1 diff --git a/Documentation/sysrq.txt b/Documentation/sysrq.txt
2 index 3a3b30ac2a75..9e0745cafbd8 100644
3 --- a/Documentation/sysrq.txt
4 +++ b/Documentation/sysrq.txt
5 @@ -59,10 +59,17 @@ On PowerPC - Press 'ALT - Print Screen (or F13) - <command key>,
6 On other - If you know of the key combos for other architectures, please
7 let me know so I can add them to this section.
9 -On all - write a character to /proc/sysrq-trigger. e.g.:
11 +On all - write a character to /proc/sysrq-trigger, e.g.:
12 echo t > /proc/sysrq-trigger
14 +On all - Enable network SysRq by writing a cookie to icmp_echo_sysrq, e.g.
15 + echo 0x01020304 >/proc/sys/net/ipv4/icmp_echo_sysrq
16 + Send an ICMP echo request with this pattern plus the particular
17 + SysRq command key. Example:
18 + # ping -c1 -s57 -p0102030468
19 + will trigger the SysRq-H (help) command.
22 * What are the 'command' keys?
23 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
24 'b' - Will immediately reboot the system without syncing or unmounting
25 diff --git a/Documentation/trace/histograms.txt b/Documentation/trace/histograms.txt
27 index 000000000000..6f2aeabf7faa
29 +++ b/Documentation/trace/histograms.txt
31 + Using the Linux Kernel Latency Histograms
34 +This document gives a short explanation how to enable, configure and use
35 +latency histograms. Latency histograms are primarily relevant in the
36 +context of real-time enabled kernels (CONFIG_PREEMPT/CONFIG_PREEMPT_RT)
37 +and are used in the quality management of the Linux real-time
41 +* Purpose of latency histograms
43 +A latency histogram continuously accumulates the frequencies of latency
44 +data. There are two types of histograms
45 +- potential sources of latencies
46 +- effective latencies
49 +* Potential sources of latencies
51 +Potential sources of latencies are code segments where interrupts,
52 +preemption or both are disabled (aka critical sections). To create
53 +histograms of potential sources of latency, the kernel stores the time
54 +stamp at the start of a critical section, determines the time elapsed
55 +when the end of the section is reached, and increments the frequency
56 +counter of that latency value - irrespective of whether any concurrently
57 +running process is affected by latency or not.
58 +- Configuration items (in the Kernel hacking/Tracers submenu)
59 + CONFIG_INTERRUPT_OFF_LATENCY
60 + CONFIG_PREEMPT_OFF_LATENCY
63 +* Effective latencies
65 +Effective latencies are actually occuring during wakeup of a process. To
66 +determine effective latencies, the kernel stores the time stamp when a
67 +process is scheduled to be woken up, and determines the duration of the
68 +wakeup time shortly before control is passed over to this process. Note
69 +that the apparent latency in user space may be somewhat longer, since the
70 +process may be interrupted after control is passed over to it but before
71 +the execution in user space takes place. Simply measuring the interval
72 +between enqueuing and wakeup may also not appropriate in cases when a
73 +process is scheduled as a result of a timer expiration. The timer may have
74 +missed its deadline, e.g. due to disabled interrupts, but this latency
75 +would not be registered. Therefore, the offsets of missed timers are
76 +recorded in a separate histogram. If both wakeup latency and missed timer
77 +offsets are configured and enabled, a third histogram may be enabled that
78 +records the overall latency as a sum of the timer latency, if any, and the
79 +wakeup latency. This histogram is called "timerandwakeup".
80 +- Configuration items (in the Kernel hacking/Tracers submenu)
81 + CONFIG_WAKEUP_LATENCY
82 + CONFIG_MISSED_TIMER_OFSETS
87 +The interface to the administration of the latency histograms is located
88 +in the debugfs file system. To mount it, either enter
90 +mount -t sysfs nodev /sys
91 +mount -t debugfs nodev /sys/kernel/debug
93 +from shell command line level, or add
95 +nodev /sys sysfs defaults 0 0
96 +nodev /sys/kernel/debug debugfs defaults 0 0
98 +to the file /etc/fstab. All latency histogram related files are then
99 +available in the directory /sys/kernel/debug/tracing/latency_hist. A
100 +particular histogram type is enabled by writing non-zero to the related
101 +variable in the /sys/kernel/debug/tracing/latency_hist/enable directory.
102 +Select "preemptirqsoff" for the histograms of potential sources of
103 +latencies and "wakeup" for histograms of effective latencies etc. The
104 +histogram data - one per CPU - are available in the files
106 +/sys/kernel/debug/tracing/latency_hist/preemptoff/CPUx
107 +/sys/kernel/debug/tracing/latency_hist/irqsoff/CPUx
108 +/sys/kernel/debug/tracing/latency_hist/preemptirqsoff/CPUx
109 +/sys/kernel/debug/tracing/latency_hist/wakeup/CPUx
110 +/sys/kernel/debug/tracing/latency_hist/wakeup/sharedprio/CPUx
111 +/sys/kernel/debug/tracing/latency_hist/missed_timer_offsets/CPUx
112 +/sys/kernel/debug/tracing/latency_hist/timerandwakeup/CPUx
114 +The histograms are reset by writing non-zero to the file "reset" in a
115 +particular latency directory. To reset all latency data, use
119 +TRACINGDIR=/sys/kernel/debug/tracing
120 +HISTDIR=$TRACINGDIR/latency_hist
125 + for i in `find . | grep /reset$`
134 +Latency data are stored with a resolution of one microsecond. The
135 +maximum latency is 10,240 microseconds. The data are only valid, if the
136 +overflow register is empty. Every output line contains the latency in
137 +microseconds in the first row and the number of samples in the second
138 +row. To display only lines with a positive latency count, use, for
141 +grep -v " 0$" /sys/kernel/debug/tracing/latency_hist/preemptoff/CPU0
143 +#Minimum latency: 0 microseconds.
144 +#Average latency: 0 microseconds.
145 +#Maximum latency: 25 microseconds.
146 +#Total samples: 3104770694
147 +#There are 0 samples greater or equal than 10240 microseconds
176 +* Wakeup latency of a selected process
178 +To only collect wakeup latency data of a particular process, write the
179 +PID of the requested process to
181 +/sys/kernel/debug/tracing/latency_hist/wakeup/pid
183 +PIDs are not considered, if this variable is set to 0.
186 +* Details of the process with the highest wakeup latency so far
188 +Selected data of the process that suffered from the highest wakeup
189 +latency that occurred in a particular CPU are available in the file
191 +/sys/kernel/debug/tracing/latency_hist/wakeup/max_latency-CPUx.
193 +In addition, other relevant system data at the time when the
194 +latency occurred are given.
196 +The format of the data is (all in one line):
197 +<PID> <Priority> <Latency> (<Timeroffset>) <Command> \
198 +<- <PID> <Priority> <Command> <Timestamp>
200 +The value of <Timeroffset> is only relevant in the combined timer
201 +and wakeup latency recording. In the wakeup recording, it is
202 +always 0, in the missed_timer_offsets recording, it is the same
205 +When retrospectively searching for the origin of a latency and
206 +tracing was not enabled, it may be helpful to know the name and
207 +some basic data of the task that (finally) was switching to the
208 +late real-tlme task. In addition to the victim's data, also the
209 +data of the possible culprit are therefore displayed after the
212 +Finally, the timestamp of the time when the latency occurred
213 +in <seconds>.<microseconds> after the most recent system boot
216 +These data are also reset when the wakeup histogram is reset.
217 diff --git a/arch/Kconfig b/arch/Kconfig
218 index 659bdd079277..099fc0f5155e 100644
221 @@ -9,6 +9,7 @@ config OPROFILE
222 tristate "OProfile system profiling"
224 depends on HAVE_OPROFILE
225 + depends on !PREEMPT_RT_FULL
227 select RING_BUFFER_ALLOW_SWAP
229 @@ -52,6 +53,7 @@ config KPROBES
231 bool "Optimize very unlikely/likely branches"
232 depends on HAVE_ARCH_JUMP_LABEL
233 + depends on (!INTERRUPT_OFF_HIST && !PREEMPT_OFF_HIST && !WAKEUP_LATENCY_HIST && !MISSED_TIMER_OFFSETS_HIST)
235 This option enables a transparent branch optimization that
236 makes certain almost-always-true or almost-always-false branch
237 diff --git a/arch/arm/Kconfig b/arch/arm/Kconfig
238 index b5d529fdffab..5715844e83e3 100644
239 --- a/arch/arm/Kconfig
240 +++ b/arch/arm/Kconfig
241 @@ -36,7 +36,7 @@ config ARM
242 select HAVE_ARCH_AUDITSYSCALL if (AEABI && !OABI_COMPAT)
243 select HAVE_ARCH_BITREVERSE if (CPU_32v7M || CPU_32v7) && !CPU_32v6
244 select HAVE_ARCH_HARDENED_USERCOPY
245 - select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32 && MMU
246 + select HAVE_ARCH_JUMP_LABEL if !XIP_KERNEL && !CPU_ENDIAN_BE32 && MMU && !PREEMPT_RT_BASE
247 select HAVE_ARCH_KGDB if !CPU_ENDIAN_BE32 && MMU
248 select HAVE_ARCH_MMAP_RND_BITS if MMU
249 select HAVE_ARCH_SECCOMP_FILTER if (AEABI && !OABI_COMPAT)
250 @@ -75,6 +75,7 @@ config ARM
251 select HAVE_PERF_EVENTS
252 select HAVE_PERF_REGS
253 select HAVE_PERF_USER_STACK_DUMP
254 + select HAVE_PREEMPT_LAZY
255 select HAVE_RCU_TABLE_FREE if (SMP && ARM_LPAE)
256 select HAVE_REGS_AND_STACK_ACCESS_API
257 select HAVE_SYSCALL_TRACEPOINTS
258 diff --git a/arch/arm/include/asm/irq.h b/arch/arm/include/asm/irq.h
259 index e53638c8ed8a..6095a1649865 100644
260 --- a/arch/arm/include/asm/irq.h
261 +++ b/arch/arm/include/asm/irq.h
266 +#include <linux/cpumask.h>
270 extern void migrate_irqs(void);
271 diff --git a/arch/arm/include/asm/switch_to.h b/arch/arm/include/asm/switch_to.h
272 index 12ebfcc1d539..c962084605bc 100644
273 --- a/arch/arm/include/asm/switch_to.h
274 +++ b/arch/arm/include/asm/switch_to.h
277 #include <linux/thread_info.h>
279 +#if defined CONFIG_PREEMPT_RT_FULL && defined CONFIG_HIGHMEM
280 +void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p);
283 +switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) { }
287 * For v7 SMP cores running a preemptible kernel we may be pre-empted
288 * during a TLB maintenance operation, so execute an inner-shareable dsb
289 @@ -25,6 +32,7 @@ extern struct task_struct *__switch_to(struct task_struct *, struct thread_info
290 #define switch_to(prev,next,last) \
292 __complete_pending_tlbi(); \
293 + switch_kmaps(prev, next); \
294 last = __switch_to(prev,task_thread_info(prev), task_thread_info(next)); \
297 diff --git a/arch/arm/include/asm/thread_info.h b/arch/arm/include/asm/thread_info.h
298 index 776757d1604a..1f36a4eccc72 100644
299 --- a/arch/arm/include/asm/thread_info.h
300 +++ b/arch/arm/include/asm/thread_info.h
301 @@ -49,6 +49,7 @@ struct cpu_context_save {
303 unsigned long flags; /* low level flags */
304 int preempt_count; /* 0 => preemptable, <0 => bug */
305 + int preempt_lazy_count; /* 0 => preemptable, <0 => bug */
306 mm_segment_t addr_limit; /* address limit */
307 struct task_struct *task; /* main task structure */
309 @@ -142,7 +143,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
310 #define TIF_SYSCALL_TRACE 4 /* syscall trace active */
311 #define TIF_SYSCALL_AUDIT 5 /* syscall auditing active */
312 #define TIF_SYSCALL_TRACEPOINT 6 /* syscall tracepoint instrumentation */
313 -#define TIF_SECCOMP 7 /* seccomp syscall filtering active */
314 +#define TIF_SECCOMP 8 /* seccomp syscall filtering active */
315 +#define TIF_NEED_RESCHED_LAZY 7
317 #define TIF_NOHZ 12 /* in adaptive nohz mode */
318 #define TIF_USING_IWMMXT 17
319 @@ -152,6 +154,7 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
320 #define _TIF_SIGPENDING (1 << TIF_SIGPENDING)
321 #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED)
322 #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME)
323 +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY)
324 #define _TIF_UPROBE (1 << TIF_UPROBE)
325 #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE)
326 #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
327 @@ -167,7 +170,8 @@ extern int vfp_restore_user_hwstate(struct user_vfp __user *,
328 * Change these and you break ASM code in entry-common.S
330 #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \
331 - _TIF_NOTIFY_RESUME | _TIF_UPROBE)
332 + _TIF_NOTIFY_RESUME | _TIF_UPROBE | \
333 + _TIF_NEED_RESCHED_LAZY)
335 #endif /* __KERNEL__ */
336 #endif /* __ASM_ARM_THREAD_INFO_H */
337 diff --git a/arch/arm/kernel/asm-offsets.c b/arch/arm/kernel/asm-offsets.c
338 index 608008229c7d..3866da3f7bb7 100644
339 --- a/arch/arm/kernel/asm-offsets.c
340 +++ b/arch/arm/kernel/asm-offsets.c
341 @@ -65,6 +65,7 @@ int main(void)
343 DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
344 DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count));
345 + DEFINE(TI_PREEMPT_LAZY, offsetof(struct thread_info, preempt_lazy_count));
346 DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit));
347 DEFINE(TI_TASK, offsetof(struct thread_info, task));
348 DEFINE(TI_CPU, offsetof(struct thread_info, cpu));
349 diff --git a/arch/arm/kernel/entry-armv.S b/arch/arm/kernel/entry-armv.S
350 index 9f157e7c51e7..468e224d76aa 100644
351 --- a/arch/arm/kernel/entry-armv.S
352 +++ b/arch/arm/kernel/entry-armv.S
353 @@ -220,11 +220,18 @@ ENDPROC(__dabt_svc)
355 #ifdef CONFIG_PREEMPT
356 ldr r8, [tsk, #TI_PREEMPT] @ get preempt count
357 - ldr r0, [tsk, #TI_FLAGS] @ get flags
358 teq r8, #0 @ if preempt count != 0
359 + bne 1f @ return from exeption
360 + ldr r0, [tsk, #TI_FLAGS] @ get flags
361 + tst r0, #_TIF_NEED_RESCHED @ if NEED_RESCHED is set
362 + blne svc_preempt @ preempt!
364 + ldr r8, [tsk, #TI_PREEMPT_LAZY] @ get preempt lazy count
365 + teq r8, #0 @ if preempt lazy count != 0
366 movne r0, #0 @ force flags to 0
367 - tst r0, #_TIF_NEED_RESCHED
368 + tst r0, #_TIF_NEED_RESCHED_LAZY
373 svc_exit r5, irq = 1 @ return from exception
374 @@ -239,8 +246,14 @@ ENDPROC(__irq_svc)
375 1: bl preempt_schedule_irq @ irq en/disable is done inside
376 ldr r0, [tsk, #TI_FLAGS] @ get new tasks TI_FLAGS
377 tst r0, #_TIF_NEED_RESCHED
379 + tst r0, #_TIF_NEED_RESCHED_LAZY
382 + ldr r0, [tsk, #TI_PREEMPT_LAZY] @ get preempt lazy count
383 + teq r0, #0 @ if preempt lazy count != 0
390 diff --git a/arch/arm/kernel/entry-common.S b/arch/arm/kernel/entry-common.S
391 index 10c3283d6c19..8872937862cc 100644
392 --- a/arch/arm/kernel/entry-common.S
393 +++ b/arch/arm/kernel/entry-common.S
396 disable_irq_notrace @ disable interrupts
397 ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing
398 - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK
399 + tst r1, #((_TIF_SYSCALL_WORK | _TIF_WORK_MASK) & ~_TIF_SECCOMP)
400 + bne fast_work_pending
401 + tst r1, #_TIF_SECCOMP
402 bne fast_work_pending
404 /* perform architecture specific actions before user return */
405 @@ -62,8 +64,11 @@ ENDPROC(ret_fast_syscall)
406 str r0, [sp, #S_R0 + S_OFF]! @ save returned r0
407 disable_irq_notrace @ disable interrupts
408 ldr r1, [tsk, #TI_FLAGS] @ re-check for syscall tracing
409 - tst r1, #_TIF_SYSCALL_WORK | _TIF_WORK_MASK
410 + tst r1, #((_TIF_SYSCALL_WORK | _TIF_WORK_MASK) & ~_TIF_SECCOMP)
412 + tst r1, #_TIF_SECCOMP
416 ENDPROC(ret_fast_syscall)
418 diff --git a/arch/arm/kernel/patch.c b/arch/arm/kernel/patch.c
419 index 69bda1a5707e..1f665acaa6a9 100644
420 --- a/arch/arm/kernel/patch.c
421 +++ b/arch/arm/kernel/patch.c
422 @@ -15,7 +15,7 @@ struct patch {
426 -static DEFINE_SPINLOCK(patch_lock);
427 +static DEFINE_RAW_SPINLOCK(patch_lock);
429 static void __kprobes *patch_map(void *addr, int fixmap, unsigned long *flags)
430 __acquires(&patch_lock)
431 @@ -32,7 +32,7 @@ static void __kprobes *patch_map(void *addr, int fixmap, unsigned long *flags)
435 - spin_lock_irqsave(&patch_lock, *flags);
436 + raw_spin_lock_irqsave(&patch_lock, *flags);
438 __acquire(&patch_lock);
440 @@ -47,7 +47,7 @@ static void __kprobes patch_unmap(int fixmap, unsigned long *flags)
441 clear_fixmap(fixmap);
444 - spin_unlock_irqrestore(&patch_lock, *flags);
445 + raw_spin_unlock_irqrestore(&patch_lock, *flags);
447 __release(&patch_lock);
449 diff --git a/arch/arm/kernel/process.c b/arch/arm/kernel/process.c
450 index 91d2d5b01414..750550098b59 100644
451 --- a/arch/arm/kernel/process.c
452 +++ b/arch/arm/kernel/process.c
453 @@ -322,6 +322,30 @@ unsigned long arch_randomize_brk(struct mm_struct *mm)
458 + * CONFIG_SPLIT_PTLOCK_CPUS results in a page->ptl lock. If the lock is not
459 + * initialized by pgtable_page_ctor() then a coredump of the vector page will
462 +static int __init vectors_user_mapping_init_page(void)
465 + unsigned long addr = 0xffff0000;
470 + pgd = pgd_offset_k(addr);
471 + pud = pud_offset(pgd, addr);
472 + pmd = pmd_offset(pud, addr);
473 + page = pmd_page(*(pmd));
475 + pgtable_page_ctor(page);
479 +late_initcall(vectors_user_mapping_init_page);
481 #ifdef CONFIG_KUSER_HELPERS
483 * The vectors page is always readable from user space for the
484 diff --git a/arch/arm/kernel/signal.c b/arch/arm/kernel/signal.c
485 index 7b8f2141427b..96541e00b74a 100644
486 --- a/arch/arm/kernel/signal.c
487 +++ b/arch/arm/kernel/signal.c
488 @@ -572,7 +572,8 @@ do_work_pending(struct pt_regs *regs, unsigned int thread_flags, int syscall)
490 trace_hardirqs_off();
492 - if (likely(thread_flags & _TIF_NEED_RESCHED)) {
493 + if (likely(thread_flags & (_TIF_NEED_RESCHED |
494 + _TIF_NEED_RESCHED_LAZY))) {
497 if (unlikely(!user_mode(regs)))
498 diff --git a/arch/arm/kernel/smp.c b/arch/arm/kernel/smp.c
499 index 7dd14e8395e6..4cd7e3d98035 100644
500 --- a/arch/arm/kernel/smp.c
501 +++ b/arch/arm/kernel/smp.c
502 @@ -234,8 +234,6 @@ int __cpu_disable(void)
504 local_flush_tlb_all();
506 - clear_tasks_mm_cpumask(cpu);
511 @@ -251,6 +249,9 @@ void __cpu_die(unsigned int cpu)
512 pr_err("CPU%u: cpu didn't die\n", cpu);
516 + clear_tasks_mm_cpumask(cpu);
518 pr_notice("CPU%u: shutdown\n", cpu);
521 diff --git a/arch/arm/kernel/unwind.c b/arch/arm/kernel/unwind.c
522 index 0bee233fef9a..314cfb232a63 100644
523 --- a/arch/arm/kernel/unwind.c
524 +++ b/arch/arm/kernel/unwind.c
525 @@ -93,7 +93,7 @@ extern const struct unwind_idx __start_unwind_idx[];
526 static const struct unwind_idx *__origin_unwind_idx;
527 extern const struct unwind_idx __stop_unwind_idx[];
529 -static DEFINE_SPINLOCK(unwind_lock);
530 +static DEFINE_RAW_SPINLOCK(unwind_lock);
531 static LIST_HEAD(unwind_tables);
533 /* Convert a prel31 symbol to an absolute address */
534 @@ -201,7 +201,7 @@ static const struct unwind_idx *unwind_find_idx(unsigned long addr)
535 /* module unwind tables */
536 struct unwind_table *table;
538 - spin_lock_irqsave(&unwind_lock, flags);
539 + raw_spin_lock_irqsave(&unwind_lock, flags);
540 list_for_each_entry(table, &unwind_tables, list) {
541 if (addr >= table->begin_addr &&
542 addr < table->end_addr) {
543 @@ -213,7 +213,7 @@ static const struct unwind_idx *unwind_find_idx(unsigned long addr)
547 - spin_unlock_irqrestore(&unwind_lock, flags);
548 + raw_spin_unlock_irqrestore(&unwind_lock, flags);
551 pr_debug("%s: idx = %p\n", __func__, idx);
552 @@ -529,9 +529,9 @@ struct unwind_table *unwind_table_add(unsigned long start, unsigned long size,
553 tab->begin_addr = text_addr;
554 tab->end_addr = text_addr + text_size;
556 - spin_lock_irqsave(&unwind_lock, flags);
557 + raw_spin_lock_irqsave(&unwind_lock, flags);
558 list_add_tail(&tab->list, &unwind_tables);
559 - spin_unlock_irqrestore(&unwind_lock, flags);
560 + raw_spin_unlock_irqrestore(&unwind_lock, flags);
564 @@ -543,9 +543,9 @@ void unwind_table_del(struct unwind_table *tab)
568 - spin_lock_irqsave(&unwind_lock, flags);
569 + raw_spin_lock_irqsave(&unwind_lock, flags);
570 list_del(&tab->list);
571 - spin_unlock_irqrestore(&unwind_lock, flags);
572 + raw_spin_unlock_irqrestore(&unwind_lock, flags);
576 diff --git a/arch/arm/kvm/arm.c b/arch/arm/kvm/arm.c
577 index 19b5f5c1c0ff..82aa639e6737 100644
578 --- a/arch/arm/kvm/arm.c
579 +++ b/arch/arm/kvm/arm.c
580 @@ -619,7 +619,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
581 * involves poking the GIC, which must be done in a
582 * non-preemptible context.
586 kvm_pmu_flush_hwstate(vcpu);
587 kvm_timer_flush_hwstate(vcpu);
588 kvm_vgic_flush_hwstate(vcpu);
589 @@ -640,7 +640,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
590 kvm_pmu_sync_hwstate(vcpu);
591 kvm_timer_sync_hwstate(vcpu);
592 kvm_vgic_sync_hwstate(vcpu);
598 @@ -696,7 +696,7 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run)
600 kvm_vgic_sync_hwstate(vcpu);
605 ret = handle_exit(vcpu, run, ret);
607 diff --git a/arch/arm/mach-exynos/platsmp.c b/arch/arm/mach-exynos/platsmp.c
608 index 98ffe1e62ad5..df9769ddece5 100644
609 --- a/arch/arm/mach-exynos/platsmp.c
610 +++ b/arch/arm/mach-exynos/platsmp.c
611 @@ -229,7 +229,7 @@ static void __iomem *scu_base_addr(void)
612 return (void __iomem *)(S5P_VA_SCU);
615 -static DEFINE_SPINLOCK(boot_lock);
616 +static DEFINE_RAW_SPINLOCK(boot_lock);
618 static void exynos_secondary_init(unsigned int cpu)
620 @@ -242,8 +242,8 @@ static void exynos_secondary_init(unsigned int cpu)
622 * Synchronise with the boot thread.
624 - spin_lock(&boot_lock);
625 - spin_unlock(&boot_lock);
626 + raw_spin_lock(&boot_lock);
627 + raw_spin_unlock(&boot_lock);
630 int exynos_set_boot_addr(u32 core_id, unsigned long boot_addr)
631 @@ -307,7 +307,7 @@ static int exynos_boot_secondary(unsigned int cpu, struct task_struct *idle)
632 * Set synchronisation state between this boot processor
633 * and the secondary one
635 - spin_lock(&boot_lock);
636 + raw_spin_lock(&boot_lock);
639 * The secondary processor is waiting to be released from
640 @@ -334,7 +334,7 @@ static int exynos_boot_secondary(unsigned int cpu, struct task_struct *idle)
643 printk(KERN_ERR "cpu1 power enable failed");
644 - spin_unlock(&boot_lock);
645 + raw_spin_unlock(&boot_lock);
649 @@ -380,7 +380,7 @@ static int exynos_boot_secondary(unsigned int cpu, struct task_struct *idle)
650 * calibrations, then wait for it to finish
653 - spin_unlock(&boot_lock);
654 + raw_spin_unlock(&boot_lock);
656 return pen_release != -1 ? ret : 0;
658 diff --git a/arch/arm/mach-hisi/platmcpm.c b/arch/arm/mach-hisi/platmcpm.c
659 index 4b653a8cb75c..b03d5a922cb1 100644
660 --- a/arch/arm/mach-hisi/platmcpm.c
661 +++ b/arch/arm/mach-hisi/platmcpm.c
664 static void __iomem *sysctrl, *fabric;
665 static int hip04_cpu_table[HIP04_MAX_CLUSTERS][HIP04_MAX_CPUS_PER_CLUSTER];
666 -static DEFINE_SPINLOCK(boot_lock);
667 +static DEFINE_RAW_SPINLOCK(boot_lock);
668 static u32 fabric_phys_addr;
670 * [0]: bootwrapper physical address
671 @@ -113,7 +113,7 @@ static int hip04_boot_secondary(unsigned int l_cpu, struct task_struct *idle)
672 if (cluster >= HIP04_MAX_CLUSTERS || cpu >= HIP04_MAX_CPUS_PER_CLUSTER)
675 - spin_lock_irq(&boot_lock);
676 + raw_spin_lock_irq(&boot_lock);
678 if (hip04_cpu_table[cluster][cpu])
680 @@ -147,7 +147,7 @@ static int hip04_boot_secondary(unsigned int l_cpu, struct task_struct *idle)
683 hip04_cpu_table[cluster][cpu]++;
684 - spin_unlock_irq(&boot_lock);
685 + raw_spin_unlock_irq(&boot_lock);
689 @@ -162,11 +162,11 @@ static void hip04_cpu_die(unsigned int l_cpu)
690 cpu = MPIDR_AFFINITY_LEVEL(mpidr, 0);
691 cluster = MPIDR_AFFINITY_LEVEL(mpidr, 1);
693 - spin_lock(&boot_lock);
694 + raw_spin_lock(&boot_lock);
695 hip04_cpu_table[cluster][cpu]--;
696 if (hip04_cpu_table[cluster][cpu] == 1) {
697 /* A power_up request went ahead of us. */
698 - spin_unlock(&boot_lock);
699 + raw_spin_unlock(&boot_lock);
701 } else if (hip04_cpu_table[cluster][cpu] > 1) {
702 pr_err("Cluster %d CPU%d boots multiple times\n", cluster, cpu);
703 @@ -174,7 +174,7 @@ static void hip04_cpu_die(unsigned int l_cpu)
706 last_man = hip04_cluster_is_down(cluster);
707 - spin_unlock(&boot_lock);
708 + raw_spin_unlock(&boot_lock);
710 /* Since it's Cortex A15, disable L2 prefetching. */
712 @@ -203,7 +203,7 @@ static int hip04_cpu_kill(unsigned int l_cpu)
713 cpu >= HIP04_MAX_CPUS_PER_CLUSTER);
715 count = TIMEOUT_MSEC / POLL_MSEC;
716 - spin_lock_irq(&boot_lock);
717 + raw_spin_lock_irq(&boot_lock);
718 for (tries = 0; tries < count; tries++) {
719 if (hip04_cpu_table[cluster][cpu])
721 @@ -211,10 +211,10 @@ static int hip04_cpu_kill(unsigned int l_cpu)
722 data = readl_relaxed(sysctrl + SC_CPU_RESET_STATUS(cluster));
723 if (data & CORE_WFI_STATUS(cpu))
725 - spin_unlock_irq(&boot_lock);
726 + raw_spin_unlock_irq(&boot_lock);
727 /* Wait for clean L2 when the whole cluster is down. */
729 - spin_lock_irq(&boot_lock);
730 + raw_spin_lock_irq(&boot_lock);
734 @@ -231,10 +231,10 @@ static int hip04_cpu_kill(unsigned int l_cpu)
736 if (hip04_cluster_is_down(cluster))
737 hip04_set_snoop_filter(cluster, 0);
738 - spin_unlock_irq(&boot_lock);
739 + raw_spin_unlock_irq(&boot_lock);
742 - spin_unlock_irq(&boot_lock);
743 + raw_spin_unlock_irq(&boot_lock);
747 diff --git a/arch/arm/mach-omap2/omap-smp.c b/arch/arm/mach-omap2/omap-smp.c
748 index b4de3da6dffa..b52893319d75 100644
749 --- a/arch/arm/mach-omap2/omap-smp.c
750 +++ b/arch/arm/mach-omap2/omap-smp.c
751 @@ -64,7 +64,7 @@ static const struct omap_smp_config omap5_cfg __initconst = {
752 .startup_addr = omap5_secondary_startup,
755 -static DEFINE_SPINLOCK(boot_lock);
756 +static DEFINE_RAW_SPINLOCK(boot_lock);
758 void __iomem *omap4_get_scu_base(void)
760 @@ -131,8 +131,8 @@ static void omap4_secondary_init(unsigned int cpu)
762 * Synchronise with the boot thread.
764 - spin_lock(&boot_lock);
765 - spin_unlock(&boot_lock);
766 + raw_spin_lock(&boot_lock);
767 + raw_spin_unlock(&boot_lock);
770 static int omap4_boot_secondary(unsigned int cpu, struct task_struct *idle)
771 @@ -146,7 +146,7 @@ static int omap4_boot_secondary(unsigned int cpu, struct task_struct *idle)
772 * Set synchronisation state between this boot processor
773 * and the secondary one
775 - spin_lock(&boot_lock);
776 + raw_spin_lock(&boot_lock);
779 * Update the AuxCoreBoot0 with boot state for secondary core.
780 @@ -223,7 +223,7 @@ static int omap4_boot_secondary(unsigned int cpu, struct task_struct *idle)
781 * Now the secondary core is starting up let it run its
782 * calibrations, then wait for it to finish
784 - spin_unlock(&boot_lock);
785 + raw_spin_unlock(&boot_lock);
789 diff --git a/arch/arm/mach-prima2/platsmp.c b/arch/arm/mach-prima2/platsmp.c
790 index 0875b99add18..18b6d98d2581 100644
791 --- a/arch/arm/mach-prima2/platsmp.c
792 +++ b/arch/arm/mach-prima2/platsmp.c
795 static void __iomem *clk_base;
797 -static DEFINE_SPINLOCK(boot_lock);
798 +static DEFINE_RAW_SPINLOCK(boot_lock);
800 static void sirfsoc_secondary_init(unsigned int cpu)
802 @@ -36,8 +36,8 @@ static void sirfsoc_secondary_init(unsigned int cpu)
804 * Synchronise with the boot thread.
806 - spin_lock(&boot_lock);
807 - spin_unlock(&boot_lock);
808 + raw_spin_lock(&boot_lock);
809 + raw_spin_unlock(&boot_lock);
812 static const struct of_device_id clk_ids[] = {
813 @@ -75,7 +75,7 @@ static int sirfsoc_boot_secondary(unsigned int cpu, struct task_struct *idle)
814 /* make sure write buffer is drained */
817 - spin_lock(&boot_lock);
818 + raw_spin_lock(&boot_lock);
821 * The secondary processor is waiting to be released from
822 @@ -107,7 +107,7 @@ static int sirfsoc_boot_secondary(unsigned int cpu, struct task_struct *idle)
823 * now the secondary core is starting up let it run its
824 * calibrations, then wait for it to finish
826 - spin_unlock(&boot_lock);
827 + raw_spin_unlock(&boot_lock);
829 return pen_release != -1 ? -ENOSYS : 0;
831 diff --git a/arch/arm/mach-qcom/platsmp.c b/arch/arm/mach-qcom/platsmp.c
832 index 5494c9e0c909..e8ce157d3548 100644
833 --- a/arch/arm/mach-qcom/platsmp.c
834 +++ b/arch/arm/mach-qcom/platsmp.c
837 extern void secondary_startup_arm(void);
839 -static DEFINE_SPINLOCK(boot_lock);
840 +static DEFINE_RAW_SPINLOCK(boot_lock);
842 #ifdef CONFIG_HOTPLUG_CPU
843 static void qcom_cpu_die(unsigned int cpu)
844 @@ -60,8 +60,8 @@ static void qcom_secondary_init(unsigned int cpu)
846 * Synchronise with the boot thread.
848 - spin_lock(&boot_lock);
849 - spin_unlock(&boot_lock);
850 + raw_spin_lock(&boot_lock);
851 + raw_spin_unlock(&boot_lock);
854 static int scss_release_secondary(unsigned int cpu)
855 @@ -284,7 +284,7 @@ static int qcom_boot_secondary(unsigned int cpu, int (*func)(unsigned int))
856 * set synchronisation state between this boot processor
857 * and the secondary one
859 - spin_lock(&boot_lock);
860 + raw_spin_lock(&boot_lock);
863 * Send the secondary CPU a soft interrupt, thereby causing
864 @@ -297,7 +297,7 @@ static int qcom_boot_secondary(unsigned int cpu, int (*func)(unsigned int))
865 * now the secondary core is starting up let it run its
866 * calibrations, then wait for it to finish
868 - spin_unlock(&boot_lock);
869 + raw_spin_unlock(&boot_lock);
873 diff --git a/arch/arm/mach-spear/platsmp.c b/arch/arm/mach-spear/platsmp.c
874 index 8d1e2d551786..7fa56cc78118 100644
875 --- a/arch/arm/mach-spear/platsmp.c
876 +++ b/arch/arm/mach-spear/platsmp.c
877 @@ -32,7 +32,7 @@ static void write_pen_release(int val)
878 sync_cache_w(&pen_release);
881 -static DEFINE_SPINLOCK(boot_lock);
882 +static DEFINE_RAW_SPINLOCK(boot_lock);
884 static void __iomem *scu_base = IOMEM(VA_SCU_BASE);
886 @@ -47,8 +47,8 @@ static void spear13xx_secondary_init(unsigned int cpu)
888 * Synchronise with the boot thread.
890 - spin_lock(&boot_lock);
891 - spin_unlock(&boot_lock);
892 + raw_spin_lock(&boot_lock);
893 + raw_spin_unlock(&boot_lock);
896 static int spear13xx_boot_secondary(unsigned int cpu, struct task_struct *idle)
897 @@ -59,7 +59,7 @@ static int spear13xx_boot_secondary(unsigned int cpu, struct task_struct *idle)
898 * set synchronisation state between this boot processor
899 * and the secondary one
901 - spin_lock(&boot_lock);
902 + raw_spin_lock(&boot_lock);
905 * The secondary processor is waiting to be released from
906 @@ -84,7 +84,7 @@ static int spear13xx_boot_secondary(unsigned int cpu, struct task_struct *idle)
907 * now the secondary core is starting up let it run its
908 * calibrations, then wait for it to finish
910 - spin_unlock(&boot_lock);
911 + raw_spin_unlock(&boot_lock);
913 return pen_release != -1 ? -ENOSYS : 0;
915 diff --git a/arch/arm/mach-sti/platsmp.c b/arch/arm/mach-sti/platsmp.c
916 index ea5a2277ee46..b988e081ac79 100644
917 --- a/arch/arm/mach-sti/platsmp.c
918 +++ b/arch/arm/mach-sti/platsmp.c
919 @@ -35,7 +35,7 @@ static void write_pen_release(int val)
920 sync_cache_w(&pen_release);
923 -static DEFINE_SPINLOCK(boot_lock);
924 +static DEFINE_RAW_SPINLOCK(boot_lock);
926 static void sti_secondary_init(unsigned int cpu)
928 @@ -48,8 +48,8 @@ static void sti_secondary_init(unsigned int cpu)
930 * Synchronise with the boot thread.
932 - spin_lock(&boot_lock);
933 - spin_unlock(&boot_lock);
934 + raw_spin_lock(&boot_lock);
935 + raw_spin_unlock(&boot_lock);
938 static int sti_boot_secondary(unsigned int cpu, struct task_struct *idle)
939 @@ -60,7 +60,7 @@ static int sti_boot_secondary(unsigned int cpu, struct task_struct *idle)
940 * set synchronisation state between this boot processor
941 * and the secondary one
943 - spin_lock(&boot_lock);
944 + raw_spin_lock(&boot_lock);
947 * The secondary processor is waiting to be released from
948 @@ -91,7 +91,7 @@ static int sti_boot_secondary(unsigned int cpu, struct task_struct *idle)
949 * now the secondary core is starting up let it run its
950 * calibrations, then wait for it to finish
952 - spin_unlock(&boot_lock);
953 + raw_spin_unlock(&boot_lock);
955 return pen_release != -1 ? -ENOSYS : 0;
957 diff --git a/arch/arm/mm/fault.c b/arch/arm/mm/fault.c
958 index 0122ad1a6027..926b1be48043 100644
959 --- a/arch/arm/mm/fault.c
960 +++ b/arch/arm/mm/fault.c
961 @@ -430,6 +430,9 @@ do_translation_fault(unsigned long addr, unsigned int fsr,
962 if (addr < TASK_SIZE)
963 return do_page_fault(addr, fsr, regs);
965 + if (interrupts_enabled(regs))
966 + local_irq_enable();
971 @@ -497,6 +500,9 @@ do_translation_fault(unsigned long addr, unsigned int fsr,
973 do_sect_fault(unsigned long addr, unsigned int fsr, struct pt_regs *regs)
975 + if (interrupts_enabled(regs))
976 + local_irq_enable();
978 do_bad_area(addr, fsr, regs);
981 diff --git a/arch/arm/mm/highmem.c b/arch/arm/mm/highmem.c
982 index d02f8187b1cc..542692dbd40a 100644
983 --- a/arch/arm/mm/highmem.c
984 +++ b/arch/arm/mm/highmem.c
985 @@ -34,6 +34,11 @@ static inline pte_t get_fixmap_pte(unsigned long vaddr)
989 +static unsigned int fixmap_idx(int type)
991 + return FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id();
994 void *kmap(struct page *page)
997 @@ -54,12 +59,13 @@ EXPORT_SYMBOL(kunmap);
999 void *kmap_atomic(struct page *page)
1001 + pte_t pte = mk_pte(page, kmap_prot);
1003 unsigned long vaddr;
1007 - preempt_disable();
1008 + preempt_disable_nort();
1009 pagefault_disable();
1010 if (!PageHighMem(page))
1011 return page_address(page);
1012 @@ -79,7 +85,7 @@ void *kmap_atomic(struct page *page)
1014 type = kmap_atomic_idx_push();
1016 - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id();
1017 + idx = fixmap_idx(type);
1018 vaddr = __fix_to_virt(idx);
1019 #ifdef CONFIG_DEBUG_HIGHMEM
1021 @@ -93,7 +99,10 @@ void *kmap_atomic(struct page *page)
1022 * in place, so the contained TLB flush ensures the TLB is updated
1023 * with the new mapping.
1025 - set_fixmap_pte(idx, mk_pte(page, kmap_prot));
1026 +#ifdef CONFIG_PREEMPT_RT_FULL
1027 + current->kmap_pte[type] = pte;
1029 + set_fixmap_pte(idx, pte);
1031 return (void *)vaddr;
1033 @@ -106,44 +115,75 @@ void __kunmap_atomic(void *kvaddr)
1035 if (kvaddr >= (void *)FIXADDR_START) {
1036 type = kmap_atomic_idx();
1037 - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id();
1038 + idx = fixmap_idx(type);
1040 if (cache_is_vivt())
1041 __cpuc_flush_dcache_area((void *)vaddr, PAGE_SIZE);
1042 +#ifdef CONFIG_PREEMPT_RT_FULL
1043 + current->kmap_pte[type] = __pte(0);
1045 #ifdef CONFIG_DEBUG_HIGHMEM
1046 BUG_ON(vaddr != __fix_to_virt(idx));
1047 - set_fixmap_pte(idx, __pte(0));
1049 (void) idx; /* to kill a warning */
1051 + set_fixmap_pte(idx, __pte(0));
1052 kmap_atomic_idx_pop();
1053 } else if (vaddr >= PKMAP_ADDR(0) && vaddr < PKMAP_ADDR(LAST_PKMAP)) {
1054 /* this address was obtained through kmap_high_get() */
1055 kunmap_high(pte_page(pkmap_page_table[PKMAP_NR(vaddr)]));
1059 + preempt_enable_nort();
1061 EXPORT_SYMBOL(__kunmap_atomic);
1063 void *kmap_atomic_pfn(unsigned long pfn)
1065 + pte_t pte = pfn_pte(pfn, kmap_prot);
1066 unsigned long vaddr;
1068 struct page *page = pfn_to_page(pfn);
1070 - preempt_disable();
1071 + preempt_disable_nort();
1072 pagefault_disable();
1073 if (!PageHighMem(page))
1074 return page_address(page);
1076 type = kmap_atomic_idx_push();
1077 - idx = FIX_KMAP_BEGIN + type + KM_TYPE_NR * smp_processor_id();
1078 + idx = fixmap_idx(type);
1079 vaddr = __fix_to_virt(idx);
1080 #ifdef CONFIG_DEBUG_HIGHMEM
1081 BUG_ON(!pte_none(get_fixmap_pte(vaddr)));
1083 - set_fixmap_pte(idx, pfn_pte(pfn, kmap_prot));
1084 +#ifdef CONFIG_PREEMPT_RT_FULL
1085 + current->kmap_pte[type] = pte;
1087 + set_fixmap_pte(idx, pte);
1089 return (void *)vaddr;
1091 +#if defined CONFIG_PREEMPT_RT_FULL
1092 +void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p)
1097 + * Clear @prev's kmap_atomic mappings
1099 + for (i = 0; i < prev_p->kmap_idx; i++) {
1100 + int idx = fixmap_idx(i);
1102 + set_fixmap_pte(idx, __pte(0));
1105 + * Restore @next_p's kmap_atomic mappings
1107 + for (i = 0; i < next_p->kmap_idx; i++) {
1108 + int idx = fixmap_idx(i);
1110 + if (!pte_none(next_p->kmap_pte[i]))
1111 + set_fixmap_pte(idx, next_p->kmap_pte[i]);
1115 diff --git a/arch/arm/plat-versatile/platsmp.c b/arch/arm/plat-versatile/platsmp.c
1116 index c2366510187a..6b60f582b738 100644
1117 --- a/arch/arm/plat-versatile/platsmp.c
1118 +++ b/arch/arm/plat-versatile/platsmp.c
1119 @@ -32,7 +32,7 @@ static void write_pen_release(int val)
1120 sync_cache_w(&pen_release);
1123 -static DEFINE_SPINLOCK(boot_lock);
1124 +static DEFINE_RAW_SPINLOCK(boot_lock);
1126 void versatile_secondary_init(unsigned int cpu)
1128 @@ -45,8 +45,8 @@ void versatile_secondary_init(unsigned int cpu)
1130 * Synchronise with the boot thread.
1132 - spin_lock(&boot_lock);
1133 - spin_unlock(&boot_lock);
1134 + raw_spin_lock(&boot_lock);
1135 + raw_spin_unlock(&boot_lock);
1138 int versatile_boot_secondary(unsigned int cpu, struct task_struct *idle)
1139 @@ -57,7 +57,7 @@ int versatile_boot_secondary(unsigned int cpu, struct task_struct *idle)
1140 * Set synchronisation state between this boot processor
1141 * and the secondary one
1143 - spin_lock(&boot_lock);
1144 + raw_spin_lock(&boot_lock);
1147 * This is really belt and braces; we hold unintended secondary
1148 @@ -87,7 +87,7 @@ int versatile_boot_secondary(unsigned int cpu, struct task_struct *idle)
1149 * now the secondary core is starting up let it run its
1150 * calibrations, then wait for it to finish
1152 - spin_unlock(&boot_lock);
1153 + raw_spin_unlock(&boot_lock);
1155 return pen_release != -1 ? -ENOSYS : 0;
1157 diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
1158 index 969ef880d234..1182fe883771 100644
1159 --- a/arch/arm64/Kconfig
1160 +++ b/arch/arm64/Kconfig
1161 @@ -91,6 +91,7 @@ config ARM64
1162 select HAVE_PERF_EVENTS
1163 select HAVE_PERF_REGS
1164 select HAVE_PERF_USER_STACK_DUMP
1165 + select HAVE_PREEMPT_LAZY
1166 select HAVE_REGS_AND_STACK_ACCESS_API
1167 select HAVE_RCU_TABLE_FREE
1168 select HAVE_SYSCALL_TRACEPOINTS
1169 @@ -694,7 +695,7 @@ config XEN_DOM0
1172 bool "Xen guest support on ARM64"
1173 - depends on ARM64 && OF
1174 + depends on ARM64 && OF && !PREEMPT_RT_FULL
1178 diff --git a/arch/arm64/include/asm/thread_info.h b/arch/arm64/include/asm/thread_info.h
1179 index e9ea5a6bd449..6c500ad63c6a 100644
1180 --- a/arch/arm64/include/asm/thread_info.h
1181 +++ b/arch/arm64/include/asm/thread_info.h
1182 @@ -49,6 +49,7 @@ struct thread_info {
1183 mm_segment_t addr_limit; /* address limit */
1184 struct task_struct *task; /* main task structure */
1185 int preempt_count; /* 0 => preemptable, <0 => bug */
1186 + int preempt_lazy_count; /* 0 => preemptable, <0 => bug */
1190 @@ -112,6 +113,7 @@ static inline struct thread_info *current_thread_info(void)
1191 #define TIF_NEED_RESCHED 1
1192 #define TIF_NOTIFY_RESUME 2 /* callback before returning to user */
1193 #define TIF_FOREIGN_FPSTATE 3 /* CPU's FP state is not current's */
1194 +#define TIF_NEED_RESCHED_LAZY 4
1196 #define TIF_SYSCALL_TRACE 8
1197 #define TIF_SYSCALL_AUDIT 9
1198 @@ -127,6 +129,7 @@ static inline struct thread_info *current_thread_info(void)
1199 #define _TIF_NEED_RESCHED (1 << TIF_NEED_RESCHED)
1200 #define _TIF_NOTIFY_RESUME (1 << TIF_NOTIFY_RESUME)
1201 #define _TIF_FOREIGN_FPSTATE (1 << TIF_FOREIGN_FPSTATE)
1202 +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY)
1203 #define _TIF_NOHZ (1 << TIF_NOHZ)
1204 #define _TIF_SYSCALL_TRACE (1 << TIF_SYSCALL_TRACE)
1205 #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
1206 @@ -135,7 +138,9 @@ static inline struct thread_info *current_thread_info(void)
1207 #define _TIF_32BIT (1 << TIF_32BIT)
1209 #define _TIF_WORK_MASK (_TIF_NEED_RESCHED | _TIF_SIGPENDING | \
1210 - _TIF_NOTIFY_RESUME | _TIF_FOREIGN_FPSTATE)
1211 + _TIF_NOTIFY_RESUME | _TIF_FOREIGN_FPSTATE | \
1212 + _TIF_NEED_RESCHED_LAZY)
1213 +#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY)
1215 #define _TIF_SYSCALL_WORK (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \
1216 _TIF_SYSCALL_TRACEPOINT | _TIF_SECCOMP | \
1217 diff --git a/arch/arm64/kernel/asm-offsets.c b/arch/arm64/kernel/asm-offsets.c
1218 index 4a2f0f0fef32..6bf2bc17c400 100644
1219 --- a/arch/arm64/kernel/asm-offsets.c
1220 +++ b/arch/arm64/kernel/asm-offsets.c
1221 @@ -38,6 +38,7 @@ int main(void)
1223 DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
1224 DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count));
1225 + DEFINE(TI_PREEMPT_LAZY, offsetof(struct thread_info, preempt_lazy_count));
1226 DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit));
1227 DEFINE(TI_TASK, offsetof(struct thread_info, task));
1228 DEFINE(TI_CPU, offsetof(struct thread_info, cpu));
1229 diff --git a/arch/arm64/kernel/entry.S b/arch/arm64/kernel/entry.S
1230 index 79b0fe24d5b7..f3c959ade308 100644
1231 --- a/arch/arm64/kernel/entry.S
1232 +++ b/arch/arm64/kernel/entry.S
1233 @@ -428,11 +428,16 @@ ENDPROC(el1_sync)
1235 #ifdef CONFIG_PREEMPT
1236 ldr w24, [tsk, #TI_PREEMPT] // get preempt count
1237 - cbnz w24, 1f // preempt count != 0
1238 + cbnz w24, 2f // preempt count != 0
1239 ldr x0, [tsk, #TI_FLAGS] // get flags
1240 - tbz x0, #TIF_NEED_RESCHED, 1f // needs rescheduling?
1242 + tbnz x0, #TIF_NEED_RESCHED, 1f // needs rescheduling?
1244 + ldr w24, [tsk, #TI_PREEMPT_LAZY] // get preempt lazy count
1245 + cbnz w24, 2f // preempt lazy count != 0
1246 + tbz x0, #TIF_NEED_RESCHED_LAZY, 2f // needs rescheduling?
1251 #ifdef CONFIG_TRACE_IRQFLAGS
1252 bl trace_hardirqs_on
1253 @@ -446,6 +451,7 @@ ENDPROC(el1_irq)
1254 1: bl preempt_schedule_irq // irq en/disable is done inside
1255 ldr x0, [tsk, #TI_FLAGS] // get new tasks TI_FLAGS
1256 tbnz x0, #TIF_NEED_RESCHED, 1b // needs rescheduling?
1257 + tbnz x0, #TIF_NEED_RESCHED_LAZY, 1b // needs rescheduling?
1261 diff --git a/arch/arm64/kernel/signal.c b/arch/arm64/kernel/signal.c
1262 index 404dd67080b9..639dc6d12e72 100644
1263 --- a/arch/arm64/kernel/signal.c
1264 +++ b/arch/arm64/kernel/signal.c
1265 @@ -409,7 +409,7 @@ asmlinkage void do_notify_resume(struct pt_regs *regs,
1267 trace_hardirqs_off();
1269 - if (thread_flags & _TIF_NEED_RESCHED) {
1270 + if (thread_flags & _TIF_NEED_RESCHED_MASK) {
1274 diff --git a/arch/mips/Kconfig b/arch/mips/Kconfig
1275 index b3c5bde43d34..8122bf058de0 100644
1276 --- a/arch/mips/Kconfig
1277 +++ b/arch/mips/Kconfig
1278 @@ -2514,7 +2514,7 @@ config MIPS_ASID_BITS_VARIABLE
1281 bool "High Memory Support"
1282 - depends on 32BIT && CPU_SUPPORTS_HIGHMEM && SYS_SUPPORTS_HIGHMEM && !CPU_MIPS32_3_5_EVA
1283 + depends on 32BIT && CPU_SUPPORTS_HIGHMEM && SYS_SUPPORTS_HIGHMEM && !CPU_MIPS32_3_5_EVA && !PREEMPT_RT_FULL
1285 config CPU_SUPPORTS_HIGHMEM
1287 diff --git a/arch/powerpc/Kconfig b/arch/powerpc/Kconfig
1288 index 65fba4c34cd7..4b5ba68910e0 100644
1289 --- a/arch/powerpc/Kconfig
1290 +++ b/arch/powerpc/Kconfig
1291 @@ -52,10 +52,11 @@ config LOCKDEP_SUPPORT
1293 config RWSEM_GENERIC_SPINLOCK
1295 + default y if PREEMPT_RT_FULL
1297 config RWSEM_XCHGADD_ALGORITHM
1300 + default y if !PREEMPT_RT_FULL
1302 config GENERIC_LOCKBREAK
1304 @@ -134,6 +135,7 @@ config PPC
1305 select ARCH_HAS_TICK_BROADCAST if GENERIC_CLOCKEVENTS_BROADCAST
1306 select GENERIC_STRNCPY_FROM_USER
1307 select GENERIC_STRNLEN_USER
1308 + select HAVE_PREEMPT_LAZY
1309 select HAVE_MOD_ARCH_SPECIFIC
1310 select MODULES_USE_ELF_RELA
1311 select CLONE_BACKWARDS
1312 @@ -321,7 +323,7 @@ menu "Kernel options"
1315 bool "High memory support"
1317 + depends on PPC32 && !PREEMPT_RT_FULL
1319 source kernel/Kconfig.hz
1320 source kernel/Kconfig.preempt
1321 diff --git a/arch/powerpc/include/asm/thread_info.h b/arch/powerpc/include/asm/thread_info.h
1322 index 87e4b2d8dcd4..981e501a4359 100644
1323 --- a/arch/powerpc/include/asm/thread_info.h
1324 +++ b/arch/powerpc/include/asm/thread_info.h
1325 @@ -43,6 +43,8 @@ struct thread_info {
1326 int cpu; /* cpu we're on */
1327 int preempt_count; /* 0 => preemptable,
1329 + int preempt_lazy_count; /* 0 => preemptable,
1331 unsigned long local_flags; /* private flags for thread */
1332 #ifdef CONFIG_LIVEPATCH
1333 unsigned long *livepatch_sp;
1334 @@ -88,8 +90,7 @@ static inline struct thread_info *current_thread_info(void)
1335 #define TIF_SYSCALL_TRACE 0 /* syscall trace active */
1336 #define TIF_SIGPENDING 1 /* signal pending */
1337 #define TIF_NEED_RESCHED 2 /* rescheduling necessary */
1338 -#define TIF_POLLING_NRFLAG 3 /* true if poll_idle() is polling
1339 - TIF_NEED_RESCHED */
1340 +#define TIF_NEED_RESCHED_LAZY 3 /* lazy rescheduling necessary */
1341 #define TIF_32BIT 4 /* 32 bit binary */
1342 #define TIF_RESTORE_TM 5 /* need to restore TM FP/VEC/VSX */
1343 #define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */
1344 @@ -107,6 +108,8 @@ static inline struct thread_info *current_thread_info(void)
1345 #if defined(CONFIG_PPC64)
1346 #define TIF_ELF2ABI 18 /* function descriptors must die! */
1348 +#define TIF_POLLING_NRFLAG 19 /* true if poll_idle() is polling
1349 + TIF_NEED_RESCHED */
1351 /* as above, but as bit values */
1352 #define _TIF_SYSCALL_TRACE (1<<TIF_SYSCALL_TRACE)
1353 @@ -125,14 +128,16 @@ static inline struct thread_info *current_thread_info(void)
1354 #define _TIF_SYSCALL_TRACEPOINT (1<<TIF_SYSCALL_TRACEPOINT)
1355 #define _TIF_EMULATE_STACK_STORE (1<<TIF_EMULATE_STACK_STORE)
1356 #define _TIF_NOHZ (1<<TIF_NOHZ)
1357 +#define _TIF_NEED_RESCHED_LAZY (1<<TIF_NEED_RESCHED_LAZY)
1358 #define _TIF_SYSCALL_DOTRACE (_TIF_SYSCALL_TRACE | _TIF_SYSCALL_AUDIT | \
1359 _TIF_SECCOMP | _TIF_SYSCALL_TRACEPOINT | \
1362 #define _TIF_USER_WORK_MASK (_TIF_SIGPENDING | _TIF_NEED_RESCHED | \
1363 _TIF_NOTIFY_RESUME | _TIF_UPROBE | \
1365 + _TIF_RESTORE_TM | _TIF_NEED_RESCHED_LAZY)
1366 #define _TIF_PERSYSCALL_MASK (_TIF_RESTOREALL|_TIF_NOERROR)
1367 +#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY)
1369 /* Bits in local_flags */
1370 /* Don't move TLF_NAPPING without adjusting the code in entry_32.S */
1371 diff --git a/arch/powerpc/kernel/asm-offsets.c b/arch/powerpc/kernel/asm-offsets.c
1372 index c833d88c423d..96e9fbc3f684 100644
1373 --- a/arch/powerpc/kernel/asm-offsets.c
1374 +++ b/arch/powerpc/kernel/asm-offsets.c
1375 @@ -156,6 +156,7 @@ int main(void)
1376 DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
1377 DEFINE(TI_LOCAL_FLAGS, offsetof(struct thread_info, local_flags));
1378 DEFINE(TI_PREEMPT, offsetof(struct thread_info, preempt_count));
1379 + DEFINE(TI_PREEMPT_LAZY, offsetof(struct thread_info, preempt_lazy_count));
1380 DEFINE(TI_TASK, offsetof(struct thread_info, task));
1381 DEFINE(TI_CPU, offsetof(struct thread_info, cpu));
1383 diff --git a/arch/powerpc/kernel/entry_32.S b/arch/powerpc/kernel/entry_32.S
1384 index 3841d749a430..6dbaeff192b9 100644
1385 --- a/arch/powerpc/kernel/entry_32.S
1386 +++ b/arch/powerpc/kernel/entry_32.S
1387 @@ -835,7 +835,14 @@ user_exc_return: /* r10 contains MSR_KERNEL here */
1388 cmpwi 0,r0,0 /* if non-zero, just restore regs and return */
1390 andi. r8,r8,_TIF_NEED_RESCHED
1392 + lwz r0,TI_PREEMPT_LAZY(r9)
1393 + cmpwi 0,r0,0 /* if non-zero, just restore regs and return */
1395 + lwz r0,TI_FLAGS(r9)
1396 + andi. r0,r0,_TIF_NEED_RESCHED_LAZY
1400 andi. r0,r3,MSR_EE /* interrupts off? */
1401 beq restore /* don't schedule if so */
1402 @@ -846,11 +853,11 @@ user_exc_return: /* r10 contains MSR_KERNEL here */
1404 bl trace_hardirqs_off
1406 -1: bl preempt_schedule_irq
1407 +2: bl preempt_schedule_irq
1408 CURRENT_THREAD_INFO(r9, r1)
1410 - andi. r0,r3,_TIF_NEED_RESCHED
1412 + andi. r0,r3,_TIF_NEED_RESCHED_MASK
1414 #ifdef CONFIG_TRACE_IRQFLAGS
1415 /* And now, to properly rebalance the above, we tell lockdep they
1416 * are being turned back on, which will happen when we return
1417 @@ -1171,7 +1178,7 @@ END_FTR_SECTION_IFSET(CPU_FTR_NEED_PAIRED_STWCX)
1418 #endif /* !(CONFIG_4xx || CONFIG_BOOKE) */
1420 do_work: /* r10 contains MSR_KERNEL here */
1421 - andi. r0,r9,_TIF_NEED_RESCHED
1422 + andi. r0,r9,_TIF_NEED_RESCHED_MASK
1425 do_resched: /* r10 contains MSR_KERNEL here */
1426 @@ -1192,7 +1199,7 @@ do_resched: /* r10 contains MSR_KERNEL here */
1427 MTMSRD(r10) /* disable interrupts */
1428 CURRENT_THREAD_INFO(r9, r1)
1430 - andi. r0,r9,_TIF_NEED_RESCHED
1431 + andi. r0,r9,_TIF_NEED_RESCHED_MASK
1433 andi. r0,r9,_TIF_USER_WORK_MASK
1435 diff --git a/arch/powerpc/kernel/entry_64.S b/arch/powerpc/kernel/entry_64.S
1436 index 6432d4bf08c8..5509a26f1070 100644
1437 --- a/arch/powerpc/kernel/entry_64.S
1438 +++ b/arch/powerpc/kernel/entry_64.S
1439 @@ -656,7 +656,7 @@ _GLOBAL(ret_from_except_lite)
1443 -1: andi. r0,r4,_TIF_NEED_RESCHED
1444 +1: andi. r0,r4,_TIF_NEED_RESCHED_MASK
1446 bl restore_interrupts
1448 @@ -718,10 +718,18 @@ _GLOBAL(ret_from_except_lite)
1450 #ifdef CONFIG_PREEMPT
1451 /* Check if we need to preempt */
1452 - andi. r0,r4,_TIF_NEED_RESCHED
1454 - /* Check that preempt_count() == 0 and interrupts are enabled */
1455 lwz r8,TI_PREEMPT(r9)
1456 + cmpwi 0,r8,0 /* if non-zero, just restore regs and return */
1458 + andi. r0,r4,_TIF_NEED_RESCHED
1461 + andi. r0,r4,_TIF_NEED_RESCHED_LAZY
1463 + lwz r8,TI_PREEMPT_LAZY(r9)
1465 + /* Check that preempt_count() == 0 and interrupts are enabled */
1470 @@ -738,7 +746,7 @@ _GLOBAL(ret_from_except_lite)
1471 /* Re-test flags and eventually loop */
1472 CURRENT_THREAD_INFO(r9, r1)
1474 - andi. r0,r4,_TIF_NEED_RESCHED
1475 + andi. r0,r4,_TIF_NEED_RESCHED_MASK
1479 diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
1480 index 3c05c311e35e..f83f6ac1274d 100644
1481 --- a/arch/powerpc/kernel/irq.c
1482 +++ b/arch/powerpc/kernel/irq.c
1483 @@ -638,6 +638,7 @@ void irq_ctx_init(void)
1487 +#ifndef CONFIG_PREEMPT_RT_FULL
1488 void do_softirq_own_stack(void)
1490 struct thread_info *curtp, *irqtp;
1491 @@ -655,6 +656,7 @@ void do_softirq_own_stack(void)
1493 set_bits(irqtp->flags, &curtp->flags);
1497 irq_hw_number_t virq_to_hw(unsigned int virq)
1499 diff --git a/arch/powerpc/kernel/misc_32.S b/arch/powerpc/kernel/misc_32.S
1500 index 030d72df5dd5..b471a709e100 100644
1501 --- a/arch/powerpc/kernel/misc_32.S
1502 +++ b/arch/powerpc/kernel/misc_32.S
1504 * We store the saved ksp_limit in the unused part
1505 * of the STACK_FRAME_OVERHEAD
1507 +#ifndef CONFIG_PREEMPT_RT_FULL
1508 _GLOBAL(call_do_softirq)
1511 @@ -57,6 +58,7 @@ _GLOBAL(call_do_softirq)
1512 stw r10,THREAD+KSP_LIMIT(r2)
1518 * void call_do_irq(struct pt_regs *regs, struct thread_info *irqtp);
1519 diff --git a/arch/powerpc/kernel/misc_64.S b/arch/powerpc/kernel/misc_64.S
1520 index 4f178671f230..39e7d84a3492 100644
1521 --- a/arch/powerpc/kernel/misc_64.S
1522 +++ b/arch/powerpc/kernel/misc_64.S
1527 +#ifndef CONFIG_PREEMPT_RT_FULL
1528 _GLOBAL(call_do_softirq)
1531 @@ -41,6 +42,7 @@ _GLOBAL(call_do_softirq)
1537 _GLOBAL(call_do_irq)
1539 diff --git a/arch/powerpc/kvm/Kconfig b/arch/powerpc/kvm/Kconfig
1540 index 029be26b5a17..9528089ea142 100644
1541 --- a/arch/powerpc/kvm/Kconfig
1542 +++ b/arch/powerpc/kvm/Kconfig
1543 @@ -175,6 +175,7 @@ config KVM_E500MC
1545 bool "KVM in-kernel MPIC emulation"
1546 depends on KVM && E500
1547 + depends on !PREEMPT_RT_FULL
1548 select HAVE_KVM_IRQCHIP
1549 select HAVE_KVM_IRQFD
1550 select HAVE_KVM_IRQ_ROUTING
1551 diff --git a/arch/powerpc/platforms/ps3/device-init.c b/arch/powerpc/platforms/ps3/device-init.c
1552 index e48462447ff0..2670cee66064 100644
1553 --- a/arch/powerpc/platforms/ps3/device-init.c
1554 +++ b/arch/powerpc/platforms/ps3/device-init.c
1555 @@ -752,7 +752,7 @@ static int ps3_notification_read_write(struct ps3_notification_device *dev,
1557 pr_debug("%s:%u: notification %s issued\n", __func__, __LINE__, op);
1559 - res = wait_event_interruptible(dev->done.wait,
1560 + res = swait_event_interruptible(dev->done.wait,
1561 dev->done.done || kthread_should_stop());
1562 if (kthread_should_stop())
1564 diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
1565 index 6c0378c0b8b5..abd58b4dff97 100644
1566 --- a/arch/sh/kernel/irq.c
1567 +++ b/arch/sh/kernel/irq.c
1568 @@ -147,6 +147,7 @@ void irq_ctx_exit(int cpu)
1569 hardirq_ctx[cpu] = NULL;
1572 +#ifndef CONFIG_PREEMPT_RT_FULL
1573 void do_softirq_own_stack(void)
1575 struct thread_info *curctx;
1576 @@ -174,6 +175,7 @@ void do_softirq_own_stack(void)
1577 "r5", "r6", "r7", "r8", "r9", "r15", "t", "pr"
1582 static inline void handle_one_irq(unsigned int irq)
1584 diff --git a/arch/sparc/Kconfig b/arch/sparc/Kconfig
1585 index 165ecdd24d22..b68a464a22be 100644
1586 --- a/arch/sparc/Kconfig
1587 +++ b/arch/sparc/Kconfig
1588 @@ -194,12 +194,10 @@ config NR_CPUS
1589 source kernel/Kconfig.hz
1591 config RWSEM_GENERIC_SPINLOCK
1593 - default y if SPARC32
1594 + def_bool PREEMPT_RT_FULL
1596 config RWSEM_XCHGADD_ALGORITHM
1598 - default y if SPARC64
1599 + def_bool !RWSEM_GENERIC_SPINLOCK && !PREEMPT_RT_FULL
1601 config GENERIC_HWEIGHT
1603 diff --git a/arch/sparc/kernel/irq_64.c b/arch/sparc/kernel/irq_64.c
1604 index 34a7930b76ef..773740521008 100644
1605 --- a/arch/sparc/kernel/irq_64.c
1606 +++ b/arch/sparc/kernel/irq_64.c
1607 @@ -854,6 +854,7 @@ void __irq_entry handler_irq(int pil, struct pt_regs *regs)
1608 set_irq_regs(old_regs);
1611 +#ifndef CONFIG_PREEMPT_RT_FULL
1612 void do_softirq_own_stack(void)
1614 void *orig_sp, *sp = softirq_stack[smp_processor_id()];
1615 @@ -868,6 +869,7 @@ void do_softirq_own_stack(void)
1616 __asm__ __volatile__("mov %0, %%sp"
1621 #ifdef CONFIG_HOTPLUG_CPU
1622 void fixup_irqs(void)
1623 diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
1624 index bada636d1065..f8a995c90c01 100644
1625 --- a/arch/x86/Kconfig
1626 +++ b/arch/x86/Kconfig
1627 @@ -17,6 +17,7 @@ config X86_64
1631 + select HAVE_PREEMPT_LAZY
1632 select ACPI_LEGACY_TABLES_LOOKUP if ACPI
1633 select ACPI_SYSTEM_POWER_STATES_SUPPORT if ACPI
1635 @@ -232,8 +233,11 @@ config ARCH_MAY_HAVE_PC_FDC
1637 depends on ISA_DMA_API
1639 +config RWSEM_GENERIC_SPINLOCK
1640 + def_bool PREEMPT_RT_FULL
1642 config RWSEM_XCHGADD_ALGORITHM
1644 + def_bool !RWSEM_GENERIC_SPINLOCK && !PREEMPT_RT_FULL
1646 config GENERIC_CALIBRATE_DELAY
1648 @@ -897,7 +901,7 @@ config IOMMU_HELPER
1650 bool "Enable Maximum number of SMP Processors and NUMA Nodes"
1651 depends on X86_64 && SMP && DEBUG_KERNEL
1652 - select CPUMASK_OFFSTACK
1653 + select CPUMASK_OFFSTACK if !PREEMPT_RT_FULL
1655 Enable maximum number of CPUS and NUMA Nodes for this architecture.
1657 diff --git a/arch/x86/crypto/aesni-intel_glue.c b/arch/x86/crypto/aesni-intel_glue.c
1658 index aa8b0672f87a..2429414bfc71 100644
1659 --- a/arch/x86/crypto/aesni-intel_glue.c
1660 +++ b/arch/x86/crypto/aesni-intel_glue.c
1661 @@ -372,14 +372,14 @@ static int ecb_encrypt(struct blkcipher_desc *desc,
1662 err = blkcipher_walk_virt(desc, &walk);
1663 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1665 - kernel_fpu_begin();
1666 while ((nbytes = walk.nbytes)) {
1667 + kernel_fpu_begin();
1668 aesni_ecb_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr,
1669 - nbytes & AES_BLOCK_MASK);
1670 + nbytes & AES_BLOCK_MASK);
1672 nbytes &= AES_BLOCK_SIZE - 1;
1673 err = blkcipher_walk_done(desc, &walk, nbytes);
1679 @@ -396,14 +396,14 @@ static int ecb_decrypt(struct blkcipher_desc *desc,
1680 err = blkcipher_walk_virt(desc, &walk);
1681 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1683 - kernel_fpu_begin();
1684 while ((nbytes = walk.nbytes)) {
1685 + kernel_fpu_begin();
1686 aesni_ecb_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr,
1687 nbytes & AES_BLOCK_MASK);
1689 nbytes &= AES_BLOCK_SIZE - 1;
1690 err = blkcipher_walk_done(desc, &walk, nbytes);
1696 @@ -420,14 +420,14 @@ static int cbc_encrypt(struct blkcipher_desc *desc,
1697 err = blkcipher_walk_virt(desc, &walk);
1698 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1700 - kernel_fpu_begin();
1701 while ((nbytes = walk.nbytes)) {
1702 + kernel_fpu_begin();
1703 aesni_cbc_enc(ctx, walk.dst.virt.addr, walk.src.virt.addr,
1704 nbytes & AES_BLOCK_MASK, walk.iv);
1706 nbytes &= AES_BLOCK_SIZE - 1;
1707 err = blkcipher_walk_done(desc, &walk, nbytes);
1713 @@ -444,14 +444,14 @@ static int cbc_decrypt(struct blkcipher_desc *desc,
1714 err = blkcipher_walk_virt(desc, &walk);
1715 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1717 - kernel_fpu_begin();
1718 while ((nbytes = walk.nbytes)) {
1719 + kernel_fpu_begin();
1720 aesni_cbc_dec(ctx, walk.dst.virt.addr, walk.src.virt.addr,
1721 nbytes & AES_BLOCK_MASK, walk.iv);
1723 nbytes &= AES_BLOCK_SIZE - 1;
1724 err = blkcipher_walk_done(desc, &walk, nbytes);
1730 @@ -503,18 +503,20 @@ static int ctr_crypt(struct blkcipher_desc *desc,
1731 err = blkcipher_walk_virt_block(desc, &walk, AES_BLOCK_SIZE);
1732 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1734 - kernel_fpu_begin();
1735 while ((nbytes = walk.nbytes) >= AES_BLOCK_SIZE) {
1736 + kernel_fpu_begin();
1737 aesni_ctr_enc_tfm(ctx, walk.dst.virt.addr, walk.src.virt.addr,
1738 nbytes & AES_BLOCK_MASK, walk.iv);
1740 nbytes &= AES_BLOCK_SIZE - 1;
1741 err = blkcipher_walk_done(desc, &walk, nbytes);
1744 + kernel_fpu_begin();
1745 ctr_crypt_final(ctx, &walk);
1747 err = blkcipher_walk_done(desc, &walk, 0);
1753 diff --git a/arch/x86/crypto/cast5_avx_glue.c b/arch/x86/crypto/cast5_avx_glue.c
1754 index 8648158f3916..d7699130ee36 100644
1755 --- a/arch/x86/crypto/cast5_avx_glue.c
1756 +++ b/arch/x86/crypto/cast5_avx_glue.c
1757 @@ -59,7 +59,7 @@ static inline void cast5_fpu_end(bool fpu_enabled)
1758 static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk,
1761 - bool fpu_enabled = false;
1763 struct cast5_ctx *ctx = crypto_blkcipher_ctx(desc->tfm);
1764 const unsigned int bsize = CAST5_BLOCK_SIZE;
1765 unsigned int nbytes;
1766 @@ -75,7 +75,7 @@ static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk,
1767 u8 *wsrc = walk->src.virt.addr;
1768 u8 *wdst = walk->dst.virt.addr;
1770 - fpu_enabled = cast5_fpu_begin(fpu_enabled, nbytes);
1771 + fpu_enabled = cast5_fpu_begin(false, nbytes);
1773 /* Process multi-block batch */
1774 if (nbytes >= bsize * CAST5_PARALLEL_BLOCKS) {
1775 @@ -103,10 +103,9 @@ static int ecb_crypt(struct blkcipher_desc *desc, struct blkcipher_walk *walk,
1776 } while (nbytes >= bsize);
1779 + cast5_fpu_end(fpu_enabled);
1780 err = blkcipher_walk_done(desc, walk, nbytes);
1783 - cast5_fpu_end(fpu_enabled);
1787 @@ -227,7 +226,7 @@ static unsigned int __cbc_decrypt(struct blkcipher_desc *desc,
1788 static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst,
1789 struct scatterlist *src, unsigned int nbytes)
1791 - bool fpu_enabled = false;
1793 struct blkcipher_walk walk;
1796 @@ -236,12 +235,11 @@ static int cbc_decrypt(struct blkcipher_desc *desc, struct scatterlist *dst,
1797 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1799 while ((nbytes = walk.nbytes)) {
1800 - fpu_enabled = cast5_fpu_begin(fpu_enabled, nbytes);
1801 + fpu_enabled = cast5_fpu_begin(false, nbytes);
1802 nbytes = __cbc_decrypt(desc, &walk);
1803 + cast5_fpu_end(fpu_enabled);
1804 err = blkcipher_walk_done(desc, &walk, nbytes);
1807 - cast5_fpu_end(fpu_enabled);
1811 @@ -311,7 +309,7 @@ static unsigned int __ctr_crypt(struct blkcipher_desc *desc,
1812 static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst,
1813 struct scatterlist *src, unsigned int nbytes)
1815 - bool fpu_enabled = false;
1817 struct blkcipher_walk walk;
1820 @@ -320,13 +318,12 @@ static int ctr_crypt(struct blkcipher_desc *desc, struct scatterlist *dst,
1821 desc->flags &= ~CRYPTO_TFM_REQ_MAY_SLEEP;
1823 while ((nbytes = walk.nbytes) >= CAST5_BLOCK_SIZE) {
1824 - fpu_enabled = cast5_fpu_begin(fpu_enabled, nbytes);
1825 + fpu_enabled = cast5_fpu_begin(false, nbytes);
1826 nbytes = __ctr_crypt(desc, &walk);
1827 + cast5_fpu_end(fpu_enabled);
1828 err = blkcipher_walk_done(desc, &walk, nbytes);
1831 - cast5_fpu_end(fpu_enabled);
1834 ctr_crypt_final(desc, &walk);
1835 err = blkcipher_walk_done(desc, &walk, 0);
1836 diff --git a/arch/x86/crypto/glue_helper.c b/arch/x86/crypto/glue_helper.c
1837 index 6a85598931b5..3a506ce7ed93 100644
1838 --- a/arch/x86/crypto/glue_helper.c
1839 +++ b/arch/x86/crypto/glue_helper.c
1840 @@ -39,7 +39,7 @@ static int __glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx,
1841 void *ctx = crypto_blkcipher_ctx(desc->tfm);
1842 const unsigned int bsize = 128 / 8;
1843 unsigned int nbytes, i, func_bytes;
1844 - bool fpu_enabled = false;
1848 err = blkcipher_walk_virt(desc, walk);
1849 @@ -49,7 +49,7 @@ static int __glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx,
1850 u8 *wdst = walk->dst.virt.addr;
1852 fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit,
1853 - desc, fpu_enabled, nbytes);
1854 + desc, false, nbytes);
1856 for (i = 0; i < gctx->num_funcs; i++) {
1857 func_bytes = bsize * gctx->funcs[i].num_blocks;
1858 @@ -71,10 +71,10 @@ static int __glue_ecb_crypt_128bit(const struct common_glue_ctx *gctx,
1862 + glue_fpu_end(fpu_enabled);
1863 err = blkcipher_walk_done(desc, walk, nbytes);
1866 - glue_fpu_end(fpu_enabled);
1870 @@ -194,7 +194,7 @@ int glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx,
1871 struct scatterlist *src, unsigned int nbytes)
1873 const unsigned int bsize = 128 / 8;
1874 - bool fpu_enabled = false;
1876 struct blkcipher_walk walk;
1879 @@ -203,12 +203,12 @@ int glue_cbc_decrypt_128bit(const struct common_glue_ctx *gctx,
1881 while ((nbytes = walk.nbytes)) {
1882 fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit,
1883 - desc, fpu_enabled, nbytes);
1884 + desc, false, nbytes);
1885 nbytes = __glue_cbc_decrypt_128bit(gctx, desc, &walk);
1886 + glue_fpu_end(fpu_enabled);
1887 err = blkcipher_walk_done(desc, &walk, nbytes);
1890 - glue_fpu_end(fpu_enabled);
1893 EXPORT_SYMBOL_GPL(glue_cbc_decrypt_128bit);
1894 @@ -277,7 +277,7 @@ int glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx,
1895 struct scatterlist *src, unsigned int nbytes)
1897 const unsigned int bsize = 128 / 8;
1898 - bool fpu_enabled = false;
1900 struct blkcipher_walk walk;
1903 @@ -286,13 +286,12 @@ int glue_ctr_crypt_128bit(const struct common_glue_ctx *gctx,
1905 while ((nbytes = walk.nbytes) >= bsize) {
1906 fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit,
1907 - desc, fpu_enabled, nbytes);
1908 + desc, false, nbytes);
1909 nbytes = __glue_ctr_crypt_128bit(gctx, desc, &walk);
1910 + glue_fpu_end(fpu_enabled);
1911 err = blkcipher_walk_done(desc, &walk, nbytes);
1914 - glue_fpu_end(fpu_enabled);
1917 glue_ctr_crypt_final_128bit(
1918 gctx->funcs[gctx->num_funcs - 1].fn_u.ctr, desc, &walk);
1919 @@ -347,7 +346,7 @@ int glue_xts_crypt_128bit(const struct common_glue_ctx *gctx,
1920 void *tweak_ctx, void *crypt_ctx)
1922 const unsigned int bsize = 128 / 8;
1923 - bool fpu_enabled = false;
1925 struct blkcipher_walk walk;
1928 @@ -360,21 +359,21 @@ int glue_xts_crypt_128bit(const struct common_glue_ctx *gctx,
1930 /* set minimum length to bsize, for tweak_fn */
1931 fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit,
1932 - desc, fpu_enabled,
1934 nbytes < bsize ? bsize : nbytes);
1936 /* calculate first value of T */
1937 tweak_fn(tweak_ctx, walk.iv, walk.iv);
1938 + glue_fpu_end(fpu_enabled);
1941 + fpu_enabled = glue_fpu_begin(bsize, gctx->fpu_blocks_limit,
1942 + desc, false, nbytes);
1943 nbytes = __glue_xts_crypt_128bit(gctx, crypt_ctx, desc, &walk);
1945 + glue_fpu_end(fpu_enabled);
1946 err = blkcipher_walk_done(desc, &walk, nbytes);
1947 nbytes = walk.nbytes;
1950 - glue_fpu_end(fpu_enabled);
1954 EXPORT_SYMBOL_GPL(glue_xts_crypt_128bit);
1955 diff --git a/arch/x86/entry/common.c b/arch/x86/entry/common.c
1956 index bdd9cc59d20f..56d01a339ba4 100644
1957 --- a/arch/x86/entry/common.c
1958 +++ b/arch/x86/entry/common.c
1959 @@ -129,7 +129,7 @@ static long syscall_trace_enter(struct pt_regs *regs)
1961 #define EXIT_TO_USERMODE_LOOP_FLAGS \
1962 (_TIF_SIGPENDING | _TIF_NOTIFY_RESUME | _TIF_UPROBE | \
1963 - _TIF_NEED_RESCHED | _TIF_USER_RETURN_NOTIFY)
1964 + _TIF_NEED_RESCHED_MASK | _TIF_USER_RETURN_NOTIFY)
1966 static void exit_to_usermode_loop(struct pt_regs *regs, u32 cached_flags)
1968 @@ -145,9 +145,16 @@ static void exit_to_usermode_loop(struct pt_regs *regs, u32 cached_flags)
1969 /* We have work to do. */
1972 - if (cached_flags & _TIF_NEED_RESCHED)
1973 + if (cached_flags & _TIF_NEED_RESCHED_MASK)
1976 +#ifdef ARCH_RT_DELAYS_SIGNAL_SEND
1977 + if (unlikely(current->forced_info.si_signo)) {
1978 + struct task_struct *t = current;
1979 + force_sig_info(t->forced_info.si_signo, &t->forced_info, t);
1980 + t->forced_info.si_signo = 0;
1983 if (cached_flags & _TIF_UPROBE)
1984 uprobe_notify_resume(regs);
1986 diff --git a/arch/x86/entry/entry_32.S b/arch/x86/entry/entry_32.S
1987 index edba8606b99a..4a3389535fc6 100644
1988 --- a/arch/x86/entry/entry_32.S
1989 +++ b/arch/x86/entry/entry_32.S
1990 @@ -308,8 +308,25 @@ END(ret_from_exception)
1991 ENTRY(resume_kernel)
1992 DISABLE_INTERRUPTS(CLBR_ANY)
1994 + # preempt count == 0 + NEED_RS set?
1995 cmpl $0, PER_CPU_VAR(__preempt_count)
1996 +#ifndef CONFIG_PREEMPT_LAZY
2001 + # atleast preempt count == 0 ?
2002 + cmpl $_PREEMPT_ENABLED,PER_CPU_VAR(__preempt_count)
2005 + movl PER_CPU_VAR(current_task), %ebp
2006 + cmpl $0,TASK_TI_preempt_lazy_count(%ebp) # non-zero preempt_lazy_count ?
2009 + testl $_TIF_NEED_RESCHED_LAZY, TASK_TI_flags(%ebp)
2013 testl $X86_EFLAGS_IF, PT_EFLAGS(%esp) # interrupts off (exception path) ?
2015 call preempt_schedule_irq
2016 diff --git a/arch/x86/entry/entry_64.S b/arch/x86/entry/entry_64.S
2017 index ef766a358b37..28401f826ab1 100644
2018 --- a/arch/x86/entry/entry_64.S
2019 +++ b/arch/x86/entry/entry_64.S
2020 @@ -546,7 +546,23 @@ GLOBAL(retint_user)
2021 bt $9, EFLAGS(%rsp) /* were interrupts off? */
2023 0: cmpl $0, PER_CPU_VAR(__preempt_count)
2024 +#ifndef CONFIG_PREEMPT_LAZY
2027 + jz do_preempt_schedule_irq
2029 + # atleast preempt count == 0 ?
2030 + cmpl $_PREEMPT_ENABLED,PER_CPU_VAR(__preempt_count)
2033 + movq PER_CPU_VAR(current_task), %rcx
2034 + cmpl $0, TASK_TI_preempt_lazy_count(%rcx)
2037 + bt $TIF_NEED_RESCHED_LAZY,TASK_TI_flags(%rcx)
2039 +do_preempt_schedule_irq:
2041 call preempt_schedule_irq
2044 @@ -894,6 +910,7 @@ EXPORT_SYMBOL(native_load_gs_index)
2048 +#ifndef CONFIG_PREEMPT_RT_FULL
2049 /* Call softirq on interrupt stack. Interrupts are off. */
2050 ENTRY(do_softirq_own_stack)
2052 @@ -906,6 +923,7 @@ ENTRY(do_softirq_own_stack)
2053 decl PER_CPU_VAR(irq_count)
2055 END(do_softirq_own_stack)
2059 idtentry xen_hypervisor_callback xen_do_hypervisor_callback has_error_code=0
2060 diff --git a/arch/x86/include/asm/preempt.h b/arch/x86/include/asm/preempt.h
2061 index 17f218645701..11bd1b7ee6eb 100644
2062 --- a/arch/x86/include/asm/preempt.h
2063 +++ b/arch/x86/include/asm/preempt.h
2064 @@ -79,17 +79,46 @@ static __always_inline void __preempt_count_sub(int val)
2065 * a decrement which hits zero means we have no preempt_count and should
2068 -static __always_inline bool __preempt_count_dec_and_test(void)
2069 +static __always_inline bool ____preempt_count_dec_and_test(void)
2071 GEN_UNARY_RMWcc("decl", __preempt_count, __percpu_arg(0), e);
2074 +static __always_inline bool __preempt_count_dec_and_test(void)
2076 + if (____preempt_count_dec_and_test())
2078 +#ifdef CONFIG_PREEMPT_LAZY
2079 + if (current_thread_info()->preempt_lazy_count)
2081 + return test_thread_flag(TIF_NEED_RESCHED_LAZY);
2088 * Returns true when we need to resched and can (barring IRQ state).
2090 static __always_inline bool should_resched(int preempt_offset)
2092 +#ifdef CONFIG_PREEMPT_LAZY
2095 + tmp = raw_cpu_read_4(__preempt_count);
2096 + if (tmp == preempt_offset)
2099 + /* preempt count == 0 ? */
2100 + tmp &= ~PREEMPT_NEED_RESCHED;
2103 + if (current_thread_info()->preempt_lazy_count)
2105 + return test_thread_flag(TIF_NEED_RESCHED_LAZY);
2107 return unlikely(raw_cpu_read_4(__preempt_count) == preempt_offset);
2111 #ifdef CONFIG_PREEMPT
2112 diff --git a/arch/x86/include/asm/signal.h b/arch/x86/include/asm/signal.h
2113 index 8af22be0fe61..d1328789b759 100644
2114 --- a/arch/x86/include/asm/signal.h
2115 +++ b/arch/x86/include/asm/signal.h
2116 @@ -27,6 +27,19 @@ typedef struct {
2117 #define SA_IA32_ABI 0x02000000u
2118 #define SA_X32_ABI 0x01000000u
2121 + * Because some traps use the IST stack, we must keep preemption
2122 + * disabled while calling do_trap(), but do_trap() may call
2123 + * force_sig_info() which will grab the signal spin_locks for the
2124 + * task, which in PREEMPT_RT_FULL are mutexes. By defining
2125 + * ARCH_RT_DELAYS_SIGNAL_SEND the force_sig_info() will set
2126 + * TIF_NOTIFY_RESUME and set up the signal to be sent on exit of the
2129 +#if defined(CONFIG_PREEMPT_RT_FULL)
2130 +#define ARCH_RT_DELAYS_SIGNAL_SEND
2133 #ifndef CONFIG_COMPAT
2134 typedef sigset_t compat_sigset_t;
2136 diff --git a/arch/x86/include/asm/stackprotector.h b/arch/x86/include/asm/stackprotector.h
2137 index 58505f01962f..02fa39652cd6 100644
2138 --- a/arch/x86/include/asm/stackprotector.h
2139 +++ b/arch/x86/include/asm/stackprotector.h
2142 static __always_inline void boot_init_stack_canary(void)
2145 + u64 uninitialized_var(canary);
2148 #ifdef CONFIG_X86_64
2149 @@ -70,8 +70,15 @@ static __always_inline void boot_init_stack_canary(void)
2150 * of randomness. The TSC only matters for very early init,
2151 * there it already has some randomness on most systems. Later
2152 * on during the bootup the random pool has true entropy too.
2154 + * For preempt-rt we need to weaken the randomness a bit, as
2155 + * we can't call into the random generator from atomic context
2156 + * due to locking constraints. We just leave canary
2157 + * uninitialized and use the TSC based randomness on top of it.
2159 +#ifndef CONFIG_PREEMPT_RT_FULL
2160 get_random_bytes(&canary, sizeof(canary));
2163 canary += tsc + (tsc << 32UL);
2165 diff --git a/arch/x86/include/asm/thread_info.h b/arch/x86/include/asm/thread_info.h
2166 index ad6f5eb07a95..5ceb3a1c2b1a 100644
2167 --- a/arch/x86/include/asm/thread_info.h
2168 +++ b/arch/x86/include/asm/thread_info.h
2169 @@ -54,11 +54,14 @@ struct task_struct;
2171 struct thread_info {
2172 unsigned long flags; /* low level flags */
2173 + int preempt_lazy_count; /* 0 => lazy preemptable
2177 #define INIT_THREAD_INFO(tsk) \
2180 + .preempt_lazy_count = 0, \
2183 #define init_stack (init_thread_union.stack)
2184 @@ -67,6 +70,10 @@ struct thread_info {
2186 #include <asm/asm-offsets.h>
2188 +#define GET_THREAD_INFO(reg) \
2189 + _ASM_MOV PER_CPU_VAR(cpu_current_top_of_stack),reg ; \
2190 + _ASM_SUB $(THREAD_SIZE),reg ;
2195 @@ -85,6 +92,7 @@ struct thread_info {
2196 #define TIF_SYSCALL_EMU 6 /* syscall emulation active */
2197 #define TIF_SYSCALL_AUDIT 7 /* syscall auditing active */
2198 #define TIF_SECCOMP 8 /* secure computing */
2199 +#define TIF_NEED_RESCHED_LAZY 9 /* lazy rescheduling necessary */
2200 #define TIF_USER_RETURN_NOTIFY 11 /* notify kernel of userspace return */
2201 #define TIF_UPROBE 12 /* breakpointed or singlestepping */
2202 #define TIF_NOTSC 16 /* TSC is not accessible in userland */
2203 @@ -108,6 +116,7 @@ struct thread_info {
2204 #define _TIF_SYSCALL_EMU (1 << TIF_SYSCALL_EMU)
2205 #define _TIF_SYSCALL_AUDIT (1 << TIF_SYSCALL_AUDIT)
2206 #define _TIF_SECCOMP (1 << TIF_SECCOMP)
2207 +#define _TIF_NEED_RESCHED_LAZY (1 << TIF_NEED_RESCHED_LAZY)
2208 #define _TIF_USER_RETURN_NOTIFY (1 << TIF_USER_RETURN_NOTIFY)
2209 #define _TIF_UPROBE (1 << TIF_UPROBE)
2210 #define _TIF_NOTSC (1 << TIF_NOTSC)
2211 @@ -143,6 +152,8 @@ struct thread_info {
2212 #define _TIF_WORK_CTXSW_PREV (_TIF_WORK_CTXSW|_TIF_USER_RETURN_NOTIFY)
2213 #define _TIF_WORK_CTXSW_NEXT (_TIF_WORK_CTXSW)
2215 +#define _TIF_NEED_RESCHED_MASK (_TIF_NEED_RESCHED | _TIF_NEED_RESCHED_LAZY)
2217 #define STACK_WARN (THREAD_SIZE/8)
2220 diff --git a/arch/x86/include/asm/uv/uv_bau.h b/arch/x86/include/asm/uv/uv_bau.h
2221 index 57ab86d94d64..35d25e27180f 100644
2222 --- a/arch/x86/include/asm/uv/uv_bau.h
2223 +++ b/arch/x86/include/asm/uv/uv_bau.h
2224 @@ -624,9 +624,9 @@ struct bau_control {
2225 cycles_t send_message;
2226 cycles_t period_end;
2227 cycles_t period_time;
2228 - spinlock_t uvhub_lock;
2229 - spinlock_t queue_lock;
2230 - spinlock_t disable_lock;
2231 + raw_spinlock_t uvhub_lock;
2232 + raw_spinlock_t queue_lock;
2233 + raw_spinlock_t disable_lock;
2236 int max_concurr_const;
2237 @@ -815,15 +815,15 @@ static inline int atom_asr(short i, struct atomic_short *v)
2238 * to be lowered below the current 'v'. atomic_add_unless can only stop
2241 -static inline int atomic_inc_unless_ge(spinlock_t *lock, atomic_t *v, int u)
2242 +static inline int atomic_inc_unless_ge(raw_spinlock_t *lock, atomic_t *v, int u)
2245 + raw_spin_lock(lock);
2246 if (atomic_read(v) >= u) {
2247 - spin_unlock(lock);
2248 + raw_spin_unlock(lock);
2252 - spin_unlock(lock);
2253 + raw_spin_unlock(lock);
2257 diff --git a/arch/x86/kernel/acpi/boot.c b/arch/x86/kernel/acpi/boot.c
2258 index 931ced8ca345..167975ac8af7 100644
2259 --- a/arch/x86/kernel/acpi/boot.c
2260 +++ b/arch/x86/kernel/acpi/boot.c
2261 @@ -87,7 +87,9 @@ static u64 acpi_lapic_addr __initdata = APIC_DEFAULT_PHYS_BASE;
2265 +#ifdef CONFIG_X86_IO_APIC
2266 static DEFINE_MUTEX(acpi_ioapic_lock);
2269 /* --------------------------------------------------------------------------
2270 Boot-time Configuration
2271 diff --git a/arch/x86/kernel/apic/io_apic.c b/arch/x86/kernel/apic/io_apic.c
2272 index d1e25564b3c1..67e585fa801f 100644
2273 --- a/arch/x86/kernel/apic/io_apic.c
2274 +++ b/arch/x86/kernel/apic/io_apic.c
2275 @@ -1712,7 +1712,8 @@ static bool io_apic_level_ack_pending(struct mp_chip_data *data)
2276 static inline bool ioapic_irqd_mask(struct irq_data *data)
2278 /* If we are moving the irq we need to mask it */
2279 - if (unlikely(irqd_is_setaffinity_pending(data))) {
2280 + if (unlikely(irqd_is_setaffinity_pending(data) &&
2281 + !irqd_irq_inprogress(data))) {
2282 mask_ioapic_irq(data);
2285 diff --git a/arch/x86/kernel/asm-offsets.c b/arch/x86/kernel/asm-offsets.c
2286 index c62e015b126c..0cc71257fca6 100644
2287 --- a/arch/x86/kernel/asm-offsets.c
2288 +++ b/arch/x86/kernel/asm-offsets.c
2289 @@ -36,6 +36,7 @@ void common(void) {
2292 OFFSET(TASK_TI_flags, task_struct, thread_info.flags);
2293 + OFFSET(TASK_TI_preempt_lazy_count, task_struct, thread_info.preempt_lazy_count);
2294 OFFSET(TASK_addr_limit, task_struct, thread.addr_limit);
2297 @@ -91,4 +92,5 @@ void common(void) {
2300 DEFINE(PTREGS_SIZE, sizeof(struct pt_regs));
2301 + DEFINE(_PREEMPT_ENABLED, PREEMPT_ENABLED);
2303 diff --git a/arch/x86/kernel/cpu/mcheck/mce.c b/arch/x86/kernel/cpu/mcheck/mce.c
2304 index a7fdf453d895..e3a0e969a66e 100644
2305 --- a/arch/x86/kernel/cpu/mcheck/mce.c
2306 +++ b/arch/x86/kernel/cpu/mcheck/mce.c
2308 #include <linux/debugfs.h>
2309 #include <linux/irq_work.h>
2310 #include <linux/export.h>
2311 +#include <linux/jiffies.h>
2312 +#include <linux/swork.h>
2313 #include <linux/jump_label.h>
2315 #include <asm/processor.h>
2316 @@ -1317,7 +1319,7 @@ void mce_log_therm_throt_event(__u64 status)
2317 static unsigned long check_interval = INITIAL_CHECK_INTERVAL;
2319 static DEFINE_PER_CPU(unsigned long, mce_next_interval); /* in jiffies */
2320 -static DEFINE_PER_CPU(struct timer_list, mce_timer);
2321 +static DEFINE_PER_CPU(struct hrtimer, mce_timer);
2323 static unsigned long mce_adjust_timer_default(unsigned long interval)
2325 @@ -1326,32 +1328,18 @@ static unsigned long mce_adjust_timer_default(unsigned long interval)
2327 static unsigned long (*mce_adjust_timer)(unsigned long interval) = mce_adjust_timer_default;
2329 -static void __restart_timer(struct timer_list *t, unsigned long interval)
2330 +static enum hrtimer_restart __restart_timer(struct hrtimer *timer, unsigned long interval)
2332 - unsigned long when = jiffies + interval;
2333 - unsigned long flags;
2335 - local_irq_save(flags);
2337 - if (timer_pending(t)) {
2338 - if (time_before(when, t->expires))
2339 - mod_timer(t, when);
2341 - t->expires = round_jiffies(when);
2342 - add_timer_on(t, smp_processor_id());
2345 - local_irq_restore(flags);
2347 + return HRTIMER_NORESTART;
2348 + hrtimer_forward_now(timer, ns_to_ktime(jiffies_to_nsecs(interval)));
2349 + return HRTIMER_RESTART;
2352 -static void mce_timer_fn(unsigned long data)
2353 +static enum hrtimer_restart mce_timer_fn(struct hrtimer *timer)
2355 - struct timer_list *t = this_cpu_ptr(&mce_timer);
2356 - int cpu = smp_processor_id();
2359 - WARN_ON(cpu != data);
2361 iv = __this_cpu_read(mce_next_interval);
2363 if (mce_available(this_cpu_ptr(&cpu_info))) {
2364 @@ -1374,7 +1362,7 @@ static void mce_timer_fn(unsigned long data)
2367 __this_cpu_write(mce_next_interval, iv);
2368 - __restart_timer(t, iv);
2369 + return __restart_timer(timer, iv);
2373 @@ -1382,7 +1370,7 @@ static void mce_timer_fn(unsigned long data)
2375 void mce_timer_kick(unsigned long interval)
2377 - struct timer_list *t = this_cpu_ptr(&mce_timer);
2378 + struct hrtimer *t = this_cpu_ptr(&mce_timer);
2379 unsigned long iv = __this_cpu_read(mce_next_interval);
2381 __restart_timer(t, interval);
2382 @@ -1397,7 +1385,7 @@ static void mce_timer_delete_all(void)
2385 for_each_online_cpu(cpu)
2386 - del_timer_sync(&per_cpu(mce_timer, cpu));
2387 + hrtimer_cancel(&per_cpu(mce_timer, cpu));
2390 static void mce_do_trigger(struct work_struct *work)
2391 @@ -1407,6 +1395,56 @@ static void mce_do_trigger(struct work_struct *work)
2393 static DECLARE_WORK(mce_trigger_work, mce_do_trigger);
2395 +static void __mce_notify_work(struct swork_event *event)
2397 + /* Not more than two messages every minute */
2398 + static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2);
2400 + /* wake processes polling /dev/mcelog */
2401 + wake_up_interruptible(&mce_chrdev_wait);
2404 + * There is no risk of missing notifications because
2405 + * work_pending is always cleared before the function is
2408 + if (mce_helper[0] && !work_pending(&mce_trigger_work))
2409 + schedule_work(&mce_trigger_work);
2411 + if (__ratelimit(&ratelimit))
2412 + pr_info(HW_ERR "Machine check events logged\n");
2415 +#ifdef CONFIG_PREEMPT_RT_FULL
2416 +static bool notify_work_ready __read_mostly;
2417 +static struct swork_event notify_work;
2419 +static int mce_notify_work_init(void)
2423 + err = swork_get();
2427 + INIT_SWORK(¬ify_work, __mce_notify_work);
2428 + notify_work_ready = true;
2432 +static void mce_notify_work(void)
2434 + if (notify_work_ready)
2435 + swork_queue(¬ify_work);
2438 +static void mce_notify_work(void)
2440 + __mce_notify_work(NULL);
2442 +static inline int mce_notify_work_init(void) { return 0; }
2446 * Notify the user(s) about new machine check events.
2447 * Can be called from interrupt context, but not from machine check/NMI
2448 @@ -1414,19 +1452,8 @@ static DECLARE_WORK(mce_trigger_work, mce_do_trigger);
2450 int mce_notify_irq(void)
2452 - /* Not more than two messages every minute */
2453 - static DEFINE_RATELIMIT_STATE(ratelimit, 60*HZ, 2);
2455 if (test_and_clear_bit(0, &mce_need_notify)) {
2456 - /* wake processes polling /dev/mcelog */
2457 - wake_up_interruptible(&mce_chrdev_wait);
2459 - if (mce_helper[0])
2460 - schedule_work(&mce_trigger_work);
2462 - if (__ratelimit(&ratelimit))
2463 - pr_info(HW_ERR "Machine check events logged\n");
2465 + mce_notify_work();
2469 @@ -1732,7 +1759,7 @@ static void __mcheck_cpu_clear_vendor(struct cpuinfo_x86 *c)
2473 -static void mce_start_timer(unsigned int cpu, struct timer_list *t)
2474 +static void mce_start_timer(unsigned int cpu, struct hrtimer *t)
2476 unsigned long iv = check_interval * HZ;
2478 @@ -1741,16 +1768,17 @@ static void mce_start_timer(unsigned int cpu, struct timer_list *t)
2480 per_cpu(mce_next_interval, cpu) = iv;
2482 - t->expires = round_jiffies(jiffies + iv);
2483 - add_timer_on(t, cpu);
2484 + hrtimer_start_range_ns(t, ns_to_ktime(jiffies_to_usecs(iv) * 1000ULL),
2485 + 0, HRTIMER_MODE_REL_PINNED);
2488 static void __mcheck_cpu_init_timer(void)
2490 - struct timer_list *t = this_cpu_ptr(&mce_timer);
2491 + struct hrtimer *t = this_cpu_ptr(&mce_timer);
2492 unsigned int cpu = smp_processor_id();
2494 - setup_pinned_timer(t, mce_timer_fn, cpu);
2495 + hrtimer_init(t, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
2496 + t->function = mce_timer_fn;
2497 mce_start_timer(cpu, t);
2500 @@ -2475,6 +2503,8 @@ static void mce_disable_cpu(void *h)
2501 if (!mce_available(raw_cpu_ptr(&cpu_info)))
2504 + hrtimer_cancel(this_cpu_ptr(&mce_timer));
2506 if (!(action & CPU_TASKS_FROZEN))
2509 @@ -2497,6 +2527,7 @@ static void mce_reenable_cpu(void *h)
2511 wrmsrl(msr_ops.ctl(i), b->ctl);
2513 + __mcheck_cpu_init_timer();
2516 /* Get notified when a cpu comes on/off. Be hotplug friendly. */
2517 @@ -2504,7 +2535,6 @@ static int
2518 mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
2520 unsigned int cpu = (unsigned long)hcpu;
2521 - struct timer_list *t = &per_cpu(mce_timer, cpu);
2523 switch (action & ~CPU_TASKS_FROZEN) {
2525 @@ -2524,11 +2554,9 @@ mce_cpu_callback(struct notifier_block *nfb, unsigned long action, void *hcpu)
2527 case CPU_DOWN_PREPARE:
2528 smp_call_function_single(cpu, mce_disable_cpu, &action, 1);
2529 - del_timer_sync(t);
2531 case CPU_DOWN_FAILED:
2532 smp_call_function_single(cpu, mce_reenable_cpu, &action, 1);
2533 - mce_start_timer(cpu, t);
2537 @@ -2567,6 +2595,10 @@ static __init int mcheck_init_device(void)
2541 + err = mce_notify_work_init();
2545 if (!zalloc_cpumask_var(&mce_device_initialized, GFP_KERNEL)) {
2548 diff --git a/arch/x86/kernel/irq_32.c b/arch/x86/kernel/irq_32.c
2549 index 1f38d9a4d9de..053bf3b2ef39 100644
2550 --- a/arch/x86/kernel/irq_32.c
2551 +++ b/arch/x86/kernel/irq_32.c
2552 @@ -127,6 +127,7 @@ void irq_ctx_init(int cpu)
2553 cpu, per_cpu(hardirq_stack, cpu), per_cpu(softirq_stack, cpu));
2556 +#ifndef CONFIG_PREEMPT_RT_FULL
2557 void do_softirq_own_stack(void)
2559 struct irq_stack *irqstk;
2560 @@ -143,6 +144,7 @@ void do_softirq_own_stack(void)
2562 call_on_stack(__do_softirq, isp);
2566 bool handle_irq(struct irq_desc *desc, struct pt_regs *regs)
2568 diff --git a/arch/x86/kernel/process_32.c b/arch/x86/kernel/process_32.c
2569 index bd7be8efdc4c..b3b0a7f7b1ca 100644
2570 --- a/arch/x86/kernel/process_32.c
2571 +++ b/arch/x86/kernel/process_32.c
2573 #include <linux/uaccess.h>
2574 #include <linux/io.h>
2575 #include <linux/kdebug.h>
2576 +#include <linux/highmem.h>
2578 #include <asm/pgtable.h>
2579 #include <asm/ldt.h>
2580 @@ -195,6 +196,35 @@ start_thread(struct pt_regs *regs, unsigned long new_ip, unsigned long new_sp)
2582 EXPORT_SYMBOL_GPL(start_thread);
2584 +#ifdef CONFIG_PREEMPT_RT_FULL
2585 +static void switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p)
2590 + * Clear @prev's kmap_atomic mappings
2592 + for (i = 0; i < prev_p->kmap_idx; i++) {
2593 + int idx = i + KM_TYPE_NR * smp_processor_id();
2594 + pte_t *ptep = kmap_pte - idx;
2596 + kpte_clear_flush(ptep, __fix_to_virt(FIX_KMAP_BEGIN + idx));
2599 + * Restore @next_p's kmap_atomic mappings
2601 + for (i = 0; i < next_p->kmap_idx; i++) {
2602 + int idx = i + KM_TYPE_NR * smp_processor_id();
2604 + if (!pte_none(next_p->kmap_pte[i]))
2605 + set_pte(kmap_pte - idx, next_p->kmap_pte[i]);
2610 +switch_kmaps(struct task_struct *prev_p, struct task_struct *next_p) { }
2615 * switch_to(x,y) should switch tasks from x to y.
2616 @@ -271,6 +301,8 @@ __switch_to(struct task_struct *prev_p, struct task_struct *next_p)
2617 task_thread_info(next_p)->flags & _TIF_WORK_CTXSW_NEXT))
2618 __switch_to_xtra(prev_p, next_p, tss);
2620 + switch_kmaps(prev_p, next_p);
2623 * Leave lazy mode, flushing any hypercalls made here.
2624 * This must be done before restoring TLS segments so
2625 diff --git a/arch/x86/kvm/lapic.c b/arch/x86/kvm/lapic.c
2626 index 3f05c044720b..fe68afd37162 100644
2627 --- a/arch/x86/kvm/lapic.c
2628 +++ b/arch/x86/kvm/lapic.c
2629 @@ -1939,6 +1939,7 @@ int kvm_create_lapic(struct kvm_vcpu *vcpu)
2630 hrtimer_init(&apic->lapic_timer.timer, CLOCK_MONOTONIC,
2631 HRTIMER_MODE_ABS_PINNED);
2632 apic->lapic_timer.timer.function = apic_timer_fn;
2633 + apic->lapic_timer.timer.irqsafe = 1;
2636 * APIC is created enabled. This will prevent kvm_lapic_set_base from
2637 diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
2638 index 731044efb195..d2905d9881f0 100644
2639 --- a/arch/x86/kvm/x86.c
2640 +++ b/arch/x86/kvm/x86.c
2641 @@ -5933,6 +5933,13 @@ int kvm_arch_init(void *opaque)
2645 +#ifdef CONFIG_PREEMPT_RT_FULL
2646 + if (!boot_cpu_has(X86_FEATURE_CONSTANT_TSC)) {
2647 + printk(KERN_ERR "RT requires X86_FEATURE_CONSTANT_TSC\n");
2648 + return -EOPNOTSUPP;
2652 r = kvm_mmu_module_init();
2654 goto out_free_percpu;
2655 diff --git a/arch/x86/mm/highmem_32.c b/arch/x86/mm/highmem_32.c
2656 index 6d18b70ed5a9..f752724c22e8 100644
2657 --- a/arch/x86/mm/highmem_32.c
2658 +++ b/arch/x86/mm/highmem_32.c
2659 @@ -32,10 +32,11 @@ EXPORT_SYMBOL(kunmap);
2661 void *kmap_atomic_prot(struct page *page, pgprot_t prot)
2663 + pte_t pte = mk_pte(page, prot);
2664 unsigned long vaddr;
2667 - preempt_disable();
2668 + preempt_disable_nort();
2669 pagefault_disable();
2671 if (!PageHighMem(page))
2672 @@ -45,7 +46,10 @@ void *kmap_atomic_prot(struct page *page, pgprot_t prot)
2673 idx = type + KM_TYPE_NR*smp_processor_id();
2674 vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
2675 BUG_ON(!pte_none(*(kmap_pte-idx)));
2676 - set_pte(kmap_pte-idx, mk_pte(page, prot));
2677 +#ifdef CONFIG_PREEMPT_RT_FULL
2678 + current->kmap_pte[type] = pte;
2680 + set_pte(kmap_pte-idx, pte);
2681 arch_flush_lazy_mmu_mode();
2683 return (void *)vaddr;
2684 @@ -88,6 +92,9 @@ void __kunmap_atomic(void *kvaddr)
2685 * is a bad idea also, in case the page changes cacheability
2686 * attributes or becomes a protected page in a hypervisor.
2688 +#ifdef CONFIG_PREEMPT_RT_FULL
2689 + current->kmap_pte[type] = __pte(0);
2691 kpte_clear_flush(kmap_pte-idx, vaddr);
2692 kmap_atomic_idx_pop();
2693 arch_flush_lazy_mmu_mode();
2694 @@ -100,7 +107,7 @@ void __kunmap_atomic(void *kvaddr)
2699 + preempt_enable_nort();
2701 EXPORT_SYMBOL(__kunmap_atomic);
2703 diff --git a/arch/x86/mm/iomap_32.c b/arch/x86/mm/iomap_32.c
2704 index ada98b39b8ad..585f6829653b 100644
2705 --- a/arch/x86/mm/iomap_32.c
2706 +++ b/arch/x86/mm/iomap_32.c
2707 @@ -56,6 +56,7 @@ EXPORT_SYMBOL_GPL(iomap_free);
2709 void *kmap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
2711 + pte_t pte = pfn_pte(pfn, prot);
2712 unsigned long vaddr;
2715 @@ -65,7 +66,12 @@ void *kmap_atomic_prot_pfn(unsigned long pfn, pgprot_t prot)
2716 type = kmap_atomic_idx_push();
2717 idx = type + KM_TYPE_NR * smp_processor_id();
2718 vaddr = __fix_to_virt(FIX_KMAP_BEGIN + idx);
2719 - set_pte(kmap_pte - idx, pfn_pte(pfn, prot));
2720 + WARN_ON(!pte_none(*(kmap_pte - idx)));
2722 +#ifdef CONFIG_PREEMPT_RT_FULL
2723 + current->kmap_pte[type] = pte;
2725 + set_pte(kmap_pte - idx, pte);
2726 arch_flush_lazy_mmu_mode();
2728 return (void *)vaddr;
2729 @@ -113,6 +119,9 @@ iounmap_atomic(void __iomem *kvaddr)
2730 * is a bad idea also, in case the page changes cacheability
2731 * attributes or becomes a protected page in a hypervisor.
2733 +#ifdef CONFIG_PREEMPT_RT_FULL
2734 + current->kmap_pte[type] = __pte(0);
2736 kpte_clear_flush(kmap_pte-idx, vaddr);
2737 kmap_atomic_idx_pop();
2739 diff --git a/arch/x86/mm/pageattr.c b/arch/x86/mm/pageattr.c
2740 index e3353c97d086..01664968555c 100644
2741 --- a/arch/x86/mm/pageattr.c
2742 +++ b/arch/x86/mm/pageattr.c
2743 @@ -214,7 +214,15 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
2744 int in_flags, struct page **pages)
2746 unsigned int i, level;
2747 +#ifdef CONFIG_PREEMPT
2749 + * Avoid wbinvd() because it causes latencies on all CPUs,
2750 + * regardless of any CPU isolation that may be in effect.
2752 + unsigned long do_wbinvd = 0;
2754 unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
2757 BUG_ON(irqs_disabled());
2759 diff --git a/arch/x86/platform/uv/tlb_uv.c b/arch/x86/platform/uv/tlb_uv.c
2760 index 9e42842e924a..5398f97172f9 100644
2761 --- a/arch/x86/platform/uv/tlb_uv.c
2762 +++ b/arch/x86/platform/uv/tlb_uv.c
2763 @@ -748,9 +748,9 @@ static void destination_plugged(struct bau_desc *bau_desc,
2765 quiesce_local_uvhub(hmaster);
2767 - spin_lock(&hmaster->queue_lock);
2768 + raw_spin_lock(&hmaster->queue_lock);
2769 reset_with_ipi(&bau_desc->distribution, bcp);
2770 - spin_unlock(&hmaster->queue_lock);
2771 + raw_spin_unlock(&hmaster->queue_lock);
2773 end_uvhub_quiesce(hmaster);
2775 @@ -770,9 +770,9 @@ static void destination_timeout(struct bau_desc *bau_desc,
2777 quiesce_local_uvhub(hmaster);
2779 - spin_lock(&hmaster->queue_lock);
2780 + raw_spin_lock(&hmaster->queue_lock);
2781 reset_with_ipi(&bau_desc->distribution, bcp);
2782 - spin_unlock(&hmaster->queue_lock);
2783 + raw_spin_unlock(&hmaster->queue_lock);
2785 end_uvhub_quiesce(hmaster);
2787 @@ -793,7 +793,7 @@ static void disable_for_period(struct bau_control *bcp, struct ptc_stats *stat)
2790 hmaster = bcp->uvhub_master;
2791 - spin_lock(&hmaster->disable_lock);
2792 + raw_spin_lock(&hmaster->disable_lock);
2793 if (!bcp->baudisabled) {
2794 stat->s_bau_disabled++;
2796 @@ -806,7 +806,7 @@ static void disable_for_period(struct bau_control *bcp, struct ptc_stats *stat)
2800 - spin_unlock(&hmaster->disable_lock);
2801 + raw_spin_unlock(&hmaster->disable_lock);
2804 static void count_max_concurr(int stat, struct bau_control *bcp,
2805 @@ -869,7 +869,7 @@ static void record_send_stats(cycles_t time1, cycles_t time2,
2807 static void uv1_throttle(struct bau_control *hmaster, struct ptc_stats *stat)
2809 - spinlock_t *lock = &hmaster->uvhub_lock;
2810 + raw_spinlock_t *lock = &hmaster->uvhub_lock;
2813 v = &hmaster->active_descriptor_count;
2814 @@ -1002,7 +1002,7 @@ static int check_enable(struct bau_control *bcp, struct ptc_stats *stat)
2815 struct bau_control *hmaster;
2817 hmaster = bcp->uvhub_master;
2818 - spin_lock(&hmaster->disable_lock);
2819 + raw_spin_lock(&hmaster->disable_lock);
2820 if (bcp->baudisabled && (get_cycles() >= bcp->set_bau_on_time)) {
2821 stat->s_bau_reenabled++;
2822 for_each_present_cpu(tcpu) {
2823 @@ -1014,10 +1014,10 @@ static int check_enable(struct bau_control *bcp, struct ptc_stats *stat)
2824 tbcp->period_giveups = 0;
2827 - spin_unlock(&hmaster->disable_lock);
2828 + raw_spin_unlock(&hmaster->disable_lock);
2831 - spin_unlock(&hmaster->disable_lock);
2832 + raw_spin_unlock(&hmaster->disable_lock);
2836 @@ -1940,9 +1940,9 @@ static void __init init_per_cpu_tunables(void)
2837 bcp->cong_reps = congested_reps;
2838 bcp->disabled_period = sec_2_cycles(disabled_period);
2839 bcp->giveup_limit = giveup_limit;
2840 - spin_lock_init(&bcp->queue_lock);
2841 - spin_lock_init(&bcp->uvhub_lock);
2842 - spin_lock_init(&bcp->disable_lock);
2843 + raw_spin_lock_init(&bcp->queue_lock);
2844 + raw_spin_lock_init(&bcp->uvhub_lock);
2845 + raw_spin_lock_init(&bcp->disable_lock);
2849 diff --git a/arch/x86/platform/uv/uv_time.c b/arch/x86/platform/uv/uv_time.c
2850 index b333fc45f9ec..8b85916e6986 100644
2851 --- a/arch/x86/platform/uv/uv_time.c
2852 +++ b/arch/x86/platform/uv/uv_time.c
2853 @@ -57,7 +57,7 @@ static DEFINE_PER_CPU(struct clock_event_device, cpu_ced);
2855 /* There is one of these allocated per node */
2856 struct uv_rtc_timer_head {
2858 + raw_spinlock_t lock;
2859 /* next cpu waiting for timer, local node relative: */
2861 /* number of cpus on this node: */
2862 @@ -177,7 +177,7 @@ static __init int uv_rtc_allocate_timers(void)
2863 uv_rtc_deallocate_timers();
2866 - spin_lock_init(&head->lock);
2867 + raw_spin_lock_init(&head->lock);
2868 head->ncpus = uv_blade_nr_possible_cpus(bid);
2869 head->next_cpu = -1;
2870 blade_info[bid] = head;
2871 @@ -231,7 +231,7 @@ static int uv_rtc_set_timer(int cpu, u64 expires)
2872 unsigned long flags;
2875 - spin_lock_irqsave(&head->lock, flags);
2876 + raw_spin_lock_irqsave(&head->lock, flags);
2878 next_cpu = head->next_cpu;
2880 @@ -243,12 +243,12 @@ static int uv_rtc_set_timer(int cpu, u64 expires)
2881 if (uv_setup_intr(cpu, expires)) {
2883 uv_rtc_find_next_timer(head, pnode);
2884 - spin_unlock_irqrestore(&head->lock, flags);
2885 + raw_spin_unlock_irqrestore(&head->lock, flags);
2890 - spin_unlock_irqrestore(&head->lock, flags);
2891 + raw_spin_unlock_irqrestore(&head->lock, flags);
2895 @@ -267,7 +267,7 @@ static int uv_rtc_unset_timer(int cpu, int force)
2896 unsigned long flags;
2899 - spin_lock_irqsave(&head->lock, flags);
2900 + raw_spin_lock_irqsave(&head->lock, flags);
2902 if ((head->next_cpu == bcpu && uv_read_rtc(NULL) >= *t) || force)
2904 @@ -279,7 +279,7 @@ static int uv_rtc_unset_timer(int cpu, int force)
2905 uv_rtc_find_next_timer(head, pnode);
2908 - spin_unlock_irqrestore(&head->lock, flags);
2909 + raw_spin_unlock_irqrestore(&head->lock, flags);
2913 @@ -299,13 +299,18 @@ static int uv_rtc_unset_timer(int cpu, int force)
2914 static cycle_t uv_read_rtc(struct clocksource *cs)
2916 unsigned long offset;
2919 + preempt_disable();
2920 if (uv_get_min_hub_revision_id() == 1)
2923 offset = (uv_blade_processor_id() * L1_CACHE_BYTES) % PAGE_SIZE;
2925 - return (cycle_t)uv_read_local_mmr(UVH_RTC | offset);
2926 + cycles = (cycle_t)uv_read_local_mmr(UVH_RTC | offset);
2933 diff --git a/block/blk-core.c b/block/blk-core.c
2934 index 14d7c0740dc0..dfd905bea77c 100644
2935 --- a/block/blk-core.c
2936 +++ b/block/blk-core.c
2937 @@ -125,6 +125,9 @@ void blk_rq_init(struct request_queue *q, struct request *rq)
2939 INIT_LIST_HEAD(&rq->queuelist);
2940 INIT_LIST_HEAD(&rq->timeout_list);
2941 +#ifdef CONFIG_PREEMPT_RT_FULL
2942 + INIT_WORK(&rq->work, __blk_mq_complete_request_remote_work);
2946 rq->__sector = (sector_t) -1;
2947 @@ -233,7 +236,7 @@ EXPORT_SYMBOL(blk_start_queue_async);
2949 void blk_start_queue(struct request_queue *q)
2951 - WARN_ON(!irqs_disabled());
2952 + WARN_ON_NONRT(!irqs_disabled());
2954 queue_flag_clear(QUEUE_FLAG_STOPPED, q);
2956 @@ -659,7 +662,7 @@ int blk_queue_enter(struct request_queue *q, bool nowait)
2960 - ret = wait_event_interruptible(q->mq_freeze_wq,
2961 + ret = swait_event_interruptible(q->mq_freeze_wq,
2962 !atomic_read(&q->mq_freeze_depth) ||
2963 blk_queue_dying(q));
2964 if (blk_queue_dying(q))
2965 @@ -679,7 +682,7 @@ static void blk_queue_usage_counter_release(struct percpu_ref *ref)
2966 struct request_queue *q =
2967 container_of(ref, struct request_queue, q_usage_counter);
2969 - wake_up_all(&q->mq_freeze_wq);
2970 + swake_up_all(&q->mq_freeze_wq);
2973 static void blk_rq_timed_out_timer(unsigned long data)
2974 @@ -748,7 +751,7 @@ struct request_queue *blk_alloc_queue_node(gfp_t gfp_mask, int node_id)
2975 q->bypass_depth = 1;
2976 __set_bit(QUEUE_FLAG_BYPASS, &q->queue_flags);
2978 - init_waitqueue_head(&q->mq_freeze_wq);
2979 + init_swait_queue_head(&q->mq_freeze_wq);
2982 * Init percpu_ref in atomic mode so that it's faster to shutdown.
