-aufs4.7 kbuild patch
-
-diff --git a/fs/Kconfig b/fs/Kconfig
-index b8fcb41..78adefb 100644
---- a/fs/Kconfig
-+++ b/fs/Kconfig
-@@ -236,6 +236,7 @@ source "fs/pstore/Kconfig"
- source "fs/sysv/Kconfig"
- source "fs/ufs/Kconfig"
- source "fs/exofs/Kconfig"
-+source "fs/aufs/Kconfig"
-
- endif # MISC_FILESYSTEMS
-
-diff --git a/fs/Makefile b/fs/Makefile
-index 85b6e13..e7bb164 100644
---- a/fs/Makefile
-+++ b/fs/Makefile
-@@ -128,3 +128,4 @@ obj-y += exofs/ # Multiple modules
- obj-$(CONFIG_CEPH_FS) += ceph/
- obj-$(CONFIG_PSTORE) += pstore/
- obj-$(CONFIG_EFIVAR_FS) += efivarfs/
-+obj-$(CONFIG_AUFS_FS) += aufs/
-diff --git a/include/uapi/linux/Kbuild b/include/uapi/linux/Kbuild
-index ec10cfe..800211b 100644
---- a/include/uapi/linux/Kbuild
-+++ b/include/uapi/linux/Kbuild
-@@ -59,6 +59,7 @@ header-y += atmsvc.h
- header-y += atm_tcp.h
- header-y += atm_zatm.h
- header-y += audit.h
-+header-y += aufs_type.h
- header-y += auto_fs4.h
- header-y += auto_fs.h
- header-y += auxvec.h
-aufs4.7 base patch
-
-diff --git a/MAINTAINERS b/MAINTAINERS
-index 8c20323..d170184 100644
---- a/MAINTAINERS
-+++ b/MAINTAINERS
-@@ -2213,6 +2213,19 @@ F: include/linux/audit.h
- F: include/uapi/linux/audit.h
- F: kernel/audit*
-
-+AUFS (advanced multi layered unification filesystem) FILESYSTEM
-+M: "J. R. Okajima" <hooanon05g@gmail.com>
-+L: linux-unionfs@vger.kernel.org
-+L: aufs-users@lists.sourceforge.net (members only)
-+W: http://aufs.sourceforge.net
-+T: git://github.com/sfjro/aufs4-linux.git
-+S: Supported
-+F: Documentation/filesystems/aufs/
-+F: Documentation/ABI/testing/debugfs-aufs
-+F: Documentation/ABI/testing/sysfs-aufs
-+F: fs/aufs/
-+F: include/uapi/linux/aufs_type.h
-+
- AUXILIARY DISPLAY DRIVERS
- M: Miguel Ojeda Sandonis <miguel.ojeda.sandonis@gmail.com>
- W: http://miguelojeda.es/auxdisplay.htm
-diff --git a/drivers/block/loop.c b/drivers/block/loop.c
-index 1fa8cc2..7339e65 100644
---- a/drivers/block/loop.c
-+++ b/drivers/block/loop.c
-@@ -712,6 +712,24 @@ static inline int is_loop_device(struct file *file)
- return i && S_ISBLK(i->i_mode) && MAJOR(i->i_rdev) == LOOP_MAJOR;
- }
-
-+/*
-+ * for AUFS
-+ * no get/put for file.
-+ */
-+struct file *loop_backing_file(struct super_block *sb)
-+{
-+ struct file *ret;
-+ struct loop_device *l;
-+
-+ ret = NULL;
-+ if (MAJOR(sb->s_dev) == LOOP_MAJOR) {
-+ l = sb->s_bdev->bd_disk->private_data;
-+ ret = l->lo_backing_file;
-+ }
-+ return ret;
-+}
-+EXPORT_SYMBOL_GPL(loop_backing_file);
-+
- /* loop sysfs attributes */
-
- static ssize_t loop_attr_show(struct device *dev, char *page,
-diff --git a/fs/dcache.c b/fs/dcache.c
-index d6847d7..c3c0b6d 100644
---- a/fs/dcache.c
-+++ b/fs/dcache.c
-@@ -1202,7 +1202,7 @@ enum d_walk_ret {
- *
- * The @enter() and @finish() callbacks are called with d_lock held.
- */
--static void d_walk(struct dentry *parent, void *data,
-+void d_walk(struct dentry *parent, void *data,
- enum d_walk_ret (*enter)(void *, struct dentry *),
- void (*finish)(void *))
- {
-diff --git a/fs/fcntl.c b/fs/fcntl.c
-index 350a2c8..6f42279 100644
---- a/fs/fcntl.c
-+++ b/fs/fcntl.c
-@@ -29,7 +29,7 @@
-
- #define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
-
--static int setfl(int fd, struct file * filp, unsigned long arg)
-+int setfl(int fd, struct file * filp, unsigned long arg)
- {
- struct inode * inode = file_inode(filp);
- int error = 0;
-@@ -60,6 +60,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
-
- if (filp->f_op->check_flags)
- error = filp->f_op->check_flags(arg);
-+ if (!error && filp->f_op->setfl)
-+ error = filp->f_op->setfl(filp, arg);
- if (error)
- return error;
-
-diff --git a/fs/inode.c b/fs/inode.c
-index 4ccbc21..aa6d071 100644
---- a/fs/inode.c
-+++ b/fs/inode.c
-@@ -1591,7 +1591,7 @@ EXPORT_SYMBOL(generic_update_time);
- * This does the actual work of updating an inodes time or version. Must have
- * had called mnt_want_write() before calling this.
- */
--static int update_time(struct inode *inode, struct timespec *time, int flags)
-+int update_time(struct inode *inode, struct timespec *time, int flags)
- {
- int (*update_time)(struct inode *, struct timespec *, int);
-
-diff --git a/fs/read_write.c b/fs/read_write.c
-index 933b53a..2d13282 100644
---- a/fs/read_write.c
-+++ b/fs/read_write.c
-@@ -515,6 +515,28 @@ ssize_t __vfs_write(struct file *file, const char __user *p, size_t count,
- }
- EXPORT_SYMBOL(__vfs_write);
-
-+vfs_readf_t vfs_readf(struct file *file)
-+{
-+ const struct file_operations *fop = file->f_op;
-+
-+ if (fop->read)
-+ return fop->read;
-+ if (fop->read_iter)
-+ return new_sync_read;
-+ return ERR_PTR(-ENOSYS);
-+}
-+
-+vfs_writef_t vfs_writef(struct file *file)
-+{
-+ const struct file_operations *fop = file->f_op;
-+
-+ if (fop->write)
-+ return fop->write;
-+ if (fop->write_iter)
-+ return new_sync_write;
-+ return ERR_PTR(-ENOSYS);
-+}
-+
- ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t *pos)
- {
- mm_segment_t old_fs;
-diff --git a/fs/splice.c b/fs/splice.c
-index dd9bf7e..9326c2a 100644
---- a/fs/splice.c
-+++ b/fs/splice.c
-@@ -1111,8 +1111,8 @@ EXPORT_SYMBOL(generic_splice_sendpage);
- /*
- * Attempt to initiate a splice from pipe to file.
- */
--static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
-- loff_t *ppos, size_t len, unsigned int flags)
-+long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags)
- {
- ssize_t (*splice_write)(struct pipe_inode_info *, struct file *,
- loff_t *, size_t, unsigned int);
-@@ -1128,9 +1128,9 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
- /*
- * Attempt to initiate a splice from a file to a pipe.
- */
--static long do_splice_to(struct file *in, loff_t *ppos,
-- struct pipe_inode_info *pipe, size_t len,
-- unsigned int flags)
-+long do_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags)
- {
- ssize_t (*splice_read)(struct file *, loff_t *,
- struct pipe_inode_info *, size_t, unsigned int);
-diff --git a/include/linux/file.h b/include/linux/file.h
-index 7444f5f..bdac0be 100644
---- a/include/linux/file.h
-+++ b/include/linux/file.h
-@@ -19,6 +19,7 @@ struct dentry;
- struct path;
- extern struct file *alloc_file(struct path *, fmode_t mode,
- const struct file_operations *fop);
-+extern struct file *get_empty_filp(void);
-
- static inline void fput_light(struct file *file, int fput_needed)
- {
-diff --git a/include/linux/fs.h b/include/linux/fs.h
-index dd28814..b689a48 100644
---- a/include/linux/fs.h
-+++ b/include/linux/fs.h
-@@ -1306,6 +1306,7 @@ extern void fasync_free(struct fasync_struct *);
- /* can be called from interrupts */
- extern void kill_fasync(struct fasync_struct **, int, int);
-
-+extern int setfl(int fd, struct file * filp, unsigned long arg);
- extern void __f_setown(struct file *filp, struct pid *, enum pid_type, int force);
- extern void f_setown(struct file *filp, unsigned long arg, int force);
- extern void f_delown(struct file *filp);
-@@ -1690,6 +1691,7 @@ struct file_operations {
- ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
- unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
- int (*check_flags)(int);
-+ int (*setfl)(struct file *, unsigned long);
- int (*flock) (struct file *, int, struct file_lock *);
- ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
- ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
-@@ -1750,6 +1752,12 @@ ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
- struct iovec *fast_pointer,
- struct iovec **ret_pointer);
-
-+typedef ssize_t (*vfs_readf_t)(struct file *, char __user *, size_t, loff_t *);
-+typedef ssize_t (*vfs_writef_t)(struct file *, const char __user *, size_t,
-+ loff_t *);
-+vfs_readf_t vfs_readf(struct file *file);
-+vfs_writef_t vfs_writef(struct file *file);
-+
- extern ssize_t __vfs_read(struct file *, char __user *, size_t, loff_t *);
- extern ssize_t __vfs_write(struct file *, const char __user *, size_t, loff_t *);
- extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *);
-@@ -2105,6 +2113,7 @@ extern int current_umask(void);
- extern void ihold(struct inode * inode);
- extern void iput(struct inode *);
- extern int generic_update_time(struct inode *, struct timespec *, int);
-+extern int update_time(struct inode *, struct timespec *, int);
-
- /* /sys/fs */
- extern struct kobject *fs_kobj;
-diff --git a/include/linux/splice.h b/include/linux/splice.h
-index da2751d..2e0fca6 100644
---- a/include/linux/splice.h
-+++ b/include/linux/splice.h
-@@ -83,4 +83,10 @@ extern void splice_shrink_spd(struct splice_pipe_desc *);
- extern void spd_release_page(struct splice_pipe_desc *, unsigned int);
-
- extern const struct pipe_buf_operations page_cache_pipe_buf_ops;
-+
-+extern long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags);
-+extern long do_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags);
- #endif
-aufs4.7 mmap patch
-
-diff --git a/fs/proc/base.c b/fs/proc/base.c
-index a11eb71..8f10865 100644
---- a/fs/proc/base.c
-+++ b/fs/proc/base.c
-@@ -1939,7 +1939,7 @@ static int map_files_get_link(struct dentry *dentry, struct path *path)
- down_read(&mm->mmap_sem);
- vma = find_exact_vma(mm, vm_start, vm_end);
- if (vma && vma->vm_file) {
-- *path = vma->vm_file->f_path;
-+ *path = vma_pr_or_file(vma)->f_path;
- path_get(path);
- rc = 0;
- }
-diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
-index f8595e8..cb8eda0 100644
---- a/fs/proc/nommu.c
-+++ b/fs/proc/nommu.c
-@@ -45,7 +45,10 @@ static int nommu_region_show(struct seq_file *m, struct vm_region *region)
- file = region->vm_file;
-
- if (file) {
-- struct inode *inode = file_inode(region->vm_file);
-+ struct inode *inode;
-+
-+ file = vmr_pr_or_file(region);
-+ inode = file_inode(file);
- dev = inode->i_sb->s_dev;
- ino = inode->i_ino;
- }
-diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
-index 4648c7f..061cb85 100644
---- a/fs/proc/task_mmu.c
-+++ b/fs/proc/task_mmu.c
-@@ -298,7 +298,10 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
- const char *name = NULL;
-
- if (file) {
-- struct inode *inode = file_inode(vma->vm_file);
-+ struct inode *inode;
-+
-+ file = vma_pr_or_file(vma);
-+ inode = file_inode(file);
- dev = inode->i_sb->s_dev;
- ino = inode->i_ino;
- pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT;
-@@ -1624,7 +1627,7 @@ static int show_numa_map(struct seq_file *m, void *v, int is_pid)
- struct proc_maps_private *proc_priv = &numa_priv->proc_maps;
- struct vm_area_struct *vma = v;
- struct numa_maps *md = &numa_priv->md;
-- struct file *file = vma->vm_file;
-+ struct file *file = vma_pr_or_file(vma);
- struct mm_struct *mm = vma->vm_mm;
- struct mm_walk walk = {
- .hugetlb_entry = gather_hugetlb_stats,
-diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
-index faacb0c..17b43be 100644
---- a/fs/proc/task_nommu.c
-+++ b/fs/proc/task_nommu.c
-@@ -163,7 +163,10 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma,
- file = vma->vm_file;
-
- if (file) {
-- struct inode *inode = file_inode(vma->vm_file);
-+ struct inode *inode;
-+
-+ file = vma_pr_or_file(vma);
-+ inode = file_inode(file);
- dev = inode->i_sb->s_dev;
- ino = inode->i_ino;
- pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
-diff --git a/include/linux/mm.h b/include/linux/mm.h
-index ece042d..1e24513 100644
---- a/include/linux/mm.h
-+++ b/include/linux/mm.h
-@@ -1239,6 +1239,28 @@ static inline int fixup_user_fault(struct task_struct *tsk,
- }
- #endif
-
-+extern void vma_do_file_update_time(struct vm_area_struct *, const char[], int);
-+extern struct file *vma_do_pr_or_file(struct vm_area_struct *, const char[],
-+ int);
-+extern void vma_do_get_file(struct vm_area_struct *, const char[], int);
-+extern void vma_do_fput(struct vm_area_struct *, const char[], int);
-+
-+#define vma_file_update_time(vma) vma_do_file_update_time(vma, __func__, \
-+ __LINE__)
-+#define vma_pr_or_file(vma) vma_do_pr_or_file(vma, __func__, \
-+ __LINE__)
-+#define vma_get_file(vma) vma_do_get_file(vma, __func__, __LINE__)
-+#define vma_fput(vma) vma_do_fput(vma, __func__, __LINE__)
-+
-+#ifndef CONFIG_MMU
-+extern struct file *vmr_do_pr_or_file(struct vm_region *, const char[], int);
-+extern void vmr_do_fput(struct vm_region *, const char[], int);
-+
-+#define vmr_pr_or_file(region) vmr_do_pr_or_file(region, __func__, \
-+ __LINE__)
-+#define vmr_fput(region) vmr_do_fput(region, __func__, __LINE__)
-+#endif /* !CONFIG_MMU */
-+
- extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len, int write);
- extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
- void *buf, int len, int write);
-diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
-index ca3e517..10bc491 100644
---- a/include/linux/mm_types.h
-+++ b/include/linux/mm_types.h
-@@ -274,6 +274,7 @@ struct vm_region {
- unsigned long vm_top; /* region allocated to here */
- unsigned long vm_pgoff; /* the offset in vm_file corresponding to vm_start */
- struct file *vm_file; /* the backing file or NULL */
-+ struct file *vm_prfile; /* the virtual backing file or NULL */
-
- int vm_usage; /* region usage count (access under nommu_region_sem) */
- bool vm_icache_flushed : 1; /* true if the icache has been flushed for
-@@ -348,6 +349,7 @@ struct vm_area_struct {
- unsigned long vm_pgoff; /* Offset (within vm_file) in PAGE_SIZE
- units */
- struct file * vm_file; /* File we map to (can be NULL). */
-+ struct file *vm_prfile; /* shadow of vm_file */
- void * vm_private_data; /* was vm_pte (shared mem) */
-
- #ifndef CONFIG_MMU
-diff --git a/kernel/fork.c b/kernel/fork.c
-index 4a7ec0c..8c8f7ac 100644
---- a/kernel/fork.c
-+++ b/kernel/fork.c
-@@ -479,7 +479,7 @@ static int dup_mmap(struct mm_struct *mm, struct mm_struct *oldmm)
- struct inode *inode = file_inode(file);
- struct address_space *mapping = file->f_mapping;
-
-- get_file(file);
-+ vma_get_file(tmp);
- if (tmp->vm_flags & VM_DENYWRITE)
- atomic_dec(&inode->i_writecount);
- i_mmap_lock_write(mapping);
-diff --git a/mm/Makefile b/mm/Makefile
-index 78c6f7d..aea4230 100644
---- a/mm/Makefile
-+++ b/mm/Makefile
-@@ -37,7 +37,7 @@ obj-y := filemap.o mempool.o oom_kill.o \
- mm_init.o mmu_context.o percpu.o slab_common.o \
- compaction.o vmacache.o \
- interval_tree.o list_lru.o workingset.o \
-- debug.o $(mmu-y)
-+ prfile.o debug.o $(mmu-y)
-
- obj-y += init-mm.o
-
-diff --git a/mm/filemap.c b/mm/filemap.c
-index 20f3b1f..ee827ce 100644
---- a/mm/filemap.c
-+++ b/mm/filemap.c
-@@ -2208,7 +2208,7 @@ int filemap_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf)
- int ret = VM_FAULT_LOCKED;
-
- sb_start_pagefault(inode->i_sb);
-- file_update_time(vma->vm_file);
-+ vma_file_update_time(vma);
- lock_page(page);
- if (page->mapping != inode->i_mapping) {
- unlock_page(page);
-diff --git a/mm/memory.c b/mm/memory.c
-index 9e04681..06980d1 100644
---- a/mm/memory.c
-+++ b/mm/memory.c
-@@ -2100,7 +2100,7 @@ static inline int wp_page_reuse(struct mm_struct *mm,
- }
-
- if (!page_mkwrite)
-- file_update_time(vma->vm_file);
-+ vma_file_update_time(vma);
- }
-
- return VM_FAULT_WRITE;
-diff --git a/mm/mmap.c b/mm/mmap.c
-index de2c176..b7f391c 100644
---- a/mm/mmap.c
-+++ b/mm/mmap.c
-@@ -162,7 +162,7 @@ static struct vm_area_struct *remove_vma(struct vm_area_struct *vma)
- if (vma->vm_ops && vma->vm_ops->close)
- vma->vm_ops->close(vma);
- if (vma->vm_file)
-- fput(vma->vm_file);
-+ vma_fput(vma);
- mpol_put(vma_policy(vma));
- kmem_cache_free(vm_area_cachep, vma);
- return next;
-@@ -782,7 +782,7 @@ again: remove_next = 1 + (end > next->vm_end);
- if (remove_next) {
- if (file) {
- uprobe_munmap(next, next->vm_start, next->vm_end);
-- fput(file);
-+ vma_fput(vma);
- }
- if (next->anon_vma)
- anon_vma_merge(vma, next);
-@@ -1563,8 +1563,8 @@ out:
- return addr;
-
- unmap_and_free_vma:
-+ vma_fput(vma);
- vma->vm_file = NULL;
-- fput(file);
-
- /* Undo any partial mapping done by a device driver. */
- unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end);
-@@ -2358,7 +2358,7 @@ static int __split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
- goto out_free_mpol;
-
- if (new->vm_file)
-- get_file(new->vm_file);
-+ vma_get_file(new);
-
- if (new->vm_ops && new->vm_ops->open)
- new->vm_ops->open(new);
-@@ -2377,7 +2377,7 @@ static int __split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
- if (new->vm_ops && new->vm_ops->close)
- new->vm_ops->close(new);
- if (new->vm_file)
-- fput(new->vm_file);
-+ vma_fput(new);
- unlink_anon_vmas(new);
- out_free_mpol:
- mpol_put(vma_policy(new));
-@@ -2528,7 +2528,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
- struct vm_area_struct *vma;
- unsigned long populate = 0;
- unsigned long ret = -EINVAL;
-- struct file *file;
-+ struct file *file, *prfile;
-
- pr_warn_once("%s (%d) uses deprecated remap_file_pages() syscall. See Documentation/vm/remap_file_pages.txt.\n",
- current->comm, current->pid);
-@@ -2597,10 +2597,27 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
- }
- }
-
-- file = get_file(vma->vm_file);
-+ vma_get_file(vma);
-+ file = vma->vm_file;
-+ prfile = vma->vm_prfile;
- ret = do_mmap_pgoff(vma->vm_file, start, size,
- prot, flags, pgoff, &populate);
-+ if (!IS_ERR_VALUE(ret) && file && prfile) {
-+ struct vm_area_struct *new_vma;
-+
-+ new_vma = find_vma(mm, ret);
-+ if (!new_vma->vm_prfile)
-+ new_vma->vm_prfile = prfile;
-+ if (new_vma != vma)
-+ get_file(prfile);
-+ }
-+ /*
-+ * two fput()s instead of vma_fput(vma),
-+ * coz vma may not be available anymore.
-+ */
- fput(file);
-+ if (prfile)
-+ fput(prfile);
- out:
- up_write(&mm->mmap_sem);
- if (populate)
-@@ -2873,7 +2890,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
- if (anon_vma_clone(new_vma, vma))
- goto out_free_mempol;
- if (new_vma->vm_file)
-- get_file(new_vma->vm_file);
-+ vma_get_file(new_vma);
- if (new_vma->vm_ops && new_vma->vm_ops->open)
- new_vma->vm_ops->open(new_vma);
- vma_link(mm, new_vma, prev, rb_link, rb_parent);
-diff --git a/mm/nommu.c b/mm/nommu.c
-index c2e588802..c39edc4 100644
---- a/mm/nommu.c
-+++ b/mm/nommu.c
-@@ -644,7 +644,7 @@ static void __put_nommu_region(struct vm_region *region)
- up_write(&nommu_region_sem);
-
- if (region->vm_file)
-- fput(region->vm_file);
-+ vmr_fput(region);
-
- /* IO memory and memory shared directly out of the pagecache
- * from ramfs/tmpfs mustn't be released here */
-@@ -802,7 +802,7 @@ static void delete_vma(struct mm_struct *mm, struct vm_area_struct *vma)
- if (vma->vm_ops && vma->vm_ops->close)
- vma->vm_ops->close(vma);
- if (vma->vm_file)
-- fput(vma->vm_file);
-+ vma_fput(vma);
- put_nommu_region(vma->vm_region);
- kmem_cache_free(vm_area_cachep, vma);
- }
-@@ -1328,7 +1328,7 @@ unsigned long do_mmap(struct file *file,
- goto error_just_free;
- }
- }
-- fput(region->vm_file);
-+ vmr_fput(region);
- kmem_cache_free(vm_region_jar, region);
- region = pregion;
- result = start;
-@@ -1403,10 +1403,10 @@ error_just_free:
- up_write(&nommu_region_sem);
- error:
- if (region->vm_file)
-- fput(region->vm_file);
-+ vmr_fput(region);
- kmem_cache_free(vm_region_jar, region);
- if (vma->vm_file)
-- fput(vma->vm_file);
-+ vma_fput(vma);
- kmem_cache_free(vm_area_cachep, vma);
- return ret;
-
-diff --git a/mm/prfile.c b/mm/prfile.c
-new file mode 100644
-index 0000000..b323b8a
---- /dev/null
-+++ b/mm/prfile.c
-@@ -0,0 +1,86 @@
-+/*
-+ * Mainly for aufs which mmap(2) diffrent file and wants to print different path
-+ * in /proc/PID/maps.
-+ * Call these functions via macros defined in linux/mm.h.
-+ *
-+ * See Documentation/filesystems/aufs/design/06mmap.txt
-+ *
-+ * Copyright (c) 2014 Junjro R. Okajima
-+ * Copyright (c) 2014 Ian Campbell
-+ */
-+
-+#include <linux/mm.h>
-+#include <linux/file.h>
-+#include <linux/fs.h>
-+
-+/* #define PRFILE_TRACE */
-+static inline void prfile_trace(struct file *f, struct file *pr,
-+ const char func[], int line, const char func2[])
-+{
-+#ifdef PRFILE_TRACE
-+ if (pr)
-+ pr_info("%s:%d: %s, %s\n", func, line, func2,
-+ f ? (char *)f->f_path.dentry->d_name.name : "(null)");
-+#endif
-+}
-+
-+void vma_do_file_update_time(struct vm_area_struct *vma, const char func[],
-+ int line)
-+{
-+ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ file_update_time(f);
-+ if (f && pr)
-+ file_update_time(pr);
-+}
-+
-+struct file *vma_do_pr_or_file(struct vm_area_struct *vma, const char func[],
-+ int line)
-+{
-+ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ return (f && pr) ? pr : f;
-+}
-+
-+void vma_do_get_file(struct vm_area_struct *vma, const char func[], int line)
-+{
-+ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ get_file(f);
-+ if (f && pr)
-+ get_file(pr);
-+}
-+
-+void vma_do_fput(struct vm_area_struct *vma, const char func[], int line)
-+{
-+ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ fput(f);
-+ if (f && pr)
-+ fput(pr);
-+}
-+
-+#ifndef CONFIG_MMU
-+struct file *vmr_do_pr_or_file(struct vm_region *region, const char func[],
-+ int line)
-+{
-+ struct file *f = region->vm_file, *pr = region->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ return (f && pr) ? pr : f;
-+}
-+
-+void vmr_do_fput(struct vm_region *region, const char func[], int line)
-+{
-+ struct file *f = region->vm_file, *pr = region->vm_prfile;
-+
-+ prfile_trace(f, pr, func, line, __func__);
-+ fput(f);
-+ if (f && pr)
-+ fput(pr);
-+}
-+#endif /* !CONFIG_MMU */
-aufs4.7 standalone patch
-
-diff --git a/fs/dcache.c b/fs/dcache.c
-index c3c0b6d..c99d2d2 100644
---- a/fs/dcache.c
-+++ b/fs/dcache.c
-@@ -1310,6 +1310,7 @@ rename_retry:
- seq = 1;
- goto again;
- }
-+EXPORT_SYMBOL_GPL(d_walk);
-
- /*
- * Search for at least 1 mount point in the dentry's subdirs.
-diff --git a/fs/exec.c b/fs/exec.c
-index 887c1c9..40e8767 100644
---- a/fs/exec.c
-+++ b/fs/exec.c
-@@ -104,6 +104,7 @@ bool path_noexec(const struct path *path)
- return (path->mnt->mnt_flags & MNT_NOEXEC) ||
- (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC);
- }
-+EXPORT_SYMBOL_GPL(path_noexec);
-
- #ifdef CONFIG_USELIB
- /*
-diff --git a/fs/fcntl.c b/fs/fcntl.c
-index 6f42279..04fd33c 100644
---- a/fs/fcntl.c
-+++ b/fs/fcntl.c
-@@ -82,6 +82,7 @@ int setfl(int fd, struct file * filp, unsigned long arg)
- out:
- return error;
- }
-+EXPORT_SYMBOL_GPL(setfl);
-
- static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
- int force)
-diff --git a/fs/file_table.c b/fs/file_table.c
-index ad17e05..ae9f267 100644
---- a/fs/file_table.c
-+++ b/fs/file_table.c
-@@ -147,6 +147,7 @@ over:
- }
- return ERR_PTR(-ENFILE);
- }
-+EXPORT_SYMBOL_GPL(get_empty_filp);
-
- /**
- * alloc_file - allocate and initialize a 'struct file'
-@@ -258,6 +259,7 @@ void flush_delayed_fput(void)
- {
- delayed_fput(NULL);
- }
-+EXPORT_SYMBOL_GPL(flush_delayed_fput);
-
- static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput);
-
-@@ -300,6 +302,7 @@ void __fput_sync(struct file *file)
- }
-
- EXPORT_SYMBOL(fput);
-+EXPORT_SYMBOL_GPL(__fput_sync);
-
- void put_filp(struct file *file)
- {
-@@ -308,6 +311,7 @@ void put_filp(struct file *file)
- file_free(file);
- }
- }
-+EXPORT_SYMBOL_GPL(put_filp);
-
- void __init files_init(void)
- {
-diff --git a/fs/inode.c b/fs/inode.c
-index aa6d071..f336032 100644
---- a/fs/inode.c
-+++ b/fs/inode.c
-@@ -1600,6 +1600,7 @@ int update_time(struct inode *inode, struct timespec *time, int flags)
-
- return update_time(inode, time, flags);
- }
-+EXPORT_SYMBOL_GPL(update_time);
-
- /**
- * touch_atime - update the access time
-diff --git a/fs/namespace.c b/fs/namespace.c
-index 419f746..9c0e0af 100644
---- a/fs/namespace.c
-+++ b/fs/namespace.c
-@@ -463,6 +463,7 @@ void __mnt_drop_write(struct vfsmount *mnt)
- mnt_dec_writers(real_mount(mnt));
- preempt_enable();
- }
-+EXPORT_SYMBOL_GPL(__mnt_drop_write);
-
- /**
- * mnt_drop_write - give up write access to a mount
-@@ -1812,6 +1813,7 @@ int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg,
- }
- return 0;
- }
-+EXPORT_SYMBOL_GPL(iterate_mounts);
-
- static void cleanup_group_ids(struct mount *mnt, struct mount *end)
- {
-diff --git a/fs/notify/group.c b/fs/notify/group.c
-index 3e2dd85..b17cb4b 100644
---- a/fs/notify/group.c
-+++ b/fs/notify/group.c
-@@ -22,6 +22,7 @@
- #include <linux/srcu.h>
- #include <linux/rculist.h>
- #include <linux/wait.h>
-+#include <linux/module.h>
-
- #include <linux/fsnotify_backend.h>
- #include "fsnotify.h"
-@@ -81,6 +82,7 @@ void fsnotify_get_group(struct fsnotify_group *group)
- {
- atomic_inc(&group->refcnt);
- }
-+EXPORT_SYMBOL_GPL(fsnotify_get_group);
-
- /*
- * Drop a reference to a group. Free it if it's through.
-@@ -90,6 +92,7 @@ void fsnotify_put_group(struct fsnotify_group *group)
- if (atomic_dec_and_test(&group->refcnt))
- fsnotify_final_destroy_group(group);
- }
-+EXPORT_SYMBOL_GPL(fsnotify_put_group);
-
- /*
- * Create a new fsnotify_group and hold a reference for the group returned.
-@@ -118,6 +121,7 @@ struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
-
- return group;
- }
-+EXPORT_SYMBOL_GPL(fsnotify_alloc_group);
-
- int fsnotify_fasync(int fd, struct file *file, int on)
- {
-diff --git a/fs/notify/mark.c b/fs/notify/mark.c
-index d3fea0b..5fc06ad 100644
---- a/fs/notify/mark.c
-+++ b/fs/notify/mark.c
-@@ -113,6 +113,7 @@ void fsnotify_put_mark(struct fsnotify_mark *mark)
- mark->free_mark(mark);
- }
- }
-+EXPORT_SYMBOL_GPL(fsnotify_put_mark);
-
- /* Calculate mask of events for a list of marks */
- u32 fsnotify_recalc_mask(struct hlist_head *head)
-@@ -230,6 +231,7 @@ void fsnotify_destroy_mark(struct fsnotify_mark *mark,
- mutex_unlock(&group->mark_mutex);
- fsnotify_free_mark(mark);
- }
-+EXPORT_SYMBOL_GPL(fsnotify_destroy_mark);
-
- void fsnotify_destroy_marks(struct hlist_head *head, spinlock_t *lock)
- {
-@@ -415,6 +417,7 @@ err:
-
- return ret;
- }
-+EXPORT_SYMBOL_GPL(fsnotify_add_mark);
-
- int fsnotify_add_mark(struct fsnotify_mark *mark, struct fsnotify_group *group,
- struct inode *inode, struct vfsmount *mnt, int allow_dups)
-@@ -533,6 +536,7 @@ void fsnotify_init_mark(struct fsnotify_mark *mark,
- atomic_set(&mark->refcnt, 1);
- mark->free_mark = free_mark;
- }
-+EXPORT_SYMBOL_GPL(fsnotify_init_mark);
-
- /*
- * Destroy all marks in destroy_list, waits for SRCU period to finish before
-diff --git a/fs/open.c b/fs/open.c
-index 93ae3cd..d25b9bd 100644
---- a/fs/open.c
-+++ b/fs/open.c
-@@ -64,6 +64,7 @@ int do_truncate(struct dentry *dentry, loff_t length, unsigned int time_attrs,
- inode_unlock(dentry->d_inode);
- return ret;
- }
-+EXPORT_SYMBOL_GPL(do_truncate);
-
- long vfs_truncate(const struct path *path, loff_t length)
- {
-@@ -678,6 +679,7 @@ int open_check_o_direct(struct file *f)
- }
- return 0;
- }
-+EXPORT_SYMBOL_GPL(open_check_o_direct);
-
- static int do_dentry_open(struct file *f,
- struct inode *inode,
-diff --git a/fs/read_write.c b/fs/read_write.c
-index 2d13282..260c4a4 100644
---- a/fs/read_write.c
-+++ b/fs/read_write.c
-@@ -525,6 +525,7 @@ vfs_readf_t vfs_readf(struct file *file)
- return new_sync_read;
- return ERR_PTR(-ENOSYS);
- }
-+EXPORT_SYMBOL_GPL(vfs_readf);
-
- vfs_writef_t vfs_writef(struct file *file)
- {
-@@ -536,6 +537,7 @@ vfs_writef_t vfs_writef(struct file *file)
- return new_sync_write;
- return ERR_PTR(-ENOSYS);
- }
-+EXPORT_SYMBOL_GPL(vfs_writef);
-
- ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t *pos)
- {
-diff --git a/fs/splice.c b/fs/splice.c
-index 9326c2a..0606690 100644
---- a/fs/splice.c
-+++ b/fs/splice.c
-@@ -1124,6 +1124,7 @@ long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
-
- return splice_write(pipe, out, ppos, len, flags);
- }
-+EXPORT_SYMBOL_GPL(do_splice_from);
-
- /*
- * Attempt to initiate a splice from a file to a pipe.
