, dev, pirq);
++ pci_write_config_byte(dev, PCI_INTERRUPT_LINE, dev->irq);
++ }
++
++ /*
++ * Find the best IRQ to assign: use the one
++ * reported by the device if possible.
++ */
++ newirq = dev->irq;
++ if (newirq && !((1 << newirq) & mask)) {
++ if ( pci_probe & PCI_USE_PIRQ_MASK) newirq = 0;
++ else printk("\n" KERN_WARNING
++ "PCI: IRQ %i for device %s doesn't match PIRQ mask "
++ "- try pci=usepirqmask\n" KERN_DEBUG, newirq,
++ pci_name(dev));
++ }
++ if (!newirq && assign) {
++ for (i = 0; i < 16; i++) {
++ if (!(mask & (1 << i)))
++ continue;
++ if (pirq_penalty[i] < pirq_penalty[newirq] && can_request_irq(i, SA_SHIRQ))
++ newirq = i;
++ }
++ }
++ DBG(" -> newirq=%d", newirq);
++
++ /* Check if it is hardcoded */
++ if ((pirq & 0xf0) == 0xf0) {
++ irq = pirq & 0xf;
++ DBG(" -> hardcoded IRQ %d\n", irq);
++ msg = "Hardcoded";
++ } else if ( r->get && (irq = r->get(pirq_router_dev, dev, pirq)) && \
++ ((!(pci_probe & PCI_USE_PIRQ_MASK)) || ((1 << irq) & mask)) ) {
++ DBG(" -> got IRQ %d\n", irq);
++ msg = "Found";
++ } else if (newirq && r->set && (dev->class >> 8) != PCI_CLASS_DISPLAY_VGA) {
++ DBG(" -> assigning IRQ %d", newirq);
++ if (r->set(pirq_router_dev, dev, pirq, newirq)) {
++ eisa_set_level_irq(newirq);
++ DBG(" ... OK\n");
++ msg = "Assigned";
++ irq = newirq;
++ }
++ }
++
++ if (!irq) {
++ DBG(" ... failed\n");
++ if (newirq && mask == (1 << newirq)) {
++ msg = "Guessed";
++ irq = newirq;
++ } else
++ return 0;
++ }
++ printk(KERN_INFO "PCI: %s IRQ %d for device %s\n", msg, irq, pci_name(dev));
++
++ /* Update IRQ for all devices with the same pirq value */
++ while ((dev2 = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev2)) != NULL) {
++ pci_read_config_byte(dev2, PCI_INTERRUPT_PIN, &pin);
++ if (!pin)
++ continue;
++ pin--;
++ info = pirq_get_info(dev2);
++ if (!info)
++ continue;
++ if (info->irq[pin].link == pirq) {
++ /* We refuse to override the dev->irq information. Give a warning! */
++ if ( dev2->irq && dev2->irq != irq && \
++ (!(pci_probe & PCI_USE_PIRQ_MASK) || \
++ ((1 << dev2->irq) & mask)) ) {
++#ifndef CONFIG_PCI_MSI
++ printk(KERN_INFO "IRQ routing conflict for %s, have irq %d, want irq %d\n",
++ pci_name(dev2), dev2->irq, irq);
++#endif
++ continue;
++ }
++ dev2->irq = irq;
++ pirq_penalty[irq]++;
++ if (dev != dev2)
++ printk(KERN_INFO "PCI: Sharing IRQ %d with %s\n", irq, pci_name(dev2));
++ }
++ }
++ return 1;
++}
++
++static void __init pcibios_fixup_irqs(void)
++{
++ struct pci_dev *dev = NULL;
++ u8 pin;
++
++ DBG(KERN_DEBUG "PCI: IRQ fixup\n");
++ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
++ /*
++ * If the BIOS has set an out of range IRQ number, just ignore it.
++ * Also keep track of which IRQ's are already in use.
++ */
++ if (dev->irq >= 16) {
++ DBG(KERN_DEBUG "%s: ignoring bogus IRQ %d\n", pci_name(dev), dev->irq);
++ dev->irq = 0;
++ }
++ /* If the IRQ is already assigned to a PCI device, ignore its ISA use penalty */
++ if (pirq_penalty[dev->irq] >= 100 && pirq_penalty[dev->irq] < 100000)
++ pirq_penalty[dev->irq] = 0;
++ pirq_penalty[dev->irq]++;
++ }
++
++ dev = NULL;
++ while ((dev = pci_get_device(PCI_ANY_ID, PCI_ANY_ID, dev)) != NULL) {
++ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
++#ifdef CONFIG_X86_IO_APIC
++ /*
++ * Recalculate IRQ numbers if we use the I/O APIC.
++ */
++ if (io_apic_assign_pci_irqs)
++ {
++ int irq;
++
++ if (pin) {
++ pin--; /* interrupt pins are numbered starting from 1 */
++ irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, PCI_SLOT(dev->devfn), pin);
++ /*
++ * Busses behind bridges are typically not listed in the MP-table.
++ * In this case we have to look up the IRQ based on the parent bus,
++ * parent slot, and pin number. The SMP code detects such bridged
++ * busses itself so we should get into this branch reliably.
++ */
++ if (irq < 0 && dev->bus->parent) { /* go back to the bridge */
++ struct pci_dev * bridge = dev->bus->self;
++
++ pin = (pin + PCI_SLOT(dev->devfn)) % 4;
++ irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
++ PCI_SLOT(bridge->devfn), pin);
++ if (irq >= 0)
++ printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
++ pci_name(bridge), 'A' + pin, irq);
++ }
++ if (irq >= 0) {
++ if (use_pci_vector() &&
++ !platform_legacy_irq(irq))
++ irq = IO_APIC_VECTOR(irq);
++
++ printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
++ pci_name(dev), 'A' + pin, irq);
++ dev->irq = irq;
++ }
++ }
++ }
++#endif
++ /*
++ * Still no IRQ? Try to lookup one...
++ */
++ if (pin && !dev->irq)
++ pcibios_lookup_irq(dev, 0);
++ }
++}
++
++/*
++ * Work around broken HP Pavilion Notebooks which assign USB to
++ * IRQ 9 even though it is actually wired to IRQ 11
++ */
++static int __init fix_broken_hp_bios_irq9(struct dmi_system_id *d)
++{
++ if (!broken_hp_bios_irq9) {
++ broken_hp_bios_irq9 = 1;
++ printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
++ }
++ return 0;
++}
++
++/*
++ * Work around broken Acer TravelMate 360 Notebooks which assign
++ * Cardbus to IRQ 11 even though it is actually wired to IRQ 10
++ */
++static int __init fix_acer_tm360_irqrouting(struct dmi_system_id *d)
++{
++ if (!acer_tm360_irqrouting) {
++ acer_tm360_irqrouting = 1;
++ printk(KERN_INFO "%s detected - fixing broken IRQ routing\n", d->ident);
++ }
++ return 0;
++}
++
++static struct dmi_system_id __initdata pciirq_dmi_table[] = {
++ {
++ .callback = fix_broken_hp_bios_irq9,
++ .ident = "HP Pavilion N5400 Series Laptop",
++ .matches = {
++ DMI_MATCH(DMI_SYS_VENDOR, "Hewlett-Packard"),
++ DMI_MATCH(DMI_BIOS_VERSION, "GE.M1.03"),
++ DMI_MATCH(DMI_PRODUCT_VERSION, "HP Pavilion Notebook Model GE"),
++ DMI_MATCH(DMI_BOARD_VERSION, "OmniBook N32N-736"),
++ },
++ },
++ {
++ .callback = fix_acer_tm360_irqrouting,
++ .ident = "Acer TravelMate 36x Laptop",
++ .matches = {
++ DMI_MATCH(DMI_SYS_VENDOR, "Acer"),
++ DMI_MATCH(DMI_PRODUCT_NAME, "TravelMate 360"),
++ },
++ },
++ { }
++};
++
++static int __init pcibios_irq_init(void)
++{
++ DBG(KERN_DEBUG "PCI: IRQ init\n");
++
++ if (pcibios_enable_irq || raw_pci_ops == NULL)
++ return 0;
++
++ dmi_check_system(pciirq_dmi_table);
++
++ pirq_table = pirq_find_routing_table();
++
++#ifdef CONFIG_PCI_BIOS
++ if (!pirq_table && (pci_probe & PCI_BIOS_IRQ_SCAN))
++ pirq_table = pcibios_get_irq_routing_table();
++#endif
++ if (pirq_table) {
++ pirq_peer_trick();
++ pirq_find_router(&pirq_router);
++ if (pirq_table->exclusive_irqs) {
++ int i;
++ for (i=0; i<16; i++)
++ if (!(pirq_table->exclusive_irqs & (1 << i)))
++ pirq_penalty[i] += 100;
++ }
++ /* If we're using the I/O APIC, avoid using the PCI IRQ routing table */
++ if (io_apic_assign_pci_irqs)
++ pirq_table = NULL;
++ }
++
++ pcibios_enable_irq = pirq_enable_irq;
++
++ pcibios_fixup_irqs();
++ return 0;
++}
++
++subsys_initcall(pcibios_irq_init);
++
++
++static void pirq_penalize_isa_irq(int irq, int active)
++{
++ /*
++ * If any ISAPnP device reports an IRQ in its list of possible
++ * IRQ's, we try to avoid assigning it to PCI devices.
++ */
++ if (irq < 16) {
++ if (active)
++ pirq_penalty[irq] += 1000;
++ else
++ pirq_penalty[irq] += 100;
++ }
++}
++
++void pcibios_penalize_isa_irq(int irq, int active)
++{
++#ifdef CONFIG_ACPI
++ if (!acpi_noirq)
++ acpi_penalize_isa_irq(irq, active);
++ else
++#endif
++ pirq_penalize_isa_irq(irq, active);
++}
++
++static int pirq_enable_irq(struct pci_dev *dev)
++{
++ u8 pin;
++ struct pci_dev *temp_dev;
++
++ pci_read_config_byte(dev, PCI_INTERRUPT_PIN, &pin);
++ if (pin && !pcibios_lookup_irq(dev, 1) && !dev->irq) {
++ char *msg = "";
++
++ pin--; /* interrupt pins are numbered starting from 1 */
++
++ if (io_apic_assign_pci_irqs) {
++ int irq;
++
++ irq = IO_APIC_get_PCI_irq_vector(dev->bus->number, PCI_SLOT(dev->devfn), pin);
++ /*
++ * Busses behind bridges are typically not listed in the MP-table.
++ * In this case we have to look up the IRQ based on the parent bus,
++ * parent slot, and pin number. The SMP code detects such bridged
++ * busses itself so we should get into this branch reliably.
++ */
++ temp_dev = dev;
++ while (irq < 0 && dev->bus->parent) { /* go back to the bridge */
++ struct pci_dev * bridge = dev->bus->self;
++
++ pin = (pin + PCI_SLOT(dev->devfn)) % 4;
++ irq = IO_APIC_get_PCI_irq_vector(bridge->bus->number,
++ PCI_SLOT(bridge->devfn), pin);
++ if (irq >= 0)
++ printk(KERN_WARNING "PCI: using PPB %s[%c] to get irq %d\n",
++ pci_name(bridge), 'A' + pin, irq);
++ dev = bridge;
++ }
++ dev = temp_dev;
++ if (irq >= 0) {
++#ifdef CONFIG_PCI_MSI
++ if (!platform_legacy_irq(irq))
++ irq = IO_APIC_VECTOR(irq);
++#endif
++ printk(KERN_INFO "PCI->APIC IRQ transform: %s[%c] -> IRQ %d\n",
++ pci_name(dev), 'A' + pin, irq);
++ dev->irq = irq;
++ return 0;
++ } else
++ msg = " Probably buggy MP table.";
++ } else if (pci_probe & PCI_BIOS_IRQ_SCAN)
++ msg = "";
++ else
++ msg = " Please try using pci=biosirq.";
++
++ /* With IDE legacy devices the IRQ lookup failure is not a problem.. */
++ if (dev->class >> 8 == PCI_CLASS_STORAGE_IDE && !(dev->class & 0x5))
++ return 0;
++
++ printk(KERN_WARNING "PCI: No IRQ known for interrupt pin %c of device %s.%s\n",
++ 'A' + pin, pci_name(dev), msg);
++ }
++ return 0;
++}
++
++int pci_vector_resources(int last, int nr_released)
++{
++ int count = nr_released;
++
++ int next = last;
++ int offset = (last % 8);
++
++ while (next < FIRST_SYSTEM_VECTOR) {
++ next += 8;
++#ifdef CONFIG_X86_64
++ if (next == IA32_SYSCALL_VECTOR)
++ continue;
++#else
++ if (next == SYSCALL_VECTOR)
++ continue;
++#endif
++ count++;
++ if (next >= FIRST_SYSTEM_VECTOR) {
++ if (offset%8) {
++ next = FIRST_DEVICE_VECTOR + offset;
++ offset++;
++ continue;
++ }
++ count--;
++ }
++ }
++
++ return count;
++}
+diff -urNp linux-2.6/arch/i386/pci/Makefile new/arch/i386/pci/Makefile
+--- linux-2.6/arch/i386/pci/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/pci/Makefile 2006-05-09 12:32:37.000000000 +0200
+@@ -4,6 +4,10 @@ obj-$(CONFIG_PCI_BIOS) += pcbios.o
+ obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o direct.o
+ obj-$(CONFIG_PCI_DIRECT) += direct.o
+
++# pcifront should be after pcbios.o, mmconfig.o, and direct.o as it should only
++# take over if direct access to the PCI bus is unavailable
++obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront.o
++
+ pci-y := fixup.o
+ pci-$(CONFIG_ACPI) += acpi.o
+ pci-y += legacy.o irq.o
+@@ -12,3 +16,8 @@ pci-$(CONFIG_X86_VISWS) := visws.o fixu
+ pci-$(CONFIG_X86_NUMAQ) := numa.o irq.o
+
+ obj-y += $(pci-y) common.o
++
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/arch/i386/pci/pcifront.c new/arch/i386/pci/pcifront.c
+--- linux-2.6/arch/i386/pci/pcifront.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/i386/pci/pcifront.c 2006-05-09 12:32:37.000000000 +0200
+@@ -0,0 +1,55 @@
++/*
++ * PCI Frontend Stub - puts some "dummy" functions in to the Linux x86 PCI core
++ * to support the Xen PCI Frontend's operation
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/pci.h>
++#include <asm/acpi.h>
++#include "pci.h"
++
++static int pcifront_enable_irq(struct pci_dev *dev)
++{
++ u8 irq;
++ pci_read_config_byte(dev, PCI_INTERRUPT_LINE, &irq);
++ dev->irq = irq;
++
++ return 0;
++}
++
++extern u8 pci_cache_line_size;
++
++static int __init pcifront_x86_stub_init(void)
++{
++ struct cpuinfo_x86 *c = &boot_cpu_data;
++
++ /* Only install our method if we haven't found real hardware already */
++ if (raw_pci_ops)
++ return 0;
++
++ printk(KERN_INFO "PCI: setting up Xen PCI frontend stub\n");
++
++ /* Copied from arch/i386/pci/common.c */
++ pci_cache_line_size = 32 >> 2;
++ if (c->x86 >= 6 && c->x86_vendor == X86_VENDOR_AMD)
++ pci_cache_line_size = 64 >> 2; /* K7 & K8 */
++ else if (c->x86 > 6 && c->x86_vendor == X86_VENDOR_INTEL)
++ pci_cache_line_size = 128 >> 2; /* P4 */
++
++ /* On x86, we need to disable the normal IRQ routing table and
++ * just ask the backend
++ */
++ pcibios_enable_irq = pcifront_enable_irq;
++ pcibios_disable_irq = NULL;
++
++#ifdef CONFIG_ACPI
++ /* Keep ACPI out of the picture */
++ acpi_noirq = 1;
++#endif
++
++ return 0;
++}
++
++arch_initcall(pcifront_x86_stub_init);
+diff -urNp linux-2.6/arch/i386/power/Makefile new/arch/i386/power/Makefile
+--- linux-2.6/arch/i386/power/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/i386/power/Makefile 2006-05-09 12:32:37.000000000 +0200
+@@ -1,2 +1,4 @@
+-obj-$(CONFIG_PM) += cpu.o
++obj-$(CONFIG_PM_LEGACY) += cpu.o
++obj-$(CONFIG_SOFTWARE_SUSPEND) += cpu.o
++obj-$(CONFIG_ACPI_SLEEP) += cpu.o
+ obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o
+diff -urNp linux-2.6/arch/ia64/hp/sim/Makefile new/arch/ia64/hp/sim/Makefile
+--- linux-2.6/arch/ia64/hp/sim/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/hp/sim/Makefile 2006-05-09 12:32:37.000000000 +0200
+@@ -14,3 +14,5 @@ obj-$(CONFIG_HP_SIMETH) += simeth.o
+ obj-$(CONFIG_HP_SIMSERIAL) += simserial.o
+ obj-$(CONFIG_HP_SIMSERIAL_CONSOLE) += hpsim_console.o
+ obj-$(CONFIG_HP_SIMSCSI) += simscsi.o
++obj-$(CONFIG_XEN) += simserial.o
++obj-$(CONFIG_XEN) += hpsim_console.o
+diff -urNp linux-2.6/arch/ia64/Kconfig new/arch/ia64/Kconfig
+--- linux-2.6/arch/ia64/Kconfig 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/Kconfig 2006-06-28 14:32:13.000000000 +0200
+@@ -58,6 +58,27 @@ config GENERIC_IOMAP
+ bool
+ default y
+
++config XEN
++ bool "Xen hypervisor support"
++ default y
++ help
++ Enable Xen hypervisor support. Resulting kernel runs
++ both as a guest OS on Xen and natively on hardware.
++
++config XEN_IA64_DOM0_VP
++ bool "dom0 vp model"
++ depends on XEN
++ default n
++ help
++ dom0 vp model
++
++config XEN_IA64_DOM0_NON_VP
++ bool
++ depends on XEN && !XEN_IA64_DOM0_VP
++ default y
++ help
++ dom0 P=M model
++
+ config SCHED_NO_NO_OMIT_FRAME_POINTER
+ bool
+ default y
+@@ -506,3 +527,40 @@ source "arch/ia64/Kconfig.debug"
+ source "security/Kconfig"
+
+ source "crypto/Kconfig"
++
++#
++# override default values of drivers/xen/Kconfig
++#
++if XEN
++config XEN_UTIL
++ default n if XEN_IA64_DOM0_VP
++
++config HAVE_ARCH_ALLOC_SKB
++ default n if !XEN_IA64_DOM0_VP
++
++config HAVE_ARCH_DEV_ALLOC_SKB
++ default n if !XEN_IA64_DOM0_VP
++
++config XEN_BALLOON
++ default n if !XEN_IA64_DOM0_VP
++
++config XEN_SKBUFF
++ default n if !XEN_IA64_DOM0_VP
++
++config XEN_NETDEV_BACKEND
++ default n if !XEN_IA64_DOM0_VP
++
++config XEN_NETDEV_FRONTEND
++ default n if !XEN_IA64_DOM0_VP
++
++config XEN_DEVMEM
++ default n
++
++config XEN_REBOOT
++ default n
++
++config XEN_SMPBOOT
++ default n
++endif
++
++source "drivers/xen/Kconfig"
+diff -urNp linux-2.6/arch/ia64/kernel/entry.S new/arch/ia64/kernel/entry.S
+--- linux-2.6/arch/ia64/kernel/entry.S 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/entry.S 2006-05-09 12:32:38.000000000 +0200
+@@ -181,7 +181,7 @@ END(sys_clone)
+ * called. The code starting at .map relies on this. The rest of the code
+ * doesn't care about the interrupt masking status.
+ */
+-GLOBAL_ENTRY(ia64_switch_to)
++GLOBAL_ENTRY(__ia64_switch_to)
+ .prologue
+ alloc r16=ar.pfs,1,0,0,0
+ DO_SAVE_SWITCH_STACK
+@@ -235,7 +235,7 @@ GLOBAL_ENTRY(ia64_switch_to)
+ ;;
+ srlz.d
+ br.cond.sptk .done
+-END(ia64_switch_to)
++END(__ia64_switch_to)
+
+ /*
+ * Note that interrupts are enabled during save_switch_stack and load_switch_stack. This
+@@ -376,7 +376,7 @@ END(save_switch_stack)
+ * - b7 holds address to return to
+ * - must not touch r8-r11
+ */
+-ENTRY(load_switch_stack)
++GLOBAL_ENTRY(load_switch_stack)
+ .prologue
+ .altrp b7
+
+@@ -511,7 +511,7 @@ END(clone)
+ * because some system calls (such as ia64_execve) directly
+ * manipulate ar.pfs.
+ */
+-GLOBAL_ENTRY(ia64_trace_syscall)
++GLOBAL_ENTRY(__ia64_trace_syscall)
+ PT_REGS_UNWIND_INFO(0)
+ /*
+ * We need to preserve the scratch registers f6-f11 in case the system
+@@ -583,7 +583,7 @@ strace_error:
+ (p6) mov r10=-1
+ (p6) mov r8=r9
+ br.cond.sptk .strace_save_retval
+-END(ia64_trace_syscall)
++END(__ia64_trace_syscall)
+
+ /*
+ * When traced and returning from sigreturn, we invoke syscall_trace but then
+@@ -636,8 +636,11 @@ GLOBAL_ENTRY(ia64_ret_from_syscall)
+ adds r2=PT(R8)+16,sp // r2 = &pt_regs.r8
+ mov r10=r0 // clear error indication in r10
+ (p7) br.cond.spnt handle_syscall_error // handle potential syscall failure
++ ;;
++ // don't fall through, ia64_leave_syscall may be #define'd
++ br.cond.sptk.few ia64_leave_syscall
++ ;;
+ END(ia64_ret_from_syscall)
+- // fall through
+ /*
+ * ia64_leave_syscall(): Same as ia64_leave_kernel, except that it doesn't
+ * need to switch to bank 0 and doesn't restore the scratch registers.
+@@ -682,7 +685,7 @@ END(ia64_ret_from_syscall)
+ * ar.csd: cleared
+ * ar.ssd: cleared
+ */
+-ENTRY(ia64_leave_syscall)
++GLOBAL_ENTRY(__ia64_leave_syscall)
+ PT_REGS_UNWIND_INFO(0)
+ /*
+ * work.need_resched etc. mustn't get changed by this CPU before it returns to
+@@ -790,7 +793,7 @@ ENTRY(ia64_leave_syscall)
+ mov.m ar.ssd=r0 // M2 clear ar.ssd
+ mov f11=f0 // F clear f11
+ br.cond.sptk.many rbs_switch // B
+-END(ia64_leave_syscall)
++END(__ia64_leave_syscall)
+
+ #ifdef CONFIG_IA32_SUPPORT
+ GLOBAL_ENTRY(ia64_ret_from_ia32_execve)
+@@ -802,10 +805,13 @@ GLOBAL_ENTRY(ia64_ret_from_ia32_execve)
+ st8.spill [r2]=r8 // store return value in slot for r8 and set unat bit
+ .mem.offset 8,0
+ st8.spill [r3]=r0 // clear error indication in slot for r10 and set unat bit
++ ;;
++ // don't fall through, ia64_leave_kernel may be #define'd
++ br.cond.sptk.few ia64_leave_kernel
++ ;;
+ END(ia64_ret_from_ia32_execve)
+- // fall through
+ #endif /* CONFIG_IA32_SUPPORT */
+-GLOBAL_ENTRY(ia64_leave_kernel)
++GLOBAL_ENTRY(__ia64_leave_kernel)
+ PT_REGS_UNWIND_INFO(0)
+ /*
+ * work.need_resched etc. mustn't get changed by this CPU before it returns to
+@@ -1136,7 +1142,7 @@ skip_rbs_switch:
+ ld8 r10=[r3]
+ br.cond.sptk.many .work_processed_syscall // re-check
+
+-END(ia64_leave_kernel)
++END(__ia64_leave_kernel)
+
+ ENTRY(handle_syscall_error)
+ /*
+@@ -1176,7 +1182,7 @@ END(ia64_invoke_schedule_tail)
+ * be set up by the caller. We declare 8 input registers so the system call
+ * args get preserved, in case we need to restart a system call.
+ */
+-ENTRY(notify_resume_user)
++GLOBAL_ENTRY(notify_resume_user)
+ .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(8)
+ alloc loc1=ar.pfs,8,2,3,0 // preserve all eight input regs in case of syscall restart!
+ mov r9=ar.unat
+@@ -1264,7 +1270,7 @@ ENTRY(sys_rt_sigreturn)
+ adds sp=16,sp
+ ;;
+ ld8 r9=[sp] // load new ar.unat
+- mov.sptk b7=r8,ia64_leave_kernel
++ mov.sptk b7=r8,__ia64_leave_kernel
+ ;;
+ mov ar.unat=r9
+ br.many b7
+diff -urNp linux-2.6/arch/ia64/kernel/gate.lds.S new/arch/ia64/kernel/gate.lds.S
+--- linux-2.6/arch/ia64/kernel/gate.lds.S 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/gate.lds.S 1970-01-01 01:00:00.000000000 +0100
+@@ -1,96 +0,0 @@
+-/*
+- * Linker script for gate DSO. The gate pages are an ELF shared object prelinked to its
+- * virtual address, with only one read-only segment and one execute-only segment (both fit
+- * in one page). This script controls its layout.
+- */
+-
+-#include <linux/config.h>
+-
+-#include <asm/system.h>
+-
+-SECTIONS
+-{
+- . = GATE_ADDR + SIZEOF_HEADERS;
+-
+- .hash : { *(.hash) } :readable
+- .dynsym : { *(.dynsym) }
+- .dynstr : { *(.dynstr) }
+- .gnu.version : { *(.gnu.version) }
+- .gnu.version_d : { *(.gnu.version_d) }
+- .gnu.version_r : { *(.gnu.version_r) }
+- .dynamic : { *(.dynamic) } :readable :dynamic
+-
+- /*
+- * This linker script is used both with -r and with -shared. For the layouts to match,
+- * we need to skip more than enough space for the dynamic symbol table et al. If this
+- * amount is insufficient, ld -shared will barf. Just increase it here.
+- */
+- . = GATE_ADDR + 0x500;
+-
+- .data.patch : {
+- __start_gate_mckinley_e9_patchlist = .;
+- *(.data.patch.mckinley_e9)
+- __end_gate_mckinley_e9_patchlist = .;
+-
+- __start_gate_vtop_patchlist = .;
+- *(.data.patch.vtop)
+- __end_gate_vtop_patchlist = .;
+-
+- __start_gate_fsyscall_patchlist = .;
+- *(.data.patch.fsyscall_table)
+- __end_gate_fsyscall_patchlist = .;
+-
+- __start_gate_brl_fsys_bubble_down_patchlist = .;
+- *(.data.patch.brl_fsys_bubble_down)
+- __end_gate_brl_fsys_bubble_down_patchlist = .;
+- } :readable
+- .IA_64.unwind_info : { *(.IA_64.unwind_info*) }
+- .IA_64.unwind : { *(.IA_64.unwind*) } :readable :unwind
+-#ifdef HAVE_BUGGY_SEGREL
+- .text (GATE_ADDR + PAGE_SIZE) : { *(.text) *(.text.*) } :readable
+-#else
+- . = ALIGN (PERCPU_PAGE_SIZE) + (. & (PERCPU_PAGE_SIZE - 1));
+- .text : { *(.text) *(.text.*) } :epc
+-#endif
+-
+- /DISCARD/ : {
+- *(.got.plt) *(.got)
+- *(.data .data.* .gnu.linkonce.d.*)
+- *(.dynbss)
+- *(.bss .bss.* .gnu.linkonce.b.*)
+- *(__ex_table)
+- *(__mca_table)
+- }
+-}
+-
+-/*
+- * We must supply the ELF program headers explicitly to get just one
+- * PT_LOAD segment, and set the flags explicitly to make segments read-only.
+- */
+-PHDRS
+-{
+- readable PT_LOAD FILEHDR PHDRS FLAGS(4); /* PF_R */
+-#ifndef HAVE_BUGGY_SEGREL
+- epc PT_LOAD FILEHDR PHDRS FLAGS(1); /* PF_X */
+-#endif
+- dynamic PT_DYNAMIC FLAGS(4); /* PF_R */
+- unwind 0x70000001; /* PT_IA_64_UNWIND, but ld doesn't match the name */
+-}
+-
+-/*
+- * This controls what symbols we export from the DSO.
+- */
+-VERSION
+-{
+- LINUX_2.5 {
+- global:
+- __kernel_syscall_via_break;
+- __kernel_syscall_via_epc;
+- __kernel_sigtramp;
+-
+- local: *;
+- };
+-}
+-
+-/* The ELF entry point can be used to set the AT_SYSINFO value. */
+-ENTRY(__kernel_syscall_via_epc)
+diff -urNp linux-2.6/arch/ia64/kernel/gate.S new/arch/ia64/kernel/gate.S
+--- linux-2.6/arch/ia64/kernel/gate.S 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/gate.S 1970-01-01 01:00:00.000000000 +0100
+@@ -1,376 +0,0 @@
+-/*
+- * This file contains the code that gets mapped at the upper end of each task's text
+- * region. For now, it contains the signal trampoline code only.
+- *
+- * Copyright (C) 1999-2003 Hewlett-Packard Co
+- * David Mosberger-Tang <davidm@hpl.hp.com>
+- */
+-
+-#include <linux/config.h>
+-
+-#include <asm/asmmacro.h>
+-#include <asm/errno.h>
+-#include <asm/asm-offsets.h>
+-#include <asm/sigcontext.h>
+-#include <asm/system.h>
+-#include <asm/unistd.h>
+-
+-/*
+- * We can't easily refer to symbols inside the kernel. To avoid full runtime relocation,
+- * complications with the linker (which likes to create PLT stubs for branches
+- * to targets outside the shared object) and to avoid multi-phase kernel builds, we
+- * simply create minimalistic "patch lists" in special ELF sections.
+- */
+- .section ".data.patch.fsyscall_table", "a"
+- .previous
+-#define LOAD_FSYSCALL_TABLE(reg) \
+-[1:] movl reg=0; \
+- .xdata4 ".data.patch.fsyscall_table", 1b-.
+-
+- .section ".data.patch.brl_fsys_bubble_down", "a"
+- .previous
+-#define BRL_COND_FSYS_BUBBLE_DOWN(pr) \
+-[1:](pr)brl.cond.sptk 0; \
+- .xdata4 ".data.patch.brl_fsys_bubble_down", 1b-.
+-
+-GLOBAL_ENTRY(__kernel_syscall_via_break)
+- .prologue
+- .altrp b6
+- .body
+- /*
+- * Note: for (fast) syscall restart to work, the break instruction must be
+- * the first one in the bundle addressed by syscall_via_break.
+- */
+-{ .mib
+- break 0x100000
+- nop.i 0
+- br.ret.sptk.many b6
+-}
+-END(__kernel_syscall_via_break)
+-
+-/*
+- * On entry:
+- * r11 = saved ar.pfs
+- * r15 = system call #
+- * b0 = saved return address
+- * b6 = return address
+- * On exit:
+- * r11 = saved ar.pfs
+- * r15 = system call #
+- * b0 = saved return address
+- * all other "scratch" registers: undefined
+- * all "preserved" registers: same as on entry
+- */
+-
+-GLOBAL_ENTRY(__kernel_syscall_via_epc)
+- .prologue
+- .altrp b6
+- .body
+-{
+- /*
+- * Note: the kernel cannot assume that the first two instructions in this
+- * bundle get executed. The remaining code must be safe even if
+- * they do not get executed.
+- */
+- adds r17=-1024,r15 // A
+- mov r10=0 // A default to successful syscall execution
+- epc // B causes split-issue
+-}
+- ;;
+- rsm psr.be | psr.i // M2 (5 cyc to srlz.d)
+- LOAD_FSYSCALL_TABLE(r14) // X
+- ;;
+- mov r16=IA64_KR(CURRENT) // M2 (12 cyc)
+- shladd r18=r17,3,r14 // A
+- mov r19=NR_syscalls-1 // A
+- ;;
+- lfetch [r18] // M0|1
+- mov r29=psr // M2 (12 cyc)
+- // If r17 is a NaT, p6 will be zero
+- cmp.geu p6,p7=r19,r17 // A (sysnr > 0 && sysnr < 1024+NR_syscalls)?
+- ;;
+- mov r21=ar.fpsr // M2 (12 cyc)
+- tnat.nz p10,p9=r15 // I0
+- mov.i r26=ar.pfs // I0 (would stall anyhow due to srlz.d...)
+- ;;
+- srlz.d // M0 (forces split-issue) ensure PSR.BE==0
+-(p6) ld8 r18=[r18] // M0|1
+- nop.i 0
+- ;;
+- nop.m 0
+-(p6) tbit.z.unc p8,p0=r18,0 // I0 (dual-issues with "mov b7=r18"!)
+- nop.i 0
+- ;;
+-(p8) ssm psr.i
+-(p6) mov b7=r18 // I0
+-(p8) br.dptk.many b7 // B
+-
+- mov r27=ar.rsc // M2 (12 cyc)
+-/*
+- * brl.cond doesn't work as intended because the linker would convert this branch
+- * into a branch to a PLT. Perhaps there will be a way to avoid this with some
+- * future version of the linker. In the meantime, we just use an indirect branch
+- * instead.
+- */
+-#ifdef CONFIG_ITANIUM
+-(p6) add r14=-8,r14 // r14 <- addr of fsys_bubble_down entry
+- ;;
+-(p6) ld8 r14=[r14] // r14 <- fsys_bubble_down
+- ;;
+-(p6) mov b7=r14
+-(p6) br.sptk.many b7
+-#else
+- BRL_COND_FSYS_BUBBLE_DOWN(p6)
+-#endif
+- ssm psr.i
+- mov r10=-1
+-(p10) mov r8=EINVAL
+-(p9) mov r8=ENOSYS
+- FSYS_RETURN
+-END(__kernel_syscall_via_epc)
+-
+-# define ARG0_OFF (16 + IA64_SIGFRAME_ARG0_OFFSET)
+-# define ARG1_OFF (16 + IA64_SIGFRAME_ARG1_OFFSET)
+-# define ARG2_OFF (16 + IA64_SIGFRAME_ARG2_OFFSET)
+-# define SIGHANDLER_OFF (16 + IA64_SIGFRAME_HANDLER_OFFSET)
+-# define SIGCONTEXT_OFF (16 + IA64_SIGFRAME_SIGCONTEXT_OFFSET)
+-
+-# define FLAGS_OFF IA64_SIGCONTEXT_FLAGS_OFFSET
+-# define CFM_OFF IA64_SIGCONTEXT_CFM_OFFSET
+-# define FR6_OFF IA64_SIGCONTEXT_FR6_OFFSET
+-# define BSP_OFF IA64_SIGCONTEXT_AR_BSP_OFFSET
+-# define RNAT_OFF IA64_SIGCONTEXT_AR_RNAT_OFFSET
+-# define UNAT_OFF IA64_SIGCONTEXT_AR_UNAT_OFFSET
+-# define FPSR_OFF IA64_SIGCONTEXT_AR_FPSR_OFFSET
+-# define PR_OFF IA64_SIGCONTEXT_PR_OFFSET
+-# define RP_OFF IA64_SIGCONTEXT_IP_OFFSET
+-# define SP_OFF IA64_SIGCONTEXT_R12_OFFSET
+-# define RBS_BASE_OFF IA64_SIGCONTEXT_RBS_BASE_OFFSET
+-# define LOADRS_OFF IA64_SIGCONTEXT_LOADRS_OFFSET
+-# define base0 r2
+-# define base1 r3
+- /*
+- * When we get here, the memory stack looks like this:
+- *
+- * +===============================+
+- * | |
+- * // struct sigframe //
+- * | |
+- * +-------------------------------+ <-- sp+16
+- * | 16 byte of scratch |
+- * | space |
+- * +-------------------------------+ <-- sp
+- *
+- * The register stack looks _exactly_ the way it looked at the time the signal
+- * occurred. In other words, we're treading on a potential mine-field: each
+- * incoming general register may be a NaT value (including sp, in which case the
+- * process ends up dying with a SIGSEGV).
+- *
+- * The first thing need to do is a cover to get the registers onto the backing
+- * store. Once that is done, we invoke the signal handler which may modify some
+- * of the machine state. After returning from the signal handler, we return
+- * control to the previous context by executing a sigreturn system call. A signal
+- * handler may call the rt_sigreturn() function to directly return to a given
+- * sigcontext. However, the user-level sigreturn() needs to do much more than
+- * calling the rt_sigreturn() system call as it needs to unwind the stack to
+- * restore preserved registers that may have been saved on the signal handler's
+- * call stack.
+- */
+-
+-#define SIGTRAMP_SAVES \
+- .unwabi 3, 's'; /* mark this as a sigtramp handler (saves scratch regs) */ \
+- .unwabi @svr4, 's'; /* backwards compatibility with old unwinders (remove in v2.7) */ \
+- .savesp ar.unat, UNAT_OFF+SIGCONTEXT_OFF; \
+- .savesp ar.fpsr, FPSR_OFF+SIGCONTEXT_OFF; \
+- .savesp pr, PR_OFF+SIGCONTEXT_OFF; \
+- .savesp rp, RP_OFF+SIGCONTEXT_OFF; \
+- .savesp ar.pfs, CFM_OFF+SIGCONTEXT_OFF; \
+- .vframesp SP_OFF+SIGCONTEXT_OFF
+-
+-GLOBAL_ENTRY(__kernel_sigtramp)
+- // describe the state that is active when we get here:
+- .prologue
+- SIGTRAMP_SAVES
+- .body
+-
+- .label_state 1
+-
+- adds base0=SIGHANDLER_OFF,sp
+- adds base1=RBS_BASE_OFF+SIGCONTEXT_OFF,sp
+- br.call.sptk.many rp=1f
+-1:
+- ld8 r17=[base0],(ARG0_OFF-SIGHANDLER_OFF) // get pointer to signal handler's plabel
+- ld8 r15=[base1] // get address of new RBS base (or NULL)
+- cover // push args in interrupted frame onto backing store
+- ;;
+- cmp.ne p1,p0=r15,r0 // do we need to switch rbs? (note: pr is saved by kernel)
+- mov.m r9=ar.bsp // fetch ar.bsp
+- .spillsp.p p1, ar.rnat, RNAT_OFF+SIGCONTEXT_OFF
+-(p1) br.cond.spnt setup_rbs // yup -> (clobbers p8, r14-r16, and r18-r20)
+-back_from_setup_rbs:
+- alloc r8=ar.pfs,0,0,3,0
+- ld8 out0=[base0],16 // load arg0 (signum)
+- adds base1=(ARG1_OFF-(RBS_BASE_OFF+SIGCONTEXT_OFF)),base1
+- ;;
+- ld8 out1=[base1] // load arg1 (siginfop)
+- ld8 r10=[r17],8 // get signal handler entry point
+- ;;
+- ld8 out2=[base0] // load arg2 (sigcontextp)
+- ld8 gp=[r17] // get signal handler's global pointer
+- adds base0=(BSP_OFF+SIGCONTEXT_OFF),sp
+- ;;
+- .spillsp ar.bsp, BSP_OFF+SIGCONTEXT_OFF
+- st8 [base0]=r9 // save sc_ar_bsp
+- adds base0=(FR6_OFF+SIGCONTEXT_OFF),sp
+- adds base1=(FR6_OFF+16+SIGCONTEXT_OFF),sp
+- ;;
+- stf.spill [base0]=f6,32
+- stf.spill [base1]=f7,32
+- ;;
+- stf.spill [base0]=f8,32
+- stf.spill [base1]=f9,32
+- mov b6=r10
+- ;;
+- stf.spill [base0]=f10,32
+- stf.spill [base1]=f11,32
+- ;;
+- stf.spill [base0]=f12,32
+- stf.spill [base1]=f13,32
+- ;;
+- stf.spill [base0]=f14,32
+- stf.spill [base1]=f15,32
+- br.call.sptk.many rp=b6 // call the signal handler
+-.ret0: adds base0=(BSP_OFF+SIGCONTEXT_OFF),sp
+- ;;
+- ld8 r15=[base0] // fetch sc_ar_bsp
+- mov r14=ar.bsp
+- ;;
+- cmp.ne p1,p0=r14,r15 // do we need to restore the rbs?
+-(p1) br.cond.spnt restore_rbs // yup -> (clobbers r14-r18, f6 & f7)
+- ;;
+-back_from_restore_rbs:
+- adds base0=(FR6_OFF+SIGCONTEXT_OFF),sp
+- adds base1=(FR6_OFF+16+SIGCONTEXT_OFF),sp
+- ;;
+- ldf.fill f6=[base0],32
+- ldf.fill f7=[base1],32
+- ;;
+- ldf.fill f8=[base0],32
+- ldf.fill f9=[base1],32
+- ;;
+- ldf.fill f10=[base0],32
+- ldf.fill f11=[base1],32
+- ;;
+- ldf.fill f12=[base0],32
+- ldf.fill f13=[base1],32
+- ;;
+- ldf.fill f14=[base0],32
+- ldf.fill f15=[base1],32
+- mov r15=__NR_rt_sigreturn
+- .restore sp // pop .prologue
+- break __BREAK_SYSCALL
+-
+- .prologue
+- SIGTRAMP_SAVES
+-setup_rbs:
+- mov ar.rsc=0 // put RSE into enforced lazy mode
+- ;;
+- .save ar.rnat, r19
+- mov r19=ar.rnat // save RNaT before switching backing store area
+- adds r14=(RNAT_OFF+SIGCONTEXT_OFF),sp
+-
+- mov r18=ar.bspstore
+- mov ar.bspstore=r15 // switch over to new register backing store area
+- ;;
+-
+- .spillsp ar.rnat, RNAT_OFF+SIGCONTEXT_OFF
+- st8 [r14]=r19 // save sc_ar_rnat
+- .body
+- mov.m r16=ar.bsp // sc_loadrs <- (new bsp - new bspstore) << 16
+- adds r14=(LOADRS_OFF+SIGCONTEXT_OFF),sp
+- ;;
+- invala
+- sub r15=r16,r15
+- extr.u r20=r18,3,6
+- ;;
+- mov ar.rsc=0xf // set RSE into eager mode, pl 3
+- cmp.eq p8,p0=63,r20
+- shl r15=r15,16
+- ;;
+- st8 [r14]=r15 // save sc_loadrs
+-(p8) st8 [r18]=r19 // if bspstore points at RNaT slot, store RNaT there now
+- .restore sp // pop .prologue
+- br.cond.sptk back_from_setup_rbs
+-
+- .prologue
+- SIGTRAMP_SAVES
+- .spillsp ar.rnat, RNAT_OFF+SIGCONTEXT_OFF
+- .body
+-restore_rbs:
+- // On input:
+- // r14 = bsp1 (bsp at the time of return from signal handler)
+- // r15 = bsp0 (bsp at the time the signal occurred)
+- //
+- // Here, we need to calculate bspstore0, the value that ar.bspstore needs
+- // to be set to, based on bsp0 and the size of the dirty partition on
+- // the alternate stack (sc_loadrs >> 16). This can be done with the
+- // following algorithm:
+- //
+- // bspstore0 = rse_skip_regs(bsp0, -rse_num_regs(bsp1 - (loadrs >> 19), bsp1));
+- //
+- // This is what the code below does.
+- //
+- alloc r2=ar.pfs,0,0,0,0 // alloc null frame
+- adds r16=(LOADRS_OFF+SIGCONTEXT_OFF),sp
+- adds r18=(RNAT_OFF+SIGCONTEXT_OFF),sp
+- ;;
+- ld8 r17=[r16]
+- ld8 r16=[r18] // get new rnat
+- extr.u r18=r15,3,6 // r18 <- rse_slot_num(bsp0)
+- ;;
+- mov ar.rsc=r17 // put RSE into enforced lazy mode
+- shr.u r17=r17,16
+- ;;
+- sub r14=r14,r17 // r14 (bspstore1) <- bsp1 - (sc_loadrs >> 16)
+- shr.u r17=r17,3 // r17 <- (sc_loadrs >> 19)
+- ;;
+- loadrs // restore dirty partition
+- extr.u r14=r14,3,6 // r14 <- rse_slot_num(bspstore1)
+- ;;
+- add r14=r14,r17 // r14 <- rse_slot_num(bspstore1) + (sc_loadrs >> 19)
+- ;;
+- shr.u r14=r14,6 // r14 <- (rse_slot_num(bspstore1) + (sc_loadrs >> 19))/0x40
+- ;;
+- sub r14=r14,r17 // r14 <- -rse_num_regs(bspstore1, bsp1)
+- movl r17=0x8208208208208209
+- ;;
+- add r18=r18,r14 // r18 (delta) <- rse_slot_num(bsp0) - rse_num_regs(bspstore1,bsp1)
+- setf.sig f7=r17
+- cmp.lt p7,p0=r14,r0 // p7 <- (r14 < 0)?
+- ;;
+-(p7) adds r18=-62,r18 // delta -= 62
+- ;;
+- setf.sig f6=r18
+- ;;
+- xmpy.h f6=f6,f7
+- ;;
+- getf.sig r17=f6
+- ;;
+- add r17=r17,r18
+- shr r18=r18,63
+- ;;
+- shr r17=r17,5
+- ;;
+- sub r17=r17,r18 // r17 = delta/63
+- ;;
+- add r17=r14,r17 // r17 <- delta/63 - rse_num_regs(bspstore1, bsp1)
+- ;;
+- shladd r15=r17,3,r15 // r15 <- bsp0 + 8*(delta/63 - rse_num_regs(bspstore1, bsp1))
+- ;;
+- mov ar.bspstore=r15 // switch back to old register backing store area
+- ;;
+- mov ar.rnat=r16 // restore RNaT
+- mov ar.rsc=0xf // (will be restored later on from sc_ar_rsc)
+- // invala not necessary as that will happen when returning to user-mode
+- br.cond.sptk back_from_restore_rbs
+-END(__kernel_sigtramp)
+diff -urNp linux-2.6/arch/ia64/kernel/head.S new/arch/ia64/kernel/head.S
+--- linux-2.6/arch/ia64/kernel/head.S 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/head.S 2006-05-09 12:32:38.000000000 +0200
+@@ -363,6 +363,12 @@ start_ap:
+ ;;
+ (isBP) st8 [r2]=r28 // save the address of the boot param area passed by the bootloader
+
++#ifdef CONFIG_XEN
++ // Note: isBP is used by the subprogram.
++ br.call.sptk.many rp=early_xen_setup
++ ;;
++#endif
++
+ #ifdef CONFIG_SMP
+ (isAP) br.call.sptk.many rp=start_secondary
+ .ret0:
+diff -urNp linux-2.6/arch/ia64/kernel/iosapic.c new/arch/ia64/kernel/iosapic.c
+--- linux-2.6/arch/ia64/kernel/iosapic.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/iosapic.c 2006-06-28 14:32:13.000000000 +0200
+@@ -160,6 +160,65 @@ static unsigned char pcat_compat __devin
+ static int iosapic_kmalloc_ok;
+ static LIST_HEAD(free_rte_list);
+
++#ifdef CONFIG_XEN
++#include <xen/interface/xen.h>
++#include <xen/interface/physdev.h>
++#include <asm/hypervisor.h>
++static inline unsigned int xen_iosapic_read(char __iomem *iosapic, unsigned int reg)
++{
++ struct physdev_apic apic_op;
++ int ret;
++
++ apic_op.apic_physbase = (unsigned long)iosapic -
++ __IA64_UNCACHED_OFFSET;
++ apic_op.reg = reg;
++ ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
++ if (ret)
++ return ret;
++ return apic_op.value;
++}
++
++static inline void xen_iosapic_write(char __iomem *iosapic, unsigned int reg, u32 val)
++{
++ struct physdev_apic apic_op;
++
++ apic_op.apic_physbase = (unsigned long)iosapic -
++ __IA64_UNCACHED_OFFSET;
++ apic_op.reg = reg;
++ apic_op.value = val;
++ HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op);
++}
++
++static inline unsigned int iosapic_read(char __iomem *iosapic, unsigned int reg)
++{
++ if (!is_running_on_xen()) {
++ writel(reg, iosapic + IOSAPIC_REG_SELECT);
++ return readl(iosapic + IOSAPIC_WINDOW);
++ } else
++ return xen_iosapic_read(iosapic, reg);
++}
++
++static inline void iosapic_write(char __iomem *iosapic, unsigned int reg, u32 val)
++{
++ if (!is_running_on_xen()) {
++ writel(reg, iosapic + IOSAPIC_REG_SELECT);
++ writel(val, iosapic + IOSAPIC_WINDOW);
++ } else
++ xen_iosapic_write(iosapic, reg, val);
++}
++
++int xen_assign_irq_vector(int irq)
++{
++ struct physdev_irq irq_op;
++
++ irq_op.irq = irq;
++ if (HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op))
++ return -ENOSPC;
++
++ return irq_op.vector;
++}
++#endif /* XEN */
++
+ /*
+ * Find an IOSAPIC associated with a GSI
+ */
+@@ -654,6 +713,9 @@ register_intr (unsigned int gsi, int vec
+ iosapic_intr_info[vector].dmode = delivery;
+ iosapic_intr_info[vector].trigger = trigger;
+
++ if (is_running_on_xen())
++ return 0;
++
+ if (trigger == IOSAPIC_EDGE)
+ irq_type = &irq_type_iosapic_edge;
+ else
+@@ -1016,6 +1078,9 @@ iosapic_system_init (int system_pcat_com
+ }
+
+ pcat_compat = system_pcat_compat;
++ if (is_running_on_xen())
++ return;
++
+ if (pcat_compat) {
+ /*
+ * Disable the compatibility mode interrupts (8259 style),
+diff -urNp linux-2.6/arch/ia64/kernel/irq_ia64.c new/arch/ia64/kernel/irq_ia64.c
+--- linux-2.6/arch/ia64/kernel/irq_ia64.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/irq_ia64.c 2006-06-28 14:32:13.000000000 +0200
+@@ -66,6 +66,13 @@ int
+ assign_irq_vector (int irq)
+ {
+ int pos, vector;
++
++#ifdef CONFIG_XEN
++ if (is_running_on_xen()) {
++ extern int xen_assign_irq_vector(int);
++ return xen_assign_irq_vector(irq);
++ }
++#endif
+ again:
+ pos = find_first_zero_bit(ia64_vector_mask, IA64_NUM_DEVICE_VECTORS);
+ vector = IA64_FIRST_DEVICE_VECTOR + pos;
+@@ -224,6 +231,157 @@ static struct irqaction ipi_irqaction =
+ };
+ #endif
+
++#ifdef CONFIG_XEN
++#include <xen/evtchn.h>
++#include <xen/interface/callback.h>
++
++static char timer_name[NR_CPUS][15];
++static char ipi_name[NR_CPUS][15];
++static char resched_name[NR_CPUS][15];
++
++struct saved_irq {
++ unsigned int irq;
++ struct irqaction *action;
++};
++/* 16 should be far optimistic value, since only several percpu irqs
++ * are registered early.
++ */
++#define MAX_LATE_IRQ 16
++static struct saved_irq saved_percpu_irqs[MAX_LATE_IRQ];
++static unsigned short late_irq_cnt = 0;
++static unsigned short saved_irq_cnt = 0;
++static int xen_slab_ready = 0;
++
++/* Dummy stub. Though we may check RESCHEDULE_VECTOR before __do_IRQ,
++ * it ends up to issue several memory accesses upon percpu data and
++ * thus adds unnecessary traffic to other paths.
++ */
++static irqreturn_t
++handle_reschedule(int irq, void *dev_id, struct pt_regs *regs)
++{
++
++ return IRQ_HANDLED;
++}
++
++static struct irqaction resched_irqaction = {
++ .handler = handle_reschedule,
++ .flags = SA_INTERRUPT,
++ .name = "RESCHED"
++};
++
++/*
++ * This is xen version percpu irq registration, which needs bind
++ * to xen specific evtchn sub-system. One trick here is that xen
++ * evtchn binding interface depends on kmalloc because related
++ * port needs to be freed at device/cpu down. So we cache the
++ * registration on BSP before slab is ready and then deal them
++ * at later point. For rest instances happening after slab ready,
++ * we hook them to xen evtchn immediately.
++ *
++ * FIXME: MCA is not supported by far, and thus "nomca" boot param is
++ * required.
++ */
++static void
++xen_register_percpu_irq (unsigned int irq, struct irqaction *action, int save)
++{
++ unsigned int cpu = smp_processor_id();
++ int ret = 0;
++
++ if (xen_slab_ready) {
++ switch (irq) {
++ case IA64_TIMER_VECTOR:
++ sprintf(timer_name[cpu], "%s%d", action->name, cpu);
++ ret = bind_virq_to_irqhandler(VIRQ_ITC, cpu,
++ action->handler, action->flags,
++ timer_name[cpu], action->dev_id);
++ printk(KERN_INFO "register VIRQ_ITC (%s) to xen irq (%d)\n", timer_name[cpu], ret);
++ break;
++ case IA64_IPI_RESCHEDULE:
++ sprintf(resched_name[cpu], "%s%d", action->name, cpu);
++ ret = bind_ipi_to_irqhandler(RESCHEDULE_VECTOR, cpu,
++ action->handler, action->flags,
++ resched_name[cpu], action->dev_id);
++ printk(KERN_INFO "register RESCHEDULE_VECTOR (%s) to xen irq (%d)\n", resched_name[cpu], ret);
++ break;
++ case IA64_IPI_VECTOR:
++ sprintf(ipi_name[cpu], "%s%d", action->name, cpu);
++ ret = bind_ipi_to_irqhandler(IPI_VECTOR, cpu,
++ action->handler, action->flags,
++ ipi_name[cpu], action->dev_id);
++ printk(KERN_INFO "register IPI_VECTOR (%s) to xen irq (%d)\n", ipi_name[cpu], ret);
++ break;
++ case IA64_SPURIOUS_INT_VECTOR:
++ break;
++ default:
++ printk(KERN_WARNING "Percpu irq %d is unsupported by xen!\n", irq);
++ break;
++ }
++ BUG_ON(ret < 0);
++ }
++
++ /* For BSP, we cache registered percpu irqs, and then re-walk
++ * them when initializing APs
++ */
++ if (!cpu && save) {
++ BUG_ON(saved_irq_cnt == MAX_LATE_IRQ);
++ saved_percpu_irqs[saved_irq_cnt].irq = irq;
++ saved_percpu_irqs[saved_irq_cnt].action = action;
++ saved_irq_cnt++;
++ if (!xen_slab_ready)
++ late_irq_cnt++;
++ }
++}
++
++static void
++xen_bind_early_percpu_irq (void)
++{
++ int i;
++
++ xen_slab_ready = 1;
++ /* There's no race when accessing this cached array, since only
++ * BSP will face with such step shortly
++ */
++ for (i = 0; i < late_irq_cnt; i++)
++ xen_register_percpu_irq(saved_percpu_irqs[i].irq,
++ saved_percpu_irqs[i].action, 0);
++}
++
++/* FIXME: There's no obvious point to check whether slab is ready. So
++ * a hack is used here by utilizing a late time hook.
++ */
++extern void (*late_time_init)(void);
++extern char xen_event_callback;
++extern void xen_init_IRQ(void);
++
++DECLARE_PER_CPU(int, ipi_to_irq[NR_IPIS]);
++void xen_smp_intr_init(void)
++{
++#ifdef CONFIG_SMP
++ unsigned int cpu = smp_processor_id();
++ unsigned int i = 0;
++ struct callback_register event = {
++ .type = CALLBACKTYPE_event,
++ .address = (unsigned long)&xen_event_callback,
++ };
++ static cpumask_t registered_cpumask;
++
++ if (!cpu)
++ return;
++
++ /* This should be piggyback when setup vcpu guest context */
++ BUG_ON(HYPERVISOR_callback_op(CALLBACKOP_register, &event));
++
++ if (!cpu_isset(cpu, registered_cpumask)) {
++ cpu_set(cpu, registered_cpumask);
++ for (i = 0; i < saved_irq_cnt; i++)
++ xen_register_percpu_irq(saved_percpu_irqs[i].irq,
++ saved_percpu_irqs[i].action,
++ 0);
++ }
++#endif /* CONFIG_SMP */
++}
++#endif /* CONFIG_XEN */
++
+ void
+ register_percpu_irq (ia64_vector vec, struct irqaction *action)
+ {
+@@ -232,6 +390,10 @@ register_percpu_irq (ia64_vector vec, st
+
+ for (irq = 0; irq < NR_IRQS; ++irq)
+ if (irq_to_vector(irq) == vec) {
++#ifdef CONFIG_XEN
++ if (is_running_on_xen())
++ return xen_register_percpu_irq(vec, action, 1);
++#endif
+ desc = irq_descp(irq);
+ desc->status |= IRQ_PER_CPU;
+ desc->handler = &irq_type_ia64_lsapic;
+@@ -243,6 +405,21 @@ register_percpu_irq (ia64_vector vec, st
+ void __init
+ init_IRQ (void)
+ {
++#ifdef CONFIG_XEN
++ /* Maybe put into platform_irq_init later */
++ if (is_running_on_xen()) {
++ struct callback_register event = {
++ .type = CALLBACKTYPE_event,
++ .address = (unsigned long)&xen_event_callback,
++ };
++ xen_init_IRQ();
++ BUG_ON(HYPERVISOR_callback_op(CALLBACKOP_register, &event));
++ late_time_init = xen_bind_early_percpu_irq;
++#ifdef CONFIG_SMP
++ register_percpu_irq(IA64_IPI_RESCHEDULE, &resched_irqaction);
++#endif /* CONFIG_SMP */
++ }
++#endif /* CONFIG_XEN */
+ register_percpu_irq(IA64_SPURIOUS_INT_VECTOR, NULL);
+ #ifdef CONFIG_SMP
+ register_percpu_irq(IA64_IPI_VECTOR, &ipi_irqaction);
+@@ -260,6 +437,37 @@ ia64_send_ipi (int cpu, int vector, int
+ unsigned long ipi_data;
+ unsigned long phys_cpu_id;
+
++#ifdef CONFIG_XEN
++ if (is_running_on_xen()) {
++ int irq = -1;
++
++ /* TODO: we need to call vcpu_up here */
++ if (unlikely(vector == ap_wakeup_vector)) {
++ extern void xen_send_ipi (int cpu, int vec);
++ xen_send_ipi (cpu, vector);
++ //vcpu_prepare_and_up(cpu);
++ return;
++ }
++
++ switch(vector) {
++ case IA64_IPI_VECTOR:
++ irq = per_cpu(ipi_to_irq, cpu)[IPI_VECTOR];
++ break;
++ case IA64_IPI_RESCHEDULE:
++ irq = per_cpu(ipi_to_irq, cpu)[RESCHEDULE_VECTOR];
++ break;
++ default:
++ printk(KERN_WARNING"Unsupported IPI type 0x%x\n", vector);
++ irq = 0;
++ break;
++ }
++
++ BUG_ON(irq < 0);
++ notify_remote_via_irq(irq);
++ return;
++ }
++#endif /* CONFIG_XEN */
++
+ #ifdef CONFIG_SMP
+ phys_cpu_id = cpu_physical_id(cpu);
+ #else
+diff -urNp linux-2.6/arch/ia64/kernel/pal.S new/arch/ia64/kernel/pal.S
+--- linux-2.6/arch/ia64/kernel/pal.S 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/pal.S 2006-05-09 12:32:39.000000000 +0200
+@@ -16,6 +16,7 @@
+ #include <asm/processor.h>
+
+ .data
++ .globl pal_entry_point
+ pal_entry_point:
+ data8 ia64_pal_default_handler
+ .text
+@@ -53,7 +54,7 @@ END(ia64_pal_default_handler)
+ * in4 1 ==> clear psr.ic, 0 ==> don't clear psr.ic
+ *
+ */
+-GLOBAL_ENTRY(ia64_pal_call_static)
++GLOBAL_ENTRY(__ia64_pal_call_static)
+ .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(5)
+ alloc loc1 = ar.pfs,5,5,0,0
+ movl loc2 = pal_entry_point
+@@ -90,7 +91,7 @@ GLOBAL_ENTRY(ia64_pal_call_static)
+ ;;
+ srlz.d // seralize restoration of psr.l
+ br.ret.sptk.many b0
+-END(ia64_pal_call_static)
++END(__ia64_pal_call_static)
+
+ /*
+ * Make a PAL call using the stacked registers calling convention.
+diff -urNp linux-2.6/arch/ia64/kernel/patch.c new/arch/ia64/kernel/patch.c
+--- linux-2.6/arch/ia64/kernel/patch.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/patch.c 1970-01-01 01:00:00.000000000 +0100
+@@ -1,197 +0,0 @@
+-/*
+- * Instruction-patching support.
+- *
+- * Copyright (C) 2003 Hewlett-Packard Co
+- * David Mosberger-Tang <davidm@hpl.hp.com>
+- */
+-#include <linux/init.h>
+-#include <linux/string.h>
+-
+-#include <asm/patch.h>
+-#include <asm/processor.h>
+-#include <asm/sections.h>
+-#include <asm/system.h>
+-#include <asm/unistd.h>
+-
+-/*
+- * This was adapted from code written by Tony Luck:
+- *
+- * The 64-bit value in a "movl reg=value" is scattered between the two words of the bundle
+- * like this:
+- *
+- * 6 6 5 4 3 2 1
+- * 3210987654321098765432109876543210987654321098765432109876543210
+- * ABBBBBBBBBBBBBBBBBBBBBBBCCCCCCCCCCCCCCCCCCDEEEEEFFFFFFFFFGGGGGGG
+- *
+- * CCCCCCCCCCCCCCCCCCxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx
+- * xxxxAFFFFFFFFFEEEEEDxGGGGGGGxxxxxxxxxxxxxBBBBBBBBBBBBBBBBBBBBBBB
+- */
+-static u64
+-get_imm64 (u64 insn_addr)
+-{
+- u64 *p = (u64 *) (insn_addr & -16); /* mask out slot number */
+-
+- return ( (p[1] & 0x0800000000000000UL) << 4) | /*A*/
+- ((p[1] & 0x00000000007fffffUL) << 40) | /*B*/
+- ((p[0] & 0xffffc00000000000UL) >> 24) | /*C*/
+- ((p[1] & 0x0000100000000000UL) >> 23) | /*D*/
+- ((p[1] & 0x0003e00000000000UL) >> 29) | /*E*/
+- ((p[1] & 0x07fc000000000000UL) >> 43) | /*F*/
+- ((p[1] & 0x000007f000000000UL) >> 36); /*G*/
+-}
+-
+-/* Patch instruction with "val" where "mask" has 1 bits. */
+-void
+-ia64_patch (u64 insn_addr, u64 mask, u64 val)
+-{
+- u64 m0, m1, v0, v1, b0, b1, *b = (u64 *) (insn_addr & -16);
+-# define insn_mask ((1UL << 41) - 1)
+- unsigned long shift;
+-
+- b0 = b[0]; b1 = b[1];
+- shift = 5 + 41 * (insn_addr % 16); /* 5 bits of template, then 3 x 41-bit instructions */
+- if (shift >= 64) {
+- m1 = mask << (shift - 64);
+- v1 = val << (shift - 64);
+- } else {
+- m0 = mask << shift; m1 = mask >> (64 - shift);
+- v0 = val << shift; v1 = val >> (64 - shift);
+- b[0] = (b0 & ~m0) | (v0 & m0);
+- }
+- b[1] = (b1 & ~m1) | (v1 & m1);
+-}
+-
+-void
+-ia64_patch_imm64 (u64 insn_addr, u64 val)
+-{
+- /* The assembler may generate offset pointing to either slot 1
+- or slot 2 for a long (2-slot) instruction, occupying slots 1
+- and 2. */
+- insn_addr &= -16UL;
+- ia64_patch(insn_addr + 2,
+- 0x01fffefe000UL, ( ((val & 0x8000000000000000UL) >> 27) /* bit 63 -> 36 */
+- | ((val & 0x0000000000200000UL) << 0) /* bit 21 -> 21 */
+- | ((val & 0x00000000001f0000UL) << 6) /* bit 16 -> 22 */
+- | ((val & 0x000000000000ff80UL) << 20) /* bit 7 -> 27 */
+- | ((val & 0x000000000000007fUL) << 13) /* bit 0 -> 13 */));
+- ia64_patch(insn_addr + 1, 0x1ffffffffffUL, val >> 22);
+-}
+-
+-void
+-ia64_patch_imm60 (u64 insn_addr, u64 val)
+-{
+- /* The assembler may generate offset pointing to either slot 1
+- or slot 2 for a long (2-slot) instruction, occupying slots 1
+- and 2. */
+- insn_addr &= -16UL;
+- ia64_patch(insn_addr + 2,
+- 0x011ffffe000UL, ( ((val & 0x0800000000000000UL) >> 23) /* bit 59 -> 36 */
+- | ((val & 0x00000000000fffffUL) << 13) /* bit 0 -> 13 */));
+- ia64_patch(insn_addr + 1, 0x1fffffffffcUL, val >> 18);
+-}
+-
+-/*
+- * We need sometimes to load the physical address of a kernel
+- * object. Often we can convert the virtual address to physical
+- * at execution time, but sometimes (either for performance reasons
+- * or during error recovery) we cannot to this. Patch the marked
+- * bundles to load the physical address.
+- */
+-void __init
+-ia64_patch_vtop (unsigned long start, unsigned long end)
+-{
+- s32 *offp = (s32 *) start;
+- u64 ip;
+-
+- while (offp < (s32 *) end) {
+- ip = (u64) offp + *offp;
+-
+- /* replace virtual address with corresponding physical address: */
+- ia64_patch_imm64(ip, ia64_tpa(get_imm64(ip)));
+- ia64_fc((void *) ip);
+- ++offp;
+- }
+- ia64_sync_i();
+- ia64_srlz_i();
+-}
+-
+-void __init
+-ia64_patch_mckinley_e9 (unsigned long start, unsigned long end)
+-{
+- static int first_time = 1;
+- int need_workaround;
+- s32 *offp = (s32 *) start;
+- u64 *wp;
+-
+- need_workaround = (local_cpu_data->family == 0x1f && local_cpu_data->model == 0);
+-
+- if (first_time) {
+- first_time = 0;
+- if (need_workaround)
+- printk(KERN_INFO "Leaving McKinley Errata 9 workaround enabled\n");
+- else
+- printk(KERN_INFO "McKinley Errata 9 workaround not needed; "
+- "disabling it\n");
+- }
+- if (need_workaround)
+- return;
+-
+- while (offp < (s32 *) end) {
+- wp = (u64 *) ia64_imva((char *) offp + *offp);
+- wp[0] = 0x0000000100000000UL; /* nop.m 0; nop.i 0; nop.i 0 */
+- wp[1] = 0x0004000000000200UL;
+- wp[2] = 0x0000000100000011UL; /* nop.m 0; nop.i 0; br.ret.sptk.many b6 */
+- wp[3] = 0x0084006880000200UL;
+- ia64_fc(wp); ia64_fc(wp + 2);
+- ++offp;
+- }
+- ia64_sync_i();
+- ia64_srlz_i();
+-}
+-
+-static void __init
+-patch_fsyscall_table (unsigned long start, unsigned long end)
+-{
+- extern unsigned long fsyscall_table[NR_syscalls];
+- s32 *offp = (s32 *) start;
+- u64 ip;
+-
+- while (offp < (s32 *) end) {
+- ip = (u64) ia64_imva((char *) offp + *offp);
+- ia64_patch_imm64(ip, (u64) fsyscall_table);
+- ia64_fc((void *) ip);
+- ++offp;
+- }
+- ia64_sync_i();
+- ia64_srlz_i();
+-}
+-
+-static void __init
+-patch_brl_fsys_bubble_down (unsigned long start, unsigned long end)
+-{
+- extern char fsys_bubble_down[];
+- s32 *offp = (s32 *) start;
+- u64 ip;
+-
+- while (offp < (s32 *) end) {
+- ip = (u64) offp + *offp;
+- ia64_patch_imm60((u64) ia64_imva((void *) ip),
+- (u64) (fsys_bubble_down - (ip & -16)) / 16);
+- ia64_fc((void *) ip);
+- ++offp;
+- }
+- ia64_sync_i();
+- ia64_srlz_i();
+-}
+-
+-void __init
+-ia64_patch_gate (void)
+-{
+-# define START(name) ((unsigned long) __start_gate_##name##_patchlist)
+-# define END(name) ((unsigned long)__end_gate_##name##_patchlist)
+-
+- patch_fsyscall_table(START(fsyscall), END(fsyscall));
+- patch_brl_fsys_bubble_down(START(brl_fsys_bubble_down), END(brl_fsys_bubble_down));
+- ia64_patch_vtop(START(vtop), END(vtop));
+- ia64_patch_mckinley_e9(START(mckinley_e9), END(mckinley_e9));
+-}
+diff -urNp linux-2.6/arch/ia64/kernel/setup.c new/arch/ia64/kernel/setup.c
+--- linux-2.6/arch/ia64/kernel/setup.c 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/kernel/setup.c 2006-06-28 14:32:13.000000000 +0200
+@@ -61,6 +61,10 @@
+ #include <asm/system.h>
+ #include <asm/unistd.h>
+ #include <asm/system.h>
++#ifdef CONFIG_XEN
++#include <asm/hypervisor.h>
++#endif
++#include <linux/dma-mapping.h>
+
+ #if defined(CONFIG_SMP) && (IA64_CPU_SIZE > PAGE_SIZE)
+ # error "struct cpuinfo_ia64 too big!"
+@@ -243,6 +247,14 @@ reserve_memory (void)
+ rsvd_region[n].end = (unsigned long) ia64_imva(_end);
+ n++;
+
++#ifdef CONFIG_XEN
++ if (is_running_on_xen()) {
++ rsvd_region[n].start = (unsigned long)__va((HYPERVISOR_shared_info->arch.start_info_pfn << PAGE_SHIFT));
++ rsvd_region[n].end = rsvd_region[n].start + PAGE_SIZE;
++ n++;
++ }
++#endif
++
+ #ifdef CONFIG_BLK_DEV_INITRD
+ if (ia64_boot_param->initrd_start) {
+ rsvd_region[n].start = (unsigned long)__va(ia64_boot_param->initrd_start);
+@@ -260,6 +272,7 @@ reserve_memory (void)
+ n++;
+
+ num_rsvd_regions = n;
++ BUG_ON(IA64_MAX_RSVD_REGIONS + 1 < n);
+
+ sort_regions(rsvd_region, num_rsvd_regions);
+ }
+@@ -333,6 +346,16 @@ early_console_setup (char *cmdline)
+ {
+ int earlycons = 0;
+
++#ifdef CONFIG_XEN
++#ifndef CONFIG_IA64_HP_SIM
++ if (is_running_on_xen()) {
++ extern struct console hpsim_cons;
++ hpsim_cons.flags |= CON_BOOT;
++ register_console(&hpsim_cons);
++ earlycons++;
++ }
++#endif
++#endif
+ #ifdef CONFIG_SERIAL_SGI_L1_CONSOLE
+ {
+ extern int sn_serial_console_early_setup(void);
+@@ -402,6 +425,11 @@ setup_arch (char **cmdline_p)
+ {
+ unw_init();
+
++#ifdef CONFIG_XEN
++ if (is_running_on_xen())
++ setup_xen_features();
++#endif
++
+ ia64_patch_vtop((u64) __start___vtop_patchlist, (u64) __end___vtop_patchlist);
+
+ *cmdline_p = __va(ia64_boot_param->command_line);
+@@ -478,6 +506,29 @@ setup_arch (char **cmdline_p)
+ conswitchp = &vga_con;
+ # endif
+ }
++#ifdef CONFIG_XEN
++ if (is_running_on_xen()) {
++ shared_info_t *s = HYPERVISOR_shared_info;
++
++ xen_start_info = __va(s->arch.start_info_pfn << PAGE_SHIFT);
++ xen_start_info->flags = s->arch.flags;
++
++ printk("Running on Xen! start_info_pfn=0x%lx nr_pages=%ld "
++ "flags=0x%x\n", s->arch.start_info_pfn,
++ xen_start_info->nr_pages, xen_start_info->flags);
++
++ /* xen_start_info isn't setup yet, get the flags manually */
++ if (s->arch.flags & SIF_INITDOMAIN) {
++ if (!(s->arch.flags & SIF_PRIVILEGED))
++ panic("Xen granted us console access "
++ "but not privileged status");
++ } else {
++ extern int console_use_vt;
++ conswitchp = NULL;
++ console_use_vt = 0;
++ }
++ }
++#endif
+ #endif
+
+ /* enable IA-64 Machine Check Abort Handling unless disabled */
+@@ -486,6 +537,7 @@ setup_arch (char **cmdline_p)
+
+ platform_setup(cmdline_p);
+ paging_init();
++ contiguous_bitmap_init(max_pfn);
+ }
+
+ /*
+@@ -870,6 +922,15 @@ cpu_init (void)
+ /* size of physical stacked register partition plus 8 bytes: */
+ __get_cpu_var(ia64_phys_stacked_size_p8) = num_phys_stacked*8 + 8;
+ platform_cpu_init();
++
++#ifdef CONFIG_XEN
++ /* Need to be moved into platform_cpu_init later */
++ if (is_running_on_xen()) {
++ extern void xen_smp_intr_init(void);
++ xen_smp_intr_init();
++ }
++#endif
++
+ pm_idle = default_idle;
+ }
+
+diff -urNp linux-2.6/arch/ia64/Makefile new/arch/ia64/Makefile
+--- linux-2.6/arch/ia64/Makefile 2006-07-03 14:14:15.000000000 +0200
++++ new/arch/ia64/Makefile 2006-05-09 12:32:37.000000000 +0200
+@@ -45,6 +45,12 @@ ifeq ($(call cc-version),0304)
+ endif
+
+ CFLAGS += $(cflags-y)
++
++cppflags-$(CONFIG_XEN) += \
++ -D__XEN_INTERFACE_VERSION__=$(CONFIG_XEN_INTERFACE_VERSION)
++
++CPPFLAGS += $(cppflags-y)
++
+ head-y := arch/ia64/kernel/head.o arch/ia64/kernel/init_task.o
+
+ libs-y += arch/ia64/lib/
+@@ -55,9 +61,15 @@ core-$(CONFIG_IA64_GENERIC) += arch/ia6
+ core-$(CONFIG_IA64_HP_ZX1) += arch/ia64/dig/
+ core-$(CONFIG_IA64_HP_ZX1_SWIOTLB) += arch/ia64/dig/
+ core-$(CONFIG_IA64_SGI_SN2) += arch/ia64/sn/
++core-$(CONFIG_XEN) += arch/ia64/xen/
+
+ drivers-$(CONFIG_PCI) += arch/ia64/pci/
++ifneq ($(CONFIG_XEN),y)
+ drivers-$(CONFIG_IA64_HP_SIM) += arch/ia64/hp/sim/
++endif
++ifneq ($(CONFIG_IA64_GENERIC),y)
++drivers-$(CONFIG_XEN) += arch/ia64/hp/sim/
++endif
+ drivers-$(CONFIG_IA64_HP_ZX1) += arch/ia64/hp/common/ arch/ia64/hp/zx1/
+ drivers-$(CONFIG_IA64_HP_ZX1_SWIOTLB) += arch/ia64/hp/common/ arch/ia64/hp/zx1/
+ drivers-$(CONFIG_IA64_GENERIC) += arch/ia64/hp/common/ arch/ia64/hp/zx1/ arch/ia64/hp/sim/ arch/ia64/sn/
+@@ -71,6 +83,8 @@ all: compressed unwcheck
+
+ compressed: vmlinux.gz
+
++vmlinuz: vmlinux.gz
++
+ vmlinux.gz: vmlinux
+ $(Q)$(MAKE) $(build)=$(boot) $@
+
+@@ -85,8 +99,8 @@ CLEAN_FILES += vmlinux.gz bootloader
+ boot: lib/lib.a vmlinux
+ $(Q)$(MAKE) $(build)=$(boot) $@
+
+-install: vmlinux.gz
+- sh $(srctree)/arch/ia64/install.sh $(KERNELRELEASE) $< System.map "$(INSTALL_PATH)"
++install:
++ -yes | sh $(srctree)/arch/ia64/install.sh $(KERNELRELEASE) vmlinux.gz System.map "$(INSTALL_PATH)"
+
+ define archhelp
+ echo '* compressed - Build compressed kernel image'
+diff -urNp linux-2.6/arch/ia64/mm/ioremap.c new/arch/ia64/mm/ioremap.c
+--- linux-2.6/arch/ia64/mm/ioremap.c 2006-07-03 14:14:16.000000000 +0200
++++ new/arch/ia64/mm/ioremap.c 2006-05-09 12:32:39.000000000 +0200
+@@ -15,6 +15,9 @@
+ static inline void __iomem *
+ __ioremap (unsigned long offset, unsigned long size)
+ {
++#ifdef CONFIG_XEN
++ offset = HYPERVISOR_ioremap(offset, size);
++#endif
+ return (void __iomem *) (__IA64_UNCACHED_OFFSET | offset);
+ }
+
+diff -urNp linux-2.6/arch/ia64/xen/drivers/README new/arch/ia64/xen/drivers/README
+--- linux-2.6/arch/ia64/xen/drivers/README 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/drivers/README 2006-05-09 12:32:40.000000000 +0200
+@@ -0,0 +1,2 @@
++This is a temporary location for source/Makefiles that need to be
++patched/reworked in drivers/xen to work with xenlinux/ia64.
+diff -urNp linux-2.6/arch/ia64/xen/hypercall.S new/arch/ia64/xen/hypercall.S
+--- linux-2.6/arch/ia64/xen/hypercall.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/hypercall.S 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,353 @@
++/*
++ * Support routines for Xen hypercalls
++ *
++ * Copyright (C) 2005 Dan Magenheimer <dan.magenheimer@hp.com>
++ */
++
++#include <linux/config.h>
++#include <asm/processor.h>
++#include <asm/asmmacro.h>
++
++/* To clear vpsr.ic, vpsr.i needs to be cleared first */
++#define XEN_CLEAR_PSR_IC \
++ mov r14=1; \
++ movl r15=XSI_PSR_I_ADDR; \
++ movl r2=XSI_PSR_IC; \
++ ;; \
++ ld8 r15=[r15]; \
++ ld4 r3=[r2]; \
++ ;; \
++ ld1 r16=[r15]; \
++ ;; \
++ st1 [r15]=r14; \
++ st4 [r2]=r0; \
++ ;;
++
++/* First restore vpsr.ic, and then vpsr.i */
++#define XEN_RESTORE_PSR_IC \
++ st4 [r2]=r3; \
++ st1 [r15]=r16; \
++ ;;
++
++GLOBAL_ENTRY(xen_get_ivr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=cr.ivr;;
++(p7) br.ret.sptk.many rp
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_IVR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_get_ivr)
++
++GLOBAL_ENTRY(xen_get_tpr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=cr.tpr;;
++(p7) br.ret.sptk.many rp
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_TPR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_get_tpr)
++
++GLOBAL_ENTRY(xen_set_tpr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov cr.tpr=r32;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_SET_TPR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_set_tpr)
++
++GLOBAL_ENTRY(xen_eoi)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov cr.eoi=r0;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_EOI
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_eoi)
++
++GLOBAL_ENTRY(xen_thash)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) thash r8=r32;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_THASH
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_thash)
++
++GLOBAL_ENTRY(xen_set_itm)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov cr.itm=r32;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_SET_ITM
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_set_itm)
++
++GLOBAL_ENTRY(xen_ptcga)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) ptc.ga r32,r33;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ mov r9=r33
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_PTC_GA
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_ptcga)
++
++GLOBAL_ENTRY(xen_get_rr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=rr[r32];;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_RR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_get_rr)
++
++GLOBAL_ENTRY(xen_set_rr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov rr[r32]=r33;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ mov r9=r33
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_SET_RR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_set_rr)
++
++GLOBAL_ENTRY(xen_set_kr)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.ne p7,p0=r8,r0;;
++(p7) br.cond.spnt.few 1f;
++ ;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar0=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar1=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar2=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar3=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar4=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar5=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar6=r9
++(p7) br.ret.sptk.many rp;;
++ cmp.eq p7,p0=r8,r0
++ adds r8=-1,r8;;
++(p7) mov ar7=r9
++(p7) br.ret.sptk.many rp;;
++
++1: mov r8=r32
++ mov r9=r33
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_SET_KR
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_set_kr)
++
++GLOBAL_ENTRY(xen_fc)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) fc r32;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_FC
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_fc)
++
++GLOBAL_ENTRY(xen_get_cpuid)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=cpuid[r32];;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_CPUID
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_get_cpuid)
++
++GLOBAL_ENTRY(xen_get_pmd)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=pmd[r32];;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_PMD
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_get_pmd)
++
++#ifdef CONFIG_IA32_SUPPORT
++GLOBAL_ENTRY(xen_get_eflag)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov r8=ar24;;
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_GET_EFLAG
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_get_eflag)
++
++// some bits aren't set if pl!=0, see SDM vol1 3.1.8
++GLOBAL_ENTRY(xen_set_eflag)
++ movl r8=running_on_xen;;
++ ld4 r8=[r8];;
++ cmp.eq p7,p0=r8,r0;;
++(p7) mov ar24=r32
++(p7) br.ret.sptk.many rp
++ ;;
++ mov r8=r32
++ ;;
++ XEN_CLEAR_PSR_IC
++ ;;
++ XEN_HYPER_SET_EFLAG
++ ;;
++ XEN_RESTORE_PSR_IC
++ ;;
++ br.ret.sptk.many rp
++END(xen_set_eflag)
++#endif
++
++GLOBAL_ENTRY(xen_send_ipi)
++ mov r14=r32
++ mov r15=r33
++ mov r2=0x400
++ break 0x1000
++ ;;
++ br.ret.sptk.many rp
++ ;;
++END(xen_send_ipi)
+diff -urNp linux-2.6/arch/ia64/xen/hypervisor.c new/arch/ia64/xen/hypervisor.c
+--- linux-2.6/arch/ia64/xen/hypervisor.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/hypervisor.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,784 @@
++/******************************************************************************
++ * include/asm-ia64/shadow.h
++ *
++ * Copyright (c) 2006 Isaku Yamahata <yamahata at valinux co jp>
++ * VA Linux Systems Japan K.K.
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++ *
++ */
++
++//#include <linux/kernel.h>
++#include <linux/spinlock.h>
++#include <linux/bootmem.h>
++#include <linux/module.h>
++#include <linux/vmalloc.h>
++#include <asm/page.h>
++#include <asm/hypervisor.h>
++#include <asm/hypercall.h>
++#include <xen/interface/memory.h>
++#include <xen/balloon.h>
++
++shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)XSI_BASE;
++EXPORT_SYMBOL(HYPERVISOR_shared_info);
++
++start_info_t *xen_start_info;
++
++int running_on_xen;
++EXPORT_SYMBOL(running_on_xen);
++
++//XXX xen/ia64 copy_from_guest() is broken.
++// This is a temporal work around until it is fixed.
++// used by balloon.c netfront.c
++
++// get_xen_guest_handle is defined only when __XEN_TOOLS__ is defined
++// if the definition in arch-ia64.h is changed, this must be updated.
++#define get_xen_guest_handle(val, hnd) do { val = (hnd).p; } while (0)
++
++int
++ia64_xenmem_reservation_op(unsigned long op,
++ struct xen_memory_reservation* reservation__)
++{
++ struct xen_memory_reservation reservation = *reservation__;
++ unsigned long* frame_list;
++ unsigned long nr_extents = reservation__->nr_extents;
++ int ret = 0;
++ get_xen_guest_handle(frame_list, reservation__->extent_start);
++
++ BUG_ON(op != XENMEM_increase_reservation &&
++ op != XENMEM_decrease_reservation &&
++ op != XENMEM_populate_physmap);
++
++ while (nr_extents > 0) {
++ int tmp_ret;
++ volatile unsigned long dummy;
++
++ set_xen_guest_handle(reservation.extent_start, frame_list);
++ reservation.nr_extents = nr_extents;
++
++ dummy = frame_list[0];// re-install tlb entry before hypercall
++ tmp_ret = ____HYPERVISOR_memory_op(op, &reservation);
++ if (tmp_ret < 0) {
++ if (ret == 0) {
++ ret = tmp_ret;
++ }
++ break;
++ }
++ if (tmp_ret == 0) {
++ //XXX dirty work around for skbuff_ctor()
++ // of a non-privileged domain,
++ if ((op == XENMEM_increase_reservation ||
++ op == XENMEM_populate_physmap) &&
++ !(xen_start_info->flags & SIF_PRIVILEGED) &&
++ reservation.extent_order > 0)
++ return ret;
++ }
++ frame_list += tmp_ret;
++ nr_extents -= tmp_ret;
++ ret += tmp_ret;
++ }
++ return ret;
++}
++
++//XXX same as i386, x86_64 contiguous_bitmap_set(), contiguous_bitmap_clear()
++// move those to lib/contiguous_bitmap?
++//XXX discontigmem/sparsemem
++
++/*
++ * Bitmap is indexed by page number. If bit is set, the page is part of a
++ * xen_create_contiguous_region() area of memory.
++ */
++unsigned long *contiguous_bitmap;
++
++void
++contiguous_bitmap_init(unsigned long end_pfn)
++{
++ unsigned long size = (end_pfn + 2 * BITS_PER_LONG) >> 3;
++ contiguous_bitmap = alloc_bootmem_low_pages(size);
++ BUG_ON(!contiguous_bitmap);
++ memset(contiguous_bitmap, 0, size);
++}
++
++#if 0
++int
++contiguous_bitmap_test(void* p)
++{
++ return test_bit(__pa(p) >> PAGE_SHIFT, contiguous_bitmap);
++}
++#endif
++
++static void contiguous_bitmap_set(
++ unsigned long first_page, unsigned long nr_pages)
++{
++ unsigned long start_off, end_off, curr_idx, end_idx;
++
++ curr_idx = first_page / BITS_PER_LONG;
++ start_off = first_page & (BITS_PER_LONG-1);
++ end_idx = (first_page + nr_pages) / BITS_PER_LONG;
++ end_off = (first_page + nr_pages) & (BITS_PER_LONG-1);
++
++ if (curr_idx == end_idx) {
++ contiguous_bitmap[curr_idx] |=
++ ((1UL<<end_off)-1) & -(1UL<<start_off);
++ } else {
++ contiguous_bitmap[curr_idx] |= -(1UL<<start_off);
++ while ( ++curr_idx < end_idx )
++ contiguous_bitmap[curr_idx] = ~0UL;
++ contiguous_bitmap[curr_idx] |= (1UL<<end_off)-1;
++ }
++}
++
++static void contiguous_bitmap_clear(
++ unsigned long first_page, unsigned long nr_pages)
++{
++ unsigned long start_off, end_off, curr_idx, end_idx;
++
++ curr_idx = first_page / BITS_PER_LONG;
++ start_off = first_page & (BITS_PER_LONG-1);
++ end_idx = (first_page + nr_pages) / BITS_PER_LONG;
++ end_off = (first_page + nr_pages) & (BITS_PER_LONG-1);
++
++ if (curr_idx == end_idx) {
++ contiguous_bitmap[curr_idx] &=
++ -(1UL<<end_off) | ((1UL<<start_off)-1);
++ } else {
++ contiguous_bitmap[curr_idx] &= (1UL<<start_off)-1;
++ while ( ++curr_idx != end_idx )
++ contiguous_bitmap[curr_idx] = 0;
++ contiguous_bitmap[curr_idx] &= -(1UL<<end_off);
++ }
++}
++
++static unsigned long
++HYPERVISOR_populate_physmap(unsigned long gpfn, unsigned int extent_order,
++ unsigned int address_bits)
++{
++ unsigned long ret;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = 1,
++ .address_bits = address_bits,
++ .extent_order = extent_order,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, &gpfn);
++ ret = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
++ // it may fail on non-privileged domain with extent_order > 0.
++ BUG_ON(ret != 1 &&
++ !(ret == 0 && !(xen_start_info->flags & SIF_PRIVILEGED) &&
++ extent_order > 0));
++ if (ret != 1)
++ return -EINVAL;//XXX
++ return 0;
++}
++
++static unsigned long
++HYPERVISOR_remove_physmap(unsigned long gpfn, unsigned int extent_order)
++{
++ unsigned long ret;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = 1,
++ .address_bits = 0,
++ .extent_order = extent_order,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, &gpfn);
++ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
++ BUG_ON(ret != 1);
++ return 0;
++}
++
++/* Ensure multi-page extents are contiguous in machine memory. */
++int
++__xen_create_contiguous_region(unsigned long vstart,
++ unsigned int order, unsigned int address_bits)
++{
++ unsigned long error = 0;
++ unsigned long gphys = __pa(vstart);
++ unsigned long start_gpfn = gphys >> PAGE_SHIFT;
++ unsigned long num_gpfn = 1 << order;
++ unsigned long i;
++ unsigned long flags;
++
++ scrub_pages(vstart, num_gpfn);
++
++ balloon_lock(flags);
++
++ error = HYPERVISOR_remove_physmap(start_gpfn, order);
++ if (error) {
++ goto fail;
++ }
++
++ error = HYPERVISOR_populate_physmap(start_gpfn, order, address_bits);
++ if (error) {
++ goto fail;
++ }
++ contiguous_bitmap_set(start_gpfn, num_gpfn);
++#if 0
++ {
++ unsigned long mfn;
++ unsigned long mfn_prev = ~0UL;
++ for (i = 0; i < num_gpfn; i++) {
++ mfn = pfn_to_mfn_for_dma(start_gpfn + i);
++ if (mfn_prev != ~0UL && mfn != mfn_prev + 1) {
++ xprintk("\n");
++ xprintk("%s:%d order %d "
++ "start 0x%lx bus 0x%lx machine 0x%lx\n",
++ __func__, __LINE__, order,
++ vstart, virt_to_bus((void*)vstart),
++ phys_to_machine_for_dma(gphys));
++ xprintk("mfn: ");
++ for (i = 0; i < num_gpfn; i++) {
++ mfn = pfn_to_mfn_for_dma(start_gpfn + i);
++ xprintk("0x%lx ", mfn);
++ }
++ xprintk("\n");
++ goto out;
++ }
++ mfn_prev = mfn;
++ }
++ }
++#endif
++out:
++ balloon_unlock(flags);
++ return error;
++
++fail:
++ for (i = 0; i < num_gpfn; i++) {
++ error = HYPERVISOR_populate_physmap(start_gpfn + i, 0, 0);
++ if (error) {
++ BUG();//XXX
++ }
++ }
++ goto out;
++}
++
++void
++__xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
++{
++ unsigned long flags;
++ unsigned long error = 0;
++ unsigned long start_gpfn = __pa(vstart) >> PAGE_SHIFT;
++ unsigned long num_gpfn = 1UL << order;
++ unsigned long* gpfns;
++ struct xen_memory_reservation reservation;
++ unsigned long i;
++
++ gpfns = kmalloc(sizeof(gpfns[0]) * num_gpfn,
++ GFP_KERNEL | __GFP_NOFAIL);
++ for (i = 0; i < num_gpfn; i++) {
++ gpfns[i] = start_gpfn + i;
++ }
++
++ scrub_pages(vstart, num_gpfn);
++
++ balloon_lock(flags);
++
++ contiguous_bitmap_clear(start_gpfn, num_gpfn);
++ error = HYPERVISOR_remove_physmap(start_gpfn, order);
++ if (error) {
++ goto fail;
++ }
++
++ set_xen_guest_handle(reservation.extent_start, gpfns);
++ reservation.nr_extents = num_gpfn;
++ reservation.address_bits = 0;
++ reservation.extent_order = 0;
++ reservation.domid = DOMID_SELF;
++ error = HYPERVISOR_memory_op(XENMEM_populate_physmap, &reservation);
++ if (error != num_gpfn) {
++ error = -EFAULT;//XXX
++ goto fail;
++ }
++ error = 0;
++out:
++ balloon_unlock(flags);
++ kfree(gpfns);
++ if (error) {
++ // error can't be returned.
++ BUG();//XXX
++ }
++ return;
++
++fail:
++ for (i = 0; i < num_gpfn; i++) {
++ int tmp_error;// don't overwrite error.
++ tmp_error = HYPERVISOR_populate_physmap(start_gpfn + i, 0, 0);
++ if (tmp_error) {
++ BUG();//XXX
++ }
++ }
++ goto out;
++}
++
++
++///////////////////////////////////////////////////////////////////////////
++// grant table hack
++// cmd: GNTTABOP_xxx
++
++#include <linux/mm.h>
++#include <xen/interface/xen.h>
++#include <xen/gnttab.h>
++
++static void
++gnttab_map_grant_ref_pre(struct gnttab_map_grant_ref *uop)
++{
++ uint32_t flags;
++
++ flags = uop->flags;
++
++ if (flags & GNTMAP_host_map) {
++ if (flags & GNTMAP_application_map) {
++ xprintd("GNTMAP_application_map is not supported yet: flags 0x%x\n", flags);
++ BUG();
++ }
++ if (flags & GNTMAP_contains_pte) {
++ xprintd("GNTMAP_contains_pte is not supported yet flags 0x%x\n", flags);
++ BUG();
++ }
++ } else if (flags & GNTMAP_device_map) {
++ xprintd("GNTMAP_device_map is not supported yet 0x%x\n", flags);
++ BUG();//XXX not yet. actually this flag is not used.
++ } else {
++ BUG();
++ }
++}
++
++int
++HYPERVISOR_grant_table_op(unsigned int cmd, void *uop, unsigned int count)
++{
++ if (cmd == GNTTABOP_map_grant_ref) {
++ unsigned int i;
++ for (i = 0; i < count; i++) {
++ gnttab_map_grant_ref_pre(
++ (struct gnttab_map_grant_ref*)uop + i);
++ }
++ }
++
++ return ____HYPERVISOR_grant_table_op(cmd, uop, count);
++}
++
++
++///////////////////////////////////////////////////////////////////////////
++// PageForeign(), SetPageForeign(), ClearPageForeign()
++
++struct address_space xen_ia64_foreign_dummy_mapping;
++
++///////////////////////////////////////////////////////////////////////////
++// foreign mapping
++#include <linux/efi.h>
++#include <asm/meminit.h> // for IA64_GRANULE_SIZE, GRANULEROUND{UP,DOWN}()
++
++static unsigned long privcmd_resource_min = 0;
++// Xen/ia64 currently can handle pseudo physical address bits up to
++// (PAGE_SHIFT * 3)
++static unsigned long privcmd_resource_max = GRANULEROUNDDOWN((1UL << (PAGE_SHIFT * 3)) - 1);
++static unsigned long privcmd_resource_align = IA64_GRANULE_SIZE;
++
++static unsigned long
++md_end_addr(const efi_memory_desc_t *md)
++{
++ return md->phys_addr + (md->num_pages << EFI_PAGE_SHIFT);
++}
++
++#define XEN_IA64_PRIVCMD_LEAST_GAP_SIZE (1024 * 1024 * 1024UL)
++static int
++xen_ia64_privcmd_check_size(unsigned long start, unsigned long end)
++{
++ return (start < end &&
++ (end - start) > XEN_IA64_PRIVCMD_LEAST_GAP_SIZE);
++}
++
++static int __init
++xen_ia64_privcmd_init(void)
++{
++ void *efi_map_start, *efi_map_end, *p;
++ u64 efi_desc_size;
++ efi_memory_desc_t *md;
++ unsigned long tmp_min;
++ unsigned long tmp_max;
++ unsigned long gap_size;
++ unsigned long prev_end;
++
++ if (!is_running_on_xen())
++ return -1;
++
++ efi_map_start = __va(ia64_boot_param->efi_memmap);
++ efi_map_end = efi_map_start + ia64_boot_param->efi_memmap_size;
++ efi_desc_size = ia64_boot_param->efi_memdesc_size;
++
++ // at first check the used highest address
++ for (p = efi_map_start; p < efi_map_end; p += efi_desc_size) {
++ // nothing
++ }
++ md = p - efi_desc_size;
++ privcmd_resource_min = GRANULEROUNDUP(md_end_addr(md));
++ if (xen_ia64_privcmd_check_size(privcmd_resource_min,
++ privcmd_resource_max)) {
++ goto out;
++ }
++
++ // the used highest address is too large. try to find the largest gap.
++ tmp_min = privcmd_resource_max;
++ tmp_max = 0;
++ gap_size = 0;
++ prev_end = 0;
++ for (p = efi_map_start;
++ p < efi_map_end - efi_desc_size;
++ p += efi_desc_size) {
++ unsigned long end;
++ efi_memory_desc_t* next;
++ unsigned long next_start;
++
++ md = p;
++ end = md_end_addr(md);
++ if (end > privcmd_resource_max) {
++ break;
++ }
++ if (end < prev_end) {
++ // work around.
++ // Xen may pass incompletely sorted memory
++ // descriptors like
++ // [x, x + length]
++ // [x, x]
++ // this order should be reversed.
++ continue;
++ }
++ next = p + efi_desc_size;
++ next_start = next->phys_addr;
++ if (next_start > privcmd_resource_max) {
++ next_start = privcmd_resource_max;
++ }
++ if (end < next_start && gap_size < (next_start - end)) {
++ tmp_min = end;
++ tmp_max = next_start;
++ gap_size = tmp_max - tmp_min;
++ }
++ prev_end = end;
++ }
++
++ privcmd_resource_min = GRANULEROUNDUP(tmp_min);
++ if (xen_ia64_privcmd_check_size(privcmd_resource_min, tmp_max)) {
++ privcmd_resource_max = tmp_max;
++ goto out;
++ }
++
++ privcmd_resource_min = tmp_min;
++ privcmd_resource_max = tmp_max;
++ if (!xen_ia64_privcmd_check_size(privcmd_resource_min,
++ privcmd_resource_max)) {
++ // Any large enough gap isn't found.
++ // go ahead anyway with the warning hoping that large region
++ // won't be requested.
++ printk(KERN_WARNING "xen privcmd: large enough region for privcmd mmap is not found.\n");
++ }
++
++out:
++ printk(KERN_INFO "xen privcmd uses pseudo physical addr range [0x%lx, 0x%lx] (%ldMB)\n",
++ privcmd_resource_min, privcmd_resource_max,
++ (privcmd_resource_max - privcmd_resource_min) >> 20);
++ BUG_ON(privcmd_resource_min >= privcmd_resource_max);
++ return 0;
++}
++late_initcall(xen_ia64_privcmd_init);
++
++struct xen_ia64_privcmd_entry {
++ atomic_t map_count;
++#define INVALID_GPFN (~0UL)
++ unsigned long gpfn;
++};
++
++struct xen_ia64_privcmd_range {
++ atomic_t ref_count;
++ unsigned long pgoff; // in PAGE_SIZE
++ struct resource* res;
++
++ unsigned long num_entries;
++ struct xen_ia64_privcmd_entry entries[0];
++};
++
++struct xen_ia64_privcmd_vma {
++ struct xen_ia64_privcmd_range* range;
++
++ unsigned long num_entries;
++ struct xen_ia64_privcmd_entry* entries;
++};
++
++static void
++xen_ia64_privcmd_init_entry(struct xen_ia64_privcmd_entry* entry)
++{
++ atomic_set(&entry->map_count, 0);
++ entry->gpfn = INVALID_GPFN;
++}
++
++static int
++xen_ia64_privcmd_entry_mmap(struct vm_area_struct* vma,
++ unsigned long addr,
++ struct xen_ia64_privcmd_range* privcmd_range,
++ int i,
++ unsigned long mfn,
++ pgprot_t prot,
++ domid_t domid)
++{
++ int error = 0;
++ struct xen_ia64_privcmd_entry* entry = &privcmd_range->entries[i];
++ unsigned long gpfn;
++ unsigned long flags;
++
++ BUG_ON((addr & ~PAGE_MASK) != 0);
++ BUG_ON(mfn == INVALID_MFN);
++
++ if (entry->gpfn != INVALID_GPFN) {
++ error = -EBUSY;
++ goto out;
++ }
++ gpfn = (privcmd_range->res->start >> PAGE_SHIFT) + i;
++
++ flags = ASSIGN_writable;
++ if (pgprot_val(prot) == PROT_READ) {
++ flags = ASSIGN_readonly;
++ }
++ error = HYPERVISOR_add_physmap(gpfn, mfn, flags, domid);
++ if (error != 0) {
++ goto out;
++ }
++
++ prot = vma->vm_page_prot;
++ error = remap_pfn_range(vma, addr, gpfn, 1 << PAGE_SHIFT, prot);
++ if (error != 0) {
++ error = HYPERVISOR_zap_physmap(gpfn, 0);
++ if (error) {
++ BUG();//XXX
++ }
++ } else {
++ atomic_inc(&entry->map_count);
++ entry->gpfn = gpfn;
++ }
++
++out:
++ return error;
++}
++
++static void
++xen_ia64_privcmd_entry_munmap(struct xen_ia64_privcmd_range* privcmd_range,
++ int i)
++{
++ struct xen_ia64_privcmd_entry* entry = &privcmd_range->entries[i];
++ unsigned long gpfn = entry->gpfn;
++ //gpfn = (privcmd_range->res->start >> PAGE_SHIFT) +
++ // (vma->vm_pgoff - privcmd_range->pgoff);
++ int error;
++
++ error = HYPERVISOR_zap_physmap(gpfn, 0);
++ if (error) {
++ BUG();//XXX
++ }
++ entry->gpfn = INVALID_GPFN;
++}
++
++static void
++xen_ia64_privcmd_entry_open(struct xen_ia64_privcmd_range* privcmd_range,
++ int i)
++{
++ struct xen_ia64_privcmd_entry* entry = &privcmd_range->entries[i];
++ if (entry->gpfn != INVALID_GPFN) {
++ atomic_inc(&entry->map_count);
++ } else {
++ BUG_ON(atomic_read(&entry->map_count) != 0);
++ }
++}
++
++static void
++xen_ia64_privcmd_entry_close(struct xen_ia64_privcmd_range* privcmd_range,
++ int i)
++{
++ struct xen_ia64_privcmd_entry* entry = &privcmd_range->entries[i];
++ if (entry->gpfn != INVALID_GPFN &&
++ atomic_dec_and_test(&entry->map_count)) {
++ xen_ia64_privcmd_entry_munmap(privcmd_range, i);
++ }
++}
++
++static void xen_ia64_privcmd_vma_open(struct vm_area_struct* vma);
++static void xen_ia64_privcmd_vma_close(struct vm_area_struct* vma);
++
++struct vm_operations_struct xen_ia64_privcmd_vm_ops = {
++ .open = &xen_ia64_privcmd_vma_open,
++ .close = &xen_ia64_privcmd_vma_close,
++};
++
++static void
++__xen_ia64_privcmd_vma_open(struct vm_area_struct* vma,
++ struct xen_ia64_privcmd_vma* privcmd_vma,
++ struct xen_ia64_privcmd_range* privcmd_range)
++{
++ unsigned long entry_offset = vma->vm_pgoff - privcmd_range->pgoff;
++ unsigned long num_entries = (vma->vm_end - vma->vm_start) >> PAGE_SHIFT;
++ unsigned long i;
++
++ BUG_ON(entry_offset < 0);
++ BUG_ON(entry_offset + num_entries > privcmd_range->num_entries);
++
++ privcmd_vma->range = privcmd_range;
++ privcmd_vma->num_entries = num_entries;
++ privcmd_vma->entries = &privcmd_range->entries[entry_offset];
++ vma->vm_private_data = privcmd_vma;
++ for (i = 0; i < privcmd_vma->num_entries; i++) {
++ xen_ia64_privcmd_entry_open(privcmd_range, entry_offset + i);
++ }
++
++ vma->vm_private_data = privcmd_vma;
++ vma->vm_ops = &xen_ia64_privcmd_vm_ops;
++}
++
++static void
++xen_ia64_privcmd_vma_open(struct vm_area_struct* vma)
++{
++ struct xen_ia64_privcmd_vma* privcmd_vma = (struct xen_ia64_privcmd_vma*)vma->vm_private_data;
++ struct xen_ia64_privcmd_range* privcmd_range = privcmd_vma->range;
++
++ atomic_inc(&privcmd_range->ref_count);
++ // vm_op->open() can't fail.
++ privcmd_vma = kmalloc(sizeof(*privcmd_vma), GFP_KERNEL | __GFP_NOFAIL);
++
++ __xen_ia64_privcmd_vma_open(vma, privcmd_vma, privcmd_range);
++}
++
++static void
++xen_ia64_privcmd_vma_close(struct vm_area_struct* vma)
++{
++ struct xen_ia64_privcmd_vma* privcmd_vma =
++ (struct xen_ia64_privcmd_vma*)vma->vm_private_data;
++ struct xen_ia64_privcmd_range* privcmd_range = privcmd_vma->range;
++ unsigned long entry_offset = vma->vm_pgoff - privcmd_range->pgoff;
++ unsigned long i;
++
++ for (i = 0; i < privcmd_vma->num_entries; i++) {
++ xen_ia64_privcmd_entry_close(privcmd_range, entry_offset + i);
++ }
++ vma->vm_private_data = NULL;
++ kfree(privcmd_vma);
++
++ if (atomic_dec_and_test(&privcmd_range->ref_count)) {
++#if 1
++ for (i = 0; i < privcmd_range->num_entries; i++) {
++ struct xen_ia64_privcmd_entry* entry =
++ &privcmd_range->entries[i];
++ BUG_ON(atomic_read(&entry->map_count) != 0);
++ BUG_ON(entry->gpfn != INVALID_GPFN);
++ }
++#endif
++ release_resource(privcmd_range->res);
++ kfree(privcmd_range->res);
++ vfree(privcmd_range);
++ }
++}
++
++int
++privcmd_mmap(struct file * file, struct vm_area_struct * vma)
++{
++ int error;
++ unsigned long size = vma->vm_end - vma->vm_start;
++ unsigned long num_entries = size >> PAGE_SHIFT;
++ struct xen_ia64_privcmd_range* privcmd_range = NULL;
++ struct xen_ia64_privcmd_vma* privcmd_vma = NULL;
++ struct resource* res = NULL;
++ unsigned long i;
++ BUG_ON(!is_running_on_xen());
++
++ BUG_ON(file->private_data != NULL);
++
++ error = -ENOMEM;
++ privcmd_range =
++ vmalloc(sizeof(*privcmd_range) +
++ sizeof(privcmd_range->entries[0]) * num_entries);
++ if (privcmd_range == NULL) {
++ goto out_enomem0;
++ }
++ privcmd_vma = kmalloc(sizeof(*privcmd_vma), GFP_KERNEL);
++ if (privcmd_vma == NULL) {
++ goto out_enomem1;
++ }
++ res = kzalloc(sizeof(*res), GFP_KERNEL);
++ if (res == NULL) {
++ goto out_enomem1;
++ }
++ res->name = "Xen privcmd mmap";
++ error = allocate_resource(&iomem_resource, res, size,
++ privcmd_resource_min, privcmd_resource_max,
++ privcmd_resource_align, NULL, NULL);
++ if (error) {
++ goto out_enomem1;
++ }
++ privcmd_range->res = res;
++
++ /* DONTCOPY is essential for Xen as copy_page_range is broken. */
++ vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY | VM_PFNMAP;
++
++ atomic_set(&privcmd_range->ref_count, 1);
++ privcmd_range->pgoff = vma->vm_pgoff;
++ privcmd_range->num_entries = num_entries;
++ for (i = 0; i < privcmd_range->num_entries; i++) {
++ xen_ia64_privcmd_init_entry(&privcmd_range->entries[i]);
++ }
++
++ __xen_ia64_privcmd_vma_open(vma, privcmd_vma, privcmd_range);
++ return 0;
++
++out_enomem1:
++ kfree(res);
++ kfree(privcmd_vma);
++out_enomem0:
++ vfree(privcmd_range);
++ return error;
++}
++
++int
++direct_remap_pfn_range(struct vm_area_struct *vma,
++ unsigned long address, // process virtual address
++ unsigned long mfn, // mfn, mfn + 1, ... mfn + size/PAGE_SIZE
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid) // target domain
++{
++ struct xen_ia64_privcmd_vma* privcmd_vma =
++ (struct xen_ia64_privcmd_vma*)vma->vm_private_data;
++ struct xen_ia64_privcmd_range* privcmd_range = privcmd_vma->range;
++ unsigned long entry_offset = vma->vm_pgoff - privcmd_range->pgoff;
++
++ unsigned long i;
++ unsigned long offset;
++ int error = 0;
++ BUG_ON(!is_running_on_xen());
++
++#if 0
++ if (prot != vm->vm_page_prot) {
++ return -EINVAL;
++ }
++#endif
++
++ i = (address - vma->vm_start) >> PAGE_SHIFT;
++ for (offset = 0; offset < size; offset += PAGE_SIZE) {
++ error = xen_ia64_privcmd_entry_mmap(vma, (address + offset) & PAGE_MASK, privcmd_range, entry_offset + i, mfn, prot, domid);
++ if (error != 0) {
++ break;
++ }
++
++ i++;
++ mfn++;
++ }
++
++ return error;
++}
++
+diff -urNp linux-2.6/arch/ia64/xen/Makefile new/arch/ia64/xen/Makefile
+--- linux-2.6/arch/ia64/xen/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/Makefile 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,8 @@
++#
++# Makefile for Xen components
++#
++
++obj-y := hypercall.o xenivt.o xenentry.o xensetup.o xenpal.o xenhpski.o
++
++obj-$(CONFIG_XEN_IA64_DOM0_VP) += hypervisor.o pci-dma-xen.o util.o
++pci-dma-xen-$(CONFIG_XEN_IA64_DOM0_VP) := ../../i386/kernel/pci-dma-xen.o
+diff -urNp linux-2.6/arch/ia64/xen/util.c new/arch/ia64/xen/util.c
+--- linux-2.6/arch/ia64/xen/util.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/util.c 2006-06-28 14:32:13.000000000 +0200
+@@ -0,0 +1,130 @@
++/******************************************************************************
++ * arch/ia64/xen/util.c
++ * This file is the ia64 counterpart of drivers/xen/util.c
++ *
++ * Copyright (c) 2006 Isaku Yamahata <yamahata at valinux co jp>
++ * VA Linux Systems Japan K.K.
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++ *
++ */
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <linux/module.h>
++#include <linux/slab.h>
++#include <linux/vmalloc.h>
++#include <asm/uaccess.h>
++#include <xen/driver_util.h>
++
++struct vm_struct *alloc_vm_area(unsigned long size)
++{
++ int order;
++ unsigned long virt;
++ unsigned long nr_pages;
++ struct vm_struct* area;
++
++ order = get_order(size);
++ virt = __get_free_pages(GFP_KERNEL, order);
++ if (virt == 0) {
++ goto err0;
++ }
++ nr_pages = 1 << order;
++ scrub_pages(virt, nr_pages);
++
++ area = kmalloc(sizeof(*area), GFP_KERNEL);
++ if (area == NULL) {
++ goto err1;
++ }
++
++ area->flags = VM_IOREMAP;//XXX
++ area->addr = (void*)virt;
++ area->size = size;
++ area->pages = NULL; //XXX
++ area->nr_pages = nr_pages;
++ area->phys_addr = __pa(virt);
++
++ return area;
++
++err1:
++ free_pages(virt, order);
++err0:
++ return NULL;
++
++}
++EXPORT_SYMBOL_GPL(alloc_vm_area);
++
++void free_vm_area(struct vm_struct *area)
++{
++ unsigned int order = get_order(area->size);
++ unsigned long i;
++
++ // This area is used for foreign page mappping.
++ // So underlying machine page may not be assigned.
++ for (i = 0; i < (1 << order); i++) {
++ unsigned long ret;
++ unsigned long gpfn = (area->phys_addr >> PAGE_SHIFT) + i;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = 1,
++ .address_bits = 0,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, &gpfn);
++ ret = HYPERVISOR_memory_op(XENMEM_populate_physmap,
++ &reservation);
++ BUG_ON(ret != 1);
++ }
++ free_pages((unsigned long)area->addr, order);
++ kfree(area);
++}
++EXPORT_SYMBOL_GPL(free_vm_area);
++
++void lock_vm_area(struct vm_struct *area)
++{
++ // nothing
++}
++EXPORT_SYMBOL_GPL(lock_vm_area);
++
++void unlock_vm_area(struct vm_struct *area)
++{
++ // nothing
++}
++EXPORT_SYMBOL_GPL(unlock_vm_area);
++
++#ifndef CONFIG_XEN_IA64_DOM0_VP
++/* We just need a range of legal va here, though finally identity
++ * mapped one is instead used for gnttab mapping.
++ */
++unsigned long alloc_empty_foreign_map_page_range(unsigned long pages)
++{
++ struct vm_struct *vma;
++
++ if ( (vma = get_vm_area(PAGE_SIZE * pages, VM_ALLOC)) == NULL )
++ return NULL;
++
++ return (unsigned long)vma->addr;
++}
++#endif
++
++/*
++ * Local variables:
++ * c-file-style: "linux"
++ * indent-tabs-mode: t
++ * c-indent-level: 8
++ * c-basic-offset: 8
++ * tab-width: 8
++ * End:
++ */
+diff -urNp linux-2.6/arch/ia64/xen/xenentry.S new/arch/ia64/xen/xenentry.S
+--- linux-2.6/arch/ia64/xen/xenentry.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xenentry.S 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,883 @@
++/*
++ * ia64/xen/entry.S
++ *
++ * Alternate kernel routines for Xen. Heavily leveraged from
++ * ia64/kernel/entry.S
++ *
++ * Copyright (C) 2005 Hewlett-Packard Co
++ * Dan Magenheimer <dan.magenheimer@.hp.com>
++ */
++
++#include <linux/config.h>
++
++#include <asm/asmmacro.h>
++#include <asm/cache.h>
++#include <asm/errno.h>
++#include <asm/kregs.h>
++#include <asm/asm-offsets.h>
++#include <asm/pgtable.h>
++#include <asm/percpu.h>
++#include <asm/processor.h>
++#include <asm/thread_info.h>
++#include <asm/unistd.h>
++
++#ifdef CONFIG_XEN
++#include "xenminstate.h"
++#else
++#include "minstate.h"
++#endif
++
++/*
++ * prev_task <- ia64_switch_to(struct task_struct *next)
++ * With Ingo's new scheduler, interrupts are disabled when this routine gets
++ * called. The code starting at .map relies on this. The rest of the code
++ * doesn't care about the interrupt masking status.
++ */
++#ifdef CONFIG_XEN
++GLOBAL_ENTRY(xen_switch_to)
++ .prologue
++ alloc r16=ar.pfs,1,0,0,0
++ movl r22=running_on_xen;;
++ ld4 r22=[r22];;
++ cmp.eq p7,p0=r22,r0
++(p7) br.cond.sptk.many __ia64_switch_to;;
++#else
++GLOBAL_ENTRY(ia64_switch_to)
++ .prologue
++ alloc r16=ar.pfs,1,0,0,0
++#endif
++ DO_SAVE_SWITCH_STACK
++ .body
++
++ adds r22=IA64_TASK_THREAD_KSP_OFFSET,r13
++ movl r25=init_task
++ mov r27=IA64_KR(CURRENT_STACK)
++ adds r21=IA64_TASK_THREAD_KSP_OFFSET,in0
++ dep r20=0,in0,61,3 // physical address of "next"
++ ;;
++ st8 [r22]=sp // save kernel stack pointer of old task
++ shr.u r26=r20,IA64_GRANULE_SHIFT
++ cmp.eq p7,p6=r25,in0
++ ;;
++#ifdef CONFIG_XEN
++ movl r8=XSI_PSR_IC
++ ;;
++ st4 [r8]=r0 // force psr.ic off for hyperprivop(s)
++ ;;
++#endif
++ /*
++ * If we've already mapped this task's page, we can skip doing it again.
++ */
++(p6) cmp.eq p7,p6=r26,r27
++(p6) br.cond.dpnt .map
++ ;;
++.done:
++#ifdef CONFIG_XEN
++ // psr.ic already off
++ // update "current" application register
++ mov r8=IA64_KR_CURRENT
++ mov r9=in0;;
++ XEN_HYPER_SET_KR
++ ld8 sp=[r21] // load kernel stack pointer of new task
++ movl r27=XSI_PSR_IC
++ mov r8=1
++ ;;
++ st4 [r27]=r8 // psr.ic back on
++#else
++ ld8 sp=[r21] // load kernel stack pointer of new task
++ mov IA64_KR(CURRENT)=in0 // update "current" application register
++#endif
++ mov r8=r13 // return pointer to previously running task
++ mov r13=in0 // set "current" pointer
++ ;;
++ DO_LOAD_SWITCH_STACK
++
++#ifdef CONFIG_SMP
++ sync.i // ensure "fc"s done by this CPU are visible on other CPUs
++#endif
++ br.ret.sptk.many rp // boogie on out in new context
++
++.map:
++#ifdef CONFIG_XEN
++ // psr.ic already off
++#else
++ rsm psr.ic // interrupts (psr.i) are already disabled here
++#endif
++ movl r25=PAGE_KERNEL
++ ;;
++ srlz.d
++ or r23=r25,r20 // construct PA | page properties
++ mov r25=IA64_GRANULE_SHIFT<<2
++ ;;
++#ifdef CONFIG_XEN
++ movl r8=XSI_ITIR
++ ;;
++ st8 [r8]=r25
++ ;;
++ movl r8=XSI_IFA
++ ;;
++ st8 [r8]=in0 // VA of next task...
++ ;;
++ mov r25=IA64_TR_CURRENT_STACK
++ // remember last page we mapped...
++ mov r8=IA64_KR_CURRENT_STACK
++ mov r9=r26;;
++ XEN_HYPER_SET_KR;;
++#else
++ mov cr.itir=r25
++ mov cr.ifa=in0 // VA of next task...
++ ;;
++ mov r25=IA64_TR_CURRENT_STACK
++ mov IA64_KR(CURRENT_STACK)=r26 // remember last page we mapped...
++#endif
++ ;;
++ itr.d dtr[r25]=r23 // wire in new mapping...
++#ifndef CONFIG_XEN
++ ssm psr.ic // reenable the psr.ic bit
++ ;;
++ srlz.d
++#endif
++ br.cond.sptk .done
++#ifdef CONFIG_XEN
++END(xen_switch_to)
++#else
++END(ia64_switch_to)
++#endif
++
++ /*
++ * Invoke a system call, but do some tracing before and after the call.
++ * We MUST preserve the current register frame throughout this routine
++ * because some system calls (such as ia64_execve) directly
++ * manipulate ar.pfs.
++ */
++#ifdef CONFIG_XEN
++GLOBAL_ENTRY(xen_trace_syscall)
++ PT_REGS_UNWIND_INFO(0)
++ movl r16=running_on_xen;;
++ ld4 r16=[r16];;
++ cmp.eq p7,p0=r16,r0
++(p7) br.cond.sptk.many __ia64_trace_syscall;;
++#else
++GLOBAL_ENTRY(ia64_trace_syscall)
++ PT_REGS_UNWIND_INFO(0)
++#endif
++ /*
++ * We need to preserve the scratch registers f6-f11 in case the system
++ * call is sigreturn.
++ */
++ adds r16=PT(F6)+16,sp
++ adds r17=PT(F7)+16,sp
++ ;;
++ stf.spill [r16]=f6,32
++ stf.spill [r17]=f7,32
++ ;;
++ stf.spill [r16]=f8,32
++ stf.spill [r17]=f9,32
++ ;;
++ stf.spill [r16]=f10
++ stf.spill [r17]=f11
++ br.call.sptk.many rp=syscall_trace_enter // give parent a chance to catch syscall args
++ adds r16=PT(F6)+16,sp
++ adds r17=PT(F7)+16,sp
++ ;;
++ ldf.fill f6=[r16],32
++ ldf.fill f7=[r17],32
++ ;;
++ ldf.fill f8=[r16],32
++ ldf.fill f9=[r17],32
++ ;;
++ ldf.fill f10=[r16]
++ ldf.fill f11=[r17]
++ // the syscall number may have changed, so re-load it and re-calculate the
++ // syscall entry-point:
++ adds r15=PT(R15)+16,sp // r15 = &pt_regs.r15 (syscall #)
++ ;;
++ ld8 r15=[r15]
++ mov r3=NR_syscalls - 1
++ ;;
++ adds r15=-1024,r15
++ movl r16=sys_call_table
++ ;;
++ shladd r20=r15,3,r16 // r20 = sys_call_table + 8*(syscall-1024)
++ cmp.leu p6,p7=r15,r3
++ ;;
++(p6) ld8 r20=[r20] // load address of syscall entry point
++(p7) movl r20=sys_ni_syscall
++ ;;
++ mov b6=r20
++ br.call.sptk.many rp=b6 // do the syscall
++.strace_check_retval:
++ cmp.lt p6,p0=r8,r0 // syscall failed?
++ adds r2=PT(R8)+16,sp // r2 = &pt_regs.r8
++ adds r3=PT(R10)+16,sp // r3 = &pt_regs.r10
++ mov r10=0
++(p6) br.cond.sptk strace_error // syscall failed ->
++ ;; // avoid RAW on r10
++.strace_save_retval:
++.mem.offset 0,0; st8.spill [r2]=r8 // store return value in slot for r8
++.mem.offset 8,0; st8.spill [r3]=r10 // clear error indication in slot for r10
++ br.call.sptk.many rp=syscall_trace_leave // give parent a chance to catch return value
++.ret3:
++(pUStk) cmp.eq.unc p6,p0=r0,r0 // p6 <- pUStk
++ br.cond.sptk .work_pending_syscall_end
++
++strace_error:
++ ld8 r3=[r2] // load pt_regs.r8
++ sub r9=0,r8 // negate return value to get errno value
++ ;;
++ cmp.ne p6,p0=r3,r0 // is pt_regs.r8!=0?
++ adds r3=16,r2 // r3=&pt_regs.r10
++ ;;
++(p6) mov r10=-1
++(p6) mov r8=r9
++ br.cond.sptk .strace_save_retval
++#ifdef CONFIG_XEN
++END(xen_trace_syscall)
++#else
++END(ia64_trace_syscall)
++#endif
++
++/*
++ * ia64_leave_syscall(): Same as ia64_leave_kernel, except that it doesn't
++ * need to switch to bank 0 and doesn't restore the scratch registers.
++ * To avoid leaking kernel bits, the scratch registers are set to
++ * the following known-to-be-safe values:
++ *
++ * r1: restored (global pointer)
++ * r2: cleared
++ * r3: 1 (when returning to user-level)
++ * r8-r11: restored (syscall return value(s))
++ * r12: restored (user-level stack pointer)
++ * r13: restored (user-level thread pointer)
++ * r14: set to __kernel_syscall_via_epc
++ * r15: restored (syscall #)
++ * r16-r17: cleared
++ * r18: user-level b6
++ * r19: cleared
++ * r20: user-level ar.fpsr
++ * r21: user-level b0
++ * r22: cleared
++ * r23: user-level ar.bspstore
++ * r24: user-level ar.rnat
++ * r25: user-level ar.unat
++ * r26: user-level ar.pfs
++ * r27: user-level ar.rsc
++ * r28: user-level ip
++ * r29: user-level psr
++ * r30: user-level cfm
++ * r31: user-level pr
++ * f6-f11: cleared
++ * pr: restored (user-level pr)
++ * b0: restored (user-level rp)
++ * b6: restored
++ * b7: set to __kernel_syscall_via_epc
++ * ar.unat: restored (user-level ar.unat)
++ * ar.pfs: restored (user-level ar.pfs)
++ * ar.rsc: restored (user-level ar.rsc)
++ * ar.rnat: restored (user-level ar.rnat)
++ * ar.bspstore: restored (user-level ar.bspstore)
++ * ar.fpsr: restored (user-level ar.fpsr)
++ * ar.ccv: cleared
++ * ar.csd: cleared
++ * ar.ssd: cleared
++ */
++#ifdef CONFIG_XEN
++GLOBAL_ENTRY(xen_leave_syscall)
++ PT_REGS_UNWIND_INFO(0)
++ movl r22=running_on_xen;;
++ ld4 r22=[r22];;
++ cmp.eq p7,p0=r22,r0
++(p7) br.cond.sptk.many __ia64_leave_syscall;;
++#else
++ENTRY(ia64_leave_syscall)
++ PT_REGS_UNWIND_INFO(0)
++#endif
++ /*
++ * work.need_resched etc. mustn't get changed by this CPU before it returns to
++ * user- or fsys-mode, hence we disable interrupts early on.
++ *
++ * p6 controls whether current_thread_info()->flags needs to be check for
++ * extra work. We always check for extra work when returning to user-level.
++ * With CONFIG_PREEMPT, we also check for extra work when the preempt_count
++ * is 0. After extra work processing has been completed, execution
++ * resumes at .work_processed_syscall with p6 set to 1 if the extra-work-check
++ * needs to be redone.
++ */
++#ifdef CONFIG_PREEMPT
++ rsm psr.i // disable interrupts
++ cmp.eq pLvSys,p0=r0,r0 // pLvSys=1: leave from syscall
++(pKStk) adds r20=TI_PRE_COUNT+IA64_TASK_SIZE,r13
++ ;;
++ .pred.rel.mutex pUStk,pKStk
++(pKStk) ld4 r21=[r20] // r21 <- preempt_count
++(pUStk) mov r21=0 // r21 <- 0
++ ;;
++ cmp.eq p6,p0=r21,r0 // p6 <- pUStk || (preempt_count == 0)
++#else /* !CONFIG_PREEMPT */
++#ifdef CONFIG_XEN
++ movl r2=XSI_PSR_I_ADDR
++ mov r18=1
++ ;;
++ ld8 r2=[r2]
++ ;;
++(pUStk) st1 [r2]=r18
++#else
++(pUStk) rsm psr.i
++#endif
++ cmp.eq pLvSys,p0=r0,r0 // pLvSys=1: leave from syscall
++(pUStk) cmp.eq.unc p6,p0=r0,r0 // p6 <- pUStk
++#endif
++.work_processed_syscall:
++ adds r2=PT(LOADRS)+16,r12
++ adds r3=PT(AR_BSPSTORE)+16,r12
++ adds r18=TI_FLAGS+IA64_TASK_SIZE,r13
++ ;;
++(p6) ld4 r31=[r18] // load current_thread_info()->flags
++ ld8 r19=[r2],PT(B6)-PT(LOADRS) // load ar.rsc value for "loadrs"
++ nop.i 0
++ ;;
++ mov r16=ar.bsp // M2 get existing backing store pointer
++ ld8 r18=[r2],PT(R9)-PT(B6) // load b6
++(p6) and r15=TIF_WORK_MASK,r31 // any work other than TIF_SYSCALL_TRACE?
++ ;;
++ ld8 r23=[r3],PT(R11)-PT(AR_BSPSTORE) // load ar.bspstore (may be garbage)
++(p6) cmp4.ne.unc p6,p0=r15, r0 // any special work pending?
++(p6) br.cond.spnt .work_pending_syscall
++ ;;
++ // start restoring the state saved on the kernel stack (struct pt_regs):
++ ld8 r9=[r2],PT(CR_IPSR)-PT(R9)
++ ld8 r11=[r3],PT(CR_IIP)-PT(R11)
++(pNonSys) break 0 // bug check: we shouldn't be here if pNonSys is TRUE!
++ ;;
++ invala // M0|1 invalidate ALAT
++#ifdef CONFIG_XEN
++ movl r28=XSI_PSR_I_ADDR
++ movl r29=XSI_PSR_IC
++ ;;
++ ld8 r28=[r28]
++ mov r30=1
++ ;;
++ st1 [r28]=r30
++ st4 [r29]=r0 // note: clears both vpsr.i and vpsr.ic!
++ ;;
++#else
++ rsm psr.i | psr.ic // M2 turn off interrupts and interruption collection
++#endif
++ cmp.eq p9,p0=r0,r0 // A set p9 to indicate that we should restore cr.ifs
++
++ ld8 r29=[r2],16 // M0|1 load cr.ipsr
++ ld8 r28=[r3],16 // M0|1 load cr.iip
++ mov r22=r0 // A clear r22
++ ;;
++ ld8 r30=[r2],16 // M0|1 load cr.ifs
++ ld8 r25=[r3],16 // M0|1 load ar.unat
++(pUStk) add r14=IA64_TASK_THREAD_ON_USTACK_OFFSET,r13
++ ;;
++ ld8 r26=[r2],PT(B0)-PT(AR_PFS) // M0|1 load ar.pfs
++(pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled
++ nop 0
++ ;;
++ ld8 r21=[r2],PT(AR_RNAT)-PT(B0) // M0|1 load b0
++ ld8 r27=[r3],PT(PR)-PT(AR_RSC) // M0|1 load ar.rsc
++ mov f6=f0 // F clear f6
++ ;;
++ ld8 r24=[r2],PT(AR_FPSR)-PT(AR_RNAT) // M0|1 load ar.rnat (may be garbage)
++ ld8 r31=[r3],PT(R1)-PT(PR) // M0|1 load predicates
++ mov f7=f0 // F clear f7
++ ;;
++ ld8 r20=[r2],PT(R12)-PT(AR_FPSR) // M0|1 load ar.fpsr
++ ld8.fill r1=[r3],16 // M0|1 load r1
++(pUStk) mov r17=1 // A
++ ;;
++(pUStk) st1 [r14]=r17 // M2|3
++ ld8.fill r13=[r3],16 // M0|1
++ mov f8=f0 // F clear f8
++ ;;
++ ld8.fill r12=[r2] // M0|1 restore r12 (sp)
++ ld8.fill r15=[r3] // M0|1 restore r15
++ mov b6=r18 // I0 restore b6
++
++ addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0 // A
++ mov f9=f0 // F clear f9
++(pKStk) br.cond.dpnt.many skip_rbs_switch // B
++
++ srlz.d // M0 ensure interruption collection is off (for cover)
++ shr.u r18=r19,16 // I0|1 get byte size of existing "dirty" partition
++#ifdef CONFIG_XEN
++ XEN_HYPER_COVER;
++#else
++ cover // B add current frame into dirty partition & set cr.ifs
++#endif
++ ;;
++(pUStk) ld4 r17=[r17] // M0|1 r17 = cpu_data->phys_stacked_size_p8
++ mov r19=ar.bsp // M2 get new backing store pointer
++ mov f10=f0 // F clear f10
++
++ nop.m 0
++ movl r14=__kernel_syscall_via_epc // X
++ ;;
++ mov.m ar.csd=r0 // M2 clear ar.csd
++ mov.m ar.ccv=r0 // M2 clear ar.ccv
++ mov b7=r14 // I0 clear b7 (hint with __kernel_syscall_via_epc)
++
++ mov.m ar.ssd=r0 // M2 clear ar.ssd
++ mov f11=f0 // F clear f11
++ br.cond.sptk.many rbs_switch // B
++#ifdef CONFIG_XEN
++END(xen_leave_syscall)
++#else
++END(ia64_leave_syscall)
++#endif
++
++#ifdef CONFIG_XEN
++GLOBAL_ENTRY(xen_leave_kernel)
++ PT_REGS_UNWIND_INFO(0)
++ movl r22=running_on_xen;;
++ ld4 r22=[r22];;
++ cmp.eq p7,p0=r22,r0
++(p7) br.cond.sptk.many __ia64_leave_kernel;;
++#else
++GLOBAL_ENTRY(ia64_leave_kernel)
++ PT_REGS_UNWIND_INFO(0)
++#endif
++ /*
++ * work.need_resched etc. mustn't get changed by this CPU before it returns to
++ * user- or fsys-mode, hence we disable interrupts early on.
++ *
++ * p6 controls whether current_thread_info()->flags needs to be check for
++ * extra work. We always check for extra work when returning to user-level.
++ * With CONFIG_PREEMPT, we also check for extra work when the preempt_count
++ * is 0. After extra work processing has been completed, execution
++ * resumes at .work_processed_syscall with p6 set to 1 if the extra-work-check
++ * needs to be redone.
++ */
++#ifdef CONFIG_PREEMPT
++ rsm psr.i // disable interrupts
++ cmp.eq p0,pLvSys=r0,r0 // pLvSys=0: leave from kernel
++(pKStk) adds r20=TI_PRE_COUNT+IA64_TASK_SIZE,r13
++ ;;
++ .pred.rel.mutex pUStk,pKStk
++(pKStk) ld4 r21=[r20] // r21 <- preempt_count
++(pUStk) mov r21=0 // r21 <- 0
++ ;;
++ cmp.eq p6,p0=r21,r0 // p6 <- pUStk || (preempt_count == 0)
++#else
++#ifdef CONFIG_XEN
++(pUStk) movl r17=XSI_PSR_I_ADDR
++(pUStk) mov r31=1
++ ;;
++(pUStk) ld8 r17=[r17]
++ ;;
++(pUStk) st1 [r17]=r31
++ ;;
++#else
++(pUStk) rsm psr.i
++#endif
++ cmp.eq p0,pLvSys=r0,r0 // pLvSys=0: leave from kernel
++(pUStk) cmp.eq.unc p6,p0=r0,r0 // p6 <- pUStk
++#endif
++.work_processed_kernel:
++ adds r17=TI_FLAGS+IA64_TASK_SIZE,r13
++ ;;
++(p6) ld4 r31=[r17] // load current_thread_info()->flags
++ adds r21=PT(PR)+16,r12
++ ;;
++
++ lfetch [r21],PT(CR_IPSR)-PT(PR)
++ adds r2=PT(B6)+16,r12
++ adds r3=PT(R16)+16,r12
++ ;;
++ lfetch [r21]
++ ld8 r28=[r2],8 // load b6
++ adds r29=PT(R24)+16,r12
++
++ ld8.fill r16=[r3],PT(AR_CSD)-PT(R16)
++ adds r30=PT(AR_CCV)+16,r12
++(p6) and r19=TIF_WORK_MASK,r31 // any work other than TIF_SYSCALL_TRACE?
++ ;;
++ ld8.fill r24=[r29]
++ ld8 r15=[r30] // load ar.ccv
++(p6) cmp4.ne.unc p6,p0=r19, r0 // any special work pending?
++ ;;
++ ld8 r29=[r2],16 // load b7
++ ld8 r30=[r3],16 // load ar.csd
++(p6) br.cond.spnt .work_pending
++ ;;
++ ld8 r31=[r2],16 // load ar.ssd
++ ld8.fill r8=[r3],16
++ ;;
++ ld8.fill r9=[r2],16
++ ld8.fill r10=[r3],PT(R17)-PT(R10)
++ ;;
++ ld8.fill r11=[r2],PT(R18)-PT(R11)
++ ld8.fill r17=[r3],16
++ ;;
++ ld8.fill r18=[r2],16
++ ld8.fill r19=[r3],16
++ ;;
++ ld8.fill r20=[r2],16
++ ld8.fill r21=[r3],16
++ mov ar.csd=r30
++ mov ar.ssd=r31
++ ;;
++#ifdef CONFIG_XEN
++ movl r23=XSI_PSR_I_ADDR
++ movl r22=XSI_PSR_IC
++ ;;
++ ld8 r23=[r23]
++ mov r25=1
++ ;;
++ st1 [r23]=r25
++ st4 [r22]=r0 // note: clears both vpsr.i and vpsr.ic!
++ ;;
++#else
++ rsm psr.i | psr.ic // initiate turning off of interrupt and interruption collection
++#endif
++ invala // invalidate ALAT
++ ;;
++ ld8.fill r22=[r2],24
++ ld8.fill r23=[r3],24
++ mov b6=r28
++ ;;
++ ld8.fill r25=[r2],16
++ ld8.fill r26=[r3],16
++ mov b7=r29
++ ;;
++ ld8.fill r27=[r2],16
++ ld8.fill r28=[r3],16
++ ;;
++ ld8.fill r29=[r2],16
++ ld8.fill r30=[r3],24
++ ;;
++ ld8.fill r31=[r2],PT(F9)-PT(R31)
++ adds r3=PT(F10)-PT(F6),r3
++ ;;
++ ldf.fill f9=[r2],PT(F6)-PT(F9)
++ ldf.fill f10=[r3],PT(F8)-PT(F10)
++ ;;
++ ldf.fill f6=[r2],PT(F7)-PT(F6)
++ ;;
++ ldf.fill f7=[r2],PT(F11)-PT(F7)
++ ldf.fill f8=[r3],32
++ ;;
++ srlz.d // ensure that inter. collection is off (VHPT is don't care, since text is pinned)
++ mov ar.ccv=r15
++ ;;
++ ldf.fill f11=[r2]
++#ifdef CONFIG_XEN
++ ;;
++ // r16-r31 all now hold bank1 values
++ movl r2=XSI_BANK1_R16
++ movl r3=XSI_BANK1_R16+8
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r16,16
++.mem.offset 8,0; st8.spill [r3]=r17,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r18,16
++.mem.offset 8,0; st8.spill [r3]=r19,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r20,16
++.mem.offset 8,0; st8.spill [r3]=r21,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r22,16
++.mem.offset 8,0; st8.spill [r3]=r23,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r24,16
++.mem.offset 8,0; st8.spill [r3]=r25,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r26,16
++.mem.offset 8,0; st8.spill [r3]=r27,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r28,16
++.mem.offset 8,0; st8.spill [r3]=r29,16
++ ;;
++.mem.offset 0,0; st8.spill [r2]=r30,16
++.mem.offset 8,0; st8.spill [r3]=r31,16
++ ;;
++ movl r2=XSI_BANKNUM;;
++ st4 [r2]=r0;
++#else
++ bsw.0 // switch back to bank 0 (no stop bit required beforehand...)
++#endif
++ ;;
++(pUStk) mov r18=IA64_KR(CURRENT)// M2 (12 cycle read latency)
++ adds r16=PT(CR_IPSR)+16,r12
++ adds r17=PT(CR_IIP)+16,r12
++
++(pKStk) mov r22=psr // M2 read PSR now that interrupts are disabled
++ nop.i 0
++ nop.i 0
++ ;;
++ ld8 r29=[r16],16 // load cr.ipsr
++ ld8 r28=[r17],16 // load cr.iip
++ ;;
++ ld8 r30=[r16],16 // load cr.ifs
++ ld8 r25=[r17],16 // load ar.unat
++ ;;
++ ld8 r26=[r16],16 // load ar.pfs
++ ld8 r27=[r17],16 // load ar.rsc
++ cmp.eq p9,p0=r0,r0 // set p9 to indicate that we should restore cr.ifs
++ ;;
++ ld8 r24=[r16],16 // load ar.rnat (may be garbage)
++ ld8 r23=[r17],16 // load ar.bspstore (may be garbage)
++ ;;
++ ld8 r31=[r16],16 // load predicates
++ ld8 r21=[r17],16 // load b0
++ ;;
++ ld8 r19=[r16],16 // load ar.rsc value for "loadrs"
++ ld8.fill r1=[r17],16 // load r1
++ ;;
++ ld8.fill r12=[r16],16
++ ld8.fill r13=[r17],16
++(pUStk) adds r18=IA64_TASK_THREAD_ON_USTACK_OFFSET,r18
++ ;;
++ ld8 r20=[r16],16 // ar.fpsr
++ ld8.fill r15=[r17],16
++ ;;
++ ld8.fill r14=[r16],16
++ ld8.fill r2=[r17]
++(pUStk) mov r17=1
++ ;;
++ ld8.fill r3=[r16]
++(pUStk) st1 [r18]=r17 // restore current->thread.on_ustack
++ shr.u r18=r19,16 // get byte size of existing "dirty" partition
++ ;;
++ mov r16=ar.bsp // get existing backing store pointer
++ addl r17=THIS_CPU(ia64_phys_stacked_size_p8),r0
++ ;;
++ ld4 r17=[r17] // r17 = cpu_data->phys_stacked_size_p8
++(pKStk) br.cond.dpnt skip_rbs_switch
++
++ /*
++ * Restore user backing store.
++ *
++ * NOTE: alloc, loadrs, and cover can't be predicated.
++ */
++(pNonSys) br.cond.dpnt dont_preserve_current_frame
++
++#ifdef CONFIG_XEN
++ XEN_HYPER_COVER;
++#else
++ cover // add current frame into dirty partition and set cr.ifs
++#endif
++ ;;
++ mov r19=ar.bsp // get new backing store pointer
++rbs_switch:
++ sub r16=r16,r18 // krbs = old bsp - size of dirty partition
++ cmp.ne p9,p0=r0,r0 // clear p9 to skip restore of cr.ifs
++ ;;
++ sub r19=r19,r16 // calculate total byte size of dirty partition
++ add r18=64,r18 // don't force in0-in7 into memory...
++ ;;
++ shl r19=r19,16 // shift size of dirty partition into loadrs position
++ ;;
++dont_preserve_current_frame:
++ /*
++ * To prevent leaking bits between the kernel and user-space,
++ * we must clear the stacked registers in the "invalid" partition here.
++ * Not pretty, but at least it's fast (3.34 registers/cycle on Itanium,
++ * 5 registers/cycle on McKinley).
++ */
++# define pRecurse p6
++# define pReturn p7
++#ifdef CONFIG_ITANIUM
++# define Nregs 10
++#else
++# define Nregs 14
++#endif
++ alloc loc0=ar.pfs,2,Nregs-2,2,0
++ shr.u loc1=r18,9 // RNaTslots <= floor(dirtySize / (64*8))
++ sub r17=r17,r18 // r17 = (physStackedSize + 8) - dirtySize
++ ;;
++ mov ar.rsc=r19 // load ar.rsc to be used for "loadrs"
++ shladd in0=loc1,3,r17
++ mov in1=0
++ ;;
++ TEXT_ALIGN(32)
++rse_clear_invalid:
++#ifdef CONFIG_ITANIUM
++ // cycle 0
++ { .mii
++ alloc loc0=ar.pfs,2,Nregs-2,2,0
++ cmp.lt pRecurse,p0=Nregs*8,in0 // if more than Nregs regs left to clear, (re)curse
++ add out0=-Nregs*8,in0
++}{ .mfb
++ add out1=1,in1 // increment recursion count
++ nop.f 0
++ nop.b 0 // can't do br.call here because of alloc (WAW on CFM)
++ ;;
++}{ .mfi // cycle 1
++ mov loc1=0
++ nop.f 0
++ mov loc2=0
++}{ .mib
++ mov loc3=0
++ mov loc4=0
++(pRecurse) br.call.sptk.many b0=rse_clear_invalid
++
++}{ .mfi // cycle 2
++ mov loc5=0
++ nop.f 0
++ cmp.ne pReturn,p0=r0,in1 // if recursion count != 0, we need to do a br.ret
++}{ .mib
++ mov loc6=0
++ mov loc7=0
++(pReturn) br.ret.sptk.many b0
++}
++#else /* !CONFIG_ITANIUM */
++ alloc loc0=ar.pfs,2,Nregs-2,2,0
++ cmp.lt pRecurse,p0=Nregs*8,in0 // if more than Nregs regs left to clear, (re)curse
++ add out0=-Nregs*8,in0
++ add out1=1,in1 // increment recursion count
++ mov loc1=0
++ mov loc2=0
++ ;;
++ mov loc3=0
++ mov loc4=0
++ mov loc5=0
++ mov loc6=0
++ mov loc7=0
++(pRecurse) br.call.dptk.few b0=rse_clear_invalid
++ ;;
++ mov loc8=0
++ mov loc9=0
++ cmp.ne pReturn,p0=r0,in1 // if recursion count != 0, we need to do a br.ret
++ mov loc10=0
++ mov loc11=0
++(pReturn) br.ret.dptk.many b0
++#endif /* !CONFIG_ITANIUM */
++# undef pRecurse
++# undef pReturn
++ ;;
++ alloc r17=ar.pfs,0,0,0,0 // drop current register frame
++ ;;
++ loadrs
++ ;;
++skip_rbs_switch:
++ mov ar.unat=r25 // M2
++(pKStk) extr.u r22=r22,21,1 // I0 extract current value of psr.pp from r22
++(pLvSys)mov r19=r0 // A clear r19 for leave_syscall, no-op otherwise
++ ;;
++(pUStk) mov ar.bspstore=r23 // M2
++(pKStk) dep r29=r22,r29,21,1 // I0 update ipsr.pp with psr.pp
++(pLvSys)mov r16=r0 // A clear r16 for leave_syscall, no-op otherwise
++ ;;
++#ifdef CONFIG_XEN
++ movl r25=XSI_IPSR
++ ;;
++ st8[r25]=r29,XSI_IFS-XSI_IPSR
++ ;;
++#else
++ mov cr.ipsr=r29 // M2
++#endif
++ mov ar.pfs=r26 // I0
++(pLvSys)mov r17=r0 // A clear r17 for leave_syscall, no-op otherwise
++
++#ifdef CONFIG_XEN
++(p9) st8 [r25]=r30
++ ;;
++ adds r25=XSI_IIP-XSI_IFS,r25
++ ;;
++#else
++(p9) mov cr.ifs=r30 // M2
++#endif
++ mov b0=r21 // I0
++(pLvSys)mov r18=r0 // A clear r18 for leave_syscall, no-op otherwise
++
++ mov ar.fpsr=r20 // M2
++#ifdef CONFIG_XEN
++ st8 [r25]=r28
++#else
++ mov cr.iip=r28 // M2
++#endif
++ nop 0
++ ;;
++(pUStk) mov ar.rnat=r24 // M2 must happen with RSE in lazy mode
++ nop 0
++(pLvSys)mov r2=r0
++
++ mov ar.rsc=r27 // M2
++ mov pr=r31,-1 // I0
++#ifdef CONFIG_XEN
++ ;;
++ XEN_HYPER_RFI;
++#else
++ rfi // B
++#endif
++
++ /*
++ * On entry:
++ * r20 = ¤t->thread_info->pre_count (if CONFIG_PREEMPT)
++ * r31 = current->thread_info->flags
++ * On exit:
++ * p6 = TRUE if work-pending-check needs to be redone
++ */
++.work_pending_syscall:
++ add r2=-8,r2
++ add r3=-8,r3
++ ;;
++ st8 [r2]=r8
++ st8 [r3]=r10
++.work_pending:
++ tbit.nz p6,p0=r31,TIF_SIGDELAYED // signal delayed from MCA/INIT/NMI/PMI context?
++(p6) br.cond.sptk.few .sigdelayed
++ ;;
++ tbit.z p6,p0=r31,TIF_NEED_RESCHED // current_thread_info()->need_resched==0?
++(p6) br.cond.sptk.few .notify
++#ifdef CONFIG_PREEMPT
++(pKStk) dep r21=-1,r0,PREEMPT_ACTIVE_BIT,1
++ ;;
++(pKStk) st4 [r20]=r21
++ ssm psr.i // enable interrupts
++#endif
++ br.call.spnt.many rp=schedule
++.ret9: cmp.eq p6,p0=r0,r0 // p6 <- 1
++#ifdef CONFIG_XEN
++ movl r2=XSI_PSR_I_ADDR
++ mov r20=1
++ ;;
++ ld8 r2=[r2]
++ ;;
++ st1 [r2]=r20
++#else
++ rsm psr.i // disable interrupts
++#endif
++ ;;
++#ifdef CONFIG_PREEMPT
++(pKStk) adds r20=TI_PRE_COUNT+IA64_TASK_SIZE,r13
++ ;;
++(pKStk) st4 [r20]=r0 // preempt_count() <- 0
++#endif
++(pLvSys)br.cond.sptk.few .work_pending_syscall_end
++ br.cond.sptk.many .work_processed_kernel // re-check
++
++.notify:
++(pUStk) br.call.spnt.many rp=notify_resume_user
++.ret10: cmp.ne p6,p0=r0,r0 // p6 <- 0
++(pLvSys)br.cond.sptk.few .work_pending_syscall_end
++ br.cond.sptk.many .work_processed_kernel // don't re-check
++
++// There is a delayed signal that was detected in MCA/INIT/NMI/PMI context where
++// it could not be delivered. Deliver it now. The signal might be for us and
++// may set TIF_SIGPENDING, so redrive ia64_leave_* after processing the delayed
++// signal.
++
++.sigdelayed:
++ br.call.sptk.many rp=do_sigdelayed
++ cmp.eq p6,p0=r0,r0 // p6 <- 1, always re-check
++(pLvSys)br.cond.sptk.few .work_pending_syscall_end
++ br.cond.sptk.many .work_processed_kernel // re-check
++
++.work_pending_syscall_end:
++ adds r2=PT(R8)+16,r12
++ adds r3=PT(R10)+16,r12
++ ;;
++ ld8 r8=[r2]
++ ld8 r10=[r3]
++ br.cond.sptk.many .work_processed_syscall // re-check
++
++#ifdef CONFIG_XEN
++END(xen_leave_kernel)
++#else
++END(ia64_leave_kernel)
++#endif
+diff -urNp linux-2.6/arch/ia64/xen/xenhpski.c new/arch/ia64/xen/xenhpski.c
+--- linux-2.6/arch/ia64/xen/xenhpski.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xenhpski.c 2006-05-09 12:32:40.000000000 +0200
+@@ -0,0 +1,19 @@
++
++extern unsigned long xen_get_cpuid(int);
++
++int
++running_on_sim(void)
++{
++ int i;
++ long cpuid[6];
++
++ for (i = 0; i < 5; ++i)
++ cpuid[i] = xen_get_cpuid(i);
++ if ((cpuid[0] & 0xff) != 'H') return 0;
++ if ((cpuid[3] & 0xff) != 0x4) return 0;
++ if (((cpuid[3] >> 8) & 0xff) != 0x0) return 0;
++ if (((cpuid[3] >> 16) & 0xff) != 0x0) return 0;
++ if (((cpuid[3] >> 24) & 0x7) != 0x7) return 0;
++ return 1;
++}
++
+diff -urNp linux-2.6/arch/ia64/xen/xenivt.S new/arch/ia64/xen/xenivt.S
+--- linux-2.6/arch/ia64/xen/xenivt.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xenivt.S 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,2171 @@
++/*
++ * arch/ia64/xen/ivt.S
++ *
++ * Copyright (C) 2005 Hewlett-Packard Co
++ * Dan Magenheimer <dan.magenheimer@hp.com>
++ */
++/*
++ * This file defines the interruption vector table used by the CPU.
++ * It does not include one entry per possible cause of interruption.
++ *
++ * The first 20 entries of the table contain 64 bundles each while the
++ * remaining 48 entries contain only 16 bundles each.
++ *
++ * The 64 bundles are used to allow inlining the whole handler for critical
++ * interruptions like TLB misses.
++ *
++ * For each entry, the comment is as follows:
++ *
++ * // 0x1c00 Entry 7 (size 64 bundles) Data Key Miss (12,51)
++ * entry offset ----/ / / / /
++ * entry number ---------/ / / /
++ * size of the entry -------------/ / /
++ * vector name -------------------------------------/ /
++ * interruptions triggering this vector ----------------------/
++ *
++ * The table is 32KB in size and must be aligned on 32KB boundary.
++ * (The CPU ignores the 15 lower bits of the address)
++ *
++ * Table is based upon EAS2.6 (Oct 1999)
++ */
++
++#include <linux/config.h>
++
++#include <asm/asmmacro.h>
++#include <asm/break.h>
++#include <asm/ia32.h>
++#include <asm/kregs.h>
++#include <asm/asm-offsets.h>
++#include <asm/pgtable.h>
++#include <asm/processor.h>
++#include <asm/ptrace.h>
++#include <asm/system.h>
++#include <asm/thread_info.h>
++#include <asm/unistd.h>
++#include <asm/errno.h>
++
++#ifdef CONFIG_XEN
++#define ia64_ivt xen_ivt
++#endif
++
++#if 1
++# define PSR_DEFAULT_BITS psr.ac
++#else
++# define PSR_DEFAULT_BITS 0
++#endif
++
++#if 0
++ /*
++ * This lets you track the last eight faults that occurred on the CPU. Make sure ar.k2 isn't
++ * needed for something else before enabling this...
++ */
++# define DBG_FAULT(i) mov r16=ar.k2;; shl r16=r16,8;; add r16=(i),r16;;mov ar.k2=r16
++#else
++# define DBG_FAULT(i)
++#endif
++
++#define MINSTATE_VIRT /* needed by minstate.h */
++#include "xenminstate.h"
++
++#define FAULT(n) \
++ mov r31=pr; \
++ mov r19=n;; /* prepare to save predicates */ \
++ br.sptk.many dispatch_to_fault_handler
++
++ .section .text.ivt,"ax"
++
++ .align 32768 // align on 32KB boundary
++ .global ia64_ivt
++ia64_ivt:
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x0000 Entry 0 (size 64 bundles) VHPT Translation (8,20,47)
++ENTRY(vhpt_miss)
++ DBG_FAULT(0)
++ /*
++ * The VHPT vector is invoked when the TLB entry for the virtual page table
++ * is missing. This happens only as a result of a previous
++ * (the "original") TLB miss, which may either be caused by an instruction
++ * fetch or a data access (or non-access).
++ *
++ * What we do here is normal TLB miss handing for the _original_ miss,
++ * followed by inserting the TLB entry for the virtual page table page
++ * that the VHPT walker was attempting to access. The latter gets
++ * inserted as long as page table entry above pte level have valid
++ * mappings for the faulting address. The TLB entry for the original
++ * miss gets inserted only if the pte entry indicates that the page is
++ * present.
++ *
++ * do_page_fault gets invoked in the following cases:
++ * - the faulting virtual address uses unimplemented address bits
++ * - the faulting virtual address has no valid page table mapping
++ */
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++#ifdef CONFIG_HUGETLB_PAGE
++ movl r18=PAGE_SHIFT
++ movl r25=XSI_ITIR
++ ;;
++ ld8 r25=[r25]
++#endif
++ ;;
++#else
++ mov r16=cr.ifa // get address that caused the TLB miss
++#ifdef CONFIG_HUGETLB_PAGE
++ movl r18=PAGE_SHIFT
++ mov r25=cr.itir
++#endif
++#endif
++ ;;
++#ifdef CONFIG_XEN
++ XEN_HYPER_RSM_PSR_DT;
++#else
++ rsm psr.dt // use physical addressing for data
++#endif
++ mov r31=pr // save the predicate registers
++ mov r19=IA64_KR(PT_BASE) // get page table base address
++ shl r21=r16,3 // shift bit 60 into sign bit
++ shr.u r17=r16,61 // get the region number into r17
++ ;;
++ shr.u r22=r21,3
++#ifdef CONFIG_HUGETLB_PAGE
++ extr.u r26=r25,2,6
++ ;;
++ cmp.ne p8,p0=r18,r26
++ sub r27=r26,r18
++ ;;
++(p8) dep r25=r18,r25,2,6
++(p8) shr r22=r22,r27
++#endif
++ ;;
++ cmp.eq p6,p7=5,r17 // is IFA pointing into to region 5?
++ shr.u r18=r22,PGDIR_SHIFT // get bottom portion of pgd index bit
++ ;;
++(p7) dep r17=r17,r19,(PAGE_SHIFT-3),3 // put region number bits in place
++
++ srlz.d
++ LOAD_PHYSICAL(p6, r19, swapper_pg_dir) // region 5 is rooted at swapper_pg_dir
++
++ .pred.rel "mutex", p6, p7
++(p6) shr.u r21=r21,PGDIR_SHIFT+PAGE_SHIFT
++(p7) shr.u r21=r21,PGDIR_SHIFT+PAGE_SHIFT-3
++ ;;
++(p6) dep r17=r18,r19,3,(PAGE_SHIFT-3) // r17=pgd_offset for region 5
++(p7) dep r17=r18,r17,3,(PAGE_SHIFT-6) // r17=pgd_offset for region[0-4]
++ cmp.eq p7,p6=0,r21 // unused address bits all zeroes?
++#ifdef CONFIG_PGTABLE_4
++ shr.u r28=r22,PUD_SHIFT // shift pud index into position
++#else
++ shr.u r18=r22,PMD_SHIFT // shift pmd index into position
++#endif
++ ;;
++ ld8 r17=[r17] // get *pgd (may be 0)
++ ;;
++(p7) cmp.eq p6,p7=r17,r0 // was pgd_present(*pgd) == NULL?
++#ifdef CONFIG_PGTABLE_4
++ dep r28=r28,r17,3,(PAGE_SHIFT-3) // r28=pud_offset(pgd,addr)
++ ;;
++ shr.u r18=r22,PMD_SHIFT // shift pmd index into position
++(p7) ld8 r29=[r28] // get *pud (may be 0)
++ ;;
++(p7) cmp.eq.or.andcm p6,p7=r29,r0 // was pud_present(*pud) == NULL?
++ dep r17=r18,r29,3,(PAGE_SHIFT-3) // r17=pmd_offset(pud,addr)
++#else
++ dep r17=r18,r17,3,(PAGE_SHIFT-3) // r17=pmd_offset(pgd,addr)
++#endif
++ ;;
++(p7) ld8 r20=[r17] // get *pmd (may be 0)
++ shr.u r19=r22,PAGE_SHIFT // shift pte index into position
++ ;;
++(p7) cmp.eq.or.andcm p6,p7=r20,r0 // was pmd_present(*pmd) == NULL?
++ dep r21=r19,r20,3,(PAGE_SHIFT-3) // r21=pte_offset(pmd,addr)
++ ;;
++(p7) ld8 r18=[r21] // read *pte
++#ifdef CONFIG_XEN
++ movl r19=XSI_ISR
++ ;;
++ ld8 r19=[r19]
++#else
++ mov r19=cr.isr // cr.isr bit 32 tells us if this is an insn miss
++#endif
++ ;;
++(p7) tbit.z p6,p7=r18,_PAGE_P_BIT // page present bit cleared?
++#ifdef CONFIG_XEN
++ movl r22=XSI_IHA
++ ;;
++ ld8 r22=[r22]
++#else
++ mov r22=cr.iha // get the VHPT address that caused the TLB miss
++#endif
++ ;; // avoid RAW on p7
++(p7) tbit.nz.unc p10,p11=r19,32 // is it an instruction TLB miss?
++ dep r23=0,r20,0,PAGE_SHIFT // clear low bits to get page address
++ ;;
++#ifdef CONFIG_XEN
++ mov r24=r8
++ mov r8=r18
++ ;;
++(p10) XEN_HYPER_ITC_I
++ ;;
++(p11) XEN_HYPER_ITC_D
++ ;;
++ mov r8=r24
++ ;;
++#else
++(p10) itc.i r18 // insert the instruction TLB entry
++(p11) itc.d r18 // insert the data TLB entry
++#endif
++(p6) br.cond.spnt.many page_fault // handle bad address/page not present (page fault)
++#ifdef CONFIG_XEN
++ movl r24=XSI_IFA
++ ;;
++ st8 [r24]=r22
++ ;;
++#else
++ mov cr.ifa=r22
++#endif
++
++#ifdef CONFIG_HUGETLB_PAGE
++(p8) mov cr.itir=r25 // change to default page-size for VHPT
++#endif
++
++ /*
++ * Now compute and insert the TLB entry for the virtual page table. We never
++ * execute in a page table page so there is no need to set the exception deferral
++ * bit.
++ */
++ adds r24=__DIRTY_BITS_NO_ED|_PAGE_PL_0|_PAGE_AR_RW,r23
++ ;;
++#ifdef CONFIG_XEN
++(p7) mov r25=r8
++(p7) mov r8=r24
++ ;;
++(p7) XEN_HYPER_ITC_D
++ ;;
++(p7) mov r8=r25
++ ;;
++#else
++(p7) itc.d r24
++#endif
++ ;;
++#ifdef CONFIG_SMP
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++
++ /*
++ * Re-check pagetable entry. If they changed, we may have received a ptc.g
++ * between reading the pagetable and the "itc". If so, flush the entry we
++ * inserted and retry. At this point, we have:
++ *
++ * r28 = equivalent of pud_offset(pgd, ifa)
++ * r17 = equivalent of pmd_offset(pud, ifa)
++ * r21 = equivalent of pte_offset(pmd, ifa)
++ *
++ * r29 = *pud
++ * r20 = *pmd
++ * r18 = *pte
++ */
++ ld8 r25=[r21] // read *pte again
++ ld8 r26=[r17] // read *pmd again
++#ifdef CONFIG_PGTABLE_4
++ ld8 r19=[r28] // read *pud again
++#endif
++ cmp.ne p6,p7=r0,r0
++ ;;
++ cmp.ne.or.andcm p6,p7=r26,r20 // did *pmd change
++#ifdef CONFIG_PGTABLE_4
++ cmp.ne.or.andcm p6,p7=r19,r29 // did *pud change
++#endif
++ mov r27=PAGE_SHIFT<<2
++ ;;
++(p6) ptc.l r22,r27 // purge PTE page translation
++(p7) cmp.ne.or.andcm p6,p7=r25,r18 // did *pte change
++ ;;
++(p6) ptc.l r16,r27 // purge translation
++#endif
++
++ mov pr=r31,-1 // restore predicate registers
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(vhpt_miss)
++
++ .org ia64_ivt+0x400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x0400 Entry 1 (size 64 bundles) ITLB (21)
++ENTRY(itlb_miss)
++ DBG_FAULT(1)
++ /*
++ * The ITLB handler accesses the PTE via the virtually mapped linear
++ * page table. If a nested TLB miss occurs, we switch into physical
++ * mode, walk the page table, and then re-execute the PTE read and
++ * go on normally after that.
++ */
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++#else
++ mov r16=cr.ifa // get virtual address
++#endif
++ mov r29=b0 // save b0
++ mov r31=pr // save predicates
++.itlb_fault:
++#ifdef CONFIG_XEN
++ movl r17=XSI_IHA
++ ;;
++ ld8 r17=[r17] // get virtual address of L3 PTE
++#else
++ mov r17=cr.iha // get virtual address of PTE
++#endif
++ movl r30=1f // load nested fault continuation point
++ ;;
++1: ld8 r18=[r17] // read *pte
++ ;;
++ mov b0=r29
++ tbit.z p6,p0=r18,_PAGE_P_BIT // page present bit cleared?
++(p6) br.cond.spnt page_fault
++ ;;
++#ifdef CONFIG_XEN
++ mov r19=r8
++ mov r8=r18
++ ;;
++ XEN_HYPER_ITC_I
++ ;;
++ mov r8=r19
++#else
++ itc.i r18
++#endif
++ ;;
++#ifdef CONFIG_SMP
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++
++ ld8 r19=[r17] // read *pte again and see if same
++ mov r20=PAGE_SHIFT<<2 // setup page size for purge
++ ;;
++ cmp.ne p7,p0=r18,r19
++ ;;
++(p7) ptc.l r16,r20
++#endif
++ mov pr=r31,-1
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(itlb_miss)
++
++ .org ia64_ivt+0x0800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x0800 Entry 2 (size 64 bundles) DTLB (9,48)
++ENTRY(dtlb_miss)
++ DBG_FAULT(2)
++ /*
++ * The DTLB handler accesses the PTE via the virtually mapped linear
++ * page table. If a nested TLB miss occurs, we switch into physical
++ * mode, walk the page table, and then re-execute the PTE read and
++ * go on normally after that.
++ */
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++#else
++ mov r16=cr.ifa // get virtual address
++#endif
++ mov r29=b0 // save b0
++ mov r31=pr // save predicates
++dtlb_fault:
++#ifdef CONFIG_XEN
++ movl r17=XSI_IHA
++ ;;
++ ld8 r17=[r17] // get virtual address of L3 PTE
++#else
++ mov r17=cr.iha // get virtual address of PTE
++#endif
++ movl r30=1f // load nested fault continuation point
++ ;;
++1: ld8 r18=[r17] // read *pte
++ ;;
++ mov b0=r29
++ tbit.z p6,p0=r18,_PAGE_P_BIT // page present bit cleared?
++(p6) br.cond.spnt page_fault
++ ;;
++#ifdef CONFIG_XEN
++ mov r19=r8
++ mov r8=r18
++ ;;
++ XEN_HYPER_ITC_D
++ ;;
++ mov r8=r19
++ ;;
++#else
++ itc.d r18
++#endif
++ ;;
++#ifdef CONFIG_SMP
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++
++ ld8 r19=[r17] // read *pte again and see if same
++ mov r20=PAGE_SHIFT<<2 // setup page size for purge
++ ;;
++ cmp.ne p7,p0=r18,r19
++ ;;
++(p7) ptc.l r16,r20
++#endif
++ mov pr=r31,-1
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(dtlb_miss)
++
++ .org ia64_ivt+0x0c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x0c00 Entry 3 (size 64 bundles) Alt ITLB (19)
++ENTRY(alt_itlb_miss)
++ DBG_FAULT(3)
++#ifdef CONFIG_XEN
++ movl r31=XSI_IPSR
++ ;;
++ ld8 r21=[r31],XSI_IFA-XSI_IPSR // get ipsr, point to ifa
++ movl r17=PAGE_KERNEL
++ ;;
++ ld8 r16=[r31] // get ifa
++#else
++ mov r16=cr.ifa // get address that caused the TLB miss
++ movl r17=PAGE_KERNEL
++ mov r21=cr.ipsr
++#endif
++ movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
++ mov r31=pr
++ ;;
++#ifdef CONFIG_DISABLE_VHPT
++ shr.u r22=r16,61 // get the region number into r21
++ ;;
++ cmp.gt p8,p0=6,r22 // user mode
++ ;;
++#ifndef CONFIG_XEN
++(p8) thash r17=r16
++ ;;
++(p8) mov cr.iha=r17
++#endif
++(p8) mov r29=b0 // save b0
++(p8) br.cond.dptk .itlb_fault
++#endif
++ extr.u r23=r21,IA64_PSR_CPL0_BIT,2 // extract psr.cpl
++ and r19=r19,r16 // clear ed, reserved bits, and PTE control bits
++ shr.u r18=r16,57 // move address bit 61 to bit 4
++ ;;
++ andcm r18=0x10,r18 // bit 4=~address-bit(61)
++ cmp.ne p8,p0=r0,r23 // psr.cpl != 0?
++ or r19=r17,r19 // insert PTE control bits into r19
++ ;;
++ or r19=r19,r18 // set bit 4 (uncached) if the access was to region 6
++(p8) br.cond.spnt page_fault
++ ;;
++#ifdef CONFIG_XEN
++ mov r18=r8
++ mov r8=r19
++ ;;
++ XEN_HYPER_ITC_I
++ ;;
++ mov r8=r18
++ ;;
++ mov pr=r31,-1
++ ;;
++ XEN_HYPER_RFI;
++#else
++ itc.i r19 // insert the TLB entry
++ mov pr=r31,-1
++ rfi
++#endif
++END(alt_itlb_miss)
++
++ .org ia64_ivt+0x1000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x1000 Entry 4 (size 64 bundles) Alt DTLB (7,46)
++ENTRY(alt_dtlb_miss)
++ DBG_FAULT(4)
++#ifdef CONFIG_XEN
++ movl r31=XSI_IPSR
++ ;;
++ ld8 r21=[r31],XSI_ISR-XSI_IPSR // get ipsr, point to isr
++ movl r17=PAGE_KERNEL
++ ;;
++ ld8 r20=[r31],XSI_IFA-XSI_ISR // get isr, point to ifa
++ movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
++ ;;
++ ld8 r16=[r31] // get ifa
++#else
++ mov r16=cr.ifa // get address that caused the TLB miss
++ movl r17=PAGE_KERNEL
++ mov r20=cr.isr
++ movl r19=(((1 << IA64_MAX_PHYS_BITS) - 1) & ~0xfff)
++ mov r21=cr.ipsr
++#endif
++ mov r31=pr
++ ;;
++#ifdef CONFIG_DISABLE_VHPT
++ shr.u r22=r16,61 // get the region number into r21
++ ;;
++ cmp.gt p8,p0=6,r22 // access to region 0-5
++ ;;
++#ifndef CONFIG_XEN
++(p8) thash r17=r16
++ ;;
++(p8) mov cr.iha=r17
++#endif
++(p8) mov r29=b0 // save b0
++(p8) br.cond.dptk dtlb_fault
++#endif
++ extr.u r23=r21,IA64_PSR_CPL0_BIT,2 // extract psr.cpl
++ and r22=IA64_ISR_CODE_MASK,r20 // get the isr.code field
++ tbit.nz p6,p7=r20,IA64_ISR_SP_BIT // is speculation bit on?
++ shr.u r18=r16,57 // move address bit 61 to bit 4
++ and r19=r19,r16 // clear ed, reserved bits, and PTE control bits
++ tbit.nz p9,p0=r20,IA64_ISR_NA_BIT // is non-access bit on?
++ ;;
++ andcm r18=0x10,r18 // bit 4=~address-bit(61)
++ cmp.ne p8,p0=r0,r23
++(p9) cmp.eq.or.andcm p6,p7=IA64_ISR_CODE_LFETCH,r22 // check isr.code field
++(p8) br.cond.spnt page_fault
++
++ dep r21=-1,r21,IA64_PSR_ED_BIT,1
++ or r19=r19,r17 // insert PTE control bits into r19
++ ;;
++ or r19=r19,r18 // set bit 4 (uncached) if the access was to region 6
++(p6) mov cr.ipsr=r21
++ ;;
++#ifdef CONFIG_XEN
++(p7) mov r18=r8
++(p7) mov r8=r19
++ ;;
++(p7) XEN_HYPER_ITC_D
++ ;;
++(p7) mov r8=r18
++ ;;
++ mov pr=r31,-1
++ ;;
++ XEN_HYPER_RFI;
++#else
++(p7) itc.d r19 // insert the TLB entry
++ mov pr=r31,-1
++ rfi
++#endif
++END(alt_dtlb_miss)
++
++ .org ia64_ivt+0x1400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x1400 Entry 5 (size 64 bundles) Data nested TLB (6,45)
++ENTRY(nested_dtlb_miss)
++ /*
++ * In the absence of kernel bugs, we get here when the virtually mapped linear
++ * page table is accessed non-speculatively (e.g., in the Dirty-bit, Instruction
++ * Access-bit, or Data Access-bit faults). If the DTLB entry for the virtual page
++ * table is missing, a nested TLB miss fault is triggered and control is
++ * transferred to this point. When this happens, we lookup the pte for the
++ * faulting address by walking the page table in physical mode and return to the
++ * continuation point passed in register r30 (or call page_fault if the address is
++ * not mapped).
++ *
++ * Input: r16: faulting address
++ * r29: saved b0
++ * r30: continuation address
++ * r31: saved pr
++ *
++ * Output: r17: physical address of PTE of faulting address
++ * r29: saved b0
++ * r30: continuation address
++ * r31: saved pr
++ *
++ * Clobbered: b0, r18, r19, r21, r22, psr.dt (cleared)
++ */
++#ifdef CONFIG_XEN
++ XEN_HYPER_RSM_PSR_DT;
++#else
++ rsm psr.dt // switch to using physical data addressing
++#endif
++ mov r19=IA64_KR(PT_BASE) // get the page table base address
++ shl r21=r16,3 // shift bit 60 into sign bit
++#ifdef CONFIG_XEN
++ movl r18=XSI_ITIR
++ ;;
++ ld8 r18=[r18]
++#else
++ mov r18=cr.itir
++#endif
++ ;;
++ shr.u r17=r16,61 // get the region number into r17
++ extr.u r18=r18,2,6 // get the faulting page size
++ ;;
++ cmp.eq p6,p7=5,r17 // is faulting address in region 5?
++ add r22=-PAGE_SHIFT,r18 // adjustment for hugetlb address
++ add r18=PGDIR_SHIFT-PAGE_SHIFT,r18
++ ;;
++ shr.u r22=r16,r22
++ shr.u r18=r16,r18
++(p7) dep r17=r17,r19,(PAGE_SHIFT-3),3 // put region number bits in place
++
++ srlz.d
++ LOAD_PHYSICAL(p6, r19, swapper_pg_dir) // region 5 is rooted at swapper_pg_dir
++
++ .pred.rel "mutex", p6, p7
++(p6) shr.u r21=r21,PGDIR_SHIFT+PAGE_SHIFT
++(p7) shr.u r21=r21,PGDIR_SHIFT+PAGE_SHIFT-3
++ ;;
++(p6) dep r17=r18,r19,3,(PAGE_SHIFT-3) // r17=pgd_offset for region 5
++(p7) dep r17=r18,r17,3,(PAGE_SHIFT-6) // r17=pgd_offset for region[0-4]
++ cmp.eq p7,p6=0,r21 // unused address bits all zeroes?
++#ifdef CONFIG_PGTABLE_4
++ shr.u r18=r22,PUD_SHIFT // shift pud index into position
++#else
++ shr.u r18=r22,PMD_SHIFT // shift pmd index into position
++#endif
++ ;;
++ ld8 r17=[r17] // get *pgd (may be 0)
++ ;;
++(p7) cmp.eq p6,p7=r17,r0 // was pgd_present(*pgd) == NULL?
++ dep r17=r18,r17,3,(PAGE_SHIFT-3) // r17=p[u|m]d_offset(pgd,addr)
++ ;;
++#ifdef CONFIG_PGTABLE_4
++(p7) ld8 r17=[r17] // get *pud (may be 0)
++ shr.u r18=r22,PMD_SHIFT // shift pmd index into position
++ ;;
++(p7) cmp.eq.or.andcm p6,p7=r17,r0 // was pud_present(*pud) == NULL?
++ dep r17=r18,r17,3,(PAGE_SHIFT-3) // r17=pmd_offset(pud,addr)
++ ;;
++#endif
++(p7) ld8 r17=[r17] // get *pmd (may be 0)
++ shr.u r19=r22,PAGE_SHIFT // shift pte index into position
++ ;;
++(p7) cmp.eq.or.andcm p6,p7=r17,r0 // was pmd_present(*pmd) == NULL?
++ dep r17=r19,r17,3,(PAGE_SHIFT-3) // r17=pte_offset(pmd,addr);
++(p6) br.cond.spnt page_fault
++ mov b0=r30
++ br.sptk.many b0 // return to continuation point
++END(nested_dtlb_miss)
++
++ .org ia64_ivt+0x1800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x1800 Entry 6 (size 64 bundles) Instruction Key Miss (24)
++ENTRY(ikey_miss)
++ DBG_FAULT(6)
++ FAULT(6)
++END(ikey_miss)
++
++ //-----------------------------------------------------------------------------------
++ // call do_page_fault (predicates are in r31, psr.dt may be off, r16 is faulting address)
++ENTRY(page_fault)
++#ifdef CONFIG_XEN
++ XEN_HYPER_SSM_PSR_DT
++#else
++ ssm psr.dt
++ ;;
++ srlz.i
++#endif
++ ;;
++ SAVE_MIN_WITH_COVER
++ alloc r15=ar.pfs,0,0,3,0
++#ifdef CONFIG_XEN
++ movl r3=XSI_ISR
++ ;;
++ ld8 out1=[r3],XSI_IFA-XSI_ISR // get vcr.isr, point to ifa
++ ;;
++ ld8 out0=[r3] // get vcr.ifa
++ mov r14=1
++ ;;
++ add r3=XSI_PSR_IC-XSI_IFA, r3 // point to vpsr.ic
++ ;;
++ st4 [r3]=r14 // vpsr.ic = 1
++ adds r3=8,r2 // set up second base pointer
++ ;;
++#else
++ mov out0=cr.ifa
++ mov out1=cr.isr
++ adds r3=8,r2 // set up second base pointer
++ ;;
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collectin is on
++ ;;
++#endif
++#ifdef CONFIG_XEN
++ br.cond.sptk.many xen_page_fault
++ ;;
++done_xen_page_fault:
++#endif
++(p15) ssm psr.i // restore psr.i
++ movl r14=ia64_leave_kernel
++ ;;
++ SAVE_REST
++ mov rp=r14
++ ;;
++ adds out2=16,r12 // out2 = pointer to pt_regs
++ br.call.sptk.many b6=ia64_do_page_fault // ignore return address
++END(page_fault)
++
++ .org ia64_ivt+0x1c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x1c00 Entry 7 (size 64 bundles) Data Key Miss (12,51)
++ENTRY(dkey_miss)
++ DBG_FAULT(7)
++ FAULT(7)
++#ifdef CONFIG_XEN
++ // Leaving this code inline above results in an IVT section overflow
++ // There is no particular reason for this code to be here...
++xen_page_fault:
++(p15) movl r3=XSI_PSR_I_ADDR
++ ;;
++(p15) ld8 r3=[r3]
++ ;;
++(p15) st1 [r3]=r0,XSI_PEND-XSI_PSR_I_ADDR // if (p15) vpsr.i = 1
++ mov r14=r0
++ ;;
++(p15) ld4 r14=[r3] // if (pending_interrupts)
++ adds r3=8,r2 // re-set up second base pointer
++ ;;
++(p15) cmp.ne p15,p0=r14,r0
++ ;;
++ br.cond.sptk.many done_xen_page_fault
++ ;;
++#endif
++END(dkey_miss)
++
++ .org ia64_ivt+0x2000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x2000 Entry 8 (size 64 bundles) Dirty-bit (54)
++ENTRY(dirty_bit)
++ DBG_FAULT(8)
++ /*
++ * What we do here is to simply turn on the dirty bit in the PTE. We need to
++ * update both the page-table and the TLB entry. To efficiently access the PTE,
++ * we address it through the virtual page table. Most likely, the TLB entry for
++ * the relevant virtual page table page is still present in the TLB so we can
++ * normally do this without additional TLB misses. In case the necessary virtual
++ * page table TLB entry isn't present, we take a nested TLB miss hit where we look
++ * up the physical address of the L3 PTE and then continue at label 1 below.
++ */
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++ ;;
++#else
++ mov r16=cr.ifa // get the address that caused the fault
++#endif
++ movl r30=1f // load continuation point in case of nested fault
++ ;;
++#ifdef CONFIG_XEN
++ mov r18=r8;
++ mov r8=r16;
++ XEN_HYPER_THASH;;
++ mov r17=r8;
++ mov r8=r18;;
++#else
++ thash r17=r16 // compute virtual address of L3 PTE
++#endif
++ mov r29=b0 // save b0 in case of nested fault
++ mov r31=pr // save pr
++#ifdef CONFIG_SMP
++ mov r28=ar.ccv // save ar.ccv
++ ;;
++1: ld8 r18=[r17]
++ ;; // avoid RAW on r18
++ mov ar.ccv=r18 // set compare value for cmpxchg
++ or r25=_PAGE_D|_PAGE_A,r18 // set the dirty and accessed bits
++ tbit.z p7,p6 = r18,_PAGE_P_BIT // Check present bit
++ ;;
++(p6) cmpxchg8.acq r26=[r17],r25,ar.ccv // Only update if page is present
++ mov r24=PAGE_SHIFT<<2
++ ;;
++(p6) cmp.eq p6,p7=r26,r18 // Only compare if page is present
++ ;;
++#ifdef CONFIG_XEN
++(p6) mov r18=r8
++(p6) mov r8=r25
++ ;;
++(p6) XEN_HYPER_ITC_D
++ ;;
++(p6) mov r8=r18
++#else
++(p6) itc.d r25 // install updated PTE
++#endif
++ ;;
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++
++ ld8 r18=[r17] // read PTE again
++ ;;
++ cmp.eq p6,p7=r18,r25 // is it same as the newly installed
++ ;;
++(p7) ptc.l r16,r24
++ mov b0=r29 // restore b0
++ mov ar.ccv=r28
++#else
++ ;;
++1: ld8 r18=[r17]
++ ;; // avoid RAW on r18
++ or r18=_PAGE_D|_PAGE_A,r18 // set the dirty and accessed bits
++ mov b0=r29 // restore b0
++ ;;
++ st8 [r17]=r18 // store back updated PTE
++ itc.d r18 // install updated PTE
++#endif
++ mov pr=r31,-1 // restore pr
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(dirty_bit)
++
++ .org ia64_ivt+0x2400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x2400 Entry 9 (size 64 bundles) Instruction Access-bit (27)
++ENTRY(iaccess_bit)
++ DBG_FAULT(9)
++ // Like Entry 8, except for instruction access
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++ ;;
++#else
++ mov r16=cr.ifa // get the address that caused the fault
++#endif
++ movl r30=1f // load continuation point in case of nested fault
++ mov r31=pr // save predicates
++#ifdef CONFIG_ITANIUM
++ /*
++ * Erratum 10 (IFA may contain incorrect address) has "NoFix" status.
++ */
++ mov r17=cr.ipsr
++ ;;
++ mov r18=cr.iip
++ tbit.z p6,p0=r17,IA64_PSR_IS_BIT // IA64 instruction set?
++ ;;
++(p6) mov r16=r18 // if so, use cr.iip instead of cr.ifa
++#endif /* CONFIG_ITANIUM */
++ ;;
++#ifdef CONFIG_XEN
++ mov r18=r8;
++ mov r8=r16;
++ XEN_HYPER_THASH;;
++ mov r17=r8;
++ mov r8=r18;;
++#else
++ thash r17=r16 // compute virtual address of L3 PTE
++#endif
++ mov r29=b0 // save b0 in case of nested fault)
++#ifdef CONFIG_SMP
++ mov r28=ar.ccv // save ar.ccv
++ ;;
++1: ld8 r18=[r17]
++ ;;
++ mov ar.ccv=r18 // set compare value for cmpxchg
++ or r25=_PAGE_A,r18 // set the accessed bit
++ tbit.z p7,p6 = r18,_PAGE_P_BIT // Check present bit
++ ;;
++(p6) cmpxchg8.acq r26=[r17],r25,ar.ccv // Only if page present
++ mov r24=PAGE_SHIFT<<2
++ ;;
++(p6) cmp.eq p6,p7=r26,r18 // Only if page present
++ ;;
++#ifdef CONFIG_XEN
++ mov r26=r8
++ mov r8=r25
++ ;;
++(p6) XEN_HYPER_ITC_I
++ ;;
++ mov r8=r26
++ ;;
++#else
++(p6) itc.i r25 // install updated PTE
++#endif
++ ;;
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++
++ ld8 r18=[r17] // read PTE again
++ ;;
++ cmp.eq p6,p7=r18,r25 // is it same as the newly installed
++ ;;
++(p7) ptc.l r16,r24
++ mov b0=r29 // restore b0
++ mov ar.ccv=r28
++#else /* !CONFIG_SMP */
++ ;;
++1: ld8 r18=[r17]
++ ;;
++ or r18=_PAGE_A,r18 // set the accessed bit
++ mov b0=r29 // restore b0
++ ;;
++ st8 [r17]=r18 // store back updated PTE
++ itc.i r18 // install updated PTE
++#endif /* !CONFIG_SMP */
++ mov pr=r31,-1
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(iaccess_bit)
++
++ .org ia64_ivt+0x2800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x2800 Entry 10 (size 64 bundles) Data Access-bit (15,55)
++ENTRY(daccess_bit)
++ DBG_FAULT(10)
++ // Like Entry 8, except for data access
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++ ;;
++#else
++ mov r16=cr.ifa // get the address that caused the fault
++#endif
++ movl r30=1f // load continuation point in case of nested fault
++ ;;
++#ifdef CONFIG_XEN
++ mov r18=r8
++ mov r8=r16
++ XEN_HYPER_THASH
++ ;;
++ mov r17=r8
++ mov r8=r18
++ ;;
++#else
++ thash r17=r16 // compute virtual address of L3 PTE
++#endif
++ mov r31=pr
++ mov r29=b0 // save b0 in case of nested fault)
++#ifdef CONFIG_SMP
++ mov r28=ar.ccv // save ar.ccv
++ ;;
++1: ld8 r18=[r17]
++ ;; // avoid RAW on r18
++ mov ar.ccv=r18 // set compare value for cmpxchg
++ or r25=_PAGE_A,r18 // set the dirty bit
++ tbit.z p7,p6 = r18,_PAGE_P_BIT // Check present bit
++ ;;
++(p6) cmpxchg8.acq r26=[r17],r25,ar.ccv // Only if page is present
++ mov r24=PAGE_SHIFT<<2
++ ;;
++(p6) cmp.eq p6,p7=r26,r18 // Only if page is present
++ ;;
++#ifdef CONFIG_XEN
++ mov r26=r8
++ mov r8=r25
++ ;;
++(p6) XEN_HYPER_ITC_D
++ ;;
++ mov r8=r26
++ ;;
++#else
++(p6) itc.d r25 // install updated PTE
++#endif
++ /*
++ * Tell the assemblers dependency-violation checker that the above "itc" instructions
++ * cannot possibly affect the following loads:
++ */
++ dv_serialize_data
++ ;;
++ ld8 r18=[r17] // read PTE again
++ ;;
++ cmp.eq p6,p7=r18,r25 // is it same as the newly installed
++ ;;
++(p7) ptc.l r16,r24
++ mov ar.ccv=r28
++#else
++ ;;
++1: ld8 r18=[r17]
++ ;; // avoid RAW on r18
++ or r18=_PAGE_A,r18 // set the accessed bit
++ ;;
++ st8 [r17]=r18 // store back updated PTE
++ itc.d r18 // install updated PTE
++#endif
++ mov b0=r29 // restore b0
++ mov pr=r31,-1
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI
++ dv_serialize_data
++#else
++ rfi
++#endif
++END(daccess_bit)
++
++ .org ia64_ivt+0x2c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x2c00 Entry 11 (size 64 bundles) Break instruction (33)
++ENTRY(break_fault)
++ /*
++ * The streamlined system call entry/exit paths only save/restore the initial part
++ * of pt_regs. This implies that the callers of system-calls must adhere to the
++ * normal procedure calling conventions.
++ *
++ * Registers to be saved & restored:
++ * CR registers: cr.ipsr, cr.iip, cr.ifs
++ * AR registers: ar.unat, ar.pfs, ar.rsc, ar.rnat, ar.bspstore, ar.fpsr
++ * others: pr, b0, b6, loadrs, r1, r11, r12, r13, r15
++ * Registers to be restored only:
++ * r8-r11: output value from the system call.
++ *
++ * During system call exit, scratch registers (including r15) are modified/cleared
++ * to prevent leaking bits from kernel to user level.
++ */
++ DBG_FAULT(11)
++ mov.m r16=IA64_KR(CURRENT) // M2 r16 <- current task (12 cyc)
++#ifdef CONFIG_XEN
++ movl r22=XSI_IPSR
++ ;;
++ ld8 r29=[r22],XSI_IIM-XSI_IPSR // get ipsr, point to iip
++#else
++ mov r29=cr.ipsr // M2 (12 cyc)
++#endif
++ mov r31=pr // I0 (2 cyc)
++
++#ifdef CONFIG_XEN
++ ;;
++ ld8 r17=[r22],XSI_IIP-XSI_IIM
++#else
++ mov r17=cr.iim // M2 (2 cyc)
++#endif
++ mov.m r27=ar.rsc // M2 (12 cyc)
++ mov r18=__IA64_BREAK_SYSCALL // A
++
++ mov.m ar.rsc=0 // M2
++ mov.m r21=ar.fpsr // M2 (12 cyc)
++ mov r19=b6 // I0 (2 cyc)
++ ;;
++ mov.m r23=ar.bspstore // M2 (12 cyc)
++ mov.m r24=ar.rnat // M2 (5 cyc)
++ mov.i r26=ar.pfs // I0 (2 cyc)
++
++ invala // M0|1
++ nop.m 0 // M
++ mov r20=r1 // A save r1
++
++ nop.m 0
++ movl r30=sys_call_table // X
++
++#ifdef CONFIG_XEN
++ ld8 r28=[r22]
++#else
++ mov r28=cr.iip // M2 (2 cyc)
++#endif
++ cmp.eq p0,p7=r18,r17 // I0 is this a system call?
++(p7) br.cond.spnt non_syscall // B no ->
++ //
++ // From this point on, we are definitely on the syscall-path
++ // and we can use (non-banked) scratch registers.
++ //
++///////////////////////////////////////////////////////////////////////
++ mov r1=r16 // A move task-pointer to "addl"-addressable reg
++ mov r2=r16 // A setup r2 for ia64_syscall_setup
++ add r9=TI_FLAGS+IA64_TASK_SIZE,r16 // A r9 = ¤t_thread_info()->flags
++
++ adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16
++ adds r15=-1024,r15 // A subtract 1024 from syscall number
++ mov r3=NR_syscalls - 1
++ ;;
++ ld1.bias r17=[r16] // M0|1 r17 = current->thread.on_ustack flag
++ ld4 r9=[r9] // M0|1 r9 = current_thread_info()->flags
++ extr.u r8=r29,41,2 // I0 extract ei field from cr.ipsr
++
++ shladd r30=r15,3,r30 // A r30 = sys_call_table + 8*(syscall-1024)
++ addl r22=IA64_RBS_OFFSET,r1 // A compute base of RBS
++ cmp.leu p6,p7=r15,r3 // A syscall number in range?
++ ;;
++
++ lfetch.fault.excl.nt1 [r22] // M0|1 prefetch RBS
++(p6) ld8 r30=[r30] // M0|1 load address of syscall entry point
++ tnat.nz.or p7,p0=r15 // I0 is syscall nr a NaT?
++
++ mov.m ar.bspstore=r22 // M2 switch to kernel RBS
++ cmp.eq p8,p9=2,r8 // A isr.ei==2?
++ ;;
++
++(p8) mov r8=0 // A clear ei to 0
++(p7) movl r30=sys_ni_syscall // X
++
++(p8) adds r28=16,r28 // A switch cr.iip to next bundle
++(p9) adds r8=1,r8 // A increment ei to next slot
++ nop.i 0
++ ;;
++
++ mov.m r25=ar.unat // M2 (5 cyc)
++ dep r29=r8,r29,41,2 // I0 insert new ei into cr.ipsr
++ adds r15=1024,r15 // A restore original syscall number
++ //
++ // If any of the above loads miss in L1D, we'll stall here until
++ // the data arrives.
++ //
++///////////////////////////////////////////////////////////////////////
++ st1 [r16]=r0 // M2|3 clear current->thread.on_ustack flag
++ mov b6=r30 // I0 setup syscall handler branch reg early
++ cmp.eq pKStk,pUStk=r0,r17 // A were we on kernel stacks already?
++
++ and r9=_TIF_SYSCALL_TRACEAUDIT,r9 // A mask trace or audit
++ mov r18=ar.bsp // M2 (12 cyc)
++(pKStk) br.cond.spnt .break_fixup // B we're already in kernel-mode -- fix up RBS
++ ;;
++.back_from_break_fixup:
++(pUStk) addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1 // A compute base of memory stack
++ cmp.eq p14,p0=r9,r0 // A are syscalls being traced/audited?
++ br.call.sptk.many b7=ia64_syscall_setup // B
++1:
++ mov ar.rsc=0x3 // M2 set eager mode, pl 0, LE, loadrs=0
++ nop 0
++#ifdef CONFIG_XEN
++ mov r2=b0; br.call.sptk b0=xen_bsw1;; mov b0=r2;;
++#else
++ bsw.1 // B (6 cyc) regs are saved, switch to bank 1
++#endif
++ ;;
++
++#ifdef CONFIG_XEN
++ movl r16=XSI_PSR_IC
++ mov r3=1
++ ;;
++ st4 [r16]=r3,XSI_PSR_I_ADDR-XSI_PSR_IC // vpsr.ic = 1
++#else
++ ssm psr.ic | PSR_DEFAULT_BITS // M2 now it's safe to re-enable intr.-collection
++#endif
++ movl r3=ia64_ret_from_syscall // X
++ ;;
++
++ srlz.i // M0 ensure interruption collection is on
++ mov rp=r3 // I0 set the real return addr
++(p10) br.cond.spnt.many ia64_ret_from_syscall // B return if bad call-frame or r15 is a NaT
++
++#ifdef CONFIG_XEN
++(p15) ld8 r16=[r16] // vpsr.i
++ ;;
++(p15) st1 [r16]=r0,XSI_PEND-XSI_PSR_I_ADDR // if (p15) vpsr.i = 1
++ mov r2=r0
++ ;;
++(p15) ld4 r2=[r16] // if (pending_interrupts)
++ ;;
++ cmp.ne p6,p0=r2,r0
++ ;;
++(p6) ssm psr.i // do a real ssm psr.i
++#else
++(p15) ssm psr.i // M2 restore psr.i
++#endif
++(p14) br.call.sptk.many b6=b6 // B invoke syscall-handker (ignore return addr)
++ br.cond.spnt.many ia64_trace_syscall // B do syscall-tracing thingamagic
++ // NOT REACHED
++///////////////////////////////////////////////////////////////////////
++ // On entry, we optimistically assumed that we're coming from user-space.
++ // For the rare cases where a system-call is done from within the kernel,
++ // we fix things up at this point:
++.break_fixup:
++ add r1=-IA64_PT_REGS_SIZE,sp // A allocate space for pt_regs structure
++ mov ar.rnat=r24 // M2 restore kernel's AR.RNAT
++ ;;
++ mov ar.bspstore=r23 // M2 restore kernel's AR.BSPSTORE
++ br.cond.sptk .back_from_break_fixup
++END(break_fault)
++
++ .org ia64_ivt+0x3000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x3000 Entry 12 (size 64 bundles) External Interrupt (4)
++ENTRY(interrupt)
++ DBG_FAULT(12)
++ mov r31=pr // prepare to save predicates
++ ;;
++ SAVE_MIN_WITH_COVER // uses r31; defines r2 and r3
++#ifdef CONFIG_XEN
++ movl r3=XSI_PSR_IC
++ mov r14=1
++ ;;
++ st4 [r3]=r14
++#else
++ ssm psr.ic | PSR_DEFAULT_BITS
++#endif
++ ;;
++ adds r3=8,r2 // set up second base pointer for SAVE_REST
++ srlz.i // ensure everybody knows psr.ic is back on
++ ;;
++ SAVE_REST
++ ;;
++ alloc r14=ar.pfs,0,0,2,0 // must be first in an insn group
++#ifdef CONFIG_XEN
++ ;;
++ br.call.sptk.many rp=xen_get_ivr
++ ;;
++ mov out0=r8 // pass cr.ivr as first arg
++#else
++ mov out0=cr.ivr // pass cr.ivr as first arg
++#endif
++ add out1=16,sp // pass pointer to pt_regs as second arg
++ ;;
++ srlz.d // make sure we see the effect of cr.ivr
++ movl r14=ia64_leave_kernel
++ ;;
++ mov rp=r14
++ br.call.sptk.many b6=ia64_handle_irq
++END(interrupt)
++
++ .org ia64_ivt+0x3400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x3400 Entry 13 (size 64 bundles) Reserved
++ DBG_FAULT(13)
++ FAULT(13)
++
++ .org ia64_ivt+0x3800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x3800 Entry 14 (size 64 bundles) Reserved
++ DBG_FAULT(14)
++ FAULT(14)
++
++ /*
++ * There is no particular reason for this code to be here, other than that
++ * there happens to be space here that would go unused otherwise. If this
++ * fault ever gets "unreserved", simply moved the following code to a more
++ * suitable spot...
++ *
++ * ia64_syscall_setup() is a separate subroutine so that it can
++ * allocate stacked registers so it can safely demine any
++ * potential NaT values from the input registers.
++ *
++ * On entry:
++ * - executing on bank 0 or bank 1 register set (doesn't matter)
++ * - r1: stack pointer
++ * - r2: current task pointer
++ * - r3: preserved
++ * - r11: original contents (saved ar.pfs to be saved)
++ * - r12: original contents (sp to be saved)
++ * - r13: original contents (tp to be saved)
++ * - r15: original contents (syscall # to be saved)
++ * - r18: saved bsp (after switching to kernel stack)
++ * - r19: saved b6
++ * - r20: saved r1 (gp)
++ * - r21: saved ar.fpsr
++ * - r22: kernel's register backing store base (krbs_base)
++ * - r23: saved ar.bspstore
++ * - r24: saved ar.rnat
++ * - r25: saved ar.unat
++ * - r26: saved ar.pfs
++ * - r27: saved ar.rsc
++ * - r28: saved cr.iip
++ * - r29: saved cr.ipsr
++ * - r31: saved pr
++ * - b0: original contents (to be saved)
++ * On exit:
++ * - p10: TRUE if syscall is invoked with more than 8 out
++ * registers or r15's Nat is true
++ * - r1: kernel's gp
++ * - r3: preserved (same as on entry)
++ * - r8: -EINVAL if p10 is true
++ * - r12: points to kernel stack
++ * - r13: points to current task
++ * - r14: preserved (same as on entry)
++ * - p13: preserved
++ * - p15: TRUE if interrupts need to be re-enabled
++ * - ar.fpsr: set to kernel settings
++ * - b6: preserved (same as on entry)
++ */
++#ifndef CONFIG_XEN
++GLOBAL_ENTRY(ia64_syscall_setup)
++#if PT(B6) != 0
++# error This code assumes that b6 is the first field in pt_regs.
++#endif
++ st8 [r1]=r19 // save b6
++ add r16=PT(CR_IPSR),r1 // initialize first base pointer
++ add r17=PT(R11),r1 // initialize second base pointer
++ ;;
++ alloc r19=ar.pfs,8,0,0,0 // ensure in0-in7 are writable
++ st8 [r16]=r29,PT(AR_PFS)-PT(CR_IPSR) // save cr.ipsr
++ tnat.nz p8,p0=in0
++
++ st8.spill [r17]=r11,PT(CR_IIP)-PT(R11) // save r11
++ tnat.nz p9,p0=in1
++(pKStk) mov r18=r0 // make sure r18 isn't NaT
++ ;;
++
++ st8 [r16]=r26,PT(CR_IFS)-PT(AR_PFS) // save ar.pfs
++ st8 [r17]=r28,PT(AR_UNAT)-PT(CR_IIP) // save cr.iip
++ mov r28=b0 // save b0 (2 cyc)
++ ;;
++
++ st8 [r17]=r25,PT(AR_RSC)-PT(AR_UNAT) // save ar.unat
++ dep r19=0,r19,38,26 // clear all bits but 0..37 [I0]
++(p8) mov in0=-1
++ ;;
++
++ st8 [r16]=r19,PT(AR_RNAT)-PT(CR_IFS) // store ar.pfs.pfm in cr.ifs
++ extr.u r11=r19,7,7 // I0 // get sol of ar.pfs
++ and r8=0x7f,r19 // A // get sof of ar.pfs
++
++ st8 [r17]=r27,PT(AR_BSPSTORE)-PT(AR_RSC)// save ar.rsc
++ tbit.nz p15,p0=r29,IA64_PSR_I_BIT // I0
++(p9) mov in1=-1
++ ;;
++
++(pUStk) sub r18=r18,r22 // r18=RSE.ndirty*8
++ tnat.nz p10,p0=in2
++ add r11=8,r11
++ ;;
++(pKStk) adds r16=PT(PR)-PT(AR_RNAT),r16 // skip over ar_rnat field
++(pKStk) adds r17=PT(B0)-PT(AR_BSPSTORE),r17 // skip over ar_bspstore field
++ tnat.nz p11,p0=in3
++ ;;
++(p10) mov in2=-1
++ tnat.nz p12,p0=in4 // [I0]
++(p11) mov in3=-1
++ ;;
++(pUStk) st8 [r16]=r24,PT(PR)-PT(AR_RNAT) // save ar.rnat
++(pUStk) st8 [r17]=r23,PT(B0)-PT(AR_BSPSTORE) // save ar.bspstore
++ shl r18=r18,16 // compute ar.rsc to be used for "loadrs"
++ ;;
++ st8 [r16]=r31,PT(LOADRS)-PT(PR) // save predicates
++ st8 [r17]=r28,PT(R1)-PT(B0) // save b0
++ tnat.nz p13,p0=in5 // [I0]
++ ;;
++ st8 [r16]=r18,PT(R12)-PT(LOADRS) // save ar.rsc value for "loadrs"
++ st8.spill [r17]=r20,PT(R13)-PT(R1) // save original r1
++(p12) mov in4=-1
++ ;;
++
++.mem.offset 0,0; st8.spill [r16]=r12,PT(AR_FPSR)-PT(R12) // save r12
++.mem.offset 8,0; st8.spill [r17]=r13,PT(R15)-PT(R13) // save r13
++(p13) mov in5=-1
++ ;;
++ st8 [r16]=r21,PT(R8)-PT(AR_FPSR) // save ar.fpsr
++ tnat.nz p13,p0=in6
++ cmp.lt p10,p9=r11,r8 // frame size can't be more than local+8
++ ;;
++ mov r8=1
++(p9) tnat.nz p10,p0=r15
++ adds r12=-16,r1 // switch to kernel memory stack (with 16 bytes of scratch)
++
++ st8.spill [r17]=r15 // save r15
++ tnat.nz p8,p0=in7
++ nop.i 0
++
++ mov r13=r2 // establish `current'
++ movl r1=__gp // establish kernel global pointer
++ ;;
++ st8 [r16]=r8 // ensure pt_regs.r8 != 0 (see handle_syscall_error)
++(p13) mov in6=-1
++(p8) mov in7=-1
++
++ cmp.eq pSys,pNonSys=r0,r0 // set pSys=1, pNonSys=0
++ movl r17=FPSR_DEFAULT
++ ;;
++ mov.m ar.fpsr=r17 // set ar.fpsr to kernel default value
++(p10) mov r8=-EINVAL
++ br.ret.sptk.many b7
++END(ia64_syscall_setup)
++#endif
++
++ .org ia64_ivt+0x3c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x3c00 Entry 15 (size 64 bundles) Reserved
++ DBG_FAULT(15)
++ FAULT(15)
++
++ /*
++ * Squatting in this space ...
++ *
++ * This special case dispatcher for illegal operation faults allows preserved
++ * registers to be modified through a callback function (asm only) that is handed
++ * back from the fault handler in r8. Up to three arguments can be passed to the
++ * callback function by returning an aggregate with the callback as its first
++ * element, followed by the arguments.
++ */
++ENTRY(dispatch_illegal_op_fault)
++ .prologue
++ .body
++ SAVE_MIN_WITH_COVER
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collection is on
++ ;;
++(p15) ssm psr.i // restore psr.i
++ adds r3=8,r2 // set up second base pointer for SAVE_REST
++ ;;
++ alloc r14=ar.pfs,0,0,1,0 // must be first in insn group
++ mov out0=ar.ec
++ ;;
++ SAVE_REST
++ PT_REGS_UNWIND_INFO(0)
++ ;;
++ br.call.sptk.many rp=ia64_illegal_op_fault
++.ret0: ;;
++ alloc r14=ar.pfs,0,0,3,0 // must be first in insn group
++ mov out0=r9
++ mov out1=r10
++ mov out2=r11
++ movl r15=ia64_leave_kernel
++ ;;
++ mov rp=r15
++ mov b6=r8
++ ;;
++ cmp.ne p6,p0=0,r8
++(p6) br.call.dpnt.many b6=b6 // call returns to ia64_leave_kernel
++ br.sptk.many ia64_leave_kernel
++END(dispatch_illegal_op_fault)
++
++ .org ia64_ivt+0x4000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x4000 Entry 16 (size 64 bundles) Reserved
++ DBG_FAULT(16)
++ FAULT(16)
++
++ .org ia64_ivt+0x4400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x4400 Entry 17 (size 64 bundles) Reserved
++ DBG_FAULT(17)
++ FAULT(17)
++
++ENTRY(non_syscall)
++ mov ar.rsc=r27 // restore ar.rsc before SAVE_MIN_WITH_COVER
++ ;;
++ SAVE_MIN_WITH_COVER
++
++ // There is no particular reason for this code to be here, other than that
++ // there happens to be space here that would go unused otherwise. If this
++ // fault ever gets "unreserved", simply moved the following code to a more
++ // suitable spot...
++
++ alloc r14=ar.pfs,0,0,2,0
++ mov out0=cr.iim
++ add out1=16,sp
++ adds r3=8,r2 // set up second base pointer for SAVE_REST
++
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collection is on
++ ;;
++(p15) ssm psr.i // restore psr.i
++ movl r15=ia64_leave_kernel
++ ;;
++ SAVE_REST
++ mov rp=r15
++ ;;
++ br.call.sptk.many b6=ia64_bad_break // avoid WAW on CFM and ignore return addr
++END(non_syscall)
++
++ .org ia64_ivt+0x4800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x4800 Entry 18 (size 64 bundles) Reserved
++ DBG_FAULT(18)
++ FAULT(18)
++
++ /*
++ * There is no particular reason for this code to be here, other than that
++ * there happens to be space here that would go unused otherwise. If this
++ * fault ever gets "unreserved", simply moved the following code to a more
++ * suitable spot...
++ */
++
++ENTRY(dispatch_unaligned_handler)
++ SAVE_MIN_WITH_COVER
++ ;;
++ alloc r14=ar.pfs,0,0,2,0 // now it's safe (must be first in insn group!)
++ mov out0=cr.ifa
++ adds out1=16,sp
++
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collection is on
++ ;;
++(p15) ssm psr.i // restore psr.i
++ adds r3=8,r2 // set up second base pointer
++ ;;
++ SAVE_REST
++ movl r14=ia64_leave_kernel
++ ;;
++ mov rp=r14
++ br.sptk.many ia64_prepare_handle_unaligned
++END(dispatch_unaligned_handler)
++
++ .org ia64_ivt+0x4c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x4c00 Entry 19 (size 64 bundles) Reserved
++ DBG_FAULT(19)
++ FAULT(19)
++
++ /*
++ * There is no particular reason for this code to be here, other than that
++ * there happens to be space here that would go unused otherwise. If this
++ * fault ever gets "unreserved", simply moved the following code to a more
++ * suitable spot...
++ */
++
++ENTRY(dispatch_to_fault_handler)
++ /*
++ * Input:
++ * psr.ic: off
++ * r19: fault vector number (e.g., 24 for General Exception)
++ * r31: contains saved predicates (pr)
++ */
++ SAVE_MIN_WITH_COVER_R19
++ alloc r14=ar.pfs,0,0,5,0
++ mov out0=r15
++#ifdef CONFIG_XEN
++ movl out1=XSI_ISR
++ ;;
++ adds out2=XSI_IFA-XSI_ISR,out1
++ adds out3=XSI_IIM-XSI_ISR,out1
++ adds out4=XSI_ITIR-XSI_ISR,out1
++ ;;
++ ld8 out1=[out1]
++ ld8 out2=[out2]
++ ld8 out3=[out4]
++ ld8 out4=[out4]
++ ;;
++#else
++ mov out1=cr.isr
++ mov out2=cr.ifa
++ mov out3=cr.iim
++ mov out4=cr.itir
++ ;;
++#endif
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collection is on
++ ;;
++(p15) ssm psr.i // restore psr.i
++ adds r3=8,r2 // set up second base pointer for SAVE_REST
++ ;;
++ SAVE_REST
++ movl r14=ia64_leave_kernel
++ ;;
++ mov rp=r14
++ br.call.sptk.many b6=ia64_fault
++END(dispatch_to_fault_handler)
++
++//
++// --- End of long entries, Beginning of short entries
++//
++
++ .org ia64_ivt+0x5000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5000 Entry 20 (size 16 bundles) Page Not Present (10,22,49)
++ENTRY(page_not_present)
++ DBG_FAULT(20)
++ mov r16=cr.ifa
++ rsm psr.dt
++ /*
++ * The Linux page fault handler doesn't expect non-present pages to be in
++ * the TLB. Flush the existing entry now, so we meet that expectation.
++ */
++ mov r17=PAGE_SHIFT<<2
++ ;;
++ ptc.l r16,r17
++ ;;
++ mov r31=pr
++ srlz.d
++ br.sptk.many page_fault
++END(page_not_present)
++
++ .org ia64_ivt+0x5100
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5100 Entry 21 (size 16 bundles) Key Permission (13,25,52)
++ENTRY(key_permission)
++ DBG_FAULT(21)
++ mov r16=cr.ifa
++ rsm psr.dt
++ mov r31=pr
++ ;;
++ srlz.d
++ br.sptk.many page_fault
++END(key_permission)
++
++ .org ia64_ivt+0x5200
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5200 Entry 22 (size 16 bundles) Instruction Access Rights (26)
++ENTRY(iaccess_rights)
++ DBG_FAULT(22)
++ mov r16=cr.ifa
++ rsm psr.dt
++ mov r31=pr
++ ;;
++ srlz.d
++ br.sptk.many page_fault
++END(iaccess_rights)
++
++ .org ia64_ivt+0x5300
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5300 Entry 23 (size 16 bundles) Data Access Rights (14,53)
++ENTRY(daccess_rights)
++ DBG_FAULT(23)
++#ifdef CONFIG_XEN
++ movl r16=XSI_IFA
++ ;;
++ ld8 r16=[r16]
++ ;;
++ XEN_HYPER_RSM_PSR_DT
++#else
++ mov r16=cr.ifa
++ rsm psr.dt
++#endif
++ mov r31=pr
++ ;;
++ srlz.d
++ br.sptk.many page_fault
++END(daccess_rights)
++
++ .org ia64_ivt+0x5400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5400 Entry 24 (size 16 bundles) General Exception (5,32,34,36,38,39)
++ENTRY(general_exception)
++ DBG_FAULT(24)
++ mov r16=cr.isr
++ mov r31=pr
++ ;;
++ cmp4.eq p6,p0=0,r16
++(p6) br.sptk.many dispatch_illegal_op_fault
++ ;;
++ mov r19=24 // fault number
++ br.sptk.many dispatch_to_fault_handler
++END(general_exception)
++
++ .org ia64_ivt+0x5500
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5500 Entry 25 (size 16 bundles) Disabled FP-Register (35)
++ENTRY(disabled_fp_reg)
++ DBG_FAULT(25)
++ rsm psr.dfh // ensure we can access fph
++ ;;
++ srlz.d
++ mov r31=pr
++ mov r19=25
++ br.sptk.many dispatch_to_fault_handler
++END(disabled_fp_reg)
++
++ .org ia64_ivt+0x5600
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5600 Entry 26 (size 16 bundles) Nat Consumption (11,23,37,50)
++ENTRY(nat_consumption)
++ DBG_FAULT(26)
++
++ mov r16=cr.ipsr
++ mov r17=cr.isr
++ mov r31=pr // save PR
++ ;;
++ and r18=0xf,r17 // r18 = cr.ipsr.code{3:0}
++ tbit.z p6,p0=r17,IA64_ISR_NA_BIT
++ ;;
++ cmp.ne.or p6,p0=IA64_ISR_CODE_LFETCH,r18
++ dep r16=-1,r16,IA64_PSR_ED_BIT,1
++(p6) br.cond.spnt 1f // branch if (cr.ispr.na == 0 || cr.ipsr.code{3:0} != LFETCH)
++ ;;
++ mov cr.ipsr=r16 // set cr.ipsr.na
++ mov pr=r31,-1
++ ;;
++ rfi
++
++1: mov pr=r31,-1
++ ;;
++ FAULT(26)
++END(nat_consumption)
++
++ .org ia64_ivt+0x5700
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5700 Entry 27 (size 16 bundles) Speculation (40)
++ENTRY(speculation_vector)
++ DBG_FAULT(27)
++ /*
++ * A [f]chk.[as] instruction needs to take the branch to the recovery code but
++ * this part of the architecture is not implemented in hardware on some CPUs, such
++ * as Itanium. Thus, in general we need to emulate the behavior. IIM contains
++ * the relative target (not yet sign extended). So after sign extending it we
++ * simply add it to IIP. We also need to reset the EI field of the IPSR to zero,
++ * i.e., the slot to restart into.
++ *
++ * cr.imm contains zero_ext(imm21)
++ */
++ mov r18=cr.iim
++ ;;
++ mov r17=cr.iip
++ shl r18=r18,43 // put sign bit in position (43=64-21)
++ ;;
++
++ mov r16=cr.ipsr
++ shr r18=r18,39 // sign extend (39=43-4)
++ ;;
++
++ add r17=r17,r18 // now add the offset
++ ;;
++ mov cr.iip=r17
++ dep r16=0,r16,41,2 // clear EI
++ ;;
++
++ mov cr.ipsr=r16
++ ;;
++
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI;
++#else
++ rfi // and go back
++#endif
++END(speculation_vector)
++
++ .org ia64_ivt+0x5800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5800 Entry 28 (size 16 bundles) Reserved
++ DBG_FAULT(28)
++ FAULT(28)
++
++ .org ia64_ivt+0x5900
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5900 Entry 29 (size 16 bundles) Debug (16,28,56)
++ENTRY(debug_vector)
++ DBG_FAULT(29)
++ FAULT(29)
++END(debug_vector)
++
++ .org ia64_ivt+0x5a00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5a00 Entry 30 (size 16 bundles) Unaligned Reference (57)
++ENTRY(unaligned_access)
++ DBG_FAULT(30)
++ mov r31=pr // prepare to save predicates
++ ;;
++ br.sptk.many dispatch_unaligned_handler
++END(unaligned_access)
++
++ .org ia64_ivt+0x5b00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5b00 Entry 31 (size 16 bundles) Unsupported Data Reference (57)
++ENTRY(unsupported_data_reference)
++ DBG_FAULT(31)
++ FAULT(31)
++END(unsupported_data_reference)
++
++ .org ia64_ivt+0x5c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5c00 Entry 32 (size 16 bundles) Floating-Point Fault (64)
++ENTRY(floating_point_fault)
++ DBG_FAULT(32)
++ FAULT(32)
++END(floating_point_fault)
++
++ .org ia64_ivt+0x5d00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5d00 Entry 33 (size 16 bundles) Floating Point Trap (66)
++ENTRY(floating_point_trap)
++ DBG_FAULT(33)
++ FAULT(33)
++END(floating_point_trap)
++
++ .org ia64_ivt+0x5e00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5e00 Entry 34 (size 16 bundles) Lower Privilege Transfer Trap (66)
++ENTRY(lower_privilege_trap)
++ DBG_FAULT(34)
++ FAULT(34)
++END(lower_privilege_trap)
++
++ .org ia64_ivt+0x5f00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x5f00 Entry 35 (size 16 bundles) Taken Branch Trap (68)
++ENTRY(taken_branch_trap)
++ DBG_FAULT(35)
++ FAULT(35)
++END(taken_branch_trap)
++
++ .org ia64_ivt+0x6000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6000 Entry 36 (size 16 bundles) Single Step Trap (69)
++ENTRY(single_step_trap)
++ DBG_FAULT(36)
++ FAULT(36)
++END(single_step_trap)
++
++ .org ia64_ivt+0x6100
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6100 Entry 37 (size 16 bundles) Reserved
++ DBG_FAULT(37)
++ FAULT(37)
++
++ .org ia64_ivt+0x6200
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6200 Entry 38 (size 16 bundles) Reserved
++ DBG_FAULT(38)
++ FAULT(38)
++
++ .org ia64_ivt+0x6300
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6300 Entry 39 (size 16 bundles) Reserved
++ DBG_FAULT(39)
++ FAULT(39)
++
++ .org ia64_ivt+0x6400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6400 Entry 40 (size 16 bundles) Reserved
++ DBG_FAULT(40)
++ FAULT(40)
++
++ .org ia64_ivt+0x6500
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6500 Entry 41 (size 16 bundles) Reserved
++ DBG_FAULT(41)
++ FAULT(41)
++
++ .org ia64_ivt+0x6600
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6600 Entry 42 (size 16 bundles) Reserved
++ DBG_FAULT(42)
++ FAULT(42)
++
++ .org ia64_ivt+0x6700
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6700 Entry 43 (size 16 bundles) Reserved
++ DBG_FAULT(43)
++ FAULT(43)
++
++ .org ia64_ivt+0x6800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6800 Entry 44 (size 16 bundles) Reserved
++ DBG_FAULT(44)
++ FAULT(44)
++
++ .org ia64_ivt+0x6900
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6900 Entry 45 (size 16 bundles) IA-32 Exeception (17,18,29,41,42,43,44,58,60,61,62,72,73,75,76,77)
++ENTRY(ia32_exception)
++ DBG_FAULT(45)
++ FAULT(45)
++END(ia32_exception)
++
++ .org ia64_ivt+0x6a00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6a00 Entry 46 (size 16 bundles) IA-32 Intercept (30,31,59,70,71)
++ENTRY(ia32_intercept)
++ DBG_FAULT(46)
++#ifdef CONFIG_IA32_SUPPORT
++ mov r31=pr
++ mov r16=cr.isr
++ ;;
++ extr.u r17=r16,16,8 // get ISR.code
++ mov r18=ar.eflag
++ mov r19=cr.iim // old eflag value
++ ;;
++ cmp.ne p6,p0=2,r17
++(p6) br.cond.spnt 1f // not a system flag fault
++ xor r16=r18,r19
++ ;;
++ extr.u r17=r16,18,1 // get the eflags.ac bit
++ ;;
++ cmp.eq p6,p0=0,r17
++(p6) br.cond.spnt 1f // eflags.ac bit didn't change
++ ;;
++ mov pr=r31,-1 // restore predicate registers
++#ifdef CONFIG_XEN
++ XEN_HYPER_RFI;
++#else
++ rfi
++#endif
++
++1:
++#endif // CONFIG_IA32_SUPPORT
++ FAULT(46)
++END(ia32_intercept)
++
++ .org ia64_ivt+0x6b00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6b00 Entry 47 (size 16 bundles) IA-32 Interrupt (74)
++ENTRY(ia32_interrupt)
++ DBG_FAULT(47)
++#ifdef CONFIG_IA32_SUPPORT
++ mov r31=pr
++ br.sptk.many dispatch_to_ia32_handler
++#else
++ FAULT(47)
++#endif
++END(ia32_interrupt)
++
++ .org ia64_ivt+0x6c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6c00 Entry 48 (size 16 bundles) Reserved
++ DBG_FAULT(48)
++ FAULT(48)
++
++ .org ia64_ivt+0x6d00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6d00 Entry 49 (size 16 bundles) Reserved
++ DBG_FAULT(49)
++ FAULT(49)
++
++ .org ia64_ivt+0x6e00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6e00 Entry 50 (size 16 bundles) Reserved
++ DBG_FAULT(50)
++ FAULT(50)
++
++ .org ia64_ivt+0x6f00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x6f00 Entry 51 (size 16 bundles) Reserved
++ DBG_FAULT(51)
++ FAULT(51)
++
++ .org ia64_ivt+0x7000
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7000 Entry 52 (size 16 bundles) Reserved
++ DBG_FAULT(52)
++ FAULT(52)
++
++ .org ia64_ivt+0x7100
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7100 Entry 53 (size 16 bundles) Reserved
++ DBG_FAULT(53)
++ FAULT(53)
++
++ .org ia64_ivt+0x7200
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7200 Entry 54 (size 16 bundles) Reserved
++ DBG_FAULT(54)
++ FAULT(54)
++
++ .org ia64_ivt+0x7300
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7300 Entry 55 (size 16 bundles) Reserved
++ DBG_FAULT(55)
++ FAULT(55)
++
++ .org ia64_ivt+0x7400
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7400 Entry 56 (size 16 bundles) Reserved
++ DBG_FAULT(56)
++ FAULT(56)
++
++ .org ia64_ivt+0x7500
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7500 Entry 57 (size 16 bundles) Reserved
++ DBG_FAULT(57)
++ FAULT(57)
++
++ .org ia64_ivt+0x7600
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7600 Entry 58 (size 16 bundles) Reserved
++ DBG_FAULT(58)
++ FAULT(58)
++
++ .org ia64_ivt+0x7700
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7700 Entry 59 (size 16 bundles) Reserved
++ DBG_FAULT(59)
++ FAULT(59)
++
++ .org ia64_ivt+0x7800
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7800 Entry 60 (size 16 bundles) Reserved
++ DBG_FAULT(60)
++ FAULT(60)
++
++ .org ia64_ivt+0x7900
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7900 Entry 61 (size 16 bundles) Reserved
++ DBG_FAULT(61)
++ FAULT(61)
++
++ .org ia64_ivt+0x7a00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7a00 Entry 62 (size 16 bundles) Reserved
++ DBG_FAULT(62)
++ FAULT(62)
++
++ .org ia64_ivt+0x7b00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7b00 Entry 63 (size 16 bundles) Reserved
++ DBG_FAULT(63)
++ FAULT(63)
++
++ .org ia64_ivt+0x7c00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7c00 Entry 64 (size 16 bundles) Reserved
++ DBG_FAULT(64)
++ FAULT(64)
++
++ .org ia64_ivt+0x7d00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7d00 Entry 65 (size 16 bundles) Reserved
++ DBG_FAULT(65)
++ FAULT(65)
++
++ .org ia64_ivt+0x7e00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7e00 Entry 66 (size 16 bundles) Reserved
++ DBG_FAULT(66)
++ FAULT(66)
++
++#ifdef CONFIG_XEN
++ /*
++ * There is no particular reason for this code to be here, other than that
++ * there happens to be space here that would go unused otherwise. If this
++ * fault ever gets "unreserved", simply moved the following code to a more
++ * suitable spot...
++ */
++
++GLOBAL_ENTRY(xen_bsw1)
++ /* FIXME: THIS CODE IS NOT NaT SAFE! */
++ movl r30=XSI_BANKNUM;
++ mov r31=1;;
++ st4 [r30]=r31;
++ movl r30=XSI_BANK1_R16;
++ movl r31=XSI_BANK1_R16+8;;
++ ld8 r16=[r30],16; ld8 r17=[r31],16;;
++ ld8 r18=[r30],16; ld8 r19=[r31],16;;
++ ld8 r20=[r30],16; ld8 r21=[r31],16;;
++ ld8 r22=[r30],16; ld8 r23=[r31],16;;
++ ld8 r24=[r30],16; ld8 r25=[r31],16;;
++ ld8 r26=[r30],16; ld8 r27=[r31],16;;
++ ld8 r28=[r30],16; ld8 r29=[r31],16;;
++ ld8 r30=[r30]; ld8 r31=[r31];;
++ br.ret.sptk.many b0
++END(xen_bsw1)
++#endif
++
++ .org ia64_ivt+0x7f00
++/////////////////////////////////////////////////////////////////////////////////////////
++// 0x7f00 Entry 67 (size 16 bundles) Reserved
++ DBG_FAULT(67)
++ FAULT(67)
++
++#ifdef CONFIG_IA32_SUPPORT
++
++ /*
++ * There is no particular reason for this code to be here, other than that
++ * there happens to be space here that would go unused otherwise. If this
++ * fault ever gets "unreserved", simply moved the following code to a more
++ * suitable spot...
++ */
++
++ // IA32 interrupt entry point
++
++ENTRY(dispatch_to_ia32_handler)
++ SAVE_MIN
++ ;;
++ mov r14=cr.isr
++ ssm psr.ic | PSR_DEFAULT_BITS
++ ;;
++ srlz.i // guarantee that interruption collection is on
++ ;;
++(p15) ssm psr.i
++ adds r3=8,r2 // Base pointer for SAVE_REST
++ ;;
++ SAVE_REST
++ ;;
++ mov r15=0x80
++ shr r14=r14,16 // Get interrupt number
++ ;;
++ cmp.ne p6,p0=r14,r15
++(p6) br.call.dpnt.many b6=non_ia32_syscall
++
++ adds r14=IA64_PT_REGS_R8_OFFSET + 16,sp // 16 byte hole per SW conventions
++ adds r15=IA64_PT_REGS_R1_OFFSET + 16,sp
++ ;;
++ cmp.eq pSys,pNonSys=r0,r0 // set pSys=1, pNonSys=0
++ ld8 r8=[r14] // get r8
++ ;;
++ st8 [r15]=r8 // save original EAX in r1 (IA32 procs don't use the GP)
++ ;;
++ alloc r15=ar.pfs,0,0,6,0 // must first in an insn group
++ ;;
++ ld4 r8=[r14],8 // r8 == eax (syscall number)
++ mov r15=IA32_NR_syscalls
++ ;;
++ cmp.ltu.unc p6,p7=r8,r15
++ ld4 out1=[r14],8 // r9 == ecx
++ ;;
++ ld4 out2=[r14],8 // r10 == edx
++ ;;
++ ld4 out0=[r14] // r11 == ebx
++ adds r14=(IA64_PT_REGS_R13_OFFSET) + 16,sp
++ ;;
++ ld4 out5=[r14],PT(R14)-PT(R13) // r13 == ebp
++ ;;
++ ld4 out3=[r14],PT(R15)-PT(R14) // r14 == esi
++ adds r2=TI_FLAGS+IA64_TASK_SIZE,r13
++ ;;
++ ld4 out4=[r14] // r15 == edi
++ movl r16=ia32_syscall_table
++ ;;
++(p6) shladd r16=r8,3,r16 // force ni_syscall if not valid syscall number
++ ld4 r2=[r2] // r2 = current_thread_info()->flags
++ ;;
++ ld8 r16=[r16]
++ and r2=_TIF_SYSCALL_TRACEAUDIT,r2 // mask trace or audit
++ ;;
++ mov b6=r16
++ movl r15=ia32_ret_from_syscall
++ cmp.eq p8,p0=r2,r0
++ ;;
++ mov rp=r15
++(p8) br.call.sptk.many b6=b6
++ br.cond.sptk ia32_trace_syscall
++
++non_ia32_syscall:
++ alloc r15=ar.pfs,0,0,2,0
++ mov out0=r14 // interrupt #
++ add out1=16,sp // pointer to pt_regs
++ ;; // avoid WAW on CFM
++ br.call.sptk.many rp=ia32_bad_interrupt
++.ret1: movl r15=ia64_leave_kernel
++ ;;
++ mov rp=r15
++ br.ret.sptk.many rp
++END(dispatch_to_ia32_handler)
++#endif /* CONFIG_IA32_SUPPORT */
++
++#ifdef CONFIG_XEN
++ .section .text,"ax"
++GLOBAL_ENTRY(xen_event_callback)
++ mov r31=pr // prepare to save predicates
++ ;;
++ SAVE_MIN_WITH_COVER // uses r31; defines r2 and r3
++ ;;
++ movl r3=XSI_PSR_IC
++ mov r14=1
++ ;;
++ st4 [r3]=r14
++ ;;
++ adds r3=8,r2 // set up second base pointer for SAVE_REST
++ srlz.i // ensure everybody knows psr.ic is back on
++ ;;
++ SAVE_REST
++ ;;
++ alloc r14=ar.pfs,0,0,1,0 // must be first in an insn group
++ add out0=16,sp // pass pointer to pt_regs as first arg
++ ;;
++ srlz.d // make sure we see the effect of cr.ivr
++ movl r14=ia64_leave_kernel
++ ;;
++ mov rp=r14
++ br.call.sptk.many b6=evtchn_do_upcall
++END(xen_event_callback)
++#endif
+diff -urNp linux-2.6/arch/ia64/xen/xenminstate.h new/arch/ia64/xen/xenminstate.h
+--- linux-2.6/arch/ia64/xen/xenminstate.h 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xenminstate.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,369 @@
++#include <linux/config.h>
++
++#include <asm/cache.h>
++
++#ifdef CONFIG_XEN
++#include "../kernel/entry.h"
++#else
++#include "entry.h"
++#endif
++
++/*
++ * For ivt.s we want to access the stack virtually so we don't have to disable translation
++ * on interrupts.
++ *
++ * On entry:
++ * r1: pointer to current task (ar.k6)
++ */
++#define MINSTATE_START_SAVE_MIN_VIRT \
++(pUStk) mov ar.rsc=0; /* set enforced lazy mode, pl 0, little-endian, loadrs=0 */ \
++ ;; \
++(pUStk) mov.m r24=ar.rnat; \
++(pUStk) addl r22=IA64_RBS_OFFSET,r1; /* compute base of RBS */ \
++(pKStk) mov r1=sp; /* get sp */ \
++ ;; \
++(pUStk) lfetch.fault.excl.nt1 [r22]; \
++(pUStk) addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1; /* compute base of memory stack */ \
++(pUStk) mov r23=ar.bspstore; /* save ar.bspstore */ \
++ ;; \
++(pUStk) mov ar.bspstore=r22; /* switch to kernel RBS */ \
++(pKStk) addl r1=-IA64_PT_REGS_SIZE,r1; /* if in kernel mode, use sp (r12) */ \
++ ;; \
++(pUStk) mov r18=ar.bsp; \
++(pUStk) mov ar.rsc=0x3; /* set eager mode, pl 0, little-endian, loadrs=0 */ \
++
++#define MINSTATE_END_SAVE_MIN_VIRT \
++ bsw.1; /* switch back to bank 1 (must be last in insn group) */ \
++ ;;
++
++/*
++ * For mca_asm.S we want to access the stack physically since the state is saved before we
++ * go virtual and don't want to destroy the iip or ipsr.
++ */
++#define MINSTATE_START_SAVE_MIN_PHYS \
++(pKStk) mov r3=IA64_KR(PER_CPU_DATA);; \
++(pKStk) addl r3=THIS_CPU(ia64_mca_data),r3;; \
++(pKStk) ld8 r3 = [r3];; \
++(pKStk) addl r3=IA64_MCA_CPU_INIT_STACK_OFFSET,r3;; \
++(pKStk) addl sp=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r3; \
++(pUStk) mov ar.rsc=0; /* set enforced lazy mode, pl 0, little-endian, loadrs=0 */ \
++(pUStk) addl r22=IA64_RBS_OFFSET,r1; /* compute base of register backing store */ \
++ ;; \
++(pUStk) mov r24=ar.rnat; \
++(pUStk) addl r1=IA64_STK_OFFSET-IA64_PT_REGS_SIZE,r1; /* compute base of memory stack */ \
++(pUStk) mov r23=ar.bspstore; /* save ar.bspstore */ \
++(pUStk) dep r22=-1,r22,61,3; /* compute kernel virtual addr of RBS */ \
++ ;; \
++(pKStk) addl r1=-IA64_PT_REGS_SIZE,r1; /* if in kernel mode, use sp (r12) */ \
++(pUStk) mov ar.bspstore=r22; /* switch to kernel RBS */ \
++ ;; \
++(pUStk) mov r18=ar.bsp; \
++(pUStk) mov ar.rsc=0x3; /* set eager mode, pl 0, little-endian, loadrs=0 */ \
++
++#define MINSTATE_END_SAVE_MIN_PHYS \
++ dep r12=-1,r12,61,3; /* make sp a kernel virtual address */ \
++ ;;
++
++#ifdef MINSTATE_VIRT
++# define MINSTATE_GET_CURRENT(reg) mov reg=IA64_KR(CURRENT)
++# define MINSTATE_START_SAVE_MIN MINSTATE_START_SAVE_MIN_VIRT
++# define MINSTATE_END_SAVE_MIN MINSTATE_END_SAVE_MIN_VIRT
++#endif
++
++#ifdef MINSTATE_PHYS
++# define MINSTATE_GET_CURRENT(reg) mov reg=IA64_KR(CURRENT);; tpa reg=reg
++# define MINSTATE_START_SAVE_MIN MINSTATE_START_SAVE_MIN_PHYS
++# define MINSTATE_END_SAVE_MIN MINSTATE_END_SAVE_MIN_PHYS
++#endif
++
++/*
++ * DO_SAVE_MIN switches to the kernel stacks (if necessary) and saves
++ * the minimum state necessary that allows us to turn psr.ic back
++ * on.
++ *
++ * Assumed state upon entry:
++ * psr.ic: off
++ * r31: contains saved predicates (pr)
++ *
++ * Upon exit, the state is as follows:
++ * psr.ic: off
++ * r2 = points to &pt_regs.r16
++ * r8 = contents of ar.ccv
++ * r9 = contents of ar.csd
++ * r10 = contents of ar.ssd
++ * r11 = FPSR_DEFAULT
++ * r12 = kernel sp (kernel virtual address)
++ * r13 = points to current task_struct (kernel virtual address)
++ * p15 = TRUE if psr.i is set in cr.ipsr
++ * predicate registers (other than p2, p3, and p15), b6, r3, r14, r15:
++ * preserved
++ * CONFIG_XEN note: p6/p7 are not preserved
++ *
++ * Note that psr.ic is NOT turned on by this macro. This is so that
++ * we can pass interruption state as arguments to a handler.
++ */
++#ifdef CONFIG_XEN
++#define DO_SAVE_MIN(COVER,SAVE_IFS,EXTRA) \
++ MINSTATE_GET_CURRENT(r16); /* M (or M;;I) */ \
++ mov r27=ar.rsc; /* M */ \
++ mov r20=r1; /* A */ \
++ mov r25=ar.unat; /* M */ \
++ /* mov r29=cr.ipsr; /* M */ \
++ movl r29=XSI_IPSR;; \
++ ld8 r29=[r29];; \
++ mov r26=ar.pfs; /* I */ \
++ /* mov r28=cr.iip; /* M */ \
++ movl r28=XSI_IIP;; \
++ ld8 r28=[r28];; \
++ mov r21=ar.fpsr; /* M */ \
++ COVER; /* B;; (or nothing) */ \
++ ;; \
++ adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16; \
++ ;; \
++ ld1 r17=[r16]; /* load current->thread.on_ustack flag */ \
++ st1 [r16]=r0; /* clear current->thread.on_ustack flag */ \
++ adds r1=-IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 \
++ /* switch from user to kernel RBS: */ \
++ ;; \
++ invala; /* M */ \
++ /* SAVE_IFS; /* see xen special handling below */ \
++ cmp.eq pKStk,pUStk=r0,r17; /* are we in kernel mode already? */ \
++ ;; \
++ MINSTATE_START_SAVE_MIN \
++ adds r17=2*L1_CACHE_BYTES,r1; /* really: biggest cache-line size */ \
++ adds r16=PT(CR_IPSR),r1; \
++ ;; \
++ lfetch.fault.excl.nt1 [r17],L1_CACHE_BYTES; \
++ st8 [r16]=r29; /* save cr.ipsr */ \
++ ;; \
++ lfetch.fault.excl.nt1 [r17]; \
++ tbit.nz p15,p0=r29,IA64_PSR_I_BIT; \
++ mov r29=b0 \
++ ;; \
++ adds r16=PT(R8),r1; /* initialize first base pointer */ \
++ adds r17=PT(R9),r1; /* initialize second base pointer */ \
++(pKStk) mov r18=r0; /* make sure r18 isn't NaT */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r8,16; \
++.mem.offset 8,0; st8.spill [r17]=r9,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r10,24; \
++.mem.offset 8,0; st8.spill [r17]=r11,24; \
++ ;; \
++ /* xen special handling for possibly lazy cover */ \
++ movl r8=XSI_INCOMPL_REGFR; \
++ ;; \
++ ld4 r30=[r8]; \
++ ;; \
++ /* set XSI_INCOMPL_REGFR 0 */ \
++ st4 [r8]=r0; \
++ cmp.eq p6,p7=r30,r0; \
++ ;; /* not sure if this stop bit is necessary */ \
++(p6) adds r8=XSI_PRECOVER_IFS-XSI_INCOMPL_REGFR,r8; \
++(p7) adds r8=XSI_IFS-XSI_INCOMPL_REGFR,r8; \
++ ;; \
++ ld8 r30=[r8]; \
++ ;; \
++ st8 [r16]=r28,16; /* save cr.iip */ \
++ st8 [r17]=r30,16; /* save cr.ifs */ \
++(pUStk) sub r18=r18,r22; /* r18=RSE.ndirty*8 */ \
++ mov r8=ar.ccv; \
++ mov r9=ar.csd; \
++ mov r10=ar.ssd; \
++ movl r11=FPSR_DEFAULT; /* L-unit */ \
++ ;; \
++ st8 [r16]=r25,16; /* save ar.unat */ \
++ st8 [r17]=r26,16; /* save ar.pfs */ \
++ shl r18=r18,16; /* compute ar.rsc to be used for "loadrs" */ \
++ ;; \
++ st8 [r16]=r27,16; /* save ar.rsc */ \
++(pUStk) st8 [r17]=r24,16; /* save ar.rnat */ \
++(pKStk) adds r17=16,r17; /* skip over ar_rnat field */ \
++ ;; /* avoid RAW on r16 & r17 */ \
++(pUStk) st8 [r16]=r23,16; /* save ar.bspstore */ \
++ st8 [r17]=r31,16; /* save predicates */ \
++(pKStk) adds r16=16,r16; /* skip over ar_bspstore field */ \
++ ;; \
++ st8 [r16]=r29,16; /* save b0 */ \
++ st8 [r17]=r18,16; /* save ar.rsc value for "loadrs" */ \
++ cmp.eq pNonSys,pSys=r0,r0 /* initialize pSys=0, pNonSys=1 */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r20,16; /* save original r1 */ \
++.mem.offset 8,0; st8.spill [r17]=r12,16; \
++ adds r12=-16,r1; /* switch to kernel memory stack (with 16 bytes of scratch) */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r13,16; \
++.mem.offset 8,0; st8.spill [r17]=r21,16; /* save ar.fpsr */ \
++ mov r13=IA64_KR(CURRENT); /* establish `current' */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r15,16; \
++.mem.offset 8,0; st8.spill [r17]=r14,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r2,16; \
++.mem.offset 8,0; st8.spill [r17]=r3,16; \
++ ;; \
++ EXTRA; \
++ mov r2=b0; br.call.sptk b0=xen_bsw1;; mov b0=r2; \
++ adds r2=IA64_PT_REGS_R16_OFFSET,r1; \
++ ;; \
++ movl r1=__gp; /* establish kernel global pointer */ \
++ ;; \
++ /* MINSTATE_END_SAVE_MIN */
++#else
++#define DO_SAVE_MIN(COVER,SAVE_IFS,EXTRA) \
++ MINSTATE_GET_CURRENT(r16); /* M (or M;;I) */ \
++ mov r27=ar.rsc; /* M */ \
++ mov r20=r1; /* A */ \
++ mov r25=ar.unat; /* M */ \
++ mov r29=cr.ipsr; /* M */ \
++ mov r26=ar.pfs; /* I */ \
++ mov r28=cr.iip; /* M */ \
++ mov r21=ar.fpsr; /* M */ \
++ COVER; /* B;; (or nothing) */ \
++ ;; \
++ adds r16=IA64_TASK_THREAD_ON_USTACK_OFFSET,r16; \
++ ;; \
++ ld1 r17=[r16]; /* load current->thread.on_ustack flag */ \
++ st1 [r16]=r0; /* clear current->thread.on_ustack flag */ \
++ adds r1=-IA64_TASK_THREAD_ON_USTACK_OFFSET,r16 \
++ /* switch from user to kernel RBS: */ \
++ ;; \
++ invala; /* M */ \
++ SAVE_IFS; \
++ cmp.eq pKStk,pUStk=r0,r17; /* are we in kernel mode already? */ \
++ ;; \
++ MINSTATE_START_SAVE_MIN \
++ adds r17=2*L1_CACHE_BYTES,r1; /* really: biggest cache-line size */ \
++ adds r16=PT(CR_IPSR),r1; \
++ ;; \
++ lfetch.fault.excl.nt1 [r17],L1_CACHE_BYTES; \
++ st8 [r16]=r29; /* save cr.ipsr */ \
++ ;; \
++ lfetch.fault.excl.nt1 [r17]; \
++ tbit.nz p15,p0=r29,IA64_PSR_I_BIT; \
++ mov r29=b0 \
++ ;; \
++ adds r16=PT(R8),r1; /* initialize first base pointer */ \
++ adds r17=PT(R9),r1; /* initialize second base pointer */ \
++(pKStk) mov r18=r0; /* make sure r18 isn't NaT */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r8,16; \
++.mem.offset 8,0; st8.spill [r17]=r9,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r10,24; \
++.mem.offset 8,0; st8.spill [r17]=r11,24; \
++ ;; \
++ st8 [r16]=r28,16; /* save cr.iip */ \
++ st8 [r17]=r30,16; /* save cr.ifs */ \
++(pUStk) sub r18=r18,r22; /* r18=RSE.ndirty*8 */ \
++ mov r8=ar.ccv; \
++ mov r9=ar.csd; \
++ mov r10=ar.ssd; \
++ movl r11=FPSR_DEFAULT; /* L-unit */ \
++ ;; \
++ st8 [r16]=r25,16; /* save ar.unat */ \
++ st8 [r17]=r26,16; /* save ar.pfs */ \
++ shl r18=r18,16; /* compute ar.rsc to be used for "loadrs" */ \
++ ;; \
++ st8 [r16]=r27,16; /* save ar.rsc */ \
++(pUStk) st8 [r17]=r24,16; /* save ar.rnat */ \
++(pKStk) adds r17=16,r17; /* skip over ar_rnat field */ \
++ ;; /* avoid RAW on r16 & r17 */ \
++(pUStk) st8 [r16]=r23,16; /* save ar.bspstore */ \
++ st8 [r17]=r31,16; /* save predicates */ \
++(pKStk) adds r16=16,r16; /* skip over ar_bspstore field */ \
++ ;; \
++ st8 [r16]=r29,16; /* save b0 */ \
++ st8 [r17]=r18,16; /* save ar.rsc value for "loadrs" */ \
++ cmp.eq pNonSys,pSys=r0,r0 /* initialize pSys=0, pNonSys=1 */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r20,16; /* save original r1 */ \
++.mem.offset 8,0; st8.spill [r17]=r12,16; \
++ adds r12=-16,r1; /* switch to kernel memory stack (with 16 bytes of scratch) */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r13,16; \
++.mem.offset 8,0; st8.spill [r17]=r21,16; /* save ar.fpsr */ \
++ mov r13=IA64_KR(CURRENT); /* establish `current' */ \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r15,16; \
++.mem.offset 8,0; st8.spill [r17]=r14,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r16]=r2,16; \
++.mem.offset 8,0; st8.spill [r17]=r3,16; \
++ adds r2=IA64_PT_REGS_R16_OFFSET,r1; \
++ ;; \
++ EXTRA; \
++ movl r1=__gp; /* establish kernel global pointer */ \
++ ;; \
++ MINSTATE_END_SAVE_MIN
++#endif
++
++/*
++ * SAVE_REST saves the remainder of pt_regs (with psr.ic on).
++ *
++ * Assumed state upon entry:
++ * psr.ic: on
++ * r2: points to &pt_regs.r16
++ * r3: points to &pt_regs.r17
++ * r8: contents of ar.ccv
++ * r9: contents of ar.csd
++ * r10: contents of ar.ssd
++ * r11: FPSR_DEFAULT
++ *
++ * Registers r14 and r15 are guaranteed not to be touched by SAVE_REST.
++ */
++#define SAVE_REST \
++.mem.offset 0,0; st8.spill [r2]=r16,16; \
++.mem.offset 8,0; st8.spill [r3]=r17,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r18,16; \
++.mem.offset 8,0; st8.spill [r3]=r19,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r20,16; \
++.mem.offset 8,0; st8.spill [r3]=r21,16; \
++ mov r18=b6; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r22,16; \
++.mem.offset 8,0; st8.spill [r3]=r23,16; \
++ mov r19=b7; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r24,16; \
++.mem.offset 8,0; st8.spill [r3]=r25,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r26,16; \
++.mem.offset 8,0; st8.spill [r3]=r27,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r28,16; \
++.mem.offset 8,0; st8.spill [r3]=r29,16; \
++ ;; \
++.mem.offset 0,0; st8.spill [r2]=r30,16; \
++.mem.offset 8,0; st8.spill [r3]=r31,32; \
++ ;; \
++ mov ar.fpsr=r11; /* M-unit */ \
++ st8 [r2]=r8,8; /* ar.ccv */ \
++ adds r24=PT(B6)-PT(F7),r3; \
++ ;; \
++ stf.spill [r2]=f6,32; \
++ stf.spill [r3]=f7,32; \
++ ;; \
++ stf.spill [r2]=f8,32; \
++ stf.spill [r3]=f9,32; \
++ ;; \
++ stf.spill [r2]=f10; \
++ stf.spill [r3]=f11; \
++ adds r25=PT(B7)-PT(F11),r3; \
++ ;; \
++ st8 [r24]=r18,16; /* b6 */ \
++ st8 [r25]=r19,16; /* b7 */ \
++ ;; \
++ st8 [r24]=r9; /* ar.csd */ \
++ st8 [r25]=r10; /* ar.ssd */ \
++ ;;
++
++#define SAVE_MIN_WITH_COVER DO_SAVE_MIN(cover, mov r30=cr.ifs,)
++#define SAVE_MIN_WITH_COVER_R19 DO_SAVE_MIN(cover, mov r30=cr.ifs, mov r15=r19)
++#ifdef CONFIG_XEN
++#define SAVE_MIN break 0;; /* FIXME: non-cover version only for ia32 support? */
++#else
++#define SAVE_MIN DO_SAVE_MIN( , mov r30=r0, )
++#endif
+diff -urNp linux-2.6/arch/ia64/xen/xenpal.S new/arch/ia64/xen/xenpal.S
+--- linux-2.6/arch/ia64/xen/xenpal.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xenpal.S 2006-05-09 12:32:40.000000000 +0200
+@@ -0,0 +1,76 @@
++/*
++ * ia64/xen/xenpal.S
++ *
++ * Alternate PAL routines for Xen. Heavily leveraged from
++ * ia64/kernel/pal.S
++ *
++ * Copyright (C) 2005 Hewlett-Packard Co
++ * Dan Magenheimer <dan.magenheimer@.hp.com>
++ */
++
++#include <asm/asmmacro.h>
++#include <asm/processor.h>
++
++GLOBAL_ENTRY(xen_pal_call_static)
++ .prologue ASM_UNW_PRLG_RP|ASM_UNW_PRLG_PFS, ASM_UNW_PRLG_GRSAVE(5)
++ alloc loc1 = ar.pfs,5,5,0,0
++#ifdef CONFIG_XEN
++ movl r22=running_on_xen;;
++ ld4 r22=[r22];;
++ cmp.eq p7,p0=r22,r0
++(p7) br.cond.spnt.many __ia64_pal_call_static;;
++#endif
++ movl loc2 = pal_entry_point
++1: {
++ mov r28 = in0
++ mov r29 = in1
++ mov r8 = ip
++ }
++ ;;
++ ld8 loc2 = [loc2] // loc2 <- entry point
++ tbit.nz p6,p7 = in4, 0
++ adds r8 = 1f-1b,r8
++ mov loc4=ar.rsc // save RSE configuration
++ ;;
++ mov ar.rsc=0 // put RSE in enforced lazy, LE mode
++ mov loc3 = psr
++ mov loc0 = rp
++ .body
++ mov r30 = in2
++
++#ifdef CONFIG_XEN
++ // this is low priority for paravirtualization, but is called
++ // from the idle loop so confuses privop counting
++ movl r31=XSI_PSR_IC
++ ;;
++(p6) st4 [r31]=r0
++ ;;
++(p7) adds r31=XSI_PSR_I_ADDR-XSI_PSR_IC,r31
++(p7) mov r22=1
++ ;;
++(p7) ld8 r31=[r31]
++ ;;
++(p7) st1 [r31]=r22
++ ;;
++ mov r31 = in3
++ mov b7 = loc2
++ ;;
++#else
++(p6) rsm psr.i | psr.ic
++ mov r31 = in3
++ mov b7 = loc2
++
++(p7) rsm psr.i
++ ;;
++(p6) srlz.i
++#endif
++ mov rp = r8
++ br.cond.sptk.many b7
++1: mov psr.l = loc3
++ mov ar.rsc = loc4 // restore RSE configuration
++ mov ar.pfs = loc1
++ mov rp = loc0
++ ;;
++ srlz.d // seralize restoration of psr.l
++ br.ret.sptk.many b0
++END(xen_pal_call_static)
+diff -urNp linux-2.6/arch/ia64/xen/xensetup.S new/arch/ia64/xen/xensetup.S
+--- linux-2.6/arch/ia64/xen/xensetup.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/ia64/xen/xensetup.S 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,24 @@
++/*
++ * Support routines for Xen
++ *
++ * Copyright (C) 2005 Dan Magenheimer <dan.magenheimer@hp.com>
++ */
++
++#include <linux/config.h>
++#include <asm/processor.h>
++#include <asm/asmmacro.h>
++
++#define isBP p3 // are we the Bootstrap Processor?
++
++ .text
++GLOBAL_ENTRY(early_xen_setup)
++ mov r8=ar.rsc // Initialized in head.S
++(isBP) movl r9=running_on_xen;;
++ extr.u r8=r8,2,2;; // Extract pl fields
++ cmp.ne p7,p0=r8,r0;; // p7: running on xen
++(p7) mov r8=1 // booleanize.
++(p7) movl r10=xen_ivt;;
++(isBP) st4 [r9]=r8
++(p7) mov cr.iva=r10
++ br.ret.sptk.many rp;;
++END(early_xen_setup)
+diff -urNp linux-2.6/arch/um/kernel/physmem.c new/arch/um/kernel/physmem.c
+--- linux-2.6/arch/um/kernel/physmem.c 2006-07-03 14:14:29.000000000 +0200
++++ new/arch/um/kernel/physmem.c 2006-05-09 12:33:14.000000000 +0200
+@@ -226,7 +226,7 @@ EXPORT_SYMBOL(physmem_forget_descriptor)
+ EXPORT_SYMBOL(physmem_remove_mapping);
+ EXPORT_SYMBOL(physmem_subst_mapping);
+
+-void arch_free_page(struct page *page, int order)
++int arch_free_page(struct page *page, int order)
+ {
+ void *virt;
+ int i;
+@@ -235,6 +235,8 @@ void arch_free_page(struct page *page, i
+ virt = __va(page_to_phys(page + i));
+ physmem_remove_mapping(virt);
+ }
++
++ return 0;
+ }
+
+ int is_remapped(void *virt)
+diff -urNp linux-2.6/arch/x86_64/ia32/ia32entry-xen.S new/arch/x86_64/ia32/ia32entry-xen.S
+--- linux-2.6/arch/x86_64/ia32/ia32entry-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/ia32/ia32entry-xen.S 2006-05-23 18:37:10.000000000 +0200
+@@ -0,0 +1,726 @@
++/*
++ * Compatibility mode system call entry point for x86-64.
++ *
++ * Copyright 2000-2002 Andi Kleen, SuSE Labs.
++ */
++
++#include <asm/dwarf2.h>
++#include <asm/calling.h>
++#include <asm/asm-offsets.h>
++#include <asm/current.h>
++#include <asm/errno.h>
++#include <asm/ia32_unistd.h>
++#include <asm/thread_info.h>
++#include <asm/segment.h>
++#include <asm/vsyscall32.h>
++#include <linux/linkage.h>
++
++#define __XEN_X86_64 1
++
++#define IA32_NR_syscalls ((ia32_syscall_end - ia32_sys_call_table)/8)
++
++ .macro IA32_ARG_FIXUP noebp=0
++ movl %edi,%r8d
++ .if \noebp
++ .else
++ movl %ebp,%r9d
++ .endif
++ xchg %ecx,%esi
++ movl %ebx,%edi
++ movl %edx,%edx /* zero extension */
++ .endm
++
++ /* clobbers %eax */
++ .macro CLEAR_RREGS
++ xorl %eax,%eax
++ movq %rax,R11(%rsp)
++ movq %rax,R10(%rsp)
++ movq %rax,R9(%rsp)
++ movq %rax,R8(%rsp)
++ .endm
++
++#if defined (__XEN_X86_64)
++#include "../kernel/xen_entry.S"
++
++#define __swapgs
++#define __cli
++#define __sti
++#else
++/*
++ * Use the native instructions
++ */
++#define __swapgs swapgs
++#define __cli cli
++#define __sti sti
++#endif
++
++ .macro CFI_STARTPROC32 simple
++ CFI_STARTPROC \simple
++ CFI_UNDEFINED r8
++ CFI_UNDEFINED r9
++ CFI_UNDEFINED r10
++ CFI_UNDEFINED r11
++ CFI_UNDEFINED r12
++ CFI_UNDEFINED r13
++ CFI_UNDEFINED r14
++ CFI_UNDEFINED r15
++ .endm
++
++/*
++ * 32bit SYSENTER instruction entry.
++ *
++ * Arguments:
++ * %eax System call number.
++ * %ebx Arg1
++ * %ecx Arg2
++ * %edx Arg3
++ * %esi Arg4
++ * %edi Arg5
++ * %ebp user stack
++ * 0(%ebp) Arg6
++ *
++ * Interrupts off.
++ *
++ * This is purely a fast path. For anything complicated we use the int 0x80
++ * path below. Set up a complete hardware stack frame to share code
++ * with the int 0x80 path.
++ */
++ENTRY(ia32_sysenter_target)
++ CFI_STARTPROC32 simple
++ CFI_DEF_CFA rsp,0
++ CFI_REGISTER rsp,rbp
++ __swapgs
++ movq %gs:pda_kernelstack, %rsp
++ addq $(PDA_STACKOFFSET),%rsp
++ XEN_UNBLOCK_EVENTS(%r11)
++ __sti
++ movl %ebp,%ebp /* zero extension */
++ pushq $__USER32_DS
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET ss,0*/
++ pushq %rbp
++ CFI_ADJUST_CFA_OFFSET 8
++ CFI_REL_OFFSET rsp,0
++ pushfq
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET rflags,0*/
++ movl $VSYSCALL32_SYSEXIT, %r10d
++ CFI_REGISTER rip,r10
++ pushq $__USER32_CS
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET cs,0*/
++ movl %eax, %eax
++ pushq %r10
++ CFI_ADJUST_CFA_OFFSET 8
++ CFI_REL_OFFSET rip,0
++ pushq %rax
++ CFI_ADJUST_CFA_OFFSET 8
++ cld
++ SAVE_ARGS 0,0,1
++ /* no need to do an access_ok check here because rbp has been
++ 32bit zero extended */
++1: movl (%rbp),%r9d
++ .section __ex_table,"a"
++ .quad 1b,ia32_badarg
++ .previous
++ GET_THREAD_INFO(%r10)
++ orl $TS_COMPAT,threadinfo_status(%r10)
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
++ CFI_REMEMBER_STATE
++ jnz sysenter_tracesys
++sysenter_do_call:
++ cmpl $(IA32_NR_syscalls-1),%eax
++ ja ia32_badsys
++ IA32_ARG_FIXUP 1
++ call *ia32_sys_call_table(,%rax,8)
++ movq %rax,RAX-ARGOFFSET(%rsp)
++ GET_THREAD_INFO(%r10)
++ XEN_BLOCK_EVENTS(%r11)
++ __cli
++ testl $_TIF_ALLWORK_MASK,threadinfo_flags(%r10)
++ jnz int_ret_from_sys_call
++ andl $~TS_COMPAT,threadinfo_status(%r10)
++ /* clear IF, that popfq doesn't enable interrupts early */
++ andl $~0x200,EFLAGS-R11(%rsp)
++ RESTORE_ARGS 1,24,1,1,1,1
++ popfq
++ CFI_ADJUST_CFA_OFFSET -8
++ /*CFI_RESTORE rflags*/
++ popq %rcx /* User %esp */
++ CFI_ADJUST_CFA_OFFSET -8
++ CFI_REGISTER rsp,rcx
++ movl $VSYSCALL32_SYSEXIT,%edx /* User %eip */
++ CFI_REGISTER rip,rdx
++ __swapgs
++ XEN_UNBLOCK_EVENTS(%r11)
++ __sti /* sti only takes effect after the next instruction */
++ /* sysexit */
++ .byte 0xf, 0x35 /* TBD */
++
++sysenter_tracesys:
++ CFI_RESTORE_STATE
++ SAVE_REST
++ CLEAR_RREGS
++ movq $-ENOSYS,RAX(%rsp) /* really needed? */
++ movq %rsp,%rdi /* &pt_regs -> arg1 */
++ call syscall_trace_enter
++ LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
++ RESTORE_REST
++ movl %ebp, %ebp
++ /* no need to do an access_ok check here because rbp has been
++ 32bit zero extended */
++1: movl (%rbp),%r9d
++ .section __ex_table,"a"
++ .quad 1b,ia32_badarg
++ .previous
++ jmp sysenter_do_call
++ CFI_ENDPROC
++
++/*
++ * 32bit SYSCALL instruction entry.
++ *
++ * Arguments:
++ * %eax System call number.
++ * %ebx Arg1
++ * %ecx return EIP
++ * %edx Arg3
++ * %esi Arg4
++ * %edi Arg5
++ * %ebp Arg2 [note: not saved in the stack frame, should not be touched]
++ * %esp user stack
++ * 0(%esp) Arg6
++ *
++ * Interrupts off.
++ *
++ * This is purely a fast path. For anything complicated we use the int 0x80
++ * path below. Set up a complete hardware stack frame to share code
++ * with the int 0x80 path.
++ */
++ENTRY(ia32_cstar_target)
++ CFI_STARTPROC32 simple
++ CFI_DEF_CFA rsp,0
++ CFI_REGISTER rip,rcx
++ /*CFI_REGISTER rflags,r11*/
++ __swapgs
++ movl %esp,%r8d
++ CFI_REGISTER rsp,r8
++ movq %gs:pda_kernelstack,%rsp
++ XEN_UNBLOCK_EVENTS(%r11)
++ __sti
++ SAVE_ARGS 8,1,1
++ movl %eax,%eax /* zero extension */
++ movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
++ movq %rcx,RIP-ARGOFFSET(%rsp)
++ CFI_REL_OFFSET rip,RIP-ARGOFFSET
++ movq %rbp,RCX-ARGOFFSET(%rsp) /* this lies slightly to ptrace */
++ movl %ebp,%ecx
++ movq $__USER32_CS,CS-ARGOFFSET(%rsp)
++ movq $__USER32_DS,SS-ARGOFFSET(%rsp)
++ movq %r11,EFLAGS-ARGOFFSET(%rsp)
++ /*CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/
++ movq %r8,RSP-ARGOFFSET(%rsp)
++ CFI_REL_OFFSET rsp,RSP-ARGOFFSET
++ /* no need to do an access_ok check here because r8 has been
++ 32bit zero extended */
++ /* hardware stack frame is complete now */
++1: movl (%r8),%r9d
++ .section __ex_table,"a"
++ .quad 1b,ia32_badarg
++ .previous
++ GET_THREAD_INFO(%r10)
++ orl $TS_COMPAT,threadinfo_status(%r10)
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
++ CFI_REMEMBER_STATE
++ jnz cstar_tracesys
++cstar_do_call:
++ cmpl $IA32_NR_syscalls-1,%eax
++ ja ia32_badsys
++ IA32_ARG_FIXUP 1
++ call *ia32_sys_call_table(,%rax,8)
++ movq %rax,RAX-ARGOFFSET(%rsp)
++ GET_THREAD_INFO(%r10)
++ XEN_BLOCK_EVENTS(%r11)
++ __cli
++ testl $_TIF_ALLWORK_MASK,threadinfo_flags(%r10)
++ jnz int_ret_from_sys_call
++ andl $~TS_COMPAT,threadinfo_status(%r10)
++ RESTORE_ARGS 1,-ARG_SKIP,1,1,1
++ movl RIP-ARGOFFSET(%rsp),%ecx
++ CFI_REGISTER rip,rcx
++ movl EFLAGS-ARGOFFSET(%rsp),%r11d
++ /*CFI_REGISTER rflags,r11*/
++ movl RSP-ARGOFFSET(%rsp),%esp
++ CFI_RESTORE rsp
++ __swapgs
++ sysretl /* TBD */
++
++cstar_tracesys:
++ CFI_RESTORE_STATE
++ SAVE_REST
++ CLEAR_RREGS
++ movq $-ENOSYS,RAX(%rsp) /* really needed? */
++ movq %rsp,%rdi /* &pt_regs -> arg1 */
++ call syscall_trace_enter
++ LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
++ RESTORE_REST
++ movl RSP-ARGOFFSET(%rsp), %r8d
++ /* no need to do an access_ok check here because r8 has been
++ 32bit zero extended */
++1: movl (%r8),%r9d
++ .section __ex_table,"a"
++ .quad 1b,ia32_badarg
++ .previous
++ jmp cstar_do_call
++
++ia32_badarg:
++ movq $-EFAULT,%rax
++ jmp ia32_sysret
++ CFI_ENDPROC
++
++/*
++ * Emulated IA32 system calls via int 0x80.
++ *
++ * Arguments:
++ * %eax System call number.
++ * %ebx Arg1
++ * %ecx Arg2
++ * %edx Arg3
++ * %esi Arg4
++ * %edi Arg5
++ * %ebp Arg6 [note: not saved in the stack frame, should not be touched]
++ *
++ * Notes:
++ * Uses the same stack frame as the x86-64 version.
++ * All registers except %eax must be saved (but ptrace may violate that)
++ * Arguments are zero extended. For system calls that want sign extension and
++ * take long arguments a wrapper is needed. Most calls can just be called
++ * directly.
++ * Assumes it is only called from user space and entered with interrupts off.
++ */
++
++ENTRY(ia32_syscall)
++ CFI_STARTPROC simple
++ CFI_DEF_CFA rsp,SS+8-RIP
++ /*CFI_REL_OFFSET ss,SS-RIP*/
++ CFI_REL_OFFSET rsp,RSP-RIP
++ /*CFI_REL_OFFSET rflags,EFLAGS-RIP*/
++ /*CFI_REL_OFFSET cs,CS-RIP*/
++ CFI_REL_OFFSET rip,RIP-RIP
++ __swapgs
++ XEN_UNBLOCK_EVENTS(%r11)
++ __sti
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x10,%rsp /* skip rcx and r11 */
++ movl %eax,%eax
++ pushq %rax
++ CFI_ADJUST_CFA_OFFSET 8
++ cld
++/* 1: jmp 1b */
++ /* note the registers are not zero extended to the sf.
++ this could be a problem. */
++ SAVE_ARGS 0,0,1
++ GET_THREAD_INFO(%r10)
++ orl $TS_COMPAT,threadinfo_status(%r10)
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%r10)
++ jnz ia32_tracesys
++ia32_do_syscall:
++ cmpl $(IA32_NR_syscalls-1),%eax
++ ja ia32_badsys
++ IA32_ARG_FIXUP
++ call *ia32_sys_call_table(,%rax,8) # xxx: rip relative
++ia32_sysret:
++ movq %rax,RAX-ARGOFFSET(%rsp)
++ jmp int_ret_from_sys_call
++
++ia32_tracesys:
++ SAVE_REST
++ movq $-ENOSYS,RAX(%rsp) /* really needed? */
++ movq %rsp,%rdi /* &pt_regs -> arg1 */
++ call syscall_trace_enter
++ LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
++ RESTORE_REST
++ jmp ia32_do_syscall
++
++ia32_badsys:
++ movq $0,ORIG_RAX-ARGOFFSET(%rsp)
++ movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
++ jmp int_ret_from_sys_call
++
++ni_syscall:
++ movq %rax,%rdi
++ jmp sys32_ni_syscall
++
++quiet_ni_syscall:
++ movq $-ENOSYS,%rax
++ ret
++ CFI_ENDPROC
++
++ .macro PTREGSCALL label, func, arg
++ .globl \label
++\label:
++ leaq \func(%rip),%rax
++ leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
++ jmp ia32_ptregs_common
++ .endm
++
++ CFI_STARTPROC32
++
++ PTREGSCALL stub32_rt_sigreturn, sys32_rt_sigreturn, %rdi
++ PTREGSCALL stub32_sigreturn, sys32_sigreturn, %rdi
++ PTREGSCALL stub32_sigaltstack, sys32_sigaltstack, %rdx
++ PTREGSCALL stub32_sigsuspend, sys32_sigsuspend, %rcx
++ PTREGSCALL stub32_execve, sys32_execve, %rcx
++ PTREGSCALL stub32_fork, sys_fork, %rdi
++ PTREGSCALL stub32_clone, sys32_clone, %rdx
++ PTREGSCALL stub32_vfork, sys_vfork, %rdi
++ PTREGSCALL stub32_iopl, sys_iopl, %rsi
++ PTREGSCALL stub32_rt_sigsuspend, sys_rt_sigsuspend, %rdx
++
++ENTRY(ia32_ptregs_common)
++ popq %r11
++ CFI_ENDPROC
++ CFI_STARTPROC32 simple
++ CFI_DEF_CFA rsp,SS+8-ARGOFFSET
++ CFI_REL_OFFSET rax,RAX-ARGOFFSET
++ CFI_REL_OFFSET rcx,RCX-ARGOFFSET
++ CFI_REL_OFFSET rdx,RDX-ARGOFFSET
++ CFI_REL_OFFSET rsi,RSI-ARGOFFSET
++ CFI_REL_OFFSET rdi,RDI-ARGOFFSET
++ CFI_REL_OFFSET rip,RIP-ARGOFFSET
++/* CFI_REL_OFFSET cs,CS-ARGOFFSET*/
++/* CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/
++ CFI_REL_OFFSET rsp,RSP-ARGOFFSET
++/* CFI_REL_OFFSET ss,SS-ARGOFFSET*/
++ SAVE_REST
++ call *%rax
++ RESTORE_REST
++ jmp ia32_sysret /* misbalances the return cache */
++ CFI_ENDPROC
++
++ .section .rodata,"a"
++ .align 8
++ .globl ia32_sys_call_table
++ia32_sys_call_table:
++ .quad sys_restart_syscall
++ .quad sys_exit
++ .quad stub32_fork
++ .quad sys_read
++ .quad sys_write
++ .quad compat_sys_open /* 5 */
++ .quad sys_close
++ .quad sys32_waitpid
++ .quad sys_creat
++ .quad sys_link
++ .quad sys_unlink /* 10 */
++ .quad stub32_execve
++ .quad sys_chdir
++ .quad compat_sys_time
++ .quad sys_mknod
++ .quad sys_chmod /* 15 */
++ .quad sys_lchown16
++ .quad quiet_ni_syscall /* old break syscall holder */
++ .quad sys_stat
++ .quad sys32_lseek
++ .quad sys_getpid /* 20 */
++ .quad compat_sys_mount /* mount */
++ .quad sys_oldumount /* old_umount */
++ .quad sys_setuid16
++ .quad sys_getuid16
++ .quad compat_sys_stime /* stime */ /* 25 */
++ .quad sys32_ptrace /* ptrace */
++ .quad sys_alarm
++ .quad sys_fstat /* (old)fstat */
++ .quad sys_pause
++ .quad compat_sys_utime /* 30 */
++ .quad quiet_ni_syscall /* old stty syscall holder */
++ .quad quiet_ni_syscall /* old gtty syscall holder */
++ .quad sys_access
++ .quad sys_nice
++ .quad quiet_ni_syscall /* 35 */ /* old ftime syscall holder */
++ .quad sys_sync
++ .quad sys32_kill
++ .quad sys_rename
++ .quad sys_mkdir
++ .quad sys_rmdir /* 40 */
++ .quad sys_dup
++ .quad sys32_pipe
++ .quad compat_sys_times
++ .quad quiet_ni_syscall /* old prof syscall holder */
++ .quad sys_brk /* 45 */
++ .quad sys_setgid16
++ .quad sys_getgid16
++ .quad sys_signal
++ .quad sys_geteuid16
++ .quad sys_getegid16 /* 50 */
++ .quad sys_acct
++ .quad sys_umount /* new_umount */
++ .quad quiet_ni_syscall /* old lock syscall holder */
++ .quad compat_sys_ioctl
++ .quad compat_sys_fcntl64 /* 55 */
++ .quad quiet_ni_syscall /* old mpx syscall holder */
++ .quad sys_setpgid
++ .quad quiet_ni_syscall /* old ulimit syscall holder */
++ .quad sys32_olduname
++ .quad sys_umask /* 60 */
++ .quad sys_chroot
++ .quad sys32_ustat
++ .quad sys_dup2
++ .quad sys_getppid
++ .quad sys_getpgrp /* 65 */
++ .quad sys_setsid
++ .quad sys32_sigaction
++ .quad sys_sgetmask
++ .quad sys_ssetmask
++ .quad sys_setreuid16 /* 70 */
++ .quad sys_setregid16
++ .quad stub32_sigsuspend
++ .quad compat_sys_sigpending
++ .quad sys_sethostname
++ .quad compat_sys_setrlimit /* 75 */
++ .quad compat_sys_old_getrlimit /* old_getrlimit */
++ .quad compat_sys_getrusage
++ .quad sys32_gettimeofday
++ .quad sys32_settimeofday
++ .quad sys_getgroups16 /* 80 */
++ .quad sys_setgroups16
++ .quad sys32_old_select
++ .quad sys_symlink
++ .quad sys_lstat
++ .quad sys_readlink /* 85 */
++#ifdef CONFIG_IA32_AOUT
++ .quad sys_uselib
++#else
++ .quad quiet_ni_syscall
++#endif
++ .quad sys_swapon
++ .quad sys_reboot
++ .quad compat_sys_old_readdir
++ .quad sys32_mmap /* 90 */
++ .quad sys_munmap
++ .quad sys_truncate
++ .quad sys_ftruncate
++ .quad sys_fchmod
++ .quad sys_fchown16 /* 95 */
++ .quad sys_getpriority
++ .quad sys_setpriority
++ .quad quiet_ni_syscall /* old profil syscall holder */
++ .quad compat_sys_statfs
++ .quad compat_sys_fstatfs /* 100 */
++ .quad sys_ioperm
++ .quad compat_sys_socketcall
++ .quad sys_syslog
++ .quad compat_sys_setitimer
++ .quad compat_sys_getitimer /* 105 */
++ .quad compat_sys_newstat
++ .quad compat_sys_newlstat
++ .quad compat_sys_newfstat
++ .quad sys32_uname
++ .quad stub32_iopl /* 110 */
++ .quad sys_vhangup
++ .quad quiet_ni_syscall /* old "idle" system call */
++ .quad sys32_vm86_warning /* vm86old */
++ .quad compat_sys_wait4
++ .quad sys_swapoff /* 115 */
++ .quad sys32_sysinfo
++ .quad sys32_ipc
++ .quad sys_fsync
++ .quad stub32_sigreturn
++ .quad stub32_clone /* 120 */
++ .quad sys_setdomainname
++ .quad sys_uname
++ .quad sys_modify_ldt
++ .quad compat_sys_adjtimex
++ .quad sys32_mprotect /* 125 */
++ .quad compat_sys_sigprocmask
++ .quad quiet_ni_syscall /* create_module */
++ .quad sys_init_module
++ .quad sys_delete_module
++ .quad quiet_ni_syscall /* 130 get_kernel_syms */
++ .quad sys_quotactl
++ .quad sys_getpgid
++ .quad sys_fchdir
++ .quad quiet_ni_syscall /* bdflush */
++ .quad sys_sysfs /* 135 */
++ .quad sys_personality
++ .quad quiet_ni_syscall /* for afs_syscall */
++ .quad sys_setfsuid16
++ .quad sys_setfsgid16
++ .quad sys_llseek /* 140 */
++ .quad compat_sys_getdents
++ .quad compat_sys_select
++ .quad sys_flock
++ .quad sys_msync
++ .quad compat_sys_readv /* 145 */
++ .quad compat_sys_writev
++ .quad sys_getsid
++ .quad sys_fdatasync
++ .quad sys32_sysctl /* sysctl */
++ .quad sys_mlock /* 150 */
++ .quad sys_munlock
++ .quad sys_mlockall
++ .quad sys_munlockall
++ .quad sys_sched_setparam
++ .quad sys_sched_getparam /* 155 */
++ .quad sys_sched_setscheduler
++ .quad sys_sched_getscheduler
++ .quad sys_sched_yield
++ .quad sys_sched_get_priority_max
++ .quad sys_sched_get_priority_min /* 160 */
++ .quad sys_sched_rr_get_interval
++ .quad compat_sys_nanosleep
++ .quad sys_mremap
++ .quad sys_setresuid16
++ .quad sys_getresuid16 /* 165 */
++ .quad sys32_vm86_warning /* vm86 */
++ .quad quiet_ni_syscall /* query_module */
++ .quad sys_poll
++ .quad compat_sys_nfsservctl
++ .quad sys_setresgid16 /* 170 */
++ .quad sys_getresgid16
++ .quad sys_prctl
++ .quad stub32_rt_sigreturn
++ .quad sys32_rt_sigaction
++ .quad sys32_rt_sigprocmask /* 175 */
++ .quad sys32_rt_sigpending
++ .quad compat_sys_rt_sigtimedwait
++ .quad sys32_rt_sigqueueinfo
++ .quad stub32_rt_sigsuspend
++ .quad sys32_pread /* 180 */
++ .quad sys32_pwrite
++ .quad sys_chown16
++ .quad sys_getcwd
++ .quad sys_capget
++ .quad sys_capset
++ .quad stub32_sigaltstack
++ .quad sys32_sendfile
++ .quad quiet_ni_syscall /* streams1 */
++ .quad quiet_ni_syscall /* streams2 */
++ .quad stub32_vfork /* 190 */
++ .quad compat_sys_getrlimit
++ .quad sys32_mmap2
++ .quad sys32_truncate64
++ .quad sys32_ftruncate64
++ .quad sys32_stat64 /* 195 */
++ .quad sys32_lstat64
++ .quad sys32_fstat64
++ .quad sys_lchown
++ .quad sys_getuid
++ .quad sys_getgid /* 200 */
++ .quad sys_geteuid
++ .quad sys_getegid
++ .quad sys_setreuid
++ .quad sys_setregid
++ .quad sys_getgroups /* 205 */
++ .quad sys_setgroups
++ .quad sys_fchown
++ .quad sys_setresuid
++ .quad sys_getresuid
++ .quad sys_setresgid /* 210 */
++ .quad sys_getresgid
++ .quad sys_chown
++ .quad sys_setuid
++ .quad sys_setgid
++ .quad sys_setfsuid /* 215 */
++ .quad sys_setfsgid
++ .quad sys_pivot_root
++ .quad sys_mincore
++ .quad sys_madvise
++ .quad compat_sys_getdents64 /* 220 getdents64 */
++ .quad compat_sys_fcntl64
++ .quad quiet_ni_syscall /* tux */
++ .quad quiet_ni_syscall /* security */
++ .quad sys_gettid
++ .quad sys_readahead /* 225 */
++ .quad sys_setxattr
++ .quad sys_lsetxattr
++ .quad sys_fsetxattr
++ .quad sys_getxattr
++ .quad sys_lgetxattr /* 230 */
++ .quad sys_fgetxattr
++ .quad sys_listxattr
++ .quad sys_llistxattr
++ .quad sys_flistxattr
++ .quad sys_removexattr /* 235 */
++ .quad sys_lremovexattr
++ .quad sys_fremovexattr
++ .quad sys_tkill
++ .quad sys_sendfile64
++ .quad compat_sys_futex /* 240 */
++ .quad compat_sys_sched_setaffinity
++ .quad compat_sys_sched_getaffinity
++ .quad sys32_set_thread_area
++ .quad sys32_get_thread_area
++ .quad compat_sys_io_setup /* 245 */
++ .quad sys_io_destroy
++ .quad compat_sys_io_getevents
++ .quad compat_sys_io_submit
++ .quad sys_io_cancel
++ .quad sys_fadvise64 /* 250 */
++ .quad quiet_ni_syscall /* free_huge_pages */
++ .quad sys_exit_group
++ .quad sys32_lookup_dcookie
++ .quad sys_epoll_create
++ .quad sys_epoll_ctl /* 255 */
++ .quad sys_epoll_wait
++ .quad sys_remap_file_pages
++ .quad sys_set_tid_address
++ .quad compat_sys_timer_create
++ .quad compat_sys_timer_settime /* 260 */
++ .quad compat_sys_timer_gettime
++ .quad sys_timer_getoverrun
++ .quad sys_timer_delete
++ .quad compat_sys_clock_settime
++ .quad compat_sys_clock_gettime /* 265 */
++ .quad compat_sys_clock_getres
++ .quad compat_sys_clock_nanosleep
++ .quad compat_sys_statfs64
++ .quad compat_sys_fstatfs64
++ .quad sys_tgkill /* 270 */
++ .quad compat_sys_utimes
++ .quad sys32_fadvise64_64
++ .quad quiet_ni_syscall /* sys_vserver */
++ .quad sys_mbind
++ .quad compat_sys_get_mempolicy /* 275 */
++ .quad sys_set_mempolicy
++ .quad compat_sys_mq_open
++ .quad sys_mq_unlink
++ .quad compat_sys_mq_timedsend
++ .quad compat_sys_mq_timedreceive /* 280 */
++ .quad compat_sys_mq_notify
++ .quad compat_sys_mq_getsetattr
++ .quad compat_sys_kexec_load /* reserved for kexec */
++ .quad compat_sys_waitid
++ .quad quiet_ni_syscall /* 285: sys_altroot */
++ .quad sys_add_key
++ .quad sys_request_key
++ .quad sys_keyctl
++ .quad sys_ioprio_set
++ .quad sys_ioprio_get /* 290 */
++ .quad sys_inotify_init
++ .quad sys_inotify_add_watch
++ .quad sys_inotify_rm_watch
++ .quad sys_migrate_pages
++ .quad compat_sys_openat /* 295 */
++ .quad sys_mkdirat
++ .quad sys_mknodat
++ .quad sys_fchownat
++ .quad compat_sys_futimesat
++ .quad sys32_fstatat /* 300 */
++ .quad sys_unlinkat
++ .quad sys_renameat
++ .quad sys_linkat
++ .quad sys_symlinkat
++ .quad sys_readlinkat /* 305 */
++ .quad sys_fchmodat
++ .quad sys_faccessat
++ .quad quiet_ni_syscall /* pselect6 for now */
++ .quad quiet_ni_syscall /* ppoll for now */
++ .quad sys_unshare /* 310 */
++ .quad compat_sys_set_robust_list
++ .quad compat_sys_get_robust_list
++ .quad sys_splice
++ .quad sys_sync_file_range
++ .quad sys_tee
++ .quad compat_sys_vmsplice
++ia32_syscall_end:
+diff -urNp linux-2.6/arch/x86_64/ia32/Makefile new/arch/x86_64/ia32/Makefile
+--- linux-2.6/arch/x86_64/ia32/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/ia32/Makefile 2006-05-23 18:37:10.000000000 +0200
+@@ -23,9 +23,25 @@ quiet_cmd_syscall = SYSCALL $@
+ -Wl,-soname=linux-gate.so.1 -o $@ \
+ -Wl,-T,$(filter-out FORCE,$^)
+
++$(obj)/vsyscall-int80.so \
+ $(obj)/vsyscall-sysenter.so $(obj)/vsyscall-syscall.so: \
+ $(obj)/vsyscall-%.so: $(src)/vsyscall.lds $(obj)/vsyscall-%.o FORCE
+ $(call if_changed,syscall)
+
+-AFLAGS_vsyscall-sysenter.o = -m32 -Wa,-32
+-AFLAGS_vsyscall-syscall.o = -m32 -Wa,-32
++AFLAGS_vsyscall-sysenter.o = -m32 -Wa,-32 -Iarch/i386/kernel
++AFLAGS_vsyscall-syscall.o = -m32 -Wa,-32 -Iarch/i386/kernel
++
++ifdef CONFIG_XEN
++AFLAGS_vsyscall-int80.o = -m32 -Wa,-32 -Iarch/i386/kernel
++CFLAGS_syscall32-xen.o += -DUSE_INT80
++AFLAGS_syscall32_syscall-xen.o += -DUSE_INT80
++
++$(obj)/syscall32_syscall-xen.o: \
++ $(foreach F,int80 sysenter syscall,$(obj)/vsyscall-$F.so)
++
++targets := $(foreach F,int80 sysenter syscall,vsyscall-$F.o vsyscall-$F.so)
++
++include $(srctree)/scripts/Makefile.xen
++
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/arch/x86_64/ia32/syscall32_syscall-xen.S new/arch/x86_64/ia32/syscall32_syscall-xen.S
+--- linux-2.6/arch/x86_64/ia32/syscall32_syscall-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/ia32/syscall32_syscall-xen.S 2006-05-09 12:33:17.000000000 +0200
+@@ -0,0 +1,28 @@
++/* 32bit VDSOs mapped into user space. */
++
++ .section ".init.data","aw"
++
++#ifdef USE_INT80
++
++ .globl syscall32_int80
++ .globl syscall32_int80_end
++
++syscall32_int80:
++ .incbin "arch/x86_64/ia32/vsyscall-int80.so"
++syscall32_int80_end:
++
++#endif
++
++ .globl syscall32_syscall
++ .globl syscall32_syscall_end
++
++syscall32_syscall:
++ .incbin "arch/x86_64/ia32/vsyscall-syscall.so"
++syscall32_syscall_end:
++
++ .globl syscall32_sysenter
++ .globl syscall32_sysenter_end
++
++syscall32_sysenter:
++ .incbin "arch/x86_64/ia32/vsyscall-sysenter.so"
++syscall32_sysenter_end:
+diff -urNp linux-2.6/arch/x86_64/ia32/syscall32-xen.c new/arch/x86_64/ia32/syscall32-xen.c
+--- linux-2.6/arch/x86_64/ia32/syscall32-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/ia32/syscall32-xen.c 2006-05-09 12:33:17.000000000 +0200
+@@ -0,0 +1,128 @@
++/* Copyright 2002,2003 Andi Kleen, SuSE Labs */
++
++/* vsyscall handling for 32bit processes. Map a stub page into it
++ on demand because 32bit cannot reach the kernel's fixmaps */
++
++#include <linux/mm.h>
++#include <linux/string.h>
++#include <linux/kernel.h>
++#include <linux/gfp.h>
++#include <linux/init.h>
++#include <linux/stringify.h>
++#include <linux/security.h>
++#include <asm/proto.h>
++#include <asm/tlbflush.h>
++#include <asm/ia32_unistd.h>
++
++#ifdef USE_INT80
++extern unsigned char syscall32_int80[], syscall32_int80_end[];
++#endif
++extern unsigned char syscall32_syscall[], syscall32_syscall_end[];
++extern unsigned char syscall32_sysenter[], syscall32_sysenter_end[];
++extern int sysctl_vsyscall32;
++
++char *syscall32_page;
++#ifndef USE_INT80
++static int use_sysenter = -1;
++#endif
++
++static struct page *
++syscall32_nopage(struct vm_area_struct *vma, unsigned long adr, int *type)
++{
++ struct page *p = virt_to_page(adr - vma->vm_start + syscall32_page);
++ get_page(p);
++ return p;
++}
++
++/* Prevent VMA merging */
++static void syscall32_vma_close(struct vm_area_struct *vma)
++{
++}
++
++static struct vm_operations_struct syscall32_vm_ops = {
++ .close = syscall32_vma_close,
++ .nopage = syscall32_nopage,
++};
++
++struct linux_binprm;
++
++/* Setup a VMA at program startup for the vsyscall page */
++int syscall32_setup_pages(struct linux_binprm *bprm, int exstack)
++{
++ int npages = (VSYSCALL32_END - VSYSCALL32_BASE) >> PAGE_SHIFT;
++ struct vm_area_struct *vma;
++ struct mm_struct *mm = current->mm;
++ int ret;
++
++ vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
++ if (!vma)
++ return -ENOMEM;
++
++ memset(vma, 0, sizeof(struct vm_area_struct));
++ /* Could randomize here */
++ vma->vm_start = VSYSCALL32_BASE;
++ vma->vm_end = VSYSCALL32_END;
++ /* MAYWRITE to allow gdb to COW and set breakpoints */
++ vma->vm_flags = VM_READ|VM_EXEC|VM_MAYREAD|VM_MAYEXEC|VM_MAYWRITE;
++ vma->vm_flags |= mm->def_flags;
++ vma->vm_page_prot = protection_map[vma->vm_flags & 7];
++ vma->vm_ops = &syscall32_vm_ops;
++ vma->vm_mm = mm;
++
++ down_write(&mm->mmap_sem);
++ if ((ret = insert_vm_struct(mm, vma))) {
++ up_write(&mm->mmap_sem);
++ kmem_cache_free(vm_area_cachep, vma);
++ return ret;
++ }
++ mm->total_vm += npages;
++ up_write(&mm->mmap_sem);
++ return 0;
++}
++
++static int __init init_syscall32(void)
++{
++ syscall32_page = (void *)get_zeroed_page(GFP_KERNEL);
++ if (!syscall32_page)
++ panic("Cannot allocate syscall32 page");
++
++#ifdef USE_INT80
++ /*
++ * At this point we use int 0x80.
++ */
++ memcpy(syscall32_page, syscall32_int80,
++ syscall32_int80_end - syscall32_int80);
++#else
++ if (use_sysenter > 0) {
++ memcpy(syscall32_page, syscall32_sysenter,
++ syscall32_sysenter_end - syscall32_sysenter);
++ } else {
++ memcpy(syscall32_page, syscall32_syscall,
++ syscall32_syscall_end - syscall32_syscall);
++ }
++#endif
++ return 0;
++}
++
++/*
++ * This must be done early in case we have an initrd containing 32-bit
++ * binaries (e.g., hotplug). This could be pushed upstream to arch/x86_64.
++ */
++core_initcall(init_syscall32);
++
++/* May not be __init: called during resume */
++void syscall32_cpu_init(void)
++{
++#ifndef USE_INT80
++ if (use_sysenter < 0)
++ use_sysenter = (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL);
++
++ /* Load these always in case some future AMD CPU supports
++ SYSENTER from compat mode too. */
++ checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS);
++ checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL);
++ checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target);
++
++ wrmsrl(MSR_CSTAR, ia32_cstar_target);
++#endif
++}
+diff -urNp linux-2.6/arch/x86_64/ia32/vsyscall-int80.S new/arch/x86_64/ia32/vsyscall-int80.S
+--- linux-2.6/arch/x86_64/ia32/vsyscall-int80.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/ia32/vsyscall-int80.S 2006-05-09 12:33:17.000000000 +0200
+@@ -0,0 +1,58 @@
++/*
++ * Code for the vsyscall page. This version uses the old int $0x80 method.
++ *
++ * NOTE:
++ * 1) __kernel_vsyscall _must_ be first in this page.
++ * 2) there are alignment constraints on this stub, see vsyscall-sigreturn.S
++ * for details.
++ */
++#include <asm/ia32_unistd.h>
++#include <asm/asm-offsets.h>
++
++ .code32
++ .text
++ .section .text.vsyscall,"ax"
++ .globl __kernel_vsyscall
++ .type __kernel_vsyscall,@function
++__kernel_vsyscall:
++.LSTART_vsyscall:
++ int $0x80
++ ret
++.LEND_vsyscall:
++ .size __kernel_vsyscall,.-.LSTART_vsyscall
++ .previous
++
++ .section .eh_frame,"a",@progbits
++.LSTARTFRAME:
++ .long .LENDCIE-.LSTARTCIE
++.LSTARTCIE:
++ .long 0 /* CIE ID */
++ .byte 1 /* Version number */
++ .string "zR" /* NUL-terminated augmentation string */
++ .uleb128 1 /* Code alignment factor */
++ .sleb128 -4 /* Data alignment factor */
++ .byte 8 /* Return address register column */
++ .uleb128 1 /* Augmentation value length */
++ .byte 0x1b /* DW_EH_PE_pcrel|DW_EH_PE_sdata4. */
++ .byte 0x0c /* DW_CFA_def_cfa */
++ .uleb128 4
++ .uleb128 4
++ .byte 0x88 /* DW_CFA_offset, column 0x8 */
++ .uleb128 1
++ .align 4
++.LENDCIE:
++
++ .long .LENDFDE1-.LSTARTFDE1 /* Length FDE */
++.LSTARTFDE1:
++ .long .LSTARTFDE1-.LSTARTFRAME /* CIE pointer */
++ .long .LSTART_vsyscall-. /* PC-relative start address */
++ .long .LEND_vsyscall-.LSTART_vsyscall
++ .uleb128 0 /* Augmentation length */
++ .align 4
++.LENDFDE1:
++
++/*
++ * Get the common code for the sigreturn entry points.
++ */
++#define SYSCALL_ENTER_KERNEL int $0x80
++#include "vsyscall-sigreturn.S"
+diff -urNp linux-2.6/arch/x86_64/ia32/vsyscall-sigreturn.S new/arch/x86_64/ia32/vsyscall-sigreturn.S
+--- linux-2.6/arch/x86_64/ia32/vsyscall-sigreturn.S 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/ia32/vsyscall-sigreturn.S 2006-05-09 12:33:17.000000000 +0200
+@@ -139,5 +139,5 @@ __kernel_rt_sigreturn:
+ .align 4
+ .LENDFDE3:
+
+-#include "../../i386/kernel/vsyscall-note.S"
++#include <vsyscall-note.S>
+
+diff -urNp linux-2.6/arch/x86_64/Kconfig new/arch/x86_64/Kconfig
+--- linux-2.6/arch/x86_64/Kconfig 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/Kconfig 2006-05-09 12:33:17.000000000 +0200
+@@ -123,6 +123,22 @@ config GENERIC_CPU
+
+ endchoice
+
++config X86_64_XEN
++ bool "Enable Xen compatible kernel"
++ select SWIOTLB
++ help
++ This option will compile a kernel compatible with Xen hypervisor
++
++config X86_NO_TSS
++ bool
++ depends on X86_64_XEN
++ default y
++
++config X86_NO_IDT
++ bool
++ depends on X86_64_XEN
++ default y
++
+ #
+ # Define implied options from the CPU selection here
+ #
+@@ -143,6 +159,7 @@ config X86_INTERNODE_CACHE_BYTES
+
+ config X86_TSC
+ bool
++ depends on !X86_64_XEN
+ default y
+
+ config X86_GOOD_APIC
+@@ -185,7 +202,7 @@ config X86_CPUID
+
+ config X86_HT
+ bool
+- depends on SMP && !MK8
++ depends on SMP && !MK8 && !X86_64_XEN
+ default y
+
+ config MATH_EMULATION
+@@ -199,14 +216,22 @@ config EISA
+
+ config X86_IO_APIC
+ bool
++ depends !XEN_UNPRIVILEGED_GUEST
+ default y
+
++config X86_XEN_GENAPIC
++ bool
++ depends X86_64_XEN
++ default XEN_PRIVILEGED_GUEST || SMP
++
+ config X86_LOCAL_APIC
+ bool
++ depends !XEN_UNPRIVILEGED_GUEST
+ default y
+
+ config MTRR
+ bool "MTRR (Memory Type Range Register) support"
++ depends on !XEN_UNPRIVILEGED_GUEST
+ ---help---
+ On Intel P6 family processors (Pentium Pro, Pentium II and later)
+ the Memory Type Range Registers (MTRRs) may be used to control
+@@ -247,7 +272,7 @@ config SMP
+
+ config SCHED_SMT
+ bool "SMT (Hyperthreading) scheduler support"
+- depends on SMP
++ depends on SMP && !X86_64_XEN
+ default n
+ help
+ SMT scheduler support improves the CPU scheduler's decision making
+@@ -257,7 +282,7 @@ config SCHED_SMT
+
+ config SCHED_MC
+ bool "Multi-core scheduler support"
+- depends on SMP
++ depends on SMP && !X86_64_XEN
+ default y
+ help
+ Multi-core scheduler support improves the CPU scheduler's decision
+@@ -268,7 +293,7 @@ source "kernel/Kconfig.preempt"
+
+ config NUMA
+ bool "Non Uniform Memory Access (NUMA) Support"
+- depends on SMP
++ depends on SMP && !X86_64_XEN
+ help
+ Enable NUMA (Non Uniform Memory Access) support. The kernel
+ will try to allocate memory used by a CPU on the local memory
+@@ -352,6 +377,7 @@ config NR_CPUS
+ int "Maximum number of CPUs (2-256)"
+ range 2 255
+ depends on SMP
++ default "16" if X86_64_XEN
+ default "8"
+ help
+ This allows you to specify the maximum number of CPUs which this
+@@ -372,6 +398,7 @@ config HOTPLUG_CPU
+
+ config HPET_TIMER
+ bool
++ depends on !X86_64_XEN
+ default y
+ help
+ Use the IA-PC HPET (High Precision Event Timer) to manage
+@@ -389,7 +416,7 @@ config GART_IOMMU
+ bool "K8 GART IOMMU support"
+ default y
+ select SWIOTLB
+- depends on PCI
++ depends on PCI && !X86_64_XEN
+ help
+ Support for hardware IOMMU in AMD's Opteron/Athlon64 Processors
+ and for the bounce buffering software IOMMU.
+@@ -409,6 +436,7 @@ config SWIOTLB
+
+ config X86_MCE
+ bool "Machine check support" if EMBEDDED
++ depends on !X86_64_XEN
+ default y
+ help
+ Include a machine check error handler to report hardware errors.
+@@ -434,7 +462,7 @@ config X86_MCE_AMD
+
+ config KEXEC
+ bool "kexec system call (EXPERIMENTAL)"
+- depends on EXPERIMENTAL
++ depends on EXPERIMENTAL && !X86_64_XEN
+ help
+ kexec is a system call that implements the ability to shutdown your
+ current kernel, and to start another kernel. It is like a reboot
+@@ -525,8 +553,11 @@ config GENERIC_PENDING_IRQ
+ default y
+
+ menu "Power management options"
++ depends on !XEN_UNPRIVILEGED_GUEST
+
++if !X86_64_XEN
+ source kernel/power/Kconfig
++endif
+
+ source "drivers/acpi/Kconfig"
+
+@@ -549,6 +580,21 @@ config PCI_MMCONFIG
+ bool "Support mmconfig PCI config space access"
+ depends on PCI && ACPI
+
++config XEN_PCIDEV_FRONTEND
++ bool "Xen PCI Frontend"
++ depends on PCI && X86_64_XEN
++ default y
++ help
++ The PCI device frontend driver allows the kernel to import arbitrary
++ PCI devices from a PCI backend to support PCI driver domains.
++
++config XEN_PCIDEV_FE_DEBUG
++ bool "Xen PCI Frontend Debugging"
++ depends on XEN_PCIDEV_FRONTEND
++ default n
++ help
++ Enables some debug statements within the PCI Frontend.
++
+ source "drivers/pci/pcie/Kconfig"
+
+ source "drivers/pci/Kconfig"
+@@ -619,4 +665,6 @@ source "security/Kconfig"
+
+ source "crypto/Kconfig"
+
++source "drivers/xen/Kconfig"
++
+ source "lib/Kconfig"
+diff -urNp linux-2.6/arch/x86_64/kernel/acpi/Makefile new/arch/x86_64/kernel/acpi/Makefile
+--- linux-2.6/arch/x86_64/kernel/acpi/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/acpi/Makefile 2006-05-09 12:33:17.000000000 +0200
+@@ -6,3 +6,4 @@ ifneq ($(CONFIG_ACPI_PROCESSOR),)
+ obj-y += processor.o
+ endif
+
++boot-$(CONFIG_XEN) := ../../../i386/kernel/acpi/boot-xen.o
+diff -urNp linux-2.6/arch/x86_64/kernel/apic-xen.c new/arch/x86_64/kernel/apic-xen.c
+--- linux-2.6/arch/x86_64/kernel/apic-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/apic-xen.c 2006-05-09 12:33:17.000000000 +0200
+@@ -0,0 +1,198 @@
++/*
++ * Local APIC handling, local APIC timers
++ *
++ * (c) 1999, 2000 Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
++ * thanks to Eric Gilmore
++ * and Rolf G. Tews
++ * for testing these extensively.
++ * Maciej W. Rozycki : Various updates and fixes.
++ * Mikael Pettersson : Power Management for UP-APIC.
++ * Pavel Machek and
++ * Mikael Pettersson : PM converted to driver model.
++ */
++
++#include <linux/config.h>
++#include <linux/init.h>
++
++#include <linux/mm.h>
++#include <linux/delay.h>
++#include <linux/bootmem.h>
++#include <linux/smp_lock.h>
++#include <linux/interrupt.h>
++#include <linux/mc146818rtc.h>
++#include <linux/kernel_stat.h>
++#include <linux/sysdev.h>
++#include <linux/module.h>
++
++#include <asm/atomic.h>
++#include <asm/smp.h>
++#include <asm/mtrr.h>
++#include <asm/mpspec.h>
++#include <asm/desc.h>
++#include <asm/arch_hooks.h>
++#include <asm/hpet.h>
++#include <asm/idle.h>
++
++int apic_verbosity;
++
++/*
++ * 'what should we do if we get a hw irq event on an illegal vector'.