2983 @@ -3177,7 +3180,7 @@ static void queue_unplugged(struct request_queue *q, unsigned int depth,
2984 blk_run_queue_async(q);
2987 - spin_unlock(q->queue_lock);
2988 + spin_unlock_irq(q->queue_lock);
2991 static void flush_plug_callbacks(struct blk_plug *plug, bool from_schedule)
2992 @@ -3225,7 +3228,6 @@ EXPORT_SYMBOL(blk_check_plugged);
2993 void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
2995 struct request_queue *q;
2996 - unsigned long flags;
3000 @@ -3245,11 +3247,6 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
3005 - * Save and disable interrupts here, to avoid doing it for every
3006 - * queue lock we have to take.
3008 - local_irq_save(flags);
3009 while (!list_empty(&list)) {
3010 rq = list_entry_rq(list.next);
3011 list_del_init(&rq->queuelist);
3012 @@ -3262,7 +3259,7 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
3013 queue_unplugged(q, depth, from_schedule);
3016 - spin_lock(q->queue_lock);
3017 + spin_lock_irq(q->queue_lock);
3021 @@ -3289,8 +3286,6 @@ void blk_flush_plug_list(struct blk_plug *plug, bool from_schedule)
3024 queue_unplugged(q, depth, from_schedule);
3026 - local_irq_restore(flags);
3029 void blk_finish_plug(struct blk_plug *plug)
3030 diff --git a/block/blk-ioc.c b/block/blk-ioc.c
3031 index 381cb50a673c..dc8785233d94 100644
3032 --- a/block/blk-ioc.c
3033 +++ b/block/blk-ioc.c
3035 #include <linux/bio.h>
3036 #include <linux/blkdev.h>
3037 #include <linux/slab.h>
3038 +#include <linux/delay.h>
3042 @@ -109,7 +110,7 @@ static void ioc_release_fn(struct work_struct *work)
3043 spin_unlock(q->queue_lock);
3045 spin_unlock_irqrestore(&ioc->lock, flags);
3048 spin_lock_irqsave_nested(&ioc->lock, flags, 1);
3051 @@ -187,7 +188,7 @@ void put_io_context_active(struct io_context *ioc)
3052 spin_unlock(icq->q->queue_lock);
3054 spin_unlock_irqrestore(&ioc->lock, flags);
3060 diff --git a/block/blk-mq.c b/block/blk-mq.c
3061 index 81caceb96c3c..b12b0ab005a9 100644
3062 --- a/block/blk-mq.c
3063 +++ b/block/blk-mq.c
3064 @@ -72,7 +72,7 @@ EXPORT_SYMBOL_GPL(blk_mq_freeze_queue_start);
3066 static void blk_mq_freeze_queue_wait(struct request_queue *q)
3068 - wait_event(q->mq_freeze_wq, percpu_ref_is_zero(&q->q_usage_counter));
3069 + swait_event(q->mq_freeze_wq, percpu_ref_is_zero(&q->q_usage_counter));
3073 @@ -110,7 +110,7 @@ void blk_mq_unfreeze_queue(struct request_queue *q)
3074 WARN_ON_ONCE(freeze_depth < 0);
3075 if (!freeze_depth) {
3076 percpu_ref_reinit(&q->q_usage_counter);
3077 - wake_up_all(&q->mq_freeze_wq);
3078 + swake_up_all(&q->mq_freeze_wq);
3081 EXPORT_SYMBOL_GPL(blk_mq_unfreeze_queue);
3082 @@ -129,7 +129,7 @@ void blk_mq_wake_waiters(struct request_queue *q)
3083 * dying, we need to ensure that processes currently waiting on
3084 * the queue are notified as well.
3086 - wake_up_all(&q->mq_freeze_wq);
3087 + swake_up_all(&q->mq_freeze_wq);
3090 bool blk_mq_can_queue(struct blk_mq_hw_ctx *hctx)
3091 @@ -177,6 +177,9 @@ static void blk_mq_rq_ctx_init(struct request_queue *q, struct blk_mq_ctx *ctx,
3095 +#ifdef CONFIG_PREEMPT_RT_FULL
3096 + INIT_WORK(&rq->work, __blk_mq_complete_request_remote_work);
3098 INIT_LIST_HEAD(&rq->timeout_list);
3101 @@ -345,6 +348,17 @@ void blk_mq_end_request(struct request *rq, int error)
3103 EXPORT_SYMBOL(blk_mq_end_request);
3105 +#ifdef CONFIG_PREEMPT_RT_FULL
3107 +void __blk_mq_complete_request_remote_work(struct work_struct *work)
3109 + struct request *rq = container_of(work, struct request, work);
3111 + rq->q->softirq_done_fn(rq);
3116 static void __blk_mq_complete_request_remote(void *data)
3118 struct request *rq = data;
3119 @@ -352,6 +366,8 @@ static void __blk_mq_complete_request_remote(void *data)
3120 rq->q->softirq_done_fn(rq);
3125 static void blk_mq_ipi_complete_request(struct request *rq)
3127 struct blk_mq_ctx *ctx = rq->mq_ctx;
3128 @@ -363,19 +379,23 @@ static void blk_mq_ipi_complete_request(struct request *rq)
3133 + cpu = get_cpu_light();
3134 if (!test_bit(QUEUE_FLAG_SAME_FORCE, &rq->q->queue_flags))
3135 shared = cpus_share_cache(cpu, ctx->cpu);
3137 if (cpu != ctx->cpu && !shared && cpu_online(ctx->cpu)) {
3138 +#ifdef CONFIG_PREEMPT_RT_FULL
3139 + schedule_work_on(ctx->cpu, &rq->work);
3141 rq->csd.func = __blk_mq_complete_request_remote;
3144 smp_call_function_single_async(ctx->cpu, &rq->csd);
3147 rq->q->softirq_done_fn(rq);
3153 static void __blk_mq_complete_request(struct request *rq)
3154 @@ -915,14 +935,14 @@ void blk_mq_run_hw_queue(struct blk_mq_hw_ctx *hctx, bool async)
3157 if (!async && !(hctx->flags & BLK_MQ_F_BLOCKING)) {
3158 - int cpu = get_cpu();
3159 + int cpu = get_cpu_light();
3160 if (cpumask_test_cpu(cpu, hctx->cpumask)) {
3161 __blk_mq_run_hw_queue(hctx);
3171 kblockd_schedule_work_on(blk_mq_hctx_next_cpu(hctx), &hctx->run_work);
3172 diff --git a/block/blk-mq.h b/block/blk-mq.h
3173 index e5d25249028c..1e846b842eab 100644
3174 --- a/block/blk-mq.h
3175 +++ b/block/blk-mq.h
3176 @@ -72,12 +72,12 @@ static inline struct blk_mq_ctx *__blk_mq_get_ctx(struct request_queue *q,
3178 static inline struct blk_mq_ctx *blk_mq_get_ctx(struct request_queue *q)
3180 - return __blk_mq_get_ctx(q, get_cpu());
3181 + return __blk_mq_get_ctx(q, get_cpu_light());
3184 static inline void blk_mq_put_ctx(struct blk_mq_ctx *ctx)
3190 struct blk_mq_alloc_data {
3191 diff --git a/block/blk-softirq.c b/block/blk-softirq.c
3192 index 06cf9807f49a..c40342643ca0 100644
3193 --- a/block/blk-softirq.c
3194 +++ b/block/blk-softirq.c
3195 @@ -51,6 +51,7 @@ static void trigger_softirq(void *data)
3196 raise_softirq_irqoff(BLOCK_SOFTIRQ);
3198 local_irq_restore(flags);
3199 + preempt_check_resched_rt();
3203 @@ -89,6 +90,7 @@ static int blk_softirq_cpu_dead(unsigned int cpu)
3204 this_cpu_ptr(&blk_cpu_done));
3205 raise_softirq_irqoff(BLOCK_SOFTIRQ);
3207 + preempt_check_resched_rt();
3211 @@ -141,6 +143,7 @@ void __blk_complete_request(struct request *req)
3214 local_irq_restore(flags);
3215 + preempt_check_resched_rt();
3219 diff --git a/block/bounce.c b/block/bounce.c
3220 index 1cb5dd3a5da1..2f1ec8a67cbe 100644
3221 --- a/block/bounce.c
3222 +++ b/block/bounce.c
3223 @@ -55,11 +55,11 @@ static void bounce_copy_vec(struct bio_vec *to, unsigned char *vfrom)
3224 unsigned long flags;
3227 - local_irq_save(flags);
3228 + local_irq_save_nort(flags);
3229 vto = kmap_atomic(to->bv_page);
3230 memcpy(vto + to->bv_offset, vfrom, to->bv_len);
3232 - local_irq_restore(flags);
3233 + local_irq_restore_nort(flags);
3236 #else /* CONFIG_HIGHMEM */
3237 diff --git a/crypto/algapi.c b/crypto/algapi.c
3238 index 1fad2a6b3bbb..ecb7315426a9 100644
3239 --- a/crypto/algapi.c
3240 +++ b/crypto/algapi.c
3241 @@ -719,13 +719,13 @@ EXPORT_SYMBOL_GPL(crypto_spawn_tfm2);
3243 int crypto_register_notifier(struct notifier_block *nb)
3245 - return blocking_notifier_chain_register(&crypto_chain, nb);
3246 + return srcu_notifier_chain_register(&crypto_chain, nb);
3248 EXPORT_SYMBOL_GPL(crypto_register_notifier);
3250 int crypto_unregister_notifier(struct notifier_block *nb)
3252 - return blocking_notifier_chain_unregister(&crypto_chain, nb);
3253 + return srcu_notifier_chain_unregister(&crypto_chain, nb);
3255 EXPORT_SYMBOL_GPL(crypto_unregister_notifier);
3257 diff --git a/crypto/api.c b/crypto/api.c
3258 index bbc147cb5dec..bc1a848f02ec 100644
3261 @@ -31,7 +31,7 @@ EXPORT_SYMBOL_GPL(crypto_alg_list);
3262 DECLARE_RWSEM(crypto_alg_sem);
3263 EXPORT_SYMBOL_GPL(crypto_alg_sem);
3265 -BLOCKING_NOTIFIER_HEAD(crypto_chain);
3266 +SRCU_NOTIFIER_HEAD(crypto_chain);
3267 EXPORT_SYMBOL_GPL(crypto_chain);
3269 static struct crypto_alg *crypto_larval_wait(struct crypto_alg *alg);
3270 @@ -236,10 +236,10 @@ int crypto_probing_notify(unsigned long val, void *v)
3274 - ok = blocking_notifier_call_chain(&crypto_chain, val, v);
3275 + ok = srcu_notifier_call_chain(&crypto_chain, val, v);
3276 if (ok == NOTIFY_DONE) {
3277 request_module("cryptomgr");
3278 - ok = blocking_notifier_call_chain(&crypto_chain, val, v);
3279 + ok = srcu_notifier_call_chain(&crypto_chain, val, v);
3283 diff --git a/crypto/internal.h b/crypto/internal.h
3284 index 7eefcdb00227..0ecc7f5a2f40 100644
3285 --- a/crypto/internal.h
3286 +++ b/crypto/internal.h
3287 @@ -47,7 +47,7 @@ struct crypto_larval {
3289 extern struct list_head crypto_alg_list;
3290 extern struct rw_semaphore crypto_alg_sem;
3291 -extern struct blocking_notifier_head crypto_chain;
3292 +extern struct srcu_notifier_head crypto_chain;
3294 #ifdef CONFIG_PROC_FS
3295 void __init crypto_init_proc(void);
3296 @@ -146,7 +146,7 @@ static inline int crypto_is_moribund(struct crypto_alg *alg)
3298 static inline void crypto_notify(unsigned long val, void *v)
3300 - blocking_notifier_call_chain(&crypto_chain, val, v);
3301 + srcu_notifier_call_chain(&crypto_chain, val, v);
3304 #endif /* _CRYPTO_INTERNAL_H */
3305 diff --git a/drivers/acpi/acpica/acglobal.h b/drivers/acpi/acpica/acglobal.h
3306 index 750fa824d42c..441edf51484a 100644
3307 --- a/drivers/acpi/acpica/acglobal.h
3308 +++ b/drivers/acpi/acpica/acglobal.h
3309 @@ -116,7 +116,7 @@ ACPI_GLOBAL(u8, acpi_gbl_global_lock_pending);
3312 ACPI_GLOBAL(acpi_spinlock, acpi_gbl_gpe_lock); /* For GPE data structs and registers */
3313 -ACPI_GLOBAL(acpi_spinlock, acpi_gbl_hardware_lock); /* For ACPI H/W except GPE registers */
3314 +ACPI_GLOBAL(acpi_raw_spinlock, acpi_gbl_hardware_lock); /* For ACPI H/W except GPE registers */
3315 ACPI_GLOBAL(acpi_spinlock, acpi_gbl_reference_count_lock);
3317 /* Mutex for _OSI support */
3318 diff --git a/drivers/acpi/acpica/hwregs.c b/drivers/acpi/acpica/hwregs.c
3319 index 3b7fb99362b6..696bf8e62afb 100644
3320 --- a/drivers/acpi/acpica/hwregs.c
3321 +++ b/drivers/acpi/acpica/hwregs.c
3322 @@ -363,14 +363,14 @@ acpi_status acpi_hw_clear_acpi_status(void)
3323 ACPI_BITMASK_ALL_FIXED_STATUS,
3324 ACPI_FORMAT_UINT64(acpi_gbl_xpm1a_status.address)));
3326 - lock_flags = acpi_os_acquire_lock(acpi_gbl_hardware_lock);
3327 + raw_spin_lock_irqsave(acpi_gbl_hardware_lock, lock_flags);
3329 /* Clear the fixed events in PM1 A/B */
3331 status = acpi_hw_register_write(ACPI_REGISTER_PM1_STATUS,
3332 ACPI_BITMASK_ALL_FIXED_STATUS);
3334 - acpi_os_release_lock(acpi_gbl_hardware_lock, lock_flags);
3335 + raw_spin_unlock_irqrestore(acpi_gbl_hardware_lock, lock_flags);
3337 if (ACPI_FAILURE(status)) {
3339 diff --git a/drivers/acpi/acpica/hwxface.c b/drivers/acpi/acpica/hwxface.c
3340 index 98c26ff39409..6e236f2ea791 100644
3341 --- a/drivers/acpi/acpica/hwxface.c
3342 +++ b/drivers/acpi/acpica/hwxface.c
3343 @@ -373,7 +373,7 @@ acpi_status acpi_write_bit_register(u32 register_id, u32 value)
3344 return_ACPI_STATUS(AE_BAD_PARAMETER);
3347 - lock_flags = acpi_os_acquire_lock(acpi_gbl_hardware_lock);
3348 + raw_spin_lock_irqsave(acpi_gbl_hardware_lock, lock_flags);
3351 * At this point, we know that the parent register is one of the
3352 @@ -434,7 +434,7 @@ acpi_status acpi_write_bit_register(u32 register_id, u32 value)
3356 - acpi_os_release_lock(acpi_gbl_hardware_lock, lock_flags);
3357 + raw_spin_unlock_irqrestore(acpi_gbl_hardware_lock, lock_flags);
3358 return_ACPI_STATUS(status);
3361 diff --git a/drivers/acpi/acpica/utmutex.c b/drivers/acpi/acpica/utmutex.c
3362 index 15073375bd00..357e7ca5a587 100644
3363 --- a/drivers/acpi/acpica/utmutex.c
3364 +++ b/drivers/acpi/acpica/utmutex.c
3365 @@ -88,7 +88,7 @@ acpi_status acpi_ut_mutex_initialize(void)
3366 return_ACPI_STATUS (status);
3369 - status = acpi_os_create_lock (&acpi_gbl_hardware_lock);
3370 + status = acpi_os_create_raw_lock (&acpi_gbl_hardware_lock);
3371 if (ACPI_FAILURE (status)) {
3372 return_ACPI_STATUS (status);
3374 @@ -145,7 +145,7 @@ void acpi_ut_mutex_terminate(void)
3375 /* Delete the spinlocks */
3377 acpi_os_delete_lock(acpi_gbl_gpe_lock);
3378 - acpi_os_delete_lock(acpi_gbl_hardware_lock);
3379 + acpi_os_delete_raw_lock(acpi_gbl_hardware_lock);
3380 acpi_os_delete_lock(acpi_gbl_reference_count_lock);
3382 /* Delete the reader/writer lock */
3383 diff --git a/drivers/ata/libata-sff.c b/drivers/ata/libata-sff.c
3384 index 051b6158d1b7..7ad293bef6ed 100644
3385 --- a/drivers/ata/libata-sff.c
3386 +++ b/drivers/ata/libata-sff.c
3387 @@ -678,9 +678,9 @@ unsigned int ata_sff_data_xfer_noirq(struct ata_device *dev, unsigned char *buf,
3388 unsigned long flags;
3389 unsigned int consumed;
3391 - local_irq_save(flags);
3392 + local_irq_save_nort(flags);
3393 consumed = ata_sff_data_xfer32(dev, buf, buflen, rw);
3394 - local_irq_restore(flags);
3395 + local_irq_restore_nort(flags);
3399 @@ -719,7 +719,7 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
3400 unsigned long flags;
3402 /* FIXME: use a bounce buffer */
3403 - local_irq_save(flags);
3404 + local_irq_save_nort(flags);
3405 buf = kmap_atomic(page);
3407 /* do the actual data transfer */
3408 @@ -727,7 +727,7 @@ static void ata_pio_sector(struct ata_queued_cmd *qc)
3412 - local_irq_restore(flags);
3413 + local_irq_restore_nort(flags);
3415 buf = page_address(page);
3416 ap->ops->sff_data_xfer(qc->dev, buf + offset, qc->sect_size,
3417 @@ -864,7 +864,7 @@ static int __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes)
3418 unsigned long flags;
3420 /* FIXME: use bounce buffer */
3421 - local_irq_save(flags);
3422 + local_irq_save_nort(flags);
3423 buf = kmap_atomic(page);
3425 /* do the actual data transfer */
3426 @@ -872,7 +872,7 @@ static int __atapi_pio_bytes(struct ata_queued_cmd *qc, unsigned int bytes)
3430 - local_irq_restore(flags);
3431 + local_irq_restore_nort(flags);
3433 buf = page_address(page);
3434 consumed = ap->ops->sff_data_xfer(dev, buf + offset,
3435 diff --git a/drivers/block/zram/zcomp.c b/drivers/block/zram/zcomp.c
3436 index 4b5cd3a7b2b6..fa8329ad79fd 100644
3437 --- a/drivers/block/zram/zcomp.c
3438 +++ b/drivers/block/zram/zcomp.c
3439 @@ -118,12 +118,19 @@ ssize_t zcomp_available_show(const char *comp, char *buf)
3441 struct zcomp_strm *zcomp_stream_get(struct zcomp *comp)
3443 - return *get_cpu_ptr(comp->stream);
3444 + struct zcomp_strm *zstrm;
3446 + zstrm = *this_cpu_ptr(comp->stream);
3447 + spin_lock(&zstrm->zcomp_lock);
3451 void zcomp_stream_put(struct zcomp *comp)
3453 - put_cpu_ptr(comp->stream);
3454 + struct zcomp_strm *zstrm;
3456 + zstrm = *this_cpu_ptr(comp->stream);
3457 + spin_unlock(&zstrm->zcomp_lock);
3460 int zcomp_compress(struct zcomp_strm *zstrm,
3461 @@ -174,6 +181,7 @@ static int __zcomp_cpu_notifier(struct zcomp *comp,
3462 pr_err("Can't allocate a compression stream\n");
3465 + spin_lock_init(&zstrm->zcomp_lock);
3466 *per_cpu_ptr(comp->stream, cpu) = zstrm;
3469 diff --git a/drivers/block/zram/zcomp.h b/drivers/block/zram/zcomp.h
3470 index 478cac2ed465..f7a6efdc3285 100644
3471 --- a/drivers/block/zram/zcomp.h
3472 +++ b/drivers/block/zram/zcomp.h
3473 @@ -14,6 +14,7 @@ struct zcomp_strm {
3474 /* compression/decompression buffer */
3476 struct crypto_comp *tfm;
3477 + spinlock_t zcomp_lock;
3480 /* dynamic per-device compression frontend */
3481 diff --git a/drivers/block/zram/zram_drv.c b/drivers/block/zram/zram_drv.c
3482 index d2ef51ca9cf4..05e749736560 100644
3483 --- a/drivers/block/zram/zram_drv.c
3484 +++ b/drivers/block/zram/zram_drv.c
3485 @@ -528,6 +528,8 @@ static struct zram_meta *zram_meta_alloc(char *pool_name, u64 disksize)
3489 + zram_meta_init_table_locks(meta, disksize);
3494 @@ -575,28 +577,28 @@ static int zram_decompress_page(struct zram *zram, char *mem, u32 index)
3495 struct zram_meta *meta = zram->meta;
3496 unsigned long handle;
3498 + struct zcomp_strm *zstrm;
3500 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3501 + zram_lock_table(&meta->table[index]);
3502 handle = meta->table[index].handle;
3503 size = zram_get_obj_size(meta, index);
3505 if (!handle || zram_test_flag(meta, index, ZRAM_ZERO)) {
3506 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3507 + zram_unlock_table(&meta->table[index]);
3512 + zstrm = zcomp_stream_get(zram->comp);
3513 cmem = zs_map_object(meta->mem_pool, handle, ZS_MM_RO);
3514 if (size == PAGE_SIZE) {
3515 copy_page(mem, cmem);
3517 - struct zcomp_strm *zstrm = zcomp_stream_get(zram->comp);
3519 ret = zcomp_decompress(zstrm, cmem, size, mem);
3520 - zcomp_stream_put(zram->comp);
3522 zs_unmap_object(meta->mem_pool, handle);
3523 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3524 + zcomp_stream_put(zram->comp);
3525 + zram_unlock_table(&meta->table[index]);
3527 /* Should NEVER happen. Return bio error if it does. */
3528 if (unlikely(ret)) {
3529 @@ -616,14 +618,14 @@ static int zram_bvec_read(struct zram *zram, struct bio_vec *bvec,
3530 struct zram_meta *meta = zram->meta;
3531 page = bvec->bv_page;
3533 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3534 + zram_lock_table(&meta->table[index]);
3535 if (unlikely(!meta->table[index].handle) ||
3536 zram_test_flag(meta, index, ZRAM_ZERO)) {
3537 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3538 + zram_unlock_table(&meta->table[index]);
3539 handle_zero_page(bvec);
3542 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3543 + zram_unlock_table(&meta->table[index]);
3545 if (is_partial_io(bvec))
3546 /* Use a temporary buffer to decompress the page */
3547 @@ -700,10 +702,10 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
3549 kunmap_atomic(user_mem);
3550 /* Free memory associated with this sector now. */
3551 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3552 + zram_lock_table(&meta->table[index]);
3553 zram_free_page(zram, index);
3554 zram_set_flag(meta, index, ZRAM_ZERO);
3555 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3556 + zram_unlock_table(&meta->table[index]);
3558 atomic64_inc(&zram->stats.zero_pages);
3560 @@ -794,12 +796,12 @@ static int zram_bvec_write(struct zram *zram, struct bio_vec *bvec, u32 index,
3561 * Free memory associated with this sector
3562 * before overwriting unused sectors.
3564 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3565 + zram_lock_table(&meta->table[index]);
3566 zram_free_page(zram, index);
3568 meta->table[index].handle = handle;
3569 zram_set_obj_size(meta, index, clen);
3570 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3571 + zram_unlock_table(&meta->table[index]);
3574 atomic64_add(clen, &zram->stats.compr_data_size);
3575 @@ -842,9 +844,9 @@ static void zram_bio_discard(struct zram *zram, u32 index,
3578 while (n >= PAGE_SIZE) {
3579 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3580 + zram_lock_table(&meta->table[index]);
3581 zram_free_page(zram, index);
3582 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3583 + zram_unlock_table(&meta->table[index]);
3584 atomic64_inc(&zram->stats.notify_free);
3587 @@ -973,9 +975,9 @@ static void zram_slot_free_notify(struct block_device *bdev,
3588 zram = bdev->bd_disk->private_data;
3591 - bit_spin_lock(ZRAM_ACCESS, &meta->table[index].value);
3592 + zram_lock_table(&meta->table[index]);
3593 zram_free_page(zram, index);
3594 - bit_spin_unlock(ZRAM_ACCESS, &meta->table[index].value);
3595 + zram_unlock_table(&meta->table[index]);
3596 atomic64_inc(&zram->stats.notify_free);
3599 diff --git a/drivers/block/zram/zram_drv.h b/drivers/block/zram/zram_drv.h
3600 index 74fcf10da374..fd4020c99b9e 100644
3601 --- a/drivers/block/zram/zram_drv.h
3602 +++ b/drivers/block/zram/zram_drv.h
3603 @@ -73,6 +73,9 @@ enum zram_pageflags {
3604 struct zram_table_entry {
3605 unsigned long handle;
3606 unsigned long value;
3607 +#ifdef CONFIG_PREEMPT_RT_BASE
3613 @@ -120,4 +123,42 @@ struct zram {
3615 bool claim; /* Protected by bdev->bd_mutex */
3618 +#ifndef CONFIG_PREEMPT_RT_BASE
3619 +static inline void zram_lock_table(struct zram_table_entry *table)
3621 + bit_spin_lock(ZRAM_ACCESS, &table->value);
3624 +static inline void zram_unlock_table(struct zram_table_entry *table)
3626 + bit_spin_unlock(ZRAM_ACCESS, &table->value);
3629 +static inline void zram_meta_init_table_locks(struct zram_meta *meta, u64 disksize) { }
3630 +#else /* CONFIG_PREEMPT_RT_BASE */
3631 +static inline void zram_lock_table(struct zram_table_entry *table)
3633 + spin_lock(&table->lock);
3634 + __set_bit(ZRAM_ACCESS, &table->value);
3637 +static inline void zram_unlock_table(struct zram_table_entry *table)
3639 + __clear_bit(ZRAM_ACCESS, &table->value);
3640 + spin_unlock(&table->lock);
3643 +static inline void zram_meta_init_table_locks(struct zram_meta *meta, u64 disksize)
3645 + size_t num_pages = disksize >> PAGE_SHIFT;
3648 + for (index = 0; index < num_pages; index++) {
3649 + spinlock_t *lock = &meta->table[index].lock;
3650 + spin_lock_init(lock);
3653 +#endif /* CONFIG_PREEMPT_RT_BASE */
3656 diff --git a/drivers/char/random.c b/drivers/char/random.c
3657 index d6876d506220..0c60b1e54579 100644
3658 --- a/drivers/char/random.c
3659 +++ b/drivers/char/random.c
3660 @@ -1028,8 +1028,6 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
3662 long delta, delta2, delta3;
3664 - preempt_disable();
3666 sample.jiffies = jiffies;
3667 sample.cycles = random_get_entropy();
3669 @@ -1070,7 +1068,6 @@ static void add_timer_randomness(struct timer_rand_state *state, unsigned num)
3671 credit_entropy_bits(r, min_t(int, fls(delta>>1), 11));
3676 void add_input_randomness(unsigned int type, unsigned int code,
3677 @@ -1123,28 +1120,27 @@ static __u32 get_reg(struct fast_pool *f, struct pt_regs *regs)
3678 return *(ptr + f->reg_idx++);
3681 -void add_interrupt_randomness(int irq, int irq_flags)
3682 +void add_interrupt_randomness(int irq, int irq_flags, __u64 ip)
3684 struct entropy_store *r;
3685 struct fast_pool *fast_pool = this_cpu_ptr(&irq_randomness);
3686 - struct pt_regs *regs = get_irq_regs();
3687 unsigned long now = jiffies;
3688 cycles_t cycles = random_get_entropy();
3689 __u32 c_high, j_high;
3695 - cycles = get_reg(fast_pool, regs);
3696 + cycles = get_reg(fast_pool, NULL);
3697 c_high = (sizeof(cycles) > 4) ? cycles >> 32 : 0;
3698 j_high = (sizeof(now) > 4) ? now >> 32 : 0;
3699 fast_pool->pool[0] ^= cycles ^ j_high ^ irq;
3700 fast_pool->pool[1] ^= now ^ c_high;
3701 - ip = regs ? instruction_pointer(regs) : _RET_IP_;
3704 fast_pool->pool[2] ^= ip;
3705 fast_pool->pool[3] ^= (sizeof(ip) > 4) ? ip >> 32 :
3706 - get_reg(fast_pool, regs);
3707 + get_reg(fast_pool, NULL);
3709 fast_mix(fast_pool);
3710 add_interrupt_bench(cycles);
3711 diff --git a/drivers/clocksource/tcb_clksrc.c b/drivers/clocksource/tcb_clksrc.c
3712 index 4da2af9694a2..5b6f57f500b8 100644
3713 --- a/drivers/clocksource/tcb_clksrc.c
3714 +++ b/drivers/clocksource/tcb_clksrc.c
3716 * this 32 bit free-running counter. the second channel is not used.
3718 * - The third channel may be used to provide a 16-bit clockevent
3719 - * source, used in either periodic or oneshot mode. This runs
3720 - * at 32 KiHZ, and can handle delays of up to two seconds.
3721 + * source, used in either periodic or oneshot mode.
3723 * A boot clocksource and clockevent source are also currently needed,
3724 * unless the relevant platforms (ARM/AT91, AVR32/AT32) are changed so
3725 @@ -74,6 +73,8 @@ static struct clocksource clksrc = {
3726 struct tc_clkevt_device {
3727 struct clock_event_device clkevt;
3734 @@ -82,15 +83,26 @@ static struct tc_clkevt_device *to_tc_clkevt(struct clock_event_device *clkevt)
3735 return container_of(clkevt, struct tc_clkevt_device, clkevt);
3738 -/* For now, we always use the 32K clock ... this optimizes for NO_HZ,
3739 - * because using one of the divided clocks would usually mean the
3740 - * tick rate can never be less than several dozen Hz (vs 0.5 Hz).
3742 - * A divided clock could be good for high resolution timers, since
3743 - * 30.5 usec resolution can seem "low".
3745 static u32 timer_clock;
3747 +static void tc_clk_disable(struct clock_event_device *d)
3749 + struct tc_clkevt_device *tcd = to_tc_clkevt(d);
3751 + clk_disable(tcd->clk);
3752 + tcd->clk_enabled = false;
3755 +static void tc_clk_enable(struct clock_event_device *d)
3757 + struct tc_clkevt_device *tcd = to_tc_clkevt(d);
3759 + if (tcd->clk_enabled)
3761 + clk_enable(tcd->clk);
3762 + tcd->clk_enabled = true;
3765 static int tc_shutdown(struct clock_event_device *d)
3767 struct tc_clkevt_device *tcd = to_tc_clkevt(d);
3768 @@ -98,8 +110,14 @@ static int tc_shutdown(struct clock_event_device *d)
3770 __raw_writel(0xff, regs + ATMEL_TC_REG(2, IDR));
3771 __raw_writel(ATMEL_TC_CLKDIS, regs + ATMEL_TC_REG(2, CCR));
3775 +static int tc_shutdown_clk_off(struct clock_event_device *d)
3778 if (!clockevent_state_detached(d))
3779 - clk_disable(tcd->clk);
3780 + tc_clk_disable(d);
3784 @@ -112,9 +130,9 @@ static int tc_set_oneshot(struct clock_event_device *d)
3785 if (clockevent_state_oneshot(d) || clockevent_state_periodic(d))
3788 - clk_enable(tcd->clk);
3791 - /* slow clock, count up to RC, then irq and stop */
3792 + /* count up to RC, then irq and stop */
3793 __raw_writel(timer_clock | ATMEL_TC_CPCSTOP | ATMEL_TC_WAVE |
3794 ATMEL_TC_WAVESEL_UP_AUTO, regs + ATMEL_TC_REG(2, CMR));
3795 __raw_writel(ATMEL_TC_CPCS, regs + ATMEL_TC_REG(2, IER));
3796 @@ -134,12 +152,12 @@ static int tc_set_periodic(struct clock_event_device *d)
3797 /* By not making the gentime core emulate periodic mode on top
3798 * of oneshot, we get lower overhead and improved accuracy.
3800 - clk_enable(tcd->clk);
3803 - /* slow clock, count up to RC, then irq and restart */
3804 + /* count up to RC, then irq and restart */
3805 __raw_writel(timer_clock | ATMEL_TC_WAVE | ATMEL_TC_WAVESEL_UP_AUTO,
3806 regs + ATMEL_TC_REG(2, CMR));
3807 - __raw_writel((32768 + HZ / 2) / HZ, tcaddr + ATMEL_TC_REG(2, RC));
3808 + __raw_writel((tcd->freq + HZ / 2) / HZ, tcaddr + ATMEL_TC_REG(2, RC));
3810 /* Enable clock and interrupts on RC compare */
3811 __raw_writel(ATMEL_TC_CPCS, regs + ATMEL_TC_REG(2, IER));
3812 @@ -166,9 +184,13 @@ static struct tc_clkevt_device clkevt = {
3813 .features = CLOCK_EVT_FEAT_PERIODIC |
3814 CLOCK_EVT_FEAT_ONESHOT,
3815 /* Should be lower than at91rm9200's system timer */
3816 +#ifdef CONFIG_ATMEL_TCB_CLKSRC_USE_SLOW_CLOCK
3821 .set_next_event = tc_next_event,
3822 - .set_state_shutdown = tc_shutdown,
3823 + .set_state_shutdown = tc_shutdown_clk_off,
3824 .set_state_periodic = tc_set_periodic,
3825 .set_state_oneshot = tc_set_oneshot,
3827 @@ -188,8 +210,9 @@ static irqreturn_t ch2_irq(int irq, void *handle)
3831 -static int __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
3832 +static int __init setup_clkevents(struct atmel_tc *tc, int divisor_idx)
3834 + unsigned divisor = atmel_tc_divisors[divisor_idx];
3836 struct clk *t2_clk = tc->clk[2];
3837 int irq = tc->irq[2];
3838 @@ -210,7 +233,11 @@ static int __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
3839 clkevt.regs = tc->regs;
3840 clkevt.clk = t2_clk;
3842 - timer_clock = clk32k_divisor_idx;
3843 + timer_clock = divisor_idx;
3845 + clkevt.freq = 32768;
3847 + clkevt.freq = clk_get_rate(t2_clk) / divisor;
3849 clkevt.clkevt.cpumask = cpumask_of(0);
3851 @@ -221,7 +248,7 @@ static int __init setup_clkevents(struct atmel_tc *tc, int clk32k_divisor_idx)
3855 - clockevents_config_and_register(&clkevt.clkevt, 32768, 1, 0xffff);
3856 + clockevents_config_and_register(&clkevt.clkevt, clkevt.freq, 1, 0xffff);
3860 @@ -358,7 +385,11 @@ static int __init tcb_clksrc_init(void)
3861 goto err_disable_t1;
3863 /* channel 2: periodic and oneshot timer support */
3864 +#ifdef CONFIG_ATMEL_TCB_CLKSRC_USE_SLOW_CLOCK
3865 ret = setup_clkevents(tc, clk32k_divisor_idx);
3867 + ret = setup_clkevents(tc, best_divisor_idx);
3870 goto err_unregister_clksrc;
3872 diff --git a/drivers/clocksource/timer-atmel-pit.c b/drivers/clocksource/timer-atmel-pit.c
3873 index 6555821bbdae..93288849b2bd 100644
3874 --- a/drivers/clocksource/timer-atmel-pit.c
3875 +++ b/drivers/clocksource/timer-atmel-pit.c
3876 @@ -46,6 +46,7 @@ struct pit_data {
3880 + bool irq_requested;
3884 @@ -96,15 +97,29 @@ static int pit_clkevt_shutdown(struct clock_event_device *dev)
3886 /* disable irq, leaving the clocksource active */
3887 pit_write(data->base, AT91_PIT_MR, (data->cycle - 1) | AT91_PIT_PITEN);
3888 + if (data->irq_requested) {
3889 + free_irq(data->irq, data);
3890 + data->irq_requested = false;
3895 +static irqreturn_t at91sam926x_pit_interrupt(int irq, void *dev_id);
3897 * Clockevent device: interrupts every 1/HZ (== pit_cycles * MCK/16)
3899 static int pit_clkevt_set_periodic(struct clock_event_device *dev)
3901 struct pit_data *data = clkevt_to_pit_data(dev);
3904 + ret = request_irq(data->irq, at91sam926x_pit_interrupt,
3905 + IRQF_SHARED | IRQF_TIMER | IRQF_IRQPOLL,
3906 + "at91_tick", data);
3908 + panic(pr_fmt("Unable to setup IRQ\n"));
3910 + data->irq_requested = true;
3912 /* update clocksource counter */
3913 data->cnt += data->cycle * PIT_PICNT(pit_read(data->base, AT91_PIT_PIVR));
3914 @@ -230,15 +245,6 @@ static int __init at91sam926x_pit_dt_init(struct device_node *node)
3918 - /* Set up irq handler */
3919 - ret = request_irq(data->irq, at91sam926x_pit_interrupt,
3920 - IRQF_SHARED | IRQF_TIMER | IRQF_IRQPOLL,
3921 - "at91_tick", data);
3923 - pr_err("Unable to setup IRQ\n");
3927 /* Set up and register clockevents */
3928 data->clkevt.name = "pit";
3929 data->clkevt.features = CLOCK_EVT_FEAT_PERIODIC;
3930 diff --git a/drivers/clocksource/timer-atmel-st.c b/drivers/clocksource/timer-atmel-st.c
3931 index e90ab5b63a90..9e124087c55f 100644
3932 --- a/drivers/clocksource/timer-atmel-st.c
3933 +++ b/drivers/clocksource/timer-atmel-st.c
3934 @@ -115,18 +115,29 @@ static void clkdev32k_disable_and_flush_irq(void)
3935 last_crtr = read_CRTR();
3938 +static int atmel_st_irq;
3940 static int clkevt32k_shutdown(struct clock_event_device *evt)
3942 clkdev32k_disable_and_flush_irq();
3944 regmap_write(regmap_st, AT91_ST_IER, irqmask);
3945 + free_irq(atmel_st_irq, regmap_st);
3949 static int clkevt32k_set_oneshot(struct clock_event_device *dev)
3953 clkdev32k_disable_and_flush_irq();
3955 + ret = request_irq(atmel_st_irq, at91rm9200_timer_interrupt,
3956 + IRQF_SHARED | IRQF_TIMER | IRQF_IRQPOLL,
3957 + "at91_tick", regmap_st);
3959 + panic(pr_fmt("Unable to setup IRQ\n"));
3962 * ALM for oneshot irqs, set by next_event()
3963 * before 32 seconds have passed.
3964 @@ -139,8 +150,16 @@ static int clkevt32k_set_oneshot(struct clock_event_device *dev)
3966 static int clkevt32k_set_periodic(struct clock_event_device *dev)
3970 clkdev32k_disable_and_flush_irq();
3972 + ret = request_irq(atmel_st_irq, at91rm9200_timer_interrupt,
3973 + IRQF_SHARED | IRQF_TIMER | IRQF_IRQPOLL,
3974 + "at91_tick", regmap_st);
3976 + panic(pr_fmt("Unable to setup IRQ\n"));
3978 /* PIT for periodic irqs; fixed rate of 1/HZ */
3979 irqmask = AT91_ST_PITS;
3980 regmap_write(regmap_st, AT91_ST_PIMR, timer_latch);
3981 @@ -198,7 +217,7 @@ static int __init atmel_st_timer_init(struct device_node *node)
3984 unsigned int sclk_rate, val;
3988 regmap_st = syscon_node_to_regmap(node);
3989 if (IS_ERR(regmap_st)) {
3990 @@ -212,21 +231,12 @@ static int __init atmel_st_timer_init(struct device_node *node)
3991 regmap_read(regmap_st, AT91_ST_SR, &val);
3993 /* Get the interrupts property */
3994 - irq = irq_of_parse_and_map(node, 0);
3996 + atmel_st_irq = irq_of_parse_and_map(node, 0);
3997 + if (!atmel_st_irq) {
3998 pr_err("Unable to get IRQ from DT\n");
4002 - /* Make IRQs happen for the system timer */
4003 - ret = request_irq(irq, at91rm9200_timer_interrupt,
4004 - IRQF_SHARED | IRQF_TIMER | IRQF_IRQPOLL,
4005 - "at91_tick", regmap_st);
4007 - pr_err("Unable to setup IRQ\n");
4011 sclk = of_clk_get(node, 0);
4013 pr_err("Unable to get slow clock\n");
4014 diff --git a/drivers/connector/cn_proc.c b/drivers/connector/cn_proc.c
4015 index a782ce87715c..19d265948526 100644
4016 --- a/drivers/connector/cn_proc.c
4017 +++ b/drivers/connector/cn_proc.c
4019 #include <linux/pid_namespace.h>
4021 #include <linux/cn_proc.h>
4022 +#include <linux/locallock.h>
4025 * Size of a cn_msg followed by a proc_event structure. Since the
4026 @@ -54,10 +55,11 @@ static struct cb_id cn_proc_event_id = { CN_IDX_PROC, CN_VAL_PROC };
4028 /* proc_event_counts is used as the sequence number of the netlink message */
4029 static DEFINE_PER_CPU(__u32, proc_event_counts) = { 0 };
4030 +static DEFINE_LOCAL_IRQ_LOCK(send_msg_lock);
4032 static inline void send_msg(struct cn_msg *msg)
4034 - preempt_disable();
4035 + local_lock(send_msg_lock);
4037 msg->seq = __this_cpu_inc_return(proc_event_counts) - 1;
4038 ((struct proc_event *)msg->data)->cpu = smp_processor_id();
4039 @@ -70,7 +72,7 @@ static inline void send_msg(struct cn_msg *msg)
4041 cn_netlink_send(msg, 0, CN_IDX_PROC, GFP_NOWAIT);
4044 + local_unlock(send_msg_lock);
4047 void proc_fork_connector(struct task_struct *task)
4048 diff --git a/drivers/cpufreq/Kconfig.x86 b/drivers/cpufreq/Kconfig.x86
4049 index adbd1de1cea5..1fac5074f2cf 100644
4050 --- a/drivers/cpufreq/Kconfig.x86
4051 +++ b/drivers/cpufreq/Kconfig.x86
4052 @@ -124,7 +124,7 @@ config X86_POWERNOW_K7_ACPI
4054 config X86_POWERNOW_K8
4055 tristate "AMD Opteron/Athlon64 PowerNow!"
4056 - depends on ACPI && ACPI_PROCESSOR && X86_ACPI_CPUFREQ
4057 + depends on ACPI && ACPI_PROCESSOR && X86_ACPI_CPUFREQ && !PREEMPT_RT_BASE
4059 This adds the CPUFreq driver for K8/early Opteron/Athlon64 processors.
4060 Support for K10 and newer processors is now in acpi-cpufreq.
4061 diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
4062 index 0c400f852a76..97d5f6193751 100644
4063 --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
4064 +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
4065 @@ -1537,7 +1537,9 @@ execbuf_submit(struct i915_execbuffer_params *params,
4069 +#ifndef CONFIG_PREEMPT_RT_BASE
4070 trace_i915_gem_ring_dispatch(params->request, params->dispatch_flags);
4073 i915_gem_execbuffer_move_to_active(vmas, params->request);
4075 diff --git a/drivers/gpu/drm/i915/i915_gem_shrinker.c b/drivers/gpu/drm/i915/i915_gem_shrinker.c
4076 index 1c237d02f30b..9e9b4404c0d7 100644
4077 --- a/drivers/gpu/drm/i915/i915_gem_shrinker.c
4078 +++ b/drivers/gpu/drm/i915/i915_gem_shrinker.c
4079 @@ -40,7 +40,7 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
4080 if (!mutex_is_locked(mutex))
4083 -#if defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_MUTEX_SPIN_ON_OWNER)
4084 +#if (defined(CONFIG_DEBUG_MUTEXES) || defined(CONFIG_MUTEX_SPIN_ON_OWNER)) && !defined(CONFIG_PREEMPT_RT_BASE)
4085 return mutex->owner == task;
4087 /* Since UP may be pre-empted, we cannot assume that we own the lock */
4088 diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
4089 index 3fc286cd1157..252a1117b103 100644
4090 --- a/drivers/gpu/drm/i915/i915_irq.c
4091 +++ b/drivers/gpu/drm/i915/i915_irq.c
4092 @@ -812,6 +812,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
4093 spin_lock_irqsave(&dev_priv->uncore.lock, irqflags);
4095 /* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
4096 + preempt_disable_rt();
4098 /* Get optional system timestamp before query. */
4100 @@ -863,6 +864,7 @@ static int i915_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
4101 *etime = ktime_get();
4103 /* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
4104 + preempt_enable_rt();
4106 spin_unlock_irqrestore(&dev_priv->uncore.lock, irqflags);
4108 diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
4109 index b9be8a6141d8..3162feddabe8 100644
4110 --- a/drivers/gpu/drm/i915/intel_display.c
4111 +++ b/drivers/gpu/drm/i915/intel_display.c
4112 @@ -12141,7 +12141,7 @@ void intel_check_page_flip(struct drm_i915_private *dev_priv, int pipe)
4113 struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
4114 struct intel_flip_work *work;
4116 - WARN_ON(!in_interrupt());
4117 + WARN_ON_NONRT(!in_interrupt());
4121 diff --git a/drivers/gpu/drm/i915/intel_sprite.c b/drivers/gpu/drm/i915/intel_sprite.c
4122 index dbed12c484c9..5c540b78e8b5 100644
4123 --- a/drivers/gpu/drm/i915/intel_sprite.c
4124 +++ b/drivers/gpu/drm/i915/intel_sprite.c
4126 #include <drm/drm_rect.h>
4127 #include <drm/drm_atomic.h>
4128 #include <drm/drm_plane_helper.h>
4129 +#include <linux/locallock.h>
4130 #include "intel_drv.h"
4131 #include "intel_frontbuffer.h"
4132 #include <drm/i915_drm.h>
4133 @@ -65,6 +66,8 @@ int intel_usecs_to_scanlines(const struct drm_display_mode *adjusted_mode,
4134 1000 * adjusted_mode->crtc_htotal);
4137 +static DEFINE_LOCAL_IRQ_LOCK(pipe_update_lock);
4140 * intel_pipe_update_start() - start update of a set of display registers
4141 * @crtc: the crtc of which the registers are going to be updated
4142 @@ -95,7 +98,7 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
4143 min = vblank_start - intel_usecs_to_scanlines(adjusted_mode, 100);
4144 max = vblank_start - 1;
4146 - local_irq_disable();
4147 + local_lock_irq(pipe_update_lock);
4149 if (min <= 0 || max <= 0)
4151 @@ -125,11 +128,11 @@ void intel_pipe_update_start(struct intel_crtc *crtc)
4155 - local_irq_enable();
4156 + local_unlock_irq(pipe_update_lock);
4158 timeout = schedule_timeout(timeout);
4160 - local_irq_disable();
4161 + local_lock_irq(pipe_update_lock);
4164 finish_wait(wq, &wait);
4165 @@ -181,7 +184,7 @@ void intel_pipe_update_end(struct intel_crtc *crtc, struct intel_flip_work *work
4166 crtc->base.state->event = NULL;
4169 - local_irq_enable();
4170 + local_unlock_irq(pipe_update_lock);
4172 if (crtc->debug.start_vbl_count &&
4173 crtc->debug.start_vbl_count != end_vbl_count) {
4174 diff --git a/drivers/gpu/drm/msm/msm_gem_shrinker.c b/drivers/gpu/drm/msm/msm_gem_shrinker.c
4175 index 192b2d3a79cb..d5372a207326 100644
4176 --- a/drivers/gpu/drm/msm/msm_gem_shrinker.c
4177 +++ b/drivers/gpu/drm/msm/msm_gem_shrinker.c
4178 @@ -23,7 +23,7 @@ static bool mutex_is_locked_by(struct mutex *mutex, struct task_struct *task)
4179 if (!mutex_is_locked(mutex))
4182 -#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_MUTEXES)
4183 +#if (defined(CONFIG_SMP) || defined(CONFIG_DEBUG_MUTEXES)) && !defined(CONFIG_PREEMPT_RT_BASE)
4184 return mutex->owner == task;
4186 /* Since UP may be pre-empted, we cannot assume that we own the lock */
4187 diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
4188 index cdb8cb568c15..b6d7fd964cbc 100644
4189 --- a/drivers/gpu/drm/radeon/radeon_display.c
4190 +++ b/drivers/gpu/drm/radeon/radeon_display.c
4191 @@ -1845,6 +1845,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
4192 struct radeon_device *rdev = dev->dev_private;
4194 /* preempt_disable_rt() should go right here in PREEMPT_RT patchset. */
4195 + preempt_disable_rt();
4197 /* Get optional system timestamp before query. */
4199 @@ -1937,6 +1938,7 @@ int radeon_get_crtc_scanoutpos(struct drm_device *dev, unsigned int pipe,
4200 *etime = ktime_get();
4202 /* preempt_enable_rt() should go right here in PREEMPT_RT patchset. */
4203 + preempt_enable_rt();
4205 /* Decode into vertical and horizontal scanout position. */
4206 *vpos = position & 0x1fff;
4207 diff --git a/drivers/hv/vmbus_drv.c b/drivers/hv/vmbus_drv.c
4208 index 0276d2ef06ee..8868045eabde 100644
4209 --- a/drivers/hv/vmbus_drv.c
4210 +++ b/drivers/hv/vmbus_drv.c
4211 @@ -761,6 +761,8 @@ static void vmbus_isr(void)
4213 struct hv_message *msg;
4214 union hv_synic_event_flags *event;
4215 + struct pt_regs *regs = get_irq_regs();
4216 + u64 ip = regs ? instruction_pointer(regs) : 0;
4217 bool handled = false;
4219 page_addr = hv_context.synic_event_page[cpu];
4220 @@ -808,7 +810,7 @@ static void vmbus_isr(void)
4221 tasklet_schedule(hv_context.msg_dpc[cpu]);
4224 - add_interrupt_randomness(HYPERVISOR_CALLBACK_VECTOR, 0);
4225 + add_interrupt_randomness(HYPERVISOR_CALLBACK_VECTOR, 0, ip);
4229 diff --git a/drivers/ide/alim15x3.c b/drivers/ide/alim15x3.c
4230 index 36f76e28a0bf..394f142f90c7 100644
4231 --- a/drivers/ide/alim15x3.c
4232 +++ b/drivers/ide/alim15x3.c
4233 @@ -234,7 +234,7 @@ static int init_chipset_ali15x3(struct pci_dev *dev)
4235 isa_dev = pci_get_device(PCI_VENDOR_ID_AL, PCI_DEVICE_ID_AL_M1533, NULL);
4237 - local_irq_save(flags);
4238 + local_irq_save_nort(flags);
4240 if (m5229_revision < 0xC2) {
4242 @@ -325,7 +325,7 @@ static int init_chipset_ali15x3(struct pci_dev *dev)
4245 pci_dev_put(isa_dev);
4246 - local_irq_restore(flags);
4247 + local_irq_restore_nort(flags);
4251 diff --git a/drivers/ide/hpt366.c b/drivers/ide/hpt366.c
4252 index 0ceae5cbd89a..c212e85d7f3e 100644
4253 --- a/drivers/ide/hpt366.c
4254 +++ b/drivers/ide/hpt366.c
4255 @@ -1236,7 +1236,7 @@ static int init_dma_hpt366(ide_hwif_t *hwif,
4257 dma_old = inb(base + 2);
4259 - local_irq_save(flags);
4260 + local_irq_save_nort(flags);
4263 pci_read_config_byte(dev, hwif->channel ? 0x4b : 0x43, &masterdma);
4264 @@ -1247,7 +1247,7 @@ static int init_dma_hpt366(ide_hwif_t *hwif,
4265 if (dma_new != dma_old)
4266 outb(dma_new, base + 2);
4268 - local_irq_restore(flags);
4269 + local_irq_restore_nort(flags);
4271 printk(KERN_INFO " %s: BM-DMA at 0x%04lx-0x%04lx\n",
4272 hwif->name, base, base + 7);
4273 diff --git a/drivers/ide/ide-io-std.c b/drivers/ide/ide-io-std.c
4274 index 19763977568c..4169433faab5 100644
4275 --- a/drivers/ide/ide-io-std.c
4276 +++ b/drivers/ide/ide-io-std.c
4277 @@ -175,7 +175,7 @@ void ide_input_data(ide_drive_t *drive, struct ide_cmd *cmd, void *buf,
4278 unsigned long uninitialized_var(flags);
4280 if ((io_32bit & 2) && !mmio) {
4281 - local_irq_save(flags);
4282 + local_irq_save_nort(flags);
4283 ata_vlb_sync(io_ports->nsect_addr);
4286 @@ -186,7 +186,7 @@ void ide_input_data(ide_drive_t *drive, struct ide_cmd *cmd, void *buf,
4287 insl(data_addr, buf, words);
4289 if ((io_32bit & 2) && !mmio)
4290 - local_irq_restore(flags);
4291 + local_irq_restore_nort(flags);
4293 if (((len + 1) & 3) < 2)
4295 @@ -219,7 +219,7 @@ void ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, void *buf,
4296 unsigned long uninitialized_var(flags);
4298 if ((io_32bit & 2) && !mmio) {
4299 - local_irq_save(flags);
4300 + local_irq_save_nort(flags);
4301 ata_vlb_sync(io_ports->nsect_addr);
4304 @@ -230,7 +230,7 @@ void ide_output_data(ide_drive_t *drive, struct ide_cmd *cmd, void *buf,
4305 outsl(data_addr, buf, words);
4307 if ((io_32bit & 2) && !mmio)
4308 - local_irq_restore(flags);
4309 + local_irq_restore_nort(flags);
4311 if (((len + 1) & 3) < 2)
4313 diff --git a/drivers/ide/ide-io.c b/drivers/ide/ide-io.c
4314 index 669ea1e45795..e12e43e62245 100644
4315 --- a/drivers/ide/ide-io.c
4316 +++ b/drivers/ide/ide-io.c
4317 @@ -659,7 +659,7 @@ void ide_timer_expiry (unsigned long data)
4318 /* disable_irq_nosync ?? */
4319 disable_irq(hwif->irq);
4320 /* local CPU only, as if we were handling an interrupt */
4321 - local_irq_disable();
4322 + local_irq_disable_nort();
4323 if (hwif->polling) {
4324 startstop = handler(drive);
4325 } else if (drive_is_ready(drive)) {
4326 diff --git a/drivers/ide/ide-iops.c b/drivers/ide/ide-iops.c
4327 index 376f2dc410c5..f014dd1b73dc 100644
4328 --- a/drivers/ide/ide-iops.c
4329 +++ b/drivers/ide/ide-iops.c
4330 @@ -129,12 +129,12 @@ int __ide_wait_stat(ide_drive_t *drive, u8 good, u8 bad,
4331 if ((stat & ATA_BUSY) == 0)
4334 - local_irq_restore(flags);
4335 + local_irq_restore_nort(flags);
4340 - local_irq_restore(flags);
4341 + local_irq_restore_nort(flags);
4344 * Allow status to settle, then read it again.
4345 diff --git a/drivers/ide/ide-probe.c b/drivers/ide/ide-probe.c
4346 index 0b63facd1d87..4ceba37afc0c 100644
4347 --- a/drivers/ide/ide-probe.c
4348 +++ b/drivers/ide/ide-probe.c
4349 @@ -196,10 +196,10 @@ static void do_identify(ide_drive_t *drive, u8 cmd, u16 *id)
4352 /* local CPU only; some systems need this */
4353 - local_irq_save(flags);
4354 + local_irq_save_nort(flags);
4355 /* read 512 bytes of id info */
4356 hwif->tp_ops->input_data(drive, NULL, id, SECTOR_SIZE);
4357 - local_irq_restore(flags);
4358 + local_irq_restore_nort(flags);
4360 drive->dev_flags |= IDE_DFLAG_ID_READ;
4362 diff --git a/drivers/ide/ide-taskfile.c b/drivers/ide/ide-taskfile.c
4363 index a716693417a3..be0568c722d6 100644
4364 --- a/drivers/ide/ide-taskfile.c
4365 +++ b/drivers/ide/ide-taskfile.c
4366 @@ -250,7 +250,7 @@ void ide_pio_bytes(ide_drive_t *drive, struct ide_cmd *cmd,
4368 page_is_high = PageHighMem(page);
4370 - local_irq_save(flags);
4371 + local_irq_save_nort(flags);
4373 buf = kmap_atomic(page) + offset;
4375 @@ -271,7 +271,7 @@ void ide_pio_bytes(ide_drive_t *drive, struct ide_cmd *cmd,
4379 - local_irq_restore(flags);
4380 + local_irq_restore_nort(flags);
4384 @@ -414,7 +414,7 @@ static ide_startstop_t pre_task_out_intr(ide_drive_t *drive,
4387 if ((drive->dev_flags & IDE_DFLAG_UNMASK) == 0)
4388 - local_irq_disable();
4389 + local_irq_disable_nort();
4391 ide_set_handler(drive, &task_pio_intr, WAIT_WORSTCASE);
4393 diff --git a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
4394 index fddff403d5d2..cca1bb4fbfe3 100644
4395 --- a/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
4396 +++ b/drivers/infiniband/ulp/ipoib/ipoib_multicast.c
4397 @@ -902,7 +902,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
4399 ipoib_dbg_mcast(priv, "restarting multicast task\n");
4401 - local_irq_save(flags);
4402 + local_irq_save_nort(flags);
4403 netif_addr_lock(dev);
4404 spin_lock(&priv->lock);
4406 @@ -984,7 +984,7 @@ void ipoib_mcast_restart_task(struct work_struct *work)
4408 spin_unlock(&priv->lock);
4409 netif_addr_unlock(dev);
4410 - local_irq_restore(flags);
4411 + local_irq_restore_nort(flags);
4414 * make sure the in-flight joins have finished before we attempt
4415 diff --git a/drivers/input/gameport/gameport.c b/drivers/input/gameport/gameport.c
4416 index 4a2a9e370be7..e970d9afd179 100644
4417 --- a/drivers/input/gameport/gameport.c
4418 +++ b/drivers/input/gameport/gameport.c
4419 @@ -91,13 +91,13 @@ static int gameport_measure_speed(struct gameport *gameport)
4422 for (i = 0; i < 50; i++) {
4423 - local_irq_save(flags);
4424 + local_irq_save_nort(flags);
4425 t1 = ktime_get_ns();
4426 for (t = 0; t < 50; t++)
4427 gameport_read(gameport);
4428 t2 = ktime_get_ns();
4429 t3 = ktime_get_ns();
4430 - local_irq_restore(flags);
4431 + local_irq_restore_nort(flags);
4433 t = (t2 - t1) - (t3 - t2);
4435 @@ -124,12 +124,12 @@ static int old_gameport_measure_speed(struct gameport *gameport)
4438 for(i = 0; i < 50; i++) {
4439 - local_irq_save(flags);
4440 + local_irq_save_nort(flags);
4442 for (t = 0; t < 50; t++) gameport_read(gameport);
4445 - local_irq_restore(flags);
4446 + local_irq_restore_nort(flags);
4448 if ((t = DELTA(t2,t1) - DELTA(t3,t2)) < tx) tx = t;
4450 @@ -148,11 +148,11 @@ static int old_gameport_measure_speed(struct gameport *gameport)
4453 for(i = 0; i < 50; i++) {
4454 - local_irq_save(flags);
4455 + local_irq_save_nort(flags);
4457 for (t = 0; t < 50; t++) gameport_read(gameport);
4459 - local_irq_restore(flags);
4460 + local_irq_restore_nort(flags);
4462 if (t2 - t1 < tx) tx = t2 - t1;
4464 diff --git a/drivers/iommu/amd_iommu.c b/drivers/iommu/amd_iommu.c
4465 index 11a13b5be73a..baaed0ac274b 100644
4466 --- a/drivers/iommu/amd_iommu.c
4467 +++ b/drivers/iommu/amd_iommu.c
4468 @@ -1923,10 +1923,10 @@ static int __attach_device(struct iommu_dev_data *dev_data,
4472 - * Must be called with IRQs disabled. Warn here to detect early
4474 + * Must be called with IRQs disabled on a non RT kernel. Warn here to
4475 + * detect early when its not.
4477 - WARN_ON(!irqs_disabled());
4478 + WARN_ON_NONRT(!irqs_disabled());
4481 spin_lock(&domain->lock);
4482 @@ -2094,10 +2094,10 @@ static void __detach_device(struct iommu_dev_data *dev_data)
4483 struct protection_domain *domain;
4486 - * Must be called with IRQs disabled. Warn here to detect early
4488 + * Must be called with IRQs disabled on a non RT kernel. Warn here to
4489 + * detect early when its not.
4491 - WARN_ON(!irqs_disabled());
4492 + WARN_ON_NONRT(!irqs_disabled());
4494 if (WARN_ON(!dev_data->domain))
4496 diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c
4497 index d82637ab09fd..ebe41d30c093 100644
4498 --- a/drivers/iommu/intel-iommu.c
4499 +++ b/drivers/iommu/intel-iommu.c
4500 @@ -479,7 +479,7 @@ struct deferred_flush_data {
4501 struct deferred_flush_table *tables;
4504 -DEFINE_PER_CPU(struct deferred_flush_data, deferred_flush);
4505 +static DEFINE_PER_CPU(struct deferred_flush_data, deferred_flush);
4507 /* bitmap for indexing intel_iommus */
4508 static int g_num_of_iommus;
4509 @@ -3715,10 +3715,8 @@ static void add_unmap(struct dmar_domain *dom, unsigned long iova_pfn,
4510 struct intel_iommu *iommu;
4511 struct deferred_flush_entry *entry;
4512 struct deferred_flush_data *flush_data;
4513 - unsigned int cpuid;
4515 - cpuid = get_cpu();
4516 - flush_data = per_cpu_ptr(&deferred_flush, cpuid);
4517 + flush_data = raw_cpu_ptr(&deferred_flush);
4519 /* Flush all CPUs' entries to avoid deferring too much. If
4520 * this becomes a bottleneck, can just flush us, and rely on
4521 @@ -3751,8 +3749,6 @@ static void add_unmap(struct dmar_domain *dom, unsigned long iova_pfn,
4524 spin_unlock_irqrestore(&flush_data->lock, flags);
4529 static void intel_unmap(struct device *dev, dma_addr_t dev_addr, size_t size)
4530 diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
4531 index e23001bfcfee..359d5d169ec0 100644
4532 --- a/drivers/iommu/iova.c
4533 +++ b/drivers/iommu/iova.c
4535 #include <linux/slab.h>
4536 #include <linux/smp.h>
4537 #include <linux/bitops.h>
4538 +#include <linux/cpu.h>
4540 static bool iova_rcache_insert(struct iova_domain *iovad,
4542 @@ -420,10 +421,8 @@ alloc_iova_fast(struct iova_domain *iovad, unsigned long size,
4544 /* Try replenishing IOVAs by flushing rcache. */
4545 flushed_rcache = true;
4546 - preempt_disable();
4547 for_each_online_cpu(cpu)
4548 free_cpu_cached_iovas(cpu, iovad);
4553 @@ -751,7 +750,7 @@ static bool __iova_rcache_insert(struct iova_domain *iovad,
4554 bool can_insert = false;
4555 unsigned long flags;
4557 - cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches);
4558 + cpu_rcache = raw_cpu_ptr(rcache->cpu_rcaches);
4559 spin_lock_irqsave(&cpu_rcache->lock, flags);
4561 if (!iova_magazine_full(cpu_rcache->loaded)) {
4562 @@ -781,7 +780,6 @@ static bool __iova_rcache_insert(struct iova_domain *iovad,
4563 iova_magazine_push(cpu_rcache->loaded, iova_pfn);
4565 spin_unlock_irqrestore(&cpu_rcache->lock, flags);
4566 - put_cpu_ptr(rcache->cpu_rcaches);
4569 iova_magazine_free_pfns(mag_to_free, iovad);
4570 @@ -815,7 +813,7 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache,
4571 bool has_pfn = false;
4572 unsigned long flags;
4574 - cpu_rcache = get_cpu_ptr(rcache->cpu_rcaches);
4575 + cpu_rcache = raw_cpu_ptr(rcache->cpu_rcaches);
4576 spin_lock_irqsave(&cpu_rcache->lock, flags);
4578 if (!iova_magazine_empty(cpu_rcache->loaded)) {
4579 @@ -837,7 +835,6 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache,
4580 iova_pfn = iova_magazine_pop(cpu_rcache->loaded, limit_pfn);
4582 spin_unlock_irqrestore(&cpu_rcache->lock, flags);
4583 - put_cpu_ptr(rcache->cpu_rcaches);
4587 diff --git a/drivers/leds/trigger/Kconfig b/drivers/leds/trigger/Kconfig
4588 index 3f9ddb9fafa7..09da5b6b44a1 100644
4589 --- a/drivers/leds/trigger/Kconfig
4590 +++ b/drivers/leds/trigger/Kconfig
4591 @@ -69,7 +69,7 @@ config LEDS_TRIGGER_BACKLIGHT
4593 config LEDS_TRIGGER_CPU
4594 bool "LED CPU Trigger"
4595 - depends on LEDS_TRIGGERS
4596 + depends on LEDS_TRIGGERS && !PREEMPT_RT_BASE
4598 This allows LEDs to be controlled by active CPUs. This shows
4599 the active CPUs across an array of LEDs so you can see which
4600 diff --git a/drivers/md/bcache/Kconfig b/drivers/md/bcache/Kconfig
4601 index 4d200883c505..98b64ed5cb81 100644
4602 --- a/drivers/md/bcache/Kconfig
4603 +++ b/drivers/md/bcache/Kconfig
4607 tristate "Block device as cache"
4608 + depends on !PREEMPT_RT_FULL
4610 Allows a block device to be used as cache for other devices; uses
4611 a btree for indexing and the layout is optimized for SSDs.
4612 diff --git a/drivers/md/dm-rq.c b/drivers/md/dm-rq.c
4613 index 2c965424d383..2c8877f50626 100644
4614 --- a/drivers/md/dm-rq.c
4615 +++ b/drivers/md/dm-rq.c
4616 @@ -842,7 +842,7 @@ static void dm_old_request_fn(struct request_queue *q)
4617 /* Establish tio->ti before queuing work (map_tio_request) */
4619 kthread_queue_work(&md->kworker, &tio->work);
4620 - BUG_ON(!irqs_disabled());
4621 + BUG_ON_NONRT(!irqs_disabled());
4625 diff --git a/drivers/md/raid5.c b/drivers/md/raid5.c
4626 index cce6057b9aca..fa2c4de32a64 100644
4627 --- a/drivers/md/raid5.c
4628 +++ b/drivers/md/raid5.c
4629 @@ -1928,8 +1928,9 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
4630 struct raid5_percpu *percpu;
4634 + cpu = get_cpu_light();
4635 percpu = per_cpu_ptr(conf->percpu, cpu);
4636 + spin_lock(&percpu->lock);
4637 if (test_bit(STRIPE_OP_BIOFILL, &ops_request)) {
4638 ops_run_biofill(sh);
4640 @@ -1985,7 +1986,8 @@ static void raid_run_ops(struct stripe_head *sh, unsigned long ops_request)
4641 if (test_and_clear_bit(R5_Overlap, &dev->flags))
4642 wake_up(&sh->raid_conf->wait_for_overlap);
4645 + spin_unlock(&percpu->lock);
4649 static struct stripe_head *alloc_stripe(struct kmem_cache *sc, gfp_t gfp,
4650 @@ -6391,6 +6393,7 @@ static int raid456_cpu_up_prepare(unsigned int cpu, struct hlist_node *node)
4654 + spin_lock_init(&per_cpu_ptr(conf->percpu, cpu)->lock);
4658 @@ -6401,7 +6404,6 @@ static int raid5_alloc_percpu(struct r5conf *conf)
4659 conf->percpu = alloc_percpu(struct raid5_percpu);
4663 err = cpuhp_state_add_instance(CPUHP_MD_RAID5_PREPARE, &conf->node);
4665 conf->scribble_disks = max(conf->raid_disks,
4666 diff --git a/drivers/md/raid5.h b/drivers/md/raid5.h
4667 index 57ec49f0839e..0739604990b7 100644
4668 --- a/drivers/md/raid5.h
4669 +++ b/drivers/md/raid5.h
4670 @@ -504,6 +504,7 @@ struct r5conf {
4671 int recovery_disabled;
4672 /* per cpu variables */
4673 struct raid5_percpu {
4674 + spinlock_t lock; /* Protection for -RT */
4675 struct page *spare_page; /* Used when checking P/Q in raid6 */
4676 struct flex_array *scribble; /* space for constructing buffer
4677 * lists and performing address
4678 diff --git a/drivers/misc/Kconfig b/drivers/misc/Kconfig
4679 index 64971baf11fa..215e91e36198 100644
4680 --- a/drivers/misc/Kconfig
4681 +++ b/drivers/misc/Kconfig
4682 @@ -54,6 +54,7 @@ config AD525X_DPOT_SPI
4684 bool "Atmel AT32/AT91 Timer/Counter Library"
4685 depends on (AVR32 || ARCH_AT91)
4686 + default y if PREEMPT_RT_FULL
4688 Select this if you want a library to allocate the Timer/Counter
4689 blocks found on many Atmel processors. This facilitates using
4690 @@ -69,8 +70,7 @@ config ATMEL_TCB_CLKSRC
4691 are combined to make a single 32-bit timer.
4693 When GENERIC_CLOCKEVENTS is defined, the third timer channel
4694 - may be used as a clock event device supporting oneshot mode
4695 - (delays of up to two seconds) based on the 32 KiHz clock.
4696 + may be used as a clock event device supporting oneshot mode.
4698 config ATMEL_TCB_CLKSRC_BLOCK
4700 @@ -84,6 +84,15 @@ config ATMEL_TCB_CLKSRC_BLOCK
4701 TC can be used for other purposes, such as PWM generation and
4704 +config ATMEL_TCB_CLKSRC_USE_SLOW_CLOCK
4705 + bool "TC Block use 32 KiHz clock"
4706 + depends on ATMEL_TCB_CLKSRC
4707 + default y if !PREEMPT_RT_FULL
4709 + Select this to use 32 KiHz base clock rate as TC block clock
4710 + source for clock events.
4714 tristate "Dummy IRQ handler"
4716 diff --git a/drivers/mmc/host/mmci.c b/drivers/mmc/host/mmci.c
4717 index df990bb8c873..1a162709a85e 100644
4718 --- a/drivers/mmc/host/mmci.c
4719 +++ b/drivers/mmc/host/mmci.c
4720 @@ -1147,15 +1147,12 @@ static irqreturn_t mmci_pio_irq(int irq, void *dev_id)
4721 struct sg_mapping_iter *sg_miter = &host->sg_miter;
4722 struct variant_data *variant = host->variant;
4723 void __iomem *base = host->base;
4724 - unsigned long flags;
4727 status = readl(base + MMCISTATUS);
4729 dev_dbg(mmc_dev(host->mmc), "irq1 (pio) %08x\n", status);
4731 - local_irq_save(flags);
4734 unsigned int remain, len;
4736 @@ -1195,8 +1192,6 @@ static irqreturn_t mmci_pio_irq(int irq, void *dev_id)
4738 sg_miter_stop(sg_miter);
4740 - local_irq_restore(flags);
4743 * If we have less than the fifo 'half-full' threshold to transfer,
4744 * trigger a PIO interrupt as soon as any data is available.
4745 diff --git a/drivers/net/ethernet/3com/3c59x.c b/drivers/net/ethernet/3com/3c59x.c
4746 index 9133e7926da5..63afb921ed40 100644
4747 --- a/drivers/net/ethernet/3com/3c59x.c
4748 +++ b/drivers/net/ethernet/3com/3c59x.c
4749 @@ -842,9 +842,9 @@ static void poll_vortex(struct net_device *dev)
4751 struct vortex_private *vp = netdev_priv(dev);
4752 unsigned long flags;
4753 - local_irq_save(flags);
4754 + local_irq_save_nort(flags);
4755 (vp->full_bus_master_rx ? boomerang_interrupt:vortex_interrupt)(dev->irq,dev);
4756 - local_irq_restore(flags);
4757 + local_irq_restore_nort(flags);
4761 @@ -1910,12 +1910,12 @@ static void vortex_tx_timeout(struct net_device *dev)
4762 * Block interrupts because vortex_interrupt does a bare spin_lock()
4764 unsigned long flags;
4765 - local_irq_save(flags);
4766 + local_irq_save_nort(flags);
4767 if (vp->full_bus_master_tx)
4768 boomerang_interrupt(dev->irq, dev);
4770 vortex_interrupt(dev->irq, dev);
4771 - local_irq_restore(flags);
4772 + local_irq_restore_nort(flags);
4776 diff --git a/drivers/net/ethernet/realtek/8139too.c b/drivers/net/ethernet/realtek/8139too.c
4777 index da4c2d8a4173..1420dfb56bac 100644
4778 --- a/drivers/net/ethernet/realtek/8139too.c
4779 +++ b/drivers/net/ethernet/realtek/8139too.c
4780 @@ -2233,7 +2233,7 @@ static void rtl8139_poll_controller(struct net_device *dev)
4781 struct rtl8139_private *tp = netdev_priv(dev);
4782 const int irq = tp->pci_dev->irq;
4785 + disable_irq_nosync(irq);
4786 rtl8139_interrupt(irq, dev);
4789 diff --git a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
4790 index bca6935a94db..d7a35ee34d03 100644
4791 --- a/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
4792 +++ b/drivers/net/wireless/intersil/orinoco/orinoco_usb.c
4793 @@ -697,7 +697,7 @@ static void ezusb_req_ctx_wait(struct ezusb_priv *upriv,
4794 while (!ctx->done.done && msecs--)
4797 - wait_event_interruptible(ctx->done.wait,
4798 + swait_event_interruptible(ctx->done.wait,
4802 diff --git a/drivers/pci/access.c b/drivers/pci/access.c
4803 index d11cdbb8fba3..223bbb9acb03 100644
4804 --- a/drivers/pci/access.c
4805 +++ b/drivers/pci/access.c
4806 @@ -672,7 +672,7 @@ void pci_cfg_access_unlock(struct pci_dev *dev)
4807 WARN_ON(!dev->block_cfg_access);
4809 dev->block_cfg_access = 0;
4810 - wake_up_all(&pci_cfg_wait);
4811 + wake_up_all_locked(&pci_cfg_wait);
4812 raw_spin_unlock_irqrestore(&pci_lock, flags);
4814 EXPORT_SYMBOL_GPL(pci_cfg_access_unlock);
4815 diff --git a/drivers/pinctrl/qcom/pinctrl-msm.c b/drivers/pinctrl/qcom/pinctrl-msm.c
4816 index 775c88303017..f8e9e1c2b2f6 100644
4817 --- a/drivers/pinctrl/qcom/pinctrl-msm.c
4818 +++ b/drivers/pinctrl/qcom/pinctrl-msm.c
4819 @@ -61,7 +61,7 @@ struct msm_pinctrl {
4820 struct notifier_block restart_nb;
4824 + raw_spinlock_t lock;
4826 DECLARE_BITMAP(dual_edge_irqs, MAX_NR_GPIO);
4827 DECLARE_BITMAP(enabled_irqs, MAX_NR_GPIO);
4828 @@ -153,14 +153,14 @@ static int msm_pinmux_set_mux(struct pinctrl_dev *pctldev,
4829 if (WARN_ON(i == g->nfuncs))
4832 - spin_lock_irqsave(&pctrl->lock, flags);
4833 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4835 val = readl(pctrl->regs + g->ctl_reg);
4837 val |= i << g->mux_bit;
4838 writel(val, pctrl->regs + g->ctl_reg);
4840 - spin_unlock_irqrestore(&pctrl->lock, flags);
4841 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4845 @@ -323,14 +323,14 @@ static int msm_config_group_set(struct pinctrl_dev *pctldev,
4847 case PIN_CONFIG_OUTPUT:
4848 /* set output value */
4849 - spin_lock_irqsave(&pctrl->lock, flags);
4850 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4851 val = readl(pctrl->regs + g->io_reg);
4853 val |= BIT(g->out_bit);
4855 val &= ~BIT(g->out_bit);
4856 writel(val, pctrl->regs + g->io_reg);
4857 - spin_unlock_irqrestore(&pctrl->lock, flags);
4858 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4862 @@ -351,12 +351,12 @@ static int msm_config_group_set(struct pinctrl_dev *pctldev,
4866 - spin_lock_irqsave(&pctrl->lock, flags);
4867 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4868 val = readl(pctrl->regs + g->ctl_reg);
4869 val &= ~(mask << bit);
4871 writel(val, pctrl->regs + g->ctl_reg);
4872 - spin_unlock_irqrestore(&pctrl->lock, flags);
4873 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4877 @@ -384,13 +384,13 @@ static int msm_gpio_direction_input(struct gpio_chip *chip, unsigned offset)
4879 g = &pctrl->soc->groups[offset];
4881 - spin_lock_irqsave(&pctrl->lock, flags);
4882 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4884 val = readl(pctrl->regs + g->ctl_reg);
4885 val &= ~BIT(g->oe_bit);
4886 writel(val, pctrl->regs + g->ctl_reg);
4888 - spin_unlock_irqrestore(&pctrl->lock, flags);
4889 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4893 @@ -404,7 +404,7 @@ static int msm_gpio_direction_output(struct gpio_chip *chip, unsigned offset, in
4895 g = &pctrl->soc->groups[offset];
4897 - spin_lock_irqsave(&pctrl->lock, flags);
4898 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4900 val = readl(pctrl->regs + g->io_reg);
4902 @@ -417,7 +417,7 @@ static int msm_gpio_direction_output(struct gpio_chip *chip, unsigned offset, in
4903 val |= BIT(g->oe_bit);
4904 writel(val, pctrl->regs + g->ctl_reg);
4906 - spin_unlock_irqrestore(&pctrl->lock, flags);
4907 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4911 @@ -443,7 +443,7 @@ static void msm_gpio_set(struct gpio_chip *chip, unsigned offset, int value)
4913 g = &pctrl->soc->groups[offset];
4915 - spin_lock_irqsave(&pctrl->lock, flags);
4916 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4918 val = readl(pctrl->regs + g->io_reg);
4920 @@ -452,7 +452,7 @@ static void msm_gpio_set(struct gpio_chip *chip, unsigned offset, int value)
4921 val &= ~BIT(g->out_bit);
4922 writel(val, pctrl->regs + g->io_reg);
4924 - spin_unlock_irqrestore(&pctrl->lock, flags);
4925 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4928 #ifdef CONFIG_DEBUG_FS
4929 @@ -571,7 +571,7 @@ static void msm_gpio_irq_mask(struct irq_data *d)
4931 g = &pctrl->soc->groups[d->hwirq];
4933 - spin_lock_irqsave(&pctrl->lock, flags);
4934 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4936 val = readl(pctrl->regs + g->intr_cfg_reg);
4937 val &= ~BIT(g->intr_enable_bit);
4938 @@ -579,7 +579,7 @@ static void msm_gpio_irq_mask(struct irq_data *d)
4940 clear_bit(d->hwirq, pctrl->enabled_irqs);
4942 - spin_unlock_irqrestore(&pctrl->lock, flags);
4943 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4946 static void msm_gpio_irq_unmask(struct irq_data *d)
4947 @@ -592,7 +592,7 @@ static void msm_gpio_irq_unmask(struct irq_data *d)
4949 g = &pctrl->soc->groups[d->hwirq];
4951 - spin_lock_irqsave(&pctrl->lock, flags);
4952 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4954 val = readl(pctrl->regs + g->intr_status_reg);
4955 val &= ~BIT(g->intr_status_bit);
4956 @@ -604,7 +604,7 @@ static void msm_gpio_irq_unmask(struct irq_data *d)
4958 set_bit(d->hwirq, pctrl->enabled_irqs);
4960 - spin_unlock_irqrestore(&pctrl->lock, flags);
4961 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4964 static void msm_gpio_irq_ack(struct irq_data *d)
4965 @@ -617,7 +617,7 @@ static void msm_gpio_irq_ack(struct irq_data *d)
4967 g = &pctrl->soc->groups[d->hwirq];
4969 - spin_lock_irqsave(&pctrl->lock, flags);
4970 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4972 val = readl(pctrl->regs + g->intr_status_reg);
4973 if (g->intr_ack_high)
4974 @@ -629,7 +629,7 @@ static void msm_gpio_irq_ack(struct irq_data *d)
4975 if (test_bit(d->hwirq, pctrl->dual_edge_irqs))
4976 msm_gpio_update_dual_edge_pos(pctrl, g, d);
4978 - spin_unlock_irqrestore(&pctrl->lock, flags);
4979 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4982 static int msm_gpio_irq_set_type(struct irq_data *d, unsigned int type)
4983 @@ -642,7 +642,7 @@ static int msm_gpio_irq_set_type(struct irq_data *d, unsigned int type)
4985 g = &pctrl->soc->groups[d->hwirq];
4987 - spin_lock_irqsave(&pctrl->lock, flags);
4988 + raw_spin_lock_irqsave(&pctrl->lock, flags);
4991 * For hw without possibility of detecting both edges
4992 @@ -716,7 +716,7 @@ static int msm_gpio_irq_set_type(struct irq_data *d, unsigned int type)
4993 if (test_bit(d->hwirq, pctrl->dual_edge_irqs))
4994 msm_gpio_update_dual_edge_pos(pctrl, g, d);
4996 - spin_unlock_irqrestore(&pctrl->lock, flags);
4997 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
4999 if (type & (IRQ_TYPE_LEVEL_LOW | IRQ_TYPE_LEVEL_HIGH))
5000 irq_set_handler_locked(d, handle_level_irq);
5001 @@ -732,11 +732,11 @@ static int msm_gpio_irq_set_wake(struct irq_data *d, unsigned int on)
5002 struct msm_pinctrl *pctrl = gpiochip_get_data(gc);
5003 unsigned long flags;
5005 - spin_lock_irqsave(&pctrl->lock, flags);
5006 + raw_spin_lock_irqsave(&pctrl->lock, flags);
5008 irq_set_irq_wake(pctrl->irq, on);
5010 - spin_unlock_irqrestore(&pctrl->lock, flags);
5011 + raw_spin_unlock_irqrestore(&pctrl->lock, flags);
5015 @@ -882,7 +882,7 @@ int msm_pinctrl_probe(struct platform_device *pdev,
5016 pctrl->soc = soc_data;
5017 pctrl->chip = msm_gpio_template;
5019 - spin_lock_init(&pctrl->lock);
5020 + raw_spin_lock_init(&pctrl->lock);
5022 res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
5023 pctrl->regs = devm_ioremap_resource(&pdev->dev, res);
5024 diff --git a/drivers/scsi/fcoe/fcoe.c b/drivers/scsi/fcoe/fcoe.c
5025 index 9bd41a35a78a..8e2d436c2e3f 100644
5026 --- a/drivers/scsi/fcoe/fcoe.c
5027 +++ b/drivers/scsi/fcoe/fcoe.c
5028 @@ -1455,11 +1455,11 @@ static int fcoe_rcv(struct sk_buff *skb, struct net_device *netdev,
5029 static int fcoe_alloc_paged_crc_eof(struct sk_buff *skb, int tlen)
5031 struct fcoe_percpu_s *fps;
5033 + int rc, cpu = get_cpu_light();
5035 - fps = &get_cpu_var(fcoe_percpu);
5036 + fps = &per_cpu(fcoe_percpu, cpu);
5037 rc = fcoe_get_paged_crc_eof(skb, tlen, fps);
5038 - put_cpu_var(fcoe_percpu);
5043 @@ -1646,11 +1646,11 @@ static inline int fcoe_filter_frames(struct fc_lport *lport,
5047 - stats = per_cpu_ptr(lport->stats, get_cpu());
5048 + stats = per_cpu_ptr(lport->stats, get_cpu_light());
5049 stats->InvalidCRCCount++;
5050 if (stats->InvalidCRCCount < 5)
5051 printk(KERN_WARNING "fcoe: dropping frame with CRC error\n");
5057 @@ -1693,7 +1693,7 @@ static void fcoe_recv_frame(struct sk_buff *skb)
5059 hp = (struct fcoe_hdr *) skb_network_header(skb);
5061 - stats = per_cpu_ptr(lport->stats, get_cpu());
5062 + stats = per_cpu_ptr(lport->stats, get_cpu_light());
5063 if (unlikely(FC_FCOE_DECAPS_VER(hp) != FC_FCOE_VER)) {
5064 if (stats->ErrorFrames < 5)
5065 printk(KERN_WARNING "fcoe: FCoE version "
5066 @@ -1725,13 +1725,13 @@ static void fcoe_recv_frame(struct sk_buff *skb)
5069 if (!fcoe_filter_frames(lport, fp)) {
5072 fc_exch_recv(lport, fp);
5076 stats->ErrorFrames++;
5082 diff --git a/drivers/scsi/fcoe/fcoe_ctlr.c b/drivers/scsi/fcoe/fcoe_ctlr.c
5083 index dcf36537a767..1a1f2e46452c 100644
5084 --- a/drivers/scsi/fcoe/fcoe_ctlr.c
5085 +++ b/drivers/scsi/fcoe/fcoe_ctlr.c
5086 @@ -834,7 +834,7 @@ static unsigned long fcoe_ctlr_age_fcfs(struct fcoe_ctlr *fip)
5088 INIT_LIST_HEAD(&del_list);
5090 - stats = per_cpu_ptr(fip->lp->stats, get_cpu());
5091 + stats = per_cpu_ptr(fip->lp->stats, get_cpu_light());
5093 list_for_each_entry_safe(fcf, next, &fip->fcfs, list) {
5094 deadline = fcf->time + fcf->fka_period + fcf->fka_period / 2;
5095 @@ -870,7 +870,7 @@ static unsigned long fcoe_ctlr_age_fcfs(struct fcoe_ctlr *fip)
5096 sel_time = fcf->time;
5102 list_for_each_entry_safe(fcf, next, &del_list, list) {
5103 /* Removes fcf from current list */
5104 diff --git a/drivers/scsi/libfc/fc_exch.c b/drivers/scsi/libfc/fc_exch.c
5105 index 16ca31ad5ec0..c3987347e762 100644
5106 --- a/drivers/scsi/libfc/fc_exch.c
5107 +++ b/drivers/scsi/libfc/fc_exch.c
5108 @@ -814,10 +814,10 @@ static struct fc_exch *fc_exch_em_alloc(struct fc_lport *lport,
5110 memset(ep, 0, sizeof(*ep));
5113 + cpu = get_cpu_light();
5114 pool = per_cpu_ptr(mp->pool, cpu);
5115 spin_lock_bh(&pool->lock);
5119 /* peek cache of free slot */
5120 if (pool->left != FC_XID_UNKNOWN) {
5121 diff --git a/drivers/scsi/libsas/sas_ata.c b/drivers/scsi/libsas/sas_ata.c
5122 index 763f012fdeca..d0f61b595470 100644
5123 --- a/drivers/scsi/libsas/sas_ata.c
5124 +++ b/drivers/scsi/libsas/sas_ata.c
5125 @@ -190,7 +190,7 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
5126 /* TODO: audit callers to ensure they are ready for qc_issue to
5127 * unconditionally re-enable interrupts
5129 - local_irq_save(flags);
5130 + local_irq_save_nort(flags);
5131 spin_unlock(ap->lock);
5133 /* If the device fell off, no sense in issuing commands */
5134 @@ -252,7 +252,7 @@ static unsigned int sas_ata_qc_issue(struct ata_queued_cmd *qc)
5137 spin_lock(ap->lock);
5138 - local_irq_restore(flags);
5139 + local_irq_restore_nort(flags);
5143 diff --git a/drivers/scsi/qla2xxx/qla_inline.h b/drivers/scsi/qla2xxx/qla_inline.h
5144 index edc48f3b8230..ee5c6f9dfb6f 100644
5145 --- a/drivers/scsi/qla2xxx/qla_inline.h
5146 +++ b/drivers/scsi/qla2xxx/qla_inline.h
5147 @@ -59,12 +59,12 @@ qla2x00_poll(struct rsp_que *rsp)
5149 unsigned long flags;
5150 struct qla_hw_data *ha = rsp->hw;
5151 - local_irq_save(flags);
5152 + local_irq_save_nort(flags);
5153 if (IS_P3P_TYPE(ha))
5154 qla82xx_poll(0, rsp);
5156 ha->isp_ops->intr_handler(0, rsp);
5157 - local_irq_restore(flags);
5158 + local_irq_restore_nort(flags);
5161 static inline uint8_t *
5162 diff --git a/drivers/scsi/qla2xxx/qla_isr.c b/drivers/scsi/qla2xxx/qla_isr.c
5163 index 068c4e47fac9..a2090f640397 100644
5164 --- a/drivers/scsi/qla2xxx/qla_isr.c
5165 +++ b/drivers/scsi/qla2xxx/qla_isr.c
5166 @@ -3125,7 +3125,11 @@ qla24xx_enable_msix(struct qla_hw_data *ha, struct rsp_que *rsp)
5169 kref_get(&qentry->irq_notify.kref);
5170 +#ifdef CONFIG_PREEMPT_RT_BASE
5171 + swork_queue(&qentry->irq_notify.swork);
5173 schedule_work(&qentry->irq_notify.work);
5178 diff --git a/drivers/thermal/x86_pkg_temp_thermal.c b/drivers/thermal/x86_pkg_temp_thermal.c
5179 index 95f4c1bcdb4c..0be934799bff 100644
5180 --- a/drivers/thermal/x86_pkg_temp_thermal.c
5181 +++ b/drivers/thermal/x86_pkg_temp_thermal.c
5183 #include <linux/pm.h>
5184 #include <linux/thermal.h>
5185 #include <linux/debugfs.h>
5186 +#include <linux/swork.h>
5187 #include <asm/cpu_device_id.h>
5188 #include <asm/mce.h>
5190 @@ -353,7 +354,7 @@ static void pkg_temp_thermal_threshold_work_fn(struct work_struct *work)
5194 -static int pkg_temp_thermal_platform_thermal_notify(__u64 msr_val)
5195 +static void platform_thermal_notify_work(struct swork_event *event)
5197 unsigned long flags;
5198 int cpu = smp_processor_id();
5199 @@ -370,7 +371,7 @@ static int pkg_temp_thermal_platform_thermal_notify(__u64 msr_val)
5200 pkg_work_scheduled[phy_id]) {
5201 disable_pkg_thres_interrupt();
5202 spin_unlock_irqrestore(&pkg_work_lock, flags);
5206 pkg_work_scheduled[phy_id] = 1;
5207 spin_unlock_irqrestore(&pkg_work_lock, flags);
5208 @@ -379,9 +380,48 @@ static int pkg_temp_thermal_platform_thermal_notify(__u64 msr_val)
5209 schedule_delayed_work_on(cpu,
5210 &per_cpu(pkg_temp_thermal_threshold_work, cpu),
5211 msecs_to_jiffies(notify_delay_ms));
5214 +#ifdef CONFIG_PREEMPT_RT_FULL
5215 +static struct swork_event notify_work;
5217 +static int thermal_notify_work_init(void)
5221 + err = swork_get();
5225 + INIT_SWORK(¬ify_work, platform_thermal_notify_work);
5229 +static void thermal_notify_work_cleanup(void)
5234 +static int pkg_temp_thermal_platform_thermal_notify(__u64 msr_val)
5236 + swork_queue(¬ify_work);
5240 +#else /* !CONFIG_PREEMPT_RT_FULL */
5242 +static int thermal_notify_work_init(void) { return 0; }
5244 +static void thermal_notify_work_cleanup(void) { }
5246 +static int pkg_temp_thermal_platform_thermal_notify(__u64 msr_val)
5248 + platform_thermal_notify_work(NULL);
5252 +#endif /* CONFIG_PREEMPT_RT_FULL */
5254 static int find_siblings_cpu(int cpu)
5257 @@ -585,6 +625,9 @@ static int __init pkg_temp_thermal_init(void)
5258 if (!x86_match_cpu(pkg_temp_thermal_ids))
5261 + if (!thermal_notify_work_init())
5264 spin_lock_init(&pkg_work_lock);
5265 platform_thermal_package_notify =
5266 pkg_temp_thermal_platform_thermal_notify;
5267 @@ -609,7 +652,7 @@ static int __init pkg_temp_thermal_init(void)
5268 kfree(pkg_work_scheduled);
5269 platform_thermal_package_notify = NULL;
5270 platform_thermal_package_rate_control = NULL;
5272 + thermal_notify_work_cleanup();
5276 @@ -634,6 +677,7 @@ static void __exit pkg_temp_thermal_exit(void)
5277 mutex_unlock(&phy_dev_list_mutex);
5278 platform_thermal_package_notify = NULL;
5279 platform_thermal_package_rate_control = NULL;
5280 + thermal_notify_work_cleanup();
5281 for_each_online_cpu(i)
5282 cancel_delayed_work_sync(
5283 &per_cpu(pkg_temp_thermal_threshold_work, i));
5284 diff --git a/drivers/tty/serial/8250/8250_core.c b/drivers/tty/serial/8250/8250_core.c
5285 index e8819aa20415..dd7f9bf45d6c 100644
5286 --- a/drivers/tty/serial/8250/8250_core.c
5287 +++ b/drivers/tty/serial/8250/8250_core.c
5288 @@ -58,7 +58,16 @@ static struct uart_driver serial8250_reg;
5290 static unsigned int skip_txen_test; /* force skip of txen test at init time */
5292 -#define PASS_LIMIT 512
5294 + * On -rt we can have a more delays, and legitimately
5295 + * so - so don't drop work spuriously and spam the
5298 +#ifdef CONFIG_PREEMPT_RT_FULL
5299 +# define PASS_LIMIT 1000000
5301 +# define PASS_LIMIT 512
5304 #include <asm/serial.h>
5306 diff --git a/drivers/tty/serial/8250/8250_port.c b/drivers/tty/serial/8250/8250_port.c
5307 index 080d5a59d0a7..eecc4f111473 100644
5308 --- a/drivers/tty/serial/8250/8250_port.c
5309 +++ b/drivers/tty/serial/8250/8250_port.c
5311 #include <linux/nmi.h>
5312 #include <linux/mutex.h>
5313 #include <linux/slab.h>
5314 +#include <linux/kdb.h>
5315 #include <linux/uaccess.h>
5316 #include <linux/pm_runtime.h>
5317 #include <linux/timer.h>
5318 @@ -3144,9 +3145,9 @@ void serial8250_console_write(struct uart_8250_port *up, const char *s,
5320 serial8250_rpm_get(up);
5323 + if (port->sysrq || oops_in_progress)
5325 - else if (oops_in_progress)
5326 + else if (in_kdb_printk())
5327 locked = spin_trylock_irqsave(&port->lock, flags);
5329 spin_lock_irqsave(&port->lock, flags);
5330 diff --git a/drivers/tty/serial/amba-pl011.c b/drivers/tty/serial/amba-pl011.c
5331 index e2c33b9528d8..53af53c43e8c 100644
5332 --- a/drivers/tty/serial/amba-pl011.c
5333 +++ b/drivers/tty/serial/amba-pl011.c
5334 @@ -2194,13 +2194,19 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
5336 clk_enable(uap->clk);
5338 - local_irq_save(flags);
5340 + * local_irq_save(flags);
5342 + * This local_irq_save() is nonsense. If we come in via sysrq
5343 + * handling then interrupts are already disabled. Aside of
5344 + * that the port.sysrq check is racy on SMP regardless.
5346 if (uap->port.sysrq)
5348 else if (oops_in_progress)
5349 - locked = spin_trylock(&uap->port.lock);
5350 + locked = spin_trylock_irqsave(&uap->port.lock, flags);
5352 - spin_lock(&uap->port.lock);
5353 + spin_lock_irqsave(&uap->port.lock, flags);
5356 * First save the CR then disable the interrupts
5357 @@ -2224,8 +2230,7 @@ pl011_console_write(struct console *co, const char *s, unsigned int count)
5358 pl011_write(old_cr, uap, REG_CR);
5361 - spin_unlock(&uap->port.lock);
5362 - local_irq_restore(flags);
5363 + spin_unlock_irqrestore(&uap->port.lock, flags);
5365 clk_disable(uap->clk);
5367 diff --git a/drivers/tty/serial/omap-serial.c b/drivers/tty/serial/omap-serial.c
5368 index a2a529994ba5..0ee7c4c518df 100644
5369 --- a/drivers/tty/serial/omap-serial.c
5370 +++ b/drivers/tty/serial/omap-serial.c
5371 @@ -1257,13 +1257,10 @@ serial_omap_console_write(struct console *co, const char *s,
5373 pm_runtime_get_sync(up->dev);
5375 - local_irq_save(flags);
5376 - if (up->port.sysrq)
5378 - else if (oops_in_progress)
5379 - locked = spin_trylock(&up->port.lock);
5380 + if (up->port.sysrq || oops_in_progress)
5381 + locked = spin_trylock_irqsave(&up->port.lock, flags);
5383 - spin_lock(&up->port.lock);
5384 + spin_lock_irqsave(&up->port.lock, flags);
5387 * First save the IER then disable the interrupts
5388 @@ -1292,8 +1289,7 @@ serial_omap_console_write(struct console *co, const char *s,
5389 pm_runtime_mark_last_busy(up->dev);
5390 pm_runtime_put_autosuspend(up->dev);
5392 - spin_unlock(&up->port.lock);
5393 - local_irq_restore(flags);
5394 + spin_unlock_irqrestore(&up->port.lock, flags);
5398 diff --git a/drivers/usb/core/hcd.c b/drivers/usb/core/hcd.c
5399 index 479e223f9cff..3418a54b4131 100644
5400 --- a/drivers/usb/core/hcd.c
5401 +++ b/drivers/usb/core/hcd.c
5402 @@ -1761,9 +1761,9 @@ static void __usb_hcd_giveback_urb(struct urb *urb)
5403 * and no one may trigger the above deadlock situation when
5404 * running complete() in tasklet.
5406 - local_irq_save(flags);
5407 + local_irq_save_nort(flags);
5409 - local_irq_restore(flags);
5410 + local_irq_restore_nort(flags);
5412 usb_anchor_resume_wakeups(anchor);
5413 atomic_dec(&urb->use_count);
5414 diff --git a/drivers/usb/gadget/function/f_fs.c b/drivers/usb/gadget/function/f_fs.c
5415 index 8d412d8b1f29..176491dd739e 100644
5416 --- a/drivers/usb/gadget/function/f_fs.c
5417 +++ b/drivers/usb/gadget/function/f_fs.c
5418 @@ -1593,7 +1593,7 @@ static void ffs_data_put(struct ffs_data *ffs)
5419 pr_info("%s(): freeing\n", __func__);
5420 ffs_data_clear(ffs);
5421 BUG_ON(waitqueue_active(&ffs->ev.waitq) ||
5422 - waitqueue_active(&ffs->ep0req_completion.wait));
5423 + swait_active(&ffs->ep0req_completion.wait));
5424 kfree(ffs->dev_name);
5427 diff --git a/drivers/usb/gadget/legacy/inode.c b/drivers/usb/gadget/legacy/inode.c
5428 index 1468d8f085a3..6aae3ae25c18 100644
5429 --- a/drivers/usb/gadget/legacy/inode.c
5430 +++ b/drivers/usb/gadget/legacy/inode.c
5431 @@ -346,7 +346,7 @@ ep_io (struct ep_data *epdata, void *buf, unsigned len)
5432 spin_unlock_irq (&epdata->dev->lock);
5434 if (likely (value == 0)) {
5435 - value = wait_event_interruptible (done.wait, done.done);
5436 + value = swait_event_interruptible (done.wait, done.done);
5438 spin_lock_irq (&epdata->dev->lock);
5439 if (likely (epdata->ep != NULL)) {
5440 @@ -355,7 +355,7 @@ ep_io (struct ep_data *epdata, void *buf, unsigned len)
5441 usb_ep_dequeue (epdata->ep, epdata->req);
5442 spin_unlock_irq (&epdata->dev->lock);
5444 - wait_event (done.wait, done.done);
5445 + swait_event (done.wait, done.done);
5446 if (epdata->status == -ECONNRESET)
5447 epdata->status = -EINTR;
5449 diff --git a/fs/aio.c b/fs/aio.c
5450 index 428484f2f841..2b02e2eb2158 100644
5454 #include <linux/ramfs.h>
5455 #include <linux/percpu-refcount.h>
5456 #include <linux/mount.h>
5457 +#include <linux/swork.h>
5459 #include <asm/kmap_types.h>
5460 #include <asm/uaccess.h>
5461 @@ -115,7 +116,7 @@ struct kioctx {
5462 struct page **ring_pages;
5465 - struct work_struct free_work;
5466 + struct swork_event free_work;
5469 * signals when all in-flight requests are done
5470 @@ -258,6 +259,7 @@ static int __init aio_setup(void)
5472 .kill_sb = kill_anon_super,
5474 + BUG_ON(swork_get());
5475 aio_mnt = kern_mount(&aio_fs);
5476 if (IS_ERR(aio_mnt))
5477 panic("Failed to create aio fs mount.");
5478 @@ -581,9 +583,9 @@ static int kiocb_cancel(struct aio_kiocb *kiocb)
5479 return cancel(&kiocb->common);
5482 -static void free_ioctx(struct work_struct *work)
5483 +static void free_ioctx(struct swork_event *sev)
5485 - struct kioctx *ctx = container_of(work, struct kioctx, free_work);
5486 + struct kioctx *ctx = container_of(sev, struct kioctx, free_work);
5488 pr_debug("freeing %p\n", ctx);
5490 @@ -602,8 +604,8 @@ static void free_ioctx_reqs(struct percpu_ref *ref)
5491 if (ctx->rq_wait && atomic_dec_and_test(&ctx->rq_wait->count))
5492 complete(&ctx->rq_wait->comp);
5494 - INIT_WORK(&ctx->free_work, free_ioctx);
5495 - schedule_work(&ctx->free_work);
5496 + INIT_SWORK(&ctx->free_work, free_ioctx);
5497 + swork_queue(&ctx->free_work);
5501 @@ -611,9 +613,9 @@ static void free_ioctx_reqs(struct percpu_ref *ref)
5502 * and ctx->users has dropped to 0, so we know no more kiocbs can be submitted -
5503 * now it's safe to cancel any that need to be.
5505 -static void free_ioctx_users(struct percpu_ref *ref)
5506 +static void free_ioctx_users_work(struct swork_event *sev)
5508 - struct kioctx *ctx = container_of(ref, struct kioctx, users);
5509 + struct kioctx *ctx = container_of(sev, struct kioctx, free_work);
5510 struct aio_kiocb *req;
5512 spin_lock_irq(&ctx->ctx_lock);
5513 @@ -632,6 +634,14 @@ static void free_ioctx_users(struct percpu_ref *ref)
5514 percpu_ref_put(&ctx->reqs);
5517 +static void free_ioctx_users(struct percpu_ref *ref)
5519 + struct kioctx *ctx = container_of(ref, struct kioctx, users);
5521 + INIT_SWORK(&ctx->free_work, free_ioctx_users_work);
5522 + swork_queue(&ctx->free_work);
5525 static int ioctx_add_table(struct kioctx *ctx, struct mm_struct *mm)
5528 diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
5529 index a1fba4285277..3796769b4cd1 100644
5530 --- a/fs/autofs4/autofs_i.h
5531 +++ b/fs/autofs4/autofs_i.h
5533 #include <linux/sched.h>
5534 #include <linux/mount.h>
5535 #include <linux/namei.h>
5536 +#include <linux/delay.h>
5537 #include <asm/current.h>
5538 #include <linux/uaccess.h>
5540 diff --git a/fs/autofs4/expire.c b/fs/autofs4/expire.c
5541 index d8e6d421c27f..2e689ab1306b 100644
5542 --- a/fs/autofs4/expire.c
5543 +++ b/fs/autofs4/expire.c
5544 @@ -148,7 +148,7 @@ static struct dentry *get_next_positive_dentry(struct dentry *prev,
5545 parent = p->d_parent;
5546 if (!spin_trylock(&parent->d_lock)) {
5547 spin_unlock(&p->d_lock);
5552 spin_unlock(&p->d_lock);
5553 diff --git a/fs/buffer.c b/fs/buffer.c
5554 index b205a629001d..5646afc022ba 100644
5557 @@ -301,8 +301,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
5558 * decide that the page is now completely done.
5560 first = page_buffers(page);
5561 - local_irq_save(flags);
5562 - bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
5563 + flags = bh_uptodate_lock_irqsave(first);
5564 clear_buffer_async_read(bh);
5567 @@ -315,8 +314,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
5569 tmp = tmp->b_this_page;
5570 } while (tmp != bh);
5571 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
5572 - local_irq_restore(flags);
5573 + bh_uptodate_unlock_irqrestore(first, flags);
5576 * If none of the buffers had errors and they are all
5577 @@ -328,9 +326,7 @@ static void end_buffer_async_read(struct buffer_head *bh, int uptodate)
5581 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
5582 - local_irq_restore(flags);
5584 + bh_uptodate_unlock_irqrestore(first, flags);
5588 @@ -358,8 +354,7 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
5591 first = page_buffers(page);
5592 - local_irq_save(flags);
5593 - bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
5594 + flags = bh_uptodate_lock_irqsave(first);
5596 clear_buffer_async_write(bh);
5598 @@ -371,15 +366,12 @@ void end_buffer_async_write(struct buffer_head *bh, int uptodate)
5600 tmp = tmp->b_this_page;
5602 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
5603 - local_irq_restore(flags);
5604 + bh_uptodate_unlock_irqrestore(first, flags);
5605 end_page_writeback(page);
5609 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
5610 - local_irq_restore(flags);
5612 + bh_uptodate_unlock_irqrestore(first, flags);
5614 EXPORT_SYMBOL(end_buffer_async_write);
5616 @@ -3383,6 +3375,7 @@ struct buffer_head *alloc_buffer_head(gfp_t gfp_flags)
5617 struct buffer_head *ret = kmem_cache_zalloc(bh_cachep, gfp_flags);
5619 INIT_LIST_HEAD(&ret->b_assoc_buffers);
5620 + buffer_head_init_locks(ret);
5622 __this_cpu_inc(bh_accounting.nr);
5624 diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
5625 index a27fc8791551..791aecb7c1ac 100644
5626 --- a/fs/cifs/readdir.c
5627 +++ b/fs/cifs/readdir.c
5628 @@ -80,7 +80,7 @@ cifs_prime_dcache(struct dentry *parent, struct qstr *name,
5629 struct inode *inode;
5630 struct super_block *sb = parent->d_sb;
5631 struct cifs_sb_info *cifs_sb = CIFS_SB(sb);
5632 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
5633 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
5635 cifs_dbg(FYI, "%s: for %s\n", __func__, name->name);
5637 diff --git a/fs/dcache.c b/fs/dcache.c
5638 index 4485a48f4091..691039a6a872 100644
5642 #include <linux/mm.h>
5643 #include <linux/fs.h>
5644 #include <linux/fsnotify.h>
5645 +#include <linux/delay.h>
5646 #include <linux/slab.h>
5647 #include <linux/init.h>
5648 #include <linux/hash.h>
5649 @@ -750,6 +751,8 @@ static inline bool fast_dput(struct dentry *dentry)
5651 void dput(struct dentry *dentry)
5653 + struct dentry *parent;
5655 if (unlikely(!dentry))
5658 @@ -788,9 +791,18 @@ void dput(struct dentry *dentry)
5662 - dentry = dentry_kill(dentry);
5665 + parent = dentry_kill(dentry);
5669 + if (parent == dentry) {
5670 + /* the task with the highest priority won't schedule */
5671 + r = cond_resched();
5680 @@ -2324,7 +2336,7 @@ void d_delete(struct dentry * dentry)
5681 if (dentry->d_lockref.count == 1) {
5682 if (!spin_trylock(&inode->i_lock)) {
5683 spin_unlock(&dentry->d_lock);
5688 dentry->d_flags &= ~DCACHE_CANT_MOUNT;
5689 @@ -2384,21 +2396,24 @@ static inline void end_dir_add(struct inode *dir, unsigned n)
5691 static void d_wait_lookup(struct dentry *dentry)
5693 - if (d_in_lookup(dentry)) {
5694 - DECLARE_WAITQUEUE(wait, current);
5695 - add_wait_queue(dentry->d_wait, &wait);
5697 - set_current_state(TASK_UNINTERRUPTIBLE);
5698 - spin_unlock(&dentry->d_lock);
5700 - spin_lock(&dentry->d_lock);
5701 - } while (d_in_lookup(dentry));
5703 + struct swait_queue __wait;
5705 + if (!d_in_lookup(dentry))
5708 + INIT_LIST_HEAD(&__wait.task_list);
5710 + prepare_to_swait(dentry->d_wait, &__wait, TASK_UNINTERRUPTIBLE);
5711 + spin_unlock(&dentry->d_lock);
5713 + spin_lock(&dentry->d_lock);
5714 + } while (d_in_lookup(dentry));
5715 + finish_swait(dentry->d_wait, &__wait);
5718 struct dentry *d_alloc_parallel(struct dentry *parent,
5719 const struct qstr *name,
5720 - wait_queue_head_t *wq)
5721 + struct swait_queue_head *wq)
5723 unsigned int hash = name->hash;
5724 struct hlist_bl_head *b = in_lookup_hash(parent, hash);
5725 @@ -2507,7 +2522,7 @@ void __d_lookup_done(struct dentry *dentry)
5727 dentry->d_flags &= ~DCACHE_PAR_LOOKUP;
5728 __hlist_bl_del(&dentry->d_u.d_in_lookup_hash);
5729 - wake_up_all(dentry->d_wait);
5730 + swake_up_all(dentry->d_wait);
5731 dentry->d_wait = NULL;
5733 INIT_HLIST_NODE(&dentry->d_u.d_alias);
5734 @@ -3604,6 +3619,11 @@ EXPORT_SYMBOL(d_genocide);
5736 void __init vfs_caches_init_early(void)
5740 + for (i = 0; i < ARRAY_SIZE(in_lookup_hashtable); i++)
5741 + INIT_HLIST_BL_HEAD(&in_lookup_hashtable[i]);
5743 dcache_init_early();
5746 diff --git a/fs/eventpoll.c b/fs/eventpoll.c
5747 index 10db91218933..42af0a06f657 100644
5748 --- a/fs/eventpoll.c
5749 +++ b/fs/eventpoll.c
5750 @@ -510,12 +510,12 @@ static int ep_poll_wakeup_proc(void *priv, void *cookie, int call_nests)
5752 static void ep_poll_safewake(wait_queue_head_t *wq)
5754 - int this_cpu = get_cpu();
5755 + int this_cpu = get_cpu_light();
5757 ep_call_nested(&poll_safewake_ncalls, EP_MAX_NESTS,
5758 ep_poll_wakeup_proc, NULL, wq, (void *) (long) this_cpu);
5764 static void ep_remove_wait_queue(struct eppoll_entry *pwq)
5765 diff --git a/fs/exec.c b/fs/exec.c
5766 index 67e86571685a..fe14cdd84016 100644
5769 @@ -1017,12 +1017,14 @@ static int exec_mmap(struct mm_struct *mm)
5773 + preempt_disable_rt();
5774 active_mm = tsk->active_mm;
5776 tsk->active_mm = mm;
5777 activate_mm(active_mm, mm);
5778 tsk->mm->vmacache_seqnum = 0;
5779 vmacache_flush(tsk);
5780 + preempt_enable_rt();
5783 up_read(&old_mm->mmap_sem);
5784 diff --git a/fs/fuse/dir.c b/fs/fuse/dir.c
5785 index 642c57b8de7b..8494b9308333 100644
5788 @@ -1191,7 +1191,7 @@ static int fuse_direntplus_link(struct file *file,
5789 struct inode *dir = d_inode(parent);
5790 struct fuse_conn *fc;
5791 struct inode *inode;
5792 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
5793 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
5797 diff --git a/fs/jbd2/checkpoint.c b/fs/jbd2/checkpoint.c
5798 index 684996c8a3a4..6e18a06aaabe 100644
5799 --- a/fs/jbd2/checkpoint.c
5800 +++ b/fs/jbd2/checkpoint.c
5801 @@ -116,6 +116,8 @@ void __jbd2_log_wait_for_space(journal_t *journal)
5802 nblocks = jbd2_space_needed(journal);
5803 while (jbd2_log_space_left(journal) < nblocks) {
5804 write_unlock(&journal->j_state_lock);
5805 + if (current->plug)
5807 mutex_lock(&journal->j_checkpoint_mutex);
5810 diff --git a/fs/locks.c b/fs/locks.c
5811 index 22c5b4aa4961..269c6a44449a 100644
5814 @@ -935,7 +935,7 @@ static int flock_lock_inode(struct inode *inode, struct file_lock *request)
5818 - percpu_down_read_preempt_disable(&file_rwsem);
5819 + percpu_down_read(&file_rwsem);
5820 spin_lock(&ctx->flc_lock);
5821 if (request->fl_flags & FL_ACCESS)
5823 @@ -976,7 +976,7 @@ static int flock_lock_inode(struct inode *inode, struct file_lock *request)
5826 spin_unlock(&ctx->flc_lock);
5827 - percpu_up_read_preempt_enable(&file_rwsem);
5828 + percpu_up_read(&file_rwsem);
5830 locks_free_lock(new_fl);
5831 locks_dispose_list(&dispose);
5832 @@ -1013,7 +1013,7 @@ static int posix_lock_inode(struct inode *inode, struct file_lock *request,
5833 new_fl2 = locks_alloc_lock();
5836 - percpu_down_read_preempt_disable(&file_rwsem);
5837 + percpu_down_read(&file_rwsem);
5838 spin_lock(&ctx->flc_lock);
5840 * New lock request. Walk all POSIX locks and look for conflicts. If
5841 @@ -1185,7 +1185,7 @@ static int posix_lock_inode(struct inode *inode, struct file_lock *request,
5844 spin_unlock(&ctx->flc_lock);
5845 - percpu_up_read_preempt_enable(&file_rwsem);
5846 + percpu_up_read(&file_rwsem);
5848 * Free any unused locks.