-@@ -1153,6 +1154,7 @@ long do_splice_to(struct file *in, loff_t *ppos,
-
- return splice_read(in, ppos, pipe, len, flags);
- }
-+EXPORT_SYMBOL_GPL(do_splice_to);
-
- /**
- * splice_direct_to_actor - splices data directly between two non-pipes
-diff --git a/fs/xattr.c b/fs/xattr.c
-index 4beafc4..e118715 100644
---- a/fs/xattr.c
-+++ b/fs/xattr.c
-@@ -207,6 +207,7 @@ vfs_getxattr_alloc(struct dentry *dentry, const char *name, char **xattr_value,
- *xattr_value = value;
- return error;
- }
-+EXPORT_SYMBOL_GPL(vfs_getxattr_alloc);
-
- ssize_t
- vfs_getxattr(struct dentry *dentry, const char *name, void *value, size_t size)
-diff --git a/kernel/task_work.c b/kernel/task_work.c
-index 53fa971..bce3211 100644
---- a/kernel/task_work.c
-+++ b/kernel/task_work.c
-@@ -118,3 +118,4 @@ void task_work_run(void)
- } while (work);
- }
- }
-+EXPORT_SYMBOL_GPL(task_work_run);
-diff --git a/security/commoncap.c b/security/commoncap.c
-index e7fadde..6423e53 100644
---- a/security/commoncap.c
-+++ b/security/commoncap.c
-@@ -1058,12 +1058,14 @@ int cap_mmap_addr(unsigned long addr)
- }
- return ret;
- }
-+EXPORT_SYMBOL_GPL(cap_mmap_addr);
-
- int cap_mmap_file(struct file *file, unsigned long reqprot,
- unsigned long prot, unsigned long flags)
- {
- return 0;
- }
-+EXPORT_SYMBOL_GPL(cap_mmap_file);
-
- #ifdef CONFIG_SECURITY
-
-diff --git a/security/device_cgroup.c b/security/device_cgroup.c
-index 03c1652..f88c84b 100644
---- a/security/device_cgroup.c
-+++ b/security/device_cgroup.c
-@@ -7,6 +7,7 @@
- #include <linux/device_cgroup.h>
- #include <linux/cgroup.h>
- #include <linux/ctype.h>
-+#include <linux/export.h>
- #include <linux/list.h>
- #include <linux/uaccess.h>
- #include <linux/seq_file.h>
-@@ -849,6 +850,7 @@ int __devcgroup_inode_permission(struct inode *inode, int mask)
- return __devcgroup_check_permission(type, imajor(inode), iminor(inode),
- access);
- }
-+EXPORT_SYMBOL_GPL(__devcgroup_inode_permission);
-
- int devcgroup_inode_mknod(int mode, dev_t dev)
- {
-diff --git a/security/security.c b/security/security.c
-index 7095693..da7fe2c 100644
---- a/security/security.c
-+++ b/security/security.c
-@@ -434,6 +434,7 @@ int security_path_rmdir(const struct path *dir, struct dentry *dentry)
- return 0;
- return call_int_hook(path_rmdir, 0, dir, dentry);
- }
-+EXPORT_SYMBOL_GPL(security_path_rmdir);
-
- int security_path_unlink(const struct path *dir, struct dentry *dentry)
- {
-@@ -450,6 +451,7 @@ int security_path_symlink(const struct path *dir, struct dentry *dentry,
- return 0;
- return call_int_hook(path_symlink, 0, dir, dentry, old_name);
- }
-+EXPORT_SYMBOL_GPL(security_path_symlink);
-
- int security_path_link(struct dentry *old_dentry, const struct path *new_dir,
- struct dentry *new_dentry)
-@@ -458,6 +460,7 @@ int security_path_link(struct dentry *old_dentry, const struct path *new_dir,
- return 0;
- return call_int_hook(path_link, 0, old_dentry, new_dir, new_dentry);
- }
-+EXPORT_SYMBOL_GPL(security_path_link);
-
- int security_path_rename(const struct path *old_dir, struct dentry *old_dentry,
- const struct path *new_dir, struct dentry *new_dentry,
-@@ -485,6 +488,7 @@ int security_path_truncate(const struct path *path)
- return 0;
- return call_int_hook(path_truncate, 0, path);
- }
-+EXPORT_SYMBOL_GPL(security_path_truncate);
-
- int security_path_chmod(const struct path *path, umode_t mode)
- {
-@@ -492,6 +496,7 @@ int security_path_chmod(const struct path *path, umode_t mode)
- return 0;
- return call_int_hook(path_chmod, 0, path, mode);
- }
-+EXPORT_SYMBOL_GPL(security_path_chmod);
-
- int security_path_chown(const struct path *path, kuid_t uid, kgid_t gid)
- {
-@@ -499,6 +504,7 @@ int security_path_chown(const struct path *path, kuid_t uid, kgid_t gid)
- return 0;
- return call_int_hook(path_chown, 0, path, uid, gid);
- }
-+EXPORT_SYMBOL_GPL(security_path_chown);
-
- int security_path_chroot(const struct path *path)
- {
-@@ -584,6 +590,7 @@ int security_inode_readlink(struct dentry *dentry)
- return 0;
- return call_int_hook(inode_readlink, 0, dentry);
- }
-+EXPORT_SYMBOL_GPL(security_inode_readlink);
-
- int security_inode_follow_link(struct dentry *dentry, struct inode *inode,
- bool rcu)
-@@ -599,6 +606,7 @@ int security_inode_permission(struct inode *inode, int mask)
- return 0;
- return call_int_hook(inode_permission, 0, inode, mask);
- }
-+EXPORT_SYMBOL_GPL(security_inode_permission);
-
- int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
- {
-@@ -737,6 +745,7 @@ int security_file_permission(struct file *file, int mask)
-
- return fsnotify_perm(file, mask);
- }
-+EXPORT_SYMBOL_GPL(security_file_permission);
-
- int security_file_alloc(struct file *file)
- {
-@@ -796,6 +805,7 @@ int security_mmap_file(struct file *file, unsigned long prot,
- return ret;
- return ima_file_mmap(file, prot);
- }
-+EXPORT_SYMBOL_GPL(security_mmap_file);
-
- int security_mmap_addr(unsigned long addr)
- {
-diff -urN /usr/share/empty/Documentation/ABI/testing/debugfs-aufs linux/Documentation/ABI/testing/debugfs-aufs
---- /usr/share/empty/Documentation/ABI/testing/debugfs-aufs 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/ABI/testing/debugfs-aufs 2015-09-24 10:47:58.244719488 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/ABI/testing/debugfs-aufs linux-4.14/Documentation/ABI/testing/debugfs-aufs
+--- linux-4.14/Documentation/ABI/testing/debugfs-aufs 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/ABI/testing/debugfs-aufs 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,50 @@
+What: /debug/aufs/si_<id>/
+Date: March 2009
+ be created.
+ When the aufs mount option 'noxino' is specified, it
+ will be empty. About XINO files, see the aufs manual.
-diff -urN /usr/share/empty/Documentation/ABI/testing/sysfs-aufs linux/Documentation/ABI/testing/sysfs-aufs
---- /usr/share/empty/Documentation/ABI/testing/sysfs-aufs 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/ABI/testing/sysfs-aufs 2015-09-24 10:47:58.244719488 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/ABI/testing/sysfs-aufs linux-4.14/Documentation/ABI/testing/sysfs-aufs
+--- linux-4.14/Documentation/ABI/testing/sysfs-aufs 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/ABI/testing/sysfs-aufs 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,31 @@
+What: /sys/fs/aufs/si_<id>/
+Date: March 2009
+ even if it is the default path.
+ When the aufs mount option 'noxino' is specified, it
+ will be empty. About XINO files, see the aufs manual.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/01intro.txt linux/Documentation/filesystems/aufs/design/01intro.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/01intro.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/01intro.txt 2016-07-25 19:05:34.807826401 +0200
-@@ -0,0 +1,170 @@
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/README linux-4.14/Documentation/filesystems/aufs/README
+--- linux-4.14/Documentation/filesystems/aufs/README 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/README 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,393 @@
++
++Aufs4 -- advanced multi layered unification filesystem version 4.x
++http://aufs.sf.net
++Junjiro R. Okajima
++
++
++0. Introduction
++----------------------------------------
++In the early days, aufs was entirely re-designed and re-implemented
++Unionfs Version 1.x series. Adding many original ideas, approaches,
++improvements and implementations, it becomes totally different from
++Unionfs while keeping the basic features.
++Recently, Unionfs Version 2.x series begin taking some of the same
++approaches to aufs1's.
++Unionfs is being developed by Professor Erez Zadok at Stony Brook
++University and his team.
++
++Aufs4 supports linux-4.0 and later, and for linux-3.x series try aufs3.
++If you want older kernel version support, try aufs2-2.6.git or
++aufs2-standalone.git repository, aufs1 from CVS on SourceForge.
++
++Note: it becomes clear that "Aufs was rejected. Let's give it up."
++ According to Christoph Hellwig, linux rejects all union-type
++ filesystems but UnionMount.
++<http://marc.info/?l=linux-kernel&m=123938533724484&w=2>
++
++PS. Al Viro seems have a plan to merge aufs as well as overlayfs and
++ UnionMount, and he pointed out an issue around a directory mutex
++ lock and aufs addressed it. But it is still unsure whether aufs will
++ be merged (or any other union solution).
++<http://marc.info/?l=linux-kernel&m=136312705029295&w=1>
++
++
++1. Features
++----------------------------------------
++- unite several directories into a single virtual filesystem. The member
++ directory is called as a branch.
++- you can specify the permission flags to the branch, which are 'readonly',
++ 'readwrite' and 'whiteout-able.'
++- by upper writable branch, internal copyup and whiteout, files/dirs on
++ readonly branch are modifiable logically.
++- dynamic branch manipulation, add, del.
++- etc...
++
++Also there are many enhancements in aufs, such as:
++- test only the highest one for the directory permission (dirperm1)
++- copyup on open (coo=)
++- 'move' policy for copy-up between two writable branches, after
++ checking free space.
++- xattr, acl
++- readdir(3) in userspace.
++- keep inode number by external inode number table
++- keep the timestamps of file/dir in internal copyup operation
++- seekable directory, supporting NFS readdir.
++- whiteout is hardlinked in order to reduce the consumption of inodes
++ on branch
++- do not copyup, nor create a whiteout when it is unnecessary
++- revert a single systemcall when an error occurs in aufs
++- remount interface instead of ioctl
++- maintain /etc/mtab by an external command, /sbin/mount.aufs.
++- loopback mounted filesystem as a branch
++- kernel thread for removing the dir who has a plenty of whiteouts
++- support copyup sparse file (a file which has a 'hole' in it)
++- default permission flags for branches
++- selectable permission flags for ro branch, whether whiteout can
++ exist or not
++- export via NFS.
++- support <sysfs>/fs/aufs and <debugfs>/aufs.
++- support multiple writable branches, some policies to select one
++ among multiple writable branches.
++- a new semantics for link(2) and rename(2) to support multiple
++ writable branches.
++- no glibc changes are required.
++- pseudo hardlink (hardlink over branches)
++- allow a direct access manually to a file on branch, e.g. bypassing aufs.
++ including NFS or remote filesystem branch.
++- userspace wrapper for pathconf(3)/fpathconf(3) with _PC_LINK_MAX.
++- and more...
++
++Currently these features are dropped temporary from aufs4.
++See design/08plan.txt in detail.
++- nested mount, i.e. aufs as readonly no-whiteout branch of another aufs
++ (robr)
++- statistics of aufs thread (/sys/fs/aufs/stat)
++
++Features or just an idea in the future (see also design/*.txt),
++- reorder the branch index without del/re-add.
++- permanent xino files for NFSD
++- an option for refreshing the opened files after add/del branches
++- light version, without branch manipulation. (unnecessary?)
++- copyup in userspace
++- inotify in userspace
++- readv/writev
++
++
++2. Download
++----------------------------------------
++There are three GIT trees for aufs4, aufs4-linux.git,
++aufs4-standalone.git, and aufs-util.git. Note that there is no "4" in
++"aufs-util.git."
++While the aufs-util is always necessary, you need either of aufs4-linux
++or aufs4-standalone.
++
++The aufs4-linux tree includes the whole linux mainline GIT tree,
++git://git.kernel.org/.../torvalds/linux.git.
++And you cannot select CONFIG_AUFS_FS=m for this version, eg. you cannot
++build aufs4 as an external kernel module.
++Several extra patches are not included in this tree. Only
++aufs4-standalone tree contains them. They are described in the later
++section "Configuration and Compilation."
++
++On the other hand, the aufs4-standalone tree has only aufs source files
++and necessary patches, and you can select CONFIG_AUFS_FS=m.
++But you need to apply all aufs patches manually.
++
++You will find GIT branches whose name is in form of "aufs4.x" where "x"
++represents the linux kernel version, "linux-4.x". For instance,
++"aufs4.0" is for linux-4.0. For latest "linux-4.x-rcN", use
++"aufs4.x-rcN" branch.
++
++o aufs4-linux tree
++$ git clone --reference /your/linux/git/tree \
++ git://github.com/sfjro/aufs4-linux.git aufs4-linux.git
++- if you don't have linux GIT tree, then remove "--reference ..."
++$ cd aufs4-linux.git
++$ git checkout origin/aufs4.0
++
++Or You may want to directly git-pull aufs into your linux GIT tree, and
++leave the patch-work to GIT.
++$ cd /your/linux/git/tree
++$ git remote add aufs4 git://github.com/sfjro/aufs4-linux.git
++$ git fetch aufs4
++$ git checkout -b my4.0 v4.0
++$ (add your local change...)
++$ git pull aufs4 aufs4.0
++- now you have v4.0 + your_changes + aufs4.0 in you my4.0 branch.
++- you may need to solve some conflicts between your_changes and
++ aufs4.0. in this case, git-rerere is recommended so that you can
++ solve the similar conflicts automatically when you upgrade to 4.1 or
++ later in the future.
++
++o aufs4-standalone tree
++$ git clone git://github.com/sfjro/aufs4-standalone.git aufs4-standalone.git
++$ cd aufs4-standalone.git
++$ git checkout origin/aufs4.0
++
++o aufs-util tree
++$ git clone git://git.code.sf.net/p/aufs/aufs-util aufs-util.git
++- note that the public aufs-util.git is on SourceForge instead of
++ GitHUB.
++$ cd aufs-util.git
++$ git checkout origin/aufs4.0
++
++Note: The 4.x-rcN branch is to be used with `rc' kernel versions ONLY.
++The minor version number, 'x' in '4.x', of aufs may not always
++follow the minor version number of the kernel.
++Because changes in the kernel that cause the use of a new
++minor version number do not always require changes to aufs-util.
++
++Since aufs-util has its own minor version number, you may not be
++able to find a GIT branch in aufs-util for your kernel's
++exact minor version number.
++In this case, you should git-checkout the branch for the
++nearest lower number.
++
++For (an unreleased) example:
++If you are using "linux-4.10" and the "aufs4.10" branch
++does not exist in aufs-util repository, then "aufs4.9", "aufs4.8"
++or something numerically smaller is the branch for your kernel.
++
++Also you can view all branches by
++ $ git branch -a
++
++
++3. Configuration and Compilation
++----------------------------------------
++Make sure you have git-checkout'ed the correct branch.
++
++For aufs4-linux tree,
++- enable CONFIG_AUFS_FS.
++- set other aufs configurations if necessary.
++
++For aufs4-standalone tree,
++There are several ways to build.
++
++1.
++- apply ./aufs4-kbuild.patch to your kernel source files.
++- apply ./aufs4-base.patch too.
++- apply ./aufs4-mmap.patch too.
++- apply ./aufs4-standalone.patch too, if you have a plan to set
++ CONFIG_AUFS_FS=m. otherwise you don't need ./aufs4-standalone.patch.
++- copy ./{Documentation,fs,include/uapi/linux/aufs_type.h} files to your
++ kernel source tree. Never copy $PWD/include/uapi/linux/Kbuild.
++- enable CONFIG_AUFS_FS, you can select either
++ =m or =y.
++- and build your kernel as usual.
++- install the built kernel.
++ Note: Since linux-3.9, every filesystem module requires an alias
++ "fs-<fsname>". You should make sure that "fs-aufs" is listed in your
++ modules.aliases file if you set CONFIG_AUFS_FS=m.
++- install the header files too by "make headers_install" to the
++ directory where you specify. By default, it is $PWD/usr.
++ "make help" shows a brief note for headers_install.
++- and reboot your system.
++
++2.
++- module only (CONFIG_AUFS_FS=m).
++- apply ./aufs4-base.patch to your kernel source files.
++- apply ./aufs4-mmap.patch too.
++- apply ./aufs4-standalone.patch too.
++- build your kernel, don't forget "make headers_install", and reboot.
++- edit ./config.mk and set other aufs configurations if necessary.
++ Note: You should read $PWD/fs/aufs/Kconfig carefully which describes
++ every aufs configurations.
++- build the module by simple "make".
++ Note: Since linux-3.9, every filesystem module requires an alias
++ "fs-<fsname>". You should make sure that "fs-aufs" is listed in your
++ modules.aliases file.
++- you can specify ${KDIR} make variable which points to your kernel
++ source tree.
++- install the files
++ + run "make install" to install the aufs module, or copy the built
++ $PWD/aufs.ko to /lib/modules/... and run depmod -a (or reboot simply).
++ + run "make install_headers" (instead of headers_install) to install
++ the modified aufs header file (you can specify DESTDIR which is
++ available in aufs standalone version's Makefile only), or copy
++ $PWD/usr/include/linux/aufs_type.h to /usr/include/linux or wherever
++ you like manually. By default, the target directory is $PWD/usr.
++- no need to apply aufs4-kbuild.patch, nor copying source files to your
++ kernel source tree.
++
++Note: The header file aufs_type.h is necessary to build aufs-util
++ as well as "make headers_install" in the kernel source tree.
++ headers_install is subject to be forgotten, but it is essentially
++ necessary, not only for building aufs-util.
++ You may not meet problems without headers_install in some older
++ version though.
++
++And then,
++- read README in aufs-util, build and install it
++- note that your distribution may contain an obsoleted version of
++ aufs_type.h in /usr/include/linux or something. When you build aufs
++ utilities, make sure that your compiler refers the correct aufs header
++ file which is built by "make headers_install."
++- if you want to use readdir(3) in userspace or pathconf(3) wrapper,
++ then run "make install_ulib" too. And refer to the aufs manual in
++ detail.
++
++There several other patches in aufs4-standalone.git. They are all
++optional. When you meet some problems, they will help you.
++- aufs4-loopback.patch
++ Supports a nested loopback mount in a branch-fs. This patch is
++ unnecessary until aufs produces a message like "you may want to try
++ another patch for loopback file".
++- vfs-ino.patch
++ Modifies a system global kernel internal function get_next_ino() in
++ order to stop assigning 0 for an inode-number. Not directly related to
++ aufs, but recommended generally.
++- tmpfs-idr.patch
++ Keeps the tmpfs inode number as the lowest value. Effective to reduce
++ the size of aufs XINO files for tmpfs branch. Also it prevents the
++ duplication of inode number, which is important for backup tools and
++ other utilities. When you find aufs XINO files for tmpfs branch
++ growing too much, try this patch.
++- lockdep-debug.patch
++ Because aufs is not only an ordinary filesystem (callee of VFS), but
++ also a caller of VFS functions for branch filesystems, subclassing of
++ the internal locks for LOCKDEP is necessary. LOCKDEP is a debugging
++ feature of linux kernel. If you enable CONFIG_LOCKDEP, then you will
++ need to apply this debug patch to expand several constant values.
++ If don't know what LOCKDEP, then you don't have apply this patch.
++
++
++4. Usage
++----------------------------------------
++At first, make sure aufs-util are installed, and please read the aufs
++manual, aufs.5 in aufs-util.git tree.
++$ man -l aufs.5
++
++And then,
++$ mkdir /tmp/rw /tmp/aufs
++# mount -t aufs -o br=/tmp/rw:${HOME} none /tmp/aufs
++
++Here is another example. The result is equivalent.
++# mount -t aufs -o br=/tmp/rw=rw:${HOME}=ro none /tmp/aufs
++ Or
++# mount -t aufs -o br:/tmp/rw none /tmp/aufs
++# mount -o remount,append:${HOME} /tmp/aufs
++
++Then, you can see whole tree of your home dir through /tmp/aufs. If
++you modify a file under /tmp/aufs, the one on your home directory is
++not affected, instead the same named file will be newly created under
++/tmp/rw. And all of your modification to a file will be applied to
++the one under /tmp/rw. This is called the file based Copy on Write
++(COW) method.
++Aufs mount options are described in aufs.5.
++If you run chroot or something and make your aufs as a root directory,
++then you need to customize the shutdown script. See the aufs manual in
++detail.
++
++Additionally, there are some sample usages of aufs which are a
++diskless system with network booting, and LiveCD over NFS.
++See sample dir in CVS tree on SourceForge.
++
++
++5. Contact
++----------------------------------------
++When you have any problems or strange behaviour in aufs, please let me
++know with:
++- /proc/mounts (instead of the output of mount(8))
++- /sys/module/aufs/*
++- /sys/fs/aufs/* (if you have them)
++- /debug/aufs/* (if you have them)
++- linux kernel version
++ if your kernel is not plain, for example modified by distributor,
++ the url where i can download its source is necessary too.
++- aufs version which was printed at loading the module or booting the
++ system, instead of the date you downloaded.
++- configuration (define/undefine CONFIG_AUFS_xxx)
++- kernel configuration or /proc/config.gz (if you have it)
++- behaviour which you think to be incorrect
++- actual operation, reproducible one is better
++- mailto: aufs-users at lists.sourceforge.net
++
++Usually, I don't watch the Public Areas(Bugs, Support Requests, Patches,
++and Feature Requests) on SourceForge. Please join and write to
++aufs-users ML.
++
++
++6. Acknowledgements
++----------------------------------------
++Thanks to everyone who have tried and are using aufs, whoever
++have reported a bug or any feedback.
++
++Especially donators:
++Tomas Matejicek(slax.org) made a donation (much more than once).
++ Since Apr 2010, Tomas M (the author of Slax and Linux Live
++ scripts) is making "doubling" donations.
++ Unfortunately I cannot list all of the donators, but I really
++ appreciate.
++ It ends Aug 2010, but the ordinary donation URL is still available.
++ <http://sourceforge.net/donate/index.php?group_id=167503>
++Dai Itasaka made a donation (2007/8).
++Chuck Smith made a donation (2008/4, 10 and 12).
++Henk Schoneveld made a donation (2008/9).
++Chih-Wei Huang, ASUS, CTC donated Eee PC 4G (2008/10).
++Francois Dupoux made a donation (2008/11).
++Bruno Cesar Ribas and Luis Carlos Erpen de Bona, C3SL serves public
++ aufs2 GIT tree (2009/2).
++William Grant made a donation (2009/3).
++Patrick Lane made a donation (2009/4).
++The Mail Archive (mail-archive.com) made donations (2009/5).
++Nippy Networks (Ed Wildgoose) made a donation (2009/7).
++New Dream Network, LLC (www.dreamhost.com) made a donation (2009/11).
++Pavel Pronskiy made a donation (2011/2).
++Iridium and Inmarsat satellite phone retailer (www.mailasail.com), Nippy
++ Networks (Ed Wildgoose) made a donation for hardware (2011/3).
++Max Lekomcev (DOM-TV project) made a donation (2011/7, 12, 2012/3, 6 and
++11).
++Sam Liddicott made a donation (2011/9).
++Era Scarecrow made a donation (2013/4).
++Bor Ratajc made a donation (2013/4).
++Alessandro Gorreta made a donation (2013/4).
++POIRETTE Marc made a donation (2013/4).
++Alessandro Gorreta made a donation (2013/4).
++lauri kasvandik made a donation (2013/5).
++"pemasu from Finland" made a donation (2013/7).
++The Parted Magic Project made a donation (2013/9 and 11).
++Pavel Barta made a donation (2013/10).
++Nikolay Pertsev made a donation (2014/5).
++James B made a donation (2014/7 and 2015/7).
++Stefano Di Biase made a donation (2014/8).
++Daniel Epellei made a donation (2015/1).
++OmegaPhil made a donation (2016/1).
++Tomasz Szewczyk made a donation (2016/4).
++James Burry made a donation (2016/12).
++
++Thank you very much.
++Donations are always, including future donations, very important and
++helpful for me to keep on developing aufs.
++
++
++7.
++----------------------------------------
++If you are an experienced user, no explanation is needed. Aufs is
++just a linux filesystem.
++
++
++Enjoy!
++
++# Local variables: ;
++# mode: text;
++# End: ;
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/01intro.txt linux-4.14/Documentation/filesystems/aufs/design/01intro.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/01intro.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/01intro.txt 2021-02-24 21:42:43.437781422 +0100
+@@ -0,0 +1,171 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+Introduction
+----------------------------------------
+
-+aufs [ei ju: ef es] | [a u f s]
++aufs [ei ju: ef es] | /ey-yoo-ef-es/ | [a u f s]
+1. abbrev. for "advanced multi-layered unification filesystem".
+2. abbrev. for "another unionfs".
+3. abbrev. for "auf das" in German which means "on the" in English.
+ Ex. "Butter aufs Brot"(G) means "butter onto bread"(E).
+ But "Filesystem aufs Filesystem" is hard to understand.
++4. abbrev. for "African Urban Fashion Show".
+
+AUFS is a filesystem with features:
+- multi layered stackable unification filesystem, the member directory
+Some people may think it is better to pass such work to user space
+helper, instead of doing in kernel space. Actually I am still thinking
+about it. But currently I have implemented it in kernel space.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/02struct.txt linux/Documentation/filesystems/aufs/design/02struct.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/02struct.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/02struct.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/02struct.txt linux-4.14/Documentation/filesystems/aufs/design/02struct.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/02struct.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/02struct.txt 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,258 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+- etc.
+
+For this purpose, use "aumvdown" command in aufs-util.git.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/03atomic_open.txt linux/Documentation/filesystems/aufs/design/03atomic_open.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/03atomic_open.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/03atomic_open.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/03atomic_open.txt linux-4.14/Documentation/filesystems/aufs/design/03atomic_open.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/03atomic_open.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/03atomic_open.txt 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,85 @@
+
-+# Copyright (C) 2015-2016 Junjiro R. Okajima
++# Copyright (C) 2015-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ ->atomic_open() are lost. in the ordinary case, the checks are
+ done by VFS:do_last(), lookup_open() and atomic_open(). some can
+ be implemented in aufs, but not all I am afraid.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/03lookup.txt linux/Documentation/filesystems/aufs/design/03lookup.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/03lookup.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/03lookup.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/03lookup.txt linux-4.14/Documentation/filesystems/aufs/design/03lookup.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/03lookup.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/03lookup.txt 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,113 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ test, and skip the revalidation in step 4. It is useful and improves
+ aufs performance when system surely hide the aufs branches from user,
+ by over-mounting something (or another method).
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/04branch.txt linux/Documentation/filesystems/aufs/design/04branch.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/04branch.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/04branch.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/04branch.txt linux-4.14/Documentation/filesystems/aufs/design/04branch.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/04branch.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/04branch.txt 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,74 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ - a file on the branch is mmap-ed.
+ - a regular file on the branch is opened for write and there is no
+ same named entry on the upper branch.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/05wbr_policy.txt linux/Documentation/filesystems/aufs/design/05wbr_policy.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/05wbr_policy.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/05wbr_policy.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/05wbr_policy.txt linux-4.14/Documentation/filesystems/aufs/design/05wbr_policy.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/05wbr_policy.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/05wbr_policy.txt 2021-02-24 21:42:43.437781422 +0100
@@ -0,0 +1,64 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ where the source and the target exists and selects the higher
+ one. If the selected branch is readonly, then aufs follows the
+ copyup policy.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/06fhsm.txt linux/Documentation/filesystems/aufs/design/06fhsm.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/06fhsm.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/06fhsm.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/06dirren.dot linux-4.14/Documentation/filesystems/aufs/design/06dirren.dot
+--- linux-4.14/Documentation/filesystems/aufs/design/06dirren.dot 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/06dirren.dot 2021-02-24 21:42:43.437781422 +0100
+@@ -0,0 +1,31 @@
++
++// to view this graph, run dot(1) command in GRAPHVIZ.
++
++digraph G {
++node [shape=box];
++whinfo [label="detailed info file\n(lower_brid_root-hinum, h_inum, namelen, old name)"];
++
++node [shape=oval];
++
++aufs_rename -> whinfo [label="store/remove"];
++
++node [shape=oval];
++inode_list [label="h_inum list in branch\ncache"];
++
++node [shape=box];
++whinode [label="h_inum list file"];
++
++node [shape=oval];
++brmgmt [label="br_add/del/mod/umount"];
++
++brmgmt -> inode_list [label="create/remove"];
++brmgmt -> whinode [label="load/store"];
++
++inode_list -> whinode [style=dashed,dir=both];
++
++aufs_rename -> inode_list [label="add/del"];
++
++aufs_lookup -> inode_list [label="search"];
++
++aufs_lookup -> whinfo [label="load/remove"];
++}
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/06dirren.txt linux-4.14/Documentation/filesystems/aufs/design/06dirren.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/06dirren.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/06dirren.txt 2021-02-24 21:42:43.437781422 +0100
+@@ -0,0 +1,102 @@
++
++# Copyright (C) 2017 Junjiro R. Okajima
++#
++# This program is free software; you can redistribute it and/or modify
++# it under the terms of the GNU General Public License as published by
++# the Free Software Foundation; either version 2 of the License, or
++# (at your option) any later version.
++#
++# This program is distributed in the hope that it will be useful,
++# but WITHOUT ANY WARRANTY; without even the implied warranty of
++# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++# GNU General Public License for more details.
++#
++# You should have received a copy of the GNU General Public License
++# along with this program. If not, see <http://www.gnu.org/licenses/>.
++
++Special handling for renaming a directory (DIRREN)
++----------------------------------------------------------------------
++First, let's assume we have a simple usecase.
++
++- /u = /rw + /ro
++- /rw/dirA exists
++- /ro/dirA and /ro/dirA/file exist too
++- there is no dirB on both branches
++- a user issues rename("dirA", "dirB")
++
++Now, what should aufs behave against this rename(2)?
++There are a few possible cases.
++
++A. returns EROFS.
++ since dirA exists on a readonly branch which cannot be renamed.
++B. returns EXDEV.
++ it is possible to copy-up dirA (only the dir itself), but the child
++ entries ("file" in this case) should not be. it must be a bad
++ approach to copy-up recursively.
++C. returns a success.
++ even the branch /ro is readonly, aufs tries renaming it. Obviously it
++ is a violation of aufs' policy.
++D. construct an extra information which indicates that /ro/dirA should
++ be handled as the name of dirB.
++ overlayfs has a similar feature called REDIRECT.
++
++Until now, aufs implements the case B only which returns EXDEV, and
++expects the userspace application behaves like mv(1) which tries
++issueing rename(2) recursively.
++
++A new aufs feature called DIRREN is introduced which implements the case
++D. There are several "extra information" added.
++
++1. detailed info per renamed directory
++ path: /rw/dirB/$AUFS_WH_DR_INFO_PFX.<lower branch-id>
++2. the inode-number list of directories on a branch
++ path: /rw/dirB/$AUFS_WH_DR_BRHINO
++
++The filename of "detailed info per directory" represents the lower
++branch, and its format is
++- a type of the branch id
++ one of these.
++ + uuid (not implemented yet)
++ + fsid
++ + dev
++- the inode-number of the branch root dir
++
++And it contains these info in a single regular file.
++- magic number
++- branch's inode-number of the logically renamed dir
++- the name of the before-renamed dir
++
++The "detailed info per directory" file is created in aufs rename(2), and
++loaded in any lookup.
++The info is considered in lookup for the matching case only. Here
++"matching" means that the root of branch (in the info filename) is same
++to the current looking-up branch. After looking-up the before-renamed
++name, the inode-number is compared. And the matched dentry is used.
++
++The "inode-number list of directories" is a regular file which contains
++simply the inode-numbers on the branch. The file is created or updated
++in removing the branch, and loaded in adding the branch. Its lifetime is
++equal to the branch.
++The list is refered in lookup, and when the current target inode is
++found in the list, the aufs tries loading the "detailed info per
++directory" and get the changed and valid name of the dir.
++
++Theoretically these "extra informaiton" may be able to be put into XATTR
++in the dir inode. But aufs doesn't choose this way because
++1. XATTR may not be supported by the branch (or its configuration)
++2. XATTR may have its size limit.
++3. XATTR may be less easy to convert than a regular file, when the
++ format of the info is changed in the future.
++At the same time, I agree that the regular file approach is much slower
++than XATTR approach. So, in the future, aufs may take the XATTR or other
++better approach.
++
++This DIRREN feature is enabled by aufs configuration, and is activated
++by a new mount option.
++
++For the more complicated case, there is a work with UDBA option, which
++is to dected the direct access to the branches (by-passing aufs) and to
++maintain the cashes in aufs. Since a single cached aufs dentry may
++contains two names, before- and after-rename, the name comparision in
++UDBA handler may not work correctly. In this case, the behaviour will be
++equivalen to udba=reval case.
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/06fhsm.txt linux-4.14/Documentation/filesystems/aufs/design/06fhsm.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/06fhsm.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/06fhsm.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,120 @@
+
-+# Copyright (C) 2011-2016 Junjiro R. Okajima
++# Copyright (C) 2011-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+
+And of course, in every step, an error may happen. So the operation
+should restore the original file state after an error happens.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/06mmap.txt linux/Documentation/filesystems/aufs/design/06mmap.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/06mmap.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/06mmap.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/06mmap.txt linux-4.14/Documentation/filesystems/aufs/design/06mmap.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/06mmap.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/06mmap.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,72 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ equivalent to vm_prfile described above.
+
+I have to give up this "looks-smater" approach.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/06xattr.txt linux/Documentation/filesystems/aufs/design/06xattr.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/06xattr.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/06xattr.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/06xattr.txt linux-4.14/Documentation/filesystems/aufs/design/06xattr.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/06xattr.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/06xattr.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,96 @@
+
-+# Copyright (C) 2014-2016 Junjiro R. Okajima
++# Copyright (C) 2014-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+Some contradiction may happen I am afraid.
+Do we need another attribute to stop copying XATTR? I am unsure. For
+now, aufs implements the branch attributes to ignore the error.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/07export.txt linux/Documentation/filesystems/aufs/design/07export.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/07export.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/07export.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/07export.txt linux-4.14/Documentation/filesystems/aufs/design/07export.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/07export.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/07export.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,58 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+ convert it into ESTALE for NFSD.
+- readdir(): call lockdep_on/off() because filldir in NFSD calls
+ lookup_one_len(), vfs_getattr(), encode_fh() and others.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/08shwh.txt linux/Documentation/filesystems/aufs/design/08shwh.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/08shwh.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/08shwh.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/08shwh.txt linux-4.14/Documentation/filesystems/aufs/design/08shwh.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/08shwh.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/08shwh.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,52 @@
+
-+# Copyright (C) 2005-2016 Junjiro R. Okajima
++# Copyright (C) 2005-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+
+This new squashfs archive can be stored on the boot device and the
+initramfs will use it to replace the old one at the next boot.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/10dynop.txt linux/Documentation/filesystems/aufs/design/10dynop.txt
---- /usr/share/empty/Documentation/filesystems/aufs/design/10dynop.txt 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/design/10dynop.txt 2016-07-25 19:05:34.807826401 +0200
+diff -urNp -x '*.orig' linux-4.14/Documentation/filesystems/aufs/design/10dynop.txt linux-4.14/Documentation/filesystems/aufs/design/10dynop.txt
+--- linux-4.14/Documentation/filesystems/aufs/design/10dynop.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/Documentation/filesystems/aufs/design/10dynop.txt 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,47 @@
+
-+# Copyright (C) 2010-2016 Junjiro R. Okajima
++# Copyright (C) 2010-2017 Junjiro R. Okajima
+#
+# This program is free software; you can redistribute it and/or modify
+# it under the terms of the GNU General Public License as published by
+XIP (DAX) mainly.
+Currently this approach is applied to address_space_operations for
+regular files only.
-diff -urN /usr/share/empty/Documentation/filesystems/aufs/README linux/Documentation/filesystems/aufs/README
---- /usr/share/empty/Documentation/filesystems/aufs/README 1970-01-01 01:00:00.000000000 +0100
-+++ linux/Documentation/filesystems/aufs/README 2016-07-25 19:05:34.807826401 +0200
-@@ -0,0 +1,392 @@
-+
-+Aufs4 -- advanced multi layered unification filesystem version 4.x
-+http://aufs.sf.net
-+Junjiro R. Okajima
-+
-+
-+0. Introduction
-+----------------------------------------
-+In the early days, aufs was entirely re-designed and re-implemented
-+Unionfs Version 1.x series. Adding many original ideas, approaches,
-+improvements and implementations, it becomes totally different from
-+Unionfs while keeping the basic features.
-+Recently, Unionfs Version 2.x series begin taking some of the same
-+approaches to aufs1's.
-+Unionfs is being developed by Professor Erez Zadok at Stony Brook
-+University and his team.
-+
-+Aufs4 supports linux-4.0 and later, and for linux-3.x series try aufs3.
-+If you want older kernel version support, try aufs2-2.6.git or
-+aufs2-standalone.git repository, aufs1 from CVS on SourceForge.
-+
-+Note: it becomes clear that "Aufs was rejected. Let's give it up."
-+ According to Christoph Hellwig, linux rejects all union-type
-+ filesystems but UnionMount.
-+<http://marc.info/?l=linux-kernel&m=123938533724484&w=2>
-+
-+PS. Al Viro seems have a plan to merge aufs as well as overlayfs and
-+ UnionMount, and he pointed out an issue around a directory mutex
-+ lock and aufs addressed it. But it is still unsure whether aufs will
-+ be merged (or any other union solution).
-+<http://marc.info/?l=linux-kernel&m=136312705029295&w=1>
-+
-+
-+1. Features
-+----------------------------------------
-+- unite several directories into a single virtual filesystem. The member
-+ directory is called as a branch.
-+- you can specify the permission flags to the branch, which are 'readonly',
-+ 'readwrite' and 'whiteout-able.'
-+- by upper writable branch, internal copyup and whiteout, files/dirs on
-+ readonly branch are modifiable logically.
-+- dynamic branch manipulation, add, del.
-+- etc...
-+
-+Also there are many enhancements in aufs, such as:
-+- test only the highest one for the directory permission (dirperm1)
-+- copyup on open (coo=)
-+- 'move' policy for copy-up between two writable branches, after
-+ checking free space.
-+- xattr, acl
-+- readdir(3) in userspace.
-+- keep inode number by external inode number table
-+- keep the timestamps of file/dir in internal copyup operation
-+- seekable directory, supporting NFS readdir.
-+- whiteout is hardlinked in order to reduce the consumption of inodes
-+ on branch
-+- do not copyup, nor create a whiteout when it is unnecessary
-+- revert a single systemcall when an error occurs in aufs
-+- remount interface instead of ioctl
-+- maintain /etc/mtab by an external command, /sbin/mount.aufs.
-+- loopback mounted filesystem as a branch
-+- kernel thread for removing the dir who has a plenty of whiteouts
-+- support copyup sparse file (a file which has a 'hole' in it)
-+- default permission flags for branches
-+- selectable permission flags for ro branch, whether whiteout can
-+ exist or not
-+- export via NFS.