++ * each architecture has to answer this themselves.
++ */
++void ack_bad_irq(unsigned int irq)
++{
++ printk("unexpected IRQ trap at vector %02x\n", irq);
++ /*
++ * Currently unexpected vectors happen only on SMP and APIC.
++ * We _must_ ack these because every local APIC has only N
++ * irq slots per priority level, and a 'hanging, unacked' IRQ
++ * holds up an irq slot - in excessive cases (when multiple
++ * unexpected vectors occur) that might lock up the APIC
++ * completely.
++ * But don't ack when the APIC is disabled. -AK
++ */
++ if (!disable_apic)
++ ack_APIC_irq();
++}
++
++int setup_profiling_timer(unsigned int multiplier)
++{
++ return -EINVAL;
++}
++
++void smp_local_timer_interrupt(struct pt_regs *regs)
++{
++ profile_tick(CPU_PROFILING, regs);
++#ifndef CONFIG_XEN
++#ifdef CONFIG_SMP
++ update_process_times(user_mode(regs));
++#endif
++#endif
++ /*
++ * We take the 'long' return path, and there every subsystem
++ * grabs the appropriate locks (kernel lock/ irq lock).
++ *
++ * we might want to decouple profiling from the 'long path',
++ * and do the profiling totally in assembly.
++ *
++ * Currently this isn't too much of an issue (performance wise),
++ * we can take more than 100K local irqs per second on a 100 MHz P5.
++ */
++}
++
++/*
++ * Local APIC timer interrupt. This is the most natural way for doing
++ * local interrupts, but local timer interrupts can be emulated by
++ * broadcast interrupts too. [in case the hw doesn't support APIC timers]
++ *
++ * [ if a single-CPU system runs an SMP kernel then we call the local
++ * interrupt as well. Thus we cannot inline the local irq ... ]
++ */
++void smp_apic_timer_interrupt(struct pt_regs *regs)
++{
++ /*
++ * the NMI deadlock-detector uses this.
++ */
++ add_pda(apic_timer_irqs, 1);
++
++ /*
++ * NOTE! We'd better ACK the irq immediately,
++ * because timer handling can be slow.
++ */
++ ack_APIC_irq();
++ /*
++ * update_process_times() expects us to have done irq_enter().
++ * Besides, if we don't timer interrupts ignore the global
++ * interrupt lock, which is the WrongThing (tm) to do.
++ */
++ exit_idle();
++ irq_enter();
++ smp_local_timer_interrupt(regs);
++ irq_exit();
++}
++
++/*
++ * This interrupt should _never_ happen with our APIC/SMP architecture
++ */
++asmlinkage void smp_spurious_interrupt(void)
++{
++ unsigned int v;
++ exit_idle();
++ irq_enter();
++ /*
++ * Check if this really is a spurious interrupt and ACK it
++ * if it is a vectored one. Just in case...
++ * Spurious interrupts should not be ACKed.
++ */
++ v = apic_read(APIC_ISR + ((SPURIOUS_APIC_VECTOR & ~0x1f) >> 1));
++ if (v & (1 << (SPURIOUS_APIC_VECTOR & 0x1f)))
++ ack_APIC_irq();
++
++#if 0
++ static unsigned long last_warning;
++ static unsigned long skipped;
++
++ /* see sw-dev-man vol 3, chapter 7.4.13.5 */
++ if (time_before(last_warning+30*HZ,jiffies)) {
++ printk(KERN_INFO "spurious APIC interrupt on CPU#%d, %ld skipped.\n",
++ smp_processor_id(), skipped);
++ last_warning = jiffies;
++ skipped = 0;
++ } else {
++ skipped++;
++ }
++#endif
++ irq_exit();
++}
++
++/*
++ * This interrupt should never happen with our APIC/SMP architecture
++ */
++
++asmlinkage void smp_error_interrupt(void)
++{
++ unsigned int v, v1;
++
++ exit_idle();
++ irq_enter();
++ /* First tickle the hardware, only then report what went on. -- REW */
++ v = apic_read(APIC_ESR);
++ apic_write(APIC_ESR, 0);
++ v1 = apic_read(APIC_ESR);
++ ack_APIC_irq();
++ atomic_inc(&irq_err_count);
++
++ /* Here is what the APIC error bits mean:
++ 0: Send CS error
++ 1: Receive CS error
++ 2: Send accept error
++ 3: Receive accept error
++ 4: Reserved
++ 5: Send illegal vector
++ 6: Received illegal vector
++ 7: Illegal register address
++ */
++ printk (KERN_DEBUG "APIC error on CPU%d: %02x(%02x)\n",
++ smp_processor_id(), v , v1);
++ irq_exit();
++}
++
++int disable_apic;
++
++/*
++ * This initializes the IO-APIC and APIC hardware if this is
++ * a UP kernel.
++ */
++int __init APIC_init_uniprocessor (void)
++{
++#ifdef CONFIG_X86_IO_APIC
++ if (smp_found_config)
++ if (!skip_ioapic_setup && nr_ioapics)
++ setup_IO_APIC();
++#endif
++
++ return 1;
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/asm-offsets.c new/arch/x86_64/kernel/asm-offsets.c
+--- linux-2.6/arch/x86_64/kernel/asm-offsets.c 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/asm-offsets.c 2006-05-09 12:33:18.000000000 +0200
+@@ -66,7 +66,9 @@ int main(void)
+ DEFINE(pbe_address, offsetof(struct pbe, address));
+ DEFINE(pbe_orig_address, offsetof(struct pbe, orig_address));
+ DEFINE(pbe_next, offsetof(struct pbe, next));
++#ifndef CONFIG_X86_NO_TSS
+ BLANK();
+ DEFINE(TSS_ist, offsetof(struct tss_struct, ist));
++#endif
+ return 0;
+ }
+diff -urNp linux-2.6/arch/x86_64/kernel/e820-xen.c new/arch/x86_64/kernel/e820-xen.c
+--- linux-2.6/arch/x86_64/kernel/e820-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/e820-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,753 @@
++/*
++ * Handle the memory map.
++ * The functions here do the job until bootmem takes over.
++ * $Id$
++ *
++ * Getting sanitize_e820_map() in sync with i386 version by applying change:
++ * - Provisions for empty E820 memory regions (reported by certain BIOSes).
++ * Alex Achenbach <xela@slit.de>, December 2002.
++ * Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
++ *
++ */
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/types.h>
++#include <linux/init.h>
++#include <linux/bootmem.h>
++#include <linux/ioport.h>
++#include <linux/string.h>
++#include <linux/kexec.h>
++#include <linux/module.h>
++
++#include <asm/page.h>
++#include <asm/e820.h>
++#include <asm/proto.h>
++#include <asm/bootsetup.h>
++#include <asm/sections.h>
++#include <xen/interface/memory.h>
++
++/*
++ * PFN of last memory page.
++ */
++unsigned long end_pfn;
++EXPORT_SYMBOL(end_pfn);
++
++/*
++ * end_pfn only includes RAM, while end_pfn_map includes all e820 entries.
++ * The direct mapping extends to end_pfn_map, so that we can directly access
++ * apertures, ACPI and other tables without having to play with fixmaps.
++ */
++unsigned long end_pfn_map;
++
++/*
++ * Last pfn which the user wants to use.
++ */
++unsigned long end_user_pfn = MAXMEM>>PAGE_SHIFT;
++
++extern struct resource code_resource, data_resource;
++
++/* Check for some hardcoded bad areas that early boot is not allowed to touch */
++static inline int bad_addr(unsigned long *addrp, unsigned long size)
++{
++ unsigned long addr = *addrp, last = addr + size;
++
++#ifndef CONFIG_XEN
++ /* various gunk below that needed for SMP startup */
++ if (addr < 0x8000) {
++ *addrp = 0x8000;
++ return 1;
++ }
++
++ /* direct mapping tables of the kernel */
++ if (last >= table_start<<PAGE_SHIFT && addr < table_end<<PAGE_SHIFT) {
++ *addrp = table_end << PAGE_SHIFT;
++ return 1;
++ }
++
++ /* initrd */
++#ifdef CONFIG_BLK_DEV_INITRD
++ if (LOADER_TYPE && INITRD_START && last >= INITRD_START &&
++ addr < INITRD_START+INITRD_SIZE) {
++ *addrp = INITRD_START + INITRD_SIZE;
++ return 1;
++ }
++#endif
++ /* kernel code + 640k memory hole (later should not be needed, but
++ be paranoid for now) */
++ if (last >= 640*1024 && addr < __pa_symbol(&_end)) {
++ *addrp = __pa_symbol(&_end);
++ return 1;
++ }
++
++ if (last >= ebda_addr && addr < ebda_addr + ebda_size) {
++ *addrp = ebda_addr + ebda_size;
++ return 1;
++ }
++
++ /* XXX ramdisk image here? */
++#else
++ if (last < (table_end<<PAGE_SHIFT)) {
++ *addrp = table_end << PAGE_SHIFT;
++ return 1;
++ }
++#endif
++ return 0;
++}
++
++#ifndef CONFIG_XEN
++/*
++ * This function checks if any part of the range <start,end> is mapped
++ * with type.
++ */
++int __meminit
++e820_any_mapped(unsigned long start, unsigned long end, unsigned type)
++{
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ if (type && ei->type != type)
++ continue;
++ if (ei->addr >= end || ei->addr + ei->size <= start)
++ continue;
++ return 1;
++ }
++ return 0;
++}
++#endif
++
++/*
++ * This function checks if the entire range <start,end> is mapped with type.
++ *
++ * Note: this function only works correct if the e820 table is sorted and
++ * not-overlapping, which is the case
++ */
++int __init e820_all_mapped(unsigned long start, unsigned long end, unsigned type)
++{
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ if (type && ei->type != type)
++ continue;
++ /* is the region (part) in overlap with the current region ?*/
++ if (ei->addr >= end || ei->addr + ei->size <= start)
++ continue;
++
++ /* if the region is at the beginning of <start,end> we move
++ * start to the end of the region since it's ok until there
++ */
++ if (ei->addr <= start)
++ start = ei->addr + ei->size;
++ /* if start is now at or beyond end, we're done, full coverage */
++ if (start >= end)
++ return 1; /* we're done */
++ }
++ return 0;
++}
++
++/*
++ * Find a free area in a specific range.
++ */
++unsigned long __init find_e820_area(unsigned long start, unsigned long end, unsigned size)
++{
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ unsigned long addr = ei->addr, last;
++ if (ei->type != E820_RAM)
++ continue;
++ if (addr < start)
++ addr = start;
++ if (addr > ei->addr + ei->size)
++ continue;
++ while (bad_addr(&addr, size) && addr+size < ei->addr + ei->size)
++ ;
++ last = addr + size;
++ if (last > ei->addr + ei->size)
++ continue;
++ if (last > end)
++ continue;
++ return addr;
++ }
++ return -1UL;
++}
++
++/*
++ * Free bootmem based on the e820 table for a node.
++ */
++void __init e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end)
++{
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ unsigned long last, addr;
++
++ if (ei->type != E820_RAM ||
++ ei->addr+ei->size <= start ||
++ ei->addr >= end)
++ continue;
++
++ addr = round_up(ei->addr, PAGE_SIZE);
++ if (addr < start)
++ addr = start;
++
++ last = round_down(ei->addr + ei->size, PAGE_SIZE);
++ if (last >= end)
++ last = end;
++
++ if (last > addr && last-addr >= PAGE_SIZE)
++ free_bootmem_node(pgdat, addr, last-addr);
++ }
++}
++
++/*
++ * Find the highest page frame number we have available
++ */
++unsigned long __init e820_end_of_ram(void)
++{
++ int i;
++ unsigned long end_pfn = 0;
++
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ unsigned long start, end;
++
++ start = round_up(ei->addr, PAGE_SIZE);
++ end = round_down(ei->addr + ei->size, PAGE_SIZE);
++ if (start >= end)
++ continue;
++ if (ei->type == E820_RAM) {
++ if (end > end_pfn<<PAGE_SHIFT)
++ end_pfn = end>>PAGE_SHIFT;
++ } else {
++ if (end > end_pfn_map<<PAGE_SHIFT)
++ end_pfn_map = end>>PAGE_SHIFT;
++ }
++ }
++
++ if (end_pfn > end_pfn_map)
++ end_pfn_map = end_pfn;
++ if (end_pfn_map > MAXMEM>>PAGE_SHIFT)
++ end_pfn_map = MAXMEM>>PAGE_SHIFT;
++ if (end_pfn > end_user_pfn)
++ end_pfn = end_user_pfn;
++ if (end_pfn > end_pfn_map)
++ end_pfn = end_pfn_map;
++
++ return end_pfn;
++}
++
++/*
++ * Compute how much memory is missing in a range.
++ * Unlike the other functions in this file the arguments are in page numbers.
++ */
++unsigned long __init
++e820_hole_size(unsigned long start_pfn, unsigned long end_pfn)
++{
++ unsigned long ram = 0;
++ unsigned long start = start_pfn << PAGE_SHIFT;
++ unsigned long end = end_pfn << PAGE_SHIFT;
++ int i;
++ for (i = 0; i < e820.nr_map; i++) {
++ struct e820entry *ei = &e820.map[i];
++ unsigned long last, addr;
++
++ if (ei->type != E820_RAM ||
++ ei->addr+ei->size <= start ||
++ ei->addr >= end)
++ continue;
++
++ addr = round_up(ei->addr, PAGE_SIZE);
++ if (addr < start)
++ addr = start;
++
++ last = round_down(ei->addr + ei->size, PAGE_SIZE);
++ if (last >= end)
++ last = end;
++
++ if (last > addr)
++ ram += last - addr;
++ }
++ return ((end - start) - ram) >> PAGE_SHIFT;
++}
++
++/*
++ * Mark e820 reserved areas as busy for the resource manager.
++ */
++void __init e820_reserve_resources(struct e820entry *e820, int nr_map)
++{
++ int i;
++ for (i = 0; i < nr_map; i++) {
++ struct resource *res;
++ res = alloc_bootmem_low(sizeof(struct resource));
++ switch (e820[i].type) {
++ case E820_RAM: res->name = "System RAM"; break;
++ case E820_ACPI: res->name = "ACPI Tables"; break;
++ case E820_NVS: res->name = "ACPI Non-volatile Storage"; break;
++ default: res->name = "reserved";
++ }
++ res->start = e820[i].addr;
++ res->end = res->start + e820[i].size - 1;
++ res->flags = IORESOURCE_MEM | IORESOURCE_BUSY;
++ request_resource(&iomem_resource, res);
++ if (e820[i].type == E820_RAM) {
++ /*
++ * We don't know which RAM region contains kernel data,
++ * so we try it repeatedly and let the resource manager
++ * test it.
++ */
++ request_resource(res, &code_resource);
++ request_resource(res, &data_resource);
++#ifdef CONFIG_KEXEC
++ request_resource(res, &crashk_res);
++#endif
++ }
++ }
++}
++
++/*
++ * Add a memory region to the kernel e820 map.
++ */
++void __init add_memory_region(unsigned long start, unsigned long size, int type)
++{
++ int x = e820.nr_map;
++
++ if (x == E820MAX) {
++ printk(KERN_ERR "Ooops! Too many entries in the memory map!\n");
++ return;
++ }
++
++ e820.map[x].addr = start;
++ e820.map[x].size = size;
++ e820.map[x].type = type;
++ e820.nr_map++;
++}
++
++void __init e820_print_map(char *who)
++{
++ int i;
++
++ for (i = 0; i < e820.nr_map; i++) {
++ printk(" %s: %016Lx - %016Lx ", who,
++ (unsigned long long) e820.map[i].addr,
++ (unsigned long long) (e820.map[i].addr + e820.map[i].size));
++ switch (e820.map[i].type) {
++ case E820_RAM: printk("(usable)\n");
++ break;
++ case E820_RESERVED:
++ printk("(reserved)\n");
++ break;
++ case E820_ACPI:
++ printk("(ACPI data)\n");
++ break;
++ case E820_NVS:
++ printk("(ACPI NVS)\n");
++ break;
++ default: printk("type %u\n", e820.map[i].type);
++ break;
++ }
++ }
++}
++
++/*
++ * Sanitize the BIOS e820 map.
++ *
++ * Some e820 responses include overlapping entries. The following
++ * replaces the original e820 map with a new one, removing overlaps.
++ *
++ */
++static int __init sanitize_e820_map(struct e820entry * biosmap, char * pnr_map)
++{
++ struct change_member {
++ struct e820entry *pbios; /* pointer to original bios entry */
++ unsigned long long addr; /* address for this change point */
++ };
++ static struct change_member change_point_list[2*E820MAX] __initdata;
++ static struct change_member *change_point[2*E820MAX] __initdata;
++ static struct e820entry *overlap_list[E820MAX] __initdata;
++ static struct e820entry new_bios[E820MAX] __initdata;
++ struct change_member *change_tmp;
++ unsigned long current_type, last_type;
++ unsigned long long last_addr;
++ int chgidx, still_changing;
++ int overlap_entries;
++ int new_bios_entry;
++ int old_nr, new_nr, chg_nr;
++ int i;
++
++ /*
++ Visually we're performing the following (1,2,3,4 = memory types)...
++
++ Sample memory map (w/overlaps):
++ ____22__________________
++ ______________________4_
++ ____1111________________
++ _44_____________________
++ 11111111________________
++ ____________________33__
++ ___________44___________
++ __________33333_________
++ ______________22________
++ ___________________2222_
++ _________111111111______
++ _____________________11_
++ _________________4______
++
++ Sanitized equivalent (no overlap):
++ 1_______________________
++ _44_____________________
++ ___1____________________
++ ____22__________________
++ ______11________________
++ _________1______________
++ __________3_____________
++ ___________44___________
++ _____________33_________
++ _______________2________
++ ________________1_______
++ _________________4______
++ ___________________2____
++ ____________________33__
++ ______________________4_
++ */
++
++ /* if there's only one memory region, don't bother */
++ if (*pnr_map < 2)
++ return -1;
++
++ old_nr = *pnr_map;
++
++ /* bail out if we find any unreasonable addresses in bios map */
++ for (i=0; i<old_nr; i++)
++ if (biosmap[i].addr + biosmap[i].size < biosmap[i].addr)
++ return -1;
++
++ /* create pointers for initial change-point information (for sorting) */
++ for (i=0; i < 2*old_nr; i++)
++ change_point[i] = &change_point_list[i];
++
++ /* record all known change-points (starting and ending addresses),
++ omitting those that are for empty memory regions */
++ chgidx = 0;
++ for (i=0; i < old_nr; i++) {
++ if (biosmap[i].size != 0) {
++ change_point[chgidx]->addr = biosmap[i].addr;
++ change_point[chgidx++]->pbios = &biosmap[i];
++ change_point[chgidx]->addr = biosmap[i].addr + biosmap[i].size;
++ change_point[chgidx++]->pbios = &biosmap[i];
++ }
++ }
++ chg_nr = chgidx;
++
++ /* sort change-point list by memory addresses (low -> high) */
++ still_changing = 1;
++ while (still_changing) {
++ still_changing = 0;
++ for (i=1; i < chg_nr; i++) {
++ /* if <current_addr> > <last_addr>, swap */
++ /* or, if current=<start_addr> & last=<end_addr>, swap */
++ if ((change_point[i]->addr < change_point[i-1]->addr) ||
++ ((change_point[i]->addr == change_point[i-1]->addr) &&
++ (change_point[i]->addr == change_point[i]->pbios->addr) &&
++ (change_point[i-1]->addr != change_point[i-1]->pbios->addr))
++ )
++ {
++ change_tmp = change_point[i];
++ change_point[i] = change_point[i-1];
++ change_point[i-1] = change_tmp;
++ still_changing=1;
++ }
++ }
++ }
++
++ /* create a new bios memory map, removing overlaps */
++ overlap_entries=0; /* number of entries in the overlap table */
++ new_bios_entry=0; /* index for creating new bios map entries */
++ last_type = 0; /* start with undefined memory type */
++ last_addr = 0; /* start with 0 as last starting address */
++ /* loop through change-points, determining affect on the new bios map */
++ for (chgidx=0; chgidx < chg_nr; chgidx++)
++ {
++ /* keep track of all overlapping bios entries */
++ if (change_point[chgidx]->addr == change_point[chgidx]->pbios->addr)
++ {
++ /* add map entry to overlap list (> 1 entry implies an overlap) */
++ overlap_list[overlap_entries++]=change_point[chgidx]->pbios;
++ }
++ else
++ {
++ /* remove entry from list (order independent, so swap with last) */
++ for (i=0; i<overlap_entries; i++)
++ {
++ if (overlap_list[i] == change_point[chgidx]->pbios)
++ overlap_list[i] = overlap_list[overlap_entries-1];
++ }
++ overlap_entries--;
++ }
++ /* if there are overlapping entries, decide which "type" to use */
++ /* (larger value takes precedence -- 1=usable, 2,3,4,4+=unusable) */
++ current_type = 0;
++ for (i=0; i<overlap_entries; i++)
++ if (overlap_list[i]->type > current_type)
++ current_type = overlap_list[i]->type;
++ /* continue building up new bios map based on this information */
++ if (current_type != last_type) {
++ if (last_type != 0) {
++ new_bios[new_bios_entry].size =
++ change_point[chgidx]->addr - last_addr;
++ /* move forward only if the new size was non-zero */
++ if (new_bios[new_bios_entry].size != 0)
++ if (++new_bios_entry >= E820MAX)
++ break; /* no more space left for new bios entries */
++ }
++ if (current_type != 0) {
++ new_bios[new_bios_entry].addr = change_point[chgidx]->addr;
++ new_bios[new_bios_entry].type = current_type;
++ last_addr=change_point[chgidx]->addr;
++ }
++ last_type = current_type;
++ }
++ }
++ new_nr = new_bios_entry; /* retain count for new bios entries */
++
++ /* copy new bios mapping into original location */
++ memcpy(biosmap, new_bios, new_nr*sizeof(struct e820entry));
++ *pnr_map = new_nr;
++
++ return 0;
++}
++
++/*
++ * Copy the BIOS e820 map into a safe place.
++ *
++ * Sanity-check it while we're at it..
++ *
++ * If we're lucky and live on a modern system, the setup code
++ * will have given us a memory map that we can use to properly
++ * set up memory. If we aren't, we'll fake a memory map.
++ *
++ * We check to see that the memory map contains at least 2 elements
++ * before we'll use it, because the detection code in setup.S may
++ * not be perfect and most every PC known to man has two memory
++ * regions: one from 0 to 640k, and one from 1mb up. (The IBM
++ * thinkpad 560x, for example, does not cooperate with the memory
++ * detection code.)
++ */
++static int __init copy_e820_map(struct e820entry * biosmap, int nr_map)
++{
++#ifndef CONFIG_XEN
++ /* Only one memory region (or negative)? Ignore it */
++ if (nr_map < 2)
++ return -1;
++#else
++ BUG_ON(nr_map < 1);
++#endif
++
++ do {
++ unsigned long start = biosmap->addr;
++ unsigned long size = biosmap->size;
++ unsigned long end = start + size;
++ unsigned long type = biosmap->type;
++
++ /* Overflow in 64 bits? Ignore the memory map. */
++ if (start > end)
++ return -1;
++
++#ifndef CONFIG_XEN
++ /*
++ * Some BIOSes claim RAM in the 640k - 1M region.
++ * Not right. Fix it up.
++ *
++ * This should be removed on Hammer which is supposed to not
++ * have non e820 covered ISA mappings there, but I had some strange
++ * problems so it stays for now. -AK
++ */
++ if (type == E820_RAM) {
++ if (start < 0x100000ULL && end > 0xA0000ULL) {
++ if (start < 0xA0000ULL)
++ add_memory_region(start, 0xA0000ULL-start, type);
++ if (end <= 0x100000ULL)
++ continue;
++ start = 0x100000ULL;
++ size = end - start;
++ }
++ }
++#endif
++
++ add_memory_region(start, size, type);
++ } while (biosmap++,--nr_map);
++ return 0;
++}
++
++#ifndef CONFIG_XEN
++void __init setup_memory_region(void)
++{
++ char *who = "BIOS-e820";
++
++ /*
++ * Try to copy the BIOS-supplied E820-map.
++ *
++ * Otherwise fake a memory map; one section from 0k->640k,
++ * the next section from 1mb->appropriate_mem_k
++ */
++ sanitize_e820_map(E820_MAP, &E820_MAP_NR);
++ if (copy_e820_map(E820_MAP, E820_MAP_NR) < 0) {
++ unsigned long mem_size;
++
++ /* compare results from other methods and take the greater */
++ if (ALT_MEM_K < EXT_MEM_K) {
++ mem_size = EXT_MEM_K;
++ who = "BIOS-88";
++ } else {
++ mem_size = ALT_MEM_K;
++ who = "BIOS-e801";
++ }
++
++ e820.nr_map = 0;
++ add_memory_region(0, LOWMEMSIZE(), E820_RAM);
++ add_memory_region(HIGH_MEMORY, mem_size << 10, E820_RAM);
++ }
++ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
++ e820_print_map(who);
++}
++
++#else /* CONFIG_XEN */
++
++void __init setup_memory_region(void)
++{
++ int rc;
++ struct xen_memory_map memmap;
++ /*
++ * This is rather large for a stack variable but this early in
++ * the boot process we know we have plenty slack space.
++ */
++ struct e820entry map[E820MAX];
++
++ memmap.nr_entries = E820MAX;
++ set_xen_guest_handle(memmap.buffer, map);
++
++ rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
++ if ( rc == -ENOSYS ) {
++ memmap.nr_entries = 1;
++ map[0].addr = 0ULL;
++ map[0].size = xen_start_info->nr_pages << PAGE_SHIFT;
++ /* 8MB slack (to balance backend allocations). */
++ map[0].size += 8 << 20;
++ map[0].type = E820_RAM;
++ rc = 0;
++ }
++ BUG_ON(rc);
++
++ sanitize_e820_map(map, (char *)&memmap.nr_entries);
++
++ BUG_ON(copy_e820_map(map, (char)memmap.nr_entries) < 0);
++
++ printk(KERN_INFO "BIOS-provided physical RAM map:\n");
++ e820_print_map("Xen");
++}
++#endif
++
++void __init parse_memopt(char *p, char **from)
++{
++ int i;
++ unsigned long current_end;
++ unsigned long end;
++
++ end_user_pfn = memparse(p, from);
++ end_user_pfn >>= PAGE_SHIFT;
++
++ end = end_user_pfn<<PAGE_SHIFT;
++ i = e820.nr_map-1;
++ current_end = e820.map[i].addr + e820.map[i].size;
++
++ if (current_end < end) {
++ /*
++ * The e820 map ends before our requested size so
++ * extend the final entry to the requested address.
++ */
++ if (e820.map[i].type == E820_RAM)
++ e820.map[i].size = end - e820.map[i].addr;
++ else
++ add_memory_region(current_end, end - current_end, E820_RAM);
++ }
++}
++
++void __init parse_memmapopt(char *p, char **from)
++{
++ unsigned long long start_at, mem_size;
++
++ mem_size = memparse(p, from);
++ p = *from;
++ if (*p == '@') {
++ start_at = memparse(p+1, from);
++ add_memory_region(start_at, mem_size, E820_RAM);
++ } else if (*p == '#') {
++ start_at = memparse(p+1, from);
++ add_memory_region(start_at, mem_size, E820_ACPI);
++ } else if (*p == '$') {
++ start_at = memparse(p+1, from);
++ add_memory_region(start_at, mem_size, E820_RESERVED);
++ } else {
++ end_user_pfn = (mem_size >> PAGE_SHIFT);
++ }
++ p = *from;
++}
++
++unsigned long pci_mem_start = 0xaeedbabe;
++
++/*
++ * Search for the biggest gap in the low 32 bits of the e820
++ * memory space. We pass this space to PCI to assign MMIO resources
++ * for hotplug or unconfigured devices in.
++ * Hopefully the BIOS let enough space left.
++ */
++__init void e820_setup_gap(struct e820entry *e820, int nr_map)
++{
++ unsigned long gapstart, gapsize, round;
++ unsigned long last;
++ int i;
++ int found = 0;
++
++ last = 0x100000000ull;
++ gapstart = 0x10000000;
++ gapsize = 0x400000;
++ i = nr_map;
++ while (--i >= 0) {
++ unsigned long long start = e820[i].addr;
++ unsigned long long end = start + e820[i].size;
++
++ /*
++ * Since "last" is at most 4GB, we know we'll
++ * fit in 32 bits if this condition is true
++ */
++ if (last > end) {
++ unsigned long gap = last - end;
++
++ if (gap > gapsize) {
++ gapsize = gap;
++ gapstart = end;
++ found = 1;
++ }
++ }
++ if (start < last)
++ last = start;
++ }
++
++ if (!found) {
++ gapstart = (end_pfn << PAGE_SHIFT) + 1024*1024;
++ printk(KERN_ERR "PCI: Warning: Cannot find a gap in the 32bit address range\n"
++ KERN_ERR "PCI: Unassigned devices with 32bit resource registers may break!\n");
++ }
++
++ /*
++ * See how much we want to round up: start off with
++ * rounding to the next 1MB area.
++ */
++ round = 0x100000;
++ while ((gapsize >> 4) > round)
++ round += round;
++ /* Fun with two's complement */
++ pci_mem_start = (gapstart + round) & -round;
++
++ printk(KERN_INFO "Allocating PCI resources starting at %lx (gap: %lx:%lx)\n",
++ pci_mem_start, gapstart, gapsize);
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/early_printk-xen.c new/arch/x86_64/kernel/early_printk-xen.c
+--- linux-2.6/arch/x86_64/kernel/early_printk-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/early_printk-xen.c 2006-05-09 12:33:18.000000000 +0200
+@@ -0,0 +1,304 @@
++#include <linux/config.h>
++#include <linux/console.h>
++#include <linux/kernel.h>
++#include <linux/init.h>
++#include <linux/string.h>
++#include <linux/tty.h>
++#include <asm/io.h>
++#include <asm/processor.h>
++#include <asm/fcntl.h>
++
++/* Simple VGA output */
++
++#ifdef __i386__
++#include <asm/setup.h>
++#define VGABASE (__ISA_IO_base + 0xb8000)
++#else
++#include <asm/bootsetup.h>
++#define VGABASE ((void __iomem *)0xffffffff800b8000UL)
++#endif
++
++static int max_ypos = 25, max_xpos = 80;
++static int current_ypos = 25, current_xpos = 0;
++
++#ifndef CONFIG_XEN
++static void early_vga_write(struct console *con, const char *str, unsigned n)
++{
++ char c;
++ int i, k, j;
++
++ while ((c = *str++) != '\0' && n-- > 0) {
++ if (current_ypos >= max_ypos) {
++ /* scroll 1 line up */
++ for (k = 1, j = 0; k < max_ypos; k++, j++) {
++ for (i = 0; i < max_xpos; i++) {
++ writew(readw(VGABASE+2*(max_xpos*k+i)),
++ VGABASE + 2*(max_xpos*j + i));
++ }
++ }
++ for (i = 0; i < max_xpos; i++)
++ writew(0x720, VGABASE + 2*(max_xpos*j + i));
++ current_ypos = max_ypos-1;
++ }
++ if (c == '\n') {
++ current_xpos = 0;
++ current_ypos++;
++ } else if (c != '\r') {
++ writew(((0x7 << 8) | (unsigned short) c),
++ VGABASE + 2*(max_xpos*current_ypos +
++ current_xpos++));
++ if (current_xpos >= max_xpos) {
++ current_xpos = 0;
++ current_ypos++;
++ }
++ }
++ }
++}
++
++static struct console early_vga_console = {
++ .name = "earlyvga",
++ .write = early_vga_write,
++ .flags = CON_PRINTBUFFER,
++ .index = -1,
++};
++
++/* Serial functions loosely based on a similar package from Klaus P. Gerlicher */
++
++static int early_serial_base = 0x3f8; /* ttyS0 */
++
++#define XMTRDY 0x20
++
++#define DLAB 0x80
++
++#define TXR 0 /* Transmit register (WRITE) */
++#define RXR 0 /* Receive register (READ) */
++#define IER 1 /* Interrupt Enable */
++#define IIR 2 /* Interrupt ID */
++#define FCR 2 /* FIFO control */
++#define LCR 3 /* Line control */
++#define MCR 4 /* Modem control */
++#define LSR 5 /* Line Status */
++#define MSR 6 /* Modem Status */
++#define DLL 0 /* Divisor Latch Low */
++#define DLH 1 /* Divisor latch High */
++
++static int early_serial_putc(unsigned char ch)
++{
++ unsigned timeout = 0xffff;
++ while ((inb(early_serial_base + LSR) & XMTRDY) == 0 && --timeout)
++ cpu_relax();
++ outb(ch, early_serial_base + TXR);
++ return timeout ? 0 : -1;
++}
++
++static void early_serial_write(struct console *con, const char *s, unsigned n)
++{
++ while (*s && n-- > 0) {
++ early_serial_putc(*s);
++ if (*s == '\n')
++ early_serial_putc('\r');
++ s++;
++ }
++}
++
++#define DEFAULT_BAUD 9600
++
++static __init void early_serial_init(char *s)
++{
++ unsigned char c;
++ unsigned divisor;
++ unsigned baud = DEFAULT_BAUD;
++ char *e;
++
++ if (*s == ',')
++ ++s;
++
++ if (*s) {
++ unsigned port;
++ if (!strncmp(s,"0x",2)) {
++ early_serial_base = simple_strtoul(s, &e, 16);
++ } else {
++ static int bases[] = { 0x3f8, 0x2f8 };
++
++ if (!strncmp(s,"ttyS",4))
++ s += 4;
++ port = simple_strtoul(s, &e, 10);
++ if (port > 1 || s == e)
++ port = 0;
++ early_serial_base = bases[port];
++ }
++ s += strcspn(s, ",");
++ if (*s == ',')
++ s++;
++ }
++
++ outb(0x3, early_serial_base + LCR); /* 8n1 */
++ outb(0, early_serial_base + IER); /* no interrupt */
++ outb(0, early_serial_base + FCR); /* no fifo */
++ outb(0x3, early_serial_base + MCR); /* DTR + RTS */
++
++ if (*s) {
++ baud = simple_strtoul(s, &e, 0);
++ if (baud == 0 || s == e)
++ baud = DEFAULT_BAUD;
++ }
++
++ divisor = 115200 / baud;
++ c = inb(early_serial_base + LCR);
++ outb(c | DLAB, early_serial_base + LCR);
++ outb(divisor & 0xff, early_serial_base + DLL);
++ outb((divisor >> 8) & 0xff, early_serial_base + DLH);
++ outb(c & ~DLAB, early_serial_base + LCR);
++}
++
++#else /* CONFIG_XEN */
++
++#undef SCREEN_INFO
++#define SCREEN_INFO screen_info
++extern struct screen_info screen_info;
++
++static void
++early_serial_write(struct console *con, const char *s, unsigned count)
++{
++ int n;
++
++ while (count > 0) {
++ n = HYPERVISOR_console_io(CONSOLEIO_write, count, (char *)s);
++ if (n <= 0)
++ break;
++ count -= n;
++ s += n;
++ }
++}
++
++static __init void early_serial_init(char *s)
++{
++ current_xpos = 0;
++}
++
++/*
++ * No early VGA console on Xen, as we do not have convenient ISA-space
++ * mappings. Someone should fix this for domain 0. For now, use fake serial.
++ */
++#define early_vga_console early_serial_console
++
++#endif
++
++static struct console early_serial_console = {
++ .name = "earlyser",
++ .write = early_serial_write,
++ .flags = CON_PRINTBUFFER,
++ .index = -1,
++};
++
++/* Console interface to a host file on AMD's SimNow! */
++
++static int simnow_fd;
++
++enum {
++ MAGIC1 = 0xBACCD00A,
++ MAGIC2 = 0xCA110000,
++ XOPEN = 5,
++ XWRITE = 4,
++};
++
++static noinline long simnow(long cmd, long a, long b, long c)
++{
++ long ret;
++ asm volatile("cpuid" :
++ "=a" (ret) :
++ "b" (a), "c" (b), "d" (c), "0" (MAGIC1), "D" (cmd + MAGIC2));
++ return ret;
++}
++
++void __init simnow_init(char *str)
++{
++ char *fn = "klog";
++ if (*str == '=')
++ fn = ++str;
++ /* error ignored */
++ simnow_fd = simnow(XOPEN, (unsigned long)fn, O_WRONLY|O_APPEND|O_CREAT, 0644);
++}
++
++static void simnow_write(struct console *con, const char *s, unsigned n)
++{
++ simnow(XWRITE, simnow_fd, (unsigned long)s, n);
++}
++
++static struct console simnow_console = {
++ .name = "simnow",
++ .write = simnow_write,
++ .flags = CON_PRINTBUFFER,
++ .index = -1,
++};
++
++/* Direct interface for emergencies */
++struct console *early_console = &early_vga_console;
++static int early_console_initialized = 0;
++
++void early_printk(const char *fmt, ...)
++{
++ char buf[512];
++ int n;
++ va_list ap;
++
++ va_start(ap,fmt);
++ n = vscnprintf(buf,512,fmt,ap);
++ early_console->write(early_console,buf,n);
++ va_end(ap);
++}
++
++static int __initdata keep_early;
++
++int __init setup_early_printk(char *opt)
++{
++ char *space;
++ char buf[256];
++
++ if (early_console_initialized)
++ return 1;
++
++ strlcpy(buf,opt,sizeof(buf));
++ space = strchr(buf, ' ');
++ if (space)
++ *space = 0;
++
++ if (strstr(buf,"keep"))
++ keep_early = 1;
++
++ if (!strncmp(buf, "serial", 6)) {
++ early_serial_init(buf + 6);
++ early_console = &early_serial_console;
++ } else if (!strncmp(buf, "ttyS", 4)) {
++ early_serial_init(buf);
++ early_console = &early_serial_console;
++ } else if (!strncmp(buf, "vga", 3)
++ && SCREEN_INFO.orig_video_isVGA == 1) {
++ max_xpos = SCREEN_INFO.orig_video_cols;
++ max_ypos = SCREEN_INFO.orig_video_lines;
++ current_ypos = SCREEN_INFO.orig_y;
++ early_console = &early_vga_console;
++ } else if (!strncmp(buf, "simnow", 6)) {
++ simnow_init(buf + 6);
++ early_console = &simnow_console;
++ keep_early = 1;
++ }
++ early_console_initialized = 1;
++ register_console(early_console);
++ return 0;
++}
++
++void __init disable_early_printk(void)
++{
++ if (!early_console_initialized || !early_console)
++ return;
++ if (!keep_early) {
++ printk("disabling early console\n");
++ unregister_console(early_console);
++ early_console_initialized = 0;
++ } else {
++ printk("keeping early console\n");
++ }
++}
++
++__setup("earlyprintk=", setup_early_printk);
+diff -urNp linux-2.6/arch/x86_64/kernel/entry.S new/arch/x86_64/kernel/entry.S
+--- linux-2.6/arch/x86_64/kernel/entry.S 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/entry.S 2006-06-05 15:54:32.000000000 +0200
+@@ -596,7 +596,7 @@ retint_kernel:
+ */
+ .macro apicinterrupt num,func
+ INTR_FRAME
+- pushq $\num-256
++ pushq $~(\num)
+ CFI_ADJUST_CFA_OFFSET 8
+ interrupt \func
+ jmp ret_from_intr
+diff -urNp linux-2.6/arch/x86_64/kernel/entry-xen.S new/arch/x86_64/kernel/entry-xen.S
+--- linux-2.6/arch/x86_64/kernel/entry-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/entry-xen.S 2006-05-09 12:33:18.000000000 +0200
+@@ -0,0 +1,1141 @@
++/*
++ * linux/arch/x86_64/entry.S
++ *
++ * Copyright (C) 1991, 1992 Linus Torvalds
++ * Copyright (C) 2000, 2001, 2002 Andi Kleen SuSE Labs
++ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
++ *
++ * $Id$
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Asit Mallick <asit.k.mallick@intel.com>
++ * Modified for Xen
++ */
++
++/*
++ * entry.S contains the system-call and fault low-level handling routines.
++ *
++ * NOTE: This code handles signal-recognition, which happens every time
++ * after an interrupt and after each system call.
++ *
++ * Normal syscalls and interrupts don't save a full stack frame, this is
++ * only done for syscall tracing, signals or fork/exec et.al.
++ *
++ * A note on terminology:
++ * - top of stack: Architecture defined interrupt frame from SS to RIP
++ * at the top of the kernel process stack.
++ * - partial stack frame: partially saved registers upto R11.
++ * - full stack frame: Like partial stack frame, but all register saved.
++ *
++ * TODO:
++ * - schedule it carefully for the final hardware.
++ */
++
++#define ASSEMBLY 1
++#include <linux/config.h>
++#ifdef CONFIG_DEBUG_INFO
++#undef CONFIG_DEBUG_INFO
++#endif
++#include <linux/linkage.h>
++#include <asm/segment.h>
++#include <asm/smp.h>
++#include <asm/cache.h>
++#include <asm/errno.h>
++#include <asm/dwarf2.h>
++#include <asm/calling.h>
++#include <asm/asm-offsets.h>
++#include <asm/msr.h>
++#include <asm/unistd.h>
++#include <asm/thread_info.h>
++#include <asm/hw_irq.h>
++#include <asm/page.h>
++#include <asm/errno.h>
++#include <xen/interface/arch-x86_64.h>
++#include <xen/interface/features.h>
++
++#include "irq_vectors.h"
++
++#include "xen_entry.S"
++
++ .code64
++
++#ifndef CONFIG_PREEMPT
++#define retint_kernel retint_restore_args
++#endif
++
++NMI_MASK = 0x80000000
++
++/*
++ * C code is not supposed to know about undefined top of stack. Every time
++ * a C function with an pt_regs argument is called from the SYSCALL based
++ * fast path FIXUP_TOP_OF_STACK is needed.
++ * RESTORE_TOP_OF_STACK syncs the syscall state after any possible ptregs
++ * manipulation.
++ */
++
++ /* %rsp:at FRAMEEND */
++ .macro FIXUP_TOP_OF_STACK tmp
++ movq $__USER_CS,CS(%rsp)
++ movq $-1,RCX(%rsp)
++ .endm
++
++ .macro RESTORE_TOP_OF_STACK tmp,offset=0
++ .endm
++
++ .macro FAKE_STACK_FRAME child_rip
++ /* push in order ss, rsp, eflags, cs, rip */
++ xorl %eax, %eax
++ pushq %rax /* ss */
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET ss,0*/
++ pushq %rax /* rsp */
++ CFI_ADJUST_CFA_OFFSET 8
++ CFI_REL_OFFSET rsp,0
++ pushq $(1<<9) /* eflags - interrupts on */
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET rflags,0*/
++ pushq $__KERNEL_CS /* cs */
++ CFI_ADJUST_CFA_OFFSET 8
++ /*CFI_REL_OFFSET cs,0*/
++ pushq \child_rip /* rip */
++ CFI_ADJUST_CFA_OFFSET 8
++ CFI_REL_OFFSET rip,0
++ pushq %rax /* orig rax */
++ CFI_ADJUST_CFA_OFFSET 8
++ .endm
++
++ .macro UNFAKE_STACK_FRAME
++ addq $8*6, %rsp
++ CFI_ADJUST_CFA_OFFSET -(6*8)
++ .endm
++
++ .macro CFI_DEFAULT_STACK start=1
++ .if \start
++ CFI_STARTPROC simple
++ CFI_DEF_CFA rsp,SS+8
++ .else
++ CFI_DEF_CFA_OFFSET SS+8
++ .endif
++ CFI_REL_OFFSET r15,R15
++ CFI_REL_OFFSET r14,R14
++ CFI_REL_OFFSET r13,R13
++ CFI_REL_OFFSET r12,R12
++ CFI_REL_OFFSET rbp,RBP
++ CFI_REL_OFFSET rbx,RBX
++ CFI_REL_OFFSET r11,R11
++ CFI_REL_OFFSET r10,R10
++ CFI_REL_OFFSET r9,R9
++ CFI_REL_OFFSET r8,R8
++ CFI_REL_OFFSET rax,RAX
++ CFI_REL_OFFSET rcx,RCX
++ CFI_REL_OFFSET rdx,RDX
++ CFI_REL_OFFSET rsi,RSI
++ CFI_REL_OFFSET rdi,RDI
++ CFI_REL_OFFSET rip,RIP
++ /*CFI_REL_OFFSET cs,CS*/
++ /*CFI_REL_OFFSET rflags,EFLAGS*/
++ CFI_REL_OFFSET rsp,RSP
++ /*CFI_REL_OFFSET ss,SS*/
++ .endm
++
++ /*
++ * Must be consistent with the definition in arch-x86_64.h:
++ * struct iret_context {
++ * u64 rax, r11, rcx, flags, rip, cs, rflags, rsp, ss;
++ * };
++ * #define VGCF_IN_SYSCALL (1<<8)
++ */
++ .macro HYPERVISOR_IRET flag
++ testb $3,1*8(%rsp)
++ jnz 2f
++ testl $NMI_MASK,2*8(%rsp)
++ jnz 2f
++
++ testb $1,(xen_features+XENFEAT_supervisor_mode_kernel)
++ jnz 1f
++
++ /* Direct iret to kernel space. Correct CS and SS. */
++ orb $3,1*8(%rsp)
++ orb $3,4*8(%rsp)
++1: iretq
++
++2: /* Slow iret via hypervisor. */
++ andl $~NMI_MASK, 16(%rsp)
++ pushq $\flag
++ jmp hypercall_page + (__HYPERVISOR_iret * 32)
++ .endm
++
++ .macro SWITCH_TO_KERNEL ssoff,adjust=0
++ jc 1f
++ orb $1,\ssoff-\adjust+4(%rsp)
++1:
++ .endm
++
++/*
++ * A newly forked process directly context switches into this.
++ */
++/* rdi: prev */
++ENTRY(ret_from_fork)
++ CFI_DEFAULT_STACK
++ call schedule_tail
++ GET_THREAD_INFO(%rcx)
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%rcx)
++ jnz rff_trace
++rff_action:
++ RESTORE_REST
++ testl $3,CS-ARGOFFSET(%rsp) # from kernel_thread?
++ je int_ret_from_sys_call
++ testl $_TIF_IA32,threadinfo_flags(%rcx)
++ jnz int_ret_from_sys_call
++ RESTORE_TOP_OF_STACK %rdi,ARGOFFSET
++ jmp ret_from_sys_call
++rff_trace:
++ movq %rsp,%rdi
++ call syscall_trace_leave
++ GET_THREAD_INFO(%rcx)
++ jmp rff_action
++ CFI_ENDPROC
++
++/*
++ * System call entry. Upto 6 arguments in registers are supported.
++ *
++ * SYSCALL does not save anything on the stack and does not change the
++ * stack pointer.
++ */
++
++/*
++ * Register setup:
++ * rax system call number
++ * rdi arg0
++ * rcx return address for syscall/sysret, C arg3
++ * rsi arg1
++ * rdx arg2
++ * r10 arg3 (--> moved to rcx for C)
++ * r8 arg4
++ * r9 arg5
++ * r11 eflags for syscall/sysret, temporary for C
++ * r12-r15,rbp,rbx saved by C code, not touched.
++ *
++ * Interrupts are off on entry.
++ * Only called from user space.
++ *
++ * XXX if we had a free scratch register we could save the RSP into the stack frame
++ * and report it properly in ps. Unfortunately we haven't.
++ *
++ * When user can change the frames always force IRET. That is because
++ * it deals with uncanonical addresses better. SYSRET has trouble
++ * with them due to bugs in both AMD and Intel CPUs.
++ */
++
++ENTRY(system_call)
++ CFI_STARTPROC simple
++ CFI_DEF_CFA rsp,0
++ CFI_REGISTER rip,rcx
++ /*CFI_REGISTER rflags,r11*/
++ SAVE_ARGS -8,0
++ movq %rax,ORIG_RAX-ARGOFFSET(%rsp)
++ XEN_UNBLOCK_EVENTS(%r11)
++ GET_THREAD_INFO(%rcx)
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SECCOMP),threadinfo_flags(%rcx)
++ CFI_REMEMBER_STATE
++ jnz tracesys
++ cmpq $__NR_syscall_max,%rax
++ ja badsys
++ movq %r10,%rcx
++ call *sys_call_table(,%rax,8) # XXX: rip relative
++ movq %rax,RAX-ARGOFFSET(%rsp)
++/*
++ * Syscall return path ending with SYSRET (fast path)
++ * Has incomplete stack frame and undefined top of stack.
++ */
++ .globl ret_from_sys_call
++ret_from_sys_call:
++ movl $_TIF_ALLWORK_MASK,%edi
++ /* edi: flagmask */
++sysret_check:
++ GET_THREAD_INFO(%rcx)
++ XEN_BLOCK_EVENTS(%rsi)
++ movl threadinfo_flags(%rcx),%edx
++ andl %edi,%edx
++ CFI_REMEMBER_STATE
++ jnz sysret_careful
++ XEN_UNBLOCK_EVENTS(%rsi)
++ CFI_REGISTER rip,rcx
++ RESTORE_ARGS 0,8,0
++ /*CFI_REGISTER rflags,r11*/
++ HYPERVISOR_IRET VGCF_IN_SYSCALL
++
++ /* Handle reschedules */
++ /* edx: work, edi: workmask */
++sysret_careful:
++ CFI_RESTORE_STATE
++ bt $TIF_NEED_RESCHED,%edx
++ jnc sysret_signal
++ XEN_BLOCK_EVENTS(%rsi)
++ pushq %rdi
++ CFI_ADJUST_CFA_OFFSET 8
++ call schedule
++ popq %rdi
++ CFI_ADJUST_CFA_OFFSET -8
++ jmp sysret_check
++
++ /* Handle a signal */
++sysret_signal:
++/* sti */
++ XEN_UNBLOCK_EVENTS(%rsi)
++ testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
++ jz 1f
++
++ /* Really a signal */
++ /* edx: work flags (arg3) */
++ leaq do_notify_resume(%rip),%rax
++ leaq -ARGOFFSET(%rsp),%rdi # &pt_regs -> arg1
++ xorl %esi,%esi # oldset -> arg2
++ call ptregscall_common
++1: movl $_TIF_NEED_RESCHED,%edi
++ /* Use IRET because user could have changed frame. This
++ works because ptregscall_common has called FIXUP_TOP_OF_STACK. */
++ cli
++ jmp int_with_check
++
++badsys:
++ movq $-ENOSYS,RAX-ARGOFFSET(%rsp)
++ jmp ret_from_sys_call
++
++ /* Do syscall tracing */
++tracesys:
++ CFI_RESTORE_STATE
++ SAVE_REST
++ movq $-ENOSYS,RAX(%rsp)
++ FIXUP_TOP_OF_STACK %rdi
++ movq %rsp,%rdi
++ call syscall_trace_enter
++ LOAD_ARGS ARGOFFSET /* reload args from stack in case ptrace changed it */
++ RESTORE_REST
++ cmpq $__NR_syscall_max,%rax
++ ja 1f
++ movq %r10,%rcx /* fixup for C */
++ call *sys_call_table(,%rax,8)
++ movq %rax,RAX-ARGOFFSET(%rsp)
++1: SAVE_REST
++ movq %rsp,%rdi
++ call syscall_trace_leave
++ RESTORE_TOP_OF_STACK %rbx
++ RESTORE_REST
++ /* Use IRET because user could have changed frame */
++ jmp int_ret_from_sys_call
++ CFI_ENDPROC
++
++/*
++ * Syscall return path ending with IRET.
++ * Has correct top of stack, but partial stack frame.
++ */
++ENTRY(int_ret_from_sys_call)
++ CFI_STARTPROC simple
++ CFI_DEF_CFA rsp,SS+8-ARGOFFSET
++ /*CFI_REL_OFFSET ss,SS-ARGOFFSET*/
++ CFI_REL_OFFSET rsp,RSP-ARGOFFSET
++ /*CFI_REL_OFFSET rflags,EFLAGS-ARGOFFSET*/
++ /*CFI_REL_OFFSET cs,CS-ARGOFFSET*/
++ CFI_REL_OFFSET rip,RIP-ARGOFFSET
++ CFI_REL_OFFSET rdx,RDX-ARGOFFSET
++ CFI_REL_OFFSET rcx,RCX-ARGOFFSET
++ CFI_REL_OFFSET rax,RAX-ARGOFFSET
++ CFI_REL_OFFSET rdi,RDI-ARGOFFSET
++ CFI_REL_OFFSET rsi,RSI-ARGOFFSET
++ CFI_REL_OFFSET r8,R8-ARGOFFSET
++ CFI_REL_OFFSET r9,R9-ARGOFFSET
++ CFI_REL_OFFSET r10,R10-ARGOFFSET
++ CFI_REL_OFFSET r11,R11-ARGOFFSET
++ XEN_BLOCK_EVENTS(%rsi)
++ testb $3,CS-ARGOFFSET(%rsp)
++ jnz 1f
++ /* Need to set the proper %ss (not NULL) for ring 3 iretq */
++ movl $__KERNEL_DS,SS-ARGOFFSET(%rsp)
++ jmp retint_restore_args # retrun from ring3 kernel
++1:
++ movl $_TIF_ALLWORK_MASK,%edi
++ /* edi: mask to check */
++int_with_check:
++ GET_THREAD_INFO(%rcx)
++ movl threadinfo_flags(%rcx),%edx
++ andl %edi,%edx
++ jnz int_careful
++ andl $~TS_COMPAT,threadinfo_status(%rcx)
++ jmp retint_restore_args
++
++ /* Either reschedule or signal or syscall exit tracking needed. */
++ /* First do a reschedule test. */
++ /* edx: work, edi: workmask */
++int_careful:
++ bt $TIF_NEED_RESCHED,%edx
++ jnc int_very_careful
++/* sti */
++ XEN_UNBLOCK_EVENTS(%rsi)
++ pushq %rdi
++ CFI_ADJUST_CFA_OFFSET 8
++ call schedule
++ popq %rdi
++ CFI_ADJUST_CFA_OFFSET -8
++ cli
++ jmp int_with_check
++
++ /* handle signals and tracing -- both require a full stack frame */
++int_very_careful:
++/* sti */
++ XEN_UNBLOCK_EVENTS(%rsi)
++ SAVE_REST
++ /* Check for syscall exit trace */
++ testl $(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edx
++ jz int_signal
++ pushq %rdi
++ CFI_ADJUST_CFA_OFFSET 8
++ leaq 8(%rsp),%rdi # &ptregs -> arg1
++ call syscall_trace_leave
++ popq %rdi
++ CFI_ADJUST_CFA_OFFSET -8
++ andl $~(_TIF_SYSCALL_TRACE|_TIF_SYSCALL_AUDIT|_TIF_SINGLESTEP),%edi
++ cli
++ jmp int_restore_rest
++
++int_signal:
++ testl $(_TIF_NOTIFY_RESUME|_TIF_SIGPENDING|_TIF_SINGLESTEP),%edx
++ jz 1f
++ movq %rsp,%rdi # &ptregs -> arg1
++ xorl %esi,%esi # oldset -> arg2
++ call do_notify_resume
++1: movl $_TIF_NEED_RESCHED,%edi
++int_restore_rest:
++ RESTORE_REST
++ cli
++ jmp int_with_check
++ CFI_ENDPROC
++
++/*
++ * Certain special system calls that need to save a complete full stack frame.
++ */
++
++ .macro PTREGSCALL label,func,arg
++ .globl \label
++\label:
++ leaq \func(%rip),%rax
++ leaq -ARGOFFSET+8(%rsp),\arg /* 8 for return address */
++ jmp ptregscall_common
++ .endm
++
++ CFI_STARTPROC
++
++ PTREGSCALL stub_clone, sys_clone, %r8
++ PTREGSCALL stub_fork, sys_fork, %rdi
++ PTREGSCALL stub_vfork, sys_vfork, %rdi
++ PTREGSCALL stub_rt_sigsuspend, sys_rt_sigsuspend, %rdx
++ PTREGSCALL stub_sigaltstack, sys_sigaltstack, %rdx
++ PTREGSCALL stub_iopl, sys_iopl, %rsi
++
++ENTRY(ptregscall_common)
++ popq %r11
++ CFI_ADJUST_CFA_OFFSET -8
++ CFI_REGISTER rip, r11
++ SAVE_REST
++ movq %r11, %r15
++ CFI_REGISTER rip, r15
++ FIXUP_TOP_OF_STACK %r11
++ call *%rax
++ RESTORE_TOP_OF_STACK %r11
++ movq %r15, %r11
++ CFI_REGISTER rip, r11
++ RESTORE_REST
++ pushq %r11
++ CFI_ADJUST_CFA_OFFSET 8
++ CFI_REL_OFFSET rip, 0
++ ret
++ CFI_ENDPROC
++
++ENTRY(stub_execve)
++ CFI_STARTPROC
++ popq %r11
++ CFI_ADJUST_CFA_OFFSET -8
++ CFI_REGISTER rip, r11
++ SAVE_REST
++ FIXUP_TOP_OF_STACK %r11
++ call sys_execve
++ RESTORE_TOP_OF_STACK %r11
++ movq %rax,RAX(%rsp)
++ RESTORE_REST
++ jmp int_ret_from_sys_call
++ CFI_ENDPROC
++
++/*
++ * sigreturn is special because it needs to restore all registers on return.
++ * This cannot be done with SYSRET, so use the IRET return path instead.
++ */
++ENTRY(stub_rt_sigreturn)
++ CFI_STARTPROC
++ addq $8, %rsp
++ CFI_ADJUST_CFA_OFFSET -8
++ SAVE_REST
++ movq %rsp,%rdi
++ FIXUP_TOP_OF_STACK %r11
++ call sys_rt_sigreturn
++ movq %rax,RAX(%rsp) # fixme, this could be done at the higher layer
++ RESTORE_REST
++ jmp int_ret_from_sys_call
++ CFI_ENDPROC
++
++/*
++ * initial frame state for interrupts and exceptions
++ */
++ .macro _frame ref
++ CFI_STARTPROC simple
++ CFI_DEF_CFA rsp,SS+8-\ref
++ /*CFI_REL_OFFSET ss,SS-\ref*/
++ CFI_REL_OFFSET rsp,RSP-\ref
++ /*CFI_REL_OFFSET rflags,EFLAGS-\ref*/
++ /*CFI_REL_OFFSET cs,CS-\ref*/
++ CFI_REL_OFFSET rip,RIP-\ref
++ .endm
++
++/* initial frame state for interrupts (and exceptions without error code) */
++#define INTR_FRAME _frame RIP
++/* initial frame state for exceptions with error code (and interrupts with
++ vector already pushed) */
++#define XCPT_FRAME _frame ORIG_RAX
++
++/*
++ * Interrupt exit.
++ *
++ */
++
++retint_check:
++ movl threadinfo_flags(%rcx),%edx
++ andl %edi,%edx
++ CFI_REMEMBER_STATE
++ jnz retint_careful
++retint_restore_args:
++ movl EFLAGS-REST_SKIP(%rsp), %eax
++ shr $9, %eax # EAX[0] == IRET_EFLAGS.IF
++ XEN_GET_VCPU_INFO(%rsi)
++ andb evtchn_upcall_mask(%rsi),%al
++ andb $1,%al # EAX[0] == IRET_EFLAGS.IF & event_mask
++ jnz restore_all_enable_events # != 0 => enable event delivery
++ XEN_PUT_VCPU_INFO(%rsi)
++
++ RESTORE_ARGS 0,8,0
++ HYPERVISOR_IRET 0
++
++ /* edi: workmask, edx: work */
++retint_careful:
++ CFI_RESTORE_STATE
++ bt $TIF_NEED_RESCHED,%edx
++ jnc retint_signal
++ XEN_UNBLOCK_EVENTS(%rsi)
++/* sti */
++ pushq %rdi
++ CFI_ADJUST_CFA_OFFSET 8
++ call schedule
++ popq %rdi
++ CFI_ADJUST_CFA_OFFSET -8
++ XEN_BLOCK_EVENTS(%rsi)
++ GET_THREAD_INFO(%rcx)
++/* cli */
++ jmp retint_check
++
++retint_signal:
++ testl $(_TIF_SIGPENDING|_TIF_NOTIFY_RESUME|_TIF_SINGLESTEP),%edx
++ jz retint_restore_args
++ XEN_UNBLOCK_EVENTS(%rsi)
++ SAVE_REST
++ movq $-1,ORIG_RAX(%rsp)
++ xorl %esi,%esi # oldset
++ movq %rsp,%rdi # &pt_regs
++ call do_notify_resume
++ RESTORE_REST
++ XEN_BLOCK_EVENTS(%rsi)
++ movl $_TIF_NEED_RESCHED,%edi
++ GET_THREAD_INFO(%rcx)
++ jmp retint_check
++
++#ifdef CONFIG_PREEMPT
++ /* Returning to kernel space. Check if we need preemption */
++ /* rcx: threadinfo. interrupts off. */
++ .p2align
++retint_kernel:
++ cmpl $0,threadinfo_preempt_count(%rcx)
++ jnz retint_restore_args
++ bt $TIF_NEED_RESCHED,threadinfo_flags(%rcx)
++ jnc retint_restore_args
++ bt $9,EFLAGS-ARGOFFSET(%rsp) /* interrupts off? */
++ jnc retint_restore_args
++ call preempt_schedule_irq
++ jmp retint_kernel /* check again */
++#endif
++ CFI_ENDPROC
++
++/*
++ * APIC interrupts.
++ */
++ .macro apicinterrupt num,func
++ INTR_FRAME
++ pushq $~(\num)
++ CFI_ADJUST_CFA_OFFSET 8
++ interrupt \func
++ jmp error_entry
++ CFI_ENDPROC
++ .endm
++
++#ifndef CONFIG_XEN
++ENTRY(thermal_interrupt)
++ apicinterrupt THERMAL_APIC_VECTOR,smp_thermal_interrupt
++
++ENTRY(threshold_interrupt)
++ apicinterrupt THRESHOLD_APIC_VECTOR,mce_threshold_interrupt
++
++#ifdef CONFIG_SMP
++ENTRY(reschedule_interrupt)
++ apicinterrupt RESCHEDULE_VECTOR,smp_reschedule_interrupt
++
++ .macro INVALIDATE_ENTRY num
++ENTRY(invalidate_interrupt\num)
++ apicinterrupt INVALIDATE_TLB_VECTOR_START+\num,smp_invalidate_interrupt
++ .endm
++
++ INVALIDATE_ENTRY 0
++ INVALIDATE_ENTRY 1
++ INVALIDATE_ENTRY 2
++ INVALIDATE_ENTRY 3
++ INVALIDATE_ENTRY 4
++ INVALIDATE_ENTRY 5
++ INVALIDATE_ENTRY 6
++ INVALIDATE_ENTRY 7
++
++ENTRY(call_function_interrupt)
++ apicinterrupt CALL_FUNCTION_VECTOR,smp_call_function_interrupt
++#endif
++
++#ifdef CONFIG_X86_LOCAL_APIC
++ENTRY(apic_timer_interrupt)
++ apicinterrupt LOCAL_TIMER_VECTOR,smp_apic_timer_interrupt
++
++ENTRY(error_interrupt)
++ apicinterrupt ERROR_APIC_VECTOR,smp_error_interrupt
++
++ENTRY(spurious_interrupt)
++ apicinterrupt SPURIOUS_APIC_VECTOR,smp_spurious_interrupt
++#endif
++#endif /* !CONFIG_XEN */
++
++/*
++ * Exception entry points.
++ */
++ .macro zeroentry sym
++ INTR_FRAME
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x10,%rsp /* skip rcx and r11 */
++ pushq $0 /* push error code/oldrax */
++ CFI_ADJUST_CFA_OFFSET 8
++ pushq %rax /* push real oldrax to the rdi slot */
++ CFI_ADJUST_CFA_OFFSET 8
++ leaq \sym(%rip),%rax
++ jmp error_entry
++ CFI_ENDPROC
++ .endm
++
++ .macro errorentry sym
++ XCPT_FRAME
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x10,%rsp /* rsp points to the error code */
++ pushq %rax
++ CFI_ADJUST_CFA_OFFSET 8
++ leaq \sym(%rip),%rax
++ jmp error_entry
++ CFI_ENDPROC
++ .endm
++
++#if 0 /* not XEN */
++ /* error code is on the stack already */
++ /* handle NMI like exceptions that can happen everywhere */
++ .macro paranoidentry sym, ist=0
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x10,%rsp /* skip rcx and r11 */
++ SAVE_ALL
++ cld
++#if 0 /* not XEN */
++ movl $1,%ebx
++ movl $MSR_GS_BASE,%ecx
++ rdmsr
++ testl %edx,%edx
++ js 1f
++ swapgs
++ xorl %ebx,%ebx
++1:
++#endif
++ .if \ist
++ movq %gs:pda_data_offset, %rbp
++ .endif
++ movq %rsp,%rdi
++ movq ORIG_RAX(%rsp),%rsi
++ movq $-1,ORIG_RAX(%rsp)
++ .if \ist
++ subq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
++ .endif
++ call \sym
++ .if \ist
++ addq $EXCEPTION_STKSZ, per_cpu__init_tss + TSS_ist + (\ist - 1) * 8(%rbp)
++ .endif
++/* cli */
++ XEN_BLOCK_EVENTS(%rsi)
++ .endm
++#endif
++
++/*
++ * Exception entry point. This expects an error code/orig_rax on the stack
++ * and the exception handler in %rax.
++ */
++ENTRY(error_entry)
++ _frame RDI
++ /* rdi slot contains rax, oldrax contains error code */
++ cld
++ subq $14*8,%rsp
++ CFI_ADJUST_CFA_OFFSET (14*8)
++ movq %rsi,13*8(%rsp)
++ CFI_REL_OFFSET rsi,RSI
++ movq 14*8(%rsp),%rsi /* load rax from rdi slot */
++ movq %rdx,12*8(%rsp)
++ CFI_REL_OFFSET rdx,RDX
++ movq %rcx,11*8(%rsp)
++ CFI_REL_OFFSET rcx,RCX
++ movq %rsi,10*8(%rsp) /* store rax */
++ CFI_REL_OFFSET rax,RAX
++ movq %r8, 9*8(%rsp)
++ CFI_REL_OFFSET r8,R8
++ movq %r9, 8*8(%rsp)
++ CFI_REL_OFFSET r9,R9
++ movq %r10,7*8(%rsp)
++ CFI_REL_OFFSET r10,R10
++ movq %r11,6*8(%rsp)
++ CFI_REL_OFFSET r11,R11
++ movq %rbx,5*8(%rsp)
++ CFI_REL_OFFSET rbx,RBX
++ movq %rbp,4*8(%rsp)
++ CFI_REL_OFFSET rbp,RBP
++ movq %r12,3*8(%rsp)
++ CFI_REL_OFFSET r12,R12
++ movq %r13,2*8(%rsp)
++ CFI_REL_OFFSET r13,R13
++ movq %r14,1*8(%rsp)
++ CFI_REL_OFFSET r14,R14
++ movq %r15,(%rsp)
++ CFI_REL_OFFSET r15,R15
++#if 0
++ cmpl $__KERNEL_CS,CS(%rsp)
++ je error_kernelspace
++#endif
++error_call_handler:
++ movq %rdi, RDI(%rsp)
++ movq %rsp,%rdi
++ movq ORIG_RAX(%rsp),%rsi # get error code
++ movq $-1,ORIG_RAX(%rsp)
++ call *%rax
++error_exit:
++ RESTORE_REST
++/* cli */
++ XEN_BLOCK_EVENTS(%rsi)
++ GET_THREAD_INFO(%rcx)
++ testb $3,CS-ARGOFFSET(%rsp)
++ jz retint_kernel
++ movl threadinfo_flags(%rcx),%edx
++ movl $_TIF_WORK_MASK,%edi
++ andl %edi,%edx
++ jnz retint_careful
++ jmp retint_restore_args
++
++error_kernelspace:
++ /*
++ * We need to re-write the logic here because we don't do iretq to
++ * to return to user mode. It's still possible that we get trap/fault
++ * in the kernel (when accessing buffers pointed to by system calls,
++ * for example).
++ *
++ */
++#if 0
++ incl %ebx
++ /* There are two places in the kernel that can potentially fault with
++ usergs. Handle them here. The exception handlers after
++ iret run with kernel gs again, so don't set the user space flag.
++ B stepping K8s sometimes report an truncated RIP for IRET
++ exceptions returning to compat mode. Check for these here too. */
++ leaq iret_label(%rip),%rbp
++ cmpq %rbp,RIP(%rsp)
++ je error_swapgs
++ movl %ebp,%ebp /* zero extend */
++ cmpq %rbp,RIP(%rsp)
++ je error_swapgs
++ cmpq $gs_change,RIP(%rsp)
++ je error_swapgs
++ jmp error_sti
++#endif
++
++ENTRY(hypervisor_callback)
++ zeroentry do_hypervisor_callback
++
++/*
++ * Copied from arch/xen/i386/kernel/entry.S
++ */
++# A note on the "critical region" in our callback handler.
++# We want to avoid stacking callback handlers due to events occurring
++# during handling of the last event. To do this, we keep events disabled
++# until we've done all processing. HOWEVER, we must enable events before
++# popping the stack frame (can't be done atomically) and so it would still
++# be possible to get enough handler activations to overflow the stack.
++# Although unlikely, bugs of that kind are hard to track down, so we'd
++# like to avoid the possibility.
++# So, on entry to the handler we detect whether we interrupted an
++# existing activation in its critical region -- if so, we pop the current
++# activation and restart the handler using the previous one.
++ENTRY(do_hypervisor_callback) # do_hypervisor_callback(struct *pt_regs)
++# Since we don't modify %rdi, evtchn_do_upall(struct *pt_regs) will
++# see the correct pointer to the pt_regs
++ movq %rdi, %rsp # we don't return, adjust the stack frame
++11: movq %gs:pda_irqstackptr,%rax
++ incl %gs:pda_irqcount
++ cmovzq %rax,%rsp
++ pushq %rdi
++ call evtchn_do_upcall
++ popq %rsp
++ decl %gs:pda_irqcount
++ jmp error_exit
++
++#ifdef CONFIG_X86_LOCAL_APIC
++KPROBE_ENTRY(nmi)
++ zeroentry do_nmi_callback
++ENTRY(do_nmi_callback)
++ addq $8, %rsp
++ call do_nmi
++ orl $NMI_MASK,EFLAGS(%rsp)
++ RESTORE_REST
++ XEN_BLOCK_EVENTS(%rsi)
++ GET_THREAD_INFO(%rcx)
++ jmp retint_restore_args
++ .previous .text
++#endif
++
++ ALIGN
++restore_all_enable_events:
++ XEN_UNBLOCK_EVENTS(%rsi) # %rsi is already set up...
++
++scrit: /**** START OF CRITICAL REGION ****/
++ XEN_TEST_PENDING(%rsi)
++ jnz 14f # process more events if necessary...
++ XEN_PUT_VCPU_INFO(%rsi)
++ RESTORE_ARGS 0,8,0
++ HYPERVISOR_IRET 0
++
++14: XEN_LOCKED_BLOCK_EVENTS(%rsi)
++ XEN_PUT_VCPU_INFO(%rsi)
++ SAVE_REST
++ movq %rsp,%rdi # set the argument again
++ jmp 11b
++ecrit: /**** END OF CRITICAL REGION ****/
++# At this point, unlike on x86-32, we don't do the fixup to simplify the
++# code and the stack frame is more complex on x86-64.
++# When the kernel is interrupted in the critical section, the kernel
++# will do IRET in that case, and everything will be restored at that point,
++# i.e. it just resumes from the next instruction interrupted with the same context.
++
++# Hypervisor uses this for application faults while it executes.
++# We get here for two reasons:
++# 1. Fault while reloading DS, ES, FS or GS
++# 2. Fault while executing IRET
++# Category 1 we do not need to fix up as Xen has already reloaded all segment
++# registers that could be reloaded and zeroed the others.
++# Category 2 we fix up by killing the current process. We cannot use the
++# normal Linux return path in this case because if we use the IRET hypercall
++# to pop the stack frame we end up in an infinite loop of failsafe callbacks.
++# We distinguish between categories by comparing each saved segment register
++# with its current contents: any discrepancy means we in category 1.
++ENTRY(failsafe_callback)
++ movw %ds,%cx
++ cmpw %cx,0x10(%rsp)
++ jne 1f
++ movw %es,%cx
++ cmpw %cx,0x18(%rsp)
++ jne 1f
++ movw %fs,%cx
++ cmpw %cx,0x20(%rsp)
++ jne 1f
++ movw %gs,%cx
++ cmpw %cx,0x28(%rsp)
++ jne 1f
++ /* All segments match their saved values => Category 2 (Bad IRET). */
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x30,%rsp
++ movq $-9999,%rdi /* better code? */
++ jmp do_exit
++1: /* Segment mismatch => Category 1 (Bad segment). Retry the IRET. */
++ movq (%rsp),%rcx
++ movq 8(%rsp),%r11
++ addq $0x30,%rsp
++ pushq $0
++ SAVE_ALL
++ jmp error_exit
++#if 0
++ .section __ex_table,"a"
++ .align 8
++ .quad gs_change,bad_gs
++ .previous
++ .section .fixup,"ax"
++ /* running with kernelgs */
++bad_gs:
++/* swapgs */ /* switch back to user gs */
++ xorl %eax,%eax
++ movl %eax,%gs
++ jmp 2b
++ .previous
++#endif
++
++/*
++ * Create a kernel thread.
++ *
++ * C extern interface:
++ * extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags)
++ *
++ * asm input arguments:
++ * rdi: fn, rsi: arg, rdx: flags
++ */
++ENTRY(kernel_thread)
++ CFI_STARTPROC
++ FAKE_STACK_FRAME $child_rip
++ SAVE_ALL
++
++ # rdi: flags, rsi: usp, rdx: will be &pt_regs
++ movq %rdx,%rdi
++ orq kernel_thread_flags(%rip),%rdi
++ movq $-1, %rsi
++ movq %rsp, %rdx
++
++ xorl %r8d,%r8d
++ xorl %r9d,%r9d
++
++ # clone now
++ call do_fork
++ movq %rax,RAX(%rsp)
++ xorl %edi,%edi
++
++ /*
++ * It isn't worth to check for reschedule here,
++ * so internally to the x86_64 port you can rely on kernel_thread()
++ * not to reschedule the child before returning, this avoids the need
++ * of hacks for example to fork off the per-CPU idle tasks.
++ * [Hopefully no generic code relies on the reschedule -AK]
++ */
++ RESTORE_ALL
++ UNFAKE_STACK_FRAME
++ ret
++ CFI_ENDPROC
++
++
++child_rip:
++ /*
++ * Here we are in the child and the registers are set as they were
++ * at kernel_thread() invocation in the parent.
++ */
++ movq %rdi, %rax
++ movq %rsi, %rdi
++ call *%rax
++ # exit
++ xorl %edi, %edi
++ call do_exit
++
++/*
++ * execve(). This function needs to use IRET, not SYSRET, to set up all state properly.
++ *
++ * C extern interface:
++ * extern long execve(char *name, char **argv, char **envp)
++ *
++ * asm input arguments:
++ * rdi: name, rsi: argv, rdx: envp
++ *
++ * We want to fallback into:
++ * extern long sys_execve(char *name, char **argv,char **envp, struct pt_regs regs)
++ *
++ * do_sys_execve asm fallback arguments:
++ * rdi: name, rsi: argv, rdx: envp, fake frame on the stack
++ */
++ENTRY(execve)
++ CFI_STARTPROC
++ FAKE_STACK_FRAME $0
++ SAVE_ALL
++ call sys_execve
++ movq %rax, RAX(%rsp)
++ RESTORE_REST
++ testq %rax,%rax
++ jne 1f
++ jmp int_ret_from_sys_call
++1: RESTORE_ARGS
++ UNFAKE_STACK_FRAME
++ ret
++ CFI_ENDPROC
++
++KPROBE_ENTRY(page_fault)
++ errorentry do_page_fault
++ .previous .text
++
++ENTRY(coprocessor_error)
++ zeroentry do_coprocessor_error
++
++ENTRY(simd_coprocessor_error)
++ zeroentry do_simd_coprocessor_error
++
++ENTRY(device_not_available)
++ zeroentry math_state_restore
++
++ /* runs on exception stack */
++KPROBE_ENTRY(debug)
++ INTR_FRAME
++/* pushq $0
++ CFI_ADJUST_CFA_OFFSET 8 */
++ zeroentry do_debug
++/* jmp paranoid_exit */
++ CFI_ENDPROC
++ .previous .text
++
++#if 0
++ /* runs on exception stack */
++KPROBE_ENTRY(nmi)
++ INTR_FRAME
++ pushq $-1
++ CFI_ADJUST_CFA_OFFSET 8
++ paranoidentry do_nmi
++ /*
++ * "Paranoid" exit path from exception stack.
++ * Paranoid because this is used by NMIs and cannot take
++ * any kernel state for granted.
++ * We don't do kernel preemption checks here, because only
++ * NMI should be common and it does not enable IRQs and
++ * cannot get reschedule ticks.
++ */
++ /* ebx: no swapgs flag */
++paranoid_exit:
++ testl %ebx,%ebx /* swapgs needed? */
++ jnz paranoid_restore
++ testl $3,CS(%rsp)
++ jnz paranoid_userspace
++paranoid_swapgs:
++ swapgs
++paranoid_restore:
++ RESTORE_ALL 8
++ iretq
++paranoid_userspace:
++ GET_THREAD_INFO(%rcx)
++ movl threadinfo_flags(%rcx),%ebx
++ andl $_TIF_WORK_MASK,%ebx
++ jz paranoid_swapgs
++ movq %rsp,%rdi /* &pt_regs */
++ call sync_regs
++ movq %rax,%rsp /* switch stack for scheduling */
++ testl $_TIF_NEED_RESCHED,%ebx
++ jnz paranoid_schedule
++ movl %ebx,%edx /* arg3: thread flags */
++ sti
++ xorl %esi,%esi /* arg2: oldset */
++ movq %rsp,%rdi /* arg1: &pt_regs */
++ call do_notify_resume
++ cli
++ jmp paranoid_userspace
++paranoid_schedule:
++ sti
++ call schedule
++ cli
++ jmp paranoid_userspace
++ CFI_ENDPROC
++ .previous .text
++#endif
++
++KPROBE_ENTRY(int3)
++ INTR_FRAME
++/* pushq $0
++ CFI_ADJUST_CFA_OFFSET 8 */
++ zeroentry do_int3
++/* jmp paranoid_exit */
++ CFI_ENDPROC
++ .previous .text
++
++ENTRY(overflow)
++ zeroentry do_overflow
++
++ENTRY(bounds)
++ zeroentry do_bounds
++
++ENTRY(invalid_op)
++ zeroentry do_invalid_op
++
++ENTRY(coprocessor_segment_overrun)
++ zeroentry do_coprocessor_segment_overrun
++
++ENTRY(reserved)
++ zeroentry do_reserved
++
++#if 0
++ /* runs on exception stack */
++ENTRY(double_fault)
++ XCPT_FRAME
++ paranoidentry do_double_fault
++ jmp paranoid_exit
++ CFI_ENDPROC
++#endif
++
++ENTRY(invalid_TSS)
++ errorentry do_invalid_TSS
++
++ENTRY(segment_not_present)
++ errorentry do_segment_not_present
++
++ /* runs on exception stack */
++ENTRY(stack_segment)
++ XCPT_FRAME
++ errorentry do_stack_segment
++ CFI_ENDPROC
++
++KPROBE_ENTRY(general_protection)
++ errorentry do_general_protection
++ .previous .text
++
++ENTRY(alignment_check)
++ errorentry do_alignment_check
++
++ENTRY(divide_error)
++ zeroentry do_divide_error
++
++ENTRY(spurious_interrupt_bug)
++ zeroentry do_spurious_interrupt_bug
++
++#ifdef CONFIG_X86_MCE
++ /* runs on exception stack */
++ENTRY(machine_check)
++ INTR_FRAME
++ pushq $0
++ CFI_ADJUST_CFA_OFFSET 8
++ paranoidentry do_machine_check
++ jmp paranoid_exit
++ CFI_ENDPROC
++#endif
++
++ENTRY(call_softirq)
++ CFI_STARTPROC
++ movq %gs:pda_irqstackptr,%rax
++ movq %rsp,%rdx
++ CFI_DEF_CFA_REGISTER rdx
++ incl %gs:pda_irqcount
++ cmove %rax,%rsp
++ pushq %rdx
++ /*todo CFI_DEF_CFA_EXPRESSION ...*/
++ call __do_softirq
++ popq %rsp
++ CFI_DEF_CFA_REGISTER rsp
++ decl %gs:pda_irqcount
++ ret
++ CFI_ENDPROC
+diff -urNp linux-2.6/arch/x86_64/kernel/genapic_xen.c new/arch/x86_64/kernel/genapic_xen.c
+--- linux-2.6/arch/x86_64/kernel/genapic_xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/genapic_xen.c 2006-05-09 12:33:18.000000000 +0200
+@@ -0,0 +1,162 @@
++/*
++ * Copyright 2004 James Cleverdon, IBM.
++ * Subject to the GNU Public License, v.2
++ *
++ * Xen APIC subarch code. Maximum 8 CPUs, logical delivery.
++ *
++ * Hacked for x86-64 by James Cleverdon from i386 architecture code by
++ * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and
++ * James Cleverdon.
++ *
++ * Hacked to pieces for Xen by Chris Wright.
++ */
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <linux/cpumask.h>
++#include <linux/string.h>
++#include <linux/kernel.h>
++#include <linux/ctype.h>
++#include <linux/init.h>
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++#include <asm/smp.h>
++#include <asm/ipi.h>
++#else
++#include <asm/apic.h>
++#include <asm/apicdef.h>
++#include <asm/genapic.h>
++#endif
++#include <xen/evtchn.h>
++
++DECLARE_PER_CPU(int, ipi_to_irq[NR_IPIS]);
++
++static inline void __send_IPI_one(unsigned int cpu, int vector)
++{
++ int irq = per_cpu(ipi_to_irq, cpu)[vector];
++ BUG_ON(irq < 0);
++ notify_remote_via_irq(irq);
++}
++
++void xen_send_IPI_shortcut(unsigned int shortcut, int vector, unsigned int dest)
++{
++ int cpu;
++
++ switch (shortcut) {
++ case APIC_DEST_SELF:
++ __send_IPI_one(smp_processor_id(), vector);
++ break;
++ case APIC_DEST_ALLBUT:
++ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
++ if (cpu == smp_processor_id())
++ continue;
++ if (cpu_isset(cpu, cpu_online_map)) {
++ __send_IPI_one(cpu, vector);
++ }
++ }
++ break;
++ case APIC_DEST_ALLINC:
++ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
++ if (cpu_isset(cpu, cpu_online_map)) {
++ __send_IPI_one(cpu, vector);
++ }
++ }
++ break;
++ default:
++ printk("XXXXXX __send_IPI_shortcut %08x vector %d\n", shortcut,
++ vector);
++ break;
++ }
++}
++
++static cpumask_t xen_target_cpus(void)
++{
++ return cpu_online_map;
++}
++
++/*
++ * Set up the logical destination ID.
++ * Do nothing, not called now.
++ */
++static void xen_init_apic_ldr(void)
++{
++ Dprintk("%s\n", __FUNCTION__);
++ return;
++}
++
++static void xen_send_IPI_allbutself(int vector)
++{
++ /*
++ * if there are no other CPUs in the system then
++ * we get an APIC send error if we try to broadcast.
++ * thus we have to avoid sending IPIs in this case.
++ */
++ Dprintk("%s\n", __FUNCTION__);
++ if (num_online_cpus() > 1)
++ xen_send_IPI_shortcut(APIC_DEST_ALLBUT, vector, APIC_DEST_LOGICAL);
++}
++
++static void xen_send_IPI_all(int vector)
++{
++ Dprintk("%s\n", __FUNCTION__);
++ xen_send_IPI_shortcut(APIC_DEST_ALLINC, vector, APIC_DEST_LOGICAL);
++}
++
++static void xen_send_IPI_mask(cpumask_t cpumask, int vector)
++{
++ unsigned long mask = cpus_addr(cpumask)[0];
++ unsigned int cpu;
++ unsigned long flags;
++
++ Dprintk("%s\n", __FUNCTION__);
++ local_irq_save(flags);
++ WARN_ON(mask & ~cpus_addr(cpu_online_map)[0]);
++
++ for (cpu = 0; cpu < NR_CPUS; ++cpu) {
++ if (cpu_isset(cpu, cpumask)) {
++ __send_IPI_one(cpu, vector);
++ }
++ }
++ local_irq_restore(flags);
++}
++
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++static int xen_apic_id_registered(void)
++{
++ /* better be set */
++ Dprintk("%s\n", __FUNCTION__);
++ return physid_isset(smp_processor_id(), phys_cpu_present_map);
++}
++#endif
++
++static unsigned int xen_cpu_mask_to_apicid(cpumask_t cpumask)
++{
++ Dprintk("%s\n", __FUNCTION__);
++ return cpus_addr(cpumask)[0] & APIC_ALL_CPUS;
++}
++
++static unsigned int phys_pkg_id(int index_msb)
++{
++ u32 ebx;
++
++ Dprintk("%s\n", __FUNCTION__);
++ ebx = cpuid_ebx(1);
++ return ((ebx >> 24) & 0xFF) >> index_msb;
++}
++
++struct genapic apic_xen = {
++ .name = "xen",
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++ .int_delivery_mode = dest_LowestPrio,
++#endif
++ .int_dest_mode = (APIC_DEST_LOGICAL != 0),
++ .int_delivery_dest = APIC_DEST_LOGICAL | APIC_DM_LOWEST,
++ .target_cpus = xen_target_cpus,
++#ifdef CONFIG_XEN_PRIVILEGED_GUEST
++ .apic_id_registered = xen_apic_id_registered,
++#endif
++ .init_apic_ldr = xen_init_apic_ldr,
++ .send_IPI_all = xen_send_IPI_all,
++ .send_IPI_allbutself = xen_send_IPI_allbutself,
++ .send_IPI_mask = xen_send_IPI_mask,
++ .cpu_mask_to_apicid = xen_cpu_mask_to_apicid,
++ .phys_pkg_id = phys_pkg_id,
++};
+diff -urNp linux-2.6/arch/x86_64/kernel/genapic-xen.c new/arch/x86_64/kernel/genapic-xen.c
+--- linux-2.6/arch/x86_64/kernel/genapic-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/genapic-xen.c 2006-05-09 12:33:18.000000000 +0200
+@@ -0,0 +1,144 @@
++/*
++ * Copyright 2004 James Cleverdon, IBM.
++ * Subject to the GNU Public License, v.2
++ *
++ * Generic APIC sub-arch probe layer.
++ *
++ * Hacked for x86-64 by James Cleverdon from i386 architecture code by
++ * Martin Bligh, Andi Kleen, James Bottomley, John Stultz, and
++ * James Cleverdon.
++ */
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <linux/cpumask.h>
++#include <linux/string.h>
++#include <linux/kernel.h>
++#include <linux/ctype.h>
++#include <linux/init.h>
++#include <linux/module.h>
++
++#include <asm/smp.h>
++#include <asm/ipi.h>
++
++#if defined(CONFIG_ACPI)
++#include <acpi/acpi_bus.h>
++#endif
++
++/* which logical CPU number maps to which CPU (physical APIC ID) */
++u8 x86_cpu_to_apicid[NR_CPUS] __read_mostly = { [0 ... NR_CPUS-1] = BAD_APICID };
++EXPORT_SYMBOL(x86_cpu_to_apicid);
++u8 x86_cpu_to_log_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
++
++extern struct genapic apic_cluster;
++extern struct genapic apic_flat;
++extern struct genapic apic_physflat;
++
++#ifndef CONFIG_XEN
++struct genapic *genapic = &apic_flat;
++#else
++extern struct genapic apic_xen;
++struct genapic *genapic = &apic_xen;
++#endif
++
++
++/*
++ * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
++ */
++void __init clustered_apic_check(void)
++{
++#ifndef CONFIG_XEN
++ long i;
++ u8 clusters, max_cluster;
++ u8 id;
++ u8 cluster_cnt[NUM_APIC_CLUSTERS];
++ int max_apic = 0;
++
++#if defined(CONFIG_ACPI)
++ /*
++ * Some x86_64 machines use physical APIC mode regardless of how many
++ * procs/clusters are present (x86_64 ES7000 is an example).
++ */
++ if (acpi_fadt.revision > FADT2_REVISION_ID)
++ if (acpi_fadt.force_apic_physical_destination_mode) {
++ genapic = &apic_cluster;
++ goto print;
++ }
++#endif
++
++ memset(cluster_cnt, 0, sizeof(cluster_cnt));
++ for (i = 0; i < NR_CPUS; i++) {
++ id = bios_cpu_apicid[i];
++ if (id == BAD_APICID)
++ continue;
++ if (id > max_apic)
++ max_apic = id;
++ cluster_cnt[APIC_CLUSTERID(id)]++;
++ }
++
++ /* Don't use clustered mode on AMD platforms. */
++ if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
++ genapic = &apic_physflat;
++#ifndef CONFIG_HOTPLUG_CPU
++ /* In the CPU hotplug case we cannot use broadcast mode
++ because that opens a race when a CPU is removed.
++ Stay at physflat mode in this case.
++ It is bad to do this unconditionally though. Once
++ we have ACPI platform support for CPU hotplug
++ we should detect hotplug capablity from ACPI tables and
++ only do this when really needed. -AK */
++ if (max_apic <= 8)
++ genapic = &apic_flat;
++#endif
++ goto print;
++ }
++
++ clusters = 0;
++ max_cluster = 0;
++
++ for (i = 0; i < NUM_APIC_CLUSTERS; i++) {
++ if (cluster_cnt[i] > 0) {
++ ++clusters;
++ if (cluster_cnt[i] > max_cluster)
++ max_cluster = cluster_cnt[i];
++ }
++ }
++
++ /*
++ * If we have clusters <= 1 and CPUs <= 8 in cluster 0, then flat mode,
++ * else if max_cluster <= 4 and cluster_cnt[15] == 0, clustered logical
++ * else physical mode.
++ * (We don't use lowest priority delivery + HW APIC IRQ steering, so
++ * can ignore the clustered logical case and go straight to physical.)
++ */
++ if (clusters <= 1 && max_cluster <= 8 && cluster_cnt[0] == max_cluster) {
++#ifdef CONFIG_HOTPLUG_CPU
++ /* Don't use APIC shortcuts in CPU hotplug to avoid races */
++ genapic = &apic_physflat;
++#else
++ genapic = &apic_flat;
++#endif
++ } else
++ genapic = &apic_cluster;
++
++print:
++#else
++ /* hardcode to xen apic functions */
++ genapic = &apic_xen;
++#endif
++ printk(KERN_INFO "Setting APIC routing to %s\n", genapic->name);
++}
++
++/* Same for both flat and clustered. */
++
++#ifdef CONFIG_XEN
++extern void xen_send_IPI_shortcut(unsigned int shortcut, int vector, unsigned int dest);
++#endif
++
++void send_IPI_self(int vector)
++{
++#ifndef CONFIG_XEN
++ __send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL);
++#else
++ xen_send_IPI_shortcut(APIC_DEST_SELF, vector, APIC_DEST_PHYSICAL);
++#endif
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/head64-xen.c new/arch/x86_64/kernel/head64-xen.c
+--- linux-2.6/arch/x86_64/kernel/head64-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/head64-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,159 @@
++/*
++ * linux/arch/x86_64/kernel/head64.c -- prepare to run common code
++ *
++ * Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
++ *
++ * $Id$
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Modified for Xen.
++ */
++
++#include <linux/init.h>
++#include <linux/linkage.h>
++#include <linux/types.h>
++#include <linux/kernel.h>
++#include <linux/string.h>
++#include <linux/percpu.h>
++#include <linux/module.h>
++
++#include <asm/processor.h>
++#include <asm/proto.h>
++#include <asm/smp.h>
++#include <asm/bootsetup.h>
++#include <asm/setup.h>
++#include <asm/desc.h>
++#include <asm/pgtable.h>
++#include <asm/sections.h>
++
++unsigned long start_pfn;
++
++/* Don't add a printk in there. printk relies on the PDA which is not initialized
++ yet. */
++#if 0
++static void __init clear_bss(void)
++{
++ memset(__bss_start, 0,
++ (unsigned long) __bss_stop - (unsigned long) __bss_start);
++}
++#endif
++
++#define NEW_CL_POINTER 0x228 /* Relative to real mode data */
++#define OLD_CL_MAGIC_ADDR 0x90020
++#define OLD_CL_MAGIC 0xA33F
++#define OLD_CL_BASE_ADDR 0x90000
++#define OLD_CL_OFFSET 0x90022
++
++extern char saved_command_line[];
++
++static void __init copy_bootdata(char *real_mode_data)
++{
++#ifndef CONFIG_XEN
++ int new_data;
++ char * command_line;
++
++ memcpy(x86_boot_params, real_mode_data, BOOT_PARAM_SIZE);
++ new_data = *(int *) (x86_boot_params + NEW_CL_POINTER);
++ if (!new_data) {
++ if (OLD_CL_MAGIC != * (u16 *) OLD_CL_MAGIC_ADDR) {
++ printk("so old bootloader that it does not support commandline?!\n");
++ return;
++ }
++ new_data = OLD_CL_BASE_ADDR + * (u16 *) OLD_CL_OFFSET;
++ printk("old bootloader convention, maybe loadlin?\n");
++ }
++ command_line = (char *) ((u64)(new_data));
++ memcpy(saved_command_line, command_line, COMMAND_LINE_SIZE);
++#else
++ int max_cmdline;
++
++ if ((max_cmdline = MAX_GUEST_CMDLINE) > COMMAND_LINE_SIZE)
++ max_cmdline = COMMAND_LINE_SIZE;
++ memcpy(saved_command_line, xen_start_info->cmd_line, max_cmdline);
++ saved_command_line[max_cmdline-1] = '\0';
++#endif
++ printk("Bootdata ok (command line is %s)\n", saved_command_line);
++}
++
++static void __init setup_boot_cpu_data(void)
++{
++ unsigned int dummy, eax;
++
++ /* get vendor info */
++ cpuid(0, (unsigned int *)&boot_cpu_data.cpuid_level,
++ (unsigned int *)&boot_cpu_data.x86_vendor_id[0],
++ (unsigned int *)&boot_cpu_data.x86_vendor_id[8],
++ (unsigned int *)&boot_cpu_data.x86_vendor_id[4]);
++
++ /* get cpu type */
++ cpuid(1, &eax, &dummy, &dummy,
++ (unsigned int *) &boot_cpu_data.x86_capability);
++ boot_cpu_data.x86 = (eax >> 8) & 0xf;
++ boot_cpu_data.x86_model = (eax >> 4) & 0xf;
++ boot_cpu_data.x86_mask = eax & 0xf;
++}
++
++#include <xen/interface/memory.h>
++unsigned long *machine_to_phys_mapping;
++EXPORT_SYMBOL(machine_to_phys_mapping);
++unsigned int machine_to_phys_order;
++EXPORT_SYMBOL(machine_to_phys_order);
++
++void __init x86_64_start_kernel(char * real_mode_data)
++{
++ struct xen_machphys_mapping mapping;
++ unsigned long machine_to_phys_nr_ents;
++ char *s;
++ int i;
++
++ xen_start_info = (struct start_info *)real_mode_data;
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ phys_to_machine_mapping =
++ (unsigned long *)xen_start_info->mfn_list;
++ start_pfn = (__pa(xen_start_info->pt_base) >> PAGE_SHIFT) +
++ xen_start_info->nr_pt_frames;
++ }
++
++
++ machine_to_phys_mapping = (unsigned long *)MACH2PHYS_VIRT_START;
++ machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
++ if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
++ machine_to_phys_mapping = (unsigned long *)mapping.v_start;
++ machine_to_phys_nr_ents = mapping.max_mfn + 1;
++ }
++ while ((1UL << machine_to_phys_order) < machine_to_phys_nr_ents )
++ machine_to_phys_order++;
++
++#if 0
++ for (i = 0; i < 256; i++)
++ set_intr_gate(i, early_idt_handler);
++ asm volatile("lidt %0" :: "m" (idt_descr));
++#endif
++
++ for (i = 0; i < NR_CPUS; i++)
++ cpu_pda(i) = &boot_cpu_pda[i];
++
++ pda_init(0);
++ copy_bootdata(real_mode_data);
++#ifdef CONFIG_SMP
++ cpu_set(0, cpu_online_map);
++#endif
++ s = strstr(saved_command_line, "earlyprintk=");
++ if (s != NULL)
++ setup_early_printk(strchr(s, '=') + 1);
++#ifdef CONFIG_NUMA
++ s = strstr(saved_command_line, "numa=");
++ if (s != NULL)
++ numa_setup(s+5);
++#endif
++#ifdef CONFIG_X86_IO_APIC
++ if (strstr(saved_command_line, "disableapic"))
++ disable_apic = 1;
++#endif
++ /* You need early console to see that */
++ if (__pa_symbol(&_end) >= KERNEL_TEXT_SIZE)
++ panic("Kernel too big for kernel mapping\n");
++
++ setup_boot_cpu_data();
++ start_kernel();
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/head-xen.S new/arch/x86_64/kernel/head-xen.S
+--- linux-2.6/arch/x86_64/kernel/head-xen.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/head-xen.S 2006-05-23 18:42:18.000000000 +0200
+@@ -0,0 +1,176 @@
++/*
++ * linux/arch/x86_64/kernel/head.S -- start in 32bit and switch to 64bit
++ *
++ * Copyright (C) 2000 Andrea Arcangeli <andrea@suse.de> SuSE
++ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
++ * Copyright (C) 2000 Karsten Keil <kkeil@suse.de>
++ * Copyright (C) 2001,2002 Andi Kleen <ak@suse.de>
++ *
++ * $Id$
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Modified for Xen
++ */
++
++
++#include <linux/linkage.h>
++#include <linux/threads.h>
++#include <linux/init.h>
++#include <asm/desc.h>
++#include <asm/segment.h>
++#include <asm/page.h>
++#include <asm/msr.h>
++#include <asm/cache.h>
++
++ .text
++ .section .bootstrap.text
++ .code64
++#define VIRT_ENTRY_OFFSET 0x0
++.org VIRT_ENTRY_OFFSET
++ .globl startup_64
++startup_64:
++ENTRY(_start)
++ movq $(init_thread_union+THREAD_SIZE-8),%rsp
++ /* zero EFLAGS after setting rsp */
++ pushq $0
++ popfq
++
++ /* rsi is pointer to startup info structure.
++ pass it to C */
++ movq %rsi,%rdi
++ jmp x86_64_start_kernel
++
++ENTRY(stext)
++ENTRY(_stext)
++
++ $page = 0
++#define NEXT_PAGE(name) \
++ $page = $page + 1; \
++ .org $page * 0x1000; \
++ phys_/**/name = $page * 0x1000 + __PHYSICAL_START; \
++ENTRY(name)
++
++NEXT_PAGE(init_level4_pgt)
++ /* This gets initialized in x86_64_start_kernel */
++ .fill 512,8,0
++
++ /*
++ * We update two pgd entries to make kernel and user pgd consistent
++ * at pgd_populate(). It can be used for kernel modules. So we place
++ * this page here for those cases to avoid memory corruption.
++ * We also use this page to establish the initiali mapping for
++ * vsyscall area.
++ */
++NEXT_PAGE(init_level4_user_pgt)
++ .fill 512,8,0
++
++NEXT_PAGE(level3_kernel_pgt)
++ .fill 512,8,0
++
++ /*
++ * This is used for vsyscall area mapping as we have a different
++ * level4 page table for user.
++ */
++NEXT_PAGE(level3_user_pgt)
++ .fill 512,8,0
++
++NEXT_PAGE(level2_kernel_pgt)
++ .fill 512,8,0
++
++NEXT_PAGE(hypercall_page)
++ .fill 512,8,0
++
++#undef NEXT_PAGE
++
++ .data
++
++ .align 16
++ .globl cpu_gdt_descr
++cpu_gdt_descr:
++ .word gdt_end-cpu_gdt_table-1
++gdt:
++ .quad cpu_gdt_table
++#ifdef CONFIG_SMP
++ .rept NR_CPUS-1
++ .word 0
++ .quad 0
++ .endr
++#endif
++
++/* We need valid kernel segments for data and code in long mode too
++ * IRET will check the segment types kkeil 2000/10/28
++ * Also sysret mandates a special GDT layout
++ */
++
++ .section .data.page_aligned, "aw"
++ .align PAGE_SIZE
++
++/* The TLS descriptors are currently at a different place compared to i386.
++ Hopefully nobody expects them at a fixed place (Wine?) */
++
++ENTRY(cpu_gdt_table)
++ .quad 0x0000000000000000 /* NULL descriptor */
++ .quad 0x0 /* unused */
++ .quad 0x00af9a000000ffff /* __KERNEL_CS */
++ .quad 0x00cf92000000ffff /* __KERNEL_DS */
++ .quad 0x00cffa000000ffff /* __USER32_CS */
++ .quad 0x00cff2000000ffff /* __USER_DS, __USER32_DS */
++ .quad 0x00affa000000ffff /* __USER_CS */
++ .quad 0x00cf9a000000ffff /* __KERNEL32_CS */
++ .quad 0,0 /* TSS */
++ .quad 0,0 /* LDT */
++ .quad 0,0,0 /* three TLS descriptors */
++ .quad 0 /* unused */
++gdt_end:
++ /* asm/segment.h:GDT_ENTRIES must match this */
++ /* This should be a multiple of the cache line size */
++ /* GDTs of other CPUs are now dynamically allocated */
++
++ /* zero the remaining page */
++ .fill PAGE_SIZE / 8 - GDT_ENTRIES,8,0
++
++ .section .bss, "aw", @nobits
++ .align L1_CACHE_BYTES
++ENTRY(idt_table)
++ .skip 256 * 16
++
++ .section .bss.page_aligned, "aw", @nobits
++ .align PAGE_SIZE
++ENTRY(empty_zero_page)
++ .skip PAGE_SIZE
++
++/*
++ * __xen_guest information
++ */
++.macro utoh value
++ .if (\value) < 0 || (\value) >= 0x10
++ utoh (((\value)>>4)&0x0fffffffffffffff)
++ .endif
++ .if ((\value) & 0xf) < 10
++ .byte '0' + ((\value) & 0xf)
++ .else
++ .byte 'A' + ((\value) & 0xf) - 10
++ .endif
++.endm
++
++.section __xen_guest
++ .ascii "GUEST_OS=linux,GUEST_VER=2.6"
++ .ascii ",XEN_VER=xen-3.0"
++ .ascii ",VIRT_BASE=0x"
++ utoh __START_KERNEL_map
++#ifdef CONFIG_XEN_COMPAT_030002
++ .ascii ",ELF_PADDR_OFFSET=0x"
++ utoh __START_KERNEL_map
++#else
++ .ascii ",ELF_PADDR_OFFSET=0x0"
++#endif /* !CONFIG_XEN_COMPAT_030002 */
++ .ascii ",VIRT_ENTRY=0x"
++ utoh (__START_KERNEL_map + __PHYSICAL_START + VIRT_ENTRY_OFFSET)
++ .ascii ",HYPERCALL_PAGE=0x"
++ utoh (phys_hypercall_page >> PAGE_SHIFT)
++ .ascii ",FEATURES=writable_page_tables"
++ .ascii "|writable_descriptor_tables"
++ .ascii "|auto_translated_physmap"
++ .ascii "|supervisor_mode_kernel"
++ .ascii ",LOADER=generic"
++ .byte 0
+diff -urNp linux-2.6/arch/x86_64/kernel/init_task.c new/arch/x86_64/kernel/init_task.c
+--- linux-2.6/arch/x86_64/kernel/init_task.c 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/init_task.c 2006-05-09 12:33:19.000000000 +0200
+@@ -37,6 +37,8 @@ union thread_union init_thread_union
+ struct task_struct init_task = INIT_TASK(init_task);
+
+ EXPORT_SYMBOL(init_task);
++
++#ifndef CONFIG_X86_NO_TSS
+ /*
+ * per-CPU TSS segments. Threads are completely 'soft' on Linux,
+ * no more per-task TSS's. The TSS size is kept cacheline-aligned
+@@ -45,5 +47,6 @@ EXPORT_SYMBOL(init_task);
+ * on exact cacheline boundaries, to eliminate cacheline ping-pong.
+ */
+ DEFINE_PER_CPU(struct tss_struct, init_tss) ____cacheline_internodealigned_in_smp = INIT_TSS;
++#endif
+
+ #define ALIGN_TO_4K __attribute__((section(".data.init_task")))
+diff -urNp linux-2.6/arch/x86_64/kernel/io_apic-xen.c new/arch/x86_64/kernel/io_apic-xen.c
+--- linux-2.6/arch/x86_64/kernel/io_apic-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/io_apic-xen.c 2006-05-23 18:37:10.000000000 +0200
+@@ -0,0 +1,2232 @@
++/*
++ * Intel IO-APIC support for multi-Pentium hosts.
++ *
++ * Copyright (C) 1997, 1998, 1999, 2000 Ingo Molnar, Hajnalka Szabo
++ *
++ * Many thanks to Stig Venaas for trying out countless experimental
++ * patches and reporting/debugging problems patiently!
++ *
++ * (c) 1999, Multiple IO-APIC support, developed by
++ * Ken-ichi Yaku <yaku@css1.kbnes.nec.co.jp> and
++ * Hidemi Kishimoto <kisimoto@css1.kbnes.nec.co.jp>,
++ * further tested and cleaned up by Zach Brown <zab@redhat.com>
++ * and Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Maciej W. Rozycki : Bits for genuine 82489DX APICs;
++ * thanks to Eric Gilmore
++ * and Rolf G. Tews
++ * for testing these extensively
++ * Paul Diefenbaugh : Added full ACPI support
++ */
++
++#include <linux/mm.h>
++#include <linux/interrupt.h>
++#include <linux/init.h>
++#include <linux/delay.h>
++#include <linux/sched.h>
++#include <linux/config.h>
++#include <linux/smp_lock.h>
++#include <linux/mc146818rtc.h>
++#include <linux/acpi.h>
++#include <linux/sysdev.h>
++#ifdef CONFIG_ACPI
++#include <acpi/acpi_bus.h>
++#endif
++
++#include <asm/io.h>
++#include <asm/smp.h>
++#include <asm/desc.h>
++#include <asm/proto.h>
++#include <asm/mach_apic.h>
++#include <asm/acpi.h>
++#include <asm/dma.h>
++
++#define __apicdebuginit __init
++
++int sis_apic_bug; /* not actually supported, dummy for compile */
++
++static int no_timer_check;
++
++int disable_timer_pin_1 __initdata;
++
++#ifndef CONFIG_XEN
++int timer_over_8254 __initdata = 0;
++
++/* Where if anywhere is the i8259 connect in external int mode */
++static struct { int pin, apic; } ioapic_i8259 = { -1, -1 };
++#endif
++
++static DEFINE_SPINLOCK(ioapic_lock);
++
++/*
++ * # of IRQ routing registers
++ */
++int nr_ioapic_registers[MAX_IO_APICS];
++
++/*
++ * Rough estimation of how many shared IRQs there are, can
++ * be changed anytime.
++ */
++#define MAX_PLUS_SHARED_IRQS NR_IRQ_VECTORS
++#define PIN_MAP_SIZE (MAX_PLUS_SHARED_IRQS + NR_IRQS)
++
++/*
++ * This is performance-critical, we want to do it O(1)
++ *
++ * the indexing order of this array favors 1:1 mappings
++ * between pins and IRQs.
++ */
++
++static struct irq_pin_list {
++ short apic, pin, next;
++} irq_2_pin[PIN_MAP_SIZE];
++
++int vector_irq[NR_VECTORS] __read_mostly = { [0 ... NR_VECTORS - 1] = -1};
++#ifdef CONFIG_PCI_MSI
++#define vector_to_irq(vector) \
++ (platform_legacy_irq(vector) ? vector : vector_irq[vector])
++#else
++#define vector_to_irq(vector) (vector)
++#endif
++
++#ifdef CONFIG_XEN
++
++#include <xen/interface/xen.h>
++#include <xen/interface/physdev.h>
++
++/* Fake i8259 */
++#define make_8259A_irq(_irq) (io_apic_irqs &= ~(1UL<<(_irq)))
++#define disable_8259A_irq(_irq) ((void)0)
++#define i8259A_irq_pending(_irq) (0)
++
++unsigned long io_apic_irqs;
++
++static inline unsigned int xen_io_apic_read(unsigned int apic, unsigned int reg)
++{
++ struct physdev_apic apic_op;
++ int ret;
++
++ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
++ apic_op.reg = reg;
++ ret = HYPERVISOR_physdev_op(PHYSDEVOP_apic_read, &apic_op);
++ if (ret)
++ return ret;
++ return apic_op.value;
++}
++
++static inline void xen_io_apic_write(unsigned int apic, unsigned int reg, unsigned int value)
++{
++ struct physdev_apic apic_op;
++
++ apic_op.apic_physbase = mp_ioapics[apic].mpc_apicaddr;
++ apic_op.reg = reg;
++ apic_op.value = value;
++ HYPERVISOR_physdev_op(PHYSDEVOP_apic_write, &apic_op);
++}
++
++#define io_apic_read(a,r) xen_io_apic_read(a,r)
++#define io_apic_write(a,r,v) xen_io_apic_write(a,r,v)
++
++#define clear_IO_APIC() ((void)0)
++
++#else
++
++#ifdef CONFIG_SMP
++static void set_ioapic_affinity_irq(unsigned int irq, cpumask_t mask)
++{
++ unsigned long flags;
++ unsigned int dest;
++ cpumask_t tmp;
++
++ cpus_and(tmp, mask, cpu_online_map);
++ if (cpus_empty(tmp))
++ tmp = TARGET_CPUS;
++
++ cpus_and(mask, tmp, CPU_MASK_ALL);
++
++ dest = cpu_mask_to_apicid(mask);
++
++ /*
++ * Only the high 8 bits are valid.
++ */
++ dest = SET_APIC_LOGICAL_ID(dest);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ __DO_ACTION(1, = dest, )
++ set_irq_info(irq, mask);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++#endif
++
++#endif /* !CONFIG_XEN */
++
++/*
++ * The common case is 1:1 IRQ<->pin mappings. Sometimes there are
++ * shared ISA-space IRQs, so we have to support them. We are super
++ * fast in the common case, and fast for shared ISA-space IRQs.
++ */
++static void add_pin_to_irq(unsigned int irq, int apic, int pin)
++{
++ static int first_free_entry = NR_IRQS;
++ struct irq_pin_list *entry = irq_2_pin + irq;
++
++ BUG_ON(irq >= NR_IRQS);
++ while (entry->next)
++ entry = irq_2_pin + entry->next;
++
++ if (entry->pin != -1) {
++ entry->next = first_free_entry;
++ entry = irq_2_pin + entry->next;
++ if (++first_free_entry >= PIN_MAP_SIZE)
++ panic("io_apic.c: ran out of irq_2_pin entries!");
++ }
++ entry->apic = apic;
++ entry->pin = pin;
++}
++
++#ifndef CONFIG_XEN
++#define __DO_ACTION(R, ACTION, FINAL) \
++ \
++{ \
++ int pin; \
++ struct irq_pin_list *entry = irq_2_pin + irq; \
++ \
++ BUG_ON(irq >= NR_IRQS); \
++ for (;;) { \
++ unsigned int reg; \
++ pin = entry->pin; \
++ if (pin == -1) \
++ break; \
++ reg = io_apic_read(entry->apic, 0x10 + R + pin*2); \
++ reg ACTION; \
++ io_apic_modify(entry->apic, reg); \
++ if (!entry->next) \
++ break; \
++ entry = irq_2_pin + entry->next; \
++ } \
++ FINAL; \
++}
++
++#define DO_ACTION(name,R,ACTION, FINAL) \
++ \
++ static void name##_IO_APIC_irq (unsigned int irq) \
++ __DO_ACTION(R, ACTION, FINAL)
++
++DO_ACTION( __mask, 0, |= 0x00010000, io_apic_sync(entry->apic) )
++ /* mask = 1 */
++DO_ACTION( __unmask, 0, &= 0xfffeffff, )
++ /* mask = 0 */
++
++static void mask_IO_APIC_irq (unsigned int irq)
++{
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ __mask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void unmask_IO_APIC_irq (unsigned int irq)
++{
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ __unmask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void clear_IO_APIC_pin(unsigned int apic, unsigned int pin)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ /* Check delivery_mode to be sure we're not clearing an SMI pin */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int*)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ *(((int*)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ if (entry.delivery_mode == dest_SMI)
++ return;
++ /*
++ * Disable it in the IO-APIC irq-routing table:
++ */
++ memset(&entry, 0, sizeof(entry));
++ entry.mask = 1;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry) + 0));
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry) + 1));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++static void clear_IO_APIC (void)
++{
++ int apic, pin;
++
++ for (apic = 0; apic < nr_ioapics; apic++)
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++)
++ clear_IO_APIC_pin(apic, pin);
++}
++
++#endif /* !CONFIG_XEN */
++
++static u8 gsi_2_irq[NR_IRQ_VECTORS] = { [0 ... NR_IRQ_VECTORS-1] = 0xFF };
++
++/*
++ * support for broken MP BIOSs, enables hand-redirection of PIRQ0-7 to
++ * specific CPU-side IRQs.
++ */
++
++#define MAX_PIRQS 8
++static int pirq_entries [MAX_PIRQS];
++static int pirqs_enabled;
++int skip_ioapic_setup;
++int ioapic_force;
++
++/* dummy parsing: see setup.c */
++
++static int __init disable_ioapic_setup(char *str)
++{
++ skip_ioapic_setup = 1;
++ return 1;
++}
++
++static int __init enable_ioapic_setup(char *str)
++{
++ ioapic_force = 1;
++ skip_ioapic_setup = 0;
++ return 1;
++}
++
++__setup("noapic", disable_ioapic_setup);
++__setup("apic", enable_ioapic_setup);
++
++#ifndef CONFIG_XEN
++static int __init setup_disable_8254_timer(char *s)
++{
++ timer_over_8254 = -1;
++ return 1;
++}
++static int __init setup_enable_8254_timer(char *s)
++{
++ timer_over_8254 = 2;
++ return 1;
++}
++
++__setup("disable_8254_timer", setup_disable_8254_timer);
++__setup("enable_8254_timer", setup_enable_8254_timer);
++#endif /* !CONFIG_XEN */
++
++#include <asm/pci-direct.h>
++#include <linux/pci_ids.h>
++#include <linux/pci.h>
++
++/* Temporary Hack. Nvidia and VIA boards currently only work with IO-APIC
++ off. Check for an Nvidia or VIA PCI bridge and turn it off.
++ Use pci direct infrastructure because this runs before the PCI subsystem.
++
++ Can be overwritten with "apic"
++
++ And another hack to disable the IOMMU on VIA chipsets.
++
++ ... and others. Really should move this somewhere else.
++
++ Kludge-O-Rama. */
++void __init check_ioapic(void)
++{
++ int num,slot,func;
++ /* Poor man's PCI discovery */
++ for (num = 0; num < 32; num++) {
++ for (slot = 0; slot < 32; slot++) {
++ for (func = 0; func < 8; func++) {
++ u32 class;
++ u32 vendor;
++ u8 type;
++ class = read_pci_config(num,slot,func,
++ PCI_CLASS_REVISION);
++ if (class == 0xffffffff)
++ break;
++
++ if ((class >> 16) != PCI_CLASS_BRIDGE_PCI)
++ continue;
++
++ vendor = read_pci_config(num, slot, func,
++ PCI_VENDOR_ID);
++ vendor &= 0xffff;
++ switch (vendor) {
++ case PCI_VENDOR_ID_VIA:
++#ifdef CONFIG_GART_IOMMU
++ if ((end_pfn > MAX_DMA32_PFN ||
++ force_iommu) &&
++ !iommu_aperture_allowed) {
++ printk(KERN_INFO
++ "Looks like a VIA chipset. Disabling IOMMU. Override with \"iommu=allowed\"\n");
++ iommu_aperture_disabled = 1;
++ }
++#endif
++ return;
++ case PCI_VENDOR_ID_NVIDIA:
++#ifdef CONFIG_ACPI
++ /* All timer overrides on Nvidia
++ seem to be wrong. Skip them. */
++ acpi_skip_timer_override = 1;
++ printk(KERN_INFO
++ "Nvidia board detected. Ignoring ACPI timer override.\n");
++#endif
++ /* RED-PEN skip them on mptables too? */
++ return;
++ case PCI_VENDOR_ID_ATI:
++
++ /* This should be actually default, but
++ for 2.6.16 let's do it for ATI only where
++ it's really needed. */
++#ifndef CONFIG_XEN
++ if (timer_over_8254 == 1) {
++ timer_over_8254 = 0;
++ printk(KERN_INFO
++ "ATI board detected. Disabling timer routing over 8254.\n");
++ }
++#endif
++ return;
++ }
++
++
++ /* No multi-function device? */
++ type = read_pci_config_byte(num,slot,func,
++ PCI_HEADER_TYPE);
++ if (!(type & 0x80))
++ break;
++ }
++ }
++ }
++}
++
++static int __init ioapic_pirq_setup(char *str)
++{
++ int i, max;
++ int ints[MAX_PIRQS+1];
++
++ get_options(str, ARRAY_SIZE(ints), ints);
++
++ for (i = 0; i < MAX_PIRQS; i++)
++ pirq_entries[i] = -1;
++
++ pirqs_enabled = 1;
++ apic_printk(APIC_VERBOSE, "PIRQ redirection, working around broken MP-BIOS.\n");
++ max = MAX_PIRQS;
++ if (ints[0] < MAX_PIRQS)
++ max = ints[0];
++
++ for (i = 0; i < max; i++) {
++ apic_printk(APIC_VERBOSE, "... PIRQ%d -> IRQ %d\n", i, ints[i+1]);
++ /*
++ * PIRQs are mapped upside down, usually.
++ */
++ pirq_entries[MAX_PIRQS-i-1] = ints[i+1];
++ }
++ return 1;
++}
++
++__setup("pirq=", ioapic_pirq_setup);
++
++/*
++ * Find the IRQ entry number of a certain pin.
++ */
++static int find_irq_entry(int apic, int pin, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++)
++ if (mp_irqs[i].mpc_irqtype == type &&
++ (mp_irqs[i].mpc_dstapic == mp_ioapics[apic].mpc_apicid ||
++ mp_irqs[i].mpc_dstapic == MP_APIC_ALL) &&
++ mp_irqs[i].mpc_dstirq == pin)
++ return i;
++
++ return -1;
++}
++
++#ifndef CONFIG_XEN
++/*
++ * Find the pin to which IRQ[irq] (ISA) is connected
++ */
++static int __init find_isa_irq_pin(int irq, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
++ (mp_irqs[i].mpc_irqtype == type) &&
++ (mp_irqs[i].mpc_srcbusirq == irq))
++
++ return mp_irqs[i].mpc_dstirq;
++ }
++ return -1;
++}
++
++static int __init find_isa_irq_apic(int irq, int type)
++{
++ int i;
++
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_ISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_EISA ||
++ mp_bus_id_to_type[lbus] == MP_BUS_MCA) &&
++ (mp_irqs[i].mpc_irqtype == type) &&
++ (mp_irqs[i].mpc_srcbusirq == irq))
++ break;
++ }
++ if (i < mp_irq_entries) {
++ int apic;
++ for(apic = 0; apic < nr_ioapics; apic++) {
++ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic)
++ return apic;
++ }
++ }
++
++ return -1;
++}
++#endif
++
++/*
++ * Find a specific PCI IRQ entry.
++ * Not an __init, possibly needed by modules
++ */
++static int pin_2_irq(int idx, int apic, int pin);
++
++int IO_APIC_get_PCI_irq_vector(int bus, int slot, int pin)
++{
++ int apic, i, best_guess = -1;
++
++ apic_printk(APIC_DEBUG, "querying PCI -> IRQ mapping bus:%d, slot:%d, pin:%d.\n",
++ bus, slot, pin);
++ if (mp_bus_id_to_pci_bus[bus] == -1) {
++ apic_printk(APIC_VERBOSE, "PCI BIOS passed nonexistent PCI bus %d!\n", bus);
++ return -1;
++ }
++ for (i = 0; i < mp_irq_entries; i++) {
++ int lbus = mp_irqs[i].mpc_srcbus;
++
++ for (apic = 0; apic < nr_ioapics; apic++)
++ if (mp_ioapics[apic].mpc_apicid == mp_irqs[i].mpc_dstapic ||
++ mp_irqs[i].mpc_dstapic == MP_APIC_ALL)
++ break;
++
++ if ((mp_bus_id_to_type[lbus] == MP_BUS_PCI) &&
++ !mp_irqs[i].mpc_irqtype &&
++ (bus == lbus) &&
++ (slot == ((mp_irqs[i].mpc_srcbusirq >> 2) & 0x1f))) {
++ int irq = pin_2_irq(i,apic,mp_irqs[i].mpc_dstirq);
++
++ if (!(apic || IO_APIC_IRQ(irq)))
++ continue;
++
++ if (pin == (mp_irqs[i].mpc_srcbusirq & 3))
++ return irq;
++ /*
++ * Use the first all-but-pin matching entry as a
++ * best-guess fuzzy result for broken mptables.
++ */
++ if (best_guess < 0)
++ best_guess = irq;
++ }
++ }
++ BUG_ON(best_guess >= NR_IRQS);
++ return best_guess;
++}
++
++/*
++ * EISA Edge/Level control register, ELCR
++ */
++static int EISA_ELCR(unsigned int irq)
++{
++ if (irq < 16) {
++ unsigned int port = 0x4d0 + (irq >> 3);
++ return (inb(port) >> (irq & 7)) & 1;
++ }
++ apic_printk(APIC_VERBOSE, "Broken MPtable reports ISA irq %d\n", irq);
++ return 0;
++}
++
++/* EISA interrupts are always polarity zero and can be edge or level
++ * trigger depending on the ELCR value. If an interrupt is listed as
++ * EISA conforming in the MP table, that means its trigger type must
++ * be read in from the ELCR */
++
++#define default_EISA_trigger(idx) (EISA_ELCR(mp_irqs[idx].mpc_srcbusirq))
++#define default_EISA_polarity(idx) (0)
++
++/* ISA interrupts are always polarity zero edge triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_ISA_trigger(idx) (0)
++#define default_ISA_polarity(idx) (0)
++
++/* PCI interrupts are always polarity one level triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_PCI_trigger(idx) (1)
++#define default_PCI_polarity(idx) (1)
++
++/* MCA interrupts are always polarity zero level triggered,
++ * when listed as conforming in the MP table. */
++
++#define default_MCA_trigger(idx) (1)
++#define default_MCA_polarity(idx) (0)
++
++static int __init MPBIOS_polarity(int idx)
++{
++ int bus = mp_irqs[idx].mpc_srcbus;
++ int polarity;
++
++ /*
++ * Determine IRQ line polarity (high active or low active):
++ */
++ switch (mp_irqs[idx].mpc_irqflag & 3)
++ {
++ case 0: /* conforms, ie. bus-type dependent polarity */
++ {
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ {
++ polarity = default_ISA_polarity(idx);
++ break;
++ }
++ case MP_BUS_EISA: /* EISA pin */
++ {
++ polarity = default_EISA_polarity(idx);
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ polarity = default_PCI_polarity(idx);
++ break;
++ }
++ case MP_BUS_MCA: /* MCA pin */
++ {
++ polarity = default_MCA_polarity(idx);
++ break;
++ }
++ default:
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ }
++ break;
++ }
++ case 1: /* high active */
++ {
++ polarity = 0;
++ break;
++ }
++ case 2: /* reserved */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ case 3: /* low active */
++ {
++ polarity = 1;
++ break;
++ }
++ default: /* invalid */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ polarity = 1;
++ break;
++ }
++ }
++ return polarity;
++}
++
++static int MPBIOS_trigger(int idx)
++{
++ int bus = mp_irqs[idx].mpc_srcbus;
++ int trigger;
++
++ /*
++ * Determine IRQ trigger mode (edge or level sensitive):
++ */
++ switch ((mp_irqs[idx].mpc_irqflag>>2) & 3)
++ {
++ case 0: /* conforms, ie. bus-type dependent */
++ {
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ {
++ trigger = default_ISA_trigger(idx);
++ break;
++ }
++ case MP_BUS_EISA: /* EISA pin */
++ {
++ trigger = default_EISA_trigger(idx);
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ trigger = default_PCI_trigger(idx);
++ break;
++ }
++ case MP_BUS_MCA: /* MCA pin */
++ {
++ trigger = default_MCA_trigger(idx);
++ break;
++ }
++ default:
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 1;
++ break;
++ }
++ }
++ break;
++ }
++ case 1: /* edge */
++ {
++ trigger = 0;
++ break;
++ }
++ case 2: /* reserved */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 1;
++ break;
++ }
++ case 3: /* level */
++ {
++ trigger = 1;
++ break;
++ }
++ default: /* invalid */
++ {
++ printk(KERN_WARNING "broken BIOS!!\n");
++ trigger = 0;
++ break;
++ }
++ }
++ return trigger;
++}
++
++static inline int irq_polarity(int idx)
++{
++ return MPBIOS_polarity(idx);
++}
++
++static inline int irq_trigger(int idx)
++{
++ return MPBIOS_trigger(idx);
++}
++
++static int next_irq = 16;
++
++/*
++ * gsi_irq_sharing -- Name overload! "irq" can be either a legacy IRQ
++ * in the range 0-15, a linux IRQ in the range 0-223, or a GSI number
++ * from ACPI, which can reach 800 in large boxen.
++ *
++ * Compact the sparse GSI space into a sequential IRQ series and reuse
++ * vectors if possible.
++ */
++int gsi_irq_sharing(int gsi)
++{
++ int i, tries, vector;
++
++ BUG_ON(gsi >= NR_IRQ_VECTORS);
++
++ if (platform_legacy_irq(gsi))
++ return gsi;
++
++ if (gsi_2_irq[gsi] != 0xFF)
++ return (int)gsi_2_irq[gsi];
++
++ tries = NR_IRQS;
++ try_again:
++ vector = assign_irq_vector(gsi);
++
++ /*
++ * Sharing vectors means sharing IRQs, so scan irq_vectors for previous
++ * use of vector and if found, return that IRQ. However, we never want
++ * to share legacy IRQs, which usually have a different trigger mode
++ * than PCI.
++ */
++ for (i = 0; i < NR_IRQS; i++)
++ if (IO_APIC_VECTOR(i) == vector)
++ break;
++ if (platform_legacy_irq(i)) {
++ if (--tries >= 0) {
++ IO_APIC_VECTOR(i) = 0;
++ goto try_again;
++ }
++ panic("gsi_irq_sharing: didn't find an IRQ using vector 0x%02X for GSI %d", vector, gsi);
++ }
++ if (i < NR_IRQS) {
++ gsi_2_irq[gsi] = i;
++ printk(KERN_INFO "GSI %d sharing vector 0x%02X and IRQ %d\n",
++ gsi, vector, i);
++ return i;
++ }
++
++ i = next_irq++;
++ BUG_ON(i >= NR_IRQS);
++ gsi_2_irq[gsi] = i;
++ IO_APIC_VECTOR(i) = vector;
++ printk(KERN_INFO "GSI %d assigned vector 0x%02X and IRQ %d\n",
++ gsi, vector, i);
++ return i;
++}
++
++static int pin_2_irq(int idx, int apic, int pin)
++{
++ int irq, i;
++ int bus = mp_irqs[idx].mpc_srcbus;
++
++ /*
++ * Debugging check, we are in big trouble if this message pops up!
++ */
++ if (mp_irqs[idx].mpc_dstirq != pin)
++ printk(KERN_ERR "broken BIOS or MPTABLE parser, ayiee!!\n");
++
++ switch (mp_bus_id_to_type[bus])
++ {
++ case MP_BUS_ISA: /* ISA pin */
++ case MP_BUS_EISA:
++ case MP_BUS_MCA:
++ {
++ irq = mp_irqs[idx].mpc_srcbusirq;
++ break;
++ }
++ case MP_BUS_PCI: /* PCI pin */
++ {
++ /*
++ * PCI IRQs are mapped in order
++ */
++ i = irq = 0;
++ while (i < apic)
++ irq += nr_ioapic_registers[i++];
++ irq += pin;
++ irq = gsi_irq_sharing(irq);
++ break;
++ }
++ default:
++ {
++ printk(KERN_ERR "unknown bus type %d.\n",bus);
++ irq = 0;
++ break;
++ }
++ }
++ BUG_ON(irq >= NR_IRQS);
++
++ /*
++ * PCI IRQ command line redirection. Yes, limits are hardcoded.
++ */
++ if ((pin >= 16) && (pin <= 23)) {
++ if (pirq_entries[pin-16] != -1) {
++ if (!pirq_entries[pin-16]) {
++ apic_printk(APIC_VERBOSE, "disabling PIRQ%d\n", pin-16);
++ } else {
++ irq = pirq_entries[pin-16];
++ apic_printk(APIC_VERBOSE, "using PIRQ%d -> IRQ %d\n",
++ pin-16, irq);
++ }
++ }
++ }
++ BUG_ON(irq >= NR_IRQS);
++ return irq;
++}
++
++static inline int IO_APIC_irq_trigger(int irq)
++{
++ int apic, idx, pin;
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++ idx = find_irq_entry(apic,pin,mp_INT);
++ if ((idx != -1) && (irq == pin_2_irq(idx,apic,pin)))
++ return irq_trigger(idx);
++ }
++ }
++ /*
++ * nonexistent IRQs are edge default
++ */
++ return 0;
++}
++
++/* irq_vectors is indexed by the sum of all RTEs in all I/O APICs. */
++u8 irq_vector[NR_IRQ_VECTORS] __read_mostly;
++
++int assign_irq_vector(int irq)
++{
++ struct physdev_irq irq_op;
++
++ BUG_ON(irq != AUTO_ASSIGN && (unsigned)irq >= NR_IRQ_VECTORS);
++ if (irq != AUTO_ASSIGN && IO_APIC_VECTOR(irq) > 0)
++ return IO_APIC_VECTOR(irq);
++
++ irq_op.irq = irq;
++ if (HYPERVISOR_physdev_op(PHYSDEVOP_alloc_irq_vector, &irq_op))
++ return -ENOSPC;
++
++ vector_irq[irq_op.vector] = irq;
++ if (irq != AUTO_ASSIGN)
++ IO_APIC_VECTOR(irq) = irq_op.vector;
++
++ return irq_op.vector;
++}
++
++extern void (*interrupt[NR_IRQS])(void);
++#ifndef CONFIG_XEN
++static struct hw_interrupt_type ioapic_level_type;
++static struct hw_interrupt_type ioapic_edge_type;
++
++#define IOAPIC_AUTO -1
++#define IOAPIC_EDGE 0
++#define IOAPIC_LEVEL 1
++
++static inline void ioapic_register_intr(int irq, int vector, unsigned long trigger)
++{
++ if (use_pci_vector() && !platform_legacy_irq(irq)) {
++ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
++ trigger == IOAPIC_LEVEL)
++ irq_desc[vector].handler = &ioapic_level_type;
++ else
++ irq_desc[vector].handler = &ioapic_edge_type;
++ set_intr_gate(vector, interrupt[vector]);
++ } else {
++ if ((trigger == IOAPIC_AUTO && IO_APIC_irq_trigger(irq)) ||
++ trigger == IOAPIC_LEVEL)
++ irq_desc[irq].handler = &ioapic_level_type;
++ else
++ irq_desc[irq].handler = &ioapic_edge_type;
++ set_intr_gate(vector, interrupt[irq]);
++ }
++}
++#else
++#define ioapic_register_intr(_irq,_vector,_trigger) ((void)0)
++#endif /* !CONFIG_XEN */
++
++static void __init setup_IO_APIC_irqs(void)
++{
++ struct IO_APIC_route_entry entry;
++ int apic, pin, idx, irq, first_notcon = 1, vector;
++ unsigned long flags;
++
++ apic_printk(APIC_VERBOSE, KERN_DEBUG "init IO_APIC IRQs\n");
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++
++ /*
++ * add it to the IO-APIC irq-routing table:
++ */
++ memset(&entry,0,sizeof(entry));
++
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.dest_mode = INT_DEST_MODE;
++ entry.mask = 0; /* enable IRQ */
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++
++ idx = find_irq_entry(apic,pin,mp_INT);
++ if (idx == -1) {
++ if (first_notcon) {
++ apic_printk(APIC_VERBOSE, KERN_DEBUG " IO-APIC (apicid-pin) %d-%d", mp_ioapics[apic].mpc_apicid, pin);
++ first_notcon = 0;
++ } else
++ apic_printk(APIC_VERBOSE, ", %d-%d", mp_ioapics[apic].mpc_apicid, pin);
++ continue;
++ }
++
++ entry.trigger = irq_trigger(idx);
++ entry.polarity = irq_polarity(idx);
++
++ if (irq_trigger(idx)) {
++ entry.trigger = 1;
++ entry.mask = 1;
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++ }
++
++ irq = pin_2_irq(idx, apic, pin);
++ add_pin_to_irq(irq, apic, pin);
++
++ if (/* !apic && */ !IO_APIC_IRQ(irq))
++ continue;
++
++ if (IO_APIC_IRQ(irq)) {
++ vector = assign_irq_vector(irq);
++ entry.vector = vector;
++
++ ioapic_register_intr(irq, vector, IOAPIC_AUTO);
++ if (!apic && (irq < 16))
++ disable_8259A_irq(irq);
++ }
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
++ set_native_irq_info(irq, TARGET_CPUS);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ }
++ }
++
++ if (!first_notcon)
++ apic_printk(APIC_VERBOSE," not connected.\n");
++}
++
++#ifndef CONFIG_XEN
++/*
++ * Set up the 8259A-master output pin as broadcast to all
++ * CPUs.
++ */
++static void __init setup_ExtINT_IRQ0_pin(unsigned int apic, unsigned int pin, int vector)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ memset(&entry,0,sizeof(entry));
++
++ disable_8259A_irq(0);
++
++ /* mask LVT0 */
++ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
++
++ /*
++ * We use logical delivery to get the timer IRQ
++ * to the first CPU.
++ */
++ entry.dest_mode = INT_DEST_MODE;
++ entry.mask = 0; /* unmask IRQ now */
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.polarity = 0;
++ entry.trigger = 0;
++ entry.vector = vector;
++
++ /*
++ * The timer IRQ doesn't have to know that behind the
++ * scene we have a 8259A-master in AEOI mode ...
++ */
++ irq_desc[0].handler = &ioapic_edge_type;
++
++ /*
++ * Add it to the IO-APIC irq-routing table:
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(apic, 0x10+2*pin, *(((int *)&entry)+0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ enable_8259A_irq(0);
++}
++
++void __init UNEXPECTED_IO_APIC(void)
++{
++}
++
++void __apicdebuginit print_IO_APIC(void)
++{
++ int apic, i;
++ union IO_APIC_reg_00 reg_00;
++ union IO_APIC_reg_01 reg_01;
++ union IO_APIC_reg_02 reg_02;
++ unsigned long flags;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "number of MP IRQ sources: %d.\n", mp_irq_entries);
++ for (i = 0; i < nr_ioapics; i++)
++ printk(KERN_DEBUG "number of IO-APIC #%d registers: %d.\n",
++ mp_ioapics[i].mpc_apicid, nr_ioapic_registers[i]);
++
++ /*
++ * We are a bit conservative about what we expect. We have to
++ * know about every hardware change ASAP.
++ */
++ printk(KERN_INFO "testing the IO APIC.......................\n");
++
++ for (apic = 0; apic < nr_ioapics; apic++) {
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ reg_01.raw = io_apic_read(apic, 1);
++ if (reg_01.bits.version >= 0x10)
++ reg_02.raw = io_apic_read(apic, 2);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ printk("\n");
++ printk(KERN_DEBUG "IO APIC #%d......\n", mp_ioapics[apic].mpc_apicid);
++ printk(KERN_DEBUG ".... register #00: %08X\n", reg_00.raw);
++ printk(KERN_DEBUG "....... : physical APIC id: %02X\n", reg_00.bits.ID);
++ if (reg_00.bits.__reserved_1 || reg_00.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++
++ printk(KERN_DEBUG ".... register #01: %08X\n", *(int *)®_01);
++ printk(KERN_DEBUG "....... : max redirection entries: %04X\n", reg_01.bits.entries);
++ if ( (reg_01.bits.entries != 0x0f) && /* older (Neptune) boards */
++ (reg_01.bits.entries != 0x17) && /* typical ISA+PCI boards */
++ (reg_01.bits.entries != 0x1b) && /* Compaq Proliant boards */
++ (reg_01.bits.entries != 0x1f) && /* dual Xeon boards */
++ (reg_01.bits.entries != 0x22) && /* bigger Xeon boards */
++ (reg_01.bits.entries != 0x2E) &&
++ (reg_01.bits.entries != 0x3F) &&
++ (reg_01.bits.entries != 0x03)
++ )
++ UNEXPECTED_IO_APIC();
++
++ printk(KERN_DEBUG "....... : PRQ implemented: %X\n", reg_01.bits.PRQ);
++ printk(KERN_DEBUG "....... : IO APIC version: %04X\n", reg_01.bits.version);
++ if ( (reg_01.bits.version != 0x01) && /* 82489DX IO-APICs */
++ (reg_01.bits.version != 0x02) && /* 82801BA IO-APICs (ICH2) */
++ (reg_01.bits.version != 0x10) && /* oldest IO-APICs */
++ (reg_01.bits.version != 0x11) && /* Pentium/Pro IO-APICs */
++ (reg_01.bits.version != 0x13) && /* Xeon IO-APICs */
++ (reg_01.bits.version != 0x20) /* Intel P64H (82806 AA) */
++ )
++ UNEXPECTED_IO_APIC();
++ if (reg_01.bits.__reserved_1 || reg_01.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++
++ if (reg_01.bits.version >= 0x10) {
++ printk(KERN_DEBUG ".... register #02: %08X\n", reg_02.raw);
++ printk(KERN_DEBUG "....... : arbitration: %02X\n", reg_02.bits.arbitration);
++ if (reg_02.bits.__reserved_1 || reg_02.bits.__reserved_2)
++ UNEXPECTED_IO_APIC();
++ }
++
++ printk(KERN_DEBUG ".... IRQ redirection table:\n");
++
++ printk(KERN_DEBUG " NR Log Phy Mask Trig IRR Pol"
++ " Stat Dest Deli Vect: \n");
++
++ for (i = 0; i <= reg_01.bits.entries; i++) {
++ struct IO_APIC_route_entry entry;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry)+0) = io_apic_read(apic, 0x10+i*2);
++ *(((int *)&entry)+1) = io_apic_read(apic, 0x11+i*2);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ printk(KERN_DEBUG " %02x %03X %02X ",
++ i,
++ entry.dest.logical.logical_dest,
++ entry.dest.physical.physical_dest
++ );
++
++ printk("%1d %1d %1d %1d %1d %1d %1d %02X\n",
++ entry.mask,
++ entry.trigger,
++ entry.irr,
++ entry.polarity,
++ entry.delivery_status,
++ entry.dest_mode,
++ entry.delivery_mode,
++ entry.vector
++ );
++ }
++ }
++ if (use_pci_vector())
++ printk(KERN_INFO "Using vector-based indexing\n");
++ printk(KERN_DEBUG "IRQ to pin mappings:\n");
++ for (i = 0; i < NR_IRQS; i++) {
++ struct irq_pin_list *entry = irq_2_pin + i;
++ if (entry->pin < 0)
++ continue;
++ if (use_pci_vector() && !platform_legacy_irq(i))
++ printk(KERN_DEBUG "IRQ%d ", IO_APIC_VECTOR(i));
++ else
++ printk(KERN_DEBUG "IRQ%d ", i);
++ for (;;) {
++ printk("-> %d:%d", entry->apic, entry->pin);
++ if (!entry->next)
++ break;
++ entry = irq_2_pin + entry->next;
++ }
++ printk("\n");
++ }
++
++ printk(KERN_INFO ".................................... done.\n");
++
++ return;
++}
++
++#if 0
++
++static __apicdebuginit void print_APIC_bitfield (int base)
++{
++ unsigned int v;
++ int i, j;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "0123456789abcdef0123456789abcdef\n" KERN_DEBUG);
++ for (i = 0; i < 8; i++) {
++ v = apic_read(base + i*0x10);
++ for (j = 0; j < 32; j++) {
++ if (v & (1<<j))
++ printk("1");
++ else
++ printk("0");
++ }
++ printk("\n");
++ }
++}
++
++void __apicdebuginit print_local_APIC(void * dummy)
++{
++ unsigned int v, ver, maxlvt;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk("\n" KERN_DEBUG "printing local APIC contents on CPU#%d/%d:\n",
++ smp_processor_id(), hard_smp_processor_id());
++ v = apic_read(APIC_ID);
++ printk(KERN_INFO "... APIC ID: %08x (%01x)\n", v, GET_APIC_ID(v));
++ v = apic_read(APIC_LVR);
++ printk(KERN_INFO "... APIC VERSION: %08x\n", v);
++ ver = GET_APIC_VERSION(v);
++ maxlvt = get_maxlvt();
++
++ v = apic_read(APIC_TASKPRI);
++ printk(KERN_DEBUG "... APIC TASKPRI: %08x (%02x)\n", v, v & APIC_TPRI_MASK);
++
++ v = apic_read(APIC_ARBPRI);
++ printk(KERN_DEBUG "... APIC ARBPRI: %08x (%02x)\n", v,
++ v & APIC_ARBPRI_MASK);
++ v = apic_read(APIC_PROCPRI);
++ printk(KERN_DEBUG "... APIC PROCPRI: %08x\n", v);
++
++ v = apic_read(APIC_EOI);
++ printk(KERN_DEBUG "... APIC EOI: %08x\n", v);
++ v = apic_read(APIC_RRR);
++ printk(KERN_DEBUG "... APIC RRR: %08x\n", v);
++ v = apic_read(APIC_LDR);
++ printk(KERN_DEBUG "... APIC LDR: %08x\n", v);
++ v = apic_read(APIC_DFR);
++ printk(KERN_DEBUG "... APIC DFR: %08x\n", v);
++ v = apic_read(APIC_SPIV);
++ printk(KERN_DEBUG "... APIC SPIV: %08x\n", v);
++
++ printk(KERN_DEBUG "... APIC ISR field:\n");
++ print_APIC_bitfield(APIC_ISR);
++ printk(KERN_DEBUG "... APIC TMR field:\n");
++ print_APIC_bitfield(APIC_TMR);
++ printk(KERN_DEBUG "... APIC IRR field:\n");
++ print_APIC_bitfield(APIC_IRR);
++
++ v = apic_read(APIC_ESR);
++ printk(KERN_DEBUG "... APIC ESR: %08x\n", v);
++
++ v = apic_read(APIC_ICR);
++ printk(KERN_DEBUG "... APIC ICR: %08x\n", v);
++ v = apic_read(APIC_ICR2);
++ printk(KERN_DEBUG "... APIC ICR2: %08x\n", v);
++
++ v = apic_read(APIC_LVTT);
++ printk(KERN_DEBUG "... APIC LVTT: %08x\n", v);
++
++ if (maxlvt > 3) { /* PC is LVT#4. */
++ v = apic_read(APIC_LVTPC);
++ printk(KERN_DEBUG "... APIC LVTPC: %08x\n", v);
++ }
++ v = apic_read(APIC_LVT0);
++ printk(KERN_DEBUG "... APIC LVT0: %08x\n", v);
++ v = apic_read(APIC_LVT1);
++ printk(KERN_DEBUG "... APIC LVT1: %08x\n", v);
++
++ if (maxlvt > 2) { /* ERR is LVT#3. */
++ v = apic_read(APIC_LVTERR);
++ printk(KERN_DEBUG "... APIC LVTERR: %08x\n", v);
++ }
++
++ v = apic_read(APIC_TMICT);
++ printk(KERN_DEBUG "... APIC TMICT: %08x\n", v);
++ v = apic_read(APIC_TMCCT);
++ printk(KERN_DEBUG "... APIC TMCCT: %08x\n", v);
++ v = apic_read(APIC_TDCR);
++ printk(KERN_DEBUG "... APIC TDCR: %08x\n", v);
++ printk("\n");
++}
++
++void print_all_local_APICs (void)
++{
++ on_each_cpu(print_local_APIC, NULL, 1, 1);
++}
++
++void __apicdebuginit print_PIC(void)
++{
++ unsigned int v;
++ unsigned long flags;
++
++ if (apic_verbosity == APIC_QUIET)
++ return;
++
++ printk(KERN_DEBUG "\nprinting PIC contents\n");
++
++ spin_lock_irqsave(&i8259A_lock, flags);
++
++ v = inb(0xa1) << 8 | inb(0x21);
++ printk(KERN_DEBUG "... PIC IMR: %04x\n", v);
++
++ v = inb(0xa0) << 8 | inb(0x20);
++ printk(KERN_DEBUG "... PIC IRR: %04x\n", v);
++
++ outb(0x0b,0xa0);
++ outb(0x0b,0x20);
++ v = inb(0xa0) << 8 | inb(0x20);
++ outb(0x0a,0xa0);
++ outb(0x0a,0x20);
++
++ spin_unlock_irqrestore(&i8259A_lock, flags);
++
++ printk(KERN_DEBUG "... PIC ISR: %04x\n", v);
++
++ v = inb(0x4d1) << 8 | inb(0x4d0);
++ printk(KERN_DEBUG "... PIC ELCR: %04x\n", v);
++}
++
++#endif /* 0 */
++
++#else
++void __init print_IO_APIC(void) { }
++#endif /* !CONFIG_XEN */
++
++static void __init enable_IO_APIC(void)
++{
++ union IO_APIC_reg_01 reg_01;
++#ifndef CONFIG_XEN
++ int i8259_apic, i8259_pin;
++#endif
++ int i, apic;
++ unsigned long flags;
++
++ for (i = 0; i < PIN_MAP_SIZE; i++) {
++ irq_2_pin[i].pin = -1;
++ irq_2_pin[i].next = 0;
++ }
++ if (!pirqs_enabled)
++ for (i = 0; i < MAX_PIRQS; i++)
++ pirq_entries[i] = -1;
++
++ /*
++ * The number of IO-APIC IRQ registers (== #pins):
++ */
++ for (apic = 0; apic < nr_ioapics; apic++) {
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(apic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ nr_ioapic_registers[apic] = reg_01.bits.entries+1;
++ }
++#ifndef CONFIG_XEN
++ for(apic = 0; apic < nr_ioapics; apic++) {
++ int pin;
++ /* See if any of the pins is in ExtINT mode */
++ for (pin = 0; pin < nr_ioapic_registers[apic]; pin++) {
++ struct IO_APIC_route_entry entry;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ *(((int *)&entry) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++
++ /* If the interrupt line is enabled and in ExtInt mode
++ * I have found the pin where the i8259 is connected.
++ */
++ if ((entry.mask == 0) && (entry.delivery_mode == dest_ExtINT)) {
++ ioapic_i8259.apic = apic;
++ ioapic_i8259.pin = pin;
++ goto found_i8259;
++ }
++ }
++ }
++ found_i8259:
++ /* Look to see what if the MP table has reported the ExtINT */
++ i8259_pin = find_isa_irq_pin(0, mp_ExtINT);
++ i8259_apic = find_isa_irq_apic(0, mp_ExtINT);
++ /* Trust the MP table if nothing is setup in the hardware */
++ if ((ioapic_i8259.pin == -1) && (i8259_pin >= 0)) {
++ printk(KERN_WARNING "ExtINT not setup in hardware but reported by MP table\n");
++ ioapic_i8259.pin = i8259_pin;
++ ioapic_i8259.apic = i8259_apic;
++ }
++ /* Complain if the MP table and the hardware disagree */
++ if (((ioapic_i8259.apic != i8259_apic) || (ioapic_i8259.pin != i8259_pin)) &&
++ (i8259_pin >= 0) && (ioapic_i8259.pin >= 0))
++ {
++ printk(KERN_WARNING "ExtINT in hardware and MP table differ\n");
++ }
++#endif
++
++ /*
++ * Do not trust the IO-APIC being empty at bootup
++ */
++ clear_IO_APIC();
++}
++
++/*
++ * Not an __init, needed by the reboot code
++ */
++void disable_IO_APIC(void)
++{
++ /*
++ * Clear the IO-APIC before rebooting:
++ */
++ clear_IO_APIC();
++
++#ifndef CONFIG_XEN
++ /*
++ * If the i8259 is routed through an IOAPIC
++ * Put that IOAPIC in virtual wire mode
++ * so legacy interrupts can be delivered.
++ */
++ if (ioapic_i8259.pin != -1) {
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ memset(&entry, 0, sizeof(entry));
++ entry.mask = 0; /* Enabled */
++ entry.trigger = 0; /* Edge */
++ entry.irr = 0;
++ entry.polarity = 0; /* High */
++ entry.delivery_status = 0;
++ entry.dest_mode = 0; /* Physical */
++ entry.delivery_mode = dest_ExtINT; /* ExtInt */
++ entry.vector = 0;
++ entry.dest.physical.physical_dest =
++ GET_APIC_ID(apic_read(APIC_ID));
++
++ /*
++ * Add it to the IO-APIC irq-routing table:
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(ioapic_i8259.apic, 0x11+2*ioapic_i8259.pin,
++ *(((int *)&entry)+1));
++ io_apic_write(ioapic_i8259.apic, 0x10+2*ioapic_i8259.pin,
++ *(((int *)&entry)+0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ }
++
++ disconnect_bsp_APIC(ioapic_i8259.pin != -1);
++#endif
++}
++
++/*
++ * function to set the IO-APIC physical IDs based on the
++ * values stored in the MPC table.
++ *
++ * by Matt Domsch <Matt_Domsch@dell.com> Tue Dec 21 12:25:05 CST 1999
++ */
++
++#ifndef CONFIG_XEN
++static void __init setup_ioapic_ids_from_mpc (void)
++{
++ union IO_APIC_reg_00 reg_00;
++ int apic;
++ int i;
++ unsigned char old_id;
++ unsigned long flags;
++
++ /*
++ * Set the IOAPIC ID to the value stored in the MPC table.
++ */
++ for (apic = 0; apic < nr_ioapics; apic++) {
++
++ /* Read the register 0 value */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ old_id = mp_ioapics[apic].mpc_apicid;
++
++
++ printk(KERN_INFO "Using IO-APIC %d\n", mp_ioapics[apic].mpc_apicid);
++
++
++ /*
++ * We need to adjust the IRQ routing table
++ * if the ID changed.
++ */
++ if (old_id != mp_ioapics[apic].mpc_apicid)
++ for (i = 0; i < mp_irq_entries; i++)
++ if (mp_irqs[i].mpc_dstapic == old_id)
++ mp_irqs[i].mpc_dstapic
++ = mp_ioapics[apic].mpc_apicid;
++
++ /*
++ * Read the right value from the MPC table and
++ * write it into the ID register.
++ */
++ apic_printk(APIC_VERBOSE,KERN_INFO "...changing IO-APIC physical APIC ID to %d ...",
++ mp_ioapics[apic].mpc_apicid);
++
++ reg_00.bits.ID = mp_ioapics[apic].mpc_apicid;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0, reg_00.raw);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ /*
++ * Sanity check
++ */
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(apic, 0);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ if (reg_00.bits.ID != mp_ioapics[apic].mpc_apicid)
++ printk("could not set ID!\n");
++ else
++ apic_printk(APIC_VERBOSE," ok.\n");
++ }
++}
++#else
++static void __init setup_ioapic_ids_from_mpc(void) { }
++#endif
++
++/*
++ * There is a nasty bug in some older SMP boards, their mptable lies
++ * about the timer IRQ. We do the following to work around the situation:
++ *
++ * - timer IRQ defaults to IO-APIC IRQ
++ * - if this function detects that timer IRQs are defunct, then we fall
++ * back to ISA timer IRQs
++ */
++#ifndef CONFIG_XEN
++static int __init timer_irq_works(void)
++{
++ unsigned long t1 = jiffies;
++
++ local_irq_enable();
++ /* Let ten ticks pass... */
++ mdelay((10 * 1000) / HZ);
++
++ /*
++ * Expect a few ticks at least, to be sure some possible
++ * glue logic does not lock up after one or two first
++ * ticks in a non-ExtINT mode. Also the local APIC
++ * might have cached one ExtINT interrupt. Finally, at
++ * least one tick may be lost due to delays.
++ */
++
++ /* jiffies wrap? */
++ if (jiffies - t1 > 4)
++ return 1;
++ return 0;
++}
++
++/*
++ * In the SMP+IOAPIC case it might happen that there are an unspecified
++ * number of pending IRQ events unhandled. These cases are very rare,
++ * so we 'resend' these IRQs via IPIs, to the same CPU. It's much
++ * better to do it this way as thus we do not have to be aware of
++ * 'pending' interrupts in the IRQ path, except at this point.
++ */
++/*
++ * Edge triggered needs to resend any interrupt
++ * that was delayed but this is now handled in the device
++ * independent code.
++ */
++
++/*
++ * Starting up a edge-triggered IO-APIC interrupt is
++ * nasty - we need to make sure that we get the edge.
++ * If it is already asserted for some reason, we need
++ * return 1 to indicate that is was pending.
++ *
++ * This is not complete - we should be able to fake
++ * an edge even if it isn't on the 8259A...
++ */
++
++static unsigned int startup_edge_ioapic_irq(unsigned int irq)
++{
++ int was_pending = 0;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ if (irq < 16) {
++ disable_8259A_irq(irq);
++ if (i8259A_irq_pending(irq))
++ was_pending = 1;
++ }
++ __unmask_IO_APIC_irq(irq);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return was_pending;
++}
++
++/*
++ * Once we have recorded IRQ_PENDING already, we can mask the
++ * interrupt for real. This prevents IRQ storms from unhandled
++ * devices.
++ */
++static void ack_edge_ioapic_irq(unsigned int irq)
++{
++ move_irq(irq);
++ if ((irq_desc[irq].status & (IRQ_PENDING | IRQ_DISABLED))
++ == (IRQ_PENDING | IRQ_DISABLED))
++ mask_IO_APIC_irq(irq);
++ ack_APIC_irq();
++}
++
++/*
++ * Level triggered interrupts can just be masked,
++ * and shutting down and starting up the interrupt
++ * is the same as enabling and disabling them -- except
++ * with a startup need to return a "was pending" value.
++ *
++ * Level triggered interrupts are special because we
++ * do not touch any IO-APIC register while handling
++ * them. We ack the APIC in the end-IRQ handler, not
++ * in the start-IRQ-handler. Protection against reentrance
++ * from the same interrupt is still provided, both by the
++ * generic IRQ layer and by the fact that an unacked local
++ * APIC does not accept IRQs.
++ */
++static unsigned int startup_level_ioapic_irq (unsigned int irq)
++{
++ unmask_IO_APIC_irq(irq);
++
++ return 0; /* don't check for pending */
++}
++
++static void end_level_ioapic_irq (unsigned int irq)
++{
++ move_irq(irq);
++ ack_APIC_irq();
++}
++
++#ifdef CONFIG_PCI_MSI
++static unsigned int startup_edge_ioapic_vector(unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ return startup_edge_ioapic_irq(irq);
++}
++
++static void ack_edge_ioapic_vector(unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ move_native_irq(vector);
++ ack_edge_ioapic_irq(irq);
++}
++
++static unsigned int startup_level_ioapic_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ return startup_level_ioapic_irq (irq);
++}
++
++static void end_level_ioapic_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ move_native_irq(vector);
++ end_level_ioapic_irq(irq);
++}
++
++static void mask_IO_APIC_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ mask_IO_APIC_irq(irq);
++}
++
++static void unmask_IO_APIC_vector (unsigned int vector)
++{
++ int irq = vector_to_irq(vector);
++
++ unmask_IO_APIC_irq(irq);
++}
++
++#ifdef CONFIG_SMP
++static void set_ioapic_affinity_vector (unsigned int vector,
++ cpumask_t cpu_mask)
++{
++ int irq = vector_to_irq(vector);
++
++ set_native_irq_info(vector, cpu_mask);
++ set_ioapic_affinity_irq(irq, cpu_mask);
++}
++#endif // CONFIG_SMP
++#endif // CONFIG_PCI_MSI
++
++/*
++ * Level and edge triggered IO-APIC interrupts need different handling,
++ * so we use two separate IRQ descriptors. Edge triggered IRQs can be
++ * handled with the level-triggered descriptor, but that one has slightly
++ * more overhead. Level-triggered interrupts cannot be handled with the
++ * edge-triggered handler, without risking IRQ storms and other ugly
++ * races.
++ */
++
++static struct hw_interrupt_type ioapic_edge_type __read_mostly = {
++ .typename = "IO-APIC-edge",
++ .startup = startup_edge_ioapic,
++ .shutdown = shutdown_edge_ioapic,
++ .enable = enable_edge_ioapic,
++ .disable = disable_edge_ioapic,
++ .ack = ack_edge_ioapic,
++ .end = end_edge_ioapic,
++#ifdef CONFIG_SMP
++ .set_affinity = set_ioapic_affinity,
++#endif
++};
++
++static struct hw_interrupt_type ioapic_level_type __read_mostly = {
++ .typename = "IO-APIC-level",
++ .startup = startup_level_ioapic,
++ .shutdown = shutdown_level_ioapic,
++ .enable = enable_level_ioapic,
++ .disable = disable_level_ioapic,
++ .ack = mask_and_ack_level_ioapic,
++ .end = end_level_ioapic,
++#ifdef CONFIG_SMP
++ .set_affinity = set_ioapic_affinity,
++#endif
++};
++#endif /* !CONFIG_XEN */
++
++static inline void init_IO_APIC_traps(void)
++{
++ int irq;
++
++ /*
++ * NOTE! The local APIC isn't very good at handling
++ * multiple interrupts at the same interrupt level.
++ * As the interrupt level is determined by taking the
++ * vector number and shifting that right by 4, we
++ * want to spread these out a bit so that they don't
++ * all fall in the same interrupt level.
++ *
++ * Also, we've got to be careful not to trash gate
++ * 0x80, because int 0x80 is hm, kind of importantish. ;)
++ */
++ for (irq = 0; irq < NR_IRQS ; irq++) {
++ int tmp = irq;
++ if (use_pci_vector()) {
++ if (!platform_legacy_irq(tmp))
++ if ((tmp = vector_to_irq(tmp)) == -1)
++ continue;
++ }
++ if (IO_APIC_IRQ(tmp) && !IO_APIC_VECTOR(tmp)) {
++ /*
++ * Hmm.. We don't have an entry for this,
++ * so default to an old-fashioned 8259
++ * interrupt if we can..
++ */
++ if (irq < 16)
++ make_8259A_irq(irq);
++#ifndef CONFIG_XEN
++ else
++ /* Strange. Oh, well.. */
++ irq_desc[irq].handler = &no_irq_type;
++#endif
++ }
++ }
++}
++
++#ifndef CONFIG_XEN
++static void enable_lapic_irq (unsigned int irq)
++{
++ unsigned long v;
++
++ v = apic_read(APIC_LVT0);
++ apic_write(APIC_LVT0, v & ~APIC_LVT_MASKED);
++}
++
++static void disable_lapic_irq (unsigned int irq)
++{
++ unsigned long v;
++
++ v = apic_read(APIC_LVT0);
++ apic_write(APIC_LVT0, v | APIC_LVT_MASKED);
++}
++
++static void ack_lapic_irq (unsigned int irq)
++{
++ ack_APIC_irq();
++}
++
++static void end_lapic_irq (unsigned int i) { /* nothing */ }
++
++static struct hw_interrupt_type lapic_irq_type __read_mostly = {
++ .typename = "local-APIC-edge",
++ .startup = NULL, /* startup_irq() not used for IRQ0 */
++ .shutdown = NULL, /* shutdown_irq() not used for IRQ0 */
++ .enable = enable_lapic_irq,
++ .disable = disable_lapic_irq,
++ .ack = ack_lapic_irq,
++ .end = end_lapic_irq,
++};
++
++static void setup_nmi (void)
++{
++ /*
++ * Dirty trick to enable the NMI watchdog ...
++ * We put the 8259A master into AEOI mode and
++ * unmask on all local APICs LVT0 as NMI.
++ *
++ * The idea to use the 8259A in AEOI mode ('8259A Virtual Wire')
++ * is from Maciej W. Rozycki - so we do not have to EOI from
++ * the NMI handler or the timer interrupt.
++ */
++ printk(KERN_INFO "activating NMI Watchdog ...");
++
++ enable_NMI_through_LVT0(NULL);
++
++ printk(" done.\n");
++}
++
++/*
++ * This looks a bit hackish but it's about the only one way of sending
++ * a few INTA cycles to 8259As and any associated glue logic. ICR does
++ * not support the ExtINT mode, unfortunately. We need to send these
++ * cycles as some i82489DX-based boards have glue logic that keeps the
++ * 8259A interrupt line asserted until INTA. --macro
++ */
++static inline void unlock_ExtINT_logic(void)
++{
++ int apic, pin, i;
++ struct IO_APIC_route_entry entry0, entry1;
++ unsigned char save_control, save_freq_select;
++ unsigned long flags;
++
++ pin = find_isa_irq_pin(8, mp_INT);
++ apic = find_isa_irq_apic(8, mp_INT);
++ if (pin == -1)
++ return;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ *(((int *)&entry0) + 1) = io_apic_read(apic, 0x11 + 2 * pin);
++ *(((int *)&entry0) + 0) = io_apic_read(apic, 0x10 + 2 * pin);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++ clear_IO_APIC_pin(apic, pin);
++
++ memset(&entry1, 0, sizeof(entry1));
++
++ entry1.dest_mode = 0; /* physical delivery */
++ entry1.mask = 0; /* unmask IRQ now */
++ entry1.dest.physical.physical_dest = hard_smp_processor_id();
++ entry1.delivery_mode = dest_ExtINT;
++ entry1.polarity = entry0.polarity;
++ entry1.trigger = 0;
++ entry1.vector = 0;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry1) + 1));
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry1) + 0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ save_control = CMOS_READ(RTC_CONTROL);
++ save_freq_select = CMOS_READ(RTC_FREQ_SELECT);
++ CMOS_WRITE((save_freq_select & ~RTC_RATE_SELECT) | 0x6,
++ RTC_FREQ_SELECT);
++ CMOS_WRITE(save_control | RTC_PIE, RTC_CONTROL);
++
++ i = 100;
++ while (i-- > 0) {
++ mdelay(10);
++ if ((CMOS_READ(RTC_INTR_FLAGS) & RTC_PF) == RTC_PF)
++ i -= 10;
++ }
++
++ CMOS_WRITE(save_control, RTC_CONTROL);
++ CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
++ clear_IO_APIC_pin(apic, pin);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(apic, 0x11 + 2 * pin, *(((int *)&entry0) + 1));
++ io_apic_write(apic, 0x10 + 2 * pin, *(((int *)&entry0) + 0));
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++}
++
++int timer_uses_ioapic_pin_0;
++
++/*
++ * This code may look a bit paranoid, but it's supposed to cooperate with
++ * a wide range of boards and BIOS bugs. Fortunately only the timer IRQ
++ * is so screwy. Thanks to Brian Perkins for testing/hacking this beast
++ * fanatically on his truly buggy board.
++ *
++ * FIXME: really need to revamp this for modern platforms only.
++ */
++static inline void check_timer(void)
++{
++ int apic1, pin1, apic2, pin2;
++ int vector;
++
++ /*
++ * get/set the timer IRQ vector:
++ */
++ disable_8259A_irq(0);
++ vector = assign_irq_vector(0);
++ set_intr_gate(vector, interrupt[0]);
++
++ /*
++ * Subtle, code in do_timer_interrupt() expects an AEOI
++ * mode for the 8259A whenever interrupts are routed
++ * through I/O APICs. Also IRQ0 has to be enabled in
++ * the 8259A which implies the virtual wire has to be
++ * disabled in the local APIC.
++ */
++ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_EXTINT);
++ init_8259A(1);
++ if (timer_over_8254 > 0)
++ enable_8259A_irq(0);
++
++ pin1 = find_isa_irq_pin(0, mp_INT);
++ apic1 = find_isa_irq_apic(0, mp_INT);
++ pin2 = ioapic_i8259.pin;
++ apic2 = ioapic_i8259.apic;
++
++ if (pin1 == 0)
++ timer_uses_ioapic_pin_0 = 1;
++
++ apic_printk(APIC_VERBOSE,KERN_INFO "..TIMER: vector=0x%02X apic1=%d pin1=%d apic2=%d pin2=%d\n",
++ vector, apic1, pin1, apic2, pin2);
++
++ if (pin1 != -1) {
++ /*
++ * Ok, does IRQ0 through the IOAPIC work?
++ */
++ unmask_IO_APIC_irq(0);
++ if (!no_timer_check && timer_irq_works()) {
++ nmi_watchdog_default();
++ if (nmi_watchdog == NMI_IO_APIC) {
++ disable_8259A_irq(0);
++ setup_nmi();
++ enable_8259A_irq(0);
++ }
++ if (disable_timer_pin_1 > 0)
++ clear_IO_APIC_pin(0, pin1);
++ return;
++ }
++ clear_IO_APIC_pin(apic1, pin1);
++ apic_printk(APIC_QUIET,KERN_ERR "..MP-BIOS bug: 8254 timer not "
++ "connected to IO-APIC\n");
++ }
++
++ apic_printk(APIC_VERBOSE,KERN_INFO "...trying to set up timer (IRQ0) "
++ "through the 8259A ... ");
++ if (pin2 != -1) {
++ apic_printk(APIC_VERBOSE,"\n..... (found apic %d pin %d) ...",
++ apic2, pin2);
++ /*
++ * legacy devices should be connected to IO APIC #0
++ */
++ setup_ExtINT_IRQ0_pin(apic2, pin2, vector);
++ if (timer_irq_works()) {
++ apic_printk(APIC_VERBOSE," works.\n");
++ nmi_watchdog_default();
++ if (nmi_watchdog == NMI_IO_APIC) {
++ setup_nmi();
++ }
++ return;
++ }
++ /*
++ * Cleanup, just in case ...
++ */
++ clear_IO_APIC_pin(apic2, pin2);
++ }
++ apic_printk(APIC_VERBOSE," failed.\n");
++
++ if (nmi_watchdog == NMI_IO_APIC) {
++ printk(KERN_WARNING "timer doesn't work through the IO-APIC - disabling NMI Watchdog!\n");
++ nmi_watchdog = 0;
++ }
++
++ apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as Virtual Wire IRQ...");
++
++ disable_8259A_irq(0);
++ irq_desc[0].handler = &lapic_irq_type;
++ apic_write(APIC_LVT0, APIC_DM_FIXED | vector); /* Fixed mode */
++ enable_8259A_irq(0);
++
++ if (timer_irq_works()) {
++ apic_printk(APIC_VERBOSE," works.\n");
++ return;
++ }
++ apic_write(APIC_LVT0, APIC_LVT_MASKED | APIC_DM_FIXED | vector);
++ apic_printk(APIC_VERBOSE," failed.\n");
++
++ apic_printk(APIC_VERBOSE, KERN_INFO "...trying to set up timer as ExtINT IRQ...");
++
++ init_8259A(0);
++ make_8259A_irq(0);
++ apic_write(APIC_LVT0, APIC_DM_EXTINT);
++
++ unlock_ExtINT_logic();
++
++ if (timer_irq_works()) {
++ apic_printk(APIC_VERBOSE," works.\n");
++ return;
++ }
++ apic_printk(APIC_VERBOSE," failed :(.\n");
++ panic("IO-APIC + timer doesn't work! Try using the 'noapic' kernel parameter\n");
++}
++#else
++int timer_uses_ioapic_pin_0;
++#define check_timer() ((void)0)
++#endif /* !CONFIG_XEN */
++
++static int __init notimercheck(char *s)
++{
++ no_timer_check = 1;
++ return 1;
++}
++__setup("no_timer_check", notimercheck);
++
++/*
++ *
++ * IRQ's that are handled by the PIC in the MPS IOAPIC case.
++ * - IRQ2 is the cascade IRQ, and cannot be a io-apic IRQ.
++ * Linux doesn't really care, as it's not actually used
++ * for any interrupt handling anyway.
++ */
++#define PIC_IRQS (1<<2)
++
++void __init setup_IO_APIC(void)
++{
++ enable_IO_APIC();
++
++ if (acpi_ioapic)
++ io_apic_irqs = ~0; /* all IRQs go through IOAPIC */
++ else
++ io_apic_irqs = ~PIC_IRQS;
++
++ apic_printk(APIC_VERBOSE, "ENABLING IO-APIC IRQs\n");
++
++ /*
++ * Set up the IO-APIC IRQ routing table.
++ */
++ if (!acpi_ioapic)
++ setup_ioapic_ids_from_mpc();
++#ifndef CONFIG_XEN
++ sync_Arb_IDs();
++#endif /* !CONFIG_XEN */
++ setup_IO_APIC_irqs();
++ init_IO_APIC_traps();
++ check_timer();
++ if (!acpi_ioapic)
++ print_IO_APIC();
++}
++
++struct sysfs_ioapic_data {
++ struct sys_device dev;
++ struct IO_APIC_route_entry entry[0];
++};
++static struct sysfs_ioapic_data * mp_ioapic_data[MAX_IO_APICS];
++
++static int ioapic_suspend(struct sys_device *dev, pm_message_t state)
++{
++ struct IO_APIC_route_entry *entry;
++ struct sysfs_ioapic_data *data;
++ unsigned long flags;
++ int i;
++
++ data = container_of(dev, struct sysfs_ioapic_data, dev);
++ entry = data->entry;
++ spin_lock_irqsave(&ioapic_lock, flags);
++ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
++ *(((int *)entry) + 1) = io_apic_read(dev->id, 0x11 + 2 * i);
++ *(((int *)entry) + 0) = io_apic_read(dev->id, 0x10 + 2 * i);
++ }
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++static int ioapic_resume(struct sys_device *dev)
++{
++ struct IO_APIC_route_entry *entry;
++ struct sysfs_ioapic_data *data;
++ unsigned long flags;
++ union IO_APIC_reg_00 reg_00;
++ int i;
++
++ data = container_of(dev, struct sysfs_ioapic_data, dev);
++ entry = data->entry;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_00.raw = io_apic_read(dev->id, 0);
++ if (reg_00.bits.ID != mp_ioapics[dev->id].mpc_apicid) {
++ reg_00.bits.ID = mp_ioapics[dev->id].mpc_apicid;
++ io_apic_write(dev->id, 0, reg_00.raw);
++ }
++ for (i = 0; i < nr_ioapic_registers[dev->id]; i ++, entry ++ ) {
++ io_apic_write(dev->id, 0x11+2*i, *(((int *)entry)+1));
++ io_apic_write(dev->id, 0x10+2*i, *(((int *)entry)+0));
++ }
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++static struct sysdev_class ioapic_sysdev_class = {
++ set_kset_name("ioapic"),
++ .suspend = ioapic_suspend,
++ .resume = ioapic_resume,
++};
++
++static int __init ioapic_init_sysfs(void)
++{
++ struct sys_device * dev;
++ int i, size, error = 0;
++
++ error = sysdev_class_register(&ioapic_sysdev_class);
++ if (error)
++ return error;
++
++ for (i = 0; i < nr_ioapics; i++ ) {
++ size = sizeof(struct sys_device) + nr_ioapic_registers[i]
++ * sizeof(struct IO_APIC_route_entry);
++ mp_ioapic_data[i] = kmalloc(size, GFP_KERNEL);
++ if (!mp_ioapic_data[i]) {
++ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
++ continue;
++ }
++ memset(mp_ioapic_data[i], 0, size);
++ dev = &mp_ioapic_data[i]->dev;
++ dev->id = i;
++ dev->cls = &ioapic_sysdev_class;
++ error = sysdev_register(dev);
++ if (error) {
++ kfree(mp_ioapic_data[i]);
++ mp_ioapic_data[i] = NULL;
++ printk(KERN_ERR "Can't suspend/resume IOAPIC %d\n", i);
++ continue;
++ }
++ }
++
++ return 0;
++}
++
++device_initcall(ioapic_init_sysfs);
++
++/* --------------------------------------------------------------------------
++ ACPI-based IOAPIC Configuration
++ -------------------------------------------------------------------------- */
++
++#ifdef CONFIG_ACPI
++
++#define IO_APIC_MAX_ID 0xFE
++
++int __init io_apic_get_version (int ioapic)
++{
++ union IO_APIC_reg_01 reg_01;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(ioapic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return reg_01.bits.version;
++}
++
++
++int __init io_apic_get_redir_entries (int ioapic)
++{
++ union IO_APIC_reg_01 reg_01;
++ unsigned long flags;
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ reg_01.raw = io_apic_read(ioapic, 1);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return reg_01.bits.entries;
++}
++
++
++int io_apic_set_pci_routing (int ioapic, int pin, int irq, int edge_level, int active_high_low)
++{
++ struct IO_APIC_route_entry entry;
++ unsigned long flags;
++
++ if (!IO_APIC_IRQ(irq)) {
++ apic_printk(APIC_QUIET,KERN_ERR "IOAPIC[%d]: Invalid reference to IRQ 0\n",
++ ioapic);
++ return -EINVAL;
++ }
++
++ /*
++ * Generate a PCI IRQ routing entry and program the IOAPIC accordingly.
++ * Note that we mask (disable) IRQs now -- these get enabled when the
++ * corresponding device driver registers for this IRQ.
++ */
++
++ memset(&entry,0,sizeof(entry));
++
++ entry.delivery_mode = INT_DELIVERY_MODE;
++ entry.dest_mode = INT_DEST_MODE;
++ entry.dest.logical.logical_dest = cpu_mask_to_apicid(TARGET_CPUS);
++ entry.trigger = edge_level;
++ entry.polarity = active_high_low;
++ entry.mask = 1; /* Disabled (masked) */
++
++ irq = gsi_irq_sharing(irq);
++ /*
++ * IRQs < 16 are already in the irq_2_pin[] map
++ */
++ if (irq >= 16)
++ add_pin_to_irq(irq, ioapic, pin);
++
++ entry.vector = assign_irq_vector(irq);
++
++ apic_printk(APIC_VERBOSE,KERN_DEBUG "IOAPIC[%d]: Set PCI routing entry (%d-%d -> 0x%x -> "
++ "IRQ %d Mode:%i Active:%i)\n", ioapic,
++ mp_ioapics[ioapic].mpc_apicid, pin, entry.vector, irq,
++ edge_level, active_high_low);
++
++ ioapic_register_intr(irq, entry.vector, edge_level);
++
++ if (!ioapic && (irq < 16))
++ disable_8259A_irq(irq);
++
++ spin_lock_irqsave(&ioapic_lock, flags);
++ io_apic_write(ioapic, 0x11+2*pin, *(((int *)&entry)+1));
++ io_apic_write(ioapic, 0x10+2*pin, *(((int *)&entry)+0));
++ set_native_irq_info(use_pci_vector() ? entry.vector : irq, TARGET_CPUS);
++ spin_unlock_irqrestore(&ioapic_lock, flags);
++
++ return 0;
++}
++
++#endif /* CONFIG_ACPI */
++
++
++#ifndef CONFIG_XEN
++/*
++ * This function currently is only a helper for the i386 smp boot process where
++ * we need to reprogram the ioredtbls to cater for the cpus which have come online
++ * so mask in all cases should simply be TARGET_CPUS
++ */
++#ifdef CONFIG_SMP
++void __init setup_ioapic_dest(void)
++{
++ int pin, ioapic, irq, irq_entry;
++
++ if (skip_ioapic_setup == 1)
++ return;
++
++ for (ioapic = 0; ioapic < nr_ioapics; ioapic++) {
++ for (pin = 0; pin < nr_ioapic_registers[ioapic]; pin++) {
++ irq_entry = find_irq_entry(ioapic, pin, mp_INT);
++ if (irq_entry == -1)
++ continue;
++ irq = pin_2_irq(irq_entry, ioapic, pin);
++ set_ioapic_affinity_irq(irq, TARGET_CPUS);
++ }
++
++ }
++}
++#endif
++#endif /* !CONFIG_XEN */
+diff -urNp linux-2.6/arch/x86_64/kernel/ioport-xen.c new/arch/x86_64/kernel/ioport-xen.c
+--- linux-2.6/arch/x86_64/kernel/ioport-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/ioport-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,99 @@
++/*
++ * linux/arch/x86_64/kernel/ioport.c
++ *
++ * This contains the io-permission bitmap code - written by obz, with changes
++ * by Linus.
++ */
++
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/capability.h>
++#include <linux/errno.h>
++#include <linux/types.h>
++#include <linux/ioport.h>
++#include <linux/mm.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/stddef.h>
++#include <linux/slab.h>
++#include <linux/thread_info.h>
++#include <xen/interface/physdev.h>
++
++/* Set EXTENT bits starting at BASE in BITMAP to value TURN_ON. */
++static void set_bitmap(unsigned long *bitmap, unsigned int base, unsigned int extent, int new_value)
++{
++ int i;
++
++ if (new_value)
++ for (i = base; i < base + extent; i++)
++ __set_bit(i, bitmap);
++ else
++ for (i = base; i < base + extent; i++)
++ clear_bit(i, bitmap);
++}
++
++/*
++ * this changes the io permissions bitmap in the current task.
++ */
++asmlinkage long sys_ioperm(unsigned long from, unsigned long num, int turn_on)
++{
++ struct thread_struct * t = ¤t->thread;
++ unsigned long *bitmap;
++ struct physdev_set_iobitmap set_iobitmap;
++
++ if ((from + num <= from) || (from + num > IO_BITMAP_BITS))
++ return -EINVAL;
++ if (turn_on && !capable(CAP_SYS_RAWIO))
++ return -EPERM;
++
++ /*
++ * If it's the first ioperm() call in this thread's lifetime, set the
++ * IO bitmap up. ioperm() is much less timing critical than clone(),
++ * this is why we delay this operation until now:
++ */
++ if (!t->io_bitmap_ptr) {
++ bitmap = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
++ if (!bitmap)
++ return -ENOMEM;
++
++ memset(bitmap, 0xff, IO_BITMAP_BYTES);
++ t->io_bitmap_ptr = bitmap;
++
++ set_iobitmap.bitmap = (char *)bitmap;
++ set_iobitmap.nr_ports = IO_BITMAP_BITS;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap, &set_iobitmap);
++ }
++
++ set_bitmap(t->io_bitmap_ptr, from, num, !turn_on);
++
++ return 0;
++}
++
++/*
++ * sys_iopl has to be used when you want to access the IO ports
++ * beyond the 0x3ff range: to get the full 65536 ports bitmapped
++ * you'd need 8kB of bitmaps/process, which is a bit excessive.
++ *
++ */
++
++asmlinkage long sys_iopl(unsigned int new_iopl, struct pt_regs *regs)
++{
++ unsigned int old_iopl = current->thread.iopl;
++ struct physdev_set_iopl set_iopl;
++
++ if (new_iopl > 3)
++ return -EINVAL;
++
++ /* Need "raw I/O" privileges for direct port access. */
++ if ((new_iopl > old_iopl) && !capable(CAP_SYS_RAWIO))
++ return -EPERM;
++
++ /* Change our version of the privilege levels. */
++ current->thread.iopl = new_iopl;
++
++ /* Force the change at ring 0. */
++ set_iopl.iopl = (new_iopl == 0) ? 1 : new_iopl;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
++
++ return 0;
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/irq.c new/arch/x86_64/kernel/irq.c
+--- linux-2.6/arch/x86_64/kernel/irq.c 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/irq.c 2006-05-09 12:33:19.000000000 +0200
+@@ -91,8 +91,8 @@ skip:
+ */
+ asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
+ {
+- /* high bits used in ret_from_ code */
+- unsigned irq = regs->orig_rax & 0xff;
++ /* high bit used in ret_from_ code */
++ unsigned irq = ~regs->orig_rax;
+
+ exit_idle();
+ irq_enter();
+diff -urNp linux-2.6/arch/x86_64/kernel/irq-xen.c new/arch/x86_64/kernel/irq-xen.c
+--- linux-2.6/arch/x86_64/kernel/irq-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/irq-xen.c 2006-05-09 12:33:19.000000000 +0200
+@@ -0,0 +1,163 @@
++/*
++ * linux/arch/x86_64/kernel/irq.c
++ *
++ * Copyright (C) 1992, 1998 Linus Torvalds, Ingo Molnar
++ *
++ * This file contains the lowest level x86_64-specific interrupt
++ * entry and irq statistics code. All the remaining irq logic is
++ * done by the generic kernel/irq/ code and in the
++ * x86_64-specific irq controller code. (e.g. i8259.c and
++ * io_apic.c.)
++ */
++
++#include <linux/kernel_stat.h>
++#include <linux/interrupt.h>
++#include <linux/seq_file.h>
++#include <linux/module.h>
++#include <linux/delay.h>
++#include <asm/uaccess.h>
++#include <asm/io_apic.h>
++#include <asm/idle.h>
++
++atomic_t irq_err_count;
++#ifdef CONFIG_X86_IO_APIC
++#ifdef APIC_MISMATCH_DEBUG
++atomic_t irq_mis_count;
++#endif
++#endif
++
++/*
++ * Generic, controller-independent functions:
++ */
++
++int show_interrupts(struct seq_file *p, void *v)
++{
++ int i = *(loff_t *) v, j;
++ struct irqaction * action;
++ unsigned long flags;
++
++ if (i == 0) {
++ seq_printf(p, " ");
++ for_each_online_cpu(j)
++ seq_printf(p, "CPU%d ",j);
++ seq_putc(p, '\n');
++ }
++
++ if (i < NR_IRQS) {
++ spin_lock_irqsave(&irq_desc[i].lock, flags);
++ action = irq_desc[i].action;
++ if (!action)
++ goto skip;
++ seq_printf(p, "%3d: ",i);
++#ifndef CONFIG_SMP
++ seq_printf(p, "%10u ", kstat_irqs(i));
++#else
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
++#endif
++ seq_printf(p, " %14s", irq_desc[i].handler->typename);
++
++ seq_printf(p, " %s", action->name);
++ for (action=action->next; action; action = action->next)
++ seq_printf(p, ", %s", action->name);
++ seq_putc(p, '\n');
++skip:
++ spin_unlock_irqrestore(&irq_desc[i].lock, flags);
++ } else if (i == NR_IRQS) {
++ seq_printf(p, "NMI: ");
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
++ seq_putc(p, '\n');
++#ifdef CONFIG_X86_LOCAL_APIC
++ seq_printf(p, "LOC: ");
++ for_each_online_cpu(j)
++ seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
++ seq_putc(p, '\n');
++#endif
++ seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
++#ifdef CONFIG_X86_IO_APIC
++#ifdef APIC_MISMATCH_DEBUG
++ seq_printf(p, "MIS: %10u\n", atomic_read(&irq_mis_count));
++#endif
++#endif
++ }
++ return 0;
++}
++
++/*
++ * do_IRQ handles all normal device IRQ's (the special
++ * SMP cross-CPU interrupts have their own specific
++ * handlers).
++ */
++asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
++{
++ /* high bit used in ret_from_ code */
++ unsigned irq = ~regs->orig_rax;
++
++ exit_idle();
++ irq_enter();
++
++ __do_IRQ(irq, regs);
++ irq_exit();
++
++ return 1;
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++void fixup_irqs(cpumask_t map)
++{
++ unsigned int irq;
++ static int warned;
++
++ for (irq = 0; irq < NR_IRQS; irq++) {
++ cpumask_t mask;
++ if (irq == 2)
++ continue;
++
++ cpus_and(mask, irq_affinity[irq], map);
++ if (any_online_cpu(mask) == NR_CPUS) {
++ printk("Breaking affinity for irq %i\n", irq);
++ mask = map;
++ }
++ if (irq_desc[irq].handler->set_affinity)
++ irq_desc[irq].handler->set_affinity(irq, mask);
++ else if (irq_desc[irq].action && !(warned++))
++ printk("Cannot set affinity for irq %i\n", irq);
++ }
++
++ /* That doesn't seem sufficient. Give it 1ms. */
++ local_irq_enable();
++ mdelay(1);
++ local_irq_disable();
++}
++#endif
++
++extern void call_softirq(void);
++
++asmlinkage void do_softirq(void)
++{
++ __u32 pending;
++ unsigned long flags;
++
++ if (in_interrupt())
++ return;
++
++ local_irq_save(flags);
++ pending = local_softirq_pending();
++ /* Switch to interrupt stack */
++ if (pending)
++ call_softirq();
++ local_irq_restore(flags);
++}
++EXPORT_SYMBOL(do_softirq);
++
++#ifndef CONFIG_X86_LOCAL_APIC
++/*
++ * 'what should we do if we get a hw irq event on an illegal vector'.
++ * each architecture has to answer this themselves.
++ */
++void ack_bad_irq(unsigned int irq)
++{
++ printk("unexpected IRQ trap at vector %02x\n", irq);
++}
++#endif
+diff -urNp linux-2.6/arch/x86_64/kernel/ldt-xen.c new/arch/x86_64/kernel/ldt-xen.c
+--- linux-2.6/arch/x86_64/kernel/ldt-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/ldt-xen.c 2006-05-09 12:33:19.000000000 +0200
+@@ -0,0 +1,282 @@
++/*
++ * linux/arch/x86_64/kernel/ldt.c
++ *
++ * Copyright (C) 1992 Krishna Balasubramanian and Linus Torvalds
++ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
++ * Copyright (C) 2002 Andi Kleen
++ *
++ * This handles calls from both 32bit and 64bit mode.
++ */
++
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/string.h>
++#include <linux/mm.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/vmalloc.h>
++#include <linux/slab.h>
++
++#include <asm/uaccess.h>
++#include <asm/system.h>
++#include <asm/ldt.h>
++#include <asm/desc.h>
++#include <asm/proto.h>
++#include <asm/pgalloc.h>
++
++#ifdef CONFIG_SMP /* avoids "defined but not used" warnig */
++static void flush_ldt(void *null)
++{
++ if (current->active_mm)
++ load_LDT(¤t->active_mm->context);
++}
++#endif
++
++static int alloc_ldt(mm_context_t *pc, unsigned mincount, int reload)
++{
++ void *oldldt;
++ void *newldt;
++ unsigned oldsize;
++
++ if (mincount <= (unsigned)pc->size)
++ return 0;
++ oldsize = pc->size;
++ mincount = (mincount+511)&(~511);
++ if (mincount*LDT_ENTRY_SIZE > PAGE_SIZE)
++ newldt = vmalloc(mincount*LDT_ENTRY_SIZE);
++ else
++ newldt = kmalloc(mincount*LDT_ENTRY_SIZE, GFP_KERNEL);
++
++ if (!newldt)
++ return -ENOMEM;
++
++ if (oldsize)
++ memcpy(newldt, pc->ldt, oldsize*LDT_ENTRY_SIZE);
++ oldldt = pc->ldt;
++ memset(newldt+oldsize*LDT_ENTRY_SIZE, 0, (mincount-oldsize)*LDT_ENTRY_SIZE);
++ wmb();
++ pc->ldt = newldt;
++ wmb();
++ pc->size = mincount;
++ wmb();
++ if (reload) {
++#ifdef CONFIG_SMP
++ cpumask_t mask;
++
++ preempt_disable();
++#endif
++ make_pages_readonly(
++ pc->ldt,
++ (pc->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ load_LDT(pc);
++#ifdef CONFIG_SMP
++ mask = cpumask_of_cpu(smp_processor_id());
++ if (!cpus_equal(current->mm->cpu_vm_mask, mask))
++ smp_call_function(flush_ldt, NULL, 1, 1);
++ preempt_enable();
++#endif
++ }
++ if (oldsize) {
++ make_pages_writable(
++ oldldt,
++ (oldsize * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ if (oldsize*LDT_ENTRY_SIZE > PAGE_SIZE)
++ vfree(oldldt);
++ else
++ kfree(oldldt);
++ }
++ return 0;
++}
++
++static inline int copy_ldt(mm_context_t *new, mm_context_t *old)
++{
++ int err = alloc_ldt(new, old->size, 0);
++ if (err < 0)
++ return err;
++ memcpy(new->ldt, old->ldt, old->size*LDT_ENTRY_SIZE);
++ make_pages_readonly(
++ new->ldt,
++ (new->size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ return 0;
++}
++
++/*
++ * we do not have to muck with descriptors here, that is
++ * done in switch_mm() as needed.
++ */
++int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
++{
++ struct mm_struct * old_mm;
++ int retval = 0;
++
++ memset(&mm->context, 0, sizeof(mm->context));
++ init_MUTEX(&mm->context.sem);
++ old_mm = current->mm;
++ if (old_mm && old_mm->context.size > 0) {
++ down(&old_mm->context.sem);
++ retval = copy_ldt(&mm->context, &old_mm->context);
++ up(&old_mm->context.sem);
++ }
++ if (retval == 0) {
++ spin_lock(&mm_unpinned_lock);
++ list_add(&mm->context.unpinned, &mm_unpinned);
++ spin_unlock(&mm_unpinned_lock);
++ }
++ return retval;
++}
++
++/*
++ *
++ * Don't touch the LDT register - we're already in the next thread.
++ */
++void destroy_context(struct mm_struct *mm)
++{
++ if (mm->context.size) {
++ if (mm == current->active_mm)
++ clear_LDT();
++ make_pages_writable(
++ mm->context.ldt,
++ (mm->context.size * LDT_ENTRY_SIZE) / PAGE_SIZE,
++ XENFEAT_writable_descriptor_tables);
++ if (mm->context.size*LDT_ENTRY_SIZE > PAGE_SIZE)
++ vfree(mm->context.ldt);
++ else
++ kfree(mm->context.ldt);
++ mm->context.size = 0;
++ }
++ if (!mm->context.pinned) {
++ spin_lock(&mm_unpinned_lock);
++ list_del(&mm->context.unpinned);
++ spin_unlock(&mm_unpinned_lock);
++ }
++}
++
++static int read_ldt(void __user * ptr, unsigned long bytecount)
++{
++ int err;
++ unsigned long size;
++ struct mm_struct * mm = current->mm;
++
++ if (!mm->context.size)
++ return 0;
++ if (bytecount > LDT_ENTRY_SIZE*LDT_ENTRIES)
++ bytecount = LDT_ENTRY_SIZE*LDT_ENTRIES;
++
++ down(&mm->context.sem);
++ size = mm->context.size*LDT_ENTRY_SIZE;
++ if (size > bytecount)
++ size = bytecount;
++
++ err = 0;
++ if (copy_to_user(ptr, mm->context.ldt, size))
++ err = -EFAULT;
++ up(&mm->context.sem);
++ if (err < 0)
++ goto error_return;
++ if (size != bytecount) {
++ /* zero-fill the rest */
++ if (clear_user(ptr+size, bytecount-size) != 0) {
++ err = -EFAULT;
++ goto error_return;
++ }
++ }
++ return bytecount;
++error_return:
++ return err;
++}
++
++static int read_default_ldt(void __user * ptr, unsigned long bytecount)
++{
++ /* Arbitrary number */
++ /* x86-64 default LDT is all zeros */
++ if (bytecount > 128)
++ bytecount = 128;
++ if (clear_user(ptr, bytecount))
++ return -EFAULT;
++ return bytecount;
++}
++
++static int write_ldt(void __user * ptr, unsigned long bytecount, int oldmode)
++{
++ struct task_struct *me = current;
++ struct mm_struct * mm = me->mm;
++ __u32 entry_1, entry_2, *lp;
++ unsigned long mach_lp;
++ int error;
++ struct user_desc ldt_info;
++
++ error = -EINVAL;
++
++ if (bytecount != sizeof(ldt_info))
++ goto out;
++ error = -EFAULT;
++ if (copy_from_user(&ldt_info, ptr, bytecount))
++ goto out;
++
++ error = -EINVAL;
++ if (ldt_info.entry_number >= LDT_ENTRIES)
++ goto out;
++ if (ldt_info.contents == 3) {
++ if (oldmode)
++ goto out;
++ if (ldt_info.seg_not_present == 0)
++ goto out;
++ }
++
++ down(&mm->context.sem);
++ if (ldt_info.entry_number >= (unsigned)mm->context.size) {
++ error = alloc_ldt(¤t->mm->context, ldt_info.entry_number+1, 1);
++ if (error < 0)
++ goto out_unlock;
++ }
++
++ lp = (__u32 *) ((ldt_info.entry_number << 3) + (char *) mm->context.ldt);
++ mach_lp = arbitrary_virt_to_machine(lp);
++
++ /* Allow LDTs to be cleared by the user. */
++ if (ldt_info.base_addr == 0 && ldt_info.limit == 0) {
++ if (oldmode || LDT_empty(&ldt_info)) {
++ entry_1 = 0;
++ entry_2 = 0;
++ goto install;
++ }
++ }
++
++ entry_1 = LDT_entry_a(&ldt_info);
++ entry_2 = LDT_entry_b(&ldt_info);
++ if (oldmode)
++ entry_2 &= ~(1 << 20);
++
++ /* Install the new entry ... */
++install:
++ error = HYPERVISOR_update_descriptor(mach_lp, (unsigned long)((entry_1 | (unsigned long) entry_2 << 32)));
++
++out_unlock:
++ up(&mm->context.sem);
++out:
++ return error;
++}
++
++asmlinkage int sys_modify_ldt(int func, void __user *ptr, unsigned long bytecount)
++{
++ int ret = -ENOSYS;
++
++ switch (func) {
++ case 0:
++ ret = read_ldt(ptr, bytecount);
++ break;
++ case 1:
++ ret = write_ldt(ptr, bytecount, 1);
++ break;
++ case 2:
++ ret = read_default_ldt(ptr, bytecount);
++ break;
++ case 0x11:
++ ret = write_ldt(ptr, bytecount, 0);
++ break;
++ }
++ return ret;
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/Makefile new/arch/x86_64/kernel/Makefile
+--- linux-2.6/arch/x86_64/kernel/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/Makefile 2006-05-09 12:33:17.000000000 +0200
+@@ -20,11 +20,13 @@ obj-$(CONFIG_MICROCODE) += microcode.o
+ obj-$(CONFIG_X86_CPUID) += cpuid.o
+ obj-$(CONFIG_SMP) += smp.o smpboot.o trampoline.o
+ obj-$(CONFIG_X86_LOCAL_APIC) += apic.o nmi.o
++obj-$(CONFIG_X86_XEN_GENAPIC) += genapic.o genapic_xen.o
+ obj-$(CONFIG_X86_IO_APIC) += io_apic.o mpparse.o \
+ genapic.o genapic_cluster.o genapic_flat.o
+ obj-$(CONFIG_KEXEC) += machine_kexec.o relocate_kernel.o crash.o
+ obj-$(CONFIG_CRASH_DUMP) += crash_dump.o
+-obj-$(CONFIG_PM) += suspend.o
++obj-$(CONFIG_SOFTWARE_SUSPEND) += suspend.o
++obj-$(CONFIG_ACPI_SLEEP) += suspend.o
+ obj-$(CONFIG_SUSPEND_SHARED) += suspend_asm.o
+ obj-$(CONFIG_CPU_FREQ) += cpufreq/
+ obj-$(CONFIG_EARLY_PRINTK) += early_printk.o
+@@ -49,3 +51,18 @@ intel_cacheinfo-y += ../../i386/kernel/
+ quirks-y += ../../i386/kernel/quirks.o
+ i8237-y += ../../i386/kernel/i8237.o
+ msr-$(subst m,y,$(CONFIG_X86_MSR)) += ../../i386/kernel/msr.o
++
++ifdef CONFIG_XEN
++time-y += ../../i386/kernel/time-xen.o
++pci-dma-y += ../../i386/kernel/pci-dma-xen.o
++microcode-$(subst m,y,$(CONFIG_MICROCODE)) := ../../i386/kernel/microcode-xen.o
++quirks-y := ../../i386/kernel/quirks-xen.o
++
++n-obj-xen := i8259.o reboot.o i8237.o smpboot.o trampoline.o
++
++include $(srctree)/scripts/Makefile.xen
++
++obj-y := $(call filterxen, $(obj-y), $(n-obj-xen))
++obj-y := $(call cherrypickxen, $(obj-y))
++extra-y := $(call cherrypickxen, $(extra-y))
++endif
+diff -urNp linux-2.6/arch/x86_64/kernel/mpparse-xen.c new/arch/x86_64/kernel/mpparse-xen.c
+--- linux-2.6/arch/x86_64/kernel/mpparse-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/mpparse-xen.c 2006-05-23 18:37:10.000000000 +0200
+@@ -0,0 +1,1012 @@
++/*
++ * Intel Multiprocessor Specification 1.1 and 1.4
++ * compliant MP-table parsing routines.
++ *
++ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
++ * (c) 1998, 1999, 2000 Ingo Molnar <mingo@redhat.com>
++ *
++ * Fixes
++ * Erich Boleyn : MP v1.4 and additional changes.
++ * Alan Cox : Added EBDA scanning
++ * Ingo Molnar : various cleanups and rewrites
++ * Maciej W. Rozycki: Bits for default MP configurations
++ * Paul Diefenbaugh: Added full ACPI support
++ */
++
++#include <linux/mm.h>
++#include <linux/init.h>
++#include <linux/delay.h>
++#include <linux/config.h>
++#include <linux/bootmem.h>
++#include <linux/smp_lock.h>
++#include <linux/kernel_stat.h>
++#include <linux/mc146818rtc.h>
++#include <linux/acpi.h>
++#include <linux/module.h>
++
++#include <asm/smp.h>
++#include <asm/mtrr.h>
++#include <asm/mpspec.h>
++#include <asm/pgalloc.h>
++#include <asm/io_apic.h>
++#include <asm/proto.h>
++#include <asm/acpi.h>
++
++/* Have we found an MP table */
++int smp_found_config;
++unsigned int __initdata maxcpus = NR_CPUS;
++
++int acpi_found_madt;
++
++/*
++ * Various Linux-internal data structures created from the
++ * MP-table.
++ */
++unsigned char apic_version [MAX_APICS];
++unsigned char mp_bus_id_to_type [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
++int mp_bus_id_to_pci_bus [MAX_MP_BUSSES] = { [0 ... MAX_MP_BUSSES-1] = -1 };
++
++static int mp_current_pci_id = 0;
++/* I/O APIC entries */
++struct mpc_config_ioapic mp_ioapics[MAX_IO_APICS];
++
++/* # of MP IRQ source entries */
++struct mpc_config_intsrc mp_irqs[MAX_IRQ_SOURCES];
++
++/* MP IRQ source entries */
++int mp_irq_entries;
++
++int nr_ioapics;
++int pic_mode;
++unsigned long mp_lapic_addr = 0;
++
++
++
++/* Processor that is doing the boot up */
++unsigned int boot_cpu_id = -1U;
++/* Internal processor count */
++unsigned int num_processors __initdata = 0;
++
++unsigned disabled_cpus __initdata;
++
++/* Bitmask of physically existing CPUs */
++physid_mask_t phys_cpu_present_map = PHYSID_MASK_NONE;
++
++/* ACPI MADT entry parsing functions */
++#ifdef CONFIG_ACPI
++extern struct acpi_boot_flags acpi_boot;
++#ifdef CONFIG_X86_LOCAL_APIC
++extern int acpi_parse_lapic (acpi_table_entry_header *header);
++extern int acpi_parse_lapic_addr_ovr (acpi_table_entry_header *header);
++extern int acpi_parse_lapic_nmi (acpi_table_entry_header *header);
++#endif /*CONFIG_X86_LOCAL_APIC*/
++#ifdef CONFIG_X86_IO_APIC
++extern int acpi_parse_ioapic (acpi_table_entry_header *header);
++#endif /*CONFIG_X86_IO_APIC*/
++#endif /*CONFIG_ACPI*/
++
++u8 bios_cpu_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
++
++
++/*
++ * Intel MP BIOS table parsing routines:
++ */
++
++/*
++ * Checksum an MP configuration block.
++ */
++
++static int __init mpf_checksum(unsigned char *mp, int len)
++{
++ int sum = 0;
++
++ while (len--)
++ sum += *mp++;
++
++ return sum & 0xFF;
++}
++
++#ifndef CONFIG_XEN
++static void __cpuinit MP_processor_info (struct mpc_config_processor *m)
++{
++ int cpu;
++ unsigned char ver;
++ cpumask_t tmp_map;
++
++ if (!(m->mpc_cpuflag & CPU_ENABLED)) {
++ disabled_cpus++;
++ return;
++ }
++
++ printk(KERN_INFO "Processor #%d %d:%d APIC version %d\n",
++ m->mpc_apicid,
++ (m->mpc_cpufeature & CPU_FAMILY_MASK)>>8,
++ (m->mpc_cpufeature & CPU_MODEL_MASK)>>4,
++ m->mpc_apicver);
++
++ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
++ Dprintk(" Bootup CPU\n");
++ boot_cpu_id = m->mpc_apicid;
++ }
++ if (num_processors >= NR_CPUS) {
++ printk(KERN_WARNING "WARNING: NR_CPUS limit of %i reached."
++ " Processor ignored.\n", NR_CPUS);
++ return;
++ }
++
++ num_processors++;
++ cpus_complement(tmp_map, cpu_present_map);
++ cpu = first_cpu(tmp_map);
++
++#if MAX_APICS < 255
++ if ((int)m->mpc_apicid > MAX_APICS) {
++ printk(KERN_ERR "Processor #%d INVALID. (Max ID: %d).\n",
++ m->mpc_apicid, MAX_APICS);
++ return;
++ }
++#endif
++ ver = m->mpc_apicver;
++
++ physid_set(m->mpc_apicid, phys_cpu_present_map);
++ /*
++ * Validate version
++ */
++ if (ver == 0x0) {
++ printk(KERN_ERR "BIOS bug, APIC version is 0 for CPU#%d! fixing up to 0x10. (tell your hw vendor)\n", m->mpc_apicid);
++ ver = 0x10;
++ }
++ apic_version[m->mpc_apicid] = ver;
++ if (m->mpc_cpuflag & CPU_BOOTPROCESSOR) {
++ /*
++ * bios_cpu_apicid is required to have processors listed
++ * in same order as logical cpu numbers. Hence the first
++ * entry is BSP, and so on.
++ */
++ cpu = 0;
++ }
++ bios_cpu_apicid[cpu] = m->mpc_apicid;
++ x86_cpu_to_apicid[cpu] = m->mpc_apicid;
++
++ cpu_set(cpu, cpu_possible_map);
++ cpu_set(cpu, cpu_present_map);
++}
++#else
++void __init MP_processor_info (struct mpc_config_processor *m)
++{
++ num_processors++;
++}
++#endif /* CONFIG_XEN */
++
++static void __init MP_bus_info (struct mpc_config_bus *m)
++{
++ char str[7];
++
++ memcpy(str, m->mpc_bustype, 6);
++ str[6] = 0;
++ Dprintk("Bus #%d is %s\n", m->mpc_busid, str);
++
++ if (strncmp(str, "ISA", 3) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_ISA;
++ } else if (strncmp(str, "EISA", 4) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_EISA;
++ } else if (strncmp(str, "PCI", 3) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_PCI;
++ mp_bus_id_to_pci_bus[m->mpc_busid] = mp_current_pci_id;
++ mp_current_pci_id++;
++ } else if (strncmp(str, "MCA", 3) == 0) {
++ mp_bus_id_to_type[m->mpc_busid] = MP_BUS_MCA;
++ } else {
++ printk(KERN_ERR "Unknown bustype %s\n", str);
++ }
++}
++
++static void __init MP_ioapic_info (struct mpc_config_ioapic *m)
++{
++ if (!(m->mpc_flags & MPC_APIC_USABLE))
++ return;
++
++ printk("I/O APIC #%d Version %d at 0x%X.\n",
++ m->mpc_apicid, m->mpc_apicver, m->mpc_apicaddr);
++ if (nr_ioapics >= MAX_IO_APICS) {
++ printk(KERN_ERR "Max # of I/O APICs (%d) exceeded (found %d).\n",
++ MAX_IO_APICS, nr_ioapics);
++ panic("Recompile kernel with bigger MAX_IO_APICS!.\n");
++ }
++ if (!m->mpc_apicaddr) {
++ printk(KERN_ERR "WARNING: bogus zero I/O APIC address"
++ " found in MP table, skipping!\n");
++ return;
++ }
++ mp_ioapics[nr_ioapics] = *m;
++ nr_ioapics++;
++}
++
++static void __init MP_intsrc_info (struct mpc_config_intsrc *m)
++{
++ mp_irqs [mp_irq_entries] = *m;
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d,"
++ " IRQ %02x, APIC ID %x, APIC INT %02x\n",
++ m->mpc_irqtype, m->mpc_irqflag & 3,
++ (m->mpc_irqflag >> 2) & 3, m->mpc_srcbus,
++ m->mpc_srcbusirq, m->mpc_dstapic, m->mpc_dstirq);
++ if (++mp_irq_entries >= MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!!\n");
++}
++
++static void __init MP_lintsrc_info (struct mpc_config_lintsrc *m)
++{
++ Dprintk("Lint: type %d, pol %d, trig %d, bus %d,"
++ " IRQ %02x, APIC ID %x, APIC LINT %02x\n",
++ m->mpc_irqtype, m->mpc_irqflag & 3,
++ (m->mpc_irqflag >> 2) &3, m->mpc_srcbusid,
++ m->mpc_srcbusirq, m->mpc_destapic, m->mpc_destapiclint);
++ /*
++ * Well it seems all SMP boards in existence
++ * use ExtINT/LVT1 == LINT0 and
++ * NMI/LVT2 == LINT1 - the following check
++ * will show us if this assumptions is false.
++ * Until then we do not have to add baggage.
++ */
++ if ((m->mpc_irqtype == mp_ExtINT) &&
++ (m->mpc_destapiclint != 0))
++ BUG();
++ if ((m->mpc_irqtype == mp_NMI) &&
++ (m->mpc_destapiclint != 1))
++ BUG();
++}
++
++/*
++ * Read/parse the MPC
++ */
++
++static int __init smp_read_mpc(struct mp_config_table *mpc)
++{
++ char str[16];
++ int count=sizeof(*mpc);
++ unsigned char *mpt=((unsigned char *)mpc)+count;
++
++ if (memcmp(mpc->mpc_signature,MPC_SIGNATURE,4)) {
++ printk("SMP mptable: bad signature [%c%c%c%c]!\n",
++ mpc->mpc_signature[0],
++ mpc->mpc_signature[1],
++ mpc->mpc_signature[2],
++ mpc->mpc_signature[3]);
++ return 0;
++ }
++ if (mpf_checksum((unsigned char *)mpc,mpc->mpc_length)) {
++ printk("SMP mptable: checksum error!\n");
++ return 0;
++ }
++ if (mpc->mpc_spec!=0x01 && mpc->mpc_spec!=0x04) {
++ printk(KERN_ERR "SMP mptable: bad table version (%d)!!\n",
++ mpc->mpc_spec);
++ return 0;
++ }
++ if (!mpc->mpc_lapic) {
++ printk(KERN_ERR "SMP mptable: null local APIC address!\n");
++ return 0;
++ }
++ memcpy(str,mpc->mpc_oem,8);
++ str[8]=0;
++ printk(KERN_INFO "OEM ID: %s ",str);
++
++ memcpy(str,mpc->mpc_productid,12);
++ str[12]=0;
++ printk("Product ID: %s ",str);
++
++ printk("APIC at: 0x%X\n",mpc->mpc_lapic);
++
++ /* save the local APIC address, it might be non-default */
++ if (!acpi_lapic)
++ mp_lapic_addr = mpc->mpc_lapic;
++
++ /*
++ * Now process the configuration blocks.
++ */
++ while (count < mpc->mpc_length) {
++ switch(*mpt) {
++ case MP_PROCESSOR:
++ {
++ struct mpc_config_processor *m=
++ (struct mpc_config_processor *)mpt;
++ if (!acpi_lapic)
++ MP_processor_info(m);
++ mpt += sizeof(*m);
++ count += sizeof(*m);
++ break;
++ }
++ case MP_BUS:
++ {
++ struct mpc_config_bus *m=
++ (struct mpc_config_bus *)mpt;
++ MP_bus_info(m);
++ mpt += sizeof(*m);
++ count += sizeof(*m);
++ break;
++ }
++ case MP_IOAPIC:
++ {
++ struct mpc_config_ioapic *m=
++ (struct mpc_config_ioapic *)mpt;
++ MP_ioapic_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ case MP_INTSRC:
++ {
++ struct mpc_config_intsrc *m=
++ (struct mpc_config_intsrc *)mpt;
++
++ MP_intsrc_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ case MP_LINTSRC:
++ {
++ struct mpc_config_lintsrc *m=
++ (struct mpc_config_lintsrc *)mpt;
++ MP_lintsrc_info(m);
++ mpt+=sizeof(*m);
++ count+=sizeof(*m);
++ break;
++ }
++ }
++ }
++ clustered_apic_check();
++ if (!num_processors)
++ printk(KERN_ERR "SMP mptable: no processors registered!\n");
++ return num_processors;
++}
++
++static int __init ELCR_trigger(unsigned int irq)
++{
++ unsigned int port;
++
++ port = 0x4d0 + (irq >> 3);
++ return (inb(port) >> (irq & 7)) & 1;
++}
++
++static void __init construct_default_ioirq_mptable(int mpc_default_type)
++{
++ struct mpc_config_intsrc intsrc;
++ int i;
++ int ELCR_fallback = 0;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqflag = 0; /* conforming */
++ intsrc.mpc_srcbus = 0;
++ intsrc.mpc_dstapic = mp_ioapics[0].mpc_apicid;
++
++ intsrc.mpc_irqtype = mp_INT;
++
++ /*
++ * If true, we have an ISA/PCI system with no IRQ entries
++ * in the MP table. To prevent the PCI interrupts from being set up
++ * incorrectly, we try to use the ELCR. The sanity check to see if
++ * there is good ELCR data is very simple - IRQ0, 1, 2 and 13 can
++ * never be level sensitive, so we simply see if the ELCR agrees.
++ * If it does, we assume it's valid.
++ */
++ if (mpc_default_type == 5) {
++ printk(KERN_INFO "ISA/PCI bus type with no IRQ information... falling back to ELCR\n");
++
++ if (ELCR_trigger(0) || ELCR_trigger(1) || ELCR_trigger(2) || ELCR_trigger(13))
++ printk(KERN_ERR "ELCR contains invalid data... not using ELCR\n");
++ else {
++ printk(KERN_INFO "Using ELCR to identify PCI interrupts\n");
++ ELCR_fallback = 1;
++ }
++ }
++
++ for (i = 0; i < 16; i++) {
++ switch (mpc_default_type) {
++ case 2:
++ if (i == 0 || i == 13)
++ continue; /* IRQ0 & IRQ13 not connected */
++ /* fall through */
++ default:
++ if (i == 2)
++ continue; /* IRQ2 is never connected */
++ }
++
++ if (ELCR_fallback) {
++ /*
++ * If the ELCR indicates a level-sensitive interrupt, we
++ * copy that information over to the MP table in the
++ * irqflag field (level sensitive, active high polarity).
++ */
++ if (ELCR_trigger(i))
++ intsrc.mpc_irqflag = 13;
++ else
++ intsrc.mpc_irqflag = 0;
++ }
++
++ intsrc.mpc_srcbusirq = i;
++ intsrc.mpc_dstirq = i ? i : 2; /* IRQ0 to INTIN2 */
++ MP_intsrc_info(&intsrc);
++ }
++
++ intsrc.mpc_irqtype = mp_ExtINT;
++ intsrc.mpc_srcbusirq = 0;
++ intsrc.mpc_dstirq = 0; /* 8259A to INTIN0 */
++ MP_intsrc_info(&intsrc);
++}
++
++static inline void __init construct_default_ISA_mptable(int mpc_default_type)
++{
++ struct mpc_config_processor processor;
++ struct mpc_config_bus bus;
++ struct mpc_config_ioapic ioapic;
++ struct mpc_config_lintsrc lintsrc;
++ int linttypes[2] = { mp_ExtINT, mp_NMI };
++ int i;
++
++ /*
++ * local APIC has default address
++ */
++ mp_lapic_addr = APIC_DEFAULT_PHYS_BASE;
++
++ /*
++ * 2 CPUs, numbered 0 & 1.
++ */
++ processor.mpc_type = MP_PROCESSOR;
++ /* Either an integrated APIC or a discrete 82489DX. */
++ processor.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
++ processor.mpc_cpuflag = CPU_ENABLED;
++ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
++ (boot_cpu_data.x86_model << 4) |
++ boot_cpu_data.x86_mask;
++ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
++ processor.mpc_reserved[0] = 0;
++ processor.mpc_reserved[1] = 0;
++ for (i = 0; i < 2; i++) {
++ processor.mpc_apicid = i;
++ MP_processor_info(&processor);
++ }
++
++ bus.mpc_type = MP_BUS;
++ bus.mpc_busid = 0;
++ switch (mpc_default_type) {
++ default:
++ printk(KERN_ERR "???\nUnknown standard configuration %d\n",
++ mpc_default_type);
++ /* fall through */
++ case 1:
++ case 5:
++ memcpy(bus.mpc_bustype, "ISA ", 6);
++ break;
++ case 2:
++ case 6:
++ case 3:
++ memcpy(bus.mpc_bustype, "EISA ", 6);
++ break;
++ case 4:
++ case 7:
++ memcpy(bus.mpc_bustype, "MCA ", 6);
++ }
++ MP_bus_info(&bus);
++ if (mpc_default_type > 4) {
++ bus.mpc_busid = 1;
++ memcpy(bus.mpc_bustype, "PCI ", 6);
++ MP_bus_info(&bus);
++ }
++
++ ioapic.mpc_type = MP_IOAPIC;
++ ioapic.mpc_apicid = 2;
++ ioapic.mpc_apicver = mpc_default_type > 4 ? 0x10 : 0x01;
++ ioapic.mpc_flags = MPC_APIC_USABLE;
++ ioapic.mpc_apicaddr = 0xFEC00000;
++ MP_ioapic_info(&ioapic);
++
++ /*
++ * We set up most of the low 16 IO-APIC pins according to MPS rules.
++ */
++ construct_default_ioirq_mptable(mpc_default_type);
++
++ lintsrc.mpc_type = MP_LINTSRC;
++ lintsrc.mpc_irqflag = 0; /* conforming */
++ lintsrc.mpc_srcbusid = 0;
++ lintsrc.mpc_srcbusirq = 0;
++ lintsrc.mpc_destapic = MP_APIC_ALL;
++ for (i = 0; i < 2; i++) {
++ lintsrc.mpc_irqtype = linttypes[i];
++ lintsrc.mpc_destapiclint = i;
++ MP_lintsrc_info(&lintsrc);
++ }
++}
++
++static struct intel_mp_floating *mpf_found;
++
++/*
++ * Scan the memory blocks for an SMP configuration block.
++ */
++void __init get_smp_config (void)
++{
++ struct intel_mp_floating *mpf = mpf_found;
++
++ /*
++ * ACPI supports both logical (e.g. Hyper-Threading) and physical
++ * processors, where MPS only supports physical.
++ */
++ if (acpi_lapic && acpi_ioapic) {
++ printk(KERN_INFO "Using ACPI (MADT) for SMP configuration information\n");
++ return;
++ }
++ else if (acpi_lapic)
++ printk(KERN_INFO "Using ACPI for processor (LAPIC) configuration information\n");
++
++ printk("Intel MultiProcessor Specification v1.%d\n", mpf->mpf_specification);
++ if (mpf->mpf_feature2 & (1<<7)) {
++ printk(KERN_INFO " IMCR and PIC compatibility mode.\n");
++ pic_mode = 1;
++ } else {
++ printk(KERN_INFO " Virtual Wire compatibility mode.\n");
++ pic_mode = 0;
++ }
++
++ /*
++ * Now see if we need to read further.
++ */
++ if (mpf->mpf_feature1 != 0) {
++
++ printk(KERN_INFO "Default MP configuration #%d\n", mpf->mpf_feature1);
++ construct_default_ISA_mptable(mpf->mpf_feature1);
++
++ } else if (mpf->mpf_physptr) {
++
++ /*
++ * Read the physical hardware table. Anything here will
++ * override the defaults.
++ */
++ if (!smp_read_mpc(isa_bus_to_virt(mpf->mpf_physptr))) {
++ smp_found_config = 0;
++ printk(KERN_ERR "BIOS bug, MP table errors detected!...\n");
++ printk(KERN_ERR "... disabling SMP support. (tell your hw vendor)\n");
++ return;
++ }
++ /*
++ * If there are no explicit MP IRQ entries, then we are
++ * broken. We set up most of the low 16 IO-APIC pins to
++ * ISA defaults and hope it will work.
++ */
++ if (!mp_irq_entries) {
++ struct mpc_config_bus bus;
++
++ printk(KERN_ERR "BIOS bug, no explicit IRQ entries, using default mptable. (tell your hw vendor)\n");
++
++ bus.mpc_type = MP_BUS;
++ bus.mpc_busid = 0;
++ memcpy(bus.mpc_bustype, "ISA ", 6);
++ MP_bus_info(&bus);
++
++ construct_default_ioirq_mptable(0);
++ }
++
++ } else
++ BUG();
++
++ printk(KERN_INFO "Processors: %d\n", num_processors);
++ /*
++ * Only use the first configuration found.
++ */
++}
++
++static int __init smp_scan_config (unsigned long base, unsigned long length)
++{
++ extern void __bad_mpf_size(void);
++ unsigned int *bp = isa_bus_to_virt(base);
++ struct intel_mp_floating *mpf;
++
++ Dprintk("Scan SMP from %p for %ld bytes.\n", bp,length);
++ if (sizeof(*mpf) != 16)
++ __bad_mpf_size();
++
++ while (length > 0) {
++ mpf = (struct intel_mp_floating *)bp;
++ if ((*bp == SMP_MAGIC_IDENT) &&
++ (mpf->mpf_length == 1) &&
++ !mpf_checksum((unsigned char *)bp, 16) &&
++ ((mpf->mpf_specification == 1)
++ || (mpf->mpf_specification == 4)) ) {
++
++ smp_found_config = 1;
++ mpf_found = mpf;
++ return 1;
++ }
++ bp += 4;
++ length -= 16;
++ }
++ return 0;
++}
++
++void __init find_intel_smp (void)
++{
++ unsigned int address;
++
++ /*
++ * FIXME: Linux assumes you have 640K of base ram..
++ * this continues the error...
++ *
++ * 1) Scan the bottom 1K for a signature
++ * 2) Scan the top 1K of base RAM
++ * 3) Scan the 64K of bios
++ */
++ if (smp_scan_config(0x0,0x400) ||
++ smp_scan_config(639*0x400,0x400) ||
++ smp_scan_config(0xF0000,0x10000))
++ return;
++ /*
++ * If it is an SMP machine we should know now, unless the
++ * configuration is in an EISA/MCA bus machine with an
++ * extended bios data area.
++ *
++ * there is a real-mode segmented pointer pointing to the
++ * 4K EBDA area at 0x40E, calculate and scan it here.
++ *
++ * NOTE! There are Linux loaders that will corrupt the EBDA
++ * area, and as such this kind of SMP config may be less
++ * trustworthy, simply because the SMP table may have been
++ * stomped on during early boot. These loaders are buggy and
++ * should be fixed.
++ */
++
++ address = *(unsigned short *)phys_to_virt(0x40E);
++ address <<= 4;
++ if (smp_scan_config(address, 0x1000))
++ return;
++
++ /* If we have come this far, we did not find an MP table */
++ printk(KERN_INFO "No mptable found.\n");
++}
++
++/*
++ * - Intel MP Configuration Table
++ */
++void __init find_smp_config (void)
++{
++#ifdef CONFIG_X86_LOCAL_APIC
++ find_intel_smp();
++#endif
++}
++
++
++/* --------------------------------------------------------------------------
++ ACPI-based MP Configuration
++ -------------------------------------------------------------------------- */
++
++#ifdef CONFIG_ACPI
++
++void __init mp_register_lapic_address (
++ u64 address)
++{
++#ifndef CONFIG_XEN
++ mp_lapic_addr = (unsigned long) address;
++
++ set_fixmap_nocache(FIX_APIC_BASE, mp_lapic_addr);
++
++ if (boot_cpu_id == -1U)
++ boot_cpu_id = GET_APIC_ID(apic_read(APIC_ID));
++
++ Dprintk("Boot CPU = %d\n", boot_cpu_physical_apicid);
++#endif
++}
++
++
++void __cpuinit mp_register_lapic (
++ u8 id,
++ u8 enabled)
++{
++ struct mpc_config_processor processor;
++ int boot_cpu = 0;
++
++ if (id >= MAX_APICS) {
++ printk(KERN_WARNING "Processor #%d invalid (max %d)\n",
++ id, MAX_APICS);
++ return;
++ }
++
++ if (id == boot_cpu_physical_apicid)
++ boot_cpu = 1;
++
++#ifndef CONFIG_XEN
++ processor.mpc_type = MP_PROCESSOR;
++ processor.mpc_apicid = id;
++ processor.mpc_apicver = GET_APIC_VERSION(apic_read(APIC_LVR));
++ processor.mpc_cpuflag = (enabled ? CPU_ENABLED : 0);
++ processor.mpc_cpuflag |= (boot_cpu ? CPU_BOOTPROCESSOR : 0);
++ processor.mpc_cpufeature = (boot_cpu_data.x86 << 8) |
++ (boot_cpu_data.x86_model << 4) | boot_cpu_data.x86_mask;
++ processor.mpc_featureflag = boot_cpu_data.x86_capability[0];
++ processor.mpc_reserved[0] = 0;
++ processor.mpc_reserved[1] = 0;
++#endif
++
++ MP_processor_info(&processor);
++}
++
++#ifdef CONFIG_X86_IO_APIC
++
++#define MP_ISA_BUS 0
++#define MP_MAX_IOAPIC_PIN 127
++
++static struct mp_ioapic_routing {
++ int apic_id;
++ int gsi_start;
++ int gsi_end;
++ u32 pin_programmed[4];
++} mp_ioapic_routing[MAX_IO_APICS];
++
++
++static int mp_find_ioapic (
++ int gsi)
++{
++ int i = 0;
++
++ /* Find the IOAPIC that manages this GSI. */
++ for (i = 0; i < nr_ioapics; i++) {
++ if ((gsi >= mp_ioapic_routing[i].gsi_start)
++ && (gsi <= mp_ioapic_routing[i].gsi_end))
++ return i;
++ }
++
++ printk(KERN_ERR "ERROR: Unable to locate IOAPIC for GSI %d\n", gsi);
++
++ return -1;
++}
++
++
++void __init mp_register_ioapic (
++ u8 id,
++ u32 address,
++ u32 gsi_base)
++{
++ int idx = 0;
++
++ if (nr_ioapics >= MAX_IO_APICS) {
++ printk(KERN_ERR "ERROR: Max # of I/O APICs (%d) exceeded "
++ "(found %d)\n", MAX_IO_APICS, nr_ioapics);
++ panic("Recompile kernel with bigger MAX_IO_APICS!\n");
++ }
++ if (!address) {
++ printk(KERN_ERR "WARNING: Bogus (zero) I/O APIC address"
++ " found in MADT table, skipping!\n");
++ return;
++ }
++
++ idx = nr_ioapics++;
++
++ mp_ioapics[idx].mpc_type = MP_IOAPIC;
++ mp_ioapics[idx].mpc_flags = MPC_APIC_USABLE;
++ mp_ioapics[idx].mpc_apicaddr = address;
++
++#ifndef CONFIG_XEN
++ set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
++#endif
++ mp_ioapics[idx].mpc_apicid = id;
++ mp_ioapics[idx].mpc_apicver = io_apic_get_version(idx);
++
++ /*
++ * Build basic IRQ lookup table to facilitate gsi->io_apic lookups
++ * and to prevent reprogramming of IOAPIC pins (PCI IRQs).
++ */
++ mp_ioapic_routing[idx].apic_id = mp_ioapics[idx].mpc_apicid;
++ mp_ioapic_routing[idx].gsi_start = gsi_base;
++ mp_ioapic_routing[idx].gsi_end = gsi_base +
++ io_apic_get_redir_entries(idx);
++
++ printk(KERN_INFO "IOAPIC[%d]: apic_id %d, version %d, address 0x%x, "
++ "GSI %d-%d\n", idx, mp_ioapics[idx].mpc_apicid,
++ mp_ioapics[idx].mpc_apicver, mp_ioapics[idx].mpc_apicaddr,
++ mp_ioapic_routing[idx].gsi_start,
++ mp_ioapic_routing[idx].gsi_end);
++
++ return;
++}
++
++
++void __init mp_override_legacy_irq (
++ u8 bus_irq,
++ u8 polarity,
++ u8 trigger,
++ u32 gsi)
++{
++ struct mpc_config_intsrc intsrc;
++ int ioapic = -1;
++ int pin = -1;
++
++ /*
++ * Convert 'gsi' to 'ioapic.pin'.
++ */
++ ioapic = mp_find_ioapic(gsi);
++ if (ioapic < 0)
++ return;
++ pin = gsi - mp_ioapic_routing[ioapic].gsi_start;
++
++ /*
++ * TBD: This check is for faulty timer entries, where the override
++ * erroneously sets the trigger to level, resulting in a HUGE
++ * increase of timer interrupts!
++ */
++ if ((bus_irq == 0) && (trigger == 3))
++ trigger = 1;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqtype = mp_INT;
++ intsrc.mpc_irqflag = (trigger << 2) | polarity;
++ intsrc.mpc_srcbus = MP_ISA_BUS;
++ intsrc.mpc_srcbusirq = bus_irq; /* IRQ */
++ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid; /* APIC ID */
++ intsrc.mpc_dstirq = pin; /* INTIN# */
++
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, %d-%d\n",
++ intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
++ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
++ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic, intsrc.mpc_dstirq);
++
++ mp_irqs[mp_irq_entries] = intsrc;
++ if (++mp_irq_entries == MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!\n");
++
++ return;
++}
++
++
++void __init mp_config_acpi_legacy_irqs (void)
++{
++ struct mpc_config_intsrc intsrc;
++ int i = 0;
++ int ioapic = -1;
++
++ /*
++ * Fabricate the legacy ISA bus (bus #31).
++ */
++ mp_bus_id_to_type[MP_ISA_BUS] = MP_BUS_ISA;
++ Dprintk("Bus #%d is ISA\n", MP_ISA_BUS);
++
++ /*
++ * Locate the IOAPIC that manages the ISA IRQs (0-15).
++ */
++ ioapic = mp_find_ioapic(0);
++ if (ioapic < 0)
++ return;
++
++ intsrc.mpc_type = MP_INTSRC;
++ intsrc.mpc_irqflag = 0; /* Conforming */
++ intsrc.mpc_srcbus = MP_ISA_BUS;
++ intsrc.mpc_dstapic = mp_ioapics[ioapic].mpc_apicid;
++
++ /*
++ * Use the default configuration for the IRQs 0-15. Unless
++ * overridden by (MADT) interrupt source override entries.
++ */
++ for (i = 0; i < 16; i++) {
++ int idx;
++
++ for (idx = 0; idx < mp_irq_entries; idx++) {
++ struct mpc_config_intsrc *irq = mp_irqs + idx;
++
++ /* Do we already have a mapping for this ISA IRQ? */
++ if (irq->mpc_srcbus == MP_ISA_BUS && irq->mpc_srcbusirq == i)
++ break;
++
++ /* Do we already have a mapping for this IOAPIC pin */
++ if ((irq->mpc_dstapic == intsrc.mpc_dstapic) &&
++ (irq->mpc_dstirq == i))
++ break;
++ }
++
++ if (idx != mp_irq_entries) {
++ printk(KERN_DEBUG "ACPI: IRQ%d used by override.\n", i);
++ continue; /* IRQ already used */
++ }
++
++ intsrc.mpc_irqtype = mp_INT;
++ intsrc.mpc_srcbusirq = i; /* Identity mapped */
++ intsrc.mpc_dstirq = i;
++
++ Dprintk("Int: type %d, pol %d, trig %d, bus %d, irq %d, "
++ "%d-%d\n", intsrc.mpc_irqtype, intsrc.mpc_irqflag & 3,
++ (intsrc.mpc_irqflag >> 2) & 3, intsrc.mpc_srcbus,
++ intsrc.mpc_srcbusirq, intsrc.mpc_dstapic,
++ intsrc.mpc_dstirq);
++
++ mp_irqs[mp_irq_entries] = intsrc;
++ if (++mp_irq_entries == MAX_IRQ_SOURCES)
++ panic("Max # of irq sources exceeded!\n");
++ }
++
++ return;
++}
++
++#define MAX_GSI_NUM 4096
++
++int mp_register_gsi(u32 gsi, int triggering, int polarity)
++{
++ int ioapic = -1;
++ int ioapic_pin = 0;
++ int idx, bit = 0;
++ static int pci_irq = 16;
++ /*
++ * Mapping between Global System Interrupts, which
++ * represent all possible interrupts, to the IRQs
++ * assigned to actual devices.
++ */
++ static int gsi_to_irq[MAX_GSI_NUM];
++
++ if (acpi_irq_model != ACPI_IRQ_MODEL_IOAPIC)
++ return gsi;
++
++ /* Don't set up the ACPI SCI because it's already set up */
++ if (acpi_fadt.sci_int == gsi)
++ return gsi;
++
++ ioapic = mp_find_ioapic(gsi);
++ if (ioapic < 0) {
++ printk(KERN_WARNING "No IOAPIC for GSI %u\n", gsi);
++ return gsi;
++ }
++
++ ioapic_pin = gsi - mp_ioapic_routing[ioapic].gsi_start;
++
++ /*
++ * Avoid pin reprogramming. PRTs typically include entries
++ * with redundant pin->gsi mappings (but unique PCI devices);
++ * we only program the IOAPIC on the first.
++ */
++ bit = ioapic_pin % 32;
++ idx = (ioapic_pin < 32) ? 0 : (ioapic_pin / 32);
++ if (idx > 3) {
++ printk(KERN_ERR "Invalid reference to IOAPIC pin "
++ "%d-%d\n", mp_ioapic_routing[ioapic].apic_id,
++ ioapic_pin);
++ return gsi;
++ }
++ if ((1<<bit) & mp_ioapic_routing[ioapic].pin_programmed[idx]) {
++ Dprintk(KERN_DEBUG "Pin %d-%d already programmed\n",
++ mp_ioapic_routing[ioapic].apic_id, ioapic_pin);
++ return gsi_to_irq[gsi];
++ }
++
++ mp_ioapic_routing[ioapic].pin_programmed[idx] |= (1<<bit);
++
++ if (triggering == ACPI_LEVEL_SENSITIVE) {
++ /*
++ * For PCI devices assign IRQs in order, avoiding gaps
++ * due to unused I/O APIC pins.
++ */
++ int irq = gsi;
++ if (gsi < MAX_GSI_NUM) {
++ /*
++ * Retain the VIA chipset work-around (gsi > 15), but
++ * avoid a problem where the 8254 timer (IRQ0) is setup
++ * via an override (so it's not on pin 0 of the ioapic),
++ * and at the same time, the pin 0 interrupt is a PCI
++ * type. The gsi > 15 test could cause these two pins
++ * to be shared as IRQ0, and they are not shareable.
++ * So test for this condition, and if necessary, avoid
++ * the pin collision.
++ */
++ if (gsi > 15 || (gsi == 0 && !timer_uses_ioapic_pin_0))
++ gsi = pci_irq++;
++ /*
++ * Don't assign IRQ used by ACPI SCI
++ */
++ if (gsi == acpi_fadt.sci_int)
++ gsi = pci_irq++;
++ gsi_to_irq[irq] = gsi;
++ } else {
++ printk(KERN_ERR "GSI %u is too high\n", gsi);
++ return gsi;
++ }
++ }
++
++ io_apic_set_pci_routing(ioapic, ioapic_pin, gsi,
++ triggering == ACPI_EDGE_SENSITIVE ? 0 : 1,
++ polarity == ACPI_ACTIVE_HIGH ? 0 : 1);
++ return gsi;
++}
++
++#endif /*CONFIG_X86_IO_APIC*/
++#endif /*CONFIG_ACPI*/
+diff -urNp linux-2.6/arch/x86_64/kernel/pci-swiotlb-xen.c new/arch/x86_64/kernel/pci-swiotlb-xen.c
+--- linux-2.6/arch/x86_64/kernel/pci-swiotlb-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/pci-swiotlb-xen.c 2006-05-09 12:33:20.000000000 +0200
+@@ -0,0 +1,54 @@
++/* Glue code to lib/swiotlb.c */
++
++#include <linux/pci.h>
++#include <linux/cache.h>
++#include <linux/module.h>
++#include <asm/dma-mapping.h>
++#include <asm/proto.h>
++#include <asm/swiotlb.h>
++#include <asm/dma.h>
++
++#if 0
++int swiotlb __read_mostly;
++EXPORT_SYMBOL(swiotlb);
++#endif
++
++struct dma_mapping_ops swiotlb_dma_ops = {
++#if 0
++ .mapping_error = swiotlb_dma_mapping_error,
++ .alloc_coherent = swiotlb_alloc_coherent,
++ .free_coherent = swiotlb_free_coherent,
++ .map_single = swiotlb_map_single,
++ .unmap_single = swiotlb_unmap_single,
++ .sync_single_for_cpu = swiotlb_sync_single_for_cpu,
++ .sync_single_for_device = swiotlb_sync_single_for_device,
++ .sync_single_range_for_cpu = swiotlb_sync_single_range_for_cpu,
++ .sync_single_range_for_device = swiotlb_sync_single_range_for_device,
++ .sync_sg_for_cpu = swiotlb_sync_sg_for_cpu,
++ .sync_sg_for_device = swiotlb_sync_sg_for_device,
++ .map_sg = swiotlb_map_sg,
++ .unmap_sg = swiotlb_unmap_sg,
++ .dma_supported = NULL,
++#endif
++};
++
++void pci_swiotlb_init(void)
++{
++#if 0
++ /* don't initialize swiotlb if iommu=off (no_iommu=1) */
++ if (!iommu_aperture && !no_iommu &&
++ (end_pfn > MAX_DMA32_PFN || force_iommu))
++ swiotlb = 1;
++ if (swiotlb) {
++ printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n");
++ swiotlb_init();
++ dma_ops = &swiotlb_dma_ops;
++ }
++#else
++ swiotlb_init();
++ if (swiotlb) {
++ printk(KERN_INFO "PCI-DMA: Using software bounce buffering for IO (SWIOTLB)\n");
++ dma_ops = &swiotlb_dma_ops;
++ }
++#endif
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/process-xen.c new/arch/x86_64/kernel/process-xen.c
+--- linux-2.6/arch/x86_64/kernel/process-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/process-xen.c 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,791 @@
++/*
++ * linux/arch/x86-64/kernel/process.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ *
++ * Pentium III FXSR, SSE support
++ * Gareth Hughes <gareth@valinux.com>, May 2000
++ *
++ * X86-64 port
++ * Andi Kleen.
++ *
++ * CPU hotplug support - ashok.raj@intel.com
++ * $Id$
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Modified for Xen
++ */
++
++/*
++ * This file handles the architecture-dependent parts of process handling..
++ */
++
++#include <stdarg.h>
++
++#include <linux/cpu.h>
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/mm.h>
++#include <linux/elfcore.h>
++#include <linux/smp.h>
++#include <linux/slab.h>
++#include <linux/user.h>
++#include <linux/module.h>
++#include <linux/a.out.h>
++#include <linux/interrupt.h>
++#include <linux/delay.h>
++#include <linux/ptrace.h>
++#include <linux/utsname.h>
++#include <linux/random.h>
++#include <linux/notifier.h>
++#include <linux/kprobes.h>
++
++#include <asm/uaccess.h>
++#include <asm/pgtable.h>
++#include <asm/system.h>
++#include <asm/io.h>
++#include <asm/processor.h>
++#include <asm/i387.h>
++#include <asm/mmu_context.h>
++#include <asm/pda.h>
++#include <asm/prctl.h>
++#include <asm/kdebug.h>
++#include <xen/interface/dom0_ops.h>
++#include <xen/interface/physdev.h>
++#include <xen/interface/vcpu.h>
++#include <asm/desc.h>
++#include <asm/proto.h>
++#include <asm/hardirq.h>
++#include <asm/ia32.h>
++#include <asm/idle.h>
++
++#include <xen/cpu_hotplug.h>
++
++asmlinkage extern void ret_from_fork(void);
++
++unsigned long kernel_thread_flags = CLONE_VM | CLONE_UNTRACED;
++
++unsigned long boot_option_idle_override = 0;
++EXPORT_SYMBOL(boot_option_idle_override);
++
++/*
++ * Powermanagement idle function, if any..
++ */
++void (*pm_idle)(void);
++static DEFINE_PER_CPU(unsigned int, cpu_idle_state);
++
++static ATOMIC_NOTIFIER_HEAD(idle_notifier);
++
++void idle_notifier_register(struct notifier_block *n)
++{
++ atomic_notifier_chain_register(&idle_notifier, n);
++}
++EXPORT_SYMBOL_GPL(idle_notifier_register);
++
++void idle_notifier_unregister(struct notifier_block *n)
++{
++ atomic_notifier_chain_unregister(&idle_notifier, n);
++}
++EXPORT_SYMBOL(idle_notifier_unregister);
++
++enum idle_state { CPU_IDLE, CPU_NOT_IDLE };
++static DEFINE_PER_CPU(enum idle_state, idle_state) = CPU_NOT_IDLE;
++
++void enter_idle(void)
++{
++ __get_cpu_var(idle_state) = CPU_IDLE;
++ atomic_notifier_call_chain(&idle_notifier, IDLE_START, NULL);
++}
++
++static void __exit_idle(void)
++{
++ __get_cpu_var(idle_state) = CPU_NOT_IDLE;
++ atomic_notifier_call_chain(&idle_notifier, IDLE_END, NULL);
++}
++
++/* Called from interrupts to signify idle end */
++void exit_idle(void)
++{
++ if (current->pid | read_pda(irqcount))
++ return;
++ __exit_idle();
++}
++
++/* XXX XEN doesn't use default_idle(), poll_idle(). Use xen_idle() instead. */
++void xen_idle(void)
++{
++ local_irq_disable();
++
++ if (need_resched())
++ local_irq_enable();
++ else {
++ clear_thread_flag(TIF_POLLING_NRFLAG);
++ smp_mb__after_clear_bit();
++ safe_halt();
++ set_thread_flag(TIF_POLLING_NRFLAG);
++ }
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++static inline void play_dead(void)
++{
++ idle_task_exit();
++ local_irq_disable();
++ cpu_clear(smp_processor_id(), cpu_initialized);
++ preempt_enable_no_resched();
++ HYPERVISOR_vcpu_op(VCPUOP_down, smp_processor_id(), NULL);
++ cpu_bringup();
++}
++#else
++static inline void play_dead(void)
++{
++ BUG();
++}
++#endif /* CONFIG_HOTPLUG_CPU */
++
++/*
++ * The idle thread. There's no useful work to be
++ * done, so just try to conserve power and have a
++ * low exit latency (ie sit in a loop waiting for
++ * somebody to say that they'd like to reschedule)
++ */
++void cpu_idle (void)
++{
++ set_thread_flag(TIF_POLLING_NRFLAG);
++
++ /* endless idle loop with no priority at all */
++ while (1) {
++ while (!need_resched()) {
++ if (__get_cpu_var(cpu_idle_state))
++ __get_cpu_var(cpu_idle_state) = 0;
++ rmb();
++
++ if (cpu_is_offline(smp_processor_id()))
++ play_dead();
++ enter_idle();
++ xen_idle();
++ __exit_idle();
++ }
++
++ preempt_enable_no_resched();
++ schedule();
++ preempt_disable();
++ }
++}
++
++void cpu_idle_wait(void)
++{
++ unsigned int cpu, this_cpu = get_cpu();
++ cpumask_t map;
++
++ set_cpus_allowed(current, cpumask_of_cpu(this_cpu));
++ put_cpu();
++
++ cpus_clear(map);
++ for_each_online_cpu(cpu) {
++ per_cpu(cpu_idle_state, cpu) = 1;
++ cpu_set(cpu, map);
++ }
++
++ __get_cpu_var(cpu_idle_state) = 0;
++
++ wmb();
++ do {
++ ssleep(1);
++ for_each_online_cpu(cpu) {
++ if (cpu_isset(cpu, map) &&
++ !per_cpu(cpu_idle_state, cpu))
++ cpu_clear(cpu, map);
++ }
++ cpus_and(map, map, cpu_online_map);
++ } while (!cpus_empty(map));
++}
++EXPORT_SYMBOL_GPL(cpu_idle_wait);
++
++/* XXX XEN doesn't use mwait_idle(), select_idle_routine(), idle_setup(). */
++/* Always use xen_idle() instead. */
++void __cpuinit select_idle_routine(const struct cpuinfo_x86 *c) {}
++
++/* Prints also some state that isn't saved in the pt_regs */
++void __show_regs(struct pt_regs * regs)
++{
++ unsigned long fs, gs, shadowgs;
++ unsigned int fsindex,gsindex;
++ unsigned int ds,cs,es;
++
++ printk("\n");
++ print_modules();
++ printk("Pid: %d, comm: %.20s %s %s %.*s\n",
++ current->pid, current->comm, print_tainted(),
++ system_utsname.release,
++ (int)strcspn(system_utsname.version, " "),
++ system_utsname.version);
++ printk("RIP: %04lx:[<%016lx>] ", regs->cs & 0xffff, regs->rip);
++ printk_address(regs->rip);
++ printk("\nRSP: %04lx:%016lx EFLAGS: %08lx\n", regs->ss, regs->rsp,
++ regs->eflags);
++ printk("RAX: %016lx RBX: %016lx RCX: %016lx\n",
++ regs->rax, regs->rbx, regs->rcx);
++ printk("RDX: %016lx RSI: %016lx RDI: %016lx\n",
++ regs->rdx, regs->rsi, regs->rdi);
++ printk("RBP: %016lx R08: %016lx R09: %016lx\n",
++ regs->rbp, regs->r8, regs->r9);
++ printk("R10: %016lx R11: %016lx R12: %016lx\n",
++ regs->r10, regs->r11, regs->r12);
++ printk("R13: %016lx R14: %016lx R15: %016lx\n",
++ regs->r13, regs->r14, regs->r15);
++
++ asm("mov %%ds,%0" : "=r" (ds));
++ asm("mov %%cs,%0" : "=r" (cs));
++ asm("mov %%es,%0" : "=r" (es));
++ asm("mov %%fs,%0" : "=r" (fsindex));
++ asm("mov %%gs,%0" : "=r" (gsindex));
++
++ rdmsrl(MSR_FS_BASE, fs);
++ rdmsrl(MSR_GS_BASE, gs);
++ rdmsrl(MSR_KERNEL_GS_BASE, shadowgs);
++
++ printk("FS: %016lx(%04x) GS:%016lx(%04x) knlGS:%016lx\n",
++ fs,fsindex,gs,gsindex,shadowgs);
++ printk("CS: %04x DS: %04x ES: %04x\n", cs, ds, es);
++
++}
++
++void show_regs(struct pt_regs *regs)
++{
++ printk("CPU %d:", smp_processor_id());
++ __show_regs(regs);
++ show_trace(®s->rsp);
++}
++
++/*
++ * Free current thread data structures etc..
++ */
++void exit_thread(void)
++{
++ struct task_struct *me = current;
++ struct thread_struct *t = &me->thread;
++
++ if (me->thread.io_bitmap_ptr) {
++#ifndef CONFIG_X86_NO_TSS
++ struct tss_struct *tss = &per_cpu(init_tss, get_cpu());
++#endif
++#ifdef CONFIG_XEN
++ struct physdev_set_iobitmap iobmp_op = { 0 };
++#endif
++
++ kfree(t->io_bitmap_ptr);
++ t->io_bitmap_ptr = NULL;
++ /*
++ * Careful, clear this in the TSS too:
++ */
++#ifndef CONFIG_X86_NO_TSS
++ memset(tss->io_bitmap, 0xff, t->io_bitmap_max);
++ put_cpu();
++#endif
++#ifdef CONFIG_XEN
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iobitmap, &iobmp_op);
++#endif
++ t->io_bitmap_max = 0;
++ }
++}
++
++void load_gs_index(unsigned gs)
++{
++ HYPERVISOR_set_segment_base(SEGBASE_GS_USER_SEL, gs);
++}
++
++void flush_thread(void)
++{
++ struct task_struct *tsk = current;
++ struct thread_info *t = current_thread_info();
++
++ if (t->flags & _TIF_ABI_PENDING)
++ t->flags ^= (_TIF_ABI_PENDING | _TIF_IA32);
++
++ tsk->thread.debugreg0 = 0;
++ tsk->thread.debugreg1 = 0;
++ tsk->thread.debugreg2 = 0;
++ tsk->thread.debugreg3 = 0;
++ tsk->thread.debugreg6 = 0;
++ tsk->thread.debugreg7 = 0;
++ memset(tsk->thread.tls_array, 0, sizeof(tsk->thread.tls_array));
++ /*
++ * Forget coprocessor state..
++ */
++ clear_fpu(tsk);
++ clear_used_math();
++}
++
++void release_thread(struct task_struct *dead_task)
++{
++ if (dead_task->mm) {
++ if (dead_task->mm->context.size) {
++ printk("WARNING: dead process %8s still has LDT? <%p/%d>\n",
++ dead_task->comm,
++ dead_task->mm->context.ldt,
++ dead_task->mm->context.size);
++ BUG();
++ }
++ }
++}
++
++static inline void set_32bit_tls(struct task_struct *t, int tls, u32 addr)
++{
++ struct user_desc ud = {
++ .base_addr = addr,
++ .limit = 0xfffff,
++ .contents = (3 << 3), /* user */
++ .seg_32bit = 1,
++ .limit_in_pages = 1,
++ .useable = 1,
++ };
++ struct n_desc_struct *desc = (void *)t->thread.tls_array;
++ desc += tls;
++ desc->a = LDT_entry_a(&ud);
++ desc->b = LDT_entry_b(&ud);
++}
++
++static inline u32 read_32bit_tls(struct task_struct *t, int tls)
++{
++ struct desc_struct *desc = (void *)t->thread.tls_array;
++ desc += tls;
++ return desc->base0 |
++ (((u32)desc->base1) << 16) |
++ (((u32)desc->base2) << 24);
++}
++
++/*
++ * This gets called before we allocate a new thread and copy
++ * the current task into it.
++ */
++void prepare_to_copy(struct task_struct *tsk)
++{
++ unlazy_fpu(tsk);
++}
++
++int copy_thread(int nr, unsigned long clone_flags, unsigned long rsp,
++ unsigned long unused,
++ struct task_struct * p, struct pt_regs * regs)
++{
++ int err;
++ struct pt_regs * childregs;
++ struct task_struct *me = current;
++
++ childregs = ((struct pt_regs *)
++ (THREAD_SIZE + task_stack_page(p))) - 1;
++ *childregs = *regs;
++
++ childregs->rax = 0;
++ childregs->rsp = rsp;
++ if (rsp == ~0UL)
++ childregs->rsp = (unsigned long)childregs;
++
++ p->thread.rsp = (unsigned long) childregs;
++ p->thread.rsp0 = (unsigned long) (childregs+1);
++ p->thread.userrsp = me->thread.userrsp;
++
++ set_tsk_thread_flag(p, TIF_FORK);
++
++ p->thread.fs = me->thread.fs;
++ p->thread.gs = me->thread.gs;
++
++ asm("mov %%gs,%0" : "=m" (p->thread.gsindex));
++ asm("mov %%fs,%0" : "=m" (p->thread.fsindex));
++ asm("mov %%es,%0" : "=m" (p->thread.es));
++ asm("mov %%ds,%0" : "=m" (p->thread.ds));
++
++ if (unlikely(me->thread.io_bitmap_ptr != NULL)) {
++ p->thread.io_bitmap_ptr = kmalloc(IO_BITMAP_BYTES, GFP_KERNEL);
++ if (!p->thread.io_bitmap_ptr) {
++ p->thread.io_bitmap_max = 0;
++ return -ENOMEM;
++ }
++ memcpy(p->thread.io_bitmap_ptr, me->thread.io_bitmap_ptr,
++ IO_BITMAP_BYTES);
++ }
++
++ /*
++ * Set a new TLS for the child thread?
++ */
++ if (clone_flags & CLONE_SETTLS) {
++#ifdef CONFIG_IA32_EMULATION
++ if (test_thread_flag(TIF_IA32))
++ err = ia32_child_tls(p, childregs);
++ else
++#endif
++ err = do_arch_prctl(p, ARCH_SET_FS, childregs->r8);
++ if (err)
++ goto out;
++ }
++ p->thread.iopl = current->thread.iopl;
++
++ err = 0;
++out:
++ if (err && p->thread.io_bitmap_ptr) {
++ kfree(p->thread.io_bitmap_ptr);
++ p->thread.io_bitmap_max = 0;
++ }
++ return err;
++}
++
++static inline void __save_init_fpu( struct task_struct *tsk )
++{
++ asm volatile( "rex64 ; fxsave %0 ; fnclex"
++ : "=m" (tsk->thread.i387.fxsave));
++ tsk->thread_info->status &= ~TS_USEDFPU;
++}
++
++/*
++ * switch_to(x,y) should switch tasks from x to y.
++ *
++ * This could still be optimized:
++ * - fold all the options into a flag word and test it with a single test.
++ * - could test fs/gs bitsliced
++ *
++ * Kprobes not supported here. Set the probe on schedule instead.
++ */
++__kprobes struct task_struct *
++__switch_to(struct task_struct *prev_p, struct task_struct *next_p)
++{
++ struct thread_struct *prev = &prev_p->thread,
++ *next = &next_p->thread;
++ int cpu = smp_processor_id();
++#ifndef CONFIG_X86_NO_TSS
++ struct tss_struct *tss = &per_cpu(init_tss, cpu);
++#endif
++ struct physdev_set_iopl iopl_op;
++ struct physdev_set_iobitmap iobmp_op;
++ multicall_entry_t _mcl[8], *mcl = _mcl;
++
++ /*
++ * This is basically '__unlazy_fpu', except that we queue a
++ * multicall to indicate FPU task switch, rather than
++ * synchronously trapping to Xen.
++ */
++ if (prev_p->thread_info->status & TS_USEDFPU) {
++ __save_init_fpu(prev_p); /* _not_ save_init_fpu() */
++ mcl->op = __HYPERVISOR_fpu_taskswitch;
++ mcl->args[0] = 1;
++ mcl++;
++ }
++
++ /*
++ * Reload esp0, LDT and the page table pointer:
++ */
++ mcl->op = __HYPERVISOR_stack_switch;
++ mcl->args[0] = __KERNEL_DS;
++ mcl->args[1] = next->rsp0;
++ mcl++;
++
++ /*
++ * Load the per-thread Thread-Local Storage descriptor.
++ * This is load_TLS(next, cpu) with multicalls.
++ */
++#define C(i) do { \
++ if (unlikely(next->tls_array[i] != prev->tls_array[i])) { \
++ mcl->op = __HYPERVISOR_update_descriptor; \
++ mcl->args[0] = virt_to_machine( \
++ &cpu_gdt(cpu)[GDT_ENTRY_TLS_MIN + i]); \
++ mcl->args[1] = next->tls_array[i]; \
++ mcl++; \
++ } \
++} while (0)
++ C(0); C(1); C(2);
++#undef C
++
++ if (unlikely(prev->iopl != next->iopl)) {
++ iopl_op.iopl = (next->iopl == 0) ? 1 : next->iopl;
++ mcl->op = __HYPERVISOR_physdev_op;
++ mcl->args[0] = PHYSDEVOP_set_iopl;
++ mcl->args[1] = (unsigned long)&iopl_op;
++ mcl++;
++ }
++
++ if (unlikely(prev->io_bitmap_ptr || next->io_bitmap_ptr)) {
++ iobmp_op.bitmap = (char *)next->io_bitmap_ptr;
++ iobmp_op.nr_ports = next->io_bitmap_ptr ? IO_BITMAP_BITS : 0;
++ mcl->op = __HYPERVISOR_physdev_op;
++ mcl->args[0] = PHYSDEVOP_set_iobitmap;
++ mcl->args[1] = (unsigned long)&iobmp_op;
++ mcl++;
++ }
++
++ (void)HYPERVISOR_multicall(_mcl, mcl - _mcl);
++ /*
++ * Switch DS and ES.
++ * This won't pick up thread selector changes, but I guess that is ok.
++ */
++ if (unlikely(next->es))
++ loadsegment(es, next->es);
++
++ if (unlikely(next->ds))
++ loadsegment(ds, next->ds);
++
++ /*
++ * Switch FS and GS.
++ */
++ if (unlikely(next->fsindex))
++ loadsegment(fs, next->fsindex);
++
++ if (next->fs)
++ HYPERVISOR_set_segment_base(SEGBASE_FS, next->fs);
++
++ if (unlikely(next->gsindex))
++ load_gs_index(next->gsindex);
++
++ if (next->gs)
++ HYPERVISOR_set_segment_base(SEGBASE_GS_USER, next->gs);
++
++ /*
++ * Switch the PDA.
++ */
++ prev->userrsp = read_pda(oldrsp);
++ write_pda(oldrsp, next->userrsp);
++ write_pda(pcurrent, next_p);
++ write_pda(kernelstack,
++ task_stack_page(next_p) + THREAD_SIZE - PDA_STACKOFFSET);
++
++ /*
++ * Now maybe reload the debug registers
++ */
++ if (unlikely(next->debugreg7)) {
++ set_debugreg(next->debugreg0, 0);
++ set_debugreg(next->debugreg1, 1);
++ set_debugreg(next->debugreg2, 2);
++ set_debugreg(next->debugreg3, 3);
++ /* no 4 and 5 */
++ set_debugreg(next->debugreg6, 6);
++ set_debugreg(next->debugreg7, 7);
++ }
++
++ return prev_p;
++}
++
++/*
++ * sys_execve() executes a new program.
++ */
++asmlinkage
++long sys_execve(char __user *name, char __user * __user *argv,
++ char __user * __user *envp, struct pt_regs regs)
++{
++ long error;
++ char * filename;
++
++ filename = getname(name);
++ error = PTR_ERR(filename);
++ if (IS_ERR(filename))
++ return error;
++ error = do_execve(filename, argv, envp, ®s);
++ if (error == 0) {
++ task_lock(current);
++ current->ptrace &= ~PT_DTRACE;
++ task_unlock(current);
++ }
++ putname(filename);
++ return error;
++}
++
++void set_personality_64bit(void)
++{
++ /* inherit personality from parent */
++
++ /* Make sure to be in 64bit mode */
++ clear_thread_flag(TIF_IA32);
++
++ /* TBD: overwrites user setup. Should have two bits.
++ But 64bit processes have always behaved this way,
++ so it's not too bad. The main problem is just that
++ 32bit childs are affected again. */
++ current->personality &= ~READ_IMPLIES_EXEC;
++}
++
++asmlinkage long sys_fork(struct pt_regs *regs)
++{
++ return do_fork(SIGCHLD, regs->rsp, regs, 0, NULL, NULL);
++}
++
++asmlinkage long
++sys_clone(unsigned long clone_flags, unsigned long newsp,
++ void __user *parent_tid, void __user *child_tid, struct pt_regs *regs)
++{
++ if (!newsp)
++ newsp = regs->rsp;
++ return do_fork(clone_flags, newsp, regs, 0, parent_tid, child_tid);
++}
++
++/*
++ * This is trivial, and on the face of it looks like it
++ * could equally well be done in user mode.
++ *
++ * Not so, for quite unobvious reasons - register pressure.
++ * In user mode vfork() cannot have a stack frame, and if
++ * done by calling the "clone()" system call directly, you
++ * do not have enough call-clobbered registers to hold all
++ * the information you need.
++ */
++asmlinkage long sys_vfork(struct pt_regs *regs)
++{
++ return do_fork(CLONE_VFORK | CLONE_VM | SIGCHLD, regs->rsp, regs, 0,
++ NULL, NULL);
++}
++
++unsigned long get_wchan(struct task_struct *p)
++{
++ unsigned long stack;
++ u64 fp,rip;
++ int count = 0;
++
++ if (!p || p == current || p->state==TASK_RUNNING)
++ return 0;
++ stack = (unsigned long)task_stack_page(p);
++ if (p->thread.rsp < stack || p->thread.rsp > stack+THREAD_SIZE)
++ return 0;
++ fp = *(u64 *)(p->thread.rsp);
++ do {
++ if (fp < (unsigned long)stack ||
++ fp > (unsigned long)stack+THREAD_SIZE)
++ return 0;
++ rip = *(u64 *)(fp+8);
++ if (!in_sched_functions(rip))
++ return rip;
++ fp = *(u64 *)fp;
++ } while (count++ < 16);
++ return 0;
++}
++
++long do_arch_prctl(struct task_struct *task, int code, unsigned long addr)
++{
++ int ret = 0;
++ int doit = task == current;
++ int cpu;
++
++ switch (code) {
++ case ARCH_SET_GS:
++ if (addr >= TASK_SIZE_OF(task))
++ return -EPERM;
++ cpu = get_cpu();
++ /* handle small bases via the GDT because that's faster to
++ switch. */
++ if (addr <= 0xffffffff) {
++ set_32bit_tls(task, GS_TLS, addr);
++ if (doit) {
++ load_TLS(&task->thread, cpu);
++ load_gs_index(GS_TLS_SEL);
++ }
++ task->thread.gsindex = GS_TLS_SEL;
++ task->thread.gs = 0;
++ } else {
++ task->thread.gsindex = 0;
++ task->thread.gs = addr;
++ if (doit) {
++ load_gs_index(0);
++ ret = HYPERVISOR_set_segment_base(
++ SEGBASE_GS_USER, addr);
++ }
++ }
++ put_cpu();
++ break;
++ case ARCH_SET_FS:
++ /* Not strictly needed for fs, but do it for symmetry
++ with gs */
++ if (addr >= TASK_SIZE_OF(task))
++ return -EPERM;
++ cpu = get_cpu();
++ /* handle small bases via the GDT because that's faster to
++ switch. */
++ if (addr <= 0xffffffff) {
++ set_32bit_tls(task, FS_TLS, addr);
++ if (doit) {
++ load_TLS(&task->thread, cpu);
++ asm volatile("movl %0,%%fs" :: "r"(FS_TLS_SEL));
++ }
++ task->thread.fsindex = FS_TLS_SEL;
++ task->thread.fs = 0;
++ } else {
++ task->thread.fsindex = 0;
++ task->thread.fs = addr;
++ if (doit) {
++ /* set the selector to 0 to not confuse
++ __switch_to */
++ asm volatile("movl %0,%%fs" :: "r" (0));
++ ret = HYPERVISOR_set_segment_base(SEGBASE_FS,
++ addr);
++ }
++ }
++ put_cpu();
++ break;
++ case ARCH_GET_FS: {
++ unsigned long base;
++ if (task->thread.fsindex == FS_TLS_SEL)
++ base = read_32bit_tls(task, FS_TLS);
++ else if (doit)
++ rdmsrl(MSR_FS_BASE, base);
++ else
++ base = task->thread.fs;
++ ret = put_user(base, (unsigned long __user *)addr);
++ break;
++ }
++ case ARCH_GET_GS: {
++ unsigned long base;
++ unsigned gsindex;
++ if (task->thread.gsindex == GS_TLS_SEL)
++ base = read_32bit_tls(task, GS_TLS);
++ else if (doit) {
++ asm("movl %%gs,%0" : "=r" (gsindex));
++ if (gsindex)
++ rdmsrl(MSR_KERNEL_GS_BASE, base);
++ else
++ base = task->thread.gs;
++ }
++ else
++ base = task->thread.gs;
++ ret = put_user(base, (unsigned long __user *)addr);
++ break;
++ }
++
++ default:
++ ret = -EINVAL;
++ break;
++ }
++
++ return ret;
++}
++
++long sys_arch_prctl(int code, unsigned long addr)
++{
++ return do_arch_prctl(current, code, addr);
++}
++
++/*
++ * Capture the user space registers if the task is not running (in user space)
++ */
++int dump_task_regs(struct task_struct *tsk, elf_gregset_t *regs)
++{
++ struct pt_regs *pp, ptregs;
++
++ pp = task_pt_regs(tsk);
++
++ ptregs = *pp;
++ ptregs.cs &= 0xffff;
++ ptregs.ss &= 0xffff;
++
++ elf_core_copy_regs(regs, &ptregs);
++
++ boot_option_idle_override = 1;
++ return 1;
++}
++
++unsigned long arch_align_stack(unsigned long sp)
++{
++ if (randomize_va_space)
++ sp -= get_random_int() % 8192;
++ return sp & ~0xf;
++}
++
++#ifndef CONFIG_SMP
++void _restore_vcpu(void)
++{
++}
++#endif
+diff -urNp linux-2.6/arch/x86_64/kernel/setup64-xen.c new/arch/x86_64/kernel/setup64-xen.c
+--- linux-2.6/arch/x86_64/kernel/setup64-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/setup64-xen.c 2006-05-09 12:33:21.000000000 +0200
+@@ -0,0 +1,360 @@
++/*
++ * X86-64 specific CPU setup.
++ * Copyright (C) 1995 Linus Torvalds
++ * Copyright 2001, 2002, 2003 SuSE Labs / Andi Kleen.
++ * See setup.c for older changelog.
++ * $Id$
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Modified for Xen
++ *
++ */
++#include <linux/config.h>
++#include <linux/init.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/string.h>
++#include <linux/bootmem.h>
++#include <linux/bitops.h>
++#include <linux/module.h>
++#include <asm/bootsetup.h>
++#include <asm/pda.h>
++#include <asm/pgtable.h>
++#include <asm/processor.h>
++#include <asm/desc.h>
++#include <asm/atomic.h>
++#include <asm/mmu_context.h>
++#include <asm/smp.h>
++#include <asm/i387.h>
++#include <asm/percpu.h>
++#include <asm/proto.h>
++#include <asm/sections.h>
++#ifdef CONFIG_XEN
++#include <asm/hypervisor.h>
++#endif
++
++char x86_boot_params[BOOT_PARAM_SIZE] __initdata = {0,};
++
++cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
++
++struct x8664_pda *_cpu_pda[NR_CPUS] __read_mostly;
++struct x8664_pda boot_cpu_pda[NR_CPUS] __cacheline_aligned;
++
++#ifndef CONFIG_X86_NO_IDT
++struct desc_ptr idt_descr = { 256 * 16 - 1, (unsigned long) idt_table };
++#endif
++
++char boot_cpu_stack[IRQSTACKSIZE] __attribute__((section(".bss.page_aligned")));
++
++unsigned long __supported_pte_mask __read_mostly = ~0UL;
++static int do_not_nx __cpuinitdata = 0;
++
++/* noexec=on|off
++Control non executable mappings for 64bit processes.
++
++on Enable(default)
++off Disable
++*/
++int __init nonx_setup(char *str)
++{
++ if (!strncmp(str, "on", 2)) {
++ __supported_pte_mask |= _PAGE_NX;
++ do_not_nx = 0;
++ } else if (!strncmp(str, "off", 3)) {
++ do_not_nx = 1;
++ __supported_pte_mask &= ~_PAGE_NX;
++ }
++ return 1;
++}
++__setup("noexec=", nonx_setup); /* parsed early actually */
++
++int force_personality32 = 0;
++
++/* noexec32=on|off
++Control non executable heap for 32bit processes.
++To control the stack too use noexec=off
++
++on PROT_READ does not imply PROT_EXEC for 32bit processes
++off PROT_READ implies PROT_EXEC (default)
++*/
++static int __init nonx32_setup(char *str)
++{
++ if (!strcmp(str, "on"))
++ force_personality32 &= ~READ_IMPLIES_EXEC;
++ else if (!strcmp(str, "off"))
++ force_personality32 |= READ_IMPLIES_EXEC;
++ return 1;
++}
++__setup("noexec32=", nonx32_setup);
++
++/*
++ * Great future plan:
++ * Declare PDA itself and support (irqstack,tss,pgd) as per cpu data.
++ * Always point %gs to its beginning
++ */
++void __init setup_per_cpu_areas(void)
++{
++ int i;
++ unsigned long size;
++
++#ifdef CONFIG_HOTPLUG_CPU
++ prefill_possible_map();
++#endif
++
++ /* Copy section for each CPU (we discard the original) */
++ size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
++#ifdef CONFIG_MODULES
++ if (size < PERCPU_ENOUGH_ROOM)
++ size = PERCPU_ENOUGH_ROOM;
++#endif
++
++ for_each_cpu_mask (i, cpu_possible_map) {
++ char *ptr;
++
++ if (!NODE_DATA(cpu_to_node(i))) {
++ printk("cpu with no node %d, num_online_nodes %d\n",
++ i, num_online_nodes());
++ ptr = alloc_bootmem(size);
++ } else {
++ ptr = alloc_bootmem_node(NODE_DATA(cpu_to_node(i)), size);
++ }
++ if (!ptr)
++ panic("Cannot allocate cpu data for CPU %d\n", i);
++ cpu_pda(i)->data_offset = ptr - __per_cpu_start;
++ memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
++ }
++}
++
++#ifdef CONFIG_XEN
++static void switch_pt(void)
++{
++ xen_pt_switch(__pa(init_level4_pgt));
++ xen_new_user_pt(__pa(init_level4_user_pgt));
++}
++
++void __cpuinit cpu_gdt_init(struct desc_ptr *gdt_descr)
++{
++ unsigned long frames[16];
++ unsigned long va;
++ int f;
++
++ for (va = gdt_descr->address, f = 0;
++ va < gdt_descr->address + gdt_descr->size;
++ va += PAGE_SIZE, f++) {
++ frames[f] = virt_to_mfn(va);
++ make_page_readonly(
++ (void *)va, XENFEAT_writable_descriptor_tables);
++ }
++ if (HYPERVISOR_set_gdt(frames, gdt_descr->size /
++ sizeof (struct desc_struct)))
++ BUG();
++}
++#else
++static void switch_pt(void)
++{
++ asm volatile("movq %0,%%cr3" :: "r" (__pa_symbol(&init_level4_pgt)));
++}
++
++void __init cpu_gdt_init(struct desc_ptr *gdt_descr)
++{
++ asm volatile("lgdt %0" :: "m" (*gdt_descr));
++ asm volatile("lidt %0" :: "m" (idt_descr));
++}
++#endif
++
++void pda_init(int cpu)
++{
++ struct x8664_pda *pda = cpu_pda(cpu);
++
++ /* Setup up data that may be needed in __get_free_pages early */
++ asm volatile("movl %0,%%fs ; movl %0,%%gs" :: "r" (0));
++#ifndef CONFIG_XEN
++ wrmsrl(MSR_GS_BASE, pda);
++#else
++ HYPERVISOR_set_segment_base(SEGBASE_GS_KERNEL, (unsigned long)pda);
++#endif
++ pda->cpunumber = cpu;
++ pda->irqcount = -1;
++ pda->kernelstack =
++ (unsigned long)stack_thread_info() - PDA_STACKOFFSET + THREAD_SIZE;
++ pda->active_mm = &init_mm;
++ pda->mmu_state = 0;
++
++ if (cpu == 0) {
++#ifdef CONFIG_XEN
++ xen_init_pt();
++#endif
++ /* others are initialized in smpboot.c */
++ pda->pcurrent = &init_task;
++ pda->irqstackptr = boot_cpu_stack;
++ } else {
++ pda->irqstackptr = (char *)
++ __get_free_pages(GFP_ATOMIC, IRQSTACK_ORDER);
++ if (!pda->irqstackptr)
++ panic("cannot allocate irqstack for cpu %d", cpu);
++ }
++
++ switch_pt();
++
++ pda->irqstackptr += IRQSTACKSIZE-64;
++}
++
++#ifndef CONFIG_X86_NO_TSS
++char boot_exception_stacks[(N_EXCEPTION_STACKS - 1) * EXCEPTION_STKSZ + DEBUG_STKSZ]
++__attribute__((section(".bss.page_aligned")));
++#endif
++
++/* May not be marked __init: used by software suspend */
++void syscall_init(void)
++{
++#ifndef CONFIG_XEN
++ /*
++ * LSTAR and STAR live in a bit strange symbiosis.
++ * They both write to the same internal register. STAR allows to set CS/DS
++ * but only a 32bit target. LSTAR sets the 64bit rip.
++ */
++ wrmsrl(MSR_STAR, ((u64)__USER32_CS)<<48 | ((u64)__KERNEL_CS)<<32);
++ wrmsrl(MSR_LSTAR, system_call);
++
++ /* Flags to clear on syscall */
++ wrmsrl(MSR_SYSCALL_MASK, EF_TF|EF_DF|EF_IE|0x3000);
++#endif
++#ifdef CONFIG_IA32_EMULATION
++ syscall32_cpu_init ();
++#endif
++}
++
++void __cpuinit check_efer(void)
++{
++ unsigned long efer;
++
++ rdmsrl(MSR_EFER, efer);
++ if (!(efer & EFER_NX) || do_not_nx) {
++ __supported_pte_mask &= ~_PAGE_NX;
++ }
++}
++
++/*
++ * cpu_init() initializes state that is per-CPU. Some data is already
++ * initialized (naturally) in the bootstrap process, such as the GDT
++ * and IDT. We reload them nevertheless, this function acts as a
++ * 'CPU state barrier', nothing should get across.
++ * A lot of state is already set up in PDA init.
++ */
++void __cpuinit cpu_init (void)
++{
++ int cpu = stack_smp_processor_id();
++#ifndef CONFIG_X86_NO_TSS
++ struct tss_struct *t = &per_cpu(init_tss, cpu);
++ unsigned long v;
++ char *estacks = NULL;
++ unsigned i;
++#endif
++ struct task_struct *me;
++
++ /* CPU 0 is initialised in head64.c */
++ if (cpu != 0) {
++ pda_init(cpu);
++ zap_low_mappings(cpu);
++ }
++#ifndef CONFIG_X86_NO_TSS
++ else
++ estacks = boot_exception_stacks;
++#endif
++
++ me = current;
++
++ if (cpu_test_and_set(cpu, cpu_initialized))
++ panic("CPU#%d already initialized!\n", cpu);
++
++ printk("Initializing CPU#%d\n", cpu);
++
++ clear_in_cr4(X86_CR4_VME|X86_CR4_PVI|X86_CR4_TSD|X86_CR4_DE);
++
++ /*
++ * Initialize the per-CPU GDT with the boot GDT,
++ * and set up the GDT descriptor:
++ */
++#ifndef CONFIG_XEN
++ if (cpu)
++ memcpy(cpu_gdt(cpu), cpu_gdt_table, GDT_SIZE);
++#endif
++
++ cpu_gdt_descr[cpu].size = GDT_SIZE;
++ cpu_gdt_init(&cpu_gdt_descr[cpu]);
++
++ memset(me->thread.tls_array, 0, GDT_ENTRY_TLS_ENTRIES * 8);
++ syscall_init();
++
++ wrmsrl(MSR_FS_BASE, 0);
++ wrmsrl(MSR_KERNEL_GS_BASE, 0);
++ barrier();
++
++ check_efer();
++
++#ifndef CONFIG_X86_NO_TSS
++ /*
++ * set up and load the per-CPU TSS
++ */
++ for (v = 0; v < N_EXCEPTION_STACKS; v++) {
++ if (cpu) {
++ static const unsigned int order[N_EXCEPTION_STACKS] = {
++ [0 ... N_EXCEPTION_STACKS - 1] = EXCEPTION_STACK_ORDER,
++ [DEBUG_STACK - 1] = DEBUG_STACK_ORDER
++ };
++
++ estacks = (char *)__get_free_pages(GFP_ATOMIC, order[v]);
++ if (!estacks)
++ panic("Cannot allocate exception stack %ld %d\n",
++ v, cpu);
++ }
++ switch (v + 1) {
++#if DEBUG_STKSZ > EXCEPTION_STKSZ
++ case DEBUG_STACK:
++ cpu_pda(cpu)->debugstack = (unsigned long)estacks;
++ estacks += DEBUG_STKSZ;
++ break;
++#endif
++ default:
++ estacks += EXCEPTION_STKSZ;
++ break;
++ }
++ t->ist[v] = (unsigned long)estacks;
++ }
++
++ t->io_bitmap_base = offsetof(struct tss_struct, io_bitmap);
++ /*
++ * <= is required because the CPU will access up to
++ * 8 bits beyond the end of the IO permission bitmap.
++ */
++ for (i = 0; i <= IO_BITMAP_LONGS; i++)
++ t->io_bitmap[i] = ~0UL;
++#endif
++
++ atomic_inc(&init_mm.mm_count);
++ me->active_mm = &init_mm;
++ if (me->mm)
++ BUG();
++ enter_lazy_tlb(&init_mm, me);
++
++#ifndef CONFIG_X86_NO_TSS
++ set_tss_desc(cpu, t);
++#endif
++#ifndef CONFIG_XEN
++ load_TR_desc();
++#endif
++ load_LDT(&init_mm.context);
++
++ /*
++ * Clear all 6 debug registers:
++ */
++
++ set_debugreg(0UL, 0);
++ set_debugreg(0UL, 1);
++ set_debugreg(0UL, 2);
++ set_debugreg(0UL, 3);
++ set_debugreg(0UL, 6);
++ set_debugreg(0UL, 7);
++
++ fpu_init();
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/setup-xen.c new/arch/x86_64/kernel/setup-xen.c
+--- linux-2.6/arch/x86_64/kernel/setup-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/setup-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,1710 @@
++/*
++ * linux/arch/x86-64/kernel/setup.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ *
++ * Nov 2001 Dave Jones <davej@suse.de>
++ * Forked from i386 setup code.
++ *
++ * $Id$
++ */
++
++/*
++ * This file handles the architecture-dependent parts of initialization
++ */
++
++#include <linux/errno.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/mm.h>
++#include <linux/stddef.h>
++#include <linux/unistd.h>
++#include <linux/ptrace.h>
++#include <linux/slab.h>
++#include <linux/user.h>
++#include <linux/a.out.h>
++#include <linux/tty.h>
++#include <linux/ioport.h>
++#include <linux/delay.h>
++#include <linux/config.h>
++#include <linux/init.h>
++#include <linux/initrd.h>
++#include <linux/highmem.h>
++#include <linux/bootmem.h>
++#include <linux/module.h>
++#include <asm/processor.h>
++#include <linux/console.h>
++#include <linux/seq_file.h>
++#include <linux/crash_dump.h>
++#include <linux/root_dev.h>
++#include <linux/pci.h>
++#include <linux/acpi.h>
++#include <linux/kallsyms.h>
++#include <linux/edd.h>
++#include <linux/mmzone.h>
++#include <linux/kexec.h>
++#include <linux/cpufreq.h>
++#include <linux/dmi.h>
++#include <linux/dma-mapping.h>
++#include <linux/ctype.h>
++
++#include <asm/mtrr.h>
++#include <asm/uaccess.h>
++#include <asm/system.h>
++#include <asm/io.h>
++#include <asm/smp.h>
++#include <asm/msr.h>
++#include <asm/desc.h>
++#include <video/edid.h>
++#include <asm/e820.h>
++#include <asm/dma.h>
++#include <asm/mpspec.h>
++#include <asm/mmu_context.h>
++#include <asm/bootsetup.h>
++#include <asm/proto.h>
++#include <asm/setup.h>
++#include <asm/mach_apic.h>
++#include <asm/numa.h>
++#include <asm/swiotlb.h>
++#include <asm/sections.h>
++#include <asm/gart-mapping.h>
++#include <asm/dmi.h>
++#ifdef CONFIG_XEN
++#include <linux/percpu.h>
++#include <xen/interface/physdev.h>
++#include "setup_arch_pre.h"
++#include <asm/hypervisor.h>
++#include <xen/interface/nmi.h>
++#include <xen/features.h>
++#define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT)
++#define PFN_PHYS(x) ((x) << PAGE_SHIFT)
++#include <asm/mach-xen/setup_arch_post.h>
++#include <xen/interface/memory.h>
++
++extern unsigned long start_pfn;
++extern struct edid_info edid_info;
++
++shared_info_t *HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
++EXPORT_SYMBOL(HYPERVISOR_shared_info);
++
++extern char hypercall_page[PAGE_SIZE];
++EXPORT_SYMBOL(hypercall_page);
++
++/* Allows setting of maximum possible memory size */
++unsigned long xen_override_max_pfn;
++
++static int xen_panic_event(struct notifier_block *, unsigned long, void *);
++static struct notifier_block xen_panic_block = {
++ xen_panic_event, NULL, 0 /* try to go last */
++};
++
++unsigned long *phys_to_machine_mapping;
++unsigned long *pfn_to_mfn_frame_list_list, *pfn_to_mfn_frame_list[512];
++
++EXPORT_SYMBOL(phys_to_machine_mapping);
++
++DEFINE_PER_CPU(multicall_entry_t, multicall_list[8]);
++DEFINE_PER_CPU(int, nr_multicall_ents);
++
++/* Raw start-of-day parameters from the hypervisor. */
++start_info_t *xen_start_info;
++EXPORT_SYMBOL(xen_start_info);
++#endif
++
++/*
++ * Machine setup..
++ */
++
++struct cpuinfo_x86 boot_cpu_data __read_mostly;
++
++unsigned long mmu_cr4_features;
++
++int acpi_disabled;
++EXPORT_SYMBOL(acpi_disabled);
++#ifdef CONFIG_ACPI
++extern int __initdata acpi_ht;
++extern acpi_interrupt_flags acpi_sci_flags;
++int __initdata acpi_force = 0;
++#endif
++
++int acpi_numa __initdata;
++
++/* Boot loader ID as an integer, for the benefit of proc_dointvec */
++int bootloader_type;
++
++unsigned long saved_video_mode;
++
++/*
++ * Early DMI memory
++ */
++int dmi_alloc_index;
++char dmi_alloc_data[DMI_MAX_DATA];
++
++/*
++ * Setup options
++ */
++struct screen_info screen_info;
++struct sys_desc_table_struct {
++ unsigned short length;
++ unsigned char table[0];
++};
++
++struct edid_info edid_info;
++struct e820map e820;
++
++extern int root_mountflags;
++
++char command_line[COMMAND_LINE_SIZE];
++
++struct resource standard_io_resources[] = {
++ { .name = "dma1", .start = 0x00, .end = 0x1f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "pic1", .start = 0x20, .end = 0x21,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "timer0", .start = 0x40, .end = 0x43,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "timer1", .start = 0x50, .end = 0x53,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "keyboard", .start = 0x60, .end = 0x6f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "dma page reg", .start = 0x80, .end = 0x8f,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "pic2", .start = 0xa0, .end = 0xa1,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "dma2", .start = 0xc0, .end = 0xdf,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO },
++ { .name = "fpu", .start = 0xf0, .end = 0xff,
++ .flags = IORESOURCE_BUSY | IORESOURCE_IO }
++};
++
++#define STANDARD_IO_RESOURCES \
++ (sizeof standard_io_resources / sizeof standard_io_resources[0])
++
++#define IORESOURCE_RAM (IORESOURCE_BUSY | IORESOURCE_MEM)
++
++struct resource data_resource = {
++ .name = "Kernel data",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_RAM,
++};
++struct resource code_resource = {
++ .name = "Kernel code",
++ .start = 0,
++ .end = 0,
++ .flags = IORESOURCE_RAM,
++};
++
++#define IORESOURCE_ROM (IORESOURCE_BUSY | IORESOURCE_READONLY | IORESOURCE_MEM)
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST) || !defined(CONFIG_XEN)
++static struct resource system_rom_resource = {
++ .name = "System ROM",
++ .start = 0xf0000,
++ .end = 0xfffff,
++ .flags = IORESOURCE_ROM,
++};
++
++static struct resource extension_rom_resource = {
++ .name = "Extension ROM",
++ .start = 0xe0000,
++ .end = 0xeffff,
++ .flags = IORESOURCE_ROM,
++};
++
++static struct resource adapter_rom_resources[] = {
++ { .name = "Adapter ROM", .start = 0xc8000, .end = 0,
++ .flags = IORESOURCE_ROM },
++ { .name = "Adapter ROM", .start = 0, .end = 0,
++ .flags = IORESOURCE_ROM },
++ { .name = "Adapter ROM", .start = 0, .end = 0,
++ .flags = IORESOURCE_ROM },
++ { .name = "Adapter ROM", .start = 0, .end = 0,
++ .flags = IORESOURCE_ROM },
++ { .name = "Adapter ROM", .start = 0, .end = 0,
++ .flags = IORESOURCE_ROM },
++ { .name = "Adapter ROM", .start = 0, .end = 0,
++ .flags = IORESOURCE_ROM }
++};
++#endif
++
++#define ADAPTER_ROM_RESOURCES \
++ (sizeof adapter_rom_resources / sizeof adapter_rom_resources[0])
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST) || !defined(CONFIG_XEN)
++static struct resource video_rom_resource = {
++ .name = "Video ROM",
++ .start = 0xc0000,
++ .end = 0xc7fff,
++ .flags = IORESOURCE_ROM,
++};
++#endif
++
++static struct resource video_ram_resource = {
++ .name = "Video RAM area",
++ .start = 0xa0000,
++ .end = 0xbffff,
++ .flags = IORESOURCE_RAM,
++};
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST) || !defined(CONFIG_XEN)
++#define romsignature(x) (*(unsigned short *)(x) == 0xaa55)
++
++static int __init romchecksum(unsigned char *rom, unsigned long length)
++{
++ unsigned char *p, sum = 0;
++
++ for (p = rom; p < rom + length; p++)
++ sum += *p;
++ return sum == 0;
++}
++
++static void __init probe_roms(void)
++{
++ unsigned long start, length, upper;
++ unsigned char *rom;
++ int i;
++
++ /* video rom */
++ upper = adapter_rom_resources[0].start;
++ for (start = video_rom_resource.start; start < upper; start += 2048) {
++ rom = isa_bus_to_virt(start);
++ if (!romsignature(rom))
++ continue;
++
++ video_rom_resource.start = start;
++
++ /* 0 < length <= 0x7f * 512, historically */
++ length = rom[2] * 512;
++
++ /* if checksum okay, trust length byte */
++ if (length && romchecksum(rom, length))
++ video_rom_resource.end = start + length - 1;
++
++ request_resource(&iomem_resource, &video_rom_resource);
++ break;
++ }
++
++ start = (video_rom_resource.end + 1 + 2047) & ~2047UL;
++ if (start < upper)
++ start = upper;
++
++ /* system rom */
++ request_resource(&iomem_resource, &system_rom_resource);
++ upper = system_rom_resource.start;
++
++ /* check for extension rom (ignore length byte!) */
++ rom = isa_bus_to_virt(extension_rom_resource.start);
++ if (romsignature(rom)) {
++ length = extension_rom_resource.end - extension_rom_resource.start + 1;
++ if (romchecksum(rom, length)) {
++ request_resource(&iomem_resource, &extension_rom_resource);
++ upper = extension_rom_resource.start;
++ }
++ }
++
++ /* check for adapter roms on 2k boundaries */
++ for (i = 0; i < ADAPTER_ROM_RESOURCES && start < upper; start += 2048) {
++ rom = isa_bus_to_virt(start);
++ if (!romsignature(rom))
++ continue;
++
++ /* 0 < length <= 0x7f * 512, historically */
++ length = rom[2] * 512;
++
++ /* but accept any length that fits if checksum okay */
++ if (!length || start + length > upper || !romchecksum(rom, length))
++ continue;
++
++ adapter_rom_resources[i].start = start;
++ adapter_rom_resources[i].end = start + length - 1;
++ request_resource(&iomem_resource, &adapter_rom_resources[i]);
++
++ start = adapter_rom_resources[i++].end & ~2047UL;
++ }
++}
++#endif
++
++/* Check for full argument with no trailing characters */
++static int fullarg(char *p, char *arg)
++{
++ int l = strlen(arg);
++ return !memcmp(p, arg, l) && (p[l] == 0 || isspace(p[l]));
++}
++
++static __init void parse_cmdline_early (char ** cmdline_p)
++{
++ char c = ' ', *to = command_line, *from = COMMAND_LINE;
++ int len = 0;
++ int userdef = 0;
++
++ for (;;) {
++ if (c != ' ')
++ goto next_char;
++
++#ifdef CONFIG_SMP
++ /*
++ * If the BIOS enumerates physical processors before logical,
++ * maxcpus=N at enumeration-time can be used to disable HT.
++ */
++ else if (!memcmp(from, "maxcpus=", 8)) {
++ extern unsigned int maxcpus;
++
++ maxcpus = simple_strtoul(from + 8, NULL, 0);
++ }
++#endif
++#ifdef CONFIG_ACPI
++ /* "acpi=off" disables both ACPI table parsing and interpreter init */
++ if (fullarg(from,"acpi=off"))
++ disable_acpi();
++
++ if (fullarg(from, "acpi=force")) {
++ /* add later when we do DMI horrors: */
++ acpi_force = 1;
++ acpi_disabled = 0;
++ }
++
++ /* acpi=ht just means: do ACPI MADT parsing
++ at bootup, but don't enable the full ACPI interpreter */
++ if (fullarg(from, "acpi=ht")) {
++ if (!acpi_force)
++ disable_acpi();
++ acpi_ht = 1;
++ }
++ else if (fullarg(from, "pci=noacpi"))
++ acpi_disable_pci();
++ else if (fullarg(from, "acpi=noirq"))
++ acpi_noirq_set();
++
++ else if (fullarg(from, "acpi_sci=edge"))
++ acpi_sci_flags.trigger = 1;
++ else if (fullarg(from, "acpi_sci=level"))
++ acpi_sci_flags.trigger = 3;
++ else if (fullarg(from, "acpi_sci=high"))
++ acpi_sci_flags.polarity = 1;
++ else if (fullarg(from, "acpi_sci=low"))
++ acpi_sci_flags.polarity = 3;
++
++ /* acpi=strict disables out-of-spec workarounds */
++ else if (fullarg(from, "acpi=strict")) {
++ acpi_strict = 1;
++ }
++#ifdef CONFIG_X86_IO_APIC
++ else if (fullarg(from, "acpi_skip_timer_override"))
++ acpi_skip_timer_override = 1;
++#endif
++#endif
++
++#ifndef CONFIG_XEN
++ if (fullarg(from, "disable_timer_pin_1"))
++ disable_timer_pin_1 = 1;
++ if (fullarg(from, "enable_timer_pin_1"))
++ disable_timer_pin_1 = -1;
++
++ if (fullarg(from, "nolapic") || fullarg(from, "disableapic")) {
++ clear_bit(X86_FEATURE_APIC, boot_cpu_data.x86_capability);
++ disable_apic = 1;
++ }
++
++ if (fullarg(from, "noapic"))
++ skip_ioapic_setup = 1;
++
++ if (fullarg(from,"apic")) {
++ skip_ioapic_setup = 0;
++ ioapic_force = 1;
++ }
++#endif
++
++ if (!memcmp(from, "mem=", 4))
++ parse_memopt(from+4, &from);
++
++ if (!memcmp(from, "memmap=", 7)) {
++ /* exactmap option is for used defined memory */
++ if (!memcmp(from+7, "exactmap", 8)) {
++#ifdef CONFIG_CRASH_DUMP
++ /* If we are doing a crash dump, we
++ * still need to know the real mem
++ * size before original memory map is
++ * reset.
++ */
++ saved_max_pfn = e820_end_of_ram();
++#endif
++ from += 8+7;
++ end_pfn_map = 0;
++ e820.nr_map = 0;
++ userdef = 1;
++ }
++ else {
++ parse_memmapopt(from+7, &from);
++ userdef = 1;
++ }
++ }
++
++#ifdef CONFIG_NUMA
++ if (!memcmp(from, "numa=", 5))
++ numa_setup(from+5);
++#endif
++
++ if (!memcmp(from,"iommu=",6)) {
++ iommu_setup(from+6);
++ }
++
++ if (fullarg(from,"oops=panic"))
++ panic_on_oops = 1;
++
++ if (!memcmp(from, "noexec=", 7))
++ nonx_setup(from + 7);
++
++#ifdef CONFIG_KEXEC
++ /* crashkernel=size@addr specifies the location to reserve for
++ * a crash kernel. By reserving this memory we guarantee
++ * that linux never set's it up as a DMA target.
++ * Useful for holding code to do something appropriate
++ * after a kernel panic.
++ */
++ else if (!memcmp(from, "crashkernel=", 12)) {
++ unsigned long size, base;
++ size = memparse(from+12, &from);
++ if (*from == '@') {
++ base = memparse(from+1, &from);
++ /* FIXME: Do I want a sanity check
++ * to validate the memory range?
++ */
++ crashk_res.start = base;
++ crashk_res.end = base + size - 1;
++ }
++ }
++#endif
++
++#ifdef CONFIG_PROC_VMCORE
++ /* elfcorehdr= specifies the location of elf core header
++ * stored by the crashed kernel. This option will be passed
++ * by kexec loader to the capture kernel.
++ */
++ else if(!memcmp(from, "elfcorehdr=", 11))
++ elfcorehdr_addr = memparse(from+11, &from);
++#endif
++
++#if defined(CONFIG_HOTPLUG_CPU) && !defined(CONFIG_XEN)
++ else if (!memcmp(from, "additional_cpus=", 16))
++ setup_additional_cpus(from+16);
++#endif
++
++ next_char:
++ c = *(from++);
++ if (!c)
++ break;
++ if (COMMAND_LINE_SIZE <= ++len)
++ break;
++ *(to++) = c;
++ }
++ if (userdef) {
++ printk(KERN_INFO "user-defined physical RAM map:\n");
++ e820_print_map("user");
++ }
++ *to = '\0';
++ *cmdline_p = command_line;
++}
++
++#ifndef CONFIG_NUMA
++static void __init
++contig_initmem_init(unsigned long start_pfn, unsigned long end_pfn)
++{
++ unsigned long bootmap_size, bootmap;
++
++ bootmap_size = bootmem_bootmap_pages(end_pfn)<<PAGE_SHIFT;
++ bootmap = find_e820_area(0, end_pfn<<PAGE_SHIFT, bootmap_size);
++ if (bootmap == -1L)
++ panic("Cannot find bootmem map of size %ld\n",bootmap_size);
++ bootmap_size = init_bootmem(bootmap >> PAGE_SHIFT, end_pfn);
++#ifdef CONFIG_XEN
++ e820_bootmem_free(NODE_DATA(0), 0, xen_start_info->nr_pages<<PAGE_SHIFT);
++#else
++ e820_bootmem_free(NODE_DATA(0), 0, end_pfn << PAGE_SHIFT);
++#endif
++ reserve_bootmem(bootmap, bootmap_size);
++}
++#endif
++
++/* Use inline assembly to define this because the nops are defined
++ as inline assembly strings in the include files and we cannot
++ get them easily into strings. */
++asm("\t.data\nk8nops: "
++ K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
++ K8_NOP7 K8_NOP8);
++
++extern unsigned char k8nops[];
++static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
++ NULL,
++ k8nops,
++ k8nops + 1,
++ k8nops + 1 + 2,
++ k8nops + 1 + 2 + 3,
++ k8nops + 1 + 2 + 3 + 4,
++ k8nops + 1 + 2 + 3 + 4 + 5,
++ k8nops + 1 + 2 + 3 + 4 + 5 + 6,
++ k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
++};
++
++extern char __vsyscall_0;
++
++/* Replace instructions with better alternatives for this CPU type.
++
++ This runs before SMP is initialized to avoid SMP problems with
++ self modifying code. This implies that assymetric systems where
++ APs have less capabilities than the boot processor are not handled.
++ In this case boot with "noreplacement". */
++void apply_alternatives(void *start, void *end)
++{
++ struct alt_instr *a;
++ int diff, i, k;
++ for (a = start; (void *)a < end; a++) {
++ u8 *instr;
++
++ if (!boot_cpu_has(a->cpuid))
++ continue;
++
++ BUG_ON(a->replacementlen > a->instrlen);
++ instr = a->instr;
++ /* vsyscall code is not mapped yet. resolve it manually. */
++ if (instr >= (u8 *)VSYSCALL_START && instr < (u8*)VSYSCALL_END)
++ instr -= VSYSCALL_START - (unsigned long)&__vsyscall_0;
++ __inline_memcpy(instr, a->replacement, a->replacementlen);
++ diff = a->instrlen - a->replacementlen;
++
++ /* Pad the rest with nops */
++ for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
++ k = diff;
++ if (k > ASM_NOP_MAX)
++ k = ASM_NOP_MAX;
++ __inline_memcpy(instr + i, k8_nops[k], k);
++ }
++ }
++}
++
++static int no_replacement __initdata = 0;
++
++void __init alternative_instructions(void)
++{
++ extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
++ if (no_replacement)
++ return;
++ apply_alternatives(__alt_instructions, __alt_instructions_end);
++}
++
++static int __init noreplacement_setup(char *s)
++{
++ no_replacement = 1;
++ return 1;
++}
++
++__setup("noreplacement", noreplacement_setup);
++
++#if defined(CONFIG_EDD) || defined(CONFIG_EDD_MODULE)
++struct edd edd;
++#ifdef CONFIG_EDD_MODULE
++EXPORT_SYMBOL(edd);
++#endif
++/**
++ * copy_edd() - Copy the BIOS EDD information
++ * from boot_params into a safe place.