5850 @@ -1460,7 +1460,7 @@ int __break_lease(struct inode *inode, unsigned int mode, unsigned int type)
5854 - percpu_down_read_preempt_disable(&file_rwsem);
5855 + percpu_down_read(&file_rwsem);
5856 spin_lock(&ctx->flc_lock);
5858 time_out_leases(inode, &dispose);
5859 @@ -1512,13 +1512,13 @@ int __break_lease(struct inode *inode, unsigned int mode, unsigned int type)
5860 locks_insert_block(fl, new_fl);
5861 trace_break_lease_block(inode, new_fl);
5862 spin_unlock(&ctx->flc_lock);
5863 - percpu_up_read_preempt_enable(&file_rwsem);
5864 + percpu_up_read(&file_rwsem);
5866 locks_dispose_list(&dispose);
5867 error = wait_event_interruptible_timeout(new_fl->fl_wait,
5868 !new_fl->fl_next, break_time);
5870 - percpu_down_read_preempt_disable(&file_rwsem);
5871 + percpu_down_read(&file_rwsem);
5872 spin_lock(&ctx->flc_lock);
5873 trace_break_lease_unblock(inode, new_fl);
5874 locks_delete_block(new_fl);
5875 @@ -1535,7 +1535,7 @@ int __break_lease(struct inode *inode, unsigned int mode, unsigned int type)
5878 spin_unlock(&ctx->flc_lock);
5879 - percpu_up_read_preempt_enable(&file_rwsem);
5880 + percpu_up_read(&file_rwsem);
5881 locks_dispose_list(&dispose);
5882 locks_free_lock(new_fl);
5884 @@ -1609,7 +1609,7 @@ int fcntl_getlease(struct file *filp)
5886 ctx = smp_load_acquire(&inode->i_flctx);
5887 if (ctx && !list_empty_careful(&ctx->flc_lease)) {
5888 - percpu_down_read_preempt_disable(&file_rwsem);
5889 + percpu_down_read(&file_rwsem);
5890 spin_lock(&ctx->flc_lock);
5891 time_out_leases(inode, &dispose);
5892 list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
5893 @@ -1619,7 +1619,7 @@ int fcntl_getlease(struct file *filp)
5896 spin_unlock(&ctx->flc_lock);
5897 - percpu_up_read_preempt_enable(&file_rwsem);
5898 + percpu_up_read(&file_rwsem);
5900 locks_dispose_list(&dispose);
5902 @@ -1694,7 +1694,7 @@ generic_add_lease(struct file *filp, long arg, struct file_lock **flp, void **pr
5906 - percpu_down_read_preempt_disable(&file_rwsem);
5907 + percpu_down_read(&file_rwsem);
5908 spin_lock(&ctx->flc_lock);
5909 time_out_leases(inode, &dispose);
5910 error = check_conflicting_open(dentry, arg, lease->fl_flags);
5911 @@ -1765,7 +1765,7 @@ generic_add_lease(struct file *filp, long arg, struct file_lock **flp, void **pr
5912 lease->fl_lmops->lm_setup(lease, priv);
5914 spin_unlock(&ctx->flc_lock);
5915 - percpu_up_read_preempt_enable(&file_rwsem);
5916 + percpu_up_read(&file_rwsem);
5917 locks_dispose_list(&dispose);
5919 inode_unlock(inode);
5920 @@ -1788,7 +1788,7 @@ static int generic_delete_lease(struct file *filp, void *owner)
5924 - percpu_down_read_preempt_disable(&file_rwsem);
5925 + percpu_down_read(&file_rwsem);
5926 spin_lock(&ctx->flc_lock);
5927 list_for_each_entry(fl, &ctx->flc_lease, fl_list) {
5928 if (fl->fl_file == filp &&
5929 @@ -1801,7 +1801,7 @@ static int generic_delete_lease(struct file *filp, void *owner)
5931 error = fl->fl_lmops->lm_change(victim, F_UNLCK, &dispose);
5932 spin_unlock(&ctx->flc_lock);
5933 - percpu_up_read_preempt_enable(&file_rwsem);
5934 + percpu_up_read(&file_rwsem);
5935 locks_dispose_list(&dispose);
5938 @@ -2532,13 +2532,13 @@ locks_remove_lease(struct file *filp, struct file_lock_context *ctx)
5939 if (list_empty(&ctx->flc_lease))
5942 - percpu_down_read_preempt_disable(&file_rwsem);
5943 + percpu_down_read(&file_rwsem);
5944 spin_lock(&ctx->flc_lock);
5945 list_for_each_entry_safe(fl, tmp, &ctx->flc_lease, fl_list)
5946 if (filp == fl->fl_file)
5947 lease_modify(fl, F_UNLCK, &dispose);
5948 spin_unlock(&ctx->flc_lock);
5949 - percpu_up_read_preempt_enable(&file_rwsem);
5950 + percpu_up_read(&file_rwsem);
5952 locks_dispose_list(&dispose);
5954 diff --git a/fs/namei.c b/fs/namei.c
5955 index 5b4eed221530..9c8dd3c83a80 100644
5958 @@ -1629,7 +1629,7 @@ static struct dentry *lookup_slow(const struct qstr *name,
5960 struct dentry *dentry = ERR_PTR(-ENOENT), *old;
5961 struct inode *inode = dir->d_inode;
5962 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
5963 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
5965 inode_lock_shared(inode);
5966 /* Don't go there if it's already dead */
5967 @@ -3086,7 +3086,7 @@ static int lookup_open(struct nameidata *nd, struct path *path,
5968 struct dentry *dentry;
5969 int error, create_error = 0;
5970 umode_t mode = op->mode;
5971 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
5972 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
5974 if (unlikely(IS_DEADDIR(dir_inode)))
5976 diff --git a/fs/namespace.c b/fs/namespace.c
5977 index 7cea503ae06d..cb15f5397991 100644
5978 --- a/fs/namespace.c
5979 +++ b/fs/namespace.c
5981 #include <linux/mnt_namespace.h>
5982 #include <linux/user_namespace.h>
5983 #include <linux/namei.h>
5984 +#include <linux/delay.h>
5985 #include <linux/security.h>
5986 #include <linux/idr.h>
5987 #include <linux/init.h> /* init_rootfs */
5988 @@ -356,8 +357,11 @@ int __mnt_want_write(struct vfsmount *m)
5989 * incremented count after it has set MNT_WRITE_HOLD.
5992 - while (ACCESS_ONCE(mnt->mnt.mnt_flags) & MNT_WRITE_HOLD)
5994 + while (ACCESS_ONCE(mnt->mnt.mnt_flags) & MNT_WRITE_HOLD) {
5997 + preempt_disable();
6000 * After the slowpath clears MNT_WRITE_HOLD, mnt_is_readonly will
6001 * be set to match its requirements. So we must not load that until
6002 diff --git a/fs/nfs/delegation.c b/fs/nfs/delegation.c
6003 index dff600ae0d74..d726d2e09353 100644
6004 --- a/fs/nfs/delegation.c
6005 +++ b/fs/nfs/delegation.c
6006 @@ -150,11 +150,11 @@ static int nfs_delegation_claim_opens(struct inode *inode,
6008 /* Block nfs4_proc_unlck */
6009 mutex_lock(&sp->so_delegreturn_mutex);
6010 - seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
6011 + seq = read_seqbegin(&sp->so_reclaim_seqlock);
6012 err = nfs4_open_delegation_recall(ctx, state, stateid, type);
6014 err = nfs_delegation_claim_locks(ctx, state, stateid);
6015 - if (!err && read_seqcount_retry(&sp->so_reclaim_seqcount, seq))
6016 + if (!err && read_seqretry(&sp->so_reclaim_seqlock, seq))
6018 mutex_unlock(&sp->so_delegreturn_mutex);
6019 put_nfs_open_context(ctx);
6020 diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
6021 index 53e02b8bd9bd..a66e7d77cfbb 100644
6024 @@ -485,7 +485,7 @@ static
6025 void nfs_prime_dcache(struct dentry *parent, struct nfs_entry *entry)
6027 struct qstr filename = QSTR_INIT(entry->name, entry->len);
6028 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
6029 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
6030 struct dentry *dentry;
6031 struct dentry *alias;
6032 struct inode *dir = d_inode(parent);
6033 @@ -1487,7 +1487,7 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
6034 struct file *file, unsigned open_flags,
6035 umode_t mode, int *opened)
6037 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
6038 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
6039 struct nfs_open_context *ctx;
6041 struct iattr attr = { .ia_valid = ATTR_OPEN };
6042 @@ -1802,7 +1802,11 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry)
6044 trace_nfs_rmdir_enter(dir, dentry);
6045 if (d_really_is_positive(dentry)) {
6046 +#ifdef CONFIG_PREEMPT_RT_BASE
6047 + down(&NFS_I(d_inode(dentry))->rmdir_sem);
6049 down_write(&NFS_I(d_inode(dentry))->rmdir_sem);
6051 error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
6052 /* Ensure the VFS deletes this inode */
6054 @@ -1812,7 +1816,11 @@ int nfs_rmdir(struct inode *dir, struct dentry *dentry)
6056 nfs_dentry_handle_enoent(dentry);
6058 +#ifdef CONFIG_PREEMPT_RT_BASE
6059 + up(&NFS_I(d_inode(dentry))->rmdir_sem);
6061 up_write(&NFS_I(d_inode(dentry))->rmdir_sem);
6064 error = NFS_PROTO(dir)->rmdir(dir, &dentry->d_name);
6065 trace_nfs_rmdir_exit(dir, dentry, error);
6066 diff --git a/fs/nfs/inode.c b/fs/nfs/inode.c
6067 index bf4ec5ecc97e..36cd5fc9192c 100644
6068 --- a/fs/nfs/inode.c
6069 +++ b/fs/nfs/inode.c
6070 @@ -1957,7 +1957,11 @@ static void init_once(void *foo)
6071 nfsi->nrequests = 0;
6072 nfsi->commit_info.ncommit = 0;
6073 atomic_set(&nfsi->commit_info.rpcs_out, 0);
6074 +#ifdef CONFIG_PREEMPT_RT_BASE
6075 + sema_init(&nfsi->rmdir_sem, 1);
6077 init_rwsem(&nfsi->rmdir_sem);
6079 nfs4_init_once(nfsi);
6082 diff --git a/fs/nfs/nfs4_fs.h b/fs/nfs/nfs4_fs.h
6083 index 1452177c822d..f43b01d54c59 100644
6084 --- a/fs/nfs/nfs4_fs.h
6085 +++ b/fs/nfs/nfs4_fs.h
6086 @@ -111,7 +111,7 @@ struct nfs4_state_owner {
6087 unsigned long so_flags;
6088 struct list_head so_states;
6089 struct nfs_seqid_counter so_seqid;
6090 - seqcount_t so_reclaim_seqcount;
6091 + seqlock_t so_reclaim_seqlock;
6092 struct mutex so_delegreturn_mutex;
6095 diff --git a/fs/nfs/nfs4proc.c b/fs/nfs/nfs4proc.c
6096 index 78ff8b63d5f7..3573653fd5cc 100644
6097 --- a/fs/nfs/nfs4proc.c
6098 +++ b/fs/nfs/nfs4proc.c
6099 @@ -2698,7 +2698,7 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
6103 - seq = raw_seqcount_begin(&sp->so_reclaim_seqcount);
6104 + seq = raw_seqcount_begin(&sp->so_reclaim_seqlock.seqcount);
6106 ret = _nfs4_proc_open(opendata);
6108 @@ -2736,7 +2736,7 @@ static int _nfs4_open_and_get_state(struct nfs4_opendata *opendata,
6110 if (d_inode(dentry) == state->inode) {
6111 nfs_inode_attach_open_context(ctx);
6112 - if (read_seqcount_retry(&sp->so_reclaim_seqcount, seq))
6113 + if (read_seqretry(&sp->so_reclaim_seqlock, seq))
6114 nfs4_schedule_stateid_recovery(server, state);
6117 diff --git a/fs/nfs/nfs4state.c b/fs/nfs/nfs4state.c
6118 index 0959c9661662..dabd834d7686 100644
6119 --- a/fs/nfs/nfs4state.c
6120 +++ b/fs/nfs/nfs4state.c
6121 @@ -488,7 +488,7 @@ nfs4_alloc_state_owner(struct nfs_server *server,
6122 nfs4_init_seqid_counter(&sp->so_seqid);
6123 atomic_set(&sp->so_count, 1);
6124 INIT_LIST_HEAD(&sp->so_lru);
6125 - seqcount_init(&sp->so_reclaim_seqcount);
6126 + seqlock_init(&sp->so_reclaim_seqlock);
6127 mutex_init(&sp->so_delegreturn_mutex);
6130 @@ -1497,8 +1497,12 @@ static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, const struct nfs
6131 * recovering after a network partition or a reboot from a
6132 * server that doesn't support a grace period.
6134 +#ifdef CONFIG_PREEMPT_RT_FULL
6135 + write_seqlock(&sp->so_reclaim_seqlock);
6137 + write_seqcount_begin(&sp->so_reclaim_seqlock.seqcount);
6139 spin_lock(&sp->so_lock);
6140 - raw_write_seqcount_begin(&sp->so_reclaim_seqcount);
6142 list_for_each_entry(state, &sp->so_states, open_states) {
6143 if (!test_and_clear_bit(ops->state_flag_bit, &state->flags))
6144 @@ -1567,14 +1571,20 @@ static int nfs4_reclaim_open_state(struct nfs4_state_owner *sp, const struct nfs
6145 spin_lock(&sp->so_lock);
6148 - raw_write_seqcount_end(&sp->so_reclaim_seqcount);
6149 spin_unlock(&sp->so_lock);
6150 +#ifdef CONFIG_PREEMPT_RT_FULL
6151 + write_sequnlock(&sp->so_reclaim_seqlock);
6153 + write_seqcount_end(&sp->so_reclaim_seqlock.seqcount);
6157 nfs4_put_open_state(state);
6158 - spin_lock(&sp->so_lock);
6159 - raw_write_seqcount_end(&sp->so_reclaim_seqcount);
6160 - spin_unlock(&sp->so_lock);
6161 +#ifdef CONFIG_PREEMPT_RT_FULL
6162 + write_sequnlock(&sp->so_reclaim_seqlock);
6164 + write_seqcount_end(&sp->so_reclaim_seqlock.seqcount);
6169 diff --git a/fs/nfs/unlink.c b/fs/nfs/unlink.c
6170 index 191aa577dd1f..58990c8f52e0 100644
6171 --- a/fs/nfs/unlink.c
6172 +++ b/fs/nfs/unlink.c
6174 #include <linux/sunrpc/clnt.h>
6175 #include <linux/nfs_fs.h>
6176 #include <linux/sched.h>
6177 -#include <linux/wait.h>
6178 +#include <linux/swait.h>
6179 #include <linux/namei.h>
6180 #include <linux/fsnotify.h>
6182 @@ -51,6 +51,29 @@ static void nfs_async_unlink_done(struct rpc_task *task, void *calldata)
6183 rpc_restart_call_prepare(task);
6186 +#ifdef CONFIG_PREEMPT_RT_BASE
6187 +static void nfs_down_anon(struct semaphore *sema)
6192 +static void nfs_up_anon(struct semaphore *sema)
6198 +static void nfs_down_anon(struct rw_semaphore *rwsem)
6200 + down_read_non_owner(rwsem);
6203 +static void nfs_up_anon(struct rw_semaphore *rwsem)
6205 + up_read_non_owner(rwsem);
6210 * nfs_async_unlink_release - Release the sillydelete data.
6211 * @task: rpc_task of the sillydelete
6212 @@ -64,7 +87,7 @@ static void nfs_async_unlink_release(void *calldata)
6213 struct dentry *dentry = data->dentry;
6214 struct super_block *sb = dentry->d_sb;
6216 - up_read_non_owner(&NFS_I(d_inode(dentry->d_parent))->rmdir_sem);
6217 + nfs_up_anon(&NFS_I(d_inode(dentry->d_parent))->rmdir_sem);
6218 d_lookup_done(dentry);
6219 nfs_free_unlinkdata(data);
6221 @@ -117,10 +140,10 @@ static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data)
6222 struct inode *dir = d_inode(dentry->d_parent);
6223 struct dentry *alias;
6225 - down_read_non_owner(&NFS_I(dir)->rmdir_sem);
6226 + nfs_down_anon(&NFS_I(dir)->rmdir_sem);
6227 alias = d_alloc_parallel(dentry->d_parent, &data->args.name, &data->wq);
6228 if (IS_ERR(alias)) {
6229 - up_read_non_owner(&NFS_I(dir)->rmdir_sem);
6230 + nfs_up_anon(&NFS_I(dir)->rmdir_sem);
6233 if (!d_in_lookup(alias)) {
6234 @@ -142,7 +165,7 @@ static int nfs_call_unlink(struct dentry *dentry, struct nfs_unlinkdata *data)
6236 spin_unlock(&alias->d_lock);
6238 - up_read_non_owner(&NFS_I(dir)->rmdir_sem);
6239 + nfs_up_anon(&NFS_I(dir)->rmdir_sem);
6241 * If we'd displaced old cached devname, free it. At that
6242 * point dentry is definitely not a root, so we won't need
6243 @@ -182,7 +205,7 @@ nfs_async_unlink(struct dentry *dentry, const struct qstr *name)
6246 data->res.dir_attr = &data->dir_attr;
6247 - init_waitqueue_head(&data->wq);
6248 + init_swait_queue_head(&data->wq);
6251 spin_lock(&dentry->d_lock);
6252 diff --git a/fs/ntfs/aops.c b/fs/ntfs/aops.c
6253 index fe251f187ff8..e89da4fb14c2 100644
6254 --- a/fs/ntfs/aops.c
6255 +++ b/fs/ntfs/aops.c
6256 @@ -92,13 +92,13 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
6258 if (file_ofs < init_size)
6259 ofs = init_size - file_ofs;
6260 - local_irq_save(flags);
6261 + local_irq_save_nort(flags);
6262 kaddr = kmap_atomic(page);
6263 memset(kaddr + bh_offset(bh) + ofs, 0,
6265 flush_dcache_page(page);
6266 kunmap_atomic(kaddr);
6267 - local_irq_restore(flags);
6268 + local_irq_restore_nort(flags);
6271 clear_buffer_uptodate(bh);
6272 @@ -107,8 +107,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
6273 "0x%llx.", (unsigned long long)bh->b_blocknr);
6275 first = page_buffers(page);
6276 - local_irq_save(flags);
6277 - bit_spin_lock(BH_Uptodate_Lock, &first->b_state);
6278 + flags = bh_uptodate_lock_irqsave(first);
6279 clear_buffer_async_read(bh);
6282 @@ -123,8 +122,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
6284 tmp = tmp->b_this_page;
6285 } while (tmp != bh);
6286 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
6287 - local_irq_restore(flags);
6288 + bh_uptodate_unlock_irqrestore(first, flags);
6290 * If none of the buffers had errors then we can set the page uptodate,
6291 * but we first have to perform the post read mst fixups, if the
6292 @@ -145,13 +143,13 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
6293 recs = PAGE_SIZE / rec_size;
6294 /* Should have been verified before we got here... */
6296 - local_irq_save(flags);
6297 + local_irq_save_nort(flags);
6298 kaddr = kmap_atomic(page);
6299 for (i = 0; i < recs; i++)
6300 post_read_mst_fixup((NTFS_RECORD*)(kaddr +
6301 i * rec_size), rec_size);
6302 kunmap_atomic(kaddr);
6303 - local_irq_restore(flags);
6304 + local_irq_restore_nort(flags);
6305 flush_dcache_page(page);
6306 if (likely(page_uptodate && !PageError(page)))
6307 SetPageUptodate(page);
6308 @@ -159,9 +157,7 @@ static void ntfs_end_buffer_async_read(struct buffer_head *bh, int uptodate)
6312 - bit_spin_unlock(BH_Uptodate_Lock, &first->b_state);
6313 - local_irq_restore(flags);
6315 + bh_uptodate_unlock_irqrestore(first, flags);
6319 diff --git a/fs/proc/base.c b/fs/proc/base.c
6320 index ca651ac00660..41d9dc789285 100644
6321 --- a/fs/proc/base.c
6322 +++ b/fs/proc/base.c
6323 @@ -1834,7 +1834,7 @@ bool proc_fill_cache(struct file *file, struct dir_context *ctx,
6325 child = d_hash_and_lookup(dir, &qname);
6327 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
6328 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
6329 child = d_alloc_parallel(dir, &qname, &wq);
6331 goto end_instantiate;
6332 diff --git a/fs/proc/proc_sysctl.c b/fs/proc/proc_sysctl.c
6333 index d4e37acd4821..000cea46434a 100644
6334 --- a/fs/proc/proc_sysctl.c
6335 +++ b/fs/proc/proc_sysctl.c
6336 @@ -632,7 +632,7 @@ static bool proc_sys_fill_cache(struct file *file,
6338 child = d_lookup(dir, &qname);
6340 - DECLARE_WAIT_QUEUE_HEAD_ONSTACK(wq);
6341 + DECLARE_SWAIT_QUEUE_HEAD_ONSTACK(wq);
6342 child = d_alloc_parallel(dir, &qname, &wq);
6345 diff --git a/fs/timerfd.c b/fs/timerfd.c
6346 index 9ae4abb4110b..8644b67c48fd 100644
6349 @@ -460,7 +460,10 @@ static int do_timerfd_settime(int ufd, int flags,
6352 spin_unlock_irq(&ctx->wqh.lock);
6355 + hrtimer_wait_for_timer(&ctx->t.alarm.timer);
6357 + hrtimer_wait_for_timer(&ctx->t.tmr);
6361 diff --git a/include/acpi/platform/aclinux.h b/include/acpi/platform/aclinux.h
6362 index e861a24f06f2..b5c97d3059c7 100644
6363 --- a/include/acpi/platform/aclinux.h
6364 +++ b/include/acpi/platform/aclinux.h
6367 #define acpi_cache_t struct kmem_cache
6368 #define acpi_spinlock spinlock_t *
6369 +#define acpi_raw_spinlock raw_spinlock_t *
6370 #define acpi_cpu_flags unsigned long
6372 /* Use native linux version of acpi_os_allocate_zeroed */
6373 @@ -151,6 +152,20 @@
6374 #define ACPI_USE_ALTERNATE_PROTOTYPE_acpi_os_get_thread_id
6375 #define ACPI_USE_ALTERNATE_PROTOTYPE_acpi_os_create_lock
6377 +#define acpi_os_create_raw_lock(__handle) \
6379 + raw_spinlock_t *lock = ACPI_ALLOCATE(sizeof(*lock)); \
6382 + *(__handle) = lock; \
6383 + raw_spin_lock_init(*(__handle)); \
6385 + lock ? AE_OK : AE_NO_MEMORY; \
6388 +#define acpi_os_delete_raw_lock(__handle) kfree(__handle)
6392 * OSL interfaces used by debugger/disassembler
6394 diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
6395 index 6f96247226a4..fa53a21263c2 100644
6396 --- a/include/asm-generic/bug.h
6397 +++ b/include/asm-generic/bug.h
6398 @@ -215,6 +215,20 @@ void __warn(const char *file, int line, void *caller, unsigned taint,
6399 # define WARN_ON_SMP(x) ({0;})
6402 +#ifdef CONFIG_PREEMPT_RT_BASE
6403 +# define BUG_ON_RT(c) BUG_ON(c)
6404 +# define BUG_ON_NONRT(c) do { } while (0)
6405 +# define WARN_ON_RT(condition) WARN_ON(condition)
6406 +# define WARN_ON_NONRT(condition) do { } while (0)
6407 +# define WARN_ON_ONCE_NONRT(condition) do { } while (0)
6409 +# define BUG_ON_RT(c) do { } while (0)
6410 +# define BUG_ON_NONRT(c) BUG_ON(c)
6411 +# define WARN_ON_RT(condition) do { } while (0)
6412 +# define WARN_ON_NONRT(condition) WARN_ON(condition)
6413 +# define WARN_ON_ONCE_NONRT(condition) WARN_ON_ONCE(condition)
6416 #endif /* __ASSEMBLY__ */
6419 diff --git a/include/linux/blk-mq.h b/include/linux/blk-mq.h
6420 index 535ab2e13d2e..cfc246899473 100644
6421 --- a/include/linux/blk-mq.h
6422 +++ b/include/linux/blk-mq.h
6423 @@ -209,7 +209,7 @@ static inline u16 blk_mq_unique_tag_to_tag(u32 unique_tag)
6424 return unique_tag & BLK_MQ_UNIQUE_TAG_MASK;
6428 +void __blk_mq_complete_request_remote_work(struct work_struct *work);
6429 int blk_mq_request_started(struct request *rq);
6430 void blk_mq_start_request(struct request *rq);
6431 void blk_mq_end_request(struct request *rq, int error);
6432 diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h
6433 index f6a816129856..ec7a4676f8a8 100644
6434 --- a/include/linux/blkdev.h
6435 +++ b/include/linux/blkdev.h
6436 @@ -89,6 +89,7 @@ struct request {
6437 struct list_head queuelist;
6439 struct call_single_data csd;
6440 + struct work_struct work;
6444 @@ -467,7 +468,7 @@ struct request_queue {
6445 struct throtl_data *td;
6447 struct rcu_head rcu_head;
6448 - wait_queue_head_t mq_freeze_wq;
6449 + struct swait_queue_head mq_freeze_wq;
6450 struct percpu_ref q_usage_counter;
6451 struct list_head all_q_node;
6453 diff --git a/include/linux/bottom_half.h b/include/linux/bottom_half.h
6454 index 8fdcb783197d..d07dbeec7bc1 100644
6455 --- a/include/linux/bottom_half.h
6456 +++ b/include/linux/bottom_half.h
6459 #include <linux/preempt.h>
6461 +#ifdef CONFIG_PREEMPT_RT_FULL
6463 +extern void __local_bh_disable(void);
6464 +extern void _local_bh_enable(void);
6465 +extern void __local_bh_enable(void);
6467 +static inline void local_bh_disable(void)
6469 + __local_bh_disable();
6472 +static inline void __local_bh_disable_ip(unsigned long ip, unsigned int cnt)
6474 + __local_bh_disable();
6477 +static inline void local_bh_enable(void)
6479 + __local_bh_enable();
6482 +static inline void __local_bh_enable_ip(unsigned long ip, unsigned int cnt)
6484 + __local_bh_enable();
6487 +static inline void local_bh_enable_ip(unsigned long ip)
6489 + __local_bh_enable();
6494 #ifdef CONFIG_TRACE_IRQFLAGS
6495 extern void __local_bh_disable_ip(unsigned long ip, unsigned int cnt);
6497 @@ -30,5 +63,6 @@ static inline void local_bh_enable(void)
6499 __local_bh_enable_ip(_THIS_IP_, SOFTIRQ_DISABLE_OFFSET);
6503 #endif /* _LINUX_BH_H */
6504 diff --git a/include/linux/buffer_head.h b/include/linux/buffer_head.h
6505 index ebbacd14d450..be5e87f6360a 100644
6506 --- a/include/linux/buffer_head.h
6507 +++ b/include/linux/buffer_head.h
6508 @@ -75,8 +75,50 @@ struct buffer_head {
6509 struct address_space *b_assoc_map; /* mapping this buffer is
6511 atomic_t b_count; /* users using this buffer_head */
6512 +#ifdef CONFIG_PREEMPT_RT_BASE
6513 + spinlock_t b_uptodate_lock;
6514 +#if IS_ENABLED(CONFIG_JBD2)
6515 + spinlock_t b_state_lock;
6516 + spinlock_t b_journal_head_lock;
6521 +static inline unsigned long bh_uptodate_lock_irqsave(struct buffer_head *bh)
6523 + unsigned long flags;
6525 +#ifndef CONFIG_PREEMPT_RT_BASE
6526 + local_irq_save(flags);
6527 + bit_spin_lock(BH_Uptodate_Lock, &bh->b_state);
6529 + spin_lock_irqsave(&bh->b_uptodate_lock, flags);
6535 +bh_uptodate_unlock_irqrestore(struct buffer_head *bh, unsigned long flags)
6537 +#ifndef CONFIG_PREEMPT_RT_BASE
6538 + bit_spin_unlock(BH_Uptodate_Lock, &bh->b_state);
6539 + local_irq_restore(flags);
6541 + spin_unlock_irqrestore(&bh->b_uptodate_lock, flags);
6545 +static inline void buffer_head_init_locks(struct buffer_head *bh)
6547 +#ifdef CONFIG_PREEMPT_RT_BASE
6548 + spin_lock_init(&bh->b_uptodate_lock);
6549 +#if IS_ENABLED(CONFIG_JBD2)
6550 + spin_lock_init(&bh->b_state_lock);
6551 + spin_lock_init(&bh->b_journal_head_lock);
6557 * macro tricks to expand the set_buffer_foo(), clear_buffer_foo()
6558 * and buffer_foo() functions.
6559 diff --git a/include/linux/cgroup-defs.h b/include/linux/cgroup-defs.h
6560 index 5b17de62c962..56027cc01a56 100644
6561 --- a/include/linux/cgroup-defs.h
6562 +++ b/include/linux/cgroup-defs.h
6564 #include <linux/percpu-refcount.h>
6565 #include <linux/percpu-rwsem.h>
6566 #include <linux/workqueue.h>
6567 +#include <linux/swork.h>
6569 #ifdef CONFIG_CGROUPS
6571 @@ -137,6 +138,7 @@ struct cgroup_subsys_state {
6572 /* percpu_ref killing and RCU release */
6573 struct rcu_head rcu_head;
6574 struct work_struct destroy_work;
6575 + struct swork_event destroy_swork;
6579 diff --git a/include/linux/completion.h b/include/linux/completion.h
6580 index 5d5aaae3af43..3bca1590e29f 100644
6581 --- a/include/linux/completion.h
6582 +++ b/include/linux/completion.h
6584 * Atomic wait-for-completion handler data structures.
6585 * See kernel/sched/completion.c for details.
6588 -#include <linux/wait.h>
6589 +#include <linux/swait.h>
6592 * struct completion - structure used to maintain state for a "completion"
6597 - wait_queue_head_t wait;
6598 + struct swait_queue_head wait;
6601 #define COMPLETION_INITIALIZER(work) \
6602 - { 0, __WAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
6603 + { 0, __SWAIT_QUEUE_HEAD_INITIALIZER((work).wait) }
6605 #define COMPLETION_INITIALIZER_ONSTACK(work) \
6606 ({ init_completion(&work); work; })
6607 @@ -73,7 +72,7 @@ struct completion {
6608 static inline void init_completion(struct completion *x)
6611 - init_waitqueue_head(&x->wait);
6612 + init_swait_queue_head(&x->wait);
6616 diff --git a/include/linux/cpu.h b/include/linux/cpu.h
6617 index e571128ad99a..5e52d28c20c1 100644
6618 --- a/include/linux/cpu.h
6619 +++ b/include/linux/cpu.h
6620 @@ -182,6 +182,8 @@ extern void get_online_cpus(void);
6621 extern void put_online_cpus(void);
6622 extern void cpu_hotplug_disable(void);
6623 extern void cpu_hotplug_enable(void);
6624 +extern void pin_current_cpu(void);
6625 +extern void unpin_current_cpu(void);
6626 #define hotcpu_notifier(fn, pri) cpu_notifier(fn, pri)
6627 #define __hotcpu_notifier(fn, pri) __cpu_notifier(fn, pri)
6628 #define register_hotcpu_notifier(nb) register_cpu_notifier(nb)
6629 @@ -199,6 +201,8 @@ static inline void cpu_hotplug_done(void) {}
6630 #define put_online_cpus() do { } while (0)
6631 #define cpu_hotplug_disable() do { } while (0)
6632 #define cpu_hotplug_enable() do { } while (0)
6633 +static inline void pin_current_cpu(void) { }
6634 +static inline void unpin_current_cpu(void) { }
6635 #define hotcpu_notifier(fn, pri) do { (void)(fn); } while (0)
6636 #define __hotcpu_notifier(fn, pri) do { (void)(fn); } while (0)
6637 /* These aren't inline functions due to a GCC bug. */
6638 diff --git a/include/linux/dcache.h b/include/linux/dcache.h
6639 index 5beed7b30561..61cab7ef458e 100644
6640 --- a/include/linux/dcache.h
6641 +++ b/include/linux/dcache.h
6643 #include <linux/rcupdate.h>
6644 #include <linux/lockref.h>
6645 #include <linux/stringhash.h>
6646 +#include <linux/wait.h>
6650 @@ -100,7 +101,7 @@ struct dentry {
6653 struct list_head d_lru; /* LRU list */
6654 - wait_queue_head_t *d_wait; /* in-lookup ones only */
6655 + struct swait_queue_head *d_wait; /* in-lookup ones only */
6657 struct list_head d_child; /* child of parent list */
6658 struct list_head d_subdirs; /* our children */
6659 @@ -230,7 +231,7 @@ extern void d_set_d_op(struct dentry *dentry, const struct dentry_operations *op
6660 extern struct dentry * d_alloc(struct dentry *, const struct qstr *);
6661 extern struct dentry * d_alloc_pseudo(struct super_block *, const struct qstr *);
6662 extern struct dentry * d_alloc_parallel(struct dentry *, const struct qstr *,
6663 - wait_queue_head_t *);
6664 + struct swait_queue_head *);
6665 extern struct dentry * d_splice_alias(struct inode *, struct dentry *);
6666 extern struct dentry * d_add_ci(struct dentry *, struct inode *, struct qstr *);
6667 extern struct dentry * d_exact_alias(struct dentry *, struct inode *);
6668 diff --git a/include/linux/delay.h b/include/linux/delay.h
6669 index a6ecb34cf547..37caab306336 100644
6670 --- a/include/linux/delay.h
6671 +++ b/include/linux/delay.h
6672 @@ -52,4 +52,10 @@ static inline void ssleep(unsigned int seconds)
6673 msleep(seconds * 1000);
6676 +#ifdef CONFIG_PREEMPT_RT_FULL
6677 +extern void cpu_chill(void);
6679 +# define cpu_chill() cpu_relax()
6682 #endif /* defined(_LINUX_DELAY_H) */
6683 diff --git a/include/linux/highmem.h b/include/linux/highmem.h
6684 index bb3f3297062a..a117a33ef72c 100644
6685 --- a/include/linux/highmem.h
6686 +++ b/include/linux/highmem.h
6688 #include <linux/mm.h>
6689 #include <linux/uaccess.h>
6690 #include <linux/hardirq.h>
6691 +#include <linux/sched.h>
6693 #include <asm/cacheflush.h>
6695 @@ -65,7 +66,7 @@ static inline void kunmap(struct page *page)
6697 static inline void *kmap_atomic(struct page *page)
6699 - preempt_disable();
6700 + preempt_disable_nort();
6701 pagefault_disable();
6702 return page_address(page);
6704 @@ -74,7 +75,7 @@ static inline void *kmap_atomic(struct page *page)
6705 static inline void __kunmap_atomic(void *addr)
6709 + preempt_enable_nort();
6712 #define kmap_atomic_pfn(pfn) kmap_atomic(pfn_to_page(pfn))
6713 @@ -86,32 +87,51 @@ static inline void __kunmap_atomic(void *addr)
6715 #if defined(CONFIG_HIGHMEM) || defined(CONFIG_X86_32)
6717 +#ifndef CONFIG_PREEMPT_RT_FULL
6718 DECLARE_PER_CPU(int, __kmap_atomic_idx);
6721 static inline int kmap_atomic_idx_push(void)
6723 +#ifndef CONFIG_PREEMPT_RT_FULL
6724 int idx = __this_cpu_inc_return(__kmap_atomic_idx) - 1;
6726 -#ifdef CONFIG_DEBUG_HIGHMEM
6727 +# ifdef CONFIG_DEBUG_HIGHMEM
6728 WARN_ON_ONCE(in_irq() && !irqs_disabled());
6729 BUG_ON(idx >= KM_TYPE_NR);
6734 + current->kmap_idx++;
6735 + BUG_ON(current->kmap_idx > KM_TYPE_NR);
6736 + return current->kmap_idx - 1;
6740 static inline int kmap_atomic_idx(void)
6742 +#ifndef CONFIG_PREEMPT_RT_FULL
6743 return __this_cpu_read(__kmap_atomic_idx) - 1;
6745 + return current->kmap_idx - 1;
6749 static inline void kmap_atomic_idx_pop(void)
6751 -#ifdef CONFIG_DEBUG_HIGHMEM
6752 +#ifndef CONFIG_PREEMPT_RT_FULL
6753 +# ifdef CONFIG_DEBUG_HIGHMEM
6754 int idx = __this_cpu_dec_return(__kmap_atomic_idx);
6759 __this_cpu_dec(__kmap_atomic_idx);
6762 + current->kmap_idx--;
6763 +# ifdef CONFIG_DEBUG_HIGHMEM
6764 + BUG_ON(current->kmap_idx < 0);
6769 diff --git a/include/linux/hrtimer.h b/include/linux/hrtimer.h
6770 index 5e00f80b1535..a34e10b55cde 100644
6771 --- a/include/linux/hrtimer.h
6772 +++ b/include/linux/hrtimer.h
6773 @@ -87,6 +87,9 @@ enum hrtimer_restart {
6774 * @function: timer expiry callback function
6775 * @base: pointer to the timer base (per cpu and per clock)
6776 * @state: state information (See bit values above)
6777 + * @cb_entry: list entry to defer timers from hardirq context
6778 + * @irqsafe: timer can run in hardirq context
6779 + * @praecox: timer expiry time if expired at the time of programming
6780 * @is_rel: Set if the timer was armed relative
6781 * @start_pid: timer statistics field to store the pid of the task which
6783 @@ -103,6 +106,11 @@ struct hrtimer {
6784 enum hrtimer_restart (*function)(struct hrtimer *);
6785 struct hrtimer_clock_base *base;
6787 + struct list_head cb_entry;
6789 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
6793 #ifdef CONFIG_TIMER_STATS
6795 @@ -123,11 +131,7 @@ struct hrtimer_sleeper {
6796 struct task_struct *task;
6799 -#ifdef CONFIG_64BIT
6800 # define HRTIMER_CLOCK_BASE_ALIGN 64
6802 -# define HRTIMER_CLOCK_BASE_ALIGN 32
6806 * struct hrtimer_clock_base - the timer base for a specific clock
6807 @@ -136,6 +140,7 @@ struct hrtimer_sleeper {
6808 * timer to a base on another cpu.
6809 * @clockid: clock id for per_cpu support
6810 * @active: red black tree root node for the active timers
6811 + * @expired: list head for deferred timers.
6812 * @get_time: function to retrieve the current time of the clock
6813 * @offset: offset of this clock to the monotonic base
6815 @@ -144,6 +149,7 @@ struct hrtimer_clock_base {
6818 struct timerqueue_head active;
6819 + struct list_head expired;
6820 ktime_t (*get_time)(void);
6822 } __attribute__((__aligned__(HRTIMER_CLOCK_BASE_ALIGN)));
6823 @@ -187,6 +193,7 @@ struct hrtimer_cpu_base {
6824 raw_spinlock_t lock;
6826 struct hrtimer *running;
6827 + struct hrtimer *running_soft;
6829 unsigned int active_bases;
6830 unsigned int clock_was_set_seq;
6831 @@ -203,6 +210,9 @@ struct hrtimer_cpu_base {
6832 unsigned int nr_hangs;
6833 unsigned int max_hang_time;
6835 +#ifdef CONFIG_PREEMPT_RT_BASE
6836 + wait_queue_head_t wait;
6838 struct hrtimer_clock_base clock_base[HRTIMER_MAX_CLOCK_BASES];
6839 } ____cacheline_aligned;
6841 @@ -412,6 +422,13 @@ static inline void hrtimer_restart(struct hrtimer *timer)
6842 hrtimer_start_expires(timer, HRTIMER_MODE_ABS);
6845 +/* Softirq preemption could deadlock timer removal */
6846 +#ifdef CONFIG_PREEMPT_RT_BASE
6847 + extern void hrtimer_wait_for_timer(const struct hrtimer *timer);
6849 +# define hrtimer_wait_for_timer(timer) do { cpu_relax(); } while (0)
6853 extern ktime_t __hrtimer_get_remaining(const struct hrtimer *timer, bool adjust);
6855 @@ -436,9 +453,15 @@ static inline int hrtimer_is_queued(struct hrtimer *timer)
6856 * Helper function to check, whether the timer is running the callback
6859 -static inline int hrtimer_callback_running(struct hrtimer *timer)
6860 +static inline int hrtimer_callback_running(const struct hrtimer *timer)
6862 - return timer->base->cpu_base->running == timer;
6863 + if (timer->base->cpu_base->running == timer)
6865 +#ifdef CONFIG_PREEMPT_RT_BASE
6866 + if (timer->base->cpu_base->running_soft == timer)
6872 /* Forward a hrtimer so it expires after now: */
6873 diff --git a/include/linux/idr.h b/include/linux/idr.h
6874 index 083d61e92706..5899796f50cb 100644
6875 --- a/include/linux/idr.h
6876 +++ b/include/linux/idr.h
6877 @@ -95,10 +95,14 @@ bool idr_is_empty(struct idr *idp);
6878 * Each idr_preload() should be matched with an invocation of this
6879 * function. See idr_preload() for details.
6881 +#ifdef CONFIG_PREEMPT_RT_FULL
6882 +void idr_preload_end(void);
6884 static inline void idr_preload_end(void)
6891 * idr_find - return pointer for given id
6892 diff --git a/include/linux/init_task.h b/include/linux/init_task.h
6893 index 325f649d77ff..8af70bcc799b 100644
6894 --- a/include/linux/init_task.h
6895 +++ b/include/linux/init_task.h
6896 @@ -150,6 +150,12 @@ extern struct task_group root_task_group;
6897 # define INIT_PERF_EVENTS(tsk)
6900 +#ifdef CONFIG_PREEMPT_RT_BASE
6901 +# define INIT_TIMER_LIST .posix_timer_list = NULL,
6903 +# define INIT_TIMER_LIST
6906 #ifdef CONFIG_VIRT_CPU_ACCOUNTING_GEN
6907 # define INIT_VTIME(tsk) \
6908 .vtime_seqcount = SEQCNT_ZERO(tsk.vtime_seqcount), \
6909 @@ -250,6 +256,7 @@ extern struct task_group root_task_group;
6910 .cpu_timers = INIT_CPU_TIMERS(tsk.cpu_timers), \
6911 .pi_lock = __RAW_SPIN_LOCK_UNLOCKED(tsk.pi_lock), \
6912 .timer_slack_ns = 50000, /* 50 usec default slack */ \
6915 [PIDTYPE_PID] = INIT_PID_LINK(PIDTYPE_PID), \
6916 [PIDTYPE_PGID] = INIT_PID_LINK(PIDTYPE_PGID), \
6917 diff --git a/include/linux/interrupt.h b/include/linux/interrupt.h
6918 index 72f0721f75e7..480972ae47d3 100644
6919 --- a/include/linux/interrupt.h
6920 +++ b/include/linux/interrupt.h
6922 #include <linux/hrtimer.h>
6923 #include <linux/kref.h>
6924 #include <linux/workqueue.h>
6925 +#include <linux/swork.h>
6927 #include <linux/atomic.h>
6928 #include <asm/ptrace.h>
6930 * interrupt handler after suspending interrupts. For system
6931 * wakeup devices users need to implement wakeup detection in
6932 * their interrupt handlers.
6933 + * IRQF_NO_SOFTIRQ_CALL - Do not process softirqs in the irq thread context (RT)
6935 #define IRQF_SHARED 0x00000080
6936 #define IRQF_PROBE_SHARED 0x00000100
6938 #define IRQF_NO_THREAD 0x00010000
6939 #define IRQF_EARLY_RESUME 0x00020000
6940 #define IRQF_COND_SUSPEND 0x00040000
6941 +#define IRQF_NO_SOFTIRQ_CALL 0x00080000
6943 #define IRQF_TIMER (__IRQF_TIMER | IRQF_NO_SUSPEND | IRQF_NO_THREAD)
6945 @@ -196,7 +199,7 @@ extern void devm_free_irq(struct device *dev, unsigned int irq, void *dev_id);
6946 #ifdef CONFIG_LOCKDEP
6947 # define local_irq_enable_in_hardirq() do { } while (0)
6949 -# define local_irq_enable_in_hardirq() local_irq_enable()
6950 +# define local_irq_enable_in_hardirq() local_irq_enable_nort()
6953 extern void disable_irq_nosync(unsigned int irq);
6954 @@ -216,6 +219,7 @@ extern void resume_device_irqs(void);
6955 * struct irq_affinity_notify - context for notification of IRQ affinity changes
6956 * @irq: Interrupt to which notification applies
6957 * @kref: Reference count, for internal use
6958 + * @swork: Swork item, for internal use
6959 * @work: Work item, for internal use
6960 * @notify: Function to be called on change. This will be
6961 * called in process context.
6962 @@ -227,7 +231,11 @@ extern void resume_device_irqs(void);
6963 struct irq_affinity_notify {
6966 +#ifdef CONFIG_PREEMPT_RT_BASE
6967 + struct swork_event swork;
6969 struct work_struct work;
6971 void (*notify)(struct irq_affinity_notify *, const cpumask_t *mask);
6972 void (*release)(struct kref *ref);
6974 @@ -406,9 +414,13 @@ extern int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which,
6977 #ifdef CONFIG_IRQ_FORCED_THREADING
6978 +# ifndef CONFIG_PREEMPT_RT_BASE
6979 extern bool force_irqthreads;
6981 +# define force_irqthreads (true)
6984 -#define force_irqthreads (0)
6985 +#define force_irqthreads (false)
6988 #ifndef __ARCH_SET_SOFTIRQ_PENDING
6989 @@ -465,9 +477,10 @@ struct softirq_action
6990 void (*action)(struct softirq_action *);
6993 +#ifndef CONFIG_PREEMPT_RT_FULL
6994 asmlinkage void do_softirq(void);
6995 asmlinkage void __do_softirq(void);
6997 +static inline void thread_do_softirq(void) { do_softirq(); }
6998 #ifdef __ARCH_HAS_DO_SOFTIRQ
6999 void do_softirq_own_stack(void);
7001 @@ -476,13 +489,25 @@ static inline void do_softirq_own_stack(void)
7006 +extern void thread_do_softirq(void);
7009 extern void open_softirq(int nr, void (*action)(struct softirq_action *));
7010 extern void softirq_init(void);
7011 extern void __raise_softirq_irqoff(unsigned int nr);
7012 +#ifdef CONFIG_PREEMPT_RT_FULL
7013 +extern void __raise_softirq_irqoff_ksoft(unsigned int nr);
7015 +static inline void __raise_softirq_irqoff_ksoft(unsigned int nr)
7017 + __raise_softirq_irqoff(nr);
7021 extern void raise_softirq_irqoff(unsigned int nr);
7022 extern void raise_softirq(unsigned int nr);
7023 +extern void softirq_check_pending_idle(void);
7025 DECLARE_PER_CPU(struct task_struct *, ksoftirqd);
7027 @@ -504,8 +529,9 @@ static inline struct task_struct *this_cpu_ksoftirqd(void)
7028 to be executed on some cpu at least once after this.
7029 * If the tasklet is already scheduled, but its execution is still not
7030 started, it will be executed only once.
7031 - * If this tasklet is already running on another CPU (or schedule is called
7032 - from tasklet itself), it is rescheduled for later.
7033 + * If this tasklet is already running on another CPU, it is rescheduled
7035 + * Schedule must not be called from the tasklet itself (a lockup occurs)
7036 * Tasklet is strictly serialized wrt itself, but not
7037 wrt another tasklets. If client needs some intertask synchronization,
7038 he makes it with spinlocks.
7039 @@ -530,27 +556,36 @@ struct tasklet_struct name = { NULL, 0, ATOMIC_INIT(1), func, data }
7042 TASKLET_STATE_SCHED, /* Tasklet is scheduled for execution */
7043 - TASKLET_STATE_RUN /* Tasklet is running (SMP only) */
7044 + TASKLET_STATE_RUN, /* Tasklet is running (SMP only) */
7045 + TASKLET_STATE_PENDING /* Tasklet is pending */
7049 +#define TASKLET_STATEF_SCHED (1 << TASKLET_STATE_SCHED)
7050 +#define TASKLET_STATEF_RUN (1 << TASKLET_STATE_RUN)
7051 +#define TASKLET_STATEF_PENDING (1 << TASKLET_STATE_PENDING)
7053 +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT_FULL)
7054 static inline int tasklet_trylock(struct tasklet_struct *t)
7056 return !test_and_set_bit(TASKLET_STATE_RUN, &(t)->state);
7059 +static inline int tasklet_tryunlock(struct tasklet_struct *t)
7061 + return cmpxchg(&t->state, TASKLET_STATEF_RUN, 0) == TASKLET_STATEF_RUN;
7064 static inline void tasklet_unlock(struct tasklet_struct *t)
7066 smp_mb__before_atomic();
7067 clear_bit(TASKLET_STATE_RUN, &(t)->state);
7070 -static inline void tasklet_unlock_wait(struct tasklet_struct *t)
7072 - while (test_bit(TASKLET_STATE_RUN, &(t)->state)) { barrier(); }
7074 +extern void tasklet_unlock_wait(struct tasklet_struct *t);
7077 #define tasklet_trylock(t) 1
7078 +#define tasklet_tryunlock(t) 1
7079 #define tasklet_unlock_wait(t) do { } while (0)
7080 #define tasklet_unlock(t) do { } while (0)
7082 @@ -599,12 +634,7 @@ static inline void tasklet_disable(struct tasklet_struct *t)
7086 -static inline void tasklet_enable(struct tasklet_struct *t)
7088 - smp_mb__before_atomic();
7089 - atomic_dec(&t->count);
7092 +extern void tasklet_enable(struct tasklet_struct *t);
7093 extern void tasklet_kill(struct tasklet_struct *t);
7094 extern void tasklet_kill_immediate(struct tasklet_struct *t, unsigned int cpu);
7095 extern void tasklet_init(struct tasklet_struct *t,
7096 @@ -635,6 +665,12 @@ void tasklet_hrtimer_cancel(struct tasklet_hrtimer *ttimer)
7097 tasklet_kill(&ttimer->tasklet);
7100 +#ifdef CONFIG_PREEMPT_RT_FULL
7101 +extern void softirq_early_init(void);
7103 +static inline void softirq_early_init(void) { }
7107 * Autoprobing for irqs:
7109 diff --git a/include/linux/irq.h b/include/linux/irq.h
7110 index 39e3254e5769..8ebac94fbb9f 100644
7111 --- a/include/linux/irq.h
7112 +++ b/include/linux/irq.h
7113 @@ -72,6 +72,7 @@ enum irqchip_irq_state;
7114 * IRQ_IS_POLLED - Always polled by another interrupt. Exclude
7115 * it from the spurious interrupt detection
7116 * mechanism and from core side polling.
7117 + * IRQ_NO_SOFTIRQ_CALL - No softirq processing in the irq thread context (RT)
7118 * IRQ_DISABLE_UNLAZY - Disable lazy irq disable
7121 @@ -99,13 +100,14 @@ enum {
7122 IRQ_PER_CPU_DEVID = (1 << 17),
7123 IRQ_IS_POLLED = (1 << 18),
7124 IRQ_DISABLE_UNLAZY = (1 << 19),
7125 + IRQ_NO_SOFTIRQ_CALL = (1 << 20),
7128 #define IRQF_MODIFY_MASK \
7129 (IRQ_TYPE_SENSE_MASK | IRQ_NOPROBE | IRQ_NOREQUEST | \
7130 IRQ_NOAUTOEN | IRQ_MOVE_PCNTXT | IRQ_LEVEL | IRQ_NO_BALANCING | \
7131 IRQ_PER_CPU | IRQ_NESTED_THREAD | IRQ_NOTHREAD | IRQ_PER_CPU_DEVID | \
7132 - IRQ_IS_POLLED | IRQ_DISABLE_UNLAZY)
7133 + IRQ_IS_POLLED | IRQ_DISABLE_UNLAZY | IRQ_NO_SOFTIRQ_CALL)
7135 #define IRQ_NO_BALANCING_MASK (IRQ_PER_CPU | IRQ_NO_BALANCING)
7137 diff --git a/include/linux/irq_work.h b/include/linux/irq_work.h
7138 index 47b9ebd4a74f..2543aab05daa 100644
7139 --- a/include/linux/irq_work.h
7140 +++ b/include/linux/irq_work.h
7142 #define IRQ_WORK_BUSY 2UL
7143 #define IRQ_WORK_FLAGS 3UL
7144 #define IRQ_WORK_LAZY 4UL /* Doesn't want IPI, wait for tick */
7145 +#define IRQ_WORK_HARD_IRQ 8UL /* Run hard IRQ context, even on RT */
7148 unsigned long flags;
7149 @@ -51,4 +52,10 @@ static inline bool irq_work_needs_cpu(void) { return false; }
7150 static inline void irq_work_run(void) { }
7153 +#if defined(CONFIG_IRQ_WORK) && defined(CONFIG_PREEMPT_RT_FULL)
7154 +void irq_work_tick_soft(void);
7156 +static inline void irq_work_tick_soft(void) { }
7159 #endif /* _LINUX_IRQ_WORK_H */
7160 diff --git a/include/linux/irqdesc.h b/include/linux/irqdesc.h
7161 index c9be57931b58..eeeb540971ae 100644
7162 --- a/include/linux/irqdesc.h
7163 +++ b/include/linux/irqdesc.h
7164 @@ -66,6 +66,7 @@ struct irq_desc {
7165 unsigned int irqs_unhandled;
7166 atomic_t threads_handled;
7167 int threads_handled_last;
7169 raw_spinlock_t lock;
7170 struct cpumask *percpu_enabled;
7171 const struct cpumask *percpu_affinity;
7172 diff --git a/include/linux/irqflags.h b/include/linux/irqflags.h
7173 index 5dd1272d1ab2..9b77034f7c5e 100644
7174 --- a/include/linux/irqflags.h
7175 +++ b/include/linux/irqflags.h
7177 # define trace_softirqs_enabled(p) ((p)->softirqs_enabled)
7178 # define trace_hardirq_enter() do { current->hardirq_context++; } while (0)
7179 # define trace_hardirq_exit() do { current->hardirq_context--; } while (0)
7180 -# define lockdep_softirq_enter() do { current->softirq_context++; } while (0)
7181 -# define lockdep_softirq_exit() do { current->softirq_context--; } while (0)
7182 # define INIT_TRACE_IRQFLAGS .softirqs_enabled = 1,
7184 # define trace_hardirqs_on() do { } while (0)
7186 # define trace_softirqs_enabled(p) 0
7187 # define trace_hardirq_enter() do { } while (0)
7188 # define trace_hardirq_exit() do { } while (0)
7189 +# define INIT_TRACE_IRQFLAGS
7192 +#if defined(CONFIG_TRACE_IRQFLAGS) && !defined(CONFIG_PREEMPT_RT_FULL)
7193 +# define lockdep_softirq_enter() do { current->softirq_context++; } while (0)
7194 +# define lockdep_softirq_exit() do { current->softirq_context--; } while (0)
7196 # define lockdep_softirq_enter() do { } while (0)
7197 # define lockdep_softirq_exit() do { } while (0)
7198 -# define INIT_TRACE_IRQFLAGS
7201 #if defined(CONFIG_IRQSOFF_TRACER) || \
7202 @@ -148,4 +152,23 @@
7204 #define irqs_disabled_flags(flags) raw_irqs_disabled_flags(flags)
7207 + * local_irq* variants depending on RT/!RT
7209 +#ifdef CONFIG_PREEMPT_RT_FULL
7210 +# define local_irq_disable_nort() do { } while (0)
7211 +# define local_irq_enable_nort() do { } while (0)
7212 +# define local_irq_save_nort(flags) local_save_flags(flags)
7213 +# define local_irq_restore_nort(flags) (void)(flags)
7214 +# define local_irq_disable_rt() local_irq_disable()
7215 +# define local_irq_enable_rt() local_irq_enable()
7217 +# define local_irq_disable_nort() local_irq_disable()
7218 +# define local_irq_enable_nort() local_irq_enable()
7219 +# define local_irq_save_nort(flags) local_irq_save(flags)
7220 +# define local_irq_restore_nort(flags) local_irq_restore(flags)
7221 +# define local_irq_disable_rt() do { } while (0)
7222 +# define local_irq_enable_rt() do { } while (0)
7226 diff --git a/include/linux/jbd2.h b/include/linux/jbd2.h
7227 index dfaa1f4dcb0c..d57dd06544a1 100644
7228 --- a/include/linux/jbd2.h
7229 +++ b/include/linux/jbd2.h
7230 @@ -347,32 +347,56 @@ static inline struct journal_head *bh2jh(struct buffer_head *bh)
7232 static inline void jbd_lock_bh_state(struct buffer_head *bh)
7234 +#ifndef CONFIG_PREEMPT_RT_BASE
7235 bit_spin_lock(BH_State, &bh->b_state);
7237 + spin_lock(&bh->b_state_lock);
7241 static inline int jbd_trylock_bh_state(struct buffer_head *bh)
7243 +#ifndef CONFIG_PREEMPT_RT_BASE
7244 return bit_spin_trylock(BH_State, &bh->b_state);
7246 + return spin_trylock(&bh->b_state_lock);
7250 static inline int jbd_is_locked_bh_state(struct buffer_head *bh)
7252 +#ifndef CONFIG_PREEMPT_RT_BASE
7253 return bit_spin_is_locked(BH_State, &bh->b_state);
7255 + return spin_is_locked(&bh->b_state_lock);
7259 static inline void jbd_unlock_bh_state(struct buffer_head *bh)
7261 +#ifndef CONFIG_PREEMPT_RT_BASE
7262 bit_spin_unlock(BH_State, &bh->b_state);
7264 + spin_unlock(&bh->b_state_lock);
7268 static inline void jbd_lock_bh_journal_head(struct buffer_head *bh)
7270 +#ifndef CONFIG_PREEMPT_RT_BASE
7271 bit_spin_lock(BH_JournalHead, &bh->b_state);
7273 + spin_lock(&bh->b_journal_head_lock);
7277 static inline void jbd_unlock_bh_journal_head(struct buffer_head *bh)
7279 +#ifndef CONFIG_PREEMPT_RT_BASE
7280 bit_spin_unlock(BH_JournalHead, &bh->b_state);
7282 + spin_unlock(&bh->b_journal_head_lock);
7286 #define J_ASSERT(assert) BUG_ON(!(assert))
7287 diff --git a/include/linux/kdb.h b/include/linux/kdb.h
7288 index 410decacff8f..0861bebfc188 100644
7289 --- a/include/linux/kdb.h
7290 +++ b/include/linux/kdb.h
7291 @@ -167,6 +167,7 @@ extern __printf(2, 0) int vkdb_printf(enum kdb_msgsrc src, const char *fmt,
7292 extern __printf(1, 2) int kdb_printf(const char *, ...);
7293 typedef __printf(1, 2) int (*kdb_printf_t)(const char *, ...);
7295 +#define in_kdb_printk() (kdb_trap_printk)
7296 extern void kdb_init(int level);
7298 /* Access to kdb specific polling devices */
7299 @@ -201,6 +202,7 @@ extern int kdb_register_flags(char *, kdb_func_t, char *, char *,
7300 extern int kdb_unregister(char *);
7301 #else /* ! CONFIG_KGDB_KDB */
7302 static inline __printf(1, 2) int kdb_printf(const char *fmt, ...) { return 0; }
7303 +#define in_kdb_printk() (0)
7304 static inline void kdb_init(int level) {}
7305 static inline int kdb_register(char *cmd, kdb_func_t func, char *usage,
7306 char *help, short minlen) { return 0; }
7307 diff --git a/include/linux/kernel.h b/include/linux/kernel.h
7308 index bc6ed52a39b9..7894d55e4998 100644
7309 --- a/include/linux/kernel.h
7310 +++ b/include/linux/kernel.h
7311 @@ -194,6 +194,9 @@ extern int _cond_resched(void);
7313 # define might_sleep() \
7314 do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0)
7316 +# define might_sleep_no_state_check() \
7317 + do { ___might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0)
7318 # define sched_annotate_sleep() (current->task_state_change = 0)
7320 static inline void ___might_sleep(const char *file, int line,
7321 @@ -201,6 +204,7 @@ extern int _cond_resched(void);
7322 static inline void __might_sleep(const char *file, int line,
7323 int preempt_offset) { }
7324 # define might_sleep() do { might_resched(); } while (0)
7325 +# define might_sleep_no_state_check() do { might_resched(); } while (0)
7326 # define sched_annotate_sleep() do { } while (0)
7329 @@ -488,6 +492,7 @@ extern enum system_states {
7336 #define TAINT_PROPRIETARY_MODULE 0
7337 diff --git a/include/linux/list_bl.h b/include/linux/list_bl.h
7338 index cb483305e1f5..4e5062316bb6 100644
7339 --- a/include/linux/list_bl.h
7340 +++ b/include/linux/list_bl.h
7342 #define _LINUX_LIST_BL_H
7344 #include <linux/list.h>
7345 +#include <linux/spinlock.h>
7346 #include <linux/bit_spinlock.h>
7351 struct hlist_bl_head {
7352 struct hlist_bl_node *first;
7353 +#ifdef CONFIG_PREEMPT_RT_BASE
7354 + raw_spinlock_t lock;
7358 struct hlist_bl_node {
7359 struct hlist_bl_node *next, **pprev;
7361 -#define INIT_HLIST_BL_HEAD(ptr) \
7362 - ((ptr)->first = NULL)
7364 +#ifdef CONFIG_PREEMPT_RT_BASE
7365 +#define INIT_HLIST_BL_HEAD(h) \
7367 + (h)->first = NULL; \
7368 + raw_spin_lock_init(&(h)->lock); \
7371 +#define INIT_HLIST_BL_HEAD(h) (h)->first = NULL
7374 static inline void INIT_HLIST_BL_NODE(struct hlist_bl_node *h)
7376 @@ -118,12 +130,26 @@ static inline void hlist_bl_del_init(struct hlist_bl_node *n)
7378 static inline void hlist_bl_lock(struct hlist_bl_head *b)
7380 +#ifndef CONFIG_PREEMPT_RT_BASE
7381 bit_spin_lock(0, (unsigned long *)b);
7383 + raw_spin_lock(&b->lock);
7384 +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
7385 + __set_bit(0, (unsigned long *)b);
7390 static inline void hlist_bl_unlock(struct hlist_bl_head *b)
7392 +#ifndef CONFIG_PREEMPT_RT_BASE
7393 __bit_spin_unlock(0, (unsigned long *)b);
7395 +#if defined(CONFIG_SMP) || defined(CONFIG_DEBUG_SPINLOCK)
7396 + __clear_bit(0, (unsigned long *)b);
7398 + raw_spin_unlock(&b->lock);
7402 static inline bool hlist_bl_is_locked(struct hlist_bl_head *b)
7403 diff --git a/include/linux/locallock.h b/include/linux/locallock.h
7404 new file mode 100644
7405 index 000000000000..845c77f1a5ca
7407 +++ b/include/linux/locallock.h
7409 +#ifndef _LINUX_LOCALLOCK_H
7410 +#define _LINUX_LOCALLOCK_H
7412 +#include <linux/percpu.h>
7413 +#include <linux/spinlock.h>
7415 +#ifdef CONFIG_PREEMPT_RT_BASE
7417 +#ifdef CONFIG_DEBUG_SPINLOCK
7418 +# define LL_WARN(cond) WARN_ON(cond)
7420 +# define LL_WARN(cond) do { } while (0)
7424 + * per cpu lock based substitute for local_irq_*()
7426 +struct local_irq_lock {
7428 + struct task_struct *owner;
7430 + unsigned long flags;
7433 +#define DEFINE_LOCAL_IRQ_LOCK(lvar) \
7434 + DEFINE_PER_CPU(struct local_irq_lock, lvar) = { \
7435 + .lock = __SPIN_LOCK_UNLOCKED((lvar).lock) }
7437 +#define DECLARE_LOCAL_IRQ_LOCK(lvar) \
7438 + DECLARE_PER_CPU(struct local_irq_lock, lvar)
7440 +#define local_irq_lock_init(lvar) \
7443 + for_each_possible_cpu(__cpu) \
7444 + spin_lock_init(&per_cpu(lvar, __cpu).lock); \
7448 + * spin_lock|trylock|unlock_local flavour that does not migrate disable
7449 + * used for __local_lock|trylock|unlock where get_local_var/put_local_var
7450 + * already takes care of the migrate_disable/enable
7451 + * for CONFIG_PREEMPT_BASE map to the normal spin_* calls.
7453 +#ifdef CONFIG_PREEMPT_RT_FULL
7454 +# define spin_lock_local(lock) rt_spin_lock__no_mg(lock)
7455 +# define spin_trylock_local(lock) rt_spin_trylock__no_mg(lock)
7456 +# define spin_unlock_local(lock) rt_spin_unlock__no_mg(lock)
7458 +# define spin_lock_local(lock) spin_lock(lock)
7459 +# define spin_trylock_local(lock) spin_trylock(lock)
7460 +# define spin_unlock_local(lock) spin_unlock(lock)
7463 +static inline void __local_lock(struct local_irq_lock *lv)
7465 + if (lv->owner != current) {
7466 + spin_lock_local(&lv->lock);
7467 + LL_WARN(lv->owner);
7468 + LL_WARN(lv->nestcnt);
7469 + lv->owner = current;
7474 +#define local_lock(lvar) \
7475 + do { __local_lock(&get_local_var(lvar)); } while (0)
7477 +#define local_lock_on(lvar, cpu) \
7478 + do { __local_lock(&per_cpu(lvar, cpu)); } while (0)
7480 +static inline int __local_trylock(struct local_irq_lock *lv)
7482 + if (lv->owner != current && spin_trylock_local(&lv->lock)) {
7483 + LL_WARN(lv->owner);
7484 + LL_WARN(lv->nestcnt);
7485 + lv->owner = current;
7492 +#define local_trylock(lvar) \
7495 + __locked = __local_trylock(&get_local_var(lvar)); \
7497 + put_local_var(lvar); \
7501 +static inline void __local_unlock(struct local_irq_lock *lv)
7503 + LL_WARN(lv->nestcnt == 0);
7504 + LL_WARN(lv->owner != current);
7505 + if (--lv->nestcnt)
7509 + spin_unlock_local(&lv->lock);
7512 +#define local_unlock(lvar) \
7514 + __local_unlock(this_cpu_ptr(&lvar)); \
7515 + put_local_var(lvar); \
7518 +#define local_unlock_on(lvar, cpu) \
7519 + do { __local_unlock(&per_cpu(lvar, cpu)); } while (0)
7521 +static inline void __local_lock_irq(struct local_irq_lock *lv)
7523 + spin_lock_irqsave(&lv->lock, lv->flags);
7524 + LL_WARN(lv->owner);
7525 + LL_WARN(lv->nestcnt);
7526 + lv->owner = current;
7530 +#define local_lock_irq(lvar) \
7531 + do { __local_lock_irq(&get_local_var(lvar)); } while (0)
7533 +#define local_lock_irq_on(lvar, cpu) \
7534 + do { __local_lock_irq(&per_cpu(lvar, cpu)); } while (0)
7536 +static inline void __local_unlock_irq(struct local_irq_lock *lv)
7538 + LL_WARN(!lv->nestcnt);
7539 + LL_WARN(lv->owner != current);
7542 + spin_unlock_irq(&lv->lock);
7545 +#define local_unlock_irq(lvar) \
7547 + __local_unlock_irq(this_cpu_ptr(&lvar)); \
7548 + put_local_var(lvar); \
7551 +#define local_unlock_irq_on(lvar, cpu) \
7553 + __local_unlock_irq(&per_cpu(lvar, cpu)); \
7556 +static inline int __local_lock_irqsave(struct local_irq_lock *lv)
7558 + if (lv->owner != current) {
7559 + __local_lock_irq(lv);
7567 +#define local_lock_irqsave(lvar, _flags) \
7569 + if (__local_lock_irqsave(&get_local_var(lvar))) \
7570 + put_local_var(lvar); \
7571 + _flags = __this_cpu_read(lvar.flags); \
7574 +#define local_lock_irqsave_on(lvar, _flags, cpu) \
7576 + __local_lock_irqsave(&per_cpu(lvar, cpu)); \
7577 + _flags = per_cpu(lvar, cpu).flags; \
7580 +static inline int __local_unlock_irqrestore(struct local_irq_lock *lv,
7581 + unsigned long flags)
7583 + LL_WARN(!lv->nestcnt);
7584 + LL_WARN(lv->owner != current);
7585 + if (--lv->nestcnt)
7589 + spin_unlock_irqrestore(&lv->lock, lv->flags);
7593 +#define local_unlock_irqrestore(lvar, flags) \
7595 + if (__local_unlock_irqrestore(this_cpu_ptr(&lvar), flags)) \
7596 + put_local_var(lvar); \
7599 +#define local_unlock_irqrestore_on(lvar, flags, cpu) \
7601 + __local_unlock_irqrestore(&per_cpu(lvar, cpu), flags); \
7604 +#define local_spin_trylock_irq(lvar, lock) \
7607 + local_lock_irq(lvar); \
7608 + __locked = spin_trylock(lock); \
7610 + local_unlock_irq(lvar); \
7614 +#define local_spin_lock_irq(lvar, lock) \
7616 + local_lock_irq(lvar); \
7617 + spin_lock(lock); \
7620 +#define local_spin_unlock_irq(lvar, lock) \
7622 + spin_unlock(lock); \
7623 + local_unlock_irq(lvar); \
7626 +#define local_spin_lock_irqsave(lvar, lock, flags) \
7628 + local_lock_irqsave(lvar, flags); \
7629 + spin_lock(lock); \
7632 +#define local_spin_unlock_irqrestore(lvar, lock, flags) \
7634 + spin_unlock(lock); \
7635 + local_unlock_irqrestore(lvar, flags); \
7638 +#define get_locked_var(lvar, var) \
7640 + local_lock(lvar); \
7641 + this_cpu_ptr(&var); \
7644 +#define put_locked_var(lvar, var) local_unlock(lvar);
7646 +#define local_lock_cpu(lvar) \
7648 + local_lock(lvar); \
7649 + smp_processor_id(); \
7652 +#define local_unlock_cpu(lvar) local_unlock(lvar)
7654 +#else /* PREEMPT_RT_BASE */
7656 +#define DEFINE_LOCAL_IRQ_LOCK(lvar) __typeof__(const int) lvar
7657 +#define DECLARE_LOCAL_IRQ_LOCK(lvar) extern __typeof__(const int) lvar
7659 +static inline void local_irq_lock_init(int lvar) { }
7661 +#define local_lock(lvar) preempt_disable()
7662 +#define local_unlock(lvar) preempt_enable()
7663 +#define local_lock_irq(lvar) local_irq_disable()
7664 +#define local_lock_irq_on(lvar, cpu) local_irq_disable()
7665 +#define local_unlock_irq(lvar) local_irq_enable()
7666 +#define local_unlock_irq_on(lvar, cpu) local_irq_enable()
7667 +#define local_lock_irqsave(lvar, flags) local_irq_save(flags)
7668 +#define local_unlock_irqrestore(lvar, flags) local_irq_restore(flags)
7670 +#define local_spin_trylock_irq(lvar, lock) spin_trylock_irq(lock)
7671 +#define local_spin_lock_irq(lvar, lock) spin_lock_irq(lock)
7672 +#define local_spin_unlock_irq(lvar, lock) spin_unlock_irq(lock)
7673 +#define local_spin_lock_irqsave(lvar, lock, flags) \
7674 + spin_lock_irqsave(lock, flags)
7675 +#define local_spin_unlock_irqrestore(lvar, lock, flags) \
7676 + spin_unlock_irqrestore(lock, flags)
7678 +#define get_locked_var(lvar, var) get_cpu_var(var)
7679 +#define put_locked_var(lvar, var) put_cpu_var(var)
7681 +#define local_lock_cpu(lvar) get_cpu()
7682 +#define local_unlock_cpu(lvar) put_cpu()
7687 diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
7688 index 08d947fc4c59..705fb564a605 100644
7689 --- a/include/linux/mm_types.h
7690 +++ b/include/linux/mm_types.h
7692 #include <linux/completion.h>
7693 #include <linux/cpumask.h>
7694 #include <linux/uprobes.h>
7695 +#include <linux/rcupdate.h>
7696 #include <linux/page-flags-layout.h>
7697 #include <linux/workqueue.h>
7698 #include <asm/page.h>
7699 @@ -509,6 +510,9 @@ struct mm_struct {
7700 bool tlb_flush_pending;
7702 struct uprobes_state uprobes_state;
7703 +#ifdef CONFIG_PREEMPT_RT_BASE
7704 + struct rcu_head delayed_drop;
7706 #ifdef CONFIG_X86_INTEL_MPX
7707 /* address of the bounds directory */
7708 void __user *bd_addr;
7709 diff --git a/include/linux/module.h b/include/linux/module.h
7710 index 0c3207d26ac0..5944baaa3f28 100644
7711 --- a/include/linux/module.h
7712 +++ b/include/linux/module.h
7713 @@ -496,6 +496,7 @@ static inline int module_is_live(struct module *mod)
7714 struct module *__module_text_address(unsigned long addr);
7715 struct module *__module_address(unsigned long addr);
7716 bool is_module_address(unsigned long addr);
7717 +bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr);
7718 bool is_module_percpu_address(unsigned long addr);
7719 bool is_module_text_address(unsigned long addr);
7721 @@ -663,6 +664,11 @@ static inline bool is_module_percpu_address(unsigned long addr)
7725 +static inline bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr)
7730 static inline bool is_module_text_address(unsigned long addr)
7733 diff --git a/include/linux/mutex.h b/include/linux/mutex.h
7734 index 2cb7531e7d7a..b3fdfc820216 100644
7735 --- a/include/linux/mutex.h
7736 +++ b/include/linux/mutex.h
7738 #include <asm/processor.h>
7739 #include <linux/osq_lock.h>
7741 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
7742 +# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \
7743 + , .dep_map = { .name = #lockname }
7745 +# define __DEP_MAP_MUTEX_INITIALIZER(lockname)
7748 +#ifdef CONFIG_PREEMPT_RT_FULL
7749 +# include <linux/mutex_rt.h>
7753 * Simple, straightforward mutexes with strict semantics:
7755 @@ -99,13 +110,6 @@ do { \
7756 static inline void mutex_destroy(struct mutex *lock) {}
7759 -#ifdef CONFIG_DEBUG_LOCK_ALLOC
7760 -# define __DEP_MAP_MUTEX_INITIALIZER(lockname) \
7761 - , .dep_map = { .name = #lockname }
7763 -# define __DEP_MAP_MUTEX_INITIALIZER(lockname)
7766 #define __MUTEX_INITIALIZER(lockname) \
7767 { .count = ATOMIC_INIT(1) \
7768 , .wait_lock = __SPIN_LOCK_UNLOCKED(lockname.wait_lock) \
7769 @@ -173,6 +177,8 @@ extern int __must_check mutex_lock_killable(struct mutex *lock);
7770 extern int mutex_trylock(struct mutex *lock);
7771 extern void mutex_unlock(struct mutex *lock);
7773 +#endif /* !PREEMPT_RT_FULL */
7775 extern int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock);
7777 #endif /* __LINUX_MUTEX_H */
7778 diff --git a/include/linux/mutex_rt.h b/include/linux/mutex_rt.h
7779 new file mode 100644
7780 index 000000000000..e0284edec655
7782 +++ b/include/linux/mutex_rt.h
7784 +#ifndef __LINUX_MUTEX_RT_H
7785 +#define __LINUX_MUTEX_RT_H
7787 +#ifndef __LINUX_MUTEX_H
7788 +#error "Please include mutex.h"
7791 +#include <linux/rtmutex.h>
7793 +/* FIXME: Just for __lockfunc */
7794 +#include <linux/spinlock.h>
7797 + struct rt_mutex lock;
7798 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
7799 + struct lockdep_map dep_map;
7803 +#define __MUTEX_INITIALIZER(mutexname) \
7805 + .lock = __RT_MUTEX_INITIALIZER(mutexname.lock) \
7806 + __DEP_MAP_MUTEX_INITIALIZER(mutexname) \
7809 +#define DEFINE_MUTEX(mutexname) \
7810 + struct mutex mutexname = __MUTEX_INITIALIZER(mutexname)
7812 +extern void __mutex_do_init(struct mutex *lock, const char *name, struct lock_class_key *key);
7813 +extern void __lockfunc _mutex_lock(struct mutex *lock);
7814 +extern int __lockfunc _mutex_lock_interruptible(struct mutex *lock);
7815 +extern int __lockfunc _mutex_lock_killable(struct mutex *lock);
7816 +extern void __lockfunc _mutex_lock_nested(struct mutex *lock, int subclass);
7817 +extern void __lockfunc _mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest_lock);
7818 +extern int __lockfunc _mutex_lock_interruptible_nested(struct mutex *lock, int subclass);
7819 +extern int __lockfunc _mutex_lock_killable_nested(struct mutex *lock, int subclass);
7820 +extern int __lockfunc _mutex_trylock(struct mutex *lock);
7821 +extern void __lockfunc _mutex_unlock(struct mutex *lock);
7823 +#define mutex_is_locked(l) rt_mutex_is_locked(&(l)->lock)
7824 +#define mutex_lock(l) _mutex_lock(l)
7825 +#define mutex_lock_interruptible(l) _mutex_lock_interruptible(l)
7826 +#define mutex_lock_killable(l) _mutex_lock_killable(l)
7827 +#define mutex_trylock(l) _mutex_trylock(l)
7828 +#define mutex_unlock(l) _mutex_unlock(l)
7830 +#ifdef CONFIG_DEBUG_MUTEXES
7831 +#define mutex_destroy(l) rt_mutex_destroy(&(l)->lock)
7833 +static inline void mutex_destroy(struct mutex *lock) {}
7836 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
7837 +# define mutex_lock_nested(l, s) _mutex_lock_nested(l, s)
7838 +# define mutex_lock_interruptible_nested(l, s) \
7839 + _mutex_lock_interruptible_nested(l, s)
7840 +# define mutex_lock_killable_nested(l, s) \
7841 + _mutex_lock_killable_nested(l, s)
7843 +# define mutex_lock_nest_lock(lock, nest_lock) \
7845 + typecheck(struct lockdep_map *, &(nest_lock)->dep_map); \
7846 + _mutex_lock_nest_lock(lock, &(nest_lock)->dep_map); \
7850 +# define mutex_lock_nested(l, s) _mutex_lock(l)
7851 +# define mutex_lock_interruptible_nested(l, s) \
7852 + _mutex_lock_interruptible(l)
7853 +# define mutex_lock_killable_nested(l, s) \
7854 + _mutex_lock_killable(l)
7855 +# define mutex_lock_nest_lock(lock, nest_lock) mutex_lock(lock)
7858 +# define mutex_init(mutex) \
7860 + static struct lock_class_key __key; \
7862 + rt_mutex_init(&(mutex)->lock); \
7863 + __mutex_do_init((mutex), #mutex, &__key); \
7866 +# define __mutex_init(mutex, name, key) \
7868 + rt_mutex_init(&(mutex)->lock); \
7869 + __mutex_do_init((mutex), name, key); \
7873 diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
7874 index bb9b102c15cd..a5b12b8ad196 100644
7875 --- a/include/linux/netdevice.h
7876 +++ b/include/linux/netdevice.h
7877 @@ -396,7 +396,19 @@ typedef enum rx_handler_result rx_handler_result_t;
7878 typedef rx_handler_result_t rx_handler_func_t(struct sk_buff **pskb);
7880 void __napi_schedule(struct napi_struct *n);
7883 + * When PREEMPT_RT_FULL is defined, all device interrupt handlers
7884 + * run as threads, and they can also be preempted (without PREEMPT_RT
7885 + * interrupt threads can not be preempted). Which means that calling
7886 + * __napi_schedule_irqoff() from an interrupt handler can be preempted
7887 + * and can corrupt the napi->poll_list.
7889 +#ifdef CONFIG_PREEMPT_RT_FULL
7890 +#define __napi_schedule_irqoff(n) __napi_schedule(n)
7892 void __napi_schedule_irqoff(struct napi_struct *n);
7895 static inline bool napi_disable_pending(struct napi_struct *n)
7897 @@ -2463,14 +2475,53 @@ void netdev_freemem(struct net_device *dev);
7898 void synchronize_net(void);
7899 int init_dummy_netdev(struct net_device *dev);
7901 -DECLARE_PER_CPU(int, xmit_recursion);
7902 #define XMIT_RECURSION_LIMIT 10
7903 +#ifdef CONFIG_PREEMPT_RT_FULL
7904 +static inline int dev_recursion_level(void)
7906 + return current->xmit_recursion;
7909 +static inline int xmit_rec_read(void)
7911 + return current->xmit_recursion;
7914 +static inline void xmit_rec_inc(void)
7916 + current->xmit_recursion++;
7919 +static inline void xmit_rec_dec(void)
7921 + current->xmit_recursion--;
7926 +DECLARE_PER_CPU(int, xmit_recursion);
7928 static inline int dev_recursion_level(void)
7930 return this_cpu_read(xmit_recursion);
7933 +static inline int xmit_rec_read(void)
7935 + return __this_cpu_read(xmit_recursion);
7938 +static inline void xmit_rec_inc(void)
7940 + __this_cpu_inc(xmit_recursion);
7943 +static inline void xmit_rec_dec(void)
7945 + __this_cpu_dec(xmit_recursion);
7949 struct net_device *dev_get_by_index(struct net *net, int ifindex);
7950 struct net_device *__dev_get_by_index(struct net *net, int ifindex);
7951 struct net_device *dev_get_by_index_rcu(struct net *net, int ifindex);
7952 @@ -2855,6 +2906,7 @@ struct softnet_data {
7953 unsigned int dropped;
7954 struct sk_buff_head input_pkt_queue;
7955 struct napi_struct backlog;
7956 + struct sk_buff_head tofree_queue;
7960 diff --git a/include/linux/netfilter/x_tables.h b/include/linux/netfilter/x_tables.h
7961 index 2ad1a2b289b5..b4d10155af54 100644
7962 --- a/include/linux/netfilter/x_tables.h
7963 +++ b/include/linux/netfilter/x_tables.h
7966 #include <linux/netdevice.h>
7967 #include <linux/static_key.h>
7968 +#include <linux/locallock.h>
7969 #include <uapi/linux/netfilter/x_tables.h>
7971 /* Test a struct->invflags and a boolean for inequality */
7972 @@ -300,6 +301,8 @@ void xt_free_table_info(struct xt_table_info *info);
7974 DECLARE_PER_CPU(seqcount_t, xt_recseq);
7976 +DECLARE_LOCAL_IRQ_LOCK(xt_write_lock);
7978 /* xt_tee_enabled - true if x_tables needs to handle reentrancy
7980 * Enabled if current ip(6)tables ruleset has at least one -j TEE rule.
7981 @@ -320,6 +323,9 @@ static inline unsigned int xt_write_recseq_begin(void)
7983 unsigned int addend;
7985 + /* RT protection */
7986 + local_lock(xt_write_lock);
7989 * Low order bit of sequence is set if we already
7990 * called xt_write_recseq_begin().
7991 @@ -350,6 +356,7 @@ static inline void xt_write_recseq_end(unsigned int addend)
7992 /* this is kind of a write_seqcount_end(), but addend is 0 or 1 */
7994 __this_cpu_add(xt_recseq.sequence, addend);
7995 + local_unlock(xt_write_lock);
7999 diff --git a/include/linux/nfs_fs.h b/include/linux/nfs_fs.h
8000 index 810124b33327..d54ca43d571f 100644
8001 --- a/include/linux/nfs_fs.h
8002 +++ b/include/linux/nfs_fs.h
8003 @@ -165,7 +165,11 @@ struct nfs_inode {
8005 /* Readers: in-flight sillydelete RPC calls */
8006 /* Writers: rmdir */
8007 +#ifdef CONFIG_PREEMPT_RT_BASE
8008 + struct semaphore rmdir_sem;
8010 struct rw_semaphore rmdir_sem;
8013 #if IS_ENABLED(CONFIG_NFS_V4)
8014 struct nfs4_cached_acl *nfs4_acl;
8015 diff --git a/include/linux/nfs_xdr.h b/include/linux/nfs_xdr.h
8016 index beb1e10f446e..ebaf2e7bfe29 100644
8017 --- a/include/linux/nfs_xdr.h
8018 +++ b/include/linux/nfs_xdr.h
8019 @@ -1490,7 +1490,7 @@ struct nfs_unlinkdata {
8020 struct nfs_removeargs args;
8021 struct nfs_removeres res;
8022 struct dentry *dentry;
8023 - wait_queue_head_t wq;
8024 + struct swait_queue_head wq;
8025 struct rpc_cred *cred;
8026 struct nfs_fattr dir_attr;
8028 diff --git a/include/linux/notifier.h b/include/linux/notifier.h
8029 index 4149868de4e6..babe5b9bcb91 100644
8030 --- a/include/linux/notifier.h
8031 +++ b/include/linux/notifier.h
8034 * Alan Cox <Alan.Cox@linux.org>
8038 #ifndef _LINUX_NOTIFIER_H
8039 #define _LINUX_NOTIFIER_H
8040 #include <linux/errno.h>
8042 * in srcu_notifier_call_chain(): no cache bounces and no memory barriers.
8043 * As compensation, srcu_notifier_chain_unregister() is rather expensive.
8044 * SRCU notifier chains should be used when the chain will be called very
8045 - * often but notifier_blocks will seldom be removed. Also, SRCU notifier
8046 - * chains are slightly more difficult to use because they require special
8047 - * runtime initialization.
8048 + * often but notifier_blocks will seldom be removed.
8051 struct notifier_block;
8052 @@ -90,7 +88,7 @@ struct srcu_notifier_head {
8053 (name)->head = NULL; \
8056 -/* srcu_notifier_heads must be initialized and cleaned up dynamically */
8057 +/* srcu_notifier_heads must be cleaned up dynamically */
8058 extern void srcu_init_notifier_head(struct srcu_notifier_head *nh);
8059 #define srcu_cleanup_notifier_head(name) \
8060 cleanup_srcu_struct(&(name)->srcu);
8061 @@ -103,7 +101,13 @@ extern void srcu_init_notifier_head(struct srcu_notifier_head *nh);
8063 #define RAW_NOTIFIER_INIT(name) { \
8065 -/* srcu_notifier_heads cannot be initialized statically */
8067 +#define SRCU_NOTIFIER_INIT(name, pcpu) \
8069 + .mutex = __MUTEX_INITIALIZER(name.mutex), \
8071 + .srcu = __SRCU_STRUCT_INIT(name.srcu, pcpu), \
8074 #define ATOMIC_NOTIFIER_HEAD(name) \
8075 struct atomic_notifier_head name = \
8076 @@ -115,6 +119,18 @@ extern void srcu_init_notifier_head(struct srcu_notifier_head *nh);
8077 struct raw_notifier_head name = \
8078 RAW_NOTIFIER_INIT(name)
8080 +#define _SRCU_NOTIFIER_HEAD(name, mod) \
8081 + static DEFINE_PER_CPU(struct srcu_struct_array, \
8082 + name##_head_srcu_array); \
8083 + mod struct srcu_notifier_head name = \
8084 + SRCU_NOTIFIER_INIT(name, name##_head_srcu_array)
8086 +#define SRCU_NOTIFIER_HEAD(name) \
8087 + _SRCU_NOTIFIER_HEAD(name, )
8089 +#define SRCU_NOTIFIER_HEAD_STATIC(name) \
8090 + _SRCU_NOTIFIER_HEAD(name, static)
8094 extern int atomic_notifier_chain_register(struct atomic_notifier_head *nh,
8095 @@ -184,12 +200,12 @@ static inline int notifier_to_errno(int ret)
8098 * Declared notifiers so far. I can imagine quite a few more chains
8099 - * over time (eg laptop power reset chains, reboot chain (to clean
8100 + * over time (eg laptop power reset chains, reboot chain (to clean
8101 * device units up), device [un]mount chain, module load/unload chain,
8102 - * low memory chain, screenblank chain (for plug in modular screenblankers)
8103 + * low memory chain, screenblank chain (for plug in modular screenblankers)
8104 * VC switch chains (for loadable kernel svgalib VC switch helpers) etc...
8108 /* CPU notfiers are defined in include/linux/cpu.h. */
8110 /* netdevice notifiers are defined in include/linux/netdevice.h */
8111 diff --git a/include/linux/percpu-rwsem.h b/include/linux/percpu-rwsem.h
8112 index 5b2e6159b744..ea940f451606 100644
8113 --- a/include/linux/percpu-rwsem.h
8114 +++ b/include/linux/percpu-rwsem.h
8116 #include <linux/atomic.h>
8117 #include <linux/rwsem.h>
8118 #include <linux/percpu.h>
8119 -#include <linux/wait.h>
8120 +#include <linux/swait.h>
8121 #include <linux/rcu_sync.h>
8122 #include <linux/lockdep.h>
8124 @@ -12,7 +12,7 @@ struct percpu_rw_semaphore {
8125 struct rcu_sync rss;
8126 unsigned int __percpu *read_count;
8127 struct rw_semaphore rw_sem;
8128 - wait_queue_head_t writer;
8129 + struct swait_queue_head writer;
8133 @@ -22,13 +22,13 @@ static struct percpu_rw_semaphore name = { \
8134 .rss = __RCU_SYNC_INITIALIZER(name.rss, RCU_SCHED_SYNC), \
8135 .read_count = &__percpu_rwsem_rc_##name, \
8136 .rw_sem = __RWSEM_INITIALIZER(name.rw_sem), \
8137 - .writer = __WAIT_QUEUE_HEAD_INITIALIZER(name.writer), \
8138 + .writer = __SWAIT_QUEUE_HEAD_INITIALIZER(name.writer), \
8141 extern int __percpu_down_read(struct percpu_rw_semaphore *, int);
8142 extern void __percpu_up_read(struct percpu_rw_semaphore *);
8144 -static inline void percpu_down_read_preempt_disable(struct percpu_rw_semaphore *sem)
8145 +static inline void percpu_down_read(struct percpu_rw_semaphore *sem)
8149 @@ -46,16 +46,10 @@ static inline void percpu_down_read_preempt_disable(struct percpu_rw_semaphore *
8150 __this_cpu_inc(*sem->read_count);
8151 if (unlikely(!rcu_sync_is_idle(&sem->rss)))
8152 __percpu_down_read(sem, false); /* Unconditional memory barrier */
8155 - * The barrier() prevents the compiler from
8156 + * The preempt_enable() prevents the compiler from
8157 * bleeding the critical section out.
8161 -static inline void percpu_down_read(struct percpu_rw_semaphore *sem)
8163 - percpu_down_read_preempt_disable(sem);
8167 @@ -82,13 +76,9 @@ static inline int percpu_down_read_trylock(struct percpu_rw_semaphore *sem)
8171 -static inline void percpu_up_read_preempt_enable(struct percpu_rw_semaphore *sem)
8172 +static inline void percpu_up_read(struct percpu_rw_semaphore *sem)
8175 - * The barrier() prevents the compiler from
8176 - * bleeding the critical section out.
8179 + preempt_disable();
8181 * Same as in percpu_down_read().
8183 @@ -101,12 +91,6 @@ static inline void percpu_up_read_preempt_enable(struct percpu_rw_semaphore *sem
8184 rwsem_release(&sem->rw_sem.dep_map, 1, _RET_IP_);
8187 -static inline void percpu_up_read(struct percpu_rw_semaphore *sem)
8189 - preempt_disable();
8190 - percpu_up_read_preempt_enable(sem);
8193 extern void percpu_down_write(struct percpu_rw_semaphore *);
8194 extern void percpu_up_write(struct percpu_rw_semaphore *);
8196 diff --git a/include/linux/percpu.h b/include/linux/percpu.h
8197 index 56939d3f6e53..b988bf40ad3e 100644
8198 --- a/include/linux/percpu.h
8199 +++ b/include/linux/percpu.h
8201 #define PERCPU_MODULE_RESERVE 0
8204 +#ifdef CONFIG_PREEMPT_RT_FULL
8206 +#define get_local_var(var) (*({ \
8207 + migrate_disable(); \
8208 + this_cpu_ptr(&var); }))
8210 +#define put_local_var(var) do { \
8212 + migrate_enable(); \
8215 +# define get_local_ptr(var) ({ \
8216 + migrate_disable(); \
8217 + this_cpu_ptr(var); })
8219 +# define put_local_ptr(var) do { \
8221 + migrate_enable(); \
8226 +#define get_local_var(var) get_cpu_var(var)
8227 +#define put_local_var(var) put_cpu_var(var)
8228 +#define get_local_ptr(var) get_cpu_ptr(var)
8229 +#define put_local_ptr(var) put_cpu_ptr(var)
8233 /* minimum unit size, also is the maximum supported allocation size */
8234 #define PCPU_MIN_UNIT_SIZE PFN_ALIGN(32 << 10)
8236 @@ -110,6 +139,7 @@ extern int __init pcpu_page_first_chunk(size_t reserved_size,
8239 extern void __percpu *__alloc_reserved_percpu(size_t size, size_t align);
8240 +extern bool __is_kernel_percpu_address(unsigned long addr, unsigned long *can_addr);
8241 extern bool is_kernel_percpu_address(unsigned long addr);
8243 #if !defined(CONFIG_SMP) || !defined(CONFIG_HAVE_SETUP_PER_CPU_AREA)
8244 diff --git a/include/linux/pid.h b/include/linux/pid.h
8245 index 23705a53abba..2cc64b779f03 100644
8246 --- a/include/linux/pid.h
8247 +++ b/include/linux/pid.h
8249 #define _LINUX_PID_H
8251 #include <linux/rcupdate.h>
8252 +#include <linux/atomic.h>
8256 diff --git a/include/linux/preempt.h b/include/linux/preempt.h
8257 index 75e4e30677f1..1cfb1cb72354 100644
8258 --- a/include/linux/preempt.h
8259 +++ b/include/linux/preempt.h
8261 #define HARDIRQ_OFFSET (1UL << HARDIRQ_SHIFT)
8262 #define NMI_OFFSET (1UL << NMI_SHIFT)
8264 -#define SOFTIRQ_DISABLE_OFFSET (2 * SOFTIRQ_OFFSET)
8265 +#ifndef CONFIG_PREEMPT_RT_FULL
8266 +# define SOFTIRQ_DISABLE_OFFSET (2 * SOFTIRQ_OFFSET)
8268 +# define SOFTIRQ_DISABLE_OFFSET (0)
8271 /* We use the MSB mostly because its available */
8272 #define PREEMPT_NEED_RESCHED 0x80000000
8274 #include <asm/preempt.h>
8276 #define hardirq_count() (preempt_count() & HARDIRQ_MASK)
8277 -#define softirq_count() (preempt_count() & SOFTIRQ_MASK)
8278 #define irq_count() (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_MASK \
8280 +#ifndef CONFIG_PREEMPT_RT_FULL
8281 +# define softirq_count() (preempt_count() & SOFTIRQ_MASK)
8282 +# define in_serving_softirq() (softirq_count() & SOFTIRQ_OFFSET)
8284 +# define softirq_count() (0UL)
8285 +extern int in_serving_softirq(void);
8289 * Are we doing bottom half or hardware interrupt processing?
8291 #define in_irq() (hardirq_count())
8292 #define in_softirq() (softirq_count())
8293 #define in_interrupt() (irq_count())
8294 -#define in_serving_softirq() (softirq_count() & SOFTIRQ_OFFSET)
8297 * Are we in NMI context?
8300 * The preempt_count offset after spin_lock()
8302 +#if !defined(CONFIG_PREEMPT_RT_FULL)
8303 #define PREEMPT_LOCK_OFFSET PREEMPT_DISABLE_OFFSET
8305 +#define PREEMPT_LOCK_OFFSET 0
8309 * The preempt_count offset needed for things like:
8310 @@ -140,6 +153,20 @@ extern void preempt_count_sub(int val);
8311 #define preempt_count_inc() preempt_count_add(1)
8312 #define preempt_count_dec() preempt_count_sub(1)
8314 +#ifdef CONFIG_PREEMPT_LAZY
8315 +#define add_preempt_lazy_count(val) do { preempt_lazy_count() += (val); } while (0)
8316 +#define sub_preempt_lazy_count(val) do { preempt_lazy_count() -= (val); } while (0)
8317 +#define inc_preempt_lazy_count() add_preempt_lazy_count(1)
8318 +#define dec_preempt_lazy_count() sub_preempt_lazy_count(1)
8319 +#define preempt_lazy_count() (current_thread_info()->preempt_lazy_count)
8321 +#define add_preempt_lazy_count(val) do { } while (0)
8322 +#define sub_preempt_lazy_count(val) do { } while (0)
8323 +#define inc_preempt_lazy_count() do { } while (0)
8324 +#define dec_preempt_lazy_count() do { } while (0)
8325 +#define preempt_lazy_count() (0)
8328 #ifdef CONFIG_PREEMPT_COUNT
8330 #define preempt_disable() \
8331 @@ -148,13 +175,25 @@ do { \
8335 +#define preempt_lazy_disable() \
8337 + inc_preempt_lazy_count(); \
8341 #define sched_preempt_enable_no_resched() \
8344 preempt_count_dec(); \
8347 -#define preempt_enable_no_resched() sched_preempt_enable_no_resched()
8348 +#ifdef CONFIG_PREEMPT_RT_BASE
8349 +# define preempt_enable_no_resched() sched_preempt_enable_no_resched()
8350 +# define preempt_check_resched_rt() preempt_check_resched()
8352 +# define preempt_enable_no_resched() preempt_enable()
8353 +# define preempt_check_resched_rt() barrier();
8356 #define preemptible() (preempt_count() == 0 && !irqs_disabled())
8358 @@ -179,6 +218,13 @@ do { \
8359 __preempt_schedule(); \
8362 +#define preempt_lazy_enable() \
8364 + dec_preempt_lazy_count(); \
8366 + preempt_check_resched(); \
8369 #else /* !CONFIG_PREEMPT */
8370 #define preempt_enable() \
8372 @@ -224,6 +270,7 @@ do { \
8373 #define preempt_disable_notrace() barrier()
8374 #define preempt_enable_no_resched_notrace() barrier()
8375 #define preempt_enable_notrace() barrier()
8376 +#define preempt_check_resched_rt() barrier()
8377 #define preemptible() 0
8379 #endif /* CONFIG_PREEMPT_COUNT */
8380 @@ -244,10 +291,31 @@ do { \
8382 #define preempt_fold_need_resched() \
8384 - if (tif_need_resched()) \
8385 + if (tif_need_resched_now()) \
8386 set_preempt_need_resched(); \
8389 +#ifdef CONFIG_PREEMPT_RT_FULL
8390 +# define preempt_disable_rt() preempt_disable()
8391 +# define preempt_enable_rt() preempt_enable()
8392 +# define preempt_disable_nort() barrier()
8393 +# define preempt_enable_nort() barrier()
8395 + extern void migrate_disable(void);
8396 + extern void migrate_enable(void);
8397 +# else /* CONFIG_SMP */
8398 +# define migrate_disable() barrier()
8399 +# define migrate_enable() barrier()
8400 +# endif /* CONFIG_SMP */
8402 +# define preempt_disable_rt() barrier()
8403 +# define preempt_enable_rt() barrier()
8404 +# define preempt_disable_nort() preempt_disable()
8405 +# define preempt_enable_nort() preempt_enable()
8406 +# define migrate_disable() preempt_disable()
8407 +# define migrate_enable() preempt_enable()
8410 #ifdef CONFIG_PREEMPT_NOTIFIERS
8412 struct preempt_notifier;
8413 diff --git a/include/linux/printk.h b/include/linux/printk.h
8414 index eac1af8502bb..37e647af0b0b 100644
8415 --- a/include/linux/printk.h
8416 +++ b/include/linux/printk.h
8417 @@ -126,9 +126,11 @@ struct va_format {
8418 #ifdef CONFIG_EARLY_PRINTK
8419 extern asmlinkage __printf(1, 2)
8420 void early_printk(const char *fmt, ...);
8421 +extern void printk_kill(void);
8423 static inline __printf(1, 2) __cold
8424 void early_printk(const char *s, ...) { }
8425 +static inline void printk_kill(void) { }
8428 #ifdef CONFIG_PRINTK_NMI
8429 diff --git a/include/linux/radix-tree.h b/include/linux/radix-tree.h
8430 index af3581b8a451..277295039c8f 100644
8431 --- a/include/linux/radix-tree.h
8432 +++ b/include/linux/radix-tree.h
8433 @@ -292,6 +292,8 @@ unsigned int radix_tree_gang_lookup_slot(struct radix_tree_root *root,
8434 int radix_tree_preload(gfp_t gfp_mask);
8435 int radix_tree_maybe_preload(gfp_t gfp_mask);
8436 int radix_tree_maybe_preload_order(gfp_t gfp_mask, int order);
8437 +void radix_tree_preload_end(void);
8439 void radix_tree_init(void);
8440 void *radix_tree_tag_set(struct radix_tree_root *root,
8441 unsigned long index, unsigned int tag);
8442 @@ -314,11 +316,6 @@ unsigned long radix_tree_range_tag_if_tagged(struct radix_tree_root *root,
8443 int radix_tree_tagged(struct radix_tree_root *root, unsigned int tag);
8444 unsigned long radix_tree_locate_item(struct radix_tree_root *root, void *item);
8446 -static inline void radix_tree_preload_end(void)
8452 * struct radix_tree_iter - radix tree iterator state
8454 diff --git a/include/linux/random.h b/include/linux/random.h
8455 index 7bd2403e4fef..b2df7148a42b 100644
8456 --- a/include/linux/random.h
8457 +++ b/include/linux/random.h
8458 @@ -31,7 +31,7 @@ static inline void add_latent_entropy(void) {}
8460 extern void add_input_randomness(unsigned int type, unsigned int code,
8461 unsigned int value) __latent_entropy;
8462 -extern void add_interrupt_randomness(int irq, int irq_flags) __latent_entropy;
8463 +extern void add_interrupt_randomness(int irq, int irq_flags, __u64 ip) __latent_entropy;
8465 extern void get_random_bytes(void *buf, int nbytes);
8466 extern int add_random_ready_callback(struct random_ready_callback *rdy);
8467 diff --git a/include/linux/rbtree.h b/include/linux/rbtree.h
8468 index e585018498d5..25c64474fc27 100644
8469 --- a/include/linux/rbtree.h
8470 +++ b/include/linux/rbtree.h
8473 #include <linux/kernel.h>
8474 #include <linux/stddef.h>
8475 -#include <linux/rcupdate.h>
8476 +#include <linux/rcu_assign_pointer.h>
8479 unsigned long __rb_parent_color;
8480 diff --git a/include/linux/rbtree_augmented.h b/include/linux/rbtree_augmented.h
8481 index d076183e49be..36bfb4dd57ae 100644
8482 --- a/include/linux/rbtree_augmented.h
8483 +++ b/include/linux/rbtree_augmented.h
8486 #include <linux/compiler.h>
8487 #include <linux/rbtree.h>
8488 +#include <linux/rcupdate.h>
8491 * Please note - only struct rb_augment_callbacks and the prototypes for
8492 diff --git a/include/linux/rcu_assign_pointer.h b/include/linux/rcu_assign_pointer.h
8493 new file mode 100644
8494 index 000000000000..7066962a4379
8496 +++ b/include/linux/rcu_assign_pointer.h
8498 +#ifndef __LINUX_RCU_ASSIGN_POINTER_H__
8499 +#define __LINUX_RCU_ASSIGN_POINTER_H__
8500 +#include <linux/compiler.h>
8501 +#include <asm/barrier.h>
8504 + * RCU_INITIALIZER() - statically initialize an RCU-protected global variable
8505 + * @v: The value to statically initialize with.
8507 +#define RCU_INITIALIZER(v) (typeof(*(v)) __force __rcu *)(v)
8510 + * rcu_assign_pointer() - assign to RCU-protected pointer
8511 + * @p: pointer to assign to
8512 + * @v: value to assign (publish)
8514 + * Assigns the specified value to the specified RCU-protected
8515 + * pointer, ensuring that any concurrent RCU readers will see
8516 + * any prior initialization.
8518 + * Inserts memory barriers on architectures that require them
8519 + * (which is most of them), and also prevents the compiler from
8520 + * reordering the code that initializes the structure after the pointer
8521 + * assignment. More importantly, this call documents which pointers
8522 + * will be dereferenced by RCU read-side code.
8524 + * In some special cases, you may use RCU_INIT_POINTER() instead
8525 + * of rcu_assign_pointer(). RCU_INIT_POINTER() is a bit faster due
8526 + * to the fact that it does not constrain either the CPU or the compiler.
8527 + * That said, using RCU_INIT_POINTER() when you should have used
8528 + * rcu_assign_pointer() is a very bad thing that results in
8529 + * impossible-to-diagnose memory corruption. So please be careful.
8530 + * See the RCU_INIT_POINTER() comment header for details.
8532 + * Note that rcu_assign_pointer() evaluates each of its arguments only
8533 + * once, appearances notwithstanding. One of the "extra" evaluations
8534 + * is in typeof() and the other visible only to sparse (__CHECKER__),
8535 + * neither of which actually execute the argument. As with most cpp
8536 + * macros, this execute-arguments-only-once property is important, so
8537 + * please be careful when making changes to rcu_assign_pointer() and the
8538 + * other macros that it invokes.
8540 +#define rcu_assign_pointer(p, v) \
8542 + uintptr_t _r_a_p__v = (uintptr_t)(v); \
8544 + if (__builtin_constant_p(v) && (_r_a_p__v) == (uintptr_t)NULL) \
8545 + WRITE_ONCE((p), (typeof(p))(_r_a_p__v)); \
8547 + smp_store_release(&p, RCU_INITIALIZER((typeof(p))_r_a_p__v)); \
8552 diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
8553 index 01f71e1d2e94..30cc001d0d5a 100644
8554 --- a/include/linux/rcupdate.h
8555 +++ b/include/linux/rcupdate.h
8557 #include <linux/compiler.h>
8558 #include <linux/ktime.h>
8559 #include <linux/irqflags.h>
8560 +#include <linux/rcu_assign_pointer.h>
8562 #include <asm/barrier.h>
8564 @@ -178,6 +179,9 @@ void call_rcu(struct rcu_head *head,
8566 #endif /* #else #ifdef CONFIG_PREEMPT_RCU */
8568 +#ifdef CONFIG_PREEMPT_RT_FULL
8569 +#define call_rcu_bh call_rcu
8572 * call_rcu_bh() - Queue an RCU for invocation after a quicker grace period.
8573 * @head: structure to be used for queueing the RCU updates.
8574 @@ -201,6 +205,7 @@ void call_rcu(struct rcu_head *head,
8576 void call_rcu_bh(struct rcu_head *head,
8577 rcu_callback_t func);
8581 * call_rcu_sched() - Queue an RCU for invocation after sched grace period.
8582 @@ -301,6 +306,11 @@ void synchronize_rcu(void);
8583 * types of kernel builds, the rcu_read_lock() nesting depth is unknowable.
8585 #define rcu_preempt_depth() (current->rcu_read_lock_nesting)
8586 +#ifndef CONFIG_PREEMPT_RT_FULL
8587 +#define sched_rcu_preempt_depth() rcu_preempt_depth()
8589 +static inline int sched_rcu_preempt_depth(void) { return 0; }
8592 #else /* #ifdef CONFIG_PREEMPT_RCU */
8594 @@ -326,6 +336,8 @@ static inline int rcu_preempt_depth(void)
8598 +#define sched_rcu_preempt_depth() rcu_preempt_depth()
8600 #endif /* #else #ifdef CONFIG_PREEMPT_RCU */
8602 /* Internal to kernel */
8603 @@ -505,7 +517,14 @@ extern struct lockdep_map rcu_callback_map;
8604 int debug_lockdep_rcu_enabled(void);
8606 int rcu_read_lock_held(void);
8607 +#ifdef CONFIG_PREEMPT_RT_FULL
8608 +static inline int rcu_read_lock_bh_held(void)
8610 + return rcu_read_lock_held();
8613 int rcu_read_lock_bh_held(void);
8617 * rcu_read_lock_sched_held() - might we be in RCU-sched read-side critical section?
8618 @@ -626,54 +645,6 @@ static inline void rcu_preempt_sleep_check(void)
8622 - * RCU_INITIALIZER() - statically initialize an RCU-protected global variable
8623 - * @v: The value to statically initialize with.
8625 -#define RCU_INITIALIZER(v) (typeof(*(v)) __force __rcu *)(v)
8628 - * rcu_assign_pointer() - assign to RCU-protected pointer
8629 - * @p: pointer to assign to
8630 - * @v: value to assign (publish)
8632 - * Assigns the specified value to the specified RCU-protected
8633 - * pointer, ensuring that any concurrent RCU readers will see
8634 - * any prior initialization.
8636 - * Inserts memory barriers on architectures that require them
8637 - * (which is most of them), and also prevents the compiler from
8638 - * reordering the code that initializes the structure after the pointer
8639 - * assignment. More importantly, this call documents which pointers
8640 - * will be dereferenced by RCU read-side code.
8642 - * In some special cases, you may use RCU_INIT_POINTER() instead
8643 - * of rcu_assign_pointer(). RCU_INIT_POINTER() is a bit faster due
8644 - * to the fact that it does not constrain either the CPU or the compiler.
8645 - * That said, using RCU_INIT_POINTER() when you should have used
8646 - * rcu_assign_pointer() is a very bad thing that results in
8647 - * impossible-to-diagnose memory corruption. So please be careful.
8648 - * See the RCU_INIT_POINTER() comment header for details.
8650 - * Note that rcu_assign_pointer() evaluates each of its arguments only
8651 - * once, appearances notwithstanding. One of the "extra" evaluations
8652 - * is in typeof() and the other visible only to sparse (__CHECKER__),
8653 - * neither of which actually execute the argument. As with most cpp
8654 - * macros, this execute-arguments-only-once property is important, so
8655 - * please be careful when making changes to rcu_assign_pointer() and the
8656 - * other macros that it invokes.
8658 -#define rcu_assign_pointer(p, v) \
8660 - uintptr_t _r_a_p__v = (uintptr_t)(v); \
8662 - if (__builtin_constant_p(v) && (_r_a_p__v) == (uintptr_t)NULL) \
8663 - WRITE_ONCE((p), (typeof(p))(_r_a_p__v)); \
8665 - smp_store_release(&p, RCU_INITIALIZER((typeof(p))_r_a_p__v)); \
8670 * rcu_access_pointer() - fetch RCU pointer with no dereferencing
8671 * @p: The pointer to read
8673 @@ -951,10 +922,14 @@ static inline void rcu_read_unlock(void)
8674 static inline void rcu_read_lock_bh(void)
8677 +#ifdef CONFIG_PREEMPT_RT_FULL
8681 rcu_lock_acquire(&rcu_bh_lock_map);
8682 RCU_LOCKDEP_WARN(!rcu_is_watching(),
8683 "rcu_read_lock_bh() used illegally while idle");
8688 @@ -964,10 +939,14 @@ static inline void rcu_read_lock_bh(void)
8690 static inline void rcu_read_unlock_bh(void)
8692 +#ifdef CONFIG_PREEMPT_RT_FULL
8693 + rcu_read_unlock();
8695 RCU_LOCKDEP_WARN(!rcu_is_watching(),
8696 "rcu_read_unlock_bh() used illegally while idle");
8697 rcu_lock_release(&rcu_bh_lock_map);
8703 diff --git a/include/linux/rcutree.h b/include/linux/rcutree.h
8704 index 63a4e4cf40a5..08ab12df2863 100644
8705 --- a/include/linux/rcutree.h
8706 +++ b/include/linux/rcutree.h
8707 @@ -44,7 +44,11 @@ static inline void rcu_virt_note_context_switch(int cpu)
8708 rcu_note_context_switch();
8711 +#ifdef CONFIG_PREEMPT_RT_FULL
8712 +# define synchronize_rcu_bh synchronize_rcu
8714 void synchronize_rcu_bh(void);
8716 void synchronize_sched_expedited(void);
8717 void synchronize_rcu_expedited(void);
8719 @@ -72,7 +76,11 @@ static inline void synchronize_rcu_bh_expedited(void)
8722 void rcu_barrier(void);
8723 +#ifdef CONFIG_PREEMPT_RT_FULL
8724 +# define rcu_barrier_bh rcu_barrier
8726 void rcu_barrier_bh(void);
8728 void rcu_barrier_sched(void);
8729 unsigned long get_state_synchronize_rcu(void);
8730 void cond_synchronize_rcu(unsigned long oldstate);
8731 @@ -82,17 +90,14 @@ void cond_synchronize_sched(unsigned long oldstate);
8732 extern unsigned long rcutorture_testseq;
8733 extern unsigned long rcutorture_vernum;
8734 unsigned long rcu_batches_started(void);
8735 -unsigned long rcu_batches_started_bh(void);
8736 unsigned long rcu_batches_started_sched(void);
8737 unsigned long rcu_batches_completed(void);
8738 -unsigned long rcu_batches_completed_bh(void);
8739 unsigned long rcu_batches_completed_sched(void);
8740 unsigned long rcu_exp_batches_completed(void);
8741 unsigned long rcu_exp_batches_completed_sched(void);
8742 void show_rcu_gp_kthreads(void);
8744 void rcu_force_quiescent_state(void);
8745 -void rcu_bh_force_quiescent_state(void);
8746 void rcu_sched_force_quiescent_state(void);
8748 void rcu_idle_enter(void);
8749 @@ -109,6 +114,16 @@ extern int rcu_scheduler_active __read_mostly;
8751 bool rcu_is_watching(void);
8753 +#ifndef CONFIG_PREEMPT_RT_FULL
8754 +void rcu_bh_force_quiescent_state(void);
8755 +unsigned long rcu_batches_started_bh(void);
8756 +unsigned long rcu_batches_completed_bh(void);
8758 +# define rcu_bh_force_quiescent_state rcu_force_quiescent_state
8759 +# define rcu_batches_completed_bh rcu_batches_completed
8760 +# define rcu_batches_started_bh rcu_batches_completed
8763 void rcu_all_qs(void);
8765 /* RCUtree hotplug events */
8766 diff --git a/include/linux/rtmutex.h b/include/linux/rtmutex.h
8767 index 1abba5ce2a2f..30211c627511 100644
8768 --- a/include/linux/rtmutex.h
8769 +++ b/include/linux/rtmutex.h
8771 #define __LINUX_RT_MUTEX_H
8773 #include <linux/linkage.h>
8774 +#include <linux/spinlock_types_raw.h>
8775 #include <linux/rbtree.h>
8776 -#include <linux/spinlock_types.h>
8778 extern int max_lock_depth; /* for sysctl */
8780 +#ifdef CONFIG_DEBUG_MUTEXES
8781 +#include <linux/debug_locks.h>
8785 * The rt_mutex structure
8787 @@ -31,8 +35,8 @@ struct rt_mutex {
8788 struct rb_root waiters;
8789 struct rb_node *waiters_leftmost;
8790 struct task_struct *owner;
8791 -#ifdef CONFIG_DEBUG_RT_MUTEXES
8793 +#ifdef CONFIG_DEBUG_RT_MUTEXES
8794 const char *name, *file;
8797 @@ -55,22 +59,33 @@ struct hrtimer_sleeper;
8798 # define rt_mutex_debug_check_no_locks_held(task) do { } while (0)
8801 +# define rt_mutex_init(mutex) \
8803 + raw_spin_lock_init(&(mutex)->wait_lock); \
8804 + __rt_mutex_init(mutex, #mutex); \
8807 #ifdef CONFIG_DEBUG_RT_MUTEXES
8808 # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname) \
8809 , .name = #mutexname, .file = __FILE__, .line = __LINE__
8810 -# define rt_mutex_init(mutex) __rt_mutex_init(mutex, __func__)
8811 extern void rt_mutex_debug_task_free(struct task_struct *tsk);
8813 # define __DEBUG_RT_MUTEX_INITIALIZER(mutexname)
8814 -# define rt_mutex_init(mutex) __rt_mutex_init(mutex, NULL)
8815 # define rt_mutex_debug_task_free(t) do { } while (0)
8818 -#define __RT_MUTEX_INITIALIZER(mutexname) \
8819 - { .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \
8820 +#define __RT_MUTEX_INITIALIZER_PLAIN(mutexname) \
8821 + .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(mutexname.wait_lock) \
8822 , .waiters = RB_ROOT \
8824 - __DEBUG_RT_MUTEX_INITIALIZER(mutexname)}
8825 + __DEBUG_RT_MUTEX_INITIALIZER(mutexname)
8827 +#define __RT_MUTEX_INITIALIZER(mutexname) \
8828 + { __RT_MUTEX_INITIALIZER_PLAIN(mutexname) }
8830 +#define __RT_MUTEX_INITIALIZER_SAVE_STATE(mutexname) \
8831 + { __RT_MUTEX_INITIALIZER_PLAIN(mutexname) \
8832 + , .save_state = 1 }
8834 #define DEFINE_RT_MUTEX(mutexname) \
8835 struct rt_mutex mutexname = __RT_MUTEX_INITIALIZER(mutexname)
8836 @@ -91,6 +106,7 @@ extern void rt_mutex_destroy(struct rt_mutex *lock);
8838 extern void rt_mutex_lock(struct rt_mutex *lock);
8839 extern int rt_mutex_lock_interruptible(struct rt_mutex *lock);
8840 +extern int rt_mutex_lock_killable(struct rt_mutex *lock);
8841 extern int rt_mutex_timed_lock(struct rt_mutex *lock,
8842 struct hrtimer_sleeper *timeout);
8844 diff --git a/include/linux/rwlock_rt.h b/include/linux/rwlock_rt.h
8845 new file mode 100644
8846 index 000000000000..49ed2d45d3be
8848 +++ b/include/linux/rwlock_rt.h
8850 +#ifndef __LINUX_RWLOCK_RT_H
8851 +#define __LINUX_RWLOCK_RT_H
8853 +#ifndef __LINUX_SPINLOCK_H
8854 +#error Do not include directly. Use spinlock.h
8857 +#define rwlock_init(rwl) \
8859 + static struct lock_class_key __key; \
8861 + rt_mutex_init(&(rwl)->lock); \
8862 + __rt_rwlock_init(rwl, #rwl, &__key); \
8865 +extern void __lockfunc rt_write_lock(rwlock_t *rwlock);
8866 +extern void __lockfunc rt_read_lock(rwlock_t *rwlock);
8867 +extern int __lockfunc rt_write_trylock(rwlock_t *rwlock);
8868 +extern int __lockfunc rt_write_trylock_irqsave(rwlock_t *trylock, unsigned long *flags);
8869 +extern int __lockfunc rt_read_trylock(rwlock_t *rwlock);
8870 +extern void __lockfunc rt_write_unlock(rwlock_t *rwlock);
8871 +extern void __lockfunc rt_read_unlock(rwlock_t *rwlock);
8872 +extern unsigned long __lockfunc rt_write_lock_irqsave(rwlock_t *rwlock);
8873 +extern unsigned long __lockfunc rt_read_lock_irqsave(rwlock_t *rwlock);
8874 +extern void __rt_rwlock_init(rwlock_t *rwlock, char *name, struct lock_class_key *key);
8876 +#define read_trylock(lock) __cond_lock(lock, rt_read_trylock(lock))
8877 +#define write_trylock(lock) __cond_lock(lock, rt_write_trylock(lock))
8879 +#define write_trylock_irqsave(lock, flags) \
8880 + __cond_lock(lock, rt_write_trylock_irqsave(lock, &flags))
8882 +#define read_lock_irqsave(lock, flags) \
8884 + typecheck(unsigned long, flags); \
8885 + flags = rt_read_lock_irqsave(lock); \
8888 +#define write_lock_irqsave(lock, flags) \
8890 + typecheck(unsigned long, flags); \
8891 + flags = rt_write_lock_irqsave(lock); \
8894 +#define read_lock(lock) rt_read_lock(lock)
8896 +#define read_lock_bh(lock) \
8898 + local_bh_disable(); \
8899 + rt_read_lock(lock); \
8902 +#define read_lock_irq(lock) read_lock(lock)
8904 +#define write_lock(lock) rt_write_lock(lock)
8906 +#define write_lock_bh(lock) \
8908 + local_bh_disable(); \
8909 + rt_write_lock(lock); \
8912 +#define write_lock_irq(lock) write_lock(lock)
8914 +#define read_unlock(lock) rt_read_unlock(lock)
8916 +#define read_unlock_bh(lock) \
8918 + rt_read_unlock(lock); \
8919 + local_bh_enable(); \
8922 +#define read_unlock_irq(lock) read_unlock(lock)
8924 +#define write_unlock(lock) rt_write_unlock(lock)
8926 +#define write_unlock_bh(lock) \
8928 + rt_write_unlock(lock); \
8929 + local_bh_enable(); \
8932 +#define write_unlock_irq(lock) write_unlock(lock)
8934 +#define read_unlock_irqrestore(lock, flags) \
8936 + typecheck(unsigned long, flags); \
8938 + rt_read_unlock(lock); \
8941 +#define write_unlock_irqrestore(lock, flags) \
8943 + typecheck(unsigned long, flags); \
8945 + rt_write_unlock(lock); \
8949 diff --git a/include/linux/rwlock_types.h b/include/linux/rwlock_types.h
8950 index cc0072e93e36..5317cd957292 100644
8951 --- a/include/linux/rwlock_types.h
8952 +++ b/include/linux/rwlock_types.h
8954 #ifndef __LINUX_RWLOCK_TYPES_H
8955 #define __LINUX_RWLOCK_TYPES_H
8957 +#if !defined(__LINUX_SPINLOCK_TYPES_H)
8958 +# error "Do not include directly, include spinlock_types.h"
8962 * include/linux/rwlock_types.h - generic rwlock type definitions
8964 diff --git a/include/linux/rwlock_types_rt.h b/include/linux/rwlock_types_rt.h
8965 new file mode 100644
8966 index 000000000000..51b28d775fe1
8968 +++ b/include/linux/rwlock_types_rt.h
8970 +#ifndef __LINUX_RWLOCK_TYPES_RT_H
8971 +#define __LINUX_RWLOCK_TYPES_RT_H
8973 +#ifndef __LINUX_SPINLOCK_TYPES_H
8974 +#error "Do not include directly. Include spinlock_types.h instead"
8978 + * rwlocks - rtmutex which allows single reader recursion
8981 + struct rt_mutex lock;
8983 + unsigned int break_lock;
8984 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
8985 + struct lockdep_map dep_map;
8989 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
8990 +# define RW_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname }
8992 +# define RW_DEP_MAP_INIT(lockname)
8995 +#define __RW_LOCK_UNLOCKED(name) \
8996 + { .lock = __RT_MUTEX_INITIALIZER_SAVE_STATE(name.lock), \
8997 + RW_DEP_MAP_INIT(name) }
8999 +#define DEFINE_RWLOCK(name) \
9000 + rwlock_t name = __RW_LOCK_UNLOCKED(name)
9003 diff --git a/include/linux/rwsem.h b/include/linux/rwsem.h
9004 index dd1d14250340..8e1f44ff1f2f 100644
9005 --- a/include/linux/rwsem.h
9006 +++ b/include/linux/rwsem.h
9008 #include <linux/osq_lock.h>
9011 +#ifdef CONFIG_PREEMPT_RT_FULL
9012 +#include <linux/rwsem_rt.h>
9013 +#else /* PREEMPT_RT_FULL */
9015 struct rw_semaphore;
9017 #ifdef CONFIG_RWSEM_GENERIC_SPINLOCK
9018 @@ -184,4 +188,6 @@ extern void up_read_non_owner(struct rw_semaphore *sem);
9019 # define up_read_non_owner(sem) up_read(sem)
9022 +#endif /* !PREEMPT_RT_FULL */
9024 #endif /* _LINUX_RWSEM_H */
9025 diff --git a/include/linux/rwsem_rt.h b/include/linux/rwsem_rt.h
9026 new file mode 100644
9027 index 000000000000..e26bd95a57c3
9029 +++ b/include/linux/rwsem_rt.h
9031 +#ifndef _LINUX_RWSEM_RT_H
9032 +#define _LINUX_RWSEM_RT_H
9034 +#ifndef _LINUX_RWSEM_H
9035 +#error "Include rwsem.h"
9039 + * RW-semaphores are a spinlock plus a reader-depth count.