-+- support <sysfs>/fs/aufs and <debugfs>/aufs.
-+- support multiple writable branches, some policies to select one
-+ among multiple writable branches.
-+- a new semantics for link(2) and rename(2) to support multiple
-+ writable branches.
-+- no glibc changes are required.
-+- pseudo hardlink (hardlink over branches)
-+- allow a direct access manually to a file on branch, e.g. bypassing aufs.
-+ including NFS or remote filesystem branch.
-+- userspace wrapper for pathconf(3)/fpathconf(3) with _PC_LINK_MAX.
-+- and more...
-+
-+Currently these features are dropped temporary from aufs4.
-+See design/08plan.txt in detail.
-+- nested mount, i.e. aufs as readonly no-whiteout branch of another aufs
-+ (robr)
-+- statistics of aufs thread (/sys/fs/aufs/stat)
-+
-+Features or just an idea in the future (see also design/*.txt),
-+- reorder the branch index without del/re-add.
-+- permanent xino files for NFSD
-+- an option for refreshing the opened files after add/del branches
-+- light version, without branch manipulation. (unnecessary?)
-+- copyup in userspace
-+- inotify in userspace
-+- readv/writev
-+
-+
-+2. Download
-+----------------------------------------
-+There are three GIT trees for aufs4, aufs4-linux.git,
-+aufs4-standalone.git, and aufs-util.git. Note that there is no "4" in
-+"aufs-util.git."
-+While the aufs-util is always necessary, you need either of aufs4-linux
-+or aufs4-standalone.
-+
-+The aufs4-linux tree includes the whole linux mainline GIT tree,
-+git://git.kernel.org/.../torvalds/linux.git.
-+And you cannot select CONFIG_AUFS_FS=m for this version, eg. you cannot
-+build aufs4 as an external kernel module.
-+Several extra patches are not included in this tree. Only
-+aufs4-standalone tree contains them. They are described in the later
-+section "Configuration and Compilation."
-+
-+On the other hand, the aufs4-standalone tree has only aufs source files
-+and necessary patches, and you can select CONFIG_AUFS_FS=m.
-+But you need to apply all aufs patches manually.
-+
-+You will find GIT branches whose name is in form of "aufs4.x" where "x"
-+represents the linux kernel version, "linux-4.x". For instance,
-+"aufs4.0" is for linux-4.0. For latest "linux-4.x-rcN", use
-+"aufs4.x-rcN" branch.
-+
-+o aufs4-linux tree
-+$ git clone --reference /your/linux/git/tree \
-+ git://github.com/sfjro/aufs4-linux.git aufs4-linux.git
-+- if you don't have linux GIT tree, then remove "--reference ..."
-+$ cd aufs4-linux.git
-+$ git checkout origin/aufs4.0
-+
-+Or You may want to directly git-pull aufs into your linux GIT tree, and
-+leave the patch-work to GIT.
-+$ cd /your/linux/git/tree
-+$ git remote add aufs4 git://github.com/sfjro/aufs4-linux.git
-+$ git fetch aufs4
-+$ git checkout -b my4.0 v4.0
-+$ (add your local change...)
-+$ git pull aufs4 aufs4.0
-+- now you have v4.0 + your_changes + aufs4.0 in you my4.0 branch.
-+- you may need to solve some conflicts between your_changes and
-+ aufs4.0. in this case, git-rerere is recommended so that you can
-+ solve the similar conflicts automatically when you upgrade to 4.1 or
-+ later in the future.
-+
-+o aufs4-standalone tree
-+$ git clone git://github.com/sfjro/aufs4-standalone.git aufs4-standalone.git
-+$ cd aufs4-standalone.git
-+$ git checkout origin/aufs4.0
-+
-+o aufs-util tree
-+$ git clone git://git.code.sf.net/p/aufs/aufs-util aufs-util.git
-+- note that the public aufs-util.git is on SourceForge instead of
-+ GitHUB.
-+$ cd aufs-util.git
-+$ git checkout origin/aufs4.0
-+
-+Note: The 4.x-rcN branch is to be used with `rc' kernel versions ONLY.
-+The minor version number, 'x' in '4.x', of aufs may not always
-+follow the minor version number of the kernel.
-+Because changes in the kernel that cause the use of a new
-+minor version number do not always require changes to aufs-util.
-+
-+Since aufs-util has its own minor version number, you may not be
-+able to find a GIT branch in aufs-util for your kernel's
-+exact minor version number.
-+In this case, you should git-checkout the branch for the
-+nearest lower number.
-+
-+For (an unreleased) example:
-+If you are using "linux-4.10" and the "aufs4.10" branch
-+does not exist in aufs-util repository, then "aufs4.9", "aufs4.8"
-+or something numerically smaller is the branch for your kernel.
-+
-+Also you can view all branches by
-+ $ git branch -a
+diff -urNp -x '*.orig' linux-4.14/MAINTAINERS linux-4.14/MAINTAINERS
+--- linux-4.14/MAINTAINERS 2021-02-24 21:42:34.681134057 +0100
++++ linux-4.14/MAINTAINERS 2021-02-24 21:42:43.431114768 +0100
+@@ -2465,6 +2465,19 @@ F: include/linux/audit.h
+ F: include/uapi/linux/audit.h
+ F: kernel/audit*
+
++AUFS (advanced multi layered unification filesystem) FILESYSTEM
++M: "J. R. Okajima" <hooanon05g@gmail.com>
++L: linux-unionfs@vger.kernel.org
++L: aufs-users@lists.sourceforge.net (members only)
++W: http://aufs.sourceforge.net
++T: git://github.com/sfjro/aufs4-linux.git
++S: Supported
++F: Documentation/filesystems/aufs/
++F: Documentation/ABI/testing/debugfs-aufs
++F: Documentation/ABI/testing/sysfs-aufs
++F: fs/aufs/
++F: include/uapi/linux/aufs_type.h
+
+ AUXILIARY DISPLAY DRIVERS
+ M: Miguel Ojeda Sandonis <miguel.ojeda.sandonis@gmail.com>
+ W: http://miguelojeda.es/auxdisplay.htm
+diff -urNp -x '*.orig' linux-4.14/drivers/block/loop.c linux-4.14/drivers/block/loop.c
+--- linux-4.14/drivers/block/loop.c 2021-02-24 21:42:34.864466934 +0100
++++ linux-4.14/drivers/block/loop.c 2021-02-24 21:42:43.447781402 +0100
+@@ -605,6 +605,15 @@ static inline void loop_update_dio(struc
+ lo->use_dio);
+ }
+
++static struct file *loop_real_file(struct file *file)
++{
++ struct file *f = NULL;
+
-+3. Configuration and Compilation
-+----------------------------------------
-+Make sure you have git-checkout'ed the correct branch.
++ if (file->f_path.dentry->d_sb->s_op->real_loop)
++ f = file->f_path.dentry->d_sb->s_op->real_loop(file);
++ return f;
++}
+
-+For aufs4-linux tree,
-+- enable CONFIG_AUFS_FS.
-+- set other aufs configurations if necessary.
+ static void loop_reread_partitions(struct loop_device *lo,
+ struct block_device *bdev)
+ {
+@@ -669,6 +678,7 @@ static int loop_change_fd(struct loop_de
+ unsigned int arg)
+ {
+ struct file *file, *old_file;
++ struct file *f, *virt_file = NULL, *old_virt_file;
+ struct inode *inode;
+ int error;
+
+@@ -685,6 +695,12 @@ static int loop_change_fd(struct loop_de
+ file = fget(arg);
+ if (!file)
+ goto out;
++ f = loop_real_file(file);
++ if (f) {
++ virt_file = file;
++ file = f;
++ get_file(file);
++ }
+
+ error = loop_validate_file(file, bdev);
+ if (error)
+@@ -692,6 +708,7 @@ static int loop_change_fd(struct loop_de
+
+ inode = file->f_mapping->host;
+ old_file = lo->lo_backing_file;
++ old_virt_file = lo->lo_backing_virt_file;
+
+ error = -EINVAL;
+
+@@ -703,6 +720,7 @@ static int loop_change_fd(struct loop_de
+ blk_mq_freeze_queue(lo->lo_queue);
+ mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask);
+ lo->lo_backing_file = file;
++ lo->lo_backing_virt_file = virt_file;
+ lo->old_gfp_mask = mapping_gfp_mask(file->f_mapping);
+ mapping_set_gfp_mask(file->f_mapping,
+ lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
+@@ -710,16 +728,38 @@ static int loop_change_fd(struct loop_de
+ blk_mq_unfreeze_queue(lo->lo_queue);
+
+ fput(old_file);
++ if (old_virt_file)
++ fput(old_virt_file);
+ if (lo->lo_flags & LO_FLAGS_PARTSCAN)
+ loop_reread_partitions(lo, bdev);
+ return 0;
+
+ out_putf:
+ fput(file);
++ if (virt_file)
++ fput(virt_file);
+ out:
+ return error;
+ }
+
++/*
++ * for AUFS
++ * no get/put for file.
++ */
++struct file *loop_backing_file(struct super_block *sb)
++{
++ struct file *ret;
++ struct loop_device *l;
+
-+For aufs4-standalone tree,
-+There are several ways to build.
++ ret = NULL;
++ if (MAJOR(sb->s_dev) == LOOP_MAJOR) {
++ l = sb->s_bdev->bd_disk->private_data;
++ ret = l->lo_backing_file;
++ }
++ return ret;
++}
++EXPORT_SYMBOL_GPL(loop_backing_file);
+
-+1.
-+- apply ./aufs4-kbuild.patch to your kernel source files.
-+- apply ./aufs4-base.patch too.
-+- apply ./aufs4-mmap.patch too.
-+- apply ./aufs4-standalone.patch too, if you have a plan to set
-+ CONFIG_AUFS_FS=m. otherwise you don't need ./aufs4-standalone.patch.
-+- copy ./{Documentation,fs,include/uapi/linux/aufs_type.h} files to your
-+ kernel source tree. Never copy $PWD/include/uapi/linux/Kbuild.
-+- enable CONFIG_AUFS_FS, you can select either
-+ =m or =y.
-+- and build your kernel as usual.
-+- install the built kernel.
-+ Note: Since linux-3.9, every filesystem module requires an alias
-+ "fs-<fsname>". You should make sure that "fs-aufs" is listed in your
-+ modules.aliases file if you set CONFIG_AUFS_FS=m.
-+- install the header files too by "make headers_install" to the
-+ directory where you specify. By default, it is $PWD/usr.
-+ "make help" shows a brief note for headers_install.
-+- and reboot your system.
+ /* loop sysfs attributes */
+
+ static ssize_t loop_attr_show(struct device *dev, char *page,
+@@ -885,7 +925,7 @@ static int loop_prepare_queue(struct loo
+ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
+ struct block_device *bdev, unsigned int arg)
+ {
+- struct file *file;
++ struct file *file, *f, *virt_file = NULL;
+ struct inode *inode;
+ struct address_space *mapping;
+ int lo_flags = 0;
+@@ -899,6 +939,12 @@ static int loop_set_fd(struct loop_devic
+ file = fget(arg);
+ if (!file)
+ goto out;
++ f = loop_real_file(file);
++ if (f) {
++ virt_file = file;
++ file = f;
++ get_file(file);
++ }
+
+ error = -EBUSY;
+ if (lo->lo_state != Lo_unbound)
+@@ -931,6 +977,7 @@ static int loop_set_fd(struct loop_devic
+ lo->lo_device = bdev;
+ lo->lo_flags = lo_flags;
+ lo->lo_backing_file = file;
++ lo->lo_backing_virt_file = virt_file;
+ lo->transfer = NULL;
+ lo->ioctl = NULL;
+ lo->lo_sizelimit = 0;
+@@ -964,6 +1011,8 @@ static int loop_set_fd(struct loop_devic
+
+ out_putf:
+ fput(file);
++ if (virt_file)
++ fput(virt_file);
+ out:
+ /* This is safe: open() is still holding a reference. */
+ module_put(THIS_MODULE);
+@@ -1010,6 +1059,7 @@ loop_init_xfer(struct loop_device *lo, s
+ static int loop_clr_fd(struct loop_device *lo)
+ {
+ struct file *filp = lo->lo_backing_file;
++ struct file *virt_filp = lo->lo_backing_virt_file;
+ gfp_t gfp = lo->old_gfp_mask;
+ struct block_device *bdev = lo->lo_device;
+
+@@ -1041,6 +1091,7 @@ static int loop_clr_fd(struct loop_devic
+ spin_lock_irq(&lo->lo_lock);
+ lo->lo_state = Lo_rundown;
+ lo->lo_backing_file = NULL;
++ lo->lo_backing_virt_file = NULL;
+ spin_unlock_irq(&lo->lo_lock);
+
+ loop_release_xfer(lo);
+@@ -1088,6 +1139,8 @@ static int loop_clr_fd(struct loop_devic
+ * bd_mutex which is usually taken before lo_ctl_mutex.
+ */
+ fput(filp);
++ if (virt_filp)
++ fput(virt_filp);
+ return 0;
+ }
+
+diff -urNp -x '*.orig' linux-4.14/drivers/block/loop.h linux-4.14/drivers/block/loop.h
+--- linux-4.14/drivers/block/loop.h 2021-02-24 21:42:34.864466934 +0100
++++ linux-4.14/drivers/block/loop.h 2021-02-24 21:42:43.447781402 +0100
+@@ -46,7 +46,7 @@ struct loop_device {
+ int (*ioctl)(struct loop_device *, int cmd,
+ unsigned long arg);
+
+- struct file * lo_backing_file;
++ struct file * lo_backing_file, *lo_backing_virt_file;
+ struct block_device *lo_device;
+ void *key_data;
+
+diff -urNp -x '*.orig' linux-4.14/fs/Kconfig linux-4.14/fs/Kconfig
+--- linux-4.14/fs/Kconfig 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/Kconfig 2021-02-24 21:42:43.431114768 +0100
+@@ -248,6 +248,7 @@ source "fs/pstore/Kconfig"
+ source "fs/sysv/Kconfig"
+ source "fs/ufs/Kconfig"
+ source "fs/exofs/Kconfig"
++source "fs/aufs/Kconfig"
+
+ endif # MISC_FILESYSTEMS
+
+diff -urNp -x '*.orig' linux-4.14/fs/Makefile linux-4.14/fs/Makefile
+--- linux-4.14/fs/Makefile 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/Makefile 2021-02-24 21:42:43.431114768 +0100
+@@ -129,3 +129,4 @@ obj-y += exofs/ # Multiple modules
+ obj-$(CONFIG_CEPH_FS) += ceph/
+ obj-$(CONFIG_PSTORE) += pstore/
+ obj-$(CONFIG_EFIVAR_FS) += efivarfs/
++obj-$(CONFIG_AUFS_FS) += aufs/
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/Kconfig linux-4.14/fs/aufs/Kconfig
+--- linux-4.14/fs/aufs/Kconfig 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/Kconfig 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,198 @@
++config AUFS_FS
++ tristate "Aufs (Advanced multi layered unification filesystem) support"
++ help
++ Aufs is a stackable unification filesystem such as Unionfs,
++ which unifies several directories and provides a merged single
++ directory.
++ In the early days, aufs was entirely re-designed and
++ re-implemented Unionfs Version 1.x series. Introducing many
++ original ideas, approaches and improvements, it becomes totally
++ different from Unionfs while keeping the basic features.
+
-+2.
-+- module only (CONFIG_AUFS_FS=m).
-+- apply ./aufs4-base.patch to your kernel source files.
-+- apply ./aufs4-mmap.patch too.
-+- apply ./aufs4-standalone.patch too.
-+- build your kernel, don't forget "make headers_install", and reboot.
-+- edit ./config.mk and set other aufs configurations if necessary.
-+ Note: You should read $PWD/fs/aufs/Kconfig carefully which describes
-+ every aufs configurations.
-+- build the module by simple "make".
-+ Note: Since linux-3.9, every filesystem module requires an alias
-+ "fs-<fsname>". You should make sure that "fs-aufs" is listed in your
-+ modules.aliases file.
-+- you can specify ${KDIR} make variable which points to your kernel
-+ source tree.
-+- install the files
-+ + run "make install" to install the aufs module, or copy the built
-+ $PWD/aufs.ko to /lib/modules/... and run depmod -a (or reboot simply).
-+ + run "make install_headers" (instead of headers_install) to install
-+ the modified aufs header file (you can specify DESTDIR which is
-+ available in aufs standalone version's Makefile only), or copy
-+ $PWD/usr/include/linux/aufs_type.h to /usr/include/linux or wherever
-+ you like manually. By default, the target directory is $PWD/usr.
-+- no need to apply aufs4-kbuild.patch, nor copying source files to your
-+ kernel source tree.
++if AUFS_FS
++choice
++ prompt "Maximum number of branches"
++ default AUFS_BRANCH_MAX_127
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_127
++ bool "127"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_511
++ bool "511"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_1023
++ bool "1023"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_32767
++ bool "32767"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++endchoice
+
-+Note: The header file aufs_type.h is necessary to build aufs-util
-+ as well as "make headers_install" in the kernel source tree.
-+ headers_install is subject to be forgotten, but it is essentially
-+ necessary, not only for building aufs-util.
-+ You may not meet problems without headers_install in some older
-+ version though.
++config AUFS_SBILIST
++ bool
++ depends on AUFS_MAGIC_SYSRQ || PROC_FS
++ default y
++ help
++ Automatic configuration for internal use.
++ When aufs supports Magic SysRq or /proc, enabled automatically.
+
-+And then,
-+- read README in aufs-util, build and install it
-+- note that your distribution may contain an obsoleted version of
-+ aufs_type.h in /usr/include/linux or something. When you build aufs
-+ utilities, make sure that your compiler refers the correct aufs header
-+ file which is built by "make headers_install."
-+- if you want to use readdir(3) in userspace or pathconf(3) wrapper,
-+ then run "make install_ulib" too. And refer to the aufs manual in
-+ detail.
++config AUFS_HNOTIFY
++ bool "Detect direct branch access (bypassing aufs)"
++ help
++ If you want to modify files on branches directly, eg. bypassing aufs,
++ and want aufs to detect the changes of them fully, then enable this
++ option and use 'udba=notify' mount option.
++ Currently there is only one available configuration, "fsnotify".
++ It will have a negative impact to the performance.
++ See detail in aufs.5.
+
-+There several other patches in aufs4-standalone.git. They are all
-+optional. When you meet some problems, they will help you.
-+- aufs4-loopback.patch
-+ Supports a nested loopback mount in a branch-fs. This patch is
-+ unnecessary until aufs produces a message like "you may want to try
-+ another patch for loopback file".
-+- vfs-ino.patch
-+ Modifies a system global kernel internal function get_next_ino() in
-+ order to stop assigning 0 for an inode-number. Not directly related to
-+ aufs, but recommended generally.
-+- tmpfs-idr.patch
-+ Keeps the tmpfs inode number as the lowest value. Effective to reduce
-+ the size of aufs XINO files for tmpfs branch. Also it prevents the
-+ duplication of inode number, which is important for backup tools and
-+ other utilities. When you find aufs XINO files for tmpfs branch
-+ growing too much, try this patch.
-+- lockdep-debug.patch
-+ Because aufs is not only an ordinary filesystem (callee of VFS), but
-+ also a caller of VFS functions for branch filesystems, subclassing of
-+ the internal locks for LOCKDEP is necessary. LOCKDEP is a debugging
-+ feature of linux kernel. If you enable CONFIG_LOCKDEP, then you will
-+ need to apply this debug patch to expand several constant values.
-+ If don't know what LOCKDEP, then you don't have apply this patch.
++choice
++ prompt "method" if AUFS_HNOTIFY
++ default AUFS_HFSNOTIFY
++config AUFS_HFSNOTIFY
++ bool "fsnotify"
++ select FSNOTIFY
++endchoice
+
++config AUFS_EXPORT
++ bool "NFS-exportable aufs"
++ depends on EXPORTFS
++ help
++ If you want to export your mounted aufs via NFS, then enable this
++ option. There are several requirements for this configuration.
++ See detail in aufs.5.
+
-+4. Usage
-+----------------------------------------
-+At first, make sure aufs-util are installed, and please read the aufs
-+manual, aufs.5 in aufs-util.git tree.
-+$ man -l aufs.5
++config AUFS_INO_T_64
++ bool
++ depends on AUFS_EXPORT
++ depends on 64BIT && !(ALPHA || S390)
++ default y
++ help
++ Automatic configuration for internal use.
++ /* typedef unsigned long/int __kernel_ino_t */
++ /* alpha and s390x are int */
+
-+And then,
-+$ mkdir /tmp/rw /tmp/aufs
-+# mount -t aufs -o br=/tmp/rw:${HOME} none /tmp/aufs
++config AUFS_XATTR
++ bool "support for XATTR/EA (including Security Labels)"
++ help
++ If your branch fs supports XATTR/EA and you want to make them
++ available in aufs too, then enable this opsion and specify the
++ branch attributes for EA.
++ See detail in aufs.5.
+
-+Here is another example. The result is equivalent.
-+# mount -t aufs -o br=/tmp/rw=rw:${HOME}=ro none /tmp/aufs
-+ Or
-+# mount -t aufs -o br:/tmp/rw none /tmp/aufs
-+# mount -o remount,append:${HOME} /tmp/aufs
++config AUFS_FHSM
++ bool "File-based Hierarchical Storage Management"
++ help
++ Hierarchical Storage Management (or HSM) is a well-known feature
++ in the storage world. Aufs provides this feature as file-based.
++ with multiple branches.
++ These multiple branches are prioritized, ie. the topmost one
++ should be the fastest drive and be used heavily.
+
-+Then, you can see whole tree of your home dir through /tmp/aufs. If
-+you modify a file under /tmp/aufs, the one on your home directory is
-+not affected, instead the same named file will be newly created under
-+/tmp/rw. And all of your modification to a file will be applied to
-+the one under /tmp/rw. This is called the file based Copy on Write
-+(COW) method.
-+Aufs mount options are described in aufs.5.
-+If you run chroot or something and make your aufs as a root directory,
-+then you need to customize the shutdown script. See the aufs manual in
-+detail.
++config AUFS_RDU
++ bool "Readdir in userspace"
++ help
++ Aufs has two methods to provide a merged view for a directory,
++ by a user-space library and by kernel-space natively. The latter
++ is always enabled but sometimes large and slow.
++ If you enable this option, install the library in aufs2-util
++ package, and set some environment variables for your readdir(3),
++ then the work will be handled in user-space which generally
++ shows better performance in most cases.
++ See detail in aufs.5.
+
-+Additionally, there are some sample usages of aufs which are a
-+diskless system with network booting, and LiveCD over NFS.
-+See sample dir in CVS tree on SourceForge.
++config AUFS_DIRREN
++ bool "Workaround for rename(2)-ing a directory"
++ help
++ By default, aufs returns EXDEV error in renameing a dir who has
++ his child on the lower branch, since it is a bad idea to issue
++ rename(2) internally for every lower branch. But user may not
++ accept this behaviour. So here is a workaround to allow such
++ rename(2) and store some extra infromation on the writable
++ branch. Obviously this costs high (and I don't like it).
++ To use this feature, you need to enable this configuration AND
++ to specify the mount option `dirren.'
++ See details in aufs.5 and the design documents.
+
++config AUFS_SHWH
++ bool "Show whiteouts"
++ help
++ If you want to make the whiteouts in aufs visible, then enable
++ this option and specify 'shwh' mount option. Although it may
++ sounds like philosophy or something, but in technically it
++ simply shows the name of whiteout with keeping its behaviour.
+
-+5. Contact
-+----------------------------------------
-+When you have any problems or strange behaviour in aufs, please let me
-+know with:
-+- /proc/mounts (instead of the output of mount(8))
-+- /sys/module/aufs/*
-+- /sys/fs/aufs/* (if you have them)
-+- /debug/aufs/* (if you have them)
-+- linux kernel version
-+ if your kernel is not plain, for example modified by distributor,
-+ the url where i can download its source is necessary too.
-+- aufs version which was printed at loading the module or booting the
-+ system, instead of the date you downloaded.
-+- configuration (define/undefine CONFIG_AUFS_xxx)
-+- kernel configuration or /proc/config.gz (if you have it)
-+- behaviour which you think to be incorrect
-+- actual operation, reproducible one is better
-+- mailto: aufs-users at lists.sourceforge.net
++config AUFS_BR_RAMFS
++ bool "Ramfs (initramfs/rootfs) as an aufs branch"
++ help
++ If you want to use ramfs as an aufs branch fs, then enable this
++ option. Generally tmpfs is recommended.
++ Aufs prohibited them to be a branch fs by default, because
++ initramfs becomes unusable after switch_root or something
++ generally. If you sets initramfs as an aufs branch and boot your
++ system by switch_root, you will meet a problem easily since the
++ files in initramfs may be inaccessible.
++ Unless you are going to use ramfs as an aufs branch fs without
++ switch_root or something, leave it N.
+
-+Usually, I don't watch the Public Areas(Bugs, Support Requests, Patches,
-+and Feature Requests) on SourceForge. Please join and write to
-+aufs-users ML.
++config AUFS_BR_FUSE
++ bool "Fuse fs as an aufs branch"
++ depends on FUSE_FS
++ select AUFS_POLL
++ help
++ If you want to use fuse-based userspace filesystem as an aufs
++ branch fs, then enable this option.
++ It implements the internal poll(2) operation which is
++ implemented by fuse only (curretnly).
+
++config AUFS_POLL
++ bool
++ help
++ Automatic configuration for internal use.
+
-+6. Acknowledgements
-+----------------------------------------
-+Thanks to everyone who have tried and are using aufs, whoever
-+have reported a bug or any feedback.
++config AUFS_BR_HFSPLUS
++ bool "Hfsplus as an aufs branch"
++ depends on HFSPLUS_FS
++ default y
++ help
++ If you want to use hfsplus fs as an aufs branch fs, then enable
++ this option. This option introduces a small overhead at
++ copying-up a file on hfsplus.
+
-+Especially donators:
-+Tomas Matejicek(slax.org) made a donation (much more than once).
-+ Since Apr 2010, Tomas M (the author of Slax and Linux Live
-+ scripts) is making "doubling" donations.
-+ Unfortunately I cannot list all of the donators, but I really
-+ appreciate.
-+ It ends Aug 2010, but the ordinary donation URL is still available.
-+ <http://sourceforge.net/donate/index.php?group_id=167503>
-+Dai Itasaka made a donation (2007/8).
-+Chuck Smith made a donation (2008/4, 10 and 12).
-+Henk Schoneveld made a donation (2008/9).
-+Chih-Wei Huang, ASUS, CTC donated Eee PC 4G (2008/10).
-+Francois Dupoux made a donation (2008/11).
-+Bruno Cesar Ribas and Luis Carlos Erpen de Bona, C3SL serves public
-+ aufs2 GIT tree (2009/2).
-+William Grant made a donation (2009/3).
-+Patrick Lane made a donation (2009/4).
-+The Mail Archive (mail-archive.com) made donations (2009/5).
-+Nippy Networks (Ed Wildgoose) made a donation (2009/7).
-+New Dream Network, LLC (www.dreamhost.com) made a donation (2009/11).
-+Pavel Pronskiy made a donation (2011/2).
-+Iridium and Inmarsat satellite phone retailer (www.mailasail.com), Nippy
-+ Networks (Ed Wildgoose) made a donation for hardware (2011/3).
-+Max Lekomcev (DOM-TV project) made a donation (2011/7, 12, 2012/3, 6 and
-+11).
-+Sam Liddicott made a donation (2011/9).
-+Era Scarecrow made a donation (2013/4).
-+Bor Ratajc made a donation (2013/4).
-+Alessandro Gorreta made a donation (2013/4).
-+POIRETTE Marc made a donation (2013/4).
-+Alessandro Gorreta made a donation (2013/4).
-+lauri kasvandik made a donation (2013/5).
-+"pemasu from Finland" made a donation (2013/7).
-+The Parted Magic Project made a donation (2013/9 and 11).
-+Pavel Barta made a donation (2013/10).
-+Nikolay Pertsev made a donation (2014/5).
-+James B made a donation (2014/7 and 2015/7).
-+Stefano Di Biase made a donation (2014/8).
-+Daniel Epellei made a donation (2015/1).
-+OmegaPhil made a donation (2016/1).
-+Tomasz Szewczyk made a donation (2016/4).
++config AUFS_BDEV_LOOP
++ bool
++ depends on BLK_DEV_LOOP
++ default y
++ help
++ Automatic configuration for internal use.
++ Convert =[ym] into =y.
+
-+Thank you very much.
-+Donations are always, including future donations, very important and
-+helpful for me to keep on developing aufs.
++config AUFS_DEBUG
++ bool "Debug aufs"
++ help
++ Enable this to compile aufs internal debug code.
++ It will have a negative impact to the performance.
+
++config AUFS_MAGIC_SYSRQ
++ bool
++ depends on AUFS_DEBUG && MAGIC_SYSRQ
++ default y
++ help
++ Automatic configuration for internal use.
++ When aufs supports Magic SysRq, enabled automatically.
++endif
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/Makefile linux-4.14/fs/aufs/Makefile
+--- linux-4.14/fs/aufs/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/Makefile 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,45 @@
+
-+7.
-+----------------------------------------
-+If you are an experienced user, no explanation is needed. Aufs is
-+just a linux filesystem.
++include ${src}/magic.mk
++ifeq (${CONFIG_AUFS_FS},m)
++include ${src}/conf.mk
++endif
++-include ${src}/priv_def.mk
+
++# cf. include/linux/kernel.h
++# enable pr_debug
++ccflags-y += -DDEBUG
++# sparse requires the full pathname
++ifdef M
++ccflags-y += -include ${M}/../../include/uapi/linux/aufs_type.h
++else
++ccflags-y += -include ${srctree}/include/uapi/linux/aufs_type.h
++endif
+
-+Enjoy!