++ *
++ */
++static inline void copy_edd(void)
++{
++ memcpy(edd.mbr_signature, EDD_MBR_SIGNATURE, sizeof(edd.mbr_signature));
++ memcpy(edd.edd_info, EDD_BUF, sizeof(edd.edd_info));
++ edd.mbr_signature_nr = EDD_MBR_SIG_NR;
++ edd.edd_info_nr = EDD_NR;
++}
++#else
++static inline void copy_edd(void)
++{
++}
++#endif
++
++#ifndef CONFIG_XEN
++#define EBDA_ADDR_POINTER 0x40E
++
++unsigned __initdata ebda_addr;
++unsigned __initdata ebda_size;
++
++static void discover_ebda(void)
++{
++ /*
++ * there is a real-mode segmented pointer pointing to the
++ * 4K EBDA area at 0x40E
++ */
++ ebda_addr = *(unsigned short *)EBDA_ADDR_POINTER;
++ ebda_addr <<= 4;
++
++ ebda_size = *(unsigned short *)(unsigned long)ebda_addr;
++
++ /* Round EBDA up to pages */
++ if (ebda_size == 0)
++ ebda_size = 1;
++ ebda_size <<= 10;
++ ebda_size = round_up(ebda_size + (ebda_addr & ~PAGE_MASK), PAGE_SIZE);
++ if (ebda_size > 64*1024)
++ ebda_size = 64*1024;
++}
++#endif
++
++void __init setup_arch(char **cmdline_p)
++{
++ unsigned long kernel_end;
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
++ struct e820entry *machine_e820;
++ struct xen_memory_map memmap;
++#endif
++
++#ifdef CONFIG_XEN
++ /* Register a call for panic conditions. */
++ atomic_notifier_chain_register(&panic_notifier_list, &xen_panic_block);
++
++ ROOT_DEV = MKDEV(RAMDISK_MAJOR,0);
++ kernel_end = 0; /* dummy */
++ screen_info = SCREEN_INFO;
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ /* This is drawn from a dump from vgacon:startup in
++ * standard Linux. */
++ screen_info.orig_video_mode = 3;
++ screen_info.orig_video_isVGA = 1;
++ screen_info.orig_video_lines = 25;
++ screen_info.orig_video_cols = 80;
++ screen_info.orig_video_ega_bx = 3;
++ screen_info.orig_video_points = 16;
++ } else
++ screen_info.orig_video_isVGA = 0;
++
++ edid_info = EDID_INFO;
++ saved_video_mode = SAVED_VIDEO_MODE;
++ bootloader_type = LOADER_TYPE;
++
++#ifdef CONFIG_BLK_DEV_RAM
++ rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
++ rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
++ rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
++
++
++#endif
++
++ setup_xen_features();
++
++ HYPERVISOR_vm_assist(VMASST_CMD_enable,
++ VMASST_TYPE_writable_pagetables);
++
++ ARCH_SETUP
++#else
++ ROOT_DEV = old_decode_dev(ORIG_ROOT_DEV);
++ screen_info = SCREEN_INFO;
++ edid_info = EDID_INFO;
++ saved_video_mode = SAVED_VIDEO_MODE;
++ bootloader_type = LOADER_TYPE;
++
++#ifdef CONFIG_BLK_DEV_RAM
++ rd_image_start = RAMDISK_FLAGS & RAMDISK_IMAGE_START_MASK;
++ rd_prompt = ((RAMDISK_FLAGS & RAMDISK_PROMPT_FLAG) != 0);
++ rd_doload = ((RAMDISK_FLAGS & RAMDISK_LOAD_FLAG) != 0);
++#endif
++#endif /* !CONFIG_XEN */
++ setup_memory_region();
++ copy_edd();
++
++ if (!MOUNT_ROOT_RDONLY)
++ root_mountflags &= ~MS_RDONLY;
++ init_mm.start_code = (unsigned long) &_text;
++ init_mm.end_code = (unsigned long) &_etext;
++ init_mm.end_data = (unsigned long) &_edata;
++ init_mm.brk = (unsigned long) &_end;
++
++ code_resource.start = virt_to_phys(&_text);
++ code_resource.end = virt_to_phys(&_etext)-1;
++ data_resource.start = virt_to_phys(&_etext);
++ data_resource.end = virt_to_phys(&_edata)-1;
++
++ parse_cmdline_early(cmdline_p);
++
++ early_identify_cpu(&boot_cpu_data);
++
++ /*
++ * partially used pages are not usable - thus
++ * we are rounding upwards:
++ */
++ end_pfn = e820_end_of_ram();
++ num_physpages = end_pfn; /* for pfn_valid */
++
++ check_efer();
++
++#ifndef CONFIG_XEN
++ discover_ebda();
++#endif
++
++ init_memory_mapping(0, (end_pfn_map << PAGE_SHIFT));
++
++#ifdef CONFIG_ACPI_NUMA
++ /*
++ * Parse SRAT to discover nodes.
++ */
++ acpi_numa_init();
++#endif
++
++#ifdef CONFIG_NUMA
++ numa_initmem_init(0, end_pfn);
++#else
++ contig_initmem_init(0, end_pfn);
++#endif
++
++ /* Reserve direct mapping */
++ reserve_bootmem_generic(table_start << PAGE_SHIFT,
++ (table_end - table_start) << PAGE_SHIFT);
++
++ /* reserve kernel */
++ kernel_end = round_up(__pa_symbol(&_end),PAGE_SIZE);
++ reserve_bootmem_generic(HIGH_MEMORY, kernel_end - HIGH_MEMORY);
++
++#ifdef CONFIG_XEN
++ /* reserve physmap, start info and initial page tables */
++ reserve_bootmem(kernel_end, (table_start<<PAGE_SHIFT)-kernel_end);
++#else
++ /*
++ * reserve physical page 0 - it's a special BIOS page on many boxes,
++ * enabling clean reboots, SMP operation, laptop functions.
++ */
++ reserve_bootmem_generic(0, PAGE_SIZE);
++
++ /* reserve ebda region */
++ if (ebda_addr)
++ reserve_bootmem_generic(ebda_addr, ebda_size);
++#endif
++
++#ifdef CONFIG_SMP
++ /*
++ * But first pinch a few for the stack/trampoline stuff
++ * FIXME: Don't need the extra page at 4K, but need to fix
++ * trampoline before removing it. (see the GDT stuff)
++ */
++ reserve_bootmem_generic(PAGE_SIZE, PAGE_SIZE);
++
++ /* Reserve SMP trampoline */
++ reserve_bootmem_generic(SMP_TRAMPOLINE_BASE, PAGE_SIZE);
++#endif
++
++#ifdef CONFIG_ACPI_SLEEP
++ /*
++ * Reserve low memory region for sleep support.
++ */
++ acpi_reserve_bootmem();
++#endif
++#ifdef CONFIG_XEN
++#ifdef CONFIG_BLK_DEV_INITRD
++ if (xen_start_info->mod_start) {
++ if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
++ /*reserve_bootmem_generic(INITRD_START, INITRD_SIZE);*/
++ initrd_start = INITRD_START + PAGE_OFFSET;
++ initrd_end = initrd_start+INITRD_SIZE;
++ initrd_below_start_ok = 1;
++ } else {
++ printk(KERN_ERR "initrd extends beyond end of memory "
++ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
++ (unsigned long)(INITRD_START + INITRD_SIZE),
++ (unsigned long)(end_pfn << PAGE_SHIFT));
++ initrd_start = 0;
++ }
++ }
++#endif
++#else /* CONFIG_XEN */
++#ifdef CONFIG_BLK_DEV_INITRD
++ if (LOADER_TYPE && INITRD_START) {
++ if (INITRD_START + INITRD_SIZE <= (end_pfn << PAGE_SHIFT)) {
++ reserve_bootmem_generic(INITRD_START, INITRD_SIZE);
++ initrd_start =
++ INITRD_START ? INITRD_START + PAGE_OFFSET : 0;
++ initrd_end = initrd_start+INITRD_SIZE;
++ }
++ else {
++ printk(KERN_ERR "initrd extends beyond end of memory "
++ "(0x%08lx > 0x%08lx)\ndisabling initrd\n",
++ (unsigned long)(INITRD_START + INITRD_SIZE),
++ (unsigned long)(end_pfn << PAGE_SHIFT));
++ initrd_start = 0;
++ }
++ }
++#endif
++#endif /* !CONFIG_XEN */
++#ifdef CONFIG_KEXEC
++ if (crashk_res.start != crashk_res.end) {
++ reserve_bootmem(crashk_res.start,
++ crashk_res.end - crashk_res.start + 1);
++ }
++#endif
++
++ paging_init();
++#ifdef CONFIG_X86_LOCAL_APIC
++ /*
++ * Find and reserve possible boot-time SMP configuration:
++ */
++ find_smp_config();
++#endif
++#ifdef CONFIG_XEN
++ {
++ int i, j, k, fpp;
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* Make sure we have a large enough P->M table. */
++ phys_to_machine_mapping = alloc_bootmem(
++ end_pfn * sizeof(unsigned long));
++ memset(phys_to_machine_mapping, ~0,
++ end_pfn * sizeof(unsigned long));
++ memcpy(phys_to_machine_mapping,
++ (unsigned long *)xen_start_info->mfn_list,
++ xen_start_info->nr_pages * sizeof(unsigned long));
++ free_bootmem(
++ __pa(xen_start_info->mfn_list),
++ PFN_PHYS(PFN_UP(xen_start_info->nr_pages *
++ sizeof(unsigned long))));
++
++ /*
++ * Initialise the list of the frames that specify the
++ * list of frames that make up the p2m table. Used by
++ * save/restore.
++ */
++ pfn_to_mfn_frame_list_list = alloc_bootmem(PAGE_SIZE);
++ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
++ virt_to_mfn(pfn_to_mfn_frame_list_list);
++
++ fpp = PAGE_SIZE/sizeof(unsigned long);
++ for (i=0, j=0, k=-1; i< end_pfn; i+=fpp, j++) {
++ if ((j % fpp) == 0) {
++ k++;
++ BUG_ON(k>=fpp);
++ pfn_to_mfn_frame_list[k] =
++ alloc_bootmem(PAGE_SIZE);
++ pfn_to_mfn_frame_list_list[k] =
++ virt_to_mfn(pfn_to_mfn_frame_list[k]);
++ j=0;
++ }
++ pfn_to_mfn_frame_list[k][j] =
++ virt_to_mfn(&phys_to_machine_mapping[i]);
++ }
++ HYPERVISOR_shared_info->arch.max_pfn = end_pfn;
++ }
++
++ }
++
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ dmi_scan_machine();
++
++ if ( ! (xen_start_info->flags & SIF_INITDOMAIN))
++ {
++ acpi_disabled = 1;
++#ifdef CONFIG_ACPI
++ acpi_ht = 0;
++#endif
++ }
++#endif
++
++#ifndef CONFIG_XEN
++ check_ioapic();
++#endif
++
++ zap_low_mappings(0);
++
++ /*
++ * set this early, so we dont allocate cpu0
++ * if MADT list doesnt list BSP first
++ * mpparse.c/MP_processor_info() allocates logical cpu numbers.
++ */
++ cpu_set(0, cpu_present_map);
++#ifdef CONFIG_ACPI
++ /*
++ * Initialize the ACPI boot-time table parser (gets the RSDP and SDT).
++ * Call this early for SRAT node setup.
++ */
++ acpi_boot_table_init();
++
++ /*
++ * Read APIC and some other early information from ACPI tables.
++ */
++ acpi_boot_init();
++#endif
++
++ init_cpu_to_node();
++
++#ifdef CONFIG_X86_LOCAL_APIC
++ /*
++ * get boot-time SMP configuration:
++ */
++ if (smp_found_config)
++ get_smp_config();
++#ifndef CONFIG_XEN
++ init_apic_mappings();
++#endif
++#endif
++#if defined(CONFIG_XEN) && defined(CONFIG_SMP) && !defined(CONFIG_HOTPLUG_CPU)
++ prefill_possible_map();
++#endif
++
++ /*
++ * Request address space for all standard RAM and ROM resources
++ * and also for regions reported as reserved by the e820.
++ */
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
++ probe_roms();
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ machine_e820 = alloc_bootmem_low_pages(PAGE_SIZE);
++
++ memmap.nr_entries = E820MAX;
++ set_xen_guest_handle(memmap.buffer, machine_e820);
++
++ BUG_ON(HYPERVISOR_memory_op(XENMEM_machine_memory_map, &memmap));
++
++ e820_reserve_resources(machine_e820, memmap.nr_entries);
++ } else if (!(xen_start_info->flags & SIF_INITDOMAIN))
++ e820_reserve_resources(e820.map, e820.nr_map);
++#elif defined(CONFIG_XEN)
++ e820_reserve_resources(e820.map, e820.nr_map);
++#else
++ probe_roms();
++ e820_reserve_resources(e820.map, e820.nr_map);
++#endif
++
++ request_resource(&iomem_resource, &video_ram_resource);
++
++ {
++ unsigned i;
++ /* request I/O space for devices used on all i[345]86 PCs */
++ for (i = 0; i < STANDARD_IO_RESOURCES; i++)
++ request_resource(&ioport_resource, &standard_io_resources[i]);
++ }
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ e820_setup_gap(machine_e820, memmap.nr_entries);
++ free_bootmem(__pa(machine_e820), PAGE_SIZE);
++ }
++#elif !defined(CONFIG_XEN)
++ e820_setup_gap(e820.map, e820.nr_map);
++#endif
++
++#ifdef CONFIG_GART_IOMMU
++ iommu_hole_init();
++#endif
++
++#ifdef CONFIG_XEN
++ {
++ struct physdev_set_iopl set_iopl;
++
++ set_iopl.iopl = 1;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ if (!(xen_start_info->flags & SIF_PRIVILEGED))
++ panic("Xen granted us console access "
++ "but not privileged status");
++
++#ifdef CONFIG_VT
++#if defined(CONFIG_VGA_CONSOLE)
++ conswitchp = &vga_con;
++#elif defined(CONFIG_DUMMY_CONSOLE)
++ conswitchp = &dummy_con;
++#endif
++#endif
++ } else {
++ extern int console_use_vt;
++ console_use_vt = 0;
++ }
++ }
++#else /* CONFIG_XEN */
++
++#ifdef CONFIG_VT
++#if defined(CONFIG_VGA_CONSOLE)
++ conswitchp = &vga_con;
++#elif defined(CONFIG_DUMMY_CONSOLE)
++ conswitchp = &dummy_con;
++#endif
++#endif
++
++#endif /* !CONFIG_XEN */
++}
++
++#ifdef CONFIG_XEN
++static int
++xen_panic_event(struct notifier_block *this, unsigned long event, void *ptr)
++{
++ HYPERVISOR_shutdown(SHUTDOWN_crash);
++ /* we're never actually going to get here... */
++ return NOTIFY_DONE;
++}
++#endif /* !CONFIG_XEN */
++
++
++static int __cpuinit get_model_name(struct cpuinfo_x86 *c)
++{
++ unsigned int *v;
++
++ if (c->extended_cpuid_level < 0x80000004)
++ return 0;
++
++ v = (unsigned int *) c->x86_model_id;
++ cpuid(0x80000002, &v[0], &v[1], &v[2], &v[3]);
++ cpuid(0x80000003, &v[4], &v[5], &v[6], &v[7]);
++ cpuid(0x80000004, &v[8], &v[9], &v[10], &v[11]);
++ c->x86_model_id[48] = 0;
++ return 1;
++}
++
++
++static void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
++{
++ unsigned int n, dummy, eax, ebx, ecx, edx;
++
++ n = c->extended_cpuid_level;
++
++ if (n >= 0x80000005) {
++ cpuid(0x80000005, &dummy, &ebx, &ecx, &edx);
++ printk(KERN_INFO "CPU: L1 I Cache: %dK (%d bytes/line), D cache %dK (%d bytes/line)\n",
++ edx>>24, edx&0xFF, ecx>>24, ecx&0xFF);
++ c->x86_cache_size=(ecx>>24)+(edx>>24);
++ /* On K8 L1 TLB is inclusive, so don't count it */
++ c->x86_tlbsize = 0;
++ }
++
++ if (n >= 0x80000006) {
++ cpuid(0x80000006, &dummy, &ebx, &ecx, &edx);
++ ecx = cpuid_ecx(0x80000006);
++ c->x86_cache_size = ecx >> 16;
++ c->x86_tlbsize += ((ebx >> 16) & 0xfff) + (ebx & 0xfff);
++
++ printk(KERN_INFO "CPU: L2 Cache: %dK (%d bytes/line)\n",
++ c->x86_cache_size, ecx & 0xFF);
++ }
++
++ if (n >= 0x80000007)
++ cpuid(0x80000007, &dummy, &dummy, &dummy, &c->x86_power);
++ if (n >= 0x80000008) {
++ cpuid(0x80000008, &eax, &dummy, &dummy, &dummy);
++ c->x86_virt_bits = (eax >> 8) & 0xff;
++ c->x86_phys_bits = eax & 0xff;
++ }
++}
++
++#ifdef CONFIG_NUMA
++static int nearby_node(int apicid)
++{
++ int i;
++ for (i = apicid - 1; i >= 0; i--) {
++ int node = apicid_to_node[i];
++ if (node != NUMA_NO_NODE && node_online(node))
++ return node;
++ }
++ for (i = apicid + 1; i < MAX_LOCAL_APIC; i++) {
++ int node = apicid_to_node[i];
++ if (node != NUMA_NO_NODE && node_online(node))
++ return node;
++ }
++ return first_node(node_online_map); /* Shouldn't happen */
++}
++#endif
++
++/*
++ * On a AMD dual core setup the lower bits of the APIC id distingush the cores.
++ * Assumes number of cores is a power of two.
++ */
++static void __init amd_detect_cmp(struct cpuinfo_x86 *c)
++{
++#ifdef CONFIG_SMP
++ int cpu = smp_processor_id();
++ unsigned bits;
++#ifdef CONFIG_NUMA
++ int node = 0;
++ unsigned apicid = hard_smp_processor_id();
++#endif
++
++ bits = 0;
++ while ((1 << bits) < c->x86_max_cores)
++ bits++;
++
++ /* Low order bits define the core id (index of core in socket) */
++ cpu_core_id[cpu] = phys_proc_id[cpu] & ((1 << bits)-1);
++ /* Convert the APIC ID into the socket ID */
++ phys_proc_id[cpu] = phys_pkg_id(bits);
++
++#ifdef CONFIG_NUMA
++ node = phys_proc_id[cpu];
++ if (apicid_to_node[apicid] != NUMA_NO_NODE)
++ node = apicid_to_node[apicid];
++ if (!node_online(node)) {
++ /* Two possibilities here:
++ - The CPU is missing memory and no node was created.
++ In that case try picking one from a nearby CPU
++ - The APIC IDs differ from the HyperTransport node IDs
++ which the K8 northbridge parsing fills in.
++ Assume they are all increased by a constant offset,
++ but in the same order as the HT nodeids.
++ If that doesn't result in a usable node fall back to the
++ path for the previous case. */
++ int ht_nodeid = apicid - (phys_proc_id[0] << bits);
++ if (ht_nodeid >= 0 &&
++ apicid_to_node[ht_nodeid] != NUMA_NO_NODE)
++ node = apicid_to_node[ht_nodeid];
++ /* Pick a nearby node */
++ if (!node_online(node))
++ node = nearby_node(apicid);
++ }
++ numa_set_node(cpu, node);
++
++ printk(KERN_INFO "CPU %d/%x(%d) -> Node %d -> Core %d\n",
++ cpu, apicid, c->x86_max_cores, node, cpu_core_id[cpu]);
++#endif
++#endif
++}
++
++static int __init init_amd(struct cpuinfo_x86 *c)
++{
++ int r;
++ unsigned level;
++
++#ifdef CONFIG_SMP
++ unsigned long value;
++
++ /*
++ * Disable TLB flush filter by setting HWCR.FFDIS on K8
++ * bit 6 of msr C001_0015
++ *
++ * Errata 63 for SH-B3 steppings
++ * Errata 122 for all steppings (F+ have it disabled by default)
++ */
++ if (c->x86 == 15) {
++ rdmsrl(MSR_K8_HWCR, value);
++ value |= 1 << 6;
++ wrmsrl(MSR_K8_HWCR, value);
++ }
++#endif
++
++ /* Bit 31 in normal CPUID used for nonstandard 3DNow ID;
++ 3DNow is IDd by bit 31 in extended CPUID (1*32+31) anyway */
++ clear_bit(0*32+31, &c->x86_capability);
++
++ /* On C+ stepping K8 rep microcode works well for copy/memset */
++ level = cpuid_eax(1);
++ if (c->x86 == 15 && ((level >= 0x0f48 && level < 0x0f50) || level >= 0x0f58))
++ set_bit(X86_FEATURE_REP_GOOD, &c->x86_capability);
++
++ /* Enable workaround for FXSAVE leak */
++ if (c->x86 >= 6)
++ set_bit(X86_FEATURE_FXSAVE_LEAK, &c->x86_capability);
++
++ r = get_model_name(c);
++ if (!r) {
++ switch (c->x86) {
++ case 15:
++ /* Should distinguish Models here, but this is only
++ a fallback anyways. */
++ strcpy(c->x86_model_id, "Hammer");
++ break;
++ }
++ }
++ display_cacheinfo(c);
++
++ /* c->x86_power is 8000_0007 edx. Bit 8 is constant TSC */
++ if (c->x86_power & (1<<8))
++ set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
++
++ if (c->extended_cpuid_level >= 0x80000008) {
++ c->x86_max_cores = (cpuid_ecx(0x80000008) & 0xff) + 1;
++
++ amd_detect_cmp(c);
++ }
++
++ return r;
++}
++
++static void __cpuinit detect_ht(struct cpuinfo_x86 *c)
++{
++#ifdef CONFIG_SMP
++ u32 eax, ebx, ecx, edx;
++ int index_msb, core_bits;
++ int cpu = smp_processor_id();
++
++ cpuid(1, &eax, &ebx, &ecx, &edx);
++
++
++ if (!cpu_has(c, X86_FEATURE_HT) || cpu_has(c, X86_FEATURE_CMP_LEGACY))
++ return;
++
++ smp_num_siblings = (ebx & 0xff0000) >> 16;
++
++ if (smp_num_siblings == 1) {
++ printk(KERN_INFO "CPU: Hyper-Threading is disabled\n");
++ } else if (smp_num_siblings > 1 ) {
++
++ if (smp_num_siblings > NR_CPUS) {
++ printk(KERN_WARNING "CPU: Unsupported number of the siblings %d", smp_num_siblings);
++ smp_num_siblings = 1;
++ return;
++ }
++
++ index_msb = get_count_order(smp_num_siblings);
++ phys_proc_id[cpu] = phys_pkg_id(index_msb);
++
++ printk(KERN_INFO "CPU: Physical Processor ID: %d\n",
++ phys_proc_id[cpu]);
++
++ smp_num_siblings = smp_num_siblings / c->x86_max_cores;
++
++ index_msb = get_count_order(smp_num_siblings) ;
++
++ core_bits = get_count_order(c->x86_max_cores);
++
++ cpu_core_id[cpu] = phys_pkg_id(index_msb) &
++ ((1 << core_bits) - 1);
++
++ if (c->x86_max_cores > 1)
++ printk(KERN_INFO "CPU: Processor Core ID: %d\n",
++ cpu_core_id[cpu]);
++ }
++#endif
++}
++
++/*
++ * find out the number of processor cores on the die
++ */
++static int __cpuinit intel_num_cpu_cores(struct cpuinfo_x86 *c)
++{
++ unsigned int eax;
++
++ if (c->cpuid_level < 4)
++ return 1;
++
++ __asm__("cpuid"
++ : "=a" (eax)
++ : "0" (4), "c" (0)
++ : "bx", "dx");
++
++ if (eax & 0x1f)
++ return ((eax >> 26) + 1);
++ else
++ return 1;
++}
++
++static void srat_detect_node(void)
++{
++#ifdef CONFIG_NUMA
++ unsigned node;
++ int cpu = smp_processor_id();
++
++ /* Don't do the funky fallback heuristics the AMD version employs
++ for now. */
++ node = apicid_to_node[hard_smp_processor_id()];
++ if (node == NUMA_NO_NODE)
++ node = 0;
++ numa_set_node(cpu, node);
++
++ if (acpi_numa > 0)
++ printk(KERN_INFO "CPU %d -> Node %d\n", cpu, node);
++#endif
++}
++
++static void __cpuinit init_intel(struct cpuinfo_x86 *c)
++{
++ /* Cache sizes */
++ unsigned n;
++
++ init_intel_cacheinfo(c);
++ n = c->extended_cpuid_level;
++ if (n >= 0x80000008) {
++ unsigned eax = cpuid_eax(0x80000008);
++ c->x86_virt_bits = (eax >> 8) & 0xff;
++ c->x86_phys_bits = eax & 0xff;
++ /* CPUID workaround for Intel 0F34 CPU */
++ if (c->x86_vendor == X86_VENDOR_INTEL &&
++ c->x86 == 0xF && c->x86_model == 0x3 &&
++ c->x86_mask == 0x4)
++ c->x86_phys_bits = 36;
++ }
++
++ if (c->x86 == 15)
++ c->x86_cache_alignment = c->x86_clflush_size * 2;
++ if ((c->x86 == 0xf && c->x86_model >= 0x03) ||
++ (c->x86 == 0x6 && c->x86_model >= 0x0e))
++ set_bit(X86_FEATURE_CONSTANT_TSC, &c->x86_capability);
++ set_bit(X86_FEATURE_SYNC_RDTSC, &c->x86_capability);
++ c->x86_max_cores = intel_num_cpu_cores(c);
++
++ srat_detect_node();
++}
++
++static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c)
++{
++ char *v = c->x86_vendor_id;
++
++ if (!strcmp(v, "AuthenticAMD"))
++ c->x86_vendor = X86_VENDOR_AMD;
++ else if (!strcmp(v, "GenuineIntel"))
++ c->x86_vendor = X86_VENDOR_INTEL;
++ else
++ c->x86_vendor = X86_VENDOR_UNKNOWN;
++}
++
++struct cpu_model_info {
++ int vendor;
++ int family;
++ char *model_names[16];
++};
++
++/* Do some early cpuid on the boot CPU to get some parameter that are
++ needed before check_bugs. Everything advanced is in identify_cpu
++ below. */
++void __cpuinit early_identify_cpu(struct cpuinfo_x86 *c)
++{
++ u32 tfms;
++
++ c->loops_per_jiffy = loops_per_jiffy;
++ c->x86_cache_size = -1;
++ c->x86_vendor = X86_VENDOR_UNKNOWN;
++ c->x86_model = c->x86_mask = 0; /* So far unknown... */
++ c->x86_vendor_id[0] = '\0'; /* Unset */
++ c->x86_model_id[0] = '\0'; /* Unset */
++ c->x86_clflush_size = 64;
++ c->x86_cache_alignment = c->x86_clflush_size;
++ c->x86_max_cores = 1;
++ c->extended_cpuid_level = 0;
++ memset(&c->x86_capability, 0, sizeof c->x86_capability);
++
++ /* Get vendor name */
++ cpuid(0x00000000, (unsigned int *)&c->cpuid_level,
++ (unsigned int *)&c->x86_vendor_id[0],
++ (unsigned int *)&c->x86_vendor_id[8],
++ (unsigned int *)&c->x86_vendor_id[4]);
++
++ get_cpu_vendor(c);
++
++ /* Initialize the standard set of capabilities */
++ /* Note that the vendor-specific code below might override */
++
++ /* Intel-defined flags: level 0x00000001 */
++ if (c->cpuid_level >= 0x00000001) {
++ __u32 misc;
++ cpuid(0x00000001, &tfms, &misc, &c->x86_capability[4],
++ &c->x86_capability[0]);
++ c->x86 = (tfms >> 8) & 0xf;
++ c->x86_model = (tfms >> 4) & 0xf;
++ c->x86_mask = tfms & 0xf;
++ if (c->x86 == 0xf)
++ c->x86 += (tfms >> 20) & 0xff;
++ if (c->x86 >= 0x6)
++ c->x86_model += ((tfms >> 16) & 0xF) << 4;
++ if (c->x86_capability[0] & (1<<19))
++ c->x86_clflush_size = ((misc >> 8) & 0xff) * 8;
++ } else {
++ /* Have CPUID level 0 only - unheard of */
++ c->x86 = 4;
++ }
++
++#ifdef CONFIG_SMP
++ phys_proc_id[smp_processor_id()] = (cpuid_ebx(1) >> 24) & 0xff;
++#endif
++}
++
++/*
++ * This does the hard work of actually picking apart the CPU stuff...
++ */
++void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
++{
++ int i;
++ u32 xlvl;
++
++ early_identify_cpu(c);
++
++ /* AMD-defined flags: level 0x80000001 */
++ xlvl = cpuid_eax(0x80000000);
++ c->extended_cpuid_level = xlvl;
++ if ((xlvl & 0xffff0000) == 0x80000000) {
++ if (xlvl >= 0x80000001) {
++ c->x86_capability[1] = cpuid_edx(0x80000001);
++ c->x86_capability[6] = cpuid_ecx(0x80000001);
++ }
++ if (xlvl >= 0x80000004)
++ get_model_name(c); /* Default name */
++ }
++
++ /* Transmeta-defined flags: level 0x80860001 */
++ xlvl = cpuid_eax(0x80860000);
++ if ((xlvl & 0xffff0000) == 0x80860000) {
++ /* Don't set x86_cpuid_level here for now to not confuse. */
++ if (xlvl >= 0x80860001)
++ c->x86_capability[2] = cpuid_edx(0x80860001);
++ }
++
++ c->apicid = phys_pkg_id(0);
++
++ /*
++ * Vendor-specific initialization. In this section we
++ * canonicalize the feature flags, meaning if there are
++ * features a certain CPU supports which CPUID doesn't
++ * tell us, CPUID claiming incorrect flags, or other bugs,
++ * we handle them here.
++ *
++ * At the end of this section, c->x86_capability better
++ * indicate the features this CPU genuinely supports!
++ */
++ switch (c->x86_vendor) {
++ case X86_VENDOR_AMD:
++ init_amd(c);
++ break;
++
++ case X86_VENDOR_INTEL:
++ init_intel(c);
++ break;
++
++ case X86_VENDOR_UNKNOWN:
++ default:
++ display_cacheinfo(c);
++ break;
++ }
++
++ select_idle_routine(c);
++ detect_ht(c);
++
++ /*
++ * On SMP, boot_cpu_data holds the common feature set between
++ * all CPUs; so make sure that we indicate which features are
++ * common between the CPUs. The first time this routine gets
++ * executed, c == &boot_cpu_data.
++ */
++ if (c != &boot_cpu_data) {
++ /* AND the already accumulated flags with these */
++ for (i = 0 ; i < NCAPINTS ; i++)
++ boot_cpu_data.x86_capability[i] &= c->x86_capability[i];
++ }
++
++#ifdef CONFIG_X86_MCE
++ mcheck_init(c);
++#endif
++ if (c == &boot_cpu_data)
++ mtrr_bp_init();
++ else
++ mtrr_ap_init();
++#ifdef CONFIG_NUMA
++ numa_add_cpu(smp_processor_id());
++#endif
++}
++
++
++void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
++{
++ if (c->x86_model_id[0])
++ printk("%s", c->x86_model_id);
++
++ if (c->x86_mask || c->cpuid_level >= 0)
++ printk(" stepping %02x\n", c->x86_mask);
++ else
++ printk("\n");
++}
++
++/*
++ * Get CPU information for use by the procfs.
++ */
++
++static int show_cpuinfo(struct seq_file *m, void *v)
++{
++ struct cpuinfo_x86 *c = v;
++
++ /*
++ * These flag bits must match the definitions in <asm/cpufeature.h>.
++ * NULL means this bit is undefined or reserved; either way it doesn't
++ * have meaning as far as Linux is concerned. Note that it's important
++ * to realize there is a difference between this table and CPUID -- if
++ * applications want to get the raw CPUID data, they should access
++ * /dev/cpu/<cpu_nr>/cpuid instead.
++ */
++ static char *x86_cap_flags[] = {
++ /* Intel-defined */
++ "fpu", "vme", "de", "pse", "tsc", "msr", "pae", "mce",
++ "cx8", "apic", NULL, "sep", "mtrr", "pge", "mca", "cmov",
++ "pat", "pse36", "pn", "clflush", NULL, "dts", "acpi", "mmx",
++ "fxsr", "sse", "sse2", "ss", "ht", "tm", "ia64", NULL,
++
++ /* AMD-defined */
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, "syscall", NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, "nx", NULL, "mmxext", NULL,
++ NULL, "fxsr_opt", "rdtscp", NULL, NULL, "lm", "3dnowext", "3dnow",
++
++ /* Transmeta-defined */
++ "recovery", "longrun", NULL, "lrti", NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++
++ /* Other (Linux-defined) */
++ "cxmmx", NULL, "cyrix_arr", "centaur_mcr", NULL,
++ "constant_tsc", NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++
++ /* Intel-defined (#2) */
++ "pni", NULL, NULL, "monitor", "ds_cpl", "vmx", "smx", "est",
++ "tm2", NULL, "cid", NULL, NULL, "cx16", "xtpr", NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++
++ /* VIA/Cyrix/Centaur-defined */
++ NULL, NULL, "rng", "rng_en", NULL, NULL, "ace", "ace_en",
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++
++ /* AMD-defined (#2) */
++ "lahf_lm", "cmp_legacy", "svm", NULL, "cr8_legacy", NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
++ };
++ static char *x86_power_flags[] = {
++ "ts", /* temperature sensor */
++ "fid", /* frequency id control */
++ "vid", /* voltage id control */
++ "ttp", /* thermal trip */
++ "tm",
++ "stc",
++ NULL,
++ /* nothing */ /* constant_tsc - moved to flags */
++ };
++
++
++#ifdef CONFIG_SMP
++ if (!cpu_online(c-cpu_data))
++ return 0;
++#endif
++
++ seq_printf(m,"processor\t: %u\n"
++ "vendor_id\t: %s\n"
++ "cpu family\t: %d\n"
++ "model\t\t: %d\n"
++ "model name\t: %s\n",
++ (unsigned)(c-cpu_data),
++ c->x86_vendor_id[0] ? c->x86_vendor_id : "unknown",
++ c->x86,
++ (int)c->x86_model,
++ c->x86_model_id[0] ? c->x86_model_id : "unknown");
++
++ if (c->x86_mask || c->cpuid_level >= 0)
++ seq_printf(m, "stepping\t: %d\n", c->x86_mask);
++ else
++ seq_printf(m, "stepping\t: unknown\n");
++
++ if (cpu_has(c,X86_FEATURE_TSC)) {
++ unsigned int freq = cpufreq_quick_get((unsigned)(c-cpu_data));
++ if (!freq)
++ freq = cpu_khz;
++ seq_printf(m, "cpu MHz\t\t: %u.%03u\n",
++ freq / 1000, (freq % 1000));
++ }
++
++ /* Cache size */
++ if (c->x86_cache_size >= 0)
++ seq_printf(m, "cache size\t: %d KB\n", c->x86_cache_size);
++
++#ifdef CONFIG_SMP
++ if (smp_num_siblings * c->x86_max_cores > 1) {
++ int cpu = c - cpu_data;
++ seq_printf(m, "physical id\t: %d\n", phys_proc_id[cpu]);
++ seq_printf(m, "siblings\t: %d\n", cpus_weight(cpu_core_map[cpu]));
++ seq_printf(m, "core id\t\t: %d\n", cpu_core_id[cpu]);
++ seq_printf(m, "cpu cores\t: %d\n", c->booted_cores);
++ }
++#endif
++
++ seq_printf(m,
++ "fpu\t\t: yes\n"
++ "fpu_exception\t: yes\n"
++ "cpuid level\t: %d\n"
++ "wp\t\t: yes\n"
++ "flags\t\t:",
++ c->cpuid_level);
++
++ {
++ int i;
++ for ( i = 0 ; i < 32*NCAPINTS ; i++ )
++ if (cpu_has(c, i) && x86_cap_flags[i] != NULL)
++ seq_printf(m, " %s", x86_cap_flags[i]);
++ }
++
++ seq_printf(m, "\nbogomips\t: %lu.%02lu\n",
++ c->loops_per_jiffy/(500000/HZ),
++ (c->loops_per_jiffy/(5000/HZ)) % 100);
++
++ if (c->x86_tlbsize > 0)
++ seq_printf(m, "TLB size\t: %d 4K pages\n", c->x86_tlbsize);
++ seq_printf(m, "clflush size\t: %d\n", c->x86_clflush_size);
++ seq_printf(m, "cache_alignment\t: %d\n", c->x86_cache_alignment);
++
++ seq_printf(m, "address sizes\t: %u bits physical, %u bits virtual\n",
++ c->x86_phys_bits, c->x86_virt_bits);
++
++ seq_printf(m, "power management:");
++ {
++ unsigned i;
++ for (i = 0; i < 32; i++)
++ if (c->x86_power & (1 << i)) {
++ if (i < ARRAY_SIZE(x86_power_flags) &&
++ x86_power_flags[i])
++ seq_printf(m, "%s%s",
++ x86_power_flags[i][0]?" ":"",
++ x86_power_flags[i]);
++ else
++ seq_printf(m, " [%d]", i);
++ }
++ }
++
++ seq_printf(m, "\n\n");
++
++ return 0;
++}
++
++static void *c_start(struct seq_file *m, loff_t *pos)
++{
++ return *pos < NR_CPUS ? cpu_data + *pos : NULL;
++}
++
++static void *c_next(struct seq_file *m, void *v, loff_t *pos)
++{
++ ++*pos;
++ return c_start(m, pos);
++}
++
++static void c_stop(struct seq_file *m, void *v)
++{
++}
++
++struct seq_operations cpuinfo_op = {
++ .start =c_start,
++ .next = c_next,
++ .stop = c_stop,
++ .show = show_cpuinfo,
++};
++
++#ifdef CONFIG_INPUT_PCSPKR
++#include <linux/platform_device.h>
++static __init int add_pcspkr(void)
++{
++ struct platform_device *pd;
++ int ret;
++
++ pd = platform_device_alloc("pcspkr", -1);
++ if (!pd)
++ return -ENOMEM;
++
++ ret = platform_device_add(pd);
++ if (ret)
++ platform_device_put(pd);
++
++ return ret;
++}
++device_initcall(add_pcspkr);
++#endif
+diff -urNp linux-2.6/arch/x86_64/kernel/smp.c new/arch/x86_64/kernel/smp.c
+--- linux-2.6/arch/x86_64/kernel/smp.c 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/kernel/smp.c 2006-05-09 12:33:21.000000000 +0200
+@@ -135,10 +135,10 @@ asmlinkage void smp_invalidate_interrupt
+
+ cpu = smp_processor_id();
+ /*
+- * orig_rax contains the interrupt vector - 256.
++ * orig_rax contains the negated interrupt vector.
+ * Use that to determine where the sender put the data.
+ */
+- sender = regs->orig_rax + 256 - INVALIDATE_TLB_VECTOR_START;
++ sender = ~regs->orig_rax - INVALIDATE_TLB_VECTOR_START;
+ f = &per_cpu(flush_state, sender);
+
+ if (!cpu_isset(cpu, f->flush_cpumask))
+diff -urNp linux-2.6/arch/x86_64/kernel/smp-xen.c new/arch/x86_64/kernel/smp-xen.c
+--- linux-2.6/arch/x86_64/kernel/smp-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/smp-xen.c 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,596 @@
++/*
++ * Intel SMP support routines.
++ *
++ * (c) 1995 Alan Cox, Building #3 <alan@redhat.com>
++ * (c) 1998-99, 2000 Ingo Molnar <mingo@redhat.com>
++ * (c) 2002,2003 Andi Kleen, SuSE Labs.
++ *
++ * This code is released under the GNU General Public License version 2 or
++ * later.
++ */
++
++#include <linux/init.h>
++
++#include <linux/mm.h>
++#include <linux/delay.h>
++#include <linux/spinlock.h>
++#include <linux/smp_lock.h>
++#include <linux/smp.h>
++#include <linux/kernel_stat.h>
++#include <linux/mc146818rtc.h>
++#include <linux/interrupt.h>
++
++#include <asm/mtrr.h>
++#include <asm/pgalloc.h>
++#include <asm/tlbflush.h>
++#include <asm/mach_apic.h>
++#include <asm/mmu_context.h>
++#include <asm/proto.h>
++#include <asm/apicdef.h>
++#include <asm/idle.h>
++#ifdef CONFIG_XEN
++#include <xen/evtchn.h>
++#endif
++
++#ifndef CONFIG_XEN
++/*
++ * Smarter SMP flushing macros.
++ * c/o Linus Torvalds.
++ *
++ * These mean you can really definitely utterly forget about
++ * writing to user space from interrupts. (Its not allowed anyway).
++ *
++ * Optimizations Manfred Spraul <manfred@colorfullife.com>
++ *
++ * More scalable flush, from Andi Kleen
++ *
++ * To avoid global state use 8 different call vectors.
++ * Each CPU uses a specific vector to trigger flushes on other
++ * CPUs. Depending on the received vector the target CPUs look into
++ * the right per cpu variable for the flush data.
++ *
++ * With more than 8 CPUs they are hashed to the 8 available
++ * vectors. The limited global vector space forces us to this right now.
++ * In future when interrupts are split into per CPU domains this could be
++ * fixed, at the cost of triggering multiple IPIs in some cases.
++ */
++
++union smp_flush_state {
++ struct {
++ cpumask_t flush_cpumask;
++ struct mm_struct *flush_mm;
++ unsigned long flush_va;
++#define FLUSH_ALL -1ULL
++ spinlock_t tlbstate_lock;
++ };
++ char pad[SMP_CACHE_BYTES];
++} ____cacheline_aligned;
++
++/* State is put into the per CPU data section, but padded
++ to a full cache line because other CPUs can access it and we don't
++ want false sharing in the per cpu data segment. */
++static DEFINE_PER_CPU(union smp_flush_state, flush_state);
++#endif
++
++/*
++ * We cannot call mmdrop() because we are in interrupt context,
++ * instead update mm->cpu_vm_mask.
++ */
++static inline void leave_mm(unsigned long cpu)
++{
++ if (read_pda(mmu_state) == TLBSTATE_OK)
++ BUG();
++ cpu_clear(cpu, read_pda(active_mm)->cpu_vm_mask);
++ load_cr3(swapper_pg_dir);
++}
++
++#ifndef CONFIG_XEN
++/*
++ *
++ * The flush IPI assumes that a thread switch happens in this order:
++ * [cpu0: the cpu that switches]
++ * 1) switch_mm() either 1a) or 1b)
++ * 1a) thread switch to a different mm
++ * 1a1) cpu_clear(cpu, old_mm->cpu_vm_mask);
++ * Stop ipi delivery for the old mm. This is not synchronized with
++ * the other cpus, but smp_invalidate_interrupt ignore flush ipis
++ * for the wrong mm, and in the worst case we perform a superfluous
++ * tlb flush.
++ * 1a2) set cpu mmu_state to TLBSTATE_OK
++ * Now the smp_invalidate_interrupt won't call leave_mm if cpu0
++ * was in lazy tlb mode.
++ * 1a3) update cpu active_mm
++ * Now cpu0 accepts tlb flushes for the new mm.
++ * 1a4) cpu_set(cpu, new_mm->cpu_vm_mask);
++ * Now the other cpus will send tlb flush ipis.
++ * 1a4) change cr3.
++ * 1b) thread switch without mm change
++ * cpu active_mm is correct, cpu0 already handles
++ * flush ipis.
++ * 1b1) set cpu mmu_state to TLBSTATE_OK
++ * 1b2) test_and_set the cpu bit in cpu_vm_mask.
++ * Atomically set the bit [other cpus will start sending flush ipis],
++ * and test the bit.
++ * 1b3) if the bit was 0: leave_mm was called, flush the tlb.
++ * 2) switch %%esp, ie current
++ *
++ * The interrupt must handle 2 special cases:
++ * - cr3 is changed before %%esp, ie. it cannot use current->{active_,}mm.
++ * - the cpu performs speculative tlb reads, i.e. even if the cpu only
++ * runs in kernel space, the cpu could load tlb entries for user space
++ * pages.
++ *
++ * The good news is that cpu mmu_state is local to each cpu, no
++ * write/read ordering problems.
++ */
++
++/*
++ * TLB flush IPI:
++ *
++ * 1) Flush the tlb entries if the cpu uses the mm that's being flushed.
++ * 2) Leave the mm if we are in the lazy tlb mode.
++ *
++ * Interrupts are disabled.
++ */
++
++asmlinkage void smp_invalidate_interrupt(struct pt_regs *regs)
++{
++ int cpu;
++ int sender;
++ union smp_flush_state *f;
++
++ cpu = smp_processor_id();
++ /*
++ * orig_rax contains the interrupt vector - 256.
++ * Use that to determine where the sender put the data.
++ */
++ sender = regs->orig_rax + 256 - INVALIDATE_TLB_VECTOR_START;
++ f = &per_cpu(flush_state, sender);
++
++ if (!cpu_isset(cpu, f->flush_cpumask))
++ goto out;
++ /*
++ * This was a BUG() but until someone can quote me the
++ * line from the intel manual that guarantees an IPI to
++ * multiple CPUs is retried _only_ on the erroring CPUs
++ * its staying as a return
++ *
++ * BUG();
++ */
++
++ if (f->flush_mm == read_pda(active_mm)) {
++ if (read_pda(mmu_state) == TLBSTATE_OK) {
++ if (f->flush_va == FLUSH_ALL)
++ local_flush_tlb();
++ else
++ __flush_tlb_one(f->flush_va);
++ } else
++ leave_mm(cpu);
++ }
++out:
++ ack_APIC_irq();
++ cpu_clear(cpu, f->flush_cpumask);
++}
++
++static void flush_tlb_others(cpumask_t cpumask, struct mm_struct *mm,
++ unsigned long va)
++{
++ int sender;
++ union smp_flush_state *f;
++
++ /* Caller has disabled preemption */
++ sender = smp_processor_id() % NUM_INVALIDATE_TLB_VECTORS;
++ f = &per_cpu(flush_state, sender);
++
++ /* Could avoid this lock when
++ num_online_cpus() <= NUM_INVALIDATE_TLB_VECTORS, but it is
++ probably not worth checking this for a cache-hot lock. */
++ spin_lock(&f->tlbstate_lock);
++
++ f->flush_mm = mm;
++ f->flush_va = va;
++ cpus_or(f->flush_cpumask, cpumask, f->flush_cpumask);
++
++ /*
++ * We have to send the IPI only to
++ * CPUs affected.
++ */
++ send_IPI_mask(cpumask, INVALIDATE_TLB_VECTOR_START + sender);
++
++ while (!cpus_empty(f->flush_cpumask))
++ cpu_relax();
++
++ f->flush_mm = NULL;
++ f->flush_va = 0;
++ spin_unlock(&f->tlbstate_lock);
++}
++
++int __cpuinit init_smp_flush(void)
++{
++ int i;
++ for_each_cpu_mask(i, cpu_possible_map) {
++ spin_lock_init(&per_cpu(flush_state.tlbstate_lock, i));
++ }
++ return 0;
++}
++
++core_initcall(init_smp_flush);
++
++void flush_tlb_current_task(void)
++{
++ struct mm_struct *mm = current->mm;
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ local_flush_tlb();
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
++ preempt_enable();
++}
++
++void flush_tlb_mm (struct mm_struct * mm)
++{
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ if (current->active_mm == mm) {
++ if (current->mm)
++ local_flush_tlb();
++ else
++ leave_mm(smp_processor_id());
++ }
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, FLUSH_ALL);
++
++ preempt_enable();
++}
++
++void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
++{
++ struct mm_struct *mm = vma->vm_mm;
++ cpumask_t cpu_mask;
++
++ preempt_disable();
++ cpu_mask = mm->cpu_vm_mask;
++ cpu_clear(smp_processor_id(), cpu_mask);
++
++ if (current->active_mm == mm) {
++ if(current->mm)
++ __flush_tlb_one(va);
++ else
++ leave_mm(smp_processor_id());
++ }
++
++ if (!cpus_empty(cpu_mask))
++ flush_tlb_others(cpu_mask, mm, va);
++
++ preempt_enable();
++}
++
++static void do_flush_tlb_all(void* info)
++{
++ unsigned long cpu = smp_processor_id();
++
++ __flush_tlb_all();
++ if (read_pda(mmu_state) == TLBSTATE_LAZY)
++ leave_mm(cpu);
++}
++
++void flush_tlb_all(void)
++{
++ on_each_cpu(do_flush_tlb_all, NULL, 1, 1);
++}
++#else
++asmlinkage void smp_invalidate_interrupt (void)
++{ return; }
++void flush_tlb_current_task(void)
++{ xen_tlb_flush_mask(¤t->mm->cpu_vm_mask); }
++void flush_tlb_mm (struct mm_struct * mm)
++{ xen_tlb_flush_mask(&mm->cpu_vm_mask); }
++void flush_tlb_page(struct vm_area_struct * vma, unsigned long va)
++{ xen_invlpg_mask(&vma->vm_mm->cpu_vm_mask, va); }
++void flush_tlb_all(void)
++{ xen_tlb_flush_all(); }
++#endif /* Xen */
++
++/*
++ * this function sends a 'reschedule' IPI to another CPU.
++ * it goes straight through and wastes no time serializing
++ * anything. Worst case is that we lose a reschedule ...
++ */
++
++void smp_send_reschedule(int cpu)
++{
++ send_IPI_mask(cpumask_of_cpu(cpu), RESCHEDULE_VECTOR);
++}
++
++/*
++ * Structure and data for smp_call_function(). This is designed to minimise
++ * static memory requirements. It also looks cleaner.
++ */
++static DEFINE_SPINLOCK(call_lock);
++
++struct call_data_struct {
++ void (*func) (void *info);
++ void *info;
++ atomic_t started;
++ atomic_t finished;
++ int wait;
++};
++
++static struct call_data_struct * call_data;
++
++void lock_ipi_call_lock(void)
++{
++ spin_lock_irq(&call_lock);
++}
++
++void unlock_ipi_call_lock(void)
++{
++ spin_unlock_irq(&call_lock);
++}
++
++/*
++ * this function sends a 'generic call function' IPI to one other CPU
++ * in the system.
++ *
++ * cpu is a standard Linux logical CPU number.
++ */
++static void
++__smp_call_function_single(int cpu, void (*func) (void *info), void *info,
++ int nonatomic, int wait)
++{
++ struct call_data_struct data;
++ int cpus = 1;
++
++ data.func = func;
++ data.info = info;
++ atomic_set(&data.started, 0);
++ data.wait = wait;
++ if (wait)
++ atomic_set(&data.finished, 0);
++
++ call_data = &data;
++ wmb();
++ /* Send a message to all other CPUs and wait for them to respond */
++ send_IPI_mask(cpumask_of_cpu(cpu), CALL_FUNCTION_VECTOR);
++
++ /* Wait for response */
++ while (atomic_read(&data.started) != cpus)
++ cpu_relax();
++
++ if (!wait)
++ return;
++
++ while (atomic_read(&data.finished) != cpus)
++ cpu_relax();
++}
++
++/*
++ * smp_call_function_single - Run a function on another CPU
++ * @func: The function to run. This must be fast and non-blocking.
++ * @info: An arbitrary pointer to pass to the function.
++ * @nonatomic: Currently unused.
++ * @wait: If true, wait until function has completed on other CPUs.
++ *
++ * Retrurns 0 on success, else a negative status code.
++ *
++ * Does not return until the remote CPU is nearly ready to execute <func>
++ * or is or has executed.
++ */
++
++int smp_call_function_single (int cpu, void (*func) (void *info), void *info,
++ int nonatomic, int wait)
++{
++ /* prevent preemption and reschedule on another processor */
++ int me = get_cpu();
++ if (cpu == me) {
++ WARN_ON(1);
++ put_cpu();
++ return -EBUSY;
++ }
++ spin_lock_bh(&call_lock);
++ __smp_call_function_single(cpu, func, info, nonatomic, wait);
++ spin_unlock_bh(&call_lock);
++ put_cpu();
++ return 0;
++}
++
++/*
++ * this function sends a 'generic call function' IPI to all other CPUs
++ * in the system.
++ */
++static void __smp_call_function (void (*func) (void *info), void *info,
++ int nonatomic, int wait)
++{
++ struct call_data_struct data;
++ int cpus = num_online_cpus()-1;
++
++ if (!cpus)
++ return;
++
++ data.func = func;
++ data.info = info;
++ atomic_set(&data.started, 0);
++ data.wait = wait;
++ if (wait)
++ atomic_set(&data.finished, 0);
++
++ call_data = &data;
++ wmb();
++ /* Send a message to all other CPUs and wait for them to respond */
++ send_IPI_allbutself(CALL_FUNCTION_VECTOR);
++
++ /* Wait for response */
++ while (atomic_read(&data.started) != cpus)
++#ifndef CONFIG_XEN
++ cpu_relax();
++#else
++ barrier();
++#endif
++
++ if (!wait)
++ return;
++
++ while (atomic_read(&data.finished) != cpus)
++#ifndef CONFIG_XEN
++ cpu_relax();
++#else
++ barrier();
++#endif
++}
++
++/*
++ * smp_call_function - run a function on all other CPUs.
++ * @func: The function to run. This must be fast and non-blocking.
++ * @info: An arbitrary pointer to pass to the function.
++ * @nonatomic: currently unused.
++ * @wait: If true, wait (atomically) until function has completed on other
++ * CPUs.
++ *
++ * Returns 0 on success, else a negative status code. Does not return until
++ * remote CPUs are nearly ready to execute func or are or have executed.
++ *
++ * You must not call this function with disabled interrupts or from a
++ * hardware interrupt handler or from a bottom half handler.
++ * Actually there are a few legal cases, like panic.
++ */
++int smp_call_function (void (*func) (void *info), void *info, int nonatomic,
++ int wait)
++{
++ spin_lock(&call_lock);
++ __smp_call_function(func,info,nonatomic,wait);
++ spin_unlock(&call_lock);
++ return 0;
++}
++
++void smp_stop_cpu(void)
++{
++ unsigned long flags;
++ /*
++ * Remove this CPU:
++ */
++ cpu_clear(smp_processor_id(), cpu_online_map);
++ local_irq_save(flags);
++#ifndef CONFIG_XEN
++ disable_local_APIC();
++#endif
++ local_irq_restore(flags);
++}
++
++static void smp_really_stop_cpu(void *dummy)
++{
++ smp_stop_cpu();
++ for (;;)
++ halt();
++}
++
++void smp_send_stop(void)
++{
++ int nolock = 0;
++#ifndef CONFIG_XEN
++ if (reboot_force)
++ return;
++#endif
++ /* Don't deadlock on the call lock in panic */
++ if (!spin_trylock(&call_lock)) {
++ /* ignore locking because we have paniced anyways */
++ nolock = 1;
++ }
++ __smp_call_function(smp_really_stop_cpu, NULL, 0, 0);
++ if (!nolock)
++ spin_unlock(&call_lock);
++
++ local_irq_disable();
++#ifndef CONFIG_XEN
++ disable_local_APIC();
++#endif
++ local_irq_enable();
++}
++
++/*
++ * Reschedule call back. Nothing to do,
++ * all the work is done automatically when
++ * we return from the interrupt.
++ */
++#ifndef CONFIG_XEN
++asmlinkage void smp_reschedule_interrupt(void)
++#else
++asmlinkage irqreturn_t smp_reschedule_interrupt(void)
++#endif
++{
++#ifndef CONFIG_XEN
++ ack_APIC_irq();
++#else
++ return IRQ_HANDLED;
++#endif
++}
++
++#ifndef CONFIG_XEN
++asmlinkage void smp_call_function_interrupt(void)
++#else
++asmlinkage irqreturn_t smp_call_function_interrupt(void)
++#endif
++{
++ void (*func) (void *info) = call_data->func;
++ void *info = call_data->info;
++ int wait = call_data->wait;
++
++#ifndef CONFIG_XEN
++ ack_APIC_irq();
++#endif
++ /*
++ * Notify initiating CPU that I've grabbed the data and am
++ * about to execute the function
++ */
++ mb();
++ atomic_inc(&call_data->started);
++ /*
++ * At this point the info structure may be out of scope unless wait==1
++ */
++ exit_idle();
++ irq_enter();
++ (*func)(info);
++ irq_exit();
++ if (wait) {
++ mb();
++ atomic_inc(&call_data->finished);
++ }
++#ifdef CONFIG_XEN
++ return IRQ_HANDLED;
++#endif
++}
++
++int safe_smp_processor_id(void)
++{
++#ifdef CONFIG_XEN
++ return smp_processor_id();
++#else
++ int apicid, i;
++
++ if (disable_apic)
++ return 0;
++
++ apicid = hard_smp_processor_id();
++ if (x86_cpu_to_apicid[apicid] == apicid)
++ return apicid;
++
++ for (i = 0; i < NR_CPUS; ++i) {
++ if (x86_cpu_to_apicid[i] == apicid)
++ return i;
++ }
++
++ /* No entries in x86_cpu_to_apicid? Either no MPS|ACPI,
++ * or called too early. Either way, we must be CPU 0. */
++ if (x86_cpu_to_apicid[0] == BAD_APICID)
++ return 0;
++
++ return 0; /* Should not happen */
++#endif
++}
+diff -urNp linux-2.6/arch/x86_64/kernel/traps-xen.c new/arch/x86_64/kernel/traps-xen.c
+--- linux-2.6/arch/x86_64/kernel/traps-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/traps-xen.c 2006-05-23 18:37:10.000000000 +0200
+@@ -0,0 +1,1035 @@
++/*
++ * linux/arch/x86-64/traps.c
++ *
++ * Copyright (C) 1991, 1992 Linus Torvalds
++ * Copyright (C) 2000, 2001, 2002 Andi Kleen, SuSE Labs
++ *
++ * Pentium III FXSR, SSE support
++ * Gareth Hughes <gareth@valinux.com>, May 2000
++ *
++ * $Id$
++ */
++
++/*
++ * 'Traps.c' handles hardware traps and faults after we have saved some
++ * state in 'entry.S'.
++ */
++#include <linux/config.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/ptrace.h>
++#include <linux/timer.h>
++#include <linux/mm.h>
++#include <linux/init.h>
++#include <linux/delay.h>
++#include <linux/spinlock.h>
++#include <linux/interrupt.h>
++#include <linux/module.h>
++#include <linux/moduleparam.h>
++#include <linux/nmi.h>
++#include <linux/kprobes.h>
++#include <linux/kexec.h>
++
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/io.h>
++#include <asm/atomic.h>
++#include <asm/debugreg.h>
++#include <asm/desc.h>
++#include <asm/i387.h>
++#include <asm/kdebug.h>
++#include <asm/processor.h>
++
++#include <asm/smp.h>
++#include <asm/pgalloc.h>
++#include <asm/pda.h>
++#include <asm/proto.h>
++#include <asm/nmi.h>
++
++asmlinkage void divide_error(void);
++asmlinkage void debug(void);
++asmlinkage void nmi(void);
++asmlinkage void int3(void);
++asmlinkage void overflow(void);
++asmlinkage void bounds(void);
++asmlinkage void invalid_op(void);
++asmlinkage void device_not_available(void);
++asmlinkage void double_fault(void);
++asmlinkage void coprocessor_segment_overrun(void);
++asmlinkage void invalid_TSS(void);
++asmlinkage void segment_not_present(void);
++asmlinkage void stack_segment(void);
++asmlinkage void general_protection(void);
++asmlinkage void page_fault(void);
++asmlinkage void coprocessor_error(void);
++asmlinkage void simd_coprocessor_error(void);
++asmlinkage void reserved(void);
++asmlinkage void alignment_check(void);
++asmlinkage void machine_check(void);
++asmlinkage void spurious_interrupt_bug(void);
++
++ATOMIC_NOTIFIER_HEAD(die_chain);
++
++int register_die_notifier(struct notifier_block *nb)
++{
++ vmalloc_sync_all();
++ return atomic_notifier_chain_register(&die_chain, nb);
++}
++EXPORT_SYMBOL(register_die_notifier);
++
++int unregister_die_notifier(struct notifier_block *nb)
++{
++ return atomic_notifier_chain_unregister(&die_chain, nb);
++}
++EXPORT_SYMBOL(unregister_die_notifier);
++
++static inline void conditional_sti(struct pt_regs *regs)
++{
++ if (regs->eflags & X86_EFLAGS_IF)
++ local_irq_enable();
++}
++
++static inline void preempt_conditional_sti(struct pt_regs *regs)
++{
++ preempt_disable();
++ if (regs->eflags & X86_EFLAGS_IF)
++ local_irq_enable();
++}
++
++static inline void preempt_conditional_cli(struct pt_regs *regs)
++{
++ if (regs->eflags & X86_EFLAGS_IF)
++ local_irq_disable();
++ preempt_enable_no_resched();
++}
++
++static int kstack_depth_to_print = 10;
++
++#ifdef CONFIG_KALLSYMS
++#include <linux/kallsyms.h>
++int printk_address(unsigned long address)
++{
++ unsigned long offset = 0, symsize;
++ const char *symname;
++ char *modname;
++ char *delim = ":";
++ char namebuf[128];
++
++ symname = kallsyms_lookup(address, &symsize, &offset, &modname, namebuf);
++ if (!symname)
++ return printk("[<%016lx>]", address);
++ if (!modname)
++ modname = delim = "";
++ return printk("<%016lx>{%s%s%s%s%+ld}",
++ address, delim, modname, delim, symname, offset);
++}
++#else
++int printk_address(unsigned long address)
++{
++ return printk("[<%016lx>]", address);
++}
++#endif
++
++static unsigned long *in_exception_stack(unsigned cpu, unsigned long stack,
++ unsigned *usedp, const char **idp)
++{
++#ifndef CONFIG_X86_NO_TSS
++ static char ids[][8] = {
++ [DEBUG_STACK - 1] = "#DB",
++ [NMI_STACK - 1] = "NMI",
++ [DOUBLEFAULT_STACK - 1] = "#DF",
++ [STACKFAULT_STACK - 1] = "#SS",
++ [MCE_STACK - 1] = "#MC",
++#if DEBUG_STKSZ > EXCEPTION_STKSZ
++ [N_EXCEPTION_STACKS ... N_EXCEPTION_STACKS + DEBUG_STKSZ / EXCEPTION_STKSZ - 2] = "#DB[?]"
++#endif
++ };
++ unsigned k;
++
++ for (k = 0; k < N_EXCEPTION_STACKS; k++) {
++ unsigned long end;
++
++ switch (k + 1) {
++#if DEBUG_STKSZ > EXCEPTION_STKSZ
++ case DEBUG_STACK:
++ end = cpu_pda(cpu)->debugstack + DEBUG_STKSZ;
++ break;
++#endif
++ default:
++ end = per_cpu(init_tss, cpu).ist[k];
++ break;
++ }
++ if (stack >= end)
++ continue;
++ if (stack >= end - EXCEPTION_STKSZ) {
++ if (*usedp & (1U << k))
++ break;
++ *usedp |= 1U << k;
++ *idp = ids[k];
++ return (unsigned long *)end;
++ }
++#if DEBUG_STKSZ > EXCEPTION_STKSZ
++ if (k == DEBUG_STACK - 1 && stack >= end - DEBUG_STKSZ) {
++ unsigned j = N_EXCEPTION_STACKS - 1;
++
++ do {
++ ++j;
++ end -= EXCEPTION_STKSZ;
++ ids[j][4] = '1' + (j - N_EXCEPTION_STACKS);
++ } while (stack < end - EXCEPTION_STKSZ);
++ if (*usedp & (1U << j))
++ break;
++ *usedp |= 1U << j;
++ *idp = ids[j];
++ return (unsigned long *)end;
++ }
++#endif
++ }
++#endif
++ return NULL;
++}
++
++/*
++ * x86-64 can have upto three kernel stacks:
++ * process stack
++ * interrupt stack
++ * severe exception (double fault, nmi, stack fault, debug, mce) hardware stack
++ */
++
++void show_trace(unsigned long *stack)
++{
++ const unsigned cpu = safe_smp_processor_id();
++ unsigned long *irqstack_end = (unsigned long *)cpu_pda(cpu)->irqstackptr;
++ int i;
++ unsigned used = 0;
++
++ printk("\nCall Trace:");
++
++#define HANDLE_STACK(cond) \
++ do while (cond) { \
++ unsigned long addr = *stack++; \
++ if (kernel_text_address(addr)) { \
++ if (i > 50) { \
++ printk("\n "); \
++ i = 0; \
++ } \
++ else \
++ i += printk(" "); \
++ /* \
++ * If the address is either in the text segment of the \
++ * kernel, or in the region which contains vmalloc'ed \
++ * memory, it *may* be the address of a calling \
++ * routine; if so, print it so that someone tracing \
++ * down the cause of the crash will be able to figure \
++ * out the call path that was taken. \
++ */ \
++ i += printk_address(addr); \
++ } \
++ } while (0)
++
++ for(i = 11; ; ) {
++ const char *id;
++ unsigned long *estack_end;
++ estack_end = in_exception_stack(cpu, (unsigned long)stack,
++ &used, &id);
++
++ if (estack_end) {
++ i += printk(" <%s>", id);
++ HANDLE_STACK (stack < estack_end);
++ i += printk(" <EOE>");
++ stack = (unsigned long *) estack_end[-2];
++ continue;
++ }
++ if (irqstack_end) {
++ unsigned long *irqstack;
++ irqstack = irqstack_end -
++ (IRQSTACKSIZE - 64) / sizeof(*irqstack);
++
++ if (stack >= irqstack && stack < irqstack_end) {
++ i += printk(" <IRQ>");
++ HANDLE_STACK (stack < irqstack_end);
++ stack = (unsigned long *) (irqstack_end[-1]);
++ irqstack_end = NULL;
++ i += printk(" <EOI>");
++ continue;
++ }
++ }
++ break;
++ }
++
++ HANDLE_STACK (((long) stack & (THREAD_SIZE-1)) != 0);
++#undef HANDLE_STACK
++ printk("\n");
++}
++
++void show_stack(struct task_struct *tsk, unsigned long * rsp)
++{
++ unsigned long *stack;
++ int i;
++ const int cpu = safe_smp_processor_id();
++ unsigned long *irqstack_end = (unsigned long *) (cpu_pda(cpu)->irqstackptr);
++ unsigned long *irqstack = (unsigned long *) (cpu_pda(cpu)->irqstackptr - IRQSTACKSIZE);
++
++ // debugging aid: "show_stack(NULL, NULL);" prints the
++ // back trace for this cpu.
++
++ if (rsp == NULL) {
++ if (tsk)
++ rsp = (unsigned long *)tsk->thread.rsp;
++ else
++ rsp = (unsigned long *)&rsp;
++ }
++
++ stack = rsp;
++ for(i=0; i < kstack_depth_to_print; i++) {
++ if (stack >= irqstack && stack <= irqstack_end) {
++ if (stack == irqstack_end) {
++ stack = (unsigned long *) (irqstack_end[-1]);
++ printk(" <EOI> ");
++ }
++ } else {
++ if (((long) stack & (THREAD_SIZE-1)) == 0)
++ break;
++ }
++ if (i && ((i % 4) == 0))
++ printk("\n ");
++ printk("%016lx ", *stack++);
++ touch_nmi_watchdog();
++ }
++ show_trace((unsigned long *)rsp);
++}
++
++/*
++ * The architecture-independent dump_stack generator
++ */
++void dump_stack(void)
++{
++ unsigned long dummy;
++ show_trace(&dummy);
++}
++
++EXPORT_SYMBOL(dump_stack);
++
++void show_registers(struct pt_regs *regs)
++{
++ int i;
++ int in_kernel = !user_mode(regs);
++ unsigned long rsp;
++ const int cpu = safe_smp_processor_id();
++ struct task_struct *cur = cpu_pda(cpu)->pcurrent;
++
++ rsp = regs->rsp;
++
++ printk("CPU %d ", cpu);
++ __show_regs(regs);
++ printk("Process %s (pid: %d, threadinfo %p, task %p)\n",
++ cur->comm, cur->pid, task_thread_info(cur), cur);
++
++ /*
++ * When in-kernel, we also print out the stack and code at the
++ * time of the fault..
++ */
++ if (in_kernel) {
++
++ printk("Stack: ");
++ show_stack(NULL, (unsigned long*)rsp);
++
++ printk("\nCode: ");
++ if (regs->rip < PAGE_OFFSET)
++ goto bad;
++
++ for (i=0; i<20; i++) {
++ unsigned char c;
++ if (__get_user(c, &((unsigned char*)regs->rip)[i])) {
++bad:
++ printk(" Bad RIP value.");
++ break;
++ }
++ printk("%02x ", c);
++ }
++ }
++ printk("\n");
++}
++
++void handle_BUG(struct pt_regs *regs)
++{
++ struct bug_frame f;
++ long len;
++ const char *prefix = "";
++
++ if (user_mode(regs))
++ return;
++ if (__copy_from_user(&f, (const void __user *) regs->rip,
++ sizeof(struct bug_frame)))
++ return;
++ if (f.filename >= 0 ||
++ f.ud2[0] != 0x0f || f.ud2[1] != 0x0b)
++ return;
++ len = __strnlen_user((char *)(long)f.filename, PATH_MAX) - 1;
++ if (len < 0 || len >= PATH_MAX)
++ f.filename = (int)(long)"unmapped filename";
++ else if (len > 50) {
++ f.filename += len - 50;
++ prefix = "...";
++ }
++ printk("----------- [cut here ] --------- [please bite here ] ---------\n");
++ printk(KERN_ALERT "Kernel BUG at %s%.50s:%d\n", prefix, (char *)(long)f.filename, f.line);
++}
++
++#ifdef CONFIG_BUG
++void out_of_line_bug(void)
++{
++ BUG();
++}
++#endif
++
++static DEFINE_SPINLOCK(die_lock);
++static int die_owner = -1;
++static unsigned int die_nest_count;
++
++unsigned __kprobes long oops_begin(void)
++{
++ int cpu = safe_smp_processor_id();
++ unsigned long flags;
++
++ /* racy, but better than risking deadlock. */
++ local_irq_save(flags);
++ if (!spin_trylock(&die_lock)) {
++ if (cpu == die_owner)
++ /* nested oops. should stop eventually */;
++ else
++ spin_lock(&die_lock);
++ }
++ die_nest_count++;
++ die_owner = cpu;
++ console_verbose();
++ bust_spinlocks(1);
++ return flags;
++}
++
++void __kprobes oops_end(unsigned long flags)
++{
++ die_owner = -1;
++ bust_spinlocks(0);
++ die_nest_count--;
++ if (die_nest_count)
++ /* We still own the lock */
++ local_irq_restore(flags);
++ else
++ /* Nest count reaches zero, release the lock. */
++ spin_unlock_irqrestore(&die_lock, flags);
++ if (panic_on_oops)
++ panic("Oops");
++}
++
++void __kprobes __die(const char * str, struct pt_regs * regs, long err)
++{
++ static int die_counter;
++ printk(KERN_EMERG "%s: %04lx [%u] ", str, err & 0xffff,++die_counter);
++#ifdef CONFIG_PREEMPT
++ printk("PREEMPT ");
++#endif
++#ifdef CONFIG_SMP
++ printk("SMP ");
++#endif
++#ifdef CONFIG_DEBUG_PAGEALLOC
++ printk("DEBUG_PAGEALLOC");
++#endif
++ printk("\n");
++ notify_die(DIE_OOPS, str, regs, err, current->thread.trap_no, SIGSEGV);
++ show_registers(regs);
++ /* Executive summary in case the oops scrolled away */
++ printk(KERN_ALERT "RIP ");
++ printk_address(regs->rip);
++ printk(" RSP <%016lx>\n", regs->rsp);
++ if (kexec_should_crash(current))
++ crash_kexec(regs);
++}
++
++void die(const char * str, struct pt_regs * regs, long err)
++{
++ unsigned long flags = oops_begin();
++
++ handle_BUG(regs);
++ __die(str, regs, err);
++ oops_end(flags);
++ do_exit(SIGSEGV);
++}
++
++#ifdef CONFIG_X86_LOCAL_APIC
++void __kprobes die_nmi(char *str, struct pt_regs *regs)
++{
++ unsigned long flags = oops_begin();
++
++ /*
++ * We are in trouble anyway, lets at least try
++ * to get a message out.
++ */
++ printk(str, safe_smp_processor_id());
++ show_registers(regs);
++ if (kexec_should_crash(current))
++ crash_kexec(regs);
++ if (panic_on_timeout || panic_on_oops)
++ panic("nmi watchdog");
++ printk("console shuts up ...\n");
++ oops_end(flags);
++ nmi_exit();
++ local_irq_enable();
++ do_exit(SIGSEGV);
++}
++#endif
++
++static void __kprobes do_trap(int trapnr, int signr, char *str,
++ struct pt_regs * regs, long error_code,
++ siginfo_t *info)
++{
++ struct task_struct *tsk = current;
++
++ conditional_sti(regs);
++
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = trapnr;
++
++ if (user_mode(regs)) {
++ if (exception_trace && unhandled_signal(tsk, signr))
++ printk(KERN_INFO
++ "%s[%d] trap %s rip:%lx rsp:%lx error:%lx\n",
++ tsk->comm, tsk->pid, str,
++ regs->rip, regs->rsp, error_code);
++
++ if (info)
++ force_sig_info(signr, info, tsk);
++ else
++ force_sig(signr, tsk);
++ return;
++ }
++
++
++ /* kernel trap */
++ {
++ const struct exception_table_entry *fixup;
++ fixup = search_exception_tables(regs->rip);
++ if (fixup)
++ regs->rip = fixup->fixup;
++ else
++ die(str, regs, error_code);
++ return;
++ }
++}
++
++#define DO_ERROR(trapnr, signr, str, name) \
++asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, regs, error_code, NULL); \
++}
++
++#define DO_ERROR_INFO(trapnr, signr, str, name, sicode, siaddr) \
++asmlinkage void do_##name(struct pt_regs * regs, long error_code) \
++{ \
++ siginfo_t info; \
++ info.si_signo = signr; \
++ info.si_errno = 0; \
++ info.si_code = sicode; \
++ info.si_addr = (void __user *)siaddr; \
++ if (notify_die(DIE_TRAP, str, regs, error_code, trapnr, signr) \
++ == NOTIFY_STOP) \
++ return; \
++ do_trap(trapnr, signr, str, regs, error_code, &info); \
++}
++
++DO_ERROR_INFO( 0, SIGFPE, "divide error", divide_error, FPE_INTDIV, regs->rip)
++DO_ERROR( 4, SIGSEGV, "overflow", overflow)
++DO_ERROR( 5, SIGSEGV, "bounds", bounds)
++DO_ERROR_INFO( 6, SIGILL, "invalid opcode", invalid_op, ILL_ILLOPN, regs->rip)
++DO_ERROR( 7, SIGSEGV, "device not available", device_not_available)
++DO_ERROR( 9, SIGFPE, "coprocessor segment overrun", coprocessor_segment_overrun)
++DO_ERROR(10, SIGSEGV, "invalid TSS", invalid_TSS)
++DO_ERROR(11, SIGBUS, "segment not present", segment_not_present)
++DO_ERROR_INFO(17, SIGBUS, "alignment check", alignment_check, BUS_ADRALN, 0)
++DO_ERROR(18, SIGSEGV, "reserved", reserved)
++DO_ERROR(12, SIGBUS, "stack segment", stack_segment)
++
++asmlinkage void do_double_fault(struct pt_regs * regs, long error_code)
++{
++ static const char str[] = "double fault";
++ struct task_struct *tsk = current;
++
++ /* Return not checked because double check cannot be ignored */
++ notify_die(DIE_TRAP, str, regs, error_code, 8, SIGSEGV);
++
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = 8;
++
++ /* This is always a kernel trap and never fixable (and thus must
++ never return). */
++ for (;;)
++ die(str, regs, error_code);
++}
++
++asmlinkage void __kprobes do_general_protection(struct pt_regs * regs,
++ long error_code)
++{
++ struct task_struct *tsk = current;
++
++ conditional_sti(regs);
++
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = 13;
++
++ if (user_mode(regs)) {
++ if (exception_trace && unhandled_signal(tsk, SIGSEGV))
++ printk(KERN_INFO
++ "%s[%d] general protection rip:%lx rsp:%lx error:%lx\n",
++ tsk->comm, tsk->pid,
++ regs->rip, regs->rsp, error_code);
++
++ force_sig(SIGSEGV, tsk);
++ return;
++ }
++
++ /* kernel gp */
++ {
++ const struct exception_table_entry *fixup;
++ fixup = search_exception_tables(regs->rip);
++ if (fixup) {
++ regs->rip = fixup->fixup;
++ return;
++ }
++ if (notify_die(DIE_GPF, "general protection fault", regs,
++ error_code, 13, SIGSEGV) == NOTIFY_STOP)
++ return;
++ die("general protection fault", regs, error_code);
++ }
++}
++
++static __kprobes void
++mem_parity_error(unsigned char reason, struct pt_regs * regs)
++{
++ printk("Uhhuh. NMI received. Dazed and confused, but trying to continue\n");
++ printk("You probably have a hardware problem with your RAM chips\n");
++
++#if 0 /* XEN */
++ /* Clear and disable the memory parity error line. */
++ reason = (reason & 0xf) | 4;
++ outb(reason, 0x61);
++#endif /* XEN */
++}
++
++static __kprobes void
++io_check_error(unsigned char reason, struct pt_regs * regs)
++{
++ printk("NMI: IOCK error (debug interrupt?)\n");
++ show_registers(regs);
++
++#if 0 /* XEN */
++ /* Re-enable the IOCK line, wait for a few seconds */
++ reason = (reason & 0xf) | 8;
++ outb(reason, 0x61);
++ mdelay(2000);
++ reason &= ~8;
++ outb(reason, 0x61);
++#endif /* XEN */
++}
++
++static __kprobes void
++unknown_nmi_error(unsigned char reason, struct pt_regs * regs)
++{ printk("Uhhuh. NMI received for unknown reason %02x.\n", reason);
++ printk("Dazed and confused, but trying to continue\n");
++ printk("Do you have a strange power saving mode enabled?\n");
++}
++
++/* Runs on IST stack. This code must keep interrupts off all the time.
++ Nested NMIs are prevented by the CPU. */
++asmlinkage __kprobes void default_do_nmi(struct pt_regs *regs)
++{
++ unsigned char reason = 0;
++ int cpu;
++
++ cpu = smp_processor_id();
++
++ /* Only the BSP gets external NMIs from the system. */
++ if (!cpu)
++ reason = get_nmi_reason();
++
++ if (!(reason & 0xc0)) {
++ if (notify_die(DIE_NMI_IPI, "nmi_ipi", regs, reason, 2, SIGINT)
++ == NOTIFY_STOP)
++ return;
++#ifdef CONFIG_X86_LOCAL_APIC
++ /*
++ * Ok, so this is none of the documented NMI sources,
++ * so it must be the NMI watchdog.
++ */
++ if (nmi_watchdog > 0) {
++ nmi_watchdog_tick(regs,reason);
++ return;
++ }
++#endif
++ unknown_nmi_error(reason, regs);
++ return;
++ }
++ if (notify_die(DIE_NMI, "nmi", regs, reason, 2, SIGINT) == NOTIFY_STOP)
++ return;
++
++ /* AK: following checks seem to be broken on modern chipsets. FIXME */
++
++ if (reason & 0x80)
++ mem_parity_error(reason, regs);
++ if (reason & 0x40)
++ io_check_error(reason, regs);
++}
++
++/* runs on IST stack. */
++asmlinkage void __kprobes do_int3(struct pt_regs * regs, long error_code)
++{
++ if (notify_die(DIE_INT3, "int3", regs, error_code, 3, SIGTRAP) == NOTIFY_STOP) {
++ return;
++ }
++ do_trap(3, SIGTRAP, "int3", regs, error_code, NULL);
++ return;
++}
++
++/* Help handler running on IST stack to switch back to user stack
++ for scheduling or signal handling. The actual stack switch is done in
++ entry.S */
++asmlinkage __kprobes struct pt_regs *sync_regs(struct pt_regs *eregs)
++{
++ struct pt_regs *regs = eregs;
++ /* Did already sync */
++ if (eregs == (struct pt_regs *)eregs->rsp)
++ ;
++ /* Exception from user space */
++ else if (user_mode(eregs))
++ regs = task_pt_regs(current);
++ /* Exception from kernel and interrupts are enabled. Move to
++ kernel process stack. */
++ else if (eregs->eflags & X86_EFLAGS_IF)
++ regs = (struct pt_regs *)(eregs->rsp -= sizeof(struct pt_regs));
++ if (eregs != regs)
++ *regs = *eregs;
++ return regs;
++}
++
++/* runs on IST stack. */
++asmlinkage void __kprobes do_debug(struct pt_regs * regs,
++ unsigned long error_code)
++{
++ unsigned long condition;
++ struct task_struct *tsk = current;
++ siginfo_t info;
++
++ get_debugreg(condition, 6);
++
++ if (notify_die(DIE_DEBUG, "debug", regs, condition, error_code,
++ SIGTRAP) == NOTIFY_STOP)
++ return;
++
++ preempt_conditional_sti(regs);
++
++ /* Mask out spurious debug traps due to lazy DR7 setting */
++ if (condition & (DR_TRAP0|DR_TRAP1|DR_TRAP2|DR_TRAP3)) {
++ if (!tsk->thread.debugreg7) {
++ goto clear_dr7;
++ }
++ }
++
++ tsk->thread.debugreg6 = condition;
++
++ /* Mask out spurious TF errors due to lazy TF clearing */
++ if (condition & DR_STEP) {
++ /*
++ * The TF error should be masked out only if the current
++ * process is not traced and if the TRAP flag has been set
++ * previously by a tracing process (condition detected by
++ * the PT_DTRACE flag); remember that the i386 TRAP flag
++ * can be modified by the process itself in user mode,
++ * allowing programs to debug themselves without the ptrace()
++ * interface.
++ */
++ if (!user_mode(regs))
++ goto clear_TF_reenable;
++ /*
++ * Was the TF flag set by a debugger? If so, clear it now,
++ * so that register information is correct.
++ */
++ if (tsk->ptrace & PT_DTRACE) {
++ regs->eflags &= ~TF_MASK;
++ tsk->ptrace &= ~PT_DTRACE;
++ }
++ }
++
++ /* Ok, finally something we can handle */
++ tsk->thread.trap_no = 1;
++ tsk->thread.error_code = error_code;
++ info.si_signo = SIGTRAP;
++ info.si_errno = 0;
++ info.si_code = TRAP_BRKPT;
++ info.si_addr = user_mode(regs) ? (void __user *)regs->rip : NULL;
++ force_sig_info(SIGTRAP, &info, tsk);
++
++clear_dr7:
++ set_debugreg(0UL, 7);
++ preempt_conditional_cli(regs);
++ return;
++
++clear_TF_reenable:
++ set_tsk_thread_flag(tsk, TIF_SINGLESTEP);
++ regs->eflags &= ~TF_MASK;
++ preempt_conditional_cli(regs);
++}
++
++static int kernel_math_error(struct pt_regs *regs, const char *str, int trapnr)
++{
++ const struct exception_table_entry *fixup;
++ fixup = search_exception_tables(regs->rip);
++ if (fixup) {
++ regs->rip = fixup->fixup;
++ return 1;
++ }
++ notify_die(DIE_GPF, str, regs, 0, trapnr, SIGFPE);
++ /* Illegal floating point operation in the kernel */
++ current->thread.trap_no = trapnr;
++ die(str, regs, 0);
++ return 0;
++}
++
++/*
++ * Note that we play around with the 'TS' bit in an attempt to get
++ * the correct behaviour even in the presence of the asynchronous
++ * IRQ13 behaviour
++ */
++asmlinkage void do_coprocessor_error(struct pt_regs *regs)
++{
++ void __user *rip = (void __user *)(regs->rip);
++ struct task_struct * task;
++ siginfo_t info;
++ unsigned short cwd, swd;
++
++ conditional_sti(regs);
++ if (!user_mode(regs) &&
++ kernel_math_error(regs, "kernel x87 math error", 16))
++ return;
++
++ /*
++ * Save the info for the exception handler and clear the error.
++ */
++ task = current;
++ save_init_fpu(task);
++ task->thread.trap_no = 16;
++ task->thread.error_code = 0;
++ info.si_signo = SIGFPE;
++ info.si_errno = 0;
++ info.si_code = __SI_FAULT;
++ info.si_addr = rip;
++ /*
++ * (~cwd & swd) will mask out exceptions that are not set to unmasked
++ * status. 0x3f is the exception bits in these regs, 0x200 is the
++ * C1 reg you need in case of a stack fault, 0x040 is the stack
++ * fault bit. We should only be taking one exception at a time,
++ * so if this combination doesn't produce any single exception,
++ * then we have a bad program that isn't synchronizing its FPU usage
++ * and it will suffer the consequences since we won't be able to
++ * fully reproduce the context of the exception
++ */
++ cwd = get_fpu_cwd(task);
++ swd = get_fpu_swd(task);
++ switch (swd & ~cwd & 0x3f) {
++ case 0x000:
++ default:
++ break;
++ case 0x001: /* Invalid Op */
++ /*
++ * swd & 0x240 == 0x040: Stack Underflow
++ * swd & 0x240 == 0x240: Stack Overflow
++ * User must clear the SF bit (0x40) if set
++ */
++ info.si_code = FPE_FLTINV;
++ break;
++ case 0x002: /* Denormalize */
++ case 0x010: /* Underflow */
++ info.si_code = FPE_FLTUND;
++ break;
++ case 0x004: /* Zero Divide */
++ info.si_code = FPE_FLTDIV;
++ break;
++ case 0x008: /* Overflow */
++ info.si_code = FPE_FLTOVF;
++ break;
++ case 0x020: /* Precision */
++ info.si_code = FPE_FLTRES;
++ break;
++ }
++ force_sig_info(SIGFPE, &info, task);
++}
++
++asmlinkage void bad_intr(void)
++{
++ printk("bad interrupt");
++}
++
++asmlinkage void do_simd_coprocessor_error(struct pt_regs *regs)
++{
++ void __user *rip = (void __user *)(regs->rip);
++ struct task_struct * task;
++ siginfo_t info;
++ unsigned short mxcsr;
++
++ conditional_sti(regs);
++ if (!user_mode(regs) &&
++ kernel_math_error(regs, "kernel simd math error", 19))
++ return;
++
++ /*
++ * Save the info for the exception handler and clear the error.
++ */
++ task = current;
++ save_init_fpu(task);
++ task->thread.trap_no = 19;
++ task->thread.error_code = 0;
++ info.si_signo = SIGFPE;
++ info.si_errno = 0;
++ info.si_code = __SI_FAULT;
++ info.si_addr = rip;
++ /*
++ * The SIMD FPU exceptions are handled a little differently, as there
++ * is only a single status/control register. Thus, to determine which
++ * unmasked exception was caught we must mask the exception mask bits
++ * at 0x1f80, and then use these to mask the exception bits at 0x3f.
++ */
++ mxcsr = get_fpu_mxcsr(task);
++ switch (~((mxcsr & 0x1f80) >> 7) & (mxcsr & 0x3f)) {
++ case 0x000:
++ default:
++ break;
++ case 0x001: /* Invalid Op */
++ info.si_code = FPE_FLTINV;
++ break;
++ case 0x002: /* Denormalize */
++ case 0x010: /* Underflow */
++ info.si_code = FPE_FLTUND;
++ break;
++ case 0x004: /* Zero Divide */
++ info.si_code = FPE_FLTDIV;
++ break;
++ case 0x008: /* Overflow */
++ info.si_code = FPE_FLTOVF;
++ break;
++ case 0x020: /* Precision */
++ info.si_code = FPE_FLTRES;
++ break;
++ }
++ force_sig_info(SIGFPE, &info, task);
++}
++
++asmlinkage void do_spurious_interrupt_bug(struct pt_regs * regs)
++{
++}
++
++#if 0
++asmlinkage void __attribute__((weak)) smp_thermal_interrupt(void)
++{
++}
++#endif
++
++asmlinkage void __attribute__((weak)) mce_threshold_interrupt(void)
++{
++}
++
++/*
++ * 'math_state_restore()' saves the current math information in the
++ * old math state array, and gets the new ones from the current task
++ *
++ * Careful.. There are problems with IBM-designed IRQ13 behaviour.
++ * Don't touch unless you *really* know how it works.
++ */
++asmlinkage void math_state_restore(void)
++{
++ struct task_struct *me = current;
++ /* clts(); */ /* 'clts' is done for us by Xen during virtual trap. */
++
++ if (!used_math())
++ init_fpu(me);
++ restore_fpu_checking(&me->thread.i387.fxsave);
++ task_thread_info(me)->status |= TS_USEDFPU;
++}
++
++
++/*
++ * NB. All these are "interrupt gates" (i.e. events_mask is set) because we
++ * specify <dpl>|4 in the second field.
++ */
++static trap_info_t trap_table[] = {
++ { 0, 0|4, __KERNEL_CS, (unsigned long)divide_error },
++ { 1, 0|4, __KERNEL_CS, (unsigned long)debug },
++ { 3, 3|4, __KERNEL_CS, (unsigned long)int3 },
++ { 4, 3|4, __KERNEL_CS, (unsigned long)overflow },
++ { 5, 0|4, __KERNEL_CS, (unsigned long)bounds },
++ { 6, 0|4, __KERNEL_CS, (unsigned long)invalid_op },
++ { 7, 0|4, __KERNEL_CS, (unsigned long)device_not_available },
++ { 9, 0|4, __KERNEL_CS, (unsigned long)coprocessor_segment_overrun},
++ { 10, 0|4, __KERNEL_CS, (unsigned long)invalid_TSS },
++ { 11, 0|4, __KERNEL_CS, (unsigned long)segment_not_present },
++ { 12, 0|4, __KERNEL_CS, (unsigned long)stack_segment },
++ { 13, 0|4, __KERNEL_CS, (unsigned long)general_protection },
++ { 14, 0|4, __KERNEL_CS, (unsigned long)page_fault },
++ { 15, 0|4, __KERNEL_CS, (unsigned long)spurious_interrupt_bug },
++ { 16, 0|4, __KERNEL_CS, (unsigned long)coprocessor_error },
++ { 17, 0|4, __KERNEL_CS, (unsigned long)alignment_check },
++#ifdef CONFIG_X86_MCE
++ { 18, 0|4, __KERNEL_CS, (unsigned long)machine_check },
++#endif
++ { 19, 0|4, __KERNEL_CS, (unsigned long)simd_coprocessor_error },
++#ifdef CONFIG_IA32_EMULATION
++ { IA32_SYSCALL_VECTOR, 3|4, __KERNEL_CS, (unsigned long)ia32_syscall},
++#endif
++ { 0, 0, 0, 0 }
++};
++
++void __init trap_init(void)
++{
++ int ret;
++
++ ret = HYPERVISOR_set_trap_table(trap_table);
++
++ if (ret)
++ printk("HYPERVISOR_set_trap_table faild: error %d\n",
++ ret);
++
++ /*
++ * Should be a barrier for any external CPU state.
++ */
++ cpu_init();
++}
++
++void smp_trap_init(trap_info_t *trap_ctxt)
++{
++ trap_info_t *t = trap_table;
++
++ for (t = trap_table; t->address; t++) {
++ trap_ctxt[t->vector].flags = t->flags;
++ trap_ctxt[t->vector].cs = t->cs;
++ trap_ctxt[t->vector].address = t->address;
++ }
++}
++
++
++/* Actual parsing is done early in setup.c. */
++static int __init oops_dummy(char *s)
++{
++ panic_on_oops = 1;
++ return 1;
++}
++__setup("oops=", oops_dummy);
++
++static int __init kstack_setup(char *s)
++{
++ kstack_depth_to_print = simple_strtoul(s,NULL,0);
++ return 1;
++}
++__setup("kstack=", kstack_setup);
++
+diff -urNp linux-2.6/arch/x86_64/kernel/vsyscall-xen.c new/arch/x86_64/kernel/vsyscall-xen.c
+--- linux-2.6/arch/x86_64/kernel/vsyscall-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/vsyscall-xen.c 2006-05-09 12:33:21.000000000 +0200
+@@ -0,0 +1,239 @@
++/*
++ * linux/arch/x86_64/kernel/vsyscall.c
++ *
++ * Copyright (C) 2001 Andrea Arcangeli <andrea@suse.de> SuSE
++ * Copyright 2003 Andi Kleen, SuSE Labs.
++ *
++ * Thanks to hpa@transmeta.com for some useful hint.
++ * Special thanks to Ingo Molnar for his early experience with
++ * a different vsyscall implementation for Linux/IA32 and for the name.
++ *
++ * vsyscall 1 is located at -10Mbyte, vsyscall 2 is located
++ * at virtual address -10Mbyte+1024bytes etc... There are at max 4
++ * vsyscalls. One vsyscall can reserve more than 1 slot to avoid
++ * jumping out of line if necessary. We cannot add more with this
++ * mechanism because older kernels won't return -ENOSYS.
++ * If we want more than four we need a vDSO.
++ *
++ * Note: the concept clashes with user mode linux. If you use UML and
++ * want per guest time just set the kernel.vsyscall64 sysctl to 0.
++ */
++
++#include <linux/time.h>
++#include <linux/init.h>
++#include <linux/kernel.h>
++#include <linux/timer.h>
++#include <linux/seqlock.h>
++#include <linux/jiffies.h>
++#include <linux/sysctl.h>
++
++#include <asm/vsyscall.h>
++#include <asm/pgtable.h>
++#include <asm/page.h>
++#include <asm/fixmap.h>
++#include <asm/errno.h>
++#include <asm/io.h>
++
++#define __vsyscall(nr) __attribute__ ((unused,__section__(".vsyscall_" #nr)))
++
++int __sysctl_vsyscall __section_sysctl_vsyscall = 1;
++seqlock_t __xtime_lock __section_xtime_lock = SEQLOCK_UNLOCKED;
++
++#include <asm/unistd.h>
++
++static __always_inline void timeval_normalize(struct timeval * tv)
++{
++ time_t __sec;
++
++ __sec = tv->tv_usec / 1000000;
++ if (__sec) {
++ tv->tv_usec %= 1000000;
++ tv->tv_sec += __sec;
++ }
++}
++
++static __always_inline void do_vgettimeofday(struct timeval * tv)
++{
++ long sequence, t;
++ unsigned long sec, usec;
++
++ do {
++ sequence = read_seqbegin(&__xtime_lock);
++
++ sec = __xtime.tv_sec;
++ usec = (__xtime.tv_nsec / 1000) +
++ (__jiffies - __wall_jiffies) * (1000000 / HZ);
++
++ if (__vxtime.mode != VXTIME_HPET) {
++ t = get_cycles_sync();
++ if (t < __vxtime.last_tsc)
++ t = __vxtime.last_tsc;
++ usec += ((t - __vxtime.last_tsc) *
++ __vxtime.tsc_quot) >> 32;
++ /* See comment in x86_64 do_gettimeofday. */
++ } else {
++ usec += ((readl((void *)fix_to_virt(VSYSCALL_HPET) + 0xf0) -
++ __vxtime.last) * __vxtime.quot) >> 32;
++ }
++ } while (read_seqretry(&__xtime_lock, sequence));
++
++ tv->tv_sec = sec + usec / 1000000;
++ tv->tv_usec = usec % 1000000;
++}
++
++/* RED-PEN may want to readd seq locking, but then the variable should be write-once. */
++static __always_inline void do_get_tz(struct timezone * tz)
++{
++ *tz = __sys_tz;
++}
++
++static __always_inline int gettimeofday(struct timeval *tv, struct timezone *tz)
++{
++ int ret;
++ asm volatile("vsysc2: syscall"
++ : "=a" (ret)
++ : "0" (__NR_gettimeofday),"D" (tv),"S" (tz) : __syscall_clobber );
++ return ret;
++}
++
++static __always_inline long time_syscall(long *t)
++{
++ long secs;
++ asm volatile("vsysc1: syscall"
++ : "=a" (secs)
++ : "0" (__NR_time),"D" (t) : __syscall_clobber);
++ return secs;
++}
++
++int __vsyscall(0) vgettimeofday(struct timeval * tv, struct timezone * tz)
++{
++ if (unlikely(!__sysctl_vsyscall))
++ return gettimeofday(tv,tz);
++ if (tv)
++ do_vgettimeofday(tv);
++ if (tz)
++ do_get_tz(tz);
++ return 0;
++}
++
++/* This will break when the xtime seconds get inaccurate, but that is
++ * unlikely */
++time_t __vsyscall(1) vtime(time_t *t)
++{
++ if (unlikely(!__sysctl_vsyscall))
++ return time_syscall(t);
++ else if (t)
++ *t = __xtime.tv_sec;
++ return __xtime.tv_sec;
++}
++
++long __vsyscall(2) venosys_0(void)
++{
++ return -ENOSYS;
++}
++
++long __vsyscall(3) venosys_1(void)
++{
++ return -ENOSYS;
++}
++
++#ifdef CONFIG_SYSCTL
++
++#define SYSCALL 0x050f
++#define NOP2 0x9090
++
++/*
++ * NOP out syscall in vsyscall page when not needed.
++ */
++static int vsyscall_sysctl_change(ctl_table *ctl, int write, struct file * filp,
++ void __user *buffer, size_t *lenp, loff_t *ppos)
++{
++ extern u16 vsysc1, vsysc2;
++ u16 *map1, *map2;
++ int ret = proc_dointvec(ctl, write, filp, buffer, lenp, ppos);
++ if (!write)
++ return ret;
++ /* gcc has some trouble with __va(__pa()), so just do it this
++ way. */
++ map1 = ioremap(__pa_symbol(&vsysc1), 2);
++ if (!map1)
++ return -ENOMEM;
++ map2 = ioremap(__pa_symbol(&vsysc2), 2);
++ if (!map2) {
++ ret = -ENOMEM;
++ goto out;
++ }
++ if (!sysctl_vsyscall) {
++ *map1 = SYSCALL;
++ *map2 = SYSCALL;
++ } else {
++ *map1 = NOP2;
++ *map2 = NOP2;
++ }
++ iounmap(map2);
++out:
++ iounmap(map1);
++ return ret;
++}
++
++static int vsyscall_sysctl_nostrat(ctl_table *t, int __user *name, int nlen,
++ void __user *oldval, size_t __user *oldlenp,
++ void __user *newval, size_t newlen,
++ void **context)
++{
++ return -ENOSYS;
++}
++
++static ctl_table kernel_table2[] = {
++ { .ctl_name = 99, .procname = "vsyscall64",
++ .data = &sysctl_vsyscall, .maxlen = sizeof(int), .mode = 0644,
++ .strategy = vsyscall_sysctl_nostrat,
++ .proc_handler = vsyscall_sysctl_change },
++ { 0, }
++};
++
++static ctl_table kernel_root_table2[] = {
++ { .ctl_name = CTL_KERN, .procname = "kernel", .mode = 0555,
++ .child = kernel_table2 },
++ { 0 },
++};
++
++#endif
++
++static void __init map_vsyscall(void)
++{
++ extern char __vsyscall_0;
++ unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0);
++
++ __set_fixmap(VSYSCALL_FIRST_PAGE, physaddr_page0, PAGE_KERNEL_VSYSCALL);
++}
++
++#ifdef CONFIG_XEN
++static void __init map_vsyscall_user(void)
++{
++ extern void __set_fixmap_user(enum fixed_addresses, unsigned long, pgprot_t);
++ extern char __vsyscall_0;
++ unsigned long physaddr_page0 = __pa_symbol(&__vsyscall_0);
++
++ __set_fixmap_user(VSYSCALL_FIRST_PAGE, physaddr_page0, PAGE_KERNEL_VSYSCALL);
++}
++#endif
++
++static int __init vsyscall_init(void)
++{
++ BUG_ON(((unsigned long) &vgettimeofday !=
++ VSYSCALL_ADDR(__NR_vgettimeofday)));
++ BUG_ON((unsigned long) &vtime != VSYSCALL_ADDR(__NR_vtime));
++ BUG_ON((VSYSCALL_ADDR(0) != __fix_to_virt(VSYSCALL_FIRST_PAGE)));
++ map_vsyscall();
++#ifdef CONFIG_XEN
++ map_vsyscall_user();
++ sysctl_vsyscall = 0; /* disable vgettimeofay() */
++#endif
++#ifdef CONFIG_SYSCTL
++ register_sysctl_table(kernel_root_table2, 0);
++#endif
++ return 0;
++}
++
++__initcall(vsyscall_init);
+diff -urNp linux-2.6/arch/x86_64/kernel/x8664_ksyms-xen.c new/arch/x86_64/kernel/x8664_ksyms-xen.c
+--- linux-2.6/arch/x86_64/kernel/x8664_ksyms-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/x8664_ksyms-xen.c 2006-05-09 12:33:21.000000000 +0200
+@@ -0,0 +1,155 @@
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/smp.h>
++#include <linux/user.h>
++#include <linux/sched.h>
++#include <linux/in6.h>
++#include <linux/interrupt.h>
++#include <linux/smp_lock.h>
++#include <linux/pm.h>
++#include <linux/pci.h>
++#include <linux/apm_bios.h>
++#include <linux/kernel.h>
++#include <linux/string.h>
++#include <linux/syscalls.h>
++#include <linux/tty.h>
++
++#include <asm/semaphore.h>
++#include <asm/processor.h>
++#include <asm/i387.h>
++#include <asm/uaccess.h>
++#include <asm/checksum.h>
++#include <asm/io.h>
++#include <asm/delay.h>
++#include <asm/irq.h>
++#include <asm/mmx.h>
++#include <asm/desc.h>
++#include <asm/pgtable.h>
++#include <asm/pgalloc.h>
++#include <asm/nmi.h>
++#include <asm/kdebug.h>
++#include <asm/unistd.h>
++#include <asm/tlbflush.h>
++#include <asm/kdebug.h>
++
++#ifdef CONFIG_SMP
++extern void __write_lock_failed(rwlock_t *rw);
++extern void __read_lock_failed(rwlock_t *rw);
++#endif
++
++/* platform dependent support */
++EXPORT_SYMBOL(boot_cpu_data);
++//EXPORT_SYMBOL(dump_fpu);
++EXPORT_SYMBOL(kernel_thread);
++EXPORT_SYMBOL(pm_idle);
++EXPORT_SYMBOL(pm_power_off);
++
++EXPORT_SYMBOL(__down_failed);
++EXPORT_SYMBOL(__down_failed_interruptible);
++EXPORT_SYMBOL(__down_failed_trylock);
++EXPORT_SYMBOL(__up_wakeup);
++/* Networking helper routines. */
++EXPORT_SYMBOL(csum_partial_copy_nocheck);
++EXPORT_SYMBOL(ip_compute_csum);
++/* Delay loops */
++EXPORT_SYMBOL(__udelay);
++EXPORT_SYMBOL(__ndelay);
++EXPORT_SYMBOL(__delay);
++EXPORT_SYMBOL(__const_udelay);
++
++EXPORT_SYMBOL(__get_user_1);
++EXPORT_SYMBOL(__get_user_2);
++EXPORT_SYMBOL(__get_user_4);
++EXPORT_SYMBOL(__get_user_8);
++EXPORT_SYMBOL(__put_user_1);
++EXPORT_SYMBOL(__put_user_2);
++EXPORT_SYMBOL(__put_user_4);
++EXPORT_SYMBOL(__put_user_8);
++
++EXPORT_SYMBOL(strncpy_from_user);
++EXPORT_SYMBOL(__strncpy_from_user);
++EXPORT_SYMBOL(clear_user);
++EXPORT_SYMBOL(__clear_user);
++EXPORT_SYMBOL(copy_user_generic);
++EXPORT_SYMBOL(copy_from_user);
++EXPORT_SYMBOL(copy_to_user);
++EXPORT_SYMBOL(copy_in_user);
++EXPORT_SYMBOL(strnlen_user);
++
++#ifdef CONFIG_PCI
++EXPORT_SYMBOL(pci_mem_start);
++#endif
++
++EXPORT_SYMBOL(copy_page);
++EXPORT_SYMBOL(clear_page);
++
++EXPORT_SYMBOL(_cpu_pda);
++#ifdef CONFIG_SMP
++EXPORT_SYMBOL(__write_lock_failed);
++EXPORT_SYMBOL(__read_lock_failed);
++
++EXPORT_SYMBOL(smp_call_function);
++#endif
++
++#ifdef CONFIG_VT
++EXPORT_SYMBOL(screen_info);
++#endif
++
++#ifdef CONFIG_X86_LOCAL_APIC
++EXPORT_SYMBOL_GPL(set_nmi_callback);
++EXPORT_SYMBOL_GPL(unset_nmi_callback);
++#endif
++
++/* Export string functions. We normally rely on gcc builtin for most of these,
++ but gcc sometimes decides not to inline them. */
++#undef memcpy
++#undef memset
++#undef memmove
++
++extern void * memset(void *,int,__kernel_size_t);
++extern size_t strlen(const char *);
++extern void * memmove(void * dest,const void *src,size_t count);
++extern void * memcpy(void *,const void *,__kernel_size_t);
++extern void * __memcpy(void *,const void *,__kernel_size_t);
++
++EXPORT_SYMBOL(memset);
++EXPORT_SYMBOL(memmove);
++EXPORT_SYMBOL(memcpy);
++EXPORT_SYMBOL(__memcpy);
++
++#ifdef CONFIG_RWSEM_XCHGADD_ALGORITHM
++/* prototypes are wrong, these are assembly with custom calling functions */
++extern void rwsem_down_read_failed_thunk(void);
++extern void rwsem_wake_thunk(void);
++extern void rwsem_downgrade_thunk(void);
++extern void rwsem_down_write_failed_thunk(void);
++EXPORT_SYMBOL(rwsem_down_read_failed_thunk);
++EXPORT_SYMBOL(rwsem_wake_thunk);
++EXPORT_SYMBOL(rwsem_downgrade_thunk);
++EXPORT_SYMBOL(rwsem_down_write_failed_thunk);
++#endif
++
++EXPORT_SYMBOL(empty_zero_page);
++
++EXPORT_SYMBOL(die_chain);
++
++#ifdef CONFIG_SMP
++EXPORT_SYMBOL(cpu_sibling_map);
++EXPORT_SYMBOL(smp_num_siblings);
++#endif
++
++#ifdef CONFIG_BUG
++EXPORT_SYMBOL(out_of_line_bug);
++#endif
++
++EXPORT_SYMBOL(init_level4_pgt);
++
++extern unsigned long __supported_pte_mask;
++EXPORT_SYMBOL(__supported_pte_mask);
++
++#ifdef CONFIG_SMP
++EXPORT_SYMBOL(flush_tlb_page);
++#endif
++
++EXPORT_SYMBOL(load_gs_index);
++
+diff -urNp linux-2.6/arch/x86_64/kernel/xen_entry.S new/arch/x86_64/kernel/xen_entry.S
+--- linux-2.6/arch/x86_64/kernel/xen_entry.S 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/kernel/xen_entry.S 2006-05-09 12:33:21.000000000 +0200
+@@ -0,0 +1,40 @@
++/*
++ * Copied from arch/xen/i386/kernel/entry.S
++ */
++/* Offsets into shared_info_t. */
++#define evtchn_upcall_pending /* 0 */
++#define evtchn_upcall_mask 1
++
++#define sizeof_vcpu_shift 6
++
++#ifdef CONFIG_SMP
++//#define preempt_disable(reg) incl threadinfo_preempt_count(reg)
++//#define preempt_enable(reg) decl threadinfo_preempt_count(reg)
++#define preempt_disable(reg)
++#define preempt_enable(reg)
++#define XEN_GET_VCPU_INFO(reg) preempt_disable(%rbp) ; \
++ movq %gs:pda_cpunumber,reg ; \
++ shl $32, reg ; \
++ shr $32-sizeof_vcpu_shift,reg ; \
++ addq HYPERVISOR_shared_info,reg
++#define XEN_PUT_VCPU_INFO(reg) preempt_enable(%rbp) ; \
++#define XEN_PUT_VCPU_INFO_fixup .byte 0xff,0xff,0xff
++#else
++#define XEN_GET_VCPU_INFO(reg) movq HYPERVISOR_shared_info,reg
++#define XEN_PUT_VCPU_INFO(reg)
++#define XEN_PUT_VCPU_INFO_fixup
++#endif
++
++#define XEN_LOCKED_BLOCK_EVENTS(reg) movb $1,evtchn_upcall_mask(reg)
++#define XEN_LOCKED_UNBLOCK_EVENTS(reg) movb $0,evtchn_upcall_mask(reg)
++#define XEN_BLOCK_EVENTS(reg) XEN_GET_VCPU_INFO(reg) ; \
++ XEN_LOCKED_BLOCK_EVENTS(reg) ; \
++ XEN_PUT_VCPU_INFO(reg)
++#define XEN_UNBLOCK_EVENTS(reg) XEN_GET_VCPU_INFO(reg) ; \
++ XEN_LOCKED_UNBLOCK_EVENTS(reg) ; \
++ XEN_PUT_VCPU_INFO(reg)
++#define XEN_TEST_PENDING(reg) testb $0xFF,evtchn_upcall_pending(reg)
++
++VGCF_IN_SYSCALL = (1<<8)
++
++
+diff -urNp linux-2.6/arch/x86_64/Makefile new/arch/x86_64/Makefile
+--- linux-2.6/arch/x86_64/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/Makefile 2006-05-09 12:33:17.000000000 +0200
+@@ -31,6 +31,10 @@ cflags-$(CONFIG_MK8) += $(call cc-option
+ cflags-$(CONFIG_MPSC) += $(call cc-option,-march=nocona)
+ cflags-$(CONFIG_GENERIC_CPU) += $(call cc-option,-mtune=generic)
+
++cppflags-$(CONFIG_XEN) += \
++ -D__XEN_INTERFACE_VERSION__=$(CONFIG_XEN_INTERFACE_VERSION)
++CPPFLAGS += $(cppflags-y)
++
+ cflags-y += -m64
+ cflags-y += -mno-red-zone
+ cflags-y += -mcmodel=kernel
+@@ -72,6 +76,21 @@ boot := arch/x86_64/boot
+ PHONY += bzImage bzlilo install archmrproper \
+ fdimage fdimage144 fdimage288 isoimage archclean
+
++ifdef CONFIG_XEN
++CPPFLAGS := -Iinclude$(if $(KBUILD_SRC),2)/asm/mach-xen $(CPPFLAGS)
++head-y := arch/x86_64/kernel/head-xen.o arch/x86_64/kernel/head64-xen.o arch/x86_64/kernel/init_task.o
++LDFLAGS_vmlinux := -e _start
++boot := arch/i386/boot-xen
++.PHONY: vmlinuz
++#Default target when executing "make"
++all: vmlinuz
++
++vmlinuz: vmlinux
++ $(Q)$(MAKE) $(build)=$(boot) $@
++
++install:
++ $(Q)$(MAKE) $(build)=$(boot) XENGUEST=$(XENGUEST) $@
++else
+ #Default target when executing "make"
+ all: bzImage
+
+@@ -92,6 +111,7 @@ fdimage fdimage144 fdimage288 isoimage:
+
+ install:
+ $(Q)$(MAKE) $(build)=$(boot) BOOTIMAGE=$(BOOTIMAGE) $@
++endif
+
+ archclean:
+ $(Q)$(MAKE) $(clean)=$(boot)
+diff -urNp linux-2.6/arch/x86_64/mm/fault-xen.c new/arch/x86_64/mm/fault-xen.c
+--- linux-2.6/arch/x86_64/mm/fault-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/mm/fault-xen.c 2006-06-28 16:31:30.000000000 +0200
+@@ -0,0 +1,690 @@
++/*
++ * linux/arch/x86-64/mm/fault.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ * Copyright (C) 2001,2002 Andi Kleen, SuSE Labs.
++ */
++
++#include <linux/config.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/ptrace.h>
++#include <linux/mman.h>
++#include <linux/mm.h>
++#include <linux/smp.h>
++#include <linux/smp_lock.h>
++#include <linux/interrupt.h>
++#include <linux/init.h>
++#include <linux/tty.h>
++#include <linux/vt_kern.h> /* For unblank_screen() */
++#include <linux/compiler.h>
++#include <linux/module.h>
++#include <linux/kprobes.h>
++
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/pgalloc.h>
++#include <asm/smp.h>
++#include <asm/tlbflush.h>
++#include <asm/proto.h>
++#include <asm/kdebug.h>
++#include <asm-generic/sections.h>
++
++/* Page fault error code bits */
++#define PF_PROT (1<<0) /* or no page found */
++#define PF_WRITE (1<<1)
++#define PF_USER (1<<2)
++#define PF_RSVD (1<<3)
++#define PF_INSTR (1<<4)
++
++void bust_spinlocks(int yes)
++{
++ int loglevel_save = console_loglevel;
++ if (yes) {
++ oops_in_progress = 1;
++ } else {
++#ifdef CONFIG_VT
++ unblank_screen();
++#endif
++ oops_in_progress = 0;
++ /*
++ * OK, the message is on the console. Now we call printk()
++ * without oops_in_progress set so that printk will give klogd
++ * a poke. Hold onto your hats...
++ */
++ console_loglevel = 15; /* NMI oopser may have shut the console up */
++ printk(" ");
++ console_loglevel = loglevel_save;
++ }
++}
++
++/* Sometimes the CPU reports invalid exceptions on prefetch.
++ Check that here and ignore.
++ Opcode checker based on code by Richard Brunner */
++static noinline int is_prefetch(struct pt_regs *regs, unsigned long addr,
++ unsigned long error_code)
++{
++ unsigned char *instr;
++ int scan_more = 1;
++ int prefetch = 0;
++ unsigned char *max_instr;
++
++ /* If it was a exec fault ignore */
++ if (error_code & PF_INSTR)
++ return 0;
++
++ instr = (unsigned char *)convert_rip_to_linear(current, regs);
++ max_instr = instr + 15;
++
++ if (user_mode(regs) && instr >= (unsigned char *)TASK_SIZE)
++ return 0;
++
++ while (scan_more && instr < max_instr) {
++ unsigned char opcode;
++ unsigned char instr_hi;
++ unsigned char instr_lo;
++
++ if (__get_user(opcode, instr))
++ break;
++
++ instr_hi = opcode & 0xf0;
++ instr_lo = opcode & 0x0f;
++ instr++;
++
++ switch (instr_hi) {
++ case 0x20:
++ case 0x30:
++ /* Values 0x26,0x2E,0x36,0x3E are valid x86
++ prefixes. In long mode, the CPU will signal
++ invalid opcode if some of these prefixes are
++ present so we will never get here anyway */
++ scan_more = ((instr_lo & 7) == 0x6);
++ break;
++
++ case 0x40:
++ /* In AMD64 long mode, 0x40 to 0x4F are valid REX prefixes
++ Need to figure out under what instruction mode the
++ instruction was issued ... */
++ /* Could check the LDT for lm, but for now it's good
++ enough to assume that long mode only uses well known
++ segments or kernel. */
++ scan_more = (!user_mode(regs)) || (regs->cs == __USER_CS);
++ break;
++
++ case 0x60:
++ /* 0x64 thru 0x67 are valid prefixes in all modes. */
++ scan_more = (instr_lo & 0xC) == 0x4;
++ break;
++ case 0xF0:
++ /* 0xF0, 0xF2, and 0xF3 are valid prefixes in all modes. */
++ scan_more = !instr_lo || (instr_lo>>1) == 1;
++ break;
++ case 0x00:
++ /* Prefetch instruction is 0x0F0D or 0x0F18 */
++ scan_more = 0;
++ if (__get_user(opcode, instr))
++ break;
++ prefetch = (instr_lo == 0xF) &&
++ (opcode == 0x0D || opcode == 0x18);
++ break;
++ default:
++ scan_more = 0;
++ break;
++ }
++ }
++ return prefetch;
++}
++
++static int bad_address(void *p)
++{
++ unsigned long dummy;
++ return __get_user(dummy, (unsigned long *)p);
++}
++
++void dump_pagetable(unsigned long address)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++ pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK);
++ pgd += pgd_index(address);
++ if (bad_address(pgd)) goto bad;
++ printk("PGD %lx ", pgd_val(*pgd));
++ if (!pgd_present(*pgd)) goto ret;
++
++ pud = __pud_offset_k((pud_t *)pgd_page(*pgd), address);
++ if (bad_address(pud)) goto bad;
++ printk("PUD %lx ", pud_val(*pud));
++ if (!pud_present(*pud)) goto ret;
++
++ pmd = pmd_offset(pud, address);
++ if (bad_address(pmd)) goto bad;
++ printk("PMD %lx ", pmd_val(*pmd));
++ if (!pmd_present(*pmd)) goto ret;
++
++ pte = pte_offset_kernel(pmd, address);
++ if (bad_address(pte)) goto bad;
++ printk("PTE %lx", pte_val(*pte));
++ret:
++ printk("\n");
++ return;
++bad:
++ printk("BAD\n");
++}
++
++static const char errata93_warning[] =
++KERN_ERR "******* Your BIOS seems to not contain a fix for K8 errata #93\n"
++KERN_ERR "******* Working around it, but it may cause SEGVs or burn power.\n"
++KERN_ERR "******* Please consider a BIOS update.\n"
++KERN_ERR "******* Disabling USB legacy in the BIOS may also help.\n";
++
++/* Workaround for K8 erratum #93 & buggy BIOS.
++ BIOS SMM functions are required to use a specific workaround
++ to avoid corruption of the 64bit RIP register on C stepping K8.
++ A lot of BIOS that didn't get tested properly miss this.
++ The OS sees this as a page fault with the upper 32bits of RIP cleared.
++ Try to work around it here.
++ Note we only handle faults in kernel here. */
++
++static int is_errata93(struct pt_regs *regs, unsigned long address)
++{
++ static int warned;
++ if (address != regs->rip)
++ return 0;
++ if ((address >> 32) != 0)
++ return 0;
++ address |= 0xffffffffUL << 32;
++ if ((address >= (u64)_stext && address <= (u64)_etext) ||
++ (address >= MODULES_VADDR && address <= MODULES_END)) {
++ if (!warned) {
++ printk(errata93_warning);
++ warned = 1;
++ }
++ regs->rip = address;
++ return 1;
++ }
++ return 0;
++}
++
++int unhandled_signal(struct task_struct *tsk, int sig)
++{
++ if (tsk->pid == 1)
++ return 1;
++ if (tsk->ptrace & PT_PTRACED)
++ return 0;
++ return (tsk->sighand->action[sig-1].sa.sa_handler == SIG_IGN) ||
++ (tsk->sighand->action[sig-1].sa.sa_handler == SIG_DFL);
++}
++
++static noinline void pgtable_bad(unsigned long address, struct pt_regs *regs,
++ unsigned long error_code)
++{
++ unsigned long flags = oops_begin();
++ struct task_struct *tsk;
++
++ printk(KERN_ALERT "%s: Corrupted page table at address %lx\n",
++ current->comm, address);
++ dump_pagetable(address);
++ tsk = current;
++ tsk->thread.cr2 = address;
++ tsk->thread.trap_no = 14;
++ tsk->thread.error_code = error_code;
++ __die("Bad pagetable", regs, error_code);
++ oops_end(flags);
++ do_exit(SIGKILL);
++}
++
++/*
++ * Handle a fault on the vmalloc area
++ *
++ * This assumes no large pages in there.
++ */
++static int vmalloc_fault(unsigned long address)
++{
++ pgd_t *pgd, *pgd_ref;
++ pud_t *pud, *pud_ref;
++ pmd_t *pmd, *pmd_ref;
++ pte_t *pte, *pte_ref;
++
++ /* Copy kernel mappings over when needed. This can also
++ happen within a race in page table update. In the later
++ case just flush. */
++
++ /* On Xen the line below does not always work. Needs investigating! */
++ /*pgd = pgd_offset(current->mm ?: &init_mm, address);*/
++ pgd = __va(read_cr3() & PHYSICAL_PAGE_MASK);
++ pgd += pgd_index(address);
++ pgd_ref = pgd_offset_k(address);
++ if (pgd_none(*pgd_ref))
++ return -1;
++ if (pgd_none(*pgd))
++ set_pgd(pgd, *pgd_ref);
++ else
++ BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
++
++ /* Below here mismatches are bugs because these lower tables
++ are shared */
++
++ pud = pud_offset(pgd, address);
++ pud_ref = pud_offset(pgd_ref, address);
++ if (pud_none(*pud_ref))
++ return -1;
++ if (pud_none(*pud) || pud_page(*pud) != pud_page(*pud_ref))
++ BUG();
++ pmd = pmd_offset(pud, address);
++ pmd_ref = pmd_offset(pud_ref, address);
++ if (pmd_none(*pmd_ref))
++ return -1;
++ if (pmd_none(*pmd) || pmd_page(*pmd) != pmd_page(*pmd_ref))
++ BUG();
++ pte_ref = pte_offset_kernel(pmd_ref, address);
++ if (!pte_present(*pte_ref))
++ return -1;
++ pte = pte_offset_kernel(pmd, address);
++ /* Don't use pte_page here, because the mappings can point
++ outside mem_map, and the NUMA hash lookup cannot handle
++ that. */
++ if (!pte_present(*pte) || pte_pfn(*pte) != pte_pfn(*pte_ref))
++ BUG();
++ return 0;
++}
++
++int page_fault_trace = 0;
++int exception_trace = 1;
++
++
++#define MEM_VERBOSE 1
++
++#ifdef MEM_VERBOSE
++#define MEM_LOG(_f, _a...) \
++ printk("fault.c:[%d]-> " _f "\n", \
++ __LINE__ , ## _a )
++#else
++#define MEM_LOG(_f, _a...) ((void)0)
++#endif
++
++static int spurious_fault(struct pt_regs *regs,
++ unsigned long address,
++ unsigned long error_code)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++#ifdef CONFIG_XEN
++ /* Faults in hypervisor area are never spurious. */
++ if ((address >= HYPERVISOR_VIRT_START) &&
++ (address < HYPERVISOR_VIRT_END))
++ return 0;
++#endif
++
++ /* Reserved-bit violation or user access to kernel space? */
++ if (error_code & (PF_RSVD|PF_USER))
++ return 0;
++
++ pgd = init_mm.pgd + pgd_index(address);
++ if (!pgd_present(*pgd))
++ return 0;
++
++ pud = pud_offset(pgd, address);
++ if (!pud_present(*pud))
++ return 0;
++
++ pmd = pmd_offset(pud, address);
++ if (!pmd_present(*pmd))
++ return 0;
++
++ pte = pte_offset_kernel(pmd, address);
++ if (!pte_present(*pte))
++ return 0;
++ if ((error_code & PF_WRITE) && !pte_write(*pte))
++ return 0;
++ if ((error_code & PF_INSTR) && (pte_val(*pte) & _PAGE_NX))
++ return 0;
++
++ return 1;
++}
++
++/*
++ * This routine handles page faults. It determines the address,
++ * and the problem, and then passes it off to one of the appropriate
++ * routines.
++ */
++asmlinkage void __kprobes do_page_fault(struct pt_regs *regs,
++ unsigned long error_code)
++{
++ struct task_struct *tsk;
++ struct mm_struct *mm;
++ struct vm_area_struct * vma;
++ unsigned long address;
++ const struct exception_table_entry *fixup;
++ int write;
++ unsigned long flags;
++ siginfo_t info;
++
++ if (!user_mode(regs))
++ error_code &= ~PF_USER; /* means kernel */
++
++ tsk = current;
++ mm = tsk->mm;
++ prefetchw(&mm->mmap_sem);
++
++ /* get the address */
++ address = HYPERVISOR_shared_info->vcpu_info[
++ smp_processor_id()].arch.cr2;
++
++ info.si_code = SEGV_MAPERR;
++
++
++ /*
++ * We fault-in kernel-space virtual memory on-demand. The
++ * 'reference' page table is init_mm.pgd.
++ *
++ * NOTE! We MUST NOT take any locks for this case. We may
++ * be in an interrupt or a critical region, and should
++ * only copy the information from the master page table,
++ * nothing more.
++ *
++ * This verifies that the fault happens in kernel space
++ * (error_code & 4) == 0, and that the fault was not a
++ * protection error (error_code & 9) == 0.
++ */
++ if (unlikely(address >= TASK_SIZE64)) {
++ /*
++ * Don't check for the module range here: its PML4
++ * is always initialized because it's shared with the main
++ * kernel text. Only vmalloc may need PML4 syncups.
++ */
++ if (!(error_code & (PF_RSVD|PF_USER|PF_PROT)) &&
++ address >= PAGE_OFFSET) {
++ if (vmalloc_fault(address) >= 0)
++ return;
++ }
++ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
++ SIGSEGV) == NOTIFY_STOP)
++ return;
++ /* Can take a spurious fault if mapping changes R/O -> R/W. */
++ if (spurious_fault(regs, address, error_code))
++ return;
++ /*
++ * Don't take the mm semaphore here. If we fixup a prefetch
++ * fault we could otherwise deadlock.
++ */
++ goto bad_area_nosemaphore;
++ }
++
++ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
++ SIGSEGV) == NOTIFY_STOP)
++ return;
++
++ if (likely(regs->eflags & X86_EFLAGS_IF))
++ local_irq_enable();
++
++ if (unlikely(page_fault_trace))
++ printk("pagefault rip:%lx rsp:%lx cs:%lu ss:%lu address %lx error %lx\n",
++ regs->rip,regs->rsp,regs->cs,regs->ss,address,error_code);
++
++ if (unlikely(error_code & PF_RSVD))
++ pgtable_bad(address, regs, error_code);
++
++ /*
++ * If we're in an interrupt or have no user
++ * context, we must not take the fault..
++ */
++ if (unlikely(in_atomic() || !mm))
++ goto bad_area_nosemaphore;
++
++ again:
++ /* When running in the kernel we expect faults to occur only to
++ * addresses in user space. All other faults represent errors in the
++ * kernel and should generate an OOPS. Unfortunatly, in the case of an
++ * erroneous fault occuring in a code path which already holds mmap_sem
++ * we will deadlock attempting to validate the fault against the
++ * address space. Luckily the kernel only validly references user
++ * space from well defined areas of code, which are listed in the
++ * exceptions table.
++ *
++ * As the vast majority of faults will be valid we will only perform
++ * the source reference check when there is a possibilty of a deadlock.
++ * Attempt to lock the address space, if we cannot we then validate the
++ * source. If this is invalid we can skip the address space check,
++ * thus avoiding the deadlock.
++ */
++ if (!down_read_trylock(&mm->mmap_sem)) {
++ if ((error_code & PF_USER) == 0 &&
++ !search_exception_tables(regs->rip))
++ goto bad_area_nosemaphore;
++ down_read(&mm->mmap_sem);
++ }
++
++ vma = find_vma(mm, address);
++ if (!vma)
++ goto bad_area;
++ if (likely(vma->vm_start <= address))
++ goto good_area;
++ if (!(vma->vm_flags & VM_GROWSDOWN))
++ goto bad_area;
++ if (error_code & 4) {
++ // XXX: align red zone size with ABI
++ if (address + 128 < regs->rsp)
++ goto bad_area;
++ }
++ if (expand_stack(vma, address))
++ goto bad_area;
++/*
++ * Ok, we have a good vm_area for this memory access, so
++ * we can handle it..
++ */
++good_area:
++ info.si_code = SEGV_ACCERR;
++ write = 0;
++ switch (error_code & (PF_PROT|PF_WRITE)) {
++ default: /* 3: write, present */
++ /* fall through */
++ case PF_WRITE: /* write, not present */
++ if (!(vma->vm_flags & VM_WRITE))
++ goto bad_area;
++ write++;
++ break;
++ case PF_PROT: /* read, present */
++ goto bad_area;
++ case 0: /* read, not present */
++ if (!(vma->vm_flags & (VM_READ | VM_EXEC)))
++ goto bad_area;
++ }
++
++ /*
++ * If for any reason at all we couldn't handle the fault,
++ * make sure we exit gracefully rather than endlessly redo
++ * the fault.
++ */
++ switch (handle_mm_fault(mm, vma, address, write)) {
++ case VM_FAULT_MINOR:
++ tsk->min_flt++;
++ break;
++ case VM_FAULT_MAJOR:
++ tsk->maj_flt++;
++ break;
++ case VM_FAULT_SIGBUS:
++ goto do_sigbus;
++ default:
++ goto out_of_memory;
++ }
++
++ up_read(&mm->mmap_sem);
++ return;
++
++/*
++ * Something tried to access memory that isn't in our memory map..
++ * Fix it, but check if it's kernel or user first..
++ */
++bad_area:
++ up_read(&mm->mmap_sem);
++
++bad_area_nosemaphore:
++ /* User mode accesses just cause a SIGSEGV */
++ if (error_code & PF_USER) {
++ if (is_prefetch(regs, address, error_code))
++ return;
++
++ /* Work around K8 erratum #100 K8 in compat mode
++ occasionally jumps to illegal addresses >4GB. We
++ catch this here in the page fault handler because
++ these addresses are not reachable. Just detect this
++ case and return. Any code segment in LDT is
++ compatibility mode. */
++ if ((regs->cs == __USER32_CS || (regs->cs & (1<<2))) &&
++ (address >> 32))
++ return;
++
++ if (exception_trace && unhandled_signal(tsk, SIGSEGV)) {
++ printk(
++ "%s%s[%d]: segfault at %016lx rip %016lx rsp %016lx error %lx\n",
++ tsk->pid > 1 ? KERN_INFO : KERN_EMERG,
++ tsk->comm, tsk->pid, address, regs->rip,
++ regs->rsp, error_code);
++ }
++
++ tsk->thread.cr2 = address;
++ /* Kernel addresses are always protection faults */
++ tsk->thread.error_code = error_code | (address >= TASK_SIZE);
++ tsk->thread.trap_no = 14;
++ info.si_signo = SIGSEGV;
++ info.si_errno = 0;
++ /* info.si_code has been set above */
++ info.si_addr = (void __user *)address;
++ force_sig_info(SIGSEGV, &info, tsk);
++ return;
++ }
++
++no_context:
++
++ /* Are we prepared to handle this kernel fault? */
++ fixup = search_exception_tables(regs->rip);
++ if (fixup) {
++ regs->rip = fixup->fixup;
++ return;
++ }
++
++ /*
++ * Hall of shame of CPU/BIOS bugs.
++ */
++
++ if (is_prefetch(regs, address, error_code))
++ return;
++
++ if (is_errata93(regs, address))
++ return;
++
++/*
++ * Oops. The kernel tried to access some bad page. We'll have to
++ * terminate things with extreme prejudice.
++ */
++
++ flags = oops_begin();
++
++ if (address < PAGE_SIZE)
++ printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
++ else
++ printk(KERN_ALERT "Unable to handle kernel paging request");
++ printk(" at %016lx RIP: \n" KERN_ALERT,address);
++ printk_address(regs->rip);
++ printk("\n");
++ dump_pagetable(address);
++ tsk->thread.cr2 = address;
++ tsk->thread.trap_no = 14;
++ tsk->thread.error_code = error_code;
++ __die("Oops", regs, error_code);
++ /* Executive summary in case the body of the oops scrolled away */
++ printk(KERN_EMERG "CR2: %016lx\n", address);
++ oops_end(flags);
++ do_exit(SIGKILL);
++
++/*
++ * We ran out of memory, or some other thing happened to us that made
++ * us unable to handle the page fault gracefully.
++ */
++out_of_memory:
++ up_read(&mm->mmap_sem);
++ if (current->pid == 1) {
++ yield();
++ goto again;
++ }
++ printk("VM: killing process %s\n", tsk->comm);
++ if (error_code & 4)
++ do_exit(SIGKILL);
++ goto no_context;
++
++do_sigbus:
++ up_read(&mm->mmap_sem);
++
++ /* Kernel mode? Handle exceptions or die */
++ if (!(error_code & PF_USER))
++ goto no_context;
++
++ tsk->thread.cr2 = address;
++ tsk->thread.error_code = error_code;
++ tsk->thread.trap_no = 14;
++ info.si_signo = SIGBUS;
++ info.si_errno = 0;
++ info.si_code = BUS_ADRERR;
++ info.si_addr = (void __user *)address;
++ force_sig_info(SIGBUS, &info, tsk);
++ return;
++}
++
++DEFINE_SPINLOCK(pgd_lock);
++struct page *pgd_list;
++
++void vmalloc_sync_all(void)
++{
++ /* Note that races in the updates of insync and start aren't
++ problematic:
++ insync can only get set bits added, and updates to start are only
++ improving performance (without affecting correctness if undone). */
++ static DECLARE_BITMAP(insync, PTRS_PER_PGD);
++ static unsigned long start = VMALLOC_START & PGDIR_MASK;
++ unsigned long address;
++
++ for (address = start; address <= VMALLOC_END; address += PGDIR_SIZE) {
++ if (!test_bit(pgd_index(address), insync)) {
++ const pgd_t *pgd_ref = pgd_offset_k(address);
++ struct page *page;
++
++ if (pgd_none(*pgd_ref))
++ continue;
++ spin_lock(&pgd_lock);
++ for (page = pgd_list; page;
++ page = (struct page *)page->index) {
++ pgd_t *pgd;
++ pgd = (pgd_t *)page_address(page) + pgd_index(address);
++ if (pgd_none(*pgd))
++ set_pgd(pgd, *pgd_ref);
++ else
++ BUG_ON(pgd_page(*pgd) != pgd_page(*pgd_ref));
++ }
++ spin_unlock(&pgd_lock);
++ set_bit(pgd_index(address), insync);
++ }
++ if (address == start)
++ start = address + PGDIR_SIZE;
++ }
++ /* Check that there is no need to do the same for the modules area. */
++ BUILD_BUG_ON(!(MODULES_VADDR > __START_KERNEL));
++ BUILD_BUG_ON(!(((MODULES_END - 1) & PGDIR_MASK) ==
++ (__START_KERNEL & PGDIR_MASK)));
++}
++
++static int __init enable_pagefaulttrace(char *str)
++{
++ page_fault_trace = 1;
++ return 1;
++}
++__setup("pagefaulttrace", enable_pagefaulttrace);
+diff -urNp linux-2.6/arch/x86_64/mm/init-xen.c new/arch/x86_64/mm/init-xen.c
+--- linux-2.6/arch/x86_64/mm/init-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/mm/init-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,1202 @@
++/*
++ * linux/arch/x86_64/mm/init.c
++ *
++ * Copyright (C) 1995 Linus Torvalds
++ * Copyright (C) 2000 Pavel Machek <pavel@suse.cz>
++ * Copyright (C) 2002,2003 Andi Kleen <ak@suse.de>
++ *
++ * Jun Nakajima <jun.nakajima@intel.com>
++ * Modified for Xen.
++ */
++
++#include <linux/config.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/ptrace.h>
++#include <linux/mman.h>
++#include <linux/mm.h>
++#include <linux/swap.h>
++#include <linux/smp.h>
++#include <linux/init.h>
++#include <linux/pagemap.h>
++#include <linux/bootmem.h>
++#include <linux/proc_fs.h>
++#include <linux/pci.h>
++#include <linux/dma-mapping.h>
++#include <linux/module.h>
++#include <linux/memory_hotplug.h>
++
++#include <asm/processor.h>
++#include <asm/system.h>
++#include <asm/uaccess.h>
++#include <asm/pgtable.h>
++#include <asm/pgalloc.h>
++#include <asm/dma.h>
++#include <asm/fixmap.h>
++#include <asm/e820.h>
++#include <asm/apic.h>
++#include <asm/tlb.h>
++#include <asm/mmu_context.h>
++#include <asm/proto.h>
++#include <asm/smp.h>
++#include <asm/sections.h>
++#include <asm/dma-mapping.h>
++#include <asm/swiotlb.h>
++
++#include <xen/features.h>
++
++#ifndef Dprintk
++#define Dprintk(x...)
++#endif
++
++struct dma_mapping_ops* dma_ops;
++EXPORT_SYMBOL(dma_ops);
++
++extern unsigned long *contiguous_bitmap;
++
++static unsigned long dma_reserve __initdata;
++
++DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
++extern unsigned long start_pfn;
++
++/*
++ * Use this until direct mapping is established, i.e. before __va() is
++ * available in init_memory_mapping().
++ */
++
++#define addr_to_page(addr, page) \
++ (addr) &= PHYSICAL_PAGE_MASK; \
++ (page) = ((unsigned long *) ((unsigned long) \
++ (((mfn_to_pfn((addr) >> PAGE_SHIFT)) << PAGE_SHIFT) + \
++ __START_KERNEL_map)))
++
++static void early_make_page_readonly(void *va, unsigned int feature)
++{
++ unsigned long addr, _va = (unsigned long)va;
++ pte_t pte, *ptep;
++ unsigned long *page = (unsigned long *) init_level4_pgt;
++
++ if (xen_feature(feature))
++ return;
++
++ addr = (unsigned long) page[pgd_index(_va)];
++ addr_to_page(addr, page);
++
++ addr = page[pud_index(_va)];
++ addr_to_page(addr, page);
++
++ addr = page[pmd_index(_va)];
++ addr_to_page(addr, page);
++
++ ptep = (pte_t *) &page[pte_index(_va)];
++
++ pte.pte = ptep->pte & ~_PAGE_RW;
++ if (HYPERVISOR_update_va_mapping(_va, pte, 0))
++ BUG();
++}
++
++void make_page_readonly(void *va, unsigned int feature)
++{
++ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t pte, *ptep;
++ unsigned long addr = (unsigned long) va;
++
++ if (xen_feature(feature))
++ return;
++
++ pgd = pgd_offset_k(addr);
++ pud = pud_offset(pgd, addr);
++ pmd = pmd_offset(pud, addr);
++ ptep = pte_offset_kernel(pmd, addr);
++
++ pte.pte = ptep->pte & ~_PAGE_RW;
++ if (HYPERVISOR_update_va_mapping(addr, pte, 0))
++ xen_l1_entry_update(ptep, pte); /* fallback */
++
++ if ((addr >= VMALLOC_START) && (addr < VMALLOC_END))
++ make_page_readonly(__va(pte_pfn(pte) << PAGE_SHIFT), feature);
++}
++
++void make_page_writable(void *va, unsigned int feature)
++{
++ pgd_t *pgd; pud_t *pud; pmd_t *pmd; pte_t pte, *ptep;
++ unsigned long addr = (unsigned long) va;
++
++ if (xen_feature(feature))
++ return;
++
++ pgd = pgd_offset_k(addr);
++ pud = pud_offset(pgd, addr);
++ pmd = pmd_offset(pud, addr);
++ ptep = pte_offset_kernel(pmd, addr);
++
++ pte.pte = ptep->pte | _PAGE_RW;
++ if (HYPERVISOR_update_va_mapping(addr, pte, 0))
++ xen_l1_entry_update(ptep, pte); /* fallback */
++
++ if ((addr >= VMALLOC_START) && (addr < VMALLOC_END))
++ make_page_writable(__va(pte_pfn(pte) << PAGE_SHIFT), feature);
++}
++
++void make_pages_readonly(void *va, unsigned nr, unsigned int feature)
++{
++ if (xen_feature(feature))
++ return;
++
++ while (nr-- != 0) {
++ make_page_readonly(va, feature);
++ va = (void*)((unsigned long)va + PAGE_SIZE);
++ }
++}
++
++void make_pages_writable(void *va, unsigned nr, unsigned int feature)
++{
++ if (xen_feature(feature))
++ return;
++
++ while (nr-- != 0) {
++ make_page_writable(va, feature);
++ va = (void*)((unsigned long)va + PAGE_SIZE);
++ }
++}
++
++/*
++ * NOTE: pagetable_init alloc all the fixmap pagetables contiguous on the
++ * physical space so we can cache the place of the first one and move
++ * around without checking the pgd every time.
++ */
++
++void show_mem(void)
++{
++ long i, total = 0, reserved = 0;
++ long shared = 0, cached = 0;
++ pg_data_t *pgdat;
++ struct page *page;
++
++ printk(KERN_INFO "Mem-info:\n");
++ show_free_areas();
++ printk(KERN_INFO "Free swap: %6ldkB\n", nr_swap_pages<<(PAGE_SHIFT-10));
++
++ for_each_online_pgdat(pgdat) {
++ for (i = 0; i < pgdat->node_spanned_pages; ++i) {
++ page = pfn_to_page(pgdat->node_start_pfn + i);
++ total++;
++ if (PageReserved(page))
++ reserved++;
++ else if (PageSwapCache(page))
++ cached++;
++ else if (page_count(page))
++ shared += page_count(page) - 1;
++ }
++ }
++ printk(KERN_INFO "%lu pages of RAM\n", total);
++ printk(KERN_INFO "%lu reserved pages\n",reserved);
++ printk(KERN_INFO "%lu pages shared\n",shared);
++ printk(KERN_INFO "%lu pages swap cached\n",cached);
++}
++
++/* References to section boundaries */
++
++int after_bootmem;
++
++static __init void *spp_getpage(void)
++{
++ void *ptr;
++ if (after_bootmem)
++ ptr = (void *) get_zeroed_page(GFP_ATOMIC);
++ else
++ ptr = alloc_bootmem_pages(PAGE_SIZE);
++ if (!ptr || ((unsigned long)ptr & ~PAGE_MASK))
++ panic("set_pte_phys: cannot allocate page data %s\n", after_bootmem?"after bootmem":"");
++
++ Dprintk("spp_getpage %p\n", ptr);
++ return ptr;
++}
++
++#define pgd_offset_u(address) (pgd_t *)(init_level4_user_pgt + pgd_index(address))
++
++static inline pud_t *pud_offset_u(unsigned long address)
++{
++ pud_t *pud = level3_user_pgt;
++
++ return pud + pud_index(address);
++}
++
++static __init void set_pte_phys(unsigned long vaddr,
++ unsigned long phys, pgprot_t prot, int user_mode)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte, new_pte;
++
++ Dprintk("set_pte_phys %lx to %lx\n", vaddr, phys);
++
++ pgd = (user_mode ? pgd_offset_u(vaddr) : pgd_offset_k(vaddr));
++ if (pgd_none(*pgd)) {
++ printk("PGD FIXMAP MISSING, it should be setup in head.S!\n");
++ return;
++ }
++ pud = (user_mode ? pud_offset_u(vaddr) : pud_offset(pgd, vaddr));
++ if (pud_none(*pud)) {
++ pmd = (pmd_t *) spp_getpage();
++ make_page_readonly(pmd, XENFEAT_writable_page_tables);
++ set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
++ if (pmd != pmd_offset(pud, 0)) {
++ printk("PAGETABLE BUG #01! %p <-> %p\n", pmd, pmd_offset(pud,0));
++ return;
++ }
++ }
++ pmd = pmd_offset(pud, vaddr);
++ if (pmd_none(*pmd)) {
++ pte = (pte_t *) spp_getpage();
++ make_page_readonly(pte, XENFEAT_writable_page_tables);
++ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
++ if (pte != pte_offset_kernel(pmd, 0)) {
++ printk("PAGETABLE BUG #02!\n");
++ return;
++ }
++ }
++ new_pte = pfn_pte(phys >> PAGE_SHIFT, prot);
++
++ pte = pte_offset_kernel(pmd, vaddr);
++ if (!pte_none(*pte) &&
++ pte_val(*pte) != (pte_val(new_pte) & __supported_pte_mask))
++ pte_ERROR(*pte);
++ set_pte(pte, new_pte);
++
++ /*
++ * It's enough to flush this one mapping.
++ * (PGE mappings get flushed as well)
++ */
++ __flush_tlb_one(vaddr);
++}
++
++static void set_pte_phys_ma(unsigned long vaddr,
++ unsigned long phys, pgprot_t prot)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte, new_pte;
++
++ Dprintk("set_pte_phys %lx to %lx\n", vaddr, phys);
++
++ pgd = pgd_offset_k(vaddr);
++ if (pgd_none(*pgd)) {
++ printk("PGD FIXMAP MISSING, it should be setup in head.S!\n");
++ return;
++ }
++ pud = pud_offset(pgd, vaddr);
++ if (pud_none(*pud)) {
++
++ pmd = (pmd_t *) spp_getpage();
++ make_page_readonly(pmd, XENFEAT_writable_page_tables);
++
++ set_pud(pud, __pud(__pa(pmd) | _KERNPG_TABLE | _PAGE_USER));
++
++ if (pmd != pmd_offset(pud, 0)) {
++ printk("PAGETABLE BUG #01! %p <-> %p\n", pmd, pmd_offset(pud,0));
++ return;
++ }
++ }
++ pmd = pmd_offset(pud, vaddr);
++
++ if (pmd_none(*pmd)) {
++ pte = (pte_t *) spp_getpage();
++ make_page_readonly(pte, XENFEAT_writable_page_tables);
++
++ set_pmd(pmd, __pmd(__pa(pte) | _KERNPG_TABLE | _PAGE_USER));
++ if (pte != pte_offset_kernel(pmd, 0)) {
++ printk("PAGETABLE BUG #02!\n");
++ return;
++ }
++ }
++
++ new_pte = pfn_pte_ma(phys >> PAGE_SHIFT, prot);
++ pte = pte_offset_kernel(pmd, vaddr);
++
++ /*
++ * Note that the pte page is already RO, thus we want to use
++ * xen_l1_entry_update(), not set_pte().
++ */
++ xen_l1_entry_update(pte,
++ pfn_pte_ma(phys >> PAGE_SHIFT, prot));
++
++ /*
++ * It's enough to flush this one mapping.
++ * (PGE mappings get flushed as well)
++ */
++ __flush_tlb_one(vaddr);
++}
++
++#define SET_FIXMAP_KERNEL 0
++#define SET_FIXMAP_USER 1
++
++/* NOTE: this is meant to be run only at boot */
++void __init
++__set_fixmap (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
++{
++ unsigned long address = __fix_to_virt(idx);
++
++ if (idx >= __end_of_fixed_addresses) {
++ printk("Invalid __set_fixmap\n");
++ return;
++ }
++ switch (idx) {
++ case VSYSCALL_FIRST_PAGE:
++ set_pte_phys(address, phys, prot, SET_FIXMAP_KERNEL);
++ break;
++ default:
++ set_pte_phys_ma(address, phys, prot);
++ break;
++ }
++}
++
++/*
++ * At this point it only supports vsyscall area.
++ */
++void __set_fixmap_user (enum fixed_addresses idx, unsigned long phys, pgprot_t prot)
++{
++ unsigned long address = __fix_to_virt(idx);
++
++ if (idx >= __end_of_fixed_addresses) {
++ printk("Invalid __set_fixmap\n");
++ return;
++ }
++
++ set_pte_phys(address, phys, prot, SET_FIXMAP_USER);
++}
++
++unsigned long __initdata table_start, table_end;
++
++#ifndef CONFIG_XEN
++extern pmd_t temp_boot_pmds[];
++
++static struct temp_map {
++ pmd_t *pmd;
++ void *address;
++ int allocated;
++} temp_mappings[] __initdata = {
++ { &temp_boot_pmds[0], (void *)(40UL * 1024 * 1024) },
++ { &temp_boot_pmds[1], (void *)(42UL * 1024 * 1024) },
++ {}
++};
++#endif /* !CONFIG_XEN */
++
++unsigned long get_machine_pfn(unsigned long addr)
++{
++ pud_t* pud = pud_offset_k(NULL, addr);
++ pmd_t* pmd = pmd_offset(pud, addr);
++ pte_t *pte = pte_offset_kernel(pmd, addr);
++
++ return pte_mfn(*pte);
++}
++
++static __meminit void *alloc_static_page(unsigned long *phys)
++{
++ unsigned long va = (start_pfn << PAGE_SHIFT) + __START_KERNEL_map;
++
++ if (after_bootmem) {
++ void *adr = (void *)get_zeroed_page(GFP_ATOMIC);
++
++ *phys = __pa(adr);
++ return adr;
++ }
++
++ *phys = start_pfn << PAGE_SHIFT;
++ start_pfn++;
++ memset((void *)va, 0, PAGE_SIZE);
++ return (void *)va;
++}
++
++#define PTE_SIZE PAGE_SIZE
++
++static inline void __set_pte(pte_t *dst, pte_t val)
++{
++ *dst = val;
++}
++
++static inline int make_readonly(unsigned long paddr)
++{
++ int readonly = 0;
++
++ /* Make new page tables read-only. */
++ if (!xen_feature(XENFEAT_writable_page_tables)
++ && (paddr >= (table_start << PAGE_SHIFT))
++ && (paddr < (table_end << PAGE_SHIFT)))
++ readonly = 1;
++ /* Make old page tables read-only. */
++ if (!xen_feature(XENFEAT_writable_page_tables)
++ && (paddr >= (xen_start_info->pt_base - __START_KERNEL_map))
++ && (paddr < (start_pfn << PAGE_SHIFT)))
++ readonly = 1;
++
++ /*
++ * No need for writable mapping of kernel image. This also ensures that
++ * page and descriptor tables embedded inside don't have writable
++ * mappings.
++ */
++ if ((paddr >= __pa_symbol(&_text)) && (paddr < __pa_symbol(&_end)))
++ readonly = 1;
++
++ return readonly;
++}
++
++#ifndef CONFIG_XEN
++/* Must run before zap_low_mappings */
++__init void *early_ioremap(unsigned long addr, unsigned long size)
++{
++ unsigned long map = round_down(addr, LARGE_PAGE_SIZE);
++
++ /* actually usually some more */
++ if (size >= LARGE_PAGE_SIZE) {
++ printk("SMBIOS area too long %lu\n", size);
++ return NULL;
++ }
++ set_pmd(temp_mappings[0].pmd, __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
++ map += LARGE_PAGE_SIZE;
++ set_pmd(temp_mappings[1].pmd, __pmd(map | _KERNPG_TABLE | _PAGE_PSE));
++ __flush_tlb();
++ return temp_mappings[0].address + (addr & (LARGE_PAGE_SIZE-1));
++}
++
++/* To avoid virtual aliases later */
++__init void early_iounmap(void *addr, unsigned long size)
++{
++ if ((void *)round_down((unsigned long)addr, LARGE_PAGE_SIZE) != temp_mappings[0].address)
++ printk("early_iounmap: bad address %p\n", addr);
++ set_pmd(temp_mappings[0].pmd, __pmd(0));
++ set_pmd(temp_mappings[1].pmd, __pmd(0));
++ __flush_tlb();
++}
++#endif /* !CONFIG_XEN */
++
++static void __meminit
++phys_pmd_init(pmd_t *pmd, unsigned long address, unsigned long end)
++{
++ int i, k;
++
++ for (i = 0; i < PTRS_PER_PMD; pmd++, i++) {
++ unsigned long pte_phys;
++ pte_t *pte, *pte_save;
++
++ if (address >= end) {
++ for (; i < PTRS_PER_PMD; i++, pmd++)
++ set_pmd(pmd, __pmd(0));
++ break;
++ }
++ pte = alloc_static_page(&pte_phys);
++ pte_save = pte;
++ for (k = 0; k < PTRS_PER_PTE; pte++, k++, address += PTE_SIZE) {
++ if ((address >= end) ||
++ ((address >> PAGE_SHIFT) >=
++ xen_start_info->nr_pages)) {
++ __set_pte(pte, __pte(0));
++ continue;
++ }
++ if (make_readonly(address)) {
++ __set_pte(pte,
++ __pte(address | (_KERNPG_TABLE & ~_PAGE_RW)));
++ continue;
++ }
++ __set_pte(pte, __pte(address | _KERNPG_TABLE));
++ }
++ pte = pte_save;
++ early_make_page_readonly(pte, XENFEAT_writable_page_tables);
++ set_pmd(pmd, __pmd(pte_phys | _KERNPG_TABLE));
++ }
++}
++
++static void __meminit
++phys_pmd_update(pud_t *pud, unsigned long address, unsigned long end)
++{
++ pmd_t *pmd = pmd_offset(pud, (unsigned long)__va(address));
++
++ if (pmd_none(*pmd)) {
++ spin_lock(&init_mm.page_table_lock);
++ phys_pmd_init(pmd, address, end);
++ spin_unlock(&init_mm.page_table_lock);
++ __flush_tlb_all();
++ }
++}
++
++static void __meminit phys_pud_init(pud_t *pud, unsigned long address, unsigned long end)
++{
++ long i = pud_index(address);
++
++ pud = pud + i;
++
++ if (after_bootmem && pud_val(*pud)) {
++ phys_pmd_update(pud, address, end);
++ return;
++ }
++
++ for (; i < PTRS_PER_PUD; pud++, i++) {
++ unsigned long paddr, pmd_phys;
++ pmd_t *pmd;
++
++ paddr = (address & PGDIR_MASK) + i*PUD_SIZE;
++ if (paddr >= end)
++ break;
++
++ pmd = alloc_static_page(&pmd_phys);
++ early_make_page_readonly(pmd, XENFEAT_writable_page_tables);
++ spin_lock(&init_mm.page_table_lock);
++ set_pud(pud, __pud(pmd_phys | _KERNPG_TABLE));
++ phys_pmd_init(pmd, paddr, end);
++ spin_unlock(&init_mm.page_table_lock);
++ }
++ __flush_tlb();
++}
++
++void __init xen_init_pt(void)
++{
++ unsigned long addr, *page;
++
++ memset((void *)init_level4_pgt, 0, PAGE_SIZE);
++ memset((void *)level3_kernel_pgt, 0, PAGE_SIZE);
++ memset((void *)level2_kernel_pgt, 0, PAGE_SIZE);
++
++ /* Find the initial pte page that was built for us. */
++ page = (unsigned long *)xen_start_info->pt_base;
++ addr = page[pgd_index(__START_KERNEL_map)];
++ addr_to_page(addr, page);
++ addr = page[pud_index(__START_KERNEL_map)];
++ addr_to_page(addr, page);
++
++ /* Construct mapping of initial pte page in our own directories. */
++ init_level4_pgt[pgd_index(__START_KERNEL_map)] =
++ mk_kernel_pgd(__pa_symbol(level3_kernel_pgt));
++ level3_kernel_pgt[pud_index(__START_KERNEL_map)] =
++ __pud(__pa_symbol(level2_kernel_pgt) |
++ _KERNPG_TABLE | _PAGE_USER);
++ memcpy((void *)level2_kernel_pgt, page, PAGE_SIZE);
++
++ early_make_page_readonly(init_level4_pgt,
++ XENFEAT_writable_page_tables);
++ early_make_page_readonly(init_level4_user_pgt,
++ XENFEAT_writable_page_tables);
++ early_make_page_readonly(level3_kernel_pgt,
++ XENFEAT_writable_page_tables);
++ early_make_page_readonly(level3_user_pgt,
++ XENFEAT_writable_page_tables);
++ early_make_page_readonly(level2_kernel_pgt,
++ XENFEAT_writable_page_tables);
++
++ xen_pgd_pin(__pa_symbol(init_level4_pgt));
++ xen_pgd_pin(__pa_symbol(init_level4_user_pgt));
++
++ set_pgd((pgd_t *)(init_level4_user_pgt + 511),
++ mk_kernel_pgd(__pa_symbol(level3_user_pgt)));
++}
++
++void __init extend_init_mapping(unsigned long tables_space)
++{
++ unsigned long va = __START_KERNEL_map;
++ unsigned long phys, addr, *pte_page;
++ pmd_t *pmd;
++ pte_t *pte, new_pte;
++ unsigned long *page = (unsigned long *)init_level4_pgt;
++
++ addr = page[pgd_index(va)];
++ addr_to_page(addr, page);
++ addr = page[pud_index(va)];
++ addr_to_page(addr, page);
++
++ /* Kill mapping of low 1MB. */
++ while (va < (unsigned long)&_text) {
++ HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0);
++ va += PAGE_SIZE;
++ }
++
++ /* Ensure init mappings cover kernel text/data and initial tables. */
++ while (va < (__START_KERNEL_map
++ + (start_pfn << PAGE_SHIFT)
++ + tables_space)) {
++ pmd = (pmd_t *)&page[pmd_index(va)];
++ if (pmd_none(*pmd)) {
++ pte_page = alloc_static_page(&phys);
++ early_make_page_readonly(
++ pte_page, XENFEAT_writable_page_tables);
++ set_pmd(pmd, __pmd(phys | _KERNPG_TABLE | _PAGE_USER));
++ } else {
++ addr = page[pmd_index(va)];
++ addr_to_page(addr, pte_page);
++ }
++ pte = (pte_t *)&pte_page[pte_index(va)];
++ if (pte_none(*pte)) {
++ new_pte = pfn_pte(
++ (va - __START_KERNEL_map) >> PAGE_SHIFT,
++ __pgprot(_KERNPG_TABLE | _PAGE_USER));
++ xen_l1_entry_update(pte, new_pte);
++ }
++ va += PAGE_SIZE;
++ }
++
++ /* Finally, blow away any spurious initial mappings. */
++ while (1) {
++ pmd = (pmd_t *)&page[pmd_index(va)];
++ if (pmd_none(*pmd))
++ break;
++ HYPERVISOR_update_va_mapping(va, __pte_ma(0), 0);
++ va += PAGE_SIZE;
++ }
++}
++
++static void __init find_early_table_space(unsigned long end)
++{
++ unsigned long puds, pmds, ptes, tables;
++
++ puds = (end + PUD_SIZE - 1) >> PUD_SHIFT;
++ pmds = (end + PMD_SIZE - 1) >> PMD_SHIFT;
++ ptes = (end + PTE_SIZE - 1) >> PAGE_SHIFT;
++
++ tables = round_up(puds * 8, PAGE_SIZE) +
++ round_up(pmds * 8, PAGE_SIZE) +
++ round_up(ptes * 8, PAGE_SIZE);
++
++ extend_init_mapping(tables);
++
++ table_start = start_pfn;
++ table_end = table_start + (tables>>PAGE_SHIFT);
++
++ early_printk("kernel direct mapping tables up to %lx @ %lx-%lx\n",
++ end, table_start << PAGE_SHIFT, table_end << PAGE_SHIFT);
++}
++
++/* Setup the direct mapping of the physical memory at PAGE_OFFSET.
++ This runs before bootmem is initialized and gets pages directly from the
++ physical memory. To access them they are temporarily mapped. */
++void __meminit init_memory_mapping(unsigned long start, unsigned long end)
++{
++ unsigned long next;
++
++ Dprintk("init_memory_mapping\n");
++
++ /*
++ * Find space for the kernel direct mapping tables.
++ * Later we should allocate these tables in the local node of the memory
++ * mapped. Unfortunately this is done currently before the nodes are
++ * discovered.
++ */
++ if (!after_bootmem)
++ find_early_table_space(end);
++
++ start = (unsigned long)__va(start);
++ end = (unsigned long)__va(end);
++
++ for (; start < end; start = next) {
++ unsigned long pud_phys;
++ pgd_t *pgd = pgd_offset_k(start);
++ pud_t *pud;
++
++ if (after_bootmem) {
++ pud = pud_offset_k(pgd, start & PGDIR_MASK);
++ make_page_readonly(pud, XENFEAT_writable_page_tables);
++ pud_phys = __pa(pud);
++ } else {
++ pud = alloc_static_page(&pud_phys);
++ early_make_page_readonly(pud, XENFEAT_writable_page_tables);
++ }
++ next = start + PGDIR_SIZE;
++ if (next > end)
++ next = end;
++ phys_pud_init(pud, __pa(start), __pa(next));
++ if (!after_bootmem)
++ set_pgd(pgd_offset_k(start), mk_kernel_pgd(pud_phys));
++ }
++
++ if (!after_bootmem) {
++ BUG_ON(start_pfn != table_end);
++
++ /* Re-vector virtual addresses pointing into the initial
++ mapping to the just-established permanent ones. */
++ xen_start_info = __va(__pa(xen_start_info));
++ xen_start_info->pt_base = (unsigned long)
++ __va(__pa(xen_start_info->pt_base));
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ phys_to_machine_mapping =
++ __va(__pa(xen_start_info->mfn_list));
++ xen_start_info->mfn_list = (unsigned long)
++ phys_to_machine_mapping;
++ }
++ if (xen_start_info->mod_start)
++ xen_start_info->mod_start = (unsigned long)
++ __va(__pa(xen_start_info->mod_start));
++
++ /* Destroy the Xen-created mappings beyond the kernel image as
++ * well as the temporary mappings created above. Prevents
++ * overlap with modules area (if init mapping is very big).
++ */
++ start = PAGE_ALIGN((unsigned long)_end);
++ end = __START_KERNEL_map + (table_end << PAGE_SHIFT);
++ for (; start < end; start += PAGE_SIZE)
++ WARN_ON(HYPERVISOR_update_va_mapping(
++ start, __pte_ma(0), 0));
++ }
++
++ __flush_tlb_all();
++}
++
++void __cpuinit zap_low_mappings(int cpu)
++{
++ /* this is not required for Xen */
++#if 0
++ swap_low_mappings();
++#endif
++}
++
++/* Compute zone sizes for the DMA and DMA32 zones in a node. */
++__init void
++size_zones(unsigned long *z, unsigned long *h,
++ unsigned long start_pfn, unsigned long end_pfn)
++{
++ int i;
++#ifndef CONFIG_XEN
++ unsigned long w;
++#endif
++
++ for (i = 0; i < MAX_NR_ZONES; i++)
++ z[i] = 0;
++
++#ifndef CONFIG_XEN
++ if (start_pfn < MAX_DMA_PFN)
++ z[ZONE_DMA] = MAX_DMA_PFN - start_pfn;
++ if (start_pfn < MAX_DMA32_PFN) {
++ unsigned long dma32_pfn = MAX_DMA32_PFN;
++ if (dma32_pfn > end_pfn)
++ dma32_pfn = end_pfn;
++ z[ZONE_DMA32] = dma32_pfn - start_pfn;
++ }
++ z[ZONE_NORMAL] = end_pfn - start_pfn;
++
++ /* Remove lower zones from higher ones. */
++ w = 0;
++ for (i = 0; i < MAX_NR_ZONES; i++) {
++ if (z[i])
++ z[i] -= w;
++ w += z[i];
++ }
++
++ /* Compute holes */
++ w = start_pfn;
++ for (i = 0; i < MAX_NR_ZONES; i++) {
++ unsigned long s = w;
++ w += z[i];
++ h[i] = e820_hole_size(s, w);
++ }
++
++ /* Add the space pace needed for mem_map to the holes too. */
++ for (i = 0; i < MAX_NR_ZONES; i++)
++ h[i] += (z[i] * sizeof(struct page)) / PAGE_SIZE;
++
++ /* The 16MB DMA zone has the kernel and other misc mappings.
++ Account them too */
++ if (h[ZONE_DMA]) {
++ h[ZONE_DMA] += dma_reserve;
++ if (h[ZONE_DMA] >= z[ZONE_DMA]) {
++ printk(KERN_WARNING
++ "Kernel too large and filling up ZONE_DMA?\n");
++ h[ZONE_DMA] = z[ZONE_DMA];
++ }
++ }
++#else
++ z[ZONE_DMA] = end_pfn;
++ for (i = 0; i < MAX_NR_ZONES; i++)
++ h[i] = 0;
++#endif
++}
++
++#ifndef CONFIG_NUMA
++void __init paging_init(void)
++{
++ unsigned long zones[MAX_NR_ZONES], holes[MAX_NR_ZONES];
++ int i;
++
++ memory_present(0, 0, end_pfn);
++ sparse_init();
++ size_zones(zones, holes, 0, end_pfn);
++ free_area_init_node(0, NODE_DATA(0), zones,
++ __pa(PAGE_OFFSET) >> PAGE_SHIFT, holes);
++
++ /* Switch to the real shared_info page, and clear the
++ * dummy page. */
++ set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
++ HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
++ memset(empty_zero_page, 0, sizeof(empty_zero_page));
++
++ init_mm.context.pinned = 1;
++
++ /* Setup mapping of lower 1st MB */
++ for (i = 0; i < NR_FIX_ISAMAPS; i++)
++ if (xen_start_info->flags & SIF_PRIVILEGED)
++ set_fixmap(FIX_ISAMAP_BEGIN - i, i * PAGE_SIZE);
++ else
++ __set_fixmap(FIX_ISAMAP_BEGIN - i,
++ virt_to_mfn(empty_zero_page) << PAGE_SHIFT,
++ PAGE_KERNEL_RO);
++}
++#endif
++
++/* Unmap a kernel mapping if it exists. This is useful to avoid prefetches
++ from the CPU leading to inconsistent cache lines. address and size
++ must be aligned to 2MB boundaries.
++ Does nothing when the mapping doesn't exist. */
++void __init clear_kernel_mapping(unsigned long address, unsigned long size)
++{
++ unsigned long end = address + size;
++
++ BUG_ON(address & ~LARGE_PAGE_MASK);
++ BUG_ON(size & ~LARGE_PAGE_MASK);
++
++ for (; address < end; address += LARGE_PAGE_SIZE) {
++ pgd_t *pgd = pgd_offset_k(address);
++ pud_t *pud;
++ pmd_t *pmd;
++ if (pgd_none(*pgd))
++ continue;
++ pud = pud_offset(pgd, address);
++ if (pud_none(*pud))
++ continue;
++ pmd = pmd_offset(pud, address);
++ if (!pmd || pmd_none(*pmd))
++ continue;
++ if (0 == (pmd_val(*pmd) & _PAGE_PSE)) {
++ /* Could handle this, but it should not happen currently. */
++ printk(KERN_ERR
++ "clear_kernel_mapping: mapping has been split. will leak memory\n");
++ pmd_ERROR(*pmd);
++ }
++ set_pmd(pmd, __pmd(0));
++ }
++ __flush_tlb_all();
++}
++
++/*
++ * Memory hotplug specific functions
++ */
++#if defined(CONFIG_ACPI_HOTPLUG_MEMORY) || defined(CONFIG_ACPI_HOTPLUG_MEMORY_MODULE)
++
++void online_page(struct page *page)
++{
++ ClearPageReserved(page);
++ init_page_count(page);
++ __free_page(page);
++ totalram_pages++;
++ num_physpages++;
++}
++
++#ifndef CONFIG_MEMORY_HOTPLUG
++/*
++ * Memory Hotadd without sparsemem. The mem_maps have been allocated in advance,
++ * just online the pages.
++ */
++int __add_pages(struct zone *z, unsigned long start_pfn, unsigned long nr_pages)
++{
++ int err = -EIO;
++ unsigned long pfn;
++ unsigned long total = 0, mem = 0;
++ for (pfn = start_pfn; pfn < start_pfn + nr_pages; pfn++) {
++ if (pfn_valid(pfn)) {
++ online_page(pfn_to_page(pfn));
++ err = 0;
++ mem++;
++ }
++ total++;
++ }
++ if (!err) {
++ z->spanned_pages += total;
++ z->present_pages += mem;
++ z->zone_pgdat->node_spanned_pages += total;
++ z->zone_pgdat->node_present_pages += mem;
++ }
++ return err;
++}
++#endif
++
++/*
++ * Memory is added always to NORMAL zone. This means you will never get
++ * additional DMA/DMA32 memory.
++ */
++int add_memory(u64 start, u64 size)
++{
++ struct pglist_data *pgdat = NODE_DATA(0);
++ struct zone *zone = pgdat->node_zones + MAX_NR_ZONES-2;
++ unsigned long start_pfn = start >> PAGE_SHIFT;
++ unsigned long nr_pages = size >> PAGE_SHIFT;
++ int ret;
++
++ ret = __add_pages(zone, start_pfn, nr_pages);
++ if (ret)
++ goto error;
++
++ init_memory_mapping(start, (start + size -1));
++
++ return ret;
++error:
++ printk("%s: Problem encountered in __add_pages!\n", __func__);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(add_memory);
++
++int remove_memory(u64 start, u64 size)
++{
++ return -EINVAL;
++}
++EXPORT_SYMBOL_GPL(remove_memory);
++
++#endif
++
++static struct kcore_list kcore_mem, kcore_vmalloc, kcore_kernel, kcore_modules,
++ kcore_vsyscall;
++
++void __init mem_init(void)
++{
++ long codesize, reservedpages, datasize, initsize;
++ unsigned long pfn;
++
++ contiguous_bitmap = alloc_bootmem_low_pages(
++ (end_pfn + 2*BITS_PER_LONG) >> 3);
++ BUG_ON(!contiguous_bitmap);
++ memset(contiguous_bitmap, 0, (end_pfn + 2*BITS_PER_LONG) >> 3);
++
++#if defined(CONFIG_SWIOTLB)
++ pci_swiotlb_init();
++#endif
++ no_iommu_init();
++
++ /* How many end-of-memory variables you have, grandma! */
++ max_low_pfn = end_pfn;
++ max_pfn = end_pfn;
++ num_physpages = end_pfn;
++ high_memory = (void *) __va(end_pfn * PAGE_SIZE);
++
++ /* clear the zero-page */
++ memset(empty_zero_page, 0, PAGE_SIZE);
++
++ reservedpages = 0;
++
++ /* this will put all low memory onto the freelists */
++#ifdef CONFIG_NUMA
++ totalram_pages = numa_free_all_bootmem();
++#else
++ totalram_pages = free_all_bootmem();
++#endif
++ /* XEN: init and count pages outside initial allocation. */
++ for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) {
++ ClearPageReserved(&mem_map[pfn]);
++ init_page_count(&mem_map[pfn]);
++ totalram_pages++;
++ }
++ reservedpages = end_pfn - totalram_pages - e820_hole_size(0, end_pfn);
++
++ after_bootmem = 1;
++
++ codesize = (unsigned long) &_etext - (unsigned long) &_text;
++ datasize = (unsigned long) &_edata - (unsigned long) &_etext;
++ initsize = (unsigned long) &__init_end - (unsigned long) &__init_begin;
++
++ /* Register memory areas for /proc/kcore */
++ kclist_add(&kcore_mem, __va(0), max_low_pfn << PAGE_SHIFT);
++ kclist_add(&kcore_vmalloc, (void *)VMALLOC_START,
++ VMALLOC_END-VMALLOC_START);
++ kclist_add(&kcore_kernel, &_stext, _end - _stext);
++ kclist_add(&kcore_modules, (void *)MODULES_VADDR, MODULES_LEN);
++ kclist_add(&kcore_vsyscall, (void *)VSYSCALL_START,
++ VSYSCALL_END - VSYSCALL_START);
++
++ printk("Memory: %luk/%luk available (%ldk kernel code, %ldk reserved, %ldk data, %ldk init)\n",
++ (unsigned long) nr_free_pages() << (PAGE_SHIFT-10),
++ end_pfn << (PAGE_SHIFT-10),
++ codesize >> 10,
++ reservedpages << (PAGE_SHIFT-10),
++ datasize >> 10,
++ initsize >> 10);
++
++#ifndef CONFIG_XEN
++#ifdef CONFIG_SMP
++ /*
++ * Sync boot_level4_pgt mappings with the init_level4_pgt
++ * except for the low identity mappings which are already zapped
++ * in init_level4_pgt. This sync-up is essential for AP's bringup
++ */
++ memcpy(boot_level4_pgt+1, init_level4_pgt+1, (PTRS_PER_PGD-1)*sizeof(pgd_t));
++#endif
++#endif
++}
++
++void free_initmem(void)
++{
++#ifdef __DO_LATER__
++ /*
++ * Some pages can be pinned, but some are not. Unpinning such pages
++ * triggers BUG().
++ */
++ unsigned long addr;
++
++ addr = (unsigned long)(&__init_begin);
++ for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
++ ClearPageReserved(virt_to_page(addr));
++ init_page_count(virt_to_page(addr));
++ memset((void *)(addr & ~(PAGE_SIZE-1)), 0xcc, PAGE_SIZE);
++ make_page_writable(
++ __va(__pa(addr)), XENFEAT_writable_page_tables);
++ /*
++ * Make pages from __PAGE_OFFSET address as well
++ */
++ make_page_writable(
++ (void *)addr, XENFEAT_writable_page_tables);
++ free_page(addr);
++ totalram_pages++;
++ }
++ memset(__initdata_begin, 0xba, __initdata_end - __initdata_begin);
++ printk ("Freeing unused kernel memory: %luk freed\n", (__init_end - __init_begin) >> 10);
++#endif
++}
++
++#ifdef CONFIG_DEBUG_RODATA
++
++extern char __start_rodata, __end_rodata;
++void mark_rodata_ro(void)
++{
++ unsigned long addr = (unsigned long)&__start_rodata;
++
++ for (; addr < (unsigned long)&__end_rodata; addr += PAGE_SIZE)
++ change_page_attr_addr(addr, 1, PAGE_KERNEL_RO);
++
++ printk ("Write protecting the kernel read-only data: %luk\n",
++ (&__end_rodata - &__start_rodata) >> 10);
++
++ /*
++ * change_page_attr_addr() requires a global_flush_tlb() call after it.
++ * We do this after the printk so that if something went wrong in the
++ * change, the printk gets out at least to give a better debug hint
++ * of who is the culprit.
++ */
++ global_flush_tlb();
++}
++#endif
++
++#ifdef CONFIG_BLK_DEV_INITRD
++void free_initrd_mem(unsigned long start, unsigned long end)
++{
++ if (start >= end)
++ return;
++ printk ("Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
++ for (; start < end; start += PAGE_SIZE) {
++ ClearPageReserved(virt_to_page(start));
++ init_page_count(virt_to_page(start));
++ free_page(start);
++ totalram_pages++;
++ }
++}
++#endif
++
++void __init reserve_bootmem_generic(unsigned long phys, unsigned len)
++{
++ /* Should check here against the e820 map to avoid double free */
++#ifdef CONFIG_NUMA
++ int nid = phys_to_nid(phys);
++ reserve_bootmem_node(NODE_DATA(nid), phys, len);
++#else
++ reserve_bootmem(phys, len);
++#endif
++ if (phys+len <= MAX_DMA_PFN*PAGE_SIZE)
++ dma_reserve += len / PAGE_SIZE;
++}
++
++int kern_addr_valid(unsigned long addr)
++{
++ unsigned long above = ((long)addr) >> __VIRTUAL_MASK_SHIFT;
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++
++ if (above != 0 && above != -1UL)
++ return 0;
++
++ pgd = pgd_offset_k(addr);
++ if (pgd_none(*pgd))
++ return 0;
++
++ pud = pud_offset_k(pgd, addr);
++ if (pud_none(*pud))
++ return 0;
++
++ pmd = pmd_offset(pud, addr);
++ if (pmd_none(*pmd))
++ return 0;
++ if (pmd_large(*pmd))
++ return pfn_valid(pmd_pfn(*pmd));
++
++ pte = pte_offset_kernel(pmd, addr);
++ if (pte_none(*pte))
++ return 0;
++ return pfn_valid(pte_pfn(*pte));
++}
++
++#ifdef CONFIG_SYSCTL
++#include <linux/sysctl.h>
++
++extern int exception_trace, page_fault_trace;
++
++static ctl_table debug_table2[] = {
++ { 99, "exception-trace", &exception_trace, sizeof(int), 0644, NULL,
++ proc_dointvec },
++ { 0, }
++};
++
++static ctl_table debug_root_table2[] = {
++ { .ctl_name = CTL_DEBUG, .procname = "debug", .mode = 0555,
++ .child = debug_table2 },
++ { 0 },
++};
++
++static __init int x8664_sysctl_init(void)
++{
++ register_sysctl_table(debug_root_table2, 1);
++ return 0;
++}
++__initcall(x8664_sysctl_init);
++#endif
++
++/* A pseudo VMAs to allow ptrace access for the vsyscall page. This only
++ covers the 64bit vsyscall page now. 32bit has a real VMA now and does
++ not need special handling anymore. */
++
++static struct vm_area_struct gate_vma = {
++ .vm_start = VSYSCALL_START,
++ .vm_end = VSYSCALL_END,
++ .vm_page_prot = PAGE_READONLY
++};
++
++struct vm_area_struct *get_gate_vma(struct task_struct *tsk)
++{
++#ifdef CONFIG_IA32_EMULATION
++ if (test_tsk_thread_flag(tsk, TIF_IA32))
++ return NULL;
++#endif
++ return &gate_vma;
++}
++
++int in_gate_area(struct task_struct *task, unsigned long addr)
++{
++ struct vm_area_struct *vma = get_gate_vma(task);
++ if (!vma)
++ return 0;
++ return (addr >= vma->vm_start) && (addr < vma->vm_end);
++}
++
++/* Use this when you have no reliable task/vma, typically from interrupt
++ * context. It is less reliable than using the task's vma and may give
++ * false positives.
++ */
++int in_gate_area_no_task(unsigned long addr)
++{
++ return (addr >= VSYSCALL_START) && (addr < VSYSCALL_END);
++}
+diff -urNp linux-2.6/arch/x86_64/mm/Makefile new/arch/x86_64/mm/Makefile
+--- linux-2.6/arch/x86_64/mm/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/mm/Makefile 2006-05-09 12:33:21.000000000 +0200
+@@ -9,3 +9,13 @@ obj-$(CONFIG_K8_NUMA) += k8topology.o
+ obj-$(CONFIG_ACPI_NUMA) += srat.o
+
+ hugetlbpage-y = ../../i386/mm/hugetlbpage.o
++
++ifdef CONFIG_XEN
++include $(srctree)/scripts/Makefile.xen
++
++ioremap-y += ../../i386/mm/ioremap-xen.o
++hypervisor-y += ../../i386/mm/hypervisor.o
++obj-y += hypervisor.o
++
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/arch/x86_64/mm/pageattr-xen.c new/arch/x86_64/mm/pageattr-xen.c
+--- linux-2.6/arch/x86_64/mm/pageattr-xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/arch/x86_64/mm/pageattr-xen.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,396 @@
++/*
++ * Copyright 2002 Andi Kleen, SuSE Labs.
++ * Thanks to Ben LaHaise for precious feedback.
++ */
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <linux/sched.h>
++#include <linux/highmem.h>
++#include <linux/module.h>
++#include <linux/slab.h>
++#include <asm/uaccess.h>
++#include <asm/processor.h>
++#include <asm/tlbflush.h>
++#include <asm/io.h>
++
++#ifdef CONFIG_XEN
++#include <asm/pgalloc.h>
++#include <asm/mmu_context.h>
++
++LIST_HEAD(mm_unpinned);
++DEFINE_SPINLOCK(mm_unpinned_lock);
++
++static inline void mm_walk_set_prot(void *pt, pgprot_t flags)
++{
++ struct page *page = virt_to_page(pt);
++ unsigned long pfn = page_to_pfn(page);
++
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__va(pfn << PAGE_SHIFT),
++ pfn_pte(pfn, flags), 0));
++}
++
++static void mm_walk(struct mm_struct *mm, pgprot_t flags)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++ int g,u,m;
++
++ pgd = mm->pgd;
++ /*
++ * Cannot iterate up to USER_PTRS_PER_PGD as these pagetables may not
++ * be the 'current' task's pagetables (e.g., current may be 32-bit,
++ * but the pagetables may be for a 64-bit task).
++ * Subtracting 1 from TASK_SIZE64 means the loop limit is correct
++ * regardless of whether TASK_SIZE64 is a multiple of PGDIR_SIZE.
++ */
++ for (g = 0; g <= ((TASK_SIZE64-1) / PGDIR_SIZE); g++, pgd++) {
++ if (pgd_none(*pgd))
++ continue;
++ pud = pud_offset(pgd, 0);
++ if (PTRS_PER_PUD > 1) /* not folded */
++ mm_walk_set_prot(pud,flags);
++ for (u = 0; u < PTRS_PER_PUD; u++, pud++) {
++ if (pud_none(*pud))
++ continue;
++ pmd = pmd_offset(pud, 0);
++ if (PTRS_PER_PMD > 1) /* not folded */
++ mm_walk_set_prot(pmd,flags);
++ for (m = 0; m < PTRS_PER_PMD; m++, pmd++) {
++ if (pmd_none(*pmd))
++ continue;
++ pte = pte_offset_kernel(pmd,0);
++ mm_walk_set_prot(pte,flags);
++ }
++ }
++ }
++}
++
++void mm_pin(struct mm_struct *mm)
++{
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++
++ spin_lock(&mm->page_table_lock);
++
++ mm_walk(mm, PAGE_KERNEL_RO);
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)mm->pgd,
++ pfn_pte(virt_to_phys(mm->pgd)>>PAGE_SHIFT, PAGE_KERNEL_RO),
++ UVMF_TLB_FLUSH));
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__user_pgd(mm->pgd),
++ pfn_pte(virt_to_phys(__user_pgd(mm->pgd))>>PAGE_SHIFT, PAGE_KERNEL_RO),
++ UVMF_TLB_FLUSH));
++ xen_pgd_pin(__pa(mm->pgd)); /* kernel */
++ xen_pgd_pin(__pa(__user_pgd(mm->pgd))); /* user */
++ mm->context.pinned = 1;
++ spin_lock(&mm_unpinned_lock);
++ list_del(&mm->context.unpinned);
++ spin_unlock(&mm_unpinned_lock);
++
++ spin_unlock(&mm->page_table_lock);
++}
++
++void mm_unpin(struct mm_struct *mm)
++{
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++
++ spin_lock(&mm->page_table_lock);
++
++ xen_pgd_unpin(__pa(mm->pgd));
++ xen_pgd_unpin(__pa(__user_pgd(mm->pgd)));
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)mm->pgd,
++ pfn_pte(virt_to_phys(mm->pgd)>>PAGE_SHIFT, PAGE_KERNEL), 0));
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__user_pgd(mm->pgd),
++ pfn_pte(virt_to_phys(__user_pgd(mm->pgd))>>PAGE_SHIFT, PAGE_KERNEL), 0));
++ mm_walk(mm, PAGE_KERNEL);
++ xen_tlb_flush();
++ mm->context.pinned = 0;
++ spin_lock(&mm_unpinned_lock);
++ list_add(&mm->context.unpinned, &mm_unpinned);
++ spin_unlock(&mm_unpinned_lock);
++
++ spin_unlock(&mm->page_table_lock);
++}
++
++void mm_pin_all(void)
++{
++ if (xen_feature(XENFEAT_writable_page_tables))
++ return;
++
++ while (!list_empty(&mm_unpinned))
++ mm_pin(list_entry(mm_unpinned.next, struct mm_struct,
++ context.unpinned));
++}
++
++void _arch_dup_mmap(struct mm_struct *mm)
++{
++ if (!mm->context.pinned)
++ mm_pin(mm);
++}
++
++void _arch_exit_mmap(struct mm_struct *mm)
++{
++ struct task_struct *tsk = current;
++
++ task_lock(tsk);
++
++ /*
++ * We aggressively remove defunct pgd from cr3. We execute unmap_vmas()
++ * *much* faster this way, as no tlb flushes means bigger wrpt batches.
++ */
++ if ( tsk->active_mm == mm )
++ {
++ tsk->active_mm = &init_mm;
++ atomic_inc(&init_mm.mm_count);
++
++ switch_mm(mm, &init_mm, tsk);
++
++ atomic_dec(&mm->mm_count);
++ BUG_ON(atomic_read(&mm->mm_count) == 0);
++ }
++
++ task_unlock(tsk);
++
++ if ( mm->context.pinned && (atomic_read(&mm->mm_count) == 1) )
++ mm_unpin(mm);
++}
++
++void pte_free(struct page *pte)
++{
++ unsigned long va = (unsigned long)__va(page_to_pfn(pte)<<PAGE_SHIFT);
++
++ if (!pte_write(*virt_to_ptep(va)))
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ va, pfn_pte(page_to_pfn(pte), PAGE_KERNEL), 0));
++ __free_page(pte);
++}
++#endif /* CONFIG_XEN */
++
++static inline pte_t *lookup_address(unsigned long address)
++{
++ pgd_t *pgd = pgd_offset_k(address);
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t *pte;
++ if (pgd_none(*pgd))
++ return NULL;
++ pud = pud_offset(pgd, address);
++ if (!pud_present(*pud))
++ return NULL;
++ pmd = pmd_offset(pud, address);
++ if (!pmd_present(*pmd))
++ return NULL;
++ if (pmd_large(*pmd))
++ return (pte_t *)pmd;
++ pte = pte_offset_kernel(pmd, address);
++ if (pte && !pte_present(*pte))
++ pte = NULL;
++ return pte;
++}
++
++static struct page *split_large_page(unsigned long address, pgprot_t prot,
++ pgprot_t ref_prot)
++{
++ int i;
++ unsigned long addr;
++ struct page *base = alloc_pages(GFP_KERNEL, 0);
++ pte_t *pbase;
++ if (!base)
++ return NULL;
++ /*
++ * page_private is used to track the number of entries in
++ * the page table page have non standard attributes.
++ */
++ SetPagePrivate(base);
++ page_private(base) = 0;
++
++ address = __pa(address);
++ addr = address & LARGE_PAGE_MASK;
++ pbase = (pte_t *)page_address(base);
++ for (i = 0; i < PTRS_PER_PTE; i++, addr += PAGE_SIZE) {
++ pbase[i] = pfn_pte(addr >> PAGE_SHIFT,
++ addr == address ? prot : ref_prot);
++ }
++ return base;
++}
++
++
++static void flush_kernel_map(void *address)
++{
++ if (0 && address && cpu_has_clflush) {
++ /* is this worth it? */
++ int i;
++ for (i = 0; i < PAGE_SIZE; i += boot_cpu_data.x86_clflush_size)
++ asm volatile("clflush (%0)" :: "r" (address + i));
++ } else
++ asm volatile("wbinvd":::"memory");
++ if (address)
++ __flush_tlb_one(address);
++ else
++ __flush_tlb_all();
++}
++
++
++static inline void flush_map(unsigned long address)
++{
++ on_each_cpu(flush_kernel_map, (void *)address, 1, 1);
++}
++
++static struct page *deferred_pages; /* protected by init_mm.mmap_sem */
++
++static inline void save_page(struct page *fpage)
++{
++ fpage->lru.next = (struct list_head *)deferred_pages;
++ deferred_pages = fpage;
++}
++
++/*
++ * No more special protections in this 2/4MB area - revert to a
++ * large page again.
++ */
++static void revert_page(unsigned long address, pgprot_t ref_prot)
++{
++ pgd_t *pgd;
++ pud_t *pud;
++ pmd_t *pmd;
++ pte_t large_pte;
++
++ pgd = pgd_offset_k(address);
++ BUG_ON(pgd_none(*pgd));
++ pud = pud_offset(pgd,address);
++ BUG_ON(pud_none(*pud));
++ pmd = pmd_offset(pud, address);
++ BUG_ON(pmd_val(*pmd) & _PAGE_PSE);
++ pgprot_val(ref_prot) |= _PAGE_PSE;
++ large_pte = mk_pte_phys(__pa(address) & LARGE_PAGE_MASK, ref_prot);
++ set_pte((pte_t *)pmd, large_pte);
++}
++
++static int
++__change_page_attr(unsigned long address, unsigned long pfn, pgprot_t prot,
++ pgprot_t ref_prot)
++{
++ pte_t *kpte;
++ struct page *kpte_page;
++ unsigned kpte_flags;
++ pgprot_t ref_prot2;
++ kpte = lookup_address(address);
++ if (!kpte) return 0;
++ kpte_page = virt_to_page(((unsigned long)kpte) & PAGE_MASK);
++ kpte_flags = pte_val(*kpte);
++ if (pgprot_val(prot) != pgprot_val(ref_prot)) {
++ if ((kpte_flags & _PAGE_PSE) == 0) {
++ set_pte(kpte, pfn_pte(pfn, prot));
++ } else {
++ /*
++ * split_large_page will take the reference for this
++ * change_page_attr on the split page.
++ */
++
++ struct page *split;
++ ref_prot2 = __pgprot(pgprot_val(pte_pgprot(*lookup_address(address))) & ~(1<<_PAGE_BIT_PSE));
++
++ split = split_large_page(address, prot, ref_prot2);
++ if (!split)
++ return -ENOMEM;
++ set_pte(kpte,mk_pte(split, ref_prot2));
++ kpte_page = split;
++ }
++ page_private(kpte_page)++;
++ } else if ((kpte_flags & _PAGE_PSE) == 0) {
++ set_pte(kpte, pfn_pte(pfn, ref_prot));
++ BUG_ON(page_private(kpte_page) == 0);
++ page_private(kpte_page)--;
++ } else
++ BUG();
++
++ /* on x86-64 the direct mapping set at boot is not using 4k pages */
++ /*
++ * ..., but the XEN guest kernels (currently) do:
++ * If the pte was reserved, it means it was created at boot
++ * time (not via split_large_page) and in turn we must not
++ * replace it with a large page.
++ */
++#ifndef CONFIG_XEN
++ BUG_ON(PageReserved(kpte_page));
++#endif
++ if (page_private(kpte_page) == 0) {
++ save_page(kpte_page);
++ revert_page(address, ref_prot);
++ }
++ return 0;
++}
++
++/*
++ * Change the page attributes of an page in the linear mapping.
++ *
++ * This should be used when a page is mapped with a different caching policy
++ * than write-back somewhere - some CPUs do not like it when mappings with
++ * different caching policies exist. This changes the page attributes of the
++ * in kernel linear mapping too.
++ *
++ * The caller needs to ensure that there are no conflicting mappings elsewhere.
++ * This function only deals with the kernel linear map.
++ *
++ * Caller must call global_flush_tlb() after this.
++ */
++int change_page_attr_addr(unsigned long address, int numpages, pgprot_t prot)
++{
++ int err = 0;
++ int i;
++
++ down_write(&init_mm.mmap_sem);
++ for (i = 0; i < numpages; i++, address += PAGE_SIZE) {
++ unsigned long pfn = __pa(address) >> PAGE_SHIFT;
++
++ err = __change_page_attr(address, pfn, prot, PAGE_KERNEL);
++ if (err)
++ break;
++ /* Handle kernel mapping too which aliases part of the
++ * lowmem */
++ if (__pa(address) < KERNEL_TEXT_SIZE) {
++ unsigned long addr2;
++ pgprot_t prot2 = prot;
++ addr2 = __START_KERNEL_map + __pa(address);
++ pgprot_val(prot2) &= ~_PAGE_NX;
++ err = __change_page_attr(addr2, pfn, prot2, PAGE_KERNEL_EXEC);
++ }
++ }
++ up_write(&init_mm.mmap_sem);
++ return err;
++}
++
++/* Don't call this for MMIO areas that may not have a mem_map entry */
++int change_page_attr(struct page *page, int numpages, pgprot_t prot)
++{
++ unsigned long addr = (unsigned long)page_address(page);
++ return change_page_attr_addr(addr, numpages, prot);
++}
++
++void global_flush_tlb(void)
++{
++ struct page *dpage;
++
++ down_read(&init_mm.mmap_sem);
++ dpage = xchg(&deferred_pages, NULL);
++ up_read(&init_mm.mmap_sem);
++
++ flush_map((dpage && !dpage->lru.next) ? (unsigned long)page_address(dpage) : 0);
++ while (dpage) {
++ struct page *tmp = dpage;
++ dpage = (struct page *)dpage->lru.next;
++ ClearPagePrivate(tmp);
++ __free_page(tmp);
++ }
++}
++
++EXPORT_SYMBOL(change_page_attr);
++EXPORT_SYMBOL(global_flush_tlb);
+diff -urNp linux-2.6/arch/x86_64/oprofile/Makefile new/arch/x86_64/oprofile/Makefile
+--- linux-2.6/arch/x86_64/oprofile/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/oprofile/Makefile 2006-05-09 12:33:21.000000000 +0200
+@@ -11,9 +11,12 @@ DRIVER_OBJS = $(addprefix ../../../drive
+ oprofilefs.o oprofile_stats.o \
+ timer_int.o )
+
++ifdef CONFIG_XEN
++OPROFILE-y := xenoprof.o
++else
+ OPROFILE-y := init.o backtrace.o
+ OPROFILE-$(CONFIG_X86_LOCAL_APIC) += nmi_int.o op_model_athlon.o op_model_p4.o \
+ op_model_ppro.o
+ OPROFILE-$(CONFIG_X86_IO_APIC) += nmi_timer_int.o
+-
++endif
+ oprofile-y = $(DRIVER_OBJS) $(addprefix ../../i386/oprofile/, $(OPROFILE-y))
+diff -urNp linux-2.6/arch/x86_64/pci/Makefile new/arch/x86_64/pci/Makefile
+--- linux-2.6/arch/x86_64/pci/Makefile 2006-07-03 14:14:30.000000000 +0200
++++ new/arch/x86_64/pci/Makefile 2006-05-09 12:33:21.000000000 +0200
+@@ -15,11 +15,23 @@ obj-$(CONFIG_PCI_MMCONFIG) += mmconfig.o
+
+ obj-$(CONFIG_NUMA) += k8-bus.o
+
++# pcifront should be after mmconfig.o and direct.o as it should only
++# take over if direct access to the PCI bus is unavailable
++obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront.o
++
+ direct-y += ../../i386/pci/direct.o
+ acpi-y += ../../i386/pci/acpi.o
++pcifront-y += ../../i386/pci/pcifront.o
+ legacy-y += ../../i386/pci/legacy.o
+ irq-y += ../../i386/pci/irq.o
+ common-y += ../../i386/pci/common.o
+ fixup-y += ../../i386/pci/fixup.o
+ i386-y += ../../i386/pci/i386.o
+ init-y += ../../i386/pci/init.o
++
++ifdef CONFIG_XEN
++irq-y := ../../i386/pci/irq-xen.o
++include $(srctree)/scripts/Makefile.xen
++
++obj-y := $(call cherrypickxen, $(obj-y))
++endif
+diff -urNp linux-2.6/Documentation/networking/netdevices.txt new/Documentation/networking/netdevices.txt
+--- linux-2.6/Documentation/networking/netdevices.txt 2006-07-03 14:14:08.000000000 +0200
++++ new/Documentation/networking/netdevices.txt 2006-07-07 15:10:03.000000000 +0200
+@@ -42,9 +42,9 @@ dev->get_stats:
+ Context: nominally process, but don't sleep inside an rwlock
+
+ dev->hard_start_xmit:
+- Synchronization: dev->xmit_lock spinlock.
++ Synchronization: netif_tx_lock spinlock.
+ When the driver sets NETIF_F_LLTX in dev->features this will be
+- called without holding xmit_lock. In this case the driver
++ called without holding netif_tx_lock. In this case the driver
+ has to lock by itself when needed. It is recommended to use a try lock
+ for this and return -1 when the spin lock fails.
+ The locking there should also properly protect against
+@@ -62,12 +62,12 @@ dev->hard_start_xmit:
+ Only valid when NETIF_F_LLTX is set.
+
+ dev->tx_timeout:
+- Synchronization: dev->xmit_lock spinlock.
++ Synchronization: netif_tx_lock spinlock.
+ Context: BHs disabled
+ Notes: netif_queue_stopped() is guaranteed true
+
+ dev->set_multicast_list:
+- Synchronization: dev->xmit_lock spinlock.
++ Synchronization: netif_tx_lock spinlock.
+ Context: BHs disabled
+
+ dev->poll:
+diff -urNp linux-2.6/drivers/acpi/Kconfig new/drivers/acpi/Kconfig
+--- linux-2.6/drivers/acpi/Kconfig 2006-07-03 14:14:31.000000000 +0200
++++ new/drivers/acpi/Kconfig 2006-05-09 12:33:23.000000000 +0200
+@@ -46,7 +46,7 @@ if ACPI
+
+ config ACPI_SLEEP
+ bool "Sleep States"
+- depends on X86 && (!SMP || SUSPEND_SMP)
++ depends on X86 && (!SMP || SUSPEND_SMP) && !XEN
+ depends on PM
+ default y
+ ---help---
+@@ -300,6 +300,7 @@ config ACPI_SYSTEM
+ config X86_PM_TIMER
+ bool "Power Management Timer Support" if EMBEDDED
+ depends on X86
++ depends on !XEN
+ default y
+ help
+ The Power Management Timer is available on all ACPI-capable,
+diff -urNp linux-2.6/drivers/char/mem.c new/drivers/char/mem.c
+--- linux-2.6/drivers/char/mem.c 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/mem.c 2006-05-09 12:33:33.000000000 +0200
+@@ -103,6 +103,7 @@ static inline int valid_mmap_phys_addr_r
+ }
+ #endif
+
++#ifndef ARCH_HAS_DEV_MEM
+ /*
+ * This funcion reads the *physical* memory. The f_pos points directly to the
+ * memory location.
+@@ -225,6 +226,7 @@ static ssize_t write_mem(struct file * f
+ *ppos += written;
+ return written;
+ }
++#endif
+
+ #ifndef __HAVE_PHYS_MEM_ACCESS_PROT
+ static pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
+@@ -778,6 +780,7 @@ static int open_port(struct inode * inod
+ #define open_kmem open_mem
+ #define open_oldmem open_mem
+
++#ifndef ARCH_HAS_DEV_MEM
+ static struct file_operations mem_fops = {
+ .llseek = memory_lseek,
+ .read = read_mem,
+@@ -785,6 +788,9 @@ static struct file_operations mem_fops =
+ .mmap = mmap_mem,
+ .open = open_mem,
+ };
++#else
++extern struct file_operations mem_fops;
++#endif
+
+ static struct file_operations kmem_fops = {
+ .llseek = memory_lseek,
+diff -urNp linux-2.6/drivers/char/tpm/Kconfig new/drivers/char/tpm/Kconfig
+--- linux-2.6/drivers/char/tpm/Kconfig 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/tpm/Kconfig 2006-07-07 15:17:16.000000000 +0200
+@@ -31,7 +31,7 @@ config TCG_TIS
+
+ config TCG_NSC
+ tristate "National Semiconductor TPM Interface"
+- depends on TCG_TPM && PNPACPI
++ depends on TCG_TPM && PNPACPI && !XEN_UNPRIVILEGED_GUEST
+ ---help---
+ If you have a TPM security chip from National Semicondutor
+ say Yes and it will be accessible from within Linux. To
+@@ -58,5 +58,13 @@ config TCG_INFINEON
+ Further information on this driver and the supported hardware
+ can be found at http://www.prosec.rub.de/tpm
+
+-endmenu
++config TCG_XEN
++ tristate "XEN TPM Interface"
++ depends on TCG_TPM && XEN
++ ---help---
++ If you want to make TPM support available to a Xen user domain,
++ say Yes and it will be accessible from within Linux.
++ To compile this driver as a module, choose M here; the module
++ will be called tpm_xenu.
+
++endmenu
+diff -urNp linux-2.6/drivers/char/tpm/Makefile new/drivers/char/tpm/Makefile
+--- linux-2.6/drivers/char/tpm/Makefile 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/tpm/Makefile 2006-07-07 15:17:21.000000000 +0200
+@@ -9,3 +9,5 @@ obj-$(CONFIG_TCG_TIS) += tpm_tis.o
+ obj-$(CONFIG_TCG_NSC) += tpm_nsc.o
+ obj-$(CONFIG_TCG_ATMEL) += tpm_atmel.o
+ obj-$(CONFIG_TCG_INFINEON) += tpm_infineon.o
++obj-$(CONFIG_TCG_XEN) += tpm_xenu.o
++tpm_xenu-y = tpm_xen.o tpm_vtpm.o
+diff -urNp linux-2.6/drivers/char/tpm/tpm.c new/drivers/char/tpm/tpm.c
+--- linux-2.6/drivers/char/tpm/tpm.c 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/tpm/tpm.c 2006-07-07 15:19:52.000000000 +0200
+@@ -30,7 +30,9 @@
+
+ enum tpm_const {
+ TPM_MINOR = 224, /* officially assigned */
++#ifndef CONFIG_XEN
+ TPM_BUFSIZE = 2048,
++#endif
+ TPM_NUM_DEVICES = 256,
+ };
+
+@@ -331,7 +333,11 @@ static void timeout_work(void *ptr)
+
+ down(&chip->buffer_mutex);
+ atomic_set(&chip->data_pending, 0);
++#ifndef CONFIG_XEN
+ memset(chip->data_buffer, 0, TPM_BUFSIZE);
++#else
++ memset(chip->data_buffer, 0, get_chip_buffersize(chip));
++#endif
+ up(&chip->buffer_mutex);
+ }
+
+@@ -921,7 +927,12 @@ int tpm_open(struct inode *inode, struct
+
+ spin_unlock(&driver_lock);
+
++#ifndef CONFIG_XEN
+ chip->data_buffer = kmalloc(TPM_BUFSIZE * sizeof(u8), GFP_KERNEL);
++#else
++ chip->data_buffer = kmalloc(get_chip_buffersize(chip) * sizeof(u8),
++ GFP_KERNEL);
++#endif
+ if (chip->data_buffer == NULL) {
+ chip->num_opens--;
+ put_device(chip->dev);
+@@ -969,8 +980,13 @@ ssize_t tpm_write(struct file *file, con
+
+ down(&chip->buffer_mutex);
+
++#ifndef CONFIG_XEN
+ if (in_size > TPM_BUFSIZE)
+ in_size = TPM_BUFSIZE;
++#else
++ if (in_size > get_chip_buffersize(chip))
++ in_size = get_chip_buffersize(chip);
++#endif
+
+ if (copy_from_user
+ (chip->data_buffer, (void __user *) buf, in_size)) {
+@@ -979,9 +995,17 @@ ssize_t tpm_write(struct file *file, con
+ }
+
+ /* atomic tpm command send and result receive */
++#ifndef CONFIG_XEN
+ out_size = tpm_transmit(chip, chip->data_buffer, TPM_BUFSIZE);
++#else
++ out_size = tpm_transmit(chip, chip->data_buffer,
++ get_chip_buffersize(chip));
++#endif
+
+ atomic_set(&chip->data_pending, out_size);
++#ifdef CONFIG_XEN
++ atomic_set(&chip->data_position, 0);
++#endif
+ up(&chip->buffer_mutex);
+
+ /* Set a timeout by which the reader must come claim the result */
+@@ -996,21 +1020,52 @@ ssize_t tpm_read(struct file *file, char
+ {
+ struct tpm_chip *chip = file->private_data;
+ int ret_size;
++#ifdef CONFIG_XEN
++ int pos, pending = 0;
++#endif
+
++#ifndef CONFIG_XEN
+ del_singleshot_timer_sync(&chip->user_read_timer);
+ flush_scheduled_work();
++#endif
+ ret_size = atomic_read(&chip->data_pending);
++#ifndef CONFIG_XEN
+ atomic_set(&chip->data_pending, 0);
++#endif
+ if (ret_size > 0) { /* relay data */
+ if (size < ret_size)
+ ret_size = size;
+
++#ifdef CONFIG_XEN
++ pos = atomic_read(&chip->data_position);
++#endif
+ down(&chip->buffer_mutex);
++#ifndef CONFIG_XEN
+ if (copy_to_user(buf, chip->data_buffer, ret_size))
++#else
++ if (copy_to_user(buf, &chip->data_buffer[pos], ret_size)) {
++#endif
+ ret_size = -EFAULT;
++#ifdef CONFIG_XEN
++ } else {
++ pending = atomic_read(&chip->data_pending) - ret_size;
++ if ( pending ) {
++ atomic_set(&chip->data_pending, pending);
++ atomic_set(&chip->data_position,
++ pos+ret_size);
++ }
++ }
++#endif
+ up(&chip->buffer_mutex);
+ }
+
++#ifdef CONFIG_XEN
++ if ( ret_size <= 0 || pending == 0 ) {
++ atomic_set(&chip->data_pending, 0);
++ del_singleshot_timer_sync(&chip->user_read_timer);
++ flush_scheduled_work();
++ }
++#endif
+ return ret_size;
+ }
+ EXPORT_SYMBOL_GPL(tpm_read);
+diff -urNp linux-2.6/drivers/char/tpm/tpm.h new/drivers/char/tpm/tpm.h
+--- linux-2.6/drivers/char/tpm/tpm.h 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/tpm/tpm.h 2006-07-07 15:21:18.000000000 +0200
+@@ -61,6 +61,7 @@ struct tpm_vendor_specific {
+ const u8 req_complete_mask;
+ const u8 req_complete_val;
+ const u8 req_canceled;
++ u32 buffersize;
+ void __iomem *iobase; /* ioremapped address */
+ unsigned long base; /* TPM base address */
+
+@@ -94,6 +95,7 @@ struct tpm_chip {
+ /* Data passed to and from the tpm via the read/write calls */
+ u8 *data_buffer;
+ atomic_t data_pending;
++ atomic_t data_position;
+ struct semaphore buffer_mutex;
+
+ struct timer_list user_read_timer; /* user needs to claim result */
+@@ -121,6 +123,11 @@ static inline void tpm_write_index(int b
+ outb(value & 0xFF, base+1);
+ }
+
++static inline u32 get_chip_buffersize(struct tpm_chip *chip)
++{
++ return chip->vendor.buffersize;
++}
++
+ extern void tpm_get_timeouts(struct tpm_chip *);
+ extern void tpm_gen_interrupt(struct tpm_chip *);
+ extern void tpm_continue_selftest(struct tpm_chip *);
+diff -urNp linux-2.6/drivers/char/tpm/tpm_vtpm.c new/drivers/char/tpm/tpm_vtpm.c
+--- linux-2.6/drivers/char/tpm/tpm_vtpm.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/char/tpm/tpm_vtpm.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,547 @@
++/*
++ * Copyright (C) 2006 IBM Corporation
++ *
++ * Authors:
++ * Stefan Berger <stefanb@us.ibm.com>
++ *
++ * Generic device driver part for device drivers in a virtualized
++ * environment.
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License as
++ * published by the Free Software Foundation, version 2 of the
++ * License.
++ *
++ */
++
++#include <asm/uaccess.h>
++#include <linux/list.h>
++#include <linux/device.h>
++#include <linux/interrupt.h>
++#include <linux/platform_device.h>
++#include "tpm.h"
++#include "tpm_vtpm.h"
++
++/* read status bits */
++enum {
++ STATUS_BUSY = 0x01,
++ STATUS_DATA_AVAIL = 0x02,
++ STATUS_READY = 0x04
++};
++
++struct transmission {
++ struct list_head next;
++
++ unsigned char *request;
++ size_t request_len;
++ size_t request_buflen;
++
++ unsigned char *response;
++ size_t response_len;
++ size_t response_buflen;
++
++ unsigned int flags;
++};
++
++enum {
++ TRANSMISSION_FLAG_WAS_QUEUED = 0x1
++};
++
++
++enum {
++ DATAEX_FLAG_QUEUED_ONLY = 0x1
++};
++
++
++/* local variables */
++
++/* local function prototypes */
++static int _vtpm_send_queued(struct tpm_chip *chip);
++
++
++/* =============================================================
++ * Some utility functions
++ * =============================================================
++ */
++static void vtpm_state_init(struct vtpm_state *vtpms)
++{
++ vtpms->current_request = NULL;
++ spin_lock_init(&vtpms->req_list_lock);
++ init_waitqueue_head(&vtpms->req_wait_queue);
++ INIT_LIST_HEAD(&vtpms->queued_requests);
++
++ vtpms->current_response = NULL;
++ spin_lock_init(&vtpms->resp_list_lock);
++ init_waitqueue_head(&vtpms->resp_wait_queue);
++
++ vtpms->disconnect_time = jiffies;
++}
++
++
++static inline struct transmission *transmission_alloc(void)
++{
++ return kzalloc(sizeof(struct transmission), GFP_ATOMIC);
++}
++
++static unsigned char *
++transmission_set_req_buffer(struct transmission *t,
++ unsigned char *buffer, size_t len)
++{
++ if (t->request_buflen < len) {
++ kfree(t->request);
++ t->request = kmalloc(len, GFP_KERNEL);
++ if (!t->request) {
++ t->request_buflen = 0;
++ return NULL;
++ }
++ t->request_buflen = len;
++ }
++
++ memcpy(t->request, buffer, len);
++ t->request_len = len;
++
++ return t->request;
++}
++
++static unsigned char *
++transmission_set_res_buffer(struct transmission *t,
++ const unsigned char *buffer, size_t len)
++{
++ if (t->response_buflen < len) {
++ kfree(t->response);
++ t->response = kmalloc(len, GFP_ATOMIC);
++ if (!t->response) {
++ t->response_buflen = 0;
++ return NULL;
++ }
++ t->response_buflen = len;
++ }
++
++ memcpy(t->response, buffer, len);
++ t->response_len = len;
++
++ return t->response;
++}
++
++static inline void transmission_free(struct transmission *t)
++{
++ kfree(t->request);
++ kfree(t->response);
++ kfree(t);
++}
++
++/* =============================================================
++ * Interface with the lower layer driver
++ * =============================================================
++ */
++/*
++ * Lower layer uses this function to make a response available.
++ */
++int vtpm_vd_recv(const struct tpm_chip *chip,
++ const unsigned char *buffer, size_t count,
++ void *ptr)
++{
++ unsigned long flags;
++ int ret_size = 0;
++ struct transmission *t;
++ struct vtpm_state *vtpms;
++
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ /*
++ * The list with requests must contain one request
++ * only and the element there must be the one that
++ * was passed to me from the front-end.
++ */
++ spin_lock_irqsave(&vtpms->resp_list_lock, flags);
++ if (vtpms->current_request != ptr) {
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ return 0;
++ }
++
++ if ((t = vtpms->current_request)) {
++ transmission_free(t);
++ vtpms->current_request = NULL;
++ }
++
++ t = transmission_alloc();
++ if (t) {
++ if (!transmission_set_res_buffer(t, buffer, count)) {
++ transmission_free(t);
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ return -ENOMEM;
++ }
++ ret_size = count;
++ vtpms->current_response = t;
++ wake_up_interruptible(&vtpms->resp_wait_queue);
++ }
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++
++ return ret_size;
++}
++
++
++/*
++ * Lower layer indicates its status (connected/disconnected)
++ */
++void vtpm_vd_status(const struct tpm_chip *chip, u8 vd_status)
++{
++ struct vtpm_state *vtpms;
++
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ vtpms->vd_status = vd_status;
++ if ((vtpms->vd_status & TPM_VD_STATUS_CONNECTED) == 0) {
++ vtpms->disconnect_time = jiffies;
++ }
++}
++
++/* =============================================================
++ * Interface with the generic TPM driver
++ * =============================================================
++ */
++static int vtpm_recv(struct tpm_chip *chip, u8 *buf, size_t count)
++{
++ int rc = 0;
++ unsigned long flags;
++ struct vtpm_state *vtpms;
++
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ /*
++ * Check if the previous operation only queued the command
++ * In this case there won't be a response, so I just
++ * return from here and reset that flag. In any other
++ * case I should receive a response from the back-end.
++ */
++ spin_lock_irqsave(&vtpms->resp_list_lock, flags);
++ if ((vtpms->flags & DATAEX_FLAG_QUEUED_ONLY) != 0) {
++ vtpms->flags &= ~DATAEX_FLAG_QUEUED_ONLY;
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ /*
++ * The first few commands (measurements) must be
++ * queued since it might not be possible to talk to the
++ * TPM, yet.
++ * Return a response of up to 30 '0's.
++ */
++
++ count = min_t(size_t, count, 30);
++ memset(buf, 0x0, count);
++ return count;
++ }
++ /*
++ * Check whether something is in the responselist and if
++ * there's nothing in the list wait for something to appear.
++ */
++
++ if (!vtpms->current_response) {
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ interruptible_sleep_on_timeout(&vtpms->resp_wait_queue,
++ 1000);
++ spin_lock_irqsave(&vtpms->resp_list_lock ,flags);
++ }
++
++ if (vtpms->current_response) {
++ struct transmission *t = vtpms->current_response;
++ vtpms->current_response = NULL;
++ rc = min(count, t->response_len);
++ memcpy(buf, t->response, rc);
++ transmission_free(t);
++ }
++
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ return rc;
++}
++
++static int vtpm_send(struct tpm_chip *chip, u8 *buf, size_t count)
++{
++ int rc = 0;
++ unsigned long flags;
++ struct transmission *t = transmission_alloc();
++ struct vtpm_state *vtpms;
++
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ if (!t)
++ return -ENOMEM;
++ /*
++ * If there's a current request, it must be the
++ * previous request that has timed out.
++ */
++ spin_lock_irqsave(&vtpms->req_list_lock, flags);
++ if (vtpms->current_request != NULL) {
++ printk("WARNING: Sending although there is a request outstanding.\n"
++ " Previous request must have timed out.\n");
++ transmission_free(vtpms->current_request);
++ vtpms->current_request = NULL;
++ }
++ spin_unlock_irqrestore(&vtpms->req_list_lock, flags);
++
++ /*
++ * Queue the packet if the driver below is not
++ * ready, yet, or there is any packet already
++ * in the queue.
++ * If the driver below is ready, unqueue all
++ * packets first before sending our current
++ * packet.
++ * For each unqueued packet, except for the
++ * last (=current) packet, call the function
++ * tpm_xen_recv to wait for the response to come
++ * back.
++ */
++ if ((vtpms->vd_status & TPM_VD_STATUS_CONNECTED) == 0) {
++ if (time_after(jiffies,
++ vtpms->disconnect_time + HZ * 10)) {
++ rc = -ENOENT;
++ } else {
++ goto queue_it;
++ }
++ } else {
++ /*
++ * Send all queued packets.
++ */
++ if (_vtpm_send_queued(chip) == 0) {
++
++ vtpms->current_request = t;
++
++ rc = vtpm_vd_send(vtpms->tpm_private,
++ buf,
++ count,
++ t);
++ /*
++ * The generic TPM driver will call
++ * the function to receive the response.
++ */
++ if (rc < 0) {
++ vtpms->current_request = NULL;
++ goto queue_it;
++ }
++ } else {
++queue_it:
++ if (!transmission_set_req_buffer(t, buf, count)) {
++ transmission_free(t);
++ rc = -ENOMEM;
++ goto exit;
++ }
++ /*
++ * An error occurred. Don't event try
++ * to send the current request. Just
++ * queue it.
++ */
++ spin_lock_irqsave(&vtpms->req_list_lock, flags);
++ vtpms->flags |= DATAEX_FLAG_QUEUED_ONLY;
++ list_add_tail(&t->next, &vtpms->queued_requests);
++ spin_unlock_irqrestore(&vtpms->req_list_lock, flags);
++ }
++ }
++
++exit:
++ return rc;
++}
++
++
++/*
++ * Send all queued requests.
++ */
++static int _vtpm_send_queued(struct tpm_chip *chip)
++{
++ int rc;
++ int error = 0;
++ long flags;
++ unsigned char buffer[1];
++ struct vtpm_state *vtpms;
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ spin_lock_irqsave(&vtpms->req_list_lock, flags);
++
++ while (!list_empty(&vtpms->queued_requests)) {
++ /*
++ * Need to dequeue them.
++ * Read the result into a dummy buffer.
++ */
++ struct transmission *qt = (struct transmission *)
++ vtpms->queued_requests.next;
++ list_del(&qt->next);
++ vtpms->current_request = qt;
++ spin_unlock_irqrestore(&vtpms->req_list_lock, flags);
++
++ rc = vtpm_vd_send(vtpms->tpm_private,
++ qt->request,
++ qt->request_len,
++ qt);
++
++ if (rc < 0) {
++ spin_lock_irqsave(&vtpms->req_list_lock, flags);
++ if ((qt = vtpms->current_request) != NULL) {
++ /*
++ * requeue it at the beginning
++ * of the list
++ */
++ list_add(&qt->next,
++ &vtpms->queued_requests);
++ }
++ vtpms->current_request = NULL;
++ error = 1;
++ break;
++ }
++ /*
++ * After this point qt is not valid anymore!
++ * It is freed when the front-end is delivering
++ * the data by calling tpm_recv
++ */
++ /*
++ * Receive response into provided dummy buffer
++ */
++ rc = vtpm_recv(chip, buffer, sizeof(buffer));
++ spin_lock_irqsave(&vtpms->req_list_lock, flags);
++ }
++
++ spin_unlock_irqrestore(&vtpms->req_list_lock, flags);
++
++ return error;
++}
++
++static void vtpm_cancel(struct tpm_chip *chip)
++{
++ unsigned long flags;
++ struct vtpm_state *vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ spin_lock_irqsave(&vtpms->resp_list_lock,flags);
++
++ if (!vtpms->current_response && vtpms->current_request) {
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ interruptible_sleep_on(&vtpms->resp_wait_queue);
++ spin_lock_irqsave(&vtpms->resp_list_lock,flags);
++ }
++
++ if (vtpms->current_response) {
++ struct transmission *t = vtpms->current_response;
++ vtpms->current_response = NULL;
++ transmission_free(t);
++ }
++
++ spin_unlock_irqrestore(&vtpms->resp_list_lock,flags);
++}
++
++static u8 vtpm_status(struct tpm_chip *chip)
++{
++ u8 rc = 0;
++ unsigned long flags;
++ struct vtpm_state *vtpms;
++
++ vtpms = (struct vtpm_state *)chip_get_private(chip);
++
++ spin_lock_irqsave(&vtpms->resp_list_lock, flags);
++ /*
++ * Data are available if:
++ * - there's a current response
++ * - the last packet was queued only (this is fake, but necessary to
++ * get the generic TPM layer to call the receive function.)
++ */
++ if (vtpms->current_response ||
++ 0 != (vtpms->flags & DATAEX_FLAG_QUEUED_ONLY)) {
++ rc = STATUS_DATA_AVAIL;
++ } else if (!vtpms->current_response && !vtpms->current_request) {
++ rc = STATUS_READY;
++ }
++
++ spin_unlock_irqrestore(&vtpms->resp_list_lock, flags);
++ return rc;
++}
++
++static struct file_operations vtpm_ops = {
++ .owner = THIS_MODULE,
++ .llseek = no_llseek,
++ .open = tpm_open,
++ .read = tpm_read,
++ .write = tpm_write,
++ .release = tpm_release,
++};
++
++static DEVICE_ATTR(pubek, S_IRUGO, tpm_show_pubek, NULL);
++static DEVICE_ATTR(pcrs, S_IRUGO, tpm_show_pcrs, NULL);
++static DEVICE_ATTR(enabled, S_IRUGO, tpm_show_enabled, NULL);
++static DEVICE_ATTR(active, S_IRUGO, tpm_show_active, NULL);
++static DEVICE_ATTR(owned, S_IRUGO, tpm_show_owned, NULL);
++static DEVICE_ATTR(temp_deactivated, S_IRUGO, tpm_show_temp_deactivated,
++ NULL);
++static DEVICE_ATTR(caps, S_IRUGO, tpm_show_caps, NULL);
++static DEVICE_ATTR(cancel, S_IWUSR |S_IWGRP, NULL, tpm_store_cancel);
++
++static struct attribute *vtpm_attrs[] = {
++ &dev_attr_pubek.attr,
++ &dev_attr_pcrs.attr,
++ &dev_attr_enabled.attr,
++ &dev_attr_active.attr,
++ &dev_attr_owned.attr,
++ &dev_attr_temp_deactivated.attr,
++ &dev_attr_caps.attr,
++ &dev_attr_cancel.attr,
++ NULL,
++};
++
++static struct attribute_group vtpm_attr_grp = { .attrs = vtpm_attrs };
++
++#define TPM_LONG_TIMEOUT (10 * 60 * HZ)
++
++static struct tpm_vendor_specific tpm_vtpm = {
++ .recv = vtpm_recv,
++ .send = vtpm_send,
++ .cancel = vtpm_cancel,
++ .status = vtpm_status,
++ .req_complete_mask = STATUS_BUSY | STATUS_DATA_AVAIL,
++ .req_complete_val = STATUS_DATA_AVAIL,
++ .req_canceled = STATUS_READY,
++ .attr_group = &vtpm_attr_grp,
++ .miscdev = {
++ .fops = &vtpm_ops,
++ },
++ .duration = {
++ TPM_LONG_TIMEOUT,
++ TPM_LONG_TIMEOUT,
++ TPM_LONG_TIMEOUT,
++ },
++};
++
++struct tpm_chip *init_vtpm(struct device *dev,
++ struct tpm_virtual_device *tvd,
++ struct tpm_private *tp)
++{
++ long rc;
++ struct tpm_chip *chip;
++ struct vtpm_state *vtpms;
++
++ vtpms = kzalloc(sizeof(struct vtpm_state), GFP_KERNEL);
++ if (!vtpms)
++ return ERR_PTR(-ENOMEM);
++
++ vtpm_state_init(vtpms);
++ vtpms->tpmvd = tvd;
++ vtpms->tpm_private = tp;
++
++ if (tvd)
++ tpm_vtpm.buffersize = tvd->max_tx_size;
++
++ chip = tpm_register_hardware(dev, &tpm_vtpm);
++ if (!chip) {
++ rc = -ENODEV;
++ goto err_free_mem;
++ }
++
++ chip_set_private(chip, vtpms);
++
++ return chip;
++
++err_free_mem:
++ kfree(vtpms);
++
++ return ERR_PTR(rc);
++}
++
++void cleanup_vtpm(struct device *dev)
++{
++ struct tpm_chip *chip = dev_get_drvdata(dev);
++ struct vtpm_state *vtpms = (struct vtpm_state*)chip_get_private(chip);
++ tpm_remove_hardware(dev);
++ kfree(vtpms);
++}
+diff -urNp linux-2.6/drivers/char/tpm/tpm_vtpm.h new/drivers/char/tpm/tpm_vtpm.h
+--- linux-2.6/drivers/char/tpm/tpm_vtpm.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/char/tpm/tpm_vtpm.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,68 @@
++#ifndef TPM_VTPM_H
++#define TPM_VTPM_H
++
++struct tpm_chip;
++struct tpm_private;
++
++struct tpm_virtual_device {
++ /*
++ * This field indicates the maximum size the driver can
++ * transfer in one chunk. It is filled in by the front-end
++ * driver and should be propagated to the generic tpm driver
++ * for allocation of buffers.
++ */
++ unsigned int max_tx_size;
++};
++
++struct vtpm_state {
++ struct transmission *current_request;
++ spinlock_t req_list_lock;
++ wait_queue_head_t req_wait_queue;
++
++ struct list_head queued_requests;
++
++ struct transmission *current_response;
++ spinlock_t resp_list_lock;
++ wait_queue_head_t resp_wait_queue; // processes waiting for responses
++
++ u8 vd_status;
++ u8 flags;
++
++ unsigned long disconnect_time;
++
++ struct tpm_virtual_device *tpmvd;
++
++ /*
++ * The following is a private structure of the underlying
++ * driver. It is passed as parameter in the send function.
++ */
++ struct tpm_private *tpm_private;
++};
++
++
++enum vdev_status {
++ TPM_VD_STATUS_DISCONNECTED = 0x0,
++ TPM_VD_STATUS_CONNECTED = 0x1
++};
++
++/* this function is called from tpm_vtpm.c */
++int vtpm_vd_send(struct tpm_private * tp,
++ const u8 * buf, size_t count, void *ptr);
++
++/* these functions are offered by tpm_vtpm.c */
++struct tpm_chip *init_vtpm(struct device *,
++ struct tpm_virtual_device *,
++ struct tpm_private *);
++void cleanup_vtpm(struct device *);
++int vtpm_vd_recv(const struct tpm_chip* chip,
++ const unsigned char *buffer, size_t count, void *ptr);
++void vtpm_vd_status(const struct tpm_chip *, u8 status);
++
++static inline struct tpm_private *tpm_private_from_dev(struct device *dev)
++{
++ struct tpm_chip *chip = dev_get_drvdata(dev);
++ struct vtpm_state *vtpms = chip_get_private(chip);
++ return vtpms->tpm_private;
++}
++
++#endif
+diff -urNp linux-2.6/drivers/char/tpm/tpm_xen.c new/drivers/char/tpm/tpm_xen.c
+--- linux-2.6/drivers/char/tpm/tpm_xen.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/char/tpm/tpm_xen.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,758 @@
++/*
++ * Copyright (c) 2005, IBM Corporation
++ *
++ * Author: Stefan Berger, stefanb@us.ibm.com
++ * Grant table support: Mahadevan Gomathisankaran
++ *
++ * This code has been derived from drivers/xen/netfront/netfront.c
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/errno.h>
++#include <linux/err.h>
++#include <linux/interrupt.h>
++#include <linux/mutex.h>
++#include <asm/uaccess.h>
++#include <xen/evtchn.h>
++#include <xen/interface/grant_table.h>
++#include <xen/interface/io/tpmif.h>
++#include <xen/xenbus.h>
++#include "tpm.h"
++#include "tpm_vtpm.h"
++
++#undef DEBUG
++
++/* local structures */
++struct tpm_private {
++ struct tpm_chip *chip;
++
++ tpmif_tx_interface_t *tx;
++ atomic_t refcnt;
++ unsigned int evtchn;
++ unsigned int irq;
++ u8 is_connected;
++ u8 is_suspended;
++
++ spinlock_t tx_lock;
++
++ struct tx_buffer *tx_buffers[TPMIF_TX_RING_SIZE];
++
++ atomic_t tx_busy;
++ void *tx_remember;
++
++ domid_t backend_id;
++ wait_queue_head_t wait_q;
++
++ struct xenbus_device *dev;
++ int ring_ref;
++};
++
++struct tx_buffer {
++ unsigned int size; // available space in data
++ unsigned int len; // used space in data
++ unsigned char *data; // pointer to a page
++};
++
++
++/* locally visible variables */
++static grant_ref_t gref_head;
++static struct tpm_private *my_priv;
++
++/* local function prototypes */
++static irqreturn_t tpmif_int(int irq,
++ void *tpm_priv,
++ struct pt_regs *ptregs);
++static void tpmif_rx_action(unsigned long unused);
++static int tpmif_connect(struct xenbus_device *dev,
++ struct tpm_private *tp,
++ domid_t domid);
++static DECLARE_TASKLET(tpmif_rx_tasklet, tpmif_rx_action, 0);
++static int tpmif_allocate_tx_buffers(struct tpm_private *tp);
++static void tpmif_free_tx_buffers(struct tpm_private *tp);
++static void tpmif_set_connected_state(struct tpm_private *tp,
++ u8 newstate);
++static int tpm_xmit(struct tpm_private *tp,
++ const u8 * buf, size_t count, int userbuffer,
++ void *remember);
++static void destroy_tpmring(struct tpm_private *tp);
++void __exit tpmif_exit(void);
++
++#define DPRINTK(fmt, args...) \
++ pr_debug("xen_tpm_fr (%s:%d) " fmt, __FUNCTION__, __LINE__, ##args)
++#define IPRINTK(fmt, args...) \
++ printk(KERN_INFO "xen_tpm_fr: " fmt, ##args)
++#define WPRINTK(fmt, args...) \
++ printk(KERN_WARNING "xen_tpm_fr: " fmt, ##args)
++
++#define GRANT_INVALID_REF 0
++
++
++static inline int
++tx_buffer_copy(struct tx_buffer *txb, const u8 * src, int len,
++ int isuserbuffer)
++{
++ int copied = len;
++
++ if (len > txb->size) {
++ copied = txb->size;
++ }
++ if (isuserbuffer) {
++ if (copy_from_user(txb->data, src, copied))
++ return -EFAULT;
++ } else {
++ memcpy(txb->data, src, copied);
++ }
++ txb->len = len;
++ return copied;
++}
++
++static inline struct tx_buffer *tx_buffer_alloc(void)
++{
++ struct tx_buffer *txb = kzalloc(sizeof (struct tx_buffer),
++ GFP_KERNEL);
++
++ if (txb) {
++ txb->len = 0;
++ txb->size = PAGE_SIZE;
++ txb->data = (unsigned char *)__get_free_page(GFP_KERNEL);
++ if (txb->data == NULL) {
++ kfree(txb);
++ txb = NULL;
++ }
++ }
++ return txb;
++}
++
++
++static inline void tx_buffer_free(struct tx_buffer *txb)
++{
++ if (txb) {
++ free_page((long)txb->data);
++ kfree(txb);
++ }
++}
++
++/**************************************************************
++ Utility function for the tpm_private structure
++**************************************************************/
++static inline void tpm_private_init(struct tpm_private *tp)
++{
++ spin_lock_init(&tp->tx_lock);
++ init_waitqueue_head(&tp->wait_q);
++ atomic_set(&tp->refcnt, 1);
++}
++
++static inline void tpm_private_put(void)
++{
++ if ( atomic_dec_and_test(&my_priv->refcnt)) {
++ tpmif_free_tx_buffers(my_priv);
++ kfree(my_priv);
++ my_priv = NULL;
++ }
++}
++
++static struct tpm_private *tpm_private_get(void)
++{
++ int err;
++ if (!my_priv) {
++ my_priv = kzalloc(sizeof(struct tpm_private), GFP_KERNEL);
++ if (my_priv) {
++ tpm_private_init(my_priv);
++ err = tpmif_allocate_tx_buffers(my_priv);
++ if (err < 0) {
++ tpm_private_put();
++ }
++ }
++ } else {
++ atomic_inc(&my_priv->refcnt);
++ }
++ return my_priv;
++}
++
++/**************************************************************
++
++ The interface to let the tpm plugin register its callback
++ function and send data to another partition using this module
++
++**************************************************************/
++
++static DEFINE_MUTEX(suspend_lock);
++/*
++ * Send data via this module by calling this function
++ */
++int vtpm_vd_send(struct tpm_private *tp,
++ const u8 * buf, size_t count, void *ptr)
++{
++ int sent;
++
++ mutex_lock(&suspend_lock);
++ sent = tpm_xmit(tp, buf, count, 0, ptr);
++ mutex_unlock(&suspend_lock);
++
++ return sent;
++}
++
++/**************************************************************
++ XENBUS support code
++**************************************************************/
++
++static int setup_tpmring(struct xenbus_device *dev,
++ struct tpm_private *tp)
++{
++ tpmif_tx_interface_t *sring;
++ int err;
++
++ tp->ring_ref = GRANT_INVALID_REF;
++
++ sring = (void *)__get_free_page(GFP_KERNEL);
++ if (!sring) {
++ xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring");
++ return -ENOMEM;
++ }
++ tp->tx = sring;
++
++ err = xenbus_grant_ring(dev, virt_to_mfn(tp->tx));
++ if (err < 0) {
++ free_page((unsigned long)sring);
++ tp->tx = NULL;
++ xenbus_dev_fatal(dev, err, "allocating grant reference");
++ goto fail;
++ }
++ tp->ring_ref = err;
++
++ err = tpmif_connect(dev, tp, dev->otherend_id);
++ if (err)
++ goto fail;
++
++ return 0;
++fail:
++ destroy_tpmring(tp);
++ return err;
++}
++
++
++static void destroy_tpmring(struct tpm_private *tp)
++{
++ tpmif_set_connected_state(tp, 0);
++
++ if (tp->ring_ref != GRANT_INVALID_REF) {
++ gnttab_end_foreign_access(tp->ring_ref, 0,
++ (unsigned long)tp->tx);
++ tp->ring_ref = GRANT_INVALID_REF;
++ tp->tx = NULL;
++ }
++
++ if (tp->irq)
++ unbind_from_irqhandler(tp->irq, tp);
++
++ tp->evtchn = tp->irq = 0;
++}
++
++
++static int talk_to_backend(struct xenbus_device *dev,
++ struct tpm_private *tp)
++{
++ const char *message = NULL;
++ int err;
++ struct xenbus_transaction xbt;
++
++ err = setup_tpmring(dev, tp);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "setting up ring");
++ goto out;
++ }
++
++again:
++ err = xenbus_transaction_start(&xbt);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "starting transaction");
++ goto destroy_tpmring;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename,
++ "ring-ref","%u", tp->ring_ref);
++ if (err) {
++ message = "writing ring-ref";
++ goto abort_transaction;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename,
++ "event-channel", "%u", tp->evtchn);
++ if (err) {
++ message = "writing event-channel";
++ goto abort_transaction;
++ }
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err == -EAGAIN)
++ goto again;
++ if (err) {
++ xenbus_dev_fatal(dev, err, "completing transaction");
++ goto destroy_tpmring;
++ }
++
++ xenbus_switch_state(dev, XenbusStateConnected);
++
++ return 0;
++
++abort_transaction:
++ xenbus_transaction_end(xbt, 1);
++ if (message)
++ xenbus_dev_error(dev, err, "%s", message);
++destroy_tpmring:
++ destroy_tpmring(tp);
++out:
++ return err;
++}
++
++/**
++ * Callback received when the backend's state changes.
++ */
++static void backend_changed(struct xenbus_device *dev,
++ enum xenbus_state backend_state)
++{
++ struct tpm_private *tp = tpm_private_from_dev(&dev->dev);
++ DPRINTK("\n");
++
++ switch (backend_state) {
++ case XenbusStateInitialising:
++ case XenbusStateInitWait:
++ case XenbusStateInitialised:
++ case XenbusStateUnknown:
++ break;
++
++ case XenbusStateConnected:
++ tpmif_set_connected_state(tp, 1);
++ break;
++
++ case XenbusStateClosing:
++ tpmif_set_connected_state(tp, 0);
++ break;
++
++ case XenbusStateClosed:
++ if (tp->is_suspended == 0) {
++ device_unregister(&dev->dev);
++ }
++ xenbus_switch_state(dev, XenbusStateClosed);
++ break;
++ }
++}
++
++struct tpm_virtual_device tvd = {
++ .max_tx_size = PAGE_SIZE * TPMIF_TX_RING_SIZE,
++};
++
++static int tpmfront_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err;
++ int handle;
++ struct tpm_private *tp = tpm_private_get();
++
++ if (!tp)
++ return -ENOMEM;
++
++ tp->chip = init_vtpm(&dev->dev, &tvd, tp);
++
++ if (IS_ERR(tp->chip)) {
++ return PTR_ERR(tp->chip);
++ }
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename,
++ "handle", "%i", &handle);
++ if (XENBUS_EXIST_ERR(err))
++ return err;
++
++ if (err < 0) {
++ xenbus_dev_fatal(dev,err,"reading virtual-device");
++ return err;
++ }
++
++ tp->dev = dev;
++
++ err = talk_to_backend(dev, tp);
++ if (err) {
++ tpm_private_put();
++ return err;
++ }
++ return 0;
++}
++
++
++static int tpmfront_remove(struct xenbus_device *dev)
++{
++ struct tpm_private *tp = tpm_private_from_dev(&dev->dev);
++ destroy_tpmring(tp);
++ cleanup_vtpm(&dev->dev);
++ return 0;
++}
++
++static int tpmfront_suspend(struct xenbus_device *dev)
++{
++ struct tpm_private *tp = tpm_private_from_dev(&dev->dev);
++ u32 ctr;
++ /* lock, so no app can send */
++ mutex_lock(&suspend_lock);
++ tp->is_suspended = 1;
++
++ for (ctr = 0; atomic_read(&tp->tx_busy) && ctr <= 25; ctr++) {
++ if ((ctr % 10) == 0)
++ printk("TPM-FE [INFO]: Waiting for outstanding request.\n");
++ /*
++ * Wait for a request to be responded to.
++ */
++ interruptible_sleep_on_timeout(&tp->wait_q, 100);
++ }
++ xenbus_switch_state(dev, XenbusStateClosing);
++
++ if (atomic_read(&tp->tx_busy)) {
++ /*
++ * A temporary work-around.
++ */
++ printk("TPM-FE [WARNING]: Resetting busy flag.");
++ atomic_set(&tp->tx_busy, 0);
++ }
++
++ return 0;
++}
++
++static int tpmfront_resume(struct xenbus_device *dev)
++{
++ struct tpm_private *tp = tpm_private_from_dev(&dev->dev);
++ destroy_tpmring(tp);
++ return talk_to_backend(dev, tp);
++}
++
++static int tpmif_connect(struct xenbus_device *dev,
++ struct tpm_private *tp,
++ domid_t domid)
++{
++ int err;
++
++ tp->backend_id = domid;
++
++ err = xenbus_alloc_evtchn(dev, &tp->evtchn);
++ if (err)
++ return err;
++
++ err = bind_evtchn_to_irqhandler(tp->evtchn,
++ tpmif_int, SA_SAMPLE_RANDOM, "tpmif",
++ tp);
++ if (err <= 0) {
++ WPRINTK("bind_evtchn_to_irqhandler failed (err=%d)\n", err);
++ return err;
++ }
++
++ tp->irq = err;
++ return 0;
++}
++
++static struct xenbus_device_id tpmfront_ids[] = {
++ { "vtpm" },
++ { "" }
++};
++
++static struct xenbus_driver tpmfront = {
++ .name = "vtpm",
++ .owner = THIS_MODULE,
++ .ids = tpmfront_ids,
++ .probe = tpmfront_probe,
++ .remove = tpmfront_remove,
++ .resume = tpmfront_resume,
++ .otherend_changed = backend_changed,
++ .suspend = tpmfront_suspend,
++};
++
++static void __init init_tpm_xenbus(void)
++{
++ xenbus_register_frontend(&tpmfront);
++}
++
++static void __exit exit_tpm_xenbus(void)
++{
++ xenbus_unregister_driver(&tpmfront);
++}
++
++static int tpmif_allocate_tx_buffers(struct tpm_private *tp)
++{
++ unsigned int i;
++
++ for (i = 0; i < TPMIF_TX_RING_SIZE; i++) {
++ tp->tx_buffers[i] = tx_buffer_alloc();
++ if (!tp->tx_buffers[i]) {
++ tpmif_free_tx_buffers(tp);
++ return -ENOMEM;
++ }
++ }
++ return 0;
++}
++
++static void tpmif_free_tx_buffers(struct tpm_private *tp)
++{
++ unsigned int i;
++
++ for (i = 0; i < TPMIF_TX_RING_SIZE; i++) {
++ tx_buffer_free(tp->tx_buffers[i]);
++ }
++}
++
++static void tpmif_rx_action(unsigned long priv)
++{
++ struct tpm_private *tp = (struct tpm_private *)priv;
++
++ int i = 0;
++ unsigned int received;
++ unsigned int offset = 0;
++ u8 *buffer;
++ tpmif_tx_request_t *tx;
++ tx = &tp->tx->ring[i].req;
++
++ atomic_set(&tp->tx_busy, 0);
++ wake_up_interruptible(&tp->wait_q);
++
++ received = tx->size;
++
++ buffer = kmalloc(received, GFP_ATOMIC);
++ if (NULL == buffer) {
++ goto exit;
++ }
++
++ for (i = 0; i < TPMIF_TX_RING_SIZE && offset < received; i++) {
++ struct tx_buffer *txb = tp->tx_buffers[i];
++ tpmif_tx_request_t *tx;
++ unsigned int tocopy;
++
++ tx = &tp->tx->ring[i].req;
++ tocopy = tx->size;
++ if (tocopy > PAGE_SIZE) {
++ tocopy = PAGE_SIZE;
++ }
++
++ memcpy(&buffer[offset], txb->data, tocopy);
++
++ gnttab_release_grant_reference(&gref_head, tx->ref);
++
++ offset += tocopy;
++ }
++
++ vtpm_vd_recv(tp->chip, buffer, received, tp->tx_remember);
++ kfree(buffer);
++
++exit:
++
++ return;
++}
++
++
++static irqreturn_t tpmif_int(int irq, void *tpm_priv, struct pt_regs *ptregs)
++{
++ struct tpm_private *tp = tpm_priv;
++ unsigned long flags;
++
++ spin_lock_irqsave(&tp->tx_lock, flags);
++ tpmif_rx_tasklet.data = (unsigned long)tp;
++ tasklet_schedule(&tpmif_rx_tasklet);
++ spin_unlock_irqrestore(&tp->tx_lock, flags);
++
++ return IRQ_HANDLED;
++}
++
++
++static int tpm_xmit(struct tpm_private *tp,
++ const u8 * buf, size_t count, int isuserbuffer,
++ void *remember)
++{
++ tpmif_tx_request_t *tx;
++ TPMIF_RING_IDX i;
++ unsigned int offset = 0;
++
++ spin_lock_irq(&tp->tx_lock);
++
++ if (unlikely(atomic_read(&tp->tx_busy))) {
++ printk("tpm_xmit: There's an outstanding request/response "
++ "on the way!\n");
++ spin_unlock_irq(&tp->tx_lock);
++ return -EBUSY;
++ }
++
++ if (tp->is_connected != 1) {
++ spin_unlock_irq(&tp->tx_lock);
++ return -EIO;
++ }
++
++ for (i = 0; count > 0 && i < TPMIF_TX_RING_SIZE; i++) {
++ struct tx_buffer *txb = tp->tx_buffers[i];
++ int copied;
++
++ if (NULL == txb) {
++ DPRINTK("txb (i=%d) is NULL. buffers initilized?\n"
++ "Not transmitting anything!\n", i);
++ spin_unlock_irq(&tp->tx_lock);
++ return -EFAULT;
++ }
++ copied = tx_buffer_copy(txb, &buf[offset], count,
++ isuserbuffer);
++ if (copied < 0) {
++ /* An error occurred */
++ spin_unlock_irq(&tp->tx_lock);
++ return copied;
++ }
++ count -= copied;
++ offset += copied;
++
++ tx = &tp->tx->ring[i].req;
++
++ tx->addr = virt_to_machine(txb->data);
++ tx->size = txb->len;
++
++ DPRINTK("First 4 characters sent by TPM-FE are 0x%02x 0x%02x 0x%02x 0x%02x\n",
++ txb->data[0],txb->data[1],txb->data[2],txb->data[3]);
++
++ /* get the granttable reference for this page */
++ tx->ref = gnttab_claim_grant_reference(&gref_head);
++
++ if (-ENOSPC == tx->ref) {
++ spin_unlock_irq(&tp->tx_lock);
++ DPRINTK(" Grant table claim reference failed in func:%s line:%d file:%s\n", __FUNCTION__, __LINE__, __FILE__);
++ return -ENOSPC;
++ }
++ gnttab_grant_foreign_access_ref( tx->ref,
++ tp->backend_id,
++ (tx->addr >> PAGE_SHIFT),
++ 0 /*RW*/);
++ wmb();
++ }
++
++ atomic_set(&tp->tx_busy, 1);
++ tp->tx_remember = remember;
++
++ mb();
++
++ DPRINTK("Notifying backend via event channel %d\n",
++ tp->evtchn);
++
++ notify_remote_via_irq(tp->irq);
++
++ spin_unlock_irq(&tp->tx_lock);
++ return offset;
++}
++
++
++static void tpmif_notify_upperlayer(struct tpm_private *tp)
++{
++ /*
++ * Notify upper layer about the state of the connection
++ * to the BE.
++ */
++ if (tp->is_connected) {
++ vtpm_vd_status(tp->chip, TPM_VD_STATUS_CONNECTED);
++ } else {
++ vtpm_vd_status(tp->chip, TPM_VD_STATUS_DISCONNECTED);
++ }
++}
++
++
++static void tpmif_set_connected_state(struct tpm_private *tp, u8 is_connected)
++{
++ /*
++ * Don't notify upper layer if we are in suspend mode and
++ * should disconnect - assumption is that we will resume
++ * The mutex keeps apps from sending.
++ */
++ if (is_connected == 0 && tp->is_suspended == 1) {
++ return;
++ }
++
++ /*
++ * Unlock the mutex if we are connected again
++ * after being suspended - now resuming.
++ * This also removes the suspend state.
++ */
++ if (is_connected == 1 && tp->is_suspended == 1) {
++ tp->is_suspended = 0;
++ /* unlock, so apps can resume sending */
++ mutex_unlock(&suspend_lock);
++ }
++
++ if (is_connected != tp->is_connected) {
++ tp->is_connected = is_connected;
++ tpmif_notify_upperlayer(tp);
++ }
++}
++
++
++
++/* =================================================================
++ * Initialization function.
++ * =================================================================
++ */
++
++
++static int __init tpmif_init(void)
++{
++ long rc = 0;
++ struct tpm_private *tp;
++
++ if ((xen_start_info->flags & SIF_INITDOMAIN)) {
++ return -EPERM;
++ }
++
++ tp = tpm_private_get();
++ if (!tp) {
++ rc = -ENOMEM;
++ goto failexit;
++ }
++
++ IPRINTK("Initialising the vTPM driver.\n");
++ if ( gnttab_alloc_grant_references ( TPMIF_TX_RING_SIZE,
++ &gref_head ) < 0) {
++ rc = -EFAULT;
++ goto gnttab_alloc_failed;
++ }
++
++ init_tpm_xenbus();
++ return 0;
++
++gnttab_alloc_failed:
++ tpm_private_put();
++failexit:
++
++ return (int)rc;
++}
++
++
++void __exit tpmif_exit(void)
++{
++ exit_tpm_xenbus();
++ tpm_private_put();
++ gnttab_free_grant_references(gref_head);
++}
++
++module_init(tpmif_init);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/char/tty_io.c new/drivers/char/tty_io.c
+--- linux-2.6/drivers/char/tty_io.c 2006-07-03 14:14:34.000000000 +0200
++++ new/drivers/char/tty_io.c 2006-05-27 02:49:07.000000000 +0200
+@@ -132,6 +132,8 @@ LIST_HEAD(tty_drivers); /* linked list
+ vt.c for deeply disgusting hack reasons */
+ DEFINE_MUTEX(tty_mutex);
+
++int console_use_vt = 1;
++
+ #ifdef CONFIG_UNIX98_PTYS
+ extern struct tty_driver *ptm_driver; /* Unix98 pty masters; for /dev/ptmx */
+ extern int pty_limit; /* Config limit on Unix98 ptys */
+@@ -2060,7 +2062,7 @@ retry_open:
+ goto got_driver;
+ }
+ #ifdef CONFIG_VT
+- if (device == MKDEV(TTY_MAJOR,0)) {
++ if (console_use_vt && (device == MKDEV(TTY_MAJOR,0))) {
+ extern struct tty_driver *console_driver;
+ driver = console_driver;
+ index = fg_console;
+@@ -3258,6 +3260,8 @@ static int __init tty_init(void)
+ #endif
+
+ #ifdef CONFIG_VT
++ if (!console_use_vt)
++ goto out_vt;
+ cdev_init(&vc0_cdev, &console_fops);
+ if (cdev_add(&vc0_cdev, MKDEV(TTY_MAJOR, 0), 1) ||
+ register_chrdev_region(MKDEV(TTY_MAJOR, 0), 1, "/dev/vc/0") < 0)
+@@ -3266,6 +3270,7 @@ static int __init tty_init(void)
+ class_device_create(tty_class, NULL, MKDEV(TTY_MAJOR, 0), NULL, "tty0");
+
+ vty_init();
++ out_vt:
+ #endif
+ return 0;
+ }
+diff -urNp linux-2.6/drivers/firmware/Kconfig new/drivers/firmware/Kconfig
+--- linux-2.6/drivers/firmware/Kconfig 2006-07-03 14:14:35.000000000 +0200
++++ new/drivers/firmware/Kconfig 2006-05-09 12:33:37.000000000 +0200
+@@ -8,7 +8,7 @@ menu "Firmware Drivers"
+ config EDD
+ tristate "BIOS Enhanced Disk Drive calls determine boot disk (EXPERIMENTAL)"
+ depends on EXPERIMENTAL
+- depends on !IA64
++ depends on !IA64 && !XEN
+ help
+ Say Y or M here if you want to enable BIOS Enhanced Disk Drive
+ Services real mode BIOS calls to determine which disk
+diff -urNp linux-2.6/drivers/ide/ide-lib.c new/drivers/ide/ide-lib.c
+--- linux-2.6/drivers/ide/ide-lib.c 2006-07-03 14:14:36.000000000 +0200
++++ new/drivers/ide/ide-lib.c 2006-05-27 02:49:07.000000000 +0200
+@@ -410,10 +410,10 @@ void ide_toggle_bounce(ide_drive_t *driv
+ {
+ u64 addr = BLK_BOUNCE_HIGH; /* dma64_addr_t */
+
+- if (!PCI_DMA_BUS_IS_PHYS) {
+- addr = BLK_BOUNCE_ANY;
+- } else if (on && drive->media == ide_disk) {
+- if (HWIF(drive)->pci_dev)
++ if (on && drive->media == ide_disk) {
++ if (!PCI_DMA_BUS_IS_PHYS)
++ addr = BLK_BOUNCE_ANY;
++ else if (HWIF(drive)->pci_dev)
+ addr = HWIF(drive)->pci_dev->dma_mask;
+ }
+
+diff -urNp linux-2.6/drivers/infiniband/ulp/ipoib/ipoib_multicast.c new/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
+--- linux-2.6/drivers/infiniband/ulp/ipoib/ipoib_multicast.c 2006-07-03 14:14:40.000000000 +0200
++++ new/drivers/infiniband/ulp/ipoib/ipoib_multicast.c 2006-07-07 15:21:25.000000000 +0200
+@@ -821,7 +821,8 @@ void ipoib_mcast_restart_task(void *dev_
+
+ ipoib_mcast_stop_thread(dev, 0);
+
+- spin_lock_irqsave(&dev->xmit_lock, flags);
++ local_irq_save(flags);
++ netif_tx_lock(dev);
+ spin_lock(&priv->lock);
+
+ /*
+@@ -896,7 +897,8 @@ void ipoib_mcast_restart_task(void *dev_
+ }
+
+ spin_unlock(&priv->lock);
+- spin_unlock_irqrestore(&dev->xmit_lock, flags);
++ netif_tx_unlock(dev);
++ local_irq_restore(flags);
+
+ /* We have to cancel outside of the spinlock */
+ list_for_each_entry_safe(mcast, tmcast, &remove_list, list) {
+diff -urNp linux-2.6/drivers/Makefile new/drivers/Makefile
+--- linux-2.6/drivers/Makefile 2006-07-03 14:14:31.000000000 +0200
++++ new/drivers/Makefile 2006-05-09 12:33:23.000000000 +0200
+@@ -31,6 +31,7 @@ obj-y += base/ block/ misc/ mfd/ net/
+ obj-$(CONFIG_NUBUS) += nubus/
+ obj-$(CONFIG_ATM) += atm/
+ obj-$(CONFIG_PPC_PMAC) += macintosh/
++obj-$(CONFIG_XEN) += xen/
+ obj-$(CONFIG_IDE) += ide/
+ obj-$(CONFIG_FC4) += fc4/
+ obj-$(CONFIG_SCSI) += scsi/
+diff -urNp linux-2.6/drivers/media/dvb/dvb-core/dvb_net.c new/drivers/media/dvb/dvb-core/dvb_net.c
+--- linux-2.6/drivers/media/dvb/dvb-core/dvb_net.c 2006-07-03 14:14:43.000000000 +0200
++++ new/drivers/media/dvb/dvb-core/dvb_net.c 2006-07-07 15:21:25.000000000 +0200
+@@ -1052,7 +1052,7 @@ static void wq_set_multicast_list (void
+
+ dvb_net_feed_stop(dev);
+ priv->rx_mode = RX_MODE_UNI;
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+
+ if (dev->flags & IFF_PROMISC) {
+ dprintk("%s: promiscuous mode\n", dev->name);
+@@ -1077,7 +1077,7 @@ static void wq_set_multicast_list (void
+ }
+ }
+
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ dvb_net_feed_start(dev);
+ }
+
+diff -urNp linux-2.6/drivers/net/8139cp.c new/drivers/net/8139cp.c
+--- linux-2.6/drivers/net/8139cp.c 2006-07-03 14:14:45.000000000 +0200
++++ new/drivers/net/8139cp.c 2006-07-07 15:21:26.000000000 +0200
+@@ -792,7 +792,7 @@ static int cp_start_xmit (struct sk_buff
+ entry = cp->tx_head;
+ eor = (entry == (CP_TX_RING_SIZE - 1)) ? RingEnd : 0;
+ if (dev->features & NETIF_F_TSO)
+- mss = skb_shinfo(skb)->tso_size;
++ mss = skb_shinfo(skb)->gso_size;
+
+ if (skb_shinfo(skb)->nr_frags == 0) {
+ struct cp_desc *txd = &cp->tx_ring[entry];
+diff -urNp linux-2.6/drivers/net/bnx2.c new/drivers/net/bnx2.c
+--- linux-2.6/drivers/net/bnx2.c 2006-07-03 14:14:45.000000000 +0200
++++ new/drivers/net/bnx2.c 2006-07-07 15:21:26.000000000 +0200
+@@ -1638,7 +1638,7 @@ bnx2_tx_int(struct bnx2 *bp)
+ skb = tx_buf->skb;
+ #ifdef BCM_TSO
+ /* partial BD completions possible with TSO packets */
+- if (skb_shinfo(skb)->tso_size) {
++ if (skb_shinfo(skb)->gso_size) {
+ u16 last_idx, last_ring_idx;
+
+ last_idx = sw_cons +
+@@ -2009,7 +2009,7 @@ bnx2_poll(struct net_device *dev, int *b
+ return 1;
+ }
+
+-/* Called with rtnl_lock from vlan functions and also dev->xmit_lock
++/* Called with rtnl_lock from vlan functions and also netif_tx_lock
+ * from set_multicast.
+ */
+ static void
+@@ -4252,7 +4252,7 @@ bnx2_vlan_rx_kill_vid(struct net_device
+ }
+ #endif
+
+-/* Called with dev->xmit_lock.
++/* Called with netif_tx_lock.
+ * hard_start_xmit is pseudo-lockless - a lock is only required when
+ * the tx queue is full. This way, we get the benefit of lockless
+ * operations most of the time without the complexities to handle
+@@ -4290,7 +4290,7 @@ bnx2_start_xmit(struct sk_buff *skb, str
+ (TX_BD_FLAGS_VLAN_TAG | (vlan_tx_tag_get(skb) << 16));
+ }
+ #ifdef BCM_TSO
+- if ((mss = skb_shinfo(skb)->tso_size) &&
++ if ((mss = skb_shinfo(skb)->gso_size) &&
+ (skb->len > (bp->dev->mtu + ETH_HLEN))) {
+ u32 tcp_opt_len, ip_tcp_len;
+
+diff -urNp linux-2.6/drivers/net/bonding/bond_main.c new/drivers/net/bonding/bond_main.c
+--- linux-2.6/drivers/net/bonding/bond_main.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/bonding/bond_main.c 2006-07-07 15:21:27.000000000 +0200
+@@ -1199,8 +1199,7 @@ int bond_sethwaddr(struct net_device *bo
+ }
+
+ #define BOND_INTERSECT_FEATURES \
+- (NETIF_F_SG|NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM|\
+- NETIF_F_TSO|NETIF_F_UFO)
++ (NETIF_F_SG | NETIF_F_ALL_CSUM | NETIF_F_TSO | NETIF_F_UFO)
+
+ /*
+ * Compute the common dev->feature set available to all slaves. Some
+@@ -1218,9 +1217,7 @@ static int bond_compute_features(struct
+ features &= (slave->dev->features & BOND_INTERSECT_FEATURES);
+
+ if ((features & NETIF_F_SG) &&
+- !(features & (NETIF_F_IP_CSUM |
+- NETIF_F_NO_CSUM |
+- NETIF_F_HW_CSUM)))
++ !(features & NETIF_F_ALL_CSUM))
+ features &= ~NETIF_F_SG;
+
+ /*
+@@ -4191,7 +4188,7 @@ static int bond_init(struct net_device *
+ */
+ bond_dev->features |= NETIF_F_VLAN_CHALLENGED;
+
+- /* don't acquire bond device's xmit_lock when
++ /* don't acquire bond device's netif_tx_lock when
+ * transmitting */
+ bond_dev->features |= NETIF_F_LLTX;
+
+diff -urNp linux-2.6/drivers/net/chelsio/sge.c new/drivers/net/chelsio/sge.c
+--- linux-2.6/drivers/net/chelsio/sge.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/chelsio/sge.c 2006-07-07 15:21:27.000000000 +0200
+@@ -1418,7 +1418,7 @@ int t1_start_xmit(struct sk_buff *skb, s
+ struct cpl_tx_pkt *cpl;
+
+ #ifdef NETIF_F_TSO
+- if (skb_shinfo(skb)->tso_size) {
++ if (skb_shinfo(skb)->gso_size) {
+ int eth_type;
+ struct cpl_tx_pkt_lso *hdr;
+
+@@ -1433,7 +1433,7 @@ int t1_start_xmit(struct sk_buff *skb, s
+ hdr->ip_hdr_words = skb->nh.iph->ihl;
+ hdr->tcp_hdr_words = skb->h.th->doff;
+ hdr->eth_type_mss = htons(MK_ETH_TYPE_MSS(eth_type,
+- skb_shinfo(skb)->tso_size));
++ skb_shinfo(skb)->gso_size));
+ hdr->len = htonl(skb->len - sizeof(*hdr));
+ cpl = (struct cpl_tx_pkt *)hdr;
+ sge->stats.tx_lso_pkts++;
+diff -urNp linux-2.6/drivers/net/e1000/e1000_main.c new/drivers/net/e1000/e1000_main.c
+--- linux-2.6/drivers/net/e1000/e1000_main.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/e1000/e1000_main.c 2006-07-07 16:40:08.000000000 +0200
+@@ -2413,7 +2413,7 @@ e1000_tso(struct e1000_adapter *adapter,
+ uint8_t ipcss, ipcso, tucss, tucso, hdr_len;
+ int err;
+
+- if (skb_shinfo(skb)->tso_size) {
++ if (skb_shinfo(skb)->gso_size) {
+ if (skb_header_cloned(skb)) {
+ err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
+ if (err)
+@@ -2421,7 +2421,7 @@ e1000_tso(struct e1000_adapter *adapter,
+ }
+
+ hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2));
+- mss = skb_shinfo(skb)->tso_size;
++ mss = skb_shinfo(skb)->gso_size;
+ if (skb->protocol == ntohs(ETH_P_IP)) {
+ skb->nh.iph->tot_len = 0;
+ skb->nh.iph->check = 0;
+@@ -2538,7 +2538,7 @@ e1000_tx_map(struct e1000_adapter *adapt
+ * tso gets written back prematurely before the data is fully
+ * DMA'd to the controller */
+ if (!skb->data_len && tx_ring->last_tx_tso &&
+- !skb_shinfo(skb)->tso_size) {
++ !skb_shinfo(skb)->gso_size) {
+ tx_ring->last_tx_tso = 0;
+ size -= 4;
+ }
+@@ -2776,7 +2776,7 @@ e1000_xmit_frame(struct sk_buff *skb, st
+ }
+
+ #ifdef NETIF_F_TSO
+- mss = skb_shinfo(skb)->tso_size;
++ mss = skb_shinfo(skb)->gso_size;
+ /* The controller does a simple calculation to
+ * make sure there is enough room in the FIFO before
+ * initiating the DMA for each buffer. The calc is:
+@@ -2826,7 +2826,7 @@ e1000_xmit_frame(struct sk_buff *skb, st
+ #ifdef NETIF_F_TSO
+ /* Controller Erratum workaround */
+ if (!skb->data_len && tx_ring->last_tx_tso &&
+- !skb_shinfo(skb)->tso_size)
++ !skb_shinfo(skb)->gso_size)
+ count++;
+ #endif
+
+diff -urNp linux-2.6/drivers/net/forcedeth.c new/drivers/net/forcedeth.c
+--- linux-2.6/drivers/net/forcedeth.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/forcedeth.c 2006-07-07 15:57:14.000000000 +0200
+@@ -533,9 +533,9 @@ typedef union _ring_type {
+ * critical parts:
+ * - rx is (pseudo-) lockless: it relies on the single-threading provided
+ * by the arch code for interrupts.
+- * - tx setup is lockless: it relies on dev->xmit_lock. Actual submission
++ * - tx setup is lockless: it relies on netif_tx_lock. Actual submission
+ * needs dev->priv->lock :-(
+- * - set_multicast_list: preparation lockless, relies on dev->xmit_lock.
++ * - set_multicast_list: preparation lockless, relies on netif_tx_lock.
+ */
+
+ /* in dev: base, irq */
+@@ -1213,7 +1213,7 @@ static void drain_ring(struct net_device
+
+ /*
+ * nv_start_xmit: dev->hard_start_xmit function
+- * Called with dev->xmit_lock held.
++ * Called with netif_tx_lock held.
+ */
+ static int nv_start_xmit(struct sk_buff *skb, struct net_device *dev)
+ {
+@@ -1303,8 +1303,8 @@ static int nv_start_xmit(struct sk_buff
+ np->tx_skbuff[nr] = skb;
+
+ #ifdef NETIF_F_TSO
+- if (skb_shinfo(skb)->tso_size)
+- tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->tso_size << NV_TX2_TSO_SHIFT);
++ if (skb_shinfo(skb)->gso_size)
++ tx_flags_extra = NV_TX2_TSO | (skb_shinfo(skb)->gso_size << NV_TX2_TSO_SHIFT);
+ else
+ #endif
+ tx_flags_extra = (skb->ip_summed == CHECKSUM_HW ? (NV_TX2_CHECKSUM_L3|NV_TX2_CHECKSUM_L4) : 0);
+@@ -1407,7 +1407,7 @@ static void nv_tx_done(struct net_device
+
+ /*
+ * nv_tx_timeout: dev->tx_timeout function
+- * Called with dev->xmit_lock held.
++ * Called with netif_tx_lock held.
+ */
+ static void nv_tx_timeout(struct net_device *dev)
+ {
+@@ -1736,8 +1736,8 @@ static int nv_change_mtu(struct net_devi
+ * guessed, there is probably a simpler approach.
+ * Changing the MTU is a rare event, it shouldn't matter.
+ */
+- nv_disable_irq(dev);
+- spin_lock_bh(&dev->xmit_lock);
++ disable_irq(dev->irq);
++ netif_tx_lock_bh(dev);
+ spin_lock(&np->lock);
+ /* stop engines */
+ nv_stop_rx(dev);
+@@ -1768,8 +1768,8 @@ static int nv_change_mtu(struct net_devi
+ nv_start_rx(dev);
+ nv_start_tx(dev);
+ spin_unlock(&np->lock);
+- spin_unlock_bh(&dev->xmit_lock);
+- nv_enable_irq(dev);
++ netif_tx_unlock_bh(dev);
++ enable_irq(dev->irq);
+ }
+ return 0;
+ }
+@@ -1803,7 +1803,7 @@ static int nv_set_mac_address(struct net
+ memcpy(dev->dev_addr, macaddr->sa_data, ETH_ALEN);
+
+ if (netif_running(dev)) {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ spin_lock_irq(&np->lock);
+
+ /* stop rx engine */
+@@ -1815,7 +1815,7 @@ static int nv_set_mac_address(struct net
+ /* restart rx engine */
+ nv_start_rx(dev);
+ spin_unlock_irq(&np->lock);
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ } else {
+ nv_copy_mac_to_hw(dev);
+ }
+@@ -1824,7 +1824,7 @@ static int nv_set_mac_address(struct net
+
+ /*
+ * nv_set_multicast: dev->set_multicast function
+- * Called with dev->xmit_lock held.
++ * Called with netif_tx_lock held.
+ */
+ static void nv_set_multicast(struct net_device *dev)
+ {
+diff -urNp linux-2.6/drivers/net/hamradio/6pack.c new/drivers/net/hamradio/6pack.c
+--- linux-2.6/drivers/net/hamradio/6pack.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/hamradio/6pack.c 2006-07-07 15:10:03.000000000 +0200
+@@ -308,9 +308,9 @@ static int sp_set_mac_address(struct net
+ {
+ struct sockaddr_ax25 *sa = addr;
+
+- spin_lock_irq(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+- spin_unlock_irq(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+
+ return 0;
+ }
+@@ -767,9 +767,9 @@ static int sixpack_ioctl(struct tty_stru
+ break;
+ }
+
+- spin_lock_irq(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ memcpy(dev->dev_addr, &addr, AX25_ADDR_LEN);
+- spin_unlock_irq(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+
+ err = 0;
+ break;
+diff -urNp linux-2.6/drivers/net/hamradio/mkiss.c new/drivers/net/hamradio/mkiss.c
+--- linux-2.6/drivers/net/hamradio/mkiss.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/hamradio/mkiss.c 2006-07-07 15:57:37.000000000 +0200
+@@ -357,9 +357,9 @@ static int ax_set_mac_address(struct net
+ {
+ struct sockaddr_ax25 *sa = addr;
+
+- spin_lock_irq(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ memcpy(dev->dev_addr, &sa->sax25_call, AX25_ADDR_LEN);
+- spin_unlock_irq(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+
+ return 0;
+ }
+@@ -886,9 +886,9 @@ static int mkiss_ioctl(struct tty_struct
+ break;
+ }
+
+- spin_lock_irq(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ memcpy(dev->dev_addr, addr, AX25_ADDR_LEN);
+- spin_unlock_irq(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+
+ err = 0;
+ break;
+diff -urNp linux-2.6/drivers/net/ifb.c new/drivers/net/ifb.c
+--- linux-2.6/drivers/net/ifb.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/ifb.c 2006-07-07 15:10:03.000000000 +0200
+@@ -76,13 +76,13 @@ static void ri_tasklet(unsigned long dev
+ dp->st_task_enter++;
+ if ((skb = skb_peek(&dp->tq)) == NULL) {
+ dp->st_txq_refl_try++;
+- if (spin_trylock(&_dev->xmit_lock)) {
++ if (netif_tx_trylock(_dev)) {
+ dp->st_rxq_enter++;
+ while ((skb = skb_dequeue(&dp->rq)) != NULL) {
+ skb_queue_tail(&dp->tq, skb);
+ dp->st_rx2tx_tran++;
+ }
+- spin_unlock(&_dev->xmit_lock);
++ netif_tx_unlock(_dev);
+ } else {
+ /* reschedule */
+ dp->st_rxq_notenter++;
+@@ -110,7 +110,7 @@ static void ri_tasklet(unsigned long dev
+ }
+ }
+
+- if (spin_trylock(&_dev->xmit_lock)) {
++ if (netif_tx_trylock(_dev)) {
+ dp->st_rxq_check++;
+ if ((skb = skb_peek(&dp->rq)) == NULL) {
+ dp->tasklet_pending = 0;
+@@ -118,10 +118,10 @@ static void ri_tasklet(unsigned long dev
+ netif_wake_queue(_dev);
+ } else {
+ dp->st_rxq_rsch++;
+- spin_unlock(&_dev->xmit_lock);
++ netif_tx_unlock(_dev);
+ goto resched;
+ }
+- spin_unlock(&_dev->xmit_lock);
++ netif_tx_unlock(_dev);
+ } else {
+ resched:
+ dp->tasklet_pending = 1;
+diff -urNp linux-2.6/drivers/net/irda/vlsi_ir.c new/drivers/net/irda/vlsi_ir.c
+--- linux-2.6/drivers/net/irda/vlsi_ir.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/irda/vlsi_ir.c 2006-07-07 15:57:37.000000000 +0200
+@@ -959,7 +959,7 @@ static int vlsi_hard_start_xmit(struct s
+ || (now.tv_sec==ready.tv_sec && now.tv_usec>=ready.tv_usec))
+ break;
+ udelay(100);
+- /* must not sleep here - we are called under xmit_lock! */
++ /* must not sleep here - called under netif_tx_lock! */
+ }
+ }
+
+diff -urNp linux-2.6/drivers/net/ixgb/ixgb_main.c new/drivers/net/ixgb/ixgb_main.c
+--- linux-2.6/drivers/net/ixgb/ixgb_main.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/ixgb/ixgb_main.c 2006-07-07 15:57:38.000000000 +0200
+@@ -1168,7 +1168,7 @@ ixgb_tso(struct ixgb_adapter *adapter, s
+ uint16_t ipcse, tucse, mss;
+ int err;
+
+- if(likely(skb_shinfo(skb)->tso_size)) {
++ if(likely(skb_shinfo(skb)->gso_size)) {
+ if (skb_header_cloned(skb)) {
+ err = pskb_expand_head(skb, 0, 0, GFP_ATOMIC);
+ if (err)
+@@ -1176,7 +1176,7 @@ ixgb_tso(struct ixgb_adapter *adapter, s
+ }
+
+ hdr_len = ((skb->h.raw - skb->data) + (skb->h.th->doff << 2));
+- mss = skb_shinfo(skb)->tso_size;
++ mss = skb_shinfo(skb)->gso_size;
+ skb->nh.iph->tot_len = 0;
+ skb->nh.iph->check = 0;
+ skb->h.th->check = ~csum_tcpudp_magic(skb->nh.iph->saddr,
+diff -urNp linux-2.6/drivers/net/loopback.c new/drivers/net/loopback.c
+--- linux-2.6/drivers/net/loopback.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/loopback.c 2006-07-07 15:57:38.000000000 +0200
+@@ -74,7 +74,7 @@ static void emulate_large_send_offload(s
+ struct iphdr *iph = skb->nh.iph;
+ struct tcphdr *th = (struct tcphdr*)(skb->nh.raw + (iph->ihl * 4));
+ unsigned int doffset = (iph->ihl + th->doff) * 4;
+- unsigned int mtu = skb_shinfo(skb)->tso_size + doffset;
++ unsigned int mtu = skb_shinfo(skb)->gso_size + doffset;
+ unsigned int offset = 0;
+ u32 seq = ntohl(th->seq);
+ u16 id = ntohs(iph->id);
+@@ -139,7 +139,7 @@ static int loopback_xmit(struct sk_buff
+ #endif
+
+ #ifdef LOOPBACK_TSO
+- if (skb_shinfo(skb)->tso_size) {
++ if (skb_shinfo(skb)->gso_size) {
+ BUG_ON(skb->protocol != htons(ETH_P_IP));
+ BUG_ON(skb->nh.iph->protocol != IPPROTO_TCP);
+
+diff -urNp linux-2.6/drivers/net/mv643xx_eth.c new/drivers/net/mv643xx_eth.c
+--- linux-2.6/drivers/net/mv643xx_eth.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/mv643xx_eth.c 2006-07-07 15:57:38.000000000 +0200
+@@ -1200,7 +1200,7 @@ static int mv643xx_eth_start_xmit(struct
+ }
+
+ if (has_tiny_unaligned_frags(skb)) {
+- if ((skb_linearize(skb, GFP_ATOMIC) != 0)) {
++ if (__skb_linearize(skb)) {
+ stats->tx_dropped++;
+ printk(KERN_DEBUG "%s: failed to linearize tiny "
+ "unaligned fragment\n", dev->name);
+diff -urNp linux-2.6/drivers/net/natsemi.c new/drivers/net/natsemi.c
+--- linux-2.6/drivers/net/natsemi.c 2006-07-03 14:14:46.000000000 +0200
++++ new/drivers/net/natsemi.c 2006-07-07 15:57:39.000000000 +0200
+@@ -318,12 +318,12 @@ performance critical codepaths:
+ The rx process only runs in the interrupt handler. Access from outside
+ the interrupt handler is only permitted after disable_irq().
+
+-The rx process usually runs under the dev->xmit_lock. If np->intr_tx_reap
++The rx process usually runs under the netif_tx_lock. If np->intr_tx_reap
+ is set, then access is permitted under spin_lock_irq(&np->lock).
+
+ Thus configuration functions that want to access everything must call
+ disable_irq(dev->irq);
+- spin_lock_bh(dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ spin_lock_irq(&np->lock);
+
+ IV. Notes
+diff -urNp linux-2.6/drivers/net/r8169.c new/drivers/net/r8169.c
+--- linux-2.6/drivers/net/r8169.c 2006-07-03 14:14:47.000000000 +0200
++++ new/drivers/net/r8169.c 2006-07-07 15:57:39.000000000 +0200
+@@ -2171,7 +2171,7 @@ static int rtl8169_xmit_frags(struct rtl
+ static inline u32 rtl8169_tso_csum(struct sk_buff *skb, struct net_device *dev)
+ {
+ if (dev->features & NETIF_F_TSO) {
+- u32 mss = skb_shinfo(skb)->tso_size;
++ u32 mss = skb_shinfo(skb)->gso_size;
+
+ if (mss)
+ return LargeSend | ((mss & MSSMask) << MSSShift);
+diff -urNp linux-2.6/drivers/net/s2io.c new/drivers/net/s2io.c
+--- linux-2.6/drivers/net/s2io.c 2006-07-03 14:14:47.000000000 +0200
++++ new/drivers/net/s2io.c 2006-07-07 15:57:39.000000000 +0200
+@@ -3564,8 +3564,8 @@ static int s2io_xmit(struct sk_buff *skb
+ txdp->Control_1 = 0;
+ txdp->Control_2 = 0;
+ #ifdef NETIF_F_TSO
+- mss = skb_shinfo(skb)->tso_size;
+- if (mss) {
++ mss = skb_shinfo(skb)->gso_size;
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_TCPV4) {
+ txdp->Control_1 |= TXD_TCP_LSO_EN;
+ txdp->Control_1 |= TXD_TCP_LSO_MSS(mss);
+ }
+@@ -3585,10 +3585,10 @@ static int s2io_xmit(struct sk_buff *skb
+ }
+
+ frg_len = skb->len - skb->data_len;
+- if (skb_shinfo(skb)->ufo_size) {
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4) {
+ int ufo_size;
+
+- ufo_size = skb_shinfo(skb)->ufo_size;
++ ufo_size = skb_shinfo(skb)->gso_size;
+ ufo_size &= ~7;
+ txdp->Control_1 |= TXD_UFO_EN;
+ txdp->Control_1 |= TXD_UFO_MSS(ufo_size);
+@@ -3614,7 +3614,7 @@ static int s2io_xmit(struct sk_buff *skb
+ txdp->Host_Control = (unsigned long) skb;
+ txdp->Control_1 |= TXD_BUFFER0_SIZE(frg_len);
+
+- if (skb_shinfo(skb)->ufo_size)
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4)
+ txdp->Control_1 |= TXD_UFO_EN;
+
+ frg_cnt = skb_shinfo(skb)->nr_frags;
+@@ -3629,12 +3629,12 @@ static int s2io_xmit(struct sk_buff *skb
+ (sp->pdev, frag->page, frag->page_offset,
+ frag->size, PCI_DMA_TODEVICE);
+ txdp->Control_1 = TXD_BUFFER0_SIZE(frag->size);
+- if (skb_shinfo(skb)->ufo_size)
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4)
+ txdp->Control_1 |= TXD_UFO_EN;
+ }
+ txdp->Control_1 |= TXD_GATHER_CODE_LAST;
+
+- if (skb_shinfo(skb)->ufo_size)
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4)
+ frg_cnt++; /* as Txd0 was used for inband header */
+
+ tx_fifo = mac_control->tx_FIFO_start[queue];
+@@ -3648,7 +3648,7 @@ static int s2io_xmit(struct sk_buff *skb
+ if (mss)
+ val64 |= TX_FIFO_SPECIAL_FUNC;
+ #endif
+- if (skb_shinfo(skb)->ufo_size)
++ if (skb_shinfo(skb)->gso_type == SKB_GSO_UDPV4)
+ val64 |= TX_FIFO_SPECIAL_FUNC;
+ writeq(val64, &tx_fifo->List_Control);
+
+diff -urNp linux-2.6/drivers/net/sky2.c new/drivers/net/sky2.c
+--- linux-2.6/drivers/net/sky2.c 2006-07-03 14:14:47.000000000 +0200
++++ new/drivers/net/sky2.c 2006-07-07 15:57:39.000000000 +0200
+@@ -1160,7 +1160,7 @@ static unsigned tx_le_req(const struct s
+ count = sizeof(dma_addr_t) / sizeof(u32);
+ count += skb_shinfo(skb)->nr_frags * count;
+
+- if (skb_shinfo(skb)->tso_size)
++ if (skb_shinfo(skb)->gso_size)
+ ++count;
+
+ if (skb->ip_summed == CHECKSUM_HW)
+@@ -1232,7 +1232,7 @@ static int sky2_xmit_frame(struct sk_buf
+ }
+
+ /* Check for TCP Segmentation Offload */
+- mss = skb_shinfo(skb)->tso_size;
++ mss = skb_shinfo(skb)->gso_size;
+ if (mss != 0) {
+ /* just drop the packet if non-linear expansion fails */
+ if (skb_header_cloned(skb) &&
+diff -urNp linux-2.6/drivers/net/tg3.c new/drivers/net/tg3.c
+--- linux-2.6/drivers/net/tg3.c 2006-07-03 14:14:47.000000000 +0200
++++ new/drivers/net/tg3.c 2006-07-07 16:41:44.000000000 +0200
+@@ -3743,7 +3743,7 @@ static int tg3_start_xmit(struct sk_buff
+ #if TG3_TSO_SUPPORT != 0
+ mss = 0;
+ if (skb->len > (tp->dev->mtu + ETH_HLEN) &&
+- (mss = skb_shinfo(skb)->tso_size) != 0) {
++ (mss = skb_shinfo(skb)->gso_size) != 0) {
+ int tcp_opt_len, ip_tcp_len;
+
+ if (skb_header_cloned(skb) &&
+@@ -3871,7 +3871,7 @@ static int tg3_start_xmit_dma_bug(struct
+ #if TG3_TSO_SUPPORT != 0
+ mss = 0;
+ if (skb->len > (tp->dev->mtu + ETH_HLEN) &&
+- (mss = skb_shinfo(skb)->tso_size) != 0) {
++ (mss = skb_shinfo(skb)->gso_size) != 0) {
+ int tcp_opt_len, ip_tcp_len;
+
+ if (skb_header_cloned(skb) &&
+diff -urNp linux-2.6/drivers/net/tulip/winbond-840.c new/drivers/net/tulip/winbond-840.c
+--- linux-2.6/drivers/net/tulip/winbond-840.c 2006-07-03 14:14:48.000000000 +0200
++++ new/drivers/net/tulip/winbond-840.c 2006-07-07 15:57:40.000000000 +0200
+@@ -1605,11 +1605,11 @@ static void __devexit w840_remove1 (stru
+ * - get_stats:
+ * spin_lock_irq(np->lock), doesn't touch hw if not present
+ * - hard_start_xmit:
+- * netif_stop_queue + spin_unlock_wait(&dev->xmit_lock);
++ * synchronize_irq + netif_tx_disable;
+ * - tx_timeout:
+- * netif_device_detach + spin_unlock_wait(&dev->xmit_lock);
++ * netif_device_detach + netif_tx_disable;
+ * - set_multicast_list
+- * netif_device_detach + spin_unlock_wait(&dev->xmit_lock);
++ * netif_device_detach + netif_tx_disable;
+ * - interrupt handler
+ * doesn't touch hw if not present, synchronize_irq waits for
+ * running instances of the interrupt handler.
+@@ -1635,11 +1635,10 @@ static int w840_suspend (struct pci_dev
+ netif_device_detach(dev);
+ update_csr6(dev, 0);
+ iowrite32(0, ioaddr + IntrEnable);
+- netif_stop_queue(dev);
+ spin_unlock_irq(&np->lock);
+
+- spin_unlock_wait(&dev->xmit_lock);
+ synchronize_irq(dev->irq);
++ netif_tx_disable(dev);
+
+ np->stats.rx_missed_errors += ioread32(ioaddr + RxMissed) & 0xffff;
+
+diff -urNp linux-2.6/drivers/net/typhoon.c new/drivers/net/typhoon.c
+--- linux-2.6/drivers/net/typhoon.c 2006-07-03 14:14:48.000000000 +0200
++++ new/drivers/net/typhoon.c 2006-07-07 15:57:40.000000000 +0200
+@@ -340,7 +340,7 @@ enum state_values {
+ #endif
+
+ #if defined(NETIF_F_TSO)
+-#define skb_tso_size(x) (skb_shinfo(x)->tso_size)
++#define skb_tso_size(x) (skb_shinfo(x)->gso_size)
+ #define TSO_NUM_DESCRIPTORS 2
+ #define TSO_OFFLOAD_ON TYPHOON_OFFLOAD_TCP_SEGMENT
+ #else
+diff -urNp linux-2.6/drivers/net/via-velocity.c new/drivers/net/via-velocity.c
+--- linux-2.6/drivers/net/via-velocity.c 2006-07-03 14:14:48.000000000 +0200
++++ new/drivers/net/via-velocity.c 2006-07-07 15:10:03.000000000 +0200
+@@ -1899,6 +1899,13 @@ static int velocity_xmit(struct sk_buff
+
+ int pktlen = skb->len;
+
++#ifdef VELOCITY_ZERO_COPY_SUPPORT
++ if (skb_shinfo(skb)->nr_frags > 6 && __skb_linearize(skb)) {
++ kfree_skb(skb);
++ return 0;
++ }
++#endif
++
+ spin_lock_irqsave(&vptr->lock, flags);
+
+ index = vptr->td_curr[qnum];
+@@ -1914,8 +1921,6 @@ static int velocity_xmit(struct sk_buff
+ */
+ if (pktlen < ETH_ZLEN) {
+ /* Cannot occur until ZC support */
+- if(skb_linearize(skb, GFP_ATOMIC))
+- return 0;
+ pktlen = ETH_ZLEN;
+ memcpy(tdinfo->buf, skb->data, skb->len);
+ memset(tdinfo->buf + skb->len, 0, ETH_ZLEN - skb->len);
+@@ -1933,7 +1938,6 @@ static int velocity_xmit(struct sk_buff
+ int nfrags = skb_shinfo(skb)->nr_frags;
+ tdinfo->skb = skb;
+ if (nfrags > 6) {
+- skb_linearize(skb, GFP_ATOMIC);
+ memcpy(tdinfo->buf, skb->data, skb->len);
+ tdinfo->skb_dma[0] = tdinfo->buf_dma;
+ td_ptr->tdesc0.pktsize =
+diff -urNp linux-2.6/drivers/net/wireless/orinoco.c new/drivers/net/wireless/orinoco.c
+--- linux-2.6/drivers/net/wireless/orinoco.c 2006-07-03 14:14:49.000000000 +0200
++++ new/drivers/net/wireless/orinoco.c 2006-07-07 15:57:40.000000000 +0200
+@@ -1833,7 +1833,9 @@ static int __orinoco_program_rids(struct
+ /* Set promiscuity / multicast*/
+ priv->promiscuous = 0;
+ priv->mc_count = 0;
+- __orinoco_set_multicast_list(dev); /* FIXME: what about the xmit_lock */
++
++ /* FIXME: what about netif_tx_lock */
++ __orinoco_set_multicast_list(dev);
+
+ return 0;
+ }
+diff -urNp linux-2.6/drivers/oprofile/buffer_sync.c new/drivers/oprofile/buffer_sync.c
+--- linux-2.6/drivers/oprofile/buffer_sync.c 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/buffer_sync.c 2006-07-07 15:58:36.000000000 +0200
+@@ -6,6 +6,10 @@
+ *
+ * @author John Levon <levon@movementarian.org>
+ *
++ * Modified by Aravind Menon for Xen
++ * These modifications are:
++ * Copyright (C) 2005 Hewlett-Packard Co.
++ *
+ * This is the core of the buffer management. Each
+ * CPU buffer is processed and entered into the
+ * global event buffer. Such processing is necessary
+@@ -275,15 +279,30 @@ static void add_cpu_switch(int i)
+ last_cookie = INVALID_COOKIE;
+ }
+
+-static void add_kernel_ctx_switch(unsigned int in_kernel)
++static void add_cpu_mode_switch(unsigned int cpu_mode)
+ {
+ add_event_entry(ESCAPE_CODE);
+- if (in_kernel)
+- add_event_entry(KERNEL_ENTER_SWITCH_CODE);
+- else
+- add_event_entry(KERNEL_EXIT_SWITCH_CODE);
++ switch (cpu_mode) {
++ case CPU_MODE_USER:
++ add_event_entry(USER_ENTER_SWITCH_CODE);
++ break;
++ case CPU_MODE_KERNEL:
++ add_event_entry(KERNEL_ENTER_SWITCH_CODE);
++ break;
++ case CPU_MODE_XEN:
++ add_event_entry(XEN_ENTER_SWITCH_CODE);
++ break;
++ case CPU_MODE_PASSIVE_START:
++ add_event_entry(PASSIVE_START_CODE);
++ break;
++ case CPU_MODE_PASSIVE_STOP:
++ add_event_entry(PASSIVE_STOP_CODE);
++ break;
++ default:
++ break;
++ }
+ }
+-
++
+ static void
+ add_user_ctx_switch(struct task_struct const * task, unsigned long cookie)
+ {
+@@ -348,9 +367,9 @@ static int add_us_sample(struct mm_struc
+ * for later lookup from userspace.
+ */
+ static int
+-add_sample(struct mm_struct * mm, struct op_sample * s, int in_kernel)
++add_sample(struct mm_struct * mm, struct op_sample * s, int cpu_mode)
+ {
+- if (in_kernel) {
++ if (cpu_mode >= CPU_MODE_KERNEL) {
+ add_sample_entry(s->eip, s->event);
+ return 1;
+ } else if (mm) {
+@@ -496,10 +515,11 @@ void sync_buffer(int cpu)
+ struct mm_struct *mm = NULL;
+ struct task_struct * new;
+ unsigned long cookie = 0;
+- int in_kernel = 1;
++ int cpu_mode = 1;
+ unsigned int i;
+ sync_buffer_state state = sb_buffer_start;
+ unsigned long available;
++ int domain_switch = NO_DOMAIN_SWITCH;
+
+ down(&buffer_sem);
+
+@@ -513,12 +533,19 @@ void sync_buffer(int cpu)
+ struct op_sample * s = &cpu_buf->buffer[cpu_buf->tail_pos];
+
+ if (is_code(s->eip)) {
+- if (s->event <= CPU_IS_KERNEL) {
++ if (s->event < CPU_TRACE_BEGIN) {
+ /* kernel/userspace switch */
+- in_kernel = s->event;
++ cpu_mode = s->event;
+ if (state == sb_buffer_start)
+ state = sb_sample_start;
+- add_kernel_ctx_switch(s->event);
++
++ if (s->event == CPU_MODE_PASSIVE_START)
++ domain_switch = DOMAIN_SWITCH_START_EVENT1;
++ else if (s->event == CPU_MODE_PASSIVE_STOP)
++ domain_switch = DOMAIN_SWITCH_STOP_EVENT1;
++
++ if (domain_switch != DOMAIN_SWITCH_START_EVENT2)
++ add_cpu_mode_switch(s->event);
+ } else if (s->event == CPU_TRACE_BEGIN) {
+ state = sb_bt_start;
+ add_trace_begin();
+@@ -535,11 +562,20 @@ void sync_buffer(int cpu)
+ add_user_ctx_switch(new, cookie);
+ }
+ } else {
+- if (state >= sb_bt_start &&
+- !add_sample(mm, s, in_kernel)) {
+- if (state == sb_bt_start) {
+- state = sb_bt_ignore;
+- atomic_inc(&oprofile_stats.bt_lost_no_mapping);
++ if (domain_switch == DOMAIN_SWITCH_START_EVENT1) {
++ add_event_entry(s->event);
++ domain_switch = DOMAIN_SWITCH_START_EVENT2;
++ } else if (domain_switch == DOMAIN_SWITCH_START_EVENT1) {
++ add_sample_entry(s->eip, s->event);
++ } else if (domain_switch == DOMAIN_SWITCH_STOP_EVENT1) {
++ domain_switch = NO_DOMAIN_SWITCH;
++ } else {
++ if (state >= sb_bt_start &&
++ !add_sample(mm, s, cpu_mode)) {
++ if (state == sb_bt_start) {
++ state = sb_bt_ignore;
++ atomic_inc(&oprofile_stats.bt_lost_no_mapping);
++ }
+ }
+ }
+ }
+diff -urNp linux-2.6/drivers/oprofile/buffer_sync.h new/drivers/oprofile/buffer_sync.h
+--- linux-2.6/drivers/oprofile/buffer_sync.h 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/buffer_sync.h 2006-07-07 15:10:03.000000000 +0200
+@@ -9,6 +9,11 @@
+
+ #ifndef OPROFILE_BUFFER_SYNC_H
+ #define OPROFILE_BUFFER_SYNC_H
++
++#define NO_DOMAIN_SWITCH -1
++#define DOMAIN_SWITCH_START_EVENT1 0
++#define DOMAIN_SWITCH_START_EVENT2 1
++#define DOMAIN_SWITCH_STOP_EVENT1 2
+
+ /* add the necessary profiling hooks */
+ int sync_start(void);
+diff -urNp linux-2.6/drivers/oprofile/cpu_buffer.c new/drivers/oprofile/cpu_buffer.c
+--- linux-2.6/drivers/oprofile/cpu_buffer.c 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/cpu_buffer.c 2006-07-07 16:00:19.000000000 +0200
+@@ -6,6 +6,10 @@
+ *
+ * @author John Levon <levon@movementarian.org>
+ *
++ * Modified by Aravind Menon for Xen
++ * These modifications are:
++ * Copyright (C) 2005 Hewlett-Packard Co.
++ *
+ * Each CPU has a local buffer that stores PC value/event
+ * pairs. We also log context switches when we notice them.
+ * Eventually each CPU's buffer is processed into the global
+@@ -57,7 +61,7 @@ int alloc_cpu_buffers(void)
+ goto fail;
+
+ b->last_task = NULL;
+- b->last_is_kernel = -1;
++ b->last_cpu_mode = -1;
+ b->tracing = 0;
+ b->buffer_size = buffer_size;
+ b->tail_pos = 0;
+@@ -113,7 +117,7 @@ void cpu_buffer_reset(struct oprofile_cp
+ * collected will populate the buffer with proper
+ * values to initialize the buffer
+ */
+- cpu_buf->last_is_kernel = -1;
++ cpu_buf->last_cpu_mode = -1;
+ cpu_buf->last_task = NULL;
+ }
+
+@@ -163,13 +167,13 @@ add_code(struct oprofile_cpu_buffer * bu
+ * because of the head/tail separation of the writer and reader
+ * of the CPU buffer.
+ *
+- * is_kernel is needed because on some architectures you cannot
++ * cpu_mode is needed because on some architectures you cannot
+ * tell if you are in kernel or user space simply by looking at
+- * pc. We tag this in the buffer by generating kernel enter/exit
+- * events whenever is_kernel changes
++ * pc. We tag this in the buffer by generating kernel/user (and xen)
++ * enter events whenever cpu_mode changes
+ */
+ static int log_sample(struct oprofile_cpu_buffer * cpu_buf, unsigned long pc,
+- int is_kernel, unsigned long event)
++ int cpu_mode, unsigned long event)
+ {
+ struct task_struct * task;
+
+@@ -180,16 +184,16 @@ static int log_sample(struct oprofile_cp
+ return 0;
+ }
+
+- is_kernel = !!is_kernel;
++ WARN_ON(cpu_mode > CPU_MODE_XEN);
+
+ task = current;
+
+ /* notice a switch from user->kernel or vice versa */
+- if (cpu_buf->last_is_kernel != is_kernel) {
+- cpu_buf->last_is_kernel = is_kernel;
+- add_code(cpu_buf, is_kernel);
++ if (cpu_buf->last_cpu_mode != cpu_mode) {
++ cpu_buf->last_cpu_mode = cpu_mode;
++ add_code(cpu_buf, cpu_mode);
+ }
+-
++
+ /* notice a task switch */
+ if (cpu_buf->last_task != task) {
+ cpu_buf->last_task = task;
+diff -urNp linux-2.6/drivers/oprofile/cpu_buffer.h new/drivers/oprofile/cpu_buffer.h
+--- linux-2.6/drivers/oprofile/cpu_buffer.h 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/cpu_buffer.h 2006-07-07 16:00:46.000000000 +0200
+@@ -36,7 +36,7 @@ struct oprofile_cpu_buffer {
+ volatile unsigned long tail_pos;
+ unsigned long buffer_size;
+ struct task_struct * last_task;
+- int last_is_kernel;
++ int last_cpu_mode;
+ int tracing;
+ struct op_sample * buffer;
+ unsigned long sample_received;
+@@ -51,7 +51,13 @@ extern struct oprofile_cpu_buffer cpu_bu
+ void cpu_buffer_reset(struct oprofile_cpu_buffer * cpu_buf);
+
+ /* transient events for the CPU buffer -> event buffer */
+-#define CPU_IS_KERNEL 1
+-#define CPU_TRACE_BEGIN 2
++#define CPU_MODE_USER 0
++#define CPU_MODE_KERNEL 1
++#define CPU_MODE_XEN 2
++#define CPU_MODE_PASSIVE_START 3
++#define CPU_MODE_PASSIVE_STOP 4
++#define CPU_TRACE_BEGIN 5
++
++#define IGNORED_PC 0
+
+ #endif /* OPROFILE_CPU_BUFFER_H */
+diff -urNp linux-2.6/drivers/oprofile/event_buffer.h new/drivers/oprofile/event_buffer.h
+--- linux-2.6/drivers/oprofile/event_buffer.h 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/event_buffer.h 2006-07-07 16:01:13.000000000 +0200
+@@ -29,11 +29,14 @@ void wake_up_buffer_waiter(void);
+ #define CPU_SWITCH_CODE 2
+ #define COOKIE_SWITCH_CODE 3
+ #define KERNEL_ENTER_SWITCH_CODE 4
+-#define KERNEL_EXIT_SWITCH_CODE 5
++#define USER_ENTER_SWITCH_CODE 5
+ #define MODULE_LOADED_CODE 6
+ #define CTX_TGID_CODE 7
+ #define TRACE_BEGIN_CODE 8
+ #define TRACE_END_CODE 9
++#define XEN_ENTER_SWITCH_CODE 10
++#define PASSIVE_START_CODE 11
++#define PASSIVE_STOP_CODE 12
+
+ #define INVALID_COOKIE ~0UL
+ #define NO_COOKIE 0UL
+diff -urNp linux-2.6/drivers/oprofile/oprof.c new/drivers/oprofile/oprof.c
+--- linux-2.6/drivers/oprofile/oprof.c 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/oprof.c 2006-07-07 16:01:51.000000000 +0200
+@@ -5,6 +5,10 @@
+ * @remark Read the file COPYING
+ *
+ * @author John Levon <levon@movementarian.org>
++ *
++ * Modified by Aravind Menon for Xen
++ * These modifications are:
++ * Copyright (C) 2005 Hewlett-Packard Co.
+ */
+
+ #include <linux/kernel.h>
+@@ -19,7 +23,7 @@
+ #include "cpu_buffer.h"
+ #include "buffer_sync.h"
+ #include "oprofile_stats.h"
+-
++
+ struct oprofile_operations oprofile_ops;
+
+ unsigned long oprofile_started;
+@@ -33,6 +37,32 @@ static DECLARE_MUTEX(start_sem);
+ */
+ static int timer = 0;
+
++int oprofile_set_active(int active_domains[], unsigned int adomains)
++{
++ int err;
++
++ if (!oprofile_ops.set_active)
++ return -EINVAL;
++
++ down(&start_sem);
++ err = oprofile_ops.set_active(active_domains, adomains);
++ up(&start_sem);
++ return err;
++}
++
++int oprofile_set_passive(int passive_domains[], unsigned int pdomains)
++{
++ int err;
++
++ if (!oprofile_ops.set_passive)
++ return -EINVAL;
++
++ down(&start_sem);
++ err = oprofile_ops.set_passive(passive_domains, pdomains);
++ up(&start_sem);
++ return err;
++}
++
+ int oprofile_setup(void)
+ {
+ int err;
+diff -urNp linux-2.6/drivers/oprofile/oprof.h new/drivers/oprofile/oprof.h
+--- linux-2.6/drivers/oprofile/oprof.h 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/oprof.h 2006-07-07 16:02:16.000000000 +0200
+@@ -35,5 +35,8 @@ void oprofile_create_files(struct super_
+ void oprofile_timer_init(struct oprofile_operations * ops);
+
+ int oprofile_set_backtrace(unsigned long depth);
++
++int oprofile_set_active(int active_domains[], unsigned int adomains);
++int oprofile_set_passive(int passive_domains[], unsigned int pdomains);
+
+ #endif /* OPROF_H */
+diff -urNp linux-2.6/drivers/oprofile/oprofile_files.c new/drivers/oprofile/oprofile_files.c
+--- linux-2.6/drivers/oprofile/oprofile_files.c 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/oprofile/oprofile_files.c 2006-07-07 16:04:26.000000000 +0200
+@@ -5,15 +5,21 @@
+ * @remark Read the file COPYING
+ *
+ * @author John Levon <levon@movementarian.org>
++ *
++ * Modified by Aravind Menon for Xen
++ * These modifications are:
++ * Copyright (C) 2005 Hewlett-Packard Co.