9041 + * Note that the semantics are different from the usual
9042 + * Linux rw-sems, in PREEMPT_RT mode we do not allow
9043 + * multiple readers to hold the lock at once, we only allow
9044 + * a read-lock owner to read-lock recursively. This is
9045 + * better for latency, makes the implementation inherently
9046 + * fair and makes it simpler as well.
9049 +#include <linux/rtmutex.h>
9051 +struct rw_semaphore {
9052 + struct rt_mutex lock;
9054 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
9055 + struct lockdep_map dep_map;
9059 +#define __RWSEM_INITIALIZER(name) \
9060 + { .lock = __RT_MUTEX_INITIALIZER(name.lock), \
9061 + RW_DEP_MAP_INIT(name) }
9063 +#define DECLARE_RWSEM(lockname) \
9064 + struct rw_semaphore lockname = __RWSEM_INITIALIZER(lockname)
9066 +extern void __rt_rwsem_init(struct rw_semaphore *rwsem, const char *name,
9067 + struct lock_class_key *key);
9069 +#define __rt_init_rwsem(sem, name, key) \
9071 + rt_mutex_init(&(sem)->lock); \
9072 + __rt_rwsem_init((sem), (name), (key));\
9075 +#define __init_rwsem(sem, name, key) __rt_init_rwsem(sem, name, key)
9077 +# define rt_init_rwsem(sem) \
9079 + static struct lock_class_key __key; \
9081 + __rt_init_rwsem((sem), #sem, &__key); \
9084 +extern void rt_down_write(struct rw_semaphore *rwsem);
9085 +extern int rt_down_write_killable(struct rw_semaphore *rwsem);
9086 +extern void rt_down_read_nested(struct rw_semaphore *rwsem, int subclass);
9087 +extern void rt_down_write_nested(struct rw_semaphore *rwsem, int subclass);
9088 +extern int rt_down_write_killable_nested(struct rw_semaphore *rwsem,
9090 +extern void rt_down_write_nested_lock(struct rw_semaphore *rwsem,
9091 + struct lockdep_map *nest);
9092 +extern void rt__down_read(struct rw_semaphore *rwsem);
9093 +extern void rt_down_read(struct rw_semaphore *rwsem);
9094 +extern int rt_down_write_trylock(struct rw_semaphore *rwsem);
9095 +extern int rt__down_read_trylock(struct rw_semaphore *rwsem);
9096 +extern int rt_down_read_trylock(struct rw_semaphore *rwsem);
9097 +extern void __rt_up_read(struct rw_semaphore *rwsem);
9098 +extern void rt_up_read(struct rw_semaphore *rwsem);
9099 +extern void rt_up_write(struct rw_semaphore *rwsem);
9100 +extern void rt_downgrade_write(struct rw_semaphore *rwsem);
9102 +#define init_rwsem(sem) rt_init_rwsem(sem)
9103 +#define rwsem_is_locked(s) rt_mutex_is_locked(&(s)->lock)
9105 +static inline int rwsem_is_contended(struct rw_semaphore *sem)
9107 + /* rt_mutex_has_waiters() */
9108 + return !RB_EMPTY_ROOT(&sem->lock.waiters);
9111 +static inline void __down_read(struct rw_semaphore *sem)
9113 + rt__down_read(sem);
9116 +static inline void down_read(struct rw_semaphore *sem)
9118 + rt_down_read(sem);
9121 +static inline int __down_read_trylock(struct rw_semaphore *sem)
9123 + return rt__down_read_trylock(sem);
9126 +static inline int down_read_trylock(struct rw_semaphore *sem)
9128 + return rt_down_read_trylock(sem);
9131 +static inline void down_write(struct rw_semaphore *sem)
9133 + rt_down_write(sem);
9136 +static inline int down_write_killable(struct rw_semaphore *sem)
9138 + return rt_down_write_killable(sem);
9141 +static inline int down_write_trylock(struct rw_semaphore *sem)
9143 + return rt_down_write_trylock(sem);
9146 +static inline void __up_read(struct rw_semaphore *sem)
9148 + __rt_up_read(sem);
9151 +static inline void up_read(struct rw_semaphore *sem)
9156 +static inline void up_write(struct rw_semaphore *sem)
9161 +static inline void downgrade_write(struct rw_semaphore *sem)
9163 + rt_downgrade_write(sem);
9166 +static inline void down_read_nested(struct rw_semaphore *sem, int subclass)
9168 + return rt_down_read_nested(sem, subclass);
9171 +static inline void down_write_nested(struct rw_semaphore *sem, int subclass)
9173 + rt_down_write_nested(sem, subclass);
9176 +static inline int down_write_killable_nested(struct rw_semaphore *sem,
9179 + return rt_down_write_killable_nested(sem, subclass);
9182 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
9183 +static inline void down_write_nest_lock(struct rw_semaphore *sem,
9184 + struct rw_semaphore *nest_lock)
9186 + rt_down_write_nested_lock(sem, &nest_lock->dep_map);
9191 +static inline void down_write_nest_lock(struct rw_semaphore *sem,
9192 + struct rw_semaphore *nest_lock)
9194 + rt_down_write_nested_lock(sem, NULL);
9198 diff --git a/include/linux/sched.h b/include/linux/sched.h
9199 index 75d9a57e212e..8cb7df0f56e3 100644
9200 --- a/include/linux/sched.h
9201 +++ b/include/linux/sched.h
9202 @@ -26,6 +26,7 @@ struct sched_param {
9203 #include <linux/nodemask.h>
9204 #include <linux/mm_types.h>
9205 #include <linux/preempt.h>
9206 +#include <asm/kmap_types.h>
9208 #include <asm/page.h>
9209 #include <asm/ptrace.h>
9210 @@ -243,10 +244,7 @@ extern char ___assert_task_state[1 - 2*!!(
9211 TASK_UNINTERRUPTIBLE | __TASK_STOPPED | \
9212 __TASK_TRACED | EXIT_ZOMBIE | EXIT_DEAD)
9214 -#define task_is_traced(task) ((task->state & __TASK_TRACED) != 0)
9215 #define task_is_stopped(task) ((task->state & __TASK_STOPPED) != 0)
9216 -#define task_is_stopped_or_traced(task) \
9217 - ((task->state & (__TASK_STOPPED | __TASK_TRACED)) != 0)
9218 #define task_contributes_to_load(task) \
9219 ((task->state & TASK_UNINTERRUPTIBLE) != 0 && \
9220 (task->flags & PF_FROZEN) == 0 && \
9221 @@ -312,6 +310,11 @@ extern char ___assert_task_state[1 - 2*!!(
9225 +#define __set_current_state_no_track(state_value) \
9226 + do { current->state = (state_value); } while (0)
9227 +#define set_current_state_no_track(state_value) \
9228 + set_mb(current->state, (state_value))
9230 /* Task command name length */
9231 #define TASK_COMM_LEN 16
9233 @@ -1013,8 +1016,18 @@ struct wake_q_head {
9234 struct wake_q_head name = { WAKE_Q_TAIL, &name.first }
9236 extern void wake_q_add(struct wake_q_head *head,
9237 - struct task_struct *task);
9238 -extern void wake_up_q(struct wake_q_head *head);
9239 + struct task_struct *task);
9240 +extern void __wake_up_q(struct wake_q_head *head, bool sleeper);
9242 +static inline void wake_up_q(struct wake_q_head *head)
9244 + __wake_up_q(head, false);
9247 +static inline void wake_up_q_sleeper(struct wake_q_head *head)
9249 + __wake_up_q(head, true);
9253 * sched-domains (multiprocessor balancing) declarations:
9254 @@ -1481,6 +1494,7 @@ struct task_struct {
9255 struct thread_info thread_info;
9257 volatile long state; /* -1 unrunnable, 0 runnable, >0 stopped */
9258 + volatile long saved_state; /* saved state for "spinlock sleepers" */
9261 unsigned int flags; /* per process flags, defined below */
9262 @@ -1520,6 +1534,12 @@ struct task_struct {
9265 unsigned int policy;
9266 +#ifdef CONFIG_PREEMPT_RT_FULL
9267 + int migrate_disable;
9268 +# ifdef CONFIG_SCHED_DEBUG
9269 + int migrate_disable_atomic;
9272 int nr_cpus_allowed;
9273 cpumask_t cpus_allowed;
9275 @@ -1654,6 +1674,9 @@ struct task_struct {
9277 struct task_cputime cputime_expires;
9278 struct list_head cpu_timers[3];
9279 +#ifdef CONFIG_PREEMPT_RT_BASE
9280 + struct task_struct *posix_timer_list;
9283 /* process credentials */
9284 const struct cred __rcu *ptracer_cred; /* Tracer's credentials at attach */
9285 @@ -1685,10 +1708,15 @@ struct task_struct {
9286 /* signal handlers */
9287 struct signal_struct *signal;
9288 struct sighand_struct *sighand;
9289 + struct sigqueue *sigqueue_cache;
9291 sigset_t blocked, real_blocked;
9292 sigset_t saved_sigmask; /* restored if set_restore_sigmask() was used */
9293 struct sigpending pending;
9294 +#ifdef CONFIG_PREEMPT_RT_FULL
9295 + /* TODO: move me into ->restart_block ? */
9296 + struct siginfo forced_info;
9299 unsigned long sas_ss_sp;
9301 @@ -1917,6 +1945,12 @@ struct task_struct {
9302 /* bitmask and counter of trace recursion */
9303 unsigned long trace_recursion;
9304 #endif /* CONFIG_TRACING */
9305 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
9306 + u64 preempt_timestamp_hist;
9307 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
9308 + long timer_offset;
9312 /* Coverage collection mode enabled for this task (0 if disabled). */
9313 enum kcov_mode kcov_mode;
9314 @@ -1942,9 +1976,23 @@ struct task_struct {
9315 unsigned int sequential_io;
9316 unsigned int sequential_io_avg;
9318 +#ifdef CONFIG_PREEMPT_RT_BASE
9319 + struct rcu_head put_rcu;
9320 + int softirq_nestcnt;
9321 + unsigned int softirqs_raised;
9323 +#ifdef CONFIG_PREEMPT_RT_FULL
9324 +# if defined CONFIG_HIGHMEM || defined CONFIG_X86_32
9326 + pte_t kmap_pte[KM_TYPE_NR];
9329 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
9330 unsigned long task_state_change;
9332 +#ifdef CONFIG_PREEMPT_RT_FULL
9333 + int xmit_recursion;
9335 int pagefault_disabled;
9337 struct task_struct *oom_reaper_list;
9338 @@ -1984,14 +2032,6 @@ static inline struct vm_struct *task_stack_vm_area(const struct task_struct *t)
9342 -/* Future-safe accessor for struct task_struct's cpus_allowed. */
9343 -#define tsk_cpus_allowed(tsk) (&(tsk)->cpus_allowed)
9345 -static inline int tsk_nr_cpus_allowed(struct task_struct *p)
9347 - return p->nr_cpus_allowed;
9350 #define TNF_MIGRATED 0x01
9351 #define TNF_NO_GROUP 0x02
9352 #define TNF_SHARED 0x04
9353 @@ -2207,6 +2247,15 @@ extern struct pid *cad_pid;
9354 extern void free_task(struct task_struct *tsk);
9355 #define get_task_struct(tsk) do { atomic_inc(&(tsk)->usage); } while(0)
9357 +#ifdef CONFIG_PREEMPT_RT_BASE
9358 +extern void __put_task_struct_cb(struct rcu_head *rhp);
9360 +static inline void put_task_struct(struct task_struct *t)
9362 + if (atomic_dec_and_test(&t->usage))
9363 + call_rcu(&t->put_rcu, __put_task_struct_cb);
9366 extern void __put_task_struct(struct task_struct *t);
9368 static inline void put_task_struct(struct task_struct *t)
9369 @@ -2214,6 +2263,7 @@ static inline void put_task_struct(struct task_struct *t)
9370 if (atomic_dec_and_test(&t->usage))
9371 __put_task_struct(t);
9375 struct task_struct *task_rcu_dereference(struct task_struct **ptask);
9376 struct task_struct *try_get_task_struct(struct task_struct **ptask);
9377 @@ -2255,6 +2305,7 @@ extern void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut,
9381 +#define PF_IN_SOFTIRQ 0x00000001 /* Task is serving softirq */
9382 #define PF_EXITING 0x00000004 /* getting shut down */
9383 #define PF_EXITPIDONE 0x00000008 /* pi exit done on shut down */
9384 #define PF_VCPU 0x00000010 /* I'm a virtual CPU */
9385 @@ -2423,6 +2474,10 @@ extern void do_set_cpus_allowed(struct task_struct *p,
9387 extern int set_cpus_allowed_ptr(struct task_struct *p,
9388 const struct cpumask *new_mask);
9389 +int migrate_me(void);
9390 +void tell_sched_cpu_down_begin(int cpu);
9391 +void tell_sched_cpu_down_done(int cpu);
9394 static inline void do_set_cpus_allowed(struct task_struct *p,
9395 const struct cpumask *new_mask)
9396 @@ -2435,6 +2490,9 @@ static inline int set_cpus_allowed_ptr(struct task_struct *p,
9400 +static inline int migrate_me(void) { return 0; }
9401 +static inline void tell_sched_cpu_down_begin(int cpu) { }
9402 +static inline void tell_sched_cpu_down_done(int cpu) { }
9405 #ifdef CONFIG_NO_HZ_COMMON
9406 @@ -2673,6 +2731,7 @@ extern void xtime_update(unsigned long ticks);
9408 extern int wake_up_state(struct task_struct *tsk, unsigned int state);
9409 extern int wake_up_process(struct task_struct *tsk);
9410 +extern int wake_up_lock_sleeper(struct task_struct * tsk);
9411 extern void wake_up_new_task(struct task_struct *tsk);
9413 extern void kick_process(struct task_struct *tsk);
9414 @@ -2881,6 +2940,17 @@ static inline void mmdrop(struct mm_struct *mm)
9418 +#ifdef CONFIG_PREEMPT_RT_BASE
9419 +extern void __mmdrop_delayed(struct rcu_head *rhp);
9420 +static inline void mmdrop_delayed(struct mm_struct *mm)
9422 + if (atomic_dec_and_test(&mm->mm_count))
9423 + call_rcu(&mm->delayed_drop, __mmdrop_delayed);
9426 +# define mmdrop_delayed(mm) mmdrop(mm)
9429 static inline void mmdrop_async_fn(struct work_struct *work)
9431 struct mm_struct *mm = container_of(work, struct mm_struct, async_put_work);
9432 @@ -3273,6 +3343,43 @@ static inline int test_tsk_need_resched(struct task_struct *tsk)
9433 return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED));
9436 +#ifdef CONFIG_PREEMPT_LAZY
9437 +static inline void set_tsk_need_resched_lazy(struct task_struct *tsk)
9439 + set_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY);
9442 +static inline void clear_tsk_need_resched_lazy(struct task_struct *tsk)
9444 + clear_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY);
9447 +static inline int test_tsk_need_resched_lazy(struct task_struct *tsk)
9449 + return unlikely(test_tsk_thread_flag(tsk,TIF_NEED_RESCHED_LAZY));
9452 +static inline int need_resched_lazy(void)
9454 + return test_thread_flag(TIF_NEED_RESCHED_LAZY);
9457 +static inline int need_resched_now(void)
9459 + return test_thread_flag(TIF_NEED_RESCHED);
9463 +static inline void clear_tsk_need_resched_lazy(struct task_struct *tsk) { }
9464 +static inline int need_resched_lazy(void) { return 0; }
9466 +static inline int need_resched_now(void)
9468 + return test_thread_flag(TIF_NEED_RESCHED);
9473 static inline int restart_syscall(void)
9475 set_tsk_thread_flag(current, TIF_SIGPENDING);
9476 @@ -3304,6 +3411,51 @@ static inline int signal_pending_state(long state, struct task_struct *p)
9477 return (state & TASK_INTERRUPTIBLE) || __fatal_signal_pending(p);
9480 +static inline bool __task_is_stopped_or_traced(struct task_struct *task)
9482 + if (task->state & (__TASK_STOPPED | __TASK_TRACED))
9484 +#ifdef CONFIG_PREEMPT_RT_FULL
9485 + if (task->saved_state & (__TASK_STOPPED | __TASK_TRACED))
9491 +static inline bool task_is_stopped_or_traced(struct task_struct *task)
9493 + bool traced_stopped;
9495 +#ifdef CONFIG_PREEMPT_RT_FULL
9496 + unsigned long flags;
9498 + raw_spin_lock_irqsave(&task->pi_lock, flags);
9499 + traced_stopped = __task_is_stopped_or_traced(task);
9500 + raw_spin_unlock_irqrestore(&task->pi_lock, flags);
9502 + traced_stopped = __task_is_stopped_or_traced(task);
9504 + return traced_stopped;
9507 +static inline bool task_is_traced(struct task_struct *task)
9509 + bool traced = false;
9511 + if (task->state & __TASK_TRACED)
9513 +#ifdef CONFIG_PREEMPT_RT_FULL
9514 + /* in case the task is sleeping on tasklist_lock */
9515 + raw_spin_lock_irq(&task->pi_lock);
9516 + if (task->state & __TASK_TRACED)
9518 + else if (task->saved_state & __TASK_TRACED)
9520 + raw_spin_unlock_irq(&task->pi_lock);
9526 * cond_resched() and cond_resched_lock(): latency reduction via
9527 * explicit rescheduling in places that are safe. The return
9528 @@ -3329,12 +3481,16 @@ extern int __cond_resched_lock(spinlock_t *lock);
9529 __cond_resched_lock(lock); \
9532 +#ifndef CONFIG_PREEMPT_RT_FULL
9533 extern int __cond_resched_softirq(void);
9535 #define cond_resched_softirq() ({ \
9536 ___might_sleep(__FILE__, __LINE__, SOFTIRQ_DISABLE_OFFSET); \
9537 __cond_resched_softirq(); \
9540 +# define cond_resched_softirq() cond_resched()
9543 static inline void cond_resched_rcu(void)
9545 @@ -3509,6 +3665,31 @@ static inline void set_task_cpu(struct task_struct *p, unsigned int cpu)
9547 #endif /* CONFIG_SMP */
9549 +static inline int __migrate_disabled(struct task_struct *p)
9551 +#ifdef CONFIG_PREEMPT_RT_FULL
9552 + return p->migrate_disable;
9558 +/* Future-safe accessor for struct task_struct's cpus_allowed. */
9559 +static inline const struct cpumask *tsk_cpus_allowed(struct task_struct *p)
9561 + if (__migrate_disabled(p))
9562 + return cpumask_of(task_cpu(p));
9564 + return &p->cpus_allowed;
9567 +static inline int tsk_nr_cpus_allowed(struct task_struct *p)
9569 + if (__migrate_disabled(p))
9571 + return p->nr_cpus_allowed;
9574 extern long sched_setaffinity(pid_t pid, const struct cpumask *new_mask);
9575 extern long sched_getaffinity(pid_t pid, struct cpumask *mask);
9577 diff --git a/include/linux/seqlock.h b/include/linux/seqlock.h
9578 index ead97654c4e9..3d7223ffdd3b 100644
9579 --- a/include/linux/seqlock.h
9580 +++ b/include/linux/seqlock.h
9581 @@ -220,20 +220,30 @@ static inline int read_seqcount_retry(const seqcount_t *s, unsigned start)
9582 return __read_seqcount_retry(s, start);
9587 -static inline void raw_write_seqcount_begin(seqcount_t *s)
9588 +static inline void __raw_write_seqcount_begin(seqcount_t *s)
9594 -static inline void raw_write_seqcount_end(seqcount_t *s)
9595 +static inline void raw_write_seqcount_begin(seqcount_t *s)
9597 + preempt_disable_rt();
9598 + __raw_write_seqcount_begin(s);
9601 +static inline void __raw_write_seqcount_end(seqcount_t *s)
9607 +static inline void raw_write_seqcount_end(seqcount_t *s)
9609 + __raw_write_seqcount_end(s);
9610 + preempt_enable_rt();
9614 * raw_write_seqcount_barrier - do a seq write barrier
9615 * @s: pointer to seqcount_t
9616 @@ -428,10 +438,32 @@ typedef struct {
9618 * Read side functions for starting and finalizing a read side section.
9620 +#ifndef CONFIG_PREEMPT_RT_FULL
9621 static inline unsigned read_seqbegin(const seqlock_t *sl)
9623 return read_seqcount_begin(&sl->seqcount);
9627 + * Starvation safe read side for RT
9629 +static inline unsigned read_seqbegin(seqlock_t *sl)
9634 + ret = ACCESS_ONCE(sl->seqcount.sequence);
9635 + if (unlikely(ret & 1)) {
9637 + * Take the lock and let the writer proceed (i.e. evtl
9638 + * boost it), otherwise we could loop here forever.
9640 + spin_unlock_wait(&sl->lock);
9647 static inline unsigned read_seqretry(const seqlock_t *sl, unsigned start)
9649 @@ -446,36 +478,45 @@ static inline unsigned read_seqretry(const seqlock_t *sl, unsigned start)
9650 static inline void write_seqlock(seqlock_t *sl)
9652 spin_lock(&sl->lock);
9653 - write_seqcount_begin(&sl->seqcount);
9654 + __raw_write_seqcount_begin(&sl->seqcount);
9657 +static inline int try_write_seqlock(seqlock_t *sl)
9659 + if (spin_trylock(&sl->lock)) {
9660 + __raw_write_seqcount_begin(&sl->seqcount);
9666 static inline void write_sequnlock(seqlock_t *sl)
9668 - write_seqcount_end(&sl->seqcount);
9669 + __raw_write_seqcount_end(&sl->seqcount);
9670 spin_unlock(&sl->lock);
9673 static inline void write_seqlock_bh(seqlock_t *sl)
9675 spin_lock_bh(&sl->lock);
9676 - write_seqcount_begin(&sl->seqcount);
9677 + __raw_write_seqcount_begin(&sl->seqcount);
9680 static inline void write_sequnlock_bh(seqlock_t *sl)
9682 - write_seqcount_end(&sl->seqcount);
9683 + __raw_write_seqcount_end(&sl->seqcount);
9684 spin_unlock_bh(&sl->lock);
9687 static inline void write_seqlock_irq(seqlock_t *sl)
9689 spin_lock_irq(&sl->lock);
9690 - write_seqcount_begin(&sl->seqcount);
9691 + __raw_write_seqcount_begin(&sl->seqcount);
9694 static inline void write_sequnlock_irq(seqlock_t *sl)
9696 - write_seqcount_end(&sl->seqcount);
9697 + __raw_write_seqcount_end(&sl->seqcount);
9698 spin_unlock_irq(&sl->lock);
9701 @@ -484,7 +525,7 @@ static inline unsigned long __write_seqlock_irqsave(seqlock_t *sl)
9702 unsigned long flags;
9704 spin_lock_irqsave(&sl->lock, flags);
9705 - write_seqcount_begin(&sl->seqcount);
9706 + __raw_write_seqcount_begin(&sl->seqcount);
9710 @@ -494,7 +535,7 @@ static inline unsigned long __write_seqlock_irqsave(seqlock_t *sl)
9712 write_sequnlock_irqrestore(seqlock_t *sl, unsigned long flags)
9714 - write_seqcount_end(&sl->seqcount);
9715 + __raw_write_seqcount_end(&sl->seqcount);
9716 spin_unlock_irqrestore(&sl->lock, flags);
9719 diff --git a/include/linux/signal.h b/include/linux/signal.h
9720 index b63f63eaa39c..295540fdfc72 100644
9721 --- a/include/linux/signal.h
9722 +++ b/include/linux/signal.h
9723 @@ -233,6 +233,7 @@ static inline void init_sigpending(struct sigpending *sig)
9726 extern void flush_sigqueue(struct sigpending *queue);
9727 +extern void flush_task_sigqueue(struct task_struct *tsk);
9729 /* Test if 'sig' is valid signal. Use this instead of testing _NSIG directly */
9730 static inline int valid_signal(unsigned long sig)
9731 diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
9732 index 32810f279f8e..0db6e31161f6 100644
9733 --- a/include/linux/skbuff.h
9734 +++ b/include/linux/skbuff.h
9735 @@ -284,6 +284,7 @@ struct sk_buff_head {
9739 + raw_spinlock_t raw_lock;
9743 @@ -1573,6 +1574,12 @@ static inline void skb_queue_head_init(struct sk_buff_head *list)
9744 __skb_queue_head_init(list);
9747 +static inline void skb_queue_head_init_raw(struct sk_buff_head *list)
9749 + raw_spin_lock_init(&list->raw_lock);
9750 + __skb_queue_head_init(list);
9753 static inline void skb_queue_head_init_class(struct sk_buff_head *list,
9754 struct lock_class_key *class)
9756 diff --git a/include/linux/smp.h b/include/linux/smp.h
9757 index 8e0cb7a0f836..b16ca967ad80 100644
9758 --- a/include/linux/smp.h
9759 +++ b/include/linux/smp.h
9760 @@ -185,6 +185,9 @@ static inline void smp_init(void) { }
9761 #define get_cpu() ({ preempt_disable(); smp_processor_id(); })
9762 #define put_cpu() preempt_enable()
9764 +#define get_cpu_light() ({ migrate_disable(); smp_processor_id(); })
9765 +#define put_cpu_light() migrate_enable()
9768 * Callback to arch code if there's nosmp or maxcpus=0 on the
9769 * boot command line:
9770 diff --git a/include/linux/spinlock.h b/include/linux/spinlock.h
9771 index 47dd0cebd204..02928fa5499d 100644
9772 --- a/include/linux/spinlock.h
9773 +++ b/include/linux/spinlock.h
9774 @@ -271,7 +271,11 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock)
9775 #define raw_spin_can_lock(lock) (!raw_spin_is_locked(lock))
9777 /* Include rwlock functions */
9778 -#include <linux/rwlock.h>
9779 +#ifdef CONFIG_PREEMPT_RT_FULL
9780 +# include <linux/rwlock_rt.h>
9782 +# include <linux/rwlock.h>
9786 * Pull the _spin_*()/_read_*()/_write_*() functions/declarations:
9787 @@ -282,6 +286,10 @@ static inline void do_raw_spin_unlock(raw_spinlock_t *lock) __releases(lock)
9788 # include <linux/spinlock_api_up.h>
9791 +#ifdef CONFIG_PREEMPT_RT_FULL
9792 +# include <linux/spinlock_rt.h>
9793 +#else /* PREEMPT_RT_FULL */
9796 * Map the spin_lock functions to the raw variants for PREEMPT_RT=n
9798 @@ -347,6 +355,12 @@ static __always_inline void spin_unlock(spinlock_t *lock)
9799 raw_spin_unlock(&lock->rlock);
9802 +static __always_inline int spin_unlock_no_deboost(spinlock_t *lock)
9804 + raw_spin_unlock(&lock->rlock);
9808 static __always_inline void spin_unlock_bh(spinlock_t *lock)
9810 raw_spin_unlock_bh(&lock->rlock);
9811 @@ -416,4 +430,6 @@ extern int _atomic_dec_and_lock(atomic_t *atomic, spinlock_t *lock);
9812 #define atomic_dec_and_lock(atomic, lock) \
9813 __cond_lock(lock, _atomic_dec_and_lock(atomic, lock))
9815 +#endif /* !PREEMPT_RT_FULL */
9817 #endif /* __LINUX_SPINLOCK_H */
9818 diff --git a/include/linux/spinlock_api_smp.h b/include/linux/spinlock_api_smp.h
9819 index 5344268e6e62..043263f30e81 100644
9820 --- a/include/linux/spinlock_api_smp.h
9821 +++ b/include/linux/spinlock_api_smp.h
9822 @@ -189,6 +189,8 @@ static inline int __raw_spin_trylock_bh(raw_spinlock_t *lock)
9826 -#include <linux/rwlock_api_smp.h>
9827 +#ifndef CONFIG_PREEMPT_RT_FULL
9828 +# include <linux/rwlock_api_smp.h>
9831 #endif /* __LINUX_SPINLOCK_API_SMP_H */
9832 diff --git a/include/linux/spinlock_rt.h b/include/linux/spinlock_rt.h
9833 new file mode 100644
9834 index 000000000000..3534cff3dd08
9836 +++ b/include/linux/spinlock_rt.h
9838 +#ifndef __LINUX_SPINLOCK_RT_H
9839 +#define __LINUX_SPINLOCK_RT_H
9841 +#ifndef __LINUX_SPINLOCK_H
9842 +#error Do not include directly. Use spinlock.h
9845 +#include <linux/bug.h>
9848 +__rt_spin_lock_init(spinlock_t *lock, char *name, struct lock_class_key *key);
9850 +#define spin_lock_init(slock) \
9852 + static struct lock_class_key __key; \
9854 + rt_mutex_init(&(slock)->lock); \
9855 + __rt_spin_lock_init(slock, #slock, &__key); \
9858 +void __lockfunc rt_spin_lock__no_mg(spinlock_t *lock);
9859 +void __lockfunc rt_spin_unlock__no_mg(spinlock_t *lock);
9860 +int __lockfunc rt_spin_trylock__no_mg(spinlock_t *lock);
9862 +extern void __lockfunc rt_spin_lock(spinlock_t *lock);
9863 +extern unsigned long __lockfunc rt_spin_lock_trace_flags(spinlock_t *lock);
9864 +extern void __lockfunc rt_spin_lock_nested(spinlock_t *lock, int subclass);
9865 +extern void __lockfunc rt_spin_unlock(spinlock_t *lock);
9866 +extern int __lockfunc rt_spin_unlock_no_deboost(spinlock_t *lock);
9867 +extern void __lockfunc rt_spin_unlock_wait(spinlock_t *lock);
9868 +extern int __lockfunc rt_spin_trylock_irqsave(spinlock_t *lock, unsigned long *flags);
9869 +extern int __lockfunc rt_spin_trylock_bh(spinlock_t *lock);
9870 +extern int __lockfunc rt_spin_trylock(spinlock_t *lock);
9871 +extern int atomic_dec_and_spin_lock(atomic_t *atomic, spinlock_t *lock);
9874 + * lockdep-less calls, for derived types like rwlock:
9875 + * (for trylock they can use rt_mutex_trylock() directly.
9877 +extern void __lockfunc __rt_spin_lock__no_mg(struct rt_mutex *lock);
9878 +extern void __lockfunc __rt_spin_lock(struct rt_mutex *lock);
9879 +extern void __lockfunc __rt_spin_unlock(struct rt_mutex *lock);
9881 +#define spin_lock(lock) rt_spin_lock(lock)
9883 +#define spin_lock_bh(lock) \
9885 + local_bh_disable(); \
9886 + rt_spin_lock(lock); \
9889 +#define spin_lock_irq(lock) spin_lock(lock)
9891 +#define spin_do_trylock(lock) __cond_lock(lock, rt_spin_trylock(lock))
9893 +#define spin_trylock(lock) \
9896 + __locked = spin_do_trylock(lock); \
9900 +#ifdef CONFIG_LOCKDEP
9901 +# define spin_lock_nested(lock, subclass) \
9903 + rt_spin_lock_nested(lock, subclass); \
9906 +#define spin_lock_bh_nested(lock, subclass) \
9908 + local_bh_disable(); \
9909 + rt_spin_lock_nested(lock, subclass); \
9912 +# define spin_lock_irqsave_nested(lock, flags, subclass) \
9914 + typecheck(unsigned long, flags); \
9916 + rt_spin_lock_nested(lock, subclass); \
9919 +# define spin_lock_nested(lock, subclass) spin_lock(lock)
9920 +# define spin_lock_bh_nested(lock, subclass) spin_lock_bh(lock)
9922 +# define spin_lock_irqsave_nested(lock, flags, subclass) \
9924 + typecheck(unsigned long, flags); \
9926 + spin_lock(lock); \
9930 +#define spin_lock_irqsave(lock, flags) \
9932 + typecheck(unsigned long, flags); \
9934 + spin_lock(lock); \
9937 +static inline unsigned long spin_lock_trace_flags(spinlock_t *lock)
9939 + unsigned long flags = 0;
9940 +#ifdef CONFIG_TRACE_IRQFLAGS
9941 + flags = rt_spin_lock_trace_flags(lock);
9943 + spin_lock(lock); /* lock_local */
9948 +/* FIXME: we need rt_spin_lock_nest_lock */
9949 +#define spin_lock_nest_lock(lock, nest_lock) spin_lock_nested(lock, 0)
9951 +#define spin_unlock(lock) rt_spin_unlock(lock)
9952 +#define spin_unlock_no_deboost(lock) rt_spin_unlock_no_deboost(lock)
9954 +#define spin_unlock_bh(lock) \
9956 + rt_spin_unlock(lock); \
9957 + local_bh_enable(); \
9960 +#define spin_unlock_irq(lock) spin_unlock(lock)
9962 +#define spin_unlock_irqrestore(lock, flags) \
9964 + typecheck(unsigned long, flags); \
9966 + spin_unlock(lock); \
9969 +#define spin_trylock_bh(lock) __cond_lock(lock, rt_spin_trylock_bh(lock))
9970 +#define spin_trylock_irq(lock) spin_trylock(lock)
9972 +#define spin_trylock_irqsave(lock, flags) \
9973 + rt_spin_trylock_irqsave(lock, &(flags))
9975 +#define spin_unlock_wait(lock) rt_spin_unlock_wait(lock)
9977 +#ifdef CONFIG_GENERIC_LOCKBREAK
9978 +# define spin_is_contended(lock) ((lock)->break_lock)
9980 +# define spin_is_contended(lock) (((void)(lock), 0))
9983 +static inline int spin_can_lock(spinlock_t *lock)
9985 + return !rt_mutex_is_locked(&lock->lock);
9988 +static inline int spin_is_locked(spinlock_t *lock)
9990 + return rt_mutex_is_locked(&lock->lock);
9993 +static inline void assert_spin_locked(spinlock_t *lock)
9995 + BUG_ON(!spin_is_locked(lock));
9998 +#define atomic_dec_and_lock(atomic, lock) \
9999 + atomic_dec_and_spin_lock(atomic, lock)
10002 diff --git a/include/linux/spinlock_types.h b/include/linux/spinlock_types.h
10003 index 73548eb13a5d..10bac715ea96 100644
10004 --- a/include/linux/spinlock_types.h
10005 +++ b/include/linux/spinlock_types.h
10007 * Released under the General Public License (GPL).
10010 -#if defined(CONFIG_SMP)
10011 -# include <asm/spinlock_types.h>
10012 +#include <linux/spinlock_types_raw.h>
10014 +#ifndef CONFIG_PREEMPT_RT_FULL
10015 +# include <linux/spinlock_types_nort.h>
10016 +# include <linux/rwlock_types.h>
10018 -# include <linux/spinlock_types_up.h>
10019 +# include <linux/rtmutex.h>
10020 +# include <linux/spinlock_types_rt.h>
10021 +# include <linux/rwlock_types_rt.h>
10024 -#include <linux/lockdep.h>
10026 -typedef struct raw_spinlock {
10027 - arch_spinlock_t raw_lock;
10028 -#ifdef CONFIG_GENERIC_LOCKBREAK
10029 - unsigned int break_lock;
10031 -#ifdef CONFIG_DEBUG_SPINLOCK
10032 - unsigned int magic, owner_cpu;
10035 -#ifdef CONFIG_DEBUG_LOCK_ALLOC
10036 - struct lockdep_map dep_map;
10040 -#define SPINLOCK_MAGIC 0xdead4ead
10042 -#define SPINLOCK_OWNER_INIT ((void *)-1L)
10044 -#ifdef CONFIG_DEBUG_LOCK_ALLOC
10045 -# define SPIN_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname }
10047 -# define SPIN_DEP_MAP_INIT(lockname)
10050 -#ifdef CONFIG_DEBUG_SPINLOCK
10051 -# define SPIN_DEBUG_INIT(lockname) \
10052 - .magic = SPINLOCK_MAGIC, \
10053 - .owner_cpu = -1, \
10054 - .owner = SPINLOCK_OWNER_INIT,
10056 -# define SPIN_DEBUG_INIT(lockname)
10059 -#define __RAW_SPIN_LOCK_INITIALIZER(lockname) \
10061 - .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \
10062 - SPIN_DEBUG_INIT(lockname) \
10063 - SPIN_DEP_MAP_INIT(lockname) }
10065 -#define __RAW_SPIN_LOCK_UNLOCKED(lockname) \
10066 - (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname)
10068 -#define DEFINE_RAW_SPINLOCK(x) raw_spinlock_t x = __RAW_SPIN_LOCK_UNLOCKED(x)
10070 -typedef struct spinlock {
10072 - struct raw_spinlock rlock;
10074 -#ifdef CONFIG_DEBUG_LOCK_ALLOC
10075 -# define LOCK_PADSIZE (offsetof(struct raw_spinlock, dep_map))
10077 - u8 __padding[LOCK_PADSIZE];
10078 - struct lockdep_map dep_map;
10084 -#define __SPIN_LOCK_INITIALIZER(lockname) \
10085 - { { .rlock = __RAW_SPIN_LOCK_INITIALIZER(lockname) } }
10087 -#define __SPIN_LOCK_UNLOCKED(lockname) \
10088 - (spinlock_t ) __SPIN_LOCK_INITIALIZER(lockname)
10090 -#define DEFINE_SPINLOCK(x) spinlock_t x = __SPIN_LOCK_UNLOCKED(x)
10092 -#include <linux/rwlock_types.h>
10094 #endif /* __LINUX_SPINLOCK_TYPES_H */
10095 diff --git a/include/linux/spinlock_types_nort.h b/include/linux/spinlock_types_nort.h
10096 new file mode 100644
10097 index 000000000000..f1dac1fb1d6a
10099 +++ b/include/linux/spinlock_types_nort.h
10101 +#ifndef __LINUX_SPINLOCK_TYPES_NORT_H
10102 +#define __LINUX_SPINLOCK_TYPES_NORT_H
10104 +#ifndef __LINUX_SPINLOCK_TYPES_H
10105 +#error "Do not include directly. Include spinlock_types.h instead"
10109 + * The non RT version maps spinlocks to raw_spinlocks
10111 +typedef struct spinlock {
10113 + struct raw_spinlock rlock;
10115 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
10116 +# define LOCK_PADSIZE (offsetof(struct raw_spinlock, dep_map))
10118 + u8 __padding[LOCK_PADSIZE];
10119 + struct lockdep_map dep_map;
10125 +#define __SPIN_LOCK_INITIALIZER(lockname) \
10126 + { { .rlock = __RAW_SPIN_LOCK_INITIALIZER(lockname) } }
10128 +#define __SPIN_LOCK_UNLOCKED(lockname) \
10129 + (spinlock_t ) __SPIN_LOCK_INITIALIZER(lockname)
10131 +#define DEFINE_SPINLOCK(x) spinlock_t x = __SPIN_LOCK_UNLOCKED(x)
10134 diff --git a/include/linux/spinlock_types_raw.h b/include/linux/spinlock_types_raw.h
10135 new file mode 100644
10136 index 000000000000..edffc4d53fc9
10138 +++ b/include/linux/spinlock_types_raw.h
10140 +#ifndef __LINUX_SPINLOCK_TYPES_RAW_H
10141 +#define __LINUX_SPINLOCK_TYPES_RAW_H
10143 +#if defined(CONFIG_SMP)
10144 +# include <asm/spinlock_types.h>
10146 +# include <linux/spinlock_types_up.h>
10149 +#include <linux/lockdep.h>
10151 +typedef struct raw_spinlock {
10152 + arch_spinlock_t raw_lock;
10153 +#ifdef CONFIG_GENERIC_LOCKBREAK
10154 + unsigned int break_lock;
10156 +#ifdef CONFIG_DEBUG_SPINLOCK
10157 + unsigned int magic, owner_cpu;
10160 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
10161 + struct lockdep_map dep_map;
10165 +#define SPINLOCK_MAGIC 0xdead4ead
10167 +#define SPINLOCK_OWNER_INIT ((void *)-1L)
10169 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
10170 +# define SPIN_DEP_MAP_INIT(lockname) .dep_map = { .name = #lockname }
10172 +# define SPIN_DEP_MAP_INIT(lockname)
10175 +#ifdef CONFIG_DEBUG_SPINLOCK
10176 +# define SPIN_DEBUG_INIT(lockname) \
10177 + .magic = SPINLOCK_MAGIC, \
10178 + .owner_cpu = -1, \
10179 + .owner = SPINLOCK_OWNER_INIT,
10181 +# define SPIN_DEBUG_INIT(lockname)
10184 +#define __RAW_SPIN_LOCK_INITIALIZER(lockname) \
10186 + .raw_lock = __ARCH_SPIN_LOCK_UNLOCKED, \
10187 + SPIN_DEBUG_INIT(lockname) \
10188 + SPIN_DEP_MAP_INIT(lockname) }
10190 +#define __RAW_SPIN_LOCK_UNLOCKED(lockname) \
10191 + (raw_spinlock_t) __RAW_SPIN_LOCK_INITIALIZER(lockname)
10193 +#define DEFINE_RAW_SPINLOCK(x) raw_spinlock_t x = __RAW_SPIN_LOCK_UNLOCKED(x)
10196 diff --git a/include/linux/spinlock_types_rt.h b/include/linux/spinlock_types_rt.h
10197 new file mode 100644
10198 index 000000000000..3e3d8c5f7a9a
10200 +++ b/include/linux/spinlock_types_rt.h
10202 +#ifndef __LINUX_SPINLOCK_TYPES_RT_H
10203 +#define __LINUX_SPINLOCK_TYPES_RT_H
10205 +#ifndef __LINUX_SPINLOCK_TYPES_H
10206 +#error "Do not include directly. Include spinlock_types.h instead"
10209 +#include <linux/cache.h>
10212 + * PREEMPT_RT: spinlocks - an RT mutex plus lock-break field:
10214 +typedef struct spinlock {
10215 + struct rt_mutex lock;
10216 + unsigned int break_lock;
10217 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
10218 + struct lockdep_map dep_map;
10222 +#ifdef CONFIG_DEBUG_RT_MUTEXES
10223 +# define __RT_SPIN_INITIALIZER(name) \
10225 + .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(name.wait_lock), \
10226 + .save_state = 1, \
10227 + .file = __FILE__, \
10228 + .line = __LINE__ , \
10231 +# define __RT_SPIN_INITIALIZER(name) \
10233 + .wait_lock = __RAW_SPIN_LOCK_UNLOCKED(name.wait_lock), \
10234 + .save_state = 1, \
10239 +.wait_list = PLIST_HEAD_INIT_RAW((name).lock.wait_list, (name).lock.wait_lock)
10242 +#define __SPIN_LOCK_UNLOCKED(name) \
10243 + { .lock = __RT_SPIN_INITIALIZER(name.lock), \
10244 + SPIN_DEP_MAP_INIT(name) }
10246 +#define DEFINE_SPINLOCK(name) \
10247 + spinlock_t name = __SPIN_LOCK_UNLOCKED(name)
10250 diff --git a/include/linux/srcu.h b/include/linux/srcu.h
10251 index dc8eb63c6568..e793d3a257da 100644
10252 --- a/include/linux/srcu.h
10253 +++ b/include/linux/srcu.h
10254 @@ -84,10 +84,10 @@ int init_srcu_struct(struct srcu_struct *sp);
10256 void process_srcu(struct work_struct *work);
10258 -#define __SRCU_STRUCT_INIT(name) \
10259 +#define __SRCU_STRUCT_INIT(name, pcpu_name) \
10261 .completed = -300, \
10262 - .per_cpu_ref = &name##_srcu_array, \
10263 + .per_cpu_ref = &pcpu_name, \
10264 .queue_lock = __SPIN_LOCK_UNLOCKED(name.queue_lock), \
10265 .running = false, \
10266 .batch_queue = RCU_BATCH_INIT(name.batch_queue), \
10267 @@ -119,7 +119,7 @@ void process_srcu(struct work_struct *work);
10269 #define __DEFINE_SRCU(name, is_static) \
10270 static DEFINE_PER_CPU(struct srcu_struct_array, name##_srcu_array);\
10271 - is_static struct srcu_struct name = __SRCU_STRUCT_INIT(name)
10272 + is_static struct srcu_struct name = __SRCU_STRUCT_INIT(name, name##_srcu_array)
10273 #define DEFINE_SRCU(name) __DEFINE_SRCU(name, /* not static */)
10274 #define DEFINE_STATIC_SRCU(name) __DEFINE_SRCU(name, static)
10276 diff --git a/include/linux/suspend.h b/include/linux/suspend.h
10277 index d9718378a8be..e81e6dc7dcb1 100644
10278 --- a/include/linux/suspend.h
10279 +++ b/include/linux/suspend.h
10280 @@ -193,6 +193,12 @@ struct platform_freeze_ops {
10284 +#if defined(CONFIG_SUSPEND) || defined(CONFIG_HIBERNATION)
10285 +extern bool pm_in_action;
10287 +# define pm_in_action false
10290 #ifdef CONFIG_SUSPEND
10292 * suspend_set_ops - set platform dependent suspend operations
10293 diff --git a/include/linux/swait.h b/include/linux/swait.h
10294 index c1f9c62a8a50..83f004a72320 100644
10295 --- a/include/linux/swait.h
10296 +++ b/include/linux/swait.h
10297 @@ -87,6 +87,7 @@ static inline int swait_active(struct swait_queue_head *q)
10298 extern void swake_up(struct swait_queue_head *q);
10299 extern void swake_up_all(struct swait_queue_head *q);
10300 extern void swake_up_locked(struct swait_queue_head *q);
10301 +extern void swake_up_all_locked(struct swait_queue_head *q);
10303 extern void __prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait);
10304 extern void prepare_to_swait(struct swait_queue_head *q, struct swait_queue *wait, int state);
10305 diff --git a/include/linux/swap.h b/include/linux/swap.h
10306 index 55ff5593c193..52bf5477dc92 100644
10307 --- a/include/linux/swap.h
10308 +++ b/include/linux/swap.h
10310 #include <linux/fs.h>
10311 #include <linux/atomic.h>
10312 #include <linux/page-flags.h>
10313 +#include <linux/locallock.h>
10314 #include <asm/page.h>
10316 struct notifier_block;
10317 @@ -247,7 +248,8 @@ struct swap_info_struct {
10318 void *workingset_eviction(struct address_space *mapping, struct page *page);
10319 bool workingset_refault(void *shadow);
10320 void workingset_activation(struct page *page);
10321 -extern struct list_lru workingset_shadow_nodes;
10322 +extern struct list_lru __workingset_shadow_nodes;
10323 +DECLARE_LOCAL_IRQ_LOCK(workingset_shadow_lock);
10325 static inline unsigned int workingset_node_pages(struct radix_tree_node *node)
10327 @@ -292,6 +294,7 @@ extern unsigned long nr_free_pagecache_pages(void);
10330 /* linux/mm/swap.c */
10331 +DECLARE_LOCAL_IRQ_LOCK(swapvec_lock);
10332 extern void lru_cache_add(struct page *);
10333 extern void lru_cache_add_anon(struct page *page);
10334 extern void lru_cache_add_file(struct page *page);
10335 diff --git a/include/linux/swork.h b/include/linux/swork.h
10336 new file mode 100644
10337 index 000000000000..f175fa9a6016
10339 +++ b/include/linux/swork.h
10341 +#ifndef _LINUX_SWORK_H
10342 +#define _LINUX_SWORK_H
10344 +#include <linux/list.h>
10346 +struct swork_event {
10347 + struct list_head item;
10348 + unsigned long flags;
10349 + void (*func)(struct swork_event *);
10352 +static inline void INIT_SWORK(struct swork_event *event,
10353 + void (*func)(struct swork_event *))
10355 + event->flags = 0;
10356 + event->func = func;
10359 +bool swork_queue(struct swork_event *sev);
10361 +int swork_get(void);
10362 +void swork_put(void);
10364 +#endif /* _LINUX_SWORK_H */
10365 diff --git a/include/linux/thread_info.h b/include/linux/thread_info.h
10366 index 2873baf5372a..eb1a108f17ca 100644
10367 --- a/include/linux/thread_info.h
10368 +++ b/include/linux/thread_info.h
10369 @@ -107,7 +107,17 @@ static inline int test_ti_thread_flag(struct thread_info *ti, int flag)
10370 #define test_thread_flag(flag) \
10371 test_ti_thread_flag(current_thread_info(), flag)
10373 -#define tif_need_resched() test_thread_flag(TIF_NEED_RESCHED)
10374 +#ifdef CONFIG_PREEMPT_LAZY
10375 +#define tif_need_resched() (test_thread_flag(TIF_NEED_RESCHED) || \
10376 + test_thread_flag(TIF_NEED_RESCHED_LAZY))
10377 +#define tif_need_resched_now() (test_thread_flag(TIF_NEED_RESCHED))
10378 +#define tif_need_resched_lazy() test_thread_flag(TIF_NEED_RESCHED_LAZY))
10381 +#define tif_need_resched() test_thread_flag(TIF_NEED_RESCHED)
10382 +#define tif_need_resched_now() test_thread_flag(TIF_NEED_RESCHED)
10383 +#define tif_need_resched_lazy() 0
10386 #ifndef CONFIG_HAVE_ARCH_WITHIN_STACK_FRAMES
10387 static inline int arch_within_stack_frames(const void * const stack,
10388 diff --git a/include/linux/timer.h b/include/linux/timer.h
10389 index 51d601f192d4..83cea629efe1 100644
10390 --- a/include/linux/timer.h
10391 +++ b/include/linux/timer.h
10392 @@ -241,7 +241,7 @@ extern void add_timer(struct timer_list *timer);
10394 extern int try_to_del_timer_sync(struct timer_list *timer);
10397 +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT_FULL)
10398 extern int del_timer_sync(struct timer_list *timer);
10400 # define del_timer_sync(t) del_timer(t)
10401 diff --git a/include/linux/trace_events.h b/include/linux/trace_events.h
10402 index be007610ceb0..15154b13a53b 100644
10403 --- a/include/linux/trace_events.h
10404 +++ b/include/linux/trace_events.h
10405 @@ -56,6 +56,9 @@ struct trace_entry {
10406 unsigned char flags;
10407 unsigned char preempt_count;
10409 + unsigned short migrate_disable;
10410 + unsigned short padding;
10411 + unsigned char preempt_lazy_count;
10414 #define TRACE_EVENT_TYPE_MAX \
10415 diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h
10416 index f30c187ed785..83bf0f798426 100644
10417 --- a/include/linux/uaccess.h
10418 +++ b/include/linux/uaccess.h
10419 @@ -24,6 +24,7 @@ static __always_inline void pagefault_disabled_dec(void)
10421 static inline void pagefault_disable(void)
10423 + migrate_disable();
10424 pagefault_disabled_inc();
10426 * make sure to have issued the store before a pagefault
10427 @@ -40,6 +41,7 @@ static inline void pagefault_enable(void)
10430 pagefault_disabled_dec();
10431 + migrate_enable();
10435 diff --git a/include/linux/uprobes.h b/include/linux/uprobes.h
10436 index 4a29c75b146e..0a294e950df8 100644
10437 --- a/include/linux/uprobes.h
10438 +++ b/include/linux/uprobes.h
10440 #include <linux/errno.h>
10441 #include <linux/rbtree.h>
10442 #include <linux/types.h>
10443 +#include <linux/wait.h>
10445 struct vm_area_struct;
10447 diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
10448 index 613771909b6e..e28c5a43229d 100644
10449 --- a/include/linux/vmstat.h
10450 +++ b/include/linux/vmstat.h
10451 @@ -33,7 +33,9 @@ DECLARE_PER_CPU(struct vm_event_state, vm_event_states);
10453 static inline void __count_vm_event(enum vm_event_item item)
10455 + preempt_disable_rt();
10456 raw_cpu_inc(vm_event_states.event[item]);
10457 + preempt_enable_rt();
10460 static inline void count_vm_event(enum vm_event_item item)
10461 @@ -43,7 +45,9 @@ static inline void count_vm_event(enum vm_event_item item)
10463 static inline void __count_vm_events(enum vm_event_item item, long delta)
10465 + preempt_disable_rt();
10466 raw_cpu_add(vm_event_states.event[item], delta);
10467 + preempt_enable_rt();
10470 static inline void count_vm_events(enum vm_event_item item, long delta)
10471 diff --git a/include/linux/wait.h b/include/linux/wait.h
10472 index 2408e8d5c05c..db50d6609195 100644
10473 --- a/include/linux/wait.h
10474 +++ b/include/linux/wait.h
10476 #include <linux/spinlock.h>
10477 #include <asm/current.h>
10478 #include <uapi/linux/wait.h>
10479 +#include <linux/atomic.h>
10481 typedef struct __wait_queue wait_queue_t;
10482 typedef int (*wait_queue_func_t)(wait_queue_t *wait, unsigned mode, int flags, void *key);
10483 diff --git a/include/net/dst.h b/include/net/dst.h
10484 index 6835d224d47b..55a5a9698f14 100644
10485 --- a/include/net/dst.h
10486 +++ b/include/net/dst.h
10487 @@ -446,7 +446,7 @@ static inline void dst_confirm(struct dst_entry *dst)
10488 static inline int dst_neigh_output(struct dst_entry *dst, struct neighbour *n,
10489 struct sk_buff *skb)
10491 - const struct hh_cache *hh;
10492 + struct hh_cache *hh;
10494 if (dst->pending_confirm) {
10495 unsigned long now = jiffies;
10496 diff --git a/include/net/gen_stats.h b/include/net/gen_stats.h
10497 index 231e121cc7d9..d125222b979d 100644
10498 --- a/include/net/gen_stats.h
10499 +++ b/include/net/gen_stats.h
10501 #include <linux/socket.h>
10502 #include <linux/rtnetlink.h>
10503 #include <linux/pkt_sched.h>
10504 +#include <net/net_seq_lock.h>
10506 struct gnet_stats_basic_cpu {
10507 struct gnet_stats_basic_packed bstats;
10508 @@ -33,11 +34,11 @@ int gnet_stats_start_copy_compat(struct sk_buff *skb, int type,
10509 spinlock_t *lock, struct gnet_dump *d,
10512 -int gnet_stats_copy_basic(const seqcount_t *running,
10513 +int gnet_stats_copy_basic(net_seqlock_t *running,
10514 struct gnet_dump *d,
10515 struct gnet_stats_basic_cpu __percpu *cpu,
10516 struct gnet_stats_basic_packed *b);
10517 -void __gnet_stats_copy_basic(const seqcount_t *running,
10518 +void __gnet_stats_copy_basic(net_seqlock_t *running,
10519 struct gnet_stats_basic_packed *bstats,
10520 struct gnet_stats_basic_cpu __percpu *cpu,
10521 struct gnet_stats_basic_packed *b);
10522 @@ -55,14 +56,14 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
10523 struct gnet_stats_basic_cpu __percpu *cpu_bstats,
10524 struct gnet_stats_rate_est64 *rate_est,
10525 spinlock_t *stats_lock,
10526 - seqcount_t *running, struct nlattr *opt);
10527 + net_seqlock_t *running, struct nlattr *opt);
10528 void gen_kill_estimator(struct gnet_stats_basic_packed *bstats,
10529 struct gnet_stats_rate_est64 *rate_est);
10530 int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
10531 struct gnet_stats_basic_cpu __percpu *cpu_bstats,
10532 struct gnet_stats_rate_est64 *rate_est,
10533 spinlock_t *stats_lock,
10534 - seqcount_t *running, struct nlattr *opt);
10535 + net_seqlock_t *running, struct nlattr *opt);
10536 bool gen_estimator_active(const struct gnet_stats_basic_packed *bstats,
10537 const struct gnet_stats_rate_est64 *rate_est);
10539 diff --git a/include/net/neighbour.h b/include/net/neighbour.h
10540 index 8b683841e574..bf656008f6e7 100644
10541 --- a/include/net/neighbour.h
10542 +++ b/include/net/neighbour.h
10543 @@ -446,7 +446,7 @@ static inline int neigh_hh_bridge(struct hh_cache *hh, struct sk_buff *skb)
10547 -static inline int neigh_hh_output(const struct hh_cache *hh, struct sk_buff *skb)
10548 +static inline int neigh_hh_output(struct hh_cache *hh, struct sk_buff *skb)
10552 @@ -501,7 +501,7 @@ struct neighbour_cb {
10554 #define NEIGH_CB(skb) ((struct neighbour_cb *)(skb)->cb)
10556 -static inline void neigh_ha_snapshot(char *dst, const struct neighbour *n,
10557 +static inline void neigh_ha_snapshot(char *dst, struct neighbour *n,
10558 const struct net_device *dev)
10561 diff --git a/include/net/net_seq_lock.h b/include/net/net_seq_lock.h
10562 new file mode 100644
10563 index 000000000000..a7034298a82a
10565 +++ b/include/net/net_seq_lock.h
10567 +#ifndef __NET_NET_SEQ_LOCK_H__
10568 +#define __NET_NET_SEQ_LOCK_H__
10570 +#ifdef CONFIG_PREEMPT_RT_BASE
10571 +# define net_seqlock_t seqlock_t
10572 +# define net_seq_begin(__r) read_seqbegin(__r)
10573 +# define net_seq_retry(__r, __s) read_seqretry(__r, __s)
10576 +# define net_seqlock_t seqcount_t
10577 +# define net_seq_begin(__r) read_seqcount_begin(__r)
10578 +# define net_seq_retry(__r, __s) read_seqcount_retry(__r, __s)
10582 diff --git a/include/net/netns/ipv4.h b/include/net/netns/ipv4.h
10583 index 7adf4386ac8f..d3fd5c357268 100644
10584 --- a/include/net/netns/ipv4.h
10585 +++ b/include/net/netns/ipv4.h
10586 @@ -69,6 +69,7 @@ struct netns_ipv4 {
10588 int sysctl_icmp_echo_ignore_all;
10589 int sysctl_icmp_echo_ignore_broadcasts;
10590 + int sysctl_icmp_echo_sysrq;
10591 int sysctl_icmp_ignore_bogus_error_responses;
10592 int sysctl_icmp_ratelimit;
10593 int sysctl_icmp_ratemask;
10594 diff --git a/include/net/sch_generic.h b/include/net/sch_generic.h
10595 index e6aa0a249672..b57736f2a8a3 100644
10596 --- a/include/net/sch_generic.h
10597 +++ b/include/net/sch_generic.h
10599 #include <linux/dynamic_queue_limits.h>
10600 #include <net/gen_stats.h>
10601 #include <net/rtnetlink.h>
10602 +#include <net/net_seq_lock.h>
10605 struct qdisc_walker;
10606 @@ -86,7 +87,7 @@ struct Qdisc {
10607 struct sk_buff *gso_skb ____cacheline_aligned_in_smp;
10608 struct qdisc_skb_head q;
10609 struct gnet_stats_basic_packed bstats;
10610 - seqcount_t running;
10611 + net_seqlock_t running;
10612 struct gnet_stats_queue qstats;
10613 unsigned long state;
10614 struct Qdisc *next_sched;
10615 @@ -98,13 +99,22 @@ struct Qdisc {
10616 spinlock_t busylock ____cacheline_aligned_in_smp;
10619 -static inline bool qdisc_is_running(const struct Qdisc *qdisc)
10620 +static inline bool qdisc_is_running(struct Qdisc *qdisc)
10622 +#ifdef CONFIG_PREEMPT_RT_BASE
10623 + return spin_is_locked(&qdisc->running.lock) ? true : false;
10625 return (raw_read_seqcount(&qdisc->running) & 1) ? true : false;
10629 static inline bool qdisc_run_begin(struct Qdisc *qdisc)
10631 +#ifdef CONFIG_PREEMPT_RT_BASE
10632 + if (try_write_seqlock(&qdisc->running))
10636 if (qdisc_is_running(qdisc))
10638 /* Variant of write_seqcount_begin() telling lockdep a trylock
10639 @@ -113,11 +123,16 @@ static inline bool qdisc_run_begin(struct Qdisc *qdisc)
10640 raw_write_seqcount_begin(&qdisc->running);
10641 seqcount_acquire(&qdisc->running.dep_map, 0, 1, _RET_IP_);
10646 static inline void qdisc_run_end(struct Qdisc *qdisc)
10648 +#ifdef CONFIG_PREEMPT_RT_BASE
10649 + write_sequnlock(&qdisc->running);
10651 write_seqcount_end(&qdisc->running);
10655 static inline bool qdisc_may_bulk(const struct Qdisc *qdisc)
10656 @@ -308,7 +323,7 @@ static inline spinlock_t *qdisc_root_sleeping_lock(const struct Qdisc *qdisc)
10657 return qdisc_lock(root);
10660 -static inline seqcount_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc)
10661 +static inline net_seqlock_t *qdisc_root_sleeping_running(const struct Qdisc *qdisc)
10663 struct Qdisc *root = qdisc_root_sleeping(qdisc);
10665 diff --git a/include/trace/events/hist.h b/include/trace/events/hist.h
10666 new file mode 100644
10667 index 000000000000..f7710de1b1f3
10669 +++ b/include/trace/events/hist.h
10671 +#undef TRACE_SYSTEM
10672 +#define TRACE_SYSTEM hist
10674 +#if !defined(_TRACE_HIST_H) || defined(TRACE_HEADER_MULTI_READ)
10675 +#define _TRACE_HIST_H
10677 +#include "latency_hist.h"
10678 +#include <linux/tracepoint.h>
10680 +#if !defined(CONFIG_PREEMPT_OFF_HIST) && !defined(CONFIG_INTERRUPT_OFF_HIST)
10681 +#define trace_preemptirqsoff_hist(a, b)
10682 +#define trace_preemptirqsoff_hist_rcuidle(a, b)
10684 +TRACE_EVENT(preemptirqsoff_hist,
10686 + TP_PROTO(int reason, int starthist),
10688 + TP_ARGS(reason, starthist),
10690 + TP_STRUCT__entry(
10691 + __field(int, reason)
10692 + __field(int, starthist)
10696 + __entry->reason = reason;
10697 + __entry->starthist = starthist;
10700 + TP_printk("reason=%s starthist=%s", getaction(__entry->reason),
10701 + __entry->starthist ? "start" : "stop")
10705 +#ifndef CONFIG_MISSED_TIMER_OFFSETS_HIST
10706 +#define trace_hrtimer_interrupt(a, b, c, d)
10708 +TRACE_EVENT(hrtimer_interrupt,
10710 + TP_PROTO(int cpu, long long offset, struct task_struct *curr,
10711 + struct task_struct *task),
10713 + TP_ARGS(cpu, offset, curr, task),
10715 + TP_STRUCT__entry(
10716 + __field(int, cpu)
10717 + __field(long long, offset)
10718 + __array(char, ccomm, TASK_COMM_LEN)
10719 + __field(int, cprio)
10720 + __array(char, tcomm, TASK_COMM_LEN)
10721 + __field(int, tprio)
10725 + __entry->cpu = cpu;
10726 + __entry->offset = offset;
10727 + memcpy(__entry->ccomm, curr->comm, TASK_COMM_LEN);
10728 + __entry->cprio = curr->prio;
10729 + memcpy(__entry->tcomm, task != NULL ? task->comm : "<none>",
10730 + task != NULL ? TASK_COMM_LEN : 7);
10731 + __entry->tprio = task != NULL ? task->prio : -1;
10734 + TP_printk("cpu=%d offset=%lld curr=%s[%d] thread=%s[%d]",
10735 + __entry->cpu, __entry->offset, __entry->ccomm,
10736 + __entry->cprio, __entry->tcomm, __entry->tprio)
10740 +#endif /* _TRACE_HIST_H */
10742 +/* This part must be outside protection */
10743 +#include <trace/define_trace.h>
10744 diff --git a/include/trace/events/latency_hist.h b/include/trace/events/latency_hist.h
10745 new file mode 100644
10746 index 000000000000..d3f2fbd560b1
10748 +++ b/include/trace/events/latency_hist.h
10750 +#ifndef _LATENCY_HIST_H
10751 +#define _LATENCY_HIST_H
10753 +enum hist_action {
10762 +static char *actions[] = {
10771 +static inline char *getaction(int action)
10773 + if (action >= 0 && action <= sizeof(actions)/sizeof(actions[0]))
10774 + return actions[action];
10775 + return "unknown";
10778 +#endif /* _LATENCY_HIST_H */
10779 diff --git a/init/Kconfig b/init/Kconfig
10780 index 34407f15e6d3..2ce33a32e65d 100644
10783 @@ -506,7 +506,7 @@ config TINY_RCU
10786 bool "Make expert-level adjustments to RCU configuration"
10788 + default y if PREEMPT_RT_FULL
10790 This option needs to be enabled if you wish to make
10791 expert-level adjustments to RCU configuration. By default,
10792 @@ -623,7 +623,7 @@ config RCU_FANOUT_LEAF
10794 config RCU_FAST_NO_HZ
10795 bool "Accelerate last non-dyntick-idle CPU's grace periods"
10796 - depends on NO_HZ_COMMON && SMP && RCU_EXPERT
10797 + depends on NO_HZ_COMMON && SMP && RCU_EXPERT && !PREEMPT_RT_FULL
10800 This option permits CPUs to enter dynticks-idle state even if
10801 @@ -650,7 +650,7 @@ config TREE_RCU_TRACE
10803 bool "Enable RCU priority boosting"
10804 depends on RT_MUTEXES && PREEMPT_RCU && RCU_EXPERT
10806 + default y if PREEMPT_RT_FULL
10808 This option boosts the priority of preempted RCU readers that
10809 block the current preemptible RCU grace period for too long.
10810 @@ -781,19 +781,6 @@ config RCU_NOCB_CPU_ALL
10814 -config RCU_EXPEDITE_BOOT
10818 - This option enables expedited grace periods at boot time,
10819 - as if rcu_expedite_gp() had been invoked early in boot.
10820 - The corresponding rcu_unexpedite_gp() is invoked from
10821 - rcu_end_inkernel_boot(), which is intended to be invoked
10822 - at the end of the kernel-only boot sequence, just before
10825 - Accept the default if unsure.
10827 endmenu # "RCU Subsystem"
10830 @@ -1064,6 +1051,7 @@ config CFS_BANDWIDTH
10831 config RT_GROUP_SCHED
10832 bool "Group scheduling for SCHED_RR/FIFO"
10833 depends on CGROUP_SCHED
10834 + depends on !PREEMPT_RT_FULL
10837 This feature lets you explicitly allocate real CPU bandwidth
10838 @@ -1772,6 +1760,7 @@ choice
10842 + depends on !PREEMPT_RT_FULL
10843 select HAVE_HARDENED_USERCOPY_ALLOCATOR
10845 The regular slab allocator that is established and known to work
10846 @@ -1792,6 +1781,7 @@ config SLUB
10849 bool "SLOB (Simple Allocator)"
10850 + depends on !PREEMPT_RT_FULL
10852 SLOB replaces the stock allocator with a drastically simpler
10853 allocator. SLOB is generally more space efficient but
10854 @@ -1810,7 +1800,7 @@ config SLAB_FREELIST_RANDOM
10856 config SLUB_CPU_PARTIAL
10858 - depends on SLUB && SMP
10859 + depends on SLUB && SMP && !PREEMPT_RT_FULL
10860 bool "SLUB per cpu partial cache"
10862 Per cpu partial caches accellerate objects allocation and freeing
10863 diff --git a/init/Makefile b/init/Makefile
10864 index c4fb45525d08..821190dfaa75 100644
10865 --- a/init/Makefile
10866 +++ b/init/Makefile
10867 @@ -35,4 +35,4 @@ $(obj)/version.o: include/generated/compile.h
10868 include/generated/compile.h: FORCE
10869 @$($(quiet)chk_compile.h)
10870 $(Q)$(CONFIG_SHELL) $(srctree)/scripts/mkcompile_h $@ \
10871 - "$(UTS_MACHINE)" "$(CONFIG_SMP)" "$(CONFIG_PREEMPT)" "$(CC) $(KBUILD_CFLAGS)"
10872 + "$(UTS_MACHINE)" "$(CONFIG_SMP)" "$(CONFIG_PREEMPT)" "$(CONFIG_PREEMPT_RT_FULL)" "$(CC) $(KBUILD_CFLAGS)"
10873 diff --git a/init/main.c b/init/main.c
10874 index 2858be732f6d..3c97c3c91d88 100644
10877 @@ -507,6 +507,7 @@ asmlinkage __visible void __init start_kernel(void)
10878 setup_command_line(command_line);
10879 setup_nr_cpu_ids();
10880 setup_per_cpu_areas();
10881 + softirq_early_init();
10882 boot_cpu_state_init();
10883 smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
10885 diff --git a/ipc/sem.c b/ipc/sem.c
10886 index 10b94bc59d4a..b8360eaacc7a 100644
10889 @@ -712,6 +712,13 @@ static int perform_atomic_semop(struct sem_array *sma, struct sem_queue *q)
10890 static void wake_up_sem_queue_prepare(struct list_head *pt,
10891 struct sem_queue *q, int error)
10893 +#ifdef CONFIG_PREEMPT_RT_BASE
10894 + struct task_struct *p = q->sleeper;
10895 + get_task_struct(p);
10896 + q->status = error;
10897 + wake_up_process(p);
10898 + put_task_struct(p);
10900 if (list_empty(pt)) {
10902 * Hold preempt off so that we don't get preempted and have the
10903 @@ -723,6 +730,7 @@ static void wake_up_sem_queue_prepare(struct list_head *pt,
10906 list_add_tail(&q->list, pt);
10911 @@ -736,6 +744,7 @@ static void wake_up_sem_queue_prepare(struct list_head *pt,
10913 static void wake_up_sem_queue_do(struct list_head *pt)
10915 +#ifndef CONFIG_PREEMPT_RT_BASE
10916 struct sem_queue *q, *t;
10919 @@ -748,6 +757,7 @@ static void wake_up_sem_queue_do(struct list_head *pt)
10926 static void unlink_queue(struct sem_array *sma, struct sem_queue *q)
10927 diff --git a/kernel/Kconfig.locks b/kernel/Kconfig.locks
10928 index ebdb0043203a..b9e6aa7e5aa6 100644
10929 --- a/kernel/Kconfig.locks
10930 +++ b/kernel/Kconfig.locks
10931 @@ -225,11 +225,11 @@ config ARCH_SUPPORTS_ATOMIC_RMW
10933 config MUTEX_SPIN_ON_OWNER
10935 - depends on SMP && !DEBUG_MUTEXES && ARCH_SUPPORTS_ATOMIC_RMW
10936 + depends on SMP && !DEBUG_MUTEXES && ARCH_SUPPORTS_ATOMIC_RMW && !PREEMPT_RT_FULL
10938 config RWSEM_SPIN_ON_OWNER
10940 - depends on SMP && RWSEM_XCHGADD_ALGORITHM && ARCH_SUPPORTS_ATOMIC_RMW
10941 + depends on SMP && RWSEM_XCHGADD_ALGORITHM && ARCH_SUPPORTS_ATOMIC_RMW && !PREEMPT_RT_FULL
10943 config LOCK_SPIN_ON_OWNER
10945 diff --git a/kernel/Kconfig.preempt b/kernel/Kconfig.preempt
10946 index 3f9c97419f02..11dbe26a8279 100644
10947 --- a/kernel/Kconfig.preempt
10948 +++ b/kernel/Kconfig.preempt
10952 + select PREEMPT_COUNT
10954 +config PREEMPT_RT_BASE
10958 +config HAVE_PREEMPT_LAZY
10961 +config PREEMPT_LAZY
10962 + def_bool y if HAVE_PREEMPT_LAZY && PREEMPT_RT_FULL
10965 prompt "Preemption Model"
10966 @@ -33,9 +46,9 @@ config PREEMPT_VOLUNTARY
10968 Select this if you are building a kernel for a desktop system.
10971 +config PREEMPT__LL
10972 bool "Preemptible Kernel (Low-Latency Desktop)"
10973 - select PREEMPT_COUNT
10975 select UNINLINE_SPIN_UNLOCK if !ARCH_INLINE_SPIN_UNLOCK
10977 This option reduces the latency of the kernel by making
10978 @@ -52,6 +65,22 @@ config PREEMPT
10979 embedded system with latency requirements in the milliseconds
10982 +config PREEMPT_RTB
10983 + bool "Preemptible Kernel (Basic RT)"
10984 + select PREEMPT_RT_BASE
10986 + This option is basically the same as (Low-Latency Desktop) but
10987 + enables changes which are preliminary for the full preemptible
10990 +config PREEMPT_RT_FULL
10991 + bool "Fully Preemptible Kernel (RT)"
10992 + depends on IRQ_FORCED_THREADING
10993 + select PREEMPT_RT_BASE
10994 + select PREEMPT_RCU
10996 + All and everything
11000 config PREEMPT_COUNT
11001 diff --git a/kernel/cgroup.c b/kernel/cgroup.c
11002 index 4e2f3de0e40b..6401eb5fe140 100644
11003 --- a/kernel/cgroup.c
11004 +++ b/kernel/cgroup.c
11005 @@ -5040,10 +5040,10 @@ static void css_free_rcu_fn(struct rcu_head *rcu_head)
11006 queue_work(cgroup_destroy_wq, &css->destroy_work);
11009 -static void css_release_work_fn(struct work_struct *work)
11010 +static void css_release_work_fn(struct swork_event *sev)
11012 struct cgroup_subsys_state *css =
11013 - container_of(work, struct cgroup_subsys_state, destroy_work);
11014 + container_of(sev, struct cgroup_subsys_state, destroy_swork);
11015 struct cgroup_subsys *ss = css->ss;
11016 struct cgroup *cgrp = css->cgroup;
11018 @@ -5086,8 +5086,8 @@ static void css_release(struct percpu_ref *ref)
11019 struct cgroup_subsys_state *css =
11020 container_of(ref, struct cgroup_subsys_state, refcnt);
11022 - INIT_WORK(&css->destroy_work, css_release_work_fn);
11023 - queue_work(cgroup_destroy_wq, &css->destroy_work);
11024 + INIT_SWORK(&css->destroy_swork, css_release_work_fn);
11025 + swork_queue(&css->destroy_swork);
11028 static void init_and_link_css(struct cgroup_subsys_state *css,
11029 @@ -5739,6 +5739,7 @@ static int __init cgroup_wq_init(void)
11031 cgroup_destroy_wq = alloc_workqueue("cgroup_destroy", 0, 1);
11032 BUG_ON(!cgroup_destroy_wq);
11033 + BUG_ON(swork_get());
11036 * Used to destroy pidlists and separate to serve as flush domain.
11037 diff --git a/kernel/cpu.c b/kernel/cpu.c
11038 index 217fd2e7f435..69444f1bc924 100644
11041 @@ -239,6 +239,289 @@ static struct {
11042 #define cpuhp_lock_acquire() lock_map_acquire(&cpu_hotplug.dep_map)
11043 #define cpuhp_lock_release() lock_map_release(&cpu_hotplug.dep_map)
11046 + * hotplug_pcp - per cpu hotplug descriptor
11047 + * @unplug: set when pin_current_cpu() needs to sync tasks
11048 + * @sync_tsk: the task that waits for tasks to finish pinned sections
11049 + * @refcount: counter of tasks in pinned sections
11050 + * @grab_lock: set when the tasks entering pinned sections should wait
11051 + * @synced: notifier for @sync_tsk to tell cpu_down it's finished
11052 + * @mutex: the mutex to make tasks wait (used when @grab_lock is true)
11053 + * @mutex_init: zero if the mutex hasn't been initialized yet.
11055 + * Although @unplug and @sync_tsk may point to the same task, the @unplug
11056 + * is used as a flag and still exists after @sync_tsk has exited and
11057 + * @sync_tsk set to NULL.
11059 +struct hotplug_pcp {
11060 + struct task_struct *unplug;
11061 + struct task_struct *sync_tsk;
11064 + struct completion synced;
11065 + struct completion unplug_wait;
11066 +#ifdef CONFIG_PREEMPT_RT_FULL
11068 + * Note, on PREEMPT_RT, the hotplug lock must save the state of
11069 + * the task, otherwise the mutex will cause the task to fail
11070 + * to sleep when required. (Because it's called from migrate_disable())
11072 + * The spinlock_t on PREEMPT_RT is a mutex that saves the task's
11077 + struct mutex mutex;
11082 +#ifdef CONFIG_PREEMPT_RT_FULL
11083 +# define hotplug_lock(hp) rt_spin_lock__no_mg(&(hp)->lock)
11084 +# define hotplug_unlock(hp) rt_spin_unlock__no_mg(&(hp)->lock)
11086 +# define hotplug_lock(hp) mutex_lock(&(hp)->mutex)
11087 +# define hotplug_unlock(hp) mutex_unlock(&(hp)->mutex)
11090 +static DEFINE_PER_CPU(struct hotplug_pcp, hotplug_pcp);
11093 + * pin_current_cpu - Prevent the current cpu from being unplugged
11095 + * Lightweight version of get_online_cpus() to prevent cpu from being
11096 + * unplugged when code runs in a migration disabled region.
11098 + * Must be called with preemption disabled (preempt_count = 1)!
11100 +void pin_current_cpu(void)
11102 + struct hotplug_pcp *hp;
11106 + hp = this_cpu_ptr(&hotplug_pcp);
11108 + if (!hp->unplug || hp->refcount || force || preempt_count() > 1 ||
11109 + hp->unplug == current) {
11113 + if (hp->grab_lock) {
11114 + preempt_enable();
11115 + hotplug_lock(hp);
11116 + hotplug_unlock(hp);
11118 + preempt_enable();
11120 + * Try to push this task off of this CPU.
11122 + if (!migrate_me()) {
11123 + preempt_disable();
11124 + hp = this_cpu_ptr(&hotplug_pcp);
11125 + if (!hp->grab_lock) {
11127 + * Just let it continue it's already pinned
11128 + * or about to sleep.
11133 + preempt_enable();
11136 + preempt_disable();
11141 + * unpin_current_cpu - Allow unplug of current cpu
11143 + * Must be called with preemption or interrupts disabled!
11145 +void unpin_current_cpu(void)
11147 + struct hotplug_pcp *hp = this_cpu_ptr(&hotplug_pcp);
11149 + WARN_ON(hp->refcount <= 0);
11151 + /* This is safe. sync_unplug_thread is pinned to this cpu */
11152 + if (!--hp->refcount && hp->unplug && hp->unplug != current)
11153 + wake_up_process(hp->unplug);
11156 +static void wait_for_pinned_cpus(struct hotplug_pcp *hp)
11158 + set_current_state(TASK_UNINTERRUPTIBLE);
11159 + while (hp->refcount) {
11160 + schedule_preempt_disabled();
11161 + set_current_state(TASK_UNINTERRUPTIBLE);
11165 +static int sync_unplug_thread(void *data)
11167 + struct hotplug_pcp *hp = data;
11169 + wait_for_completion(&hp->unplug_wait);
11170 + preempt_disable();
11171 + hp->unplug = current;
11172 + wait_for_pinned_cpus(hp);
11175 + * This thread will synchronize the cpu_down() with threads
11176 + * that have pinned the CPU. When the pinned CPU count reaches
11177 + * zero, we inform the cpu_down code to continue to the next step.
11179 + set_current_state(TASK_UNINTERRUPTIBLE);
11180 + preempt_enable();
11181 + complete(&hp->synced);
11184 + * If all succeeds, the next step will need tasks to wait till
11185 + * the CPU is offline before continuing. To do this, the grab_lock
11186 + * is set and tasks going into pin_current_cpu() will block on the
11187 + * mutex. But we still need to wait for those that are already in
11188 + * pinned CPU sections. If the cpu_down() failed, the kthread_should_stop()
11189 + * will kick this thread out.
11191 + while (!hp->grab_lock && !kthread_should_stop()) {
11193 + set_current_state(TASK_UNINTERRUPTIBLE);
11196 + /* Make sure grab_lock is seen before we see a stale completion */
11200 + * Now just before cpu_down() enters stop machine, we need to make
11201 + * sure all tasks that are in pinned CPU sections are out, and new
11202 + * tasks will now grab the lock, keeping them from entering pinned
11205 + if (!kthread_should_stop()) {
11206 + preempt_disable();
11207 + wait_for_pinned_cpus(hp);
11208 + preempt_enable();
11209 + complete(&hp->synced);
11212 + set_current_state(TASK_UNINTERRUPTIBLE);
11213 + while (!kthread_should_stop()) {
11215 + set_current_state(TASK_UNINTERRUPTIBLE);
11217 + set_current_state(TASK_RUNNING);
11220 + * Force this thread off this CPU as it's going down and
11221 + * we don't want any more work on this CPU.
11223 + current->flags &= ~PF_NO_SETAFFINITY;
11224 + set_cpus_allowed_ptr(current, cpu_present_mask);
11229 +static void __cpu_unplug_sync(struct hotplug_pcp *hp)
11231 + wake_up_process(hp->sync_tsk);
11232 + wait_for_completion(&hp->synced);
11235 +static void __cpu_unplug_wait(unsigned int cpu)
11237 + struct hotplug_pcp *hp = &per_cpu(hotplug_pcp, cpu);
11239 + complete(&hp->unplug_wait);
11240 + wait_for_completion(&hp->synced);
11244 + * Start the sync_unplug_thread on the target cpu and wait for it to
11247 +static int cpu_unplug_begin(unsigned int cpu)
11249 + struct hotplug_pcp *hp = &per_cpu(hotplug_pcp, cpu);
11252 + /* Protected by cpu_hotplug.lock */
11253 + if (!hp->mutex_init) {
11254 +#ifdef CONFIG_PREEMPT_RT_FULL
11255 + spin_lock_init(&hp->lock);
11257 + mutex_init(&hp->mutex);
11259 + hp->mutex_init = 1;
11262 + /* Inform the scheduler to migrate tasks off this CPU */
11263 + tell_sched_cpu_down_begin(cpu);
11265 + init_completion(&hp->synced);
11266 + init_completion(&hp->unplug_wait);
11268 + hp->sync_tsk = kthread_create(sync_unplug_thread, hp, "sync_unplug/%d", cpu);
11269 + if (IS_ERR(hp->sync_tsk)) {
11270 + err = PTR_ERR(hp->sync_tsk);
11271 + hp->sync_tsk = NULL;
11274 + kthread_bind(hp->sync_tsk, cpu);
11277 + * Wait for tasks to get out of the pinned sections,
11278 + * it's still OK if new tasks enter. Some CPU notifiers will
11279 + * wait for tasks that are going to enter these sections and
11280 + * we must not have them block.
11282 + wake_up_process(hp->sync_tsk);
11286 +static void cpu_unplug_sync(unsigned int cpu)
11288 + struct hotplug_pcp *hp = &per_cpu(hotplug_pcp, cpu);
11290 + init_completion(&hp->synced);
11291 + /* The completion needs to be initialzied before setting grab_lock */
11294 + /* Grab the mutex before setting grab_lock */
11295 + hotplug_lock(hp);
11296 + hp->grab_lock = 1;
11299 + * The CPU notifiers have been completed.
11300 + * Wait for tasks to get out of pinned CPU sections and have new
11301 + * tasks block until the CPU is completely down.
11303 + __cpu_unplug_sync(hp);
11305 + /* All done with the sync thread */
11306 + kthread_stop(hp->sync_tsk);
11307 + hp->sync_tsk = NULL;
11310 +static void cpu_unplug_done(unsigned int cpu)
11312 + struct hotplug_pcp *hp = &per_cpu(hotplug_pcp, cpu);
11314 + hp->unplug = NULL;
11315 + /* Let all tasks know cpu unplug is finished before cleaning up */
11318 + if (hp->sync_tsk)
11319 + kthread_stop(hp->sync_tsk);
11321 + if (hp->grab_lock) {
11322 + hotplug_unlock(hp);
11323 + /* protected by cpu_hotplug.lock */
11324 + hp->grab_lock = 0;
11326 + tell_sched_cpu_down_done(cpu);
11329 void get_online_cpus(void)
11331 @@ -789,10 +1072,14 @@ static int takedown_cpu(unsigned int cpu)
11332 struct cpuhp_cpu_state *st = per_cpu_ptr(&cpuhp_state, cpu);
11335 + __cpu_unplug_wait(cpu);
11336 /* Park the smpboot threads */
11337 kthread_park(per_cpu_ptr(&cpuhp_state, cpu)->thread);
11338 smpboot_park_threads(cpu);
11340 + /* Notifiers are done. Don't let any more tasks pin this CPU. */
11341 + cpu_unplug_sync(cpu);
11344 * Prevent irq alloc/free while the dying cpu reorganizes the
11345 * interrupt affinities.
11346 @@ -877,6 +1164,9 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen,
11347 struct cpuhp_cpu_state *st = per_cpu_ptr(&cpuhp_state, cpu);
11348 int prev_state, ret = 0;
11349 bool hasdied = false;
11351 + cpumask_var_t cpumask;
11352 + cpumask_var_t cpumask_org;
11354 if (num_online_cpus() == 1)
11356 @@ -884,7 +1174,34 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen,
11357 if (!cpu_present(cpu))
11360 + /* Move the downtaker off the unplug cpu */
11361 + if (!alloc_cpumask_var(&cpumask, GFP_KERNEL))
11363 + if (!alloc_cpumask_var(&cpumask_org, GFP_KERNEL)) {
11364 + free_cpumask_var(cpumask);
11368 + cpumask_copy(cpumask_org, tsk_cpus_allowed(current));
11369 + cpumask_andnot(cpumask, cpu_online_mask, cpumask_of(cpu));
11370 + set_cpus_allowed_ptr(current, cpumask);
11371 + free_cpumask_var(cpumask);
11372 + migrate_disable();
11373 + mycpu = smp_processor_id();
11374 + if (mycpu == cpu) {
11375 + printk(KERN_ERR "Yuck! Still on unplug CPU\n!");
11376 + migrate_enable();
11378 + goto restore_cpus;
11381 + migrate_enable();
11382 cpu_hotplug_begin();
11383 + ret = cpu_unplug_begin(cpu);
11385 + printk("cpu_unplug_begin(%d) failed\n", cpu);
11389 cpuhp_tasks_frozen = tasks_frozen;
11391 @@ -923,10 +1240,15 @@ static int __ref _cpu_down(unsigned int cpu, int tasks_frozen,
11393 hasdied = prev_state != st->state && st->state == CPUHP_OFFLINE;
11395 + cpu_unplug_done(cpu);
11397 cpu_hotplug_done();
11398 /* This post dead nonsense must die */
11399 if (!ret && hasdied)
11400 cpu_notify_nofail(CPU_POST_DEAD, cpu);
11402 + set_cpus_allowed_ptr(current, cpumask_org);
11403 + free_cpumask_var(cpumask_org);
11407 diff --git a/kernel/cpuset.c b/kernel/cpuset.c
11408 index 29f815d2ef7e..341b17f24f95 100644
11409 --- a/kernel/cpuset.c
11410 +++ b/kernel/cpuset.c
11411 @@ -284,7 +284,7 @@ static struct cpuset top_cpuset = {
11414 static DEFINE_MUTEX(cpuset_mutex);
11415 -static DEFINE_SPINLOCK(callback_lock);
11416 +static DEFINE_RAW_SPINLOCK(callback_lock);
11418 static struct workqueue_struct *cpuset_migrate_mm_wq;
11420 @@ -907,9 +907,9 @@ static void update_cpumasks_hier(struct cpuset *cs, struct cpumask *new_cpus)
11424 - spin_lock_irq(&callback_lock);
11425 + raw_spin_lock_irq(&callback_lock);
11426 cpumask_copy(cp->effective_cpus, new_cpus);
11427 - spin_unlock_irq(&callback_lock);
11428 + raw_spin_unlock_irq(&callback_lock);
11430 WARN_ON(!cgroup_subsys_on_dfl(cpuset_cgrp_subsys) &&
11431 !cpumask_equal(cp->cpus_allowed, cp->effective_cpus));
11432 @@ -974,9 +974,9 @@ static int update_cpumask(struct cpuset *cs, struct cpuset *trialcs,
11436 - spin_lock_irq(&callback_lock);
11437 + raw_spin_lock_irq(&callback_lock);
11438 cpumask_copy(cs->cpus_allowed, trialcs->cpus_allowed);
11439 - spin_unlock_irq(&callback_lock);
11440 + raw_spin_unlock_irq(&callback_lock);
11442 /* use trialcs->cpus_allowed as a temp variable */
11443 update_cpumasks_hier(cs, trialcs->cpus_allowed);
11444 @@ -1176,9 +1176,9 @@ static void update_nodemasks_hier(struct cpuset *cs, nodemask_t *new_mems)
11448 - spin_lock_irq(&callback_lock);
11449 + raw_spin_lock_irq(&callback_lock);
11450 cp->effective_mems = *new_mems;
11451 - spin_unlock_irq(&callback_lock);
11452 + raw_spin_unlock_irq(&callback_lock);
11454 WARN_ON(!cgroup_subsys_on_dfl(cpuset_cgrp_subsys) &&
11455 !nodes_equal(cp->mems_allowed, cp->effective_mems));
11456 @@ -1246,9 +1246,9 @@ static int update_nodemask(struct cpuset *cs, struct cpuset *trialcs,
11460 - spin_lock_irq(&callback_lock);
11461 + raw_spin_lock_irq(&callback_lock);
11462 cs->mems_allowed = trialcs->mems_allowed;
11463 - spin_unlock_irq(&callback_lock);
11464 + raw_spin_unlock_irq(&callback_lock);
11466 /* use trialcs->mems_allowed as a temp variable */
11467 update_nodemasks_hier(cs, &trialcs->mems_allowed);
11468 @@ -1339,9 +1339,9 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs,
11469 spread_flag_changed = ((is_spread_slab(cs) != is_spread_slab(trialcs))
11470 || (is_spread_page(cs) != is_spread_page(trialcs)));
11472 - spin_lock_irq(&callback_lock);
11473 + raw_spin_lock_irq(&callback_lock);
11474 cs->flags = trialcs->flags;
11475 - spin_unlock_irq(&callback_lock);
11476 + raw_spin_unlock_irq(&callback_lock);
11478 if (!cpumask_empty(trialcs->cpus_allowed) && balance_flag_changed)
11479 rebuild_sched_domains_locked();
11480 @@ -1756,7 +1756,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v)
11481 cpuset_filetype_t type = seq_cft(sf)->private;
11484 - spin_lock_irq(&callback_lock);
11485 + raw_spin_lock_irq(&callback_lock);
11489 @@ -1775,7 +1775,7 @@ static int cpuset_common_seq_show(struct seq_file *sf, void *v)
11493 - spin_unlock_irq(&callback_lock);
11494 + raw_spin_unlock_irq(&callback_lock);
11498 @@ -1989,12 +1989,12 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
11502 - spin_lock_irq(&callback_lock);
11503 + raw_spin_lock_irq(&callback_lock);
11504 if (cgroup_subsys_on_dfl(cpuset_cgrp_subsys)) {
11505 cpumask_copy(cs->effective_cpus, parent->effective_cpus);
11506 cs->effective_mems = parent->effective_mems;
11508 - spin_unlock_irq(&callback_lock);
11509 + raw_spin_unlock_irq(&callback_lock);
11511 if (!test_bit(CGRP_CPUSET_CLONE_CHILDREN, &css->cgroup->flags))
11513 @@ -2021,12 +2021,12 @@ static int cpuset_css_online(struct cgroup_subsys_state *css)
11517 - spin_lock_irq(&callback_lock);
11518 + raw_spin_lock_irq(&callback_lock);
11519 cs->mems_allowed = parent->mems_allowed;
11520 cs->effective_mems = parent->mems_allowed;
11521 cpumask_copy(cs->cpus_allowed, parent->cpus_allowed);
11522 cpumask_copy(cs->effective_cpus, parent->cpus_allowed);
11523 - spin_unlock_irq(&callback_lock);
11524 + raw_spin_unlock_irq(&callback_lock);
11526 mutex_unlock(&cpuset_mutex);
11528 @@ -2065,7 +2065,7 @@ static void cpuset_css_free(struct cgroup_subsys_state *css)
11529 static void cpuset_bind(struct cgroup_subsys_state *root_css)
11531 mutex_lock(&cpuset_mutex);
11532 - spin_lock_irq(&callback_lock);
11533 + raw_spin_lock_irq(&callback_lock);
11535 if (cgroup_subsys_on_dfl(cpuset_cgrp_subsys)) {
11536 cpumask_copy(top_cpuset.cpus_allowed, cpu_possible_mask);
11537 @@ -2076,7 +2076,7 @@ static void cpuset_bind(struct cgroup_subsys_state *root_css)
11538 top_cpuset.mems_allowed = top_cpuset.effective_mems;
11541 - spin_unlock_irq(&callback_lock);
11542 + raw_spin_unlock_irq(&callback_lock);
11543 mutex_unlock(&cpuset_mutex);
11546 @@ -2177,12 +2177,12 @@ hotplug_update_tasks_legacy(struct cpuset *cs,
11550 - spin_lock_irq(&callback_lock);
11551 + raw_spin_lock_irq(&callback_lock);
11552 cpumask_copy(cs->cpus_allowed, new_cpus);
11553 cpumask_copy(cs->effective_cpus, new_cpus);
11554 cs->mems_allowed = *new_mems;
11555 cs->effective_mems = *new_mems;
11556 - spin_unlock_irq(&callback_lock);
11557 + raw_spin_unlock_irq(&callback_lock);
11560 * Don't call update_tasks_cpumask() if the cpuset becomes empty,
11561 @@ -2219,10 +2219,10 @@ hotplug_update_tasks(struct cpuset *cs,
11562 if (nodes_empty(*new_mems))
11563 *new_mems = parent_cs(cs)->effective_mems;
11565 - spin_lock_irq(&callback_lock);
11566 + raw_spin_lock_irq(&callback_lock);
11567 cpumask_copy(cs->effective_cpus, new_cpus);
11568 cs->effective_mems = *new_mems;
11569 - spin_unlock_irq(&callback_lock);
11570 + raw_spin_unlock_irq(&callback_lock);
11573 update_tasks_cpumask(cs);
11574 @@ -2308,21 +2308,21 @@ static void cpuset_hotplug_workfn(struct work_struct *work)
11576 /* synchronize cpus_allowed to cpu_active_mask */
11577 if (cpus_updated) {
11578 - spin_lock_irq(&callback_lock);
11579 + raw_spin_lock_irq(&callback_lock);
11581 cpumask_copy(top_cpuset.cpus_allowed, &new_cpus);
11582 cpumask_copy(top_cpuset.effective_cpus, &new_cpus);
11583 - spin_unlock_irq(&callback_lock);
11584 + raw_spin_unlock_irq(&callback_lock);
11585 /* we don't mess with cpumasks of tasks in top_cpuset */
11588 /* synchronize mems_allowed to N_MEMORY */
11589 if (mems_updated) {
11590 - spin_lock_irq(&callback_lock);
11591 + raw_spin_lock_irq(&callback_lock);
11593 top_cpuset.mems_allowed = new_mems;
11594 top_cpuset.effective_mems = new_mems;
11595 - spin_unlock_irq(&callback_lock);
11596 + raw_spin_unlock_irq(&callback_lock);
11597 update_tasks_nodemask(&top_cpuset);
11600 @@ -2420,11 +2420,11 @@ void cpuset_cpus_allowed(struct task_struct *tsk, struct cpumask *pmask)
11602 unsigned long flags;
11604 - spin_lock_irqsave(&callback_lock, flags);
11605 + raw_spin_lock_irqsave(&callback_lock, flags);
11607 guarantee_online_cpus(task_cs(tsk), pmask);
11609 - spin_unlock_irqrestore(&callback_lock, flags);
11610 + raw_spin_unlock_irqrestore(&callback_lock, flags);
11613 void cpuset_cpus_allowed_fallback(struct task_struct *tsk)
11614 @@ -2472,11 +2472,11 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk)
11616 unsigned long flags;
11618 - spin_lock_irqsave(&callback_lock, flags);
11619 + raw_spin_lock_irqsave(&callback_lock, flags);
11621 guarantee_online_mems(task_cs(tsk), &mask);
11623 - spin_unlock_irqrestore(&callback_lock, flags);
11624 + raw_spin_unlock_irqrestore(&callback_lock, flags);
11628 @@ -2568,14 +2568,14 @@ bool __cpuset_node_allowed(int node, gfp_t gfp_mask)
11631 /* Not hardwall and node outside mems_allowed: scan up cpusets */
11632 - spin_lock_irqsave(&callback_lock, flags);
11633 + raw_spin_lock_irqsave(&callback_lock, flags);
11636 cs = nearest_hardwall_ancestor(task_cs(current));
11637 allowed = node_isset(node, cs->mems_allowed);
11640 - spin_unlock_irqrestore(&callback_lock, flags);
11641 + raw_spin_unlock_irqrestore(&callback_lock, flags);
11645 diff --git a/kernel/debug/kdb/kdb_io.c b/kernel/debug/kdb/kdb_io.c
11646 index fc1ef736253c..83c666537a7a 100644
11647 --- a/kernel/debug/kdb/kdb_io.c
11648 +++ b/kernel/debug/kdb/kdb_io.c
11649 @@ -554,7 +554,6 @@ int vkdb_printf(enum kdb_msgsrc src, const char *fmt, va_list ap)
11652 int logging, saved_loglevel = 0;
11653 - int saved_trap_printk;
11654 int got_printf_lock = 0;
11657 @@ -565,8 +564,6 @@ int vkdb_printf(enum kdb_msgsrc src, const char *fmt, va_list ap)
11658 unsigned long uninitialized_var(flags);
11661 - saved_trap_printk = kdb_trap_printk;
11662 - kdb_trap_printk = 0;
11664 /* Serialize kdb_printf if multiple cpus try to write at once.
11665 * But if any cpu goes recursive in kdb, just print the output,
11666 @@ -855,7 +852,6 @@ int vkdb_printf(enum kdb_msgsrc src, const char *fmt, va_list ap)
11668 __release(kdb_printf_lock);
11670 - kdb_trap_printk = saved_trap_printk;
11674 @@ -865,9 +861,11 @@ int kdb_printf(const char *fmt, ...)
11678 + kdb_trap_printk++;
11680 r = vkdb_printf(KDB_MSGSRC_INTERNAL, fmt, ap);
11682 + kdb_trap_printk--;
11686 diff --git a/kernel/events/core.c b/kernel/events/core.c
11687 index 4b3323151a2f..e89a1a4d59cd 100644
11688 --- a/kernel/events/core.c
11689 +++ b/kernel/events/core.c
11690 @@ -1050,6 +1050,7 @@ static void __perf_mux_hrtimer_init(struct perf_cpu_context *cpuctx, int cpu)
11691 raw_spin_lock_init(&cpuctx->hrtimer_lock);
11692 hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS_PINNED);
11693 timer->function = perf_mux_hrtimer_handler;
11694 + timer->irqsafe = 1;
11697 static int perf_mux_hrtimer_restart(struct perf_cpu_context *cpuctx)
11698 @@ -8363,6 +8364,7 @@ static void perf_swevent_init_hrtimer(struct perf_event *event)
11700 hrtimer_init(&hwc->hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
11701 hwc->hrtimer.function = perf_swevent_hrtimer;
11702 + hwc->hrtimer.irqsafe = 1;
11705 * Since hrtimers have a fixed rate, we can do a static freq->period
11706 diff --git a/kernel/exit.c b/kernel/exit.c
11707 index 3076f3089919..fb2ebcf3ca7c 100644
11708 --- a/kernel/exit.c
11709 +++ b/kernel/exit.c
11710 @@ -143,7 +143,7 @@ static void __exit_signal(struct task_struct *tsk)
11711 * Do this under ->siglock, we can race with another thread
11712 * doing sigqueue_free() if we have SIGQUEUE_PREALLOC signals.
11714 - flush_sigqueue(&tsk->pending);
11715 + flush_task_sigqueue(tsk);
11716 tsk->sighand = NULL;
11717 spin_unlock(&sighand->siglock);
11719 diff --git a/kernel/fork.c b/kernel/fork.c
11720 index ba8a01564985..47784f8aed37 100644
11721 --- a/kernel/fork.c
11722 +++ b/kernel/fork.c
11724 #include <linux/compiler.h>
11725 #include <linux/sysctl.h>
11726 #include <linux/kcov.h>
11727 +#include <linux/kprobes.h>
11729 #include <asm/pgtable.h>
11730 #include <asm/pgalloc.h>
11731 @@ -376,13 +377,24 @@ static inline void put_signal_struct(struct signal_struct *sig)
11732 if (atomic_dec_and_test(&sig->sigcnt))
11733 free_signal_struct(sig);
11736 +#ifdef CONFIG_PREEMPT_RT_BASE
11739 void __put_task_struct(struct task_struct *tsk)
11741 WARN_ON(!tsk->exit_state);
11742 WARN_ON(atomic_read(&tsk->usage));
11743 WARN_ON(tsk == current);
11746 + * Remove function-return probe instances associated with this
11747 + * task and put them back on the free list.
11749 + kprobe_flush_task(tsk);
11751 + /* Task is done with its stack. */
11752 + put_task_stack(tsk);
11755 task_numa_free(tsk);
11756 security_task_free(tsk);
11757 @@ -393,7 +405,18 @@ void __put_task_struct(struct task_struct *tsk)
11758 if (!profile_handoff_task(tsk))
11761 +#ifndef CONFIG_PREEMPT_RT_BASE
11762 EXPORT_SYMBOL_GPL(__put_task_struct);
11764 +void __put_task_struct_cb(struct rcu_head *rhp)
11766 + struct task_struct *tsk = container_of(rhp, struct task_struct, put_rcu);
11768 + __put_task_struct(tsk);
11771 +EXPORT_SYMBOL_GPL(__put_task_struct_cb);
11774 void __init __weak arch_task_cache_init(void) { }
11776 @@ -852,6 +875,19 @@ void __mmdrop(struct mm_struct *mm)
11778 EXPORT_SYMBOL_GPL(__mmdrop);
11780 +#ifdef CONFIG_PREEMPT_RT_BASE
11782 + * RCU callback for delayed mm drop. Not strictly rcu, but we don't
11783 + * want another facility to make this work.
11785 +void __mmdrop_delayed(struct rcu_head *rhp)
11787 + struct mm_struct *mm = container_of(rhp, struct mm_struct, delayed_drop);
11793 static inline void __mmput(struct mm_struct *mm)
11795 VM_BUG_ON(atomic_read(&mm->mm_users));
11796 @@ -1426,6 +1462,9 @@ static void rt_mutex_init_task(struct task_struct *p)
11798 static void posix_cpu_timers_init(struct task_struct *tsk)
11800 +#ifdef CONFIG_PREEMPT_RT_BASE
11801 + tsk->posix_timer_list = NULL;
11803 tsk->cputime_expires.prof_exp = 0;
11804 tsk->cputime_expires.virt_exp = 0;
11805 tsk->cputime_expires.sched_exp = 0;
11806 @@ -1552,6 +1591,7 @@ static __latent_entropy struct task_struct *copy_process(
11807 spin_lock_init(&p->alloc_lock);
11809 init_sigpending(&p->pending);
11810 + p->sigqueue_cache = NULL;
11812 p->utime = p->stime = p->gtime = 0;
11813 p->utimescaled = p->stimescaled = 0;
11814 diff --git a/kernel/futex.c b/kernel/futex.c
11815 index 38b68c2735c5..6450a8d81667 100644
11816 --- a/kernel/futex.c
11817 +++ b/kernel/futex.c
11818 @@ -904,7 +904,9 @@ void exit_pi_state_list(struct task_struct *curr)
11819 * task still owns the PI-state:
11821 if (head->next != next) {
11822 + raw_spin_unlock_irq(&curr->pi_lock);
11823 spin_unlock(&hb->lock);
11824 + raw_spin_lock_irq(&curr->pi_lock);
11828 @@ -1299,6 +1301,7 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this,
11829 struct futex_pi_state *pi_state = this->pi_state;
11830 u32 uninitialized_var(curval), newval;
11832 + WAKE_Q(wake_sleeper_q);
11836 @@ -1365,7 +1368,8 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this,
11838 raw_spin_unlock_irq(&pi_state->pi_mutex.wait_lock);
11840 - deboost = rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q);
11841 + deboost = rt_mutex_futex_unlock(&pi_state->pi_mutex, &wake_q,
11842 + &wake_sleeper_q);
11845 * First unlock HB so the waiter does not spin on it once he got woken
11846 @@ -1373,8 +1377,9 @@ static int wake_futex_pi(u32 __user *uaddr, u32 uval, struct futex_q *this,
11847 * deboost first (and lose our higher priority), then the task might get
11848 * scheduled away before the wake up can take place.
11850 - spin_unlock(&hb->lock);
11851 + deboost |= spin_unlock_no_deboost(&hb->lock);
11852 wake_up_q(&wake_q);
11853 + wake_up_q_sleeper(&wake_sleeper_q);
11855 rt_mutex_adjust_prio(current);
11857 @@ -1924,6 +1929,16 @@ static int futex_requeue(u32 __user *uaddr1, unsigned int flags,
11858 requeue_pi_wake_futex(this, &key2, hb2);
11861 + } else if (ret == -EAGAIN) {
11863 + * Waiter was woken by timeout or
11864 + * signal and has set pi_blocked_on to
11865 + * PI_WAKEUP_INPROGRESS before we
11866 + * tried to enqueue it on the rtmutex.
11868 + this->pi_state = NULL;
11869 + put_pi_state(pi_state);
11873 * rt_mutex_start_proxy_lock() detected a
11874 @@ -2814,7 +2829,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
11875 struct hrtimer_sleeper timeout, *to = NULL;
11876 struct rt_mutex_waiter rt_waiter;
11877 struct rt_mutex *pi_mutex = NULL;
11878 - struct futex_hash_bucket *hb;
11879 + struct futex_hash_bucket *hb, *hb2;
11880 union futex_key key2 = FUTEX_KEY_INIT;
11881 struct futex_q q = futex_q_init;
11883 @@ -2839,10 +2854,7 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
11884 * The waiter is allocated on our stack, manipulated by the requeue
11885 * code while we sleep on uaddr.
11887 - debug_rt_mutex_init_waiter(&rt_waiter);
11888 - RB_CLEAR_NODE(&rt_waiter.pi_tree_entry);
11889 - RB_CLEAR_NODE(&rt_waiter.tree_entry);
11890 - rt_waiter.task = NULL;
11891 + rt_mutex_init_waiter(&rt_waiter, false);
11893 ret = get_futex_key(uaddr2, flags & FLAGS_SHARED, &key2, VERIFY_WRITE);
11894 if (unlikely(ret != 0))
11895 @@ -2873,20 +2885,55 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
11896 /* Queue the futex_q, drop the hb lock, wait for wakeup. */
11897 futex_wait_queue_me(hb, &q, to);
11899 - spin_lock(&hb->lock);
11900 - ret = handle_early_requeue_pi_wakeup(hb, &q, &key2, to);
11901 - spin_unlock(&hb->lock);
11903 - goto out_put_keys;
11905 + * On RT we must avoid races with requeue and trying to block
11906 + * on two mutexes (hb->lock and uaddr2's rtmutex) by
11907 + * serializing access to pi_blocked_on with pi_lock.
11909 + raw_spin_lock_irq(¤t->pi_lock);
11910 + if (current->pi_blocked_on) {
11912 + * We have been requeued or are in the process of
11913 + * being requeued.
11915 + raw_spin_unlock_irq(¤t->pi_lock);
11918 + * Setting pi_blocked_on to PI_WAKEUP_INPROGRESS
11919 + * prevents a concurrent requeue from moving us to the
11920 + * uaddr2 rtmutex. After that we can safely acquire
11921 + * (and possibly block on) hb->lock.
11923 + current->pi_blocked_on = PI_WAKEUP_INPROGRESS;
11924 + raw_spin_unlock_irq(¤t->pi_lock);
11926 + spin_lock(&hb->lock);
11929 + * Clean up pi_blocked_on. We might leak it otherwise
11930 + * when we succeeded with the hb->lock in the fast
11933 + raw_spin_lock_irq(¤t->pi_lock);
11934 + current->pi_blocked_on = NULL;
11935 + raw_spin_unlock_irq(¤t->pi_lock);
11937 + ret = handle_early_requeue_pi_wakeup(hb, &q, &key2, to);
11938 + spin_unlock(&hb->lock);
11940 + goto out_put_keys;
11944 - * In order for us to be here, we know our q.key == key2, and since
11945 - * we took the hb->lock above, we also know that futex_requeue() has
11946 - * completed and we no longer have to concern ourselves with a wakeup
11947 - * race with the atomic proxy lock acquisition by the requeue code. The
11948 - * futex_requeue dropped our key1 reference and incremented our key2
11949 - * reference count.
11950 + * In order to be here, we have either been requeued, are in
11951 + * the process of being requeued, or requeue successfully
11952 + * acquired uaddr2 on our behalf. If pi_blocked_on was
11953 + * non-null above, we may be racing with a requeue. Do not
11954 + * rely on q->lock_ptr to be hb2->lock until after blocking on
11955 + * hb->lock or hb2->lock. The futex_requeue dropped our key1
11956 + * reference and incremented our key2 reference count.
11958 + hb2 = hash_futex(&key2);
11960 /* Check if the requeue code acquired the second futex for us. */
11961 if (!q.rt_waiter) {
11962 @@ -2895,14 +2942,15 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
11963 * did a lock-steal - fix up the PI-state in that case.
11965 if (q.pi_state && (q.pi_state->owner != current)) {
11966 - spin_lock(q.lock_ptr);
11967 + spin_lock(&hb2->lock);
11968 + BUG_ON(&hb2->lock != q.lock_ptr);
11969 ret = fixup_pi_state_owner(uaddr2, &q, current);
11971 * Drop the reference to the pi state which
11972 * the requeue_pi() code acquired for us.
11974 put_pi_state(q.pi_state);
11975 - spin_unlock(q.lock_ptr);
11976 + spin_unlock(&hb2->lock);
11980 @@ -2915,7 +2963,8 @@ static int futex_wait_requeue_pi(u32 __user *uaddr, unsigned int flags,
11981 ret = rt_mutex_finish_proxy_lock(pi_mutex, to, &rt_waiter);
11982 debug_rt_mutex_free_waiter(&rt_waiter);
11984 - spin_lock(q.lock_ptr);
11985 + spin_lock(&hb2->lock);
11986 + BUG_ON(&hb2->lock != q.lock_ptr);
11988 * Fixup the pi_state owner and possibly acquire the lock if we
11990 diff --git a/kernel/irq/handle.c b/kernel/irq/handle.c
11991 index d3f24905852c..f87aa8fdcc51 100644
11992 --- a/kernel/irq/handle.c
11993 +++ b/kernel/irq/handle.c
11994 @@ -181,10 +181,16 @@ irqreturn_t handle_irq_event_percpu(struct irq_desc *desc)
11996 irqreturn_t retval;
11997 unsigned int flags = 0;
11998 + struct pt_regs *regs = get_irq_regs();
11999 + u64 ip = regs ? instruction_pointer(regs) : 0;
12001 retval = __handle_irq_event_percpu(desc, &flags);
12003 - add_interrupt_randomness(desc->irq_data.irq, flags);
12004 +#ifdef CONFIG_PREEMPT_RT_FULL
12005 + desc->random_ip = ip;
12007 + add_interrupt_randomness(desc->irq_data.irq, flags, ip);
12011 note_interrupt(desc, retval);
12012 diff --git a/kernel/irq/manage.c b/kernel/irq/manage.c
12013 index 6b669593e7eb..e357bf6c59d5 100644
12014 --- a/kernel/irq/manage.c
12015 +++ b/kernel/irq/manage.c
12017 #include "internals.h"
12019 #ifdef CONFIG_IRQ_FORCED_THREADING
12020 +# ifndef CONFIG_PREEMPT_RT_BASE
12021 __read_mostly bool force_irqthreads;
12023 static int __init setup_forced_irqthreads(char *arg)
12024 @@ -30,6 +31,7 @@ static int __init setup_forced_irqthreads(char *arg)
12027 early_param("threadirqs", setup_forced_irqthreads);
12031 static void __synchronize_hardirq(struct irq_desc *desc)
12032 @@ -233,7 +235,12 @@ int irq_set_affinity_locked(struct irq_data *data, const struct cpumask *mask,
12034 if (desc->affinity_notify) {
12035 kref_get(&desc->affinity_notify->kref);
12037 +#ifdef CONFIG_PREEMPT_RT_BASE
12038 + swork_queue(&desc->affinity_notify->swork);
12040 schedule_work(&desc->affinity_notify->work);
12043 irqd_set(data, IRQD_AFFINITY_SET);
12045 @@ -271,10 +278,8 @@ int irq_set_affinity_hint(unsigned int irq, const struct cpumask *m)
12047 EXPORT_SYMBOL_GPL(irq_set_affinity_hint);
12049 -static void irq_affinity_notify(struct work_struct *work)
12050 +static void _irq_affinity_notify(struct irq_affinity_notify *notify)
12052 - struct irq_affinity_notify *notify =
12053 - container_of(work, struct irq_affinity_notify, work);
12054 struct irq_desc *desc = irq_to_desc(notify->irq);
12055 cpumask_var_t cpumask;
12056 unsigned long flags;
12057 @@ -296,6 +301,35 @@ static void irq_affinity_notify(struct work_struct *work)
12058 kref_put(¬ify->kref, notify->release);
12061 +#ifdef CONFIG_PREEMPT_RT_BASE
12062 +static void init_helper_thread(void)
12064 + static int init_sworker_once;
12066 + if (init_sworker_once)
12068 + if (WARN_ON(swork_get()))
12070 + init_sworker_once = 1;
12073 +static void irq_affinity_notify(struct swork_event *swork)
12075 + struct irq_affinity_notify *notify =
12076 + container_of(swork, struct irq_affinity_notify, swork);
12077 + _irq_affinity_notify(notify);
12082 +static void irq_affinity_notify(struct work_struct *work)
12084 + struct irq_affinity_notify *notify =
12085 + container_of(work, struct irq_affinity_notify, work);
12086 + _irq_affinity_notify(notify);
12091 * irq_set_affinity_notifier - control notification of IRQ affinity changes
12092 * @irq: Interrupt for which to enable/disable notification
12093 @@ -324,7 +358,12 @@ irq_set_affinity_notifier(unsigned int irq, struct irq_affinity_notify *notify)
12096 kref_init(¬ify->kref);
12097 +#ifdef CONFIG_PREEMPT_RT_BASE
12098 + INIT_SWORK(¬ify->swork, irq_affinity_notify);
12099 + init_helper_thread();
12101 INIT_WORK(¬ify->work, irq_affinity_notify);
12105 raw_spin_lock_irqsave(&desc->lock, flags);
12106 @@ -879,7 +918,15 @@ irq_forced_thread_fn(struct irq_desc *desc, struct irqaction *action)
12107 local_bh_disable();
12108 ret = action->thread_fn(action->irq, action->dev_id);
12109 irq_finalize_oneshot(desc, action);
12110 - local_bh_enable();
12112 + * Interrupts which have real time requirements can be set up
12113 + * to avoid softirq processing in the thread handler. This is
12114 + * safe as these interrupts do not raise soft interrupts.