++obj-$(CONFIG_AUFS_FS) += aufs.o
++aufs-y := module.o sbinfo.o super.o branch.o xino.o sysaufs.o opts.o \
++ wkq.o vfsub.o dcsub.o \
++ cpup.o whout.o wbr_policy.o \
++ dinfo.o dentry.o \
++ dynop.o \
++ finfo.o file.o f_op.o \
++ dir.o vdir.o \
++ iinfo.o inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o \
++ mvdown.o ioctl.o
+
-+# Local variables: ;
-+# mode: text;
-+# End: ;
-diff -urN /usr/share/empty/fs/aufs/aufs.h linux/fs/aufs/aufs.h
---- /usr/share/empty/fs/aufs/aufs.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/aufs.h 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,59 @@
++# all are boolean
++aufs-$(CONFIG_PROC_FS) += procfs.o plink.o
++aufs-$(CONFIG_SYSFS) += sysfs.o
++aufs-$(CONFIG_DEBUG_FS) += dbgaufs.o
++aufs-$(CONFIG_AUFS_BDEV_LOOP) += loop.o
++aufs-$(CONFIG_AUFS_HNOTIFY) += hnotify.o
++aufs-$(CONFIG_AUFS_HFSNOTIFY) += hfsnotify.o
++aufs-$(CONFIG_AUFS_EXPORT) += export.o
++aufs-$(CONFIG_AUFS_XATTR) += xattr.o
++aufs-$(CONFIG_FS_POSIX_ACL) += posix_acl.o
++aufs-$(CONFIG_AUFS_DIRREN) += dirren.o
++aufs-$(CONFIG_AUFS_FHSM) += fhsm.o
++aufs-$(CONFIG_AUFS_POLL) += poll.o
++aufs-$(CONFIG_AUFS_RDU) += rdu.o
++aufs-$(CONFIG_AUFS_BR_HFSPLUS) += hfsplus.o
++aufs-$(CONFIG_AUFS_DEBUG) += debug.o
++aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/aufs.h linux-4.14/fs/aufs/aufs.h
+--- linux-4.14/fs/aufs/aufs.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/aufs.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,60 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#include "dbgaufs.h"
+#include "dentry.h"
+#include "dir.h"
++#include "dirren.h"
+#include "dynop.h"
+#include "file.h"
+#include "fstype.h"
++#include "hbl.h"
+#include "inode.h"
+#include "loop.h"
+#include "module.h"
+#include "opts.h"
+#include "rwsem.h"
-+#include "spl.h"
+#include "super.h"
+#include "sysaufs.h"
+#include "vfsub.h"
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_H__ */
-diff -urN /usr/share/empty/fs/aufs/branch.c linux/fs/aufs/branch.c
---- /usr/share/empty/fs/aufs/branch.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/branch.c 2016-08-17 18:01:06.095221547 +0200
-@@ -0,0 +1,1409 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/branch.c linux-4.14/fs/aufs/branch.c
+--- linux-4.14/fs/aufs/branch.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/branch.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,1432 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ struct au_dykey **key;
+
+ au_hnotify_fin_br(br);
++ /* always, regardless the mount option */
++ au_dr_hino_free(&br->br_dirren);
+
+ if (br->br_xino.xi_file)
+ fput(br->br_xino.xi_file);
-+ mutex_destroy(&br->br_xino.xi_nondir_mtx);
++ for (i = br->br_xino.xi_nondir.total - 1; i >= 0; i--)
++ AuDebugOn(br->br_xino.xi_nondir.array[i]);
++ kfree(br->br_xino.xi_nondir.array);
+
+ AuDebugOn(au_br_count(br));
+ au_br_count_fin(br);
+
+ if (br->br_fhsm) {
+ au_br_fhsm_fin(br->br_fhsm);
-+ au_delayed_kfree(br->br_fhsm);
++ kfree(br->br_fhsm);
+ }
+
+ key = br->br_dykey;
+ lockdep_off();
+ path_put(&br->br_path);
+ lockdep_on();
-+ if (wbr)
-+ au_delayed_kfree(wbr);
-+ au_delayed_kfree(br);
++ kfree(wbr);
++ kfree(br);
+}
+
+/*
+ int err;
+
+ err = -ENOMEM;
-+ root = sb->s_root;
+ add_branch = kzalloc(sizeof(*add_branch), GFP_NOFS);
+ if (unlikely(!add_branch))
+ goto out;
++ add_branch->br_xino.xi_nondir.total = 8; /* initial size */
++ add_branch->br_xino.xi_nondir.array
++ = kcalloc(add_branch->br_xino.xi_nondir.total, sizeof(ino_t),
++ GFP_NOFS);
++ if (unlikely(!add_branch->br_xino.xi_nondir.array))
++ goto out_br;
+
+ err = au_hnotify_init_br(add_branch, perm);
+ if (unlikely(err))
-+ goto out_br;
++ goto out_xinondir;
+
+ if (au_br_writable(perm)) {
+ /* may be freed separately at changing the branch permission */
+ goto out_wbr;
+ }
+
-+ err = au_sbr_realloc(au_sbi(sb), new_nbranch);
++ root = sb->s_root;
++ err = au_sbr_realloc(au_sbi(sb), new_nbranch, /*may_shrink*/0);
+ if (!err)
-+ err = au_di_realloc(au_di(root), new_nbranch);
++ err = au_di_realloc(au_di(root), new_nbranch, /*may_shrink*/0);
+ if (!err) {
+ inode = d_inode(root);
-+ err = au_hinode_realloc(au_ii(inode), new_nbranch);
++ err = au_hinode_realloc(au_ii(inode), new_nbranch,
++ /*may_shrink*/0);
+ }
+ if (!err)
+ return add_branch; /* success */
+
+out_wbr:
-+ if (add_branch->br_wbr)
-+ au_delayed_kfree(add_branch->br_wbr);
++ kfree(add_branch->br_wbr);
+out_hnotify:
+ au_hnotify_fin_br(add_branch);
++out_xinondir:
++ kfree(add_branch->br_xino.xi_nondir.array);
+out_br:
-+ au_delayed_kfree(add_branch);
++ kfree(add_branch);
+out:
+ return ERR_PTR(err);
+}
+ br->br_perm = old_perm;
+
+ if (!err && wbr && !au_br_writable(new_perm)) {
-+ au_delayed_kfree(wbr);
++ kfree(wbr);
+ br->br_wbr = NULL;
+ }
+
+ struct inode *h_inode;
+
+ err = 0;
-+ mutex_init(&br->br_xino.xi_nondir_mtx);
++ spin_lock_init(&br->br_xino.xi_nondir.spin);
++ init_waitqueue_head(&br->br_xino.xi_nondir.wqh);
+ br->br_perm = add->perm;
+ br->br_path = add->path; /* set first, path_get() later */
+ spin_lock_init(&br->br_dykey_lock);
+ br->br_id = au_new_br_id(sb);
+ AuDebugOn(br->br_id < 0);
+
++ /* always, regardless the given option */
++ err = au_dr_br_init(sb, br, &add->path);
++ if (unlikely(err))
++ goto out_err;
++
+ if (au_br_writable(add->perm)) {
+ err = au_wbr_init(br, sb, add->perm);
+ if (unlikely(err))
+{
+ unsigned long long n;
+ struct file **p, *f;
-+ struct au_sphlhead *files;
++ struct hlist_bl_head *files;
++ struct hlist_bl_node *pos;
+ struct au_finfo *finfo;
+
+ n = 0;
+ p = a;
+ files = &au_sbi(sb)->si_files;
-+ spin_lock(&files->spin);
-+ hlist_for_each_entry(finfo, &files->head, fi_hlist) {
++ hlist_bl_lock(files);
++ hlist_bl_for_each_entry(finfo, pos, files, fi_hlist) {
+ f = finfo->fi_file;
+ if (file_count(f)
+ && !special_file(file_inode(f)->i_mode)) {
+ AuDebugOn(n > max);
+ }
+ }
-+ spin_unlock(&files->spin);
++ hlist_bl_unlock(files);
+
+ return n;
+}
+ sbinfo->si_branch[0 + bbot] = NULL;
+ sbinfo->si_bbot--;
+
-+ p = krealloc(sbinfo->si_branch, sizeof(*p) * bbot, AuGFP_SBILIST);
++ p = au_krealloc(sbinfo->si_branch, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
+ if (p)
+ sbinfo->si_branch = p;
+ /* harmless error */
+ /* au_h_dentry_init(au_hdentry(dinfo, bbot); */
+ dinfo->di_bbot--;
+
-+ p = krealloc(dinfo->di_hdentry, sizeof(*p) * bbot, AuGFP_SBILIST);
++ p = au_krealloc(dinfo->di_hdentry, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
+ if (p)
+ dinfo->di_hdentry = p;
+ /* harmless error */
+ /* au_hinode_init(au_hinode(iinfo, bbot)); */
+ iinfo->ii_bbot--;
+
-+ p = krealloc(iinfo->ii_hinode, sizeof(*p) * bbot, AuGFP_SBILIST);
++ p = au_krealloc(iinfo->ii_hinode, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
+ if (p)
+ iinfo->ii_hinode = p;
+ /* harmless error */
+ au_br_do_del_hip(au_ii(inode), bindex, bbot);
+ au_sbilist_unlock();
+
++ /* ignore an error */
++ au_dr_br_fin(sb, br); /* always, regardless the mount option */
++
+ dput(h_root);
+ iput(h_inode);
+ au_br_do_free(br);
+ if (br->br_wbr) {
+ err = au_wbr_init(br, sb, mod->perm);
+ if (unlikely(err)) {
-+ au_delayed_kfree(br->br_wbr);
++ kfree(br->br_wbr);
+ br->br_wbr = NULL;
+ }
+ }
+ if (!au_br_fhsm(mod->perm)) {
+ /* fhsm --> non-fhsm */
+ au_br_fhsm_fin(br->br_fhsm);
-+ au_delayed_kfree(br->br_fhsm);
++ kfree(br->br_fhsm);
+ br->br_fhsm = NULL;
+ }
+ } else if (au_br_fhsm(mod->perm))
+ goto out; /* success */
+
+out_bf:
-+ if (bf)
-+ au_delayed_kfree(bf);
++ kfree(bf);
+out:
+ AuTraceErr(err);
+ return err;
+
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/branch.h linux/fs/aufs/branch.h
---- /usr/share/empty/fs/aufs/branch.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/branch.h 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,309 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/branch.h linux-4.14/fs/aufs/branch.h
+--- linux-4.14/fs/aufs/branch.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/branch.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,333 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#ifdef __KERNEL__
+
+#include <linux/mount.h>
++#include "dirren.h"
+#include "dynop.h"
+#include "rwsem.h"
+#include "super.h"
+/* a xino file */
+struct au_xino_file {
+ struct file *xi_file;
-+ struct mutex xi_nondir_mtx;
++ struct {
++ spinlock_t spin;
++ ino_t *array;
++ int total;
++ /* reserved for future use */
++ /* unsigned long *bitmap; */
++ wait_queue_head_t wqh;
++ } xi_nondir;
+
+ /* todo: make xino files an array to support huge inode number */
+
+ /* entries under sysfs per mount-point */
+ struct au_brsysfs br_sysfs[AuBrSysfs_Last];
+#endif
++
++ struct au_dr_br br_dirren;
+};
+
+/* ---------------------------------------------------------------------- */
+
+static inline int au_br_rdonly(struct au_branch *br)
+{
-+ return ((au_br_sb(br)->s_flags & MS_RDONLY)
++ return (sb_rdonly(au_br_sb(br))
+ || !au_br_writable(br->br_perm))
+ ? -EROFS : 0;
+}
+struct file *au_xino_def(struct super_block *sb);
+int au_xino_path(struct seq_file *seq, struct file *file);
+
++void au_xinondir_leave(struct super_block *sb, aufs_bindex_t bindex,
++ ino_t h_ino, int idx);
++int au_xinondir_enter(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ int *idx);
++
+/* ---------------------------------------------------------------------- */
+
+/* Superblock to branch */
+
+/* ---------------------------------------------------------------------- */
+
++#define wbr_wh_read_lock(wbr) au_rw_read_lock(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_write_lock(wbr) au_rw_write_lock(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_read_trylock(wbr) au_rw_read_trylock(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_write_trylock(wbr) au_rw_write_trylock(&(wbr)->wbr_wh_rwsem)
+/*
-+ * wbr_wh_read_lock, wbr_wh_write_lock
-+ * wbr_wh_read_unlock, wbr_wh_write_unlock, wbr_wh_downgrade_lock
-+ */
-+AuSimpleRwsemFuncs(wbr_wh, struct au_wbr *wbr, &wbr->wbr_wh_rwsem);
++#define wbr_wh_read_trylock_nested(wbr) \
++ au_rw_read_trylock_nested(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_write_trylock_nested(wbr) \
++ au_rw_write_trylock_nested(&(wbr)->wbr_wh_rwsem)
++*/
+
-+#define WbrWhMustNoWaiters(wbr) AuRwMustNoWaiters(&wbr->wbr_wh_rwsem)
-+#define WbrWhMustAnyLock(wbr) AuRwMustAnyLock(&wbr->wbr_wh_rwsem)
-+#define WbrWhMustWriteLock(wbr) AuRwMustWriteLock(&wbr->wbr_wh_rwsem)
++#define wbr_wh_read_unlock(wbr) au_rw_read_unlock(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_write_unlock(wbr) au_rw_write_unlock(&(wbr)->wbr_wh_rwsem)
++#define wbr_wh_downgrade_lock(wbr) au_rw_dgrade_lock(&(wbr)->wbr_wh_rwsem)
++
++#define WbrWhMustNoWaiters(wbr) AuRwMustNoWaiters(&(wbr)->wbr_wh_rwsem)
++#define WbrWhMustAnyLock(wbr) AuRwMustAnyLock(&(wbr)->wbr_wh_rwsem)
++#define WbrWhMustWriteLock(wbr) AuRwMustWriteLock(&(wbr)->wbr_wh_rwsem)
+
+/* ---------------------------------------------------------------------- */
+
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_BRANCH_H__ */
-diff -urN /usr/share/empty/fs/aufs/conf.mk linux/fs/aufs/conf.mk
---- /usr/share/empty/fs/aufs/conf.mk 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/conf.mk 2015-09-24 10:47:58.248052907 +0200
-@@ -0,0 +1,38 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/conf.mk linux-4.14/fs/aufs/conf.mk
+--- linux-4.14/fs/aufs/conf.mk 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/conf.mk 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,39 @@
+
+AuConfStr = CONFIG_AUFS_FS=${CONFIG_AUFS_FS}
+
+ XATTR \
+ FHSM \
+ RDU \
++ DIRREN \
+ SHWH \
+ BR_RAMFS \
+ BR_FUSE POLL \
+${obj}/sysfs.o: ${AuConfName}
+
+-include ${srctree}/${src}/conf_priv.mk
-diff -urN /usr/share/empty/fs/aufs/cpup.c linux/fs/aufs/cpup.c
---- /usr/share/empty/fs/aufs/cpup.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/cpup.c 2016-08-17 18:01:06.095221547 +0200
-@@ -0,0 +1,1383 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/cpup.c linux-4.14/fs/aufs/cpup.c
+--- linux-4.14/fs/aufs/cpup.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/cpup.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,1443 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ dst->f_pos = 0;
+ err = au_do_copy_file(dst, src, len, buf, blksize);
+ if (do_kfree)
-+ au_delayed_kfree(buf);
++ kfree(buf);
+ else
-+ au_delayed_free_page((unsigned long)buf);
++ free_page((unsigned long)buf);
++
++out:
++ return err;
++}
++
++static int au_do_copy(struct file *dst, struct file *src, loff_t len)
++{
++ int err;
++ struct super_block *h_src_sb;
++ struct inode *h_src_inode;
++
++ h_src_inode = file_inode(src);
++ h_src_sb = h_src_inode->i_sb;
++
++ /* XFS acquires inode_lock */
++ if (!au_test_xfs(h_src_sb))
++ err = au_copy_file(dst, src, len);
++ else {
++ inode_unlock_shared(h_src_inode);
++ err = au_copy_file(dst, src, len);
++ vfsub_inode_lock_shared_nested(h_src_inode, AuLsc_I_CHILD);
++ }
++
++ return err;
++}
++
++static int au_clone_or_copy(struct file *dst, struct file *src, loff_t len)
++{
++ int err;
++ struct super_block *h_src_sb;
++ struct inode *h_src_inode;
++
++ h_src_inode = file_inode(src);
++ h_src_sb = h_src_inode->i_sb;
++ if (h_src_sb != file_inode(dst)->i_sb
++ || !dst->f_op->clone_file_range) {
++ err = au_do_copy(dst, src, len);
++ goto out;
++ }
++
++ if (!au_test_nfs(h_src_sb)) {
++ inode_unlock_shared(h_src_inode);
++ err = vfsub_clone_file_range(src, dst, len);
++ vfsub_inode_lock_shared_nested(h_src_inode, AuLsc_I_CHILD);
++ } else
++ err = vfsub_clone_file_range(src, dst, len);
++ /* older XFS has a condition in cloning */
++ if (unlikely(err != -EOPNOTSUPP))
++ goto out;
++
++ /* the backend fs on NFS may not support cloning */
++ err = au_do_copy(dst, src, len);
+
+out:
++ AuTraceErr(err);
+ return err;
+}
+
+ .label = &&out_src
+ }
+ };
-+ struct super_block *sb;
++ struct super_block *sb, *h_src_sb;
++ struct inode *h_src_inode;
+ struct task_struct *tsk = current;
+
+ /* bsrc branch can be ro/rw. */
+ }
+
+ /* try stopping to update while we copyup */
-+ IMustLock(d_inode(file[SRC].dentry));
-+ err = au_copy_file(file[DST].file, file[SRC].file, cpg->len);
++ h_src_inode = d_inode(file[SRC].dentry);
++ h_src_sb = h_src_inode->i_sb;
++ if (!au_test_nfs(h_src_sb))
++ IMustLock(h_src_inode);
++ err = au_clone_or_copy(file[DST].file, file[SRC].file, cpg->len);
+
+ /* i wonder if we had O_NO_DELAY_FPUT flag */
+ if (tsk->flags & PF_KTHREAD)
+ cpg->len = l;
+ if (cpg->len) {
+ /* try stopping to update while we are referencing */
-+ inode_lock_nested(h_src_inode, AuLsc_I_CHILD);
++ vfsub_inode_lock_shared_nested(h_src_inode, AuLsc_I_CHILD);
+ au_pin_hdir_unlock(cpg->pin);
+
+ h_path.dentry = au_h_dptr(cpg->dentry, cpg->bsrc);
+ h_path.mnt = au_sbr_mnt(cpg->dentry->d_sb, cpg->bsrc);
+ h_src_attr->iflags = h_src_inode->i_flags;
+ if (!au_test_nfs(h_src_inode->i_sb))
-+ err = vfs_getattr(&h_path, &h_src_attr->st);
++ err = vfsub_getattr(&h_path, &h_src_attr->st);
+ else {
-+ inode_unlock(h_src_inode);
-+ err = vfs_getattr(&h_path, &h_src_attr->st);
-+ inode_lock_nested(h_src_inode, AuLsc_I_CHILD);
++ inode_unlock_shared(h_src_inode);
++ err = vfsub_getattr(&h_path, &h_src_attr->st);
++ vfsub_inode_lock_shared_nested(h_src_inode,
++ AuLsc_I_CHILD);
+ }
+ if (unlikely(err)) {
-+ inode_unlock(h_src_inode);
++ inode_unlock_shared(h_src_inode);
+ goto out;
+ }
+ h_src_attr->valid = 1;
-+ err = au_cp_regular(cpg);
-+ inode_unlock(h_src_inode);
++ if (!au_test_nfs(h_src_inode->i_sb)) {
++ err = au_cp_regular(cpg);
++ inode_unlock_shared(h_src_inode);
++ } else {
++ inode_unlock_shared(h_src_inode);
++ err = au_cp_regular(cpg);
++ }
+ rerr = au_pin_hdir_relock(cpg->pin);
+ if (!err && rerr)
+ err = rerr;
+ char *k;
+ char __user *u;
+ } sym;
-+ struct inode *h_inode = d_inode(h_src);
-+ const struct inode_operations *h_iop = h_inode->i_op;
-+
-+ err = -ENOSYS;
-+ if (unlikely(!h_iop->readlink))
-+ goto out;
+
+ err = -ENOMEM;
+ sym.k = (void *)__get_free_page(GFP_NOFS);
+ /* unnecessary to support mmap_sem since symlink is not mmap-able */
+ old_fs = get_fs();
+ set_fs(KERNEL_DS);
-+ symlen = h_iop->readlink(h_src, sym.u, PATH_MAX);
++ symlen = vfs_readlink(h_src, sym.u, PATH_MAX);
+ err = symlen;
+ set_fs(old_fs);
+
+ sym.k[symlen] = 0;
+ err = vfsub_symlink(h_dir, h_path, sym.k);
+ }
-+ au_delayed_free_page((unsigned long)sym.k);
++ free_page((unsigned long)sym.k);
+
+out:
+ return err;
+ IMustLock(h_dir);
+ AuDbg("%pd %pd\n", h_dentry, h_path->dentry);
+ /* no delegation since it is just created */
-+ err = vfsub_rename(h_dir, h_dentry, h_dir, h_path, /*delegated*/NULL);
++ err = vfsub_rename(h_dir, h_dentry, h_dir, h_path, /*delegated*/NULL,
++ /*flags*/0);
+ dput(h_path->dentry);
+
+out:
+ * @len is for truncating when it is -1 copyup the entire file.
+ * in link/rename cases, @dst_parent may be different from the real one.
+ * basic->bsrc can be larger than basic->bdst.
++ * aufs doesn't touch the credential so
++ * security_inode_copy_up{,_xattr}() are unnecrssary.
+ */
+static int au_cpup_single(struct au_cp_generic *cpg, struct dentry *dst_parent)
+{
+ }
+out_parent:
+ dput(dst_parent);
-+ au_delayed_kfree(a);
++ kfree(a);
+out:
+ return err;
+}
+ dput(parent);
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/cpup.h linux/fs/aufs/cpup.h
---- /usr/share/empty/fs/aufs/cpup.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/cpup.h 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,94 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/cpup.h linux-4.14/fs/aufs/cpup.h
+--- linux-4.14/fs/aufs/cpup.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/cpup.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,99 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#define AuCpup_RWDST (1 << 5) /* force write target even if
+ the branch is marked as RO */
+
++#ifndef CONFIG_AUFS_BR_HFSPLUS
++#undef AuCpup_HOPEN
++#define AuCpup_HOPEN 0
++#endif
++
+#define au_ftest_cpup(flags, name) ((flags) & AuCpup_##name)
+#define au_fset_cpup(flags, name) \
+ do { (flags) |= AuCpup_##name; } while (0)
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_CPUP_H__ */
-diff -urN /usr/share/empty/fs/aufs/dbgaufs.c linux/fs/aufs/dbgaufs.c
---- /usr/share/empty/fs/aufs/dbgaufs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dbgaufs.c 2016-08-17 18:01:06.095221547 +0200
-@@ -0,0 +1,435 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dbgaufs.c linux-4.14/fs/aufs/dbgaufs.c
+--- linux-4.14/fs/aufs/dbgaufs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dbgaufs.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,437 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+static int dbgaufs_xi_release(struct inode *inode __maybe_unused,
+ struct file *file)
+{
-+ au_delayed_kfree(file->private_data);
++ kfree(file->private_data);
+ return 0;
+}
+
+ if (!xf)
+ goto out;
+
-+ err = vfs_getattr(&xf->f_path, &st);
++ err = vfsub_getattr(&xf->f_path, &st);
+ if (!err) {
+ if (do_fcnt)
+ p->n = snprintf
-+ (p->a, sizeof(p->a), "%ld, %llux%lu %lld\n",
++ (p->a, sizeof(p->a), "%ld, %llux%u %lld\n",
+ (long)file_count(xf), st.blocks, st.blksize,
+ (long long)st.size);
+ else
-+ p->n = snprintf(p->a, sizeof(p->a), "%llux%lu %lld\n",
++ p->n = snprintf(p->a, sizeof(p->a), "%llux%u %lld\n",
+ st.blocks, st.blksize,
+ (long long)st.size);
+ AuDebugOn(p->n >= sizeof(p->a));
+static int dbgaufs_plink_release(struct inode *inode __maybe_unused,
+ struct file *file)
+{
-+ au_delayed_free_page((unsigned long)file->private_data);
++ free_page((unsigned long)file->private_data);
+ return 0;
+}
+
+ struct dbgaufs_plink_arg *p;
+ struct au_sbinfo *sbinfo;
+ struct super_block *sb;
-+ struct au_sphlhead *sphl;
++ struct hlist_bl_head *hbl;
+
+ err = -ENOMEM;
+ p = (void *)get_zeroed_page(GFP_NOFS);
+ limit -= n;
+
+ sum = 0;
-+ for (i = 0, sphl = sbinfo->si_plink;
-+ i < AuPlink_NHASH;
-+ i++, sphl++) {
-+ n = au_sphl_count(sphl);
++ for (i = 0, hbl = sbinfo->si_plink; i < AuPlink_NHASH;
++ i++, hbl++) {
++ n = au_hbl_count(hbl);
+ sum += n;
+
+ n = snprintf(p->a + p->n, limit, "%lu ", n);
+ goto out; /* success */
+
+out_free:
-+ au_delayed_free_page((unsigned long)p);
++ free_page((unsigned long)p);
+out:
+ return err;
+}
+ for (; bindex <= bbot; bindex++) {
+ br = au_sbr(sb, bindex);
+ xi = &br->br_xino;
++ /* debugfs acquires the parent i_mutex */
++ lockdep_off();
+ debugfs_remove(xi->xi_dbgaufs);
++ lockdep_on();
+ xi->xi_dbgaufs = NULL;
+ }
+}
+ err = 0;
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/dbgaufs.h linux/fs/aufs/dbgaufs.h
---- /usr/share/empty/fs/aufs/dbgaufs.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dbgaufs.h 2016-07-25 19:05:34.811159821 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dbgaufs.h linux-4.14/fs/aufs/dbgaufs.h
+--- linux-4.14/fs/aufs/dbgaufs.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dbgaufs.h 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,48 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#endif /* __KERNEL__ */
+#endif /* __DBGAUFS_H__ */
-diff -urN /usr/share/empty/fs/aufs/dcsub.c linux/fs/aufs/dcsub.c
---- /usr/share/empty/fs/aufs/dcsub.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dcsub.c 2016-08-17 18:01:06.101888388 +0200
-@@ -0,0 +1,224 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dcsub.c linux-4.14/fs/aufs/dcsub.c
+--- linux-4.14/fs/aufs/dcsub.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dcsub.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,225 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ p = dpage->dentries;
+ for (i = 0; i < dpage->ndentry; i++)
+ dput(*p++);
-+ au_delayed_free_page((unsigned long)dpage->dentries);
++ free_page((unsigned long)dpage->dentries);
+}
+
+int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp)
+ return 0; /* success */
+
+out_dpages:
-+ au_delayed_kfree(dpages->dpages);
++ kfree(dpages->dpages);
+out:
+ return err;
+}
+ p = dpages->dpages;
+ for (i = 0; i < dpages->ndpage; i++)
+ au_dpage_free(p++);
-+ au_delayed_kfree(dpages->dpages);
++ kfree(dpages->dpages);
+}
+
+static int au_dpages_append(struct au_dcsub_pages *dpages,
+ err = -ENOMEM;
+ sz = dpages->ndpage * sizeof(*dpages->dpages);
+ p = au_kzrealloc(dpages->dpages, sz,
-+ sz + sizeof(*dpages->dpages), gfp);
++ sz + sizeof(*dpages->dpages), gfp,
++ /*may_shrink*/0);
+ if (unlikely(!p))
+ goto out;
+
+
+ return path_is_under(path + 0, path + 1);
+}
-diff -urN /usr/share/empty/fs/aufs/dcsub.h linux/fs/aufs/dcsub.h
---- /usr/share/empty/fs/aufs/dcsub.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dcsub.h 2016-07-25 19:05:34.811159821 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dcsub.h linux-4.14/fs/aufs/dcsub.h
+--- linux-4.14/fs/aufs/dcsub.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dcsub.h 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,136 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DCSUB_H__ */
-diff -urN /usr/share/empty/fs/aufs/debug.c linux/fs/aufs/debug.c
---- /usr/share/empty/fs/aufs/debug.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/debug.c 2016-08-17 18:01:06.101888388 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/debug.c linux-4.14/fs/aufs/debug.c
+--- linux-4.14/fs/aufs/debug.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/debug.c 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,440 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ au_br_count_init(&a->fake);
+ err = do_pri_br(-1, &a->fake);
+ au_br_count_fin(&a->fake);
-+ au_delayed_kfree(a);
++ kfree(a);
+ dpri("dev 0x%x\n", sb->s_dev);
+ if (err || !au_test_aufs(sb))
+ return;
+ return;
+ dpri("nw %d, gen %u, kobj %d\n",
+ atomic_read(&sbinfo->si_nowait.nw_len), sbinfo->si_generation,
-+ atomic_read(&sbinfo->si_kobj.kref.refcount));
++ kref_read(&sbinfo->si_kobj.kref));
+ for (bindex = 0; bindex <= sbinfo->si_bbot; bindex++)
+ do_pri_br(bindex, sbinfo->si_branch[0 + bindex]);
+}
+
+ return 0;
+}
-diff -urN /usr/share/empty/fs/aufs/debug.h linux/fs/aufs/debug.h
---- /usr/share/empty/fs/aufs/debug.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/debug.h 2016-07-25 19:05:34.811159821 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/debug.h linux-4.14/fs/aufs/debug.h
+--- linux-4.14/fs/aufs/debug.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/debug.h 2021-02-24 21:42:43.441114748 +0100
@@ -0,0 +1,225 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DEBUG_H__ */
-diff -urN /usr/share/empty/fs/aufs/dentry.c linux/fs/aufs/dentry.c
---- /usr/share/empty/fs/aufs/dentry.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dentry.c 2016-08-17 18:01:06.111888648 +0200
-@@ -0,0 +1,1128 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dentry.c linux-4.14/fs/aufs/dentry.c
+--- linux-4.14/fs/aufs/dentry.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dentry.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,1152 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#include <linux/namei.h>
+#include "aufs.h"
+
-+struct au_do_lookup_args {
-+ unsigned int flags;
-+ mode_t type;
-+};
-+
+/*
+ * returns positive/negative dentry, NULL or an error.
+ * NULL means whiteout-ed or not-found.
+ */
+static struct dentry*
+au_do_lookup(struct dentry *h_parent, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct qstr *wh_name,
-+ struct au_do_lookup_args *args)
++ aufs_bindex_t bindex, struct au_do_lookup_args *args)
+{
+ struct dentry *h_dentry;
+ struct inode *h_inode;
+ br = au_sbr(dentry->d_sb, bindex);
+ wh_able = !!au_br_whable(br->br_perm);
+ if (wh_able)
-+ wh_found = au_wh_test(h_parent, wh_name, /*try_sio*/0);
++ wh_found = au_wh_test(h_parent, &args->whname, ignore_perm);
+ h_dentry = ERR_PTR(wh_found);
+ if (!wh_found)
+ goto real_lookup;
+
+real_lookup:
+ if (!ignore_perm)
-+ h_dentry = vfsub_lkup_one(&dentry->d_name, h_parent);
++ h_dentry = vfsub_lkup_one(args->name, h_parent);
+ else
-+ h_dentry = au_sio_lkup_one(&dentry->d_name, h_parent);
++ h_dentry = au_sio_lkup_one(args->name, h_parent);
+ if (IS_ERR(h_dentry)) {
+ if (PTR_ERR(h_dentry) == -ENAMETOOLONG
+ && !allow_neg)
+ } else if (wh_found
+ || (args->type && args->type != (h_inode->i_mode & S_IFMT)))
+ goto out_neg;
++ else if (au_ftest_lkup(args->flags, DIRREN)
++ /* && h_inode */
++ && !au_dr_lkup_h_ino(args, bindex, h_inode->i_ino)) {
++ AuDbg("b%d %pd ignored hi%llu\n", bindex, h_dentry,
++ (unsigned long long)h_inode->i_ino);
++ goto out_neg;
++ }
+
+ if (au_dbbot(dentry) <= bindex)
+ au_set_dbbot(dentry, bindex);
+ || (d_really_is_positive(dentry) && !d_is_dir(dentry)))
+ goto out; /* success */
+
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ vfsub_inode_lock_shared_nested(h_inode, AuLsc_I_CHILD);
+ opq = au_diropq_test(h_dentry);
-+ inode_unlock(h_inode);
++ inode_unlock_shared(h_inode);
+ if (opq > 0)
+ au_set_dbdiropq(dentry, bindex);
+ else if (unlikely(opq < 0)) {
+{
+ int npositive, err;
+ aufs_bindex_t bindex, btail, bdiropq;
-+ unsigned char isdir, dirperm1;
-+ struct qstr whname;
++ unsigned char isdir, dirperm1, dirren;
+ struct au_do_lookup_args args = {
-+ .flags = flags
++ .flags = flags,
++ .name = &dentry->d_name
+ };
-+ const struct qstr *name = &dentry->d_name;
+ struct dentry *parent;
+ struct super_block *sb;
+
+ sb = dentry->d_sb;
-+ err = au_test_shwh(sb, name);
++ err = au_test_shwh(sb, args.name);
+ if (unlikely(err))
+ goto out;
+
-+ err = au_wh_name_alloc(&whname, name);
++ err = au_wh_name_alloc(&args.whname, args.name);
+ if (unlikely(err))
+ goto out;
+
+ isdir = !!d_is_dir(dentry);
+ dirperm1 = !!au_opt_test(au_mntflags(sb), DIRPERM1);
++ dirren = !!au_opt_test(au_mntflags(sb), DIRREN);
++ if (dirren)
++ au_fset_lkup(args.flags, DIRREN);
+
+ npositive = 0;
+ parent = dget_parent(dentry);
+ for (bindex = btop; bindex <= btail; bindex++) {
+ struct dentry *h_parent, *h_dentry;
+ struct inode *h_inode, *h_dir;
++ struct au_branch *br;
+
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (h_dentry) {
+ if (!h_parent || !d_is_dir(h_parent))
+ continue;
+
++ if (dirren) {
++ /* if the inum matches, then use the prepared name */
++ err = au_dr_lkup_name(&args, bindex);
++ if (unlikely(err))
++ goto out_parent;
++ }
++
+ h_dir = d_inode(h_parent);
-+ inode_lock_nested(h_dir, AuLsc_I_PARENT);
-+ h_dentry = au_do_lookup(h_parent, dentry, bindex, &whname,
-+ &args);
-+ inode_unlock(h_dir);
++ vfsub_inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
++ h_dentry = au_do_lookup(h_parent, dentry, bindex, &args);
++ inode_unlock_shared(h_dir);
+ err = PTR_ERR(h_dentry);
+ if (IS_ERR(h_dentry))
+ goto out_parent;
+ if (bdiropq >= 0 && bdiropq <= bindex)
+ break;
+ }
++ br = au_sbr(sb, bindex);
++ if (dirren
++ && au_dr_hino_test_add(&br->br_dirren, h_inode->i_ino,
++ /*add_ent*/NULL)) {
++ /* prepare next name to lookup */
++ err = au_dr_lkup(&args, dentry, bindex);
++ if (unlikely(err))
++ goto out_parent;
++ }
+ }
+
+ if (npositive) {
+
+out_parent:
+ dput(parent);
-+ au_delayed_kfree(whname.name);
++ kfree(args.whname.name);
++ if (dirren)
++ au_dr_lkup_fin(&args);
+out:
+ return err;
+}
+
+int au_refresh_dentry(struct dentry *dentry, struct dentry *parent)
+{
-+ int err, ebrange;
++ int err, ebrange, nbr;
+ unsigned int sigen;
+ struct au_dinfo *dinfo, *tmp;
+ struct super_block *sb;
+ if (unlikely(err))
+ goto out;
+
++ nbr = au_sbbot(sb) + 1;
+ dinfo = au_di(dentry);
-+ err = au_di_realloc(dinfo, au_sbbot(sb) + 1);
++ err = au_di_realloc(dinfo, nbr, /*may_shrink*/0);
+ if (unlikely(err))
+ goto out;
+ ebrange = au_dbrange_test(dentry);
+ au_dbg_verify_dinode(dentry);
+ AuTraceErr(err);
+ }
++ au_di_realloc(dinfo, nbr, /*may_shrink*/1); /* harmless if err */
+ au_rw_write_unlock(&tmp->di_rwsem);
+ au_di_free(tmp);
+ if (unlikely(err))
+
+/* todo: remove this */
+static int h_d_revalidate(struct dentry *dentry, struct inode *inode,
-+ unsigned int flags, int do_udba)
++ unsigned int flags, int do_udba, int dirren)
+{
+ int err;
+ umode_t mode, h_mode;
+ && !is_root
+ && ((!h_nfs
+ && (unhashed != !!d_unhashed(h_dentry)
-+ || (!tmpfile
++ || (!tmpfile && !dirren
+ && !au_qstreq(name, h_name))
+ ))
+ || (h_nfs
+{
+ int valid, err;
+ unsigned int sigen;
-+ unsigned char do_udba;
++ unsigned char do_udba, dirren;
+ struct super_block *sb;
+ struct inode *inode;
+
+ }
+ }
+
-+ err = h_d_revalidate(dentry, inode, flags, do_udba);
++ dirren = !!au_opt_test(au_mntflags(sb), DIRREN);
++ err = h_d_revalidate(dentry, inode, flags, do_udba, dirren);
+ if (unlikely(!err && do_udba && au_dbtop(dentry) < 0)) {
+ err = -EIO;
+ AuDbg("both of real entry and whiteout found, %p, err %d\n",
+const struct dentry_operations aufs_dop_noreval = {
+ .d_release = aufs_d_release
+};
-diff -urN /usr/share/empty/fs/aufs/dentry.h linux/fs/aufs/dentry.h
---- /usr/share/empty/fs/aufs/dentry.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dentry.h 2016-08-17 18:01:06.111888648 +0200
-@@ -0,0 +1,255 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dentry.h linux-4.14/fs/aufs/dentry.h
+--- linux-4.14/fs/aufs/dentry.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dentry.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,266 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#ifdef __KERNEL__
+
+#include <linux/dcache.h>
++#include "dirren.h"