+ */
+
+ #include <linux/fs.h>
+ #include <linux/oprofile.h>
++#include <asm/uaccess.h>
++#include <linux/ctype.h>
+
+ #include "event_buffer.h"
+ #include "oprofile_stats.h"
+ #include "oprof.h"
+-
++
+ unsigned long fs_buffer_size = 131072;
+ unsigned long fs_cpu_buffer_size = 8192;
+ unsigned long fs_buffer_watershed = 32768; /* FIXME: tune */
+@@ -117,11 +123,206 @@ static ssize_t dump_write(struct file *
+ static struct file_operations dump_fops = {
+ .write = dump_write,
+ };
+-
++
++#define TMPBUFSIZE 512
++
++#ifndef MAX_OPROF_DOMAINS
++#define MAX_OPROF_DOMAINS 25
++#endif
++
++static unsigned int adomains = 0;
++static int active_domains[MAX_OPROF_DOMAINS + 1];
++static DEFINE_MUTEX(adom_mutex);
++
++static ssize_t adomain_write(struct file * file, char const __user * buf,
++ size_t count, loff_t * offset)
++{
++ char *tmpbuf;
++ char *startp, *endp;
++ int i;
++ unsigned long val;
++ ssize_t retval = count;
++
++ if (*offset)
++ return -EINVAL;
++ if (count > TMPBUFSIZE - 1)
++ return -EINVAL;
++
++ if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL)))
++ return -ENOMEM;
++
++ if (copy_from_user(tmpbuf, buf, count)) {
++ kfree(tmpbuf);
++ return -EFAULT;
++ }
++ tmpbuf[count] = 0;
++
++ mutex_lock(&adom_mutex);
++
++ startp = tmpbuf;
++ /* Parse one more than MAX_OPROF_DOMAINS, for easy error checking */
++ for (i = 0; i <= MAX_OPROF_DOMAINS; i++) {
++ val = simple_strtoul(startp, &endp, 0);
++ if (endp == startp)
++ break;
++ while (ispunct(*endp) || isspace(*endp))
++ endp++;
++ active_domains[i] = val;
++ if (active_domains[i] != val)
++ /* Overflow, force error below */
++ i = MAX_OPROF_DOMAINS + 1;
++ startp = endp;
++ }
++ /* Force error on trailing junk */
++ adomains = *startp ? MAX_OPROF_DOMAINS + 1 : i;
++
++ kfree(tmpbuf);
++
++ if (adomains > MAX_OPROF_DOMAINS
++ || oprofile_set_active(active_domains, adomains)) {
++ adomains = 0;
++ retval = -EINVAL;
++ }
++
++ mutex_unlock(&adom_mutex);
++ return retval;
++}
++
++static ssize_t adomain_read(struct file * file, char __user * buf,
++ size_t count, loff_t * offset)
++{
++ char * tmpbuf;
++ size_t len;
++ int i;
++ ssize_t retval;
++
++ if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL)))
++ return -ENOMEM;
++
++ mutex_lock(&adom_mutex);
++
++ len = 0;
++ for (i = 0; i < adomains; i++)
++ len += snprintf(tmpbuf + len,
++ len < TMPBUFSIZE ? TMPBUFSIZE - len : 0,
++ "%u ", active_domains[i]);
++ WARN_ON(len > TMPBUFSIZE);
++ if (len != 0 && len <= TMPBUFSIZE)
++ tmpbuf[len-1] = '\n';
++
++ mutex_unlock(&adom_mutex);
++
++ retval = simple_read_from_buffer(buf, count, offset, tmpbuf, len);
++
++ kfree(tmpbuf);
++ return retval;
++}
++
++
++static struct file_operations active_domain_ops = {
++ .read = adomain_read,
++ .write = adomain_write,
++};
++
++static unsigned int pdomains = 0;
++static int passive_domains[MAX_OPROF_DOMAINS];
++static DEFINE_MUTEX(pdom_mutex);
++
++static ssize_t pdomain_write(struct file * file, char const __user * buf,
++ size_t count, loff_t * offset)
++{
++ char *tmpbuf;
++ char *startp, *endp;
++ int i;
++ unsigned long val;
++ ssize_t retval = count;
++
++ if (*offset)
++ return -EINVAL;
++ if (count > TMPBUFSIZE - 1)
++ return -EINVAL;
++
++ if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL)))
++ return -ENOMEM;
++
++ if (copy_from_user(tmpbuf, buf, count)) {
++ kfree(tmpbuf);
++ return -EFAULT;
++ }
++ tmpbuf[count] = 0;
++
++ mutex_lock(&pdom_mutex);
++
++ startp = tmpbuf;
++ /* Parse one more than MAX_OPROF_DOMAINS, for easy error checking */
++ for (i = 0; i <= MAX_OPROF_DOMAINS; i++) {
++ val = simple_strtoul(startp, &endp, 0);
++ if (endp == startp)
++ break;
++ while (ispunct(*endp) || isspace(*endp))
++ endp++;
++ passive_domains[i] = val;
++ if (passive_domains[i] != val)
++ /* Overflow, force error below */
++ i = MAX_OPROF_DOMAINS + 1;
++ startp = endp;
++ }
++ /* Force error on trailing junk */
++ pdomains = *startp ? MAX_OPROF_DOMAINS + 1 : i;
++
++ kfree(tmpbuf);
++
++ if (pdomains > MAX_OPROF_DOMAINS
++ || oprofile_set_passive(passive_domains, pdomains)) {
++ pdomains = 0;
++ retval = -EINVAL;
++ }
++
++ mutex_unlock(&pdom_mutex);
++ return retval;
++}
++
++static ssize_t pdomain_read(struct file * file, char __user * buf,
++ size_t count, loff_t * offset)
++{
++ char * tmpbuf;
++ size_t len;
++ int i;
++ ssize_t retval;
++
++ if (!(tmpbuf = kmalloc(TMPBUFSIZE, GFP_KERNEL)))
++ return -ENOMEM;
++
++ mutex_lock(&pdom_mutex);
++
++ len = 0;
++ for (i = 0; i < pdomains; i++)
++ len += snprintf(tmpbuf + len,
++ len < TMPBUFSIZE ? TMPBUFSIZE - len : 0,
++ "%u ", passive_domains[i]);
++ WARN_ON(len > TMPBUFSIZE);
++ if (len != 0 && len <= TMPBUFSIZE)
++ tmpbuf[len-1] = '\n';
++
++ mutex_unlock(&pdom_mutex);
++
++ retval = simple_read_from_buffer(buf, count, offset, tmpbuf, len);
++
++ kfree(tmpbuf);
++ return retval;
++}
++
++static struct file_operations passive_domain_ops = {
++ .read = pdomain_read,
++ .write = pdomain_write,
++};
++
+ void oprofile_create_files(struct super_block * sb, struct dentry * root)
+ {
+ oprofilefs_create_file(sb, root, "enable", &enable_fops);
+ oprofilefs_create_file_perm(sb, root, "dump", &dump_fops, 0666);
++ oprofilefs_create_file(sb, root, "active_domains", &active_domain_ops);
++ oprofilefs_create_file(sb, root, "passive_domains", &passive_domain_ops);
+ oprofilefs_create_file(sb, root, "buffer", &event_buffer_fops);
+ oprofilefs_create_ulong(sb, root, "buffer_size", &fs_buffer_size);
+ oprofilefs_create_ulong(sb, root, "buffer_watershed", &fs_buffer_watershed);
+diff -urNp linux-2.6/drivers/pci/Kconfig new/drivers/pci/Kconfig
+--- linux-2.6/drivers/pci/Kconfig 2006-07-03 14:14:50.000000000 +0200
++++ new/drivers/pci/Kconfig 2006-05-09 12:34:08.000000000 +0200
+@@ -5,6 +5,7 @@ config PCI_MSI
+ bool "Message Signaled Interrupts (MSI and MSI-X)"
+ depends on PCI
+ depends on (X86_LOCAL_APIC && X86_IO_APIC) || IA64
++ depends on !XEN
+ help
+ This allows device drivers to enable MSI (Message Signaled
+ Interrupts). Message Signaled Interrupts enable a device to
+diff -urNp linux-2.6/drivers/s390/net/qeth_eddp.c new/drivers/s390/net/qeth_eddp.c
+--- linux-2.6/drivers/s390/net/qeth_eddp.c 2006-07-03 14:14:51.000000000 +0200
++++ new/drivers/s390/net/qeth_eddp.c 2006-07-07 16:04:32.000000000 +0200
+@@ -420,7 +420,7 @@ __qeth_eddp_fill_context_tcp(struct qeth
+ }
+ tcph = eddp->skb->h.th;
+ while (eddp->skb_offset < eddp->skb->len) {
+- data_len = min((int)skb_shinfo(eddp->skb)->tso_size,
++ data_len = min((int)skb_shinfo(eddp->skb)->gso_size,
+ (int)(eddp->skb->len - eddp->skb_offset));
+ /* prepare qdio hdr */
+ if (eddp->qh.hdr.l2.id == QETH_HEADER_TYPE_LAYER2){
+@@ -515,20 +515,20 @@ qeth_eddp_calc_num_pages(struct qeth_edd
+
+ QETH_DBF_TEXT(trace, 5, "eddpcanp");
+ /* can we put multiple skbs in one page? */
+- skbs_per_page = PAGE_SIZE / (skb_shinfo(skb)->tso_size + hdr_len);
++ skbs_per_page = PAGE_SIZE / (skb_shinfo(skb)->gso_size + hdr_len);
+ if (skbs_per_page > 1){
+- ctx->num_pages = (skb_shinfo(skb)->tso_segs + 1) /
++ ctx->num_pages = (skb_shinfo(skb)->gso_segs + 1) /
+ skbs_per_page + 1;
+ ctx->elements_per_skb = 1;
+ } else {
+ /* no -> how many elements per skb? */
+- ctx->elements_per_skb = (skb_shinfo(skb)->tso_size + hdr_len +
++ ctx->elements_per_skb = (skb_shinfo(skb)->gso_size + hdr_len +
+ PAGE_SIZE) >> PAGE_SHIFT;
+ ctx->num_pages = ctx->elements_per_skb *
+- (skb_shinfo(skb)->tso_segs + 1);
++ (skb_shinfo(skb)->gso_segs + 1);
+ }
+ ctx->num_elements = ctx->elements_per_skb *
+- (skb_shinfo(skb)->tso_segs + 1);
++ (skb_shinfo(skb)->gso_segs + 1);
+ }
+
+ static inline struct qeth_eddp_context *
+diff -urNp linux-2.6/drivers/s390/net/qeth_main.c new/drivers/s390/net/qeth_main.c
+--- linux-2.6/drivers/s390/net/qeth_main.c 2006-07-03 14:14:51.000000000 +0200
++++ new/drivers/s390/net/qeth_main.c 2006-07-07 16:39:53.000000000 +0200
+@@ -4417,7 +4417,6 @@ qeth_send_packet(struct qeth_card *card,
+ struct qeth_eddp_context *ctx = NULL;
+ int tx_bytes = skb->len;
+ unsigned short nr_frags = skb_shinfo(skb)->nr_frags;
+- unsigned short tso_size = skb_shinfo(skb)->tso_size;
+ int rc;
+
+ QETH_DBF_TEXT(trace, 6, "sendpkt");
+@@ -4453,7 +4452,7 @@ qeth_send_packet(struct qeth_card *card,
+ queue = card->qdio.out_qs
+ [qeth_get_priority_queue(card, skb, ipv, cast_type)];
+
+- if (skb_shinfo(skb)->tso_size)
++ if (skb_shinfo(skb)->gso_size)
+ large_send = card->options.large_send;
+
+ /*are we able to do TSO ? If so ,prepare and send it from here */
+@@ -4500,7 +4499,7 @@ qeth_send_packet(struct qeth_card *card,
+ card->stats.tx_packets++;
+ card->stats.tx_bytes += tx_bytes;
+ #ifdef CONFIG_QETH_PERF_STATS
+- if (tso_size &&
++ if (skb_shinfo(skb)->gso_size &&
+ !(large_send == QETH_LARGE_SEND_NO)) {
+ card->perf_stats.large_send_bytes += tx_bytes;
+ card->perf_stats.large_send_cnt++;
+diff -urNp linux-2.6/drivers/s390/net/qeth_tso.h new/drivers/s390/net/qeth_tso.h
+--- linux-2.6/drivers/s390/net/qeth_tso.h 2006-07-03 14:14:51.000000000 +0200
++++ new/drivers/s390/net/qeth_tso.h 2006-07-07 16:05:51.000000000 +0200
+@@ -51,7 +51,7 @@ qeth_tso_fill_header(struct qeth_card *c
+ hdr->ext.hdr_version = 1;
+ hdr->ext.hdr_len = 28;
+ /*insert non-fix values */
+- hdr->ext.mss = skb_shinfo(skb)->tso_size;
++ hdr->ext.mss = skb_shinfo(skb)->gso_size;
+ hdr->ext.dg_hdr_len = (__u16)(iph->ihl*4 + tcph->doff*4);
+ hdr->ext.payload_len = (__u16)(skb->len - hdr->ext.dg_hdr_len -
+ sizeof(struct qeth_hdr_tso));
+diff -urNp linux-2.6/drivers/serial/Kconfig new/drivers/serial/Kconfig
+--- linux-2.6/drivers/serial/Kconfig 2006-07-03 14:14:57.000000000 +0200
++++ new/drivers/serial/Kconfig 2006-05-09 12:34:23.000000000 +0200
+@@ -11,6 +11,7 @@ menu "Serial drivers"
+ config SERIAL_8250
+ tristate "8250/16550 and compatible serial support"
+ depends on (BROKEN || !SPARC)
++ depends on !XEN_DISABLE_SERIAL
+ select SERIAL_CORE
+ ---help---
+ This selects whether you want to include the driver for the standard
+diff -urNp linux-2.6/drivers/video/Kconfig new/drivers/video/Kconfig
+--- linux-2.6/drivers/video/Kconfig 2006-07-03 14:15:00.000000000 +0200
++++ new/drivers/video/Kconfig 2006-05-09 12:34:31.000000000 +0200
+@@ -513,7 +513,7 @@ config FB_HGA_ACCEL
+
+ config VIDEO_SELECT
+ bool
+- depends on (FB = y) && X86
++ depends on (FB = y) && X86 && !XEN
+ default y
+
+ config FB_SGIVW
+diff -urNp linux-2.6/drivers/xen/balloon/balloon.c new/drivers/xen/balloon/balloon.c
+--- linux-2.6/drivers/xen/balloon/balloon.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/balloon/balloon.c 2006-07-07 16:05:51.000000000 +0200
+@@ -0,0 +1,608 @@
++/******************************************************************************
++ * balloon.c
++ *
++ * Xen balloon driver - enables returning/claiming memory to/from Xen.
++ *
++ * Copyright (c) 2003, B Dragovic
++ * Copyright (c) 2003-2004, M Williamson, K Fraser
++ * Copyright (c) 2005 Dan M. Smith, IBM Corporation
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/module.h>
++#include <linux/sched.h>
++#include <linux/errno.h>
++#include <linux/mm.h>
++#include <linux/mman.h>
++#include <linux/smp_lock.h>
++#include <linux/pagemap.h>
++#include <linux/bootmem.h>
++#include <linux/highmem.h>
++#include <linux/vmalloc.h>
++#include <xen/xen_proc.h>
++#include <asm/hypervisor.h>
++#include <xen/balloon.h>
++#include <xen/interface/memory.h>
++#include <asm/pgalloc.h>
++#include <asm/pgtable.h>
++#include <asm/uaccess.h>
++#include <asm/tlb.h>
++#include <linux/list.h>
++
++#include <xen/xenbus.h>
++
++#define PAGES2KB(_p) ((_p)<<(PAGE_SHIFT-10))
++
++#ifdef CONFIG_PROC_FS
++static struct proc_dir_entry *balloon_pde;
++#endif
++
++static DECLARE_MUTEX(balloon_mutex);
++
++/*
++ * Protects atomic reservation decrease/increase against concurrent increases.
++ * Also protects non-atomic updates of current_pages and driver_pages, and
++ * balloon lists.
++ */
++DEFINE_SPINLOCK(balloon_lock);
++
++/* We aim for 'current allocation' == 'target allocation'. */
++static unsigned long current_pages;
++static unsigned long target_pages;
++
++/* We increase/decrease in batches which fit in a page */
++static unsigned long frame_list[PAGE_SIZE / sizeof(unsigned long)];
++
++/* VM /proc information for memory */
++extern unsigned long totalram_pages;
++
++/* We may hit the hard limit in Xen. If we do then we remember it. */
++static unsigned long hard_limit;
++
++/*
++ * Drivers may alter the memory reservation independently, but they must
++ * inform the balloon driver so that we can avoid hitting the hard limit.
++ */
++static unsigned long driver_pages;
++
++/* List of ballooned pages, threaded through the mem_map array. */
++static LIST_HEAD(ballooned_pages);
++static unsigned long balloon_low, balloon_high;
++
++/* Main work function, always executed in process context. */
++static void balloon_process(void *unused);
++static DECLARE_WORK(balloon_worker, balloon_process, NULL);
++static struct timer_list balloon_timer;
++
++/* When ballooning out (allocating memory to return to Xen) we don't really
++ want the kernel to try too hard since that can trigger the oom killer. */
++#define GFP_BALLOON \
++ (GFP_HIGHUSER | __GFP_NOWARN | __GFP_NORETRY | __GFP_NOMEMALLOC)
++
++#define PAGE_TO_LIST(p) (&(p)->lru)
++#define LIST_TO_PAGE(l) list_entry((l), struct page, lru)
++#define UNLIST_PAGE(p) \
++ do { \
++ list_del(PAGE_TO_LIST(p)); \
++ PAGE_TO_LIST(p)->next = NULL; \
++ PAGE_TO_LIST(p)->prev = NULL; \
++ } while(0)
++
++#define IPRINTK(fmt, args...) \
++ printk(KERN_INFO "xen_mem: " fmt, ##args)
++#define WPRINTK(fmt, args...) \
++ printk(KERN_WARNING "xen_mem: " fmt, ##args)
++
++/* balloon_append: add the given page to the balloon. */
++static void balloon_append(struct page *page)
++{
++ /* Lowmem is re-populated first, so highmem pages go at list tail. */
++ if (PageHighMem(page)) {
++ list_add_tail(PAGE_TO_LIST(page), &ballooned_pages);
++ balloon_high++;
++ } else {
++ list_add(PAGE_TO_LIST(page), &ballooned_pages);
++ balloon_low++;
++ }
++}
++
++/* balloon_retrieve: rescue a page from the balloon, if it is not empty. */
++static struct page *balloon_retrieve(void)
++{
++ struct page *page;
++
++ if (list_empty(&ballooned_pages))
++ return NULL;
++
++ page = LIST_TO_PAGE(ballooned_pages.next);
++ UNLIST_PAGE(page);
++
++ if (PageHighMem(page))
++ balloon_high--;
++ else
++ balloon_low--;
++
++ return page;
++}
++
++static struct page *balloon_first_page(void)
++{
++ if (list_empty(&ballooned_pages))
++ return NULL;
++ return LIST_TO_PAGE(ballooned_pages.next);
++}
++
++static struct page *balloon_next_page(struct page *page)
++{
++ struct list_head *next = PAGE_TO_LIST(page)->next;
++ if (next == &ballooned_pages)
++ return NULL;
++ return LIST_TO_PAGE(next);
++}
++
++static void balloon_alarm(unsigned long unused)
++{
++ schedule_work(&balloon_worker);
++}
++
++static unsigned long current_target(void)
++{
++ unsigned long target = min(target_pages, hard_limit);
++ if (target > (current_pages + balloon_low + balloon_high))
++ target = current_pages + balloon_low + balloon_high;
++ return target;
++}
++
++static int increase_reservation(unsigned long nr_pages)
++{
++ unsigned long pfn, i, flags;
++ struct page *page;
++ long rc;
++ struct xen_memory_reservation reservation = {
++ .address_bits = 0,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ };
++
++ if (nr_pages > ARRAY_SIZE(frame_list))
++ nr_pages = ARRAY_SIZE(frame_list);
++
++ balloon_lock(flags);
++
++ page = balloon_first_page();
++ for (i = 0; i < nr_pages; i++) {
++ BUG_ON(page == NULL);
++ frame_list[i] = page_to_pfn(page);;
++ page = balloon_next_page(page);
++ }
++
++ set_xen_guest_handle(reservation.extent_start, frame_list);
++ reservation.nr_extents = nr_pages;
++ rc = HYPERVISOR_memory_op(
++ XENMEM_populate_physmap, &reservation);
++ if (rc < nr_pages) {
++ if (rc > 0) {
++ int ret;
++
++ /* We hit the Xen hard limit: reprobe. */
++ reservation.nr_extents = rc;
++ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
++ &reservation);
++ BUG_ON(ret != rc);
++ }
++ if (rc >= 0)
++ hard_limit = current_pages + rc - driver_pages;
++ goto out;
++ }
++
++ for (i = 0; i < nr_pages; i++) {
++ page = balloon_retrieve();
++ BUG_ON(page == NULL);
++
++ pfn = page_to_pfn(page);
++ BUG_ON(!xen_feature(XENFEAT_auto_translated_physmap) &&
++ phys_to_machine_mapping_valid(pfn));
++
++ /* Update P->M and M->P tables. */
++ set_phys_to_machine(pfn, frame_list[i]);
++ xen_machphys_update(frame_list[i], pfn);
++
++ /* Link back into the page tables if not highmem. */
++ if (pfn < max_low_pfn) {
++ int ret;
++ ret = HYPERVISOR_update_va_mapping(
++ (unsigned long)__va(pfn << PAGE_SHIFT),
++ pfn_pte_ma(frame_list[i], PAGE_KERNEL),
++ 0);
++ BUG_ON(ret);
++ }
++
++ /* Relinquish the page back to the allocator. */
++ ClearPageReserved(page);
++ init_page_count(page);
++ __free_page(page);
++ }
++
++ current_pages += nr_pages;
++ totalram_pages = current_pages;
++
++ out:
++ balloon_unlock(flags);
++
++ return 0;
++}
++
++static int decrease_reservation(unsigned long nr_pages)
++{
++ unsigned long pfn, i, flags;
++ struct page *page;
++ void *v;
++ int need_sleep = 0;
++ int ret;
++ struct xen_memory_reservation reservation = {
++ .address_bits = 0,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ };
++
++ if (nr_pages > ARRAY_SIZE(frame_list))
++ nr_pages = ARRAY_SIZE(frame_list);
++
++ for (i = 0; i < nr_pages; i++) {
++ if ((page = alloc_page(GFP_BALLOON)) == NULL) {
++ nr_pages = i;
++ need_sleep = 1;
++ break;
++ }
++
++ pfn = page_to_pfn(page);
++ frame_list[i] = pfn_to_mfn(pfn);
++
++ if (!PageHighMem(page)) {
++ v = phys_to_virt(pfn << PAGE_SHIFT);
++ scrub_pages(v, 1);
++ ret = HYPERVISOR_update_va_mapping(
++ (unsigned long)v, __pte_ma(0), 0);
++ BUG_ON(ret);
++ }
++#ifdef CONFIG_XEN_SCRUB_PAGES
++ else {
++ v = kmap(page);
++ scrub_pages(v, 1);
++ kunmap(page);
++ }
++#endif
++ }
++
++ /* Ensure that ballooned highmem pages don't have kmaps. */
++ kmap_flush_unused();
++ flush_tlb_all();
++
++ balloon_lock(flags);
++
++ /* No more mappings: invalidate P2M and add to balloon. */
++ for (i = 0; i < nr_pages; i++) {
++ pfn = mfn_to_pfn(frame_list[i]);
++ set_phys_to_machine(pfn, INVALID_P2M_ENTRY);
++ balloon_append(pfn_to_page(pfn));
++ }
++
++ set_xen_guest_handle(reservation.extent_start, frame_list);
++ reservation.nr_extents = nr_pages;
++ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
++ BUG_ON(ret != nr_pages);
++
++ current_pages -= nr_pages;
++ totalram_pages = current_pages;
++
++ balloon_unlock(flags);
++
++ return need_sleep;
++}
++
++/*
++ * We avoid multiple worker processes conflicting via the balloon mutex.
++ * We may of course race updates of the target counts (which are protected
++ * by the balloon lock), or with changes to the Xen hard limit, but we will
++ * recover from these in time.
++ */
++static void balloon_process(void *unused)
++{
++ int need_sleep = 0;
++ long credit;
++
++ down(&balloon_mutex);
++
++ do {
++ credit = current_target() - current_pages;
++ if (credit > 0)
++ need_sleep = (increase_reservation(credit) != 0);
++ if (credit < 0)
++ need_sleep = (decrease_reservation(-credit) != 0);
++
++#ifndef CONFIG_PREEMPT
++ if (need_resched())
++ schedule();
++#endif
++ } while ((credit != 0) && !need_sleep);
++
++ /* Schedule more work if there is some still to be done. */
++ if (current_target() != current_pages)
++ mod_timer(&balloon_timer, jiffies + HZ);
++
++ up(&balloon_mutex);
++}
++
++/* Resets the Xen limit, sets new target, and kicks off processing. */
++static void set_new_target(unsigned long target)
++{
++ /* No need for lock. Not read-modify-write updates. */
++ hard_limit = ~0UL;
++ target_pages = target;
++ schedule_work(&balloon_worker);
++}
++
++static struct xenbus_watch target_watch =
++{
++ .node = "memory/target"
++};
++
++/* React to a change in the target key */
++static void watch_target(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ unsigned long long new_target;
++ int err;
++
++ err = xenbus_scanf(XBT_NIL, "memory", "target", "%llu", &new_target);
++ if (err != 1) {
++ /* This is ok (for domain0 at least) - so just return */
++ return;
++ }
++
++ /* The given memory/target value is in KiB, so it needs converting to
++ * pages. PAGE_SHIFT converts bytes to pages, hence PAGE_SHIFT - 10.
++ */
++ set_new_target(new_target >> (PAGE_SHIFT - 10));
++}
++
++static int balloon_init_watcher(struct notifier_block *notifier,
++ unsigned long event,
++ void *data)
++{
++ int err;
++
++ err = register_xenbus_watch(&target_watch);
++ if (err)
++ printk(KERN_ERR "Failed to set balloon watcher\n");
++
++ return NOTIFY_DONE;
++}
++
++#ifdef CONFIG_PROC_FS
++static int balloon_write(struct file *file, const char __user *buffer,
++ unsigned long count, void *data)
++{
++ char memstring[64], *endchar;
++ unsigned long long target_bytes;
++
++ if (!capable(CAP_SYS_ADMIN))
++ return -EPERM;
++
++ if (count <= 1)
++ return -EBADMSG; /* runt */
++ if (count > sizeof(memstring))
++ return -EFBIG; /* too long */
++
++ if (copy_from_user(memstring, buffer, count))
++ return -EFAULT;
++ memstring[sizeof(memstring)-1] = '\0';
++
++ target_bytes = memparse(memstring, &endchar);
++ set_new_target(target_bytes >> PAGE_SHIFT);
++
++ return count;
++}
++
++static int balloon_read(char *page, char **start, off_t off,
++ int count, int *eof, void *data)
++{
++ int len;
++
++ len = sprintf(
++ page,
++ "Current allocation: %8lu kB\n"
++ "Requested target: %8lu kB\n"
++ "Low-mem balloon: %8lu kB\n"
++ "High-mem balloon: %8lu kB\n"
++ "Xen hard limit: ",
++ PAGES2KB(current_pages), PAGES2KB(target_pages),
++ PAGES2KB(balloon_low), PAGES2KB(balloon_high));
++
++ if (hard_limit != ~0UL) {
++ len += sprintf(
++ page + len,
++ "%8lu kB (inc. %8lu kB driver headroom)\n",
++ PAGES2KB(hard_limit), PAGES2KB(driver_pages));
++ } else {
++ len += sprintf(
++ page + len,
++ " ??? kB\n");
++ }
++
++ *eof = 1;
++ return len;
++}
++#endif
++
++static struct notifier_block xenstore_notifier;
++
++static int __init balloon_init(void)
++{
++ unsigned long pfn;
++ struct page *page;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ IPRINTK("Initialising balloon driver.\n");
++
++ current_pages = min(xen_start_info->nr_pages, max_pfn);
++ totalram_pages = current_pages;
++ target_pages = current_pages;
++ balloon_low = 0;
++ balloon_high = 0;
++ driver_pages = 0UL;
++ hard_limit = ~0UL;
++
++ init_timer(&balloon_timer);
++ balloon_timer.data = 0;
++ balloon_timer.function = balloon_alarm;
++
++#ifdef CONFIG_PROC_FS
++ if ((balloon_pde = create_xen_proc_entry("balloon", 0644)) == NULL) {
++ WPRINTK("Unable to create /proc/xen/balloon.\n");
++ return -1;
++ }
++
++ balloon_pde->read_proc = balloon_read;
++ balloon_pde->write_proc = balloon_write;
++#endif
++
++ /* Initialise the balloon with excess memory space. */
++ for (pfn = xen_start_info->nr_pages; pfn < max_pfn; pfn++) {
++ page = pfn_to_page(pfn);
++ if (!PageReserved(page))
++ balloon_append(page);
++ }
++
++ target_watch.callback = watch_target;
++ xenstore_notifier.notifier_call = balloon_init_watcher;
++
++ register_xenstore_notifier(&xenstore_notifier);
++
++ return 0;
++}
++
++subsys_initcall(balloon_init);
++
++void balloon_update_driver_allowance(long delta)
++{
++ unsigned long flags;
++
++ balloon_lock(flags);
++ driver_pages += delta;
++ balloon_unlock(flags);
++}
++
++static int dealloc_pte_fn(
++ pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
++{
++ unsigned long mfn = pte_mfn(*pte);
++ int ret;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = 1,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, &mfn);
++ set_pte_at(&init_mm, addr, pte, __pte_ma(0));
++ set_phys_to_machine(__pa(addr) >> PAGE_SHIFT, INVALID_P2M_ENTRY);
++ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation, &reservation);
++ BUG_ON(ret != 1);
++ return 0;
++}
++
++struct page *balloon_alloc_empty_page_range(unsigned long nr_pages)
++{
++ unsigned long vstart, flags;
++ unsigned int order = get_order(nr_pages * PAGE_SIZE);
++ int ret;
++ unsigned long i;
++ struct page *page;
++
++ vstart = __get_free_pages(GFP_KERNEL, order);
++ if (vstart == 0)
++ return NULL;
++
++ scrub_pages(vstart, 1 << order);
++
++ balloon_lock(flags);
++ if (xen_feature(XENFEAT_auto_translated_physmap)) {
++ unsigned long gmfn = __pa(vstart) >> PAGE_SHIFT;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = 1,
++ .extent_order = order,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, &gmfn);
++ ret = HYPERVISOR_memory_op(XENMEM_decrease_reservation,
++ &reservation);
++ BUG_ON(ret != 1);
++ } else {
++ ret = apply_to_page_range(&init_mm, vstart, PAGE_SIZE << order,
++ dealloc_pte_fn, NULL);
++ BUG_ON(ret);
++ }
++ current_pages -= 1UL << order;
++ totalram_pages = current_pages;
++ balloon_unlock(flags);
++
++ schedule_work(&balloon_worker);
++
++ flush_tlb_all();
++
++ page = virt_to_page(vstart);
++
++ for (i = 0; i < (1UL << order); i++)
++ init_page_count(page + i);
++
++ return page;
++}
++
++void balloon_dealloc_empty_page_range(
++ struct page *page, unsigned long nr_pages)
++{
++ unsigned long i, flags;
++ unsigned int order = get_order(nr_pages * PAGE_SIZE);
++
++ balloon_lock(flags);
++ for (i = 0; i < (1UL << order); i++) {
++ BUG_ON(page_count(page + i) != 1);
++ balloon_append(page + i);
++ }
++ balloon_unlock(flags);
++
++ schedule_work(&balloon_worker);
++}
++
++EXPORT_SYMBOL_GPL(balloon_update_driver_allowance);
++EXPORT_SYMBOL_GPL(balloon_alloc_empty_page_range);
++EXPORT_SYMBOL_GPL(balloon_dealloc_empty_page_range);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/balloon/Makefile new/drivers/xen/balloon/Makefile
+--- linux-2.6/drivers/xen/balloon/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/balloon/Makefile 2006-05-09 12:34:36.000000000 +0200
+@@ -0,0 +1,2 @@
++
++obj-y += balloon.o
+diff -urNp linux-2.6/drivers/xen/blkback/blkback.c new/drivers/xen/blkback/blkback.c
+--- linux-2.6/drivers/xen/blkback/blkback.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/blkback.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,568 @@
++/******************************************************************************
++ * arch/xen/drivers/blkif/backend/main.c
++ *
++ * Back-end of the driver for virtual block devices. This portion of the
++ * driver exports a 'unified' block-device interface that can be accessed
++ * by any operating system that implements a compatible front end. A
++ * reference front-end implementation can be found in:
++ * arch/xen/drivers/blkif/frontend
++ *
++ * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
++ * Copyright (c) 2005, Christopher Clark
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/spinlock.h>
++#include <linux/kthread.h>
++#include <linux/list.h>
++#include <xen/balloon.h>
++#include <asm/hypervisor.h>
++#include "common.h"
++
++/*
++ * These are rather arbitrary. They are fairly large because adjacent requests
++ * pulled from a communication ring are quite likely to end up being part of
++ * the same scatter/gather request at the disc.
++ *
++ * ** TRY INCREASING 'blkif_reqs' IF WRITE SPEEDS SEEM TOO LOW **
++ *
++ * This will increase the chances of being able to write whole tracks.
++ * 64 should be enough to keep us competitive with Linux.
++ */
++static int blkif_reqs = 64;
++module_param_named(reqs, blkif_reqs, int, 0);
++MODULE_PARM_DESC(reqs, "Number of blkback requests to allocate");
++
++static int mmap_pages;
++
++/* Run-time switchable: /sys/module/blkback/parameters/ */
++static unsigned int log_stats = 0;
++static unsigned int debug_lvl = 0;
++module_param(log_stats, int, 0644);
++module_param(debug_lvl, int, 0644);
++
++/*
++ * Each outstanding request that we've passed to the lower device layers has a
++ * 'pending_req' allocated to it. Each buffer_head that completes decrements
++ * the pendcnt towards zero. When it hits zero, the specified domain has a
++ * response queued for it, with the saved 'id' passed back.
++ */
++typedef struct {
++ blkif_t *blkif;
++ unsigned long id;
++ int nr_pages;
++ atomic_t pendcnt;
++ unsigned short operation;
++ int status;
++ struct list_head free_list;
++} pending_req_t;
++
++static pending_req_t *pending_reqs;
++static struct list_head pending_free;
++static DEFINE_SPINLOCK(pending_free_lock);
++static DECLARE_WAIT_QUEUE_HEAD(pending_free_wq);
++
++#define BLKBACK_INVALID_HANDLE (~0)
++
++static unsigned long mmap_vstart;
++static unsigned long *pending_vaddrs;
++static grant_handle_t *pending_grant_handles;
++
++static inline int vaddr_pagenr(pending_req_t *req, int seg)
++{
++ return (req - pending_reqs) * BLKIF_MAX_SEGMENTS_PER_REQUEST + seg;
++}
++
++static inline unsigned long vaddr(pending_req_t *req, int seg)
++{
++ return pending_vaddrs[vaddr_pagenr(req, seg)];
++}
++
++#define pending_handle(_req, _seg) \
++ (pending_grant_handles[vaddr_pagenr(_req, _seg)])
++
++
++static int do_block_io_op(blkif_t *blkif);
++static void dispatch_rw_block_io(blkif_t *blkif,
++ blkif_request_t *req,
++ pending_req_t *pending_req);
++static void make_response(blkif_t *blkif, unsigned long id,
++ unsigned short op, int st);
++
++/******************************************************************
++ * misc small helpers
++ */
++static pending_req_t* alloc_req(void)
++{
++ pending_req_t *req = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&pending_free_lock, flags);
++ if (!list_empty(&pending_free)) {
++ req = list_entry(pending_free.next, pending_req_t, free_list);
++ list_del(&req->free_list);
++ }
++ spin_unlock_irqrestore(&pending_free_lock, flags);
++ return req;
++}
++
++static void free_req(pending_req_t *req)
++{
++ unsigned long flags;
++ int was_empty;
++
++ spin_lock_irqsave(&pending_free_lock, flags);
++ was_empty = list_empty(&pending_free);
++ list_add(&req->free_list, &pending_free);
++ spin_unlock_irqrestore(&pending_free_lock, flags);
++ if (was_empty)
++ wake_up(&pending_free_wq);
++}
++
++static void unplug_queue(blkif_t *blkif)
++{
++ if (blkif->plug == NULL)
++ return;
++ if (blkif->plug->unplug_fn)
++ blkif->plug->unplug_fn(blkif->plug);
++ blk_put_queue(blkif->plug);
++ blkif->plug = NULL;
++}
++
++static void plug_queue(blkif_t *blkif, struct bio *bio)
++{
++ request_queue_t *q = bdev_get_queue(bio->bi_bdev);
++
++ if (q == blkif->plug)
++ return;
++ unplug_queue(blkif);
++ blk_get_queue(q);
++ blkif->plug = q;
++}
++
++static void fast_flush_area(pending_req_t *req)
++{
++ struct gnttab_unmap_grant_ref unmap[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++ unsigned int i, invcount = 0;
++ grant_handle_t handle;
++ int ret;
++
++ for (i = 0; i < req->nr_pages; i++) {
++ handle = pending_handle(req, i);
++ if (handle == BLKBACK_INVALID_HANDLE)
++ continue;
++ gnttab_set_unmap_op(&unmap[i], vaddr(req, i), GNTMAP_host_map,
++ handle);
++ pending_handle(req, i) = BLKBACK_INVALID_HANDLE;
++ invcount++;
++ }
++
++ ret = HYPERVISOR_grant_table_op(
++ GNTTABOP_unmap_grant_ref, unmap, invcount);
++ BUG_ON(ret);
++}
++
++/******************************************************************
++ * SCHEDULER FUNCTIONS
++ */
++
++static void print_stats(blkif_t *blkif)
++{
++ printk(KERN_DEBUG "%s: oo %3d | rd %4d | wr %4d\n",
++ current->comm, blkif->st_oo_req,
++ blkif->st_rd_req, blkif->st_wr_req);
++ blkif->st_print = jiffies + msecs_to_jiffies(10 * 1000);
++ blkif->st_rd_req = 0;
++ blkif->st_wr_req = 0;
++ blkif->st_oo_req = 0;
++}
++
++int blkif_schedule(void *arg)
++{
++ blkif_t *blkif = arg;
++
++ blkif_get(blkif);
++
++ if (debug_lvl)
++ printk(KERN_DEBUG "%s: started\n", current->comm);
++
++ while (!kthread_should_stop()) {
++ wait_event_interruptible(
++ blkif->wq,
++ blkif->waiting_reqs || kthread_should_stop());
++ wait_event_interruptible(
++ pending_free_wq,
++ !list_empty(&pending_free) || kthread_should_stop());
++
++ blkif->waiting_reqs = 0;
++ smp_mb(); /* clear flag *before* checking for work */
++
++ if (do_block_io_op(blkif))
++ blkif->waiting_reqs = 1;
++ unplug_queue(blkif);
++
++ if (log_stats && time_after(jiffies, blkif->st_print))
++ print_stats(blkif);
++ }
++
++ if (log_stats)
++ print_stats(blkif);
++ if (debug_lvl)
++ printk(KERN_DEBUG "%s: exiting\n", current->comm);
++
++ blkif->xenblkd = NULL;
++ blkif_put(blkif);
++
++ return 0;
++}
++
++/******************************************************************
++ * COMPLETION CALLBACK -- Called as bh->b_end_io()
++ */
++
++static void __end_block_io_op(pending_req_t *pending_req, int uptodate)
++{
++ /* An error fails the entire request. */
++ if (!uptodate) {
++ DPRINTK("Buffer not up-to-date at end of operation\n");
++ pending_req->status = BLKIF_RSP_ERROR;
++ }
++
++ if (atomic_dec_and_test(&pending_req->pendcnt)) {
++ fast_flush_area(pending_req);
++ make_response(pending_req->blkif, pending_req->id,
++ pending_req->operation, pending_req->status);
++ blkif_put(pending_req->blkif);
++ free_req(pending_req);
++ }
++}
++
++static int end_block_io_op(struct bio *bio, unsigned int done, int error)
++{
++ if (bio->bi_size != 0)
++ return 1;
++ __end_block_io_op(bio->bi_private, !error);
++ bio_put(bio);
++ return error;
++}
++
++
++/******************************************************************************
++ * NOTIFICATION FROM GUEST OS.
++ */
++
++static void blkif_notify_work(blkif_t *blkif)
++{
++ blkif->waiting_reqs = 1;
++ wake_up(&blkif->wq);
++}
++
++irqreturn_t blkif_be_int(int irq, void *dev_id, struct pt_regs *regs)
++{
++ blkif_notify_work(dev_id);
++ return IRQ_HANDLED;
++}
++
++
++
++/******************************************************************
++ * DOWNWARD CALLS -- These interface with the block-device layer proper.
++ */
++
++static int do_block_io_op(blkif_t *blkif)
++{
++ blkif_back_ring_t *blk_ring = &blkif->blk_ring;
++ blkif_request_t *req;
++ pending_req_t *pending_req;
++ RING_IDX rc, rp;
++ int more_to_do = 0;
++
++ rc = blk_ring->req_cons;
++ rp = blk_ring->sring->req_prod;
++ rmb(); /* Ensure we see queued requests up to 'rp'. */
++
++ while ((rc != rp) && !RING_REQUEST_CONS_OVERFLOW(blk_ring, rc)) {
++
++ pending_req = alloc_req();
++ if (NULL == pending_req) {
++ blkif->st_oo_req++;
++ more_to_do = 1;
++ break;
++ }
++
++ req = RING_GET_REQUEST(blk_ring, rc);
++ blk_ring->req_cons = ++rc; /* before make_response() */
++
++ switch (req->operation) {
++ case BLKIF_OP_READ:
++ blkif->st_rd_req++;
++ dispatch_rw_block_io(blkif, req, pending_req);
++ break;
++ case BLKIF_OP_WRITE:
++ blkif->st_wr_req++;
++ dispatch_rw_block_io(blkif, req, pending_req);
++ break;
++ default:
++ DPRINTK("error: unknown block io operation [%d]\n",
++ req->operation);
++ make_response(blkif, req->id, req->operation,
++ BLKIF_RSP_ERROR);
++ free_req(pending_req);
++ break;
++ }
++ }
++ return more_to_do;
++}
++
++static void dispatch_rw_block_io(blkif_t *blkif,
++ blkif_request_t *req,
++ pending_req_t *pending_req)
++{
++ extern void ll_rw_block(int rw, int nr, struct buffer_head * bhs[]);
++ int operation = (req->operation == BLKIF_OP_WRITE) ? WRITE : READ;
++ struct gnttab_map_grant_ref map[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++ struct phys_req preq;
++ struct {
++ unsigned long buf; unsigned int nsec;
++ } seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++ unsigned int nseg;
++ struct bio *bio = NULL, *biolist[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++ int ret, i, nbio = 0;
++
++ /* Check that number of segments is sane. */
++ nseg = req->nr_segments;
++ if (unlikely(nseg == 0) ||
++ unlikely(nseg > BLKIF_MAX_SEGMENTS_PER_REQUEST)) {
++ DPRINTK("Bad number of segments in request (%d)\n", nseg);
++ goto fail_response;
++ }
++
++ preq.dev = req->handle;
++ preq.sector_number = req->sector_number;
++ preq.nr_sects = 0;
++
++ pending_req->blkif = blkif;
++ pending_req->id = req->id;
++ pending_req->operation = operation;
++ pending_req->status = BLKIF_RSP_OKAY;
++ pending_req->nr_pages = nseg;
++
++ for (i = 0; i < nseg; i++) {
++ uint32_t flags;
++
++ seg[i].nsec = req->seg[i].last_sect -
++ req->seg[i].first_sect + 1;
++
++ if ((req->seg[i].last_sect >= (PAGE_SIZE >> 9)) ||
++ (seg[i].nsec <= 0))
++ goto fail_response;
++ preq.nr_sects += seg[i].nsec;
++
++ flags = GNTMAP_host_map;
++ if ( operation == WRITE )
++ flags |= GNTMAP_readonly;
++ gnttab_set_map_op(&map[i], vaddr(pending_req, i), flags,
++ req->seg[i].gref, blkif->domid);
++ }
++
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, map, nseg);
++ BUG_ON(ret);
++
++ for (i = 0; i < nseg; i++) {
++ if (unlikely(map[i].status != 0)) {
++ DPRINTK("invalid buffer -- could not remap it\n");
++ goto fail_flush;
++ }
++
++ pending_handle(pending_req, i) = map[i].handle;
++#ifdef CONFIG_XEN_IA64_DOM0_NON_VP
++ pending_vaddrs[vaddr_pagenr(pending_req, i)] =
++ (unsigned long)gnttab_map_vaddr(map[i]);
++#else
++ set_phys_to_machine(__pa(vaddr(
++ pending_req, i)) >> PAGE_SHIFT,
++ FOREIGN_FRAME(map[i].dev_bus_addr >> PAGE_SHIFT));
++#endif
++ seg[i].buf = map[i].dev_bus_addr |
++ (req->seg[i].first_sect << 9);
++ }
++
++ if (vbd_translate(&preq, blkif, operation) != 0) {
++ DPRINTK("access denied: %s of [%llu,%llu] on dev=%04x\n",
++ operation == READ ? "read" : "write",
++ preq.sector_number,
++ preq.sector_number + preq.nr_sects, preq.dev);
++ goto fail_flush;
++ }
++
++ for (i = 0; i < nseg; i++) {
++ if (((int)preq.sector_number|(int)seg[i].nsec) &
++ ((bdev_hardsect_size(preq.bdev) >> 9) - 1)) {
++ DPRINTK("Misaligned I/O request from domain %d",
++ blkif->domid);
++ goto fail_put_bio;
++ }
++
++ while ((bio == NULL) ||
++ (bio_add_page(bio,
++ virt_to_page(vaddr(pending_req, i)),
++ seg[i].nsec << 9,
++ seg[i].buf & ~PAGE_MASK) == 0)) {
++ bio = biolist[nbio++] = bio_alloc(GFP_KERNEL, nseg-i);
++ if (unlikely(bio == NULL))
++ goto fail_put_bio;
++
++ bio->bi_bdev = preq.bdev;
++ bio->bi_private = pending_req;
++ bio->bi_end_io = end_block_io_op;
++ bio->bi_sector = preq.sector_number;
++ }
++
++ preq.sector_number += seg[i].nsec;
++ }
++
++ plug_queue(blkif, bio);
++ atomic_set(&pending_req->pendcnt, nbio);
++ blkif_get(blkif);
++
++ for (i = 0; i < nbio; i++)
++ submit_bio(operation, biolist[i]);
++
++ return;
++
++ fail_put_bio:
++ for (i = 0; i < (nbio-1); i++)
++ bio_put(biolist[i]);
++ fail_flush:
++ fast_flush_area(pending_req);
++ fail_response:
++ make_response(blkif, req->id, req->operation, BLKIF_RSP_ERROR);
++ free_req(pending_req);
++}
++
++
++
++/******************************************************************
++ * MISCELLANEOUS SETUP / TEARDOWN / DEBUGGING
++ */
++
++
++static void make_response(blkif_t *blkif, unsigned long id,
++ unsigned short op, int st)
++{
++ blkif_response_t *resp;
++ unsigned long flags;
++ blkif_back_ring_t *blk_ring = &blkif->blk_ring;
++ int more_to_do = 0;
++ int notify;
++
++ spin_lock_irqsave(&blkif->blk_ring_lock, flags);
++
++ /* Place on the response ring for the relevant domain. */
++ resp = RING_GET_RESPONSE(blk_ring, blk_ring->rsp_prod_pvt);
++ resp->id = id;
++ resp->operation = op;
++ resp->status = st;
++ blk_ring->rsp_prod_pvt++;
++ RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(blk_ring, notify);
++
++ if (blk_ring->rsp_prod_pvt == blk_ring->req_cons) {
++ /*
++ * Tail check for pending requests. Allows frontend to avoid
++ * notifications if requests are already in flight (lower
++ * overheads and promotes batching).
++ */
++ RING_FINAL_CHECK_FOR_REQUESTS(blk_ring, more_to_do);
++
++ } else if (RING_HAS_UNCONSUMED_REQUESTS(blk_ring)) {
++ more_to_do = 1;
++
++ }
++ spin_unlock_irqrestore(&blkif->blk_ring_lock, flags);
++
++ if (more_to_do)
++ blkif_notify_work(blkif);
++ if (notify)
++ notify_remote_via_irq(blkif->irq);
++}
++
++static int __init blkif_init(void)
++{
++ struct page *page;
++ int i;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ mmap_pages = blkif_reqs * BLKIF_MAX_SEGMENTS_PER_REQUEST;
++ pending_reqs = kmalloc(sizeof(pending_reqs[0]) *
++ blkif_reqs, GFP_KERNEL);
++ pending_grant_handles = kmalloc(sizeof(pending_grant_handles[0]) *
++ mmap_pages, GFP_KERNEL);
++ pending_vaddrs = kmalloc(sizeof(pending_vaddrs[0]) *
++ mmap_pages, GFP_KERNEL);
++ if (!pending_reqs || !pending_grant_handles || !pending_vaddrs) {
++ kfree(pending_reqs);
++ kfree(pending_grant_handles);
++ kfree(pending_vaddrs);
++ printk("%s: out of memory\n", __FUNCTION__);
++ return -ENOMEM;
++ }
++
++ blkif_interface_init();
++
++#ifdef CONFIG_XEN_IA64_DOM0_NON_VP
++ extern unsigned long alloc_empty_foreign_map_page_range(
++ unsigned long pages);
++ mmap_vstart = (unsigned long)
++ alloc_empty_foreign_map_page_range(mmap_pages);
++#else /* ! ia64 */
++ page = balloon_alloc_empty_page_range(mmap_pages);
++ BUG_ON(page == NULL);
++ mmap_vstart = (unsigned long)pfn_to_kaddr(page_to_pfn(page));
++#endif
++ printk("%s: reqs=%d, pages=%d, mmap_vstart=0x%lx\n",
++ __FUNCTION__, blkif_reqs, mmap_pages, mmap_vstart);
++ BUG_ON(mmap_vstart == 0);
++ for (i = 0; i < mmap_pages; i++) {
++ pending_vaddrs[i] = mmap_vstart + (i << PAGE_SHIFT);
++ pending_grant_handles[i] = BLKBACK_INVALID_HANDLE;
++ }
++
++ memset(pending_reqs, 0, sizeof(pending_reqs));
++ INIT_LIST_HEAD(&pending_free);
++
++ for (i = 0; i < blkif_reqs; i++)
++ list_add_tail(&pending_reqs[i].free_list, &pending_free);
++
++ blkif_xenbus_init();
++
++ return 0;
++}
++
++module_init(blkif_init);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/blkback/common.h new/drivers/xen/blkback/common.h
+--- linux-2.6/drivers/xen/blkback/common.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/common.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,134 @@
++/*
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __BLKIF__BACKEND__COMMON_H__
++#define __BLKIF__BACKEND__COMMON_H__
++
++#include <linux/config.h>
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/interrupt.h>
++#include <linux/slab.h>
++#include <linux/blkdev.h>
++#include <linux/vmalloc.h>
++#include <linux/wait.h>
++#include <asm/io.h>
++#include <asm/setup.h>
++#include <asm/pgalloc.h>
++#include <xen/evtchn.h>
++#include <asm/hypervisor.h>
++#include <xen/interface/io/blkif.h>
++#include <xen/interface/io/ring.h>
++#include <xen/gnttab.h>
++#include <xen/driver_util.h>
++
++#define DPRINTK(_f, _a...) \
++ pr_debug("(file=%s, line=%d) " _f, \
++ __FILE__ , __LINE__ , ## _a )
++
++struct vbd {
++ blkif_vdev_t handle; /* what the domain refers to this vbd as */
++ unsigned char readonly; /* Non-zero -> read-only */
++ unsigned char type; /* VDISK_xxx */
++ u32 pdevice; /* phys device that this vbd maps to */
++ struct block_device *bdev;
++};
++
++struct backend_info;
++
++typedef struct blkif_st {
++ /* Unique identifier for this interface. */
++ domid_t domid;
++ unsigned int handle;
++ /* Physical parameters of the comms window. */
++ unsigned int evtchn;
++ unsigned int irq;
++ /* Comms information. */
++ blkif_back_ring_t blk_ring;
++ struct vm_struct *blk_ring_area;
++ /* The VBD attached to this interface. */
++ struct vbd vbd;
++ /* Back pointer to the backend_info. */
++ struct backend_info *be;
++ /* Private fields. */
++ spinlock_t blk_ring_lock;
++ atomic_t refcnt;
++
++ wait_queue_head_t wq;
++ struct task_struct *xenblkd;
++ unsigned int waiting_reqs;
++ request_queue_t *plug;
++
++ /* statistics */
++ unsigned long st_print;
++ int st_rd_req;
++ int st_wr_req;
++ int st_oo_req;
++
++ wait_queue_head_t waiting_to_free;
++
++ grant_handle_t shmem_handle;
++ grant_ref_t shmem_ref;
++} blkif_t;
++
++blkif_t *blkif_alloc(domid_t domid);
++void blkif_disconnect(blkif_t *blkif);
++void blkif_free(blkif_t *blkif);
++int blkif_map(blkif_t *blkif, unsigned long shared_page, unsigned int evtchn);
++
++#define blkif_get(_b) (atomic_inc(&(_b)->refcnt))
++#define blkif_put(_b) \
++ do { \
++ if (atomic_dec_and_test(&(_b)->refcnt)) \
++ wake_up(&(_b)->waiting_to_free);\
++ } while (0)
++
++/* Create a vbd. */
++int vbd_create(blkif_t *blkif, blkif_vdev_t vdevice, unsigned major,
++ unsigned minor, int readonly);
++void vbd_free(struct vbd *vbd);
++
++unsigned long vbd_size(struct vbd *vbd);
++unsigned int vbd_info(struct vbd *vbd);
++unsigned long vbd_secsize(struct vbd *vbd);
++
++struct phys_req {
++ unsigned short dev;
++ unsigned short nr_sects;
++ struct block_device *bdev;
++ blkif_sector_t sector_number;
++};
++
++int vbd_translate(struct phys_req *req, blkif_t *blkif, int operation);
++
++void blkif_interface_init(void);
++
++void blkif_xenbus_init(void);
++
++irqreturn_t blkif_be_int(int irq, void *dev_id, struct pt_regs *regs);
++int blkif_schedule(void *arg);
++
++#endif /* __BLKIF__BACKEND__COMMON_H__ */
+diff -urNp linux-2.6/drivers/xen/blkback/interface.c new/drivers/xen/blkback/interface.c
+--- linux-2.6/drivers/xen/blkback/interface.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/interface.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,177 @@
++/******************************************************************************
++ * arch/xen/drivers/blkif/backend/interface.c
++ *
++ * Block-device interface management.
++ *
++ * Copyright (c) 2004, Keir Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include "common.h"
++#include <xen/evtchn.h>
++#include <linux/kthread.h>
++
++static kmem_cache_t *blkif_cachep;
++
++blkif_t *blkif_alloc(domid_t domid)
++{
++ blkif_t *blkif;
++
++ blkif = kmem_cache_alloc(blkif_cachep, GFP_KERNEL);
++ if (!blkif)
++ return ERR_PTR(-ENOMEM);
++
++ memset(blkif, 0, sizeof(*blkif));
++ blkif->domid = domid;
++ spin_lock_init(&blkif->blk_ring_lock);
++ atomic_set(&blkif->refcnt, 1);
++ init_waitqueue_head(&blkif->wq);
++ blkif->st_print = jiffies;
++ init_waitqueue_head(&blkif->waiting_to_free);
++
++ return blkif;
++}
++
++static int map_frontend_page(blkif_t *blkif, unsigned long shared_page)
++{
++ struct gnttab_map_grant_ref op;
++ int ret;
++
++ gnttab_set_map_op(&op, (unsigned long)blkif->blk_ring_area->addr,
++ GNTMAP_host_map, shared_page, blkif->domid);
++
++ lock_vm_area(blkif->blk_ring_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1);
++ unlock_vm_area(blkif->blk_ring_area);
++ BUG_ON(ret);
++
++ if (op.status) {
++ DPRINTK(" Grant table operation failure !\n");
++ return op.status;
++ }
++
++ blkif->shmem_ref = shared_page;
++ blkif->shmem_handle = op.handle;
++
++#ifdef CONFIG_XEN_IA64_DOM0_NON_VP
++ /* on some arch's, map_grant_ref behaves like mmap, in that the
++ * passed address is a hint and a different address may be returned */
++ blkif->blk_ring_area->addr = gnttab_map_vaddr(op);
++#endif
++
++ return 0;
++}
++
++static void unmap_frontend_page(blkif_t *blkif)
++{
++ struct gnttab_unmap_grant_ref op;
++ int ret;
++
++ gnttab_set_unmap_op(&op, (unsigned long)blkif->blk_ring_area->addr,
++ GNTMAP_host_map, blkif->shmem_handle);
++
++ lock_vm_area(blkif->blk_ring_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1);
++ unlock_vm_area(blkif->blk_ring_area);
++ BUG_ON(ret);
++}
++
++int blkif_map(blkif_t *blkif, unsigned long shared_page, unsigned int evtchn)
++{
++ blkif_sring_t *sring;
++ int err;
++ struct evtchn_bind_interdomain bind_interdomain;
++
++ /* Already connected through? */
++ if (blkif->irq)
++ return 0;
++
++ if ( (blkif->blk_ring_area = alloc_vm_area(PAGE_SIZE)) == NULL )
++ return -ENOMEM;
++
++ err = map_frontend_page(blkif, shared_page);
++ if (err) {
++ free_vm_area(blkif->blk_ring_area);
++ return err;
++ }
++
++ bind_interdomain.remote_dom = blkif->domid;
++ bind_interdomain.remote_port = evtchn;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
++ &bind_interdomain);
++ if (err) {
++ unmap_frontend_page(blkif);
++ free_vm_area(blkif->blk_ring_area);
++ return err;
++ }
++
++ blkif->evtchn = bind_interdomain.local_port;
++
++ sring = (blkif_sring_t *)blkif->blk_ring_area->addr;
++ BACK_RING_INIT(&blkif->blk_ring, sring, PAGE_SIZE);
++
++ blkif->irq = bind_evtchn_to_irqhandler(
++ blkif->evtchn, blkif_be_int, 0, "blkif-backend", blkif);
++
++ return 0;
++}
++
++void blkif_disconnect(blkif_t *blkif)
++{
++ if (blkif->xenblkd) {
++ kthread_stop(blkif->xenblkd);
++ blkif->xenblkd = NULL;
++ }
++
++ atomic_dec(&blkif->refcnt);
++ wait_event(blkif->waiting_to_free, atomic_read(&blkif->refcnt) == 0);
++ atomic_inc(&blkif->refcnt);
++
++ if (blkif->irq) {
++ unbind_from_irqhandler(blkif->irq, blkif);
++ blkif->irq = 0;
++ }
++
++ if (blkif->blk_ring.sring) {
++ unmap_frontend_page(blkif);
++ free_vm_area(blkif->blk_ring_area);
++ blkif->blk_ring.sring = NULL;
++ }
++}
++
++void blkif_free(blkif_t *blkif)
++{
++ if (!atomic_dec_and_test(&blkif->refcnt))
++ BUG();
++ kmem_cache_free(blkif_cachep, blkif);
++}
++
++void __init blkif_interface_init(void)
++{
++ blkif_cachep = kmem_cache_create("blkif_cache", sizeof(blkif_t),
++ 0, 0, NULL, NULL);
++}
+diff -urNp linux-2.6/drivers/xen/blkback/Makefile new/drivers/xen/blkback/Makefile
+--- linux-2.6/drivers/xen/blkback/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/Makefile 2006-05-09 12:34:36.000000000 +0200
+@@ -0,0 +1,3 @@
++obj-$(CONFIG_XEN_BLKDEV_BACKEND) := blkbk.o
++
++blkbk-y := blkback.o xenbus.o interface.o vbd.o
+diff -urNp linux-2.6/drivers/xen/blkback/vbd.c new/drivers/xen/blkback/vbd.c
+--- linux-2.6/drivers/xen/blkback/vbd.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/vbd.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,119 @@
++/******************************************************************************
++ * blkback/vbd.c
++ *
++ * Routines for managing virtual block devices (VBDs).
++ *
++ * Copyright (c) 2003-2005, Keir Fraser & Steve Hand
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include "common.h"
++#include <xen/xenbus.h>
++
++#define vbd_sz(_v) ((_v)->bdev->bd_part ? \
++ (_v)->bdev->bd_part->nr_sects : (_v)->bdev->bd_disk->capacity)
++
++unsigned long vbd_size(struct vbd *vbd)
++{
++ return vbd_sz(vbd);
++}
++
++unsigned int vbd_info(struct vbd *vbd)
++{
++ return vbd->type | (vbd->readonly?VDISK_READONLY:0);
++}
++
++unsigned long vbd_secsize(struct vbd *vbd)
++{
++ return bdev_hardsect_size(vbd->bdev);
++}
++
++int vbd_create(blkif_t *blkif, blkif_vdev_t handle, unsigned major,
++ unsigned minor, int readonly)
++{
++ struct vbd *vbd;
++ struct block_device *bdev;
++
++ vbd = &blkif->vbd;
++ vbd->handle = handle;
++ vbd->readonly = readonly;
++ vbd->type = 0;
++
++ vbd->pdevice = MKDEV(major, minor);
++
++ bdev = open_by_devnum(vbd->pdevice,
++ vbd->readonly ? FMODE_READ : FMODE_WRITE);
++
++ if (IS_ERR(bdev)) {
++ DPRINTK("vbd_creat: device %08x could not be opened.\n",
++ vbd->pdevice);
++ return -ENOENT;
++ }
++
++ vbd->bdev = bdev;
++
++ if (vbd->bdev->bd_disk == NULL) {
++ DPRINTK("vbd_creat: device %08x doesn't exist.\n",
++ vbd->pdevice);
++ vbd_free(vbd);
++ return -ENOENT;
++ }
++
++ if (vbd->bdev->bd_disk->flags & GENHD_FL_CD)
++ vbd->type |= VDISK_CDROM;
++ if (vbd->bdev->bd_disk->flags & GENHD_FL_REMOVABLE)
++ vbd->type |= VDISK_REMOVABLE;
++
++ DPRINTK("Successful creation of handle=%04x (dom=%u)\n",
++ handle, blkif->domid);
++ return 0;
++}
++
++void vbd_free(struct vbd *vbd)
++{
++ if (vbd->bdev)
++ blkdev_put(vbd->bdev);
++ vbd->bdev = NULL;
++}
++
++int vbd_translate(struct phys_req *req, blkif_t *blkif, int operation)
++{
++ struct vbd *vbd = &blkif->vbd;
++ int rc = -EACCES;
++
++ if ((operation == WRITE) && vbd->readonly)
++ goto out;
++
++ if (unlikely((req->sector_number + req->nr_sects) > vbd_sz(vbd)))
++ goto out;
++
++ req->dev = vbd->pdevice;
++ req->bdev = vbd->bdev;
++ rc = 0;
++
++ out:
++ return rc;
++}
+diff -urNp linux-2.6/drivers/xen/blkback/xenbus.c new/drivers/xen/blkback/xenbus.c
+--- linux-2.6/drivers/xen/blkback/xenbus.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkback/xenbus.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,460 @@
++/* Xenbus code for blkif backend
++ Copyright (C) 2005 Rusty Russell <rusty@rustcorp.com.au>
++ Copyright (C) 2005 XenSource Ltd
++
++ This program is free software; you can redistribute it and/or modify
++ it under the terms of the GNU General Public License as published by
++ the Free Software Foundation; either version 2 of the License, or
++ (at your option) any later version.
++
++ This program is distributed in the hope that it will be useful,
++ but WITHOUT ANY WARRANTY; without even the implied warranty of
++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ GNU General Public License for more details.
++
++ You should have received a copy of the GNU General Public License
++ along with this program; if not, write to the Free Software
++ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++*/
++
++#include <stdarg.h>
++#include <linux/module.h>
++#include <linux/kthread.h>
++#include <xen/xenbus.h>
++#include "common.h"
++
++#undef DPRINTK
++#define DPRINTK(fmt, args...) \
++ pr_debug("blkback/xenbus (%s:%d) " fmt ".\n", \
++ __FUNCTION__, __LINE__, ##args)
++
++struct backend_info
++{
++ struct xenbus_device *dev;
++ blkif_t *blkif;
++ struct xenbus_watch backend_watch;
++ unsigned major;
++ unsigned minor;
++ char *mode;
++};
++
++static void connect(struct backend_info *);
++static int connect_ring(struct backend_info *);
++static void backend_changed(struct xenbus_watch *, const char **,
++ unsigned int);
++
++static void update_blkif_status(blkif_t *blkif)
++{
++ int err;
++
++ /* Not ready to connect? */
++ if (!blkif->irq || !blkif->vbd.bdev)
++ return;
++
++ /* Already connected? */
++ if (blkif->be->dev->state == XenbusStateConnected)
++ return;
++
++ /* Attempt to connect: exit if we fail to. */
++ connect(blkif->be);
++ if (blkif->be->dev->state != XenbusStateConnected)
++ return;
++
++ blkif->xenblkd = kthread_run(blkif_schedule, blkif,
++ "xvd %d %02x:%02x",
++ blkif->domid,
++ blkif->be->major, blkif->be->minor);
++ if (IS_ERR(blkif->xenblkd)) {
++ err = PTR_ERR(blkif->xenblkd);
++ blkif->xenblkd = NULL;
++ xenbus_dev_error(blkif->be->dev, err, "start xenblkd");
++ }
++}
++
++
++/****************************************************************
++ * sysfs interface for VBD I/O requests
++ */
++
++#define VBD_SHOW(name, format, args...) \
++ static ssize_t show_##name(struct device *_dev, \
++ struct device_attribute *attr, \
++ char *buf) \
++ { \
++ struct xenbus_device *dev = to_xenbus_device(_dev); \
++ struct backend_info *be = dev->dev.driver_data; \
++ \
++ return sprintf(buf, format, ##args); \
++ } \
++ DEVICE_ATTR(name, S_IRUGO, show_##name, NULL)
++
++VBD_SHOW(oo_req, "%d\n", be->blkif->st_oo_req);
++VBD_SHOW(rd_req, "%d\n", be->blkif->st_rd_req);
++VBD_SHOW(wr_req, "%d\n", be->blkif->st_wr_req);
++
++static struct attribute *vbdstat_attrs[] = {
++ &dev_attr_oo_req.attr,
++ &dev_attr_rd_req.attr,
++ &dev_attr_wr_req.attr,
++ NULL
++};
++
++static struct attribute_group vbdstat_group = {
++ .name = "statistics",
++ .attrs = vbdstat_attrs,
++};
++
++VBD_SHOW(physical_device, "%x:%x\n", be->major, be->minor);
++VBD_SHOW(mode, "%s\n", be->mode);
++
++int xenvbd_sysfs_addif(struct xenbus_device *dev)
++{
++ int error;
++
++ error = device_create_file(&dev->dev, &dev_attr_physical_device);
++ if (error)
++ goto fail1;
++
++ error = device_create_file(&dev->dev, &dev_attr_mode);
++ if (error)
++ goto fail2;
++
++ error = sysfs_create_group(&dev->dev.kobj, &vbdstat_group);
++ if (error)
++ goto fail3;
++
++ return 0;
++
++fail3: sysfs_remove_group(&dev->dev.kobj, &vbdstat_group);
++fail2: device_remove_file(&dev->dev, &dev_attr_mode);
++fail1: device_remove_file(&dev->dev, &dev_attr_physical_device);
++ return error;
++}
++
++void xenvbd_sysfs_delif(struct xenbus_device *dev)
++{
++ sysfs_remove_group(&dev->dev.kobj, &vbdstat_group);
++ device_remove_file(&dev->dev, &dev_attr_mode);
++ device_remove_file(&dev->dev, &dev_attr_physical_device);
++}
++
++static int blkback_remove(struct xenbus_device *dev)
++{
++ struct backend_info *be = dev->dev.driver_data;
++
++ DPRINTK("");
++
++ if (be->backend_watch.node) {
++ unregister_xenbus_watch(&be->backend_watch);
++ kfree(be->backend_watch.node);
++ be->backend_watch.node = NULL;
++ }
++
++ if (be->blkif) {
++ blkif_disconnect(be->blkif);
++ vbd_free(&be->blkif->vbd);
++ blkif_free(be->blkif);
++ be->blkif = NULL;
++ }
++
++ if (be->major || be->minor)
++ xenvbd_sysfs_delif(dev);
++
++ kfree(be);
++ dev->dev.driver_data = NULL;
++ return 0;
++}
++
++
++/**
++ * Entry point to this code when a new device is created. Allocate the basic
++ * structures, and watch the store waiting for the hotplug scripts to tell us
++ * the device's physical major and minor numbers. Switch to InitWait.
++ */
++static int blkback_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err;
++ struct backend_info *be = kzalloc(sizeof(struct backend_info),
++ GFP_KERNEL);
++ if (!be) {
++ xenbus_dev_fatal(dev, -ENOMEM,
++ "allocating backend structure");
++ return -ENOMEM;
++ }
++ be->dev = dev;
++ dev->dev.driver_data = be;
++
++ be->blkif = blkif_alloc(dev->otherend_id);
++ if (IS_ERR(be->blkif)) {
++ err = PTR_ERR(be->blkif);
++ be->blkif = NULL;
++ xenbus_dev_fatal(dev, err, "creating block interface");
++ goto fail;
++ }
++
++ /* setup back pointer */
++ be->blkif->be = be;
++
++ err = xenbus_watch_path2(dev, dev->nodename, "physical-device",
++ &be->backend_watch, backend_changed);
++ if (err)
++ goto fail;
++
++ err = xenbus_switch_state(dev, XenbusStateInitWait);
++ if (err)
++ goto fail;
++
++ return 0;
++
++fail:
++ DPRINTK("failed");
++ blkback_remove(dev);
++ return err;
++}
++
++
++/**
++ * Callback received when the hotplug scripts have placed the physical-device
++ * node. Read it and the mode node, and create a vbd. If the frontend is
++ * ready, connect.
++ */
++static void backend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ int err;
++ unsigned major;
++ unsigned minor;
++ struct backend_info *be
++ = container_of(watch, struct backend_info, backend_watch);
++ struct xenbus_device *dev = be->dev;
++
++ DPRINTK("");
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename, "physical-device", "%x:%x",
++ &major, &minor);
++ if (XENBUS_EXIST_ERR(err)) {
++ /* Since this watch will fire once immediately after it is
++ registered, we expect this. Ignore it, and wait for the
++ hotplug scripts. */
++ return;
++ }
++ if (err != 2) {
++ xenbus_dev_fatal(dev, err, "reading physical-device");
++ return;
++ }
++
++ if ((be->major || be->minor) &&
++ ((be->major != major) || (be->minor != minor))) {
++ printk(KERN_WARNING
++ "blkback: changing physical device (from %x:%x to "
++ "%x:%x) not supported.\n", be->major, be->minor,
++ major, minor);
++ return;
++ }
++
++ be->mode = xenbus_read(XBT_NIL, dev->nodename, "mode", NULL);
++ if (IS_ERR(be->mode)) {
++ err = PTR_ERR(be->mode);
++ be->mode = NULL;
++ xenbus_dev_fatal(dev, err, "reading mode");
++ return;
++ }
++
++ if (be->major == 0 && be->minor == 0) {
++ /* Front end dir is a number, which is used as the handle. */
++
++ char *p = strrchr(dev->otherend, '/') + 1;
++ long handle = simple_strtoul(p, NULL, 0);
++
++ be->major = major;
++ be->minor = minor;
++
++ err = vbd_create(be->blkif, handle, major, minor,
++ (NULL == strchr(be->mode, 'w')));
++ if (err) {
++ be->major = be->minor = 0;
++ xenbus_dev_fatal(dev, err, "creating vbd structure");
++ return;
++ }
++
++ err = xenvbd_sysfs_addif(dev);
++ if (err) {
++ vbd_free(&be->blkif->vbd);
++ be->major = be->minor = 0;
++ xenbus_dev_fatal(dev, err, "creating sysfs entries");
++ return;
++ }
++
++ /* We're potentially connected now */
++ update_blkif_status(be->blkif);
++ }
++}
++
++
++/**
++ * Callback received when the frontend's state changes.
++ */
++static void frontend_changed(struct xenbus_device *dev,
++ enum xenbus_state frontend_state)
++{
++ struct backend_info *be = dev->dev.driver_data;
++ int err;
++
++ DPRINTK("");
++
++ switch (frontend_state) {
++ case XenbusStateInitialising:
++ break;
++
++ case XenbusStateInitialised:
++ case XenbusStateConnected:
++ /* Ensure we connect even when two watches fire in
++ close successsion and we miss the intermediate value
++ of frontend_state. */
++ if (dev->state == XenbusStateConnected)
++ break;
++
++ err = connect_ring(be);
++ if (err)
++ break;
++ update_blkif_status(be->blkif);
++ break;
++
++ case XenbusStateClosing:
++ blkif_disconnect(be->blkif);
++ xenbus_switch_state(dev, XenbusStateClosing);
++ break;
++
++ case XenbusStateClosed:
++ device_unregister(&dev->dev);
++ break;
++
++ case XenbusStateUnknown:
++ case XenbusStateInitWait:
++ default:
++ xenbus_dev_fatal(dev, -EINVAL, "saw state %d at frontend",
++ frontend_state);
++ break;
++ }
++}
++
++
++/* ** Connection ** */
++
++
++/**
++ * Write the physical details regarding the block device to the store, and
++ * switch to Connected state.
++ */
++static void connect(struct backend_info *be)
++{
++ struct xenbus_transaction xbt;
++ int err;
++ struct xenbus_device *dev = be->dev;
++
++ DPRINTK("%s", dev->otherend);
++
++ /* Supply the information about the device the frontend needs */
++again:
++ err = xenbus_transaction_start(&xbt);
++
++ if (err) {
++ xenbus_dev_fatal(dev, err, "starting transaction");
++ return;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename, "sectors", "%lu",
++ vbd_size(&be->blkif->vbd));
++ if (err) {
++ xenbus_dev_fatal(dev, err, "writing %s/sectors",
++ dev->nodename);
++ goto abort;
++ }
++
++ /* FIXME: use a typename instead */
++ err = xenbus_printf(xbt, dev->nodename, "info", "%u",
++ vbd_info(&be->blkif->vbd));
++ if (err) {
++ xenbus_dev_fatal(dev, err, "writing %s/info",
++ dev->nodename);
++ goto abort;
++ }
++ err = xenbus_printf(xbt, dev->nodename, "sector-size", "%lu",
++ vbd_secsize(&be->blkif->vbd));
++ if (err) {
++ xenbus_dev_fatal(dev, err, "writing %s/sector-size",
++ dev->nodename);
++ goto abort;
++ }
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err == -EAGAIN)
++ goto again;
++ if (err)
++ xenbus_dev_fatal(dev, err, "ending transaction");
++
++ err = xenbus_switch_state(dev, XenbusStateConnected);
++ if (err)
++ xenbus_dev_fatal(dev, err, "switching to Connected state",
++ dev->nodename);
++
++ return;
++ abort:
++ xenbus_transaction_end(xbt, 1);
++}
++
++
++static int connect_ring(struct backend_info *be)
++{
++ struct xenbus_device *dev = be->dev;
++ unsigned long ring_ref;
++ unsigned int evtchn;
++ int err;
++
++ DPRINTK("%s", dev->otherend);
++
++ err = xenbus_gather(XBT_NIL, dev->otherend, "ring-ref", "%lu", &ring_ref,
++ "event-channel", "%u", &evtchn, NULL);
++ if (err) {
++ xenbus_dev_fatal(dev, err,
++ "reading %s/ring-ref and event-channel",
++ dev->otherend);
++ return err;
++ }
++
++ /* Map the shared frame, irq etc. */
++ err = blkif_map(be->blkif, ring_ref, evtchn);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "mapping ring-ref %lu port %u",
++ ring_ref, evtchn);
++ return err;
++ }
++
++ return 0;
++}
++
++
++/* ** Driver Registration ** */
++
++
++static struct xenbus_device_id blkback_ids[] = {
++ { "vbd" },
++ { "" }
++};
++
++
++static struct xenbus_driver blkback = {
++ .name = "vbd",
++ .owner = THIS_MODULE,
++ .ids = blkback_ids,
++ .probe = blkback_probe,
++ .remove = blkback_remove,
++ .otherend_changed = frontend_changed
++};
++
++
++void blkif_xenbus_init(void)
++{
++ xenbus_register_backend(&blkback);
++}
+diff -urNp linux-2.6/drivers/xen/blkfront/blkfront.c new/drivers/xen/blkfront/blkfront.c
+--- linux-2.6/drivers/xen/blkfront/blkfront.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkfront/blkfront.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,841 @@
++/******************************************************************************
++ * blkfront.c
++ *
++ * XenLinux virtual block-device driver.
++ *
++ * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
++ * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
++ * Copyright (c) 2004, Christian Limpach
++ * Copyright (c) 2004, Andrew Warfield
++ * Copyright (c) 2005, Christopher Clark
++ * Copyright (c) 2005, XenSource Ltd
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/version.h>
++#include "block.h"
++#include <linux/cdrom.h>
++#include <linux/sched.h>
++#include <linux/interrupt.h>
++#include <scsi/scsi.h>
++#include <xen/evtchn.h>
++#include <xen/xenbus.h>
++#include <xen/interface/grant_table.h>
++#include <xen/gnttab.h>
++#include <asm/hypervisor.h>
++
++#define BLKIF_STATE_DISCONNECTED 0
++#define BLKIF_STATE_CONNECTED 1
++#define BLKIF_STATE_SUSPENDED 2
++
++#define MAXIMUM_OUTSTANDING_BLOCK_REQS \
++ (BLKIF_MAX_SEGMENTS_PER_REQUEST * BLK_RING_SIZE)
++#define GRANT_INVALID_REF 0
++
++static void connect(struct blkfront_info *);
++static void blkfront_closing(struct xenbus_device *);
++static int blkfront_remove(struct xenbus_device *);
++static int talk_to_backend(struct xenbus_device *, struct blkfront_info *);
++static int setup_blkring(struct xenbus_device *, struct blkfront_info *);
++
++static void kick_pending_request_queues(struct blkfront_info *);
++
++static irqreturn_t blkif_int(int irq, void *dev_id, struct pt_regs *ptregs);
++static void blkif_restart_queue(void *arg);
++static void blkif_recover(struct blkfront_info *);
++static void blkif_completion(struct blk_shadow *);
++static void blkif_free(struct blkfront_info *, int);
++
++
++/**
++ * Entry point to this code when a new device is created. Allocate the basic
++ * structures and the ring buffer for communication with the backend, and
++ * inform the backend of the appropriate details for those. Switch to
++ * Initialised state.
++ */
++static int blkfront_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err, vdevice, i;
++ struct blkfront_info *info;
++
++ /* FIXME: Use dynamic device id if this is not set. */
++ err = xenbus_scanf(XBT_NIL, dev->nodename,
++ "virtual-device", "%i", &vdevice);
++ if (err != 1) {
++ xenbus_dev_fatal(dev, err, "reading virtual-device");
++ return err;
++ }
++
++ info = kzalloc(sizeof(*info), GFP_KERNEL);
++ if (!info) {
++ xenbus_dev_fatal(dev, -ENOMEM, "allocating info structure");
++ return -ENOMEM;
++ }
++
++ info->xbdev = dev;
++ info->vdevice = vdevice;
++ info->connected = BLKIF_STATE_DISCONNECTED;
++ INIT_WORK(&info->work, blkif_restart_queue, (void *)info);
++
++ for (i = 0; i < BLK_RING_SIZE; i++)
++ info->shadow[i].req.id = i+1;
++ info->shadow[BLK_RING_SIZE-1].req.id = 0x0fffffff;
++
++ /* Front end dir is a number, which is used as the id. */
++ info->handle = simple_strtoul(strrchr(dev->nodename,'/')+1, NULL, 0);
++ dev->dev.driver_data = info;
++
++ err = talk_to_backend(dev, info);
++ if (err) {
++ kfree(info);
++ dev->dev.driver_data = NULL;
++ return err;
++ }
++
++ return 0;
++}
++
++
++/**
++ * We are reconnecting to the backend, due to a suspend/resume, or a backend
++ * driver restart. We tear down our blkif structure and recreate it, but
++ * leave the device-layer structures intact so that this is transparent to the
++ * rest of the kernel.
++ */
++static int blkfront_resume(struct xenbus_device *dev)
++{
++ struct blkfront_info *info = dev->dev.driver_data;
++ int err;
++
++ DPRINTK("blkfront_resume: %s\n", dev->nodename);
++
++ blkif_free(info, 1);
++
++ err = talk_to_backend(dev, info);
++ if (!err)
++ blkif_recover(info);
++
++ return err;
++}
++
++
++/* Common code used when first setting up, and when resuming. */
++static int talk_to_backend(struct xenbus_device *dev,
++ struct blkfront_info *info)
++{
++ const char *message = NULL;
++ struct xenbus_transaction xbt;
++ int err;
++
++ /* Create shared ring, alloc event channel. */
++ err = setup_blkring(dev, info);
++ if (err)
++ goto out;
++
++again:
++ err = xenbus_transaction_start(&xbt);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "starting transaction");
++ goto destroy_blkring;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename,
++ "ring-ref","%u", info->ring_ref);
++ if (err) {
++ message = "writing ring-ref";
++ goto abort_transaction;
++ }
++ err = xenbus_printf(xbt, dev->nodename,
++ "event-channel", "%u", info->evtchn);
++ if (err) {
++ message = "writing event-channel";
++ goto abort_transaction;
++ }
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err) {
++ if (err == -EAGAIN)
++ goto again;
++ xenbus_dev_fatal(dev, err, "completing transaction");
++ goto destroy_blkring;
++ }
++
++ xenbus_switch_state(dev, XenbusStateInitialised);
++
++ return 0;
++
++ abort_transaction:
++ xenbus_transaction_end(xbt, 1);
++ if (message)
++ xenbus_dev_fatal(dev, err, "%s", message);
++ destroy_blkring:
++ blkif_free(info, 0);
++ out:
++ return err;
++}
++
++
++static int setup_blkring(struct xenbus_device *dev,
++ struct blkfront_info *info)
++{
++ blkif_sring_t *sring;
++ int err;
++
++ info->ring_ref = GRANT_INVALID_REF;
++
++ sring = (blkif_sring_t *)__get_free_page(GFP_KERNEL);
++ if (!sring) {
++ xenbus_dev_fatal(dev, -ENOMEM, "allocating shared ring");
++ return -ENOMEM;
++ }
++ SHARED_RING_INIT(sring);
++ FRONT_RING_INIT(&info->ring, sring, PAGE_SIZE);
++
++ err = xenbus_grant_ring(dev, virt_to_mfn(info->ring.sring));
++ if (err < 0) {
++ free_page((unsigned long)sring);
++ info->ring.sring = NULL;
++ goto fail;
++ }
++ info->ring_ref = err;
++
++ err = xenbus_alloc_evtchn(dev, &info->evtchn);
++ if (err)
++ goto fail;
++
++ err = bind_evtchn_to_irqhandler(
++ info->evtchn, blkif_int, SA_SAMPLE_RANDOM, "blkif", info);
++ if (err <= 0) {
++ xenbus_dev_fatal(dev, err,
++ "bind_evtchn_to_irqhandler failed");
++ goto fail;
++ }
++ info->irq = err;
++
++ return 0;
++fail:
++ blkif_free(info, 0);
++ return err;
++}
++
++
++/**
++ * Callback received when the backend's state changes.
++ */
++static void backend_changed(struct xenbus_device *dev,
++ enum xenbus_state backend_state)
++{
++ struct blkfront_info *info = dev->dev.driver_data;
++ struct block_device *bd;
++
++ DPRINTK("blkfront:backend_changed.\n");
++
++ switch (backend_state) {
++ case XenbusStateUnknown:
++ case XenbusStateInitialising:
++ case XenbusStateInitWait:
++ case XenbusStateInitialised:
++ case XenbusStateClosed:
++ break;
++
++ case XenbusStateConnected:
++ connect(info);
++ break;
++
++ case XenbusStateClosing:
++ bd = bdget(info->dev);
++ if (bd == NULL)
++ xenbus_dev_fatal(dev, -ENODEV, "bdget failed");
++
++ mutex_lock(&bd->bd_mutex);
++ if (info->users > 0)
++ xenbus_dev_error(dev, -EBUSY,
++ "Device in use; refusing to close");
++ else
++ blkfront_closing(dev);
++ mutex_unlock(&bd->bd_mutex);
++ bdput(bd);
++ break;
++ }
++}
++
++
++/* ** Connection ** */
++
++
++/*
++ * Invoked when the backend is finally 'ready' (and has told produced
++ * the details about the physical device - #sectors, size, etc).
++ */
++static void connect(struct blkfront_info *info)
++{
++ unsigned long sectors, sector_size;
++ unsigned int binfo;
++ int err;
++
++ if ((info->connected == BLKIF_STATE_CONNECTED) ||
++ (info->connected == BLKIF_STATE_SUSPENDED) )
++ return;
++
++ DPRINTK("blkfront.c:connect:%s.\n", info->xbdev->otherend);
++
++ err = xenbus_gather(XBT_NIL, info->xbdev->otherend,
++ "sectors", "%lu", §ors,
++ "info", "%u", &binfo,
++ "sector-size", "%lu", §or_size,
++ NULL);
++ if (err) {
++ xenbus_dev_fatal(info->xbdev, err,
++ "reading backend fields at %s",
++ info->xbdev->otherend);
++ return;
++ }
++
++ err = xlvbd_add(sectors, info->vdevice, binfo, sector_size, info);
++ if (err) {
++ xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s",
++ info->xbdev->otherend);
++ return;
++ }
++
++ (void)xenbus_switch_state(info->xbdev, XenbusStateConnected);
++
++ /* Kick pending requests. */
++ spin_lock_irq(&blkif_io_lock);
++ info->connected = BLKIF_STATE_CONNECTED;
++ kick_pending_request_queues(info);
++ spin_unlock_irq(&blkif_io_lock);
++
++ add_disk(info->gd);
++}
++
++/**
++ * Handle the change of state of the backend to Closing. We must delete our
++ * device-layer structures now, to ensure that writes are flushed through to
++ * the backend. Once is this done, we can switch to Closed in
++ * acknowledgement.
++ */
++static void blkfront_closing(struct xenbus_device *dev)
++{
++ struct blkfront_info *info = dev->dev.driver_data;
++ unsigned long flags;
++
++ DPRINTK("blkfront_closing: %s removed\n", dev->nodename);
++
++ if (info->rq == NULL)
++ return;
++
++ spin_lock_irqsave(&blkif_io_lock, flags);
++ /* No more blkif_request(). */
++ blk_stop_queue(info->rq);
++ /* No more gnttab callback work. */
++ gnttab_cancel_free_callback(&info->callback);
++ flush_scheduled_work();
++ spin_unlock_irqrestore(&blkif_io_lock, flags);
++
++ xlvbd_del(info);
++
++ xenbus_switch_state(dev, XenbusStateClosed);
++}
++
++
++static int blkfront_remove(struct xenbus_device *dev)
++{
++ struct blkfront_info *info = dev->dev.driver_data;
++
++ DPRINTK("blkfront_remove: %s removed\n", dev->nodename);
++
++ blkif_free(info, 0);
++
++ kfree(info);
++
++ return 0;
++}
++
++
++static inline int GET_ID_FROM_FREELIST(
++ struct blkfront_info *info)
++{
++ unsigned long free = info->shadow_free;
++ BUG_ON(free > BLK_RING_SIZE);
++ info->shadow_free = info->shadow[free].req.id;
++ info->shadow[free].req.id = 0x0fffffee; /* debug */
++ return free;
++}
++
++static inline void ADD_ID_TO_FREELIST(
++ struct blkfront_info *info, unsigned long id)
++{
++ info->shadow[id].req.id = info->shadow_free;
++ info->shadow[id].request = 0;
++ info->shadow_free = id;
++}
++
++static inline void flush_requests(struct blkfront_info *info)
++{
++ int notify;
++
++ RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&info->ring, notify);
++
++ if (notify)
++ notify_remote_via_irq(info->irq);
++}
++
++static void kick_pending_request_queues(struct blkfront_info *info)
++{
++ if (!RING_FULL(&info->ring)) {
++ /* Re-enable calldowns. */
++ blk_start_queue(info->rq);
++ /* Kick things off immediately. */
++ do_blkif_request(info->rq);
++ }
++}
++
++static void blkif_restart_queue(void *arg)
++{
++ struct blkfront_info *info = (struct blkfront_info *)arg;
++ spin_lock_irq(&blkif_io_lock);
++ if (info->connected == BLKIF_STATE_CONNECTED)
++ kick_pending_request_queues(info);
++ spin_unlock_irq(&blkif_io_lock);
++}
++
++static void blkif_restart_queue_callback(void *arg)
++{
++ struct blkfront_info *info = (struct blkfront_info *)arg;
++ schedule_work(&info->work);
++}
++
++int blkif_open(struct inode *inode, struct file *filep)
++{
++ struct blkfront_info *info = inode->i_bdev->bd_disk->private_data;
++ info->users++;
++ return 0;
++}
++
++
++int blkif_release(struct inode *inode, struct file *filep)
++{
++ struct blkfront_info *info = inode->i_bdev->bd_disk->private_data;
++ info->users--;
++ if (info->users == 0) {
++ /* Check whether we have been instructed to close. We will
++ have ignored this request initially, as the device was
++ still mounted. */
++ struct xenbus_device * dev = info->xbdev;
++ enum xenbus_state state = xenbus_read_driver_state(dev->otherend);
++
++ if (state == XenbusStateClosing)
++ blkfront_closing(dev);
++ }
++ return 0;
++}
++
++
++int blkif_ioctl(struct inode *inode, struct file *filep,
++ unsigned command, unsigned long argument)
++{
++ int i;
++
++ DPRINTK_IOCTL("command: 0x%x, argument: 0x%lx, dev: 0x%04x\n",
++ command, (long)argument, inode->i_rdev);
++
++ switch (command) {
++ case CDROMMULTISESSION:
++ DPRINTK("FIXME: support multisession CDs later\n");
++ for (i = 0; i < sizeof(struct cdrom_multisession); i++)
++ if (put_user(0, (char __user *)(argument + i)))
++ return -EFAULT;
++ return 0;
++
++ default:
++ /*printk(KERN_ALERT "ioctl %08x not supported by Xen blkdev\n",
++ command);*/
++ return -EINVAL; /* same return as native Linux */
++ }
++
++ return 0;
++}
++
++
++int blkif_getgeo(struct block_device *bd, struct hd_geometry *hg)
++{
++ /* We don't have real geometry info, but let's at least return
++ values consistent with the size of the device */
++ sector_t nsect = get_capacity(bd->bd_disk);
++ sector_t cylinders = nsect;
++
++ hg->heads = 0xff;
++ hg->sectors = 0x3f;
++ sector_div(cylinders, hg->heads * hg->sectors);
++ hg->cylinders = cylinders;
++ if ((sector_t)(hg->cylinders + 1) * hg->heads * hg->sectors < nsect)
++ hg->cylinders = 0xffff;
++ return 0;
++}
++
++
++/*
++ * blkif_queue_request
++ *
++ * request block io
++ *
++ * id: for guest use only.
++ * operation: BLKIF_OP_{READ,WRITE,PROBE}
++ * buffer: buffer to read/write into. this should be a
++ * virtual address in the guest os.
++ */
++static int blkif_queue_request(struct request *req)
++{
++ struct blkfront_info *info = req->rq_disk->private_data;
++ unsigned long buffer_mfn;
++ blkif_request_t *ring_req;
++ struct bio *bio;
++ struct bio_vec *bvec;
++ int idx;
++ unsigned long id;
++ unsigned int fsect, lsect;
++ int ref;
++ grant_ref_t gref_head;
++
++ if (unlikely(info->connected != BLKIF_STATE_CONNECTED))
++ return 1;
++
++ if (gnttab_alloc_grant_references(
++ BLKIF_MAX_SEGMENTS_PER_REQUEST, &gref_head) < 0) {
++ gnttab_request_free_callback(
++ &info->callback,
++ blkif_restart_queue_callback,
++ info,
++ BLKIF_MAX_SEGMENTS_PER_REQUEST);
++ return 1;
++ }
++
++ /* Fill out a communications ring structure. */
++ ring_req = RING_GET_REQUEST(&info->ring, info->ring.req_prod_pvt);
++ id = GET_ID_FROM_FREELIST(info);
++ info->shadow[id].request = (unsigned long)req;
++
++ ring_req->id = id;
++ ring_req->operation = rq_data_dir(req) ?
++ BLKIF_OP_WRITE : BLKIF_OP_READ;
++ ring_req->sector_number = (blkif_sector_t)req->sector;
++ ring_req->handle = info->handle;
++
++ ring_req->nr_segments = 0;
++ rq_for_each_bio (bio, req) {
++ bio_for_each_segment (bvec, bio, idx) {
++ BUG_ON(ring_req->nr_segments
++ == BLKIF_MAX_SEGMENTS_PER_REQUEST);
++ buffer_mfn = page_to_phys(bvec->bv_page) >> PAGE_SHIFT;
++ fsect = bvec->bv_offset >> 9;
++ lsect = fsect + (bvec->bv_len >> 9) - 1;
++ /* install a grant reference. */
++ ref = gnttab_claim_grant_reference(&gref_head);
++ BUG_ON(ref == -ENOSPC);
++
++ gnttab_grant_foreign_access_ref(
++ ref,
++ info->xbdev->otherend_id,
++ buffer_mfn,
++ rq_data_dir(req) );
++
++ info->shadow[id].frame[ring_req->nr_segments] =
++ mfn_to_pfn(buffer_mfn);
++
++ ring_req->seg[ring_req->nr_segments] =
++ (struct blkif_request_segment) {
++ .gref = ref,
++ .first_sect = fsect,
++ .last_sect = lsect };
++
++ ring_req->nr_segments++;
++ }
++ }
++
++ info->ring.req_prod_pvt++;
++
++ /* Keep a private copy so we can reissue requests when recovering. */
++ info->shadow[id].req = *ring_req;
++
++ gnttab_free_grant_references(gref_head);
++
++ return 0;
++}
++
++/*
++ * do_blkif_request
++ * read a block; request is in a request queue
++ */
++void do_blkif_request(request_queue_t *rq)
++{
++ struct blkfront_info *info = NULL;
++ struct request *req;
++ int queued;
++
++ DPRINTK("Entered do_blkif_request\n");
++
++ queued = 0;
++
++ while ((req = elv_next_request(rq)) != NULL) {
++ info = req->rq_disk->private_data;
++ if (!blk_fs_request(req)) {
++ end_request(req, 0);
++ continue;
++ }
++
++ if (RING_FULL(&info->ring))
++ goto wait;
++
++ DPRINTK("do_blk_req %p: cmd %p, sec %lx, "
++ "(%u/%li) buffer:%p [%s]\n",
++ req, req->cmd, req->sector, req->current_nr_sectors,
++ req->nr_sectors, req->buffer,
++ rq_data_dir(req) ? "write" : "read");
++
++
++ blkdev_dequeue_request(req);
++ if (blkif_queue_request(req)) {
++ blk_requeue_request(rq, req);
++ wait:
++ /* Avoid pointless unplugs. */
++ blk_stop_queue(rq);
++ break;
++ }
++
++ queued++;
++ }
++
++ if (queued != 0)
++ flush_requests(info);
++}
++
++
++static irqreturn_t blkif_int(int irq, void *dev_id, struct pt_regs *ptregs)
++{
++ struct request *req;
++ blkif_response_t *bret;
++ RING_IDX i, rp;
++ unsigned long flags;
++ struct blkfront_info *info = (struct blkfront_info *)dev_id;
++
++ spin_lock_irqsave(&blkif_io_lock, flags);
++
++ if (unlikely(info->connected != BLKIF_STATE_CONNECTED)) {
++ spin_unlock_irqrestore(&blkif_io_lock, flags);
++ return IRQ_HANDLED;
++ }
++
++ again:
++ rp = info->ring.sring->rsp_prod;
++ rmb(); /* Ensure we see queued responses up to 'rp'. */
++
++ for (i = info->ring.rsp_cons; i != rp; i++) {
++ unsigned long id;
++ int ret;
++
++ bret = RING_GET_RESPONSE(&info->ring, i);
++ id = bret->id;
++ req = (struct request *)info->shadow[id].request;
++
++ blkif_completion(&info->shadow[id]);
++
++ ADD_ID_TO_FREELIST(info, id);
++
++ switch (bret->operation) {
++ case BLKIF_OP_READ:
++ case BLKIF_OP_WRITE:
++ if (unlikely(bret->status != BLKIF_RSP_OKAY))
++ DPRINTK("Bad return from blkdev data "
++ "request: %x\n", bret->status);
++
++ ret = end_that_request_first(
++ req, (bret->status == BLKIF_RSP_OKAY),
++ req->hard_nr_sectors);
++ BUG_ON(ret);
++ end_that_request_last(
++ req, (bret->status == BLKIF_RSP_OKAY));
++ break;
++ default:
++ BUG();
++ }
++ }
++
++ info->ring.rsp_cons = i;
++
++ if (i != info->ring.req_prod_pvt) {
++ int more_to_do;
++ RING_FINAL_CHECK_FOR_RESPONSES(&info->ring, more_to_do);
++ if (more_to_do)
++ goto again;
++ } else
++ info->ring.sring->rsp_event = i + 1;
++
++ kick_pending_request_queues(info);
++
++ spin_unlock_irqrestore(&blkif_io_lock, flags);
++
++ return IRQ_HANDLED;
++}
++
++static void blkif_free(struct blkfront_info *info, int suspend)
++{
++ /* Prevent new requests being issued until we fix things up. */
++ spin_lock_irq(&blkif_io_lock);
++ info->connected = suspend ?
++ BLKIF_STATE_SUSPENDED : BLKIF_STATE_DISCONNECTED;
++ /* No more blkif_request(). */
++ if (info->rq)
++ blk_stop_queue(info->rq);
++ /* No more gnttab callback work. */
++ gnttab_cancel_free_callback(&info->callback);
++ flush_scheduled_work();
++ spin_unlock_irq(&blkif_io_lock);
++
++ /* Free resources associated with old device channel. */
++ if (info->ring_ref != GRANT_INVALID_REF) {
++ gnttab_end_foreign_access(info->ring_ref, 0,
++ (unsigned long)info->ring.sring);
++ info->ring_ref = GRANT_INVALID_REF;
++ info->ring.sring = NULL;
++ }
++ if (info->irq)
++ unbind_from_irqhandler(info->irq, info);
++ info->evtchn = info->irq = 0;
++
++}
++
++static void blkif_completion(struct blk_shadow *s)
++{
++ int i;
++ for (i = 0; i < s->req.nr_segments; i++)
++ gnttab_end_foreign_access(s->req.seg[i].gref, 0, 0UL);
++}
++
++static void blkif_recover(struct blkfront_info *info)
++{
++ int i;
++ blkif_request_t *req;
++ struct blk_shadow *copy;
++ int j;
++
++ /* Stage 1: Make a safe copy of the shadow state. */
++ copy = kmalloc(sizeof(info->shadow), GFP_KERNEL | __GFP_NOFAIL);
++ memcpy(copy, info->shadow, sizeof(info->shadow));
++
++ /* Stage 2: Set up free list. */
++ memset(&info->shadow, 0, sizeof(info->shadow));
++ for (i = 0; i < BLK_RING_SIZE; i++)
++ info->shadow[i].req.id = i+1;
++ info->shadow_free = info->ring.req_prod_pvt;
++ info->shadow[BLK_RING_SIZE-1].req.id = 0x0fffffff;
++
++ /* Stage 3: Find pending requests and requeue them. */
++ for (i = 0; i < BLK_RING_SIZE; i++) {
++ /* Not in use? */
++ if (copy[i].request == 0)
++ continue;
++
++ /* Grab a request slot and copy shadow state into it. */
++ req = RING_GET_REQUEST(
++ &info->ring, info->ring.req_prod_pvt);
++ *req = copy[i].req;
++
++ /* We get a new request id, and must reset the shadow state. */
++ req->id = GET_ID_FROM_FREELIST(info);
++ memcpy(&info->shadow[req->id], ©[i], sizeof(copy[i]));
++
++ /* Rewrite any grant references invalidated by susp/resume. */
++ for (j = 0; j < req->nr_segments; j++)
++ gnttab_grant_foreign_access_ref(
++ req->seg[j].gref,
++ info->xbdev->otherend_id,
++ pfn_to_mfn(info->shadow[req->id].frame[j]),
++ rq_data_dir(
++ (struct request *)
++ info->shadow[req->id].request));
++ info->shadow[req->id].req = *req;
++
++ info->ring.req_prod_pvt++;
++ }
++
++ kfree(copy);
++
++ (void)xenbus_switch_state(info->xbdev, XenbusStateConnected);
++
++ spin_lock_irq(&blkif_io_lock);
++
++ /* Now safe for us to use the shared ring */
++ info->connected = BLKIF_STATE_CONNECTED;
++
++ /* Send off requeued requests */
++ flush_requests(info);
++
++ /* Kick any other new requests queued since we resumed */
++ kick_pending_request_queues(info);
++
++ spin_unlock_irq(&blkif_io_lock);
++}
++
++
++/* ** Driver Registration ** */
++
++
++static struct xenbus_device_id blkfront_ids[] = {
++ { "vbd" },
++ { "" }
++};
++
++
++static struct xenbus_driver blkfront = {
++ .name = "vbd",
++ .owner = THIS_MODULE,
++ .ids = blkfront_ids,
++ .probe = blkfront_probe,
++ .remove = blkfront_remove,
++ .resume = blkfront_resume,
++ .otherend_changed = backend_changed,
++};
++
++
++static int __init xlblk_init(void)
++{
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ return xenbus_register_frontend(&blkfront);
++}
++module_init(xlblk_init);
++
++
++static void xlblk_exit(void)
++{
++ return xenbus_unregister_driver(&blkfront);
++}
++module_exit(xlblk_exit);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/blkfront/block.h new/drivers/xen/blkfront/block.h
+--- linux-2.6/drivers/xen/blkfront/block.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkfront/block.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,156 @@
++/******************************************************************************
++ * block.h
++ *
++ * Shared definitions between all levels of XenLinux Virtual block devices.
++ *
++ * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
++ * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
++ * Copyright (c) 2004-2005, Christian Limpach
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __XEN_DRIVERS_BLOCK_H__
++#define __XEN_DRIVERS_BLOCK_H__
++
++#include <linux/config.h>
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/fs.h>
++#include <linux/hdreg.h>
++#include <linux/blkdev.h>
++#include <linux/major.h>
++#include <linux/devfs_fs_kernel.h>
++#include <asm/hypervisor.h>
++#include <xen/xenbus.h>
++#include <xen/gnttab.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/io/blkif.h>
++#include <xen/interface/io/ring.h>
++#include <asm/io.h>
++#include <asm/atomic.h>
++#include <asm/uaccess.h>
++
++#if 1
++#define IPRINTK(fmt, args...) \
++ printk(KERN_INFO "xen_blk: " fmt, ##args)
++#else
++#define IPRINTK(fmt, args...) ((void)0)
++#endif
++
++#if 1
++#define WPRINTK(fmt, args...) \
++ printk(KERN_WARNING "xen_blk: " fmt, ##args)
++#else
++#define WPRINTK(fmt, args...) ((void)0)
++#endif
++
++#define DPRINTK(_f, _a...) pr_debug(_f, ## _a)
++
++#if 0
++#define DPRINTK_IOCTL(_f, _a...) printk(KERN_ALERT _f, ## _a)
++#else
++#define DPRINTK_IOCTL(_f, _a...) ((void)0)
++#endif
++
++struct xlbd_type_info
++{
++ int partn_shift;
++ int disks_per_major;
++ char *devname;
++ char *diskname;
++};
++
++struct xlbd_major_info
++{
++ int major;
++ int index;
++ int usage;
++ struct xlbd_type_info *type;
++};
++
++struct blk_shadow {
++ blkif_request_t req;
++ unsigned long request;
++ unsigned long frame[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++};
++
++#define BLK_RING_SIZE __RING_SIZE((blkif_sring_t *)0, PAGE_SIZE)
++
++/*
++ * We have one of these per vbd, whether ide, scsi or 'other'. They
++ * hang in private_data off the gendisk structure. We may end up
++ * putting all kinds of interesting stuff here :-)
++ */
++struct blkfront_info
++{
++ struct xenbus_device *xbdev;
++ dev_t dev;
++ struct gendisk *gd;
++ int vdevice;
++ blkif_vdev_t handle;
++ int connected;
++ int ring_ref;
++ blkif_front_ring_t ring;
++ unsigned int evtchn, irq;
++ struct xlbd_major_info *mi;
++ request_queue_t *rq;
++ struct work_struct work;
++ struct gnttab_free_callback callback;
++ struct blk_shadow shadow[BLK_RING_SIZE];
++ unsigned long shadow_free;
++
++ /**
++ * The number of people holding this device open. We won't allow a
++ * hot-unplug unless this is 0.
++ */
++ int users;
++};
++
++extern spinlock_t blkif_io_lock;
++
++extern int blkif_open(struct inode *inode, struct file *filep);
++extern int blkif_release(struct inode *inode, struct file *filep);
++extern int blkif_ioctl(struct inode *inode, struct file *filep,
++ unsigned command, unsigned long argument);
++extern int blkif_getgeo(struct block_device *, struct hd_geometry *);
++extern int blkif_check(dev_t dev);
++extern int blkif_revalidate(dev_t dev);
++extern void do_blkif_request (request_queue_t *rq);
++
++/* Virtual block-device subsystem. */
++/* Note that xlvbd_add doesn't call add_disk for you: you're expected
++ to call add_disk on info->gd once the disk is properly connected
++ up. */
++int xlvbd_add(blkif_sector_t capacity, int device,
++ u16 vdisk_info, u16 sector_size, struct blkfront_info *info);
++void xlvbd_del(struct blkfront_info *info);
++
++#endif /* __XEN_DRIVERS_BLOCK_H__ */
+diff -urNp linux-2.6/drivers/xen/blkfront/Kconfig new/drivers/xen/blkfront/Kconfig
+--- linux-2.6/drivers/xen/blkfront/Kconfig 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkfront/Kconfig 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,6 @@
++
++config XENBLOCK
++ tristate "Block device driver"
++ depends on ARCH_XEN
++ help
++ Block device driver for Xen
+diff -urNp linux-2.6/drivers/xen/blkfront/Makefile new/drivers/xen/blkfront/Makefile
+--- linux-2.6/drivers/xen/blkfront/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkfront/Makefile 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,5 @@
++
++obj-$(CONFIG_XEN_BLKDEV_FRONTEND) := xenblk.o
++
++xenblk-objs := blkfront.o vbd.o
++
+diff -urNp linux-2.6/drivers/xen/blkfront/vbd.c new/drivers/xen/blkfront/vbd.c
+--- linux-2.6/drivers/xen/blkfront/vbd.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/blkfront/vbd.c 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,318 @@
++/******************************************************************************
++ * vbd.c
++ *
++ * XenLinux virtual block-device driver (xvd).
++ *
++ * Copyright (c) 2003-2004, Keir Fraser & Steve Hand
++ * Modifications by Mark A. Williamson are (c) Intel Research Cambridge
++ * Copyright (c) 2004-2005, Christian Limpach
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include "block.h"
++#include <linux/blkdev.h>
++#include <linux/list.h>
++
++#define BLKIF_MAJOR(dev) ((dev)>>8)
++#define BLKIF_MINOR(dev) ((dev) & 0xff)
++
++/*
++ * For convenience we distinguish between ide, scsi and 'other' (i.e.,
++ * potentially combinations of the two) in the naming scheme and in a few other
++ * places.
++ */
++
++#define NUM_IDE_MAJORS 10
++#define NUM_SCSI_MAJORS 9
++#define NUM_VBD_MAJORS 1
++
++static struct xlbd_type_info xlbd_ide_type = {
++ .partn_shift = 6,
++ .disks_per_major = 2,
++ .devname = "ide",
++ .diskname = "hd",
++};
++
++static struct xlbd_type_info xlbd_scsi_type = {
++ .partn_shift = 4,
++ .disks_per_major = 16,
++ .devname = "sd",
++ .diskname = "sd",
++};
++
++static struct xlbd_type_info xlbd_vbd_type = {
++ .partn_shift = 4,
++ .disks_per_major = 16,
++ .devname = "xvd",
++ .diskname = "xvd",
++};
++
++static struct xlbd_major_info *major_info[NUM_IDE_MAJORS + NUM_SCSI_MAJORS +
++ NUM_VBD_MAJORS];
++
++#define XLBD_MAJOR_IDE_START 0
++#define XLBD_MAJOR_SCSI_START (NUM_IDE_MAJORS)
++#define XLBD_MAJOR_VBD_START (NUM_IDE_MAJORS + NUM_SCSI_MAJORS)
++
++#define XLBD_MAJOR_IDE_RANGE XLBD_MAJOR_IDE_START ... XLBD_MAJOR_SCSI_START - 1
++#define XLBD_MAJOR_SCSI_RANGE XLBD_MAJOR_SCSI_START ... XLBD_MAJOR_VBD_START - 1
++#define XLBD_MAJOR_VBD_RANGE XLBD_MAJOR_VBD_START ... XLBD_MAJOR_VBD_START + NUM_VBD_MAJORS - 1
++
++/* Information about our VBDs. */
++#define MAX_VBDS 64
++static LIST_HEAD(vbds_list);
++
++static struct block_device_operations xlvbd_block_fops =
++{
++ .owner = THIS_MODULE,
++ .open = blkif_open,
++ .release = blkif_release,
++ .ioctl = blkif_ioctl,
++ .getgeo = blkif_getgeo
++};
++
++DEFINE_SPINLOCK(blkif_io_lock);
++
++static struct xlbd_major_info *
++xlbd_alloc_major_info(int major, int minor, int index)
++{
++ struct xlbd_major_info *ptr;
++
++ ptr = kzalloc(sizeof(struct xlbd_major_info), GFP_KERNEL);
++ if (ptr == NULL)
++ return NULL;
++
++ ptr->major = major;
++
++ switch (index) {
++ case XLBD_MAJOR_IDE_RANGE:
++ ptr->type = &xlbd_ide_type;
++ ptr->index = index - XLBD_MAJOR_IDE_START;
++ break;
++ case XLBD_MAJOR_SCSI_RANGE:
++ ptr->type = &xlbd_scsi_type;
++ ptr->index = index - XLBD_MAJOR_SCSI_START;
++ break;
++ case XLBD_MAJOR_VBD_RANGE:
++ ptr->type = &xlbd_vbd_type;
++ ptr->index = index - XLBD_MAJOR_VBD_START;
++ break;
++ }
++
++ printk("Registering block device major %i\n", ptr->major);
++ if (register_blkdev(ptr->major, ptr->type->devname)) {
++ WPRINTK("can't get major %d with name %s\n",
++ ptr->major, ptr->type->devname);
++ kfree(ptr);
++ return NULL;
++ }
++
++ devfs_mk_dir(ptr->type->devname);
++ major_info[index] = ptr;
++ return ptr;
++}
++
++static struct xlbd_major_info *
++xlbd_get_major_info(int vdevice)
++{
++ struct xlbd_major_info *mi;
++ int major, minor, index;
++
++ major = BLKIF_MAJOR(vdevice);
++ minor = BLKIF_MINOR(vdevice);
++
++ switch (major) {
++ case IDE0_MAJOR: index = 0; break;
++ case IDE1_MAJOR: index = 1; break;
++ case IDE2_MAJOR: index = 2; break;
++ case IDE3_MAJOR: index = 3; break;
++ case IDE4_MAJOR: index = 4; break;
++ case IDE5_MAJOR: index = 5; break;
++ case IDE6_MAJOR: index = 6; break;
++ case IDE7_MAJOR: index = 7; break;
++ case IDE8_MAJOR: index = 8; break;
++ case IDE9_MAJOR: index = 9; break;
++ case SCSI_DISK0_MAJOR: index = 10; break;
++ case SCSI_DISK1_MAJOR ... SCSI_DISK7_MAJOR:
++ index = 11 + major - SCSI_DISK1_MAJOR;
++ break;
++ case SCSI_CDROM_MAJOR: index = 18; break;
++ default: index = 19; break;
++ }
++
++ mi = ((major_info[index] != NULL) ? major_info[index] :
++ xlbd_alloc_major_info(major, minor, index));
++ if (mi)
++ mi->usage++;
++ return mi;
++}
++
++static void
++xlbd_put_major_info(struct xlbd_major_info *mi)
++{
++ mi->usage--;
++ /* XXX: release major if 0 */
++}
++
++static int
++xlvbd_init_blk_queue(struct gendisk *gd, u16 sector_size)
++{
++ request_queue_t *rq;
++
++ rq = blk_init_queue(do_blkif_request, &blkif_io_lock);
++ if (rq == NULL)
++ return -1;
++
++ elevator_init(rq, "noop");
++
++ /* Hard sector size and max sectors impersonate the equiv. hardware. */
++ blk_queue_hardsect_size(rq, sector_size);
++ blk_queue_max_sectors(rq, 512);
++
++ /* Each segment in a request is up to an aligned page in size. */
++ blk_queue_segment_boundary(rq, PAGE_SIZE - 1);
++ blk_queue_max_segment_size(rq, PAGE_SIZE);
++
++ /* Ensure a merged request will fit in a single I/O ring slot. */
++ blk_queue_max_phys_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST);
++ blk_queue_max_hw_segments(rq, BLKIF_MAX_SEGMENTS_PER_REQUEST);
++
++ /* Make sure buffer addresses are sector-aligned. */
++ blk_queue_dma_alignment(rq, 511);
++
++ gd->queue = rq;
++
++ return 0;
++}
++
++static int
++xlvbd_alloc_gendisk(int minor, blkif_sector_t capacity, int vdevice,
++ u16 vdisk_info, u16 sector_size,
++ struct blkfront_info *info)
++{
++ struct gendisk *gd;
++ struct xlbd_major_info *mi;
++ int nr_minors = 1;
++ int err = -ENODEV;
++
++ BUG_ON(info->gd != NULL);
++ BUG_ON(info->mi != NULL);
++ BUG_ON(info->rq != NULL);
++
++ mi = xlbd_get_major_info(vdevice);
++ if (mi == NULL)
++ goto out;
++ info->mi = mi;
++
++ if ((minor & ((1 << mi->type->partn_shift) - 1)) == 0)
++ nr_minors = 1 << mi->type->partn_shift;
++
++ gd = alloc_disk(nr_minors);
++ if (gd == NULL)
++ goto out;
++
++ if (nr_minors > 1)
++ sprintf(gd->disk_name, "%s%c", mi->type->diskname,
++ 'a' + mi->index * mi->type->disks_per_major +
++ (minor >> mi->type->partn_shift));
++ else
++ sprintf(gd->disk_name, "%s%c%d", mi->type->diskname,
++ 'a' + mi->index * mi->type->disks_per_major +
++ (minor >> mi->type->partn_shift),
++ minor & ((1 << mi->type->partn_shift) - 1));
++
++ gd->major = mi->major;
++ gd->first_minor = minor;
++ gd->fops = &xlvbd_block_fops;
++ gd->private_data = info;
++ gd->driverfs_dev = &(info->xbdev->dev);
++ set_capacity(gd, capacity);
++
++ if (xlvbd_init_blk_queue(gd, sector_size)) {
++ del_gendisk(gd);
++ goto out;
++ }
++
++ info->rq = gd->queue;
++
++ if (vdisk_info & VDISK_READONLY)
++ set_disk_ro(gd, 1);
++
++ if (vdisk_info & VDISK_REMOVABLE)
++ gd->flags |= GENHD_FL_REMOVABLE;
++
++ if (vdisk_info & VDISK_CDROM)
++ gd->flags |= GENHD_FL_CD;
++
++ info->gd = gd;
++
++ return 0;
++
++ out:
++ if (mi)
++ xlbd_put_major_info(mi);
++ info->mi = NULL;
++ return err;
++}
++
++int
++xlvbd_add(blkif_sector_t capacity, int vdevice, u16 vdisk_info,
++ u16 sector_size, struct blkfront_info *info)
++{
++ struct block_device *bd;
++ int err = 0;
++
++ info->dev = MKDEV(BLKIF_MAJOR(vdevice), BLKIF_MINOR(vdevice));
++
++ bd = bdget(info->dev);
++ if (bd == NULL)
++ return -ENODEV;
++
++ err = xlvbd_alloc_gendisk(BLKIF_MINOR(vdevice), capacity, vdevice,
++ vdisk_info, sector_size, info);
++
++ bdput(bd);
++ return err;
++}
++
++void
++xlvbd_del(struct blkfront_info *info)
++{
++ if (info->mi == NULL)
++ return;
++
++ BUG_ON(info->gd == NULL);
++ del_gendisk(info->gd);
++ put_disk(info->gd);
++ info->gd = NULL;
++
++ xlbd_put_major_info(info->mi);
++ info->mi = NULL;
++
++ BUG_ON(info->rq == NULL);
++ blk_cleanup_queue(info->rq);
++ info->rq = NULL;
++}
+diff -urNp linux-2.6/drivers/xen/char/Makefile new/drivers/xen/char/Makefile
+--- linux-2.6/drivers/xen/char/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/char/Makefile 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,2 @@
++
++obj-y := mem.o
+diff -urNp linux-2.6/drivers/xen/char/mem.c new/drivers/xen/char/mem.c
+--- linux-2.6/drivers/xen/char/mem.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/char/mem.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,182 @@
++/*
++ * Originally from linux/drivers/char/mem.c
++ *
++ * Copyright (C) 1991, 1992 Linus Torvalds
++ *
++ * Added devfs support.
++ * Jan-11-1998, C. Scott Ananian <cananian@alumni.princeton.edu>
++ * Shared /dev/zero mmaping support, Feb 2000, Kanoj Sarcar <kanoj@sgi.com>
++ */
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <linux/miscdevice.h>
++#include <linux/slab.h>
++#include <linux/vmalloc.h>
++#include <linux/mman.h>
++#include <linux/random.h>
++#include <linux/init.h>
++#include <linux/raw.h>
++#include <linux/tty.h>
++#include <linux/capability.h>
++#include <linux/smp_lock.h>
++#include <linux/devfs_fs_kernel.h>
++#include <linux/ptrace.h>
++#include <linux/device.h>
++#include <asm/pgalloc.h>
++#include <asm/uaccess.h>
++#include <asm/io.h>
++#include <asm/hypervisor.h>
++
++static inline int uncached_access(struct file *file)
++{
++ if (file->f_flags & O_SYNC)
++ return 1;
++ /* Xen sets correct MTRR type on non-RAM for us. */
++ return 0;
++}
++
++/*
++ * This funcion reads the *physical* memory. The f_pos points directly to the
++ * memory location.
++ */
++static ssize_t read_mem(struct file * file, char __user * buf,
++ size_t count, loff_t *ppos)
++{
++ unsigned long p = *ppos, ignored;
++ ssize_t read = 0, sz;
++ void __iomem *v;
++
++ while (count > 0) {
++ /*
++ * Handle first page in case it's not aligned
++ */
++ if (-p & (PAGE_SIZE - 1))
++ sz = -p & (PAGE_SIZE - 1);
++ else
++ sz = PAGE_SIZE;
++
++ sz = min_t(unsigned long, sz, count);
++
++ if ((v = ioremap(p, sz)) == NULL) {
++ /*
++ * Some programs (e.g., dmidecode) groove off into weird RAM
++ * areas where no tables can possibly exist (because Xen will
++ * have stomped on them!). These programs get rather upset if
++ * we let them know that Xen failed their access, so we fake
++ * out a read of all zeroes. :-)
++ */
++ if (clear_user(buf, count))
++ return -EFAULT;
++ read += count;
++ break;
++ }
++
++ ignored = copy_to_user(buf, v, sz);
++ iounmap(v);
++ if (ignored)
++ return -EFAULT;
++ buf += sz;
++ p += sz;
++ count -= sz;
++ read += sz;
++ }
++
++ *ppos += read;
++ return read;
++}
++
++static ssize_t write_mem(struct file * file, const char __user * buf,
++ size_t count, loff_t *ppos)
++{
++ unsigned long p = *ppos, ignored;
++ ssize_t written = 0, sz;
++ void __iomem *v;
++
++ while (count > 0) {
++ /*
++ * Handle first page in case it's not aligned
++ */
++ if (-p & (PAGE_SIZE - 1))
++ sz = -p & (PAGE_SIZE - 1);
++ else
++ sz = PAGE_SIZE;
++
++ sz = min_t(unsigned long, sz, count);
++
++ if ((v = ioremap(p, sz)) == NULL)
++ break;
++
++ ignored = copy_from_user(v, buf, sz);
++ iounmap(v);
++ if (ignored) {
++ written += sz - ignored;
++ if (written)
++ break;
++ return -EFAULT;
++ }
++ buf += sz;
++ p += sz;
++ count -= sz;
++ written += sz;
++ }
++
++ *ppos += written;
++ return written;
++}
++
++static int mmap_mem(struct file * file, struct vm_area_struct * vma)
++{
++ size_t size = vma->vm_end - vma->vm_start;
++
++ if (uncached_access(file))
++ vma->vm_page_prot = pgprot_noncached(vma->vm_page_prot);
++
++ /* We want to return the real error code, not EAGAIN. */
++ return direct_remap_pfn_range(vma, vma->vm_start, vma->vm_pgoff,
++ size, vma->vm_page_prot, DOMID_IO);
++}
++
++/*
++ * The memory devices use the full 32/64 bits of the offset, and so we cannot
++ * check against negative addresses: they are ok. The return value is weird,
++ * though, in that case (0).
++ *
++ * also note that seeking relative to the "end of file" isn't supported:
++ * it has no meaning, so it returns -EINVAL.
++ */
++static loff_t memory_lseek(struct file * file, loff_t offset, int orig)
++{
++ loff_t ret;
++
++ mutex_lock(&file->f_dentry->d_inode->i_mutex);
++ switch (orig) {
++ case 0:
++ file->f_pos = offset;
++ ret = file->f_pos;
++ force_successful_syscall_return();
++ break;
++ case 1:
++ file->f_pos += offset;
++ ret = file->f_pos;
++ force_successful_syscall_return();
++ break;
++ default:
++ ret = -EINVAL;
++ }
++ mutex_unlock(&file->f_dentry->d_inode->i_mutex);
++ return ret;
++}
++
++static int open_mem(struct inode * inode, struct file * filp)
++{
++ return capable(CAP_SYS_RAWIO) ? 0 : -EPERM;
++}
++
++struct file_operations mem_fops = {
++ .llseek = memory_lseek,
++ .read = read_mem,
++ .write = write_mem,
++ .mmap = mmap_mem,
++ .open = open_mem,
++};
+diff -urNp linux-2.6/drivers/xen/console/console.c new/drivers/xen/console/console.c
+--- linux-2.6/drivers/xen/console/console.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/console/console.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,640 @@
++/******************************************************************************
++ * console.c
++ *
++ * Virtual console driver.
++ *
++ * Copyright (c) 2002-2004, K A Fraser.
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/errno.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/interrupt.h>
++#include <linux/tty.h>
++#include <linux/tty_flip.h>
++#include <linux/serial.h>
++#include <linux/major.h>
++#include <linux/ptrace.h>
++#include <linux/ioport.h>
++#include <linux/mm.h>
++#include <linux/slab.h>
++#include <linux/init.h>
++#include <linux/console.h>
++#include <linux/bootmem.h>
++#include <linux/sysrq.h>
++#include <asm/io.h>
++#include <asm/irq.h>
++#include <asm/uaccess.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/event_channel.h>
++#include <asm/hypervisor.h>
++#include <xen/evtchn.h>
++#include <xen/xencons.h>
++
++/*
++ * Modes:
++ * 'xencons=off' [XC_OFF]: Console is disabled.
++ * 'xencons=tty' [XC_TTY]: Console attached to '/dev/tty[0-9]+'.
++ * 'xencons=ttyS' [XC_SERIAL]: Console attached to '/dev/ttyS[0-9]+'.
++ * [XC_DEFAULT]: DOM0 -> XC_SERIAL ; all others -> XC_TTY.
++ *
++ * NB. In mode XC_TTY, we create dummy consoles for tty2-63. This suppresses
++ * warnings from standard distro startup scripts.
++ */
++static enum { XC_OFF, XC_DEFAULT, XC_TTY, XC_SERIAL } xc_mode = XC_DEFAULT;
++static int xc_num = -1;
++
++#ifdef CONFIG_MAGIC_SYSRQ
++static unsigned long sysrq_requested;
++extern int sysrq_enabled;
++#endif
++
++static int __init xencons_setup(char *str)
++{
++ char *q;
++ int n;
++
++ if (!strncmp(str, "ttyS", 4))
++ xc_mode = XC_SERIAL;
++ else if (!strncmp(str, "tty", 3))
++ xc_mode = XC_TTY;
++ else if (!strncmp(str, "off", 3))
++ xc_mode = XC_OFF;
++
++ switch (xc_mode) {
++ case XC_SERIAL:
++ n = simple_strtol(str+4, &q, 10);
++ if (q > (str + 4))
++ xc_num = n;
++ break;
++ case XC_TTY:
++ n = simple_strtol(str+3, &q, 10);
++ if (q > (str + 3))
++ xc_num = n;
++ break;
++ default:
++ break;
++ }
++
++ return 1;
++}
++__setup("xencons=", xencons_setup);
++
++/* The kernel and user-land drivers share a common transmit buffer. */
++static unsigned int wbuf_size = 4096;
++#define WBUF_MASK(_i) ((_i)&(wbuf_size-1))
++static char *wbuf;
++static unsigned int wc, wp; /* write_cons, write_prod */
++
++static int __init xencons_bufsz_setup(char *str)
++{
++ unsigned int goal;
++ goal = simple_strtoul(str, NULL, 0);
++ if (goal) {
++ goal = roundup_pow_of_two(goal);
++ if (wbuf_size < goal)
++ wbuf_size = goal;
++ }
++ return 1;
++}
++__setup("xencons_bufsz=", xencons_bufsz_setup);
++
++/* This lock protects accesses to the common transmit buffer. */
++static DEFINE_SPINLOCK(xencons_lock);
++
++/* Common transmit-kick routine. */
++static void __xencons_tx_flush(void);
++
++static struct tty_driver *xencons_driver;
++
++/******************** Kernel console driver ********************************/
++
++static void kcons_write(struct console *c, const char *s, unsigned int count)
++{
++ int i = 0;
++ unsigned long flags;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++
++ while (i < count) {
++ for (; i < count; i++) {
++ if ((wp - wc) >= (wbuf_size - 1))
++ break;
++ if ((wbuf[WBUF_MASK(wp++)] = s[i]) == '\n')
++ wbuf[WBUF_MASK(wp++)] = '\r';
++ }
++
++ __xencons_tx_flush();
++ }
++
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static void kcons_write_dom0(struct console *c, const char *s, unsigned int count)
++{
++
++ while (count > 0) {
++ int rc;
++ rc = HYPERVISOR_console_io( CONSOLEIO_write, count, (char *)s);
++ if (rc <= 0)
++ break;
++ count -= rc;
++ s += rc;
++ }
++}
++
++static struct tty_driver *kcons_device(struct console *c, int *index)
++{
++ *index = 0;
++ return xencons_driver;
++}
++
++static struct console kcons_info = {
++ .device = kcons_device,
++ .flags = CON_PRINTBUFFER,
++ .index = -1,
++};
++
++#define __RETCODE 0
++static int __init xen_console_init(void)
++{
++ if (!is_running_on_xen())
++ return __RETCODE;
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ if (xc_mode == XC_DEFAULT)
++ xc_mode = XC_SERIAL;
++ kcons_info.write = kcons_write_dom0;
++ if (xc_mode == XC_SERIAL)
++ kcons_info.flags |= CON_ENABLED;
++ } else {
++ if (xc_mode == XC_DEFAULT)
++ xc_mode = XC_TTY;
++ kcons_info.write = kcons_write;
++ }
++
++ switch (xc_mode) {
++ case XC_SERIAL:
++ strcpy(kcons_info.name, "ttyS");
++ if (xc_num == -1)
++ xc_num = 0;
++ break;
++
++ case XC_TTY:
++ strcpy(kcons_info.name, "tty");
++ if (xc_num == -1)
++ xc_num = 1;
++ break;
++
++ default:
++ return __RETCODE;
++ }
++
++ wbuf = alloc_bootmem(wbuf_size);
++
++ register_console(&kcons_info);
++
++ return __RETCODE;
++}
++console_initcall(xen_console_init);
++
++/*** Useful function for console debugging -- goes straight to Xen. ***/
++asmlinkage int xprintk(const char *fmt, ...)
++{
++ va_list args;
++ int printk_len;
++ static char printk_buf[1024];
++
++ /* Emit the output into the temporary buffer */
++ va_start(args, fmt);
++ printk_len = vsnprintf(printk_buf, sizeof(printk_buf), fmt, args);
++ va_end(args);
++
++ /* Send the processed output directly to Xen. */
++ kcons_write_dom0(NULL, printk_buf, printk_len);
++
++ return 0;
++}
++
++/*** Forcibly flush console data before dying. ***/
++void xencons_force_flush(void)
++{
++ int sz;
++
++ /* Emergency console is synchronous, so there's nothing to flush. */
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ return;
++
++ /* Spin until console data is flushed through to the daemon. */
++ while (wc != wp) {
++ int sent = 0;
++ if ((sz = wp - wc) == 0)
++ continue;
++ sent = xencons_ring_send(&wbuf[WBUF_MASK(wc)], sz);
++ if (sent > 0)
++ wc += sent;
++ }
++}
++
++
++/******************** User-space console driver (/dev/console) ************/
++
++#define DRV(_d) (_d)
++#define TTY_INDEX(_tty) ((_tty)->index)
++
++static struct termios *xencons_termios[MAX_NR_CONSOLES];
++static struct termios *xencons_termios_locked[MAX_NR_CONSOLES];
++static struct tty_struct *xencons_tty;
++static int xencons_priv_irq;
++static char x_char;
++
++void xencons_rx(char *buf, unsigned len, struct pt_regs *regs)
++{
++ int i;
++ unsigned long flags;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ if (xencons_tty == NULL)
++ goto out;
++
++ for (i = 0; i < len; i++) {
++#ifdef CONFIG_MAGIC_SYSRQ
++ if (sysrq_enabled) {
++ if (buf[i] == '\x0f') { /* ^O */
++ sysrq_requested = jiffies;
++ continue; /* don't print the sysrq key */
++ } else if (sysrq_requested) {
++ unsigned long sysrq_timeout =
++ sysrq_requested + HZ*2;
++ sysrq_requested = 0;
++ if (time_before(jiffies, sysrq_timeout)) {
++ spin_unlock_irqrestore(
++ &xencons_lock, flags);
++ handle_sysrq(
++ buf[i], regs, xencons_tty);
++ spin_lock_irqsave(
++ &xencons_lock, flags);
++ continue;
++ }
++ }
++ }
++#endif
++ tty_insert_flip_char(xencons_tty, buf[i], 0);
++ }
++ tty_flip_buffer_push(xencons_tty);
++
++ out:
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static void __xencons_tx_flush(void)
++{
++ int sent, sz, work_done = 0;
++
++ if (x_char) {
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ kcons_write_dom0(NULL, &x_char, 1);
++ else
++ while (x_char)
++ if (xencons_ring_send(&x_char, 1) == 1)
++ break;
++ x_char = 0;
++ work_done = 1;
++ }
++
++ while (wc != wp) {
++ sz = wp - wc;
++ if (sz > (wbuf_size - WBUF_MASK(wc)))
++ sz = wbuf_size - WBUF_MASK(wc);
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ kcons_write_dom0(NULL, &wbuf[WBUF_MASK(wc)], sz);
++ wc += sz;
++ } else {
++ sent = xencons_ring_send(&wbuf[WBUF_MASK(wc)], sz);
++ if (sent == 0)
++ break;
++ wc += sent;
++ }
++ work_done = 1;
++ }
++
++ if (work_done && (xencons_tty != NULL)) {
++ wake_up_interruptible(&xencons_tty->write_wait);
++ if ((xencons_tty->flags & (1 << TTY_DO_WRITE_WAKEUP)) &&
++ (xencons_tty->ldisc.write_wakeup != NULL))
++ (xencons_tty->ldisc.write_wakeup)(xencons_tty);
++ }
++}
++
++void xencons_tx(void)
++{
++ unsigned long flags;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ __xencons_tx_flush();
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++/* Privileged receive callback and transmit kicker. */
++static irqreturn_t xencons_priv_interrupt(int irq, void *dev_id,
++ struct pt_regs *regs)
++{
++ static char rbuf[16];
++ int l;
++
++ while ((l = HYPERVISOR_console_io(CONSOLEIO_read, 16, rbuf)) > 0)
++ xencons_rx(rbuf, l, regs);
++
++ xencons_tx();
++
++ return IRQ_HANDLED;
++}
++
++static int xencons_write_room(struct tty_struct *tty)
++{
++ return wbuf_size - (wp - wc);
++}
++
++static int xencons_chars_in_buffer(struct tty_struct *tty)
++{
++ return wp - wc;
++}
++
++static void xencons_send_xchar(struct tty_struct *tty, char ch)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ x_char = ch;
++ __xencons_tx_flush();
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static void xencons_throttle(struct tty_struct *tty)
++{
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ if (I_IXOFF(tty))
++ xencons_send_xchar(tty, STOP_CHAR(tty));
++}
++
++static void xencons_unthrottle(struct tty_struct *tty)
++{
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ if (I_IXOFF(tty)) {
++ if (x_char != 0)
++ x_char = 0;
++ else
++ xencons_send_xchar(tty, START_CHAR(tty));
++ }
++}
++
++static void xencons_flush_buffer(struct tty_struct *tty)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ wc = wp = 0;
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static inline int __xencons_put_char(int ch)
++{
++ char _ch = (char)ch;
++ if ((wp - wc) == wbuf_size)
++ return 0;
++ wbuf[WBUF_MASK(wp++)] = _ch;
++ return 1;
++}
++
++static int xencons_write(
++ struct tty_struct *tty,
++ const unsigned char *buf,
++ int count)
++{
++ int i;
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return count;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++
++ for (i = 0; i < count; i++)
++ if (!__xencons_put_char(buf[i]))
++ break;
++
++ if (i != 0)
++ __xencons_tx_flush();
++
++ spin_unlock_irqrestore(&xencons_lock, flags);
++
++ return i;
++}
++
++static void xencons_put_char(struct tty_struct *tty, u_char ch)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ (void)__xencons_put_char(ch);
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static void xencons_flush_chars(struct tty_struct *tty)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ __xencons_tx_flush();
++ spin_unlock_irqrestore(&xencons_lock, flags);
++}
++
++static void xencons_wait_until_sent(struct tty_struct *tty, int timeout)
++{
++ unsigned long orig_jiffies = jiffies;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ while (DRV(tty->driver)->chars_in_buffer(tty)) {
++ set_current_state(TASK_INTERRUPTIBLE);
++ schedule_timeout(1);
++ if (signal_pending(current))
++ break;
++ if (timeout && time_after(jiffies, orig_jiffies + timeout))
++ break;
++ }
++
++ set_current_state(TASK_RUNNING);
++}
++
++static int xencons_open(struct tty_struct *tty, struct file *filp)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return 0;
++
++ spin_lock_irqsave(&xencons_lock, flags);
++ tty->driver_data = NULL;
++ if (xencons_tty == NULL)
++ xencons_tty = tty;
++ __xencons_tx_flush();
++ spin_unlock_irqrestore(&xencons_lock, flags);
++
++ return 0;
++}
++
++static void xencons_close(struct tty_struct *tty, struct file *filp)
++{
++ unsigned long flags;
++
++ if (TTY_INDEX(tty) != 0)
++ return;
++
++ if (tty->count == 1) {
++ tty->closing = 1;
++ tty_wait_until_sent(tty, 0);
++ if (DRV(tty->driver)->flush_buffer != NULL)
++ DRV(tty->driver)->flush_buffer(tty);
++ if (tty->ldisc.flush_buffer != NULL)
++ tty->ldisc.flush_buffer(tty);
++ tty->closing = 0;
++ spin_lock_irqsave(&xencons_lock, flags);
++ xencons_tty = NULL;
++ spin_unlock_irqrestore(&xencons_lock, flags);
++ }
++}
++
++static struct tty_operations xencons_ops = {
++ .open = xencons_open,
++ .close = xencons_close,
++ .write = xencons_write,
++ .write_room = xencons_write_room,
++ .put_char = xencons_put_char,
++ .flush_chars = xencons_flush_chars,
++ .chars_in_buffer = xencons_chars_in_buffer,
++ .send_xchar = xencons_send_xchar,
++ .flush_buffer = xencons_flush_buffer,
++ .throttle = xencons_throttle,
++ .unthrottle = xencons_unthrottle,
++ .wait_until_sent = xencons_wait_until_sent,
++};
++
++static int __init xencons_init(void)
++{
++ int rc;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ if (xc_mode == XC_OFF)
++ return 0;
++
++ xencons_ring_init();
++
++ xencons_driver = alloc_tty_driver((xc_mode == XC_SERIAL) ?
++ 1 : MAX_NR_CONSOLES);
++ if (xencons_driver == NULL)
++ return -ENOMEM;
++
++ DRV(xencons_driver)->name = "xencons";
++ DRV(xencons_driver)->major = TTY_MAJOR;
++ DRV(xencons_driver)->type = TTY_DRIVER_TYPE_SERIAL;
++ DRV(xencons_driver)->subtype = SERIAL_TYPE_NORMAL;
++ DRV(xencons_driver)->init_termios = tty_std_termios;
++ DRV(xencons_driver)->flags =
++ TTY_DRIVER_REAL_RAW |
++ TTY_DRIVER_RESET_TERMIOS |
++ TTY_DRIVER_NO_DEVFS;
++ DRV(xencons_driver)->termios = xencons_termios;
++ DRV(xencons_driver)->termios_locked = xencons_termios_locked;
++
++ if (xc_mode == XC_SERIAL) {
++ DRV(xencons_driver)->name = "ttyS";
++ DRV(xencons_driver)->minor_start = 64 + xc_num;
++ DRV(xencons_driver)->name_base = 0 + xc_num;
++ } else {
++ DRV(xencons_driver)->name = "tty";
++ DRV(xencons_driver)->minor_start = xc_num;
++ DRV(xencons_driver)->name_base = xc_num;
++ }
++
++ tty_set_operations(xencons_driver, &xencons_ops);
++
++ if ((rc = tty_register_driver(DRV(xencons_driver))) != 0) {
++ printk("WARNING: Failed to register Xen virtual "
++ "console driver as '%s%d'\n",
++ DRV(xencons_driver)->name,
++ DRV(xencons_driver)->name_base);
++ put_tty_driver(xencons_driver);
++ xencons_driver = NULL;
++ return rc;
++ }
++
++ tty_register_device(xencons_driver, 0, NULL);
++
++ if (xen_start_info->flags & SIF_INITDOMAIN) {
++ xencons_priv_irq = bind_virq_to_irqhandler(
++ VIRQ_CONSOLE,
++ 0,
++ xencons_priv_interrupt,
++ 0,
++ "console",
++ NULL);
++ BUG_ON(xencons_priv_irq < 0);
++ }
++
++ printk("Xen virtual console successfully installed as %s%d\n",
++ DRV(xencons_driver)->name,
++ DRV(xencons_driver)->name_base );
++
++ return 0;
++}
++
++module_init(xencons_init);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/console/Makefile new/drivers/xen/console/Makefile
+--- linux-2.6/drivers/xen/console/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/console/Makefile 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,2 @@
++
++obj-y := console.o xencons_ring.o
+diff -urNp linux-2.6/drivers/xen/console/xencons_ring.c new/drivers/xen/console/xencons_ring.c
+--- linux-2.6/drivers/xen/console/xencons_ring.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/console/xencons_ring.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,141 @@
++/*
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/errno.h>
++#include <linux/signal.h>
++#include <linux/sched.h>
++#include <linux/interrupt.h>
++#include <linux/tty.h>
++#include <linux/tty_flip.h>
++#include <linux/serial.h>
++#include <linux/major.h>
++#include <linux/ptrace.h>
++#include <linux/ioport.h>
++#include <linux/mm.h>
++#include <linux/slab.h>
++
++#include <asm/hypervisor.h>
++#include <xen/evtchn.h>
++#include <xen/xencons.h>
++#include <linux/wait.h>
++#include <linux/interrupt.h>
++#include <linux/sched.h>
++#include <linux/err.h>
++#include <xen/interface/io/console.h>
++
++static int xencons_irq;
++
++static inline struct xencons_interface *xencons_interface(void)
++{
++ return mfn_to_virt(xen_start_info->console_mfn);
++}
++
++static inline void notify_daemon(void)
++{
++ /* Use evtchn: this is called early, before irq is set up. */
++ notify_remote_via_evtchn(xen_start_info->console_evtchn);
++}
++
++int xencons_ring_send(const char *data, unsigned len)
++{
++ int sent = 0;
++ struct xencons_interface *intf = xencons_interface();
++ XENCONS_RING_IDX cons, prod;
++
++ cons = intf->out_cons;
++ prod = intf->out_prod;
++ mb();
++ BUG_ON((prod - cons) > sizeof(intf->out));
++
++ while ((sent < len) && ((prod - cons) < sizeof(intf->out)))
++ intf->out[MASK_XENCONS_IDX(prod++, intf->out)] = data[sent++];
++
++ wmb();
++ intf->out_prod = prod;
++
++ notify_daemon();
++
++ return sent;
++}
++
++static irqreturn_t handle_input(int irq, void *unused, struct pt_regs *regs)
++{
++ struct xencons_interface *intf = xencons_interface();
++ XENCONS_RING_IDX cons, prod;
++
++ cons = intf->in_cons;
++ prod = intf->in_prod;
++ mb();
++ BUG_ON((prod - cons) > sizeof(intf->in));
++
++ while (cons != prod) {
++ xencons_rx(intf->in+MASK_XENCONS_IDX(cons,intf->in), 1, regs);
++ cons++;
++ }
++
++ mb();
++ intf->in_cons = cons;
++
++ notify_daemon();
++
++ xencons_tx();
++
++ return IRQ_HANDLED;
++}
++
++int xencons_ring_init(void)
++{
++ int err;
++
++ if (xencons_irq)
++ unbind_from_irqhandler(xencons_irq, NULL);
++ xencons_irq = 0;
++
++ if (!xen_start_info->console_evtchn)
++ return 0;
++
++ err = bind_evtchn_to_irqhandler(
++ xen_start_info->console_evtchn,
++ handle_input, 0, "xencons", NULL);
++ if (err <= 0) {
++ printk(KERN_ERR "XEN console request irq failed %i\n", err);
++ return err;
++ }
++
++ xencons_irq = err;
++
++ /* In case we have in-flight data after save/restore... */
++ notify_daemon();
++
++ return 0;
++}
++
++void xencons_resume(void)
++{
++ (void)xencons_ring_init();
++}
+diff -urNp linux-2.6/drivers/xen/core/cpu_hotplug.c new/drivers/xen/core/cpu_hotplug.c
+--- linux-2.6/drivers/xen/core/cpu_hotplug.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/cpu_hotplug.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,188 @@
++#include <linux/config.h>
++#include <linux/init.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/notifier.h>
++#include <linux/cpu.h>
++#include <xen/cpu_hotplug.h>
++#include <xen/xenbus.h>
++
++/*
++ * Set of CPUs that remote admin software will allow us to bring online.
++ * Notified to us via xenbus.
++ */
++static cpumask_t xenbus_allowed_cpumask;
++
++/* Set of CPUs that local admin will allow us to bring online. */
++static cpumask_t local_allowed_cpumask = CPU_MASK_ALL;
++
++static int local_cpu_hotplug_request(void)
++{
++ /*
++ * We assume a CPU hotplug request comes from local admin if it is made
++ * via a userspace process (i.e., one with a real mm_struct).
++ */
++ return (current->mm != NULL);
++}
++
++static void vcpu_hotplug(unsigned int cpu)
++{
++ int err;
++ char dir[32], state[32];
++
++ if ((cpu >= NR_CPUS) || !cpu_possible(cpu))
++ return;
++
++ sprintf(dir, "cpu/%d", cpu);
++ err = xenbus_scanf(XBT_NIL, dir, "availability", "%s", state);
++ if (err != 1) {
++ printk(KERN_ERR "XENBUS: Unable to read cpu state\n");
++ return;
++ }
++
++ if (strcmp(state, "online") == 0) {
++ cpu_set(cpu, xenbus_allowed_cpumask);
++ (void)cpu_up(cpu);
++ } else if (strcmp(state, "offline") == 0) {
++ cpu_clear(cpu, xenbus_allowed_cpumask);
++ (void)cpu_down(cpu);
++ } else {
++ printk(KERN_ERR "XENBUS: unknown state(%s) on CPU%d\n",
++ state, cpu);
++ }
++}
++
++static void handle_vcpu_hotplug_event(
++ struct xenbus_watch *watch, const char **vec, unsigned int len)
++{
++ int cpu;
++ char *cpustr;
++ const char *node = vec[XS_WATCH_PATH];
++
++ if ((cpustr = strstr(node, "cpu/")) != NULL) {
++ sscanf(cpustr, "cpu/%d", &cpu);
++ vcpu_hotplug(cpu);
++ }
++}
++
++static int smpboot_cpu_notify(struct notifier_block *notifier,
++ unsigned long action, void *hcpu)
++{
++ int cpu = (long)hcpu;
++
++ /*
++ * We do this in a callback notifier rather than __cpu_disable()
++ * because local_cpu_hotplug_request() does not work in the latter
++ * as it's always executed from within a stopmachine kthread.
++ */
++ if ((action == CPU_DOWN_PREPARE) && local_cpu_hotplug_request())
++ cpu_clear(cpu, local_allowed_cpumask);
++
++ return NOTIFY_OK;
++}
++
++static int setup_cpu_watcher(struct notifier_block *notifier,
++ unsigned long event, void *data)
++{
++ int i;
++
++ static struct xenbus_watch cpu_watch = {
++ .node = "cpu",
++ .callback = handle_vcpu_hotplug_event,
++ .flags = XBWF_new_thread };
++ (void)register_xenbus_watch(&cpu_watch);
++
++ if (!(xen_start_info->flags & SIF_INITDOMAIN)) {
++ for_each_cpu(i)
++ vcpu_hotplug(i);
++ printk(KERN_INFO "Brought up %ld CPUs\n",
++ (long)num_online_cpus());
++ }
++
++ return NOTIFY_DONE;
++}
++
++static int __init setup_vcpu_hotplug_event(void)
++{
++ static struct notifier_block hotplug_cpu = {
++ .notifier_call = smpboot_cpu_notify };
++ static struct notifier_block xsn_cpu = {
++ .notifier_call = setup_cpu_watcher };
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ register_cpu_notifier(&hotplug_cpu);
++ register_xenstore_notifier(&xsn_cpu);
++
++ return 0;
++}
++
++arch_initcall(setup_vcpu_hotplug_event);
++
++int smp_suspend(void)
++{
++ int i, err;
++
++ lock_cpu_hotplug();
++
++ /*
++ * Take all other CPUs offline. We hold the hotplug mutex to
++ * avoid other processes bringing up CPUs under our feet.
++ */
++ while (num_online_cpus() > 1) {
++ unlock_cpu_hotplug();
++ for_each_online_cpu(i) {
++ if (i == 0)
++ continue;
++ err = cpu_down(i);
++ if (err) {
++ printk(KERN_CRIT "Failed to take all CPUs "
++ "down: %d.\n", err);
++ for_each_cpu(i)
++ vcpu_hotplug(i);
++ return err;
++ }
++ }
++ lock_cpu_hotplug();
++ }
++
++ return 0;
++}
++
++void smp_resume(void)
++{
++ int cpu;
++
++ for_each_cpu(cpu)
++ cpu_initialize_context(cpu);
++
++ unlock_cpu_hotplug();
++
++ for_each_cpu(cpu)
++ vcpu_hotplug(cpu);
++}
++
++int cpu_up_check(unsigned int cpu)
++{
++ int rc = 0;
++
++ if (local_cpu_hotplug_request()) {
++ cpu_set(cpu, local_allowed_cpumask);
++ if (!cpu_isset(cpu, xenbus_allowed_cpumask)) {
++ printk("%s: attempt to bring up CPU %u disallowed by "
++ "remote admin.\n", __FUNCTION__, cpu);
++ rc = -EBUSY;
++ }
++ } else if (!cpu_isset(cpu, local_allowed_cpumask) ||
++ !cpu_isset(cpu, xenbus_allowed_cpumask)) {
++ rc = -EBUSY;
++ }
++
++ return rc;
++}
++
++void init_xenbus_allowed_cpumask(void)
++{
++ xenbus_allowed_cpumask = cpu_present_map;
++}
+diff -urNp linux-2.6/drivers/xen/core/evtchn.c new/drivers/xen/core/evtchn.c
+--- linux-2.6/drivers/xen/core/evtchn.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/evtchn.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,853 @@
++/******************************************************************************
++ * evtchn.c
++ *
++ * Communication via Xen event channels.
++ *
++ * Copyright (c) 2002-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/irq.h>
++#include <linux/interrupt.h>
++#include <linux/sched.h>
++#include <linux/kernel_stat.h>
++#include <linux/version.h>
++#include <asm/atomic.h>
++#include <asm/system.h>
++#include <asm/ptrace.h>
++#include <asm/synch_bitops.h>
++#include <xen/evtchn.h>
++#include <xen/interface/event_channel.h>
++#include <xen/interface/physdev.h>
++#include <asm/hypervisor.h>
++#include <linux/mc146818rtc.h> /* RTC_IRQ */
++
++/*
++ * This lock protects updates to the following mapping and reference-count
++ * arrays. The lock does not need to be acquired to read the mapping tables.
++ */
++static DEFINE_SPINLOCK(irq_mapping_update_lock);
++
++/* IRQ <-> event-channel mappings. */
++static int evtchn_to_irq[NR_EVENT_CHANNELS] = {
++ [0 ... NR_EVENT_CHANNELS-1] = -1 };
++
++/* Packed IRQ information: binding type, sub-type index, and event channel. */
++static u32 irq_info[NR_IRQS];
++
++/* Binding types. */
++enum { IRQT_UNBOUND, IRQT_PIRQ, IRQT_VIRQ, IRQT_IPI, IRQT_EVTCHN };
++
++/* Constructor for packed IRQ information. */
++static inline u32 mk_irq_info(u32 type, u32 index, u32 evtchn)
++{
++ return ((type << 24) | (index << 16) | evtchn);
++}
++
++/* Convenient shorthand for packed representation of an unbound IRQ. */
++#define IRQ_UNBOUND mk_irq_info(IRQT_UNBOUND, 0, 0)
++
++/*
++ * Accessors for packed IRQ information.
++ */
++
++static inline unsigned int evtchn_from_irq(int irq)
++{
++ return (u16)(irq_info[irq]);
++}
++
++static inline unsigned int index_from_irq(int irq)
++{
++ return (u8)(irq_info[irq] >> 16);
++}
++
++static inline unsigned int type_from_irq(int irq)
++{
++ return (u8)(irq_info[irq] >> 24);
++}
++
++/* IRQ <-> VIRQ mapping. */
++DEFINE_PER_CPU(int, virq_to_irq[NR_VIRQS]) = {[0 ... NR_VIRQS-1] = -1};
++
++/* IRQ <-> IPI mapping. */
++#ifndef NR_IPIS
++#define NR_IPIS 1
++#endif
++DEFINE_PER_CPU(int, ipi_to_irq[NR_IPIS]) = {[0 ... NR_IPIS-1] = -1};
++
++/* Reference counts for bindings to IRQs. */
++static int irq_bindcount[NR_IRQS];
++
++/* Bitmap indicating which PIRQs require Xen to be notified on unmask. */
++static unsigned long pirq_needs_eoi[NR_PIRQS/sizeof(unsigned long)];
++
++#ifdef CONFIG_SMP
++
++static u8 cpu_evtchn[NR_EVENT_CHANNELS];
++static unsigned long cpu_evtchn_mask[NR_CPUS][NR_EVENT_CHANNELS/BITS_PER_LONG];
++
++static inline unsigned long active_evtchns(unsigned int cpu, shared_info_t *sh,
++ unsigned int idx)
++{
++ return (sh->evtchn_pending[idx] &
++ cpu_evtchn_mask[cpu][idx] &
++ ~sh->evtchn_mask[idx]);
++}
++
++static void bind_evtchn_to_cpu(unsigned int chn, unsigned int cpu)
++{
++ int irq = evtchn_to_irq[chn];
++
++ BUG_ON(irq == -1);
++ set_native_irq_info(irq, cpumask_of_cpu(cpu));
++
++ clear_bit(chn, (unsigned long *)cpu_evtchn_mask[cpu_evtchn[chn]]);
++ set_bit(chn, (unsigned long *)cpu_evtchn_mask[cpu]);
++ cpu_evtchn[chn] = cpu;
++}
++
++static void init_evtchn_cpu_bindings(void)
++{
++ int i;
++
++ /* By default all event channels notify CPU#0. */
++ for (i = 0; i < NR_IRQS; i++)
++ set_native_irq_info(i, cpumask_of_cpu(0));
++
++ memset(cpu_evtchn, 0, sizeof(cpu_evtchn));
++ memset(cpu_evtchn_mask[0], ~0, sizeof(cpu_evtchn_mask[0]));
++}
++
++static inline unsigned int cpu_from_evtchn(unsigned int evtchn)
++{
++ return cpu_evtchn[evtchn];
++}
++
++#else
++
++static inline unsigned long active_evtchns(unsigned int cpu, shared_info_t *sh,
++ unsigned int idx)
++{
++ return (sh->evtchn_pending[idx] & ~sh->evtchn_mask[idx]);
++}
++
++static void bind_evtchn_to_cpu(unsigned int chn, unsigned int cpu)
++{
++}
++
++static void init_evtchn_cpu_bindings(void)
++{
++}
++
++static inline unsigned int cpu_from_evtchn(unsigned int evtchn)
++{
++ return 0;
++}
++
++#endif
++
++/* Upcall to generic IRQ layer. */
++#ifdef CONFIG_X86
++extern fastcall unsigned int do_IRQ(struct pt_regs *regs);
++void __init xen_init_IRQ(void);
++void __init init_IRQ(void)
++{
++ irq_ctx_init(0);
++ xen_init_IRQ();
++}
++#if defined (__i386__)
++static inline void exit_idle(void) {}
++#define IRQ_REG orig_eax
++#elif defined (__x86_64__)
++#include <asm/idle.h>
++#define IRQ_REG orig_rax
++#endif
++#define do_IRQ(irq, regs) do { \
++ (regs)->IRQ_REG = ~(irq); \
++ do_IRQ((regs)); \
++} while (0)
++#endif
++
++/* Xen will never allocate port zero for any purpose. */
++#define VALID_EVTCHN(chn) ((chn) != 0)
++
++/*
++ * Force a proper event-channel callback from Xen after clearing the
++ * callback mask. We do this in a very simple manner, by making a call
++ * down into Xen. The pending flag will be checked by Xen on return.
++ */
++void force_evtchn_callback(void)
++{
++ (void)HYPERVISOR_xen_version(0, NULL);
++}
++/* Not a GPL symbol: used in ubiquitous macros, so too restrictive. */
++EXPORT_SYMBOL(force_evtchn_callback);
++
++/* NB. Interrupts are disabled on entry. */
++asmlinkage void evtchn_do_upcall(struct pt_regs *regs)
++{
++ unsigned long l1, l2;
++ unsigned int l1i, l2i, port;
++ int irq, cpu = smp_processor_id();
++ shared_info_t *s = HYPERVISOR_shared_info;
++ vcpu_info_t *vcpu_info = &s->vcpu_info[cpu];
++
++ vcpu_info->evtchn_upcall_pending = 0;
++
++#ifndef CONFIG_X86 /* No need for a barrier -- XCHG is a barrier on x86. */
++ /* Clear master pending flag /before/ clearing selector flag. */
++ rmb();
++#endif
++ l1 = xchg(&vcpu_info->evtchn_pending_sel, 0);
++ while (l1 != 0) {
++ l1i = __ffs(l1);
++ l1 &= ~(1UL << l1i);
++
++ while ((l2 = active_evtchns(cpu, s, l1i)) != 0) {
++ l2i = __ffs(l2);
++
++ port = (l1i * BITS_PER_LONG) + l2i;
++ if ((irq = evtchn_to_irq[port]) != -1)
++ do_IRQ(irq, regs);
++ else {
++ exit_idle();
++ evtchn_device_upcall(port);
++ }
++ }
++ }
++}
++
++static int find_unbound_irq(void)
++{
++ int irq;
++
++ /* Only allocate from dynirq range */
++ for (irq = DYNIRQ_BASE; irq < NR_IRQS; irq++)
++ if (irq_bindcount[irq] == 0)
++ break;
++
++ if (irq == NR_IRQS)
++ panic("No available IRQ to bind to: increase NR_IRQS!\n");
++
++ return irq;
++}
++
++static int bind_evtchn_to_irq(unsigned int evtchn)
++{
++ int irq;
++
++ spin_lock(&irq_mapping_update_lock);
++
++ if ((irq = evtchn_to_irq[evtchn]) == -1) {
++ irq = find_unbound_irq();
++ evtchn_to_irq[evtchn] = irq;
++ irq_info[irq] = mk_irq_info(IRQT_EVTCHN, 0, evtchn);
++ }
++
++ irq_bindcount[irq]++;
++
++ spin_unlock(&irq_mapping_update_lock);
++
++ return irq;
++}
++
++static int bind_virq_to_irq(unsigned int virq, unsigned int cpu)
++{
++ struct evtchn_bind_virq bind_virq;
++ int evtchn, irq;
++
++ spin_lock(&irq_mapping_update_lock);
++
++ if ((irq = per_cpu(virq_to_irq, cpu)[virq]) == -1) {
++ bind_virq.virq = virq;
++ bind_virq.vcpu = cpu;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
++ &bind_virq) != 0)
++ BUG();
++ evtchn = bind_virq.port;
++
++ irq = find_unbound_irq();
++ evtchn_to_irq[evtchn] = irq;
++ irq_info[irq] = mk_irq_info(IRQT_VIRQ, virq, evtchn);
++
++ per_cpu(virq_to_irq, cpu)[virq] = irq;
++
++ bind_evtchn_to_cpu(evtchn, cpu);
++ }
++
++ irq_bindcount[irq]++;
++
++ spin_unlock(&irq_mapping_update_lock);
++
++ return irq;
++}
++
++static int bind_ipi_to_irq(unsigned int ipi, unsigned int cpu)
++{
++ struct evtchn_bind_ipi bind_ipi;
++ int evtchn, irq;
++
++ spin_lock(&irq_mapping_update_lock);
++
++ if ((irq = per_cpu(ipi_to_irq, cpu)[ipi]) == -1) {
++ bind_ipi.vcpu = cpu;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
++ &bind_ipi) != 0)
++ BUG();
++ evtchn = bind_ipi.port;
++
++ irq = find_unbound_irq();
++ evtchn_to_irq[evtchn] = irq;
++ irq_info[irq] = mk_irq_info(IRQT_IPI, ipi, evtchn);
++
++ per_cpu(ipi_to_irq, cpu)[ipi] = irq;
++
++ bind_evtchn_to_cpu(evtchn, cpu);
++ }
++
++ irq_bindcount[irq]++;
++
++ spin_unlock(&irq_mapping_update_lock);
++
++ return irq;
++}
++
++static void unbind_from_irq(unsigned int irq)
++{
++ struct evtchn_close close;
++ int evtchn = evtchn_from_irq(irq);
++
++ spin_lock(&irq_mapping_update_lock);
++
++ if ((--irq_bindcount[irq] == 0) && VALID_EVTCHN(evtchn)) {
++ close.port = evtchn;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_close, &close) != 0)
++ BUG();
++
++ switch (type_from_irq(irq)) {
++ case IRQT_VIRQ:
++ per_cpu(virq_to_irq, cpu_from_evtchn(evtchn))
++ [index_from_irq(irq)] = -1;
++ break;
++ case IRQT_IPI:
++ per_cpu(ipi_to_irq, cpu_from_evtchn(evtchn))
++ [index_from_irq(irq)] = -1;
++ break;
++ default:
++ break;
++ }
++
++ /* Closed ports are implicitly re-bound to VCPU0. */
++ bind_evtchn_to_cpu(evtchn, 0);
++
++ evtchn_to_irq[evtchn] = -1;
++ irq_info[irq] = IRQ_UNBOUND;
++ }
++
++ spin_unlock(&irq_mapping_update_lock);
++}
++
++int bind_evtchn_to_irqhandler(
++ unsigned int evtchn,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id)
++{
++ unsigned int irq;
++ int retval;
++
++ irq = bind_evtchn_to_irq(evtchn);
++ retval = request_irq(irq, handler, irqflags, devname, dev_id);
++ if (retval != 0) {
++ unbind_from_irq(irq);
++ return retval;
++ }
++
++ return irq;
++}
++EXPORT_SYMBOL_GPL(bind_evtchn_to_irqhandler);
++
++int bind_virq_to_irqhandler(
++ unsigned int virq,
++ unsigned int cpu,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id)
++{
++ unsigned int irq;
++ int retval;
++
++ irq = bind_virq_to_irq(virq, cpu);
++ retval = request_irq(irq, handler, irqflags, devname, dev_id);
++ if (retval != 0) {
++ unbind_from_irq(irq);
++ return retval;
++ }
++
++ return irq;
++}
++EXPORT_SYMBOL_GPL(bind_virq_to_irqhandler);
++
++int bind_ipi_to_irqhandler(
++ unsigned int ipi,
++ unsigned int cpu,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id)
++{
++ unsigned int irq;
++ int retval;
++
++ irq = bind_ipi_to_irq(ipi, cpu);
++ retval = request_irq(irq, handler, irqflags, devname, dev_id);
++ if (retval != 0) {
++ unbind_from_irq(irq);
++ return retval;
++ }
++
++ return irq;
++}
++EXPORT_SYMBOL_GPL(bind_ipi_to_irqhandler);
++
++void unbind_from_irqhandler(unsigned int irq, void *dev_id)
++{
++ free_irq(irq, dev_id);
++ unbind_from_irq(irq);
++}
++EXPORT_SYMBOL_GPL(unbind_from_irqhandler);
++
++/* Rebind an evtchn so that it gets delivered to a specific cpu */
++static void rebind_irq_to_cpu(unsigned irq, unsigned tcpu)
++{
++ struct evtchn_bind_vcpu bind_vcpu;
++ int evtchn = evtchn_from_irq(irq);
++
++ if (!VALID_EVTCHN(evtchn))
++ return;
++
++ /* Send future instances of this interrupt to other vcpu. */
++ bind_vcpu.port = evtchn;
++ bind_vcpu.vcpu = tcpu;
++
++ /*
++ * If this fails, it usually just indicates that we're dealing with a
++ * virq or IPI channel, which don't actually need to be rebound. Ignore
++ * it, but don't do the xenlinux-level rebind in that case.
++ */
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_vcpu, &bind_vcpu) >= 0)
++ bind_evtchn_to_cpu(evtchn, tcpu);
++}
++
++
++static void set_affinity_irq(unsigned irq, cpumask_t dest)
++{
++ unsigned tcpu = first_cpu(dest);
++ rebind_irq_to_cpu(irq, tcpu);
++}
++
++/*
++ * Interface to generic handling in irq.c
++ */
++
++static unsigned int startup_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ unmask_evtchn(evtchn);
++ return 0;
++}
++
++static void shutdown_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ mask_evtchn(evtchn);
++}
++
++static void enable_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ unmask_evtchn(evtchn);
++}
++
++static void disable_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ mask_evtchn(evtchn);
++}
++
++static void ack_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ move_native_irq(irq);
++
++ if (VALID_EVTCHN(evtchn)) {
++ mask_evtchn(evtchn);
++ clear_evtchn(evtchn);
++ }
++}
++
++static void end_dynirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn) && !(irq_desc[irq].status & IRQ_DISABLED))
++ unmask_evtchn(evtchn);
++}
++
++static struct hw_interrupt_type dynirq_type = {
++ "Dynamic-irq",
++ startup_dynirq,
++ shutdown_dynirq,
++ enable_dynirq,
++ disable_dynirq,
++ ack_dynirq,
++ end_dynirq,
++ set_affinity_irq
++};
++
++static inline void pirq_unmask_notify(int pirq)
++{
++ struct physdev_eoi eoi = { .irq = pirq };
++ if (unlikely(test_bit(pirq, &pirq_needs_eoi[0])))
++ (void)HYPERVISOR_physdev_op(PHYSDEVOP_eoi, &eoi);
++}
++
++static inline void pirq_query_unmask(int pirq)
++{
++ struct physdev_irq_status_query irq_status;
++ irq_status.irq = pirq;
++ (void)HYPERVISOR_physdev_op(PHYSDEVOP_irq_status_query, &irq_status);
++ clear_bit(pirq, &pirq_needs_eoi[0]);
++ if (irq_status.flags & XENIRQSTAT_needs_eoi)
++ set_bit(pirq, &pirq_needs_eoi[0]);
++}
++
++/*
++ * On startup, if there is no action associated with the IRQ then we are
++ * probing. In this case we should not share with others as it will confuse us.
++ */
++#define probing_irq(_irq) (irq_desc[(_irq)].action == NULL)
++
++static unsigned int startup_pirq(unsigned int irq)
++{
++ struct evtchn_bind_pirq bind_pirq;
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ goto out;
++
++ bind_pirq.pirq = irq;
++ /* NB. We are happy to share unless we are probing. */
++ bind_pirq.flags = probing_irq(irq) ? 0 : BIND_PIRQ__WILL_SHARE;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_pirq, &bind_pirq) != 0) {
++ if (!probing_irq(irq))
++ printk(KERN_INFO "Failed to obtain physical IRQ %d\n",
++ irq);
++ return 0;
++ }
++ evtchn = bind_pirq.port;
++
++ pirq_query_unmask(irq_to_pirq(irq));
++
++ evtchn_to_irq[evtchn] = irq;
++ bind_evtchn_to_cpu(evtchn, 0);
++ irq_info[irq] = mk_irq_info(IRQT_PIRQ, irq, evtchn);
++
++ out:
++ unmask_evtchn(evtchn);
++ pirq_unmask_notify(irq_to_pirq(irq));
++
++ return 0;
++}
++
++static void shutdown_pirq(unsigned int irq)
++{
++ struct evtchn_close close;
++ int evtchn = evtchn_from_irq(irq);
++
++ if (!VALID_EVTCHN(evtchn))
++ return;
++
++ mask_evtchn(evtchn);
++
++ close.port = evtchn;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_close, &close) != 0)
++ BUG();
++
++ bind_evtchn_to_cpu(evtchn, 0);
++ evtchn_to_irq[evtchn] = -1;
++ irq_info[irq] = IRQ_UNBOUND;
++}
++
++static void enable_pirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn)) {
++ unmask_evtchn(evtchn);
++ pirq_unmask_notify(irq_to_pirq(irq));
++ }
++}
++
++static void disable_pirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ mask_evtchn(evtchn);
++}
++
++static void ack_pirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ move_native_irq(irq);
++
++ if (VALID_EVTCHN(evtchn)) {
++ mask_evtchn(evtchn);
++ clear_evtchn(evtchn);
++ }
++}
++
++static void end_pirq(unsigned int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn) && !(irq_desc[irq].status & IRQ_DISABLED)) {
++ unmask_evtchn(evtchn);
++ pirq_unmask_notify(irq_to_pirq(irq));
++ }
++}
++
++static struct hw_interrupt_type pirq_type = {
++ "Phys-irq",
++ startup_pirq,
++ shutdown_pirq,
++ enable_pirq,
++ disable_pirq,
++ ack_pirq,
++ end_pirq,
++ set_affinity_irq
++};
++
++int irq_ignore_unhandled(unsigned int irq)
++{
++ struct physdev_irq_status_query irq_status = { .irq = irq };
++
++ if (!is_running_on_xen())
++ return 0;
++
++ (void)HYPERVISOR_physdev_op(PHYSDEVOP_irq_status_query, &irq_status);
++ return !!(irq_status.flags & XENIRQSTAT_shared);
++}
++
++void resend_irq_on_evtchn(struct hw_interrupt_type *h, unsigned int i)
++{
++ int evtchn = evtchn_from_irq(i);
++ shared_info_t *s = HYPERVISOR_shared_info;
++ if (!VALID_EVTCHN(evtchn))
++ return;
++ BUG_ON(!synch_test_bit(evtchn, &s->evtchn_mask[0]));
++ synch_set_bit(evtchn, &s->evtchn_pending[0]);
++}
++
++void notify_remote_via_irq(int irq)
++{
++ int evtchn = evtchn_from_irq(irq);
++
++ if (VALID_EVTCHN(evtchn))
++ notify_remote_via_evtchn(evtchn);
++}
++EXPORT_SYMBOL_GPL(notify_remote_via_irq);
++
++void mask_evtchn(int port)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ synch_set_bit(port, &s->evtchn_mask[0]);
++}
++EXPORT_SYMBOL_GPL(mask_evtchn);
++
++void unmask_evtchn(int port)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ unsigned int cpu = smp_processor_id();
++ vcpu_info_t *vcpu_info = &s->vcpu_info[cpu];
++
++ BUG_ON(!irqs_disabled());
++
++ /* Slow path (hypercall) if this is a non-local port. */
++ if (unlikely(cpu != cpu_from_evtchn(port))) {
++ struct evtchn_unmask unmask = { .port = port };
++ (void)HYPERVISOR_event_channel_op(EVTCHNOP_unmask, &unmask);
++ return;
++ }
++
++ synch_clear_bit(port, &s->evtchn_mask[0]);
++
++ /*
++ * The following is basically the equivalent of 'hw_resend_irq'. Just
++ * like a real IO-APIC we 'lose the interrupt edge' if the channel is
++ * masked.
++ */
++ if (synch_test_bit(port, &s->evtchn_pending[0]) &&
++ !synch_test_and_set_bit(port / BITS_PER_LONG,
++ &vcpu_info->evtchn_pending_sel))
++ vcpu_info->evtchn_upcall_pending = 1;
++}
++EXPORT_SYMBOL_GPL(unmask_evtchn);
++
++void irq_resume(void)
++{
++ struct evtchn_bind_virq bind_virq;
++ struct evtchn_bind_ipi bind_ipi;
++ int cpu, pirq, virq, ipi, irq, evtchn;
++
++ init_evtchn_cpu_bindings();
++
++ /* New event-channel space is not 'live' yet. */
++ for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
++ mask_evtchn(evtchn);
++
++ /* Check that no PIRQs are still bound. */
++ for (pirq = 0; pirq < NR_PIRQS; pirq++)
++ BUG_ON(irq_info[pirq_to_irq(pirq)] != IRQ_UNBOUND);
++
++ /* Secondary CPUs must have no VIRQ or IPI bindings. */
++ for_each_possible_cpu(cpu) {
++ if (cpu == 0)
++ continue;
++ for (virq = 0; virq < NR_VIRQS; virq++)
++ BUG_ON(per_cpu(virq_to_irq, cpu)[virq] != -1);
++ for (ipi = 0; ipi < NR_IPIS; ipi++)
++ BUG_ON(per_cpu(ipi_to_irq, cpu)[ipi] != -1);
++ }
++
++ /* No IRQ <-> event-channel mappings. */
++ for (irq = 0; irq < NR_IRQS; irq++)
++ irq_info[irq] &= ~0xFFFF; /* zap event-channel binding */
++ for (evtchn = 0; evtchn < NR_EVENT_CHANNELS; evtchn++)
++ evtchn_to_irq[evtchn] = -1;
++
++ /* Primary CPU: rebind VIRQs automatically. */
++ for (virq = 0; virq < NR_VIRQS; virq++) {
++ if ((irq = per_cpu(virq_to_irq, 0)[virq]) == -1)
++ continue;
++
++ BUG_ON(irq_info[irq] != mk_irq_info(IRQT_VIRQ, virq, 0));
++
++ /* Get a new binding from Xen. */
++ bind_virq.virq = virq;
++ bind_virq.vcpu = 0;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
++ &bind_virq) != 0)
++ BUG();
++ evtchn = bind_virq.port;
++
++ /* Record the new mapping. */
++ evtchn_to_irq[evtchn] = irq;
++ irq_info[irq] = mk_irq_info(IRQT_VIRQ, virq, evtchn);
++
++ /* Ready for use. */
++ unmask_evtchn(evtchn);
++ }
++
++ /* Primary CPU: rebind IPIs automatically. */
++ for (ipi = 0; ipi < NR_IPIS; ipi++) {
++ if ((irq = per_cpu(ipi_to_irq, 0)[ipi]) == -1)
++ continue;
++
++ BUG_ON(irq_info[irq] != mk_irq_info(IRQT_IPI, ipi, 0));
++
++ /* Get a new binding from Xen. */
++ bind_ipi.vcpu = 0;
++ if (HYPERVISOR_event_channel_op(EVTCHNOP_bind_ipi,
++ &bind_ipi) != 0)
++ BUG();
++ evtchn = bind_ipi.port;
++
++ /* Record the new mapping. */
++ evtchn_to_irq[evtchn] = irq;
++ irq_info[irq] = mk_irq_info(IRQT_IPI, ipi, evtchn);
++
++ /* Ready for use. */
++ unmask_evtchn(evtchn);
++ }
++}
++
++void __init xen_init_IRQ(void)
++{
++ int i;
++
++ init_evtchn_cpu_bindings();
++
++ /* No event channels are 'live' right now. */
++ for (i = 0; i < NR_EVENT_CHANNELS; i++)
++ mask_evtchn(i);
++
++ /* No IRQ -> event-channel mappings. */
++ for (i = 0; i < NR_IRQS; i++)
++ irq_info[i] = IRQ_UNBOUND;
++
++ /* Dynamic IRQ space is currently unbound. Zero the refcnts. */
++ for (i = 0; i < NR_DYNIRQS; i++) {
++ irq_bindcount[dynirq_to_irq(i)] = 0;
++
++ irq_desc[dynirq_to_irq(i)].status = IRQ_DISABLED;
++ irq_desc[dynirq_to_irq(i)].action = NULL;
++ irq_desc[dynirq_to_irq(i)].depth = 1;
++ irq_desc[dynirq_to_irq(i)].handler = &dynirq_type;
++ }
++
++ /* Phys IRQ space is statically bound (1:1 mapping). Nail refcnts. */
++ for (i = 0; i < NR_PIRQS; i++) {
++ irq_bindcount[pirq_to_irq(i)] = 1;
++
++#ifdef RTC_IRQ
++ /* If not domain 0, force our RTC driver to fail its probe. */
++ if ((i == RTC_IRQ) &&
++ !(xen_start_info->flags & SIF_INITDOMAIN))
++ continue;
++#endif
++
++ irq_desc[pirq_to_irq(i)].status = IRQ_DISABLED;
++ irq_desc[pirq_to_irq(i)].action = NULL;
++ irq_desc[pirq_to_irq(i)].depth = 1;
++ irq_desc[pirq_to_irq(i)].handler = &pirq_type;
++ }
++}
+diff -urNp linux-2.6/drivers/xen/core/features.c new/drivers/xen/core/features.c
+--- linux-2.6/drivers/xen/core/features.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/features.c 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,30 @@
++/******************************************************************************
++ * features.c
++ *
++ * Xen feature flags.
++ *
++ * Copyright (c) 2006, Ian Campbell, XenSource Inc.
++ */
++#include <linux/types.h>
++#include <linux/cache.h>
++#include <linux/module.h>
++#include <asm/hypervisor.h>
++#include <xen/features.h>
++
++u8 xen_features[XENFEAT_NR_SUBMAPS * 32] __read_mostly;
++/* Not a GPL symbol: used in ubiquitous macros, so too restrictive. */
++EXPORT_SYMBOL(xen_features);
++
++void setup_xen_features(void)
++{
++ xen_feature_info_t fi;
++ int i, j;
++
++ for (i = 0; i < XENFEAT_NR_SUBMAPS; i++) {
++ fi.submap_idx = i;
++ if (HYPERVISOR_xen_version(XENVER_get_features, &fi) < 0)
++ break;
++ for (j=0; j<32; j++)
++ xen_features[i*32+j] = !!(fi.submap & 1<<j);
++ }
++}
+diff -urNp linux-2.6/drivers/xen/core/gnttab.c new/drivers/xen/core/gnttab.c
+--- linux-2.6/drivers/xen/core/gnttab.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/gnttab.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,442 @@
++/******************************************************************************
++ * gnttab.c
++ *
++ * Granting foreign access to our memory reservation.
++ *
++ * Copyright (c) 2005, Christopher Clark
++ * Copyright (c) 2004-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/sched.h>
++#include <linux/mm.h>
++#include <linux/vmalloc.h>
++#include <xen/interface/xen.h>
++#include <xen/gnttab.h>
++#include <asm/pgtable.h>
++#include <asm/uaccess.h>
++#include <asm/synch_bitops.h>
++
++/* External tools reserve first few grant table entries. */
++#define NR_RESERVED_ENTRIES 8
++
++#define NR_GRANT_ENTRIES \
++ (NR_GRANT_FRAMES * PAGE_SIZE / sizeof(struct grant_entry))
++#define GNTTAB_LIST_END (NR_GRANT_ENTRIES + 1)
++
++static grant_ref_t gnttab_list[NR_GRANT_ENTRIES];
++static int gnttab_free_count;
++static grant_ref_t gnttab_free_head;
++static DEFINE_SPINLOCK(gnttab_list_lock);
++
++static struct grant_entry *shared;
++
++static struct gnttab_free_callback *gnttab_free_callback_list;
++
++static int get_free_entries(int count)
++{
++ unsigned long flags;
++ int ref;
++ grant_ref_t head;
++ spin_lock_irqsave(&gnttab_list_lock, flags);
++ if (gnttab_free_count < count) {
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++ return -1;
++ }
++ ref = head = gnttab_free_head;
++ gnttab_free_count -= count;
++ while (count-- > 1)
++ head = gnttab_list[head];
++ gnttab_free_head = gnttab_list[head];
++ gnttab_list[head] = GNTTAB_LIST_END;
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++ return ref;
++}
++
++#define get_free_entry() get_free_entries(1)
++
++static void do_free_callbacks(void)
++{
++ struct gnttab_free_callback *callback, *next;
++
++ callback = gnttab_free_callback_list;
++ gnttab_free_callback_list = NULL;
++
++ while (callback != NULL) {
++ next = callback->next;
++ if (gnttab_free_count >= callback->count) {
++ callback->next = NULL;
++ callback->fn(callback->arg);
++ } else {
++ callback->next = gnttab_free_callback_list;
++ gnttab_free_callback_list = callback;
++ }
++ callback = next;
++ }
++}
++
++static inline void check_free_callbacks(void)
++{
++ if (unlikely(gnttab_free_callback_list))
++ do_free_callbacks();
++}
++
++static void put_free_entry(grant_ref_t ref)
++{
++ unsigned long flags;
++ spin_lock_irqsave(&gnttab_list_lock, flags);
++ gnttab_list[ref] = gnttab_free_head;
++ gnttab_free_head = ref;
++ gnttab_free_count++;
++ check_free_callbacks();
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++}
++
++/*
++ * Public grant-issuing interface functions
++ */
++
++int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
++ int readonly)
++{
++ int ref;
++
++ if (unlikely((ref = get_free_entry()) == -1))
++ return -ENOSPC;
++
++ shared[ref].frame = frame;
++ shared[ref].domid = domid;
++ wmb();
++ shared[ref].flags = GTF_permit_access | (readonly ? GTF_readonly : 0);
++
++ return ref;
++}
++EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access);
++
++void gnttab_grant_foreign_access_ref(grant_ref_t ref, domid_t domid,
++ unsigned long frame, int readonly)
++{
++ shared[ref].frame = frame;
++ shared[ref].domid = domid;
++ wmb();
++ shared[ref].flags = GTF_permit_access | (readonly ? GTF_readonly : 0);
++}
++EXPORT_SYMBOL_GPL(gnttab_grant_foreign_access_ref);
++
++
++int gnttab_query_foreign_access(grant_ref_t ref)
++{
++ u16 nflags;
++
++ nflags = shared[ref].flags;
++
++ return (nflags & (GTF_reading|GTF_writing));
++}
++EXPORT_SYMBOL_GPL(gnttab_query_foreign_access);
++
++int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly)
++{
++ u16 flags, nflags;
++
++ nflags = shared[ref].flags;
++ do {
++ if ((flags = nflags) & (GTF_reading|GTF_writing)) {
++ printk(KERN_ALERT "WARNING: g.e. still in use!\n");
++ return 0;
++ }
++ } while ((nflags = synch_cmpxchg_subword(&shared[ref].flags, flags, 0)) !=
++ flags);
++
++ return 1;
++}
++EXPORT_SYMBOL_GPL(gnttab_end_foreign_access_ref);
++
++void gnttab_end_foreign_access(grant_ref_t ref, int readonly,
++ unsigned long page)
++{
++ if (gnttab_end_foreign_access_ref(ref, readonly)) {
++ put_free_entry(ref);
++ if (page != 0)
++ free_page(page);
++ } else {
++ /* XXX This needs to be fixed so that the ref and page are
++ placed on a list to be freed up later. */
++ printk(KERN_WARNING
++ "WARNING: leaking g.e. and page still in use!\n");
++ }
++}
++EXPORT_SYMBOL_GPL(gnttab_end_foreign_access);
++
++int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn)
++{
++ int ref;
++
++ if (unlikely((ref = get_free_entry()) == -1))
++ return -ENOSPC;
++ gnttab_grant_foreign_transfer_ref(ref, domid, pfn);
++
++ return ref;
++}
++EXPORT_SYMBOL_GPL(gnttab_grant_foreign_transfer);
++
++void gnttab_grant_foreign_transfer_ref(grant_ref_t ref, domid_t domid,
++ unsigned long pfn)
++{
++ shared[ref].frame = pfn;
++ shared[ref].domid = domid;
++ wmb();
++ shared[ref].flags = GTF_accept_transfer;
++}
++EXPORT_SYMBOL_GPL(gnttab_grant_foreign_transfer_ref);
++
++unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref)
++{
++ unsigned long frame;
++ u16 flags;
++
++ /*
++ * If a transfer is not even yet started, try to reclaim the grant
++ * reference and return failure (== 0).
++ */
++ while (!((flags = shared[ref].flags) & GTF_transfer_committed)) {
++ if (synch_cmpxchg_subword(&shared[ref].flags, flags, 0) == flags)
++ return 0;
++ cpu_relax();
++ }
++
++ /* If a transfer is in progress then wait until it is completed. */
++ while (!(flags & GTF_transfer_completed)) {
++ flags = shared[ref].flags;
++ cpu_relax();
++ }
++
++ /* Read the frame number /after/ reading completion status. */
++ rmb();
++ frame = shared[ref].frame;
++ BUG_ON(frame == 0);
++
++ return frame;
++}
++EXPORT_SYMBOL_GPL(gnttab_end_foreign_transfer_ref);
++
++unsigned long gnttab_end_foreign_transfer(grant_ref_t ref)
++{
++ unsigned long frame = gnttab_end_foreign_transfer_ref(ref);
++ put_free_entry(ref);
++ return frame;
++}
++EXPORT_SYMBOL_GPL(gnttab_end_foreign_transfer);
++
++void gnttab_free_grant_reference(grant_ref_t ref)
++{
++ put_free_entry(ref);
++}
++EXPORT_SYMBOL_GPL(gnttab_free_grant_reference);
++
++void gnttab_free_grant_references(grant_ref_t head)
++{
++ grant_ref_t ref;
++ unsigned long flags;
++ int count = 1;
++ if (head == GNTTAB_LIST_END)
++ return;
++ spin_lock_irqsave(&gnttab_list_lock, flags);
++ ref = head;
++ while (gnttab_list[ref] != GNTTAB_LIST_END) {
++ ref = gnttab_list[ref];
++ count++;
++ }
++ gnttab_list[ref] = gnttab_free_head;
++ gnttab_free_head = head;
++ gnttab_free_count += count;
++ check_free_callbacks();
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++}
++EXPORT_SYMBOL_GPL(gnttab_free_grant_references);
++
++int gnttab_alloc_grant_references(u16 count, grant_ref_t *head)
++{
++ int h = get_free_entries(count);
++
++ if (h == -1)
++ return -ENOSPC;
++
++ *head = h;
++
++ return 0;
++}
++EXPORT_SYMBOL_GPL(gnttab_alloc_grant_references);
++
++int gnttab_empty_grant_references(const grant_ref_t *private_head)
++{
++ return (*private_head == GNTTAB_LIST_END);
++}
++EXPORT_SYMBOL_GPL(gnttab_empty_grant_references);
++
++int gnttab_claim_grant_reference(grant_ref_t *private_head)
++{
++ grant_ref_t g = *private_head;
++ if (unlikely(g == GNTTAB_LIST_END))
++ return -ENOSPC;
++ *private_head = gnttab_list[g];
++ return g;
++}
++EXPORT_SYMBOL_GPL(gnttab_claim_grant_reference);
++
++void gnttab_release_grant_reference(grant_ref_t *private_head,
++ grant_ref_t release)
++{
++ gnttab_list[release] = *private_head;
++ *private_head = release;
++}
++EXPORT_SYMBOL_GPL(gnttab_release_grant_reference);
++
++void gnttab_request_free_callback(struct gnttab_free_callback *callback,
++ void (*fn)(void *), void *arg, u16 count)
++{
++ unsigned long flags;
++ spin_lock_irqsave(&gnttab_list_lock, flags);
++ if (callback->next)
++ goto out;
++ callback->fn = fn;
++ callback->arg = arg;
++ callback->count = count;
++ callback->next = gnttab_free_callback_list;
++ gnttab_free_callback_list = callback;
++ check_free_callbacks();
++out:
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++}
++EXPORT_SYMBOL_GPL(gnttab_request_free_callback);
++
++void gnttab_cancel_free_callback(struct gnttab_free_callback *callback)
++{
++ struct gnttab_free_callback **pcb;
++ unsigned long flags;
++
++ spin_lock_irqsave(&gnttab_list_lock, flags);
++ for (pcb = &gnttab_free_callback_list; *pcb; pcb = &(*pcb)->next) {
++ if (*pcb == callback) {
++ *pcb = callback->next;
++ break;
++ }
++ }
++ spin_unlock_irqrestore(&gnttab_list_lock, flags);
++}
++EXPORT_SYMBOL_GPL(gnttab_cancel_free_callback);
++
++#ifndef __ia64__
++static int map_pte_fn(pte_t *pte, struct page *pmd_page,
++ unsigned long addr, void *data)
++{
++ unsigned long **frames = (unsigned long **)data;
++
++ set_pte_at(&init_mm, addr, pte, pfn_pte_ma((*frames)[0], PAGE_KERNEL));
++ (*frames)++;
++ return 0;
++}
++
++static int unmap_pte_fn(pte_t *pte, struct page *pmd_page,
++ unsigned long addr, void *data)
++{
++
++ set_pte_at(&init_mm, addr, pte, __pte(0));
++ return 0;
++}
++#endif
++
++int gnttab_resume(void)
++{
++ struct gnttab_setup_table setup;
++ unsigned long frames[NR_GRANT_FRAMES];
++ int rc;
++#ifndef __ia64__
++ void *pframes = frames;
++ struct vm_struct *area;
++#endif
++
++ setup.dom = DOMID_SELF;
++ setup.nr_frames = NR_GRANT_FRAMES;
++ set_xen_guest_handle(setup.frame_list, frames);
++
++ rc = HYPERVISOR_grant_table_op(GNTTABOP_setup_table, &setup, 1);
++ if (rc == -ENOSYS)
++ return -ENOSYS;
++
++ BUG_ON(rc || setup.status);
++
++#ifndef __ia64__
++ if (shared == NULL) {
++ area = get_vm_area(PAGE_SIZE * NR_GRANT_FRAMES, VM_IOREMAP);
++ BUG_ON(area == NULL);
++ shared = area->addr;
++ }
++ rc = apply_to_page_range(&init_mm, (unsigned long)shared,
++ PAGE_SIZE * NR_GRANT_FRAMES,
++ map_pte_fn, &pframes);
++ BUG_ON(rc);
++#else
++ shared = __va(frames[0] << PAGE_SHIFT);
++ printk("grant table at %p\n", shared);
++#endif
++
++ return 0;
++}
++
++int gnttab_suspend(void)
++{
++
++#ifndef __ia64__
++ apply_to_page_range(&init_mm, (unsigned long)shared,
++ PAGE_SIZE * NR_GRANT_FRAMES,
++ unmap_pte_fn, NULL);
++#endif
++
++ return 0;
++}
++
++static int __init gnttab_init(void)
++{
++ int i;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ if (gnttab_resume() < 0)
++ return -ENODEV;
++
++ for (i = NR_RESERVED_ENTRIES; i < NR_GRANT_ENTRIES; i++)
++ gnttab_list[i] = i + 1;
++ gnttab_free_count = NR_GRANT_ENTRIES - NR_RESERVED_ENTRIES;
++ gnttab_free_head = NR_RESERVED_ENTRIES;
++
++ printk("Grant table initialized\n");
++ return 0;
++}
++
++core_initcall(gnttab_init);
+diff -urNp linux-2.6/drivers/xen/core/hypervisor_sysfs.c new/drivers/xen/core/hypervisor_sysfs.c
+--- linux-2.6/drivers/xen/core/hypervisor_sysfs.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/hypervisor_sysfs.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,60 @@
++/*
++ * copyright (c) 2006 IBM Corporation
++ * Authored by: Mike D. Day <ncmike@us.ibm.com>
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License version 2 as
++ * published by the Free Software Foundation.
++ */
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/module.h>
++#include <linux/kobject.h>
++#include <xen/hypervisor_sysfs.h>
++
++decl_subsys(hypervisor, NULL, NULL);
++
++static ssize_t hyp_sysfs_show(struct kobject *kobj,
++ struct attribute *attr,
++ char *buffer)
++{
++ struct hyp_sysfs_attr *hyp_attr;
++ hyp_attr = container_of(attr, struct hyp_sysfs_attr, attr);
++ if (hyp_attr->show)
++ return hyp_attr->show(hyp_attr, buffer);
++ return 0;
++}
++
++static ssize_t hyp_sysfs_store(struct kobject *kobj,
++ struct attribute *attr,
++ const char *buffer,
++ size_t len)
++{
++ struct hyp_sysfs_attr *hyp_attr;
++ hyp_attr = container_of(attr, struct hyp_sysfs_attr, attr);
++ if (hyp_attr->store)
++ return hyp_attr->store(hyp_attr, buffer, len);
++ return 0;
++}
++
++struct sysfs_ops hyp_sysfs_ops = {
++ .show = hyp_sysfs_show,
++ .store = hyp_sysfs_store,
++};
++
++static struct kobj_type hyp_sysfs_kobj_type = {
++ .sysfs_ops = &hyp_sysfs_ops,
++};
++
++static int __init hypervisor_subsys_init(void)
++{
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ hypervisor_subsys.kset.kobj.ktype = &hyp_sysfs_kobj_type;
++ return subsystem_register(&hypervisor_subsys);
++}
++
++device_initcall(hypervisor_subsys_init);
++EXPORT_SYMBOL_GPL(hypervisor_subsys);
+diff -urNp linux-2.6/drivers/xen/core/Makefile new/drivers/xen/core/Makefile
+--- linux-2.6/drivers/xen/core/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/Makefile 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,13 @@
++#
++# Makefile for the linux kernel.
++#
++
++obj-y := evtchn.o gnttab.o features.o
++
++obj-$(CONFIG_PROC_FS) += xen_proc.o
++obj-$(CONFIG_SYSFS) += hypervisor_sysfs.o
++obj-$(CONFIG_HOTPLUG_CPU) += cpu_hotplug.o
++obj-$(CONFIG_XEN_SYSFS) += xen_sysfs.o
++obj-$(CONFIG_XEN_SKBUFF) += skbuff.o
++obj-$(CONFIG_XEN_REBOOT) += reboot.o
++obj-$(CONFIG_XEN_SMPBOOT) += smpboot.o
+diff -urNp linux-2.6/drivers/xen/core/reboot.c new/drivers/xen/core/reboot.c
+--- linux-2.6/drivers/xen/core/reboot.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/reboot.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,364 @@
++#define __KERNEL_SYSCALLS__
++#include <linux/version.h>
++#include <linux/kernel.h>
++#include <linux/mm.h>
++#include <linux/unistd.h>
++#include <linux/module.h>
++#include <linux/reboot.h>
++#include <linux/sysrq.h>
++#include <linux/stringify.h>
++#include <asm/irq.h>
++#include <asm/mmu_context.h>
++#include <xen/evtchn.h>
++#include <asm/hypervisor.h>
++#include <xen/interface/dom0_ops.h>
++#include <xen/xenbus.h>
++#include <linux/cpu.h>
++#include <linux/kthread.h>
++#include <xen/gnttab.h>
++#include <xen/xencons.h>
++#include <xen/cpu_hotplug.h>
++
++#if defined(__i386__) || defined(__x86_64__)
++/*
++ * Power off function, if any
++ */
++void (*pm_power_off)(void);
++EXPORT_SYMBOL(pm_power_off);
++#endif
++
++extern void ctrl_alt_del(void);
++
++#define SHUTDOWN_INVALID -1
++#define SHUTDOWN_POWEROFF 0
++#define SHUTDOWN_SUSPEND 2
++/* Code 3 is SHUTDOWN_CRASH, which we don't use because the domain can only
++ * report a crash, not be instructed to crash!
++ * HALT is the same as POWEROFF, as far as we're concerned. The tools use
++ * the distinction when we return the reason code to them.
++ */
++#define SHUTDOWN_HALT 4
++
++void machine_emergency_restart(void)
++{
++ /* We really want to get pending console data out before we die. */
++ xencons_force_flush();
++ HYPERVISOR_shutdown(SHUTDOWN_reboot);
++}
++
++void machine_restart(char * __unused)
++{
++ machine_emergency_restart();
++}
++
++void machine_halt(void)
++{
++ machine_power_off();
++}
++
++void machine_power_off(void)
++{
++ /* We really want to get pending console data out before we die. */
++ xencons_force_flush();
++#if defined(__i386__) || defined(__x86_64__)
++ if (pm_power_off)
++ pm_power_off();
++#endif
++ HYPERVISOR_shutdown(SHUTDOWN_poweroff);
++}
++
++int reboot_thru_bios = 0; /* for dmi_scan.c */
++EXPORT_SYMBOL(machine_restart);
++EXPORT_SYMBOL(machine_halt);
++EXPORT_SYMBOL(machine_power_off);
++
++
++/******************************************************************************
++ * Stop/pickle callback handling.
++ */
++
++/* Ignore multiple shutdown requests. */
++static int shutting_down = SHUTDOWN_INVALID;
++static void __shutdown_handler(void *unused);
++static DECLARE_WORK(shutdown_work, __shutdown_handler, NULL);
++
++/* Ensure we run on the idle task page tables so that we will
++ switch page tables before running user space. This is needed
++ on architectures with separate kernel and user page tables
++ because the user page table pointer is not saved/restored. */
++static void switch_idle_mm(void)
++{
++ struct mm_struct *mm = current->active_mm;
++
++ if (mm == &init_mm)
++ return;
++
++ atomic_inc(&init_mm.mm_count);
++ switch_mm(mm, &init_mm, current);
++ current->active_mm = &init_mm;
++ mmdrop(mm);
++}
++
++static int __do_suspend(void *ignore)
++{
++ int i, j, k, fpp, err;
++
++ extern unsigned long max_pfn;
++ extern unsigned long *pfn_to_mfn_frame_list_list;
++ extern unsigned long *pfn_to_mfn_frame_list[];
++
++ extern void time_resume(void);
++
++ BUG_ON(smp_processor_id() != 0);
++ BUG_ON(in_interrupt());
++
++ if (xen_feature(XENFEAT_auto_translated_physmap)) {
++ printk(KERN_WARNING "Cannot suspend in "
++ "auto_translated_physmap mode.\n");
++ return -EOPNOTSUPP;
++ }
++
++ err = smp_suspend();
++ if (err)
++ return err;
++
++ xenbus_suspend();
++
++ preempt_disable();
++
++#ifdef __i386__
++ kmem_cache_shrink(pgd_cache);
++#endif
++ mm_pin_all();
++
++ __cli();
++ preempt_enable();
++
++ gnttab_suspend();
++
++ HYPERVISOR_shared_info = (shared_info_t *)empty_zero_page;
++ clear_fixmap(FIX_SHARED_INFO);
++
++ xen_start_info->store_mfn = mfn_to_pfn(xen_start_info->store_mfn);
++ xen_start_info->console_mfn = mfn_to_pfn(xen_start_info->console_mfn);
++
++ /*
++ * We'll stop somewhere inside this hypercall. When it returns,
++ * we'll start resuming after the restore.
++ */
++ HYPERVISOR_suspend(virt_to_mfn(xen_start_info));
++
++ shutting_down = SHUTDOWN_INVALID;
++
++ set_fixmap(FIX_SHARED_INFO, xen_start_info->shared_info);
++
++ HYPERVISOR_shared_info = (shared_info_t *)fix_to_virt(FIX_SHARED_INFO);
++
++ memset(empty_zero_page, 0, PAGE_SIZE);
++
++ HYPERVISOR_shared_info->arch.pfn_to_mfn_frame_list_list =
++ virt_to_mfn(pfn_to_mfn_frame_list_list);
++
++ fpp = PAGE_SIZE/sizeof(unsigned long);
++ for (i = 0, j = 0, k = -1; i < max_pfn; i += fpp, j++) {
++ if ((j % fpp) == 0) {
++ k++;
++ pfn_to_mfn_frame_list_list[k] =
++ virt_to_mfn(pfn_to_mfn_frame_list[k]);
++ j = 0;
++ }
++ pfn_to_mfn_frame_list[k][j] =
++ virt_to_mfn(&phys_to_machine_mapping[i]);
++ }
++ HYPERVISOR_shared_info->arch.max_pfn = max_pfn;
++
++ gnttab_resume();
++
++ irq_resume();
++
++ time_resume();
++
++ switch_idle_mm();
++
++ __sti();
++
++ xencons_resume();
++
++ xenbus_resume();
++
++ smp_resume();
++
++ return err;
++}
++
++static int shutdown_process(void *__unused)
++{
++ static char *envp[] = { "HOME=/", "TERM=linux",
++ "PATH=/sbin:/usr/sbin:/bin:/usr/bin", NULL };
++ static char *poweroff_argv[] = { "/sbin/poweroff", NULL };
++
++ extern asmlinkage long sys_reboot(int magic1, int magic2,
++ unsigned int cmd, void *arg);
++
++ if ((shutting_down == SHUTDOWN_POWEROFF) ||
++ (shutting_down == SHUTDOWN_HALT)) {
++ if (execve("/sbin/poweroff", poweroff_argv, envp) < 0) {
++ sys_reboot(LINUX_REBOOT_MAGIC1,
++ LINUX_REBOOT_MAGIC2,
++ LINUX_REBOOT_CMD_POWER_OFF,
++ NULL);
++ }
++ }
++
++ shutting_down = SHUTDOWN_INVALID; /* could try again */
++
++ return 0;
++}
++
++static int kthread_create_on_cpu(int (*f)(void *arg),
++ void *arg,
++ const char *name,
++ int cpu)
++{
++ struct task_struct *p;
++ p = kthread_create(f, arg, name);
++ if (IS_ERR(p))
++ return PTR_ERR(p);
++ kthread_bind(p, cpu);
++ wake_up_process(p);
++ return 0;
++}
++
++static void __shutdown_handler(void *unused)
++{
++ int err;
++
++ if (shutting_down != SHUTDOWN_SUSPEND)
++ err = kernel_thread(shutdown_process, NULL,
++ CLONE_FS | CLONE_FILES);
++ else
++ err = kthread_create_on_cpu(__do_suspend, NULL, "suspend", 0);
++
++ if (err < 0) {
++ printk(KERN_WARNING "Error creating shutdown process (%d): "
++ "retrying...\n", -err);
++ schedule_delayed_work(&shutdown_work, HZ/2);
++ }
++}
++
++static void shutdown_handler(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ char *str;
++ struct xenbus_transaction xbt;
++ int err;
++
++ if (shutting_down != SHUTDOWN_INVALID)
++ return;
++
++ again:
++ err = xenbus_transaction_start(&xbt);
++ if (err)
++ return;
++ str = (char *)xenbus_read(xbt, "control", "shutdown", NULL);
++ /* Ignore read errors and empty reads. */
++ if (XENBUS_IS_ERR_READ(str)) {
++ xenbus_transaction_end(xbt, 1);
++ return;
++ }
++
++ xenbus_write(xbt, "control", "shutdown", "");
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err == -EAGAIN) {
++ kfree(str);
++ goto again;
++ }
++
++ if (strcmp(str, "poweroff") == 0)
++ shutting_down = SHUTDOWN_POWEROFF;
++ else if (strcmp(str, "reboot") == 0)
++ ctrl_alt_del();
++ else if (strcmp(str, "suspend") == 0)
++ shutting_down = SHUTDOWN_SUSPEND;
++ else if (strcmp(str, "halt") == 0)
++ shutting_down = SHUTDOWN_HALT;
++ else {
++ printk("Ignoring shutdown request: %s\n", str);
++ shutting_down = SHUTDOWN_INVALID;
++ }
++
++ if (shutting_down != SHUTDOWN_INVALID)
++ schedule_work(&shutdown_work);
++
++ kfree(str);
++}
++
++static void sysrq_handler(struct xenbus_watch *watch, const char **vec,
++ unsigned int len)
++{
++ char sysrq_key = '\0';
++ struct xenbus_transaction xbt;
++ int err;
++
++ again:
++ err = xenbus_transaction_start(&xbt);
++ if (err)
++ return;
++ if (!xenbus_scanf(xbt, "control", "sysrq", "%c", &sysrq_key)) {
++ printk(KERN_ERR "Unable to read sysrq code in "
++ "control/sysrq\n");
++ xenbus_transaction_end(xbt, 1);
++ return;
++ }
++
++ if (sysrq_key != '\0')
++ xenbus_printf(xbt, "control", "sysrq", "%c", '\0');
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err == -EAGAIN)
++ goto again;
++
++#ifdef CONFIG_MAGIC_SYSRQ
++ if (sysrq_key != '\0')
++ handle_sysrq(sysrq_key, NULL, NULL);
++#endif
++}
++
++static struct xenbus_watch shutdown_watch = {
++ .node = "control/shutdown",
++ .callback = shutdown_handler
++};
++
++static struct xenbus_watch sysrq_watch = {
++ .node ="control/sysrq",
++ .callback = sysrq_handler
++};
++
++static int setup_shutdown_watcher(struct notifier_block *notifier,
++ unsigned long event,
++ void *data)
++{
++ int err;
++
++ err = register_xenbus_watch(&shutdown_watch);
++ if (err)
++ printk(KERN_ERR "Failed to set shutdown watcher\n");
++
++ err = register_xenbus_watch(&sysrq_watch);
++ if (err)
++ printk(KERN_ERR "Failed to set sysrq watcher\n");
++
++ return NOTIFY_DONE;
++}
++
++static int __init setup_shutdown_event(void)
++{
++ static struct notifier_block xenstore_notifier = {
++ .notifier_call = setup_shutdown_watcher
++ };
++ register_xenstore_notifier(&xenstore_notifier);
++ return 0;
++}
++
++subsys_initcall(setup_shutdown_event);
+diff -urNp linux-2.6/drivers/xen/core/skbuff.c new/drivers/xen/core/skbuff.c
+--- linux-2.6/drivers/xen/core/skbuff.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/skbuff.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,141 @@
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/version.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/netdevice.h>
++#include <linux/inetdevice.h>
++#include <linux/etherdevice.h>
++#include <linux/skbuff.h>
++#include <linux/init.h>
++#include <asm/io.h>
++#include <asm/page.h>
++#include <asm/hypervisor.h>
++
++/* Referenced in netback.c. */
++/*static*/ kmem_cache_t *skbuff_cachep;
++EXPORT_SYMBOL(skbuff_cachep);
++
++#define MAX_SKBUFF_ORDER 4
++static kmem_cache_t *skbuff_order_cachep[MAX_SKBUFF_ORDER + 1];
++
++static struct {
++ int size;
++ kmem_cache_t *cachep;
++} skbuff_small[] = { { 512, NULL }, { 2048, NULL } };
++
++struct sk_buff *__alloc_skb(unsigned int length, gfp_t gfp_mask,
++ int fclone)
++{
++ int order, i;
++ kmem_cache_t *cachep;
++
++ length = SKB_DATA_ALIGN(length) + sizeof(struct skb_shared_info);
++
++ if (length <= skbuff_small[ARRAY_SIZE(skbuff_small)-1].size) {
++ for (i = 0; skbuff_small[i].size < length; i++)
++ continue;
++ cachep = skbuff_small[i].cachep;
++ } else {
++ order = get_order(length);
++ if (order > MAX_SKBUFF_ORDER) {
++ printk(KERN_ALERT "Attempt to allocate order %d "
++ "skbuff. Increase MAX_SKBUFF_ORDER.\n", order);
++ return NULL;
++ }
++ cachep = skbuff_order_cachep[order];
++ }
++
++ length -= sizeof(struct skb_shared_info);
++
++ return alloc_skb_from_cache(cachep, length, gfp_mask, fclone);
++}
++
++struct sk_buff *__dev_alloc_skb(unsigned int length, gfp_t gfp_mask)
++{
++ struct sk_buff *skb;
++ int order;
++
++ length = SKB_DATA_ALIGN(length + 16);
++ order = get_order(length + sizeof(struct skb_shared_info));
++ if (order > MAX_SKBUFF_ORDER) {
++ printk(KERN_ALERT "Attempt to allocate order %d skbuff. "
++ "Increase MAX_SKBUFF_ORDER.\n", order);
++ return NULL;
++ }
++
++ skb = alloc_skb_from_cache(
++ skbuff_order_cachep[order], length, gfp_mask, 0);
++ if (skb != NULL)
++ skb_reserve(skb, 16);
++
++ return skb;
++}
++
++static void skbuff_ctor(void *buf, kmem_cache_t *cachep, unsigned long unused)
++{
++ int order = 0;
++
++ while (skbuff_order_cachep[order] != cachep)
++ order++;
++
++ /* Do our best to allocate contiguous memory but fall back to IOMMU. */
++ if (order != 0)
++ (void)xen_create_contiguous_region(
++ (unsigned long)buf, order, 0);
++
++ scrub_pages(buf, 1 << order);
++}
++
++static void skbuff_dtor(void *buf, kmem_cache_t *cachep, unsigned long unused)
++{
++ int order = 0;
++
++ while (skbuff_order_cachep[order] != cachep)
++ order++;
++
++ if (order != 0)
++ xen_destroy_contiguous_region((unsigned long)buf, order);
++}
++
++static int __init skbuff_init(void)
++{
++ static char name[MAX_SKBUFF_ORDER + 1][20];
++ static char small_name[ARRAY_SIZE(skbuff_small)][20];
++ unsigned long size;
++ int i, order;
++
++ for (i = 0; i < ARRAY_SIZE(skbuff_small); i++) {
++ size = skbuff_small[i].size;
++ sprintf(small_name[i], "xen-skb-%lu", size);
++ /*
++ * No ctor/dtor: objects do not span page boundaries, and they
++ * are only used on transmit path so no need for scrubbing.
++ */
++ skbuff_small[i].cachep = kmem_cache_create(
++ small_name[i], size, size, 0, NULL, NULL);
++ }
++
++ for (order = 0; order <= MAX_SKBUFF_ORDER; order++) {
++ size = PAGE_SIZE << order;
++ sprintf(name[order], "xen-skb-%lu", size);
++ if (is_running_on_xen() &&
++ (xen_start_info->flags & SIF_PRIVILEGED))
++ skbuff_order_cachep[order] = kmem_cache_create(
++ name[order], size, size, 0,
++ skbuff_ctor, skbuff_dtor);
++ else
++ skbuff_order_cachep[order] = kmem_cache_create(
++ name[order], size, size, 0, NULL, NULL);
++
++ }
++
++ skbuff_cachep = skbuff_order_cachep[0];
++
++ return 0;
++}
++core_initcall(skbuff_init);
++
++EXPORT_SYMBOL(__dev_alloc_skb);
+diff -urNp linux-2.6/drivers/xen/core/smpboot.c new/drivers/xen/core/smpboot.c
+--- linux-2.6/drivers/xen/core/smpboot.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/smpboot.c 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,433 @@
++/*
++ * Xen SMP booting functions
++ *
++ * See arch/i386/kernel/smpboot.c for copyright and credits for derived
++ * portions of this file.
++ */
++
++#include <linux/module.h>
++#include <linux/config.h>
++#include <linux/init.h>
++#include <linux/kernel.h>
++#include <linux/mm.h>
++#include <linux/sched.h>
++#include <linux/kernel_stat.h>
++#include <linux/smp_lock.h>
++#include <linux/irq.h>
++#include <linux/bootmem.h>
++#include <linux/notifier.h>
++#include <linux/cpu.h>
++#include <linux/percpu.h>
++#include <asm/desc.h>
++#include <asm/arch_hooks.h>
++#include <asm/pgalloc.h>
++#include <xen/evtchn.h>
++#include <xen/interface/vcpu.h>
++#include <xen/cpu_hotplug.h>
++#include <xen/xenbus.h>
++
++extern irqreturn_t smp_reschedule_interrupt(int, void *, struct pt_regs *);
++extern irqreturn_t smp_call_function_interrupt(int, void *, struct pt_regs *);
++
++extern void local_setup_timer(unsigned int cpu);
++extern void local_teardown_timer(unsigned int cpu);
++
++extern void hypervisor_callback(void);
++extern void failsafe_callback(void);
++extern void system_call(void);
++extern void smp_trap_init(trap_info_t *);
++
++/* Number of siblings per CPU package */
++int smp_num_siblings = 1;
++int phys_proc_id[NR_CPUS]; /* Package ID of each logical CPU */
++EXPORT_SYMBOL(phys_proc_id);
++int cpu_core_id[NR_CPUS]; /* Core ID of each logical CPU */
++EXPORT_SYMBOL(cpu_core_id);
++#if defined(__i386__)
++int cpu_llc_id[NR_CPUS] __cpuinitdata = {[0 ... NR_CPUS-1] = BAD_APICID};
++#elif defined(__x86_64__)
++u8 cpu_llc_id[NR_CPUS] __cpuinitdata = {[0 ... NR_CPUS-1] = BAD_APICID};
++#endif
++
++cpumask_t cpu_online_map;
++EXPORT_SYMBOL(cpu_online_map);
++cpumask_t cpu_possible_map;
++EXPORT_SYMBOL(cpu_possible_map);
++
++struct cpuinfo_x86 cpu_data[NR_CPUS] __cacheline_aligned;
++EXPORT_SYMBOL(cpu_data);
++
++#ifdef CONFIG_HOTPLUG_CPU
++DEFINE_PER_CPU(int, cpu_state) = { 0 };
++#endif
++
++static DEFINE_PER_CPU(int, resched_irq);
++static DEFINE_PER_CPU(int, callfunc_irq);
++static char resched_name[NR_CPUS][15];
++static char callfunc_name[NR_CPUS][15];
++
++u8 cpu_2_logical_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = BAD_APICID };
++
++void *xquad_portio;
++
++cpumask_t cpu_sibling_map[NR_CPUS] __cacheline_aligned;
++cpumask_t cpu_core_map[NR_CPUS] __cacheline_aligned;
++EXPORT_SYMBOL(cpu_core_map);
++
++#if defined(__i386__)
++u8 x86_cpu_to_apicid[NR_CPUS] = { [0 ... NR_CPUS-1] = 0xff };
++EXPORT_SYMBOL(x86_cpu_to_apicid);
++#elif !defined(CONFIG_X86_IO_APIC)
++unsigned int maxcpus = NR_CPUS;
++#endif
++
++void __init prefill_possible_map(void)
++{
++ int i, rc;
++
++ for (i = 0; i < NR_CPUS; i++) {
++ rc = HYPERVISOR_vcpu_op(VCPUOP_is_up, i, NULL);
++ if (rc >= 0)
++ cpu_set(i, cpu_possible_map);
++ }
++}
++
++void __init smp_alloc_memory(void)
++{
++}
++
++static inline void
++set_cpu_sibling_map(int cpu)
++{
++ phys_proc_id[cpu] = cpu;
++ cpu_core_id[cpu] = 0;
++
++ cpu_sibling_map[cpu] = cpumask_of_cpu(cpu);
++ cpu_core_map[cpu] = cpumask_of_cpu(cpu);
++
++ cpu_data[cpu].booted_cores = 1;
++}
++
++static void xen_smp_intr_init(unsigned int cpu)
++{
++ sprintf(resched_name[cpu], "resched%d", cpu);
++ per_cpu(resched_irq, cpu) =
++ bind_ipi_to_irqhandler(
++ RESCHEDULE_VECTOR,
++ cpu,
++ smp_reschedule_interrupt,
++ SA_INTERRUPT,
++ resched_name[cpu],
++ NULL);
++ BUG_ON(per_cpu(resched_irq, cpu) < 0);
++
++ sprintf(callfunc_name[cpu], "callfunc%d", cpu);
++ per_cpu(callfunc_irq, cpu) =
++ bind_ipi_to_irqhandler(
++ CALL_FUNCTION_VECTOR,
++ cpu,
++ smp_call_function_interrupt,
++ SA_INTERRUPT,
++ callfunc_name[cpu],
++ NULL);
++ BUG_ON(per_cpu(callfunc_irq, cpu) < 0);
++
++ if (cpu != 0)
++ local_setup_timer(cpu);
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++static void xen_smp_intr_exit(unsigned int cpu)
++{
++ if (cpu != 0)
++ local_teardown_timer(cpu);
++
++ unbind_from_irqhandler(per_cpu(resched_irq, cpu), NULL);
++ unbind_from_irqhandler(per_cpu(callfunc_irq, cpu), NULL);
++}
++#endif
++
++void cpu_bringup(void)
++{
++ cpu_init();
++ touch_softlockup_watchdog();
++ preempt_disable();
++ local_irq_enable();
++}
++
++static void cpu_bringup_and_idle(void)
++{
++ cpu_bringup();
++ cpu_idle();
++}
++
++void cpu_initialize_context(unsigned int cpu)
++{
++ vcpu_guest_context_t ctxt;
++ struct task_struct *idle = idle_task(cpu);
++#ifdef __x86_64__
++ struct desc_ptr *gdt_descr = &cpu_gdt_descr[cpu];
++#else
++ struct Xgt_desc_struct *gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
++#endif
++
++ if (cpu == 0)
++ return;
++
++ memset(&ctxt, 0, sizeof(ctxt));
++
++ ctxt.flags = VGCF_IN_KERNEL;
++ ctxt.user_regs.ds = __USER_DS;
++ ctxt.user_regs.es = __USER_DS;
++ ctxt.user_regs.fs = 0;
++ ctxt.user_regs.gs = 0;
++ ctxt.user_regs.ss = __KERNEL_DS;
++ ctxt.user_regs.eip = (unsigned long)cpu_bringup_and_idle;
++ ctxt.user_regs.eflags = X86_EFLAGS_IF | 0x1000; /* IOPL_RING1 */
++
++ memset(&ctxt.fpu_ctxt, 0, sizeof(ctxt.fpu_ctxt));
++
++ smp_trap_init(ctxt.trap_ctxt);
++
++ ctxt.ldt_ents = 0;
++
++ ctxt.gdt_frames[0] = virt_to_mfn(gdt_descr->address);
++ ctxt.gdt_ents = gdt_descr->size / 8;
++
++#ifdef __i386__
++ ctxt.user_regs.cs = __KERNEL_CS;
++ ctxt.user_regs.esp = idle->thread.esp0 - sizeof(struct pt_regs);
++
++ ctxt.kernel_ss = __KERNEL_DS;
++ ctxt.kernel_sp = idle->thread.esp0;
++
++ ctxt.event_callback_cs = __KERNEL_CS;
++ ctxt.event_callback_eip = (unsigned long)hypervisor_callback;
++ ctxt.failsafe_callback_cs = __KERNEL_CS;
++ ctxt.failsafe_callback_eip = (unsigned long)failsafe_callback;
++
++ ctxt.ctrlreg[3] = xen_pfn_to_cr3(virt_to_mfn(swapper_pg_dir));
++#else /* __x86_64__ */
++ ctxt.user_regs.cs = __KERNEL_CS;
++ ctxt.user_regs.esp = idle->thread.rsp0 - sizeof(struct pt_regs);
++
++ ctxt.kernel_ss = __KERNEL_DS;
++ ctxt.kernel_sp = idle->thread.rsp0;
++
++ ctxt.event_callback_eip = (unsigned long)hypervisor_callback;
++ ctxt.failsafe_callback_eip = (unsigned long)failsafe_callback;
++ ctxt.syscall_callback_eip = (unsigned long)system_call;
++
++ ctxt.ctrlreg[3] = xen_pfn_to_cr3(virt_to_mfn(init_level4_pgt));
++
++ ctxt.gs_base_kernel = (unsigned long)(cpu_pda(cpu));
++#endif
++
++ BUG_ON(HYPERVISOR_vcpu_op(VCPUOP_initialise, cpu, &ctxt));
++}
++
++void __init smp_prepare_cpus(unsigned int max_cpus)
++{
++ int cpu;
++ struct task_struct *idle;
++#ifdef __x86_64__
++ struct desc_ptr *gdt_descr;
++#else
++ struct Xgt_desc_struct *gdt_descr;
++#endif
++
++ boot_cpu_data.apicid = 0;
++ cpu_data[0] = boot_cpu_data;
++
++ cpu_2_logical_apicid[0] = 0;
++ x86_cpu_to_apicid[0] = 0;
++
++ current_thread_info()->cpu = 0;
++
++ for (cpu = 0; cpu < NR_CPUS; cpu++) {
++ cpus_clear(cpu_sibling_map[cpu]);
++ cpus_clear(cpu_core_map[cpu]);
++ }
++
++ set_cpu_sibling_map(0);
++
++ xen_smp_intr_init(0);
++
++ for_each_cpu_mask (cpu, cpu_possible_map) {
++ if (cpu == 0)
++ continue;
++
++#ifdef __x86_64__
++ gdt_descr = &cpu_gdt_descr[cpu];
++#else
++ gdt_descr = &per_cpu(cpu_gdt_descr, cpu);
++#endif
++ gdt_descr->address = get_zeroed_page(GFP_KERNEL);
++ if (unlikely(!gdt_descr->address)) {
++ printk(KERN_CRIT "CPU%d failed to allocate GDT\n", cpu);
++ continue;
++ }
++ gdt_descr->size = GDT_SIZE;
++ memcpy((void *)gdt_descr->address, cpu_gdt_table, GDT_SIZE);
++ make_page_readonly(
++ (void *)gdt_descr->address,
++ XENFEAT_writable_descriptor_tables);
++
++ cpu_data[cpu] = boot_cpu_data;
++ cpu_data[cpu].apicid = cpu;
++
++ cpu_2_logical_apicid[cpu] = cpu;
++ x86_cpu_to_apicid[cpu] = cpu;
++
++ idle = fork_idle(cpu);
++ if (IS_ERR(idle))
++ panic("failed fork for CPU %d", cpu);
++
++#ifdef __x86_64__
++ cpu_pda(cpu)->pcurrent = idle;
++ cpu_pda(cpu)->cpunumber = cpu;
++ clear_ti_thread_flag(idle->thread_info, TIF_FORK);
++#endif
++
++ irq_ctx_init(cpu);
++
++#ifdef CONFIG_HOTPLUG_CPU
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ cpu_set(cpu, cpu_present_map);
++#else
++ cpu_set(cpu, cpu_present_map);
++#endif
++
++ cpu_initialize_context(cpu);
++ }
++
++ init_xenbus_allowed_cpumask();
++
++ /* Currently, Xen gives no dynamic NUMA/HT info. */
++ for (cpu = 1; cpu < NR_CPUS; cpu++) {
++ cpu_sibling_map[cpu] = cpumask_of_cpu(cpu);
++ cpu_core_map[cpu] = cpumask_of_cpu(cpu);
++ }
++
++#ifdef CONFIG_X86_IO_APIC
++ /*
++ * Here we can be sure that there is an IO-APIC in the system. Let's
++ * go and set it up:
++ */
++ if (!skip_ioapic_setup && nr_ioapics)
++ setup_IO_APIC();
++#endif
++}
++
++void __devinit smp_prepare_boot_cpu(void)
++{
++}
++
++#ifdef CONFIG_HOTPLUG_CPU
++
++/*
++ * Initialize cpu_present_map late to skip SMP boot code in init/main.c.
++ * But do it early enough to catch critical for_each_present_cpu() loops
++ * in i386-specific code.
++ */
++static int __init initialize_cpu_present_map(void)
++{
++ cpu_present_map = cpu_possible_map;
++ return 0;
++}
++core_initcall(initialize_cpu_present_map);
++
++static void
++remove_siblinginfo(int cpu)
++{
++ phys_proc_id[cpu] = BAD_APICID;
++ cpu_core_id[cpu] = BAD_APICID;
++
++ cpus_clear(cpu_sibling_map[cpu]);
++ cpus_clear(cpu_core_map[cpu]);
++
++ cpu_data[cpu].booted_cores = 0;
++}
++
++int __cpu_disable(void)
++{
++ cpumask_t map = cpu_online_map;
++ int cpu = smp_processor_id();
++
++ if (cpu == 0)
++ return -EBUSY;
++
++ remove_siblinginfo(cpu);
++
++ cpu_clear(cpu, map);
++ fixup_irqs(map);
++ cpu_clear(cpu, cpu_online_map);
++
++ return 0;
++}
++
++void __cpu_die(unsigned int cpu)
++{
++ while (HYPERVISOR_vcpu_op(VCPUOP_is_up, cpu, NULL)) {
++ current->state = TASK_UNINTERRUPTIBLE;
++ schedule_timeout(HZ/10);
++ }
++
++ xen_smp_intr_exit(cpu);
++
++#ifdef __i386__
++ if (num_online_cpus() == 1)
++ alternatives_smp_switch(0);
++#endif
++}
++
++#else /* !CONFIG_HOTPLUG_CPU */
++
++int __cpu_disable(void)
++{
++ return -ENOSYS;
++}
++
++void __cpu_die(unsigned int cpu)
++{
++ BUG();
++}
++
++#endif /* CONFIG_HOTPLUG_CPU */
++
++int __devinit __cpu_up(unsigned int cpu)
++{
++ int rc;
++
++ rc = cpu_up_check(cpu);
++ if (rc)
++ return rc;
++
++#ifdef __i386__
++ if (num_online_cpus() == 1)
++ alternatives_smp_switch(1);
++#endif
++
++ /* This must be done before setting cpu_online_map */
++ set_cpu_sibling_map(cpu);
++ wmb();
++
++ xen_smp_intr_init(cpu);
++ cpu_set(cpu, cpu_online_map);
++
++ rc = HYPERVISOR_vcpu_op(VCPUOP_up, cpu, NULL);
++ BUG_ON(rc);
++
++ return 0;
++}
++
++void __init smp_cpus_done(unsigned int max_cpus)
++{
++}
++
++#ifndef CONFIG_X86_LOCAL_APIC
++int setup_profiling_timer(unsigned int multiplier)
++{
++ return -EINVAL;
++}
++#endif
+diff -urNp linux-2.6/drivers/xen/core/xen_proc.c new/drivers/xen/core/xen_proc.c
+--- linux-2.6/drivers/xen/core/xen_proc.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/xen_proc.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,19 @@
++
++#include <linux/config.h>
++#include <linux/proc_fs.h>
++#include <xen/xen_proc.h>
++
++static struct proc_dir_entry *xen_base;
++
++struct proc_dir_entry *create_xen_proc_entry(const char *name, mode_t mode)
++{
++ if ( xen_base == NULL )
++ if ( (xen_base = proc_mkdir("xen", &proc_root)) == NULL )
++ panic("Couldn't create /proc/xen");
++ return create_proc_entry(name, mode, xen_base);
++}
++
++void remove_xen_proc_entry(const char *name)
++{
++ remove_proc_entry(name, xen_base);
++}
+diff -urNp linux-2.6/drivers/xen/core/xen_sysfs.c new/drivers/xen/core/xen_sysfs.c
+--- linux-2.6/drivers/xen/core/xen_sysfs.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/core/xen_sysfs.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,379 @@
++/*
++ * copyright (c) 2006 IBM Corporation
++ * Authored by: Mike D. Day <ncmike@us.ibm.com>
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License version 2 as
++ * published by the Free Software Foundation.
++ */
++
++#include <linux/config.h>
++#include <linux/err.h>
++#include <linux/kernel.h>
++#include <linux/module.h>
++#include <linux/init.h>
++#include <asm/hypervisor.h>
++#include <xen/features.h>
++#include <xen/hypervisor_sysfs.h>
++#include <xen/xenbus.h>
++
++MODULE_LICENSE("GPL");
++MODULE_AUTHOR("Mike D. Day <ncmike@us.ibm.com>");
++
++static ssize_t type_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ return sprintf(buffer, "xen\n");
++}
++
++HYPERVISOR_ATTR_RO(type);
++
++static int __init xen_sysfs_type_init(void)
++{
++ return sysfs_create_file(&hypervisor_subsys.kset.kobj, &type_attr.attr);
++}
++
++static void xen_sysfs_type_destroy(void)
++{
++ sysfs_remove_file(&hypervisor_subsys.kset.kobj, &type_attr.attr);
++}
++
++/* xen version attributes */
++static ssize_t major_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int version = HYPERVISOR_xen_version(XENVER_version, NULL);
++ if (version)
++ return sprintf(buffer, "%d\n", version >> 16);
++ return -ENODEV;
++}
++
++HYPERVISOR_ATTR_RO(major);
++
++static ssize_t minor_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int version = HYPERVISOR_xen_version(XENVER_version, NULL);
++ if (version)
++ return sprintf(buffer, "%d\n", version & 0xff);
++ return -ENODEV;
++}
++
++HYPERVISOR_ATTR_RO(minor);
++
++static ssize_t extra_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ char *extra;
++
++ extra = kmalloc(XEN_EXTRAVERSION_LEN, GFP_KERNEL);
++ if (extra) {
++ ret = HYPERVISOR_xen_version(XENVER_extraversion, extra);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", extra);
++ kfree(extra);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(extra);
++
++static struct attribute *version_attrs[] = {
++ &major_attr.attr,
++ &minor_attr.attr,
++ &extra_attr.attr,
++ NULL
++};
++
++static struct attribute_group version_group = {
++ .name = "version",
++ .attrs = version_attrs,
++};
++
++static int __init xen_sysfs_version_init(void)
++{
++ return sysfs_create_group(&hypervisor_subsys.kset.kobj,
++ &version_group);
++}
++
++static void xen_sysfs_version_destroy(void)
++{
++ sysfs_remove_group(&hypervisor_subsys.kset.kobj, &version_group);
++}
++
++/* UUID */
++
++static ssize_t uuid_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ char *vm, *val;
++ int ret;
++
++ vm = xenbus_read(XBT_NIL, "vm", "", NULL);
++ if (IS_ERR(vm))
++ return PTR_ERR(vm);
++ val = xenbus_read(XBT_NIL, vm, "uuid", NULL);
++ kfree(vm);
++ if (IS_ERR(val))
++ return PTR_ERR(val);
++ ret = sprintf(buffer, "%s\n", val);
++ kfree(val);
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(uuid);
++
++static int __init xen_sysfs_uuid_init(void)
++{
++ return sysfs_create_file(&hypervisor_subsys.kset.kobj, &uuid_attr.attr);
++}
++
++static void xen_sysfs_uuid_destroy(void)
++{
++ sysfs_remove_file(&hypervisor_subsys.kset.kobj, &uuid_attr.attr);
++}
++
++/* xen compilation attributes */
++
++static ssize_t compiler_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ struct xen_compile_info *info;
++
++ info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
++ if (info) {
++ ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", info->compiler);
++ kfree(info);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(compiler);
++
++static ssize_t compiled_by_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ struct xen_compile_info *info;
++
++ info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
++ if (info) {
++ ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", info->compile_by);
++ kfree(info);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(compiled_by);
++
++static ssize_t compile_date_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ struct xen_compile_info *info;
++
++ info = kmalloc(sizeof(struct xen_compile_info), GFP_KERNEL);
++ if (info) {
++ ret = HYPERVISOR_xen_version(XENVER_compile_info, info);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", info->compile_date);
++ kfree(info);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(compile_date);
++
++static struct attribute *xen_compile_attrs[] = {
++ &compiler_attr.attr,
++ &compiled_by_attr.attr,
++ &compile_date_attr.attr,
++ NULL
++};
++
++static struct attribute_group xen_compilation_group = {
++ .name = "compilation",
++ .attrs = xen_compile_attrs,
++};
++
++int __init static xen_compilation_init(void)
++{
++ return sysfs_create_group(&hypervisor_subsys.kset.kobj,
++ &xen_compilation_group);
++}
++
++static void xen_compilation_destroy(void)
++{
++ sysfs_remove_group(&hypervisor_subsys.kset.kobj,
++ &xen_compilation_group);
++}
++
++/* xen properties info */
++
++static ssize_t capabilities_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ char *caps;
++
++ caps = kmalloc(XEN_CAPABILITIES_INFO_LEN, GFP_KERNEL);
++ if (caps) {
++ ret = HYPERVISOR_xen_version(XENVER_capabilities, caps);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", caps);
++ kfree(caps);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(capabilities);
++
++static ssize_t changeset_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ char *cset;
++
++ cset = kmalloc(XEN_CHANGESET_INFO_LEN, GFP_KERNEL);
++ if (cset) {
++ ret = HYPERVISOR_xen_version(XENVER_changeset, cset);
++ if (!ret)
++ ret = sprintf(buffer, "%s\n", cset);
++ kfree(cset);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(changeset);
++
++static ssize_t virtual_start_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret = -ENOMEM;
++ struct xen_platform_parameters *parms;
++
++ parms = kmalloc(sizeof(struct xen_platform_parameters), GFP_KERNEL);
++ if (parms) {
++ ret = HYPERVISOR_xen_version(XENVER_platform_parameters,
++ parms);
++ if (!ret)
++ ret = sprintf(buffer, "%lx\n", parms->virt_start);
++ kfree(parms);
++ }
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(virtual_start);
++
++static ssize_t pagesize_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ int ret;
++
++ ret = HYPERVISOR_xen_version(XENVER_pagesize, NULL);
++ if (ret > 0)
++ ret = sprintf(buffer, "%x\n", ret);
++
++ return ret;
++}
++
++HYPERVISOR_ATTR_RO(pagesize);
++
++/* eventually there will be several more features to export */
++static ssize_t xen_feature_show(int index, char *buffer)
++{
++ int ret = -ENOMEM;
++ struct xen_feature_info *info;
++
++ info = kmalloc(sizeof(struct xen_feature_info), GFP_KERNEL);
++ if (info) {
++ info->submap_idx = index;
++ ret = HYPERVISOR_xen_version(XENVER_get_features, info);
++ if (!ret)
++ ret = sprintf(buffer, "%d\n", info->submap);
++ kfree(info);
++ }
++
++ return ret;
++}
++
++static ssize_t writable_pt_show(struct hyp_sysfs_attr *attr, char *buffer)
++{
++ return xen_feature_show(XENFEAT_writable_page_tables, buffer);
++}
++
++HYPERVISOR_ATTR_RO(writable_pt);
++
++static struct attribute *xen_properties_attrs[] = {
++ &capabilities_attr.attr,
++ &changeset_attr.attr,
++ &virtual_start_attr.attr,
++ &pagesize_attr.attr,
++ &writable_pt_attr.attr,
++ NULL
++};
++
++static struct attribute_group xen_properties_group = {
++ .name = "properties",
++ .attrs = xen_properties_attrs,
++};
++
++static int __init xen_properties_init(void)
++{
++ return sysfs_create_group(&hypervisor_subsys.kset.kobj,
++ &xen_properties_group);
++}
++
++static void xen_properties_destroy(void)
++{
++ sysfs_remove_group(&hypervisor_subsys.kset.kobj,
++ &xen_properties_group);
++}
++
++static int __init hyper_sysfs_init(void)
++{
++ int ret;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ ret = xen_sysfs_type_init();
++ if (ret)
++ goto out;
++ ret = xen_sysfs_version_init();
++ if (ret)
++ goto version_out;
++ ret = xen_compilation_init();
++ if (ret)
++ goto comp_out;
++ ret = xen_sysfs_uuid_init();
++ if (ret)
++ goto uuid_out;
++ ret = xen_properties_init();
++ if (!ret)
++ goto out;
++
++ xen_sysfs_uuid_destroy();
++uuid_out:
++ xen_compilation_destroy();
++comp_out:
++ xen_sysfs_version_destroy();
++version_out:
++ xen_sysfs_type_destroy();
++out:
++ return ret;
++}
++
++static void hyper_sysfs_exit(void)
++{
++ xen_properties_destroy();
++ xen_compilation_destroy();
++ xen_sysfs_uuid_destroy();
++ xen_sysfs_version_destroy();
++ xen_sysfs_type_destroy();
++
++}
++
++module_init(hyper_sysfs_init);
++module_exit(hyper_sysfs_exit);
+diff -urNp linux-2.6/drivers/xen/evtchn/evtchn.c new/drivers/xen/evtchn/evtchn.c
+--- linux-2.6/drivers/xen/evtchn/evtchn.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/evtchn/evtchn.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,458 @@
++/******************************************************************************
++ * evtchn.c
++ *
++ * Driver for receiving and demuxing event-channel signals.
++ *
++ * Copyright (c) 2004-2005, K A Fraser
++ * Multi-process extensions Copyright (c) 2004, Steven Smith
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/fs.h>
++#include <linux/errno.h>
++#include <linux/miscdevice.h>
++#include <linux/major.h>
++#include <linux/proc_fs.h>
++#include <linux/stat.h>
++#include <linux/poll.h>
++#include <linux/irq.h>
++#include <linux/init.h>
++#include <linux/gfp.h>
++#include <xen/evtchn.h>
++#include <xen/public/evtchn.h>
++
++struct per_user_data {
++ /* Notification ring, accessed via /dev/xen/evtchn. */
++#define EVTCHN_RING_SIZE (PAGE_SIZE / sizeof(evtchn_port_t))
++#define EVTCHN_RING_MASK(_i) ((_i)&(EVTCHN_RING_SIZE-1))
++ evtchn_port_t *ring;
++ unsigned int ring_cons, ring_prod, ring_overflow;
++
++ /* Processes wait on this queue when ring is empty. */
++ wait_queue_head_t evtchn_wait;
++ struct fasync_struct *evtchn_async_queue;
++};
++
++/* Who's bound to each port? */
++static struct per_user_data *port_user[NR_EVENT_CHANNELS];
++static spinlock_t port_user_lock;
++
++void evtchn_device_upcall(int port)
++{
++ struct per_user_data *u;
++
++ spin_lock(&port_user_lock);
++
++ mask_evtchn(port);
++ clear_evtchn(port);
++
++ if ((u = port_user[port]) != NULL) {
++ if ((u->ring_prod - u->ring_cons) < EVTCHN_RING_SIZE) {
++ u->ring[EVTCHN_RING_MASK(u->ring_prod)] = port;
++ if (u->ring_cons == u->ring_prod++) {
++ wake_up_interruptible(&u->evtchn_wait);
++ kill_fasync(&u->evtchn_async_queue,
++ SIGIO, POLL_IN);
++ }
++ } else {
++ u->ring_overflow = 1;
++ }
++ }
++
++ spin_unlock(&port_user_lock);
++}
++
++static ssize_t evtchn_read(struct file *file, char __user *buf,
++ size_t count, loff_t *ppos)
++{
++ int rc;
++ unsigned int c, p, bytes1 = 0, bytes2 = 0;
++ struct per_user_data *u = file->private_data;
++
++ /* Whole number of ports. */
++ count &= ~(sizeof(evtchn_port_t)-1);
++
++ if (count == 0)
++ return 0;
++
++ if (count > PAGE_SIZE)
++ count = PAGE_SIZE;
++
++ for (;;) {
++ if (u->ring_overflow)
++ return -EFBIG;
++
++ if ((c = u->ring_cons) != (p = u->ring_prod))
++ break;
++
++ if (file->f_flags & O_NONBLOCK)
++ return -EAGAIN;
++
++ rc = wait_event_interruptible(
++ u->evtchn_wait, u->ring_cons != u->ring_prod);
++ if (rc)
++ return rc;
++ }
++
++ /* Byte lengths of two chunks. Chunk split (if any) is at ring wrap. */
++ if (((c ^ p) & EVTCHN_RING_SIZE) != 0) {
++ bytes1 = (EVTCHN_RING_SIZE - EVTCHN_RING_MASK(c)) *
++ sizeof(evtchn_port_t);
++ bytes2 = EVTCHN_RING_MASK(p) * sizeof(evtchn_port_t);
++ } else {
++ bytes1 = (p - c) * sizeof(evtchn_port_t);
++ bytes2 = 0;
++ }
++
++ /* Truncate chunks according to caller's maximum byte count. */
++ if (bytes1 > count) {
++ bytes1 = count;
++ bytes2 = 0;
++ } else if ((bytes1 + bytes2) > count) {
++ bytes2 = count - bytes1;
++ }
++
++ if (copy_to_user(buf, &u->ring[EVTCHN_RING_MASK(c)], bytes1) ||
++ ((bytes2 != 0) &&
++ copy_to_user(&buf[bytes1], &u->ring[0], bytes2)))
++ return -EFAULT;
++
++ u->ring_cons += (bytes1 + bytes2) / sizeof(evtchn_port_t);
++
++ return bytes1 + bytes2;
++}
++
++static ssize_t evtchn_write(struct file *file, const char __user *buf,
++ size_t count, loff_t *ppos)
++{
++ int rc, i;
++ evtchn_port_t *kbuf = (evtchn_port_t *)__get_free_page(GFP_KERNEL);
++ struct per_user_data *u = file->private_data;
++
++ if (kbuf == NULL)
++ return -ENOMEM;
++
++ /* Whole number of ports. */
++ count &= ~(sizeof(evtchn_port_t)-1);
++
++ if (count == 0) {
++ rc = 0;
++ goto out;
++ }
++
++ if (count > PAGE_SIZE)
++ count = PAGE_SIZE;
++
++ if (copy_from_user(kbuf, buf, count) != 0) {
++ rc = -EFAULT;
++ goto out;
++ }
++
++ spin_lock_irq(&port_user_lock);
++ for (i = 0; i < (count/sizeof(evtchn_port_t)); i++)
++ if ((kbuf[i] < NR_EVENT_CHANNELS) && (port_user[kbuf[i]] == u))
++ unmask_evtchn(kbuf[i]);
++ spin_unlock_irq(&port_user_lock);
++
++ rc = count;
++
++ out:
++ free_page((unsigned long)kbuf);
++ return rc;
++}
++
++static void evtchn_bind_to_user(struct per_user_data *u, int port)
++{
++ spin_lock_irq(&port_user_lock);
++ BUG_ON(port_user[port] != NULL);
++ port_user[port] = u;
++ unmask_evtchn(port);
++ spin_unlock_irq(&port_user_lock);
++}
++
++static int evtchn_ioctl(struct inode *inode, struct file *file,
++ unsigned int cmd, unsigned long arg)
++{
++ int rc;
++ struct per_user_data *u = file->private_data;
++ void __user *uarg = (void __user *) arg;
++
++ switch (cmd) {
++ case IOCTL_EVTCHN_BIND_VIRQ: {
++ struct ioctl_evtchn_bind_virq bind;
++ struct evtchn_bind_virq bind_virq;
++
++ rc = -EFAULT;
++ if (copy_from_user(&bind, uarg, sizeof(bind)))
++ break;
++
++ bind_virq.virq = bind.virq;
++ bind_virq.vcpu = 0;
++ rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_virq,
++ &bind_virq);
++ if (rc != 0)
++ break;
++
++ rc = bind_virq.port;
++ evtchn_bind_to_user(u, rc);
++ break;
++ }
++
++ case IOCTL_EVTCHN_BIND_INTERDOMAIN: {
++ struct ioctl_evtchn_bind_interdomain bind;
++ struct evtchn_bind_interdomain bind_interdomain;
++
++ rc = -EFAULT;
++ if (copy_from_user(&bind, uarg, sizeof(bind)))
++ break;
++
++ bind_interdomain.remote_dom = bind.remote_domain;
++ bind_interdomain.remote_port = bind.remote_port;
++ rc = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
++ &bind_interdomain);
++ if (rc != 0)
++ break;
++
++ rc = bind_interdomain.local_port;
++ evtchn_bind_to_user(u, rc);
++ break;
++ }
++
++ case IOCTL_EVTCHN_BIND_UNBOUND_PORT: {
++ struct ioctl_evtchn_bind_unbound_port bind;
++ struct evtchn_alloc_unbound alloc_unbound;
++
++ rc = -EFAULT;
++ if (copy_from_user(&bind, uarg, sizeof(bind)))
++ break;
++
++ alloc_unbound.dom = DOMID_SELF;
++ alloc_unbound.remote_dom = bind.remote_domain;
++ rc = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound,
++ &alloc_unbound);
++ if (rc != 0)
++ break;
++
++ rc = alloc_unbound.port;
++ evtchn_bind_to_user(u, rc);
++ break;
++ }
++
++ case IOCTL_EVTCHN_UNBIND: {
++ struct ioctl_evtchn_unbind unbind;
++ struct evtchn_close close;
++ int ret;
++
++ rc = -EFAULT;
++ if (copy_from_user(&unbind, uarg, sizeof(unbind)))
++ break;
++
++ rc = -EINVAL;
++ if (unbind.port >= NR_EVENT_CHANNELS)
++ break;
++
++ spin_lock_irq(&port_user_lock);
++
++ rc = -ENOTCONN;
++ if (port_user[unbind.port] != u) {
++ spin_unlock_irq(&port_user_lock);
++ break;
++ }
++
++ port_user[unbind.port] = NULL;
++ mask_evtchn(unbind.port);
++
++ spin_unlock_irq(&port_user_lock);
++
++ close.port = unbind.port;
++ ret = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close);
++ BUG_ON(ret);
++
++ rc = 0;
++ break;
++ }
++
++ case IOCTL_EVTCHN_NOTIFY: {
++ struct ioctl_evtchn_notify notify;
++
++ rc = -EFAULT;
++ if (copy_from_user(¬ify, uarg, sizeof(notify)))
++ break;
++
++ if (notify.port >= NR_EVENT_CHANNELS) {
++ rc = -EINVAL;
++ } else if (port_user[notify.port] != u) {
++ rc = -ENOTCONN;
++ } else {
++ notify_remote_via_evtchn(notify.port);
++ rc = 0;
++ }
++ break;
++ }
++
++ case IOCTL_EVTCHN_RESET: {
++ /* Initialise the ring to empty. Clear errors. */
++ spin_lock_irq(&port_user_lock);
++ u->ring_cons = u->ring_prod = u->ring_overflow = 0;
++ spin_unlock_irq(&port_user_lock);
++ rc = 0;
++ break;
++ }
++
++ default:
++ rc = -ENOSYS;
++ break;
++ }
++
++ return rc;
++}
++
++static unsigned int evtchn_poll(struct file *file, poll_table *wait)
++{
++ unsigned int mask = POLLOUT | POLLWRNORM;
++ struct per_user_data *u = file->private_data;
++
++ poll_wait(file, &u->evtchn_wait, wait);
++ if (u->ring_cons != u->ring_prod)
++ mask |= POLLIN | POLLRDNORM;
++ if (u->ring_overflow)
++ mask = POLLERR;
++ return mask;
++}
++
++static int evtchn_fasync(int fd, struct file *filp, int on)
++{
++ struct per_user_data *u = filp->private_data;
++ return fasync_helper(fd, filp, on, &u->evtchn_async_queue);
++}
++
++static int evtchn_open(struct inode *inode, struct file *filp)
++{
++ struct per_user_data *u;
++
++ if ((u = kmalloc(sizeof(*u), GFP_KERNEL)) == NULL)
++ return -ENOMEM;
++
++ memset(u, 0, sizeof(*u));
++ init_waitqueue_head(&u->evtchn_wait);
++
++ u->ring = (evtchn_port_t *)__get_free_page(GFP_KERNEL);
++ if (u->ring == NULL) {
++ kfree(u);
++ return -ENOMEM;
++ }
++
++ filp->private_data = u;
++
++ return 0;
++}
++
++static int evtchn_release(struct inode *inode, struct file *filp)
++{
++ int i;
++ struct per_user_data *u = filp->private_data;
++ struct evtchn_close close;
++
++ spin_lock_irq(&port_user_lock);
++
++ free_page((unsigned long)u->ring);
++
++ for (i = 0; i < NR_EVENT_CHANNELS; i++) {
++ int ret;
++ if (port_user[i] != u)
++ continue;
++
++ port_user[i] = NULL;
++ mask_evtchn(i);
++
++ close.port = i;
++ ret = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close);
++ BUG_ON(ret);
++ }
++
++ spin_unlock_irq(&port_user_lock);
++
++ kfree(u);
++
++ return 0;
++}
++
++static struct file_operations evtchn_fops = {
++ .owner = THIS_MODULE,
++ .read = evtchn_read,
++ .write = evtchn_write,
++ .ioctl = evtchn_ioctl,
++ .poll = evtchn_poll,
++ .fasync = evtchn_fasync,
++ .open = evtchn_open,
++ .release = evtchn_release,
++};
++
++static struct miscdevice evtchn_miscdev = {
++ .minor = EVTCHN_MINOR,
++ .name = "evtchn",
++ .fops = &evtchn_fops,
++ .devfs_name = "misc/evtchn",
++};
++
++static int __init evtchn_init(void)
++{
++ int err;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ spin_lock_init(&port_user_lock);
++ memset(port_user, 0, sizeof(port_user));
++
++ /* Create '/dev/misc/evtchn'. */
++ err = misc_register(&evtchn_miscdev);
++ if (err != 0) {
++ printk(KERN_ALERT "Could not register /dev/misc/evtchn\n");
++ return err;
++ }
++
++ printk("Event-channel device installed.\n");
++
++ return 0;
++}
++
++static void evtchn_cleanup(void)
++{
++ misc_deregister(&evtchn_miscdev);
++}
++
++module_init(evtchn_init);
++module_exit(evtchn_cleanup);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/evtchn/Makefile new/drivers/xen/evtchn/Makefile
+--- linux-2.6/drivers/xen/evtchn/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/evtchn/Makefile 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,2 @@
++
++obj-y := evtchn.o
+diff -urNp linux-2.6/drivers/xen/Kconfig new/drivers/xen/Kconfig
+--- linux-2.6/drivers/xen/Kconfig 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/Kconfig 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,248 @@
++#
++# This Kconfig describe xen options
++#
++
++mainmenu "Xen Configuration"
++
++config XEN
++ bool
++ default y if X86_XEN || X86_64_XEN
++ help
++ This is the Linux Xen port.
++
++if XEN
++config XEN_INTERFACE_VERSION
++ hex
++ default 0x00030202
++
++menu "XEN"
++
++config XEN_PRIVILEGED_GUEST
++ bool "Privileged Guest (domain 0)"
++ depends XEN
++ default n
++ help
++ Support for privileged operation (domain 0)
++
++config XEN_UNPRIVILEGED_GUEST
++ bool
++ default !XEN_PRIVILEGED_GUEST
++
++config XEN_PRIVCMD
++ bool
++ depends on PROC_FS
++ default y
++
++config XEN_BACKEND
++ tristate "Backend driver support"
++ default y
++ help
++ Support for backend device drivers that provide I/O services
++ to other virtual machines.
++
++config XEN_PCIDEV_BACKEND
++ tristate "PCI device backend driver"
++ depends on PCI && XEN_BACKEND
++ default XEN_PRIVILEGED_GUEST
++ help
++ The PCI device backend driver allows the kernel to export arbitrary
++ PCI devices to other guests. If you select this to be a module, you
++ will need to make sure no other driver has bound to the device(s)
++ you want to make visible to other guests.
++
++choice
++ prompt "PCI Backend Mode"
++ depends on XEN_PCIDEV_BACKEND
++ default XEN_PCIDEV_BACKEND_VPCI
++
++config XEN_PCIDEV_BACKEND_VPCI
++ bool "Virtual PCI"
++ ---help---
++ This PCI Backend hides the true PCI topology and makes the frontend
++ think there is a single PCI bus with only the exported devices on it.
++ For example, a device at 03:05.0 will be re-assigned to 00:00.0. A
++ second device at 02:1a.0 will be re-assigned to 00:01.0.
++
++config XEN_PCIDEV_BACKEND_PASS
++ bool "Passthrough"
++ ---help---
++ This PCI Backend provides a real view of the PCI topology to the
++ frontend (for example, a device at 06:01.b will still appear at
++ 06:01.b to the frontend). This is similar to how Xen 2.0.x exposed
++ PCI devices to its driver domains. This may be required for drivers
++ which depend on finding their hardward in certain bus/slot
++ locations.
++
++endchoice
++
++config XEN_PCIDEV_BE_DEBUG
++ bool "PCI Backend Debugging"
++ depends on XEN_PCIDEV_BACKEND
++ default n
++
++config XEN_BLKDEV_BACKEND
++ tristate "Block-device backend driver"
++ depends on XEN_BACKEND
++ default y
++ help
++ The block-device backend driver allows the kernel to export its
++ block devices to other guests via a high-performance shared-memory
++ interface.
++
++config XEN_XENBUS_DEV
++ bool
++ depends on PROC_FS
++ default y
++
++config XEN_NETDEV_BACKEND
++ tristate "Network-device backend driver"
++ depends on XEN_BACKEND && NET
++ default y
++ help
++ The network-device backend driver allows the kernel to export its
++ network devices to other guests via a high-performance shared-memory
++ interface.
++
++config XEN_NETDEV_PIPELINED_TRANSMITTER
++ bool "Pipelined transmitter (DANGEROUS)"
++ depends on XEN_NETDEV_BACKEND
++ default n
++ help
++ If the net backend is a dumb domain, such as a transparent Ethernet
++ bridge with no local IP interface, it is safe to say Y here to get
++ slightly lower network overhead.
++ If the backend has a local IP interface; or may be doing smart things
++ like reassembling packets to perform firewall filtering; or if you
++ are unsure; or if you experience network hangs when this option is
++ enabled; then you must say N here.
++
++config XEN_NETDEV_LOOPBACK
++ tristate "Network-device loopback driver"
++ depends on XEN_NETDEV_BACKEND
++ default y
++ help
++ A two-interface loopback device to emulate a local netfront-netback
++ connection.
++
++config XEN_TPMDEV_BACKEND
++ tristate "TPM-device backend driver"
++ depends on XEN_BACKEND
++ default n
++ help
++ The TPM-device backend driver
++
++config XEN_TPMDEV_CLOSE_IF_VTPM_FAILS
++ bool "TPM backend closes upon vTPM failure"
++ depends on XEN_TPMDEV_BACKEND
++ default n
++ help
++ The TPM backend closes the channel if the vTPM in userspace indicates
++ a failure. The corresponding domain's channel will be closed.
++ Say Y if you want this feature.
++
++config XEN_BLKDEV_FRONTEND
++ tristate "Block-device frontend driver"
++ depends on XEN
++ default y
++ help
++ The block-device frontend driver allows the kernel to access block
++ devices mounted within another guest OS. Unless you are building a
++ dedicated device-driver domain, or your master control domain
++ (domain 0), then you almost certainly want to say Y here.
++
++config XEN_NETDEV_FRONTEND
++ tristate "Network-device frontend driver"
++ depends on XEN && NET
++ default y
++ help
++ The network-device frontend driver allows the kernel to access
++ network interfaces within another guest OS. Unless you are building a
++ dedicated device-driver domain, or your master control domain
++ (domain 0), then you almost certainly want to say Y here.
++
++config XEN_SCRUB_PAGES
++ bool "Scrub memory before freeing it to Xen"
++ default y
++ help
++ Erase memory contents before freeing it back to Xen's global
++ pool. This ensures that any secrets contained within that
++ memory (e.g., private keys) cannot be found by other guests that
++ may be running on the machine. Most people will want to say Y here.
++ If security is not a concern then you may increase performance by
++ saying N.
++
++config XEN_DISABLE_SERIAL
++ bool "Disable serial port drivers"
++ default y
++ help
++ Disable serial port drivers, allowing the Xen console driver
++ to provide a serial console at ttyS0.
++
++config XEN_SYSFS
++ tristate "Export Xen attributes in sysfs"
++ depends on SYSFS
++ default y
++ help
++ Xen hypervisor attributes will show up under /sys/hypervisor/.
++
++choice
++ prompt "Xen version compatibility"
++ default XEN_COMPAT_030002_AND_LATER
++
++ config XEN_COMPAT_030002_AND_LATER
++ bool "3.0.2 and later"
++
++ config XEN_COMPAT_LATEST_ONLY
++ bool "no compatibility code"
++
++endchoice
++
++config XEN_COMPAT_030002
++ bool
++ default XEN_COMPAT_030002_AND_LATER
++
++endmenu
++
++config HAVE_ARCH_ALLOC_SKB
++ bool
++ default y
++
++config HAVE_ARCH_DEV_ALLOC_SKB
++ bool
++ default y
++
++config HAVE_IRQ_IGNORE_UNHANDLED
++ bool
++ default y
++
++config NO_IDLE_HZ
++ bool
++ default y
++
++config XEN_UTIL
++ bool
++ default y
++
++config XEN_BALLOON
++ bool
++ default y
++
++config XEN_DEVMEM
++ bool
++ default y
++
++config XEN_SKBUFF
++ bool
++ default y
++ depends on NET
++
++config XEN_REBOOT
++ bool
++ default y
++
++config XEN_SMPBOOT
++ bool
++ default y
++ depends on SMP
++
++endif
+diff -urNp linux-2.6/drivers/xen/Makefile new/drivers/xen/Makefile
+--- linux-2.6/drivers/xen/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/Makefile 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,16 @@
++obj-y += core/
++obj-y += console/
++obj-y += evtchn/
++obj-y += privcmd/
++obj-y += xenbus/
++
++obj-$(CONFIG_XEN_UTIL) += util.o
++obj-$(CONFIG_XEN_BALLOON) += balloon/
++obj-$(CONFIG_XEN_DEVMEM) += char/
++obj-$(CONFIG_XEN_BLKDEV_BACKEND) += blkback/
++obj-$(CONFIG_XEN_NETDEV_BACKEND) += netback/
++obj-$(CONFIG_XEN_TPMDEV_BACKEND) += tpmback/
++obj-$(CONFIG_XEN_BLKDEV_FRONTEND) += blkfront/
++obj-$(CONFIG_XEN_NETDEV_FRONTEND) += netfront/
++obj-$(CONFIG_XEN_PCIDEV_BACKEND) += pciback/
++obj-$(CONFIG_XEN_PCIDEV_FRONTEND) += pcifront/
+diff -urNp linux-2.6/drivers/xen/netback/common.h new/drivers/xen/netback/common.h
+--- linux-2.6/drivers/xen/netback/common.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/common.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,124 @@
++/******************************************************************************
++ * arch/xen/drivers/netif/backend/common.h
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __NETIF__BACKEND__COMMON_H__
++#define __NETIF__BACKEND__COMMON_H__
++
++#include <linux/config.h>
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/interrupt.h>
++#include <linux/slab.h>
++#include <linux/ip.h>
++#include <linux/in.h>
++#include <linux/netdevice.h>
++#include <linux/etherdevice.h>
++#include <linux/wait.h>
++#include <xen/evtchn.h>
++#include <xen/interface/io/netif.h>
++#include <asm/io.h>
++#include <asm/pgalloc.h>
++#include <xen/interface/grant_table.h>
++#include <xen/gnttab.h>
++#include <xen/driver_util.h>
++
++#define DPRINTK(_f, _a...) \
++ pr_debug("(file=%s, line=%d) " _f, \
++ __FILE__ , __LINE__ , ## _a )
++#define IPRINTK(fmt, args...) \
++ printk(KERN_INFO "xen_net: " fmt, ##args)
++#define WPRINTK(fmt, args...) \
++ printk(KERN_WARNING "xen_net: " fmt, ##args)
++
++typedef struct netif_st {
++ /* Unique identifier for this interface. */
++ domid_t domid;
++ unsigned int handle;
++
++ u8 fe_dev_addr[6];
++
++ /* Physical parameters of the comms window. */
++ grant_handle_t tx_shmem_handle;
++ grant_ref_t tx_shmem_ref;
++ grant_handle_t rx_shmem_handle;
++ grant_ref_t rx_shmem_ref;
++ unsigned int evtchn;
++ unsigned int irq;
++
++ /* The shared rings and indexes. */
++ netif_tx_back_ring_t tx;
++ netif_rx_back_ring_t rx;
++ struct vm_struct *tx_comms_area;
++ struct vm_struct *rx_comms_area;
++
++ /* Allow netif_be_start_xmit() to peek ahead in the rx request ring. */
++ RING_IDX rx_req_cons_peek;
++
++ /* Transmit shaping: allow 'credit_bytes' every 'credit_usec'. */
++ unsigned long credit_bytes;
++ unsigned long credit_usec;
++ unsigned long remaining_credit;
++ struct timer_list credit_timeout;
++
++ /* Miscellaneous private stuff. */
++ enum { DISCONNECTED, DISCONNECTING, CONNECTED } status;
++ int active;
++ struct list_head list; /* scheduling list */
++ atomic_t refcnt;
++ struct net_device *dev;
++ struct net_device_stats stats;
++
++ wait_queue_head_t waiting_to_free;
++} netif_t;
++
++#define NET_TX_RING_SIZE __RING_SIZE((netif_tx_sring_t *)0, PAGE_SIZE)
++#define NET_RX_RING_SIZE __RING_SIZE((netif_rx_sring_t *)0, PAGE_SIZE)
++
++void netif_disconnect(netif_t *netif);
++
++netif_t *netif_alloc(domid_t domid, unsigned int handle, u8 be_mac[ETH_ALEN]);
++int netif_map(netif_t *netif, unsigned long tx_ring_ref,
++ unsigned long rx_ring_ref, unsigned int evtchn);
++
++#define netif_get(_b) (atomic_inc(&(_b)->refcnt))
++#define netif_put(_b) \
++ do { \
++ if ( atomic_dec_and_test(&(_b)->refcnt) ) \
++ wake_up(&(_b)->waiting_to_free); \
++ } while (0)
++
++void netif_xenbus_init(void);
++
++void netif_schedule_work(netif_t *netif);
++void netif_deschedule_work(netif_t *netif);
++
++int netif_be_start_xmit(struct sk_buff *skb, struct net_device *dev);
++struct net_device_stats *netif_be_get_stats(struct net_device *dev);
++irqreturn_t netif_be_int(int irq, void *dev_id, struct pt_regs *regs);
++
++#endif /* __NETIF__BACKEND__COMMON_H__ */
+diff -urNp linux-2.6/drivers/xen/netback/interface.c new/drivers/xen/netback/interface.c
+--- linux-2.6/drivers/xen/netback/interface.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/interface.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,314 @@
++/******************************************************************************
++ * arch/xen/drivers/netif/backend/interface.c
++ *
++ * Network-device interface management.
++ *
++ * Copyright (c) 2004-2005, Keir Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include "common.h"
++#include <linux/ethtool.h>
++#include <linux/rtnetlink.h>
++
++static void __netif_up(netif_t *netif)
++{
++ struct net_device *dev = netif->dev;
++ netif_tx_lock_bh(dev);
++ netif->active = 1;
++ netif_tx_unlock_bh(dev);
++ enable_irq(netif->irq);
++ netif_schedule_work(netif);
++}
++
++static void __netif_down(netif_t *netif)
++{
++ struct net_device *dev = netif->dev;
++ disable_irq(netif->irq);
++ netif_tx_lock_bh(dev);
++ netif->active = 0;
++ netif_tx_unlock_bh(dev);
++ netif_deschedule_work(netif);
++}
++
++static int net_open(struct net_device *dev)
++{
++ netif_t *netif = netdev_priv(dev);
++ if (netif->status == CONNECTED)
++ __netif_up(netif);
++ netif_start_queue(dev);
++ return 0;
++}
++
++static int net_close(struct net_device *dev)
++{
++ netif_t *netif = netdev_priv(dev);
++ netif_stop_queue(dev);
++ if (netif->status == CONNECTED)
++ __netif_down(netif);
++ return 0;
++}
++
++static struct ethtool_ops network_ethtool_ops =
++{
++ .get_tx_csum = ethtool_op_get_tx_csum,
++ .set_tx_csum = ethtool_op_set_tx_csum,
++};
++
++netif_t *netif_alloc(domid_t domid, unsigned int handle, u8 be_mac[ETH_ALEN])
++{
++ int err = 0, i;
++ struct net_device *dev;
++ netif_t *netif;
++ char name[IFNAMSIZ] = {};
++
++ snprintf(name, IFNAMSIZ - 1, "vif%u.%u", domid, handle);
++ dev = alloc_netdev(sizeof(netif_t), name, ether_setup);
++ if (dev == NULL) {
++ DPRINTK("Could not create netif: out of memory\n");
++ return ERR_PTR(-ENOMEM);
++ }
++
++ netif = netdev_priv(dev);
++ memset(netif, 0, sizeof(*netif));
++ netif->domid = domid;
++ netif->handle = handle;
++ netif->status = DISCONNECTED;
++ atomic_set(&netif->refcnt, 1);
++ init_waitqueue_head(&netif->waiting_to_free);
++ netif->dev = dev;
++
++ netif->credit_bytes = netif->remaining_credit = ~0UL;
++ netif->credit_usec = 0UL;
++ init_timer(&netif->credit_timeout);
++
++ dev->hard_start_xmit = netif_be_start_xmit;
++ dev->get_stats = netif_be_get_stats;
++ dev->open = net_open;
++ dev->stop = net_close;
++ dev->features = NETIF_F_IP_CSUM;
++
++ SET_ETHTOOL_OPS(dev, &network_ethtool_ops);
++
++ /* Disable queuing. */
++ dev->tx_queue_len = 0;
++
++ for (i = 0; i < ETH_ALEN; i++)
++ if (be_mac[i] != 0)
++ break;
++ if (i == ETH_ALEN) {
++ /*
++ * Initialise a dummy MAC address. We choose the numerically
++ * largest non-broadcast address to prevent the address getting
++ * stolen by an Ethernet bridge for STP purposes.
++ * (FE:FF:FF:FF:FF:FF)
++ */
++ memset(dev->dev_addr, 0xFF, ETH_ALEN);
++ dev->dev_addr[0] &= ~0x01;
++ } else
++ memcpy(dev->dev_addr, be_mac, ETH_ALEN);
++
++ rtnl_lock();
++ err = register_netdevice(dev);
++ rtnl_unlock();
++ if (err) {
++ DPRINTK("Could not register new net device %s: err=%d\n",
++ dev->name, err);
++ free_netdev(dev);
++ return ERR_PTR(err);
++ }
++
++ DPRINTK("Successfully created netif\n");
++ return netif;
++}
++
++static int map_frontend_pages(
++ netif_t *netif, grant_ref_t tx_ring_ref, grant_ref_t rx_ring_ref)
++{
++ struct gnttab_map_grant_ref op;
++ int ret;
++
++ gnttab_set_map_op(&op, (unsigned long)netif->tx_comms_area->addr,
++ GNTMAP_host_map, tx_ring_ref, netif->domid);
++
++ lock_vm_area(netif->tx_comms_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1);
++ unlock_vm_area(netif->tx_comms_area);
++ BUG_ON(ret);
++
++ if (op.status) {
++ DPRINTK(" Gnttab failure mapping tx_ring_ref!\n");
++ return op.status;
++ }
++
++ netif->tx_shmem_ref = tx_ring_ref;
++ netif->tx_shmem_handle = op.handle;
++
++ gnttab_set_map_op(&op, (unsigned long)netif->rx_comms_area->addr,
++ GNTMAP_host_map, rx_ring_ref, netif->domid);
++
++ lock_vm_area(netif->rx_comms_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1);
++ unlock_vm_area(netif->rx_comms_area);
++ BUG_ON(ret);
++
++ if (op.status) {
++ DPRINTK(" Gnttab failure mapping rx_ring_ref!\n");
++ return op.status;
++ }
++
++ netif->rx_shmem_ref = rx_ring_ref;
++ netif->rx_shmem_handle = op.handle;
++
++ return 0;
++}
++
++static void unmap_frontend_pages(netif_t *netif)
++{
++ struct gnttab_unmap_grant_ref op;
++ int ret;
++
++ gnttab_set_unmap_op(&op, (unsigned long)netif->tx_comms_area->addr,
++ GNTMAP_host_map, netif->tx_shmem_handle);
++
++ lock_vm_area(netif->tx_comms_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1);
++ unlock_vm_area(netif->tx_comms_area);
++ BUG_ON(ret);
++
++ gnttab_set_unmap_op(&op, (unsigned long)netif->rx_comms_area->addr,
++ GNTMAP_host_map, netif->rx_shmem_handle);
++
++ lock_vm_area(netif->rx_comms_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1);
++ unlock_vm_area(netif->rx_comms_area);
++ BUG_ON(ret);
++}
++
++int netif_map(netif_t *netif, unsigned long tx_ring_ref,
++ unsigned long rx_ring_ref, unsigned int evtchn)
++{
++ int err = -ENOMEM;
++ netif_tx_sring_t *txs;
++ netif_rx_sring_t *rxs;
++ struct evtchn_bind_interdomain bind_interdomain;
++
++ /* Already connected through? */
++ if (netif->irq)
++ return 0;
++
++ netif->tx_comms_area = alloc_vm_area(PAGE_SIZE);
++ if (netif->tx_comms_area == NULL)
++ return -ENOMEM;
++ netif->rx_comms_area = alloc_vm_area(PAGE_SIZE);
++ if (netif->rx_comms_area == NULL)
++ goto err_rx;
++
++ err = map_frontend_pages(netif, tx_ring_ref, rx_ring_ref);
++ if (err)
++ goto err_map;
++
++ bind_interdomain.remote_dom = netif->domid;
++ bind_interdomain.remote_port = evtchn;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
++ &bind_interdomain);
++ if (err)
++ goto err_hypervisor;
++
++ netif->evtchn = bind_interdomain.local_port;
++
++ netif->irq = bind_evtchn_to_irqhandler(
++ netif->evtchn, netif_be_int, 0, netif->dev->name, netif);
++ disable_irq(netif->irq);
++
++ txs = (netif_tx_sring_t *)netif->tx_comms_area->addr;
++ BACK_RING_INIT(&netif->tx, txs, PAGE_SIZE);
++
++ rxs = (netif_rx_sring_t *)
++ ((char *)netif->rx_comms_area->addr);
++ BACK_RING_INIT(&netif->rx, rxs, PAGE_SIZE);
++
++ netif->rx_req_cons_peek = 0;
++
++ netif_get(netif);
++ wmb(); /* Other CPUs see new state before interface is started. */
++
++ rtnl_lock();
++ netif->status = CONNECTED;
++ wmb();
++ if (netif_running(netif->dev))
++ __netif_up(netif);
++ rtnl_unlock();
++
++ return 0;
++err_hypervisor:
++ unmap_frontend_pages(netif);
++err_map:
++ free_vm_area(netif->rx_comms_area);
++err_rx:
++ free_vm_area(netif->tx_comms_area);
++ return err;
++}
++
++static void netif_free(netif_t *netif)
++{
++ atomic_dec(&netif->refcnt);
++ wait_event(netif->waiting_to_free, atomic_read(&netif->refcnt) == 0);
++
++ if (netif->irq)
++ unbind_from_irqhandler(netif->irq, netif);
++
++ unregister_netdev(netif->dev);
++
++ if (netif->tx.sring) {
++ unmap_frontend_pages(netif);
++ free_vm_area(netif->tx_comms_area);
++ free_vm_area(netif->rx_comms_area);
++ }
++
++ free_netdev(netif->dev);
++}
++
++void netif_disconnect(netif_t *netif)
++{
++ switch (netif->status) {
++ case CONNECTED:
++ rtnl_lock();
++ netif->status = DISCONNECTING;
++ wmb();
++ if (netif_running(netif->dev))
++ __netif_down(netif);
++ rtnl_unlock();
++ netif_put(netif);
++ /* fall through */
++ case DISCONNECTED:
++ netif_free(netif);
++ break;
++ default:
++ BUG();
++ }
++}
+diff -urNp linux-2.6/drivers/xen/netback/loopback.c new/drivers/xen/netback/loopback.c
+--- linux-2.6/drivers/xen/netback/loopback.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/loopback.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,260 @@
++/******************************************************************************
++ * netback/loopback.c
++ *
++ * A two-interface loopback device to emulate a local netfront-netback
++ * connection. This ensures that local packet delivery looks identical
++ * to inter-domain delivery. Most importantly, packets delivered locally
++ * originating from other domains will get *copied* when they traverse this
++ * driver. This prevents unbounded delays in socket-buffer queues from
++ * causing the netback driver to "seize up".
++ *
++ * This driver creates a symmetric pair of loopback interfaces with names
++ * vif0.0 and veth0. The intention is that 'vif0.0' is bound to an Ethernet
++ * bridge, just like a proper netback interface, while a local IP interface
++ * is configured on 'veth0'.
++ *
++ * As with a real netback interface, vif0.0 is configured with a suitable
++ * dummy MAC address. No default is provided for veth0: a reasonable strategy
++ * is to transfer eth0's MAC address to veth0, and give eth0 a dummy address
++ * (to avoid confusing the Etherbridge).
++ *
++ * Copyright (c) 2005 K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/netdevice.h>
++#include <linux/inetdevice.h>
++#include <linux/etherdevice.h>
++#include <linux/skbuff.h>
++#include <linux/ethtool.h>
++#include <net/dst.h>
++
++static int nloopbacks = 8;
++module_param(nloopbacks, int, 0);
++MODULE_PARM_DESC(nloopbacks, "Number of netback-loopback devices to create");
++
++struct net_private {
++ struct net_device *loopback_dev;
++ struct net_device_stats stats;
++};
++
++static int loopback_open(struct net_device *dev)
++{
++ struct net_private *np = netdev_priv(dev);
++ memset(&np->stats, 0, sizeof(np->stats));
++ netif_start_queue(dev);
++ return 0;
++}
++
++static int loopback_close(struct net_device *dev)
++{
++ netif_stop_queue(dev);
++ return 0;
++}
++
++static int loopback_start_xmit(struct sk_buff *skb, struct net_device *dev)
++{
++ struct net_private *np = netdev_priv(dev);
++
++ dst_release(skb->dst);
++ skb->dst = NULL;
++
++ skb_orphan(skb);
++
++ np->stats.tx_bytes += skb->len;
++ np->stats.tx_packets++;
++
++ /* Switch to loopback context. */
++ dev = np->loopback_dev;
++ np = netdev_priv(dev);
++
++ np->stats.rx_bytes += skb->len;
++ np->stats.rx_packets++;
++
++ if (skb->ip_summed == CHECKSUM_HW) {
++ /* Defer checksum calculation. */
++ skb->proto_csum_blank = 1;
++ /* Must be a local packet: assert its integrity. */
++ skb->proto_data_valid = 1;
++ }
++
++ skb->ip_summed = skb->proto_data_valid ?
++ CHECKSUM_UNNECESSARY : CHECKSUM_NONE;
++
++ skb->pkt_type = PACKET_HOST; /* overridden by eth_type_trans() */
++ skb->protocol = eth_type_trans(skb, dev);
++ skb->dev = dev;
++ dev->last_rx = jiffies;
++ netif_rx(skb);
++
++ return 0;
++}
++
++static struct net_device_stats *loopback_get_stats(struct net_device *dev)
++{
++ struct net_private *np = netdev_priv(dev);
++ return &np->stats;
++}
++
++static struct ethtool_ops network_ethtool_ops =
++{
++ .get_tx_csum = ethtool_op_get_tx_csum,
++ .set_tx_csum = ethtool_op_set_tx_csum,
++ .get_sg = ethtool_op_get_sg,
++ .set_sg = ethtool_op_set_sg,
++ .get_tso = ethtool_op_get_tso,
++ .set_tso = ethtool_op_set_tso,
++};
++
++/*
++ * Nothing to do here. Virtual interface is point-to-point and the
++ * physical interface is probably promiscuous anyway.
++ */
++static void loopback_set_multicast_list(struct net_device *dev)
++{
++}
++
++static void loopback_construct(struct net_device *dev, struct net_device *lo)
++{
++ struct net_private *np = netdev_priv(dev);
++
++ np->loopback_dev = lo;
++
++ dev->open = loopback_open;
++ dev->stop = loopback_close;
++ dev->hard_start_xmit = loopback_start_xmit;
++ dev->get_stats = loopback_get_stats;
++ dev->set_multicast_list = loopback_set_multicast_list;
++ dev->change_mtu = NULL; /* allow arbitrary mtu */
++
++ dev->tx_queue_len = 0;
++
++ dev->features = (NETIF_F_HIGHDMA |
++ NETIF_F_LLTX |
++ NETIF_F_TSO |
++ NETIF_F_SG |
++ NETIF_F_IP_CSUM);
++
++ SET_ETHTOOL_OPS(dev, &network_ethtool_ops);
++
++ /*
++ * We do not set a jumbo MTU on the interface. Otherwise the network
++ * stack will try to send large packets that will get dropped by the
++ * Ethernet bridge (unless the physical Ethernet interface is
++ * configured to transfer jumbo packets). If a larger MTU is desired
++ * then the system administrator can specify it using the 'ifconfig'
++ * command.
++ */
++ /*dev->mtu = 16*1024;*/
++}
++
++static int __init make_loopback(int i)
++{
++ struct net_device *dev1, *dev2;
++ char dev_name[IFNAMSIZ];
++ int err = -ENOMEM;
++
++ sprintf(dev_name, "vif0.%d", i);
++ dev1 = alloc_netdev(sizeof(struct net_private), dev_name, ether_setup);
++ if (!dev1)
++ return err;
++
++ sprintf(dev_name, "veth%d", i);
++ dev2 = alloc_netdev(sizeof(struct net_private), dev_name, ether_setup);
++ if (!dev2)
++ goto fail_netdev2;
++
++ loopback_construct(dev1, dev2);
++ loopback_construct(dev2, dev1);
++
++ /*
++ * Initialise a dummy MAC address for the 'dummy backend' interface. We
++ * choose the numerically largest non-broadcast address to prevent the
++ * address getting stolen by an Ethernet bridge for STP purposes.
++ */
++ memset(dev1->dev_addr, 0xFF, ETH_ALEN);
++ dev1->dev_addr[0] &= ~0x01;
++
++ if ((err = register_netdev(dev1)) != 0)
++ goto fail;
++
++ if ((err = register_netdev(dev2)) != 0) {
++ unregister_netdev(dev1);
++ goto fail;
++ }
++
++ return 0;
++
++ fail:
++ free_netdev(dev2);
++ fail_netdev2:
++ free_netdev(dev1);
++ return err;
++}
++
++static void __init clean_loopback(int i)
++{
++ struct net_device *dev1, *dev2;
++ char dev_name[IFNAMSIZ];
++
++ sprintf(dev_name, "vif0.%d", i);
++ dev1 = dev_get_by_name(dev_name);
++ sprintf(dev_name, "veth%d", i);
++ dev2 = dev_get_by_name(dev_name);
++ if (dev1 && dev2) {
++ unregister_netdev(dev2);
++ unregister_netdev(dev1);
++ free_netdev(dev2);
++ free_netdev(dev1);
++ }
++}
++
++static int __init loopback_init(void)
++{
++ int i, err = 0;
++
++ for (i = 0; i < nloopbacks; i++)
++ if ((err = make_loopback(i)) != 0)
++ break;
++
++ return err;
++}
++
++module_init(loopback_init);
++
++static void __exit loopback_exit(void)
++{
++ int i;
++
++ for (i = nloopbacks; i-- > 0; )
++ clean_loopback(i);
++}
++
++module_exit(loopback_exit);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/netback/Makefile new/drivers/xen/netback/Makefile
+--- linux-2.6/drivers/xen/netback/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/Makefile 2006-05-09 12:34:37.000000000 +0200
+@@ -0,0 +1,5 @@
++obj-$(CONFIG_XEN_NETDEV_BACKEND) := netbk.o
++obj-$(CONFIG_XEN_NETDEV_LOOPBACK) += netloop.o
++
++netbk-y := netback.o xenbus.o interface.o
++netloop-y := loopback.o
+diff -urNp linux-2.6/drivers/xen/netback/netback.c new/drivers/xen/netback/netback.c
+--- linux-2.6/drivers/xen/netback/netback.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/netback.c 2006-07-07 16:05:52.000000000 +0200
+@@ -0,0 +1,1107 @@
++/******************************************************************************
++ * drivers/xen/netback/netback.c
++ *
++ * Back-end of the driver for virtual network devices. This portion of the
++ * driver exports a 'unified' network-device interface that can be accessed
++ * by any operating system that implements a compatible front end. A
++ * reference front-end implementation can be found in:
++ * drivers/xen/netfront/netfront.c
++ *
++ * Copyright (c) 2002-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include "common.h"
++#include <xen/balloon.h>
++#include <xen/interface/memory.h>
++
++/*#define NETBE_DEBUG_INTERRUPT*/
++
++static void netif_idx_release(u16 pending_idx);
++static void netif_page_release(struct page *page);
++static void make_tx_response(netif_t *netif,
++ netif_tx_request_t *txp,
++ s8 st);
++static int make_rx_response(netif_t *netif,
++ u16 id,
++ s8 st,
++ u16 offset,
++ u16 size,
++ u16 flags);
++
++static void net_tx_action(unsigned long unused);
++static DECLARE_TASKLET(net_tx_tasklet, net_tx_action, 0);
++
++static void net_rx_action(unsigned long unused);
++static DECLARE_TASKLET(net_rx_tasklet, net_rx_action, 0);
++
++static struct timer_list net_timer;
++
++#define MAX_PENDING_REQS 256
++
++static struct sk_buff_head rx_queue;
++static multicall_entry_t rx_mcl[NET_RX_RING_SIZE+1];
++static mmu_update_t rx_mmu[NET_RX_RING_SIZE];
++static gnttab_transfer_t grant_rx_op[NET_RX_RING_SIZE];
++static unsigned char rx_notify[NR_IRQS];
++
++static unsigned long mmap_vstart;
++#define MMAP_VADDR(_req) (mmap_vstart + ((_req) * PAGE_SIZE))
++
++#define PKT_PROT_LEN 64
++
++static struct {
++ netif_tx_request_t req;
++ netif_t *netif;
++} pending_tx_info[MAX_PENDING_REQS];
++static u16 pending_ring[MAX_PENDING_REQS];
++typedef unsigned int PEND_RING_IDX;
++#define MASK_PEND_IDX(_i) ((_i)&(MAX_PENDING_REQS-1))
++static PEND_RING_IDX pending_prod, pending_cons;
++#define NR_PENDING_REQS (MAX_PENDING_REQS - pending_prod + pending_cons)
++
++/* Freed TX SKBs get batched on this ring before return to pending_ring. */
++static u16 dealloc_ring[MAX_PENDING_REQS];
++static PEND_RING_IDX dealloc_prod, dealloc_cons;
++
++static struct sk_buff_head tx_queue;
++
++static grant_handle_t grant_tx_handle[MAX_PENDING_REQS];
++static gnttab_unmap_grant_ref_t tx_unmap_ops[MAX_PENDING_REQS];
++static gnttab_map_grant_ref_t tx_map_ops[MAX_PENDING_REQS];
++
++static struct list_head net_schedule_list;
++static spinlock_t net_schedule_list_lock;
++
++#define MAX_MFN_ALLOC 64
++static unsigned long mfn_list[MAX_MFN_ALLOC];
++static unsigned int alloc_index = 0;
++static DEFINE_SPINLOCK(mfn_lock);
++
++static unsigned long alloc_mfn(void)
++{
++ unsigned long mfn = 0, flags;
++ struct xen_memory_reservation reservation = {
++ .nr_extents = MAX_MFN_ALLOC,
++ .extent_order = 0,
++ .domid = DOMID_SELF
++ };
++ set_xen_guest_handle(reservation.extent_start, mfn_list);
++ spin_lock_irqsave(&mfn_lock, flags);
++ if ( unlikely(alloc_index == 0) )
++ alloc_index = HYPERVISOR_memory_op(
++ XENMEM_increase_reservation, &reservation);
++ if ( alloc_index != 0 )
++ mfn = mfn_list[--alloc_index];
++ spin_unlock_irqrestore(&mfn_lock, flags);
++ return mfn;
++}
++
++static inline void maybe_schedule_tx_action(void)
++{
++ smp_mb();
++ if ((NR_PENDING_REQS < (MAX_PENDING_REQS/2)) &&
++ !list_empty(&net_schedule_list))
++ tasklet_schedule(&net_tx_tasklet);
++}
++
++/*
++ * A gross way of confirming the origin of an skb data page. The slab
++ * allocator abuses a field in the page struct to cache the kmem_cache_t ptr.
++ */
++static inline int is_xen_skb(struct sk_buff *skb)
++{
++ extern kmem_cache_t *skbuff_cachep;
++ kmem_cache_t *cp = (kmem_cache_t *)virt_to_page(skb->head)->lru.next;
++ return (cp == skbuff_cachep);
++}
++
++int netif_be_start_xmit(struct sk_buff *skb, struct net_device *dev)
++{
++ netif_t *netif = netdev_priv(dev);
++
++ BUG_ON(skb->dev != dev);
++
++ /* Drop the packet if the target domain has no receive buffers. */
++ if (!netif->active ||
++ (netif->rx_req_cons_peek == netif->rx.sring->req_prod) ||
++ ((netif->rx_req_cons_peek - netif->rx.rsp_prod_pvt) ==
++ NET_RX_RING_SIZE))
++ goto drop;
++
++ /*
++ * We do not copy the packet unless:
++ * 1. The data is shared; or
++ * 2. The data is not allocated from our special cache.
++ * NB. We also couldn't cope with fragmented packets, but we won't get
++ * any because we not advertise the NETIF_F_SG feature.
++ */
++ if (skb_shared(skb) || skb_cloned(skb) || !is_xen_skb(skb)) {
++ int hlen = skb->data - skb->head;
++ int ret;
++ struct sk_buff *nskb = dev_alloc_skb(hlen + skb->len);
++ if ( unlikely(nskb == NULL) )
++ goto drop;
++ skb_reserve(nskb, hlen);
++ __skb_put(nskb, skb->len);
++ ret = skb_copy_bits(skb, -hlen, nskb->data - hlen,
++ skb->len + hlen);
++ BUG_ON(ret);
++ /* Copy only the header fields we use in this driver. */
++ nskb->dev = skb->dev;
++ nskb->ip_summed = skb->ip_summed;
++ nskb->proto_data_valid = skb->proto_data_valid;
++ dev_kfree_skb(skb);
++ skb = nskb;
++ }
++
++ netif->rx_req_cons_peek++;
++ netif_get(netif);
++
++ skb_queue_tail(&rx_queue, skb);
++ tasklet_schedule(&net_rx_tasklet);
++
++ return 0;
++
++ drop:
++ netif->stats.tx_dropped++;
++ dev_kfree_skb(skb);
++ return 0;
++}
++
++#if 0
++static void xen_network_done_notify(void)
++{
++ static struct net_device *eth0_dev = NULL;
++ if (unlikely(eth0_dev == NULL))
++ eth0_dev = __dev_get_by_name("eth0");
++ netif_rx_schedule(eth0_dev);
++}
++/*
++ * Add following to poll() function in NAPI driver (Tigon3 is example):
++ * if ( xen_network_done() )
++ * tg3_enable_ints(tp);
++ */
++int xen_network_done(void)
++{
++ return skb_queue_empty(&rx_queue);
++}
++#endif
++
++static void net_rx_action(unsigned long unused)
++{
++ netif_t *netif = NULL;
++ s8 status;
++ u16 size, id, irq, flags;
++ multicall_entry_t *mcl;
++ mmu_update_t *mmu;
++ gnttab_transfer_t *gop;
++ unsigned long vdata, old_mfn, new_mfn;
++ struct sk_buff_head rxq;
++ struct sk_buff *skb;
++ u16 notify_list[NET_RX_RING_SIZE];
++ int notify_nr = 0;
++ int ret;
++
++ skb_queue_head_init(&rxq);
++
++ mcl = rx_mcl;
++ mmu = rx_mmu;
++ gop = grant_rx_op;
++
++ while ((skb = skb_dequeue(&rx_queue)) != NULL) {
++ netif = netdev_priv(skb->dev);
++ vdata = (unsigned long)skb->data;
++ old_mfn = virt_to_mfn(vdata);
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* Memory squeeze? Back off for an arbitrary while. */
++ if ((new_mfn = alloc_mfn()) == 0) {
++ if ( net_ratelimit() )
++ WPRINTK("Memory squeeze in netback "
++ "driver.\n");
++ mod_timer(&net_timer, jiffies + HZ);
++ skb_queue_head(&rx_queue, skb);
++ break;
++ }
++ /*
++ * Set the new P2M table entry before reassigning
++ * the old data page. Heed the comment in
++ * pgtable-2level.h:pte_page(). :-)
++ */
++ set_phys_to_machine(
++ __pa(skb->data) >> PAGE_SHIFT,
++ new_mfn);
++
++ MULTI_update_va_mapping(mcl, vdata,
++ pfn_pte_ma(new_mfn,
++ PAGE_KERNEL), 0);
++ mcl++;
++
++ mmu->ptr = ((maddr_t)new_mfn << PAGE_SHIFT) |
++ MMU_MACHPHYS_UPDATE;
++ mmu->val = __pa(vdata) >> PAGE_SHIFT;
++ mmu++;
++ }
++
++ gop->mfn = old_mfn;
++ gop->domid = netif->domid;
++ gop->ref = RING_GET_REQUEST(
++ &netif->rx, netif->rx.req_cons)->gref;
++ netif->rx.req_cons++;
++ gop++;
++
++ __skb_queue_tail(&rxq, skb);
++
++ /* Filled the batch queue? */
++ if ((gop - grant_rx_op) == ARRAY_SIZE(grant_rx_op))
++ break;
++ }
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ if (mcl == rx_mcl)
++ return;
++
++ mcl[-1].args[MULTI_UVMFLAGS_INDEX] = UVMF_TLB_FLUSH|UVMF_ALL;
++
++ if (mmu - rx_mmu) {
++ mcl->op = __HYPERVISOR_mmu_update;
++ mcl->args[0] = (unsigned long)rx_mmu;
++ mcl->args[1] = mmu - rx_mmu;
++ mcl->args[2] = 0;
++ mcl->args[3] = DOMID_SELF;
++ mcl++;
++ }
++
++ ret = HYPERVISOR_multicall(rx_mcl, mcl - rx_mcl);
++ BUG_ON(ret != 0);
++ }
++
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_transfer, grant_rx_op,
++ gop - grant_rx_op);
++ BUG_ON(ret != 0);
++
++ mcl = rx_mcl;
++ gop = grant_rx_op;
++ while ((skb = __skb_dequeue(&rxq)) != NULL) {
++ netif = netdev_priv(skb->dev);
++ size = skb->tail - skb->data;
++
++ atomic_set(&(skb_shinfo(skb)->dataref), 1);
++ skb_shinfo(skb)->nr_frags = 0;
++ skb_shinfo(skb)->frag_list = NULL;
++
++ netif->stats.tx_bytes += size;
++ netif->stats.tx_packets++;
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* The update_va_mapping() must not fail. */
++ BUG_ON(mcl->result != 0);
++ mcl++;
++ }
++
++ /* Check the reassignment error code. */
++ status = NETIF_RSP_OKAY;
++ if (gop->status != 0) {
++ DPRINTK("Bad status %d from grant transfer to DOM%u\n",
++ gop->status, netif->domid);
++ /*
++ * Page no longer belongs to us unless GNTST_bad_page,
++ * but that should be a fatal error anyway.
++ */
++ BUG_ON(gop->status == GNTST_bad_page);
++ status = NETIF_RSP_ERROR;
++ }
++ irq = netif->irq;
++ id = RING_GET_REQUEST(&netif->rx, netif->rx.rsp_prod_pvt)->id;
++ flags = 0;
++ if (skb->ip_summed == CHECKSUM_HW) /* local packet? */
++ flags |= NETRXF_csum_blank | NETRXF_data_validated;
++ else if (skb->proto_data_valid) /* remote but checksummed? */
++ flags |= NETRXF_data_validated;
++ if (make_rx_response(netif, id, status,
++ (unsigned long)skb->data & ~PAGE_MASK,
++ size, flags) &&
++ (rx_notify[irq] == 0)) {
++ rx_notify[irq] = 1;
++ notify_list[notify_nr++] = irq;
++ }
++
++ netif_put(netif);
++ dev_kfree_skb(skb);
++ gop++;
++ }
++
++ while (notify_nr != 0) {
++ irq = notify_list[--notify_nr];
++ rx_notify[irq] = 0;
++ notify_remote_via_irq(irq);
++ }
++
++ /* More work to do? */
++ if (!skb_queue_empty(&rx_queue) && !timer_pending(&net_timer))
++ tasklet_schedule(&net_rx_tasklet);
++#if 0
++ else
++ xen_network_done_notify();
++#endif
++}
++
++static void net_alarm(unsigned long unused)
++{
++ tasklet_schedule(&net_rx_tasklet);
++}
++
++struct net_device_stats *netif_be_get_stats(struct net_device *dev)
++{
++ netif_t *netif = netdev_priv(dev);
++ return &netif->stats;
++}
++
++static int __on_net_schedule_list(netif_t *netif)
++{
++ return netif->list.next != NULL;
++}
++
++static void remove_from_net_schedule_list(netif_t *netif)
++{
++ spin_lock_irq(&net_schedule_list_lock);
++ if (likely(__on_net_schedule_list(netif))) {
++ list_del(&netif->list);
++ netif->list.next = NULL;
++ netif_put(netif);
++ }
++ spin_unlock_irq(&net_schedule_list_lock);
++}
++
++static void add_to_net_schedule_list_tail(netif_t *netif)
++{
++ if (__on_net_schedule_list(netif))
++ return;
++
++ spin_lock_irq(&net_schedule_list_lock);
++ if (!__on_net_schedule_list(netif) && netif->active) {
++ list_add_tail(&netif->list, &net_schedule_list);
++ netif_get(netif);
++ }
++ spin_unlock_irq(&net_schedule_list_lock);
++}
++
++/*
++ * Note on CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER:
++ * If this driver is pipelining transmit requests then we can be very
++ * aggressive in avoiding new-packet notifications -- frontend only needs to
++ * send a notification if there are no outstanding unreceived responses.
++ * If we may be buffer transmit buffers for any reason then we must be rather
++ * more conservative and treat this as the final check for pending work.