12116 + if (irq_settings_no_softirq_call(desc))
12117 + _local_bh_enable();
12119 + local_bh_enable();
12123 @@ -976,6 +1023,12 @@ static int irq_thread(void *data)
12124 if (action_ret == IRQ_WAKE_THREAD)
12125 irq_wake_secondary(desc, action);
12127 +#ifdef CONFIG_PREEMPT_RT_FULL
12128 + migrate_disable();
12129 + add_interrupt_randomness(action->irq, 0,
12130 + desc->random_ip ^ (unsigned long) action);
12131 + migrate_enable();
12133 wake_threads_waitq(desc);
12136 @@ -1336,6 +1389,9 @@ __setup_irq(unsigned int irq, struct irq_desc *desc, struct irqaction *new)
12137 irqd_set(&desc->irq_data, IRQD_NO_BALANCING);
12140 + if (new->flags & IRQF_NO_SOFTIRQ_CALL)
12141 + irq_settings_set_no_softirq_call(desc);
12143 /* Set default affinity mask once everything is setup */
12144 setup_affinity(desc, mask);
12146 @@ -2061,7 +2117,7 @@ EXPORT_SYMBOL_GPL(irq_get_irqchip_state);
12147 * This call sets the internal irqchip state of an interrupt,
12148 * depending on the value of @which.
12150 - * This function should be called with preemption disabled if the
12151 + * This function should be called with migration disabled if the
12152 * interrupt controller has per-cpu registers.
12154 int irq_set_irqchip_state(unsigned int irq, enum irqchip_irq_state which,
12155 diff --git a/kernel/irq/settings.h b/kernel/irq/settings.h
12156 index 320579d89091..2df2d4445b1e 100644
12157 --- a/kernel/irq/settings.h
12158 +++ b/kernel/irq/settings.h
12159 @@ -16,6 +16,7 @@ enum {
12160 _IRQ_PER_CPU_DEVID = IRQ_PER_CPU_DEVID,
12161 _IRQ_IS_POLLED = IRQ_IS_POLLED,
12162 _IRQ_DISABLE_UNLAZY = IRQ_DISABLE_UNLAZY,
12163 + _IRQ_NO_SOFTIRQ_CALL = IRQ_NO_SOFTIRQ_CALL,
12164 _IRQF_MODIFY_MASK = IRQF_MODIFY_MASK,
12167 @@ -30,6 +31,7 @@ enum {
12168 #define IRQ_PER_CPU_DEVID GOT_YOU_MORON
12169 #define IRQ_IS_POLLED GOT_YOU_MORON
12170 #define IRQ_DISABLE_UNLAZY GOT_YOU_MORON
12171 +#define IRQ_NO_SOFTIRQ_CALL GOT_YOU_MORON
12172 #undef IRQF_MODIFY_MASK
12173 #define IRQF_MODIFY_MASK GOT_YOU_MORON
12175 @@ -40,6 +42,16 @@ irq_settings_clr_and_set(struct irq_desc *desc, u32 clr, u32 set)
12176 desc->status_use_accessors |= (set & _IRQF_MODIFY_MASK);
12179 +static inline bool irq_settings_no_softirq_call(struct irq_desc *desc)
12181 + return desc->status_use_accessors & _IRQ_NO_SOFTIRQ_CALL;
12184 +static inline void irq_settings_set_no_softirq_call(struct irq_desc *desc)
12186 + desc->status_use_accessors |= _IRQ_NO_SOFTIRQ_CALL;
12189 static inline bool irq_settings_is_per_cpu(struct irq_desc *desc)
12191 return desc->status_use_accessors & _IRQ_PER_CPU;
12192 diff --git a/kernel/irq/spurious.c b/kernel/irq/spurious.c
12193 index 5707f97a3e6a..73f38dc7a7fb 100644
12194 --- a/kernel/irq/spurious.c
12195 +++ b/kernel/irq/spurious.c
12196 @@ -442,6 +442,10 @@ MODULE_PARM_DESC(noirqdebug, "Disable irq lockup detection when true");
12198 static int __init irqfixup_setup(char *str)
12200 +#ifdef CONFIG_PREEMPT_RT_BASE
12201 + pr_warn("irqfixup boot option not supported w/ CONFIG_PREEMPT_RT_BASE\n");
12205 printk(KERN_WARNING "Misrouted IRQ fixup support enabled.\n");
12206 printk(KERN_WARNING "This may impact system performance.\n");
12207 @@ -454,6 +458,10 @@ module_param(irqfixup, int, 0644);
12209 static int __init irqpoll_setup(char *str)
12211 +#ifdef CONFIG_PREEMPT_RT_BASE
12212 + pr_warn("irqpoll boot option not supported w/ CONFIG_PREEMPT_RT_BASE\n");
12216 printk(KERN_WARNING "Misrouted IRQ fixup and polling support "
12218 diff --git a/kernel/irq_work.c b/kernel/irq_work.c
12219 index bcf107ce0854..2899ba0d23d1 100644
12220 --- a/kernel/irq_work.c
12221 +++ b/kernel/irq_work.c
12223 #include <linux/cpu.h>
12224 #include <linux/notifier.h>
12225 #include <linux/smp.h>
12226 +#include <linux/interrupt.h>
12227 #include <asm/processor.h>
12230 @@ -65,6 +66,8 @@ void __weak arch_irq_work_raise(void)
12232 bool irq_work_queue_on(struct irq_work *work, int cpu)
12234 + struct llist_head *list;
12236 /* All work should have been flushed before going offline */
12237 WARN_ON_ONCE(cpu_is_offline(cpu));
12239 @@ -75,7 +78,12 @@ bool irq_work_queue_on(struct irq_work *work, int cpu)
12240 if (!irq_work_claim(work))
12243 - if (llist_add(&work->llnode, &per_cpu(raised_list, cpu)))
12244 + if (IS_ENABLED(CONFIG_PREEMPT_RT_FULL) && !(work->flags & IRQ_WORK_HARD_IRQ))
12245 + list = &per_cpu(lazy_list, cpu);
12247 + list = &per_cpu(raised_list, cpu);
12249 + if (llist_add(&work->llnode, list))
12250 arch_send_call_function_single_ipi(cpu);
12253 @@ -86,6 +94,9 @@ EXPORT_SYMBOL_GPL(irq_work_queue_on);
12254 /* Enqueue the irq work @work on the current CPU */
12255 bool irq_work_queue(struct irq_work *work)
12257 + struct llist_head *list;
12258 + bool lazy_work, realtime = IS_ENABLED(CONFIG_PREEMPT_RT_FULL);
12260 /* Only queue if not already pending */
12261 if (!irq_work_claim(work))
12263 @@ -93,13 +104,15 @@ bool irq_work_queue(struct irq_work *work)
12264 /* Queue the entry and raise the IPI if needed. */
12267 - /* If the work is "lazy", handle it from next tick if any */
12268 - if (work->flags & IRQ_WORK_LAZY) {
12269 - if (llist_add(&work->llnode, this_cpu_ptr(&lazy_list)) &&
12270 - tick_nohz_tick_stopped())
12271 - arch_irq_work_raise();
12273 - if (llist_add(&work->llnode, this_cpu_ptr(&raised_list)))
12274 + lazy_work = work->flags & IRQ_WORK_LAZY;
12276 + if (lazy_work || (realtime && !(work->flags & IRQ_WORK_HARD_IRQ)))
12277 + list = this_cpu_ptr(&lazy_list);
12279 + list = this_cpu_ptr(&raised_list);
12281 + if (llist_add(&work->llnode, list)) {
12282 + if (!lazy_work || tick_nohz_tick_stopped())
12283 arch_irq_work_raise();
12286 @@ -116,9 +129,8 @@ bool irq_work_needs_cpu(void)
12287 raised = this_cpu_ptr(&raised_list);
12288 lazy = this_cpu_ptr(&lazy_list);
12290 - if (llist_empty(raised) || arch_irq_work_has_interrupt())
12291 - if (llist_empty(lazy))
12293 + if (llist_empty(raised) && llist_empty(lazy))
12296 /* All work should have been flushed before going offline */
12297 WARN_ON_ONCE(cpu_is_offline(smp_processor_id()));
12298 @@ -132,7 +144,7 @@ static void irq_work_run_list(struct llist_head *list)
12299 struct irq_work *work;
12300 struct llist_node *llnode;
12302 - BUG_ON(!irqs_disabled());
12303 + BUG_ON_NONRT(!irqs_disabled());
12305 if (llist_empty(list))
12307 @@ -169,7 +181,16 @@ static void irq_work_run_list(struct llist_head *list)
12308 void irq_work_run(void)
12310 irq_work_run_list(this_cpu_ptr(&raised_list));
12311 - irq_work_run_list(this_cpu_ptr(&lazy_list));
12312 + if (IS_ENABLED(CONFIG_PREEMPT_RT_FULL)) {
12314 + * NOTE: we raise softirq via IPI for safety,
12315 + * and execute in irq_work_tick() to move the
12316 + * overhead from hard to soft irq context.
12318 + if (!llist_empty(this_cpu_ptr(&lazy_list)))
12319 + raise_softirq(TIMER_SOFTIRQ);
12321 + irq_work_run_list(this_cpu_ptr(&lazy_list));
12323 EXPORT_SYMBOL_GPL(irq_work_run);
12325 @@ -179,8 +200,17 @@ void irq_work_tick(void)
12327 if (!llist_empty(raised) && !arch_irq_work_has_interrupt())
12328 irq_work_run_list(raised);
12330 + if (!IS_ENABLED(CONFIG_PREEMPT_RT_FULL))
12331 + irq_work_run_list(this_cpu_ptr(&lazy_list));
12334 +#if defined(CONFIG_IRQ_WORK) && defined(CONFIG_PREEMPT_RT_FULL)
12335 +void irq_work_tick_soft(void)
12337 irq_work_run_list(this_cpu_ptr(&lazy_list));
12342 * Synchronize against the irq_work @entry, ensures the entry is not
12343 diff --git a/kernel/ksysfs.c b/kernel/ksysfs.c
12344 index ee1bc1bb8feb..ddef07958840 100644
12345 --- a/kernel/ksysfs.c
12346 +++ b/kernel/ksysfs.c
12347 @@ -136,6 +136,15 @@ KERNEL_ATTR_RO(vmcoreinfo);
12349 #endif /* CONFIG_KEXEC_CORE */
12351 +#if defined(CONFIG_PREEMPT_RT_FULL)
12352 +static ssize_t realtime_show(struct kobject *kobj,
12353 + struct kobj_attribute *attr, char *buf)
12355 + return sprintf(buf, "%d\n", 1);
12357 +KERNEL_ATTR_RO(realtime);
12360 /* whether file capabilities are enabled */
12361 static ssize_t fscaps_show(struct kobject *kobj,
12362 struct kobj_attribute *attr, char *buf)
12363 @@ -225,6 +234,9 @@ static struct attribute * kernel_attrs[] = {
12364 &rcu_expedited_attr.attr,
12365 &rcu_normal_attr.attr,
12367 +#ifdef CONFIG_PREEMPT_RT_FULL
12368 + &realtime_attr.attr,
12373 diff --git a/kernel/locking/Makefile b/kernel/locking/Makefile
12374 index 6f88e352cd4f..5e27fb1079e7 100644
12375 --- a/kernel/locking/Makefile
12376 +++ b/kernel/locking/Makefile
12378 # and is generally not a function of system call inputs.
12379 KCOV_INSTRUMENT := n
12381 -obj-y += mutex.o semaphore.o rwsem.o percpu-rwsem.o
12382 +obj-y += semaphore.o percpu-rwsem.o
12384 ifdef CONFIG_FUNCTION_TRACER
12385 CFLAGS_REMOVE_lockdep.o = $(CC_FLAGS_FTRACE)
12386 @@ -11,7 +11,11 @@ CFLAGS_REMOVE_mutex-debug.o = $(CC_FLAGS_FTRACE)
12387 CFLAGS_REMOVE_rtmutex-debug.o = $(CC_FLAGS_FTRACE)
12390 +ifneq ($(CONFIG_PREEMPT_RT_FULL),y)
12392 obj-$(CONFIG_DEBUG_MUTEXES) += mutex-debug.o
12395 obj-$(CONFIG_LOCKDEP) += lockdep.o
12396 ifeq ($(CONFIG_PROC_FS),y)
12397 obj-$(CONFIG_LOCKDEP) += lockdep_proc.o
12398 @@ -24,7 +28,10 @@ obj-$(CONFIG_RT_MUTEXES) += rtmutex.o
12399 obj-$(CONFIG_DEBUG_RT_MUTEXES) += rtmutex-debug.o
12400 obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock.o
12401 obj-$(CONFIG_DEBUG_SPINLOCK) += spinlock_debug.o
12402 +ifneq ($(CONFIG_PREEMPT_RT_FULL),y)
12403 obj-$(CONFIG_RWSEM_GENERIC_SPINLOCK) += rwsem-spinlock.o
12404 obj-$(CONFIG_RWSEM_XCHGADD_ALGORITHM) += rwsem-xadd.o
12406 +obj-$(CONFIG_PREEMPT_RT_FULL) += rt.o
12407 obj-$(CONFIG_QUEUED_RWLOCKS) += qrwlock.o
12408 obj-$(CONFIG_LOCK_TORTURE_TEST) += locktorture.o
12409 diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
12410 index 4d7ffc0a0d00..3d157b3128eb 100644
12411 --- a/kernel/locking/lockdep.c
12412 +++ b/kernel/locking/lockdep.c
12413 @@ -658,6 +658,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
12414 struct lockdep_subclass_key *key;
12415 struct hlist_head *hash_head;
12416 struct lock_class *class;
12417 + bool is_static = false;
12419 if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
12421 @@ -671,10 +672,23 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
12424 * Static locks do not have their class-keys yet - for them the key
12425 - * is the lock object itself:
12426 + * is the lock object itself. If the lock is in the per cpu area,
12427 + * the canonical address of the lock (per cpu offset removed) is
12430 - if (unlikely(!lock->key))
12431 - lock->key = (void *)lock;
12432 + if (unlikely(!lock->key)) {
12433 + unsigned long can_addr, addr = (unsigned long)lock;
12435 + if (__is_kernel_percpu_address(addr, &can_addr))
12436 + lock->key = (void *)can_addr;
12437 + else if (__is_module_percpu_address(addr, &can_addr))
12438 + lock->key = (void *)can_addr;
12439 + else if (static_obj(lock))
12440 + lock->key = (void *)lock;
12442 + return ERR_PTR(-EINVAL);
12443 + is_static = true;
12447 * NOTE: the class-key must be unique. For dynamic locks, a static
12448 @@ -706,7 +720,7 @@ look_up_lock_class(struct lockdep_map *lock, unsigned int subclass)
12453 + return is_static || static_obj(lock->key) ? NULL : ERR_PTR(-EINVAL);
12457 @@ -724,19 +738,18 @@ register_lock_class(struct lockdep_map *lock, unsigned int subclass, int force)
12458 DEBUG_LOCKS_WARN_ON(!irqs_disabled());
12460 class = look_up_lock_class(lock, subclass);
12461 - if (likely(class))
12462 + if (likely(!IS_ERR_OR_NULL(class)))
12463 goto out_set_class_cache;
12466 * Debug-check: all keys must be persistent!
12468 - if (!static_obj(lock->key)) {
12470 + if (IS_ERR(class)) {
12472 printk("INFO: trying to register non-static key.\n");
12473 printk("the code is fine but needs lockdep annotation.\n");
12474 printk("turning off the locking correctness validator.\n");
12480 @@ -3410,7 +3423,7 @@ static int match_held_lock(struct held_lock *hlock, struct lockdep_map *lock)
12481 * Clearly if the lock hasn't been acquired _ever_, we're not
12482 * holding it either, so report failure.
12485 + if (IS_ERR_OR_NULL(class))
12489 @@ -3689,6 +3702,7 @@ static void check_flags(unsigned long flags)
12493 +#ifndef CONFIG_PREEMPT_RT_FULL
12495 * We dont accurately track softirq state in e.g.
12496 * hardirq contexts (such as on 4KSTACKS), so only
12497 @@ -3703,6 +3717,7 @@ static void check_flags(unsigned long flags)
12498 DEBUG_LOCKS_WARN_ON(!current->softirqs_enabled);
12504 print_irqtrace_events(current);
12505 @@ -4159,7 +4174,7 @@ void lockdep_reset_lock(struct lockdep_map *lock)
12506 * If the class exists we look it up and zap it:
12508 class = look_up_lock_class(lock, j);
12510 + if (!IS_ERR_OR_NULL(class))
12514 diff --git a/kernel/locking/locktorture.c b/kernel/locking/locktorture.c
12515 index f8c5af52a131..788068773e61 100644
12516 --- a/kernel/locking/locktorture.c
12517 +++ b/kernel/locking/locktorture.c
12519 #include <linux/kthread.h>
12520 #include <linux/sched/rt.h>
12521 #include <linux/spinlock.h>
12522 -#include <linux/rwlock.h>
12523 #include <linux/mutex.h>
12524 #include <linux/rwsem.h>
12525 #include <linux/smp.h>
12526 diff --git a/kernel/locking/percpu-rwsem.c b/kernel/locking/percpu-rwsem.c
12527 index ce182599cf2e..2ad3a1e8344c 100644
12528 --- a/kernel/locking/percpu-rwsem.c
12529 +++ b/kernel/locking/percpu-rwsem.c
12530 @@ -18,7 +18,7 @@ int __percpu_init_rwsem(struct percpu_rw_semaphore *sem,
12531 /* ->rw_sem represents the whole percpu_rw_semaphore for lockdep */
12532 rcu_sync_init(&sem->rss, RCU_SCHED_SYNC);
12533 __init_rwsem(&sem->rw_sem, name, rwsem_key);
12534 - init_waitqueue_head(&sem->writer);
12535 + init_swait_queue_head(&sem->writer);
12536 sem->readers_block = 0;
12539 @@ -103,7 +103,7 @@ void __percpu_up_read(struct percpu_rw_semaphore *sem)
12540 __this_cpu_dec(*sem->read_count);
12542 /* Prod writer to recheck readers_active */
12543 - wake_up(&sem->writer);
12544 + swake_up(&sem->writer);
12546 EXPORT_SYMBOL_GPL(__percpu_up_read);
12548 @@ -160,7 +160,7 @@ void percpu_down_write(struct percpu_rw_semaphore *sem)
12551 /* Wait for all now active readers to complete. */
12552 - wait_event(sem->writer, readers_active_check(sem));
12553 + swait_event(sem->writer, readers_active_check(sem));
12555 EXPORT_SYMBOL_GPL(percpu_down_write);
12557 diff --git a/kernel/locking/rt.c b/kernel/locking/rt.c
12558 new file mode 100644
12559 index 000000000000..665754c00e1e
12561 +++ b/kernel/locking/rt.c
12566 + * Real-Time Preemption Support
12568 + * started by Ingo Molnar:
12570 + * Copyright (C) 2004-2006 Red Hat, Inc., Ingo Molnar <mingo@redhat.com>
12571 + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner <tglx@timesys.com>
12573 + * historic credit for proving that Linux spinlocks can be implemented via
12574 + * RT-aware mutexes goes to many people: The Pmutex project (Dirk Grambow
12575 + * and others) who prototyped it on 2.4 and did lots of comparative
12576 + * research and analysis; TimeSys, for proving that you can implement a
12577 + * fully preemptible kernel via the use of IRQ threading and mutexes;
12578 + * Bill Huey for persuasively arguing on lkml that the mutex model is the
12579 + * right one; and to MontaVista, who ported pmutexes to 2.6.
12581 + * This code is a from-scratch implementation and is not based on pmutexes,
12582 + * but the idea of converting spinlocks to mutexes is used here too.
12584 + * lock debugging, locking tree, deadlock detection:
12586 + * Copyright (C) 2004, LynuxWorks, Inc., Igor Manyilov, Bill Huey
12587 + * Released under the General Public License (GPL).
12589 + * Includes portions of the generic R/W semaphore implementation from:
12591 + * Copyright (c) 2001 David Howells (dhowells@redhat.com).
12592 + * - Derived partially from idea by Andrea Arcangeli <andrea@suse.de>
12593 + * - Derived also from comments by Linus
12595 + * Pending ownership of locks and ownership stealing:
12597 + * Copyright (C) 2005, Kihon Technologies Inc., Steven Rostedt
12599 + * (also by Steven Rostedt)
12600 + * - Converted single pi_lock to individual task locks.
12602 + * By Esben Nielsen:
12603 + * Doing priority inheritance with help of the scheduler.
12605 + * Copyright (C) 2006, Timesys Corp., Thomas Gleixner <tglx@timesys.com>
12606 + * - major rework based on Esben Nielsens initial patch
12607 + * - replaced thread_info references by task_struct refs
12608 + * - removed task->pending_owner dependency
12609 + * - BKL drop/reacquire for semaphore style locks to avoid deadlocks
12610 + * in the scheduler return path as discussed with Steven Rostedt
12612 + * Copyright (C) 2006, Kihon Technologies Inc.
12613 + * Steven Rostedt <rostedt@goodmis.org>
12614 + * - debugged and patched Thomas Gleixner's rework.
12615 + * - added back the cmpxchg to the rework.
12616 + * - turned atomic require back on for SMP.
12619 +#include <linux/spinlock.h>
12620 +#include <linux/rtmutex.h>
12621 +#include <linux/sched.h>
12622 +#include <linux/delay.h>
12623 +#include <linux/module.h>
12624 +#include <linux/kallsyms.h>
12625 +#include <linux/syscalls.h>
12626 +#include <linux/interrupt.h>
12627 +#include <linux/plist.h>
12628 +#include <linux/fs.h>
12629 +#include <linux/futex.h>
12630 +#include <linux/hrtimer.h>
12632 +#include "rtmutex_common.h"
12635 + * struct mutex functions
12637 +void __mutex_do_init(struct mutex *mutex, const char *name,
12638 + struct lock_class_key *key)
12640 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
12642 + * Make sure we are not reinitializing a held lock:
12644 + debug_check_no_locks_freed((void *)mutex, sizeof(*mutex));
12645 + lockdep_init_map(&mutex->dep_map, name, key, 0);
12647 + mutex->lock.save_state = 0;
12649 +EXPORT_SYMBOL(__mutex_do_init);
12651 +void __lockfunc _mutex_lock(struct mutex *lock)
12653 + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_);
12654 + rt_mutex_lock(&lock->lock);
12656 +EXPORT_SYMBOL(_mutex_lock);
12658 +int __lockfunc _mutex_lock_interruptible(struct mutex *lock)
12662 + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_);
12663 + ret = rt_mutex_lock_interruptible(&lock->lock);
12665 + mutex_release(&lock->dep_map, 1, _RET_IP_);
12668 +EXPORT_SYMBOL(_mutex_lock_interruptible);
12670 +int __lockfunc _mutex_lock_killable(struct mutex *lock)
12674 + mutex_acquire(&lock->dep_map, 0, 0, _RET_IP_);
12675 + ret = rt_mutex_lock_killable(&lock->lock);
12677 + mutex_release(&lock->dep_map, 1, _RET_IP_);
12680 +EXPORT_SYMBOL(_mutex_lock_killable);
12682 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
12683 +void __lockfunc _mutex_lock_nested(struct mutex *lock, int subclass)
12685 + mutex_acquire_nest(&lock->dep_map, subclass, 0, NULL, _RET_IP_);
12686 + rt_mutex_lock(&lock->lock);
12688 +EXPORT_SYMBOL(_mutex_lock_nested);
12690 +void __lockfunc _mutex_lock_nest_lock(struct mutex *lock, struct lockdep_map *nest)
12692 + mutex_acquire_nest(&lock->dep_map, 0, 0, nest, _RET_IP_);
12693 + rt_mutex_lock(&lock->lock);
12695 +EXPORT_SYMBOL(_mutex_lock_nest_lock);
12697 +int __lockfunc _mutex_lock_interruptible_nested(struct mutex *lock, int subclass)
12701 + mutex_acquire_nest(&lock->dep_map, subclass, 0, NULL, _RET_IP_);
12702 + ret = rt_mutex_lock_interruptible(&lock->lock);
12704 + mutex_release(&lock->dep_map, 1, _RET_IP_);
12707 +EXPORT_SYMBOL(_mutex_lock_interruptible_nested);
12709 +int __lockfunc _mutex_lock_killable_nested(struct mutex *lock, int subclass)
12713 + mutex_acquire(&lock->dep_map, subclass, 0, _RET_IP_);
12714 + ret = rt_mutex_lock_killable(&lock->lock);
12716 + mutex_release(&lock->dep_map, 1, _RET_IP_);
12719 +EXPORT_SYMBOL(_mutex_lock_killable_nested);
12722 +int __lockfunc _mutex_trylock(struct mutex *lock)
12724 + int ret = rt_mutex_trylock(&lock->lock);
12727 + mutex_acquire(&lock->dep_map, 0, 1, _RET_IP_);
12731 +EXPORT_SYMBOL(_mutex_trylock);
12733 +void __lockfunc _mutex_unlock(struct mutex *lock)
12735 + mutex_release(&lock->dep_map, 1, _RET_IP_);
12736 + rt_mutex_unlock(&lock->lock);
12738 +EXPORT_SYMBOL(_mutex_unlock);
12741 + * rwlock_t functions
12743 +int __lockfunc rt_write_trylock(rwlock_t *rwlock)
12747 + migrate_disable();
12748 + ret = rt_mutex_trylock(&rwlock->lock);
12750 + rwlock_acquire(&rwlock->dep_map, 0, 1, _RET_IP_);
12752 + migrate_enable();
12756 +EXPORT_SYMBOL(rt_write_trylock);
12758 +int __lockfunc rt_write_trylock_irqsave(rwlock_t *rwlock, unsigned long *flags)
12763 + ret = rt_write_trylock(rwlock);
12766 +EXPORT_SYMBOL(rt_write_trylock_irqsave);
12768 +int __lockfunc rt_read_trylock(rwlock_t *rwlock)
12770 + struct rt_mutex *lock = &rwlock->lock;
12774 + * recursive read locks succeed when current owns the lock,
12775 + * but not when read_depth == 0 which means that the lock is
12778 + if (rt_mutex_owner(lock) != current) {
12779 + migrate_disable();
12780 + ret = rt_mutex_trylock(lock);
12782 + rwlock_acquire(&rwlock->dep_map, 0, 1, _RET_IP_);
12784 + migrate_enable();
12786 + } else if (!rwlock->read_depth) {
12791 + rwlock->read_depth++;
12795 +EXPORT_SYMBOL(rt_read_trylock);
12797 +void __lockfunc rt_write_lock(rwlock_t *rwlock)
12799 + rwlock_acquire(&rwlock->dep_map, 0, 0, _RET_IP_);
12800 + __rt_spin_lock(&rwlock->lock);
12802 +EXPORT_SYMBOL(rt_write_lock);
12804 +void __lockfunc rt_read_lock(rwlock_t *rwlock)
12806 + struct rt_mutex *lock = &rwlock->lock;
12810 + * recursive read locks succeed when current owns the lock
12812 + if (rt_mutex_owner(lock) != current) {
12813 + rwlock_acquire(&rwlock->dep_map, 0, 0, _RET_IP_);
12814 + __rt_spin_lock(lock);
12816 + rwlock->read_depth++;
12819 +EXPORT_SYMBOL(rt_read_lock);
12821 +void __lockfunc rt_write_unlock(rwlock_t *rwlock)
12823 + /* NOTE: we always pass in '1' for nested, for simplicity */
12824 + rwlock_release(&rwlock->dep_map, 1, _RET_IP_);
12825 + __rt_spin_unlock(&rwlock->lock);
12826 + migrate_enable();
12828 +EXPORT_SYMBOL(rt_write_unlock);
12830 +void __lockfunc rt_read_unlock(rwlock_t *rwlock)
12832 + /* Release the lock only when read_depth is down to 0 */
12833 + if (--rwlock->read_depth == 0) {
12834 + rwlock_release(&rwlock->dep_map, 1, _RET_IP_);
12835 + __rt_spin_unlock(&rwlock->lock);
12836 + migrate_enable();
12839 +EXPORT_SYMBOL(rt_read_unlock);
12841 +unsigned long __lockfunc rt_write_lock_irqsave(rwlock_t *rwlock)
12843 + rt_write_lock(rwlock);
12847 +EXPORT_SYMBOL(rt_write_lock_irqsave);
12849 +unsigned long __lockfunc rt_read_lock_irqsave(rwlock_t *rwlock)
12851 + rt_read_lock(rwlock);
12855 +EXPORT_SYMBOL(rt_read_lock_irqsave);
12857 +void __rt_rwlock_init(rwlock_t *rwlock, char *name, struct lock_class_key *key)
12859 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
12861 + * Make sure we are not reinitializing a held lock:
12863 + debug_check_no_locks_freed((void *)rwlock, sizeof(*rwlock));
12864 + lockdep_init_map(&rwlock->dep_map, name, key, 0);
12866 + rwlock->lock.save_state = 1;
12867 + rwlock->read_depth = 0;
12869 +EXPORT_SYMBOL(__rt_rwlock_init);
12875 +void rt_up_write(struct rw_semaphore *rwsem)
12877 + rwsem_release(&rwsem->dep_map, 1, _RET_IP_);
12878 + rt_mutex_unlock(&rwsem->lock);
12880 +EXPORT_SYMBOL(rt_up_write);
12882 +void __rt_up_read(struct rw_semaphore *rwsem)
12884 + if (--rwsem->read_depth == 0)
12885 + rt_mutex_unlock(&rwsem->lock);
12888 +void rt_up_read(struct rw_semaphore *rwsem)
12890 + rwsem_release(&rwsem->dep_map, 1, _RET_IP_);
12891 + __rt_up_read(rwsem);
12893 +EXPORT_SYMBOL(rt_up_read);
12896 + * downgrade a write lock into a read lock
12897 + * - just wake up any readers at the front of the queue
12899 +void rt_downgrade_write(struct rw_semaphore *rwsem)
12901 + BUG_ON(rt_mutex_owner(&rwsem->lock) != current);
12902 + rwsem->read_depth = 1;
12904 +EXPORT_SYMBOL(rt_downgrade_write);
12906 +int rt_down_write_trylock(struct rw_semaphore *rwsem)
12908 + int ret = rt_mutex_trylock(&rwsem->lock);
12911 + rwsem_acquire(&rwsem->dep_map, 0, 1, _RET_IP_);
12914 +EXPORT_SYMBOL(rt_down_write_trylock);
12916 +void rt_down_write(struct rw_semaphore *rwsem)
12918 + rwsem_acquire(&rwsem->dep_map, 0, 0, _RET_IP_);
12919 + rt_mutex_lock(&rwsem->lock);
12921 +EXPORT_SYMBOL(rt_down_write);
12923 +int rt_down_write_killable(struct rw_semaphore *rwsem)
12927 + rwsem_acquire(&rwsem->dep_map, 0, 0, _RET_IP_);
12928 + ret = rt_mutex_lock_killable(&rwsem->lock);
12930 + rwsem_release(&rwsem->dep_map, 1, _RET_IP_);
12933 +EXPORT_SYMBOL(rt_down_write_killable);
12935 +int rt_down_write_killable_nested(struct rw_semaphore *rwsem, int subclass)
12939 + rwsem_acquire(&rwsem->dep_map, subclass, 0, _RET_IP_);
12940 + ret = rt_mutex_lock_killable(&rwsem->lock);
12942 + rwsem_release(&rwsem->dep_map, 1, _RET_IP_);
12945 +EXPORT_SYMBOL(rt_down_write_killable_nested);
12947 +void rt_down_write_nested(struct rw_semaphore *rwsem, int subclass)
12949 + rwsem_acquire(&rwsem->dep_map, subclass, 0, _RET_IP_);
12950 + rt_mutex_lock(&rwsem->lock);
12952 +EXPORT_SYMBOL(rt_down_write_nested);
12954 +void rt_down_write_nested_lock(struct rw_semaphore *rwsem,
12955 + struct lockdep_map *nest)
12957 + rwsem_acquire_nest(&rwsem->dep_map, 0, 0, nest, _RET_IP_);
12958 + rt_mutex_lock(&rwsem->lock);
12960 +EXPORT_SYMBOL(rt_down_write_nested_lock);
12962 +int rt__down_read_trylock(struct rw_semaphore *rwsem)
12964 + struct rt_mutex *lock = &rwsem->lock;
12968 + * recursive read locks succeed when current owns the rwsem,
12969 + * but not when read_depth == 0 which means that the rwsem is
12972 + if (rt_mutex_owner(lock) != current)
12973 + ret = rt_mutex_trylock(&rwsem->lock);
12974 + else if (!rwsem->read_depth)
12978 + rwsem->read_depth++;
12983 +int rt_down_read_trylock(struct rw_semaphore *rwsem)
12987 + ret = rt__down_read_trylock(rwsem);
12989 + rwsem_acquire(&rwsem->dep_map, 0, 1, _RET_IP_);
12993 +EXPORT_SYMBOL(rt_down_read_trylock);
12995 +void rt__down_read(struct rw_semaphore *rwsem)
12997 + struct rt_mutex *lock = &rwsem->lock;
12999 + if (rt_mutex_owner(lock) != current)
13000 + rt_mutex_lock(&rwsem->lock);
13001 + rwsem->read_depth++;
13003 +EXPORT_SYMBOL(rt__down_read);
13005 +static void __rt_down_read(struct rw_semaphore *rwsem, int subclass)
13007 + rwsem_acquire_read(&rwsem->dep_map, subclass, 0, _RET_IP_);
13008 + rt__down_read(rwsem);
13011 +void rt_down_read(struct rw_semaphore *rwsem)
13013 + __rt_down_read(rwsem, 0);
13015 +EXPORT_SYMBOL(rt_down_read);
13017 +void rt_down_read_nested(struct rw_semaphore *rwsem, int subclass)
13019 + __rt_down_read(rwsem, subclass);
13021 +EXPORT_SYMBOL(rt_down_read_nested);
13023 +void __rt_rwsem_init(struct rw_semaphore *rwsem, const char *name,
13024 + struct lock_class_key *key)
13026 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
13028 + * Make sure we are not reinitializing a held lock:
13030 + debug_check_no_locks_freed((void *)rwsem, sizeof(*rwsem));
13031 + lockdep_init_map(&rwsem->dep_map, name, key, 0);
13033 + rwsem->read_depth = 0;
13034 + rwsem->lock.save_state = 0;
13036 +EXPORT_SYMBOL(__rt_rwsem_init);
13039 + * atomic_dec_and_mutex_lock - return holding mutex if we dec to 0
13040 + * @cnt: the atomic which we are to dec
13041 + * @lock: the mutex to return holding if we dec to 0
13043 + * return true and hold lock if we dec to 0, return false otherwise
13045 +int atomic_dec_and_mutex_lock(atomic_t *cnt, struct mutex *lock)
13047 + /* dec if we can't possibly hit 0 */
13048 + if (atomic_add_unless(cnt, -1, 1))
13050 + /* we might hit 0, so take the lock */
13051 + mutex_lock(lock);
13052 + if (!atomic_dec_and_test(cnt)) {
13053 + /* when we actually did the dec, we didn't hit 0 */
13054 + mutex_unlock(lock);
13057 + /* we hit 0, and we hold the lock */
13060 +EXPORT_SYMBOL(atomic_dec_and_mutex_lock);
13061 diff --git a/kernel/locking/rtmutex.c b/kernel/locking/rtmutex.c
13062 index 2c49d76f96c3..4f1a7663c34d 100644
13063 --- a/kernel/locking/rtmutex.c
13064 +++ b/kernel/locking/rtmutex.c
13066 * Copyright (C) 2005-2006 Timesys Corp., Thomas Gleixner <tglx@timesys.com>
13067 * Copyright (C) 2005 Kihon Technologies Inc., Steven Rostedt
13068 * Copyright (C) 2006 Esben Nielsen
13069 + * Adaptive Spinlocks:
13070 + * Copyright (C) 2008 Novell, Inc., Gregory Haskins, Sven Dietrich,
13071 + * and Peter Morreale,
13072 + * Adaptive Spinlocks simplification:
13073 + * Copyright (C) 2008 Red Hat, Inc., Steven Rostedt <srostedt@redhat.com>
13075 * See Documentation/locking/rt-mutex-design.txt for details.
13078 #include <linux/sched/rt.h>
13079 #include <linux/sched/deadline.h>
13080 #include <linux/timer.h>
13081 +#include <linux/ww_mutex.h>
13083 #include "rtmutex_common.h"
13085 @@ -133,6 +139,12 @@ static void fixup_rt_mutex_waiters(struct rt_mutex *lock)
13086 WRITE_ONCE(*p, owner & ~RT_MUTEX_HAS_WAITERS);
13089 +static int rt_mutex_real_waiter(struct rt_mutex_waiter *waiter)
13091 + return waiter && waiter != PI_WAKEUP_INPROGRESS &&
13092 + waiter != PI_REQUEUE_INPROGRESS;
13096 * We can speed up the acquire/release, if there's no debugging state to be
13098 @@ -414,6 +426,14 @@ static bool rt_mutex_cond_detect_deadlock(struct rt_mutex_waiter *waiter,
13099 return debug_rt_mutex_detect_deadlock(waiter, chwalk);
13102 +static void rt_mutex_wake_waiter(struct rt_mutex_waiter *waiter)
13104 + if (waiter->savestate)
13105 + wake_up_lock_sleeper(waiter->task);
13107 + wake_up_process(waiter->task);
13111 * Max number of times we'll walk the boosting chain:
13113 @@ -421,7 +441,8 @@ int max_lock_depth = 1024;
13115 static inline struct rt_mutex *task_blocked_on_lock(struct task_struct *p)
13117 - return p->pi_blocked_on ? p->pi_blocked_on->lock : NULL;
13118 + return rt_mutex_real_waiter(p->pi_blocked_on) ?
13119 + p->pi_blocked_on->lock : NULL;
13123 @@ -557,7 +578,7 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
13124 * reached or the state of the chain has changed while we
13125 * dropped the locks.
13128 + if (!rt_mutex_real_waiter(waiter))
13129 goto out_unlock_pi;
13132 @@ -719,13 +740,16 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
13133 * follow here. This is the end of the chain we are walking.
13135 if (!rt_mutex_owner(lock)) {
13136 + struct rt_mutex_waiter *lock_top_waiter;
13139 * If the requeue [7] above changed the top waiter,
13140 * then we need to wake the new top waiter up to try
13143 - if (prerequeue_top_waiter != rt_mutex_top_waiter(lock))
13144 - wake_up_process(rt_mutex_top_waiter(lock)->task);
13145 + lock_top_waiter = rt_mutex_top_waiter(lock);
13146 + if (prerequeue_top_waiter != lock_top_waiter)
13147 + rt_mutex_wake_waiter(lock_top_waiter);
13148 raw_spin_unlock_irq(&lock->wait_lock);
13151 @@ -818,6 +842,25 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
13156 +#define STEAL_NORMAL 0
13157 +#define STEAL_LATERAL 1
13160 + * Note that RT tasks are excluded from lateral-steals to prevent the
13161 + * introduction of an unbounded latency
13163 +static inline int lock_is_stealable(struct task_struct *task,
13164 + struct task_struct *pendowner, int mode)
13166 + if (mode == STEAL_NORMAL || rt_task(task)) {
13167 + if (task->prio >= pendowner->prio)
13169 + } else if (task->prio > pendowner->prio)
13175 * Try to take an rt-mutex
13177 @@ -828,8 +871,9 @@ static int rt_mutex_adjust_prio_chain(struct task_struct *task,
13178 * @waiter: The waiter that is queued to the lock's wait tree if the
13179 * callsite called task_blocked_on_lock(), otherwise NULL
13181 -static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
13182 - struct rt_mutex_waiter *waiter)
13183 +static int __try_to_take_rt_mutex(struct rt_mutex *lock,
13184 + struct task_struct *task,
13185 + struct rt_mutex_waiter *waiter, int mode)
13188 * Before testing whether we can acquire @lock, we set the
13189 @@ -866,8 +910,10 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
13190 * If waiter is not the highest priority waiter of
13193 - if (waiter != rt_mutex_top_waiter(lock))
13194 + if (waiter != rt_mutex_top_waiter(lock)) {
13195 + /* XXX lock_is_stealable() ? */
13200 * We can acquire the lock. Remove the waiter from the
13201 @@ -885,14 +931,10 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
13202 * not need to be dequeued.
13204 if (rt_mutex_has_waiters(lock)) {
13206 - * If @task->prio is greater than or equal to
13207 - * the top waiter priority (kernel view),
13210 - if (task->prio >= rt_mutex_top_waiter(lock)->prio)
13212 + struct task_struct *pown = rt_mutex_top_waiter(lock)->task;
13214 + if (task != pown && !lock_is_stealable(task, pown, mode))
13217 * The current top waiter stays enqueued. We
13218 * don't have to change anything in the lock
13219 @@ -941,6 +983,433 @@ static int try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
13223 +#ifdef CONFIG_PREEMPT_RT_FULL
13225 + * preemptible spin_lock functions:
13227 +static inline void rt_spin_lock_fastlock(struct rt_mutex *lock,
13228 + void (*slowfn)(struct rt_mutex *lock,
13232 + might_sleep_no_state_check();
13235 + migrate_disable();
13237 + if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current)))
13238 + rt_mutex_deadlock_account_lock(lock, current);
13240 + slowfn(lock, do_mig_dis);
13243 +static inline int rt_spin_lock_fastunlock(struct rt_mutex *lock,
13244 + int (*slowfn)(struct rt_mutex *lock))
13246 + if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) {
13247 + rt_mutex_deadlock_account_unlock(current);
13250 + return slowfn(lock);
13254 + * Note that owner is a speculative pointer and dereferencing relies
13255 + * on rcu_read_lock() and the check against the lock owner.
13257 +static int adaptive_wait(struct rt_mutex *lock,
13258 + struct task_struct *owner)
13264 + if (owner != rt_mutex_owner(lock))
13267 + * Ensure that owner->on_cpu is dereferenced _after_
13268 + * checking the above to be valid.
13271 + if (!owner->on_cpu) {
13277 + rcu_read_unlock();
13281 +static int adaptive_wait(struct rt_mutex *lock,
13282 + struct task_struct *orig_owner)
13288 +static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
13289 + struct rt_mutex_waiter *waiter,
13290 + struct task_struct *task,
13291 + enum rtmutex_chainwalk chwalk);
13293 + * Slow path lock function spin_lock style: this variant is very
13294 + * careful not to miss any non-lock wakeups.
13296 + * We store the current state under p->pi_lock in p->saved_state and
13297 + * the try_to_wake_up() code handles this accordingly.
13299 +static void noinline __sched rt_spin_lock_slowlock(struct rt_mutex *lock,
13302 + struct task_struct *lock_owner, *self = current;
13303 + struct rt_mutex_waiter waiter, *top_waiter;
13304 + unsigned long flags;
13307 + rt_mutex_init_waiter(&waiter, true);
13309 + raw_spin_lock_irqsave(&lock->wait_lock, flags);
13311 + if (__try_to_take_rt_mutex(lock, self, NULL, STEAL_LATERAL)) {
13312 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13316 + BUG_ON(rt_mutex_owner(lock) == self);
13319 + * We save whatever state the task is in and we'll restore it
13320 + * after acquiring the lock taking real wakeups into account
13321 + * as well. We are serialized via pi_lock against wakeups. See
13322 + * try_to_wake_up().
13324 + raw_spin_lock(&self->pi_lock);
13325 + self->saved_state = self->state;
13326 + __set_current_state_no_track(TASK_UNINTERRUPTIBLE);
13327 + raw_spin_unlock(&self->pi_lock);
13329 + ret = task_blocks_on_rt_mutex(lock, &waiter, self, RT_MUTEX_MIN_CHAINWALK);
13333 + /* Try to acquire the lock again. */
13334 + if (__try_to_take_rt_mutex(lock, self, &waiter, STEAL_LATERAL))
13337 + top_waiter = rt_mutex_top_waiter(lock);
13338 + lock_owner = rt_mutex_owner(lock);
13340 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13342 + debug_rt_mutex_print_deadlock(&waiter);
13344 + if (top_waiter != &waiter || adaptive_wait(lock, lock_owner)) {
13346 + migrate_enable();
13349 + migrate_disable();
13352 + raw_spin_lock_irqsave(&lock->wait_lock, flags);
13354 + raw_spin_lock(&self->pi_lock);
13355 + __set_current_state_no_track(TASK_UNINTERRUPTIBLE);
13356 + raw_spin_unlock(&self->pi_lock);
13360 + * Restore the task state to current->saved_state. We set it
13361 + * to the original state above and the try_to_wake_up() code
13362 + * has possibly updated it when a real (non-rtmutex) wakeup
13363 + * happened while we were blocked. Clear saved_state so
13364 + * try_to_wakeup() does not get confused.
13366 + raw_spin_lock(&self->pi_lock);
13367 + __set_current_state_no_track(self->saved_state);
13368 + self->saved_state = TASK_RUNNING;
13369 + raw_spin_unlock(&self->pi_lock);
13372 + * try_to_take_rt_mutex() sets the waiter bit
13373 + * unconditionally. We might have to fix that up:
13375 + fixup_rt_mutex_waiters(lock);
13377 + BUG_ON(rt_mutex_has_waiters(lock) && &waiter == rt_mutex_top_waiter(lock));
13378 + BUG_ON(!RB_EMPTY_NODE(&waiter.tree_entry));
13380 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13382 + debug_rt_mutex_free_waiter(&waiter);
13385 +static void mark_wakeup_next_waiter(struct wake_q_head *wake_q,
13386 + struct wake_q_head *wake_sleeper_q,
13387 + struct rt_mutex *lock);
13389 + * Slow path to release a rt_mutex spin_lock style
13391 +static int noinline __sched rt_spin_lock_slowunlock(struct rt_mutex *lock)
13393 + unsigned long flags;
13395 + WAKE_Q(wake_sleeper_q);
13397 + raw_spin_lock_irqsave(&lock->wait_lock, flags);
13399 + debug_rt_mutex_unlock(lock);
13401 + rt_mutex_deadlock_account_unlock(current);
13403 + if (!rt_mutex_has_waiters(lock)) {
13404 + lock->owner = NULL;
13405 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13409 + mark_wakeup_next_waiter(&wake_q, &wake_sleeper_q, lock);
13411 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13412 + wake_up_q(&wake_q);
13413 + wake_up_q_sleeper(&wake_sleeper_q);
13415 + /* Undo pi boosting.when necessary */
13416 + rt_mutex_adjust_prio(current);
13420 +static int noinline __sched rt_spin_lock_slowunlock_no_deboost(struct rt_mutex *lock)
13422 + unsigned long flags;
13424 + WAKE_Q(wake_sleeper_q);
13426 + raw_spin_lock_irqsave(&lock->wait_lock, flags);
13428 + debug_rt_mutex_unlock(lock);
13430 + rt_mutex_deadlock_account_unlock(current);
13432 + if (!rt_mutex_has_waiters(lock)) {
13433 + lock->owner = NULL;
13434 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13438 + mark_wakeup_next_waiter(&wake_q, &wake_sleeper_q, lock);
13440 + raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13441 + wake_up_q(&wake_q);
13442 + wake_up_q_sleeper(&wake_sleeper_q);
13446 +void __lockfunc rt_spin_lock__no_mg(spinlock_t *lock)
13448 + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock, false);
13449 + spin_acquire(&lock->dep_map, 0, 0, _RET_IP_);
13451 +EXPORT_SYMBOL(rt_spin_lock__no_mg);
13453 +void __lockfunc rt_spin_lock(spinlock_t *lock)
13455 + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock, true);
13456 + spin_acquire(&lock->dep_map, 0, 0, _RET_IP_);
13458 +EXPORT_SYMBOL(rt_spin_lock);
13460 +void __lockfunc __rt_spin_lock(struct rt_mutex *lock)
13462 + rt_spin_lock_fastlock(lock, rt_spin_lock_slowlock, true);
13464 +EXPORT_SYMBOL(__rt_spin_lock);
13466 +void __lockfunc __rt_spin_lock__no_mg(struct rt_mutex *lock)
13468 + rt_spin_lock_fastlock(lock, rt_spin_lock_slowlock, false);
13470 +EXPORT_SYMBOL(__rt_spin_lock__no_mg);
13472 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
13473 +void __lockfunc rt_spin_lock_nested(spinlock_t *lock, int subclass)
13475 + spin_acquire(&lock->dep_map, subclass, 0, _RET_IP_);
13476 + rt_spin_lock_fastlock(&lock->lock, rt_spin_lock_slowlock, true);
13478 +EXPORT_SYMBOL(rt_spin_lock_nested);
13481 +void __lockfunc rt_spin_unlock__no_mg(spinlock_t *lock)
13483 + /* NOTE: we always pass in '1' for nested, for simplicity */
13484 + spin_release(&lock->dep_map, 1, _RET_IP_);
13485 + rt_spin_lock_fastunlock(&lock->lock, rt_spin_lock_slowunlock);
13487 +EXPORT_SYMBOL(rt_spin_unlock__no_mg);
13489 +void __lockfunc rt_spin_unlock(spinlock_t *lock)
13491 + /* NOTE: we always pass in '1' for nested, for simplicity */
13492 + spin_release(&lock->dep_map, 1, _RET_IP_);
13493 + rt_spin_lock_fastunlock(&lock->lock, rt_spin_lock_slowunlock);
13494 + migrate_enable();
13496 +EXPORT_SYMBOL(rt_spin_unlock);
13498 +int __lockfunc rt_spin_unlock_no_deboost(spinlock_t *lock)
13502 + /* NOTE: we always pass in '1' for nested, for simplicity */
13503 + spin_release(&lock->dep_map, 1, _RET_IP_);
13504 + ret = rt_spin_lock_fastunlock(&lock->lock, rt_spin_lock_slowunlock_no_deboost);
13505 + migrate_enable();
13509 +void __lockfunc __rt_spin_unlock(struct rt_mutex *lock)
13511 + rt_spin_lock_fastunlock(lock, rt_spin_lock_slowunlock);
13513 +EXPORT_SYMBOL(__rt_spin_unlock);
13516 + * Wait for the lock to get unlocked: instead of polling for an unlock
13517 + * (like raw spinlocks do), we lock and unlock, to force the kernel to
13518 + * schedule if there's contention:
13520 +void __lockfunc rt_spin_unlock_wait(spinlock_t *lock)
13523 + spin_unlock(lock);
13525 +EXPORT_SYMBOL(rt_spin_unlock_wait);
13527 +int __lockfunc rt_spin_trylock__no_mg(spinlock_t *lock)
13531 + ret = rt_mutex_trylock(&lock->lock);
13533 + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_);
13536 +EXPORT_SYMBOL(rt_spin_trylock__no_mg);
13538 +int __lockfunc rt_spin_trylock(spinlock_t *lock)
13542 + migrate_disable();
13543 + ret = rt_mutex_trylock(&lock->lock);
13545 + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_);
13547 + migrate_enable();
13550 +EXPORT_SYMBOL(rt_spin_trylock);
13552 +int __lockfunc rt_spin_trylock_bh(spinlock_t *lock)
13556 + local_bh_disable();
13557 + ret = rt_mutex_trylock(&lock->lock);
13559 + migrate_disable();
13560 + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_);
13562 + local_bh_enable();
13565 +EXPORT_SYMBOL(rt_spin_trylock_bh);
13567 +int __lockfunc rt_spin_trylock_irqsave(spinlock_t *lock, unsigned long *flags)
13572 + ret = rt_mutex_trylock(&lock->lock);
13574 + migrate_disable();
13575 + spin_acquire(&lock->dep_map, 0, 1, _RET_IP_);
13579 +EXPORT_SYMBOL(rt_spin_trylock_irqsave);
13581 +int atomic_dec_and_spin_lock(atomic_t *atomic, spinlock_t *lock)
13583 + /* Subtract 1 from counter unless that drops it to 0 (ie. it was 1) */
13584 + if (atomic_add_unless(atomic, -1, 1))
13586 + rt_spin_lock(lock);
13587 + if (atomic_dec_and_test(atomic))
13589 + rt_spin_unlock(lock);
13592 +EXPORT_SYMBOL(atomic_dec_and_spin_lock);
13595 +__rt_spin_lock_init(spinlock_t *lock, char *name, struct lock_class_key *key)
13597 +#ifdef CONFIG_DEBUG_LOCK_ALLOC
13599 + * Make sure we are not reinitializing a held lock:
13601 + debug_check_no_locks_freed((void *)lock, sizeof(*lock));
13602 + lockdep_init_map(&lock->dep_map, name, key, 0);
13605 +EXPORT_SYMBOL(__rt_spin_lock_init);
13607 +#endif /* PREEMPT_RT_FULL */
13609 +#ifdef CONFIG_PREEMPT_RT_FULL
13610 + static inline int __sched
13611 +__mutex_lock_check_stamp(struct rt_mutex *lock, struct ww_acquire_ctx *ctx)
13613 + struct ww_mutex *ww = container_of(lock, struct ww_mutex, base.lock);
13614 + struct ww_acquire_ctx *hold_ctx = ACCESS_ONCE(ww->ctx);
13619 + if (unlikely(ctx == hold_ctx))
13620 + return -EALREADY;
13622 + if (ctx->stamp - hold_ctx->stamp <= LONG_MAX &&
13623 + (ctx->stamp != hold_ctx->stamp || ctx > hold_ctx)) {
13624 +#ifdef CONFIG_DEBUG_MUTEXES
13625 + DEBUG_LOCKS_WARN_ON(ctx->contending_lock);
13626 + ctx->contending_lock = ww;
13634 + static inline int __sched
13635 +__mutex_lock_check_stamp(struct rt_mutex *lock, struct ww_acquire_ctx *ctx)
13644 +try_to_take_rt_mutex(struct rt_mutex *lock, struct task_struct *task,
13645 + struct rt_mutex_waiter *waiter)
13647 + return __try_to_take_rt_mutex(lock, task, waiter, STEAL_NORMAL);
13651 * Task blocks on lock.
13653 @@ -971,6 +1440,23 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
13656 raw_spin_lock(&task->pi_lock);
13659 + * In the case of futex requeue PI, this will be a proxy
13660 + * lock. The task will wake unaware that it is enqueueed on
13661 + * this lock. Avoid blocking on two locks and corrupting
13662 + * pi_blocked_on via the PI_WAKEUP_INPROGRESS
13663 + * flag. futex_wait_requeue_pi() sets this when it wakes up
13664 + * before requeue (due to a signal or timeout). Do not enqueue
13665 + * the task if PI_WAKEUP_INPROGRESS is set.
13667 + if (task != current && task->pi_blocked_on == PI_WAKEUP_INPROGRESS) {
13668 + raw_spin_unlock(&task->pi_lock);
13672 + BUG_ON(rt_mutex_real_waiter(task->pi_blocked_on));
13674 __rt_mutex_adjust_prio(task);
13675 waiter->task = task;
13676 waiter->lock = lock;
13677 @@ -994,7 +1480,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
13678 rt_mutex_enqueue_pi(owner, waiter);
13680 __rt_mutex_adjust_prio(owner);
13681 - if (owner->pi_blocked_on)
13682 + if (rt_mutex_real_waiter(owner->pi_blocked_on))
13684 } else if (rt_mutex_cond_detect_deadlock(waiter, chwalk)) {
13686 @@ -1036,6 +1522,7 @@ static int task_blocks_on_rt_mutex(struct rt_mutex *lock,
13687 * Called with lock->wait_lock held and interrupts disabled.
13689 static void mark_wakeup_next_waiter(struct wake_q_head *wake_q,
13690 + struct wake_q_head *wake_sleeper_q,
13691 struct rt_mutex *lock)
13693 struct rt_mutex_waiter *waiter;
13694 @@ -1064,7 +1551,10 @@ static void mark_wakeup_next_waiter(struct wake_q_head *wake_q,
13696 raw_spin_unlock(¤t->pi_lock);
13698 - wake_q_add(wake_q, waiter->task);
13699 + if (waiter->savestate)
13700 + wake_q_add(wake_sleeper_q, waiter->task);
13702 + wake_q_add(wake_q, waiter->task);
13706 @@ -1078,7 +1568,7 @@ static void remove_waiter(struct rt_mutex *lock,
13708 bool is_top_waiter = (waiter == rt_mutex_top_waiter(lock));
13709 struct task_struct *owner = rt_mutex_owner(lock);
13710 - struct rt_mutex *next_lock;
13711 + struct rt_mutex *next_lock = NULL;
13713 raw_spin_lock(¤t->pi_lock);
13714 rt_mutex_dequeue(lock, waiter);
13715 @@ -1102,7 +1592,8 @@ static void remove_waiter(struct rt_mutex *lock,
13716 __rt_mutex_adjust_prio(owner);
13718 /* Store the lock on which owner is blocked or NULL */
13719 - next_lock = task_blocked_on_lock(owner);
13720 + if (rt_mutex_real_waiter(owner->pi_blocked_on))
13721 + next_lock = task_blocked_on_lock(owner);
13723 raw_spin_unlock(&owner->pi_lock);
13725 @@ -1138,17 +1629,17 @@ void rt_mutex_adjust_pi(struct task_struct *task)
13726 raw_spin_lock_irqsave(&task->pi_lock, flags);
13728 waiter = task->pi_blocked_on;
13729 - if (!waiter || (waiter->prio == task->prio &&
13730 + if (!rt_mutex_real_waiter(waiter) || (waiter->prio == task->prio &&
13731 !dl_prio(task->prio))) {
13732 raw_spin_unlock_irqrestore(&task->pi_lock, flags);
13735 next_lock = waiter->lock;
13736 - raw_spin_unlock_irqrestore(&task->pi_lock, flags);
13738 /* gets dropped in rt_mutex_adjust_prio_chain()! */
13739 get_task_struct(task);
13741 + raw_spin_unlock_irqrestore(&task->pi_lock, flags);
13742 rt_mutex_adjust_prio_chain(task, RT_MUTEX_MIN_CHAINWALK, NULL,
13743 next_lock, NULL, task);
13745 @@ -1166,7 +1657,8 @@ void rt_mutex_adjust_pi(struct task_struct *task)
13747 __rt_mutex_slowlock(struct rt_mutex *lock, int state,
13748 struct hrtimer_sleeper *timeout,
13749 - struct rt_mutex_waiter *waiter)
13750 + struct rt_mutex_waiter *waiter,
13751 + struct ww_acquire_ctx *ww_ctx)
13755 @@ -1189,6 +1681,12 @@ __rt_mutex_slowlock(struct rt_mutex *lock, int state,
13759 + if (ww_ctx && ww_ctx->acquired > 0) {
13760 + ret = __mutex_lock_check_stamp(lock, ww_ctx);
13765 raw_spin_unlock_irq(&lock->wait_lock);
13767 debug_rt_mutex_print_deadlock(waiter);
13768 @@ -1223,21 +1721,96 @@ static void rt_mutex_handle_deadlock(int res, int detect_deadlock,
13772 +static __always_inline void ww_mutex_lock_acquired(struct ww_mutex *ww,
13773 + struct ww_acquire_ctx *ww_ctx)
13775 +#ifdef CONFIG_DEBUG_MUTEXES
13777 + * If this WARN_ON triggers, you used ww_mutex_lock to acquire,
13778 + * but released with a normal mutex_unlock in this call.
13780 + * This should never happen, always use ww_mutex_unlock.
13782 + DEBUG_LOCKS_WARN_ON(ww->ctx);
13785 + * Not quite done after calling ww_acquire_done() ?
13787 + DEBUG_LOCKS_WARN_ON(ww_ctx->done_acquire);
13789 + if (ww_ctx->contending_lock) {
13791 + * After -EDEADLK you tried to
13792 + * acquire a different ww_mutex? Bad!
13794 + DEBUG_LOCKS_WARN_ON(ww_ctx->contending_lock != ww);
13797 + * You called ww_mutex_lock after receiving -EDEADLK,
13798 + * but 'forgot' to unlock everything else first?
13800 + DEBUG_LOCKS_WARN_ON(ww_ctx->acquired > 0);
13801 + ww_ctx->contending_lock = NULL;
13805 + * Naughty, using a different class will lead to undefined behavior!
13807 + DEBUG_LOCKS_WARN_ON(ww_ctx->ww_class != ww->ww_class);
13809 + ww_ctx->acquired++;
13812 +#ifdef CONFIG_PREEMPT_RT_FULL
13813 +static void ww_mutex_account_lock(struct rt_mutex *lock,
13814 + struct ww_acquire_ctx *ww_ctx)
13816 + struct ww_mutex *ww = container_of(lock, struct ww_mutex, base.lock);
13817 + struct rt_mutex_waiter *waiter, *n;
13820 + * This branch gets optimized out for the common case,
13821 + * and is only important for ww_mutex_lock.
13823 + ww_mutex_lock_acquired(ww, ww_ctx);
13824 + ww->ctx = ww_ctx;
13827 + * Give any possible sleeping processes the chance to wake up,
13828 + * so they can recheck if they have to back off.
13830 + rbtree_postorder_for_each_entry_safe(waiter, n, &lock->waiters,
13832 + /* XXX debug rt mutex waiter wakeup */
13834 + BUG_ON(waiter->lock != lock);
13835 + rt_mutex_wake_waiter(waiter);
13841 +static void ww_mutex_account_lock(struct rt_mutex *lock,
13842 + struct ww_acquire_ctx *ww_ctx)
13849 * Slow path lock function:
13852 rt_mutex_slowlock(struct rt_mutex *lock, int state,
13853 struct hrtimer_sleeper *timeout,
13854 - enum rtmutex_chainwalk chwalk)
13855 + enum rtmutex_chainwalk chwalk,
13856 + struct ww_acquire_ctx *ww_ctx)
13858 struct rt_mutex_waiter waiter;
13859 unsigned long flags;
13862 - debug_rt_mutex_init_waiter(&waiter);
13863 - RB_CLEAR_NODE(&waiter.pi_tree_entry);
13864 - RB_CLEAR_NODE(&waiter.tree_entry);
13865 + rt_mutex_init_waiter(&waiter, false);
13868 * Technically we could use raw_spin_[un]lock_irq() here, but this can
13869 @@ -1251,6 +1824,8 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state,
13871 /* Try to acquire the lock again: */
13872 if (try_to_take_rt_mutex(lock, current, NULL)) {
13874 + ww_mutex_account_lock(lock, ww_ctx);
13875 raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13878 @@ -1265,13 +1840,23 @@ rt_mutex_slowlock(struct rt_mutex *lock, int state,
13881 /* sleep on the mutex */
13882 - ret = __rt_mutex_slowlock(lock, state, timeout, &waiter);
13883 + ret = __rt_mutex_slowlock(lock, state, timeout, &waiter,
13885 + else if (ww_ctx) {
13886 + /* ww_mutex received EDEADLK, let it become EALREADY */
13887 + ret = __mutex_lock_check_stamp(lock, ww_ctx);
13891 if (unlikely(ret)) {
13892 __set_current_state(TASK_RUNNING);
13893 if (rt_mutex_has_waiters(lock))
13894 remove_waiter(lock, &waiter);
13895 - rt_mutex_handle_deadlock(ret, chwalk, &waiter);
13896 + /* ww_mutex want to report EDEADLK/EALREADY, let them */
13898 + rt_mutex_handle_deadlock(ret, chwalk, &waiter);
13899 + } else if (ww_ctx) {
13900 + ww_mutex_account_lock(lock, ww_ctx);
13904 @@ -1331,7 +1916,8 @@ static inline int rt_mutex_slowtrylock(struct rt_mutex *lock)
13905 * Return whether the current task needs to undo a potential priority boosting.
13907 static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock,
13908 - struct wake_q_head *wake_q)
13909 + struct wake_q_head *wake_q,
13910 + struct wake_q_head *wake_sleeper_q)
13912 unsigned long flags;
13914 @@ -1387,7 +1973,7 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock,
13916 * Queue the next waiter for wakeup once we release the wait_lock.
13918 - mark_wakeup_next_waiter(wake_q, lock);
13919 + mark_wakeup_next_waiter(wake_q, wake_sleeper_q, lock);
13921 raw_spin_unlock_irqrestore(&lock->wait_lock, flags);
13923 @@ -1403,31 +1989,36 @@ static bool __sched rt_mutex_slowunlock(struct rt_mutex *lock,
13926 rt_mutex_fastlock(struct rt_mutex *lock, int state,
13927 + struct ww_acquire_ctx *ww_ctx,
13928 int (*slowfn)(struct rt_mutex *lock, int state,
13929 struct hrtimer_sleeper *timeout,
13930 - enum rtmutex_chainwalk chwalk))
13931 + enum rtmutex_chainwalk chwalk,
13932 + struct ww_acquire_ctx *ww_ctx))
13934 if (likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) {
13935 rt_mutex_deadlock_account_lock(lock, current);
13938 - return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK);
13939 + return slowfn(lock, state, NULL, RT_MUTEX_MIN_CHAINWALK,
13944 rt_mutex_timed_fastlock(struct rt_mutex *lock, int state,
13945 struct hrtimer_sleeper *timeout,
13946 enum rtmutex_chainwalk chwalk,
13947 + struct ww_acquire_ctx *ww_ctx,
13948 int (*slowfn)(struct rt_mutex *lock, int state,
13949 struct hrtimer_sleeper *timeout,
13950 - enum rtmutex_chainwalk chwalk))
13951 + enum rtmutex_chainwalk chwalk,
13952 + struct ww_acquire_ctx *ww_ctx))
13954 if (chwalk == RT_MUTEX_MIN_CHAINWALK &&
13955 likely(rt_mutex_cmpxchg_acquire(lock, NULL, current))) {
13956 rt_mutex_deadlock_account_lock(lock, current);
13959 - return slowfn(lock, state, timeout, chwalk);
13960 + return slowfn(lock, state, timeout, chwalk, ww_ctx);
13964 @@ -1444,17 +2035,20 @@ rt_mutex_fasttrylock(struct rt_mutex *lock,
13966 rt_mutex_fastunlock(struct rt_mutex *lock,
13967 bool (*slowfn)(struct rt_mutex *lock,
13968 - struct wake_q_head *wqh))
13969 + struct wake_q_head *wqh,
13970 + struct wake_q_head *wq_sleeper))
13973 + WAKE_Q(wake_sleeper_q);
13975 if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) {
13976 rt_mutex_deadlock_account_unlock(current);
13979 - bool deboost = slowfn(lock, &wake_q);
13980 + bool deboost = slowfn(lock, &wake_q, &wake_sleeper_q);
13982 wake_up_q(&wake_q);
13983 + wake_up_q_sleeper(&wake_sleeper_q);
13985 /* Undo pi boosting if necessary: */
13987 @@ -1471,7 +2065,7 @@ void __sched rt_mutex_lock(struct rt_mutex *lock)
13991 - rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, rt_mutex_slowlock);
13992 + rt_mutex_fastlock(lock, TASK_UNINTERRUPTIBLE, NULL, rt_mutex_slowlock);
13994 EXPORT_SYMBOL_GPL(rt_mutex_lock);
13996 @@ -1488,7 +2082,7 @@ int __sched rt_mutex_lock_interruptible(struct rt_mutex *lock)
14000 - return rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, rt_mutex_slowlock);
14001 + return rt_mutex_fastlock(lock, TASK_INTERRUPTIBLE, NULL, rt_mutex_slowlock);
14003 EXPORT_SYMBOL_GPL(rt_mutex_lock_interruptible);
14005 @@ -1501,11 +2095,30 @@ int rt_mutex_timed_futex_lock(struct rt_mutex *lock,
14008 return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout,
14009 - RT_MUTEX_FULL_CHAINWALK,
14010 + RT_MUTEX_FULL_CHAINWALK, NULL,
14011 rt_mutex_slowlock);
14015 + * rt_mutex_lock_killable - lock a rt_mutex killable
14017 + * @lock: the rt_mutex to be locked
14018 + * @detect_deadlock: deadlock detection on/off
14022 + * -EINTR when interrupted by a signal
14023 + * -EDEADLK when the lock would deadlock (when deadlock detection is on)
14025 +int __sched rt_mutex_lock_killable(struct rt_mutex *lock)
14029 + return rt_mutex_fastlock(lock, TASK_KILLABLE, NULL, rt_mutex_slowlock);
14031 +EXPORT_SYMBOL_GPL(rt_mutex_lock_killable);
14034 * rt_mutex_timed_lock - lock a rt_mutex interruptible
14035 * the timeout structure is provided
14037 @@ -1525,6 +2138,7 @@ rt_mutex_timed_lock(struct rt_mutex *lock, struct hrtimer_sleeper *timeout)
14039 return rt_mutex_timed_fastlock(lock, TASK_INTERRUPTIBLE, timeout,
14040 RT_MUTEX_MIN_CHAINWALK,
14042 rt_mutex_slowlock);
14044 EXPORT_SYMBOL_GPL(rt_mutex_timed_lock);
14045 @@ -1542,7 +2156,11 @@ EXPORT_SYMBOL_GPL(rt_mutex_timed_lock);
14047 int __sched rt_mutex_trylock(struct rt_mutex *lock)
14049 +#ifdef CONFIG_PREEMPT_RT_FULL
14050 + if (WARN_ON_ONCE(in_irq() || in_nmi()))
14052 if (WARN_ON_ONCE(in_irq() || in_nmi() || in_serving_softirq()))
14056 return rt_mutex_fasttrylock(lock, rt_mutex_slowtrylock);
14057 @@ -1568,13 +2186,14 @@ EXPORT_SYMBOL_GPL(rt_mutex_unlock);
14060 bool __sched rt_mutex_futex_unlock(struct rt_mutex *lock,
14061 - struct wake_q_head *wqh)
14062 + struct wake_q_head *wqh,
14063 + struct wake_q_head *wq_sleeper)
14065 if (likely(rt_mutex_cmpxchg_release(lock, current, NULL))) {
14066 rt_mutex_deadlock_account_unlock(current);
14069 - return rt_mutex_slowunlock(lock, wqh);
14070 + return rt_mutex_slowunlock(lock, wqh, wq_sleeper);
14074 @@ -1607,13 +2226,12 @@ EXPORT_SYMBOL_GPL(rt_mutex_destroy);
14075 void __rt_mutex_init(struct rt_mutex *lock, const char *name)
14077 lock->owner = NULL;
14078 - raw_spin_lock_init(&lock->wait_lock);
14079 lock->waiters = RB_ROOT;
14080 lock->waiters_leftmost = NULL;
14082 debug_rt_mutex_init(lock, name);
14084 -EXPORT_SYMBOL_GPL(__rt_mutex_init);
14085 +EXPORT_SYMBOL(__rt_mutex_init);
14088 * rt_mutex_init_proxy_locked - initialize and lock a rt_mutex on behalf of a
14089 @@ -1628,7 +2246,7 @@ EXPORT_SYMBOL_GPL(__rt_mutex_init);
14090 void rt_mutex_init_proxy_locked(struct rt_mutex *lock,
14091 struct task_struct *proxy_owner)
14093 - __rt_mutex_init(lock, NULL);
14094 + rt_mutex_init(lock);
14095 debug_rt_mutex_proxy_lock(lock, proxy_owner);
14096 rt_mutex_set_owner(lock, proxy_owner);
14097 rt_mutex_deadlock_account_lock(lock, proxy_owner);
14098 @@ -1676,6 +2294,35 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
14102 +#ifdef CONFIG_PREEMPT_RT_FULL
14104 + * In PREEMPT_RT there's an added race.
14105 + * If the task, that we are about to requeue, times out,
14106 + * it can set the PI_WAKEUP_INPROGRESS. This tells the requeue
14107 + * to skip this task. But right after the task sets
14108 + * its pi_blocked_on to PI_WAKEUP_INPROGRESS it can then
14109 + * block on the spin_lock(&hb->lock), which in RT is an rtmutex.
14110 + * This will replace the PI_WAKEUP_INPROGRESS with the actual
14111 + * lock that it blocks on. We *must not* place this task
14112 + * on this proxy lock in that case.
14114 + * To prevent this race, we first take the task's pi_lock
14115 + * and check if it has updated its pi_blocked_on. If it has,
14116 + * we assume that it woke up and we return -EAGAIN.
14117 + * Otherwise, we set the task's pi_blocked_on to
14118 + * PI_REQUEUE_INPROGRESS, so that if the task is waking up
14119 + * it will know that we are in the process of requeuing it.
14121 + raw_spin_lock(&task->pi_lock);
14122 + if (task->pi_blocked_on) {
14123 + raw_spin_unlock(&task->pi_lock);
14124 + raw_spin_unlock_irq(&lock->wait_lock);
14127 + task->pi_blocked_on = PI_REQUEUE_INPROGRESS;
14128 + raw_spin_unlock(&task->pi_lock);
14131 /* We enforce deadlock detection for futexes */
14132 ret = task_blocks_on_rt_mutex(lock, waiter, task,
14133 RT_MUTEX_FULL_CHAINWALK);
14134 @@ -1690,7 +2337,7 @@ int rt_mutex_start_proxy_lock(struct rt_mutex *lock,
14138 - if (unlikely(ret))
14139 + if (ret && rt_mutex_has_waiters(lock))
14140 remove_waiter(lock, waiter);
14142 raw_spin_unlock_irq(&lock->wait_lock);
14143 @@ -1746,7 +2393,7 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock,
14144 set_current_state(TASK_INTERRUPTIBLE);
14146 /* sleep on the mutex */
14147 - ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter);
14148 + ret = __rt_mutex_slowlock(lock, TASK_INTERRUPTIBLE, to, waiter, NULL);
14151 remove_waiter(lock, waiter);
14152 @@ -1761,3 +2408,89 @@ int rt_mutex_finish_proxy_lock(struct rt_mutex *lock,
14158 +ww_mutex_deadlock_injection(struct ww_mutex *lock, struct ww_acquire_ctx *ctx)
14160 +#ifdef CONFIG_DEBUG_WW_MUTEX_SLOWPATH
14163 + if (ctx->deadlock_inject_countdown-- == 0) {
14164 + tmp = ctx->deadlock_inject_interval;
14165 + if (tmp > UINT_MAX/4)
14168 + tmp = tmp*2 + tmp + tmp/2;
14170 + ctx->deadlock_inject_interval = tmp;
14171 + ctx->deadlock_inject_countdown = tmp;
14172 + ctx->contending_lock = lock;
14174 + ww_mutex_unlock(lock);
14183 +#ifdef CONFIG_PREEMPT_RT_FULL
14185 +__ww_mutex_lock_interruptible(struct ww_mutex *lock, struct ww_acquire_ctx *ww_ctx)
14191 + mutex_acquire_nest(&lock->base.dep_map, 0, 0, &ww_ctx->dep_map, _RET_IP_);
14192 + ret = rt_mutex_slowlock(&lock->base.lock, TASK_INTERRUPTIBLE, NULL, 0, ww_ctx);
14194 + mutex_release(&lock->base.dep_map, 1, _RET_IP_);
14195 + else if (!ret && ww_ctx->acquired > 1)
14196 + return ww_mutex_deadlock_injection(lock, ww_ctx);
14200 +EXPORT_SYMBOL_GPL(__ww_mutex_lock_interruptible);
14203 +__ww_mutex_lock(struct ww_mutex *lock, struct ww_acquire_ctx *ww_ctx)
14209 + mutex_acquire_nest(&lock->base.dep_map, 0, 0, &ww_ctx->dep_map, _RET_IP_);
14210 + ret = rt_mutex_slowlock(&lock->base.lock, TASK_UNINTERRUPTIBLE, NULL, 0, ww_ctx);
14212 + mutex_release(&lock->base.dep_map, 1, _RET_IP_);
14213 + else if (!ret && ww_ctx->acquired > 1)
14214 + return ww_mutex_deadlock_injection(lock, ww_ctx);
14218 +EXPORT_SYMBOL_GPL(__ww_mutex_lock);
14220 +void __sched ww_mutex_unlock(struct ww_mutex *lock)
14222 + int nest = !!lock->ctx;
14225 + * The unlocking fastpath is the 0->1 transition from 'locked'
14226 + * into 'unlocked' state:
14229 +#ifdef CONFIG_DEBUG_MUTEXES
14230 + DEBUG_LOCKS_WARN_ON(!lock->ctx->acquired);
14232 + if (lock->ctx->acquired > 0)
14233 + lock->ctx->acquired--;
14234 + lock->ctx = NULL;
14237 + mutex_release(&lock->base.dep_map, nest, _RET_IP_);
14238 + rt_mutex_unlock(&lock->base.lock);
14240 +EXPORT_SYMBOL(ww_mutex_unlock);
14242 diff --git a/kernel/locking/rtmutex_common.h b/kernel/locking/rtmutex_common.h
14243 index e317e1cbb3eb..f457c7574920 100644
14244 --- a/kernel/locking/rtmutex_common.h
14245 +++ b/kernel/locking/rtmutex_common.h
14246 @@ -27,6 +27,7 @@ struct rt_mutex_waiter {
14247 struct rb_node pi_tree_entry;
14248 struct task_struct *task;
14249 struct rt_mutex *lock;
14251 #ifdef CONFIG_DEBUG_RT_MUTEXES
14253 struct pid *deadlock_task_pid;
14254 @@ -98,6 +99,9 @@ enum rtmutex_chainwalk {
14256 * PI-futex support (proxy locking functions, etc.):
14258 +#define PI_WAKEUP_INPROGRESS ((struct rt_mutex_waiter *) 1)
14259 +#define PI_REQUEUE_INPROGRESS ((struct rt_mutex_waiter *) 2)
14261 extern struct task_struct *rt_mutex_next_owner(struct rt_mutex *lock);
14262 extern void rt_mutex_init_proxy_locked(struct rt_mutex *lock,
14263 struct task_struct *proxy_owner);
14264 @@ -111,7 +115,8 @@ extern int rt_mutex_finish_proxy_lock(struct rt_mutex *lock,
14265 struct rt_mutex_waiter *waiter);
14266 extern int rt_mutex_timed_futex_lock(struct rt_mutex *l, struct hrtimer_sleeper *to);
14267 extern bool rt_mutex_futex_unlock(struct rt_mutex *lock,
14268 - struct wake_q_head *wqh);
14269 + struct wake_q_head *wqh,
14270 + struct wake_q_head *wq_sleeper);
14271 extern void rt_mutex_adjust_prio(struct task_struct *task);
14273 #ifdef CONFIG_DEBUG_RT_MUTEXES
14274 @@ -120,4 +125,14 @@ extern void rt_mutex_adjust_prio(struct task_struct *task);
14275 # include "rtmutex.h"
14278 +static inline void
14279 +rt_mutex_init_waiter(struct rt_mutex_waiter *waiter, bool savestate)
14281 + debug_rt_mutex_init_waiter(waiter);
14282 + waiter->task = NULL;
14283 + waiter->savestate = savestate;
14284 + RB_CLEAR_NODE(&waiter->pi_tree_entry);
14285 + RB_CLEAR_NODE(&waiter->tree_entry);
14289 diff --git a/kernel/locking/spinlock.c b/kernel/locking/spinlock.c
14290 index db3ccb1dd614..909779647bd1 100644
14291 --- a/kernel/locking/spinlock.c
14292 +++ b/kernel/locking/spinlock.c
14293 @@ -124,8 +124,11 @@ void __lockfunc __raw_##op##_lock_bh(locktype##_t *lock) \
14294 * __[spin|read|write]_lock_bh()
14296 BUILD_LOCK_OPS(spin, raw_spinlock);
14298 +#ifndef CONFIG_PREEMPT_RT_FULL
14299 BUILD_LOCK_OPS(read, rwlock);
14300 BUILD_LOCK_OPS(write, rwlock);
14305 @@ -209,6 +212,8 @@ void __lockfunc _raw_spin_unlock_bh(raw_spinlock_t *lock)
14306 EXPORT_SYMBOL(_raw_spin_unlock_bh);
14309 +#ifndef CONFIG_PREEMPT_RT_FULL
14311 #ifndef CONFIG_INLINE_READ_TRYLOCK
14312 int __lockfunc _raw_read_trylock(rwlock_t *lock)
14314 @@ -353,6 +358,8 @@ void __lockfunc _raw_write_unlock_bh(rwlock_t *lock)
14315 EXPORT_SYMBOL(_raw_write_unlock_bh);
14318 +#endif /* !PREEMPT_RT_FULL */
14320 #ifdef CONFIG_DEBUG_LOCK_ALLOC
14322 void __lockfunc _raw_spin_lock_nested(raw_spinlock_t *lock, int subclass)
14323 diff --git a/kernel/locking/spinlock_debug.c b/kernel/locking/spinlock_debug.c
14324 index 0374a596cffa..94970338d518 100644
14325 --- a/kernel/locking/spinlock_debug.c
14326 +++ b/kernel/locking/spinlock_debug.c
14327 @@ -31,6 +31,7 @@ void __raw_spin_lock_init(raw_spinlock_t *lock, const char *name,
14329 EXPORT_SYMBOL(__raw_spin_lock_init);
14331 +#ifndef CONFIG_PREEMPT_RT_FULL
14332 void __rwlock_init(rwlock_t *lock, const char *name,
14333 struct lock_class_key *key)
14335 @@ -48,6 +49,7 @@ void __rwlock_init(rwlock_t *lock, const char *name,
14338 EXPORT_SYMBOL(__rwlock_init);
14341 static void spin_dump(raw_spinlock_t *lock, const char *msg)
14343 @@ -159,6 +161,7 @@ void do_raw_spin_unlock(raw_spinlock_t *lock)
14344 arch_spin_unlock(&lock->raw_lock);
14347 +#ifndef CONFIG_PREEMPT_RT_FULL
14348 static void rwlock_bug(rwlock_t *lock, const char *msg)
14350 if (!debug_locks_off())
14351 @@ -300,3 +303,5 @@ void do_raw_write_unlock(rwlock_t *lock)
14352 debug_write_unlock(lock);
14353 arch_write_unlock(&lock->raw_lock);
14357 diff --git a/kernel/module.c b/kernel/module.c
14358 index 0e54d5bf0097..3483a3743b44 100644
14359 --- a/kernel/module.c
14360 +++ b/kernel/module.c
14361 @@ -660,16 +660,7 @@ static void percpu_modcopy(struct module *mod,
14362 memcpy(per_cpu_ptr(mod->percpu, cpu), from, size);
14366 - * is_module_percpu_address - test whether address is from module static percpu
14367 - * @addr: address to test
14369 - * Test whether @addr belongs to module static percpu area.
14372 - * %true if @addr is from module static percpu area
14374 -bool is_module_percpu_address(unsigned long addr)
14375 +bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr)
14377 struct module *mod;
14379 @@ -683,9 +674,11 @@ bool is_module_percpu_address(unsigned long addr)
14381 for_each_possible_cpu(cpu) {
14382 void *start = per_cpu_ptr(mod->percpu, cpu);
14383 + void *va = (void *)addr;
14385 - if ((void *)addr >= start &&
14386 - (void *)addr < start + mod->percpu_size) {
14387 + if (va >= start && va < start + mod->percpu_size) {
14389 + *can_addr = (unsigned long) (va - start);
14393 @@ -696,6 +689,20 @@ bool is_module_percpu_address(unsigned long addr)
14398 + * is_module_percpu_address - test whether address is from module static percpu
14399 + * @addr: address to test
14401 + * Test whether @addr belongs to module static percpu area.
14404 + * %true if @addr is from module static percpu area
14406 +bool is_module_percpu_address(unsigned long addr)
14408 + return __is_module_percpu_address(addr, NULL);
14411 #else /* ... !CONFIG_SMP */
14413 static inline void __percpu *mod_percpu(struct module *mod)
14414 @@ -727,6 +734,11 @@ bool is_module_percpu_address(unsigned long addr)
14418 +bool __is_module_percpu_address(unsigned long addr, unsigned long *can_addr)
14423 #endif /* CONFIG_SMP */
14425 #define MODINFO_ATTR(field) \
14426 diff --git a/kernel/panic.c b/kernel/panic.c
14427 index e6480e20379e..7e9c1918a94e 100644
14428 --- a/kernel/panic.c
14429 +++ b/kernel/panic.c
14430 @@ -482,9 +482,11 @@ static u64 oops_id;
14432 static int init_oops_id(void)
14434 +#ifndef CONFIG_PREEMPT_RT_FULL
14436 get_random_bytes(&oops_id, sizeof(oops_id));
14442 diff --git a/kernel/power/hibernate.c b/kernel/power/hibernate.c
14443 index b26dbc48c75b..968255f27a33 100644
14444 --- a/kernel/power/hibernate.c
14445 +++ b/kernel/power/hibernate.c
14446 @@ -286,6 +286,8 @@ static int create_image(int platform_mode)
14448 local_irq_disable();
14450 + system_state = SYSTEM_SUSPEND;
14452 error = syscore_suspend();
14454 printk(KERN_ERR "PM: Some system devices failed to power down, "
14455 @@ -317,6 +319,7 @@ static int create_image(int platform_mode)
14459 + system_state = SYSTEM_RUNNING;
14460 local_irq_enable();
14463 @@ -446,6 +449,7 @@ static int resume_target_kernel(bool platform_mode)
14466 local_irq_disable();
14467 + system_state = SYSTEM_SUSPEND;
14469 error = syscore_suspend();
14471 @@ -479,6 +483,7 @@ static int resume_target_kernel(bool platform_mode)
14475 + system_state = SYSTEM_RUNNING;
14476 local_irq_enable();
14479 @@ -564,6 +569,7 @@ int hibernation_platform_enter(void)
14482 local_irq_disable();
14483 + system_state = SYSTEM_SUSPEND;
14485 if (pm_wakeup_pending()) {
14487 @@ -576,6 +582,7 @@ int hibernation_platform_enter(void)
14491 + system_state = SYSTEM_RUNNING;
14492 local_irq_enable();
14495 @@ -676,6 +683,10 @@ static int load_image_and_restore(void)
14499 +#ifndef CONFIG_SUSPEND
14500 +bool pm_in_action;
14504 * hibernate - Carry out system hibernation, including saving the image.
14506 @@ -689,6 +700,8 @@ int hibernate(void)
14510 + pm_in_action = true;
14512 lock_system_sleep();
14513 /* The snapshot device should not be opened while we're running */
14514 if (!atomic_add_unless(&snapshot_device_available, -1, 0)) {
14515 @@ -766,6 +779,7 @@ int hibernate(void)
14516 atomic_inc(&snapshot_device_available);
14518 unlock_system_sleep();
14519 + pm_in_action = false;
14523 diff --git a/kernel/power/suspend.c b/kernel/power/suspend.c
14524 index 6ccb08f57fcb..c8cbb5ed2fe3 100644
14525 --- a/kernel/power/suspend.c
14526 +++ b/kernel/power/suspend.c
14527 @@ -369,6 +369,8 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
14528 arch_suspend_disable_irqs();
14529 BUG_ON(!irqs_disabled());
14531 + system_state = SYSTEM_SUSPEND;
14533 error = syscore_suspend();
14535 *wakeup = pm_wakeup_pending();
14536 @@ -385,6 +387,8 @@ static int suspend_enter(suspend_state_t state, bool *wakeup)
14540 + system_state = SYSTEM_RUNNING;
14542 arch_suspend_enable_irqs();
14543 BUG_ON(irqs_disabled());
14545 @@ -527,6 +531,8 @@ static int enter_state(suspend_state_t state)
14549 +bool pm_in_action;
14552 * pm_suspend - Externally visible function for suspending the system.
14553 * @state: System sleep state to enter.
14554 @@ -541,6 +547,8 @@ int pm_suspend(suspend_state_t state)
14555 if (state <= PM_SUSPEND_ON || state >= PM_SUSPEND_MAX)
14558 + pm_in_action = true;
14560 error = enter_state(state);
14562 suspend_stats.fail++;
14563 @@ -548,6 +556,7 @@ int pm_suspend(suspend_state_t state)
14565 suspend_stats.success++;
14567 + pm_in_action = false;
14570 EXPORT_SYMBOL(pm_suspend);
14571 diff --git a/kernel/printk/printk.c b/kernel/printk/printk.c
14572 index 9c5b231684d0..cf15bdb6855b 100644
14573 --- a/kernel/printk/printk.c
14574 +++ b/kernel/printk/printk.c
14575 @@ -351,6 +351,65 @@ __packed __aligned(4)
14577 DEFINE_RAW_SPINLOCK(logbuf_lock);
14579 +#ifdef CONFIG_EARLY_PRINTK
14580 +struct console *early_console;
14582 +static void early_vprintk(const char *fmt, va_list ap)
14584 + if (early_console) {
14586 + int n = vscnprintf(buf, sizeof(buf), fmt, ap);
14588 + early_console->write(early_console, buf, n);
14592 +asmlinkage void early_printk(const char *fmt, ...)
14596 + va_start(ap, fmt);
14597 + early_vprintk(fmt, ap);
14602 + * This is independent of any log levels - a global
14603 + * kill switch that turns off all of printk.
14605 + * Used by the NMI watchdog if early-printk is enabled.
14607 +static bool __read_mostly printk_killswitch;
14609 +static int __init force_early_printk_setup(char *str)
14611 + printk_killswitch = true;
14614 +early_param("force_early_printk", force_early_printk_setup);
14616 +void printk_kill(void)
14618 + printk_killswitch = true;
14621 +#ifdef CONFIG_PRINTK
14622 +static int forced_early_printk(const char *fmt, va_list ap)
14624 + if (!printk_killswitch)
14626 + early_vprintk(fmt, ap);
14632 +static inline int forced_early_printk(const char *fmt, va_list ap)
14638 #ifdef CONFIG_PRINTK
14639 DECLARE_WAIT_QUEUE_HEAD(log_wait);
14640 /* the next printk record to read by syslog(READ) or /proc/kmsg */
14641 @@ -1337,6 +1396,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear)
14645 + int attempts = 0;
14647 text = kmalloc(LOG_LINE_MAX + PREFIX_MAX, GFP_KERNEL);
14649 @@ -1348,6 +1408,14 @@ static int syslog_print_all(char __user *buf, int size, bool clear)
14652 enum log_flags prev;
14656 + if (attempts > 10) {
14663 * Find first record that fits, including all following records,
14664 @@ -1363,6 +1431,14 @@ static int syslog_print_all(char __user *buf, int size, bool clear)
14666 idx = log_next(idx);
14669 + if (num_msg > 5) {
14671 + raw_spin_unlock_irq(&logbuf_lock);
14672 + raw_spin_lock_irq(&logbuf_lock);
14673 + if (clear_seq < log_first_seq)
14678 /* move first record forward until length fits into the buffer */
14679 @@ -1376,6 +1452,14 @@ static int syslog_print_all(char __user *buf, int size, bool clear)
14681 idx = log_next(idx);
14684 + if (num_msg > 5) {
14686 + raw_spin_unlock_irq(&logbuf_lock);
14687 + raw_spin_lock_irq(&logbuf_lock);
14688 + if (clear_seq < log_first_seq)
14693 /* last message fitting into this dump */
14694 @@ -1416,6 +1500,7 @@ static int syslog_print_all(char __user *buf, int size, bool clear)
14695 clear_seq = log_next_seq;
14696 clear_idx = log_next_idx;
14699 raw_spin_unlock_irq(&logbuf_lock);
14702 @@ -1569,6 +1654,12 @@ static void call_console_drivers(int level,
14703 if (!console_drivers)
14706 + if (IS_ENABLED(CONFIG_PREEMPT_RT_BASE)) {
14707 + if (in_irq() || in_nmi())
14711 + migrate_disable();
14712 for_each_console(con) {
14713 if (exclusive_console && con != exclusive_console)
14715 @@ -1584,6 +1675,7 @@ static void call_console_drivers(int level,
14717 con->write(con, text, len);
14719 + migrate_enable();
14723 @@ -1781,6 +1873,13 @@ asmlinkage int vprintk_emit(int facility, int level,
14724 /* cpu currently holding logbuf_lock in this function */
14725 static unsigned int logbuf_cpu = UINT_MAX;
14728 + * Fall back to early_printk if a debugging subsystem has
14729 + * killed printk output
14731 + if (unlikely(forced_early_printk(fmt, args)))
14734 if (level == LOGLEVEL_SCHED) {
14735 level = LOGLEVEL_DEFAULT;
14737 @@ -1885,13 +1984,23 @@ asmlinkage int vprintk_emit(int facility, int level,
14739 /* If called from the scheduler, we can not call up(). */
14741 + int may_trylock = 1;
14744 +#ifdef CONFIG_PREEMPT_RT_FULL
14746 + * we can't take a sleeping lock with IRQs or preeption disabled
14747 + * so we can't print in these contexts
14749 + if (!(preempt_count() == 0 && !irqs_disabled()))
14753 * Try to acquire and then immediately release the console
14754 * semaphore. The release will print out buffers and wake up
14755 * /dev/kmsg and syslog() users.
14757 - if (console_trylock())
14758 + if (may_trylock && console_trylock())
14762 @@ -2014,26 +2123,6 @@ DEFINE_PER_CPU(printk_func_t, printk_func);
14764 #endif /* CONFIG_PRINTK */
14766 -#ifdef CONFIG_EARLY_PRINTK
14767 -struct console *early_console;
14769 -asmlinkage __visible void early_printk(const char *fmt, ...)
14775 - if (!early_console)
14778 - va_start(ap, fmt);
14779 - n = vscnprintf(buf, sizeof(buf), fmt, ap);
14782 - early_console->write(early_console, buf, n);
14786 static int __add_preferred_console(char *name, int idx, char *options,
14789 @@ -2303,11 +2392,16 @@ static void console_cont_flush(char *text, size_t size)
14792 len = cont_print_text(text, size);
14793 +#ifdef CONFIG_PREEMPT_RT_FULL
14794 + raw_spin_unlock_irqrestore(&logbuf_lock, flags);
14795 + call_console_drivers(cont.level, NULL, 0, text, len);
14797 raw_spin_unlock(&logbuf_lock);
14798 stop_critical_timings();
14799 call_console_drivers(cont.level, NULL, 0, text, len);
14800 start_critical_timings();
14801 local_irq_restore(flags);
14805 raw_spin_unlock_irqrestore(&logbuf_lock, flags);
14806 @@ -2431,13 +2525,17 @@ void console_unlock(void)
14807 console_idx = log_next(console_idx);
14809 console_prev = msg->flags;
14810 +#ifdef CONFIG_PREEMPT_RT_FULL
14811 + raw_spin_unlock_irqrestore(&logbuf_lock, flags);
14812 + call_console_drivers(level, ext_text, ext_len, text, len);
14814 raw_spin_unlock(&logbuf_lock);
14816 stop_critical_timings(); /* don't trace print latency */
14817 call_console_drivers(level, ext_text, ext_len, text, len);
14818 start_critical_timings();
14819 local_irq_restore(flags);
14822 if (do_cond_resched)
14825 @@ -2489,6 +2587,11 @@ void console_unblank(void)
14829 + if (IS_ENABLED(CONFIG_PREEMPT_RT_BASE)) {
14830 + if (in_irq() || in_nmi())
14835 * console_unblank can no longer be called in interrupt context unless
14836 * oops_in_progress is set to 1..
14837 diff --git a/kernel/ptrace.c b/kernel/ptrace.c
14838 index 49ba7c1ade9d..44f44b47ec07 100644
14839 --- a/kernel/ptrace.c
14840 +++ b/kernel/ptrace.c
14841 @@ -166,7 +166,14 @@ static bool ptrace_freeze_traced(struct task_struct *task)
14843 spin_lock_irq(&task->sighand->siglock);
14844 if (task_is_traced(task) && !__fatal_signal_pending(task)) {
14845 - task->state = __TASK_TRACED;
14846 + unsigned long flags;
14848 + raw_spin_lock_irqsave(&task->pi_lock, flags);
14849 + if (task->state & __TASK_TRACED)
14850 + task->state = __TASK_TRACED;
14852 + task->saved_state = __TASK_TRACED;
14853 + raw_spin_unlock_irqrestore(&task->pi_lock, flags);
14856 spin_unlock_irq(&task->sighand->siglock);
14857 diff --git a/kernel/rcu/rcutorture.c b/kernel/rcu/rcutorture.c
14858 index bf08fee53dc7..eeb8ce4ad7b6 100644
14859 --- a/kernel/rcu/rcutorture.c
14860 +++ b/kernel/rcu/rcutorture.c
14861 @@ -404,6 +404,7 @@ static struct rcu_torture_ops rcu_ops = {
14865 +#ifndef CONFIG_PREEMPT_RT_FULL
14867 * Definitions for rcu_bh torture testing.
14869 @@ -443,6 +444,12 @@ static struct rcu_torture_ops rcu_bh_ops = {
14874 +static struct rcu_torture_ops rcu_bh_ops = {
14875 + .ttype = INVALID_RCU_FLAVOR,
14880 * Don't even think about trying any of these in real life!!!
14881 * The names includes "busted", and they really means it!
14882 diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
14883 index 10f62c6f48e7..dbee19478f09 100644
14884 --- a/kernel/rcu/tree.c
14885 +++ b/kernel/rcu/tree.c
14887 #include <linux/random.h>
14888 #include <linux/trace_events.h>
14889 #include <linux/suspend.h>
14890 +#include <linux/delay.h>
14891 +#include <linux/gfp.h>
14892 +#include <linux/oom.h>
14893 +#include <linux/smpboot.h>
14894 +#include "../time/tick-internal.h"
14898 @@ -260,6 +265,19 @@ void rcu_sched_qs(void)
14899 this_cpu_ptr(&rcu_sched_data), true);
14902 +#ifdef CONFIG_PREEMPT_RT_FULL
14903 +static void rcu_preempt_qs(void);
14905 +void rcu_bh_qs(void)
14907 + unsigned long flags;
14909 + /* Callers to this function, rcu_preempt_qs(), must disable irqs. */
14910 + local_irq_save(flags);
14911 + rcu_preempt_qs();
14912 + local_irq_restore(flags);
14915 void rcu_bh_qs(void)
14917 if (__this_cpu_read(rcu_bh_data.cpu_no_qs.s)) {
14918 @@ -269,6 +287,7 @@ void rcu_bh_qs(void)
14919 __this_cpu_write(rcu_bh_data.cpu_no_qs.b.norm, false);
14924 static DEFINE_PER_CPU(int, rcu_sched_qs_mask);
14926 @@ -449,11 +468,13 @@ EXPORT_SYMBOL_GPL(rcu_batches_started_sched);
14928 * Return the number of RCU BH batches started thus far for debug & stats.
14930 +#ifndef CONFIG_PREEMPT_RT_FULL
14931 unsigned long rcu_batches_started_bh(void)
14933 return rcu_bh_state.gpnum;
14935 EXPORT_SYMBOL_GPL(rcu_batches_started_bh);
14939 * Return the number of RCU batches completed thus far for debug & stats.
14940 @@ -473,6 +494,7 @@ unsigned long rcu_batches_completed_sched(void)
14942 EXPORT_SYMBOL_GPL(rcu_batches_completed_sched);
14944 +#ifndef CONFIG_PREEMPT_RT_FULL
14946 * Return the number of RCU BH batches completed thus far for debug & stats.
14948 @@ -481,6 +503,7 @@ unsigned long rcu_batches_completed_bh(void)
14949 return rcu_bh_state.completed;
14951 EXPORT_SYMBOL_GPL(rcu_batches_completed_bh);
14955 * Return the number of RCU expedited batches completed thus far for
14956 @@ -504,6 +527,7 @@ unsigned long rcu_exp_batches_completed_sched(void)
14958 EXPORT_SYMBOL_GPL(rcu_exp_batches_completed_sched);
14960 +#ifndef CONFIG_PREEMPT_RT_FULL
14962 * Force a quiescent state.
14964 @@ -522,6 +546,13 @@ void rcu_bh_force_quiescent_state(void)
14966 EXPORT_SYMBOL_GPL(rcu_bh_force_quiescent_state);
14969 +void rcu_force_quiescent_state(void)
14972 +EXPORT_SYMBOL_GPL(rcu_force_quiescent_state);
14976 * Force a quiescent state for RCU-sched.
14978 @@ -572,9 +603,11 @@ void rcutorture_get_gp_data(enum rcutorture_type test_type, int *flags,
14982 +#ifndef CONFIG_PREEMPT_RT_FULL
14983 case RCU_BH_FLAVOR:
14984 rsp = &rcu_bh_state;
14987 case RCU_SCHED_FLAVOR:
14988 rsp = &rcu_sched_state;
14990 @@ -3016,18 +3049,17 @@ __rcu_process_callbacks(struct rcu_state *rsp)
14992 * Do RCU core processing for the current CPU.
14994 -static __latent_entropy void rcu_process_callbacks(struct softirq_action *unused)
14995 +static __latent_entropy void rcu_process_callbacks(void)
14997 struct rcu_state *rsp;
14999 if (cpu_is_offline(smp_processor_id()))
15001 - trace_rcu_utilization(TPS("Start RCU core"));
15002 for_each_rcu_flavor(rsp)
15003 __rcu_process_callbacks(rsp);
15004 - trace_rcu_utilization(TPS("End RCU core"));
15007 +static DEFINE_PER_CPU(struct task_struct *, rcu_cpu_kthread_task);
15009 * Schedule RCU callback invocation. If the specified type of RCU
15010 * does not support RCU priority boosting, just do a direct call,
15011 @@ -3039,19 +3071,106 @@ static void invoke_rcu_callbacks(struct rcu_state *rsp, struct rcu_data *rdp)
15013 if (unlikely(!READ_ONCE(rcu_scheduler_fully_active)))
15015 - if (likely(!rsp->boost)) {
15016 - rcu_do_batch(rsp, rdp);
15019 - invoke_rcu_callbacks_kthread();
15020 + rcu_do_batch(rsp, rdp);
15023 +static void rcu_wake_cond(struct task_struct *t, int status)
15026 + * If the thread is yielding, only wake it when this
15027 + * is invoked from idle
15029 + if (t && (status != RCU_KTHREAD_YIELDING || is_idle_task(current)))
15030 + wake_up_process(t);
15034 + * Wake up this CPU's rcuc kthread to do RCU core processing.
15036 static void invoke_rcu_core(void)
15038 - if (cpu_online(smp_processor_id()))
15039 - raise_softirq(RCU_SOFTIRQ);
15040 + unsigned long flags;
15041 + struct task_struct *t;
15043 + if (!cpu_online(smp_processor_id()))
15045 + local_irq_save(flags);
15046 + __this_cpu_write(rcu_cpu_has_work, 1);
15047 + t = __this_cpu_read(rcu_cpu_kthread_task);
15048 + if (t != NULL && current != t)
15049 + rcu_wake_cond(t, __this_cpu_read(rcu_cpu_kthread_status));
15050 + local_irq_restore(flags);
15053 +static void rcu_cpu_kthread_park(unsigned int cpu)
15055 + per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU;
15058 +static int rcu_cpu_kthread_should_run(unsigned int cpu)
15060 + return __this_cpu_read(rcu_cpu_has_work);
15064 + * Per-CPU kernel thread that invokes RCU callbacks. This replaces the
15065 + * RCU softirq used in flavors and configurations of RCU that do not
15066 + * support RCU priority boosting.
15068 +static void rcu_cpu_kthread(unsigned int cpu)
15070 + unsigned int *statusp = this_cpu_ptr(&rcu_cpu_kthread_status);
15071 + char work, *workp = this_cpu_ptr(&rcu_cpu_has_work);
15074 + for (spincnt = 0; spincnt < 10; spincnt++) {
15075 + trace_rcu_utilization(TPS("Start CPU kthread@rcu_wait"));
15076 + local_bh_disable();
15077 + *statusp = RCU_KTHREAD_RUNNING;
15078 + this_cpu_inc(rcu_cpu_kthread_loops);
15079 + local_irq_disable();
15082 + local_irq_enable();
15084 + rcu_process_callbacks();
15085 + local_bh_enable();
15086 + if (*workp == 0) {
15087 + trace_rcu_utilization(TPS("End CPU kthread@rcu_wait"));
15088 + *statusp = RCU_KTHREAD_WAITING;
15092 + *statusp = RCU_KTHREAD_YIELDING;
15093 + trace_rcu_utilization(TPS("Start CPU kthread@rcu_yield"));
15094 + schedule_timeout_interruptible(2);
15095 + trace_rcu_utilization(TPS("End CPU kthread@rcu_yield"));
15096 + *statusp = RCU_KTHREAD_WAITING;
15099 +static struct smp_hotplug_thread rcu_cpu_thread_spec = {
15100 + .store = &rcu_cpu_kthread_task,
15101 + .thread_should_run = rcu_cpu_kthread_should_run,
15102 + .thread_fn = rcu_cpu_kthread,
15103 + .thread_comm = "rcuc/%u",
15104 + .setup = rcu_cpu_kthread_setup,
15105 + .park = rcu_cpu_kthread_park,
15109 + * Spawn per-CPU RCU core processing kthreads.
15111 +static int __init rcu_spawn_core_kthreads(void)
15115 + for_each_possible_cpu(cpu)
15116 + per_cpu(rcu_cpu_has_work, cpu) = 0;
15117 + BUG_ON(smpboot_register_percpu_thread(&rcu_cpu_thread_spec));
15120 +early_initcall(rcu_spawn_core_kthreads);
15123 * Handle any core-RCU processing required by a call_rcu() invocation.
15125 @@ -3195,6 +3314,7 @@ void call_rcu_sched(struct rcu_head *head, rcu_callback_t func)
15127 EXPORT_SYMBOL_GPL(call_rcu_sched);
15129 +#ifndef CONFIG_PREEMPT_RT_FULL
15131 * Queue an RCU callback for invocation after a quicker grace period.
15133 @@ -3203,6 +3323,7 @@ void call_rcu_bh(struct rcu_head *head, rcu_callback_t func)
15134 __call_rcu(head, func, &rcu_bh_state, -1, 0);
15136 EXPORT_SYMBOL_GPL(call_rcu_bh);
15140 * Queue an RCU callback for lazy invocation after a grace period.
15141 @@ -3294,6 +3415,7 @@ void synchronize_sched(void)
15143 EXPORT_SYMBOL_GPL(synchronize_sched);
15145 +#ifndef CONFIG_PREEMPT_RT_FULL
15147 * synchronize_rcu_bh - wait until an rcu_bh grace period has elapsed.
15149 @@ -3320,6 +3442,7 @@ void synchronize_rcu_bh(void)
15150 wait_rcu_gp(call_rcu_bh);
15152 EXPORT_SYMBOL_GPL(synchronize_rcu_bh);
15156 * get_state_synchronize_rcu - Snapshot current RCU state
15157 @@ -3698,6 +3821,7 @@ static void _rcu_barrier(struct rcu_state *rsp)
15158 mutex_unlock(&rsp->barrier_mutex);
15161 +#ifndef CONFIG_PREEMPT_RT_FULL
15163 * rcu_barrier_bh - Wait until all in-flight call_rcu_bh() callbacks complete.
15165 @@ -3706,6 +3830,7 @@ void rcu_barrier_bh(void)
15166 _rcu_barrier(&rcu_bh_state);
15168 EXPORT_SYMBOL_GPL(rcu_barrier_bh);
15172 * rcu_barrier_sched - Wait for in-flight call_rcu_sched() callbacks.
15173 @@ -4227,12 +4352,13 @@ void __init rcu_init(void)
15175 rcu_bootup_announce();
15176 rcu_init_geometry();
15177 +#ifndef CONFIG_PREEMPT_RT_FULL
15178 rcu_init_one(&rcu_bh_state);
15180 rcu_init_one(&rcu_sched_state);
15182 rcu_dump_rcu_node_tree(&rcu_sched_state);
15183 __rcu_init_preempt();
15184 - open_softirq(RCU_SOFTIRQ, rcu_process_callbacks);
15187 * We don't need protection against CPU-hotplug here because
15188 diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
15189 index e99a5234d9ed..958ac107062c 100644
15190 --- a/kernel/rcu/tree.h
15191 +++ b/kernel/rcu/tree.h
15192 @@ -588,18 +588,18 @@ extern struct list_head rcu_struct_flavors;
15194 extern struct rcu_state rcu_sched_state;
15196 +#ifndef CONFIG_PREEMPT_RT_FULL
15197 extern struct rcu_state rcu_bh_state;
15200 #ifdef CONFIG_PREEMPT_RCU
15201 extern struct rcu_state rcu_preempt_state;
15202 #endif /* #ifdef CONFIG_PREEMPT_RCU */
15204 -#ifdef CONFIG_RCU_BOOST
15205 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_status);
15206 DECLARE_PER_CPU(int, rcu_cpu_kthread_cpu);
15207 DECLARE_PER_CPU(unsigned int, rcu_cpu_kthread_loops);
15208 DECLARE_PER_CPU(char, rcu_cpu_has_work);
15209 -#endif /* #ifdef CONFIG_RCU_BOOST */
15211 #ifndef RCU_TREE_NONCORE
15213 @@ -619,10 +619,9 @@ void call_rcu(struct rcu_head *head, rcu_callback_t func);
15214 static void __init __rcu_init_preempt(void);
15215 static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags);
15216 static void rcu_preempt_boost_start_gp(struct rcu_node *rnp);
15217 -static void invoke_rcu_callbacks_kthread(void);
15218 static bool rcu_is_callbacks_kthread(void);
15219 +static void rcu_cpu_kthread_setup(unsigned int cpu);
15220 #ifdef CONFIG_RCU_BOOST
15221 -static void rcu_preempt_do_callbacks(void);
15222 static int rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
15223 struct rcu_node *rnp);
15224 #endif /* #ifdef CONFIG_RCU_BOOST */
15225 diff --git a/kernel/rcu/tree_plugin.h b/kernel/rcu/tree_plugin.h
15226 index 56583e764ebf..7c656f8e192f 100644
15227 --- a/kernel/rcu/tree_plugin.h
15228 +++ b/kernel/rcu/tree_plugin.h
15229 @@ -24,25 +24,10 @@
15230 * Paul E. McKenney <paulmck@linux.vnet.ibm.com>
15233 -#include <linux/delay.h>
15234 -#include <linux/gfp.h>
15235 -#include <linux/oom.h>
15236 -#include <linux/smpboot.h>
15237 -#include "../time/tick-internal.h"
15239 #ifdef CONFIG_RCU_BOOST
15241 #include "../locking/rtmutex_common.h"
15244 - * Control variables for per-CPU and per-rcu_node kthreads. These
15245 - * handle all flavors of RCU.
15247 -static DEFINE_PER_CPU(struct task_struct *, rcu_cpu_kthread_task);
15248 -DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_status);
15249 -DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_loops);
15250 -DEFINE_PER_CPU(char, rcu_cpu_has_work);
15252 #else /* #ifdef CONFIG_RCU_BOOST */
15255 @@ -55,6 +40,14 @@ DEFINE_PER_CPU(char, rcu_cpu_has_work);
15257 #endif /* #else #ifdef CONFIG_RCU_BOOST */
15260 + * Control variables for per-CPU and per-rcu_node kthreads. These
15261 + * handle all flavors of RCU.
15263 +DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_status);
15264 +DEFINE_PER_CPU(unsigned int, rcu_cpu_kthread_loops);
15265 +DEFINE_PER_CPU(char, rcu_cpu_has_work);
15267 #ifdef CONFIG_RCU_NOCB_CPU
15268 static cpumask_var_t rcu_nocb_mask; /* CPUs to have callbacks offloaded. */
15269 static bool have_rcu_nocb_mask; /* Was rcu_nocb_mask allocated? */
15270 @@ -426,7 +419,7 @@ void rcu_read_unlock_special(struct task_struct *t)
15273 /* Hardware IRQ handlers cannot block, complain if they get here. */
15274 - if (in_irq() || in_serving_softirq()) {
15275 + if (preempt_count() & (HARDIRQ_MASK | SOFTIRQ_OFFSET)) {
15276 lockdep_rcu_suspicious(__FILE__, __LINE__,
15277 "rcu_read_unlock() from irq or softirq with blocking in critical section!!!\n");
15278 pr_alert("->rcu_read_unlock_special: %#x (b: %d, enq: %d nq: %d)\n",
15279 @@ -632,15 +625,6 @@ static void rcu_preempt_check_callbacks(void)
15280 t->rcu_read_unlock_special.b.need_qs = true;
15283 -#ifdef CONFIG_RCU_BOOST
15285 -static void rcu_preempt_do_callbacks(void)
15287 - rcu_do_batch(rcu_state_p, this_cpu_ptr(rcu_data_p));
15290 -#endif /* #ifdef CONFIG_RCU_BOOST */
15293 * Queue a preemptible-RCU callback for invocation after a grace period.
15295 @@ -829,6 +813,19 @@ void exit_rcu(void)
15297 #endif /* #else #ifdef CONFIG_PREEMPT_RCU */
15300 + * If boosting, set rcuc kthreads to realtime priority.
15302 +static void rcu_cpu_kthread_setup(unsigned int cpu)
15304 +#ifdef CONFIG_RCU_BOOST
15305 + struct sched_param sp;
15307 + sp.sched_priority = kthread_prio;
15308 + sched_setscheduler_nocheck(current, SCHED_FIFO, &sp);
15309 +#endif /* #ifdef CONFIG_RCU_BOOST */
15312 #ifdef CONFIG_RCU_BOOST
15314 #include "../locking/rtmutex_common.h"
15315 @@ -860,16 +857,6 @@ static void rcu_initiate_boost_trace(struct rcu_node *rnp)
15317 #endif /* #else #ifdef CONFIG_RCU_TRACE */
15319 -static void rcu_wake_cond(struct task_struct *t, int status)
15322 - * If the thread is yielding, only wake it when this
15323 - * is invoked from idle
15325 - if (status != RCU_KTHREAD_YIELDING || is_idle_task(current))
15326 - wake_up_process(t);
15330 * Carry out RCU priority boosting on the task indicated by ->exp_tasks
15331 * or ->boost_tasks, advancing the pointer to the next task in the
15332 @@ -1013,23 +1000,6 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
15336 - * Wake up the per-CPU kthread to invoke RCU callbacks.
15338 -static void invoke_rcu_callbacks_kthread(void)
15340 - unsigned long flags;
15342 - local_irq_save(flags);
15343 - __this_cpu_write(rcu_cpu_has_work, 1);
15344 - if (__this_cpu_read(rcu_cpu_kthread_task) != NULL &&
15345 - current != __this_cpu_read(rcu_cpu_kthread_task)) {
15346 - rcu_wake_cond(__this_cpu_read(rcu_cpu_kthread_task),
15347 - __this_cpu_read(rcu_cpu_kthread_status));
15349 - local_irq_restore(flags);
15353 * Is the current CPU running the RCU-callbacks kthread?
15354 * Caller must have preemption disabled.
15356 @@ -1083,67 +1053,6 @@ static int rcu_spawn_one_boost_kthread(struct rcu_state *rsp,
15360 -static void rcu_kthread_do_work(void)
15362 - rcu_do_batch(&rcu_sched_state, this_cpu_ptr(&rcu_sched_data));
15363 - rcu_do_batch(&rcu_bh_state, this_cpu_ptr(&rcu_bh_data));
15364 - rcu_preempt_do_callbacks();
15367 -static void rcu_cpu_kthread_setup(unsigned int cpu)
15369 - struct sched_param sp;
15371 - sp.sched_priority = kthread_prio;
15372 - sched_setscheduler_nocheck(current, SCHED_FIFO, &sp);
15375 -static void rcu_cpu_kthread_park(unsigned int cpu)
15377 - per_cpu(rcu_cpu_kthread_status, cpu) = RCU_KTHREAD_OFFCPU;
15380 -static int rcu_cpu_kthread_should_run(unsigned int cpu)
15382 - return __this_cpu_read(rcu_cpu_has_work);
15386 - * Per-CPU kernel thread that invokes RCU callbacks. This replaces the
15387 - * RCU softirq used in flavors and configurations of RCU that do not
15388 - * support RCU priority boosting.