+#include "rwsem.h"
+
+struct au_hdentry {
+ struct au_rwsem di_rwsem;
+ aufs_bindex_t di_btop, di_bbot, di_bwh, di_bdiropq;
+ unsigned char di_tmpfile; /* to allow the different name */
-+ union {
-+ struct au_hdentry *di_hdentry;
-+ struct llist_node di_lnode; /* delayed free */
-+ };
++ struct au_hdentry *di_hdentry;
+} ____cacheline_aligned_in_smp;
+
+/* ---------------------------------------------------------------------- */
+/* flags for au_lkup_dentry() */
+#define AuLkup_ALLOW_NEG 1
+#define AuLkup_IGNORE_PERM (1 << 1)
++#define AuLkup_DIRREN (1 << 2)
+#define au_ftest_lkup(flags, name) ((flags) & AuLkup_##name)
+#define au_fset_lkup(flags, name) \
+ do { (flags) |= AuLkup_##name; } while (0)
+#define au_fclr_lkup(flags, name) \
+ do { (flags) &= ~AuLkup_##name; } while (0)
+
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuLkup_DIRREN
++#define AuLkup_DIRREN 0
++#endif
++
++struct au_do_lookup_args {
++ unsigned int flags;
++ mode_t type;
++ struct qstr whname, *name;
++ struct au_dr_lookup dirren;
++};
++
+/* ---------------------------------------------------------------------- */
+
+/* dentry.c */
+void au_di_cp(struct au_dinfo *dst, struct au_dinfo *src);
+int au_di_init(struct dentry *dentry);
+void au_di_fin(struct dentry *dentry);
-+int au_di_realloc(struct au_dinfo *dinfo, int nbr);
++int au_di_realloc(struct au_dinfo *dinfo, int nbr, int may_shrink);
+
+void di_read_lock(struct dentry *d, int flags, unsigned int lsc);
+void di_read_unlock(struct dentry *d, int flags);
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DENTRY_H__ */
-diff -urN /usr/share/empty/fs/aufs/dinfo.c linux/fs/aufs/dinfo.c
---- /usr/share/empty/fs/aufs/dinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dinfo.c 2016-08-17 18:01:06.111888648 +0200
-@@ -0,0 +1,552 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dinfo.c linux-4.14/fs/aufs/dinfo.c
+--- linux-4.14/fs/aufs/dinfo.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dinfo.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,553 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ goto out;
+ }
+
-+ au_cache_dfree_dinfo(dinfo);
++ au_cache_free_dinfo(dinfo);
+ dinfo = NULL;
+
+out:
+ while (bindex++ <= bbot)
+ au_hdput(p++);
+ }
-+ au_delayed_kfree(dinfo->di_hdentry);
-+ au_cache_dfree_dinfo(dinfo);
++ kfree(dinfo->di_hdentry);
++ au_cache_free_dinfo(dinfo);
+}
+
+void au_di_swap(struct au_dinfo *a, struct au_dinfo *b)
+ au_di_free(dinfo);
+}
+
-+int au_di_realloc(struct au_dinfo *dinfo, int nbr)
++int au_di_realloc(struct au_dinfo *dinfo, int nbr, int may_shrink)
+{
+ int err, sz;
+ struct au_hdentry *hdp;
+ sz = sizeof(*hdp) * (dinfo->di_bbot + 1);
+ if (!sz)
+ sz = sizeof(*hdp);
-+ hdp = au_kzrealloc(dinfo->di_hdentry, sz, sizeof(*hdp) * nbr, GFP_NOFS);
++ hdp = au_kzrealloc(dinfo->di_hdentry, sz, sizeof(*hdp) * nbr, GFP_NOFS,
++ may_shrink);
+ if (hdp) {
+ dinfo->di_hdentry = hdp;
+ err = 0;
+ || d_inode(d1) == d_inode(d2)
+ || d1->d_sb != d2->d_sb);
+
-+ if (isdir && au_test_subdir(d1, d2)) {
++ if ((isdir && au_test_subdir(d1, d2))
++ || d1 < d2) {
+ di_write_lock_child(d1);
+ di_write_lock_child2(d2);
+ } else {
-+ /* there should be no races */
+ di_write_lock_child(d2);
+ di_write_lock_child2(d1);
+ }
+ || d_inode(d1) == d_inode(d2)
+ || d1->d_sb != d2->d_sb);
+
-+ if (isdir && au_test_subdir(d1, d2)) {
++ if ((isdir && au_test_subdir(d1, d2))
++ || d1 < d2) {
+ di_write_lock_parent(d1);
+ di_write_lock_parent2(d2);
+ } else {
-+ /* there should be no races */
+ di_write_lock_parent(d2);
+ di_write_lock_parent2(d1);
+ }
+ return bindex;
+ return -1;
+}
-diff -urN /usr/share/empty/fs/aufs/dir.c linux/fs/aufs/dir.c
---- /usr/share/empty/fs/aufs/dir.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dir.c 2016-08-17 18:01:06.111888648 +0200
-@@ -0,0 +1,762 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dir.c linux-4.14/fs/aufs/dir.c
+--- linux-4.14/fs/aufs/dir.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dir.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,759 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+out:
+ dput(a->dentry);
+ au_nwt_done(&au_sbi(sb)->si_nowait);
-+ au_delayed_kfree(arg);
++ kfree(arg);
+}
+
+void au_dir_ts(struct inode *dir, aufs_bindex_t bindex)
+ if (unlikely(wkq_err)) {
+ pr_err("wkq %d\n", wkq_err);
+ dput(dentry);
-+ au_delayed_kfree(arg);
++ kfree(arg);
+ }
+
+out:
+ };
+ err = au_do_open(file, &args);
+ if (unlikely(err))
-+ au_delayed_kfree(fidir);
++ kfree(fidir);
+ }
+ si_read_unlock(sb);
+ return err;
+ struct au_fidir *fidir;
+ struct au_hfile *hf;
+ aufs_bindex_t bindex, bbot;
-+ int execed, delayed;
+
-+ delayed = (current->flags & PF_KTHREAD) || in_interrupt();
+ finfo = au_fi(file);
+ fidir = finfo->fi_hdir;
+ if (fidir) {
-+ au_sphl_del(&finfo->fi_hlist,
-+ &au_sbi(file->f_path.dentry->d_sb)->si_files);
++ au_hbl_del(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
+ vdir_cache = fidir->fd_vdir_cache; /* lock-free */
+ if (vdir_cache)
-+ au_vdir_free(vdir_cache, delayed);
++ au_vdir_free(vdir_cache);
+
+ bindex = finfo->fi_btop;
+ if (bindex >= 0) {
-+ execed = vfsub_file_execed(file);
+ hf = fidir->fd_hfile + bindex;
+ /*
+ * calls fput() instead of filp_close(),
+ bbot = fidir->fd_bbot;
+ for (; bindex <= bbot; bindex++, hf++)
+ if (hf->hf_file)
-+ au_hfput(hf, execed);
++ au_hfput(hf, /*execed*/0);
+ }
-+ au_delayed_kfree(fidir);
++ kfree(fidir);
+ finfo->fi_hdir = NULL;
+ }
-+ au_finfo_fin(file, delayed);
++ au_finfo_fin(file);
+ return 0;
+}
+
+ struct super_block *sb;
+ struct inode *inode;
+
-+ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1);
++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1, /*fi_lsc*/0);
+ if (unlikely(err))
+ goto out;
+
+
+ sb = dentry->d_sb;
+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1);
++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1, /*fi_lsc*/0);
+ if (unlikely(err))
+ goto out;
+ err = au_alive_dir(dentry);
+ h_dentry = au_h_dptr(dentry, arg->bindex);
+ h_inode = d_inode(h_dentry);
+ /* todo: i_mode changes anytime? */
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ vfsub_inode_lock_shared_nested(h_inode, AuLsc_I_CHILD);
+ err = au_test_h_perm_sio(h_inode, MAY_EXEC | MAY_READ);
-+ inode_unlock(h_inode);
++ inode_unlock_shared(h_inode);
+ if (!err)
+ err = do_test_empty(dentry, arg);
+ else {
+ .flush = aufs_flush_dir,
+ .fsync = aufs_fsync_dir
+};
-diff -urN /usr/share/empty/fs/aufs/dir.h linux/fs/aufs/dir.h
---- /usr/share/empty/fs/aufs/dir.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dir.h 2016-08-17 18:01:06.111888648 +0200
-@@ -0,0 +1,137 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dir.h linux-4.14/fs/aufs/dir.h
+--- linux-4.14/fs/aufs/dir.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dir.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,131 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+struct au_vdir_dehstr {
+ struct hlist_node hash;
-+ union {
-+ struct au_vdir_destr *str;
-+ struct llist_node lnode; /* delayed free */
-+ };
++ struct au_vdir_destr *str;
+} ____cacheline_aligned_in_smp;
+
+struct au_vdir_de {
+
+ unsigned long vd_version;
+ unsigned int vd_deblk_sz;
-+ union {
-+ unsigned long vd_jiffy;
-+ struct llist_node vd_lnode; /* delayed free */
-+ };
++ unsigned long vd_jiffy;
+} ____cacheline_aligned_in_smp;
+
+/* ---------------------------------------------------------------------- */
+int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino,
+ unsigned int d_type, aufs_bindex_t bindex,
+ unsigned char shwh);
-+void au_vdir_free(struct au_vdir *vdir, int atonce);
++void au_vdir_free(struct au_vdir *vdir);
+int au_vdir_init(struct file *file);
+int au_vdir_fill_de(struct file *file, struct dir_context *ctx);
+
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DIR_H__ */
-diff -urN /usr/share/empty/fs/aufs/dynop.c linux/fs/aufs/dynop.c
---- /usr/share/empty/fs/aufs/dynop.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dynop.c 2016-08-17 18:01:21.295617591 +0200
-@@ -0,0 +1,369 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dirren.c linux-4.14/fs/aufs/dirren.c
+--- linux-4.14/fs/aufs/dirren.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dirren.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,1315 @@
+/*
-+ * Copyright (C) 2010-2016 Junjiro R. Okajima
++ * Copyright (C) 2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * dynamically customizable operations for regular files
++ * special handling in renaming a directoy
++ * in order to support looking-up the before-renamed name on the lower readonly
++ * branches
++ */
++
++#include <linux/byteorder/generic.h>
++#include "aufs.h"
++
++static void au_dr_hino_del(struct au_dr_br *dr, struct au_dr_hino *ent)
++{
++ int idx;
++
++ idx = au_dr_ihash(ent->dr_h_ino);
++ au_hbl_del(&ent->dr_hnode, dr->dr_h_ino + idx);
++}
++
++static int au_dr_hino_test_empty(struct au_dr_br *dr)
++{
++ int ret, i;
++ struct hlist_bl_head *hbl;
++
++ ret = 1;
++ for (i = 0; ret && i < AuDirren_NHASH; i++) {
++ hbl = dr->dr_h_ino + i;
++ hlist_bl_lock(hbl);
++ ret &= hlist_bl_empty(hbl);
++ hlist_bl_unlock(hbl);
++ }
++
++ return ret;
++}
++
++static struct au_dr_hino *au_dr_hino_find(struct au_dr_br *dr, ino_t ino)
++{
++ struct au_dr_hino *found, *ent;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ int idx;
++
++ found = NULL;
++ idx = au_dr_ihash(ino);
++ hbl = dr->dr_h_ino + idx;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(ent, pos, hbl, dr_hnode)
++ if (ent->dr_h_ino == ino) {
++ found = ent;
++ break;
++ }
++ hlist_bl_unlock(hbl);
++
++ return found;
++}
++
++int au_dr_hino_test_add(struct au_dr_br *dr, ino_t ino,
++ struct au_dr_hino *add_ent)
++{
++ int found, idx;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dr_hino *ent;
++
++ found = 0;
++ idx = au_dr_ihash(ino);
++ hbl = dr->dr_h_ino + idx;
++#if 0
++ {
++ struct hlist_bl_node *tmp;
++
++ hlist_bl_for_each_entry_safe(ent, pos, tmp, hbl, dr_hnode)
++ AuDbg("hi%llu\n", (unsigned long long)ent->dr_h_ino);
++ }
++#endif
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(ent, pos, hbl, dr_hnode)
++ if (ent->dr_h_ino == ino) {
++ found = 1;
++ break;
++ }
++ if (!found && add_ent)
++ hlist_bl_add_head(&add_ent->dr_hnode, hbl);
++ hlist_bl_unlock(hbl);
++
++ if (!found && add_ent)
++ AuDbg("i%llu added\n", (unsigned long long)add_ent->dr_h_ino);
++
++ return found;
++}
++
++void au_dr_hino_free(struct au_dr_br *dr)
++{
++ int i;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
++ struct au_dr_hino *ent;
++
++ /* SiMustWriteLock(sb); */
++
++ for (i = 0; i < AuDirren_NHASH; i++) {
++ hbl = dr->dr_h_ino + i;
++ /* no spinlock since sbinfo must be write-locked */
++ hlist_bl_for_each_entry_safe(ent, pos, tmp, hbl, dr_hnode)
++ kfree(ent);
++ INIT_HLIST_BL_HEAD(hbl);
++ }
++}
++
++/* returns the number of inodes or an error */
++static int au_dr_hino_store(struct super_block *sb, struct au_branch *br,
++ struct file *hinofile)
++{
++ int err, i;
++ ssize_t ssz;
++ loff_t pos, oldsize;
++ __be64 u64;
++ struct inode *hinoinode;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *n1, *n2;
++ struct au_dr_hino *ent;
++
++ SiMustWriteLock(sb);
++ AuDebugOn(!au_br_writable(br->br_perm));
++
++ hinoinode = file_inode(hinofile);
++ oldsize = i_size_read(hinoinode);
++
++ err = 0;
++ pos = 0;
++ hbl = br->br_dirren.dr_h_ino;
++ for (i = 0; !err && i < AuDirren_NHASH; i++, hbl++) {
++ /* no bit-lock since sbinfo must be write-locked */
++ hlist_bl_for_each_entry_safe(ent, n1, n2, hbl, dr_hnode) {
++ AuDbg("hi%llu, %pD2\n",
++ (unsigned long long)ent->dr_h_ino, hinofile);
++ u64 = cpu_to_be64(ent->dr_h_ino);
++ ssz = vfsub_write_k(hinofile, &u64, sizeof(u64), &pos);
++ if (ssz == sizeof(u64))
++ continue;
++
++ /* write error */
++ pr_err("ssz %zd, %pD2\n", ssz, hinofile);
++ err = -ENOSPC;
++ if (ssz < 0)
++ err = ssz;
++ break;
++ }
++ }
++ /* regardless the error */
++ if (pos < oldsize) {
++ err = vfsub_trunc(&hinofile->f_path, pos, /*attr*/0, hinofile);
++ AuTraceErr(err);
++ }
++
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_dr_hino_load(struct au_dr_br *dr, struct file *hinofile)
++{
++ int err, hidx;
++ ssize_t ssz;
++ size_t sz, n;
++ loff_t pos;
++ uint64_t u64;
++ struct au_dr_hino *ent;
++ struct inode *hinoinode;
++ struct hlist_bl_head *hbl;
++
++ err = 0;
++ pos = 0;
++ hbl = dr->dr_h_ino;
++ hinoinode = file_inode(hinofile);
++ sz = i_size_read(hinoinode);
++ AuDebugOn(sz % sizeof(u64));
++ n = sz / sizeof(u64);
++ while (n--) {
++ ssz = vfsub_read_k(hinofile, &u64, sizeof(u64), &pos);
++ if (unlikely(ssz != sizeof(u64))) {
++ pr_err("ssz %zd, %pD2\n", ssz, hinofile);
++ err = -EINVAL;
++ if (ssz < 0)
++ err = ssz;
++ goto out_free;
++ }
++
++ ent = kmalloc(sizeof(*ent), GFP_NOFS);
++ if (!ent) {
++ err = -ENOMEM;
++ AuTraceErr(err);
++ goto out_free;
++ }
++ ent->dr_h_ino = be64_to_cpu((__force __be64)u64);
++ AuDbg("hi%llu, %pD2\n",
++ (unsigned long long)ent->dr_h_ino, hinofile);
++ hidx = au_dr_ihash(ent->dr_h_ino);
++ au_hbl_add(&ent->dr_hnode, hbl + hidx);
++ }
++ goto out; /* success */
++
++out_free:
++ au_dr_hino_free(dr);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/*
++ * @bindex/@br is a switch to distinguish whether suspending hnotify or not.
++ * @path is a switch to distinguish load and store.
+ */
++static int au_dr_hino(struct super_block *sb, aufs_bindex_t bindex,
++ struct au_branch *br, const struct path *path)
++{
++ int err, flags;
++ unsigned char load, suspend;
++ struct file *hinofile;
++ struct au_hinode *hdir;
++ struct inode *dir, *delegated;
++ struct path hinopath;
++ struct qstr hinoname = QSTR_INIT(AUFS_WH_DR_BRHINO,
++ sizeof(AUFS_WH_DR_BRHINO) - 1);
++
++ AuDebugOn(bindex < 0 && !br);
++ AuDebugOn(bindex >= 0 && br);
++
++ err = -EINVAL;
++ suspend = !br;
++ if (suspend)
++ br = au_sbr(sb, bindex);
++ load = !!path;
++ if (!load) {
++ path = &br->br_path;
++ AuDebugOn(!au_br_writable(br->br_perm));
++ if (unlikely(!au_br_writable(br->br_perm)))
++ goto out;
++ }
++
++ hdir = NULL;
++ if (suspend) {
++ dir = d_inode(sb->s_root);
++ hdir = au_hinode(au_ii(dir), bindex);
++ dir = hdir->hi_inode;
++ au_hn_inode_lock_nested(hdir, AuLsc_I_CHILD);
++ } else {
++ dir = d_inode(path->dentry);
++ inode_lock_nested(dir, AuLsc_I_CHILD);
++ }
++ hinopath.dentry = vfsub_lkup_one(&hinoname, path->dentry);
++ err = PTR_ERR(hinopath.dentry);
++ if (IS_ERR(hinopath.dentry))
++ goto out_unlock;
++
++ err = 0;
++ flags = O_RDONLY;
++ if (load) {
++ if (d_is_negative(hinopath.dentry))
++ goto out_dput; /* success */
++ } else {
++ if (au_dr_hino_test_empty(&br->br_dirren)) {
++ if (d_is_positive(hinopath.dentry)) {
++ delegated = NULL;
++ err = vfsub_unlink(dir, &hinopath, &delegated,
++ /*force*/0);
++ AuTraceErr(err);
++ if (unlikely(err))
++ pr_err("ignored err %d, %pd2\n",
++ err, hinopath.dentry);
++ if (unlikely(err == -EWOULDBLOCK))
++ iput(delegated);
++ err = 0;
++ }
++ goto out_dput;
++ } else if (!d_is_positive(hinopath.dentry)) {
++ err = vfsub_create(dir, &hinopath, 0600,
++ /*want_excl*/false);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out_dput;
++ }
++ flags = O_WRONLY;
++ }
++ hinopath.mnt = path->mnt;
++ hinofile = vfsub_dentry_open(&hinopath, flags);
++ if (suspend)
++ au_hn_inode_unlock(hdir);
++ else
++ inode_unlock(dir);
++ dput(hinopath.dentry);
++ AuTraceErrPtr(hinofile);
++ if (IS_ERR(hinofile)) {
++ err = PTR_ERR(hinofile);
++ goto out;
++ }
+
-+#include "aufs.h"
++ if (load)
++ err = au_dr_hino_load(&br->br_dirren, hinofile);
++ else
++ err = au_dr_hino_store(sb, br, hinofile);
++ fput(hinofile);
++ goto out;
+
-+#define DyPrSym(key) AuDbgSym(key->dk_op.dy_hop)
++out_dput:
++ dput(hinopath.dentry);
++out_unlock:
++ if (suspend)
++ au_hn_inode_unlock(hdir);
++ else
++ inode_unlock(dir);
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+/*
-+ * How large will these lists be?
-+ * Usually just a few elements, 20-30 at most for each, I guess.
-+ */
-+static struct au_sphlhead dynop[AuDyLast];
++/* ---------------------------------------------------------------------- */
+
-+static struct au_dykey *dy_gfind_get(struct au_sphlhead *sphl, const void *h_op)
++static int au_dr_brid_init(struct au_dr_brid *brid, const struct path *path)
+{
-+ struct au_dykey *key, *tmp;
-+ struct hlist_head *head;
-+
-+ key = NULL;
-+ head = &sphl->head;
-+ rcu_read_lock();
-+ hlist_for_each_entry_rcu(tmp, head, dk_hnode)
-+ if (tmp->dk_op.dy_hop == h_op) {
-+ key = tmp;
-+ kref_get(&key->dk_kref);
-+ break;
-+ }
-+ rcu_read_unlock();
++ int err;
++ struct kstatfs kstfs;
++ dev_t dev;
++ struct dentry *dentry;
++ struct super_block *sb;
+
-+ return key;
-+}
++ err = vfs_statfs((void *)path, &kstfs);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out;
+
-+static struct au_dykey *dy_bradd(struct au_branch *br, struct au_dykey *key)
-+{
-+ struct au_dykey **k, *found;
-+ const void *h_op = key->dk_op.dy_hop;
-+ int i;
++ /* todo: support for UUID */
+
-+ found = NULL;
-+ k = br->br_dykey;
-+ for (i = 0; i < AuBrDynOp; i++)
-+ if (k[i]) {
-+ if (k[i]->dk_op.dy_hop == h_op) {
-+ found = k[i];
-+ break;
-+ }
-+ } else
-+ break;
-+ if (!found) {
-+ spin_lock(&br->br_dykey_lock);
-+ for (; i < AuBrDynOp; i++)
-+ if (k[i]) {
-+ if (k[i]->dk_op.dy_hop == h_op) {
-+ found = k[i];
-+ break;
-+ }
-+ } else {
-+ k[i] = key;
-+ break;
-+ }
-+ spin_unlock(&br->br_dykey_lock);
-+ BUG_ON(i == AuBrDynOp); /* expand the array */
++ if (kstfs.f_fsid.val[0] || kstfs.f_fsid.val[1]) {
++ brid->type = AuBrid_FSID;
++ brid->fsid = kstfs.f_fsid;
++ } else {
++ dentry = path->dentry;
++ sb = dentry->d_sb;
++ dev = sb->s_dev;
++ if (dev) {
++ brid->type = AuBrid_DEV;
++ brid->dev = dev;
++ }
+ }
+
-+ return found;
++out:
++ return err;
+}
+
-+/* kref_get() if @key is already added */
-+static struct au_dykey *dy_gadd(struct au_sphlhead *sphl, struct au_dykey *key)
++int au_dr_br_init(struct super_block *sb, struct au_branch *br,
++ const struct path *path)
+{
-+ struct au_dykey *tmp, *found;
-+ struct hlist_head *head;
-+ const void *h_op = key->dk_op.dy_hop;
++ int err, i;
++ struct au_dr_br *dr;
++ struct hlist_bl_head *hbl;
+
-+ found = NULL;
-+ head = &sphl->head;
-+ spin_lock(&sphl->spin);
-+ hlist_for_each_entry(tmp, head, dk_hnode)
-+ if (tmp->dk_op.dy_hop == h_op) {
-+ kref_get(&tmp->dk_kref);
-+ found = tmp;
-+ break;
-+ }
-+ if (!found)
-+ hlist_add_head_rcu(&key->dk_hnode, head);
-+ spin_unlock(&sphl->spin);
++ dr = &br->br_dirren;
++ hbl = dr->dr_h_ino;
++ for (i = 0; i < AuDirren_NHASH; i++, hbl++)
++ INIT_HLIST_BL_HEAD(hbl);
+
-+ if (!found)
-+ DyPrSym(key);
-+ return found;
-+}
++ err = au_dr_brid_init(&dr->dr_brid, path);
++ if (unlikely(err))
++ goto out;
+
-+static void dy_free_rcu(struct rcu_head *rcu)
-+{
-+ struct au_dykey *key;
++ if (au_opt_test(au_mntflags(sb), DIRREN))
++ err = au_dr_hino(sb, /*bindex*/-1, br, path);
+
-+ key = container_of(rcu, struct au_dykey, dk_rcu);
-+ DyPrSym(key);
-+ kfree(key); /* not delayed */
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static void dy_free(struct kref *kref)
++int au_dr_br_fin(struct super_block *sb, struct au_branch *br)
+{
-+ struct au_dykey *key;
-+ struct au_sphlhead *sphl;
++ int err;
+
-+ key = container_of(kref, struct au_dykey, dk_kref);
-+ sphl = dynop + key->dk_op.dy_type;
-+ au_sphl_del_rcu(&key->dk_hnode, sphl);
-+ call_rcu(&key->dk_rcu, dy_free_rcu);
-+}
++ err = 0;
++ if (au_br_writable(br->br_perm))
++ err = au_dr_hino(sb, /*bindex*/-1, br, /*path*/NULL);
++ if (!err)
++ au_dr_hino_free(&br->br_dirren);
+
-+void au_dy_put(struct au_dykey *key)
-+{
-+ kref_put(&key->dk_kref, dy_free);
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+#define DyDbgSize(cnt, op) AuDebugOn(cnt != sizeof(op)/sizeof(void *))
-+
-+#ifdef CONFIG_AUFS_DEBUG
-+#define DyDbgDeclare(cnt) unsigned int cnt = 0
-+#define DyDbgInc(cnt) do { cnt++; } while (0)
-+#else
-+#define DyDbgDeclare(cnt) do {} while (0)
-+#define DyDbgInc(cnt) do {} while (0)
-+#endif
-+
-+#define DySet(func, dst, src, h_op, h_sb) do { \
-+ DyDbgInc(cnt); \
-+ if (h_op->func) { \
-+ if (src.func) \
-+ dst.func = src.func; \
-+ else \
-+ AuDbg("%s %s\n", au_sbtype(h_sb), #func); \
-+ } \
-+} while (0)
++static int au_brid_str(struct au_dr_brid *brid, struct inode *h_inode,
++ char *buf, size_t sz)
++{
++ int err;
++ unsigned int major, minor;
++ char *p;
+
-+#define DySetForce(func, dst, src) do { \
-+ AuDebugOn(!src.func); \
-+ DyDbgInc(cnt); \
-+ dst.func = src.func; \
-+} while (0)
++ p = buf;
++ err = snprintf(p, sz, "%d_", brid->type);
++ AuDebugOn(err > sz);
++ p += err;
++ sz -= err;
++ switch (brid->type) {
++ case AuBrid_Unset:
++ return -EINVAL;
++ case AuBrid_UUID:
++ err = snprintf(p, sz, "%pU", brid->uuid.b);
++ break;
++ case AuBrid_FSID:
++ err = snprintf(p, sz, "%08x-%08x",
++ brid->fsid.val[0], brid->fsid.val[1]);
++ break;
++ case AuBrid_DEV:
++ major = MAJOR(brid->dev);
++ minor = MINOR(brid->dev);
++ if (major <= 0xff && minor <= 0xff)
++ err = snprintf(p, sz, "%02x%02x", major, minor);
++ else
++ err = snprintf(p, sz, "%03x:%05x", major, minor);
++ break;
++ }
++ AuDebugOn(err > sz);
++ p += err;
++ sz -= err;
++ err = snprintf(p, sz, "_%llu", (unsigned long long)h_inode->i_ino);
++ AuDebugOn(err > sz);
++ p += err;
++ sz -= err;
+
-+#define DySetAop(func) \
-+ DySet(func, dyaop->da_op, aufs_aop, h_aop, h_sb)
-+#define DySetAopForce(func) \
-+ DySetForce(func, dyaop->da_op, aufs_aop)
++ return p - buf;
++}
+
-+static void dy_aop(struct au_dykey *key, const void *h_op,
-+ struct super_block *h_sb __maybe_unused)
++static int au_drinfo_name(struct au_branch *br, char *name, int len)
+{
-+ struct au_dyaop *dyaop = (void *)key;
-+ const struct address_space_operations *h_aop = h_op;
-+ DyDbgDeclare(cnt);
-+
-+ AuDbg("%s\n", au_sbtype(h_sb));
++ int rlen;
++ struct dentry *br_dentry;
++ struct inode *br_inode;
+
-+ DySetAop(writepage);
-+ DySetAopForce(readpage); /* force */
-+ DySetAop(writepages);
-+ DySetAop(set_page_dirty);
-+ DySetAop(readpages);
-+ DySetAop(write_begin);
-+ DySetAop(write_end);
-+ DySetAop(bmap);
-+ DySetAop(invalidatepage);
-+ DySetAop(releasepage);
-+ DySetAop(freepage);
-+ /* this one will be changed according to an aufs mount option */
-+ DySetAop(direct_IO);
-+ DySetAop(migratepage);
-+ DySetAop(launder_page);
-+ DySetAop(is_partially_uptodate);
-+ DySetAop(is_dirty_writeback);
-+ DySetAop(error_remove_page);
-+ DySetAop(swap_activate);
-+ DySetAop(swap_deactivate);
++ br_dentry = au_br_dentry(br);
++ br_inode = d_inode(br_dentry);
++ rlen = au_brid_str(&br->br_dirren.dr_brid, br_inode, name, len);
++ AuDebugOn(rlen >= AUFS_DIRREN_ENV_VAL_SZ);
++ AuDebugOn(rlen > len);
+
-+ DyDbgSize(cnt, *h_aop);
++ return rlen;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static void dy_bug(struct kref *kref)
++/*
++ * from the given @h_dentry, construct drinfo at @*fdata.
++ * when the size of @*fdata is not enough, reallocate and return new @fdata and
++ * @allocated.
++ */
++static int au_drinfo_construct(struct au_drinfo_fdata **fdata,
++ struct dentry *h_dentry,
++ unsigned char *allocated)
+{
-+ BUG();
++ int err, v;
++ struct au_drinfo_fdata *f, *p;
++ struct au_drinfo *drinfo;
++ struct inode *h_inode;
++ struct qstr *qname;
++
++ err = 0;
++ f = *fdata;
++ h_inode = d_inode(h_dentry);
++ qname = &h_dentry->d_name;
++ drinfo = &f->drinfo;
++ drinfo->ino = (__force uint64_t)cpu_to_be64(h_inode->i_ino);
++ drinfo->oldnamelen = qname->len;
++ if (*allocated < sizeof(*f) + qname->len) {
++ v = roundup_pow_of_two(*allocated + qname->len);
++ p = au_krealloc(f, v, GFP_NOFS, /*may_shrink*/0);
++ if (unlikely(!p)) {
++ err = -ENOMEM;
++ AuTraceErr(err);
++ goto out;
++ }
++ f = p;
++ *fdata = f;
++ *allocated = v;
++ drinfo = &f->drinfo;
++ }
++ memcpy(drinfo->oldname, qname->name, qname->len);
++ AuDbg("i%llu, %.*s\n",
++ be64_to_cpu((__force __be64)drinfo->ino), drinfo->oldnamelen,
++ drinfo->oldname);
++
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static struct au_dykey *dy_get(struct au_dynop *op, struct au_branch *br)
++/* callers have to free the return value */
++static struct au_drinfo *au_drinfo_read_k(struct file *file, ino_t h_ino)
+{
-+ struct au_dykey *key, *old;
-+ struct au_sphlhead *sphl;
-+ struct op {
-+ unsigned int sz;
-+ void (*set)(struct au_dykey *key, const void *h_op,
-+ struct super_block *h_sb __maybe_unused);
-+ };
-+ static const struct op a[] = {
-+ [AuDy_AOP] = {
-+ .sz = sizeof(struct au_dyaop),
-+ .set = dy_aop
-+ }
-+ };
-+ const struct op *p;
++ struct au_drinfo *ret, *drinfo;
++ struct au_drinfo_fdata fdata;
++ int len;
++ loff_t pos;
++ ssize_t ssz;
+
-+ sphl = dynop + op->dy_type;
-+ key = dy_gfind_get(sphl, op->dy_hop);
-+ if (key)
-+ goto out_add; /* success */
++ ret = ERR_PTR(-EIO);
++ pos = 0;
++ ssz = vfsub_read_k(file, &fdata, sizeof(fdata), &pos);
++ if (unlikely(ssz != sizeof(fdata))) {
++ AuIOErr("ssz %zd, %u, %pD2\n",
++ ssz, (unsigned int)sizeof(fdata), file);
++ goto out;
++ }
+
-+ p = a + op->dy_type;
-+ key = kzalloc(p->sz, GFP_NOFS);
-+ if (unlikely(!key)) {
-+ key = ERR_PTR(-ENOMEM);
++ fdata.magic = ntohl((__force __be32)fdata.magic);
++ switch (fdata.magic) {
++ case AUFS_DRINFO_MAGIC_V1:
++ break;
++ default:
++ AuIOErr("magic-num 0x%x, 0x%x, %pD2\n",
++ fdata.magic, AUFS_DRINFO_MAGIC_V1, file);
+ goto out;
+ }
+
-+ key->dk_op.dy_hop = op->dy_hop;
-+ kref_init(&key->dk_kref);
-+ p->set(key, op->dy_hop, au_br_sb(br));
-+ old = dy_gadd(sphl, key);
-+ if (old) {
-+ au_delayed_kfree(key);
-+ key = old;
++ drinfo = &fdata.drinfo;
++ len = drinfo->oldnamelen;
++ if (!len) {
++ AuIOErr("broken drinfo %pD2\n", file);
++ goto out;
+ }
+
-+out_add:
-+ old = dy_bradd(br, key);
-+ if (old)
-+ /* its ref-count should never be zero here */
-+ kref_put(&key->dk_kref, dy_bug);
++ ret = NULL;
++ drinfo->ino = be64_to_cpu((__force __be64)drinfo->ino);
++ if (unlikely(h_ino && drinfo->ino != h_ino)) {
++ AuDbg("ignored i%llu, i%llu, %pD2\n",
++ (unsigned long long)drinfo->ino,
++ (unsigned long long)h_ino, file);
++ goto out; /* success */
++ }
++
++ ret = kmalloc(sizeof(*ret) + len, GFP_NOFS);
++ if (unlikely(!ret)) {
++ ret = ERR_PTR(-ENOMEM);
++ AuTraceErrPtr(ret);
++ goto out;
++ }
++
++ *ret = *drinfo;
++ ssz = vfsub_read_k(file, (void *)ret->oldname, len, &pos);
++ if (unlikely(ssz != len)) {
++ kfree(ret);
++ ret = ERR_PTR(-EIO);
++ AuIOErr("ssz %zd, %u, %pD2\n", ssz, len, file);
++ goto out;
++ }
++
++ AuDbg("oldname %.*s\n", ret->oldnamelen, ret->oldname);
++
+out:
-+ return key;
++ return ret;
+}
+
+/* ---------------------------------------------------------------------- */
-+/*
-+ * Aufs prohibits O_DIRECT by defaut even if the branch supports it.
-+ * This behaviour is necessary to return an error from open(O_DIRECT) instead
-+ * of the succeeding I/O. The dio mount option enables O_DIRECT and makes
-+ * open(O_DIRECT) always succeed, but the succeeding I/O may return an error.
-+ * See the aufs manual in detail.
-+ */
-+static void dy_adx(struct au_dyaop *dyaop, int do_dx)
-+{
-+ if (!do_dx)
-+ dyaop->da_op.direct_IO = NULL;
-+ else
-+ dyaop->da_op.direct_IO = aufs_aop.direct_IO;
-+}
+
-+static struct au_dyaop *dy_aget(struct au_branch *br,
-+ const struct address_space_operations *h_aop,
-+ int do_dx)
-+{
-+ struct au_dyaop *dyaop;
-+ struct au_dynop op;
++/* in order to be revertible */
++struct au_drinfo_rev_elm {
++ int created;
++ struct dentry *info_dentry;
++ struct au_drinfo *info_last;
++};
+
-+ op.dy_type = AuDy_AOP;
-+ op.dy_haop = h_aop;
-+ dyaop = (void *)dy_get(&op, br);
-+ if (IS_ERR(dyaop))
-+ goto out;
-+ dy_adx(dyaop, do_dx);
++struct au_drinfo_rev {
++ unsigned char already;
++ aufs_bindex_t nelm;
++ struct au_drinfo_rev_elm elm[0];
++};
+
-+out:
-+ return dyaop;
-+}
++/* todo: isn't it too large? */
++struct au_drinfo_store {
++ struct path h_ppath;
++ struct dentry *h_dentry;
++ struct au_drinfo_fdata *fdata;
++ char *infoname; /* inside of whname, just after PFX */
++ char whname[sizeof(AUFS_WH_DR_INFO_PFX) + AUFS_DIRREN_ENV_VAL_SZ];
++ aufs_bindex_t btgt, btail;
++ unsigned char no_sio,
++ allocated, /* current size of *fdata */
++ infonamelen, /* room size for p */
++ whnamelen, /* length of the genarated name */
++ renameback; /* renamed back */
++};
+
-+int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode)
++/* on rename(2) error, the caller should revert it using @elm */
++static int au_drinfo_do_store(struct au_drinfo_store *w,
++ struct au_drinfo_rev_elm *elm)
+{
-+ int err, do_dx;
-+ struct super_block *sb;
-+ struct au_branch *br;
-+ struct au_dyaop *dyaop;
++ int err, len;
++ ssize_t ssz;
++ loff_t pos;
++ struct path infopath = {
++ .mnt = w->h_ppath.mnt
++ };
++ struct inode *h_dir, *h_inode, *delegated;
++ struct file *infofile;
++ struct qstr *qname;
+
-+ AuDebugOn(!S_ISREG(h_inode->i_mode));
-+ IiMustWriteLock(inode);
++ AuDebugOn(elm
++ && memcmp(elm, page_address(ZERO_PAGE(0)), sizeof(*elm)));
+
-+ sb = inode->i_sb;
-+ br = au_sbr(sb, bindex);
-+ do_dx = !!au_opt_test(au_mntflags(sb), DIO);
-+ dyaop = dy_aget(br, h_inode->i_mapping->a_ops, do_dx);
-+ err = PTR_ERR(dyaop);
-+ if (IS_ERR(dyaop))
-+ /* unnecessary to call dy_fput() */
++ infopath.dentry = vfsub_lookup_one_len(w->whname, w->h_ppath.dentry,
++ w->whnamelen);
++ AuTraceErrPtr(infopath.dentry);
++ if (IS_ERR(infopath.dentry)) {
++ err = PTR_ERR(infopath.dentry);
+ goto out;
++ }
+
+ err = 0;
-+ inode->i_mapping->a_ops = &dyaop->da_op;
++ h_dir = d_inode(w->h_ppath.dentry);
++ if (elm && d_is_negative(infopath.dentry)) {
++ err = vfsub_create(h_dir, &infopath, 0600, /*want_excl*/true);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out_dput;
++ elm->created = 1;
++ elm->info_dentry = dget(infopath.dentry);
++ }
++
++ infofile = vfsub_dentry_open(&infopath, O_RDWR);
++ AuTraceErrPtr(infofile);
++ if (IS_ERR(infofile)) {
++ err = PTR_ERR(infofile);
++ goto out_dput;
++ }
++
++ h_inode = d_inode(infopath.dentry);
++ if (elm && i_size_read(h_inode)) {
++ h_inode = d_inode(w->h_dentry);
++ elm->info_last = au_drinfo_read_k(infofile, h_inode->i_ino);
++ AuTraceErrPtr(elm->info_last);
++ if (IS_ERR(elm->info_last)) {
++ err = PTR_ERR(elm->info_last);
++ elm->info_last = NULL;
++ AuDebugOn(elm->info_dentry);
++ goto out_fput;
++ }
++ }
++
++ if (elm && w->renameback) {
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, &infopath, &delegated, /*force*/0);
++ AuTraceErr(err);
++ if (unlikely(err == -EWOULDBLOCK))
++ iput(delegated);
++ goto out_fput;
++ }
++
++ pos = 0;
++ qname = &w->h_dentry->d_name;
++ len = sizeof(*w->fdata) + qname->len;
++ if (!elm)
++ len = sizeof(*w->fdata) + w->fdata->drinfo.oldnamelen;
++ ssz = vfsub_write_k(infofile, w->fdata, len, &pos);
++ if (ssz == len) {
++ AuDbg("hi%llu, %.*s\n", w->fdata->drinfo.ino,
++ w->fdata->drinfo.oldnamelen, w->fdata->drinfo.oldname);
++ goto out_fput; /* success */
++ } else {
++ err = -EIO;
++ if (ssz < 0)
++ err = ssz;
++ /* the caller should revert it using @elm */
++ }
+
++out_fput:
++ fput(infofile);
++out_dput:
++ dput(infopath.dentry);
+out:
++ AuTraceErr(err);
+ return err;
+}
+
-+/*
-+ * Is it safe to replace a_ops during the inode/file is in operation?
-+ * Yes, I hope so.