++ */
++void netif_schedule_work(netif_t *netif)
++{
++ int more_to_do;
++
++#ifdef CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER
++ more_to_do = RING_HAS_UNCONSUMED_REQUESTS(&netif->tx);
++#else
++ RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, more_to_do);
++#endif
++
++ if (more_to_do) {
++ add_to_net_schedule_list_tail(netif);
++ maybe_schedule_tx_action();
++ }
++}
++
++void netif_deschedule_work(netif_t *netif)
++{
++ remove_from_net_schedule_list(netif);
++}
++
++
++static void tx_credit_callback(unsigned long data)
++{
++ netif_t *netif = (netif_t *)data;
++ netif->remaining_credit = netif->credit_bytes;
++ netif_schedule_work(netif);
++}
++
++inline static void net_tx_action_dealloc(void)
++{
++ gnttab_unmap_grant_ref_t *gop;
++ u16 pending_idx;
++ PEND_RING_IDX dc, dp;
++ netif_t *netif;
++ int ret;
++
++ dc = dealloc_cons;
++ dp = dealloc_prod;
++
++ /* Ensure we see all indexes enqueued by netif_idx_release(). */
++ smp_rmb();
++
++ /*
++ * Free up any grants we have finished using
++ */
++ gop = tx_unmap_ops;
++ while (dc != dp) {
++ pending_idx = dealloc_ring[MASK_PEND_IDX(dc++)];
++ gnttab_set_unmap_op(gop, MMAP_VADDR(pending_idx),
++ GNTMAP_host_map,
++ grant_tx_handle[pending_idx]);
++ gop++;
++ }
++ ret = HYPERVISOR_grant_table_op(
++ GNTTABOP_unmap_grant_ref, tx_unmap_ops, gop - tx_unmap_ops);
++ BUG_ON(ret);
++
++ while (dealloc_cons != dp) {
++ pending_idx = dealloc_ring[MASK_PEND_IDX(dealloc_cons++)];
++
++ netif = pending_tx_info[pending_idx].netif;
++
++ make_tx_response(netif, &pending_tx_info[pending_idx].req,
++ NETIF_RSP_OKAY);
++
++ pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx;
++
++ netif_put(netif);
++ }
++}
++
++static void netbk_tx_err(netif_t *netif, netif_tx_request_t *txp, RING_IDX end)
++{
++ RING_IDX cons = netif->tx.req_cons;
++
++ do {
++ make_tx_response(netif, txp, NETIF_RSP_ERROR);
++ if (++cons >= end)
++ break;
++ txp = RING_GET_REQUEST(&netif->tx, cons);
++ } while (1);
++ netif->tx.req_cons = cons;
++ netif_schedule_work(netif);
++ netif_put(netif);
++}
++
++static int netbk_count_requests(netif_t *netif, netif_tx_request_t *txp,
++ int work_to_do)
++{
++ netif_tx_request_t *first = txp;
++ RING_IDX cons = netif->tx.req_cons;
++ int frags = 0;
++
++ while (txp->flags & NETTXF_more_data) {
++ if (frags >= work_to_do) {
++ DPRINTK("Need more frags\n");
++ return -frags;
++ }
++
++ txp = RING_GET_REQUEST(&netif->tx, cons + frags);
++ if (txp->size > first->size) {
++ DPRINTK("Frags galore\n");
++ return -frags;
++ }
++
++ first->size -= txp->size;
++ frags++;
++
++ if (unlikely((txp->offset + txp->size) > PAGE_SIZE)) {
++ DPRINTK("txp->offset: %x, size: %u\n",
++ txp->offset, txp->size);
++ return -frags;
++ }
++ }
++
++ return frags;
++}
++
++static gnttab_map_grant_ref_t *netbk_get_requests(netif_t *netif,
++ struct sk_buff *skb,
++ gnttab_map_grant_ref_t *mop)
++{
++ struct skb_shared_info *shinfo = skb_shinfo(skb);
++ skb_frag_t *frags = shinfo->frags;
++ netif_tx_request_t *txp;
++ unsigned long pending_idx = *((u16 *)skb->data);
++ RING_IDX cons = netif->tx.req_cons;
++ int i, start;
++
++ /* Skip first skb fragment if it is on same page as header fragment. */
++ start = ((unsigned long)shinfo->frags[0].page == pending_idx);
++
++ for (i = start; i < shinfo->nr_frags; i++) {
++ txp = RING_GET_REQUEST(&netif->tx, cons++);
++ pending_idx = pending_ring[MASK_PEND_IDX(pending_cons++)];
++
++ gnttab_set_map_op(mop++, MMAP_VADDR(pending_idx),
++ GNTMAP_host_map | GNTMAP_readonly,
++ txp->gref, netif->domid);
++
++ memcpy(&pending_tx_info[pending_idx].req, txp, sizeof(*txp));
++ netif_get(netif);
++ pending_tx_info[pending_idx].netif = netif;
++ frags[i].page = (void *)pending_idx;
++ }
++
++ return mop;
++}
++
++static int netbk_tx_check_mop(struct sk_buff *skb,
++ gnttab_map_grant_ref_t **mopp)
++{
++ gnttab_map_grant_ref_t *mop = *mopp;
++ int pending_idx = *((u16 *)skb->data);
++ netif_t *netif = pending_tx_info[pending_idx].netif;
++ netif_tx_request_t *txp;
++ struct skb_shared_info *shinfo = skb_shinfo(skb);
++ int nr_frags = shinfo->nr_frags;
++ int i, err, start;
++
++ /* Check status of header. */
++ err = mop->status;
++ if (unlikely(err)) {
++ txp = &pending_tx_info[pending_idx].req;
++ make_tx_response(netif, txp, NETIF_RSP_ERROR);
++ pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx;
++ netif_put(netif);
++ } else {
++ set_phys_to_machine(
++ __pa(MMAP_VADDR(pending_idx)) >> PAGE_SHIFT,
++ FOREIGN_FRAME(mop->dev_bus_addr >> PAGE_SHIFT));
++ grant_tx_handle[pending_idx] = mop->handle;
++ }
++
++ /* Skip first skb fragment if it is on same page as header fragment. */
++ start = ((unsigned long)shinfo->frags[0].page == pending_idx);
++
++ for (i = start; i < nr_frags; i++) {
++ int j, newerr;
++
++ pending_idx = (unsigned long)shinfo->frags[i].page;
++
++ /* Check error status: if okay then remember grant handle. */
++ newerr = (++mop)->status;
++ if (likely(!newerr)) {
++ set_phys_to_machine(
++ __pa(MMAP_VADDR(pending_idx))>>PAGE_SHIFT,
++ FOREIGN_FRAME(mop->dev_bus_addr>>PAGE_SHIFT));
++ grant_tx_handle[pending_idx] = mop->handle;
++ /* Had a previous error? Invalidate this fragment. */
++ if (unlikely(err))
++ netif_idx_release(pending_idx);
++ continue;
++ }
++
++ /* Error on this fragment: respond to client with an error. */
++ txp = &pending_tx_info[pending_idx].req;
++ make_tx_response(netif, txp, NETIF_RSP_ERROR);
++ pending_ring[MASK_PEND_IDX(pending_prod++)] = pending_idx;
++ netif_put(netif);
++
++ /* Not the first error? Preceding frags already invalidated. */
++ if (err)
++ continue;
++
++ /* First error: invalidate header and preceding fragments. */
++ pending_idx = *((u16 *)skb->data);
++ netif_idx_release(pending_idx);
++ for (j = start; j < i; j++) {
++ pending_idx = (unsigned long)shinfo->frags[i].page;
++ netif_idx_release(pending_idx);
++ }
++
++ /* Remember the error: invalidate all subsequent fragments. */
++ err = newerr;
++ }
++
++ *mopp = mop + 1;
++ return err;
++}
++
++static void netbk_fill_frags(struct sk_buff *skb)
++{
++ struct skb_shared_info *shinfo = skb_shinfo(skb);
++ int nr_frags = shinfo->nr_frags;
++ int i;
++
++ for (i = 0; i < nr_frags; i++) {
++ skb_frag_t *frag = shinfo->frags + i;
++ netif_tx_request_t *txp;
++ unsigned long pending_idx;
++
++ pending_idx = (unsigned long)frag->page;
++ txp = &pending_tx_info[pending_idx].req;
++ frag->page = virt_to_page(MMAP_VADDR(pending_idx));
++ frag->size = txp->size;
++ frag->page_offset = txp->offset;
++
++ skb->len += txp->size;
++ skb->data_len += txp->size;
++ skb->truesize += txp->size;
++ }
++}
++
++int netbk_get_extras(netif_t *netif, struct netif_extra_info *extras,
++ int work_to_do)
++{
++ struct netif_extra_info *extra;
++ RING_IDX cons = netif->tx.req_cons;
++
++ do {
++ if (unlikely(work_to_do-- <= 0)) {
++ DPRINTK("Missing extra info\n");
++ return -EBADR;
++ }
++
++ extra = (struct netif_extra_info *)
++ RING_GET_REQUEST(&netif->tx, cons);
++ if (unlikely(!extra->type ||
++ extra->type >= XEN_NETIF_EXTRA_TYPE_MAX)) {
++ netif->tx.req_cons = ++cons;
++ DPRINTK("Invalid extra type: %d\n", extra->type);
++ return -EINVAL;
++ }
++
++ memcpy(&extras[extra->type - 1], extra, sizeof(*extra));
++ netif->tx.req_cons = ++cons;
++ } while (extra->flags & XEN_NETIF_EXTRA_FLAG_MORE);
++
++ return work_to_do;
++}
++
++/* Called after netfront has transmitted */
++static void net_tx_action(unsigned long unused)
++{
++ struct list_head *ent;
++ struct sk_buff *skb;
++ netif_t *netif;
++ netif_tx_request_t txreq;
++ struct netif_extra_info extras[XEN_NETIF_EXTRA_TYPE_MAX - 1];
++ u16 pending_idx;
++ RING_IDX i;
++ gnttab_map_grant_ref_t *mop;
++ unsigned int data_len;
++ int ret, work_to_do;
++
++ if (dealloc_cons != dealloc_prod)
++ net_tx_action_dealloc();
++
++ mop = tx_map_ops;
++ while (((NR_PENDING_REQS + MAX_SKB_FRAGS) < MAX_PENDING_REQS) &&
++ !list_empty(&net_schedule_list)) {
++ /* Get a netif from the list with work to do. */
++ ent = net_schedule_list.next;
++ netif = list_entry(ent, netif_t, list);
++ netif_get(netif);
++ remove_from_net_schedule_list(netif);
++
++ RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, work_to_do);
++ if (!work_to_do) {
++ netif_put(netif);
++ continue;
++ }
++
++ i = netif->tx.req_cons;
++ rmb(); /* Ensure that we see the request before we copy it. */
++ memcpy(&txreq, RING_GET_REQUEST(&netif->tx, i), sizeof(txreq));
++ /* Credit-based scheduling. */
++ if (txreq.size > netif->remaining_credit) {
++ unsigned long now = jiffies;
++ unsigned long next_credit =
++ netif->credit_timeout.expires +
++ msecs_to_jiffies(netif->credit_usec / 1000);
++
++ /* Timer could already be pending in rare cases. */
++ if (timer_pending(&netif->credit_timeout))
++ break;
++
++ /* Passed the point where we can replenish credit? */
++ if (time_after_eq(now, next_credit)) {
++ netif->credit_timeout.expires = now;
++ netif->remaining_credit = netif->credit_bytes;
++ }
++
++ /* Still too big to send right now? Set a callback. */
++ if (txreq.size > netif->remaining_credit) {
++ netif->remaining_credit = 0;
++ netif->credit_timeout.data =
++ (unsigned long)netif;
++ netif->credit_timeout.function =
++ tx_credit_callback;
++ __mod_timer(&netif->credit_timeout,
++ next_credit);
++ break;
++ }
++ }
++ netif->remaining_credit -= txreq.size;
++
++ work_to_do--;
++ netif->tx.req_cons = ++i;
++
++ memset(extras, 0, sizeof(extras));
++ if (txreq.flags & NETTXF_extra_info) {
++ work_to_do = netbk_get_extras(netif, extras,
++ work_to_do);
++ if (unlikely(work_to_do < 0)) {
++ netbk_tx_err(netif, &txreq, 0);
++ continue;
++ }
++ i = netif->tx.req_cons;
++ }
++
++ ret = netbk_count_requests(netif, &txreq, work_to_do);
++ if (unlikely(ret < 0)) {
++ netbk_tx_err(netif, &txreq, i - ret);
++ continue;
++ }
++ i += ret;
++
++ if (unlikely(ret > MAX_SKB_FRAGS)) {
++ DPRINTK("Too many frags\n");
++ netbk_tx_err(netif, &txreq, i);
++ continue;
++ }
++
++ if (unlikely(txreq.size < ETH_HLEN)) {
++ DPRINTK("Bad packet size: %d\n", txreq.size);
++ netbk_tx_err(netif, &txreq, i);
++ continue;
++ }
++
++ /* No crossing a page as the payload mustn't fragment. */
++ if (unlikely((txreq.offset + txreq.size) > PAGE_SIZE)) {
++ DPRINTK("txreq.offset: %x, size: %u, end: %lu\n",
++ txreq.offset, txreq.size,
++ (txreq.offset &~PAGE_MASK) + txreq.size);
++ netbk_tx_err(netif, &txreq, i);
++ continue;
++ }
++
++ pending_idx = pending_ring[MASK_PEND_IDX(pending_cons)];
++
++ data_len = (txreq.size > PKT_PROT_LEN &&
++ ret < MAX_SKB_FRAGS) ?
++ PKT_PROT_LEN : txreq.size;
++
++ skb = alloc_skb(data_len+16, GFP_ATOMIC);
++ if (unlikely(skb == NULL)) {
++ DPRINTK("Can't allocate a skb in start_xmit.\n");
++ netbk_tx_err(netif, &txreq, i);
++ break;
++ }
++
++ /* Packets passed to netif_rx() must have some headroom. */
++ skb_reserve(skb, 16);
++
++ if (extras[XEN_NETIF_EXTRA_TYPE_GSO - 1].type) {
++ struct netif_extra_info *gso;
++ gso = &extras[XEN_NETIF_EXTRA_TYPE_GSO - 1];
++
++ /* Currently on TCPv4 S.O. is supported. */
++ if (gso->u.gso.type != XEN_NETIF_GSO_TCPV4) {
++ DPRINTK("Bad GSO type %d.\n", gso->u.gso.type);
++ kfree_skb(skb);
++ netbk_tx_err(netif, &txreq, i);
++ break;
++ }
++
++ skb_shinfo(skb)->gso_size = gso->u.gso.size;
++ skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
++
++ /* Header must be checked, and gso_segs computed. */
++ skb_shinfo(skb)->gso_type |= SKB_GSO_DODGY;
++ skb_shinfo(skb)->gso_segs = 0;
++ }
++
++ gnttab_set_map_op(mop, MMAP_VADDR(pending_idx),
++ GNTMAP_host_map | GNTMAP_readonly,
++ txreq.gref, netif->domid);
++ mop++;
++
++ memcpy(&pending_tx_info[pending_idx].req,
++ &txreq, sizeof(txreq));
++ pending_tx_info[pending_idx].netif = netif;
++ *((u16 *)skb->data) = pending_idx;
++
++ __skb_put(skb, data_len);
++
++ skb_shinfo(skb)->nr_frags = ret;
++ if (data_len < txreq.size) {
++ skb_shinfo(skb)->nr_frags++;
++ skb_shinfo(skb)->frags[0].page =
++ (void *)(unsigned long)pending_idx;
++ }
++
++ __skb_queue_tail(&tx_queue, skb);
++
++ pending_cons++;
++
++ mop = netbk_get_requests(netif, skb, mop);
++
++ netif->tx.req_cons = i;
++ netif_schedule_work(netif);
++
++ if ((mop - tx_map_ops) >= ARRAY_SIZE(tx_map_ops))
++ break;
++ }
++
++ if (mop == tx_map_ops)
++ return;
++
++ ret = HYPERVISOR_grant_table_op(
++ GNTTABOP_map_grant_ref, tx_map_ops, mop - tx_map_ops);
++ BUG_ON(ret);
++
++ mop = tx_map_ops;
++ while ((skb = __skb_dequeue(&tx_queue)) != NULL) {
++ netif_tx_request_t *txp;
++
++ pending_idx = *((u16 *)skb->data);
++ netif = pending_tx_info[pending_idx].netif;
++ txp = &pending_tx_info[pending_idx].req;
++
++ /* Check the remap error code. */
++ if (unlikely(netbk_tx_check_mop(skb, &mop))) {
++ printk(KERN_ALERT "#### netback grant fails\n");
++ skb_shinfo(skb)->nr_frags = 0;
++ kfree_skb(skb);
++ continue;
++ }
++
++ data_len = skb->len;
++ memcpy(skb->data,
++ (void *)(MMAP_VADDR(pending_idx)|txp->offset),
++ data_len);
++ if (data_len < txp->size) {
++ /* Append the packet payload as a fragment. */
++ txp->offset += data_len;
++ txp->size -= data_len;
++ } else {
++ /* Schedule a response immediately. */
++ netif_idx_release(pending_idx);
++ }
++
++ /*
++ * Old frontends do not assert data_validated but we
++ * can infer it from csum_blank so test both flags.
++ */
++ if (txp->flags & (NETTXF_data_validated|NETTXF_csum_blank)) {
++ skb->ip_summed = CHECKSUM_UNNECESSARY;
++ skb->proto_data_valid = 1;
++ } else {
++ skb->ip_summed = CHECKSUM_NONE;
++ skb->proto_data_valid = 0;
++ }
++ skb->proto_csum_blank = !!(txp->flags & NETTXF_csum_blank);
++
++ netbk_fill_frags(skb);
++
++ skb->dev = netif->dev;
++ skb->protocol = eth_type_trans(skb, skb->dev);
++
++ netif->stats.rx_bytes += skb->len;
++ netif->stats.rx_packets++;
++
++ netif_rx(skb);
++ netif->dev->last_rx = jiffies;
++ }
++}
++
++static void netif_idx_release(u16 pending_idx)
++{
++ static DEFINE_SPINLOCK(_lock);
++ unsigned long flags;
++
++ spin_lock_irqsave(&_lock, flags);
++ dealloc_ring[MASK_PEND_IDX(dealloc_prod)] = pending_idx;
++ /* Sync with net_tx_action_dealloc: insert idx /then/ incr producer. */
++ smp_wmb();
++ dealloc_prod++;
++ spin_unlock_irqrestore(&_lock, flags);
++
++ tasklet_schedule(&net_tx_tasklet);
++}
++
++static void netif_page_release(struct page *page)
++{
++ u16 pending_idx = page - virt_to_page(mmap_vstart);
++
++ /* Ready for next use. */
++ init_page_count(page);
++
++ netif_idx_release(pending_idx);
++}
++
++irqreturn_t netif_be_int(int irq, void *dev_id, struct pt_regs *regs)
++{
++ netif_t *netif = dev_id;
++ add_to_net_schedule_list_tail(netif);
++ maybe_schedule_tx_action();
++ return IRQ_HANDLED;
++}
++
++static void make_tx_response(netif_t *netif,
++ netif_tx_request_t *txp,
++ s8 st)
++{
++ RING_IDX i = netif->tx.rsp_prod_pvt;
++ netif_tx_response_t *resp;
++ int notify;
++
++ resp = RING_GET_RESPONSE(&netif->tx, i);
++ resp->id = txp->id;
++ resp->status = st;
++
++ if (txp->flags & NETTXF_extra_info)
++ RING_GET_RESPONSE(&netif->tx, ++i)->status = NETIF_RSP_NULL;
++
++ netif->tx.rsp_prod_pvt = ++i;
++ RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&netif->tx, notify);
++ if (notify)
++ notify_remote_via_irq(netif->irq);
++
++#ifdef CONFIG_XEN_NETDEV_PIPELINED_TRANSMITTER
++ if (i == netif->tx.req_cons) {
++ int more_to_do;
++ RING_FINAL_CHECK_FOR_REQUESTS(&netif->tx, more_to_do);
++ if (more_to_do)
++ add_to_net_schedule_list_tail(netif);
++ }
++#endif
++}
++
++static int make_rx_response(netif_t *netif,
++ u16 id,
++ s8 st,
++ u16 offset,
++ u16 size,
++ u16 flags)
++{
++ RING_IDX i = netif->rx.rsp_prod_pvt;
++ netif_rx_response_t *resp;
++ int notify;
++
++ resp = RING_GET_RESPONSE(&netif->rx, i);
++ resp->offset = offset;
++ resp->flags = flags;
++ resp->id = id;
++ resp->status = (s16)size;
++ if (st < 0)
++ resp->status = (s16)st;
++
++ netif->rx.rsp_prod_pvt = ++i;
++ RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(&netif->rx, notify);
++
++ return notify;
++}
++
++#ifdef NETBE_DEBUG_INTERRUPT
++static irqreturn_t netif_be_dbg(int irq, void *dev_id, struct pt_regs *regs)
++{
++ struct list_head *ent;
++ netif_t *netif;
++ int i = 0;
++
++ printk(KERN_ALERT "netif_schedule_list:\n");
++ spin_lock_irq(&net_schedule_list_lock);
++
++ list_for_each (ent, &net_schedule_list) {
++ netif = list_entry(ent, netif_t, list);
++ printk(KERN_ALERT " %d: private(rx_req_cons=%08x "
++ "rx_resp_prod=%08x\n",
++ i, netif->rx.req_cons, netif->rx.rsp_prod_pvt);
++ printk(KERN_ALERT " tx_req_cons=%08x tx_resp_prod=%08x)\n",
++ netif->tx.req_cons, netif->tx.rsp_prod_pvt);
++ printk(KERN_ALERT " shared(rx_req_prod=%08x "
++ "rx_resp_prod=%08x\n",
++ netif->rx.sring->req_prod, netif->rx.sring->rsp_prod);
++ printk(KERN_ALERT " rx_event=%08x tx_req_prod=%08x\n",
++ netif->rx.sring->rsp_event, netif->tx.sring->req_prod);
++ printk(KERN_ALERT " tx_resp_prod=%08x, tx_event=%08x)\n",
++ netif->tx.sring->rsp_prod, netif->tx.sring->rsp_event);
++ i++;
++ }
++
++ spin_unlock_irq(&net_schedule_list_lock);
++ printk(KERN_ALERT " ** End of netif_schedule_list **\n");
++
++ return IRQ_HANDLED;
++}
++#endif
++
++static int __init netback_init(void)
++{
++ int i;
++ struct page *page;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ /* We can increase reservation by this much in net_rx_action(). */
++ balloon_update_driver_allowance(NET_RX_RING_SIZE);
++
++ skb_queue_head_init(&rx_queue);
++ skb_queue_head_init(&tx_queue);
++
++ init_timer(&net_timer);
++ net_timer.data = 0;
++ net_timer.function = net_alarm;
++
++ page = balloon_alloc_empty_page_range(MAX_PENDING_REQS);
++ BUG_ON(page == NULL);
++ mmap_vstart = (unsigned long)pfn_to_kaddr(page_to_pfn(page));
++
++ for (i = 0; i < MAX_PENDING_REQS; i++) {
++ page = virt_to_page(MMAP_VADDR(i));
++ init_page_count(page);
++ SetPageForeign(page, netif_page_release);
++ }
++
++ pending_cons = 0;
++ pending_prod = MAX_PENDING_REQS;
++ for (i = 0; i < MAX_PENDING_REQS; i++)
++ pending_ring[i] = i;
++
++ spin_lock_init(&net_schedule_list_lock);
++ INIT_LIST_HEAD(&net_schedule_list);
++
++ netif_xenbus_init();
++
++#ifdef NETBE_DEBUG_INTERRUPT
++ (void)bind_virq_to_irqhandler(
++ VIRQ_DEBUG,
++ 0,
++ netif_be_dbg,
++ SA_SHIRQ,
++ "net-be-dbg",
++ &netif_be_dbg);
++#endif
++
++ return 0;
++}
++
++module_init(netback_init);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/netback/xenbus.c new/drivers/xen/netback/xenbus.c
+--- linux-2.6/drivers/xen/netback/xenbus.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netback/xenbus.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,404 @@
++/* Xenbus code for netif backend
++ Copyright (C) 2005 Rusty Russell <rusty@rustcorp.com.au>
++ Copyright (C) 2005 XenSource Ltd
++
++ This program is free software; you can redistribute it and/or modify
++ it under the terms of the GNU General Public License as published by
++ the Free Software Foundation; either version 2 of the License, or
++ (at your option) any later version.
++
++ This program is distributed in the hope that it will be useful,
++ but WITHOUT ANY WARRANTY; without even the implied warranty of
++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ GNU General Public License for more details.
++
++ You should have received a copy of the GNU General Public License
++ along with this program; if not, write to the Free Software
++ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++*/
++
++#include <stdarg.h>
++#include <linux/module.h>
++#include <xen/xenbus.h>
++#include "common.h"
++
++#if 0
++#undef DPRINTK
++#define DPRINTK(fmt, args...) \
++ printk("netback/xenbus (%s:%d) " fmt ".\n", __FUNCTION__, __LINE__, ##args)
++#endif
++
++struct backend_info
++{
++ struct xenbus_device *dev;
++ netif_t *netif;
++ struct xenbus_watch backend_watch;
++ enum xenbus_state frontend_state;
++};
++
++static int connect_rings(struct backend_info *);
++static void connect(struct backend_info *);
++static void maybe_connect(struct backend_info *);
++static void backend_changed(struct xenbus_watch *, const char **,
++ unsigned int);
++
++static int netback_remove(struct xenbus_device *dev)
++{
++ struct backend_info *be = dev->dev.driver_data;
++
++ if (be->backend_watch.node) {
++ unregister_xenbus_watch(&be->backend_watch);
++ kfree(be->backend_watch.node);
++ be->backend_watch.node = NULL;
++ }
++ if (be->netif) {
++ netif_disconnect(be->netif);
++ be->netif = NULL;
++ }
++ kfree(be);
++ dev->dev.driver_data = NULL;
++ return 0;
++}
++
++
++/**
++ * Entry point to this code when a new device is created. Allocate the basic
++ * structures, and watch the store waiting for the hotplug scripts to tell us
++ * the device's handle. Switch to InitWait.
++ */
++static int netback_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ const char *message;
++ struct xenbus_transaction xbt;
++ int err;
++ struct backend_info *be = kzalloc(sizeof(struct backend_info),
++ GFP_KERNEL);
++ if (!be) {
++ xenbus_dev_fatal(dev, -ENOMEM,
++ "allocating backend structure");
++ return -ENOMEM;
++ }
++
++ be->dev = dev;
++ dev->dev.driver_data = be;
++
++ err = xenbus_watch_path2(dev, dev->nodename, "handle",
++ &be->backend_watch, backend_changed);
++ if (err)
++ goto fail;
++
++ do {
++ err = xenbus_transaction_start(&xbt);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "starting transaction");
++ goto fail;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename, "feature-sg", "%d", 1);
++ if (err) {
++ message = "writing feature-sg";
++ goto abort_transaction;
++ }
++
++#if 0 /* KAF: After the protocol is finalised. */
++ err = xenbus_printf(xbt, dev->nodename, "feature-tso", "%d", 1);
++ if (err) {
++ message = "writing feature-tso";
++ goto abort_transaction;
++ }
++#endif
++
++ err = xenbus_transaction_end(xbt, 0);
++ } while (err == -EAGAIN);
++
++ if (err) {
++ xenbus_dev_fatal(dev, err, "completing transaction");
++ goto fail;
++ }
++
++ err = xenbus_switch_state(dev, XenbusStateInitWait);
++ if (err) {
++ goto fail;
++ }
++
++ return 0;
++
++abort_transaction:
++ xenbus_transaction_end(xbt, 1);
++ xenbus_dev_fatal(dev, err, "%s", message);
++fail:
++ DPRINTK("failed");
++ netback_remove(dev);
++ return err;
++}
++
++
++/**
++ * Handle the creation of the hotplug script environment. We add the script
++ * and vif variables to the environment, for the benefit of the vif-* hotplug
++ * scripts.
++ */
++static int netback_uevent(struct xenbus_device *xdev, char **envp,
++ int num_envp, char *buffer, int buffer_size)
++{
++ struct backend_info *be = xdev->dev.driver_data;
++ netif_t *netif = be->netif;
++ int i = 0, length = 0;
++ char *val;
++
++ DPRINTK("netback_uevent");
++
++ val = xenbus_read(XBT_NIL, xdev->nodename, "script", NULL);
++ if (IS_ERR(val)) {
++ int err = PTR_ERR(val);
++ xenbus_dev_fatal(xdev, err, "reading script");
++ return err;
++ }
++ else {
++ add_uevent_var(envp, num_envp, &i, buffer, buffer_size,
++ &length, "script=%s", val);
++ kfree(val);
++ }
++
++ add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
++ "vif=%s", netif->dev->name);
++
++ envp[i] = NULL;
++
++ return 0;
++}
++
++
++/**
++ * Callback received when the hotplug scripts have placed the handle node.
++ * Read it, and create a netif structure. If the frontend is ready, connect.
++ */
++static void backend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ int err;
++ long handle;
++ struct backend_info *be
++ = container_of(watch, struct backend_info, backend_watch);
++ struct xenbus_device *dev = be->dev;
++
++ DPRINTK("");
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename, "handle", "%li", &handle);
++ if (XENBUS_EXIST_ERR(err)) {
++ /* Since this watch will fire once immediately after it is
++ registered, we expect this. Ignore it, and wait for the
++ hotplug scripts. */
++ return;
++ }
++ if (err != 1) {
++ xenbus_dev_fatal(dev, err, "reading handle");
++ return;
++ }
++
++ if (be->netif == NULL) {
++ u8 be_mac[ETH_ALEN] = { 0, 0, 0, 0, 0, 0 };
++
++ be->netif = netif_alloc(dev->otherend_id, handle, be_mac);
++ if (IS_ERR(be->netif)) {
++ err = PTR_ERR(be->netif);
++ be->netif = NULL;
++ xenbus_dev_fatal(dev, err, "creating interface");
++ return;
++ }
++
++ kobject_uevent(&dev->dev.kobj, KOBJ_ONLINE);
++
++ maybe_connect(be);
++ }
++}
++
++
++/**
++ * Callback received when the frontend's state changes.
++ */
++static void frontend_changed(struct xenbus_device *dev,
++ enum xenbus_state frontend_state)
++{
++ struct backend_info *be = dev->dev.driver_data;
++
++ DPRINTK("");
++
++ be->frontend_state = frontend_state;
++
++ switch (frontend_state) {
++ case XenbusStateInitialising:
++ case XenbusStateInitialised:
++ break;
++
++ case XenbusStateConnected:
++ maybe_connect(be);
++ break;
++
++ case XenbusStateClosing:
++ xenbus_switch_state(dev, XenbusStateClosing);
++ break;
++
++ case XenbusStateClosed:
++ if (be->netif != NULL)
++ kobject_uevent(&dev->dev.kobj, KOBJ_OFFLINE);
++ device_unregister(&dev->dev);
++ break;
++
++ case XenbusStateUnknown:
++ case XenbusStateInitWait:
++ default:
++ xenbus_dev_fatal(dev, -EINVAL, "saw state %d at frontend",
++ frontend_state);
++ break;
++ }
++}
++
++
++/* ** Connection ** */
++
++
++static void maybe_connect(struct backend_info *be)
++{
++ if (be->netif && (be->frontend_state == XenbusStateConnected))
++ connect(be);
++}
++
++static void xen_net_read_rate(struct xenbus_device *dev,
++ unsigned long *bytes, unsigned long *usec)
++{
++ char *s, *e;
++ unsigned long b, u;
++ char *ratestr;
++
++ /* Default to unlimited bandwidth. */
++ *bytes = ~0UL;
++ *usec = 0;
++
++ ratestr = xenbus_read(XBT_NIL, dev->nodename, "rate", NULL);
++ if (IS_ERR(ratestr))
++ return;
++
++ s = ratestr;
++ b = simple_strtoul(s, &e, 10);
++ if ((s == e) || (*e != ','))
++ goto fail;
++
++ s = e + 1;
++ u = simple_strtoul(s, &e, 10);
++ if ((s == e) || (*e != '\0'))
++ goto fail;
++
++ *bytes = b;
++ *usec = u;
++
++ kfree(ratestr);
++ return;
++
++ fail:
++ WPRINTK("Failed to parse network rate limit. Traffic unlimited.\n");
++ kfree(ratestr);
++}
++
++static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
++{
++ char *s, *e, *macstr;
++ int i;
++
++ macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL);
++ if (IS_ERR(macstr))
++ return PTR_ERR(macstr);
++
++ for (i = 0; i < ETH_ALEN; i++) {
++ mac[i] = simple_strtoul(s, &e, 16);
++ if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) {
++ kfree(macstr);
++ return -ENOENT;
++ }
++ s = e+1;
++ }
++
++ kfree(macstr);
++ return 0;
++}
++
++static void connect(struct backend_info *be)
++{
++ int err;
++ struct xenbus_device *dev = be->dev;
++
++ err = connect_rings(be);
++ if (err)
++ return;
++
++ err = xen_net_read_mac(dev, be->netif->fe_dev_addr);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
++ return;
++ }
++
++ xen_net_read_rate(dev, &be->netif->credit_bytes,
++ &be->netif->credit_usec);
++ be->netif->remaining_credit = be->netif->credit_bytes;
++
++ xenbus_switch_state(dev, XenbusStateConnected);
++}
++
++
++static int connect_rings(struct backend_info *be)
++{
++ struct xenbus_device *dev = be->dev;
++ unsigned long tx_ring_ref, rx_ring_ref;
++ unsigned int evtchn;
++ int err;
++
++ DPRINTK("");
++
++ err = xenbus_gather(XBT_NIL, dev->otherend,
++ "tx-ring-ref", "%lu", &tx_ring_ref,
++ "rx-ring-ref", "%lu", &rx_ring_ref,
++ "event-channel", "%u", &evtchn, NULL);
++ if (err) {
++ xenbus_dev_fatal(dev, err,
++ "reading %s/ring-ref and event-channel",
++ dev->otherend);
++ return err;
++ }
++
++ /* Map the shared frame, irq etc. */
++ err = netif_map(be->netif, tx_ring_ref, rx_ring_ref, evtchn);
++ if (err) {
++ xenbus_dev_fatal(dev, err,
++ "mapping shared-frames %lu/%lu port %u",
++ tx_ring_ref, rx_ring_ref, evtchn);
++ return err;
++ }
++ return 0;
++}
++
++
++/* ** Driver Registration ** */
++
++
++static struct xenbus_device_id netback_ids[] = {
++ { "vif" },
++ { "" }
++};
++
++
++static struct xenbus_driver netback = {
++ .name = "vif",
++ .owner = THIS_MODULE,
++ .ids = netback_ids,
++ .probe = netback_probe,
++ .remove = netback_remove,
++ .uevent = netback_uevent,
++ .otherend_changed = frontend_changed,
++};
++
++
++void netif_xenbus_init(void)
++{
++ xenbus_register_backend(&netback);
++}
+diff -urNp linux-2.6/drivers/xen/netfront/Kconfig new/drivers/xen/netfront/Kconfig
+--- linux-2.6/drivers/xen/netfront/Kconfig 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netfront/Kconfig 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,6 @@
++
++config XENNET
++ tristate "Xen network driver"
++ depends on NETDEVICES && ARCH_XEN
++ help
++ Network driver for Xen
+diff -urNp linux-2.6/drivers/xen/netfront/Makefile new/drivers/xen/netfront/Makefile
+--- linux-2.6/drivers/xen/netfront/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netfront/Makefile 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,4 @@
++
++obj-$(CONFIG_XEN_NETDEV_FRONTEND) := xennet.o
++
++xennet-objs := netfront.o
+diff -urNp linux-2.6/drivers/xen/netfront/netfront.c new/drivers/xen/netfront/netfront.c
+--- linux-2.6/drivers/xen/netfront/netfront.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/netfront/netfront.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,1599 @@
++/******************************************************************************
++ * Virtual network driver for conversing with remote driver backends.
++ *
++ * Copyright (c) 2002-2005, K A Fraser
++ * Copyright (c) 2005, XenSource Ltd
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/module.h>
++#include <linux/version.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/netdevice.h>
++#include <linux/inetdevice.h>
++#include <linux/etherdevice.h>
++#include <linux/skbuff.h>
++#include <linux/init.h>
++#include <linux/bitops.h>
++#include <linux/ethtool.h>
++#include <linux/in.h>
++#include <linux/if_ether.h>
++#include <net/sock.h>
++#include <net/pkt_sched.h>
++#include <net/arp.h>
++#include <net/route.h>
++#include <asm/io.h>
++#include <asm/uaccess.h>
++#include <xen/evtchn.h>
++#include <xen/xenbus.h>
++#include <xen/interface/io/netif.h>
++#include <xen/interface/memory.h>
++#include <xen/balloon.h>
++#include <asm/page.h>
++#include <asm/uaccess.h>
++#include <xen/interface/grant_table.h>
++#include <xen/gnttab.h>
++
++#define GRANT_INVALID_REF 0
++
++#define NET_TX_RING_SIZE __RING_SIZE((struct netif_tx_sring *)0, PAGE_SIZE)
++#define NET_RX_RING_SIZE __RING_SIZE((struct netif_rx_sring *)0, PAGE_SIZE)
++
++static inline void init_skb_shinfo(struct sk_buff *skb)
++{
++ atomic_set(&(skb_shinfo(skb)->dataref), 1);
++ skb_shinfo(skb)->nr_frags = 0;
++ skb_shinfo(skb)->frag_list = NULL;
++}
++
++struct netfront_info {
++ struct list_head list;
++ struct net_device *netdev;
++
++ struct net_device_stats stats;
++
++ struct netif_tx_front_ring tx;
++ struct netif_rx_front_ring rx;
++
++ spinlock_t tx_lock;
++ spinlock_t rx_lock;
++
++ unsigned int handle;
++ unsigned int evtchn, irq;
++
++ /* Receive-ring batched refills. */
++#define RX_MIN_TARGET 8
++#define RX_DFL_MIN_TARGET 64
++#define RX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256)
++ unsigned rx_min_target, rx_max_target, rx_target;
++ struct sk_buff_head rx_batch;
++
++ struct timer_list rx_refill_timer;
++
++ /*
++ * {tx,rx}_skbs store outstanding skbuffs. The first entry in each
++ * array is an index into a chain of free entries.
++ */
++ struct sk_buff *tx_skbs[NET_TX_RING_SIZE+1];
++ struct sk_buff *rx_skbs[NET_RX_RING_SIZE+1];
++
++#define TX_MAX_TARGET min_t(int, NET_RX_RING_SIZE, 256)
++ grant_ref_t gref_tx_head;
++ grant_ref_t grant_tx_ref[NET_TX_RING_SIZE + 1];
++ grant_ref_t gref_rx_head;
++ grant_ref_t grant_rx_ref[NET_TX_RING_SIZE + 1];
++
++ struct xenbus_device *xbdev;
++ int tx_ring_ref;
++ int rx_ring_ref;
++ u8 mac[ETH_ALEN];
++
++ unsigned long rx_pfn_array[NET_RX_RING_SIZE];
++ struct multicall_entry rx_mcl[NET_RX_RING_SIZE+1];
++ struct mmu_update rx_mmu[NET_RX_RING_SIZE];
++};
++
++/*
++ * Access macros for acquiring freeing slots in {tx,rx}_skbs[].
++ */
++
++static inline void add_id_to_freelist(struct sk_buff **list, unsigned short id)
++{
++ list[id] = list[0];
++ list[0] = (void *)(unsigned long)id;
++}
++
++static inline unsigned short get_id_from_freelist(struct sk_buff **list)
++{
++ unsigned int id = (unsigned int)(unsigned long)list[0];
++ list[0] = list[id];
++ return id;
++}
++
++#define DPRINTK(fmt, args...) \
++ pr_debug("netfront (%s:%d) " fmt, \
++ __FUNCTION__, __LINE__, ##args)
++#define IPRINTK(fmt, args...) \
++ printk(KERN_INFO "netfront: " fmt, ##args)
++#define WPRINTK(fmt, args...) \
++ printk(KERN_WARNING "netfront: " fmt, ##args)
++
++static int talk_to_backend(struct xenbus_device *, struct netfront_info *);
++static int setup_device(struct xenbus_device *, struct netfront_info *);
++static struct net_device *create_netdev(int, struct xenbus_device *);
++
++static void netfront_closing(struct xenbus_device *);
++
++static void end_access(int, void *);
++static void netif_disconnect_backend(struct netfront_info *);
++static void close_netdev(struct netfront_info *);
++static void netif_free(struct netfront_info *);
++
++static void network_connect(struct net_device *);
++static void network_tx_buf_gc(struct net_device *);
++static void network_alloc_rx_buffers(struct net_device *);
++static int send_fake_arp(struct net_device *);
++
++static irqreturn_t netif_int(int irq, void *dev_id, struct pt_regs *ptregs);
++
++#ifdef CONFIG_SYSFS
++static int xennet_sysfs_addif(struct net_device *netdev);
++static void xennet_sysfs_delif(struct net_device *netdev);
++#else /* !CONFIG_SYSFS */
++#define xennet_sysfs_addif(dev) (0)
++#define xennet_sysfs_delif(dev) do { } while(0)
++#endif
++
++static inline int xennet_can_sg(struct net_device *dev)
++{
++ return dev->features & NETIF_F_SG;
++}
++
++/**
++ * Entry point to this code when a new device is created. Allocate the basic
++ * structures and the ring buffers for communication with the backend, and
++ * inform the backend of the appropriate details for those. Switch to
++ * Connected state.
++ */
++static int __devinit netfront_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err;
++ struct net_device *netdev;
++ struct netfront_info *info;
++ unsigned int handle;
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename, "handle", "%u", &handle);
++ if (err != 1) {
++ xenbus_dev_fatal(dev, err, "reading handle");
++ return err;
++ }
++
++ netdev = create_netdev(handle, dev);
++ if (IS_ERR(netdev)) {
++ err = PTR_ERR(netdev);
++ xenbus_dev_fatal(dev, err, "creating netdev");
++ return err;
++ }
++
++ info = netdev_priv(netdev);
++ dev->dev.driver_data = info;
++
++ err = talk_to_backend(dev, info);
++ if (err) {
++ xennet_sysfs_delif(info->netdev);
++ unregister_netdev(netdev);
++ free_netdev(netdev);
++ dev->dev.driver_data = NULL;
++ return err;
++ }
++
++ return 0;
++}
++
++
++/**
++ * We are reconnecting to the backend, due to a suspend/resume, or a backend
++ * driver restart. We tear down our netif structure and recreate it, but
++ * leave the device-layer structures intact so that this is transparent to the
++ * rest of the kernel.
++ */
++static int netfront_resume(struct xenbus_device *dev)
++{
++ struct netfront_info *info = dev->dev.driver_data;
++
++ DPRINTK("%s\n", dev->nodename);
++
++ netif_disconnect_backend(info);
++ return talk_to_backend(dev, info);
++}
++
++static int xen_net_read_mac(struct xenbus_device *dev, u8 mac[])
++{
++ char *s, *e, *macstr;
++ int i;
++
++ macstr = s = xenbus_read(XBT_NIL, dev->nodename, "mac", NULL);
++ if (IS_ERR(macstr))
++ return PTR_ERR(macstr);
++
++ for (i = 0; i < ETH_ALEN; i++) {
++ mac[i] = simple_strtoul(s, &e, 16);
++ if ((s == e) || (*e != ((i == ETH_ALEN-1) ? '\0' : ':'))) {
++ kfree(macstr);
++ return -ENOENT;
++ }
++ s = e+1;
++ }
++
++ kfree(macstr);
++ return 0;
++}
++
++/* Common code used when first setting up, and when resuming. */
++static int talk_to_backend(struct xenbus_device *dev,
++ struct netfront_info *info)
++{
++ const char *message;
++ struct xenbus_transaction xbt;
++ int err;
++
++ err = xen_net_read_mac(dev, info->mac);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "parsing %s/mac", dev->nodename);
++ goto out;
++ }
++
++ /* Create shared ring, alloc event channel. */
++ err = setup_device(dev, info);
++ if (err)
++ goto out;
++
++again:
++ err = xenbus_transaction_start(&xbt);
++ if (err) {
++ xenbus_dev_fatal(dev, err, "starting transaction");
++ goto destroy_ring;
++ }
++
++ err = xenbus_printf(xbt, dev->nodename, "tx-ring-ref","%u",
++ info->tx_ring_ref);
++ if (err) {
++ message = "writing tx ring-ref";
++ goto abort_transaction;
++ }
++ err = xenbus_printf(xbt, dev->nodename, "rx-ring-ref","%u",
++ info->rx_ring_ref);
++ if (err) {
++ message = "writing rx ring-ref";
++ goto abort_transaction;
++ }
++ err = xenbus_printf(xbt, dev->nodename,
++ "event-channel", "%u", info->evtchn);
++ if (err) {
++ message = "writing event-channel";
++ goto abort_transaction;
++ }
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err) {
++ if (err == -EAGAIN)
++ goto again;
++ xenbus_dev_fatal(dev, err, "completing transaction");
++ goto destroy_ring;
++ }
++
++ return 0;
++
++ abort_transaction:
++ xenbus_transaction_end(xbt, 1);
++ xenbus_dev_fatal(dev, err, "%s", message);
++ destroy_ring:
++ netif_free(info);
++ out:
++ return err;
++}
++
++
++static int setup_device(struct xenbus_device *dev, struct netfront_info *info)
++{
++ struct netif_tx_sring *txs;
++ struct netif_rx_sring *rxs;
++ int err;
++ struct net_device *netdev = info->netdev;
++
++ info->tx_ring_ref = GRANT_INVALID_REF;
++ info->rx_ring_ref = GRANT_INVALID_REF;
++ info->rx.sring = NULL;
++ info->tx.sring = NULL;
++ info->irq = 0;
++
++ txs = (struct netif_tx_sring *)get_zeroed_page(GFP_KERNEL);
++ if (!txs) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(dev, err, "allocating tx ring page");
++ goto fail;
++ }
++ SHARED_RING_INIT(txs);
++ FRONT_RING_INIT(&info->tx, txs, PAGE_SIZE);
++
++ err = xenbus_grant_ring(dev, virt_to_mfn(txs));
++ if (err < 0) {
++ free_page((unsigned long)txs);
++ goto fail;
++ }
++ info->tx_ring_ref = err;
++
++ rxs = (struct netif_rx_sring *)get_zeroed_page(GFP_KERNEL);
++ if (!rxs) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(dev, err, "allocating rx ring page");
++ goto fail;
++ }
++ SHARED_RING_INIT(rxs);
++ FRONT_RING_INIT(&info->rx, rxs, PAGE_SIZE);
++
++ err = xenbus_grant_ring(dev, virt_to_mfn(rxs));
++ if (err < 0) {
++ free_page((unsigned long)rxs);
++ goto fail;
++ }
++ info->rx_ring_ref = err;
++
++ err = xenbus_alloc_evtchn(dev, &info->evtchn);
++ if (err)
++ goto fail;
++
++ memcpy(netdev->dev_addr, info->mac, ETH_ALEN);
++ err = bind_evtchn_to_irqhandler(info->evtchn, netif_int,
++ SA_SAMPLE_RANDOM, netdev->name, netdev);
++ if (err < 0)
++ goto fail;
++ info->irq = err;
++ return 0;
++
++ fail:
++ netif_free(info);
++ return err;
++}
++
++
++/**
++ * Callback received when the backend's state changes.
++ */
++static void backend_changed(struct xenbus_device *dev,
++ enum xenbus_state backend_state)
++{
++ struct netfront_info *np = dev->dev.driver_data;
++ struct net_device *netdev = np->netdev;
++
++ DPRINTK("\n");
++
++ switch (backend_state) {
++ case XenbusStateInitialising:
++ case XenbusStateInitialised:
++ case XenbusStateConnected:
++ case XenbusStateUnknown:
++ case XenbusStateClosed:
++ break;
++
++ case XenbusStateInitWait:
++ network_connect(netdev);
++ xenbus_switch_state(dev, XenbusStateConnected);
++ (void)send_fake_arp(netdev);
++ break;
++
++ case XenbusStateClosing:
++ netfront_closing(dev);
++ break;
++ }
++}
++
++
++/** Send a packet on a net device to encourage switches to learn the
++ * MAC. We send a fake ARP request.
++ *
++ * @param dev device
++ * @return 0 on success, error code otherwise
++ */
++static int send_fake_arp(struct net_device *dev)
++{
++ struct sk_buff *skb;
++ u32 src_ip, dst_ip;
++
++ dst_ip = INADDR_BROADCAST;
++ src_ip = inet_select_addr(dev, dst_ip, RT_SCOPE_LINK);
++
++ /* No IP? Then nothing to do. */
++ if (src_ip == 0)
++ return 0;
++
++ skb = arp_create(ARPOP_REPLY, ETH_P_ARP,
++ dst_ip, dev, src_ip,
++ /*dst_hw*/ NULL, /*src_hw*/ NULL,
++ /*target_hw*/ dev->dev_addr);
++ if (skb == NULL)
++ return -ENOMEM;
++
++ return dev_queue_xmit(skb);
++}
++
++
++static int network_open(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++
++ memset(&np->stats, 0, sizeof(np->stats));
++
++ network_alloc_rx_buffers(dev);
++ np->rx.sring->rsp_event = np->rx.rsp_cons + 1;
++
++ netif_start_queue(dev);
++
++ return 0;
++}
++
++static inline int netfront_tx_slot_available(struct netfront_info *np)
++{
++ return RING_FREE_REQUESTS(&np->tx) >= MAX_SKB_FRAGS + 2;
++}
++
++static inline void network_maybe_wake_tx(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++
++ if (unlikely(netif_queue_stopped(dev)) &&
++ netfront_tx_slot_available(np) &&
++ likely(netif_running(dev)))
++ netif_wake_queue(dev);
++}
++
++static void network_tx_buf_gc(struct net_device *dev)
++{
++ RING_IDX cons, prod;
++ unsigned short id;
++ struct netfront_info *np = netdev_priv(dev);
++ struct sk_buff *skb;
++
++ if (unlikely(!netif_carrier_ok(dev)))
++ return;
++
++ do {
++ prod = np->tx.sring->rsp_prod;
++ rmb(); /* Ensure we see responses up to 'rp'. */
++
++ for (cons = np->tx.rsp_cons; cons != prod; cons++) {
++ struct netif_tx_response *txrsp;
++
++ txrsp = RING_GET_RESPONSE(&np->tx, cons);
++ if (txrsp->status == NETIF_RSP_NULL)
++ continue;
++
++ id = txrsp->id;
++ skb = np->tx_skbs[id];
++ if (unlikely(gnttab_query_foreign_access(
++ np->grant_tx_ref[id]) != 0)) {
++ printk(KERN_ALERT "network_tx_buf_gc: warning "
++ "-- grant still in use by backend "
++ "domain.\n");
++ BUG();
++ }
++ gnttab_end_foreign_access_ref(
++ np->grant_tx_ref[id], GNTMAP_readonly);
++ gnttab_release_grant_reference(
++ &np->gref_tx_head, np->grant_tx_ref[id]);
++ np->grant_tx_ref[id] = GRANT_INVALID_REF;
++ add_id_to_freelist(np->tx_skbs, id);
++ dev_kfree_skb_irq(skb);
++ }
++
++ np->tx.rsp_cons = prod;
++
++ /*
++ * Set a new event, then check for race with update of tx_cons.
++ * Note that it is essential to schedule a callback, no matter
++ * how few buffers are pending. Even if there is space in the
++ * transmit ring, higher layers may be blocked because too much
++ * data is outstanding: in such cases notification from Xen is
++ * likely to be the only kick that we'll get.
++ */
++ np->tx.sring->rsp_event =
++ prod + ((np->tx.sring->req_prod - prod) >> 1) + 1;
++ mb();
++ } while ((cons == prod) && (prod != np->tx.sring->rsp_prod));
++
++ network_maybe_wake_tx(dev);
++}
++
++
++static void rx_refill_timeout(unsigned long data)
++{
++ struct net_device *dev = (struct net_device *)data;
++ netif_rx_schedule(dev);
++}
++
++
++static void network_alloc_rx_buffers(struct net_device *dev)
++{
++ unsigned short id;
++ struct netfront_info *np = netdev_priv(dev);
++ struct sk_buff *skb;
++ int i, batch_target;
++ RING_IDX req_prod = np->rx.req_prod_pvt;
++ struct xen_memory_reservation reservation;
++ grant_ref_t ref;
++
++ if (unlikely(!netif_carrier_ok(dev)))
++ return;
++
++ /*
++ * Allocate skbuffs greedily, even though we batch updates to the
++ * receive ring. This creates a less bursty demand on the memory
++ * allocator, so should reduce the chance of failed allocation requests
++ * both for ourself and for other kernel subsystems.
++ */
++ batch_target = np->rx_target - (req_prod - np->rx.rsp_cons);
++ for (i = skb_queue_len(&np->rx_batch); i < batch_target; i++) {
++ /*
++ * Subtract dev_alloc_skb headroom (16 bytes) and shared info
++ * tailroom then round down to SKB_DATA_ALIGN boundary.
++ */
++ skb = __dev_alloc_skb(
++ ((PAGE_SIZE - sizeof(struct skb_shared_info)) &
++ (-SKB_DATA_ALIGN(1))) - 16,
++ GFP_ATOMIC|__GFP_NOWARN);
++ if (skb == NULL) {
++ /* Any skbuffs queued for refill? Force them out. */
++ if (i != 0)
++ goto refill;
++ /* Could not allocate any skbuffs. Try again later. */
++ mod_timer(&np->rx_refill_timer,
++ jiffies + (HZ/10));
++ return;
++ }
++ __skb_queue_tail(&np->rx_batch, skb);
++ }
++
++ /* Is the batch large enough to be worthwhile? */
++ if (i < (np->rx_target/2))
++ return;
++
++ /* Adjust our fill target if we risked running out of buffers. */
++ if (((req_prod - np->rx.sring->rsp_prod) < (np->rx_target / 4)) &&
++ ((np->rx_target *= 2) > np->rx_max_target))
++ np->rx_target = np->rx_max_target;
++
++ refill:
++ for (i = 0; ; i++) {
++ if ((skb = __skb_dequeue(&np->rx_batch)) == NULL)
++ break;
++
++ skb->dev = dev;
++
++ id = get_id_from_freelist(np->rx_skbs);
++
++ np->rx_skbs[id] = skb;
++
++ RING_GET_REQUEST(&np->rx, req_prod + i)->id = id;
++ ref = gnttab_claim_grant_reference(&np->gref_rx_head);
++ BUG_ON((signed short)ref < 0);
++ np->grant_rx_ref[id] = ref;
++ gnttab_grant_foreign_transfer_ref(ref,
++ np->xbdev->otherend_id,
++ __pa(skb->head)>>PAGE_SHIFT);
++ RING_GET_REQUEST(&np->rx, req_prod + i)->gref = ref;
++ np->rx_pfn_array[i] = virt_to_mfn(skb->head);
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* Remove this page before passing back to Xen. */
++ set_phys_to_machine(__pa(skb->head) >> PAGE_SHIFT,
++ INVALID_P2M_ENTRY);
++ MULTI_update_va_mapping(np->rx_mcl+i,
++ (unsigned long)skb->head,
++ __pte(0), 0);
++ }
++ }
++
++ /* Tell the ballon driver what is going on. */
++ balloon_update_driver_allowance(i);
++
++ set_xen_guest_handle(reservation.extent_start, np->rx_pfn_array);
++ reservation.nr_extents = i;
++ reservation.extent_order = 0;
++ reservation.address_bits = 0;
++ reservation.domid = DOMID_SELF;
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* After all PTEs have been zapped, flush the TLB. */
++ np->rx_mcl[i-1].args[MULTI_UVMFLAGS_INDEX] =
++ UVMF_TLB_FLUSH|UVMF_ALL;
++
++ /* Give away a batch of pages. */
++ np->rx_mcl[i].op = __HYPERVISOR_memory_op;
++ np->rx_mcl[i].args[0] = XENMEM_decrease_reservation;
++ np->rx_mcl[i].args[1] = (unsigned long)&reservation;
++
++ /* Zap PTEs and give away pages in one big multicall. */
++ (void)HYPERVISOR_multicall(np->rx_mcl, i+1);
++
++ /* Check return status of HYPERVISOR_memory_op(). */
++ if (unlikely(np->rx_mcl[i].result != i))
++ panic("Unable to reduce memory reservation\n");
++ } else
++ if (HYPERVISOR_memory_op(XENMEM_decrease_reservation,
++ &reservation) != i)
++ panic("Unable to reduce memory reservation\n");
++
++ /* Above is a suitable barrier to ensure backend will see requests. */
++ np->rx.req_prod_pvt = req_prod + i;
++ RING_PUSH_REQUESTS(&np->rx);
++}
++
++static void xennet_make_frags(struct sk_buff *skb, struct net_device *dev,
++ struct netif_tx_request *tx)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ char *data = skb->data;
++ unsigned long mfn;
++ RING_IDX prod = np->tx.req_prod_pvt;
++ int frags = skb_shinfo(skb)->nr_frags;
++ unsigned int offset = offset_in_page(data);
++ unsigned int len = skb_headlen(skb);
++ unsigned int id;
++ grant_ref_t ref;
++ int i;
++
++ while (len > PAGE_SIZE - offset) {
++ tx->size = PAGE_SIZE - offset;
++ tx->flags |= NETTXF_more_data;
++ len -= tx->size;
++ data += tx->size;
++ offset = 0;
++
++ id = get_id_from_freelist(np->tx_skbs);
++ np->tx_skbs[id] = skb_get(skb);
++ tx = RING_GET_REQUEST(&np->tx, prod++);
++ tx->id = id;
++ ref = gnttab_claim_grant_reference(&np->gref_tx_head);
++ BUG_ON((signed short)ref < 0);
++
++ mfn = virt_to_mfn(data);
++ gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id,
++ mfn, GNTMAP_readonly);
++
++ tx->gref = np->grant_tx_ref[id] = ref;
++ tx->offset = offset;
++ tx->size = len;
++ tx->flags = 0;
++ }
++
++ for (i = 0; i < frags; i++) {
++ skb_frag_t *frag = skb_shinfo(skb)->frags + i;
++
++ tx->flags |= NETTXF_more_data;
++
++ id = get_id_from_freelist(np->tx_skbs);
++ np->tx_skbs[id] = skb_get(skb);
++ tx = RING_GET_REQUEST(&np->tx, prod++);
++ tx->id = id;
++ ref = gnttab_claim_grant_reference(&np->gref_tx_head);
++ BUG_ON((signed short)ref < 0);
++
++ mfn = pfn_to_mfn(page_to_pfn(frag->page));
++ gnttab_grant_foreign_access_ref(ref, np->xbdev->otherend_id,
++ mfn, GNTMAP_readonly);
++
++ tx->gref = np->grant_tx_ref[id] = ref;
++ tx->offset = frag->page_offset;
++ tx->size = frag->size;
++ tx->flags = 0;
++ }
++
++ np->tx.req_prod_pvt = prod;
++}
++
++static int network_start_xmit(struct sk_buff *skb, struct net_device *dev)
++{
++ unsigned short id;
++ struct netfront_info *np = netdev_priv(dev);
++ struct netif_tx_request *tx;
++ struct netif_extra_info *extra;
++ char *data = skb->data;
++ RING_IDX i;
++ grant_ref_t ref;
++ unsigned long mfn;
++ int notify;
++ int frags = skb_shinfo(skb)->nr_frags;
++ unsigned int offset = offset_in_page(data);
++ unsigned int len = skb_headlen(skb);
++
++ frags += (offset + len + PAGE_SIZE - 1) / PAGE_SIZE;
++ if (unlikely(frags > MAX_SKB_FRAGS + 1)) {
++ printk(KERN_ALERT "xennet: skb rides the rocket: %d frags\n",
++ frags);
++ dump_stack();
++ goto drop;
++ }
++
++ spin_lock_irq(&np->tx_lock);
++
++ if (unlikely(!netif_carrier_ok(dev) ||
++ (frags > 1 && !xennet_can_sg(dev)) ||
++ netif_needs_gso(dev, skb))) {
++ spin_unlock_irq(&np->tx_lock);
++ goto drop;
++ }
++
++ i = np->tx.req_prod_pvt;
++
++ id = get_id_from_freelist(np->tx_skbs);
++ np->tx_skbs[id] = skb;
++
++ tx = RING_GET_REQUEST(&np->tx, i);
++
++ tx->id = id;
++ ref = gnttab_claim_grant_reference(&np->gref_tx_head);
++ BUG_ON((signed short)ref < 0);
++ mfn = virt_to_mfn(data);
++ gnttab_grant_foreign_access_ref(
++ ref, np->xbdev->otherend_id, mfn, GNTMAP_readonly);
++ tx->gref = np->grant_tx_ref[id] = ref;
++ tx->offset = offset;
++ tx->size = len;
++
++ tx->flags = 0;
++ extra = NULL;
++
++ if (skb->ip_summed == CHECKSUM_HW) /* local packet? */
++ tx->flags |= NETTXF_csum_blank | NETTXF_data_validated;
++ if (skb->proto_data_valid) /* remote but checksummed? */
++ tx->flags |= NETTXF_data_validated;
++
++ if (skb_shinfo(skb)->gso_size) {
++ struct netif_extra_info *gso = (struct netif_extra_info *)
++ RING_GET_REQUEST(&np->tx, ++i);
++
++ if (extra)
++ extra->flags |= XEN_NETIF_EXTRA_FLAG_MORE;
++ else
++ tx->flags |= NETTXF_extra_info;
++
++ gso->u.gso.size = skb_shinfo(skb)->gso_size;
++ gso->u.gso.type = XEN_NETIF_GSO_TCPV4;
++
++ gso->type = XEN_NETIF_EXTRA_TYPE_GSO;
++ gso->flags = 0;
++ extra = gso;
++ }
++
++ np->tx.req_prod_pvt = i + 1;
++
++ xennet_make_frags(skb, dev, tx);
++ tx->size = skb->len;
++
++ RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(&np->tx, notify);
++ if (notify)
++ notify_remote_via_irq(np->irq);
++
++ network_tx_buf_gc(dev);
++
++ if (!netfront_tx_slot_available(np))
++ netif_stop_queue(dev);
++
++ spin_unlock_irq(&np->tx_lock);
++
++ np->stats.tx_bytes += skb->len;
++ np->stats.tx_packets++;
++
++ return 0;
++
++ drop:
++ np->stats.tx_dropped++;
++ dev_kfree_skb(skb);
++ return 0;
++}
++
++static irqreturn_t netif_int(int irq, void *dev_id, struct pt_regs *ptregs)
++{
++ struct net_device *dev = dev_id;
++ struct netfront_info *np = netdev_priv(dev);
++ unsigned long flags;
++
++ spin_lock_irqsave(&np->tx_lock, flags);
++ network_tx_buf_gc(dev);
++ spin_unlock_irqrestore(&np->tx_lock, flags);
++
++ if (RING_HAS_UNCONSUMED_RESPONSES(&np->rx) &&
++ likely(netif_running(dev)))
++ netif_rx_schedule(dev);
++
++ return IRQ_HANDLED;
++}
++
++
++static int netif_poll(struct net_device *dev, int *pbudget)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ struct sk_buff *skb, *nskb;
++ struct netif_rx_response *rx;
++ RING_IDX i, rp;
++ struct mmu_update *mmu = np->rx_mmu;
++ struct multicall_entry *mcl = np->rx_mcl;
++ int work_done, budget, more_to_do = 1;
++ struct sk_buff_head rxq;
++ unsigned long flags;
++ unsigned long mfn;
++ grant_ref_t ref;
++
++ spin_lock(&np->rx_lock);
++
++ if (unlikely(!netif_carrier_ok(dev))) {
++ spin_unlock(&np->rx_lock);
++ return 0;
++ }
++
++ skb_queue_head_init(&rxq);
++
++ if ((budget = *pbudget) > dev->quota)
++ budget = dev->quota;
++ rp = np->rx.sring->rsp_prod;
++ rmb(); /* Ensure we see queued responses up to 'rp'. */
++
++ for (i = np->rx.rsp_cons, work_done = 0;
++ (i != rp) && (work_done < budget);
++ i++, work_done++) {
++ rx = RING_GET_RESPONSE(&np->rx, i);
++
++ /*
++ * This definitely indicates a bug, either in this driver or in
++ * the backend driver. In future this should flag the bad
++ * situation to the system controller to reboot the backed.
++ */
++ if ((ref = np->grant_rx_ref[rx->id]) == GRANT_INVALID_REF) {
++ WPRINTK("Bad rx response id %d.\n", rx->id);
++ work_done--;
++ continue;
++ }
++
++ /* Memory pressure, insufficient buffer headroom, ... */
++ if ((mfn = gnttab_end_foreign_transfer_ref(ref)) == 0) {
++ if (net_ratelimit())
++ WPRINTK("Unfulfilled rx req (id=%d, st=%d).\n",
++ rx->id, rx->status);
++ RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->id =
++ rx->id;
++ RING_GET_REQUEST(&np->rx, np->rx.req_prod_pvt)->gref =
++ ref;
++ np->rx.req_prod_pvt++;
++ RING_PUSH_REQUESTS(&np->rx);
++ work_done--;
++ continue;
++ }
++
++ gnttab_release_grant_reference(&np->gref_rx_head, ref);
++ np->grant_rx_ref[rx->id] = GRANT_INVALID_REF;
++
++ skb = np->rx_skbs[rx->id];
++ add_id_to_freelist(np->rx_skbs, rx->id);
++
++ /* NB. We handle skb overflow later. */
++ skb->data = skb->head + rx->offset;
++ skb->len = rx->status;
++ skb->tail = skb->data + skb->len;
++
++ /*
++ * Old backends do not assert data_validated but we
++ * can infer it from csum_blank so test both flags.
++ */
++ if (rx->flags & (NETRXF_data_validated|NETRXF_csum_blank)) {
++ skb->ip_summed = CHECKSUM_UNNECESSARY;
++ skb->proto_data_valid = 1;
++ } else {
++ skb->ip_summed = CHECKSUM_NONE;
++ skb->proto_data_valid = 0;
++ }
++ skb->proto_csum_blank = !!(rx->flags & NETRXF_csum_blank);
++
++ np->stats.rx_packets++;
++ np->stats.rx_bytes += rx->status;
++
++ if (!xen_feature(XENFEAT_auto_translated_physmap)) {
++ /* Remap the page. */
++ MULTI_update_va_mapping(mcl, (unsigned long)skb->head,
++ pfn_pte_ma(mfn, PAGE_KERNEL),
++ 0);
++ mcl++;
++ mmu->ptr = ((maddr_t)mfn << PAGE_SHIFT)
++ | MMU_MACHPHYS_UPDATE;
++ mmu->val = __pa(skb->head) >> PAGE_SHIFT;
++ mmu++;
++
++ set_phys_to_machine(__pa(skb->head) >> PAGE_SHIFT,
++ mfn);
++ }
++
++ __skb_queue_tail(&rxq, skb);
++ }
++
++ /* Some pages are no longer absent... */
++ balloon_update_driver_allowance(-work_done);
++
++ /* Do all the remapping work, and M2P updates, in one big hypercall. */
++ if (likely((mcl - np->rx_mcl) != 0)) {
++ mcl->op = __HYPERVISOR_mmu_update;
++ mcl->args[0] = (unsigned long)np->rx_mmu;
++ mcl->args[1] = mmu - np->rx_mmu;
++ mcl->args[2] = 0;
++ mcl->args[3] = DOMID_SELF;
++ mcl++;
++ (void)HYPERVISOR_multicall(np->rx_mcl, mcl - np->rx_mcl);
++ }
++
++ while ((skb = __skb_dequeue(&rxq)) != NULL) {
++ if (skb->len > (dev->mtu + ETH_HLEN + 4)) {
++ if (net_ratelimit())
++ printk(KERN_INFO "Received packet too big for "
++ "MTU (%d > %d)\n",
++ skb->len - ETH_HLEN - 4, dev->mtu);
++ skb->len = 0;
++ skb->tail = skb->data;
++ init_skb_shinfo(skb);
++ dev_kfree_skb(skb);
++ continue;
++ }
++
++ /*
++ * Enough room in skbuff for the data we were passed? Also,
++ * Linux expects at least 16 bytes headroom in each rx buffer.
++ */
++ if (unlikely(skb->tail > skb->end) ||
++ unlikely((skb->data - skb->head) < 16)) {
++ if (net_ratelimit()) {
++ if (skb->tail > skb->end)
++ printk(KERN_INFO "Received packet "
++ "is %zd bytes beyond tail.\n",
++ skb->tail - skb->end);
++ else
++ printk(KERN_INFO "Received packet "
++ "is %zd bytes before head.\n",
++ 16 - (skb->data - skb->head));
++ }
++
++ nskb = __dev_alloc_skb(skb->len + 2,
++ GFP_ATOMIC|__GFP_NOWARN);
++ if (nskb != NULL) {
++ skb_reserve(nskb, 2);
++ skb_put(nskb, skb->len);
++ memcpy(nskb->data, skb->data, skb->len);
++ /* Copy any other fields we already set up. */
++ nskb->dev = skb->dev;
++ nskb->ip_summed = skb->ip_summed;
++ nskb->proto_data_valid = skb->proto_data_valid;
++ nskb->proto_csum_blank = skb->proto_csum_blank;
++ }
++
++ /* Reinitialise and then destroy the old skbuff. */
++ skb->len = 0;
++ skb->tail = skb->data;
++ init_skb_shinfo(skb);
++ dev_kfree_skb(skb);
++
++ /* Switch old for new, if we copied the buffer. */
++ if ((skb = nskb) == NULL)
++ continue;
++ }
++
++ /* Set the shinfo area, which is hidden behind the data. */
++ init_skb_shinfo(skb);
++ /* Ethernet work: Delayed to here as it peeks the header. */
++ skb->protocol = eth_type_trans(skb, dev);
++
++ /* Pass it up. */
++ netif_receive_skb(skb);
++ dev->last_rx = jiffies;
++ }
++
++ np->rx.rsp_cons = i;
++
++ /* If we get a callback with very few responses, reduce fill target. */
++ /* NB. Note exponential increase, linear decrease. */
++ if (((np->rx.req_prod_pvt - np->rx.sring->rsp_prod) >
++ ((3*np->rx_target) / 4)) &&
++ (--np->rx_target < np->rx_min_target))
++ np->rx_target = np->rx_min_target;
++
++ network_alloc_rx_buffers(dev);
++
++ *pbudget -= work_done;
++ dev->quota -= work_done;
++
++ if (work_done < budget) {
++ local_irq_save(flags);
++
++ RING_FINAL_CHECK_FOR_RESPONSES(&np->rx, more_to_do);
++ if (!more_to_do)
++ __netif_rx_complete(dev);
++
++ local_irq_restore(flags);
++ }
++
++ spin_unlock(&np->rx_lock);
++
++ return more_to_do;
++}
++
++
++static int network_close(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ netif_stop_queue(np->netdev);
++ return 0;
++}
++
++
++static struct net_device_stats *network_get_stats(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ return &np->stats;
++}
++
++static int xennet_change_mtu(struct net_device *dev, int mtu)
++{
++ int max = xennet_can_sg(dev) ? 65535 - ETH_HLEN : ETH_DATA_LEN;
++
++ if (mtu > max)
++ return -EINVAL;
++ dev->mtu = mtu;
++ return 0;
++}
++
++static int xennet_set_sg(struct net_device *dev, u32 data)
++{
++ if (data) {
++ struct netfront_info *np = netdev_priv(dev);
++ int val;
++
++ if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-sg",
++ "%d", &val) < 0)
++ val = 0;
++ if (!val)
++ return -ENOSYS;
++ } else if (dev->mtu > ETH_DATA_LEN)
++ dev->mtu = ETH_DATA_LEN;
++
++ return ethtool_op_set_sg(dev, data);
++}
++
++static int xennet_set_tso(struct net_device *dev, u32 data)
++{
++ if (data) {
++ struct netfront_info *np = netdev_priv(dev);
++ int val;
++
++ if (xenbus_scanf(XBT_NIL, np->xbdev->otherend, "feature-tso",
++ "%d", &val) < 0)
++ val = 0;
++#if 0 /* KAF: After the protocol is finalised. */
++ if (!val)
++#endif
++ return -ENOSYS;
++ }
++
++ return ethtool_op_set_tso(dev, data);
++}
++
++static void xennet_set_features(struct net_device *dev)
++{
++ if (!xennet_set_sg(dev, 1))
++ xennet_set_tso(dev, 1);
++}
++
++static void network_connect(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ int i, requeue_idx;
++ struct sk_buff *skb;
++
++ xennet_set_features(dev);
++
++ spin_lock_irq(&np->tx_lock);
++ spin_lock(&np->rx_lock);
++
++ /*
++ * Recovery procedure:
++ * NB. Freelist index entries are always going to be less than
++ * PAGE_OFFSET, whereas pointers to skbs will always be equal or
++ * greater than PAGE_OFFSET: we use this property to distinguish
++ * them.
++ */
++
++ /* Step 1: Discard all pending TX packet fragments. */
++ for (requeue_idx = 0, i = 1; i <= NET_TX_RING_SIZE; i++) {
++ if ((unsigned long)np->tx_skbs[i] < PAGE_OFFSET)
++ continue;
++
++ skb = np->tx_skbs[i];
++ gnttab_end_foreign_access_ref(
++ np->grant_tx_ref[i], GNTMAP_readonly);
++ gnttab_release_grant_reference(
++ &np->gref_tx_head, np->grant_tx_ref[i]);
++ np->grant_tx_ref[i] = GRANT_INVALID_REF;
++ add_id_to_freelist(np->tx_skbs, i);
++ dev_kfree_skb_irq(skb);
++ }
++
++ /* Step 2: Rebuild the RX buffer freelist and the RX ring itself. */
++ for (requeue_idx = 0, i = 1; i <= NET_RX_RING_SIZE; i++) {
++ if ((unsigned long)np->rx_skbs[i] < PAGE_OFFSET)
++ continue;
++ gnttab_grant_foreign_transfer_ref(
++ np->grant_rx_ref[i], np->xbdev->otherend_id,
++ __pa(np->rx_skbs[i]->data) >> PAGE_SHIFT);
++ RING_GET_REQUEST(&np->rx, requeue_idx)->gref =
++ np->grant_rx_ref[i];
++ RING_GET_REQUEST(&np->rx, requeue_idx)->id = i;
++ requeue_idx++;
++ }
++
++ np->rx.req_prod_pvt = requeue_idx;
++ RING_PUSH_REQUESTS(&np->rx);
++
++ /*
++ * Step 3: All public and private state should now be sane. Get
++ * ready to start sending and receiving packets and give the driver
++ * domain a kick because we've probably just requeued some
++ * packets.
++ */
++ netif_carrier_on(dev);
++ notify_remote_via_irq(np->irq);
++ network_tx_buf_gc(dev);
++ network_alloc_rx_buffers(dev);
++
++ spin_unlock(&np->rx_lock);
++ spin_unlock_irq(&np->tx_lock);
++}
++
++static void netif_uninit(struct net_device *dev)
++{
++ struct netfront_info *np = netdev_priv(dev);
++ gnttab_free_grant_references(np->gref_tx_head);
++ gnttab_free_grant_references(np->gref_rx_head);
++}
++
++static struct ethtool_ops network_ethtool_ops =
++{
++ .get_tx_csum = ethtool_op_get_tx_csum,
++ .set_tx_csum = ethtool_op_set_tx_csum,
++ .get_sg = ethtool_op_get_sg,
++ .set_sg = xennet_set_sg,
++ .get_tso = ethtool_op_get_tso,
++ .set_tso = xennet_set_tso,
++};
++
++#ifdef CONFIG_SYSFS
++static ssize_t show_rxbuf_min(struct class_device *cd, char *buf)
++{
++ struct net_device *netdev = container_of(cd, struct net_device,
++ class_dev);
++ struct netfront_info *info = netdev_priv(netdev);
++
++ return sprintf(buf, "%u\n", info->rx_min_target);
++}
++
++static ssize_t store_rxbuf_min(struct class_device *cd,
++ const char *buf, size_t len)
++{
++ struct net_device *netdev = container_of(cd, struct net_device,
++ class_dev);
++ struct netfront_info *np = netdev_priv(netdev);
++ char *endp;
++ unsigned long target;
++
++ if (!capable(CAP_NET_ADMIN))
++ return -EPERM;
++
++ target = simple_strtoul(buf, &endp, 0);
++ if (endp == buf)
++ return -EBADMSG;
++
++ if (target < RX_MIN_TARGET)
++ target = RX_MIN_TARGET;
++ if (target > RX_MAX_TARGET)
++ target = RX_MAX_TARGET;
++
++ spin_lock(&np->rx_lock);
++ if (target > np->rx_max_target)
++ np->rx_max_target = target;
++ np->rx_min_target = target;
++ if (target > np->rx_target)
++ np->rx_target = target;
++
++ network_alloc_rx_buffers(netdev);
++
++ spin_unlock(&np->rx_lock);
++ return len;
++}
++
++static ssize_t show_rxbuf_max(struct class_device *cd, char *buf)
++{
++ struct net_device *netdev = container_of(cd, struct net_device,
++ class_dev);
++ struct netfront_info *info = netdev_priv(netdev);
++
++ return sprintf(buf, "%u\n", info->rx_max_target);
++}
++
++static ssize_t store_rxbuf_max(struct class_device *cd,
++ const char *buf, size_t len)
++{
++ struct net_device *netdev = container_of(cd, struct net_device,
++ class_dev);
++ struct netfront_info *np = netdev_priv(netdev);
++ char *endp;
++ unsigned long target;
++
++ if (!capable(CAP_NET_ADMIN))
++ return -EPERM;
++
++ target = simple_strtoul(buf, &endp, 0);
++ if (endp == buf)
++ return -EBADMSG;
++
++ if (target < RX_MIN_TARGET)
++ target = RX_MIN_TARGET;
++ if (target > RX_MAX_TARGET)
++ target = RX_MAX_TARGET;
++
++ spin_lock(&np->rx_lock);
++ if (target < np->rx_min_target)
++ np->rx_min_target = target;
++ np->rx_max_target = target;
++ if (target < np->rx_target)
++ np->rx_target = target;
++
++ network_alloc_rx_buffers(netdev);
++
++ spin_unlock(&np->rx_lock);
++ return len;
++}
++
++static ssize_t show_rxbuf_cur(struct class_device *cd, char *buf)
++{
++ struct net_device *netdev = container_of(cd, struct net_device,
++ class_dev);
++ struct netfront_info *info = netdev_priv(netdev);
++
++ return sprintf(buf, "%u\n", info->rx_target);
++}
++
++static const struct class_device_attribute xennet_attrs[] = {
++ __ATTR(rxbuf_min, S_IRUGO|S_IWUSR, show_rxbuf_min, store_rxbuf_min),
++ __ATTR(rxbuf_max, S_IRUGO|S_IWUSR, show_rxbuf_max, store_rxbuf_max),
++ __ATTR(rxbuf_cur, S_IRUGO, show_rxbuf_cur, NULL),
++};
++
++static int xennet_sysfs_addif(struct net_device *netdev)
++{
++ int i;
++ int error = 0;
++
++ for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) {
++ error = class_device_create_file(&netdev->class_dev,
++ &xennet_attrs[i]);
++ if (error)
++ goto fail;
++ }
++ return 0;
++
++ fail:
++ while (--i >= 0)
++ class_device_remove_file(&netdev->class_dev,
++ &xennet_attrs[i]);
++ return error;
++}
++
++static void xennet_sysfs_delif(struct net_device *netdev)
++{
++ int i;
++
++ for (i = 0; i < ARRAY_SIZE(xennet_attrs); i++) {
++ class_device_remove_file(&netdev->class_dev,
++ &xennet_attrs[i]);
++ }
++}
++
++#endif /* CONFIG_SYSFS */
++
++
++/*
++ * Nothing to do here. Virtual interface is point-to-point and the
++ * physical interface is probably promiscuous anyway.
++ */
++static void network_set_multicast_list(struct net_device *dev)
++{
++}
++
++/** Create a network device.
++ * @param handle device handle
++ * @param val return parameter for created device
++ * @return 0 on success, error code otherwise
++ */
++static struct net_device * __devinit create_netdev(int handle,
++ struct xenbus_device *dev)
++{
++ int i, err = 0;
++ struct net_device *netdev = NULL;
++ struct netfront_info *np = NULL;
++
++ netdev = alloc_etherdev(sizeof(struct netfront_info));
++ if (!netdev) {
++ printk(KERN_WARNING "%s> alloc_etherdev failed.\n",
++ __FUNCTION__);
++ return ERR_PTR(-ENOMEM);
++ }
++
++ np = netdev_priv(netdev);
++ np->handle = handle;
++ np->xbdev = dev;
++
++ netif_carrier_off(netdev);
++
++ spin_lock_init(&np->tx_lock);
++ spin_lock_init(&np->rx_lock);
++
++ skb_queue_head_init(&np->rx_batch);
++ np->rx_target = RX_DFL_MIN_TARGET;
++ np->rx_min_target = RX_DFL_MIN_TARGET;
++ np->rx_max_target = RX_MAX_TARGET;
++
++ init_timer(&np->rx_refill_timer);
++ np->rx_refill_timer.data = (unsigned long)netdev;
++ np->rx_refill_timer.function = rx_refill_timeout;
++
++ /* Initialise {tx,rx}_skbs as a free chain containing every entry. */
++ for (i = 0; i <= NET_TX_RING_SIZE; i++) {
++ np->tx_skbs[i] = (void *)((unsigned long) i+1);
++ np->grant_tx_ref[i] = GRANT_INVALID_REF;
++ }
++
++ for (i = 0; i <= NET_RX_RING_SIZE; i++) {
++ np->rx_skbs[i] = (void *)((unsigned long) i+1);
++ np->grant_rx_ref[i] = GRANT_INVALID_REF;
++ }
++
++ /* A grant for every tx ring slot */
++ if (gnttab_alloc_grant_references(TX_MAX_TARGET,
++ &np->gref_tx_head) < 0) {
++ printk(KERN_ALERT "#### netfront can't alloc tx grant refs\n");
++ err = -ENOMEM;
++ goto exit;
++ }
++ /* A grant for every rx ring slot */
++ if (gnttab_alloc_grant_references(RX_MAX_TARGET,
++ &np->gref_rx_head) < 0) {
++ printk(KERN_ALERT "#### netfront can't alloc rx grant refs\n");
++ err = -ENOMEM;
++ goto exit_free_tx;
++ }
++
++ netdev->open = network_open;
++ netdev->hard_start_xmit = network_start_xmit;
++ netdev->stop = network_close;
++ netdev->get_stats = network_get_stats;
++ netdev->poll = netif_poll;
++ netdev->set_multicast_list = network_set_multicast_list;
++ netdev->uninit = netif_uninit;
++ netdev->change_mtu = xennet_change_mtu;
++ netdev->weight = 64;
++ netdev->features = NETIF_F_IP_CSUM;
++
++ SET_ETHTOOL_OPS(netdev, &network_ethtool_ops);
++ SET_MODULE_OWNER(netdev);
++ SET_NETDEV_DEV(netdev, &dev->dev);
++
++ err = register_netdev(netdev);
++ if (err) {
++ printk(KERN_WARNING "%s> register_netdev err=%d\n",
++ __FUNCTION__, err);
++ goto exit_free_rx;
++ }
++
++ err = xennet_sysfs_addif(netdev);
++ if (err) {
++ /* This can be non-fatal: it only means no tuning parameters */
++ printk(KERN_WARNING "%s> add sysfs failed err=%d\n",
++ __FUNCTION__, err);
++ }
++
++ np->netdev = netdev;
++
++ return netdev;
++
++
++ exit_free_rx:
++ gnttab_free_grant_references(np->gref_rx_head);
++ exit_free_tx:
++ gnttab_free_grant_references(np->gref_tx_head);
++ exit:
++ free_netdev(netdev);
++ return ERR_PTR(err);
++}
++
++/*
++ * We use this notifier to send out a fake ARP reply to reset switches and
++ * router ARP caches when an IP interface is brought up on a VIF.
++ */
++static int
++inetdev_notify(struct notifier_block *this, unsigned long event, void *ptr)
++{
++ struct in_ifaddr *ifa = (struct in_ifaddr *)ptr;
++ struct net_device *dev = ifa->ifa_dev->dev;
++
++ /* UP event and is it one of our devices? */
++ if (event == NETDEV_UP && dev->open == network_open)
++ (void)send_fake_arp(dev);
++
++ return NOTIFY_DONE;
++}
++
++
++/* ** Close down ** */
++
++
++/**
++ * Handle the change of state of the backend to Closing. We must delete our
++ * device-layer structures now, to ensure that writes are flushed through to
++ * the backend. Once is this done, we can switch to Closed in
++ * acknowledgement.
++ */
++static void netfront_closing(struct xenbus_device *dev)
++{
++ struct netfront_info *info = dev->dev.driver_data;
++
++ DPRINTK("netfront_closing: %s removed\n", dev->nodename);
++
++ close_netdev(info);
++
++ xenbus_switch_state(dev, XenbusStateClosed);
++}
++
++
++static int __devexit netfront_remove(struct xenbus_device *dev)
++{
++ struct netfront_info *info = dev->dev.driver_data;
++
++ DPRINTK("%s\n", dev->nodename);
++
++ netif_disconnect_backend(info);
++ free_netdev(info->netdev);
++
++ return 0;
++}
++
++
++static void close_netdev(struct netfront_info *info)
++{
++ del_timer_sync(&info->rx_refill_timer);
++
++ xennet_sysfs_delif(info->netdev);
++ unregister_netdev(info->netdev);
++}
++
++
++static void netif_disconnect_backend(struct netfront_info *info)
++{
++ /* Stop old i/f to prevent errors whilst we rebuild the state. */
++ spin_lock_irq(&info->tx_lock);
++ spin_lock(&info->rx_lock);
++ netif_carrier_off(info->netdev);
++ spin_unlock(&info->rx_lock);
++ spin_unlock_irq(&info->tx_lock);
++
++ if (info->irq)
++ unbind_from_irqhandler(info->irq, info->netdev);
++ info->evtchn = info->irq = 0;
++
++ end_access(info->tx_ring_ref, info->tx.sring);
++ end_access(info->rx_ring_ref, info->rx.sring);
++ info->tx_ring_ref = GRANT_INVALID_REF;
++ info->rx_ring_ref = GRANT_INVALID_REF;
++ info->tx.sring = NULL;
++ info->rx.sring = NULL;
++}
++
++
++static void netif_free(struct netfront_info *info)
++{
++ close_netdev(info);
++ netif_disconnect_backend(info);
++ free_netdev(info->netdev);
++}
++
++
++static void end_access(int ref, void *page)
++{
++ if (ref != GRANT_INVALID_REF)
++ gnttab_end_foreign_access(ref, 0, (unsigned long)page);
++}
++
++
++/* ** Driver registration ** */
++
++
++static struct xenbus_device_id netfront_ids[] = {
++ { "vif" },
++ { "" }
++};
++
++
++static struct xenbus_driver netfront = {
++ .name = "vif",
++ .owner = THIS_MODULE,
++ .ids = netfront_ids,
++ .probe = netfront_probe,
++ .remove = __devexit_p(netfront_remove),
++ .resume = netfront_resume,
++ .otherend_changed = backend_changed,
++};
++
++
++static struct notifier_block notifier_inetdev = {
++ .notifier_call = inetdev_notify,
++ .next = NULL,
++ .priority = 0
++};
++
++static int __init netif_init(void)
++{
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ return 0;
++
++ IPRINTK("Initialising virtual ethernet driver.\n");
++
++ (void)register_inetaddr_notifier(¬ifier_inetdev);
++
++ return xenbus_register_frontend(&netfront);
++}
++module_init(netif_init);
++
++
++static void __exit netif_exit(void)
++{
++ unregister_inetaddr_notifier(¬ifier_inetdev);
++
++ return xenbus_unregister_driver(&netfront);
++}
++module_exit(netif_exit);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space.c new/drivers/xen/pciback/conf_space.c
+--- linux-2.6/drivers/xen/pciback/conf_space.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,403 @@
++/*
++ * PCI Backend - Functions for creating a virtual configuration space for
++ * exported PCI Devices.
++ * It's dangerous to allow PCI Driver Domains to change their
++ * device's resources (memory, i/o ports, interrupts). We need to
++ * restrict changes to certain PCI Configuration registers:
++ * BARs, INTERRUPT_PIN, most registers in the header...
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/kernel.h>
++#include <linux/pci.h>
++#include "pciback.h"
++#include "conf_space.h"
++
++static int permissive = 0;
++module_param(permissive, bool, 0644);
++
++#define DEFINE_PCI_CONFIG(op,size,type) \
++int pciback_##op##_config_##size \
++(struct pci_dev *dev, int offset, type value, void *data) \
++{ \
++ return pci_##op##_config_##size (dev, offset, value); \
++}
++
++DEFINE_PCI_CONFIG(read, byte, u8 *)
++DEFINE_PCI_CONFIG(read, word, u16 *)
++DEFINE_PCI_CONFIG(read, dword, u32 *)
++
++DEFINE_PCI_CONFIG(write, byte, u8)
++DEFINE_PCI_CONFIG(write, word, u16)
++DEFINE_PCI_CONFIG(write, dword, u32)
++
++static int conf_space_read(struct pci_dev *dev,
++ struct config_field_entry *entry, int offset,
++ u32 * value)
++{
++ int ret = 0;
++ struct config_field *field = entry->field;
++
++ *value = 0;
++
++ switch (field->size) {
++ case 1:
++ if (field->u.b.read)
++ ret = field->u.b.read(dev, offset, (u8 *) value,
++ entry->data);
++ break;
++ case 2:
++ if (field->u.w.read)
++ ret = field->u.w.read(dev, offset, (u16 *) value,
++ entry->data);
++ break;
++ case 4:
++ if (field->u.dw.read)
++ ret = field->u.dw.read(dev, offset, value, entry->data);
++ break;
++ }
++ return ret;
++}
++
++static int conf_space_write(struct pci_dev *dev,
++ struct config_field_entry *entry, int offset,
++ u32 value)
++{
++ int ret = 0;
++ struct config_field *field = entry->field;
++
++ switch (field->size) {
++ case 1:
++ if (field->u.b.write)
++ ret = field->u.b.write(dev, offset, (u8) value,
++ entry->data);
++ break;
++ case 2:
++ if (field->u.w.write)
++ ret = field->u.w.write(dev, offset, (u16) value,
++ entry->data);
++ break;
++ case 4:
++ if (field->u.dw.write)
++ ret = field->u.dw.write(dev, offset, value,
++ entry->data);
++ break;
++ }
++ return ret;
++}
++
++static inline u32 get_mask(int size)
++{
++ if (size == 1)
++ return 0xff;
++ else if (size == 2)
++ return 0xffff;
++ else
++ return 0xffffffff;
++}
++
++static inline int valid_request(int offset, int size)
++{
++ /* Validate request (no un-aligned requests) */
++ if ((size == 1 || size == 2 || size == 4) && (offset % size) == 0)
++ return 1;
++ return 0;
++}
++
++static inline u32 merge_value(u32 val, u32 new_val, u32 new_val_mask,
++ int offset)
++{
++ if (offset >= 0) {
++ new_val_mask <<= (offset * 8);
++ new_val <<= (offset * 8);
++ } else {
++ new_val_mask >>= (offset * -8);
++ new_val >>= (offset * -8);
++ }
++ val = (val & ~new_val_mask) | (new_val & new_val_mask);
++
++ return val;
++}
++
++static int pcibios_err_to_errno(int err)
++{
++ switch (err) {
++ case PCIBIOS_SUCCESSFUL:
++ return XEN_PCI_ERR_success;
++ case PCIBIOS_DEVICE_NOT_FOUND:
++ return XEN_PCI_ERR_dev_not_found;
++ case PCIBIOS_BAD_REGISTER_NUMBER:
++ return XEN_PCI_ERR_invalid_offset;
++ case PCIBIOS_FUNC_NOT_SUPPORTED:
++ return XEN_PCI_ERR_not_implemented;
++ case PCIBIOS_SET_FAILED:
++ return XEN_PCI_ERR_access_denied;
++ }
++ return err;
++}
++
++int pciback_config_read(struct pci_dev *dev, int offset, int size,
++ u32 * ret_val)
++{
++ int err = 0;
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++ struct config_field_entry *cfg_entry;
++ struct config_field *field;
++ int req_start, req_end, field_start, field_end;
++ /* if read fails for any reason, return 0 (as if device didn't respond) */
++ u32 value = 0, tmp_val;
++
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG "pciback: %s: read %d bytes at 0x%x\n",
++ pci_name(dev), size, offset);
++
++ if (!valid_request(offset, size)) {
++ err = XEN_PCI_ERR_invalid_offset;
++ goto out;
++ }
++
++ /* Get the real value first, then modify as appropriate */
++ switch (size) {
++ case 1:
++ err = pci_read_config_byte(dev, offset, (u8 *) & value);
++ break;
++ case 2:
++ err = pci_read_config_word(dev, offset, (u16 *) & value);
++ break;
++ case 4:
++ err = pci_read_config_dword(dev, offset, &value);
++ break;
++ }
++
++ list_for_each_entry(cfg_entry, &dev_data->config_fields, list) {
++ field = cfg_entry->field;
++
++ req_start = offset;
++ req_end = offset + size;
++ field_start = OFFSET(cfg_entry);
++ field_end = OFFSET(cfg_entry) + field->size;
++
++ if ((req_start >= field_start && req_start < field_end)
++ || (req_end > field_start && req_end <= field_end)) {
++ err = conf_space_read(dev, cfg_entry, field_start,
++ &tmp_val);
++ if (err)
++ goto out;
++
++ value = merge_value(value, tmp_val,
++ get_mask(field->size),
++ field_start - req_start);
++ }
++ }
++
++ out:
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG "pciback: %s: read %d bytes at 0x%x = %x\n",
++ pci_name(dev), size, offset, value);
++
++ *ret_val = value;
++ return pcibios_err_to_errno(err);
++}
++
++int pciback_config_write(struct pci_dev *dev, int offset, int size, u32 value)
++{
++ int err = 0, handled = 0;
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++ struct config_field_entry *cfg_entry;
++ struct config_field *field;
++ u32 tmp_val;
++ int req_start, req_end, field_start, field_end;
++
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG
++ "pciback: %s: write request %d bytes at 0x%x = %x\n",
++ pci_name(dev), size, offset, value);
++
++ if (!valid_request(offset, size))
++ return XEN_PCI_ERR_invalid_offset;
++
++ list_for_each_entry(cfg_entry, &dev_data->config_fields, list) {
++ field = cfg_entry->field;
++
++ req_start = offset;
++ req_end = offset + size;
++ field_start = OFFSET(cfg_entry);
++ field_end = OFFSET(cfg_entry) + field->size;
++
++ if ((req_start >= field_start && req_start < field_end)
++ || (req_end > field_start && req_end <= field_end)) {
++ tmp_val = 0;
++
++ err = pciback_config_read(dev, field_start,
++ field->size, &tmp_val);
++ if (err)
++ break;
++
++ tmp_val = merge_value(tmp_val, value, get_mask(size),
++ req_start - field_start);
++
++ err = conf_space_write(dev, cfg_entry, field_start,
++ tmp_val);
++
++ /* handled is set true here, but not every byte
++ * may have been written! Properly detecting if
++ * every byte is handled is unnecessary as the
++ * flag is used to detect devices that need
++ * special helpers to work correctly.
++ */
++ handled = 1;
++ }
++ }
++
++ if (!handled && !err) {
++ /* By default, anything not specificially handled above is
++ * read-only. The permissive flag changes this behavior so
++ * that anything not specifically handled above is writable.
++ * This means that some fields may still be read-only because
++ * they have entries in the config_field list that intercept
++ * the write and do nothing. */
++ if (permissive) {
++ switch (size) {
++ case 1:
++ err = pci_write_config_byte(dev, offset,
++ (u8)value);
++ break;
++ case 2:
++ err = pci_write_config_word(dev, offset,
++ (u16)value);
++ break;
++ case 4:
++ err = pci_write_config_dword(dev, offset,
++ (u32)value);
++ break;
++ }
++ } else if (!dev_data->warned_on_write) {
++ dev_data->warned_on_write = 1;
++ dev_warn(&dev->dev, "Driver wrote to a read-only "
++ "configuration space field!\n");
++ dev_warn(&dev->dev, "Write at offset 0x%x size %d\n",
++ offset, size);
++ dev_warn(&dev->dev, "This may be harmless, but if\n");
++ dev_warn(&dev->dev, "you have problems with your "
++ "device:\n");
++ dev_warn(&dev->dev, "1) see the permissive "
++ "attribute in sysfs.\n");
++ dev_warn(&dev->dev, "2) report problems to the "
++ "xen-devel mailing list along\n");
++ dev_warn(&dev->dev, " with details of your device "
++ "obtained from lspci.\n");
++ }
++ }
++
++ return pcibios_err_to_errno(err);
++}
++
++void pciback_config_reset_dev(struct pci_dev *dev)
++{
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++ struct config_field_entry *cfg_entry;
++ struct config_field *field;
++
++ dev_dbg(&dev->dev, "resetting virtual configuration space\n");
++
++ list_for_each_entry(cfg_entry, &dev_data->config_fields, list) {
++ field = cfg_entry->field;
++
++ if (field->reset)
++ field->reset(dev, OFFSET(cfg_entry), cfg_entry->data);
++ }
++}
++
++void pciback_config_free_dev(struct pci_dev *dev)
++{
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++ struct config_field_entry *cfg_entry, *t;
++ struct config_field *field;
++
++ dev_dbg(&dev->dev, "free-ing virtual configuration space fields\n");
++
++ list_for_each_entry_safe(cfg_entry, t, &dev_data->config_fields, list) {
++ list_del(&cfg_entry->list);
++
++ field = cfg_entry->field;
++
++ if (field->release)
++ field->release(dev, OFFSET(cfg_entry), cfg_entry->data);
++
++ kfree(cfg_entry);
++ }
++}
++
++int pciback_config_add_field_offset(struct pci_dev *dev,
++ struct config_field *field,
++ unsigned int offset)
++{
++ int err = 0;
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++ struct config_field_entry *cfg_entry;
++ void *tmp;
++
++ cfg_entry = kmalloc(sizeof(*cfg_entry), GFP_KERNEL);
++ if (!cfg_entry) {
++ err = -ENOMEM;
++ goto out;
++ }
++
++ cfg_entry->data = NULL;
++ cfg_entry->field = field;
++ cfg_entry->base_offset = offset;
++
++ if (field->init) {
++ tmp = field->init(dev, OFFSET(cfg_entry));
++
++ if (IS_ERR(tmp)) {
++ err = PTR_ERR(tmp);
++ goto out;
++ }
++
++ cfg_entry->data = tmp;
++ }
++
++ dev_dbg(&dev->dev, "added config field at offset 0x%02x\n",
++ OFFSET(cfg_entry));
++ list_add_tail(&cfg_entry->list, &dev_data->config_fields);
++
++ out:
++ if (err)
++ kfree(cfg_entry);
++
++ return err;
++}
++
++/* This sets up the device's virtual configuration space to keep track of
++ * certain registers (like the base address registers (BARs) so that we can
++ * keep the client from manipulating them directly.
++ */
++int pciback_config_init_dev(struct pci_dev *dev)
++{
++ int err = 0;
++ struct pciback_dev_data *dev_data = pci_get_drvdata(dev);
++
++ dev_dbg(&dev->dev, "initializing virtual configuration space\n");
++
++ INIT_LIST_HEAD(&dev_data->config_fields);
++
++ err = pciback_config_header_add_fields(dev);
++ if (err)
++ goto out;
++
++ err = pciback_config_capability_add_fields(dev);
++
++ out:
++ return err;
++}
++
++int pciback_config_init(void)
++{
++ int err;
++
++ err = pciback_config_capability_init();
++
++ return err;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space_capability.c new/drivers/xen/pciback/conf_space_capability.c
+--- linux-2.6/drivers/xen/pciback/conf_space_capability.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space_capability.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,71 @@
++/*
++ * PCI Backend - Handles the virtual fields found on the capability lists
++ * in the configuration space.
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/kernel.h>
++#include <linux/pci.h>
++#include "pciback.h"
++#include "conf_space.h"
++#include "conf_space_capability.h"
++
++static LIST_HEAD(capabilities);
++
++static struct config_field caplist_header[] = {
++ {
++ .offset = PCI_CAP_LIST_ID,
++ .size = 2, /* encompass PCI_CAP_LIST_ID & PCI_CAP_LIST_NEXT */
++ .u.w.read = pciback_read_config_word,
++ .u.w.write = NULL,
++ },
++ {
++ .size = 0,
++ },
++};
++
++static inline void register_capability(struct pciback_config_capability *cap)
++{
++ list_add_tail(&cap->cap_list, &capabilities);
++}
++
++int pciback_config_capability_add_fields(struct pci_dev *dev)
++{
++ int err = 0;
++ struct pciback_config_capability *cap;
++ int cap_offset;
++
++ list_for_each_entry(cap, &capabilities, cap_list) {
++ cap_offset = pci_find_capability(dev, cap->capability);
++ if (cap_offset) {
++ dev_dbg(&dev->dev, "Found capability 0x%x at 0x%x\n",
++ cap->capability, cap_offset);
++
++ err = pciback_config_add_fields_offset(dev,
++ caplist_header,
++ cap_offset);
++ if (err)
++ goto out;
++ err = pciback_config_add_fields_offset(dev,
++ cap->fields,
++ cap_offset);
++ if (err)
++ goto out;
++ }
++ }
++
++ out:
++ return err;
++}
++
++extern struct pciback_config_capability pciback_config_capability_vpd;
++extern struct pciback_config_capability pciback_config_capability_pm;
++
++int pciback_config_capability_init(void)
++{
++ register_capability(&pciback_config_capability_vpd);
++ register_capability(&pciback_config_capability_pm);
++
++ return 0;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space_capability.h new/drivers/xen/pciback/conf_space_capability.h
+--- linux-2.6/drivers/xen/pciback/conf_space_capability.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space_capability.h 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,23 @@
++/*
++ * PCI Backend - Data structures for special overlays for structures on
++ * the capability list.
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#ifndef __PCIBACK_CONFIG_CAPABILITY_H__
++#define __PCIBACK_CONFIG_CAPABILITY_H__
++
++#include <linux/pci.h>
++#include <linux/list.h>
++
++struct pciback_config_capability {
++ struct list_head cap_list;
++
++ int capability;
++
++ /* If the device has the capability found above, add these fields */
++ struct config_field *fields;
++};
++
++#endif
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space_capability_pm.c new/drivers/xen/pciback/conf_space_capability_pm.c
+--- linux-2.6/drivers/xen/pciback/conf_space_capability_pm.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space_capability_pm.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,113 @@
++/*
++ * PCI Backend - Configuration space overlay for power management
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/pci.h>
++#include "conf_space.h"
++#include "conf_space_capability.h"
++
++static int pm_caps_read(struct pci_dev *dev, int offset, u16 *value,
++ void *data)
++{
++ int err;
++ u16 real_value;
++
++ err = pci_read_config_word(dev, offset, &real_value);
++ if (err)
++ goto out;
++
++ *value = real_value & ~PCI_PM_CAP_PME_MASK;
++
++ out:
++ return err;
++}
++
++/* PM_OK_BITS specifies the bits that the driver domain is allowed to change.
++ * Can't allow driver domain to enable PMEs - they're shared */
++#define PM_OK_BITS (PCI_PM_CTRL_PME_STATUS|PCI_PM_CTRL_DATA_SEL_MASK)
++
++static int pm_ctrl_write(struct pci_dev *dev, int offset, u16 new_value,
++ void *data)
++{
++ int err;
++ u16 cur_value;
++ pci_power_t new_state;
++
++ /* Handle setting power state separately */
++ new_state = (pci_power_t)(new_value & PCI_PM_CTRL_STATE_MASK);
++
++ err = pci_read_config_word(dev, offset, &cur_value);
++ if (err)
++ goto out;
++
++ new_value &= PM_OK_BITS;
++ if ((cur_value & PM_OK_BITS) != new_value) {
++ new_value = (cur_value & ~PM_OK_BITS) | new_value;
++ err = pci_write_config_word(dev, offset, new_value);
++ if (err)
++ goto out;
++ }
++
++ /* Let pci core handle the power management change */
++ dev_dbg(&dev->dev, "set power state to %x\n", new_state);
++ err = pci_set_power_state(dev, new_state);
++ if (err)
++ err = PCIBIOS_SET_FAILED;
++
++ out:
++ return err;
++}
++
++/* Ensure PMEs are disabled */
++static void *pm_ctrl_init(struct pci_dev *dev, int offset)
++{
++ int err;
++ u16 value;
++
++ err = pci_read_config_word(dev, offset, &value);
++ if (err)
++ goto out;
++
++ if (value & PCI_PM_CTRL_PME_ENABLE) {
++ value &= ~PCI_PM_CTRL_PME_ENABLE;
++ err = pci_write_config_word(dev, offset, value);
++ }
++
++ out:
++ return ERR_PTR(err);
++}
++
++static struct config_field caplist_pm[] = {
++ {
++ .offset = PCI_PM_PMC,
++ .size = 2,
++ .u.w.read = pm_caps_read,
++ },
++ {
++ .offset = PCI_PM_CTRL,
++ .size = 2,
++ .init = pm_ctrl_init,
++ .u.w.read = pciback_read_config_word,
++ .u.w.write = pm_ctrl_write,
++ },
++ {
++ .offset = PCI_PM_PPB_EXTENSIONS,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ },
++ {
++ .offset = PCI_PM_DATA_REGISTER,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ },
++ {
++ .size = 0,
++ },
++};
++
++struct pciback_config_capability pciback_config_capability_pm = {
++ .capability = PCI_CAP_ID_PM,
++ .fields = caplist_pm,
++};
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space_capability_vpd.c new/drivers/xen/pciback/conf_space_capability_vpd.c
+--- linux-2.6/drivers/xen/pciback/conf_space_capability_vpd.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space_capability_vpd.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,42 @@
++/*
++ * PCI Backend - Configuration space overlay for Vital Product Data
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/pci.h>
++#include "conf_space.h"
++#include "conf_space_capability.h"
++
++static int vpd_address_write(struct pci_dev *dev, int offset, u16 value,
++ void *data)
++{
++ /* Disallow writes to the vital product data */
++ if (value & PCI_VPD_ADDR_F)
++ return PCIBIOS_SET_FAILED;
++ else
++ return pci_write_config_word(dev, offset, value);
++}
++
++static struct config_field caplist_vpd[] = {
++ {
++ .offset = PCI_VPD_ADDR,
++ .size = 2,
++ .u.w.read = pciback_read_config_word,
++ .u.w.write = vpd_address_write,
++ },
++ {
++ .offset = PCI_VPD_DATA,
++ .size = 4,
++ .u.dw.read = pciback_read_config_dword,
++ .u.dw.write = NULL,
++ },
++ {
++ .size = 0,
++ },
++};
++
++struct pciback_config_capability pciback_config_capability_vpd = {
++ .capability = PCI_CAP_ID_VPD,
++ .fields = caplist_vpd,
++};
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space.h new/drivers/xen/pciback/conf_space.h
+--- linux-2.6/drivers/xen/pciback/conf_space.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space.h 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,123 @@
++/*
++ * PCI Backend - Common data structures for overriding the configuration space
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#ifndef __XEN_PCIBACK_CONF_SPACE_H__
++#define __XEN_PCIBACK_CONF_SPACE_H__
++
++#include <linux/list.h>
++#include <linux/err.h>
++
++/* conf_field_init can return an errno in a ptr with ERR_PTR() */
++typedef void *(*conf_field_init) (struct pci_dev * dev, int offset);
++typedef void (*conf_field_reset) (struct pci_dev * dev, int offset, void *data);
++typedef void (*conf_field_free) (struct pci_dev * dev, int offset, void *data);
++
++typedef int (*conf_dword_write) (struct pci_dev * dev, int offset, u32 value,
++ void *data);
++typedef int (*conf_word_write) (struct pci_dev * dev, int offset, u16 value,
++ void *data);
++typedef int (*conf_byte_write) (struct pci_dev * dev, int offset, u8 value,
++ void *data);
++typedef int (*conf_dword_read) (struct pci_dev * dev, int offset, u32 * value,
++ void *data);
++typedef int (*conf_word_read) (struct pci_dev * dev, int offset, u16 * value,
++ void *data);
++typedef int (*conf_byte_read) (struct pci_dev * dev, int offset, u8 * value,
++ void *data);
++
++/* These are the fields within the configuration space which we
++ * are interested in intercepting reads/writes to and changing their
++ * values.
++ */
++struct config_field {
++ unsigned int offset;
++ unsigned int size;
++ conf_field_init init;
++ conf_field_reset reset;
++ conf_field_free release;
++ union {
++ struct {
++ conf_dword_write write;
++ conf_dword_read read;
++ } dw;
++ struct {
++ conf_word_write write;
++ conf_word_read read;
++ } w;
++ struct {
++ conf_byte_write write;
++ conf_byte_read read;
++ } b;
++ } u;
++};
++
++struct config_field_entry {
++ struct list_head list;
++ struct config_field *field;
++ unsigned int base_offset;
++ void *data;
++};
++
++#define OFFSET(cfg_entry) ((cfg_entry)->base_offset+(cfg_entry)->field->offset)
++
++/* Add fields to a device - the add_fields macro expects to get a pointer to
++ * the first entry in an array (of which the ending is marked by size==0)
++ */
++int pciback_config_add_field_offset(struct pci_dev *dev,
++ struct config_field *field,
++ unsigned int offset);
++
++static inline int pciback_config_add_field(struct pci_dev *dev,
++ struct config_field *field)
++{
++ return pciback_config_add_field_offset(dev, field, 0);
++}
++
++static inline int pciback_config_add_fields(struct pci_dev *dev,
++ struct config_field *field)
++{
++ int i, err = 0;
++ for (i = 0; field[i].size != 0; i++) {
++ err = pciback_config_add_field(dev, &field[i]);
++ if (err)
++ break;
++ }
++ return err;
++}
++
++static inline int pciback_config_add_fields_offset(struct pci_dev *dev,
++ struct config_field *field,
++ unsigned int offset)
++{
++ int i, err = 0;
++ for (i = 0; field[i].size != 0; i++) {
++ err = pciback_config_add_field_offset(dev, &field[i], offset);
++ if (err)
++ break;
++ }
++ return err;
++}
++
++/* Read/Write the real configuration space */
++int pciback_read_config_byte(struct pci_dev *dev, int offset, u8 * value,
++ void *data);
++int pciback_read_config_word(struct pci_dev *dev, int offset, u16 * value,
++ void *data);
++int pciback_read_config_dword(struct pci_dev *dev, int offset, u32 * value,
++ void *data);
++int pciback_write_config_byte(struct pci_dev *dev, int offset, u8 value,
++ void *data);
++int pciback_write_config_word(struct pci_dev *dev, int offset, u16 value,
++ void *data);
++int pciback_write_config_dword(struct pci_dev *dev, int offset, u32 value,
++ void *data);
++
++int pciback_config_capability_init(void);
++
++int pciback_config_header_add_fields(struct pci_dev *dev);
++int pciback_config_capability_add_fields(struct pci_dev *dev);
++
++#endif /* __XEN_PCIBACK_CONF_SPACE_H__ */
+diff -urNp linux-2.6/drivers/xen/pciback/conf_space_header.c new/drivers/xen/pciback/conf_space_header.c
+--- linux-2.6/drivers/xen/pciback/conf_space_header.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/conf_space_header.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,299 @@
++/*
++ * PCI Backend - Handles the virtual fields in the configuration space headers.
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/kernel.h>
++#include <linux/pci.h>
++#include "pciback.h"
++#include "conf_space.h"
++
++struct pci_bar_info {
++ u32 val;
++ u32 len_val;
++ int which;
++};
++
++#define is_enable_cmd(value) ((value)&(PCI_COMMAND_MEMORY|PCI_COMMAND_IO))
++#define is_master_cmd(value) ((value)&PCI_COMMAND_MASTER)
++
++static int command_write(struct pci_dev *dev, int offset, u16 value, void *data)
++{
++ if (!dev->is_enabled && is_enable_cmd(value)) {
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG "pciback: %s: enable\n",
++ pci_name(dev));
++ pci_enable_device(dev);
++ } else if (dev->is_enabled && !is_enable_cmd(value)) {
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG "pciback: %s: disable\n",
++ pci_name(dev));
++ pci_disable_device(dev);
++ }
++
++ if (!dev->is_busmaster && is_master_cmd(value)) {
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG "pciback: %s: set bus master\n",
++ pci_name(dev));
++ pci_set_master(dev);
++ }
++
++ if (value & PCI_COMMAND_INVALIDATE) {
++ if (unlikely(verbose_request))
++ printk(KERN_DEBUG
++ "pciback: %s: enable memory-write-invalidate\n",
++ pci_name(dev));
++ pci_set_mwi(dev);
++ }
++
++ return pci_write_config_word(dev, offset, value);
++}
++
++static int rom_write(struct pci_dev *dev, int offset, u32 value, void *data)
++{
++ struct pci_bar_info *bar = data;
++
++ if (unlikely(!bar)) {
++ printk(KERN_WARNING "pciback: driver data not found for %s\n",
++ pci_name(dev));
++ return XEN_PCI_ERR_op_failed;
++ }
++
++ /* A write to obtain the length must happen as a 32-bit write.
++ * This does not (yet) support writing individual bytes
++ */
++ if (value == ~PCI_ROM_ADDRESS_ENABLE)
++ bar->which = 1;
++ else
++ bar->which = 0;
++
++ /* Do we need to support enabling/disabling the rom address here? */
++
++ return 0;
++}
++
++/* For the BARs, only allow writes which write ~0 or
++ * the correct resource information
++ * (Needed for when the driver probes the resource usage)
++ */
++static int bar_write(struct pci_dev *dev, int offset, u32 value, void *data)
++{
++ struct pci_bar_info *bar = data;
++
++ if (unlikely(!bar)) {
++ printk(KERN_WARNING "pciback: driver data not found for %s\n",
++ pci_name(dev));
++ return XEN_PCI_ERR_op_failed;
++ }
++
++ /* A write to obtain the length must happen as a 32-bit write.
++ * This does not (yet) support writing individual bytes
++ */
++ if (value == ~0)
++ bar->which = 1;
++ else
++ bar->which = 0;
++
++ return 0;
++}
++
++static int bar_read(struct pci_dev *dev, int offset, u32 * value, void *data)
++{
++ struct pci_bar_info *bar = data;
++
++ if (unlikely(!bar)) {
++ printk(KERN_WARNING "pciback: driver data not found for %s\n",
++ pci_name(dev));
++ return XEN_PCI_ERR_op_failed;
++ }
++
++ *value = bar->which ? bar->len_val : bar->val;
++
++ return 0;
++}
++
++static inline void read_dev_bar(struct pci_dev *dev,
++ struct pci_bar_info *bar_info, int offset,
++ u32 len_mask)
++{
++ pci_read_config_dword(dev, offset, &bar_info->val);
++ pci_write_config_dword(dev, offset, len_mask);
++ pci_read_config_dword(dev, offset, &bar_info->len_val);
++ pci_write_config_dword(dev, offset, bar_info->val);
++}
++
++static void *bar_init(struct pci_dev *dev, int offset)
++{
++ struct pci_bar_info *bar = kmalloc(sizeof(*bar), GFP_KERNEL);
++
++ if (!bar)
++ return ERR_PTR(-ENOMEM);
++
++ read_dev_bar(dev, bar, offset, ~0);
++ bar->which = 0;
++
++ return bar;
++}
++
++static void *rom_init(struct pci_dev *dev, int offset)
++{
++ struct pci_bar_info *bar = kmalloc(sizeof(*bar), GFP_KERNEL);
++
++ if (!bar)
++ return ERR_PTR(-ENOMEM);
++
++ read_dev_bar(dev, bar, offset, ~PCI_ROM_ADDRESS_ENABLE);
++ bar->which = 0;
++
++ return bar;
++}
++
++static void bar_reset(struct pci_dev *dev, int offset, void *data)
++{
++ struct pci_bar_info *bar = data;
++
++ bar->which = 0;
++}
++
++static void bar_release(struct pci_dev *dev, int offset, void *data)
++{
++ kfree(data);
++}
++
++static int interrupt_read(struct pci_dev *dev, int offset, u8 * value,
++ void *data)
++{
++ *value = (u8) dev->irq;
++
++ return 0;
++}
++
++static int bist_write(struct pci_dev *dev, int offset, u8 value, void *data)
++{
++ u8 cur_value;
++ int err;
++
++ err = pci_read_config_byte(dev, offset, &cur_value);
++ if (err)
++ goto out;
++
++ if ((cur_value & ~PCI_BIST_START) == (value & ~PCI_BIST_START)
++ || value == PCI_BIST_START)
++ err = pci_write_config_byte(dev, offset, value);
++
++ out:
++ return err;
++}
++
++static struct config_field header_common[] = {
++ {
++ .offset = PCI_COMMAND,
++ .size = 2,
++ .u.w.read = pciback_read_config_word,
++ .u.w.write = command_write,
++ },
++ {
++ .offset = PCI_INTERRUPT_LINE,
++ .size = 1,
++ .u.b.read = interrupt_read,
++ },
++ {
++ .offset = PCI_INTERRUPT_PIN,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ },
++ {
++ /* Any side effects of letting driver domain control cache line? */
++ .offset = PCI_CACHE_LINE_SIZE,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ .u.b.write = pciback_write_config_byte,
++ },
++ {
++ .offset = PCI_LATENCY_TIMER,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ },
++ {
++ .offset = PCI_BIST,
++ .size = 1,
++ .u.b.read = pciback_read_config_byte,
++ .u.b.write = bist_write,
++ },
++ {
++ .size = 0,
++ },
++};
++
++#define CFG_FIELD_BAR(reg_offset) \
++ { \
++ .offset = reg_offset, \
++ .size = 4, \
++ .init = bar_init, \
++ .reset = bar_reset, \
++ .release = bar_release, \
++ .u.dw.read = bar_read, \
++ .u.dw.write = bar_write, \
++ }
++
++#define CFG_FIELD_ROM(reg_offset) \
++ { \
++ .offset = reg_offset, \
++ .size = 4, \
++ .init = rom_init, \
++ .reset = bar_reset, \
++ .release = bar_release, \
++ .u.dw.read = bar_read, \
++ .u.dw.write = rom_write, \
++ }
++
++static struct config_field header_0[] = {
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_0),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_1),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_2),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_3),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_4),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_5),
++ CFG_FIELD_ROM(PCI_ROM_ADDRESS),
++ {
++ .size = 0,
++ },
++};
++
++static struct config_field header_1[] = {
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_0),
++ CFG_FIELD_BAR(PCI_BASE_ADDRESS_1),
++ CFG_FIELD_ROM(PCI_ROM_ADDRESS1),
++ {
++ .size = 0,
++ },
++};
++
++int pciback_config_header_add_fields(struct pci_dev *dev)
++{
++ int err;
++
++ err = pciback_config_add_fields(dev, header_common);
++ if (err)
++ goto out;
++
++ switch (dev->hdr_type) {
++ case PCI_HEADER_TYPE_NORMAL:
++ err = pciback_config_add_fields(dev, header_0);
++ break;
++
++ case PCI_HEADER_TYPE_BRIDGE:
++ err = pciback_config_add_fields(dev, header_1);
++ break;
++
++ default:
++ err = -EINVAL;
++ printk(KERN_ERR "pciback: %s: Unsupported header type %d!\n",
++ pci_name(dev), dev->hdr_type);
++ break;
++ }
++
++ out:
++ return err;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/Makefile new/drivers/xen/pciback/Makefile
+--- linux-2.6/drivers/xen/pciback/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/Makefile 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,13 @@
++obj-$(CONFIG_XEN_PCIDEV_BACKEND) += pciback.o
++
++pciback-y := pci_stub.o pciback_ops.o xenbus.o
++pciback-y += conf_space.o conf_space_header.o \
++ conf_space_capability.o \
++ conf_space_capability_vpd.o \
++ conf_space_capability_pm.o
++pciback-$(CONFIG_XEN_PCIDEV_BACKEND_VPCI) += vpci.o
++pciback-$(CONFIG_XEN_PCIDEV_BACKEND_PASS) += passthrough.o
++
++ifeq ($(CONFIG_XEN_PCIDEV_BE_DEBUG),y)
++EXTRA_CFLAGS += -DDEBUG
++endif
+diff -urNp linux-2.6/drivers/xen/pciback/passthrough.c new/drivers/xen/pciback/passthrough.c
+--- linux-2.6/drivers/xen/pciback/passthrough.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/passthrough.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,157 @@
++/*
++ * PCI Backend - Provides restricted access to the real PCI bus topology
++ * to the frontend
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/list.h>
++#include <linux/pci.h>
++#include <linux/spinlock.h>
++#include "pciback.h"
++
++struct passthrough_dev_data {
++ /* Access to dev_list must be protected by lock */
++ struct list_head dev_list;
++ spinlock_t lock;
++};
++
++struct pci_dev *pciback_get_pci_dev(struct pciback_device *pdev,
++ unsigned int domain, unsigned int bus,
++ unsigned int devfn)
++{
++ struct passthrough_dev_data *dev_data = pdev->pci_dev_data;
++ struct pci_dev_entry *dev_entry;
++ struct pci_dev *dev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&dev_data->lock, flags);
++
++ list_for_each_entry(dev_entry, &dev_data->dev_list, list) {
++ if (domain == (unsigned int)pci_domain_nr(dev_entry->dev->bus)
++ && bus == (unsigned int)dev_entry->dev->bus->number
++ && devfn == dev_entry->dev->devfn) {
++ dev = dev_entry->dev;
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&dev_data->lock, flags);
++
++ return dev;
++}
++
++int pciback_add_pci_dev(struct pciback_device *pdev, struct pci_dev *dev)
++{
++ struct passthrough_dev_data *dev_data = pdev->pci_dev_data;
++ struct pci_dev_entry *dev_entry;
++ unsigned long flags;
++
++ dev_entry = kmalloc(sizeof(*dev_entry), GFP_KERNEL);
++ if (!dev_entry)
++ return -ENOMEM;
++ dev_entry->dev = dev;
++
++ spin_lock_irqsave(&dev_data->lock, flags);
++ list_add_tail(&dev_entry->list, &dev_data->dev_list);
++ spin_unlock_irqrestore(&dev_data->lock, flags);
++
++ return 0;
++}
++
++void pciback_release_pci_dev(struct pciback_device *pdev, struct pci_dev *dev)
++{
++ struct passthrough_dev_data *dev_data = pdev->pci_dev_data;
++ struct pci_dev_entry *dev_entry, *t;
++ struct pci_dev *found_dev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&dev_data->lock, flags);
++
++ list_for_each_entry_safe(dev_entry, t, &dev_data->dev_list, list) {
++ if (dev_entry->dev == dev) {
++ list_del(&dev_entry->list);
++ found_dev = dev_entry->dev;
++ kfree(dev_entry);
++ }
++ }
++
++ spin_unlock_irqrestore(&dev_data->lock, flags);
++
++ if (found_dev)
++ pcistub_put_pci_dev(found_dev);
++}
++
++int pciback_init_devices(struct pciback_device *pdev)
++{
++ struct passthrough_dev_data *dev_data;
++
++ dev_data = kmalloc(sizeof(*dev_data), GFP_KERNEL);
++ if (!dev_data)
++ return -ENOMEM;
++
++ spin_lock_init(&dev_data->lock);
++
++ INIT_LIST_HEAD(&dev_data->dev_list);
++
++ pdev->pci_dev_data = dev_data;
++
++ return 0;
++}
++
++int pciback_publish_pci_roots(struct pciback_device *pdev,
++ publish_pci_root_cb publish_root_cb)
++{
++ int err = 0;
++ struct passthrough_dev_data *dev_data = pdev->pci_dev_data;
++ struct pci_dev_entry *dev_entry, *e;
++ struct pci_dev *dev;
++ int found;
++ unsigned int domain, bus;
++
++ spin_lock(&dev_data->lock);
++
++ list_for_each_entry(dev_entry, &dev_data->dev_list, list) {
++ /* Only publish this device as a root if none of its
++ * parent bridges are exported
++ */
++ found = 0;
++ dev = dev_entry->dev->bus->self;
++ for (; !found && dev != NULL; dev = dev->bus->self) {
++ list_for_each_entry(e, &dev_data->dev_list, list) {
++ if (dev == e->dev) {
++ found = 1;
++ break;
++ }
++ }
++ }
++
++ domain = (unsigned int)pci_domain_nr(dev_entry->dev->bus);
++ bus = (unsigned int)dev_entry->dev->bus->number;
++
++ if (!found) {
++ err = publish_root_cb(pdev, domain, bus);
++ if (err)
++ break;
++ }
++ }
++
++ spin_unlock(&dev_data->lock);
++
++ return err;
++}
++
++void pciback_release_devices(struct pciback_device *pdev)
++{
++ struct passthrough_dev_data *dev_data = pdev->pci_dev_data;
++ struct pci_dev_entry *dev_entry, *t;
++
++ list_for_each_entry_safe(dev_entry, t, &dev_data->dev_list, list) {
++ list_del(&dev_entry->list);
++ pcistub_put_pci_dev(dev_entry->dev);
++ kfree(dev_entry);
++ }
++
++ kfree(dev_data);
++ pdev->pci_dev_data = NULL;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/pciback.h new/drivers/xen/pciback/pciback.h
+--- linux-2.6/drivers/xen/pciback/pciback.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/pciback.h 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,91 @@
++/*
++ * PCI Backend Common Data Structures & Function Declarations
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#ifndef __XEN_PCIBACK_H__
++#define __XEN_PCIBACK_H__
++
++#include <linux/pci.h>
++#include <linux/interrupt.h>
++#include <xen/xenbus.h>
++#include <linux/list.h>
++#include <linux/spinlock.h>
++#include <linux/workqueue.h>
++#include <asm/atomic.h>
++#include <xen/interface/io/pciif.h>
++
++struct pci_dev_entry {
++ struct list_head list;
++ struct pci_dev *dev;
++};
++
++#define _PDEVF_op_active (0)
++#define PDEVF_op_active (1<<(_PDEVF_op_active))
++
++struct pciback_device {
++ void *pci_dev_data;
++ spinlock_t dev_lock;
++
++ struct xenbus_device *xdev;
++
++ struct xenbus_watch be_watch;
++ u8 be_watching;
++
++ int evtchn_irq;
++
++ struct vm_struct *sh_area;
++ struct xen_pci_sharedinfo *sh_info;
++
++ unsigned long flags;
++
++ struct work_struct op_work;
++};
++
++struct pciback_dev_data {
++ struct list_head config_fields;
++ int warned_on_write;
++};
++
++/* Get/Put PCI Devices that are hidden from the PCI Backend Domain */
++struct pci_dev *pcistub_get_pci_dev_by_slot(struct pciback_device *pdev,
++ int domain, int bus,
++ int slot, int func);
++struct pci_dev *pcistub_get_pci_dev(struct pciback_device *pdev,
++ struct pci_dev *dev);
++void pcistub_put_pci_dev(struct pci_dev *dev);
++
++/* Ensure a device is turned off or reset */
++void pciback_reset_device(struct pci_dev *pdev);
++
++/* Access a virtual configuration space for a PCI device */
++int pciback_config_init(void);
++int pciback_config_init_dev(struct pci_dev *dev);
++void pciback_config_reset_dev(struct pci_dev *dev);
++void pciback_config_free_dev(struct pci_dev *dev);
++int pciback_config_read(struct pci_dev *dev, int offset, int size,
++ u32 * ret_val);
++int pciback_config_write(struct pci_dev *dev, int offset, int size, u32 value);
++
++/* Handle requests for specific devices from the frontend */
++typedef int (*publish_pci_root_cb) (struct pciback_device * pdev,
++ unsigned int domain, unsigned int bus);
++int pciback_add_pci_dev(struct pciback_device *pdev, struct pci_dev *dev);
++void pciback_release_pci_dev(struct pciback_device *pdev, struct pci_dev *dev);
++struct pci_dev *pciback_get_pci_dev(struct pciback_device *pdev,
++ unsigned int domain, unsigned int bus,
++ unsigned int devfn);
++int pciback_init_devices(struct pciback_device *pdev);
++int pciback_publish_pci_roots(struct pciback_device *pdev,
++ publish_pci_root_cb cb);
++void pciback_release_devices(struct pciback_device *pdev);
++
++/* Handles events from front-end */
++irqreturn_t pciback_handle_event(int irq, void *dev_id, struct pt_regs *regs);
++void pciback_do_op(void *data);
++
++int pciback_xenbus_register(void);
++void pciback_xenbus_unregister(void);
++
++extern int verbose_request;
++#endif
+diff -urNp linux-2.6/drivers/xen/pciback/pciback_ops.c new/drivers/xen/pciback/pciback_ops.c
+--- linux-2.6/drivers/xen/pciback/pciback_ops.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/pciback_ops.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,95 @@
++/*
++ * PCI Backend Operations - respond to PCI requests from Frontend
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <asm/bitops.h>
++#include <xen/evtchn.h>
++#include "pciback.h"
++
++int verbose_request = 0;
++module_param(verbose_request, int, 0644);
++
++/* Ensure a device is "turned off" and ready to be exported.
++ * (Also see pciback_config_reset to ensure virtual configuration space is
++ * ready to be re-exported)
++ */
++void pciback_reset_device(struct pci_dev *dev)
++{
++ u16 cmd;
++
++ /* Disable devices (but not bridges) */
++ if (dev->hdr_type == PCI_HEADER_TYPE_NORMAL) {
++ pci_disable_device(dev);
++
++ pci_write_config_word(dev, PCI_COMMAND, 0);
++
++ dev->is_enabled = 0;
++ dev->is_busmaster = 0;
++ } else {
++ pci_read_config_word(dev, PCI_COMMAND, &cmd);
++ if (cmd & (PCI_COMMAND_INVALIDATE)) {
++ cmd &= ~(PCI_COMMAND_INVALIDATE);
++ pci_write_config_word(dev, PCI_COMMAND, cmd);
++
++ dev->is_busmaster = 0;
++ }
++ }
++}
++
++static inline void test_and_schedule_op(struct pciback_device *pdev)
++{
++ /* Check that frontend is requesting an operation and that we are not
++ * already processing a request */
++ if (test_bit(_XEN_PCIF_active, (unsigned long *)&pdev->sh_info->flags)
++ && !test_and_set_bit(_PDEVF_op_active, &pdev->flags))
++ schedule_work(&pdev->op_work);
++}
++
++/* Performing the configuration space reads/writes must not be done in atomic
++ * context because some of the pci_* functions can sleep (mostly due to ACPI
++ * use of semaphores). This function is intended to be called from a work
++ * queue in process context taking a struct pciback_device as a parameter */
++void pciback_do_op(void *data)
++{
++ struct pciback_device *pdev = data;
++ struct pci_dev *dev;
++ struct xen_pci_op *op = &pdev->sh_info->op;
++
++ dev = pciback_get_pci_dev(pdev, op->domain, op->bus, op->devfn);
++
++ if (dev == NULL)
++ op->err = XEN_PCI_ERR_dev_not_found;
++ else if (op->cmd == XEN_PCI_OP_conf_read)
++ op->err = pciback_config_read(dev, op->offset, op->size,
++ &op->value);
++ else if (op->cmd == XEN_PCI_OP_conf_write)
++ op->err = pciback_config_write(dev, op->offset, op->size,
++ op->value);
++ else
++ op->err = XEN_PCI_ERR_not_implemented;
++
++ /* Tell the driver domain that we're done. */
++ wmb();
++ clear_bit(_XEN_PCIF_active, (unsigned long *)&pdev->sh_info->flags);
++ notify_remote_via_irq(pdev->evtchn_irq);
++
++ /* Mark that we're done. */
++ smp_mb__before_clear_bit(); /* /after/ clearing PCIF_active */
++ clear_bit(_PDEVF_op_active, &pdev->flags);
++ smp_mb__after_clear_bit(); /* /before/ final check for work */
++
++ /* Check to see if the driver domain tried to start another request in
++ * between clearing _XEN_PCIF_active and clearing _PDEVF_op_active. */
++ test_and_schedule_op(pdev);
++}
++
++irqreturn_t pciback_handle_event(int irq, void *dev_id, struct pt_regs *regs)
++{
++ struct pciback_device *pdev = dev_id;
++
++ test_and_schedule_op(pdev);
++
++ return IRQ_HANDLED;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/pci_stub.c new/drivers/xen/pciback/pci_stub.c
+--- linux-2.6/drivers/xen/pciback/pci_stub.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/pci_stub.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,690 @@
++/*
++ * PCI Stub Driver - Grabs devices in backend to be exported later
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/list.h>
++#include <linux/spinlock.h>
++#include <linux/kref.h>
++#include <asm/atomic.h>
++#include "pciback.h"
++
++static char *pci_devs_to_hide = NULL;
++module_param_named(hide, pci_devs_to_hide, charp, 0444);
++
++struct pcistub_device_id {
++ struct list_head slot_list;
++ int domain;
++ unsigned char bus;
++ unsigned int devfn;
++};
++static LIST_HEAD(pcistub_device_ids);
++static DEFINE_SPINLOCK(device_ids_lock);
++
++struct pcistub_device {
++ struct kref kref;
++ struct list_head dev_list;
++ spinlock_t lock;
++
++ struct pci_dev *dev;
++ struct pciback_device *pdev; /* non-NULL if struct pci_dev is in use */
++};
++/* Access to pcistub_devices & seized_devices lists and the initialize_devices
++ * flag must be locked with pcistub_devices_lock
++ */
++static DEFINE_SPINLOCK(pcistub_devices_lock);
++static LIST_HEAD(pcistub_devices);
++
++/* wait for device_initcall before initializing our devices
++ * (see pcistub_init_devices_late)
++ */
++static int initialize_devices = 0;
++static LIST_HEAD(seized_devices);
++
++static struct pcistub_device *pcistub_device_alloc(struct pci_dev *dev)
++{
++ struct pcistub_device *psdev;
++
++ dev_dbg(&dev->dev, "pcistub_device_alloc\n");
++
++ psdev = kzalloc(sizeof(*psdev), GFP_ATOMIC);
++ if (!psdev)
++ return NULL;
++
++ psdev->dev = pci_dev_get(dev);
++ if (!psdev->dev) {
++ kfree(psdev);
++ return NULL;
++ }
++
++ kref_init(&psdev->kref);
++ spin_lock_init(&psdev->lock);
++
++ return psdev;
++}
++
++/* Don't call this directly as it's called by pcistub_device_put */
++static void pcistub_device_release(struct kref *kref)
++{
++ struct pcistub_device *psdev;
++
++ psdev = container_of(kref, struct pcistub_device, kref);
++
++ dev_dbg(&psdev->dev->dev, "pcistub_device_release\n");
++
++ /* Clean-up the device */
++ pciback_reset_device(psdev->dev);
++ pciback_config_free_dev(psdev->dev);
++ kfree(pci_get_drvdata(psdev->dev));
++ pci_set_drvdata(psdev->dev, NULL);
++
++ pci_dev_put(psdev->dev);
++
++ kfree(psdev);
++}
++
++static inline void pcistub_device_get(struct pcistub_device *psdev)
++{
++ kref_get(&psdev->kref);
++}
++
++static inline void pcistub_device_put(struct pcistub_device *psdev)
++{
++ kref_put(&psdev->kref, pcistub_device_release);
++}
++
++static struct pci_dev *pcistub_device_get_pci_dev(struct pciback_device *pdev,
++ struct pcistub_device *psdev)
++{
++ struct pci_dev *pci_dev = NULL;
++ unsigned long flags;
++
++ pcistub_device_get(psdev);
++
++ spin_lock_irqsave(&psdev->lock, flags);
++ if (!psdev->pdev) {
++ psdev->pdev = pdev;
++ pci_dev = psdev->dev;
++ }
++ spin_unlock_irqrestore(&psdev->lock, flags);
++
++ if (!pci_dev)
++ pcistub_device_put(psdev);
++
++ return pci_dev;
++}
++
++struct pci_dev *pcistub_get_pci_dev_by_slot(struct pciback_device *pdev,
++ int domain, int bus,
++ int slot, int func)
++{
++ struct pcistub_device *psdev;
++ struct pci_dev *found_dev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ list_for_each_entry(psdev, &pcistub_devices, dev_list) {
++ if (psdev->dev != NULL
++ && domain == pci_domain_nr(psdev->dev->bus)
++ && bus == psdev->dev->bus->number
++ && PCI_DEVFN(slot, func) == psdev->dev->devfn) {
++ found_dev = pcistub_device_get_pci_dev(pdev, psdev);
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++ return found_dev;
++}
++
++struct pci_dev *pcistub_get_pci_dev(struct pciback_device *pdev,
++ struct pci_dev *dev)
++{
++ struct pcistub_device *psdev;
++ struct pci_dev *found_dev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ list_for_each_entry(psdev, &pcistub_devices, dev_list) {
++ if (psdev->dev == dev) {
++ found_dev = pcistub_device_get_pci_dev(pdev, psdev);
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++ return found_dev;
++}
++
++void pcistub_put_pci_dev(struct pci_dev *dev)
++{
++ struct pcistub_device *psdev, *found_psdev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ list_for_each_entry(psdev, &pcistub_devices, dev_list) {
++ if (psdev->dev == dev) {
++ found_psdev = psdev;
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ /* Cleanup our device
++ * (so it's ready for the next domain)
++ */
++ pciback_reset_device(found_psdev->dev);
++ pciback_config_reset_dev(found_psdev->dev);
++
++ spin_lock_irqsave(&found_psdev->lock, flags);
++ found_psdev->pdev = NULL;
++ spin_unlock_irqrestore(&found_psdev->lock, flags);
++
++ pcistub_device_put(found_psdev);
++}
++
++static int __devinit pcistub_match_one(struct pci_dev *dev,
++ struct pcistub_device_id *pdev_id)
++{
++ /* Match the specified device by domain, bus, slot, func and also if
++ * any of the device's parent bridges match.
++ */
++ for (; dev != NULL; dev = dev->bus->self) {
++ if (pci_domain_nr(dev->bus) == pdev_id->domain
++ && dev->bus->number == pdev_id->bus
++ && dev->devfn == pdev_id->devfn)
++ return 1;
++ }
++
++ return 0;
++}
++
++static int __devinit pcistub_match(struct pci_dev *dev)
++{
++ struct pcistub_device_id *pdev_id;
++ unsigned long flags;
++ int found = 0;
++
++ spin_lock_irqsave(&device_ids_lock, flags);
++ list_for_each_entry(pdev_id, &pcistub_device_ids, slot_list) {
++ if (pcistub_match_one(dev, pdev_id)) {
++ found = 1;
++ break;
++ }
++ }
++ spin_unlock_irqrestore(&device_ids_lock, flags);
++
++ return found;
++}
++
++static int __devinit pcistub_init_device(struct pci_dev *dev)
++{
++ struct pciback_dev_data *dev_data;
++ int err = 0;
++
++ dev_dbg(&dev->dev, "initializing...\n");
++
++ /* The PCI backend is not intended to be a module (or to work with
++ * removable PCI devices (yet). If it were, pciback_config_free()
++ * would need to be called somewhere to free the memory allocated
++ * here and then to call kfree(pci_get_drvdata(psdev->dev)).
++ */
++ dev_data = kzalloc(sizeof(*dev_data), GFP_ATOMIC);
++ if (!dev_data) {
++ err = -ENOMEM;
++ goto out;
++ }
++ pci_set_drvdata(dev, dev_data);
++
++ dev_dbg(&dev->dev, "initializing config\n");
++ err = pciback_config_init_dev(dev);
++ if (err)
++ goto out;
++
++ /* HACK: Force device (& ACPI) to determine what IRQ it's on - we
++ * must do this here because pcibios_enable_device may specify
++ * the pci device's true irq (and possibly its other resources)
++ * if they differ from what's in the configuration space.
++ * This makes the assumption that the device's resources won't
++ * change after this point (otherwise this code may break!)
++ */
++ dev_dbg(&dev->dev, "enabling device\n");
++ err = pci_enable_device(dev);
++ if (err)
++ goto config_release;
++
++ /* Now disable the device (this also ensures some private device
++ * data is setup before we export)
++ */
++ dev_dbg(&dev->dev, "reset device\n");
++ pciback_reset_device(dev);
++
++ return 0;
++
++ config_release:
++ pciback_config_free_dev(dev);
++
++ out:
++ pci_set_drvdata(dev, NULL);
++ kfree(dev_data);
++ return err;
++}
++
++/*
++ * Because some initialization still happens on
++ * devices during fs_initcall, we need to defer
++ * full initialization of our devices until
++ * device_initcall.
++ */
++static int __init pcistub_init_devices_late(void)
++{
++ struct pcistub_device *psdev;
++ unsigned long flags;
++ int err = 0;
++
++ pr_debug("pciback: pcistub_init_devices_late\n");
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ while (!list_empty(&seized_devices)) {
++ psdev = container_of(seized_devices.next,
++ struct pcistub_device, dev_list);
++ list_del(&psdev->dev_list);
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ err = pcistub_init_device(psdev->dev);
++ if (err) {
++ dev_err(&psdev->dev->dev,
++ "error %d initializing device\n", err);
++ kfree(psdev);
++ psdev = NULL;
++ }
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ if (psdev)
++ list_add_tail(&psdev->dev_list, &pcistub_devices);
++ }
++
++ initialize_devices = 1;
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ return 0;
++}
++
++static int __devinit pcistub_seize(struct pci_dev *dev)
++{
++ struct pcistub_device *psdev;
++ unsigned long flags;
++ int err = 0;
++
++ psdev = pcistub_device_alloc(dev);
++ if (!psdev)
++ return -ENOMEM;
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ if (initialize_devices) {
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ /* don't want irqs disabled when calling pcistub_init_device */
++ err = pcistub_init_device(psdev->dev);
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ if (!err)
++ list_add(&psdev->dev_list, &pcistub_devices);
++ } else {
++ dev_dbg(&dev->dev, "deferring initialization\n");
++ list_add(&psdev->dev_list, &seized_devices);
++ }
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ if (err)
++ pcistub_device_put(psdev);
++
++ return err;
++}
++
++static int __devinit pcistub_probe(struct pci_dev *dev,
++ const struct pci_device_id *id)
++{
++ int err = 0;
++
++ dev_dbg(&dev->dev, "probing...\n");
++
++ if (pcistub_match(dev)) {
++
++ if (dev->hdr_type != PCI_HEADER_TYPE_NORMAL
++ && dev->hdr_type != PCI_HEADER_TYPE_BRIDGE) {
++ dev_err(&dev->dev, "can't export pci devices that "
++ "don't have a normal (0) or bridge (1) "
++ "header type!\n");
++ err = -ENODEV;
++ goto out;
++ }
++
++ dev_info(&dev->dev, "seizing device\n");
++ err = pcistub_seize(dev);
++ } else
++ /* Didn't find the device */
++ err = -ENODEV;
++
++ out:
++ return err;
++}
++
++static void pcistub_remove(struct pci_dev *dev)
++{
++ struct pcistub_device *psdev, *found_psdev = NULL;
++ unsigned long flags;
++
++ dev_dbg(&dev->dev, "removing\n");
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++
++ list_for_each_entry(psdev, &pcistub_devices, dev_list) {
++ if (psdev->dev == dev) {
++ found_psdev = psdev;
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ if (found_psdev) {
++ dev_dbg(&dev->dev, "found device to remove - in use? %p\n",
++ found_psdev->pdev);
++
++ if (found_psdev->pdev) {
++ printk(KERN_WARNING "pciback: ****** removing device "
++ "%s while still in-use! ******\n",
++ pci_name(found_psdev->dev));
++ printk(KERN_WARNING "pciback: ****** driver domain may "
++ "still access this device's i/o resources!\n");
++ printk(KERN_WARNING "pciback: ****** shutdown driver "
++ "domain before binding device\n");
++ printk(KERN_WARNING "pciback: ****** to other drivers "
++ "or domains\n");
++
++ pciback_release_pci_dev(found_psdev->pdev,
++ found_psdev->dev);
++ }
++
++ spin_lock_irqsave(&pcistub_devices_lock, flags);
++ list_del(&found_psdev->dev_list);
++ spin_unlock_irqrestore(&pcistub_devices_lock, flags);
++
++ /* the final put for releasing from the list */
++ pcistub_device_put(found_psdev);
++ }
++}
++
++static struct pci_device_id pcistub_ids[] = {
++ {
++ .vendor = PCI_ANY_ID,
++ .device = PCI_ANY_ID,
++ .subvendor = PCI_ANY_ID,
++ .subdevice = PCI_ANY_ID,
++ },
++ {0,},
++};
++
++/*
++ * Note: There is no MODULE_DEVICE_TABLE entry here because this isn't
++ * for a normal device. I don't want it to be loaded automatically.
++ */
++
++static struct pci_driver pciback_pci_driver = {
++ .name = "pciback",
++ .id_table = pcistub_ids,
++ .probe = pcistub_probe,
++ .remove = pcistub_remove,
++};
++
++static inline int str_to_slot(const char *buf, int *domain, int *bus,
++ int *slot, int *func)
++{
++ int err;
++
++ err = sscanf(buf, " %x:%x:%x.%x", domain, bus, slot, func);
++ if (err == 4)
++ return 0;
++ else if (err < 0)
++ return -EINVAL;
++
++ /* try again without domain */
++ *domain = 0;
++ err = sscanf(buf, " %x:%x.%x", bus, slot, func);
++ if (err == 3)
++ return 0;
++
++ return -EINVAL;
++}
++
++static int pcistub_device_id_add(int domain, int bus, int slot, int func)
++{
++ struct pcistub_device_id *pci_dev_id;
++ unsigned long flags;
++
++ pci_dev_id = kmalloc(sizeof(*pci_dev_id), GFP_KERNEL);
++ if (!pci_dev_id)
++ return -ENOMEM;
++
++ pci_dev_id->domain = domain;
++ pci_dev_id->bus = bus;
++ pci_dev_id->devfn = PCI_DEVFN(slot, func);
++
++ pr_debug("pciback: wants to seize %04x:%02x:%02x.%01x\n",
++ domain, bus, slot, func);
++
++ spin_lock_irqsave(&device_ids_lock, flags);
++ list_add_tail(&pci_dev_id->slot_list, &pcistub_device_ids);
++ spin_unlock_irqrestore(&device_ids_lock, flags);
++
++ return 0;
++}
++
++static int pcistub_device_id_remove(int domain, int bus, int slot, int func)
++{
++ struct pcistub_device_id *pci_dev_id, *t;
++ int devfn = PCI_DEVFN(slot, func);
++ int err = -ENOENT;
++ unsigned long flags;
++
++ spin_lock_irqsave(&device_ids_lock, flags);
++ list_for_each_entry_safe(pci_dev_id, t, &pcistub_device_ids, slot_list) {
++
++ if (pci_dev_id->domain == domain
++ && pci_dev_id->bus == bus && pci_dev_id->devfn == devfn) {
++ /* Don't break; here because it's possible the same
++ * slot could be in the list more than once
++ */
++ list_del(&pci_dev_id->slot_list);
++ kfree(pci_dev_id);
++
++ err = 0;
++
++ pr_debug("pciback: removed %04x:%02x:%02x.%01x from "
++ "seize list\n", domain, bus, slot, func);
++ }
++ }
++ spin_unlock_irqrestore(&device_ids_lock, flags);
++
++ return err;
++}
++
++static ssize_t pcistub_slot_add(struct device_driver *drv, const char *buf,
++ size_t count)
++{
++ int domain, bus, slot, func;
++ int err;
++
++ err = str_to_slot(buf, &domain, &bus, &slot, &func);
++ if (err)
++ goto out;
++
++ err = pcistub_device_id_add(domain, bus, slot, func);
++
++ out:
++ if (!err)
++ err = count;
++ return err;
++}
++
++DRIVER_ATTR(new_slot, S_IWUSR, NULL, pcistub_slot_add);
++
++static ssize_t pcistub_slot_remove(struct device_driver *drv, const char *buf,
++ size_t count)
++{
++ int domain, bus, slot, func;
++ int err;
++
++ err = str_to_slot(buf, &domain, &bus, &slot, &func);
++ if (err)
++ goto out;
++
++ err = pcistub_device_id_remove(domain, bus, slot, func);
++
++ out:
++ if (!err)
++ err = count;
++ return err;
++}
++
++DRIVER_ATTR(remove_slot, S_IWUSR, NULL, pcistub_slot_remove);
++
++static ssize_t pcistub_slot_show(struct device_driver *drv, char *buf)
++{
++ struct pcistub_device_id *pci_dev_id;
++ size_t count = 0;
++ unsigned long flags;
++
++ spin_lock_irqsave(&device_ids_lock, flags);
++ list_for_each_entry(pci_dev_id, &pcistub_device_ids, slot_list) {
++ if (count >= PAGE_SIZE)
++ break;
++
++ count += scnprintf(buf + count, PAGE_SIZE - count,
++ "%04x:%02x:%02x.%01x\n",
++ pci_dev_id->domain, pci_dev_id->bus,
++ PCI_SLOT(pci_dev_id->devfn),
++ PCI_FUNC(pci_dev_id->devfn));
++ }
++ spin_unlock_irqrestore(&device_ids_lock, flags);
++
++ return count;
++}
++
++DRIVER_ATTR(slots, S_IRUSR, pcistub_slot_show, NULL);
++
++static int __init pcistub_init(void)
++{
++ int pos = 0;
++ int err = 0;
++ int domain, bus, slot, func;
++ int parsed;
++
++ if (pci_devs_to_hide && *pci_devs_to_hide) {
++ do {
++ parsed = 0;
++
++ err = sscanf(pci_devs_to_hide + pos,
++ " (%x:%x:%x.%x) %n",
++ &domain, &bus, &slot, &func, &parsed);
++ if (err != 4) {
++ domain = 0;
++ err = sscanf(pci_devs_to_hide + pos,
++ " (%x:%x.%x) %n",
++ &bus, &slot, &func, &parsed);
++ if (err != 3)
++ goto parse_error;
++ }
++
++ err = pcistub_device_id_add(domain, bus, slot, func);
++ if (err)
++ goto out;
++
++ /* if parsed<=0, we've reached the end of the string */
++ pos += parsed;
++ } while (parsed > 0 && pci_devs_to_hide[pos]);
++ }
++
++ /* If we're the first PCI Device Driver to register, we're the
++ * first one to get offered PCI devices as they become
++ * available (and thus we can be the first to grab them)
++ */
++ err = pci_register_driver(&pciback_pci_driver);
++ if (err < 0)
++ goto out;
++
++ driver_create_file(&pciback_pci_driver.driver, &driver_attr_new_slot);
++ driver_create_file(&pciback_pci_driver.driver,
++ &driver_attr_remove_slot);
++ driver_create_file(&pciback_pci_driver.driver, &driver_attr_slots);
++
++ out:
++ return err;
++
++ parse_error:
++ printk(KERN_ERR "pciback: Error parsing pci_devs_to_hide at \"%s\"\n",
++ pci_devs_to_hide + pos);
++ return -EINVAL;
++}
++
++#ifndef MODULE
++/*
++ * fs_initcall happens before device_initcall
++ * so pciback *should* get called first (b/c we
++ * want to suck up any device before other drivers
++ * get a chance by being the first pci device
++ * driver to register)
++ */
++fs_initcall(pcistub_init);
++#endif
++
++static int __init pciback_init(void)
++{
++ int err;
++
++ err = pciback_config_init();
++ if (err)
++ return err;
++
++#ifdef MODULE
++ err = pcistub_init();
++ if (err < 0)
++ return err;
++#endif
++
++ pcistub_init_devices_late();
++ pciback_xenbus_register();
++
++ return 0;
++}
++
++static void __exit pciback_cleanup(void)
++{
++ pciback_xenbus_unregister();
++
++ driver_remove_file(&pciback_pci_driver.driver, &driver_attr_new_slot);
++ driver_remove_file(&pciback_pci_driver.driver,
++ &driver_attr_remove_slot);
++ driver_remove_file(&pciback_pci_driver.driver, &driver_attr_slots);
++
++ pci_unregister_driver(&pciback_pci_driver);
++}
++
++module_init(pciback_init);
++module_exit(pciback_cleanup);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/pciback/vpci.c new/drivers/xen/pciback/vpci.c
+--- linux-2.6/drivers/xen/pciback/vpci.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/vpci.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,204 @@
++/*
++ * PCI Backend - Provides a Virtual PCI bus (with real devices)
++ * to the frontend
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++
++#include <linux/list.h>
++#include <linux/slab.h>
++#include <linux/pci.h>
++#include <linux/spinlock.h>
++#include "pciback.h"
++
++#define PCI_SLOT_MAX 32
++
++struct vpci_dev_data {
++ /* Access to dev_list must be protected by lock */
++ struct list_head dev_list[PCI_SLOT_MAX];
++ spinlock_t lock;
++};
++
++static inline struct list_head *list_first(struct list_head *head)
++{
++ return head->next;
++}
++
++struct pci_dev *pciback_get_pci_dev(struct pciback_device *pdev,
++ unsigned int domain, unsigned int bus,
++ unsigned int devfn)
++{
++ struct pci_dev_entry *entry;
++ struct pci_dev *dev = NULL;
++ struct vpci_dev_data *vpci_dev = pdev->pci_dev_data;
++ unsigned long flags;
++
++ if (domain != 0 || bus != 0)
++ return NULL;
++
++ if (PCI_SLOT(devfn) < PCI_SLOT_MAX) {
++ spin_lock_irqsave(&vpci_dev->lock, flags);
++
++ list_for_each_entry(entry,
++ &vpci_dev->dev_list[PCI_SLOT(devfn)],
++ list) {
++ if (PCI_FUNC(entry->dev->devfn) == PCI_FUNC(devfn)) {
++ dev = entry->dev;
++ break;
++ }
++ }
++
++ spin_unlock_irqrestore(&vpci_dev->lock, flags);
++ }
++ return dev;
++}
++
++static inline int match_slot(struct pci_dev *l, struct pci_dev *r)
++{
++ if (pci_domain_nr(l->bus) == pci_domain_nr(r->bus)
++ && l->bus == r->bus && PCI_SLOT(l->devfn) == PCI_SLOT(r->devfn))
++ return 1;
++
++ return 0;
++}
++
++int pciback_add_pci_dev(struct pciback_device *pdev, struct pci_dev *dev)
++{
++ int err = 0, slot;
++ struct pci_dev_entry *t, *dev_entry;
++ struct vpci_dev_data *vpci_dev = pdev->pci_dev_data;
++ unsigned long flags;
++
++ if ((dev->class >> 24) == PCI_BASE_CLASS_BRIDGE) {
++ err = -EFAULT;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Can't export bridges on the virtual PCI bus");
++ goto out;
++ }
++
++ dev_entry = kmalloc(sizeof(*dev_entry), GFP_KERNEL);
++ if (!dev_entry) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error adding entry to virtual PCI bus");
++ goto out;
++ }
++
++ dev_entry->dev = dev;
++
++ spin_lock_irqsave(&vpci_dev->lock, flags);
++
++ /* Keep multi-function devices together on the virtual PCI bus */
++ for (slot = 0; slot < PCI_SLOT_MAX; slot++) {
++ if (!list_empty(&vpci_dev->dev_list[slot])) {
++ t = list_entry(list_first(&vpci_dev->dev_list[slot]),
++ struct pci_dev_entry, list);
++
++ if (match_slot(dev, t->dev)) {
++ pr_info("pciback: vpci: %s: "
++ "assign to virtual slot %d func %d\n",
++ pci_name(dev), slot,
++ PCI_FUNC(dev->devfn));
++ list_add_tail(&dev_entry->list,
++ &vpci_dev->dev_list[slot]);
++ goto unlock;
++ }
++ }
++ }
++
++ /* Assign to a new slot on the virtual PCI bus */
++ for (slot = 0; slot < PCI_SLOT_MAX; slot++) {
++ if (list_empty(&vpci_dev->dev_list[slot])) {
++ printk(KERN_INFO
++ "pciback: vpci: %s: assign to virtual slot %d\n",
++ pci_name(dev), slot);
++ list_add_tail(&dev_entry->list,
++ &vpci_dev->dev_list[slot]);
++ goto unlock;
++ }
++ }
++
++ err = -ENOMEM;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "No more space on root virtual PCI bus");
++
++ unlock:
++ spin_unlock_irqrestore(&vpci_dev->lock, flags);
++ out:
++ return err;
++}
++
++void pciback_release_pci_dev(struct pciback_device *pdev, struct pci_dev *dev)
++{
++ int slot;
++ struct vpci_dev_data *vpci_dev = pdev->pci_dev_data;
++ struct pci_dev *found_dev = NULL;
++ unsigned long flags;
++
++ spin_lock_irqsave(&vpci_dev->lock, flags);
++
++ for (slot = 0; slot < PCI_SLOT_MAX; slot++) {
++ struct pci_dev_entry *e, *tmp;
++ list_for_each_entry_safe(e, tmp, &vpci_dev->dev_list[slot],
++ list) {
++ if (e->dev == dev) {
++ list_del(&e->list);
++ found_dev = e->dev;
++ kfree(e);
++ goto out;
++ }
++ }
++ }
++
++ out:
++ spin_unlock_irqrestore(&vpci_dev->lock, flags);
++
++ if (found_dev)
++ pcistub_put_pci_dev(found_dev);
++}
++
++int pciback_init_devices(struct pciback_device *pdev)
++{
++ int slot;
++ struct vpci_dev_data *vpci_dev;
++
++ vpci_dev = kmalloc(sizeof(*vpci_dev), GFP_KERNEL);
++ if (!vpci_dev)
++ return -ENOMEM;
++
++ spin_lock_init(&vpci_dev->lock);
++
++ for (slot = 0; slot < PCI_SLOT_MAX; slot++) {
++ INIT_LIST_HEAD(&vpci_dev->dev_list[slot]);
++ }
++
++ pdev->pci_dev_data = vpci_dev;
++
++ return 0;
++}
++
++int pciback_publish_pci_roots(struct pciback_device *pdev,
++ publish_pci_root_cb publish_cb)
++{
++ /* The Virtual PCI bus has only one root */
++ return publish_cb(pdev, 0, 0);
++}
++
++void pciback_release_devices(struct pciback_device *pdev)
++{
++ int slot;
++ struct vpci_dev_data *vpci_dev = pdev->pci_dev_data;
++
++ for (slot = 0; slot < PCI_SLOT_MAX; slot++) {
++ struct pci_dev_entry *e, *tmp;
++ list_for_each_entry_safe(e, tmp, &vpci_dev->dev_list[slot],
++ list) {
++ list_del(&e->list);
++ pcistub_put_pci_dev(e->dev);
++ kfree(e);
++ }
++ }
++
++ kfree(vpci_dev);
++ pdev->pci_dev_data = NULL;
++}
+diff -urNp linux-2.6/drivers/xen/pciback/xenbus.c new/drivers/xen/pciback/xenbus.c
+--- linux-2.6/drivers/xen/pciback/xenbus.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pciback/xenbus.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,454 @@
++/*
++ * PCI Backend Xenbus Setup - handles setup with frontend and xend
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/list.h>
++#include <linux/vmalloc.h>
++#include <xen/xenbus.h>
++#include <xen/evtchn.h>
++#include "pciback.h"
++
++#define INVALID_EVTCHN_IRQ (-1)
++
++static struct pciback_device *alloc_pdev(struct xenbus_device *xdev)
++{
++ struct pciback_device *pdev;
++
++ pdev = kzalloc(sizeof(struct pciback_device), GFP_KERNEL);
++ if (pdev == NULL)
++ goto out;
++ dev_dbg(&xdev->dev, "allocated pdev @ 0x%p\n", pdev);
++
++ pdev->xdev = xdev;
++ xdev->dev.driver_data = pdev;
++
++ spin_lock_init(&pdev->dev_lock);
++
++ pdev->sh_area = NULL;
++ pdev->sh_info = NULL;
++ pdev->evtchn_irq = INVALID_EVTCHN_IRQ;
++ pdev->be_watching = 0;
++
++ INIT_WORK(&pdev->op_work, pciback_do_op, pdev);
++
++ if (pciback_init_devices(pdev)) {
++ kfree(pdev);
++ pdev = NULL;
++ }
++ out:
++ return pdev;
++}
++
++static void free_pdev(struct pciback_device *pdev)
++{
++ if (pdev->be_watching)
++ unregister_xenbus_watch(&pdev->be_watch);
++
++ /* Ensure the guest can't trigger our handler before removing devices */
++ if (pdev->evtchn_irq != INVALID_EVTCHN_IRQ)
++ unbind_from_irqhandler(pdev->evtchn_irq, pdev);
++
++ /* If the driver domain started an op, make sure we complete it or
++ * delete it before releasing the shared memory */
++ cancel_delayed_work(&pdev->op_work);
++ flush_scheduled_work();
++
++ if (pdev->sh_info)
++ xenbus_unmap_ring_vfree(pdev->xdev, pdev->sh_area);
++
++ pciback_release_devices(pdev);
++
++ pdev->xdev->dev.driver_data = NULL;
++ pdev->xdev = NULL;
++
++ kfree(pdev);
++}
++
++static int pciback_do_attach(struct pciback_device *pdev, int gnt_ref,
++ int remote_evtchn)
++{
++ int err = 0;
++ int evtchn;
++ struct vm_struct *area;
++
++ dev_dbg(&pdev->xdev->dev,
++ "Attaching to frontend resources - gnt_ref=%d evtchn=%d\n",
++ gnt_ref, remote_evtchn);
++
++ area = xenbus_map_ring_valloc(pdev->xdev, gnt_ref);
++ if (IS_ERR(area)) {
++ err = PTR_ERR(area);
++ goto out;
++ }
++ pdev->sh_area = area;
++ pdev->sh_info = area->addr;
++
++ err = xenbus_bind_evtchn(pdev->xdev, remote_evtchn, &evtchn);
++ if (err)
++ goto out;
++
++ err = bind_evtchn_to_irqhandler(evtchn, pciback_handle_event,
++ SA_SAMPLE_RANDOM, "pciback", pdev);
++ if (err < 0) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error binding event channel to IRQ");
++ goto out;
++ }
++ pdev->evtchn_irq = err;
++ err = 0;
++
++ dev_dbg(&pdev->xdev->dev, "Attached!\n");
++ out:
++ return err;
++}
++
++static int pciback_attach(struct pciback_device *pdev)
++{
++ int err = 0;
++ int gnt_ref, remote_evtchn;
++ char *magic = NULL;
++
++ spin_lock(&pdev->dev_lock);
++
++ /* Make sure we only do this setup once */
++ if (xenbus_read_driver_state(pdev->xdev->nodename) !=
++ XenbusStateInitialised)
++ goto out;
++
++ /* Wait for frontend to state that it has published the configuration */
++ if (xenbus_read_driver_state(pdev->xdev->otherend) !=
++ XenbusStateInitialised)
++ goto out;
++
++ dev_dbg(&pdev->xdev->dev, "Reading frontend config\n");
++
++ err = xenbus_gather(XBT_NIL, pdev->xdev->otherend,
++ "pci-op-ref", "%u", &gnt_ref,
++ "event-channel", "%u", &remote_evtchn,
++ "magic", NULL, &magic, NULL);
++ if (err) {
++ /* If configuration didn't get read correctly, wait longer */
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error reading configuration from frontend");
++ goto out;
++ }
++
++ if (magic == NULL || strcmp(magic, XEN_PCI_MAGIC) != 0) {
++ xenbus_dev_fatal(pdev->xdev, -EFAULT,
++ "version mismatch (%s/%s) with pcifront - "
++ "halting pciback",
++ magic, XEN_PCI_MAGIC);
++ goto out;
++ }
++
++ err = pciback_do_attach(pdev, gnt_ref, remote_evtchn);
++ if (err)
++ goto out;
++
++ dev_dbg(&pdev->xdev->dev, "Connecting...\n");
++
++ err = xenbus_switch_state(pdev->xdev, XenbusStateConnected);
++ if (err)
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error switching to connected state!");
++
++ dev_dbg(&pdev->xdev->dev, "Connected? %d\n", err);
++ out:
++ spin_unlock(&pdev->dev_lock);
++
++ if (magic)
++ kfree(magic);
++
++ return err;
++}
++
++static void pciback_frontend_changed(struct xenbus_device *xdev,
++ enum xenbus_state fe_state)
++{
++ struct pciback_device *pdev = xdev->dev.driver_data;
++
++ dev_dbg(&xdev->dev, "fe state changed %d\n", fe_state);
++
++ switch (fe_state) {
++ case XenbusStateInitialised:
++ pciback_attach(pdev);
++ break;
++
++ case XenbusStateClosing:
++ xenbus_switch_state(xdev, XenbusStateClosing);
++ break;
++
++ case XenbusStateClosed:
++ dev_dbg(&xdev->dev, "frontend is gone! unregister device\n");
++ device_unregister(&xdev->dev);
++ break;
++
++ default:
++ break;
++ }
++}
++
++static int pciback_publish_pci_root(struct pciback_device *pdev,
++ unsigned int domain, unsigned int bus)
++{
++ unsigned int d, b;
++ int i, root_num, len, err;
++ char str[64];
++
++ dev_dbg(&pdev->xdev->dev, "Publishing pci roots\n");
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->nodename,
++ "root_num", "%d", &root_num);
++ if (err == 0 || err == -ENOENT)
++ root_num = 0;
++ else if (err < 0)
++ goto out;
++
++ /* Verify that we haven't already published this pci root */
++ for (i = 0; i < root_num; i++) {
++ len = snprintf(str, sizeof(str), "root-%d", i);
++ if (unlikely(len >= (sizeof(str) - 1))) {
++ err = -ENOMEM;
++ goto out;
++ }
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->nodename,
++ str, "%x:%x", &d, &b);
++ if (err < 0)
++ goto out;
++ if (err != 2) {
++ err = -EINVAL;
++ goto out;
++ }
++
++ if (d == domain && b == bus) {
++ err = 0;
++ goto out;
++ }
++ }
++
++ len = snprintf(str, sizeof(str), "root-%d", root_num);
++ if (unlikely(len >= (sizeof(str) - 1))) {
++ err = -ENOMEM;
++ goto out;
++ }
++
++ dev_dbg(&pdev->xdev->dev, "writing root %d at %04x:%02x\n",
++ root_num, domain, bus);
++
++ err = xenbus_printf(XBT_NIL, pdev->xdev->nodename, str,
++ "%04x:%02x", domain, bus);
++ if (err)
++ goto out;
++
++ err = xenbus_printf(XBT_NIL, pdev->xdev->nodename,
++ "root_num", "%d", (root_num + 1));
++
++ out:
++ return err;
++}
++
++static int pciback_export_device(struct pciback_device *pdev,
++ int domain, int bus, int slot, int func)
++{
++ struct pci_dev *dev;
++ int err = 0;
++
++ dev_dbg(&pdev->xdev->dev, "exporting dom %x bus %x slot %x func %x\n",
++ domain, bus, slot, func);
++
++ dev = pcistub_get_pci_dev_by_slot(pdev, domain, bus, slot, func);
++ if (!dev) {
++ err = -EINVAL;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Couldn't locate PCI device "
++ "(%04x:%02x:%02x.%01x)! "
++ "perhaps already in-use?",
++ domain, bus, slot, func);
++ goto out;
++ }
++
++ err = pciback_add_pci_dev(pdev, dev);
++ if (err)
++ goto out;
++
++ /* TODO: It'd be nice to export a bridge and have all of its children
++ * get exported with it. This may be best done in xend (which will
++ * have to calculate resource usage anyway) but we probably want to
++ * put something in here to ensure that if a bridge gets given to a
++ * driver domain, that all devices under that bridge are not given
++ * to other driver domains (as he who controls the bridge can disable
++ * it and stop the other devices from working).
++ */
++ out:
++ return err;
++}
++
++static int pciback_setup_backend(struct pciback_device *pdev)
++{
++ /* Get configuration from xend (if available now) */
++ int domain, bus, slot, func;
++ int err = 0;
++ int i, num_devs;
++ char dev_str[64];
++
++ spin_lock(&pdev->dev_lock);
++
++ /* It's possible we could get the call to setup twice, so make sure
++ * we're not already connected.
++ */
++ if (xenbus_read_driver_state(pdev->xdev->nodename) !=
++ XenbusStateInitWait)
++ goto out;
++
++ dev_dbg(&pdev->xdev->dev, "getting be setup\n");
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->nodename, "num_devs", "%d",
++ &num_devs);
++ if (err != 1) {
++ if (err >= 0)
++ err = -EINVAL;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error reading number of devices");
++ goto out;
++ }
++
++ for (i = 0; i < num_devs; i++) {
++ int l = snprintf(dev_str, sizeof(dev_str), "dev-%d", i);
++ if (unlikely(l >= (sizeof(dev_str) - 1))) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "String overflow while reading "
++ "configuration");
++ goto out;
++ }
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->nodename, dev_str,
++ "%x:%x:%x.%x", &domain, &bus, &slot, &func);
++ if (err < 0) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error reading device configuration");
++ goto out;
++ }
++ if (err != 4) {
++ err = -EINVAL;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error parsing pci device "
++ "configuration");
++ goto out;
++ }
++
++ err = pciback_export_device(pdev, domain, bus, slot, func);
++ if (err)
++ goto out;
++ }
++
++ err = pciback_publish_pci_roots(pdev, pciback_publish_pci_root);
++ if (err) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error while publish PCI root buses "
++ "for frontend");
++ goto out;
++ }
++
++ err = xenbus_switch_state(pdev->xdev, XenbusStateInitialised);
++ if (err)
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error switching to initialised state!");
++
++ out:
++ spin_unlock(&pdev->dev_lock);
++
++ if (!err)
++ /* see if pcifront is already configured (if not, we'll wait) */
++ pciback_attach(pdev);
++
++ return err;
++}
++
++static void pciback_be_watch(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ struct pciback_device *pdev =
++ container_of(watch, struct pciback_device, be_watch);
++
++ switch (xenbus_read_driver_state(pdev->xdev->nodename)) {
++ case XenbusStateInitWait:
++ pciback_setup_backend(pdev);
++ break;
++
++ default:
++ break;
++ }
++}
++
++static int pciback_xenbus_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err = 0;
++ struct pciback_device *pdev = alloc_pdev(dev);
++
++ if (pdev == NULL) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(dev, err,
++ "Error allocating pciback_device struct");
++ goto out;
++ }
++
++ /* wait for xend to configure us */
++ err = xenbus_switch_state(dev, XenbusStateInitWait);
++ if (err)
++ goto out;
++
++ /* watch the backend node for backend configuration information */
++ err = xenbus_watch_path(dev, dev->nodename, &pdev->be_watch,
++ pciback_be_watch);
++ if (err)
++ goto out;
++ pdev->be_watching = 1;
++
++ /* We need to force a call to our callback here in case
++ * xend already configured us!
++ */
++ pciback_be_watch(&pdev->be_watch, NULL, 0);
++
++ out:
++ return err;
++}
++
++static int pciback_xenbus_remove(struct xenbus_device *dev)
++{
++ struct pciback_device *pdev = dev->dev.driver_data;
++
++ if (pdev != NULL)
++ free_pdev(pdev);
++
++ return 0;
++}
++
++static struct xenbus_device_id xenpci_ids[] = {
++ {"pci"},
++ {{0}},
++};
++
++static struct xenbus_driver xenbus_pciback_driver = {
++ .name = "pciback",
++ .owner = THIS_MODULE,
++ .ids = xenpci_ids,
++ .probe = pciback_xenbus_probe,
++ .remove = pciback_xenbus_remove,
++ .otherend_changed = pciback_frontend_changed,
++};
++
++int __init pciback_xenbus_register(void)
++{
++ return xenbus_register_backend(&xenbus_pciback_driver);
++}
++
++void __exit pciback_xenbus_unregister(void)
++{
++ xenbus_unregister_driver(&xenbus_pciback_driver);
++}
+diff -urNp linux-2.6/drivers/xen/pcifront/Makefile new/drivers/xen/pcifront/Makefile
+--- linux-2.6/drivers/xen/pcifront/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pcifront/Makefile 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,7 @@
++obj-y += pcifront.o
++
++pcifront-y := pci_op.o xenbus.o pci.o
++
++ifeq ($(CONFIG_XEN_PCIDEV_FE_DEBUG),y)
++EXTRA_CFLAGS += -DDEBUG
++endif
+diff -urNp linux-2.6/drivers/xen/pcifront/pci.c new/drivers/xen/pcifront/pci.c
+--- linux-2.6/drivers/xen/pcifront/pci.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pcifront/pci.c 2006-05-09 12:34:38.000000000 +0200
+@@ -0,0 +1,46 @@
++/*
++ * PCI Frontend Operations - ensure only one PCI frontend runs at a time
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/pci.h>
++#include <linux/spinlock.h>
++#include "pcifront.h"
++
++DEFINE_SPINLOCK(pcifront_dev_lock);
++static struct pcifront_device *pcifront_dev = NULL;
++
++int pcifront_connect(struct pcifront_device *pdev)
++{
++ int err = 0;
++
++ spin_lock(&pcifront_dev_lock);
++
++ if (!pcifront_dev) {
++ dev_info(&pdev->xdev->dev, "Installing PCI frontend\n");
++ pcifront_dev = pdev;
++ }
++ else {
++ dev_err(&pdev->xdev->dev, "PCI frontend already installed!\n");
++ err = -EEXIST;
++ }
++
++ spin_unlock(&pcifront_dev_lock);
++
++ return err;
++}
++
++void pcifront_disconnect(struct pcifront_device *pdev)
++{
++ spin_lock(&pcifront_dev_lock);
++
++ if (pdev == pcifront_dev) {
++ dev_info(&pdev->xdev->dev,
++ "Disconnecting PCI Frontend Buses\n");
++ pcifront_dev = NULL;
++ }
++
++ spin_unlock(&pcifront_dev_lock);
++}
+diff -urNp linux-2.6/drivers/xen/pcifront/pcifront.h new/drivers/xen/pcifront/pcifront.h
+--- linux-2.6/drivers/xen/pcifront/pcifront.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pcifront/pcifront.h 2006-05-09 12:34:39.000000000 +0200
+@@ -0,0 +1,40 @@
++/*
++ * PCI Frontend - Common data structures & function declarations
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#ifndef __XEN_PCIFRONT_H__
++#define __XEN_PCIFRONT_H__
++
++#include <linux/spinlock.h>
++#include <linux/pci.h>
++#include <xen/xenbus.h>
++#include <xen/interface/io/pciif.h>
++#include <xen/pcifront.h>
++
++struct pci_bus_entry {
++ struct list_head list;
++ struct pci_bus *bus;
++};
++
++struct pcifront_device {
++ struct xenbus_device *xdev;
++ struct list_head root_buses;
++ spinlock_t dev_lock;
++
++ int evtchn;
++ int gnt_ref;
++
++ /* Lock this when doing any operations in sh_info */
++ spinlock_t sh_info_lock;
++ struct xen_pci_sharedinfo *sh_info;
++};
++
++int pcifront_connect(struct pcifront_device *pdev);
++void pcifront_disconnect(struct pcifront_device *pdev);
++
++int pcifront_scan_root(struct pcifront_device *pdev,
++ unsigned int domain, unsigned int bus);
++void pcifront_free_roots(struct pcifront_device *pdev);
++
++#endif /* __XEN_PCIFRONT_H__ */
+diff -urNp linux-2.6/drivers/xen/pcifront/pci_op.c new/drivers/xen/pcifront/pci_op.c
+--- linux-2.6/drivers/xen/pcifront/pci_op.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pcifront/pci_op.c 2006-05-09 12:34:39.000000000 +0200
+@@ -0,0 +1,272 @@
++/*
++ * PCI Frontend Operations - Communicates with frontend
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/version.h>
++#include <linux/init.h>
++#include <linux/pci.h>
++#include <linux/spinlock.h>
++#include <linux/time.h>
++#include <xen/evtchn.h>
++#include "pcifront.h"
++
++static int verbose_request = 0;
++module_param(verbose_request, int, 0644);
++
++static int errno_to_pcibios_err(int errno)
++{
++ switch (errno) {
++ case XEN_PCI_ERR_success:
++ return PCIBIOS_SUCCESSFUL;
++
++ case XEN_PCI_ERR_dev_not_found:
++ return PCIBIOS_DEVICE_NOT_FOUND;
++
++ case XEN_PCI_ERR_invalid_offset:
++ case XEN_PCI_ERR_op_failed:
++ return PCIBIOS_BAD_REGISTER_NUMBER;
++
++ case XEN_PCI_ERR_not_implemented:
++ return PCIBIOS_FUNC_NOT_SUPPORTED;
++
++ case XEN_PCI_ERR_access_denied:
++ return PCIBIOS_SET_FAILED;
++ }
++ return errno;
++}
++
++static int do_pci_op(struct pcifront_device *pdev, struct xen_pci_op *op)
++{
++ int err = 0;
++ struct xen_pci_op *active_op = &pdev->sh_info->op;
++ unsigned long irq_flags;
++ evtchn_port_t port = pdev->evtchn;
++ s64 ns, ns_timeout;
++ struct timeval tv;
++
++ spin_lock_irqsave(&pdev->sh_info_lock, irq_flags);
++
++ memcpy(active_op, op, sizeof(struct xen_pci_op));
++
++ /* Go */
++ wmb();
++ set_bit(_XEN_PCIF_active, (unsigned long *)&pdev->sh_info->flags);
++ notify_remote_via_evtchn(port);
++
++ /*
++ * We set a poll timeout of 3 seconds but give up on return after
++ * 2 seconds. It is better to time out too late rather than too early
++ * (in the latter case we end up continually re-executing poll() with a
++ * timeout in the past). 1s difference gives plenty of slack for error.
++ */
++ do_gettimeofday(&tv);
++ ns_timeout = timeval_to_ns(&tv) + 2 * NSEC_PER_SEC;
++
++ clear_evtchn(port);
++
++ while (test_bit(_XEN_PCIF_active,
++ (unsigned long *)&pdev->sh_info->flags)) {
++ if (HYPERVISOR_poll(&port, 1, jiffies + 3*HZ))
++ BUG();
++ clear_evtchn(port);
++ do_gettimeofday(&tv);
++ ns = timeval_to_ns(&tv);
++ if (ns > ns_timeout) {
++ dev_err(&pdev->xdev->dev,
++ "pciback not responding!!!\n");
++ clear_bit(_XEN_PCIF_active,
++ (unsigned long *)&pdev->sh_info->flags);
++ err = XEN_PCI_ERR_dev_not_found;
++ goto out;
++ }
++ }
++
++ memcpy(op, active_op, sizeof(struct xen_pci_op));
++
++ err = op->err;
++ out:
++ spin_unlock_irqrestore(&pdev->sh_info_lock, irq_flags);
++ return err;
++}
++
++/* Access to this function is spinlocked in drivers/pci/access.c */
++static int pcifront_bus_read(struct pci_bus *bus, unsigned int devfn,
++ int where, int size, u32 * val)
++{
++ int err = 0;
++ struct xen_pci_op op = {
++ .cmd = XEN_PCI_OP_conf_read,
++ .domain = pci_domain_nr(bus),
++ .bus = bus->number,
++ .devfn = devfn,
++ .offset = where,
++ .size = size,
++ };
++ struct pcifront_sd *sd = bus->sysdata;
++ struct pcifront_device *pdev = sd->pdev;
++
++ if (verbose_request)
++ dev_info(&pdev->xdev->dev,
++ "read dev=%04x:%02x:%02x.%01x - offset %x size %d\n",
++ pci_domain_nr(bus), bus->number, PCI_SLOT(devfn),
++ PCI_FUNC(devfn), where, size);
++
++ err = do_pci_op(pdev, &op);
++
++ if (likely(!err)) {
++ if (verbose_request)
++ dev_info(&pdev->xdev->dev, "read got back value %x\n",
++ op.value);
++
++ *val = op.value;
++ } else if (err == -ENODEV) {
++ /* No device here, pretend that it just returned 0 */
++ err = 0;
++ *val = 0;
++ }
++
++ return errno_to_pcibios_err(err);
++}
++
++/* Access to this function is spinlocked in drivers/pci/access.c */
++static int pcifront_bus_write(struct pci_bus *bus, unsigned int devfn,
++ int where, int size, u32 val)
++{
++ struct xen_pci_op op = {
++ .cmd = XEN_PCI_OP_conf_write,
++ .domain = pci_domain_nr(bus),
++ .bus = bus->number,
++ .devfn = devfn,
++ .offset = where,
++ .size = size,
++ .value = val,
++ };
++ struct pcifront_sd *sd = bus->sysdata;
++ struct pcifront_device *pdev = sd->pdev;
++
++ if (verbose_request)
++ dev_info(&pdev->xdev->dev,
++ "write dev=%04x:%02x:%02x.%01x - "
++ "offset %x size %d val %x\n",
++ pci_domain_nr(bus), bus->number,
++ PCI_SLOT(devfn), PCI_FUNC(devfn), where, size, val);
++
++ return errno_to_pcibios_err(do_pci_op(pdev, &op));
++}
++
++struct pci_ops pcifront_bus_ops = {
++ .read = pcifront_bus_read,
++ .write = pcifront_bus_write,
++};
++
++/* Claim resources for the PCI frontend as-is, backend won't allow changes */
++static void pcifront_claim_resource(struct pci_dev *dev, void *data)
++{
++ struct pcifront_device *pdev = data;
++ int i;
++ struct resource *r;
++
++ for (i = 0; i < PCI_NUM_RESOURCES; i++) {
++ r = &dev->resource[i];
++
++ if (!r->parent && r->start && r->flags) {
++ dev_dbg(&pdev->xdev->dev, "claiming resource %s/%d\n",
++ pci_name(dev), i);
++ pci_claim_resource(dev, i);
++ }
++ }
++}
++
++int pcifront_scan_root(struct pcifront_device *pdev,
++ unsigned int domain, unsigned int bus)
++{
++ struct pci_bus *b;
++ struct pcifront_sd *sd = NULL;
++ struct pci_bus_entry *bus_entry = NULL;
++ int err = 0;
++
++#ifndef CONFIG_PCI_DOMAINS
++ if (domain != 0) {
++ dev_err(&pdev->xdev->dev,
++ "PCI Root in non-zero PCI Domain! domain=%d\n", domain);
++ dev_err(&pdev->xdev->dev,
++ "Please compile with CONFIG_PCI_DOMAINS\n");
++ err = -EINVAL;
++ goto err_out;
++ }
++#endif
++
++ dev_info(&pdev->xdev->dev, "Creating PCI Frontend Bus %04x:%02x\n",
++ domain, bus);
++
++ bus_entry = kmalloc(sizeof(*bus_entry), GFP_KERNEL);
++ sd = kmalloc(sizeof(*sd), GFP_KERNEL);
++ if (!bus_entry || !sd) {
++ err = -ENOMEM;
++ goto err_out;
++ }
++ sd->domain = domain;
++ sd->pdev = pdev;
++
++ b = pci_scan_bus_parented(&pdev->xdev->dev, bus, &pcifront_bus_ops, sd);
++ if (!b) {
++ dev_err(&pdev->xdev->dev, "Error creating PCI Frontend Bus!\n");
++ err = -ENOMEM;
++ goto err_out;
++ }
++ bus_entry->bus = b;
++
++ list_add(&bus_entry->list, &pdev->root_buses);
++
++ /* Claim resources before going "live" with our devices */
++ pci_walk_bus(b, pcifront_claim_resource, pdev);
++
++ pci_bus_add_devices(b);
++
++ return 0;
++
++ err_out:
++ kfree(bus_entry);
++ kfree(sd);
++
++ return err;
++}
++
++static void free_root_bus_devs(struct pci_bus *bus)
++{
++ struct pci_dev *dev;
++
++ spin_lock(&pci_bus_lock);
++ while (!list_empty(&bus->devices)) {
++ dev = container_of(bus->devices.next, struct pci_dev, bus_list);
++ spin_unlock(&pci_bus_lock);
++
++ dev_dbg(&dev->dev, "removing device\n");
++ pci_remove_bus_device(dev);
++
++ spin_lock(&pci_bus_lock);
++ }
++ spin_unlock(&pci_bus_lock);
++}
++
++void pcifront_free_roots(struct pcifront_device *pdev)
++{
++ struct pci_bus_entry *bus_entry, *t;
++
++ dev_dbg(&pdev->xdev->dev, "cleaning up root buses\n");
++
++ list_for_each_entry_safe(bus_entry, t, &pdev->root_buses, list) {
++ list_del(&bus_entry->list);
++
++ free_root_bus_devs(bus_entry->bus);
++
++ kfree(bus_entry->bus->sysdata);
++
++ device_unregister(bus_entry->bus->bridge);
++ pci_remove_bus(bus_entry->bus);
++
++ kfree(bus_entry);
++ }
++}
+diff -urNp linux-2.6/drivers/xen/pcifront/xenbus.c new/drivers/xen/pcifront/xenbus.c
+--- linux-2.6/drivers/xen/pcifront/xenbus.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/pcifront/xenbus.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,294 @@
++/*
++ * PCI Frontend Xenbus Setup - handles setup with backend (imports page/evtchn)
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#include <linux/module.h>
++#include <linux/init.h>
++#include <linux/mm.h>
++#include <xen/xenbus.h>
++#include "pcifront.h"
++
++#define INVALID_GRANT_REF (0)
++#define INVALID_EVTCHN (-1)
++
++static struct pcifront_device *alloc_pdev(struct xenbus_device *xdev)
++{
++ struct pcifront_device *pdev;
++
++ pdev = kmalloc(sizeof(struct pcifront_device), GFP_KERNEL);
++ if (pdev == NULL)
++ goto out;
++
++ pdev->sh_info =
++ (struct xen_pci_sharedinfo *)__get_free_page(GFP_KERNEL);
++ if (pdev->sh_info == NULL) {
++ kfree(pdev);
++ pdev = NULL;
++ goto out;
++ }
++ pdev->sh_info->flags = 0;
++
++ xdev->dev.driver_data = pdev;
++ pdev->xdev = xdev;
++
++ INIT_LIST_HEAD(&pdev->root_buses);
++
++ spin_lock_init(&pdev->dev_lock);
++ spin_lock_init(&pdev->sh_info_lock);
++
++ pdev->evtchn = INVALID_EVTCHN;
++ pdev->gnt_ref = INVALID_GRANT_REF;
++
++ dev_dbg(&xdev->dev, "Allocated pdev @ 0x%p pdev->sh_info @ 0x%p\n",
++ pdev, pdev->sh_info);
++ out:
++ return pdev;
++}
++
++static void free_pdev(struct pcifront_device *pdev)
++{
++ dev_dbg(&pdev->xdev->dev, "freeing pdev @ 0x%p\n", pdev);
++
++ pcifront_free_roots(pdev);
++
++ if (pdev->evtchn != INVALID_EVTCHN)
++ xenbus_free_evtchn(pdev->xdev, pdev->evtchn);
++
++ if (pdev->gnt_ref != INVALID_GRANT_REF)
++ gnttab_end_foreign_access(pdev->gnt_ref, 0,
++ (unsigned long)pdev->sh_info);
++
++ pdev->xdev->dev.driver_data = NULL;
++
++ kfree(pdev);
++}
++
++static int pcifront_publish_info(struct pcifront_device *pdev)
++{
++ int err = 0;
++ struct xenbus_transaction trans;
++
++ err = xenbus_grant_ring(pdev->xdev, virt_to_mfn(pdev->sh_info));
++ if (err < 0)
++ goto out;
++
++ pdev->gnt_ref = err;
++
++ err = xenbus_alloc_evtchn(pdev->xdev, &pdev->evtchn);
++ if (err)
++ goto out;
++
++ do_publish:
++ err = xenbus_transaction_start(&trans);
++ if (err) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error writing configuration for backend "
++ "(start transaction)");
++ goto out;
++ }
++
++ err = xenbus_printf(trans, pdev->xdev->nodename,
++ "pci-op-ref", "%u", pdev->gnt_ref);
++ if (!err)
++ err = xenbus_printf(trans, pdev->xdev->nodename,
++ "event-channel", "%u", pdev->evtchn);
++ if (!err)
++ err = xenbus_printf(trans, pdev->xdev->nodename,
++ "magic", XEN_PCI_MAGIC);
++
++ if (err) {
++ xenbus_transaction_end(trans, 1);
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error writing configuration for backend");
++ goto out;
++ } else {
++ err = xenbus_transaction_end(trans, 0);
++ if (err == -EAGAIN)
++ goto do_publish;
++ else if (err) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error completing transaction "
++ "for backend");
++ goto out;
++ }
++ }
++
++ xenbus_switch_state(pdev->xdev, XenbusStateInitialised);
++
++ dev_dbg(&pdev->xdev->dev, "publishing successful!\n");
++
++ out:
++ return err;
++}
++
++static int pcifront_try_connect(struct pcifront_device *pdev)
++{
++ int err = -EFAULT;
++ int i, num_roots, len;
++ char str[64];
++ unsigned int domain, bus;
++
++ spin_lock(&pdev->dev_lock);
++
++ /* Only connect once */
++ if (xenbus_read_driver_state(pdev->xdev->nodename) !=
++ XenbusStateInitialised)
++ goto out;
++
++ err = pcifront_connect(pdev);
++ if (err) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error connecting PCI Frontend");
++ goto out;
++ }
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend,
++ "root_num", "%d", &num_roots);
++ if (err == -ENOENT) {
++ xenbus_dev_error(pdev->xdev, err,
++ "No PCI Roots found, trying 0000:00");
++ err = pcifront_scan_root(pdev, 0, 0);
++ num_roots = 0;
++ } else if (err != 1) {
++ if (err == 0)
++ err = -EINVAL;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error reading number of PCI roots");
++ goto out;
++ }
++
++ for (i = 0; i < num_roots; i++) {
++ len = snprintf(str, sizeof(str), "root-%d", i);
++ if (unlikely(len >= (sizeof(str) - 1))) {
++ err = -ENOMEM;
++ goto out;
++ }
++
++ err = xenbus_scanf(XBT_NIL, pdev->xdev->otherend, str,
++ "%x:%x", &domain, &bus);
++ if (err != 2) {
++ if (err >= 0)
++ err = -EINVAL;
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error reading PCI root %d", i);
++ goto out;
++ }
++
++ err = pcifront_scan_root(pdev, domain, bus);
++ if (err) {
++ xenbus_dev_fatal(pdev->xdev, err,
++ "Error scanning PCI root %04x:%02x",
++ domain, bus);
++ goto out;
++ }
++ }
++
++ err = xenbus_switch_state(pdev->xdev, XenbusStateConnected);
++ if (err)
++ goto out;
++
++ out:
++ spin_unlock(&pdev->dev_lock);
++ return err;
++}
++
++static int pcifront_try_disconnect(struct pcifront_device *pdev)
++{
++ int err = 0;
++ enum xenbus_state prev_state;
++
++ spin_lock(&pdev->dev_lock);
++
++ prev_state = xenbus_read_driver_state(pdev->xdev->nodename);
++
++ if (prev_state < XenbusStateClosing)
++ err = xenbus_switch_state(pdev->xdev, XenbusStateClosing);
++
++ if (!err && prev_state == XenbusStateConnected)
++ pcifront_disconnect(pdev);
++
++ spin_unlock(&pdev->dev_lock);
++
++ return err;
++}
++
++static void pcifront_backend_changed(struct xenbus_device *xdev,
++ enum xenbus_state be_state)
++{
++ struct pcifront_device *pdev = xdev->dev.driver_data;
++
++ switch (be_state) {
++ case XenbusStateClosing:
++ dev_warn(&xdev->dev, "backend going away!\n");
++ pcifront_try_disconnect(pdev);
++ break;
++
++ case XenbusStateClosed:
++ dev_warn(&xdev->dev, "backend went away!\n");
++ pcifront_try_disconnect(pdev);
++
++ device_unregister(&pdev->xdev->dev);
++ break;
++
++ case XenbusStateConnected:
++ pcifront_try_connect(pdev);
++ break;
++
++ default:
++ break;
++ }
++}
++
++static int pcifront_xenbus_probe(struct xenbus_device *xdev,
++ const struct xenbus_device_id *id)
++{
++ int err = 0;
++ struct pcifront_device *pdev = alloc_pdev(xdev);
++
++ if (pdev == NULL) {
++ err = -ENOMEM;
++ xenbus_dev_fatal(xdev, err,
++ "Error allocating pcifront_device struct");
++ goto out;
++ }
++
++ err = pcifront_publish_info(pdev);
++
++ out:
++ return err;
++}
++
++static int pcifront_xenbus_remove(struct xenbus_device *xdev)
++{
++ if (xdev->dev.driver_data)
++ free_pdev(xdev->dev.driver_data);
++
++ return 0;
++}
++
++static struct xenbus_device_id xenpci_ids[] = {
++ {"pci"},
++ {{0}},
++};
++
++static struct xenbus_driver xenbus_pcifront_driver = {
++ .name = "pcifront",
++ .owner = THIS_MODULE,
++ .ids = xenpci_ids,
++ .probe = pcifront_xenbus_probe,
++ .remove = pcifront_xenbus_remove,
++ .otherend_changed = pcifront_backend_changed,
++};
++
++static int __init pcifront_init(void)
++{
++ int err = 0;
++
++ err = xenbus_register_frontend(&xenbus_pcifront_driver);
++
++ return err;
++}
++
++/* Initialize after the Xen PCI Frontend Stub is initialized */
++subsys_initcall(pcifront_init);
+diff -urNp linux-2.6/drivers/xen/privcmd/Makefile new/drivers/xen/privcmd/Makefile
+--- linux-2.6/drivers/xen/privcmd/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/privcmd/Makefile 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,2 @@
++
++obj-$(CONFIG_XEN_PRIVCMD) := privcmd.o
+diff -urNp linux-2.6/drivers/xen/privcmd/privcmd.c new/drivers/xen/privcmd/privcmd.c
+--- linux-2.6/drivers/xen/privcmd/privcmd.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/privcmd/privcmd.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,285 @@
++/******************************************************************************
++ * privcmd.c
++ *
++ * Interface to privileged domain-0 commands.
++ *
++ * Copyright (c) 2002-2004, K A Fraser, B Dragovic
++ */
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/sched.h>
++#include <linux/slab.h>
++#include <linux/string.h>
++#include <linux/errno.h>
++#include <linux/mm.h>
++#include <linux/mman.h>
++#include <linux/swap.h>
++#include <linux/smp_lock.h>
++#include <linux/highmem.h>
++#include <linux/pagemap.h>
++#include <linux/seq_file.h>
++#include <linux/kthread.h>
++#include <asm/hypervisor.h>
++
++#include <asm/pgalloc.h>
++#include <asm/pgtable.h>
++#include <asm/uaccess.h>
++#include <asm/tlb.h>
++#include <asm/hypervisor.h>
++#include <xen/public/privcmd.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/dom0_ops.h>
++#include <xen/xen_proc.h>
++
++static struct proc_dir_entry *privcmd_intf;
++static struct proc_dir_entry *capabilities_intf;
++
++#define NR_HYPERCALLS 64
++static DECLARE_BITMAP(hypercall_permission_map, NR_HYPERCALLS);
++
++static int privcmd_ioctl(struct inode *inode, struct file *file,
++ unsigned int cmd, unsigned long data)
++{
++ int ret = -ENOSYS;
++ void __user *udata = (void __user *) data;
++
++ switch (cmd) {
++ case IOCTL_PRIVCMD_HYPERCALL: {
++ privcmd_hypercall_t hypercall;
++
++ if (copy_from_user(&hypercall, udata, sizeof(hypercall)))
++ return -EFAULT;
++
++ /* Check hypercall number for validity. */
++ if (hypercall.op >= NR_HYPERCALLS)
++ return -EINVAL;
++ if (!test_bit(hypercall.op, hypercall_permission_map))
++ return -EINVAL;
++
++#if defined(__i386__)
++ __asm__ __volatile__ (
++ "pushl %%ebx; pushl %%ecx; pushl %%edx; "
++ "pushl %%esi; pushl %%edi; "
++ "movl 8(%%eax),%%ebx ;"
++ "movl 16(%%eax),%%ecx ;"
++ "movl 24(%%eax),%%edx ;"
++ "movl 32(%%eax),%%esi ;"
++ "movl 40(%%eax),%%edi ;"
++ "movl (%%eax),%%eax ;"
++ "shll $5,%%eax ;"
++ "addl $hypercall_page,%%eax ;"
++ "call *%%eax ;"
++ "popl %%edi; popl %%esi; popl %%edx; "
++ "popl %%ecx; popl %%ebx"
++ : "=a" (ret) : "0" (&hypercall) : "memory" );
++#elif defined (__x86_64__)
++ {
++ long ign1, ign2, ign3;
++ __asm__ __volatile__ (
++ "movq %8,%%r10; movq %9,%%r8;"
++ "shlq $5,%%rax ;"
++ "addq $hypercall_page,%%rax ;"
++ "call *%%rax"
++ : "=a" (ret), "=D" (ign1),
++ "=S" (ign2), "=d" (ign3)
++ : "0" ((unsigned long)hypercall.op),
++ "1" ((unsigned long)hypercall.arg[0]),
++ "2" ((unsigned long)hypercall.arg[1]),
++ "3" ((unsigned long)hypercall.arg[2]),
++ "g" ((unsigned long)hypercall.arg[3]),
++ "g" ((unsigned long)hypercall.arg[4])
++ : "r8", "r10", "memory" );
++ }
++#elif defined (__ia64__)
++ __asm__ __volatile__ (
++ ";; mov r14=%2; mov r15=%3; "
++ "mov r16=%4; mov r17=%5; mov r18=%6;"
++ "mov r2=%1; break 0x1000;; mov %0=r8 ;;"
++ : "=r" (ret)
++ : "r" (hypercall.op),
++ "r" (hypercall.arg[0]),
++ "r" (hypercall.arg[1]),
++ "r" (hypercall.arg[2]),
++ "r" (hypercall.arg[3]),
++ "r" (hypercall.arg[4])
++ : "r14","r15","r16","r17","r18","r2","r8","memory");
++#endif
++ }
++ break;
++
++#if defined(CONFIG_XEN_PRIVILEGED_GUEST)
++ case IOCTL_PRIVCMD_MMAP: {
++#define PRIVCMD_MMAP_SZ 32
++ privcmd_mmap_t mmapcmd;
++ privcmd_mmap_entry_t msg[PRIVCMD_MMAP_SZ];
++ privcmd_mmap_entry_t __user *p;
++ int i, rc;
++
++ if (copy_from_user(&mmapcmd, udata, sizeof(mmapcmd)))
++ return -EFAULT;
++
++ p = mmapcmd.entry;
++
++ for (i = 0; i < mmapcmd.num;
++ i += PRIVCMD_MMAP_SZ, p += PRIVCMD_MMAP_SZ) {
++ int j, n = ((mmapcmd.num-i)>PRIVCMD_MMAP_SZ)?
++ PRIVCMD_MMAP_SZ:(mmapcmd.num-i);
++
++ if (copy_from_user(&msg, p,
++ n*sizeof(privcmd_mmap_entry_t)))
++ return -EFAULT;
++
++ for (j = 0; j < n; j++) {
++ struct vm_area_struct *vma =
++ find_vma( current->mm, msg[j].va );
++
++ if (!vma)
++ return -EINVAL;
++
++ if (msg[j].va > PAGE_OFFSET)
++ return -EINVAL;
++
++ if ((msg[j].va + (msg[j].npages << PAGE_SHIFT))
++ > vma->vm_end )
++ return -EINVAL;
++
++ if ((rc = direct_remap_pfn_range(
++ vma,
++ msg[j].va&PAGE_MASK,
++ msg[j].mfn,
++ msg[j].npages<<PAGE_SHIFT,
++ vma->vm_page_prot,
++ mmapcmd.dom)) < 0)
++ return rc;
++ }
++ }
++ ret = 0;
++ }
++ break;
++
++ case IOCTL_PRIVCMD_MMAPBATCH: {
++ privcmd_mmapbatch_t m;
++ struct vm_area_struct *vma = NULL;
++ xen_pfn_t __user *p;
++ unsigned long addr, mfn;
++ int i;
++
++ if (copy_from_user(&m, udata, sizeof(m))) {
++ ret = -EFAULT;
++ goto batch_err;
++ }
++
++ if (m.dom == DOMID_SELF) {
++ ret = -EINVAL;
++ goto batch_err;
++ }
++
++ vma = find_vma(current->mm, m.addr);
++ if (!vma) {
++ ret = -EINVAL;
++ goto batch_err;
++ }
++
++ if (m.addr > PAGE_OFFSET) {
++ ret = -EFAULT;
++ goto batch_err;
++ }
++
++ if ((m.addr + (m.num<<PAGE_SHIFT)) > vma->vm_end) {
++ ret = -EFAULT;
++ goto batch_err;
++ }
++
++ p = m.arr;
++ addr = m.addr;
++ for (i = 0; i < m.num; i++, addr += PAGE_SIZE, p++) {
++ if (get_user(mfn, p))
++ return -EFAULT;
++
++ ret = direct_remap_pfn_range(vma, addr & PAGE_MASK,
++ mfn, PAGE_SIZE,
++ vma->vm_page_prot, m.dom);
++ if (ret < 0)
++ put_user(0xF0000000 | mfn, p);
++ }
++
++ ret = 0;
++ break;
++
++ batch_err:
++ printk("batch_err ret=%d vma=%p addr=%lx "
++ "num=%d arr=%p %lx-%lx\n",
++ ret, vma, (unsigned long)m.addr, m.num, m.arr,
++ vma ? vma->vm_start : 0, vma ? vma->vm_end : 0);
++ break;
++ }
++ break;
++#endif
++
++ default:
++ ret = -EINVAL;
++ break;
++ }
++
++ return ret;
++}
++
++#ifndef HAVE_ARCH_PRIVCMD_MMAP
++static int privcmd_mmap(struct file * file, struct vm_area_struct * vma)
++{
++ /* DONTCOPY is essential for Xen as copy_page_range is broken. */
++ vma->vm_flags |= VM_RESERVED | VM_IO | VM_DONTCOPY;
++
++ return 0;
++}
++#endif
++
++static struct file_operations privcmd_file_ops = {
++ .ioctl = privcmd_ioctl,
++ .mmap = privcmd_mmap,
++};
++
++static int capabilities_read(char *page, char **start, off_t off,
++ int count, int *eof, void *data)
++{
++ int len = 0;
++ *page = 0;
++
++ if (xen_start_info->flags & SIF_INITDOMAIN)
++ len = sprintf( page, "control_d\n" );
++
++ *eof = 1;
++ return len;
++}
++
++static int __init privcmd_init(void)
++{
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ /* Set of hypercalls that privileged applications may execute. */
++ set_bit(__HYPERVISOR_acm_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_dom0_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_event_channel_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_memory_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_mmu_update, hypercall_permission_map);
++ set_bit(__HYPERVISOR_mmuext_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_xen_version, hypercall_permission_map);
++ set_bit(__HYPERVISOR_sched_op, hypercall_permission_map);
++ set_bit(__HYPERVISOR_sched_op_compat, hypercall_permission_map);
++ set_bit(__HYPERVISOR_event_channel_op_compat,
++ hypercall_permission_map);
++
++ privcmd_intf = create_xen_proc_entry("privcmd", 0400);
++ if (privcmd_intf != NULL)
++ privcmd_intf->proc_fops = &privcmd_file_ops;
++
++ capabilities_intf = create_xen_proc_entry("capabilities", 0400 );
++ if (capabilities_intf != NULL)
++ capabilities_intf->read_proc = capabilities_read;
++
++ return 0;
++}
++
++__initcall(privcmd_init);
+diff -urNp linux-2.6/drivers/xen/tpmback/common.h new/drivers/xen/tpmback/common.h
+--- linux-2.6/drivers/xen/tpmback/common.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/tpmback/common.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,85 @@
++/******************************************************************************
++ * drivers/xen/tpmback/common.h
++ */
++
++#ifndef __NETIF__BACKEND__COMMON_H__
++#define __NETIF__BACKEND__COMMON_H__
++
++#include <linux/config.h>
++#include <linux/version.h>
++#include <linux/module.h>
++#include <linux/interrupt.h>
++#include <linux/slab.h>
++#include <xen/evtchn.h>
++#include <xen/driver_util.h>
++#include <xen/interface/grant_table.h>
++#include <xen/interface/io/tpmif.h>
++#include <asm/io.h>
++#include <asm/pgalloc.h>
++
++#define DPRINTK(_f, _a...) \
++ pr_debug("(file=%s, line=%d) " _f, \
++ __FILE__ , __LINE__ , ## _a )
++
++struct backend_info;
++
++typedef struct tpmif_st {
++ struct list_head tpmif_list;
++ /* Unique identifier for this interface. */
++ domid_t domid;
++ unsigned int handle;
++
++ /* Physical parameters of the comms window. */
++ unsigned int evtchn;
++ unsigned int irq;
++
++ /* The shared rings and indexes. */
++ tpmif_tx_interface_t *tx;
++ struct vm_struct *tx_area;
++
++ /* Miscellaneous private stuff. */
++ enum { DISCONNECTED, DISCONNECTING, CONNECTED } status;
++ int active;
++
++ struct tpmif_st *hash_next;
++ struct list_head list; /* scheduling list */
++ atomic_t refcnt;
++
++ struct backend_info *bi;
++ unsigned long mmap_vstart;
++
++ grant_handle_t shmem_handle;
++ grant_ref_t shmem_ref;
++ struct page *pagerange;
++
++ char devname[20];
++} tpmif_t;
++
++void tpmif_disconnect_complete(tpmif_t * tpmif);
++tpmif_t *tpmif_find(domid_t domid, struct backend_info *bi);
++void tpmif_interface_init(void);
++void tpmif_interface_exit(void);
++void tpmif_schedule_work(tpmif_t * tpmif);
++void tpmif_deschedule_work(tpmif_t * tpmif);
++void tpmif_xenbus_init(void);
++void tpmif_xenbus_exit(void);
++int tpmif_map(tpmif_t *tpmif, unsigned long shared_page, unsigned int evtchn);
++irqreturn_t tpmif_be_int(int irq, void *dev_id, struct pt_regs *regs);
++
++long int tpmback_get_instance(struct backend_info *bi);
++
++int vtpm_release_packets(tpmif_t * tpmif, int send_msgs);
++
++
++#define tpmif_get(_b) (atomic_inc(&(_b)->refcnt))
++#define tpmif_put(_b) \
++ do { \
++ if (atomic_dec_and_test(&(_b)->refcnt)) \
++ tpmif_disconnect_complete(_b); \
++ } while (0)
++
++extern int num_frontends;
++
++#define MMAP_VADDR(t,_req) ((t)->mmap_vstart + ((_req) * PAGE_SIZE))
++
++#endif /* __TPMIF__BACKEND__COMMON_H__ */
+diff -urNp linux-2.6/drivers/xen/tpmback/interface.c new/drivers/xen/tpmback/interface.c
+--- linux-2.6/drivers/xen/tpmback/interface.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/tpmback/interface.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,177 @@
++ /*****************************************************************************
++ * drivers/xen/tpmback/interface.c
++ *
++ * Vritual TPM interface management.
++ *
++ * Copyright (c) 2005, IBM Corporation
++ *
++ * Author: Stefan Berger, stefanb@us.ibm.com
++ *
++ * This code has been derived from drivers/xen/netback/interface.c
++ * Copyright (c) 2004, Keir Fraser
++ */
++
++#include "common.h"
++#include <xen/balloon.h>
++#include <xen/gnttab.h>
++
++static kmem_cache_t *tpmif_cachep;
++int num_frontends = 0;
++
++LIST_HEAD(tpmif_list);
++
++static tpmif_t *alloc_tpmif(domid_t domid, struct backend_info *bi)
++{
++ tpmif_t *tpmif;
++
++ tpmif = kmem_cache_alloc(tpmif_cachep, GFP_KERNEL);
++ if (!tpmif)
++ return ERR_PTR(-ENOMEM);
++
++ memset(tpmif, 0, sizeof (*tpmif));
++ tpmif->domid = domid;
++ tpmif->status = DISCONNECTED;
++ tpmif->bi = bi;
++ snprintf(tpmif->devname, sizeof(tpmif->devname), "tpmif%d", domid);
++ atomic_set(&tpmif->refcnt, 1);
++
++ tpmif->pagerange = balloon_alloc_empty_page_range(TPMIF_TX_RING_SIZE);
++ BUG_ON(tpmif->pagerange == NULL);
++ tpmif->mmap_vstart = (unsigned long)pfn_to_kaddr(
++ page_to_pfn(tpmif->pagerange));
++
++ list_add(&tpmif->tpmif_list, &tpmif_list);
++ num_frontends++;
++
++ return tpmif;
++}
++
++static void free_tpmif(tpmif_t * tpmif)
++{
++ num_frontends--;
++ list_del(&tpmif->tpmif_list);
++ balloon_dealloc_empty_page_range(tpmif->pagerange, TPMIF_TX_RING_SIZE);
++ kmem_cache_free(tpmif_cachep, tpmif);
++}
++
++tpmif_t *tpmif_find(domid_t domid, struct backend_info *bi)
++{
++ tpmif_t *tpmif;
++
++ list_for_each_entry(tpmif, &tpmif_list, tpmif_list) {
++ if (tpmif->bi == bi) {
++ if (tpmif->domid == domid) {
++ tpmif_get(tpmif);
++ return tpmif;
++ } else {
++ return ERR_PTR(-EEXIST);
++ }
++ }
++ }
++
++ return alloc_tpmif(domid, bi);
++}
++
++static int map_frontend_page(tpmif_t *tpmif, unsigned long shared_page)
++{
++ int ret;
++ struct gnttab_map_grant_ref op;
++
++ gnttab_set_map_op(&op, (unsigned long)tpmif->tx_area->addr,
++ GNTMAP_host_map, shared_page, tpmif->domid);
++
++ lock_vm_area(tpmif->tx_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1);
++ unlock_vm_area(tpmif->tx_area);
++ BUG_ON(ret);
++
++ if (op.status) {
++ DPRINTK(" Grant table operation failure !\n");
++ return op.status;
++ }
++
++ tpmif->shmem_ref = shared_page;
++ tpmif->shmem_handle = op.handle;
++
++ return 0;
++}
++
++static void unmap_frontend_page(tpmif_t *tpmif)
++{
++ struct gnttab_unmap_grant_ref op;
++ int ret;
++
++ gnttab_set_unmap_op(&op, (unsigned long)tpmif->tx_area->addr,
++ GNTMAP_host_map, tpmif->shmem_handle);
++
++ lock_vm_area(tpmif->tx_area);
++ ret = HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1);
++ unlock_vm_area(tpmif->tx_area);
++ BUG_ON(ret);
++}
++
++int tpmif_map(tpmif_t *tpmif, unsigned long shared_page, unsigned int evtchn)
++{
++ int err;
++ struct evtchn_bind_interdomain bind_interdomain;
++
++ if (tpmif->irq) {
++ return 0;
++ }
++
++ if ((tpmif->tx_area = alloc_vm_area(PAGE_SIZE)) == NULL)
++ return -ENOMEM;
++
++ err = map_frontend_page(tpmif, shared_page);
++ if (err) {
++ free_vm_area(tpmif->tx_area);
++ return err;
++ }
++
++
++ bind_interdomain.remote_dom = tpmif->domid;
++ bind_interdomain.remote_port = evtchn;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
++ &bind_interdomain);
++ if (err) {
++ unmap_frontend_page(tpmif);
++ free_vm_area(tpmif->tx_area);
++ return err;
++ }
++
++ tpmif->evtchn = bind_interdomain.local_port;
++
++ tpmif->tx = (tpmif_tx_interface_t *)tpmif->tx_area->addr;
++
++ tpmif->irq = bind_evtchn_to_irqhandler(
++ tpmif->evtchn, tpmif_be_int, 0, tpmif->devname, tpmif);
++ tpmif->shmem_ref = shared_page;
++ tpmif->active = 1;
++
++ return 0;
++}
++
++void tpmif_disconnect_complete(tpmif_t *tpmif)
++{
++ if (tpmif->irq)
++ unbind_from_irqhandler(tpmif->irq, tpmif);
++
++ if (tpmif->tx) {
++ unmap_frontend_page(tpmif);
++ free_vm_area(tpmif->tx_area);
++ }
++
++ free_tpmif(tpmif);
++}
++
++void __init tpmif_interface_init(void)
++{
++ tpmif_cachep = kmem_cache_create("tpmif_cache", sizeof (tpmif_t),
++ 0, 0, NULL, NULL);
++}
++
++void __exit tpmif_interface_exit(void)
++{
++ kmem_cache_destroy(tpmif_cachep);
++}
+diff -urNp linux-2.6/drivers/xen/tpmback/Makefile new/drivers/xen/tpmback/Makefile
+--- linux-2.6/drivers/xen/tpmback/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/tpmback/Makefile 2006-05-09 12:34:39.000000000 +0200
+@@ -0,0 +1,4 @@
++
++obj-$(CONFIG_XEN_TPMDEV_BACKEND) += tpmbk.o
++
++tpmbk-y += tpmback.o interface.o xenbus.o
+diff -urNp linux-2.6/drivers/xen/tpmback/tpmback.c new/drivers/xen/tpmback/tpmback.c
+--- linux-2.6/drivers/xen/tpmback/tpmback.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/tpmback/tpmback.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,983 @@
++/******************************************************************************
++ * drivers/xen/tpmback/tpmback.c
++ *
++ * Copyright (c) 2005, IBM Corporation
++ *
++ * Author: Stefan Berger, stefanb@us.ibm.com
++ * Grant table support: Mahadevan Gomathisankaran
++ *
++ * This code has been derived from drivers/xen/netback/netback.c
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ */
++
++#include "common.h"
++#include <xen/evtchn.h>
++
++#include <linux/types.h>
++#include <linux/list.h>
++#include <linux/miscdevice.h>
++#include <linux/poll.h>
++#include <asm/uaccess.h>
++#include <xen/xenbus.h>
++#include <xen/interface/grant_table.h>
++#include <xen/gnttab.h>
++
++/* local data structures */
++struct data_exchange {
++ struct list_head pending_pak;
++ struct list_head current_pak;
++ unsigned int copied_so_far;
++ u8 has_opener:1;
++ u8 aborted:1;
++ rwlock_t pak_lock; // protects all of the previous fields
++ wait_queue_head_t wait_queue;
++};
++
++struct vtpm_resp_hdr {
++ uint32_t instance_no;
++ uint16_t tag_no;
++ uint32_t len_no;
++ uint32_t ordinal_no;
++} __attribute__ ((packed));
++
++struct packet {
++ struct list_head next;
++ unsigned int data_len;
++ u8 *data_buffer;
++ tpmif_t *tpmif;
++ u32 tpm_instance;
++ u8 req_tag;
++ u32 last_read;
++ u8 flags;
++ struct timer_list processing_timer;
++};
++
++enum {
++ PACKET_FLAG_DISCARD_RESPONSE = 1,
++ PACKET_FLAG_CHECK_RESPONSESTATUS = 2,
++};
++
++/* local variables */
++static struct data_exchange dataex;
++
++/* local function prototypes */
++static int _packet_write(struct packet *pak,
++ const char *data, size_t size, int userbuffer);
++static void processing_timeout(unsigned long ptr);
++static int packet_read_shmem(struct packet *pak,
++ tpmif_t * tpmif,
++ u32 offset,
++ char *buffer, int isuserbuffer, u32 left);
++static int vtpm_queue_packet(struct packet *pak);
++
++/***************************************************************
++ Buffer copying fo user and kernel space buffes.
++***************************************************************/
++static inline int copy_from_buffer(void *to,
++ const void *from, unsigned long size,
++ int isuserbuffer)
++{
++ if (isuserbuffer) {
++ if (copy_from_user(to, (void __user *)from, size))
++ return -EFAULT;
++ } else {
++ memcpy(to, from, size);
++ }
++ return 0;
++}
++
++static inline int copy_to_buffer(void *to,
++ const void *from, unsigned long size,
++ int isuserbuffer)
++{
++ if (isuserbuffer) {
++ if (copy_to_user((void __user *)to, from, size))
++ return -EFAULT;
++ } else {
++ memcpy(to, from, size);
++ }
++ return 0;
++}
++
++
++static void dataex_init(struct data_exchange *dataex)
++{
++ INIT_LIST_HEAD(&dataex->pending_pak);
++ INIT_LIST_HEAD(&dataex->current_pak);
++ dataex->has_opener = 0;
++ rwlock_init(&dataex->pak_lock);
++ init_waitqueue_head(&dataex->wait_queue);
++}
++
++/***************************************************************
++ Packet-related functions
++***************************************************************/
++
++static struct packet *packet_find_instance(struct list_head *head,
++ u32 tpm_instance)
++{
++ struct packet *pak;
++ struct list_head *p;
++
++ /*
++ * traverse the list of packets and return the first
++ * one with the given instance number
++ */
++ list_for_each(p, head) {
++ pak = list_entry(p, struct packet, next);
++
++ if (pak->tpm_instance == tpm_instance) {
++ return pak;
++ }
++ }
++ return NULL;
++}
++
++static struct packet *packet_find_packet(struct list_head *head, void *packet)
++{
++ struct packet *pak;
++ struct list_head *p;
++
++ /*
++ * traverse the list of packets and return the first
++ * one with the given instance number
++ */
++ list_for_each(p, head) {
++ pak = list_entry(p, struct packet, next);
++
++ if (pak == packet) {
++ return pak;
++ }
++ }
++ return NULL;
++}
++
++static struct packet *packet_alloc(tpmif_t * tpmif,
++ u32 size, u8 req_tag, u8 flags)
++{
++ struct packet *pak = NULL;
++ pak = kzalloc(sizeof (struct packet), GFP_ATOMIC);
++ if (NULL != pak) {
++ if (tpmif) {
++ pak->tpmif = tpmif;
++ pak->tpm_instance = tpmback_get_instance(tpmif->bi);
++ tpmif_get(tpmif);
++ }
++ pak->data_len = size;
++ pak->req_tag = req_tag;
++ pak->last_read = 0;
++ pak->flags = flags;
++
++ /*
++ * cannot do tpmif_get(tpmif); bad things happen
++ * on the last tpmif_put()
++ */
++ init_timer(&pak->processing_timer);
++ pak->processing_timer.function = processing_timeout;
++ pak->processing_timer.data = (unsigned long)pak;
++ }
++ return pak;
++}
++
++static void inline packet_reset(struct packet *pak)
++{
++ pak->last_read = 0;
++}
++
++static void packet_free(struct packet *pak)
++{
++ if (timer_pending(&pak->processing_timer)) {
++ BUG();
++ }
++
++ if (pak->tpmif)
++ tpmif_put(pak->tpmif);
++ kfree(pak->data_buffer);
++ /*
++ * cannot do tpmif_put(pak->tpmif); bad things happen
++ * on the last tpmif_put()
++ */
++ kfree(pak);
++}
++
++static int packet_set(struct packet *pak,
++ const unsigned char *buffer, u32 size)
++{
++ int rc = 0;
++ unsigned char *buf = kmalloc(size, GFP_KERNEL);
++
++ if (buf) {
++ pak->data_buffer = buf;
++ memcpy(buf, buffer, size);
++ pak->data_len = size;
++ } else {
++ rc = -ENOMEM;
++ }
++ return rc;
++}
++
++/*
++ * Write data to the shared memory and send it to the FE.
++ */
++static int packet_write(struct packet *pak,
++ const char *data, size_t size, int isuserbuffer)
++{
++ int rc = 0;
++
++ if ((pak->flags & PACKET_FLAG_CHECK_RESPONSESTATUS)) {
++#ifdef CONFIG_XEN_TPMDEV_CLOSE_IF_VTPM_FAILS
++ u32 res;
++
++ if (copy_from_buffer(&res,
++ &data[2 + 4], sizeof (res),
++ isuserbuffer)) {
++ return -EFAULT;
++ }
++
++ if (res != 0) {
++ /*
++ * Close down this device. Should have the
++ * FE notified about closure.
++ */
++ if (!pak->tpmif) {
++ return -EFAULT;
++ }
++ pak->tpmif->status = DISCONNECTING;
++ }
++#endif
++ }
++
++ if (0 != (pak->flags & PACKET_FLAG_DISCARD_RESPONSE)) {
++ /* Don't send a respone to this packet. Just acknowledge it. */
++ rc = size;
++ } else {
++ rc = _packet_write(pak, data, size, isuserbuffer);
++ }
++
++ return rc;
++}
++
++int _packet_write(struct packet *pak,
++ const char *data, size_t size, int isuserbuffer)
++{
++ /*
++ * Write into the shared memory pages directly
++ * and send it to the front end.
++ */
++ tpmif_t *tpmif = pak->tpmif;
++ grant_handle_t handle;
++ int rc = 0;
++ unsigned int i = 0;
++ unsigned int offset = 0;
++
++ if (tpmif == NULL) {
++ return -EFAULT;
++ }
++
++ if (tpmif->status == DISCONNECTED) {
++ return size;
++ }
++
++ while (offset < size && i < TPMIF_TX_RING_SIZE) {
++ unsigned int tocopy;
++ struct gnttab_map_grant_ref map_op;
++ struct gnttab_unmap_grant_ref unmap_op;
++ tpmif_tx_request_t *tx;
++
++ tx = &tpmif->tx->ring[i].req;
++
++ if (0 == tx->addr) {
++ DPRINTK("ERROR: Buffer for outgoing packet NULL?! i=%d\n", i);
++ return 0;
++ }
++
++ gnttab_set_map_op(&map_op, MMAP_VADDR(tpmif, i),
++ GNTMAP_host_map, tx->ref, tpmif->domid);
++
++ if (unlikely(HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
++ &map_op, 1))) {
++ BUG();
++ }
++
++ handle = map_op.handle;
++
++ if (map_op.status) {
++ DPRINTK(" Grant table operation failure !\n");
++ return 0;
++ }
++
++ tocopy = min_t(size_t, size - offset, PAGE_SIZE);
++
++ if (copy_from_buffer((void *)(MMAP_VADDR(tpmif, i) |
++ (tx->addr & ~PAGE_MASK)),
++ &data[offset], tocopy, isuserbuffer)) {
++ tpmif_put(tpmif);
++ return -EFAULT;
++ }
++ tx->size = tocopy;
++
++ gnttab_set_unmap_op(&unmap_op, MMAP_VADDR(tpmif, i),
++ GNTMAP_host_map, handle);
++
++ if (unlikely
++ (HYPERVISOR_grant_table_op
++ (GNTTABOP_unmap_grant_ref, &unmap_op, 1))) {
++ BUG();
++ }
++
++ offset += tocopy;
++ i++;
++ }
++
++ rc = offset;
++ DPRINTK("Notifying frontend via irq %d\n", tpmif->irq);
++ notify_remote_via_irq(tpmif->irq);
++
++ return rc;
++}
++
++/*
++ * Read data from the shared memory and copy it directly into the
++ * provided buffer. Advance the read_last indicator which tells
++ * how many bytes have already been read.
++ */
++static int packet_read(struct packet *pak, size_t numbytes,
++ char *buffer, size_t buffersize, int isuserbuffer)
++{
++ tpmif_t *tpmif = pak->tpmif;
++
++ /*
++ * Read 'numbytes' of data from the buffer. The first 4
++ * bytes are the instance number in network byte order,
++ * after that come the data from the shared memory buffer.
++ */
++ u32 to_copy;
++ u32 offset = 0;
++ u32 room_left = buffersize;
++
++ if (pak->last_read < 4) {
++ /*
++ * copy the instance number into the buffer
++ */
++ u32 instance_no = htonl(pak->tpm_instance);
++ u32 last_read = pak->last_read;
++
++ to_copy = min_t(size_t, 4 - last_read, numbytes);
++
++ if (copy_to_buffer(&buffer[0],
++ &(((u8 *) & instance_no)[last_read]),
++ to_copy, isuserbuffer)) {
++ return -EFAULT;
++ }
++
++ pak->last_read += to_copy;
++ offset += to_copy;
++ room_left -= to_copy;
++ }
++
++ /*
++ * If the packet has a data buffer appended, read from it...
++ */
++
++ if (room_left > 0) {
++ if (pak->data_buffer) {
++ u32 to_copy = min_t(u32, pak->data_len - offset, room_left);
++ u32 last_read = pak->last_read - 4;
++
++ if (copy_to_buffer(&buffer[offset],
++ &pak->data_buffer[last_read],
++ to_copy, isuserbuffer)) {
++ return -EFAULT;
++ }
++ pak->last_read += to_copy;
++ offset += to_copy;
++ } else {
++ offset = packet_read_shmem(pak,
++ tpmif,
++ offset,
++ buffer,
++ isuserbuffer, room_left);
++ }
++ }
++ return offset;
++}
++
++static int packet_read_shmem(struct packet *pak,
++ tpmif_t * tpmif,
++ u32 offset, char *buffer, int isuserbuffer,
++ u32 room_left)
++{
++ u32 last_read = pak->last_read - 4;
++ u32 i = (last_read / PAGE_SIZE);
++ u32 pg_offset = last_read & (PAGE_SIZE - 1);
++ u32 to_copy;
++ grant_handle_t handle;
++
++ tpmif_tx_request_t *tx;
++
++ tx = &tpmif->tx->ring[0].req;
++ /*
++ * Start copying data at the page with index 'index'
++ * and within that page at offset 'offset'.
++ * Copy a maximum of 'room_left' bytes.
++ */
++ to_copy = min_t(u32, PAGE_SIZE - pg_offset, room_left);
++ while (to_copy > 0) {
++ void *src;
++ struct gnttab_map_grant_ref map_op;
++ struct gnttab_unmap_grant_ref unmap_op;
++
++ tx = &tpmif->tx->ring[i].req;
++
++ gnttab_set_map_op(&map_op, MMAP_VADDR(tpmif, i),
++ GNTMAP_host_map, tx->ref, tpmif->domid);
++
++ if (unlikely(HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref,
++ &map_op, 1))) {
++ BUG();
++ }
++
++ if (map_op.status) {
++ DPRINTK(" Grant table operation failure !\n");
++ return -EFAULT;
++ }
++
++ handle = map_op.handle;
++
++ if (to_copy > tx->size) {
++ /*
++ * User requests more than what's available
++ */
++ to_copy = min_t(u32, tx->size, to_copy);
++ }
++
++ DPRINTK("Copying from mapped memory at %08lx\n",
++ (unsigned long)(MMAP_VADDR(tpmif, i) |
++ (tx->addr & ~PAGE_MASK)));
++
++ src = (void *)(MMAP_VADDR(tpmif, i) |
++ ((tx->addr & ~PAGE_MASK) + pg_offset));
++ if (copy_to_buffer(&buffer[offset],
++ src, to_copy, isuserbuffer)) {
++ return -EFAULT;
++ }
++
++ DPRINTK("Data from TPM-FE of domain %d are %d %d %d %d\n",
++ tpmif->domid, buffer[offset], buffer[offset + 1],
++ buffer[offset + 2], buffer[offset + 3]);
++
++ gnttab_set_unmap_op(&unmap_op, MMAP_VADDR(tpmif, i),
++ GNTMAP_host_map, handle);
++
++ if (unlikely
++ (HYPERVISOR_grant_table_op
++ (GNTTABOP_unmap_grant_ref, &unmap_op, 1))) {
++ BUG();
++ }
++
++ offset += to_copy;
++ pg_offset = 0;
++ last_read += to_copy;
++ room_left -= to_copy;
++
++ to_copy = min_t(u32, PAGE_SIZE, room_left);
++ i++;
++ } /* while (to_copy > 0) */
++ /*
++ * Adjust the last_read pointer
++ */
++ pak->last_read = last_read + 4;
++ return offset;
++}
++
++/* ============================================================
++ * The file layer for reading data from this device
++ * ============================================================
++ */
++static int vtpm_op_open(struct inode *inode, struct file *f)
++{
++ int rc = 0;
++ unsigned long flags;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ if (dataex.has_opener == 0) {
++ dataex.has_opener = 1;
++ } else {
++ rc = -EPERM;
++ }
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ return rc;
++}
++
++static ssize_t vtpm_op_read(struct file *file,
++ char __user * data, size_t size, loff_t * offset)
++{
++ int ret_size = -ENODATA;
++ struct packet *pak = NULL;
++ unsigned long flags;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ if (dataex.aborted) {
++ dataex.aborted = 0;
++ dataex.copied_so_far = 0;
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ return -EIO;
++ }
++
++ if (list_empty(&dataex.pending_pak)) {
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ wait_event_interruptible(dataex.wait_queue,
++ !list_empty(&dataex.pending_pak));
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ dataex.copied_so_far = 0;
++ }
++
++ if (!list_empty(&dataex.pending_pak)) {
++ unsigned int left;
++
++ pak = list_entry(dataex.pending_pak.next, struct packet, next);
++ left = pak->data_len - dataex.copied_so_far;
++ list_del(&pak->next);
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++
++ DPRINTK("size given by app: %d, available: %d\n", size, left);
++
++ ret_size = min_t(size_t, size, left);
++
++ ret_size = packet_read(pak, ret_size, data, size, 1);
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++
++ if (ret_size < 0) {
++ del_singleshot_timer_sync(&pak->processing_timer);
++ packet_free(pak);
++ dataex.copied_so_far = 0;
++ } else {
++ DPRINTK("Copied %d bytes to user buffer\n", ret_size);
++
++ dataex.copied_so_far += ret_size;
++ if (dataex.copied_so_far >= pak->data_len + 4) {
++ DPRINTK("All data from this packet given to app.\n");
++ /* All data given to app */
++
++ del_singleshot_timer_sync(&pak->
++ processing_timer);
++ list_add_tail(&pak->next, &dataex.current_pak);
++ /*
++ * The more fontends that are handled at the same time,
++ * the more time we give the TPM to process the request.
++ */
++ mod_timer(&pak->processing_timer,
++ jiffies + (num_frontends * 60 * HZ));
++ dataex.copied_so_far = 0;
++ } else {
++ list_add(&pak->next, &dataex.pending_pak);
++ }
++ }
++ }
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++
++ DPRINTK("Returning result from read to app: %d\n", ret_size);
++
++ return ret_size;
++}
++
++/*
++ * Write operation - only works after a previous read operation!
++ */
++static ssize_t vtpm_op_write(struct file *file,
++ const char __user * data, size_t size,
++ loff_t * offset)
++{
++ struct packet *pak;
++ int rc = 0;
++ unsigned int off = 4;
++ unsigned long flags;
++ struct vtpm_resp_hdr vrh;
++
++ /*
++ * Minimum required packet size is:
++ * 4 bytes for instance number
++ * 2 bytes for tag
++ * 4 bytes for paramSize
++ * 4 bytes for the ordinal
++ * sum: 14 bytes
++ */
++ if (size < sizeof (vrh))
++ return -EFAULT;
++
++ if (copy_from_user(&vrh, data, sizeof (vrh)))
++ return -EFAULT;
++
++ /* malformed packet? */
++ if ((off + ntohl(vrh.len_no)) != size)
++ return -EFAULT;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ pak = packet_find_instance(&dataex.current_pak,
++ ntohl(vrh.instance_no));
++
++ if (pak == NULL) {
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ DPRINTK(KERN_ALERT "No associated packet! (inst=%d)\n",
++ ntohl(vrh.instance_no));
++ return -EFAULT;
++ }
++
++ del_singleshot_timer_sync(&pak->processing_timer);
++ list_del(&pak->next);
++
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++
++ /*
++ * The first 'offset' bytes must be the instance number - skip them.
++ */
++ size -= off;
++
++ rc = packet_write(pak, &data[off], size, 1);
++
++ if (rc > 0) {
++ /* I neglected the first 4 bytes */
++ rc += off;
++ }
++ packet_free(pak);
++ return rc;
++}
++
++static int vtpm_op_release(struct inode *inode, struct file *file)
++{
++ unsigned long flags;
++
++ vtpm_release_packets(NULL, 1);
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ dataex.has_opener = 0;
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ return 0;
++}
++
++static unsigned int vtpm_op_poll(struct file *file,
++ struct poll_table_struct *pts)
++{
++ unsigned int flags = POLLOUT | POLLWRNORM;
++
++ poll_wait(file, &dataex.wait_queue, pts);
++ if (!list_empty(&dataex.pending_pak)) {
++ flags |= POLLIN | POLLRDNORM;
++ }
++ return flags;
++}
++
++static struct file_operations vtpm_ops = {
++ .owner = THIS_MODULE,
++ .llseek = no_llseek,
++ .open = vtpm_op_open,
++ .read = vtpm_op_read,
++ .write = vtpm_op_write,
++ .release = vtpm_op_release,
++ .poll = vtpm_op_poll,
++};
++
++static struct miscdevice vtpms_miscdevice = {
++ .minor = 225,
++ .name = "vtpm",
++ .fops = &vtpm_ops,
++};
++
++/***************************************************************
++ Utility functions
++***************************************************************/
++
++static int tpm_send_fail_message(struct packet *pak, u8 req_tag)
++{
++ int rc;
++ static const unsigned char tpm_error_message_fail[] = {
++ 0x00, 0x00,
++ 0x00, 0x00, 0x00, 0x0a,
++ 0x00, 0x00, 0x00, 0x09 /* TPM_FAIL */
++ };
++ unsigned char buffer[sizeof (tpm_error_message_fail)];
++
++ memcpy(buffer, tpm_error_message_fail,
++ sizeof (tpm_error_message_fail));
++ /*
++ * Insert the right response tag depending on the given tag
++ * All response tags are '+3' to the request tag.
++ */
++ buffer[1] = req_tag + 3;
++
++ /*
++ * Write the data to shared memory and notify the front-end
++ */
++ rc = packet_write(pak, buffer, sizeof (buffer), 0);
++
++ return rc;
++}
++
++static int _vtpm_release_packets(struct list_head *head,
++ tpmif_t * tpmif, int send_msgs)
++{
++ int aborted = 0;
++ int c = 0;
++ struct packet *pak;
++ struct list_head *pos, *tmp;
++
++ list_for_each_safe(pos, tmp, head) {
++ pak = list_entry(pos, struct packet, next);
++ c += 1;
++
++ if (tpmif == NULL || pak->tpmif == tpmif) {
++ int can_send = 0;
++
++ del_singleshot_timer_sync(&pak->processing_timer);
++ list_del(&pak->next);
++
++ if (pak->tpmif && pak->tpmif->status == CONNECTED) {
++ can_send = 1;
++ }
++
++ if (send_msgs && can_send) {
++ tpm_send_fail_message(pak, pak->req_tag);
++ }
++ packet_free(pak);
++ if (c == 1)
++ aborted = 1;
++ }
++ }
++ return aborted;
++}
++
++int vtpm_release_packets(tpmif_t * tpmif, int send_msgs)
++{
++ unsigned long flags;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++
++ dataex.aborted = _vtpm_release_packets(&dataex.pending_pak,
++ tpmif,
++ send_msgs);
++ _vtpm_release_packets(&dataex.current_pak, tpmif, send_msgs);
++
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++ return 0;
++}
++
++static int vtpm_queue_packet(struct packet *pak)
++{
++ int rc = 0;
++
++ if (dataex.has_opener) {
++ unsigned long flags;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ list_add_tail(&pak->next, &dataex.pending_pak);
++ /* give the TPM some time to pick up the request */
++ mod_timer(&pak->processing_timer, jiffies + (30 * HZ));
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++
++ wake_up_interruptible(&dataex.wait_queue);
++ } else {
++ rc = -EFAULT;
++ }
++ return rc;
++}
++
++static int vtpm_receive(tpmif_t * tpmif, u32 size)
++{
++ int rc = 0;
++ unsigned char buffer[10];
++ __be32 *native_size;
++ struct packet *pak = packet_alloc(tpmif, size, 0, 0);
++
++ if (!pak)
++ return -ENOMEM;
++ /*
++ * Read 10 bytes from the received buffer to test its
++ * content for validity.
++ */
++ if (sizeof (buffer) != packet_read(pak,
++ sizeof (buffer), buffer,
++ sizeof (buffer), 0)) {
++ goto failexit;
++ }
++ /*
++ * Reset the packet read pointer so we can read all its
++ * contents again.
++ */
++ packet_reset(pak);
++
++ native_size = (__force __be32 *) (&buffer[4 + 2]);
++ /*
++ * Verify that the size of the packet is correct
++ * as indicated and that there's actually someone reading packets.
++ * The minimum size of the packet is '10' for tag, size indicator
++ * and ordinal.
++ */
++ if (size < 10 ||
++ be32_to_cpu(*native_size) != size ||
++ 0 == dataex.has_opener || tpmif->status != CONNECTED) {
++ rc = -EINVAL;
++ goto failexit;
++ } else {
++ rc = vtpm_queue_packet(pak);
++ if (rc < 0)
++ goto failexit;
++ }
++ return 0;
++
++ failexit:
++ if (pak) {
++ tpm_send_fail_message(pak, buffer[4 + 1]);
++ packet_free(pak);
++ }
++ return rc;
++}
++
++/*
++ * Timeout function that gets invoked when a packet has not been processed
++ * during the timeout period.
++ * The packet must be on a list when this function is invoked. This
++ * also means that once its taken off a list, the timer must be
++ * destroyed as well.
++ */
++static void processing_timeout(unsigned long ptr)
++{
++ struct packet *pak = (struct packet *)ptr;
++ unsigned long flags;
++
++ write_lock_irqsave(&dataex.pak_lock, flags);
++ /*
++ * The packet needs to be searched whether it
++ * is still on the list.
++ */
++ if (pak == packet_find_packet(&dataex.pending_pak, pak) ||
++ pak == packet_find_packet(&dataex.current_pak, pak)) {
++ list_del(&pak->next);
++ if ((pak->flags & PACKET_FLAG_DISCARD_RESPONSE) == 0) {
++ tpm_send_fail_message(pak, pak->req_tag);
++ }
++ packet_free(pak);
++ }
++
++ write_unlock_irqrestore(&dataex.pak_lock, flags);
++}
++
++static void tpm_tx_action(unsigned long unused);
++static DECLARE_TASKLET(tpm_tx_tasklet, tpm_tx_action, 0);
++
++static struct list_head tpm_schedule_list;
++static spinlock_t tpm_schedule_list_lock;
++
++static inline void maybe_schedule_tx_action(void)
++{
++ smp_mb();
++ tasklet_schedule(&tpm_tx_tasklet);
++}
++
++static inline int __on_tpm_schedule_list(tpmif_t * tpmif)
++{
++ return tpmif->list.next != NULL;
++}
++
++static void remove_from_tpm_schedule_list(tpmif_t * tpmif)
++{
++ spin_lock_irq(&tpm_schedule_list_lock);
++ if (likely(__on_tpm_schedule_list(tpmif))) {
++ list_del(&tpmif->list);
++ tpmif->list.next = NULL;
++ tpmif_put(tpmif);
++ }
++ spin_unlock_irq(&tpm_schedule_list_lock);
++}
++
++static void add_to_tpm_schedule_list_tail(tpmif_t * tpmif)
++{
++ if (__on_tpm_schedule_list(tpmif))
++ return;
++
++ spin_lock_irq(&tpm_schedule_list_lock);
++ if (!__on_tpm_schedule_list(tpmif) && tpmif->active) {
++ list_add_tail(&tpmif->list, &tpm_schedule_list);
++ tpmif_get(tpmif);
++ }
++ spin_unlock_irq(&tpm_schedule_list_lock);
++}
++
++void tpmif_schedule_work(tpmif_t * tpmif)
++{
++ add_to_tpm_schedule_list_tail(tpmif);
++ maybe_schedule_tx_action();
++}
++
++void tpmif_deschedule_work(tpmif_t * tpmif)
++{
++ remove_from_tpm_schedule_list(tpmif);
++}
++
++static void tpm_tx_action(unsigned long unused)
++{
++ struct list_head *ent;
++ tpmif_t *tpmif;
++ tpmif_tx_request_t *tx;
++
++ DPRINTK("%s: Getting data from front-end(s)!\n", __FUNCTION__);
++
++ while (!list_empty(&tpm_schedule_list)) {
++ /* Get a tpmif from the list with work to do. */
++ ent = tpm_schedule_list.next;
++ tpmif = list_entry(ent, tpmif_t, list);
++ tpmif_get(tpmif);
++ remove_from_tpm_schedule_list(tpmif);
++
++ tx = &tpmif->tx->ring[0].req;
++
++ /* pass it up */
++ vtpm_receive(tpmif, tx->size);
++
++ tpmif_put(tpmif);
++ }
++}
++
++irqreturn_t tpmif_be_int(int irq, void *dev_id, struct pt_regs *regs)
++{
++ tpmif_t *tpmif = (tpmif_t *) dev_id;
++
++ add_to_tpm_schedule_list_tail(tpmif);
++ maybe_schedule_tx_action();
++ return IRQ_HANDLED;
++}
++
++static int __init tpmback_init(void)
++{
++ int rc;
++
++ if ((rc = misc_register(&vtpms_miscdevice)) != 0) {
++ printk(KERN_ALERT
++ "Could not register misc device for TPM BE.\n");
++ return rc;
++ }
++
++ dataex_init(&dataex);
++
++ spin_lock_init(&tpm_schedule_list_lock);
++ INIT_LIST_HEAD(&tpm_schedule_list);
++
++ tpmif_interface_init();
++ tpmif_xenbus_init();
++
++ printk(KERN_ALERT "Successfully initialized TPM backend driver.\n");
++
++ return 0;
++}
++
++module_init(tpmback_init);
++
++void __exit tpmback_exit(void)
++{
++ vtpm_release_packets(NULL, 0);
++ tpmif_xenbus_exit();
++ tpmif_interface_exit();
++ misc_deregister(&vtpms_miscdevice);
++}
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/tpmback/xenbus.c new/drivers/xen/tpmback/xenbus.c
+--- linux-2.6/drivers/xen/tpmback/xenbus.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/tpmback/xenbus.c 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,291 @@
++/* Xenbus code for tpmif backend
++ Copyright (C) 2005 IBM Corporation
++ Copyright (C) 2005 Rusty Russell <rusty@rustcorp.com.au>
++
++ This program is free software; you can redistribute it and/or modify
++ it under the terms of the GNU General Public License as published by
++ the Free Software Foundation; either version 2 of the License, or
++ (at your option) any later version.
++
++ This program is distributed in the hope that it will be useful,
++ but WITHOUT ANY WARRANTY; without even the implied warranty of
++ MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ GNU General Public License for more details.
++
++ You should have received a copy of the GNU General Public License
++ along with this program; if not, write to the Free Software
++ Foundation, Inc., 59 Temple Place, Suite 330, Boston, MA 02111-1307 USA
++*/
++#include <stdarg.h>
++#include <linux/module.h>
++#include <xen/xenbus.h>
++#include "common.h"
++
++struct backend_info
++{
++ struct xenbus_device *dev;
++
++ /* our communications channel */
++ tpmif_t *tpmif;
++
++ long int frontend_id;
++ long int instance; // instance of TPM
++ u8 is_instance_set;// whether instance number has been set
++
++ /* watch front end for changes */
++ struct xenbus_watch backend_watch;
++ enum xenbus_state frontend_state;
++};
++
++static void maybe_connect(struct backend_info *be);
++static void connect(struct backend_info *be);
++static int connect_ring(struct backend_info *be);
++static void backend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len);
++static void frontend_changed(struct xenbus_device *dev,
++ enum xenbus_state frontend_state);
++
++long int tpmback_get_instance(struct backend_info *bi)
++{
++ long int res = -1;
++ if (bi && bi->is_instance_set)
++ res = bi->instance;
++ return res;
++}
++
++static int tpmback_remove(struct xenbus_device *dev)
++{
++ struct backend_info *be = dev->dev.driver_data;
++
++ if (!be) return 0;
++
++ if (be->backend_watch.node) {
++ unregister_xenbus_watch(&be->backend_watch);
++ kfree(be->backend_watch.node);
++ be->backend_watch.node = NULL;
++ }
++ if (be->tpmif) {
++ be->tpmif->bi = NULL;
++ vtpm_release_packets(be->tpmif, 0);
++ tpmif_put(be->tpmif);
++ be->tpmif = NULL;
++ }
++ kfree(be);
++ dev->dev.driver_data = NULL;
++ return 0;
++}
++
++static int tpmback_probe(struct xenbus_device *dev,
++ const struct xenbus_device_id *id)
++{
++ int err;
++ struct backend_info *be = kzalloc(sizeof(struct backend_info),
++ GFP_KERNEL);
++
++ if (!be) {
++ xenbus_dev_fatal(dev, -ENOMEM,
++ "allocating backend structure");
++ return -ENOMEM;
++ }
++
++ be->is_instance_set = 0;
++ be->dev = dev;
++ dev->dev.driver_data = be;
++
++ err = xenbus_watch_path2(dev, dev->nodename,
++ "instance", &be->backend_watch,
++ backend_changed);
++ if (err) {
++ goto fail;
++ }
++
++ err = xenbus_switch_state(dev, XenbusStateInitWait);
++ if (err) {
++ goto fail;
++ }
++ return 0;
++fail:
++ tpmback_remove(dev);
++ return err;
++}
++
++
++static void backend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ int err;
++ long instance;
++ struct backend_info *be
++ = container_of(watch, struct backend_info, backend_watch);
++ struct xenbus_device *dev = be->dev;
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename,
++ "instance","%li", &instance);
++ if (XENBUS_EXIST_ERR(err)) {
++ return;
++ }
++
++ if (err != 1) {
++ xenbus_dev_fatal(dev, err, "reading instance");
++ return;
++ }
++
++ if (be->is_instance_set == 0) {
++ be->instance = instance;
++ be->is_instance_set = 1;
++ }
++}
++
++
++static void frontend_changed(struct xenbus_device *dev,
++ enum xenbus_state frontend_state)
++{
++ struct backend_info *be = dev->dev.driver_data;
++ int err;
++
++ be->frontend_state = frontend_state;
++
++ switch (frontend_state) {
++ case XenbusStateInitialising:
++ case XenbusStateInitialised:
++ break;
++
++ case XenbusStateConnected:
++ err = connect_ring(be);
++ if (err) {
++ return;
++ }
++ maybe_connect(be);
++ break;
++
++ case XenbusStateClosing:
++ be->instance = -1;
++ break;
++
++ case XenbusStateClosed:
++ device_unregister(&be->dev->dev);
++ tpmback_remove(dev);
++ break;
++
++ case XenbusStateUnknown:
++ case XenbusStateInitWait:
++ default:
++ xenbus_dev_fatal(dev, -EINVAL,
++ "saw state %d at frontend",
++ frontend_state);
++ break;
++ }
++}
++
++
++
++static void maybe_connect(struct backend_info *be)
++{
++ if (be->tpmif == NULL || be->tpmif->status == CONNECTED)
++ return;
++
++ connect(be);
++}
++
++
++static void connect(struct backend_info *be)
++{
++ struct xenbus_transaction xbt;
++ int err;
++ struct xenbus_device *dev = be->dev;
++ unsigned long ready = 1;
++
++again:
++ err = xenbus_transaction_start(&xbt);
++ if (err) {
++ xenbus_dev_fatal(be->dev, err, "starting transaction");
++ return;
++ }
++
++ err = xenbus_printf(xbt, be->dev->nodename,
++ "ready", "%lu", ready);
++ if (err) {
++ xenbus_dev_fatal(be->dev, err, "writing 'ready'");
++ goto abort;
++ }
++
++ err = xenbus_transaction_end(xbt, 0);
++ if (err == -EAGAIN)
++ goto again;
++ if (err)
++ xenbus_dev_fatal(be->dev, err, "end of transaction");
++
++ err = xenbus_switch_state(dev, XenbusStateConnected);
++ if (!err)
++ be->tpmif->status = CONNECTED;
++ return;
++abort:
++ xenbus_transaction_end(xbt, 1);
++}
++
++
++static int connect_ring(struct backend_info *be)
++{
++ struct xenbus_device *dev = be->dev;
++ unsigned long ring_ref;
++ unsigned int evtchn;
++ int err;
++
++ err = xenbus_gather(XBT_NIL, dev->otherend,
++ "ring-ref", "%lu", &ring_ref,
++ "event-channel", "%u", &evtchn, NULL);
++ if (err) {
++ xenbus_dev_error(dev, err,
++ "reading %s/ring-ref and event-channel",
++ dev->otherend);
++ return err;
++ }
++
++ if (!be->tpmif) {
++ be->tpmif = tpmif_find(dev->otherend_id, be);
++ if (IS_ERR(be->tpmif)) {
++ err = PTR_ERR(be->tpmif);
++ be->tpmif = NULL;
++ xenbus_dev_fatal(dev,err,"creating vtpm interface");
++ return err;
++ }
++ }
++
++ if (be->tpmif != NULL) {
++ err = tpmif_map(be->tpmif, ring_ref, evtchn);
++ if (err) {
++ xenbus_dev_error(dev, err,
++ "mapping shared-frame %lu port %u",
++ ring_ref, evtchn);
++ return err;
++ }
++ }
++ return 0;
++}
++
++
++static struct xenbus_device_id tpmback_ids[] = {
++ { "vtpm" },
++ { "" }
++};
++
++
++static struct xenbus_driver tpmback = {
++ .name = "vtpm",
++ .owner = THIS_MODULE,
++ .ids = tpmback_ids,
++ .probe = tpmback_probe,
++ .remove = tpmback_remove,
++ .otherend_changed = frontend_changed,
++};
++
++
++void tpmif_xenbus_init(void)
++{
++ xenbus_register_backend(&tpmback);
++}
++
++void tpmif_xenbus_exit(void)
++{
++ xenbus_unregister_driver(&tpmback);
++}
+diff -urNp linux-2.6/drivers/xen/util.c new/drivers/xen/util.c
+--- linux-2.6/drivers/xen/util.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/util.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,70 @@
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <linux/module.h>
++#include <linux/slab.h>
++#include <linux/vmalloc.h>
++#include <asm/uaccess.h>
++#include <xen/driver_util.h>
++
++static int f(pte_t *pte, struct page *pmd_page, unsigned long addr, void *data)
++{
++ /* apply_to_page_range() does all the hard work. */
++ return 0;
++}
++
++struct vm_struct *alloc_vm_area(unsigned long size)
++{
++ struct vm_struct *area;
++
++ area = get_vm_area(size, VM_IOREMAP);
++ if (area == NULL)
++ return NULL;
++
++ /*
++ * This ensures that page tables are constructed for this region
++ * of kernel virtual address space and mapped into init_mm.
++ */
++ if (apply_to_page_range(&init_mm, (unsigned long)area->addr,
++ area->size, f, NULL)) {
++ free_vm_area(area);
++ return NULL;
++ }
++
++ return area;
++}
++EXPORT_SYMBOL_GPL(alloc_vm_area);
++
++void free_vm_area(struct vm_struct *area)
++{
++ struct vm_struct *ret;
++ ret = remove_vm_area(area->addr);
++ BUG_ON(ret != area);
++ kfree(area);
++}
++EXPORT_SYMBOL_GPL(free_vm_area);
++
++void lock_vm_area(struct vm_struct *area)
++{
++ unsigned long i;
++ char c;
++
++ /*
++ * Prevent context switch to a lazy mm that doesn't have this area
++ * mapped into its page tables.
++ */
++ preempt_disable();
++
++ /*
++ * Ensure that the page tables are mapped into the current mm. The
++ * page-fault path will copy the page directory pointers from init_mm.
++ */
++ for (i = 0; i < area->size; i += PAGE_SIZE)
++ (void)__get_user(c, (char __user *)area->addr + i);
++}
++EXPORT_SYMBOL_GPL(lock_vm_area);
++
++void unlock_vm_area(struct vm_struct *area)
++{
++ preempt_enable();
++}
++EXPORT_SYMBOL_GPL(unlock_vm_area);
+diff -urNp linux-2.6/drivers/xen/xenbus/Makefile new/drivers/xen/xenbus/Makefile
+--- linux-2.6/drivers/xen/xenbus/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/Makefile 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,12 @@
++obj-y += xenbus.o
++obj-$(CONFIG_XEN_BACKEND) += xenbus_be.o
++
++xenbus_be-objs =
++xenbus_be-objs += xenbus_backend_client.o
++
++xenbus-objs =
++xenbus-objs += xenbus_client.o
++xenbus-objs += xenbus_comms.o
++xenbus-objs += xenbus_xs.o
++xenbus-objs += xenbus_probe.o
++obj-$(CONFIG_XEN_XENBUS_DEV) += xenbus_dev.o
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_backend_client.c new/drivers/xen/xenbus/xenbus_backend_client.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_backend_client.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_backend_client.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,135 @@
++/******************************************************************************
++ * Backend-client-facing interface for the Xenbus driver. In other words, the
++ * interface between the Xenbus and the device-specific code in the backend
++ * driver.
++ *
++ * Copyright (C) 2005-2006 XenSource Ltd
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/err.h>
++#include <xen/gnttab.h>
++#include <xen/xenbus.h>
++#include <xen/driver_util.h>
++
++/* Based on Rusty Russell's skeleton driver's map_page */
++struct vm_struct *xenbus_map_ring_valloc(struct xenbus_device *dev, int gnt_ref)
++{
++ struct gnttab_map_grant_ref op;
++ struct vm_struct *area;
++
++ area = alloc_vm_area(PAGE_SIZE);
++ if (!area)
++ return ERR_PTR(-ENOMEM);
++
++ gnttab_set_map_op(&op, (unsigned long)area->addr, GNTMAP_host_map,
++ gnt_ref, dev->otherend_id);
++
++ lock_vm_area(area);
++ BUG_ON(HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1));
++ unlock_vm_area(area);
++
++ if (op.status != GNTST_okay) {
++ free_vm_area(area);
++ xenbus_dev_fatal(dev, op.status,
++ "mapping in shared page %d from domain %d",
++ gnt_ref, dev->otherend_id);
++ BUG_ON(!IS_ERR(ERR_PTR(op.status)));
++ return ERR_PTR(op.status);
++ }
++
++ /* Stuff the handle in an unused field */
++ area->phys_addr = (unsigned long)op.handle;
++
++ return area;
++}
++EXPORT_SYMBOL_GPL(xenbus_map_ring_valloc);
++
++
++int xenbus_map_ring(struct xenbus_device *dev, int gnt_ref,
++ grant_handle_t *handle, void *vaddr)
++{
++ struct gnttab_map_grant_ref op;
++
++ gnttab_set_map_op(&op, (unsigned long)vaddr, GNTMAP_host_map,
++ gnt_ref, dev->otherend_id);
++ BUG_ON(HYPERVISOR_grant_table_op(GNTTABOP_map_grant_ref, &op, 1));
++
++ if (op.status != GNTST_okay) {
++ xenbus_dev_fatal(dev, op.status,
++ "mapping in shared page %d from domain %d",
++ gnt_ref, dev->otherend_id);
++ } else
++ *handle = op.handle;
++
++ return op.status;
++}
++EXPORT_SYMBOL_GPL(xenbus_map_ring);
++
++
++/* Based on Rusty Russell's skeleton driver's unmap_page */
++int xenbus_unmap_ring_vfree(struct xenbus_device *dev, struct vm_struct *area)
++{
++ struct gnttab_unmap_grant_ref op;
++
++ gnttab_set_unmap_op(&op, (unsigned long)area->addr, GNTMAP_host_map,
++ (grant_handle_t)area->phys_addr);
++
++ lock_vm_area(area);
++ BUG_ON(HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1));
++ unlock_vm_area(area);
++
++ if (op.status == GNTST_okay)
++ free_vm_area(area);
++ else
++ xenbus_dev_error(dev, op.status,
++ "unmapping page at handle %d error %d",
++ (int16_t)area->phys_addr, op.status);
++
++ return op.status;
++}
++EXPORT_SYMBOL_GPL(xenbus_unmap_ring_vfree);
++
++
++int xenbus_unmap_ring(struct xenbus_device *dev,
++ grant_handle_t handle, void *vaddr)
++{
++ struct gnttab_unmap_grant_ref op;
++
++ gnttab_set_unmap_op(&op, (unsigned long)vaddr, GNTMAP_host_map,
++ handle);
++ BUG_ON(HYPERVISOR_grant_table_op(GNTTABOP_unmap_grant_ref, &op, 1));
++
++ if (op.status != GNTST_okay)
++ xenbus_dev_error(dev, op.status,
++ "unmapping page at handle %d error %d",
++ handle, op.status);
++
++ return op.status;
++}
++EXPORT_SYMBOL_GPL(xenbus_unmap_ring);
++
++MODULE_LICENSE("Dual BSD/GPL");
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_client.c new/drivers/xen/xenbus/xenbus_client.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_client.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_client.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,281 @@
++/******************************************************************************
++ * Client-facing interface for the Xenbus driver. In other words, the
++ * interface between the Xenbus and the device-specific code, be it the
++ * frontend or the backend of that driver.
++ *
++ * Copyright (C) 2005 XenSource Ltd
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <xen/evtchn.h>
++#include <xen/gnttab.h>
++#include <xen/xenbus.h>
++#include <xen/driver_util.h>
++
++/* xenbus_probe.c */
++extern char *kasprintf(const char *fmt, ...);
++
++#define DPRINTK(fmt, args...) \
++ pr_debug("xenbus_client (%s:%d) " fmt ".\n", __FUNCTION__, __LINE__, ##args)
++
++int xenbus_watch_path(struct xenbus_device *dev, const char *path,
++ struct xenbus_watch *watch,
++ void (*callback)(struct xenbus_watch *,
++ const char **, unsigned int))
++{
++ int err;
++
++ watch->node = path;
++ watch->callback = callback;
++
++ err = register_xenbus_watch(watch);
++
++ if (err) {
++ watch->node = NULL;
++ watch->callback = NULL;
++ xenbus_dev_fatal(dev, err, "adding watch on %s", path);
++ }
++
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_watch_path);
++
++
++int xenbus_watch_path2(struct xenbus_device *dev, const char *path,
++ const char *path2, struct xenbus_watch *watch,
++ void (*callback)(struct xenbus_watch *,
++ const char **, unsigned int))
++{
++ int err;
++ char *state = kasprintf("%s/%s", path, path2);
++ if (!state) {
++ xenbus_dev_fatal(dev, -ENOMEM, "allocating path for watch");
++ return -ENOMEM;
++ }
++ err = xenbus_watch_path(dev, state, watch, callback);
++
++ if (err)
++ kfree(state);
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_watch_path2);
++
++
++int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state state)
++{
++ /* We check whether the state is currently set to the given value, and
++ if not, then the state is set. We don't want to unconditionally
++ write the given state, because we don't want to fire watches
++ unnecessarily. Furthermore, if the node has gone, we don't write
++ to it, as the device will be tearing down, and we don't want to
++ resurrect that directory.
++
++ Note that, because of this cached value of our state, this function
++ will not work inside a Xenstore transaction (something it was
++ trying to in the past) because dev->state would not get reset if
++ the transaction was aborted.
++
++ */
++
++ int current_state;
++ int err;
++
++ if (state == dev->state)
++ return 0;
++
++ err = xenbus_scanf(XBT_NIL, dev->nodename, "state", "%d",
++ ¤t_state);
++ if (err != 1)
++ return 0;
++
++ err = xenbus_printf(XBT_NIL, dev->nodename, "state", "%d", state);
++ if (err) {
++ if (state != XenbusStateClosing) /* Avoid looping */
++ xenbus_dev_fatal(dev, err, "writing new state");
++ return err;
++ }
++
++ dev->state = state;
++
++ return 0;
++}
++EXPORT_SYMBOL_GPL(xenbus_switch_state);
++
++
++/**
++ * Return the path to the error node for the given device, or NULL on failure.
++ * If the value returned is non-NULL, then it is the caller's to kfree.
++ */
++static char *error_path(struct xenbus_device *dev)
++{
++ return kasprintf("error/%s", dev->nodename);
++}
++
++
++void _dev_error(struct xenbus_device *dev, int err, const char *fmt,
++ va_list ap)
++{
++ int ret;
++ unsigned int len;
++ char *printf_buffer = NULL, *path_buffer = NULL;
++
++#define PRINTF_BUFFER_SIZE 4096
++ printf_buffer = kmalloc(PRINTF_BUFFER_SIZE, GFP_KERNEL);
++ if (printf_buffer == NULL)
++ goto fail;
++
++ len = sprintf(printf_buffer, "%i ", -err);
++ ret = vsnprintf(printf_buffer+len, PRINTF_BUFFER_SIZE-len, fmt, ap);
++
++ BUG_ON(len + ret > PRINTF_BUFFER_SIZE-1);
++
++ dev_err(&dev->dev, "%s\n", printf_buffer);
++
++ path_buffer = error_path(dev);
++
++ if (path_buffer == NULL) {
++ printk("xenbus: failed to write error node for %s (%s)\n",
++ dev->nodename, printf_buffer);
++ goto fail;
++ }
++
++ if (xenbus_write(XBT_NIL, path_buffer, "error", printf_buffer) != 0) {
++ printk("xenbus: failed to write error node for %s (%s)\n",
++ dev->nodename, printf_buffer);
++ goto fail;
++ }
++
++fail:
++ if (printf_buffer)
++ kfree(printf_buffer);
++ if (path_buffer)
++ kfree(path_buffer);
++}
++
++
++void xenbus_dev_error(struct xenbus_device *dev, int err, const char *fmt,
++ ...)
++{
++ va_list ap;
++
++ va_start(ap, fmt);
++ _dev_error(dev, err, fmt, ap);
++ va_end(ap);
++}
++EXPORT_SYMBOL_GPL(xenbus_dev_error);
++
++
++void xenbus_dev_fatal(struct xenbus_device *dev, int err, const char *fmt,
++ ...)
++{
++ va_list ap;
++
++ va_start(ap, fmt);
++ _dev_error(dev, err, fmt, ap);
++ va_end(ap);
++
++ xenbus_switch_state(dev, XenbusStateClosing);
++}
++EXPORT_SYMBOL_GPL(xenbus_dev_fatal);
++
++
++int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn)
++{
++ int err = gnttab_grant_foreign_access(dev->otherend_id, ring_mfn, 0);
++ if (err < 0)
++ xenbus_dev_fatal(dev, err, "granting access to ring page");
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_grant_ring);
++
++
++int xenbus_alloc_evtchn(struct xenbus_device *dev, int *port)
++{
++ struct evtchn_alloc_unbound alloc_unbound;
++ int err;
++
++ alloc_unbound.dom = DOMID_SELF;
++ alloc_unbound.remote_dom = dev->otherend_id;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound,
++ &alloc_unbound);
++ if (err)
++ xenbus_dev_fatal(dev, err, "allocating event channel");
++ else
++ *port = alloc_unbound.port;
++
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_alloc_evtchn);
++
++
++int xenbus_bind_evtchn(struct xenbus_device *dev, int remote_port, int *port)
++{
++ struct evtchn_bind_interdomain bind_interdomain;
++ int err;
++
++ bind_interdomain.remote_dom = dev->otherend_id;
++ bind_interdomain.remote_port = remote_port,
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_bind_interdomain,
++ &bind_interdomain);
++ if (err)
++ xenbus_dev_fatal(dev, err,
++ "binding to event channel %d from domain %d",
++ remote_port, dev->otherend_id);
++ else
++ *port = bind_interdomain.local_port;
++
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_bind_evtchn);
++
++
++int xenbus_free_evtchn(struct xenbus_device *dev, int port)
++{
++ struct evtchn_close close;
++ int err;
++
++ close.port = port;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_close, &close);
++ if (err)
++ xenbus_dev_error(dev, err, "freeing event channel %d", port);
++
++ return err;
++}
++
++
++enum xenbus_state xenbus_read_driver_state(const char *path)
++{
++ enum xenbus_state result;
++ int err = xenbus_gather(XBT_NIL, path, "state", "%d", &result, NULL);
++ if (err)
++ result = XenbusStateClosed;
++
++ return result;
++}
++EXPORT_SYMBOL_GPL(xenbus_read_driver_state);
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_comms.c new/drivers/xen/xenbus/xenbus_comms.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_comms.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_comms.c 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,208 @@
++/******************************************************************************
++ * xenbus_comms.c
++ *
++ * Low level code to talks to Xen Store: ringbuffer and event channel.
++ *
++ * Copyright (C) 2005 Rusty Russell, IBM Corporation
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <asm/hypervisor.h>
++#include <xen/evtchn.h>
++#include <linux/wait.h>
++#include <linux/interrupt.h>
++#include <linux/sched.h>
++#include <linux/err.h>
++#include <xen/xenbus.h>
++#include "xenbus_comms.h"
++
++static int xenbus_irq;
++
++extern void xenbus_probe(void *);
++extern int xenstored_ready;
++static DECLARE_WORK(probe_work, xenbus_probe, NULL);
++
++DECLARE_WAIT_QUEUE_HEAD(xb_waitq);
++
++static inline struct xenstore_domain_interface *xenstore_domain_interface(void)
++{
++ return mfn_to_virt(xen_start_info->store_mfn);
++}
++
++static irqreturn_t wake_waiting(int irq, void *unused, struct pt_regs *regs)
++{
++ if (unlikely(xenstored_ready == 0)) {
++ xenstored_ready = 1;
++ schedule_work(&probe_work);
++ }
++
++ wake_up(&xb_waitq);
++ return IRQ_HANDLED;
++}
++
++static int check_indexes(XENSTORE_RING_IDX cons, XENSTORE_RING_IDX prod)
++{
++ return ((prod - cons) <= XENSTORE_RING_SIZE);
++}
++
++static void *get_output_chunk(XENSTORE_RING_IDX cons,
++ XENSTORE_RING_IDX prod,
++ char *buf, uint32_t *len)
++{
++ *len = XENSTORE_RING_SIZE - MASK_XENSTORE_IDX(prod);
++ if ((XENSTORE_RING_SIZE - (prod - cons)) < *len)
++ *len = XENSTORE_RING_SIZE - (prod - cons);
++ return buf + MASK_XENSTORE_IDX(prod);
++}
++
++static const void *get_input_chunk(XENSTORE_RING_IDX cons,
++ XENSTORE_RING_IDX prod,
++ const char *buf, uint32_t *len)
++{
++ *len = XENSTORE_RING_SIZE - MASK_XENSTORE_IDX(cons);
++ if ((prod - cons) < *len)
++ *len = prod - cons;
++ return buf + MASK_XENSTORE_IDX(cons);
++}
++
++int xb_write(const void *data, unsigned len)
++{
++ struct xenstore_domain_interface *intf = xenstore_domain_interface();
++ XENSTORE_RING_IDX cons, prod;
++ int rc;
++
++ while (len != 0) {
++ void *dst;
++ unsigned int avail;
++
++ rc = wait_event_interruptible(
++ xb_waitq,
++ (intf->req_prod - intf->req_cons) !=
++ XENSTORE_RING_SIZE);
++ if (rc < 0)
++ return rc;
++
++ /* Read indexes, then verify. */
++ cons = intf->req_cons;
++ prod = intf->req_prod;
++ mb();
++ if (!check_indexes(cons, prod)) {
++ intf->req_cons = intf->req_prod = 0;
++ return -EIO;
++ }
++
++ dst = get_output_chunk(cons, prod, intf->req, &avail);
++ if (avail == 0)
++ continue;
++ if (avail > len)
++ avail = len;
++
++ memcpy(dst, data, avail);
++ data += avail;
++ len -= avail;
++
++ /* Other side must not see new header until data is there. */
++ wmb();
++ intf->req_prod += avail;
++
++ /* This implies mb() before other side sees interrupt. */
++ notify_remote_via_evtchn(xen_start_info->store_evtchn);
++ }
++
++ return 0;
++}
++
++int xb_read(void *data, unsigned len)
++{
++ struct xenstore_domain_interface *intf = xenstore_domain_interface();
++ XENSTORE_RING_IDX cons, prod;
++ int rc;
++
++ while (len != 0) {
++ unsigned int avail;
++ const char *src;
++
++ rc = wait_event_interruptible(
++ xb_waitq,
++ intf->rsp_cons != intf->rsp_prod);
++ if (rc < 0)
++ return rc;
++
++ /* Read indexes, then verify. */
++ cons = intf->rsp_cons;
++ prod = intf->rsp_prod;
++ mb();
++ if (!check_indexes(cons, prod)) {
++ intf->rsp_cons = intf->rsp_prod = 0;
++ return -EIO;
++ }
++
++ src = get_input_chunk(cons, prod, intf->rsp, &avail);
++ if (avail == 0)
++ continue;
++ if (avail > len)
++ avail = len;
++
++ /* We must read header before we read data. */
++ rmb();
++
++ memcpy(data, src, avail);
++ data += avail;
++ len -= avail;
++
++ /* Other side must not see free space until we've copied out */
++ mb();
++ intf->rsp_cons += avail;
++
++ pr_debug("Finished read of %i bytes (%i to go)\n", avail, len);
++
++ /* Implies mb(): they will see new header. */
++ notify_remote_via_evtchn(xen_start_info->store_evtchn);
++ }
++
++ return 0;
++}
++
++/* Set up interrupt handler off store event channel. */
++int xb_init_comms(void)
++{
++ int err;
++
++ if (xenbus_irq)
++ unbind_from_irqhandler(xenbus_irq, &xb_waitq);
++
++ err = bind_evtchn_to_irqhandler(
++ xen_start_info->store_evtchn, wake_waiting,
++ 0, "xenbus", &xb_waitq);
++ if (err <= 0) {
++ printk(KERN_ERR "XENBUS request irq failed %i\n", err);
++ return err;
++ }
++
++ xenbus_irq = err;
++
++ return 0;
++}
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_comms.h new/drivers/xen/xenbus/xenbus_comms.h
+--- linux-2.6/drivers/xen/xenbus/xenbus_comms.h 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_comms.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,43 @@
++/*
++ * Private include for xenbus communications.
++ *
++ * Copyright (C) 2005 Rusty Russell, IBM Corporation
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef _XENBUS_COMMS_H
++#define _XENBUS_COMMS_H
++
++int xs_init(void);
++int xb_init_comms(void);
++
++/* Low level routines. */
++int xb_write(const void *data, unsigned len);
++int xb_read(void *data, unsigned len);
++int xs_input_avail(void);
++extern wait_queue_head_t xb_waitq;
++
++#endif /* _XENBUS_COMMS_H */
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_dev.c new/drivers/xen/xenbus/xenbus_dev.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_dev.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_dev.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,245 @@
++/*
++ * xenbus_dev.c
++ *
++ * Driver giving user-space access to the kernel's xenbus connection
++ * to xenstore.
++ *
++ * Copyright (c) 2005, Christian Limpach
++ * Copyright (c) 2005, Rusty Russell, IBM Corporation
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/errno.h>
++#include <linux/uio.h>
++#include <linux/notifier.h>
++#include <linux/wait.h>
++#include <linux/fs.h>
++#include <linux/poll.h>
++
++#include "xenbus_comms.h"
++
++#include <asm/uaccess.h>
++#include <asm/hypervisor.h>
++#include <xen/xenbus.h>
++#include <xen/xen_proc.h>
++#include <asm/hypervisor.h>
++
++struct xenbus_dev_transaction {
++ struct list_head list;
++ struct xenbus_transaction handle;
++};
++
++struct xenbus_dev_data {
++ /* In-progress transaction. */
++ struct list_head transactions;
++
++ /* Partial request. */
++ unsigned int len;
++ union {
++ struct xsd_sockmsg msg;
++ char buffer[PAGE_SIZE];
++ } u;
++
++ /* Response queue. */
++#define MASK_READ_IDX(idx) ((idx)&(PAGE_SIZE-1))
++ char read_buffer[PAGE_SIZE];
++ unsigned int read_cons, read_prod;
++ wait_queue_head_t read_waitq;
++};
++
++static struct proc_dir_entry *xenbus_dev_intf;
++
++static ssize_t xenbus_dev_read(struct file *filp,
++ char __user *ubuf,
++ size_t len, loff_t *ppos)
++{
++ struct xenbus_dev_data *u = filp->private_data;
++ int i;
++
++ if (wait_event_interruptible(u->read_waitq,
++ u->read_prod != u->read_cons))
++ return -EINTR;
++
++ for (i = 0; i < len; i++) {
++ if (u->read_cons == u->read_prod)
++ break;
++ put_user(u->read_buffer[MASK_READ_IDX(u->read_cons)], ubuf+i);
++ u->read_cons++;
++ }
++
++ return i;
++}
++
++static void queue_reply(struct xenbus_dev_data *u,
++ char *data, unsigned int len)
++{
++ int i;
++
++ for (i = 0; i < len; i++, u->read_prod++)
++ u->read_buffer[MASK_READ_IDX(u->read_prod)] = data[i];
++
++ BUG_ON((u->read_prod - u->read_cons) > sizeof(u->read_buffer));
++
++ wake_up(&u->read_waitq);
++}
++
++static ssize_t xenbus_dev_write(struct file *filp,
++ const char __user *ubuf,
++ size_t len, loff_t *ppos)
++{
++ struct xenbus_dev_data *u = filp->private_data;
++ struct xenbus_dev_transaction *trans = NULL;
++ uint32_t msg_type;
++ void *reply;
++
++ if ((len + u->len) > sizeof(u->u.buffer))
++ return -EINVAL;
++
++ if (copy_from_user(u->u.buffer + u->len, ubuf, len) != 0)
++ return -EFAULT;
++
++ u->len += len;
++ if (u->len < (sizeof(u->u.msg) + u->u.msg.len))
++ return len;
++
++ msg_type = u->u.msg.type;
++
++ switch (msg_type) {
++ case XS_TRANSACTION_START:
++ case XS_TRANSACTION_END:
++ case XS_DIRECTORY:
++ case XS_READ:
++ case XS_GET_PERMS:
++ case XS_RELEASE:
++ case XS_GET_DOMAIN_PATH:
++ case XS_WRITE:
++ case XS_MKDIR:
++ case XS_RM:
++ case XS_SET_PERMS:
++ if (msg_type == XS_TRANSACTION_START) {
++ trans = kmalloc(sizeof(*trans), GFP_KERNEL);
++ if (!trans)
++ return -ENOMEM;
++ }
++
++ reply = xenbus_dev_request_and_reply(&u->u.msg);
++ if (IS_ERR(reply)) {
++ kfree(trans);
++ return PTR_ERR(reply);
++ }
++
++ if (msg_type == XS_TRANSACTION_START) {
++ trans->handle.id = simple_strtoul(reply, NULL, 0);
++ list_add(&trans->list, &u->transactions);
++ } else if (msg_type == XS_TRANSACTION_END) {
++ list_for_each_entry(trans, &u->transactions, list)
++ if (trans->handle.id == u->u.msg.tx_id)
++ break;
++ BUG_ON(&trans->list == &u->transactions);
++ list_del(&trans->list);
++ kfree(trans);
++ }
++ queue_reply(u, (char *)&u->u.msg, sizeof(u->u.msg));
++ queue_reply(u, (char *)reply, u->u.msg.len);
++ kfree(reply);
++ break;
++
++ default:
++ return -EINVAL;
++ }
++
++ u->len = 0;
++ return len;
++}
++
++static int xenbus_dev_open(struct inode *inode, struct file *filp)
++{
++ struct xenbus_dev_data *u;
++
++ if (xen_start_info->store_evtchn == 0)
++ return -ENOENT;
++
++ nonseekable_open(inode, filp);
++
++ u = kzalloc(sizeof(*u), GFP_KERNEL);
++ if (u == NULL)
++ return -ENOMEM;
++
++ INIT_LIST_HEAD(&u->transactions);
++ init_waitqueue_head(&u->read_waitq);
++
++ filp->private_data = u;
++
++ return 0;
++}
++
++static int xenbus_dev_release(struct inode *inode, struct file *filp)
++{
++ struct xenbus_dev_data *u = filp->private_data;
++ struct xenbus_dev_transaction *trans, *tmp;
++
++ list_for_each_entry_safe(trans, tmp, &u->transactions, list) {
++ xenbus_transaction_end(trans->handle, 1);
++ list_del(&trans->list);
++ kfree(trans);
++ }
++
++ kfree(u);
++
++ return 0;
++}
++
++static unsigned int xenbus_dev_poll(struct file *file, poll_table *wait)
++{
++ struct xenbus_dev_data *u = file->private_data;
++
++ poll_wait(file, &u->read_waitq, wait);
++ if (u->read_cons != u->read_prod)
++ return POLLIN | POLLRDNORM;
++ return 0;
++}
++
++static struct file_operations xenbus_dev_file_ops = {
++ .read = xenbus_dev_read,
++ .write = xenbus_dev_write,
++ .open = xenbus_dev_open,
++ .release = xenbus_dev_release,
++ .poll = xenbus_dev_poll,
++};
++
++static int __init
++xenbus_dev_init(void)
++{
++ xenbus_dev_intf = create_xen_proc_entry("xenbus", 0400);
++ if (xenbus_dev_intf)
++ xenbus_dev_intf->proc_fops = &xenbus_dev_file_ops;
++
++ return 0;
++}
++
++__initcall(xenbus_dev_init);
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_probe.c new/drivers/xen/xenbus/xenbus_probe.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_probe.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_probe.c 2006-07-07 16:05:52.000000000 +0200
+@@ -0,0 +1,1107 @@
++/******************************************************************************
++ * Talks to Xen Store to figure out what devices we have.
++ *
++ * Copyright (C) 2005 Rusty Russell, IBM Corporation
++ * Copyright (C) 2005 Mike Wray, Hewlett-Packard
++ * Copyright (C) 2005, 2006 XenSource Ltd
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#define DPRINTK(fmt, args...) \
++ pr_debug("xenbus_probe (%s:%d) " fmt ".\n", \
++ __FUNCTION__, __LINE__, ##args)
++
++#include <linux/kernel.h>
++#include <linux/err.h>
++#include <linux/string.h>
++#include <linux/ctype.h>
++#include <linux/fcntl.h>
++#include <linux/mm.h>
++#include <linux/notifier.h>
++#include <linux/kthread.h>
++
++#include <asm/io.h>
++#include <asm/page.h>
++#include <asm/pgtable.h>
++#include <asm/hypervisor.h>
++#include <xen/xenbus.h>
++#include <xen/xen_proc.h>
++#include <xen/evtchn.h>
++#include <xen/features.h>
++
++#include "xenbus_comms.h"
++
++extern struct mutex xenwatch_mutex;
++
++static BLOCKING_NOTIFIER_HEAD(xenstore_notifier_list);
++
++/* If something in array of ids matches this device, return it. */
++static const struct xenbus_device_id *
++match_device(const struct xenbus_device_id *arr, struct xenbus_device *dev)
++{
++ for (; *arr->devicetype != '\0'; arr++) {
++ if (!strcmp(arr->devicetype, dev->devicetype))
++ return arr;
++ }
++ return NULL;
++}
++
++static int xenbus_match(struct device *_dev, struct device_driver *_drv)
++{
++ struct xenbus_driver *drv = to_xenbus_driver(_drv);
++
++ if (!drv->ids)
++ return 0;
++
++ return match_device(drv->ids, to_xenbus_device(_dev)) != NULL;
++}
++
++struct xen_bus_type
++{
++ char *root;
++ unsigned int levels;
++ int (*get_bus_id)(char bus_id[BUS_ID_SIZE], const char *nodename);
++ int (*probe)(const char *type, const char *dir);
++ struct bus_type bus;
++ struct device dev;
++};
++
++
++/* device/<type>/<id> => <type>-<id> */
++static int frontend_bus_id(char bus_id[BUS_ID_SIZE], const char *nodename)
++{
++ nodename = strchr(nodename, '/');
++ if (!nodename || strlen(nodename + 1) >= BUS_ID_SIZE) {
++ printk(KERN_WARNING "XENBUS: bad frontend %s\n", nodename);
++ return -EINVAL;
++ }
++
++ strlcpy(bus_id, nodename + 1, BUS_ID_SIZE);
++ if (!strchr(bus_id, '/')) {
++ printk(KERN_WARNING "XENBUS: bus_id %s no slash\n", bus_id);
++ return -EINVAL;
++ }
++ *strchr(bus_id, '/') = '-';
++ return 0;
++}
++
++
++static void free_otherend_details(struct xenbus_device *dev)
++{
++ kfree(dev->otherend);
++ dev->otherend = NULL;
++}
++
++
++static void free_otherend_watch(struct xenbus_device *dev)
++{
++ if (dev->otherend_watch.node) {
++ unregister_xenbus_watch(&dev->otherend_watch);
++ kfree(dev->otherend_watch.node);
++ dev->otherend_watch.node = NULL;
++ }
++}
++
++
++static int read_otherend_details(struct xenbus_device *xendev,
++ char *id_node, char *path_node)
++{
++ int err = xenbus_gather(XBT_NIL, xendev->nodename,
++ id_node, "%i", &xendev->otherend_id,
++ path_node, NULL, &xendev->otherend,
++ NULL);
++ if (err) {
++ xenbus_dev_fatal(xendev, err,
++ "reading other end details from %s",
++ xendev->nodename);
++ return err;
++ }
++ if (strlen(xendev->otherend) == 0 ||
++ !xenbus_exists(XBT_NIL, xendev->otherend, "")) {
++ xenbus_dev_fatal(xendev, -ENOENT, "missing other end from %s",
++ xendev->nodename);
++ free_otherend_details(xendev);
++ return -ENOENT;
++ }
++
++ return 0;
++}
++
++
++static int read_backend_details(struct xenbus_device *xendev)
++{
++ return read_otherend_details(xendev, "backend-id", "backend");
++}
++
++
++static int read_frontend_details(struct xenbus_device *xendev)
++{
++ return read_otherend_details(xendev, "frontend-id", "frontend");
++}
++
++
++/* Bus type for frontend drivers. */
++static int xenbus_probe_frontend(const char *type, const char *name);
++static struct xen_bus_type xenbus_frontend = {
++ .root = "device",
++ .levels = 2, /* device/type/<id> */
++ .get_bus_id = frontend_bus_id,
++ .probe = xenbus_probe_frontend,
++ .bus = {
++ .name = "xen",
++ .match = xenbus_match,
++ },
++ .dev = {
++ .bus_id = "xen",
++ },
++};
++
++/* backend/<type>/<fe-uuid>/<id> => <type>-<fe-domid>-<id> */
++static int backend_bus_id(char bus_id[BUS_ID_SIZE], const char *nodename)
++{
++ int domid, err;
++ const char *devid, *type, *frontend;
++ unsigned int typelen;
++
++ type = strchr(nodename, '/');
++ if (!type)
++ return -EINVAL;
++ type++;
++ typelen = strcspn(type, "/");
++ if (!typelen || type[typelen] != '/')
++ return -EINVAL;
++
++ devid = strrchr(nodename, '/') + 1;
++
++ err = xenbus_gather(XBT_NIL, nodename, "frontend-id", "%i", &domid,
++ "frontend", NULL, &frontend,
++ NULL);
++ if (err)
++ return err;
++ if (strlen(frontend) == 0)
++ err = -ERANGE;
++ if (!err && !xenbus_exists(XBT_NIL, frontend, ""))
++ err = -ENOENT;
++
++ kfree(frontend);
++
++ if (err)
++ return err;
++
++ if (snprintf(bus_id, BUS_ID_SIZE,
++ "%.*s-%i-%s", typelen, type, domid, devid) >= BUS_ID_SIZE)
++ return -ENOSPC;
++ return 0;
++}
++
++static int xenbus_uevent_backend(struct device *dev, char **envp,
++ int num_envp, char *buffer, int buffer_size);
++static int xenbus_probe_backend(const char *type, const char *domid);
++static struct xen_bus_type xenbus_backend = {
++ .root = "backend",
++ .levels = 3, /* backend/type/<frontend>/<id> */
++ .get_bus_id = backend_bus_id,
++ .probe = xenbus_probe_backend,
++ .bus = {
++ .name = "xen-backend",
++ .match = xenbus_match,
++ .uevent = xenbus_uevent_backend,
++ },
++ .dev = {
++ .bus_id = "xen-backend",
++ },
++};
++
++static int xenbus_uevent_backend(struct device *dev, char **envp,
++ int num_envp, char *buffer, int buffer_size)
++{
++ struct xenbus_device *xdev;
++ struct xenbus_driver *drv;
++ int i = 0;
++ int length = 0;
++
++ DPRINTK("");
++
++ if (dev == NULL)
++ return -ENODEV;
++
++ xdev = to_xenbus_device(dev);
++ if (xdev == NULL)
++ return -ENODEV;
++
++ /* stuff we want to pass to /sbin/hotplug */
++ add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
++ "XENBUS_TYPE=%s", xdev->devicetype);
++
++ add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
++ "XENBUS_PATH=%s", xdev->nodename);
++
++ add_uevent_var(envp, num_envp, &i, buffer, buffer_size, &length,
++ "XENBUS_BASE_PATH=%s", xenbus_backend.root);
++
++ /* terminate, set to next free slot, shrink available space */
++ envp[i] = NULL;
++ envp = &envp[i];
++ num_envp -= i;
++ buffer = &buffer[length];
++ buffer_size -= length;
++
++ if (dev->driver) {
++ drv = to_xenbus_driver(dev->driver);
++ if (drv && drv->uevent)
++ return drv->uevent(xdev, envp, num_envp, buffer,
++ buffer_size);
++ }
++
++ return 0;
++}
++
++static void otherend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ struct xenbus_device *dev =
++ container_of(watch, struct xenbus_device, otherend_watch);
++ struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
++ enum xenbus_state state;
++
++ /* Protect us against watches firing on old details when the otherend
++ details change, say immediately after a resume. */
++ if (!dev->otherend ||
++ strncmp(dev->otherend, vec[XS_WATCH_PATH],
++ strlen(dev->otherend))) {
++ DPRINTK("Ignoring watch at %s", vec[XS_WATCH_PATH]);
++ return;
++ }
++
++ state = xenbus_read_driver_state(dev->otherend);
++
++ DPRINTK("state is %d, %s, %s",
++ state, dev->otherend_watch.node, vec[XS_WATCH_PATH]);
++ if (drv->otherend_changed)
++ drv->otherend_changed(dev, state);
++}
++
++
++static int talk_to_otherend(struct xenbus_device *dev)
++{
++ struct xenbus_driver *drv = to_xenbus_driver(dev->dev.driver);
++
++ free_otherend_watch(dev);
++ free_otherend_details(dev);
++
++ return drv->read_otherend_details(dev);
++}
++
++
++static int watch_otherend(struct xenbus_device *dev)
++{
++ return xenbus_watch_path2(dev, dev->otherend, "state",
++ &dev->otherend_watch, otherend_changed);
++}
++
++
++static int xenbus_dev_probe(struct device *_dev)
++{
++ struct xenbus_device *dev = to_xenbus_device(_dev);
++ struct xenbus_driver *drv = to_xenbus_driver(_dev->driver);
++ const struct xenbus_device_id *id;
++ int err;
++
++ DPRINTK("");
++
++ if (!drv->probe) {
++ err = -ENODEV;
++ goto fail;
++ }
++
++ id = match_device(drv->ids, dev);
++ if (!id) {
++ err = -ENODEV;
++ goto fail;
++ }
++
++ err = talk_to_otherend(dev);
++ if (err) {
++ printk(KERN_WARNING
++ "xenbus_probe: talk_to_otherend on %s failed.\n",
++ dev->nodename);
++ return err;
++ }
++
++ err = drv->probe(dev, id);
++ if (err)
++ goto fail;
++
++ err = watch_otherend(dev);
++ if (err) {
++ printk(KERN_WARNING
++ "xenbus_probe: watch_otherend on %s failed.\n",
++ dev->nodename);
++ return err;
++ }
++
++ return 0;
++fail:
++ xenbus_dev_error(dev, err, "xenbus_dev_probe on %s", dev->nodename);
++ xenbus_switch_state(dev, XenbusStateClosed);
++ return -ENODEV;
++}
++
++static int xenbus_dev_remove(struct device *_dev)
++{
++ struct xenbus_device *dev = to_xenbus_device(_dev);
++ struct xenbus_driver *drv = to_xenbus_driver(_dev->driver);
++
++ DPRINTK("");
++
++ free_otherend_watch(dev);
++ free_otherend_details(dev);
++
++ if (drv->remove)
++ drv->remove(dev);
++
++ xenbus_switch_state(dev, XenbusStateClosed);
++ return 0;
++}
++
++static int xenbus_register_driver_common(struct xenbus_driver *drv,
++ struct xen_bus_type *bus)
++{
++ int ret;
++
++ drv->driver.name = drv->name;
++ drv->driver.bus = &bus->bus;
++ drv->driver.owner = drv->owner;
++ drv->driver.probe = xenbus_dev_probe;
++ drv->driver.remove = xenbus_dev_remove;
++
++ mutex_lock(&xenwatch_mutex);
++ ret = driver_register(&drv->driver);
++ mutex_unlock(&xenwatch_mutex);
++ return ret;
++}
++
++int xenbus_register_frontend(struct xenbus_driver *drv)
++{
++ drv->read_otherend_details = read_backend_details;
++
++ return xenbus_register_driver_common(drv, &xenbus_frontend);
++}
++EXPORT_SYMBOL_GPL(xenbus_register_frontend);
++
++int xenbus_register_backend(struct xenbus_driver *drv)
++{
++ drv->read_otherend_details = read_frontend_details;
++
++ return xenbus_register_driver_common(drv, &xenbus_backend);
++}
++EXPORT_SYMBOL_GPL(xenbus_register_backend);
++
++void xenbus_unregister_driver(struct xenbus_driver *drv)
++{
++ driver_unregister(&drv->driver);
++}
++EXPORT_SYMBOL_GPL(xenbus_unregister_driver);
++
++struct xb_find_info
++{
++ struct xenbus_device *dev;
++ const char *nodename;
++};
++
++static int cmp_dev(struct device *dev, void *data)
++{
++ struct xenbus_device *xendev = to_xenbus_device(dev);
++ struct xb_find_info *info = data;
++
++ if (!strcmp(xendev->nodename, info->nodename)) {
++ info->dev = xendev;
++ get_device(dev);
++ return 1;
++ }
++ return 0;
++}
++
++struct xenbus_device *xenbus_device_find(const char *nodename,
++ struct bus_type *bus)
++{
++ struct xb_find_info info = { .dev = NULL, .nodename = nodename };
++
++ bus_for_each_dev(bus, NULL, &info, cmp_dev);
++ return info.dev;
++}
++
++static int cleanup_dev(struct device *dev, void *data)
++{
++ struct xenbus_device *xendev = to_xenbus_device(dev);
++ struct xb_find_info *info = data;
++ int len = strlen(info->nodename);
++
++ DPRINTK("%s", info->nodename);
++
++ /* Match the info->nodename path, or any subdirectory of that path. */
++ if (strncmp(xendev->nodename, info->nodename, len))
++ return 0;
++
++ /* If the node name is longer, ensure it really is a subdirectory. */
++ if ((strlen(xendev->nodename) > len) && (xendev->nodename[len] != '/'))
++ return 0;
++
++ info->dev = xendev;
++ get_device(dev);
++ return 1;
++}
++
++static void xenbus_cleanup_devices(const char *path, struct bus_type *bus)
++{
++ struct xb_find_info info = { .nodename = path };
++
++ do {
++ info.dev = NULL;
++ bus_for_each_dev(bus, NULL, &info, cleanup_dev);
++ if (info.dev) {
++ device_unregister(&info.dev->dev);
++ put_device(&info.dev->dev);
++ }
++ } while (info.dev);
++}
++
++static void xenbus_dev_release(struct device *dev)
++{
++ if (dev)
++ kfree(to_xenbus_device(dev));
++}
++
++/* Simplified asprintf. */
++char *kasprintf(const char *fmt, ...)
++{
++ va_list ap;
++ unsigned int len;
++ char *p, dummy[1];
++
++ va_start(ap, fmt);
++ /* FIXME: vsnprintf has a bug, NULL should work */
++ len = vsnprintf(dummy, 0, fmt, ap);
++ va_end(ap);
++
++ p = kmalloc(len + 1, GFP_KERNEL);
++ if (!p)
++ return NULL;
++ va_start(ap, fmt);
++ vsprintf(p, fmt, ap);
++ va_end(ap);
++ return p;
++}
++
++static ssize_t xendev_show_nodename(struct device *dev,
++ struct device_attribute *attr, char *buf)
++{
++ return sprintf(buf, "%s\n", to_xenbus_device(dev)->nodename);
++}
++DEVICE_ATTR(nodename, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_nodename, NULL);
++
++static ssize_t xendev_show_devtype(struct device *dev,
++ struct device_attribute *attr, char *buf)
++{
++ return sprintf(buf, "%s\n", to_xenbus_device(dev)->devicetype);
++}
++DEVICE_ATTR(devtype, S_IRUSR | S_IRGRP | S_IROTH, xendev_show_devtype, NULL);
++
++
++static int xenbus_probe_node(struct xen_bus_type *bus,
++ const char *type,
++ const char *nodename)
++{
++ int err;
++ struct xenbus_device *xendev;
++ size_t stringlen;
++ char *tmpstring;
++
++ enum xenbus_state state = xenbus_read_driver_state(nodename);
++
++ if (state != XenbusStateInitialising) {
++ /* Device is not new, so ignore it. This can happen if a
++ device is going away after switching to Closed. */
++ return 0;
++ }
++
++ stringlen = strlen(nodename) + 1 + strlen(type) + 1;
++ xendev = kzalloc(sizeof(*xendev) + stringlen, GFP_KERNEL);
++ if (!xendev)
++ return -ENOMEM;
++
++ /* Copy the strings into the extra space. */
++
++ tmpstring = (char *)(xendev + 1);
++ strcpy(tmpstring, nodename);
++ xendev->nodename = tmpstring;
++
++ tmpstring += strlen(tmpstring) + 1;
++ strcpy(tmpstring, type);
++ xendev->devicetype = tmpstring;
++
++ xendev->dev.parent = &bus->dev;
++ xendev->dev.bus = &bus->bus;
++ xendev->dev.release = xenbus_dev_release;
++
++ err = bus->get_bus_id(xendev->dev.bus_id, xendev->nodename);
++ if (err)
++ goto fail;
++
++ /* Register with generic device framework. */
++ err = device_register(&xendev->dev);
++ if (err)
++ goto fail;
++
++ device_create_file(&xendev->dev, &dev_attr_nodename);
++ device_create_file(&xendev->dev, &dev_attr_devtype);
++
++ return 0;
++fail:
++ kfree(xendev);
++ return err;
++}
++
++/* device/<typename>/<name> */
++static int xenbus_probe_frontend(const char *type, const char *name)
++{
++ char *nodename;
++ int err;
++
++ nodename = kasprintf("%s/%s/%s", xenbus_frontend.root, type, name);
++ if (!nodename)
++ return -ENOMEM;
++
++ DPRINTK("%s", nodename);
++
++ err = xenbus_probe_node(&xenbus_frontend, type, nodename);
++ kfree(nodename);
++ return err;
++}
++
++/* backend/<typename>/<frontend-uuid>/<name> */
++static int xenbus_probe_backend_unit(const char *dir,
++ const char *type,
++ const char *name)
++{
++ char *nodename;
++ int err;
++
++ nodename = kasprintf("%s/%s", dir, name);
++ if (!nodename)
++ return -ENOMEM;
++
++ DPRINTK("%s\n", nodename);
++
++ err = xenbus_probe_node(&xenbus_backend, type, nodename);
++ kfree(nodename);
++ return err;
++}
++
++/* backend/<typename>/<frontend-domid> */
++static int xenbus_probe_backend(const char *type, const char *domid)
++{
++ char *nodename;
++ int err = 0;
++ char **dir;
++ unsigned int i, dir_n = 0;
++
++ DPRINTK("");
++
++ nodename = kasprintf("%s/%s/%s", xenbus_backend.root, type, domid);
++ if (!nodename)
++ return -ENOMEM;
++
++ dir = xenbus_directory(XBT_NIL, nodename, "", &dir_n);
++ if (IS_ERR(dir)) {
++ kfree(nodename);
++ return PTR_ERR(dir);
++ }
++
++ for (i = 0; i < dir_n; i++) {
++ err = xenbus_probe_backend_unit(nodename, type, dir[i]);
++ if (err)
++ break;
++ }
++ kfree(dir);
++ kfree(nodename);
++ return err;
++}
++
++static int xenbus_probe_device_type(struct xen_bus_type *bus, const char *type)
++{
++ int err = 0;
++ char **dir;
++ unsigned int dir_n = 0;
++ int i;
++
++ dir = xenbus_directory(XBT_NIL, bus->root, type, &dir_n);
++ if (IS_ERR(dir))
++ return PTR_ERR(dir);
++
++ for (i = 0; i < dir_n; i++) {
++ err = bus->probe(type, dir[i]);
++ if (err)
++ break;
++ }
++ kfree(dir);
++ return err;
++}
++
++static int xenbus_probe_devices(struct xen_bus_type *bus)
++{
++ int err = 0;
++ char **dir;
++ unsigned int i, dir_n;
++
++ dir = xenbus_directory(XBT_NIL, bus->root, "", &dir_n);
++ if (IS_ERR(dir))
++ return PTR_ERR(dir);
++
++ for (i = 0; i < dir_n; i++) {
++ err = xenbus_probe_device_type(bus, dir[i]);
++ if (err)
++ break;
++ }
++ kfree(dir);
++ return err;
++}
++
++static unsigned int char_count(const char *str, char c)
++{
++ unsigned int i, ret = 0;
++
++ for (i = 0; str[i]; i++)
++ if (str[i] == c)
++ ret++;
++ return ret;
++}
++
++static int strsep_len(const char *str, char c, unsigned int len)
++{
++ unsigned int i;
++
++ for (i = 0; str[i]; i++)
++ if (str[i] == c) {
++ if (len == 0)
++ return i;
++ len--;
++ }
++ return (len == 0) ? i : -ERANGE;
++}
++
++static void dev_changed(const char *node, struct xen_bus_type *bus)
++{
++ int exists, rootlen;
++ struct xenbus_device *dev;
++ char type[BUS_ID_SIZE];
++ const char *p, *root;
++
++ if (char_count(node, '/') < 2)
++ return;
++
++ exists = xenbus_exists(XBT_NIL, node, "");
++ if (!exists) {
++ xenbus_cleanup_devices(node, &bus->bus);
++ return;
++ }
++
++ /* backend/<type>/... or device/<type>/... */
++ p = strchr(node, '/') + 1;
++ snprintf(type, BUS_ID_SIZE, "%.*s", (int)strcspn(p, "/"), p);
++ type[BUS_ID_SIZE-1] = '\0';
++
++ rootlen = strsep_len(node, '/', bus->levels);
++ if (rootlen < 0)
++ return;
++ root = kasprintf("%.*s", rootlen, node);
++ if (!root)
++ return;
++
++ dev = xenbus_device_find(root, &bus->bus);
++ if (!dev)
++ xenbus_probe_node(bus, type, root);
++ else
++ put_device(&dev->dev);
++
++ kfree(root);
++}
++
++static void frontend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ DPRINTK("");
++
++ dev_changed(vec[XS_WATCH_PATH], &xenbus_frontend);
++}
++
++static void backend_changed(struct xenbus_watch *watch,
++ const char **vec, unsigned int len)
++{
++ DPRINTK("");
++
++ dev_changed(vec[XS_WATCH_PATH], &xenbus_backend);
++}
++
++/* We watch for devices appearing and vanishing. */
++static struct xenbus_watch fe_watch = {
++ .node = "device",
++ .callback = frontend_changed,
++};
++
++static struct xenbus_watch be_watch = {
++ .node = "backend",
++ .callback = backend_changed,
++};
++
++static int suspend_dev(struct device *dev, void *data)
++{
++ int err = 0;
++ struct xenbus_driver *drv;
++ struct xenbus_device *xdev;
++
++ DPRINTK("");
++
++ if (dev->driver == NULL)
++ return 0;
++ drv = to_xenbus_driver(dev->driver);
++ xdev = container_of(dev, struct xenbus_device, dev);
++ if (drv->suspend)
++ err = drv->suspend(xdev);
++ if (err)
++ printk(KERN_WARNING
++ "xenbus: suspend %s failed: %i\n", dev->bus_id, err);
++ return 0;
++}
++
++static int resume_dev(struct device *dev, void *data)
++{
++ int err;
++ struct xenbus_driver *drv;
++ struct xenbus_device *xdev;
++
++ DPRINTK("");
++
++ if (dev->driver == NULL)
++ return 0;
++
++ drv = to_xenbus_driver(dev->driver);
++ xdev = container_of(dev, struct xenbus_device, dev);
++
++ err = talk_to_otherend(xdev);
++ if (err) {
++ printk(KERN_WARNING
++ "xenbus: resume (talk_to_otherend) %s failed: %i\n",
++ dev->bus_id, err);
++ return err;
++ }
++
++ xdev->state = XenbusStateInitialising;
++
++ if (drv->resume) {
++ err = drv->resume(xdev);
++ if (err) {
++ printk(KERN_WARNING
++ "xenbus: resume %s failed: %i\n",
++ dev->bus_id, err);
++ return err;
++ }
++ }
++
++ err = watch_otherend(xdev);
++ if (err) {
++ printk(KERN_WARNING
++ "xenbus_probe: resume (watch_otherend) %s failed: "
++ "%d.\n", dev->bus_id, err);
++ return err;
++ }
++
++ return 0;
++}
++
++void xenbus_suspend(void)
++{
++ DPRINTK("");
++
++ bus_for_each_dev(&xenbus_frontend.bus, NULL, NULL, suspend_dev);
++ bus_for_each_dev(&xenbus_backend.bus, NULL, NULL, suspend_dev);
++ xs_suspend();
++}
++EXPORT_SYMBOL_GPL(xenbus_suspend);
++
++void xenbus_resume(void)
++{
++ xb_init_comms();
++ xs_resume();
++ bus_for_each_dev(&xenbus_frontend.bus, NULL, NULL, resume_dev);
++ bus_for_each_dev(&xenbus_backend.bus, NULL, NULL, resume_dev);
++}
++EXPORT_SYMBOL_GPL(xenbus_resume);
++
++
++/* A flag to determine if xenstored is 'ready' (i.e. has started) */
++int xenstored_ready = 0;
++
++
++int register_xenstore_notifier(struct notifier_block *nb)
++{
++ int ret = 0;
++
++ if (xenstored_ready > 0)
++ ret = nb->notifier_call(nb, 0, NULL);
++ else
++ blocking_notifier_chain_register(&xenstore_notifier_list, nb);
++
++ return ret;
++}
++EXPORT_SYMBOL_GPL(register_xenstore_notifier);
++
++void unregister_xenstore_notifier(struct notifier_block *nb)
++{
++ blocking_notifier_chain_unregister(&xenstore_notifier_list, nb);
++}
++EXPORT_SYMBOL_GPL(unregister_xenstore_notifier);
++
++
++void xenbus_probe(void *unused)
++{
++ BUG_ON((xenstored_ready <= 0));
++
++ /* Enumerate devices in xenstore. */
++ xenbus_probe_devices(&xenbus_frontend);
++ xenbus_probe_devices(&xenbus_backend);
++
++ /* Watch for changes. */
++ register_xenbus_watch(&fe_watch);
++ register_xenbus_watch(&be_watch);
++
++ /* Notify others that xenstore is up */
++ blocking_notifier_call_chain(&xenstore_notifier_list, 0, NULL);
++}
++
++
++#ifdef CONFIG_PROC_FS
++static struct file_operations xsd_kva_fops;
++static struct proc_dir_entry *xsd_kva_intf;
++static struct proc_dir_entry *xsd_port_intf;
++
++static int xsd_kva_mmap(struct file *file, struct vm_area_struct *vma)
++{
++ size_t size = vma->vm_end - vma->vm_start;
++
++ if ((size > PAGE_SIZE) || (vma->vm_pgoff != 0))
++ return -EINVAL;
++
++ if (remap_pfn_range(vma, vma->vm_start,
++ mfn_to_pfn(xen_start_info->store_mfn),
++ size, vma->vm_page_prot))
++ return -EAGAIN;
++
++ return 0;
++}
++
++static int xsd_kva_read(char *page, char **start, off_t off,
++ int count, int *eof, void *data)
++{
++ int len;
++
++ len = sprintf(page, "0x%p", mfn_to_virt(xen_start_info->store_mfn));
++ *eof = 1;
++ return len;
++}
++
++static int xsd_port_read(char *page, char **start, off_t off,
++ int count, int *eof, void *data)
++{
++ int len;
++
++ len = sprintf(page, "%d", xen_start_info->store_evtchn);
++ *eof = 1;
++ return len;
++}
++#endif
++
++
++static int __init xenbus_probe_init(void)
++{
++ int err = 0, dom0;
++ unsigned long page = 0;
++
++ DPRINTK("");
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ /* Register ourselves with the kernel bus subsystem */
++ bus_register(&xenbus_frontend.bus);
++ bus_register(&xenbus_backend.bus);
++
++ /*
++ * Domain0 doesn't have a store_evtchn or store_mfn yet.