15390 -static void rcu_cpu_kthread(unsigned int cpu)
15392 - unsigned int *statusp = this_cpu_ptr(&rcu_cpu_kthread_status);
15393 - char work, *workp = this_cpu_ptr(&rcu_cpu_has_work);
15396 - for (spincnt = 0; spincnt < 10; spincnt++) {
15397 - trace_rcu_utilization(TPS("Start CPU kthread@rcu_wait"));
15398 - local_bh_disable();
15399 - *statusp = RCU_KTHREAD_RUNNING;
15400 - this_cpu_inc(rcu_cpu_kthread_loops);
15401 - local_irq_disable();
15404 - local_irq_enable();
15406 - rcu_kthread_do_work();
15407 - local_bh_enable();
15408 - if (*workp == 0) {
15409 - trace_rcu_utilization(TPS("End CPU kthread@rcu_wait"));
15410 - *statusp = RCU_KTHREAD_WAITING;
15414 - *statusp = RCU_KTHREAD_YIELDING;
15415 - trace_rcu_utilization(TPS("Start CPU kthread@rcu_yield"));
15416 - schedule_timeout_interruptible(2);
15417 - trace_rcu_utilization(TPS("End CPU kthread@rcu_yield"));
15418 - *statusp = RCU_KTHREAD_WAITING;
15422 * Set the per-rcu_node kthread's affinity to cover all CPUs that are
15423 * served by the rcu_node in question. The CPU hotplug lock is still
15424 @@ -1174,26 +1083,12 @@ static void rcu_boost_kthread_setaffinity(struct rcu_node *rnp, int outgoingcpu)
15425 free_cpumask_var(cm);
15428 -static struct smp_hotplug_thread rcu_cpu_thread_spec = {
15429 - .store = &rcu_cpu_kthread_task,
15430 - .thread_should_run = rcu_cpu_kthread_should_run,
15431 - .thread_fn = rcu_cpu_kthread,
15432 - .thread_comm = "rcuc/%u",
15433 - .setup = rcu_cpu_kthread_setup,
15434 - .park = rcu_cpu_kthread_park,
15438 * Spawn boost kthreads -- called as soon as the scheduler is running.
15440 static void __init rcu_spawn_boost_kthreads(void)
15442 struct rcu_node *rnp;
15445 - for_each_possible_cpu(cpu)
15446 - per_cpu(rcu_cpu_has_work, cpu) = 0;
15447 - BUG_ON(smpboot_register_percpu_thread(&rcu_cpu_thread_spec));
15448 rcu_for_each_leaf_node(rcu_state_p, rnp)
15449 (void)rcu_spawn_one_boost_kthread(rcu_state_p, rnp);
15451 @@ -1216,11 +1111,6 @@ static void rcu_initiate_boost(struct rcu_node *rnp, unsigned long flags)
15452 raw_spin_unlock_irqrestore_rcu_node(rnp, flags);
15455 -static void invoke_rcu_callbacks_kthread(void)
15460 static bool rcu_is_callbacks_kthread(void)
15463 @@ -1244,7 +1134,7 @@ static void rcu_prepare_kthreads(int cpu)
15465 #endif /* #else #ifdef CONFIG_RCU_BOOST */
15467 -#if !defined(CONFIG_RCU_FAST_NO_HZ)
15468 +#if !defined(CONFIG_RCU_FAST_NO_HZ) || defined(CONFIG_PREEMPT_RT_FULL)
15471 * Check to see if any future RCU-related work will need to be done
15472 @@ -1261,7 +1151,9 @@ int rcu_needs_cpu(u64 basemono, u64 *nextevt)
15473 return IS_ENABLED(CONFIG_RCU_NOCB_CPU_ALL)
15474 ? 0 : rcu_cpu_has_callbacks(NULL);
15476 +#endif /* !defined(CONFIG_RCU_FAST_NO_HZ) || defined(CONFIG_PREEMPT_RT_FULL) */
15478 +#if !defined(CONFIG_RCU_FAST_NO_HZ)
15480 * Because we do not have RCU_FAST_NO_HZ, don't bother cleaning up
15482 @@ -1357,6 +1249,8 @@ static bool __maybe_unused rcu_try_advance_all_cbs(void)
15486 +#ifndef CONFIG_PREEMPT_RT_FULL
15489 * Allow the CPU to enter dyntick-idle mode unless it has callbacks ready
15490 * to invoke. If the CPU has callbacks, try to advance them. Tell the
15491 @@ -1402,6 +1296,7 @@ int rcu_needs_cpu(u64 basemono, u64 *nextevt)
15492 *nextevt = basemono + dj * TICK_NSEC;
15495 +#endif /* #ifndef CONFIG_PREEMPT_RT_FULL */
15498 * Prepare a CPU for idle from an RCU perspective. The first major task
15499 diff --git a/kernel/rcu/update.c b/kernel/rcu/update.c
15500 index 4f6db7e6a117..ee02e1e1b3e5 100644
15501 --- a/kernel/rcu/update.c
15502 +++ b/kernel/rcu/update.c
15504 #ifndef CONFIG_TINY_RCU
15505 module_param(rcu_expedited, int, 0);
15506 module_param(rcu_normal, int, 0);
15507 -static int rcu_normal_after_boot;
15508 +static int rcu_normal_after_boot = IS_ENABLED(CONFIG_PREEMPT_RT_FULL);
15509 module_param(rcu_normal_after_boot, int, 0);
15510 #endif /* #ifndef CONFIG_TINY_RCU */
15512 @@ -132,8 +132,7 @@ bool rcu_gp_is_normal(void)
15514 EXPORT_SYMBOL_GPL(rcu_gp_is_normal);
15516 -static atomic_t rcu_expedited_nesting =
15517 - ATOMIC_INIT(IS_ENABLED(CONFIG_RCU_EXPEDITE_BOOT) ? 1 : 0);
15518 +static atomic_t rcu_expedited_nesting = ATOMIC_INIT(1);
15521 * Should normal grace-period primitives be expedited? Intended for
15522 @@ -182,8 +181,7 @@ EXPORT_SYMBOL_GPL(rcu_unexpedite_gp);
15524 void rcu_end_inkernel_boot(void)
15526 - if (IS_ENABLED(CONFIG_RCU_EXPEDITE_BOOT))
15527 - rcu_unexpedite_gp();
15528 + rcu_unexpedite_gp();
15529 if (rcu_normal_after_boot)
15530 WRITE_ONCE(rcu_normal, 1);
15532 @@ -298,6 +296,7 @@ int rcu_read_lock_held(void)
15534 EXPORT_SYMBOL_GPL(rcu_read_lock_held);
15536 +#ifndef CONFIG_PREEMPT_RT_FULL
15538 * rcu_read_lock_bh_held() - might we be in RCU-bh read-side critical section?
15540 @@ -324,6 +323,7 @@ int rcu_read_lock_bh_held(void)
15541 return in_softirq() || irqs_disabled();
15543 EXPORT_SYMBOL_GPL(rcu_read_lock_bh_held);
15546 #endif /* #ifdef CONFIG_DEBUG_LOCK_ALLOC */
15548 diff --git a/kernel/sched/Makefile b/kernel/sched/Makefile
15549 index 5e59b832ae2b..7337a7f60e3f 100644
15550 --- a/kernel/sched/Makefile
15551 +++ b/kernel/sched/Makefile
15552 @@ -17,7 +17,7 @@ endif
15554 obj-y += core.o loadavg.o clock.o cputime.o
15555 obj-y += idle_task.o fair.o rt.o deadline.o stop_task.o
15556 -obj-y += wait.o swait.o completion.o idle.o
15557 +obj-y += wait.o swait.o swork.o completion.o idle.o
15558 obj-$(CONFIG_SMP) += cpupri.o cpudeadline.o
15559 obj-$(CONFIG_SCHED_AUTOGROUP) += auto_group.o
15560 obj-$(CONFIG_SCHEDSTATS) += stats.o
15561 diff --git a/kernel/sched/completion.c b/kernel/sched/completion.c
15562 index 8d0f35debf35..b62cf6400fe0 100644
15563 --- a/kernel/sched/completion.c
15564 +++ b/kernel/sched/completion.c
15565 @@ -30,10 +30,10 @@ void complete(struct completion *x)
15567 unsigned long flags;
15569 - spin_lock_irqsave(&x->wait.lock, flags);
15570 + raw_spin_lock_irqsave(&x->wait.lock, flags);
15572 - __wake_up_locked(&x->wait, TASK_NORMAL, 1);
15573 - spin_unlock_irqrestore(&x->wait.lock, flags);
15574 + swake_up_locked(&x->wait);
15575 + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
15577 EXPORT_SYMBOL(complete);
15579 @@ -50,10 +50,10 @@ void complete_all(struct completion *x)
15581 unsigned long flags;
15583 - spin_lock_irqsave(&x->wait.lock, flags);
15584 + raw_spin_lock_irqsave(&x->wait.lock, flags);
15585 x->done += UINT_MAX/2;
15586 - __wake_up_locked(&x->wait, TASK_NORMAL, 0);
15587 - spin_unlock_irqrestore(&x->wait.lock, flags);
15588 + swake_up_all_locked(&x->wait);
15589 + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
15591 EXPORT_SYMBOL(complete_all);
15593 @@ -62,20 +62,20 @@ do_wait_for_common(struct completion *x,
15594 long (*action)(long), long timeout, int state)
15597 - DECLARE_WAITQUEUE(wait, current);
15598 + DECLARE_SWAITQUEUE(wait);
15600 - __add_wait_queue_tail_exclusive(&x->wait, &wait);
15601 + __prepare_to_swait(&x->wait, &wait);
15603 if (signal_pending_state(state, current)) {
15604 timeout = -ERESTARTSYS;
15607 __set_current_state(state);
15608 - spin_unlock_irq(&x->wait.lock);
15609 + raw_spin_unlock_irq(&x->wait.lock);
15610 timeout = action(timeout);
15611 - spin_lock_irq(&x->wait.lock);
15612 + raw_spin_lock_irq(&x->wait.lock);
15613 } while (!x->done && timeout);
15614 - __remove_wait_queue(&x->wait, &wait);
15615 + __finish_swait(&x->wait, &wait);
15619 @@ -89,9 +89,9 @@ __wait_for_common(struct completion *x,
15623 - spin_lock_irq(&x->wait.lock);
15624 + raw_spin_lock_irq(&x->wait.lock);
15625 timeout = do_wait_for_common(x, action, timeout, state);
15626 - spin_unlock_irq(&x->wait.lock);
15627 + raw_spin_unlock_irq(&x->wait.lock);
15631 @@ -277,12 +277,12 @@ bool try_wait_for_completion(struct completion *x)
15632 if (!READ_ONCE(x->done))
15635 - spin_lock_irqsave(&x->wait.lock, flags);
15636 + raw_spin_lock_irqsave(&x->wait.lock, flags);
15641 - spin_unlock_irqrestore(&x->wait.lock, flags);
15642 + raw_spin_unlock_irqrestore(&x->wait.lock, flags);
15645 EXPORT_SYMBOL(try_wait_for_completion);
15646 @@ -311,7 +311,7 @@ bool completion_done(struct completion *x)
15647 * after it's acquired the lock.
15650 - spin_unlock_wait(&x->wait.lock);
15651 + raw_spin_unlock_wait(&x->wait.lock);
15654 EXPORT_SYMBOL(completion_done);
15655 diff --git a/kernel/sched/core.c b/kernel/sched/core.c
15656 index 154fd689fe02..a6aa5801b21e 100644
15657 --- a/kernel/sched/core.c
15658 +++ b/kernel/sched/core.c
15659 @@ -129,7 +129,11 @@ const_debug unsigned int sysctl_sched_features =
15660 * Number of tasks to iterate in a single balance run.
15661 * Limited because this is done with IRQs disabled.
15663 +#ifndef CONFIG_PREEMPT_RT_FULL
15664 const_debug unsigned int sysctl_sched_nr_migrate = 32;
15666 +const_debug unsigned int sysctl_sched_nr_migrate = 8;
15670 * period over which we average the RT time consumption, measured
15671 @@ -345,6 +349,7 @@ static void init_rq_hrtick(struct rq *rq)
15673 hrtimer_init(&rq->hrtick_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
15674 rq->hrtick_timer.function = hrtick;
15675 + rq->hrtick_timer.irqsafe = 1;
15677 #else /* CONFIG_SCHED_HRTICK */
15678 static inline void hrtick_clear(struct rq *rq)
15679 @@ -449,7 +454,7 @@ void wake_q_add(struct wake_q_head *head, struct task_struct *task)
15680 head->lastp = &node->next;
15683 -void wake_up_q(struct wake_q_head *head)
15684 +void __wake_up_q(struct wake_q_head *head, bool sleeper)
15686 struct wake_q_node *node = head->first;
15688 @@ -466,7 +471,10 @@ void wake_up_q(struct wake_q_head *head)
15689 * wake_up_process() implies a wmb() to pair with the queueing
15690 * in wake_q_add() so as not to miss wakeups.
15692 - wake_up_process(task);
15694 + wake_up_lock_sleeper(task);
15696 + wake_up_process(task);
15697 put_task_struct(task);
15700 @@ -502,6 +510,38 @@ void resched_curr(struct rq *rq)
15701 trace_sched_wake_idle_without_ipi(cpu);
15704 +#ifdef CONFIG_PREEMPT_LAZY
15705 +void resched_curr_lazy(struct rq *rq)
15707 + struct task_struct *curr = rq->curr;
15710 + if (!sched_feat(PREEMPT_LAZY)) {
15711 + resched_curr(rq);
15715 + lockdep_assert_held(&rq->lock);
15717 + if (test_tsk_need_resched(curr))
15720 + if (test_tsk_need_resched_lazy(curr))
15723 + set_tsk_need_resched_lazy(curr);
15725 + cpu = cpu_of(rq);
15726 + if (cpu == smp_processor_id())
15729 + /* NEED_RESCHED_LAZY must be visible before we test polling */
15731 + if (!tsk_is_polling(curr))
15732 + smp_send_reschedule(cpu);
15736 void resched_cpu(int cpu)
15738 struct rq *rq = cpu_rq(cpu);
15739 @@ -525,11 +565,14 @@ void resched_cpu(int cpu)
15741 int get_nohz_timer_target(void)
15743 - int i, cpu = smp_processor_id();
15745 struct sched_domain *sd;
15747 + preempt_disable_rt();
15748 + cpu = smp_processor_id();
15750 if (!idle_cpu(cpu) && is_housekeeping_cpu(cpu))
15752 + goto preempt_en_rt;
15755 for_each_domain(cpu, sd) {
15756 @@ -548,6 +591,8 @@ int get_nohz_timer_target(void)
15757 cpu = housekeeping_any_cpu();
15761 + preempt_enable_rt();
15765 @@ -1100,6 +1145,11 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
15767 lockdep_assert_held(&p->pi_lock);
15769 + if (__migrate_disabled(p)) {
15770 + cpumask_copy(&p->cpus_allowed, new_mask);
15774 queued = task_on_rq_queued(p);
15775 running = task_current(rq, p);
15777 @@ -1122,6 +1172,84 @@ void do_set_cpus_allowed(struct task_struct *p, const struct cpumask *new_mask)
15778 set_curr_task(rq, p);
15781 +static DEFINE_PER_CPU(struct cpumask, sched_cpumasks);
15782 +static DEFINE_MUTEX(sched_down_mutex);
15783 +static cpumask_t sched_down_cpumask;
15785 +void tell_sched_cpu_down_begin(int cpu)
15787 + mutex_lock(&sched_down_mutex);
15788 + cpumask_set_cpu(cpu, &sched_down_cpumask);
15789 + mutex_unlock(&sched_down_mutex);
15792 +void tell_sched_cpu_down_done(int cpu)
15794 + mutex_lock(&sched_down_mutex);
15795 + cpumask_clear_cpu(cpu, &sched_down_cpumask);
15796 + mutex_unlock(&sched_down_mutex);
15800 + * migrate_me - try to move the current task off this cpu
15802 + * Used by the pin_current_cpu() code to try to get tasks
15803 + * to move off the current CPU as it is going down.
15804 + * It will only move the task if the task isn't pinned to
15805 + * the CPU (with migrate_disable, affinity or NO_SETAFFINITY)
15806 + * and the task has to be in a RUNNING state. Otherwise the
15807 + * movement of the task will wake it up (change its state
15808 + * to running) when the task did not expect it.
15810 + * Returns 1 if it succeeded in moving the current task
15813 +int migrate_me(void)
15815 + struct task_struct *p = current;
15816 + struct migration_arg arg;
15817 + struct cpumask *cpumask;
15818 + struct cpumask *mask;
15819 + unsigned int dest_cpu;
15820 + struct rq_flags rf;
15824 + * We can not migrate tasks bounded to a CPU or tasks not
15825 + * running. The movement of the task will wake it up.
15827 + if (p->flags & PF_NO_SETAFFINITY || p->state)
15830 + mutex_lock(&sched_down_mutex);
15831 + rq = task_rq_lock(p, &rf);
15833 + cpumask = this_cpu_ptr(&sched_cpumasks);
15834 + mask = &p->cpus_allowed;
15836 + cpumask_andnot(cpumask, mask, &sched_down_cpumask);
15838 + if (!cpumask_weight(cpumask)) {
15839 + /* It's only on this CPU? */
15840 + task_rq_unlock(rq, p, &rf);
15841 + mutex_unlock(&sched_down_mutex);
15845 + dest_cpu = cpumask_any_and(cpu_active_mask, cpumask);
15848 + arg.dest_cpu = dest_cpu;
15850 + task_rq_unlock(rq, p, &rf);
15852 + stop_one_cpu(cpu_of(rq), migration_cpu_stop, &arg);
15853 + tlb_migrate_finish(p->mm);
15854 + mutex_unlock(&sched_down_mutex);
15860 * Change a given task's CPU affinity. Migrate the thread to a
15861 * proper CPU and schedule it away if the CPU it's executing on
15862 @@ -1179,7 +1307,7 @@ static int __set_cpus_allowed_ptr(struct task_struct *p,
15865 /* Can the task run on the task's current CPU? If so, we're done */
15866 - if (cpumask_test_cpu(task_cpu(p), new_mask))
15867 + if (cpumask_test_cpu(task_cpu(p), new_mask) || __migrate_disabled(p))
15870 dest_cpu = cpumask_any_and(cpu_valid_mask, new_mask);
15871 @@ -1366,6 +1494,18 @@ int migrate_swap(struct task_struct *cur, struct task_struct *p)
15875 +static bool check_task_state(struct task_struct *p, long match_state)
15877 + bool match = false;
15879 + raw_spin_lock_irq(&p->pi_lock);
15880 + if (p->state == match_state || p->saved_state == match_state)
15882 + raw_spin_unlock_irq(&p->pi_lock);
15888 * wait_task_inactive - wait for a thread to unschedule.
15890 @@ -1410,7 +1550,7 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
15891 * is actually now running somewhere else!
15893 while (task_running(rq, p)) {
15894 - if (match_state && unlikely(p->state != match_state))
15895 + if (match_state && !check_task_state(p, match_state))
15899 @@ -1425,7 +1565,8 @@ unsigned long wait_task_inactive(struct task_struct *p, long match_state)
15900 running = task_running(rq, p);
15901 queued = task_on_rq_queued(p);
15903 - if (!match_state || p->state == match_state)
15904 + if (!match_state || p->state == match_state ||
15905 + p->saved_state == match_state)
15906 ncsw = p->nvcsw | LONG_MIN; /* sets MSB */
15907 task_rq_unlock(rq, p, &rf);
15909 @@ -1680,10 +1821,6 @@ static inline void ttwu_activate(struct rq *rq, struct task_struct *p, int en_fl
15911 activate_task(rq, p, en_flags);
15912 p->on_rq = TASK_ON_RQ_QUEUED;
15914 - /* if a worker is waking up, notify workqueue */
15915 - if (p->flags & PF_WQ_WORKER)
15916 - wq_worker_waking_up(p, cpu_of(rq));
15920 @@ -2018,8 +2155,27 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
15922 smp_mb__before_spinlock();
15923 raw_spin_lock_irqsave(&p->pi_lock, flags);
15924 - if (!(p->state & state))
15925 + if (!(p->state & state)) {
15927 + * The task might be running due to a spinlock sleeper
15928 + * wakeup. Check the saved state and set it to running
15929 + * if the wakeup condition is true.
15931 + if (!(wake_flags & WF_LOCK_SLEEPER)) {
15932 + if (p->saved_state & state) {
15933 + p->saved_state = TASK_RUNNING;
15941 + * If this is a regular wakeup, then we can unconditionally
15942 + * clear the saved state of a "lock sleeper".
15944 + if (!(wake_flags & WF_LOCK_SLEEPER))
15945 + p->saved_state = TASK_RUNNING;
15947 trace_sched_waking(p);
15949 @@ -2102,53 +2258,6 @@ try_to_wake_up(struct task_struct *p, unsigned int state, int wake_flags)
15953 - * try_to_wake_up_local - try to wake up a local task with rq lock held
15954 - * @p: the thread to be awakened
15955 - * @cookie: context's cookie for pinning
15957 - * Put @p on the run-queue if it's not already there. The caller must
15958 - * ensure that this_rq() is locked, @p is bound to this_rq() and not
15959 - * the current task.
15961 -static void try_to_wake_up_local(struct task_struct *p, struct pin_cookie cookie)
15963 - struct rq *rq = task_rq(p);
15965 - if (WARN_ON_ONCE(rq != this_rq()) ||
15966 - WARN_ON_ONCE(p == current))
15969 - lockdep_assert_held(&rq->lock);
15971 - if (!raw_spin_trylock(&p->pi_lock)) {
15973 - * This is OK, because current is on_cpu, which avoids it being
15974 - * picked for load-balance and preemption/IRQs are still
15975 - * disabled avoiding further scheduler activity on it and we've
15976 - * not yet picked a replacement task.
15978 - lockdep_unpin_lock(&rq->lock, cookie);
15979 - raw_spin_unlock(&rq->lock);
15980 - raw_spin_lock(&p->pi_lock);
15981 - raw_spin_lock(&rq->lock);
15982 - lockdep_repin_lock(&rq->lock, cookie);
15985 - if (!(p->state & TASK_NORMAL))
15988 - trace_sched_waking(p);
15990 - if (!task_on_rq_queued(p))
15991 - ttwu_activate(rq, p, ENQUEUE_WAKEUP);
15993 - ttwu_do_wakeup(rq, p, 0, cookie);
15994 - ttwu_stat(p, smp_processor_id(), 0);
15996 - raw_spin_unlock(&p->pi_lock);
16000 * wake_up_process - Wake up a specific process
16001 * @p: The process to be woken up.
16003 @@ -2166,6 +2275,18 @@ int wake_up_process(struct task_struct *p)
16005 EXPORT_SYMBOL(wake_up_process);
16008 + * wake_up_lock_sleeper - Wake up a specific process blocked on a "sleeping lock"
16009 + * @p: The process to be woken up.
16011 + * Same as wake_up_process() above, but wake_flags=WF_LOCK_SLEEPER to indicate
16012 + * the nature of the wakeup.
16014 +int wake_up_lock_sleeper(struct task_struct *p)
16016 + return try_to_wake_up(p, TASK_ALL, WF_LOCK_SLEEPER);
16019 int wake_up_state(struct task_struct *p, unsigned int state)
16021 return try_to_wake_up(p, state, 0);
16022 @@ -2442,6 +2563,9 @@ int sched_fork(unsigned long clone_flags, struct task_struct *p)
16025 init_task_preempt_count(p);
16026 +#ifdef CONFIG_HAVE_PREEMPT_LAZY
16027 + task_thread_info(p)->preempt_lazy_count = 0;
16030 plist_node_init(&p->pushable_tasks, MAX_PRIO);
16031 RB_CLEAR_NODE(&p->pushable_dl_tasks);
16032 @@ -2770,21 +2894,16 @@ static struct rq *finish_task_switch(struct task_struct *prev)
16033 finish_arch_post_lock_switch();
16035 fire_sched_in_preempt_notifiers(current);
16037 + * We use mmdrop_delayed() here so we don't have to do the
16038 + * full __mmdrop() when we are the last user.
16042 + mmdrop_delayed(mm);
16043 if (unlikely(prev_state == TASK_DEAD)) {
16044 if (prev->sched_class->task_dead)
16045 prev->sched_class->task_dead(prev);
16048 - * Remove function-return probe instances associated with this
16049 - * task and put them back on the free list.
16051 - kprobe_flush_task(prev);
16053 - /* Task is done with its stack. */
16054 - put_task_stack(prev);
16056 put_task_struct(prev);
16059 @@ -3252,6 +3371,77 @@ static inline void schedule_debug(struct task_struct *prev)
16060 schedstat_inc(this_rq()->sched_count);
16063 +#if defined(CONFIG_PREEMPT_RT_FULL) && defined(CONFIG_SMP)
16065 +void migrate_disable(void)
16067 + struct task_struct *p = current;
16069 + if (in_atomic() || irqs_disabled()) {
16070 +#ifdef CONFIG_SCHED_DEBUG
16071 + p->migrate_disable_atomic++;
16076 +#ifdef CONFIG_SCHED_DEBUG
16077 + if (unlikely(p->migrate_disable_atomic)) {
16083 + if (p->migrate_disable) {
16084 + p->migrate_disable++;
16088 + preempt_disable();
16089 + preempt_lazy_disable();
16090 + pin_current_cpu();
16091 + p->migrate_disable = 1;
16092 + preempt_enable();
16094 +EXPORT_SYMBOL(migrate_disable);
16096 +void migrate_enable(void)
16098 + struct task_struct *p = current;
16100 + if (in_atomic() || irqs_disabled()) {
16101 +#ifdef CONFIG_SCHED_DEBUG
16102 + p->migrate_disable_atomic--;
16107 +#ifdef CONFIG_SCHED_DEBUG
16108 + if (unlikely(p->migrate_disable_atomic)) {
16113 + WARN_ON_ONCE(p->migrate_disable <= 0);
16115 + if (p->migrate_disable > 1) {
16116 + p->migrate_disable--;
16120 + preempt_disable();
16122 + * Clearing migrate_disable causes tsk_cpus_allowed to
16123 + * show the tasks original cpu affinity.
16125 + p->migrate_disable = 0;
16127 + unpin_current_cpu();
16128 + preempt_enable();
16129 + preempt_lazy_enable();
16131 +EXPORT_SYMBOL(migrate_enable);
16135 * Pick up the highest-prio task:
16137 @@ -3368,19 +3558,6 @@ static void __sched notrace __schedule(bool preempt)
16139 deactivate_task(rq, prev, DEQUEUE_SLEEP);
16143 - * If a worker went to sleep, notify and ask workqueue
16144 - * whether it wants to wake up a task to maintain
16147 - if (prev->flags & PF_WQ_WORKER) {
16148 - struct task_struct *to_wakeup;
16150 - to_wakeup = wq_worker_sleeping(prev);
16152 - try_to_wake_up_local(to_wakeup, cookie);
16155 switch_count = &prev->nvcsw;
16157 @@ -3390,6 +3567,7 @@ static void __sched notrace __schedule(bool preempt)
16159 next = pick_next_task(rq, prev, cookie);
16160 clear_tsk_need_resched(prev);
16161 + clear_tsk_need_resched_lazy(prev);
16162 clear_preempt_need_resched();
16163 rq->clock_skip_update = 0;
16165 @@ -3437,9 +3615,20 @@ void __noreturn do_task_dead(void)
16167 static inline void sched_submit_work(struct task_struct *tsk)
16169 - if (!tsk->state || tsk_is_pi_blocked(tsk))
16173 + * If a worker went to sleep, notify and ask workqueue whether
16174 + * it wants to wake up a task to maintain concurrency.
16176 + if (tsk->flags & PF_WQ_WORKER)
16177 + wq_worker_sleeping(tsk);
16180 + if (tsk_is_pi_blocked(tsk))
16184 * If we are going to sleep and we have plugged IO queued,
16185 * make sure to submit it to avoid deadlocks.
16187 @@ -3447,6 +3636,12 @@ static inline void sched_submit_work(struct task_struct *tsk)
16188 blk_schedule_flush_plug(tsk);
16191 +static void sched_update_worker(struct task_struct *tsk)
16193 + if (tsk->flags & PF_WQ_WORKER)
16194 + wq_worker_running(tsk);
16197 asmlinkage __visible void __sched schedule(void)
16199 struct task_struct *tsk = current;
16200 @@ -3457,6 +3652,7 @@ asmlinkage __visible void __sched schedule(void)
16202 sched_preempt_enable_no_resched();
16203 } while (need_resched());
16204 + sched_update_worker(tsk);
16206 EXPORT_SYMBOL(schedule);
16208 @@ -3520,6 +3716,30 @@ static void __sched notrace preempt_schedule_common(void)
16209 } while (need_resched());
16212 +#ifdef CONFIG_PREEMPT_LAZY
16214 + * If TIF_NEED_RESCHED is then we allow to be scheduled away since this is
16215 + * set by a RT task. Oterwise we try to avoid beeing scheduled out as long as
16216 + * preempt_lazy_count counter >0.
16218 +static __always_inline int preemptible_lazy(void)
16220 + if (test_thread_flag(TIF_NEED_RESCHED))
16222 + if (current_thread_info()->preempt_lazy_count)
16229 +static inline int preemptible_lazy(void)
16236 #ifdef CONFIG_PREEMPT
16238 * this is the entry point to schedule() from in-kernel preemption
16239 @@ -3534,7 +3754,8 @@ asmlinkage __visible void __sched notrace preempt_schedule(void)
16241 if (likely(!preemptible()))
16244 + if (!preemptible_lazy())
16246 preempt_schedule_common();
16248 NOKPROBE_SYMBOL(preempt_schedule);
16249 @@ -3561,6 +3782,9 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
16250 if (likely(!preemptible()))
16253 + if (!preemptible_lazy())
16258 * Because the function tracer can trace preempt_count_sub()
16259 @@ -3583,7 +3807,16 @@ asmlinkage __visible void __sched notrace preempt_schedule_notrace(void)
16260 * an infinite recursion.
16262 prev_ctx = exception_enter();
16264 + * The add/subtract must not be traced by the function
16265 + * tracer. But we still want to account for the
16266 + * preempt off latency tracer. Since the _notrace versions
16267 + * of add/subtract skip the accounting for latency tracer
16268 + * we must force it manually.
16270 + start_critical_timings();
16272 + stop_critical_timings();
16273 exception_exit(prev_ctx);
16275 preempt_latency_stop(1);
16276 @@ -4939,6 +5172,7 @@ int __cond_resched_lock(spinlock_t *lock)
16278 EXPORT_SYMBOL(__cond_resched_lock);
16280 +#ifndef CONFIG_PREEMPT_RT_FULL
16281 int __sched __cond_resched_softirq(void)
16283 BUG_ON(!in_softirq());
16284 @@ -4952,6 +5186,7 @@ int __sched __cond_resched_softirq(void)
16287 EXPORT_SYMBOL(__cond_resched_softirq);
16291 * yield - yield the current processor to other threads.
16292 @@ -5315,7 +5550,9 @@ void init_idle(struct task_struct *idle, int cpu)
16294 /* Set the preempt count _outside_ the spinlocks! */
16295 init_idle_preempt_count(idle, cpu);
16297 +#ifdef CONFIG_HAVE_PREEMPT_LAZY
16298 + task_thread_info(idle)->preempt_lazy_count = 0;
16301 * The idle tasks have their own, simple scheduling class:
16303 @@ -5458,6 +5695,8 @@ void sched_setnuma(struct task_struct *p, int nid)
16304 #endif /* CONFIG_NUMA_BALANCING */
16306 #ifdef CONFIG_HOTPLUG_CPU
16307 +static DEFINE_PER_CPU(struct mm_struct *, idle_last_mm);
16310 * Ensures that the idle task is using init_mm right before its cpu goes
16312 @@ -5472,7 +5711,12 @@ void idle_task_exit(void)
16313 switch_mm_irqs_off(mm, &init_mm, current);
16314 finish_arch_post_lock_switch();
16318 + * Defer the cleanup to an alive cpu. On RT we can neither
16319 + * call mmdrop() nor mmdrop_delayed() from here.
16321 + per_cpu(idle_last_mm, smp_processor_id()) = mm;
16326 @@ -7418,6 +7662,10 @@ int sched_cpu_dying(unsigned int cpu)
16327 update_max_interval();
16328 nohz_balance_exit_idle(cpu);
16330 + if (per_cpu(idle_last_mm, cpu)) {
16331 + mmdrop_delayed(per_cpu(idle_last_mm, cpu));
16332 + per_cpu(idle_last_mm, cpu) = NULL;
16337 @@ -7698,7 +7946,7 @@ void __init sched_init(void)
16338 #ifdef CONFIG_DEBUG_ATOMIC_SLEEP
16339 static inline int preempt_count_equals(int preempt_offset)
16341 - int nested = preempt_count() + rcu_preempt_depth();
16342 + int nested = preempt_count() + sched_rcu_preempt_depth();
16344 return (nested == preempt_offset);
16346 diff --git a/kernel/sched/deadline.c b/kernel/sched/deadline.c
16347 index 37e2449186c4..e00accf92a4b 100644
16348 --- a/kernel/sched/deadline.c
16349 +++ b/kernel/sched/deadline.c
16350 @@ -687,6 +687,7 @@ void init_dl_task_timer(struct sched_dl_entity *dl_se)
16352 hrtimer_init(timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
16353 timer->function = dl_task_timer;
16354 + timer->irqsafe = 1;
16358 @@ -1729,12 +1730,11 @@ static void switched_to_dl(struct rq *rq, struct task_struct *p)
16360 if (tsk_nr_cpus_allowed(p) > 1 && rq->dl.overloaded)
16361 queue_push_tasks(rq);
16364 if (dl_task(rq->curr))
16365 check_preempt_curr_dl(rq, p, 0);
16372 diff --git a/kernel/sched/debug.c b/kernel/sched/debug.c
16373 index fa178b62ea79..935224123441 100644
16374 --- a/kernel/sched/debug.c
16375 +++ b/kernel/sched/debug.c
16376 @@ -558,6 +558,9 @@ void print_rt_rq(struct seq_file *m, int cpu, struct rt_rq *rt_rq)
16381 + P(rt_nr_migratory);
16386 @@ -953,6 +956,10 @@ void proc_sched_show_task(struct task_struct *p, struct seq_file *m)
16390 +#ifdef CONFIG_PREEMPT_RT_FULL
16391 + P(migrate_disable);
16393 + P(nr_cpus_allowed);
16394 #undef PN_SCHEDSTAT
16397 diff --git a/kernel/sched/fair.c b/kernel/sched/fair.c
16398 index c242944f5cbd..4aeb2e2e41bc 100644
16399 --- a/kernel/sched/fair.c
16400 +++ b/kernel/sched/fair.c
16401 @@ -3518,7 +3518,7 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
16402 ideal_runtime = sched_slice(cfs_rq, curr);
16403 delta_exec = curr->sum_exec_runtime - curr->prev_sum_exec_runtime;
16404 if (delta_exec > ideal_runtime) {
16405 - resched_curr(rq_of(cfs_rq));
16406 + resched_curr_lazy(rq_of(cfs_rq));
16408 * The current task ran long enough, ensure it doesn't get
16409 * re-elected due to buddy favours.
16410 @@ -3542,7 +3542,7 @@ check_preempt_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr)
16413 if (delta > ideal_runtime)
16414 - resched_curr(rq_of(cfs_rq));
16415 + resched_curr_lazy(rq_of(cfs_rq));
16419 @@ -3684,7 +3684,7 @@ entity_tick(struct cfs_rq *cfs_rq, struct sched_entity *curr, int queued)
16420 * validating it and just reschedule.
16423 - resched_curr(rq_of(cfs_rq));
16424 + resched_curr_lazy(rq_of(cfs_rq));
16428 @@ -3866,7 +3866,7 @@ static void __account_cfs_rq_runtime(struct cfs_rq *cfs_rq, u64 delta_exec)
16429 * hierarchy can be throttled
16431 if (!assign_cfs_rq_runtime(cfs_rq) && likely(cfs_rq->curr))
16432 - resched_curr(rq_of(cfs_rq));
16433 + resched_curr_lazy(rq_of(cfs_rq));
16436 static __always_inline
16437 @@ -4494,7 +4494,7 @@ static void hrtick_start_fair(struct rq *rq, struct task_struct *p)
16441 - resched_curr(rq);
16442 + resched_curr_lazy(rq);
16445 hrtick_start(rq, delta);
16446 @@ -5905,7 +5905,7 @@ static void check_preempt_wakeup(struct rq *rq, struct task_struct *p, int wake_
16450 - resched_curr(rq);
16451 + resched_curr_lazy(rq);
16453 * Only set the backward buddy when the current task is still
16454 * on the rq. This can happen when a wakeup gets interleaved
16455 @@ -8631,7 +8631,7 @@ static void task_fork_fair(struct task_struct *p)
16456 * 'current' within the tree based on its new key value.
16458 swap(curr->vruntime, se->vruntime);
16459 - resched_curr(rq);
16460 + resched_curr_lazy(rq);
16463 se->vruntime -= cfs_rq->min_vruntime;
16464 @@ -8655,7 +8655,7 @@ prio_changed_fair(struct rq *rq, struct task_struct *p, int oldprio)
16466 if (rq->curr == p) {
16467 if (p->prio > oldprio)
16468 - resched_curr(rq);
16469 + resched_curr_lazy(rq);
16471 check_preempt_curr(rq, p, 0);
16473 diff --git a/kernel/sched/features.h b/kernel/sched/features.h
16474 index 69631fa46c2f..6d28fcd08872 100644
16475 --- a/kernel/sched/features.h
16476 +++ b/kernel/sched/features.h
16477 @@ -45,11 +45,19 @@ SCHED_FEAT(LB_BIAS, true)
16479 SCHED_FEAT(NONTASK_CAPACITY, true)
16481 +#ifdef CONFIG_PREEMPT_RT_FULL
16482 +SCHED_FEAT(TTWU_QUEUE, false)
16483 +# ifdef CONFIG_PREEMPT_LAZY
16484 +SCHED_FEAT(PREEMPT_LAZY, true)
16489 * Queue remote wakeups on the target CPU and process them
16490 * using the scheduler IPI. Reduces rq->lock contention/bounces.
16492 SCHED_FEAT(TTWU_QUEUE, true)
16495 #ifdef HAVE_RT_PUSH_IPI
16497 diff --git a/kernel/sched/rt.c b/kernel/sched/rt.c
16498 index 2516b8df6dbb..b0691f4e7d49 100644
16499 --- a/kernel/sched/rt.c
16500 +++ b/kernel/sched/rt.c
16501 @@ -47,6 +47,7 @@ void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime)
16503 hrtimer_init(&rt_b->rt_period_timer,
16504 CLOCK_MONOTONIC, HRTIMER_MODE_REL);
16505 + rt_b->rt_period_timer.irqsafe = 1;
16506 rt_b->rt_period_timer.function = sched_rt_period_timer;
16509 @@ -101,6 +102,7 @@ void init_rt_rq(struct rt_rq *rt_rq)
16510 rt_rq->push_cpu = nr_cpu_ids;
16511 raw_spin_lock_init(&rt_rq->push_lock);
16512 init_irq_work(&rt_rq->push_work, push_irq_work_func);
16513 + rt_rq->push_work.flags |= IRQ_WORK_HARD_IRQ;
16515 #endif /* CONFIG_SMP */
16516 /* We start is dequeued state, because no RT tasks are queued */
16517 @@ -2198,10 +2200,9 @@ static void switched_to_rt(struct rq *rq, struct task_struct *p)
16519 if (tsk_nr_cpus_allowed(p) > 1 && rq->rt.overloaded)
16520 queue_push_tasks(rq);
16522 +#endif /* CONFIG_SMP */
16523 if (p->prio < rq->curr->prio)
16525 -#endif /* CONFIG_SMP */
16529 diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h
16530 index 055f935d4421..19324ac27026 100644
16531 --- a/kernel/sched/sched.h
16532 +++ b/kernel/sched/sched.h
16533 @@ -1163,6 +1163,7 @@ static inline void finish_lock_switch(struct rq *rq, struct task_struct *prev)
16534 #define WF_SYNC 0x01 /* waker goes to sleep after wakeup */
16535 #define WF_FORK 0x02 /* child wakeup after fork */
16536 #define WF_MIGRATED 0x4 /* internal use, task got migrated */
16537 +#define WF_LOCK_SLEEPER 0x08 /* wakeup spinlock "sleeper" */
16540 * To aid in avoiding the subversion of "niceness" due to uneven distribution
16541 @@ -1346,6 +1347,15 @@ extern void init_sched_fair_class(void);
16542 extern void resched_curr(struct rq *rq);
16543 extern void resched_cpu(int cpu);
16545 +#ifdef CONFIG_PREEMPT_LAZY
16546 +extern void resched_curr_lazy(struct rq *rq);
16548 +static inline void resched_curr_lazy(struct rq *rq)
16550 + resched_curr(rq);
16554 extern struct rt_bandwidth def_rt_bandwidth;
16555 extern void init_rt_bandwidth(struct rt_bandwidth *rt_b, u64 period, u64 runtime);
16557 diff --git a/kernel/sched/swait.c b/kernel/sched/swait.c
16558 index 82f0dff90030..ef027ff3250a 100644
16559 --- a/kernel/sched/swait.c
16560 +++ b/kernel/sched/swait.c
16562 #include <linux/sched.h>
16563 #include <linux/swait.h>
16564 +#include <linux/suspend.h>
16566 void __init_swait_queue_head(struct swait_queue_head *q, const char *name,
16567 struct lock_class_key *key)
16568 @@ -29,6 +30,25 @@ void swake_up_locked(struct swait_queue_head *q)
16570 EXPORT_SYMBOL(swake_up_locked);
16572 +void swake_up_all_locked(struct swait_queue_head *q)
16574 + struct swait_queue *curr;
16577 + while (!list_empty(&q->task_list)) {
16579 + curr = list_first_entry(&q->task_list, typeof(*curr),
16581 + wake_up_process(curr->task);
16582 + list_del_init(&curr->task_list);
16585 + if (pm_in_action)
16587 + WARN(wakes > 2, "complete_all() with %d waiters\n", wakes);
16589 +EXPORT_SYMBOL(swake_up_all_locked);
16591 void swake_up(struct swait_queue_head *q)
16593 unsigned long flags;
16594 @@ -54,6 +74,7 @@ void swake_up_all(struct swait_queue_head *q)
16595 if (!swait_active(q))
16598 + WARN_ON(irqs_disabled());
16599 raw_spin_lock_irq(&q->lock);
16600 list_splice_init(&q->task_list, &tmp);
16601 while (!list_empty(&tmp)) {
16602 diff --git a/kernel/sched/swork.c b/kernel/sched/swork.c
16603 new file mode 100644
16604 index 000000000000..1950f40ca725
16606 +++ b/kernel/sched/swork.c
16609 + * Copyright (C) 2014 BMW Car IT GmbH, Daniel Wagner daniel.wagner@bmw-carit.de
16611 + * Provides a framework for enqueuing callbacks from irq context
16612 + * PREEMPT_RT_FULL safe. The callbacks are executed in kthread context.
16615 +#include <linux/swait.h>
16616 +#include <linux/swork.h>
16617 +#include <linux/kthread.h>
16618 +#include <linux/slab.h>
16619 +#include <linux/spinlock.h>
16620 +#include <linux/export.h>
16622 +#define SWORK_EVENT_PENDING (1 << 0)
16624 +static DEFINE_MUTEX(worker_mutex);
16625 +static struct sworker *glob_worker;
16628 + struct list_head events;
16629 + struct swait_queue_head wq;
16631 + raw_spinlock_t lock;
16633 + struct task_struct *task;
16637 +static bool swork_readable(struct sworker *worker)
16641 + if (kthread_should_stop())
16644 + raw_spin_lock_irq(&worker->lock);
16645 + r = !list_empty(&worker->events);
16646 + raw_spin_unlock_irq(&worker->lock);
16651 +static int swork_kthread(void *arg)
16653 + struct sworker *worker = arg;
16656 + swait_event_interruptible(worker->wq,
16657 + swork_readable(worker));
16658 + if (kthread_should_stop())
16661 + raw_spin_lock_irq(&worker->lock);
16662 + while (!list_empty(&worker->events)) {
16663 + struct swork_event *sev;
16665 + sev = list_first_entry(&worker->events,
16666 + struct swork_event, item);
16667 + list_del(&sev->item);
16668 + raw_spin_unlock_irq(&worker->lock);
16670 + WARN_ON_ONCE(!test_and_clear_bit(SWORK_EVENT_PENDING,
16673 + raw_spin_lock_irq(&worker->lock);
16675 + raw_spin_unlock_irq(&worker->lock);
16680 +static struct sworker *swork_create(void)
16682 + struct sworker *worker;
16684 + worker = kzalloc(sizeof(*worker), GFP_KERNEL);
16686 + return ERR_PTR(-ENOMEM);
16688 + INIT_LIST_HEAD(&worker->events);
16689 + raw_spin_lock_init(&worker->lock);
16690 + init_swait_queue_head(&worker->wq);
16692 + worker->task = kthread_run(swork_kthread, worker, "kswork");
16693 + if (IS_ERR(worker->task)) {
16695 + return ERR_PTR(-ENOMEM);
16701 +static void swork_destroy(struct sworker *worker)
16703 + kthread_stop(worker->task);
16705 + WARN_ON(!list_empty(&worker->events));
16710 + * swork_queue - queue swork
16712 + * Returns %false if @work was already on a queue, %true otherwise.
16714 + * The work is queued and processed on a random CPU
16716 +bool swork_queue(struct swork_event *sev)
16718 + unsigned long flags;
16720 + if (test_and_set_bit(SWORK_EVENT_PENDING, &sev->flags))
16723 + raw_spin_lock_irqsave(&glob_worker->lock, flags);
16724 + list_add_tail(&sev->item, &glob_worker->events);
16725 + raw_spin_unlock_irqrestore(&glob_worker->lock, flags);
16727 + swake_up(&glob_worker->wq);
16730 +EXPORT_SYMBOL_GPL(swork_queue);
16733 + * swork_get - get an instance of the sworker
16735 + * Returns an negative error code if the initialization if the worker did not
16736 + * work, %0 otherwise.
16739 +int swork_get(void)
16741 + struct sworker *worker;
16743 + mutex_lock(&worker_mutex);
16744 + if (!glob_worker) {
16745 + worker = swork_create();
16746 + if (IS_ERR(worker)) {
16747 + mutex_unlock(&worker_mutex);
16751 + glob_worker = worker;
16754 + glob_worker->refs++;
16755 + mutex_unlock(&worker_mutex);
16759 +EXPORT_SYMBOL_GPL(swork_get);
16762 + * swork_put - puts an instance of the sworker
16764 + * Will destroy the sworker thread. This function must not be called until all
16765 + * queued events have been completed.
16767 +void swork_put(void)
16769 + mutex_lock(&worker_mutex);
16771 + glob_worker->refs--;
16772 + if (glob_worker->refs > 0)
16775 + swork_destroy(glob_worker);
16776 + glob_worker = NULL;
16778 + mutex_unlock(&worker_mutex);
16780 +EXPORT_SYMBOL_GPL(swork_put);
16781 diff --git a/kernel/signal.c b/kernel/signal.c
16782 index 75761acc77cf..ae0773c76bb0 100644
16783 --- a/kernel/signal.c
16784 +++ b/kernel/signal.c
16786 #include <linux/export.h>
16787 #include <linux/init.h>
16788 #include <linux/sched.h>
16789 +#include <linux/sched/rt.h>
16790 #include <linux/fs.h>
16791 #include <linux/tty.h>
16792 #include <linux/binfmts.h>
16793 @@ -352,13 +353,30 @@ static bool task_participate_group_stop(struct task_struct *task)
16797 +static inline struct sigqueue *get_task_cache(struct task_struct *t)
16799 + struct sigqueue *q = t->sigqueue_cache;
16801 + if (cmpxchg(&t->sigqueue_cache, q, NULL) != q)
16806 +static inline int put_task_cache(struct task_struct *t, struct sigqueue *q)
16808 + if (cmpxchg(&t->sigqueue_cache, NULL, q) == NULL)
16814 * allocate a new signal queue record
16815 * - this may be called without locks if and only if t == current, otherwise an
16816 * appropriate lock must be held to stop the target task from exiting
16818 static struct sigqueue *
16819 -__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimit)
16820 +__sigqueue_do_alloc(int sig, struct task_struct *t, gfp_t flags,
16821 + int override_rlimit, int fromslab)
16823 struct sigqueue *q = NULL;
16824 struct user_struct *user;
16825 @@ -375,7 +393,10 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi
16826 if (override_rlimit ||
16827 atomic_read(&user->sigpending) <=
16828 task_rlimit(t, RLIMIT_SIGPENDING)) {
16829 - q = kmem_cache_alloc(sigqueue_cachep, flags);
16831 + q = get_task_cache(t);
16833 + q = kmem_cache_alloc(sigqueue_cachep, flags);
16835 print_dropped_signal(sig);
16837 @@ -392,6 +413,13 @@ __sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags, int override_rlimi
16841 +static struct sigqueue *
16842 +__sigqueue_alloc(int sig, struct task_struct *t, gfp_t flags,
16843 + int override_rlimit)
16845 + return __sigqueue_do_alloc(sig, t, flags, override_rlimit, 0);
16848 static void __sigqueue_free(struct sigqueue *q)
16850 if (q->flags & SIGQUEUE_PREALLOC)
16851 @@ -401,6 +429,21 @@ static void __sigqueue_free(struct sigqueue *q)
16852 kmem_cache_free(sigqueue_cachep, q);
16855 +static void sigqueue_free_current(struct sigqueue *q)
16857 + struct user_struct *up;
16859 + if (q->flags & SIGQUEUE_PREALLOC)
16863 + if (rt_prio(current->normal_prio) && !put_task_cache(current, q)) {
16864 + atomic_dec(&up->sigpending);
16867 + __sigqueue_free(q);
16870 void flush_sigqueue(struct sigpending *queue)
16872 struct sigqueue *q;
16873 @@ -414,6 +457,21 @@ void flush_sigqueue(struct sigpending *queue)
16877 + * Called from __exit_signal. Flush tsk->pending and
16878 + * tsk->sigqueue_cache
16880 +void flush_task_sigqueue(struct task_struct *tsk)
16882 + struct sigqueue *q;
16884 + flush_sigqueue(&tsk->pending);
16886 + q = get_task_cache(tsk);
16888 + kmem_cache_free(sigqueue_cachep, q);
16892 * Flush all pending signals for this kthread.
16894 void flush_signals(struct task_struct *t)
16895 @@ -525,7 +583,7 @@ static void collect_signal(int sig, struct sigpending *list, siginfo_t *info)
16897 list_del_init(&first->list);
16898 copy_siginfo(info, &first->info);
16899 - __sigqueue_free(first);
16900 + sigqueue_free_current(first);
16903 * Ok, it wasn't in the queue. This must be
16904 @@ -560,6 +618,8 @@ int dequeue_signal(struct task_struct *tsk, sigset_t *mask, siginfo_t *info)
16908 + WARN_ON_ONCE(tsk != current);
16910 /* We only dequeue private signals from ourselves, we don't let
16911 * signalfd steal them
16913 @@ -1156,8 +1216,8 @@ int do_send_sig_info(int sig, struct siginfo *info, struct task_struct *p,
16914 * We don't want to have recursive SIGSEGV's etc, for example,
16915 * that is why we also clear SIGNAL_UNKILLABLE.
16918 -force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
16920 +do_force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
16922 unsigned long int flags;
16923 int ret, blocked, ignored;
16924 @@ -1182,6 +1242,39 @@ force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
16928 +int force_sig_info(int sig, struct siginfo *info, struct task_struct *t)
16931 + * On some archs, PREEMPT_RT has to delay sending a signal from a trap
16932 + * since it can not enable preemption, and the signal code's spin_locks
16933 + * turn into mutexes. Instead, it must set TIF_NOTIFY_RESUME which will
16934 + * send the signal on exit of the trap.
16936 +#ifdef ARCH_RT_DELAYS_SIGNAL_SEND
16937 + if (in_atomic()) {
16938 + if (WARN_ON_ONCE(t != current))
16940 + if (WARN_ON_ONCE(t->forced_info.si_signo))
16943 + if (is_si_special(info)) {
16944 + WARN_ON_ONCE(info != SEND_SIG_PRIV);
16945 + t->forced_info.si_signo = sig;
16946 + t->forced_info.si_errno = 0;
16947 + t->forced_info.si_code = SI_KERNEL;
16948 + t->forced_info.si_pid = 0;
16949 + t->forced_info.si_uid = 0;
16951 + t->forced_info = *info;
16954 + set_tsk_thread_flag(t, TIF_NOTIFY_RESUME);
16958 + return do_force_sig_info(sig, info, t);
16962 * Nuke all other threads in the group.
16964 @@ -1216,12 +1309,12 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
16965 * Disable interrupts early to avoid deadlocks.
16966 * See rcu_read_unlock() comment header for details.
16968 - local_irq_save(*flags);
16969 + local_irq_save_nort(*flags);
16971 sighand = rcu_dereference(tsk->sighand);
16972 if (unlikely(sighand == NULL)) {
16974 - local_irq_restore(*flags);
16975 + local_irq_restore_nort(*flags);
16979 @@ -1242,7 +1335,7 @@ struct sighand_struct *__lock_task_sighand(struct task_struct *tsk,
16981 spin_unlock(&sighand->siglock);
16983 - local_irq_restore(*flags);
16984 + local_irq_restore_nort(*flags);
16988 @@ -1485,7 +1578,8 @@ EXPORT_SYMBOL(kill_pid);
16990 struct sigqueue *sigqueue_alloc(void)
16992 - struct sigqueue *q = __sigqueue_alloc(-1, current, GFP_KERNEL, 0);
16993 + /* Preallocated sigqueue objects always from the slabcache ! */
16994 + struct sigqueue *q = __sigqueue_do_alloc(-1, current, GFP_KERNEL, 0, 1);
16997 q->flags |= SIGQUEUE_PREALLOC;
16998 @@ -1846,15 +1940,7 @@ static void ptrace_stop(int exit_code, int why, int clear_code, siginfo_t *info)
16999 if (gstop_done && ptrace_reparented(current))
17000 do_notify_parent_cldstop(current, false, why);
17003 - * Don't want to allow preemption here, because
17004 - * sys_ptrace() needs this task to be inactive.
17006 - * XXX: implement read_unlock_no_resched().
17008 - preempt_disable();
17009 read_unlock(&tasklist_lock);
17010 - preempt_enable_no_resched();
17011 freezable_schedule();
17014 diff --git a/kernel/softirq.c b/kernel/softirq.c
17015 index 744fa611cae0..819bd7cf5ad0 100644
17016 --- a/kernel/softirq.c
17017 +++ b/kernel/softirq.c
17018 @@ -21,10 +21,12 @@
17019 #include <linux/freezer.h>
17020 #include <linux/kthread.h>
17021 #include <linux/rcupdate.h>
17022 +#include <linux/delay.h>
17023 #include <linux/ftrace.h>
17024 #include <linux/smp.h>
17025 #include <linux/smpboot.h>
17026 #include <linux/tick.h>
17027 +#include <linux/locallock.h>
17028 #include <linux/irq.h>
17030 #define CREATE_TRACE_POINTS
17031 @@ -56,12 +58,108 @@ EXPORT_SYMBOL(irq_stat);
17032 static struct softirq_action softirq_vec[NR_SOFTIRQS] __cacheline_aligned_in_smp;
17034 DEFINE_PER_CPU(struct task_struct *, ksoftirqd);
17035 +#ifdef CONFIG_PREEMPT_RT_FULL
17036 +#define TIMER_SOFTIRQS ((1 << TIMER_SOFTIRQ) | (1 << HRTIMER_SOFTIRQ))
17037 +DEFINE_PER_CPU(struct task_struct *, ktimer_softirqd);
17040 const char * const softirq_to_name[NR_SOFTIRQS] = {
17041 "HI", "TIMER", "NET_TX", "NET_RX", "BLOCK", "IRQ_POLL",
17042 "TASKLET", "SCHED", "HRTIMER", "RCU"
17045 +#ifdef CONFIG_NO_HZ_COMMON
17046 +# ifdef CONFIG_PREEMPT_RT_FULL
17048 +struct softirq_runner {
17049 + struct task_struct *runner[NR_SOFTIRQS];
17052 +static DEFINE_PER_CPU(struct softirq_runner, softirq_runners);
17054 +static inline void softirq_set_runner(unsigned int sirq)
17056 + struct softirq_runner *sr = this_cpu_ptr(&softirq_runners);
17058 + sr->runner[sirq] = current;
17061 +static inline void softirq_clr_runner(unsigned int sirq)
17063 + struct softirq_runner *sr = this_cpu_ptr(&softirq_runners);
17065 + sr->runner[sirq] = NULL;
17069 + * On preempt-rt a softirq running context might be blocked on a
17070 + * lock. There might be no other runnable task on this CPU because the
17071 + * lock owner runs on some other CPU. So we have to go into idle with
17072 + * the pending bit set. Therefor we need to check this otherwise we
17073 + * warn about false positives which confuses users and defeats the
17074 + * whole purpose of this test.
17076 + * This code is called with interrupts disabled.
17078 +void softirq_check_pending_idle(void)
17080 + static int rate_limit;
17081 + struct softirq_runner *sr = this_cpu_ptr(&softirq_runners);
17085 + if (rate_limit >= 10)
17088 + warnpending = local_softirq_pending() & SOFTIRQ_STOP_IDLE_MASK;
17089 + for (i = 0; i < NR_SOFTIRQS; i++) {
17090 + struct task_struct *tsk = sr->runner[i];
17093 + * The wakeup code in rtmutex.c wakes up the task
17094 + * _before_ it sets pi_blocked_on to NULL under
17095 + * tsk->pi_lock. So we need to check for both: state
17096 + * and pi_blocked_on.
17099 + raw_spin_lock(&tsk->pi_lock);
17100 + if (tsk->pi_blocked_on || tsk->state == TASK_RUNNING) {
17101 + /* Clear all bits pending in that task */
17102 + warnpending &= ~(tsk->softirqs_raised);
17103 + warnpending &= ~(1 << i);
17105 + raw_spin_unlock(&tsk->pi_lock);
17109 + if (warnpending) {
17110 + printk(KERN_ERR "NOHZ: local_softirq_pending %02x\n",
17117 + * On !PREEMPT_RT we just printk rate limited:
17119 +void softirq_check_pending_idle(void)
17121 + static int rate_limit;
17123 + if (rate_limit < 10 &&
17124 + (local_softirq_pending() & SOFTIRQ_STOP_IDLE_MASK)) {
17125 + printk(KERN_ERR "NOHZ: local_softirq_pending %02x\n",
17126 + local_softirq_pending());
17132 +#else /* !CONFIG_NO_HZ_COMMON */
17133 +static inline void softirq_set_runner(unsigned int sirq) { }
17134 +static inline void softirq_clr_runner(unsigned int sirq) { }
17138 * we cannot loop indefinitely here to avoid userspace starvation,
17139 * but we also don't want to introduce a worst case 1/HZ latency
17140 @@ -77,6 +175,38 @@ static void wakeup_softirqd(void)
17141 wake_up_process(tsk);
17144 +#ifdef CONFIG_PREEMPT_RT_FULL
17145 +static void wakeup_timer_softirqd(void)
17147 + /* Interrupts are disabled: no need to stop preemption */
17148 + struct task_struct *tsk = __this_cpu_read(ktimer_softirqd);
17150 + if (tsk && tsk->state != TASK_RUNNING)
17151 + wake_up_process(tsk);
17155 +static void handle_softirq(unsigned int vec_nr)
17157 + struct softirq_action *h = softirq_vec + vec_nr;
17160 + prev_count = preempt_count();
17162 + kstat_incr_softirqs_this_cpu(vec_nr);
17164 + trace_softirq_entry(vec_nr);
17166 + trace_softirq_exit(vec_nr);
17167 + if (unlikely(prev_count != preempt_count())) {
17168 + pr_err("huh, entered softirq %u %s %p with preempt_count %08x, exited with %08x?\n",
17169 + vec_nr, softirq_to_name[vec_nr], h->action,
17170 + prev_count, preempt_count());
17171 + preempt_count_set(prev_count);
17175 +#ifndef CONFIG_PREEMPT_RT_FULL
17177 * If ksoftirqd is scheduled, we do not want to process pending softirqs
17178 * right now. Let ksoftirqd handle this at its own rate, to get fairness.
17179 @@ -88,6 +218,47 @@ static bool ksoftirqd_running(void)
17180 return tsk && (tsk->state == TASK_RUNNING);
17183 +static inline int ksoftirqd_softirq_pending(void)
17185 + return local_softirq_pending();
17188 +static void handle_pending_softirqs(u32 pending)
17190 + struct softirq_action *h = softirq_vec;
17193 + local_irq_enable();
17197 + while ((softirq_bit = ffs(pending))) {
17198 + unsigned int vec_nr;
17200 + h += softirq_bit - 1;
17201 + vec_nr = h - softirq_vec;
17202 + handle_softirq(vec_nr);
17205 + pending >>= softirq_bit;
17209 + local_irq_disable();
17212 +static void run_ksoftirqd(unsigned int cpu)
17214 + local_irq_disable();
17215 + if (ksoftirqd_softirq_pending()) {
17217 + local_irq_enable();
17218 + cond_resched_rcu_qs();
17221 + local_irq_enable();
17225 * preempt_count and SOFTIRQ_OFFSET usage:
17226 * - preempt_count is changed by SOFTIRQ_OFFSET on entering or leaving
17227 @@ -243,10 +414,8 @@ asmlinkage __visible void __softirq_entry __do_softirq(void)
17228 unsigned long end = jiffies + MAX_SOFTIRQ_TIME;
17229 unsigned long old_flags = current->flags;
17230 int max_restart = MAX_SOFTIRQ_RESTART;
17231 - struct softirq_action *h;
17237 * Mask out PF_MEMALLOC s current task context is borrowed for the
17238 @@ -265,36 +434,7 @@ asmlinkage __visible void __softirq_entry __do_softirq(void)
17239 /* Reset the pending bitmask before enabling irqs */
17240 set_softirq_pending(0);
17242 - local_irq_enable();
17246 - while ((softirq_bit = ffs(pending))) {
17247 - unsigned int vec_nr;
17250 - h += softirq_bit - 1;
17252 - vec_nr = h - softirq_vec;
17253 - prev_count = preempt_count();
17255 - kstat_incr_softirqs_this_cpu(vec_nr);
17257 - trace_softirq_entry(vec_nr);
17259 - trace_softirq_exit(vec_nr);
17260 - if (unlikely(prev_count != preempt_count())) {
17261 - pr_err("huh, entered softirq %u %s %p with preempt_count %08x, exited with %08x?\n",
17262 - vec_nr, softirq_to_name[vec_nr], h->action,
17263 - prev_count, preempt_count());
17264 - preempt_count_set(prev_count);
17267 - pending >>= softirq_bit;
17271 - local_irq_disable();
17272 + handle_pending_softirqs(pending);
17274 pending = local_softirq_pending();
17276 @@ -331,6 +471,309 @@ asmlinkage __visible void do_softirq(void)
17280 + * This function must run with irqs disabled!
17282 +void raise_softirq_irqoff(unsigned int nr)
17284 + __raise_softirq_irqoff(nr);
17287 + * If we're in an interrupt or softirq, we're done
17288 + * (this also catches softirq-disabled code). We will
17289 + * actually run the softirq once we return from
17290 + * the irq or softirq.
17292 + * Otherwise we wake up ksoftirqd to make sure we
17293 + * schedule the softirq soon.
17295 + if (!in_interrupt())
17296 + wakeup_softirqd();
17299 +void __raise_softirq_irqoff(unsigned int nr)
17301 + trace_softirq_raise(nr);
17302 + or_softirq_pending(1UL << nr);
17305 +static inline void local_bh_disable_nort(void) { local_bh_disable(); }
17306 +static inline void _local_bh_enable_nort(void) { _local_bh_enable(); }
17307 +static void ksoftirqd_set_sched_params(unsigned int cpu) { }
17309 +#else /* !PREEMPT_RT_FULL */
17312 + * On RT we serialize softirq execution with a cpu local lock per softirq
17314 +static DEFINE_PER_CPU(struct local_irq_lock [NR_SOFTIRQS], local_softirq_locks);
17316 +void __init softirq_early_init(void)
17320 + for (i = 0; i < NR_SOFTIRQS; i++)
17321 + local_irq_lock_init(local_softirq_locks[i]);
17324 +static void lock_softirq(int which)
17326 + local_lock(local_softirq_locks[which]);
17329 +static void unlock_softirq(int which)
17331 + local_unlock(local_softirq_locks[which]);
17334 +static void do_single_softirq(int which)
17336 + unsigned long old_flags = current->flags;
17338 + current->flags &= ~PF_MEMALLOC;
17339 + vtime_account_irq_enter(current);
17340 + current->flags |= PF_IN_SOFTIRQ;
17341 + lockdep_softirq_enter();
17342 + local_irq_enable();
17343 + handle_softirq(which);
17344 + local_irq_disable();
17345 + lockdep_softirq_exit();
17346 + current->flags &= ~PF_IN_SOFTIRQ;
17347 + vtime_account_irq_enter(current);
17348 + tsk_restore_flags(current, old_flags, PF_MEMALLOC);
17352 + * Called with interrupts disabled. Process softirqs which were raised
17353 + * in current context (or on behalf of ksoftirqd).
17355 +static void do_current_softirqs(void)
17357 + while (current->softirqs_raised) {
17358 + int i = __ffs(current->softirqs_raised);
17359 + unsigned int pending, mask = (1U << i);
17361 + current->softirqs_raised &= ~mask;
17362 + local_irq_enable();
17365 + * If the lock is contended, we boost the owner to
17366 + * process the softirq or leave the critical section
17370 + local_irq_disable();
17371 + softirq_set_runner(i);
17373 + * Check with the local_softirq_pending() bits,
17374 + * whether we need to process this still or if someone
17375 + * else took care of it.
17377 + pending = local_softirq_pending();
17378 + if (pending & mask) {
17379 + set_softirq_pending(pending & ~mask);
17380 + do_single_softirq(i);
17382 + softirq_clr_runner(i);
17383 + WARN_ON(current->softirq_nestcnt != 1);
17384 + local_irq_enable();
17385 + unlock_softirq(i);
17386 + local_irq_disable();
17390 +void __local_bh_disable(void)
17392 + if (++current->softirq_nestcnt == 1)
17393 + migrate_disable();
17395 +EXPORT_SYMBOL(__local_bh_disable);
17397 +void __local_bh_enable(void)
17399 + if (WARN_ON(current->softirq_nestcnt == 0))
17402 + local_irq_disable();
17403 + if (current->softirq_nestcnt == 1 && current->softirqs_raised)
17404 + do_current_softirqs();
17405 + local_irq_enable();
17407 + if (--current->softirq_nestcnt == 0)
17408 + migrate_enable();
17410 +EXPORT_SYMBOL(__local_bh_enable);
17412 +void _local_bh_enable(void)
17414 + if (WARN_ON(current->softirq_nestcnt == 0))
17416 + if (--current->softirq_nestcnt == 0)
17417 + migrate_enable();
17419 +EXPORT_SYMBOL(_local_bh_enable);
17421 +int in_serving_softirq(void)
17423 + return current->flags & PF_IN_SOFTIRQ;
17425 +EXPORT_SYMBOL(in_serving_softirq);
17427 +/* Called with preemption disabled */
17428 +static void run_ksoftirqd(unsigned int cpu)
17430 + local_irq_disable();
17431 + current->softirq_nestcnt++;
17433 + do_current_softirqs();
17434 + current->softirq_nestcnt--;
17435 + local_irq_enable();
17436 + cond_resched_rcu_qs();
17440 + * Called from netif_rx_ni(). Preemption enabled, but migration
17441 + * disabled. So the cpu can't go away under us.
17443 +void thread_do_softirq(void)
17445 + if (!in_serving_softirq() && current->softirqs_raised) {
17446 + current->softirq_nestcnt++;
17447 + do_current_softirqs();
17448 + current->softirq_nestcnt--;
17452 +static void do_raise_softirq_irqoff(unsigned int nr)
17454 + unsigned int mask;
17456 + mask = 1UL << nr;
17458 + trace_softirq_raise(nr);
17459 + or_softirq_pending(mask);
17462 + * If we are not in a hard interrupt and inside a bh disabled
17463 + * region, we simply raise the flag on current. local_bh_enable()
17464 + * will make sure that the softirq is executed. Otherwise we
17465 + * delegate it to ksoftirqd.
17467 + if (!in_irq() && current->softirq_nestcnt)
17468 + current->softirqs_raised |= mask;
17469 + else if (!__this_cpu_read(ksoftirqd) || !__this_cpu_read(ktimer_softirqd))
17472 + if (mask & TIMER_SOFTIRQS)
17473 + __this_cpu_read(ktimer_softirqd)->softirqs_raised |= mask;
17475 + __this_cpu_read(ksoftirqd)->softirqs_raised |= mask;
17478 +static void wakeup_proper_softirq(unsigned int nr)
17480 + if ((1UL << nr) & TIMER_SOFTIRQS)
17481 + wakeup_timer_softirqd();
17483 + wakeup_softirqd();
17486 +void __raise_softirq_irqoff(unsigned int nr)
17488 + do_raise_softirq_irqoff(nr);
17489 + if (!in_irq() && !current->softirq_nestcnt)
17490 + wakeup_proper_softirq(nr);
17494 + * Same as __raise_softirq_irqoff() but will process them in ksoftirqd
17496 +void __raise_softirq_irqoff_ksoft(unsigned int nr)
17498 + unsigned int mask;
17500 + if (WARN_ON_ONCE(!__this_cpu_read(ksoftirqd) ||
17501 + !__this_cpu_read(ktimer_softirqd)))
17503 + mask = 1UL << nr;
17505 + trace_softirq_raise(nr);
17506 + or_softirq_pending(mask);
17507 + if (mask & TIMER_SOFTIRQS)
17508 + __this_cpu_read(ktimer_softirqd)->softirqs_raised |= mask;
17510 + __this_cpu_read(ksoftirqd)->softirqs_raised |= mask;
17511 + wakeup_proper_softirq(nr);
17515 + * This function must run with irqs disabled!
17517 +void raise_softirq_irqoff(unsigned int nr)
17519 + do_raise_softirq_irqoff(nr);
17522 + * If we're in an hard interrupt we let irq return code deal
17523 + * with the wakeup of ksoftirqd.
17528 + * If we are in thread context but outside of a bh disabled
17529 + * region, we need to wake ksoftirqd as well.
17531 + * CHECKME: Some of the places which do that could be wrapped
17532 + * into local_bh_disable/enable pairs. Though it's unclear
17533 + * whether this is worth the effort. To find those places just
17534 + * raise a WARN() if the condition is met.
17536 + if (!current->softirq_nestcnt)
17537 + wakeup_proper_softirq(nr);
17540 +static inline int ksoftirqd_softirq_pending(void)
17542 + return current->softirqs_raised;
17545 +static inline void local_bh_disable_nort(void) { }
17546 +static inline void _local_bh_enable_nort(void) { }
17548 +static inline void ksoftirqd_set_sched_params(unsigned int cpu)
17550 + /* Take over all but timer pending softirqs when starting */
17551 + local_irq_disable();
17552 + current->softirqs_raised = local_softirq_pending() & ~TIMER_SOFTIRQS;
17553 + local_irq_enable();
17556 +static inline void ktimer_softirqd_set_sched_params(unsigned int cpu)
17558 + struct sched_param param = { .sched_priority = 1 };
17560 + sched_setscheduler(current, SCHED_FIFO, ¶m);
17562 + /* Take over timer pending softirqs when starting */
17563 + local_irq_disable();
17564 + current->softirqs_raised = local_softirq_pending() & TIMER_SOFTIRQS;
17565 + local_irq_enable();
17568 +static inline void ktimer_softirqd_clr_sched_params(unsigned int cpu,
17571 + struct sched_param param = { .sched_priority = 0 };
17573 + sched_setscheduler(current, SCHED_NORMAL, ¶m);
17576 +static int ktimer_softirqd_should_run(unsigned int cpu)
17578 + return current->softirqs_raised;
17581 +#endif /* PREEMPT_RT_FULL */
17583 * Enter an interrupt context.
17585 void irq_enter(void)
17586 @@ -341,9 +784,9 @@ void irq_enter(void)
17587 * Prevent raise_softirq from needlessly waking up ksoftirqd
17588 * here, as softirq will be serviced on return from interrupt.
17590 - local_bh_disable();
17591 + local_bh_disable_nort();
17593 - _local_bh_enable();
17594 + _local_bh_enable_nort();
17598 @@ -351,6 +794,7 @@ void irq_enter(void)
17600 static inline void invoke_softirq(void)
17602 +#ifndef CONFIG_PREEMPT_RT_FULL
17603 if (ksoftirqd_running())
17606 @@ -373,6 +817,18 @@ static inline void invoke_softirq(void)
17610 +#else /* PREEMPT_RT_FULL */
17611 + unsigned long flags;
17613 + local_irq_save(flags);
17614 + if (__this_cpu_read(ksoftirqd) &&
17615 + __this_cpu_read(ksoftirqd)->softirqs_raised)
17616 + wakeup_softirqd();
17617 + if (__this_cpu_read(ktimer_softirqd) &&
17618 + __this_cpu_read(ktimer_softirqd)->softirqs_raised)
17619 + wakeup_timer_softirqd();
17620 + local_irq_restore(flags);
17624 static inline void tick_irq_exit(void)
17625 @@ -409,26 +865,6 @@ void irq_exit(void)
17626 trace_hardirq_exit(); /* must be last! */
17630 - * This function must run with irqs disabled!
17632 -inline void raise_softirq_irqoff(unsigned int nr)
17634 - __raise_softirq_irqoff(nr);
17637 - * If we're in an interrupt or softirq, we're done
17638 - * (this also catches softirq-disabled code). We will
17639 - * actually run the softirq once we return from
17640 - * the irq or softirq.
17642 - * Otherwise we wake up ksoftirqd to make sure we
17643 - * schedule the softirq soon.
17645 - if (!in_interrupt())
17646 - wakeup_softirqd();
17649 void raise_softirq(unsigned int nr)
17651 unsigned long flags;
17652 @@ -438,12 +874,6 @@ void raise_softirq(unsigned int nr)
17653 local_irq_restore(flags);
17656 -void __raise_softirq_irqoff(unsigned int nr)
17658 - trace_softirq_raise(nr);
17659 - or_softirq_pending(1UL << nr);
17662 void open_softirq(int nr, void (*action)(struct softirq_action *))
17664 softirq_vec[nr].action = action;
17665 @@ -460,15 +890,45 @@ struct tasklet_head {
17666 static DEFINE_PER_CPU(struct tasklet_head, tasklet_vec);
17667 static DEFINE_PER_CPU(struct tasklet_head, tasklet_hi_vec);
17669 +static void inline
17670 +__tasklet_common_schedule(struct tasklet_struct *t, struct tasklet_head *head, unsigned int nr)
17672 + if (tasklet_trylock(t)) {
17674 + /* We may have been preempted before tasklet_trylock
17675 + * and __tasklet_action may have already run.
17676 + * So double check the sched bit while the takslet
17677 + * is locked before adding it to the list.
17679 + if (test_bit(TASKLET_STATE_SCHED, &t->state)) {
17682 + head->tail = &(t->next);
17683 + raise_softirq_irqoff(nr);
17684 + tasklet_unlock(t);
17686 + /* This is subtle. If we hit the corner case above
17687 + * It is possible that we get preempted right here,
17688 + * and another task has successfully called
17689 + * tasklet_schedule(), then this function, and
17690 + * failed on the trylock. Thus we must be sure
17691 + * before releasing the tasklet lock, that the
17692 + * SCHED_BIT is clear. Otherwise the tasklet
17693 + * may get its SCHED_BIT set, but not added to the
17696 + if (!tasklet_tryunlock(t))
17702 void __tasklet_schedule(struct tasklet_struct *t)
17704 unsigned long flags;
17706 local_irq_save(flags);
17708 - *__this_cpu_read(tasklet_vec.tail) = t;
17709 - __this_cpu_write(tasklet_vec.tail, &(t->next));
17710 - raise_softirq_irqoff(TASKLET_SOFTIRQ);
17711 + __tasklet_common_schedule(t, this_cpu_ptr(&tasklet_vec), TASKLET_SOFTIRQ);
17712 local_irq_restore(flags);
17714 EXPORT_SYMBOL(__tasklet_schedule);
17715 @@ -478,10 +938,7 @@ void __tasklet_hi_schedule(struct tasklet_struct *t)
17716 unsigned long flags;
17718 local_irq_save(flags);
17720 - *__this_cpu_read(tasklet_hi_vec.tail) = t;
17721 - __this_cpu_write(tasklet_hi_vec.tail, &(t->next));
17722 - raise_softirq_irqoff(HI_SOFTIRQ);
17723 + __tasklet_common_schedule(t, this_cpu_ptr(&tasklet_hi_vec), HI_SOFTIRQ);
17724 local_irq_restore(flags);
17726 EXPORT_SYMBOL(__tasklet_hi_schedule);
17727 @@ -490,82 +947,122 @@ void __tasklet_hi_schedule_first(struct tasklet_struct *t)
17729 BUG_ON(!irqs_disabled());
17731 - t->next = __this_cpu_read(tasklet_hi_vec.head);
17732 - __this_cpu_write(tasklet_hi_vec.head, t);
17733 - __raise_softirq_irqoff(HI_SOFTIRQ);
17734 + __tasklet_hi_schedule(t);
17736 EXPORT_SYMBOL(__tasklet_hi_schedule_first);
17738 -static __latent_entropy void tasklet_action(struct softirq_action *a)
17739 +void tasklet_enable(struct tasklet_struct *t)
17741 - struct tasklet_struct *list;
17742 + if (!atomic_dec_and_test(&t->count))
17744 + if (test_and_clear_bit(TASKLET_STATE_PENDING, &t->state))
17745 + tasklet_schedule(t);
17747 +EXPORT_SYMBOL(tasklet_enable);
17749 - local_irq_disable();
17750 - list = __this_cpu_read(tasklet_vec.head);
17751 - __this_cpu_write(tasklet_vec.head, NULL);
17752 - __this_cpu_write(tasklet_vec.tail, this_cpu_ptr(&tasklet_vec.head));
17753 - local_irq_enable();
17754 +static void __tasklet_action(struct softirq_action *a,
17755 + struct tasklet_struct *list)
17757 + int loops = 1000000;
17760 struct tasklet_struct *t = list;
17764 - if (tasklet_trylock(t)) {
17765 - if (!atomic_read(&t->count)) {
17766 - if (!test_and_clear_bit(TASKLET_STATE_SCHED,
17769 - t->func(t->data);
17770 - tasklet_unlock(t);
17773 - tasklet_unlock(t);
17775 + * Should always succeed - after a tasklist got on the
17776 + * list (after getting the SCHED bit set from 0 to 1),
17777 + * nothing but the tasklet softirq it got queued to can
17780 + if (!tasklet_trylock(t)) {
17785 - local_irq_disable();
17787 - *__this_cpu_read(tasklet_vec.tail) = t;
17788 - __this_cpu_write(tasklet_vec.tail, &(t->next));
17789 - __raise_softirq_irqoff(TASKLET_SOFTIRQ);
17790 - local_irq_enable();
17793 + * If we cannot handle the tasklet because it's disabled,
17794 + * mark it as pending. tasklet_enable() will later
17795 + * re-schedule the tasklet.
17797 + if (unlikely(atomic_read(&t->count))) {
17799 + /* implicit unlock: */
17801 + t->state = TASKLET_STATEF_PENDING;
17806 + * After this point on the tasklet might be rescheduled
17807 + * on another CPU, but it can only be added to another
17808 + * CPU's tasklet list if we unlock the tasklet (which we
17811 + if (!test_and_clear_bit(TASKLET_STATE_SCHED, &t->state))
17815 + t->func(t->data);
17818 + * Try to unlock the tasklet. We must use cmpxchg, because
17819 + * another CPU might have scheduled or disabled the tasklet.
17820 + * We only allow the STATE_RUN -> 0 transition here.
17822 + while (!tasklet_tryunlock(t)) {
17824 + * If it got disabled meanwhile, bail out:
17826 + if (atomic_read(&t->count))
17827 + goto out_disabled;
17829 + * If it got scheduled meanwhile, re-execute
17830 + * the tasklet function:
17832 + if (test_and_clear_bit(TASKLET_STATE_SCHED, &t->state))
17835 + printk("hm, tasklet state: %08lx\n", t->state);
17837 + tasklet_unlock(t);
17844 +static void tasklet_action(struct softirq_action *a)
17846 + struct tasklet_struct *list;
17848 + local_irq_disable();
17850 + list = __this_cpu_read(tasklet_vec.head);
17851 + __this_cpu_write(tasklet_vec.head, NULL);
17852 + __this_cpu_write(tasklet_vec.tail, this_cpu_ptr(&tasklet_vec.head));
17854 + local_irq_enable();
17856 + __tasklet_action(a, list);
17859 static __latent_entropy void tasklet_hi_action(struct softirq_action *a)
17861 struct tasklet_struct *list;
17863 local_irq_disable();
17865 list = __this_cpu_read(tasklet_hi_vec.head);
17866 __this_cpu_write(tasklet_hi_vec.head, NULL);
17867 __this_cpu_write(tasklet_hi_vec.tail, this_cpu_ptr(&tasklet_hi_vec.head));
17869 local_irq_enable();
17872 - struct tasklet_struct *t = list;
17874 - list = list->next;
17876 - if (tasklet_trylock(t)) {
17877 - if (!atomic_read(&t->count)) {
17878 - if (!test_and_clear_bit(TASKLET_STATE_SCHED,
17881 - t->func(t->data);
17882 - tasklet_unlock(t);
17885 - tasklet_unlock(t);
17888 - local_irq_disable();
17890 - *__this_cpu_read(tasklet_hi_vec.tail) = t;
17891 - __this_cpu_write(tasklet_hi_vec.tail, &(t->next));
17892 - __raise_softirq_irqoff(HI_SOFTIRQ);
17893 - local_irq_enable();
17895 + __tasklet_action(a, list);
17898 void tasklet_init(struct tasklet_struct *t,
17899 @@ -586,7 +1083,7 @@ void tasklet_kill(struct tasklet_struct *t)
17901 while (test_and_set_bit(TASKLET_STATE_SCHED, &t->state)) {
17905 } while (test_bit(TASKLET_STATE_SCHED, &t->state));
17907 tasklet_unlock_wait(t);
17908 @@ -660,25 +1157,26 @@ void __init softirq_init(void)
17909 open_softirq(HI_SOFTIRQ, tasklet_hi_action);
17912 +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT_FULL)
17913 +void tasklet_unlock_wait(struct tasklet_struct *t)
17915 + while (test_bit(TASKLET_STATE_RUN, &(t)->state)) {
17917 + * Hack for now to avoid this busy-loop:
17919 +#ifdef CONFIG_PREEMPT_RT_FULL
17926 +EXPORT_SYMBOL(tasklet_unlock_wait);
17929 static int ksoftirqd_should_run(unsigned int cpu)
17931 - return local_softirq_pending();
17934 -static void run_ksoftirqd(unsigned int cpu)
17936 - local_irq_disable();
17937 - if (local_softirq_pending()) {
17939 - * We can safely run softirq on inline stack, as we are not deep
17940 - * in the task stack here.
17943 - local_irq_enable();
17944 - cond_resched_rcu_qs();
17947 - local_irq_enable();
17948 + return ksoftirqd_softirq_pending();
17951 #ifdef CONFIG_HOTPLUG_CPU
17952 @@ -745,17 +1243,31 @@ static int takeover_tasklets(unsigned int cpu)
17954 static struct smp_hotplug_thread softirq_threads = {
17955 .store = &ksoftirqd,
17956 + .setup = ksoftirqd_set_sched_params,
17957 .thread_should_run = ksoftirqd_should_run,
17958 .thread_fn = run_ksoftirqd,
17959 .thread_comm = "ksoftirqd/%u",
17962 +#ifdef CONFIG_PREEMPT_RT_FULL
17963 +static struct smp_hotplug_thread softirq_timer_threads = {
17964 + .store = &ktimer_softirqd,
17965 + .setup = ktimer_softirqd_set_sched_params,
17966 + .cleanup = ktimer_softirqd_clr_sched_params,
17967 + .thread_should_run = ktimer_softirqd_should_run,
17968 + .thread_fn = run_ksoftirqd,
17969 + .thread_comm = "ktimersoftd/%u",
17973 static __init int spawn_ksoftirqd(void)
17975 cpuhp_setup_state_nocalls(CPUHP_SOFTIRQ_DEAD, "softirq:dead", NULL,
17976 takeover_tasklets);
17977 BUG_ON(smpboot_register_percpu_thread(&softirq_threads));
17979 +#ifdef CONFIG_PREEMPT_RT_FULL
17980 + BUG_ON(smpboot_register_percpu_thread(&softirq_timer_threads));
17984 early_initcall(spawn_ksoftirqd);
17985 diff --git a/kernel/stop_machine.c b/kernel/stop_machine.c
17986 index ec9ab2f01489..8b89dbedeaff 100644
17987 --- a/kernel/stop_machine.c
17988 +++ b/kernel/stop_machine.c
17989 @@ -36,7 +36,7 @@ struct cpu_stop_done {
17990 struct cpu_stopper {
17991 struct task_struct *thread;
17994 + raw_spinlock_t lock;
17995 bool enabled; /* is this stopper enabled? */
17996 struct list_head works; /* list of pending works */
17998 @@ -78,14 +78,14 @@ static bool cpu_stop_queue_work(unsigned int cpu, struct cpu_stop_work *work)
17999 unsigned long flags;
18002 - spin_lock_irqsave(&stopper->lock, flags);
18003 + raw_spin_lock_irqsave(&stopper->lock, flags);
18004 enabled = stopper->enabled;
18006 __cpu_stop_queue_work(stopper, work);
18007 else if (work->done)
18008 cpu_stop_signal_done(work->done);
18009 - spin_unlock_irqrestore(&stopper->lock, flags);
18011 + raw_spin_unlock_irqrestore(&stopper->lock, flags);
18015 @@ -231,8 +231,8 @@ static int cpu_stop_queue_two_works(int cpu1, struct cpu_stop_work *work1,
18016 struct cpu_stopper *stopper2 = per_cpu_ptr(&cpu_stopper, cpu2);
18019 - spin_lock_irq(&stopper1->lock);
18020 - spin_lock_nested(&stopper2->lock, SINGLE_DEPTH_NESTING);
18021 + raw_spin_lock_irq(&stopper1->lock);
18022 + raw_spin_lock_nested(&stopper2->lock, SINGLE_DEPTH_NESTING);
18025 if (!stopper1->enabled || !stopper2->enabled)
18026 @@ -255,8 +255,8 @@ static int cpu_stop_queue_two_works(int cpu1, struct cpu_stop_work *work1,
18027 __cpu_stop_queue_work(stopper1, work1);
18028 __cpu_stop_queue_work(stopper2, work2);
18030 - spin_unlock(&stopper2->lock);
18031 - spin_unlock_irq(&stopper1->lock);
18032 + raw_spin_unlock(&stopper2->lock);
18033 + raw_spin_unlock_irq(&stopper1->lock);
18035 if (unlikely(err == -EDEADLK)) {
18036 while (stop_cpus_in_progress)
18037 @@ -448,9 +448,9 @@ static int cpu_stop_should_run(unsigned int cpu)
18038 unsigned long flags;
18041 - spin_lock_irqsave(&stopper->lock, flags);
18042 + raw_spin_lock_irqsave(&stopper->lock, flags);
18043 run = !list_empty(&stopper->works);
18044 - spin_unlock_irqrestore(&stopper->lock, flags);
18045 + raw_spin_unlock_irqrestore(&stopper->lock, flags);
18049 @@ -461,13 +461,13 @@ static void cpu_stopper_thread(unsigned int cpu)
18053 - spin_lock_irq(&stopper->lock);
18054 + raw_spin_lock_irq(&stopper->lock);
18055 if (!list_empty(&stopper->works)) {
18056 work = list_first_entry(&stopper->works,
18057 struct cpu_stop_work, list);
18058 list_del_init(&work->list);
18060 - spin_unlock_irq(&stopper->lock);
18061 + raw_spin_unlock_irq(&stopper->lock);
18064 cpu_stop_fn_t fn = work->fn;
18065 @@ -475,6 +475,8 @@ static void cpu_stopper_thread(unsigned int cpu)
18066 struct cpu_stop_done *done = work->done;
18071 /* cpu stop callbacks must not sleep, make in_atomic() == T */
18072 preempt_count_inc();
18074 @@ -541,7 +543,7 @@ static int __init cpu_stop_init(void)
18075 for_each_possible_cpu(cpu) {
18076 struct cpu_stopper *stopper = &per_cpu(cpu_stopper, cpu);
18078 - spin_lock_init(&stopper->lock);
18079 + raw_spin_lock_init(&stopper->lock);
18080 INIT_LIST_HEAD(&stopper->works);
18083 diff --git a/kernel/time/hrtimer.c b/kernel/time/hrtimer.c
18084 index bb5ec425dfe0..8338b14ed3a3 100644
18085 --- a/kernel/time/hrtimer.c
18086 +++ b/kernel/time/hrtimer.c
18088 #include <asm/uaccess.h>
18090 #include <trace/events/timer.h>
18091 +#include <trace/events/hist.h>
18093 #include "tick-internal.h"
18095 @@ -695,6 +696,29 @@ static void hrtimer_switch_to_hres(void)
18096 retrigger_next_event(NULL);
18099 +#ifdef CONFIG_PREEMPT_RT_FULL
18101 +static struct swork_event clock_set_delay_work;
18103 +static void run_clock_set_delay(struct swork_event *event)
18108 +void clock_was_set_delayed(void)
18110 + swork_queue(&clock_set_delay_work);
18113 +static __init int create_clock_set_delay_thread(void)
18115 + WARN_ON(swork_get());
18116 + INIT_SWORK(&clock_set_delay_work, run_clock_set_delay);
18119 +early_initcall(create_clock_set_delay_thread);
18120 +#else /* PREEMPT_RT_FULL */
18122 static void clock_was_set_work(struct work_struct *work)
18125 @@ -710,6 +734,7 @@ void clock_was_set_delayed(void)
18127 schedule_work(&hrtimer_work);
18133 @@ -719,11 +744,8 @@ static inline int hrtimer_is_hres_enabled(void) { return 0; }
18134 static inline void hrtimer_switch_to_hres(void) { }
18136 hrtimer_force_reprogram(struct hrtimer_cpu_base *base, int skip_equal) { }
18137 -static inline int hrtimer_reprogram(struct hrtimer *timer,
18138 - struct hrtimer_clock_base *base)
18142 +static inline void hrtimer_reprogram(struct hrtimer *timer,
18143 + struct hrtimer_clock_base *base) { }
18144 static inline void hrtimer_init_hres(struct hrtimer_cpu_base *base) { }
18145 static inline void retrigger_next_event(void *arg) { }
18147 @@ -855,6 +877,32 @@ u64 hrtimer_forward(struct hrtimer *timer, ktime_t now, ktime_t interval)
18149 EXPORT_SYMBOL_GPL(hrtimer_forward);
18151 +#ifdef CONFIG_PREEMPT_RT_BASE
18152 +# define wake_up_timer_waiters(b) wake_up(&(b)->wait)
18155 + * hrtimer_wait_for_timer - Wait for a running timer
18157 + * @timer: timer to wait for
18159 + * The function waits in case the timers callback function is
18160 + * currently executed on the waitqueue of the timer base. The
18161 + * waitqueue is woken up after the timer callback function has
18162 + * finished execution.
18164 +void hrtimer_wait_for_timer(const struct hrtimer *timer)
18166 + struct hrtimer_clock_base *base = timer->base;
18168 + if (base && base->cpu_base && !timer->irqsafe)
18169 + wait_event(base->cpu_base->wait,
18170 + !(hrtimer_callback_running(timer)));
18174 +# define wake_up_timer_waiters(b) do { } while (0)
18178 * enqueue_hrtimer - internal function to (re)start a timer
18180 @@ -896,6 +944,11 @@ static void __remove_hrtimer(struct hrtimer *timer,
18181 if (!(state & HRTIMER_STATE_ENQUEUED))
18184 + if (unlikely(!list_empty(&timer->cb_entry))) {
18185 + list_del_init(&timer->cb_entry);
18189 if (!timerqueue_del(&base->active, &timer->node))
18190 cpu_base->active_bases &= ~(1 << base->index);
18192 @@ -991,7 +1044,16 @@ void hrtimer_start_range_ns(struct hrtimer *timer, ktime_t tim,
18193 new_base = switch_hrtimer_base(timer, base, mode & HRTIMER_MODE_PINNED);
18195 timer_stats_hrtimer_set_start_info(timer);
18196 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
18198 + ktime_t now = new_base->get_time();
18200 + if (ktime_to_ns(tim) < ktime_to_ns(now))
18201 + timer->praecox = now;
18203 + timer->praecox = ktime_set(0, 0);
18206 leftmost = enqueue_hrtimer(timer, new_base);
18209 @@ -1063,7 +1125,7 @@ int hrtimer_cancel(struct hrtimer *timer)
18214 + hrtimer_wait_for_timer(timer);
18217 EXPORT_SYMBOL_GPL(hrtimer_cancel);
18218 @@ -1127,6 +1189,7 @@ static void __hrtimer_init(struct hrtimer *timer, clockid_t clock_id,
18220 base = hrtimer_clockid_to_base(clock_id);
18221 timer->base = &cpu_base->clock_base[base];
18222 + INIT_LIST_HEAD(&timer->cb_entry);
18223 timerqueue_init(&timer->node);
18225 #ifdef CONFIG_TIMER_STATS
18226 @@ -1167,6 +1230,7 @@ bool hrtimer_active(const struct hrtimer *timer)
18227 seq = raw_read_seqcount_begin(&cpu_base->seq);
18229 if (timer->state != HRTIMER_STATE_INACTIVE ||
18230 + cpu_base->running_soft == timer ||
18231 cpu_base->running == timer)
18234 @@ -1265,10 +1329,112 @@ static void __run_hrtimer(struct hrtimer_cpu_base *cpu_base,
18235 cpu_base->running = NULL;
18238 +#ifdef CONFIG_PREEMPT_RT_BASE
18239 +static void hrtimer_rt_reprogram(int restart, struct hrtimer *timer,
18240 + struct hrtimer_clock_base *base)
18244 + if (restart != HRTIMER_NORESTART &&
18245 + !(timer->state & HRTIMER_STATE_ENQUEUED)) {
18247 + leftmost = enqueue_hrtimer(timer, base);
18250 +#ifdef CONFIG_HIGH_RES_TIMERS
18251 + if (!hrtimer_is_hres_active(timer)) {
18253 + * Kick to reschedule the next tick to handle the new timer
18254 + * on dynticks target.
18256 + if (base->cpu_base->nohz_active)
18257 + wake_up_nohz_cpu(base->cpu_base->cpu);
18260 + hrtimer_reprogram(timer, base);
18267 + * The changes in mainline which removed the callback modes from
18268 + * hrtimer are not yet working with -rt. The non wakeup_process()
18269 + * based callbacks which involve sleeping locks need to be treated
18272 +static void hrtimer_rt_run_pending(void)
18274 + enum hrtimer_restart (*fn)(struct hrtimer *);
18275 + struct hrtimer_cpu_base *cpu_base;
18276 + struct hrtimer_clock_base *base;
18277 + struct hrtimer *timer;
18278 + int index, restart;
18280 + local_irq_disable();
18281 + cpu_base = &per_cpu(hrtimer_bases, smp_processor_id());
18283 + raw_spin_lock(&cpu_base->lock);
18285 + for (index = 0; index < HRTIMER_MAX_CLOCK_BASES; index++) {
18286 + base = &cpu_base->clock_base[index];
18288 + while (!list_empty(&base->expired)) {
18289 + timer = list_first_entry(&base->expired,
18290 + struct hrtimer, cb_entry);
18293 + * Same as the above __run_hrtimer function
18294 + * just we run with interrupts enabled.
18296 + debug_deactivate(timer);
18297 + cpu_base->running_soft = timer;
18298 + raw_write_seqcount_barrier(&cpu_base->seq);
18300 + __remove_hrtimer(timer, base, HRTIMER_STATE_INACTIVE, 0);
18301 + timer_stats_account_hrtimer(timer);
18302 + fn = timer->function;
18304 + raw_spin_unlock_irq(&cpu_base->lock);
18305 + restart = fn(timer);
18306 + raw_spin_lock_irq(&cpu_base->lock);
18308 + hrtimer_rt_reprogram(restart, timer, base);
18309 + raw_write_seqcount_barrier(&cpu_base->seq);
18311 + WARN_ON_ONCE(cpu_base->running_soft != timer);
18312 + cpu_base->running_soft = NULL;
18316 + raw_spin_unlock_irq(&cpu_base->lock);
18318 + wake_up_timer_waiters(cpu_base);
18321 +static int hrtimer_rt_defer(struct hrtimer *timer)
18323 + if (timer->irqsafe)
18326 + __remove_hrtimer(timer, timer->base, timer->state, 0);
18327 + list_add_tail(&timer->cb_entry, &timer->base->expired);
18333 +static inline int hrtimer_rt_defer(struct hrtimer *timer) { return 0; }
18337 +static enum hrtimer_restart hrtimer_wakeup(struct hrtimer *timer);
18339 static void __hrtimer_run_queues(struct hrtimer_cpu_base *cpu_base, ktime_t now)
18341 struct hrtimer_clock_base *base = cpu_base->clock_base;
18342 unsigned int active = cpu_base->active_bases;
18345 for (; active; base++, active >>= 1) {
18346 struct timerqueue_node *node;
18347 @@ -1284,6 +1450,15 @@ static void __hrtimer_run_queues(struct hrtimer_cpu_base *cpu_base, ktime_t now)
18349 timer = container_of(node, struct hrtimer, node);
18351 + trace_hrtimer_interrupt(raw_smp_processor_id(),
18352 + ktime_to_ns(ktime_sub(ktime_to_ns(timer->praecox) ?
18353 + timer->praecox : hrtimer_get_expires(timer),
18356 + timer->function == hrtimer_wakeup ?
18357 + container_of(timer, struct hrtimer_sleeper,
18358 + timer)->task : NULL);
18361 * The immediate goal for using the softexpires is
18362 * minimizing wakeups, not running timers at the
18363 @@ -1299,9 +1474,14 @@ static void __hrtimer_run_queues(struct hrtimer_cpu_base *cpu_base, ktime_t now)
18364 if (basenow.tv64 < hrtimer_get_softexpires_tv64(timer))
18367 - __run_hrtimer(cpu_base, base, timer, &basenow);
18368 + if (!hrtimer_rt_defer(timer))
18369 + __run_hrtimer(cpu_base, base, timer, &basenow);
18375 + raise_softirq_irqoff(HRTIMER_SOFTIRQ);
18378 #ifdef CONFIG_HIGH_RES_TIMERS
18379 @@ -1464,16 +1644,18 @@ static enum hrtimer_restart hrtimer_wakeup(struct hrtimer *timer)
18380 void hrtimer_init_sleeper(struct hrtimer_sleeper *sl, struct task_struct *task)
18382 sl->timer.function = hrtimer_wakeup;
18383 + sl->timer.irqsafe = 1;
18386 EXPORT_SYMBOL_GPL(hrtimer_init_sleeper);
18388 -static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode)
18389 +static int __sched do_nanosleep(struct hrtimer_sleeper *t, enum hrtimer_mode mode,
18390 + unsigned long state)
18392 hrtimer_init_sleeper(t, current);
18395 - set_current_state(TASK_INTERRUPTIBLE);
18396 + set_current_state(state);
18397 hrtimer_start_expires(&t->timer, mode);
18399 if (likely(t->task))
18400 @@ -1515,7 +1697,8 @@ long __sched hrtimer_nanosleep_restart(struct restart_block *restart)
18402 hrtimer_set_expires_tv64(&t.timer, restart->nanosleep.expires);
18404 - if (do_nanosleep(&t, HRTIMER_MODE_ABS))
18405 + /* cpu_chill() does not care about restart state. */
18406 + if (do_nanosleep(&t, HRTIMER_MODE_ABS, TASK_INTERRUPTIBLE))
18409 rmtp = restart->nanosleep.rmtp;
18410 @@ -1532,8 +1715,10 @@ long __sched hrtimer_nanosleep_restart(struct restart_block *restart)
18414 -long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
18415 - const enum hrtimer_mode mode, const clockid_t clockid)
18417 +__hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
18418 + const enum hrtimer_mode mode, const clockid_t clockid,
18419 + unsigned long state)
18421 struct restart_block *restart;
18422 struct hrtimer_sleeper t;
18423 @@ -1546,7 +1731,7 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
18425 hrtimer_init_on_stack(&t.timer, clockid, mode);
18426 hrtimer_set_expires_range_ns(&t.timer, timespec_to_ktime(*rqtp), slack);
18427 - if (do_nanosleep(&t, mode))
18428 + if (do_nanosleep(&t, mode, state))
18431 /* Absolute timers do not update the rmtp value and restart: */
18432 @@ -1573,6 +1758,12 @@ long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
18436 +long hrtimer_nanosleep(struct timespec *rqtp, struct timespec __user *rmtp,
18437 + const enum hrtimer_mode mode, const clockid_t clockid)
18439 + return __hrtimer_nanosleep(rqtp, rmtp, mode, clockid, TASK_INTERRUPTIBLE);
18442 SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp,
18443 struct timespec __user *, rmtp)
18445 @@ -1587,6 +1778,26 @@ SYSCALL_DEFINE2(nanosleep, struct timespec __user *, rqtp,
18446 return hrtimer_nanosleep(&tu, rmtp, HRTIMER_MODE_REL, CLOCK_MONOTONIC);
18449 +#ifdef CONFIG_PREEMPT_RT_FULL
18451 + * Sleep for 1 ms in hope whoever holds what we want will let it go.
18453 +void cpu_chill(void)
18455 + struct timespec tu = {
18456 + .tv_nsec = NSEC_PER_MSEC,
18458 + unsigned int freeze_flag = current->flags & PF_NOFREEZE;
18460 + current->flags |= PF_NOFREEZE;
18461 + __hrtimer_nanosleep(&tu, NULL, HRTIMER_MODE_REL, CLOCK_MONOTONIC,
18462 + TASK_UNINTERRUPTIBLE);
18463 + if (!freeze_flag)
18464 + current->flags &= ~PF_NOFREEZE;
18466 +EXPORT_SYMBOL(cpu_chill);
18470 * Functions related to boot-time initialization:
18472 @@ -1598,10 +1809,14 @@ int hrtimers_prepare_cpu(unsigned int cpu)
18473 for (i = 0; i < HRTIMER_MAX_CLOCK_BASES; i++) {
18474 cpu_base->clock_base[i].cpu_base = cpu_base;
18475 timerqueue_init_head(&cpu_base->clock_base[i].active);
18476 + INIT_LIST_HEAD(&cpu_base->clock_base[i].expired);
18479 cpu_base->cpu = cpu;
18480 hrtimer_init_hres(cpu_base);
18481 +#ifdef CONFIG_PREEMPT_RT_BASE
18482 + init_waitqueue_head(&cpu_base->wait);
18487 @@ -1671,9 +1886,26 @@ int hrtimers_dead_cpu(unsigned int scpu)
18489 #endif /* CONFIG_HOTPLUG_CPU */
18491 +#ifdef CONFIG_PREEMPT_RT_BASE
18493 +static void run_hrtimer_softirq(struct softirq_action *h)
18495 + hrtimer_rt_run_pending();
18498 +static void hrtimers_open_softirq(void)
18500 + open_softirq(HRTIMER_SOFTIRQ, run_hrtimer_softirq);
18504 +static void hrtimers_open_softirq(void) { }
18507 void __init hrtimers_init(void)
18509 hrtimers_prepare_cpu(smp_processor_id());
18510 + hrtimers_open_softirq();
18514 diff --git a/kernel/time/itimer.c b/kernel/time/itimer.c
18515 index 1d5c7204ddc9..184de6751180 100644
18516 --- a/kernel/time/itimer.c
18517 +++ b/kernel/time/itimer.c
18518 @@ -213,6 +213,7 @@ int do_setitimer(int which, struct itimerval *value, struct itimerval *ovalue)
18519 /* We are sharing ->siglock with it_real_fn() */
18520 if (hrtimer_try_to_cancel(timer) < 0) {
18521 spin_unlock_irq(&tsk->sighand->siglock);
18522 + hrtimer_wait_for_timer(&tsk->signal->real_timer);
18525 expires = timeval_to_ktime(value->it_value);
18526 diff --git a/kernel/time/jiffies.c b/kernel/time/jiffies.c
18527 index 555e21f7b966..a5d6435fabbb 100644
18528 --- a/kernel/time/jiffies.c
18529 +++ b/kernel/time/jiffies.c
18530 @@ -74,7 +74,8 @@ static struct clocksource clocksource_jiffies = {
18534 -__cacheline_aligned_in_smp DEFINE_SEQLOCK(jiffies_lock);
18535 +__cacheline_aligned_in_smp DEFINE_RAW_SPINLOCK(jiffies_lock);
18536 +__cacheline_aligned_in_smp seqcount_t jiffies_seq;
18538 #if (BITS_PER_LONG < 64)
18539 u64 get_jiffies_64(void)
18540 @@ -83,9 +84,9 @@ u64 get_jiffies_64(void)
18544 - seq = read_seqbegin(&jiffies_lock);
18545 + seq = read_seqcount_begin(&jiffies_seq);
18547 - } while (read_seqretry(&jiffies_lock, seq));
18548 + } while (read_seqcount_retry(&jiffies_seq, seq));
18551 EXPORT_SYMBOL(get_jiffies_64);
18552 diff --git a/kernel/time/ntp.c b/kernel/time/ntp.c
18553 index 6df8927c58a5..05b7391bf9bd 100644
18554 --- a/kernel/time/ntp.c
18555 +++ b/kernel/time/ntp.c
18557 #include <linux/module.h>
18558 #include <linux/rtc.h>
18559 #include <linux/math64.h>
18560 +#include <linux/swork.h>
18562 #include "ntp_internal.h"
18563 #include "timekeeping_internal.h"
18564 @@ -568,10 +569,35 @@ static void sync_cmos_clock(struct work_struct *work)
18565 &sync_cmos_work, timespec64_to_jiffies(&next));
18568 +#ifdef CONFIG_PREEMPT_RT_FULL
18570 +static void run_clock_set_delay(struct swork_event *event)
18572 + queue_delayed_work(system_power_efficient_wq, &sync_cmos_work, 0);
18575 +static struct swork_event ntp_cmos_swork;
18577 +void ntp_notify_cmos_timer(void)
18579 + swork_queue(&ntp_cmos_swork);
18582 +static __init int create_cmos_delay_thread(void)
18584 + WARN_ON(swork_get());
18585 + INIT_SWORK(&ntp_cmos_swork, run_clock_set_delay);
18588 +early_initcall(create_cmos_delay_thread);
18592 void ntp_notify_cmos_timer(void)
18594 queue_delayed_work(system_power_efficient_wq, &sync_cmos_work, 0);
18596 +#endif /* CONFIG_PREEMPT_RT_FULL */
18599 void ntp_notify_cmos_timer(void) { }
18600 diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c
18601 index 39008d78927a..633f4eaca9e7 100644
18602 --- a/kernel/time/posix-cpu-timers.c
18603 +++ b/kernel/time/posix-cpu-timers.c
18607 #include <linux/sched.h>
18608 +#include <linux/sched/rt.h>
18609 #include <linux/posix-timers.h>
18610 #include <linux/errno.h>
18611 #include <linux/math64.h>
18612 @@ -620,7 +621,7 @@ static int posix_cpu_timer_set(struct k_itimer *timer, int timer_flags,
18614 * Disarm any old timer after extracting its expiry time.
18616 - WARN_ON_ONCE(!irqs_disabled());
18617 + WARN_ON_ONCE_NONRT(!irqs_disabled());
18620 old_incr = timer->it.cpu.incr;
18621 @@ -1064,7 +1065,7 @@ void posix_cpu_timer_schedule(struct k_itimer *timer)
18623 * Now re-arm for the new expiry time.
18625 - WARN_ON_ONCE(!irqs_disabled());
18626 + WARN_ON_ONCE_NONRT(!irqs_disabled());
18628 unlock_task_sighand(p, &flags);
18630 @@ -1153,13 +1154,13 @@ static inline int fastpath_timer_check(struct task_struct *tsk)
18631 * already updated our counts. We need to check if any timers fire now.
18632 * Interrupts are disabled.
18634 -void run_posix_cpu_timers(struct task_struct *tsk)
18635 +static void __run_posix_cpu_timers(struct task_struct *tsk)
18638 struct k_itimer *timer, *next;
18639 unsigned long flags;
18641 - WARN_ON_ONCE(!irqs_disabled());
18642 + WARN_ON_ONCE_NONRT(!irqs_disabled());
18645 * The fast path checks that there are no expired thread or thread
18646 @@ -1213,6 +1214,190 @@ void run_posix_cpu_timers(struct task_struct *tsk)
18650 +#ifdef CONFIG_PREEMPT_RT_BASE
18651 +#include <linux/kthread.h>
18652 +#include <linux/cpu.h>
18653 +DEFINE_PER_CPU(struct task_struct *, posix_timer_task);
18654 +DEFINE_PER_CPU(struct task_struct *, posix_timer_tasklist);
18656 +static int posix_cpu_timers_thread(void *data)
18658 + int cpu = (long)data;
18660 + BUG_ON(per_cpu(posix_timer_task,cpu) != current);
18662 + while (!kthread_should_stop()) {
18663 + struct task_struct *tsk = NULL;
18664 + struct task_struct *next = NULL;
18666 + if (cpu_is_offline(cpu))
18667 + goto wait_to_die;
18669 + /* grab task list */
18670 + raw_local_irq_disable();
18671 + tsk = per_cpu(posix_timer_tasklist, cpu);
18672 + per_cpu(posix_timer_tasklist, cpu) = NULL;
18673 + raw_local_irq_enable();
18675 + /* its possible the list is empty, just return */
18677 + set_current_state(TASK_INTERRUPTIBLE);
18679 + __set_current_state(TASK_RUNNING);
18683 + /* Process task list */
18686 + next = tsk->posix_timer_list;
18688 + /* run the task timers, clear its ptr and
18691 + __run_posix_cpu_timers(tsk);
18692 + tsk->posix_timer_list = NULL;
18693 + put_task_struct(tsk);
18695 + /* check if this is the last on the list */
18704 + /* Wait for kthread_stop */
18705 + set_current_state(TASK_INTERRUPTIBLE);
18706 + while (!kthread_should_stop()) {
18708 + set_current_state(TASK_INTERRUPTIBLE);
18710 + __set_current_state(TASK_RUNNING);
18714 +static inline int __fastpath_timer_check(struct task_struct *tsk)
18716 + /* tsk == current, ensure it is safe to use ->signal/sighand */
18717 + if (unlikely(tsk->exit_state))
18720 + if (!task_cputime_zero(&tsk->cputime_expires))
18723 + if (!task_cputime_zero(&tsk->signal->cputime_expires))
18729 +void run_posix_cpu_timers(struct task_struct *tsk)
18731 + unsigned long cpu = smp_processor_id();
18732 + struct task_struct *tasklist;
18734 + BUG_ON(!irqs_disabled());
18735 + if(!per_cpu(posix_timer_task, cpu))
18737 + /* get per-cpu references */
18738 + tasklist = per_cpu(posix_timer_tasklist, cpu);
18740 + /* check to see if we're already queued */
18741 + if (!tsk->posix_timer_list && __fastpath_timer_check(tsk)) {
18742 + get_task_struct(tsk);
18744 + tsk->posix_timer_list = tasklist;
18747 + * The list is terminated by a self-pointing
18750 + tsk->posix_timer_list = tsk;
18752 + per_cpu(posix_timer_tasklist, cpu) = tsk;
18754 + wake_up_process(per_cpu(posix_timer_task, cpu));
18759 + * posix_cpu_thread_call - callback that gets triggered when a CPU is added.
18760 + * Here we can start up the necessary migration thread for the new CPU.
18762 +static int posix_cpu_thread_call(struct notifier_block *nfb,
18763 + unsigned long action, void *hcpu)
18765 + int cpu = (long)hcpu;
18766 + struct task_struct *p;
18767 + struct sched_param param;
18769 + switch (action) {
18770 + case CPU_UP_PREPARE:
18771 + p = kthread_create(posix_cpu_timers_thread, hcpu,
18772 + "posixcputmr/%d",cpu);
18774 + return NOTIFY_BAD;
18775 + p->flags |= PF_NOFREEZE;
18776 + kthread_bind(p, cpu);
18777 + /* Must be high prio to avoid getting starved */
18778 + param.sched_priority = MAX_RT_PRIO-1;
18779 + sched_setscheduler(p, SCHED_FIFO, ¶m);
18780 + per_cpu(posix_timer_task,cpu) = p;
18783 + /* Strictly unneccessary, as first user will wake it. */
18784 + wake_up_process(per_cpu(posix_timer_task,cpu));
18786 +#ifdef CONFIG_HOTPLUG_CPU
18787 + case CPU_UP_CANCELED:
18788 + /* Unbind it from offline cpu so it can run. Fall thru. */
18789 + kthread_bind(per_cpu(posix_timer_task, cpu),
18790 + cpumask_any(cpu_online_mask));
18791 + kthread_stop(per_cpu(posix_timer_task,cpu));
18792 + per_cpu(posix_timer_task,cpu) = NULL;
18795 + kthread_stop(per_cpu(posix_timer_task,cpu));
18796 + per_cpu(posix_timer_task,cpu) = NULL;
18800 + return NOTIFY_OK;
18803 +/* Register at highest priority so that task migration (migrate_all_tasks)
18804 + * happens before everything else.
18806 +static struct notifier_block posix_cpu_thread_notifier = {
18807 + .notifier_call = posix_cpu_thread_call,
18811 +static int __init posix_cpu_thread_init(void)
18813 + void *hcpu = (void *)(long)smp_processor_id();
18814 + /* Start one for boot CPU. */
18815 + unsigned long cpu;
18817 + /* init the per-cpu posix_timer_tasklets */
18818 + for_each_possible_cpu(cpu)
18819 + per_cpu(posix_timer_tasklist, cpu) = NULL;
18821 + posix_cpu_thread_call(&posix_cpu_thread_notifier, CPU_UP_PREPARE, hcpu);
18822 + posix_cpu_thread_call(&posix_cpu_thread_notifier, CPU_ONLINE, hcpu);
18823 + register_cpu_notifier(&posix_cpu_thread_notifier);
18826 +early_initcall(posix_cpu_thread_init);
18827 +#else /* CONFIG_PREEMPT_RT_BASE */
18828 +void run_posix_cpu_timers(struct task_struct *tsk)
18830 + __run_posix_cpu_timers(tsk);
18832 +#endif /* CONFIG_PREEMPT_RT_BASE */
18835 * Set one of the process-wide special case CPU timers or RLIMIT_CPU.
18836 * The tsk->sighand->siglock must be held by the caller.
18837 diff --git a/kernel/time/posix-timers.c b/kernel/time/posix-timers.c
18838 index f2826c35e918..464a98155a0e 100644
18839 --- a/kernel/time/posix-timers.c
18840 +++ b/kernel/time/posix-timers.c
18841 @@ -506,6 +506,7 @@ static enum hrtimer_restart posix_timer_fn(struct hrtimer *timer)
18842 static struct pid *good_sigevent(sigevent_t * event)
18844 struct task_struct *rtn = current->group_leader;
18845 + int sig = event->sigev_signo;
18847 if ((event->sigev_notify & SIGEV_THREAD_ID ) &&
18848 (!(rtn = find_task_by_vpid(event->sigev_notify_thread_id)) ||
18849 @@ -514,7 +515,8 @@ static struct pid *good_sigevent(sigevent_t * event)
18852 if (((event->sigev_notify & ~SIGEV_THREAD_ID) != SIGEV_NONE) &&
18853 - ((event->sigev_signo <= 0) || (event->sigev_signo > SIGRTMAX)))
18854 + (sig <= 0 || sig > SIGRTMAX || sig_kernel_only(sig) ||
18855 + sig_kernel_coredump(sig)))
18858 return task_pid(rtn);
18859 @@ -826,6 +828,20 @@ SYSCALL_DEFINE1(timer_getoverrun, timer_t, timer_id)
18864 + * Protected by RCU!
18866 +static void timer_wait_for_callback(struct k_clock *kc, struct k_itimer *timr)
18868 +#ifdef CONFIG_PREEMPT_RT_FULL
18869 + if (kc->timer_set == common_timer_set)
18870 + hrtimer_wait_for_timer(&timr->it.real.timer);
18872 + /* FIXME: Whacky hack for posix-cpu-timers */
18873 + schedule_timeout(1);
18877 /* Set a POSIX.1b interval timer. */
18878 /* timr->it_lock is taken. */
18880 @@ -903,6 +919,7 @@ SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags,
18885 kc = clockid_to_kclock(timr->it_clock);
18886 if (WARN_ON_ONCE(!kc || !kc->timer_set))
18888 @@ -911,9 +928,12 @@ SYSCALL_DEFINE4(timer_settime, timer_t, timer_id, int, flags,
18890 unlock_timer(timr, flag);
18891 if (error == TIMER_RETRY) {
18892 + timer_wait_for_callback(kc, timr);
18893 rtn = NULL; // We already got the old time...