-+ */
-+int au_dy_irefresh(struct inode *inode)
++struct au_call_drinfo_do_store_args {
++ int *errp;
++ struct au_drinfo_store *w;
++ struct au_drinfo_rev_elm *elm;
++};
++
++static void au_call_drinfo_do_store(void *args)
+{
-+ int err;
-+ aufs_bindex_t btop;
-+ struct inode *h_inode;
++ struct au_call_drinfo_do_store_args *a = args;
+
-+ err = 0;
-+ if (S_ISREG(inode->i_mode)) {
-+ btop = au_ibtop(inode);
-+ h_inode = au_h_iptr(inode, btop);
-+ err = au_dy_iaop(inode, btop, h_inode);
-+ }
-+ return err;
++ *a->errp = au_drinfo_do_store(a->w, a->elm);
+}
+
-+void au_dy_arefresh(int do_dx)
++static int au_drinfo_store_sio(struct au_drinfo_store *w,
++ struct au_drinfo_rev_elm *elm)
+{
-+ struct au_sphlhead *sphl;
-+ struct hlist_head *head;
-+ struct au_dykey *key;
++ int err, wkq_err;
+
-+ sphl = dynop + AuDy_AOP;
-+ head = &sphl->head;
-+ spin_lock(&sphl->spin);
-+ hlist_for_each_entry(key, head, dk_hnode)
-+ dy_adx((void *)key, do_dx);
-+ spin_unlock(&sphl->spin);
-+}
++ if (w->no_sio)
++ err = au_drinfo_do_store(w, elm);
++ else {
++ struct au_call_drinfo_do_store_args a = {
++ .errp = &err,
++ .w = w,
++ .elm = elm
++ };
++ wkq_err = au_wkq_wait(au_call_drinfo_do_store, &a);
++ if (unlikely(wkq_err))
++ err = wkq_err;
++ }
++ AuTraceErr(err);
+
-+/* ---------------------------------------------------------------------- */
++ return err;
++}
+
-+void __init au_dy_init(void)
++static int au_drinfo_store_work_init(struct au_drinfo_store *w,
++ aufs_bindex_t btgt)
+{
-+ int i;
++ int err;
+
-+ /* make sure that 'struct au_dykey *' can be any type */
-+ BUILD_BUG_ON(offsetof(struct au_dyaop, da_key));
++ memset(w, 0, sizeof(*w));
++ w->allocated = roundup_pow_of_two(sizeof(*w->fdata) + 40);
++ strcpy(w->whname, AUFS_WH_DR_INFO_PFX);
++ w->infoname = w->whname + sizeof(AUFS_WH_DR_INFO_PFX) - 1;
++ w->infonamelen = sizeof(w->whname) - sizeof(AUFS_WH_DR_INFO_PFX);
++ w->btgt = btgt;
++ w->no_sio = !!uid_eq(current_fsuid(), GLOBAL_ROOT_UID);
+
-+ for (i = 0; i < AuDyLast; i++)
-+ au_sphl_init(dynop + i);
++ err = -ENOMEM;
++ w->fdata = kcalloc(1, w->allocated, GFP_NOFS);
++ if (unlikely(!w->fdata)) {
++ AuTraceErr(err);
++ goto out;
++ }
++ w->fdata->magic = (__force uint32_t)htonl(AUFS_DRINFO_MAGIC_V1);
++ err = 0;
++
++out:
++ return err;
+}
+
-+void au_dy_fin(void)
++static void au_drinfo_store_work_fin(struct au_drinfo_store *w)
+{
-+ int i;
-+
-+ for (i = 0; i < AuDyLast; i++)
-+ WARN_ON(!hlist_empty(&dynop[i].head));
++ kfree(w->fdata);
+}
-diff -urN /usr/share/empty/fs/aufs/dynop.h linux/fs/aufs/dynop.h
---- /usr/share/empty/fs/aufs/dynop.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/dynop.h 2016-08-17 18:01:06.118555489 +0200
-@@ -0,0 +1,74 @@
-+/*
-+ * Copyright (C) 2010-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
+
-+/*
-+ * dynamically customizable operations (for regular files only)
-+ */
++static void au_drinfo_store_rev(struct au_drinfo_rev *rev,
++ struct au_drinfo_store *w)
++{
++ struct au_drinfo_rev_elm *elm;
++ struct inode *h_dir, *delegated;
++ int err, nelm;
++ struct path infopath = {
++ .mnt = w->h_ppath.mnt
++ };
+
-+#ifndef __AUFS_DYNOP_H__
-+#define __AUFS_DYNOP_H__
++ h_dir = d_inode(w->h_ppath.dentry);
++ IMustLock(h_dir);
+
-+#ifdef __KERNEL__
++ err = 0;
++ elm = rev->elm;
++ for (nelm = rev->nelm; nelm > 0; nelm--, elm++) {
++ AuDebugOn(elm->created && elm->info_last);
++ if (elm->created) {
++ AuDbg("here\n");
++ delegated = NULL;
++ infopath.dentry = elm->info_dentry;
++ err = vfsub_unlink(h_dir, &infopath, &delegated,
++ !w->no_sio);
++ AuTraceErr(err);
++ if (unlikely(err == -EWOULDBLOCK))
++ iput(delegated);
++ dput(elm->info_dentry);
++ } else if (elm->info_last) {
++ AuDbg("here\n");
++ w->fdata->drinfo = *elm->info_last;
++ memcpy(w->fdata->drinfo.oldname,
++ elm->info_last->oldname,
++ elm->info_last->oldnamelen);
++ err = au_drinfo_store_sio(w, /*elm*/NULL);
++ kfree(elm->info_last);
++ }
++ if (unlikely(err))
++ AuIOErr("%d, %s\n", err, w->whname);
++ /* go on even if err */
++ }
++}
+
-+#include <linux/fs.h>
-+#include <linux/kref.h>
++/* caller has to call au_dr_rename_fin() later */
++static int au_drinfo_store(struct dentry *dentry, aufs_bindex_t btgt,
++ struct qstr *dst_name, void *_rev)
++{
++ int err, sz, nelm;
++ aufs_bindex_t bindex, btail;
++ struct au_drinfo_store work;
++ struct au_drinfo_rev *rev, **p;
++ struct au_drinfo_rev_elm *elm;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_hinode *hdir;
+
-+enum {AuDy_AOP, AuDyLast};
++ err = au_drinfo_store_work_init(&work, btgt);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out;
+
-+struct au_dynop {
-+ int dy_type;
-+ union {
-+ const void *dy_hop;
-+ const struct address_space_operations *dy_haop;
-+ };
-+};
++ err = -ENOMEM;
++ btail = au_dbtaildir(dentry);
++ nelm = btail - btgt;
++ sz = sizeof(*rev) + sizeof(*elm) * nelm;
++ rev = kcalloc(1, sz, GFP_NOFS);
++ if (unlikely(!rev)) {
++ AuTraceErr(err);
++ goto out_args;
++ }
++ rev->nelm = nelm;
++ elm = rev->elm;
++ p = _rev;
++ *p = rev;
+
-+struct au_dykey {
-+ union {
-+ struct hlist_node dk_hnode;
-+ struct rcu_head dk_rcu;
-+ };
-+ struct au_dynop dk_op;
++ err = 0;
++ sb = dentry->d_sb;
++ work.h_ppath.dentry = au_h_dptr(dentry, btgt);
++ work.h_ppath.mnt = au_sbr_mnt(sb, btgt);
++ hdir = au_hi(d_inode(dentry), btgt);
++ au_hn_inode_lock_nested(hdir, AuLsc_I_CHILD);
++ for (bindex = btgt + 1; bindex <= btail; bindex++, elm++) {
++ work.h_dentry = au_h_dptr(dentry, bindex);
++ if (!work.h_dentry)
++ continue;
+
-+ /*
-+ * during I am in the branch local array, kref is gotten. when the
-+ * branch is removed, kref is put.
-+ */
-+ struct kref dk_kref;
-+};
++ err = au_drinfo_construct(&work.fdata, work.h_dentry,
++ &work.allocated);
++ AuTraceErr(err);
++ if (unlikely(err))
++ break;
+
-+/* stop unioning since their sizes are very different from each other */
-+struct au_dyaop {
-+ struct au_dykey da_key;
-+ struct address_space_operations da_op; /* not const */
-+};
++ work.renameback = au_qstreq(&work.h_dentry->d_name, dst_name);
++ br = au_sbr(sb, bindex);
++ work.whnamelen = sizeof(AUFS_WH_DR_INFO_PFX) - 1;
++ work.whnamelen += au_drinfo_name(br, work.infoname,
++ work.infonamelen);
++ AuDbg("whname %.*s, i%llu, %.*s\n",
++ work.whnamelen, work.whname,
++ be64_to_cpu((__force __be64)work.fdata->drinfo.ino),
++ work.fdata->drinfo.oldnamelen,
++ work.fdata->drinfo.oldname);
++
++ err = au_drinfo_store_sio(&work, elm);
++ AuTraceErr(err);
++ if (unlikely(err))
++ break;
++ }
++ if (unlikely(err)) {
++ /* revert all drinfo */
++ au_drinfo_store_rev(rev, &work);
++ kfree(rev);
++ *p = NULL;
++ }
++ au_hn_inode_unlock(hdir);
++
++out_args:
++ au_drinfo_store_work_fin(&work);
++out:
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+/* dynop.c */
-+struct au_branch;
-+void au_dy_put(struct au_dykey *key);
-+int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode);
-+int au_dy_irefresh(struct inode *inode);
-+void au_dy_arefresh(int do_dio);
++int au_dr_rename(struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev)
++{
++ int err, already;
++ ino_t ino;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_dr_br *dr;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct au_dr_hino *ent;
++ struct au_drinfo_rev *rev, **p;
+
-+void __init au_dy_init(void);
-+void au_dy_fin(void);
++ AuDbg("bindex %d\n", bindex);
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_DYNOP_H__ */
-diff -urN /usr/share/empty/fs/aufs/export.c linux/fs/aufs/export.c
---- /usr/share/empty/fs/aufs/export.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/export.c 2016-08-17 18:01:06.128555749 +0200
-@@ -0,0 +1,837 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
++ err = -ENOMEM;
++ ent = kmalloc(sizeof(*ent), GFP_NOFS);
++ if (unlikely(!ent))
++ goto out;
+
-+/*
-+ * export via nfs
-+ */
++ sb = src->d_sb;
++ br = au_sbr(sb, bindex);
++ dr = &br->br_dirren;
++ h_dentry = au_h_dptr(src, bindex);
++ h_inode = d_inode(h_dentry);
++ ino = h_inode->i_ino;
++ ent->dr_h_ino = ino;
++ already = au_dr_hino_test_add(dr, ino, ent);
++ AuDbg("b%d, hi%llu, already %d\n",
++ bindex, (unsigned long long)ino, already);
++
++ err = au_drinfo_store(src, bindex, dst_name, _rev);
++ AuTraceErr(err);
++ if (!err) {
++ p = _rev;
++ rev = *p;
++ rev->already = already;
++ goto out; /* success */
++ }
+
-+#include <linux/exportfs.h>
-+#include <linux/fs_struct.h>
-+#include <linux/namei.h>
-+#include <linux/nsproxy.h>
-+#include <linux/random.h>
-+#include <linux/writeback.h>
-+#include "../fs/mount.h"
-+#include "aufs.h"
++ /* revert */
++ if (!already)
++ au_dr_hino_del(dr, ent);
++ kfree(ent);
+
-+union conv {
-+#ifdef CONFIG_AUFS_INO_T_64
-+ __u32 a[2];
-+#else
-+ __u32 a[1];
-+#endif
-+ ino_t ino;
-+};
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+static ino_t decode_ino(__u32 *a)
++void au_dr_rename_fin(struct dentry *src, aufs_bindex_t btgt, void *_rev)
+{
-+ union conv u;
++ struct au_drinfo_rev *rev;
++ struct au_drinfo_rev_elm *elm;
++ int nelm;
+
-+ BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a));
-+ u.a[0] = a[0];
-+#ifdef CONFIG_AUFS_INO_T_64
-+ u.a[1] = a[1];
-+#endif
-+ return u.ino;
++ rev = _rev;
++ elm = rev->elm;
++ for (nelm = rev->nelm; nelm > 0; nelm--, elm++) {
++ dput(elm->info_dentry);
++ kfree(elm->info_last);
++ }
++ kfree(rev);
+}
+
-+static void encode_ino(__u32 *a, ino_t ino)
++void au_dr_rename_rev(struct dentry *src, aufs_bindex_t btgt, void *_rev)
+{
-+ union conv u;
++ int err;
++ struct au_drinfo_store work;
++ struct au_drinfo_rev *rev = _rev;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct inode *h_inode;
++ struct au_dr_br *dr;
++ struct au_dr_hino *ent;
+
-+ u.ino = ino;
-+ a[0] = u.a[0];
-+#ifdef CONFIG_AUFS_INO_T_64
-+ a[1] = u.a[1];
-+#endif
++ err = au_drinfo_store_work_init(&work, btgt);
++ if (unlikely(err))
++ goto out;
++
++ sb = src->d_sb;
++ br = au_sbr(sb, btgt);
++ work.h_ppath.dentry = au_h_dptr(src, btgt);
++ work.h_ppath.mnt = au_br_mnt(br);
++ au_drinfo_store_rev(rev, &work);
++ au_drinfo_store_work_fin(&work);
++ if (rev->already)
++ goto out;
++
++ dr = &br->br_dirren;
++ h_inode = d_inode(work.h_ppath.dentry);
++ ent = au_dr_hino_find(dr, h_inode->i_ino);
++ BUG_ON(!ent);
++ au_dr_hino_del(dr, ent);
++ kfree(ent);
++
++out:
++ kfree(rev);
++ if (unlikely(err))
++ pr_err("failed to remove dirren info\n");
+}
+
-+/* NFS file handle */
-+enum {
-+ Fh_br_id,
-+ Fh_sigen,
-+#ifdef CONFIG_AUFS_INO_T_64
-+ /* support 64bit inode number */
-+ Fh_ino1,
-+ Fh_ino2,
-+ Fh_dir_ino1,
-+ Fh_dir_ino2,
-+#else
-+ Fh_ino1,
-+ Fh_dir_ino1,
-+#endif
-+ Fh_igen,
-+ Fh_h_type,
-+ Fh_tail,
++/* ---------------------------------------------------------------------- */
+
-+ Fh_ino = Fh_ino1,
-+ Fh_dir_ino = Fh_dir_ino1
++static struct au_drinfo *au_drinfo_do_load(struct path *h_ppath,
++ char *whname, int whnamelen,
++ struct dentry **info_dentry)
++{
++ struct au_drinfo *drinfo;
++ struct file *f;
++ struct inode *h_dir;
++ struct path infopath;
++ int unlocked;
++
++ AuDbg("%pd/%.*s\n", h_ppath->dentry, whnamelen, whname);
++
++ *info_dentry = NULL;
++ drinfo = NULL;
++ unlocked = 0;
++ h_dir = d_inode(h_ppath->dentry);
++ vfsub_inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
++ infopath.dentry = vfsub_lookup_one_len(whname, h_ppath->dentry,
++ whnamelen);
++ if (IS_ERR(infopath.dentry)) {
++ drinfo = (void *)infopath.dentry;
++ goto out;
++ }
++
++ if (d_is_negative(infopath.dentry))
++ goto out_dput; /* success */
++
++ infopath.mnt = h_ppath->mnt;
++ f = vfsub_dentry_open(&infopath, O_RDONLY);
++ inode_unlock_shared(h_dir);
++ unlocked = 1;
++ if (IS_ERR(f)) {
++ drinfo = (void *)f;
++ goto out_dput;
++ }
++
++ drinfo = au_drinfo_read_k(f, /*h_ino*/0);
++ if (IS_ERR_OR_NULL(drinfo))
++ goto out_fput;
++
++ AuDbg("oldname %.*s\n", drinfo->oldnamelen, drinfo->oldname);
++ *info_dentry = dget(infopath.dentry); /* keep it alive */
++
++out_fput:
++ fput(f);
++out_dput:
++ dput(infopath.dentry);
++out:
++ if (!unlocked)
++ inode_unlock_shared(h_dir);
++ AuTraceErrPtr(drinfo);
++ return drinfo;
++}
++
++struct au_drinfo_do_load_args {
++ struct au_drinfo **drinfop;
++ struct path *h_ppath;
++ char *whname;
++ int whnamelen;
++ struct dentry **info_dentry;
+};
+
-+static int au_test_anon(struct dentry *dentry)
++static void au_call_drinfo_do_load(void *args)
+{
-+ /* note: read d_flags without d_lock */
-+ return !!(dentry->d_flags & DCACHE_DISCONNECTED);
++ struct au_drinfo_do_load_args *a = args;
++
++ *a->drinfop = au_drinfo_do_load(a->h_ppath, a->whname, a->whnamelen,
++ a->info_dentry);
+}
+
-+int au_test_nfsd(void)
++struct au_drinfo_load {
++ struct path h_ppath;
++ struct qstr *qname;
++ unsigned char no_sio;
++
++ aufs_bindex_t ninfo;
++ struct au_drinfo **drinfo;
++};
++
++static int au_drinfo_load(struct au_drinfo_load *w, aufs_bindex_t bindex,
++ struct au_branch *br)
+{
-+ int ret;
-+ struct task_struct *tsk = current;
-+ char comm[sizeof(tsk->comm)];
++ int err, wkq_err, whnamelen, e;
++ char whname[sizeof(AUFS_WH_DR_INFO_PFX) + AUFS_DIRREN_ENV_VAL_SZ]
++ = AUFS_WH_DR_INFO_PFX;
++ struct au_drinfo *drinfo;
++ struct qstr oldname;
++ struct inode *h_dir, *delegated;
++ struct dentry *info_dentry;
++ struct path infopath;
++
++ whnamelen = sizeof(AUFS_WH_DR_INFO_PFX) - 1;
++ whnamelen += au_drinfo_name(br, whname + whnamelen,
++ sizeof(whname) - whnamelen);
++ if (w->no_sio)
++ drinfo = au_drinfo_do_load(&w->h_ppath, whname, whnamelen,
++ &info_dentry);
++ else {
++ struct au_drinfo_do_load_args args = {
++ .drinfop = &drinfo,
++ .h_ppath = &w->h_ppath,
++ .whname = whname,
++ .whnamelen = whnamelen,
++ .info_dentry = &info_dentry
++ };
++ wkq_err = au_wkq_wait(au_call_drinfo_do_load, &args);
++ if (unlikely(wkq_err))
++ drinfo = ERR_PTR(wkq_err);
++ }
++ err = PTR_ERR(drinfo);
++ if (IS_ERR_OR_NULL(drinfo))
++ goto out;
+
-+ ret = 0;
-+ if (tsk->flags & PF_KTHREAD) {
-+ get_task_comm(comm, tsk);
-+ ret = !strcmp(comm, "nfsd");
++ err = 0;
++ oldname.len = drinfo->oldnamelen;
++ oldname.name = drinfo->oldname;
++ if (au_qstreq(w->qname, &oldname)) {
++ /* the name is renamed back */
++ kfree(drinfo);
++ drinfo = NULL;
++
++ infopath.dentry = info_dentry;
++ infopath.mnt = w->h_ppath.mnt;
++ h_dir = d_inode(w->h_ppath.dentry);
++ delegated = NULL;
++ inode_lock_nested(h_dir, AuLsc_I_PARENT);
++ e = vfsub_unlink(h_dir, &infopath, &delegated, !w->no_sio);
++ inode_unlock(h_dir);
++ if (unlikely(e))
++ AuIOErr("ignored %d, %pd2\n", e, &infopath.dentry);
++ if (unlikely(e == -EWOULDBLOCK))
++ iput(delegated);
+ }
++ kfree(w->drinfo[bindex]);
++ w->drinfo[bindex] = drinfo;
++ dput(info_dentry);
+
-+ return ret;
++out:
++ AuTraceErr(err);
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
-+/* inode generation external table */
+
-+void au_xigen_inc(struct inode *inode)
++static void au_dr_lkup_free(struct au_drinfo **drinfo, int n)
+{
-+ loff_t pos;
-+ ssize_t sz;
-+ __u32 igen;
-+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
-+
-+ sb = inode->i_sb;
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), XINO));
-+
-+ sbinfo = au_sbi(sb);
-+ pos = inode->i_ino;
-+ pos *= sizeof(igen);
-+ igen = inode->i_generation + 1;
-+ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xigen, &igen,
-+ sizeof(igen), &pos);
-+ if (sz == sizeof(igen))
-+ return; /* success */
++ struct au_drinfo **p = drinfo;
+
-+ if (unlikely(sz >= 0))
-+ AuIOErr("xigen error (%zd)\n", sz);
++ while (n-- > 0)
++ kfree(*drinfo++);
++ kfree(p);
+}
+
-+int au_xigen_new(struct inode *inode)
++int au_dr_lkup(struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t btgt)
+{
-+ int err;
-+ loff_t pos;
-+ ssize_t sz;
++ int err, ninfo;
++ struct au_drinfo_load w;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++ struct inode *h_dir;
++ struct au_dr_hino *ent;
+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file;
+
-+ err = 0;
-+ /* todo: dirty, at mount time */
-+ if (inode->i_ino == AUFS_ROOT_INO)
-+ goto out;
-+ sb = inode->i_sb;
-+ SiMustAnyLock(sb);
-+ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
-+ goto out;
++ AuDbg("%.*s, name %.*s, whname %.*s, b%d\n",
++ AuLNPair(&dentry->d_name), AuLNPair(&lkup->dirren.dr_name),
++ AuLNPair(&lkup->whname), btgt);
+
-+ err = -EFBIG;
-+ pos = inode->i_ino;
-+ if (unlikely(au_loff_max / sizeof(inode->i_generation) - 1 < pos)) {
-+ AuIOErr1("too large i%lld\n", pos);
-+ goto out;
++ sb = dentry->d_sb;
++ bbot = au_sbbot(sb);
++ w.ninfo = bbot + 1;
++ if (!lkup->dirren.drinfo) {
++ lkup->dirren.drinfo = kcalloc(w.ninfo,
++ sizeof(*lkup->dirren.drinfo),
++ GFP_NOFS);
++ if (unlikely(!lkup->dirren.drinfo)) {
++ err = -ENOMEM;
++ goto out;
++ }
++ lkup->dirren.ninfo = w.ninfo;
+ }
-+ pos *= sizeof(inode->i_generation);
-+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ file = sbinfo->si_xigen;
-+ BUG_ON(!file);
-+
-+ if (vfsub_f_size_read(file)
-+ < pos + sizeof(inode->i_generation)) {
-+ inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next);
-+ sz = xino_fwrite(sbinfo->si_xwrite, file, &inode->i_generation,
-+ sizeof(inode->i_generation), &pos);
-+ } else
-+ sz = xino_fread(sbinfo->si_xread, file, &inode->i_generation,
-+ sizeof(inode->i_generation), &pos);
-+ if (sz == sizeof(inode->i_generation))
-+ goto out; /* success */
++ w.drinfo = lkup->dirren.drinfo;
++ w.no_sio = !!uid_eq(current_fsuid(), GLOBAL_ROOT_UID);
++ w.h_ppath.dentry = au_h_dptr(dentry, btgt);
++ AuDebugOn(!w.h_ppath.dentry);
++ w.h_ppath.mnt = au_sbr_mnt(sb, btgt);
++ w.qname = &dentry->d_name;
+
-+ err = sz;
-+ if (unlikely(sz >= 0)) {
-+ err = -EIO;
-+ AuIOErr("xigen error (%zd)\n", sz);
++ ninfo = 0;
++ for (bindex = btgt + 1; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ err = au_drinfo_load(&w, bindex, br);
++ if (unlikely(err))
++ goto out_free;
++ if (w.drinfo[bindex])
++ ninfo++;
+ }
++ if (!ninfo) {
++ br = au_sbr(sb, btgt);
++ h_dir = d_inode(w.h_ppath.dentry);
++ ent = au_dr_hino_find(&br->br_dirren, h_dir->i_ino);
++ AuDebugOn(!ent);
++ au_dr_hino_del(&br->br_dirren, ent);
++ kfree(ent);
++ }
++ goto out; /* success */
+
++out_free:
++ au_dr_lkup_free(lkup->dirren.drinfo, lkup->dirren.ninfo);
++ lkup->dirren.ninfo = 0;
++ lkup->dirren.drinfo = NULL;
+out:
++ AuTraceErr(err);
+ return err;
+}
+
-+int au_xigen_set(struct super_block *sb, struct file *base)
++void au_dr_lkup_fin(struct au_do_lookup_args *lkup)
+{
-+ int err;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file;
++ au_dr_lkup_free(lkup->dirren.drinfo, lkup->dirren.ninfo);
++}
+
-+ SiMustWriteLock(sb);
++int au_dr_lkup_name(struct au_do_lookup_args *lkup, aufs_bindex_t btgt)
++{
++ int err;
++ struct au_drinfo *drinfo;
+
-+ sbinfo = au_sbi(sb);
-+ file = au_xino_create2(base, sbinfo->si_xigen);
-+ err = PTR_ERR(file);
-+ if (IS_ERR(file))
-+ goto out;
+ err = 0;
-+ if (sbinfo->si_xigen)
-+ fput(sbinfo->si_xigen);
-+ sbinfo->si_xigen = file;
++ if (!lkup->dirren.drinfo)
++ goto out;
++ AuDebugOn(lkup->dirren.ninfo < btgt + 1);
++ drinfo = lkup->dirren.drinfo[btgt + 1];
++ if (!drinfo)
++ goto out;
++
++ kfree(lkup->whname.name);
++ lkup->whname.name = NULL;
++ lkup->dirren.dr_name.len = drinfo->oldnamelen;
++ lkup->dirren.dr_name.name = drinfo->oldname;
++ lkup->name = &lkup->dirren.dr_name;
++ err = au_wh_name_alloc(&lkup->whname, lkup->name);
++ if (!err)
++ AuDbg("name %.*s, whname %.*s, b%d\n",
++ AuLNPair(lkup->name), AuLNPair(&lkup->whname),
++ btgt);
+
+out:
++ AuTraceErr(err);
+ return err;
+}
+
-+void au_xigen_clr(struct super_block *sb)
++int au_dr_lkup_h_ino(struct au_do_lookup_args *lkup, aufs_bindex_t bindex,
++ ino_t h_ino)
+{
-+ struct au_sbinfo *sbinfo;
++ int match;
++ struct au_drinfo *drinfo;
+
-+ SiMustWriteLock(sb);
++ match = 1;
++ if (!lkup->dirren.drinfo)
++ goto out;
++ AuDebugOn(lkup->dirren.ninfo < bindex + 1);
++ drinfo = lkup->dirren.drinfo[bindex + 1];
++ if (!drinfo)
++ goto out;
+
-+ sbinfo = au_sbi(sb);
-+ if (sbinfo->si_xigen) {
-+ fput(sbinfo->si_xigen);
-+ sbinfo->si_xigen = NULL;
-+ }
++ match = (drinfo->ino == h_ino);
++ AuDbg("match %d\n", match);
++
++out:
++ return match;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino,
-+ ino_t dir_ino)
++int au_dr_opt_set(struct super_block *sb)
+{
-+ struct dentry *dentry, *d;
-+ struct inode *inode;
-+ unsigned int sigen;
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
+
-+ dentry = NULL;
-+ inode = ilookup(sb, ino);
-+ if (!inode)
-+ goto out;
++ err = 0;
++ bbot = au_sbbot(sb);
++ for (bindex = 0; !err && bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ err = au_dr_hino(sb, bindex, /*br*/NULL, &br->br_path);
++ }
+
-+ dentry = ERR_PTR(-ESTALE);
-+ sigen = au_sigen(sb);
-+ if (unlikely(au_is_bad_inode(inode)
-+ || IS_DEADDIR(inode)
-+ || sigen != au_iigen(inode, NULL)))
-+ goto out_iput;
++ return err;
++}
+
-+ dentry = NULL;
-+ if (!dir_ino || S_ISDIR(inode->i_mode))
-+ dentry = d_find_alias(inode);
-+ else {
-+ spin_lock(&inode->i_lock);
-+ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
-+ spin_lock(&d->d_lock);
-+ if (!au_test_anon(d)
-+ && d_inode(d->d_parent)->i_ino == dir_ino) {
-+ dentry = dget_dlock(d);
-+ spin_unlock(&d->d_lock);
-+ break;
-+ }
-+ spin_unlock(&d->d_lock);
-+ }
-+ spin_unlock(&inode->i_lock);
++int au_dr_opt_flush(struct super_block *sb)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++
++ err = 0;
++ bbot = au_sbbot(sb);
++ for (bindex = 0; !err && bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_writable(br->br_perm))
++ err = au_dr_hino(sb, bindex, /*br*/NULL, /*path*/NULL);
+ }
-+ if (unlikely(dentry && au_digen_test(dentry, sigen))) {
-+ /* need to refresh */
-+ dput(dentry);
-+ dentry = NULL;
++
++ return err;
++}
++
++int au_dr_opt_clr(struct super_block *sb, int no_flush)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++
++ err = 0;
++ if (!no_flush) {
++ err = au_dr_opt_flush(sb);
++ if (unlikely(err))
++ goto out;
++ }
++
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ au_dr_hino_free(&br->br_dirren);
+ }
+
-+out_iput:
-+ iput(inode);
+out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return err;
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dirren.h linux-4.14/fs/aufs/dirren.h
+--- linux-4.14/fs/aufs/dirren.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dirren.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,139 @@
++/*
++ * Copyright (C) 2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * renamed dir info
++ */
+
-+/* todo: dirty? */
-+/* if exportfs_decode_fh() passed vfsmount*, we could be happy */
++#ifndef __AUFS_DIRREN_H__
++#define __AUFS_DIRREN_H__
+
-+struct au_compare_mnt_args {
-+ /* input */
-+ struct super_block *sb;
++#ifdef __KERNEL__
+
-+ /* output */
-+ struct vfsmount *mnt;
-+};
++#include <linux/dcache.h>
++#include <linux/statfs.h>
++#include <linux/uuid.h>
++#include "hbl.h"
+
-+static int au_compare_mnt(struct vfsmount *mnt, void *arg)
-+{
-+ struct au_compare_mnt_args *a = arg;
++#define AuDirren_NHASH 100
+
-+ if (mnt->mnt_sb != a->sb)
-+ return 0;
-+ a->mnt = mntget(mnt);
-+ return 1;
-+}
++#ifdef CONFIG_AUFS_DIRREN
++enum au_brid_type {
++ AuBrid_Unset,
++ AuBrid_UUID,
++ AuBrid_FSID,
++ AuBrid_DEV
++};
+
-+static struct vfsmount *au_mnt_get(struct super_block *sb)
-+{
-+ int err;
-+ struct path root;
-+ struct au_compare_mnt_args args = {
-+ .sb = sb
++struct au_dr_brid {
++ enum au_brid_type type;
++ union {
++ uuid_t uuid; /* unimplemented yet */
++ fsid_t fsid;
++ dev_t dev;
+ };
++};
+
-+ get_fs_root(current->fs, &root);
-+ rcu_read_lock();
-+ err = iterate_mounts(au_compare_mnt, &args, root.mnt);
-+ rcu_read_unlock();
-+ path_put(&root);
-+ AuDebugOn(!err);
-+ AuDebugOn(!args.mnt);
-+ return args.mnt;
-+}
++/* 20 is the max digits length of ulong 64 */
++/* brid-type "_" uuid "_" inum */
++#define AUFS_DIRREN_FNAME_SZ (1 + 1 + UUID_STRING_LEN + 20)
++#define AUFS_DIRREN_ENV_VAL_SZ (AUFS_DIRREN_FNAME_SZ + 1 + 20)
+
-+struct au_nfsd_si_lock {
-+ unsigned int sigen;
-+ aufs_bindex_t bindex, br_id;
-+ unsigned char force_lock;
++struct au_dr_hino {
++ struct hlist_bl_node dr_hnode;
++ ino_t dr_h_ino;
+};
+
-+static int si_nfsd_read_lock(struct super_block *sb,
-+ struct au_nfsd_si_lock *nsi_lock)
-+{
-+ int err;
-+ aufs_bindex_t bindex;
++struct au_dr_br {
++ struct hlist_bl_head dr_h_ino[AuDirren_NHASH];
++ struct au_dr_brid dr_brid;
++};
+
-+ si_read_lock(sb, AuLock_FLUSH);
++struct au_dr_lookup {
++ /* dr_name is pointed by struct au_do_lookup_args.name */
++ struct qstr dr_name; /* subset of dr_info */
++ aufs_bindex_t ninfo;
++ struct au_drinfo **drinfo;
++};
++#else
++struct au_dr_hino;
++/* empty */
++struct au_dr_br { };
++struct au_dr_lookup { };
++#endif
+
-+ /* branch id may be wrapped around */
-+ err = 0;
-+ bindex = au_br_index(sb, nsi_lock->br_id);
-+ if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb))
-+ goto out; /* success */
++/* ---------------------------------------------------------------------- */
+
-+ err = -ESTALE;
-+ bindex = -1;
-+ if (!nsi_lock->force_lock)
-+ si_read_unlock(sb);
++struct au_branch;
++struct au_do_lookup_args;
++struct au_hinode;
++#ifdef CONFIG_AUFS_DIRREN
++int au_dr_hino_test_add(struct au_dr_br *dr, ino_t h_ino,
++ struct au_dr_hino *add_ent);
++void au_dr_hino_free(struct au_dr_br *dr);
++int au_dr_br_init(struct super_block *sb, struct au_branch *br,
++ const struct path *path);
++int au_dr_br_fin(struct super_block *sb, struct au_branch *br);
++int au_dr_rename(struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev);
++void au_dr_rename_fin(struct dentry *src, aufs_bindex_t btgt, void *rev);
++void au_dr_rename_rev(struct dentry *src, aufs_bindex_t bindex, void *rev);
++int au_dr_lkup(struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t bindex);
++int au_dr_lkup_name(struct au_do_lookup_args *lkup, aufs_bindex_t btgt);
++int au_dr_lkup_h_ino(struct au_do_lookup_args *lkup, aufs_bindex_t bindex,
++ ino_t h_ino);
++void au_dr_lkup_fin(struct au_do_lookup_args *lkup);
++int au_dr_opt_set(struct super_block *sb);
++int au_dr_opt_flush(struct super_block *sb);
++int au_dr_opt_clr(struct super_block *sb, int no_flush);
++#else
++AuStubInt0(au_dr_hino_test_add, struct au_dr_br *dr, ino_t h_ino,
++ struct au_dr_hino *add_ent);
++AuStubVoid(au_dr_hino_free, struct au_dr_br *dr);
++AuStubInt0(au_dr_br_init, struct super_block *sb, struct au_branch *br,
++ const struct path *path);
++AuStubInt0(au_dr_br_fin, struct super_block *sb, struct au_branch *br);
++AuStubInt0(au_dr_rename, struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev);
++AuStubVoid(au_dr_rename_fin, struct dentry *src, aufs_bindex_t btgt, void *rev);
++AuStubVoid(au_dr_rename_rev, struct dentry *src, aufs_bindex_t bindex,
++ void *rev);
++AuStubInt0(au_dr_lkup, struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t bindex);
++AuStubInt0(au_dr_lkup_name, struct au_do_lookup_args *lkup, aufs_bindex_t btgt);
++AuStubInt0(au_dr_lkup_h_ino, struct au_do_lookup_args *lkup,
++ aufs_bindex_t bindex, ino_t h_ino);
++AuStubVoid(au_dr_lkup_fin, struct au_do_lookup_args *lkup);
++AuStubInt0(au_dr_opt_set, struct super_block *sb);
++AuStubInt0(au_dr_opt_flush, struct super_block *sb);
++AuStubInt0(au_dr_opt_clr, struct super_block *sb, int no_flush);
++#endif
+
-+out:
-+ nsi_lock->bindex = bindex;
-+ return err;
++/* ---------------------------------------------------------------------- */
++
++#ifdef CONFIG_AUFS_DIRREN
++static inline int au_dr_ihash(ino_t h_ino)
++{
++ return h_ino % AuDirren_NHASH;
+}
++#else
++AuStubInt0(au_dr_ihash, ino_t h_ino);
++#endif
+
-+struct find_name_by_ino {
-+ struct dir_context ctx;
-+ int called, found;
-+ ino_t ino;
-+ char *name;
-+ int namelen;
-+};
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DIRREN_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dynop.c linux-4.14/fs/aufs/dynop.c
+--- linux-4.14/fs/aufs/dynop.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dynop.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,369 @@
++/*
++ * Copyright (C) 2010-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+static int
-+find_name_by_ino(struct dir_context *ctx, const char *name, int namelen,
-+ loff_t offset, u64 ino, unsigned int d_type)
-+{
-+ struct find_name_by_ino *a = container_of(ctx, struct find_name_by_ino,
-+ ctx);
++/*
++ * dynamically customizable operations for regular files
++ */
+
-+ a->called++;
-+ if (a->ino != ino)
-+ return 0;
++#include "aufs.h"
+
-+ memcpy(a->name, name, namelen);
-+ a->namelen = namelen;
-+ a->found = 1;
-+ return 1;
-+}
++#define DyPrSym(key) AuDbgSym(key->dk_op.dy_hop)
+
-+static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino,
-+ struct au_nfsd_si_lock *nsi_lock)
++/*
++ * How large will these lists be?
++ * Usually just a few elements, 20-30 at most for each, I guess.