++ */
++ dom0 = (xen_start_info->store_evtchn == 0);
++
++ if (dom0) {
++ struct evtchn_alloc_unbound alloc_unbound;
++
++ /* Allocate page. */
++ page = get_zeroed_page(GFP_KERNEL);
++ if (!page)
++ return -ENOMEM;
++
++ xen_start_info->store_mfn =
++ pfn_to_mfn(virt_to_phys((void *)page) >>
++ PAGE_SHIFT);
++
++ /* Next allocate a local port which xenstored can bind to */
++ alloc_unbound.dom = DOMID_SELF;
++ alloc_unbound.remote_dom = 0;
++
++ err = HYPERVISOR_event_channel_op(EVTCHNOP_alloc_unbound,
++ &alloc_unbound);
++ if (err == -ENOSYS)
++ goto err;
++ BUG_ON(err);
++ xen_start_info->store_evtchn = alloc_unbound.port;
++
++#ifdef CONFIG_PROC_FS
++ /* And finally publish the above info in /proc/xen */
++ xsd_kva_intf = create_xen_proc_entry("xsd_kva", 0600);
++ if (xsd_kva_intf) {
++ memcpy(&xsd_kva_fops, xsd_kva_intf->proc_fops,
++ sizeof(xsd_kva_fops));
++ xsd_kva_fops.mmap = xsd_kva_mmap;
++ xsd_kva_intf->proc_fops = &xsd_kva_fops;
++ xsd_kva_intf->read_proc = xsd_kva_read;
++ }
++ xsd_port_intf = create_xen_proc_entry("xsd_port", 0400);
++ if (xsd_port_intf)
++ xsd_port_intf->read_proc = xsd_port_read;
++#endif
++ } else
++ xenstored_ready = 1;
++
++ /* Initialize the interface to xenstore. */
++ err = xs_init();
++ if (err) {
++ printk(KERN_WARNING
++ "XENBUS: Error initializing xenstore comms: %i\n", err);
++ goto err;
++ }
++
++ /* Register ourselves with the kernel device subsystem */
++ device_register(&xenbus_frontend.dev);
++ device_register(&xenbus_backend.dev);
++
++ if (!dom0)
++ xenbus_probe(NULL);
++
++ return 0;
++
++ err:
++ if (page)
++ free_page(page);
++
++ /*
++ * Do not unregister the xenbus front/backend buses here. The buses
++ * must exist because front/backend drivers will use them when they are
++ * registered.
++ */
++
++ return err;
++}
++
++postcore_initcall(xenbus_probe_init);
++
++
++static int is_disconnected_device(struct device *dev, void *data)
++{
++ struct xenbus_device *xendev = to_xenbus_device(dev);
++
++ /*
++ * A device with no driver will never connect. We care only about
++ * devices which should currently be in the process of connecting.
++ */
++ if (!dev->driver)
++ return 0;
++
++ return (xendev->state != XenbusStateConnected);
++}
++
++static int exists_disconnected_device(void)
++{
++ return bus_for_each_dev(&xenbus_frontend.bus, NULL, NULL,
++ is_disconnected_device);
++}
++
++static int print_device_status(struct device *dev, void *data)
++{
++ struct xenbus_device *xendev = to_xenbus_device(dev);
++
++ if (!dev->driver) {
++ /* Information only: is this too noisy? */
++ printk(KERN_INFO "XENBUS: Device with no driver: %s\n",
++ xendev->nodename);
++ } else if (xendev->state != XenbusStateConnected) {
++ printk(KERN_WARNING "XENBUS: Timeout connecting "
++ "to device: %s (state %d)\n",
++ xendev->nodename, xendev->state);
++ }
++
++ return 0;
++}
++
++/*
++ * On a 10 second timeout, wait for all devices currently configured. We need
++ * to do this to guarantee that the filesystems and / or network devices
++ * needed for boot are available, before we can allow the boot to proceed.
++ *
++ * This needs to be on a late_initcall, to happen after the frontend device
++ * drivers have been initialised, but before the root fs is mounted.
++ *
++ * A possible improvement here would be to have the tools add a per-device
++ * flag to the store entry, indicating whether it is needed at boot time.
++ * This would allow people who knew what they were doing to accelerate their
++ * boot slightly, but of course needs tools or manual intervention to set up
++ * those flags correctly.
++ */
++static int __init wait_for_devices(void)
++{
++ unsigned long timeout = jiffies + 10*HZ;
++
++ if (!is_running_on_xen())
++ return -ENODEV;
++
++ while (time_before(jiffies, timeout) && exists_disconnected_device())
++ schedule_timeout_interruptible(HZ/10);
++
++ bus_for_each_dev(&xenbus_frontend.bus, NULL, NULL,
++ print_device_status);
++
++ return 0;
++}
++
++late_initcall(wait_for_devices);
+diff -urNp linux-2.6/drivers/xen/xenbus/xenbus_xs.c new/drivers/xen/xenbus/xenbus_xs.c
+--- linux-2.6/drivers/xen/xenbus/xenbus_xs.c 1970-01-01 01:00:00.000000000 +0100
++++ new/drivers/xen/xenbus/xenbus_xs.c 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,846 @@
++/******************************************************************************
++ * xenbus_xs.c
++ *
++ * This is the kernel equivalent of the "xs" library. We don't need everything
++ * and we use xenbus_comms for communication.
++ *
++ * Copyright (C) 2005 Rusty Russell, IBM Corporation
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#include <linux/unistd.h>
++#include <linux/errno.h>
++#include <linux/types.h>
++#include <linux/uio.h>
++#include <linux/kernel.h>
++#include <linux/string.h>
++#include <linux/err.h>
++#include <linux/slab.h>
++#include <linux/fcntl.h>
++#include <linux/kthread.h>
++#include <linux/rwsem.h>
++#include <xen/xenbus.h>
++#include "xenbus_comms.h"
++
++/* xenbus_probe.c */
++extern char *kasprintf(const char *fmt, ...);
++
++struct xs_stored_msg {
++ struct list_head list;
++
++ struct xsd_sockmsg hdr;
++
++ union {
++ /* Queued replies. */
++ struct {
++ char *body;
++ } reply;
++
++ /* Queued watch events. */
++ struct {
++ struct xenbus_watch *handle;
++ char **vec;
++ unsigned int vec_size;
++ } watch;
++ } u;
++};
++
++struct xs_handle {
++ /* A list of replies. Currently only one will ever be outstanding. */
++ struct list_head reply_list;
++ spinlock_t reply_lock;
++ wait_queue_head_t reply_waitq;
++
++ /* One request at a time. */
++ struct mutex request_mutex;
++
++ /* Protect transactions against save/restore. */
++ struct rw_semaphore suspend_mutex;
++};
++
++static struct xs_handle xs_state;
++
++/* List of registered watches, and a lock to protect it. */
++static LIST_HEAD(watches);
++static DEFINE_SPINLOCK(watches_lock);
++
++/* List of pending watch callback events, and a lock to protect it. */
++static LIST_HEAD(watch_events);
++static DEFINE_SPINLOCK(watch_events_lock);
++
++/*
++ * Details of the xenwatch callback kernel thread. The thread waits on the
++ * watch_events_waitq for work to do (queued on watch_events list). When it
++ * wakes up it acquires the xenwatch_mutex before reading the list and
++ * carrying out work.
++ */
++static pid_t xenwatch_pid;
++/* static */ DEFINE_MUTEX(xenwatch_mutex);
++static DECLARE_WAIT_QUEUE_HEAD(watch_events_waitq);
++
++static int get_error(const char *errorstring)
++{
++ unsigned int i;
++
++ for (i = 0; strcmp(errorstring, xsd_errors[i].errstring) != 0; i++) {
++ if (i == ARRAY_SIZE(xsd_errors) - 1) {
++ printk(KERN_WARNING
++ "XENBUS xen store gave: unknown error %s",
++ errorstring);
++ return EINVAL;
++ }
++ }
++ return xsd_errors[i].errnum;
++}
++
++static void *read_reply(enum xsd_sockmsg_type *type, unsigned int *len)
++{
++ struct xs_stored_msg *msg;
++ char *body;
++
++ spin_lock(&xs_state.reply_lock);
++
++ while (list_empty(&xs_state.reply_list)) {
++ spin_unlock(&xs_state.reply_lock);
++ /* XXX FIXME: Avoid synchronous wait for response here. */
++ wait_event(xs_state.reply_waitq,
++ !list_empty(&xs_state.reply_list));
++ spin_lock(&xs_state.reply_lock);
++ }
++
++ msg = list_entry(xs_state.reply_list.next,
++ struct xs_stored_msg, list);
++ list_del(&msg->list);
++
++ spin_unlock(&xs_state.reply_lock);
++
++ *type = msg->hdr.type;
++ if (len)
++ *len = msg->hdr.len;
++ body = msg->u.reply.body;
++
++ kfree(msg);
++
++ return body;
++}
++
++/* Emergency write. */
++void xenbus_debug_write(const char *str, unsigned int count)
++{
++ struct xsd_sockmsg msg = { 0 };
++
++ msg.type = XS_DEBUG;
++ msg.len = sizeof("print") + count + 1;
++
++ mutex_lock(&xs_state.request_mutex);
++ xb_write(&msg, sizeof(msg));
++ xb_write("print", sizeof("print"));
++ xb_write(str, count);
++ xb_write("", 1);
++ mutex_unlock(&xs_state.request_mutex);
++}
++
++void *xenbus_dev_request_and_reply(struct xsd_sockmsg *msg)
++{
++ void *ret;
++ struct xsd_sockmsg req_msg = *msg;
++ int err;
++
++ if (req_msg.type == XS_TRANSACTION_START)
++ down_read(&xs_state.suspend_mutex);
++
++ mutex_lock(&xs_state.request_mutex);
++
++ err = xb_write(msg, sizeof(*msg) + msg->len);
++ if (err) {
++ msg->type = XS_ERROR;
++ ret = ERR_PTR(err);
++ } else
++ ret = read_reply(&msg->type, &msg->len);
++
++ mutex_unlock(&xs_state.request_mutex);
++
++ if ((req_msg.type == XS_TRANSACTION_END) ||
++ ((req_msg.type == XS_TRANSACTION_START) &&
++ (msg->type == XS_ERROR)))
++ up_read(&xs_state.suspend_mutex);
++
++ return ret;
++}
++
++/* Send message to xs, get kmalloc'ed reply. ERR_PTR() on error. */
++static void *xs_talkv(struct xenbus_transaction t,
++ enum xsd_sockmsg_type type,
++ const struct kvec *iovec,
++ unsigned int num_vecs,
++ unsigned int *len)
++{
++ struct xsd_sockmsg msg;
++ void *ret = NULL;
++ unsigned int i;
++ int err;
++
++ msg.tx_id = t.id;
++ msg.req_id = 0;
++ msg.type = type;
++ msg.len = 0;
++ for (i = 0; i < num_vecs; i++)
++ msg.len += iovec[i].iov_len;
++
++ mutex_lock(&xs_state.request_mutex);
++
++ err = xb_write(&msg, sizeof(msg));
++ if (err) {
++ mutex_unlock(&xs_state.request_mutex);
++ return ERR_PTR(err);
++ }
++
++ for (i = 0; i < num_vecs; i++) {
++ err = xb_write(iovec[i].iov_base, iovec[i].iov_len);;
++ if (err) {
++ mutex_unlock(&xs_state.request_mutex);
++ return ERR_PTR(err);
++ }
++ }
++
++ ret = read_reply(&msg.type, len);
++
++ mutex_unlock(&xs_state.request_mutex);
++
++ if (IS_ERR(ret))
++ return ret;
++
++ if (msg.type == XS_ERROR) {
++ err = get_error(ret);
++ kfree(ret);
++ return ERR_PTR(-err);
++ }
++
++ if (msg.type != type) {
++ if (printk_ratelimit())
++ printk(KERN_WARNING
++ "XENBUS unexpected type [%d], expected [%d]\n",
++ msg.type, type);
++ kfree(ret);
++ return ERR_PTR(-EINVAL);
++ }
++ return ret;
++}
++
++/* Simplified version of xs_talkv: single message. */
++static void *xs_single(struct xenbus_transaction t,
++ enum xsd_sockmsg_type type,
++ const char *string,
++ unsigned int *len)
++{
++ struct kvec iovec;
++
++ iovec.iov_base = (void *)string;
++ iovec.iov_len = strlen(string) + 1;
++ return xs_talkv(t, type, &iovec, 1, len);
++}
++
++/* Many commands only need an ack, don't care what it says. */
++static int xs_error(char *reply)
++{
++ if (IS_ERR(reply))
++ return PTR_ERR(reply);
++ kfree(reply);
++ return 0;
++}
++
++static unsigned int count_strings(const char *strings, unsigned int len)
++{
++ unsigned int num;
++ const char *p;
++
++ for (p = strings, num = 0; p < strings + len; p += strlen(p) + 1)
++ num++;
++
++ return num;
++}
++
++/* Return the path to dir with /name appended. Buffer must be kfree()'ed. */
++static char *join(const char *dir, const char *name)
++{
++ char *buffer;
++
++ if (strlen(name) == 0)
++ buffer = kasprintf("%s", dir);
++ else
++ buffer = kasprintf("%s/%s", dir, name);
++ return (!buffer) ? ERR_PTR(-ENOMEM) : buffer;
++}
++
++static char **split(char *strings, unsigned int len, unsigned int *num)
++{
++ char *p, **ret;
++
++ /* Count the strings. */
++ *num = count_strings(strings, len);
++
++ /* Transfer to one big alloc for easy freeing. */
++ ret = kmalloc(*num * sizeof(char *) + len, GFP_KERNEL);
++ if (!ret) {
++ kfree(strings);
++ return ERR_PTR(-ENOMEM);
++ }
++ memcpy(&ret[*num], strings, len);
++ kfree(strings);
++
++ strings = (char *)&ret[*num];
++ for (p = strings, *num = 0; p < strings + len; p += strlen(p) + 1)
++ ret[(*num)++] = p;
++
++ return ret;
++}
++
++char **xenbus_directory(struct xenbus_transaction t,
++ const char *dir, const char *node, unsigned int *num)
++{
++ char *strings, *path;
++ unsigned int len;
++
++ path = join(dir, node);
++ if (IS_ERR(path))
++ return (char **)path;
++
++ strings = xs_single(t, XS_DIRECTORY, path, &len);
++ kfree(path);
++ if (IS_ERR(strings))
++ return (char **)strings;
++
++ return split(strings, len, num);
++}
++EXPORT_SYMBOL_GPL(xenbus_directory);
++
++/* Check if a path exists. Return 1 if it does. */
++int xenbus_exists(struct xenbus_transaction t,
++ const char *dir, const char *node)
++{
++ char **d;
++ int dir_n;
++
++ d = xenbus_directory(t, dir, node, &dir_n);
++ if (IS_ERR(d))
++ return 0;
++ kfree(d);
++ return 1;
++}
++EXPORT_SYMBOL_GPL(xenbus_exists);
++
++/* Get the value of a single file.
++ * Returns a kmalloced value: call free() on it after use.
++ * len indicates length in bytes.
++ */
++void *xenbus_read(struct xenbus_transaction t,
++ const char *dir, const char *node, unsigned int *len)
++{
++ char *path;
++ void *ret;
++
++ path = join(dir, node);
++ if (IS_ERR(path))
++ return (void *)path;
++
++ ret = xs_single(t, XS_READ, path, len);
++ kfree(path);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_read);
++
++/* Write the value of a single file.
++ * Returns -err on failure.
++ */
++int xenbus_write(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *string)
++{
++ const char *path;
++ struct kvec iovec[2];
++ int ret;
++
++ path = join(dir, node);
++ if (IS_ERR(path))
++ return PTR_ERR(path);
++
++ iovec[0].iov_base = (void *)path;
++ iovec[0].iov_len = strlen(path) + 1;
++ iovec[1].iov_base = (void *)string;
++ iovec[1].iov_len = strlen(string);
++
++ ret = xs_error(xs_talkv(t, XS_WRITE, iovec, ARRAY_SIZE(iovec), NULL));
++ kfree(path);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_write);
++
++/* Create a new directory. */
++int xenbus_mkdir(struct xenbus_transaction t,
++ const char *dir, const char *node)
++{
++ char *path;
++ int ret;
++
++ path = join(dir, node);
++ if (IS_ERR(path))
++ return PTR_ERR(path);
++
++ ret = xs_error(xs_single(t, XS_MKDIR, path, NULL));
++ kfree(path);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_mkdir);
++
++/* Destroy a file or directory (directories must be empty). */
++int xenbus_rm(struct xenbus_transaction t, const char *dir, const char *node)
++{
++ char *path;
++ int ret;
++
++ path = join(dir, node);
++ if (IS_ERR(path))
++ return PTR_ERR(path);
++
++ ret = xs_error(xs_single(t, XS_RM, path, NULL));
++ kfree(path);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_rm);
++
++/* Start a transaction: changes by others will not be seen during this
++ * transaction, and changes will not be visible to others until end.
++ */
++int xenbus_transaction_start(struct xenbus_transaction *t)
++{
++ char *id_str;
++
++ down_read(&xs_state.suspend_mutex);
++
++ id_str = xs_single(XBT_NIL, XS_TRANSACTION_START, "", NULL);
++ if (IS_ERR(id_str)) {
++ up_read(&xs_state.suspend_mutex);
++ return PTR_ERR(id_str);
++ }
++
++ t->id = simple_strtoul(id_str, NULL, 0);
++ kfree(id_str);
++ return 0;
++}
++EXPORT_SYMBOL_GPL(xenbus_transaction_start);
++
++/* End a transaction.
++ * If abandon is true, transaction is discarded instead of committed.
++ */
++int xenbus_transaction_end(struct xenbus_transaction t, int abort)
++{
++ char abortstr[2];
++ int err;
++
++ if (abort)
++ strcpy(abortstr, "F");
++ else
++ strcpy(abortstr, "T");
++
++ err = xs_error(xs_single(t, XS_TRANSACTION_END, abortstr, NULL));
++
++ up_read(&xs_state.suspend_mutex);
++
++ return err;
++}
++EXPORT_SYMBOL_GPL(xenbus_transaction_end);
++
++/* Single read and scanf: returns -errno or num scanned. */
++int xenbus_scanf(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *fmt, ...)
++{
++ va_list ap;
++ int ret;
++ char *val;
++
++ val = xenbus_read(t, dir, node, NULL);
++ if (IS_ERR(val))
++ return PTR_ERR(val);
++
++ va_start(ap, fmt);
++ ret = vsscanf(val, fmt, ap);
++ va_end(ap);
++ kfree(val);
++ /* Distinctive errno. */
++ if (ret == 0)
++ return -ERANGE;
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_scanf);
++
++/* Single printf and write: returns -errno or 0. */
++int xenbus_printf(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *fmt, ...)
++{
++ va_list ap;
++ int ret;
++#define PRINTF_BUFFER_SIZE 4096
++ char *printf_buffer;
++
++ printf_buffer = kmalloc(PRINTF_BUFFER_SIZE, GFP_KERNEL);
++ if (printf_buffer == NULL)
++ return -ENOMEM;
++
++ va_start(ap, fmt);
++ ret = vsnprintf(printf_buffer, PRINTF_BUFFER_SIZE, fmt, ap);
++ va_end(ap);
++
++ BUG_ON(ret > PRINTF_BUFFER_SIZE-1);
++ ret = xenbus_write(t, dir, node, printf_buffer);
++
++ kfree(printf_buffer);
++
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_printf);
++
++/* Takes tuples of names, scanf-style args, and void **, NULL terminated. */
++int xenbus_gather(struct xenbus_transaction t, const char *dir, ...)
++{
++ va_list ap;
++ const char *name;
++ int ret = 0;
++
++ va_start(ap, dir);
++ while (ret == 0 && (name = va_arg(ap, char *)) != NULL) {
++ const char *fmt = va_arg(ap, char *);
++ void *result = va_arg(ap, void *);
++ char *p;
++
++ p = xenbus_read(t, dir, name, NULL);
++ if (IS_ERR(p)) {
++ ret = PTR_ERR(p);
++ break;
++ }
++ if (fmt) {
++ if (sscanf(p, fmt, result) == 0)
++ ret = -EINVAL;
++ kfree(p);
++ } else
++ *(char **)result = p;
++ }
++ va_end(ap);
++ return ret;
++}
++EXPORT_SYMBOL_GPL(xenbus_gather);
++
++static int xs_watch(const char *path, const char *token)
++{
++ struct kvec iov[2];
++
++ iov[0].iov_base = (void *)path;
++ iov[0].iov_len = strlen(path) + 1;
++ iov[1].iov_base = (void *)token;
++ iov[1].iov_len = strlen(token) + 1;
++
++ return xs_error(xs_talkv(XBT_NIL, XS_WATCH, iov,
++ ARRAY_SIZE(iov), NULL));
++}
++
++static int xs_unwatch(const char *path, const char *token)
++{
++ struct kvec iov[2];
++
++ iov[0].iov_base = (char *)path;
++ iov[0].iov_len = strlen(path) + 1;
++ iov[1].iov_base = (char *)token;
++ iov[1].iov_len = strlen(token) + 1;
++
++ return xs_error(xs_talkv(XBT_NIL, XS_UNWATCH, iov,
++ ARRAY_SIZE(iov), NULL));
++}
++
++static struct xenbus_watch *find_watch(const char *token)
++{
++ struct xenbus_watch *i, *cmp;
++
++ cmp = (void *)simple_strtoul(token, NULL, 16);
++
++ list_for_each_entry(i, &watches, list)
++ if (i == cmp)
++ return i;
++
++ return NULL;
++}
++
++/* Register callback to watch this node. */
++int register_xenbus_watch(struct xenbus_watch *watch)
++{
++ /* Pointer in ascii is the token. */
++ char token[sizeof(watch) * 2 + 1];
++ int err;
++
++ sprintf(token, "%lX", (long)watch);
++
++ down_read(&xs_state.suspend_mutex);
++
++ spin_lock(&watches_lock);
++ BUG_ON(find_watch(token));
++ list_add(&watch->list, &watches);
++ spin_unlock(&watches_lock);
++
++ err = xs_watch(watch->node, token);
++
++ /* Ignore errors due to multiple registration. */
++ if ((err != 0) && (err != -EEXIST)) {
++ spin_lock(&watches_lock);
++ list_del(&watch->list);
++ spin_unlock(&watches_lock);
++ }
++
++ up_read(&xs_state.suspend_mutex);
++
++ return err;
++}
++EXPORT_SYMBOL_GPL(register_xenbus_watch);
++
++void unregister_xenbus_watch(struct xenbus_watch *watch)
++{
++ struct xs_stored_msg *msg, *tmp;
++ char token[sizeof(watch) * 2 + 1];
++ int err;
++
++ sprintf(token, "%lX", (long)watch);
++
++ down_read(&xs_state.suspend_mutex);
++
++ spin_lock(&watches_lock);
++ BUG_ON(!find_watch(token));
++ list_del(&watch->list);
++ spin_unlock(&watches_lock);
++
++ err = xs_unwatch(watch->node, token);
++ if (err)
++ printk(KERN_WARNING
++ "XENBUS Failed to release watch %s: %i\n",
++ watch->node, err);
++
++ up_read(&xs_state.suspend_mutex);
++
++ /* Cancel pending watch events. */
++ spin_lock(&watch_events_lock);
++ list_for_each_entry_safe(msg, tmp, &watch_events, list) {
++ if (msg->u.watch.handle != watch)
++ continue;
++ list_del(&msg->list);
++ kfree(msg->u.watch.vec);
++ kfree(msg);
++ }
++ spin_unlock(&watch_events_lock);
++
++ /* Flush any currently-executing callback, unless we are it. :-) */
++ if (current->pid != xenwatch_pid) {
++ mutex_lock(&xenwatch_mutex);
++ mutex_unlock(&xenwatch_mutex);
++ }
++}
++EXPORT_SYMBOL_GPL(unregister_xenbus_watch);
++
++void xs_suspend(void)
++{
++ down_write(&xs_state.suspend_mutex);
++ mutex_lock(&xs_state.request_mutex);
++}
++
++void xs_resume(void)
++{
++ struct xenbus_watch *watch;
++ char token[sizeof(watch) * 2 + 1];
++
++ mutex_unlock(&xs_state.request_mutex);
++
++ /* No need for watches_lock: the suspend_mutex is sufficient. */
++ list_for_each_entry(watch, &watches, list) {
++ sprintf(token, "%lX", (long)watch);
++ xs_watch(watch->node, token);
++ }
++
++ up_write(&xs_state.suspend_mutex);
++}
++
++static int xenwatch_handle_callback(void *data)
++{
++ struct xs_stored_msg *msg = data;
++
++ msg->u.watch.handle->callback(msg->u.watch.handle,
++ (const char **)msg->u.watch.vec,
++ msg->u.watch.vec_size);
++
++ kfree(msg->u.watch.vec);
++ kfree(msg);
++
++ /* Kill this kthread if we were spawned just for this callback. */
++ if (current->pid != xenwatch_pid)
++ do_exit(0);
++
++ return 0;
++}
++
++static int xenwatch_thread(void *unused)
++{
++ struct list_head *ent;
++ struct xs_stored_msg *msg;
++
++ for (;;) {
++ wait_event_interruptible(watch_events_waitq,
++ !list_empty(&watch_events));
++
++ if (kthread_should_stop())
++ break;
++
++ mutex_lock(&xenwatch_mutex);
++
++ spin_lock(&watch_events_lock);
++ ent = watch_events.next;
++ if (ent != &watch_events)
++ list_del(ent);
++ spin_unlock(&watch_events_lock);
++
++ if (ent != &watch_events) {
++ msg = list_entry(ent, struct xs_stored_msg, list);
++ if (msg->u.watch.handle->flags & XBWF_new_thread)
++ kthread_run(xenwatch_handle_callback,
++ msg, "xenwatch_cb");
++ else
++ xenwatch_handle_callback(msg);
++ }
++
++ mutex_unlock(&xenwatch_mutex);
++ }
++
++ return 0;
++}
++
++static int process_msg(void)
++{
++ struct xs_stored_msg *msg;
++ char *body;
++ int err;
++
++ msg = kmalloc(sizeof(*msg), GFP_KERNEL);
++ if (msg == NULL)
++ return -ENOMEM;
++
++ err = xb_read(&msg->hdr, sizeof(msg->hdr));
++ if (err) {
++ kfree(msg);
++ return err;
++ }
++
++ body = kmalloc(msg->hdr.len + 1, GFP_KERNEL);
++ if (body == NULL) {
++ kfree(msg);
++ return -ENOMEM;
++ }
++
++ err = xb_read(body, msg->hdr.len);
++ if (err) {
++ kfree(body);
++ kfree(msg);
++ return err;
++ }
++ body[msg->hdr.len] = '\0';
++
++ if (msg->hdr.type == XS_WATCH_EVENT) {
++ msg->u.watch.vec = split(body, msg->hdr.len,
++ &msg->u.watch.vec_size);
++ if (IS_ERR(msg->u.watch.vec)) {
++ kfree(msg);
++ return PTR_ERR(msg->u.watch.vec);
++ }
++
++ spin_lock(&watches_lock);
++ msg->u.watch.handle = find_watch(
++ msg->u.watch.vec[XS_WATCH_TOKEN]);
++ if (msg->u.watch.handle != NULL) {
++ spin_lock(&watch_events_lock);
++ list_add_tail(&msg->list, &watch_events);
++ wake_up(&watch_events_waitq);
++ spin_unlock(&watch_events_lock);
++ } else {
++ kfree(msg->u.watch.vec);
++ kfree(msg);
++ }
++ spin_unlock(&watches_lock);
++ } else {
++ msg->u.reply.body = body;
++ spin_lock(&xs_state.reply_lock);
++ list_add_tail(&msg->list, &xs_state.reply_list);
++ spin_unlock(&xs_state.reply_lock);
++ wake_up(&xs_state.reply_waitq);
++ }
++
++ return 0;
++}
++
++static int xenbus_thread(void *unused)
++{
++ int err;
++
++ for (;;) {
++ err = process_msg();
++ if (err)
++ printk(KERN_WARNING "XENBUS error %d while reading "
++ "message\n", err);
++ if (kthread_should_stop())
++ break;
++ }
++
++ return 0;
++}
++
++int xs_init(void)
++{
++ int err;
++ struct task_struct *task;
++
++ INIT_LIST_HEAD(&xs_state.reply_list);
++ spin_lock_init(&xs_state.reply_lock);
++ init_waitqueue_head(&xs_state.reply_waitq);
++
++ mutex_init(&xs_state.request_mutex);
++ init_rwsem(&xs_state.suspend_mutex);
++
++ /* Initialize the shared memory rings to talk to xenstored */
++ err = xb_init_comms();
++ if (err)
++ return err;
++
++ task = kthread_run(xenwatch_thread, NULL, "xenwatch");
++ if (IS_ERR(task))
++ return PTR_ERR(task);
++ xenwatch_pid = task->pid;
++
++ task = kthread_run(xenbus_thread, NULL, "xenbus");
++ if (IS_ERR(task))
++ return PTR_ERR(task);
++
++ return 0;
++}
+diff -urNp linux-2.6/fs/Kconfig new/fs/Kconfig
+--- linux-2.6/fs/Kconfig 2006-07-03 14:15:01.000000000 +0200
++++ new/fs/Kconfig 2006-05-09 12:34:40.000000000 +0200
+@@ -842,6 +842,7 @@ config TMPFS
+ config HUGETLBFS
+ bool "HugeTLB file system support"
+ depends X86 || IA64 || PPC64 || SPARC64 || SUPERH || BROKEN
++ depends !XEN
+ help
+ hugetlbfs is a filesystem backing for HugeTLB pages, based on
+ ramfs. For architectures that support it, say Y here and read
+diff -urNp linux-2.6/include/asm-i386/a.out.h new/include/asm-i386/a.out.h
+--- linux-2.6/include/asm-i386/a.out.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/a.out.h 2006-05-09 12:35:16.000000000 +0200
+@@ -19,7 +19,7 @@ struct exec
+
+ #ifdef __KERNEL__
+
+-#define STACK_TOP TASK_SIZE
++#define STACK_TOP (TASK_SIZE - 3*PAGE_SIZE)
+
+ #endif
+
+diff -urNp linux-2.6/include/asm-i386/apic.h new/include/asm-i386/apic.h
+--- linux-2.6/include/asm-i386/apic.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/apic.h 2006-05-09 12:35:16.000000000 +0200
+@@ -132,10 +132,12 @@ extern unsigned int nmi_watchdog;
+
+ extern int disable_timer_pin_1;
+
++#ifndef CONFIG_XEN
+ void smp_send_timer_broadcast_ipi(struct pt_regs *regs);
+ void switch_APIC_timer_to_ipi(void *cpumask);
+ void switch_ipi_to_APIC_timer(void *cpumask);
+ #define ARCH_APICTIMER_STOPS_ON_C3 1
++#endif
+
+ extern int timer_over_8254;
+
+diff -urNp linux-2.6/include/asm-i386/elf.h new/include/asm-i386/elf.h
+--- linux-2.6/include/asm-i386/elf.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/elf.h 2006-05-09 12:35:16.000000000 +0200
+@@ -129,11 +129,16 @@ extern int dump_task_extended_fpu (struc
+ #define ELF_CORE_COPY_FPREGS(tsk, elf_fpregs) dump_task_fpu(tsk, elf_fpregs)
+ #define ELF_CORE_COPY_XFPREGS(tsk, elf_xfpregs) dump_task_extended_fpu(tsk, elf_xfpregs)
+
+-#define VSYSCALL_BASE (__fix_to_virt(FIX_VSYSCALL))
++#define VSYSCALL_BASE (PAGE_OFFSET - 2*PAGE_SIZE)
+ #define VSYSCALL_EHDR ((const struct elfhdr *) VSYSCALL_BASE)
+ #define VSYSCALL_ENTRY ((unsigned long) &__kernel_vsyscall)
+ extern void __kernel_vsyscall;
+
++#define ARCH_HAS_SETUP_ADDITIONAL_PAGES
++struct linux_binprm;
++extern int arch_setup_additional_pages(struct linux_binprm *bprm,
++ int executable_stack);
++
+ #define ARCH_DLINFO \
+ do { \
+ NEW_AUX_ENT(AT_SYSINFO, VSYSCALL_ENTRY); \
+diff -urNp linux-2.6/include/asm-i386/fixmap.h new/include/asm-i386/fixmap.h
+--- linux-2.6/include/asm-i386/fixmap.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/fixmap.h 2006-05-09 12:35:16.000000000 +0200
+@@ -20,7 +20,7 @@
+ * Leave one empty page between vmalloc'ed areas and
+ * the start of the fixmap.
+ */
+-#define __FIXADDR_TOP 0xfffff000
++extern unsigned long __FIXADDR_TOP;
+
+ #ifndef __ASSEMBLY__
+ #include <linux/kernel.h>
+@@ -52,7 +52,6 @@
+ */
+ enum fixed_addresses {
+ FIX_HOLE,
+- FIX_VSYSCALL,
+ #ifdef CONFIG_X86_LOCAL_APIC
+ FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
+ #endif
+@@ -95,6 +94,8 @@ enum fixed_addresses {
+ extern void __set_fixmap (enum fixed_addresses idx,
+ unsigned long phys, pgprot_t flags);
+
++extern void set_fixaddr_top(unsigned long top);
++
+ #define set_fixmap(idx, phys) \
+ __set_fixmap(idx, phys, PAGE_KERNEL)
+ /*
+@@ -116,14 +117,6 @@ extern void __set_fixmap (enum fixed_add
+ #define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
+ #define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
+
+-/*
+- * This is the range that is readable by user mode, and things
+- * acting like user mode such as get_user_pages.
+- */
+-#define FIXADDR_USER_START (__fix_to_virt(FIX_VSYSCALL))
+-#define FIXADDR_USER_END (FIXADDR_USER_START + PAGE_SIZE)
+-
+-
+ extern void __this_fixmap_does_not_exist(void);
+
+ /*
+diff -urNp linux-2.6/include/asm-i386/mach-default/mach_traps.h new/include/asm-i386/mach-default/mach_traps.h
+--- linux-2.6/include/asm-i386/mach-default/mach_traps.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/mach-default/mach_traps.h 2006-05-09 12:35:16.000000000 +0200
+@@ -15,6 +15,18 @@ static inline void clear_mem_error(unsig
+ outb(reason, 0x61);
+ }
+
++static inline void clear_io_check_error(unsigned char reason)
++{
++ unsigned long i;
++
++ reason = (reason & 0xf) | 8;
++ outb(reason, 0x61);
++ i = 2000;
++ while (--i) udelay(1000);
++ reason &= ~8;
++ outb(reason, 0x61);
++}
++
+ static inline unsigned char get_nmi_reason(void)
+ {
+ return inb(0x61);
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/agp.h new/include/asm-i386/mach-xen/asm/agp.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/agp.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/agp.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,37 @@
++#ifndef AGP_H
++#define AGP_H 1
++
++#include <asm/pgtable.h>
++#include <asm/cacheflush.h>
++#include <asm/system.h>
++
++/*
++ * Functions to keep the agpgart mappings coherent with the MMU.
++ * The GART gives the CPU a physical alias of pages in memory. The alias region is
++ * mapped uncacheable. Make sure there are no conflicting mappings
++ * with different cachability attributes for the same page. This avoids
++ * data corruption on some CPUs.
++ */
++
++int map_page_into_agp(struct page *page);
++int unmap_page_from_agp(struct page *page);
++#define flush_agp_mappings() global_flush_tlb()
++
++/* Could use CLFLUSH here if the cpu supports it. But then it would
++ need to be called for each cacheline of the whole page so it may not be
++ worth it. Would need a page for it. */
++#define flush_agp_cache() wbinvd()
++
++/* Convert a physical address to an address suitable for the GART. */
++#define phys_to_gart(x) phys_to_machine(x)
++#define gart_to_phys(x) machine_to_phys(x)
++
++/* GATT allocation. Returns/accepts GATT kernel virtual address. */
++#define alloc_gatt_pages(order) ({ \
++ char *_t; dma_addr_t _d; \
++ _t = dma_alloc_coherent(NULL,PAGE_SIZE<<(order),&_d,GFP_KERNEL); \
++ _t; })
++#define free_gatt_pages(table, order) \
++ dma_free_coherent(NULL,PAGE_SIZE<<(order),(table),virt_to_bus(table))
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/desc.h new/include/asm-i386/mach-xen/asm/desc.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/desc.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/desc.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,164 @@
++#ifndef __ARCH_DESC_H
++#define __ARCH_DESC_H
++
++#include <asm/ldt.h>
++#include <asm/segment.h>
++
++#define CPU_16BIT_STACK_SIZE 1024
++
++#ifndef __ASSEMBLY__
++
++#include <linux/preempt.h>
++#include <linux/smp.h>
++
++#include <asm/mmu.h>
++
++extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
++
++DECLARE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
++
++struct Xgt_desc_struct {
++ unsigned short size;
++ unsigned long address __attribute__((packed));
++ unsigned short pad;
++} __attribute__ ((packed));
++
++extern struct Xgt_desc_struct idt_descr;
++DECLARE_PER_CPU(struct Xgt_desc_struct, cpu_gdt_descr);
++
++
++static inline struct desc_struct *get_cpu_gdt_table(unsigned int cpu)
++{
++ return (struct desc_struct *)per_cpu(cpu_gdt_descr, cpu).address;
++}
++
++#define load_TR_desc() __asm__ __volatile__("ltr %w0"::"q" (GDT_ENTRY_TSS*8))
++#define load_LDT_desc() __asm__ __volatile__("lldt %w0"::"q" (GDT_ENTRY_LDT*8))
++
++#define load_gdt(dtr) __asm__ __volatile("lgdt %0"::"m" (*dtr))
++#define load_idt(dtr) __asm__ __volatile("lidt %0"::"m" (*dtr))
++#define load_tr(tr) __asm__ __volatile("ltr %0"::"mr" (tr))
++#define load_ldt(ldt) __asm__ __volatile("lldt %0"::"mr" (ldt))
++
++#define store_gdt(dtr) __asm__ ("sgdt %0":"=m" (*dtr))
++#define store_idt(dtr) __asm__ ("sidt %0":"=m" (*dtr))
++#define store_tr(tr) __asm__ ("str %0":"=mr" (tr))
++#define store_ldt(ldt) __asm__ ("sldt %0":"=mr" (ldt))
++
++/*
++ * This is the ldt that every process will get unless we need
++ * something other than this.
++ */
++extern struct desc_struct default_ldt[];
++extern void set_intr_gate(unsigned int irq, void * addr);
++
++#define _set_tssldt_desc(n,addr,limit,type) \
++__asm__ __volatile__ ("movw %w3,0(%2)\n\t" \
++ "movw %w1,2(%2)\n\t" \
++ "rorl $16,%1\n\t" \
++ "movb %b1,4(%2)\n\t" \
++ "movb %4,5(%2)\n\t" \
++ "movb $0,6(%2)\n\t" \
++ "movb %h1,7(%2)\n\t" \
++ "rorl $16,%1" \
++ : "=m"(*(n)) : "q" (addr), "r"(n), "ir"(limit), "i"(type))
++
++#ifndef CONFIG_X86_NO_TSS
++static inline void __set_tss_desc(unsigned int cpu, unsigned int entry, void *addr)
++{
++ _set_tssldt_desc(&get_cpu_gdt_table(cpu)[entry], (int)addr,
++ offsetof(struct tss_struct, __cacheline_filler) - 1, 0x89);
++}
++
++#define set_tss_desc(cpu,addr) __set_tss_desc(cpu, GDT_ENTRY_TSS, addr)
++#endif
++
++static inline void set_ldt_desc(unsigned int cpu, void *addr, unsigned int size)
++{
++ _set_tssldt_desc(&get_cpu_gdt_table(cpu)[GDT_ENTRY_LDT], (int)addr, ((size << 3)-1), 0x82);
++}
++
++#define LDT_entry_a(info) \
++ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
++
++#define LDT_entry_b(info) \
++ (((info)->base_addr & 0xff000000) | \
++ (((info)->base_addr & 0x00ff0000) >> 16) | \
++ ((info)->limit & 0xf0000) | \
++ (((info)->read_exec_only ^ 1) << 9) | \
++ ((info)->contents << 10) | \
++ (((info)->seg_not_present ^ 1) << 15) | \
++ ((info)->seg_32bit << 22) | \
++ ((info)->limit_in_pages << 23) | \
++ ((info)->useable << 20) | \
++ 0x7000)
++
++#define LDT_empty(info) (\
++ (info)->base_addr == 0 && \
++ (info)->limit == 0 && \
++ (info)->contents == 0 && \
++ (info)->read_exec_only == 1 && \
++ (info)->seg_32bit == 0 && \
++ (info)->limit_in_pages == 0 && \
++ (info)->seg_not_present == 1 && \
++ (info)->useable == 0 )
++
++extern int write_ldt_entry(void *ldt, int entry, __u32 entry_a, __u32 entry_b);
++
++#if TLS_SIZE != 24
++# error update this code.
++#endif
++
++static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
++{
++#define C(i) HYPERVISOR_update_descriptor(virt_to_machine(&get_cpu_gdt_table(cpu)[GDT_ENTRY_TLS_MIN + i]), *(u64 *)&t->tls_array[i])
++ C(0); C(1); C(2);
++#undef C
++}
++
++static inline void clear_LDT(void)
++{
++ int cpu = get_cpu();
++
++ /*
++ * NB. We load the default_ldt for lcall7/27 handling on demand, as
++ * it slows down context switching. Noone uses it anyway.
++ */
++ cpu = cpu; /* XXX avoid compiler warning */
++ xen_set_ldt(0UL, 0);
++ put_cpu();
++}
++
++/*
++ * load one particular LDT into the current CPU
++ */
++static inline void load_LDT_nolock(mm_context_t *pc, int cpu)
++{
++ void *segments = pc->ldt;
++ int count = pc->size;
++
++ if (likely(!count))
++ segments = NULL;
++
++ xen_set_ldt((unsigned long)segments, count);
++}
++
++static inline void load_LDT(mm_context_t *pc)
++{
++ int cpu = get_cpu();
++ load_LDT_nolock(pc, cpu);
++ put_cpu();
++}
++
++static inline unsigned long get_desc_base(unsigned long *desc)
++{
++ unsigned long base;
++ base = ((desc[0] >> 16) & 0x0000ffff) |
++ ((desc[1] << 16) & 0x00ff0000) |
++ (desc[1] & 0xff000000);
++ return base;
++}
++
++#endif /* !__ASSEMBLY__ */
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/dma-mapping.h new/include/asm-i386/mach-xen/asm/dma-mapping.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/dma-mapping.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/dma-mapping.h 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,154 @@
++#ifndef _ASM_I386_DMA_MAPPING_H
++#define _ASM_I386_DMA_MAPPING_H
++
++/*
++ * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for
++ * documentation.
++ */
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <asm/cache.h>
++#include <asm/io.h>
++#include <asm/scatterlist.h>
++#include <asm/swiotlb.h>
++
++static inline int
++address_needs_mapping(struct device *hwdev, dma_addr_t addr)
++{
++ dma_addr_t mask = 0xffffffff;
++ /* If the device has a mask, use it, otherwise default to 32 bits */
++ if (hwdev && hwdev->dma_mask)
++ mask = *hwdev->dma_mask;
++ return (addr & ~mask) != 0;
++}
++
++static inline int
++range_straddles_page_boundary(void *p, size_t size)
++{
++ extern unsigned long *contiguous_bitmap;
++ return (((((unsigned long)p & ~PAGE_MASK) + size) > PAGE_SIZE) &&
++ !test_bit(__pa(p) >> PAGE_SHIFT, contiguous_bitmap));
++}
++
++#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
++#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
++
++void *dma_alloc_coherent(struct device *dev, size_t size,
++ dma_addr_t *dma_handle, gfp_t flag);
++
++void dma_free_coherent(struct device *dev, size_t size,
++ void *vaddr, dma_addr_t dma_handle);
++
++extern dma_addr_t
++dma_map_single(struct device *dev, void *ptr, size_t size,
++ enum dma_data_direction direction);
++
++extern void
++dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
++ enum dma_data_direction direction);
++
++extern int dma_map_sg(struct device *hwdev, struct scatterlist *sg,
++ int nents, enum dma_data_direction direction);
++extern void dma_unmap_sg(struct device *hwdev, struct scatterlist *sg,
++ int nents, enum dma_data_direction direction);
++
++extern dma_addr_t
++dma_map_page(struct device *dev, struct page *page, unsigned long offset,
++ size_t size, enum dma_data_direction direction);
++
++extern void
++dma_unmap_page(struct device *dev, dma_addr_t dma_address, size_t size,
++ enum dma_data_direction direction);
++
++extern void
++dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle, size_t size,
++ enum dma_data_direction direction);
++
++extern void
++dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle, size_t size,
++ enum dma_data_direction direction);
++
++static inline void
++dma_sync_single_range_for_cpu(struct device *dev, dma_addr_t dma_handle,
++ unsigned long offset, size_t size,
++ enum dma_data_direction direction)
++{
++ dma_sync_single_for_cpu(dev, dma_handle+offset, size, direction);
++}
++
++static inline void
++dma_sync_single_range_for_device(struct device *dev, dma_addr_t dma_handle,
++ unsigned long offset, size_t size,
++ enum dma_data_direction direction)
++{
++ dma_sync_single_for_device(dev, dma_handle+offset, size, direction);
++}
++
++static inline void
++dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_sg_for_cpu(dev,sg,nelems,direction);
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_sg_for_device(dev,sg,nelems,direction);
++ flush_write_buffers();
++}
++
++extern int
++dma_mapping_error(dma_addr_t dma_addr);
++
++extern int
++dma_supported(struct device *dev, u64 mask);
++
++static inline int
++dma_set_mask(struct device *dev, u64 mask)
++{
++ if(!dev->dma_mask || !dma_supported(dev, mask))
++ return -EIO;
++
++ *dev->dma_mask = mask;
++
++ return 0;
++}
++
++#ifdef __i386__
++static inline int
++dma_get_cache_alignment(void)
++{
++ /* no easy way to get cache size on all x86, so return the
++ * maximum possible, to be safe */
++ return (1 << INTERNODE_CACHE_SHIFT);
++}
++#endif
++
++#define dma_is_consistent(d) (1)
++
++static inline void
++dma_cache_sync(void *vaddr, size_t size,
++ enum dma_data_direction direction)
++{
++ flush_write_buffers();
++}
++
++#define ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
++extern int
++dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
++ dma_addr_t device_addr, size_t size, int flags);
++
++extern void
++dma_release_declared_memory(struct device *dev);
++
++extern void *
++dma_mark_declared_memory_occupied(struct device *dev,
++ dma_addr_t device_addr, size_t size);
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/fixmap.h new/include/asm-i386/mach-xen/asm/fixmap.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/fixmap.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/fixmap.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,156 @@
++/*
++ * fixmap.h: compile-time virtual memory allocation
++ *
++ * This file is subject to the terms and conditions of the GNU General Public
++ * License. See the file "COPYING" in the main directory of this archive
++ * for more details.
++ *
++ * Copyright (C) 1998 Ingo Molnar
++ *
++ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999
++ */
++
++#ifndef _ASM_FIXMAP_H
++#define _ASM_FIXMAP_H
++
++#include <linux/config.h>
++
++/* used by vmalloc.c, vsyscall.lds.S.
++ *
++ * Leave one empty page between vmalloc'ed areas and
++ * the start of the fixmap.
++ */
++extern unsigned long __FIXADDR_TOP;
++
++#ifndef __ASSEMBLY__
++#include <linux/kernel.h>
++#include <asm/acpi.h>
++#include <asm/apicdef.h>
++#include <asm/page.h>
++#include <xen/gnttab.h>
++#ifdef CONFIG_HIGHMEM
++#include <linux/threads.h>
++#include <asm/kmap_types.h>
++#endif
++
++/*
++ * Here we define all the compile-time 'special' virtual
++ * addresses. The point is to have a constant address at
++ * compile time, but to set the physical address only
++ * in the boot process. We allocate these special addresses
++ * from the end of virtual memory (0xfffff000) backwards.
++ * Also this lets us do fail-safe vmalloc(), we
++ * can guarantee that these special addresses and
++ * vmalloc()-ed addresses never overlap.
++ *
++ * these 'compile-time allocated' memory buffers are
++ * fixed-size 4k pages. (or larger if used with an increment
++ * highger than 1) use fixmap_set(idx,phys) to associate
++ * physical memory with fixmap indices.
++ *
++ * TLB entries of such buffers will not be flushed across
++ * task switches.
++ */
++enum fixed_addresses {
++ FIX_HOLE,
++#ifdef CONFIG_X86_LOCAL_APIC
++ FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
++#endif
++#ifdef CONFIG_X86_IO_APIC
++ FIX_IO_APIC_BASE_0,
++ FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
++#endif
++#ifdef CONFIG_X86_VISWS_APIC
++ FIX_CO_CPU, /* Cobalt timer */
++ FIX_CO_APIC, /* Cobalt APIC Redirection Table */
++ FIX_LI_PCIA, /* Lithium PCI Bridge A */
++ FIX_LI_PCIB, /* Lithium PCI Bridge B */
++#endif
++#ifdef CONFIG_X86_F00F_BUG
++ FIX_F00F_IDT, /* Virtual mapping for IDT */
++#endif
++#ifdef CONFIG_X86_CYCLONE_TIMER
++ FIX_CYCLONE_TIMER, /*cyclone timer register*/
++#endif
++#ifdef CONFIG_HIGHMEM
++ FIX_KMAP_BEGIN, /* reserved pte's for temporary kernel mappings */
++ FIX_KMAP_END = FIX_KMAP_BEGIN+(KM_TYPE_NR*NR_CPUS)-1,
++#endif
++#ifdef CONFIG_ACPI
++ FIX_ACPI_BEGIN,
++ FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
++#endif
++#ifdef CONFIG_PCI_MMCONFIG
++ FIX_PCIE_MCFG,
++#endif
++ FIX_SHARED_INFO,
++#define NR_FIX_ISAMAPS 256
++ FIX_ISAMAP_END,
++ FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
++ __end_of_permanent_fixed_addresses,
++ /* temporary boot-time mappings, used before ioremap() is functional */
++#define NR_FIX_BTMAPS 16
++ FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
++ FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS - 1,
++ FIX_WP_TEST,
++ __end_of_fixed_addresses
++};
++
++extern void __set_fixmap(enum fixed_addresses idx,
++ maddr_t phys, pgprot_t flags);
++
++extern void set_fixaddr_top(unsigned long top);
++
++#define set_fixmap(idx, phys) \
++ __set_fixmap(idx, phys, PAGE_KERNEL)
++/*
++ * Some hardware wants to get fixmapped without caching.
++ */
++#define set_fixmap_nocache(idx, phys) \
++ __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
++
++#define clear_fixmap(idx) \
++ __set_fixmap(idx, 0, __pgprot(0))
++
++#define FIXADDR_TOP ((unsigned long)__FIXADDR_TOP)
++
++#define __FIXADDR_SIZE (__end_of_permanent_fixed_addresses << PAGE_SHIFT)
++#define __FIXADDR_BOOT_SIZE (__end_of_fixed_addresses << PAGE_SHIFT)
++#define FIXADDR_START (FIXADDR_TOP - __FIXADDR_SIZE)
++#define FIXADDR_BOOT_START (FIXADDR_TOP - __FIXADDR_BOOT_SIZE)
++
++#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
++#define __virt_to_fix(x) ((FIXADDR_TOP - ((x)&PAGE_MASK)) >> PAGE_SHIFT)
++
++extern void __this_fixmap_does_not_exist(void);
++
++/*
++ * 'index to address' translation. If anyone tries to use the idx
++ * directly without tranlation, we catch the bug with a NULL-deference
++ * kernel oops. Illegal ranges of incoming indices are caught too.
++ */
++static __always_inline unsigned long fix_to_virt(const unsigned int idx)
++{
++ /*
++ * this branch gets completely eliminated after inlining,
++ * except when someone tries to use fixaddr indices in an
++ * illegal way. (such as mixing up address types or using
++ * out-of-range indices).
++ *
++ * If it doesn't get removed, the linker will complain
++ * loudly with a reasonably clear error message..
++ */
++ if (idx >= __end_of_fixed_addresses)
++ __this_fixmap_does_not_exist();
++
++ return __fix_to_virt(idx);
++}
++
++static inline unsigned long virt_to_fix(const unsigned long vaddr)
++{
++ BUG_ON(vaddr >= FIXADDR_TOP || vaddr < FIXADDR_START);
++ return __virt_to_fix(vaddr);
++}
++
++#endif /* !__ASSEMBLY__ */
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/floppy.h new/include/asm-i386/mach-xen/asm/floppy.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/floppy.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/floppy.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,147 @@
++/*
++ * Architecture specific parts of the Floppy driver
++ *
++ * This file is subject to the terms and conditions of the GNU General Public
++ * License. See the file "COPYING" in the main directory of this archive
++ * for more details.
++ *
++ * Copyright (C) 1995
++ *
++ * Modifications for Xen are Copyright (c) 2004, Keir Fraser.
++ */
++#ifndef __ASM_XEN_I386_FLOPPY_H
++#define __ASM_XEN_I386_FLOPPY_H
++
++#include <linux/vmalloc.h>
++
++/* XEN: Hit DMA paths on the head. This trick from asm-m68k/floppy.h. */
++#include <asm/dma.h>
++#undef MAX_DMA_ADDRESS
++#define MAX_DMA_ADDRESS 0
++#define CROSS_64KB(a,s) (0)
++
++#define fd_inb(port) inb_p(port)
++#define fd_outb(value,port) outb_p(value,port)
++
++#define fd_request_dma() (0)
++#define fd_free_dma() ((void)0)
++#define fd_enable_irq() enable_irq(FLOPPY_IRQ)
++#define fd_disable_irq() disable_irq(FLOPPY_IRQ)
++#define fd_free_irq() free_irq(FLOPPY_IRQ, NULL)
++#define fd_get_dma_residue() (virtual_dma_count + virtual_dma_residue)
++#define fd_dma_setup(addr, size, mode, io) vdma_dma_setup(addr, size, mode, io)
++/*
++ * Do not use vmalloc/vfree: floppy_release_irq_and_dma() gets called from
++ * softirq context via motor_off_callback. A generic bug we happen to trigger.
++ */
++#define fd_dma_mem_alloc(size) __get_free_pages(GFP_KERNEL, get_order(size))
++#define fd_dma_mem_free(addr, size) free_pages(addr, get_order(size))
++
++static int virtual_dma_count;
++static int virtual_dma_residue;
++static char *virtual_dma_addr;
++static int virtual_dma_mode;
++static int doing_pdma;
++
++static irqreturn_t floppy_hardint(int irq, void *dev_id, struct pt_regs * regs)
++{
++ register unsigned char st;
++ register int lcount;
++ register char *lptr;
++
++ if (!doing_pdma)
++ return floppy_interrupt(irq, dev_id, regs);
++
++ st = 1;
++ for(lcount=virtual_dma_count, lptr=virtual_dma_addr;
++ lcount; lcount--, lptr++) {
++ st=inb(virtual_dma_port+4) & 0xa0 ;
++ if(st != 0xa0)
++ break;
++ if(virtual_dma_mode)
++ outb_p(*lptr, virtual_dma_port+5);
++ else
++ *lptr = inb_p(virtual_dma_port+5);
++ }
++ virtual_dma_count = lcount;
++ virtual_dma_addr = lptr;
++ st = inb(virtual_dma_port+4);
++
++ if(st == 0x20)
++ return IRQ_HANDLED;
++ if(!(st & 0x20)) {
++ virtual_dma_residue += virtual_dma_count;
++ virtual_dma_count=0;
++ doing_pdma = 0;
++ floppy_interrupt(irq, dev_id, regs);
++ return IRQ_HANDLED;
++ }
++ return IRQ_HANDLED;
++}
++
++static void fd_disable_dma(void)
++{
++ doing_pdma = 0;
++ virtual_dma_residue += virtual_dma_count;
++ virtual_dma_count=0;
++}
++
++static int fd_request_irq(void)
++{
++ return request_irq(FLOPPY_IRQ, floppy_hardint,SA_INTERRUPT,
++ "floppy", NULL);
++}
++
++static int vdma_dma_setup(char *addr, unsigned long size, int mode, int io)
++{
++ doing_pdma = 1;
++ virtual_dma_port = io;
++ virtual_dma_mode = (mode == DMA_MODE_WRITE);
++ virtual_dma_addr = addr;
++ virtual_dma_count = size;
++ virtual_dma_residue = 0;
++ return 0;
++}
++
++/* XEN: This trick to force 'virtual DMA' is from include/asm-m68k/floppy.h. */
++#define FDC1 xen_floppy_init()
++static int FDC2 = -1;
++
++static int xen_floppy_init(void)
++{
++ use_virtual_dma = 1;
++ can_use_virtual_dma = 1;
++ return 0x3f0;
++}
++
++/*
++ * Floppy types are stored in the rtc's CMOS RAM and so rtc_lock
++ * is needed to prevent corrupted CMOS RAM in case "insmod floppy"
++ * coincides with another rtc CMOS user. Paul G.
++ */
++#define FLOPPY0_TYPE ({ \
++ unsigned long flags; \
++ unsigned char val; \
++ spin_lock_irqsave(&rtc_lock, flags); \
++ val = (CMOS_READ(0x10) >> 4) & 15; \
++ spin_unlock_irqrestore(&rtc_lock, flags); \
++ val; \
++})
++
++#define FLOPPY1_TYPE ({ \
++ unsigned long flags; \
++ unsigned char val; \
++ spin_lock_irqsave(&rtc_lock, flags); \
++ val = CMOS_READ(0x10) & 15; \
++ spin_unlock_irqrestore(&rtc_lock, flags); \
++ val; \
++})
++
++#define N_FDC 2
++#define N_DRIVE 8
++
++#define FLOPPY_MOTOR_MASK 0xf0
++
++#define EXTRA_FLOPPY_PARAMS
++
++#endif /* __ASM_XEN_I386_FLOPPY_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/highmem.h new/include/asm-i386/mach-xen/asm/highmem.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/highmem.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/highmem.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,81 @@
++/*
++ * highmem.h: virtual kernel memory mappings for high memory
++ *
++ * Used in CONFIG_HIGHMEM systems for memory pages which
++ * are not addressable by direct kernel virtual addresses.
++ *
++ * Copyright (C) 1999 Gerhard Wichert, Siemens AG
++ * Gerhard.Wichert@pdb.siemens.de
++ *
++ *
++ * Redesigned the x86 32-bit VM architecture to deal with
++ * up to 16 Terabyte physical memory. With current x86 CPUs
++ * we now support up to 64 Gigabytes physical RAM.
++ *
++ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
++ */
++
++#ifndef _ASM_HIGHMEM_H
++#define _ASM_HIGHMEM_H
++
++#ifdef __KERNEL__
++
++#include <linux/config.h>
++#include <linux/interrupt.h>
++#include <linux/threads.h>
++#include <asm/kmap_types.h>
++#include <asm/tlbflush.h>
++
++/* declarations for highmem.c */
++extern unsigned long highstart_pfn, highend_pfn;
++
++extern pte_t *kmap_pte;
++extern pgprot_t kmap_prot;
++extern pte_t *pkmap_page_table;
++
++/*
++ * Right now we initialize only a single pte table. It can be extended
++ * easily, subsequent pte tables have to be allocated in one physical
++ * chunk of RAM.
++ */
++#ifdef CONFIG_X86_PAE
++#define LAST_PKMAP 512
++#else
++#define LAST_PKMAP 1024
++#endif
++/*
++ * Ordering is:
++ *
++ * FIXADDR_TOP
++ * fixed_addresses
++ * FIXADDR_START
++ * temp fixed addresses
++ * FIXADDR_BOOT_START
++ * Persistent kmap area
++ * PKMAP_BASE
++ * VMALLOC_END
++ * Vmalloc area
++ * VMALLOC_START
++ * high_memory
++ */
++#define PKMAP_BASE ( (FIXADDR_BOOT_START - PAGE_SIZE*(LAST_PKMAP + 1)) & PMD_MASK )
++#define LAST_PKMAP_MASK (LAST_PKMAP-1)
++#define PKMAP_NR(virt) ((virt-PKMAP_BASE) >> PAGE_SHIFT)
++#define PKMAP_ADDR(nr) (PKMAP_BASE + ((nr) << PAGE_SHIFT))
++
++extern void * FASTCALL(kmap_high(struct page *page));
++extern void FASTCALL(kunmap_high(struct page *page));
++
++void *kmap(struct page *page);
++void kunmap(struct page *page);
++void *kmap_atomic(struct page *page, enum km_type type);
++void *kmap_atomic_pte(struct page *page, enum km_type type);
++void kunmap_atomic(void *kvaddr, enum km_type type);
++void *kmap_atomic_pfn(unsigned long pfn, enum km_type type);
++struct page *kmap_atomic_to_page(void *ptr);
++
++#define flush_cache_kmaps() do { } while (0)
++
++#endif /* __KERNEL__ */
++
++#endif /* _ASM_HIGHMEM_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/hw_irq.h new/include/asm-i386/mach-xen/asm/hw_irq.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/hw_irq.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/hw_irq.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,77 @@
++#ifndef _ASM_HW_IRQ_H
++#define _ASM_HW_IRQ_H
++
++/*
++ * linux/include/asm/hw_irq.h
++ *
++ * (C) 1992, 1993 Linus Torvalds, (C) 1997 Ingo Molnar
++ *
++ * moved some of the old arch/i386/kernel/irq.h to here. VY
++ *
++ * IRQ/IPI changes taken from work by Thomas Radke
++ * <tomsoft@informatik.tu-chemnitz.de>
++ */
++
++#include <linux/config.h>
++#include <linux/profile.h>
++#include <asm/atomic.h>
++#include <asm/irq.h>
++#include <asm/sections.h>
++
++struct hw_interrupt_type;
++
++/*
++ * Various low-level irq details needed by irq.c, process.c,
++ * time.c, io_apic.c and smp.c
++ *
++ * Interrupt entry/exit code at both C and assembly level
++ */
++
++extern u8 irq_vector[NR_IRQ_VECTORS];
++#define IO_APIC_VECTOR(irq) (irq_vector[irq])
++#define AUTO_ASSIGN -1
++
++extern void (*interrupt[NR_IRQS])(void);
++
++#ifdef CONFIG_SMP
++fastcall void reschedule_interrupt(void);
++fastcall void invalidate_interrupt(void);
++fastcall void call_function_interrupt(void);
++#endif
++
++#ifdef CONFIG_X86_LOCAL_APIC
++fastcall void apic_timer_interrupt(void);
++fastcall void error_interrupt(void);
++fastcall void spurious_interrupt(void);
++fastcall void thermal_interrupt(struct pt_regs *);
++#define platform_legacy_irq(irq) ((irq) < 16)
++#endif
++
++void disable_8259A_irq(unsigned int irq);
++void enable_8259A_irq(unsigned int irq);
++int i8259A_irq_pending(unsigned int irq);
++void make_8259A_irq(unsigned int irq);
++void init_8259A(int aeoi);
++void FASTCALL(send_IPI_self(int vector));
++void init_VISWS_APIC_irqs(void);
++void setup_IO_APIC(void);
++void disable_IO_APIC(void);
++void print_IO_APIC(void);
++int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn);
++void send_IPI(int dest, int vector);
++void setup_ioapic_dest(void);
++
++extern unsigned long io_apic_irqs;
++
++extern atomic_t irq_err_count;
++extern atomic_t irq_mis_count;
++
++#define IO_APIC_IRQ(x) (((x) >= 16) || ((1<<(x)) & io_apic_irqs))
++
++extern void resend_irq_on_evtchn(struct hw_interrupt_type *h, unsigned int i);
++static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i)
++{
++ resend_irq_on_evtchn(h, i);
++}
++
++#endif /* _ASM_HW_IRQ_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/hypercall.h new/include/asm-i386/mach-xen/asm/hypercall.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/hypercall.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/hypercall.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,372 @@
++/******************************************************************************
++ * hypercall.h
++ *
++ * Linux-specific hypervisor handling.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __HYPERCALL_H__
++#define __HYPERCALL_H__
++
++#include <linux/string.h> /* memcpy() */
++
++#ifndef __HYPERVISOR_H__
++# error "please don't include this file directly"
++#endif
++
++#define __STR(x) #x
++#define STR(x) __STR(x)
++
++#define _hypercall0(type, name) \
++({ \
++ long __res; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res) \
++ : \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall1(type, name, a1) \
++({ \
++ long __res, __ign1; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=b" (__ign1) \
++ : "1" ((long)(a1)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall2(type, name, a1, a2) \
++({ \
++ long __res, __ign1, __ign2; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=b" (__ign1), "=c" (__ign2) \
++ : "1" ((long)(a1)), "2" ((long)(a2)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall3(type, name, a1, a2, a3) \
++({ \
++ long __res, __ign1, __ign2, __ign3; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
++ "=d" (__ign3) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall4(type, name, a1, a2, a3, a4) \
++({ \
++ long __res, __ign1, __ign2, __ign3, __ign4; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
++ "=d" (__ign3), "=S" (__ign4) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)), "4" ((long)(a4)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall5(type, name, a1, a2, a3, a4, a5) \
++({ \
++ long __res, __ign1, __ign2, __ign3, __ign4, __ign5; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=b" (__ign1), "=c" (__ign2), \
++ "=d" (__ign3), "=S" (__ign4), "=D" (__ign5) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)), "4" ((long)(a4)), \
++ "5" ((long)(a5)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++static inline int
++HYPERVISOR_set_trap_table(
++ trap_info_t *table)
++{
++ return _hypercall1(int, set_trap_table, table);
++}
++
++static inline int
++HYPERVISOR_mmu_update(
++ mmu_update_t *req, int count, int *success_count, domid_t domid)
++{
++ return _hypercall4(int, mmu_update, req, count, success_count, domid);
++}
++
++static inline int
++HYPERVISOR_mmuext_op(
++ struct mmuext_op *op, int count, int *success_count, domid_t domid)
++{
++ return _hypercall4(int, mmuext_op, op, count, success_count, domid);
++}
++
++static inline int
++HYPERVISOR_set_gdt(
++ unsigned long *frame_list, int entries)
++{
++ return _hypercall2(int, set_gdt, frame_list, entries);
++}
++
++static inline int
++HYPERVISOR_stack_switch(
++ unsigned long ss, unsigned long esp)
++{
++ return _hypercall2(int, stack_switch, ss, esp);
++}
++
++static inline int
++HYPERVISOR_set_callbacks(
++ unsigned long event_selector, unsigned long event_address,
++ unsigned long failsafe_selector, unsigned long failsafe_address)
++{
++ return _hypercall4(int, set_callbacks,
++ event_selector, event_address,
++ failsafe_selector, failsafe_address);
++}
++
++static inline int
++HYPERVISOR_fpu_taskswitch(
++ int set)
++{
++ return _hypercall1(int, fpu_taskswitch, set);
++}
++
++static inline int
++HYPERVISOR_sched_op_compat(
++ int cmd, unsigned long arg)
++{
++ return _hypercall2(int, sched_op_compat, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_sched_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, sched_op, cmd, arg);
++}
++
++static inline long
++HYPERVISOR_set_timer_op(
++ u64 timeout)
++{
++ unsigned long timeout_hi = (unsigned long)(timeout>>32);
++ unsigned long timeout_lo = (unsigned long)timeout;
++ return _hypercall2(long, set_timer_op, timeout_lo, timeout_hi);
++}
++
++static inline int
++HYPERVISOR_dom0_op(
++ dom0_op_t *dom0_op)
++{
++ dom0_op->interface_version = DOM0_INTERFACE_VERSION;
++ return _hypercall1(int, dom0_op, dom0_op);
++}
++
++static inline int
++HYPERVISOR_set_debugreg(
++ int reg, unsigned long value)
++{
++ return _hypercall2(int, set_debugreg, reg, value);
++}
++
++static inline unsigned long
++HYPERVISOR_get_debugreg(
++ int reg)
++{
++ return _hypercall1(unsigned long, get_debugreg, reg);
++}
++
++static inline int
++HYPERVISOR_update_descriptor(
++ u64 ma, u64 desc)
++{
++ return _hypercall4(int, update_descriptor, ma, ma>>32, desc, desc>>32);
++}
++
++static inline int
++HYPERVISOR_memory_op(
++ unsigned int cmd, void *arg)
++{
++ return _hypercall2(int, memory_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_multicall(
++ void *call_list, int nr_calls)
++{
++ return _hypercall2(int, multicall, call_list, nr_calls);
++}
++
++static inline int
++HYPERVISOR_update_va_mapping(
++ unsigned long va, pte_t new_val, unsigned long flags)
++{
++ unsigned long pte_hi = 0;
++#ifdef CONFIG_X86_PAE
++ pte_hi = new_val.pte_high;
++#endif
++ return _hypercall4(int, update_va_mapping, va,
++ new_val.pte_low, pte_hi, flags);
++}
++
++static inline int
++HYPERVISOR_event_channel_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, event_channel_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct evtchn_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, event_channel_op_compat, &op);
++ memcpy(arg, &op.u, sizeof(op.u));
++ }
++ return rc;
++}
++
++static inline int
++HYPERVISOR_acm_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, acm_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_xen_version(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, xen_version, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_console_io(
++ int cmd, int count, char *str)
++{
++ return _hypercall3(int, console_io, cmd, count, str);
++}
++
++static inline int
++HYPERVISOR_physdev_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, physdev_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct physdev_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, physdev_op_compat, &op);
++ memcpy(arg, &op.u, sizeof(op.u));
++ }
++ return rc;
++}
++
++static inline int
++HYPERVISOR_grant_table_op(
++ unsigned int cmd, void *uop, unsigned int count)
++{
++ return _hypercall3(int, grant_table_op, cmd, uop, count);
++}
++
++static inline int
++HYPERVISOR_update_va_mapping_otherdomain(
++ unsigned long va, pte_t new_val, unsigned long flags, domid_t domid)
++{
++ unsigned long pte_hi = 0;
++#ifdef CONFIG_X86_PAE
++ pte_hi = new_val.pte_high;
++#endif
++ return _hypercall5(int, update_va_mapping_otherdomain, va,
++ new_val.pte_low, pte_hi, flags, domid);
++}
++
++static inline int
++HYPERVISOR_vm_assist(
++ unsigned int cmd, unsigned int type)
++{
++ return _hypercall2(int, vm_assist, cmd, type);
++}
++
++static inline int
++HYPERVISOR_vcpu_op(
++ int cmd, int vcpuid, void *extra_args)
++{
++ return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args);
++}
++
++static inline int
++HYPERVISOR_suspend(
++ unsigned long srec)
++{
++ struct sched_shutdown sched_shutdown = {
++ .reason = SHUTDOWN_suspend
++ };
++
++ int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown,
++ &sched_shutdown, srec);
++
++ if (rc == -ENOSYS)
++ rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown,
++ SHUTDOWN_suspend, srec);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_nmi_op(
++ unsigned long op, void *arg)
++{
++ return _hypercall2(int, nmi_op, op, arg);
++}
++
++static inline int
++HYPERVISOR_callback_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, callback_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_xenoprof_op(
++ int op, void *arg)
++{
++ return _hypercall2(int, xenoprof_op, op, arg);
++}
++
++
++#endif /* __HYPERCALL_H__ */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/hypervisor.h new/include/asm-i386/mach-xen/asm/hypervisor.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/hypervisor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/hypervisor.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,226 @@
++/******************************************************************************
++ * hypervisor.h
++ *
++ * Linux-specific hypervisor handling.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __HYPERVISOR_H__
++#define __HYPERVISOR_H__
++
++#include <linux/config.h>
++#include <linux/types.h>
++#include <linux/kernel.h>
++#include <linux/version.h>
++#include <linux/errno.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/dom0_ops.h>
++#include <xen/interface/event_channel.h>
++#include <xen/interface/physdev.h>
++#include <xen/interface/sched.h>
++#include <xen/interface/nmi.h>
++#include <asm/ptrace.h>
++#include <asm/page.h>
++#if defined(__i386__)
++# ifdef CONFIG_X86_PAE
++# include <asm-generic/pgtable-nopud.h>
++# else
++# include <asm-generic/pgtable-nopmd.h>
++# endif
++#endif
++
++extern shared_info_t *HYPERVISOR_shared_info;
++
++/* arch/xen/i386/kernel/setup.c */
++extern start_info_t *xen_start_info;
++
++/* arch/xen/kernel/evtchn.c */
++/* Force a proper event-channel callback from Xen. */
++void force_evtchn_callback(void);
++
++/* arch/xen/kernel/process.c */
++void xen_cpu_idle (void);
++
++/* arch/xen/i386/kernel/hypervisor.c */
++void do_hypervisor_callback(struct pt_regs *regs);
++
++/* arch/xen/i386/mm/hypervisor.c */
++/*
++ * NB. ptr values should be PHYSICAL, not MACHINE. 'vals' should be already
++ * be MACHINE addresses.
++ */
++
++void xen_pt_switch(unsigned long ptr);
++void xen_new_user_pt(unsigned long ptr); /* x86_64 only */
++void xen_load_gs(unsigned int selector); /* x86_64 only */
++void xen_tlb_flush(void);
++void xen_invlpg(unsigned long ptr);
++
++void xen_l1_entry_update(pte_t *ptr, pte_t val);
++void xen_l2_entry_update(pmd_t *ptr, pmd_t val);
++void xen_l3_entry_update(pud_t *ptr, pud_t val); /* x86_64/PAE */
++void xen_l4_entry_update(pgd_t *ptr, pgd_t val); /* x86_64 only */
++void xen_pgd_pin(unsigned long ptr);
++void xen_pgd_unpin(unsigned long ptr);
++
++void xen_set_ldt(unsigned long ptr, unsigned long bytes);
++void xen_machphys_update(unsigned long mfn, unsigned long pfn);
++
++#ifdef CONFIG_SMP
++#include <linux/cpumask.h>
++void xen_tlb_flush_all(void);
++void xen_invlpg_all(unsigned long ptr);
++void xen_tlb_flush_mask(cpumask_t *mask);
++void xen_invlpg_mask(cpumask_t *mask, unsigned long ptr);
++#endif
++
++/* Returns zero on success else negative errno. */
++int xen_create_contiguous_region(
++ unsigned long vstart, unsigned int order, unsigned int address_bits);
++void xen_destroy_contiguous_region(
++ unsigned long vstart, unsigned int order);
++
++/* Turn jiffies into Xen system time. */
++u64 jiffies_to_st(unsigned long jiffies);
++
++#include <asm/hypercall.h>
++
++#if defined(CONFIG_X86_64)
++#define MULTI_UVMFLAGS_INDEX 2
++#define MULTI_UVMDOMID_INDEX 3
++#else
++#define MULTI_UVMFLAGS_INDEX 3
++#define MULTI_UVMDOMID_INDEX 4
++#endif
++
++#define is_running_on_xen() 1
++
++static inline int
++HYPERVISOR_yield(
++ void)
++{
++ int rc = HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_block(
++ void)
++{
++ int rc = HYPERVISOR_sched_op(SCHEDOP_block, NULL);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_block, 0);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_shutdown(
++ unsigned int reason)
++{
++ struct sched_shutdown sched_shutdown = {
++ .reason = reason
++ };
++
++ int rc = HYPERVISOR_sched_op(SCHEDOP_shutdown, &sched_shutdown);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_shutdown, reason);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_poll(
++ evtchn_port_t *ports, unsigned int nr_ports, u64 timeout)
++{
++ int rc;
++ struct sched_poll sched_poll = {
++ .nr_ports = nr_ports,
++ .timeout = jiffies_to_st(timeout)
++ };
++ set_xen_guest_handle(sched_poll.ports, ports);
++
++ rc = HYPERVISOR_sched_op(SCHEDOP_poll, &sched_poll);
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
++
++ return rc;
++}
++
++static inline void
++MULTI_update_va_mapping(
++ multicall_entry_t *mcl, unsigned long va,
++ pte_t new_val, unsigned long flags)
++{
++ mcl->op = __HYPERVISOR_update_va_mapping;
++ mcl->args[0] = va;
++#if defined(CONFIG_X86_64)
++ mcl->args[1] = new_val.pte;
++ mcl->args[2] = flags;
++#elif defined(CONFIG_X86_PAE)
++ mcl->args[1] = new_val.pte_low;
++ mcl->args[2] = new_val.pte_high;
++ mcl->args[3] = flags;
++#else
++ mcl->args[1] = new_val.pte_low;
++ mcl->args[2] = 0;
++ mcl->args[3] = flags;
++#endif
++}
++
++static inline void
++MULTI_update_va_mapping_otherdomain(
++ multicall_entry_t *mcl, unsigned long va,
++ pte_t new_val, unsigned long flags, domid_t domid)
++{
++ mcl->op = __HYPERVISOR_update_va_mapping_otherdomain;
++ mcl->args[0] = va;
++#if defined(CONFIG_X86_64)
++ mcl->args[1] = new_val.pte;
++ mcl->args[2] = flags;
++ mcl->args[3] = domid;
++#elif defined(CONFIG_X86_PAE)
++ mcl->args[1] = new_val.pte_low;
++ mcl->args[2] = new_val.pte_high;
++ mcl->args[3] = flags;
++ mcl->args[4] = domid;
++#else
++ mcl->args[1] = new_val.pte_low;
++ mcl->args[2] = 0;
++ mcl->args[3] = flags;
++ mcl->args[4] = domid;
++#endif
++}
++
++#endif /* __HYPERVISOR_H__ */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/io.h new/include/asm-i386/mach-xen/asm/io.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/io.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/io.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,390 @@
++#ifndef _ASM_IO_H
++#define _ASM_IO_H
++
++#include <linux/config.h>
++#include <linux/string.h>
++#include <linux/compiler.h>
++
++/*
++ * This file contains the definitions for the x86 IO instructions
++ * inb/inw/inl/outb/outw/outl and the "string versions" of the same
++ * (insb/insw/insl/outsb/outsw/outsl). You can also use "pausing"
++ * versions of the single-IO instructions (inb_p/inw_p/..).
++ *
++ * This file is not meant to be obfuscating: it's just complicated
++ * to (a) handle it all in a way that makes gcc able to optimize it
++ * as well as possible and (b) trying to avoid writing the same thing
++ * over and over again with slight variations and possibly making a
++ * mistake somewhere.
++ */
++
++/*
++ * Thanks to James van Artsdalen for a better timing-fix than
++ * the two short jumps: using outb's to a nonexistent port seems
++ * to guarantee better timings even on fast machines.
++ *
++ * On the other hand, I'd like to be sure of a non-existent port:
++ * I feel a bit unsafe about using 0x80 (should be safe, though)
++ *
++ * Linus
++ */
++
++ /*
++ * Bit simplified and optimized by Jan Hubicka
++ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999.
++ *
++ * isa_memset_io, isa_memcpy_fromio, isa_memcpy_toio added,
++ * isa_read[wl] and isa_write[wl] fixed
++ * - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
++ */
++
++#define IO_SPACE_LIMIT 0xffff
++
++#define XQUAD_PORTIO_BASE 0xfe400000
++#define XQUAD_PORTIO_QUAD 0x40000 /* 256k per quad. */
++
++#ifdef __KERNEL__
++
++#include <asm-generic/iomap.h>
++
++#include <linux/vmalloc.h>
++#include <asm/fixmap.h>
++
++/*
++ * Convert a physical pointer to a virtual kernel pointer for /dev/mem
++ * access
++ */
++#define xlate_dev_mem_ptr(p) __va(p)
++
++/*
++ * Convert a virtual cached pointer to an uncached pointer
++ */
++#define xlate_dev_kmem_ptr(p) p
++
++/**
++ * virt_to_phys - map virtual addresses to physical
++ * @address: address to remap
++ *
++ * The returned physical address is the physical (CPU) mapping for
++ * the memory address given. It is only valid to use this function on
++ * addresses directly mapped or allocated via kmalloc.
++ *
++ * This function does not give bus mappings for DMA transfers. In
++ * almost all conceivable cases a device driver should not be using
++ * this function
++ */
++
++static inline unsigned long virt_to_phys(volatile void * address)
++{
++ return __pa(address);
++}
++
++/**
++ * phys_to_virt - map physical address to virtual
++ * @address: address to remap
++ *
++ * The returned virtual address is a current CPU mapping for
++ * the memory address given. It is only valid to use this function on
++ * addresses that have a kernel mapping
++ *
++ * This function does not handle bus mappings for DMA transfers. In
++ * almost all conceivable cases a device driver should not be using
++ * this function
++ */
++
++static inline void * phys_to_virt(unsigned long address)
++{
++ return __va(address);
++}
++
++/*
++ * Change "struct page" to physical address.
++ */
++#define page_to_pseudophys(page) ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
++#define page_to_phys(page) (phys_to_machine(page_to_pseudophys(page)))
++#define page_to_bus(page) (phys_to_machine(page_to_pseudophys(page)))
++
++#define bio_to_pseudophys(bio) (page_to_pseudophys(bio_page((bio))) + \
++ (unsigned long) bio_offset((bio)))
++#define bvec_to_pseudophys(bv) (page_to_pseudophys((bv)->bv_page) + \
++ (unsigned long) (bv)->bv_offset)
++
++#define BIOVEC_PHYS_MERGEABLE(vec1, vec2) \
++ (((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2))) && \
++ ((bvec_to_pseudophys((vec1)) + (vec1)->bv_len) == \
++ bvec_to_pseudophys((vec2))))
++
++extern void __iomem * __ioremap(unsigned long offset, unsigned long size, unsigned long flags);
++
++/**
++ * ioremap - map bus memory into CPU space
++ * @offset: bus address of the memory
++ * @size: size of the resource to map
++ *
++ * ioremap performs a platform specific sequence of operations to
++ * make bus memory CPU accessible via the readb/readw/readl/writeb/
++ * writew/writel functions and the other mmio helpers. The returned
++ * address is not guaranteed to be usable directly as a virtual
++ * address.
++ */
++
++static inline void __iomem * ioremap(unsigned long offset, unsigned long size)
++{
++ return __ioremap(offset, size, 0);
++}
++
++extern void __iomem * ioremap_nocache(unsigned long offset, unsigned long size);
++extern void iounmap(volatile void __iomem *addr);
++
++/*
++ * bt_ioremap() and bt_iounmap() are for temporary early boot-time
++ * mappings, before the real ioremap() is functional.
++ * A boot-time mapping is currently limited to at most 16 pages.
++ */
++extern void *bt_ioremap(unsigned long offset, unsigned long size);
++extern void bt_iounmap(void *addr, unsigned long size);
++
++/* Use early IO mappings for DMI because it's initialized early */
++#define dmi_ioremap bt_ioremap
++#define dmi_iounmap bt_iounmap
++#define dmi_alloc alloc_bootmem
++
++/*
++ * ISA I/O bus memory addresses are 1:1 with the physical address.
++ */
++#define isa_virt_to_bus(_x) isa_virt_to_bus_is_UNSUPPORTED->x
++#define isa_page_to_bus(_x) isa_page_to_bus_is_UNSUPPORTED->x
++#define isa_bus_to_virt(_x) (void *)(__fix_to_virt(FIX_ISAMAP_BEGIN) + (_x))
++
++/*
++ * However PCI ones are not necessarily 1:1 and therefore these interfaces
++ * are forbidden in portable PCI drivers.
++ *
++ * Allow them on x86 for legacy drivers, though.
++ */
++#define virt_to_bus(_x) phys_to_machine(__pa(_x))
++#define bus_to_virt(_x) __va(machine_to_phys(_x))
++
++/*
++ * readX/writeX() are used to access memory mapped devices. On some
++ * architectures the memory mapped IO stuff needs to be accessed
++ * differently. On the x86 architecture, we just read/write the
++ * memory location directly.
++ */
++
++static inline unsigned char readb(const volatile void __iomem *addr)
++{
++ return *(volatile unsigned char __force *) addr;
++}
++static inline unsigned short readw(const volatile void __iomem *addr)
++{
++ return *(volatile unsigned short __force *) addr;
++}
++static inline unsigned int readl(const volatile void __iomem *addr)
++{
++ return *(volatile unsigned int __force *) addr;
++}
++#define readb_relaxed(addr) readb(addr)
++#define readw_relaxed(addr) readw(addr)
++#define readl_relaxed(addr) readl(addr)
++#define __raw_readb readb
++#define __raw_readw readw
++#define __raw_readl readl
++
++static inline void writeb(unsigned char b, volatile void __iomem *addr)
++{
++ *(volatile unsigned char __force *) addr = b;
++}
++static inline void writew(unsigned short b, volatile void __iomem *addr)
++{
++ *(volatile unsigned short __force *) addr = b;
++}
++static inline void writel(unsigned int b, volatile void __iomem *addr)
++{
++ *(volatile unsigned int __force *) addr = b;
++}
++#define __raw_writeb writeb
++#define __raw_writew writew
++#define __raw_writel writel
++
++#define mmiowb()
++
++static inline void memset_io(volatile void __iomem *addr, unsigned char val, int count)
++{
++ memset((void __force *) addr, val, count);
++}
++static inline void memcpy_fromio(void *dst, const volatile void __iomem *src, int count)
++{
++ __memcpy(dst, (void __force *) src, count);
++}
++static inline void memcpy_toio(volatile void __iomem *dst, const void *src, int count)
++{
++ __memcpy((void __force *) dst, src, count);
++}
++
++/*
++ * ISA space is 'always mapped' on a typical x86 system, no need to
++ * explicitly ioremap() it. The fact that the ISA IO space is mapped
++ * to PAGE_OFFSET is pure coincidence - it does not mean ISA values
++ * are physical addresses. The following constant pointer can be
++ * used as the IO-area pointer (it can be iounmapped as well, so the
++ * analogy with PCI is quite large):
++ */
++#define __ISA_IO_base ((char __iomem *)(fix_to_virt(FIX_ISAMAP_BEGIN)))
++
++/*
++ * Again, i386 does not require mem IO specific function.
++ */
++
++#define eth_io_copy_and_sum(a,b,c,d) eth_copy_and_sum((a),(void __force *)(b),(c),(d))
++
++/**
++ * check_signature - find BIOS signatures
++ * @io_addr: mmio address to check
++ * @signature: signature block
++ * @length: length of signature
++ *
++ * Perform a signature comparison with the mmio address io_addr. This
++ * address should have been obtained by ioremap.
++ * Returns 1 on a match.
++ */
++
++static inline int check_signature(volatile void __iomem * io_addr,
++ const unsigned char *signature, int length)
++{
++ int retval = 0;
++ do {
++ if (readb(io_addr) != *signature)
++ goto out;
++ io_addr++;
++ signature++;
++ length--;
++ } while (length);
++ retval = 1;
++out:
++ return retval;
++}
++
++/*
++ * Cache management
++ *
++ * This needed for two cases
++ * 1. Out of order aware processors
++ * 2. Accidentally out of order processors (PPro errata #51)
++ */
++
++#if defined(CONFIG_X86_OOSTORE) || defined(CONFIG_X86_PPRO_FENCE)
++
++static inline void flush_write_buffers(void)
++{
++ __asm__ __volatile__ ("lock; addl $0,0(%%esp)": : :"memory");
++}
++
++#define dma_cache_inv(_start,_size) flush_write_buffers()
++#define dma_cache_wback(_start,_size) flush_write_buffers()
++#define dma_cache_wback_inv(_start,_size) flush_write_buffers()
++
++#else
++
++/* Nothing to do */
++
++#define dma_cache_inv(_start,_size) do { } while (0)
++#define dma_cache_wback(_start,_size) do { } while (0)
++#define dma_cache_wback_inv(_start,_size) do { } while (0)
++#define flush_write_buffers()
++
++#endif
++
++#endif /* __KERNEL__ */
++
++#ifdef SLOW_IO_BY_JUMPING
++#define __SLOW_DOWN_IO "jmp 1f; 1: jmp 1f; 1:"
++#else
++#define __SLOW_DOWN_IO "outb %%al,$0x80;"
++#endif
++
++static inline void slow_down_io(void) {
++ __asm__ __volatile__(
++ __SLOW_DOWN_IO
++#ifdef REALLY_SLOW_IO
++ __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO
++#endif
++ : : );
++}
++
++#ifdef CONFIG_X86_NUMAQ
++extern void *xquad_portio; /* Where the IO area was mapped */
++#define XQUAD_PORT_ADDR(port, quad) (xquad_portio + (XQUAD_PORTIO_QUAD*quad) + port)
++#define __BUILDIO(bwl,bw,type) \
++static inline void out##bwl##_quad(unsigned type value, int port, int quad) { \
++ if (xquad_portio) \
++ write##bwl(value, XQUAD_PORT_ADDR(port, quad)); \
++ else \
++ out##bwl##_local(value, port); \
++} \
++static inline void out##bwl(unsigned type value, int port) { \
++ out##bwl##_quad(value, port, 0); \
++} \
++static inline unsigned type in##bwl##_quad(int port, int quad) { \
++ if (xquad_portio) \
++ return read##bwl(XQUAD_PORT_ADDR(port, quad)); \
++ else \
++ return in##bwl##_local(port); \
++} \
++static inline unsigned type in##bwl(int port) { \
++ return in##bwl##_quad(port, 0); \
++}
++#else
++#define __BUILDIO(bwl,bw,type) \
++static inline void out##bwl(unsigned type value, int port) { \
++ out##bwl##_local(value, port); \
++} \
++static inline unsigned type in##bwl(int port) { \
++ return in##bwl##_local(port); \
++}
++#endif
++
++
++#define BUILDIO(bwl,bw,type) \
++static inline void out##bwl##_local(unsigned type value, int port) { \
++ __asm__ __volatile__("out" #bwl " %" #bw "0, %w1" : : "a"(value), "Nd"(port)); \
++} \
++static inline unsigned type in##bwl##_local(int port) { \
++ unsigned type value; \
++ __asm__ __volatile__("in" #bwl " %w1, %" #bw "0" : "=a"(value) : "Nd"(port)); \
++ return value; \
++} \
++static inline void out##bwl##_local_p(unsigned type value, int port) { \
++ out##bwl##_local(value, port); \
++ slow_down_io(); \
++} \
++static inline unsigned type in##bwl##_local_p(int port) { \
++ unsigned type value = in##bwl##_local(port); \
++ slow_down_io(); \
++ return value; \
++} \
++__BUILDIO(bwl,bw,type) \
++static inline void out##bwl##_p(unsigned type value, int port) { \
++ out##bwl(value, port); \
++ slow_down_io(); \
++} \
++static inline unsigned type in##bwl##_p(int port) { \
++ unsigned type value = in##bwl(port); \
++ slow_down_io(); \
++ return value; \
++} \
++static inline void outs##bwl(int port, const void *addr, unsigned long count) { \
++ __asm__ __volatile__("rep; outs" #bwl : "+S"(addr), "+c"(count) : "d"(port)); \
++} \
++static inline void ins##bwl(int port, void *addr, unsigned long count) { \
++ __asm__ __volatile__("rep; ins" #bwl : "+D"(addr), "+c"(count) : "d"(port)); \
++}
++
++BUILDIO(b,b,char)
++BUILDIO(w,w,short)
++BUILDIO(l,,int)
++
++/* We will be supplying our own /dev/mem implementation */
++#define ARCH_HAS_DEV_MEM
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/kmap_types.h new/include/asm-i386/mach-xen/asm/kmap_types.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/kmap_types.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/kmap_types.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,32 @@
++#ifndef _ASM_KMAP_TYPES_H
++#define _ASM_KMAP_TYPES_H
++
++#include <linux/config.h>
++
++#ifdef CONFIG_DEBUG_HIGHMEM
++# define D(n) __KM_FENCE_##n ,
++#else
++# define D(n)
++#endif
++
++enum km_type {
++D(0) KM_BOUNCE_READ,
++D(1) KM_SKB_SUNRPC_DATA,
++D(2) KM_SKB_DATA_SOFTIRQ,
++D(3) KM_USER0,
++D(4) KM_USER1,
++D(5) KM_BIO_SRC_IRQ,
++D(6) KM_BIO_DST_IRQ,
++D(7) KM_PTE0,
++D(8) KM_PTE1,
++D(9) KM_IRQ0,
++D(10) KM_IRQ1,
++D(11) KM_SOFTIRQ0,
++D(12) KM_SOFTIRQ1,
++D(13) KM_SWIOTLB,
++D(14) KM_TYPE_NR
++};
++
++#undef D
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/mmu_context.h new/include/asm-i386/mach-xen/asm/mmu_context.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/mmu_context.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/mmu_context.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,108 @@
++#ifndef __I386_SCHED_H
++#define __I386_SCHED_H
++
++#include <linux/config.h>
++#include <asm/desc.h>
++#include <asm/atomic.h>
++#include <asm/pgalloc.h>
++#include <asm/tlbflush.h>
++
++/*
++ * Used for LDT copy/destruction.
++ */
++int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
++void destroy_context(struct mm_struct *mm);
++
++
++static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
++{
++#if 0 /* XEN: no lazy tlb */
++ unsigned cpu = smp_processor_id();
++ if (per_cpu(cpu_tlbstate, cpu).state == TLBSTATE_OK)
++ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_LAZY;
++#endif
++}
++
++#define prepare_arch_switch(next) __prepare_arch_switch()
++
++static inline void __prepare_arch_switch(void)
++{
++ /*
++ * Save away %fs and %gs. No need to save %es and %ds, as those
++ * are always kernel segments while inside the kernel. Must
++ * happen before reload of cr3/ldt (i.e., not in __switch_to).
++ */
++ asm volatile ( "mov %%fs,%0 ; mov %%gs,%1"
++ : "=m" (current->thread.fs),
++ "=m" (current->thread.gs));
++ asm volatile ( "movl %0,%%fs ; movl %0,%%gs"
++ : : "r" (0) );
++}
++
++extern void mm_pin(struct mm_struct *mm);
++extern void mm_unpin(struct mm_struct *mm);
++void mm_pin_all(void);
++
++static inline void switch_mm(struct mm_struct *prev,
++ struct mm_struct *next,
++ struct task_struct *tsk)
++{
++ int cpu = smp_processor_id();
++ struct mmuext_op _op[2], *op = _op;
++
++ if (likely(prev != next)) {
++ BUG_ON(!test_bit(PG_pinned, &virt_to_page(next->pgd)->flags));
++
++ /* stop flush ipis for the previous mm */
++ cpu_clear(cpu, prev->cpu_vm_mask);
++#if 0 /* XEN: no lazy tlb */
++ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
++ per_cpu(cpu_tlbstate, cpu).active_mm = next;
++#endif
++ cpu_set(cpu, next->cpu_vm_mask);
++
++ /* Re-load page tables: load_cr3(next->pgd) */
++ op->cmd = MMUEXT_NEW_BASEPTR;
++ op->arg1.mfn = pfn_to_mfn(__pa(next->pgd) >> PAGE_SHIFT);
++ op++;
++
++ /*
++ * load the LDT, if the LDT is different:
++ */
++ if (unlikely(prev->context.ldt != next->context.ldt)) {
++ /* load_LDT_nolock(&next->context, cpu) */
++ op->cmd = MMUEXT_SET_LDT;
++ op->arg1.linear_addr = (unsigned long)next->context.ldt;
++ op->arg2.nr_ents = next->context.size;
++ op++;
++ }
++
++ BUG_ON(HYPERVISOR_mmuext_op(_op, op-_op, NULL, DOMID_SELF));
++ }
++#if 0 /* XEN: no lazy tlb */
++ else {
++ per_cpu(cpu_tlbstate, cpu).state = TLBSTATE_OK;
++ BUG_ON(per_cpu(cpu_tlbstate, cpu).active_mm != next);
++
++ if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
++ /* We were in lazy tlb mode and leave_mm disabled
++ * tlb flush IPI delivery. We must reload %cr3.
++ */
++ load_cr3(next->pgd);
++ load_LDT_nolock(&next->context, cpu);
++ }
++ }
++#endif
++}
++
++#define deactivate_mm(tsk, mm) \
++ asm("movl %0,%%fs ; movl %0,%%gs": :"r" (0))
++
++static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
++{
++ if (!test_bit(PG_pinned, &virt_to_page(next->pgd)->flags))
++ mm_pin(next);
++ switch_mm(prev, next, NULL);
++}
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/mmu.h new/include/asm-i386/mach-xen/asm/mmu.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/mmu.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/mmu.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,25 @@
++#ifndef __i386_MMU_H
++#define __i386_MMU_H
++
++#include <asm/semaphore.h>
++/*
++ * The i386 doesn't have a mmu context, but
++ * we put the segment information here.
++ *
++ * cpu_vm_mask is used to optimize ldt flushing.
++ */
++typedef struct {
++ int size;
++ struct semaphore sem;
++ void *ldt;
++} mm_context_t;
++
++/* mm/memory.c:exit_mmap hook */
++extern void _arch_exit_mmap(struct mm_struct *mm);
++#define arch_exit_mmap(_mm) _arch_exit_mmap(_mm)
++
++/* kernel/fork.c:dup_mmap hook */
++extern void _arch_dup_mmap(struct mm_struct *mm);
++#define arch_dup_mmap(mm, oldmm) ((void)(oldmm), _arch_dup_mmap(mm))
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/page.h new/include/asm-i386/mach-xen/asm/page.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/page.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/page.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,336 @@
++#ifndef _I386_PAGE_H
++#define _I386_PAGE_H
++
++/* PAGE_SHIFT determines the page size */
++#define PAGE_SHIFT 12
++#define PAGE_SIZE (1UL << PAGE_SHIFT)
++#define PAGE_MASK (~(PAGE_SIZE-1))
++
++#define LARGE_PAGE_MASK (~(LARGE_PAGE_SIZE-1))
++#define LARGE_PAGE_SIZE (1UL << PMD_SHIFT)
++
++#ifdef __KERNEL__
++#ifndef __ASSEMBLY__
++
++#include <linux/config.h>
++#include <linux/string.h>
++#include <linux/types.h>
++#include <linux/kernel.h>
++#include <asm/bug.h>
++#include <xen/interface/xen.h>
++#include <xen/features.h>
++#include <xen/foreign_page.h>
++
++#define arch_free_page(_page,_order) \
++({ int foreign = PageForeign(_page); \
++ if (foreign) \
++ (PageForeignDestructor(_page))(_page); \
++ foreign; \
++})
++#define HAVE_ARCH_FREE_PAGE
++
++#ifdef CONFIG_XEN_SCRUB_PAGES
++#define scrub_pages(_p,_n) memset((void *)(_p), 0, (_n) << PAGE_SHIFT)
++#else
++#define scrub_pages(_p,_n) ((void)0)
++#endif
++
++#ifdef CONFIG_X86_USE_3DNOW
++
++#include <asm/mmx.h>
++
++#define clear_page(page) mmx_clear_page((void *)(page))
++#define copy_page(to,from) mmx_copy_page(to,from)
++
++#else
++
++#define alloc_zeroed_user_highpage(vma, vaddr) alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO, vma, vaddr)
++#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
++
++/*
++ * On older X86 processors it's not a win to use MMX here it seems.
++ * Maybe the K6-III ?
++ */
++
++#define clear_page(page) memset((void *)(page), 0, PAGE_SIZE)
++#define copy_page(to,from) memcpy((void *)(to), (void *)(from), PAGE_SIZE)
++
++#endif
++
++#define clear_user_page(page, vaddr, pg) clear_page(page)
++#define copy_user_page(to, from, vaddr, pg) copy_page(to, from)
++
++/**** MACHINE <-> PHYSICAL CONVERSION MACROS ****/
++#define INVALID_P2M_ENTRY (~0UL)
++#define FOREIGN_FRAME_BIT (1UL<<31)
++#define FOREIGN_FRAME(m) ((m) | FOREIGN_FRAME_BIT)
++
++extern unsigned long *phys_to_machine_mapping;
++
++#undef machine_to_phys_mapping
++extern unsigned long *machine_to_phys_mapping;
++extern unsigned int machine_to_phys_order;
++
++static inline unsigned long pfn_to_mfn(unsigned long pfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return pfn;
++ return phys_to_machine_mapping[(unsigned int)(pfn)] &
++ ~FOREIGN_FRAME_BIT;
++}
++
++static inline int phys_to_machine_mapping_valid(unsigned long pfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return 1;
++ return (phys_to_machine_mapping[pfn] != INVALID_P2M_ENTRY);
++}
++
++static inline unsigned long mfn_to_pfn(unsigned long mfn)
++{
++ extern unsigned long max_mapnr;
++ unsigned long pfn;
++
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return mfn;
++
++ if (unlikely((mfn >> machine_to_phys_order) != 0))
++ return max_mapnr;
++
++ /* The array access can fail (e.g., device space beyond end of RAM). */
++ asm (
++ "1: movl %1,%0\n"
++ "2:\n"
++ ".section .fixup,\"ax\"\n"
++ "3: movl %2,%0\n"
++ " jmp 2b\n"
++ ".previous\n"
++ ".section __ex_table,\"a\"\n"
++ " .align 4\n"
++ " .long 1b,3b\n"
++ ".previous"
++ : "=r" (pfn)
++ : "m" (machine_to_phys_mapping[mfn]), "m" (max_mapnr) );
++
++ return pfn;
++}
++
++/*
++ * We detect special mappings in one of two ways:
++ * 1. If the MFN is an I/O page then Xen will set the m2p entry
++ * to be outside our maximum possible pseudophys range.
++ * 2. If the MFN belongs to a different domain then we will certainly
++ * not have MFN in our p2m table. Conversely, if the page is ours,
++ * then we'll have p2m(m2p(MFN))==MFN.
++ * If we detect a special mapping then it doesn't have a 'struct page'.
++ * We force !pfn_valid() by returning an out-of-range pointer.
++ *
++ * NB. These checks require that, for any MFN that is not in our reservation,
++ * there is no PFN such that p2m(PFN) == MFN. Otherwise we can get confused if
++ * we are foreign-mapping the MFN, and the other domain as m2p(MFN) == PFN.
++ * Yikes! Various places must poke in INVALID_P2M_ENTRY for safety.
++ *
++ * NB2. When deliberately mapping foreign pages into the p2m table, you *must*
++ * use FOREIGN_FRAME(). This will cause pte_pfn() to choke on it, as we
++ * require. In all the cases we care about, the FOREIGN_FRAME bit is
++ * masked (e.g., pfn_to_mfn()) so behaviour there is correct.
++ */
++static inline unsigned long mfn_to_local_pfn(unsigned long mfn)
++{
++ extern unsigned long max_mapnr;
++ unsigned long pfn = mfn_to_pfn(mfn);
++ if ((pfn < max_mapnr)
++ && !xen_feature(XENFEAT_auto_translated_physmap)
++ && (phys_to_machine_mapping[pfn] != mfn))
++ return max_mapnr; /* force !pfn_valid() */
++ return pfn;
++}
++
++static inline void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap)) {
++ BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
++ return;
++ }
++ phys_to_machine_mapping[pfn] = mfn;
++}
++
++/* Definitions for machine and pseudophysical addresses. */
++#ifdef CONFIG_X86_PAE
++typedef unsigned long long paddr_t;
++typedef unsigned long long maddr_t;
++#else
++typedef unsigned long paddr_t;
++typedef unsigned long maddr_t;
++#endif
++
++static inline maddr_t phys_to_machine(paddr_t phys)
++{
++ maddr_t machine = pfn_to_mfn(phys >> PAGE_SHIFT);
++ machine = (machine << PAGE_SHIFT) | (phys & ~PAGE_MASK);
++ return machine;
++}
++static inline paddr_t machine_to_phys(maddr_t machine)
++{
++ paddr_t phys = mfn_to_pfn(machine >> PAGE_SHIFT);
++ phys = (phys << PAGE_SHIFT) | (machine & ~PAGE_MASK);
++ return phys;
++}
++
++/*
++ * These are used to make use of C type-checking..
++ */
++extern int nx_enabled;
++#ifdef CONFIG_X86_PAE
++extern unsigned long long __supported_pte_mask;
++typedef struct { unsigned long pte_low, pte_high; } pte_t;
++typedef struct { unsigned long long pmd; } pmd_t;
++typedef struct { unsigned long long pgd; } pgd_t;
++typedef struct { unsigned long long pgprot; } pgprot_t;
++#define __pte(x) ({ unsigned long long _x = (x); \
++ if (_x & 1) _x = phys_to_machine(_x); \
++ ((pte_t) {(unsigned long)(_x), (unsigned long)(_x>>32)}); })
++#define __pgd(x) ({ unsigned long long _x = (x); \
++ (((_x)&1) ? ((pgd_t) {phys_to_machine(_x)}) : ((pgd_t) {(_x)})); })
++#define __pmd(x) ({ unsigned long long _x = (x); \
++ (((_x)&1) ? ((pmd_t) {phys_to_machine(_x)}) : ((pmd_t) {(_x)})); })
++static inline unsigned long long pte_val(pte_t x)
++{
++ unsigned long long ret;
++
++ if (x.pte_low) {
++ ret = x.pte_low | (unsigned long long)x.pte_high << 32;
++ ret = machine_to_phys(ret) | 1;
++ } else {
++ ret = 0;
++ }
++ return ret;
++}
++static inline unsigned long long pmd_val(pmd_t x)
++{
++ unsigned long long ret = x.pmd;
++ if (ret) ret = machine_to_phys(ret) | 1;
++ return ret;
++}
++static inline unsigned long long pgd_val(pgd_t x)
++{
++ unsigned long long ret = x.pgd;
++ if (ret) ret = machine_to_phys(ret) | 1;
++ return ret;
++}
++static inline unsigned long long pte_val_ma(pte_t x)
++{
++ return (unsigned long long)x.pte_high << 32 | x.pte_low;
++}
++#define HPAGE_SHIFT 21
++#else
++typedef struct { unsigned long pte_low; } pte_t;
++typedef struct { unsigned long pgd; } pgd_t;
++typedef struct { unsigned long pgprot; } pgprot_t;
++#define boot_pte_t pte_t /* or would you rather have a typedef */
++#define pte_val(x) (((x).pte_low & 1) ? machine_to_phys((x).pte_low) : \
++ (x).pte_low)
++#define pte_val_ma(x) ((x).pte_low)
++#define __pte(x) ({ unsigned long _x = (x); \
++ (((_x)&1) ? ((pte_t) {phys_to_machine(_x)}) : ((pte_t) {(_x)})); })
++#define __pgd(x) ({ unsigned long _x = (x); \
++ (((_x)&1) ? ((pgd_t) {phys_to_machine(_x)}) : ((pgd_t) {(_x)})); })
++static inline unsigned long pgd_val(pgd_t x)
++{
++ unsigned long ret = x.pgd;
++ if (ret) ret = machine_to_phys(ret) | 1;
++ return ret;
++}
++#define HPAGE_SHIFT 22
++#endif
++#define PTE_MASK PAGE_MASK
++
++#ifdef CONFIG_HUGETLB_PAGE
++#define HPAGE_SIZE ((1UL) << HPAGE_SHIFT)
++#define HPAGE_MASK (~(HPAGE_SIZE - 1))
++#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT)
++#define HAVE_ARCH_HUGETLB_UNMAPPED_AREA
++#endif
++
++#define pgprot_val(x) ((x).pgprot)
++
++#define __pte_ma(x) ((pte_t) { (x) } )
++#define __pgprot(x) ((pgprot_t) { (x) } )
++
++#endif /* !__ASSEMBLY__ */
++
++/* to align the pointer to the (next) page boundary */
++#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK)
++
++/*
++ * This handles the memory map.. We could make this a config
++ * option, but too many people screw it up, and too few need
++ * it.
++ *
++ * A __PAGE_OFFSET of 0xC0000000 means that the kernel has
++ * a virtual address space of one gigabyte, which limits the
++ * amount of physical memory you can use to about 950MB.
++ *
++ * If you want more physical memory than this then see the CONFIG_HIGHMEM4G
++ * and CONFIG_HIGHMEM64G options in the kernel configuration.
++ */
++
++#ifndef __ASSEMBLY__
++
++/*
++ * This much address space is reserved for vmalloc() and iomap()
++ * as well as fixmap mappings.
++ */
++extern unsigned int __VMALLOC_RESERVE;
++
++extern int sysctl_legacy_va_layout;
++
++extern int page_is_ram(unsigned long pagenr);
++
++#endif /* __ASSEMBLY__ */
++
++#ifdef __ASSEMBLY__
++#define __PAGE_OFFSET CONFIG_PAGE_OFFSET
++#define __PHYSICAL_START CONFIG_PHYSICAL_START
++#else
++#define __PAGE_OFFSET ((unsigned long)CONFIG_PAGE_OFFSET)
++#define __PHYSICAL_START ((unsigned long)CONFIG_PHYSICAL_START)
++#endif
++#define __KERNEL_START (__PAGE_OFFSET + __PHYSICAL_START)
++
++#ifdef CONFIG_XEN_COMPAT_030002
++#undef LOAD_OFFSET
++#define LOAD_OFFSET 0
++#endif /* CONFIG_XEN_COMPAT_030002 */
++
++#define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET)
++#define VMALLOC_RESERVE ((unsigned long)__VMALLOC_RESERVE)
++#define MAXMEM (__FIXADDR_TOP-__PAGE_OFFSET-__VMALLOC_RESERVE)
++#define __pa(x) ((unsigned long)(x)-PAGE_OFFSET)
++#define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET))
++#define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT)
++#ifdef CONFIG_FLATMEM
++#define pfn_valid(pfn) ((pfn) < max_mapnr)
++#endif /* CONFIG_FLATMEM */
++#define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
++
++#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
++
++#define VM_DATA_DEFAULT_FLAGS \
++ (VM_READ | VM_WRITE | \
++ ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
++ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
++
++/* VIRT <-> MACHINE conversion */
++#define virt_to_machine(v) (phys_to_machine(__pa(v)))
++#define virt_to_mfn(v) (pfn_to_mfn(__pa(v) >> PAGE_SHIFT))
++#define mfn_to_virt(m) (__va(mfn_to_pfn(m) << PAGE_SHIFT))
++
++#define __HAVE_ARCH_GATE_AREA 1
++
++#endif /* __KERNEL__ */
++
++#include <asm-generic/memory_model.h>
++#include <asm-generic/page.h>
++
++#endif /* _I386_PAGE_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/param.h new/include/asm-i386/mach-xen/asm/param.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/param.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/param.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,24 @@
++#ifndef _ASMi386_PARAM_H
++#define _ASMi386_PARAM_H
++
++#ifdef __KERNEL__
++# include <linux/config.h>
++# define HZ CONFIG_HZ /* Internal kernel timer frequency */
++# define USER_HZ 100 /* .. some user interfaces are in "ticks" */
++# define CLOCKS_PER_SEC (USER_HZ) /* like times() */
++#endif
++
++#ifndef HZ
++#define HZ 100
++#endif
++
++#define EXEC_PAGESIZE 4096
++
++#ifndef NOGROUP
++#define NOGROUP (-1)
++#endif
++
++#define MAXHOSTNAMELEN 64 /* max length of hostname */
++#define COMMAND_LINE_SIZE 256
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pci.h new/include/asm-i386/mach-xen/asm/pci.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pci.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pci.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,154 @@
++#ifndef __i386_PCI_H
++#define __i386_PCI_H
++
++#include <linux/config.h>
++
++#ifdef __KERNEL__
++#include <linux/mm.h> /* for struct page */
++
++/* Can be used to override the logic in pci_scan_bus for skipping
++ already-configured bus numbers - to be used for buggy BIOSes
++ or architectures with incomplete PCI setup by the loader */
++
++#ifdef CONFIG_PCI
++extern unsigned int pcibios_assign_all_busses(void);
++#else
++#define pcibios_assign_all_busses() 0
++#endif
++#define pcibios_scan_all_fns(a, b) 0
++
++extern unsigned long pci_mem_start;
++#define PCIBIOS_MIN_IO 0x1000
++#define PCIBIOS_MIN_MEM (pci_mem_start)
++
++#define PCIBIOS_MIN_CARDBUS_IO 0x4000
++
++void pcibios_config_init(void);
++struct pci_bus * pcibios_scan_root(int bus);
++
++void pcibios_set_master(struct pci_dev *dev);
++void pcibios_penalize_isa_irq(int irq, int active);
++struct irq_routing_table *pcibios_get_irq_routing_table(void);
++int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq);
++
++/* Dynamic DMA mapping stuff.
++ * i386 has everything mapped statically.
++ */
++
++#include <linux/types.h>
++#include <linux/slab.h>
++#include <asm/scatterlist.h>
++#include <linux/string.h>
++#include <asm/io.h>
++
++struct pci_dev;
++
++#ifdef CONFIG_SWIOTLB
++
++
++/* On Xen we use SWIOTLB instead of blk-specific bounce buffers. */
++#define PCI_DMA_BUS_IS_PHYS (0)
++
++#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \
++ dma_addr_t ADDR_NAME;
++#define DECLARE_PCI_UNMAP_LEN(LEN_NAME) \
++ __u32 LEN_NAME;
++#define pci_unmap_addr(PTR, ADDR_NAME) \
++ ((PTR)->ADDR_NAME)
++#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) \
++ (((PTR)->ADDR_NAME) = (VAL))
++#define pci_unmap_len(PTR, LEN_NAME) \
++ ((PTR)->LEN_NAME)
++#define pci_unmap_len_set(PTR, LEN_NAME, VAL) \
++ (((PTR)->LEN_NAME) = (VAL))
++
++#else
++
++/* The PCI address space does equal the physical memory
++ * address space. The networking and block device layers use
++ * this boolean for bounce buffer decisions.
++ */
++#define PCI_DMA_BUS_IS_PHYS (1)
++
++/* pci_unmap_{page,single} is a nop so... */
++#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
++#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
++#define pci_unmap_addr(PTR, ADDR_NAME) (0)
++#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) do { } while (0)
++#define pci_unmap_len(PTR, LEN_NAME) (0)
++#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
++
++#endif
++
++/* This is always fine. */
++#define pci_dac_dma_supported(pci_dev, mask) (1)
++
++static inline dma64_addr_t
++pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
++{
++ return ((dma64_addr_t) page_to_phys(page) +
++ (dma64_addr_t) offset);
++}
++
++static inline struct page *
++pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
++{
++ return pfn_to_page(dma_addr >> PAGE_SHIFT);
++}
++
++static inline unsigned long
++pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
++{
++ return (dma_addr & ~PAGE_MASK);
++}
++
++static inline void
++pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
++{
++}
++
++static inline void
++pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
++{
++ flush_write_buffers();
++}
++
++#define HAVE_PCI_MMAP
++extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
++ enum pci_mmap_state mmap_state, int write_combine);
++
++
++static inline void pcibios_add_platform_entries(struct pci_dev *dev)
++{
++}
++
++#ifdef CONFIG_PCI
++static inline void pci_dma_burst_advice(struct pci_dev *pdev,
++ enum pci_dma_burst_strategy *strat,
++ unsigned long *strategy_parameter)
++{
++ *strat = PCI_DMA_BURST_INFINITY;
++ *strategy_parameter = ~0UL;
++}
++#endif
++
++#endif /* __KERNEL__ */
++
++#ifdef CONFIG_XEN_PCIDEV_FRONTEND
++#include <xen/pcifront.h>
++#endif /* CONFIG_XEN_PCIDEV_FRONTEND */
++
++/* implement the pci_ DMA API in terms of the generic device dma_ one */
++#include <asm-generic/pci-dma-compat.h>
++
++/* generic pci stuff */
++#include <asm-generic/pci.h>
++
++/* On Xen we have to scan all functions since Xen hides bridges from
++ * us. If a bridge is at fn=0 and that slot has a multifunction
++ * device, we won't find the additional devices without scanning all
++ * functions. */
++#undef pcibios_scan_all_fns
++#define pcibios_scan_all_fns(a, b) 1
++
++#endif /* __i386_PCI_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgalloc.h new/include/asm-i386/mach-xen/asm/pgalloc.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgalloc.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgalloc.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,64 @@
++#ifndef _I386_PGALLOC_H
++#define _I386_PGALLOC_H
++
++#include <linux/config.h>
++#include <asm/fixmap.h>
++#include <linux/threads.h>
++#include <linux/mm.h> /* for struct page */
++#include <asm/io.h> /* for phys_to_virt and page_to_pseudophys */
++
++/* Is this pagetable pinned? */
++#define PG_pinned PG_arch_1
++
++#define pmd_populate_kernel(mm, pmd, pte) \
++ set_pmd(pmd, __pmd(_PAGE_TABLE + __pa(pte)))
++
++#define pmd_populate(mm, pmd, pte) \
++do { \
++ if (test_bit(PG_pinned, &virt_to_page((mm)->pgd)->flags)) { \
++ if (!PageHighMem(pte)) \
++ BUG_ON(HYPERVISOR_update_va_mapping( \
++ (unsigned long)__va(page_to_pfn(pte)<<PAGE_SHIFT),\
++ pfn_pte(page_to_pfn(pte), PAGE_KERNEL_RO), 0));\
++ set_pmd(pmd, __pmd(_PAGE_TABLE + \
++ ((unsigned long long)page_to_pfn(pte) << \
++ (unsigned long long) PAGE_SHIFT))); \
++ } else { \
++ *(pmd) = __pmd(_PAGE_TABLE + \
++ ((unsigned long long)page_to_pfn(pte) << \
++ (unsigned long long) PAGE_SHIFT)); \
++ } \
++} while (0)
++
++/*
++ * Allocate and free page tables.
++ */
++extern pgd_t *pgd_alloc(struct mm_struct *);
++extern void pgd_free(pgd_t *pgd);
++
++extern pte_t *pte_alloc_one_kernel(struct mm_struct *, unsigned long);
++extern struct page *pte_alloc_one(struct mm_struct *, unsigned long);
++
++static inline void pte_free_kernel(pte_t *pte)
++{
++ free_page((unsigned long)pte);
++ make_page_writable(pte, XENFEAT_writable_page_tables);
++}
++
++extern void pte_free(struct page *pte);
++
++#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
++
++#ifdef CONFIG_X86_PAE
++/*
++ * In the PAE case we free the pmds as part of the pgd.
++ */
++#define pmd_alloc_one(mm, addr) ({ BUG(); ((pmd_t *)2); })
++#define pmd_free(x) do { } while (0)
++#define __pmd_free_tlb(tlb,x) do { } while (0)
++#define pud_populate(mm, pmd, pte) BUG()
++#endif
++
++#define check_pgt_cache() do { } while (0)
++
++#endif /* _I386_PGALLOC_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgtable-2level-defs.h new/include/asm-i386/mach-xen/asm/pgtable-2level-defs.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgtable-2level-defs.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgtable-2level-defs.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,21 @@
++#ifndef _I386_PGTABLE_2LEVEL_DEFS_H
++#define _I386_PGTABLE_2LEVEL_DEFS_H
++
++#define HAVE_SHARED_KERNEL_PMD 0
++
++/*
++ * traditional i386 two-level paging structure:
++ */
++
++#define PGDIR_SHIFT 22
++#define PTRS_PER_PGD 1024
++#define PTRS_PER_PGD_NO_HV (HYPERVISOR_VIRT_START >> PGDIR_SHIFT)
++
++/*
++ * the i386 is two-level, so we don't really have any
++ * PMD directory physically.
++ */
++
++#define PTRS_PER_PTE 1024
++
++#endif /* _I386_PGTABLE_2LEVEL_DEFS_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgtable-2level.h new/include/asm-i386/mach-xen/asm/pgtable-2level.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgtable-2level.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgtable-2level.h 2006-05-23 18:37:11.000000000 +0200
+@@ -0,0 +1,88 @@
++#ifndef _I386_PGTABLE_2LEVEL_H
++#define _I386_PGTABLE_2LEVEL_H
++
++#include <asm-generic/pgtable-nopmd.h>
++
++#define pte_ERROR(e) \
++ printk("%s:%d: bad pte %08lx.\n", __FILE__, __LINE__, (e).pte_low)
++#define pgd_ERROR(e) \
++ printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
++
++/*
++ * Certain architectures need to do special things when PTEs
++ * within a page table are directly modified. Thus, the following
++ * hook is made available.
++ */
++#define set_pte(pteptr, pteval) (*(pteptr) = pteval)
++
++#define set_pte_at(_mm,addr,ptep,pteval) do { \
++ if (((_mm) != current->mm && (_mm) != &init_mm) || \
++ HYPERVISOR_update_va_mapping((addr), (pteval), 0)) \
++ set_pte((ptep), (pteval)); \
++} while (0)
++
++#define set_pte_at_sync(_mm,addr,ptep,pteval) do { \
++ if (((_mm) != current->mm && (_mm) != &init_mm) || \
++ HYPERVISOR_update_va_mapping((addr), (pteval), UVMF_INVLPG)) { \
++ set_pte((ptep), (pteval)); \
++ xen_invlpg((addr)); \
++ } \
++} while (0)
++
++#define set_pte_atomic(pteptr, pteval) set_pte(pteptr,pteval)
++
++#define set_pmd(pmdptr, pmdval) xen_l2_entry_update((pmdptr), (pmdval))
++
++#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
++#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
++
++#define ptep_get_and_clear(mm,addr,xp) __pte_ma(xchg(&(xp)->pte_low, 0))
++#define pte_same(a, b) ((a).pte_low == (b).pte_low)
++#define pte_mfn(_pte) ((_pte).pte_low >> PAGE_SHIFT)
++#define pte_pfn(_pte) mfn_to_local_pfn(pte_mfn(_pte))
++
++#define pte_page(_pte) pfn_to_page(pte_pfn(_pte))
++
++#define pte_none(x) (!(x).pte_low)
++#define pfn_pte(pfn, prot) __pte(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
++#define pfn_pte_ma(pfn, prot) __pte_ma(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
++#define pfn_pmd(pfn, prot) __pmd(((pfn) << PAGE_SHIFT) | pgprot_val(prot))
++
++/*
++ * All present user pages are user-executable:
++ */
++static inline int pte_exec(pte_t pte)
++{
++ return pte_user(pte);
++}
++
++/*
++ * All present pages are kernel-executable:
++ */
++static inline int pte_exec_kernel(pte_t pte)
++{
++ return 1;
++}
++
++/*
++ * Bits 0, 6 and 7 are taken, split up the 29 bits of offset
++ * into this range:
++ */
++#define PTE_FILE_MAX_BITS 29
++
++#define pte_to_pgoff(pte) \
++ ((((pte).pte_low >> 1) & 0x1f ) + (((pte).pte_low >> 8) << 5 ))
++
++#define pgoff_to_pte(off) \
++ ((pte_t) { (((off) & 0x1f) << 1) + (((off) >> 5) << 8) + _PAGE_FILE })
++
++/* Encode and de-code a swap entry */
++#define __swp_type(x) (((x).val >> 1) & 0x1f)
++#define __swp_offset(x) ((x).val >> 8)
++#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
++#define __pte_to_swp_entry(pte) ((swp_entry_t) { (pte).pte_low })
++#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
++
++void vmalloc_sync_all(void);
++
++#endif /* _I386_PGTABLE_2LEVEL_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgtable-3level-defs.h new/include/asm-i386/mach-xen/asm/pgtable-3level-defs.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgtable-3level-defs.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgtable-3level-defs.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,25 @@
++#ifndef _I386_PGTABLE_3LEVEL_DEFS_H
++#define _I386_PGTABLE_3LEVEL_DEFS_H
++
++#define HAVE_SHARED_KERNEL_PMD 0
++
++/*
++ * PGDIR_SHIFT determines what a top-level page table entry can map
++ */
++#define PGDIR_SHIFT 30
++#define PTRS_PER_PGD 4
++#define PTRS_PER_PGD_NO_HV 4
++
++/*
++ * PMD_SHIFT determines the size of the area a middle-level
++ * page table can map
++ */
++#define PMD_SHIFT 21
++#define PTRS_PER_PMD 512
++
++/*
++ * entries per page directory level
++ */
++#define PTRS_PER_PTE 512
++
++#endif /* _I386_PGTABLE_3LEVEL_DEFS_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgtable-3level.h new/include/asm-i386/mach-xen/asm/pgtable-3level.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgtable-3level.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgtable-3level.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,197 @@
++#ifndef _I386_PGTABLE_3LEVEL_H
++#define _I386_PGTABLE_3LEVEL_H
++
++#include <asm-generic/pgtable-nopud.h>
++
++/*
++ * Intel Physical Address Extension (PAE) Mode - three-level page
++ * tables on PPro+ CPUs.
++ *
++ * Copyright (C) 1999 Ingo Molnar <mingo@redhat.com>
++ */
++
++#define pte_ERROR(e) \
++ printk("%s:%d: bad pte %p(%08lx%08lx).\n", __FILE__, __LINE__, &(e), (e).pte_high, (e).pte_low)
++#define pmd_ERROR(e) \
++ printk("%s:%d: bad pmd %p(%016Lx).\n", __FILE__, __LINE__, &(e), pmd_val(e))
++#define pgd_ERROR(e) \
++ printk("%s:%d: bad pgd %p(%016Lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
++
++#define pud_none(pud) 0
++#define pud_bad(pud) 0
++#define pud_present(pud) 1
++
++/*
++ * Is the pte executable?
++ */
++static inline int pte_x(pte_t pte)
++{
++ return !(pte_val(pte) & _PAGE_NX);
++}
++
++/*
++ * All present user-pages with !NX bit are user-executable:
++ */
++static inline int pte_exec(pte_t pte)
++{
++ return pte_user(pte) && pte_x(pte);
++}
++/*
++ * All present pages with !NX bit are kernel-executable:
++ */
++static inline int pte_exec_kernel(pte_t pte)
++{
++ return pte_x(pte);
++}
++
++/* Rules for using set_pte: the pte being assigned *must* be
++ * either not present or in a state where the hardware will
++ * not attempt to update the pte. In places where this is
++ * not possible, use pte_get_and_clear to obtain the old pte
++ * value and then use set_pte to update it. -ben
++ */
++#define __HAVE_ARCH_SET_PTE_ATOMIC
++
++#if 1
++/* use writable pagetables */
++static inline void set_pte(pte_t *ptep, pte_t pte)
++{
++ ptep->pte_high = pte.pte_high;
++ smp_wmb();
++ ptep->pte_low = pte.pte_low;
++}
++# define set_pte_atomic(pteptr,pteval) \
++ set_64bit((unsigned long long *)(pteptr),pte_val_ma(pteval))
++#else
++/* no writable pagetables */
++# define set_pte(pteptr,pteval) \
++ xen_l1_entry_update((pteptr), (pteval))
++# define set_pte_atomic(pteptr,pteval) set_pte(pteptr,pteval)
++#endif
++
++#define set_pte_at(_mm,addr,ptep,pteval) do { \
++ if (((_mm) != current->mm && (_mm) != &init_mm) || \
++ HYPERVISOR_update_va_mapping((addr), (pteval), 0)) \
++ set_pte((ptep), (pteval)); \
++} while (0)
++
++#define set_pte_at_sync(_mm,addr,ptep,pteval) do { \
++ if (((_mm) != current->mm && (_mm) != &init_mm) || \
++ HYPERVISOR_update_va_mapping((addr), (pteval), UVMF_INVLPG)) { \
++ set_pte((ptep), (pteval)); \
++ xen_invlpg((addr)); \
++ } \
++} while (0)
++
++#define set_pmd(pmdptr,pmdval) \
++ xen_l2_entry_update((pmdptr), (pmdval))
++#define set_pud(pudptr,pudval) \
++ xen_l3_entry_update((pudptr), (pudval))
++
++/*
++ * Pentium-II erratum A13: in PAE mode we explicitly have to flush
++ * the TLB via cr3 if the top-level pgd is changed...
++ * We do not let the generic code free and clear pgd entries due to
++ * this erratum.
++ */
++static inline void pud_clear (pud_t * pud) { }
++
++#define pud_page(pud) \
++((struct page *) __va(pud_val(pud) & PAGE_MASK))
++
++#define pud_page_kernel(pud) \
++((unsigned long) __va(pud_val(pud) & PAGE_MASK))
++
++
++/* Find an entry in the second-level page table.. */
++#define pmd_offset(pud, address) ((pmd_t *) pud_page(*(pud)) + \
++ pmd_index(address))
++
++/*
++ * For PTEs and PDEs, we must clear the P-bit first when clearing a page table
++ * entry, so clear the bottom half first and enforce ordering with a compiler
++ * barrier.
++ */
++static inline void pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
++{
++ ptep->pte_low = 0;
++ smp_wmb();
++ ptep->pte_high = 0;
++}
++
++#define pmd_clear(xp)do { set_pmd(xp, __pmd(0)); } while (0)
++
++static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
++{
++ pte_t res;
++
++ /* xchg acts as a barrier before the setting of the high bits */
++ res.pte_low = xchg(&ptep->pte_low, 0);
++ res.pte_high = ptep->pte_high;
++ ptep->pte_high = 0;
++
++ return res;
++}
++
++static inline int pte_same(pte_t a, pte_t b)
++{
++ return a.pte_low == b.pte_low && a.pte_high == b.pte_high;
++}
++
++#define pte_page(x) pfn_to_page(pte_pfn(x))
++
++static inline int pte_none(pte_t pte)
++{
++ return !pte.pte_low && !pte.pte_high;
++}
++
++#define pte_mfn(_pte) (((_pte).pte_low >> PAGE_SHIFT) |\
++ (((_pte).pte_high & 0xfff) << (32-PAGE_SHIFT)))
++#define pte_pfn(_pte) mfn_to_local_pfn(pte_mfn(_pte))
++
++extern unsigned long long __supported_pte_mask;
++
++static inline pte_t pfn_pte_ma(unsigned long page_nr, pgprot_t pgprot)
++{
++ pte_t pte;
++
++ pte.pte_high = (page_nr >> (32 - PAGE_SHIFT)) | \
++ (pgprot_val(pgprot) >> 32);
++ pte.pte_high &= (__supported_pte_mask >> 32);
++ pte.pte_low = ((page_nr << PAGE_SHIFT) | pgprot_val(pgprot)) & \
++ __supported_pte_mask;
++ return pte;
++}
++
++static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
++{
++ return pfn_pte_ma(pfn_to_mfn(page_nr), pgprot);
++}
++
++static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
++{
++ BUG(); panic("needs review");
++ return __pmd((((unsigned long long)page_nr << PAGE_SHIFT) | \
++ pgprot_val(pgprot)) & __supported_pte_mask);
++}
++
++/*
++ * Bits 0, 6 and 7 are taken in the low part of the pte,
++ * put the 32 bits of offset into the high part.
++ */
++#define pte_to_pgoff(pte) ((pte).pte_high)
++#define pgoff_to_pte(off) ((pte_t) { _PAGE_FILE, (off) })
++#define PTE_FILE_MAX_BITS 32
++
++/* Encode and de-code a swap entry */
++#define __swp_type(x) (((x).val) & 0x1f)
++#define __swp_offset(x) ((x).val >> 5)
++#define __swp_entry(type, offset) ((swp_entry_t){(type) | (offset) << 5})
++#define __pte_to_swp_entry(pte) ((swp_entry_t){ (pte).pte_high })
++#define __swp_entry_to_pte(x) ((pte_t){ 0, (x).val })
++
++#define __pmd_free_tlb(tlb, x) do { } while (0)
++
++#define vmalloc_sync_all() ((void)0)
++
++#endif /* _I386_PGTABLE_3LEVEL_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/pgtable.h new/include/asm-i386/mach-xen/asm/pgtable.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/pgtable.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/pgtable.h 2006-05-23 18:37:12.000000000 +0200
+@@ -0,0 +1,509 @@
++#ifndef _I386_PGTABLE_H
++#define _I386_PGTABLE_H
++
++#include <linux/config.h>
++#include <asm/hypervisor.h>
++
++/*
++ * The Linux memory management assumes a three-level page table setup. On
++ * the i386, we use that, but "fold" the mid level into the top-level page
++ * table, so that we physically have the same two-level page table as the
++ * i386 mmu expects.
++ *
++ * This file contains the functions and defines necessary to modify and use
++ * the i386 page table tree.
++ */
++#ifndef __ASSEMBLY__
++#include <asm/processor.h>
++#include <asm/fixmap.h>
++#include <linux/threads.h>
++
++#ifndef _I386_BITOPS_H
++#include <asm/bitops.h>
++#endif
++
++#include <linux/slab.h>
++#include <linux/list.h>
++#include <linux/spinlock.h>
++
++struct mm_struct;
++struct vm_area_struct;
++
++/*
++ * ZERO_PAGE is a global shared page that is always zero: used
++ * for zero-mapped memory areas etc..
++ */
++#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
++extern unsigned long empty_zero_page[1024];
++extern pgd_t *swapper_pg_dir;
++extern kmem_cache_t *pgd_cache;
++extern kmem_cache_t *pmd_cache;
++extern spinlock_t pgd_lock;
++extern struct page *pgd_list;
++
++void pmd_ctor(void *, kmem_cache_t *, unsigned long);
++void pgd_ctor(void *, kmem_cache_t *, unsigned long);
++void pgd_dtor(void *, kmem_cache_t *, unsigned long);
++void pgtable_cache_init(void);
++void paging_init(void);
++
++/*
++ * The Linux x86 paging architecture is 'compile-time dual-mode', it
++ * implements both the traditional 2-level x86 page tables and the
++ * newer 3-level PAE-mode page tables.
++ */
++#ifdef CONFIG_X86_PAE
++# include <asm/pgtable-3level-defs.h>
++# define PMD_SIZE (1UL << PMD_SHIFT)
++# define PMD_MASK (~(PMD_SIZE-1))
++#else
++# include <asm/pgtable-2level-defs.h>
++#endif
++
++#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
++#define PGDIR_MASK (~(PGDIR_SIZE-1))
++
++#define USER_PTRS_PER_PGD (TASK_SIZE/PGDIR_SIZE)
++#define FIRST_USER_ADDRESS 0
++
++#define USER_PGD_PTRS (PAGE_OFFSET >> PGDIR_SHIFT)
++#define KERNEL_PGD_PTRS (PTRS_PER_PGD-USER_PGD_PTRS)
++
++#define TWOLEVEL_PGDIR_SHIFT 22
++#define BOOT_USER_PGD_PTRS (__PAGE_OFFSET >> TWOLEVEL_PGDIR_SHIFT)
++#define BOOT_KERNEL_PGD_PTRS (1024-BOOT_USER_PGD_PTRS)
++
++/* Just any arbitrary offset to the start of the vmalloc VM area: the
++ * current 8MB value just means that there will be a 8MB "hole" after the
++ * physical memory until the kernel virtual memory starts. That means that
++ * any out-of-bounds memory accesses will hopefully be caught.
++ * The vmalloc() routines leaves a hole of 4kB between each vmalloced
++ * area for the same reason. ;)
++ */
++#define VMALLOC_OFFSET (8*1024*1024)
++#define VMALLOC_START (((unsigned long) high_memory + vmalloc_earlyreserve + \
++ 2*VMALLOC_OFFSET-1) & ~(VMALLOC_OFFSET-1))
++#ifdef CONFIG_HIGHMEM
++# define VMALLOC_END (PKMAP_BASE-2*PAGE_SIZE)
++#else
++# define VMALLOC_END (FIXADDR_START-2*PAGE_SIZE)
++#endif
++
++/*
++ * _PAGE_PSE set in the page directory entry just means that
++ * the page directory entry points directly to a 4MB-aligned block of
++ * memory.
++ */
++#define _PAGE_BIT_PRESENT 0
++#define _PAGE_BIT_RW 1
++#define _PAGE_BIT_USER 2
++#define _PAGE_BIT_PWT 3
++#define _PAGE_BIT_PCD 4
++#define _PAGE_BIT_ACCESSED 5
++#define _PAGE_BIT_DIRTY 6
++#define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page, Pentium+, if present.. */
++#define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */
++#define _PAGE_BIT_UNUSED1 9 /* available for programmer */
++#define _PAGE_BIT_UNUSED2 10
++#define _PAGE_BIT_UNUSED3 11
++#define _PAGE_BIT_NX 63
++
++#define _PAGE_PRESENT 0x001
++#define _PAGE_RW 0x002
++#define _PAGE_USER 0x004
++#define _PAGE_PWT 0x008
++#define _PAGE_PCD 0x010
++#define _PAGE_ACCESSED 0x020
++#define _PAGE_DIRTY 0x040
++#define _PAGE_PSE 0x080 /* 4 MB (or 2MB) page, Pentium+, if present.. */
++#define _PAGE_GLOBAL 0x100 /* Global TLB entry PPro+ */
++#define _PAGE_UNUSED1 0x200 /* available for programmer */
++#define _PAGE_UNUSED2 0x400
++#define _PAGE_UNUSED3 0x800
++
++/* If _PAGE_PRESENT is clear, we use these: */
++#define _PAGE_FILE 0x040 /* nonlinear file mapping, saved PTE; unset:swap */
++#define _PAGE_PROTNONE 0x080 /* if the user mapped it with PROT_NONE;
++ pte_present gives true */
++#ifdef CONFIG_X86_PAE
++#define _PAGE_NX (1ULL<<_PAGE_BIT_NX)
++#else
++#define _PAGE_NX 0
++#endif
++
++#define _PAGE_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
++#define _KERNPG_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED | _PAGE_DIRTY)
++#define _PAGE_CHG_MASK (PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
++
++#define PAGE_NONE \
++ __pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
++#define PAGE_SHARED \
++ __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
++
++#define PAGE_SHARED_EXEC \
++ __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
++#define PAGE_COPY_NOEXEC \
++ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
++#define PAGE_COPY_EXEC \
++ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
++#define PAGE_COPY \
++ PAGE_COPY_NOEXEC
++#define PAGE_READONLY \
++ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
++#define PAGE_READONLY_EXEC \
++ __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
++
++#define _PAGE_KERNEL \
++ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX)
++#define _PAGE_KERNEL_EXEC \
++ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED)
++
++extern unsigned long long __PAGE_KERNEL, __PAGE_KERNEL_EXEC;
++#define __PAGE_KERNEL_RO (__PAGE_KERNEL & ~_PAGE_RW)
++#define __PAGE_KERNEL_NOCACHE (__PAGE_KERNEL | _PAGE_PCD)
++#define __PAGE_KERNEL_LARGE (__PAGE_KERNEL | _PAGE_PSE)
++#define __PAGE_KERNEL_LARGE_EXEC (__PAGE_KERNEL_EXEC | _PAGE_PSE)
++
++#define PAGE_KERNEL __pgprot(__PAGE_KERNEL)
++#define PAGE_KERNEL_RO __pgprot(__PAGE_KERNEL_RO)
++#define PAGE_KERNEL_EXEC __pgprot(__PAGE_KERNEL_EXEC)
++#define PAGE_KERNEL_NOCACHE __pgprot(__PAGE_KERNEL_NOCACHE)
++#define PAGE_KERNEL_LARGE __pgprot(__PAGE_KERNEL_LARGE)
++#define PAGE_KERNEL_LARGE_EXEC __pgprot(__PAGE_KERNEL_LARGE_EXEC)
++
++/*
++ * The i386 can't do page protection for execute, and considers that
++ * the same are read. Also, write permissions imply read permissions.
++ * This is the closest we can get..
++ */
++#define __P000 PAGE_NONE
++#define __P001 PAGE_READONLY
++#define __P010 PAGE_COPY
++#define __P011 PAGE_COPY
++#define __P100 PAGE_READONLY_EXEC
++#define __P101 PAGE_READONLY_EXEC
++#define __P110 PAGE_COPY_EXEC
++#define __P111 PAGE_COPY_EXEC
++
++#define __S000 PAGE_NONE
++#define __S001 PAGE_READONLY
++#define __S010 PAGE_SHARED
++#define __S011 PAGE_SHARED
++#define __S100 PAGE_READONLY_EXEC
++#define __S101 PAGE_READONLY_EXEC
++#define __S110 PAGE_SHARED_EXEC
++#define __S111 PAGE_SHARED_EXEC
++
++/*
++ * Define this if things work differently on an i386 and an i486:
++ * it will (on an i486) warn about kernel memory accesses that are
++ * done without a 'access_ok(VERIFY_WRITE,..)'
++ */
++#undef TEST_ACCESS_OK
++
++/* The boot page tables (all created as a single array) */
++extern unsigned long pg0[];
++
++#define pte_present(x) ((x).pte_low & (_PAGE_PRESENT | _PAGE_PROTNONE))
++
++/* To avoid harmful races, pmd_none(x) should check only the lower when PAE */
++#define pmd_none(x) (!(unsigned long)pmd_val(x))
++/* pmd_present doesn't just test the _PAGE_PRESENT bit since wr.p.t.
++ can temporarily clear it. */
++#define pmd_present(x) (pmd_val(x))
++#define pmd_bad(x) ((pmd_val(x) & (~PAGE_MASK & ~_PAGE_USER & ~_PAGE_PRESENT)) != (_KERNPG_TABLE & ~_PAGE_PRESENT))
++
++
++#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
++
++/*
++ * The following only work if pte_present() is true.
++ * Undefined behaviour if not..
++ */
++static inline int pte_user(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
++static inline int pte_read(pte_t pte) { return (pte).pte_low & _PAGE_USER; }
++static inline int pte_dirty(pte_t pte) { return (pte).pte_low & _PAGE_DIRTY; }
++static inline int pte_young(pte_t pte) { return (pte).pte_low & _PAGE_ACCESSED; }
++static inline int pte_write(pte_t pte) { return (pte).pte_low & _PAGE_RW; }
++static inline int pte_huge(pte_t pte) { return (pte).pte_low & _PAGE_PSE; }
++
++/*
++ * The following only works if pte_present() is not true.
++ */
++static inline int pte_file(pte_t pte) { return (pte).pte_low & _PAGE_FILE; }
++
++static inline pte_t pte_rdprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
++static inline pte_t pte_exprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_USER; return pte; }
++static inline pte_t pte_mkclean(pte_t pte) { (pte).pte_low &= ~_PAGE_DIRTY; return pte; }
++static inline pte_t pte_mkold(pte_t pte) { (pte).pte_low &= ~_PAGE_ACCESSED; return pte; }
++static inline pte_t pte_wrprotect(pte_t pte) { (pte).pte_low &= ~_PAGE_RW; return pte; }
++static inline pte_t pte_mkread(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
++static inline pte_t pte_mkexec(pte_t pte) { (pte).pte_low |= _PAGE_USER; return pte; }
++static inline pte_t pte_mkdirty(pte_t pte) { (pte).pte_low |= _PAGE_DIRTY; return pte; }
++static inline pte_t pte_mkyoung(pte_t pte) { (pte).pte_low |= _PAGE_ACCESSED; return pte; }
++static inline pte_t pte_mkwrite(pte_t pte) { (pte).pte_low |= _PAGE_RW; return pte; }
++static inline pte_t pte_mkhuge(pte_t pte) { (pte).pte_low |= _PAGE_PSE; return pte; }
++
++#ifdef CONFIG_X86_PAE
++# include <asm/pgtable-3level.h>
++#else
++# include <asm/pgtable-2level.h>
++#endif
++
++static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
++{
++ if (!pte_dirty(*ptep))
++ return 0;
++ return test_and_clear_bit(_PAGE_BIT_DIRTY, &ptep->pte_low);
++}
++
++static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
++{
++ if (!pte_young(*ptep))
++ return 0;
++ return test_and_clear_bit(_PAGE_BIT_ACCESSED, &ptep->pte_low);
++}
++
++static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
++{
++ pte_t pte;
++ if (full) {
++ pte = *ptep;
++ pte_clear(mm, addr, ptep);
++ } else {
++ pte = ptep_get_and_clear(mm, addr, ptep);
++ }
++ return pte;
++}
++
++static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
++{
++ if (pte_write(*ptep))
++ clear_bit(_PAGE_BIT_RW, &ptep->pte_low);
++}
++
++/*
++ * clone_pgd_range(pgd_t *dst, pgd_t *src, int count);
++ *
++ * dst - pointer to pgd range anwhere on a pgd page
++ * src - ""
++ * count - the number of pgds to copy.
++ *
++ * dst and src can be on the same page, but the range must not overlap,
++ * and must not cross a page boundary.
++ */
++static inline void clone_pgd_range(pgd_t *dst, pgd_t *src, int count)
++{
++ memcpy(dst, src, count * sizeof(pgd_t));
++}
++
++/*
++ * Macro to mark a page protection value as "uncacheable". On processors which do not support
++ * it, this is a no-op.
++ */
++#define pgprot_noncached(prot) ((boot_cpu_data.x86 > 3) \
++ ? (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT)) : (prot))
++
++/*
++ * Conversion functions: convert a page and protection to a page entry,
++ * and a page entry and page directory to the page they refer to.
++ */
++
++#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
++
++static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
++{
++ pte.pte_low &= _PAGE_CHG_MASK;
++ pte.pte_low |= pgprot_val(newprot);
++#ifdef CONFIG_X86_PAE
++ /*
++ * Chop off the NX bit (if present), and add the NX portion of
++ * the newprot (if present):
++ */
++ pte.pte_high &= ~(1 << (_PAGE_BIT_NX - 32));
++ pte.pte_high |= (pgprot_val(newprot) >> 32) & \
++ (__supported_pte_mask >> 32);
++#endif
++ return pte;
++}
++
++#define pmd_large(pmd) \
++((pmd_val(pmd) & (_PAGE_PSE|_PAGE_PRESENT)) == (_PAGE_PSE|_PAGE_PRESENT))
++
++/*
++ * the pgd page can be thought of an array like this: pgd_t[PTRS_PER_PGD]
++ *
++ * this macro returns the index of the entry in the pgd page which would
++ * control the given virtual address
++ */
++#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
++#define pgd_index_k(addr) pgd_index(addr)
++
++/*
++ * pgd_offset() returns a (pgd_t *)
++ * pgd_index() is used get the offset into the pgd page's array of pgd_t's;
++ */
++#define pgd_offset(mm, address) ((mm)->pgd+pgd_index(address))
++
++/*
++ * a shortcut which implies the use of the kernel's pgd, instead
++ * of a process's
++ */
++#define pgd_offset_k(address) pgd_offset(&init_mm, address)
++
++/*
++ * the pmd page can be thought of an array like this: pmd_t[PTRS_PER_PMD]
++ *
++ * this macro returns the index of the entry in the pmd page which would
++ * control the given virtual address
++ */
++#define pmd_index(address) \
++ (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
++
++/*
++ * the pte page can be thought of an array like this: pte_t[PTRS_PER_PTE]
++ *
++ * this macro returns the index of the entry in the pte page which would
++ * control the given virtual address
++ */
++#define pte_index(address) \
++ (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
++#define pte_offset_kernel(dir, address) \
++ ((pte_t *) pmd_page_kernel(*(dir)) + pte_index(address))
++
++#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
++
++#define pmd_page_kernel(pmd) \
++ ((unsigned long) __va(pmd_val(pmd) & PAGE_MASK))
++
++/*
++ * Helper function that returns the kernel pagetable entry controlling
++ * the virtual address 'address'. NULL means no pagetable entry present.
++ * NOTE: the return type is pte_t but if the pmd is PSE then we return it
++ * as a pte too.
++ */
++extern pte_t *lookup_address(unsigned long address);
++
++/*
++ * Make a given kernel text page executable/non-executable.
++ * Returns the previous executability setting of that page (which
++ * is used to restore the previous state). Used by the SMP bootup code.
++ * NOTE: this is an __init function for security reasons.
++ */
++#ifdef CONFIG_X86_PAE
++ extern int set_kernel_exec(unsigned long vaddr, int enable);
++#else
++ static inline int set_kernel_exec(unsigned long vaddr, int enable) { return 0;}
++#endif
++
++extern void noexec_setup(const char *str);
++
++#if defined(CONFIG_HIGHPTE)
++#define pte_offset_map(dir, address) \
++ ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE0) + \
++ pte_index(address))
++#define pte_offset_map_nested(dir, address) \
++ ((pte_t *)kmap_atomic_pte(pmd_page(*(dir)),KM_PTE1) + \
++ pte_index(address))
++#define pte_unmap(pte) kunmap_atomic(pte, KM_PTE0)
++#define pte_unmap_nested(pte) kunmap_atomic(pte, KM_PTE1)
++#else
++#define pte_offset_map(dir, address) \
++ ((pte_t *)page_address(pmd_page(*(dir))) + pte_index(address))
++#define pte_offset_map_nested(dir, address) pte_offset_map(dir, address)
++#define pte_unmap(pte) do { } while (0)
++#define pte_unmap_nested(pte) do { } while (0)
++#endif
++
++/*
++ * The i386 doesn't have any external MMU info: the kernel page
++ * tables contain all the necessary information.
++ *
++ * Also, we only update the dirty/accessed state if we set
++ * the dirty bit by hand in the kernel, since the hardware
++ * will do the accessed bit for us, and we don't want to
++ * race with other CPU's that might be updating the dirty
++ * bit at the same time.
++ */
++#define update_mmu_cache(vma,address,pte) do { } while (0)
++#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
++#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
++ do { \
++ if (__dirty) { \
++ if ( likely((__vma)->vm_mm == current->mm) ) { \
++ BUG_ON(HYPERVISOR_update_va_mapping((__address), (__entry), UVMF_INVLPG|UVMF_MULTI|(unsigned long)((__vma)->vm_mm->cpu_vm_mask.bits))); \
++ } else { \
++ xen_l1_entry_update((__ptep), (__entry)); \
++ flush_tlb_page((__vma), (__address)); \
++ } \
++ } \
++ } while (0)
++
++#define __HAVE_ARCH_PTEP_ESTABLISH
++#define ptep_establish(__vma, __address, __ptep, __entry) \
++do { \
++ ptep_set_access_flags(__vma, __address, __ptep, __entry, 1); \
++} while (0)
++
++#include <xen/features.h>
++void make_lowmem_page_readonly(void *va, unsigned int feature);
++void make_lowmem_page_writable(void *va, unsigned int feature);
++void make_page_readonly(void *va, unsigned int feature);
++void make_page_writable(void *va, unsigned int feature);
++void make_pages_readonly(void *va, unsigned int nr, unsigned int feature);
++void make_pages_writable(void *va, unsigned int nr, unsigned int feature);
++
++#define virt_to_ptep(__va) \
++({ \
++ pgd_t *__pgd = pgd_offset_k((unsigned long)(__va)); \
++ pud_t *__pud = pud_offset(__pgd, (unsigned long)(__va)); \
++ pmd_t *__pmd = pmd_offset(__pud, (unsigned long)(__va)); \
++ pte_offset_kernel(__pmd, (unsigned long)(__va)); \
++})
++
++#define arbitrary_virt_to_machine(__va) \
++({ \
++ maddr_t m = (maddr_t)pte_mfn(*virt_to_ptep(__va)) << PAGE_SHIFT;\
++ m | ((unsigned long)(__va) & (PAGE_SIZE-1)); \
++})
++
++#endif /* !__ASSEMBLY__ */
++
++#ifdef CONFIG_FLATMEM
++#define kern_addr_valid(addr) (1)
++#endif /* CONFIG_FLATMEM */
++
++int direct_remap_pfn_range(struct vm_area_struct *vma,
++ unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid);
++int direct_kernel_remap_pfn_range(unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid);
++int create_lookup_pte_addr(struct mm_struct *mm,
++ unsigned long address,
++ uint64_t *ptep);
++int touch_pte_range(struct mm_struct *mm,
++ unsigned long address,
++ unsigned long size);
++
++#define io_remap_pfn_range(vma,from,pfn,size,prot) \
++direct_remap_pfn_range(vma,from,pfn,size,prot,DOMID_IO)
++
++#define MK_IOSPACE_PFN(space, pfn) (pfn)
++#define GET_IOSPACE(pfn) 0
++#define GET_PFN(pfn) (pfn)
++
++#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
++#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
++#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
++#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
++#define __HAVE_ARCH_PTEP_SET_WRPROTECT
++#define __HAVE_ARCH_PTE_SAME
++#include <asm-generic/pgtable.h>
++
++#endif /* _I386_PGTABLE_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/processor.h new/include/asm-i386/mach-xen/asm/processor.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/processor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/processor.h 2006-05-23 18:37:12.000000000 +0200
+@@ -0,0 +1,753 @@
++/*
++ * include/asm-i386/processor.h
++ *
++ * Copyright (C) 1994 Linus Torvalds
++ */
++
++#ifndef __ASM_I386_PROCESSOR_H
++#define __ASM_I386_PROCESSOR_H
++
++#include <asm/vm86.h>
++#include <asm/math_emu.h>
++#include <asm/segment.h>
++#include <asm/page.h>
++#include <asm/types.h>
++#include <asm/sigcontext.h>
++#include <asm/cpufeature.h>
++#include <asm/msr.h>
++#include <asm/system.h>
++#include <linux/cache.h>
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <asm/percpu.h>
++#include <linux/cpumask.h>
++#include <xen/interface/physdev.h>
++
++/* flag for disabling the tsc */
++extern int tsc_disable;
++
++struct desc_struct {
++ unsigned long a,b;
++};
++
++#define desc_empty(desc) \
++ (!((desc)->a | (desc)->b))
++
++#define desc_equal(desc1, desc2) \
++ (((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
++/*
++ * Default implementation of macro that returns current
++ * instruction pointer ("program counter").
++ */
++#define current_text_addr() ({ void *pc; __asm__("movl $1f,%0\n1:":"=g" (pc)); pc; })
++
++/*
++ * CPU type and hardware bug flags. Kept separately for each CPU.
++ * Members of this structure are referenced in head.S, so think twice
++ * before touching them. [mj]
++ */
++
++struct cpuinfo_x86 {
++ __u8 x86; /* CPU family */
++ __u8 x86_vendor; /* CPU vendor */
++ __u8 x86_model;
++ __u8 x86_mask;
++ char wp_works_ok; /* It doesn't on 386's */
++ char hlt_works_ok; /* Problems on some 486Dx4's and old 386's */
++ char hard_math;
++ char rfu;
++ int cpuid_level; /* Maximum supported CPUID level, -1=no CPUID */
++ unsigned long x86_capability[NCAPINTS];
++ char x86_vendor_id[16];
++ char x86_model_id[64];
++ int x86_cache_size; /* in KB - valid for CPUS which support this
++ call */
++ int x86_cache_alignment; /* In bytes */
++ char fdiv_bug;
++ char f00f_bug;
++ char coma_bug;
++ char pad0;
++ int x86_power;
++ unsigned long loops_per_jiffy;
++#ifdef CONFIG_SMP
++ cpumask_t llc_shared_map; /* cpus sharing the last level cache */
++#endif
++ unsigned char x86_max_cores; /* cpuid returned max cores value */
++ unsigned char booted_cores; /* number of cores as seen by OS */
++ unsigned char apicid;
++} __attribute__((__aligned__(SMP_CACHE_BYTES)));
++
++#define X86_VENDOR_INTEL 0
++#define X86_VENDOR_CYRIX 1
++#define X86_VENDOR_AMD 2
++#define X86_VENDOR_UMC 3
++#define X86_VENDOR_NEXGEN 4
++#define X86_VENDOR_CENTAUR 5
++#define X86_VENDOR_RISE 6
++#define X86_VENDOR_TRANSMETA 7
++#define X86_VENDOR_NSC 8
++#define X86_VENDOR_NUM 9
++#define X86_VENDOR_UNKNOWN 0xff
++
++/*
++ * capabilities of CPUs
++ */
++
++extern struct cpuinfo_x86 boot_cpu_data;
++extern struct cpuinfo_x86 new_cpu_data;
++#ifndef CONFIG_X86_NO_TSS
++extern struct tss_struct doublefault_tss;
++DECLARE_PER_CPU(struct tss_struct, init_tss);
++#endif
++
++#ifdef CONFIG_SMP
++extern struct cpuinfo_x86 cpu_data[];
++#define current_cpu_data cpu_data[smp_processor_id()]
++#else
++#define cpu_data (&boot_cpu_data)
++#define current_cpu_data boot_cpu_data
++#endif
++
++extern int phys_proc_id[NR_CPUS];
++extern int cpu_core_id[NR_CPUS];
++extern int cpu_llc_id[NR_CPUS];
++extern char ignore_fpu_irq;
++
++extern void identify_cpu(struct cpuinfo_x86 *);
++extern void print_cpu_info(struct cpuinfo_x86 *);
++extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
++
++#ifdef CONFIG_X86_HT
++extern void detect_ht(struct cpuinfo_x86 *c);
++#else
++static inline void detect_ht(struct cpuinfo_x86 *c) {}
++#endif
++
++/*
++ * EFLAGS bits
++ */
++#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
++#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
++#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
++#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
++#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
++#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
++#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
++#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
++#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
++#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
++#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
++#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
++#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
++#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
++#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
++#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
++#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
++
++/*
++ * Generic CPUID function
++ * clear %ecx since some cpus (Cyrix MII) do not set or clear %ecx
++ * resulting in stale register contents being returned.
++ */
++static inline void cpuid(unsigned int op, unsigned int *eax, unsigned int *ebx, unsigned int *ecx, unsigned int *edx)
++{
++ __asm__(XEN_CPUID
++ : "=a" (*eax),
++ "=b" (*ebx),
++ "=c" (*ecx),
++ "=d" (*edx)
++ : "0" (op), "c"(0));
++}
++
++/* Some CPUID calls want 'count' to be placed in ecx */
++static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
++ int *edx)
++{
++ __asm__(XEN_CPUID
++ : "=a" (*eax),
++ "=b" (*ebx),
++ "=c" (*ecx),
++ "=d" (*edx)
++ : "0" (op), "c" (count));
++}
++
++/*
++ * CPUID functions returning a single datum
++ */
++static inline unsigned int cpuid_eax(unsigned int op)
++{
++ unsigned int eax;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax)
++ : "0" (op)
++ : "bx", "cx", "dx");
++ return eax;
++}
++static inline unsigned int cpuid_ebx(unsigned int op)
++{
++ unsigned int eax, ebx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=b" (ebx)
++ : "0" (op)
++ : "cx", "dx" );
++ return ebx;
++}
++static inline unsigned int cpuid_ecx(unsigned int op)
++{
++ unsigned int eax, ecx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=c" (ecx)
++ : "0" (op)
++ : "bx", "dx" );
++ return ecx;
++}
++static inline unsigned int cpuid_edx(unsigned int op)
++{
++ unsigned int eax, edx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=d" (edx)
++ : "0" (op)
++ : "bx", "cx");
++ return edx;
++}
++
++#define load_cr3(pgdir) write_cr3(__pa(pgdir))
++
++/*
++ * Intel CPU features in CR4
++ */
++#define X86_CR4_VME 0x0001 /* enable vm86 extensions */
++#define X86_CR4_PVI 0x0002 /* virtual interrupts flag enable */
++#define X86_CR4_TSD 0x0004 /* disable time stamp at ipl 3 */
++#define X86_CR4_DE 0x0008 /* enable debugging extensions */
++#define X86_CR4_PSE 0x0010 /* enable page size extensions */
++#define X86_CR4_PAE 0x0020 /* enable physical address extensions */
++#define X86_CR4_MCE 0x0040 /* Machine check enable */
++#define X86_CR4_PGE 0x0080 /* enable global pages */
++#define X86_CR4_PCE 0x0100 /* enable performance counters at ipl 3 */
++#define X86_CR4_OSFXSR 0x0200 /* enable fast FPU save and restore */
++#define X86_CR4_OSXMMEXCPT 0x0400 /* enable unmasked SSE exceptions */
++
++/*
++ * Save the cr4 feature set we're using (ie
++ * Pentium 4MB enable and PPro Global page
++ * enable), so that any CPU's that boot up
++ * after us can get the correct flags.
++ */
++extern unsigned long mmu_cr4_features;
++
++static inline void set_in_cr4 (unsigned long mask)
++{
++ unsigned cr4;
++ mmu_cr4_features |= mask;
++ cr4 = read_cr4();
++ cr4 |= mask;
++ write_cr4(cr4);
++}
++
++static inline void clear_in_cr4 (unsigned long mask)
++{
++ unsigned cr4;
++ mmu_cr4_features &= ~mask;
++ cr4 = read_cr4();
++ cr4 &= ~mask;
++ write_cr4(cr4);
++}
++
++/*
++ * NSC/Cyrix CPU configuration register indexes
++ */
++
++#define CX86_PCR0 0x20
++#define CX86_GCR 0xb8
++#define CX86_CCR0 0xc0
++#define CX86_CCR1 0xc1
++#define CX86_CCR2 0xc2
++#define CX86_CCR3 0xc3
++#define CX86_CCR4 0xe8
++#define CX86_CCR5 0xe9
++#define CX86_CCR6 0xea
++#define CX86_CCR7 0xeb
++#define CX86_PCR1 0xf0
++#define CX86_DIR0 0xfe
++#define CX86_DIR1 0xff
++#define CX86_ARR_BASE 0xc4
++#define CX86_RCR_BASE 0xdc
++
++/*
++ * NSC/Cyrix CPU indexed register access macros
++ */
++
++#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
++
++#define setCx86(reg, data) do { \
++ outb((reg), 0x22); \
++ outb((data), 0x23); \
++} while (0)
++
++/* Stop speculative execution */
++static inline void sync_core(void)
++{
++ int tmp;
++ asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
++}
++
++static inline void __monitor(const void *eax, unsigned long ecx,
++ unsigned long edx)
++{
++ /* "monitor %eax,%ecx,%edx;" */
++ asm volatile(
++ ".byte 0x0f,0x01,0xc8;"
++ : :"a" (eax), "c" (ecx), "d"(edx));
++}
++
++static inline void __mwait(unsigned long eax, unsigned long ecx)
++{
++ /* "mwait %eax,%ecx;" */
++ asm volatile(
++ ".byte 0x0f,0x01,0xc9;"
++ : :"a" (eax), "c" (ecx));
++}
++
++/* from system description table in BIOS. Mostly for MCA use, but
++others may find it useful. */
++extern unsigned int machine_id;
++extern unsigned int machine_submodel_id;
++extern unsigned int BIOS_revision;
++extern unsigned int mca_pentium_flag;
++
++/* Boot loader type from the setup header */
++extern int bootloader_type;
++
++/*
++ * User space process size: 3GB (default).
++ */
++#define TASK_SIZE (PAGE_OFFSET)
++
++/* This decides where the kernel will search for a free chunk of vm
++ * space during mmap's.
++ */
++#define TASK_UNMAPPED_BASE (PAGE_ALIGN(TASK_SIZE / 3))
++
++#define HAVE_ARCH_PICK_MMAP_LAYOUT
++
++/*
++ * Size of io_bitmap.
++ */
++#define IO_BITMAP_BITS 65536
++#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
++#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
++#ifndef CONFIG_X86_NO_TSS
++#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
++#endif
++#define INVALID_IO_BITMAP_OFFSET 0x8000
++#define INVALID_IO_BITMAP_OFFSET_LAZY 0x9000
++
++struct i387_fsave_struct {
++ long cwd;
++ long swd;
++ long twd;
++ long fip;
++ long fcs;
++ long foo;
++ long fos;
++ long st_space[20]; /* 8*10 bytes for each FP-reg = 80 bytes */
++ long status; /* software status information */
++};
++
++struct i387_fxsave_struct {
++ unsigned short cwd;
++ unsigned short swd;
++ unsigned short twd;
++ unsigned short fop;
++ long fip;
++ long fcs;
++ long foo;
++ long fos;
++ long mxcsr;
++ long mxcsr_mask;
++ long st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */
++ long xmm_space[32]; /* 8*16 bytes for each XMM-reg = 128 bytes */
++ long padding[56];
++} __attribute__ ((aligned (16)));
++
++struct i387_soft_struct {
++ long cwd;
++ long swd;
++ long twd;
++ long fip;
++ long fcs;
++ long foo;
++ long fos;
++ long st_space[20]; /* 8*10 bytes for each FP-reg = 80 bytes */
++ unsigned char ftop, changed, lookahead, no_update, rm, alimit;
++ struct info *info;
++ unsigned long entry_eip;
++};
++
++union i387_union {
++ struct i387_fsave_struct fsave;
++ struct i387_fxsave_struct fxsave;
++ struct i387_soft_struct soft;
++};
++
++typedef struct {
++ unsigned long seg;
++} mm_segment_t;
++
++struct thread_struct;
++
++#ifndef CONFIG_X86_NO_TSS
++struct tss_struct {
++ unsigned short back_link,__blh;
++ unsigned long esp0;
++ unsigned short ss0,__ss0h;
++ unsigned long esp1;
++ unsigned short ss1,__ss1h; /* ss1 is used to cache MSR_IA32_SYSENTER_CS */
++ unsigned long esp2;
++ unsigned short ss2,__ss2h;
++ unsigned long __cr3;
++ unsigned long eip;
++ unsigned long eflags;
++ unsigned long eax,ecx,edx,ebx;
++ unsigned long esp;
++ unsigned long ebp;
++ unsigned long esi;
++ unsigned long edi;
++ unsigned short es, __esh;
++ unsigned short cs, __csh;
++ unsigned short ss, __ssh;
++ unsigned short ds, __dsh;
++ unsigned short fs, __fsh;
++ unsigned short gs, __gsh;
++ unsigned short ldt, __ldth;
++ unsigned short trace, io_bitmap_base;
++ /*
++ * The extra 1 is there because the CPU will access an
++ * additional byte beyond the end of the IO permission
++ * bitmap. The extra byte must be all 1 bits, and must
++ * be within the limit.
++ */
++ unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
++ /*
++ * Cache the current maximum and the last task that used the bitmap:
++ */
++ unsigned long io_bitmap_max;
++ struct thread_struct *io_bitmap_owner;
++ /*
++ * pads the TSS to be cacheline-aligned (size is 0x100)
++ */
++ unsigned long __cacheline_filler[35];
++ /*
++ * .. and then another 0x100 bytes for emergency kernel stack
++ */
++ unsigned long stack[64];
++} __attribute__((packed));
++#endif
++
++#define ARCH_MIN_TASKALIGN 16
++
++struct thread_struct {
++/* cached TLS descriptors. */
++ struct desc_struct tls_array[GDT_ENTRY_TLS_ENTRIES];
++ unsigned long esp0;
++ unsigned long sysenter_cs;
++ unsigned long eip;
++ unsigned long esp;
++ unsigned long fs;
++ unsigned long gs;
++/* Hardware debugging registers */
++ unsigned long debugreg[8]; /* %%db0-7 debug registers */
++/* fault info */
++ unsigned long cr2, trap_no, error_code;
++/* floating point info */
++ union i387_union i387;
++/* virtual 86 mode info */
++ struct vm86_struct __user * vm86_info;
++ unsigned long screen_bitmap;
++ unsigned long v86flags, v86mask, saved_esp0;
++ unsigned int saved_fs, saved_gs;
++/* IO permissions */
++ unsigned long *io_bitmap_ptr;
++ unsigned long iopl;
++/* max allowed port in the bitmap, in bytes: */
++ unsigned long io_bitmap_max;
++};
++
++#define INIT_THREAD { \
++ .vm86_info = NULL, \
++ .sysenter_cs = __KERNEL_CS, \
++ .io_bitmap_ptr = NULL, \
++}
++
++#ifndef CONFIG_X86_NO_TSS
++/*
++ * Note that the .io_bitmap member must be extra-big. This is because
++ * the CPU will access an additional byte beyond the end of the IO
++ * permission bitmap. The extra byte must be all 1 bits, and must
++ * be within the limit.
++ */
++#define INIT_TSS { \
++ .esp0 = sizeof(init_stack) + (long)&init_stack, \
++ .ss0 = __KERNEL_DS, \
++ .ss1 = __KERNEL_CS, \
++ .io_bitmap_base = INVALID_IO_BITMAP_OFFSET, \
++ .io_bitmap = { [ 0 ... IO_BITMAP_LONGS] = ~0 }, \
++}
++
++static inline void __load_esp0(struct tss_struct *tss, struct thread_struct *thread)
++{
++ tss->esp0 = thread->esp0;
++ /* This can only happen when SEP is enabled, no need to test "SEP"arately */
++ if (unlikely(tss->ss1 != thread->sysenter_cs)) {
++ tss->ss1 = thread->sysenter_cs;
++ wrmsr(MSR_IA32_SYSENTER_CS, thread->sysenter_cs, 0);
++ }
++}
++#define load_esp0(tss, thread) \
++ __load_esp0(tss, thread)
++#else
++#define load_esp0(tss, thread) \
++ HYPERVISOR_stack_switch(__KERNEL_DS, (thread)->esp0)
++#endif
++
++#define start_thread(regs, new_eip, new_esp) do { \
++ __asm__("movl %0,%%fs ; movl %0,%%gs": :"r" (0)); \
++ set_fs(USER_DS); \
++ regs->xds = __USER_DS; \
++ regs->xes = __USER_DS; \
++ regs->xss = __USER_DS; \
++ regs->xcs = __USER_CS; \
++ regs->eip = new_eip; \
++ regs->esp = new_esp; \
++} while (0)
++
++/*
++ * These special macros can be used to get or set a debugging register
++ */
++#define get_debugreg(var, register) \
++ (var) = HYPERVISOR_get_debugreg((register))
++#define set_debugreg(value, register) \
++ HYPERVISOR_set_debugreg((register), (value))
++
++/*
++ * Set IOPL bits in EFLAGS from given mask
++ */
++static inline void set_iopl_mask(unsigned mask)
++{
++ struct physdev_set_iopl set_iopl;
++
++ /* Force the change at ring 0. */
++ set_iopl.iopl = (mask == 0) ? 1 : (mask >> 12) & 3;
++ HYPERVISOR_physdev_op(PHYSDEVOP_set_iopl, &set_iopl);
++}
++
++/* Forward declaration, a strange C thing */
++struct task_struct;
++struct mm_struct;
++
++/* Free all resources held by a thread. */
++extern void release_thread(struct task_struct *);
++
++/* Prepare to copy thread state - unlazy all lazy status */
++extern void prepare_to_copy(struct task_struct *tsk);
++
++/*
++ * create a kernel thread without removing it from tasklists
++ */
++extern int kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
++
++extern unsigned long thread_saved_pc(struct task_struct *tsk);
++void show_trace(struct task_struct *task, unsigned long *stack);
++
++unsigned long get_wchan(struct task_struct *p);
++
++#define THREAD_SIZE_LONGS (THREAD_SIZE/sizeof(unsigned long))
++#define KSTK_TOP(info) \
++({ \
++ unsigned long *__ptr = (unsigned long *)(info); \
++ (unsigned long)(&__ptr[THREAD_SIZE_LONGS]); \
++})
++
++/*
++ * The below -8 is to reserve 8 bytes on top of the ring0 stack.
++ * This is necessary to guarantee that the entire "struct pt_regs"
++ * is accessable even if the CPU haven't stored the SS/ESP registers
++ * on the stack (interrupt gate does not save these registers
++ * when switching to the same priv ring).
++ * Therefore beware: accessing the xss/esp fields of the
++ * "struct pt_regs" is possible, but they may contain the
++ * completely wrong values.
++ */
++#define task_pt_regs(task) \
++({ \
++ struct pt_regs *__regs__; \
++ __regs__ = (struct pt_regs *)(KSTK_TOP(task_stack_page(task))-8); \
++ __regs__ - 1; \
++})
++
++#define KSTK_EIP(task) (task_pt_regs(task)->eip)
++#define KSTK_ESP(task) (task_pt_regs(task)->esp)
++
++
++struct microcode_header {
++ unsigned int hdrver;
++ unsigned int rev;
++ unsigned int date;
++ unsigned int sig;
++ unsigned int cksum;
++ unsigned int ldrver;
++ unsigned int pf;
++ unsigned int datasize;
++ unsigned int totalsize;
++ unsigned int reserved[3];
++};
++
++struct microcode {
++ struct microcode_header hdr;
++ unsigned int bits[0];
++};
++
++typedef struct microcode microcode_t;
++typedef struct microcode_header microcode_header_t;
++
++/* microcode format is extended from prescott processors */
++struct extended_signature {
++ unsigned int sig;
++ unsigned int pf;
++ unsigned int cksum;
++};
++
++struct extended_sigtable {
++ unsigned int count;
++ unsigned int cksum;
++ unsigned int reserved[3];
++ struct extended_signature sigs[0];
++};
++
++/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
++static inline void rep_nop(void)
++{
++ __asm__ __volatile__("rep;nop": : :"memory");
++}
++
++#define cpu_relax() rep_nop()
++
++/* generic versions from gas */
++#define GENERIC_NOP1 ".byte 0x90\n"
++#define GENERIC_NOP2 ".byte 0x89,0xf6\n"
++#define GENERIC_NOP3 ".byte 0x8d,0x76,0x00\n"
++#define GENERIC_NOP4 ".byte 0x8d,0x74,0x26,0x00\n"
++#define GENERIC_NOP5 GENERIC_NOP1 GENERIC_NOP4
++#define GENERIC_NOP6 ".byte 0x8d,0xb6,0x00,0x00,0x00,0x00\n"
++#define GENERIC_NOP7 ".byte 0x8d,0xb4,0x26,0x00,0x00,0x00,0x00\n"
++#define GENERIC_NOP8 GENERIC_NOP1 GENERIC_NOP7
++
++/* Opteron nops */
++#define K8_NOP1 GENERIC_NOP1
++#define K8_NOP2 ".byte 0x66,0x90\n"
++#define K8_NOP3 ".byte 0x66,0x66,0x90\n"
++#define K8_NOP4 ".byte 0x66,0x66,0x66,0x90\n"
++#define K8_NOP5 K8_NOP3 K8_NOP2
++#define K8_NOP6 K8_NOP3 K8_NOP3
++#define K8_NOP7 K8_NOP4 K8_NOP3
++#define K8_NOP8 K8_NOP4 K8_NOP4
++
++/* K7 nops */
++/* uses eax dependencies (arbitary choice) */
++#define K7_NOP1 GENERIC_NOP1
++#define K7_NOP2 ".byte 0x8b,0xc0\n"
++#define K7_NOP3 ".byte 0x8d,0x04,0x20\n"
++#define K7_NOP4 ".byte 0x8d,0x44,0x20,0x00\n"
++#define K7_NOP5 K7_NOP4 ASM_NOP1
++#define K7_NOP6 ".byte 0x8d,0x80,0,0,0,0\n"
++#define K7_NOP7 ".byte 0x8D,0x04,0x05,0,0,0,0\n"
++#define K7_NOP8 K7_NOP7 ASM_NOP1
++
++#ifdef CONFIG_MK8
++#define ASM_NOP1 K8_NOP1
++#define ASM_NOP2 K8_NOP2
++#define ASM_NOP3 K8_NOP3
++#define ASM_NOP4 K8_NOP4
++#define ASM_NOP5 K8_NOP5
++#define ASM_NOP6 K8_NOP6
++#define ASM_NOP7 K8_NOP7
++#define ASM_NOP8 K8_NOP8
++#elif defined(CONFIG_MK7)
++#define ASM_NOP1 K7_NOP1
++#define ASM_NOP2 K7_NOP2
++#define ASM_NOP3 K7_NOP3
++#define ASM_NOP4 K7_NOP4
++#define ASM_NOP5 K7_NOP5
++#define ASM_NOP6 K7_NOP6
++#define ASM_NOP7 K7_NOP7
++#define ASM_NOP8 K7_NOP8
++#else
++#define ASM_NOP1 GENERIC_NOP1
++#define ASM_NOP2 GENERIC_NOP2
++#define ASM_NOP3 GENERIC_NOP3
++#define ASM_NOP4 GENERIC_NOP4
++#define ASM_NOP5 GENERIC_NOP5
++#define ASM_NOP6 GENERIC_NOP6
++#define ASM_NOP7 GENERIC_NOP7
++#define ASM_NOP8 GENERIC_NOP8
++#endif
++
++#define ASM_NOP_MAX 8
++
++/* Prefetch instructions for Pentium III and AMD Athlon */
++/* It's not worth to care about 3dnow! prefetches for the K6
++ because they are microcoded there and very slow.
++ However we don't do prefetches for pre XP Athlons currently
++ That should be fixed. */
++#define ARCH_HAS_PREFETCH
++static inline void prefetch(const void *x)
++{
++ alternative_input(ASM_NOP4,
++ "prefetchnta (%1)",
++ X86_FEATURE_XMM,
++ "r" (x));
++}
++
++#define ARCH_HAS_PREFETCH
++#define ARCH_HAS_PREFETCHW
++#define ARCH_HAS_SPINLOCK_PREFETCH
++
++/* 3dnow! prefetch to get an exclusive cache line. Useful for
++ spinlocks to avoid one state transition in the cache coherency protocol. */
++static inline void prefetchw(const void *x)
++{
++ alternative_input(ASM_NOP4,
++ "prefetchw (%1)",
++ X86_FEATURE_3DNOW,
++ "r" (x));
++}
++#define spin_lock_prefetch(x) prefetchw(x)
++
++extern void select_idle_routine(const struct cpuinfo_x86 *c);
++
++#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
++
++extern unsigned long boot_option_idle_override;
++extern void enable_sep_cpu(void);
++extern int sysenter_setup(void);
++
++#ifdef CONFIG_MTRR
++extern void mtrr_ap_init(void);
++extern void mtrr_bp_init(void);
++#else
++#define mtrr_ap_init() do {} while (0)
++#define mtrr_bp_init() do {} while (0)
++#endif
++
++#ifdef CONFIG_X86_MCE
++extern void mcheck_init(struct cpuinfo_x86 *c);
++#else
++#define mcheck_init(c) do {} while(0)
++#endif
++
++#endif /* __ASM_I386_PROCESSOR_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/ptrace.h new/include/asm-i386/mach-xen/asm/ptrace.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/ptrace.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/ptrace.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,90 @@
++#ifndef _I386_PTRACE_H
++#define _I386_PTRACE_H
++
++#define EBX 0
++#define ECX 1
++#define EDX 2
++#define ESI 3
++#define EDI 4
++#define EBP 5
++#define EAX 6
++#define DS 7
++#define ES 8
++#define FS 9
++#define GS 10
++#define ORIG_EAX 11
++#define EIP 12
++#define CS 13
++#define EFL 14
++#define UESP 15
++#define SS 16
++#define FRAME_SIZE 17
++
++/* this struct defines the way the registers are stored on the
++ stack during a system call. */
++
++struct pt_regs {
++ long ebx;
++ long ecx;
++ long edx;
++ long esi;
++ long edi;
++ long ebp;
++ long eax;
++ int xds;
++ int xes;
++ long orig_eax;
++ long eip;
++ int xcs;
++ long eflags;
++ long esp;
++ int xss;
++};
++
++/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
++#define PTRACE_GETREGS 12
++#define PTRACE_SETREGS 13
++#define PTRACE_GETFPREGS 14
++#define PTRACE_SETFPREGS 15
++#define PTRACE_GETFPXREGS 18
++#define PTRACE_SETFPXREGS 19
++
++#define PTRACE_OLDSETOPTIONS 21
++
++#define PTRACE_GET_THREAD_AREA 25
++#define PTRACE_SET_THREAD_AREA 26
++
++#define PTRACE_SYSEMU 31
++#define PTRACE_SYSEMU_SINGLESTEP 32
++
++#ifdef __KERNEL__
++
++#include <asm/vm86.h>
++
++struct task_struct;
++extern void send_sigtrap(struct task_struct *tsk, struct pt_regs *regs, int error_code);
++
++/*
++ * user_mode_vm(regs) determines whether a register set came from user mode.
++ * This is true if V8086 mode was enabled OR if the register set was from
++ * protected mode with RPL-3 CS value. This tricky test checks that with
++ * one comparison. Many places in the kernel can bypass this full check
++ * if they have already ruled out V8086 mode, so user_mode(regs) can be used.
++ */
++static inline int user_mode(struct pt_regs *regs)
++{
++ return (regs->xcs & 2) != 0;
++}
++static inline int user_mode_vm(struct pt_regs *regs)
++{
++ return ((regs->xcs & 2) | (regs->eflags & VM_MASK)) != 0;
++}
++#define instruction_pointer(regs) ((regs)->eip)
++#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
++extern unsigned long profile_pc(struct pt_regs *regs);
++#else
++#define profile_pc(regs) instruction_pointer(regs)
++#endif
++#endif /* __KERNEL__ */
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/scatterlist.h new/include/asm-i386/mach-xen/asm/scatterlist.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/scatterlist.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/scatterlist.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,22 @@
++#ifndef _I386_SCATTERLIST_H
++#define _I386_SCATTERLIST_H
++
++struct scatterlist {
++ struct page *page;
++ unsigned int offset;
++ unsigned int length;
++ dma_addr_t dma_address;
++ unsigned int dma_length;
++};
++
++/* These macros should be used after a pci_map_sg call has been done
++ * to get bus addresses of each of the SG entries and their lengths.
++ * You should only work with the number of sg entries pci_map_sg
++ * returns.
++ */
++#define sg_dma_address(sg) ((sg)->dma_address)
++#define sg_dma_len(sg) ((sg)->dma_length)
++
++#define ISA_DMA_THRESHOLD (0x00ffffff)
++
++#endif /* !(_I386_SCATTERLIST_H) */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/segment.h new/include/asm-i386/mach-xen/asm/segment.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/segment.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/segment.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,117 @@
++#ifndef _ASM_SEGMENT_H
++#define _ASM_SEGMENT_H
++
++/*
++ * The layout of the per-CPU GDT under Linux:
++ *
++ * 0 - null
++ * 1 - reserved
++ * 2 - reserved
++ * 3 - reserved
++ *
++ * 4 - unused <==== new cacheline
++ * 5 - unused
++ *
++ * ------- start of TLS (Thread-Local Storage) segments:
++ *
++ * 6 - TLS segment #1 [ glibc's TLS segment ]
++ * 7 - TLS segment #2 [ Wine's %fs Win32 segment ]
++ * 8 - TLS segment #3
++ * 9 - reserved
++ * 10 - reserved
++ * 11 - reserved
++ *
++ * ------- start of kernel segments:
++ *
++ * 12 - kernel code segment <==== new cacheline
++ * 13 - kernel data segment
++ * 14 - default user CS
++ * 15 - default user DS
++ * 16 - TSS
++ * 17 - LDT
++ * 18 - PNPBIOS support (16->32 gate)
++ * 19 - PNPBIOS support
++ * 20 - PNPBIOS support
++ * 21 - PNPBIOS support
++ * 22 - PNPBIOS support
++ * 23 - APM BIOS support
++ * 24 - APM BIOS support
++ * 25 - APM BIOS support
++ *
++ * 26 - ESPFIX small SS
++ * 27 - unused
++ * 28 - unused
++ * 29 - unused
++ * 30 - unused
++ * 31 - TSS for double fault handler
++ */
++#define GDT_ENTRY_TLS_ENTRIES 3
++#define GDT_ENTRY_TLS_MIN 6
++#define GDT_ENTRY_TLS_MAX (GDT_ENTRY_TLS_MIN + GDT_ENTRY_TLS_ENTRIES - 1)
++
++#define TLS_SIZE (GDT_ENTRY_TLS_ENTRIES * 8)
++
++#define GDT_ENTRY_DEFAULT_USER_CS 14
++#define __USER_CS (GDT_ENTRY_DEFAULT_USER_CS * 8 + 3)
++
++#define GDT_ENTRY_DEFAULT_USER_DS 15
++#define __USER_DS (GDT_ENTRY_DEFAULT_USER_DS * 8 + 3)
++
++#define GDT_ENTRY_KERNEL_BASE 12
++
++#define GDT_ENTRY_KERNEL_CS (GDT_ENTRY_KERNEL_BASE + 0)
++#define __KERNEL_CS (GDT_ENTRY_KERNEL_CS * 8)
++#define GET_KERNEL_CS() (__KERNEL_CS | (xen_feature(XENFEAT_supervisor_mode_kernel)?0:1) )
++
++#define GDT_ENTRY_KERNEL_DS (GDT_ENTRY_KERNEL_BASE + 1)
++#define __KERNEL_DS (GDT_ENTRY_KERNEL_DS * 8)
++#define GET_KERNEL_DS() (__KERNEL_DS | (xen_feature(XENFEAT_supervisor_mode_kernel)?0:1) )
++
++#define GDT_ENTRY_TSS (GDT_ENTRY_KERNEL_BASE + 4)
++#define GDT_ENTRY_LDT (GDT_ENTRY_KERNEL_BASE + 5)
++
++#define GDT_ENTRY_PNPBIOS_BASE (GDT_ENTRY_KERNEL_BASE + 6)
++#define GDT_ENTRY_APMBIOS_BASE (GDT_ENTRY_KERNEL_BASE + 11)
++
++#define GDT_ENTRY_ESPFIX_SS (GDT_ENTRY_KERNEL_BASE + 14)
++#define __ESPFIX_SS (GDT_ENTRY_ESPFIX_SS * 8)
++
++#define GDT_ENTRY_DOUBLEFAULT_TSS 31
++
++/*
++ * The GDT has 32 entries
++ */
++#define GDT_ENTRIES 32
++
++#define GDT_SIZE (GDT_ENTRIES * 8)
++
++/* Simple and small GDT entries for booting only */
++
++#define GDT_ENTRY_BOOT_CS 2
++#define __BOOT_CS (GDT_ENTRY_BOOT_CS * 8)
++
++#define GDT_ENTRY_BOOT_DS (GDT_ENTRY_BOOT_CS + 1)
++#define __BOOT_DS (GDT_ENTRY_BOOT_DS * 8)
++
++/* The PnP BIOS entries in the GDT */
++#define GDT_ENTRY_PNPBIOS_CS32 (GDT_ENTRY_PNPBIOS_BASE + 0)
++#define GDT_ENTRY_PNPBIOS_CS16 (GDT_ENTRY_PNPBIOS_BASE + 1)
++#define GDT_ENTRY_PNPBIOS_DS (GDT_ENTRY_PNPBIOS_BASE + 2)
++#define GDT_ENTRY_PNPBIOS_TS1 (GDT_ENTRY_PNPBIOS_BASE + 3)
++#define GDT_ENTRY_PNPBIOS_TS2 (GDT_ENTRY_PNPBIOS_BASE + 4)
++
++/* The PnP BIOS selectors */
++#define PNP_CS32 (GDT_ENTRY_PNPBIOS_CS32 * 8) /* segment for calling fn */
++#define PNP_CS16 (GDT_ENTRY_PNPBIOS_CS16 * 8) /* code segment for BIOS */
++#define PNP_DS (GDT_ENTRY_PNPBIOS_DS * 8) /* data segment for BIOS */
++#define PNP_TS1 (GDT_ENTRY_PNPBIOS_TS1 * 8) /* transfer data segment */
++#define PNP_TS2 (GDT_ENTRY_PNPBIOS_TS2 * 8) /* another data segment */
++
++/*
++ * The interrupt descriptor table has room for 256 idt's,
++ * the global descriptor table is dependent on the number
++ * of tasks we can have..
++ */
++#define IDT_ENTRIES 256
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/setup.h new/include/asm-i386/mach-xen/asm/setup.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/setup.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/setup.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,64 @@
++/*
++ * Just a place holder. We don't want to have to test x86 before
++ * we include stuff
++ */
++
++#ifndef _i386_SETUP_H
++#define _i386_SETUP_H
++
++#include <linux/pfn.h>
++
++/*
++ * Reserved space for vmalloc and iomap - defined in asm/page.h
++ */
++#define MAXMEM_PFN PFN_DOWN(MAXMEM)
++#define MAX_NONPAE_PFN (1 << 20)
++
++#define PARAM_SIZE 4096
++#define COMMAND_LINE_SIZE 256
++
++#define OLD_CL_MAGIC_ADDR 0x90020
++#define OLD_CL_MAGIC 0xA33F
++#define OLD_CL_BASE_ADDR 0x90000
++#define OLD_CL_OFFSET 0x90022
++#define NEW_CL_POINTER 0x228 /* Relative to real mode data */
++
++#ifndef __ASSEMBLY__
++/*
++ * This is set up by the setup-routine at boot-time
++ */
++extern unsigned char boot_params[PARAM_SIZE];
++
++#define PARAM (boot_params)
++#define SCREEN_INFO (*(struct screen_info *) (PARAM+0))
++#define EXT_MEM_K (*(unsigned short *) (PARAM+2))
++#define ALT_MEM_K (*(unsigned long *) (PARAM+0x1e0))
++#define E820_MAP_NR (*(char*) (PARAM+E820NR))
++#define E820_MAP ((struct e820entry *) (PARAM+E820MAP))
++#define APM_BIOS_INFO (*(struct apm_bios_info *) (PARAM+0x40))
++#define IST_INFO (*(struct ist_info *) (PARAM+0x60))
++#define DRIVE_INFO (*(struct drive_info_struct *) (PARAM+0x80))
++#define SYS_DESC_TABLE (*(struct sys_desc_table_struct*)(PARAM+0xa0))
++#define EFI_SYSTAB ((efi_system_table_t *) *((unsigned long *)(PARAM+0x1c4)))
++#define EFI_MEMDESC_SIZE (*((unsigned long *) (PARAM+0x1c8)))
++#define EFI_MEMDESC_VERSION (*((unsigned long *) (PARAM+0x1cc)))
++#define EFI_MEMMAP ((void *) *((unsigned long *)(PARAM+0x1d0)))
++#define EFI_MEMMAP_SIZE (*((unsigned long *) (PARAM+0x1d4)))
++#define MOUNT_ROOT_RDONLY (*(unsigned short *) (PARAM+0x1F2))
++#define RAMDISK_FLAGS (*(unsigned short *) (PARAM+0x1F8))
++#define VIDEO_MODE (*(unsigned short *) (PARAM+0x1FA))
++#define ORIG_ROOT_DEV (*(unsigned short *) (PARAM+0x1FC))
++#define AUX_DEVICE_INFO (*(unsigned char *) (PARAM+0x1FF))
++#define LOADER_TYPE (*(unsigned char *) (PARAM+0x210))
++#define KERNEL_START (*(unsigned long *) (PARAM+0x214))
++#define INITRD_START (__pa(xen_start_info->mod_start))
++#define INITRD_SIZE (xen_start_info->mod_len)
++#define EDID_INFO (*(struct edid_info *) (PARAM+0x440))
++#define EDD_NR (*(unsigned char *) (PARAM+EDDNR))
++#define EDD_MBR_SIG_NR (*(unsigned char *) (PARAM+EDD_MBR_SIG_NR_BUF))
++#define EDD_MBR_SIGNATURE ((unsigned int *) (PARAM+EDD_MBR_SIG_BUF))
++#define EDD_BUF ((struct edd_info *) (PARAM+EDDBUF))
++
++#endif /* __ASSEMBLY__ */
++
++#endif /* _i386_SETUP_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/smp.h new/include/asm-i386/mach-xen/asm/smp.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/smp.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/smp.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,104 @@
++#ifndef __ASM_SMP_H
++#define __ASM_SMP_H
++
++/*
++ * We need the APIC definitions automatically as part of 'smp.h'
++ */
++#ifndef __ASSEMBLY__
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/threads.h>
++#include <linux/cpumask.h>
++#endif
++
++#ifdef CONFIG_X86_LOCAL_APIC
++#ifndef __ASSEMBLY__
++#include <asm/fixmap.h>
++#include <asm/bitops.h>
++#include <asm/mpspec.h>
++#ifdef CONFIG_X86_IO_APIC
++#include <asm/io_apic.h>
++#endif
++#include <asm/apic.h>
++#endif
++#endif
++
++#define BAD_APICID 0xFFu
++#ifdef CONFIG_SMP
++#ifndef __ASSEMBLY__
++
++/*
++ * Private routines/data
++ */
++
++extern void smp_alloc_memory(void);
++extern int pic_mode;
++extern int smp_num_siblings;
++extern cpumask_t cpu_sibling_map[];
++extern cpumask_t cpu_core_map[];
++
++extern void (*mtrr_hook) (void);
++extern void zap_low_mappings (void);
++extern void lock_ipi_call_lock(void);
++extern void unlock_ipi_call_lock(void);
++
++#define MAX_APICID 256
++extern u8 x86_cpu_to_apicid[];
++
++#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu]
++
++#ifdef CONFIG_HOTPLUG_CPU
++extern void cpu_exit_clear(void);
++extern void cpu_uninit(void);
++#endif
++
++/*
++ * This function is needed by all SMP systems. It must _always_ be valid
++ * from the initial startup. We map APIC_BASE very early in page_setup(),
++ * so this is correct in the x86 case.
++ */
++#define raw_smp_processor_id() (current_thread_info()->cpu)
++
++extern cpumask_t cpu_possible_map;
++#define cpu_callin_map cpu_possible_map
++
++/* We don't mark CPUs online until __cpu_up(), so we need another measure */
++static inline int num_booting_cpus(void)
++{
++ return cpus_weight(cpu_possible_map);
++}
++
++#ifdef CONFIG_X86_LOCAL_APIC
++
++#ifdef APIC_DEFINITION
++extern int hard_smp_processor_id(void);
++#else
++#include <mach_apicdef.h>
++static inline int hard_smp_processor_id(void)
++{
++ /* we don't want to mark this access volatile - bad code generation */
++ return GET_APIC_ID(*(unsigned long *)(APIC_BASE+APIC_ID));
++}
++#endif
++
++static __inline int logical_smp_processor_id(void)
++{
++ /* we don't want to mark this access volatile - bad code generation */
++ return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
++}
++
++#endif
++
++extern int __cpu_disable(void);
++extern void __cpu_die(unsigned int cpu);
++extern void prefill_possible_map(void);
++#endif /* !__ASSEMBLY__ */
++
++#else /* CONFIG_SMP */
++
++#define cpu_physical_id(cpu) boot_cpu_physical_apicid
++
++#define NO_PROC_ID 0xFF /* No processor magic marker */
++
++#endif
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/spinlock.h new/include/asm-i386/mach-xen/asm/spinlock.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/spinlock.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/spinlock.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,200 @@
++#ifndef __ASM_SPINLOCK_H
++#define __ASM_SPINLOCK_H
++
++#include <asm/atomic.h>
++#include <asm/rwlock.h>
++#include <asm/page.h>
++#include <linux/config.h>
++#include <linux/compiler.h>
++
++/*
++ * Your basic SMP spinlocks, allowing only a single CPU anywhere
++ *
++ * Simple spin lock operations. There are two variants, one clears IRQ's
++ * on the local processor, one does not.
++ *
++ * We make no fairness assumptions. They have a cost.
++ *
++ * (the type definitions are in asm/spinlock_types.h)
++ */
++
++#define __raw_spin_is_locked(x) \
++ (*(volatile signed char *)(&(x)->slock) <= 0)
++
++#define __raw_spin_lock_string \
++ "\n1:\t" \
++ "lock ; decb %0\n\t" \
++ "jns 3f\n" \
++ "2:\t" \
++ "rep;nop\n\t" \
++ "cmpb $0,%0\n\t" \
++ "jle 2b\n\t" \
++ "jmp 1b\n" \
++ "3:\n\t"
++
++#define __raw_spin_lock_string_flags \
++ "\n1:\t" \
++ "lock ; decb %0\n\t" \
++ "jns 5f\n" \
++ "2:\t" \
++ "testl $0x200, %1\n\t" \
++ "jz 4f\n\t" \
++ "#sti\n" \
++ "3:\t" \
++ "rep;nop\n\t" \
++ "cmpb $0, %0\n\t" \
++ "jle 3b\n\t" \
++ "#cli\n\t" \
++ "jmp 1b\n" \
++ "4:\t" \
++ "rep;nop\n\t" \
++ "cmpb $0, %0\n\t" \
++ "jg 1b\n\t" \
++ "jmp 4b\n" \
++ "5:\n\t"
++
++#define __raw_spin_lock_string_up \
++ "\n\tdecb %0"
++
++static inline void __raw_spin_lock(raw_spinlock_t *lock)
++{
++ alternative_smp(
++ __raw_spin_lock_string,
++ __raw_spin_lock_string_up,
++ "=m" (lock->slock) : : "memory");
++}
++
++static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
++{
++ alternative_smp(
++ __raw_spin_lock_string_flags,
++ __raw_spin_lock_string_up,
++ "=m" (lock->slock) : "r" (flags) : "memory");
++}
++
++static inline int __raw_spin_trylock(raw_spinlock_t *lock)
++{
++ char oldval;
++ __asm__ __volatile__(
++ "xchgb %b0,%1"
++ :"=q" (oldval), "=m" (lock->slock)
++ :"0" (0) : "memory");
++ return oldval > 0;
++}
++
++/*
++ * __raw_spin_unlock based on writing $1 to the low byte.
++ * This method works. Despite all the confusion.
++ * (except on PPro SMP or if we are using OOSTORE, so we use xchgb there)
++ * (PPro errata 66, 92)
++ */
++
++#if !defined(CONFIG_X86_OOSTORE) && !defined(CONFIG_X86_PPRO_FENCE)
++
++#define __raw_spin_unlock_string \
++ "movb $1,%0" \
++ :"=m" (lock->slock) : : "memory"
++
++
++static inline void __raw_spin_unlock(raw_spinlock_t *lock)
++{
++ __asm__ __volatile__(
++ __raw_spin_unlock_string
++ );
++}
++
++#else
++
++#define __raw_spin_unlock_string \
++ "xchgb %b0, %1" \
++ :"=q" (oldval), "=m" (lock->slock) \
++ :"0" (oldval) : "memory"
++
++static inline void __raw_spin_unlock(raw_spinlock_t *lock)
++{
++ char oldval = 1;
++
++ __asm__ __volatile__(
++ __raw_spin_unlock_string
++ );
++}
++
++#endif
++
++#define __raw_spin_unlock_wait(lock) \
++ do { while (__raw_spin_is_locked(lock)) cpu_relax(); } while (0)
++
++/*
++ * Read-write spinlocks, allowing multiple readers
++ * but only one writer.
++ *
++ * NOTE! it is quite common to have readers in interrupts
++ * but no interrupt writers. For those circumstances we
++ * can "mix" irq-safe locks - any writer needs to get a
++ * irq-safe write-lock, but readers can get non-irqsafe
++ * read-locks.
++ *
++ * On x86, we implement read-write locks as a 32-bit counter
++ * with the high bit (sign) being the "contended" bit.
++ *
++ * The inline assembly is non-obvious. Think about it.
++ *
++ * Changed to use the same technique as rw semaphores. See
++ * semaphore.h for details. -ben
++ *
++ * the helpers are in arch/i386/kernel/semaphore.c
++ */
++
++/**
++ * read_can_lock - would read_trylock() succeed?
++ * @lock: the rwlock in question.
++ */
++#define __raw_read_can_lock(x) ((int)(x)->lock > 0)
++
++/**
++ * write_can_lock - would write_trylock() succeed?
++ * @lock: the rwlock in question.
++ */
++#define __raw_write_can_lock(x) ((x)->lock == RW_LOCK_BIAS)
++
++static inline void __raw_read_lock(raw_rwlock_t *rw)
++{
++ __build_read_lock(rw, "__read_lock_failed");
++}
++
++static inline void __raw_write_lock(raw_rwlock_t *rw)
++{
++ __build_write_lock(rw, "__write_lock_failed");
++}
++
++static inline int __raw_read_trylock(raw_rwlock_t *lock)
++{
++ atomic_t *count = (atomic_t *)lock;
++ atomic_dec(count);
++ if (atomic_read(count) >= 0)
++ return 1;
++ atomic_inc(count);
++ return 0;
++}
++
++static inline int __raw_write_trylock(raw_rwlock_t *lock)
++{
++ atomic_t *count = (atomic_t *)lock;
++ if (atomic_sub_and_test(RW_LOCK_BIAS, count))
++ return 1;
++ atomic_add(RW_LOCK_BIAS, count);
++ return 0;
++}
++
++static inline void __raw_read_unlock(raw_rwlock_t *rw)
++{
++ asm volatile(LOCK_PREFIX "incl %0" :"=m" (rw->lock) : : "memory");
++}
++
++static inline void __raw_write_unlock(raw_rwlock_t *rw)
++{
++ asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
++ : "=m" (rw->lock) : : "memory");
++}
++
++#endif /* __ASM_SPINLOCK_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/swiotlb.h new/include/asm-i386/mach-xen/asm/swiotlb.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/swiotlb.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/swiotlb.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,43 @@
++#ifndef _ASM_SWIOTLB_H
++#define _ASM_SWIOTLB_H 1
++
++#include <linux/config.h>
++
++/* SWIOTLB interface */
++
++extern dma_addr_t swiotlb_map_single(struct device *hwdev, void *ptr, size_t size,
++ int dir);
++extern void swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr,
++ size_t size, int dir);
++extern void swiotlb_sync_single_for_cpu(struct device *hwdev,
++ dma_addr_t dev_addr,
++ size_t size, int dir);
++extern void swiotlb_sync_single_for_device(struct device *hwdev,
++ dma_addr_t dev_addr,
++ size_t size, int dir);
++extern void swiotlb_sync_sg_for_cpu(struct device *hwdev,
++ struct scatterlist *sg, int nelems,
++ int dir);
++extern void swiotlb_sync_sg_for_device(struct device *hwdev,
++ struct scatterlist *sg, int nelems,
++ int dir);
++extern int swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg,
++ int nents, int direction);
++extern void swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg,
++ int nents, int direction);
++extern int swiotlb_dma_mapping_error(dma_addr_t dma_addr);
++extern dma_addr_t swiotlb_map_page(struct device *hwdev, struct page *page,
++ unsigned long offset, size_t size,
++ enum dma_data_direction direction);
++extern void swiotlb_unmap_page(struct device *hwdev, dma_addr_t dma_address,
++ size_t size, enum dma_data_direction direction);
++extern int swiotlb_dma_supported(struct device *hwdev, u64 mask);
++extern void swiotlb_init(void);
++
++#ifdef CONFIG_SWIOTLB
++extern int swiotlb;
++#else
++#define swiotlb 0
++#endif
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/synch_bitops.h new/include/asm-i386/mach-xen/asm/synch_bitops.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/synch_bitops.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/synch_bitops.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,143 @@
++#ifndef __XEN_SYNCH_BITOPS_H__
++#define __XEN_SYNCH_BITOPS_H__
++
++/*
++ * Copyright 1992, Linus Torvalds.
++ * Heavily modified to provide guaranteed strong synchronisation
++ * when communicating with Xen or other guest OSes running on other CPUs.
++ */
++
++#include <linux/config.h>
++
++#define ADDR (*(volatile long *) addr)
++
++static __inline__ void synch_set_bit(int nr, volatile void * addr)
++{
++ __asm__ __volatile__ (
++ "lock btsl %1,%0"
++ : "+m" (ADDR) : "Ir" (nr) : "memory" );
++}
++
++static __inline__ void synch_clear_bit(int nr, volatile void * addr)
++{
++ __asm__ __volatile__ (
++ "lock btrl %1,%0"
++ : "+m" (ADDR) : "Ir" (nr) : "memory" );
++}
++
++static __inline__ void synch_change_bit(int nr, volatile void * addr)
++{
++ __asm__ __volatile__ (
++ "lock btcl %1,%0"
++ : "+m" (ADDR) : "Ir" (nr) : "memory" );
++}
++
++static __inline__ int synch_test_and_set_bit(int nr, volatile void * addr)
++{
++ int oldbit;
++ __asm__ __volatile__ (
++ "lock btsl %2,%1\n\tsbbl %0,%0"
++ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
++ return oldbit;
++}
++
++static __inline__ int synch_test_and_clear_bit(int nr, volatile void * addr)
++{
++ int oldbit;
++ __asm__ __volatile__ (
++ "lock btrl %2,%1\n\tsbbl %0,%0"
++ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
++ return oldbit;
++}
++
++static __inline__ int synch_test_and_change_bit(int nr, volatile void * addr)
++{
++ int oldbit;
++
++ __asm__ __volatile__ (
++ "lock btcl %2,%1\n\tsbbl %0,%0"
++ : "=r" (oldbit), "+m" (ADDR) : "Ir" (nr) : "memory");
++ return oldbit;
++}
++
++struct __synch_xchg_dummy { unsigned long a[100]; };
++#define __synch_xg(x) ((struct __synch_xchg_dummy *)(x))
++
++#define synch_cmpxchg(ptr, old, new) \
++((__typeof__(*(ptr)))__synch_cmpxchg((ptr),\
++ (unsigned long)(old), \
++ (unsigned long)(new), \
++ sizeof(*(ptr))))
++
++static inline unsigned long __synch_cmpxchg(volatile void *ptr,
++ unsigned long old,
++ unsigned long new, int size)
++{
++ unsigned long prev;
++ switch (size) {
++ case 1:
++ __asm__ __volatile__("lock; cmpxchgb %b1,%2"
++ : "=a"(prev)
++ : "q"(new), "m"(*__synch_xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++ case 2:
++ __asm__ __volatile__("lock; cmpxchgw %w1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__synch_xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++#ifdef CONFIG_X86_64
++ case 4:
++ __asm__ __volatile__("lock; cmpxchgl %k1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__synch_xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++ case 8:
++ __asm__ __volatile__("lock; cmpxchgq %1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__synch_xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++#else
++ case 4:
++ __asm__ __volatile__("lock; cmpxchgl %1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__synch_xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++#endif
++ }
++ return old;
++}
++
++static __always_inline int synch_const_test_bit(int nr,
++ const volatile void * addr)
++{
++ return ((1UL << (nr & 31)) &
++ (((const volatile unsigned int *) addr)[nr >> 5])) != 0;
++}
++
++static __inline__ int synch_var_test_bit(int nr, volatile void * addr)
++{
++ int oldbit;
++ __asm__ __volatile__ (
++ "btl %2,%1\n\tsbbl %0,%0"
++ : "=r" (oldbit) : "m" (ADDR), "Ir" (nr) );
++ return oldbit;
++}
++
++#define synch_test_bit(nr,addr) \
++(__builtin_constant_p(nr) ? \
++ synch_const_test_bit((nr),(addr)) : \
++ synch_var_test_bit((nr),(addr)))
++
++#define synch_cmpxchg_subword synch_cmpxchg
++
++#endif /* __XEN_SYNCH_BITOPS_H__ */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/system.h new/include/asm-i386/mach-xen/asm/system.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/system.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/system.h 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,578 @@
++#ifndef __ASM_SYSTEM_H
++#define __ASM_SYSTEM_H
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <asm/segment.h>
++#include <asm/cpufeature.h>
++#include <linux/bitops.h> /* for LOCK_PREFIX */
++#include <asm/synch_bitops.h>
++#include <asm/hypervisor.h>
++
++#ifdef __KERNEL__
++
++#ifdef CONFIG_SMP
++#define __vcpu_id smp_processor_id()
++#else
++#define __vcpu_id 0
++#endif
++
++struct task_struct; /* one of the stranger aspects of C forward declarations.. */
++extern struct task_struct * FASTCALL(__switch_to(struct task_struct *prev, struct task_struct *next));
++
++#define switch_to(prev,next,last) do { \
++ unsigned long esi,edi; \
++ asm volatile("pushl %%ebp\n\t" \
++ "movl %%esp,%0\n\t" /* save ESP */ \
++ "movl %5,%%esp\n\t" /* restore ESP */ \
++ "movl $1f,%1\n\t" /* save EIP */ \
++ "pushl %6\n\t" /* restore EIP */ \
++ "jmp __switch_to\n" \
++ "1:\t" \
++ "popl %%ebp\n\t" \
++ :"=m" (prev->thread.esp),"=m" (prev->thread.eip), \
++ "=a" (last),"=S" (esi),"=D" (edi) \
++ :"m" (next->thread.esp),"m" (next->thread.eip), \
++ "2" (prev), "d" (next)); \
++} while (0)
++
++#define _set_base(addr,base) do { unsigned long __pr; \
++__asm__ __volatile__ ("movw %%dx,%1\n\t" \
++ "rorl $16,%%edx\n\t" \
++ "movb %%dl,%2\n\t" \
++ "movb %%dh,%3" \
++ :"=&d" (__pr) \
++ :"m" (*((addr)+2)), \
++ "m" (*((addr)+4)), \
++ "m" (*((addr)+7)), \
++ "0" (base) \
++ ); } while(0)
++
++#define _set_limit(addr,limit) do { unsigned long __lr; \
++__asm__ __volatile__ ("movw %%dx,%1\n\t" \
++ "rorl $16,%%edx\n\t" \
++ "movb %2,%%dh\n\t" \
++ "andb $0xf0,%%dh\n\t" \
++ "orb %%dh,%%dl\n\t" \
++ "movb %%dl,%2" \
++ :"=&d" (__lr) \
++ :"m" (*(addr)), \
++ "m" (*((addr)+6)), \
++ "0" (limit) \
++ ); } while(0)
++
++#define set_base(ldt,base) _set_base( ((char *)&(ldt)) , (base) )
++#define set_limit(ldt,limit) _set_limit( ((char *)&(ldt)) , ((limit)-1) )
++
++/*
++ * Load a segment. Fall back on loading the zero
++ * segment if something goes wrong..
++ */
++#define loadsegment(seg,value) \
++ asm volatile("\n" \
++ "1:\t" \
++ "mov %0,%%" #seg "\n" \
++ "2:\n" \
++ ".section .fixup,\"ax\"\n" \
++ "3:\t" \
++ "pushl $0\n\t" \
++ "popl %%" #seg "\n\t" \
++ "jmp 2b\n" \
++ ".previous\n" \
++ ".section __ex_table,\"a\"\n\t" \
++ ".align 4\n\t" \
++ ".long 1b,3b\n" \
++ ".previous" \
++ : :"rm" (value))
++
++/*
++ * Save a segment register away
++ */
++#define savesegment(seg, value) \
++ asm volatile("mov %%" #seg ",%0":"=rm" (value))
++
++/*
++ * Clear and set 'TS' bit respectively
++ */
++#define clts() (HYPERVISOR_fpu_taskswitch(0))
++#define read_cr0() ({ \
++ unsigned int __dummy; \
++ __asm__ __volatile__( \
++ "movl %%cr0,%0\n\t" \
++ :"=r" (__dummy)); \
++ __dummy; \
++})
++#define write_cr0(x) \
++ __asm__ __volatile__("movl %0,%%cr0": :"r" (x));
++
++#define read_cr2() \
++ (HYPERVISOR_shared_info->vcpu_info[smp_processor_id()].arch.cr2)
++#define write_cr2(x) \
++ __asm__ __volatile__("movl %0,%%cr2": :"r" (x));
++
++#define read_cr3() ({ \
++ unsigned int __dummy; \
++ __asm__ ( \
++ "movl %%cr3,%0\n\t" \
++ :"=r" (__dummy)); \
++ __dummy = xen_cr3_to_pfn(__dummy); \
++ mfn_to_pfn(__dummy) << PAGE_SHIFT; \
++})
++#define write_cr3(x) ({ \
++ unsigned int __dummy = pfn_to_mfn((x) >> PAGE_SHIFT); \
++ __dummy = xen_pfn_to_cr3(__dummy); \
++ __asm__ __volatile__("movl %0,%%cr3": :"r" (__dummy)); \
++})
++
++#define read_cr4() ({ \
++ unsigned int __dummy; \
++ __asm__( \
++ "movl %%cr4,%0\n\t" \
++ :"=r" (__dummy)); \
++ __dummy; \
++})
++
++#define read_cr4_safe() ({ \
++ unsigned int __dummy; \
++ /* This could fault if %cr4 does not exist */ \
++ __asm__("1: movl %%cr4, %0 \n" \
++ "2: \n" \
++ ".section __ex_table,\"a\" \n" \
++ ".long 1b,2b \n" \
++ ".previous \n" \
++ : "=r" (__dummy): "0" (0)); \
++ __dummy; \
++})
++
++#define write_cr4(x) \
++ __asm__ __volatile__("movl %0,%%cr4": :"r" (x));
++#define stts() (HYPERVISOR_fpu_taskswitch(1))
++
++#endif /* __KERNEL__ */
++
++#define wbinvd() \
++ __asm__ __volatile__ ("wbinvd": : :"memory");
++
++static inline unsigned long get_limit(unsigned long segment)
++{
++ unsigned long __limit;
++ __asm__("lsll %1,%0"
++ :"=r" (__limit):"r" (segment));
++ return __limit+1;
++}
++
++#define nop() __asm__ __volatile__ ("nop")
++
++#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
++
++#define tas(ptr) (xchg((ptr),1))
++
++struct __xchg_dummy { unsigned long a[100]; };
++#define __xg(x) ((struct __xchg_dummy *)(x))
++
++
++#ifdef CONFIG_X86_CMPXCHG64
++
++/*
++ * The semantics of XCHGCMP8B are a bit strange, this is why
++ * there is a loop and the loading of %%eax and %%edx has to
++ * be inside. This inlines well in most cases, the cached
++ * cost is around ~38 cycles. (in the future we might want
++ * to do an SIMD/3DNOW!/MMX/FPU 64-bit store here, but that
++ * might have an implicit FPU-save as a cost, so it's not
++ * clear which path to go.)
++ *
++ * cmpxchg8b must be used with the lock prefix here to allow
++ * the instruction to be executed atomically, see page 3-102
++ * of the instruction set reference 24319102.pdf. We need
++ * the reader side to see the coherent 64bit value.
++ */
++static inline void __set_64bit (unsigned long long * ptr,
++ unsigned int low, unsigned int high)
++{
++ __asm__ __volatile__ (
++ "\n1:\t"
++ "movl (%0), %%eax\n\t"
++ "movl 4(%0), %%edx\n\t"
++ "lock cmpxchg8b (%0)\n\t"
++ "jnz 1b"
++ : /* no outputs */
++ : "D"(ptr),
++ "b"(low),
++ "c"(high)
++ : "ax","dx","memory");
++}
++
++static inline void __set_64bit_constant (unsigned long long *ptr,
++ unsigned long long value)
++{
++ __set_64bit(ptr,(unsigned int)(value), (unsigned int)((value)>>32ULL));
++}
++#define ll_low(x) *(((unsigned int*)&(x))+0)
++#define ll_high(x) *(((unsigned int*)&(x))+1)
++
++static inline void __set_64bit_var (unsigned long long *ptr,
++ unsigned long long value)
++{
++ __set_64bit(ptr,ll_low(value), ll_high(value));
++}
++
++#define set_64bit(ptr,value) \
++(__builtin_constant_p(value) ? \
++ __set_64bit_constant(ptr, value) : \
++ __set_64bit_var(ptr, value) )
++
++#define _set_64bit(ptr,value) \
++(__builtin_constant_p(value) ? \
++ __set_64bit(ptr, (unsigned int)(value), (unsigned int)((value)>>32ULL) ) : \
++ __set_64bit(ptr, ll_low(value), ll_high(value)) )
++
++#endif
++
++/*
++ * Note: no "lock" prefix even on SMP: xchg always implies lock anyway
++ * Note 2: xchg has side effect, so that attribute volatile is necessary,
++ * but generally the primitive is invalid, *ptr is output argument. --ANK
++ */
++static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
++{
++ switch (size) {
++ case 1:
++ __asm__ __volatile__("xchgb %b0,%1"
++ :"=q" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ case 2:
++ __asm__ __volatile__("xchgw %w0,%1"
++ :"=r" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ case 4:
++ __asm__ __volatile__("xchgl %0,%1"
++ :"=r" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ }
++ return x;
++}
++
++/*
++ * Atomic compare and exchange. Compare OLD with MEM, if identical,
++ * store NEW in MEM. Return the initial value in MEM. Success is
++ * indicated by comparing RETURN with OLD.
++ */
++
++#ifdef CONFIG_X86_CMPXCHG
++#define __HAVE_ARCH_CMPXCHG 1
++#define cmpxchg(ptr,o,n)\
++ ((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
++ (unsigned long)(n),sizeof(*(ptr))))
++#endif
++
++static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
++ unsigned long new, int size)
++{
++ unsigned long prev;
++ switch (size) {
++ case 1:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
++ : "=a"(prev)
++ : "q"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ case 2:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ case 4:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgl %1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ }
++ return old;
++}
++
++#ifndef CONFIG_X86_CMPXCHG
++/*
++ * Building a kernel capable running on 80386. It may be necessary to
++ * simulate the cmpxchg on the 80386 CPU. For that purpose we define
++ * a function for each of the sizes we support.
++ */
++
++extern unsigned long cmpxchg_386_u8(volatile void *, u8, u8);
++extern unsigned long cmpxchg_386_u16(volatile void *, u16, u16);
++extern unsigned long cmpxchg_386_u32(volatile void *, u32, u32);
++
++static inline unsigned long cmpxchg_386(volatile void *ptr, unsigned long old,
++ unsigned long new, int size)
++{
++ switch (size) {
++ case 1:
++ return cmpxchg_386_u8(ptr, old, new);
++ case 2:
++ return cmpxchg_386_u16(ptr, old, new);
++ case 4:
++ return cmpxchg_386_u32(ptr, old, new);
++ }
++ return old;
++}
++
++#define cmpxchg(ptr,o,n) \
++({ \
++ __typeof__(*(ptr)) __ret; \
++ if (likely(boot_cpu_data.x86 > 3)) \
++ __ret = __cmpxchg((ptr), (unsigned long)(o), \
++ (unsigned long)(n), sizeof(*(ptr))); \
++ else \
++ __ret = cmpxchg_386((ptr), (unsigned long)(o), \
++ (unsigned long)(n), sizeof(*(ptr))); \
++ __ret; \
++})
++#endif
++
++#ifdef CONFIG_X86_CMPXCHG64
++
++static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long long old,
++ unsigned long long new)
++{
++ unsigned long long prev;
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchg8b %3"
++ : "=A"(prev)
++ : "b"((unsigned long)new),
++ "c"((unsigned long)(new >> 32)),
++ "m"(*__xg(ptr)),
++ "0"(old)
++ : "memory");
++ return prev;
++}
++
++#define cmpxchg64(ptr,o,n)\
++ ((__typeof__(*(ptr)))__cmpxchg64((ptr),(unsigned long long)(o),\
++ (unsigned long long)(n)))
++
++#endif
++
++/*
++ * Force strict CPU ordering.
++ * And yes, this is required on UP too when we're talking
++ * to devices.
++ *
++ * For now, "wmb()" doesn't actually do anything, as all
++ * Intel CPU's follow what Intel calls a *Processor Order*,
++ * in which all writes are seen in the program order even
++ * outside the CPU.
++ *
++ * I expect future Intel CPU's to have a weaker ordering,
++ * but I'd also expect them to finally get their act together
++ * and add some real memory barriers if so.
++ *
++ * Some non intel clones support out of order store. wmb() ceases to be a
++ * nop for these.
++ */
++
++
++/*
++ * Actually only lfence would be needed for mb() because all stores done
++ * by the kernel should be already ordered. But keep a full barrier for now.
++ */
++
++#define mb() alternative("lock; addl $0,0(%%esp)", "mfence", X86_FEATURE_XMM2)
++#define rmb() alternative("lock; addl $0,0(%%esp)", "lfence", X86_FEATURE_XMM2)
++
++/**
++ * read_barrier_depends - Flush all pending reads that subsequents reads
++ * depend on.
++ *
++ * No data-dependent reads from memory-like regions are ever reordered
++ * over this barrier. All reads preceding this primitive are guaranteed
++ * to access memory (but not necessarily other CPUs' caches) before any
++ * reads following this primitive that depend on the data return by
++ * any of the preceding reads. This primitive is much lighter weight than
++ * rmb() on most CPUs, and is never heavier weight than is
++ * rmb().
++ *
++ * These ordering constraints are respected by both the local CPU
++ * and the compiler.
++ *
++ * Ordering is not guaranteed by anything other than these primitives,
++ * not even by data dependencies. See the documentation for
++ * memory_barrier() for examples and URLs to more information.
++ *
++ * For example, the following code would force ordering (the initial
++ * value of "a" is zero, "b" is one, and "p" is "&a"):
++ *
++ * <programlisting>
++ * CPU 0 CPU 1
++ *
++ * b = 2;
++ * memory_barrier();
++ * p = &b; q = p;
++ * read_barrier_depends();
++ * d = *q;
++ * </programlisting>
++ *
++ * because the read of "*q" depends on the read of "p" and these
++ * two reads are separated by a read_barrier_depends(). However,
++ * the following code, with the same initial values for "a" and "b":
++ *
++ * <programlisting>
++ * CPU 0 CPU 1
++ *
++ * a = 2;
++ * memory_barrier();
++ * b = 3; y = b;
++ * read_barrier_depends();
++ * x = a;
++ * </programlisting>
++ *
++ * does not enforce ordering, since there is no data dependency between
++ * the read of "a" and the read of "b". Therefore, on some CPUs, such
++ * as Alpha, "y" could be set to 3 and "x" to 0. Use rmb()
++ * in cases like thiswhere there are no data dependencies.
++ **/
++
++#define read_barrier_depends() do { } while(0)
++
++#ifdef CONFIG_X86_OOSTORE
++/* Actually there are no OOO store capable CPUs for now that do SSE,
++ but make it already an possibility. */
++#define wmb() alternative("lock; addl $0,0(%%esp)", "sfence", X86_FEATURE_XMM)
++#else
++#define wmb() __asm__ __volatile__ ("": : :"memory")
++#endif
++
++#ifdef CONFIG_SMP
++#define smp_mb() mb()
++#define smp_rmb() rmb()
++#define smp_wmb() wmb()
++#define smp_read_barrier_depends() read_barrier_depends()
++#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
++#else
++#define smp_mb() barrier()
++#define smp_rmb() barrier()
++#define smp_wmb() barrier()
++#define smp_read_barrier_depends() do { } while(0)
++#define set_mb(var, value) do { var = value; barrier(); } while (0)
++#endif
++
++#define set_wmb(var, value) do { var = value; wmb(); } while (0)
++
++/* interrupt control.. */
++
++/*
++ * The use of 'barrier' in the following reflects their use as local-lock
++ * operations. Reentrancy must be prevented (e.g., __cli()) /before/ following
++ * critical operations are executed. All critical operations must complete
++ * /before/ reentrancy is permitted (e.g., __sti()). Alpha architecture also
++ * includes these barriers, for example.
++ */
++
++#define __cli() \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ _vcpu->evtchn_upcall_mask = 1; \
++ preempt_enable_no_resched(); \
++ barrier(); \
++} while (0)
++
++#define __sti() \
++do { \
++ vcpu_info_t *_vcpu; \
++ barrier(); \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ _vcpu->evtchn_upcall_mask = 0; \
++ barrier(); /* unmask then check (avoid races) */ \
++ if (unlikely(_vcpu->evtchn_upcall_pending)) \
++ force_evtchn_callback(); \
++ preempt_enable(); \
++} while (0)
++
++#define __save_flags(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ (x) = _vcpu->evtchn_upcall_mask; \
++ preempt_enable(); \
++} while (0)
++
++#define __restore_flags(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ barrier(); \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ if ((_vcpu->evtchn_upcall_mask = (x)) == 0) { \
++ barrier(); /* unmask then check (avoid races) */ \
++ if (unlikely(_vcpu->evtchn_upcall_pending)) \
++ force_evtchn_callback(); \
++ preempt_enable(); \
++ } else \
++ preempt_enable_no_resched(); \
++} while (0)
++
++void safe_halt(void);
++void halt(void);
++
++#define __save_and_cli(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ (x) = _vcpu->evtchn_upcall_mask; \
++ _vcpu->evtchn_upcall_mask = 1; \
++ preempt_enable_no_resched(); \
++ barrier(); \
++} while (0)
++
++#define local_irq_save(x) __save_and_cli(x)
++#define local_irq_restore(x) __restore_flags(x)
++#define local_save_flags(x) __save_flags(x)
++#define local_irq_disable() __cli()
++#define local_irq_enable() __sti()
++
++/* Cannot use preempt_enable() here as we would recurse in preempt_sched(). */
++#define irqs_disabled() \
++({ int ___x; \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ ___x = (_vcpu->evtchn_upcall_mask != 0); \
++ preempt_enable_no_resched(); \
++ ___x; })
++
++/*
++ * disable hlt during certain critical i/o operations
++ */
++#define HAVE_DISABLE_HLT
++void disable_hlt(void);
++void enable_hlt(void);
++
++extern int es7000_plat;
++void cpu_idle_wait(void);
++
++/*
++ * On SMP systems, when the scheduler does migration-cost autodetection,
++ * it needs a way to flush as much of the CPU's caches as possible:
++ */
++static inline void sched_cacheflush(void)
++{
++ wbinvd();
++}
++
++extern unsigned long arch_align_stack(unsigned long sp);
++extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
++
++void default_idle(void);
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/tlbflush.h new/include/asm-i386/mach-xen/asm/tlbflush.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/tlbflush.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/tlbflush.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,102 @@
++#ifndef _I386_TLBFLUSH_H
++#define _I386_TLBFLUSH_H
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <asm/processor.h>
++
++#define __flush_tlb() xen_tlb_flush()
++#define __flush_tlb_global() xen_tlb_flush()
++#define __flush_tlb_all() xen_tlb_flush()
++
++extern unsigned long pgkern_mask;
++
++#define cpu_has_invlpg (boot_cpu_data.x86 > 3)
++
++#define __flush_tlb_single(addr) xen_invlpg(addr)
++
++#define __flush_tlb_one(addr) __flush_tlb_single(addr)
++
++/*
++ * TLB flushing:
++ *
++ * - flush_tlb() flushes the current mm struct TLBs
++ * - flush_tlb_all() flushes all processes TLBs
++ * - flush_tlb_mm(mm) flushes the specified mm context TLB's
++ * - flush_tlb_page(vma, vmaddr) flushes one page
++ * - flush_tlb_range(vma, start, end) flushes a range of pages
++ * - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
++ * - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
++ *
++ * ..but the i386 has somewhat limited tlb flushing capabilities,
++ * and page-granular flushes are available only on i486 and up.
++ */
++
++#ifndef CONFIG_SMP
++
++#define flush_tlb() __flush_tlb()
++#define flush_tlb_all() __flush_tlb_all()
++#define local_flush_tlb() __flush_tlb()
++
++static inline void flush_tlb_mm(struct mm_struct *mm)
++{
++ if (mm == current->active_mm)
++ __flush_tlb();
++}
++
++static inline void flush_tlb_page(struct vm_area_struct *vma,
++ unsigned long addr)
++{
++ if (vma->vm_mm == current->active_mm)
++ __flush_tlb_one(addr);
++}
++
++static inline void flush_tlb_range(struct vm_area_struct *vma,
++ unsigned long start, unsigned long end)
++{
++ if (vma->vm_mm == current->active_mm)
++ __flush_tlb();
++}
++
++#else
++
++#include <asm/smp.h>
++
++#define local_flush_tlb() \
++ __flush_tlb()
++
++extern void flush_tlb_all(void);
++extern void flush_tlb_current_task(void);
++extern void flush_tlb_mm(struct mm_struct *);
++extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
++
++#define flush_tlb() flush_tlb_current_task()
++
++static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
++{
++ flush_tlb_mm(vma->vm_mm);
++}
++
++#define TLBSTATE_OK 1
++#define TLBSTATE_LAZY 2
++
++struct tlb_state
++{
++ struct mm_struct *active_mm;
++ int state;
++ char __cacheline_padding[L1_CACHE_BYTES-8];
++};
++DECLARE_PER_CPU(struct tlb_state, cpu_tlbstate);
++
++
++#endif
++
++#define flush_tlb_kernel_range(start, end) flush_tlb_all()
++
++static inline void flush_tlb_pgtables(struct mm_struct *mm,
++ unsigned long start, unsigned long end)
++{
++ /* i386 does not keep any page table caches in TLB */
++}
++
++#endif /* _I386_TLBFLUSH_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/asm/vga.h new/include/asm-i386/mach-xen/asm/vga.h
+--- linux-2.6/include/asm-i386/mach-xen/asm/vga.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/asm/vga.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,20 @@
++/*
++ * Access to VGA videoram
++ *
++ * (c) 1998 Martin Mares <mj@ucw.cz>
++ */
++
++#ifndef _LINUX_ASM_VGA_H_
++#define _LINUX_ASM_VGA_H_
++
++/*
++ * On the PC, we can just recalculate addresses and then
++ * access the videoram directly without any black magic.
++ */
++
++#define VGA_MAP_MEM(x) (unsigned long)isa_bus_to_virt(x)
++
++#define vga_readb(x) (*(x))
++#define vga_writeb(x,y) (*(y) = (x))
++
++#endif
+diff -urNp linux-2.6/include/asm-i386/mach-xen/irq_vectors.h new/include/asm-i386/mach-xen/irq_vectors.h
+--- linux-2.6/include/asm-i386/mach-xen/irq_vectors.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/irq_vectors.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,125 @@
++/*
++ * This file should contain #defines for all of the interrupt vector
++ * numbers used by this architecture.
++ *
++ * In addition, there are some standard defines:
++ *
++ * FIRST_EXTERNAL_VECTOR:
++ * The first free place for external interrupts
++ *
++ * SYSCALL_VECTOR:
++ * The IRQ vector a syscall makes the user to kernel transition
++ * under.
++ *
++ * TIMER_IRQ:
++ * The IRQ number the timer interrupt comes in at.
++ *
++ * NR_IRQS:
++ * The total number of interrupt vectors (including all the
++ * architecture specific interrupts) needed.
++ *
++ */
++#ifndef _ASM_IRQ_VECTORS_H
++#define _ASM_IRQ_VECTORS_H
++
++/*
++ * IDT vectors usable for external interrupt sources start
++ * at 0x20:
++ */
++#define FIRST_EXTERNAL_VECTOR 0x20
++
++#define SYSCALL_VECTOR 0x80
++
++/*
++ * Vectors 0x20-0x2f are used for ISA interrupts.
++ */
++
++#if 0
++/*
++ * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
++ *
++ * some of the following vectors are 'rare', they are merged
++ * into a single vector (CALL_FUNCTION_VECTOR) to save vector space.
++ * TLB, reschedule and local APIC vectors are performance-critical.
++ *
++ * Vectors 0xf0-0xfa are free (reserved for future Linux use).
++ */
++#define SPURIOUS_APIC_VECTOR 0xff
++#define ERROR_APIC_VECTOR 0xfe
++#define INVALIDATE_TLB_VECTOR 0xfd
++#define RESCHEDULE_VECTOR 0xfc
++#define CALL_FUNCTION_VECTOR 0xfb
++
++#define THERMAL_APIC_VECTOR 0xf0
++/*
++ * Local APIC timer IRQ vector is on a different priority level,
++ * to work around the 'lost local interrupt if more than 2 IRQ
++ * sources per level' errata.
++ */
++#define LOCAL_TIMER_VECTOR 0xef
++#endif
++
++#define SPURIOUS_APIC_VECTOR 0xff
++#define ERROR_APIC_VECTOR 0xfe
++
++/*
++ * First APIC vector available to drivers: (vectors 0x30-0xee)
++ * we start at 0x31 to spread out vectors evenly between priority
++ * levels. (0x80 is the syscall vector)
++ */
++#define FIRST_DEVICE_VECTOR 0x31
++#define FIRST_SYSTEM_VECTOR 0xef
++
++/*
++ * 16 8259A IRQ's, 208 potential APIC interrupt sources.
++ * Right now the APIC is mostly only used for SMP.
++ * 256 vectors is an architectural limit. (we can have
++ * more than 256 devices theoretically, but they will
++ * have to use shared interrupts)
++ * Since vectors 0x00-0x1f are used/reserved for the CPU,
++ * the usable vector space is 0x20-0xff (224 vectors)
++ */
++
++#define RESCHEDULE_VECTOR 0
++#define CALL_FUNCTION_VECTOR 1
++#define NR_IPIS 2
++
++/*
++ * The maximum number of vectors supported by i386 processors
++ * is limited to 256. For processors other than i386, NR_VECTORS
++ * should be changed accordingly.
++ */
++#define NR_VECTORS 256
++
++#define FPU_IRQ 13
++
++#define FIRST_VM86_IRQ 3
++#define LAST_VM86_IRQ 15
++#define invalid_vm86_irq(irq) ((irq) < 3 || (irq) > 15)
++
++/*
++ * The flat IRQ space is divided into two regions:
++ * 1. A one-to-one mapping of real physical IRQs. This space is only used
++ * if we have physical device-access privilege. This region is at the
++ * start of the IRQ space so that existing device drivers do not need
++ * to be modified to translate physical IRQ numbers into our IRQ space.
++ * 3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
++ * are bound using the provided bind/unbind functions.
++ */
++
++#define PIRQ_BASE 0
++#define NR_PIRQS 256
++
++#define DYNIRQ_BASE (PIRQ_BASE + NR_PIRQS)
++#define NR_DYNIRQS 256
++
++#define NR_IRQS (NR_PIRQS + NR_DYNIRQS)
++#define NR_IRQ_VECTORS NR_IRQS
++
++#define pirq_to_irq(_x) ((_x) + PIRQ_BASE)
++#define irq_to_pirq(_x) ((_x) - PIRQ_BASE)
++
++#define dynirq_to_irq(_x) ((_x) + DYNIRQ_BASE)
++#define irq_to_dynirq(_x) ((_x) - DYNIRQ_BASE)
++
++#endif /* _ASM_IRQ_VECTORS_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/mach_traps.h new/include/asm-i386/mach-xen/mach_traps.h
+--- linux-2.6/include/asm-i386/mach-xen/mach_traps.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/mach_traps.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,33 @@
++/*
++ * include/asm-xen/asm-i386/mach-xen/mach_traps.h
++ *
++ * Machine specific NMI handling for Xen
++ */
++#ifndef _MACH_TRAPS_H
++#define _MACH_TRAPS_H
++
++#include <linux/bitops.h>
++#include <xen/interface/nmi.h>
++
++static inline void clear_mem_error(unsigned char reason) {}
++static inline void clear_io_check_error(unsigned char reason) {}
++
++static inline unsigned char get_nmi_reason(void)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ unsigned char reason = 0;
++
++ /* construct a value which looks like it came from
++ * port 0x61.
++ */
++ if (test_bit(_XEN_NMIREASON_io_error, &s->arch.nmi_reason))
++ reason |= 0x40;
++ if (test_bit(_XEN_NMIREASON_parity_error, &s->arch.nmi_reason))
++ reason |= 0x80;
++
++ return reason;
++}
++
++static inline void reassert_nmi(void) {}
++
++#endif /* !_MACH_TRAPS_H */
+diff -urNp linux-2.6/include/asm-i386/mach-xen/setup_arch_post.h new/include/asm-i386/mach-xen/setup_arch_post.h
+--- linux-2.6/include/asm-i386/mach-xen/setup_arch_post.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/setup_arch_post.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,101 @@
++/**
++ * machine_specific_memory_setup - Hook for machine specific memory setup.
++ *
++ * Description:
++ * This is included late in kernel/setup.c so that it can make
++ * use of all of the static functions.
++ **/
++
++#include <xen/interface/callback.h>
++#include <xen/interface/memory.h>
++
++static char * __init machine_specific_memory_setup(void)
++{
++ int rc;
++ struct xen_memory_map memmap;
++ /*
++ * This is rather large for a stack variable but this early in
++ * the boot process we know we have plenty slack space.
++ */
++ struct e820entry map[E820MAX];
++
++ memmap.nr_entries = E820MAX;
++ set_xen_guest_handle(memmap.buffer, map);
++
++ rc = HYPERVISOR_memory_op(XENMEM_memory_map, &memmap);
++ if ( rc == -ENOSYS ) {
++ memmap.nr_entries = 1;
++ map[0].addr = 0ULL;
++ map[0].size = PFN_PHYS(xen_start_info->nr_pages);
++ /* 8MB slack (to balance backend allocations). */
++ map[0].size += 8ULL << 20;
++ map[0].type = E820_RAM;
++ rc = 0;
++ }
++ BUG_ON(rc);
++
++ sanitize_e820_map(map, (char *)&memmap.nr_entries);
++
++ BUG_ON(copy_e820_map(map, (char)memmap.nr_entries) < 0);
++
++ return "Xen";
++}
++
++extern void hypervisor_callback(void);
++extern void failsafe_callback(void);
++extern void nmi(void);
++
++unsigned long *machine_to_phys_mapping;
++EXPORT_SYMBOL(machine_to_phys_mapping);
++unsigned int machine_to_phys_order;
++EXPORT_SYMBOL(machine_to_phys_order);
++
++static void __init machine_specific_arch_setup(void)
++{
++ int ret;
++ struct xen_machphys_mapping mapping;
++ unsigned long machine_to_phys_nr_ents;
++ struct xen_platform_parameters pp;
++ struct callback_register event = {
++ .type = CALLBACKTYPE_event,
++ .address = { __KERNEL_CS, (unsigned long)hypervisor_callback },
++ };
++ struct callback_register failsafe = {
++ .type = CALLBACKTYPE_failsafe,
++ .address = { __KERNEL_CS, (unsigned long)failsafe_callback },
++ };
++ struct callback_register nmi_cb = {
++ .type = CALLBACKTYPE_nmi,
++ .address = { __KERNEL_CS, (unsigned long)nmi },
++ };
++
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
++ if (ret == 0)
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
++ if (ret == -ENOSYS)
++ ret = HYPERVISOR_set_callbacks(
++ event.address.cs, event.address.eip,
++ failsafe.address.cs, failsafe.address.eip);
++ BUG_ON(ret);
++
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
++ if (ret == -ENOSYS) {
++ struct xennmi_callback cb;
++
++ cb.handler_address = nmi_cb.address.eip;
++ HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
++ }
++
++ if (HYPERVISOR_xen_version(XENVER_platform_parameters,
++ &pp) == 0)
++ set_fixaddr_top(pp.virt_start - PAGE_SIZE);
++
++ machine_to_phys_mapping = (unsigned long *)MACH2PHYS_VIRT_START;
++ machine_to_phys_nr_ents = MACH2PHYS_NR_ENTRIES;
++ if (HYPERVISOR_memory_op(XENMEM_machphys_mapping, &mapping) == 0) {
++ machine_to_phys_mapping = (unsigned long *)mapping.v_start;
++ machine_to_phys_nr_ents = mapping.max_mfn + 1;
++ }
++ while ((1UL << machine_to_phys_order) < machine_to_phys_nr_ents )
++ machine_to_phys_order++;
++}
+diff -urNp linux-2.6/include/asm-i386/mach-xen/setup_arch_pre.h new/include/asm-i386/mach-xen/setup_arch_pre.h
+--- linux-2.6/include/asm-i386/mach-xen/setup_arch_pre.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-i386/mach-xen/setup_arch_pre.h 2006-05-09 12:35:17.000000000 +0200
+@@ -0,0 +1,5 @@
++/* Hook to call BIOS initialisation function */
++
++#define ARCH_SETUP machine_specific_arch_setup();
++
++static void __init machine_specific_arch_setup(void);
+diff -urNp linux-2.6/include/asm-i386/page.h new/include/asm-i386/page.h
+--- linux-2.6/include/asm-i386/page.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/page.h 2006-05-09 12:35:17.000000000 +0200
+@@ -121,7 +121,7 @@ extern int page_is_ram(unsigned long pag
+
+ #define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET)
+ #define VMALLOC_RESERVE ((unsigned long)__VMALLOC_RESERVE)
+-#define MAXMEM (-__PAGE_OFFSET-__VMALLOC_RESERVE)
++#define MAXMEM (__FIXADDR_TOP-__PAGE_OFFSET-__VMALLOC_RESERVE)
+ #define __pa(x) ((unsigned long)(x)-PAGE_OFFSET)
+ #define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET))
+ #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT)
+@@ -137,6 +137,8 @@ extern int page_is_ram(unsigned long pag
+ ((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
+ VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
+
++#define __HAVE_ARCH_GATE_AREA 1
++
+ #endif /* __KERNEL__ */
+
+ #include <asm-generic/memory_model.h>
+diff -urNp linux-2.6/include/asm-i386/pgtable-2level-defs.h new/include/asm-i386/pgtable-2level-defs.h
+--- linux-2.6/include/asm-i386/pgtable-2level-defs.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/pgtable-2level-defs.h 2006-05-09 12:35:17.000000000 +0200
+@@ -1,6 +1,8 @@
+ #ifndef _I386_PGTABLE_2LEVEL_DEFS_H
+ #define _I386_PGTABLE_2LEVEL_DEFS_H
+
++#define HAVE_SHARED_KERNEL_PMD 0
++
+ /*
+ * traditional i386 two-level paging structure:
+ */
+diff -urNp linux-2.6/include/asm-i386/pgtable-3level-defs.h new/include/asm-i386/pgtable-3level-defs.h
+--- linux-2.6/include/asm-i386/pgtable-3level-defs.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-i386/pgtable-3level-defs.h 2006-05-09 12:35:18.000000000 +0200
+@@ -1,6 +1,8 @@
+ #ifndef _I386_PGTABLE_3LEVEL_DEFS_H
+ #define _I386_PGTABLE_3LEVEL_DEFS_H
+
++#define HAVE_SHARED_KERNEL_PMD 1
++
+ /*
+ * PGDIR_SHIFT determines what a top-level page table entry can map
+ */
+diff -urNp linux-2.6/include/asm-ia64/agp.h new/include/asm-ia64/agp.h
+--- linux-2.6/include/asm-ia64/agp.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/agp.h 2006-05-09 12:35:18.000000000 +0200
+@@ -19,13 +19,44 @@
+ #define flush_agp_cache() mb()
+
+ /* Convert a physical address to an address suitable for the GART. */
++#ifndef CONFIG_XEN_IA64_DOM0_VP
+ #define phys_to_gart(x) (x)
+ #define gart_to_phys(x) (x)
++#else
++#define phys_to_gart(x) phys_to_machine_for_dma(x)
++#define gart_to_phys(x) machine_to_phys_for_dma(x)
++#endif
+
+ /* GATT allocation. Returns/accepts GATT kernel virtual address. */
++#ifndef CONFIG_XEN_IA64_DOM0_VP
+ #define alloc_gatt_pages(order) \
+ ((char *)__get_free_pages(GFP_KERNEL, (order)))
+ #define free_gatt_pages(table, order) \
+ free_pages((unsigned long)(table), (order))
++#else
++#include <asm/hypervisor.h>
++static inline char*
++alloc_gatt_pages(unsigned int order)
++{
++ unsigned long error;
++ unsigned long ret = __get_free_pages(GFP_KERNEL, (order));
++ if (ret == 0) {
++ goto out;
++ }
++ error = xen_create_contiguous_region(ret, order, 0);
++ if (error) {
++ free_pages(ret, order);
++ ret = 0;
++ }
++out:
++ return (char*)ret;
++}
++static inline void
++free_gatt_pages(void* table, unsigned int order)
++{
++ xen_destroy_contiguous_region((unsigned long)table, order);
++ free_pages((unsigned long)table, order);
++}
++#endif /* CONFIG_XEN_IA64_DOM0_VP */
+
+ #endif /* _ASM_IA64_AGP_H */
+diff -urNp linux-2.6/include/asm-ia64/dma-mapping.h new/include/asm-ia64/dma-mapping.h
+--- linux-2.6/include/asm-ia64/dma-mapping.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/dma-mapping.h 2006-06-28 14:32:14.000000000 +0200
+@@ -7,7 +7,14 @@
+ */
+ #include <linux/config.h>
+ #include <asm/machvec.h>
++#ifdef CONFIG_XEN_IA64_DOM0_VP
++/* Needed for arch/i386/kernel/swiotlb.c and arch/i386/kernel/pci-dma-xen.c */
++#include <asm/hypervisor.h>
++/* Needed for arch/i386/kernel/swiotlb.c */
++#include <asm-i386/mach-xen/asm/swiotlb.h>
++#endif
+
++#ifndef CONFIG_XEN_IA64_DOM0_VP
+ #define dma_alloc_coherent platform_dma_alloc_coherent
+ #define dma_alloc_noncoherent platform_dma_alloc_coherent /* coherent mem. is cheap */
+ #define dma_free_coherent platform_dma_free_coherent
+@@ -21,6 +28,46 @@
+ #define dma_sync_single_for_device platform_dma_sync_single_for_device
+ #define dma_sync_sg_for_device platform_dma_sync_sg_for_device
+ #define dma_mapping_error platform_dma_mapping_error
++#else
++int dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents,
++ enum dma_data_direction direction);
++void dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
++ enum dma_data_direction direction);
++int dma_supported(struct device *dev, u64 mask);
++void *dma_alloc_coherent(struct device *dev, size_t size,
++ dma_addr_t *dma_handle, gfp_t gfp);
++void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
++ dma_addr_t dma_handle);
++dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
++ enum dma_data_direction direction);
++void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
++ enum dma_data_direction direction);
++void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
++ size_t size, enum dma_data_direction direction);
++void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
++ size_t size,
++ enum dma_data_direction direction);
++int dma_mapping_error(dma_addr_t dma_addr);
++
++#define flush_write_buffers() do { } while (0)
++static inline void
++dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_sg_for_cpu(dev,sg,nelems,direction);
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
++ enum dma_data_direction direction)
++{
++ if (swiotlb)
++ swiotlb_sync_sg_for_device(dev,sg,nelems,direction);
++ flush_write_buffers();
++}
++#endif
+
+ #define dma_map_page(dev, pg, off, size, dir) \
+ dma_map_single(dev, page_address(pg) + (off), (size), (dir))
+@@ -62,4 +109,29 @@ dma_cache_sync (void *vaddr, size_t size
+
+ #define dma_is_consistent(dma_handle) (1) /* all we do is coherent memory... */
+
++#ifdef CONFIG_XEN_IA64_DOM0_VP
++/* arch/i386/kernel/swiotlb.o requires */
++void contiguous_bitmap_init(unsigned long end_pfn);
++
++static inline int
++address_needs_mapping(struct device *hwdev, dma_addr_t addr)
++{
++ dma_addr_t mask = DMA_64BIT_MASK;
++ /* If the device has a mask, use it, otherwise default to 64 bits */
++ if (hwdev && hwdev->dma_mask)
++ mask = *hwdev->dma_mask;
++ return (addr & ~mask) != 0;
++}
++
++static inline int
++range_straddles_page_boundary(void *p, size_t size)
++{
++ extern unsigned long *contiguous_bitmap;
++ return (((((unsigned long)p & ~PAGE_MASK) + size) > PAGE_SIZE) &&
++ !test_bit(__pa(p) >> PAGE_SHIFT, contiguous_bitmap));
++}
++#else
++#define contiguous_bitmap_init(end_pfn) ((void)end_pfn)
++#endif
++
+ #endif /* _ASM_IA64_DMA_MAPPING_H */
+diff -urNp linux-2.6/include/asm-ia64/fixmap.h new/include/asm-ia64/fixmap.h
+--- linux-2.6/include/asm-ia64/fixmap.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/fixmap.h 2006-05-09 12:35:18.000000000 +0200
+@@ -0,0 +1,2 @@
++#define clear_fixmap(x) do {} while (0)
++#define set_fixmap(x,y) do {} while (0)
+diff -urNp linux-2.6/include/asm-ia64/gcc_intrin.h new/include/asm-ia64/gcc_intrin.h
+--- linux-2.6/include/asm-ia64/gcc_intrin.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/gcc_intrin.h 2006-05-09 12:35:18.000000000 +0200
+@@ -26,7 +26,7 @@ extern void ia64_bad_param_for_getreg (v
+
+ register unsigned long ia64_r13 asm ("r13") __attribute_used__;
+
+-#define ia64_setreg(regnum, val) \
++#define __ia64_setreg(regnum, val) \
+ ({ \
+ switch (regnum) { \
+ case _IA64_REG_PSR_L: \
+@@ -55,7 +55,7 @@ register unsigned long ia64_r13 asm ("r1
+ } \
+ })
+
+-#define ia64_getreg(regnum) \
++#define __ia64_getreg(regnum) \
+ ({ \
+ __u64 ia64_intri_res; \
+ \
+@@ -92,7 +92,7 @@ register unsigned long ia64_r13 asm ("r1
+
+ #define ia64_hint_pause 0
+
+-#define ia64_hint(mode) \
++#define __ia64_hint(mode) \
+ ({ \
+ switch (mode) { \
+ case ia64_hint_pause: \
+@@ -374,7 +374,7 @@ register unsigned long ia64_r13 asm ("r1
+
+ #define ia64_invala() asm volatile ("invala" ::: "memory")
+
+-#define ia64_thash(addr) \
++#define __ia64_thash(addr) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("thash %0=%1" : "=r"(ia64_intri_res) : "r" (addr)); \
+@@ -394,18 +394,18 @@ register unsigned long ia64_r13 asm ("r1
+
+ #define ia64_nop(x) asm volatile ("nop %0"::"i"(x));
+
+-#define ia64_itci(addr) asm volatile ("itc.i %0;;" :: "r"(addr) : "memory")
++#define __ia64_itci(addr) asm volatile ("itc.i %0;;" :: "r"(addr) : "memory")
+
+-#define ia64_itcd(addr) asm volatile ("itc.d %0;;" :: "r"(addr) : "memory")
++#define __ia64_itcd(addr) asm volatile ("itc.d %0;;" :: "r"(addr) : "memory")
+
+
+-#define ia64_itri(trnum, addr) asm volatile ("itr.i itr[%0]=%1" \
++#define __ia64_itri(trnum, addr) asm volatile ("itr.i itr[%0]=%1" \
+ :: "r"(trnum), "r"(addr) : "memory")
+
+-#define ia64_itrd(trnum, addr) asm volatile ("itr.d dtr[%0]=%1" \
++#define __ia64_itrd(trnum, addr) asm volatile ("itr.d dtr[%0]=%1" \
+ :: "r"(trnum), "r"(addr) : "memory")
+
+-#define ia64_tpa(addr) \
++#define __ia64_tpa(addr) \
+ ({ \
+ __u64 ia64_pa; \
+ asm volatile ("tpa %0 = %1" : "=r"(ia64_pa) : "r"(addr) : "memory"); \
+@@ -415,22 +415,22 @@ register unsigned long ia64_r13 asm ("r1
+ #define __ia64_set_dbr(index, val) \
+ asm volatile ("mov dbr[%0]=%1" :: "r"(index), "r"(val) : "memory")
+
+-#define ia64_set_ibr(index, val) \
++#define __ia64_set_ibr(index, val) \
+ asm volatile ("mov ibr[%0]=%1" :: "r"(index), "r"(val) : "memory")
+
+-#define ia64_set_pkr(index, val) \
++#define __ia64_set_pkr(index, val) \
+ asm volatile ("mov pkr[%0]=%1" :: "r"(index), "r"(val) : "memory")
+
+-#define ia64_set_pmc(index, val) \
++#define __ia64_set_pmc(index, val) \
+ asm volatile ("mov pmc[%0]=%1" :: "r"(index), "r"(val) : "memory")
+
+-#define ia64_set_pmd(index, val) \
++#define __ia64_set_pmd(index, val) \
+ asm volatile ("mov pmd[%0]=%1" :: "r"(index), "r"(val) : "memory")
+
+-#define ia64_set_rr(index, val) \
++#define __ia64_set_rr(index, val) \
+ asm volatile ("mov rr[%0]=%1" :: "r"(index), "r"(val) : "memory");
+
+-#define ia64_get_cpuid(index) \
++#define __ia64_get_cpuid(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=cpuid[%r1]" : "=r"(ia64_intri_res) : "rO"(index)); \
+@@ -444,21 +444,21 @@ register unsigned long ia64_r13 asm ("r1
+ ia64_intri_res; \
+ })
+
+-#define ia64_get_ibr(index) \
++#define __ia64_get_ibr(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=ibr[%1]" : "=r"(ia64_intri_res) : "r"(index)); \
+ ia64_intri_res; \
+ })
+
+-#define ia64_get_pkr(index) \
++#define __ia64_get_pkr(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=pkr[%1]" : "=r"(ia64_intri_res) : "r"(index)); \
+ ia64_intri_res; \
+ })
+
+-#define ia64_get_pmc(index) \
++#define __ia64_get_pmc(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=pmc[%1]" : "=r"(ia64_intri_res) : "r"(index)); \
+@@ -466,48 +466,48 @@ register unsigned long ia64_r13 asm ("r1
+ })
+
+
+-#define ia64_get_pmd(index) \
++#define __ia64_get_pmd(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=pmd[%1]" : "=r"(ia64_intri_res) : "r"(index)); \
+ ia64_intri_res; \
+ })
+
+-#define ia64_get_rr(index) \
++#define __ia64_get_rr(index) \
+ ({ \
+ __u64 ia64_intri_res; \
+ asm volatile ("mov %0=rr[%1]" : "=r"(ia64_intri_res) : "r" (index)); \
+ ia64_intri_res; \
+ })
+
+-#define ia64_fc(addr) asm volatile ("fc %0" :: "r"(addr) : "memory")
++#define __ia64_fc(addr) asm volatile ("fc %0" :: "r"(addr) : "memory")
+
+
+ #define ia64_sync_i() asm volatile (";; sync.i" ::: "memory")
+
+-#define ia64_ssm(mask) asm volatile ("ssm %0":: "i"((mask)) : "memory")
+-#define ia64_rsm(mask) asm volatile ("rsm %0":: "i"((mask)) : "memory")
++#define __ia64_ssm(mask) asm volatile ("ssm %0":: "i"((mask)) : "memory")
++#define __ia64_rsm(mask) asm volatile ("rsm %0":: "i"((mask)) : "memory")
+ #define ia64_sum(mask) asm volatile ("sum %0":: "i"((mask)) : "memory")
+ #define ia64_rum(mask) asm volatile ("rum %0":: "i"((mask)) : "memory")
+
+-#define ia64_ptce(addr) asm volatile ("ptc.e %0" :: "r"(addr))
++#define __ia64_ptce(addr) asm volatile ("ptc.e %0" :: "r"(addr))
+
+-#define ia64_ptcga(addr, size) \
++#define __ia64_ptcga(addr, size) \
+ do { \
+ asm volatile ("ptc.ga %0,%1" :: "r"(addr), "r"(size) : "memory"); \
+ ia64_dv_serialize_data(); \
+ } while (0)
+
+-#define ia64_ptcl(addr, size) \
++#define __ia64_ptcl(addr, size) \
+ do { \
+ asm volatile ("ptc.l %0,%1" :: "r"(addr), "r"(size) : "memory"); \
+ ia64_dv_serialize_data(); \
+ } while (0)
+
+-#define ia64_ptri(addr, size) \
++#define __ia64_ptri(addr, size) \
+ asm volatile ("ptr.i %0,%1" :: "r"(addr), "r"(size) : "memory")
+
+-#define ia64_ptrd(addr, size) \
++#define __ia64_ptrd(addr, size) \
+ asm volatile ("ptr.d %0,%1" :: "r"(addr), "r"(size) : "memory")
+
+ /* Values for lfhint in ia64_lfetch and ia64_lfetch_fault */
+@@ -589,7 +589,7 @@ do { \
+ } \
+ })
+
+-#define ia64_intrin_local_irq_restore(x) \
++#define __ia64_intrin_local_irq_restore(x) \
+ do { \
+ asm volatile (";; cmp.ne p6,p7=%0,r0;;" \
+ "(p6) ssm psr.i;" \
+@@ -598,4 +598,6 @@ do { \
+ :: "r"((x)) : "p6", "p7", "memory"); \
+ } while (0)
+
++#define __ia64_get_psr_i() (__ia64_getreg(_IA64_REG_PSR) & 0x4000UL)
++
+ #endif /* _ASM_IA64_GCC_INTRIN_H */
+diff -urNp linux-2.6/include/asm-ia64/hw_irq.h new/include/asm-ia64/hw_irq.h
+--- linux-2.6/include/asm-ia64/hw_irq.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/hw_irq.h 2006-06-28 14:32:14.000000000 +0200
+@@ -15,7 +15,11 @@
+ #include <asm/ptrace.h>
+ #include <asm/smp.h>
+
++#ifndef CONFIG_XEN
+ typedef u8 ia64_vector;
++#else
++typedef u16 ia64_vector;
++#endif
+
+ /*
+ * 0 special
+@@ -89,6 +93,13 @@ extern void register_percpu_irq (ia64_ve
+ static inline void
+ hw_resend_irq (struct hw_interrupt_type *h, unsigned int vector)
+ {
++#ifdef CONFIG_XEN
++ extern void resend_irq_on_evtchn(struct hw_interrupt_type *h,
++ unsigned int i);
++ if (is_running_on_xen())
++ resend_irq_on_evtchn(h, vector);
++ else
++#endif /* CONFIG_XEN */
+ platform_send_ipi(smp_processor_id(), vector, IA64_IPI_DM_INT, 0);
+ }
+
+diff -urNp linux-2.6/include/asm-ia64/hypercall.h new/include/asm-ia64/hypercall.h
+--- linux-2.6/include/asm-ia64/hypercall.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/hypercall.h 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,533 @@
++/******************************************************************************
++ * hypercall.h
++ *
++ * Linux-specific hypervisor handling.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __HYPERCALL_H__
++#define __HYPERCALL_H__
++
++#include <linux/string.h> /* memcpy() */
++
++#ifndef __HYPERVISOR_H__
++# error "please don't include this file directly"
++#endif
++
++/*
++ * Assembler stubs for hyper-calls.
++ */
++
++#define _hypercall0(type, name) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name) \
++ : "r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall1(type, name, a1) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "r" ((unsigned long)(a1)) \
++ : "r14","r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall2(type, name, a1, a2) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)) \
++ : "r14","r15","r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall3(type, name, a1, a2, a3) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)), \
++ "r" ((unsigned long)(a3)) \
++ : "r14","r15","r16","r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall4(type, name, a1, a2, a3, a4) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r17=%5\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)), \
++ "r" ((unsigned long)(a3)), \
++ "r" ((unsigned long)(a4)) \
++ : "r14","r15","r16","r2","r8", \
++ "r17","memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall5(type, name, a1, a2, a3, a4, a5) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r17=%5\n" \
++ "mov r18=%6\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)), \
++ "r" ((unsigned long)(a3)), \
++ "r" ((unsigned long)(a4)), \
++ "r" ((unsigned long)(a5)) \
++ : "r14","r15","r16","r2","r8", \
++ "r17","r18","memory" ); \
++ (type)__res; \
++})
++
++static inline int
++HYPERVISOR_sched_op_compat(
++ int cmd, unsigned long arg)
++{
++ return _hypercall2(int, sched_op_compat, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_sched_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, sched_op, cmd, arg);
++}
++
++static inline long
++HYPERVISOR_set_timer_op(
++ u64 timeout)
++{
++ unsigned long timeout_hi = (unsigned long)(timeout>>32);
++ unsigned long timeout_lo = (unsigned long)timeout;
++ return _hypercall2(long, set_timer_op, timeout_lo, timeout_hi);
++}
++
++static inline int
++HYPERVISOR_dom0_op(
++ dom0_op_t *dom0_op)
++{
++ dom0_op->interface_version = DOM0_INTERFACE_VERSION;
++ return _hypercall1(int, dom0_op, dom0_op);
++}
++
++static inline int
++HYPERVISOR_multicall(
++ void *call_list, int nr_calls)
++{
++ return _hypercall2(int, multicall, call_list, nr_calls);
++}
++
++#ifndef CONFIG_XEN_IA64_DOM0_VP
++static inline int
++HYPERVISOR_memory_op(
++ unsigned int cmd, void *arg)
++{
++ return _hypercall2(int, memory_op, cmd, arg);
++}
++#else
++//XXX xen/ia64 copy_from_guest() is broken.
++// This is a temporal work around until it is fixed.
++static inline int
++____HYPERVISOR_memory_op(
++ unsigned int cmd, void *arg)
++{
++ return _hypercall2(int, memory_op, cmd, arg);
++}
++
++#include <xen/interface/memory.h>
++int ia64_xenmem_reservation_op(unsigned long op,
++ struct xen_memory_reservation* reservation__);
++static inline int
++HYPERVISOR_memory_op(
++ unsigned int cmd, void *arg)
++{
++ switch (cmd) {
++ case XENMEM_increase_reservation:
++ case XENMEM_decrease_reservation:
++ case XENMEM_populate_physmap:
++ return ia64_xenmem_reservation_op(cmd,
++ (struct xen_memory_reservation*)arg);
++ default:
++ return ____HYPERVISOR_memory_op(cmd, arg);
++ }
++ /* NOTREACHED */
++}
++#endif
++
++static inline int
++HYPERVISOR_event_channel_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, event_channel_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct evtchn_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, event_channel_op_compat, &op);
++ }
++ return rc;
++}
++
++static inline int
++HYPERVISOR_acm_op(
++ unsigned int cmd, void *arg)
++{
++ return _hypercall2(int, acm_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_xen_version(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, xen_version, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_console_io(
++ int cmd, int count, char *str)
++{
++ return _hypercall3(int, console_io, cmd, count, str);
++}
++
++static inline int
++HYPERVISOR_physdev_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, physdev_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct physdev_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, physdev_op_compat, &op);
++ }
++ return rc;
++}
++
++//XXX __HYPERVISOR_grant_table_op is used for this hypercall constant.
++static inline int
++____HYPERVISOR_grant_table_op(
++ unsigned int cmd, void *uop, unsigned int count)
++{
++ return _hypercall3(int, grant_table_op, cmd, uop, count);
++}
++#ifndef CONFIG_XEN_IA64_DOM0_VP
++#define HYPERVISOR_grant_table_op(cmd, uop, count) \
++ ____HYPERVISOR_grant_table_op((cmd), (uop), (count))
++#else
++int HYPERVISOR_grant_table_op(unsigned int cmd, void *uop, unsigned int count);
++#endif
++
++static inline int
++HYPERVISOR_vcpu_op(
++ int cmd, int vcpuid, void *extra_args)
++{
++ return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args);
++}
++
++static inline int
++HYPERVISOR_suspend(
++ unsigned long srec)
++{
++ struct sched_shutdown sched_shutdown = {
++ .reason = SHUTDOWN_suspend
++ };
++
++ int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown,
++ &sched_shutdown, srec);
++
++ if (rc == -ENOSYS)
++ rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown,
++ SHUTDOWN_suspend, srec);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_callback_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, callback_op, cmd, arg);
++}
++
++extern fastcall unsigned int __do_IRQ(unsigned int irq, struct pt_regs *regs);
++static inline void exit_idle(void) {}
++#define do_IRQ(irq, regs) ({ \
++ irq_enter(); \
++ __do_IRQ((irq), (regs)); \
++ irq_exit(); \
++})
++
++#ifdef CONFIG_XEN_IA64_DOM0_VP
++#include <linux/err.h>
++#include <asm/xen/privop.h>
++
++#define _hypercall_imm1(type, name, imm, a1) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "i" (imm), \
++ "r" ((unsigned long)(a1)) \
++ : "r14","r15","r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall_imm2(type, name, imm, a1, a2) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "i" (imm), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)) \
++ : "r14","r15","r16","r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall_imm3(type, name, imm, a1, a2, a3) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r17=%5\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "i" (imm), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)), \
++ "r" ((unsigned long)(a3)) \
++ : "r14","r15","r16","r17", \
++ "r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall_imm4(type, name, imm, a1, a2, a3, a4) \
++({ \
++ long __res; \
++ __asm__ __volatile__ (";;\n" \
++ "mov r14=%2\n" \
++ "mov r15=%3\n" \
++ "mov r16=%4\n" \
++ "mov r17=%5\n" \
++ "mov r18=%6\n" \
++ "mov r2=%1\n" \
++ "break 0x1000 ;;\n" \
++ "mov %0=r8 ;;\n" \
++ : "=r" (__res) \
++ : "i" (__HYPERVISOR_##name), \
++ "i" (imm), \
++ "r" ((unsigned long)(a1)), \
++ "r" ((unsigned long)(a2)), \
++ "r" ((unsigned long)(a3)), \
++ "r" ((unsigned long)(a4)) \
++ : "r14","r15","r16","r17","r18", \
++ "r2","r8", \
++ "memory" ); \
++ (type)__res; \
++})
++
++static inline unsigned long
++__HYPERVISOR_ioremap(unsigned long ioaddr, unsigned long size)
++{
++ return _hypercall_imm2(unsigned long, ia64_dom0vp_op,
++ IA64_DOM0VP_ioremap, ioaddr, size);
++}
++
++static inline unsigned long
++HYPERVISOR_ioremap(unsigned long ioaddr, unsigned long size)
++{
++ unsigned long ret = ioaddr;
++ if (is_running_on_xen()) {
++ ret = __HYPERVISOR_ioremap(ioaddr, size);
++ if (unlikely(ret == -ENOSYS))
++ panic("hypercall %s failed with %ld. "
++ "Please check Xen and Linux config mismatch\n",
++ __func__, -ret);
++ else if (unlikely(IS_ERR_VALUE(ret)))
++ ret = ioaddr;
++ }
++ return ret;
++}
++
++static inline unsigned long
++__HYPERVISOR_phystomach(unsigned long gpfn)
++{
++ return _hypercall_imm1(unsigned long, ia64_dom0vp_op,
++ IA64_DOM0VP_phystomach, gpfn);
++}
++
++static inline unsigned long
++HYPERVISOR_phystomach(unsigned long gpfn)
++{
++ unsigned long ret = gpfn;
++ if (is_running_on_xen()) {
++ ret = __HYPERVISOR_phystomach(gpfn);
++ }
++ return ret;
++}
++
++static inline unsigned long
++__HYPERVISOR_machtophys(unsigned long mfn)
++{
++ return _hypercall_imm1(unsigned long, ia64_dom0vp_op,
++ IA64_DOM0VP_machtophys, mfn);
++}
++
++static inline unsigned long
++HYPERVISOR_machtophys(unsigned long mfn)
++{
++ unsigned long ret = mfn;
++ if (is_running_on_xen()) {
++ ret = __HYPERVISOR_machtophys(mfn);
++ }
++ return ret;
++}
++
++static inline unsigned long
++__HYPERVISOR_zap_physmap(unsigned long gpfn, unsigned int extent_order)
++{
++ return _hypercall_imm2(unsigned long, ia64_dom0vp_op,
++ IA64_DOM0VP_zap_physmap, gpfn, extent_order);
++}
++
++static inline unsigned long
++HYPERVISOR_zap_physmap(unsigned long gpfn, unsigned int extent_order)
++{
++ unsigned long ret = 0;
++ if (is_running_on_xen()) {
++ ret = __HYPERVISOR_zap_physmap(gpfn, extent_order);
++ }
++ return ret;
++}
++
++static inline unsigned long
++__HYPERVISOR_add_physmap(unsigned long gpfn, unsigned long mfn,
++ unsigned long flags, domid_t domid)
++{
++ return _hypercall_imm4(unsigned long, ia64_dom0vp_op,
++ IA64_DOM0VP_add_physmap, gpfn, mfn, flags,
++ domid);
++}
++
++static inline unsigned long
++HYPERVISOR_add_physmap(unsigned long gpfn, unsigned long mfn,
++ unsigned long flags, domid_t domid)
++{
++ unsigned long ret = 0;
++ BUG_ON(!is_running_on_xen());//XXX
++ if (is_running_on_xen()) {
++ ret = __HYPERVISOR_add_physmap(gpfn, mfn, flags, domid);
++ }
++ return ret;
++}
++
++// for balloon driver
++#define HYPERVISOR_update_va_mapping(va, new_val, flags) (0)
++
++#else
++#define HYPERVISOR_ioremap(ioaddr, size) (ioaddr)
++#define HYPERVISOR_phystomach(gpfn) (gpfn)
++#define HYPERVISOR_machtophys(mfn) (mfn)
++#define HYPERVISOR_zap_physmap(gpfn, extent_order) (0)
++#define HYPERVISOR_add_physmap(gpfn, mfn, flags) (0)
++#endif
++#endif /* __HYPERCALL_H__ */
+diff -urNp linux-2.6/include/asm-ia64/hypervisor.h new/include/asm-ia64/hypervisor.h
+--- linux-2.6/include/asm-ia64/hypervisor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/hypervisor.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,207 @@
++/******************************************************************************
++ * hypervisor.h
++ *
++ * Linux-specific hypervisor handling.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __HYPERVISOR_H__
++#define __HYPERVISOR_H__
++
++#ifndef CONFIG_XEN
++#define is_running_on_xen() (0)
++#define HYPERVISOR_ioremap(offset, size) (offset)
++#else
++extern int running_on_xen;
++#define is_running_on_xen() (running_on_xen)
++#endif
++
++#ifdef CONFIG_XEN
++#include <linux/config.h>
++#include <linux/types.h>
++#include <linux/kernel.h>
++#include <linux/version.h>
++#include <linux/errno.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/dom0_ops.h>
++#include <xen/interface/event_channel.h>
++#include <xen/interface/physdev.h>
++#include <xen/interface/sched.h>
++#include <asm/hypercall.h>
++#include <asm/ptrace.h>
++#include <asm/page.h>
++
++extern shared_info_t *HYPERVISOR_shared_info;
++extern start_info_t *xen_start_info;
++
++void force_evtchn_callback(void);
++
++/* Turn jiffies into Xen system time. XXX Implement me. */
++#define jiffies_to_st(j) 0
++
++static inline int
++HYPERVISOR_yield(
++ void)
++{
++ int rc = HYPERVISOR_sched_op(SCHEDOP_yield, NULL);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_block(
++ void)
++{
++ int rc = HYPERVISOR_sched_op(SCHEDOP_block, NULL);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_block, 0);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_shutdown(
++ unsigned int reason)
++{
++ struct sched_shutdown sched_shutdown = {
++ .reason = reason
++ };
++
++ int rc = HYPERVISOR_sched_op(SCHEDOP_shutdown, &sched_shutdown);
++
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_shutdown, reason);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_poll(
++ evtchn_port_t *ports, unsigned int nr_ports, u64 timeout)
++{
++ struct sched_poll sched_poll = {
++ .nr_ports = nr_ports,
++ .timeout = jiffies_to_st(timeout)
++ };
++
++ int rc;
++
++ set_xen_guest_handle(sched_poll.ports, ports);
++ rc = HYPERVISOR_sched_op(SCHEDOP_poll, &sched_poll);
++ if (rc == -ENOSYS)
++ rc = HYPERVISOR_sched_op_compat(SCHEDOP_yield, 0);
++
++ return rc;
++}
++
++#include <asm/hypercall.h>
++
++// for drivers/xen/privcmd/privcmd.c
++#define machine_to_phys_mapping 0
++#ifndef CONFIG_XEN_IA64_DOM0_VP
++#define direct_remap_pfn_range(a,b,c,d,e,f) remap_pfn_range(a,b,c,d,e)
++#define pfn_to_mfn(x) (x)
++#define mfn_to_pfn(x) (x)
++#else
++struct vm_area_struct;
++int direct_remap_pfn_range(struct vm_area_struct *vma,
++ unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid);
++struct file;
++int privcmd_mmap(struct file * file, struct vm_area_struct * vma);
++#define HAVE_ARCH_PRIVCMD_MMAP
++#endif
++
++// for drivers/xen/balloon/balloon.c
++#ifdef CONFIG_XEN_SCRUB_PAGES
++#define scrub_pages(_p,_n) memset((void *)(_p), 0, (_n) << PAGE_SHIFT)
++#else
++#define scrub_pages(_p,_n) ((void)0)
++#endif
++#define pte_mfn(_x) pte_pfn(_x)
++#define __pte_ma(_x) ((pte_t) {(_x)})
++#define phys_to_machine_mapping_valid(_x) (1)
++#define pfn_pte_ma(_x,_y) __pte_ma(0)
++#ifndef CONFIG_XEN_IA64_DOM0_VP //XXX
++#define set_phys_to_machine(_x,_y) do {} while (0)
++#define xen_machphys_update(_x,_y) do {} while (0)
++#endif
++
++#ifdef CONFIG_XEN_IA64_DOM0_VP
++int __xen_create_contiguous_region(unsigned long vstart, unsigned int order, unsigned int address_bits);
++static inline int
++xen_create_contiguous_region(unsigned long vstart,
++ unsigned int order, unsigned int address_bits)
++{
++ int ret = 0;
++ if (is_running_on_xen()) {
++ ret = __xen_create_contiguous_region(vstart, order,
++ address_bits);
++ }
++ return ret;
++}
++
++void __xen_destroy_contiguous_region(unsigned long vstart, unsigned int order);
++static inline void
++xen_destroy_contiguous_region(unsigned long vstart, unsigned int order)
++{
++ if (is_running_on_xen())
++ __xen_destroy_contiguous_region(vstart, order);
++}
++
++// for netfront.c, netback.c
++#define MULTI_UVMFLAGS_INDEX 0 //XXX any value
++
++static inline void
++MULTI_update_va_mapping(
++ multicall_entry_t *mcl, unsigned long va,
++ pte_t new_val, unsigned long flags)
++{
++ mcl->op = __HYPERVISOR_update_va_mapping;
++ mcl->result = 0;
++}
++
++#else
++#define xen_create_contiguous_region(vstart, order, address_bits) (0)
++#define xen_destroy_contiguous_region(vstart, order) do {} while (0)
++#endif
++
++// for debug
++asmlinkage int xprintk(const char *fmt, ...);
++#define xprintd(fmt, ...) xprintk("%s:%d " fmt, __func__, __LINE__, \
++ ##__VA_ARGS__)
++#endif /* CONFIG_XEN */
++
++#endif /* __HYPERVISOR_H__ */
+diff -urNp linux-2.6/include/asm-ia64/intel_intrin.h new/include/asm-ia64/intel_intrin.h
+--- linux-2.6/include/asm-ia64/intel_intrin.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/intel_intrin.h 2006-05-09 12:35:18.000000000 +0200
+@@ -16,8 +16,10 @@
+ * intrinsic
+ */
+
+-#define ia64_getreg __getReg
+-#define ia64_setreg __setReg
++#define __ia64_getreg __getReg
++#define __ia64_setreg __setReg
++
++#define __ia64_hint(x)
+
+ #define ia64_hint __hint
+ #define ia64_hint_pause __hint_pause
+@@ -33,16 +35,16 @@
+ #define ia64_getf_exp __getf_exp
+ #define ia64_shrp _m64_shrp
+
+-#define ia64_tpa __tpa
++#define __ia64_tpa __tpa
+ #define ia64_invala __invala
+ #define ia64_invala_gr __invala_gr
+ #define ia64_invala_fr __invala_fr
+ #define ia64_nop __nop
+ #define ia64_sum __sum
+-#define ia64_ssm __ssm
++#define __ia64_ssm __ssm
+ #define ia64_rum __rum
+-#define ia64_rsm __rsm
+-#define ia64_fc __fc
++#define __ia64_rsm __rsm
++#define __ia64_fc __fc
+
+ #define ia64_ldfs __ldfs
+ #define ia64_ldfd __ldfd
+@@ -80,24 +82,24 @@
+
+ #define __ia64_set_dbr(index, val) \
+ __setIndReg(_IA64_REG_INDR_DBR, index, val)
+-#define ia64_set_ibr(index, val) \
++#define __ia64_set_ibr(index, val) \
+ __setIndReg(_IA64_REG_INDR_IBR, index, val)
+-#define ia64_set_pkr(index, val) \
++#define __ia64_set_pkr(index, val) \
+ __setIndReg(_IA64_REG_INDR_PKR, index, val)
+-#define ia64_set_pmc(index, val) \
++#define __ia64_set_pmc(index, val) \
+ __setIndReg(_IA64_REG_INDR_PMC, index, val)
+-#define ia64_set_pmd(index, val) \
++#define __ia64_set_pmd(index, val) \
+ __setIndReg(_IA64_REG_INDR_PMD, index, val)
+-#define ia64_set_rr(index, val) \
++#define __ia64_set_rr(index, val) \
+ __setIndReg(_IA64_REG_INDR_RR, index, val)
+
+-#define ia64_get_cpuid(index) __getIndReg(_IA64_REG_INDR_CPUID, index)
++#define __ia64_get_cpuid(index) __getIndReg(_IA64_REG_INDR_CPUID, index)
+ #define __ia64_get_dbr(index) __getIndReg(_IA64_REG_INDR_DBR, index)
+-#define ia64_get_ibr(index) __getIndReg(_IA64_REG_INDR_IBR, index)
+-#define ia64_get_pkr(index) __getIndReg(_IA64_REG_INDR_PKR, index)
+-#define ia64_get_pmc(index) __getIndReg(_IA64_REG_INDR_PMC, index)
+-#define ia64_get_pmd(index) __getIndReg(_IA64_REG_INDR_PMD, index)
+-#define ia64_get_rr(index) __getIndReg(_IA64_REG_INDR_RR, index)
++#define __ia64_get_ibr(index) __getIndReg(_IA64_REG_INDR_IBR, index)
++#define __ia64_get_pkr(index) __getIndReg(_IA64_REG_INDR_PKR, index)
++#define __ia64_get_pmc(index) __getIndReg(_IA64_REG_INDR_PMC, index)
++#define __ia64_get_pmd(index) __getIndReg(_IA64_REG_INDR_PMD, index)
++#define __ia64_get_rr(index) __getIndReg(_IA64_REG_INDR_RR, index)
+
+ #define ia64_srlz_d __dsrlz
+ #define ia64_srlz_i __isrlz
+@@ -116,18 +118,18 @@
+ #define ia64_ld8_acq __ld8_acq
+
+ #define ia64_sync_i __synci
+-#define ia64_thash __thash
+-#define ia64_ttag __ttag
+-#define ia64_itcd __itcd
+-#define ia64_itci __itci
+-#define ia64_itrd __itrd
+-#define ia64_itri __itri
+-#define ia64_ptce __ptce
+-#define ia64_ptcl __ptcl
+-#define ia64_ptcg __ptcg
+-#define ia64_ptcga __ptcga
+-#define ia64_ptri __ptri
+-#define ia64_ptrd __ptrd
++#define __ia64_thash __thash
++#define __ia64_ttag __ttag
++#define __ia64_itcd __itcd
++#define __ia64_itci __itci
++#define __ia64_itrd __itrd
++#define __ia64_itri __itri
++#define __ia64_ptce __ptce
++#define __ia64_ptcl __ptcl
++#define __ia64_ptcg __ptcg
++#define __ia64_ptcga __ptcga
++#define __ia64_ptri __ptri
++#define __ia64_ptrd __ptrd
+ #define ia64_dep_mi _m64_dep_mi
+
+ /* Values for lfhint in __lfetch and __lfetch_fault */
+@@ -142,16 +144,18 @@
+ #define ia64_lfetch_fault __lfetch_fault
+ #define ia64_lfetch_fault_excl __lfetch_fault_excl
+
+-#define ia64_intrin_local_irq_restore(x) \
++#define __ia64_intrin_local_irq_restore(x) \
+ do { \
+ if ((x) != 0) { \
+- ia64_ssm(IA64_PSR_I); \
++ __ia64_ssm(IA64_PSR_I); \
+ ia64_srlz_d(); \
+ } else { \
+- ia64_rsm(IA64_PSR_I); \
++ __ia64_rsm(IA64_PSR_I); \
+ } \
+ } while (0)
+
++#define __ia64_get_psr_i() (__ia64_getreg(_IA64_REG_PSR) & 0x4000UL)
++
+ #define __builtin_trap() __break(0);
+
+ #endif /* _ASM_IA64_INTEL_INTRIN_H */
+diff -urNp linux-2.6/include/asm-ia64/io.h new/include/asm-ia64/io.h
+--- linux-2.6/include/asm-ia64/io.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/io.h 2006-06-28 14:32:14.000000000 +0200
+@@ -66,9 +66,11 @@ extern unsigned int num_io_spaces;
+ #define PIO_RESERVED __IA64_UNCACHED_OFFSET
+ #define HAVE_ARCH_PIO_SIZE
+
++#include <asm/hypervisor.h>
+ #include <asm/intrinsics.h>
+ #include <asm/machvec.h>
+ #include <asm/page.h>
++#include <asm/privop.h>
+ #include <asm/system.h>
+ #include <asm-generic/iomap.h>
+
+@@ -95,9 +97,41 @@ extern int valid_mmap_phys_addr_range (u
+ * The following two macros are deprecated and scheduled for removal.
+ * Please use the PCI-DMA interface defined in <asm/pci.h> instead.
+ */
++#ifndef CONFIG_XEN_IA64_DOM0_VP
+ #define bus_to_virt phys_to_virt
+ #define virt_to_bus virt_to_phys
+ #define page_to_bus page_to_phys
++#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT)
++#define page_to_pseudophys(page) page_to_phys(page)
++#else
++#define bus_to_virt(bus) \
++ phys_to_virt(machine_to_phys_for_dma(bus))
++#define virt_to_bus(virt) \
++ phys_to_machine_for_dma(virt_to_phys(virt))
++#define page_to_bus(page) \
++ phys_to_machine_for_dma(page_to_pseudophys(page))
++
++#define page_to_pseudophys(page) \
++ ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
++
++/*
++ * Drivers that use page_to_phys() for bus addresses are broken.
++ * This includes:
++ * drivers/ide/cris/ide-cris.c
++ * drivers/scsi/dec_esp.c
++ */
++#define page_to_phys(page) (page_to_pseudophys(page))
++#define bvec_to_bus(bv) (page_to_bus((bv)->bv_page) + \
++ (unsigned long) (bv)->bv_offset)
++#define bio_to_pseudophys(bio) (page_to_pseudophys(bio_page((bio))) + \
++ (unsigned long) bio_offset((bio)))
++#define bvec_to_pseudophys(bv) (page_to_pseudophys((bv)->bv_page) + \
++ (unsigned long) (bv)->bv_offset)
++#define BIOVEC_PHYS_MERGEABLE(vec1, vec2) \
++ (((bvec_to_bus((vec1)) + (vec1)->bv_len) == bvec_to_bus((vec2))) && \
++ ((bvec_to_pseudophys((vec1)) + (vec1)->bv_len) == \
++ bvec_to_pseudophys((vec2))))
++#endif
+
+ # endif /* KERNEL */
+
+@@ -416,14 +450,26 @@ __writeq (unsigned long val, volatile vo
+ # define outl_p outl
+ #endif
+
+-extern void __iomem * ioremap(unsigned long offset, unsigned long size);
+-extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size);
++/*
++ * An "address" in IO memory space is not clearly either an integer or a pointer. We will
++ * accept both, thus the casts.
++ *
++ * On ia-64, we access the physical I/O memory space through the uncached kernel region.
++ */
++static inline void __iomem *
++ioremap (unsigned long offset, unsigned long size)
++{
++ offset = HYPERVISOR_ioremap(offset, size);
++ return (void __iomem *) (__IA64_UNCACHED_OFFSET | (offset));
++}
+
+ static inline void
+ iounmap (volatile void __iomem *addr)
+ {
+ }
+
++#define ioremap_nocache(o,s) ioremap(o,s)
++
+ /* Use normal IO mappings for DMI */
+ #define dmi_ioremap ioremap
+ #define dmi_iounmap(x,l) iounmap(x)
+diff -urNp linux-2.6/include/asm-ia64/iosapic.h new/include/asm-ia64/iosapic.h
+--- linux-2.6/include/asm-ia64/iosapic.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/iosapic.h 2006-05-09 12:35:18.000000000 +0200
+@@ -53,6 +53,7 @@
+
+ #define NR_IOSAPICS 256
+
++#ifndef CONFIG_XEN
+ static inline unsigned int iosapic_read(char __iomem *iosapic, unsigned int reg)
+ {
+ writel(reg, iosapic + IOSAPIC_REG_SELECT);
+@@ -64,6 +65,7 @@ static inline void iosapic_write(char __
+ writel(reg, iosapic + IOSAPIC_REG_SELECT);
+ writel(val, iosapic + IOSAPIC_WINDOW);
+ }
++#endif
+
+ static inline void iosapic_eoi(char __iomem *iosapic, u32 vector)
+ {
+diff -urNp linux-2.6/include/asm-ia64/irq.h new/include/asm-ia64/irq.h
+--- linux-2.6/include/asm-ia64/irq.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/irq.h 2006-06-07 13:15:16.000000000 +0200
+@@ -11,8 +11,39 @@
+ * 02/29/00 D.Mosberger moved most things into hw_irq.h
+ */
+
++#ifndef CONFIG_XEN
+ #define NR_IRQS 256
+ #define NR_IRQ_VECTORS NR_IRQS
++#else
++/*
++ * The flat IRQ space is divided into two regions:
++ * 1. A one-to-one mapping of real physical IRQs. This space is only used
++ * if we have physical device-access privilege. This region is at the
++ * start of the IRQ space so that existing device drivers do not need
++ * to be modified to translate physical IRQ numbers into our IRQ space.
++ * 3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
++ * are bound using the provided bind/unbind functions.
++ */
++
++#define PIRQ_BASE 0
++#define NR_PIRQS 256
++
++#define DYNIRQ_BASE (PIRQ_BASE + NR_PIRQS)
++#define NR_DYNIRQS 256
++
++#define NR_IRQS (NR_PIRQS + NR_DYNIRQS)
++#define NR_IRQ_VECTORS NR_IRQS
++
++#define pirq_to_irq(_x) ((_x) + PIRQ_BASE)
++#define irq_to_pirq(_x) ((_x) - PIRQ_BASE)
++
++#define dynirq_to_irq(_x) ((_x) + DYNIRQ_BASE)
++#define irq_to_dynirq(_x) ((_x) - DYNIRQ_BASE)
++
++#define RESCHEDULE_VECTOR 0
++#define IPI_VECTOR 1
++#define NR_IPIS 2
++#endif /* CONFIG_XEN */
+
+ /*
+ * IRQ line status macro IRQ_PER_CPU is used
+diff -urNp linux-2.6/include/asm-ia64/machvec.h new/include/asm-ia64/machvec.h
+--- linux-2.6/include/asm-ia64/machvec.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/machvec.h 2006-05-09 12:35:19.000000000 +0200
+@@ -257,6 +257,21 @@ extern void machvec_init (const char *na
+ # error Unknown configuration. Update asm-ia64/machvec.h.
+ # endif /* CONFIG_IA64_GENERIC */
+
++#ifdef CONFIG_XEN_IA64_DOM0_VP
++# define platform_dma_map_sg dma_map_sg
++# define platform_dma_unmap_sg dma_unmap_sg
++# define platform_dma_mapping_error dma_mapping_error
++# define platform_dma_supported dma_supported
++# define platform_dma_alloc_coherent dma_alloc_coherent
++# define platform_dma_free_coherent dma_free_coherent
++# define platform_dma_map_single dma_map_single
++# define platform_dma_unmap_single dma_unmap_single
++# define platform_dma_sync_single_for_cpu \
++ dma_sync_single_for_cpu
++# define platform_dma_sync_single_for_device \
++ dma_sync_single_for_device
++#endif
++
+ /*
+ * Declare default routines which aren't declared anywhere else:
+ */
+diff -urNp linux-2.6/include/asm-ia64/meminit.h new/include/asm-ia64/meminit.h
+--- linux-2.6/include/asm-ia64/meminit.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/meminit.h 2006-05-09 12:35:19.000000000 +0200
+@@ -17,10 +17,15 @@
+ * - command line string
+ * - kernel code & data
+ * - Kernel memory map built from EFI memory map
++ * - xen start info
+ *
+ * More could be added if necessary
+ */
++#ifndef CONFIG_XEN
+ #define IA64_MAX_RSVD_REGIONS 6
++#else
++#define IA64_MAX_RSVD_REGIONS 7
++#endif
+
+ struct rsvd_region {
+ unsigned long start; /* virtual address of beginning of element */
+diff -urNp linux-2.6/include/asm-ia64/page.h new/include/asm-ia64/page.h
+--- linux-2.6/include/asm-ia64/page.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/page.h 2006-06-28 14:32:14.000000000 +0200
+@@ -127,7 +127,6 @@ extern unsigned long max_low_pfn;
+ # define pfn_valid(pfn) (((pfn) >= min_low_pfn) && ((pfn) < max_low_pfn) && ia64_pfn_valid(pfn))
+ #endif
+
+-#define page_to_phys(page) (page_to_pfn(page) << PAGE_SHIFT)
+ #define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
+ #define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT)
+
+@@ -229,4 +228,127 @@ get_order (unsigned long size)
+ (((current->personality & READ_IMPLIES_EXEC) != 0) \
+ ? VM_EXEC : 0))
+
++#ifndef __ASSEMBLY__
++#ifdef CONFIG_XEN
++
++#define INVALID_P2M_ENTRY (~0UL)
++
++#ifndef CONFIG_XEN_IA64_DOM0_VP
++
++#define virt_to_machine(v) __pa(v)
++#define machine_to_virt(m) __va(m)
++#define virt_to_mfn(v) ((__pa(v)) >> PAGE_SHIFT)
++#define mfn_to_virt(m) (__va((m) << PAGE_SHIFT))
++
++#else
++
++#include <linux/kernel.h>
++#include <asm/hypervisor.h>
++#include <xen/features.h> // to compile netback, netfront
++typedef unsigned long maddr_t; // to compile netback, netfront
++
++/*
++ * XXX hack!
++ * Linux/IA64 uses PG_arch_1.
++ * This hack will be removed once PG_foreign bit is taken.
++ * #include <xen/foreign_page.h>
++ */
++#ifdef __ASM_XEN_FOREIGN_PAGE_H__
++# error "don't include include/xen/foreign_page.h!"