18894 + rcu_read_unlock();
18897 + rcu_read_unlock();
18899 if (old_setting && !error &&
18900 copy_to_user(old_setting, &old_spec, sizeof (old_spec)))
18901 @@ -951,10 +971,15 @@ SYSCALL_DEFINE1(timer_delete, timer_t, timer_id)
18906 if (timer_delete_hook(timer) == TIMER_RETRY) {
18907 unlock_timer(timer, flags);
18908 + timer_wait_for_callback(clockid_to_kclock(timer->it_clock),
18910 + rcu_read_unlock();
18913 + rcu_read_unlock();
18915 spin_lock(¤t->sighand->siglock);
18916 list_del(&timer->list);
18917 @@ -980,8 +1005,18 @@ static void itimer_delete(struct k_itimer *timer)
18919 spin_lock_irqsave(&timer->it_lock, flags);
18921 - if (timer_delete_hook(timer) == TIMER_RETRY) {
18922 + /* On RT we can race with a deletion */
18923 + if (!timer->it_signal) {
18924 unlock_timer(timer, flags);
18928 + if (timer_delete_hook(timer) == TIMER_RETRY) {
18930 + unlock_timer(timer, flags);
18931 + timer_wait_for_callback(clockid_to_kclock(timer->it_clock),
18933 + rcu_read_unlock();
18936 list_del(&timer->list);
18937 diff --git a/kernel/time/tick-broadcast-hrtimer.c b/kernel/time/tick-broadcast-hrtimer.c
18938 index 690b797f522e..fe8ba1619879 100644
18939 --- a/kernel/time/tick-broadcast-hrtimer.c
18940 +++ b/kernel/time/tick-broadcast-hrtimer.c
18941 @@ -107,5 +107,6 @@ void tick_setup_hrtimer_broadcast(void)
18943 hrtimer_init(&bctimer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
18944 bctimer.function = bc_handler;
18945 + bctimer.irqsafe = true;
18946 clockevents_register_device(&ce_broadcast_hrtimer);
18948 diff --git a/kernel/time/tick-common.c b/kernel/time/tick-common.c
18949 index 4fcd99e12aa0..5a47f2e98faf 100644
18950 --- a/kernel/time/tick-common.c
18951 +++ b/kernel/time/tick-common.c
18952 @@ -79,13 +79,15 @@ int tick_is_oneshot_available(void)
18953 static void tick_periodic(int cpu)
18955 if (tick_do_timer_cpu == cpu) {
18956 - write_seqlock(&jiffies_lock);
18957 + raw_spin_lock(&jiffies_lock);
18958 + write_seqcount_begin(&jiffies_seq);
18960 /* Keep track of the next tick event */
18961 tick_next_period = ktime_add(tick_next_period, tick_period);
18964 - write_sequnlock(&jiffies_lock);
18965 + write_seqcount_end(&jiffies_seq);
18966 + raw_spin_unlock(&jiffies_lock);
18967 update_wall_time();
18970 @@ -157,9 +159,9 @@ void tick_setup_periodic(struct clock_event_device *dev, int broadcast)
18974 - seq = read_seqbegin(&jiffies_lock);
18975 + seq = read_seqcount_begin(&jiffies_seq);
18976 next = tick_next_period;
18977 - } while (read_seqretry(&jiffies_lock, seq));
18978 + } while (read_seqcount_retry(&jiffies_seq, seq));
18980 clockevents_switch_state(dev, CLOCK_EVT_STATE_ONESHOT);
18982 diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
18983 index 3bcb61b52f6c..66d85482a96e 100644
18984 --- a/kernel/time/tick-sched.c
18985 +++ b/kernel/time/tick-sched.c
18986 @@ -62,7 +62,8 @@ static void tick_do_update_jiffies64(ktime_t now)
18989 /* Reevaluate with jiffies_lock held */
18990 - write_seqlock(&jiffies_lock);
18991 + raw_spin_lock(&jiffies_lock);
18992 + write_seqcount_begin(&jiffies_seq);
18994 delta = ktime_sub(now, last_jiffies_update);
18995 if (delta.tv64 >= tick_period.tv64) {
18996 @@ -85,10 +86,12 @@ static void tick_do_update_jiffies64(ktime_t now)
18997 /* Keep the tick_next_period variable up to date */
18998 tick_next_period = ktime_add(last_jiffies_update, tick_period);
19000 - write_sequnlock(&jiffies_lock);
19001 + write_seqcount_end(&jiffies_seq);
19002 + raw_spin_unlock(&jiffies_lock);
19005 - write_sequnlock(&jiffies_lock);
19006 + write_seqcount_end(&jiffies_seq);
19007 + raw_spin_unlock(&jiffies_lock);
19008 update_wall_time();
19011 @@ -99,12 +102,14 @@ static ktime_t tick_init_jiffy_update(void)
19015 - write_seqlock(&jiffies_lock);
19016 + raw_spin_lock(&jiffies_lock);
19017 + write_seqcount_begin(&jiffies_seq);
19018 /* Did we start the jiffies update yet ? */
19019 if (last_jiffies_update.tv64 == 0)
19020 last_jiffies_update = tick_next_period;
19021 period = last_jiffies_update;
19022 - write_sequnlock(&jiffies_lock);
19023 + write_seqcount_end(&jiffies_seq);
19024 + raw_spin_unlock(&jiffies_lock);
19028 @@ -215,6 +220,7 @@ static void nohz_full_kick_func(struct irq_work *work)
19030 static DEFINE_PER_CPU(struct irq_work, nohz_full_kick_work) = {
19031 .func = nohz_full_kick_func,
19032 + .flags = IRQ_WORK_HARD_IRQ,
19036 @@ -673,10 +679,10 @@ static ktime_t tick_nohz_stop_sched_tick(struct tick_sched *ts,
19038 /* Read jiffies and the time when jiffies were updated last */
19040 - seq = read_seqbegin(&jiffies_lock);
19041 + seq = read_seqcount_begin(&jiffies_seq);
19042 basemono = last_jiffies_update.tv64;
19043 basejiff = jiffies;
19044 - } while (read_seqretry(&jiffies_lock, seq));
19045 + } while (read_seqcount_retry(&jiffies_seq, seq));
19046 ts->last_jiffies = basejiff;
19048 if (rcu_needs_cpu(basemono, &next_rcu) ||
19049 @@ -877,14 +883,7 @@ static bool can_stop_idle_tick(int cpu, struct tick_sched *ts)
19052 if (unlikely(local_softirq_pending() && cpu_online(cpu))) {
19053 - static int ratelimit;
19055 - if (ratelimit < 10 &&
19056 - (local_softirq_pending() & SOFTIRQ_STOP_IDLE_MASK)) {
19057 - pr_warn("NOHZ: local_softirq_pending %02x\n",
19058 - (unsigned int) local_softirq_pending());
19061 + softirq_check_pending_idle();
19065 @@ -1193,6 +1192,7 @@ void tick_setup_sched_timer(void)
19066 * Emulate tick processing via per-CPU hrtimers:
19068 hrtimer_init(&ts->sched_timer, CLOCK_MONOTONIC, HRTIMER_MODE_ABS);
19069 + ts->sched_timer.irqsafe = 1;
19070 ts->sched_timer.function = tick_sched_timer;
19072 /* Get the next period (per-CPU) */
19073 diff --git a/kernel/time/timekeeping.c b/kernel/time/timekeeping.c
19074 index 46e312e9be38..fa75cf5d9253 100644
19075 --- a/kernel/time/timekeeping.c
19076 +++ b/kernel/time/timekeeping.c
19077 @@ -2328,8 +2328,10 @@ EXPORT_SYMBOL(hardpps);
19079 void xtime_update(unsigned long ticks)
19081 - write_seqlock(&jiffies_lock);
19082 + raw_spin_lock(&jiffies_lock);
19083 + write_seqcount_begin(&jiffies_seq);
19085 - write_sequnlock(&jiffies_lock);
19086 + write_seqcount_end(&jiffies_seq);
19087 + raw_spin_unlock(&jiffies_lock);
19088 update_wall_time();
19090 diff --git a/kernel/time/timekeeping.h b/kernel/time/timekeeping.h
19091 index 704f595ce83f..763a3e5121ff 100644
19092 --- a/kernel/time/timekeeping.h
19093 +++ b/kernel/time/timekeeping.h
19094 @@ -19,7 +19,8 @@ extern void timekeeping_resume(void);
19095 extern void do_timer(unsigned long ticks);
19096 extern void update_wall_time(void);
19098 -extern seqlock_t jiffies_lock;
19099 +extern raw_spinlock_t jiffies_lock;
19100 +extern seqcount_t jiffies_seq;
19102 #define CS_NAME_LEN 32
19104 diff --git a/kernel/time/timer.c b/kernel/time/timer.c
19105 index c611c47de884..cdff4411f8f6 100644
19106 --- a/kernel/time/timer.c
19107 +++ b/kernel/time/timer.c
19108 @@ -193,8 +193,11 @@ EXPORT_SYMBOL(jiffies_64);
19111 struct timer_base {
19113 + raw_spinlock_t lock;
19114 struct timer_list *running_timer;
19115 +#ifdef CONFIG_PREEMPT_RT_FULL
19116 + struct swait_queue_head wait_for_running_timer;
19119 unsigned long next_expiry;
19121 @@ -203,6 +206,8 @@ struct timer_base {
19123 DECLARE_BITMAP(pending_map, WHEEL_SIZE);
19124 struct hlist_head vectors[WHEEL_SIZE];
19125 + struct hlist_head expired_lists[LVL_DEPTH];
19126 + int expired_count;
19127 } ____cacheline_aligned;
19129 static DEFINE_PER_CPU(struct timer_base, timer_bases[NR_BASES]);
19130 @@ -948,10 +953,10 @@ static struct timer_base *lock_timer_base(struct timer_list *timer,
19132 if (!(tf & TIMER_MIGRATING)) {
19133 base = get_timer_base(tf);
19134 - spin_lock_irqsave(&base->lock, *flags);
19135 + raw_spin_lock_irqsave(&base->lock, *flags);
19136 if (timer->flags == tf)
19138 - spin_unlock_irqrestore(&base->lock, *flags);
19139 + raw_spin_unlock_irqrestore(&base->lock, *flags);
19143 @@ -1023,9 +1028,9 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
19144 /* See the comment in lock_timer_base() */
19145 timer->flags |= TIMER_MIGRATING;
19147 - spin_unlock(&base->lock);
19148 + raw_spin_unlock(&base->lock);
19150 - spin_lock(&base->lock);
19151 + raw_spin_lock(&base->lock);
19152 WRITE_ONCE(timer->flags,
19153 (timer->flags & ~TIMER_BASEMASK) | base->cpu);
19155 @@ -1050,7 +1055,7 @@ __mod_timer(struct timer_list *timer, unsigned long expires, bool pending_only)
19159 - spin_unlock_irqrestore(&base->lock, flags);
19160 + raw_spin_unlock_irqrestore(&base->lock, flags);
19164 @@ -1144,19 +1149,46 @@ void add_timer_on(struct timer_list *timer, int cpu)
19165 if (base != new_base) {
19166 timer->flags |= TIMER_MIGRATING;
19168 - spin_unlock(&base->lock);
19169 + raw_spin_unlock(&base->lock);
19171 - spin_lock(&base->lock);
19172 + raw_spin_lock(&base->lock);
19173 WRITE_ONCE(timer->flags,
19174 (timer->flags & ~TIMER_BASEMASK) | cpu);
19177 debug_activate(timer, timer->expires);
19178 internal_add_timer(base, timer);
19179 - spin_unlock_irqrestore(&base->lock, flags);
19180 + raw_spin_unlock_irqrestore(&base->lock, flags);
19182 EXPORT_SYMBOL_GPL(add_timer_on);
19184 +#ifdef CONFIG_PREEMPT_RT_FULL
19186 + * Wait for a running timer
19188 +static void wait_for_running_timer(struct timer_list *timer)
19190 + struct timer_base *base;
19191 + u32 tf = timer->flags;
19193 + if (tf & TIMER_MIGRATING)
19196 + base = get_timer_base(tf);
19197 + swait_event(base->wait_for_running_timer,
19198 + base->running_timer != timer);
19201 +# define wakeup_timer_waiters(b) swake_up_all(&(b)->wait_for_running_timer)
19203 +static inline void wait_for_running_timer(struct timer_list *timer)
19208 +# define wakeup_timer_waiters(b) do { } while (0)
19212 * del_timer - deactive a timer.
19213 * @timer: the timer to be deactivated
19214 @@ -1180,7 +1212,7 @@ int del_timer(struct timer_list *timer)
19215 if (timer_pending(timer)) {
19216 base = lock_timer_base(timer, &flags);
19217 ret = detach_if_pending(timer, base, true);
19218 - spin_unlock_irqrestore(&base->lock, flags);
19219 + raw_spin_unlock_irqrestore(&base->lock, flags);
19223 @@ -1208,13 +1240,13 @@ int try_to_del_timer_sync(struct timer_list *timer)
19224 timer_stats_timer_clear_start_info(timer);
19225 ret = detach_if_pending(timer, base, true);
19227 - spin_unlock_irqrestore(&base->lock, flags);
19228 + raw_spin_unlock_irqrestore(&base->lock, flags);
19232 EXPORT_SYMBOL(try_to_del_timer_sync);
19235 +#if defined(CONFIG_SMP) || defined(CONFIG_PREEMPT_RT_FULL)
19237 * del_timer_sync - deactivate a timer and wait for the handler to finish.
19238 * @timer: the timer to be deactivated
19239 @@ -1274,7 +1306,7 @@ int del_timer_sync(struct timer_list *timer)
19240 int ret = try_to_del_timer_sync(timer);
19244 + wait_for_running_timer(timer);
19247 EXPORT_SYMBOL(del_timer_sync);
19248 @@ -1323,7 +1355,8 @@ static void call_timer_fn(struct timer_list *timer, void (*fn)(unsigned long),
19252 -static void expire_timers(struct timer_base *base, struct hlist_head *head)
19253 +static inline void __expire_timers(struct timer_base *base,
19254 + struct hlist_head *head)
19256 while (!hlist_empty(head)) {
19257 struct timer_list *timer;
19258 @@ -1339,33 +1372,53 @@ static void expire_timers(struct timer_base *base, struct hlist_head *head)
19259 fn = timer->function;
19260 data = timer->data;
19262 - if (timer->flags & TIMER_IRQSAFE) {
19263 - spin_unlock(&base->lock);
19264 + if (!IS_ENABLED(CONFIG_PREEMPT_RT_FULL) &&
19265 + timer->flags & TIMER_IRQSAFE) {
19266 + raw_spin_unlock(&base->lock);
19267 call_timer_fn(timer, fn, data);
19268 - spin_lock(&base->lock);
19269 + base->running_timer = NULL;
19270 + raw_spin_lock(&base->lock);
19272 - spin_unlock_irq(&base->lock);
19273 + raw_spin_unlock_irq(&base->lock);
19274 call_timer_fn(timer, fn, data);
19275 - spin_lock_irq(&base->lock);
19276 + base->running_timer = NULL;
19277 + raw_spin_lock_irq(&base->lock);
19282 -static int __collect_expired_timers(struct timer_base *base,
19283 - struct hlist_head *heads)
19284 +static void expire_timers(struct timer_base *base)
19286 + struct hlist_head *head;
19288 + while (base->expired_count--) {
19289 + head = base->expired_lists + base->expired_count;
19290 + __expire_timers(base, head);
19292 + base->expired_count = 0;
19295 +static void __collect_expired_timers(struct timer_base *base)
19297 unsigned long clk = base->clk;
19298 struct hlist_head *vec;
19299 - int i, levels = 0;
19304 + * expire_timers() must be called at least once before we can
19305 + * collect more timers
19307 + if (WARN_ON(base->expired_count))
19310 for (i = 0; i < LVL_DEPTH; i++) {
19311 idx = (clk & LVL_MASK) + i * LVL_SIZE;
19313 if (__test_and_clear_bit(idx, base->pending_map)) {
19314 vec = base->vectors + idx;
19315 - hlist_move_list(vec, heads++);
19317 + hlist_move_list(vec,
19318 + &base->expired_lists[base->expired_count++]);
19320 /* Is it time to look at the next level? */
19321 if (clk & LVL_CLK_MASK)
19322 @@ -1373,7 +1426,6 @@ static int __collect_expired_timers(struct timer_base *base,
19323 /* Shift clock for the next level granularity */
19324 clk >>= LVL_CLK_SHIFT;
19329 #ifdef CONFIG_NO_HZ_COMMON
19330 @@ -1515,7 +1567,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
19331 if (cpu_is_offline(smp_processor_id()))
19334 - spin_lock(&base->lock);
19335 + raw_spin_lock(&base->lock);
19336 nextevt = __next_timer_interrupt(base);
19337 is_max_delta = (nextevt == base->clk + NEXT_TIMER_MAX_DELTA);
19338 base->next_expiry = nextevt;
19339 @@ -1543,7 +1595,7 @@ u64 get_next_timer_interrupt(unsigned long basej, u64 basem)
19340 if ((expires - basem) > TICK_NSEC)
19341 base->is_idle = true;
19343 - spin_unlock(&base->lock);
19344 + raw_spin_unlock(&base->lock);
19346 return cmp_next_hrtimer_event(basem, expires);
19348 @@ -1566,8 +1618,7 @@ void timer_clear_idle(void)
19349 base->is_idle = false;
19352 -static int collect_expired_timers(struct timer_base *base,
19353 - struct hlist_head *heads)
19354 +static void collect_expired_timers(struct timer_base *base)
19357 * NOHZ optimization. After a long idle sleep we need to forward the
19358 @@ -1584,20 +1635,49 @@ static int collect_expired_timers(struct timer_base *base,
19359 if (time_after(next, jiffies)) {
19360 /* The call site will increment clock! */
19361 base->clk = jiffies - 1;
19367 - return __collect_expired_timers(base, heads);
19368 + __collect_expired_timers(base);
19371 -static inline int collect_expired_timers(struct timer_base *base,
19372 - struct hlist_head *heads)
19373 +static inline void collect_expired_timers(struct timer_base *base)
19375 - return __collect_expired_timers(base, heads);
19376 + __collect_expired_timers(base);
19380 +static int find_expired_timers(struct timer_base *base)
19382 + const unsigned long int end_clk = jiffies;
19384 + while (!base->expired_count && time_after_eq(end_clk, base->clk)) {
19385 + collect_expired_timers(base);
19389 + return base->expired_count;
19392 +/* Called from CPU tick routine to quickly collect expired timers */
19393 +static int tick_find_expired(struct timer_base *base)
19397 + raw_spin_lock(&base->lock);
19399 + if (unlikely(time_after(jiffies, base->clk + HZ))) {
19400 + /* defer to ktimersoftd; don't spend too long in irq context */
19403 + count = find_expired_timers(base);
19405 + raw_spin_unlock(&base->lock);
19411 * Called from the timer interrupt handler to charge one tick to the current
19412 * process. user_tick is 1 if the tick is user time, 0 for system.
19413 @@ -1608,13 +1688,13 @@ void update_process_times(int user_tick)
19415 /* Note: this timer irq context must be accounted for as well. */
19416 account_process_tick(p, user_tick);
19417 + scheduler_tick();
19418 run_local_timers();
19419 rcu_check_callbacks(user_tick);
19420 -#ifdef CONFIG_IRQ_WORK
19421 +#if defined(CONFIG_IRQ_WORK)
19425 - scheduler_tick();
19426 run_posix_cpu_timers(p);
19429 @@ -1624,24 +1704,13 @@ void update_process_times(int user_tick)
19431 static inline void __run_timers(struct timer_base *base)
19433 - struct hlist_head heads[LVL_DEPTH];
19435 + raw_spin_lock_irq(&base->lock);
19437 - if (!time_after_eq(jiffies, base->clk))
19439 + while (find_expired_timers(base))
19440 + expire_timers(base);
19442 - spin_lock_irq(&base->lock);
19444 - while (time_after_eq(jiffies, base->clk)) {
19446 - levels = collect_expired_timers(base, heads);
19450 - expire_timers(base, heads + levels);
19452 - base->running_timer = NULL;
19453 - spin_unlock_irq(&base->lock);
19454 + raw_spin_unlock_irq(&base->lock);
19455 + wakeup_timer_waiters(base);
19459 @@ -1651,6 +1720,8 @@ static __latent_entropy void run_timer_softirq(struct softirq_action *h)
19461 struct timer_base *base = this_cpu_ptr(&timer_bases[BASE_STD]);
19463 + irq_work_tick_soft();
19465 __run_timers(base);
19466 if (IS_ENABLED(CONFIG_NO_HZ_COMMON) && base->nohz_active)
19467 __run_timers(this_cpu_ptr(&timer_bases[BASE_DEF]));
19468 @@ -1665,12 +1736,12 @@ void run_local_timers(void)
19470 hrtimer_run_queues();
19471 /* Raise the softirq only if required. */
19472 - if (time_before(jiffies, base->clk)) {
19473 + if (time_before(jiffies, base->clk) || !tick_find_expired(base)) {
19474 if (!IS_ENABLED(CONFIG_NO_HZ_COMMON) || !base->nohz_active)
19476 /* CPU is awake, so check the deferrable base. */
19478 - if (time_before(jiffies, base->clk))
19479 + if (time_before(jiffies, base->clk) || !tick_find_expired(base))
19482 raise_softirq(TIMER_SOFTIRQ);
19483 @@ -1836,16 +1907,17 @@ int timers_dead_cpu(unsigned int cpu)
19484 * The caller is globally serialized and nobody else
19485 * takes two locks at once, deadlock is not possible.
19487 - spin_lock_irq(&new_base->lock);
19488 - spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
19489 + raw_spin_lock_irq(&new_base->lock);
19490 + raw_spin_lock_nested(&old_base->lock, SINGLE_DEPTH_NESTING);
19492 BUG_ON(old_base->running_timer);
19493 + BUG_ON(old_base->expired_count);
19495 for (i = 0; i < WHEEL_SIZE; i++)
19496 migrate_timer_list(new_base, old_base->vectors + i);
19498 - spin_unlock(&old_base->lock);
19499 - spin_unlock_irq(&new_base->lock);
19500 + raw_spin_unlock(&old_base->lock);
19501 + raw_spin_unlock_irq(&new_base->lock);
19502 put_cpu_ptr(&timer_bases);
19505 @@ -1861,8 +1933,12 @@ static void __init init_timer_cpu(int cpu)
19506 for (i = 0; i < NR_BASES; i++) {
19507 base = per_cpu_ptr(&timer_bases[i], cpu);
19509 - spin_lock_init(&base->lock);
19510 + raw_spin_lock_init(&base->lock);
19511 base->clk = jiffies;
19512 +#ifdef CONFIG_PREEMPT_RT_FULL
19513 + init_swait_queue_head(&base->wait_for_running_timer);
19515 + base->expired_count = 0;
19519 diff --git a/kernel/trace/Kconfig b/kernel/trace/Kconfig
19520 index 2a96b063d659..812e37237eb8 100644
19521 --- a/kernel/trace/Kconfig
19522 +++ b/kernel/trace/Kconfig
19523 @@ -182,6 +182,24 @@ config IRQSOFF_TRACER
19524 enabled. This option and the preempt-off timing option can be
19525 used together or separately.)
19527 +config INTERRUPT_OFF_HIST
19528 + bool "Interrupts-off Latency Histogram"
19529 + depends on IRQSOFF_TRACER
19531 + This option generates continuously updated histograms (one per cpu)
19532 + of the duration of time periods with interrupts disabled. The
19533 + histograms are disabled by default. To enable them, write a non-zero
19536 + /sys/kernel/debug/tracing/latency_hist/enable/preemptirqsoff
19538 + If PREEMPT_OFF_HIST is also selected, additional histograms (one
19539 + per cpu) are generated that accumulate the duration of time periods
19540 + when both interrupts and preemption are disabled. The histogram data
19541 + will be located in the debug file system at
19543 + /sys/kernel/debug/tracing/latency_hist/irqsoff
19545 config PREEMPT_TRACER
19546 bool "Preemption-off Latency Tracer"
19548 @@ -206,6 +224,24 @@ config PREEMPT_TRACER
19549 enabled. This option and the irqs-off timing option can be
19550 used together or separately.)
19552 +config PREEMPT_OFF_HIST
19553 + bool "Preemption-off Latency Histogram"
19554 + depends on PREEMPT_TRACER
19556 + This option generates continuously updated histograms (one per cpu)
19557 + of the duration of time periods with preemption disabled. The
19558 + histograms are disabled by default. To enable them, write a non-zero
19561 + /sys/kernel/debug/tracing/latency_hist/enable/preemptirqsoff
19563 + If INTERRUPT_OFF_HIST is also selected, additional histograms (one
19564 + per cpu) are generated that accumulate the duration of time periods
19565 + when both interrupts and preemption are disabled. The histogram data
19566 + will be located in the debug file system at
19568 + /sys/kernel/debug/tracing/latency_hist/preemptoff
19570 config SCHED_TRACER
19571 bool "Scheduling Latency Tracer"
19572 select GENERIC_TRACER
19573 @@ -251,6 +287,74 @@ config HWLAT_TRACER
19574 file. Every time a latency is greater than tracing_thresh, it will
19575 be recorded into the ring buffer.
19577 +config WAKEUP_LATENCY_HIST
19578 + bool "Scheduling Latency Histogram"
19579 + depends on SCHED_TRACER
19581 + This option generates continuously updated histograms (one per cpu)
19582 + of the scheduling latency of the highest priority task.
19583 + The histograms are disabled by default. To enable them, write a
19584 + non-zero number to
19586 + /sys/kernel/debug/tracing/latency_hist/enable/wakeup
19588 + Two different algorithms are used, one to determine the latency of
19589 + processes that exclusively use the highest priority of the system and
19590 + another one to determine the latency of processes that share the
19591 + highest system priority with other processes. The former is used to
19592 + improve hardware and system software, the latter to optimize the
19593 + priority design of a given system. The histogram data will be
19594 + located in the debug file system at
19596 + /sys/kernel/debug/tracing/latency_hist/wakeup
19600 + /sys/kernel/debug/tracing/latency_hist/wakeup/sharedprio
19602 + If both Scheduling Latency Histogram and Missed Timer Offsets
19603 + Histogram are selected, additional histogram data will be collected
19604 + that contain, in addition to the wakeup latency, the timer latency, in
19605 + case the wakeup was triggered by an expired timer. These histograms
19606 + are available in the
19608 + /sys/kernel/debug/tracing/latency_hist/timerandwakeup
19610 + directory. They reflect the apparent interrupt and scheduling latency
19611 + and are best suitable to determine the worst-case latency of a given
19612 + system. To enable these histograms, write a non-zero number to
19614 + /sys/kernel/debug/tracing/latency_hist/enable/timerandwakeup
19616 +config MISSED_TIMER_OFFSETS_HIST
19617 + depends on HIGH_RES_TIMERS
19618 + select GENERIC_TRACER
19619 + bool "Missed Timer Offsets Histogram"
19621 + Generate a histogram of missed timer offsets in microseconds. The
19622 + histograms are disabled by default. To enable them, write a non-zero
19625 + /sys/kernel/debug/tracing/latency_hist/enable/missed_timer_offsets
19627 + The histogram data will be located in the debug file system at
19629 + /sys/kernel/debug/tracing/latency_hist/missed_timer_offsets
19631 + If both Scheduling Latency Histogram and Missed Timer Offsets
19632 + Histogram are selected, additional histogram data will be collected
19633 + that contain, in addition to the wakeup latency, the timer latency, in
19634 + case the wakeup was triggered by an expired timer. These histograms
19635 + are available in the
19637 + /sys/kernel/debug/tracing/latency_hist/timerandwakeup
19639 + directory. They reflect the apparent interrupt and scheduling latency
19640 + and are best suitable to determine the worst-case latency of a given
19641 + system. To enable these histograms, write a non-zero number to
19643 + /sys/kernel/debug/tracing/latency_hist/enable/timerandwakeup
19645 config ENABLE_DEFAULT_TRACERS
19646 bool "Trace process context switches and events"
19647 depends on !GENERIC_TRACER
19648 diff --git a/kernel/trace/Makefile b/kernel/trace/Makefile
19649 index e57980845549..83af000b783c 100644
19650 --- a/kernel/trace/Makefile
19651 +++ b/kernel/trace/Makefile
19652 @@ -38,6 +38,10 @@ obj-$(CONFIG_IRQSOFF_TRACER) += trace_irqsoff.o
19653 obj-$(CONFIG_PREEMPT_TRACER) += trace_irqsoff.o
19654 obj-$(CONFIG_SCHED_TRACER) += trace_sched_wakeup.o
19655 obj-$(CONFIG_HWLAT_TRACER) += trace_hwlat.o
19656 +obj-$(CONFIG_INTERRUPT_OFF_HIST) += latency_hist.o
19657 +obj-$(CONFIG_PREEMPT_OFF_HIST) += latency_hist.o
19658 +obj-$(CONFIG_WAKEUP_LATENCY_HIST) += latency_hist.o
19659 +obj-$(CONFIG_MISSED_TIMER_OFFSETS_HIST) += latency_hist.o
19660 obj-$(CONFIG_NOP_TRACER) += trace_nop.o
19661 obj-$(CONFIG_STACK_TRACER) += trace_stack.o
19662 obj-$(CONFIG_MMIOTRACE) += trace_mmiotrace.o
19663 diff --git a/kernel/trace/latency_hist.c b/kernel/trace/latency_hist.c
19664 new file mode 100644
19665 index 000000000000..7f6ee70dea41
19667 +++ b/kernel/trace/latency_hist.c
19670 + * kernel/trace/latency_hist.c
19672 + * Add support for histograms of preemption-off latency and
19673 + * interrupt-off latency and wakeup latency, it depends on
19674 + * Real-Time Preemption Support.
19676 + * Copyright (C) 2005 MontaVista Software, Inc.
19677 + * Yi Yang <yyang@ch.mvista.com>
19679 + * Converted to work with the new latency tracer.
19680 + * Copyright (C) 2008 Red Hat, Inc.
19681 + * Steven Rostedt <srostedt@redhat.com>
19684 +#include <linux/module.h>
19685 +#include <linux/debugfs.h>
19686 +#include <linux/seq_file.h>
19687 +#include <linux/percpu.h>
19688 +#include <linux/kallsyms.h>
19689 +#include <linux/uaccess.h>
19690 +#include <linux/sched.h>
19691 +#include <linux/sched/rt.h>
19692 +#include <linux/slab.h>
19693 +#include <linux/atomic.h>
19694 +#include <asm/div64.h>
19696 +#include "trace.h"
19697 +#include <trace/events/sched.h>
19699 +#define NSECS_PER_USECS 1000L
19701 +#define CREATE_TRACE_POINTS
19702 +#include <trace/events/hist.h>
19705 + IRQSOFF_LATENCY = 0,
19706 + PREEMPTOFF_LATENCY,
19707 + PREEMPTIRQSOFF_LATENCY,
19709 + WAKEUP_LATENCY_SHAREDPRIO,
19710 + MISSED_TIMER_OFFSETS,
19711 + TIMERANDWAKEUP_LATENCY,
19712 + MAX_LATENCY_TYPE,
19715 +#define MAX_ENTRY_NUM 10240
19717 +struct hist_data {
19718 + atomic_t hist_mode; /* 0 log, 1 don't log */
19719 + long offset; /* set it to MAX_ENTRY_NUM/2 for a bipolar scale */
19722 + unsigned long long below_hist_bound_samples;
19723 + unsigned long long above_hist_bound_samples;
19724 + long long accumulate_lat;
19725 + unsigned long long total_samples;
19726 + unsigned long long hist_array[MAX_ENTRY_NUM];
19729 +struct enable_data {
19730 + int latency_type;
19734 +static char *latency_hist_dir_root = "latency_hist";
19736 +#ifdef CONFIG_INTERRUPT_OFF_HIST
19737 +static DEFINE_PER_CPU(struct hist_data, irqsoff_hist);
19738 +static char *irqsoff_hist_dir = "irqsoff";
19739 +static DEFINE_PER_CPU(cycles_t, hist_irqsoff_start);
19740 +static DEFINE_PER_CPU(int, hist_irqsoff_counting);
19743 +#ifdef CONFIG_PREEMPT_OFF_HIST
19744 +static DEFINE_PER_CPU(struct hist_data, preemptoff_hist);
19745 +static char *preemptoff_hist_dir = "preemptoff";
19746 +static DEFINE_PER_CPU(cycles_t, hist_preemptoff_start);
19747 +static DEFINE_PER_CPU(int, hist_preemptoff_counting);
19750 +#if defined(CONFIG_PREEMPT_OFF_HIST) && defined(CONFIG_INTERRUPT_OFF_HIST)
19751 +static DEFINE_PER_CPU(struct hist_data, preemptirqsoff_hist);
19752 +static char *preemptirqsoff_hist_dir = "preemptirqsoff";
19753 +static DEFINE_PER_CPU(cycles_t, hist_preemptirqsoff_start);
19754 +static DEFINE_PER_CPU(int, hist_preemptirqsoff_counting);
19757 +#if defined(CONFIG_PREEMPT_OFF_HIST) || defined(CONFIG_INTERRUPT_OFF_HIST)
19758 +static notrace void probe_preemptirqsoff_hist(void *v, int reason, int start);
19759 +static struct enable_data preemptirqsoff_enabled_data = {
19760 + .latency_type = PREEMPTIRQSOFF_LATENCY,
19765 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
19766 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
19767 +struct maxlatproc_data {
19768 + char comm[FIELD_SIZEOF(struct task_struct, comm)];
19769 + char current_comm[FIELD_SIZEOF(struct task_struct, comm)];
19773 + int current_prio;
19775 + long timeroffset;
19776 + cycle_t timestamp;
19780 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
19781 +static DEFINE_PER_CPU(struct hist_data, wakeup_latency_hist);
19782 +static DEFINE_PER_CPU(struct hist_data, wakeup_latency_hist_sharedprio);
19783 +static char *wakeup_latency_hist_dir = "wakeup";
19784 +static char *wakeup_latency_hist_dir_sharedprio = "sharedprio";
19785 +static notrace void probe_wakeup_latency_hist_start(void *v,
19786 + struct task_struct *p);
19787 +static notrace void probe_wakeup_latency_hist_stop(void *v,
19788 + bool preempt, struct task_struct *prev, struct task_struct *next);
19789 +static notrace void probe_sched_migrate_task(void *,
19790 + struct task_struct *task, int cpu);
19791 +static struct enable_data wakeup_latency_enabled_data = {
19792 + .latency_type = WAKEUP_LATENCY,
19795 +static DEFINE_PER_CPU(struct maxlatproc_data, wakeup_maxlatproc);
19796 +static DEFINE_PER_CPU(struct maxlatproc_data, wakeup_maxlatproc_sharedprio);
19797 +static DEFINE_PER_CPU(struct task_struct *, wakeup_task);
19798 +static DEFINE_PER_CPU(int, wakeup_sharedprio);
19799 +static unsigned long wakeup_pid;
19802 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
19803 +static DEFINE_PER_CPU(struct hist_data, missed_timer_offsets);
19804 +static char *missed_timer_offsets_dir = "missed_timer_offsets";
19805 +static notrace void probe_hrtimer_interrupt(void *v, int cpu,
19806 + long long offset, struct task_struct *curr, struct task_struct *task);
19807 +static struct enable_data missed_timer_offsets_enabled_data = {
19808 + .latency_type = MISSED_TIMER_OFFSETS,
19811 +static DEFINE_PER_CPU(struct maxlatproc_data, missed_timer_offsets_maxlatproc);
19812 +static unsigned long missed_timer_offsets_pid;
19815 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) && \
19816 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
19817 +static DEFINE_PER_CPU(struct hist_data, timerandwakeup_latency_hist);
19818 +static char *timerandwakeup_latency_hist_dir = "timerandwakeup";
19819 +static struct enable_data timerandwakeup_enabled_data = {
19820 + .latency_type = TIMERANDWAKEUP_LATENCY,
19823 +static DEFINE_PER_CPU(struct maxlatproc_data, timerandwakeup_maxlatproc);
19826 +void notrace latency_hist(int latency_type, int cpu, long latency,
19827 + long timeroffset, cycle_t stop,
19828 + struct task_struct *p)
19830 + struct hist_data *my_hist;
19831 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
19832 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
19833 + struct maxlatproc_data *mp = NULL;
19836 + if (!cpu_possible(cpu) || latency_type < 0 ||
19837 + latency_type >= MAX_LATENCY_TYPE)
19840 + switch (latency_type) {
19841 +#ifdef CONFIG_INTERRUPT_OFF_HIST
19842 + case IRQSOFF_LATENCY:
19843 + my_hist = &per_cpu(irqsoff_hist, cpu);
19846 +#ifdef CONFIG_PREEMPT_OFF_HIST
19847 + case PREEMPTOFF_LATENCY:
19848 + my_hist = &per_cpu(preemptoff_hist, cpu);
19851 +#if defined(CONFIG_PREEMPT_OFF_HIST) && defined(CONFIG_INTERRUPT_OFF_HIST)
19852 + case PREEMPTIRQSOFF_LATENCY:
19853 + my_hist = &per_cpu(preemptirqsoff_hist, cpu);
19856 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
19857 + case WAKEUP_LATENCY:
19858 + my_hist = &per_cpu(wakeup_latency_hist, cpu);
19859 + mp = &per_cpu(wakeup_maxlatproc, cpu);
19861 + case WAKEUP_LATENCY_SHAREDPRIO:
19862 + my_hist = &per_cpu(wakeup_latency_hist_sharedprio, cpu);
19863 + mp = &per_cpu(wakeup_maxlatproc_sharedprio, cpu);
19866 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
19867 + case MISSED_TIMER_OFFSETS:
19868 + my_hist = &per_cpu(missed_timer_offsets, cpu);
19869 + mp = &per_cpu(missed_timer_offsets_maxlatproc, cpu);
19872 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) && \
19873 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
19874 + case TIMERANDWAKEUP_LATENCY:
19875 + my_hist = &per_cpu(timerandwakeup_latency_hist, cpu);
19876 + mp = &per_cpu(timerandwakeup_maxlatproc, cpu);
19884 + latency += my_hist->offset;
19886 + if (atomic_read(&my_hist->hist_mode) == 0)
19889 + if (latency < 0 || latency >= MAX_ENTRY_NUM) {
19891 + my_hist->below_hist_bound_samples++;
19893 + my_hist->above_hist_bound_samples++;
19895 + my_hist->hist_array[latency]++;
19897 + if (unlikely(latency > my_hist->max_lat ||
19898 + my_hist->min_lat == LONG_MAX)) {
19899 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
19900 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
19901 + if (latency_type == WAKEUP_LATENCY ||
19902 + latency_type == WAKEUP_LATENCY_SHAREDPRIO ||
19903 + latency_type == MISSED_TIMER_OFFSETS ||
19904 + latency_type == TIMERANDWAKEUP_LATENCY) {
19905 + strncpy(mp->comm, p->comm, sizeof(mp->comm));
19906 + strncpy(mp->current_comm, current->comm,
19907 + sizeof(mp->current_comm));
19908 + mp->pid = task_pid_nr(p);
19909 + mp->current_pid = task_pid_nr(current);
19910 + mp->prio = p->prio;
19911 + mp->current_prio = current->prio;
19912 + mp->latency = latency;
19913 + mp->timeroffset = timeroffset;
19914 + mp->timestamp = stop;
19917 + my_hist->max_lat = latency;
19919 + if (unlikely(latency < my_hist->min_lat))
19920 + my_hist->min_lat = latency;
19921 + my_hist->total_samples++;
19922 + my_hist->accumulate_lat += latency;
19925 +static void *l_start(struct seq_file *m, loff_t *pos)
19927 + loff_t *index_ptr = NULL;
19928 + loff_t index = *pos;
19929 + struct hist_data *my_hist = m->private;
19931 + if (index == 0) {
19932 + char minstr[32], avgstr[32], maxstr[32];
19934 + atomic_dec(&my_hist->hist_mode);
19936 + if (likely(my_hist->total_samples)) {
19937 + long avg = (long) div64_s64(my_hist->accumulate_lat,
19938 + my_hist->total_samples);
19939 + snprintf(minstr, sizeof(minstr), "%ld",
19940 + my_hist->min_lat - my_hist->offset);
19941 + snprintf(avgstr, sizeof(avgstr), "%ld",
19942 + avg - my_hist->offset);
19943 + snprintf(maxstr, sizeof(maxstr), "%ld",
19944 + my_hist->max_lat - my_hist->offset);
19946 + strcpy(minstr, "<undef>");
19947 + strcpy(avgstr, minstr);
19948 + strcpy(maxstr, minstr);
19951 + seq_printf(m, "#Minimum latency: %s microseconds\n"
19952 + "#Average latency: %s microseconds\n"
19953 + "#Maximum latency: %s microseconds\n"
19954 + "#Total samples: %llu\n"
19955 + "#There are %llu samples lower than %ld"
19956 + " microseconds.\n"
19957 + "#There are %llu samples greater or equal"
19958 + " than %ld microseconds.\n"
19959 + "#usecs\t%16s\n",
19960 + minstr, avgstr, maxstr,
19961 + my_hist->total_samples,
19962 + my_hist->below_hist_bound_samples,
19963 + -my_hist->offset,
19964 + my_hist->above_hist_bound_samples,
19965 + MAX_ENTRY_NUM - my_hist->offset,
19968 + if (index < MAX_ENTRY_NUM) {
19969 + index_ptr = kmalloc(sizeof(loff_t), GFP_KERNEL);
19971 + *index_ptr = index;
19974 + return index_ptr;
19977 +static void *l_next(struct seq_file *m, void *p, loff_t *pos)
19979 + loff_t *index_ptr = p;
19980 + struct hist_data *my_hist = m->private;
19982 + if (++*pos >= MAX_ENTRY_NUM) {
19983 + atomic_inc(&my_hist->hist_mode);
19986 + *index_ptr = *pos;
19987 + return index_ptr;
19990 +static void l_stop(struct seq_file *m, void *p)
19995 +static int l_show(struct seq_file *m, void *p)
19997 + int index = *(loff_t *) p;
19998 + struct hist_data *my_hist = m->private;
20000 + seq_printf(m, "%6ld\t%16llu\n", index - my_hist->offset,
20001 + my_hist->hist_array[index]);
20005 +static const struct seq_operations latency_hist_seq_op = {
20006 + .start = l_start,
20012 +static int latency_hist_open(struct inode *inode, struct file *file)
20016 + ret = seq_open(file, &latency_hist_seq_op);
20018 + struct seq_file *seq = file->private_data;
20019 + seq->private = inode->i_private;
20024 +static const struct file_operations latency_hist_fops = {
20025 + .open = latency_hist_open,
20026 + .read = seq_read,
20027 + .llseek = seq_lseek,
20028 + .release = seq_release,
20031 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20032 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20033 +static void clear_maxlatprocdata(struct maxlatproc_data *mp)
20035 + mp->comm[0] = mp->current_comm[0] = '\0';
20036 + mp->prio = mp->current_prio = mp->pid = mp->current_pid =
20037 + mp->latency = mp->timeroffset = -1;
20038 + mp->timestamp = 0;
20042 +static void hist_reset(struct hist_data *hist)
20044 + atomic_dec(&hist->hist_mode);
20046 + memset(hist->hist_array, 0, sizeof(hist->hist_array));
20047 + hist->below_hist_bound_samples = 0ULL;
20048 + hist->above_hist_bound_samples = 0ULL;
20049 + hist->min_lat = LONG_MAX;
20050 + hist->max_lat = LONG_MIN;
20051 + hist->total_samples = 0ULL;
20052 + hist->accumulate_lat = 0LL;
20054 + atomic_inc(&hist->hist_mode);
20058 +latency_hist_reset(struct file *file, const char __user *a,
20059 + size_t size, loff_t *off)
20062 + struct hist_data *hist = NULL;
20063 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20064 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20065 + struct maxlatproc_data *mp = NULL;
20067 + off_t latency_type = (off_t) file->private_data;
20069 + for_each_online_cpu(cpu) {
20071 + switch (latency_type) {
20072 +#ifdef CONFIG_PREEMPT_OFF_HIST
20073 + case PREEMPTOFF_LATENCY:
20074 + hist = &per_cpu(preemptoff_hist, cpu);
20077 +#ifdef CONFIG_INTERRUPT_OFF_HIST
20078 + case IRQSOFF_LATENCY:
20079 + hist = &per_cpu(irqsoff_hist, cpu);
20082 +#if defined(CONFIG_INTERRUPT_OFF_HIST) && defined(CONFIG_PREEMPT_OFF_HIST)
20083 + case PREEMPTIRQSOFF_LATENCY:
20084 + hist = &per_cpu(preemptirqsoff_hist, cpu);
20087 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20088 + case WAKEUP_LATENCY:
20089 + hist = &per_cpu(wakeup_latency_hist, cpu);
20090 + mp = &per_cpu(wakeup_maxlatproc, cpu);
20092 + case WAKEUP_LATENCY_SHAREDPRIO:
20093 + hist = &per_cpu(wakeup_latency_hist_sharedprio, cpu);
20094 + mp = &per_cpu(wakeup_maxlatproc_sharedprio, cpu);
20097 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20098 + case MISSED_TIMER_OFFSETS:
20099 + hist = &per_cpu(missed_timer_offsets, cpu);
20100 + mp = &per_cpu(missed_timer_offsets_maxlatproc, cpu);
20103 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) && \
20104 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20105 + case TIMERANDWAKEUP_LATENCY:
20106 + hist = &per_cpu(timerandwakeup_latency_hist, cpu);
20107 + mp = &per_cpu(timerandwakeup_maxlatproc, cpu);
20112 + hist_reset(hist);
20113 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20114 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20115 + if (latency_type == WAKEUP_LATENCY ||
20116 + latency_type == WAKEUP_LATENCY_SHAREDPRIO ||
20117 + latency_type == MISSED_TIMER_OFFSETS ||
20118 + latency_type == TIMERANDWAKEUP_LATENCY)
20119 + clear_maxlatprocdata(mp);
20126 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20127 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20129 +show_pid(struct file *file, char __user *ubuf, size_t cnt, loff_t *ppos)
20133 + unsigned long *this_pid = file->private_data;
20135 + r = snprintf(buf, sizeof(buf), "%lu\n", *this_pid);
20136 + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
20139 +static ssize_t do_pid(struct file *file, const char __user *ubuf,
20140 + size_t cnt, loff_t *ppos)
20143 + unsigned long pid;
20144 + unsigned long *this_pid = file->private_data;
20146 + if (cnt >= sizeof(buf))
20149 + if (copy_from_user(&buf, ubuf, cnt))
20154 + if (kstrtoul(buf, 10, &pid))
20163 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20164 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20166 +show_maxlatproc(struct file *file, char __user *ubuf, size_t cnt, loff_t *ppos)
20169 + struct maxlatproc_data *mp = file->private_data;
20170 + int strmaxlen = (TASK_COMM_LEN * 2) + (8 * 8);
20171 + unsigned long long t;
20172 + unsigned long usecs, secs;
20175 + if (mp->pid == -1 || mp->current_pid == -1) {
20176 + buf = "(none)\n";
20177 + return simple_read_from_buffer(ubuf, cnt, ppos, buf,
20181 + buf = kmalloc(strmaxlen, GFP_KERNEL);
20185 + t = ns2usecs(mp->timestamp);
20186 + usecs = do_div(t, USEC_PER_SEC);
20187 + secs = (unsigned long) t;
20188 + r = snprintf(buf, strmaxlen,
20189 + "%d %d %ld (%ld) %s <- %d %d %s %lu.%06lu\n", mp->pid,
20190 + MAX_RT_PRIO-1 - mp->prio, mp->latency, mp->timeroffset, mp->comm,
20191 + mp->current_pid, MAX_RT_PRIO-1 - mp->current_prio, mp->current_comm,
20193 + r = simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
20200 +show_enable(struct file *file, char __user *ubuf, size_t cnt, loff_t *ppos)
20203 + struct enable_data *ed = file->private_data;
20206 + r = snprintf(buf, sizeof(buf), "%d\n", ed->enabled);
20207 + return simple_read_from_buffer(ubuf, cnt, ppos, buf, r);
20211 +do_enable(struct file *file, const char __user *ubuf, size_t cnt, loff_t *ppos)
20215 + struct enable_data *ed = file->private_data;
20217 + if (cnt >= sizeof(buf))
20220 + if (copy_from_user(&buf, ubuf, cnt))
20225 + if (kstrtoul(buf, 10, &enable))
20228 + if ((enable && ed->enabled) || (!enable && !ed->enabled))
20234 + switch (ed->latency_type) {
20235 +#if defined(CONFIG_INTERRUPT_OFF_HIST) || defined(CONFIG_PREEMPT_OFF_HIST)
20236 + case PREEMPTIRQSOFF_LATENCY:
20237 + ret = register_trace_preemptirqsoff_hist(
20238 + probe_preemptirqsoff_hist, NULL);
20240 + pr_info("wakeup trace: Couldn't assign "
20241 + "probe_preemptirqsoff_hist "
20242 + "to trace_preemptirqsoff_hist\n");
20247 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20248 + case WAKEUP_LATENCY:
20249 + ret = register_trace_sched_wakeup(
20250 + probe_wakeup_latency_hist_start, NULL);
20252 + pr_info("wakeup trace: Couldn't assign "
20253 + "probe_wakeup_latency_hist_start "
20254 + "to trace_sched_wakeup\n");
20257 + ret = register_trace_sched_wakeup_new(
20258 + probe_wakeup_latency_hist_start, NULL);
20260 + pr_info("wakeup trace: Couldn't assign "
20261 + "probe_wakeup_latency_hist_start "
20262 + "to trace_sched_wakeup_new\n");
20263 + unregister_trace_sched_wakeup(
20264 + probe_wakeup_latency_hist_start, NULL);
20267 + ret = register_trace_sched_switch(
20268 + probe_wakeup_latency_hist_stop, NULL);
20270 + pr_info("wakeup trace: Couldn't assign "
20271 + "probe_wakeup_latency_hist_stop "
20272 + "to trace_sched_switch\n");
20273 + unregister_trace_sched_wakeup(
20274 + probe_wakeup_latency_hist_start, NULL);
20275 + unregister_trace_sched_wakeup_new(
20276 + probe_wakeup_latency_hist_start, NULL);
20279 + ret = register_trace_sched_migrate_task(
20280 + probe_sched_migrate_task, NULL);
20282 + pr_info("wakeup trace: Couldn't assign "
20283 + "probe_sched_migrate_task "
20284 + "to trace_sched_migrate_task\n");
20285 + unregister_trace_sched_wakeup(
20286 + probe_wakeup_latency_hist_start, NULL);
20287 + unregister_trace_sched_wakeup_new(
20288 + probe_wakeup_latency_hist_start, NULL);
20289 + unregister_trace_sched_switch(
20290 + probe_wakeup_latency_hist_stop, NULL);
20295 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20296 + case MISSED_TIMER_OFFSETS:
20297 + ret = register_trace_hrtimer_interrupt(
20298 + probe_hrtimer_interrupt, NULL);
20300 + pr_info("wakeup trace: Couldn't assign "
20301 + "probe_hrtimer_interrupt "
20302 + "to trace_hrtimer_interrupt\n");
20307 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) && \
20308 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20309 + case TIMERANDWAKEUP_LATENCY:
20310 + if (!wakeup_latency_enabled_data.enabled ||
20311 + !missed_timer_offsets_enabled_data.enabled)
20319 + switch (ed->latency_type) {
20320 +#if defined(CONFIG_INTERRUPT_OFF_HIST) || defined(CONFIG_PREEMPT_OFF_HIST)
20321 + case PREEMPTIRQSOFF_LATENCY:
20325 + unregister_trace_preemptirqsoff_hist(
20326 + probe_preemptirqsoff_hist, NULL);
20327 + for_each_online_cpu(cpu) {
20328 +#ifdef CONFIG_INTERRUPT_OFF_HIST
20329 + per_cpu(hist_irqsoff_counting,
20332 +#ifdef CONFIG_PREEMPT_OFF_HIST
20333 + per_cpu(hist_preemptoff_counting,
20336 +#if defined(CONFIG_INTERRUPT_OFF_HIST) && defined(CONFIG_PREEMPT_OFF_HIST)
20337 + per_cpu(hist_preemptirqsoff_counting,
20344 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20345 + case WAKEUP_LATENCY:
20349 + unregister_trace_sched_wakeup(
20350 + probe_wakeup_latency_hist_start, NULL);
20351 + unregister_trace_sched_wakeup_new(
20352 + probe_wakeup_latency_hist_start, NULL);
20353 + unregister_trace_sched_switch(
20354 + probe_wakeup_latency_hist_stop, NULL);
20355 + unregister_trace_sched_migrate_task(
20356 + probe_sched_migrate_task, NULL);
20358 + for_each_online_cpu(cpu) {
20359 + per_cpu(wakeup_task, cpu) = NULL;
20360 + per_cpu(wakeup_sharedprio, cpu) = 0;
20363 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20364 + timerandwakeup_enabled_data.enabled = 0;
20368 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20369 + case MISSED_TIMER_OFFSETS:
20370 + unregister_trace_hrtimer_interrupt(
20371 + probe_hrtimer_interrupt, NULL);
20372 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20373 + timerandwakeup_enabled_data.enabled = 0;
20381 + ed->enabled = enable;
20385 +static const struct file_operations latency_hist_reset_fops = {
20386 + .open = tracing_open_generic,
20387 + .write = latency_hist_reset,
20390 +static const struct file_operations enable_fops = {
20391 + .open = tracing_open_generic,
20392 + .read = show_enable,
20393 + .write = do_enable,
20396 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20397 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20398 +static const struct file_operations pid_fops = {
20399 + .open = tracing_open_generic,
20400 + .read = show_pid,
20404 +static const struct file_operations maxlatproc_fops = {
20405 + .open = tracing_open_generic,
20406 + .read = show_maxlatproc,
20410 +#if defined(CONFIG_INTERRUPT_OFF_HIST) || defined(CONFIG_PREEMPT_OFF_HIST)
20411 +static notrace void probe_preemptirqsoff_hist(void *v, int reason,
20414 + int cpu = raw_smp_processor_id();
20415 + int time_set = 0;
20418 + cycle_t uninitialized_var(start);
20420 + if (!preempt_count() && !irqs_disabled())
20423 +#ifdef CONFIG_INTERRUPT_OFF_HIST
20424 + if ((reason == IRQS_OFF || reason == TRACE_START) &&
20425 + !per_cpu(hist_irqsoff_counting, cpu)) {
20426 + per_cpu(hist_irqsoff_counting, cpu) = 1;
20427 + start = ftrace_now(cpu);
20429 + per_cpu(hist_irqsoff_start, cpu) = start;
20433 +#ifdef CONFIG_PREEMPT_OFF_HIST
20434 + if ((reason == PREEMPT_OFF || reason == TRACE_START) &&
20435 + !per_cpu(hist_preemptoff_counting, cpu)) {
20436 + per_cpu(hist_preemptoff_counting, cpu) = 1;
20437 + if (!(time_set++))
20438 + start = ftrace_now(cpu);
20439 + per_cpu(hist_preemptoff_start, cpu) = start;
20443 +#if defined(CONFIG_INTERRUPT_OFF_HIST) && defined(CONFIG_PREEMPT_OFF_HIST)
20444 + if (per_cpu(hist_irqsoff_counting, cpu) &&
20445 + per_cpu(hist_preemptoff_counting, cpu) &&
20446 + !per_cpu(hist_preemptirqsoff_counting, cpu)) {
20447 + per_cpu(hist_preemptirqsoff_counting, cpu) = 1;
20449 + start = ftrace_now(cpu);
20450 + per_cpu(hist_preemptirqsoff_start, cpu) = start;
20454 + cycle_t uninitialized_var(stop);
20456 +#ifdef CONFIG_INTERRUPT_OFF_HIST
20457 + if ((reason == IRQS_ON || reason == TRACE_STOP) &&
20458 + per_cpu(hist_irqsoff_counting, cpu)) {
20459 + cycle_t start = per_cpu(hist_irqsoff_start, cpu);
20461 + stop = ftrace_now(cpu);
20464 + long latency = ((long) (stop - start)) /
20467 + latency_hist(IRQSOFF_LATENCY, cpu, latency, 0,
20470 + per_cpu(hist_irqsoff_counting, cpu) = 0;
20474 +#ifdef CONFIG_PREEMPT_OFF_HIST
20475 + if ((reason == PREEMPT_ON || reason == TRACE_STOP) &&
20476 + per_cpu(hist_preemptoff_counting, cpu)) {
20477 + cycle_t start = per_cpu(hist_preemptoff_start, cpu);
20479 + if (!(time_set++))
20480 + stop = ftrace_now(cpu);
20482 + long latency = ((long) (stop - start)) /
20485 + latency_hist(PREEMPTOFF_LATENCY, cpu, latency,
20488 + per_cpu(hist_preemptoff_counting, cpu) = 0;
20492 +#if defined(CONFIG_INTERRUPT_OFF_HIST) && defined(CONFIG_PREEMPT_OFF_HIST)
20493 + if ((!per_cpu(hist_irqsoff_counting, cpu) ||
20494 + !per_cpu(hist_preemptoff_counting, cpu)) &&
20495 + per_cpu(hist_preemptirqsoff_counting, cpu)) {
20496 + cycle_t start = per_cpu(hist_preemptirqsoff_start, cpu);
20499 + stop = ftrace_now(cpu);
20501 + long latency = ((long) (stop - start)) /
20504 + latency_hist(PREEMPTIRQSOFF_LATENCY, cpu,
20505 + latency, 0, stop, NULL);
20507 + per_cpu(hist_preemptirqsoff_counting, cpu) = 0;
20514 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20515 +static DEFINE_RAW_SPINLOCK(wakeup_lock);
20516 +static notrace void probe_sched_migrate_task(void *v, struct task_struct *task,
20519 + int old_cpu = task_cpu(task);
20521 + if (cpu != old_cpu) {
20522 + unsigned long flags;
20523 + struct task_struct *cpu_wakeup_task;
20525 + raw_spin_lock_irqsave(&wakeup_lock, flags);
20527 + cpu_wakeup_task = per_cpu(wakeup_task, old_cpu);
20528 + if (task == cpu_wakeup_task) {
20529 + put_task_struct(cpu_wakeup_task);
20530 + per_cpu(wakeup_task, old_cpu) = NULL;
20531 + cpu_wakeup_task = per_cpu(wakeup_task, cpu) = task;
20532 + get_task_struct(cpu_wakeup_task);
20535 + raw_spin_unlock_irqrestore(&wakeup_lock, flags);
20539 +static notrace void probe_wakeup_latency_hist_start(void *v,
20540 + struct task_struct *p)
20542 + unsigned long flags;
20543 + struct task_struct *curr = current;
20544 + int cpu = task_cpu(p);
20545 + struct task_struct *cpu_wakeup_task;
20547 + raw_spin_lock_irqsave(&wakeup_lock, flags);
20549 + cpu_wakeup_task = per_cpu(wakeup_task, cpu);
20551 + if (wakeup_pid) {
20552 + if ((cpu_wakeup_task && p->prio == cpu_wakeup_task->prio) ||
20553 + p->prio == curr->prio)
20554 + per_cpu(wakeup_sharedprio, cpu) = 1;
20555 + if (likely(wakeup_pid != task_pid_nr(p)))
20558 + if (likely(!rt_task(p)) ||
20559 + (cpu_wakeup_task && p->prio > cpu_wakeup_task->prio) ||
20560 + p->prio > curr->prio)
20562 + if ((cpu_wakeup_task && p->prio == cpu_wakeup_task->prio) ||
20563 + p->prio == curr->prio)
20564 + per_cpu(wakeup_sharedprio, cpu) = 1;
20567 + if (cpu_wakeup_task)
20568 + put_task_struct(cpu_wakeup_task);
20569 + cpu_wakeup_task = per_cpu(wakeup_task, cpu) = p;
20570 + get_task_struct(cpu_wakeup_task);
20571 + cpu_wakeup_task->preempt_timestamp_hist =
20572 + ftrace_now(raw_smp_processor_id());
20574 + raw_spin_unlock_irqrestore(&wakeup_lock, flags);
20577 +static notrace void probe_wakeup_latency_hist_stop(void *v,
20578 + bool preempt, struct task_struct *prev, struct task_struct *next)
20580 + unsigned long flags;
20581 + int cpu = task_cpu(next);
20584 + struct task_struct *cpu_wakeup_task;
20586 + raw_spin_lock_irqsave(&wakeup_lock, flags);
20588 + cpu_wakeup_task = per_cpu(wakeup_task, cpu);
20590 + if (cpu_wakeup_task == NULL)
20593 + /* Already running? */
20594 + if (unlikely(current == cpu_wakeup_task))
20597 + if (next != cpu_wakeup_task) {
20598 + if (next->prio < cpu_wakeup_task->prio)
20601 + if (next->prio == cpu_wakeup_task->prio)
20602 + per_cpu(wakeup_sharedprio, cpu) = 1;
20607 + if (current->prio == cpu_wakeup_task->prio)
20608 + per_cpu(wakeup_sharedprio, cpu) = 1;
20611 + * The task we are waiting for is about to be switched to.
20612 + * Calculate latency and store it in histogram.
20614 + stop = ftrace_now(raw_smp_processor_id());
20616 + latency = ((long) (stop - next->preempt_timestamp_hist)) /
20619 + if (per_cpu(wakeup_sharedprio, cpu)) {
20620 + latency_hist(WAKEUP_LATENCY_SHAREDPRIO, cpu, latency, 0, stop,
20622 + per_cpu(wakeup_sharedprio, cpu) = 0;
20624 + latency_hist(WAKEUP_LATENCY, cpu, latency, 0, stop, next);
20625 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20626 + if (timerandwakeup_enabled_data.enabled) {
20627 + latency_hist(TIMERANDWAKEUP_LATENCY, cpu,
20628 + next->timer_offset + latency, next->timer_offset,
20635 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20636 + next->timer_offset = 0;
20638 + put_task_struct(cpu_wakeup_task);
20639 + per_cpu(wakeup_task, cpu) = NULL;
20641 + raw_spin_unlock_irqrestore(&wakeup_lock, flags);
20645 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20646 +static notrace void probe_hrtimer_interrupt(void *v, int cpu,
20647 + long long latency_ns, struct task_struct *curr,
20648 + struct task_struct *task)
20650 + if (latency_ns <= 0 && task != NULL && rt_task(task) &&
20651 + (task->prio < curr->prio ||
20652 + (task->prio == curr->prio &&
20653 + !cpumask_test_cpu(cpu, &task->cpus_allowed)))) {
20657 + if (missed_timer_offsets_pid) {
20658 + if (likely(missed_timer_offsets_pid !=
20659 + task_pid_nr(task)))
20663 + now = ftrace_now(cpu);
20664 + latency = (long) div_s64(-latency_ns, NSECS_PER_USECS);
20665 + latency_hist(MISSED_TIMER_OFFSETS, cpu, latency, latency, now,
20667 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20668 + task->timer_offset = latency;
20674 +static __init int latency_hist_init(void)
20676 + struct dentry *latency_hist_root = NULL;
20677 + struct dentry *dentry;
20678 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20679 + struct dentry *dentry_sharedprio;
20681 + struct dentry *entry;
20682 + struct dentry *enable_root;
20684 + struct hist_data *my_hist;
20686 + char *cpufmt = "CPU%d";
20687 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) || \
20688 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20689 + char *cpufmt_maxlatproc = "max_latency-CPU%d";
20690 + struct maxlatproc_data *mp = NULL;
20693 + dentry = tracing_init_dentry();
20694 + latency_hist_root = debugfs_create_dir(latency_hist_dir_root, dentry);
20695 + enable_root = debugfs_create_dir("enable", latency_hist_root);
20697 +#ifdef CONFIG_INTERRUPT_OFF_HIST
20698 + dentry = debugfs_create_dir(irqsoff_hist_dir, latency_hist_root);
20699 + for_each_possible_cpu(i) {
20700 + sprintf(name, cpufmt, i);
20701 + entry = debugfs_create_file(name, 0444, dentry,
20702 + &per_cpu(irqsoff_hist, i), &latency_hist_fops);
20703 + my_hist = &per_cpu(irqsoff_hist, i);
20704 + atomic_set(&my_hist->hist_mode, 1);
20705 + my_hist->min_lat = LONG_MAX;
20707 + entry = debugfs_create_file("reset", 0644, dentry,
20708 + (void *)IRQSOFF_LATENCY, &latency_hist_reset_fops);
20711 +#ifdef CONFIG_PREEMPT_OFF_HIST
20712 + dentry = debugfs_create_dir(preemptoff_hist_dir,
20713 + latency_hist_root);
20714 + for_each_possible_cpu(i) {
20715 + sprintf(name, cpufmt, i);
20716 + entry = debugfs_create_file(name, 0444, dentry,
20717 + &per_cpu(preemptoff_hist, i), &latency_hist_fops);
20718 + my_hist = &per_cpu(preemptoff_hist, i);
20719 + atomic_set(&my_hist->hist_mode, 1);
20720 + my_hist->min_lat = LONG_MAX;
20722 + entry = debugfs_create_file("reset", 0644, dentry,
20723 + (void *)PREEMPTOFF_LATENCY, &latency_hist_reset_fops);
20726 +#if defined(CONFIG_INTERRUPT_OFF_HIST) && defined(CONFIG_PREEMPT_OFF_HIST)
20727 + dentry = debugfs_create_dir(preemptirqsoff_hist_dir,
20728 + latency_hist_root);
20729 + for_each_possible_cpu(i) {
20730 + sprintf(name, cpufmt, i);
20731 + entry = debugfs_create_file(name, 0444, dentry,
20732 + &per_cpu(preemptirqsoff_hist, i), &latency_hist_fops);
20733 + my_hist = &per_cpu(preemptirqsoff_hist, i);
20734 + atomic_set(&my_hist->hist_mode, 1);
20735 + my_hist->min_lat = LONG_MAX;
20737 + entry = debugfs_create_file("reset", 0644, dentry,
20738 + (void *)PREEMPTIRQSOFF_LATENCY, &latency_hist_reset_fops);
20741 +#if defined(CONFIG_INTERRUPT_OFF_HIST) || defined(CONFIG_PREEMPT_OFF_HIST)
20742 + entry = debugfs_create_file("preemptirqsoff", 0644,
20743 + enable_root, (void *)&preemptirqsoff_enabled_data,
20747 +#ifdef CONFIG_WAKEUP_LATENCY_HIST
20748 + dentry = debugfs_create_dir(wakeup_latency_hist_dir,
20749 + latency_hist_root);
20750 + dentry_sharedprio = debugfs_create_dir(
20751 + wakeup_latency_hist_dir_sharedprio, dentry);
20752 + for_each_possible_cpu(i) {
20753 + sprintf(name, cpufmt, i);
20755 + entry = debugfs_create_file(name, 0444, dentry,
20756 + &per_cpu(wakeup_latency_hist, i),
20757 + &latency_hist_fops);
20758 + my_hist = &per_cpu(wakeup_latency_hist, i);
20759 + atomic_set(&my_hist->hist_mode, 1);
20760 + my_hist->min_lat = LONG_MAX;
20762 + entry = debugfs_create_file(name, 0444, dentry_sharedprio,
20763 + &per_cpu(wakeup_latency_hist_sharedprio, i),
20764 + &latency_hist_fops);
20765 + my_hist = &per_cpu(wakeup_latency_hist_sharedprio, i);
20766 + atomic_set(&my_hist->hist_mode, 1);
20767 + my_hist->min_lat = LONG_MAX;
20769 + sprintf(name, cpufmt_maxlatproc, i);
20771 + mp = &per_cpu(wakeup_maxlatproc, i);
20772 + entry = debugfs_create_file(name, 0444, dentry, mp,
20773 + &maxlatproc_fops);
20774 + clear_maxlatprocdata(mp);
20776 + mp = &per_cpu(wakeup_maxlatproc_sharedprio, i);
20777 + entry = debugfs_create_file(name, 0444, dentry_sharedprio, mp,
20778 + &maxlatproc_fops);
20779 + clear_maxlatprocdata(mp);
20781 + entry = debugfs_create_file("pid", 0644, dentry,
20782 + (void *)&wakeup_pid, &pid_fops);
20783 + entry = debugfs_create_file("reset", 0644, dentry,
20784 + (void *)WAKEUP_LATENCY, &latency_hist_reset_fops);
20785 + entry = debugfs_create_file("reset", 0644, dentry_sharedprio,
20786 + (void *)WAKEUP_LATENCY_SHAREDPRIO, &latency_hist_reset_fops);
20787 + entry = debugfs_create_file("wakeup", 0644,
20788 + enable_root, (void *)&wakeup_latency_enabled_data,
20792 +#ifdef CONFIG_MISSED_TIMER_OFFSETS_HIST
20793 + dentry = debugfs_create_dir(missed_timer_offsets_dir,
20794 + latency_hist_root);
20795 + for_each_possible_cpu(i) {
20796 + sprintf(name, cpufmt, i);
20797 + entry = debugfs_create_file(name, 0444, dentry,
20798 + &per_cpu(missed_timer_offsets, i), &latency_hist_fops);
20799 + my_hist = &per_cpu(missed_timer_offsets, i);
20800 + atomic_set(&my_hist->hist_mode, 1);
20801 + my_hist->min_lat = LONG_MAX;
20803 + sprintf(name, cpufmt_maxlatproc, i);
20804 + mp = &per_cpu(missed_timer_offsets_maxlatproc, i);
20805 + entry = debugfs_create_file(name, 0444, dentry, mp,
20806 + &maxlatproc_fops);
20807 + clear_maxlatprocdata(mp);
20809 + entry = debugfs_create_file("pid", 0644, dentry,
20810 + (void *)&missed_timer_offsets_pid, &pid_fops);
20811 + entry = debugfs_create_file("reset", 0644, dentry,
20812 + (void *)MISSED_TIMER_OFFSETS, &latency_hist_reset_fops);
20813 + entry = debugfs_create_file("missed_timer_offsets", 0644,
20814 + enable_root, (void *)&missed_timer_offsets_enabled_data,
20818 +#if defined(CONFIG_WAKEUP_LATENCY_HIST) && \
20819 + defined(CONFIG_MISSED_TIMER_OFFSETS_HIST)
20820 + dentry = debugfs_create_dir(timerandwakeup_latency_hist_dir,
20821 + latency_hist_root);
20822 + for_each_possible_cpu(i) {
20823 + sprintf(name, cpufmt, i);
20824 + entry = debugfs_create_file(name, 0444, dentry,
20825 + &per_cpu(timerandwakeup_latency_hist, i),
20826 + &latency_hist_fops);
20827 + my_hist = &per_cpu(timerandwakeup_latency_hist, i);
20828 + atomic_set(&my_hist->hist_mode, 1);
20829 + my_hist->min_lat = LONG_MAX;
20831 + sprintf(name, cpufmt_maxlatproc, i);
20832 + mp = &per_cpu(timerandwakeup_maxlatproc, i);
20833 + entry = debugfs_create_file(name, 0444, dentry, mp,
20834 + &maxlatproc_fops);
20835 + clear_maxlatprocdata(mp);
20837 + entry = debugfs_create_file("reset", 0644, dentry,
20838 + (void *)TIMERANDWAKEUP_LATENCY, &latency_hist_reset_fops);
20839 + entry = debugfs_create_file("timerandwakeup", 0644,
20840 + enable_root, (void *)&timerandwakeup_enabled_data,
20846 +device_initcall(latency_hist_init);
20847 diff --git a/kernel/trace/trace.c b/kernel/trace/trace.c
20848 index 8696ce6bf2f6..277f048a4695 100644
20849 --- a/kernel/trace/trace.c
20850 +++ b/kernel/trace/trace.c
20851 @@ -1897,6 +1897,7 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
20852 struct task_struct *tsk = current;
20854 entry->preempt_count = pc & 0xff;
20855 + entry->preempt_lazy_count = preempt_lazy_count();
20856 entry->pid = (tsk) ? tsk->pid : 0;
20858 #ifdef CONFIG_TRACE_IRQFLAGS_SUPPORT
20859 @@ -1907,8 +1908,11 @@ tracing_generic_entry_update(struct trace_entry *entry, unsigned long flags,
20860 ((pc & NMI_MASK ) ? TRACE_FLAG_NMI : 0) |
20861 ((pc & HARDIRQ_MASK) ? TRACE_FLAG_HARDIRQ : 0) |
20862 ((pc & SOFTIRQ_MASK) ? TRACE_FLAG_SOFTIRQ : 0) |
20863 - (tif_need_resched() ? TRACE_FLAG_NEED_RESCHED : 0) |
20864 + (tif_need_resched_now() ? TRACE_FLAG_NEED_RESCHED : 0) |
20865 + (need_resched_lazy() ? TRACE_FLAG_NEED_RESCHED_LAZY : 0) |
20866 (test_preempt_need_resched() ? TRACE_FLAG_PREEMPT_RESCHED : 0);
20868 + entry->migrate_disable = (tsk) ? __migrate_disabled(tsk) & 0xFF : 0;
20870 EXPORT_SYMBOL_GPL(tracing_generic_entry_update);
20872 @@ -2892,14 +2896,17 @@ get_total_entries(struct trace_buffer *buf,
20874 static void print_lat_help_header(struct seq_file *m)
20876 - seq_puts(m, "# _------=> CPU# \n"
20877 - "# / _-----=> irqs-off \n"
20878 - "# | / _----=> need-resched \n"
20879 - "# || / _---=> hardirq/softirq \n"
20880 - "# ||| / _--=> preempt-depth \n"
20881 - "# |||| / delay \n"
20882 - "# cmd pid ||||| time | caller \n"
20883 - "# \\ / ||||| \\ | / \n");
20884 + seq_puts(m, "# _--------=> CPU# \n"
20885 + "# / _-------=> irqs-off \n"
20886 + "# | / _------=> need-resched \n"
20887 + "# || / _-----=> need-resched_lazy \n"
20888 + "# ||| / _----=> hardirq/softirq \n"
20889 + "# |||| / _---=> preempt-depth \n"
20890 + "# ||||| / _--=> preempt-lazy-depth\n"
20891 + "# |||||| / _-=> migrate-disable \n"
20892 + "# ||||||| / delay \n"
20893 + "# cmd pid |||||||| time | caller \n"
20894 + "# \\ / |||||||| \\ | / \n");
20897 static void print_event_info(struct trace_buffer *buf, struct seq_file *m)
20898 @@ -2925,11 +2932,14 @@ static void print_func_help_header_irq(struct trace_buffer *buf, struct seq_file
20899 print_event_info(buf, m);
20900 seq_puts(m, "# _-----=> irqs-off\n"
20901 "# / _----=> need-resched\n"
20902 - "# | / _---=> hardirq/softirq\n"
20903 - "# || / _--=> preempt-depth\n"
20904 - "# ||| / delay\n"
20905 - "# TASK-PID CPU# |||| TIMESTAMP FUNCTION\n"
20906 - "# | | | |||| | |\n");
20907 + "# |/ _-----=> need-resched_lazy\n"
20908 + "# || / _---=> hardirq/softirq\n"
20909 + "# ||| / _--=> preempt-depth\n"
20910 + "# |||| / _-=> preempt-lazy-depth\n"
20911 + "# ||||| / _-=> migrate-disable \n"
20912 + "# |||||| / delay\n"
20913 + "# TASK-PID CPU# ||||||| TIMESTAMP FUNCTION\n"
20914 + "# | | | ||||||| | |\n");
20918 diff --git a/kernel/trace/trace.h b/kernel/trace/trace.h
20919 index fd24b1f9ac43..852b2c81be25 100644
20920 --- a/kernel/trace/trace.h
20921 +++ b/kernel/trace/trace.h
20922 @@ -124,6 +124,7 @@ struct kretprobe_trace_entry_head {
20923 * NEED_RESCHED - reschedule is requested
20924 * HARDIRQ - inside an interrupt handler
20925 * SOFTIRQ - inside a softirq handler
20926 + * NEED_RESCHED_LAZY - lazy reschedule is requested
20928 enum trace_flag_type {
20929 TRACE_FLAG_IRQS_OFF = 0x01,
20930 @@ -133,6 +134,7 @@ enum trace_flag_type {
20931 TRACE_FLAG_SOFTIRQ = 0x10,
20932 TRACE_FLAG_PREEMPT_RESCHED = 0x20,
20933 TRACE_FLAG_NMI = 0x40,
20934 + TRACE_FLAG_NEED_RESCHED_LAZY = 0x80,
20937 #define TRACE_BUF_SIZE 1024
20938 diff --git a/kernel/trace/trace_events.c b/kernel/trace/trace_events.c
20939 index 03c0a48c3ac4..0b85d516b491 100644
20940 --- a/kernel/trace/trace_events.c
20941 +++ b/kernel/trace/trace_events.c
20942 @@ -187,6 +187,8 @@ static int trace_define_common_fields(void)
20943 __common_field(unsigned char, flags);
20944 __common_field(unsigned char, preempt_count);
20945 __common_field(int, pid);
20946 + __common_field(unsigned short, migrate_disable);
20947 + __common_field(unsigned short, padding);
20951 diff --git a/kernel/trace/trace_irqsoff.c b/kernel/trace/trace_irqsoff.c
20952 index 03cdff84d026..940bd10b4406 100644
20953 --- a/kernel/trace/trace_irqsoff.c
20954 +++ b/kernel/trace/trace_irqsoff.c
20956 #include <linux/uaccess.h>
20957 #include <linux/module.h>
20958 #include <linux/ftrace.h>
20959 +#include <trace/events/hist.h>
20963 @@ -424,11 +425,13 @@ void start_critical_timings(void)
20965 if (preempt_trace() || irq_trace())
20966 start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
20967 + trace_preemptirqsoff_hist_rcuidle(TRACE_START, 1);
20969 EXPORT_SYMBOL_GPL(start_critical_timings);
20971 void stop_critical_timings(void)
20973 + trace_preemptirqsoff_hist_rcuidle(TRACE_STOP, 0);
20974 if (preempt_trace() || irq_trace())
20975 stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
20977 @@ -438,6 +441,7 @@ EXPORT_SYMBOL_GPL(stop_critical_timings);
20978 #ifdef CONFIG_PROVE_LOCKING
20979 void time_hardirqs_on(unsigned long a0, unsigned long a1)
20981 + trace_preemptirqsoff_hist_rcuidle(IRQS_ON, 0);
20982 if (!preempt_trace() && irq_trace())
20983 stop_critical_timing(a0, a1);
20985 @@ -446,6 +450,7 @@ void time_hardirqs_off(unsigned long a0, unsigned long a1)
20987 if (!preempt_trace() && irq_trace())
20988 start_critical_timing(a0, a1);
20989 + trace_preemptirqsoff_hist_rcuidle(IRQS_OFF, 1);
20992 #else /* !CONFIG_PROVE_LOCKING */
20993 @@ -471,6 +476,7 @@ inline void print_irqtrace_events(struct task_struct *curr)
20995 void trace_hardirqs_on(void)
20997 + trace_preemptirqsoff_hist(IRQS_ON, 0);
20998 if (!preempt_trace() && irq_trace())
20999 stop_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
21001 @@ -480,11 +486,13 @@ void trace_hardirqs_off(void)
21003 if (!preempt_trace() && irq_trace())
21004 start_critical_timing(CALLER_ADDR0, CALLER_ADDR1);
21005 + trace_preemptirqsoff_hist(IRQS_OFF, 1);
21007 EXPORT_SYMBOL(trace_hardirqs_off);
21009 __visible void trace_hardirqs_on_caller(unsigned long caller_addr)
21011 + trace_preemptirqsoff_hist(IRQS_ON, 0);
21012 if (!preempt_trace() && irq_trace())
21013 stop_critical_timing(CALLER_ADDR0, caller_addr);
21015 @@ -494,6 +502,7 @@ __visible void trace_hardirqs_off_caller(unsigned long caller_addr)
21017 if (!preempt_trace() && irq_trace())
21018 start_critical_timing(CALLER_ADDR0, caller_addr);
21019 + trace_preemptirqsoff_hist(IRQS_OFF, 1);
21021 EXPORT_SYMBOL(trace_hardirqs_off_caller);
21023 @@ -503,12 +512,14 @@ EXPORT_SYMBOL(trace_hardirqs_off_caller);
21024 #ifdef CONFIG_PREEMPT_TRACER
21025 void trace_preempt_on(unsigned long a0, unsigned long a1)
21027 + trace_preemptirqsoff_hist(PREEMPT_ON, 0);
21028 if (preempt_trace() && !irq_trace())
21029 stop_critical_timing(a0, a1);
21032 void trace_preempt_off(unsigned long a0, unsigned long a1)
21034 + trace_preemptirqsoff_hist(PREEMPT_ON, 1);
21035 if (preempt_trace() && !irq_trace())
21036 start_critical_timing(a0, a1);
21038 diff --git a/kernel/trace/trace_output.c b/kernel/trace/trace_output.c
21039 index 3fc20422c166..65a6dde71a7d 100644
21040 --- a/kernel/trace/trace_output.c
21041 +++ b/kernel/trace/trace_output.c
21042 @@ -386,6 +386,7 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry)
21046 + char need_resched_lazy;
21050 @@ -416,6 +417,9 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry)
21054 + need_resched_lazy =
21055 + (entry->flags & TRACE_FLAG_NEED_RESCHED_LAZY) ? 'L' : '.';
21058 (nmi && hardirq) ? 'Z' :
21060 @@ -424,14 +428,25 @@ int trace_print_lat_fmt(struct trace_seq *s, struct trace_entry *entry)
21064 - trace_seq_printf(s, "%c%c%c",
21065 - irqs_off, need_resched, hardsoft_irq);
21066 + trace_seq_printf(s, "%c%c%c%c",
21067 + irqs_off, need_resched, need_resched_lazy,
21070 if (entry->preempt_count)
21071 trace_seq_printf(s, "%x", entry->preempt_count);
21073 trace_seq_putc(s, '.');
21075 + if (entry->preempt_lazy_count)
21076 + trace_seq_printf(s, "%x", entry->preempt_lazy_count);
21078 + trace_seq_putc(s, '.');
21080 + if (entry->migrate_disable)
21081 + trace_seq_printf(s, "%x", entry->migrate_disable);
21083 + trace_seq_putc(s, '.');
21085 return !trace_seq_has_overflowed(s);
21088 diff --git a/kernel/user.c b/kernel/user.c
21089 index b069ccbfb0b0..1a2e88e98b5e 100644
21090 --- a/kernel/user.c
21091 +++ b/kernel/user.c
21092 @@ -161,11 +161,11 @@ void free_uid(struct user_struct *up)
21096 - local_irq_save(flags);
21097 + local_irq_save_nort(flags);
21098 if (atomic_dec_and_lock(&up->__count, &uidhash_lock))
21099 free_user(up, flags);
21101 - local_irq_restore(flags);
21102 + local_irq_restore_nort(flags);
21105 struct user_struct *alloc_uid(kuid_t uid)
21106 diff --git a/kernel/watchdog.c b/kernel/watchdog.c
21107 index 6d1020c03d41..70c6a2f79f7e 100644
21108 --- a/kernel/watchdog.c
21109 +++ b/kernel/watchdog.c
21110 @@ -315,6 +315,8 @@ static int is_softlockup(unsigned long touch_ts)
21112 #ifdef CONFIG_HARDLOCKUP_DETECTOR
21114 +static DEFINE_RAW_SPINLOCK(watchdog_output_lock);
21116 static struct perf_event_attr wd_hw_attr = {
21117 .type = PERF_TYPE_HARDWARE,
21118 .config = PERF_COUNT_HW_CPU_CYCLES,
21119 @@ -348,6 +350,13 @@ static void watchdog_overflow_callback(struct perf_event *event,
21120 /* only print hardlockups once */
21121 if (__this_cpu_read(hard_watchdog_warn) == true)
21124 + * If early-printk is enabled then make sure we do not
21125 + * lock up in printk() and kill console logging:
21129 + raw_spin_lock(&watchdog_output_lock);
21131 pr_emerg("Watchdog detected hard LOCKUP on cpu %d", this_cpu);
21133 @@ -365,6 +374,7 @@ static void watchdog_overflow_callback(struct perf_event *event,
21134 !test_and_set_bit(0, &hardlockup_allcpu_dumped))
21135 trigger_allbutself_cpu_backtrace();
21137 + raw_spin_unlock(&watchdog_output_lock);
21138 if (hardlockup_panic)
21139 nmi_panic(regs, "Hard LOCKUP");
21141 @@ -512,6 +522,7 @@ static void watchdog_enable(unsigned int cpu)
21142 /* kick off the timer for the hardlockup detector */
21143 hrtimer_init(hrtimer, CLOCK_MONOTONIC, HRTIMER_MODE_REL);
21144 hrtimer->function = watchdog_timer_fn;
21145 + hrtimer->irqsafe = 1;
21147 /* Enable the perf event */
21148 watchdog_nmi_enable(cpu);
21149 diff --git a/kernel/workqueue.c b/kernel/workqueue.c
21150 index 479d840db286..24eba6620a45 100644
21151 --- a/kernel/workqueue.c
21152 +++ b/kernel/workqueue.c
21154 #include <linux/nodemask.h>
21155 #include <linux/moduleparam.h>
21156 #include <linux/uaccess.h>
21157 +#include <linux/locallock.h>
21158 +#include <linux/delay.h>
21160 #include "workqueue_internal.h"
21162 @@ -121,11 +123,16 @@ enum {
21163 * cpu or grabbing pool->lock is enough for read access. If
21164 * POOL_DISASSOCIATED is set, it's identical to L.
21166 + * On RT we need the extra protection via rt_lock_idle_list() for
21167 + * the list manipulations against read access from
21168 + * wq_worker_sleeping(). All other places are nicely serialized via
21171 * A: pool->attach_mutex protected.
21173 * PL: wq_pool_mutex protected.
21175 - * PR: wq_pool_mutex protected for writes. Sched-RCU protected for reads.
21176 + * PR: wq_pool_mutex protected for writes. RCU protected for reads.
21178 * PW: wq_pool_mutex and wq->mutex protected for writes. Either for reads.
21180 @@ -134,7 +141,7 @@ enum {
21182 * WQ: wq->mutex protected.
21184 - * WR: wq->mutex protected for writes. Sched-RCU protected for reads.
21185 + * WR: wq->mutex protected for writes. RCU protected for reads.
21187 * MD: wq_mayday_lock protected.
21189 @@ -185,7 +192,7 @@ struct worker_pool {
21190 atomic_t nr_running ____cacheline_aligned_in_smp;
21193 - * Destruction of pool is sched-RCU protected to allow dereferences
21194 + * Destruction of pool is RCU protected to allow dereferences
21195 * from get_work_pool().
21197 struct rcu_head rcu;
21198 @@ -214,7 +221,7 @@ struct pool_workqueue {
21200 * Release of unbound pwq is punted to system_wq. See put_pwq()
21201 * and pwq_unbound_release_workfn() for details. pool_workqueue
21202 - * itself is also sched-RCU protected so that the first pwq can be
21203 + * itself is also RCU protected so that the first pwq can be
21204 * determined without grabbing wq->mutex.
21206 struct work_struct unbound_release_work;
21207 @@ -348,6 +355,8 @@ EXPORT_SYMBOL_GPL(system_power_efficient_wq);
21208 struct workqueue_struct *system_freezable_power_efficient_wq __read_mostly;
21209 EXPORT_SYMBOL_GPL(system_freezable_power_efficient_wq);
21211 +static DEFINE_LOCAL_IRQ_LOCK(pendingb_lock);
21213 static int worker_thread(void *__worker);
21214 static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
21216 @@ -355,20 +364,20 @@ static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
21217 #include <trace/events/workqueue.h>
21219 #define assert_rcu_or_pool_mutex() \
21220 - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held() && \
21221 + RCU_LOCKDEP_WARN(!rcu_read_lock_held() && \
21222 !lockdep_is_held(&wq_pool_mutex), \
21223 - "sched RCU or wq_pool_mutex should be held")
21224 + "RCU or wq_pool_mutex should be held")
21226 #define assert_rcu_or_wq_mutex(wq) \
21227 - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held() && \
21228 + RCU_LOCKDEP_WARN(!rcu_read_lock_held() && \
21229 !lockdep_is_held(&wq->mutex), \
21230 - "sched RCU or wq->mutex should be held")
21231 + "RCU or wq->mutex should be held")
21233 #define assert_rcu_or_wq_mutex_or_pool_mutex(wq) \
21234 - RCU_LOCKDEP_WARN(!rcu_read_lock_sched_held() && \
21235 + RCU_LOCKDEP_WARN(!rcu_read_lock_held() && \
21236 !lockdep_is_held(&wq->mutex) && \
21237 !lockdep_is_held(&wq_pool_mutex), \
21238 - "sched RCU, wq->mutex or wq_pool_mutex should be held")
21239 + "RCU, wq->mutex or wq_pool_mutex should be held")
21241 #define for_each_cpu_worker_pool(pool, cpu) \
21242 for ((pool) = &per_cpu(cpu_worker_pools, cpu)[0]; \
21243 @@ -380,7 +389,7 @@ static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
21244 * @pool: iteration cursor
21245 * @pi: integer used for iteration
21247 - * This must be called either with wq_pool_mutex held or sched RCU read
21248 + * This must be called either with wq_pool_mutex held or RCU read
21249 * locked. If the pool needs to be used beyond the locking in effect, the
21250 * caller is responsible for guaranteeing that the pool stays online.
21252 @@ -412,7 +421,7 @@ static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
21253 * @pwq: iteration cursor
21254 * @wq: the target workqueue
21256 - * This must be called either with wq->mutex held or sched RCU read locked.
21257 + * This must be called either with wq->mutex held or RCU read locked.
21258 * If the pwq needs to be used beyond the locking in effect, the caller is
21259 * responsible for guaranteeing that the pwq stays online.
21261 @@ -424,6 +433,31 @@ static void workqueue_sysfs_unregister(struct workqueue_struct *wq);
21262 if (({ assert_rcu_or_wq_mutex(wq); false; })) { } \
21265 +#ifdef CONFIG_PREEMPT_RT_BASE
21266 +static inline void rt_lock_idle_list(struct worker_pool *pool)
21268 + preempt_disable();
21270 +static inline void rt_unlock_idle_list(struct worker_pool *pool)
21272 + preempt_enable();
21274 +static inline void sched_lock_idle_list(struct worker_pool *pool) { }
21275 +static inline void sched_unlock_idle_list(struct worker_pool *pool) { }
21277 +static inline void rt_lock_idle_list(struct worker_pool *pool) { }
21278 +static inline void rt_unlock_idle_list(struct worker_pool *pool) { }
21279 +static inline void sched_lock_idle_list(struct worker_pool *pool)
21281 + spin_lock_irq(&pool->lock);
21283 +static inline void sched_unlock_idle_list(struct worker_pool *pool)
21285 + spin_unlock_irq(&pool->lock);
21290 #ifdef CONFIG_DEBUG_OBJECTS_WORK
21292 static struct debug_obj_descr work_debug_descr;
21293 @@ -548,7 +582,7 @@ static int worker_pool_assign_id(struct worker_pool *pool)
21294 * @wq: the target workqueue
21295 * @node: the node ID
21297 - * This must be called with any of wq_pool_mutex, wq->mutex or sched RCU
21298 + * This must be called with any of wq_pool_mutex, wq->mutex or RCU
21300 * If the pwq needs to be used beyond the locking in effect, the caller is
21301 * responsible for guaranteeing that the pwq stays online.
21302 @@ -692,8 +726,8 @@ static struct pool_workqueue *get_work_pwq(struct work_struct *work)
21303 * @work: the work item of interest
21305 * Pools are created and destroyed under wq_pool_mutex, and allows read
21306 - * access under sched-RCU read lock. As such, this function should be
21307 - * called under wq_pool_mutex or with preemption disabled.
21308 + * access under RCU read lock. As such, this function should be
21309 + * called under wq_pool_mutex or inside of a rcu_read_lock() region.
21311 * All fields of the returned pool are accessible as long as the above
21312 * mentioned locking is in effect. If the returned pool needs to be used
21313 @@ -830,50 +864,45 @@ static struct worker *first_idle_worker(struct worker_pool *pool)
21315 static void wake_up_worker(struct worker_pool *pool)
21317 - struct worker *worker = first_idle_worker(pool);
21318 + struct worker *worker;
21320 + rt_lock_idle_list(pool);
21322 + worker = first_idle_worker(pool);
21324 if (likely(worker))
21325 wake_up_process(worker->task);
21327 + rt_unlock_idle_list(pool);
21331 - * wq_worker_waking_up - a worker is waking up
21332 + * wq_worker_running - a worker is running again
21333 * @task: task waking up
21334 - * @cpu: CPU @task is waking up to
21336 - * This function is called during try_to_wake_up() when a worker is
21340 - * spin_lock_irq(rq->lock)
21341 + * This function is called when a worker returns from schedule()
21343 -void wq_worker_waking_up(struct task_struct *task, int cpu)
21344 +void wq_worker_running(struct task_struct *task)
21346 struct worker *worker = kthread_data(task);
21348 - if (!(worker->flags & WORKER_NOT_RUNNING)) {
21349 - WARN_ON_ONCE(worker->pool->cpu != cpu);
21350 + if (!worker->sleeping)
21352 + if (!(worker->flags & WORKER_NOT_RUNNING))
21353 atomic_inc(&worker->pool->nr_running);
21355 + worker->sleeping = 0;
21359 * wq_worker_sleeping - a worker is going to sleep
21360 * @task: task going to sleep
21362 - * This function is called during schedule() when a busy worker is
21363 - * going to sleep. Worker on the same cpu can be woken up by
21364 - * returning pointer to its task.
21367 - * spin_lock_irq(rq->lock)
21370 - * Worker task on @cpu to wake up, %NULL if none.
21371 + * This function is called from schedule() when a busy worker is
21372 + * going to sleep.
21374 -struct task_struct *wq_worker_sleeping(struct task_struct *task)
21375 +void wq_worker_sleeping(struct task_struct *task)
21377 - struct worker *worker = kthread_data(task), *to_wakeup = NULL;
21378 + struct worker *worker = kthread_data(task);
21379 struct worker_pool *pool;
21382 @@ -882,29 +911,26 @@ struct task_struct *wq_worker_sleeping(struct task_struct *task)
21383 * checking NOT_RUNNING.
21385 if (worker->flags & WORKER_NOT_RUNNING)
21389 pool = worker->pool;
21391 - /* this can only happen on the local cpu */
21392 - if (WARN_ON_ONCE(pool->cpu != raw_smp_processor_id()))
21394 + if (WARN_ON_ONCE(worker->sleeping))
21397 + worker->sleeping = 1;
21400 * The counterpart of the following dec_and_test, implied mb,
21401 * worklist not empty test sequence is in insert_work().
21402 * Please read comment there.
21404 - * NOT_RUNNING is clear. This means that we're bound to and
21405 - * running on the local cpu w/ rq lock held and preemption
21406 - * disabled, which in turn means that none else could be
21407 - * manipulating idle_list, so dereferencing idle_list without pool
21410 if (atomic_dec_and_test(&pool->nr_running) &&
21411 - !list_empty(&pool->worklist))
21412 - to_wakeup = first_idle_worker(pool);
21413 - return to_wakeup ? to_wakeup->task : NULL;
21414 + !list_empty(&pool->worklist)) {
21415 + sched_lock_idle_list(pool);
21416 + wake_up_worker(pool);
21417 + sched_unlock_idle_list(pool);
21422 @@ -1098,12 +1124,14 @@ static void put_pwq_unlocked(struct pool_workqueue *pwq)
21426 - * As both pwqs and pools are sched-RCU protected, the
21427 + * As both pwqs and pools are RCU protected, the
21428 * following lock operations are safe.
21430 - spin_lock_irq(&pwq->pool->lock);
21432 + local_spin_lock_irq(pendingb_lock, &pwq->pool->lock);
21434 - spin_unlock_irq(&pwq->pool->lock);
21435 + local_spin_unlock_irq(pendingb_lock, &pwq->pool->lock);
21436 + rcu_read_unlock();
21440 @@ -1207,7 +1235,7 @@ static int try_to_grab_pending(struct work_struct *work, bool is_dwork,
21441 struct worker_pool *pool;
21442 struct pool_workqueue *pwq;
21444 - local_irq_save(*flags);
21445 + local_lock_irqsave(pendingb_lock, *flags);
21447 /* try to steal the timer if it exists */
21449 @@ -1226,6 +1254,7 @@ static int try_to_grab_pending(struct work_struct *work, bool is_dwork,
21450 if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work)))
21455 * The queueing is in progress, or it is already queued. Try to
21456 * steal it from ->worklist without clearing WORK_STRUCT_PENDING.
21457 @@ -1264,14 +1293,16 @@ static int try_to_grab_pending(struct work_struct *work, bool is_dwork,
21458 set_work_pool_and_keep_pending(work, pool->id);
21460 spin_unlock(&pool->lock);
21461 + rcu_read_unlock();
21464 spin_unlock(&pool->lock);
21466 - local_irq_restore(*flags);
21467 + rcu_read_unlock();
21468 + local_unlock_irqrestore(pendingb_lock, *flags);
21469 if (work_is_canceling(work))
21476 @@ -1373,7 +1404,7 @@ static void __queue_work(int cpu, struct workqueue_struct *wq,
21477 * queued or lose PENDING. Grabbing PENDING and queueing should
21478 * happen with IRQ disabled.
21480 - WARN_ON_ONCE(!irqs_disabled());
21481 + WARN_ON_ONCE_NONRT(!irqs_disabled());
21483 debug_work_activate(work);
21485 @@ -1381,6 +1412,7 @@ static void __queue_work(int cpu, struct workqueue_struct *wq,
21486 if (unlikely(wq->flags & __WQ_DRAINING) &&
21487 WARN_ON_ONCE(!is_chained_work(wq)))
21491 if (req_cpu == WORK_CPU_UNBOUND)
21492 cpu = wq_select_unbound_cpu(raw_smp_processor_id());
21493 @@ -1437,10 +1469,8 @@ static void __queue_work(int cpu, struct workqueue_struct *wq,
21494 /* pwq determined, queue */
21495 trace_workqueue_queue_work(req_cpu, pwq, work);
21497 - if (WARN_ON(!list_empty(&work->entry))) {
21498 - spin_unlock(&pwq->pool->lock);
21501 + if (WARN_ON(!list_empty(&work->entry)))
21504 pwq->nr_in_flight[pwq->work_color]++;
21505 work_flags = work_color_to_flags(pwq->work_color);
21506 @@ -1458,7 +1488,9 @@ static void __queue_work(int cpu, struct workqueue_struct *wq,
21508 insert_work(pwq, work, worklist, work_flags);
21511 spin_unlock(&pwq->pool->lock);
21512 + rcu_read_unlock();
21516 @@ -1478,14 +1510,14 @@ bool queue_work_on(int cpu, struct workqueue_struct *wq,
21518 unsigned long flags;
21520 - local_irq_save(flags);
21521 + local_lock_irqsave(pendingb_lock,flags);
21523 if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))) {
21524 __queue_work(cpu, wq, work);
21528 - local_irq_restore(flags);
21529 + local_unlock_irqrestore(pendingb_lock, flags);
21532 EXPORT_SYMBOL(queue_work_on);
21533 @@ -1552,14 +1584,14 @@ bool queue_delayed_work_on(int cpu, struct workqueue_struct *wq,
21534 unsigned long flags;
21536 /* read the comment in __queue_work() */
21537 - local_irq_save(flags);
21538 + local_lock_irqsave(pendingb_lock, flags);
21540 if (!test_and_set_bit(WORK_STRUCT_PENDING_BIT, work_data_bits(work))) {
21541 __queue_delayed_work(cpu, wq, dwork, delay);
21545 - local_irq_restore(flags);
21546 + local_unlock_irqrestore(pendingb_lock, flags);
21549 EXPORT_SYMBOL(queue_delayed_work_on);
21550 @@ -1594,7 +1626,7 @@ bool mod_delayed_work_on(int cpu, struct workqueue_struct *wq,
21552 if (likely(ret >= 0)) {
21553 __queue_delayed_work(cpu, wq, dwork, delay);
21554 - local_irq_restore(flags);
21555 + local_unlock_irqrestore(pendingb_lock, flags);
21558 /* -ENOENT from try_to_grab_pending() becomes %true */
21559 @@ -1627,7 +1659,9 @@ static void worker_enter_idle(struct worker *worker)
21560 worker->last_active = jiffies;
21562 /* idle_list is LIFO */
21563 + rt_lock_idle_list(pool);
21564 list_add(&worker->entry, &pool->idle_list);
21565 + rt_unlock_idle_list(pool);
21567 if (too_many_workers(pool) && !timer_pending(&pool->idle_timer))
21568 mod_timer(&pool->idle_timer, jiffies + IDLE_WORKER_TIMEOUT);
21569 @@ -1660,7 +1694,9 @@ static void worker_leave_idle(struct worker *worker)
21571 worker_clr_flags(worker, WORKER_IDLE);
21573 + rt_lock_idle_list(pool);
21574 list_del_init(&worker->entry);
21575 + rt_unlock_idle_list(pool);
21578 static struct worker *alloc_worker(int node)
21579 @@ -1826,7 +1862,9 @@ static void destroy_worker(struct worker *worker)
21580 pool->nr_workers--;
21583 + rt_lock_idle_list(pool);
21584 list_del_init(&worker->entry);
21585 + rt_unlock_idle_list(pool);
21586 worker->flags |= WORKER_DIE;
21587 wake_up_process(worker->task);
21589 @@ -2785,14 +2823,14 @@ static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr)
21593 - local_irq_disable();
21595 pool = get_work_pool(work);
21597 - local_irq_enable();
21598 + rcu_read_unlock();
21602 - spin_lock(&pool->lock);
21603 + spin_lock_irq(&pool->lock);
21604 /* see the comment in try_to_grab_pending() with the same code */
21605 pwq = get_work_pwq(work);
21607 @@ -2821,10 +2859,11 @@ static bool start_flush_work(struct work_struct *work, struct wq_barrier *barr)
21609 lock_map_acquire_read(&pwq->wq->lockdep_map);
21610 lock_map_release(&pwq->wq->lockdep_map);
21612 + rcu_read_unlock();
21615 spin_unlock_irq(&pool->lock);
21616 + rcu_read_unlock();
21620 @@ -2911,7 +2950,7 @@ static bool __cancel_work_timer(struct work_struct *work, bool is_dwork)
21622 /* tell other tasks trying to grab @work to back off */
21623 mark_work_canceling(work);
21624 - local_irq_restore(flags);
21625 + local_unlock_irqrestore(pendingb_lock, flags);
21628 clear_work_data(work);
21629 @@ -2966,10 +3005,10 @@ EXPORT_SYMBOL_GPL(cancel_work_sync);
21631 bool flush_delayed_work(struct delayed_work *dwork)
21633 - local_irq_disable();
21634 + local_lock_irq(pendingb_lock);
21635 if (del_timer_sync(&dwork->timer))
21636 __queue_work(dwork->cpu, dwork->wq, &dwork->work);
21637 - local_irq_enable();
21638 + local_unlock_irq(pendingb_lock);
21639 return flush_work(&dwork->work);
21641 EXPORT_SYMBOL(flush_delayed_work);
21642 @@ -2987,7 +3026,7 @@ static bool __cancel_work(struct work_struct *work, bool is_dwork)
21645 set_work_pool_and_clear_pending(work, get_work_pool_id(work));
21646 - local_irq_restore(flags);
21647 + local_unlock_irqrestore(pendingb_lock, flags);
21651 @@ -3245,7 +3284,7 @@ static void rcu_free_pool(struct rcu_head *rcu)
21652 * put_unbound_pool - put a worker_pool
21653 * @pool: worker_pool to put
21655 - * Put @pool. If its refcnt reaches zero, it gets destroyed in sched-RCU
21656 + * Put @pool. If its refcnt reaches zero, it gets destroyed in RCU
21657 * safe manner. get_unbound_pool() calls this function on its failure path
21658 * and this function should be able to release pools which went through,
21659 * successfully or not, init_worker_pool().
21660 @@ -3299,8 +3338,8 @@ static void put_unbound_pool(struct worker_pool *pool)
21661 del_timer_sync(&pool->idle_timer);
21662 del_timer_sync(&pool->mayday_timer);
21664 - /* sched-RCU protected to allow dereferences from get_work_pool() */
21665 - call_rcu_sched(&pool->rcu, rcu_free_pool);
21666 + /* RCU protected to allow dereferences from get_work_pool() */
21667 + call_rcu(&pool->rcu, rcu_free_pool);
21671 @@ -3407,14 +3446,14 @@ static void pwq_unbound_release_workfn(struct work_struct *work)
21672 put_unbound_pool(pool);
21673 mutex_unlock(&wq_pool_mutex);
21675 - call_rcu_sched(&pwq->rcu, rcu_free_pwq);
21676 + call_rcu(&pwq->rcu, rcu_free_pwq);
21679 * If we're the last pwq going away, @wq is already dead and no one
21680 * is gonna access it anymore. Schedule RCU free.
21683 - call_rcu_sched(&wq->rcu, rcu_free_wq);
21684 + call_rcu(&wq->rcu, rcu_free_wq);
21688 @@ -4064,7 +4103,7 @@ void destroy_workqueue(struct workqueue_struct *wq)
21689 * The base ref is never dropped on per-cpu pwqs. Directly
21690 * schedule RCU free.
21692 - call_rcu_sched(&wq->rcu, rcu_free_wq);
21693 + call_rcu(&wq->rcu, rcu_free_wq);
21696 * We're the sole accessor of @wq at this point. Directly
21697 @@ -4157,7 +4196,8 @@ bool workqueue_congested(int cpu, struct workqueue_struct *wq)
21698 struct pool_workqueue *pwq;
21701 - rcu_read_lock_sched();
21703 + preempt_disable();
21705 if (cpu == WORK_CPU_UNBOUND)
21706 cpu = smp_processor_id();
21707 @@ -4168,7 +4208,8 @@ bool workqueue_congested(int cpu, struct workqueue_struct *wq)
21708 pwq = unbound_pwq_by_node(wq, cpu_to_node(cpu));
21710 ret = !list_empty(&pwq->delayed_works);
21711 - rcu_read_unlock_sched();
21712 + preempt_enable();
21713 + rcu_read_unlock();
21717 @@ -4194,15 +4235,15 @@ unsigned int work_busy(struct work_struct *work)
21718 if (work_pending(work))
21719 ret |= WORK_BUSY_PENDING;
21721 - local_irq_save(flags);
21723 pool = get_work_pool(work);
21725 - spin_lock(&pool->lock);
21726 + spin_lock_irqsave(&pool->lock, flags);
21727 if (find_worker_executing_work(pool, work))
21728 ret |= WORK_BUSY_RUNNING;
21729 - spin_unlock(&pool->lock);
21730 + spin_unlock_irqrestore(&pool->lock, flags);
21732 - local_irq_restore(flags);
21733 + rcu_read_unlock();
21737 @@ -4391,7 +4432,7 @@ void show_workqueue_state(void)
21738 unsigned long flags;
21741 - rcu_read_lock_sched();
21744 pr_info("Showing busy workqueues and worker pools:\n");
21746 @@ -4444,7 +4485,7 @@ void show_workqueue_state(void)
21747 spin_unlock_irqrestore(&pool->lock, flags);
21750 - rcu_read_unlock_sched();
21751 + rcu_read_unlock();
21755 @@ -4782,16 +4823,16 @@ bool freeze_workqueues_busy(void)
21756 * nr_active is monotonically decreasing. It's safe
21757 * to peek without lock.
21759 - rcu_read_lock_sched();
21761 for_each_pwq(pwq, wq) {
21762 WARN_ON_ONCE(pwq->nr_active < 0);
21763 if (pwq->nr_active) {
21765 - rcu_read_unlock_sched();
21766 + rcu_read_unlock();
21770 - rcu_read_unlock_sched();
21771 + rcu_read_unlock();
21774 mutex_unlock(&wq_pool_mutex);
21775 @@ -4981,7 +5022,8 @@ static ssize_t wq_pool_ids_show(struct device *dev,
21776 const char *delim = "";
21777 int node, written = 0;
21779 - rcu_read_lock_sched();
21780 + get_online_cpus();
21782 for_each_node(node) {
21783 written += scnprintf(buf + written, PAGE_SIZE - written,
21784 "%s%d:%d", delim, node,
21785 @@ -4989,7 +5031,8 @@ static ssize_t wq_pool_ids_show(struct device *dev,
21788 written += scnprintf(buf + written, PAGE_SIZE - written, "\n");
21789 - rcu_read_unlock_sched();
21790 + rcu_read_unlock();
21791 + put_online_cpus();
21795 diff --git a/kernel/workqueue_internal.h b/kernel/workqueue_internal.h
21796 index 8635417c587b..f000c4d6917e 100644
21797 --- a/kernel/workqueue_internal.h
21798 +++ b/kernel/workqueue_internal.h
21799 @@ -43,6 +43,7 @@ struct worker {
21800 unsigned long last_active; /* L: last active timestamp */
21801 unsigned int flags; /* X: flags */
21802 int id; /* I: worker id */
21803 + int sleeping; /* None */
21806 * Opaque string set with work_set_desc(). Printed out with task
21807 @@ -68,7 +69,7 @@ static inline struct worker *current_wq_worker(void)
21808 * Scheduler hooks for concurrency managed workqueue. Only to be used from
21809 * sched/core.c and workqueue.c.
21811 -void wq_worker_waking_up(struct task_struct *task, int cpu);
21812 -struct task_struct *wq_worker_sleeping(struct task_struct *task);
21813 +void wq_worker_running(struct task_struct *task);
21814 +void wq_worker_sleeping(struct task_struct *task);
21816 #endif /* _KERNEL_WORKQUEUE_INTERNAL_H */
21817 diff --git a/lib/Kconfig b/lib/Kconfig
21818 index 260a80e313b9..b06becb3f477 100644
21821 @@ -400,6 +400,7 @@ config CHECK_SIGNATURE
21823 config CPUMASK_OFFSTACK
21824 bool "Force CPU masks off stack" if DEBUG_PER_CPU_MAPS
21825 + depends on !PREEMPT_RT_FULL
21827 Use dynamic allocation for cpumask_var_t, instead of putting
21828 them on the stack. This is a bit more expensive, but avoids
21829 diff --git a/lib/debugobjects.c b/lib/debugobjects.c
21830 index 056052dc8e91..d8494e126de8 100644
21831 --- a/lib/debugobjects.c
21832 +++ b/lib/debugobjects.c
21833 @@ -308,7 +308,10 @@ __debug_object_init(void *addr, struct debug_obj_descr *descr, int onstack)
21834 struct debug_obj *obj;
21835 unsigned long flags;
21838 +#ifdef CONFIG_PREEMPT_RT_FULL
21839 + if (preempt_count() == 0 && !irqs_disabled())
21843 db = get_bucket((unsigned long) addr);
21845 diff --git a/lib/idr.c b/lib/idr.c
21846 index 6098336df267..9decbe914595 100644
21850 #include <linux/idr.h>
21851 #include <linux/spinlock.h>
21852 #include <linux/percpu.h>
21853 +#include <linux/locallock.h>
21855 #define MAX_IDR_SHIFT (sizeof(int) * 8 - 1)
21856 #define MAX_IDR_BIT (1U << MAX_IDR_SHIFT)
21857 @@ -45,6 +46,37 @@ static DEFINE_PER_CPU(struct idr_layer *, idr_preload_head);
21858 static DEFINE_PER_CPU(int, idr_preload_cnt);
21859 static DEFINE_SPINLOCK(simple_ida_lock);
21861 +#ifdef CONFIG_PREEMPT_RT_FULL
21862 +static DEFINE_LOCAL_IRQ_LOCK(idr_lock);
21864 +static inline void idr_preload_lock(void)
21866 + local_lock(idr_lock);
21869 +static inline void idr_preload_unlock(void)
21871 + local_unlock(idr_lock);
21874 +void idr_preload_end(void)
21876 + idr_preload_unlock();
21878 +EXPORT_SYMBOL(idr_preload_end);
21880 +static inline void idr_preload_lock(void)
21882 + preempt_disable();
21885 +static inline void idr_preload_unlock(void)
21887 + preempt_enable();
21892 /* the maximum ID which can be allocated given idr->layers */
21893 static int idr_max(int layers)
21895 @@ -115,14 +147,14 @@ static struct idr_layer *idr_layer_alloc(gfp_t gfp_mask, struct idr *layer_idr)
21896 * context. See idr_preload() for details.
21898 if (!in_interrupt()) {
21899 - preempt_disable();
21900 + idr_preload_lock();
21901 new = __this_cpu_read(idr_preload_head);
21903 __this_cpu_write(idr_preload_head, new->ary[0]);
21904 __this_cpu_dec(idr_preload_cnt);
21905 new->ary[0] = NULL;
21907 - preempt_enable();
21908 + idr_preload_unlock();
21912 @@ -366,7 +398,6 @@ static void idr_fill_slot(struct idr *idr, void *ptr, int id,
21913 idr_mark_full(pa, id);
21918 * idr_preload - preload for idr_alloc()
21919 * @gfp_mask: allocation mask to use for preloading
21920 @@ -401,7 +432,7 @@ void idr_preload(gfp_t gfp_mask)
21921 WARN_ON_ONCE(in_interrupt());
21922 might_sleep_if(gfpflags_allow_blocking(gfp_mask));
21924 - preempt_disable();
21925 + idr_preload_lock();
21928 * idr_alloc() is likely to succeed w/o full idr_layer buffer and
21929 @@ -413,9 +444,9 @@ void idr_preload(gfp_t gfp_mask)
21930 while (__this_cpu_read(idr_preload_cnt) < MAX_IDR_FREE) {
21931 struct idr_layer *new;
21933 - preempt_enable();
21934 + idr_preload_unlock();
21935 new = kmem_cache_zalloc(idr_layer_cache, gfp_mask);
21936 - preempt_disable();
21937 + idr_preload_lock();
21941 diff --git a/lib/irq_poll.c b/lib/irq_poll.c
21942 index 1d6565e81030..b23a79761df7 100644
21943 --- a/lib/irq_poll.c
21944 +++ b/lib/irq_poll.c
21945 @@ -36,6 +36,7 @@ void irq_poll_sched(struct irq_poll *iop)
21946 list_add_tail(&iop->list, this_cpu_ptr(&blk_cpu_iopoll));
21947 __raise_softirq_irqoff(IRQ_POLL_SOFTIRQ);
21948 local_irq_restore(flags);
21949 + preempt_check_resched_rt();
21951 EXPORT_SYMBOL(irq_poll_sched);
21953 @@ -71,6 +72,7 @@ void irq_poll_complete(struct irq_poll *iop)
21954 local_irq_save(flags);
21955 __irq_poll_complete(iop);
21956 local_irq_restore(flags);
21957 + preempt_check_resched_rt();
21959 EXPORT_SYMBOL(irq_poll_complete);
21961 @@ -95,6 +97,7 @@ static void __latent_entropy irq_poll_softirq(struct softirq_action *h)
21964 local_irq_enable();
21965 + preempt_check_resched_rt();
21967 /* Even though interrupts have been re-enabled, this
21968 * access is safe because interrupts can only add new
21969 @@ -132,6 +135,7 @@ static void __latent_entropy irq_poll_softirq(struct softirq_action *h)
21970 __raise_softirq_irqoff(IRQ_POLL_SOFTIRQ);
21972 local_irq_enable();
21973 + preempt_check_resched_rt();
21977 @@ -195,6 +199,7 @@ static int irq_poll_cpu_dead(unsigned int cpu)
21978 this_cpu_ptr(&blk_cpu_iopoll));
21979 __raise_softirq_irqoff(IRQ_POLL_SOFTIRQ);
21980 local_irq_enable();
21981 + preempt_check_resched_rt();
21985 diff --git a/lib/locking-selftest.c b/lib/locking-selftest.c
21986 index f3a217ea0388..4611b156ef79 100644
21987 --- a/lib/locking-selftest.c
21988 +++ b/lib/locking-selftest.c
21989 @@ -590,6 +590,8 @@ GENERATE_TESTCASE(init_held_rsem)
21990 #include "locking-selftest-spin-hardirq.h"
21991 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_spin)
21993 +#ifndef CONFIG_PREEMPT_RT_FULL
21995 #include "locking-selftest-rlock-hardirq.h"
21996 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_hard_rlock)
21998 @@ -605,9 +607,12 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_rlock)
21999 #include "locking-selftest-wlock-softirq.h"
22000 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe1_soft_wlock)
22007 +#ifndef CONFIG_PREEMPT_RT_FULL
22009 * Enabling hardirqs with a softirq-safe lock held:
22011 @@ -640,6 +645,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2A_rlock)
22018 * Enabling irqs with an irq-safe lock held:
22020 @@ -663,6 +670,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2A_rlock)
22021 #include "locking-selftest-spin-hardirq.h"
22022 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_spin)
22024 +#ifndef CONFIG_PREEMPT_RT_FULL
22026 #include "locking-selftest-rlock-hardirq.h"
22027 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_hard_rlock)
22029 @@ -678,6 +687,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_rlock)
22030 #include "locking-selftest-wlock-softirq.h"
22031 GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_wlock)
22038 @@ -709,6 +720,8 @@ GENERATE_PERMUTATIONS_2_EVENTS(irqsafe2B_soft_wlock)
22039 #include "locking-selftest-spin-hardirq.h"
22040 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_spin)
22042 +#ifndef CONFIG_PREEMPT_RT_FULL
22044 #include "locking-selftest-rlock-hardirq.h"
22045 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_hard_rlock)
22047 @@ -724,6 +737,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_rlock)
22048 #include "locking-selftest-wlock-softirq.h"
22049 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_wlock)
22056 @@ -757,6 +772,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe3_soft_wlock)
22057 #include "locking-selftest-spin-hardirq.h"
22058 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_spin)
22060 +#ifndef CONFIG_PREEMPT_RT_FULL
22062 #include "locking-selftest-rlock-hardirq.h"
22063 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_hard_rlock)
22065 @@ -772,10 +789,14 @@ GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_rlock)
22066 #include "locking-selftest-wlock-softirq.h"
22067 GENERATE_PERMUTATIONS_3_EVENTS(irqsafe4_soft_wlock)
22075 +#ifndef CONFIG_PREEMPT_RT_FULL
22078 * read-lock / write-lock irq inversion.
22080 @@ -838,6 +859,10 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_inversion_soft_wlock)
22086 +#ifndef CONFIG_PREEMPT_RT_FULL
22089 * read-lock / write-lock recursion that is actually safe.
22091 @@ -876,6 +901,8 @@ GENERATE_PERMUTATIONS_3_EVENTS(irq_read_recursion_soft)
22098 * read-lock / write-lock recursion that is unsafe.