++ */
++static struct hlist_bl_head dynop[AuDyLast];
++
++static struct au_dykey *dy_gfind_get(struct hlist_bl_head *hbl,
++ const void *h_op)
+{
-+ struct dentry *dentry, *parent;
-+ struct file *file;
-+ struct inode *dir;
-+ struct find_name_by_ino arg = {
-+ .ctx = {
-+ .actor = find_name_by_ino
++ struct au_dykey *key, *tmp;
++ struct hlist_bl_node *pos;
++
++ key = NULL;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(tmp, pos, hbl, dk_hnode)
++ if (tmp->dk_op.dy_hop == h_op) {
++ key = tmp;
++ kref_get(&key->dk_kref);
++ break;
+ }
-+ };
-+ int err;
++ hlist_bl_unlock(hbl);
+
-+ parent = path->dentry;
-+ if (nsi_lock)
-+ si_read_unlock(parent->d_sb);
-+ file = vfsub_dentry_open(path, au_dir_roflags);
-+ dentry = (void *)file;
-+ if (IS_ERR(file))
-+ goto out;
++ return key;
++}
+
-+ dentry = ERR_PTR(-ENOMEM);
-+ arg.name = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!arg.name))
-+ goto out_file;
-+ arg.ino = ino;
-+ arg.found = 0;
-+ do {
-+ arg.called = 0;
-+ /* smp_mb(); */
-+ err = vfsub_iterate_dir(file, &arg.ctx);
-+ } while (!err && !arg.found && arg.called);
-+ dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_name;
-+ /* instead of ENOENT */
-+ dentry = ERR_PTR(-ESTALE);
-+ if (!arg.found)
-+ goto out_name;
++static struct au_dykey *dy_bradd(struct au_branch *br, struct au_dykey *key)
++{
++ struct au_dykey **k, *found;
++ const void *h_op = key->dk_op.dy_hop;
++ int i;
+
-+ /* do not call vfsub_lkup_one() */
-+ dir = d_inode(parent);
-+ dentry = vfsub_lookup_one_len_unlocked(arg.name, parent, arg.namelen);
-+ AuTraceErrPtr(dentry);
-+ if (IS_ERR(dentry))
-+ goto out_name;
-+ AuDebugOn(au_test_anon(dentry));
-+ if (unlikely(d_really_is_negative(dentry))) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ENOENT);
++ found = NULL;
++ k = br->br_dykey;
++ for (i = 0; i < AuBrDynOp; i++)
++ if (k[i]) {
++ if (k[i]->dk_op.dy_hop == h_op) {
++ found = k[i];
++ break;
++ }
++ } else
++ break;
++ if (!found) {
++ spin_lock(&br->br_dykey_lock);
++ for (; i < AuBrDynOp; i++)
++ if (k[i]) {
++ if (k[i]->dk_op.dy_hop == h_op) {
++ found = k[i];
++ break;
++ }
++ } else {
++ k[i] = key;
++ break;
++ }
++ spin_unlock(&br->br_dykey_lock);
++ BUG_ON(i == AuBrDynOp); /* expand the array */
+ }
+
-+out_name:
-+ au_delayed_free_page((unsigned long)arg.name);
-+out_file:
-+ fput(file);
-+out:
-+ if (unlikely(nsi_lock
-+ && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0))
-+ if (!IS_ERR(dentry)) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ }
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return found;
+}
+
-+static struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino,
-+ ino_t dir_ino,
-+ struct au_nfsd_si_lock *nsi_lock)
++/* kref_get() if @key is already added */
++static struct au_dykey *dy_gadd(struct hlist_bl_head *hbl, struct au_dykey *key)
+{
-+ struct dentry *dentry;
-+ struct path path;
-+
-+ if (dir_ino != AUFS_ROOT_INO) {
-+ path.dentry = decode_by_ino(sb, dir_ino, 0);
-+ dentry = path.dentry;
-+ if (!path.dentry || IS_ERR(path.dentry))
-+ goto out;
-+ AuDebugOn(au_test_anon(path.dentry));
-+ } else
-+ path.dentry = dget(sb->s_root);
++ struct au_dykey *tmp, *found;
++ struct hlist_bl_node *pos;
++ const void *h_op = key->dk_op.dy_hop;
+
-+ path.mnt = au_mnt_get(sb);
-+ dentry = au_lkup_by_ino(&path, ino, nsi_lock);
-+ path_put(&path);
++ found = NULL;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(tmp, pos, hbl, dk_hnode)
++ if (tmp->dk_op.dy_hop == h_op) {
++ kref_get(&tmp->dk_kref);
++ found = tmp;
++ break;
++ }
++ if (!found)
++ hlist_bl_add_head(&key->dk_hnode, hbl);
++ hlist_bl_unlock(hbl);
+
-+out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ if (!found)
++ DyPrSym(key);
++ return found;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static int h_acceptable(void *expv, struct dentry *dentry)
++static void dy_free_rcu(struct rcu_head *rcu)
+{
-+ return 1;
++ struct au_dykey *key;
++
++ key = container_of(rcu, struct au_dykey, dk_rcu);
++ DyPrSym(key);
++ kfree(key);
+}
+
-+static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath,
-+ char *buf, int len, struct super_block *sb)
++static void dy_free(struct kref *kref)
+{
-+ char *p;
-+ int n;
-+ struct path path;
++ struct au_dykey *key;
++ struct hlist_bl_head *hbl;
+
-+ p = d_path(h_rootpath, buf, len);
-+ if (IS_ERR(p))
-+ goto out;
-+ n = strlen(p);
++ key = container_of(kref, struct au_dykey, dk_kref);
++ hbl = dynop + key->dk_op.dy_type;
++ au_hbl_del(&key->dk_hnode, hbl);
++ call_rcu(&key->dk_rcu, dy_free_rcu);
++}
+
-+ path.mnt = h_rootpath->mnt;
-+ path.dentry = h_parent;
-+ p = d_path(&path, buf, len);
-+ if (IS_ERR(p))
-+ goto out;
-+ if (n != 1)
-+ p += n;
++void au_dy_put(struct au_dykey *key)
++{
++ kref_put(&key->dk_kref, dy_free);
++}
+
-+ path.mnt = au_mnt_get(sb);
-+ path.dentry = sb->s_root;
-+ p = d_path(&path, buf, len - strlen(p));
-+ mntput(path.mnt);
-+ if (IS_ERR(p))
-+ goto out;
-+ if (n != 1)
-+ p[strlen(p)] = '/';
++/* ---------------------------------------------------------------------- */
+
-+out:
-+ AuTraceErrPtr(p);
-+ return p;
-+}
++#define DyDbgSize(cnt, op) AuDebugOn(cnt != sizeof(op)/sizeof(void *))
+
-+static
-+struct dentry *decode_by_path(struct super_block *sb, ino_t ino, __u32 *fh,
-+ int fh_len, struct au_nfsd_si_lock *nsi_lock)
-+{
-+ struct dentry *dentry, *h_parent, *root;
-+ struct super_block *h_sb;
-+ char *pathname, *p;
-+ struct vfsmount *h_mnt;
-+ struct au_branch *br;
-+ int err;
-+ struct path path;
++#ifdef CONFIG_AUFS_DEBUG
++#define DyDbgDeclare(cnt) unsigned int cnt = 0
++#define DyDbgInc(cnt) do { cnt++; } while (0)
++#else
++#define DyDbgDeclare(cnt) do {} while (0)
++#define DyDbgInc(cnt) do {} while (0)
++#endif
+
-+ br = au_sbr(sb, nsi_lock->bindex);
-+ h_mnt = au_br_mnt(br);
-+ h_sb = h_mnt->mnt_sb;
-+ /* todo: call lower fh_to_dentry()? fh_to_parent()? */
-+ lockdep_off();
-+ h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail),
-+ fh_len - Fh_tail, fh[Fh_h_type],
-+ h_acceptable, /*context*/NULL);
-+ lockdep_on();
-+ dentry = h_parent;
-+ if (unlikely(!h_parent || IS_ERR(h_parent))) {
-+ AuWarn1("%s decode_fh failed, %ld\n",
-+ au_sbtype(h_sb), PTR_ERR(h_parent));
-+ goto out;
-+ }
-+ dentry = NULL;
-+ if (unlikely(au_test_anon(h_parent))) {
-+ AuWarn1("%s decode_fh returned a disconnected dentry\n",
-+ au_sbtype(h_sb));
-+ goto out_h_parent;
-+ }
++#define DySet(func, dst, src, h_op, h_sb) do { \
++ DyDbgInc(cnt); \
++ if (h_op->func) { \
++ if (src.func) \
++ dst.func = src.func; \
++ else \
++ AuDbg("%s %s\n", au_sbtype(h_sb), #func); \
++ } \
++} while (0)
+
-+ dentry = ERR_PTR(-ENOMEM);
-+ pathname = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!pathname))
-+ goto out_h_parent;
++#define DySetForce(func, dst, src) do { \
++ AuDebugOn(!src.func); \
++ DyDbgInc(cnt); \
++ dst.func = src.func; \
++} while (0)
+
-+ root = sb->s_root;
-+ path.mnt = h_mnt;
-+ di_read_lock_parent(root, !AuLock_IR);
-+ path.dentry = au_h_dptr(root, nsi_lock->bindex);
-+ di_read_unlock(root, !AuLock_IR);
-+ p = au_build_path(h_parent, &path, pathname, PAGE_SIZE, sb);
-+ dentry = (void *)p;
-+ if (IS_ERR(p))
-+ goto out_pathname;
++#define DySetAop(func) \
++ DySet(func, dyaop->da_op, aufs_aop, h_aop, h_sb)
++#define DySetAopForce(func) \
++ DySetForce(func, dyaop->da_op, aufs_aop)
+
-+ si_read_unlock(sb);
-+ err = vfsub_kern_path(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path);
-+ dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_relock;
++static void dy_aop(struct au_dykey *key, const void *h_op,
++ struct super_block *h_sb __maybe_unused)
++{
++ struct au_dyaop *dyaop = (void *)key;
++ const struct address_space_operations *h_aop = h_op;
++ DyDbgDeclare(cnt);
+
-+ dentry = ERR_PTR(-ENOENT);
-+ AuDebugOn(au_test_anon(path.dentry));
-+ if (unlikely(d_really_is_negative(path.dentry)))
-+ goto out_path;
++ AuDbg("%s\n", au_sbtype(h_sb));
+
-+ if (ino != d_inode(path.dentry)->i_ino)
-+ dentry = au_lkup_by_ino(&path, ino, /*nsi_lock*/NULL);
-+ else
-+ dentry = dget(path.dentry);
++ DySetAop(writepage);
++ DySetAopForce(readpage); /* force */
++ DySetAop(writepages);
++ DySetAop(set_page_dirty);
++ DySetAop(readpages);
++ DySetAop(write_begin);
++ DySetAop(write_end);
++ DySetAop(bmap);
++ DySetAop(invalidatepage);
++ DySetAop(releasepage);
++ DySetAop(freepage);
++ /* this one will be changed according to an aufs mount option */
++ DySetAop(direct_IO);
++ DySetAop(migratepage);
++ DySetAop(isolate_page);
++ DySetAop(putback_page);
++ DySetAop(launder_page);
++ DySetAop(is_partially_uptodate);
++ DySetAop(is_dirty_writeback);
++ DySetAop(error_remove_page);
++ DySetAop(swap_activate);
++ DySetAop(swap_deactivate);
+
-+out_path:
-+ path_put(&path);
-+out_relock:
-+ if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0))
-+ if (!IS_ERR(dentry)) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ }
-+out_pathname:
-+ au_delayed_free_page((unsigned long)pathname);
-+out_h_parent:
-+ dput(h_parent);
-+out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ DyDbgSize(cnt, *h_aop);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *
-+aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len,
-+ int fh_type)
++static void dy_bug(struct kref *kref)
+{
-+ struct dentry *dentry;
-+ __u32 *fh = fid->raw;
-+ struct au_branch *br;
-+ ino_t ino, dir_ino;
-+ struct au_nfsd_si_lock nsi_lock = {
-+ .force_lock = 0
++ BUG();
++}
++
++static struct au_dykey *dy_get(struct au_dynop *op, struct au_branch *br)
++{
++ struct au_dykey *key, *old;
++ struct hlist_bl_head *hbl;
++ struct op {
++ unsigned int sz;
++ void (*set)(struct au_dykey *key, const void *h_op,
++ struct super_block *h_sb __maybe_unused);
++ };
++ static const struct op a[] = {
++ [AuDy_AOP] = {
++ .sz = sizeof(struct au_dyaop),
++ .set = dy_aop
++ }
+ };
++ const struct op *p;
+
-+ dentry = ERR_PTR(-ESTALE);
-+ /* it should never happen, but the file handle is unreliable */
-+ if (unlikely(fh_len < Fh_tail))
-+ goto out;
-+ nsi_lock.sigen = fh[Fh_sigen];
-+ nsi_lock.br_id = fh[Fh_br_id];
++ hbl = dynop + op->dy_type;
++ key = dy_gfind_get(hbl, op->dy_hop);
++ if (key)
++ goto out_add; /* success */
+
-+ /* branch id may be wrapped around */
-+ br = NULL;
-+ if (unlikely(si_nfsd_read_lock(sb, &nsi_lock)))
++ p = a + op->dy_type;
++ key = kzalloc(p->sz, GFP_NOFS);
++ if (unlikely(!key)) {
++ key = ERR_PTR(-ENOMEM);
+ goto out;
-+ nsi_lock.force_lock = 1;
++ }
+
-+ /* is this inode still cached? */
-+ ino = decode_ino(fh + Fh_ino);
-+ /* it should never happen */
-+ if (unlikely(ino == AUFS_ROOT_INO))
-+ goto out_unlock;
++ key->dk_op.dy_hop = op->dy_hop;
++ kref_init(&key->dk_kref);
++ p->set(key, op->dy_hop, au_br_sb(br));
++ old = dy_gadd(hbl, key);
++ if (old) {
++ kfree(key);
++ key = old;
++ }
+
-+ dir_ino = decode_ino(fh + Fh_dir_ino);
-+ dentry = decode_by_ino(sb, ino, dir_ino);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (dentry)
-+ goto accept;
++out_add:
++ old = dy_bradd(br, key);
++ if (old)
++ /* its ref-count should never be zero here */
++ kref_put(&key->dk_kref, dy_bug);
++out:
++ return key;
++}
+
-+ /* is the parent dir cached? */
-+ br = au_sbr(sb, nsi_lock.bindex);
-+ au_br_get(br);
-+ dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (dentry)
-+ goto accept;
++/* ---------------------------------------------------------------------- */
++/*
++ * Aufs prohibits O_DIRECT by defaut even if the branch supports it.
++ * This behaviour is necessary to return an error from open(O_DIRECT) instead
++ * of the succeeding I/O. The dio mount option enables O_DIRECT and makes
++ * open(O_DIRECT) always succeed, but the succeeding I/O may return an error.
++ * See the aufs manual in detail.
++ */
++static void dy_adx(struct au_dyaop *dyaop, int do_dx)
++{
++ if (!do_dx)
++ dyaop->da_op.direct_IO = NULL;
++ else
++ dyaop->da_op.direct_IO = aufs_aop.direct_IO;
++}
+
-+ /* lookup path */
-+ dentry = decode_by_path(sb, ino, fh, fh_len, &nsi_lock);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (unlikely(!dentry))
-+ /* todo?: make it ESTALE */
-+ goto out_unlock;
++static struct au_dyaop *dy_aget(struct au_branch *br,
++ const struct address_space_operations *h_aop,
++ int do_dx)
++{
++ struct au_dyaop *dyaop;
++ struct au_dynop op;
+
-+accept:
-+ if (!au_digen_test(dentry, au_sigen(sb))
-+ && d_inode(dentry)->i_generation == fh[Fh_igen])
-+ goto out_unlock; /* success */
++ op.dy_type = AuDy_AOP;
++ op.dy_haop = h_aop;
++ dyaop = (void *)dy_get(&op, br);
++ if (IS_ERR(dyaop))
++ goto out;
++ dy_adx(dyaop, do_dx);
+
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+out_unlock:
-+ if (br)
-+ au_br_put(br);
-+ si_read_unlock(sb);
+out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return dyaop;
+}
+
-+#if 0 /* reserved for future use */
-+/* support subtreecheck option */
-+static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid,
-+ int fh_len, int fh_type)
++int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode)
+{
-+ struct dentry *parent;
-+ __u32 *fh = fid->raw;
-+ ino_t dir_ino;
++ int err, do_dx;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_dyaop *dyaop;
+
-+ dir_ino = decode_ino(fh + Fh_dir_ino);
-+ parent = decode_by_ino(sb, dir_ino, 0);
-+ if (IS_ERR(parent))
++ AuDebugOn(!S_ISREG(h_inode->i_mode));
++ IiMustWriteLock(inode);
++
++ sb = inode->i_sb;
++ br = au_sbr(sb, bindex);
++ do_dx = !!au_opt_test(au_mntflags(sb), DIO);
++ dyaop = dy_aget(br, h_inode->i_mapping->a_ops, do_dx);
++ err = PTR_ERR(dyaop);
++ if (IS_ERR(dyaop))
++ /* unnecessary to call dy_fput() */
+ goto out;
-+ if (!parent)
-+ parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]),
-+ dir_ino, fh, fh_len);
++
++ err = 0;
++ inode->i_mapping->a_ops = &dyaop->da_op;
+
+out:
-+ AuTraceErrPtr(parent);
-+ return parent;
++ return err;
+}
-+#endif
-+
-+/* ---------------------------------------------------------------------- */
+
-+static int aufs_encode_fh(struct inode *inode, __u32 *fh, int *max_len,
-+ struct inode *dir)
++/*
++ * Is it safe to replace a_ops during the inode/file is in operation?
++ * Yes, I hope so.
++ */
++int au_dy_irefresh(struct inode *inode)
+{
+ int err;
-+ aufs_bindex_t bindex;
-+ struct super_block *sb, *h_sb;
-+ struct dentry *dentry, *parent, *h_parent;
-+ struct inode *h_dir;
-+ struct au_branch *br;
-+
-+ err = -ENOSPC;
-+ if (unlikely(*max_len <= Fh_tail)) {
-+ AuWarn1("NFSv2 client (max_len %d)?\n", *max_len);
-+ goto out;
-+ }
++ aufs_bindex_t btop;
++ struct inode *h_inode;
+
-+ err = FILEID_ROOT;
-+ if (inode->i_ino == AUFS_ROOT_INO) {
-+ AuDebugOn(inode->i_ino != AUFS_ROOT_INO);
-+ goto out;
++ err = 0;
++ if (S_ISREG(inode->i_mode)) {
++ btop = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, btop);
++ err = au_dy_iaop(inode, btop, h_inode);
+ }
++ return err;
++}
+
-+ h_parent = NULL;
-+ sb = inode->i_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH);
-+ if (unlikely(err))
-+ goto out;
++void au_dy_arefresh(int do_dx)
++{
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dykey *key;
+
-+#ifdef CONFIG_AUFS_DEBUG
-+ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
-+ AuWarn1("NFS-exporting requires xino\n");
-+#endif
-+ err = -EIO;
-+ parent = NULL;
-+ ii_read_lock_child(inode);
-+ bindex = au_ibtop(inode);
-+ if (!dir) {
-+ dentry = d_find_any_alias(inode);
-+ if (unlikely(!dentry))
-+ goto out_unlock;
-+ AuDebugOn(au_test_anon(dentry));
-+ parent = dget_parent(dentry);
-+ dput(dentry);
-+ if (unlikely(!parent))
-+ goto out_unlock;
-+ if (d_really_is_positive(parent))
-+ dir = d_inode(parent);
-+ }
++ hbl = dynop + AuDy_AOP;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(key, pos, hbl, dk_hnode)
++ dy_adx((void *)key, do_dx);
++ hlist_bl_unlock(hbl);
++}
+
-+ ii_read_lock_parent(dir);
-+ h_dir = au_h_iptr(dir, bindex);
-+ ii_read_unlock(dir);
-+ if (unlikely(!h_dir))
-+ goto out_parent;
-+ h_parent = d_find_any_alias(h_dir);
-+ if (unlikely(!h_parent))
-+ goto out_hparent;
++/* ---------------------------------------------------------------------- */
+
-+ err = -EPERM;
-+ br = au_sbr(sb, bindex);
-+ h_sb = au_br_sb(br);
-+ if (unlikely(!h_sb->s_export_op)) {
-+ AuErr1("%s branch is not exportable\n", au_sbtype(h_sb));
-+ goto out_hparent;
-+ }
++void __init au_dy_init(void)
++{
++ int i;
+
-+ fh[Fh_br_id] = br->br_id;
-+ fh[Fh_sigen] = au_sigen(sb);
-+ encode_ino(fh + Fh_ino, inode->i_ino);
-+ encode_ino(fh + Fh_dir_ino, dir->i_ino);
-+ fh[Fh_igen] = inode->i_generation;
++ /* make sure that 'struct au_dykey *' can be any type */
++ BUILD_BUG_ON(offsetof(struct au_dyaop, da_key));
+
-+ *max_len -= Fh_tail;
-+ fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail),
-+ max_len,
-+ /*connectable or subtreecheck*/0);
-+ err = fh[Fh_h_type];
-+ *max_len += Fh_tail;
-+ /* todo: macros? */
-+ if (err != FILEID_INVALID)
-+ err = 99;
-+ else
-+ AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb));
++ for (i = 0; i < AuDyLast; i++)
++ INIT_HLIST_BL_HEAD(dynop + i);
++}
+
-+out_hparent:
-+ dput(h_parent);
-+out_parent:
-+ dput(parent);
-+out_unlock:
-+ ii_read_unlock(inode);
-+ si_read_unlock(sb);
-+out:
-+ if (unlikely(err < 0))
-+ err = FILEID_INVALID;
-+ return err;
++void au_dy_fin(void)
++{
++ int i;
++
++ for (i = 0; i < AuDyLast; i++)
++ WARN_ON(!hlist_bl_empty(dynop + i));
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/dynop.h linux-4.14/fs/aufs/dynop.h
+--- linux-4.14/fs/aufs/dynop.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/dynop.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,74 @@
++/*
++ * Copyright (C) 2010-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * dynamically customizable operations (for regular files only)
++ */
+
-+static int aufs_commit_metadata(struct inode *inode)
-+{
-+ int err;
-+ aufs_bindex_t bindex;
-+ struct super_block *sb;
-+ struct inode *h_inode;
-+ int (*f)(struct inode *inode);
++#ifndef __AUFS_DYNOP_H__
++#define __AUFS_DYNOP_H__
+
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
-+ ii_write_lock_child(inode);
-+ bindex = au_ibtop(inode);
-+ AuDebugOn(bindex < 0);
-+ h_inode = au_h_iptr(inode, bindex);
++#ifdef __KERNEL__
+
-+ f = h_inode->i_sb->s_export_op->commit_metadata;
-+ if (f)
-+ err = f(h_inode);
-+ else {
-+ struct writeback_control wbc = {
-+ .sync_mode = WB_SYNC_ALL,
-+ .nr_to_write = 0 /* metadata only */
-+ };
++#include <linux/fs.h>
++#include <linux/kref.h>
+
-+ err = sync_inode(h_inode, &wbc);
-+ }
++enum {AuDy_AOP, AuDyLast};
+
-+ au_cpup_attr_timesizes(inode);
-+ ii_write_unlock(inode);
-+ si_read_unlock(sb);
-+ return err;
-+}
++struct au_dynop {
++ int dy_type;
++ union {
++ const void *dy_hop;
++ const struct address_space_operations *dy_haop;
++ };
++};
+
-+/* ---------------------------------------------------------------------- */
++struct au_dykey {
++ union {
++ struct hlist_bl_node dk_hnode;
++ struct rcu_head dk_rcu;
++ };
++ struct au_dynop dk_op;
+
-+static struct export_operations aufs_export_op = {
-+ .fh_to_dentry = aufs_fh_to_dentry,
-+ /* .fh_to_parent = aufs_fh_to_parent, */
-+ .encode_fh = aufs_encode_fh,
-+ .commit_metadata = aufs_commit_metadata
++ /*
++ * during I am in the branch local array, kref is gotten. when the
++ * branch is removed, kref is put.
++ */
++ struct kref dk_kref;
+};
+
-+void au_export_init(struct super_block *sb)
-+{
-+ struct au_sbinfo *sbinfo;
-+ __u32 u;
++/* stop unioning since their sizes are very different from each other */
++struct au_dyaop {
++ struct au_dykey da_key;
++ struct address_space_operations da_op; /* not const */
++};
+
-+ BUILD_BUG_ON_MSG(IS_BUILTIN(CONFIG_AUFS_FS)
-+ && IS_MODULE(CONFIG_EXPORTFS),
-+ AUFS_NAME ": unsupported configuration "
-+ "CONFIG_EXPORTFS=m and CONFIG_AUFS_FS=y");
++/* ---------------------------------------------------------------------- */
+
-+ sb->s_export_op = &aufs_export_op;
-+ sbinfo = au_sbi(sb);
-+ sbinfo->si_xigen = NULL;
-+ get_random_bytes(&u, sizeof(u));
-+ BUILD_BUG_ON(sizeof(u) != sizeof(int));
-+ atomic_set(&sbinfo->si_xigen_next, u);
-+}
-diff -urN /usr/share/empty/fs/aufs/fhsm.c linux/fs/aufs/fhsm.c
---- /usr/share/empty/fs/aufs/fhsm.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/fhsm.c 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,426 @@
++/* dynop.c */
++struct au_branch;
++void au_dy_put(struct au_dykey *key);
++int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode);
++int au_dy_irefresh(struct inode *inode);
++void au_dy_arefresh(int do_dio);
++
++void __init au_dy_init(void);
++void au_dy_fin(void);
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DYNOP_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/export.c linux-4.14/fs/aufs/export.c
+--- linux-4.14/fs/aufs/export.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/export.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,836 @@
+/*
-+ * Copyright (C) 2011-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * File-based Hierarchy Storage Management
++ * export via nfs
+ */
+
-+#include <linux/anon_inodes.h>
-+#include <linux/poll.h>
-+#include <linux/seq_file.h>
-+#include <linux/statfs.h>
++#include <linux/exportfs.h>
++#include <linux/fs_struct.h>
++#include <linux/namei.h>
++#include <linux/nsproxy.h>
++#include <linux/random.h>
++#include <linux/writeback.h>
+#include "aufs.h"
+
-+static aufs_bindex_t au_fhsm_bottom(struct super_block *sb)
-+{
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
++union conv {
++#ifdef CONFIG_AUFS_INO_T_64
++ __u32 a[2];
++#else
++ __u32 a[1];
++#endif
++ ino_t ino;
++};
+
-+ SiMustAnyLock(sb);
++static ino_t decode_ino(__u32 *a)
++{
++ union conv u;
+
-+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ AuDebugOn(!fhsm);
-+ return fhsm->fhsm_bottom;
++ BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a));
++ u.a[0] = a[0];
++#ifdef CONFIG_AUFS_INO_T_64
++ u.a[1] = a[1];
++#endif
++ return u.ino;
+}
+
-+void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex)
++static void encode_ino(__u32 *a, ino_t ino)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
-+
-+ SiMustWriteLock(sb);
++ union conv u;
+
-+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ AuDebugOn(!fhsm);
-+ fhsm->fhsm_bottom = bindex;
++ u.ino = ino;
++ a[0] = u.a[0];
++#ifdef CONFIG_AUFS_INO_T_64
++ a[1] = u.a[1];
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
++/* NFS file handle */
++enum {
++ Fh_br_id,
++ Fh_sigen,
++#ifdef CONFIG_AUFS_INO_T_64
++ /* support 64bit inode number */
++ Fh_ino1,
++ Fh_ino2,
++ Fh_dir_ino1,
++ Fh_dir_ino2,
++#else
++ Fh_ino1,
++ Fh_dir_ino1,
++#endif
++ Fh_igen,
++ Fh_h_type,
++ Fh_tail,
+
-+static int au_fhsm_test_jiffy(struct au_sbinfo *sbinfo, struct au_branch *br)
++ Fh_ino = Fh_ino1,
++ Fh_dir_ino = Fh_dir_ino1
++};
++
++static int au_test_anon(struct dentry *dentry)
+{
-+ struct au_br_fhsm *bf;
++ /* note: read d_flags without d_lock */
++ return !!(dentry->d_flags & DCACHE_DISCONNECTED);
++}
+
-+ bf = br->br_fhsm;
-+ MtxMustLock(&bf->bf_lock);
++int au_test_nfsd(void)
++{
++ int ret;
++ struct task_struct *tsk = current;
++ char comm[sizeof(tsk->comm)];
+
-+ return !bf->bf_readable
-+ || time_after(jiffies,
-+ bf->bf_jiffy + sbinfo->si_fhsm.fhsm_expire);
++ ret = 0;
++ if (tsk->flags & PF_KTHREAD) {
++ get_task_comm(comm, tsk);
++ ret = !strcmp(comm, "nfsd");
++ }
++
++ return ret;
+}
+
+/* ---------------------------------------------------------------------- */
++/* inode generation external table */
+
-+static void au_fhsm_notify(struct super_block *sb, int val)
++void au_xigen_inc(struct inode *inode)
+{
++ loff_t pos;
++ ssize_t sz;
++ __u32 igen;
++ struct super_block *sb;
+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
+
-+ SiMustAnyLock(sb);
++ sb = inode->i_sb;
++ AuDebugOn(!au_opt_test(au_mntflags(sb), XINO));
+
+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ if (au_fhsm_pid(fhsm)
-+ && atomic_read(&fhsm->fhsm_readable) != -1) {
-+ atomic_set(&fhsm->fhsm_readable, val);
-+ if (val)
-+ wake_up(&fhsm->fhsm_wqh);
-+ }
++ pos = inode->i_ino;
++ pos *= sizeof(igen);
++ igen = inode->i_generation + 1;
++ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xigen, &igen,
++ sizeof(igen), &pos);
++ if (sz == sizeof(igen))
++ return; /* success */
++
++ if (unlikely(sz >= 0))
++ AuIOErr("xigen error (%zd)\n", sz);
+}
+
-+static int au_fhsm_stfs(struct super_block *sb, aufs_bindex_t bindex,
-+ struct aufs_stfs *rstfs, int do_lock, int do_notify)
++int au_xigen_new(struct inode *inode)
+{
+ int err;
-+ struct au_branch *br;
-+ struct au_br_fhsm *bf;
-+
-+ br = au_sbr(sb, bindex);
-+ AuDebugOn(au_br_rdonly(br));
-+ bf = br->br_fhsm;
-+ AuDebugOn(!bf);
++ loff_t pos;
++ ssize_t sz;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
+
-+ if (do_lock)
-+ mutex_lock(&bf->bf_lock);
-+ else
-+ MtxMustLock(&bf->bf_lock);
++ err = 0;
++ /* todo: dirty, at mount time */
++ if (inode->i_ino == AUFS_ROOT_INO)
++ goto out;
++ sb = inode->i_sb;
++ SiMustAnyLock(sb);
++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
++ goto out;
+
-+ /* sb->s_root for NFS is unreliable */
-+ err = au_br_stfs(br, &bf->bf_stfs);
-+ if (unlikely(err)) {
-+ AuErr1("FHSM failed (%d), b%d, ignored.\n", bindex, err);
++ err = -EFBIG;
++ pos = inode->i_ino;
++ if (unlikely(au_loff_max / sizeof(inode->i_generation) - 1 < pos)) {
++ AuIOErr1("too large i%lld\n", pos);
+ goto out;
+ }
++ pos *= sizeof(inode->i_generation);
+
-+ bf->bf_jiffy = jiffies;
-+ bf->bf_readable = 1;
-+ if (do_notify)
-+ au_fhsm_notify(sb, /*val*/1);
-+ if (rstfs)
-+ *rstfs = bf->bf_stfs;
++ err = 0;
++ sbinfo = au_sbi(sb);
++ file = sbinfo->si_xigen;
++ BUG_ON(!file);
+
-+out:
-+ if (do_lock)
-+ mutex_unlock(&bf->bf_lock);
-+ au_fhsm_notify(sb, /*val*/1);
++ if (vfsub_f_size_read(file)
++ < pos + sizeof(inode->i_generation)) {
++ inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next);
++ sz = xino_fwrite(sbinfo->si_xwrite, file, &inode->i_generation,
++ sizeof(inode->i_generation), &pos);
++ } else
++ sz = xino_fread(sbinfo->si_xread, file, &inode->i_generation,
++ sizeof(inode->i_generation), &pos);
++ if (sz == sizeof(inode->i_generation))
++ goto out; /* success */
++
++ err = sz;
++ if (unlikely(sz >= 0)) {
++ err = -EIO;
++ AuIOErr("xigen error (%zd)\n", sz);
++ }
+
++out:
+ return err;
+}
+
-+void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force)
++int au_xigen_set(struct super_block *sb, struct file *base)
+{
+ int err;
+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
-+ struct au_branch *br;
-+ struct au_br_fhsm *bf;
++ struct file *file;
+
-+ AuDbg("b%d, force %d\n", bindex, force);
-+ SiMustAnyLock(sb);
++ SiMustWriteLock(sb);
+
+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ if (!au_ftest_si(sbinfo, FHSM)
-+ || fhsm->fhsm_bottom == bindex)
-+ return;
++ file = au_xino_create2(base, sbinfo->si_xigen);
++ err = PTR_ERR(file);
++ if (IS_ERR(file))
++ goto out;
++ err = 0;
++ if (sbinfo->si_xigen)
++ fput(sbinfo->si_xigen);
++ sbinfo->si_xigen = file;
+
-+ br = au_sbr(sb, bindex);
-+ bf = br->br_fhsm;
-+ AuDebugOn(!bf);
-+ mutex_lock(&bf->bf_lock);
-+ if (force
-+ || au_fhsm_pid(fhsm)
-+ || au_fhsm_test_jiffy(sbinfo, br))
-+ err = au_fhsm_stfs(sb, bindex, /*rstfs*/NULL, /*do_lock*/0,
-+ /*do_notify*/1);
-+ mutex_unlock(&bf->bf_lock);
++out:
++ return err;
+}
+
-+void au_fhsm_wrote_all(struct super_block *sb, int force)
++void au_xigen_clr(struct super_block *sb)
+{
-+ aufs_bindex_t bindex, bbot;
-+ struct au_branch *br;
++ struct au_sbinfo *sbinfo;
+
-+ /* exclude the bottom */
-+ bbot = au_fhsm_bottom(sb);
-+ for (bindex = 0; bindex < bbot; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ if (au_br_fhsm(br->br_perm))
-+ au_fhsm_wrote(sb, bindex, force);
++ SiMustWriteLock(sb);
++
++ sbinfo = au_sbi(sb);
++ if (sbinfo->si_xigen) {
++ fput(sbinfo->si_xigen);
++ sbinfo->si_xigen = NULL;
+ }
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static unsigned int au_fhsm_poll(struct file *file,
-+ struct poll_table_struct *wait)
++static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino,
++ ino_t dir_ino)
+{
-+ unsigned int mask;
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
++ struct dentry *dentry, *d;
++ struct inode *inode;
++ unsigned int sigen;
+
-+ mask = 0;
-+ sbinfo = file->private_data;
-+ fhsm = &sbinfo->si_fhsm;
-+ poll_wait(file, &fhsm->fhsm_wqh, wait);
-+ if (atomic_read(&fhsm->fhsm_readable))
-+ mask = POLLIN /* | POLLRDNORM */;
++ dentry = NULL;
++ inode = ilookup(sb, ino);
++ if (!inode)
++ goto out;
+
-+ AuTraceErr((int)mask);
-+ return mask;
++ dentry = ERR_PTR(-ESTALE);
++ sigen = au_sigen(sb);
++ if (unlikely(au_is_bad_inode(inode)
++ || IS_DEADDIR(inode)
++ || sigen != au_iigen(inode, NULL)))
++ goto out_iput;
++
++ dentry = NULL;
++ if (!dir_ino || S_ISDIR(inode->i_mode))
++ dentry = d_find_alias(inode);
++ else {
++ spin_lock(&inode->i_lock);
++ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
++ spin_lock(&d->d_lock);
++ if (!au_test_anon(d)