++#endif
++
++extern struct address_space xen_ia64_foreign_dummy_mapping;
++#define PageForeign(page) \
++ ((page)->mapping == &xen_ia64_foreign_dummy_mapping)
++
++#define SetPageForeign(page, dtor) do { \
++ set_page_private((page), (unsigned long)(dtor)); \
++ (page)->mapping = &xen_ia64_foreign_dummy_mapping; \
++ smp_rmb(); \
++} while (0)
++
++#define ClearPageForeign(page) do { \
++ (page)->mapping = NULL; \
++ smp_rmb(); \
++ set_page_private((page), 0); \
++} while (0)
++
++#define PageForeignDestructor(page) \
++ ( (void (*) (struct page *)) page_private(page) )
++
++#define arch_free_page(_page,_order) \
++({ int foreign = PageForeign(_page); \
++ if (foreign) \
++ (PageForeignDestructor(_page))(_page); \
++ foreign; \
++})
++#define HAVE_ARCH_FREE_PAGE
++
++/* XXX xen page size != page size */
++
++static inline unsigned long
++pfn_to_mfn_for_dma(unsigned long pfn)
++{
++ unsigned long mfn;
++ mfn = HYPERVISOR_phystomach(pfn);
++ BUG_ON(mfn == 0); // XXX
++ BUG_ON(mfn == INVALID_P2M_ENTRY); // XXX
++ BUG_ON(mfn == INVALID_MFN);
++ return mfn;
++}
++
++static inline unsigned long
++phys_to_machine_for_dma(unsigned long phys)
++{
++ unsigned long machine =
++ pfn_to_mfn_for_dma(phys >> PAGE_SHIFT) << PAGE_SHIFT;
++ machine |= (phys & ~PAGE_MASK);
++ return machine;
++}
++
++static inline unsigned long
++mfn_to_pfn_for_dma(unsigned long mfn)
++{
++ unsigned long pfn;
++ pfn = HYPERVISOR_machtophys(mfn);
++ BUG_ON(pfn == 0);
++ //BUG_ON(pfn == INVALID_M2P_ENTRY);
++ return pfn;
++}
++
++static inline unsigned long
++machine_to_phys_for_dma(unsigned long machine)
++{
++ unsigned long phys =
++ mfn_to_pfn_for_dma(machine >> PAGE_SHIFT) << PAGE_SHIFT;
++ phys |= (machine & ~PAGE_MASK);
++ return phys;
++}
++
++#define set_phys_to_machine(pfn, mfn) do { } while (0)
++#define xen_machphys_update(mfn, pfn) do { } while (0)
++
++/* XXX to compile set_phys_to_machine(vaddr, FOREIGN_FRAME(m)) */
++#define FOREIGN_FRAME(m) (INVALID_P2M_ENTRY)
++
++#define mfn_to_pfn(mfn) (mfn)
++#define mfn_to_virt(mfn) (__va((mfn) << PAGE_SHIFT))
++#define pfn_to_mfn(pfn) (pfn)
++#define virt_to_mfn(virt) (__pa(virt) >> PAGE_SHIFT)
++#define virt_to_machine(virt) __pa(virt) // for tpmfront.c
++
++static inline unsigned long
++mfn_to_local_pfn(unsigned long mfn)
++{
++ extern unsigned long max_mapnr;
++ unsigned long pfn = mfn_to_pfn(mfn);
++ if (!pfn_valid(pfn))
++ return INVALID_P2M_ENTRY;
++ return pfn;
++}
++
++#endif /* CONFIG_XEN_IA64_DOM0_VP */
++#endif /* CONFIG_XEN */
++#endif /* __ASSEMBLY__ */
++
+ #endif /* _ASM_IA64_PAGE_H */
+diff -urNp linux-2.6/include/asm-ia64/pal.h new/include/asm-ia64/pal.h
+--- linux-2.6/include/asm-ia64/pal.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/pal.h 2006-05-09 12:35:19.000000000 +0200
+@@ -82,6 +82,7 @@
+ #ifndef __ASSEMBLY__
+
+ #include <linux/types.h>
++#include <asm/processor.h>
+ #include <asm/fpu.h>
+
+ /*
+diff -urNp linux-2.6/include/asm-ia64/pgalloc.h new/include/asm-ia64/pgalloc.h
+--- linux-2.6/include/asm-ia64/pgalloc.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/pgalloc.h 2006-05-09 12:35:19.000000000 +0200
+@@ -126,7 +126,7 @@ static inline void pmd_free(pmd_t * pmd)
+ static inline void
+ pmd_populate(struct mm_struct *mm, pmd_t * pmd_entry, struct page *pte)
+ {
+- pmd_val(*pmd_entry) = page_to_phys(pte);
++ pmd_val(*pmd_entry) = page_to_pseudophys(pte);
+ }
+
+ static inline void
+diff -urNp linux-2.6/include/asm-ia64/privop.h new/include/asm-ia64/privop.h
+--- linux-2.6/include/asm-ia64/privop.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/privop.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,59 @@
++#ifndef _ASM_IA64_PRIVOP_H
++#define _ASM_IA64_PRIVOP_H
++
++/*
++ * Copyright (C) 2005 Hewlett-Packard Co
++ * Dan Magenheimer <dan.magenheimer@hp.com>
++ *
++ */
++
++#ifdef CONFIG_XEN
++#include <asm/xen/privop.h>
++#endif
++
++#ifndef __ASSEMBLY
++
++#ifndef IA64_PARAVIRTUALIZED
++
++#define ia64_getreg __ia64_getreg
++#define ia64_setreg __ia64_setreg
++#define ia64_hint __ia64_hint
++#define ia64_thash __ia64_thash
++#define ia64_itci __ia64_itci
++#define ia64_itcd __ia64_itcd
++#define ia64_itri __ia64_itri
++#define ia64_itrd __ia64_itrd
++#define ia64_tpa __ia64_tpa
++#define ia64_set_ibr __ia64_set_ibr
++#define ia64_set_pkr __ia64_set_pkr
++#define ia64_set_pmc __ia64_set_pmc
++#define ia64_set_pmd __ia64_set_pmd
++#define ia64_set_rr __ia64_set_rr
++#define ia64_get_cpuid __ia64_get_cpuid
++#define ia64_get_ibr __ia64_get_ibr
++#define ia64_get_pkr __ia64_get_pkr
++#define ia64_get_pmc __ia64_get_pmc
++#define ia64_get_pmd __ia64_get_pmd
++#define ia64_get_rr __ia64_get_rr
++#define ia64_fc __ia64_fc
++#define ia64_ssm __ia64_ssm
++#define ia64_rsm __ia64_rsm
++#define ia64_ptce __ia64_ptce
++#define ia64_ptcga __ia64_ptcga
++#define ia64_ptcl __ia64_ptcl
++#define ia64_ptri __ia64_ptri
++#define ia64_ptrd __ia64_ptrd
++#define ia64_get_psr_i __ia64_get_psr_i
++#define ia64_intrin_local_irq_restore __ia64_intrin_local_irq_restore
++#define ia64_pal_halt_light __ia64_pal_halt_light
++#define ia64_leave_kernel __ia64_leave_kernel
++#define ia64_leave_syscall __ia64_leave_syscall
++#define ia64_trace_syscall __ia64_trace_syscall
++#define ia64_switch_to __ia64_switch_to
++#define ia64_pal_call_static __ia64_pal_call_static
++
++#endif /* !IA64_PARAVIRTUALIZED */
++
++#endif /* !__ASSEMBLY */
++
++#endif /* _ASM_IA64_PRIVOP_H */
+diff -urNp linux-2.6/include/asm-ia64/processor.h new/include/asm-ia64/processor.h
+--- linux-2.6/include/asm-ia64/processor.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/processor.h 2006-05-09 12:35:19.000000000 +0200
+@@ -19,6 +19,7 @@
+ #include <asm/kregs.h>
+ #include <asm/ptrace.h>
+ #include <asm/ustack.h>
++#include <asm/privop.h>
+
+ #define IA64_NUM_DBG_REGS 8
+ /*
+diff -urNp linux-2.6/include/asm-ia64/synch_bitops.h new/include/asm-ia64/synch_bitops.h
+--- linux-2.6/include/asm-ia64/synch_bitops.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/synch_bitops.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,63 @@
++#ifndef __XEN_SYNCH_BITOPS_H__
++#define __XEN_SYNCH_BITOPS_H__
++
++/*
++ * Copyright 1992, Linus Torvalds.
++ * Heavily modified to provide guaranteed strong synchronisation
++ * when communicating with Xen or other guest OSes running on other CPUs.
++ */
++
++#include <linux/config.h>
++
++#define ADDR (*(volatile long *) addr)
++
++static __inline__ void synch_set_bit(int nr, volatile void * addr)
++{
++ set_bit(nr, addr);
++}
++
++static __inline__ void synch_clear_bit(int nr, volatile void * addr)
++{
++ clear_bit(nr, addr);
++}
++
++static __inline__ void synch_change_bit(int nr, volatile void * addr)
++{
++ change_bit(nr, addr);
++}
++
++static __inline__ int synch_test_and_set_bit(int nr, volatile void * addr)
++{
++ return test_and_set_bit(nr, addr);
++}
++
++static __inline__ int synch_test_and_clear_bit(int nr, volatile void * addr)
++{
++ return test_and_clear_bit(nr, addr);
++}
++
++static __inline__ int synch_test_and_change_bit(int nr, volatile void * addr)
++{
++ return test_and_change_bit(nr, addr);
++}
++
++static __inline__ int synch_const_test_bit(int nr, const volatile void * addr)
++{
++ return test_bit(nr, addr);
++}
++
++static __inline__ int synch_var_test_bit(int nr, volatile void * addr)
++{
++ return test_bit(nr, addr);
++}
++
++#define synch_cmpxchg ia64_cmpxchg4_acq
++
++#define synch_test_bit(nr,addr) \
++(__builtin_constant_p(nr) ? \
++ synch_const_test_bit((nr),(addr)) : \
++ synch_var_test_bit((nr),(addr)))
++
++#define synch_cmpxchg_subword synch_cmpxchg
++
++#endif /* __XEN_SYNCH_BITOPS_H__ */
+diff -urNp linux-2.6/include/asm-ia64/system.h new/include/asm-ia64/system.h
+--- linux-2.6/include/asm-ia64/system.h 2006-07-03 14:15:09.000000000 +0200
++++ new/include/asm-ia64/system.h 2006-05-09 12:35:20.000000000 +0200
+@@ -125,7 +125,7 @@ extern struct ia64_boot_param {
+ #define __local_irq_save(x) \
+ do { \
+ ia64_stop(); \
+- (x) = ia64_getreg(_IA64_REG_PSR); \
++ (x) = ia64_get_psr_i(); \
+ ia64_stop(); \
+ ia64_rsm(IA64_PSR_I); \
+ } while (0)
+@@ -173,7 +173,7 @@ do { \
+ #endif /* !CONFIG_IA64_DEBUG_IRQ */
+
+ #define local_irq_enable() ({ ia64_stop(); ia64_ssm(IA64_PSR_I); ia64_srlz_d(); })
+-#define local_save_flags(flags) ({ ia64_stop(); (flags) = ia64_getreg(_IA64_REG_PSR); })
++#define local_save_flags(flags) ({ ia64_stop(); (flags) = ia64_get_psr_i(); })
+
+ #define irqs_disabled() \
+ ({ \
+diff -urNp linux-2.6/include/asm-ia64/xen/privop.h new/include/asm-ia64/xen/privop.h
+--- linux-2.6/include/asm-ia64/xen/privop.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-ia64/xen/privop.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,277 @@
++#ifndef _ASM_IA64_XEN_PRIVOP_H
++#define _ASM_IA64_XEN_PRIVOP_H
++
++/*
++ * Copyright (C) 2005 Hewlett-Packard Co
++ * Dan Magenheimer <dan.magenheimer@hp.com>
++ *
++ * Paravirtualizations of privileged operations for Xen/ia64
++ *
++ */
++
++
++#include <asm/xen/asm-xsi-offsets.h>
++#include <xen/interface/arch-ia64.h>
++
++#define IA64_PARAVIRTUALIZED
++
++#ifdef __ASSEMBLY__
++#define XEN_HYPER_RFI break HYPERPRIVOP_RFI
++#define XEN_HYPER_RSM_PSR_DT break HYPERPRIVOP_RSM_DT
++#define XEN_HYPER_SSM_PSR_DT break HYPERPRIVOP_SSM_DT
++#define XEN_HYPER_COVER break HYPERPRIVOP_COVER
++#define XEN_HYPER_ITC_D break HYPERPRIVOP_ITC_D
++#define XEN_HYPER_ITC_I break HYPERPRIVOP_ITC_I
++#define XEN_HYPER_SSM_I break HYPERPRIVOP_SSM_I
++#define XEN_HYPER_GET_IVR break HYPERPRIVOP_GET_IVR
++#define XEN_HYPER_GET_TPR break HYPERPRIVOP_GET_TPR
++#define XEN_HYPER_SET_TPR break HYPERPRIVOP_SET_TPR
++#define XEN_HYPER_EOI break HYPERPRIVOP_EOI
++#define XEN_HYPER_SET_ITM break HYPERPRIVOP_SET_ITM
++#define XEN_HYPER_THASH break HYPERPRIVOP_THASH
++#define XEN_HYPER_PTC_GA break HYPERPRIVOP_PTC_GA
++#define XEN_HYPER_ITR_D break HYPERPRIVOP_ITR_D
++#define XEN_HYPER_GET_RR break HYPERPRIVOP_GET_RR
++#define XEN_HYPER_SET_RR break HYPERPRIVOP_SET_RR
++#define XEN_HYPER_SET_KR break HYPERPRIVOP_SET_KR
++#define XEN_HYPER_FC break HYPERPRIVOP_FC
++#define XEN_HYPER_GET_CPUID break HYPERPRIVOP_GET_CPUID
++#define XEN_HYPER_GET_PMD break HYPERPRIVOP_GET_PMD
++#define XEN_HYPER_GET_EFLAG break HYPERPRIVOP_GET_EFLAG
++#define XEN_HYPER_SET_EFLAG break HYPERPRIVOP_SET_EFLAG
++#endif
++
++#ifndef __ASSEMBLY__
++#define XEN_HYPER_SSM_I asm("break %0" : : "i" (HYPERPRIVOP_SSM_I))
++#define XEN_HYPER_GET_IVR asm("break %0" : : "i" (HYPERPRIVOP_GET_IVR))
++
++/************************************************/
++/* Instructions paravirtualized for correctness */
++/************************************************/
++
++/* "fc" and "thash" are privilege-sensitive instructions, meaning they
++ * may have different semantics depending on whether they are executed
++ * at PL0 vs PL!=0. When paravirtualized, these instructions mustn't
++ * be allowed to execute directly, lest incorrect semantics result. */
++extern unsigned long xen_fc(unsigned long addr);
++#define ia64_fc(addr) xen_fc((unsigned long)(addr))
++extern unsigned long xen_thash(unsigned long addr);
++#define ia64_thash(addr) xen_thash((unsigned long)(addr))
++/* Note that "ttag" and "cover" are also privilege-sensitive; "ttag"
++ * is not currently used (though it may be in a long-format VHPT system!)
++ * and the semantics of cover only change if psr.ic is off which is very
++ * rare (and currently non-existent outside of assembly code */
++
++/* There are also privilege-sensitive registers. These registers are
++ * readable at any privilege level but only writable at PL0. */
++extern unsigned long xen_get_cpuid(int index);
++#define ia64_get_cpuid(i) xen_get_cpuid(i)
++extern unsigned long xen_get_pmd(int index);
++#define ia64_get_pmd(i) xen_get_pmd(i)
++extern unsigned long xen_get_eflag(void); /* see xen_ia64_getreg */
++extern void xen_set_eflag(unsigned long); /* see xen_ia64_setreg */
++
++/************************************************/
++/* Instructions paravirtualized for performance */
++/************************************************/
++
++/* Xen uses memory-mapped virtual privileged registers for access to many
++ * performance-sensitive privileged registers. Some, like the processor
++ * status register (psr), are broken up into multiple memory locations.
++ * Others, like "pend", are abstractions based on privileged registers.
++ * "Pend" is guaranteed to be set if reading cr.ivr would return a
++ * (non-spurious) interrupt. */
++#define XSI_PSR_I \
++ (*(uint64_t *)(XSI_PSR_I_ADDR))
++#define xen_get_virtual_psr_i() \
++ (!(*(uint8_t *)(XSI_PSR_I)))
++#define xen_set_virtual_psr_i(_val) \
++ ({ *(uint8_t *)(XSI_PSR_I) = (uint8_t)(_val) ? 0:1; })
++#define xen_set_virtual_psr_ic(_val) \
++ ({ *(int *)(XSI_PSR_IC) = _val ? 1:0; })
++#define xen_get_virtual_pend() (*(int *)(XSI_PEND))
++
++/* Hyperprivops are "break" instructions with a well-defined API.
++ * In particular, the virtual psr.ic bit must be off; in this way
++ * it is guaranteed to never conflict with a linux break instruction.
++ * Normally, this is done in a xen stub but this one is frequent enough
++ * that we inline it */
++#define xen_hyper_ssm_i() \
++({ \
++ xen_set_virtual_psr_i(0); \
++ xen_set_virtual_psr_ic(0); \
++ XEN_HYPER_SSM_I; \
++})
++
++/* turning off interrupts can be paravirtualized simply by writing
++ * to a memory-mapped virtual psr.i bit (implemented as a 16-bit bool) */
++#define xen_rsm_i() xen_set_virtual_psr_i(0)
++
++/* turning on interrupts is a bit more complicated.. write to the
++ * memory-mapped virtual psr.i bit first (to avoid race condition),
++ * then if any interrupts were pending, we have to execute a hyperprivop
++ * to ensure the pending interrupt gets delivered; else we're done! */
++#define xen_ssm_i() \
++({ \
++ int old = xen_get_virtual_psr_i(); \
++ xen_set_virtual_psr_i(1); \
++ if (!old && xen_get_virtual_pend()) xen_hyper_ssm_i(); \
++})
++
++#define xen_ia64_intrin_local_irq_restore(x) \
++{ \
++ if (is_running_on_xen()) { \
++ if ((x) & IA64_PSR_I) { xen_ssm_i(); } \
++ else { xen_rsm_i(); } \
++ } \
++ else __ia64_intrin_local_irq_restore((x)); \
++}
++
++#define xen_get_psr_i() \
++( \
++ (is_running_on_xen()) ? \
++ (xen_get_virtual_psr_i() ? IA64_PSR_I : 0) \
++ : __ia64_get_psr_i() \
++)
++
++#define xen_ia64_ssm(mask) \
++{ \
++ if ((mask)==IA64_PSR_I) { \
++ if (is_running_on_xen()) { xen_ssm_i(); } \
++ else { __ia64_ssm(mask); } \
++ } \
++ else { __ia64_ssm(mask); } \
++}
++
++#define xen_ia64_rsm(mask) \
++{ \
++ if ((mask)==IA64_PSR_I) { \
++ if (is_running_on_xen()) { xen_rsm_i(); } \
++ else { __ia64_rsm(mask); } \
++ } \
++ else { __ia64_rsm(mask); } \
++}
++
++
++/* Although all privileged operations can be left to trap and will
++ * be properly handled by Xen, some are frequent enough that we use
++ * hyperprivops for performance. */
++
++extern unsigned long xen_get_ivr(void);
++extern unsigned long xen_get_tpr(void);
++extern void xen_set_itm(unsigned long);
++extern void xen_set_tpr(unsigned long);
++extern void xen_eoi(void);
++extern void xen_set_rr(unsigned long index, unsigned long val);
++extern unsigned long xen_get_rr(unsigned long index);
++extern void xen_set_kr(unsigned long index, unsigned long val);
++extern void xen_ptcga(unsigned long addr, unsigned long size);
++
++/* Note: It may look wrong to test for is_running_on_xen() in each case.
++ * However regnum is always a constant so, as written, the compiler
++ * eliminates the switch statement, whereas is_running_on_xen() must be
++ * tested dynamically. */
++#define xen_ia64_getreg(regnum) \
++({ \
++ __u64 ia64_intri_res; \
++ \
++ switch(regnum) { \
++ case _IA64_REG_CR_IVR: \
++ ia64_intri_res = (is_running_on_xen()) ? \
++ xen_get_ivr() : \
++ __ia64_getreg(regnum); \
++ break; \
++ case _IA64_REG_CR_TPR: \
++ ia64_intri_res = (is_running_on_xen()) ? \
++ xen_get_tpr() : \
++ __ia64_getreg(regnum); \
++ break; \
++ case _IA64_REG_AR_EFLAG: \
++ ia64_intri_res = (is_running_on_xen()) ? \
++ xen_get_eflag() : \
++ __ia64_getreg(regnum); \
++ break; \
++ default: \
++ ia64_intri_res = __ia64_getreg(regnum); \
++ break; \
++ } \
++ ia64_intri_res; \
++})
++
++#define xen_ia64_setreg(regnum,val) \
++({ \
++ switch(regnum) { \
++ case _IA64_REG_AR_KR0 ... _IA64_REG_AR_KR7: \
++ (is_running_on_xen()) ? \
++ xen_set_kr((regnum-_IA64_REG_AR_KR0), val) : \
++ __ia64_setreg(regnum,val); \
++ break; \
++ case _IA64_REG_CR_ITM: \
++ (is_running_on_xen()) ? \
++ xen_set_itm(val) : \
++ __ia64_setreg(regnum,val); \
++ break; \
++ case _IA64_REG_CR_TPR: \
++ (is_running_on_xen()) ? \
++ xen_set_tpr(val) : \
++ __ia64_setreg(regnum,val); \
++ break; \
++ case _IA64_REG_CR_EOI: \
++ (is_running_on_xen()) ? \
++ xen_eoi() : \
++ __ia64_setreg(regnum,val); \
++ break; \
++ case _IA64_REG_AR_EFLAG: \
++ (is_running_on_xen()) ? \
++ xen_set_eflag(val) : \
++ __ia64_setreg(regnum,val); \
++ break; \
++ default: \
++ __ia64_setreg(regnum,val); \
++ break; \
++ } \
++})
++
++#define ia64_ssm xen_ia64_ssm
++#define ia64_rsm xen_ia64_rsm
++#define ia64_intrin_local_irq_restore xen_ia64_intrin_local_irq_restore
++#define ia64_ptcga xen_ptcga
++#define ia64_set_rr(index,val) xen_set_rr(index,val)
++#define ia64_get_rr(index) xen_get_rr(index)
++#define ia64_getreg xen_ia64_getreg
++#define ia64_setreg xen_ia64_setreg
++#define ia64_get_psr_i xen_get_psr_i
++
++/* the remainder of these are not performance-sensitive so its
++ * OK to not paravirtualize and just take a privop trap and emulate */
++#define ia64_hint __ia64_hint
++#define ia64_set_pmd __ia64_set_pmd
++#define ia64_itci __ia64_itci
++#define ia64_itcd __ia64_itcd
++#define ia64_itri __ia64_itri
++#define ia64_itrd __ia64_itrd
++#define ia64_tpa __ia64_tpa
++#define ia64_set_ibr __ia64_set_ibr
++#define ia64_set_pkr __ia64_set_pkr
++#define ia64_set_pmc __ia64_set_pmc
++#define ia64_get_ibr __ia64_get_ibr
++#define ia64_get_pkr __ia64_get_pkr
++#define ia64_get_pmc __ia64_get_pmc
++#define ia64_ptce __ia64_ptce
++#define ia64_ptcl __ia64_ptcl
++#define ia64_ptri __ia64_ptri
++#define ia64_ptrd __ia64_ptrd
++
++#endif /* !__ASSEMBLY__ */
++
++/* these routines utilize privilege-sensitive or performance-sensitive
++ * privileged instructions so the code must be replaced with
++ * paravirtualized versions */
++#define ia64_pal_halt_light xen_pal_halt_light
++#define ia64_leave_kernel xen_leave_kernel
++#define ia64_leave_syscall xen_leave_syscall
++#define ia64_trace_syscall xen_trace_syscall
++#define ia64_switch_to xen_switch_to
++#define ia64_pal_call_static xen_pal_call_static
++
++#endif /* _ASM_IA64_XEN_PRIVOP_H */
+diff -urNp linux-2.6/include/asm-um/page.h new/include/asm-um/page.h
+--- linux-2.6/include/asm-um/page.h 2006-07-03 14:15:14.000000000 +0200
++++ new/include/asm-um/page.h 2006-05-09 12:35:40.000000000 +0200
+@@ -115,7 +115,7 @@ extern unsigned long uml_physmem;
+ extern struct page *arch_validate(struct page *page, gfp_t mask, int order);
+ #define HAVE_ARCH_VALIDATE
+
+-extern void arch_free_page(struct page *page, int order);
++extern int arch_free_page(struct page *page, int order);
+ #define HAVE_ARCH_FREE_PAGE
+
+ #include <asm-generic/memory_model.h>
+diff -urNp linux-2.6/include/asm-x86_64/apic.h new/include/asm-x86_64/apic.h
+--- linux-2.6/include/asm-x86_64/apic.h 2006-07-03 14:15:14.000000000 +0200
++++ new/include/asm-x86_64/apic.h 2006-05-09 12:35:41.000000000 +0200
+@@ -105,11 +105,13 @@ extern int disable_timer_pin_1;
+
+ extern void setup_threshold_lvt(unsigned long lvt_off);
+
++#ifndef CONFIG_XEN
+ void smp_send_timer_broadcast_ipi(void);
+ void switch_APIC_timer_to_ipi(void *cpumask);
+ void switch_ipi_to_APIC_timer(void *cpumask);
+
+ #define ARCH_APICTIMER_STOPS_ON_C3 1
++#endif
+
+ #endif /* CONFIG_X86_LOCAL_APIC */
+
+diff -urNp linux-2.6/include/asm-x86_64/hw_irq.h new/include/asm-x86_64/hw_irq.h
+--- linux-2.6/include/asm-x86_64/hw_irq.h 2006-07-03 14:15:15.000000000 +0200
++++ new/include/asm-x86_64/hw_irq.h 2006-05-09 12:35:41.000000000 +0200
+@@ -127,7 +127,7 @@ asmlinkage void IRQ_NAME(nr); \
+ __asm__( \
+ "\n.p2align\n" \
+ "IRQ" #nr "_interrupt:\n\t" \
+- "push $" #nr "-256 ; " \
++ "push $~(" #nr ") ; " \
+ "jmp common_interrupt");
+
+ #if defined(CONFIG_X86_IO_APIC)
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/arch_hooks.h new/include/asm-x86_64/mach-xen/asm/arch_hooks.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/arch_hooks.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/arch_hooks.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,27 @@
++#ifndef _ASM_ARCH_HOOKS_H
++#define _ASM_ARCH_HOOKS_H
++
++#include <linux/interrupt.h>
++
++/*
++ * linux/include/asm/arch_hooks.h
++ *
++ * define the architecture specific hooks
++ */
++
++/* these aren't arch hooks, they are generic routines
++ * that can be used by the hooks */
++extern void init_ISA_irqs(void);
++extern void apic_intr_init(void);
++extern void smp_intr_init(void);
++extern irqreturn_t timer_interrupt(int irq, void *dev_id, struct pt_regs *regs);
++
++/* these are the defined hooks */
++extern void intr_init_hook(void);
++extern void pre_intr_init_hook(void);
++extern void pre_setup_arch_hook(void);
++extern void trap_init_hook(void);
++extern void time_init_hook(void);
++extern void mca_nmi_hook(void);
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/bootsetup.h new/include/asm-x86_64/mach-xen/asm/bootsetup.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/bootsetup.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/bootsetup.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,42 @@
++
++#ifndef _X86_64_BOOTSETUP_H
++#define _X86_64_BOOTSETUP_H 1
++
++#define BOOT_PARAM_SIZE 4096
++extern char x86_boot_params[BOOT_PARAM_SIZE];
++
++/*
++ * This is set up by the setup-routine at boot-time
++ */
++#define PARAM ((unsigned char *)x86_boot_params)
++#define SCREEN_INFO (*(struct screen_info *) (PARAM+0))
++#define EXT_MEM_K (*(unsigned short *) (PARAM+2))
++#define ALT_MEM_K (*(unsigned int *) (PARAM+0x1e0))
++#define E820_MAP_NR (*(char*) (PARAM+E820NR))
++#define E820_MAP ((struct e820entry *) (PARAM+E820MAP))
++#define APM_BIOS_INFO (*(struct apm_bios_info *) (PARAM+0x40))
++#define DRIVE_INFO (*(struct drive_info_struct *) (PARAM+0x80))
++#define SYS_DESC_TABLE (*(struct sys_desc_table_struct*)(PARAM+0xa0))
++#define MOUNT_ROOT_RDONLY (*(unsigned short *) (PARAM+0x1F2))
++#define RAMDISK_FLAGS (*(unsigned short *) (PARAM+0x1F8))
++#define SAVED_VIDEO_MODE (*(unsigned short *) (PARAM+0x1FA))
++#define ORIG_ROOT_DEV (*(unsigned short *) (PARAM+0x1FC))
++#define AUX_DEVICE_INFO (*(unsigned char *) (PARAM+0x1FF))
++#define LOADER_TYPE (*(unsigned char *) (PARAM+0x210))
++#define KERNEL_START (*(unsigned int *) (PARAM+0x214))
++
++#define INITRD_START (__pa(xen_start_info->mod_start))
++#define INITRD_SIZE (xen_start_info->mod_len)
++#define EDID_INFO (*(struct edid_info *) (PARAM+0x440))
++
++#define EDD_NR (*(unsigned char *) (PARAM+EDDNR))
++#define EDD_MBR_SIG_NR (*(unsigned char *) (PARAM+EDD_MBR_SIG_NR_BUF))
++#define EDD_MBR_SIGNATURE ((unsigned int *) (PARAM+EDD_MBR_SIG_BUF))
++#define EDD_BUF ((struct edd_info *) (PARAM+EDDBUF))
++#define COMMAND_LINE saved_command_line
++
++#define RAMDISK_IMAGE_START_MASK 0x07FF
++#define RAMDISK_PROMPT_FLAG 0x8000
++#define RAMDISK_LOAD_FLAG 0x4000
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/desc.h new/include/asm-x86_64/mach-xen/asm/desc.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/desc.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/desc.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,263 @@
++/* Written 2000 by Andi Kleen */
++#ifndef __ARCH_DESC_H
++#define __ARCH_DESC_H
++
++#include <linux/threads.h>
++#include <asm/ldt.h>
++
++#ifndef __ASSEMBLY__
++
++#include <linux/string.h>
++#include <linux/smp.h>
++
++#include <asm/segment.h>
++#include <asm/mmu.h>
++
++// 8 byte segment descriptor
++struct desc_struct {
++ u16 limit0;
++ u16 base0;
++ unsigned base1 : 8, type : 4, s : 1, dpl : 2, p : 1;
++ unsigned limit : 4, avl : 1, l : 1, d : 1, g : 1, base2 : 8;
++} __attribute__((packed));
++
++struct n_desc_struct {
++ unsigned int a,b;
++};
++
++enum {
++ GATE_INTERRUPT = 0xE,
++ GATE_TRAP = 0xF,
++ GATE_CALL = 0xC,
++};
++
++// 16byte gate
++struct gate_struct {
++ u16 offset_low;
++ u16 segment;
++ unsigned ist : 3, zero0 : 5, type : 5, dpl : 2, p : 1;
++ u16 offset_middle;
++ u32 offset_high;
++ u32 zero1;
++} __attribute__((packed));
++
++#define PTR_LOW(x) ((unsigned long)(x) & 0xFFFF)
++#define PTR_MIDDLE(x) (((unsigned long)(x) >> 16) & 0xFFFF)
++#define PTR_HIGH(x) ((unsigned long)(x) >> 32)
++
++enum {
++ DESC_TSS = 0x9,
++ DESC_LDT = 0x2,
++};
++
++// LDT or TSS descriptor in the GDT. 16 bytes.
++struct ldttss_desc {
++ u16 limit0;
++ u16 base0;
++ unsigned base1 : 8, type : 5, dpl : 2, p : 1;
++ unsigned limit1 : 4, zero0 : 3, g : 1, base2 : 8;
++ u32 base3;
++ u32 zero1;
++} __attribute__((packed));
++
++struct desc_ptr {
++ unsigned short size;
++ unsigned long address;
++} __attribute__((packed)) ;
++
++extern struct desc_ptr idt_descr, cpu_gdt_descr[NR_CPUS];
++
++extern struct desc_struct cpu_gdt_table[GDT_ENTRIES];
++
++#define load_TR_desc() asm volatile("ltr %w0"::"r" (GDT_ENTRY_TSS*8))
++#define load_LDT_desc() asm volatile("lldt %w0"::"r" (GDT_ENTRY_LDT*8))
++
++static inline void clear_LDT(void)
++{
++ int cpu = get_cpu();
++
++ /*
++ * NB. We load the default_ldt for lcall7/27 handling on demand, as
++ * it slows down context switching. Noone uses it anyway.
++ */
++ cpu = cpu; /* XXX avoid compiler warning */
++ xen_set_ldt(0UL, 0);
++ put_cpu();
++}
++
++/*
++ * This is the ldt that every process will get unless we need
++ * something other than this.
++ */
++extern struct desc_struct default_ldt[];
++#ifndef CONFIG_X86_NO_IDT
++extern struct gate_struct idt_table[];
++#endif
++extern struct desc_ptr cpu_gdt_descr[];
++
++/* the cpu gdt accessor */
++#define cpu_gdt(_cpu) ((struct desc_struct *)cpu_gdt_descr[_cpu].address)
++
++static inline void _set_gate(void *adr, unsigned type, unsigned long func, unsigned dpl, unsigned ist)
++{
++ struct gate_struct s;
++ s.offset_low = PTR_LOW(func);
++ s.segment = __KERNEL_CS;
++ s.ist = ist;
++ s.p = 1;
++ s.dpl = dpl;
++ s.zero0 = 0;
++ s.zero1 = 0;
++ s.type = type;
++ s.offset_middle = PTR_MIDDLE(func);
++ s.offset_high = PTR_HIGH(func);
++ /* does not need to be atomic because it is only done once at setup time */
++ memcpy(adr, &s, 16);
++}
++
++#ifndef CONFIG_X86_NO_IDT
++static inline void set_intr_gate(int nr, void *func)
++{
++ BUG_ON((unsigned)nr > 0xFF);
++ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, 0);
++}
++
++static inline void set_intr_gate_ist(int nr, void *func, unsigned ist)
++{
++ BUG_ON((unsigned)nr > 0xFF);
++ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 0, ist);
++}
++
++static inline void set_system_gate(int nr, void *func)
++{
++ BUG_ON((unsigned)nr > 0xFF);
++ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, 0);
++}
++
++static inline void set_system_gate_ist(int nr, void *func, unsigned ist)
++{
++ _set_gate(&idt_table[nr], GATE_INTERRUPT, (unsigned long) func, 3, ist);
++}
++#endif
++
++static inline void set_tssldt_descriptor(void *ptr, unsigned long tss, unsigned type,
++ unsigned size)
++{
++ struct ldttss_desc d;
++ memset(&d,0,sizeof(d));
++ d.limit0 = size & 0xFFFF;
++ d.base0 = PTR_LOW(tss);
++ d.base1 = PTR_MIDDLE(tss) & 0xFF;
++ d.type = type;
++ d.p = 1;
++ d.limit1 = (size >> 16) & 0xF;
++ d.base2 = (PTR_MIDDLE(tss) >> 8) & 0xFF;
++ d.base3 = PTR_HIGH(tss);
++ memcpy(ptr, &d, 16);
++}
++
++#ifndef CONFIG_X86_NO_TSS
++static inline void set_tss_desc(unsigned cpu, void *addr)
++{
++ /*
++ * sizeof(unsigned long) coming from an extra "long" at the end
++ * of the iobitmap. See tss_struct definition in processor.h
++ *
++ * -1? seg base+limit should be pointing to the address of the
++ * last valid byte
++ */
++ set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_TSS],
++ (unsigned long)addr, DESC_TSS,
++ IO_BITMAP_OFFSET + IO_BITMAP_BYTES + sizeof(unsigned long) - 1);
++}
++#endif
++
++static inline void set_ldt_desc(unsigned cpu, void *addr, int size)
++{
++ set_tssldt_descriptor(&cpu_gdt(cpu)[GDT_ENTRY_LDT], (unsigned long)addr,
++ DESC_LDT, size * 8 - 1);
++}
++
++static inline void set_seg_base(unsigned cpu, int entry, void *base)
++{
++ struct desc_struct *d = &cpu_gdt(cpu)[entry];
++ u32 addr = (u32)(u64)base;
++ BUG_ON((u64)base >> 32);
++ d->base0 = addr & 0xffff;
++ d->base1 = (addr >> 16) & 0xff;
++ d->base2 = (addr >> 24) & 0xff;
++}
++
++#define LDT_entry_a(info) \
++ ((((info)->base_addr & 0x0000ffff) << 16) | ((info)->limit & 0x0ffff))
++/* Don't allow setting of the lm bit. It is useless anyways because
++ 64bit system calls require __USER_CS. */
++#define LDT_entry_b(info) \
++ (((info)->base_addr & 0xff000000) | \
++ (((info)->base_addr & 0x00ff0000) >> 16) | \
++ ((info)->limit & 0xf0000) | \
++ (((info)->read_exec_only ^ 1) << 9) | \
++ ((info)->contents << 10) | \
++ (((info)->seg_not_present ^ 1) << 15) | \
++ ((info)->seg_32bit << 22) | \
++ ((info)->limit_in_pages << 23) | \
++ ((info)->useable << 20) | \
++ /* ((info)->lm << 21) | */ \
++ 0x7000)
++
++#define LDT_empty(info) (\
++ (info)->base_addr == 0 && \
++ (info)->limit == 0 && \
++ (info)->contents == 0 && \
++ (info)->read_exec_only == 1 && \
++ (info)->seg_32bit == 0 && \
++ (info)->limit_in_pages == 0 && \
++ (info)->seg_not_present == 1 && \
++ (info)->useable == 0 && \
++ (info)->lm == 0)
++
++#if TLS_SIZE != 24
++# error update this code.
++#endif
++
++static inline void load_TLS(struct thread_struct *t, unsigned int cpu)
++{
++#if 0
++ u64 *gdt = (u64 *)(cpu_gdt(cpu) + GDT_ENTRY_TLS_MIN);
++ gdt[0] = t->tls_array[0];
++ gdt[1] = t->tls_array[1];
++ gdt[2] = t->tls_array[2];
++#endif
++#define C(i) \
++ HYPERVISOR_update_descriptor(virt_to_machine(&cpu_gdt(cpu)[GDT_ENTRY_TLS_MIN + i]), t->tls_array[i])
++
++ C(0); C(1); C(2);
++#undef C
++}
++
++/*
++ * load one particular LDT into the current CPU
++ */
++static inline void load_LDT_nolock (mm_context_t *pc, int cpu)
++{
++ void *segments = pc->ldt;
++ int count = pc->size;
++
++ if (likely(!count))
++ segments = NULL;
++
++ xen_set_ldt((unsigned long)segments, count);
++}
++
++static inline void load_LDT(mm_context_t *pc)
++{
++ int cpu = get_cpu();
++ load_LDT_nolock(pc, cpu);
++ put_cpu();
++}
++
++extern struct desc_ptr idt_descr;
++
++#endif /* !__ASSEMBLY__ */
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/dma-mapping.h new/include/asm-x86_64/mach-xen/asm/dma-mapping.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/dma-mapping.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/dma-mapping.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,191 @@
++#ifndef _X8664_DMA_MAPPING_H
++#define _X8664_DMA_MAPPING_H 1
++
++/*
++ * IOMMU interface. See Documentation/DMA-mapping.txt and DMA-API.txt for
++ * documentation.
++ */
++
++#include <linux/config.h>
++
++#include <asm/scatterlist.h>
++#include <asm/io.h>
++#include <asm/swiotlb.h>
++
++struct dma_mapping_ops {
++ int (*mapping_error)(dma_addr_t dma_addr);
++ void* (*alloc_coherent)(struct device *dev, size_t size,
++ dma_addr_t *dma_handle, gfp_t gfp);
++ void (*free_coherent)(struct device *dev, size_t size,
++ void *vaddr, dma_addr_t dma_handle);
++ dma_addr_t (*map_single)(struct device *hwdev, void *ptr,
++ size_t size, int direction);
++ /* like map_single, but doesn't check the device mask */
++ dma_addr_t (*map_simple)(struct device *hwdev, char *ptr,
++ size_t size, int direction);
++ void (*unmap_single)(struct device *dev, dma_addr_t addr,
++ size_t size, int direction);
++ void (*sync_single_for_cpu)(struct device *hwdev,
++ dma_addr_t dma_handle, size_t size,
++ int direction);
++ void (*sync_single_for_device)(struct device *hwdev,
++ dma_addr_t dma_handle, size_t size,
++ int direction);
++ void (*sync_single_range_for_cpu)(struct device *hwdev,
++ dma_addr_t dma_handle, unsigned long offset,
++ size_t size, int direction);
++ void (*sync_single_range_for_device)(struct device *hwdev,
++ dma_addr_t dma_handle, unsigned long offset,
++ size_t size, int direction);
++ void (*sync_sg_for_cpu)(struct device *hwdev,
++ struct scatterlist *sg, int nelems,
++ int direction);
++ void (*sync_sg_for_device)(struct device *hwdev,
++ struct scatterlist *sg, int nelems,
++ int direction);
++ int (*map_sg)(struct device *hwdev, struct scatterlist *sg,
++ int nents, int direction);
++ void (*unmap_sg)(struct device *hwdev,
++ struct scatterlist *sg, int nents,
++ int direction);
++ int (*dma_supported)(struct device *hwdev, u64 mask);
++ int is_phys;
++};
++
++extern dma_addr_t bad_dma_address;
++extern struct dma_mapping_ops* dma_ops;
++extern int iommu_merge;
++
++#if 0
++static inline int dma_mapping_error(dma_addr_t dma_addr)
++{
++ if (dma_ops->mapping_error)
++ return dma_ops->mapping_error(dma_addr);
++
++ return (dma_addr == bad_dma_address);
++}
++
++extern void *dma_alloc_coherent(struct device *dev, size_t size,
++ dma_addr_t *dma_handle, gfp_t gfp);
++extern void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
++ dma_addr_t dma_handle);
++
++static inline dma_addr_t
++dma_map_single(struct device *hwdev, void *ptr, size_t size,
++ int direction)
++{
++ return dma_ops->map_single(hwdev, ptr, size, direction);
++}
++
++static inline void
++dma_unmap_single(struct device *dev, dma_addr_t addr,size_t size,
++ int direction)
++{
++ dma_ops->unmap_single(dev, addr, size, direction);
++}
++
++#define dma_map_page(dev,page,offset,size,dir) \
++ dma_map_single((dev), page_address(page)+(offset), (size), (dir))
++
++#define dma_unmap_page dma_unmap_single
++
++static inline void
++dma_sync_single_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
++ size_t size, int direction)
++{
++ if (dma_ops->sync_single_for_cpu)
++ dma_ops->sync_single_for_cpu(hwdev, dma_handle, size,
++ direction);
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_single_for_device(struct device *hwdev, dma_addr_t dma_handle,
++ size_t size, int direction)
++{
++ if (dma_ops->sync_single_for_device)
++ dma_ops->sync_single_for_device(hwdev, dma_handle, size,
++ direction);
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dma_handle,
++ unsigned long offset, size_t size, int direction)
++{
++ if (dma_ops->sync_single_range_for_cpu) {
++ dma_ops->sync_single_range_for_cpu(hwdev, dma_handle, offset, size, direction);
++ }
++
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_single_range_for_device(struct device *hwdev, dma_addr_t dma_handle,
++ unsigned long offset, size_t size, int direction)
++{
++ if (dma_ops->sync_single_range_for_device)
++ dma_ops->sync_single_range_for_device(hwdev, dma_handle,
++ offset, size, direction);
++
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
++ int nelems, int direction)
++{
++ if (dma_ops->sync_sg_for_cpu)
++ dma_ops->sync_sg_for_cpu(hwdev, sg, nelems, direction);
++ flush_write_buffers();
++}
++
++static inline void
++dma_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
++ int nelems, int direction)
++{
++ if (dma_ops->sync_sg_for_device) {
++ dma_ops->sync_sg_for_device(hwdev, sg, nelems, direction);
++ }
++
++ flush_write_buffers();
++}
++
++static inline int
++dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents, int direction)
++{
++ return dma_ops->map_sg(hwdev, sg, nents, direction);
++}
++
++static inline void
++dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
++ int direction)
++{
++ dma_ops->unmap_sg(hwdev, sg, nents, direction);
++}
++
++extern int dma_supported(struct device *hwdev, u64 mask);
++
++/* same for gart, swiotlb, and nommu */
++static inline int dma_get_cache_alignment(void)
++{
++ return boot_cpu_data.x86_clflush_size;
++}
++
++#define dma_is_consistent(h) 1
++
++extern int dma_set_mask(struct device *dev, u64 mask);
++
++static inline void
++dma_cache_sync(void *vaddr, size_t size, enum dma_data_direction dir)
++{
++ flush_write_buffers();
++}
++
++extern struct device fallback_dev;
++extern int panic_on_overflow;
++#endif
++
++#endif /* _X8664_DMA_MAPPING_H */
++
++#include <asm-i386/mach-xen/asm/dma-mapping.h>
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/dmi.h new/include/asm-x86_64/mach-xen/asm/dmi.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/dmi.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/dmi.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,29 @@
++#ifndef _ASM_DMI_H
++#define _ASM_DMI_H 1
++
++#include <asm/io.h>
++
++extern void *dmi_ioremap(unsigned long addr, unsigned long size);
++extern void dmi_iounmap(void *addr, unsigned long size);
++extern void *bt_ioremap(unsigned long addr, unsigned long size);
++extern void bt_iounmap(void *addr, unsigned long size);
++
++#define DMI_MAX_DATA 2048
++
++extern int dmi_alloc_index;
++extern char dmi_alloc_data[DMI_MAX_DATA];
++
++/* This is so early that there is no good way to allocate dynamic memory.
++ Allocate data in an BSS array. */
++static inline void *dmi_alloc(unsigned len)
++{
++ int idx = dmi_alloc_index;
++ if ((dmi_alloc_index += len) > DMI_MAX_DATA)
++ return NULL;
++ return dmi_alloc_data + idx;
++}
++
++#define dmi_ioremap bt_ioremap
++#define dmi_iounmap bt_iounmap
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/e820.h new/include/asm-x86_64/mach-xen/asm/e820.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/e820.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/e820.h 2006-06-05 15:54:33.000000000 +0200
+@@ -0,0 +1,63 @@
++/*
++ * structures and definitions for the int 15, ax=e820 memory map
++ * scheme.
++ *
++ * In a nutshell, setup.S populates a scratch table in the
++ * empty_zero_block that contains a list of usable address/size
++ * duples. setup.c, this information is transferred into the e820map,
++ * and in init.c/numa.c, that new information is used to mark pages
++ * reserved or not.
++ */
++#ifndef __E820_HEADER
++#define __E820_HEADER
++
++#include <linux/mmzone.h>
++
++#define E820MAP 0x2d0 /* our map */
++#define E820MAX 128 /* number of entries in E820MAP */
++#define E820NR 0x1e8 /* # entries in E820MAP */
++
++#define E820_RAM 1
++#define E820_RESERVED 2
++#define E820_ACPI 3 /* usable as RAM once ACPI tables have been read */
++#define E820_NVS 4
++
++#define HIGH_MEMORY (1024*1024)
++
++#define LOWMEMSIZE() (0x9f000)
++
++#ifndef __ASSEMBLY__
++struct e820entry {
++ u64 addr; /* start of memory segment */
++ u64 size; /* size of memory segment */
++ u32 type; /* type of memory segment */
++} __attribute__((packed));
++
++struct e820map {
++ int nr_map;
++ struct e820entry map[E820MAX];
++};
++
++extern unsigned long find_e820_area(unsigned long start, unsigned long end,
++ unsigned size);
++extern void add_memory_region(unsigned long start, unsigned long size,
++ int type);
++extern void setup_memory_region(void);
++extern void contig_e820_setup(void);
++extern unsigned long e820_end_of_ram(void);
++extern void e820_reserve_resources(struct e820entry *e820, int nr_map);
++extern void e820_print_map(char *who);
++extern int e820_mapped(unsigned long start, unsigned long end, unsigned type);
++
++extern void e820_bootmem_free(pg_data_t *pgdat, unsigned long start,unsigned long end);
++extern void e820_setup_gap(struct e820entry *e820, int nr_map);
++extern unsigned long e820_hole_size(unsigned long start_pfn,
++ unsigned long end_pfn);
++
++extern void __init parse_memopt(char *p, char **end);
++extern void __init parse_memmapopt(char *p, char **end);
++
++extern struct e820map e820;
++#endif/*!__ASSEMBLY__*/
++
++#endif/*__E820_HEADER*/
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/fixmap.h new/include/asm-x86_64/mach-xen/asm/fixmap.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/fixmap.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/fixmap.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,114 @@
++/*
++ * fixmap.h: compile-time virtual memory allocation
++ *
++ * This file is subject to the terms and conditions of the GNU General Public
++ * License. See the file "COPYING" in the main directory of this archive
++ * for more details.
++ *
++ * Copyright (C) 1998 Ingo Molnar
++ */
++
++#ifndef _ASM_FIXMAP_H
++#define _ASM_FIXMAP_H
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <asm/apicdef.h>
++#include <xen/gnttab.h>
++#include <asm/page.h>
++#include <asm/vsyscall.h>
++#include <asm/vsyscall32.h>
++#include <asm/acpi.h>
++
++/*
++ * Here we define all the compile-time 'special' virtual
++ * addresses. The point is to have a constant address at
++ * compile time, but to set the physical address only
++ * in the boot process.
++ *
++ * these 'compile-time allocated' memory buffers are
++ * fixed-size 4k pages. (or larger if used with an increment
++ * highger than 1) use fixmap_set(idx,phys) to associate
++ * physical memory with fixmap indices.
++ *
++ * TLB entries of such buffers will not be flushed across
++ * task switches.
++ */
++
++enum fixed_addresses {
++ VSYSCALL_LAST_PAGE,
++ VSYSCALL_FIRST_PAGE = VSYSCALL_LAST_PAGE + ((VSYSCALL_END-VSYSCALL_START) >> PAGE_SHIFT) - 1,
++ VSYSCALL_HPET,
++ FIX_HPET_BASE,
++#ifdef CONFIG_X86_LOCAL_APIC
++ FIX_APIC_BASE, /* local (CPU) APIC) -- required for SMP or not */
++#endif
++#ifdef CONFIG_X86_IO_APIC
++ FIX_IO_APIC_BASE_0,
++ FIX_IO_APIC_BASE_END = FIX_IO_APIC_BASE_0 + MAX_IO_APICS-1,
++#endif
++#ifdef CONFIG_ACPI
++ FIX_ACPI_BEGIN,
++ FIX_ACPI_END = FIX_ACPI_BEGIN + FIX_ACPI_PAGES - 1,
++#endif
++ FIX_SHARED_INFO,
++#define NR_FIX_ISAMAPS 256
++ FIX_ISAMAP_END,
++ FIX_ISAMAP_BEGIN = FIX_ISAMAP_END + NR_FIX_ISAMAPS - 1,
++ __end_of_permanent_fixed_addresses,
++ /* temporary boot-time mappings, used before ioremap() is functional */
++#define NR_FIX_BTMAPS 16
++ FIX_BTMAP_END = __end_of_permanent_fixed_addresses,
++ FIX_BTMAP_BEGIN = FIX_BTMAP_END + NR_FIX_BTMAPS - 1,
++ __end_of_fixed_addresses
++};
++
++extern void __set_fixmap (enum fixed_addresses idx,
++ unsigned long phys, pgprot_t flags);
++
++#define set_fixmap(idx, phys) \
++ __set_fixmap(idx, phys, PAGE_KERNEL)
++/*
++ * Some hardware wants to get fixmapped without caching.
++ */
++#define set_fixmap_nocache(idx, phys) \
++ __set_fixmap(idx, phys, PAGE_KERNEL_NOCACHE)
++
++#define clear_fixmap(idx) \
++ __set_fixmap(idx, 0, __pgprot(0))
++
++#define FIXADDR_TOP (VSYSCALL_END-PAGE_SIZE)
++#define FIXADDR_SIZE (__end_of_fixed_addresses << PAGE_SHIFT)
++#define FIXADDR_START (FIXADDR_TOP - FIXADDR_SIZE)
++
++/* Only covers 32bit vsyscalls currently. Need another set for 64bit. */
++#define FIXADDR_USER_START ((unsigned long)VSYSCALL32_VSYSCALL)
++#define FIXADDR_USER_END (FIXADDR_USER_START + PAGE_SIZE)
++
++#define __fix_to_virt(x) (FIXADDR_TOP - ((x) << PAGE_SHIFT))
++
++extern void __this_fixmap_does_not_exist(void);
++
++/*
++ * 'index to address' translation. If anyone tries to use the idx
++ * directly without translation, we catch the bug with a NULL-deference
++ * kernel oops. Illegal ranges of incoming indices are caught too.
++ */
++static __always_inline unsigned long fix_to_virt(const unsigned int idx)
++{
++ /*
++ * this branch gets completely eliminated after inlining,
++ * except when someone tries to use fixaddr indices in an
++ * illegal way. (such as mixing up address types or using
++ * out-of-range indices).
++ *
++ * If it doesn't get removed, the linker will complain
++ * loudly with a reasonably clear error message..
++ */
++ if (idx >= __end_of_fixed_addresses)
++ __this_fixmap_does_not_exist();
++
++ return __fix_to_virt(idx);
++}
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/floppy.h new/include/asm-x86_64/mach-xen/asm/floppy.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/floppy.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/floppy.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,206 @@
++/*
++ * Architecture specific parts of the Floppy driver
++ *
++ * This file is subject to the terms and conditions of the GNU General Public
++ * License. See the file "COPYING" in the main directory of this archive
++ * for more details.
++ *
++ * Copyright (C) 1995
++ *
++ * Modifications for Xen are Copyright (c) 2004, Keir Fraser.
++ */
++#ifndef __ASM_XEN_X86_64_FLOPPY_H
++#define __ASM_XEN_X86_64_FLOPPY_H
++
++#include <linux/vmalloc.h>
++
++/*
++ * The DMA channel used by the floppy controller cannot access data at
++ * addresses >= 16MB
++ *
++ * Went back to the 1MB limit, as some people had problems with the floppy
++ * driver otherwise. It doesn't matter much for performance anyway, as most
++ * floppy accesses go through the track buffer.
++ */
++#define _CROSS_64KB(a,s,vdma) \
++(!(vdma) && ((unsigned long)(a)/K_64 != ((unsigned long)(a) + (s) - 1) / K_64))
++
++/* XEN: Hit DMA paths on the head. This trick from asm-m68k/floppy.h. */
++#include <asm/dma.h>
++#undef MAX_DMA_ADDRESS
++#define MAX_DMA_ADDRESS 0
++#define CROSS_64KB(a,s) (0)
++
++#define fd_inb(port) inb_p(port)
++#define fd_outb(value,port) outb_p(value,port)
++
++#define fd_request_dma() (0)
++#define fd_free_dma() ((void)0)
++#define fd_enable_irq() enable_irq(FLOPPY_IRQ)
++#define fd_disable_irq() disable_irq(FLOPPY_IRQ)
++#define fd_free_irq() free_irq(FLOPPY_IRQ, NULL)
++#define fd_get_dma_residue() vdma_get_dma_residue(FLOPPY_DMA)
++/*
++ * Do not use vmalloc/vfree: floppy_release_irq_and_dma() gets called from
++ * softirq context via motor_off_callback. A generic bug we happen to trigger.
++ */
++#define fd_dma_mem_alloc(size) __get_free_pages(GFP_KERNEL|__GFP_NORETRY, get_order(size))
++#define fd_dma_mem_free(addr, size) free_pages(addr, get_order(size))
++#define fd_dma_setup(addr, size, mode, io) vdma_dma_setup(addr, size, mode, io)
++
++static int virtual_dma_count;
++static int virtual_dma_residue;
++static char *virtual_dma_addr;
++static int virtual_dma_mode;
++static int doing_pdma;
++
++static irqreturn_t floppy_hardint(int irq, void *dev_id, struct pt_regs * regs)
++{
++ register unsigned char st;
++
++#undef TRACE_FLPY_INT
++
++#ifdef TRACE_FLPY_INT
++ static int calls=0;
++ static int bytes=0;
++ static int dma_wait=0;
++#endif
++ if (!doing_pdma)
++ return floppy_interrupt(irq, dev_id, regs);
++
++#ifdef TRACE_FLPY_INT
++ if(!calls)
++ bytes = virtual_dma_count;
++#endif
++
++ {
++ register int lcount;
++ register char *lptr;
++
++ st = 1;
++ for(lcount=virtual_dma_count, lptr=virtual_dma_addr;
++ lcount; lcount--, lptr++) {
++ st=inb(virtual_dma_port+4) & 0xa0 ;
++ if(st != 0xa0)
++ break;
++ if(virtual_dma_mode)
++ outb_p(*lptr, virtual_dma_port+5);
++ else
++ *lptr = inb_p(virtual_dma_port+5);
++ }
++ virtual_dma_count = lcount;
++ virtual_dma_addr = lptr;
++ st = inb(virtual_dma_port+4);
++ }
++
++#ifdef TRACE_FLPY_INT
++ calls++;
++#endif
++ if(st == 0x20)
++ return IRQ_HANDLED;
++ if(!(st & 0x20)) {
++ virtual_dma_residue += virtual_dma_count;
++ virtual_dma_count=0;
++#ifdef TRACE_FLPY_INT
++ printk("count=%x, residue=%x calls=%d bytes=%d dma_wait=%d\n",
++ virtual_dma_count, virtual_dma_residue, calls, bytes,
++ dma_wait);
++ calls = 0;
++ dma_wait=0;
++#endif
++ doing_pdma = 0;
++ floppy_interrupt(irq, dev_id, regs);
++ return IRQ_HANDLED;
++ }
++#ifdef TRACE_FLPY_INT
++ if(!virtual_dma_count)
++ dma_wait++;
++#endif
++ return IRQ_HANDLED;
++}
++
++static void fd_disable_dma(void)
++{
++ doing_pdma = 0;
++ virtual_dma_residue += virtual_dma_count;
++ virtual_dma_count=0;
++}
++
++static int vdma_get_dma_residue(unsigned int dummy)
++{
++ return virtual_dma_count + virtual_dma_residue;
++}
++
++
++static int fd_request_irq(void)
++{
++ return request_irq(FLOPPY_IRQ, floppy_hardint,SA_INTERRUPT,
++ "floppy", NULL);
++}
++
++#if 0
++static unsigned long vdma_mem_alloc(unsigned long size)
++{
++ return (unsigned long) vmalloc(size);
++
++}
++
++static void vdma_mem_free(unsigned long addr, unsigned long size)
++{
++ vfree((void *)addr);
++}
++#endif
++
++static int vdma_dma_setup(char *addr, unsigned long size, int mode, int io)
++{
++ doing_pdma = 1;
++ virtual_dma_port = io;
++ virtual_dma_mode = (mode == DMA_MODE_WRITE);
++ virtual_dma_addr = addr;
++ virtual_dma_count = size;
++ virtual_dma_residue = 0;
++ return 0;
++}
++
++/* XEN: This trick to force 'virtual DMA' is from include/asm-m68k/floppy.h. */
++#define FDC1 xen_floppy_init()
++static int FDC2 = -1;
++
++static int xen_floppy_init(void)
++{
++ use_virtual_dma = 1;
++ can_use_virtual_dma = 1;
++ return 0x3f0;
++}
++
++/*
++ * Floppy types are stored in the rtc's CMOS RAM and so rtc_lock
++ * is needed to prevent corrupted CMOS RAM in case "insmod floppy"
++ * coincides with another rtc CMOS user. Paul G.
++ */
++#define FLOPPY0_TYPE ({ \
++ unsigned long flags; \
++ unsigned char val; \
++ spin_lock_irqsave(&rtc_lock, flags); \
++ val = (CMOS_READ(0x10) >> 4) & 15; \
++ spin_unlock_irqrestore(&rtc_lock, flags); \
++ val; \
++})
++
++#define FLOPPY1_TYPE ({ \
++ unsigned long flags; \
++ unsigned char val; \
++ spin_lock_irqsave(&rtc_lock, flags); \
++ val = CMOS_READ(0x10) & 15; \
++ spin_unlock_irqrestore(&rtc_lock, flags); \
++ val; \
++})
++
++#define N_FDC 2
++#define N_DRIVE 8
++
++#define FLOPPY_MOTOR_MASK 0xf0
++
++#define EXTRA_FLOPPY_PARAMS
++
++#endif /* __ASM_XEN_X86_64_FLOPPY_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/hw_irq.h new/include/asm-x86_64/mach-xen/asm/hw_irq.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/hw_irq.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/hw_irq.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,145 @@
++#ifndef _ASM_HW_IRQ_H
++#define _ASM_HW_IRQ_H
++
++/*
++ * linux/include/asm/hw_irq.h
++ *
++ * (C) 1992, 1993 Linus Torvalds, (C) 1997 Ingo Molnar
++ *
++ * moved some of the old arch/i386/kernel/irq.h to here. VY
++ *
++ * IRQ/IPI changes taken from work by Thomas Radke
++ * <tomsoft@informatik.tu-chemnitz.de>
++ *
++ * hacked by Andi Kleen for x86-64.
++ *
++ * $Id$
++ */
++
++#ifndef __ASSEMBLY__
++#include <linux/config.h>
++#include <asm/atomic.h>
++#include <asm/irq.h>
++#include <linux/profile.h>
++#include <linux/smp.h>
++
++struct hw_interrupt_type;
++#endif
++
++#define NMI_VECTOR 0x02
++/*
++ * IDT vectors usable for external interrupt sources start
++ * at 0x20:
++ */
++#define FIRST_EXTERNAL_VECTOR 0x20
++
++#define IA32_SYSCALL_VECTOR 0x80
++
++
++/*
++ * Vectors 0x20-0x2f are used for ISA interrupts.
++ */
++
++/*
++ * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
++ *
++ * some of the following vectors are 'rare', they are merged
++ * into a single vector (CALL_FUNCTION_VECTOR) to save vector space.
++ * TLB, reschedule and local APIC vectors are performance-critical.
++ */
++#ifndef CONFIG_XEN
++#define SPURIOUS_APIC_VECTOR 0xff
++#define ERROR_APIC_VECTOR 0xfe
++#define RESCHEDULE_VECTOR 0xfd
++#define CALL_FUNCTION_VECTOR 0xfc
++/* fb free - please don't readd KDB here because it's useless
++ (hint - think what a NMI bit does to a vector) */
++#define THERMAL_APIC_VECTOR 0xfa
++#define THRESHOLD_APIC_VECTOR 0xf9
++/* f8 free */
++#define INVALIDATE_TLB_VECTOR_END 0xf7
++#define INVALIDATE_TLB_VECTOR_START 0xf0 /* f0-f7 used for TLB flush */
++
++#define NUM_INVALIDATE_TLB_VECTORS 8
++#endif
++
++/*
++ * Local APIC timer IRQ vector is on a different priority level,
++ * to work around the 'lost local interrupt if more than 2 IRQ
++ * sources per level' errata.
++ */
++#define LOCAL_TIMER_VECTOR 0xef
++
++/*
++ * First APIC vector available to drivers: (vectors 0x30-0xee)
++ * we start at 0x31 to spread out vectors evenly between priority
++ * levels. (0x80 is the syscall vector)
++ */
++#define FIRST_DEVICE_VECTOR 0x31
++#define FIRST_SYSTEM_VECTOR 0xef /* duplicated in irq.h */
++
++
++#ifndef __ASSEMBLY__
++extern u8 irq_vector[NR_IRQ_VECTORS];
++#define IO_APIC_VECTOR(irq) (irq_vector[irq])
++#define AUTO_ASSIGN -1
++
++/*
++ * Various low-level irq details needed by irq.c, process.c,
++ * time.c, io_apic.c and smp.c
++ *
++ * Interrupt entry/exit code at both C and assembly level
++ */
++
++extern void disable_8259A_irq(unsigned int irq);
++extern void enable_8259A_irq(unsigned int irq);
++extern int i8259A_irq_pending(unsigned int irq);
++extern void make_8259A_irq(unsigned int irq);
++extern void init_8259A(int aeoi);
++extern void FASTCALL(send_IPI_self(int vector));
++extern void init_VISWS_APIC_irqs(void);
++extern void setup_IO_APIC(void);
++extern void disable_IO_APIC(void);
++extern void print_IO_APIC(void);
++extern int IO_APIC_get_PCI_irq_vector(int bus, int slot, int fn);
++extern void send_IPI(int dest, int vector);
++extern void setup_ioapic_dest(void);
++
++extern unsigned long io_apic_irqs;
++
++extern atomic_t irq_err_count;
++extern atomic_t irq_mis_count;
++
++#define IO_APIC_IRQ(x) (((x) >= 16) || ((1<<(x)) & io_apic_irqs))
++
++#define __STR(x) #x
++#define STR(x) __STR(x)
++
++#include <asm/ptrace.h>
++
++#define IRQ_NAME2(nr) nr##_interrupt(void)
++#define IRQ_NAME(nr) IRQ_NAME2(IRQ##nr)
++
++/*
++ * SMP has a few special interrupts for IPI messages
++ */
++
++#define BUILD_IRQ(nr) \
++asmlinkage void IRQ_NAME(nr); \
++__asm__( \
++"\n.p2align\n" \
++"IRQ" #nr "_interrupt:\n\t" \
++ "push $" #nr "-256 ; " \
++ "jmp common_interrupt");
++
++extern void resend_irq_on_evtchn(struct hw_interrupt_type *h, unsigned int i);
++static inline void hw_resend_irq(struct hw_interrupt_type *h, unsigned int i)
++{
++ resend_irq_on_evtchn(h, i);
++}
++
++#define platform_legacy_irq(irq) ((irq) < 16)
++
++#endif
++
++#endif /* _ASM_HW_IRQ_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/hypercall.h new/include/asm-x86_64/mach-xen/asm/hypercall.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/hypercall.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/hypercall.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,372 @@
++/******************************************************************************
++ * hypercall.h
++ *
++ * Linux-specific hypervisor handling.
++ *
++ * Copyright (c) 2002-2004, K A Fraser
++ *
++ * 64-bit updates:
++ * Benjamin Liu <benjamin.liu@intel.com>
++ * Jun Nakajima <jun.nakajima@intel.com>
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __HYPERCALL_H__
++#define __HYPERCALL_H__
++
++#include <linux/string.h> /* memcpy() */
++
++#ifndef __HYPERVISOR_H__
++# error "please don't include this file directly"
++#endif
++
++#define __STR(x) #x
++#define STR(x) __STR(x)
++
++#define _hypercall0(type, name) \
++({ \
++ long __res; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res) \
++ : \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall1(type, name, a1) \
++({ \
++ long __res, __ign1; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=D" (__ign1) \
++ : "1" ((long)(a1)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall2(type, name, a1, a2) \
++({ \
++ long __res, __ign1, __ign2; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=D" (__ign1), "=S" (__ign2) \
++ : "1" ((long)(a1)), "2" ((long)(a2)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall3(type, name, a1, a2, a3) \
++({ \
++ long __res, __ign1, __ign2, __ign3; \
++ asm volatile ( \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
++ "=d" (__ign3) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)) \
++ : "memory" ); \
++ (type)__res; \
++})
++
++#define _hypercall4(type, name, a1, a2, a3, a4) \
++({ \
++ long __res, __ign1, __ign2, __ign3; \
++ asm volatile ( \
++ "movq %7,%%r10; " \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
++ "=d" (__ign3) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)), "g" ((long)(a4)) \
++ : "memory", "r10" ); \
++ (type)__res; \
++})
++
++#define _hypercall5(type, name, a1, a2, a3, a4, a5) \
++({ \
++ long __res, __ign1, __ign2, __ign3; \
++ asm volatile ( \
++ "movq %7,%%r10; movq %8,%%r8; " \
++ "call hypercall_page + ("STR(__HYPERVISOR_##name)" * 32)"\
++ : "=a" (__res), "=D" (__ign1), "=S" (__ign2), \
++ "=d" (__ign3) \
++ : "1" ((long)(a1)), "2" ((long)(a2)), \
++ "3" ((long)(a3)), "g" ((long)(a4)), \
++ "g" ((long)(a5)) \
++ : "memory", "r10", "r8" ); \
++ (type)__res; \
++})
++
++static inline int
++HYPERVISOR_set_trap_table(
++ trap_info_t *table)
++{
++ return _hypercall1(int, set_trap_table, table);
++}
++
++static inline int
++HYPERVISOR_mmu_update(
++ mmu_update_t *req, int count, int *success_count, domid_t domid)
++{
++ return _hypercall4(int, mmu_update, req, count, success_count, domid);
++}
++
++static inline int
++HYPERVISOR_mmuext_op(
++ struct mmuext_op *op, int count, int *success_count, domid_t domid)
++{
++ return _hypercall4(int, mmuext_op, op, count, success_count, domid);
++}
++
++static inline int
++HYPERVISOR_set_gdt(
++ unsigned long *frame_list, int entries)
++{
++ return _hypercall2(int, set_gdt, frame_list, entries);
++}
++
++static inline int
++HYPERVISOR_stack_switch(
++ unsigned long ss, unsigned long esp)
++{
++ return _hypercall2(int, stack_switch, ss, esp);
++}
++
++static inline int
++HYPERVISOR_set_callbacks(
++ unsigned long event_address, unsigned long failsafe_address,
++ unsigned long syscall_address)
++{
++ return _hypercall3(int, set_callbacks,
++ event_address, failsafe_address, syscall_address);
++}
++
++static inline int
++HYPERVISOR_fpu_taskswitch(
++ int set)
++{
++ return _hypercall1(int, fpu_taskswitch, set);
++}
++
++static inline int
++HYPERVISOR_sched_op_compat(
++ int cmd, unsigned long arg)
++{
++ return _hypercall2(int, sched_op_compat, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_sched_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, sched_op, cmd, arg);
++}
++
++static inline long
++HYPERVISOR_set_timer_op(
++ u64 timeout)
++{
++ return _hypercall1(long, set_timer_op, timeout);
++}
++
++static inline int
++HYPERVISOR_dom0_op(
++ dom0_op_t *dom0_op)
++{
++ dom0_op->interface_version = DOM0_INTERFACE_VERSION;
++ return _hypercall1(int, dom0_op, dom0_op);
++}
++
++static inline int
++HYPERVISOR_set_debugreg(
++ int reg, unsigned long value)
++{
++ return _hypercall2(int, set_debugreg, reg, value);
++}
++
++static inline unsigned long
++HYPERVISOR_get_debugreg(
++ int reg)
++{
++ return _hypercall1(unsigned long, get_debugreg, reg);
++}
++
++static inline int
++HYPERVISOR_update_descriptor(
++ unsigned long ma, unsigned long word)
++{
++ return _hypercall2(int, update_descriptor, ma, word);
++}
++
++static inline int
++HYPERVISOR_memory_op(
++ unsigned int cmd, void *arg)
++{
++ return _hypercall2(int, memory_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_multicall(
++ void *call_list, int nr_calls)
++{
++ return _hypercall2(int, multicall, call_list, nr_calls);
++}
++
++static inline int
++HYPERVISOR_update_va_mapping(
++ unsigned long va, pte_t new_val, unsigned long flags)
++{
++ return _hypercall3(int, update_va_mapping, va, new_val.pte, flags);
++}
++
++static inline int
++HYPERVISOR_event_channel_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, event_channel_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct evtchn_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, event_channel_op_compat, &op);
++ memcpy(arg, &op.u, sizeof(op.u));
++ }
++ return rc;
++}
++
++static inline int
++HYPERVISOR_acm_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, acm_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_xen_version(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, xen_version, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_console_io(
++ int cmd, int count, char *str)
++{
++ return _hypercall3(int, console_io, cmd, count, str);
++}
++
++static inline int
++HYPERVISOR_physdev_op(
++ int cmd, void *arg)
++{
++ int rc = _hypercall2(int, physdev_op, cmd, arg);
++ if (unlikely(rc == -ENOSYS)) {
++ struct physdev_op op;
++ op.cmd = cmd;
++ memcpy(&op.u, arg, sizeof(op.u));
++ rc = _hypercall1(int, physdev_op_compat, &op);
++ memcpy(arg, &op.u, sizeof(op.u));
++ }
++ return rc;
++}
++
++static inline int
++HYPERVISOR_grant_table_op(
++ unsigned int cmd, void *uop, unsigned int count)
++{
++ return _hypercall3(int, grant_table_op, cmd, uop, count);
++}
++
++static inline int
++HYPERVISOR_update_va_mapping_otherdomain(
++ unsigned long va, pte_t new_val, unsigned long flags, domid_t domid)
++{
++ return _hypercall4(int, update_va_mapping_otherdomain, va,
++ new_val.pte, flags, domid);
++}
++
++static inline int
++HYPERVISOR_vm_assist(
++ unsigned int cmd, unsigned int type)
++{
++ return _hypercall2(int, vm_assist, cmd, type);
++}
++
++static inline int
++HYPERVISOR_vcpu_op(
++ int cmd, int vcpuid, void *extra_args)
++{
++ return _hypercall3(int, vcpu_op, cmd, vcpuid, extra_args);
++}
++
++static inline int
++HYPERVISOR_set_segment_base(
++ int reg, unsigned long value)
++{
++ return _hypercall2(int, set_segment_base, reg, value);
++}
++
++static inline int
++HYPERVISOR_suspend(
++ unsigned long srec)
++{
++ struct sched_shutdown sched_shutdown = {
++ .reason = SHUTDOWN_suspend
++ };
++
++ int rc = _hypercall3(int, sched_op, SCHEDOP_shutdown,
++ &sched_shutdown, srec);
++
++ if (rc == -ENOSYS)
++ rc = _hypercall3(int, sched_op_compat, SCHEDOP_shutdown,
++ SHUTDOWN_suspend, srec);
++
++ return rc;
++}
++
++static inline int
++HYPERVISOR_nmi_op(
++ unsigned long op, void *arg)
++{
++ return _hypercall2(int, nmi_op, op, arg);
++}
++
++static inline int
++HYPERVISOR_callback_op(
++ int cmd, void *arg)
++{
++ return _hypercall2(int, callback_op, cmd, arg);
++}
++
++static inline int
++HYPERVISOR_xenoprof_op(
++ int op, void *arg)
++{
++ return _hypercall2(int, xenoprof_op, op, arg);
++}
++
++#endif /* __HYPERCALL_H__ */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/hypervisor.h new/include/asm-x86_64/mach-xen/asm/hypervisor.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/hypervisor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/hypervisor.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,2 @@
++
++#include <asm-i386/mach-xen/asm/hypervisor.h>
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/io.h new/include/asm-x86_64/mach-xen/asm/io.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/io.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/io.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,328 @@
++#ifndef _ASM_IO_H
++#define _ASM_IO_H
++
++#include <linux/config.h>
++#include <asm/fixmap.h>
++
++/*
++ * This file contains the definitions for the x86 IO instructions
++ * inb/inw/inl/outb/outw/outl and the "string versions" of the same
++ * (insb/insw/insl/outsb/outsw/outsl). You can also use "pausing"
++ * versions of the single-IO instructions (inb_p/inw_p/..).
++ *
++ * This file is not meant to be obfuscating: it's just complicated
++ * to (a) handle it all in a way that makes gcc able to optimize it
++ * as well as possible and (b) trying to avoid writing the same thing
++ * over and over again with slight variations and possibly making a
++ * mistake somewhere.
++ */
++
++/*
++ * Thanks to James van Artsdalen for a better timing-fix than
++ * the two short jumps: using outb's to a nonexistent port seems
++ * to guarantee better timings even on fast machines.
++ *
++ * On the other hand, I'd like to be sure of a non-existent port:
++ * I feel a bit unsafe about using 0x80 (should be safe, though)
++ *
++ * Linus
++ */
++
++ /*
++ * Bit simplified and optimized by Jan Hubicka
++ * Support of BIGMEM added by Gerhard Wichert, Siemens AG, July 1999.
++ *
++ * isa_memset_io, isa_memcpy_fromio, isa_memcpy_toio added,
++ * isa_read[wl] and isa_write[wl] fixed
++ * - Arnaldo Carvalho de Melo <acme@conectiva.com.br>
++ */
++
++#define __SLOW_DOWN_IO "\noutb %%al,$0x80"
++
++#ifdef REALLY_SLOW_IO
++#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO __SLOW_DOWN_IO
++#else
++#define __FULL_SLOW_DOWN_IO __SLOW_DOWN_IO
++#endif
++
++/*
++ * Talk about misusing macros..
++ */
++#define __OUT1(s,x) \
++static inline void out##s(unsigned x value, unsigned short port) {
++
++#define __OUT2(s,s1,s2) \
++__asm__ __volatile__ ("out" #s " %" s1 "0,%" s2 "1"
++
++#define __OUT(s,s1,x) \
++__OUT1(s,x) __OUT2(s,s1,"w") : : "a" (value), "Nd" (port)); } \
++__OUT1(s##_p,x) __OUT2(s,s1,"w") __FULL_SLOW_DOWN_IO : : "a" (value), "Nd" (port));} \
++
++#define __IN1(s) \
++static inline RETURN_TYPE in##s(unsigned short port) { RETURN_TYPE _v;
++
++#define __IN2(s,s1,s2) \
++__asm__ __volatile__ ("in" #s " %" s2 "1,%" s1 "0"
++
++#define __IN(s,s1,i...) \
++__IN1(s) __IN2(s,s1,"w") : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \
++__IN1(s##_p) __IN2(s,s1,"w") __FULL_SLOW_DOWN_IO : "=a" (_v) : "Nd" (port) ,##i ); return _v; } \
++
++#define __INS(s) \
++static inline void ins##s(unsigned short port, void * addr, unsigned long count) \
++{ __asm__ __volatile__ ("rep ; ins" #s \
++: "=D" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); }
++
++#define __OUTS(s) \
++static inline void outs##s(unsigned short port, const void * addr, unsigned long count) \
++{ __asm__ __volatile__ ("rep ; outs" #s \
++: "=S" (addr), "=c" (count) : "d" (port),"0" (addr),"1" (count)); }
++
++#define RETURN_TYPE unsigned char
++__IN(b,"")
++#undef RETURN_TYPE
++#define RETURN_TYPE unsigned short
++__IN(w,"")
++#undef RETURN_TYPE
++#define RETURN_TYPE unsigned int
++__IN(l,"")
++#undef RETURN_TYPE
++
++__OUT(b,"b",char)
++__OUT(w,"w",short)
++__OUT(l,,int)
++
++__INS(b)
++__INS(w)
++__INS(l)
++
++__OUTS(b)
++__OUTS(w)
++__OUTS(l)
++
++#define IO_SPACE_LIMIT 0xffff
++
++#if defined(__KERNEL__) && __x86_64__
++
++#include <linux/vmalloc.h>
++
++#ifndef __i386__
++/*
++ * Change virtual addresses to physical addresses and vv.
++ * These are pretty trivial
++ */
++static inline unsigned long virt_to_phys(volatile void * address)
++{
++ return __pa(address);
++}
++
++static inline void * phys_to_virt(unsigned long address)
++{
++ return __va(address);
++}
++
++#define virt_to_bus(_x) phys_to_machine(__pa(_x))
++#define bus_to_virt(_x) __va(machine_to_phys(_x))
++#endif
++
++/*
++ * Change "struct page" to physical address.
++ */
++#define page_to_pseudophys(page) ((dma_addr_t)page_to_pfn(page) << PAGE_SHIFT)
++#define page_to_phys(page) (phys_to_machine(page_to_pseudophys(page)))
++#define page_to_bus(page) (phys_to_machine(page_to_pseudophys(page)))
++
++#define bio_to_pseudophys(bio) (page_to_pseudophys(bio_page((bio))) + \
++ (unsigned long) bio_offset((bio)))
++#define bvec_to_pseudophys(bv) (page_to_pseudophys((bv)->bv_page) + \
++ (unsigned long) (bv)->bv_offset)
++
++#define BIOVEC_PHYS_MERGEABLE(vec1, vec2) \
++ (((bvec_to_phys((vec1)) + (vec1)->bv_len) == bvec_to_phys((vec2))) && \
++ ((bvec_to_pseudophys((vec1)) + (vec1)->bv_len) == \
++ bvec_to_pseudophys((vec2))))
++
++#include <asm-generic/iomap.h>
++
++extern void __iomem *__ioremap(unsigned long offset, unsigned long size, unsigned long flags);
++
++static inline void __iomem * ioremap (unsigned long offset, unsigned long size)
++{
++ return __ioremap(offset, size, 0);
++}
++
++extern void *early_ioremap(unsigned long addr, unsigned long size);
++extern void early_iounmap(void *addr, unsigned long size);
++
++/*
++ * This one maps high address device memory and turns off caching for that area.
++ * it's useful if some control registers are in such an area and write combining
++ * or read caching is not desirable:
++ */
++extern void __iomem * ioremap_nocache (unsigned long offset, unsigned long size);
++extern void iounmap(volatile void __iomem *addr);
++
++/*
++ * ISA I/O bus memory addresses are 1:1 with the physical address.
++ */
++
++#define isa_virt_to_bus(_x) isa_virt_to_bus_is_UNSUPPORTED->x
++#define isa_page_to_bus(_x) isa_page_to_bus_is_UNSUPPORTED->x
++#define isa_bus_to_virt(_x) (void *)(__fix_to_virt(FIX_ISAMAP_BEGIN) + (_x))
++
++/*
++ * However PCI ones are not necessarily 1:1 and therefore these interfaces
++ * are forbidden in portable PCI drivers.
++ *
++ * Allow them on x86 for legacy drivers, though.
++ */
++#define virt_to_bus(_x) phys_to_machine(__pa(_x))
++#define bus_to_virt(_x) __va(machine_to_phys(_x))
++
++/*
++ * readX/writeX() are used to access memory mapped devices. On some
++ * architectures the memory mapped IO stuff needs to be accessed
++ * differently. On the x86 architecture, we just read/write the
++ * memory location directly.
++ */
++
++static inline __u8 __readb(const volatile void __iomem *addr)
++{
++ return *(__force volatile __u8 *)addr;
++}
++static inline __u16 __readw(const volatile void __iomem *addr)
++{
++ return *(__force volatile __u16 *)addr;
++}
++static __always_inline __u32 __readl(const volatile void __iomem *addr)
++{
++ return *(__force volatile __u32 *)addr;
++}
++static inline __u64 __readq(const volatile void __iomem *addr)
++{
++ return *(__force volatile __u64 *)addr;
++}
++#define readb(x) __readb(x)
++#define readw(x) __readw(x)
++#define readl(x) __readl(x)
++#define readq(x) __readq(x)
++#define readb_relaxed(a) readb(a)
++#define readw_relaxed(a) readw(a)
++#define readl_relaxed(a) readl(a)
++#define readq_relaxed(a) readq(a)
++#define __raw_readb readb
++#define __raw_readw readw
++#define __raw_readl readl
++#define __raw_readq readq
++
++#define mmiowb()
++
++static inline void __writel(__u32 b, volatile void __iomem *addr)
++{
++ *(__force volatile __u32 *)addr = b;
++}
++static inline void __writeq(__u64 b, volatile void __iomem *addr)
++{
++ *(__force volatile __u64 *)addr = b;
++}
++static inline void __writeb(__u8 b, volatile void __iomem *addr)
++{
++ *(__force volatile __u8 *)addr = b;
++}
++static inline void __writew(__u16 b, volatile void __iomem *addr)
++{
++ *(__force volatile __u16 *)addr = b;
++}
++#define writeq(val,addr) __writeq((val),(addr))
++#define writel(val,addr) __writel((val),(addr))
++#define writew(val,addr) __writew((val),(addr))
++#define writeb(val,addr) __writeb((val),(addr))
++#define __raw_writeb writeb
++#define __raw_writew writew
++#define __raw_writel writel
++#define __raw_writeq writeq
++
++void __memcpy_fromio(void*,unsigned long,unsigned);
++void __memcpy_toio(unsigned long,const void*,unsigned);
++
++static inline void memcpy_fromio(void *to, const volatile void __iomem *from, unsigned len)
++{
++ __memcpy_fromio(to,(unsigned long)from,len);
++}
++static inline void memcpy_toio(volatile void __iomem *to, const void *from, unsigned len)
++{
++ __memcpy_toio((unsigned long)to,from,len);
++}
++
++void memset_io(volatile void __iomem *a, int b, size_t c);
++
++/*
++ * ISA space is 'always mapped' on a typical x86 system, no need to
++ * explicitly ioremap() it. The fact that the ISA IO space is mapped
++ * to PAGE_OFFSET is pure coincidence - it does not mean ISA values
++ * are physical addresses. The following constant pointer can be
++ * used as the IO-area pointer (it can be iounmapped as well, so the
++ * analogy with PCI is quite large):
++ */
++#define __ISA_IO_base ((char __iomem *)(fix_to_virt(FIX_ISAMAP_BEGIN)))
++
++/*
++ * Again, x86-64 does not require mem IO specific function.
++ */
++
++#define eth_io_copy_and_sum(a,b,c,d) eth_copy_and_sum((a),(void *)(b),(c),(d))
++
++/**
++ * check_signature - find BIOS signatures
++ * @io_addr: mmio address to check
++ * @signature: signature block
++ * @length: length of signature
++ *
++ * Perform a signature comparison with the mmio address io_addr. This
++ * address should have been obtained by ioremap.
++ * Returns 1 on a match.
++ */
++
++static inline int check_signature(void __iomem *io_addr,
++ const unsigned char *signature, int length)
++{
++ int retval = 0;
++ do {
++ if (readb(io_addr) != *signature)
++ goto out;
++ io_addr++;
++ signature++;
++ length--;
++ } while (length);
++ retval = 1;
++out:
++ return retval;
++}
++
++/* Nothing to do */
++
++#define dma_cache_inv(_start,_size) do { } while (0)
++#define dma_cache_wback(_start,_size) do { } while (0)
++#define dma_cache_wback_inv(_start,_size) do { } while (0)
++
++#define flush_write_buffers()
++
++extern int iommu_bio_merge;
++#define BIO_VMERGE_BOUNDARY iommu_bio_merge
++
++/*
++ * Convert a physical pointer to a virtual kernel pointer for /dev/mem
++ * access
++ */
++#define xlate_dev_mem_ptr(p) __va(p)
++
++/*
++ * Convert a virtual cached pointer to an uncached pointer
++ */
++#define xlate_dev_kmem_ptr(p) p
++
++#endif /* __KERNEL__ */
++
++#define ARCH_HAS_DEV_MEM
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/irq.h new/include/asm-x86_64/mach-xen/asm/irq.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/irq.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/irq.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,39 @@
++#ifndef _ASM_IRQ_H
++#define _ASM_IRQ_H
++
++/*
++ * linux/include/asm/irq.h
++ *
++ * (C) 1992, 1993 Linus Torvalds, (C) 1997 Ingo Molnar
++ *
++ * IRQ/IPI changes taken from work by Thomas Radke
++ * <tomsoft@informatik.tu-chemnitz.de>
++ */
++
++#include <linux/config.h>
++#include <linux/sched.h>
++/* include comes from machine specific directory */
++#include "irq_vectors.h"
++#include <asm/thread_info.h>
++
++static __inline__ int irq_canonicalize(int irq)
++{
++ return ((irq == 2) ? 9 : irq);
++}
++
++#ifdef CONFIG_X86_LOCAL_APIC
++#define ARCH_HAS_NMI_WATCHDOG /* See include/linux/nmi.h */
++#endif
++
++#define KDB_VECTOR 0xf9
++
++# define irq_ctx_init(cpu) do { } while (0)
++
++#ifdef CONFIG_HOTPLUG_CPU
++#include <linux/cpumask.h>
++extern void fixup_irqs(cpumask_t map);
++#endif
++
++#define __ARCH_HAS_DO_SOFTIRQ 1
++
++#endif /* _ASM_IRQ_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/mmu_context.h new/include/asm-x86_64/mach-xen/asm/mmu_context.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/mmu_context.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/mmu_context.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,136 @@
++#ifndef __X86_64_MMU_CONTEXT_H
++#define __X86_64_MMU_CONTEXT_H
++
++#include <linux/config.h>
++#include <asm/desc.h>
++#include <asm/atomic.h>
++#include <asm/pgalloc.h>
++#include <asm/page.h>
++#include <asm/pda.h>
++#include <asm/pgtable.h>
++#include <asm/tlbflush.h>
++
++/*
++ * possibly do the LDT unload here?
++ */
++int init_new_context(struct task_struct *tsk, struct mm_struct *mm);
++void destroy_context(struct mm_struct *mm);
++
++static inline void enter_lazy_tlb(struct mm_struct *mm, struct task_struct *tsk)
++{
++#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
++ if (read_pda(mmu_state) == TLBSTATE_OK)
++ write_pda(mmu_state, TLBSTATE_LAZY);
++#endif
++}
++
++#define prepare_arch_switch(next) __prepare_arch_switch()
++
++static inline void __prepare_arch_switch(void)
++{
++ /*
++ * Save away %es, %ds, %fs and %gs. Must happen before reload
++ * of cr3/ldt (i.e., not in __switch_to).
++ */
++ __asm__ __volatile__ (
++ "mov %%es,%0 ; mov %%ds,%1 ; mov %%fs,%2 ; mov %%gs,%3"
++ : "=m" (current->thread.es),
++ "=m" (current->thread.ds),
++ "=m" (current->thread.fsindex),
++ "=m" (current->thread.gsindex) );
++
++ if (current->thread.ds)
++ __asm__ __volatile__ ( "movl %0,%%ds" : : "r" (0) );
++
++ if (current->thread.es)
++ __asm__ __volatile__ ( "movl %0,%%es" : : "r" (0) );
++
++ if (current->thread.fsindex) {
++ __asm__ __volatile__ ( "movl %0,%%fs" : : "r" (0) );
++ current->thread.fs = 0;
++ }
++
++ if (current->thread.gsindex) {
++ load_gs_index(0);
++ current->thread.gs = 0;
++ }
++}
++
++extern void mm_pin(struct mm_struct *mm);
++extern void mm_unpin(struct mm_struct *mm);
++void mm_pin_all(void);
++
++static inline void load_cr3(pgd_t *pgd)
++{
++ asm volatile("movq %0,%%cr3" :: "r" (phys_to_machine(__pa(pgd))) :
++ "memory");
++}
++
++static inline void switch_mm(struct mm_struct *prev, struct mm_struct *next,
++ struct task_struct *tsk)
++{
++ unsigned cpu = smp_processor_id();
++ struct mmuext_op _op[3], *op = _op;
++
++ if (likely(prev != next)) {
++ BUG_ON(!next->context.pinned);
++
++ /* stop flush ipis for the previous mm */
++ cpu_clear(cpu, prev->cpu_vm_mask);
++#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
++ write_pda(mmu_state, TLBSTATE_OK);
++ write_pda(active_mm, next);
++#endif
++ cpu_set(cpu, next->cpu_vm_mask);
++
++ /* load_cr3(next->pgd) */
++ op->cmd = MMUEXT_NEW_BASEPTR;
++ op->arg1.mfn = pfn_to_mfn(__pa(next->pgd) >> PAGE_SHIFT);
++ op++;
++
++ /* xen_new_user_pt(__pa(__user_pgd(next->pgd))) */
++ op->cmd = MMUEXT_NEW_USER_BASEPTR;
++ op->arg1.mfn = pfn_to_mfn(__pa(__user_pgd(next->pgd)) >> PAGE_SHIFT);
++ op++;
++
++ if (unlikely(next->context.ldt != prev->context.ldt)) {
++ /* load_LDT_nolock(&next->context, cpu) */
++ op->cmd = MMUEXT_SET_LDT;
++ op->arg1.linear_addr = (unsigned long)next->context.ldt;
++ op->arg2.nr_ents = next->context.size;
++ op++;
++ }
++
++ BUG_ON(HYPERVISOR_mmuext_op(_op, op-_op, NULL, DOMID_SELF));
++ }
++#if defined(CONFIG_SMP) && !defined(CONFIG_XEN)
++ else {
++ write_pda(mmu_state, TLBSTATE_OK);
++ if (read_pda(active_mm) != next)
++ out_of_line_bug();
++ if (!cpu_test_and_set(cpu, next->cpu_vm_mask)) {
++ /* We were in lazy tlb mode and leave_mm disabled
++ * tlb flush IPI delivery. We must reload CR3
++ * to make sure to use no freed page tables.
++ */
++ load_cr3(next->pgd);
++ xen_new_user_pt(__pa(__user_pgd(next->pgd)));
++ load_LDT_nolock(&next->context, cpu);
++ }
++ }
++#endif
++}
++
++#define deactivate_mm(tsk,mm) do { \
++ load_gs_index(0); \
++ asm volatile("movl %0,%%fs"::"r"(0)); \
++} while(0)
++
++static inline void activate_mm(struct mm_struct *prev, struct mm_struct *next)
++{
++ if (!next->context.pinned)
++ mm_pin(next);
++ switch_mm(prev, next, NULL);
++}
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/mmu.h new/include/asm-x86_64/mach-xen/asm/mmu.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/mmu.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/mmu.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,37 @@
++#ifndef __x86_64_MMU_H
++#define __x86_64_MMU_H
++
++#include <linux/spinlock.h>
++#include <asm/semaphore.h>
++
++/*
++ * The x86_64 doesn't have a mmu context, but
++ * we put the segment information here.
++ *
++ * cpu_vm_mask is used to optimize ldt flushing.
++ */
++typedef struct {
++ void *ldt;
++ rwlock_t ldtlock;
++ int size;
++ struct semaphore sem;
++#ifdef CONFIG_XEN
++ unsigned pinned:1;
++ struct list_head unpinned;
++#endif
++} mm_context_t;
++
++#ifdef CONFIG_XEN
++extern struct list_head mm_unpinned;
++extern spinlock_t mm_unpinned_lock;
++
++/* mm/memory.c:exit_mmap hook */
++extern void _arch_exit_mmap(struct mm_struct *mm);
++#define arch_exit_mmap(_mm) _arch_exit_mmap(_mm)
++
++/* kernel/fork.c:dup_mmap hook */
++extern void _arch_dup_mmap(struct mm_struct *mm);
++#define arch_dup_mmap(mm, oldmm) ((void)(oldmm), _arch_dup_mmap(mm))
++#endif
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/msr.h new/include/asm-x86_64/mach-xen/asm/msr.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/msr.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/msr.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,399 @@
++#ifndef X86_64_MSR_H
++#define X86_64_MSR_H 1
++
++#ifndef __ASSEMBLY__
++/*
++ * Access to machine-specific registers (available on 586 and better only)
++ * Note: the rd* operations modify the parameters directly (without using
++ * pointer indirection), this allows gcc to optimize better
++ */
++
++#define rdmsr(msr,val1,val2) \
++ __asm__ __volatile__("rdmsr" \
++ : "=a" (val1), "=d" (val2) \
++ : "c" (msr))
++
++
++#define rdmsrl(msr,val) do { unsigned long a__,b__; \
++ __asm__ __volatile__("rdmsr" \
++ : "=a" (a__), "=d" (b__) \
++ : "c" (msr)); \
++ val = a__ | (b__<<32); \
++} while(0)
++
++#define wrmsr(msr,val1,val2) \
++ __asm__ __volatile__("wrmsr" \
++ : /* no outputs */ \
++ : "c" (msr), "a" (val1), "d" (val2))
++
++#define wrmsrl(msr,val) wrmsr(msr,(__u32)((__u64)(val)),((__u64)(val))>>32)
++
++/* wrmsr with exception handling */
++#define wrmsr_safe(msr,a,b) ({ int ret__; \
++ asm volatile("2: wrmsr ; xorl %0,%0\n" \
++ "1:\n\t" \
++ ".section .fixup,\"ax\"\n\t" \
++ "3: movl %4,%0 ; jmp 1b\n\t" \
++ ".previous\n\t" \
++ ".section __ex_table,\"a\"\n" \
++ " .align 8\n\t" \
++ " .quad 2b,3b\n\t" \
++ ".previous" \
++ : "=a" (ret__) \
++ : "c" (msr), "0" (a), "d" (b), "i" (-EFAULT)); \
++ ret__; })
++
++#define checking_wrmsrl(msr,val) wrmsr_safe(msr,(u32)(val),(u32)((val)>>32))
++
++#define rdmsr_safe(msr,a,b) \
++ ({ int ret__; \
++ asm volatile ("1: rdmsr\n" \
++ "2:\n" \
++ ".section .fixup,\"ax\"\n" \
++ "3: movl %4,%0\n" \
++ " jmp 2b\n" \
++ ".previous\n" \
++ ".section __ex_table,\"a\"\n" \
++ " .align 8\n" \
++ " .quad 1b,3b\n" \
++ ".previous":"=&bDS" (ret__), "=a"(*(a)), "=d"(*(b))\
++ :"c"(msr), "i"(-EIO), "0"(0)); \
++ ret__; })
++
++#define rdtsc(low,high) \
++ __asm__ __volatile__("rdtsc" : "=a" (low), "=d" (high))
++
++#define rdtscl(low) \
++ __asm__ __volatile__ ("rdtsc" : "=a" (low) : : "edx")
++
++#define rdtscll(val) do { \
++ unsigned int __a,__d; \
++ asm volatile("rdtsc" : "=a" (__a), "=d" (__d)); \
++ (val) = ((unsigned long)__a) | (((unsigned long)__d)<<32); \
++} while(0)
++
++#define write_tsc(val1,val2) wrmsr(0x10, val1, val2)
++
++#define rdpmc(counter,low,high) \
++ __asm__ __volatile__("rdpmc" \
++ : "=a" (low), "=d" (high) \
++ : "c" (counter))
++
++static inline void cpuid(int op, unsigned int *eax, unsigned int *ebx,
++ unsigned int *ecx, unsigned int *edx)
++{
++ __asm__(XEN_CPUID
++ : "=a" (*eax),
++ "=b" (*ebx),
++ "=c" (*ecx),
++ "=d" (*edx)
++ : "0" (op));
++}
++
++/* Some CPUID calls want 'count' to be placed in ecx */
++static inline void cpuid_count(int op, int count, int *eax, int *ebx, int *ecx,
++ int *edx)
++{
++ __asm__(XEN_CPUID
++ : "=a" (*eax),
++ "=b" (*ebx),
++ "=c" (*ecx),
++ "=d" (*edx)
++ : "0" (op), "c" (count));
++}
++
++/*
++ * CPUID functions returning a single datum
++ */
++static inline unsigned int cpuid_eax(unsigned int op)
++{
++ unsigned int eax;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax)
++ : "0" (op)
++ : "bx", "cx", "dx");
++ return eax;
++}
++static inline unsigned int cpuid_ebx(unsigned int op)
++{
++ unsigned int eax, ebx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=b" (ebx)
++ : "0" (op)
++ : "cx", "dx" );
++ return ebx;
++}
++static inline unsigned int cpuid_ecx(unsigned int op)
++{
++ unsigned int eax, ecx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=c" (ecx)
++ : "0" (op)
++ : "bx", "dx" );
++ return ecx;
++}
++static inline unsigned int cpuid_edx(unsigned int op)
++{
++ unsigned int eax, edx;
++
++ __asm__(XEN_CPUID
++ : "=a" (eax), "=d" (edx)
++ : "0" (op)
++ : "bx", "cx");
++ return edx;
++}
++
++#define MSR_IA32_UCODE_WRITE 0x79
++#define MSR_IA32_UCODE_REV 0x8b
++
++
++#endif
++
++/* AMD/K8 specific MSRs */
++#define MSR_EFER 0xc0000080 /* extended feature register */
++#define MSR_STAR 0xc0000081 /* legacy mode SYSCALL target */
++#define MSR_LSTAR 0xc0000082 /* long mode SYSCALL target */
++#define MSR_CSTAR 0xc0000083 /* compatibility mode SYSCALL target */
++#define MSR_SYSCALL_MASK 0xc0000084 /* EFLAGS mask for syscall */
++#define MSR_FS_BASE 0xc0000100 /* 64bit GS base */
++#define MSR_GS_BASE 0xc0000101 /* 64bit FS base */
++#define MSR_KERNEL_GS_BASE 0xc0000102 /* SwapGS GS shadow (or USER_GS from kernel) */
++/* EFER bits: */
++#define _EFER_SCE 0 /* SYSCALL/SYSRET */
++#define _EFER_LME 8 /* Long mode enable */
++#define _EFER_LMA 10 /* Long mode active (read-only) */
++#define _EFER_NX 11 /* No execute enable */
++
++#define EFER_SCE (1<<_EFER_SCE)
++#define EFER_LME (1<<_EFER_LME)
++#define EFER_LMA (1<<_EFER_LMA)
++#define EFER_NX (1<<_EFER_NX)
++
++/* Intel MSRs. Some also available on other CPUs */
++#define MSR_IA32_TSC 0x10
++#define MSR_IA32_PLATFORM_ID 0x17
++
++#define MSR_IA32_PERFCTR0 0xc1
++#define MSR_IA32_PERFCTR1 0xc2
++
++#define MSR_MTRRcap 0x0fe
++#define MSR_IA32_BBL_CR_CTL 0x119
++
++#define MSR_IA32_SYSENTER_CS 0x174
++#define MSR_IA32_SYSENTER_ESP 0x175
++#define MSR_IA32_SYSENTER_EIP 0x176
++
++#define MSR_IA32_MCG_CAP 0x179
++#define MSR_IA32_MCG_STATUS 0x17a
++#define MSR_IA32_MCG_CTL 0x17b
++
++#define MSR_IA32_EVNTSEL0 0x186
++#define MSR_IA32_EVNTSEL1 0x187
++
++#define MSR_IA32_DEBUGCTLMSR 0x1d9
++#define MSR_IA32_LASTBRANCHFROMIP 0x1db
++#define MSR_IA32_LASTBRANCHTOIP 0x1dc
++#define MSR_IA32_LASTINTFROMIP 0x1dd
++#define MSR_IA32_LASTINTTOIP 0x1de
++
++#define MSR_MTRRfix64K_00000 0x250
++#define MSR_MTRRfix16K_80000 0x258
++#define MSR_MTRRfix16K_A0000 0x259
++#define MSR_MTRRfix4K_C0000 0x268
++#define MSR_MTRRfix4K_C8000 0x269
++#define MSR_MTRRfix4K_D0000 0x26a
++#define MSR_MTRRfix4K_D8000 0x26b
++#define MSR_MTRRfix4K_E0000 0x26c
++#define MSR_MTRRfix4K_E8000 0x26d
++#define MSR_MTRRfix4K_F0000 0x26e
++#define MSR_MTRRfix4K_F8000 0x26f
++#define MSR_MTRRdefType 0x2ff
++
++#define MSR_IA32_MC0_CTL 0x400
++#define MSR_IA32_MC0_STATUS 0x401
++#define MSR_IA32_MC0_ADDR 0x402
++#define MSR_IA32_MC0_MISC 0x403
++
++#define MSR_P6_PERFCTR0 0xc1
++#define MSR_P6_PERFCTR1 0xc2
++#define MSR_P6_EVNTSEL0 0x186
++#define MSR_P6_EVNTSEL1 0x187
++
++/* K7/K8 MSRs. Not complete. See the architecture manual for a more complete list. */
++#define MSR_K7_EVNTSEL0 0xC0010000
++#define MSR_K7_PERFCTR0 0xC0010004
++#define MSR_K7_EVNTSEL1 0xC0010001
++#define MSR_K7_PERFCTR1 0xC0010005
++#define MSR_K7_EVNTSEL2 0xC0010002
++#define MSR_K7_PERFCTR2 0xC0010006
++#define MSR_K7_EVNTSEL3 0xC0010003
++#define MSR_K7_PERFCTR3 0xC0010007
++#define MSR_K8_TOP_MEM1 0xC001001A
++#define MSR_K8_TOP_MEM2 0xC001001D
++#define MSR_K8_SYSCFG 0xC0010010
++#define MSR_K8_HWCR 0xC0010015
++
++/* K6 MSRs */
++#define MSR_K6_EFER 0xC0000080
++#define MSR_K6_STAR 0xC0000081
++#define MSR_K6_WHCR 0xC0000082
++#define MSR_K6_UWCCR 0xC0000085
++#define MSR_K6_PSOR 0xC0000087
++#define MSR_K6_PFIR 0xC0000088
++
++/* Centaur-Hauls/IDT defined MSRs. */
++#define MSR_IDT_FCR1 0x107
++#define MSR_IDT_FCR2 0x108
++#define MSR_IDT_FCR3 0x109
++#define MSR_IDT_FCR4 0x10a
++
++#define MSR_IDT_MCR0 0x110
++#define MSR_IDT_MCR1 0x111
++#define MSR_IDT_MCR2 0x112
++#define MSR_IDT_MCR3 0x113
++#define MSR_IDT_MCR4 0x114
++#define MSR_IDT_MCR5 0x115
++#define MSR_IDT_MCR6 0x116
++#define MSR_IDT_MCR7 0x117
++#define MSR_IDT_MCR_CTRL 0x120
++
++/* VIA Cyrix defined MSRs*/
++#define MSR_VIA_FCR 0x1107
++#define MSR_VIA_LONGHAUL 0x110a
++#define MSR_VIA_RNG 0x110b
++#define MSR_VIA_BCR2 0x1147
++
++/* Intel defined MSRs. */
++#define MSR_IA32_P5_MC_ADDR 0
++#define MSR_IA32_P5_MC_TYPE 1
++#define MSR_IA32_PLATFORM_ID 0x17
++#define MSR_IA32_EBL_CR_POWERON 0x2a
++
++#define MSR_IA32_APICBASE 0x1b
++#define MSR_IA32_APICBASE_BSP (1<<8)
++#define MSR_IA32_APICBASE_ENABLE (1<<11)
++#define MSR_IA32_APICBASE_BASE (0xfffff<<12)
++
++/* P4/Xeon+ specific */
++#define MSR_IA32_MCG_EAX 0x180
++#define MSR_IA32_MCG_EBX 0x181
++#define MSR_IA32_MCG_ECX 0x182
++#define MSR_IA32_MCG_EDX 0x183
++#define MSR_IA32_MCG_ESI 0x184
++#define MSR_IA32_MCG_EDI 0x185
++#define MSR_IA32_MCG_EBP 0x186
++#define MSR_IA32_MCG_ESP 0x187
++#define MSR_IA32_MCG_EFLAGS 0x188
++#define MSR_IA32_MCG_EIP 0x189
++#define MSR_IA32_MCG_RESERVED 0x18A
++
++#define MSR_P6_EVNTSEL0 0x186
++#define MSR_P6_EVNTSEL1 0x187
++
++#define MSR_IA32_PERF_STATUS 0x198
++#define MSR_IA32_PERF_CTL 0x199
++
++#define MSR_IA32_THERM_CONTROL 0x19a
++#define MSR_IA32_THERM_INTERRUPT 0x19b
++#define MSR_IA32_THERM_STATUS 0x19c
++#define MSR_IA32_MISC_ENABLE 0x1a0
++
++#define MSR_IA32_DEBUGCTLMSR 0x1d9
++#define MSR_IA32_LASTBRANCHFROMIP 0x1db
++#define MSR_IA32_LASTBRANCHTOIP 0x1dc
++#define MSR_IA32_LASTINTFROMIP 0x1dd
++#define MSR_IA32_LASTINTTOIP 0x1de
++
++#define MSR_IA32_MC0_CTL 0x400
++#define MSR_IA32_MC0_STATUS 0x401
++#define MSR_IA32_MC0_ADDR 0x402
++#define MSR_IA32_MC0_MISC 0x403
++
++/* Pentium IV performance counter MSRs */
++#define MSR_P4_BPU_PERFCTR0 0x300
++#define MSR_P4_BPU_PERFCTR1 0x301
++#define MSR_P4_BPU_PERFCTR2 0x302
++#define MSR_P4_BPU_PERFCTR3 0x303
++#define MSR_P4_MS_PERFCTR0 0x304
++#define MSR_P4_MS_PERFCTR1 0x305
++#define MSR_P4_MS_PERFCTR2 0x306
++#define MSR_P4_MS_PERFCTR3 0x307
++#define MSR_P4_FLAME_PERFCTR0 0x308
++#define MSR_P4_FLAME_PERFCTR1 0x309
++#define MSR_P4_FLAME_PERFCTR2 0x30a
++#define MSR_P4_FLAME_PERFCTR3 0x30b
++#define MSR_P4_IQ_PERFCTR0 0x30c
++#define MSR_P4_IQ_PERFCTR1 0x30d
++#define MSR_P4_IQ_PERFCTR2 0x30e
++#define MSR_P4_IQ_PERFCTR3 0x30f
++#define MSR_P4_IQ_PERFCTR4 0x310
++#define MSR_P4_IQ_PERFCTR5 0x311
++#define MSR_P4_BPU_CCCR0 0x360
++#define MSR_P4_BPU_CCCR1 0x361
++#define MSR_P4_BPU_CCCR2 0x362
++#define MSR_P4_BPU_CCCR3 0x363
++#define MSR_P4_MS_CCCR0 0x364
++#define MSR_P4_MS_CCCR1 0x365
++#define MSR_P4_MS_CCCR2 0x366
++#define MSR_P4_MS_CCCR3 0x367
++#define MSR_P4_FLAME_CCCR0 0x368
++#define MSR_P4_FLAME_CCCR1 0x369
++#define MSR_P4_FLAME_CCCR2 0x36a
++#define MSR_P4_FLAME_CCCR3 0x36b
++#define MSR_P4_IQ_CCCR0 0x36c
++#define MSR_P4_IQ_CCCR1 0x36d
++#define MSR_P4_IQ_CCCR2 0x36e
++#define MSR_P4_IQ_CCCR3 0x36f
++#define MSR_P4_IQ_CCCR4 0x370
++#define MSR_P4_IQ_CCCR5 0x371
++#define MSR_P4_ALF_ESCR0 0x3ca
++#define MSR_P4_ALF_ESCR1 0x3cb
++#define MSR_P4_BPU_ESCR0 0x3b2
++#define MSR_P4_BPU_ESCR1 0x3b3
++#define MSR_P4_BSU_ESCR0 0x3a0
++#define MSR_P4_BSU_ESCR1 0x3a1
++#define MSR_P4_CRU_ESCR0 0x3b8
++#define MSR_P4_CRU_ESCR1 0x3b9
++#define MSR_P4_CRU_ESCR2 0x3cc
++#define MSR_P4_CRU_ESCR3 0x3cd
++#define MSR_P4_CRU_ESCR4 0x3e0
++#define MSR_P4_CRU_ESCR5 0x3e1
++#define MSR_P4_DAC_ESCR0 0x3a8
++#define MSR_P4_DAC_ESCR1 0x3a9
++#define MSR_P4_FIRM_ESCR0 0x3a4
++#define MSR_P4_FIRM_ESCR1 0x3a5
++#define MSR_P4_FLAME_ESCR0 0x3a6
++#define MSR_P4_FLAME_ESCR1 0x3a7
++#define MSR_P4_FSB_ESCR0 0x3a2
++#define MSR_P4_FSB_ESCR1 0x3a3
++#define MSR_P4_IQ_ESCR0 0x3ba
++#define MSR_P4_IQ_ESCR1 0x3bb
++#define MSR_P4_IS_ESCR0 0x3b4
++#define MSR_P4_IS_ESCR1 0x3b5
++#define MSR_P4_ITLB_ESCR0 0x3b6
++#define MSR_P4_ITLB_ESCR1 0x3b7
++#define MSR_P4_IX_ESCR0 0x3c8
++#define MSR_P4_IX_ESCR1 0x3c9
++#define MSR_P4_MOB_ESCR0 0x3aa
++#define MSR_P4_MOB_ESCR1 0x3ab
++#define MSR_P4_MS_ESCR0 0x3c0
++#define MSR_P4_MS_ESCR1 0x3c1
++#define MSR_P4_PMH_ESCR0 0x3ac
++#define MSR_P4_PMH_ESCR1 0x3ad
++#define MSR_P4_RAT_ESCR0 0x3bc
++#define MSR_P4_RAT_ESCR1 0x3bd
++#define MSR_P4_SAAT_ESCR0 0x3ae
++#define MSR_P4_SAAT_ESCR1 0x3af
++#define MSR_P4_SSU_ESCR0 0x3be
++#define MSR_P4_SSU_ESCR1 0x3bf /* guess: not defined in manual */
++#define MSR_P4_TBPU_ESCR0 0x3c2
++#define MSR_P4_TBPU_ESCR1 0x3c3
++#define MSR_P4_TC_ESCR0 0x3c4
++#define MSR_P4_TC_ESCR1 0x3c5
++#define MSR_P4_U2L_ESCR0 0x3b0
++#define MSR_P4_U2L_ESCR1 0x3b1
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/nmi.h new/include/asm-x86_64/mach-xen/asm/nmi.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/nmi.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/nmi.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,75 @@
++/*
++ * linux/include/asm-i386/nmi.h
++ */
++#ifndef ASM_NMI_H
++#define ASM_NMI_H
++
++#include <linux/pm.h>
++
++#include <xen/interface/nmi.h>
++
++struct pt_regs;
++
++typedef int (*nmi_callback_t)(struct pt_regs * regs, int cpu);
++
++/**
++ * set_nmi_callback
++ *
++ * Set a handler for an NMI. Only one handler may be
++ * set. Return 1 if the NMI was handled.
++ */
++void set_nmi_callback(nmi_callback_t callback);
++
++/**
++ * unset_nmi_callback
++ *
++ * Remove the handler previously set.
++ */
++void unset_nmi_callback(void);
++
++#ifdef CONFIG_PM
++
++/** Replace the PM callback routine for NMI. */
++struct pm_dev * set_nmi_pm_callback(pm_callback callback);
++
++/** Unset the PM callback routine back to the default. */
++void unset_nmi_pm_callback(struct pm_dev * dev);
++
++#else
++
++static inline struct pm_dev * set_nmi_pm_callback(pm_callback callback)
++{
++ return 0;
++}
++
++static inline void unset_nmi_pm_callback(struct pm_dev * dev)
++{
++}
++
++#endif /* CONFIG_PM */
++
++extern void default_do_nmi(struct pt_regs *);
++extern void die_nmi(char *str, struct pt_regs *regs);
++
++static inline unsigned char get_nmi_reason(void)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ unsigned char reason = 0;
++
++ /* construct a value which looks like it came from
++ * port 0x61.
++ */
++ if (test_bit(_XEN_NMIREASON_io_error, &s->arch.nmi_reason))
++ reason |= 0x40;
++ if (test_bit(_XEN_NMIREASON_parity_error, &s->arch.nmi_reason))
++ reason |= 0x80;
++
++ return reason;
++}
++
++extern int panic_on_timeout;
++extern int unknown_nmi_panic;
++
++extern int check_nmi_watchdog(void);
++
++#endif /* ASM_NMI_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/page.h new/include/asm-x86_64/mach-xen/asm/page.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/page.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/page.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,327 @@
++#ifndef _X86_64_PAGE_H
++#define _X86_64_PAGE_H
++
++#include <linux/config.h>
++/* #include <linux/string.h> */
++#ifndef __ASSEMBLY__
++#include <linux/kernel.h>
++#include <linux/types.h>
++#include <asm/bug.h>
++#include <xen/features.h>
++#endif
++#include <xen/interface/xen.h>
++#include <xen/foreign_page.h>
++
++#define arch_free_page(_page,_order) \
++({ int foreign = PageForeign(_page); \
++ if (foreign) \
++ (PageForeignDestructor(_page))(_page); \
++ foreign; \
++})
++#define HAVE_ARCH_FREE_PAGE
++
++#ifdef CONFIG_XEN_SCRUB_PAGES
++#define scrub_pages(_p,_n) memset((void *)(_p), 0, (_n) << PAGE_SHIFT)
++#else
++#define scrub_pages(_p,_n) ((void)0)
++#endif
++
++/* PAGE_SHIFT determines the page size */
++#define PAGE_SHIFT 12
++#ifdef __ASSEMBLY__
++#define PAGE_SIZE (0x1 << PAGE_SHIFT)
++#else
++#define PAGE_SIZE (1UL << PAGE_SHIFT)
++#endif
++#define PAGE_MASK (~(PAGE_SIZE-1))
++#define PHYSICAL_PAGE_MASK (~(PAGE_SIZE-1) & __PHYSICAL_MASK)
++
++#define THREAD_ORDER 1
++#define THREAD_SIZE (PAGE_SIZE << THREAD_ORDER)
++#define CURRENT_MASK (~(THREAD_SIZE-1))
++
++#define EXCEPTION_STACK_ORDER 0
++#define EXCEPTION_STKSZ (PAGE_SIZE << EXCEPTION_STACK_ORDER)
++
++#define DEBUG_STACK_ORDER EXCEPTION_STACK_ORDER
++#define DEBUG_STKSZ (PAGE_SIZE << DEBUG_STACK_ORDER)
++
++#define IRQSTACK_ORDER 2
++#define IRQSTACKSIZE (PAGE_SIZE << IRQSTACK_ORDER)
++
++#define STACKFAULT_STACK 1
++#define DOUBLEFAULT_STACK 2
++#define NMI_STACK 3
++#define DEBUG_STACK 4
++#define MCE_STACK 5
++#define N_EXCEPTION_STACKS 5 /* hw limit: 7 */
++
++#define LARGE_PAGE_MASK (~(LARGE_PAGE_SIZE-1))
++#define LARGE_PAGE_SIZE (1UL << PMD_SHIFT)
++
++#define HPAGE_SHIFT PMD_SHIFT
++#define HPAGE_SIZE ((1UL) << HPAGE_SHIFT)
++#define HPAGE_MASK (~(HPAGE_SIZE - 1))
++#define HUGETLB_PAGE_ORDER (HPAGE_SHIFT - PAGE_SHIFT)
++
++#ifdef __KERNEL__
++#ifndef __ASSEMBLY__
++
++extern unsigned long end_pfn;
++
++void clear_page(void *);
++void copy_page(void *, void *);
++
++#define clear_user_page(page, vaddr, pg) clear_page(page)
++#define copy_user_page(to, from, vaddr, pg) copy_page(to, from)
++
++#define alloc_zeroed_user_highpage(vma, vaddr) alloc_page_vma(GFP_HIGHUSER | __GFP_ZERO, vma, vaddr)
++#define __HAVE_ARCH_ALLOC_ZEROED_USER_HIGHPAGE
++
++/**** MACHINE <-> PHYSICAL CONVERSION MACROS ****/
++#define INVALID_P2M_ENTRY (~0UL)
++#define FOREIGN_FRAME_BIT (1UL<<63)
++#define FOREIGN_FRAME(m) ((m) | FOREIGN_FRAME_BIT)
++
++extern unsigned long *phys_to_machine_mapping;
++
++#undef machine_to_phys_mapping
++extern unsigned long *machine_to_phys_mapping;
++extern unsigned int machine_to_phys_order;
++
++static inline unsigned long pfn_to_mfn(unsigned long pfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return pfn;
++ return phys_to_machine_mapping[(unsigned int)(pfn)] &
++ ~FOREIGN_FRAME_BIT;
++}
++
++static inline int phys_to_machine_mapping_valid(unsigned long pfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return 1;
++ return (phys_to_machine_mapping[pfn] != INVALID_P2M_ENTRY);
++}
++
++static inline unsigned long mfn_to_pfn(unsigned long mfn)
++{
++ unsigned long pfn;
++
++ if (xen_feature(XENFEAT_auto_translated_physmap))
++ return mfn;
++
++ if (unlikely((mfn >> machine_to_phys_order) != 0))
++ return end_pfn;
++
++ /* The array access can fail (e.g., device space beyond end of RAM). */
++ asm (
++ "1: movq %1,%0\n"
++ "2:\n"
++ ".section .fixup,\"ax\"\n"
++ "3: movq %2,%0\n"
++ " jmp 2b\n"
++ ".previous\n"
++ ".section __ex_table,\"a\"\n"
++ " .align 8\n"
++ " .quad 1b,3b\n"
++ ".previous"
++ : "=r" (pfn)
++ : "m" (machine_to_phys_mapping[mfn]), "m" (end_pfn) );
++
++ return pfn;
++}
++
++/*
++ * We detect special mappings in one of two ways:
++ * 1. If the MFN is an I/O page then Xen will set the m2p entry
++ * to be outside our maximum possible pseudophys range.
++ * 2. If the MFN belongs to a different domain then we will certainly
++ * not have MFN in our p2m table. Conversely, if the page is ours,
++ * then we'll have p2m(m2p(MFN))==MFN.
++ * If we detect a special mapping then it doesn't have a 'struct page'.
++ * We force !pfn_valid() by returning an out-of-range pointer.
++ *
++ * NB. These checks require that, for any MFN that is not in our reservation,
++ * there is no PFN such that p2m(PFN) == MFN. Otherwise we can get confused if
++ * we are foreign-mapping the MFN, and the other domain as m2p(MFN) == PFN.
++ * Yikes! Various places must poke in INVALID_P2M_ENTRY for safety.
++ *
++ * NB2. When deliberately mapping foreign pages into the p2m table, you *must*
++ * use FOREIGN_FRAME(). This will cause pte_pfn() to choke on it, as we
++ * require. In all the cases we care about, the FOREIGN_FRAME bit is
++ * masked (e.g., pfn_to_mfn()) so behaviour there is correct.
++ */
++static inline unsigned long mfn_to_local_pfn(unsigned long mfn)
++{
++ unsigned long pfn = mfn_to_pfn(mfn);
++ if ((pfn < end_pfn)
++ && !xen_feature(XENFEAT_auto_translated_physmap)
++ && (phys_to_machine_mapping[pfn] != mfn))
++ return end_pfn; /* force !pfn_valid() */
++ return pfn;
++}
++
++
++static inline void set_phys_to_machine(unsigned long pfn, unsigned long mfn)
++{
++ if (xen_feature(XENFEAT_auto_translated_physmap)) {
++ BUG_ON(pfn != mfn && mfn != INVALID_P2M_ENTRY);
++ return;
++ }
++ phys_to_machine_mapping[pfn] = mfn;
++}
++
++/* Definitions for machine and pseudophysical addresses. */
++typedef unsigned long paddr_t;
++typedef unsigned long maddr_t;
++
++static inline maddr_t phys_to_machine(paddr_t phys)
++{
++ maddr_t machine = pfn_to_mfn(phys >> PAGE_SHIFT);
++ machine = (machine << PAGE_SHIFT) | (phys & ~PAGE_MASK);
++ return machine;
++}
++
++static inline paddr_t machine_to_phys(maddr_t machine)
++{
++ paddr_t phys = mfn_to_pfn(machine >> PAGE_SHIFT);
++ phys = (phys << PAGE_SHIFT) | (machine & ~PAGE_MASK);
++ return phys;
++}
++
++/*
++ * These are used to make use of C type-checking..
++ */
++typedef struct { unsigned long pte; } pte_t;
++typedef struct { unsigned long pmd; } pmd_t;
++typedef struct { unsigned long pud; } pud_t;
++typedef struct { unsigned long pgd; } pgd_t;
++#define PTE_MASK PHYSICAL_PAGE_MASK
++
++typedef struct { unsigned long pgprot; } pgprot_t;
++
++#define pte_val(x) (((x).pte & 1) ? machine_to_phys((x).pte) : \
++ (x).pte)
++#define pte_val_ma(x) ((x).pte)
++
++static inline unsigned long pmd_val(pmd_t x)
++{
++ unsigned long ret = x.pmd;
++ if (ret) ret = machine_to_phys(ret);
++ return ret;
++}
++
++static inline unsigned long pud_val(pud_t x)
++{
++ unsigned long ret = x.pud;
++ if (ret) ret = machine_to_phys(ret);
++ return ret;
++}
++
++static inline unsigned long pgd_val(pgd_t x)
++{
++ unsigned long ret = x.pgd;
++ if (ret) ret = machine_to_phys(ret);
++ return ret;
++}
++
++#define pgprot_val(x) ((x).pgprot)
++
++#define __pte_ma(x) ((pte_t) { (x) } )
++
++static inline pte_t __pte(unsigned long x)
++{
++ if (x & 1) x = phys_to_machine(x);
++ return ((pte_t) { (x) });
++}
++
++static inline pmd_t __pmd(unsigned long x)
++{
++ if ((x & 1)) x = phys_to_machine(x);
++ return ((pmd_t) { (x) });
++}
++
++static inline pud_t __pud(unsigned long x)
++{
++ if ((x & 1)) x = phys_to_machine(x);
++ return ((pud_t) { (x) });
++}
++
++static inline pgd_t __pgd(unsigned long x)
++{
++ if ((x & 1)) x = phys_to_machine(x);
++ return ((pgd_t) { (x) });
++}
++
++#define __pgprot(x) ((pgprot_t) { (x) } )
++
++#define __PHYSICAL_START ((unsigned long)CONFIG_PHYSICAL_START)
++#define __START_KERNEL (__START_KERNEL_map + __PHYSICAL_START)
++#define __START_KERNEL_map 0xffffffff80000000UL
++#define __PAGE_OFFSET 0xffff880000000000UL
++
++#else
++#define __PHYSICAL_START CONFIG_PHYSICAL_START
++#define __START_KERNEL (__START_KERNEL_map + __PHYSICAL_START)
++#define __START_KERNEL_map 0xffffffff80000000
++#define __PAGE_OFFSET 0xffff880000000000
++#endif /* !__ASSEMBLY__ */
++
++#ifdef CONFIG_XEN_COMPAT_030002
++#undef LOAD_OFFSET
++#define LOAD_OFFSET 0
++#endif /* CONFIG_XEN_COMPAT_030002 */
++
++/* to align the pointer to the (next) page boundary */
++#define PAGE_ALIGN(addr) (((addr)+PAGE_SIZE-1)&PAGE_MASK)
++
++/* See Documentation/x86_64/mm.txt for a description of the memory map. */
++#define __PHYSICAL_MASK_SHIFT 46
++#define __PHYSICAL_MASK ((1UL << __PHYSICAL_MASK_SHIFT) - 1)
++#define __VIRTUAL_MASK_SHIFT 48
++#define __VIRTUAL_MASK ((1UL << __VIRTUAL_MASK_SHIFT) - 1)
++
++#define KERNEL_TEXT_SIZE (40UL*1024*1024)
++#define KERNEL_TEXT_START 0xffffffff80000000UL
++
++#define PAGE_OFFSET ((unsigned long)__PAGE_OFFSET)
++
++/* Note: __pa(&symbol_visible_to_c) should be always replaced with __pa_symbol.
++ Otherwise you risk miscompilation. */
++#define __pa(x) (((unsigned long)(x)>=__START_KERNEL_map)?(unsigned long)(x) - (unsigned long)__START_KERNEL_map:(unsigned long)(x) - PAGE_OFFSET)
++/* __pa_symbol should be used for C visible symbols.
++ This seems to be the official gcc blessed way to do such arithmetic. */
++#define __pa_symbol(x) \
++ ({unsigned long v; \
++ asm("" : "=r" (v) : "0" (x)); \
++ __pa(v); })
++
++#define __va(x) ((void *)((unsigned long)(x)+PAGE_OFFSET))
++#define __boot_va(x) __va(x)
++#define __boot_pa(x) __pa(x)
++#ifdef CONFIG_FLATMEM
++#define pfn_valid(pfn) ((pfn) < end_pfn)
++#endif
++
++#define virt_to_page(kaddr) pfn_to_page(__pa(kaddr) >> PAGE_SHIFT)
++#define virt_addr_valid(kaddr) pfn_valid(__pa(kaddr) >> PAGE_SHIFT)
++#define pfn_to_kaddr(pfn) __va((pfn) << PAGE_SHIFT)
++
++/* VIRT <-> MACHINE conversion */
++#define virt_to_machine(v) (phys_to_machine(__pa(v)))
++#define virt_to_mfn(v) (pfn_to_mfn(__pa(v) >> PAGE_SHIFT))
++#define mfn_to_virt(m) (__va(mfn_to_pfn(m) << PAGE_SHIFT))
++
++#define VM_DATA_DEFAULT_FLAGS \
++ (((current->personality & READ_IMPLIES_EXEC) ? VM_EXEC : 0 ) | \
++ VM_READ | VM_WRITE | VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)
++
++#define __HAVE_ARCH_GATE_AREA 1
++
++#endif /* __KERNEL__ */
++
++#include <asm-generic/memory_model.h>
++#include <asm-generic/page.h>
++
++#endif /* _X86_64_PAGE_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/param.h new/include/asm-x86_64/mach-xen/asm/param.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/param.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/param.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,23 @@
++#ifndef _ASMx86_64_PARAM_H
++#define _ASMx86_64_PARAM_H
++
++#ifdef __KERNEL__
++# include <linux/config.h>
++# define HZ CONFIG_HZ /* Internal kernel timer frequency */
++# define USER_HZ 100 /* .. some user interfaces are in "ticks */
++# define CLOCKS_PER_SEC (USER_HZ) /* like times() */
++#endif
++
++#ifndef HZ
++#define HZ 100
++#endif
++
++#define EXEC_PAGESIZE 4096
++
++#ifndef NOGROUP
++#define NOGROUP (-1)
++#endif
++
++#define MAXHOSTNAMELEN 64 /* max length of hostname */
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/pci.h new/include/asm-x86_64/mach-xen/asm/pci.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/pci.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/pci.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,174 @@
++#ifndef __x8664_PCI_H
++#define __x8664_PCI_H
++
++#include <linux/config.h>
++#include <asm/io.h>
++
++#ifdef __KERNEL__
++
++#include <linux/mm.h> /* for struct page */
++
++/* Can be used to override the logic in pci_scan_bus for skipping
++ already-configured bus numbers - to be used for buggy BIOSes
++ or architectures with incomplete PCI setup by the loader */
++
++#ifdef CONFIG_PCI
++extern unsigned int pcibios_assign_all_busses(void);
++#else
++#define pcibios_assign_all_busses() 0
++#endif
++#define pcibios_scan_all_fns(a, b) 0
++
++extern unsigned long pci_mem_start;
++#define PCIBIOS_MIN_IO 0x1000
++#define PCIBIOS_MIN_MEM (pci_mem_start)
++
++#define PCIBIOS_MIN_CARDBUS_IO 0x4000
++
++void pcibios_config_init(void);
++struct pci_bus * pcibios_scan_root(int bus);
++extern int (*pci_config_read)(int seg, int bus, int dev, int fn, int reg, int len, u32 *value);
++extern int (*pci_config_write)(int seg, int bus, int dev, int fn, int reg, int len, u32 value);
++
++void pcibios_set_master(struct pci_dev *dev);
++void pcibios_penalize_isa_irq(int irq, int active);
++struct irq_routing_table *pcibios_get_irq_routing_table(void);
++int pcibios_set_irq_routing(struct pci_dev *dev, int pin, int irq);
++
++#include <linux/types.h>
++#include <linux/slab.h>
++#include <asm/scatterlist.h>
++#include <linux/string.h>
++#include <asm/page.h>
++#include <linux/dma-mapping.h> /* for have_iommu */
++
++extern int iommu_setup(char *opt);
++
++/* The PCI address space does equal the physical memory
++ * address space. The networking and block device layers use
++ * this boolean for bounce buffer decisions
++ *
++ * On AMD64 it mostly equals, but we set it to zero if a hardware
++ * IOMMU (gart) of sotware IOMMU (swiotlb) is available.
++ */
++#define PCI_DMA_BUS_IS_PHYS (dma_ops->is_phys)
++
++#ifdef CONFIG_GART_IOMMU
++
++/*
++ * x86-64 always supports DAC, but sometimes it is useful to force
++ * devices through the IOMMU to get automatic sg list merging.
++ * Optional right now.
++ */
++extern int iommu_sac_force;
++#define pci_dac_dma_supported(pci_dev, mask) (!iommu_sac_force)
++
++#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \
++ dma_addr_t ADDR_NAME;
++#define DECLARE_PCI_UNMAP_LEN(LEN_NAME) \
++ __u32 LEN_NAME;
++#define pci_unmap_addr(PTR, ADDR_NAME) \
++ ((PTR)->ADDR_NAME)
++#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) \
++ (((PTR)->ADDR_NAME) = (VAL))
++#define pci_unmap_len(PTR, LEN_NAME) \
++ ((PTR)->LEN_NAME)
++#define pci_unmap_len_set(PTR, LEN_NAME, VAL) \
++ (((PTR)->LEN_NAME) = (VAL))
++
++#elif defined(CONFIG_SWIOTLB)
++
++#define pci_dac_dma_supported(pci_dev, mask) 1
++
++#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \
++ dma_addr_t ADDR_NAME;
++#define DECLARE_PCI_UNMAP_LEN(LEN_NAME) \
++ __u32 LEN_NAME;
++#define pci_unmap_addr(PTR, ADDR_NAME) \
++ ((PTR)->ADDR_NAME)
++#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) \
++ (((PTR)->ADDR_NAME) = (VAL))
++#define pci_unmap_len(PTR, LEN_NAME) \
++ ((PTR)->LEN_NAME)
++#define pci_unmap_len_set(PTR, LEN_NAME, VAL) \
++ (((PTR)->LEN_NAME) = (VAL))
++
++#else
++/* No IOMMU */
++
++#define pci_dac_dma_supported(pci_dev, mask) 1
++
++#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
++#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
++#define pci_unmap_addr(PTR, ADDR_NAME) (0)
++#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL) do { } while (0)
++#define pci_unmap_len(PTR, LEN_NAME) (0)
++#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
++
++#endif
++
++#include <asm-generic/pci-dma-compat.h>
++
++static inline dma64_addr_t
++pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
++{
++ return ((dma64_addr_t) page_to_phys(page) +
++ (dma64_addr_t) offset);
++}
++
++static inline struct page *
++pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
++{
++ return virt_to_page(__va(dma_addr));
++}
++
++static inline unsigned long
++pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
++{
++ return (dma_addr & ~PAGE_MASK);
++}
++
++static inline void
++pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
++{
++}
++
++static inline void
++pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
++{
++ flush_write_buffers();
++}
++
++#ifdef CONFIG_PCI
++static inline void pci_dma_burst_advice(struct pci_dev *pdev,
++ enum pci_dma_burst_strategy *strat,
++ unsigned long *strategy_parameter)
++{
++ *strat = PCI_DMA_BURST_INFINITY;
++ *strategy_parameter = ~0UL;
++}
++#endif
++
++#define HAVE_PCI_MMAP
++extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
++ enum pci_mmap_state mmap_state, int write_combine);
++
++static inline void pcibios_add_platform_entries(struct pci_dev *dev)
++{
++}
++
++#endif /* __KERNEL__ */
++
++/* generic pci stuff */
++#ifdef CONFIG_PCI
++#include <asm-generic/pci.h>
++#endif
++
++/* On Xen we have to scan all functions since Xen hides bridges from
++ * us. If a bridge is at fn=0 and that slot has a multifunction
++ * device, we won't find the additional devices without scanning all
++ * functions. */
++#undef pcibios_scan_all_fns
++#define pcibios_scan_all_fns(a, b) 1
++
++#endif /* __x8664_PCI_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/pgalloc.h new/include/asm-x86_64/mach-xen/asm/pgalloc.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/pgalloc.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/pgalloc.h 2006-05-09 12:35:41.000000000 +0200
+@@ -0,0 +1,226 @@
++#ifndef _X86_64_PGALLOC_H
++#define _X86_64_PGALLOC_H
++
++#include <asm/fixmap.h>
++#include <asm/pda.h>
++#include <linux/threads.h>
++#include <linux/mm.h>
++#include <asm/io.h> /* for phys_to_virt and page_to_pseudophys */
++
++#include <xen/features.h>
++void make_page_readonly(void *va, unsigned int feature);
++void make_page_writable(void *va, unsigned int feature);
++void make_pages_readonly(void *va, unsigned int nr, unsigned int feature);
++void make_pages_writable(void *va, unsigned int nr, unsigned int feature);
++
++#define __user_pgd(pgd) ((pgd) + PTRS_PER_PGD)
++
++static inline void pmd_populate_kernel(struct mm_struct *mm, pmd_t *pmd, pte_t *pte)
++{
++ set_pmd(pmd, __pmd(_PAGE_TABLE | __pa(pte)));
++}
++
++static inline void pmd_populate(struct mm_struct *mm, pmd_t *pmd, struct page *pte)
++{
++ if (unlikely((mm)->context.pinned)) {
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__va(page_to_pfn(pte) << PAGE_SHIFT),
++ pfn_pte(page_to_pfn(pte), PAGE_KERNEL_RO), 0));
++ set_pmd(pmd, __pmd(_PAGE_TABLE | (page_to_pfn(pte) << PAGE_SHIFT)));
++ } else {
++ *(pmd) = __pmd(_PAGE_TABLE | (page_to_pfn(pte) << PAGE_SHIFT));
++ }
++}
++
++static inline void pud_populate(struct mm_struct *mm, pud_t *pud, pmd_t *pmd)
++{
++ if (unlikely((mm)->context.pinned)) {
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pmd,
++ pfn_pte(virt_to_phys(pmd)>>PAGE_SHIFT,
++ PAGE_KERNEL_RO), 0));
++ set_pud(pud, __pud(_PAGE_TABLE | __pa(pmd)));
++ } else {
++ *(pud) = __pud(_PAGE_TABLE | __pa(pmd));
++ }
++}
++
++/*
++ * We need to use the batch mode here, but pgd_pupulate() won't be
++ * be called frequently.
++ */
++static inline void pgd_populate(struct mm_struct *mm, pgd_t *pgd, pud_t *pud)
++{
++ if (unlikely((mm)->context.pinned)) {
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pud,
++ pfn_pte(virt_to_phys(pud)>>PAGE_SHIFT,
++ PAGE_KERNEL_RO), 0));
++ set_pgd(pgd, __pgd(_PAGE_TABLE | __pa(pud)));
++ set_pgd(__user_pgd(pgd), __pgd(_PAGE_TABLE | __pa(pud)));
++ } else {
++ *(pgd) = __pgd(_PAGE_TABLE | __pa(pud));
++ *(__user_pgd(pgd)) = *(pgd);
++ }
++}
++
++static inline void pmd_free(pmd_t *pmd)
++{
++ pte_t *ptep = virt_to_ptep(pmd);
++
++ if (!pte_write(*ptep)) {
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pmd,
++ pfn_pte(virt_to_phys(pmd)>>PAGE_SHIFT, PAGE_KERNEL),
++ 0));
++ }
++ free_page((unsigned long)pmd);
++}
++
++static inline pmd_t *pmd_alloc_one(struct mm_struct *mm, unsigned long addr)
++{
++ pmd_t *pmd = (pmd_t *) get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
++ return pmd;
++}
++
++static inline pud_t *pud_alloc_one(struct mm_struct *mm, unsigned long addr)
++{
++ pud_t *pud = (pud_t *) get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
++ return pud;
++}
++
++static inline void pud_free(pud_t *pud)
++{
++ pte_t *ptep = virt_to_ptep(pud);
++
++ if (!pte_write(*ptep)) {
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pud,
++ pfn_pte(virt_to_phys(pud)>>PAGE_SHIFT, PAGE_KERNEL),
++ 0));
++ }
++ free_page((unsigned long)pud);
++}
++
++static inline void pgd_list_add(pgd_t *pgd)
++{
++ struct page *page = virt_to_page(pgd);
++
++ spin_lock(&pgd_lock);
++ page->index = (pgoff_t)pgd_list;
++ if (pgd_list)
++ pgd_list->private = (unsigned long)&page->index;
++ pgd_list = page;
++ page->private = (unsigned long)&pgd_list;
++ spin_unlock(&pgd_lock);
++}
++
++static inline void pgd_list_del(pgd_t *pgd)
++{
++ struct page *next, **pprev, *page = virt_to_page(pgd);
++
++ spin_lock(&pgd_lock);
++ next = (struct page *)page->index;
++ pprev = (struct page **)page->private;
++ *pprev = next;
++ if (next)
++ next->private = (unsigned long)pprev;
++ spin_unlock(&pgd_lock);
++}
++
++static inline pgd_t *pgd_alloc(struct mm_struct *mm)
++{
++ /*
++ * We allocate two contiguous pages for kernel and user.
++ */
++ unsigned boundary;
++ pgd_t *pgd = (pgd_t *)__get_free_pages(GFP_KERNEL|__GFP_REPEAT, 1);
++
++ if (!pgd)
++ return NULL;
++ pgd_list_add(pgd);
++ /*
++ * Copy kernel pointers in from init.
++ * Could keep a freelist or slab cache of those because the kernel
++ * part never changes.
++ */
++ boundary = pgd_index(__PAGE_OFFSET);
++ memset(pgd, 0, boundary * sizeof(pgd_t));
++ memcpy(pgd + boundary,
++ init_level4_pgt + boundary,
++ (PTRS_PER_PGD - boundary) * sizeof(pgd_t));
++
++ memset(__user_pgd(pgd), 0, PAGE_SIZE); /* clean up user pgd */
++ /*
++ * Set level3_user_pgt for vsyscall area
++ */
++ set_pgd(__user_pgd(pgd) + pgd_index(VSYSCALL_START),
++ mk_kernel_pgd(__pa_symbol(level3_user_pgt)));
++ return pgd;
++}
++
++static inline void pgd_free(pgd_t *pgd)
++{
++ pte_t *ptep = virt_to_ptep(pgd);
++
++ if (!pte_write(*ptep)) {
++ xen_pgd_unpin(__pa(pgd));
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)pgd,
++ pfn_pte(virt_to_phys(pgd)>>PAGE_SHIFT, PAGE_KERNEL),
++ 0));
++ }
++
++ ptep = virt_to_ptep(__user_pgd(pgd));
++
++ if (!pte_write(*ptep)) {
++ xen_pgd_unpin(__pa(__user_pgd(pgd)));
++ BUG_ON(HYPERVISOR_update_va_mapping(
++ (unsigned long)__user_pgd(pgd),
++ pfn_pte(virt_to_phys(__user_pgd(pgd))>>PAGE_SHIFT,
++ PAGE_KERNEL),
++ 0));
++ }
++
++ pgd_list_del(pgd);
++ free_pages((unsigned long)pgd, 1);
++}
++
++static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm, unsigned long address)
++{
++ pte_t *pte = (pte_t *)get_zeroed_page(GFP_KERNEL|__GFP_REPEAT);
++ if (pte)
++ make_page_readonly(pte, XENFEAT_writable_page_tables);
++
++ return pte;
++}
++
++static inline struct page *pte_alloc_one(struct mm_struct *mm, unsigned long address)
++{
++ struct page *pte;
++
++ pte = alloc_pages(GFP_KERNEL|__GFP_REPEAT|__GFP_ZERO, 0);
++ return pte;
++}
++
++/* Should really implement gc for free page table pages. This could be
++ done with a reference count in struct page. */
++
++static inline void pte_free_kernel(pte_t *pte)
++{
++ BUG_ON((unsigned long)pte & (PAGE_SIZE-1));
++ make_page_writable(pte, XENFEAT_writable_page_tables);
++ free_page((unsigned long)pte);
++}
++
++extern void pte_free(struct page *pte);
++
++//#define __pte_free_tlb(tlb,pte) tlb_remove_page((tlb),(pte))
++//#define __pmd_free_tlb(tlb,x) tlb_remove_page((tlb),virt_to_page(x))
++//#define __pud_free_tlb(tlb,x) tlb_remove_page((tlb),virt_to_page(x))
++
++#define __pte_free_tlb(tlb,x) pte_free((x))
++#define __pmd_free_tlb(tlb,x) pmd_free((x))
++#define __pud_free_tlb(tlb,x) pud_free((x))
++
++#endif /* _X86_64_PGALLOC_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/pgtable.h new/include/asm-x86_64/mach-xen/asm/pgtable.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/pgtable.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/pgtable.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,564 @@
++#ifndef _X86_64_PGTABLE_H
++#define _X86_64_PGTABLE_H
++
++/*
++ * This file contains the functions and defines necessary to modify and use
++ * the x86-64 page table tree.
++ */
++#include <asm/processor.h>
++#include <asm/fixmap.h>
++#include <asm/bitops.h>
++#include <linux/threads.h>
++#include <linux/sched.h>
++#include <asm/pda.h>
++#ifdef CONFIG_XEN
++#include <asm/hypervisor.h>
++
++extern pud_t level3_user_pgt[512];
++extern pud_t init_level4_user_pgt[];
++
++extern void xen_init_pt(void);
++
++#define virt_to_ptep(__va) \
++({ \
++ pgd_t *__pgd = pgd_offset_k((unsigned long)(__va)); \
++ pud_t *__pud = pud_offset(__pgd, (unsigned long)(__va)); \
++ pmd_t *__pmd = pmd_offset(__pud, (unsigned long)(__va)); \
++ pte_offset_kernel(__pmd, (unsigned long)(__va)); \
++})
++
++#define arbitrary_virt_to_machine(__va) \
++({ \
++ maddr_t m = (maddr_t)pte_mfn(*virt_to_ptep(__va)) << PAGE_SHIFT;\
++ m | ((unsigned long)(__va) & (PAGE_SIZE-1)); \
++})
++#endif
++
++extern pud_t level3_kernel_pgt[512];
++extern pud_t level3_physmem_pgt[512];
++extern pud_t level3_ident_pgt[512];
++extern pmd_t level2_kernel_pgt[512];
++extern pgd_t init_level4_pgt[];
++extern pgd_t boot_level4_pgt[];
++extern unsigned long __supported_pte_mask;
++
++#define swapper_pg_dir init_level4_pgt
++
++extern int nonx_setup(char *str);
++extern void paging_init(void);
++extern void clear_kernel_mapping(unsigned long addr, unsigned long size);
++
++extern unsigned long pgkern_mask;
++
++/*
++ * ZERO_PAGE is a global shared page that is always zero: used
++ * for zero-mapped memory areas etc..
++ */
++extern unsigned long empty_zero_page[PAGE_SIZE/sizeof(unsigned long)];
++#define ZERO_PAGE(vaddr) (virt_to_page(empty_zero_page))
++
++/*
++ * PGDIR_SHIFT determines what a top-level page table entry can map
++ */
++#define PGDIR_SHIFT 39
++#define PTRS_PER_PGD 512
++
++/*
++ * 3rd level page
++ */
++#define PUD_SHIFT 30
++#define PTRS_PER_PUD 512
++
++/*
++ * PMD_SHIFT determines the size of the area a middle-level
++ * page table can map
++ */
++#define PMD_SHIFT 21
++#define PTRS_PER_PMD 512
++
++/*
++ * entries per page directory level
++ */
++#define PTRS_PER_PTE 512
++
++#define pte_ERROR(e) \
++ printk("%s:%d: bad pte %p(%016lx).\n", __FILE__, __LINE__, &(e), pte_val(e))
++#define pmd_ERROR(e) \
++ printk("%s:%d: bad pmd %p(%016lx).\n", __FILE__, __LINE__, &(e), pmd_val(e))
++#define pud_ERROR(e) \
++ printk("%s:%d: bad pud %p(%016lx).\n", __FILE__, __LINE__, &(e), pud_val(e))
++#define pgd_ERROR(e) \
++ printk("%s:%d: bad pgd %p(%016lx).\n", __FILE__, __LINE__, &(e), pgd_val(e))
++
++#define pgd_none(x) (!pgd_val(x))
++#define pud_none(x) (!pud_val(x))
++
++#define set_pte_batched(pteptr, pteval) \
++ queue_l1_entry_update(pteptr, (pteval))
++
++extern inline int pud_present(pud_t pud) { return !pud_none(pud); }
++
++static inline void set_pte(pte_t *dst, pte_t val)
++{
++ *dst = val;
++}
++
++#define set_pmd(pmdptr, pmdval) xen_l2_entry_update(pmdptr, (pmdval))
++#define set_pud(pudptr, pudval) xen_l3_entry_update(pudptr, (pudval))
++#define set_pgd(pgdptr, pgdval) xen_l4_entry_update(pgdptr, (pgdval))
++
++static inline void pud_clear (pud_t * pud)
++{
++ set_pud(pud, __pud(0));
++}
++
++#define __user_pgd(pgd) ((pgd) + PTRS_PER_PGD)
++
++static inline void pgd_clear (pgd_t * pgd)
++{
++ set_pgd(pgd, __pgd(0));
++ set_pgd(__user_pgd(pgd), __pgd(0));
++}
++
++#define pud_page(pud) \
++ ((unsigned long) __va(pud_val(pud) & PHYSICAL_PAGE_MASK))
++
++/*
++ * A note on implementation of this atomic 'get-and-clear' operation.
++ * This is actually very simple because Xen Linux can only run on a single
++ * processor. Therefore, we cannot race other processors setting the 'accessed'
++ * or 'dirty' bits on a page-table entry.
++ * Even if pages are shared between domains, that is not a problem because
++ * each domain will have separate page tables, with their own versions of
++ * accessed & dirty state.
++ */
++#define ptep_get_and_clear(mm,addr,xp) __pte_ma(xchg(&(xp)->pte, 0))
++
++#if 0
++static inline pte_t ptep_get_and_clear(struct mm_struct *mm, unsigned long addr, pte_t *xp)
++{
++ pte_t pte = *xp;
++ if (pte.pte)
++ set_pte(xp, __pte_ma(0));
++ return pte;
++}
++#endif
++
++struct mm_struct;
++
++static inline pte_t ptep_get_and_clear_full(struct mm_struct *mm, unsigned long addr, pte_t *ptep, int full)
++{
++ pte_t pte;
++ if (full) {
++ pte = *ptep;
++ *ptep = __pte(0);
++ } else {
++ pte = ptep_get_and_clear(mm, addr, ptep);
++ }
++ return pte;
++}
++
++#define pte_same(a, b) ((a).pte == (b).pte)
++
++#define pte_pgprot(a) (__pgprot((a).pte & ~PHYSICAL_PAGE_MASK))
++
++#define PMD_SIZE (1UL << PMD_SHIFT)
++#define PMD_MASK (~(PMD_SIZE-1))
++#define PUD_SIZE (1UL << PUD_SHIFT)
++#define PUD_MASK (~(PUD_SIZE-1))
++#define PGDIR_SIZE (1UL << PGDIR_SHIFT)
++#define PGDIR_MASK (~(PGDIR_SIZE-1))
++
++#define USER_PTRS_PER_PGD ((TASK_SIZE-1)/PGDIR_SIZE+1)
++#define FIRST_USER_ADDRESS 0
++
++#ifndef __ASSEMBLY__
++#define MAXMEM 0x3fffffffffffUL
++#define VMALLOC_START 0xffffc20000000000UL
++#define VMALLOC_END 0xffffe1ffffffffffUL
++#define MODULES_VADDR 0xffffffff88000000UL
++#define MODULES_END 0xfffffffffff00000UL
++#define MODULES_LEN (MODULES_END - MODULES_VADDR)
++
++#define _PAGE_BIT_PRESENT 0
++#define _PAGE_BIT_RW 1
++#define _PAGE_BIT_USER 2
++#define _PAGE_BIT_PWT 3
++#define _PAGE_BIT_PCD 4
++#define _PAGE_BIT_ACCESSED 5
++#define _PAGE_BIT_DIRTY 6
++#define _PAGE_BIT_PSE 7 /* 4 MB (or 2MB) page */
++#define _PAGE_BIT_GLOBAL 8 /* Global TLB entry PPro+ */
++#define _PAGE_BIT_NX 63 /* No execute: only valid after cpuid check */
++
++#define _PAGE_PRESENT 0x001
++#define _PAGE_RW 0x002
++#define _PAGE_USER 0x004
++#define _PAGE_PWT 0x008
++#define _PAGE_PCD 0x010
++#define _PAGE_ACCESSED 0x020
++#define _PAGE_DIRTY 0x040
++#define _PAGE_PSE 0x080 /* 2MB page */
++#define _PAGE_FILE 0x040 /* nonlinear file mapping, saved PTE; unset:swap */
++#define _PAGE_GLOBAL 0x100 /* Global TLB entry */
++
++#define _PAGE_PROTNONE 0x080 /* If not present */
++#define _PAGE_NX (1UL<<_PAGE_BIT_NX)
++
++#define _PAGE_TABLE (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_DIRTY)
++#define _KERNPG_TABLE _PAGE_TABLE
++
++#define _PAGE_CHG_MASK (PTE_MASK | _PAGE_ACCESSED | _PAGE_DIRTY)
++
++#define PAGE_NONE __pgprot(_PAGE_PROTNONE | _PAGE_ACCESSED)
++#define PAGE_SHARED __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
++#define PAGE_SHARED_EXEC __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER | _PAGE_ACCESSED)
++#define PAGE_COPY_NOEXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
++#define PAGE_COPY PAGE_COPY_NOEXEC
++#define PAGE_COPY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
++#define PAGE_READONLY __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_NX)
++#define PAGE_READONLY_EXEC __pgprot(_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED)
++#define __PAGE_KERNEL \
++ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX | _PAGE_USER )
++#define __PAGE_KERNEL_EXEC \
++ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_USER )
++#define __PAGE_KERNEL_NOCACHE \
++ (_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_PCD | _PAGE_ACCESSED | _PAGE_NX | _PAGE_USER )
++#define __PAGE_KERNEL_RO \
++ (_PAGE_PRESENT | _PAGE_DIRTY | _PAGE_ACCESSED | _PAGE_NX | _PAGE_USER )
++#define __PAGE_KERNEL_VSYSCALL \
++ (_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_USER )
++#define __PAGE_KERNEL_VSYSCALL_NOCACHE \
++ (_PAGE_PRESENT | _PAGE_USER | _PAGE_ACCESSED | _PAGE_PCD | _PAGE_USER )
++#define __PAGE_KERNEL_LARGE \
++ (__PAGE_KERNEL | _PAGE_PSE | _PAGE_USER )
++#define __PAGE_KERNEL_LARGE_EXEC \
++ (__PAGE_KERNEL_EXEC | _PAGE_PSE | _PAGE_USER )
++
++
++/*
++ * We don't support GLOBAL page in xenolinux64
++ */
++#define MAKE_GLOBAL(x) __pgprot((x))
++
++#define PAGE_KERNEL MAKE_GLOBAL(__PAGE_KERNEL)
++#define PAGE_KERNEL_EXEC MAKE_GLOBAL(__PAGE_KERNEL_EXEC)
++#define PAGE_KERNEL_RO MAKE_GLOBAL(__PAGE_KERNEL_RO)
++#define PAGE_KERNEL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_NOCACHE)
++#define PAGE_KERNEL_VSYSCALL32 __pgprot(__PAGE_KERNEL_VSYSCALL)
++#define PAGE_KERNEL_VSYSCALL MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL)
++#define PAGE_KERNEL_LARGE MAKE_GLOBAL(__PAGE_KERNEL_LARGE)
++#define PAGE_KERNEL_VSYSCALL_NOCACHE MAKE_GLOBAL(__PAGE_KERNEL_VSYSCALL_NOCACHE)
++
++/* xwr */
++#define __P000 PAGE_NONE
++#define __P001 PAGE_READONLY
++#define __P010 PAGE_COPY
++#define __P011 PAGE_COPY
++#define __P100 PAGE_READONLY_EXEC
++#define __P101 PAGE_READONLY_EXEC
++#define __P110 PAGE_COPY_EXEC
++#define __P111 PAGE_COPY_EXEC
++
++#define __S000 PAGE_NONE
++#define __S001 PAGE_READONLY
++#define __S010 PAGE_SHARED
++#define __S011 PAGE_SHARED
++#define __S100 PAGE_READONLY_EXEC
++#define __S101 PAGE_READONLY_EXEC
++#define __S110 PAGE_SHARED_EXEC
++#define __S111 PAGE_SHARED_EXEC
++
++static inline unsigned long pgd_bad(pgd_t pgd)
++{
++ unsigned long val = pgd_val(pgd);
++ val &= ~PTE_MASK;
++ val &= ~(_PAGE_USER | _PAGE_DIRTY);
++ return val & ~(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED);
++}
++
++static inline unsigned long pud_bad(pud_t pud)
++{
++ unsigned long val = pud_val(pud);
++ val &= ~PTE_MASK;
++ val &= ~(_PAGE_USER | _PAGE_DIRTY);
++ return val & ~(_PAGE_PRESENT | _PAGE_RW | _PAGE_ACCESSED);
++}
++
++#define set_pte_at(_mm,addr,ptep,pteval) do { \
++ if (((_mm) != current->mm && (_mm) != &init_mm) || \
++ HYPERVISOR_update_va_mapping((addr), (pteval), 0)) \
++ set_pte((ptep), (pteval)); \
++} while (0)
++
++#define pte_none(x) (!(x).pte)
++#define pte_present(x) ((x).pte & (_PAGE_PRESENT | _PAGE_PROTNONE))
++#define pte_clear(mm,addr,xp) do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
++
++#define pages_to_mb(x) ((x) >> (20-PAGE_SHIFT))
++
++#define pte_mfn(_pte) (((_pte).pte & PTE_MASK) >> PAGE_SHIFT)
++#define pte_pfn(_pte) mfn_to_local_pfn(pte_mfn(_pte))
++
++#define pte_page(x) pfn_to_page(pte_pfn(x))
++
++static inline pte_t pfn_pte(unsigned long page_nr, pgprot_t pgprot)
++{
++ pte_t pte;
++
++ (pte).pte = (pfn_to_mfn(page_nr) << PAGE_SHIFT);
++ (pte).pte |= pgprot_val(pgprot);
++ (pte).pte &= __supported_pte_mask;
++ return pte;
++}
++
++#define pfn_pte_ma(pfn, prot) __pte_ma((((pfn) << PAGE_SHIFT) | pgprot_val(prot)) & __supported_pte_mask)
++/*
++ * The following only work if pte_present() is true.
++ * Undefined behaviour if not..
++ */
++#define __pte_val(x) ((x).pte)
++
++#define __LARGE_PTE (_PAGE_PSE|_PAGE_PRESENT)
++static inline int pte_user(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
++static inline int pte_read(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
++static inline int pte_exec(pte_t pte) { return __pte_val(pte) & _PAGE_USER; }
++static inline int pte_dirty(pte_t pte) { return __pte_val(pte) & _PAGE_DIRTY; }
++static inline int pte_young(pte_t pte) { return __pte_val(pte) & _PAGE_ACCESSED; }
++static inline int pte_write(pte_t pte) { return __pte_val(pte) & _PAGE_RW; }
++static inline int pte_file(pte_t pte) { return __pte_val(pte) & _PAGE_FILE; }
++static inline int pte_huge(pte_t pte) { return __pte_val(pte) & _PAGE_PSE; }
++
++static inline pte_t pte_rdprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
++static inline pte_t pte_exprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_USER; return pte; }
++static inline pte_t pte_mkclean(pte_t pte) { __pte_val(pte) &= ~_PAGE_DIRTY; return pte; }
++static inline pte_t pte_mkold(pte_t pte) { __pte_val(pte) &= ~_PAGE_ACCESSED; return pte; }
++static inline pte_t pte_wrprotect(pte_t pte) { __pte_val(pte) &= ~_PAGE_RW; return pte; }
++static inline pte_t pte_mkread(pte_t pte) { __pte_val(pte) |= _PAGE_USER; return pte; }
++static inline pte_t pte_mkexec(pte_t pte) { __pte_val(pte) |= _PAGE_USER; return pte; }
++static inline pte_t pte_mkdirty(pte_t pte) { __pte_val(pte) |= _PAGE_DIRTY; return pte; }
++static inline pte_t pte_mkyoung(pte_t pte) { __pte_val(pte) |= _PAGE_ACCESSED; return pte; }
++static inline pte_t pte_mkwrite(pte_t pte) { __pte_val(pte) |= _PAGE_RW; return pte; }
++static inline pte_t pte_mkhuge(pte_t pte) { __pte_val(pte) |= _PAGE_PSE; return pte; }
++
++struct vm_area_struct;
++
++static inline int ptep_test_and_clear_dirty(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
++{
++ pte_t pte = *ptep;
++ int ret = pte_dirty(pte);
++ if (ret)
++ set_pte(ptep, pte_mkclean(pte));
++ return ret;
++}
++
++static inline int ptep_test_and_clear_young(struct vm_area_struct *vma, unsigned long addr, pte_t *ptep)
++{
++ pte_t pte = *ptep;
++ int ret = pte_young(pte);
++ if (ret)
++ set_pte(ptep, pte_mkold(pte));
++ return ret;
++}
++
++static inline void ptep_set_wrprotect(struct mm_struct *mm, unsigned long addr, pte_t *ptep)
++{
++ pte_t pte = *ptep;
++ if (pte_write(pte))
++ set_pte(ptep, pte_wrprotect(pte));
++}
++
++/*
++ * Macro to mark a page protection value as "uncacheable".
++ */
++#define pgprot_noncached(prot) (__pgprot(pgprot_val(prot) | _PAGE_PCD | _PAGE_PWT))
++
++static inline int pmd_large(pmd_t pte) {
++ return (pmd_val(pte) & __LARGE_PTE) == __LARGE_PTE;
++}
++
++
++/*
++ * Conversion functions: convert a page and protection to a page entry,
++ * and a page entry and page directory to the page they refer to.
++ */
++
++/*
++ * Level 4 access.
++ * Never use these in the common code.
++ */
++#define pgd_page(pgd) ((unsigned long) __va(pgd_val(pgd) & PTE_MASK))
++#define pgd_index(address) (((address) >> PGDIR_SHIFT) & (PTRS_PER_PGD-1))
++#define pgd_offset(mm, addr) ((mm)->pgd + pgd_index(addr))
++#define pgd_offset_k(address) (pgd_t *)(init_level4_pgt + pgd_index(address))
++#define pgd_present(pgd) (pgd_val(pgd) & _PAGE_PRESENT)
++#define mk_kernel_pgd(address) __pgd((address) | _KERNPG_TABLE)
++
++/* PUD - Level3 access */
++/* to find an entry in a page-table-directory. */
++#define pud_index(address) (((address) >> PUD_SHIFT) & (PTRS_PER_PUD-1))
++#define pud_offset(pgd, address) ((pud_t *) pgd_page(*(pgd)) + pud_index(address))
++static inline pud_t *__pud_offset_k(pud_t *pud, unsigned long address)
++{
++ return pud + pud_index(address);
++}
++
++/* Find correct pud via the hidden fourth level page level: */
++
++/* This accesses the reference page table of the boot cpu.
++ Other CPUs get synced lazily via the page fault handler. */
++static inline pud_t *pud_offset_k(pgd_t *pgd, unsigned long address)
++{
++ return pud_offset(pgd_offset_k(address), address);
++}
++
++/* PMD - Level 2 access */
++#define pmd_page_kernel(pmd) ((unsigned long) __va(pmd_val(pmd) & PTE_MASK))
++#define pmd_page(pmd) (pfn_to_page(pmd_val(pmd) >> PAGE_SHIFT))
++
++#define pmd_index(address) (((address) >> PMD_SHIFT) & (PTRS_PER_PMD-1))
++#define pmd_offset(dir, address) ((pmd_t *) pud_page(*(dir)) + \
++ pmd_index(address))
++#define pmd_none(x) (!pmd_val(x))
++/* pmd_present doesn't just test the _PAGE_PRESENT bit since wr.p.t.
++ can temporarily clear it. */
++#define pmd_present(x) (pmd_val(x))
++#define pmd_clear(xp) do { set_pmd(xp, __pmd(0)); } while (0)
++#define pmd_bad(x) ((pmd_val(x) & (~PAGE_MASK & ~_PAGE_PRESENT)) != (_KERNPG_TABLE & ~_PAGE_PRESENT))
++#define pfn_pmd(nr,prot) (__pmd(((nr) << PAGE_SHIFT) | pgprot_val(prot)))
++#define pmd_pfn(x) ((pmd_val(x) & __PHYSICAL_MASK) >> PAGE_SHIFT)
++
++#define pte_to_pgoff(pte) ((pte_val(pte) & PHYSICAL_PAGE_MASK) >> PAGE_SHIFT)
++#define pgoff_to_pte(off) ((pte_t) { ((off) << PAGE_SHIFT) | _PAGE_FILE })
++#define PTE_FILE_MAX_BITS __PHYSICAL_MASK_SHIFT
++
++/* PTE - Level 1 access. */
++
++/* page, protection -> pte */
++#define mk_pte(page, pgprot) pfn_pte(page_to_pfn(page), (pgprot))
++#define mk_pte_huge(entry) (pte_val(entry) |= _PAGE_PRESENT | _PAGE_PSE)
++
++/* physical address -> PTE */
++static inline pte_t mk_pte_phys(unsigned long physpage, pgprot_t pgprot)
++{
++ pte_t pte;
++ (pte).pte = physpage | pgprot_val(pgprot);
++ return pte;
++}
++
++/* Change flags of a PTE */
++static inline pte_t pte_modify(pte_t pte, pgprot_t newprot)
++{
++ (pte).pte &= _PAGE_CHG_MASK;
++ (pte).pte |= pgprot_val(newprot);
++ (pte).pte &= __supported_pte_mask;
++ return pte;
++}
++
++#define pte_index(address) \
++ (((address) >> PAGE_SHIFT) & (PTRS_PER_PTE - 1))
++#define pte_offset_kernel(dir, address) ((pte_t *) pmd_page_kernel(*(dir)) + \
++ pte_index(address))
++
++/* x86-64 always has all page tables mapped. */
++#define pte_offset_map(dir,address) pte_offset_kernel(dir,address)
++#define pte_offset_map_nested(dir,address) pte_offset_kernel(dir,address)
++#define pte_unmap(pte) /* NOP */
++#define pte_unmap_nested(pte) /* NOP */
++
++#define update_mmu_cache(vma,address,pte) do { } while (0)
++
++/* We only update the dirty/accessed state if we set
++ * the dirty bit by hand in the kernel, since the hardware
++ * will do the accessed bit for us, and we don't want to
++ * race with other CPU's that might be updating the dirty
++ * bit at the same time. */
++#define __HAVE_ARCH_PTEP_SET_ACCESS_FLAGS
++#if 0
++#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
++ do { \
++ if (__dirty) { \
++ set_pte(__ptep, __entry); \
++ flush_tlb_page(__vma, __address); \
++ } \
++ } while (0)
++#endif
++#define ptep_set_access_flags(__vma, __address, __ptep, __entry, __dirty) \
++ do { \
++ if (__dirty) { \
++ if ( likely((__vma)->vm_mm == current->mm) ) { \
++ BUG_ON(HYPERVISOR_update_va_mapping((__address), (__entry), UVMF_INVLPG|UVMF_MULTI|(unsigned long)((__vma)->vm_mm->cpu_vm_mask.bits))); \
++ } else { \
++ xen_l1_entry_update((__ptep), (__entry)); \
++ flush_tlb_page((__vma), (__address)); \
++ } \
++ } \
++ } while (0)
++
++/* Encode and de-code a swap entry */
++#define __swp_type(x) (((x).val >> 1) & 0x3f)
++#define __swp_offset(x) ((x).val >> 8)
++#define __swp_entry(type, offset) ((swp_entry_t) { ((type) << 1) | ((offset) << 8) })
++#define __pte_to_swp_entry(pte) ((swp_entry_t) { pte_val(pte) })
++#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
++
++extern spinlock_t pgd_lock;
++extern struct page *pgd_list;
++void vmalloc_sync_all(void);
++
++#endif /* !__ASSEMBLY__ */
++
++extern int kern_addr_valid(unsigned long addr);
++
++#define DOMID_LOCAL (0xFFFFU)
++
++int direct_remap_pfn_range(struct vm_area_struct *vma,
++ unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid);
++
++int direct_kernel_remap_pfn_range(unsigned long address,
++ unsigned long mfn,
++ unsigned long size,
++ pgprot_t prot,
++ domid_t domid);
++
++int create_lookup_pte_addr(struct mm_struct *mm,
++ unsigned long address,
++ uint64_t *ptep);
++
++int touch_pte_range(struct mm_struct *mm,
++ unsigned long address,
++ unsigned long size);
++
++#define io_remap_pfn_range(vma, vaddr, pfn, size, prot) \
++ direct_remap_pfn_range(vma,vaddr,pfn,size,prot,DOMID_IO)
++
++#define MK_IOSPACE_PFN(space, pfn) (pfn)
++#define GET_IOSPACE(pfn) 0
++#define GET_PFN(pfn) (pfn)
++
++#define HAVE_ARCH_UNMAPPED_AREA
++
++#define pgtable_cache_init() do { } while (0)
++#define check_pgt_cache() do { } while (0)
++
++#define PAGE_AGP PAGE_KERNEL_NOCACHE
++#define HAVE_PAGE_AGP 1
++
++/* fs/proc/kcore.c */
++#define kc_vaddr_to_offset(v) ((v) & __VIRTUAL_MASK)
++#define kc_offset_to_vaddr(o) \
++ (((o) & (1UL << (__VIRTUAL_MASK_SHIFT-1))) ? ((o) | (~__VIRTUAL_MASK)) : (o))
++
++#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_YOUNG
++#define __HAVE_ARCH_PTEP_TEST_AND_CLEAR_DIRTY
++#define __HAVE_ARCH_PTEP_GET_AND_CLEAR
++#define __HAVE_ARCH_PTEP_GET_AND_CLEAR_FULL
++#define __HAVE_ARCH_PTEP_SET_WRPROTECT
++#define __HAVE_ARCH_PTE_SAME
++#include <asm-generic/pgtable.h>
++
++#endif /* _X86_64_PGTABLE_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/processor.h new/include/asm-x86_64/mach-xen/asm/processor.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/processor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/processor.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,494 @@
++/*
++ * include/asm-x86_64/processor.h
++ *
++ * Copyright (C) 1994 Linus Torvalds
++ */
++
++#ifndef __ASM_X86_64_PROCESSOR_H
++#define __ASM_X86_64_PROCESSOR_H
++
++#include <asm/segment.h>
++#include <asm/page.h>
++#include <asm/types.h>
++#include <asm/sigcontext.h>
++#include <asm/cpufeature.h>
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <asm/msr.h>
++#include <asm/current.h>
++#include <asm/system.h>
++#include <asm/mmsegment.h>
++#include <asm/percpu.h>
++#include <linux/personality.h>
++#include <linux/cpumask.h>
++
++#define TF_MASK 0x00000100
++#define IF_MASK 0x00000200
++#define IOPL_MASK 0x00003000
++#define NT_MASK 0x00004000
++#define VM_MASK 0x00020000
++#define AC_MASK 0x00040000
++#define VIF_MASK 0x00080000 /* virtual interrupt flag */
++#define VIP_MASK 0x00100000 /* virtual interrupt pending */
++#define ID_MASK 0x00200000
++
++#define desc_empty(desc) \
++ (!((desc)->a | (desc)->b))
++
++#define desc_equal(desc1, desc2) \
++ (((desc1)->a == (desc2)->a) && ((desc1)->b == (desc2)->b))
++
++/*
++ * Default implementation of macro that returns current
++ * instruction pointer ("program counter").
++ */
++#define current_text_addr() ({ void *pc; asm volatile("leaq 1f(%%rip),%0\n1:":"=r"(pc)); pc; })
++
++/*
++ * CPU type and hardware bug flags. Kept separately for each CPU.
++ */
++
++struct cpuinfo_x86 {
++ __u8 x86; /* CPU family */
++ __u8 x86_vendor; /* CPU vendor */
++ __u8 x86_model;
++ __u8 x86_mask;
++ int cpuid_level; /* Maximum supported CPUID level, -1=no CPUID */
++ __u32 x86_capability[NCAPINTS];
++ char x86_vendor_id[16];
++ char x86_model_id[64];
++ int x86_cache_size; /* in KB */
++ int x86_clflush_size;
++ int x86_cache_alignment;
++ int x86_tlbsize; /* number of 4K pages in DTLB/ITLB combined(in pages)*/
++ __u8 x86_virt_bits, x86_phys_bits;
++ __u8 x86_max_cores; /* cpuid returned max cores value */
++ __u32 x86_power;
++ __u32 extended_cpuid_level; /* Max extended CPUID function supported */
++ unsigned long loops_per_jiffy;
++#ifdef CONFIG_SMP
++ cpumask_t llc_shared_map; /* cpus sharing the last level cache */
++#endif
++ __u8 apicid;
++ __u8 booted_cores; /* number of cores as seen by OS */
++} ____cacheline_aligned;
++
++#define X86_VENDOR_INTEL 0
++#define X86_VENDOR_CYRIX 1
++#define X86_VENDOR_AMD 2
++#define X86_VENDOR_UMC 3
++#define X86_VENDOR_NEXGEN 4
++#define X86_VENDOR_CENTAUR 5
++#define X86_VENDOR_RISE 6
++#define X86_VENDOR_TRANSMETA 7
++#define X86_VENDOR_NUM 8
++#define X86_VENDOR_UNKNOWN 0xff
++
++#ifdef CONFIG_SMP
++extern struct cpuinfo_x86 cpu_data[];
++#define current_cpu_data cpu_data[smp_processor_id()]
++#else
++#define cpu_data (&boot_cpu_data)
++#define current_cpu_data boot_cpu_data
++#endif
++
++extern char ignore_irq13;
++
++extern void identify_cpu(struct cpuinfo_x86 *);
++extern void print_cpu_info(struct cpuinfo_x86 *);
++extern unsigned int init_intel_cacheinfo(struct cpuinfo_x86 *c);
++
++/*
++ * EFLAGS bits
++ */
++#define X86_EFLAGS_CF 0x00000001 /* Carry Flag */
++#define X86_EFLAGS_PF 0x00000004 /* Parity Flag */
++#define X86_EFLAGS_AF 0x00000010 /* Auxillary carry Flag */
++#define X86_EFLAGS_ZF 0x00000040 /* Zero Flag */
++#define X86_EFLAGS_SF 0x00000080 /* Sign Flag */
++#define X86_EFLAGS_TF 0x00000100 /* Trap Flag */
++#define X86_EFLAGS_IF 0x00000200 /* Interrupt Flag */
++#define X86_EFLAGS_DF 0x00000400 /* Direction Flag */
++#define X86_EFLAGS_OF 0x00000800 /* Overflow Flag */
++#define X86_EFLAGS_IOPL 0x00003000 /* IOPL mask */
++#define X86_EFLAGS_NT 0x00004000 /* Nested Task */
++#define X86_EFLAGS_RF 0x00010000 /* Resume Flag */
++#define X86_EFLAGS_VM 0x00020000 /* Virtual Mode */
++#define X86_EFLAGS_AC 0x00040000 /* Alignment Check */
++#define X86_EFLAGS_VIF 0x00080000 /* Virtual Interrupt Flag */
++#define X86_EFLAGS_VIP 0x00100000 /* Virtual Interrupt Pending */
++#define X86_EFLAGS_ID 0x00200000 /* CPUID detection flag */
++
++/*
++ * Intel CPU features in CR4
++ */
++#define X86_CR4_VME 0x0001 /* enable vm86 extensions */
++#define X86_CR4_PVI 0x0002 /* virtual interrupts flag enable */
++#define X86_CR4_TSD 0x0004 /* disable time stamp at ipl 3 */
++#define X86_CR4_DE 0x0008 /* enable debugging extensions */
++#define X86_CR4_PSE 0x0010 /* enable page size extensions */
++#define X86_CR4_PAE 0x0020 /* enable physical address extensions */
++#define X86_CR4_MCE 0x0040 /* Machine check enable */
++#define X86_CR4_PGE 0x0080 /* enable global pages */
++#define X86_CR4_PCE 0x0100 /* enable performance counters at ipl 3 */
++#define X86_CR4_OSFXSR 0x0200 /* enable fast FPU save and restore */
++#define X86_CR4_OSXMMEXCPT 0x0400 /* enable unmasked SSE exceptions */
++
++/*
++ * Save the cr4 feature set we're using (ie
++ * Pentium 4MB enable and PPro Global page
++ * enable), so that any CPU's that boot up
++ * after us can get the correct flags.
++ */
++extern unsigned long mmu_cr4_features;
++
++static inline void set_in_cr4 (unsigned long mask)
++{
++ mmu_cr4_features |= mask;
++ __asm__("movq %%cr4,%%rax\n\t"
++ "orq %0,%%rax\n\t"
++ "movq %%rax,%%cr4\n"
++ : : "irg" (mask)
++ :"ax");
++}
++
++static inline void clear_in_cr4 (unsigned long mask)
++{
++ mmu_cr4_features &= ~mask;
++ __asm__("movq %%cr4,%%rax\n\t"
++ "andq %0,%%rax\n\t"
++ "movq %%rax,%%cr4\n"
++ : : "irg" (~mask)
++ :"ax");
++}
++
++
++/*
++ * Bus types
++ */
++#define MCA_bus 0
++#define MCA_bus__is_a_macro
++
++/*
++ * User space process size. 47bits minus one guard page.
++ */
++#define TASK_SIZE64 (0x800000000000UL - 4096)
++
++/* This decides where the kernel will search for a free chunk of vm
++ * space during mmap's.
++ */
++#define IA32_PAGE_OFFSET ((current->personality & ADDR_LIMIT_3GB) ? 0xc0000000 : 0xFFFFe000)
++
++#define TASK_SIZE (test_thread_flag(TIF_IA32) ? IA32_PAGE_OFFSET : TASK_SIZE64)
++#define TASK_SIZE_OF(child) ((test_tsk_thread_flag(child, TIF_IA32)) ? IA32_PAGE_OFFSET : TASK_SIZE64)
++
++#define TASK_UNMAPPED_BASE PAGE_ALIGN(TASK_SIZE/3)
++
++/*
++ * Size of io_bitmap.
++ */
++#define IO_BITMAP_BITS 65536
++#define IO_BITMAP_BYTES (IO_BITMAP_BITS/8)
++#define IO_BITMAP_LONGS (IO_BITMAP_BYTES/sizeof(long))
++#ifndef CONFIG_X86_NO_TSS
++#define IO_BITMAP_OFFSET offsetof(struct tss_struct,io_bitmap)
++#endif
++#define INVALID_IO_BITMAP_OFFSET 0x8000
++
++struct i387_fxsave_struct {
++ u16 cwd;
++ u16 swd;
++ u16 twd;
++ u16 fop;
++ u64 rip;
++ u64 rdp;
++ u32 mxcsr;
++ u32 mxcsr_mask;
++ u32 st_space[32]; /* 8*16 bytes for each FP-reg = 128 bytes */
++ u32 xmm_space[64]; /* 16*16 bytes for each XMM-reg = 128 bytes */
++ u32 padding[24];
++} __attribute__ ((aligned (16)));
++
++union i387_union {
++ struct i387_fxsave_struct fxsave;
++};
++
++#ifndef CONFIG_X86_NO_TSS
++struct tss_struct {
++ u32 reserved1;
++ u64 rsp0;
++ u64 rsp1;
++ u64 rsp2;
++ u64 reserved2;
++ u64 ist[7];
++ u32 reserved3;
++ u32 reserved4;
++ u16 reserved5;
++ u16 io_bitmap_base;
++ /*
++ * The extra 1 is there because the CPU will access an
++ * additional byte beyond the end of the IO permission
++ * bitmap. The extra byte must be all 1 bits, and must
++ * be within the limit. Thus we have:
++ *
++ * 128 bytes, the bitmap itself, for ports 0..0x3ff
++ * 8 bytes, for an extra "long" of ~0UL
++ */
++ unsigned long io_bitmap[IO_BITMAP_LONGS + 1];
++} __attribute__((packed)) ____cacheline_aligned;
++
++DECLARE_PER_CPU(struct tss_struct,init_tss);
++#endif
++
++extern struct cpuinfo_x86 boot_cpu_data;
++
++#ifdef CONFIG_X86_VSMP
++#define ARCH_MIN_TASKALIGN (1 << INTERNODE_CACHE_SHIFT)
++#define ARCH_MIN_MMSTRUCT_ALIGN (1 << INTERNODE_CACHE_SHIFT)
++#else
++#define ARCH_MIN_TASKALIGN 16
++#define ARCH_MIN_MMSTRUCT_ALIGN 0
++#endif
++
++struct thread_struct {
++ unsigned long rsp0;
++ unsigned long rsp;
++ unsigned long userrsp; /* Copy from PDA */
++ unsigned long fs;
++ unsigned long gs;
++ unsigned short es, ds, fsindex, gsindex;
++/* Hardware debugging registers */
++ unsigned long debugreg0;
++ unsigned long debugreg1;
++ unsigned long debugreg2;
++ unsigned long debugreg3;
++ unsigned long debugreg6;
++ unsigned long debugreg7;
++/* fault info */
++ unsigned long cr2, trap_no, error_code;
++/* floating point info */
++ union i387_union i387 __attribute__((aligned(16)));
++/* IO permissions. the bitmap could be moved into the GDT, that would make
++ switch faster for a limited number of ioperm using tasks. -AK */
++ int ioperm;
++ unsigned long *io_bitmap_ptr;
++ unsigned io_bitmap_max;
++/* cached TLS descriptors. */
++ u64 tls_array[GDT_ENTRY_TLS_ENTRIES];
++ unsigned int iopl;
++} __attribute__((aligned(16)));
++
++#define INIT_THREAD { \
++ .rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
++}
++
++#ifndef CONFIG_X86_NO_TSS
++#define INIT_TSS { \
++ .rsp0 = (unsigned long)&init_stack + sizeof(init_stack) \
++}
++#endif
++
++#define INIT_MMAP \
++{ &init_mm, 0, 0, NULL, PAGE_SHARED, VM_READ | VM_WRITE | VM_EXEC, 1, NULL, NULL }
++
++#define start_thread(regs,new_rip,new_rsp) do { \
++ asm volatile("movl %0,%%fs; movl %0,%%es; movl %0,%%ds": :"r" (0)); \
++ load_gs_index(0); \
++ (regs)->rip = (new_rip); \
++ (regs)->rsp = (new_rsp); \
++ write_pda(oldrsp, (new_rsp)); \
++ (regs)->cs = __USER_CS; \
++ (regs)->ss = __USER_DS; \
++ (regs)->eflags = 0x200; \
++ set_fs(USER_DS); \
++} while(0)
++
++#define get_debugreg(var, register) \
++ var = HYPERVISOR_get_debugreg(register)
++#define set_debugreg(value, register) \
++ HYPERVISOR_set_debugreg(register, value)
++
++struct task_struct;
++struct mm_struct;
++
++/* Free all resources held by a thread. */
++extern void release_thread(struct task_struct *);
++
++/* Prepare to copy thread state - unlazy all lazy status */
++extern void prepare_to_copy(struct task_struct *tsk);
++
++/*
++ * create a kernel thread without removing it from tasklists
++ */
++extern long kernel_thread(int (*fn)(void *), void * arg, unsigned long flags);
++
++/*
++ * Return saved PC of a blocked thread.
++ * What is this good for? it will be always the scheduler or ret_from_fork.
++ */
++#define thread_saved_pc(t) (*(unsigned long *)((t)->thread.rsp - 8))
++
++extern unsigned long get_wchan(struct task_struct *p);
++#define task_pt_regs(tsk) ((struct pt_regs *)(tsk)->thread.rsp0 - 1)
++#define KSTK_EIP(tsk) (task_pt_regs(tsk)->rip)
++#define KSTK_ESP(tsk) -1 /* sorry. doesn't work for syscall. */
++
++
++struct microcode_header {
++ unsigned int hdrver;
++ unsigned int rev;
++ unsigned int date;
++ unsigned int sig;
++ unsigned int cksum;
++ unsigned int ldrver;
++ unsigned int pf;
++ unsigned int datasize;
++ unsigned int totalsize;
++ unsigned int reserved[3];
++};
++
++struct microcode {
++ struct microcode_header hdr;
++ unsigned int bits[0];
++};
++
++typedef struct microcode microcode_t;
++typedef struct microcode_header microcode_header_t;
++
++/* microcode format is extended from prescott processors */
++struct extended_signature {
++ unsigned int sig;
++ unsigned int pf;
++ unsigned int cksum;
++};
++
++struct extended_sigtable {
++ unsigned int count;
++ unsigned int cksum;
++ unsigned int reserved[3];
++ struct extended_signature sigs[0];
++};
++
++
++#define ASM_NOP1 K8_NOP1
++#define ASM_NOP2 K8_NOP2
++#define ASM_NOP3 K8_NOP3
++#define ASM_NOP4 K8_NOP4
++#define ASM_NOP5 K8_NOP5
++#define ASM_NOP6 K8_NOP6
++#define ASM_NOP7 K8_NOP7
++#define ASM_NOP8 K8_NOP8
++
++/* Opteron nops */
++#define K8_NOP1 ".byte 0x90\n"
++#define K8_NOP2 ".byte 0x66,0x90\n"
++#define K8_NOP3 ".byte 0x66,0x66,0x90\n"
++#define K8_NOP4 ".byte 0x66,0x66,0x66,0x90\n"
++#define K8_NOP5 K8_NOP3 K8_NOP2
++#define K8_NOP6 K8_NOP3 K8_NOP3
++#define K8_NOP7 K8_NOP4 K8_NOP3
++#define K8_NOP8 K8_NOP4 K8_NOP4
++
++#define ASM_NOP_MAX 8
++
++/* REP NOP (PAUSE) is a good thing to insert into busy-wait loops. */
++static inline void rep_nop(void)
++{
++ __asm__ __volatile__("rep;nop": : :"memory");
++}
++
++/* Stop speculative execution */
++static inline void sync_core(void)
++{
++ int tmp;
++ asm volatile("cpuid" : "=a" (tmp) : "0" (1) : "ebx","ecx","edx","memory");
++}
++
++#define cpu_has_fpu 1
++
++#define ARCH_HAS_PREFETCH
++static inline void prefetch(void *x)
++{
++ asm volatile("prefetcht0 %0" :: "m" (*(unsigned long *)x));
++}
++
++#define ARCH_HAS_PREFETCHW 1
++static inline void prefetchw(void *x)
++{
++ alternative_input("prefetcht0 (%1)",
++ "prefetchw (%1)",
++ X86_FEATURE_3DNOW,
++ "r" (x));
++}
++
++#define ARCH_HAS_SPINLOCK_PREFETCH 1
++
++#define spin_lock_prefetch(x) prefetchw(x)
++
++#define cpu_relax() rep_nop()
++
++/*
++ * NSC/Cyrix CPU configuration register indexes
++ */
++#define CX86_CCR0 0xc0
++#define CX86_CCR1 0xc1
++#define CX86_CCR2 0xc2
++#define CX86_CCR3 0xc3
++#define CX86_CCR4 0xe8
++#define CX86_CCR5 0xe9
++#define CX86_CCR6 0xea
++#define CX86_CCR7 0xeb
++#define CX86_DIR0 0xfe
++#define CX86_DIR1 0xff
++#define CX86_ARR_BASE 0xc4
++#define CX86_RCR_BASE 0xdc
++
++/*
++ * NSC/Cyrix CPU indexed register access macros
++ */
++
++#define getCx86(reg) ({ outb((reg), 0x22); inb(0x23); })
++
++#define setCx86(reg, data) do { \
++ outb((reg), 0x22); \
++ outb((data), 0x23); \
++} while (0)
++
++static inline void serialize_cpu(void)
++{
++ __asm__ __volatile__ ("cpuid" : : : "ax", "bx", "cx", "dx");
++}
++
++static inline void __monitor(const void *eax, unsigned long ecx,
++ unsigned long edx)
++{
++ /* "monitor %eax,%ecx,%edx;" */
++ asm volatile(
++ ".byte 0x0f,0x01,0xc8;"
++ : :"a" (eax), "c" (ecx), "d"(edx));
++}
++
++static inline void __mwait(unsigned long eax, unsigned long ecx)
++{
++ /* "mwait %eax,%ecx;" */
++ asm volatile(
++ ".byte 0x0f,0x01,0xc9;"
++ : :"a" (eax), "c" (ecx));
++}
++
++#define stack_current() \
++({ \
++ struct thread_info *ti; \
++ asm("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \
++ ti->task; \
++})
++
++#define cache_line_size() (boot_cpu_data.x86_cache_alignment)
++
++extern unsigned long boot_option_idle_override;
++/* Boot loader type from the setup header */
++extern int bootloader_type;
++
++#define HAVE_ARCH_PICK_MMAP_LAYOUT 1
++
++#endif /* __ASM_X86_64_PROCESSOR_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/ptrace.h new/include/asm-x86_64/mach-xen/asm/ptrace.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/ptrace.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/ptrace.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,125 @@
++#ifndef _X86_64_PTRACE_H
++#define _X86_64_PTRACE_H
++
++#if defined(__ASSEMBLY__) || defined(__FRAME_OFFSETS)
++#define R15 0
++#define R14 8
++#define R13 16
++#define R12 24
++#define RBP 32
++#define RBX 40
++/* arguments: interrupts/non tracing syscalls only save upto here*/
++#define R11 48
++#define R10 56
++#define R9 64
++#define R8 72
++#define RAX 80
++#define RCX 88
++#define RDX 96
++#define RSI 104
++#define RDI 112
++#define ORIG_RAX 120 /* = ERROR */
++/* end of arguments */
++/* cpu exception frame or undefined in case of fast syscall. */
++#define RIP 128
++#define CS 136
++#define EFLAGS 144
++#define RSP 152
++#define SS 160
++#define ARGOFFSET R11
++#endif /* __ASSEMBLY__ */
++
++/* top of stack page */
++#define FRAME_SIZE 168
++
++#define PTRACE_OLDSETOPTIONS 21
++
++#ifndef __ASSEMBLY__
++
++struct pt_regs {
++ unsigned long r15;
++ unsigned long r14;
++ unsigned long r13;
++ unsigned long r12;
++ unsigned long rbp;
++ unsigned long rbx;
++/* arguments: non interrupts/non tracing syscalls only save upto here*/
++ unsigned long r11;
++ unsigned long r10;
++ unsigned long r9;
++ unsigned long r8;
++ unsigned long rax;
++ unsigned long rcx;
++ unsigned long rdx;
++ unsigned long rsi;
++ unsigned long rdi;
++ unsigned long orig_rax;
++/* end of arguments */
++/* cpu exception frame or undefined */
++ unsigned long rip;
++ unsigned long cs;
++ unsigned long eflags;
++ unsigned long rsp;
++ unsigned long ss;
++/* top of stack page */
++};
++
++#endif
++
++/* Arbitrarily choose the same ptrace numbers as used by the Sparc code. */
++#define PTRACE_GETREGS 12
++#define PTRACE_SETREGS 13
++#define PTRACE_GETFPREGS 14
++#define PTRACE_SETFPREGS 15
++#define PTRACE_GETFPXREGS 18
++#define PTRACE_SETFPXREGS 19
++
++/* only useful for access 32bit programs */
++#define PTRACE_GET_THREAD_AREA 25
++#define PTRACE_SET_THREAD_AREA 26
++
++#define PTRACE_ARCH_PRCTL 30 /* arch_prctl for child */
++
++#if defined(__KERNEL__) && !defined(__ASSEMBLY__)
++#define user_mode(regs) (!!((regs)->cs & 3))
++#define user_mode_vm(regs) user_mode(regs)
++#define instruction_pointer(regs) ((regs)->rip)
++#if defined(CONFIG_SMP) && defined(CONFIG_FRAME_POINTER)
++extern unsigned long profile_pc(struct pt_regs *regs);
++#else
++#define profile_pc(regs) instruction_pointer(regs)
++#endif
++
++void signal_fault(struct pt_regs *regs, void __user *frame, char *where);
++
++struct task_struct;
++
++extern unsigned long
++convert_rip_to_linear(struct task_struct *child, struct pt_regs *regs);
++
++enum {
++ EF_CF = 0x00000001,
++ EF_PF = 0x00000004,
++ EF_AF = 0x00000010,
++ EF_ZF = 0x00000040,
++ EF_SF = 0x00000080,
++ EF_TF = 0x00000100,
++ EF_IE = 0x00000200,
++ EF_DF = 0x00000400,
++ EF_OF = 0x00000800,
++ EF_IOPL = 0x00003000,
++ EF_IOPL_RING0 = 0x00000000,
++ EF_IOPL_RING1 = 0x00001000,
++ EF_IOPL_RING2 = 0x00002000,
++ EF_NT = 0x00004000, /* nested task */
++ EF_RF = 0x00010000, /* resume */
++ EF_VM = 0x00020000, /* virtual mode */
++ EF_AC = 0x00040000, /* alignment */
++ EF_VIF = 0x00080000, /* virtual interrupt */
++ EF_VIP = 0x00100000, /* virtual interrupt pending */
++ EF_ID = 0x00200000, /* id */
++};
++
++#endif
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/smp.h new/include/asm-x86_64/mach-xen/asm/smp.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/smp.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/smp.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,153 @@
++#ifndef __ASM_SMP_H
++#define __ASM_SMP_H
++
++/*
++ * We need the APIC definitions automatically as part of 'smp.h'
++ */
++#ifndef __ASSEMBLY__
++#include <linux/config.h>
++#include <linux/threads.h>
++#include <linux/cpumask.h>
++#include <linux/bitops.h>
++extern int disable_apic;
++#endif
++
++#ifdef CONFIG_X86_LOCAL_APIC
++#ifndef __ASSEMBLY__
++#include <asm/fixmap.h>
++#include <asm/mpspec.h>
++#ifdef CONFIG_X86_IO_APIC
++#include <asm/io_apic.h>
++#endif
++#include <asm/apic.h>
++#include <asm/thread_info.h>
++#endif
++#endif
++
++#ifdef CONFIG_SMP
++#ifndef ASSEMBLY
++
++#include <asm/pda.h>
++
++struct pt_regs;
++
++extern cpumask_t cpu_present_mask;
++extern cpumask_t cpu_possible_map;
++extern cpumask_t cpu_online_map;
++extern cpumask_t cpu_initialized;
++
++/*
++ * Private routines/data
++ */
++
++extern void smp_alloc_memory(void);
++extern volatile unsigned long smp_invalidate_needed;
++extern int pic_mode;
++extern void lock_ipi_call_lock(void);
++extern void unlock_ipi_call_lock(void);
++extern int smp_num_siblings;
++extern void smp_send_reschedule(int cpu);
++void smp_stop_cpu(void);
++extern int smp_call_function_single(int cpuid, void (*func) (void *info),
++ void *info, int retry, int wait);
++
++extern cpumask_t cpu_sibling_map[NR_CPUS];
++extern cpumask_t cpu_core_map[NR_CPUS];
++extern int phys_proc_id[NR_CPUS];
++extern int cpu_core_id[NR_CPUS];
++extern u8 cpu_llc_id[NR_CPUS];
++
++#define SMP_TRAMPOLINE_BASE 0x6000
++
++/*
++ * On x86 all CPUs are mapped 1:1 to the APIC space.
++ * This simplifies scheduling and IPI sending and
++ * compresses data structures.
++ */
++
++static inline int num_booting_cpus(void)
++{
++ return cpus_weight(cpu_possible_map);
++}
++
++#define raw_smp_processor_id() read_pda(cpunumber)
++
++#ifdef CONFIG_X86_LOCAL_APIC
++static inline int hard_smp_processor_id(void)
++{
++ /* we don't want to mark this access volatile - bad code generation */
++ return GET_APIC_ID(*(unsigned int *)(APIC_BASE+APIC_ID));
++}
++#endif
++
++extern int safe_smp_processor_id(void);
++extern int __cpu_disable(void);
++extern void __cpu_die(unsigned int cpu);
++extern void prefill_possible_map(void);
++extern unsigned num_processors;
++extern unsigned disabled_cpus;
++
++#endif /* !ASSEMBLY */
++
++#define NO_PROC_ID 0xFF /* No processor magic marker */
++
++#endif
++
++#ifndef ASSEMBLY
++/*
++ * Some lowlevel functions might want to know about
++ * the real APIC ID <-> CPU # mapping.
++ */
++extern u8 x86_cpu_to_apicid[NR_CPUS]; /* physical ID */
++extern u8 x86_cpu_to_log_apicid[NR_CPUS];
++extern u8 bios_cpu_apicid[];
++
++#ifdef CONFIG_X86_LOCAL_APIC
++static inline unsigned int cpu_mask_to_apicid(cpumask_t cpumask)
++{
++ return cpus_addr(cpumask)[0];
++}
++
++static inline int cpu_present_to_apicid(int mps_cpu)
++{
++ if (mps_cpu < NR_CPUS)
++ return (int)bios_cpu_apicid[mps_cpu];
++ else
++ return BAD_APICID;
++}
++#endif
++
++#endif /* !ASSEMBLY */
++
++#ifndef CONFIG_SMP
++#define stack_smp_processor_id() 0
++#define safe_smp_processor_id() 0
++#define cpu_logical_map(x) (x)
++#else
++#include <asm/thread_info.h>
++#define stack_smp_processor_id() \
++({ \
++ struct thread_info *ti; \
++ __asm__("andq %%rsp,%0; ":"=r" (ti) : "0" (CURRENT_MASK)); \
++ ti->cpu; \
++})
++#endif
++
++#ifndef __ASSEMBLY__
++#ifdef CONFIG_X86_LOCAL_APIC
++static __inline int logical_smp_processor_id(void)
++{
++ /* we don't want to mark this access volatile - bad code generation */
++ return GET_APIC_LOGICAL_ID(*(unsigned long *)(APIC_BASE+APIC_LDR));
++}
++#endif
++#endif
++
++#ifdef CONFIG_SMP
++#define cpu_physical_id(cpu) x86_cpu_to_apicid[cpu]
++#else
++#define cpu_physical_id(cpu) boot_cpu_id
++#endif
++
++#endif
++
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/synch_bitops.h new/include/asm-x86_64/mach-xen/asm/synch_bitops.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/synch_bitops.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/synch_bitops.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,2 @@
++
++#include <asm-i386/mach-xen/asm/synch_bitops.h>
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/system.h new/include/asm-x86_64/mach-xen/asm/system.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/system.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/system.h 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,428 @@
++#ifndef __ASM_SYSTEM_H
++#define __ASM_SYSTEM_H
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <asm/segment.h>
++#include <asm/synch_bitops.h>
++#include <asm/hypervisor.h>
++#include <xen/interface/arch-x86_64.h>
++
++#ifdef __KERNEL__
++
++#ifdef CONFIG_SMP
++#define __vcpu_id smp_processor_id()
++#else
++#define __vcpu_id 0
++#endif
++
++#ifdef CONFIG_SMP
++#define LOCK_PREFIX "lock ; "
++#else
++#define LOCK_PREFIX ""
++#endif
++
++#define __STR(x) #x
++#define STR(x) __STR(x)
++
++#define __SAVE(reg,offset) "movq %%" #reg ",(14-" #offset ")*8(%%rsp)\n\t"
++#define __RESTORE(reg,offset) "movq (14-" #offset ")*8(%%rsp),%%" #reg "\n\t"
++
++/* frame pointer must be last for get_wchan */
++#define SAVE_CONTEXT "pushq %%rbp ; movq %%rsi,%%rbp\n\t"
++#define RESTORE_CONTEXT "movq %%rbp,%%rsi ; popq %%rbp\n\t"
++
++#define __EXTRA_CLOBBER \
++ ,"rcx","rbx","rdx","r8","r9","r10","r11","r12","r13","r14","r15"
++
++#define switch_to(prev,next,last) \
++ asm volatile(SAVE_CONTEXT \
++ "movq %%rsp,%P[threadrsp](%[prev])\n\t" /* save RSP */ \
++ "movq %P[threadrsp](%[next]),%%rsp\n\t" /* restore RSP */ \
++ "call __switch_to\n\t" \
++ ".globl thread_return\n" \
++ "thread_return:\n\t" \
++ "movq %%gs:%P[pda_pcurrent],%%rsi\n\t" \
++ "movq %P[thread_info](%%rsi),%%r8\n\t" \
++ LOCK "btr %[tif_fork],%P[ti_flags](%%r8)\n\t" \
++ "movq %%rax,%%rdi\n\t" \
++ "jc ret_from_fork\n\t" \
++ RESTORE_CONTEXT \
++ : "=a" (last) \
++ : [next] "S" (next), [prev] "D" (prev), \
++ [threadrsp] "i" (offsetof(struct task_struct, thread.rsp)), \
++ [ti_flags] "i" (offsetof(struct thread_info, flags)),\
++ [tif_fork] "i" (TIF_FORK), \
++ [thread_info] "i" (offsetof(struct task_struct, thread_info)), \
++ [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent)) \
++ : "memory", "cc" __EXTRA_CLOBBER)
++
++
++extern void load_gs_index(unsigned);
++
++/*
++ * Load a segment. Fall back on loading the zero
++ * segment if something goes wrong..
++ */
++#define loadsegment(seg,value) \
++ asm volatile("\n" \
++ "1:\t" \
++ "movl %k0,%%" #seg "\n" \
++ "2:\n" \
++ ".section .fixup,\"ax\"\n" \
++ "3:\t" \
++ "movl %1,%%" #seg "\n\t" \
++ "jmp 2b\n" \
++ ".previous\n" \
++ ".section __ex_table,\"a\"\n\t" \
++ ".align 8\n\t" \
++ ".quad 1b,3b\n" \
++ ".previous" \
++ : :"r" (value), "r" (0))
++
++#ifdef __KERNEL__
++struct alt_instr {
++ __u8 *instr; /* original instruction */
++ __u8 *replacement;
++ __u8 cpuid; /* cpuid bit set for replacement */
++ __u8 instrlen; /* length of original instruction */
++ __u8 replacementlen; /* length of new instruction, <= instrlen */
++ __u8 pad[5];
++};
++#endif
++
++/*
++ * Alternative instructions for different CPU types or capabilities.
++ *
++ * This allows to use optimized instructions even on generic binary
++ * kernels.
++ *
++ * length of oldinstr must be longer or equal the length of newinstr
++ * It can be padded with nops as needed.
++ *
++ * For non barrier like inlines please define new variants
++ * without volatile and memory clobber.
++ */
++#define alternative(oldinstr, newinstr, feature) \
++ asm volatile ("661:\n\t" oldinstr "\n662:\n" \
++ ".section .altinstructions,\"a\"\n" \
++ " .align 8\n" \
++ " .quad 661b\n" /* label */ \
++ " .quad 663f\n" /* new instruction */ \
++ " .byte %c0\n" /* feature bit */ \
++ " .byte 662b-661b\n" /* sourcelen */ \
++ " .byte 664f-663f\n" /* replacementlen */ \
++ ".previous\n" \
++ ".section .altinstr_replacement,\"ax\"\n" \
++ "663:\n\t" newinstr "\n664:\n" /* replacement */ \
++ ".previous" :: "i" (feature) : "memory")
++
++/*
++ * Alternative inline assembly with input.
++ *
++ * Peculiarities:
++ * No memory clobber here.
++ * Argument numbers start with 1.
++ * Best is to use constraints that are fixed size (like (%1) ... "r")
++ * If you use variable sized constraints like "m" or "g" in the
++ * replacement make sure to pad to the worst case length.
++ */
++#define alternative_input(oldinstr, newinstr, feature, input...) \
++ asm volatile ("661:\n\t" oldinstr "\n662:\n" \
++ ".section .altinstructions,\"a\"\n" \
++ " .align 8\n" \
++ " .quad 661b\n" /* label */ \
++ " .quad 663f\n" /* new instruction */ \
++ " .byte %c0\n" /* feature bit */ \
++ " .byte 662b-661b\n" /* sourcelen */ \
++ " .byte 664f-663f\n" /* replacementlen */ \
++ ".previous\n" \
++ ".section .altinstr_replacement,\"ax\"\n" \
++ "663:\n\t" newinstr "\n664:\n" /* replacement */ \
++ ".previous" :: "i" (feature), ##input)
++
++/* Like alternative_input, but with a single output argument */
++#define alternative_io(oldinstr, newinstr, feature, output, input...) \
++ asm volatile ("661:\n\t" oldinstr "\n662:\n" \
++ ".section .altinstructions,\"a\"\n" \
++ " .align 8\n" \
++ " .quad 661b\n" /* label */ \
++ " .quad 663f\n" /* new instruction */ \
++ " .byte %c[feat]\n" /* feature bit */ \
++ " .byte 662b-661b\n" /* sourcelen */ \
++ " .byte 664f-663f\n" /* replacementlen */ \
++ ".previous\n" \
++ ".section .altinstr_replacement,\"ax\"\n" \
++ "663:\n\t" newinstr "\n664:\n" /* replacement */ \
++ ".previous" : output : [feat] "i" (feature), ##input)
++
++/*
++ * Clear and set 'TS' bit respectively
++ */
++#define clts() (HYPERVISOR_fpu_taskswitch(0))
++
++static inline unsigned long read_cr0(void)
++{
++ unsigned long cr0;
++ asm volatile("movq %%cr0,%0" : "=r" (cr0));
++ return cr0;
++}
++
++static inline void write_cr0(unsigned long val)
++{
++ asm volatile("movq %0,%%cr0" :: "r" (val));
++}
++
++#define read_cr3() ({ \
++ unsigned long __dummy; \
++ asm("movq %%cr3,%0" : "=r" (__dummy)); \
++ machine_to_phys(__dummy); \
++})
++
++static inline unsigned long read_cr4(void)
++{
++ unsigned long cr4;
++ asm("movq %%cr4,%0" : "=r" (cr4));
++ return cr4;
++}
++
++static inline void write_cr4(unsigned long val)
++{
++ asm volatile("movq %0,%%cr4" :: "r" (val));
++}
++
++#define stts() (HYPERVISOR_fpu_taskswitch(1))
++
++#define wbinvd() \
++ __asm__ __volatile__ ("wbinvd": : :"memory");
++
++/*
++ * On SMP systems, when the scheduler does migration-cost autodetection,
++ * it needs a way to flush as much of the CPU's caches as possible.
++ */
++static inline void sched_cacheflush(void)
++{
++ wbinvd();
++}
++
++#endif /* __KERNEL__ */
++
++#define nop() __asm__ __volatile__ ("nop")
++
++#define xchg(ptr,v) ((__typeof__(*(ptr)))__xchg((unsigned long)(v),(ptr),sizeof(*(ptr))))
++
++#define tas(ptr) (xchg((ptr),1))
++
++#define __xg(x) ((volatile long *)(x))
++
++static inline void set_64bit(volatile unsigned long *ptr, unsigned long val)
++{
++ *ptr = val;
++}
++
++#define _set_64bit set_64bit
++
++/*
++ * Note: no "lock" prefix even on SMP: xchg always implies lock anyway
++ * Note 2: xchg has side effect, so that attribute volatile is necessary,
++ * but generally the primitive is invalid, *ptr is output argument. --ANK
++ */
++static inline unsigned long __xchg(unsigned long x, volatile void * ptr, int size)
++{
++ switch (size) {
++ case 1:
++ __asm__ __volatile__("xchgb %b0,%1"
++ :"=q" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ case 2:
++ __asm__ __volatile__("xchgw %w0,%1"
++ :"=r" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ case 4:
++ __asm__ __volatile__("xchgl %k0,%1"
++ :"=r" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ case 8:
++ __asm__ __volatile__("xchgq %0,%1"
++ :"=r" (x)
++ :"m" (*__xg(ptr)), "0" (x)
++ :"memory");
++ break;
++ }
++ return x;
++}
++
++/*
++ * Atomic compare and exchange. Compare OLD with MEM, if identical,
++ * store NEW in MEM. Return the initial value in MEM. Success is
++ * indicated by comparing RETURN with OLD.
++ */
++
++#define __HAVE_ARCH_CMPXCHG 1
++
++static inline unsigned long __cmpxchg(volatile void *ptr, unsigned long old,
++ unsigned long new, int size)
++{
++ unsigned long prev;
++ switch (size) {
++ case 1:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgb %b1,%2"
++ : "=a"(prev)
++ : "q"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ case 2:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgw %w1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ case 4:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgl %k1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ case 8:
++ __asm__ __volatile__(LOCK_PREFIX "cmpxchgq %1,%2"
++ : "=a"(prev)
++ : "r"(new), "m"(*__xg(ptr)), "0"(old)
++ : "memory");
++ return prev;
++ }
++ return old;
++}
++
++#define cmpxchg(ptr,o,n)\
++ ((__typeof__(*(ptr)))__cmpxchg((ptr),(unsigned long)(o),\
++ (unsigned long)(n),sizeof(*(ptr))))
++
++#ifdef CONFIG_SMP
++#define smp_mb() mb()
++#define smp_rmb() rmb()
++#define smp_wmb() wmb()
++#define smp_read_barrier_depends() do {} while(0)
++#else
++#define smp_mb() barrier()
++#define smp_rmb() barrier()
++#define smp_wmb() barrier()
++#define smp_read_barrier_depends() do {} while(0)
++#endif
++
++
++/*
++ * Force strict CPU ordering.
++ * And yes, this is required on UP too when we're talking
++ * to devices.
++ */
++#define mb() asm volatile("mfence":::"memory")
++#define rmb() asm volatile("lfence":::"memory")
++
++#ifdef CONFIG_UNORDERED_IO
++#define wmb() asm volatile("sfence" ::: "memory")
++#else
++#define wmb() asm volatile("" ::: "memory")
++#endif
++#define read_barrier_depends() do {} while(0)
++#define set_mb(var, value) do { (void) xchg(&var, value); } while (0)
++#define set_wmb(var, value) do { var = value; wmb(); } while (0)
++
++#define warn_if_not_ulong(x) do { unsigned long foo; (void) (&(x) == &foo); } while (0)
++
++
++/*
++ * The use of 'barrier' in the following reflects their use as local-lock
++ * operations. Reentrancy must be prevented (e.g., __cli()) /before/ following
++ * critical operations are executed. All critical operations must complete
++ * /before/ reentrancy is permitted (e.g., __sti()). Alpha architecture also
++ * includes these barriers, for example.
++ */
++
++#define __cli() \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ _vcpu->evtchn_upcall_mask = 1; \
++ preempt_enable_no_resched(); \
++ barrier(); \
++} while (0)
++
++#define __sti() \
++do { \
++ vcpu_info_t *_vcpu; \
++ barrier(); \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ _vcpu->evtchn_upcall_mask = 0; \
++ barrier(); /* unmask then check (avoid races) */ \
++ if ( unlikely(_vcpu->evtchn_upcall_pending) ) \
++ force_evtchn_callback(); \
++ preempt_enable(); \
++} while (0)
++
++#define __save_flags(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ (x) = _vcpu->evtchn_upcall_mask; \
++ preempt_enable(); \
++} while (0)
++
++#define __restore_flags(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ barrier(); \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ if ((_vcpu->evtchn_upcall_mask = (x)) == 0) { \
++ barrier(); /* unmask then check (avoid races) */ \
++ if ( unlikely(_vcpu->evtchn_upcall_pending) ) \
++ force_evtchn_callback(); \
++ preempt_enable(); \
++ } else \
++ preempt_enable_no_resched(); \
++} while (0)
++
++#define __save_and_cli(x) \
++do { \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ (x) = _vcpu->evtchn_upcall_mask; \
++ _vcpu->evtchn_upcall_mask = 1; \
++ preempt_enable_no_resched(); \
++ barrier(); \
++} while (0)
++
++#define local_irq_save(x) __save_and_cli(x)
++#define local_irq_restore(x) __restore_flags(x)
++#define local_save_flags(x) __save_flags(x)
++#define local_irq_disable() __cli()
++#define local_irq_enable() __sti()
++
++/* Cannot use preempt_enable() here as we would recurse in preempt_sched(). */
++#define irqs_disabled() \
++({ int ___x; \
++ vcpu_info_t *_vcpu; \
++ preempt_disable(); \
++ _vcpu = &HYPERVISOR_shared_info->vcpu_info[__vcpu_id]; \
++ ___x = (_vcpu->evtchn_upcall_mask != 0); \
++ preempt_enable_no_resched(); \
++ ___x; })
++
++void safe_halt(void);
++void halt(void);
++
++void cpu_idle_wait(void);
++
++extern unsigned long arch_align_stack(unsigned long sp);
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/timer.h new/include/asm-x86_64/mach-xen/asm/timer.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/timer.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/timer.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,67 @@
++#ifndef _ASMi386_TIMER_H
++#define _ASMi386_TIMER_H
++#include <linux/init.h>
++
++/**
++ * struct timer_ops - used to define a timer source
++ *
++ * @name: name of the timer.
++ * @init: Probes and initializes the timer. Takes clock= override
++ * string as an argument. Returns 0 on success, anything else
++ * on failure.
++ * @mark_offset: called by the timer interrupt.
++ * @get_offset: called by gettimeofday(). Returns the number of microseconds
++ * since the last timer interupt.
++ * @monotonic_clock: returns the number of nanoseconds since the init of the
++ * timer.
++ * @delay: delays this many clock cycles.
++ */
++struct timer_opts {
++ char* name;
++ void (*mark_offset)(void);
++ unsigned long (*get_offset)(void);
++ unsigned long long (*monotonic_clock)(void);
++ void (*delay)(unsigned long);
++ unsigned long (*read_timer)(void);
++ int (*suspend)(pm_message_t state);
++ int (*resume)(void);
++};
++
++struct init_timer_opts {
++ int (*init)(char *override);
++ struct timer_opts *opts;
++};
++
++#define TICK_SIZE (tick_nsec / 1000)
++
++extern struct timer_opts* __init select_timer(void);
++extern void clock_fallback(void);
++void setup_pit_timer(void);
++
++/* Modifiers for buggy PIT handling */
++
++extern int pit_latch_buggy;
++
++extern struct timer_opts *cur_timer;
++extern int timer_ack;
++
++/* list of externed timers */
++extern struct timer_opts timer_none;
++extern struct timer_opts timer_pit;
++extern struct init_timer_opts timer_pit_init;
++extern struct init_timer_opts timer_tsc_init;
++#ifdef CONFIG_X86_CYCLONE_TIMER
++extern struct init_timer_opts timer_cyclone_init;
++#endif
++
++extern unsigned long calibrate_tsc(void);
++extern void init_cpu_khz(void);
++#ifdef CONFIG_HPET_TIMER
++extern struct init_timer_opts timer_hpet_init;
++extern unsigned long calibrate_tsc_hpet(unsigned long *tsc_hpet_quotient_ptr);
++#endif
++
++#ifdef CONFIG_X86_PM_TIMER
++extern struct init_timer_opts timer_pmtmr_init;
++#endif
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/tlbflush.h new/include/asm-x86_64/mach-xen/asm/tlbflush.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/tlbflush.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/tlbflush.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,104 @@
++#ifndef _X8664_TLBFLUSH_H
++#define _X8664_TLBFLUSH_H
++
++#include <linux/config.h>
++#include <linux/mm.h>
++#include <asm/processor.h>
++
++#define __flush_tlb() xen_tlb_flush()
++
++/*
++ * Global pages have to be flushed a bit differently. Not a real
++ * performance problem because this does not happen often.
++ */
++#define __flush_tlb_global() xen_tlb_flush()
++
++
++extern unsigned long pgkern_mask;
++
++#define __flush_tlb_all() __flush_tlb_global()
++
++#define __flush_tlb_one(addr) xen_invlpg((unsigned long)addr)
++
++
++/*
++ * TLB flushing:
++ *
++ * - flush_tlb() flushes the current mm struct TLBs
++ * - flush_tlb_all() flushes all processes TLBs
++ * - flush_tlb_mm(mm) flushes the specified mm context TLB's
++ * - flush_tlb_page(vma, vmaddr) flushes one page
++ * - flush_tlb_range(vma, start, end) flushes a range of pages
++ * - flush_tlb_kernel_range(start, end) flushes a range of kernel pages
++ * - flush_tlb_pgtables(mm, start, end) flushes a range of page tables
++ *
++ * x86-64 can only flush individual pages or full VMs. For a range flush
++ * we always do the full VM. Might be worth trying if for a small
++ * range a few INVLPGs in a row are a win.
++ */
++
++#ifndef CONFIG_SMP
++
++#define flush_tlb() __flush_tlb()
++#define flush_tlb_all() __flush_tlb_all()
++#define local_flush_tlb() __flush_tlb()
++
++static inline void flush_tlb_mm(struct mm_struct *mm)
++{
++ if (mm == current->active_mm)
++ __flush_tlb();
++}
++
++static inline void flush_tlb_page(struct vm_area_struct *vma,
++ unsigned long addr)
++{
++ if (vma->vm_mm == current->active_mm)
++ __flush_tlb_one(addr);
++}
++
++static inline void flush_tlb_range(struct vm_area_struct *vma,
++ unsigned long start, unsigned long end)
++{
++ if (vma->vm_mm == current->active_mm)
++ __flush_tlb();
++}
++
++#else
++
++#include <asm/smp.h>
++
++#define local_flush_tlb() \
++ __flush_tlb()
++
++extern void flush_tlb_all(void);
++extern void flush_tlb_current_task(void);
++extern void flush_tlb_mm(struct mm_struct *);
++extern void flush_tlb_page(struct vm_area_struct *, unsigned long);
++
++#define flush_tlb() flush_tlb_current_task()
++
++static inline void flush_tlb_range(struct vm_area_struct * vma, unsigned long start, unsigned long end)
++{
++ flush_tlb_mm(vma->vm_mm);
++}
++
++#define TLBSTATE_OK 1
++#define TLBSTATE_LAZY 2
++
++/* Roughly an IPI every 20MB with 4k pages for freeing page table
++ ranges. Cost is about 42k of memory for each CPU. */
++#define ARCH_FREE_PTE_NR 5350
++
++#endif
++
++#define flush_tlb_kernel_range(start, end) flush_tlb_all()
++
++static inline void flush_tlb_pgtables(struct mm_struct *mm,
++ unsigned long start, unsigned long end)
++{
++ /* x86_64 does not keep any page table caches in a software TLB.
++ The CPUs do in their hardware TLBs, but they are handled
++ by the normal TLB flushing algorithms. */
++}
++
++#endif /* _X8664_TLBFLUSH_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/vga.h new/include/asm-x86_64/mach-xen/asm/vga.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/vga.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/vga.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,20 @@
++/*
++ * Access to VGA videoram
++ *
++ * (c) 1998 Martin Mares <mj@ucw.cz>
++ */
++
++#ifndef _LINUX_ASM_VGA_H_
++#define _LINUX_ASM_VGA_H_
++
++/*
++ * On the PC, we can just recalculate addresses and then
++ * access the videoram directly without any black magic.
++ */
++
++#define VGA_MAP_MEM(x) (unsigned long)isa_bus_to_virt(x)
++
++#define vga_readb(x) (*(x))
++#define vga_writeb(x,y) (*(y) = (x))
++
++#endif
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/asm/xor.h new/include/asm-x86_64/mach-xen/asm/xor.h
+--- linux-2.6/include/asm-x86_64/mach-xen/asm/xor.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/asm/xor.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,328 @@
++/*
++ * x86-64 changes / gcc fixes from Andi Kleen.
++ * Copyright 2002 Andi Kleen, SuSE Labs.
++ *
++ * This hasn't been optimized for the hammer yet, but there are likely
++ * no advantages to be gotten from x86-64 here anyways.
++ */
++
++typedef struct { unsigned long a,b; } __attribute__((aligned(16))) xmm_store_t;
++
++/* Doesn't use gcc to save the XMM registers, because there is no easy way to
++ tell it to do a clts before the register saving. */
++#define XMMS_SAVE do { \
++ preempt_disable(); \
++ if (!(current_thread_info()->status & TS_USEDFPU)) \
++ clts(); \
++ __asm__ __volatile__ ( \
++ "movups %%xmm0,(%1) ;\n\t" \
++ "movups %%xmm1,0x10(%1) ;\n\t" \
++ "movups %%xmm2,0x20(%1) ;\n\t" \
++ "movups %%xmm3,0x30(%1) ;\n\t" \
++ : "=&r" (cr0) \
++ : "r" (xmm_save) \
++ : "memory"); \
++} while(0)
++
++#define XMMS_RESTORE do { \
++ asm volatile ( \
++ "sfence ;\n\t" \
++ "movups (%1),%%xmm0 ;\n\t" \
++ "movups 0x10(%1),%%xmm1 ;\n\t" \
++ "movups 0x20(%1),%%xmm2 ;\n\t" \
++ "movups 0x30(%1),%%xmm3 ;\n\t" \
++ : \
++ : "r" (cr0), "r" (xmm_save) \
++ : "memory"); \
++ if (!(current_thread_info()->status & TS_USEDFPU)) \
++ stts(); \
++ preempt_enable(); \
++} while(0)
++
++#define OFFS(x) "16*("#x")"
++#define PF_OFFS(x) "256+16*("#x")"
++#define PF0(x) " prefetchnta "PF_OFFS(x)"(%[p1]) ;\n"
++#define LD(x,y) " movaps "OFFS(x)"(%[p1]), %%xmm"#y" ;\n"
++#define ST(x,y) " movaps %%xmm"#y", "OFFS(x)"(%[p1]) ;\n"
++#define PF1(x) " prefetchnta "PF_OFFS(x)"(%[p2]) ;\n"
++#define PF2(x) " prefetchnta "PF_OFFS(x)"(%[p3]) ;\n"
++#define PF3(x) " prefetchnta "PF_OFFS(x)"(%[p4]) ;\n"
++#define PF4(x) " prefetchnta "PF_OFFS(x)"(%[p5]) ;\n"
++#define PF5(x) " prefetchnta "PF_OFFS(x)"(%[p6]) ;\n"
++#define XO1(x,y) " xorps "OFFS(x)"(%[p2]), %%xmm"#y" ;\n"
++#define XO2(x,y) " xorps "OFFS(x)"(%[p3]), %%xmm"#y" ;\n"
++#define XO3(x,y) " xorps "OFFS(x)"(%[p4]), %%xmm"#y" ;\n"
++#define XO4(x,y) " xorps "OFFS(x)"(%[p5]), %%xmm"#y" ;\n"
++#define XO5(x,y) " xorps "OFFS(x)"(%[p6]), %%xmm"#y" ;\n"
++
++
++static void
++xor_sse_2(unsigned long bytes, unsigned long *p1, unsigned long *p2)
++{
++ unsigned int lines = bytes >> 8;
++ unsigned long cr0;
++ xmm_store_t xmm_save[4];
++
++ XMMS_SAVE;
++
++ asm volatile (
++#undef BLOCK
++#define BLOCK(i) \
++ LD(i,0) \
++ LD(i+1,1) \
++ PF1(i) \
++ PF1(i+2) \
++ LD(i+2,2) \
++ LD(i+3,3) \
++ PF0(i+4) \
++ PF0(i+6) \
++ XO1(i,0) \
++ XO1(i+1,1) \
++ XO1(i+2,2) \
++ XO1(i+3,3) \
++ ST(i,0) \
++ ST(i+1,1) \
++ ST(i+2,2) \
++ ST(i+3,3) \
++
++
++ PF0(0)
++ PF0(2)
++
++ " .align 32 ;\n"
++ " 1: ;\n"
++
++ BLOCK(0)
++ BLOCK(4)
++ BLOCK(8)
++ BLOCK(12)
++
++ " addq %[inc], %[p1] ;\n"
++ " addq %[inc], %[p2] ;\n"
++ " decl %[cnt] ; jnz 1b"
++ : [p1] "+r" (p1), [p2] "+r" (p2), [cnt] "+r" (lines)
++ : [inc] "r" (256UL)
++ : "memory");
++
++ XMMS_RESTORE;
++}
++
++static void
++xor_sse_3(unsigned long bytes, unsigned long *p1, unsigned long *p2,
++ unsigned long *p3)
++{
++ unsigned int lines = bytes >> 8;
++ xmm_store_t xmm_save[4];
++ unsigned long cr0;
++
++ XMMS_SAVE;
++
++ __asm__ __volatile__ (
++#undef BLOCK
++#define BLOCK(i) \
++ PF1(i) \
++ PF1(i+2) \
++ LD(i,0) \
++ LD(i+1,1) \
++ LD(i+2,2) \
++ LD(i+3,3) \
++ PF2(i) \
++ PF2(i+2) \
++ PF0(i+4) \
++ PF0(i+6) \
++ XO1(i,0) \
++ XO1(i+1,1) \
++ XO1(i+2,2) \
++ XO1(i+3,3) \
++ XO2(i,0) \
++ XO2(i+1,1) \
++ XO2(i+2,2) \
++ XO2(i+3,3) \
++ ST(i,0) \
++ ST(i+1,1) \
++ ST(i+2,2) \
++ ST(i+3,3) \
++
++
++ PF0(0)
++ PF0(2)
++
++ " .align 32 ;\n"
++ " 1: ;\n"
++
++ BLOCK(0)
++ BLOCK(4)
++ BLOCK(8)
++ BLOCK(12)
++
++ " addq %[inc], %[p1] ;\n"
++ " addq %[inc], %[p2] ;\n"
++ " addq %[inc], %[p3] ;\n"
++ " decl %[cnt] ; jnz 1b"
++ : [cnt] "+r" (lines),
++ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3)
++ : [inc] "r" (256UL)
++ : "memory");
++ XMMS_RESTORE;
++}
++
++static void
++xor_sse_4(unsigned long bytes, unsigned long *p1, unsigned long *p2,
++ unsigned long *p3, unsigned long *p4)
++{
++ unsigned int lines = bytes >> 8;
++ xmm_store_t xmm_save[4];
++ unsigned long cr0;
++
++ XMMS_SAVE;
++
++ __asm__ __volatile__ (
++#undef BLOCK
++#define BLOCK(i) \
++ PF1(i) \
++ PF1(i+2) \
++ LD(i,0) \
++ LD(i+1,1) \
++ LD(i+2,2) \
++ LD(i+3,3) \
++ PF2(i) \
++ PF2(i+2) \
++ XO1(i,0) \
++ XO1(i+1,1) \
++ XO1(i+2,2) \
++ XO1(i+3,3) \
++ PF3(i) \
++ PF3(i+2) \
++ PF0(i+4) \
++ PF0(i+6) \
++ XO2(i,0) \
++ XO2(i+1,1) \
++ XO2(i+2,2) \
++ XO2(i+3,3) \
++ XO3(i,0) \
++ XO3(i+1,1) \
++ XO3(i+2,2) \
++ XO3(i+3,3) \
++ ST(i,0) \
++ ST(i+1,1) \
++ ST(i+2,2) \
++ ST(i+3,3) \
++
++
++ PF0(0)
++ PF0(2)
++
++ " .align 32 ;\n"
++ " 1: ;\n"
++
++ BLOCK(0)
++ BLOCK(4)
++ BLOCK(8)
++ BLOCK(12)
++
++ " addq %[inc], %[p1] ;\n"
++ " addq %[inc], %[p2] ;\n"
++ " addq %[inc], %[p3] ;\n"
++ " addq %[inc], %[p4] ;\n"
++ " decl %[cnt] ; jnz 1b"
++ : [cnt] "+c" (lines),
++ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4)
++ : [inc] "r" (256UL)
++ : "memory" );
++
++ XMMS_RESTORE;
++}
++
++static void
++xor_sse_5(unsigned long bytes, unsigned long *p1, unsigned long *p2,
++ unsigned long *p3, unsigned long *p4, unsigned long *p5)
++{
++ unsigned int lines = bytes >> 8;
++ xmm_store_t xmm_save[4];
++ unsigned long cr0;
++
++ XMMS_SAVE;
++
++ __asm__ __volatile__ (
++#undef BLOCK
++#define BLOCK(i) \
++ PF1(i) \
++ PF1(i+2) \
++ LD(i,0) \
++ LD(i+1,1) \
++ LD(i+2,2) \
++ LD(i+3,3) \
++ PF2(i) \
++ PF2(i+2) \
++ XO1(i,0) \
++ XO1(i+1,1) \
++ XO1(i+2,2) \
++ XO1(i+3,3) \
++ PF3(i) \
++ PF3(i+2) \
++ XO2(i,0) \
++ XO2(i+1,1) \
++ XO2(i+2,2) \
++ XO2(i+3,3) \
++ PF4(i) \
++ PF4(i+2) \
++ PF0(i+4) \
++ PF0(i+6) \
++ XO3(i,0) \
++ XO3(i+1,1) \
++ XO3(i+2,2) \
++ XO3(i+3,3) \
++ XO4(i,0) \
++ XO4(i+1,1) \
++ XO4(i+2,2) \
++ XO4(i+3,3) \
++ ST(i,0) \
++ ST(i+1,1) \
++ ST(i+2,2) \
++ ST(i+3,3) \
++
++
++ PF0(0)
++ PF0(2)
++
++ " .align 32 ;\n"
++ " 1: ;\n"
++
++ BLOCK(0)
++ BLOCK(4)
++ BLOCK(8)
++ BLOCK(12)
++
++ " addq %[inc], %[p1] ;\n"
++ " addq %[inc], %[p2] ;\n"
++ " addq %[inc], %[p3] ;\n"
++ " addq %[inc], %[p4] ;\n"
++ " addq %[inc], %[p5] ;\n"
++ " decl %[cnt] ; jnz 1b"
++ : [cnt] "+c" (lines),
++ [p1] "+r" (p1), [p2] "+r" (p2), [p3] "+r" (p3), [p4] "+r" (p4),
++ [p5] "+r" (p5)
++ : [inc] "r" (256UL)
++ : "memory");
++
++ XMMS_RESTORE;
++}
++
++static struct xor_block_template xor_block_sse = {
++ .name = "generic_sse",
++ .do_2 = xor_sse_2,
++ .do_3 = xor_sse_3,
++ .do_4 = xor_sse_4,
++ .do_5 = xor_sse_5,
++};
++
++#undef XOR_TRY_TEMPLATES
++#define XOR_TRY_TEMPLATES \
++ do { \
++ xor_speed(&xor_block_sse); \
++ } while (0)
++
++/* We force the use of the SSE xor block because it can write around L2.
++ We may also be able to load into the L1 only depending on how the cpu
++ deals with a load to a line that is being prefetched. */
++#define XOR_SELECT_TEMPLATE(FASTEST) (&xor_block_sse)
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/irq_vectors.h new/include/asm-x86_64/mach-xen/irq_vectors.h
+--- linux-2.6/include/asm-x86_64/mach-xen/irq_vectors.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/irq_vectors.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,123 @@
++/*
++ * This file should contain #defines for all of the interrupt vector
++ * numbers used by this architecture.
++ *
++ * In addition, there are some standard defines:
++ *
++ * FIRST_EXTERNAL_VECTOR:
++ * The first free place for external interrupts
++ *
++ * SYSCALL_VECTOR:
++ * The IRQ vector a syscall makes the user to kernel transition
++ * under.
++ *
++ * TIMER_IRQ:
++ * The IRQ number the timer interrupt comes in at.
++ *
++ * NR_IRQS:
++ * The total number of interrupt vectors (including all the
++ * architecture specific interrupts) needed.
++ *
++ */
++#ifndef _ASM_IRQ_VECTORS_H
++#define _ASM_IRQ_VECTORS_H
++
++/*
++ * IDT vectors usable for external interrupt sources start
++ * at 0x20:
++ */
++#define FIRST_EXTERNAL_VECTOR 0x20
++
++#define SYSCALL_VECTOR 0x80
++
++/*
++ * Vectors 0x20-0x2f are used for ISA interrupts.
++ */
++
++#if 0
++/*
++ * Special IRQ vectors used by the SMP architecture, 0xf0-0xff
++ *
++ * some of the following vectors are 'rare', they are merged
++ * into a single vector (CALL_FUNCTION_VECTOR) to save vector space.
++ * TLB, reschedule and local APIC vectors are performance-critical.
++ *
++ * Vectors 0xf0-0xfa are free (reserved for future Linux use).
++ */
++#define INVALIDATE_TLB_VECTOR 0xfd
++#define RESCHEDULE_VECTOR 0xfc
++#define CALL_FUNCTION_VECTOR 0xfb
++
++#define THERMAL_APIC_VECTOR 0xf0
++/*
++ * Local APIC timer IRQ vector is on a different priority level,
++ * to work around the 'lost local interrupt if more than 2 IRQ
++ * sources per level' errata.
++ */
++#define LOCAL_TIMER_VECTOR 0xef
++#endif
++
++#define SPURIOUS_APIC_VECTOR 0xff
++#define ERROR_APIC_VECTOR 0xfe
++
++/*
++ * First APIC vector available to drivers: (vectors 0x30-0xee)
++ * we start at 0x31 to spread out vectors evenly between priority
++ * levels. (0x80 is the syscall vector)
++ */
++#define FIRST_DEVICE_VECTOR 0x31
++#define FIRST_SYSTEM_VECTOR 0xef
++
++/*
++ * 16 8259A IRQ's, 208 potential APIC interrupt sources.
++ * Right now the APIC is mostly only used for SMP.
++ * 256 vectors is an architectural limit. (we can have
++ * more than 256 devices theoretically, but they will
++ * have to use shared interrupts)
++ * Since vectors 0x00-0x1f are used/reserved for the CPU,
++ * the usable vector space is 0x20-0xff (224 vectors)
++ */
++
++#define RESCHEDULE_VECTOR 0
++#define CALL_FUNCTION_VECTOR 1
++#define NR_IPIS 2
++
++/*
++ * The maximum number of vectors supported by i386 processors
++ * is limited to 256. For processors other than i386, NR_VECTORS
++ * should be changed accordingly.
++ */
++#define NR_VECTORS 256
++
++#define FPU_IRQ 13
++
++#define FIRST_VM86_IRQ 3
++#define LAST_VM86_IRQ 15
++#define invalid_vm86_irq(irq) ((irq) < 3 || (irq) > 15)
++
++/*
++ * The flat IRQ space is divided into two regions:
++ * 1. A one-to-one mapping of real physical IRQs. This space is only used
++ * if we have physical device-access privilege. This region is at the
++ * start of the IRQ space so that existing device drivers do not need
++ * to be modified to translate physical IRQ numbers into our IRQ space.
++ * 3. A dynamic mapping of inter-domain and Xen-sourced virtual IRQs. These
++ * are bound using the provided bind/unbind functions.
++ */
++
++#define PIRQ_BASE 0
++#define NR_PIRQS 256
++
++#define DYNIRQ_BASE (PIRQ_BASE + NR_PIRQS)
++#define NR_DYNIRQS 256
++
++#define NR_IRQS (NR_PIRQS + NR_DYNIRQS)
++#define NR_IRQ_VECTORS NR_IRQS
++
++#define pirq_to_irq(_x) ((_x) + PIRQ_BASE)
++#define irq_to_pirq(_x) ((_x) - PIRQ_BASE)
++
++#define dynirq_to_irq(_x) ((_x) + DYNIRQ_BASE)
++#define irq_to_dynirq(_x) ((_x) - DYNIRQ_BASE)
++
++#endif /* _ASM_IRQ_VECTORS_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/mach_time.h new/include/asm-x86_64/mach-xen/mach_time.h
+--- linux-2.6/include/asm-x86_64/mach-xen/mach_time.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/mach_time.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,111 @@
++/*
++ * include/asm-i386/mach-default/mach_time.h
++ *
++ * Machine specific set RTC function for generic.
++ * Split out from time.c by Osamu Tomita <tomita@cinet.co.jp>
++ */
++#ifndef _MACH_TIME_H
++#define _MACH_TIME_H
++
++#include <asm-i386/mc146818rtc.h>
++
++/* for check timing call set_rtc_mmss() 500ms */
++/* used in arch/i386/time.c::do_timer_interrupt() */
++#define USEC_AFTER 500000
++#define USEC_BEFORE 500000
++
++/*
++ * In order to set the CMOS clock precisely, set_rtc_mmss has to be
++ * called 500 ms after the second nowtime has started, because when
++ * nowtime is written into the registers of the CMOS clock, it will
++ * jump to the next second precisely 500 ms later. Check the Motorola
++ * MC146818A or Dallas DS12887 data sheet for details.
++ *
++ * BUG: This routine does not handle hour overflow properly; it just
++ * sets the minutes. Usually you'll only notice that after reboot!
++ */
++static inline int mach_set_rtc_mmss(unsigned long nowtime)
++{
++ int retval = 0;
++ int real_seconds, real_minutes, cmos_minutes;
++ unsigned char save_control, save_freq_select;
++
++ save_control = CMOS_READ(RTC_CONTROL); /* tell the clock it's being set */
++ CMOS_WRITE((save_control|RTC_SET), RTC_CONTROL);
++
++ save_freq_select = CMOS_READ(RTC_FREQ_SELECT); /* stop and reset prescaler */
++ CMOS_WRITE((save_freq_select|RTC_DIV_RESET2), RTC_FREQ_SELECT);
++
++ cmos_minutes = CMOS_READ(RTC_MINUTES);
++ if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD)
++ BCD_TO_BIN(cmos_minutes);
++
++ /*
++ * since we're only adjusting minutes and seconds,
++ * don't interfere with hour overflow. This avoids
++ * messing with unknown time zones but requires your
++ * RTC not to be off by more than 15 minutes
++ */
++ real_seconds = nowtime % 60;
++ real_minutes = nowtime / 60;
++ if (((abs(real_minutes - cmos_minutes) + 15)/30) & 1)
++ real_minutes += 30; /* correct for half hour time zone */
++ real_minutes %= 60;
++
++ if (abs(real_minutes - cmos_minutes) < 30) {
++ if (!(save_control & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
++ BIN_TO_BCD(real_seconds);
++ BIN_TO_BCD(real_minutes);
++ }
++ CMOS_WRITE(real_seconds,RTC_SECONDS);
++ CMOS_WRITE(real_minutes,RTC_MINUTES);
++ } else {
++ printk(KERN_WARNING
++ "set_rtc_mmss: can't update from %d to %d\n",
++ cmos_minutes, real_minutes);
++ retval = -1;
++ }
++
++ /* The following flags have to be released exactly in this order,
++ * otherwise the DS12887 (popular MC146818A clone with integrated
++ * battery and quartz) will not reset the oscillator and will not
++ * update precisely 500 ms later. You won't find this mentioned in
++ * the Dallas Semiconductor data sheets, but who believes data
++ * sheets anyway ... -- Markus Kuhn
++ */
++ CMOS_WRITE(save_control, RTC_CONTROL);
++ CMOS_WRITE(save_freq_select, RTC_FREQ_SELECT);
++
++ return retval;
++}
++
++static inline unsigned long mach_get_cmos_time(void)
++{
++ unsigned int year, mon, day, hour, min, sec;
++
++ do {
++ sec = CMOS_READ(RTC_SECONDS);
++ min = CMOS_READ(RTC_MINUTES);
++ hour = CMOS_READ(RTC_HOURS);
++ day = CMOS_READ(RTC_DAY_OF_MONTH);
++ mon = CMOS_READ(RTC_MONTH);
++ year = CMOS_READ(RTC_YEAR);
++ } while (sec != CMOS_READ(RTC_SECONDS));
++
++ if (!(CMOS_READ(RTC_CONTROL) & RTC_DM_BINARY) || RTC_ALWAYS_BCD) {
++ BCD_TO_BIN(sec);
++ BCD_TO_BIN(min);
++ BCD_TO_BIN(hour);
++ BCD_TO_BIN(day);
++ BCD_TO_BIN(mon);
++ BCD_TO_BIN(year);
++ }
++
++ year += 1900;
++ if (year < 1970)
++ year += 100;
++
++ return mktime(year, mon, day, hour, min, sec);
++}
++
++#endif /* !_MACH_TIME_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/mach_timer.h new/include/asm-x86_64/mach-xen/mach_timer.h
+--- linux-2.6/include/asm-x86_64/mach-xen/mach_timer.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/mach_timer.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,48 @@
++/*
++ * include/asm-i386/mach-default/mach_timer.h
++ *
++ * Machine specific calibrate_tsc() for generic.
++ * Split out from timer_tsc.c by Osamu Tomita <tomita@cinet.co.jp>
++ */
++/* ------ Calibrate the TSC -------
++ * Return 2^32 * (1 / (TSC clocks per usec)) for do_fast_gettimeoffset().
++ * Too much 64-bit arithmetic here to do this cleanly in C, and for
++ * accuracy's sake we want to keep the overhead on the CTC speaker (channel 2)
++ * output busy loop as low as possible. We avoid reading the CTC registers
++ * directly because of the awkward 8-bit access mechanism of the 82C54
++ * device.
++ */
++#ifndef _MACH_TIMER_H
++#define _MACH_TIMER_H
++
++#define CALIBRATE_LATCH (5 * LATCH)
++
++static inline void mach_prepare_counter(void)
++{
++ /* Set the Gate high, disable speaker */
++ outb((inb(0x61) & ~0x02) | 0x01, 0x61);
++
++ /*
++ * Now let's take care of CTC channel 2
++ *
++ * Set the Gate high, program CTC channel 2 for mode 0,
++ * (interrupt on terminal count mode), binary count,
++ * load 5 * LATCH count, (LSB and MSB) to begin countdown.
++ *
++ * Some devices need a delay here.
++ */
++ outb(0xb0, 0x43); /* binary, mode 0, LSB/MSB, Ch 2 */
++ outb_p(CALIBRATE_LATCH & 0xff, 0x42); /* LSB of count */
++ outb_p(CALIBRATE_LATCH >> 8, 0x42); /* MSB of count */
++}
++
++static inline void mach_countup(unsigned long *count_p)
++{
++ unsigned long count = 0;
++ do {
++ count++;
++ } while ((inb_p(0x61) & 0x20) == 0);
++ *count_p = count;
++}
++
++#endif /* !_MACH_TIMER_H */
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/setup_arch_post.h new/include/asm-x86_64/mach-xen/setup_arch_post.h
+--- linux-2.6/include/asm-x86_64/mach-xen/setup_arch_post.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/setup_arch_post.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,58 @@
++/**
++ * machine_specific_* - Hooks for machine specific setup.
++ *
++ * Description:
++ * This is included late in kernel/setup.c so that it can make
++ * use of all of the static functions.
++ **/
++
++#include <xen/interface/callback.h>
++
++extern void hypervisor_callback(void);
++extern void failsafe_callback(void);
++extern void nmi(void);
++
++static void __init machine_specific_arch_setup(void)
++{
++ int ret;
++ struct callback_register event = {
++ .type = CALLBACKTYPE_event,
++ .address = (unsigned long) hypervisor_callback,
++ };
++ struct callback_register failsafe = {
++ .type = CALLBACKTYPE_failsafe,
++ .address = (unsigned long)failsafe_callback,
++ };
++ struct callback_register syscall = {
++ .type = CALLBACKTYPE_syscall,
++ .address = (unsigned long)system_call,
++ };
++#ifdef CONFIG_X86_LOCAL_APIC
++ struct callback_register nmi_cb = {
++ .type = CALLBACKTYPE_nmi,
++ .address = (unsigned long)nmi,
++ };
++#endif
++
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &event);
++ if (ret == 0)
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &failsafe);
++ if (ret == 0)
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &syscall);
++ if (ret == -ENOSYS)
++ ret = HYPERVISOR_set_callbacks(
++ event.address,
++ failsafe.address,
++ syscall.address);
++ BUG_ON(ret);
++
++#ifdef CONFIG_X86_LOCAL_APIC
++ ret = HYPERVISOR_callback_op(CALLBACKOP_register, &nmi_cb);
++ if (ret == -ENOSYS) {
++ struct xennmi_callback cb;
++
++ cb.handler_address = nmi_cb.address;
++ HYPERVISOR_nmi_op(XENNMI_register_callback, &cb);
++ }
++#endif
++}
+diff -urNp linux-2.6/include/asm-x86_64/mach-xen/setup_arch_pre.h new/include/asm-x86_64/mach-xen/setup_arch_pre.h
+--- linux-2.6/include/asm-x86_64/mach-xen/setup_arch_pre.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/asm-x86_64/mach-xen/setup_arch_pre.h 2006-05-09 12:35:42.000000000 +0200
+@@ -0,0 +1,5 @@
++/* Hook to call BIOS initialisation function */
++
++#define ARCH_SETUP machine_specific_arch_setup();
++
++static void __init machine_specific_arch_setup(void);
+diff -urNp linux-2.6/include/linux/ethtool.h new/include/linux/ethtool.h
+--- linux-2.6/include/linux/ethtool.h 2006-07-03 14:15:16.000000000 +0200
++++ new/include/linux/ethtool.h 2006-07-07 15:10:03.000000000 +0200
+@@ -408,6 +408,8 @@ struct ethtool_ops {
+ #define ETHTOOL_GPERMADDR 0x00000020 /* Get permanent hardware address */
+ #define ETHTOOL_GUFO 0x00000021 /* Get UFO enable (ethtool_value) */
+ #define ETHTOOL_SUFO 0x00000022 /* Set UFO enable (ethtool_value) */
++#define ETHTOOL_GGSO 0x00000023 /* Get GSO enable (ethtool_value) */
++#define ETHTOOL_SGSO 0x00000024 /* Set GSO enable (ethtool_value) */
+
+ /* compatibility with older code */
+ #define SPARC_ETH_GSET ETHTOOL_GSET
+diff -urNp linux-2.6/include/linux/gfp.h new/include/linux/gfp.h
+--- linux-2.6/include/linux/gfp.h 2006-07-03 14:15:16.000000000 +0200
++++ new/include/linux/gfp.h 2006-05-09 12:35:45.000000000 +0200
+@@ -100,7 +100,11 @@ static inline int gfp_zone(gfp_t gfp)
+ */
+
+ #ifndef HAVE_ARCH_FREE_PAGE
+-static inline void arch_free_page(struct page *page, int order) { }
++/*
++ * If arch_free_page returns non-zero then the generic free_page code can
++ * immediately bail: the arch-specific function has done all the work.
++ */
++static inline int arch_free_page(struct page *page, int order) { return 0; }
+ #endif
+
+ extern struct page *
+diff -urNp linux-2.6/include/linux/highmem.h new/include/linux/highmem.h
+--- linux-2.6/include/linux/highmem.h 2006-07-03 14:15:16.000000000 +0200
++++ new/include/linux/highmem.h 2006-05-09 12:35:45.000000000 +0200
+@@ -25,10 +25,16 @@ static inline void flush_kernel_dcache_p
+
+ /* declarations for linux/mm/highmem.c */
+ unsigned int nr_free_highpages(void);
++#ifdef CONFIG_XEN
++void kmap_flush_unused(void);
++#endif
+
+ #else /* CONFIG_HIGHMEM */
+
+ static inline unsigned int nr_free_highpages(void) { return 0; }
++#ifdef CONFIG_XEN
++static inline void kmap_flush_unused(void) { }
++#endif
+
+ static inline void *kmap(struct page *page)
+ {
+diff -urNp linux-2.6/include/linux/interrupt.h new/include/linux/interrupt.h
+--- linux-2.6/include/linux/interrupt.h 2006-07-03 14:15:17.000000000 +0200
++++ new/include/linux/interrupt.h 2006-06-28 14:32:14.000000000 +0200
+@@ -58,6 +58,12 @@ extern void disable_irq(unsigned int irq
+ extern void enable_irq(unsigned int irq);
+ #endif
+
++#ifdef CONFIG_HAVE_IRQ_IGNORE_UNHANDLED
++int irq_ignore_unhandled(unsigned int irq);
++#else
++#define irq_ignore_unhandled(irq) 0
++#endif
++
+ #ifndef __ARCH_SET_SOFTIRQ_PENDING
+ #define set_softirq_pending(x) (local_softirq_pending() = (x))
+ #define or_softirq_pending(x) (local_softirq_pending() |= (x))
+diff -urNp linux-2.6/include/linux/mm.h new/include/linux/mm.h
+--- linux-2.6/include/linux/mm.h 2006-07-03 14:15:17.000000000 +0200
++++ new/include/linux/mm.h 2006-05-09 12:40:16.000000000 +0200
+@@ -166,6 +166,9 @@ extern unsigned int kobjsize(const void
+ #define VM_NONLINEAR 0x00800000 /* Is non-linear (remap_file_pages) */
+ #define VM_MAPPED_COPY 0x01000000 /* T if mapped copy of data (nommu mmap) */
+ #define VM_INSERTPAGE 0x02000000 /* The vma has had "vm_insert_page()" done on it */
++#ifdef CONFIG_XEN
++#define VM_FOREIGN 0x04000000 /* Has pages belonging to another VM */
++#endif
+
+ #ifndef VM_STACK_DEFAULT_FLAGS /* arch can override this */
+ #define VM_STACK_DEFAULT_FLAGS VM_DATA_DEFAULT_FLAGS
+@@ -1014,6 +1017,13 @@ struct page *follow_page(struct vm_area_
+ #define FOLL_GET 0x04 /* do get_page on page */
+ #define FOLL_ANON 0x08 /* give ZERO_PAGE if no pgtable */
+
++#ifdef CONFIG_XEN
++typedef int (*pte_fn_t)(pte_t *pte, struct page *pmd_page, unsigned long addr,
++ void *data);
++extern int apply_to_page_range(struct mm_struct *mm, unsigned long address,
++ unsigned long size, pte_fn_t fn, void *data);
++#endif
++
+ #ifdef CONFIG_PROC_FS
+ void vm_stat_account(struct mm_struct *, unsigned long, struct file *, long);
+ #else
+diff -urNp linux-2.6/include/linux/netdevice.h new/include/linux/netdevice.h
+--- linux-2.6/include/linux/netdevice.h 2006-07-03 14:15:17.000000000 +0200
++++ new/include/linux/netdevice.h 2006-07-07 16:50:50.000000000 +0200
+@@ -232,6 +232,7 @@ enum netdev_state_t
+ __LINK_STATE_RX_SCHED,
+ __LINK_STATE_LINKWATCH_PENDING,
+ __LINK_STATE_DORMANT,
++ __LINK_STATE_QDISC_RUNNING,
+ };
+
+
+@@ -307,9 +308,17 @@ struct net_device
+ #define NETIF_F_HW_VLAN_RX 256 /* Receive VLAN hw acceleration */
+ #define NETIF_F_HW_VLAN_FILTER 512 /* Receive filtering on VLAN */
+ #define NETIF_F_VLAN_CHALLENGED 1024 /* Device cannot handle VLAN packets */
+-#define NETIF_F_TSO 2048 /* Can offload TCP/IP segmentation */
++#define NETIF_F_GSO 2048 /* Enable software GSO. */
+ #define NETIF_F_LLTX 4096 /* LockLess TX */
+-#define NETIF_F_UFO 8192 /* Can offload UDP Large Send*/
++
++ /* Segmentation offload features */
++#define NETIF_F_GSO_SHIFT 16
++#define NETIF_F_TSO (SKB_GSO_TCPV4 << NETIF_F_GSO_SHIFT)
++#define NETIF_F_UFO (SKB_GSO_UDPV4 << NETIF_F_GSO_SHIFT)
++#define NETIF_F_GSO_ROBUST (SKB_GSO_DODGY << NETIF_F_GSO_SHIFT)
++
++#define NETIF_F_GEN_CSUM (NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
++#define NETIF_F_ALL_CSUM (NETIF_F_IP_CSUM | NETIF_F_GEN_CSUM)
+
+ struct net_device *next_sched;
+
+@@ -398,6 +407,9 @@ struct net_device
+ struct list_head qdisc_list;
+ unsigned long tx_queue_len; /* Max frames per queue allowed */
+
++ /* Partially transmitted GSO packet. */
++ struct sk_buff *gso_skb;
++
+ /* ingress path synchronizer */
+ spinlock_t ingress_lock;
+ struct Qdisc *qdisc_ingress;
+@@ -406,7 +418,7 @@ struct net_device
+ * One part is mostly used on xmit path (device)
+ */
+ /* hard_start_xmit synchronizer */
+- spinlock_t xmit_lock ____cacheline_aligned_in_smp;
++ spinlock_t _xmit_lock ____cacheline_aligned_in_smp;
+ /* cpu id of processor entered to hard_start_xmit or -1,
+ if nobody entered there.
+ */
+@@ -532,6 +544,8 @@ struct packet_type {
+ struct net_device *,
+ struct packet_type *,
+ struct net_device *);
++ struct sk_buff *(*gso_segment)(struct sk_buff *skb,
++ int features);
+ void *af_packet_priv;
+ struct list_head list;
+ };
+@@ -679,7 +693,8 @@ extern int dev_change_name(struct net_d
+ extern int dev_set_mtu(struct net_device *, int);
+ extern int dev_set_mac_address(struct net_device *,
+ struct sockaddr *);
+-extern void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev);
++extern int dev_hard_start_xmit(struct sk_buff *skb,
++ struct net_device *dev);
+
+ extern void dev_init(void);
+
+@@ -889,11 +904,43 @@ static inline void __netif_rx_complete(s
+ clear_bit(__LINK_STATE_RX_SCHED, &dev->state);
+ }
+
++static inline void netif_tx_lock(struct net_device *dev)
++{
++ spin_lock(&dev->_xmit_lock);
++ dev->xmit_lock_owner = smp_processor_id();
++}
++
++static inline void netif_tx_lock_bh(struct net_device *dev)
++{
++ spin_lock_bh(&dev->_xmit_lock);
++ dev->xmit_lock_owner = smp_processor_id();
++}
++
++static inline int netif_tx_trylock(struct net_device *dev)
++{
++ int err = spin_trylock(&dev->_xmit_lock);
++ if (!err)
++ dev->xmit_lock_owner = smp_processor_id();
++ return err;
++}
++
++static inline void netif_tx_unlock(struct net_device *dev)
++{
++ dev->xmit_lock_owner = -1;
++ spin_unlock(&dev->_xmit_lock);
++}
++
++static inline void netif_tx_unlock_bh(struct net_device *dev)
++{
++ dev->xmit_lock_owner = -1;
++ spin_unlock_bh(&dev->_xmit_lock);
++}
++
+ static inline void netif_tx_disable(struct net_device *dev)
+ {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ netif_stop_queue(dev);
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ }
+
+ /* These functions live elsewhere (drivers/net/net_init.c, but related) */
+@@ -921,6 +968,7 @@ extern int netdev_max_backlog;
+ extern int weight_p;
+ extern int netdev_set_master(struct net_device *dev, struct net_device *master);
+ extern int skb_checksum_help(struct sk_buff *skb, int inward);
++extern struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features);
+ #ifdef CONFIG_BUG
+ extern void netdev_rx_csum_fault(struct net_device *dev);
+ #else
+@@ -940,6 +988,18 @@ extern void dev_seq_stop(struct seq_file
+
+ extern void linkwatch_run_queue(void);
+
++static inline int skb_gso_ok(struct sk_buff *skb, int features)
++{
++ int feature = skb_shinfo(skb)->gso_size ?
++ skb_shinfo(skb)->gso_type << NETIF_F_GSO_SHIFT : 0;
++ return (features & feature) == feature;
++}
++
++static inline int netif_needs_gso(struct net_device *dev, struct sk_buff *skb)
++{
++ return !skb_gso_ok(skb, dev->features);
++}
++
+ #endif /* __KERNEL__ */
+
+ #endif /* _LINUX_DEV_H */
+diff -urNp linux-2.6/include/linux/oprofile.h new/include/linux/oprofile.h
+--- linux-2.6/include/linux/oprofile.h 2006-07-03 14:15:18.000000000 +0200
++++ new/include/linux/oprofile.h 2006-07-07 16:08:22.000000000 +0200
+@@ -16,6 +16,10 @@
+ #include <linux/types.h>
+ #include <linux/spinlock.h>
+ #include <asm/atomic.h>
++
++#ifdef CONFIG_XEN
++#include <xen/interface/xenoprof.h>
++#endif
+
+ struct super_block;
+ struct dentry;
+@@ -27,6 +31,11 @@ struct oprofile_operations {
+ /* create any necessary configuration files in the oprofile fs.
+ * Optional. */
+ int (*create_files)(struct super_block * sb, struct dentry * root);
++ /* setup active domains with Xen */
++ int (*set_active)(int *active_domains, unsigned int adomains);
++ /* setup passive domains with Xen */
++ int (*set_passive)(int *passive_domains, unsigned int pdomains);
++
+ /* Do any necessary interrupt setup. Optional. */
+ int (*setup)(void);
+ /* Do any necessary interrupt shutdown. Optional. */
+diff -urNp linux-2.6/include/linux/pfn.h new/include/linux/pfn.h
+--- linux-2.6/include/linux/pfn.h 2006-07-03 14:15:18.000000000 +0200
++++ new/include/linux/pfn.h 2006-05-09 12:35:49.000000000 +0200
+@@ -4,6 +4,6 @@
+ #define PFN_ALIGN(x) (((unsigned long)(x) + (PAGE_SIZE - 1)) & PAGE_MASK)
+ #define PFN_UP(x) (((x) + PAGE_SIZE-1) >> PAGE_SHIFT)
+ #define PFN_DOWN(x) ((x) >> PAGE_SHIFT)
+-#define PFN_PHYS(x) ((x) << PAGE_SHIFT)
++#define PFN_PHYS(x) ((unsigned long long)(x) << PAGE_SHIFT)
+
+ #endif
+diff -urNp linux-2.6/include/linux/skbuff.h new/include/linux/skbuff.h
+--- linux-2.6/include/linux/skbuff.h 2006-07-03 14:15:18.000000000 +0200
++++ new/include/linux/skbuff.h 2006-07-07 16:12:19.000000000 +0200
+@@ -134,9 +134,10 @@ struct skb_frag_struct {
+ struct skb_shared_info {
+ atomic_t dataref;
+ unsigned short nr_frags;
+- unsigned short tso_size;
+- unsigned short tso_segs;
+- unsigned short ufo_size;
++ unsigned short gso_size;
++ /* Warning: this field is not always filled in (UFO)! */
++ unsigned short gso_segs;
++ unsigned short gso_type;
+ unsigned int ip6_frag_id;
+ struct sk_buff *frag_list;
+ skb_frag_t frags[MAX_SKB_FRAGS];
+@@ -168,6 +169,14 @@ enum {
+ SKB_FCLONE_CLONE,
+ };
+
++enum {
++ SKB_GSO_TCPV4 = 1 << 0,
++ SKB_GSO_UDPV4 = 1 << 1,
++
++ /* This indicates the skb is from an untrusted source. */
++ SKB_GSO_DODGY = 1 << 2,
++};
++
+ /**
+ * struct sk_buff - socket buffer
+ * @next: Next buffer in list
+@@ -189,6 +198,8 @@ enum {
+ * @local_df: allow local fragmentation
+ * @cloned: Head may be cloned (check refcnt to be sure)
+ * @nohdr: Payload reference only, must not modify header
++ * @proto_data_valid: Protocol data validated since arriving at localhost
++ * @proto_csum_blank: Protocol csum must be added before leaving localhost
+ * @pkt_type: Packet class
+ * @fclone: skbuff clone status
+ * @ip_summed: Driver fed us an IP checksum
+@@ -265,7 +276,13 @@ struct sk_buff {
+ nfctinfo:3;
+ __u8 pkt_type:3,
+ fclone:2,
++#ifndef CONFIG_XEN
+ ipvs_property:1;
++#else
++ ipvs_property:1,
++ proto_data_valid:1,
++ proto_csum_blank:1;
++#endif
+ __be16 protocol;
+
+ void (*destructor)(struct sk_buff *skb);
+@@ -322,7 +339,8 @@ static inline struct sk_buff *alloc_skb_
+
+ extern struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp,
+ unsigned int size,
+- gfp_t priority);
++ gfp_t priority,
++ int fclone);
+ extern void kfree_skbmem(struct sk_buff *skb);
+ extern struct sk_buff *skb_clone(struct sk_buff *skb,
+ gfp_t priority);
+@@ -1062,7 +1080,7 @@ static inline struct sk_buff *__dev_allo
+ return skb;
+ }
+ #else
+-extern struct sk_buff *__dev_alloc_skb(unsigned int length, int gfp_mask);
++extern struct sk_buff *__dev_alloc_skb(unsigned int length, gfp_t gfp_mask);
+ #endif
+
+ /**
+@@ -1161,18 +1179,34 @@ static inline int skb_can_coalesce(struc
+ return 0;
+ }
+
++static inline int __skb_linearize(struct sk_buff *skb)
++{
++ return __pskb_pull_tail(skb, skb->data_len) ? 0 : -ENOMEM;
++}
++
+ /**
+ * skb_linearize - convert paged skb to linear one
+ * @skb: buffer to linarize
+- * @gfp: allocation mode
+ *
+ * If there is no free memory -ENOMEM is returned, otherwise zero
+ * is returned and the old skb data released.
+ */
+-extern int __skb_linearize(struct sk_buff *skb, gfp_t gfp);
+-static inline int skb_linearize(struct sk_buff *skb, gfp_t gfp)
++static inline int skb_linearize(struct sk_buff *skb)
++{
++ return skb_is_nonlinear(skb) ? __skb_linearize(skb) : 0;
++}
++
++/**
++ * skb_linearize_cow - make sure skb is linear and writable
++ * @skb: buffer to process
++ *
++ * If there is no free memory -ENOMEM is returned, otherwise zero
++ * is returned and the old skb data released.
++ */
++static inline int skb_linearize_cow(struct sk_buff *skb)
+ {
+- return __skb_linearize(skb, gfp);
++ return skb_is_nonlinear(skb) || skb_cloned(skb) ?
++ __skb_linearize(skb) : 0;
+ }
+
+ /**
+@@ -1269,6 +1303,7 @@ extern void skb_split(struct sk_b
+ struct sk_buff *skb1, const u32 len);
+
+ extern void skb_release_data(struct sk_buff *skb);
++extern struct sk_buff *skb_segment(struct sk_buff *skb, int features);
+
+ static inline void *skb_header_pointer(const struct sk_buff *skb, int offset,
+ int len, void *buffer)
+diff -urNp linux-2.6/include/net/pkt_sched.h new/include/net/pkt_sched.h
+--- linux-2.6/include/net/pkt_sched.h 2006-07-03 14:15:19.000000000 +0200
++++ new/include/net/pkt_sched.h 2006-07-07 15:10:03.000000000 +0200
+@@ -218,12 +218,13 @@ extern struct qdisc_rate_table *qdisc_ge
+ struct rtattr *tab);
+ extern void qdisc_put_rtab(struct qdisc_rate_table *tab);
+
+-extern int qdisc_restart(struct net_device *dev);
++extern void __qdisc_run(struct net_device *dev);
+
+ static inline void qdisc_run(struct net_device *dev)
+ {
+- while (!netif_queue_stopped(dev) && qdisc_restart(dev) < 0)
+- /* NOTHING */;
++ if (!netif_queue_stopped(dev) &&
++ !test_and_set_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
++ __qdisc_run(dev);
+ }
+
+ extern int tc_classify(struct sk_buff *skb, struct tcf_proto *tp,
+diff -urNp linux-2.6/include/net/protocol.h new/include/net/protocol.h
+--- linux-2.6/include/net/protocol.h 2006-07-03 14:15:19.000000000 +0200
++++ new/include/net/protocol.h 2006-07-07 15:10:03.000000000 +0200
+@@ -37,6 +37,8 @@
+ struct net_protocol {
+ int (*handler)(struct sk_buff *skb);
+ void (*err_handler)(struct sk_buff *skb, u32 info);
++ struct sk_buff *(*gso_segment)(struct sk_buff *skb,
++ int features);
+ int no_policy;
+ };
+
+diff -urNp linux-2.6/include/net/sock.h new/include/net/sock.h
+--- linux-2.6/include/net/sock.h 2006-07-03 14:15:19.000000000 +0200
++++ new/include/net/sock.h 2006-07-07 16:12:19.000000000 +0200
+@@ -1032,9 +1032,13 @@ static inline void sk_setup_caps(struct
+ {
+ __sk_dst_set(sk, dst);
+ sk->sk_route_caps = dst->dev->features;
++ if (sk->sk_route_caps & NETIF_F_GSO)
++ sk->sk_route_caps |= NETIF_F_TSO;
+ if (sk->sk_route_caps & NETIF_F_TSO) {
+ if (sock_flag(sk, SOCK_NO_LARGESEND) || dst->header_len)
+ sk->sk_route_caps &= ~NETIF_F_TSO;
++ else
++ sk->sk_route_caps |= NETIF_F_SG | NETIF_F_HW_CSUM;
+ }
+ }
+
+diff -urNp linux-2.6/include/net/tcp.h new/include/net/tcp.h
+--- linux-2.6/include/net/tcp.h 2006-07-03 14:15:19.000000000 +0200
++++ new/include/net/tcp.h 2006-07-07 16:12:20.000000000 +0200
+@@ -565,13 +565,13 @@ struct tcp_skb_cb {
+ */
+ static inline int tcp_skb_pcount(const struct sk_buff *skb)
+ {
+- return skb_shinfo(skb)->tso_segs;
++ return skb_shinfo(skb)->gso_segs;
+ }
+
+ /* This is valid iff tcp_skb_pcount() > 1. */
+ static inline int tcp_skb_mss(const struct sk_buff *skb)
+ {
+- return skb_shinfo(skb)->tso_size;
++ return skb_shinfo(skb)->gso_size;
+ }
+
+ static inline void tcp_dec_pcount_approx(__u32 *count,
+@@ -1076,6 +1076,8 @@ extern struct request_sock_ops tcp_reque
+
+ extern int tcp_v4_destroy_sock(struct sock *sk);
+
++extern struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features);
++
+ #ifdef CONFIG_PROC_FS
+ extern int tcp4_proc_init(void);
+ extern void tcp4_proc_exit(void);
+diff -urNp linux-2.6/include/xen/balloon.h new/include/xen/balloon.h
+--- linux-2.6/include/xen/balloon.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/balloon.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,63 @@
++/******************************************************************************
++ * balloon.h
++ *
++ * Xen balloon driver - enables returning/claiming memory to/from Xen.
++ *
++ * Copyright (c) 2003, B Dragovic
++ * Copyright (c) 2003-2004, M Williamson, K Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __ASM_BALLOON_H__
++#define __ASM_BALLOON_H__
++
++/*
++ * Inform the balloon driver that it should allow some slop for device-driver
++ * memory activities.
++ */
++extern void
++balloon_update_driver_allowance(
++ long delta);
++
++/* Allocate an empty low-memory page range. */
++extern struct page *
++balloon_alloc_empty_page_range(
++ unsigned long nr_pages);
++
++/* Deallocate an empty page range, adding to the balloon. */
++extern void
++balloon_dealloc_empty_page_range(
++ struct page *page, unsigned long nr_pages);
++
++/*
++ * Prevent the balloon driver from changing the memory reservation during
++ * a driver critical region.
++ */
++extern spinlock_t balloon_lock;
++#define balloon_lock(__flags) spin_lock_irqsave(&balloon_lock, __flags)
++#define balloon_unlock(__flags) spin_unlock_irqrestore(&balloon_lock, __flags)
++
++#endif /* __ASM_BALLOON_H__ */
+diff -urNp linux-2.6/include/xen/cpu_hotplug.h new/include/xen/cpu_hotplug.h
+--- linux-2.6/include/xen/cpu_hotplug.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/cpu_hotplug.h 2006-06-07 13:15:16.000000000 +0200
+@@ -0,0 +1,44 @@
++#ifndef __XEN_CPU_HOTPLUG_H__
++#define __XEN_CPU_HOTPLUG_H__
++
++#include <linux/config.h>
++#include <linux/kernel.h>
++#include <linux/cpumask.h>
++
++#if defined(CONFIG_HOTPLUG_CPU)
++
++#if defined(CONFIG_X86)
++void cpu_initialize_context(unsigned int cpu);
++#else
++#define cpu_initialize_context(cpu) ((void)0)
++#endif
++
++int cpu_up_check(unsigned int cpu);
++void init_xenbus_allowed_cpumask(void);
++int smp_suspend(void);
++void smp_resume(void);
++
++void cpu_bringup(void);
++
++#else /* !defined(CONFIG_HOTPLUG_CPU) */
++
++#define cpu_up_check(cpu) (0)
++#define init_xenbus_allowed_cpumask() ((void)0)
++
++static inline int smp_suspend(void)
++{
++ if (num_online_cpus() > 1) {
++ printk(KERN_WARNING "Can't suspend SMP guests "
++ "without CONFIG_HOTPLUG_CPU\n");
++ return -EOPNOTSUPP;
++ }
++ return 0;
++}
++
++static inline void smp_resume(void)
++{
++}
++
++#endif /* !defined(CONFIG_HOTPLUG_CPU) */
++
++#endif /* __XEN_CPU_HOTPLUG_H__ */
+diff -urNp linux-2.6/include/xen/driver_util.h new/include/xen/driver_util.h
+--- linux-2.6/include/xen/driver_util.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/driver_util.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,16 @@
++
++#ifndef __ASM_XEN_DRIVER_UTIL_H__
++#define __ASM_XEN_DRIVER_UTIL_H__
++
++#include <linux/config.h>
++#include <linux/vmalloc.h>
++
++/* Allocate/destroy a 'vmalloc' VM area. */
++extern struct vm_struct *alloc_vm_area(unsigned long size);
++extern void free_vm_area(struct vm_struct *area);
++
++/* Lock an area so that PTEs are accessible in the current address space. */
++extern void lock_vm_area(struct vm_struct *area);
++extern void unlock_vm_area(struct vm_struct *area);
++
++#endif /* __ASM_XEN_DRIVER_UTIL_H__ */
+diff -urNp linux-2.6/include/xen/evtchn.h new/include/xen/evtchn.h
+--- linux-2.6/include/xen/evtchn.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/evtchn.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,114 @@
++/******************************************************************************
++ * evtchn.h
++ *
++ * Communication via Xen event channels.
++ * Also definitions for the device that demuxes notifications to userspace.
++ *
++ * Copyright (c) 2004-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __ASM_EVTCHN_H__
++#define __ASM_EVTCHN_H__
++
++#include <linux/config.h>
++#include <linux/interrupt.h>
++#include <asm/hypervisor.h>
++#include <asm/ptrace.h>
++#include <asm/synch_bitops.h>
++#include <xen/interface/event_channel.h>
++#include <linux/smp.h>
++
++/*
++ * LOW-LEVEL DEFINITIONS
++ */
++
++/*
++ * Dynamically bind an event source to an IRQ-like callback handler.
++ * On some platforms this may not be implemented via the Linux IRQ subsystem.
++ * The IRQ argument passed to the callback handler is the same as returned
++ * from the bind call. It may not correspond to a Linux IRQ number.
++ * Returns IRQ or negative errno.
++ * UNBIND: Takes IRQ to unbind from; automatically closes the event channel.
++ */
++extern int bind_evtchn_to_irqhandler(
++ unsigned int evtchn,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id);
++extern int bind_virq_to_irqhandler(
++ unsigned int virq,
++ unsigned int cpu,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id);
++extern int bind_ipi_to_irqhandler(
++ unsigned int ipi,
++ unsigned int cpu,
++ irqreturn_t (*handler)(int, void *, struct pt_regs *),
++ unsigned long irqflags,
++ const char *devname,
++ void *dev_id);
++
++/*
++ * Common unbind function for all event sources. Takes IRQ to unbind from.
++ * Automatically closes the underlying event channel (even for bindings
++ * made with bind_evtchn_to_irqhandler()).
++ */
++extern void unbind_from_irqhandler(unsigned int irq, void *dev_id);
++
++extern void irq_resume(void);
++
++/* Entry point for notifications into Linux subsystems. */
++asmlinkage void evtchn_do_upcall(struct pt_regs *regs);
++
++/* Entry point for notifications into the userland character device. */
++extern void evtchn_device_upcall(int port);
++
++extern void mask_evtchn(int port);
++extern void unmask_evtchn(int port);
++
++static inline void clear_evtchn(int port)
++{
++ shared_info_t *s = HYPERVISOR_shared_info;
++ synch_clear_bit(port, &s->evtchn_pending[0]);
++}
++
++static inline void notify_remote_via_evtchn(int port)
++{
++ struct evtchn_send send = { .port = port };
++ (void)HYPERVISOR_event_channel_op(EVTCHNOP_send, &send);
++}
++
++/*
++ * Unlike notify_remote_via_evtchn(), this is safe to use across
++ * save/restore. Notifications on a broken connection are silently dropped.
++ */
++extern void notify_remote_via_irq(int irq);
++
++#endif /* __ASM_EVTCHN_H__ */
+diff -urNp linux-2.6/include/xen/features.h new/include/xen/features.h
+--- linux-2.6/include/xen/features.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/features.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,20 @@
++/******************************************************************************
++ * features.h
++ *
++ * Query the features reported by Xen.
++ *
++ * Copyright (c) 2006, Ian Campbell
++ */
++
++#ifndef __ASM_XEN_FEATURES_H__
++#define __ASM_XEN_FEATURES_H__
++
++#include <xen/interface/version.h>
++
++extern void setup_xen_features(void);
++
++extern u8 xen_features[XENFEAT_NR_SUBMAPS * 32];
++
++#define xen_feature(flag) (xen_features[flag])
++
++#endif /* __ASM_XEN_FEATURES_H__ */
+diff -urNp linux-2.6/include/xen/foreign_page.h new/include/xen/foreign_page.h
+--- linux-2.6/include/xen/foreign_page.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/foreign_page.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,30 @@
++/******************************************************************************
++ * foreign_page.h
++ *
++ * Provide a "foreign" page type, that is owned by a foreign allocator and
++ * not the normal buddy allocator in page_alloc.c
++ *
++ * Copyright (c) 2004, K A Fraser
++ */
++
++#ifndef __ASM_XEN_FOREIGN_PAGE_H__
++#define __ASM_XEN_FOREIGN_PAGE_H__
++
++#define PG_foreign PG_arch_1
++
++#define PageForeign(page) test_bit(PG_foreign, &(page)->flags)
++
++#define SetPageForeign(page, dtor) do { \
++ set_bit(PG_foreign, &(page)->flags); \
++ (page)->mapping = (void *)dtor; \
++} while (0)
++
++#define ClearPageForeign(page) do { \
++ clear_bit(PG_foreign, &(page)->flags); \
++ (page)->mapping = NULL; \
++} while (0)
++
++#define PageForeignDestructor(page) \
++ ( (void (*) (struct page *)) (page)->mapping )
++
++#endif /* __ASM_XEN_FOREIGN_PAGE_H__ */
+diff -urNp linux-2.6/include/xen/gnttab.h new/include/xen/gnttab.h
+--- linux-2.6/include/xen/gnttab.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/gnttab.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,151 @@
++/******************************************************************************
++ * gnttab.h
++ *
++ * Two sets of functionality:
++ * 1. Granting foreign access to our memory reservation.
++ * 2. Accessing others' memory reservations via grant references.
++ * (i.e., mechanisms for both sender and recipient of grant references)
++ *
++ * Copyright (c) 2004-2005, K A Fraser
++ * Copyright (c) 2005, Christopher Clark
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __ASM_GNTTAB_H__
++#define __ASM_GNTTAB_H__
++
++#include <linux/config.h>
++#include <asm/hypervisor.h>
++#include <xen/interface/grant_table.h>
++#include <xen/features.h>
++
++/* NR_GRANT_FRAMES must be less than or equal to that configured in Xen */
++#ifdef __ia64__
++#define NR_GRANT_FRAMES 1
++#else
++#define NR_GRANT_FRAMES 4
++#endif
++
++struct gnttab_free_callback {
++ struct gnttab_free_callback *next;
++ void (*fn)(void *);
++ void *arg;
++ u16 count;
++};
++
++int gnttab_grant_foreign_access(domid_t domid, unsigned long frame,
++ int readonly);
++
++/*
++ * End access through the given grant reference, iff the grant entry is no
++ * longer in use. Return 1 if the grant entry was freed, 0 if it is still in
++ * use.
++ */
++int gnttab_end_foreign_access_ref(grant_ref_t ref, int readonly);
++
++/*
++ * Eventually end access through the given grant reference, and once that
++ * access has been ended, free the given page too. Access will be ended
++ * immediately iff the grant entry is not in use, otherwise it will happen
++ * some time later. page may be 0, in which case no freeing will occur.
++ */
++void gnttab_end_foreign_access(grant_ref_t ref, int readonly,
++ unsigned long page);
++
++int gnttab_grant_foreign_transfer(domid_t domid, unsigned long pfn);
++
++unsigned long gnttab_end_foreign_transfer_ref(grant_ref_t ref);
++unsigned long gnttab_end_foreign_transfer(grant_ref_t ref);
++
++int gnttab_query_foreign_access(grant_ref_t ref);
++
++/*
++ * operations on reserved batches of grant references
++ */
++int gnttab_alloc_grant_references(u16 count, grant_ref_t *pprivate_head);
++
++void gnttab_free_grant_reference(grant_ref_t ref);
++
++void gnttab_free_grant_references(grant_ref_t head);
++
++int gnttab_empty_grant_references(const grant_ref_t *pprivate_head);
++
++int gnttab_claim_grant_reference(grant_ref_t *pprivate_head);
++
++void gnttab_release_grant_reference(grant_ref_t *private_head,
++ grant_ref_t release);
++
++void gnttab_request_free_callback(struct gnttab_free_callback *callback,
++ void (*fn)(void *), void *arg, u16 count);
++void gnttab_cancel_free_callback(struct gnttab_free_callback *callback);
++
++void gnttab_grant_foreign_access_ref(grant_ref_t ref, domid_t domid,
++ unsigned long frame, int readonly);
++
++void gnttab_grant_foreign_transfer_ref(grant_ref_t, domid_t domid,
++ unsigned long pfn);
++
++#ifdef __ia64__
++#define gnttab_map_vaddr(map) __va(map.dev_bus_addr)
++#else
++#define gnttab_map_vaddr(map) ((void *)(map.host_virt_addr))
++#endif
++
++int gnttab_suspend(void);
++int gnttab_resume(void);
++
++static inline void
++gnttab_set_map_op(struct gnttab_map_grant_ref *map, unsigned long addr,
++ uint32_t flags, grant_ref_t ref, domid_t domid)
++{
++ if (flags & GNTMAP_contains_pte)
++ map->host_addr = addr;
++ else if (xen_feature(XENFEAT_auto_translated_physmap))
++ map->host_addr = __pa(addr);
++ else
++ map->host_addr = addr;
++
++ map->flags = flags;
++ map->ref = ref;
++ map->dom = domid;
++}
++
++static inline void
++gnttab_set_unmap_op(struct gnttab_unmap_grant_ref *unmap, unsigned long addr,
++ uint32_t flags, grant_handle_t handle)
++{
++ if (flags & GNTMAP_contains_pte)
++ unmap->host_addr = addr;
++ else if (xen_feature(XENFEAT_auto_translated_physmap))
++ unmap->host_addr = __pa(addr);
++ else
++ unmap->host_addr = addr;
++
++ unmap->handle = handle;
++ unmap->dev_bus_addr = 0;
++}
++
++#endif /* __ASM_GNTTAB_H__ */
+diff -urNp linux-2.6/include/xen/hypervisor_sysfs.h new/include/xen/hypervisor_sysfs.h
+--- linux-2.6/include/xen/hypervisor_sysfs.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/hypervisor_sysfs.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,32 @@
++/*
++ * copyright (c) 2006 IBM Corporation
++ * Authored by: Mike D. Day <ncmike@us.ibm.com>
++ *
++ * This program is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License version 2 as
++ * published by the Free Software Foundation.
++ */
++
++#ifndef _HYP_SYSFS_H_
++#define _HYP_SYSFS_H_
++
++#include <linux/kobject.h>
++#include <linux/sysfs.h>
++
++#define HYPERVISOR_ATTR_RO(_name) \
++static struct hyp_sysfs_attr _name##_attr = __ATTR_RO(_name)
++
++#define HYPERVISOR_ATTR_RW(_name) \
++static struct hyp_sysfs_attr _name##_attr = \
++ __ATTR(_name, 0644, _name##_show, _name##_store)
++
++extern struct subsystem hypervisor_subsys;
++
++struct hyp_sysfs_attr {
++ struct attribute attr;
++ ssize_t (*show)(struct hyp_sysfs_attr *, char *);
++ ssize_t (*store)(struct hyp_sysfs_attr *, const char *, size_t);
++ void *hyp_attr_data;
++};
++
++#endif /* _HYP_SYSFS_H_ */
+diff -urNp linux-2.6/include/xen/interface/acm.h new/include/xen/interface/acm.h
+--- linux-2.6/include/xen/interface/acm.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/acm.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,188 @@
++/*
++ * acm.h: Xen access control module interface defintions
++ *
++ * Reiner Sailer <sailer@watson.ibm.com>
++ * Copyright (c) 2005, International Business Machines Corporation.
++ */
++
++#ifndef _XEN_PUBLIC_ACM_H
++#define _XEN_PUBLIC_ACM_H
++
++#include "xen.h"
++#include "sched_ctl.h"
++
++/* if ACM_DEBUG defined, all hooks should
++ * print a short trace message (comment it out
++ * when not in testing mode )
++ */
++/* #define ACM_DEBUG */
++
++#ifdef ACM_DEBUG
++# define printkd(fmt, args...) printk(fmt,## args)
++#else
++# define printkd(fmt, args...)
++#endif
++
++/* default ssid reference value if not supplied */
++#define ACM_DEFAULT_SSID 0x0
++#define ACM_DEFAULT_LOCAL_SSID 0x0
++
++/* Internal ACM ERROR types */
++#define ACM_OK 0
++#define ACM_UNDEF -1
++#define ACM_INIT_SSID_ERROR -2
++#define ACM_INIT_SOID_ERROR -3
++#define ACM_ERROR -4
++
++/* External ACCESS DECISIONS */
++#define ACM_ACCESS_PERMITTED 0
++#define ACM_ACCESS_DENIED -111
++#define ACM_NULL_POINTER_ERROR -200
++
++/* primary policy in lower 4 bits */
++#define ACM_NULL_POLICY 0
++#define ACM_CHINESE_WALL_POLICY 1
++#define ACM_SIMPLE_TYPE_ENFORCEMENT_POLICY 2
++#define ACM_POLICY_UNDEFINED 15
++
++/* combinations have secondary policy component in higher 4bit */
++#define ACM_CHINESE_WALL_AND_SIMPLE_TYPE_ENFORCEMENT_POLICY \
++ ((ACM_SIMPLE_TYPE_ENFORCEMENT_POLICY << 4) | ACM_CHINESE_WALL_POLICY)
++
++/* policy: */
++#define ACM_POLICY_NAME(X) \
++ ((X) == (ACM_NULL_POLICY)) ? "NULL" : \
++ ((X) == (ACM_CHINESE_WALL_POLICY)) ? "CHINESE WALL" : \
++ ((X) == (ACM_SIMPLE_TYPE_ENFORCEMENT_POLICY)) ? "SIMPLE TYPE ENFORCEMENT" : \
++ ((X) == (ACM_CHINESE_WALL_AND_SIMPLE_TYPE_ENFORCEMENT_POLICY)) ? "CHINESE WALL AND SIMPLE TYPE ENFORCEMENT" : \
++ "UNDEFINED"
++
++/* the following policy versions must be increased
++ * whenever the interpretation of the related
++ * policy's data structure changes
++ */
++#define ACM_POLICY_VERSION 2
++#define ACM_CHWALL_VERSION 1
++#define ACM_STE_VERSION 1
++
++/* defines a ssid reference used by xen */
++typedef uint32_t ssidref_t;
++
++/* hooks that are known to domains */
++#define ACMHOOK_none 0
++#define ACMHOOK_sharing 1
++
++/* -------security policy relevant type definitions-------- */
++
++/* type identifier; compares to "equal" or "not equal" */
++typedef uint16_t domaintype_t;
++
++/* CHINESE WALL POLICY DATA STRUCTURES
++ *
++ * current accumulated conflict type set:
++ * When a domain is started and has a type that is in
++ * a conflict set, the conflicting types are incremented in
++ * the aggregate set. When a domain is destroyed, the
++ * conflicting types to its type are decremented.
++ * If a domain has multiple types, this procedure works over
++ * all those types.
++ *
++ * conflict_aggregate_set[i] holds the number of
++ * running domains that have a conflict with type i.
++ *
++ * running_types[i] holds the number of running domains
++ * that include type i in their ssidref-referenced type set
++ *
++ * conflict_sets[i][j] is "0" if type j has no conflict
++ * with type i and is "1" otherwise.
++ */
++/* high-16 = version, low-16 = check magic */
++#define ACM_MAGIC 0x0001debc
++
++/* each offset in bytes from start of the struct they
++ * are part of */
++
++/* each buffer consists of all policy information for
++ * the respective policy given in the policy code
++ *
++ * acm_policy_buffer, acm_chwall_policy_buffer,
++ * and acm_ste_policy_buffer need to stay 32-bit aligned
++ * because we create binary policies also with external
++ * tools that assume packed representations (e.g. the java tool)
++ */
++struct acm_policy_buffer {
++ uint32_t policy_version; /* ACM_POLICY_VERSION */
++ uint32_t magic;
++ uint32_t len;
++ uint32_t policy_reference_offset;
++ uint32_t primary_policy_code;
++ uint32_t primary_buffer_offset;
++ uint32_t secondary_policy_code;
++ uint32_t secondary_buffer_offset;
++};
++
++struct acm_policy_reference_buffer {
++ uint32_t len;
++};
++
++struct acm_chwall_policy_buffer {
++ uint32_t policy_version; /* ACM_CHWALL_VERSION */
++ uint32_t policy_code;
++ uint32_t chwall_max_types;
++ uint32_t chwall_max_ssidrefs;
++ uint32_t chwall_max_conflictsets;
++ uint32_t chwall_ssid_offset;
++ uint32_t chwall_conflict_sets_offset;
++ uint32_t chwall_running_types_offset;
++ uint32_t chwall_conflict_aggregate_offset;
++};
++
++struct acm_ste_policy_buffer {
++ uint32_t policy_version; /* ACM_STE_VERSION */
++ uint32_t policy_code;
++ uint32_t ste_max_types;
++ uint32_t ste_max_ssidrefs;
++ uint32_t ste_ssid_offset;
++};
++
++struct acm_stats_buffer {
++ uint32_t magic;
++ uint32_t len;
++ uint32_t primary_policy_code;
++ uint32_t primary_stats_offset;
++ uint32_t secondary_policy_code;
++ uint32_t secondary_stats_offset;
++};
++
++struct acm_ste_stats_buffer {
++ uint32_t ec_eval_count;
++ uint32_t gt_eval_count;
++ uint32_t ec_denied_count;
++ uint32_t gt_denied_count;
++ uint32_t ec_cachehit_count;
++ uint32_t gt_cachehit_count;
++};
++
++struct acm_ssid_buffer {
++ uint32_t len;
++ ssidref_t ssidref;
++ uint32_t policy_reference_offset;
++ uint32_t primary_policy_code;
++ uint32_t primary_max_types;
++ uint32_t primary_types_offset;
++ uint32_t secondary_policy_code;
++ uint32_t secondary_max_types;
++ uint32_t secondary_types_offset;
++};
++
++#endif
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/acm_ops.h new/include/xen/interface/acm_ops.h
+--- linux-2.6/include/xen/interface/acm_ops.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/acm_ops.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,103 @@
++/*
++ * acm_ops.h: Xen access control module hypervisor commands
++ *
++ * Reiner Sailer <sailer@watson.ibm.com>
++ * Copyright (c) 2005,2006 International Business Machines Corporation.
++ */
++
++#ifndef __XEN_PUBLIC_ACM_OPS_H__
++#define __XEN_PUBLIC_ACM_OPS_H__
++
++#include "xen.h"
++#include "sched_ctl.h"
++#include "acm.h"
++
++/*
++ * Make sure you increment the interface version whenever you modify this file!
++ * This makes sure that old versions of acm tools will stop working in a
++ * well-defined way (rather than crashing the machine, for instance).
++ */
++#define ACM_INTERFACE_VERSION 0xAAAA0008
++
++/************************************************************************/
++
++/*
++ * Prototype for this hypercall is:
++ * int acm_op(int cmd, void *args)
++ * @cmd == ACMOP_??? (access control module operation).
++ * @args == Operation-specific extra arguments (NULL if none).
++ */
++
++
++#define ACMOP_setpolicy 1
++struct acm_setpolicy {
++ /* IN */
++ uint32_t interface_version;
++ XEN_GUEST_HANDLE(void) pushcache;
++ uint32_t pushcache_size;
++};
++
++
++#define ACMOP_getpolicy 2
++struct acm_getpolicy {
++ /* IN */
++ uint32_t interface_version;
++ XEN_GUEST_HANDLE(void) pullcache;
++ uint32_t pullcache_size;
++};
++
++
++#define ACMOP_dumpstats 3
++struct acm_dumpstats {
++ /* IN */
++ uint32_t interface_version;
++ XEN_GUEST_HANDLE(void) pullcache;
++ uint32_t pullcache_size;
++};
++
++
++#define ACMOP_getssid 4
++#define ACM_GETBY_ssidref 1
++#define ACM_GETBY_domainid 2
++struct acm_getssid {
++ /* IN */
++ uint32_t interface_version;
++ uint32_t get_ssid_by; /* ACM_GETBY_* */
++ union {
++ domaintype_t domainid;
++ ssidref_t ssidref;
++ } id;
++ XEN_GUEST_HANDLE(void) ssidbuf;
++ uint32_t ssidbuf_size;
++};
++
++#define ACMOP_getdecision 5
++struct acm_getdecision {
++ /* IN */
++ uint32_t interface_version;
++ uint32_t get_decision_by1; /* ACM_GETBY_* */
++ uint32_t get_decision_by2; /* ACM_GETBY_* */
++ union {
++ domaintype_t domainid;
++ ssidref_t ssidref;
++ } id1;
++ union {
++ domaintype_t domainid;
++ ssidref_t ssidref;
++ } id2;
++ uint32_t hook;
++ /* OUT */
++ uint32_t acm_decision;
++};
++
++#endif /* __XEN_PUBLIC_ACM_OPS_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/arch-ia64.h new/include/xen/interface/arch-ia64.h
+--- linux-2.6/include/xen/interface/arch-ia64.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/arch-ia64.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,420 @@
++/******************************************************************************
++ * arch-ia64/hypervisor-if.h
++ *
++ * Guest OS interface to IA64 Xen.
++ */
++
++#ifndef __HYPERVISOR_IF_IA64_H__
++#define __HYPERVISOR_IF_IA64_H__
++
++#define __DEFINE_XEN_GUEST_HANDLE(name, type) \
++ typedef struct { type *p; } __guest_handle_ ## name
++
++#define DEFINE_XEN_GUEST_HANDLE(name) __DEFINE_XEN_GUEST_HANDLE(name, name)
++#define XEN_GUEST_HANDLE(name) __guest_handle_ ## name
++#define set_xen_guest_handle(hnd, val) do { (hnd).p = val; } while (0)
++#ifdef __XEN_TOOLS__
++#define get_xen_guest_handle(val, hnd) do { val = (hnd).p; } while (0)
++#endif
++
++#ifndef __ASSEMBLY__
++/* Guest handles for primitive C types. */
++__DEFINE_XEN_GUEST_HANDLE(uchar, unsigned char);
++__DEFINE_XEN_GUEST_HANDLE(uint, unsigned int);
++__DEFINE_XEN_GUEST_HANDLE(ulong, unsigned long);
++DEFINE_XEN_GUEST_HANDLE(char);
++DEFINE_XEN_GUEST_HANDLE(int);
++DEFINE_XEN_GUEST_HANDLE(long);
++DEFINE_XEN_GUEST_HANDLE(void);
++
++typedef unsigned long xen_pfn_t;
++DEFINE_XEN_GUEST_HANDLE(xen_pfn_t);
++#endif
++
++/* Arch specific VIRQs definition */
++#define VIRQ_ITC VIRQ_ARCH_0 /* V. Virtual itc timer */
++
++/* Maximum number of virtual CPUs in multi-processor guests. */
++/* WARNING: before changing this, check that shared_info fits on a page */
++#define MAX_VIRT_CPUS 64
++
++#ifndef __ASSEMBLY__
++
++typedef unsigned long xen_ulong_t;
++
++#define MAX_NR_SECTION 32 /* at most 32 memory holes */
++struct mm_section {
++ unsigned long start; /* start of memory hole */
++ unsigned long end; /* end of memory hole */
++};
++typedef struct mm_section mm_section_t;
++
++struct pmt_entry {
++ unsigned long mfn : 56;
++ unsigned long type: 8;
++};
++typedef struct pmt_entry pmt_entry_t;
++
++#define GPFN_MEM (0UL << 56) /* Guest pfn is normal mem */
++#define GPFN_FRAME_BUFFER (1UL << 56) /* VGA framebuffer */
++#define GPFN_LOW_MMIO (2UL << 56) /* Low MMIO range */
++#define GPFN_PIB (3UL << 56) /* PIB base */
++#define GPFN_IOSAPIC (4UL << 56) /* IOSAPIC base */
++#define GPFN_LEGACY_IO (5UL << 56) /* Legacy I/O base */
++#define GPFN_GFW (6UL << 56) /* Guest Firmware */
++#define GPFN_HIGH_MMIO (7UL << 56) /* High MMIO range */
++
++#define GPFN_IO_MASK (7UL << 56) /* Guest pfn is I/O type */
++#define GPFN_INV_MASK (31UL << 59) /* Guest pfn is invalid */
++
++#define INVALID_MFN (~0UL)
++
++#define MEM_G (1UL << 30)
++#define MEM_M (1UL << 20)
++
++#define MMIO_START (3 * MEM_G)
++#define MMIO_SIZE (512 * MEM_M)
++
++#define VGA_IO_START 0xA0000UL
++#define VGA_IO_SIZE 0x20000
++
++#define LEGACY_IO_START (MMIO_START + MMIO_SIZE)
++#define LEGACY_IO_SIZE (64*MEM_M)
++
++#define IO_PAGE_START (LEGACY_IO_START + LEGACY_IO_SIZE)
++#define IO_PAGE_SIZE PAGE_SIZE
++
++#define STORE_PAGE_START (IO_PAGE_START + IO_PAGE_SIZE)
++#define STORE_PAGE_SIZE PAGE_SIZE
++
++#define IO_SAPIC_START 0xfec00000UL
++#define IO_SAPIC_SIZE 0x100000
++
++#define PIB_START 0xfee00000UL
++#define PIB_SIZE 0x200000
++
++#define GFW_START (4*MEM_G -16*MEM_M)
++#define GFW_SIZE (16*MEM_M)
++
++/*
++ * NB. This may become a 64-bit count with no shift. If this happens then the
++ * structure size will still be 8 bytes, so no other alignments will change.