22100 @@ -1858,6 +1885,7 @@ void locking_selftest(void)
22102 printk(" --------------------------------------------------------------------------\n");
22104 +#ifndef CONFIG_PREEMPT_RT_FULL
22106 * irq-context testcases:
22108 @@ -1870,6 +1898,28 @@ void locking_selftest(void)
22110 DO_TESTCASE_6x2("irq read-recursion", irq_read_recursion);
22111 // DO_TESTCASE_6x2B("irq read-recursion #2", irq_read_recursion2);
22113 + /* On -rt, we only do hardirq context test for raw spinlock */
22114 + DO_TESTCASE_1B("hard-irqs-on + irq-safe-A", irqsafe1_hard_spin, 12);
22115 + DO_TESTCASE_1B("hard-irqs-on + irq-safe-A", irqsafe1_hard_spin, 21);
22117 + DO_TESTCASE_1B("hard-safe-A + irqs-on", irqsafe2B_hard_spin, 12);
22118 + DO_TESTCASE_1B("hard-safe-A + irqs-on", irqsafe2B_hard_spin, 21);
22120 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 123);
22121 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 132);
22122 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 213);
22123 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 231);
22124 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 312);
22125 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #1", irqsafe3_hard_spin, 321);
22127 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 123);
22128 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 132);
22129 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 213);
22130 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 231);
22131 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 312);
22132 + DO_TESTCASE_1B("hard-safe-A + unsafe-B #2", irqsafe4_hard_spin, 321);
22137 diff --git a/lib/percpu_ida.c b/lib/percpu_ida.c
22138 index 6d40944960de..822a2c027e72 100644
22139 --- a/lib/percpu_ida.c
22140 +++ b/lib/percpu_ida.c
22142 #include <linux/string.h>
22143 #include <linux/spinlock.h>
22144 #include <linux/percpu_ida.h>
22145 +#include <linux/locallock.h>
22147 +static DEFINE_LOCAL_IRQ_LOCK(irq_off_lock);
22149 struct percpu_ida_cpu {
22151 @@ -148,13 +151,13 @@ int percpu_ida_alloc(struct percpu_ida *pool, int state)
22152 unsigned long flags;
22155 - local_irq_save(flags);
22156 + local_lock_irqsave(irq_off_lock, flags);
22157 tags = this_cpu_ptr(pool->tag_cpu);
22160 tag = alloc_local_tag(tags);
22161 if (likely(tag >= 0)) {
22162 - local_irq_restore(flags);
22163 + local_unlock_irqrestore(irq_off_lock, flags);
22167 @@ -173,6 +176,7 @@ int percpu_ida_alloc(struct percpu_ida *pool, int state)
22169 if (!tags->nr_free)
22170 alloc_global_tags(pool, tags);
22172 if (!tags->nr_free)
22173 steal_tags(pool, tags);
22175 @@ -184,7 +188,7 @@ int percpu_ida_alloc(struct percpu_ida *pool, int state)
22178 spin_unlock(&pool->lock);
22179 - local_irq_restore(flags);
22180 + local_unlock_irqrestore(irq_off_lock, flags);
22182 if (tag >= 0 || state == TASK_RUNNING)
22184 @@ -196,7 +200,7 @@ int percpu_ida_alloc(struct percpu_ida *pool, int state)
22188 - local_irq_save(flags);
22189 + local_lock_irqsave(irq_off_lock, flags);
22190 tags = this_cpu_ptr(pool->tag_cpu);
22192 if (state != TASK_RUNNING)
22193 @@ -221,7 +225,7 @@ void percpu_ida_free(struct percpu_ida *pool, unsigned tag)
22195 BUG_ON(tag >= pool->nr_tags);
22197 - local_irq_save(flags);
22198 + local_lock_irqsave(irq_off_lock, flags);
22199 tags = this_cpu_ptr(pool->tag_cpu);
22201 spin_lock(&tags->lock);
22202 @@ -253,7 +257,7 @@ void percpu_ida_free(struct percpu_ida *pool, unsigned tag)
22203 spin_unlock(&pool->lock);
22206 - local_irq_restore(flags);
22207 + local_unlock_irqrestore(irq_off_lock, flags);
22209 EXPORT_SYMBOL_GPL(percpu_ida_free);
22211 @@ -345,7 +349,7 @@ int percpu_ida_for_each_free(struct percpu_ida *pool, percpu_ida_cb fn,
22212 struct percpu_ida_cpu *remote;
22213 unsigned cpu, i, err = 0;
22215 - local_irq_save(flags);
22216 + local_lock_irqsave(irq_off_lock, flags);
22217 for_each_possible_cpu(cpu) {
22218 remote = per_cpu_ptr(pool->tag_cpu, cpu);
22219 spin_lock(&remote->lock);
22220 @@ -367,7 +371,7 @@ int percpu_ida_for_each_free(struct percpu_ida *pool, percpu_ida_cb fn,
22222 spin_unlock(&pool->lock);
22224 - local_irq_restore(flags);
22225 + local_unlock_irqrestore(irq_off_lock, flags);
22228 EXPORT_SYMBOL_GPL(percpu_ida_for_each_free);
22229 diff --git a/lib/radix-tree.c b/lib/radix-tree.c
22230 index 8e6d552c40dd..741da5a77fd5 100644
22231 --- a/lib/radix-tree.c
22232 +++ b/lib/radix-tree.c
22234 #include <linux/bitops.h>
22235 #include <linux/rcupdate.h>
22236 #include <linux/preempt.h> /* in_interrupt() */
22238 +#include <linux/locallock.h>
22240 /* Number of nodes in fully populated tree of given height */
22241 static unsigned long height_to_maxnodes[RADIX_TREE_MAX_PATH + 1] __read_mostly;
22242 @@ -68,6 +68,7 @@ struct radix_tree_preload {
22243 struct radix_tree_node *nodes;
22245 static DEFINE_PER_CPU(struct radix_tree_preload, radix_tree_preloads) = { 0, };
22246 +static DEFINE_LOCAL_IRQ_LOCK(radix_tree_preloads_lock);
22248 static inline void *node_to_entry(void *ptr)
22250 @@ -290,13 +291,14 @@ radix_tree_node_alloc(struct radix_tree_root *root)
22251 * succeed in getting a node here (and never reach
22252 * kmem_cache_alloc)
22254 - rtp = this_cpu_ptr(&radix_tree_preloads);
22255 + rtp = &get_locked_var(radix_tree_preloads_lock, radix_tree_preloads);
22258 rtp->nodes = ret->private_data;
22259 ret->private_data = NULL;
22262 + put_locked_var(radix_tree_preloads_lock, radix_tree_preloads);
22264 * Update the allocation stack trace as this is more useful
22266 @@ -357,14 +359,14 @@ static int __radix_tree_preload(gfp_t gfp_mask, int nr)
22268 gfp_mask &= ~__GFP_ACCOUNT;
22270 - preempt_disable();
22271 + local_lock(radix_tree_preloads_lock);
22272 rtp = this_cpu_ptr(&radix_tree_preloads);
22273 while (rtp->nr < nr) {
22274 - preempt_enable();
22275 + local_unlock(radix_tree_preloads_lock);
22276 node = kmem_cache_alloc(radix_tree_node_cachep, gfp_mask);
22279 - preempt_disable();
22280 + local_lock(radix_tree_preloads_lock);
22281 rtp = this_cpu_ptr(&radix_tree_preloads);
22282 if (rtp->nr < nr) {
22283 node->private_data = rtp->nodes;
22284 @@ -406,7 +408,7 @@ int radix_tree_maybe_preload(gfp_t gfp_mask)
22285 if (gfpflags_allow_blocking(gfp_mask))
22286 return __radix_tree_preload(gfp_mask, RADIX_TREE_PRELOAD_SIZE);
22287 /* Preloading doesn't help anything with this gfp mask, skip it */
22288 - preempt_disable();
22289 + local_lock(radix_tree_preloads_lock);
22292 EXPORT_SYMBOL(radix_tree_maybe_preload);
22293 @@ -422,7 +424,7 @@ int radix_tree_maybe_preload_order(gfp_t gfp_mask, int order)
22295 /* Preloading doesn't help anything with this gfp mask, skip it */
22296 if (!gfpflags_allow_blocking(gfp_mask)) {
22297 - preempt_disable();
22298 + local_lock(radix_tree_preloads_lock);
22302 @@ -456,6 +458,12 @@ int radix_tree_maybe_preload_order(gfp_t gfp_mask, int order)
22303 return __radix_tree_preload(gfp_mask, nr_nodes);
22306 +void radix_tree_preload_end(void)
22308 + local_unlock(radix_tree_preloads_lock);
22310 +EXPORT_SYMBOL(radix_tree_preload_end);
22313 * The maximum index which can be stored in a radix tree
22315 diff --git a/lib/scatterlist.c b/lib/scatterlist.c
22316 index 004fc70fc56a..ccc46992a517 100644
22317 --- a/lib/scatterlist.c
22318 +++ b/lib/scatterlist.c
22319 @@ -620,7 +620,7 @@ void sg_miter_stop(struct sg_mapping_iter *miter)
22320 flush_kernel_dcache_page(miter->page);
22322 if (miter->__flags & SG_MITER_ATOMIC) {
22323 - WARN_ON_ONCE(preemptible());
22324 + WARN_ON_ONCE(!pagefault_disabled());
22325 kunmap_atomic(miter->addr);
22327 kunmap(miter->page);
22328 @@ -664,7 +664,7 @@ size_t sg_copy_buffer(struct scatterlist *sgl, unsigned int nents, void *buf,
22329 if (!sg_miter_skip(&miter, skip))
22332 - local_irq_save(flags);
22333 + local_irq_save_nort(flags);
22335 while (sg_miter_next(&miter) && offset < buflen) {
22337 @@ -681,7 +681,7 @@ size_t sg_copy_buffer(struct scatterlist *sgl, unsigned int nents, void *buf,
22339 sg_miter_stop(&miter);
22341 - local_irq_restore(flags);
22342 + local_irq_restore_nort(flags);
22345 EXPORT_SYMBOL(sg_copy_buffer);
22346 diff --git a/lib/smp_processor_id.c b/lib/smp_processor_id.c
22347 index 1afec32de6f2..11fa431046a8 100644
22348 --- a/lib/smp_processor_id.c
22349 +++ b/lib/smp_processor_id.c
22350 @@ -39,8 +39,9 @@ notrace static unsigned int check_preemption_disabled(const char *what1,
22351 if (!printk_ratelimit())
22354 - printk(KERN_ERR "BUG: using %s%s() in preemptible [%08x] code: %s/%d\n",
22355 - what1, what2, preempt_count() - 1, current->comm, current->pid);
22356 + printk(KERN_ERR "BUG: using %s%s() in preemptible [%08x %08x] code: %s/%d\n",
22357 + what1, what2, preempt_count() - 1, __migrate_disabled(current),
22358 + current->comm, current->pid);
22360 print_symbol("caller is %s\n", (long)__builtin_return_address(0));
22362 diff --git a/localversion-rt b/localversion-rt
22363 new file mode 100644
22364 index 000000000000..6e44e540b927
22366 +++ b/localversion-rt
22369 diff --git a/mm/Kconfig b/mm/Kconfig
22370 index 86e3e0e74d20..77e5862a1ed2 100644
22373 @@ -410,7 +410,7 @@ config NOMMU_INITIAL_TRIM_EXCESS
22375 config TRANSPARENT_HUGEPAGE
22376 bool "Transparent Hugepage Support"
22377 - depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE
22378 + depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE && !PREEMPT_RT_FULL
22380 select RADIX_TREE_MULTIORDER
22382 diff --git a/mm/backing-dev.c b/mm/backing-dev.c
22383 index 6ff2d7744223..b5a91dd53b5f 100644
22384 --- a/mm/backing-dev.c
22385 +++ b/mm/backing-dev.c
22386 @@ -457,9 +457,9 @@ void wb_congested_put(struct bdi_writeback_congested *congested)
22388 unsigned long flags;
22390 - local_irq_save(flags);
22391 + local_irq_save_nort(flags);
22392 if (!atomic_dec_and_lock(&congested->refcnt, &cgwb_lock)) {
22393 - local_irq_restore(flags);
22394 + local_irq_restore_nort(flags);
22398 diff --git a/mm/compaction.c b/mm/compaction.c
22399 index 70e6bec46dc2..6678ed58b7c6 100644
22400 --- a/mm/compaction.c
22401 +++ b/mm/compaction.c
22402 @@ -1593,10 +1593,12 @@ static enum compact_result compact_zone(struct zone *zone, struct compact_contro
22403 block_start_pfn(cc->migrate_pfn, cc->order);
22405 if (cc->last_migrated_pfn < current_block_start) {
22407 + cpu = get_cpu_light();
22408 + local_lock_irq(swapvec_lock);
22409 lru_add_drain_cpu(cpu);
22410 + local_unlock_irq(swapvec_lock);
22411 drain_local_pages(zone);
22414 /* No more flushing until we migrate again */
22415 cc->last_migrated_pfn = 0;
22417 diff --git a/mm/filemap.c b/mm/filemap.c
22418 index d8d7df82c69a..0eac87a07892 100644
22421 @@ -159,9 +159,12 @@ static int page_cache_tree_insert(struct address_space *mapping,
22422 * node->private_list is protected by
22423 * mapping->tree_lock.
22425 - if (!list_empty(&node->private_list))
22426 - list_lru_del(&workingset_shadow_nodes,
22427 + if (!list_empty(&node->private_list)) {
22428 + local_lock(workingset_shadow_lock);
22429 + list_lru_del(&__workingset_shadow_nodes,
22430 &node->private_list);
22431 + local_unlock(workingset_shadow_lock);
22436 @@ -217,8 +220,10 @@ static void page_cache_tree_delete(struct address_space *mapping,
22437 if (!dax_mapping(mapping) && !workingset_node_pages(node) &&
22438 list_empty(&node->private_list)) {
22439 node->private_data = mapping;
22440 - list_lru_add(&workingset_shadow_nodes,
22441 - &node->private_list);
22442 + local_lock(workingset_shadow_lock);
22443 + list_lru_add(&__workingset_shadow_nodes,
22444 + &node->private_list);
22445 + local_unlock(workingset_shadow_lock);
22449 diff --git a/mm/highmem.c b/mm/highmem.c
22450 index 50b4ca6787f0..77518a3b35a1 100644
22453 @@ -29,10 +29,11 @@
22454 #include <linux/kgdb.h>
22455 #include <asm/tlbflush.h>
22458 +#ifndef CONFIG_PREEMPT_RT_FULL
22459 #if defined(CONFIG_HIGHMEM) || defined(CONFIG_X86_32)
22460 DEFINE_PER_CPU(int, __kmap_atomic_idx);
22465 * Virtual_count is not a pure "count".
22466 @@ -107,8 +108,9 @@ static inline wait_queue_head_t *get_pkmap_wait_queue_head(unsigned int color)
22467 unsigned long totalhigh_pages __read_mostly;
22468 EXPORT_SYMBOL(totalhigh_pages);
22471 +#ifndef CONFIG_PREEMPT_RT_FULL
22472 EXPORT_PER_CPU_SYMBOL(__kmap_atomic_idx);
22475 unsigned int nr_free_highpages (void)
22477 diff --git a/mm/memcontrol.c b/mm/memcontrol.c
22478 index 4c6ade54d833..ba29283aa43d 100644
22479 --- a/mm/memcontrol.c
22480 +++ b/mm/memcontrol.c
22482 #include <net/sock.h>
22483 #include <net/ip.h>
22485 +#include <linux/locallock.h>
22487 #include <asm/uaccess.h>
22489 @@ -92,6 +93,8 @@ int do_swap_account __read_mostly;
22490 #define do_swap_account 0
22493 +static DEFINE_LOCAL_IRQ_LOCK(event_lock);
22495 /* Whether legacy memory+swap accounting is active */
22496 static bool do_memsw_account(void)
22498 @@ -1692,6 +1695,7 @@ struct memcg_stock_pcp {
22499 #define FLUSHING_CACHED_CHARGE 0
22501 static DEFINE_PER_CPU(struct memcg_stock_pcp, memcg_stock);
22502 +static DEFINE_LOCAL_IRQ_LOCK(memcg_stock_ll);
22503 static DEFINE_MUTEX(percpu_charge_mutex);
22506 @@ -1714,7 +1718,7 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
22507 if (nr_pages > CHARGE_BATCH)
22510 - local_irq_save(flags);
22511 + local_lock_irqsave(memcg_stock_ll, flags);
22513 stock = this_cpu_ptr(&memcg_stock);
22514 if (memcg == stock->cached && stock->nr_pages >= nr_pages) {
22515 @@ -1722,7 +1726,7 @@ static bool consume_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
22519 - local_irq_restore(flags);
22520 + local_unlock_irqrestore(memcg_stock_ll, flags);
22524 @@ -1749,13 +1753,13 @@ static void drain_local_stock(struct work_struct *dummy)
22525 struct memcg_stock_pcp *stock;
22526 unsigned long flags;
22528 - local_irq_save(flags);
22529 + local_lock_irqsave(memcg_stock_ll, flags);
22531 stock = this_cpu_ptr(&memcg_stock);
22532 drain_stock(stock);
22533 clear_bit(FLUSHING_CACHED_CHARGE, &stock->flags);
22535 - local_irq_restore(flags);
22536 + local_unlock_irqrestore(memcg_stock_ll, flags);
22540 @@ -1767,7 +1771,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
22541 struct memcg_stock_pcp *stock;
22542 unsigned long flags;
22544 - local_irq_save(flags);
22545 + local_lock_irqsave(memcg_stock_ll, flags);
22547 stock = this_cpu_ptr(&memcg_stock);
22548 if (stock->cached != memcg) { /* reset if necessary */
22549 @@ -1776,7 +1780,7 @@ static void refill_stock(struct mem_cgroup *memcg, unsigned int nr_pages)
22551 stock->nr_pages += nr_pages;
22553 - local_irq_restore(flags);
22554 + local_unlock_irqrestore(memcg_stock_ll, flags);
22558 @@ -1792,7 +1796,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg)
22560 /* Notify other cpus that system-wide "drain" is running */
22562 - curcpu = get_cpu();
22563 + curcpu = get_cpu_light();
22564 for_each_online_cpu(cpu) {
22565 struct memcg_stock_pcp *stock = &per_cpu(memcg_stock, cpu);
22566 struct mem_cgroup *memcg;
22567 @@ -1809,7 +1813,7 @@ static void drain_all_stock(struct mem_cgroup *root_memcg)
22568 schedule_work_on(cpu, &stock->work);
22574 mutex_unlock(&percpu_charge_mutex);
22576 @@ -4548,12 +4552,12 @@ static int mem_cgroup_move_account(struct page *page,
22580 - local_irq_disable();
22581 + local_lock_irq(event_lock);
22582 mem_cgroup_charge_statistics(to, page, compound, nr_pages);
22583 memcg_check_events(to, page);
22584 mem_cgroup_charge_statistics(from, page, compound, -nr_pages);
22585 memcg_check_events(from, page);
22586 - local_irq_enable();
22587 + local_unlock_irq(event_lock);
22591 @@ -5428,10 +5432,10 @@ void mem_cgroup_commit_charge(struct page *page, struct mem_cgroup *memcg,
22593 commit_charge(page, memcg, lrucare);
22595 - local_irq_disable();
22596 + local_lock_irq(event_lock);
22597 mem_cgroup_charge_statistics(memcg, page, compound, nr_pages);
22598 memcg_check_events(memcg, page);
22599 - local_irq_enable();
22600 + local_unlock_irq(event_lock);
22602 if (do_memsw_account() && PageSwapCache(page)) {
22603 swp_entry_t entry = { .val = page_private(page) };
22604 @@ -5487,14 +5491,14 @@ static void uncharge_batch(struct mem_cgroup *memcg, unsigned long pgpgout,
22605 memcg_oom_recover(memcg);
22608 - local_irq_save(flags);
22609 + local_lock_irqsave(event_lock, flags);
22610 __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_RSS], nr_anon);
22611 __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_CACHE], nr_file);
22612 __this_cpu_sub(memcg->stat->count[MEM_CGROUP_STAT_RSS_HUGE], nr_huge);
22613 __this_cpu_add(memcg->stat->events[MEM_CGROUP_EVENTS_PGPGOUT], pgpgout);
22614 __this_cpu_add(memcg->stat->nr_page_events, nr_pages);
22615 memcg_check_events(memcg, dummy_page);
22616 - local_irq_restore(flags);
22617 + local_unlock_irqrestore(event_lock, flags);
22619 if (!mem_cgroup_is_root(memcg))
22620 css_put_many(&memcg->css, nr_pages);
22621 @@ -5649,10 +5653,10 @@ void mem_cgroup_migrate(struct page *oldpage, struct page *newpage)
22623 commit_charge(newpage, memcg, false);
22625 - local_irq_save(flags);
22626 + local_lock_irqsave(event_lock, flags);
22627 mem_cgroup_charge_statistics(memcg, newpage, compound, nr_pages);
22628 memcg_check_events(memcg, newpage);
22629 - local_irq_restore(flags);
22630 + local_unlock_irqrestore(event_lock, flags);
22633 DEFINE_STATIC_KEY_FALSE(memcg_sockets_enabled_key);
22634 @@ -5832,6 +5836,7 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry)
22636 struct mem_cgroup *memcg, *swap_memcg;
22637 unsigned short oldid;
22638 + unsigned long flags;
22640 VM_BUG_ON_PAGE(PageLRU(page), page);
22641 VM_BUG_ON_PAGE(page_count(page), page);
22642 @@ -5872,12 +5877,16 @@ void mem_cgroup_swapout(struct page *page, swp_entry_t entry)
22643 * important here to have the interrupts disabled because it is the
22644 * only synchronisation we have for udpating the per-CPU variables.
22646 + local_lock_irqsave(event_lock, flags);
22647 +#ifndef CONFIG_PREEMPT_RT_BASE
22648 VM_BUG_ON(!irqs_disabled());
22650 mem_cgroup_charge_statistics(memcg, page, false, -1);
22651 memcg_check_events(memcg, page);
22653 if (!mem_cgroup_is_root(memcg))
22654 css_put(&memcg->css);
22655 + local_unlock_irqrestore(event_lock, flags);
22659 diff --git a/mm/mmu_context.c b/mm/mmu_context.c
22660 index 6f4d27c5bb32..5cd25c745a8f 100644
22661 --- a/mm/mmu_context.c
22662 +++ b/mm/mmu_context.c
22663 @@ -23,6 +23,7 @@ void use_mm(struct mm_struct *mm)
22664 struct task_struct *tsk = current;
22667 + preempt_disable_rt();
22668 active_mm = tsk->active_mm;
22669 if (active_mm != mm) {
22670 atomic_inc(&mm->mm_count);
22671 @@ -30,6 +31,7 @@ void use_mm(struct mm_struct *mm)
22674 switch_mm(active_mm, mm, tsk);
22675 + preempt_enable_rt();
22677 #ifdef finish_arch_post_lock_switch
22678 finish_arch_post_lock_switch();
22679 diff --git a/mm/page_alloc.c b/mm/page_alloc.c
22680 index f4a02e240fb6..2e73f8cfde74 100644
22681 --- a/mm/page_alloc.c
22682 +++ b/mm/page_alloc.c
22684 #include <linux/page_ext.h>
22685 #include <linux/hugetlb.h>
22686 #include <linux/sched/rt.h>
22687 +#include <linux/locallock.h>
22688 #include <linux/page_owner.h>
22689 #include <linux/kthread.h>
22690 #include <linux/memcontrol.h>
22691 @@ -281,6 +282,18 @@ EXPORT_SYMBOL(nr_node_ids);
22692 EXPORT_SYMBOL(nr_online_nodes);
22695 +static DEFINE_LOCAL_IRQ_LOCK(pa_lock);
22697 +#ifdef CONFIG_PREEMPT_RT_BASE
22698 +# define cpu_lock_irqsave(cpu, flags) \
22699 + local_lock_irqsave_on(pa_lock, flags, cpu)
22700 +# define cpu_unlock_irqrestore(cpu, flags) \
22701 + local_unlock_irqrestore_on(pa_lock, flags, cpu)
22703 +# define cpu_lock_irqsave(cpu, flags) local_irq_save(flags)
22704 +# define cpu_unlock_irqrestore(cpu, flags) local_irq_restore(flags)
22707 int page_group_by_mobility_disabled __read_mostly;
22709 #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
22710 @@ -1072,7 +1085,7 @@ static bool bulkfree_pcp_prepare(struct page *page)
22711 #endif /* CONFIG_DEBUG_VM */
22714 - * Frees a number of pages from the PCP lists
22715 + * Frees a number of pages which have been collected from the pcp lists.
22716 * Assumes all pages on list are in same zone, and of same order.
22717 * count is the number of pages to free.
22719 @@ -1083,19 +1096,58 @@ static bool bulkfree_pcp_prepare(struct page *page)
22720 * pinned" detection logic.
22722 static void free_pcppages_bulk(struct zone *zone, int count,
22723 - struct per_cpu_pages *pcp)
22724 + struct list_head *list)
22726 - int migratetype = 0;
22727 - int batch_free = 0;
22728 unsigned long nr_scanned;
22729 bool isolated_pageblocks;
22730 + unsigned long flags;
22732 + spin_lock_irqsave(&zone->lock, flags);
22734 - spin_lock(&zone->lock);
22735 isolated_pageblocks = has_isolate_pageblock(zone);
22736 nr_scanned = node_page_state(zone->zone_pgdat, NR_PAGES_SCANNED);
22738 __mod_node_page_state(zone->zone_pgdat, NR_PAGES_SCANNED, -nr_scanned);
22740 + while (!list_empty(list)) {
22741 + struct page *page;
22742 + int mt; /* migratetype of the to-be-freed page */
22744 + page = list_first_entry(list, struct page, lru);
22745 + /* must delete as __free_one_page list manipulates */
22746 + list_del(&page->lru);
22748 + mt = get_pcppage_migratetype(page);
22749 + /* MIGRATE_ISOLATE page should not go to pcplists */
22750 + VM_BUG_ON_PAGE(is_migrate_isolate(mt), page);
22751 + /* Pageblock could have been isolated meanwhile */
22752 + if (unlikely(isolated_pageblocks))
22753 + mt = get_pageblock_migratetype(page);
22755 + if (bulkfree_pcp_prepare(page))
22758 + __free_one_page(page, page_to_pfn(page), zone, 0, mt);
22759 + trace_mm_page_pcpu_drain(page, 0, mt);
22762 + WARN_ON(count != 0);
22763 + spin_unlock_irqrestore(&zone->lock, flags);
22767 + * Moves a number of pages from the PCP lists to free list which
22768 + * is freed outside of the locked region.
22770 + * Assumes all pages on list are in same zone, and of same order.
22771 + * count is the number of pages to free.
22773 +static void isolate_pcp_pages(int count, struct per_cpu_pages *src,
22774 + struct list_head *dst)
22776 + int migratetype = 0;
22777 + int batch_free = 0;
22781 struct list_head *list;
22782 @@ -1111,7 +1163,7 @@ static void free_pcppages_bulk(struct zone *zone, int count,
22784 if (++migratetype == MIGRATE_PCPTYPES)
22786 - list = &pcp->lists[migratetype];
22787 + list = &src->lists[migratetype];
22788 } while (list_empty(list));
22790 /* This is the only non-empty list. Free them all. */
22791 @@ -1119,27 +1171,12 @@ static void free_pcppages_bulk(struct zone *zone, int count,
22792 batch_free = count;
22795 - int mt; /* migratetype of the to-be-freed page */
22797 page = list_last_entry(list, struct page, lru);
22798 - /* must delete as __free_one_page list manipulates */
22799 list_del(&page->lru);
22801 - mt = get_pcppage_migratetype(page);
22802 - /* MIGRATE_ISOLATE page should not go to pcplists */
22803 - VM_BUG_ON_PAGE(is_migrate_isolate(mt), page);
22804 - /* Pageblock could have been isolated meanwhile */
22805 - if (unlikely(isolated_pageblocks))
22806 - mt = get_pageblock_migratetype(page);
22808 - if (bulkfree_pcp_prepare(page))
22811 - __free_one_page(page, page_to_pfn(page), zone, 0, mt);
22812 - trace_mm_page_pcpu_drain(page, 0, mt);
22813 + list_add(&page->lru, dst);
22814 } while (--count && --batch_free && !list_empty(list));
22816 - spin_unlock(&zone->lock);
22819 static void free_one_page(struct zone *zone,
22820 @@ -1148,7 +1185,9 @@ static void free_one_page(struct zone *zone,
22823 unsigned long nr_scanned;
22824 - spin_lock(&zone->lock);
22825 + unsigned long flags;
22827 + spin_lock_irqsave(&zone->lock, flags);
22828 nr_scanned = node_page_state(zone->zone_pgdat, NR_PAGES_SCANNED);
22830 __mod_node_page_state(zone->zone_pgdat, NR_PAGES_SCANNED, -nr_scanned);
22831 @@ -1158,7 +1197,7 @@ static void free_one_page(struct zone *zone,
22832 migratetype = get_pfnblock_migratetype(page, pfn);
22834 __free_one_page(page, pfn, zone, order, migratetype);
22835 - spin_unlock(&zone->lock);
22836 + spin_unlock_irqrestore(&zone->lock, flags);
22839 static void __meminit __init_single_page(struct page *page, unsigned long pfn,
22840 @@ -1244,10 +1283,10 @@ static void __free_pages_ok(struct page *page, unsigned int order)
22843 migratetype = get_pfnblock_migratetype(page, pfn);
22844 - local_irq_save(flags);
22845 + local_lock_irqsave(pa_lock, flags);
22846 __count_vm_events(PGFREE, 1 << order);
22847 free_one_page(page_zone(page), page, pfn, order, migratetype);
22848 - local_irq_restore(flags);
22849 + local_unlock_irqrestore(pa_lock, flags);
22852 static void __init __free_pages_boot_core(struct page *page, unsigned int order)
22853 @@ -2246,16 +2285,18 @@ static int rmqueue_bulk(struct zone *zone, unsigned int order,
22854 void drain_zone_pages(struct zone *zone, struct per_cpu_pages *pcp)
22856 unsigned long flags;
22858 int to_drain, batch;
22860 - local_irq_save(flags);
22861 + local_lock_irqsave(pa_lock, flags);
22862 batch = READ_ONCE(pcp->batch);
22863 to_drain = min(pcp->count, batch);
22864 if (to_drain > 0) {
22865 - free_pcppages_bulk(zone, to_drain, pcp);
22866 + isolate_pcp_pages(to_drain, pcp, &dst);
22867 pcp->count -= to_drain;
22869 - local_irq_restore(flags);
22870 + local_unlock_irqrestore(pa_lock, flags);
22871 + free_pcppages_bulk(zone, to_drain, &dst);
22875 @@ -2271,16 +2312,21 @@ static void drain_pages_zone(unsigned int cpu, struct zone *zone)
22876 unsigned long flags;
22877 struct per_cpu_pageset *pset;
22878 struct per_cpu_pages *pcp;
22882 - local_irq_save(flags);
22883 + cpu_lock_irqsave(cpu, flags);
22884 pset = per_cpu_ptr(zone->pageset, cpu);
22887 - if (pcp->count) {
22888 - free_pcppages_bulk(zone, pcp->count, pcp);
22889 + count = pcp->count;
22891 + isolate_pcp_pages(count, pcp, &dst);
22894 - local_irq_restore(flags);
22895 + cpu_unlock_irqrestore(cpu, flags);
22897 + free_pcppages_bulk(zone, count, &dst);
22901 @@ -2366,8 +2412,17 @@ void drain_all_pages(struct zone *zone)
22903 cpumask_clear_cpu(cpu, &cpus_with_pcps);
22905 +#ifndef CONFIG_PREEMPT_RT_BASE
22906 on_each_cpu_mask(&cpus_with_pcps, (smp_call_func_t) drain_local_pages,
22909 + for_each_cpu(cpu, &cpus_with_pcps) {
22911 + drain_pages_zone(cpu, zone);
22913 + drain_pages(cpu);
22918 #ifdef CONFIG_HIBERNATION
22919 @@ -2427,7 +2482,7 @@ void free_hot_cold_page(struct page *page, bool cold)
22921 migratetype = get_pfnblock_migratetype(page, pfn);
22922 set_pcppage_migratetype(page, migratetype);
22923 - local_irq_save(flags);
22924 + local_lock_irqsave(pa_lock, flags);
22925 __count_vm_event(PGFREE);
22928 @@ -2453,12 +2508,17 @@ void free_hot_cold_page(struct page *page, bool cold)
22930 if (pcp->count >= pcp->high) {
22931 unsigned long batch = READ_ONCE(pcp->batch);
22932 - free_pcppages_bulk(zone, batch, pcp);
22935 + isolate_pcp_pages(batch, pcp, &dst);
22936 pcp->count -= batch;
22937 + local_unlock_irqrestore(pa_lock, flags);
22938 + free_pcppages_bulk(zone, batch, &dst);
22943 - local_irq_restore(flags);
22944 + local_unlock_irqrestore(pa_lock, flags);
22948 @@ -2600,7 +2660,7 @@ struct page *buffered_rmqueue(struct zone *preferred_zone,
22949 struct per_cpu_pages *pcp;
22950 struct list_head *list;
22952 - local_irq_save(flags);
22953 + local_lock_irqsave(pa_lock, flags);
22955 pcp = &this_cpu_ptr(zone->pageset)->pcp;
22956 list = &pcp->lists[migratetype];
22957 @@ -2627,7 +2687,7 @@ struct page *buffered_rmqueue(struct zone *preferred_zone,
22958 * allocate greater than order-1 page units with __GFP_NOFAIL.
22960 WARN_ON_ONCE((gfp_flags & __GFP_NOFAIL) && (order > 1));
22961 - spin_lock_irqsave(&zone->lock, flags);
22962 + local_spin_lock_irqsave(pa_lock, &zone->lock, flags);
22966 @@ -2639,22 +2699,24 @@ struct page *buffered_rmqueue(struct zone *preferred_zone,
22968 page = __rmqueue(zone, order, migratetype);
22969 } while (page && check_new_pages(page, order));
22970 - spin_unlock(&zone->lock);
22973 + spin_unlock(&zone->lock);
22976 __mod_zone_freepage_state(zone, -(1 << order),
22977 get_pcppage_migratetype(page));
22978 + spin_unlock(&zone->lock);
22981 __count_zid_vm_events(PGALLOC, page_zonenum(page), 1 << order);
22982 zone_statistics(preferred_zone, zone, gfp_flags);
22983 - local_irq_restore(flags);
22984 + local_unlock_irqrestore(pa_lock, flags);
22986 VM_BUG_ON_PAGE(bad_range(zone, page), page);
22990 - local_irq_restore(flags);
22991 + local_unlock_irqrestore(pa_lock, flags);
22995 @@ -6531,7 +6593,9 @@ static int page_alloc_cpu_notify(struct notifier_block *self,
22996 int cpu = (unsigned long)hcpu;
22998 if (action == CPU_DEAD || action == CPU_DEAD_FROZEN) {
22999 + local_lock_irq_on(swapvec_lock, cpu);
23000 lru_add_drain_cpu(cpu);
23001 + local_unlock_irq_on(swapvec_lock, cpu);
23005 @@ -6557,6 +6621,7 @@ static int page_alloc_cpu_notify(struct notifier_block *self,
23006 void __init page_alloc_init(void)
23008 hotcpu_notifier(page_alloc_cpu_notify, 0);
23009 + local_irq_lock_init(pa_lock);
23013 @@ -7385,7 +7450,7 @@ void zone_pcp_reset(struct zone *zone)
23014 struct per_cpu_pageset *pset;
23016 /* avoid races with drain_pages() */
23017 - local_irq_save(flags);
23018 + local_lock_irqsave(pa_lock, flags);
23019 if (zone->pageset != &boot_pageset) {
23020 for_each_online_cpu(cpu) {
23021 pset = per_cpu_ptr(zone->pageset, cpu);
23022 @@ -7394,7 +7459,7 @@ void zone_pcp_reset(struct zone *zone)
23023 free_percpu(zone->pageset);
23024 zone->pageset = &boot_pageset;
23026 - local_irq_restore(flags);
23027 + local_unlock_irqrestore(pa_lock, flags);
23030 #ifdef CONFIG_MEMORY_HOTREMOVE
23031 diff --git a/mm/percpu.c b/mm/percpu.c
23032 index 255714302394..59b529b944a9 100644
23035 @@ -1280,6 +1280,28 @@ void free_percpu(void __percpu *ptr)
23037 EXPORT_SYMBOL_GPL(free_percpu);
23039 +bool __is_kernel_percpu_address(unsigned long addr, unsigned long *can_addr)
23042 + const size_t static_size = __per_cpu_end - __per_cpu_start;
23043 + void __percpu *base = __addr_to_pcpu_ptr(pcpu_base_addr);
23044 + unsigned int cpu;
23046 + for_each_possible_cpu(cpu) {
23047 + void *start = per_cpu_ptr(base, cpu);
23048 + void *va = (void *)addr;
23050 + if (va >= start && va < start + static_size) {
23052 + *can_addr = (unsigned long) (va - start);
23057 + /* on UP, can't distinguish from other static vars, always false */
23062 * is_kernel_percpu_address - test whether address is from static percpu area
23063 * @addr: address to test
23064 @@ -1293,20 +1315,7 @@ EXPORT_SYMBOL_GPL(free_percpu);
23066 bool is_kernel_percpu_address(unsigned long addr)
23069 - const size_t static_size = __per_cpu_end - __per_cpu_start;
23070 - void __percpu *base = __addr_to_pcpu_ptr(pcpu_base_addr);
23071 - unsigned int cpu;
23073 - for_each_possible_cpu(cpu) {
23074 - void *start = per_cpu_ptr(base, cpu);
23076 - if ((void *)addr >= start && (void *)addr < start + static_size)
23080 - /* on UP, can't distinguish from other static vars, always false */
23082 + return __is_kernel_percpu_address(addr, NULL);
23086 diff --git a/mm/slab.h b/mm/slab.h
23087 index bc05fdc3edce..610cf61634f0 100644
23090 @@ -426,7 +426,11 @@ static inline void slab_post_alloc_hook(struct kmem_cache *s, gfp_t flags,
23091 * The slab lists for all objects.
23093 struct kmem_cache_node {
23094 +#ifdef CONFIG_SLUB
23095 + raw_spinlock_t list_lock;
23097 spinlock_t list_lock;
23101 struct list_head slabs_partial; /* partial list first, better asm code */
23102 diff --git a/mm/slub.c b/mm/slub.c
23103 index 7aa0e97af928..35c873f09201 100644
23106 @@ -1141,7 +1141,7 @@ static noinline int free_debug_processing(
23107 unsigned long uninitialized_var(flags);
23110 - spin_lock_irqsave(&n->list_lock, flags);
23111 + raw_spin_lock_irqsave(&n->list_lock, flags);
23114 if (s->flags & SLAB_CONSISTENCY_CHECKS) {
23115 @@ -1176,7 +1176,7 @@ static noinline int free_debug_processing(
23119 - spin_unlock_irqrestore(&n->list_lock, flags);
23120 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23122 slab_fix(s, "Object at 0x%p not freed", object);
23124 @@ -1304,6 +1304,12 @@ static inline void dec_slabs_node(struct kmem_cache *s, int node,
23126 #endif /* CONFIG_SLUB_DEBUG */
23128 +struct slub_free_list {
23129 + raw_spinlock_t lock;
23130 + struct list_head list;
23132 +static DEFINE_PER_CPU(struct slub_free_list, slub_free_list);
23135 * Hooks for other subsystems that check memory allocations. In a typical
23136 * production configuration these hooks all should produce no code at all.
23137 @@ -1527,10 +1533,17 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
23141 + bool enableirqs = false;
23143 flags &= gfp_allowed_mask;
23145 if (gfpflags_allow_blocking(flags))
23146 + enableirqs = true;
23147 +#ifdef CONFIG_PREEMPT_RT_FULL
23148 + if (system_state == SYSTEM_RUNNING)
23149 + enableirqs = true;
23152 local_irq_enable();
23154 flags |= s->allocflags;
23155 @@ -1605,7 +1618,7 @@ static struct page *allocate_slab(struct kmem_cache *s, gfp_t flags, int node)
23159 - if (gfpflags_allow_blocking(flags))
23161 local_irq_disable();
23164 @@ -1664,6 +1677,16 @@ static void __free_slab(struct kmem_cache *s, struct page *page)
23165 __free_pages(page, order);
23168 +static void free_delayed(struct list_head *h)
23170 + while(!list_empty(h)) {
23171 + struct page *page = list_first_entry(h, struct page, lru);
23173 + list_del(&page->lru);
23174 + __free_slab(page->slab_cache, page);
23178 #define need_reserve_slab_rcu \
23179 (sizeof(((struct page *)NULL)->lru) < sizeof(struct rcu_head))
23181 @@ -1695,6 +1718,12 @@ static void free_slab(struct kmem_cache *s, struct page *page)
23184 call_rcu(head, rcu_free_slab);
23185 + } else if (irqs_disabled()) {
23186 + struct slub_free_list *f = this_cpu_ptr(&slub_free_list);
23188 + raw_spin_lock(&f->lock);
23189 + list_add(&page->lru, &f->list);
23190 + raw_spin_unlock(&f->lock);
23192 __free_slab(s, page);
23194 @@ -1802,7 +1831,7 @@ static void *get_partial_node(struct kmem_cache *s, struct kmem_cache_node *n,
23195 if (!n || !n->nr_partial)
23198 - spin_lock(&n->list_lock);
23199 + raw_spin_lock(&n->list_lock);
23200 list_for_each_entry_safe(page, page2, &n->partial, lru) {
23203 @@ -1827,7 +1856,7 @@ static void *get_partial_node(struct kmem_cache *s, struct kmem_cache_node *n,
23207 - spin_unlock(&n->list_lock);
23208 + raw_spin_unlock(&n->list_lock);
23212 @@ -2073,7 +2102,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page,
23213 * that acquire_slab() will see a slab page that
23216 - spin_lock(&n->list_lock);
23217 + raw_spin_lock(&n->list_lock);
23221 @@ -2084,7 +2113,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page,
23222 * slabs from diagnostic functions will not see
23223 * any frozen slabs.
23225 - spin_lock(&n->list_lock);
23226 + raw_spin_lock(&n->list_lock);
23230 @@ -2119,7 +2148,7 @@ static void deactivate_slab(struct kmem_cache *s, struct page *page,
23234 - spin_unlock(&n->list_lock);
23235 + raw_spin_unlock(&n->list_lock);
23238 stat(s, DEACTIVATE_EMPTY);
23239 @@ -2151,10 +2180,10 @@ static void unfreeze_partials(struct kmem_cache *s,
23240 n2 = get_node(s, page_to_nid(page));
23243 - spin_unlock(&n->list_lock);
23244 + raw_spin_unlock(&n->list_lock);
23247 - spin_lock(&n->list_lock);
23248 + raw_spin_lock(&n->list_lock);
23252 @@ -2183,7 +2212,7 @@ static void unfreeze_partials(struct kmem_cache *s,
23256 - spin_unlock(&n->list_lock);
23257 + raw_spin_unlock(&n->list_lock);
23259 while (discard_page) {
23260 page = discard_page;
23261 @@ -2222,14 +2251,21 @@ static void put_cpu_partial(struct kmem_cache *s, struct page *page, int drain)
23262 pobjects = oldpage->pobjects;
23263 pages = oldpage->pages;
23264 if (drain && pobjects > s->cpu_partial) {
23265 + struct slub_free_list *f;
23266 unsigned long flags;
23267 + LIST_HEAD(tofree);
23269 * partial array is full. Move the existing
23270 * set to the per node partial list.
23272 local_irq_save(flags);
23273 unfreeze_partials(s, this_cpu_ptr(s->cpu_slab));
23274 + f = this_cpu_ptr(&slub_free_list);
23275 + raw_spin_lock(&f->lock);
23276 + list_splice_init(&f->list, &tofree);
23277 + raw_spin_unlock(&f->lock);
23278 local_irq_restore(flags);
23279 + free_delayed(&tofree);
23283 @@ -2301,7 +2337,22 @@ static bool has_cpu_slab(int cpu, void *info)
23285 static void flush_all(struct kmem_cache *s)
23287 + LIST_HEAD(tofree);
23290 on_each_cpu_cond(has_cpu_slab, flush_cpu_slab, s, 1, GFP_ATOMIC);
23291 + for_each_online_cpu(cpu) {
23292 + struct slub_free_list *f;
23294 + if (!has_cpu_slab(cpu, s))
23297 + f = &per_cpu(slub_free_list, cpu);
23298 + raw_spin_lock_irq(&f->lock);
23299 + list_splice_init(&f->list, &tofree);
23300 + raw_spin_unlock_irq(&f->lock);
23301 + free_delayed(&tofree);
23306 @@ -2356,10 +2407,10 @@ static unsigned long count_partial(struct kmem_cache_node *n,
23307 unsigned long x = 0;
23310 - spin_lock_irqsave(&n->list_lock, flags);
23311 + raw_spin_lock_irqsave(&n->list_lock, flags);
23312 list_for_each_entry(page, &n->partial, lru)
23313 x += get_count(page);
23314 - spin_unlock_irqrestore(&n->list_lock, flags);
23315 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23318 #endif /* CONFIG_SLUB_DEBUG || CONFIG_SYSFS */
23319 @@ -2497,8 +2548,10 @@ static inline void *get_freelist(struct kmem_cache *s, struct page *page)
23320 * already disabled (which is the case for bulk allocation).
23322 static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
23323 - unsigned long addr, struct kmem_cache_cpu *c)
23324 + unsigned long addr, struct kmem_cache_cpu *c,
23325 + struct list_head *to_free)
23327 + struct slub_free_list *f;
23331 @@ -2558,6 +2611,13 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
23332 VM_BUG_ON(!c->page->frozen);
23333 c->freelist = get_freepointer(s, freelist);
23334 c->tid = next_tid(c->tid);
23337 + f = this_cpu_ptr(&slub_free_list);
23338 + raw_spin_lock(&f->lock);
23339 + list_splice_init(&f->list, to_free);
23340 + raw_spin_unlock(&f->lock);
23345 @@ -2589,7 +2649,7 @@ static void *___slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
23346 deactivate_slab(s, page, get_freepointer(s, freelist));
23348 c->freelist = NULL;
23354 @@ -2601,6 +2661,7 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
23357 unsigned long flags;
23358 + LIST_HEAD(tofree);
23360 local_irq_save(flags);
23361 #ifdef CONFIG_PREEMPT
23362 @@ -2612,8 +2673,9 @@ static void *__slab_alloc(struct kmem_cache *s, gfp_t gfpflags, int node,
23363 c = this_cpu_ptr(s->cpu_slab);
23366 - p = ___slab_alloc(s, gfpflags, node, addr, c);
23367 + p = ___slab_alloc(s, gfpflags, node, addr, c, &tofree);
23368 local_irq_restore(flags);
23369 + free_delayed(&tofree);
23373 @@ -2799,7 +2861,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
23377 - spin_unlock_irqrestore(&n->list_lock, flags);
23378 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23381 prior = page->freelist;
23382 @@ -2831,7 +2893,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
23383 * Otherwise the list_lock will synchronize with
23384 * other processors updating the list of slabs.
23386 - spin_lock_irqsave(&n->list_lock, flags);
23387 + raw_spin_lock_irqsave(&n->list_lock, flags);
23391 @@ -2873,7 +2935,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
23392 add_partial(n, page, DEACTIVATE_TO_TAIL);
23393 stat(s, FREE_ADD_PARTIAL);
23395 - spin_unlock_irqrestore(&n->list_lock, flags);
23396 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23400 @@ -2888,7 +2950,7 @@ static void __slab_free(struct kmem_cache *s, struct page *page,
23401 remove_full(s, n, page);
23404 - spin_unlock_irqrestore(&n->list_lock, flags);
23405 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23406 stat(s, FREE_SLAB);
23407 discard_slab(s, page);
23409 @@ -3093,6 +3155,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
23412 struct kmem_cache_cpu *c;
23413 + LIST_HEAD(to_free);
23416 /* memcg and kmem_cache debug support */
23417 @@ -3116,7 +3179,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
23418 * of re-populating per CPU c->freelist
23420 p[i] = ___slab_alloc(s, flags, NUMA_NO_NODE,
23422 + _RET_IP_, c, &to_free);
23423 if (unlikely(!p[i]))
23426 @@ -3128,6 +3191,7 @@ int kmem_cache_alloc_bulk(struct kmem_cache *s, gfp_t flags, size_t size,
23428 c->tid = next_tid(c->tid);
23429 local_irq_enable();
23430 + free_delayed(&to_free);
23432 /* Clear memory outside IRQ disabled fastpath loop */
23433 if (unlikely(flags & __GFP_ZERO)) {
23434 @@ -3275,7 +3339,7 @@ static void
23435 init_kmem_cache_node(struct kmem_cache_node *n)
23438 - spin_lock_init(&n->list_lock);
23439 + raw_spin_lock_init(&n->list_lock);
23440 INIT_LIST_HEAD(&n->partial);
23441 #ifdef CONFIG_SLUB_DEBUG
23442 atomic_long_set(&n->nr_slabs, 0);
23443 @@ -3619,6 +3683,10 @@ static void list_slab_objects(struct kmem_cache *s, struct page *page,
23446 #ifdef CONFIG_SLUB_DEBUG
23447 +#ifdef CONFIG_PREEMPT_RT_BASE
23448 + /* XXX move out of irq-off section */
23449 + slab_err(s, page, text, s->name);
23451 void *addr = page_address(page);
23453 unsigned long *map = kzalloc(BITS_TO_LONGS(page->objects) *
23454 @@ -3639,6 +3707,7 @@ static void list_slab_objects(struct kmem_cache *s, struct page *page,
23462 @@ -3652,7 +3721,7 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n)
23463 struct page *page, *h;
23465 BUG_ON(irqs_disabled());
23466 - spin_lock_irq(&n->list_lock);
23467 + raw_spin_lock_irq(&n->list_lock);
23468 list_for_each_entry_safe(page, h, &n->partial, lru) {
23469 if (!page->inuse) {
23470 remove_partial(n, page);
23471 @@ -3662,7 +3731,7 @@ static void free_partial(struct kmem_cache *s, struct kmem_cache_node *n)
23472 "Objects remaining in %s on __kmem_cache_shutdown()");
23475 - spin_unlock_irq(&n->list_lock);
23476 + raw_spin_unlock_irq(&n->list_lock);
23478 list_for_each_entry_safe(page, h, &discard, lru)
23479 discard_slab(s, page);
23480 @@ -3920,7 +3989,7 @@ int __kmem_cache_shrink(struct kmem_cache *s, bool deactivate)
23481 for (i = 0; i < SHRINK_PROMOTE_MAX; i++)
23482 INIT_LIST_HEAD(promote + i);
23484 - spin_lock_irqsave(&n->list_lock, flags);
23485 + raw_spin_lock_irqsave(&n->list_lock, flags);
23488 * Build lists of slabs to discard or promote.
23489 @@ -3951,7 +4020,7 @@ int __kmem_cache_shrink(struct kmem_cache *s, bool deactivate)
23490 for (i = SHRINK_PROMOTE_MAX - 1; i >= 0; i--)
23491 list_splice(promote + i, &n->partial);
23493 - spin_unlock_irqrestore(&n->list_lock, flags);
23494 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23496 /* Release empty slabs */
23497 list_for_each_entry_safe(page, t, &discard, lru)
23498 @@ -4127,6 +4196,12 @@ void __init kmem_cache_init(void)
23500 static __initdata struct kmem_cache boot_kmem_cache,
23501 boot_kmem_cache_node;
23504 + for_each_possible_cpu(cpu) {
23505 + raw_spin_lock_init(&per_cpu(slub_free_list, cpu).lock);
23506 + INIT_LIST_HEAD(&per_cpu(slub_free_list, cpu).list);
23509 if (debug_guardpage_minorder())
23510 slub_max_order = 0;
23511 @@ -4335,7 +4410,7 @@ static int validate_slab_node(struct kmem_cache *s,
23513 unsigned long flags;
23515 - spin_lock_irqsave(&n->list_lock, flags);
23516 + raw_spin_lock_irqsave(&n->list_lock, flags);
23518 list_for_each_entry(page, &n->partial, lru) {
23519 validate_slab_slab(s, page, map);
23520 @@ -4357,7 +4432,7 @@ static int validate_slab_node(struct kmem_cache *s,
23521 s->name, count, atomic_long_read(&n->nr_slabs));
23524 - spin_unlock_irqrestore(&n->list_lock, flags);
23525 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23529 @@ -4545,12 +4620,12 @@ static int list_locations(struct kmem_cache *s, char *buf,
23530 if (!atomic_long_read(&n->nr_slabs))
23533 - spin_lock_irqsave(&n->list_lock, flags);
23534 + raw_spin_lock_irqsave(&n->list_lock, flags);
23535 list_for_each_entry(page, &n->partial, lru)
23536 process_slab(&t, s, page, alloc, map);
23537 list_for_each_entry(page, &n->full, lru)
23538 process_slab(&t, s, page, alloc, map);
23539 - spin_unlock_irqrestore(&n->list_lock, flags);
23540 + raw_spin_unlock_irqrestore(&n->list_lock, flags);
23543 for (i = 0; i < t.count; i++) {
23544 diff --git a/mm/swap.c b/mm/swap.c
23545 index 4dcf852e1e6d..69c3a5b24060 100644
23549 #include <linux/memcontrol.h>
23550 #include <linux/gfp.h>
23551 #include <linux/uio.h>
23552 +#include <linux/locallock.h>
23553 #include <linux/hugetlb.h>
23554 #include <linux/page_idle.h>
23556 @@ -50,6 +51,8 @@ static DEFINE_PER_CPU(struct pagevec, lru_deactivate_pvecs);
23558 static DEFINE_PER_CPU(struct pagevec, activate_page_pvecs);
23560 +static DEFINE_LOCAL_IRQ_LOCK(rotate_lock);
23561 +DEFINE_LOCAL_IRQ_LOCK(swapvec_lock);
23564 * This path almost never happens for VM activity - pages are normally
23565 @@ -240,11 +243,11 @@ void rotate_reclaimable_page(struct page *page)
23566 unsigned long flags;
23569 - local_irq_save(flags);
23570 + local_lock_irqsave(rotate_lock, flags);
23571 pvec = this_cpu_ptr(&lru_rotate_pvecs);
23572 if (!pagevec_add(pvec, page) || PageCompound(page))
23573 pagevec_move_tail(pvec);
23574 - local_irq_restore(flags);
23575 + local_unlock_irqrestore(rotate_lock, flags);
23579 @@ -294,12 +297,13 @@ void activate_page(struct page *page)
23581 page = compound_head(page);
23582 if (PageLRU(page) && !PageActive(page) && !PageUnevictable(page)) {
23583 - struct pagevec *pvec = &get_cpu_var(activate_page_pvecs);
23584 + struct pagevec *pvec = &get_locked_var(swapvec_lock,
23585 + activate_page_pvecs);
23588 if (!pagevec_add(pvec, page) || PageCompound(page))
23589 pagevec_lru_move_fn(pvec, __activate_page, NULL);
23590 - put_cpu_var(activate_page_pvecs);
23591 + put_locked_var(swapvec_lock, activate_page_pvecs);
23595 @@ -326,7 +330,7 @@ void activate_page(struct page *page)
23597 static void __lru_cache_activate_page(struct page *page)
23599 - struct pagevec *pvec = &get_cpu_var(lru_add_pvec);
23600 + struct pagevec *pvec = &get_locked_var(swapvec_lock, lru_add_pvec);
23604 @@ -348,7 +352,7 @@ static void __lru_cache_activate_page(struct page *page)
23608 - put_cpu_var(lru_add_pvec);
23609 + put_locked_var(swapvec_lock, lru_add_pvec);
23613 @@ -390,12 +394,12 @@ EXPORT_SYMBOL(mark_page_accessed);
23615 static void __lru_cache_add(struct page *page)
23617 - struct pagevec *pvec = &get_cpu_var(lru_add_pvec);
23618 + struct pagevec *pvec = &get_locked_var(swapvec_lock, lru_add_pvec);
23621 if (!pagevec_add(pvec, page) || PageCompound(page))
23622 __pagevec_lru_add(pvec);
23623 - put_cpu_var(lru_add_pvec);
23624 + put_locked_var(swapvec_lock, lru_add_pvec);
23628 @@ -593,9 +597,15 @@ void lru_add_drain_cpu(int cpu)
23629 unsigned long flags;
23631 /* No harm done if a racing interrupt already did this */
23632 - local_irq_save(flags);
23633 +#ifdef CONFIG_PREEMPT_RT_BASE
23634 + local_lock_irqsave_on(rotate_lock, flags, cpu);
23635 pagevec_move_tail(pvec);
23636 - local_irq_restore(flags);
23637 + local_unlock_irqrestore_on(rotate_lock, flags, cpu);
23639 + local_lock_irqsave(rotate_lock, flags);
23640 + pagevec_move_tail(pvec);
23641 + local_unlock_irqrestore(rotate_lock, flags);
23645 pvec = &per_cpu(lru_deactivate_file_pvecs, cpu);
23646 @@ -627,11 +637,12 @@ void deactivate_file_page(struct page *page)
23649 if (likely(get_page_unless_zero(page))) {
23650 - struct pagevec *pvec = &get_cpu_var(lru_deactivate_file_pvecs);
23651 + struct pagevec *pvec = &get_locked_var(swapvec_lock,
23652 + lru_deactivate_file_pvecs);
23654 if (!pagevec_add(pvec, page) || PageCompound(page))
23655 pagevec_lru_move_fn(pvec, lru_deactivate_file_fn, NULL);
23656 - put_cpu_var(lru_deactivate_file_pvecs);
23657 + put_locked_var(swapvec_lock, lru_deactivate_file_pvecs);
23661 @@ -646,27 +657,31 @@ void deactivate_file_page(struct page *page)
23662 void deactivate_page(struct page *page)
23664 if (PageLRU(page) && PageActive(page) && !PageUnevictable(page)) {
23665 - struct pagevec *pvec = &get_cpu_var(lru_deactivate_pvecs);
23666 + struct pagevec *pvec = &get_locked_var(swapvec_lock,
23667 + lru_deactivate_pvecs);
23670 if (!pagevec_add(pvec, page) || PageCompound(page))
23671 pagevec_lru_move_fn(pvec, lru_deactivate_fn, NULL);
23672 - put_cpu_var(lru_deactivate_pvecs);
23673 + put_locked_var(swapvec_lock, lru_deactivate_pvecs);
23677 void lru_add_drain(void)
23679 - lru_add_drain_cpu(get_cpu());
23681 + lru_add_drain_cpu(local_lock_cpu(swapvec_lock));
23682 + local_unlock_cpu(swapvec_lock);
23685 -static void lru_add_drain_per_cpu(struct work_struct *dummy)
23686 +#ifdef CONFIG_PREEMPT_RT_BASE
23687 +static inline void remote_lru_add_drain(int cpu, struct cpumask *has_work)
23690 + local_lock_on(swapvec_lock, cpu);
23691 + lru_add_drain_cpu(cpu);
23692 + local_unlock_on(swapvec_lock, cpu);
23695 -static DEFINE_PER_CPU(struct work_struct, lru_add_drain_work);
23699 * lru_add_drain_wq is used to do lru_add_drain_all() from a WQ_MEM_RECLAIM
23700 @@ -686,6 +701,22 @@ static int __init lru_init(void)
23702 early_initcall(lru_init);
23704 +static void lru_add_drain_per_cpu(struct work_struct *dummy)
23709 +static DEFINE_PER_CPU(struct work_struct, lru_add_drain_work);
23710 +static inline void remote_lru_add_drain(int cpu, struct cpumask *has_work)
23712 + struct work_struct *work = &per_cpu(lru_add_drain_work, cpu);
23714 + INIT_WORK(work, lru_add_drain_per_cpu);
23715 + queue_work_on(cpu, lru_add_drain_wq, work);
23716 + cpumask_set_cpu(cpu, has_work);
23720 void lru_add_drain_all(void)
23722 static DEFINE_MUTEX(lock);
23723 @@ -697,21 +728,18 @@ void lru_add_drain_all(void)
23724 cpumask_clear(&has_work);
23726 for_each_online_cpu(cpu) {
23727 - struct work_struct *work = &per_cpu(lru_add_drain_work, cpu);
23729 if (pagevec_count(&per_cpu(lru_add_pvec, cpu)) ||
23730 pagevec_count(&per_cpu(lru_rotate_pvecs, cpu)) ||
23731 pagevec_count(&per_cpu(lru_deactivate_file_pvecs, cpu)) ||
23732 pagevec_count(&per_cpu(lru_deactivate_pvecs, cpu)) ||
23733 - need_activate_page_drain(cpu)) {
23734 - INIT_WORK(work, lru_add_drain_per_cpu);
23735 - queue_work_on(cpu, lru_add_drain_wq, work);
23736 - cpumask_set_cpu(cpu, &has_work);
23738 + need_activate_page_drain(cpu))
23739 + remote_lru_add_drain(cpu, &has_work);
23742 +#ifndef CONFIG_PREEMPT_RT_BASE
23743 for_each_cpu(cpu, &has_work)
23744 flush_work(&per_cpu(lru_add_drain_work, cpu));
23748 mutex_unlock(&lock);
23749 diff --git a/mm/truncate.c b/mm/truncate.c
23750 index 8d8c62d89e6d..5bf1bd25d077 100644
23751 --- a/mm/truncate.c
23752 +++ b/mm/truncate.c
23753 @@ -62,9 +62,12 @@ static void clear_exceptional_entry(struct address_space *mapping,
23754 * protected by mapping->tree_lock.
23756 if (!workingset_node_shadows(node) &&
23757 - !list_empty(&node->private_list))
23758 - list_lru_del(&workingset_shadow_nodes,
23759 + !list_empty(&node->private_list)) {
23760 + local_lock(workingset_shadow_lock);
23761 + list_lru_del(&__workingset_shadow_nodes,
23762 &node->private_list);
23763 + local_unlock(workingset_shadow_lock);
23765 __radix_tree_delete_node(&mapping->page_tree, node);
23767 spin_unlock_irq(&mapping->tree_lock);
23768 diff --git a/mm/vmalloc.c b/mm/vmalloc.c
23769 index f2481cb4e6b2..db4de08fa97c 100644
23772 @@ -845,7 +845,7 @@ static void *new_vmap_block(unsigned int order, gfp_t gfp_mask)
23773 struct vmap_block *vb;
23774 struct vmap_area *va;
23775 unsigned long vb_idx;
23777 + int node, err, cpu;
23780 node = numa_node_id();
23781 @@ -888,11 +888,12 @@ static void *new_vmap_block(unsigned int order, gfp_t gfp_mask)
23783 radix_tree_preload_end();
23785 - vbq = &get_cpu_var(vmap_block_queue);
23786 + cpu = get_cpu_light();
23787 + vbq = this_cpu_ptr(&vmap_block_queue);
23788 spin_lock(&vbq->lock);
23789 list_add_tail_rcu(&vb->free_list, &vbq->free);
23790 spin_unlock(&vbq->lock);
23791 - put_cpu_var(vmap_block_queue);
23796 @@ -961,6 +962,7 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask)
23797 struct vmap_block *vb;
23798 void *vaddr = NULL;
23799 unsigned int order;
23802 BUG_ON(offset_in_page(size));
23803 BUG_ON(size > PAGE_SIZE*VMAP_MAX_ALLOC);
23804 @@ -975,7 +977,8 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask)
23805 order = get_order(size);
23808 - vbq = &get_cpu_var(vmap_block_queue);
23809 + cpu = get_cpu_light();
23810 + vbq = this_cpu_ptr(&vmap_block_queue);
23811 list_for_each_entry_rcu(vb, &vbq->free, free_list) {
23812 unsigned long pages_off;
23814 @@ -998,7 +1001,7 @@ static void *vb_alloc(unsigned long size, gfp_t gfp_mask)
23818 - put_cpu_var(vmap_block_queue);
23822 /* Allocate new block if nothing was found */
23823 diff --git a/mm/vmstat.c b/mm/vmstat.c
23824 index 604f26a4f696..312006d2db50 100644
23827 @@ -245,6 +245,7 @@ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
23831 + preempt_disable_rt();
23832 x = delta + __this_cpu_read(*p);
23834 t = __this_cpu_read(pcp->stat_threshold);
23835 @@ -254,6 +255,7 @@ void __mod_zone_page_state(struct zone *zone, enum zone_stat_item item,
23838 __this_cpu_write(*p, x);
23839 + preempt_enable_rt();
23841 EXPORT_SYMBOL(__mod_zone_page_state);
23843 @@ -265,6 +267,7 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
23847 + preempt_disable_rt();
23848 x = delta + __this_cpu_read(*p);
23850 t = __this_cpu_read(pcp->stat_threshold);
23851 @@ -274,6 +277,7 @@ void __mod_node_page_state(struct pglist_data *pgdat, enum node_stat_item item,
23854 __this_cpu_write(*p, x);
23855 + preempt_enable_rt();
23857 EXPORT_SYMBOL(__mod_node_page_state);
23859 @@ -306,6 +310,7 @@ void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
23860 s8 __percpu *p = pcp->vm_stat_diff + item;
23863 + preempt_disable_rt();
23864 v = __this_cpu_inc_return(*p);
23865 t = __this_cpu_read(pcp->stat_threshold);
23866 if (unlikely(v > t)) {
23867 @@ -314,6 +319,7 @@ void __inc_zone_state(struct zone *zone, enum zone_stat_item item)
23868 zone_page_state_add(v + overstep, zone, item);
23869 __this_cpu_write(*p, -overstep);
23871 + preempt_enable_rt();
23874 void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23875 @@ -322,6 +328,7 @@ void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23876 s8 __percpu *p = pcp->vm_node_stat_diff + item;
23879 + preempt_disable_rt();
23880 v = __this_cpu_inc_return(*p);
23881 t = __this_cpu_read(pcp->stat_threshold);
23882 if (unlikely(v > t)) {
23883 @@ -330,6 +337,7 @@ void __inc_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23884 node_page_state_add(v + overstep, pgdat, item);
23885 __this_cpu_write(*p, -overstep);
23887 + preempt_enable_rt();
23890 void __inc_zone_page_state(struct page *page, enum zone_stat_item item)
23891 @@ -350,6 +358,7 @@ void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
23892 s8 __percpu *p = pcp->vm_stat_diff + item;
23895 + preempt_disable_rt();
23896 v = __this_cpu_dec_return(*p);
23897 t = __this_cpu_read(pcp->stat_threshold);
23898 if (unlikely(v < - t)) {
23899 @@ -358,6 +367,7 @@ void __dec_zone_state(struct zone *zone, enum zone_stat_item item)
23900 zone_page_state_add(v - overstep, zone, item);
23901 __this_cpu_write(*p, overstep);
23903 + preempt_enable_rt();
23906 void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23907 @@ -366,6 +376,7 @@ void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23908 s8 __percpu *p = pcp->vm_node_stat_diff + item;
23911 + preempt_disable_rt();
23912 v = __this_cpu_dec_return(*p);
23913 t = __this_cpu_read(pcp->stat_threshold);
23914 if (unlikely(v < - t)) {
23915 @@ -374,6 +385,7 @@ void __dec_node_state(struct pglist_data *pgdat, enum node_stat_item item)
23916 node_page_state_add(v - overstep, pgdat, item);
23917 __this_cpu_write(*p, overstep);
23919 + preempt_enable_rt();
23922 void __dec_zone_page_state(struct page *page, enum zone_stat_item item)
23923 diff --git a/mm/workingset.c b/mm/workingset.c
23924 index fb1f9183d89a..7e6ef1a48cd3 100644
23925 --- a/mm/workingset.c
23926 +++ b/mm/workingset.c
23927 @@ -334,7 +334,8 @@ void workingset_activation(struct page *page)
23928 * point where they would still be useful.
23931 -struct list_lru workingset_shadow_nodes;
23932 +struct list_lru __workingset_shadow_nodes;
23933 +DEFINE_LOCAL_IRQ_LOCK(workingset_shadow_lock);
23935 static unsigned long count_shadow_nodes(struct shrinker *shrinker,
23936 struct shrink_control *sc)
23937 @@ -344,9 +345,9 @@ static unsigned long count_shadow_nodes(struct shrinker *shrinker,
23938 unsigned long pages;
23940 /* list_lru lock nests inside IRQ-safe mapping->tree_lock */
23941 - local_irq_disable();
23942 - shadow_nodes = list_lru_shrink_count(&workingset_shadow_nodes, sc);
23943 - local_irq_enable();
23944 + local_lock_irq(workingset_shadow_lock);
23945 + shadow_nodes = list_lru_shrink_count(&__workingset_shadow_nodes, sc);
23946 + local_unlock_irq(workingset_shadow_lock);
23949 pages = mem_cgroup_node_nr_lru_pages(sc->memcg, sc->nid,
23950 @@ -438,9 +439,9 @@ static enum lru_status shadow_lru_isolate(struct list_head *item,
23951 spin_unlock(&mapping->tree_lock);
23952 ret = LRU_REMOVED_RETRY;
23954 - local_irq_enable();
23955 + local_unlock_irq(workingset_shadow_lock);
23957 - local_irq_disable();
23958 + local_lock_irq(workingset_shadow_lock);
23959 spin_lock(lru_lock);
23962 @@ -451,10 +452,10 @@ static unsigned long scan_shadow_nodes(struct shrinker *shrinker,
23965 /* list_lru lock nests inside IRQ-safe mapping->tree_lock */
23966 - local_irq_disable();
23967 - ret = list_lru_shrink_walk(&workingset_shadow_nodes, sc,
23968 + local_lock_irq(workingset_shadow_lock);
23969 + ret = list_lru_shrink_walk(&__workingset_shadow_nodes, sc,
23970 shadow_lru_isolate, NULL);
23971 - local_irq_enable();
23972 + local_unlock_irq(workingset_shadow_lock);
23976 @@ -492,7 +493,7 @@ static int __init workingset_init(void)
23977 pr_info("workingset: timestamp_bits=%d max_order=%d bucket_order=%u\n",
23978 timestamp_bits, max_order, bucket_order);
23980 - ret = list_lru_init_key(&workingset_shadow_nodes, &shadow_nodes_key);
23981 + ret = list_lru_init_key(&__workingset_shadow_nodes, &shadow_nodes_key);
23984 ret = register_shrinker(&workingset_shadow_shrinker);
23985 @@ -500,7 +501,7 @@ static int __init workingset_init(void)
23989 - list_lru_destroy(&workingset_shadow_nodes);
23990 + list_lru_destroy(&__workingset_shadow_nodes);
23994 diff --git a/mm/zsmalloc.c b/mm/zsmalloc.c
23995 index b0bc023d25c5..5af6426fbcbe 100644
23996 --- a/mm/zsmalloc.c
23997 +++ b/mm/zsmalloc.c
23999 #include <linux/mount.h>
24000 #include <linux/migrate.h>
24001 #include <linux/pagemap.h>
24002 +#include <linux/locallock.h>
24004 #define ZSPAGE_MAGIC 0x58
24008 #define ZS_MAX_ZSPAGE_ORDER 2
24009 #define ZS_MAX_PAGES_PER_ZSPAGE (_AC(1, UL) << ZS_MAX_ZSPAGE_ORDER)
24011 #define ZS_HANDLE_SIZE (sizeof(unsigned long))
24013 +#ifdef CONFIG_PREEMPT_RT_FULL
24015 +struct zsmalloc_handle {
24016 + unsigned long addr;
24017 + struct mutex lock;
24020 +#define ZS_HANDLE_ALLOC_SIZE (sizeof(struct zsmalloc_handle))
24024 +#define ZS_HANDLE_ALLOC_SIZE (sizeof(unsigned long))
24028 * Object location (<PFN>, <obj_idx>) is encoded as
24029 * as single (unsigned long) handle value.
24030 @@ -327,7 +341,7 @@ static void SetZsPageMovable(struct zs_pool *pool, struct zspage *zspage) {}
24032 static int create_cache(struct zs_pool *pool)
24034 - pool->handle_cachep = kmem_cache_create("zs_handle", ZS_HANDLE_SIZE,
24035 + pool->handle_cachep = kmem_cache_create("zs_handle", ZS_HANDLE_ALLOC_SIZE,
24037 if (!pool->handle_cachep)
24039 @@ -351,10 +365,27 @@ static void destroy_cache(struct zs_pool *pool)
24041 static unsigned long cache_alloc_handle(struct zs_pool *pool, gfp_t gfp)
24043 - return (unsigned long)kmem_cache_alloc(pool->handle_cachep,
24044 - gfp & ~(__GFP_HIGHMEM|__GFP_MOVABLE));
24047 + p = kmem_cache_alloc(pool->handle_cachep,
24048 + gfp & ~(__GFP_HIGHMEM|__GFP_MOVABLE));
24049 +#ifdef CONFIG_PREEMPT_RT_FULL
24051 + struct zsmalloc_handle *zh = p;
24053 + mutex_init(&zh->lock);
24056 + return (unsigned long)p;
24059 +#ifdef CONFIG_PREEMPT_RT_FULL
24060 +static struct zsmalloc_handle *zs_get_pure_handle(unsigned long handle)
24062 + return (void *)(handle &~((1 << OBJ_TAG_BITS) - 1));
24066 static void cache_free_handle(struct zs_pool *pool, unsigned long handle)
24068 kmem_cache_free(pool->handle_cachep, (void *)handle);
24069 @@ -373,12 +404,18 @@ static void cache_free_zspage(struct zs_pool *pool, struct zspage *zspage)
24071 static void record_obj(unsigned long handle, unsigned long obj)
24073 +#ifdef CONFIG_PREEMPT_RT_FULL
24074 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24076 + WRITE_ONCE(zh->addr, obj);
24079 * lsb of @obj represents handle lock while other bits
24080 * represent object value the handle is pointing so
24081 * updating shouldn't do store tearing.
24083 WRITE_ONCE(*(unsigned long *)handle, obj);
24088 @@ -467,6 +504,7 @@ MODULE_ALIAS("zpool-zsmalloc");
24090 /* per-cpu VM mapping areas for zspage accesses that cross page boundaries */
24091 static DEFINE_PER_CPU(struct mapping_area, zs_map_area);
24092 +static DEFINE_LOCAL_IRQ_LOCK(zs_map_area_lock);
24094 static bool is_zspage_isolated(struct zspage *zspage)
24096 @@ -902,7 +940,13 @@ static unsigned long location_to_obj(struct page *page, unsigned int obj_idx)
24098 static unsigned long handle_to_obj(unsigned long handle)
24100 +#ifdef CONFIG_PREEMPT_RT_FULL
24101 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24105 return *(unsigned long *)handle;
24109 static unsigned long obj_to_head(struct page *page, void *obj)
24110 @@ -916,22 +960,46 @@ static unsigned long obj_to_head(struct page *page, void *obj)
24112 static inline int testpin_tag(unsigned long handle)
24114 +#ifdef CONFIG_PREEMPT_RT_FULL
24115 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24117 + return mutex_is_locked(&zh->lock);
24119 return bit_spin_is_locked(HANDLE_PIN_BIT, (unsigned long *)handle);
24123 static inline int trypin_tag(unsigned long handle)
24125 +#ifdef CONFIG_PREEMPT_RT_FULL
24126 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24128 + return mutex_trylock(&zh->lock);
24130 return bit_spin_trylock(HANDLE_PIN_BIT, (unsigned long *)handle);
24134 static void pin_tag(unsigned long handle)
24136 +#ifdef CONFIG_PREEMPT_RT_FULL
24137 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24139 + return mutex_lock(&zh->lock);
24141 bit_spin_lock(HANDLE_PIN_BIT, (unsigned long *)handle);
24145 static void unpin_tag(unsigned long handle)
24147 +#ifdef CONFIG_PREEMPT_RT_FULL
24148 + struct zsmalloc_handle *zh = zs_get_pure_handle(handle);
24150 + return mutex_unlock(&zh->lock);
24152 bit_spin_unlock(HANDLE_PIN_BIT, (unsigned long *)handle);
24156 static void reset_page(struct page *page)
24157 @@ -1423,7 +1491,7 @@ void *zs_map_object(struct zs_pool *pool, unsigned long handle,
24158 class = pool->size_class[class_idx];
24159 off = (class->size * obj_idx) & ~PAGE_MASK;
24161 - area = &get_cpu_var(zs_map_area);
24162 + area = &get_locked_var(zs_map_area_lock, zs_map_area);
24164 if (off + class->size <= PAGE_SIZE) {
24165 /* this object is contained entirely within a page */
24166 @@ -1477,7 +1545,7 @@ void zs_unmap_object(struct zs_pool *pool, unsigned long handle)
24168 __zs_unmap_object(area, pages, off, class->size);
24170 - put_cpu_var(zs_map_area);
24171 + put_locked_var(zs_map_area_lock, zs_map_area);
24173 migrate_read_unlock(zspage);
24175 diff --git a/net/core/dev.c b/net/core/dev.c
24176 index 60b0a6049e72..660ca3b9c60b 100644
24177 --- a/net/core/dev.c
24178 +++ b/net/core/dev.c
24179 @@ -190,6 +190,7 @@ static unsigned int napi_gen_id = NR_CPUS;
24180 static DEFINE_READ_MOSTLY_HASHTABLE(napi_hash, 8);
24182 static seqcount_t devnet_rename_seq;
24183 +static DEFINE_MUTEX(devnet_rename_mutex);
24185 static inline void dev_base_seq_inc(struct net *net)
24187 @@ -211,14 +212,14 @@ static inline struct hlist_head *dev_index_hash(struct net *net, int ifindex)
24188 static inline void rps_lock(struct softnet_data *sd)
24191 - spin_lock(&sd->input_pkt_queue.lock);
24192 + raw_spin_lock(&sd->input_pkt_queue.raw_lock);
24196 static inline void rps_unlock(struct softnet_data *sd)
24199 - spin_unlock(&sd->input_pkt_queue.lock);
24200 + raw_spin_unlock(&sd->input_pkt_queue.raw_lock);
24204 @@ -888,7 +889,8 @@ int netdev_get_name(struct net *net, char *name, int ifindex)
24205 strcpy(name, dev->name);
24207 if (read_seqcount_retry(&devnet_rename_seq, seq)) {
24209 + mutex_lock(&devnet_rename_mutex);
24210 + mutex_unlock(&devnet_rename_mutex);
24214 @@ -1157,20 +1159,17 @@ int dev_change_name(struct net_device *dev, const char *newname)
24215 if (dev->flags & IFF_UP)
24218 - write_seqcount_begin(&devnet_rename_seq);
24219 + mutex_lock(&devnet_rename_mutex);
24220 + __raw_write_seqcount_begin(&devnet_rename_seq);
24222 - if (strncmp(newname, dev->name, IFNAMSIZ) == 0) {
24223 - write_seqcount_end(&devnet_rename_seq);
24226 + if (strncmp(newname, dev->name, IFNAMSIZ) == 0)
24229 memcpy(oldname, dev->name, IFNAMSIZ);
24231 err = dev_get_valid_name(net, dev, newname);
24233 - write_seqcount_end(&devnet_rename_seq);
24239 if (oldname[0] && !strchr(oldname, '%'))
24240 netdev_info(dev, "renamed from %s\n", oldname);
24241 @@ -1183,11 +1182,12 @@ int dev_change_name(struct net_device *dev, const char *newname)
24243 memcpy(dev->name, oldname, IFNAMSIZ);
24244 dev->name_assign_type = old_assign_type;
24245 - write_seqcount_end(&devnet_rename_seq);
24251 - write_seqcount_end(&devnet_rename_seq);
24252 + __raw_write_seqcount_end(&devnet_rename_seq);
24253 + mutex_unlock(&devnet_rename_mutex);
24255 netdev_adjacent_rename_links(dev, oldname);
24257 @@ -1208,7 +1208,8 @@ int dev_change_name(struct net_device *dev, const char *newname)
24258 /* err >= 0 after dev_alloc_name() or stores the first errno */
24261 - write_seqcount_begin(&devnet_rename_seq);
24262 + mutex_lock(&devnet_rename_mutex);
24263 + __raw_write_seqcount_begin(&devnet_rename_seq);
24264 memcpy(dev->name, oldname, IFNAMSIZ);
24265 memcpy(oldname, newname, IFNAMSIZ);
24266 dev->name_assign_type = old_assign_type;
24267 @@ -1221,6 +1222,11 @@ int dev_change_name(struct net_device *dev, const char *newname)
24273 + __raw_write_seqcount_end(&devnet_rename_seq);
24274 + mutex_unlock(&devnet_rename_mutex);
24279 @@ -2258,6 +2264,7 @@ static void __netif_reschedule(struct Qdisc *q)
24280 sd->output_queue_tailp = &q->next_sched;
24281 raise_softirq_irqoff(NET_TX_SOFTIRQ);
24282 local_irq_restore(flags);
24283 + preempt_check_resched_rt();
24286 void __netif_schedule(struct Qdisc *q)
24287 @@ -2339,6 +2346,7 @@ void __dev_kfree_skb_irq(struct sk_buff *skb, enum skb_free_reason reason)
24288 __this_cpu_write(softnet_data.completion_queue, skb);
24289 raise_softirq_irqoff(NET_TX_SOFTIRQ);
24290 local_irq_restore(flags);
24291 + preempt_check_resched_rt();
24293 EXPORT_SYMBOL(__dev_kfree_skb_irq);
24295 @@ -3073,7 +3081,11 @@ static inline int __dev_xmit_skb(struct sk_buff *skb, struct Qdisc *q,
24296 * This permits qdisc->running owner to get the lock more
24297 * often and dequeue packets faster.
24299 +#ifdef CONFIG_PREEMPT_RT_FULL
24300 + contended = true;
24302 contended = qdisc_is_running(q);
24304 if (unlikely(contended))
24305 spin_lock(&q->busylock);
24307 @@ -3136,8 +3148,10 @@ static void skb_update_prio(struct sk_buff *skb)
24308 #define skb_update_prio(skb)
24311 +#ifndef CONFIG_PREEMPT_RT_FULL
24312 DEFINE_PER_CPU(int, xmit_recursion);
24313 EXPORT_SYMBOL(xmit_recursion);
24317 * dev_loopback_xmit - loop back @skb
24318 @@ -3371,8 +3385,7 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
24319 int cpu = smp_processor_id(); /* ok because BHs are off */
24321 if (txq->xmit_lock_owner != cpu) {
24322 - if (unlikely(__this_cpu_read(xmit_recursion) >
24323 - XMIT_RECURSION_LIMIT))
24324 + if (unlikely(xmit_rec_read() > XMIT_RECURSION_LIMIT))
24325 goto recursion_alert;
24327 skb = validate_xmit_skb(skb, dev);
24328 @@ -3382,9 +3395,9 @@ static int __dev_queue_xmit(struct sk_buff *skb, void *accel_priv)
24329 HARD_TX_LOCK(dev, txq, cpu);
24331 if (!netif_xmit_stopped(txq)) {
24332 - __this_cpu_inc(xmit_recursion);
24334 skb = dev_hard_start_xmit(skb, dev, txq, &rc);
24335 - __this_cpu_dec(xmit_recursion);
24337 if (dev_xmit_complete(rc)) {
24338 HARD_TX_UNLOCK(dev, txq);
24340 @@ -3758,6 +3771,7 @@ static int enqueue_to_backlog(struct sk_buff *skb, int cpu,
24343 local_irq_restore(flags);
24344 + preempt_check_resched_rt();
24346 atomic_long_inc(&skb->dev->rx_dropped);
24348 @@ -3776,7 +3790,7 @@ static int netif_rx_internal(struct sk_buff *skb)
24349 struct rps_dev_flow voidflow, *rflow = &voidflow;
24352 - preempt_disable();
24353 + migrate_disable();
24356 cpu = get_rps_cpu(skb->dev, skb, &rflow);
24357 @@ -3786,13 +3800,13 @@ static int netif_rx_internal(struct sk_buff *skb)
24358 ret = enqueue_to_backlog(skb, cpu, &rflow->last_qtail);
24361 - preempt_enable();
24362 + migrate_enable();
24366 unsigned int qtail;
24367 - ret = enqueue_to_backlog(skb, get_cpu(), &qtail);
24369 + ret = enqueue_to_backlog(skb, get_cpu_light(), &qtail);
24374 @@ -3826,11 +3840,9 @@ int netif_rx_ni(struct sk_buff *skb)
24376 trace_netif_rx_ni_entry(skb);
24378 - preempt_disable();
24379 + local_bh_disable();
24380 err = netif_rx_internal(skb);
24381 - if (local_softirq_pending())
24383 - preempt_enable();
24384 + local_bh_enable();
24388 @@ -4309,7 +4321,7 @@ static void flush_backlog(struct work_struct *work)
24389 skb_queue_walk_safe(&sd->input_pkt_queue, skb, tmp) {
24390 if (skb->dev->reg_state == NETREG_UNREGISTERING) {
24391 __skb_unlink(skb, &sd->input_pkt_queue);
24393 + __skb_queue_tail(&sd->tofree_queue, skb);
24394 input_queue_head_incr(sd);
24397 @@ -4319,11 +4331,14 @@ static void flush_backlog(struct work_struct *work)
24398 skb_queue_walk_safe(&sd->process_queue, skb, tmp) {
24399 if (skb->dev->reg_state == NETREG_UNREGISTERING) {
24400 __skb_unlink(skb, &sd->process_queue);
24402 + __skb_queue_tail(&sd->tofree_queue, skb);
24403 input_queue_head_incr(sd);
24406 + if (!skb_queue_empty(&sd->tofree_queue))
24407 + raise_softirq_irqoff(NET_RX_SOFTIRQ);
24412 static void flush_all_backlogs(void)
24413 @@ -4804,6 +4819,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd)
24414 sd->rps_ipi_list = NULL;
24416 local_irq_enable();
24417 + preempt_check_resched_rt();
24419 /* Send pending IPI's to kick RPS processing on remote cpus. */
24421 @@ -4817,6 +4833,7 @@ static void net_rps_action_and_irq_enable(struct softnet_data *sd)
24424 local_irq_enable();
24425 + preempt_check_resched_rt();
24428 static bool sd_has_rps_ipi_waiting(struct softnet_data *sd)
24429 @@ -4846,7 +4863,9 @@ static int process_backlog(struct napi_struct *napi, int quota)
24431 struct sk_buff *skb;
24433 + local_irq_disable();
24434 while ((skb = __skb_dequeue(&sd->process_queue))) {
24435 + local_irq_enable();
24437 __netif_receive_skb(skb);
24439 @@ -4854,9 +4873,9 @@ static int process_backlog(struct napi_struct *napi, int quota)
24440 if (++work >= quota)
24443 + local_irq_disable();
24446 - local_irq_disable();
24448 if (skb_queue_empty(&sd->input_pkt_queue)) {
24450 @@ -4894,9 +4913,11 @@ void __napi_schedule(struct napi_struct *n)
24451 local_irq_save(flags);
24452 ____napi_schedule(this_cpu_ptr(&softnet_data), n);
24453 local_irq_restore(flags);
24454 + preempt_check_resched_rt();
24456 EXPORT_SYMBOL(__napi_schedule);
24458 +#ifndef CONFIG_PREEMPT_RT_FULL
24460 * __napi_schedule_irqoff - schedule for receive
24461 * @n: entry to schedule
24462 @@ -4908,6 +4929,7 @@ void __napi_schedule_irqoff(struct napi_struct *n)
24463 ____napi_schedule(this_cpu_ptr(&softnet_data), n);
24465 EXPORT_SYMBOL(__napi_schedule_irqoff);
24468 void __napi_complete(struct napi_struct *n)
24470 @@ -5197,13 +5219,21 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
24471 struct softnet_data *sd = this_cpu_ptr(&softnet_data);
24472 unsigned long time_limit = jiffies + 2;
24473 int budget = netdev_budget;
24474 + struct sk_buff_head tofree_q;
24475 + struct sk_buff *skb;
24479 + __skb_queue_head_init(&tofree_q);
24481 local_irq_disable();
24482 + skb_queue_splice_init(&sd->tofree_queue, &tofree_q);
24483 list_splice_init(&sd->poll_list, &list);
24484 local_irq_enable();
24486 + while ((skb = __skb_dequeue(&tofree_q)))
24490 struct napi_struct *n;
24492 @@ -5234,7 +5264,7 @@ static __latent_entropy void net_rx_action(struct softirq_action *h)
24493 list_splice_tail(&repoll, &list);
24494 list_splice(&list, &sd->poll_list);
24495 if (!list_empty(&sd->poll_list))
24496 - __raise_softirq_irqoff(NET_RX_SOFTIRQ);
24497 + __raise_softirq_irqoff_ksoft(NET_RX_SOFTIRQ);
24499 net_rps_action_and_irq_enable(sd);
24501 @@ -7995,16 +8025,20 @@ static int dev_cpu_callback(struct notifier_block *nfb,
24503 raise_softirq_irqoff(NET_TX_SOFTIRQ);
24504 local_irq_enable();
24505 + preempt_check_resched_rt();
24507 /* Process offline CPU's input_pkt_queue */
24508 while ((skb = __skb_dequeue(&oldsd->process_queue))) {
24510 input_queue_head_incr(oldsd);
24512 - while ((skb = skb_dequeue(&oldsd->input_pkt_queue))) {
24513 + while ((skb = __skb_dequeue(&oldsd->input_pkt_queue))) {
24515 input_queue_head_incr(oldsd);
24517 + while ((skb = __skb_dequeue(&oldsd->tofree_queue))) {
24523 @@ -8309,8 +8343,9 @@ static int __init net_dev_init(void)
24525 INIT_WORK(flush, flush_backlog);
24527 - skb_queue_head_init(&sd->input_pkt_queue);
24528 - skb_queue_head_init(&sd->process_queue);
24529 + skb_queue_head_init_raw(&sd->input_pkt_queue);
24530 + skb_queue_head_init_raw(&sd->process_queue);
24531 + skb_queue_head_init_raw(&sd->tofree_queue);
24532 INIT_LIST_HEAD(&sd->poll_list);
24533 sd->output_queue_tailp = &sd->output_queue;
24535 diff --git a/net/core/filter.c b/net/core/filter.c
24536 index b391209838ef..b86e9681a88e 100644
24537 --- a/net/core/filter.c
24538 +++ b/net/core/filter.c
24539 @@ -1645,7 +1645,7 @@ static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb)
24543 - if (unlikely(__this_cpu_read(xmit_recursion) > XMIT_RECURSION_LIMIT)) {
24544 + if (unlikely(xmit_rec_read() > XMIT_RECURSION_LIMIT)) {
24545 net_crit_ratelimited("bpf: recursion limit reached on datapath, buggy bpf program?\n");
24548 @@ -1653,9 +1653,9 @@ static inline int __bpf_tx_skb(struct net_device *dev, struct sk_buff *skb)
24552 - __this_cpu_inc(xmit_recursion);
24554 ret = dev_queue_xmit(skb);
24555 - __this_cpu_dec(xmit_recursion);
24560 diff --git a/net/core/gen_estimator.c b/net/core/gen_estimator.c
24561 index cad8e791f28e..2a9364fe62a5 100644
24562 --- a/net/core/gen_estimator.c
24563 +++ b/net/core/gen_estimator.c
24564 @@ -84,7 +84,7 @@ struct gen_estimator
24565 struct gnet_stats_basic_packed *bstats;
24566 struct gnet_stats_rate_est64 *rate_est;
24567 spinlock_t *stats_lock;
24568 - seqcount_t *running;
24569 + net_seqlock_t *running;
24572 unsigned long avpps;
24573 @@ -213,7 +213,7 @@ int gen_new_estimator(struct gnet_stats_basic_packed *bstats,
24574 struct gnet_stats_basic_cpu __percpu *cpu_bstats,
24575 struct gnet_stats_rate_est64 *rate_est,
24576 spinlock_t *stats_lock,
24577 - seqcount_t *running,
24578 + net_seqlock_t *running,
24579 struct nlattr *opt)
24581 struct gen_estimator *est;
24582 @@ -309,7 +309,7 @@ int gen_replace_estimator(struct gnet_stats_basic_packed *bstats,
24583 struct gnet_stats_basic_cpu __percpu *cpu_bstats,
24584 struct gnet_stats_rate_est64 *rate_est,
24585 spinlock_t *stats_lock,
24586 - seqcount_t *running, struct nlattr *opt)
24587 + net_seqlock_t *running, struct nlattr *opt)
24589 gen_kill_estimator(bstats, rate_est);
24590 return gen_new_estimator(bstats, cpu_bstats, rate_est, stats_lock, running, opt);
24591 diff --git a/net/core/gen_stats.c b/net/core/gen_stats.c
24592 index 508e051304fb..bc3b17b78c94 100644
24593 --- a/net/core/gen_stats.c
24594 +++ b/net/core/gen_stats.c
24595 @@ -130,7 +130,7 @@ __gnet_stats_copy_basic_cpu(struct gnet_stats_basic_packed *bstats,
24599 -__gnet_stats_copy_basic(const seqcount_t *running,
24600 +__gnet_stats_copy_basic(net_seqlock_t *running,
24601 struct gnet_stats_basic_packed *bstats,
24602 struct gnet_stats_basic_cpu __percpu *cpu,
24603 struct gnet_stats_basic_packed *b)
24604 @@ -143,10 +143,10 @@ __gnet_stats_copy_basic(const seqcount_t *running,
24608 - seq = read_seqcount_begin(running);
24609 + seq = net_seq_begin(running);
24610 bstats->bytes = b->bytes;
24611 bstats->packets = b->packets;
24612 - } while (running && read_seqcount_retry(running, seq));
24613 + } while (running && net_seq_retry(running, seq));
24615 EXPORT_SYMBOL(__gnet_stats_copy_basic);
24617 @@ -164,7 +164,7 @@ EXPORT_SYMBOL(__gnet_stats_copy_basic);
24618 * if the room in the socket buffer was not sufficient.
24621 -gnet_stats_copy_basic(const seqcount_t *running,
24622 +gnet_stats_copy_basic(net_seqlock_t *running,
24623 struct gnet_dump *d,
24624 struct gnet_stats_basic_cpu __percpu *cpu,
24625 struct gnet_stats_basic_packed *b)
24626 diff --git a/net/core/skbuff.c b/net/core/skbuff.c
24627 index 1e3e0087245b..1077b39db717 100644
24628 --- a/net/core/skbuff.c
24629 +++ b/net/core/skbuff.c
24631 #include <linux/errqueue.h>
24632 #include <linux/prefetch.h>
24633 #include <linux/if_vlan.h>
24634 +#include <linux/locallock.h>
24636 #include <net/protocol.h>
24637 #include <net/dst.h>
24638 @@ -360,6 +361,8 @@ struct napi_alloc_cache {
24640 static DEFINE_PER_CPU(struct page_frag_cache, netdev_alloc_cache);
24641 static DEFINE_PER_CPU(struct napi_alloc_cache, napi_alloc_cache);
24642 +static DEFINE_LOCAL_IRQ_LOCK(netdev_alloc_lock);
24643 +static DEFINE_LOCAL_IRQ_LOCK(napi_alloc_cache_lock);
24645 static void *__netdev_alloc_frag(unsigned int fragsz, gfp_t gfp_mask)
24647 @@ -367,10 +370,10 @@ static void *__netdev_alloc_frag(unsigned int fragsz, gfp_t gfp_mask)
24648 unsigned long flags;
24651 - local_irq_save(flags);
24652 + local_lock_irqsave(netdev_alloc_lock, flags);
24653 nc = this_cpu_ptr(&netdev_alloc_cache);
24654 data = __alloc_page_frag(nc, fragsz, gfp_mask);
24655 - local_irq_restore(flags);
24656 + local_unlock_irqrestore(netdev_alloc_lock, flags);
24660 @@ -389,9 +392,13 @@ EXPORT_SYMBOL(netdev_alloc_frag);
24662 static void *__napi_alloc_frag(unsigned int fragsz, gfp_t gfp_mask)
24664 - struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
24665 + struct napi_alloc_cache *nc;
24668 - return __alloc_page_frag(&nc->page, fragsz, gfp_mask);
24669 + nc = &get_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24670 + data = __alloc_page_frag(&nc->page, fragsz, gfp_mask);
24671 + put_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24675 void *napi_alloc_frag(unsigned int fragsz)
24676 @@ -438,13 +445,13 @@ struct sk_buff *__netdev_alloc_skb(struct net_device *dev, unsigned int len,
24677 if (sk_memalloc_socks())
24678 gfp_mask |= __GFP_MEMALLOC;
24680 - local_irq_save(flags);
24681 + local_lock_irqsave(netdev_alloc_lock, flags);
24683 nc = this_cpu_ptr(&netdev_alloc_cache);
24684 data = __alloc_page_frag(nc, len, gfp_mask);
24685 pfmemalloc = nc->pfmemalloc;
24687 - local_irq_restore(flags);
24688 + local_unlock_irqrestore(netdev_alloc_lock, flags);
24690 if (unlikely(!data))
24692 @@ -485,9 +492,10 @@ EXPORT_SYMBOL(__netdev_alloc_skb);
24693 struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
24696 - struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
24697 + struct napi_alloc_cache *nc;
24698 struct sk_buff *skb;
24702 len += NET_SKB_PAD + NET_IP_ALIGN;
24704 @@ -505,7 +513,10 @@ struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
24705 if (sk_memalloc_socks())
24706 gfp_mask |= __GFP_MEMALLOC;
24708 + nc = &get_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24709 data = __alloc_page_frag(&nc->page, len, gfp_mask);
24710 + pfmemalloc = nc->page.pfmemalloc;
24711 + put_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24712 if (unlikely(!data))
24715 @@ -516,7 +527,7 @@ struct sk_buff *__napi_alloc_skb(struct napi_struct *napi, unsigned int len,
24718 /* use OR instead of assignment to avoid clearing of bits in mask */
24719 - if (nc->page.pfmemalloc)
24721 skb->pfmemalloc = 1;
24722 skb->head_frag = 1;
24724 @@ -760,23 +771,26 @@ EXPORT_SYMBOL(consume_skb);
24726 void __kfree_skb_flush(void)
24728 - struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
24729 + struct napi_alloc_cache *nc;
24731 + nc = &get_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24732 /* flush skb_cache if containing objects */
24733 if (nc->skb_count) {
24734 kmem_cache_free_bulk(skbuff_head_cache, nc->skb_count,
24738 + put_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24741 static inline void _kfree_skb_defer(struct sk_buff *skb)
24743 - struct napi_alloc_cache *nc = this_cpu_ptr(&napi_alloc_cache);
24744 + struct napi_alloc_cache *nc;
24746 /* drop skb->head and call any destructors for packet */
24747 skb_release_all(skb);
24749 + nc = &get_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24750 /* record skb to CPU local list */
24751 nc->skb_cache[nc->skb_count++] = skb;
24753 @@ -791,6 +805,7 @@ static inline void _kfree_skb_defer(struct sk_buff *skb)
24757 + put_locked_var(napi_alloc_cache_lock, napi_alloc_cache);
24759 void __kfree_skb_defer(struct sk_buff *skb)
24761 diff --git a/net/core/sock.c b/net/core/sock.c
24762 index bc6543f7de36..2c32ee79620f 100644
24763 --- a/net/core/sock.c
24764 +++ b/net/core/sock.c
24765 @@ -2488,12 +2488,11 @@ void lock_sock_nested(struct sock *sk, int subclass)
24766 if (sk->sk_lock.owned)
24768 sk->sk_lock.owned = 1;
24769 - spin_unlock(&sk->sk_lock.slock);
24770 + spin_unlock_bh(&sk->sk_lock.slock);
24772 * The sk_lock has mutex_lock() semantics here:
24774 mutex_acquire(&sk->sk_lock.dep_map, subclass, 0, _RET_IP_);
24775 - local_bh_enable();
24777 EXPORT_SYMBOL(lock_sock_nested);
24779 diff --git a/net/ipv4/icmp.c b/net/ipv4/icmp.c
24780 index 48734ee6293f..e6864ff11352 100644
24781 --- a/net/ipv4/icmp.c
24782 +++ b/net/ipv4/icmp.c
24784 #include <linux/jiffies.h>
24785 #include <linux/kernel.h>
24786 #include <linux/fcntl.h>
24787 +#include <linux/sysrq.h>
24788 #include <linux/socket.h>
24789 #include <linux/in.h>
24790 #include <linux/inet.h>
24792 #include <linux/string.h>
24793 #include <linux/netfilter_ipv4.h>
24794 #include <linux/slab.h>
24795 +#include <linux/locallock.h>
24796 #include <net/snmp.h>
24797 #include <net/ip.h>
24798 #include <net/route.h>
24799 @@ -204,6 +206,8 @@ static const struct icmp_control icmp_pointers[NR_ICMP_TYPES+1];
24801 * On SMP we have one ICMP socket per-cpu.
24803 +static DEFINE_LOCAL_IRQ_LOCK(icmp_sk_lock);
24805 static struct sock *icmp_sk(struct net *net)
24807 return *this_cpu_ptr(net->ipv4.icmp_sk);
24808 @@ -215,12 +219,14 @@ static inline struct sock *icmp_xmit_lock(struct net *net)
24810 local_bh_disable();
24812 + local_lock(icmp_sk_lock);
24815 if (unlikely(!spin_trylock(&sk->sk_lock.slock))) {
24816 /* This can happen if the output path signals a
24817 * dst_link_failure() for an outgoing ICMP packet.
24819 + local_unlock(icmp_sk_lock);
24823 @@ -230,6 +236,7 @@ static inline struct sock *icmp_xmit_lock(struct net *net)
24824 static inline void icmp_xmit_unlock(struct sock *sk)
24826 spin_unlock_bh(&sk->sk_lock.slock);
24827 + local_unlock(icmp_sk_lock);
24830 int sysctl_icmp_msgs_per_sec __read_mostly = 1000;
24831 @@ -358,6 +365,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
24833 struct sk_buff *skb;
24835 + local_lock(icmp_sk_lock);
24836 sk = icmp_sk(dev_net((*rt)->dst.dev));
24837 if (ip_append_data(sk, fl4, icmp_glue_bits, icmp_param,
24838 icmp_param->data_len+icmp_param->head_len,
24839 @@ -380,6 +388,7 @@ static void icmp_push_reply(struct icmp_bxm *icmp_param,
24840 skb->ip_summed = CHECKSUM_NONE;
24841 ip_push_pending_frames(sk, fl4);
24843 + local_unlock(icmp_sk_lock);
24847 @@ -891,6 +900,30 @@ static bool icmp_redirect(struct sk_buff *skb)
24851 + * 32bit and 64bit have different timestamp length, so we check for
24852 + * the cookie at offset 20 and verify it is repeated at offset 50
24854 +#define CO_POS0 20
24855 +#define CO_POS1 50
24856 +#define CO_SIZE sizeof(int)
24857 +#define ICMP_SYSRQ_SIZE 57
24860 + * We got a ICMP_SYSRQ_SIZE sized ping request. Check for the cookie
24861 + * pattern and if it matches send the next byte as a trigger to sysrq.
24863 +static void icmp_check_sysrq(struct net *net, struct sk_buff *skb)
24865 + int cookie = htonl(net->ipv4.sysctl_icmp_echo_sysrq);
24866 + char *p = skb->data;
24868 + if (!memcmp(&cookie, p + CO_POS0, CO_SIZE) &&
24869 + !memcmp(&cookie, p + CO_POS1, CO_SIZE) &&
24870 + p[CO_POS0 + CO_SIZE] == p[CO_POS1 + CO_SIZE])
24871 + handle_sysrq(p[CO_POS0 + CO_SIZE]);
24875 * Handle ICMP_ECHO ("ping") requests.
24877 * RFC 1122: 3.2.2.6 MUST have an echo server that answers ICMP echo
24878 @@ -917,6 +950,11 @@ static bool icmp_echo(struct sk_buff *skb)
24879 icmp_param.data_len = skb->len;
24880 icmp_param.head_len = sizeof(struct icmphdr);
24881 icmp_reply(&icmp_param, skb);
24883 + if (skb->len == ICMP_SYSRQ_SIZE &&
24884 + net->ipv4.sysctl_icmp_echo_sysrq) {
24885 + icmp_check_sysrq(net, skb);
24888 /* should there be an ICMP stat for ignored echos? */
24890 diff --git a/net/ipv4/sysctl_net_ipv4.c b/net/ipv4/sysctl_net_ipv4.c
24891 index 80bc36b25de2..215b90adfb05 100644
24892 --- a/net/ipv4/sysctl_net_ipv4.c
24893 +++ b/net/ipv4/sysctl_net_ipv4.c
24894 @@ -681,6 +681,13 @@ static struct ctl_table ipv4_net_table[] = {
24895 .proc_handler = proc_dointvec
24898 + .procname = "icmp_echo_sysrq",
24899 + .data = &init_net.ipv4.sysctl_icmp_echo_sysrq,
24900 + .maxlen = sizeof(int),
24902 + .proc_handler = proc_dointvec
24905 .procname = "icmp_ignore_bogus_error_responses",
24906 .data = &init_net.ipv4.sysctl_icmp_ignore_bogus_error_responses,
24907 .maxlen = sizeof(int),
24908 diff --git a/net/ipv4/tcp_ipv4.c b/net/ipv4/tcp_ipv4.c
24909 index 2259114c7242..829e60985a81 100644
24910 --- a/net/ipv4/tcp_ipv4.c
24911 +++ b/net/ipv4/tcp_ipv4.c
24913 #include <linux/init.h>
24914 #include <linux/times.h>
24915 #include <linux/slab.h>
24916 +#include <linux/locallock.h>
24918 #include <net/net_namespace.h>
24919 #include <net/icmp.h>
24920 @@ -564,6 +565,7 @@ void tcp_v4_send_check(struct sock *sk, struct sk_buff *skb)
24922 EXPORT_SYMBOL(tcp_v4_send_check);
24924 +static DEFINE_LOCAL_IRQ_LOCK(tcp_sk_lock);
24926 * This routine will send an RST to the other tcp.
24928 @@ -691,6 +693,8 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
24929 offsetof(struct inet_timewait_sock, tw_bound_dev_if));
24931 arg.tos = ip_hdr(skb)->tos;
24933 + local_lock(tcp_sk_lock);
24934 local_bh_disable();
24935 ip_send_unicast_reply(*this_cpu_ptr(net->ipv4.tcp_sk),
24936 skb, &TCP_SKB_CB(skb)->header.h4.opt,
24937 @@ -700,6 +704,7 @@ static void tcp_v4_send_reset(const struct sock *sk, struct sk_buff *skb)
24938 __TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
24939 __TCP_INC_STATS(net, TCP_MIB_OUTRSTS);
24941 + local_unlock(tcp_sk_lock);
24943 #ifdef CONFIG_TCP_MD5SIG
24945 @@ -775,6 +780,7 @@ static void tcp_v4_send_ack(struct net *net,
24947 arg.bound_dev_if = oif;
24949 + local_lock(tcp_sk_lock);
24950 local_bh_disable();
24951 ip_send_unicast_reply(*this_cpu_ptr(net->ipv4.tcp_sk),
24952 skb, &TCP_SKB_CB(skb)->header.h4.opt,
24953 @@ -783,6 +789,7 @@ static void tcp_v4_send_ack(struct net *net,
24955 __TCP_INC_STATS(net, TCP_MIB_OUTSEGS);
24957 + local_unlock(tcp_sk_lock);
24960 static void tcp_v4_timewait_ack(struct sock *sk, struct sk_buff *skb)
24961 diff --git a/net/mac80211/rx.c b/net/mac80211/rx.c
24962 index 2384b4aae064..bf7ab51d7035 100644
24963 --- a/net/mac80211/rx.c
24964 +++ b/net/mac80211/rx.c
24965 @@ -4166,7 +4166,7 @@ void ieee80211_rx_napi(struct ieee80211_hw *hw, struct ieee80211_sta *pubsta,
24966 struct ieee80211_supported_band *sband;
24967 struct ieee80211_rx_status *status = IEEE80211_SKB_RXCB(skb);
24969 - WARN_ON_ONCE(softirq_count() == 0);
24970 + WARN_ON_ONCE_NONRT(softirq_count() == 0);
24972 if (WARN_ON(status->band >= NUM_NL80211_BANDS))
24974 diff --git a/net/netfilter/core.c b/net/netfilter/core.c
24975 index 004af030ef1a..b64f751bda45 100644
24976 --- a/net/netfilter/core.c
24977 +++ b/net/netfilter/core.c
24978 @@ -22,12 +22,18 @@
24979 #include <linux/proc_fs.h>
24980 #include <linux/mutex.h>
24981 #include <linux/slab.h>
24982 +#include <linux/locallock.h>
24983 #include <linux/rcupdate.h>
24984 #include <net/net_namespace.h>
24985 #include <net/sock.h>
24987 #include "nf_internals.h"
24989 +#ifdef CONFIG_PREEMPT_RT_BASE
24990 +DEFINE_LOCAL_IRQ_LOCK(xt_write_lock);
24991 +EXPORT_PER_CPU_SYMBOL(xt_write_lock);
24994 static DEFINE_MUTEX(afinfo_mutex);
24996 const struct nf_afinfo __rcu *nf_afinfo[NFPROTO_NUMPROTO] __read_mostly;
24997 diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
24998 index 34de326b4f09..fe9597af0840 100644
24999 --- a/net/packet/af_packet.c
25000 +++ b/net/packet/af_packet.c
25002 #include <linux/if_packet.h>
25003 #include <linux/wireless.h>
25004 #include <linux/kernel.h>
25005 +#include <linux/delay.h>
25006 #include <linux/kmod.h>
25007 #include <linux/slab.h>
25008 #include <linux/vmalloc.h>
25009 @@ -694,7 +695,7 @@ static void prb_retire_rx_blk_timer_expired(unsigned long data)
25010 if (BLOCK_NUM_PKTS(pbd)) {
25011 while (atomic_read(&pkc->blk_fill_in_prog)) {
25012 /* Waiting for skb_copy_bits to finish... */
25018 @@ -956,7 +957,7 @@ static void prb_retire_current_block(struct tpacket_kbdq_core *pkc,
25019 if (!(status & TP_STATUS_BLK_TMO)) {
25020 while (atomic_read(&pkc->blk_fill_in_prog)) {
25021 /* Waiting for skb_copy_bits to finish... */
25026 prb_close_block(pkc, pbd, po, status);
25027 diff --git a/net/rds/ib_rdma.c b/net/rds/ib_rdma.c
25028 index 977f69886c00..f3e7a36b0396 100644
25029 --- a/net/rds/ib_rdma.c
25030 +++ b/net/rds/ib_rdma.c
25032 #include <linux/slab.h>
25033 #include <linux/rculist.h>
25034 #include <linux/llist.h>
25035 +#include <linux/delay.h>
25037 #include "rds_single_path.h"
25039 @@ -210,7 +211,7 @@ static inline void wait_clean_list_grace(void)
25040 for_each_online_cpu(cpu) {
25041 flag = &per_cpu(clean_list_grace, cpu);
25042 while (test_bit(CLEAN_LIST_BUSY_BIT, flag))
25048 diff --git a/net/rxrpc/security.c b/net/rxrpc/security.c
25049 index 7d921e56e715..13df56a738e5 100644
25050 --- a/net/rxrpc/security.c
25051 +++ b/net/rxrpc/security.c
25053 #include <keys/rxrpc-type.h>
25054 #include "ar-internal.h"
25056 -static LIST_HEAD(rxrpc_security_methods);
25057 -static DECLARE_RWSEM(rxrpc_security_sem);
25059 static const struct rxrpc_security *rxrpc_security_types[] = {
25060 [RXRPC_SECURITY_NONE] = &rxrpc_no_security,
25061 #ifdef CONFIG_RXKAD
25062 diff --git a/net/sched/sch_api.c b/net/sched/sch_api.c
25063 index 206dc24add3a..00ea9bde5bb3 100644
25064 --- a/net/sched/sch_api.c
25065 +++ b/net/sched/sch_api.c
25066 @@ -981,7 +981,7 @@ static struct Qdisc *qdisc_create(struct net_device *dev,
25067 rcu_assign_pointer(sch->stab, stab);
25069 if (tca[TCA_RATE]) {
25070 - seqcount_t *running;
25071 + net_seqlock_t *running;
25074 if (sch->flags & TCQ_F_MQROOT)
25075 diff --git a/net/sched/sch_generic.c b/net/sched/sch_generic.c
25076 index 6cfb6e9038c2..20727e1347de 100644
25077 --- a/net/sched/sch_generic.c
25078 +++ b/net/sched/sch_generic.c
25079 @@ -425,7 +425,11 @@ struct Qdisc noop_qdisc = {
25080 .ops = &noop_qdisc_ops,
25081 .q.lock = __SPIN_LOCK_UNLOCKED(noop_qdisc.q.lock),
25082 .dev_queue = &noop_netdev_queue,
25083 +#ifdef CONFIG_PREEMPT_RT_BASE
25084 + .running = __SEQLOCK_UNLOCKED(noop_qdisc.running),
25086 .running = SEQCNT_ZERO(noop_qdisc.running),
25088 .busylock = __SPIN_LOCK_UNLOCKED(noop_qdisc.busylock),
25090 EXPORT_SYMBOL(noop_qdisc);
25091 @@ -624,9 +628,17 @@ struct Qdisc *qdisc_alloc(struct netdev_queue *dev_queue,
25092 lockdep_set_class(&sch->busylock,
25093 dev->qdisc_tx_busylock ?: &qdisc_tx_busylock);
25095 +#ifdef CONFIG_PREEMPT_RT_BASE
25096 + seqlock_init(&sch->running);
25097 + lockdep_set_class(&sch->running.seqcount,
25098 + dev->qdisc_running_key ?: &qdisc_running_key);
25099 + lockdep_set_class(&sch->running.lock,
25100 + dev->qdisc_running_key ?: &qdisc_running_key);
25102 seqcount_init(&sch->running);
25103 lockdep_set_class(&sch->running,
25104 dev->qdisc_running_key ?: &qdisc_running_key);
25108 sch->enqueue = ops->enqueue;
25109 @@ -925,7 +937,7 @@ void dev_deactivate_many(struct list_head *head)
25110 /* Wait for outstanding qdisc_run calls. */
25111 list_for_each_entry(dev, head, close_list)
25112 while (some_qdisc_is_busy(dev))
25117 void dev_deactivate(struct net_device *dev)
25118 diff --git a/net/sunrpc/svc_xprt.c b/net/sunrpc/svc_xprt.c
25119 index 9c9db55a0c1e..e6583b018a72 100644
25120 --- a/net/sunrpc/svc_xprt.c
25121 +++ b/net/sunrpc/svc_xprt.c
25122 @@ -396,7 +396,7 @@ void svc_xprt_do_enqueue(struct svc_xprt *xprt)
25127 + cpu = get_cpu_light();
25128 pool = svc_pool_for_cpu(xprt->xpt_server, cpu);
25130 atomic_long_inc(&pool->sp_stats.packets);
25131 @@ -432,7 +432,7 @@ void svc_xprt_do_enqueue(struct svc_xprt *xprt)
25133 atomic_long_inc(&pool->sp_stats.threads_woken);
25134 wake_up_process(rqstp->rq_task);
25140 @@ -453,7 +453,7 @@ void svc_xprt_do_enqueue(struct svc_xprt *xprt)
25147 trace_svc_xprt_do_enqueue(xprt, rqstp);
25149 diff --git a/scripts/mkcompile_h b/scripts/mkcompile_h
25150 index 6fdc97ef6023..523e0420d7f0 100755
25151 --- a/scripts/mkcompile_h
25152 +++ b/scripts/mkcompile_h
25153 @@ -4,7 +4,8 @@ TARGET=$1
25161 vecho() { [ "${quiet}" = "silent_" ] || echo "$@" ; }
25163 @@ -57,6 +58,7 @@ UTS_VERSION="#$VERSION"
25165 if [ -n "$SMP" ] ; then CONFIG_FLAGS="SMP"; fi
25166 if [ -n "$PREEMPT" ] ; then CONFIG_FLAGS="$CONFIG_FLAGS PREEMPT"; fi
25167 +if [ -n "$RT" ] ; then CONFIG_FLAGS="$CONFIG_FLAGS RT"; fi
25168 UTS_VERSION="$UTS_VERSION $CONFIG_FLAGS $TIMESTAMP"
25170 # Truncate to maximum length
25171 diff --git a/sound/core/pcm_native.c b/sound/core/pcm_native.c
25172 index 9d33c1e85c79..3d307bda86f9 100644
25173 --- a/sound/core/pcm_native.c
25174 +++ b/sound/core/pcm_native.c
25175 @@ -135,7 +135,7 @@ EXPORT_SYMBOL_GPL(snd_pcm_stream_unlock);
25176 void snd_pcm_stream_lock_irq(struct snd_pcm_substream *substream)
25178 if (!substream->pcm->nonatomic)
25179 - local_irq_disable();
25180 + local_irq_disable_nort();
25181 snd_pcm_stream_lock(substream);
25183 EXPORT_SYMBOL_GPL(snd_pcm_stream_lock_irq);
25184 @@ -150,7 +150,7 @@ void snd_pcm_stream_unlock_irq(struct snd_pcm_substream *substream)
25186 snd_pcm_stream_unlock(substream);
25187 if (!substream->pcm->nonatomic)
25188 - local_irq_enable();
25189 + local_irq_enable_nort();
25191 EXPORT_SYMBOL_GPL(snd_pcm_stream_unlock_irq);
25193 @@ -158,7 +158,7 @@ unsigned long _snd_pcm_stream_lock_irqsave(struct snd_pcm_substream *substream)
25195 unsigned long flags = 0;
25196 if (!substream->pcm->nonatomic)
25197 - local_irq_save(flags);
25198 + local_irq_save_nort(flags);
25199 snd_pcm_stream_lock(substream);
25202 @@ -176,7 +176,7 @@ void snd_pcm_stream_unlock_irqrestore(struct snd_pcm_substream *substream,
25204 snd_pcm_stream_unlock(substream);
25205 if (!substream->pcm->nonatomic)
25206 - local_irq_restore(flags);
25207 + local_irq_restore_nort(flags);
25209 EXPORT_SYMBOL_GPL(snd_pcm_stream_unlock_irqrestore);