++ */
++struct tsc_timestamp {
++ unsigned int tsc_bits; /* 0: 32 bits read from the CPU's TSC. */
++ unsigned int tsc_bitshift; /* 4: 'tsc_bits' uses N:N+31 of TSC. */
++}; /* 8 bytes */
++typedef struct tsc_timestamp tsc_timestamp_t;
++
++struct pt_fpreg {
++ union {
++ unsigned long bits[2];
++ long double __dummy; /* force 16-byte alignment */
++ } u;
++};
++
++struct cpu_user_regs {
++ /* The following registers are saved by SAVE_MIN: */
++ unsigned long b6; /* scratch */
++ unsigned long b7; /* scratch */
++
++ unsigned long ar_csd; /* used by cmp8xchg16 (scratch) */
++ unsigned long ar_ssd; /* reserved for future use (scratch) */
++
++ unsigned long r8; /* scratch (return value register 0) */
++ unsigned long r9; /* scratch (return value register 1) */
++ unsigned long r10; /* scratch (return value register 2) */
++ unsigned long r11; /* scratch (return value register 3) */
++
++ unsigned long cr_ipsr; /* interrupted task's psr */
++ unsigned long cr_iip; /* interrupted task's instruction pointer */
++ unsigned long cr_ifs; /* interrupted task's function state */
++
++ unsigned long ar_unat; /* interrupted task's NaT register (preserved) */
++ unsigned long ar_pfs; /* prev function state */
++ unsigned long ar_rsc; /* RSE configuration */
++ /* The following two are valid only if cr_ipsr.cpl > 0: */
++ unsigned long ar_rnat; /* RSE NaT */
++ unsigned long ar_bspstore; /* RSE bspstore */
++
++ unsigned long pr; /* 64 predicate registers (1 bit each) */
++ unsigned long b0; /* return pointer (bp) */
++ unsigned long loadrs; /* size of dirty partition << 16 */
++
++ unsigned long r1; /* the gp pointer */
++ unsigned long r12; /* interrupted task's memory stack pointer */
++ unsigned long r13; /* thread pointer */
++
++ unsigned long ar_fpsr; /* floating point status (preserved) */
++ unsigned long r15; /* scratch */
++
++ /* The remaining registers are NOT saved for system calls. */
++
++ unsigned long r14; /* scratch */
++ unsigned long r2; /* scratch */
++ unsigned long r3; /* scratch */
++ unsigned long r16; /* scratch */
++ unsigned long r17; /* scratch */
++ unsigned long r18; /* scratch */
++ unsigned long r19; /* scratch */
++ unsigned long r20; /* scratch */
++ unsigned long r21; /* scratch */
++ unsigned long r22; /* scratch */
++ unsigned long r23; /* scratch */
++ unsigned long r24; /* scratch */
++ unsigned long r25; /* scratch */
++ unsigned long r26; /* scratch */
++ unsigned long r27; /* scratch */
++ unsigned long r28; /* scratch */
++ unsigned long r29; /* scratch */
++ unsigned long r30; /* scratch */
++ unsigned long r31; /* scratch */
++ unsigned long ar_ccv; /* compare/exchange value (scratch) */
++
++ /*
++ * Floating point registers that the kernel considers scratch:
++ */
++ struct pt_fpreg f6; /* scratch */
++ struct pt_fpreg f7; /* scratch */
++ struct pt_fpreg f8; /* scratch */
++ struct pt_fpreg f9; /* scratch */
++ struct pt_fpreg f10; /* scratch */
++ struct pt_fpreg f11; /* scratch */
++ unsigned long r4; /* preserved */
++ unsigned long r5; /* preserved */
++ unsigned long r6; /* preserved */
++ unsigned long r7; /* preserved */
++ unsigned long eml_unat; /* used for emulating instruction */
++ unsigned long rfi_pfs; /* used for elulating rfi */
++
++};
++typedef struct cpu_user_regs cpu_user_regs_t;
++
++union vac {
++ unsigned long value;
++ struct {
++ int a_int:1;
++ int a_from_int_cr:1;
++ int a_to_int_cr:1;
++ int a_from_psr:1;
++ int a_from_cpuid:1;
++ int a_cover:1;
++ int a_bsw:1;
++ long reserved:57;
++ };
++};
++typedef union vac vac_t;
++
++union vdc {
++ unsigned long value;
++ struct {
++ int d_vmsw:1;
++ int d_extint:1;
++ int d_ibr_dbr:1;
++ int d_pmc:1;
++ int d_to_pmd:1;
++ int d_itm:1;
++ long reserved:58;
++ };
++};
++typedef union vdc vdc_t;
++
++struct mapped_regs {
++ union vac vac;
++ union vdc vdc;
++ unsigned long virt_env_vaddr;
++ unsigned long reserved1[29];
++ unsigned long vhpi;
++ unsigned long reserved2[95];
++ union {
++ unsigned long vgr[16];
++ unsigned long bank1_regs[16]; // bank1 regs (r16-r31) when bank0 active
++ };
++ union {
++ unsigned long vbgr[16];
++ unsigned long bank0_regs[16]; // bank0 regs (r16-r31) when bank1 active
++ };
++ unsigned long vnat;
++ unsigned long vbnat;
++ unsigned long vcpuid[5];
++ unsigned long reserved3[11];
++ unsigned long vpsr;
++ unsigned long vpr;
++ unsigned long reserved4[76];
++ union {
++ unsigned long vcr[128];
++ struct {
++ unsigned long dcr; // CR0
++ unsigned long itm;
++ unsigned long iva;
++ unsigned long rsv1[5];
++ unsigned long pta; // CR8
++ unsigned long rsv2[7];
++ unsigned long ipsr; // CR16
++ unsigned long isr;
++ unsigned long rsv3;
++ unsigned long iip;
++ unsigned long ifa;
++ unsigned long itir;
++ unsigned long iipa;
++ unsigned long ifs;
++ unsigned long iim; // CR24
++ unsigned long iha;
++ unsigned long rsv4[38];
++ unsigned long lid; // CR64
++ unsigned long ivr;
++ unsigned long tpr;
++ unsigned long eoi;
++ unsigned long irr[4];
++ unsigned long itv; // CR72
++ unsigned long pmv;
++ unsigned long cmcv;
++ unsigned long rsv5[5];
++ unsigned long lrr0; // CR80
++ unsigned long lrr1;
++ unsigned long rsv6[46];
++ };
++ };
++ union {
++ unsigned long reserved5[128];
++ struct {
++ unsigned long precover_ifs;
++ unsigned long unat; // not sure if this is needed until NaT arch is done
++ int interrupt_collection_enabled; // virtual psr.ic
++ /* virtual interrupt deliverable flag is evtchn_upcall_mask in
++ * shared info area now. interrupt_mask_addr is the address
++ * of evtchn_upcall_mask for current vcpu
++ */
++ unsigned long interrupt_mask_addr;
++ int pending_interruption;
++ int incomplete_regframe; // see SDM vol2 6.8
++ unsigned long reserved5_1[4];
++ int metaphysical_mode; // 1 = use metaphys mapping, 0 = use virtual
++ int banknum; // 0 or 1, which virtual register bank is active
++ unsigned long rrs[8]; // region registers
++ unsigned long krs[8]; // kernel registers
++ unsigned long pkrs[8]; // protection key registers
++ unsigned long tmp[8]; // temp registers (e.g. for hyperprivops)
++ // FIXME: tmp[8] temp'ly being used for virtual psr.pp
++ };
++ };
++ unsigned long reserved6[3456];
++ unsigned long vmm_avail[128];
++ unsigned long reserved7[4096];
++};
++typedef struct mapped_regs mapped_regs_t;
++
++struct arch_vcpu_info {
++};
++typedef struct arch_vcpu_info arch_vcpu_info_t;
++
++typedef mapped_regs_t vpd_t;
++
++struct arch_shared_info {
++ unsigned int flags;
++ unsigned long start_info_pfn;
++
++ /* Interrupt vector for event channel. */
++ int evtchn_vector;
++};
++typedef struct arch_shared_info arch_shared_info_t;
++
++struct arch_initrd_info {
++ unsigned long start;
++ unsigned long size;
++};
++typedef struct arch_initrd_info arch_initrd_info_t;
++
++typedef unsigned long xen_callback_t;
++
++#define IA64_COMMAND_LINE_SIZE 512
++struct vcpu_guest_context {
++#define VGCF_FPU_VALID (1<<0)
++#define VGCF_VMX_GUEST (1<<1)
++#define VGCF_IN_KERNEL (1<<2)
++ unsigned long flags; /* VGCF_* flags */
++ unsigned long pt_base; /* PMT table base */
++ unsigned long share_io_pg; /* Shared page for I/O emulation */
++ unsigned long sys_pgnr; /* System pages out of domain memory */
++ unsigned long vm_assist; /* VMASST_TYPE_* bitmap, now none on IPF */
++
++ struct cpu_user_regs regs;
++ struct mapped_regs *privregs;
++ struct arch_shared_info shared;
++ struct arch_initrd_info initrd;
++ char cmdline[IA64_COMMAND_LINE_SIZE];
++};
++typedef struct vcpu_guest_context vcpu_guest_context_t;
++DEFINE_XEN_GUEST_HANDLE(vcpu_guest_context_t);
++
++// dom0 vp op
++#define __HYPERVISOR_ia64_dom0vp_op __HYPERVISOR_arch_0
++#define IA64_DOM0VP_ioremap 0 // map io space in machine
++ // address to dom0 physical
++ // address space.
++ // currently physical
++ // assignedg address equals to
++ // machine address
++#define IA64_DOM0VP_phystomach 1 // convert a pseudo physical
++ // page frame number
++ // to the corresponding
++ // machine page frame number.
++ // if no page is assigned,
++ // INVALID_MFN or GPFN_INV_MASK
++ // is returned depending on
++ // domain's non-vti/vti mode.
++#define IA64_DOM0VP_machtophys 3 // convert a machine page
++ // frame number
++ // to the corresponding
++ // pseudo physical page frame
++ // number of the caller domain
++#define IA64_DOM0VP_zap_physmap 17 // unmap and free pages
++ // contained in the specified
++ // pseudo physical region
++#define IA64_DOM0VP_add_physmap 18 // assigne machine page frane
++ // to dom0's pseudo physical
++ // address space.
++// flags for page assignement to pseudo physical address space
++#define _ASSIGN_readonly 0
++#define ASSIGN_readonly (1UL << _ASSIGN_readonly)
++#define ASSIGN_writable (0UL << _ASSIGN_readonly) // dummy flag
++
++#endif /* !__ASSEMBLY__ */
++
++/* Hyperprivops. */
++#define HYPERPRIVOP_RFI 0x1
++#define HYPERPRIVOP_RSM_DT 0x2
++#define HYPERPRIVOP_SSM_DT 0x3
++#define HYPERPRIVOP_COVER 0x4
++#define HYPERPRIVOP_ITC_D 0x5
++#define HYPERPRIVOP_ITC_I 0x6
++#define HYPERPRIVOP_SSM_I 0x7
++#define HYPERPRIVOP_GET_IVR 0x8
++#define HYPERPRIVOP_GET_TPR 0x9
++#define HYPERPRIVOP_SET_TPR 0xa
++#define HYPERPRIVOP_EOI 0xb
++#define HYPERPRIVOP_SET_ITM 0xc
++#define HYPERPRIVOP_THASH 0xd
++#define HYPERPRIVOP_PTC_GA 0xe
++#define HYPERPRIVOP_ITR_D 0xf
++#define HYPERPRIVOP_GET_RR 0x10
++#define HYPERPRIVOP_SET_RR 0x11
++#define HYPERPRIVOP_SET_KR 0x12
++#define HYPERPRIVOP_FC 0x13
++#define HYPERPRIVOP_GET_CPUID 0x14
++#define HYPERPRIVOP_GET_PMD 0x15
++#define HYPERPRIVOP_GET_EFLAG 0x16
++#define HYPERPRIVOP_SET_EFLAG 0x17
++#define HYPERPRIVOP_MAX 0x17
++
++#endif /* __HYPERVISOR_IF_IA64_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/arch-x86_32.h new/include/xen/interface/arch-x86_32.h
+--- linux-2.6/include/xen/interface/arch-x86_32.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/arch-x86_32.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,233 @@
++/******************************************************************************
++ * arch-x86_32.h
++ *
++ * Guest OS interface to x86 32-bit Xen.
++ *
++ * Copyright (c) 2004, K A Fraser
++ */
++
++#ifndef __XEN_PUBLIC_ARCH_X86_32_H__
++#define __XEN_PUBLIC_ARCH_X86_32_H__
++
++#define __DEFINE_XEN_GUEST_HANDLE(name, type) \
++ typedef struct { type *p; } __guest_handle_ ## name
++
++#define DEFINE_XEN_GUEST_HANDLE(name) __DEFINE_XEN_GUEST_HANDLE(name, name)
++#define XEN_GUEST_HANDLE(name) __guest_handle_ ## name
++#define set_xen_guest_handle(hnd, val) do { (hnd).p = val; } while (0)
++#ifdef __XEN_TOOLS__
++#define get_xen_guest_handle(val, hnd) do { val = (hnd).p; } while (0)
++#endif
++
++#ifndef __ASSEMBLY__
++/* Guest handles for primitive C types. */
++__DEFINE_XEN_GUEST_HANDLE(uchar, unsigned char);
++__DEFINE_XEN_GUEST_HANDLE(uint, unsigned int);
++__DEFINE_XEN_GUEST_HANDLE(ulong, unsigned long);
++DEFINE_XEN_GUEST_HANDLE(char);
++DEFINE_XEN_GUEST_HANDLE(int);
++DEFINE_XEN_GUEST_HANDLE(long);
++DEFINE_XEN_GUEST_HANDLE(void);
++
++typedef unsigned long xen_pfn_t;
++DEFINE_XEN_GUEST_HANDLE(xen_pfn_t);
++#endif
++
++/*
++ * SEGMENT DESCRIPTOR TABLES
++ */
++/*
++ * A number of GDT entries are reserved by Xen. These are not situated at the
++ * start of the GDT because some stupid OSes export hard-coded selector values
++ * in their ABI. These hard-coded values are always near the start of the GDT,
++ * so Xen places itself out of the way, at the far end of the GDT.
++ */
++#define FIRST_RESERVED_GDT_PAGE 14
++#define FIRST_RESERVED_GDT_BYTE (FIRST_RESERVED_GDT_PAGE * 4096)
++#define FIRST_RESERVED_GDT_ENTRY (FIRST_RESERVED_GDT_BYTE / 8)
++
++/*
++ * These flat segments are in the Xen-private section of every GDT. Since these
++ * are also present in the initial GDT, many OSes will be able to avoid
++ * installing their own GDT.
++ */
++#define FLAT_RING1_CS 0xe019 /* GDT index 259 */
++#define FLAT_RING1_DS 0xe021 /* GDT index 260 */
++#define FLAT_RING1_SS 0xe021 /* GDT index 260 */
++#define FLAT_RING3_CS 0xe02b /* GDT index 261 */
++#define FLAT_RING3_DS 0xe033 /* GDT index 262 */
++#define FLAT_RING3_SS 0xe033 /* GDT index 262 */
++
++#define FLAT_KERNEL_CS FLAT_RING1_CS
++#define FLAT_KERNEL_DS FLAT_RING1_DS
++#define FLAT_KERNEL_SS FLAT_RING1_SS
++#define FLAT_USER_CS FLAT_RING3_CS
++#define FLAT_USER_DS FLAT_RING3_DS
++#define FLAT_USER_SS FLAT_RING3_SS
++
++/* And the trap vector is... */
++#define TRAP_INSTR "int $0x82"
++
++/*
++ * Virtual addresses beyond this are not modifiable by guest OSes. The
++ * machine->physical mapping table starts at this address, read-only.
++ */
++#ifdef CONFIG_X86_PAE
++#define __HYPERVISOR_VIRT_START 0xF5800000
++#define __MACH2PHYS_VIRT_START 0xF5800000
++#define __MACH2PHYS_VIRT_END 0xF6800000
++#else
++#define __HYPERVISOR_VIRT_START 0xFC000000
++#define __MACH2PHYS_VIRT_START 0xFC000000
++#define __MACH2PHYS_VIRT_END 0xFC400000
++#endif
++
++#ifndef HYPERVISOR_VIRT_START
++#define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START)
++#endif
++
++#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START)
++#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END)
++#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>2)
++#ifndef machine_to_phys_mapping
++#define machine_to_phys_mapping ((unsigned long *)MACH2PHYS_VIRT_START)
++#endif
++
++/* Maximum number of virtual CPUs in multi-processor guests. */
++#define MAX_VIRT_CPUS 32
++
++#ifndef __ASSEMBLY__
++
++typedef unsigned long xen_ulong_t;
++
++/*
++ * Send an array of these to HYPERVISOR_set_trap_table()
++ */
++#define TI_GET_DPL(_ti) ((_ti)->flags & 3)
++#define TI_GET_IF(_ti) ((_ti)->flags & 4)
++#define TI_SET_DPL(_ti,_dpl) ((_ti)->flags |= (_dpl))
++#define TI_SET_IF(_ti,_if) ((_ti)->flags |= ((!!(_if))<<2))
++struct trap_info {
++ uint8_t vector; /* exception vector */
++ uint8_t flags; /* 0-3: privilege level; 4: clear event enable? */
++ uint16_t cs; /* code selector */
++ unsigned long address; /* code offset */
++};
++typedef struct trap_info trap_info_t;
++DEFINE_XEN_GUEST_HANDLE(trap_info_t);
++
++struct cpu_user_regs {
++ uint32_t ebx;
++ uint32_t ecx;
++ uint32_t edx;
++ uint32_t esi;
++ uint32_t edi;
++ uint32_t ebp;
++ uint32_t eax;
++ uint16_t error_code; /* private */
++ uint16_t entry_vector; /* private */
++ uint32_t eip;
++ uint16_t cs;
++ uint8_t saved_upcall_mask;
++ uint8_t _pad0;
++ uint32_t eflags; /* eflags.IF == !saved_upcall_mask */
++ uint32_t esp;
++ uint16_t ss, _pad1;
++ uint16_t es, _pad2;
++ uint16_t ds, _pad3;
++ uint16_t fs, _pad4;
++ uint16_t gs, _pad5;
++};
++typedef struct cpu_user_regs cpu_user_regs_t;
++DEFINE_XEN_GUEST_HANDLE(cpu_user_regs_t);
++
++typedef uint64_t tsc_timestamp_t; /* RDTSC timestamp */
++
++/*
++ * The following is all CPU context. Note that the fpu_ctxt block is filled
++ * in by FXSAVE if the CPU has feature FXSR; otherwise FSAVE is used.
++ */
++struct vcpu_guest_context {
++ /* FPU registers come first so they can be aligned for FXSAVE/FXRSTOR. */
++ struct { char x[512]; } fpu_ctxt; /* User-level FPU registers */
++#define VGCF_I387_VALID (1<<0)
++#define VGCF_HVM_GUEST (1<<1)
++#define VGCF_IN_KERNEL (1<<2)
++#define _VGCF_i387_valid 0
++#define VGCF_i387_valid (1<<_VGCF_i387_valid)
++#define _VGCF_hvm_guest 1
++#define VGCF_hvm_guest (1<<_VGCF_hvm_guest)
++#define _VGCF_in_kernel 2
++#define VGCF_in_kernel (1<<_VGCF_in_kernel)
++#define _VGCF_failsafe_disables_events 3
++#define VGCF_failsafe_disables_events (1<<_VGCF_failsafe_disables_events)
++ unsigned long flags; /* VGCF_* flags */
++ struct cpu_user_regs user_regs; /* User-level CPU registers */
++ struct trap_info trap_ctxt[256]; /* Virtual IDT */
++ unsigned long ldt_base, ldt_ents; /* LDT (linear address, # ents) */
++ unsigned long gdt_frames[16], gdt_ents; /* GDT (machine frames, # ents) */
++ unsigned long kernel_ss, kernel_sp; /* Virtual TSS (only SS1/SP1) */
++ unsigned long ctrlreg[8]; /* CR0-CR7 (control registers) */
++ unsigned long debugreg[8]; /* DB0-DB7 (debug registers) */
++ unsigned long event_callback_cs; /* CS:EIP of event callback */
++ unsigned long event_callback_eip;
++ unsigned long failsafe_callback_cs; /* CS:EIP of failsafe callback */
++ unsigned long failsafe_callback_eip;
++ unsigned long vm_assist; /* VMASST_TYPE_* bitmap */
++};
++typedef struct vcpu_guest_context vcpu_guest_context_t;
++DEFINE_XEN_GUEST_HANDLE(vcpu_guest_context_t);
++
++/*
++ * Page-directory addresses above 4GB do not fit into architectural %cr3.
++ * When accessing %cr3, or equivalent field in vcpu_guest_context, guests
++ * must use the following accessor macros to pack/unpack valid MFNs.
++ */
++#define xen_pfn_to_cr3(pfn) (((unsigned)(pfn) << 12) | ((unsigned)(pfn) >> 20))
++#define xen_cr3_to_pfn(cr3) (((unsigned)(cr3) >> 12) | ((unsigned)(cr3) << 20))
++
++struct arch_shared_info {
++ unsigned long max_pfn; /* max pfn that appears in table */
++ /* Frame containing list of mfns containing list of mfns containing p2m. */
++ xen_pfn_t pfn_to_mfn_frame_list_list;
++ unsigned long nmi_reason;
++};
++typedef struct arch_shared_info arch_shared_info_t;
++
++struct arch_vcpu_info {
++ unsigned long cr2;
++ unsigned long pad[5]; /* sizeof(vcpu_info_t) == 64 */
++};
++typedef struct arch_vcpu_info arch_vcpu_info_t;
++
++struct xen_callback {
++ unsigned long cs;
++ unsigned long eip;
++};
++typedef struct xen_callback xen_callback_t;
++
++#endif /* !__ASSEMBLY__ */
++
++/*
++ * Prefix forces emulation of some non-trapping instructions.
++ * Currently only CPUID.
++ */
++#ifdef __ASSEMBLY__
++#define XEN_EMULATE_PREFIX .byte 0x0f,0x0b,0x78,0x65,0x6e ;
++#define XEN_CPUID XEN_EMULATE_PREFIX cpuid
++#else
++#define XEN_EMULATE_PREFIX ".byte 0x0f,0x0b,0x78,0x65,0x6e ; "
++#define XEN_CPUID XEN_EMULATE_PREFIX "cpuid"
++#endif
++
++#endif
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/arch-x86_64.h new/include/xen/interface/arch-x86_64.h
+--- linux-2.6/include/xen/interface/arch-x86_64.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/arch-x86_64.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,299 @@
++/******************************************************************************
++ * arch-x86_64.h
++ *
++ * Guest OS interface to x86 64-bit Xen.
++ *
++ * Copyright (c) 2004, K A Fraser
++ */
++
++#ifndef __XEN_PUBLIC_ARCH_X86_64_H__
++#define __XEN_PUBLIC_ARCH_X86_64_H__
++
++#define __DEFINE_XEN_GUEST_HANDLE(name, type) \
++ typedef struct { type *p; } __guest_handle_ ## name
++
++#define DEFINE_XEN_GUEST_HANDLE(name) __DEFINE_XEN_GUEST_HANDLE(name, name)
++#define XEN_GUEST_HANDLE(name) __guest_handle_ ## name
++#define set_xen_guest_handle(hnd, val) do { (hnd).p = val; } while (0)
++#ifdef __XEN_TOOLS__
++#define get_xen_guest_handle(val, hnd) do { val = (hnd).p; } while (0)
++#endif
++
++#ifndef __ASSEMBLY__
++/* Guest handles for primitive C types. */
++__DEFINE_XEN_GUEST_HANDLE(uchar, unsigned char);
++__DEFINE_XEN_GUEST_HANDLE(uint, unsigned int);
++__DEFINE_XEN_GUEST_HANDLE(ulong, unsigned long);
++DEFINE_XEN_GUEST_HANDLE(char);
++DEFINE_XEN_GUEST_HANDLE(int);
++DEFINE_XEN_GUEST_HANDLE(long);
++DEFINE_XEN_GUEST_HANDLE(void);
++
++typedef unsigned long xen_pfn_t;
++DEFINE_XEN_GUEST_HANDLE(xen_pfn_t);
++#endif
++
++/*
++ * SEGMENT DESCRIPTOR TABLES
++ */
++/*
++ * A number of GDT entries are reserved by Xen. These are not situated at the
++ * start of the GDT because some stupid OSes export hard-coded selector values
++ * in their ABI. These hard-coded values are always near the start of the GDT,
++ * so Xen places itself out of the way, at the far end of the GDT.
++ */
++#define FIRST_RESERVED_GDT_PAGE 14
++#define FIRST_RESERVED_GDT_BYTE (FIRST_RESERVED_GDT_PAGE * 4096)
++#define FIRST_RESERVED_GDT_ENTRY (FIRST_RESERVED_GDT_BYTE / 8)
++
++/*
++ * 64-bit segment selectors
++ * These flat segments are in the Xen-private section of every GDT. Since these
++ * are also present in the initial GDT, many OSes will be able to avoid
++ * installing their own GDT.
++ */
++
++#define FLAT_RING3_CS32 0xe023 /* GDT index 260 */
++#define FLAT_RING3_CS64 0xe033 /* GDT index 261 */
++#define FLAT_RING3_DS32 0xe02b /* GDT index 262 */
++#define FLAT_RING3_DS64 0x0000 /* NULL selector */
++#define FLAT_RING3_SS32 0xe02b /* GDT index 262 */
++#define FLAT_RING3_SS64 0xe02b /* GDT index 262 */
++
++#define FLAT_KERNEL_DS64 FLAT_RING3_DS64
++#define FLAT_KERNEL_DS32 FLAT_RING3_DS32
++#define FLAT_KERNEL_DS FLAT_KERNEL_DS64
++#define FLAT_KERNEL_CS64 FLAT_RING3_CS64
++#define FLAT_KERNEL_CS32 FLAT_RING3_CS32
++#define FLAT_KERNEL_CS FLAT_KERNEL_CS64
++#define FLAT_KERNEL_SS64 FLAT_RING3_SS64
++#define FLAT_KERNEL_SS32 FLAT_RING3_SS32
++#define FLAT_KERNEL_SS FLAT_KERNEL_SS64
++
++#define FLAT_USER_DS64 FLAT_RING3_DS64
++#define FLAT_USER_DS32 FLAT_RING3_DS32
++#define FLAT_USER_DS FLAT_USER_DS64
++#define FLAT_USER_CS64 FLAT_RING3_CS64
++#define FLAT_USER_CS32 FLAT_RING3_CS32
++#define FLAT_USER_CS FLAT_USER_CS64
++#define FLAT_USER_SS64 FLAT_RING3_SS64
++#define FLAT_USER_SS32 FLAT_RING3_SS32
++#define FLAT_USER_SS FLAT_USER_SS64
++
++/* And the trap vector is... */
++#define TRAP_INSTR "syscall"
++
++#define __HYPERVISOR_VIRT_START 0xFFFF800000000000
++#define __HYPERVISOR_VIRT_END 0xFFFF880000000000
++#define __MACH2PHYS_VIRT_START 0xFFFF800000000000
++#define __MACH2PHYS_VIRT_END 0xFFFF804000000000
++
++#ifndef HYPERVISOR_VIRT_START
++#define HYPERVISOR_VIRT_START mk_unsigned_long(__HYPERVISOR_VIRT_START)
++#define HYPERVISOR_VIRT_END mk_unsigned_long(__HYPERVISOR_VIRT_END)
++#endif
++
++#define MACH2PHYS_VIRT_START mk_unsigned_long(__MACH2PHYS_VIRT_START)
++#define MACH2PHYS_VIRT_END mk_unsigned_long(__MACH2PHYS_VIRT_END)
++#define MACH2PHYS_NR_ENTRIES ((MACH2PHYS_VIRT_END-MACH2PHYS_VIRT_START)>>3)
++#ifndef machine_to_phys_mapping
++#define machine_to_phys_mapping ((unsigned long *)HYPERVISOR_VIRT_START)
++#endif
++
++/* Maximum number of virtual CPUs in multi-processor guests. */
++#define MAX_VIRT_CPUS 32
++
++#ifndef __ASSEMBLY__
++
++typedef unsigned long xen_ulong_t;
++
++/*
++ * int HYPERVISOR_set_segment_base(unsigned int which, unsigned long base)
++ * @which == SEGBASE_* ; @base == 64-bit base address
++ * Returns 0 on success.
++ */
++#define SEGBASE_FS 0
++#define SEGBASE_GS_USER 1
++#define SEGBASE_GS_KERNEL 2
++#define SEGBASE_GS_USER_SEL 3 /* Set user %gs specified in base[15:0] */
++
++/*
++ * int HYPERVISOR_iret(void)
++ * All arguments are on the kernel stack, in the following format.
++ * Never returns if successful. Current kernel context is lost.
++ * The saved CS is mapped as follows:
++ * RING0 -> RING3 kernel mode.
++ * RING1 -> RING3 kernel mode.
++ * RING2 -> RING3 kernel mode.
++ * RING3 -> RING3 user mode.
++ * However RING0 indicates that the guest kernel should return to iteself
++ * directly with
++ * orb $3,1*8(%rsp)
++ * iretq
++ * If flags contains VGCF_IN_SYSCALL:
++ * Restore RAX, RIP, RFLAGS, RSP.
++ * Discard R11, RCX, CS, SS.
++ * Otherwise:
++ * Restore RAX, R11, RCX, CS:RIP, RFLAGS, SS:RSP.
++ * All other registers are saved on hypercall entry and restored to user.
++ */
++/* Guest exited in SYSCALL context? Return to guest with SYSRET? */
++#define VGCF_IN_SYSCALL (1<<8)
++struct iret_context {
++ /* Top of stack (%rsp at point of hypercall). */
++ uint64_t rax, r11, rcx, flags, rip, cs, rflags, rsp, ss;
++ /* Bottom of iret stack frame. */
++};
++
++/*
++ * Send an array of these to HYPERVISOR_set_trap_table().
++ * N.B. As in x86/32 mode, the privilege level specifies which modes may enter
++ * a trap via a software interrupt. Since rings 1 and 2 are unavailable, we
++ * allocate privilege levels as follows:
++ * Level == 0: Noone may enter
++ * Level == 1: Kernel may enter
++ * Level == 2: Kernel may enter
++ * Level == 3: Everyone may enter
++ */
++#define TI_GET_DPL(_ti) ((_ti)->flags & 3)
++#define TI_GET_IF(_ti) ((_ti)->flags & 4)
++#define TI_SET_DPL(_ti,_dpl) ((_ti)->flags |= (_dpl))
++#define TI_SET_IF(_ti,_if) ((_ti)->flags |= ((!!(_if))<<2))
++struct trap_info {
++ uint8_t vector; /* exception vector */
++ uint8_t flags; /* 0-3: privilege level; 4: clear event enable? */
++ uint16_t cs; /* code selector */
++ unsigned long address; /* code offset */
++};
++typedef struct trap_info trap_info_t;
++DEFINE_XEN_GUEST_HANDLE(trap_info_t);
++
++#ifdef __GNUC__
++/* Anonymous union includes both 32- and 64-bit names (e.g., eax/rax). */
++#define __DECL_REG(name) union { uint64_t r ## name, e ## name; }
++#else
++/* Non-gcc sources must always use the proper 64-bit name (e.g., rax). */
++#define __DECL_REG(name) uint64_t r ## name
++#endif
++
++struct cpu_user_regs {
++ uint64_t r15;
++ uint64_t r14;
++ uint64_t r13;
++ uint64_t r12;
++ __DECL_REG(bp);
++ __DECL_REG(bx);
++ uint64_t r11;
++ uint64_t r10;
++ uint64_t r9;
++ uint64_t r8;
++ __DECL_REG(ax);
++ __DECL_REG(cx);
++ __DECL_REG(dx);
++ __DECL_REG(si);
++ __DECL_REG(di);
++ uint32_t error_code; /* private */
++ uint32_t entry_vector; /* private */
++ __DECL_REG(ip);
++ uint16_t cs, _pad0[1];
++ uint8_t saved_upcall_mask;
++ uint8_t _pad1[3];
++ __DECL_REG(flags); /* rflags.IF == !saved_upcall_mask */
++ __DECL_REG(sp);
++ uint16_t ss, _pad2[3];
++ uint16_t es, _pad3[3];
++ uint16_t ds, _pad4[3];
++ uint16_t fs, _pad5[3]; /* Non-zero => takes precedence over fs_base. */
++ uint16_t gs, _pad6[3]; /* Non-zero => takes precedence over gs_base_usr. */
++};
++typedef struct cpu_user_regs cpu_user_regs_t;
++DEFINE_XEN_GUEST_HANDLE(cpu_user_regs_t);
++
++#undef __DECL_REG
++
++typedef uint64_t tsc_timestamp_t; /* RDTSC timestamp */
++
++/*
++ * The following is all CPU context. Note that the fpu_ctxt block is filled
++ * in by FXSAVE if the CPU has feature FXSR; otherwise FSAVE is used.
++ */
++struct vcpu_guest_context {
++ /* FPU registers come first so they can be aligned for FXSAVE/FXRSTOR. */
++ struct { char x[512]; } fpu_ctxt; /* User-level FPU registers */
++#define VGCF_I387_VALID (1<<0)
++#define VGCF_HVM_GUEST (1<<1)
++#define VGCF_IN_KERNEL (1<<2)
++#define _VGCF_i387_valid 0
++#define VGCF_i387_valid (1<<_VGCF_i387_valid)
++#define _VGCF_hvm_guest 1
++#define VGCF_hvm_guest (1<<_VGCF_hvm_guest)
++#define _VGCF_in_kernel 2
++#define VGCF_in_kernel (1<<_VGCF_in_kernel)
++#define _VGCF_failsafe_disables_events 3
++#define VGCF_failsafe_disables_events (1<<_VGCF_failsafe_disables_events)
++#define _VGCF_syscall_disables_events 4
++#define VGCF_syscall_disables_events (1<<_VGCF_syscall_disables_events)
++ unsigned long flags; /* VGCF_* flags */
++ struct cpu_user_regs user_regs; /* User-level CPU registers */
++ struct trap_info trap_ctxt[256]; /* Virtual IDT */
++ unsigned long ldt_base, ldt_ents; /* LDT (linear address, # ents) */
++ unsigned long gdt_frames[16], gdt_ents; /* GDT (machine frames, # ents) */
++ unsigned long kernel_ss, kernel_sp; /* Virtual TSS (only SS1/SP1) */
++ unsigned long ctrlreg[8]; /* CR0-CR7 (control registers) */
++ unsigned long debugreg[8]; /* DB0-DB7 (debug registers) */
++ unsigned long event_callback_eip;
++ unsigned long failsafe_callback_eip;
++ unsigned long syscall_callback_eip;
++ unsigned long vm_assist; /* VMASST_TYPE_* bitmap */
++ /* Segment base addresses. */
++ uint64_t fs_base;
++ uint64_t gs_base_kernel;
++ uint64_t gs_base_user;
++};
++typedef struct vcpu_guest_context vcpu_guest_context_t;
++DEFINE_XEN_GUEST_HANDLE(vcpu_guest_context_t);
++
++#define xen_pfn_to_cr3(pfn) ((unsigned long)(pfn) << 12)
++#define xen_cr3_to_pfn(cr3) ((unsigned long)(cr3) >> 12)
++
++struct arch_shared_info {
++ unsigned long max_pfn; /* max pfn that appears in table */
++ /* Frame containing list of mfns containing list of mfns containing p2m. */
++ xen_pfn_t pfn_to_mfn_frame_list_list;
++ unsigned long nmi_reason;
++};
++typedef struct arch_shared_info arch_shared_info_t;
++
++struct arch_vcpu_info {
++ unsigned long cr2;
++ unsigned long pad; /* sizeof(vcpu_info_t) == 64 */
++};
++typedef struct arch_vcpu_info arch_vcpu_info_t;
++
++typedef unsigned long xen_callback_t;
++
++#endif /* !__ASSEMBLY__ */
++
++/*
++ * Prefix forces emulation of some non-trapping instructions.
++ * Currently only CPUID.
++ */
++#ifdef __ASSEMBLY__
++#define XEN_EMULATE_PREFIX .byte 0x0f,0x0b,0x78,0x65,0x6e ;
++#define XEN_CPUID XEN_EMULATE_PREFIX cpuid
++#else
++#define XEN_EMULATE_PREFIX ".byte 0x0f,0x0b,0x78,0x65,0x6e ; "
++#define XEN_CPUID XEN_EMULATE_PREFIX "cpuid"
++#endif
++
++#endif
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/callback.h new/include/xen/interface/callback.h
+--- linux-2.6/include/xen/interface/callback.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/callback.h 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,74 @@
++/******************************************************************************
++ * callback.h
++ *
++ * Register guest OS callbacks with Xen.
++ *
++ * Copyright (c) 2006, Ian Campbell
++ */
++
++#ifndef __XEN_PUBLIC_CALLBACK_H__
++#define __XEN_PUBLIC_CALLBACK_H__
++
++#include "xen.h"
++
++/*
++ * Prototype for this hypercall is:
++ * long callback_op(int cmd, void *extra_args)
++ * @cmd == CALLBACKOP_??? (callback operation).
++ * @extra_args == Operation-specific extra arguments (NULL if none).
++ */
++
++#define CALLBACKTYPE_event 0
++#define CALLBACKTYPE_failsafe 1
++#define CALLBACKTYPE_syscall 2 /* x86_64 only */
++/*
++ * sysenter is only available on x86_32 with the
++ * supervisor_mode_kernel option enabled.
++ */
++#define CALLBACKTYPE_sysenter 3
++#define CALLBACKTYPE_nmi 4
++
++/*
++ * Disable event deliver during callback? This flag is ignored for event and
++ * NMI callbacks: event delivery is unconditionally disabled.
++ */
++#define _CALLBACKF_mask_events 0
++#define CALLBACKF_mask_events (1U << _CALLBACKF_mask_events)
++
++/*
++ * Register a callback.
++ */
++#define CALLBACKOP_register 0
++struct callback_register {
++ uint16_t type;
++ uint16_t flags;
++ xen_callback_t address;
++};
++typedef struct callback_register callback_register_t;
++DEFINE_XEN_GUEST_HANDLE(callback_register_t);
++
++/*
++ * Unregister a callback.
++ *
++ * Not all callbacks can be unregistered. -EINVAL will be returned if
++ * you attempt to unregister such a callback.
++ */
++#define CALLBACKOP_unregister 1
++struct callback_unregister {
++ uint16_t type;
++ uint16_t _unused;
++};
++typedef struct callback_unregister callback_unregister_t;
++DEFINE_XEN_GUEST_HANDLE(callback_unregister_t);
++
++#endif /* __XEN_PUBLIC_CALLBACK_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/dom0_ops.h new/include/xen/interface/dom0_ops.h
+--- linux-2.6/include/xen/interface/dom0_ops.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/dom0_ops.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,598 @@
++/******************************************************************************
++ * dom0_ops.h
++ *
++ * Process command requests from domain-0 guest OS.
++ *
++ * Copyright (c) 2002-2003, B Dragovic
++ * Copyright (c) 2002-2004, K Fraser
++ */
++
++
++#ifndef __XEN_PUBLIC_DOM0_OPS_H__
++#define __XEN_PUBLIC_DOM0_OPS_H__
++
++#include "xen.h"
++#include "sched_ctl.h"
++
++/*
++ * Make sure you increment the interface version whenever you modify this file!
++ * This makes sure that old versions of dom0 tools will stop working in a
++ * well-defined way (rather than crashing the machine, for instance).
++ */
++#define DOM0_INTERFACE_VERSION 0x03000001
++
++/************************************************************************/
++
++#define DOM0_GETMEMLIST 2
++struct dom0_getmemlist {
++ /* IN variables. */
++ domid_t domain;
++ uint64_t max_pfns;
++ XEN_GUEST_HANDLE(xen_pfn_t) buffer;
++ /* OUT variables. */
++ uint64_t num_pfns;
++};
++typedef struct dom0_getmemlist dom0_getmemlist_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getmemlist_t);
++
++#define DOM0_SCHEDCTL 6
++ /* struct sched_ctl_cmd is from sched-ctl.h */
++typedef struct sched_ctl_cmd dom0_schedctl_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_schedctl_t);
++
++#define DOM0_ADJUSTDOM 7
++/* struct sched_adjdom_cmd is from sched-ctl.h */
++typedef struct sched_adjdom_cmd dom0_adjustdom_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_adjustdom_t);
++
++#define DOM0_CREATEDOMAIN 8
++struct dom0_createdomain {
++ /* IN parameters */
++ uint32_t ssidref;
++ xen_domain_handle_t handle;
++ /* IN/OUT parameters. */
++ /* Identifier for new domain (auto-allocate if zero is specified). */
++ domid_t domain;
++};
++typedef struct dom0_createdomain dom0_createdomain_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_createdomain_t);
++
++#define DOM0_DESTROYDOMAIN 9
++struct dom0_destroydomain {
++ /* IN variables. */
++ domid_t domain;
++};
++typedef struct dom0_destroydomain dom0_destroydomain_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_destroydomain_t);
++
++#define DOM0_PAUSEDOMAIN 10
++struct dom0_pausedomain {
++ /* IN parameters. */
++ domid_t domain;
++};
++typedef struct dom0_pausedomain dom0_pausedomain_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_pausedomain_t);
++
++#define DOM0_UNPAUSEDOMAIN 11
++struct dom0_unpausedomain {
++ /* IN parameters. */
++ domid_t domain;
++};
++typedef struct dom0_unpausedomain dom0_unpausedomain_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_unpausedomain_t);
++
++#define DOM0_GETDOMAININFO 12
++struct dom0_getdomaininfo {
++ /* IN variables. */
++ domid_t domain; /* NB. IN/OUT variable. */
++ /* OUT variables. */
++#define DOMFLAGS_DYING (1<<0) /* Domain is scheduled to die. */
++#define DOMFLAGS_SHUTDOWN (1<<2) /* The guest OS has shut down. */
++#define DOMFLAGS_PAUSED (1<<3) /* Currently paused by control software. */
++#define DOMFLAGS_BLOCKED (1<<4) /* Currently blocked pending an event. */
++#define DOMFLAGS_RUNNING (1<<5) /* Domain is currently running. */
++#define DOMFLAGS_CPUMASK 255 /* CPU to which this domain is bound. */
++#define DOMFLAGS_CPUSHIFT 8
++#define DOMFLAGS_SHUTDOWNMASK 255 /* DOMFLAGS_SHUTDOWN guest-supplied code. */
++#define DOMFLAGS_SHUTDOWNSHIFT 16
++ uint32_t flags;
++ uint64_t tot_pages;
++ uint64_t max_pages;
++ xen_pfn_t shared_info_frame; /* MFN of shared_info struct */
++ uint64_t cpu_time;
++ uint32_t nr_online_vcpus; /* Number of VCPUs currently online. */
++ uint32_t max_vcpu_id; /* Maximum VCPUID in use by this domain. */
++ uint32_t ssidref;
++ xen_domain_handle_t handle;
++};
++typedef struct dom0_getdomaininfo dom0_getdomaininfo_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getdomaininfo_t);
++
++#define DOM0_SETVCPUCONTEXT 13
++struct dom0_setvcpucontext {
++ /* IN variables. */
++ domid_t domain;
++ uint32_t vcpu;
++ /* IN/OUT parameters */
++ XEN_GUEST_HANDLE(vcpu_guest_context_t) ctxt;
++};
++typedef struct dom0_setvcpucontext dom0_setvcpucontext_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_setvcpucontext_t);
++
++#define DOM0_MSR 15
++struct dom0_msr {
++ /* IN variables. */
++ uint32_t write;
++ cpumap_t cpu_mask;
++ uint32_t msr;
++ uint32_t in1;
++ uint32_t in2;
++ /* OUT variables. */
++ uint32_t out1;
++ uint32_t out2;
++};
++typedef struct dom0_msr dom0_msr_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_msr_t);
++
++/*
++ * Set clock such that it would read <secs,nsecs> after 00:00:00 UTC,
++ * 1 January, 1970 if the current system time was <system_time>.
++ */
++#define DOM0_SETTIME 17
++struct dom0_settime {
++ /* IN variables. */
++ uint32_t secs;
++ uint32_t nsecs;
++ uint64_t system_time;
++};
++typedef struct dom0_settime dom0_settime_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_settime_t);
++
++#define DOM0_GETPAGEFRAMEINFO 18
++#define LTAB_SHIFT 28
++#define NOTAB 0 /* normal page */
++#define L1TAB (1<<LTAB_SHIFT)
++#define L2TAB (2<<LTAB_SHIFT)
++#define L3TAB (3<<LTAB_SHIFT)
++#define L4TAB (4<<LTAB_SHIFT)
++#define LPINTAB (1<<31)
++#define XTAB (0xf<<LTAB_SHIFT) /* invalid page */
++#define LTAB_MASK XTAB
++#define LTABTYPE_MASK (0x7<<LTAB_SHIFT)
++
++struct dom0_getpageframeinfo {
++ /* IN variables. */
++ xen_pfn_t mfn; /* Machine page frame number to query. */
++ domid_t domain; /* To which domain does the frame belong? */
++ /* OUT variables. */
++ /* Is the page PINNED to a type? */
++ uint32_t type; /* see above type defs */
++};
++typedef struct dom0_getpageframeinfo dom0_getpageframeinfo_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getpageframeinfo_t);
++
++/*
++ * Read console content from Xen buffer ring.
++ */
++#define DOM0_READCONSOLE 19
++struct dom0_readconsole {
++ /* IN variables. */
++ uint32_t clear; /* Non-zero -> clear after reading. */
++ /* IN/OUT variables. */
++ XEN_GUEST_HANDLE(char) buffer; /* In: Buffer start; Out: Used buffer start */
++ uint32_t count; /* In: Buffer size; Out: Used buffer size */
++};
++typedef struct dom0_readconsole dom0_readconsole_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_readconsole_t);
++
++/*
++ * Set which physical cpus a vcpu can execute on.
++ */
++#define DOM0_SETVCPUAFFINITY 20
++struct dom0_setvcpuaffinity {
++ /* IN variables. */
++ domid_t domain;
++ uint32_t vcpu;
++ cpumap_t cpumap;
++};
++typedef struct dom0_setvcpuaffinity dom0_setvcpuaffinity_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_setvcpuaffinity_t);
++
++/* Get trace buffers machine base address */
++#define DOM0_TBUFCONTROL 21
++struct dom0_tbufcontrol {
++ /* IN variables */
++#define DOM0_TBUF_GET_INFO 0
++#define DOM0_TBUF_SET_CPU_MASK 1
++#define DOM0_TBUF_SET_EVT_MASK 2
++#define DOM0_TBUF_SET_SIZE 3
++#define DOM0_TBUF_ENABLE 4
++#define DOM0_TBUF_DISABLE 5
++ uint32_t op;
++ /* IN/OUT variables */
++ cpumap_t cpu_mask;
++ uint32_t evt_mask;
++ /* OUT variables */
++ xen_pfn_t buffer_mfn;
++ uint32_t size;
++};
++typedef struct dom0_tbufcontrol dom0_tbufcontrol_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_tbufcontrol_t);
++
++/*
++ * Get physical information about the host machine
++ */
++#define DOM0_PHYSINFO 22
++struct dom0_physinfo {
++ uint32_t threads_per_core;
++ uint32_t cores_per_socket;
++ uint32_t sockets_per_node;
++ uint32_t nr_nodes;
++ uint32_t cpu_khz;
++ uint64_t total_pages;
++ uint64_t free_pages;
++ uint64_t scrub_pages;
++ uint32_t hw_cap[8];
++};
++typedef struct dom0_physinfo dom0_physinfo_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_physinfo_t);
++
++/*
++ * Get the ID of the current scheduler.
++ */
++#define DOM0_SCHED_ID 24
++struct dom0_sched_id {
++ /* OUT variable */
++ uint32_t sched_id;
++};
++typedef struct dom0_physinfo dom0_sched_id_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_sched_id_t);
++
++/*
++ * Control shadow pagetables operation
++ */
++#define DOM0_SHADOW_CONTROL 25
++
++#define DOM0_SHADOW_CONTROL_OP_OFF 0
++#define DOM0_SHADOW_CONTROL_OP_ENABLE_TEST 1
++#define DOM0_SHADOW_CONTROL_OP_ENABLE_LOGDIRTY 2
++#define DOM0_SHADOW_CONTROL_OP_ENABLE_TRANSLATE 3
++
++#define DOM0_SHADOW_CONTROL_OP_FLUSH 10 /* table ops */
++#define DOM0_SHADOW_CONTROL_OP_CLEAN 11
++#define DOM0_SHADOW_CONTROL_OP_PEEK 12
++
++struct dom0_shadow_control_stats {
++ uint32_t fault_count;
++ uint32_t dirty_count;
++ uint32_t dirty_net_count;
++ uint32_t dirty_block_count;
++};
++typedef struct dom0_shadow_control_stats dom0_shadow_control_stats_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_shadow_control_stats_t);
++
++struct dom0_shadow_control {
++ /* IN variables. */
++ domid_t domain;
++ uint32_t op;
++ XEN_GUEST_HANDLE(ulong) dirty_bitmap;
++ /* IN/OUT variables. */
++ uint64_t pages; /* size of buffer, updated with actual size */
++ /* OUT variables. */
++ struct dom0_shadow_control_stats stats;
++};
++typedef struct dom0_shadow_control dom0_shadow_control_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_shadow_control_t);
++
++#define DOM0_SETDOMAINMAXMEM 28
++struct dom0_setdomainmaxmem {
++ /* IN variables. */
++ domid_t domain;
++ uint64_t max_memkb;
++};
++typedef struct dom0_setdomainmaxmem dom0_setdomainmaxmem_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_setdomainmaxmem_t);
++
++#define DOM0_GETPAGEFRAMEINFO2 29 /* batched interface */
++struct dom0_getpageframeinfo2 {
++ /* IN variables. */
++ domid_t domain;
++ uint64_t num;
++ /* IN/OUT variables. */
++ XEN_GUEST_HANDLE(ulong) array;
++};
++typedef struct dom0_getpageframeinfo2 dom0_getpageframeinfo2_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getpageframeinfo2_t);
++
++/*
++ * Request memory range (@mfn, @mfn+@nr_mfns-1) to have type @type.
++ * On x86, @type is an architecture-defined MTRR memory type.
++ * On success, returns the MTRR that was used (@reg) and a handle that can
++ * be passed to DOM0_DEL_MEMTYPE to accurately tear down the new setting.
++ * (x86-specific).
++ */
++#define DOM0_ADD_MEMTYPE 31
++struct dom0_add_memtype {
++ /* IN variables. */
++ xen_pfn_t mfn;
++ uint64_t nr_mfns;
++ uint32_t type;
++ /* OUT variables. */
++ uint32_t handle;
++ uint32_t reg;
++};
++typedef struct dom0_add_memtype dom0_add_memtype_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_add_memtype_t);
++
++/*
++ * Tear down an existing memory-range type. If @handle is remembered then it
++ * should be passed in to accurately tear down the correct setting (in case
++ * of overlapping memory regions with differing types). If it is not known
++ * then @handle should be set to zero. In all cases @reg must be set.
++ * (x86-specific).
++ */
++#define DOM0_DEL_MEMTYPE 32
++struct dom0_del_memtype {
++ /* IN variables. */
++ uint32_t handle;
++ uint32_t reg;
++};
++typedef struct dom0_del_memtype dom0_del_memtype_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_del_memtype_t);
++
++/* Read current type of an MTRR (x86-specific). */
++#define DOM0_READ_MEMTYPE 33
++struct dom0_read_memtype {
++ /* IN variables. */
++ uint32_t reg;
++ /* OUT variables. */
++ xen_pfn_t mfn;
++ uint64_t nr_mfns;
++ uint32_t type;
++};
++typedef struct dom0_read_memtype dom0_read_memtype_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_read_memtype_t);
++
++/* Interface for controlling Xen software performance counters. */
++#define DOM0_PERFCCONTROL 34
++/* Sub-operations: */
++#define DOM0_PERFCCONTROL_OP_RESET 1 /* Reset all counters to zero. */
++#define DOM0_PERFCCONTROL_OP_QUERY 2 /* Get perfctr information. */
++struct dom0_perfc_desc {
++ char name[80]; /* name of perf counter */
++ uint32_t nr_vals; /* number of values for this counter */
++ uint32_t vals[64]; /* array of values */
++};
++typedef struct dom0_perfc_desc dom0_perfc_desc_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_perfc_desc_t);
++
++struct dom0_perfccontrol {
++ /* IN variables. */
++ uint32_t op; /* DOM0_PERFCCONTROL_OP_??? */
++ /* OUT variables. */
++ uint32_t nr_counters; /* number of counters */
++ XEN_GUEST_HANDLE(dom0_perfc_desc_t) desc; /* counter information (or NULL) */
++};
++typedef struct dom0_perfccontrol dom0_perfccontrol_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_perfccontrol_t);
++
++#define DOM0_MICROCODE 35
++struct dom0_microcode {
++ /* IN variables. */
++ XEN_GUEST_HANDLE(void) data; /* Pointer to microcode data */
++ uint32_t length; /* Length of microcode data. */
++};
++typedef struct dom0_microcode dom0_microcode_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_microcode_t);
++
++#define DOM0_IOPORT_PERMISSION 36
++struct dom0_ioport_permission {
++ domid_t domain; /* domain to be affected */
++ uint32_t first_port; /* first port int range */
++ uint32_t nr_ports; /* size of port range */
++ uint8_t allow_access; /* allow or deny access to range? */
++};
++typedef struct dom0_ioport_permission dom0_ioport_permission_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_ioport_permission_t);
++
++#define DOM0_GETVCPUCONTEXT 37
++struct dom0_getvcpucontext {
++ /* IN variables. */
++ domid_t domain; /* domain to be affected */
++ uint32_t vcpu; /* vcpu # */
++ /* OUT variables. */
++ XEN_GUEST_HANDLE(vcpu_guest_context_t) ctxt;
++};
++typedef struct dom0_getvcpucontext dom0_getvcpucontext_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getvcpucontext_t);
++
++#define DOM0_GETVCPUINFO 43
++struct dom0_getvcpuinfo {
++ /* IN variables. */
++ domid_t domain; /* domain to be affected */
++ uint32_t vcpu; /* vcpu # */
++ /* OUT variables. */
++ uint8_t online; /* currently online (not hotplugged)? */
++ uint8_t blocked; /* blocked waiting for an event? */
++ uint8_t running; /* currently scheduled on its CPU? */
++ uint64_t cpu_time; /* total cpu time consumed (ns) */
++ uint32_t cpu; /* current mapping */
++ cpumap_t cpumap; /* allowable mapping */
++};
++typedef struct dom0_getvcpuinfo dom0_getvcpuinfo_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getvcpuinfo_t);
++
++#define DOM0_GETDOMAININFOLIST 38
++struct dom0_getdomaininfolist {
++ /* IN variables. */
++ domid_t first_domain;
++ uint32_t max_domains;
++ XEN_GUEST_HANDLE(dom0_getdomaininfo_t) buffer;
++ /* OUT variables. */
++ uint32_t num_domains;
++};
++typedef struct dom0_getdomaininfolist dom0_getdomaininfolist_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_getdomaininfolist_t);
++
++#define DOM0_PLATFORM_QUIRK 39
++#define QUIRK_NOIRQBALANCING 1 /* Do not restrict IO-APIC RTE targets */
++#define QUIRK_IOAPIC_BAD_REGSEL 2 /* IO-APIC REGSEL forgets its value */
++#define QUIRK_IOAPIC_GOOD_REGSEL 3 /* IO-APIC REGSEL behaves properly */
++struct dom0_platform_quirk {
++ /* IN variables. */
++ uint32_t quirk_id;
++};
++typedef struct dom0_platform_quirk dom0_platform_quirk_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_platform_quirk_t);
++
++#define DOM0_PHYSICAL_MEMORY_MAP 40 /* Unimplemented from 3.0.3 onwards */
++struct dom0_memory_map_entry {
++ uint64_t start, end;
++ uint32_t flags; /* reserved */
++ uint8_t is_ram;
++};
++typedef struct dom0_memory_map_entry dom0_memory_map_entry_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_memory_map_entry_t);
++
++struct dom0_physical_memory_map {
++ /* IN variables. */
++ uint32_t max_map_entries;
++ /* OUT variables. */
++ uint32_t nr_map_entries;
++ XEN_GUEST_HANDLE(dom0_memory_map_entry_t) memory_map;
++};
++typedef struct dom0_physical_memory_map dom0_physical_memory_map_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_physical_memory_map_t);
++
++#define DOM0_MAX_VCPUS 41
++struct dom0_max_vcpus {
++ domid_t domain; /* domain to be affected */
++ uint32_t max; /* maximum number of vcpus */
++};
++typedef struct dom0_max_vcpus dom0_max_vcpus_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_max_vcpus_t);
++
++#define DOM0_SETDOMAINHANDLE 44
++struct dom0_setdomainhandle {
++ domid_t domain;
++ xen_domain_handle_t handle;
++};
++typedef struct dom0_setdomainhandle dom0_setdomainhandle_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_setdomainhandle_t);
++
++#define DOM0_SETDEBUGGING 45
++struct dom0_setdebugging {
++ domid_t domain;
++ uint8_t enable;
++};
++typedef struct dom0_setdebugging dom0_setdebugging_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_setdebugging_t);
++
++#define DOM0_IRQ_PERMISSION 46
++struct dom0_irq_permission {
++ domid_t domain; /* domain to be affected */
++ uint8_t pirq;
++ uint8_t allow_access; /* flag to specify enable/disable of IRQ access */
++};
++typedef struct dom0_irq_permission dom0_irq_permission_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_irq_permission_t);
++
++#define DOM0_IOMEM_PERMISSION 47
++struct dom0_iomem_permission {
++ domid_t domain; /* domain to be affected */
++ xen_pfn_t first_mfn; /* first page (physical page number) in range */
++ uint64_t nr_mfns; /* number of pages in range (>0) */
++ uint8_t allow_access; /* allow (!0) or deny (0) access to range? */
++};
++typedef struct dom0_iomem_permission dom0_iomem_permission_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_iomem_permission_t);
++
++#define DOM0_HYPERCALL_INIT 48
++struct dom0_hypercall_init {
++ domid_t domain; /* domain to be affected */
++ xen_pfn_t mfn; /* machine frame to be initialised */
++};
++typedef struct dom0_hypercall_init dom0_hypercall_init_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_hypercall_init_t);
++
++#define DOM0_DOMAIN_SETUP 49
++#define _XEN_DOMAINSETUP_hvm_guest 0
++#define XEN_DOMAINSETUP_hvm_guest (1UL<<_XEN_DOMAINSETUP_hvm_guest)
++typedef struct dom0_domain_setup {
++ domid_t domain; /* domain to be affected */
++ unsigned long flags; /* XEN_DOMAINSETUP_* */
++#ifdef __ia64__
++ unsigned long bp; /* mpaddr of boot param area */
++ unsigned long maxmem; /* Highest memory address for MDT. */
++#endif
++} dom0_domain_setup_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_domain_setup_t);
++
++#define DOM0_SETTIMEOFFSET 50
++struct dom0_settimeoffset {
++ domid_t domain;
++ int32_t time_offset_seconds; /* applied to domain wallclock time */
++};
++typedef struct dom0_settimeoffset dom0_settimeoffset_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_settimeoffset_t);
++
++struct dom0_op {
++ uint32_t cmd;
++ uint32_t interface_version; /* DOM0_INTERFACE_VERSION */
++ union {
++ struct dom0_createdomain createdomain;
++ struct dom0_pausedomain pausedomain;
++ struct dom0_unpausedomain unpausedomain;
++ struct dom0_destroydomain destroydomain;
++ struct dom0_getmemlist getmemlist;
++ struct sched_ctl_cmd schedctl;
++ struct sched_adjdom_cmd adjustdom;
++ struct dom0_setvcpucontext setvcpucontext;
++ struct dom0_getdomaininfo getdomaininfo;
++ struct dom0_getpageframeinfo getpageframeinfo;
++ struct dom0_msr msr;
++ struct dom0_settime settime;
++ struct dom0_readconsole readconsole;
++ struct dom0_setvcpuaffinity setvcpuaffinity;
++ struct dom0_tbufcontrol tbufcontrol;
++ struct dom0_physinfo physinfo;
++ struct dom0_sched_id sched_id;
++ struct dom0_shadow_control shadow_control;
++ struct dom0_setdomainmaxmem setdomainmaxmem;
++ struct dom0_getpageframeinfo2 getpageframeinfo2;
++ struct dom0_add_memtype add_memtype;
++ struct dom0_del_memtype del_memtype;
++ struct dom0_read_memtype read_memtype;
++ struct dom0_perfccontrol perfccontrol;
++ struct dom0_microcode microcode;
++ struct dom0_ioport_permission ioport_permission;
++ struct dom0_getvcpucontext getvcpucontext;
++ struct dom0_getvcpuinfo getvcpuinfo;
++ struct dom0_getdomaininfolist getdomaininfolist;
++ struct dom0_platform_quirk platform_quirk;
++ struct dom0_physical_memory_map physical_memory_map;
++ struct dom0_max_vcpus max_vcpus;
++ struct dom0_setdomainhandle setdomainhandle;
++ struct dom0_setdebugging setdebugging;
++ struct dom0_irq_permission irq_permission;
++ struct dom0_iomem_permission iomem_permission;
++ struct dom0_hypercall_init hypercall_init;
++ struct dom0_domain_setup domain_setup;
++ struct dom0_settimeoffset settimeoffset;
++ uint8_t pad[128];
++ } u;
++};
++typedef struct dom0_op dom0_op_t;
++DEFINE_XEN_GUEST_HANDLE(dom0_op_t);
++
++#endif /* __XEN_PUBLIC_DOM0_OPS_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/event_channel.h new/include/xen/interface/event_channel.h
+--- linux-2.6/include/xen/interface/event_channel.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/event_channel.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,233 @@
++/******************************************************************************
++ * event_channel.h
++ *
++ * Event channels between domains.
++ *
++ * Copyright (c) 2003-2004, K A Fraser.
++ */
++
++#ifndef __XEN_PUBLIC_EVENT_CHANNEL_H__
++#define __XEN_PUBLIC_EVENT_CHANNEL_H__
++
++/*
++ * Prototype for this hypercall is:
++ * int event_channel_op(int cmd, void *args)
++ * @cmd == EVTCHNOP_??? (event-channel operation).
++ * @args == Operation-specific extra arguments (NULL if none).
++ */
++
++typedef uint32_t evtchn_port_t;
++DEFINE_XEN_GUEST_HANDLE(evtchn_port_t);
++
++/*
++ * EVTCHNOP_alloc_unbound: Allocate a port in domain <dom> and mark as
++ * accepting interdomain bindings from domain <remote_dom>. A fresh port
++ * is allocated in <dom> and returned as <port>.
++ * NOTES:
++ * 1. If the caller is unprivileged then <dom> must be DOMID_SELF.
++ * 2. <rdom> may be DOMID_SELF, allowing loopback connections.
++ */
++#define EVTCHNOP_alloc_unbound 6
++struct evtchn_alloc_unbound {
++ /* IN parameters */
++ domid_t dom, remote_dom;
++ /* OUT parameters */
++ evtchn_port_t port;
++};
++typedef struct evtchn_alloc_unbound evtchn_alloc_unbound_t;
++
++/*
++ * EVTCHNOP_bind_interdomain: Construct an interdomain event channel between
++ * the calling domain and <remote_dom>. <remote_dom,remote_port> must identify
++ * a port that is unbound and marked as accepting bindings from the calling
++ * domain. A fresh port is allocated in the calling domain and returned as
++ * <local_port>.
++ * NOTES:
++ * 2. <remote_dom> may be DOMID_SELF, allowing loopback connections.
++ */
++#define EVTCHNOP_bind_interdomain 0
++struct evtchn_bind_interdomain {
++ /* IN parameters. */
++ domid_t remote_dom;
++ evtchn_port_t remote_port;
++ /* OUT parameters. */
++ evtchn_port_t local_port;
++};
++typedef struct evtchn_bind_interdomain evtchn_bind_interdomain_t;
++
++/*
++ * EVTCHNOP_bind_virq: Bind a local event channel to VIRQ <irq> on specified
++ * vcpu.
++ * NOTES:
++ * 1. Virtual IRQs are classified as per-vcpu or global. See the VIRQ list
++ * in xen.h for the classification of each VIRQ.
++ * 2. Global VIRQs must be allocated on VCPU0 but can subsequently be
++ * re-bound via EVTCHNOP_bind_vcpu.
++ * 3. Per-vcpu VIRQs may be bound to at most one event channel per vcpu.
++ * The allocated event channel is bound to the specified vcpu and the
++ * binding cannot be changed.
++ */
++#define EVTCHNOP_bind_virq 1
++struct evtchn_bind_virq {
++ /* IN parameters. */
++ uint32_t virq;
++ uint32_t vcpu;
++ /* OUT parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_bind_virq evtchn_bind_virq_t;
++
++/*
++ * EVTCHNOP_bind_pirq: Bind a local event channel to PIRQ <irq>.
++ * NOTES:
++ * 1. A physical IRQ may be bound to at most one event channel per domain.
++ * 2. Only a sufficiently-privileged domain may bind to a physical IRQ.
++ */
++#define EVTCHNOP_bind_pirq 2
++struct evtchn_bind_pirq {
++ /* IN parameters. */
++ uint32_t pirq;
++#define BIND_PIRQ__WILL_SHARE 1
++ uint32_t flags; /* BIND_PIRQ__* */
++ /* OUT parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_bind_pirq evtchn_bind_pirq_t;
++
++/*
++ * EVTCHNOP_bind_ipi: Bind a local event channel to receive events.
++ * NOTES:
++ * 1. The allocated event channel is bound to the specified vcpu. The binding
++ * may not be changed.
++ */
++#define EVTCHNOP_bind_ipi 7
++struct evtchn_bind_ipi {
++ uint32_t vcpu;
++ /* OUT parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_bind_ipi evtchn_bind_ipi_t;
++
++/*
++ * EVTCHNOP_close: Close a local event channel <port>. If the channel is
++ * interdomain then the remote end is placed in the unbound state
++ * (EVTCHNSTAT_unbound), awaiting a new connection.
++ */
++#define EVTCHNOP_close 3
++struct evtchn_close {
++ /* IN parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_close evtchn_close_t;
++
++/*
++ * EVTCHNOP_send: Send an event to the remote end of the channel whose local
++ * endpoint is <port>.
++ */
++#define EVTCHNOP_send 4
++struct evtchn_send {
++ /* IN parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_send evtchn_send_t;
++
++/*
++ * EVTCHNOP_status: Get the current status of the communication channel which
++ * has an endpoint at <dom, port>.
++ * NOTES:
++ * 1. <dom> may be specified as DOMID_SELF.
++ * 2. Only a sufficiently-privileged domain may obtain the status of an event
++ * channel for which <dom> is not DOMID_SELF.
++ */
++#define EVTCHNOP_status 5
++struct evtchn_status {
++ /* IN parameters */
++ domid_t dom;
++ evtchn_port_t port;
++ /* OUT parameters */
++#define EVTCHNSTAT_closed 0 /* Channel is not in use. */
++#define EVTCHNSTAT_unbound 1 /* Channel is waiting interdom connection.*/
++#define EVTCHNSTAT_interdomain 2 /* Channel is connected to remote domain. */
++#define EVTCHNSTAT_pirq 3 /* Channel is bound to a phys IRQ line. */
++#define EVTCHNSTAT_virq 4 /* Channel is bound to a virtual IRQ line */
++#define EVTCHNSTAT_ipi 5 /* Channel is bound to a virtual IPI line */
++ uint32_t status;
++ uint32_t vcpu; /* VCPU to which this channel is bound. */
++ union {
++ struct {
++ domid_t dom;
++ } unbound; /* EVTCHNSTAT_unbound */
++ struct {
++ domid_t dom;
++ evtchn_port_t port;
++ } interdomain; /* EVTCHNSTAT_interdomain */
++ uint32_t pirq; /* EVTCHNSTAT_pirq */
++ uint32_t virq; /* EVTCHNSTAT_virq */
++ } u;
++};
++typedef struct evtchn_status evtchn_status_t;
++
++/*
++ * EVTCHNOP_bind_vcpu: Specify which vcpu a channel should notify when an
++ * event is pending.
++ * NOTES:
++ * 1. IPI-bound channels always notify the vcpu specified at bind time.
++ * This binding cannot be changed.
++ * 2. Per-VCPU VIRQ channels always notify the vcpu specified at bind time.
++ * This binding cannot be changed.
++ * 3. All other channels notify vcpu0 by default. This default is set when
++ * the channel is allocated (a port that is freed and subsequently reused
++ * has its binding reset to vcpu0).
++ */
++#define EVTCHNOP_bind_vcpu 8
++struct evtchn_bind_vcpu {
++ /* IN parameters. */
++ evtchn_port_t port;
++ uint32_t vcpu;
++};
++typedef struct evtchn_bind_vcpu evtchn_bind_vcpu_t;
++
++/*
++ * EVTCHNOP_unmask: Unmask the specified local event-channel port and deliver
++ * a notification to the appropriate VCPU if an event is pending.
++ */
++#define EVTCHNOP_unmask 9
++struct evtchn_unmask {
++ /* IN parameters. */
++ evtchn_port_t port;
++};
++typedef struct evtchn_unmask evtchn_unmask_t;
++
++/*
++ * Argument to event_channel_op_compat() hypercall. Superceded by new
++ * event_channel_op() hypercall since 0x00030202.
++ */
++struct evtchn_op {
++ uint32_t cmd; /* EVTCHNOP_* */
++ union {
++ struct evtchn_alloc_unbound alloc_unbound;
++ struct evtchn_bind_interdomain bind_interdomain;
++ struct evtchn_bind_virq bind_virq;
++ struct evtchn_bind_pirq bind_pirq;
++ struct evtchn_bind_ipi bind_ipi;
++ struct evtchn_close close;
++ struct evtchn_send send;
++ struct evtchn_status status;
++ struct evtchn_bind_vcpu bind_vcpu;
++ struct evtchn_unmask unmask;
++ } u;
++};
++typedef struct evtchn_op evtchn_op_t;
++DEFINE_XEN_GUEST_HANDLE(evtchn_op_t);
++
++#endif /* __XEN_PUBLIC_EVENT_CHANNEL_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/features.h new/include/xen/interface/features.h
+--- linux-2.6/include/xen/interface/features.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/features.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,53 @@
++/******************************************************************************
++ * features.h
++ *
++ * Feature flags, reported by XENVER_get_features.
++ *
++ * Copyright (c) 2006, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_FEATURES_H__
++#define __XEN_PUBLIC_FEATURES_H__
++
++/*
++ * If set, the guest does not need to write-protect its pagetables, and can
++ * update them via direct writes.
++ */
++#define XENFEAT_writable_page_tables 0
++
++/*
++ * If set, the guest does not need to write-protect its segment descriptor
++ * tables, and can update them via direct writes.
++ */
++#define XENFEAT_writable_descriptor_tables 1
++
++/*
++ * If set, translation between the guest's 'pseudo-physical' address space
++ * and the host's machine address space are handled by the hypervisor. In this
++ * mode the guest does not need to perform phys-to/from-machine translations
++ * when performing page table operations.
++ */
++#define XENFEAT_auto_translated_physmap 2
++
++/* If set, the guest is running in supervisor mode (e.g., x86 ring 0). */
++#define XENFEAT_supervisor_mode_kernel 3
++
++/*
++ * If set, the guest does not need to allocate x86 PAE page directories
++ * below 4GB. This flag is usually implied by auto_translated_physmap.
++ */
++#define XENFEAT_pae_pgdir_above_4gb 4
++
++#define XENFEAT_NR_SUBMAPS 1
++
++#endif /* __XEN_PUBLIC_FEATURES_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/grant_table.h new/include/xen/interface/grant_table.h
+--- linux-2.6/include/xen/interface/grant_table.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/grant_table.h 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,317 @@
++/******************************************************************************
++ * grant_table.h
++ *
++ * Interface for granting foreign access to page frames, and receiving
++ * page-ownership transfers.
++ *
++ * Copyright (c) 2004, K A Fraser
++ */
++
++#ifndef __XEN_PUBLIC_GRANT_TABLE_H__
++#define __XEN_PUBLIC_GRANT_TABLE_H__
++
++
++/***********************************
++ * GRANT TABLE REPRESENTATION
++ */
++
++/* Some rough guidelines on accessing and updating grant-table entries
++ * in a concurrency-safe manner. For more information, Linux contains a
++ * reference implementation for guest OSes (arch/xen/kernel/grant_table.c).
++ *
++ * NB. WMB is a no-op on current-generation x86 processors. However, a
++ * compiler barrier will still be required.
++ *
++ * Introducing a valid entry into the grant table:
++ * 1. Write ent->domid.
++ * 2. Write ent->frame:
++ * GTF_permit_access: Frame to which access is permitted.
++ * GTF_accept_transfer: Pseudo-phys frame slot being filled by new
++ * frame, or zero if none.
++ * 3. Write memory barrier (WMB).
++ * 4. Write ent->flags, inc. valid type.
++ *
++ * Invalidating an unused GTF_permit_access entry:
++ * 1. flags = ent->flags.
++ * 2. Observe that !(flags & (GTF_reading|GTF_writing)).
++ * 3. Check result of SMP-safe CMPXCHG(&ent->flags, flags, 0).
++ * NB. No need for WMB as reuse of entry is control-dependent on success of
++ * step 3, and all architectures guarantee ordering of ctrl-dep writes.
++ *
++ * Invalidating an in-use GTF_permit_access entry:
++ * This cannot be done directly. Request assistance from the domain controller
++ * which can set a timeout on the use of a grant entry and take necessary
++ * action. (NB. This is not yet implemented!).
++ *
++ * Invalidating an unused GTF_accept_transfer entry:
++ * 1. flags = ent->flags.
++ * 2. Observe that !(flags & GTF_transfer_committed). [*]
++ * 3. Check result of SMP-safe CMPXCHG(&ent->flags, flags, 0).
++ * NB. No need for WMB as reuse of entry is control-dependent on success of
++ * step 3, and all architectures guarantee ordering of ctrl-dep writes.
++ * [*] If GTF_transfer_committed is set then the grant entry is 'committed'.
++ * The guest must /not/ modify the grant entry until the address of the
++ * transferred frame is written. It is safe for the guest to spin waiting
++ * for this to occur (detect by observing GTF_transfer_completed in
++ * ent->flags).
++ *
++ * Invalidating a committed GTF_accept_transfer entry:
++ * 1. Wait for (ent->flags & GTF_transfer_completed).
++ *
++ * Changing a GTF_permit_access from writable to read-only:
++ * Use SMP-safe CMPXCHG to set GTF_readonly, while checking !GTF_writing.
++ *
++ * Changing a GTF_permit_access from read-only to writable:
++ * Use SMP-safe bit-setting instruction.
++ */
++
++/*
++ * A grant table comprises a packed array of grant entries in one or more
++ * page frames shared between Xen and a guest.
++ * [XEN]: This field is written by Xen and read by the sharing guest.
++ * [GST]: This field is written by the guest and read by Xen.
++ */
++struct grant_entry {
++ /* GTF_xxx: various type and flag information. [XEN,GST] */
++ uint16_t flags;
++ /* The domain being granted foreign privileges. [GST] */
++ domid_t domid;
++ /*
++ * GTF_permit_access: Frame that @domid is allowed to map and access. [GST]
++ * GTF_accept_transfer: Frame whose ownership transferred by @domid. [XEN]
++ */
++ uint32_t frame;
++};
++typedef struct grant_entry grant_entry_t;
++
++/*
++ * Type of grant entry.
++ * GTF_invalid: This grant entry grants no privileges.
++ * GTF_permit_access: Allow @domid to map/access @frame.
++ * GTF_accept_transfer: Allow @domid to transfer ownership of one page frame
++ * to this guest. Xen writes the page number to @frame.
++ */
++#define GTF_invalid (0U<<0)
++#define GTF_permit_access (1U<<0)
++#define GTF_accept_transfer (2U<<0)
++#define GTF_type_mask (3U<<0)
++
++/*
++ * Subflags for GTF_permit_access.
++ * GTF_readonly: Restrict @domid to read-only mappings and accesses. [GST]
++ * GTF_reading: Grant entry is currently mapped for reading by @domid. [XEN]
++ * GTF_writing: Grant entry is currently mapped for writing by @domid. [XEN]
++ */
++#define _GTF_readonly (2)
++#define GTF_readonly (1U<<_GTF_readonly)
++#define _GTF_reading (3)
++#define GTF_reading (1U<<_GTF_reading)
++#define _GTF_writing (4)
++#define GTF_writing (1U<<_GTF_writing)
++
++/*
++ * Subflags for GTF_accept_transfer:
++ * GTF_transfer_committed: Xen sets this flag to indicate that it is committed
++ * to transferring ownership of a page frame. When a guest sees this flag
++ * it must /not/ modify the grant entry until GTF_transfer_completed is
++ * set by Xen.
++ * GTF_transfer_completed: It is safe for the guest to spin-wait on this flag
++ * after reading GTF_transfer_committed. Xen will always write the frame
++ * address, followed by ORing this flag, in a timely manner.
++ */
++#define _GTF_transfer_committed (2)
++#define GTF_transfer_committed (1U<<_GTF_transfer_committed)
++#define _GTF_transfer_completed (3)
++#define GTF_transfer_completed (1U<<_GTF_transfer_completed)
++
++
++/***********************************
++ * GRANT TABLE QUERIES AND USES
++ */
++
++/*
++ * Reference to a grant entry in a specified domain's grant table.
++ */
++typedef uint32_t grant_ref_t;
++
++/*
++ * Handle to track a mapping created via a grant reference.
++ */
++typedef uint32_t grant_handle_t;
++
++/*
++ * GNTTABOP_map_grant_ref: Map the grant entry (<dom>,<ref>) for access
++ * by devices and/or host CPUs. If successful, <handle> is a tracking number
++ * that must be presented later to destroy the mapping(s). On error, <handle>
++ * is a negative status code.
++ * NOTES:
++ * 1. If GNTPIN_map_for_dev is specified then <dev_bus_addr> is the address
++ * via which I/O devices may access the granted frame.
++ * 2. If GNTPIN_map_for_host is specified then a mapping will be added at
++ * either a host virtual address in the current address space, or at
++ * a PTE at the specified machine address. The type of mapping to
++ * perform is selected through the GNTMAP_contains_pte flag, and the
++ * address is specified in <host_addr>.
++ * 3. Mappings should only be destroyed via GNTTABOP_unmap_grant_ref. If a
++ * host mapping is destroyed by other means then it is *NOT* guaranteed
++ * to be accounted to the correct grant reference!
++ */
++#define GNTTABOP_map_grant_ref 0
++struct gnttab_map_grant_ref {
++ /* IN parameters. */
++ uint64_t host_addr;
++ uint32_t flags; /* GNTMAP_* */
++ grant_ref_t ref;
++ domid_t dom;
++ /* OUT parameters. */
++ int16_t status; /* GNTST_* */
++ grant_handle_t handle;
++ uint64_t dev_bus_addr;
++};
++typedef struct gnttab_map_grant_ref gnttab_map_grant_ref_t;
++DEFINE_XEN_GUEST_HANDLE(gnttab_map_grant_ref_t);
++
++/*
++ * GNTTABOP_unmap_grant_ref: Destroy one or more grant-reference mappings
++ * tracked by <handle>. If <host_addr> or <dev_bus_addr> is zero, that
++ * field is ignored. If non-zero, they must refer to a device/host mapping
++ * that is tracked by <handle>
++ * NOTES:
++ * 1. The call may fail in an undefined manner if either mapping is not
++ * tracked by <handle>.
++ * 3. After executing a batch of unmaps, it is guaranteed that no stale
++ * mappings will remain in the device or host TLBs.
++ */
++#define GNTTABOP_unmap_grant_ref 1
++struct gnttab_unmap_grant_ref {
++ /* IN parameters. */
++ uint64_t host_addr;
++ uint64_t dev_bus_addr;
++ grant_handle_t handle;
++ /* OUT parameters. */
++ int16_t status; /* GNTST_* */
++};
++typedef struct gnttab_unmap_grant_ref gnttab_unmap_grant_ref_t;
++DEFINE_XEN_GUEST_HANDLE(gnttab_unmap_grant_ref_t);
++
++/*
++ * GNTTABOP_setup_table: Set up a grant table for <dom> comprising at least
++ * <nr_frames> pages. The frame addresses are written to the <frame_list>.
++ * Only <nr_frames> addresses are written, even if the table is larger.
++ * NOTES:
++ * 1. <dom> may be specified as DOMID_SELF.
++ * 2. Only a sufficiently-privileged domain may specify <dom> != DOMID_SELF.
++ * 3. Xen may not support more than a single grant-table page per domain.
++ */
++#define GNTTABOP_setup_table 2
++struct gnttab_setup_table {
++ /* IN parameters. */
++ domid_t dom;
++ uint32_t nr_frames;
++ /* OUT parameters. */
++ int16_t status; /* GNTST_* */
++ XEN_GUEST_HANDLE(ulong) frame_list;
++};
++typedef struct gnttab_setup_table gnttab_setup_table_t;
++DEFINE_XEN_GUEST_HANDLE(gnttab_setup_table_t);
++
++/*
++ * GNTTABOP_dump_table: Dump the contents of the grant table to the
++ * xen console. Debugging use only.
++ */
++#define GNTTABOP_dump_table 3
++struct gnttab_dump_table {
++ /* IN parameters. */
++ domid_t dom;
++ /* OUT parameters. */
++ int16_t status; /* GNTST_* */
++};
++typedef struct gnttab_dump_table gnttab_dump_table_t;
++DEFINE_XEN_GUEST_HANDLE(gnttab_dump_table_t);
++
++/*
++ * GNTTABOP_transfer_grant_ref: Transfer <frame> to a foreign domain. The
++ * foreign domain has previously registered its interest in the transfer via
++ * <domid, ref>.
++ *
++ * Note that, even if the transfer fails, the specified page no longer belongs
++ * to the calling domain *unless* the error is GNTST_bad_page.
++ */
++#define GNTTABOP_transfer 4
++struct gnttab_transfer {
++ /* IN parameters. */
++ xen_pfn_t mfn;
++ domid_t domid;
++ grant_ref_t ref;
++ /* OUT parameters. */
++ int16_t status;
++};
++typedef struct gnttab_transfer gnttab_transfer_t;
++DEFINE_XEN_GUEST_HANDLE(gnttab_transfer_t);
++
++/*
++ * Bitfield values for update_pin_status.flags.
++ */
++ /* Map the grant entry for access by I/O devices. */
++#define _GNTMAP_device_map (0)
++#define GNTMAP_device_map (1<<_GNTMAP_device_map)
++ /* Map the grant entry for access by host CPUs. */
++#define _GNTMAP_host_map (1)
++#define GNTMAP_host_map (1<<_GNTMAP_host_map)
++ /* Accesses to the granted frame will be restricted to read-only access. */
++#define _GNTMAP_readonly (2)
++#define GNTMAP_readonly (1<<_GNTMAP_readonly)
++ /*
++ * GNTMAP_host_map subflag:
++ * 0 => The host mapping is usable only by the guest OS.
++ * 1 => The host mapping is usable by guest OS + current application.
++ */
++#define _GNTMAP_application_map (3)
++#define GNTMAP_application_map (1<<_GNTMAP_application_map)
++
++ /*
++ * GNTMAP_contains_pte subflag:
++ * 0 => This map request contains a host virtual address.
++ * 1 => This map request contains the machine addess of the PTE to update.
++ */
++#define _GNTMAP_contains_pte (4)
++#define GNTMAP_contains_pte (1<<_GNTMAP_contains_pte)
++
++/*
++ * Values for error status returns. All errors are -ve.
++ */
++#define GNTST_okay (0) /* Normal return. */
++#define GNTST_general_error (-1) /* General undefined error. */
++#define GNTST_bad_domain (-2) /* Unrecognsed domain id. */
++#define GNTST_bad_gntref (-3) /* Unrecognised or inappropriate gntref. */
++#define GNTST_bad_handle (-4) /* Unrecognised or inappropriate handle. */
++#define GNTST_bad_virt_addr (-5) /* Inappropriate virtual address to map. */
++#define GNTST_bad_dev_addr (-6) /* Inappropriate device address to unmap.*/
++#define GNTST_no_device_space (-7) /* Out of space in I/O MMU. */
++#define GNTST_permission_denied (-8) /* Not enough privilege for operation. */
++#define GNTST_bad_page (-9) /* Specified page was invalid for op. */
++
++#define GNTTABOP_error_msgs { \
++ "okay", \
++ "undefined error", \
++ "unrecognised domain id", \
++ "invalid grant reference", \
++ "invalid mapping handle", \
++ "invalid virtual address", \
++ "invalid device address", \
++ "no spare translation slot in the I/O MMU", \
++ "permission denied", \
++ "bad page" \
++}
++
++#endif /* __XEN_PUBLIC_GRANT_TABLE_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/hvm/hvm_info_table.h new/include/xen/interface/hvm/hvm_info_table.h
+--- linux-2.6/include/xen/interface/hvm/hvm_info_table.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/hvm/hvm_info_table.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,24 @@
++/******************************************************************************
++ * hvm/hvm_info_table.h
++ *
++ * HVM parameter and information table, written into guest memory map.
++ */
++
++#ifndef __XEN_PUBLIC_HVM_HVM_INFO_TABLE_H__
++#define __XEN_PUBLIC_HVM_HVM_INFO_TABLE_H__
++
++#define HVM_INFO_PFN 0x09F
++#define HVM_INFO_OFFSET 0x800
++#define HVM_INFO_PADDR ((HVM_INFO_PFN << 12) + HVM_INFO_OFFSET)
++
++struct hvm_info_table {
++ char signature[8]; /* "HVM INFO" */
++ uint32_t length;
++ uint8_t checksum;
++ uint8_t acpi_enabled;
++ uint8_t apic_enabled;
++ uint8_t pae_enabled;
++ uint32_t nr_vcpus;
++};
++
++#endif /* __XEN_PUBLIC_HVM_HVM_INFO_TABLE_H__ */
+diff -urNp linux-2.6/include/xen/interface/hvm/ioreq.h new/include/xen/interface/hvm/ioreq.h
+--- linux-2.6/include/xen/interface/hvm/ioreq.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/hvm/ioreq.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,93 @@
++/*
++ * ioreq.h: I/O request definitions for device models
++ * Copyright (c) 2004, Intel Corporation.
++ *
++ * This program is free software; you can redistribute it and/or modify it
++ * under the terms and conditions of the GNU General Public License,
++ * version 2, as published by the Free Software Foundation.
++ *
++ * This program is distributed in the hope it will be useful, but WITHOUT
++ * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
++ * FITNESS FOR A PARTICULAR PURPOSE. See the GNU General Public License for
++ * more details.
++ *
++ * You should have received a copy of the GNU General Public License along with
++ * this program; if not, write to the Free Software Foundation, Inc., 59 Temple
++ * Place - Suite 330, Boston, MA 02111-1307 USA.
++ *
++ */
++
++#ifndef _IOREQ_H_
++#define _IOREQ_H_
++
++#define IOREQ_READ 1
++#define IOREQ_WRITE 0
++
++#define STATE_INVALID 0
++#define STATE_IOREQ_READY 1
++#define STATE_IOREQ_INPROCESS 2
++#define STATE_IORESP_READY 3
++#define STATE_IORESP_HOOK 4
++
++#define IOREQ_TYPE_PIO 0 /* pio */
++#define IOREQ_TYPE_COPY 1 /* mmio ops */
++#define IOREQ_TYPE_AND 2
++#define IOREQ_TYPE_OR 3
++#define IOREQ_TYPE_XOR 4
++#define IOREQ_TYPE_XCHG 5
++
++/*
++ * VMExit dispatcher should cooperate with instruction decoder to
++ * prepare this structure and notify service OS and DM by sending
++ * virq
++ */
++struct ioreq {
++ uint64_t addr; /* physical address */
++ uint64_t size; /* size in bytes */
++ uint64_t count; /* for rep prefixes */
++ union {
++ uint64_t data; /* data */
++ void *pdata; /* pointer to data */
++ } u;
++ uint8_t state:4;
++ uint8_t pdata_valid:1; /* if 1, use pdata above */
++ uint8_t dir:1; /* 1=read, 0=write */
++ uint8_t df:1;
++ uint8_t type; /* I/O type */
++ uint64_t io_count; /* How many IO done on a vcpu */
++};
++typedef struct ioreq ioreq_t;
++
++struct global_iodata {
++ uint16_t pic_elcr;
++ uint16_t pic_irr;
++ uint16_t pic_last_irr;
++ uint16_t pic_clear_irr;
++};
++typedef struct global_iodata global_iodata_t;
++
++struct vcpu_iodata {
++ struct ioreq vp_ioreq;
++ /* Event channel port */
++ unsigned int vp_eport; /* VMX vcpu uses this to notify DM */
++ unsigned int dm_eport; /* DM uses this to notify VMX vcpu */
++};
++typedef struct vcpu_iodata vcpu_iodata_t;
++
++struct shared_iopage {
++ struct global_iodata sp_global;
++ struct vcpu_iodata vcpu_iodata[1];
++};
++typedef struct shared_iopage shared_iopage_t;
++
++#endif /* _IOREQ_H_ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/hvm/vmx_assist.h new/include/xen/interface/hvm/vmx_assist.h
+--- linux-2.6/include/xen/interface/hvm/vmx_assist.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/hvm/vmx_assist.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,98 @@
++/*
++ * vmx_assist.h: Context definitions for the VMXASSIST world switch.
++ *
++ * Leendert van Doorn, leendert@watson.ibm.com
++ * Copyright (c) 2005, International Business Machines Corporation.
++ */
++
++#ifndef _VMX_ASSIST_H_
++#define _VMX_ASSIST_H_
++
++#define VMXASSIST_BASE 0xD0000
++#define VMXASSIST_MAGIC 0x17101966
++#define VMXASSIST_MAGIC_OFFSET (VMXASSIST_BASE+8)
++
++#define VMXASSIST_NEW_CONTEXT (VMXASSIST_BASE + 12)
++#define VMXASSIST_OLD_CONTEXT (VMXASSIST_NEW_CONTEXT + 4)
++
++#ifndef __ASSEMBLY__
++
++union vmcs_arbytes {
++ struct arbyte_fields {
++ unsigned int seg_type : 4,
++ s : 1,
++ dpl : 2,
++ p : 1,
++ reserved0 : 4,
++ avl : 1,
++ reserved1 : 1,
++ default_ops_size: 1,
++ g : 1,
++ null_bit : 1,
++ reserved2 : 15;
++ } fields;
++ unsigned int bytes;
++};
++
++/*
++ * World switch state
++ */
++struct vmx_assist_context {
++ uint32_t eip; /* execution pointer */
++ uint32_t esp; /* stack pointer */
++ uint32_t eflags; /* flags register */
++ uint32_t cr0;
++ uint32_t cr3; /* page table directory */
++ uint32_t cr4;
++ uint32_t idtr_limit; /* idt */
++ uint32_t idtr_base;
++ uint32_t gdtr_limit; /* gdt */
++ uint32_t gdtr_base;
++ uint32_t cs_sel; /* cs selector */
++ uint32_t cs_limit;
++ uint32_t cs_base;
++ union vmcs_arbytes cs_arbytes;
++ uint32_t ds_sel; /* ds selector */
++ uint32_t ds_limit;
++ uint32_t ds_base;
++ union vmcs_arbytes ds_arbytes;
++ uint32_t es_sel; /* es selector */
++ uint32_t es_limit;
++ uint32_t es_base;
++ union vmcs_arbytes es_arbytes;
++ uint32_t ss_sel; /* ss selector */
++ uint32_t ss_limit;
++ uint32_t ss_base;
++ union vmcs_arbytes ss_arbytes;
++ uint32_t fs_sel; /* fs selector */
++ uint32_t fs_limit;
++ uint32_t fs_base;
++ union vmcs_arbytes fs_arbytes;
++ uint32_t gs_sel; /* gs selector */
++ uint32_t gs_limit;
++ uint32_t gs_base;
++ union vmcs_arbytes gs_arbytes;
++ uint32_t tr_sel; /* task selector */
++ uint32_t tr_limit;
++ uint32_t tr_base;
++ union vmcs_arbytes tr_arbytes;
++ uint32_t ldtr_sel; /* ldtr selector */
++ uint32_t ldtr_limit;
++ uint32_t ldtr_base;
++ union vmcs_arbytes ldtr_arbytes;
++};
++typedef struct vmx_assist_context vmx_assist_context_t;
++
++#endif /* __ASSEMBLY__ */
++
++#endif /* _VMX_ASSIST_H_ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/blkif.h new/include/xen/interface/io/blkif.h
+--- linux-2.6/include/xen/interface/io/blkif.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/blkif.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,87 @@
++/******************************************************************************
++ * blkif.h
++ *
++ * Unified block-device I/O interface for Xen guest OSes.
++ *
++ * Copyright (c) 2003-2004, Keir Fraser
++ */
++
++#ifndef __XEN_PUBLIC_IO_BLKIF_H__
++#define __XEN_PUBLIC_IO_BLKIF_H__
++
++#include "ring.h"
++#include "../grant_table.h"
++
++/*
++ * Front->back notifications: When enqueuing a new request, sending a
++ * notification can be made conditional on req_event (i.e., the generic
++ * hold-off mechanism provided by the ring macros). Backends must set
++ * req_event appropriately (e.g., using RING_FINAL_CHECK_FOR_REQUESTS()).
++ *
++ * Back->front notifications: When enqueuing a new response, sending a
++ * notification can be made conditional on rsp_event (i.e., the generic
++ * hold-off mechanism provided by the ring macros). Frontends must set
++ * rsp_event appropriately (e.g., using RING_FINAL_CHECK_FOR_RESPONSES()).
++ */
++
++#ifndef blkif_vdev_t
++#define blkif_vdev_t uint16_t
++#endif
++#define blkif_sector_t uint64_t
++
++#define BLKIF_OP_READ 0
++#define BLKIF_OP_WRITE 1
++
++/*
++ * Maximum scatter/gather segments per request.
++ * This is carefully chosen so that sizeof(blkif_ring_t) <= PAGE_SIZE.
++ * NB. This could be 12 if the ring indexes weren't stored in the same page.
++ */
++#define BLKIF_MAX_SEGMENTS_PER_REQUEST 11
++
++struct blkif_request {
++ uint8_t operation; /* BLKIF_OP_??? */
++ uint8_t nr_segments; /* number of segments */
++ blkif_vdev_t handle; /* only for read/write requests */
++ uint64_t id; /* private guest value, echoed in resp */
++ blkif_sector_t sector_number;/* start sector idx on disk (r/w only) */
++ struct blkif_request_segment {
++ grant_ref_t gref; /* reference to I/O buffer frame */
++ /* @first_sect: first sector in frame to transfer (inclusive). */
++ /* @last_sect: last sector in frame to transfer (inclusive). */
++ uint8_t first_sect, last_sect;
++ } seg[BLKIF_MAX_SEGMENTS_PER_REQUEST];
++};
++typedef struct blkif_request blkif_request_t;
++
++struct blkif_response {
++ uint64_t id; /* copied from request */
++ uint8_t operation; /* copied from request */
++ int16_t status; /* BLKIF_RSP_??? */
++};
++typedef struct blkif_response blkif_response_t;
++
++#define BLKIF_RSP_ERROR -1 /* non-specific 'error' */
++#define BLKIF_RSP_OKAY 0 /* non-specific 'okay' */
++
++/*
++ * Generate blkif ring structures and types.
++ */
++
++DEFINE_RING_TYPES(blkif, struct blkif_request, struct blkif_response);
++
++#define VDISK_CDROM 0x1
++#define VDISK_REMOVABLE 0x2
++#define VDISK_READONLY 0x4
++
++#endif /* __XEN_PUBLIC_IO_BLKIF_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/console.h new/include/xen/interface/io/console.h
+--- linux-2.6/include/xen/interface/io/console.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/console.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,33 @@
++/******************************************************************************
++ * console.h
++ *
++ * Console I/O interface for Xen guest OSes.
++ *
++ * Copyright (c) 2005, Keir Fraser
++ */
++
++#ifndef __XEN_PUBLIC_IO_CONSOLE_H__
++#define __XEN_PUBLIC_IO_CONSOLE_H__
++
++typedef uint32_t XENCONS_RING_IDX;
++
++#define MASK_XENCONS_IDX(idx, ring) ((idx) & (sizeof(ring)-1))
++
++struct xencons_interface {
++ char in[1024];
++ char out[2048];
++ XENCONS_RING_IDX in_cons, in_prod;
++ XENCONS_RING_IDX out_cons, out_prod;
++};
++
++#endif /* __XEN_PUBLIC_IO_CONSOLE_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/netif.h new/include/xen/interface/io/netif.h
+--- linux-2.6/include/xen/interface/io/netif.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/netif.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,148 @@
++/******************************************************************************
++ * netif.h
++ *
++ * Unified network-device I/O interface for Xen guest OSes.
++ *
++ * Copyright (c) 2003-2004, Keir Fraser
++ */
++
++#ifndef __XEN_PUBLIC_IO_NETIF_H__
++#define __XEN_PUBLIC_IO_NETIF_H__
++
++#include "ring.h"
++#include "../grant_table.h"
++
++/*
++ * Note that there is *never* any need to notify the backend when
++ * enqueuing receive requests (struct netif_rx_request). Notifications
++ * after enqueuing any other type of message should be conditional on
++ * the appropriate req_event or rsp_event field in the shared ring.
++ */
++
++/*
++ * This is the 'wire' format for packets:
++ * Request 1: netif_tx_request -- NETTXF_* (any flags)
++ * [Request 2: netif_tx_extra] (only if request 1 has NETTXF_extra_info)
++ * [Request 3: netif_tx_extra] (only if request 2 has XEN_NETIF_EXTRA_MORE)
++ * Request 4: netif_tx_request -- NETTXF_more_data
++ * Request 5: netif_tx_request -- NETTXF_more_data
++ * ...
++ * Request N: netif_tx_request -- 0
++ */
++
++/* Protocol checksum field is blank in the packet (hardware offload)? */
++#define _NETTXF_csum_blank (0)
++#define NETTXF_csum_blank (1U<<_NETTXF_csum_blank)
++
++/* Packet data has been validated against protocol checksum. */
++#define _NETTXF_data_validated (1)
++#define NETTXF_data_validated (1U<<_NETTXF_data_validated)
++
++/* Packet continues in the next request descriptor. */
++#define _NETTXF_more_data (2)
++#define NETTXF_more_data (1U<<_NETTXF_more_data)
++
++/* Packet to be followed by extra descriptor(s). */
++#define _NETTXF_extra_info (3)
++#define NETTXF_extra_info (1U<<_NETTXF_extra_info)
++
++struct netif_tx_request {
++ grant_ref_t gref; /* Reference to buffer page */
++ uint16_t offset; /* Offset within buffer page */
++ uint16_t flags; /* NETTXF_* */
++ uint16_t id; /* Echoed in response message. */
++ uint16_t size; /* Packet size in bytes. */
++};
++typedef struct netif_tx_request netif_tx_request_t;
++
++/* Types of netif_extra_info descriptors. */
++#define XEN_NETIF_EXTRA_TYPE_NONE (0) /* Never used - invalid */
++#define XEN_NETIF_EXTRA_TYPE_GSO (1) /* u.gso */
++#define XEN_NETIF_EXTRA_TYPE_MAX (2)
++
++/* netif_extra_info flags. */
++#define _XEN_NETIF_EXTRA_FLAG_MORE (0)
++#define XEN_NETIF_EXTRA_FLAG_MORE (1U<<_XEN_NETIF_EXTRA_FLAG_MORE)
++
++/* GSO types - only TCPv4 currently supported. */
++#define XEN_NETIF_GSO_TCPV4 (1)
++
++/*
++ * This structure needs to fit within both netif_tx_request and
++ * netif_rx_response for compatibility.
++ */
++struct netif_extra_info {
++ uint8_t type; /* XEN_NETIF_EXTRA_TYPE_* */
++ uint8_t flags; /* XEN_NETIF_EXTRA_FLAG_* */
++
++ union {
++ struct {
++ /*
++ * Maximum payload size of each segment. For example, for TCP this
++ * is just the path MSS.
++ */
++ uint16_t size;
++
++ /*
++ * GSO type. This determines the protocol of the packet and any
++ * extra features required to segment the packet properly.
++ */
++ uint16_t type; /* XEN_NETIF_GSO_* */
++ } gso;
++
++ uint16_t pad[3];
++ } u;
++};
++
++struct netif_tx_response {
++ uint16_t id;
++ int16_t status; /* NETIF_RSP_* */
++};
++typedef struct netif_tx_response netif_tx_response_t;
++
++struct netif_rx_request {
++ uint16_t id; /* Echoed in response message. */
++ grant_ref_t gref; /* Reference to incoming granted frame */
++};
++typedef struct netif_rx_request netif_rx_request_t;
++
++/* Packet data has been validated against protocol checksum. */
++#define _NETRXF_data_validated (0)
++#define NETRXF_data_validated (1U<<_NETRXF_data_validated)
++
++/* Protocol checksum field is blank in the packet (hardware offload)? */
++#define _NETRXF_csum_blank (1)
++#define NETRXF_csum_blank (1U<<_NETRXF_csum_blank)
++
++struct netif_rx_response {
++ uint16_t id;
++ uint16_t offset; /* Offset in page of start of received packet */
++ uint16_t flags; /* NETRXF_* */
++ int16_t status; /* -ve: BLKIF_RSP_* ; +ve: Rx'ed pkt size. */
++};
++typedef struct netif_rx_response netif_rx_response_t;
++
++/*
++ * Generate netif ring structures and types.
++ */
++
++DEFINE_RING_TYPES(netif_tx, struct netif_tx_request, struct netif_tx_response);
++DEFINE_RING_TYPES(netif_rx, struct netif_rx_request, struct netif_rx_response);
++
++#define NETIF_RSP_DROPPED -2
++#define NETIF_RSP_ERROR -1
++#define NETIF_RSP_OKAY 0
++/* No response: used for auxiliary requests (e.g., netif_tx_extra). */
++#define NETIF_RSP_NULL 1
++
++#endif
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/pciif.h new/include/xen/interface/io/pciif.h
+--- linux-2.6/include/xen/interface/io/pciif.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/pciif.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,55 @@
++/*
++ * PCI Backend/Frontend Common Data Structures & Macros
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#ifndef __XEN_PCI_COMMON_H__
++#define __XEN_PCI_COMMON_H__
++
++/* Be sure to bump this number if you change this file */
++#define XEN_PCI_MAGIC "7"
++
++/* xen_pci_sharedinfo flags */
++#define _XEN_PCIF_active (0)
++#define XEN_PCIF_active (1<<_XEN_PCI_active)
++
++/* xen_pci_op commands */
++#define XEN_PCI_OP_conf_read (0)
++#define XEN_PCI_OP_conf_write (1)
++
++/* xen_pci_op error numbers */
++#define XEN_PCI_ERR_success (0)
++#define XEN_PCI_ERR_dev_not_found (-1)
++#define XEN_PCI_ERR_invalid_offset (-2)
++#define XEN_PCI_ERR_access_denied (-3)
++#define XEN_PCI_ERR_not_implemented (-4)
++/* XEN_PCI_ERR_op_failed - backend failed to complete the operation */
++#define XEN_PCI_ERR_op_failed (-5)
++
++struct xen_pci_op {
++ /* IN: what action to perform: XEN_PCI_OP_* */
++ uint32_t cmd;
++
++ /* OUT: will contain an error number (if any) from errno.h */
++ int32_t err;
++
++ /* IN: which device to touch */
++ uint32_t domain; /* PCI Domain/Segment */
++ uint32_t bus;
++ uint32_t devfn;
++
++ /* IN: which configuration registers to touch */
++ int32_t offset;
++ int32_t size;
++
++ /* IN/OUT: Contains the result after a READ or the value to WRITE */
++ uint32_t value;
++};
++
++struct xen_pci_sharedinfo {
++ /* flags - XEN_PCIF_* */
++ uint32_t flags;
++ struct xen_pci_op op;
++};
++
++#endif /* __XEN_PCI_COMMON_H__ */
+diff -urNp linux-2.6/include/xen/interface/io/ring.h new/include/xen/interface/io/ring.h
+--- linux-2.6/include/xen/interface/io/ring.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/ring.h 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,273 @@
++/******************************************************************************
++ * ring.h
++ *
++ * Shared producer-consumer ring macros.
++ *
++ * Tim Deegan and Andrew Warfield November 2004.
++ */
++
++#ifndef __XEN_PUBLIC_IO_RING_H__
++#define __XEN_PUBLIC_IO_RING_H__
++
++typedef unsigned int RING_IDX;
++
++/* Round a 32-bit unsigned constant down to the nearest power of two. */
++#define __RD2(_x) (((_x) & 0x00000002) ? 0x2 : ((_x) & 0x1))
++#define __RD4(_x) (((_x) & 0x0000000c) ? __RD2((_x)>>2)<<2 : __RD2(_x))
++#define __RD8(_x) (((_x) & 0x000000f0) ? __RD4((_x)>>4)<<4 : __RD4(_x))
++#define __RD16(_x) (((_x) & 0x0000ff00) ? __RD8((_x)>>8)<<8 : __RD8(_x))
++#define __RD32(_x) (((_x) & 0xffff0000) ? __RD16((_x)>>16)<<16 : __RD16(_x))
++
++/*
++ * Calculate size of a shared ring, given the total available space for the
++ * ring and indexes (_sz), and the name tag of the request/response structure.
++ * A ring contains as many entries as will fit, rounded down to the nearest
++ * power of two (so we can mask with (size-1) to loop around).
++ */
++#define __RING_SIZE(_s, _sz) \
++ (__RD32(((_sz) - (long)&(_s)->ring + (long)(_s)) / sizeof((_s)->ring[0])))
++
++/*
++ * Macros to make the correct C datatypes for a new kind of ring.
++ *
++ * To make a new ring datatype, you need to have two message structures,
++ * let's say request_t, and response_t already defined.
++ *
++ * In a header where you want the ring datatype declared, you then do:
++ *
++ * DEFINE_RING_TYPES(mytag, request_t, response_t);
++ *
++ * These expand out to give you a set of types, as you can see below.
++ * The most important of these are:
++ *
++ * mytag_sring_t - The shared ring.
++ * mytag_front_ring_t - The 'front' half of the ring.
++ * mytag_back_ring_t - The 'back' half of the ring.
++ *
++ * To initialize a ring in your code you need to know the location and size
++ * of the shared memory area (PAGE_SIZE, for instance). To initialise
++ * the front half:
++ *
++ * mytag_front_ring_t front_ring;
++ * SHARED_RING_INIT((mytag_sring_t *)shared_page);
++ * FRONT_RING_INIT(&front_ring, (mytag_sring_t *)shared_page, PAGE_SIZE);
++ *
++ * Initializing the back follows similarly (note that only the front
++ * initializes the shared ring):
++ *
++ * mytag_back_ring_t back_ring;
++ * BACK_RING_INIT(&back_ring, (mytag_sring_t *)shared_page, PAGE_SIZE);
++ */
++
++#define DEFINE_RING_TYPES(__name, __req_t, __rsp_t) \
++ \
++/* Shared ring entry */ \
++union __name##_sring_entry { \
++ __req_t req; \
++ __rsp_t rsp; \
++}; \
++ \
++/* Shared ring page */ \
++struct __name##_sring { \
++ RING_IDX req_prod, req_event; \
++ RING_IDX rsp_prod, rsp_event; \
++ uint8_t pad[48]; \
++ union __name##_sring_entry ring[1]; /* variable-length */ \
++}; \
++ \
++/* "Front" end's private variables */ \
++struct __name##_front_ring { \
++ RING_IDX req_prod_pvt; \
++ RING_IDX rsp_cons; \
++ unsigned int nr_ents; \
++ struct __name##_sring *sring; \
++}; \
++ \
++/* "Back" end's private variables */ \
++struct __name##_back_ring { \
++ RING_IDX rsp_prod_pvt; \
++ RING_IDX req_cons; \
++ unsigned int nr_ents; \
++ struct __name##_sring *sring; \
++}; \
++ \
++/* Syntactic sugar */ \
++typedef struct __name##_sring __name##_sring_t; \
++typedef struct __name##_front_ring __name##_front_ring_t; \
++typedef struct __name##_back_ring __name##_back_ring_t
++
++/*
++ * Macros for manipulating rings.
++ *
++ * FRONT_RING_whatever works on the "front end" of a ring: here
++ * requests are pushed on to the ring and responses taken off it.
++ *
++ * BACK_RING_whatever works on the "back end" of a ring: here
++ * requests are taken off the ring and responses put on.
++ *
++ * N.B. these macros do NO INTERLOCKS OR FLOW CONTROL.
++ * This is OK in 1-for-1 request-response situations where the
++ * requestor (front end) never has more than RING_SIZE()-1
++ * outstanding requests.
++ */
++
++/* Initialising empty rings */
++#define SHARED_RING_INIT(_s) do { \
++ (_s)->req_prod = (_s)->rsp_prod = 0; \
++ (_s)->req_event = (_s)->rsp_event = 1; \
++ memset((_s)->pad, 0, sizeof((_s)->pad)); \
++} while(0)
++
++#define FRONT_RING_INIT(_r, _s, __size) do { \
++ (_r)->req_prod_pvt = 0; \
++ (_r)->rsp_cons = 0; \
++ (_r)->nr_ents = __RING_SIZE(_s, __size); \
++ (_r)->sring = (_s); \
++} while (0)
++
++#define BACK_RING_INIT(_r, _s, __size) do { \
++ (_r)->rsp_prod_pvt = 0; \
++ (_r)->req_cons = 0; \
++ (_r)->nr_ents = __RING_SIZE(_s, __size); \
++ (_r)->sring = (_s); \
++} while (0)
++
++/* Initialize to existing shared indexes -- for recovery */
++#define FRONT_RING_ATTACH(_r, _s, __size) do { \
++ (_r)->sring = (_s); \
++ (_r)->req_prod_pvt = (_s)->req_prod; \
++ (_r)->rsp_cons = (_s)->rsp_prod; \
++ (_r)->nr_ents = __RING_SIZE(_s, __size); \
++} while (0)
++
++#define BACK_RING_ATTACH(_r, _s, __size) do { \
++ (_r)->sring = (_s); \
++ (_r)->rsp_prod_pvt = (_s)->rsp_prod; \
++ (_r)->req_cons = (_s)->req_prod; \
++ (_r)->nr_ents = __RING_SIZE(_s, __size); \
++} while (0)
++
++/* How big is this ring? */
++#define RING_SIZE(_r) \
++ ((_r)->nr_ents)
++
++/* Number of free requests (for use on front side only). */
++#define RING_FREE_REQUESTS(_r) \
++ (RING_SIZE(_r) - ((_r)->req_prod_pvt - (_r)->rsp_cons))
++
++/* Test if there is an empty slot available on the front ring.
++ * (This is only meaningful from the front. )
++ */
++#define RING_FULL(_r) \
++ (RING_FREE_REQUESTS(_r) == 0)
++
++/* Test if there are outstanding messages to be processed on a ring. */
++#define RING_HAS_UNCONSUMED_RESPONSES(_r) \
++ ((_r)->sring->rsp_prod - (_r)->rsp_cons)
++
++#define RING_HAS_UNCONSUMED_REQUESTS(_r) \
++ ({ \
++ unsigned int req = (_r)->sring->req_prod - (_r)->req_cons; \
++ unsigned int rsp = RING_SIZE(_r) - \
++ ((_r)->req_cons - (_r)->rsp_prod_pvt); \
++ req < rsp ? req : rsp; \
++ })
++
++/* Direct access to individual ring elements, by index. */
++#define RING_GET_REQUEST(_r, _idx) \
++ (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].req))
++
++#define RING_GET_RESPONSE(_r, _idx) \
++ (&((_r)->sring->ring[((_idx) & (RING_SIZE(_r) - 1))].rsp))
++
++/* Loop termination condition: Would the specified index overflow the ring? */
++#define RING_REQUEST_CONS_OVERFLOW(_r, _cons) \
++ (((_cons) - (_r)->rsp_prod_pvt) >= RING_SIZE(_r))
++
++#define RING_PUSH_REQUESTS(_r) do { \
++ wmb(); /* back sees requests /before/ updated producer index */ \
++ (_r)->sring->req_prod = (_r)->req_prod_pvt; \
++} while (0)
++
++#define RING_PUSH_RESPONSES(_r) do { \
++ wmb(); /* front sees responses /before/ updated producer index */ \
++ (_r)->sring->rsp_prod = (_r)->rsp_prod_pvt; \
++} while (0)
++
++/*
++ * Notification hold-off (req_event and rsp_event):
++ *
++ * When queueing requests or responses on a shared ring, it may not always be
++ * necessary to notify the remote end. For example, if requests are in flight
++ * in a backend, the front may be able to queue further requests without
++ * notifying the back (if the back checks for new requests when it queues
++ * responses).
++ *
++ * When enqueuing requests or responses:
++ *
++ * Use RING_PUSH_{REQUESTS,RESPONSES}_AND_CHECK_NOTIFY(). The second argument
++ * is a boolean return value. True indicates that the receiver requires an
++ * asynchronous notification.
++ *
++ * After dequeuing requests or responses (before sleeping the connection):
++ *
++ * Use RING_FINAL_CHECK_FOR_REQUESTS() or RING_FINAL_CHECK_FOR_RESPONSES().
++ * The second argument is a boolean return value. True indicates that there
++ * are pending messages on the ring (i.e., the connection should not be put
++ * to sleep).
++ *
++ * These macros will set the req_event/rsp_event field to trigger a
++ * notification on the very next message that is enqueued. If you want to
++ * create batches of work (i.e., only receive a notification after several
++ * messages have been enqueued) then you will need to create a customised
++ * version of the FINAL_CHECK macro in your own code, which sets the event
++ * field appropriately.
++ */
++
++#define RING_PUSH_REQUESTS_AND_CHECK_NOTIFY(_r, _notify) do { \
++ RING_IDX __old = (_r)->sring->req_prod; \
++ RING_IDX __new = (_r)->req_prod_pvt; \
++ wmb(); /* back sees requests /before/ updated producer index */ \
++ (_r)->sring->req_prod = __new; \
++ mb(); /* back sees new requests /before/ we check req_event */ \
++ (_notify) = ((RING_IDX)(__new - (_r)->sring->req_event) < \
++ (RING_IDX)(__new - __old)); \
++} while (0)
++
++#define RING_PUSH_RESPONSES_AND_CHECK_NOTIFY(_r, _notify) do { \
++ RING_IDX __old = (_r)->sring->rsp_prod; \
++ RING_IDX __new = (_r)->rsp_prod_pvt; \
++ wmb(); /* front sees responses /before/ updated producer index */ \
++ (_r)->sring->rsp_prod = __new; \
++ mb(); /* front sees new responses /before/ we check rsp_event */ \
++ (_notify) = ((RING_IDX)(__new - (_r)->sring->rsp_event) < \
++ (RING_IDX)(__new - __old)); \
++} while (0)
++
++#define RING_FINAL_CHECK_FOR_REQUESTS(_r, _work_to_do) do { \
++ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
++ if (_work_to_do) break; \
++ (_r)->sring->req_event = (_r)->req_cons + 1; \
++ mb(); \
++ (_work_to_do) = RING_HAS_UNCONSUMED_REQUESTS(_r); \
++} while (0)
++
++#define RING_FINAL_CHECK_FOR_RESPONSES(_r, _work_to_do) do { \
++ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
++ if (_work_to_do) break; \
++ (_r)->sring->rsp_event = (_r)->rsp_cons + 1; \
++ mb(); \
++ (_work_to_do) = RING_HAS_UNCONSUMED_RESPONSES(_r); \
++} while (0)
++
++#endif /* __XEN_PUBLIC_IO_RING_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/tpmif.h new/include/xen/interface/io/tpmif.h
+--- linux-2.6/include/xen/interface/io/tpmif.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/tpmif.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,59 @@
++/******************************************************************************
++ * tpmif.h
++ *
++ * TPM I/O interface for Xen guest OSes.
++ *
++ * Copyright (c) 2005, IBM Corporation
++ *
++ * Author: Stefan Berger, stefanb@us.ibm.com
++ * Grant table support: Mahadevan Gomathisankaran
++ *
++ * This code has been derived from tools/libxc/xen/io/netif.h
++ *
++ * Copyright (c) 2003-2004, Keir Fraser
++ */
++
++#ifndef __XEN_PUBLIC_IO_TPMIF_H__
++#define __XEN_PUBLIC_IO_TPMIF_H__
++
++#include "../grant_table.h"
++
++struct tpmif_tx_request {
++ unsigned long addr; /* Machine address of packet. */
++ grant_ref_t ref; /* grant table access reference */
++ uint16_t unused;
++ uint16_t size; /* Packet size in bytes. */
++};
++typedef struct tpmif_tx_request tpmif_tx_request_t;
++
++/*
++ * The TPMIF_TX_RING_SIZE defines the number of pages the
++ * front-end and backend can exchange (= size of array).
++ */
++typedef uint32_t TPMIF_RING_IDX;
++
++#define TPMIF_TX_RING_SIZE 10
++
++/* This structure must fit in a memory page. */
++
++struct tpmif_ring {
++ struct tpmif_tx_request req;
++};
++typedef struct tpmif_ring tpmif_ring_t;
++
++struct tpmif_tx_interface {
++ struct tpmif_ring ring[TPMIF_TX_RING_SIZE];
++};
++typedef struct tpmif_tx_interface tpmif_tx_interface_t;
++
++#endif
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/io/xenbus.h new/include/xen/interface/io/xenbus.h
+--- linux-2.6/include/xen/interface/io/xenbus.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/xenbus.h 2006-06-05 15:54:33.000000000 +0200
+@@ -0,0 +1,45 @@
++/*****************************************************************************
++ * xenbus.h
++ *
++ * Xenbus protocol details.
++ *
++ * Copyright (C) 2005 XenSource Ltd.
++ */
++
++#ifndef _XEN_PUBLIC_IO_XENBUS_H
++#define _XEN_PUBLIC_IO_XENBUS_H
++
++/*
++ * The state of either end of the Xenbus, i.e. the current communication
++ * status of initialisation across the bus. States here imply nothing about
++ * the state of the connection between the driver and the kernel's device
++ * layers.
++ */
++enum xenbus_state {
++ XenbusStateUnknown = 0,
++
++ XenbusStateInitialising = 1,
++
++ /*
++ * InitWait: Finished early initialisation but waiting for information
++ * from the peer or hotplug scripts.
++ */
++ XenbusStateInitWait = 2,
++
++ /*
++ * Initialised: Waiting for a connection from the peer.
++ */
++ XenbusStateInitialised = 3,
++
++ XenbusStateConnected = 4,
++
++ /*
++ * Closing: The device is being closed due to an error or an unplug event.
++ */
++ XenbusStateClosing = 5,
++
++ XenbusStateClosed = 6
++};
++typedef enum xenbus_state XenbusState;
++
++#endif /* _XEN_PUBLIC_IO_XENBUS_H */
+diff -urNp linux-2.6/include/xen/interface/io/xs_wire.h new/include/xen/interface/io/xs_wire.h
+--- linux-2.6/include/xen/interface/io/xs_wire.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/io/xs_wire.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,97 @@
++/*
++ * Details of the "wire" protocol between Xen Store Daemon and client
++ * library or guest kernel.
++ * Copyright (C) 2005 Rusty Russell IBM Corporation
++ */
++
++#ifndef _XS_WIRE_H
++#define _XS_WIRE_H
++
++enum xsd_sockmsg_type
++{
++ XS_DEBUG,
++ XS_DIRECTORY,
++ XS_READ,
++ XS_GET_PERMS,
++ XS_WATCH,
++ XS_UNWATCH,
++ XS_TRANSACTION_START,
++ XS_TRANSACTION_END,
++ XS_INTRODUCE,
++ XS_RELEASE,
++ XS_GET_DOMAIN_PATH,
++ XS_WRITE,
++ XS_MKDIR,
++ XS_RM,
++ XS_SET_PERMS,
++ XS_WATCH_EVENT,
++ XS_ERROR,
++ XS_IS_DOMAIN_INTRODUCED
++};
++
++#define XS_WRITE_NONE "NONE"
++#define XS_WRITE_CREATE "CREATE"
++#define XS_WRITE_CREATE_EXCL "CREATE|EXCL"
++
++/* We hand errors as strings, for portability. */
++struct xsd_errors
++{
++ int errnum;
++ const char *errstring;
++};
++#define XSD_ERROR(x) { x, #x }
++static struct xsd_errors xsd_errors[] __attribute__((unused)) = {
++ XSD_ERROR(EINVAL),
++ XSD_ERROR(EACCES),
++ XSD_ERROR(EEXIST),
++ XSD_ERROR(EISDIR),
++ XSD_ERROR(ENOENT),
++ XSD_ERROR(ENOMEM),
++ XSD_ERROR(ENOSPC),
++ XSD_ERROR(EIO),
++ XSD_ERROR(ENOTEMPTY),
++ XSD_ERROR(ENOSYS),
++ XSD_ERROR(EROFS),
++ XSD_ERROR(EBUSY),
++ XSD_ERROR(EAGAIN),
++ XSD_ERROR(EISCONN)
++};
++
++struct xsd_sockmsg
++{
++ uint32_t type; /* XS_??? */
++ uint32_t req_id;/* Request identifier, echoed in daemon's response. */
++ uint32_t tx_id; /* Transaction id (0 if not related to a transaction). */
++ uint32_t len; /* Length of data following this. */
++
++ /* Generally followed by nul-terminated string(s). */
++};
++
++enum xs_watch_type
++{
++ XS_WATCH_PATH = 0,
++ XS_WATCH_TOKEN
++};
++
++/* Inter-domain shared memory communications. */
++#define XENSTORE_RING_SIZE 1024
++typedef uint32_t XENSTORE_RING_IDX;
++#define MASK_XENSTORE_IDX(idx) ((idx) & (XENSTORE_RING_SIZE-1))
++struct xenstore_domain_interface {
++ char req[XENSTORE_RING_SIZE]; /* Requests to xenstore daemon. */
++ char rsp[XENSTORE_RING_SIZE]; /* Replies and async watch events. */
++ XENSTORE_RING_IDX req_cons, req_prod;
++ XENSTORE_RING_IDX rsp_cons, rsp_prod;
++};
++
++#endif /* _XS_WIRE_H */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/memory.h new/include/xen/interface/memory.h
+--- linux-2.6/include/xen/interface/memory.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/memory.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,243 @@
++/******************************************************************************
++ * memory.h
++ *
++ * Memory reservation and information.
++ *
++ * Copyright (c) 2005, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_MEMORY_H__
++#define __XEN_PUBLIC_MEMORY_H__
++
++/*
++ * Increase or decrease the specified domain's memory reservation. Returns the
++ * number of extents successfully allocated or freed.
++ * arg == addr of struct xen_memory_reservation.
++ */
++#define XENMEM_increase_reservation 0
++#define XENMEM_decrease_reservation 1
++#define XENMEM_populate_physmap 6
++struct xen_memory_reservation {
++
++ /*
++ * XENMEM_increase_reservation:
++ * OUT: MFN (*not* GMFN) bases of extents that were allocated
++ * XENMEM_decrease_reservation:
++ * IN: GMFN bases of extents to free
++ * XENMEM_populate_physmap:
++ * IN: GPFN bases of extents to populate with memory
++ * OUT: GMFN bases of extents that were allocated
++ * (NB. This command also updates the mach_to_phys translation table)
++ */
++ XEN_GUEST_HANDLE(xen_pfn_t) extent_start;
++
++ /* Number of extents, and size/alignment of each (2^extent_order pages). */
++ xen_ulong_t nr_extents;
++ unsigned int extent_order;
++
++ /*
++ * Maximum # bits addressable by the user of the allocated region (e.g.,
++ * I/O devices often have a 32-bit limitation even in 64-bit systems). If
++ * zero then the user has no addressing restriction.
++ * This field is not used by XENMEM_decrease_reservation.
++ */
++ unsigned int address_bits;
++
++ /*
++ * Domain whose reservation is being changed.
++ * Unprivileged domains can specify only DOMID_SELF.
++ */
++ domid_t domid;
++};
++typedef struct xen_memory_reservation xen_memory_reservation_t;
++DEFINE_XEN_GUEST_HANDLE(xen_memory_reservation_t);
++
++/*
++ * An atomic exchange of memory pages. If return code is zero then
++ * @out.extent_list provides GMFNs of the newly-allocated memory.
++ * Returns zero on complete success, otherwise a negative error code.
++ * On complete success then always @nr_exchanged == @in.nr_extents.
++ * On partial success @nr_exchanged indicates how much work was done.
++ */
++#define XENMEM_exchange 11
++struct xen_memory_exchange {
++ /*
++ * [IN] Details of memory extents to be exchanged (GMFN bases).
++ * Note that @in.address_bits is ignored and unused.
++ */
++ struct xen_memory_reservation in;
++
++ /*
++ * [IN/OUT] Details of new memory extents.
++ * We require that:
++ * 1. @in.domid == @out.domid
++ * 2. @in.nr_extents << @in.extent_order ==
++ * @out.nr_extents << @out.extent_order
++ * 3. @in.extent_start and @out.extent_start lists must not overlap
++ * 4. @out.extent_start lists GPFN bases to be populated
++ * 5. @out.extent_start is overwritten with allocated GMFN bases
++ */
++ struct xen_memory_reservation out;
++
++ /*
++ * [OUT] Number of input extents that were successfully exchanged:
++ * 1. The first @nr_exchanged input extents were successfully
++ * deallocated.
++ * 2. The corresponding first entries in the output extent list correctly
++ * indicate the GMFNs that were successfully exchanged.
++ * 3. All other input and output extents are untouched.
++ * 4. If not all input exents are exchanged then the return code of this
++ * command will be non-zero.
++ * 5. THIS FIELD MUST BE INITIALISED TO ZERO BY THE CALLER!
++ */
++ xen_ulong_t nr_exchanged;
++};
++typedef struct xen_memory_exchange xen_memory_exchange_t;
++DEFINE_XEN_GUEST_HANDLE(xen_memory_exchange_t);
++
++/*
++ * Returns the maximum machine frame number of mapped RAM in this system.
++ * This command always succeeds (it never returns an error code).
++ * arg == NULL.
++ */
++#define XENMEM_maximum_ram_page 2
++
++/*
++ * Returns the current or maximum memory reservation, in pages, of the
++ * specified domain (may be DOMID_SELF). Returns -ve errcode on failure.
++ * arg == addr of domid_t.
++ */
++#define XENMEM_current_reservation 3
++#define XENMEM_maximum_reservation 4
++
++/*
++ * Returns a list of MFN bases of 2MB extents comprising the machine_to_phys
++ * mapping table. Architectures which do not have a m2p table do not implement
++ * this command.
++ * arg == addr of xen_machphys_mfn_list_t.
++ */
++#define XENMEM_machphys_mfn_list 5
++struct xen_machphys_mfn_list {
++ /*
++ * Size of the 'extent_start' array. Fewer entries will be filled if the
++ * machphys table is smaller than max_extents * 2MB.
++ */
++ unsigned int max_extents;
++
++ /*
++ * Pointer to buffer to fill with list of extent starts. If there are
++ * any large discontiguities in the machine address space, 2MB gaps in
++ * the machphys table will be represented by an MFN base of zero.
++ */
++ XEN_GUEST_HANDLE(xen_pfn_t) extent_start;
++
++ /*
++ * Number of extents written to the above array. This will be smaller
++ * than 'max_extents' if the machphys table is smaller than max_e * 2MB.
++ */
++ unsigned int nr_extents;
++};
++typedef struct xen_machphys_mfn_list xen_machphys_mfn_list_t;
++DEFINE_XEN_GUEST_HANDLE(xen_machphys_mfn_list_t);
++
++/*
++ * Returns the location in virtual address space of the machine_to_phys
++ * mapping table. Architectures which do not have a m2p table, or which do not
++ * map it by default into guest address space, do not implement this command.
++ * arg == addr of xen_machphys_mapping_t.
++ */
++#define XENMEM_machphys_mapping 12
++struct xen_machphys_mapping {
++ xen_ulong_t v_start, v_end; /* Start and end virtual addresses. */
++ xen_ulong_t max_mfn; /* Maximum MFN that can be looked up. */
++};
++typedef struct xen_machphys_mapping xen_machphys_mapping_t;
++DEFINE_XEN_GUEST_HANDLE(xen_machphys_mapping_t);
++
++/*
++ * Sets the GPFN at which a particular page appears in the specified guest's
++ * pseudophysical address space.
++ * arg == addr of xen_add_to_physmap_t.
++ */
++#define XENMEM_add_to_physmap 7
++struct xen_add_to_physmap {
++ /* Which domain to change the mapping for. */
++ domid_t domid;
++
++ /* Source mapping space. */
++#define XENMAPSPACE_shared_info 0 /* shared info page */
++#define XENMAPSPACE_grant_table 1 /* grant table page */
++ unsigned int space;
++
++ /* Index into source mapping space. */
++ xen_ulong_t idx;
++
++ /* GPFN where the source mapping page should appear. */
++ xen_pfn_t gpfn;
++};
++typedef struct xen_add_to_physmap xen_add_to_physmap_t;
++DEFINE_XEN_GUEST_HANDLE(xen_add_to_physmap_t);
++
++/*
++ * Translates a list of domain-specific GPFNs into MFNs. Returns a -ve error
++ * code on failure. This call only works for auto-translated guests.
++ */
++#define XENMEM_translate_gpfn_list 8
++struct xen_translate_gpfn_list {
++ /* Which domain to translate for? */
++ domid_t domid;
++
++ /* Length of list. */
++ xen_ulong_t nr_gpfns;
++
++ /* List of GPFNs to translate. */
++ XEN_GUEST_HANDLE(xen_pfn_t) gpfn_list;
++
++ /*
++ * Output list to contain MFN translations. May be the same as the input
++ * list (in which case each input GPFN is overwritten with the output MFN).
++ */
++ XEN_GUEST_HANDLE(xen_pfn_t) mfn_list;
++};
++typedef struct xen_translate_gpfn_list xen_translate_gpfn_list_t;
++DEFINE_XEN_GUEST_HANDLE(xen_translate_gpfn_list_t);
++
++/*
++ * Returns the pseudo-physical memory map as it was when the domain
++ * was started.
++ */
++#define XENMEM_memory_map 9
++struct xen_memory_map {
++ /*
++ * On call the number of entries which can be stored in buffer. On
++ * return the number of entries which have been stored in
++ * buffer.
++ */
++ unsigned int nr_entries;
++
++ /*
++ * Entries in the buffer are in the same format as returned by the
++ * BIOS INT 0x15 EAX=0xE820 call.
++ */
++ XEN_GUEST_HANDLE(void) buffer;
++};
++typedef struct xen_memory_map xen_memory_map_t;
++DEFINE_XEN_GUEST_HANDLE(xen_memory_map_t);
++
++/*
++ * Returns the real physical memory map. Passes the same structure as
++ * XENMEM_memory_map.
++ */
++#define XENMEM_machine_memory_map 10
++
++#endif /* __XEN_PUBLIC_MEMORY_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/nmi.h new/include/xen/interface/nmi.h
+--- linux-2.6/include/xen/interface/nmi.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/nmi.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,60 @@
++/******************************************************************************
++ * nmi.h
++ *
++ * NMI callback registration and reason codes.
++ *
++ * Copyright (c) 2005, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_NMI_H__
++#define __XEN_PUBLIC_NMI_H__
++
++/*
++ * NMI reason codes:
++ * Currently these are x86-specific, stored in arch_shared_info.nmi_reason.
++ */
++ /* I/O-check error reported via ISA port 0x61, bit 6. */
++#define _XEN_NMIREASON_io_error 0
++#define XEN_NMIREASON_io_error (1UL << _XEN_NMIREASON_io_error)
++ /* Parity error reported via ISA port 0x61, bit 7. */
++#define _XEN_NMIREASON_parity_error 1
++#define XEN_NMIREASON_parity_error (1UL << _XEN_NMIREASON_parity_error)
++ /* Unknown hardware-generated NMI. */
++#define _XEN_NMIREASON_unknown 2
++#define XEN_NMIREASON_unknown (1UL << _XEN_NMIREASON_unknown)
++
++/*
++ * long nmi_op(unsigned int cmd, void *arg)
++ * NB. All ops return zero on success, else a negative error code.
++ */
++
++/*
++ * Register NMI callback for this (calling) VCPU. Currently this only makes
++ * sense for domain 0, vcpu 0. All other callers will be returned EINVAL.
++ * arg == pointer to xennmi_callback structure.
++ */
++#define XENNMI_register_callback 0
++struct xennmi_callback {
++ unsigned long handler_address;
++ unsigned long pad;
++};
++typedef struct xennmi_callback xennmi_callback_t;
++DEFINE_XEN_GUEST_HANDLE(xennmi_callback_t);
++
++/*
++ * Deregister NMI callback for this (calling) VCPU.
++ * arg == NULL.
++ */
++#define XENNMI_unregister_callback 1
++
++#endif /* __XEN_PUBLIC_NMI_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/physdev.h new/include/xen/interface/physdev.h
+--- linux-2.6/include/xen/interface/physdev.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/physdev.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,149 @@
++
++#ifndef __XEN_PUBLIC_PHYSDEV_H__
++#define __XEN_PUBLIC_PHYSDEV_H__
++
++/*
++ * Prototype for this hypercall is:
++ * int physdev_op(int cmd, void *args)
++ * @cmd == PHYSDEVOP_??? (physdev operation).
++ * @args == Operation-specific extra arguments (NULL if none).
++ */
++
++/*
++ * Notify end-of-interrupt (EOI) for the specified IRQ.
++ * @arg == pointer to physdev_eoi structure.
++ */
++#define PHYSDEVOP_eoi 12
++struct physdev_eoi {
++ /* IN */
++ uint32_t irq;
++};
++typedef struct physdev_eoi physdev_eoi_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_eoi_t);
++
++/*
++ * Query the status of an IRQ line.
++ * @arg == pointer to physdev_irq_status_query structure.
++ */
++#define PHYSDEVOP_irq_status_query 5
++struct physdev_irq_status_query {
++ /* IN */
++ uint32_t irq;
++ /* OUT */
++ uint32_t flags; /* XENIRQSTAT_* */
++};
++typedef struct physdev_irq_status_query physdev_irq_status_query_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_irq_status_query_t);
++
++/* Need to call PHYSDEVOP_eoi when the IRQ has been serviced? */
++#define _XENIRQSTAT_needs_eoi (0)
++#define XENIRQSTAT_needs_eoi (1U<<_XENIRQSTAT_needs_eoi)
++
++/* IRQ shared by multiple guests? */
++#define _XENIRQSTAT_shared (1)
++#define XENIRQSTAT_shared (1U<<_XENIRQSTAT_shared)
++
++/*
++ * Set the current VCPU's I/O privilege level.
++ * @arg == pointer to physdev_set_iopl structure.
++ */
++#define PHYSDEVOP_set_iopl 6
++struct physdev_set_iopl {
++ /* IN */
++ uint32_t iopl;
++};
++typedef struct physdev_set_iopl physdev_set_iopl_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_set_iopl_t);
++
++/*
++ * Set the current VCPU's I/O-port permissions bitmap.
++ * @arg == pointer to physdev_set_iobitmap structure.
++ */
++#define PHYSDEVOP_set_iobitmap 7
++struct physdev_set_iobitmap {
++ /* IN */
++ uint8_t *bitmap;
++ uint32_t nr_ports;
++};
++typedef struct physdev_set_iobitmap physdev_set_iobitmap_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_set_iobitmap_t);
++
++/*
++ * Read or write an IO-APIC register.
++ * @arg == pointer to physdev_apic structure.
++ */
++#define PHYSDEVOP_apic_read 8
++#define PHYSDEVOP_apic_write 9
++struct physdev_apic {
++ /* IN */
++ unsigned long apic_physbase;
++ uint32_t reg;
++ /* IN or OUT */
++ uint32_t value;
++};
++typedef struct physdev_apic physdev_apic_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_apic_t);
++
++/*
++ * Allocate or free a physical upcall vector for the specified IRQ line.
++ * @arg == pointer to physdev_irq structure.
++ */
++#define PHYSDEVOP_alloc_irq_vector 10
++#define PHYSDEVOP_free_irq_vector 11
++struct physdev_irq {
++ /* IN */
++ uint32_t irq;
++ /* IN or OUT */
++ uint32_t vector;
++};
++typedef struct physdev_irq physdev_irq_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_irq_t);
++
++/*
++ * Argument to physdev_op_compat() hypercall. Superceded by new physdev_op()
++ * hypercall since 0x00030202.
++ */
++struct physdev_op {
++ uint32_t cmd;
++ union {
++ struct physdev_irq_status_query irq_status_query;
++ struct physdev_set_iopl set_iopl;
++ struct physdev_set_iobitmap set_iobitmap;
++ struct physdev_apic apic_op;
++ struct physdev_irq irq_op;
++ } u;
++};
++typedef struct physdev_op physdev_op_t;
++DEFINE_XEN_GUEST_HANDLE(physdev_op_t);
++
++/*
++ * Notify that some PIRQ-bound event channels have been unmasked.
++ * ** This command is obsolete since interface version 0x00030202 and is **
++ * ** unsupported by newer versions of Xen. **
++ */
++#define PHYSDEVOP_IRQ_UNMASK_NOTIFY 4
++
++/*
++ * These all-capitals physdev operation names are superceded by the new names
++ * (defined above) since interface version 0x00030202.
++ */
++#define PHYSDEVOP_IRQ_STATUS_QUERY PHYSDEVOP_irq_status_query
++#define PHYSDEVOP_SET_IOPL PHYSDEVOP_set_iopl
++#define PHYSDEVOP_SET_IOBITMAP PHYSDEVOP_set_iobitmap
++#define PHYSDEVOP_APIC_READ PHYSDEVOP_apic_read
++#define PHYSDEVOP_APIC_WRITE PHYSDEVOP_apic_write
++#define PHYSDEVOP_ASSIGN_VECTOR PHYSDEVOP_alloc_irq_vector
++#define PHYSDEVOP_IRQ_NEEDS_UNMASK_NOTIFY XENIRQSTAT_needs_eoi
++#define PHYSDEVOP_IRQ_SHARED XENIRQSTAT_shared
++
++#endif /* __XEN_PUBLIC_PHYSDEV_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/sched_ctl.h new/include/xen/interface/sched_ctl.h
+--- linux-2.6/include/xen/interface/sched_ctl.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/sched_ctl.h 2006-06-05 15:54:33.000000000 +0200
+@@ -0,0 +1,69 @@
++/******************************************************************************
++ * Generic scheduler control interface.
++ *
++ * Mark Williamson, (C) 2004 Intel Research Cambridge
++ */
++
++#ifndef __XEN_PUBLIC_SCHED_CTL_H__
++#define __XEN_PUBLIC_SCHED_CTL_H__
++
++/* Scheduler types. */
++#define SCHED_BVT 0
++#define SCHED_SEDF 4
++#define SCHED_CREDIT 5
++
++/* Set or get info? */
++#define SCHED_INFO_PUT 0
++#define SCHED_INFO_GET 1
++
++/*
++ * Generic scheduler control command - used to adjust system-wide scheduler
++ * parameters
++ */
++struct sched_ctl_cmd {
++ uint32_t sched_id;
++ uint32_t direction;
++ union {
++ struct bvt_ctl {
++ uint32_t ctx_allow;
++ } bvt;
++ } u;
++};
++
++struct sched_adjdom_cmd {
++ uint32_t sched_id;
++ uint32_t direction;
++ domid_t domain;
++ union {
++ struct bvt_adjdom {
++ uint32_t mcu_adv; /* mcu advance: inverse of weight */
++ uint32_t warpback; /* warp? */
++ int32_t warpvalue; /* warp value */
++ int64_t warpl; /* warp limit */
++ int64_t warpu; /* unwarp time requirement */
++ } bvt;
++ struct sedf_adjdom {
++ uint64_t period;
++ uint64_t slice;
++ uint64_t latency;
++ uint32_t extratime;
++ uint32_t weight;
++ } sedf;
++ struct sched_credit_adjdom {
++ uint16_t weight;
++ uint16_t cap;
++ } credit;
++ } u;
++};
++
++#endif /* __XEN_PUBLIC_SCHED_CTL_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/sched.h new/include/xen/interface/sched.h
+--- linux-2.6/include/xen/interface/sched.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/sched.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,103 @@
++/******************************************************************************
++ * sched.h
++ *
++ * Scheduler state interactions
++ *
++ * Copyright (c) 2005, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_SCHED_H__
++#define __XEN_PUBLIC_SCHED_H__
++
++#include "event_channel.h"
++
++/*
++ * The prototype for this hypercall is:
++ * long sched_op(int cmd, void *arg)
++ * @cmd == SCHEDOP_??? (scheduler operation).
++ * @arg == Operation-specific extra argument(s), as described below.
++ *
++ * Versions of Xen prior to 3.0.2 provided only the following legacy version
++ * of this hypercall, supporting only the commands yield, block and shutdown:
++ * long sched_op(int cmd, unsigned long arg)
++ * @cmd == SCHEDOP_??? (scheduler operation).
++ * @arg == 0 (SCHEDOP_yield and SCHEDOP_block)
++ * == SHUTDOWN_* code (SCHEDOP_shutdown)
++ * This legacy version is available to new guests as sched_op_compat().
++ */
++
++/*
++ * Voluntarily yield the CPU.
++ * @arg == NULL.
++ */
++#define SCHEDOP_yield 0
++
++/*
++ * Block execution of this VCPU until an event is received for processing.
++ * If called with event upcalls masked, this operation will atomically
++ * reenable event delivery and check for pending events before blocking the
++ * VCPU. This avoids a "wakeup waiting" race.
++ * @arg == NULL.
++ */
++#define SCHEDOP_block 1
++
++/*
++ * Halt execution of this domain (all VCPUs) and notify the system controller.
++ * @arg == pointer to sched_shutdown structure.
++ */
++#define SCHEDOP_shutdown 2
++struct sched_shutdown {
++ unsigned int reason; /* SHUTDOWN_* */
++};
++typedef struct sched_shutdown sched_shutdown_t;
++DEFINE_XEN_GUEST_HANDLE(sched_shutdown_t);
++
++/*
++ * Poll a set of event-channel ports. Return when one or more are pending. An
++ * optional timeout may be specified.
++ * @arg == pointer to sched_poll structure.
++ */
++#define SCHEDOP_poll 3
++struct sched_poll {
++ XEN_GUEST_HANDLE(evtchn_port_t) ports;
++ unsigned int nr_ports;
++ uint64_t timeout;
++};
++typedef struct sched_poll sched_poll_t;
++DEFINE_XEN_GUEST_HANDLE(sched_poll_t);
++
++/*
++ * Declare a shutdown for another domain. The main use of this function is
++ * in interpreting shutdown requests and reasons for fully-virtualized
++ * domains. A para-virtualized domain may use SCHEDOP_shutdown directly.
++ * @arg == pointer to sched_remote_shutdown structure.
++ */
++#define SCHEDOP_remote_shutdown 4
++struct sched_remote_shutdown {
++ domid_t domain_id; /* Remote domain ID */
++ unsigned int reason; /* SHUTDOWN_xxx reason */
++};
++typedef struct sched_remote_shutdown sched_remote_shutdown_t;
++DEFINE_XEN_GUEST_HANDLE(sched_remote_shutdown_t);
++
++/*
++ * Reason codes for SCHEDOP_shutdown. These may be interpreted by control
++ * software to determine the appropriate action. For the most part, Xen does
++ * not care about the shutdown code.
++ */
++#define SHUTDOWN_poweroff 0 /* Domain exited normally. Clean up and kill. */
++#define SHUTDOWN_reboot 1 /* Clean up, kill, and then restart. */
++#define SHUTDOWN_suspend 2 /* Clean up, save suspend info, kill. */
++#define SHUTDOWN_crash 3 /* Tell controller we've crashed. */
++
++#endif /* __XEN_PUBLIC_SCHED_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/trace.h new/include/xen/interface/trace.h
+--- linux-2.6/include/xen/interface/trace.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/trace.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,86 @@
++/******************************************************************************
++ * include/public/trace.h
++ *
++ * Mark Williamson, (C) 2004 Intel Research Cambridge
++ * Copyright (C) 2005 Bin Ren
++ */
++
++#ifndef __XEN_PUBLIC_TRACE_H__
++#define __XEN_PUBLIC_TRACE_H__
++
++/* Trace classes */
++#define TRC_CLS_SHIFT 16
++#define TRC_GEN 0x0001f000 /* General trace */
++#define TRC_SCHED 0x0002f000 /* Xen Scheduler trace */
++#define TRC_DOM0OP 0x0004f000 /* Xen DOM0 operation trace */
++#define TRC_VMX 0x0008f000 /* Xen VMX trace */
++#define TRC_MEM 0x000af000 /* Xen memory trace */
++#define TRC_ALL 0xfffff000
++
++/* Trace subclasses */
++#define TRC_SUBCLS_SHIFT 12
++/* trace subclasses for VMX */
++#define TRC_VMXEXIT 0x00081000 /* VMX exit trace */
++#define TRC_VMXTIMER 0x00082000 /* VMX timer trace */
++#define TRC_VMXINT 0x00084000 /* VMX interrupt trace */
++#define TRC_VMXIO 0x00088000 /* VMX io emulation trace */
++
++/* Trace events per class */
++
++#define TRC_SCHED_DOM_ADD (TRC_SCHED + 1)
++#define TRC_SCHED_DOM_REM (TRC_SCHED + 2)
++#define TRC_SCHED_SLEEP (TRC_SCHED + 3)
++#define TRC_SCHED_WAKE (TRC_SCHED + 4)
++#define TRC_SCHED_YIELD (TRC_SCHED + 5)
++#define TRC_SCHED_BLOCK (TRC_SCHED + 6)
++#define TRC_SCHED_SHUTDOWN (TRC_SCHED + 7)
++#define TRC_SCHED_CTL (TRC_SCHED + 8)
++#define TRC_SCHED_ADJDOM (TRC_SCHED + 9)
++#define TRC_SCHED_SWITCH (TRC_SCHED + 10)
++#define TRC_SCHED_S_TIMER_FN (TRC_SCHED + 11)
++#define TRC_SCHED_T_TIMER_FN (TRC_SCHED + 12)
++#define TRC_SCHED_DOM_TIMER_FN (TRC_SCHED + 13)
++#define TRC_SCHED_SWITCH_INFPREV (TRC_SCHED + 14)
++#define TRC_SCHED_SWITCH_INFNEXT (TRC_SCHED + 15)
++
++#define TRC_MEM_PAGE_GRANT_MAP (TRC_MEM + 1)
++#define TRC_MEM_PAGE_GRANT_UNMAP (TRC_MEM + 2)
++#define TRC_MEM_PAGE_GRANT_TRANSFER (TRC_MEM + 3)
++
++/* trace events per subclass */
++#define TRC_VMX_VMEXIT (TRC_VMXEXIT + 1)
++#define TRC_VMX_VMENTRY (TRC_VMXEXIT + 2)
++
++#define TRC_VMX_TIMER_INTR (TRC_VMXTIMER + 1)
++
++#define TRC_VMX_INT (TRC_VMXINT + 1)
++
++
++/* This structure represents a single trace buffer record. */
++struct t_rec {
++ uint64_t cycles; /* cycle counter timestamp */
++ uint32_t event; /* event ID */
++ unsigned long data[5]; /* event data items */
++};
++
++/*
++ * This structure contains the metadata for a single trace buffer. The head
++ * field, indexes into an array of struct t_rec's.
++ */
++struct t_buf {
++ uint32_t cons; /* Next item to be consumed by control tools. */
++ uint32_t prod; /* Next item to be produced by Xen. */
++ /* 'nr_recs' records follow immediately after the meta-data header. */
++};
++
++#endif /* __XEN_PUBLIC_TRACE_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/vcpu.h new/include/xen/interface/vcpu.h
+--- linux-2.6/include/xen/interface/vcpu.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/vcpu.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,121 @@
++/******************************************************************************
++ * vcpu.h
++ *
++ * VCPU initialisation, query, and hotplug.
++ *
++ * Copyright (c) 2005, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_VCPU_H__
++#define __XEN_PUBLIC_VCPU_H__
++
++/*
++ * Prototype for this hypercall is:
++ * int vcpu_op(int cmd, int vcpuid, void *extra_args)
++ * @cmd == VCPUOP_??? (VCPU operation).
++ * @vcpuid == VCPU to operate on.
++ * @extra_args == Operation-specific extra arguments (NULL if none).
++ */
++
++/*
++ * Initialise a VCPU. Each VCPU can be initialised only once. A
++ * newly-initialised VCPU will not run until it is brought up by VCPUOP_up.
++ *
++ * @extra_arg == pointer to vcpu_guest_context structure containing initial
++ * state for the VCPU.
++ */
++#define VCPUOP_initialise 0
++
++/*
++ * Bring up a VCPU. This makes the VCPU runnable. This operation will fail
++ * if the VCPU has not been initialised (VCPUOP_initialise).
++ */
++#define VCPUOP_up 1
++
++/*
++ * Bring down a VCPU (i.e., make it non-runnable).
++ * There are a few caveats that callers should observe:
++ * 1. This operation may return, and VCPU_is_up may return false, before the
++ * VCPU stops running (i.e., the command is asynchronous). It is a good
++ * idea to ensure that the VCPU has entered a non-critical loop before
++ * bringing it down. Alternatively, this operation is guaranteed
++ * synchronous if invoked by the VCPU itself.
++ * 2. After a VCPU is initialised, there is currently no way to drop all its
++ * references to domain memory. Even a VCPU that is down still holds
++ * memory references via its pagetable base pointer and GDT. It is good
++ * practise to move a VCPU onto an 'idle' or default page table, LDT and
++ * GDT before bringing it down.
++ */
++#define VCPUOP_down 2
++
++/* Returns 1 if the given VCPU is up. */
++#define VCPUOP_is_up 3
++
++/*
++ * Return information about the state and running time of a VCPU.
++ * @extra_arg == pointer to vcpu_runstate_info structure.
++ */
++#define VCPUOP_get_runstate_info 4
++struct vcpu_runstate_info {
++ /* VCPU's current state (RUNSTATE_*). */
++ int state;
++ /* When was current state entered (system time, ns)? */
++ uint64_t state_entry_time;
++ /*
++ * Time spent in each RUNSTATE_* (ns). The sum of these times is
++ * guaranteed not to drift from system time.
++ */
++ uint64_t time[4];
++};
++typedef struct vcpu_runstate_info vcpu_runstate_info_t;
++
++/* VCPU is currently running on a physical CPU. */
++#define RUNSTATE_running 0
++
++/* VCPU is runnable, but not currently scheduled on any physical CPU. */
++#define RUNSTATE_runnable 1
++
++/* VCPU is blocked (a.k.a. idle). It is therefore not runnable. */
++#define RUNSTATE_blocked 2
++
++/*
++ * VCPU is not runnable, but it is not blocked.
++ * This is a 'catch all' state for things like hotplug and pauses by the
++ * system administrator (or for critical sections in the hypervisor).
++ * RUNSTATE_blocked dominates this state (it is the preferred state).
++ */
++#define RUNSTATE_offline 3
++
++/*
++ * Register a shared memory area from which the guest may obtain its own
++ * runstate information without needing to execute a hypercall.
++ * Notes:
++ * 1. The registered address may be virtual or physical, depending on the
++ * platform. The virtual address should be registered on x86 systems.
++ * 2. Only one shared area may be registered per VCPU. The shared area is
++ * updated by the hypervisor each time the VCPU is scheduled. Thus
++ * runstate.state will always be RUNSTATE_running and
++ * runstate.state_entry_time will indicate the system time at which the
++ * VCPU was last scheduled to run.
++ * @extra_arg == pointer to vcpu_register_runstate_memory_area structure.
++ */
++#define VCPUOP_register_runstate_memory_area 5
++struct vcpu_register_runstate_memory_area {
++ union {
++ struct vcpu_runstate_info *v;
++ uint64_t p;
++ } addr;
++};
++typedef struct vcpu_register_runstate_memory_area vcpu_register_runstate_memory_area_t;
++
++#endif /* __XEN_PUBLIC_VCPU_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/version.h new/include/xen/interface/version.h
+--- linux-2.6/include/xen/interface/version.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/version.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,70 @@
++/******************************************************************************
++ * version.h
++ *
++ * Xen version, type, and compile information.
++ *
++ * Copyright (c) 2005, Nguyen Anh Quynh <aquynh@gmail.com>
++ * Copyright (c) 2005, Keir Fraser <keir@xensource.com>
++ */
++
++#ifndef __XEN_PUBLIC_VERSION_H__
++#define __XEN_PUBLIC_VERSION_H__
++
++/* NB. All ops return zero on success, except XENVER_{version,pagesize} */
++
++/* arg == NULL; returns major:minor (16:16). */
++#define XENVER_version 0
++
++/* arg == xen_extraversion_t. */
++#define XENVER_extraversion 1
++typedef char xen_extraversion_t[16];
++#define XEN_EXTRAVERSION_LEN (sizeof(xen_extraversion_t))
++
++/* arg == xen_compile_info_t. */
++#define XENVER_compile_info 2
++struct xen_compile_info {
++ char compiler[64];
++ char compile_by[16];
++ char compile_domain[32];
++ char compile_date[32];
++};
++typedef struct xen_compile_info xen_compile_info_t;
++
++#define XENVER_capabilities 3
++typedef char xen_capabilities_info_t[1024];
++#define XEN_CAPABILITIES_INFO_LEN (sizeof(xen_capabilities_info_t))
++
++#define XENVER_changeset 4
++typedef char xen_changeset_info_t[64];
++#define XEN_CHANGESET_INFO_LEN (sizeof(xen_changeset_info_t))
++
++#define XENVER_platform_parameters 5
++struct xen_platform_parameters {
++ unsigned long virt_start;
++};
++typedef struct xen_platform_parameters xen_platform_parameters_t;
++
++#define XENVER_get_features 6
++struct xen_feature_info {
++ unsigned int submap_idx; /* IN: which 32-bit submap to return */
++ uint32_t submap; /* OUT: 32-bit submap */
++};
++typedef struct xen_feature_info xen_feature_info_t;
++
++/* Declares the features reported by XENVER_get_features. */
++#include "features.h"
++
++/* arg == NULL; returns host memory page size. */
++#define XENVER_pagesize 7
++
++#endif /* __XEN_PUBLIC_VERSION_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/xen-compat.h new/include/xen/interface/xen-compat.h
+--- linux-2.6/include/xen/interface/xen-compat.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/xen-compat.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,47 @@
++/******************************************************************************
++ * xen-compat.h
++ *
++ * Guest OS interface to Xen. Compatibility layer.
++ *
++ * Copyright (c) 2006, Christian Limpach
++ */
++
++#ifndef __XEN_PUBLIC_XEN_COMPAT_H__
++#define __XEN_PUBLIC_XEN_COMPAT_H__
++
++#define __XEN_LATEST_INTERFACE_VERSION__ 0x00030202
++
++#if defined(__XEN__) || defined(__XEN_TOOLS__)
++/* Xen is built with matching headers and implements the latest interface. */
++#define __XEN_INTERFACE_VERSION__ __XEN_LATEST_INTERFACE_VERSION__
++#elif !defined(__XEN_INTERFACE_VERSION__)
++/* Guests which do not specify a version get the legacy interface. */
++#define __XEN_INTERFACE_VERSION__ 0x00000000
++#endif
++
++#if __XEN_INTERFACE_VERSION__ > __XEN_LATEST_INTERFACE_VERSION__
++#error "These header files do not support the requested interface version."
++#endif
++
++/* New sched_op hypercall introduced in 0x00030101. */
++#if __XEN_INTERFACE_VERSION__ < 0x00030101
++#undef __HYPERVISOR_sched_op
++#define __HYPERVISOR_sched_op __HYPERVISOR_sched_op_compat
++#endif
++
++/* Structural guest handles introduced in 0x00030201. */
++#if __XEN_INTERFACE_VERSION__ < 0x00030201
++#undef __DEFINE_XEN_GUEST_HANDLE
++#define __DEFINE_XEN_GUEST_HANDLE(name, type) \
++ typedef type * __guest_handle_ ## name
++#endif
++
++/* New event-channel and physdev hypercalls introduced in 0x00030202. */
++#if __XEN_INTERFACE_VERSION__ < 0x00030202
++#undef __HYPERVISOR_event_channel_op
++#define __HYPERVISOR_event_channel_op __HYPERVISOR_event_channel_op_compat
++#undef __HYPERVISOR_physdev_op
++#define __HYPERVISOR_physdev_op __HYPERVISOR_physdev_op_compat
++#endif
++
++#endif /* __XEN_PUBLIC_XEN_COMPAT_H__ */
+diff -urNp linux-2.6/include/xen/interface/xen.h new/include/xen/interface/xen.h
+--- linux-2.6/include/xen/interface/xen.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/xen.h 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,505 @@
++/******************************************************************************
++ * xen.h
++ *
++ * Guest OS interface to Xen.
++ *
++ * Copyright (c) 2004, K A Fraser
++ */
++
++#ifndef __XEN_PUBLIC_XEN_H__
++#define __XEN_PUBLIC_XEN_H__
++
++#if defined(__i386__)
++#include "arch-x86_32.h"
++#elif defined(__x86_64__)
++#include "arch-x86_64.h"
++#elif defined(__ia64__)
++#include "arch-ia64.h"
++#else
++#error "Unsupported architecture"
++#endif
++
++/*
++ * XEN "SYSTEM CALLS" (a.k.a. HYPERCALLS).
++ */
++
++/*
++ * x86_32: EAX = vector; EBX, ECX, EDX, ESI, EDI = args 1, 2, 3, 4, 5.
++ * EAX = return value
++ * (argument registers may be clobbered on return)
++ * x86_64: RAX = vector; RDI, RSI, RDX, R10, R8, R9 = args 1, 2, 3, 4, 5, 6.
++ * RAX = return value
++ * (argument registers not clobbered on return; RCX, R11 are)
++ */
++#define __HYPERVISOR_set_trap_table 0
++#define __HYPERVISOR_mmu_update 1
++#define __HYPERVISOR_set_gdt 2
++#define __HYPERVISOR_stack_switch 3
++#define __HYPERVISOR_set_callbacks 4
++#define __HYPERVISOR_fpu_taskswitch 5
++#define __HYPERVISOR_sched_op_compat 6 /* compat since 0x00030101 */
++#define __HYPERVISOR_dom0_op 7
++#define __HYPERVISOR_set_debugreg 8
++#define __HYPERVISOR_get_debugreg 9
++#define __HYPERVISOR_update_descriptor 10
++#define __HYPERVISOR_memory_op 12
++#define __HYPERVISOR_multicall 13
++#define __HYPERVISOR_update_va_mapping 14
++#define __HYPERVISOR_set_timer_op 15
++#define __HYPERVISOR_event_channel_op_compat 16 /* compat since 0x00030202 */
++#define __HYPERVISOR_xen_version 17
++#define __HYPERVISOR_console_io 18
++#define __HYPERVISOR_physdev_op_compat 19 /* compat since 0x00030202 */
++#define __HYPERVISOR_grant_table_op 20
++#define __HYPERVISOR_vm_assist 21
++#define __HYPERVISOR_update_va_mapping_otherdomain 22
++#define __HYPERVISOR_iret 23 /* x86 only */
++#define __HYPERVISOR_vcpu_op 24
++#define __HYPERVISOR_set_segment_base 25 /* x86/64 only */
++#define __HYPERVISOR_mmuext_op 26
++#define __HYPERVISOR_acm_op 27
++#define __HYPERVISOR_nmi_op 28
++#define __HYPERVISOR_sched_op 29
++#define __HYPERVISOR_callback_op 30
++#define __HYPERVISOR_xenoprof_op 31
++#define __HYPERVISOR_event_channel_op 32
++#define __HYPERVISOR_physdev_op 33
++
++/* Architecture-specific hypercall definitions. */
++#define __HYPERVISOR_arch_0 48
++#define __HYPERVISOR_arch_1 49
++#define __HYPERVISOR_arch_2 50
++#define __HYPERVISOR_arch_3 51
++#define __HYPERVISOR_arch_4 52
++#define __HYPERVISOR_arch_5 53
++#define __HYPERVISOR_arch_6 54
++#define __HYPERVISOR_arch_7 55
++
++/*
++ * VIRTUAL INTERRUPTS
++ *
++ * Virtual interrupts that a guest OS may receive from Xen.
++ *
++ * In the side comments, 'V.' denotes a per-VCPU VIRQ while 'G.' denotes a
++ * global VIRQ. The former can be bound once per VCPU and cannot be re-bound.
++ * The latter can be allocated only once per guest: they must initially be
++ * allocated to VCPU0 but can subsequently be re-bound.
++ */
++#define VIRQ_TIMER 0 /* V. Timebase update, and/or requested timeout. */
++#define VIRQ_DEBUG 1 /* V. Request guest to dump debug info. */
++#define VIRQ_CONSOLE 2 /* G. (DOM0) Bytes received on emergency console. */
++#define VIRQ_DOM_EXC 3 /* G. (DOM0) Exceptional event for some domain. */
++#define VIRQ_TBUF 4 /* G. (DOM0) Trace buffer has records available. */
++#define VIRQ_DEBUGGER 6 /* G. (DOM0) A domain has paused for debugging. */
++#define VIRQ_XENOPROF 7 /* V. XenOprofile interrupt: new sample available */
++
++/* Architecture-specific VIRQ definitions. */
++#define VIRQ_ARCH_0 16
++#define VIRQ_ARCH_1 17
++#define VIRQ_ARCH_2 18
++#define VIRQ_ARCH_3 19
++#define VIRQ_ARCH_4 20
++#define VIRQ_ARCH_5 21
++#define VIRQ_ARCH_6 22
++#define VIRQ_ARCH_7 23
++
++#define NR_VIRQS 24
++
++/*
++ * MMU-UPDATE REQUESTS
++ *
++ * HYPERVISOR_mmu_update() accepts a list of (ptr, val) pairs.
++ * A foreigndom (FD) can be specified (or DOMID_SELF for none).
++ * Where the FD has some effect, it is described below.
++ * ptr[1:0] specifies the appropriate MMU_* command.
++ *
++ * ptr[1:0] == MMU_NORMAL_PT_UPDATE:
++ * Updates an entry in a page table. If updating an L1 table, and the new
++ * table entry is valid/present, the mapped frame must belong to the FD, if
++ * an FD has been specified. If attempting to map an I/O page then the
++ * caller assumes the privilege of the FD.
++ * FD == DOMID_IO: Permit /only/ I/O mappings, at the priv level of the caller.
++ * FD == DOMID_XEN: Map restricted areas of Xen's heap space.
++ * ptr[:2] -- Machine address of the page-table entry to modify.
++ * val -- Value to write.
++ *
++ * ptr[1:0] == MMU_MACHPHYS_UPDATE:
++ * Updates an entry in the machine->pseudo-physical mapping table.
++ * ptr[:2] -- Machine address within the frame whose mapping to modify.
++ * The frame must belong to the FD, if one is specified.
++ * val -- Value to write into the mapping entry.
++ */
++#define MMU_NORMAL_PT_UPDATE 0 /* checked '*ptr = val'. ptr is MA. */
++#define MMU_MACHPHYS_UPDATE 1 /* ptr = MA of frame to modify entry for */
++
++/*
++ * MMU EXTENDED OPERATIONS
++ *
++ * HYPERVISOR_mmuext_op() accepts a list of mmuext_op structures.
++ * A foreigndom (FD) can be specified (or DOMID_SELF for none).
++ * Where the FD has some effect, it is described below.
++ *
++ * cmd: MMUEXT_(UN)PIN_*_TABLE
++ * mfn: Machine frame number to be (un)pinned as a p.t. page.
++ * The frame must belong to the FD, if one is specified.
++ *
++ * cmd: MMUEXT_NEW_BASEPTR
++ * mfn: Machine frame number of new page-table base to install in MMU.
++ *
++ * cmd: MMUEXT_NEW_USER_BASEPTR [x86/64 only]
++ * mfn: Machine frame number of new page-table base to install in MMU
++ * when in user space.
++ *
++ * cmd: MMUEXT_TLB_FLUSH_LOCAL
++ * No additional arguments. Flushes local TLB.
++ *
++ * cmd: MMUEXT_INVLPG_LOCAL
++ * linear_addr: Linear address to be flushed from the local TLB.
++ *
++ * cmd: MMUEXT_TLB_FLUSH_MULTI
++ * vcpumask: Pointer to bitmap of VCPUs to be flushed.
++ *
++ * cmd: MMUEXT_INVLPG_MULTI
++ * linear_addr: Linear address to be flushed.
++ * vcpumask: Pointer to bitmap of VCPUs to be flushed.
++ *
++ * cmd: MMUEXT_TLB_FLUSH_ALL
++ * No additional arguments. Flushes all VCPUs' TLBs.
++ *
++ * cmd: MMUEXT_INVLPG_ALL
++ * linear_addr: Linear address to be flushed from all VCPUs' TLBs.
++ *
++ * cmd: MMUEXT_FLUSH_CACHE
++ * No additional arguments. Writes back and flushes cache contents.
++ *
++ * cmd: MMUEXT_SET_LDT
++ * linear_addr: Linear address of LDT base (NB. must be page-aligned).
++ * nr_ents: Number of entries in LDT.
++ */
++#define MMUEXT_PIN_L1_TABLE 0
++#define MMUEXT_PIN_L2_TABLE 1
++#define MMUEXT_PIN_L3_TABLE 2
++#define MMUEXT_PIN_L4_TABLE 3
++#define MMUEXT_UNPIN_TABLE 4
++#define MMUEXT_NEW_BASEPTR 5
++#define MMUEXT_TLB_FLUSH_LOCAL 6
++#define MMUEXT_INVLPG_LOCAL 7
++#define MMUEXT_TLB_FLUSH_MULTI 8
++#define MMUEXT_INVLPG_MULTI 9
++#define MMUEXT_TLB_FLUSH_ALL 10
++#define MMUEXT_INVLPG_ALL 11
++#define MMUEXT_FLUSH_CACHE 12
++#define MMUEXT_SET_LDT 13
++#define MMUEXT_NEW_USER_BASEPTR 15
++
++#ifndef __ASSEMBLY__
++struct mmuext_op {
++ unsigned int cmd;
++ union {
++ /* [UN]PIN_TABLE, NEW_BASEPTR, NEW_USER_BASEPTR */
++ xen_pfn_t mfn;
++ /* INVLPG_LOCAL, INVLPG_ALL, SET_LDT */
++ unsigned long linear_addr;
++ } arg1;
++ union {
++ /* SET_LDT */
++ unsigned int nr_ents;
++ /* TLB_FLUSH_MULTI, INVLPG_MULTI */
++ void *vcpumask;
++ } arg2;
++};
++typedef struct mmuext_op mmuext_op_t;
++DEFINE_XEN_GUEST_HANDLE(mmuext_op_t);
++#endif
++
++/* These are passed as 'flags' to update_va_mapping. They can be ORed. */
++/* When specifying UVMF_MULTI, also OR in a pointer to a CPU bitmap. */
++/* UVMF_LOCAL is merely UVMF_MULTI with a NULL bitmap pointer. */
++#define UVMF_NONE (0UL<<0) /* No flushing at all. */
++#define UVMF_TLB_FLUSH (1UL<<0) /* Flush entire TLB(s). */
++#define UVMF_INVLPG (2UL<<0) /* Flush only one entry. */
++#define UVMF_FLUSHTYPE_MASK (3UL<<0)
++#define UVMF_MULTI (0UL<<2) /* Flush subset of TLBs. */
++#define UVMF_LOCAL (0UL<<2) /* Flush local TLB. */
++#define UVMF_ALL (1UL<<2) /* Flush all TLBs. */
++
++/*
++ * Commands to HYPERVISOR_console_io().
++ */
++#define CONSOLEIO_write 0
++#define CONSOLEIO_read 1
++
++/*
++ * Commands to HYPERVISOR_vm_assist().
++ */
++#define VMASST_CMD_enable 0
++#define VMASST_CMD_disable 1
++
++/* x86/32 guests: simulate full 4GB segment limits. */
++#define VMASST_TYPE_4gb_segments 0
++
++/* x86/32 guests: trap (vector 15) whenever above vmassist is used. */
++#define VMASST_TYPE_4gb_segments_notify 1
++
++/*
++ * x86 guests: support writes to bottom-level PTEs.
++ * NB1. Page-directory entries cannot be written.
++ * NB2. Guest must continue to remove all writable mappings of PTEs.
++ */
++#define VMASST_TYPE_writable_pagetables 2
++
++/* x86/PAE guests: support PDPTs above 4GB. */
++#define VMASST_TYPE_pae_extended_cr3 3
++
++#define MAX_VMASST_TYPE 3
++
++#ifndef __ASSEMBLY__
++
++typedef uint16_t domid_t;
++
++/* Domain ids >= DOMID_FIRST_RESERVED cannot be used for ordinary domains. */
++#define DOMID_FIRST_RESERVED (0x7FF0U)
++
++/* DOMID_SELF is used in certain contexts to refer to oneself. */
++#define DOMID_SELF (0x7FF0U)
++
++/*
++ * DOMID_IO is used to restrict page-table updates to mapping I/O memory.
++ * Although no Foreign Domain need be specified to map I/O pages, DOMID_IO
++ * is useful to ensure that no mappings to the OS's own heap are accidentally
++ * installed. (e.g., in Linux this could cause havoc as reference counts
++ * aren't adjusted on the I/O-mapping code path).
++ * This only makes sense in MMUEXT_SET_FOREIGNDOM, but in that context can
++ * be specified by any calling domain.
++ */
++#define DOMID_IO (0x7FF1U)
++
++/*
++ * DOMID_XEN is used to allow privileged domains to map restricted parts of
++ * Xen's heap space (e.g., the machine_to_phys table).
++ * This only makes sense in MMUEXT_SET_FOREIGNDOM, and is only permitted if
++ * the caller is privileged.
++ */
++#define DOMID_XEN (0x7FF2U)
++
++/*
++ * Send an array of these to HYPERVISOR_mmu_update().
++ * NB. The fields are natural pointer/address size for this architecture.
++ */
++struct mmu_update {
++ uint64_t ptr; /* Machine address of PTE. */
++ uint64_t val; /* New contents of PTE. */
++};
++typedef struct mmu_update mmu_update_t;
++DEFINE_XEN_GUEST_HANDLE(mmu_update_t);
++
++/*
++ * Send an array of these to HYPERVISOR_multicall().
++ * NB. The fields are natural register size for this architecture.
++ */
++struct multicall_entry {
++ unsigned long op, result;
++ unsigned long args[6];
++};
++typedef struct multicall_entry multicall_entry_t;
++DEFINE_XEN_GUEST_HANDLE(multicall_entry_t);
++
++/*
++ * Event channel endpoints per domain:
++ * 1024 if a long is 32 bits; 4096 if a long is 64 bits.
++ */
++#define NR_EVENT_CHANNELS (sizeof(unsigned long) * sizeof(unsigned long) * 64)
++
++struct vcpu_time_info {
++ /*
++ * Updates to the following values are preceded and followed by an
++ * increment of 'version'. The guest can therefore detect updates by
++ * looking for changes to 'version'. If the least-significant bit of
++ * the version number is set then an update is in progress and the guest
++ * must wait to read a consistent set of values.
++ * The correct way to interact with the version number is similar to
++ * Linux's seqlock: see the implementations of read_seqbegin/read_seqretry.
++ */
++ uint32_t version;
++ uint32_t pad0;
++ uint64_t tsc_timestamp; /* TSC at last update of time vals. */
++ uint64_t system_time; /* Time, in nanosecs, since boot. */
++ /*
++ * Current system time:
++ * system_time +
++ * ((((tsc - tsc_timestamp) << tsc_shift) * tsc_to_system_mul) >> 32)
++ * CPU frequency (Hz):
++ * ((10^9 << 32) / tsc_to_system_mul) >> tsc_shift
++ */
++ uint32_t tsc_to_system_mul;
++ int8_t tsc_shift;
++ int8_t pad1[3];
++}; /* 32 bytes */
++typedef struct vcpu_time_info vcpu_time_info_t;
++
++struct vcpu_info {
++ /*
++ * 'evtchn_upcall_pending' is written non-zero by Xen to indicate
++ * a pending notification for a particular VCPU. It is then cleared
++ * by the guest OS /before/ checking for pending work, thus avoiding
++ * a set-and-check race. Note that the mask is only accessed by Xen
++ * on the CPU that is currently hosting the VCPU. This means that the
++ * pending and mask flags can be updated by the guest without special
++ * synchronisation (i.e., no need for the x86 LOCK prefix).
++ * This may seem suboptimal because if the pending flag is set by
++ * a different CPU then an IPI may be scheduled even when the mask
++ * is set. However, note:
++ * 1. The task of 'interrupt holdoff' is covered by the per-event-
++ * channel mask bits. A 'noisy' event that is continually being
++ * triggered can be masked at source at this very precise
++ * granularity.
++ * 2. The main purpose of the per-VCPU mask is therefore to restrict
++ * reentrant execution: whether for concurrency control, or to
++ * prevent unbounded stack usage. Whatever the purpose, we expect
++ * that the mask will be asserted only for short periods at a time,
++ * and so the likelihood of a 'spurious' IPI is suitably small.
++ * The mask is read before making an event upcall to the guest: a
++ * non-zero mask therefore guarantees that the VCPU will not receive
++ * an upcall activation. The mask is cleared when the VCPU requests
++ * to block: this avoids wakeup-waiting races.
++ */
++ uint8_t evtchn_upcall_pending;
++ uint8_t evtchn_upcall_mask;
++ unsigned long evtchn_pending_sel;
++ struct arch_vcpu_info arch;
++ struct vcpu_time_info time;
++}; /* 64 bytes (x86) */
++typedef struct vcpu_info vcpu_info_t;
++
++/*
++ * Xen/kernel shared data -- pointer provided in start_info.
++ * NB. We expect that this struct is smaller than a page.
++ */
++struct shared_info {
++ struct vcpu_info vcpu_info[MAX_VIRT_CPUS];
++
++ /*
++ * A domain can create "event channels" on which it can send and receive
++ * asynchronous event notifications. There are three classes of event that
++ * are delivered by this mechanism:
++ * 1. Bi-directional inter- and intra-domain connections. Domains must
++ * arrange out-of-band to set up a connection (usually by allocating
++ * an unbound 'listener' port and avertising that via a storage service
++ * such as xenstore).
++ * 2. Physical interrupts. A domain with suitable hardware-access
++ * privileges can bind an event-channel port to a physical interrupt
++ * source.
++ * 3. Virtual interrupts ('events'). A domain can bind an event-channel
++ * port to a virtual interrupt source, such as the virtual-timer
++ * device or the emergency console.
++ *
++ * Event channels are addressed by a "port index". Each channel is
++ * associated with two bits of information:
++ * 1. PENDING -- notifies the domain that there is a pending notification
++ * to be processed. This bit is cleared by the guest.
++ * 2. MASK -- if this bit is clear then a 0->1 transition of PENDING
++ * will cause an asynchronous upcall to be scheduled. This bit is only
++ * updated by the guest. It is read-only within Xen. If a channel
++ * becomes pending while the channel is masked then the 'edge' is lost
++ * (i.e., when the channel is unmasked, the guest must manually handle
++ * pending notifications as no upcall will be scheduled by Xen).
++ *
++ * To expedite scanning of pending notifications, any 0->1 pending
++ * transition on an unmasked channel causes a corresponding bit in a
++ * per-vcpu selector word to be set. Each bit in the selector covers a
++ * 'C long' in the PENDING bitfield array.
++ */
++ unsigned long evtchn_pending[sizeof(unsigned long) * 8];
++ unsigned long evtchn_mask[sizeof(unsigned long) * 8];
++
++ /*
++ * Wallclock time: updated only by control software. Guests should base
++ * their gettimeofday() syscall on this wallclock-base value.
++ */
++ uint32_t wc_version; /* Version counter: see vcpu_time_info_t. */
++ uint32_t wc_sec; /* Secs 00:00:00 UTC, Jan 1, 1970. */
++ uint32_t wc_nsec; /* Nsecs 00:00:00 UTC, Jan 1, 1970. */
++
++ struct arch_shared_info arch;
++
++};
++typedef struct shared_info shared_info_t;
++
++/*
++ * Start-of-day memory layout for the initial domain (DOM0):
++ * 1. The domain is started within contiguous virtual-memory region.
++ * 2. The contiguous region begins and ends on an aligned 4MB boundary.
++ * 3. The region start corresponds to the load address of the OS image.
++ * If the load address is not 4MB aligned then the address is rounded down.
++ * 4. This the order of bootstrap elements in the initial virtual region:
++ * a. relocated kernel image
++ * b. initial ram disk [mod_start, mod_len]
++ * c. list of allocated page frames [mfn_list, nr_pages]
++ * d. start_info_t structure [register ESI (x86)]
++ * e. bootstrap page tables [pt_base, CR3 (x86)]
++ * f. bootstrap stack [register ESP (x86)]
++ * 5. Bootstrap elements are packed together, but each is 4kB-aligned.
++ * 6. The initial ram disk may be omitted.
++ * 7. The list of page frames forms a contiguous 'pseudo-physical' memory
++ * layout for the domain. In particular, the bootstrap virtual-memory
++ * region is a 1:1 mapping to the first section of the pseudo-physical map.
++ * 8. All bootstrap elements are mapped read-writable for the guest OS. The
++ * only exception is the bootstrap page table, which is mapped read-only.
++ * 9. There is guaranteed to be at least 512kB padding after the final
++ * bootstrap element. If necessary, the bootstrap virtual region is
++ * extended by an extra 4MB to ensure this.
++ */
++
++#define MAX_GUEST_CMDLINE 1024
++struct start_info {
++ /* THE FOLLOWING ARE FILLED IN BOTH ON INITIAL BOOT AND ON RESUME. */
++ char magic[32]; /* "xen-<version>-<platform>". */
++ unsigned long nr_pages; /* Total pages allocated to this domain. */
++ unsigned long shared_info; /* MACHINE address of shared info struct. */
++ uint32_t flags; /* SIF_xxx flags. */
++ xen_pfn_t store_mfn; /* MACHINE page number of shared page. */
++ uint32_t store_evtchn; /* Event channel for store communication. */
++ xen_pfn_t console_mfn; /* MACHINE page number of console page. */
++ uint32_t console_evtchn; /* Event channel for console messages. */
++ /* THE FOLLOWING ARE ONLY FILLED IN ON INITIAL BOOT (NOT RESUME). */
++ unsigned long pt_base; /* VIRTUAL address of page directory. */
++ unsigned long nr_pt_frames; /* Number of bootstrap p.t. frames. */
++ unsigned long mfn_list; /* VIRTUAL address of page-frame list. */
++ unsigned long mod_start; /* VIRTUAL address of pre-loaded module. */
++ unsigned long mod_len; /* Size (bytes) of pre-loaded module. */
++ int8_t cmd_line[MAX_GUEST_CMDLINE];
++};
++typedef struct start_info start_info_t;
++
++/* These flags are passed in the 'flags' field of start_info_t. */
++#define SIF_PRIVILEGED (1<<0) /* Is the domain privileged? */
++#define SIF_INITDOMAIN (1<<1) /* Is this the initial control domain? */
++
++typedef uint64_t cpumap_t;
++
++typedef uint8_t xen_domain_handle_t[16];
++
++/* Turn a plain number into a C unsigned long constant. */
++#define __mk_unsigned_long(x) x ## UL
++#define mk_unsigned_long(x) __mk_unsigned_long(x)
++
++#else /* __ASSEMBLY__ */
++
++/* In assembly code we cannot use C numeric constant suffixes. */
++#define mk_unsigned_long(x) x
++
++#endif /* !__ASSEMBLY__ */
++
++#include "xen-compat.h"
++
++#endif /* __XEN_PUBLIC_XEN_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/interface/xenoprof.h new/include/xen/interface/xenoprof.h
+--- linux-2.6/include/xen/interface/xenoprof.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/interface/xenoprof.h 2006-07-07 15:10:03.000000000 +0200
+@@ -0,0 +1,103 @@
++/******************************************************************************
++ * xenoprof.h
++ *
++ * Interface for enabling system wide profiling based on hardware performance
++ * counters
++ *
++ * Copyright (C) 2005 Hewlett-Packard Co.
++ * Written by Aravind Menon & Jose Renato Santos
++ */
++
++#ifndef __XEN_PUBLIC_XENOPROF_H__
++#define __XEN_PUBLIC_XENOPROF_H__
++
++/*
++ * Commands to HYPERVISOR_xenoprof_op().
++ */
++#define XENOPROF_init 0
++#define XENOPROF_reset_active_list 1
++#define XENOPROF_reset_passive_list 2
++#define XENOPROF_set_active 3
++#define XENOPROF_set_passive 4
++#define XENOPROF_reserve_counters 5
++#define XENOPROF_counter 6
++#define XENOPROF_setup_events 7
++#define XENOPROF_enable_virq 8
++#define XENOPROF_start 9
++#define XENOPROF_stop 10
++#define XENOPROF_disable_virq 11
++#define XENOPROF_release_counters 12
++#define XENOPROF_shutdown 13
++
++#define MAX_OPROF_EVENTS 32
++#define MAX_OPROF_DOMAINS 25
++#define XENOPROF_CPU_TYPE_SIZE 64
++
++/* Xenoprof performance events (not Xen events) */
++struct event_log {
++ uint64_t eip;
++ uint8_t mode;
++ uint8_t event;
++};
++
++/* Xenoprof buffer shared between Xen and domain - 1 per VCPU */
++struct xenoprof_buf {
++ uint32_t event_head;
++ uint32_t event_tail;
++ uint32_t event_size;
++ uint32_t vcpu_id;
++ uint64_t xen_samples;
++ uint64_t kernel_samples;
++ uint64_t user_samples;
++ uint64_t lost_samples;
++ struct event_log event_log[1];
++};
++typedef struct xenoprof_buf xenoprof_buf_t;
++DEFINE_XEN_GUEST_HANDLE(xenoprof_buf_t);
++
++struct xenoprof_init {
++ int32_t max_samples;
++ int32_t num_events;
++ int32_t is_primary;
++ int32_t nbuf;
++ int32_t bufsize;
++ uint64_t buf_maddr;
++ char cpu_type[XENOPROF_CPU_TYPE_SIZE];
++};
++typedef struct xenoprof_init xenoprof_init_t;
++DEFINE_XEN_GUEST_HANDLE(xenoprof_init_t);
++
++struct xenoprof_counter {
++ uint32_t ind;
++ uint64_t count;
++ uint32_t enabled;
++ uint32_t event;
++ uint32_t hypervisor;
++ uint32_t kernel;
++ uint32_t user;
++ uint64_t unit_mask;
++};
++typedef struct xenoprof_counter xenoprof_counter_t;
++DEFINE_XEN_GUEST_HANDLE(xenoprof_counter_t);
++
++typedef struct xenoprof_passive {
++ uint16_t domain_id;
++ int32_t max_samples;
++ int32_t nbuf;
++ int32_t bufsize;
++ uint64_t buf_maddr;
++} xenoprof_passive_t;
++DEFINE_XEN_GUEST_HANDLE(xenoprof_passive_t);
++
++
++#endif /* __XEN_PUBLIC_XENOPROF_H__ */
++
++/*
++ * Local variables:
++ * mode: C
++ * c-set-style: "BSD"
++ * c-basic-offset: 4
++ * tab-width: 4
++ * indent-tabs-mode: nil
++ * End:
++ */
+diff -urNp linux-2.6/include/xen/pcifront.h new/include/xen/pcifront.h
+--- linux-2.6/include/xen/pcifront.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/pcifront.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,39 @@
++/*
++ * PCI Frontend - arch-dependendent declarations
++ *
++ * Author: Ryan Wilson <hap9@epoch.ncsc.mil>
++ */
++#ifndef __XEN_ASM_PCIFRONT_H__
++#define __XEN_ASM_PCIFRONT_H__
++
++#include <linux/config.h>
++#include <linux/spinlock.h>
++
++#ifdef __KERNEL__
++
++struct pcifront_device;
++
++struct pcifront_sd {
++ int domain;
++ struct pcifront_device *pdev;
++};
++
++struct pci_bus;
++
++#ifdef CONFIG_PCI_DOMAINS
++static inline int pci_domain_nr(struct pci_bus *bus)
++{
++ struct pcifront_sd *sd = bus->sysdata;
++ return sd->domain;
++}
++static inline int pci_proc_domain(struct pci_bus *bus)
++{
++ return pci_domain_nr(bus);
++}
++#endif /* CONFIG_PCI_DOMAINS */
++
++extern spinlock_t pci_bus_lock;
++
++#endif /* __KERNEL__ */
++
++#endif /* __XEN_ASM_PCIFRONT_H__ */
+diff -urNp linux-2.6/include/xen/public/evtchn.h new/include/xen/public/evtchn.h
+--- linux-2.6/include/xen/public/evtchn.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/public/evtchn.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,91 @@
++/******************************************************************************
++ * evtchn.h
++ *
++ * Interface to /dev/xen/evtchn.
++ *
++ * Copyright (c) 2003-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __LINUX_PUBLIC_EVTCHN_H__
++#define __LINUX_PUBLIC_EVTCHN_H__
++
++/* /dev/xen/evtchn resides at device number major=10, minor=201 */
++#define EVTCHN_MINOR 201
++
++/*
++ * Bind a fresh port to VIRQ @virq.
++ * Return allocated port.
++ */
++#define IOCTL_EVTCHN_BIND_VIRQ \
++ _IOC(_IOC_NONE, 'E', 0, sizeof(struct ioctl_evtchn_bind_virq))
++struct ioctl_evtchn_bind_virq {
++ unsigned int virq;
++};
++
++/*
++ * Bind a fresh port to remote <@remote_domain, @remote_port>.
++ * Return allocated port.
++ */
++#define IOCTL_EVTCHN_BIND_INTERDOMAIN \
++ _IOC(_IOC_NONE, 'E', 1, sizeof(struct ioctl_evtchn_bind_interdomain))
++struct ioctl_evtchn_bind_interdomain {
++ unsigned int remote_domain, remote_port;
++};
++
++/*
++ * Allocate a fresh port for binding to @remote_domain.
++ * Return allocated port.
++ */
++#define IOCTL_EVTCHN_BIND_UNBOUND_PORT \
++ _IOC(_IOC_NONE, 'E', 2, sizeof(struct ioctl_evtchn_bind_unbound_port))
++struct ioctl_evtchn_bind_unbound_port {
++ unsigned int remote_domain;
++};
++
++/*
++ * Unbind previously allocated @port.
++ */
++#define IOCTL_EVTCHN_UNBIND \
++ _IOC(_IOC_NONE, 'E', 3, sizeof(struct ioctl_evtchn_unbind))
++struct ioctl_evtchn_unbind {
++ unsigned int port;
++};
++
++/*
++ * Unbind previously allocated @port.
++ */
++#define IOCTL_EVTCHN_NOTIFY \
++ _IOC(_IOC_NONE, 'E', 4, sizeof(struct ioctl_evtchn_notify))
++struct ioctl_evtchn_notify {
++ unsigned int port;
++};
++
++/* Clear and reinitialise the event buffer. Clear error condition. */
++#define IOCTL_EVTCHN_RESET \
++ _IOC(_IOC_NONE, 'E', 5, 0)
++
++#endif /* __LINUX_PUBLIC_EVTCHN_H__ */
+diff -urNp linux-2.6/include/xen/public/privcmd.h new/include/xen/public/privcmd.h
+--- linux-2.6/include/xen/public/privcmd.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/public/privcmd.h 2006-06-07 13:29:36.000000000 +0200
+@@ -0,0 +1,79 @@
++/******************************************************************************
++ * privcmd.h
++ *
++ * Interface to /proc/xen/privcmd.
++ *
++ * Copyright (c) 2003-2005, K A Fraser
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef __LINUX_PUBLIC_PRIVCMD_H__
++#define __LINUX_PUBLIC_PRIVCMD_H__
++
++#include <linux/types.h>
++
++#ifndef __user
++#define __user
++#endif
++
++typedef struct privcmd_hypercall
++{
++ __u64 op;
++ __u64 arg[5];
++} privcmd_hypercall_t;
++
++typedef struct privcmd_mmap_entry {
++ __u64 va;
++ __u64 mfn;
++ __u64 npages;
++} privcmd_mmap_entry_t;
++
++typedef struct privcmd_mmap {
++ int num;
++ domid_t dom; /* target domain */
++ privcmd_mmap_entry_t __user *entry;
++} privcmd_mmap_t;
++
++typedef struct privcmd_mmapbatch {
++ int num; /* number of pages to populate */
++ domid_t dom; /* target domain */
++ __u64 addr; /* virtual address */
++ xen_pfn_t __user *arr; /* array of mfns - top nibble set on err */
++} privcmd_mmapbatch_t;
++
++/*
++ * @cmd: IOCTL_PRIVCMD_HYPERCALL
++ * @arg: &privcmd_hypercall_t
++ * Return: Value returned from execution of the specified hypercall.
++ */
++#define IOCTL_PRIVCMD_HYPERCALL \
++ _IOC(_IOC_NONE, 'P', 0, sizeof(privcmd_hypercall_t))
++#define IOCTL_PRIVCMD_MMAP \
++ _IOC(_IOC_NONE, 'P', 2, sizeof(privcmd_mmap_t))
++#define IOCTL_PRIVCMD_MMAPBATCH \
++ _IOC(_IOC_NONE, 'P', 3, sizeof(privcmd_mmapbatch_t))
++
++#endif /* __LINUX_PUBLIC_PRIVCMD_H__ */
+diff -urNp linux-2.6/include/xen/xenbus.h new/include/xen/xenbus.h
+--- linux-2.6/include/xen/xenbus.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/xenbus.h 2006-06-28 14:32:14.000000000 +0200
+@@ -0,0 +1,299 @@
++/******************************************************************************
++ * xenbus.h
++ *
++ * Talks to Xen Store to figure out what devices we have.
++ *
++ * Copyright (C) 2005 Rusty Russell, IBM Corporation
++ * Copyright (C) 2005 XenSource Ltd.
++ *
++ * This program is free software; you can redistribute it and/or
++ * modify it under the terms of the GNU General Public License version 2
++ * as published by the Free Software Foundation; or, when distributed
++ * separately from the Linux kernel or incorporated into other
++ * software packages, subject to the following license:
++ *
++ * Permission is hereby granted, free of charge, to any person obtaining a copy
++ * of this source file (the "Software"), to deal in the Software without
++ * restriction, including without limitation the rights to use, copy, modify,
++ * merge, publish, distribute, sublicense, and/or sell copies of the Software,
++ * and to permit persons to whom the Software is furnished to do so, subject to
++ * the following conditions:
++ *
++ * The above copyright notice and this permission notice shall be included in
++ * all copies or substantial portions of the Software.
++ *
++ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
++ * IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
++ * FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
++ * AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
++ * LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING
++ * FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS
++ * IN THE SOFTWARE.
++ */
++
++#ifndef _XEN_XENBUS_H
++#define _XEN_XENBUS_H
++
++#include <linux/device.h>
++#include <linux/notifier.h>
++#include <linux/mutex.h>
++#include <xen/interface/xen.h>
++#include <xen/interface/grant_table.h>
++#include <xen/interface/io/xenbus.h>
++#include <xen/interface/io/xs_wire.h>
++
++/* Register callback to watch this node. */
++struct xenbus_watch
++{
++ struct list_head list;
++
++ /* Path being watched. */
++ const char *node;
++
++ /* Callback (executed in a process context with no locks held). */
++ void (*callback)(struct xenbus_watch *,
++ const char **vec, unsigned int len);
++
++ /* See XBWF_ definitions below. */
++ unsigned long flags;
++};
++
++/*
++ * Execute callback in its own kthread. Useful if the callback is long
++ * running or heavily serialised, to avoid taking out the main xenwatch thread
++ * for a long period of time (or even unwittingly causing a deadlock).
++ */
++#define XBWF_new_thread 1
++
++/* A xenbus device. */
++struct xenbus_device {
++ const char *devicetype;
++ const char *nodename;
++ const char *otherend;
++ int otherend_id;
++ struct xenbus_watch otherend_watch;
++ struct device dev;
++ enum xenbus_state state;
++};
++
++static inline struct xenbus_device *to_xenbus_device(struct device *dev)
++{
++ return container_of(dev, struct xenbus_device, dev);
++}
++
++struct xenbus_device_id
++{
++ /* .../device/<device_type>/<identifier> */
++ char devicetype[32]; /* General class of device. */
++};
++
++/* A xenbus driver. */
++struct xenbus_driver {
++ char *name;
++ struct module *owner;
++ const struct xenbus_device_id *ids;
++ int (*probe)(struct xenbus_device *dev,
++ const struct xenbus_device_id *id);
++ void (*otherend_changed)(struct xenbus_device *dev,
++ enum xenbus_state backend_state);
++ int (*remove)(struct xenbus_device *dev);
++ int (*suspend)(struct xenbus_device *dev);
++ int (*resume)(struct xenbus_device *dev);
++ int (*uevent)(struct xenbus_device *, char **, int, char *, int);
++ struct device_driver driver;
++ int (*read_otherend_details)(struct xenbus_device *dev);
++};
++
++static inline struct xenbus_driver *to_xenbus_driver(struct device_driver *drv)
++{
++ return container_of(drv, struct xenbus_driver, driver);
++}
++
++int xenbus_register_frontend(struct xenbus_driver *drv);
++int xenbus_register_backend(struct xenbus_driver *drv);
++void xenbus_unregister_driver(struct xenbus_driver *drv);
++
++struct xenbus_transaction
++{
++ u32 id;
++};
++
++/* Nil transaction ID. */
++#define XBT_NIL ((struct xenbus_transaction) { 0 })
++
++char **xenbus_directory(struct xenbus_transaction t,
++ const char *dir, const char *node, unsigned int *num);
++void *xenbus_read(struct xenbus_transaction t,
++ const char *dir, const char *node, unsigned int *len);
++int xenbus_write(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *string);
++int xenbus_mkdir(struct xenbus_transaction t,
++ const char *dir, const char *node);
++int xenbus_exists(struct xenbus_transaction t,
++ const char *dir, const char *node);
++int xenbus_rm(struct xenbus_transaction t, const char *dir, const char *node);
++int xenbus_transaction_start(struct xenbus_transaction *t);
++int xenbus_transaction_end(struct xenbus_transaction t, int abort);
++
++/* Single read and scanf: returns -errno or num scanned if > 0. */
++int xenbus_scanf(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *fmt, ...)
++ __attribute__((format(scanf, 4, 5)));
++
++/* Single printf and write: returns -errno or 0. */
++int xenbus_printf(struct xenbus_transaction t,
++ const char *dir, const char *node, const char *fmt, ...)
++ __attribute__((format(printf, 4, 5)));
++
++/* Generic read function: NULL-terminated triples of name,
++ * sprintf-style type string, and pointer. Returns 0 or errno.*/
++int xenbus_gather(struct xenbus_transaction t, const char *dir, ...);
++
++/* notifer routines for when the xenstore comes up */
++int register_xenstore_notifier(struct notifier_block *nb);
++void unregister_xenstore_notifier(struct notifier_block *nb);
++
++int register_xenbus_watch(struct xenbus_watch *watch);
++void unregister_xenbus_watch(struct xenbus_watch *watch);
++void xs_suspend(void);
++void xs_resume(void);
++
++/* Used by xenbus_dev to borrow kernel's store connection. */
++void *xenbus_dev_request_and_reply(struct xsd_sockmsg *msg);
++
++/* Called from xen core code. */
++void xenbus_suspend(void);
++void xenbus_resume(void);
++
++#define XENBUS_IS_ERR_READ(str) ({ \
++ if (!IS_ERR(str) && strlen(str) == 0) { \
++ kfree(str); \
++ str = ERR_PTR(-ERANGE); \
++ } \
++ IS_ERR(str); \
++})
++
++#define XENBUS_EXIST_ERR(err) ((err) == -ENOENT || (err) == -ERANGE)
++
++
++/**
++ * Register a watch on the given path, using the given xenbus_watch structure
++ * for storage, and the given callback function as the callback. Return 0 on
++ * success, or -errno on error. On success, the given path will be saved as
++ * watch->node, and remains the caller's to free. On error, watch->node will
++ * be NULL, the device will switch to XenbusStateClosing, and the error will
++ * be saved in the store.
++ */
++int xenbus_watch_path(struct xenbus_device *dev, const char *path,
++ struct xenbus_watch *watch,
++ void (*callback)(struct xenbus_watch *,
++ const char **, unsigned int));
++
++
++/**
++ * Register a watch on the given path/path2, using the given xenbus_watch
++ * structure for storage, and the given callback function as the callback.
++ * Return 0 on success, or -errno on error. On success, the watched path
++ * (path/path2) will be saved as watch->node, and becomes the caller's to
++ * kfree(). On error, watch->node will be NULL, so the caller has nothing to
++ * free, the device will switch to XenbusStateClosing, and the error will be
++ * saved in the store.
++ */
++int xenbus_watch_path2(struct xenbus_device *dev, const char *path,
++ const char *path2, struct xenbus_watch *watch,
++ void (*callback)(struct xenbus_watch *,
++ const char **, unsigned int));
++
++
++/**
++ * Advertise in the store a change of the given driver to the given new_state.
++ * Return 0 on success, or -errno on error. On error, the device will switch
++ * to XenbusStateClosing, and the error will be saved in the store.
++ */
++int xenbus_switch_state(struct xenbus_device *dev, enum xenbus_state new_state);
++
++
++/**
++ * Grant access to the given ring_mfn to the peer of the given device. Return
++ * 0 on success, or -errno on error. On error, the device will switch to
++ * XenbusStateClosing, and the error will be saved in the store.
++ */
++int xenbus_grant_ring(struct xenbus_device *dev, unsigned long ring_mfn);
++
++
++/**
++ * Map a page of memory into this domain from another domain's grant table.
++ * xenbus_map_ring_valloc allocates a page of virtual address space, maps the
++ * page to that address, and sets *vaddr to that address.
++ * xenbus_map_ring does not allocate the virtual address space (you must do
++ * this yourself!). It only maps in the page to the specified address.
++ * Returns 0 on success, and GNTST_* (see xen/include/interface/grant_table.h)
++ * or -ENOMEM on error. If an error is returned, device will switch to
++ * XenbusStateClosing and the error message will be saved in XenStore.
++ */
++struct vm_struct *xenbus_map_ring_valloc(struct xenbus_device *dev,
++ int gnt_ref);
++int xenbus_map_ring(struct xenbus_device *dev, int gnt_ref,
++ grant_handle_t *handle, void *vaddr);
++
++
++/**
++ * Unmap a page of memory in this domain that was imported from another domain.
++ * Use xenbus_unmap_ring_vfree if you mapped in your memory with
++ * xenbus_map_ring_valloc (it will free the virtual address space).
++ * Returns 0 on success and returns GNTST_* on error
++ * (see xen/include/interface/grant_table.h).
++ */
++int xenbus_unmap_ring_vfree(struct xenbus_device *dev, struct vm_struct *);
++int xenbus_unmap_ring(struct xenbus_device *dev,
++ grant_handle_t handle, void *vaddr);
++
++
++/**
++ * Allocate an event channel for the given xenbus_device, assigning the newly
++ * created local port to *port. Return 0 on success, or -errno on error. On
++ * error, the device will switch to XenbusStateClosing, and the error will be
++ * saved in the store.
++ */
++int xenbus_alloc_evtchn(struct xenbus_device *dev, int *port);
++
++
++/**
++ * Bind to an existing interdomain event channel in another domain. Returns 0
++ * on success and stores the local port in *port. On error, returns -errno,
++ * switches the device to XenbusStateClosing, and saves the error in XenStore.
++ */
++int xenbus_bind_evtchn(struct xenbus_device *dev, int remote_port, int *port);
++
++
++/**
++ * Free an existing event channel. Returns 0 on success or -errno on error.
++ */
++int xenbus_free_evtchn(struct xenbus_device *dev, int port);
++
++
++/**
++ * Return the state of the driver rooted at the given store path, or
++ * XenbusStateClosed if no state can be read.
++ */
++enum xenbus_state xenbus_read_driver_state(const char *path);
++
++
++/***
++ * Report the given negative errno into the store, along with the given
++ * formatted message.
++ */
++void xenbus_dev_error(struct xenbus_device *dev, int err, const char *fmt,
++ ...);
++
++
++/***
++ * Equivalent to xenbus_dev_error(dev, err, fmt, args), followed by
++ * xenbus_switch_state(dev, NULL, XenbusStateClosing) to schedule an orderly
++ * closedown of this driver and its peer.
++ */
++void xenbus_dev_fatal(struct xenbus_device *dev, int err, const char *fmt,
++ ...);
++
++
++#endif /* _XEN_XENBUS_H */
+diff -urNp linux-2.6/include/xen/xencons.h new/include/xen/xencons.h
+--- linux-2.6/include/xen/xencons.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/xencons.h 2006-05-09 12:35:56.000000000 +0200
+@@ -0,0 +1,14 @@
++#ifndef __ASM_XENCONS_H__
++#define __ASM_XENCONS_H__
++
++void xencons_force_flush(void);
++void xencons_resume(void);
++
++/* Interrupt work hooks. Receive data, or kick data out. */
++void xencons_rx(char *buf, unsigned len, struct pt_regs *regs);
++void xencons_tx(void);
++
++int xencons_ring_init(void);
++int xencons_ring_send(const char *data, unsigned len);
++
++#endif /* __ASM_XENCONS_H__ */
+diff -urNp linux-2.6/include/xen/xen_proc.h new/include/xen/xen_proc.h
+--- linux-2.6/include/xen/xen_proc.h 1970-01-01 01:00:00.000000000 +0100
++++ new/include/xen/xen_proc.h 2006-05-23 18:42:17.000000000 +0200
+@@ -0,0 +1,13 @@
++
++#ifndef __ASM_XEN_PROC_H__
++#define __ASM_XEN_PROC_H__
++
++#include <linux/config.h>
++#include <linux/proc_fs.h>
++
++extern struct proc_dir_entry *create_xen_proc_entry(
++ const char *name, mode_t mode);
++extern void remove_xen_proc_entry(
++ const char *name);
++
++#endif /* __ASM_XEN_PROC_H__ */
+diff -urNp linux-2.6/kernel/fork.c new/kernel/fork.c
+--- linux-2.6/kernel/fork.c 2006-07-03 14:15:20.000000000 +0200
++++ new/kernel/fork.c 2006-06-28 14:32:14.000000000 +0200
+@@ -272,6 +272,9 @@ static inline int dup_mmap(struct mm_str
+ if (retval)
+ goto out;
+ }
++#ifdef arch_dup_mmap
++ arch_dup_mmap(mm, oldmm);
++#endif
+ retval = 0;
+ out:
+ up_write(&mm->mmap_sem);
+diff -urNp linux-2.6/kernel/irq/manage.c new/kernel/irq/manage.c
+--- linux-2.6/kernel/irq/manage.c 2006-07-03 14:15:20.000000000 +0200
++++ new/kernel/irq/manage.c 2006-05-09 12:35:57.000000000 +0200
+@@ -204,14 +204,10 @@ int setup_irq(unsigned int irq, struct i
+ p = &desc->action;
+ if ((old = *p) != NULL) {
+ /* Can't share interrupts unless both agree to */
+- if (!(old->flags & new->flags & SA_SHIRQ))
+- goto mismatch;
+-
+-#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+- /* All handlers must agree on per-cpuness */
+- if ((old->flags & IRQ_PER_CPU) != (new->flags & IRQ_PER_CPU))
+- goto mismatch;
+-#endif
++ if (!(old->flags & new->flags & SA_SHIRQ)) {
++ spin_unlock_irqrestore(&desc->lock,flags);
++ return -EBUSY;
++ }
+
+ /* add new interrupt at end of irq queue */
+ do {
+@@ -222,10 +218,7 @@ int setup_irq(unsigned int irq, struct i
+ }
+
+ *p = new;
+-#if defined(ARCH_HAS_IRQ_PER_CPU) && defined(SA_PERCPU_IRQ)
+- if (new->flags & SA_PERCPU_IRQ)
+- desc->status |= IRQ_PER_CPU;
+-#endif
++
+ if (!shared) {
+ desc->depth = 0;
+ desc->status &= ~(IRQ_DISABLED | IRQ_AUTODETECT |
+@@ -243,14 +236,6 @@ int setup_irq(unsigned int irq, struct i
+ register_handler_proc(irq, new);
+
+ return 0;
+-
+-mismatch:
+- spin_unlock_irqrestore(&desc->lock, flags);
+- if (!(new->flags & SA_PROBEIRQ)) {
+- printk(KERN_ERR "%s: irq handler mismatch\n", __FUNCTION__);
+- dump_stack();
+- }
+- return -EBUSY;
+ }
+
+ /**
+@@ -273,7 +258,6 @@ void free_irq(unsigned int irq, void *de
+ struct irqaction **p;
+ unsigned long flags;
+
+- WARN_ON(in_interrupt());
+ if (irq >= NR_IRQS)
+ return;
+
+diff -urNp linux-2.6/kernel/irq/spurious.c new/kernel/irq/spurious.c
+--- linux-2.6/kernel/irq/spurious.c 2006-07-03 14:15:20.000000000 +0200
++++ new/kernel/irq/spurious.c 2006-06-28 14:32:14.000000000 +0200
+@@ -137,7 +137,8 @@ void note_interrupt(unsigned int irq, ir
+ struct pt_regs *regs)
+ {
+ if (action_ret != IRQ_HANDLED) {
+- desc->irqs_unhandled++;
++ if (!irq_ignore_unhandled(irq))
++ desc->irqs_unhandled++;
+ if (action_ret != IRQ_NONE)
+ report_bad_irq(irq, desc, action_ret);
+ }
+diff -urNp linux-2.6/kernel/Kconfig.preempt new/kernel/Kconfig.preempt
+--- linux-2.6/kernel/Kconfig.preempt 2006-07-03 14:15:20.000000000 +0200
++++ new/kernel/Kconfig.preempt 2006-05-09 12:35:57.000000000 +0200
+@@ -35,6 +35,7 @@ config PREEMPT_VOLUNTARY
+
+ config PREEMPT
+ bool "Preemptible Kernel (Low-Latency Desktop)"
++ depends on !XEN
+ help
+ This option reduces the latency of the kernel by making
+ all kernel code (that is not executing in a critical section)
+diff -urNp linux-2.6/kernel/timer.c new/kernel/timer.c
+--- linux-2.6/kernel/timer.c 2006-07-03 14:15:20.000000000 +0200
++++ new/kernel/timer.c 2006-07-07 16:12:20.000000000 +0200
+@@ -557,6 +557,22 @@ found:
+ if (time_before(expires, jiffies))
+ return jiffies;
+
++ /*
++ * It can happen that other CPUs service timer IRQs and increment
++ * jiffies, but we have not yet got a local timer tick to process
++ * the timer wheels. In that case, the expiry time can be before
++ * jiffies, but since the high-resolution timer here is relative to
++ * jiffies, the default expression when high-resolution timers are
++ * not active,
++ *
++ * time_before(MAX_JIFFY_OFFSET + jiffies, expires)
++ *
++ * would falsely evaluate to true. If that is the case, just
++ * return jiffies so that we can immediately fire the local timer
++ */
++ if (time_before(expires, jiffies))
++ return jiffies;
++
+ if (time_before(hr_expires, expires))
+ return hr_expires;
+
+diff -urNp linux-2.6/lib/Makefile new/lib/Makefile
+--- linux-2.6/lib/Makefile 2006-07-03 14:15:20.000000000 +0200
++++ new/lib/Makefile 2006-05-09 12:35:58.000000000 +0200
+@@ -48,6 +48,9 @@ obj-$(CONFIG_TEXTSEARCH_BM) += ts_bm.o
+ obj-$(CONFIG_TEXTSEARCH_FSM) += ts_fsm.o
+
+ obj-$(CONFIG_SWIOTLB) += swiotlb.o
++ifneq ($(CONFIG_XEN_IA64_DOM0_NON_VP),y)
++swiotlb-$(CONFIG_XEN) := ../arch/i386/kernel/swiotlb.o
++endif
+
+ hostprogs-y := gen_crc32table
+ clean-files := crc32table.h
+diff -urNp linux-2.6/mm/highmem.c new/mm/highmem.c
+--- linux-2.6/mm/highmem.c 2006-07-03 14:15:20.000000000 +0200
++++ new/mm/highmem.c 2006-05-09 12:35:59.000000000 +0200
+@@ -142,6 +142,17 @@ start:
+ return vaddr;
+ }
+
++#ifdef CONFIG_XEN
++void kmap_flush_unused(void)
++{
++ spin_lock(&kmap_lock);
++ flush_all_zero_pkmaps();
++ spin_unlock(&kmap_lock);
++}
++
++EXPORT_SYMBOL(kmap_flush_unused);
++#endif
++
+ void fastcall *kmap_high(struct page *page)
+ {
+ unsigned long vaddr;
+diff -urNp linux-2.6/mm/Kconfig new/mm/Kconfig
+--- linux-2.6/mm/Kconfig 2006-07-03 14:15:20.000000000 +0200
++++ new/mm/Kconfig 2006-05-09 12:40:16.000000000 +0200
+@@ -126,11 +126,14 @@ comment "Memory hotplug is currently inc
+ # Default to 4 for wider testing, though 8 might be more appropriate.
+ # ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock.
+ # PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes.
++# XEN on x86 architecture uses the mapping field on pagetable pages to store a
++# pointer to the destructor. This conflicts with pte_lock_deinit().
+ #
+ config SPLIT_PTLOCK_CPUS
+ int
+ default "4096" if ARM && !CPU_CACHE_VIPT
+ default "4096" if PARISC && !PA20
++ default "4096" if X86_XEN || X86_64_XEN
+ default "4"
+
+ #
+diff -urNp linux-2.6/mm/memory.c new/mm/memory.c
+--- linux-2.6/mm/memory.c 2006-07-03 14:15:20.000000000 +0200
++++ new/mm/memory.c 2006-06-28 14:32:14.000000000 +0200
+@@ -402,7 +402,8 @@ struct page *vm_normal_page(struct vm_ar
+ * and that the resulting page looks ok.
+ */
+ if (unlikely(!pfn_valid(pfn))) {
+- print_bad_pte(vma, pte, addr);
++ if (!(vma->vm_flags & VM_RESERVED))
++ print_bad_pte(vma, pte, addr);
+ return NULL;
+ }
+
+@@ -1017,6 +1018,26 @@ int get_user_pages(struct task_struct *t
+ continue;
+ }
+
++#ifdef CONFIG_XEN
++ if (vma && (vma->vm_flags & VM_FOREIGN)) {
++ struct page **map = vma->vm_private_data;
++ int offset = (start - vma->vm_start) >> PAGE_SHIFT;
++ if (map[offset] != NULL) {
++ if (pages) {
++ struct page *page = map[offset];
++
++ pages[i] = page;
++ get_page(page);
++ }
++ if (vmas)
++ vmas[i] = vma;
++ i++;
++ start += PAGE_SIZE;
++ len--;
++ continue;
++ }
++ }
++#endif
+ if (!vma || (vma->vm_flags & (VM_IO | VM_PFNMAP))
+ || !(vm_flags & vma->vm_flags))
+ return i ? : -EFAULT;
+@@ -1356,6 +1377,102 @@ int remap_pfn_range(struct vm_area_struc
+ }
+ EXPORT_SYMBOL(remap_pfn_range);
+
++#ifdef CONFIG_XEN
++static inline int apply_to_pte_range(struct mm_struct *mm, pmd_t *pmd,
++ unsigned long addr, unsigned long end,
++ pte_fn_t fn, void *data)
++{
++ pte_t *pte;
++ int err;
++ struct page *pmd_page;
++ spinlock_t *ptl;
++
++ pte = (mm == &init_mm) ?
++ pte_alloc_kernel(pmd, addr) :
++ pte_alloc_map_lock(mm, pmd, addr, &ptl);
++ if (!pte)
++ return -ENOMEM;
++
++ BUG_ON(pmd_huge(*pmd));
++
++ pmd_page = pmd_page(*pmd);
++
++ do {
++ err = fn(pte, pmd_page, addr, data);
++ if (err)
++ break;
++ } while (pte++, addr += PAGE_SIZE, addr != end);
++
++ if (mm != &init_mm)
++ pte_unmap_unlock(pte-1, ptl);
++ return err;
++}
++
++static inline int apply_to_pmd_range(struct mm_struct *mm, pud_t *pud,
++ unsigned long addr, unsigned long end,
++ pte_fn_t fn, void *data)
++{
++ pmd_t *pmd;
++ unsigned long next;
++ int err;
++
++ pmd = pmd_alloc(mm, pud, addr);
++ if (!pmd)
++ return -ENOMEM;
++ do {
++ next = pmd_addr_end(addr, end);
++ err = apply_to_pte_range(mm, pmd, addr, next, fn, data);
++ if (err)
++ break;
++ } while (pmd++, addr = next, addr != end);
++ return err;
++}
++
++static inline int apply_to_pud_range(struct mm_struct *mm, pgd_t *pgd,
++ unsigned long addr, unsigned long end,
++ pte_fn_t fn, void *data)
++{
++ pud_t *pud;
++ unsigned long next;
++ int err;
++
++ pud = pud_alloc(mm, pgd, addr);
++ if (!pud)
++ return -ENOMEM;
++ do {
++ next = pud_addr_end(addr, end);
++ err = apply_to_pmd_range(mm, pud, addr, next, fn, data);
++ if (err)
++ break;
++ } while (pud++, addr = next, addr != end);
++ return err;
++}
++
++/*
++ * Scan a region of virtual memory, filling in page tables as necessary
++ * and calling a provided function on each leaf page table.
++ */
++int apply_to_page_range(struct mm_struct *mm, unsigned long addr,
++ unsigned long size, pte_fn_t fn, void *data)
++{
++ pgd_t *pgd;
++ unsigned long next;
++ unsigned long end = addr + size;
++ int err;
++
++ BUG_ON(addr >= end);
++ pgd = pgd_offset(mm, addr);
++ do {
++ next = pgd_addr_end(addr, end);
++ err = apply_to_pud_range(mm, pgd, addr, next, fn, data);
++ if (err)
++ break;
++ } while (pgd++, addr = next, addr != end);
++ return err;
++}
++EXPORT_SYMBOL_GPL(apply_to_page_range);
++#endif
++
+ /*
+ * handle_pte_fault chooses page fault handler according to an entry
+ * which was read non-atomically. Before making any commitment, on
+diff -urNp linux-2.6/mm/mmap.c new/mm/mmap.c
+--- linux-2.6/mm/mmap.c 2006-07-03 14:15:20.000000000 +0200
++++ new/mm/mmap.c 2006-05-09 12:35:59.000000000 +0200
+@@ -1942,6 +1942,10 @@ void exit_mmap(struct mm_struct *mm)
+ unsigned long nr_accounted = 0;
+ unsigned long end;
+
++#ifdef arch_exit_mmap
++ arch_exit_mmap(mm);
++#endif
++
+ lru_add_drain();
+ flush_cache_mm(mm);
+ tlb = tlb_gather_mmu(mm, 1);
+diff -urNp linux-2.6/mm/page_alloc.c new/mm/page_alloc.c
+--- linux-2.6/mm/page_alloc.c 2006-07-03 14:15:20.000000000 +0200
++++ new/mm/page_alloc.c 2006-05-23 18:37:12.000000000 +0200
+@@ -438,7 +438,8 @@ static void __free_pages_ok(struct page
+ int i;
+ int reserved = 0;
+
+- arch_free_page(page, order);
++ if (arch_free_page(page, order))
++ return;
+ if (!PageHighMem(page))
+ mutex_debug_check_no_locks_freed(page_address(page),
+ PAGE_SIZE<<order);
+@@ -733,7 +734,8 @@ static void fastcall free_hot_cold_page(
+ struct per_cpu_pages *pcp;
+ unsigned long flags;
+
+- arch_free_page(page, 0);
++ if (arch_free_page(page, 0))
++ return;
+
+ if (PageAnon(page))
+ page->mapping = NULL;
+diff -urNp linux-2.6/net/atm/clip.c new/net/atm/clip.c
+--- linux-2.6/net/atm/clip.c 2006-07-03 14:15:20.000000000 +0200
++++ new/net/atm/clip.c 2006-07-07 16:13:30.000000000 +0200
+@@ -98,7 +98,7 @@ static void unlink_clip_vcc(struct clip_
+ printk(KERN_CRIT "!clip_vcc->entry (clip_vcc %p)\n", clip_vcc);
+ return;
+ }
+- spin_lock_bh(&entry->neigh->dev->xmit_lock); /* block clip_start_xmit() */
++ netif_tx_lock_bh(entry->neigh->dev); /* block clip_start_xmit() */
+ entry->neigh->used = jiffies;
+ for (walk = &entry->vccs; *walk; walk = &(*walk)->next)
+ if (*walk == clip_vcc) {
+@@ -120,9 +120,9 @@ static void unlink_clip_vcc(struct clip_
+ goto out;
+ }
+ printk(KERN_CRIT "ATMARP: unlink_clip_vcc failed (entry %p, vcc "
+- "0x%p)\n", entry, clip_vcc);
+- out:
+- spin_unlock_bh(&entry->neigh->dev->xmit_lock);
++ "0x%p)\n",entry,clip_vcc);
++out:
++ netif_tx_unlock_bh(entry->neigh->dev);
+ }
+
+ /* The neighbour entry n->lock is held. */
+diff -urNp linux-2.6/net/bridge/br_device.c new/net/bridge/br_device.c
+--- linux-2.6/net/bridge/br_device.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/bridge/br_device.c 2006-07-07 16:13:36.000000000 +0200
+@@ -145,9 +145,9 @@ static int br_set_tx_csum(struct net_dev
+ struct net_bridge *br = netdev_priv(dev);
+
+ if (data)
+- br->feature_mask |= NETIF_F_IP_CSUM;
++ br->feature_mask |= NETIF_F_NO_CSUM;
+ else
+- br->feature_mask &= ~NETIF_F_IP_CSUM;
++ br->feature_mask &= ~NETIF_F_ALL_CSUM;
+
+ br_features_recompute(br);
+ return 0;
+@@ -184,6 +184,6 @@ void br_dev_setup(struct net_device *dev
+ dev->set_mac_address = br_set_mac_address;
+ dev->priv_flags = IFF_EBRIDGE;
+
+- dev->features = NETIF_F_SG | NETIF_F_FRAGLIST
+- | NETIF_F_HIGHDMA | NETIF_F_TSO | NETIF_F_IP_CSUM;
++ dev->features = NETIF_F_SG | NETIF_F_FRAGLIST | NETIF_F_HIGHDMA |
++ NETIF_F_TSO | NETIF_F_NO_CSUM | NETIF_F_GSO_ROBUST;
+ }
+diff -urNp linux-2.6/net/bridge/br_forward.c new/net/bridge/br_forward.c
+--- linux-2.6/net/bridge/br_forward.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/bridge/br_forward.c 2006-07-07 16:14:07.000000000 +0200
+@@ -38,7 +38,7 @@ static inline unsigned packet_length(con
+ int br_dev_queue_push_xmit(struct sk_buff *skb)
+ {
+ /* drop mtu oversized packets except tso */
+- if (packet_length(skb) > skb->dev->mtu && !skb_shinfo(skb)->tso_size)
++ if (skb->len > skb->dev->mtu && !skb_shinfo(skb)->gso_size)
+ kfree_skb(skb);
+ else {
+ #ifdef CONFIG_BRIDGE_NETFILTER
+diff -urNp linux-2.6/net/bridge/br_if.c new/net/bridge/br_if.c
+--- linux-2.6/net/bridge/br_if.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/bridge/br_if.c 2006-07-07 16:14:17.000000000 +0200
+@@ -372,17 +372,28 @@ void br_features_recompute(struct net_br
+ struct net_bridge_port *p;
+ unsigned long features, checksum;
+
+- features = br->feature_mask &~ NETIF_F_IP_CSUM;
+- checksum = br->feature_mask & NETIF_F_IP_CSUM;
++ checksum = br->feature_mask & NETIF_F_ALL_CSUM ? NETIF_F_NO_CSUM : 0;
++ features = br->feature_mask & ~NETIF_F_ALL_CSUM;
+
+ list_for_each_entry(p, &br->port_list, list) {
+- if (!(p->dev->features
+- & (NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM)))
++ unsigned long feature = p->dev->features;
++
++ if (checksum & NETIF_F_NO_CSUM && !(feature & NETIF_F_NO_CSUM))
++ checksum ^= NETIF_F_NO_CSUM | NETIF_F_HW_CSUM;
++ if (checksum & NETIF_F_HW_CSUM && !(feature & NETIF_F_HW_CSUM))
++ checksum ^= NETIF_F_HW_CSUM | NETIF_F_IP_CSUM;
++ if (!(feature & NETIF_F_IP_CSUM))
+ checksum = 0;
+- features &= p->dev->features;
++
++ if (feature & NETIF_F_GSO)
++ feature |= NETIF_F_TSO;
++ feature |= NETIF_F_GSO;
++
++ features &= feature;
+ }
+
+- br->dev->features = features | checksum | NETIF_F_LLTX;
++ br->dev->features = features | checksum | NETIF_F_LLTX |
++ NETIF_F_GSO_ROBUST;
+ }
+
+ /* called with RTNL */
+diff -urNp linux-2.6/net/bridge/br_netfilter.c new/net/bridge/br_netfilter.c
+--- linux-2.6/net/bridge/br_netfilter.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/bridge/br_netfilter.c 2006-07-07 16:17:11.000000000 +0200
+@@ -769,7 +769,7 @@ static int br_nf_dev_queue_xmit(struct s
+ {
+ if (skb->protocol == htons(ETH_P_IP) &&
+ skb->len > skb->dev->mtu &&
+- !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size))
++ !skb_shinfo(skb)->gso_size)
+ return ip_fragment(skb, br_dev_queue_push_xmit);
+ else
+ return br_dev_queue_push_xmit(skb);
+@@ -877,8 +877,9 @@ static unsigned int ip_sabotage_out(unsi
+ struct sk_buff *skb = *pskb;
+
+ if ((out->hard_start_xmit == br_dev_xmit &&
+- okfn != br_nf_forward_finish &&
+- okfn != br_nf_local_out_finish && okfn != br_nf_dev_queue_xmit)
++ okfn != br_nf_forward_finish &&
++ okfn != br_nf_local_out_finish &&
++ okfn != br_nf_dev_queue_xmit)
+ #if defined(CONFIG_VLAN_8021Q) || defined(CONFIG_VLAN_8021Q_MODULE)
+ || ((out->priv_flags & IFF_802_1Q_VLAN) &&
+ VLAN_DEV_INFO(out)->real_dev->hard_start_xmit == br_dev_xmit)
+diff -urNp linux-2.6/net/core/dev.c new/net/core/dev.c
+--- linux-2.6/net/core/dev.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/dev.c 2006-07-07 16:17:58.000000000 +0200
+@@ -115,6 +115,13 @@
+ #include <asm/current.h>
+ #include <linux/audit.h>
+ #include <linux/vs_network.h>
++#include <linux/err.h>
++
++#ifdef CONFIG_XEN
++#include <net/ip.h>
++#include <linux/tcp.h>
++#include <linux/udp.h>
++#endif
+
+ /*
+ * The list of packet types we will receive (as opposed to discard)
+@@ -1041,7 +1048,7 @@ static inline void net_timestamp(struct
+ * taps currently in use.
+ */
+
+-void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
++static void dev_queue_xmit_nit(struct sk_buff *skb, struct net_device *dev)
+ {
+ struct packet_type *ptype;
+
+@@ -1179,6 +1186,45 @@ out:
+ return ret;
+ }
+
++/**
++ * skb_gso_segment - Perform segmentation on skb.
++ * @skb: buffer to segment
++ * @features: features for the output path (see dev->features)
++ *
++ * This function segments the given skb and returns a list of segments.
++ *
++ * It may return NULL if the skb requires no segmentation. This is
++ * only possible when GSO is used for verifying header integrity.
++ */
++struct sk_buff *skb_gso_segment(struct sk_buff *skb, int features)
++{
++ struct sk_buff *segs = ERR_PTR(-EPROTONOSUPPORT);
++ struct packet_type *ptype;
++ int type = skb->protocol;
++
++ BUG_ON(skb_shinfo(skb)->frag_list);
++ BUG_ON(skb->ip_summed != CHECKSUM_HW);
++
++ skb->mac.raw = skb->data;
++ skb->mac_len = skb->nh.raw - skb->data;
++ __skb_pull(skb, skb->mac_len);
++
++ rcu_read_lock();
++ list_for_each_entry_rcu(ptype, &ptype_base[ntohs(type) & 15], list) {
++ if (ptype->type == type && !ptype->dev && ptype->gso_segment) {
++ segs = ptype->gso_segment(skb, features);
++ break;
++ }
++ }
++ rcu_read_unlock();
++
++ __skb_push(skb, skb->data - skb->mac.raw);
++
++ return segs;
++}
++
++EXPORT_SYMBOL(skb_gso_segment);
++
+ /* Take action when hardware reception checksum errors are detected. */
+ #ifdef CONFIG_BUG
+ void netdev_rx_csum_fault(struct net_device *dev)
+@@ -1215,79 +1261,148 @@ static inline int illegal_highdma(struct
+ #define illegal_highdma(dev, skb) (0)
+ #endif
+
+-/* Keep head the same: replace data */
+-int __skb_linearize(struct sk_buff *skb, gfp_t gfp_mask)
++struct dev_gso_cb {
++ void (*destructor)(struct sk_buff *skb);
++};
++
++#define DEV_GSO_CB(skb) ((struct dev_gso_cb *)(skb)->cb)
++
++static void dev_gso_skb_destructor(struct sk_buff *skb)
+ {
+- unsigned int size;
+- u8 *data;
+- long offset;
+- struct skb_shared_info *ninfo;
+- int headerlen = skb->data - skb->head;
+- int expand = (skb->tail + skb->data_len) - skb->end;
+-
+- if (skb_shared(skb))
+- BUG();
+-
+- if (expand <= 0)
+- expand = 0;
+-
+- size = skb->end - skb->head + expand;
+- size = SKB_DATA_ALIGN(size);
+- data = kmalloc(size + sizeof(struct skb_shared_info), gfp_mask);
+- if (!data)
+- return -ENOMEM;
+-
+- /* Copy entire thing */
+- if (skb_copy_bits(skb, -headerlen, data, headerlen + skb->len))
+- BUG();
+-
+- /* Set up shinfo */
+- ninfo = (struct skb_shared_info*)(data + size);
+- atomic_set(&ninfo->dataref, 1);
+- ninfo->tso_size = skb_shinfo(skb)->tso_size;
+- ninfo->tso_segs = skb_shinfo(skb)->tso_segs;
+- ninfo->ufo_size = skb_shinfo(skb)->ufo_size;
+- ninfo->nr_frags = 0;
+- ninfo->frag_list = NULL;
+-
+- /* Offset between the two in bytes */
+- offset = data - skb->head;
+-
+- /* Free old data. */
+- skb_release_data(skb);
+-
+- skb->head = data;
+- skb->end = data + size;
+-
+- /* Set up new pointers */
+- skb->h.raw += offset;
+- skb->nh.raw += offset;
+- skb->mac.raw += offset;
+- skb->tail += offset;
+- skb->data += offset;
++ struct dev_gso_cb *cb;
++
++ do {
++ struct sk_buff *nskb = skb->next;
+
+- /* We are no longer a clone, even if we were. */
+- skb->cloned = 0;
++ skb->next = nskb->next;
++ nskb->next = NULL;
++ kfree_skb(nskb);
++ } while (skb->next);
++
++ cb = DEV_GSO_CB(skb);
++ if (cb->destructor)
++ cb->destructor(skb);
++}
++
++/**
++ * dev_gso_segment - Perform emulated hardware segmentation on skb.
++ * @skb: buffer to segment
++ *
++ * This function segments the given skb and stores the list of segments
++ * in skb->next.
++ */
++static int dev_gso_segment(struct sk_buff *skb)
++{
++ struct net_device *dev = skb->dev;
++ struct sk_buff *segs;
++ int features = dev->features & ~(illegal_highdma(dev, skb) ?
++ NETIF_F_SG : 0);
++
++ segs = skb_gso_segment(skb, features);
++
++ /* Verifying header integrity only. */
++ if (!segs)
++ return 0;
++
++ if (unlikely(IS_ERR(segs)))
++ return PTR_ERR(segs);
++
++ skb->next = segs;
++ DEV_GSO_CB(skb)->destructor = skb->destructor;
++ skb->destructor = dev_gso_skb_destructor;
+
+- skb->tail += skb->data_len;
+- skb->data_len = 0;
++ return 0;
++}
++
++int dev_hard_start_xmit(struct sk_buff *skb, struct net_device *dev)
++{
++ if (likely(!skb->next)) {
++ if (netdev_nit)
++ dev_queue_xmit_nit(skb, dev);
++
++ if (netif_needs_gso(dev, skb)) {
++ if (unlikely(dev_gso_segment(skb)))
++ goto out_kfree_skb;
++ if (skb->next)
++ goto gso;
++ }
++
++ return dev->hard_start_xmit(skb, dev);
++ }
++
++gso:
++ do {
++ struct sk_buff *nskb = skb->next;
++ int rc;
++
++ skb->next = nskb->next;
++ nskb->next = NULL;
++ rc = dev->hard_start_xmit(nskb, dev);
++ if (unlikely(rc)) {
++ nskb->next = skb->next;
++ skb->next = nskb;
++ return rc;
++ }
++ if (unlikely(netif_queue_stopped(dev) && skb->next))
++ return NETDEV_TX_BUSY;
++ } while (skb->next);
++
++ skb->destructor = DEV_GSO_CB(skb)->destructor;
++
++out_kfree_skb:
++ kfree_skb(skb);
+ return 0;
+ }
+
+ #define HARD_TX_LOCK(dev, cpu) { \
+ if ((dev->features & NETIF_F_LLTX) == 0) { \
+- spin_lock(&dev->xmit_lock); \
+- dev->xmit_lock_owner = cpu; \
++ netif_tx_lock(dev); \
+ } \
+ }
+
+ #define HARD_TX_UNLOCK(dev) { \
+ if ((dev->features & NETIF_F_LLTX) == 0) { \
+- dev->xmit_lock_owner = -1; \
+- spin_unlock(&dev->xmit_lock); \
++ netif_tx_unlock(dev); \
+ } \
+ }
+
++#ifdef CONFIG_XEN
++inline int skb_checksum_setup(struct sk_buff *skb)
++{
++ if (skb->proto_csum_blank) {
++ if (skb->protocol != htons(ETH_P_IP))
++ goto out;
++ skb->h.raw = (unsigned char *)skb->nh.iph + 4*skb->nh.iph->ihl;
++ if (skb->h.raw >= skb->tail)
++ goto out;
++ switch (skb->nh.iph->protocol) {
++ case IPPROTO_TCP:
++ skb->csum = offsetof(struct tcphdr, check);
++ break;
++ case IPPROTO_UDP:
++ skb->csum = offsetof(struct udphdr, check);
++ break;
++ default:
++ if (net_ratelimit())
++ printk(KERN_ERR "Attempting to checksum a non-"
++ "TCP/UDP packet, dropping a protocol"
++ " %d packet", skb->nh.iph->protocol);
++ goto out;
++ }
++ if ((skb->h.raw + skb->csum + 2) > skb->tail)
++ goto out;
++ skb->ip_summed = CHECKSUM_HW;
++ skb->proto_csum_blank = 0;
++ }
++ return 0;
++out:
++ return -EPROTO;
++}
++#else
++inline int skb_checksum_setup(struct sk_buff *skb) { return 0; }
++#endif
++
++
+ /**
+ * dev_queue_xmit - transmit a buffer
+ * @skb: buffer to transmit
+@@ -1319,9 +1435,19 @@ int dev_queue_xmit(struct sk_buff *skb)
+ struct Qdisc *q;
+ int rc = -ENOMEM;
+
++ /* If a checksum-deferred packet is forwarded to a device that needs a
++ * checksum, correct the pointers and force checksumming.
++ */
++ if (skb_checksum_setup(skb))
++ goto out_kfree_skb;
++
++ /* GSO will handle the following emulations directly. */
++ if (netif_needs_gso(dev, skb))
++ goto gso;
++
+ if (skb_shinfo(skb)->frag_list &&
+ !(dev->features & NETIF_F_FRAGLIST) &&
+- __skb_linearize(skb, GFP_ATOMIC))
++ __skb_linearize(skb))
+ goto out_kfree_skb;
+
+ /* Fragmented skb is linearized if device does not support SG,
+@@ -1330,25 +1456,26 @@ int dev_queue_xmit(struct sk_buff *skb)
+ */
+ if (skb_shinfo(skb)->nr_frags &&
+ (!(dev->features & NETIF_F_SG) || illegal_highdma(dev, skb)) &&
+- __skb_linearize(skb, GFP_ATOMIC))
++ __skb_linearize(skb))
+ goto out_kfree_skb;
+
+ /* If packet is not checksummed and device does not support
+ * checksumming for this protocol, complete checksumming here.
+ */
+ if (skb->ip_summed == CHECKSUM_HW &&
+- (!(dev->features & (NETIF_F_HW_CSUM | NETIF_F_NO_CSUM)) &&
++ (!(dev->features & NETIF_F_GEN_CSUM) &&
+ (!(dev->features & NETIF_F_IP_CSUM) ||
+ skb->protocol != htons(ETH_P_IP))))
+ if (skb_checksum_help(skb, 0))
+ goto out_kfree_skb;
+
++gso:
+ spin_lock_prefetch(&dev->queue_lock);
+
+ /* Disable soft irqs for various locks below. Also
+ * stops preemption for RCU.
+ */
+- local_bh_disable();
++ rcu_read_lock_bh();
+
+ /* Updates of qdisc are serialized by queue_lock.
+ * The struct Qdisc which is pointed to by qdisc is now a
+@@ -1382,8 +1509,8 @@ int dev_queue_xmit(struct sk_buff *skb)
+ /* The device has no queue. Common case for software devices:
+ loopback, all the sorts of tunnels...
+
+- Really, it is unlikely that xmit_lock protection is necessary here.
+- (f.e. loopback and IP tunnels are clean ignoring statistics
++ Really, it is unlikely that netif_tx_lock protection is necessary
++ here. (f.e. loopback and IP tunnels are clean ignoring statistics
+ counters.)
+ However, it is possible, that they rely on protection
+ made by us here.
+@@ -1399,11 +1526,8 @@ int dev_queue_xmit(struct sk_buff *skb)
+ HARD_TX_LOCK(dev, cpu);
+
+ if (!netif_queue_stopped(dev)) {
+- if (netdev_nit)
+- dev_queue_xmit_nit(skb, dev);
+-
+ rc = 0;
+- if (!dev->hard_start_xmit(skb, dev)) {
++ if (!dev_hard_start_xmit(skb, dev)) {
+ HARD_TX_UNLOCK(dev);
+ goto out;
+ }
+@@ -1422,13 +1546,13 @@ int dev_queue_xmit(struct sk_buff *skb)
+ }
+
+ rc = -ENETDOWN;
+- local_bh_enable();
++ rcu_read_unlock_bh();
+
+ out_kfree_skb:
+ kfree_skb(skb);
+ return rc;
+ out:
+- local_bh_enable();
++ rcu_read_unlock_bh();
+ return rc;
+ }
+
+@@ -1706,6 +1830,19 @@ int netif_receive_skb(struct sk_buff *sk
+ }
+ #endif
+
++#ifdef CONFIG_XEN
++ switch (skb->ip_summed) {
++ case CHECKSUM_UNNECESSARY:
++ skb->proto_data_valid = 1;
++ break;
++ case CHECKSUM_HW:
++ /* XXX Implement me. */
++ default:
++ skb->proto_data_valid = 0;
++ break;
++ }
++#endif
++
+ list_for_each_entry_rcu(ptype, &ptype_all, list) {
+ if (!ptype->dev || ptype->dev == skb->dev) {
+ if (pt_prev)
+@@ -2785,7 +2922,7 @@ int register_netdevice(struct net_device
+ BUG_ON(dev->reg_state != NETREG_UNINITIALIZED);
+
+ spin_lock_init(&dev->queue_lock);
+- spin_lock_init(&dev->xmit_lock);
++ spin_lock_init(&dev->_xmit_lock);
+ dev->xmit_lock_owner = -1;
+ #ifdef CONFIG_NET_CLS_ACT
+ spin_lock_init(&dev->ingress_lock);
+@@ -2829,9 +2966,7 @@ int register_netdevice(struct net_device
+
+ /* Fix illegal SG+CSUM combinations. */
+ if ((dev->features & NETIF_F_SG) &&
+- !(dev->features & (NETIF_F_IP_CSUM |
+- NETIF_F_NO_CSUM |
+- NETIF_F_HW_CSUM))) {
++ !(dev->features & NETIF_F_ALL_CSUM)) {
+ printk("%s: Dropping NETIF_F_SG since no checksum feature.\n",
+ dev->name);
+ dev->features &= ~NETIF_F_SG;
+@@ -3371,7 +3506,6 @@ subsys_initcall(net_dev_init);
+ EXPORT_SYMBOL(__dev_get_by_index);
+ EXPORT_SYMBOL(__dev_get_by_name);
+ EXPORT_SYMBOL(__dev_remove_pack);
+-EXPORT_SYMBOL(__skb_linearize);
+ EXPORT_SYMBOL(dev_valid_name);
+ EXPORT_SYMBOL(dev_add_pack);
+ EXPORT_SYMBOL(dev_alloc_name);
+@@ -3403,6 +3537,7 @@ EXPORT_SYMBOL(unregister_netdevice_notif
+ EXPORT_SYMBOL(net_enable_timestamp);
+ EXPORT_SYMBOL(net_disable_timestamp);
+ EXPORT_SYMBOL(dev_get_flags);
++EXPORT_SYMBOL(skb_checksum_setup);
+
+ #if defined(CONFIG_BRIDGE) || defined(CONFIG_BRIDGE_MODULE)
+ EXPORT_SYMBOL(br_handle_frame_hook);
+diff -urNp linux-2.6/net/core/dev_mcast.c new/net/core/dev_mcast.c
+--- linux-2.6/net/core/dev_mcast.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/dev_mcast.c 2006-07-07 15:10:03.000000000 +0200
+@@ -62,7 +62,7 @@
+ * Device mc lists are changed by bh at least if IPv6 is enabled,
+ * so that it must be bh protected.
+ *
+- * We block accesses to device mc filters with dev->xmit_lock.
++ * We block accesses to device mc filters with netif_tx_lock.
+ */
+
+ /*
+@@ -93,9 +93,9 @@ static void __dev_mc_upload(struct net_d
+
+ void dev_mc_upload(struct net_device *dev)
+ {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ __dev_mc_upload(dev);
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ }
+
+ /*
+@@ -107,7 +107,7 @@ int dev_mc_delete(struct net_device *dev
+ int err = 0;
+ struct dev_mc_list *dmi, **dmip;
+
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+
+ for (dmip = &dev->mc_list; (dmi = *dmip) != NULL; dmip = &dmi->next) {
+ /*
+@@ -139,13 +139,13 @@ int dev_mc_delete(struct net_device *dev
+ */
+ __dev_mc_upload(dev);
+
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ return 0;
+ }
+ }
+ err = -ENOENT;
+ done:
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ return err;
+ }
+
+@@ -160,7 +160,7 @@ int dev_mc_add(struct net_device *dev, v
+
+ dmi1 = kmalloc(sizeof(*dmi), GFP_ATOMIC);
+
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ for (dmi = dev->mc_list; dmi != NULL; dmi = dmi->next) {
+ if (memcmp(dmi->dmi_addr, addr, dmi->dmi_addrlen) == 0 &&
+ dmi->dmi_addrlen == alen) {
+@@ -176,7 +176,7 @@ int dev_mc_add(struct net_device *dev, v
+ }
+
+ if ((dmi = dmi1) == NULL) {
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ return -ENOMEM;
+ }
+ memcpy(dmi->dmi_addr, addr, alen);
+@@ -189,11 +189,11 @@ int dev_mc_add(struct net_device *dev, v
+
+ __dev_mc_upload(dev);
+
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ return 0;
+
+ done:
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ kfree(dmi1);
+ return err;
+ }
+@@ -204,7 +204,7 @@ done:
+
+ void dev_mc_discard(struct net_device *dev)
+ {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+
+ while (dev->mc_list != NULL) {
+ struct dev_mc_list *tmp = dev->mc_list;
+@@ -215,7 +215,7 @@ void dev_mc_discard(struct net_device *d
+ }
+ dev->mc_count = 0;
+
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ }
+
+ #ifdef CONFIG_PROC_FS
+@@ -250,7 +250,7 @@ static int dev_mc_seq_show(struct seq_fi
+ struct dev_mc_list *m;
+ struct net_device *dev = v;
+
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ for (m = dev->mc_list; m; m = m->next) {
+ int i;
+
+@@ -262,7 +262,7 @@ static int dev_mc_seq_show(struct seq_fi
+
+ seq_putc(seq, '\n');
+ }
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ return 0;
+ }
+
+diff -urNp linux-2.6/net/core/ethtool.c new/net/core/ethtool.c
+--- linux-2.6/net/core/ethtool.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/ethtool.c 2006-07-07 15:10:03.000000000 +0200
+@@ -30,7 +30,7 @@ u32 ethtool_op_get_link(struct net_devic
+
+ u32 ethtool_op_get_tx_csum(struct net_device *dev)
+ {
+- return (dev->features & (NETIF_F_IP_CSUM | NETIF_F_HW_CSUM)) != 0;
++ return (dev->features & NETIF_F_ALL_CSUM) != 0;
+ }
+
+ int ethtool_op_set_tx_csum(struct net_device *dev, u32 data)
+@@ -551,9 +551,7 @@ static int ethtool_set_sg(struct net_dev
+ return -EFAULT;
+
+ if (edata.data &&
+- !(dev->features & (NETIF_F_IP_CSUM |
+- NETIF_F_NO_CSUM |
+- NETIF_F_HW_CSUM)))
++ !(dev->features & NETIF_F_ALL_CSUM))
+ return -EINVAL;
+
+ return __ethtool_set_sg(dev, edata.data);
+@@ -591,7 +589,7 @@ static int ethtool_set_tso(struct net_de
+
+ static int ethtool_get_ufo(struct net_device *dev, char __user *useraddr)
+ {
+- struct ethtool_value edata = { ETHTOOL_GTSO };
++ struct ethtool_value edata = { ETHTOOL_GUFO };
+
+ if (!dev->ethtool_ops->get_ufo)
+ return -EOPNOTSUPP;
+@@ -615,6 +614,29 @@ static int ethtool_set_ufo(struct net_de
+ return dev->ethtool_ops->set_ufo(dev, edata.data);
+ }
+
++static int ethtool_get_gso(struct net_device *dev, char __user *useraddr)
++{
++ struct ethtool_value edata = { ETHTOOL_GGSO };
++
++ edata.data = dev->features & NETIF_F_GSO;
++ if (copy_to_user(useraddr, &edata, sizeof(edata)))
++ return -EFAULT;
++ return 0;
++}
++
++static int ethtool_set_gso(struct net_device *dev, char __user *useraddr)
++{
++ struct ethtool_value edata;
++
++ if (copy_from_user(&edata, useraddr, sizeof(edata)))
++ return -EFAULT;
++ if (edata.data)
++ dev->features |= NETIF_F_GSO;
++ else
++ dev->features &= ~NETIF_F_GSO;
++ return 0;
++}
++
+ static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
+ {
+ struct ethtool_test test;
+@@ -906,6 +928,12 @@ int dev_ethtool(struct ifreq *ifr)
+ case ETHTOOL_SUFO:
+ rc = ethtool_set_ufo(dev, useraddr);
+ break;
++ case ETHTOOL_GGSO:
++ rc = ethtool_get_gso(dev, useraddr);
++ break;
++ case ETHTOOL_SGSO:
++ rc = ethtool_set_gso(dev, useraddr);
++ break;
+ default:
+ rc = -EOPNOTSUPP;
+ }
+diff -urNp linux-2.6/net/core/netpoll.c new/net/core/netpoll.c
+--- linux-2.6/net/core/netpoll.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/netpoll.c 2006-07-07 16:18:04.000000000 +0200
+@@ -273,24 +273,21 @@ static void netpoll_send_skb(struct netp
+
+ do {
+ npinfo->tries--;
+- spin_lock(&np->dev->xmit_lock);
+- np->dev->xmit_lock_owner = smp_processor_id();
++ netif_tx_lock(np->dev);
+
+ /*
+ * network drivers do not expect to be called if the queue is
+ * stopped.
+ */
+ if (netif_queue_stopped(np->dev)) {
+- np->dev->xmit_lock_owner = -1;
+- spin_unlock(&np->dev->xmit_lock);
++ netif_tx_unlock(np->dev);
+ netpoll_poll(np);
+ udelay(50);
+ continue;
+ }
+
+ status = np->dev->hard_start_xmit(skb, np->dev);
+- np->dev->xmit_lock_owner = -1;
+- spin_unlock(&np->dev->xmit_lock);
++ netif_tx_unlock(np->dev);
+
+ /* success */
+ if(!status) {
+diff -urNp linux-2.6/net/core/pktgen.c new/net/core/pktgen.c
+--- linux-2.6/net/core/pktgen.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/pktgen.c 2006-07-07 16:18:55.000000000 +0200
+@@ -2896,8 +2896,8 @@ static __inline__ void pktgen_xmit(struc
+ pkt_dev->clone_count = 0; /* reset counter */
+ }
+ }
+-
+- spin_lock_bh(&odev->xmit_lock);
++
++ netif_tx_lock_bh(odev);
+ if (!netif_queue_stopped(odev)) {
+
+ atomic_inc(&(pkt_dev->skb->users));
+@@ -2942,8 +2942,8 @@ static __inline__ void pktgen_xmit(struc
+ pkt_dev->next_tx_ns = 0;
+ }
+
+- spin_unlock_bh(&odev->xmit_lock);
+-
++ netif_tx_unlock_bh(odev);
++
+ /* If pkt_dev->count is zero, then run forever */
+ if ((pkt_dev->count != 0) && (pkt_dev->sofar >= pkt_dev->count)) {
+ if (atomic_read(&(pkt_dev->skb->users)) != 1) {
+diff -urNp linux-2.6/net/core/skbuff.c new/net/core/skbuff.c
+--- linux-2.6/net/core/skbuff.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/core/skbuff.c 2006-07-07 16:20:30.000000000 +0200
+@@ -140,6 +140,7 @@ EXPORT_SYMBOL(skb_truesize_bug);
+ * Buffers may only be allocated from interrupts using a @gfp_mask of
+ * %GFP_ATOMIC.
+ */
++#ifndef CONFIG_HAVE_ARCH_ALLOC_SKB
+ struct sk_buff *__alloc_skb(unsigned int size, gfp_t gfp_mask,
+ int fclone)
+ {
+@@ -172,9 +173,9 @@ struct sk_buff *__alloc_skb(unsigned int
+ shinfo = skb_shinfo(skb);
+ atomic_set(&shinfo->dataref, 1);
+ shinfo->nr_frags = 0;
+- shinfo->tso_size = 0;
+- shinfo->tso_segs = 0;
+- shinfo->ufo_size = 0;
++ shinfo->gso_size = 0;
++ shinfo->gso_segs = 0;
++ shinfo->gso_type = 0;
+ shinfo->ip6_frag_id = 0;
+ shinfo->frag_list = NULL;
+
+@@ -194,6 +195,7 @@ nodata:
+ skb = NULL;
+ goto out;
+ }
++#endif /* !CONFIG_HAVE_ARCH_ALLOC_SKB */
+
+ /**
+ * alloc_skb_from_cache - allocate a network buffer
+@@ -211,14 +213,18 @@ nodata:
+ */
+ struct sk_buff *alloc_skb_from_cache(kmem_cache_t *cp,
+ unsigned int size,
+- gfp_t gfp_mask)
++ gfp_t gfp_mask,
++ int fclone)
+ {
++ kmem_cache_t *cache;
++ struct skb_shared_info *shinfo;
+ struct sk_buff *skb;
+ u8 *data;
+
++ cache = fclone ? skbuff_fclone_cache : skbuff_head_cache;
++
+ /* Get the HEAD */
+- skb = kmem_cache_alloc(skbuff_head_cache,
+- gfp_mask & ~__GFP_DMA);
++ skb = kmem_cache_alloc(cache, gfp_mask & ~__GFP_DMA);
+ if (!skb)
+ goto out;
+
+@@ -235,17 +241,29 @@ struct sk_buff *alloc_skb_from_cache(kme
+ skb->data = data;
+ skb->tail = data;
+ skb->end = data + size;
++ /* make sure we initialize shinfo sequentially */
++ shinfo = skb_shinfo(skb);
++ atomic_set(&shinfo->dataref, 1);
++ shinfo->nr_frags = 0;
++ shinfo->gso_size = 0;
++ shinfo->gso_segs = 0;
++ shinfo->gso_type = 0;
++ shinfo->ip6_frag_id = 0;
++ shinfo->frag_list = NULL;
+
+- atomic_set(&(skb_shinfo(skb)->dataref), 1);
+- skb_shinfo(skb)->nr_frags = 0;
+- skb_shinfo(skb)->tso_size = 0;
+- skb_shinfo(skb)->tso_segs = 0;
+- skb_shinfo(skb)->ufo_size = 0;
+- skb_shinfo(skb)->frag_list = NULL;
++ if (fclone) {
++ struct sk_buff *child = skb + 1;
++ atomic_t *fclone_ref = (atomic_t *) (child + 1);
++
++ skb->fclone = SKB_FCLONE_ORIG;
++ atomic_set(fclone_ref, 1);
++
++ child->fclone = SKB_FCLONE_UNAVAILABLE;
++ }
+ out:
+ return skb;
+ nodata:
+- kmem_cache_free(skbuff_head_cache, skb);
++ kmem_cache_free(cache, skb);
+ skb = NULL;
+ goto out;
+ }
+@@ -434,6 +453,10 @@ struct sk_buff *skb_clone(struct sk_buff
+ C(local_df);
+ n->cloned = 1;
+ n->nohdr = 0;
++#ifdef CONFIG_XEN
++ C(proto_data_valid);
++ C(proto_csum_blank);
++#endif
+ C(pkt_type);
+ C(ip_summed);
+ C(priority);
+@@ -527,9 +550,9 @@ static void copy_skb_header(struct sk_bu
+ new->tc_index = old->tc_index;
+ #endif
+ atomic_set(&new->users, 1);
+- skb_shinfo(new)->tso_size = skb_shinfo(old)->tso_size;
+- skb_shinfo(new)->tso_segs = skb_shinfo(old)->tso_segs;
+- skb_shinfo(new)->ufo_size = skb_shinfo(old)->ufo_size;
++ skb_shinfo(new)->gso_size = skb_shinfo(old)->gso_size;
++ skb_shinfo(new)->gso_segs = skb_shinfo(old)->gso_segs;
++ skb_shinfo(new)->gso_type = skb_shinfo(old)->gso_type;
+ }
+
+ /**
+@@ -1826,6 +1850,133 @@ unsigned char *skb_pull_rcsum(struct sk_
+
+ EXPORT_SYMBOL_GPL(skb_pull_rcsum);
+
++/**
++ * skb_segment - Perform protocol segmentation on skb.
++ * @skb: buffer to segment
++ * @features: features for the output path (see dev->features)
++ *
++ * This function performs segmentation on the given skb. It returns
++ * the segment at the given position. It returns NULL if there are
++ * no more segments to generate, or when an error is encountered.
++ */
++struct sk_buff *skb_segment(struct sk_buff *skb, int features)
++{
++ struct sk_buff *segs = NULL;
++ struct sk_buff *tail = NULL;
++ unsigned int mss = skb_shinfo(skb)->gso_size;
++ unsigned int doffset = skb->data - skb->mac.raw;
++ unsigned int offset = doffset;
++ unsigned int headroom;
++ unsigned int len;
++ int sg = features & NETIF_F_SG;
++ int nfrags = skb_shinfo(skb)->nr_frags;
++ int err = -ENOMEM;
++ int i = 0;
++ int pos;
++
++ __skb_push(skb, doffset);
++ headroom = skb_headroom(skb);
++ pos = skb_headlen(skb);
++
++ do {
++ struct sk_buff *nskb;
++ skb_frag_t *frag;
++ int hsize, nsize;
++ int k;
++ int size;
++
++ len = skb->len - offset;
++ if (len > mss)
++ len = mss;
++
++ hsize = skb_headlen(skb) - offset;
++ if (hsize < 0)
++ hsize = 0;
++ nsize = hsize + doffset;
++ if (nsize > len + doffset || !sg)
++ nsize = len + doffset;
++
++ nskb = alloc_skb(nsize + headroom, GFP_ATOMIC);
++ if (unlikely(!nskb))
++ goto err;
++
++ if (segs)
++ tail->next = nskb;
++ else
++ segs = nskb;
++ tail = nskb;
++
++ nskb->dev = skb->dev;
++ nskb->priority = skb->priority;
++ nskb->protocol = skb->protocol;
++ nskb->dst = dst_clone(skb->dst);
++ memcpy(nskb->cb, skb->cb, sizeof(skb->cb));
++ nskb->pkt_type = skb->pkt_type;
++ nskb->mac_len = skb->mac_len;
++
++ skb_reserve(nskb, headroom);
++ nskb->mac.raw = nskb->data;
++ nskb->nh.raw = nskb->data + skb->mac_len;
++ nskb->h.raw = nskb->nh.raw + (skb->h.raw - skb->nh.raw);
++ memcpy(skb_put(nskb, doffset), skb->data, doffset);
++
++ if (!sg) {
++ nskb->csum = skb_copy_and_csum_bits(skb, offset,
++ skb_put(nskb, len),
++ len, 0);
++ continue;
++ }
++
++ frag = skb_shinfo(nskb)->frags;
++ k = 0;
++
++ nskb->ip_summed = CHECKSUM_HW;
++ nskb->csum = skb->csum;
++ memcpy(skb_put(nskb, hsize), skb->data + offset, hsize);
++
++ while (pos < offset + len) {
++ BUG_ON(i >= nfrags);
++
++ *frag = skb_shinfo(skb)->frags[i];
++ get_page(frag->page);
++ size = frag->size;
++
++ if (pos < offset) {
++ frag->page_offset += offset - pos;
++ frag->size -= offset - pos;
++ }
++
++ k++;
++
++ if (pos + size <= offset + len) {
++ i++;
++ pos += size;
++ } else {
++ frag->size -= pos + size - (offset + len);
++ break;
++ }
++
++ frag++;
++ }
++
++ skb_shinfo(nskb)->nr_frags = k;
++ nskb->data_len = len - hsize;
++ nskb->len += nskb->data_len;
++ nskb->truesize += nskb->data_len;
++ } while ((offset += len) < skb->len);
++
++ return segs;
++
++err:
++ while ((skb = segs)) {
++ segs = skb->next;
++ kfree(skb);
++ }
++ return ERR_PTR(err);
++}
++
++EXPORT_SYMBOL_GPL(skb_segment);
++
+ void __init skb_init(void)
+ {
+ skbuff_head_cache = kmem_cache_create("skbuff_head_cache",
+diff -urNp linux-2.6/net/decnet/dn_nsp_in.c new/net/decnet/dn_nsp_in.c
+--- linux-2.6/net/decnet/dn_nsp_in.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/decnet/dn_nsp_in.c 2006-07-07 16:20:38.000000000 +0200
+@@ -801,8 +801,7 @@ got_it:
+ * We linearize everything except data segments here.
+ */
+ if (cb->nsp_flags & ~0x60) {
+- if (unlikely(skb_is_nonlinear(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0)
++ if (unlikely(skb_linearize(skb)))
+ goto free_out;
+ }
+
+diff -urNp linux-2.6/net/decnet/dn_route.c new/net/decnet/dn_route.c
+--- linux-2.6/net/decnet/dn_route.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/decnet/dn_route.c 2006-07-07 16:20:38.000000000 +0200
+@@ -629,8 +629,7 @@ int dn_route_rcv(struct sk_buff *skb, st
+ padlen);
+
+ if (flags & DN_RT_PKT_CNTL) {
+- if (unlikely(skb_is_nonlinear(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0)
++ if (unlikely(skb_linearize(skb)))
+ goto dump_it;
+
+ switch(flags & DN_RT_CNTL_MSK) {
+diff -urNp linux-2.6/net/ipv4/af_inet.c new/net/ipv4/af_inet.c
+--- linux-2.6/net/ipv4/af_inet.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/ipv4/af_inet.c 2006-07-07 16:20:38.000000000 +0200
+@@ -68,6 +68,7 @@
+ */
+
+ #include <linux/config.h>
++#include <linux/err.h>
+ #include <linux/errno.h>
+ #include <linux/types.h>
+ #include <linux/socket.h>
+@@ -1096,6 +1097,54 @@ int inet_sk_rebuild_header(struct sock *
+
+ EXPORT_SYMBOL(inet_sk_rebuild_header);
+
++static struct sk_buff *inet_gso_segment(struct sk_buff *skb, int features)
++{
++ struct sk_buff *segs = ERR_PTR(-EINVAL);
++ struct iphdr *iph;
++ struct net_protocol *ops;
++ int proto;
++ int ihl;
++ int id;
++
++ if (!pskb_may_pull(skb, sizeof(*iph)))
++ goto out;
++
++ iph = skb->nh.iph;
++ ihl = iph->ihl * 4;
++ if (ihl < sizeof(*iph))
++ goto out;
++
++ if (!pskb_may_pull(skb, ihl))
++ goto out;
++
++ skb->h.raw = __skb_pull(skb, ihl);
++ iph = skb->nh.iph;
++ id = ntohs(iph->id);
++ proto = iph->protocol & (MAX_INET_PROTOS - 1);
++ segs = ERR_PTR(-EPROTONOSUPPORT);
++
++ rcu_read_lock();
++ ops = rcu_dereference(inet_protos[proto]);
++ if (ops && ops->gso_segment)
++ segs = ops->gso_segment(skb, features);
++ rcu_read_unlock();
++
++ if (!segs || unlikely(IS_ERR(segs)))
++ goto out;
++
++ skb = segs;
++ do {
++ iph = skb->nh.iph;
++ iph->id = htons(id++);
++ iph->tot_len = htons(skb->len - skb->mac_len);
++ iph->check = 0;
++ iph->check = ip_fast_csum(skb->nh.raw, iph->ihl);
++ } while ((skb = skb->next));
++
++out:
++ return segs;
++}
++
+ #ifdef CONFIG_IP_MULTICAST
+ static struct net_protocol igmp_protocol = {
+ .handler = igmp_rcv,
+@@ -1105,6 +1154,7 @@ static struct net_protocol igmp_protocol
+ static struct net_protocol tcp_protocol = {
+ .handler = tcp_v4_rcv,
+ .err_handler = tcp_v4_err,
++ .gso_segment = tcp_tso_segment,
+ .no_policy = 1,
+ };
+
+@@ -1150,6 +1200,7 @@ static int ipv4_proc_init(void);
+ static struct packet_type ip_packet_type = {
+ .type = __constant_htons(ETH_P_IP),
+ .func = ip_rcv,
++ .gso_segment = inet_gso_segment,
+ };
+
+ static int __init inet_init(void)
+diff -urNp linux-2.6/net/ipv4/ipcomp.c new/net/ipv4/ipcomp.c
+--- linux-2.6/net/ipv4/ipcomp.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/ipv4/ipcomp.c 2006-07-07 16:20:39.000000000 +0200
+@@ -84,7 +84,7 @@ out:
+ static int ipcomp_input(struct xfrm_state *x, struct sk_buff *skb)
+ {
+ u8 nexthdr;
+- int err = 0;
++ int err = -ENOMEM;
+ struct iphdr *iph;
+ union {
+ struct iphdr iph;
+@@ -92,11 +92,8 @@ static int ipcomp_input(struct xfrm_stat
+ } tmp_iph;
+
+
+- if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0) {
+- err = -ENOMEM;
++ if (skb_linearize_cow(skb))
+ goto out;
+- }
+
+ skb->ip_summed = CHECKSUM_NONE;
+
+@@ -171,10 +168,8 @@ static int ipcomp_output(struct xfrm_sta
+ goto out_ok;
+ }
+
+- if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0) {
++ if (skb_linearize_cow(skb))
+ goto out_ok;
+- }
+
+ err = ipcomp_compress(x, skb);
+ iph = skb->nh.iph;
+diff -urNp linux-2.6/net/ipv4/ip_output.c new/net/ipv4/ip_output.c
+--- linux-2.6/net/ipv4/ip_output.c 2006-07-03 14:15:21.000000000 +0200
++++ new/net/ipv4/ip_output.c 2006-07-07 16:20:39.000000000 +0200
+@@ -210,8 +210,7 @@ static inline int ip_finish_output(struc
+ return dst_output(skb);
+ }
+ #endif
+- if (skb->len > dst_mtu(skb->dst) &&
+- !(skb_shinfo(skb)->ufo_size || skb_shinfo(skb)->tso_size))
++ if (skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->gso_size)
+ return ip_fragment(skb, ip_finish_output2);
+ else
+ return ip_finish_output2(skb);
+@@ -362,7 +361,7 @@ packet_routed:
+ }
+
+ ip_select_ident_more(iph, &rt->u.dst, sk,
+- (skb_shinfo(skb)->tso_segs ?: 1) - 1);
++ (skb_shinfo(skb)->gso_segs ?: 1) - 1);
+
+ /* Add an IP checksum. */
+ ip_send_check(iph);
+@@ -743,7 +742,8 @@ static inline int ip_ufo_append_data(str
+ (length - transhdrlen));
+ if (!err) {
+ /* specify the length of each IP datagram fragment*/
+- skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen);
++ skb_shinfo(skb)->gso_size = mtu - fragheaderlen;
++ skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4;
+ __skb_queue_tail(&sk->sk_write_queue, skb);
+
+ return 0;
+@@ -839,7 +839,7 @@ int ip_append_data(struct sock *sk,
+ */
+ if (transhdrlen &&
+ length + fragheaderlen <= mtu &&
+- rt->u.dst.dev->features&(NETIF_F_IP_CSUM|NETIF_F_NO_CSUM|NETIF_F_HW_CSUM) &&
++ rt->u.dst.dev->features & NETIF_F_ALL_CSUM &&
+ !exthdrlen)
+ csummode = CHECKSUM_HW;
+
+@@ -1086,14 +1086,16 @@ ssize_t ip_append_page(struct sock *sk,
+
+ inet->cork.length += size;
+ if ((sk->sk_protocol == IPPROTO_UDP) &&
+- (rt->u.dst.dev->features & NETIF_F_UFO))
+- skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen);
++ (rt->u.dst.dev->features & NETIF_F_UFO)) {
++ skb_shinfo(skb)->gso_size = mtu - fragheaderlen;
++ skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4;
++ }
+
+
+ while (size > 0) {
+ int i;
+
+- if (skb_shinfo(skb)->ufo_size)
++ if (skb_shinfo(skb)->gso_size)
+ len = size;
+ else {
+
+diff -urNp linux-2.6/net/ipv4/netfilter/ip_nat_proto_tcp.c new/net/ipv4/netfilter/ip_nat_proto_tcp.c
+--- linux-2.6/net/ipv4/netfilter/ip_nat_proto_tcp.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/netfilter/ip_nat_proto_tcp.c 2006-05-23 18:37:13.000000000 +0200
+@@ -129,10 +129,17 @@ tcp_manip_pkt(struct sk_buff **pskb,
+ if (hdrsize < sizeof(*hdr))
+ return 1;
+
+- hdr->check = ip_nat_cheat_check(~oldip, newip,
++#ifdef CONFIG_XEN
++ if ((*pskb)->proto_csum_blank) {
++ hdr->check = ip_nat_cheat_check(oldip, ~newip, hdr->check);
++ } else
++#endif
++ {
++ hdr->check = ip_nat_cheat_check(~oldip, newip,
+ ip_nat_cheat_check(oldport ^ 0xFFFF,
+ newport,
+ hdr->check));
++ }
+ return 1;
+ }
+
+diff -urNp linux-2.6/net/ipv4/netfilter/ip_nat_proto_udp.c new/net/ipv4/netfilter/ip_nat_proto_udp.c
+--- linux-2.6/net/ipv4/netfilter/ip_nat_proto_udp.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/netfilter/ip_nat_proto_udp.c 2006-05-23 18:37:13.000000000 +0200
+@@ -113,11 +113,19 @@ udp_manip_pkt(struct sk_buff **pskb,
+ newport = tuple->dst.u.udp.port;
+ portptr = &hdr->dest;
+ }
+- if (hdr->check) /* 0 is a special case meaning no checksum */
+- hdr->check = ip_nat_cheat_check(~oldip, newip,
++ if (hdr->check) { /* 0 is a special case meaning no checksum */
++#ifdef CONFIG_XEN
++ if ((*pskb)->proto_csum_blank) {
++ hdr->check = ip_nat_cheat_check(oldip, ~newip, hdr->check);
++ } else
++#endif
++ {
++ hdr->check = ip_nat_cheat_check(~oldip, newip,
+ ip_nat_cheat_check(*portptr ^ 0xFFFF,
+ newport,
+ hdr->check));
++ }
++ }
+ *portptr = newport;
+ return 1;
+ }
+diff -urNp linux-2.6/net/ipv4/tcp.c new/net/ipv4/tcp.c
+--- linux-2.6/net/ipv4/tcp.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/tcp.c 2006-07-07 16:20:59.000000000 +0200
+@@ -258,6 +258,7 @@
+ #include <linux/bootmem.h>
+ #include <linux/cache.h>
+ #include <linux/in.h>
++#include <linux/err.h>
+
+ #include <net/icmp.h>
+ #include <net/tcp.h>
+@@ -571,7 +572,7 @@ new_segment:
+ skb->ip_summed = CHECKSUM_HW;
+ tp->write_seq += copy;
+ TCP_SKB_CB(skb)->end_seq += copy;
+- skb_shinfo(skb)->tso_segs = 0;
++ skb_shinfo(skb)->gso_segs = 0;
+
+ if (!copied)
+ TCP_SKB_CB(skb)->flags &= ~TCPCB_FLAG_PSH;
+@@ -622,14 +623,10 @@ ssize_t tcp_sendpage(struct socket *sock
+ ssize_t res;
+ struct sock *sk = sock->sk;
+
+-#define TCP_ZC_CSUM_FLAGS (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM | NETIF_F_HW_CSUM)
+-
+ if (!(sk->sk_route_caps & NETIF_F_SG) ||
+- !(sk->sk_route_caps & TCP_ZC_CSUM_FLAGS))
++ !(sk->sk_route_caps & NETIF_F_ALL_CSUM))
+ return sock_no_sendpage(sock, page, offset, size, flags);
+
+-#undef TCP_ZC_CSUM_FLAGS
+-
+ lock_sock(sk);
+ TCP_CHECK_TIMER(sk);
+ res = do_tcp_sendpages(sk, &page, offset, size, flags);
+@@ -726,9 +723,7 @@ new_segment:
+ /*
+ * Check whether we can use HW checksum.
+ */
+- if (sk->sk_route_caps &
+- (NETIF_F_IP_CSUM | NETIF_F_NO_CSUM |
+- NETIF_F_HW_CSUM))
++ if (sk->sk_route_caps & NETIF_F_ALL_CSUM)
+ skb->ip_summed = CHECKSUM_HW;
+
+ skb_entail(sk, tp, skb);
+@@ -824,7 +819,7 @@ new_segment:
+
+ tp->write_seq += copy;
+ TCP_SKB_CB(skb)->end_seq += copy;
+- skb_shinfo(skb)->tso_segs = 0;
++ skb_shinfo(skb)->gso_segs = 0;
+
+ from += copy;
+ copied += copy;
+@@ -2071,6 +2066,77 @@ int compat_tcp_getsockopt(struct sock *s
+ EXPORT_SYMBOL(compat_tcp_getsockopt);
+ #endif
+
++struct sk_buff *tcp_tso_segment(struct sk_buff *skb, int features)
++{
++ struct sk_buff *segs = ERR_PTR(-EINVAL);
++ struct tcphdr *th;
++ unsigned thlen;
++ unsigned int seq;
++ unsigned int delta;
++ unsigned int oldlen;
++ unsigned int len;
++
++ if (!pskb_may_pull(skb, sizeof(*th)))
++ goto out;
++
++ th = skb->h.th;
++ thlen = th->doff * 4;
++ if (thlen < sizeof(*th))
++ goto out;
++
++ if (!pskb_may_pull(skb, thlen))
++ goto out;
++
++ oldlen = (u16)~skb->len;
++ __skb_pull(skb, thlen);
++
++ if (skb_gso_ok(skb, features | NETIF_F_GSO_ROBUST)) {
++ /* Packet is from an untrusted source, reset gso_segs. */
++ int mss = skb_shinfo(skb)->gso_size;
++
++ skb_shinfo(skb)->gso_segs = (skb->len + mss - 1) / mss;
++
++ segs = NULL;
++ goto out;
++ }
++
++ segs = skb_segment(skb, features);
++ if (IS_ERR(segs))
++ goto out;
++
++ len = skb_shinfo(skb)->gso_size;
++ delta = htonl(oldlen + (thlen + len));
++
++ skb = segs;
++ th = skb->h.th;
++ seq = ntohl(th->seq);
++
++ do {
++ th->fin = th->psh = 0;
++
++ th->check = ~csum_fold(th->check + delta);
++ if (skb->ip_summed != CHECKSUM_HW)
++ th->check = csum_fold(csum_partial(skb->h.raw, thlen,
++ skb->csum));
++
++ seq += len;
++ skb = skb->next;
++ th = skb->h.th;
++
++ th->seq = htonl(seq);
++ th->cwr = 0;
++ } while (skb->next);
++
++ delta = htonl(oldlen + (skb->tail - skb->h.raw) + skb->data_len);
++ th->check = ~csum_fold(th->check + delta);
++ if (skb->ip_summed != CHECKSUM_HW)
++ th->check = csum_fold(csum_partial(skb->h.raw, thlen,
++ skb->csum));
++
++out:
++ return segs;
++}
++
+ extern void __skb_cb_too_small_for_tcp(int, int);
+ extern struct tcp_congestion_ops tcp_reno;
+
+diff -urNp linux-2.6/net/ipv4/tcp_input.c new/net/ipv4/tcp_input.c
+--- linux-2.6/net/ipv4/tcp_input.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/tcp_input.c 2006-07-07 16:21:07.000000000 +0200
+@@ -1072,7 +1072,7 @@ tcp_sacktag_write_queue(struct sock *sk,
+ else
+ pkt_len = (end_seq -
+ TCP_SKB_CB(skb)->seq);
+- if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->tso_size))
++ if (tcp_fragment(sk, skb, pkt_len, skb_shinfo(skb)->gso_size))
+ break;
+ pcount = tcp_skb_pcount(skb);
+ }
+diff -urNp linux-2.6/net/ipv4/tcp_output.c new/net/ipv4/tcp_output.c
+--- linux-2.6/net/ipv4/tcp_output.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/tcp_output.c 2006-07-07 16:21:08.000000000 +0200
+@@ -511,15 +511,17 @@ static void tcp_set_skb_tso_segs(struct
+ /* Avoid the costly divide in the normal
+ * non-TSO case.
+ */
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+ } else {
+ unsigned int factor;
+
+ factor = skb->len + (mss_now - 1);
+ factor /= mss_now;
+- skb_shinfo(skb)->tso_segs = factor;
+- skb_shinfo(skb)->tso_size = mss_now;
++ skb_shinfo(skb)->gso_segs = factor;
++ skb_shinfo(skb)->gso_size = mss_now;
++ skb_shinfo(skb)->gso_type = SKB_GSO_TCPV4;
+ }
+ }
+
+@@ -910,7 +912,7 @@ static int tcp_init_tso_segs(struct sock
+
+ if (!tso_segs ||
+ (tso_segs > 1 &&
+- skb_shinfo(skb)->tso_size != mss_now)) {
++ tcp_skb_mss(skb) != mss_now)) {
+ tcp_set_skb_tso_segs(sk, skb, mss_now);
+ tso_segs = tcp_skb_pcount(skb);
+ }
+@@ -1720,8 +1722,9 @@ int tcp_retransmit_skb(struct sock *sk,
+ tp->snd_una == (TCP_SKB_CB(skb)->end_seq - 1)) {
+ if (!pskb_trim(skb, 0)) {
+ TCP_SKB_CB(skb)->seq = TCP_SKB_CB(skb)->end_seq - 1;
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+ skb->ip_summed = CHECKSUM_NONE;
+ skb->csum = 0;
+ }
+@@ -1926,8 +1929,9 @@ void tcp_send_fin(struct sock *sk)
+ skb->csum = 0;
+ TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_FIN);
+ TCP_SKB_CB(skb)->sacked = 0;
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+
+ /* FIN eats a sequence byte, write_seq advanced by tcp_queue_skb(). */
+ TCP_SKB_CB(skb)->seq = tp->write_seq;
+@@ -1959,8 +1963,9 @@ void tcp_send_active_reset(struct sock *
+ skb->csum = 0;
+ TCP_SKB_CB(skb)->flags = (TCPCB_FLAG_ACK | TCPCB_FLAG_RST);
+ TCP_SKB_CB(skb)->sacked = 0;
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+
+ /* Send it off. */
+ TCP_SKB_CB(skb)->seq = tcp_acceptable_seq(sk, tp);
+@@ -2043,8 +2048,9 @@ struct sk_buff * tcp_make_synack(struct
+ TCP_SKB_CB(skb)->seq = tcp_rsk(req)->snt_isn;
+ TCP_SKB_CB(skb)->end_seq = TCP_SKB_CB(skb)->seq + 1;
+ TCP_SKB_CB(skb)->sacked = 0;
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+ th->seq = htonl(TCP_SKB_CB(skb)->seq);
+ th->ack_seq = htonl(tcp_rsk(req)->rcv_isn + 1);
+ if (req->rcv_wnd == 0) { /* ignored for retransmitted syns */
+@@ -2148,8 +2154,9 @@ int tcp_connect(struct sock *sk)
+ TCP_SKB_CB(buff)->flags = TCPCB_FLAG_SYN;
+ TCP_ECN_send_syn(sk, tp, buff);
+ TCP_SKB_CB(buff)->sacked = 0;
+- skb_shinfo(buff)->tso_segs = 1;
+- skb_shinfo(buff)->tso_size = 0;
++ skb_shinfo(buff)->gso_segs = 1;
++ skb_shinfo(buff)->gso_size = 0;
++ skb_shinfo(buff)->gso_type = 0;
+ buff->csum = 0;
+ TCP_SKB_CB(buff)->seq = tp->write_seq++;
+ TCP_SKB_CB(buff)->end_seq = tp->write_seq;
+@@ -2253,8 +2260,9 @@ void tcp_send_ack(struct sock *sk)
+ buff->csum = 0;
+ TCP_SKB_CB(buff)->flags = TCPCB_FLAG_ACK;
+ TCP_SKB_CB(buff)->sacked = 0;
+- skb_shinfo(buff)->tso_segs = 1;
+- skb_shinfo(buff)->tso_size = 0;
++ skb_shinfo(buff)->gso_segs = 1;
++ skb_shinfo(buff)->gso_size = 0;
++ skb_shinfo(buff)->gso_type = 0;
+
+ /* Send it off, this clears delayed acks for us. */
+ TCP_SKB_CB(buff)->seq = TCP_SKB_CB(buff)->end_seq = tcp_acceptable_seq(sk, tp);
+@@ -2289,8 +2297,9 @@ static int tcp_xmit_probe_skb(struct soc
+ skb->csum = 0;
+ TCP_SKB_CB(skb)->flags = TCPCB_FLAG_ACK;
+ TCP_SKB_CB(skb)->sacked = urgent;
+- skb_shinfo(skb)->tso_segs = 1;
+- skb_shinfo(skb)->tso_size = 0;
++ skb_shinfo(skb)->gso_segs = 1;
++ skb_shinfo(skb)->gso_size = 0;
++ skb_shinfo(skb)->gso_type = 0;
+
+ /* Use a previous sequence. This should cause the other
+ * end to send an ack. Don't queue or clone SKB, just
+diff -urNp linux-2.6/net/ipv4/xfrm4_output.c new/net/ipv4/xfrm4_output.c
+--- linux-2.6/net/ipv4/xfrm4_output.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv4/xfrm4_output.c 2006-07-07 16:21:08.000000000 +0200
+@@ -9,6 +9,8 @@
+ */
+
+ #include <linux/compiler.h>
++#include <linux/if_ether.h>
++#include <linux/kernel.h>
+ #include <linux/skbuff.h>
+ #include <linux/spinlock.h>
+ #include <linux/netfilter_ipv4.h>
+@@ -17,6 +19,8 @@
+ #include <net/xfrm.h>
+ #include <net/icmp.h>
+
++extern int skb_checksum_setup(struct sk_buff *skb);
++
+ /* Add encapsulation header.
+ *
+ * In transport mode, the IP header will be moved forward to make space
+@@ -103,6 +107,10 @@ static int xfrm4_output_one(struct sk_bu
+ struct xfrm_state *x = dst->xfrm;
+ int err;
+
++ err = skb_checksum_setup(skb);
++ if (err)
++ goto error_nolock;
++
+ if (skb->ip_summed == CHECKSUM_HW) {
+ err = skb_checksum_help(skb, 0);
+ if (err)
+@@ -152,16 +160,10 @@ error_nolock:
+ goto out_exit;
+ }
+
+-static int xfrm4_output_finish(struct sk_buff *skb)
++static int xfrm4_output_finish2(struct sk_buff *skb)
+ {
+ int err;
+
+-#ifdef CONFIG_NETFILTER
+- if (!skb->dst->xfrm) {
+- IPCB(skb)->flags |= IPSKB_REROUTED;
+- return dst_output(skb);
+- }
+-#endif
+ while (likely((err = xfrm4_output_one(skb)) == 0)) {
+ nf_reset(skb);
+
+@@ -174,7 +176,7 @@ static int xfrm4_output_finish(struct sk
+ return dst_output(skb);
+
+ err = nf_hook(PF_INET, NF_IP_POST_ROUTING, &skb, NULL,
+- skb->dst->dev, xfrm4_output_finish);
++ skb->dst->dev, xfrm4_output_finish2);
+ if (unlikely(err != 1))
+ break;
+ }
+@@ -182,6 +184,48 @@ static int xfrm4_output_finish(struct sk
+ return err;
+ }
+
++static int xfrm4_output_finish(struct sk_buff *skb)
++{
++ struct sk_buff *segs;
++
++#ifdef CONFIG_NETFILTER
++ if (!skb->dst->xfrm) {
++ IPCB(skb)->flags |= IPSKB_REROUTED;
++ return dst_output(skb);
++ }
++#endif
++
++ if (!skb_shinfo(skb)->gso_size)
++ return xfrm4_output_finish2(skb);
++
++ skb->protocol = htons(ETH_P_IP);
++ segs = skb_gso_segment(skb, 0);
++ kfree_skb(skb);
++ if (unlikely(IS_ERR(segs)))
++ return PTR_ERR(segs);
++
++ do {
++ struct sk_buff *nskb = segs->next;
++ int err;
++
++ segs->next = NULL;
++ err = xfrm4_output_finish2(segs);
++
++ if (unlikely(err)) {
++ while ((segs = nskb)) {
++ nskb = segs->next;
++ segs->next = NULL;
++ kfree_skb(segs);
++ }
++ return err;
++ }
++
++ segs = nskb;
++ } while (segs);
++
++ return 0;
++}
++
+ int xfrm4_output(struct sk_buff *skb)
+ {
+ return NF_HOOK_COND(PF_INET, NF_IP_POST_ROUTING, skb, NULL, skb->dst->dev,
+diff -urNp linux-2.6/net/ipv6/addrconf.c new/net/ipv6/addrconf.c
+--- linux-2.6/net/ipv6/addrconf.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv6/addrconf.c 2006-06-28 14:32:14.000000000 +0200
+@@ -2461,6 +2461,7 @@ static void addrconf_dad_start(struct in
+ spin_lock_bh(&ifp->lock);
+
+ if (dev->flags&(IFF_NOARP|IFF_LOOPBACK) ||
++ !(dev->flags&IFF_MULTICAST) ||
+ !(ifp->flags&IFA_F_TENTATIVE)) {
+ ifp->flags &= ~IFA_F_TENTATIVE;
+ spin_unlock_bh(&ifp->lock);
+@@ -2545,6 +2546,7 @@ static void addrconf_dad_completed(struc
+ if (ifp->idev->cnf.forwarding == 0 &&
+ ifp->idev->cnf.rtr_solicits > 0 &&
+ (dev->flags&IFF_LOOPBACK) == 0 &&
++ (dev->flags & IFF_MULTICAST) &&
+ (ipv6_addr_type(&ifp->addr) & IPV6_ADDR_LINKLOCAL)) {
+ struct in6_addr all_routers;
+
+diff -urNp linux-2.6/net/ipv6/ip6_output.c new/net/ipv6/ip6_output.c
+--- linux-2.6/net/ipv6/ip6_output.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv6/ip6_output.c 2006-07-07 16:21:08.000000000 +0200
+@@ -147,7 +147,7 @@ static int ip6_output2(struct sk_buff *s
+
+ int ip6_output(struct sk_buff *skb)
+ {
+- if ((skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->ufo_size) ||
++ if ((skb->len > dst_mtu(skb->dst) && !skb_shinfo(skb)->gso_size) ||
+ dst_allfrag(skb->dst))
+ return ip6_fragment(skb, ip6_output2);
+ else
+@@ -830,8 +830,9 @@ static inline int ip6_ufo_append_data(st
+ struct frag_hdr fhdr;
+
+ /* specify the length of each IP datagram fragment*/
+- skb_shinfo(skb)->ufo_size = (mtu - fragheaderlen) -
+- sizeof(struct frag_hdr);
++ skb_shinfo(skb)->gso_size = mtu - fragheaderlen -
++ sizeof(struct frag_hdr);
++ skb_shinfo(skb)->gso_type = SKB_GSO_UDPV4;
+ ipv6_select_ident(skb, &fhdr);
+ skb_shinfo(skb)->ip6_frag_id = fhdr.identification;
+ __skb_queue_tail(&sk->sk_write_queue, skb);
+diff -urNp linux-2.6/net/ipv6/ipcomp6.c new/net/ipv6/ipcomp6.c
+--- linux-2.6/net/ipv6/ipcomp6.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv6/ipcomp6.c 2006-07-07 16:21:08.000000000 +0200
+@@ -65,7 +65,7 @@ static LIST_HEAD(ipcomp6_tfms_list);
+
+ static int ipcomp6_input(struct xfrm_state *x, struct sk_buff *skb)
+ {
+- int err = 0;
++ int err = -ENOMEM;
+ u8 nexthdr = 0;
+ int hdr_len = skb->h.raw - skb->nh.raw;
+ unsigned char *tmp_hdr = NULL;
+@@ -76,11 +76,8 @@ static int ipcomp6_input(struct xfrm_sta
+ struct crypto_tfm *tfm;
+ int cpu;
+
+- if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0) {
+- err = -ENOMEM;
++ if (skb_linearize_cow(skb))
+ goto out;
+- }
+
+ skb->ip_summed = CHECKSUM_NONE;
+
+@@ -159,10 +156,8 @@ static int ipcomp6_output(struct xfrm_st
+ goto out_ok;
+ }
+
+- if ((skb_is_nonlinear(skb) || skb_cloned(skb)) &&
+- skb_linearize(skb, GFP_ATOMIC) != 0) {
++ if (skb_linearize_cow(skb))
+ goto out_ok;
+- }
+
+ /* compression */
+ plen = skb->len - hdr_len;
+diff -urNp linux-2.6/net/ipv6/xfrm6_output.c new/net/ipv6/xfrm6_output.c
+--- linux-2.6/net/ipv6/xfrm6_output.c 2006-07-03 14:15:22.000000000 +0200
++++ new/net/ipv6/xfrm6_output.c 2006-07-07 15:10:03.000000000 +0200
+@@ -151,7 +151,7 @@ error_nolock:
+ goto out_exit;
+ }
+
+-static int xfrm6_output_finish(struct sk_buff *skb)
++static int xfrm6_output_finish2(struct sk_buff *skb)
+ {
+ int err;
+
+@@ -167,7 +167,7 @@ static int xfrm6_output_finish(struct sk
+ return dst_output(skb);
+
+ err = nf_hook(PF_INET6, NF_IP6_POST_ROUTING, &skb, NULL,
+- skb->dst->dev, xfrm6_output_finish);
++ skb->dst->dev, xfrm6_output_finish2);
+ if (unlikely(err != 1))
+ break;
+ }
+@@ -175,6 +175,41 @@ static int xfrm6_output_finish(struct sk
+ return err;
+ }
+
++static int xfrm6_output_finish(struct sk_buff *skb)
++{
++ struct sk_buff *segs;
++
++ if (!skb_shinfo(skb)->gso_size)
++ return xfrm6_output_finish2(skb);
++
++ skb->protocol = htons(ETH_P_IP);
++ segs = skb_gso_segment(skb, 0);
++ kfree_skb(skb);
++ if (unlikely(IS_ERR(segs)))
++ return PTR_ERR(segs);
++
++ do {
++ struct sk_buff *nskb = segs->next;
++ int err;
++
++ segs->next = NULL;
++ err = xfrm6_output_finish2(segs);
++
++ if (unlikely(err)) {
++ while ((segs = nskb)) {
++ nskb = segs->next;
++ segs->next = NULL;
++ kfree_skb(segs);
++ }
++ return err;
++ }
++
++ segs = nskb;
++ } while (segs);
++
++ return 0;
++}
++
+ int xfrm6_output(struct sk_buff *skb)
+ {
+ return NF_HOOK(PF_INET6, NF_IP6_POST_ROUTING, skb, NULL, skb->dst->dev,
+diff -urNp linux-2.6/net/sched/sch_teql.c new/net/sched/sch_teql.c
+--- linux-2.6/net/sched/sch_teql.c 2006-07-03 14:15:23.000000000 +0200
++++ new/net/sched/sch_teql.c 2006-07-07 15:10:03.000000000 +0200
+@@ -302,20 +302,17 @@ restart:
+
+ switch (teql_resolve(skb, skb_res, slave)) {
+ case 0:
+- if (spin_trylock(&slave->xmit_lock)) {
+- slave->xmit_lock_owner = smp_processor_id();
++ if (netif_tx_trylock(slave)) {
+ if (!netif_queue_stopped(slave) &&
+ slave->hard_start_xmit(skb, slave) == 0) {
+- slave->xmit_lock_owner = -1;
+- spin_unlock(&slave->xmit_lock);
++ netif_tx_unlock(slave);
+ master->slaves = NEXT_SLAVE(q);
+ netif_wake_queue(dev);
+ master->stats.tx_packets++;
+ master->stats.tx_bytes += len;
+ return 0;
+ }
+- slave->xmit_lock_owner = -1;
+- spin_unlock(&slave->xmit_lock);
++ netif_tx_unlock(slave);
+ }
+ if (netif_queue_stopped(dev))
+ busy = 1;
+diff -urNp linux-2.6/scripts/Makefile.xen new/scripts/Makefile.xen
+--- linux-2.6/scripts/Makefile.xen 1970-01-01 01:00:00.000000000 +0100
++++ new/scripts/Makefile.xen 2006-05-09 12:40:15.000000000 +0200
+@@ -0,0 +1,14 @@
++
++# cherrypickxen($1 = allobj)
++cherrypickxen = $(foreach var, $(1), \
++ $(shell o=$(var); \
++ c=$${o%.o}-xen.c; \
++ s=$${o%.o}-xen.S; \
++ oxen=$${o%.o}-xen.o; \
++ [ -f $(srctree)/$(src)/$${c} ] || \
++ [ -f $(srctree)/$(src)/$${s} ] \
++ && echo $$oxen \
++ || echo $(var) ) \
++ )
++# filterxen($1 = allobj, $2 = noobjs)
++filterxen = $(filter-out $(2), $(1))
+--- linux-2.6.17/drivers/block/aoe/aoenet.c~ 2006-06-18 03:49:35.000000000 +0200
++++ linux-2.6.17/drivers/block/aoe/aoenet.c 2006-09-17 03:44:30.000000000 +0200
+@@ -117,7 +117,7 @@
+ if (skb == NULL)
+ return 0;
+ if (skb_is_nonlinear(skb))
+- if (skb_linearize(skb, GFP_ATOMIC) < 0)
++ if (skb_linearize(skb) < 0)
+ goto exit;
+ if (!is_aoe_netif(ifp))
+ goto exit;
+--- linux-2.6.17/drivers/net/sk98lin/sky2.c~ 2006-09-17 04:09:56.000000000 +0200
++++ linux-2.6.17/drivers/net/sk98lin/sky2.c 2006-09-17 04:42:45.000000000 +0200
+@@ -990,7 +990,7 @@
+ SK_DBG_MSG(pAC, SK_DBGMOD_DRV, SK_DBGCAT_DRV_TX_PROGRESS,
+ ("\tGet LE\n"));
+ #ifdef NETIF_F_TSO
+- Mss = skb_shinfo(pSkPacket->pMBuf)->tso_size;
++ Mss = skb_shinfo(pSkPacket->pMBuf)->gso_size;
+ if (Mss) {
+ TcpOptLen = ((pSkPacket->pMBuf->h.th->doff - 5) * 4);
+ IpTcpLen = ((pSkPacket->pMBuf->nh.iph->ihl * 4) +
+--- linux-2.6.17/drivers/net/imq.c~ 2006-09-17 04:46:00.000000000 +0200
++++ linux-2.6.17/drivers/net/imq.c 2006-09-17 12:34:36.000000000 +0200
+@@ -201,7 +201,7 @@
+ ret = 0;
+ }
+ }
+- if (spin_is_locked(&dev->xmit_lock))
++ if (spin_is_locked(&dev->_xmit_lock))
+ netif_schedule(dev);
+ else
+
+--- linux-2.6.17/net/sched/sch_generic.c.orig 2006-09-17 19:56:31.000000000 +0200
++++ linux-2.6.17/net/sched/sch_generic.c 2006-09-17 20:03:04.000000000 +0200
+@@ -75,9 +75,9 @@
+ dev->queue_lock serializes queue accesses for this device
+ AND dev->qdisc pointer itself.
+
+- dev->xmit_lock serializes accesses to device driver.
++ netif_tx_lock serializes accesses to device driver.
+
+- dev->queue_lock and dev->xmit_lock are mutually exclusive,
++ dev->queue_lock and netif_tx_lock are mutually exclusive,
+ if one is grabbed, another must be free.
+ */
+
+@@ -93,14 +93,17 @@
+ NOTE: Called under dev->queue_lock with locally disabled BH.
+ */
+
+-int qdisc_restart(struct net_device *dev)
++static inline int qdisc_restart(struct net_device *dev)
+ {
+ struct Qdisc *q = dev->qdisc;
+ struct sk_buff *skb;
+
+ /* Dequeue packet */
+- if ((skb = q->dequeue(q)) != NULL) {
++ if (((skb = dev->gso_skb)) || ((skb = q->dequeue(q)))) {
+ unsigned nolock = (dev->features & NETIF_F_LLTX);
++
++ dev->gso_skb = NULL;
++
+ /*
+ * When the driver has LLTX set it does its own locking
+ * in start_xmit. No need to add additional overhead by
+@@ -111,7 +114,7 @@
+ * will be requeued.
+ */
+ if (!nolock) {
+- if (!spin_trylock(&dev->xmit_lock)) {
++ if (!netif_tx_trylock(dev)) {
+ collision:
+ /* So, someone grabbed the driver. */
+
+@@ -129,8 +132,6 @@
+ __get_cpu_var(netdev_rx_stat).cpu_collision++;
+ goto requeue;
+ }
+- /* Remember that the driver is grabbed by us. */
+- dev->xmit_lock_owner = smp_processor_id();
+ }
+
+ {
+@@ -139,20 +140,10 @@
+
+ if (!netif_queue_stopped(dev)) {
+ int ret;
+-
+- if (netdev_nit
+-#if defined(CONFIG_IMQ) || defined(CONFIG_IMQ_MODULE)
+- && !(skb->imq_flags & IMQ_F_ENQUEUE)
+-#endif
+- )
+-
+- dev_queue_xmit_nit(skb, dev);
+-
+- ret = dev->hard_start_xmit(skb, dev);
++ ret = dev_hard_start_xmit(skb, dev);
+ if (ret == NETDEV_TX_OK) {
+ if (!nolock) {
+- dev->xmit_lock_owner = -1;
+- spin_unlock(&dev->xmit_lock);
++ netif_tx_unlock(dev);
+ }
+ spin_lock(&dev->queue_lock);
+ return -1;
+@@ -166,8 +157,7 @@
+ /* NETDEV_TX_BUSY - we need to requeue */
+ /* Release the driver */
+ if (!nolock) {
+- dev->xmit_lock_owner = -1;
+- spin_unlock(&dev->xmit_lock);
++ netif_tx_unlock(dev);
+ }
+ spin_lock(&dev->queue_lock);
+ q = dev->qdisc;
+@@ -184,7 +174,10 @@
+ */
+
+ requeue:
+- q->ops->requeue(skb, q);
++ if (skb->next)
++ dev->gso_skb = skb;
++ else
++ q->ops->requeue(skb, q);
+ netif_schedule(dev);
+ return 1;
+ }
+@@ -192,11 +185,23 @@
+ return q->q.qlen;
+ }
+
++void __qdisc_run(struct net_device *dev)
++{
++ if (unlikely(dev->qdisc == &noop_qdisc))
++ goto out;
++
++ while (qdisc_restart(dev) < 0 && !netif_queue_stopped(dev))
++ /* NOTHING */;
++
++out:
++ clear_bit(__LINK_STATE_QDISC_RUNNING, &dev->state);
++}
++
+ static void dev_watchdog(unsigned long arg)
+ {
+ struct net_device *dev = (struct net_device *)arg;
+
+- spin_lock(&dev->xmit_lock);
++ netif_tx_lock(dev);
+ if (dev->qdisc != &noop_qdisc) {
+ if (netif_device_present(dev) &&
+ netif_running(dev) &&
+@@ -212,7 +217,7 @@
+ dev_hold(dev);
+ }
+ }
+- spin_unlock(&dev->xmit_lock);
++ netif_tx_unlock(dev);
+
+ dev_put(dev);
+ }
+@@ -236,17 +241,17 @@
+
+ static void dev_watchdog_up(struct net_device *dev)
+ {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ __netdev_watchdog_up(dev);
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ }
+
+ static void dev_watchdog_down(struct net_device *dev)
+ {
+- spin_lock_bh(&dev->xmit_lock);
++ netif_tx_lock_bh(dev);
+ if (del_timer(&dev->watchdog_timer))
+ dev_put(dev);
+- spin_unlock_bh(&dev->xmit_lock);
++ netif_tx_unlock_bh(dev);
+ }
+
+ void netif_carrier_on(struct net_device *dev)
+@@ -588,10 +593,17 @@
+
+ dev_watchdog_down(dev);
+
+- while (test_bit(__LINK_STATE_SCHED, &dev->state))
++ /* Wait for outstanding dev_queue_xmit calls. */
++ synchronize_rcu();
++
++ /* Wait for outstanding qdisc_run calls. */
++ while (test_bit(__LINK_STATE_QDISC_RUNNING, &dev->state))
+ yield();
+
+- spin_unlock_wait(&dev->xmit_lock);
++ if (dev->gso_skb) {
++ kfree_skb(dev->gso_skb);
++ dev->gso_skb = NULL;
++ }
+ }
+
+ void dev_init_scheduler(struct net_device *dev)
+@@ -633,6 +645,5 @@
+ EXPORT_SYMBOL(qdisc_alloc);
+ EXPORT_SYMBOL(qdisc_destroy);
+ EXPORT_SYMBOL(qdisc_reset);
+-EXPORT_SYMBOL(qdisc_restart);
+ EXPORT_SYMBOL(qdisc_lock_tree);
+ EXPORT_SYMBOL(qdisc_unlock_tree);
projects / packages / kernel.git / commitdiff