++ && d_inode(d->d_parent)->i_ino == dir_ino) {
++ dentry = dget_dlock(d);
++ spin_unlock(&d->d_lock);
++ break;
++ }
++ spin_unlock(&d->d_lock);
++ }
++ spin_unlock(&inode->i_lock);
++ }
++ if (unlikely(dentry && au_digen_test(dentry, sigen))) {
++ /* need to refresh */
++ dput(dentry);
++ dentry = NULL;
++ }
++
++out_iput:
++ iput(inode);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+static int au_fhsm_do_read_one(struct aufs_stbr __user *stbr,
-+ struct aufs_stfs *stfs, __s16 brid)
-+{
-+ int err;
++/* ---------------------------------------------------------------------- */
+
-+ err = copy_to_user(&stbr->stfs, stfs, sizeof(*stfs));
-+ if (!err)
-+ err = __put_user(brid, &stbr->brid);
-+ if (unlikely(err))
-+ err = -EFAULT;
++/* todo: dirty? */
++/* if exportfs_decode_fh() passed vfsmount*, we could be happy */
+
-+ return err;
-+}
++struct au_compare_mnt_args {
++ /* input */
++ struct super_block *sb;
+
-+static ssize_t au_fhsm_do_read(struct super_block *sb,
-+ struct aufs_stbr __user *stbr, size_t count)
++ /* output */
++ struct vfsmount *mnt;
++};
++
++static int au_compare_mnt(struct vfsmount *mnt, void *arg)
+{
-+ ssize_t err;
-+ int nstbr;
-+ aufs_bindex_t bindex, bbot;
-+ struct au_branch *br;
-+ struct au_br_fhsm *bf;
++ struct au_compare_mnt_args *a = arg;
+
-+ /* except the bottom branch */
-+ err = 0;
-+ nstbr = 0;
-+ bbot = au_fhsm_bottom(sb);
-+ for (bindex = 0; !err && bindex < bbot; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ if (!au_br_fhsm(br->br_perm))
-+ continue;
++ if (mnt->mnt_sb != a->sb)
++ return 0;
++ a->mnt = mntget(mnt);
++ return 1;
++}
+
-+ bf = br->br_fhsm;
-+ mutex_lock(&bf->bf_lock);
-+ if (bf->bf_readable) {
-+ err = -EFAULT;
-+ if (count >= sizeof(*stbr))
-+ err = au_fhsm_do_read_one(stbr++, &bf->bf_stfs,
-+ br->br_id);
-+ if (!err) {
-+ bf->bf_readable = 0;
-+ count -= sizeof(*stbr);
-+ nstbr++;
-+ }
-+ }
-+ mutex_unlock(&bf->bf_lock);
-+ }
-+ if (!err)
-+ err = sizeof(*stbr) * nstbr;
++static struct vfsmount *au_mnt_get(struct super_block *sb)
++{
++ int err;
++ struct path root;
++ struct au_compare_mnt_args args = {
++ .sb = sb
++ };
+
-+ return err;
++ get_fs_root(current->fs, &root);
++ rcu_read_lock();
++ err = iterate_mounts(au_compare_mnt, &args, root.mnt);
++ rcu_read_unlock();
++ path_put(&root);
++ AuDebugOn(!err);
++ AuDebugOn(!args.mnt);
++ return args.mnt;
+}
+
-+static ssize_t au_fhsm_read(struct file *file, char __user *buf, size_t count,
-+ loff_t *pos)
-+{
-+ ssize_t err;
-+ int readable;
-+ aufs_bindex_t nfhsm, bindex, bbot;
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
-+ struct au_branch *br;
-+ struct super_block *sb;
++struct au_nfsd_si_lock {
++ unsigned int sigen;
++ aufs_bindex_t bindex, br_id;
++ unsigned char force_lock;
++};
++
++static int si_nfsd_read_lock(struct super_block *sb,
++ struct au_nfsd_si_lock *nsi_lock)
++{
++ int err;
++ aufs_bindex_t bindex;
++
++ si_read_lock(sb, AuLock_FLUSH);
+
++ /* branch id may be wrapped around */
+ err = 0;
-+ sbinfo = file->private_data;
-+ fhsm = &sbinfo->si_fhsm;
-+need_data:
-+ spin_lock_irq(&fhsm->fhsm_wqh.lock);
-+ if (!atomic_read(&fhsm->fhsm_readable)) {
-+ if (vfsub_file_flags(file) & O_NONBLOCK)
-+ err = -EAGAIN;
-+ else
-+ err = wait_event_interruptible_locked_irq
-+ (fhsm->fhsm_wqh,
-+ atomic_read(&fhsm->fhsm_readable));
-+ }
-+ spin_unlock_irq(&fhsm->fhsm_wqh.lock);
-+ if (unlikely(err))
-+ goto out;
++ bindex = au_br_index(sb, nsi_lock->br_id);
++ if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb))
++ goto out; /* success */
+
-+ /* sb may already be dead */
-+ au_rw_read_lock(&sbinfo->si_rwsem);
-+ readable = atomic_read(&fhsm->fhsm_readable);
-+ if (readable > 0) {
-+ sb = sbinfo->si_sb;
-+ AuDebugOn(!sb);
-+ /* exclude the bottom branch */
-+ nfhsm = 0;
-+ bbot = au_fhsm_bottom(sb);
-+ for (bindex = 0; bindex < bbot; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ if (au_br_fhsm(br->br_perm))
-+ nfhsm++;
-+ }
-+ err = -EMSGSIZE;
-+ if (nfhsm * sizeof(struct aufs_stbr) <= count) {
-+ atomic_set(&fhsm->fhsm_readable, 0);
-+ err = au_fhsm_do_read(sbinfo->si_sb, (void __user *)buf,
-+ count);
-+ }
-+ }
-+ au_rw_read_unlock(&sbinfo->si_rwsem);
-+ if (!readable)
-+ goto need_data;
++ err = -ESTALE;
++ bindex = -1;
++ if (!nsi_lock->force_lock)
++ si_read_unlock(sb);
+
+out:
++ nsi_lock->bindex = bindex;
+ return err;
+}
+
-+static int au_fhsm_release(struct inode *inode, struct file *file)
++struct find_name_by_ino {
++ struct dir_context ctx;
++ int called, found;
++ ino_t ino;
++ char *name;
++ int namelen;
++};
++
++static int
++find_name_by_ino(struct dir_context *ctx, const char *name, int namelen,
++ loff_t offset, u64 ino, unsigned int d_type)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
++ struct find_name_by_ino *a = container_of(ctx, struct find_name_by_ino,
++ ctx);
+
-+ /* sb may already be dead */
-+ sbinfo = file->private_data;
-+ fhsm = &sbinfo->si_fhsm;
-+ spin_lock(&fhsm->fhsm_spin);
-+ fhsm->fhsm_pid = 0;
-+ spin_unlock(&fhsm->fhsm_spin);
-+ kobject_put(&sbinfo->si_kobj);
++ a->called++;
++ if (a->ino != ino)
++ return 0;
+
-+ return 0;
++ memcpy(a->name, name, namelen);
++ a->namelen = namelen;
++ a->found = 1;
++ return 1;
+}
+
-+static const struct file_operations au_fhsm_fops = {
-+ .owner = THIS_MODULE,
-+ .llseek = noop_llseek,
-+ .read = au_fhsm_read,
-+ .poll = au_fhsm_poll,
-+ .release = au_fhsm_release
-+};
-+
-+int au_fhsm_fd(struct super_block *sb, int oflags)
++static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino,
++ struct au_nfsd_si_lock *nsi_lock)
+{
-+ int err, fd;
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
-+
-+ err = -EPERM;
-+ if (unlikely(!capable(CAP_SYS_ADMIN)))
-+ goto out;
++ struct dentry *dentry, *parent;
++ struct file *file;
++ struct inode *dir;
++ struct find_name_by_ino arg = {
++ .ctx = {
++ .actor = find_name_by_ino
++ }
++ };
++ int err;
+
-+ err = -EINVAL;
-+ if (unlikely(oflags & ~(O_CLOEXEC | O_NONBLOCK)))
++ parent = path->dentry;
++ if (nsi_lock)
++ si_read_unlock(parent->d_sb);
++ file = vfsub_dentry_open(path, au_dir_roflags);
++ dentry = (void *)file;
++ if (IS_ERR(file))
+ goto out;
+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ spin_lock(&fhsm->fhsm_spin);
-+ if (!fhsm->fhsm_pid)
-+ fhsm->fhsm_pid = current->pid;
-+ else
-+ err = -EBUSY;
-+ spin_unlock(&fhsm->fhsm_spin);
++ dentry = ERR_PTR(-ENOMEM);
++ arg.name = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!arg.name))
++ goto out_file;
++ arg.ino = ino;
++ arg.found = 0;
++ do {
++ arg.called = 0;
++ /* smp_mb(); */
++ err = vfsub_iterate_dir(file, &arg.ctx);
++ } while (!err && !arg.found && arg.called);
++ dentry = ERR_PTR(err);
+ if (unlikely(err))
-+ goto out;
-+
-+ oflags |= O_RDONLY;
-+ /* oflags |= FMODE_NONOTIFY; */
-+ fd = anon_inode_getfd("[aufs_fhsm]", &au_fhsm_fops, sbinfo, oflags);
-+ err = fd;
-+ if (unlikely(fd < 0))
-+ goto out_pid;
++ goto out_name;
++ /* instead of ENOENT */
++ dentry = ERR_PTR(-ESTALE);
++ if (!arg.found)
++ goto out_name;
+
-+ /* succeed reglardless 'fhsm' status */
-+ kobject_get(&sbinfo->si_kobj);
-+ si_noflush_read_lock(sb);
-+ if (au_ftest_si(sbinfo, FHSM))
-+ au_fhsm_wrote_all(sb, /*force*/0);
-+ si_read_unlock(sb);
-+ goto out; /* success */
++ /* do not call vfsub_lkup_one() */
++ dir = d_inode(parent);
++ dentry = vfsub_lookup_one_len_unlocked(arg.name, parent, arg.namelen);
++ AuTraceErrPtr(dentry);
++ if (IS_ERR(dentry))
++ goto out_name;
++ AuDebugOn(au_test_anon(dentry));
++ if (unlikely(d_really_is_negative(dentry))) {
++ dput(dentry);
++ dentry = ERR_PTR(-ENOENT);
++ }
+
-+out_pid:
-+ spin_lock(&fhsm->fhsm_spin);
-+ fhsm->fhsm_pid = 0;
-+ spin_unlock(&fhsm->fhsm_spin);
++out_name:
++ free_page((unsigned long)arg.name);
++out_file:
++ fput(file);
+out:
-+ AuTraceErr(err);
-+ return err;
++ if (unlikely(nsi_lock
++ && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0))
++ if (!IS_ERR(dentry)) {
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++ }
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+int au_fhsm_br_alloc(struct au_branch *br)
++static struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino,
++ ino_t dir_ino,
++ struct au_nfsd_si_lock *nsi_lock)
+{
-+ int err;
++ struct dentry *dentry;
++ struct path path;
+
-+ err = 0;
-+ br->br_fhsm = kmalloc(sizeof(*br->br_fhsm), GFP_NOFS);
-+ if (br->br_fhsm)
-+ au_br_fhsm_init(br->br_fhsm);
-+ else
-+ err = -ENOMEM;
++ if (dir_ino != AUFS_ROOT_INO) {
++ path.dentry = decode_by_ino(sb, dir_ino, 0);
++ dentry = path.dentry;
++ if (!path.dentry || IS_ERR(path.dentry))
++ goto out;
++ AuDebugOn(au_test_anon(path.dentry));
++ } else
++ path.dentry = dget(sb->s_root);
+
-+ return err;
++ path.mnt = au_mnt_get(sb);
++ dentry = au_lkup_by_ino(&path, ino, nsi_lock);
++ path_put(&path);
++
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+void au_fhsm_fin(struct super_block *sb)
++static int h_acceptable(void *expv, struct dentry *dentry)
+{
-+ au_fhsm_notify(sb, /*val*/-1);
++ return 1;
+}
+
-+void au_fhsm_init(struct au_sbinfo *sbinfo)
++static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath,
++ char *buf, int len, struct super_block *sb)
+{
-+ struct au_fhsm *fhsm;
++ char *p;
++ int n;
++ struct path path;
+
-+ fhsm = &sbinfo->si_fhsm;
-+ spin_lock_init(&fhsm->fhsm_spin);
-+ init_waitqueue_head(&fhsm->fhsm_wqh);
-+ atomic_set(&fhsm->fhsm_readable, 0);
-+ fhsm->fhsm_expire
-+ = msecs_to_jiffies(AUFS_FHSM_CACHE_DEF_SEC * MSEC_PER_SEC);
-+ fhsm->fhsm_bottom = -1;
-+}
++ p = d_path(h_rootpath, buf, len);
++ if (IS_ERR(p))
++ goto out;
++ n = strlen(p);
+
-+void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec)
-+{
-+ sbinfo->si_fhsm.fhsm_expire
-+ = msecs_to_jiffies(sec * MSEC_PER_SEC);
++ path.mnt = h_rootpath->mnt;
++ path.dentry = h_parent;
++ p = d_path(&path, buf, len);
++ if (IS_ERR(p))
++ goto out;
++ if (n != 1)
++ p += n;
++
++ path.mnt = au_mnt_get(sb);
++ path.dentry = sb->s_root;
++ p = d_path(&path, buf, len - strlen(p));
++ mntput(path.mnt);
++ if (IS_ERR(p))
++ goto out;
++ if (n != 1)
++ p[strlen(p)] = '/';
++
++out:
++ AuTraceErrPtr(p);
++ return p;
+}
+
-+void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo)
++static
++struct dentry *decode_by_path(struct super_block *sb, ino_t ino, __u32 *fh,
++ int fh_len, struct au_nfsd_si_lock *nsi_lock)
+{
-+ unsigned int u;
++ struct dentry *dentry, *h_parent, *root;
++ struct super_block *h_sb;
++ char *pathname, *p;
++ struct vfsmount *h_mnt;
++ struct au_branch *br;
++ int err;
++ struct path path;
+
-+ if (!au_ftest_si(sbinfo, FHSM))
-+ return;
++ br = au_sbr(sb, nsi_lock->bindex);
++ h_mnt = au_br_mnt(br);
++ h_sb = h_mnt->mnt_sb;
++ /* todo: call lower fh_to_dentry()? fh_to_parent()? */
++ lockdep_off();
++ h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail),
++ fh_len - Fh_tail, fh[Fh_h_type],
++ h_acceptable, /*context*/NULL);
++ lockdep_on();
++ dentry = h_parent;
++ if (unlikely(!h_parent || IS_ERR(h_parent))) {
++ AuWarn1("%s decode_fh failed, %ld\n",
++ au_sbtype(h_sb), PTR_ERR(h_parent));
++ goto out;
++ }
++ dentry = NULL;
++ if (unlikely(au_test_anon(h_parent))) {
++ AuWarn1("%s decode_fh returned a disconnected dentry\n",
++ au_sbtype(h_sb));
++ goto out_h_parent;
++ }
+
-+ u = jiffies_to_msecs(sbinfo->si_fhsm.fhsm_expire) / MSEC_PER_SEC;
-+ if (u != AUFS_FHSM_CACHE_DEF_SEC)
-+ seq_printf(seq, ",fhsm_sec=%u", u);
-+}
-diff -urN /usr/share/empty/fs/aufs/file.c linux/fs/aufs/file.c
---- /usr/share/empty/fs/aufs/file.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/file.c 2016-08-17 18:01:21.295617591 +0200
-@@ -0,0 +1,845 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
++ dentry = ERR_PTR(-ENOMEM);
++ pathname = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!pathname))
++ goto out_h_parent;
+
-+/*
-+ * handling file/dir, and address_space operation
-+ */
++ root = sb->s_root;
++ path.mnt = h_mnt;
++ di_read_lock_parent(root, !AuLock_IR);
++ path.dentry = au_h_dptr(root, nsi_lock->bindex);
++ di_read_unlock(root, !AuLock_IR);
++ p = au_build_path(h_parent, &path, pathname, PAGE_SIZE, sb);
++ dentry = (void *)p;
++ if (IS_ERR(p))
++ goto out_pathname;
+
-+#ifdef CONFIG_AUFS_DEBUG
-+#include <linux/migrate.h>
-+#endif
-+#include <linux/pagemap.h>
-+#include "aufs.h"
++ si_read_unlock(sb);
++ err = vfsub_kern_path(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path);
++ dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_relock;
+
-+/* drop flags for writing */
-+unsigned int au_file_roflags(unsigned int flags)
-+{
-+ flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC);
-+ flags |= O_RDONLY | O_NOATIME;
-+ return flags;
++ dentry = ERR_PTR(-ENOENT);
++ AuDebugOn(au_test_anon(path.dentry));
++ if (unlikely(d_really_is_negative(path.dentry)))
++ goto out_path;
++
++ if (ino != d_inode(path.dentry)->i_ino)
++ dentry = au_lkup_by_ino(&path, ino, /*nsi_lock*/NULL);
++ else
++ dentry = dget(path.dentry);
++
++out_path:
++ path_put(&path);
++out_relock:
++ if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0))
++ if (!IS_ERR(dentry)) {
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++ }
++out_pathname:
++ free_page((unsigned long)pathname);
++out_h_parent:
++ dput(h_parent);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+/* common functions to regular file and dir */
-+struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
-+ struct file *file, int force_wr)
++/* ---------------------------------------------------------------------- */
++
++static struct dentry *
++aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len,
++ int fh_type)
+{
-+ struct file *h_file;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ struct super_block *sb;
++ struct dentry *dentry;
++ __u32 *fh = fid->raw;
+ struct au_branch *br;
-+ struct path h_path;
-+ int err;
++ ino_t ino, dir_ino;
++ struct au_nfsd_si_lock nsi_lock = {
++ .force_lock = 0
++ };
+
-+ /* a race condition can happen between open and unlink/rmdir */
-+ h_file = ERR_PTR(-ENOENT);
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (au_test_nfsd() && (!h_dentry || d_is_negative(h_dentry)))
-+ goto out;
-+ h_inode = d_inode(h_dentry);
-+ spin_lock(&h_dentry->d_lock);
-+ err = (!d_unhashed(dentry) && d_unlinked(h_dentry))
-+ /* || !d_inode(dentry)->i_nlink */
-+ ;
-+ spin_unlock(&h_dentry->d_lock);
-+ if (unlikely(err))
++ dentry = ERR_PTR(-ESTALE);
++ /* it should never happen, but the file handle is unreliable */
++ if (unlikely(fh_len < Fh_tail))
+ goto out;
++ nsi_lock.sigen = fh[Fh_sigen];
++ nsi_lock.br_id = fh[Fh_br_id];
+
-+ sb = dentry->d_sb;
-+ br = au_sbr(sb, bindex);
-+ err = au_br_test_oflag(flags, br);
-+ h_file = ERR_PTR(err);
-+ if (unlikely(err))
++ /* branch id may be wrapped around */
++ br = NULL;
++ if (unlikely(si_nfsd_read_lock(sb, &nsi_lock)))
+ goto out;
++ nsi_lock.force_lock = 1;
+
-+ /* drop flags for writing */
-+ if (au_test_ro(sb, bindex, d_inode(dentry))) {
-+ if (force_wr && !(flags & O_WRONLY))
-+ force_wr = 0;
-+ flags = au_file_roflags(flags);
-+ if (force_wr) {
-+ h_file = ERR_PTR(-EROFS);
-+ flags = au_file_roflags(flags);
-+ if (unlikely(vfsub_native_ro(h_inode)
-+ || IS_APPEND(h_inode)))
-+ goto out;
-+ flags &= ~O_ACCMODE;
-+ flags |= O_WRONLY;
-+ }
-+ }
-+ flags &= ~O_CREAT;
++ /* is this inode still cached? */
++ ino = decode_ino(fh + Fh_ino);
++ /* it should never happen */
++ if (unlikely(ino == AUFS_ROOT_INO))
++ goto out_unlock;
++
++ dir_ino = decode_ino(fh + Fh_dir_ino);
++ dentry = decode_by_ino(sb, ino, dir_ino);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (dentry)
++ goto accept;
++
++ /* is the parent dir cached? */
++ br = au_sbr(sb, nsi_lock.bindex);
+ au_br_get(br);
-+ h_path.dentry = h_dentry;
-+ h_path.mnt = au_br_mnt(br);
-+ h_file = vfsub_dentry_open(&h_path, flags);
-+ if (IS_ERR(h_file))
-+ goto out_br;
++ dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (dentry)
++ goto accept;
+
-+ if (flags & __FMODE_EXEC) {
-+ err = deny_write_access(h_file);
-+ if (unlikely(err)) {
-+ fput(h_file);
-+ h_file = ERR_PTR(err);
-+ goto out_br;
-+ }
-+ }
-+ fsnotify_open(h_file);
-+ goto out; /* success */
++ /* lookup path */
++ dentry = decode_by_path(sb, ino, fh, fh_len, &nsi_lock);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (unlikely(!dentry))
++ /* todo?: make it ESTALE */
++ goto out_unlock;
++
++accept:
++ if (!au_digen_test(dentry, au_sigen(sb))
++ && d_inode(dentry)->i_generation == fh[Fh_igen])
++ goto out_unlock; /* success */
++
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++out_unlock:
++ if (br)
++ au_br_put(br);
++ si_read_unlock(sb);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
++}
++
++#if 0 /* reserved for future use */
++/* support subtreecheck option */
++static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid,
++ int fh_len, int fh_type)
++{
++ struct dentry *parent;
++ __u32 *fh = fid->raw;
++ ino_t dir_ino;
++
++ dir_ino = decode_ino(fh + Fh_dir_ino);
++ parent = decode_by_ino(sb, dir_ino, 0);
++ if (IS_ERR(parent))
++ goto out;
++ if (!parent)
++ parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]),
++ dir_ino, fh, fh_len);
+
-+out_br:
-+ au_br_put(br);
+out:
-+ return h_file;
++ AuTraceErrPtr(parent);
++ return parent;
+}
++#endif
+
-+static int au_cmoo(struct dentry *dentry)
++/* ---------------------------------------------------------------------- */
++
++static int aufs_encode_fh(struct inode *inode, __u32 *fh, int *max_len,
++ struct inode *dir)
+{
-+ int err, cmoo;
-+ unsigned int udba;
-+ struct path h_path;
-+ struct au_pin pin;
-+ struct au_cp_generic cpg = {
-+ .dentry = dentry,
-+ .bdst = -1,
-+ .bsrc = -1,
-+ .len = -1,
-+ .pin = &pin,
-+ .flags = AuCpup_DTIME | AuCpup_HOPEN
-+ };
-+ struct inode *delegated;
-+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
-+ struct au_fhsm *fhsm;
-+ pid_t pid;
++ int err;
++ aufs_bindex_t bindex;
++ struct super_block *sb, *h_sb;
++ struct dentry *dentry, *parent, *h_parent;
++ struct inode *h_dir;
+ struct au_branch *br;
-+ struct dentry *parent;
-+ struct au_hinode *hdir;
-+
-+ DiMustWriteLock(dentry);
-+ IiMustWriteLock(d_inode(dentry));
+
-+ err = 0;
-+ if (IS_ROOT(dentry))
-+ goto out;
-+ cpg.bsrc = au_dbtop(dentry);
-+ if (!cpg.bsrc)
++ err = -ENOSPC;
++ if (unlikely(*max_len <= Fh_tail)) {
++ AuWarn1("NFSv2 client (max_len %d)?\n", *max_len);
+ goto out;
++ }
+
-+ sb = dentry->d_sb;
-+ sbinfo = au_sbi(sb);
-+ fhsm = &sbinfo->si_fhsm;
-+ pid = au_fhsm_pid(fhsm);
-+ if (pid
-+ && (current->pid == pid
-+ || current->real_parent->pid == pid))
++ err = FILEID_ROOT;
++ if (inode->i_ino == AUFS_ROOT_INO) {
++ AuDebugOn(inode->i_ino != AUFS_ROOT_INO);
+ goto out;
++ }
+
-+ br = au_sbr(sb, cpg.bsrc);
-+ cmoo = au_br_cmoo(br->br_perm);
-+ if (!cmoo)
-+ goto out;
-+ if (!d_is_reg(dentry))
-+ cmoo &= AuBrAttr_COO_ALL;
-+ if (!cmoo)
++ h_parent = NULL;
++ sb = inode->i_sb;
++ err = si_read_lock(sb, AuLock_FLUSH);
++ if (unlikely(err))
+ goto out;
+
-+ parent = dget_parent(dentry);
-+ di_write_lock_parent(parent);
-+ err = au_wbr_do_copyup_bu(dentry, cpg.bsrc - 1);
-+ cpg.bdst = err;
-+ if (unlikely(err < 0)) {
-+ err = 0; /* there is no upper writable branch */
-+ goto out_dgrade;
++#ifdef CONFIG_AUFS_DEBUG
++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
++ AuWarn1("NFS-exporting requires xino\n");
++#endif
++ err = -EIO;
++ parent = NULL;
++ ii_read_lock_child(inode);
++ bindex = au_ibtop(inode);
++ if (!dir) {
++ dentry = d_find_any_alias(inode);
++ if (unlikely(!dentry))
++ goto out_unlock;
++ AuDebugOn(au_test_anon(dentry));
++ parent = dget_parent(dentry);
++ dput(dentry);
++ if (unlikely(!parent))
++ goto out_unlock;
++ if (d_really_is_positive(parent))
++ dir = d_inode(parent);
+ }
-+ AuDbg("bsrc %d, bdst %d\n", cpg.bsrc, cpg.bdst);
-+
-+ /* do not respect the coo attrib for the target branch */
-+ err = au_cpup_dirs(dentry, cpg.bdst);
-+ if (unlikely(err))
-+ goto out_dgrade;
+
-+ di_downgrade_lock(parent, AuLock_IR);
-+ udba = au_opt_udba(sb);
-+ err = au_pin(&pin, dentry, cpg.bdst, udba,
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ if (unlikely(err))
++ ii_read_lock_parent(dir);
++ h_dir = au_h_iptr(dir, bindex);
++ ii_read_unlock(dir);
++ if (unlikely(!h_dir))
+ goto out_parent;
++ h_parent = d_find_any_alias(h_dir);
++ if (unlikely(!h_parent))
++ goto out_hparent;
+
-+ err = au_sio_cpup_simple(&cpg);
-+ au_unpin(&pin);
-+ if (unlikely(err))
-+ goto out_parent;
-+ if (!(cmoo & AuBrWAttr_MOO))
-+ goto out_parent; /* success */
++ err = -EPERM;
++ br = au_sbr(sb, bindex);
++ h_sb = au_br_sb(br);
++ if (unlikely(!h_sb->s_export_op)) {
++ AuErr1("%s branch is not exportable\n", au_sbtype(h_sb));
++ goto out_hparent;
++ }
+
-+ err = au_pin(&pin, dentry, cpg.bsrc, udba,
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ if (unlikely(err))
-+ goto out_parent;
++ fh[Fh_br_id] = br->br_id;
++ fh[Fh_sigen] = au_sigen(sb);
++ encode_ino(fh + Fh_ino, inode->i_ino);
++ encode_ino(fh + Fh_dir_ino, dir->i_ino);
++ fh[Fh_igen] = inode->i_generation;
+
-+ h_path.mnt = au_br_mnt(br);
-+ h_path.dentry = au_h_dptr(dentry, cpg.bsrc);
-+ hdir = au_hi(d_inode(parent), cpg.bsrc);
-+ delegated = NULL;
-+ err = vfsub_unlink(hdir->hi_inode, &h_path, &delegated, /*force*/1);
-+ au_unpin(&pin);
-+ /* todo: keep h_dentry or not? */
-+ if (unlikely(err == -EWOULDBLOCK)) {
-+ pr_warn("cannot retry for NFSv4 delegation"
-+ " for an internal unlink\n");
-+ iput(delegated);
-+ }
-+ if (unlikely(err)) {
-+ pr_err("unlink %pd after coo failed (%d), ignored\n",
-+ dentry, err);
-+ err = 0;
-+ }
-+ goto out_parent; /* success */
++ *max_len -= Fh_tail;
++ fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail),
++ max_len,
++ /*connectable or subtreecheck*/0);
++ err = fh[Fh_h_type];
++ *max_len += Fh_tail;
++ /* todo: macros? */
++ if (err != FILEID_INVALID)
++ err = 99;
++ else
++ AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb));
+
-+out_dgrade:
-+ di_downgrade_lock(parent, AuLock_IR);
++out_hparent:
++ dput(h_parent);
+out_parent:
-+ di_read_unlock(parent, AuLock_IR);
+ dput(parent);
++out_unlock:
++ ii_read_unlock(inode);
++ si_read_unlock(sb);
+out:
-+ AuTraceErr(err);
++ if (unlikely(err < 0))
++ err = FILEID_INVALID;
+ return err;
+}
+
-+int au_do_open(struct file *file, struct au_do_open_args *args)
-+{
-+ int err, no_lock = args->no_lock;
-+ struct dentry *dentry;
-+ struct au_finfo *finfo;
++/* ---------------------------------------------------------------------- */
+
-+ if (!no_lock)
-+ err = au_finfo_init(file, args->fidir);
-+ else {
-+ lockdep_off();
-+ err = au_finfo_init(file, args->fidir);
-+ lockdep_on();
-+ }
-+ if (unlikely(err))
-+ goto out;
++static int aufs_commit_metadata(struct inode *inode)
++{
++ int err;
++ aufs_bindex_t bindex;
++ struct super_block *sb;
++ struct inode *h_inode;
++ int (*f)(struct inode *inode);
+
-+ dentry = file->f_path.dentry;
-+ AuDebugOn(IS_ERR_OR_NULL(dentry));
-+ if (!no_lock) {
-+ di_write_lock_child(dentry);
-+ err = au_cmoo(dentry);
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ if (!err)
-+ err = args->open(file, vfsub_file_flags(file), NULL);
-+ di_read_unlock(dentry, AuLock_IR);
-+ } else {
-+ err = au_cmoo(dentry);
-+ if (!err)
-+ err = args->open(file, vfsub_file_flags(file),
-+ args->h_file);
-+ if (!err && au_fbtop(file) != au_dbtop(dentry))
-+ /*
-+ * cmoo happens after h_file was opened.
-+ * need to refresh file later.
-+ */
-+ atomic_dec(&au_fi(file)->fi_generation);
-+ }
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ ii_write_lock_child(inode);
++ bindex = au_ibtop(inode);
++ AuDebugOn(bindex < 0);
++ h_inode = au_h_iptr(inode, bindex);
+
-+ finfo = au_fi(file);
-+ if (!err) {
-+ finfo->fi_file = file;
-+ au_sphl_add(&finfo->fi_hlist,
-+ &au_sbi(file->f_path.dentry->d_sb)->si_files);
-+ }
-+ if (!no_lock)
-+ fi_write_unlock(file);
++ f = h_inode->i_sb->s_export_op->commit_metadata;
++ if (f)
++ err = f(h_inode);
+ else {
-+ lockdep_off();
-+ fi_write_unlock(file);
-+ lockdep_on();
-+ }
-+ if (unlikely(err)) {
-+ finfo->fi_hdir = NULL;
-+ au_finfo_fin(file, /*atonce*/0);
++ struct writeback_control wbc = {
++ .sync_mode = WB_SYNC_ALL,
++ .nr_to_write = 0 /* metadata only */
++ };
++
++ err = sync_inode(h_inode, &wbc);
+ }
+
-+out:
++ au_cpup_attr_timesizes(inode);
++ ii_write_unlock(inode);
++ si_read_unlock(sb);
+ return err;
+}
+
-+int au_reopen_nondir(struct file *file)
-+{
-+ int err;
-+ aufs_bindex_t btop;
-+ struct dentry *dentry;
-+ struct file *h_file, *h_file_tmp;
++/* ---------------------------------------------------------------------- */
+
-+ dentry = file->f_path.dentry;
-+ btop = au_dbtop(dentry);
-+ h_file_tmp = NULL;
-+ if (au_fbtop(file) == btop) {
-+ h_file = au_hf_top(file);
-+ if (file->f_mode == h_file->f_mode)
-+ return 0; /* success */
-+ h_file_tmp = h_file;
-+ get_file(h_file_tmp);
-+ au_set_h_fptr(file, btop, NULL);
-+ }
-+ AuDebugOn(au_fi(file)->fi_hdir);
-+ /*
-+ * it can happen
-+ * file exists on both of rw and ro
-+ * open --> dbtop and fbtop are both 0
-+ * prepend a branch as rw, "rw" become ro
-+ * remove rw/file
-+ * delete the top branch, "rw" becomes rw again
-+ * --> dbtop is 1, fbtop is still 0
-+ * write --> fbtop is 0 but dbtop is 1
-+ */
-+ /* AuDebugOn(au_fbtop(file) < btop); */
++static struct export_operations aufs_export_op = {
++ .fh_to_dentry = aufs_fh_to_dentry,
++ /* .fh_to_parent = aufs_fh_to_parent, */
++ .encode_fh = aufs_encode_fh,
++ .commit_metadata = aufs_commit_metadata
++};
+
-+ h_file = au_h_open(dentry, btop, vfsub_file_flags(file) & ~O_TRUNC,
-+ file, /*force_wr*/0);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file)) {
-+ if (h_file_tmp) {
-+ au_sbr_get(dentry->d_sb, btop);
-+ au_set_h_fptr(file, btop, h_file_tmp);
-+ h_file_tmp = NULL;
-+ }
-+ goto out; /* todo: close all? */
-+ }
++void au_export_init(struct super_block *sb)
++{
++ struct au_sbinfo *sbinfo;
++ __u32 u;
+
-+ err = 0;
-+ au_set_fbtop(file, btop);
-+ au_set_h_fptr(file, btop, h_file);
-+ au_update_figen(file);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
++ BUILD_BUG_ON_MSG(IS_BUILTIN(CONFIG_AUFS_FS)
++ && IS_MODULE(CONFIG_EXPORTFS),
++ AUFS_NAME ": unsupported configuration "
++ "CONFIG_EXPORTFS=m and CONFIG_AUFS_FS=y");
+
-+out:
-+ if (h_file_tmp)
-+ fput(h_file_tmp);
-+ return err;
++ sb->s_export_op = &aufs_export_op;
++ sbinfo = au_sbi(sb);
++ sbinfo->si_xigen = NULL;
++ get_random_bytes(&u, sizeof(u));
++ BUILD_BUG_ON(sizeof(u) != sizeof(int));
++ atomic_set(&sbinfo->si_xigen_next, u);
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/f_op.c linux-4.14/fs/aufs/f_op.c
+--- linux-4.14/fs/aufs/f_op.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/f_op.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,817 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * file and vm operations
++ */
+
-+static int au_reopen_wh(struct file *file, aufs_bindex_t btgt,
-+ struct dentry *hi_wh)
++#include <linux/aio.h>
++#include <linux/fs_stack.h>
++#include <linux/mman.h>
++#include <linux/security.h>
++#include "aufs.h"
++
++int au_do_open_nondir(struct file *file, int flags, struct file *h_file)
+{
+ int err;
-+ aufs_bindex_t btop;
-+ struct au_dinfo *dinfo;
-+ struct dentry *h_dentry;
-+ struct au_hdentry *hdp;
++ aufs_bindex_t bindex;
++ struct dentry *dentry, *h_dentry;
++ struct au_finfo *finfo;
++ struct inode *h_inode;
+
-+ dinfo = au_di(file->f_path.dentry);
-+ AuRwMustWriteLock(&dinfo->di_rwsem);
++ FiMustWriteLock(file);
+
-+ btop = dinfo->di_btop;
-+ dinfo->di_btop = btgt;
-+ hdp = au_hdentry(dinfo, btgt);
-+ h_dentry = hdp->hd_dentry;
-+ hdp->hd_dentry = hi_wh;
-+ err = au_reopen_nondir(file);
-+ hdp->hd_dentry = h_dentry;
-+ dinfo->di_btop = btop;
++ err = 0;
++ dentry = file->f_path.dentry;
++ AuDebugOn(IS_ERR_OR_NULL(dentry));
++ finfo = au_fi(file);
++ memset(&finfo->fi_htop, 0, sizeof(finfo->fi_htop));
++ atomic_set(&finfo->fi_mmapped, 0);
++ bindex = au_dbtop(dentry);
++ if (!h_file) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
++ if (unlikely(err))
++ goto out;
++ h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0);
++ } else {
++ h_dentry = h_file->f_path.dentry;
++ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
++ if (unlikely(err))
++ goto out;
++ get_file(h_file);
++ }
++ if (IS_ERR(h_file))
++ err = PTR_ERR(h_file);
++ else {
++ if ((flags & __O_TMPFILE)
++ && !(flags & O_EXCL)) {
++ h_inode = file_inode(h_file);
++ spin_lock(&h_inode->i_lock);
++ h_inode->i_state |= I_LINKABLE;
++ spin_unlock(&h_inode->i_lock);
++ }
++ au_set_fbtop(file, bindex);
++ au_set_h_fptr(file, bindex, h_file);
++ au_update_figen(file);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++ }
+
++out:
+ return err;
+}
+
-+static int au_ready_to_write_wh(struct file *file, loff_t len,
-+ aufs_bindex_t bcpup, struct au_pin *pin)
++static int aufs_open_nondir(struct inode *inode __maybe_unused,
++ struct file *file)
+{
+ int err;
-+ struct inode *inode, *h_inode;
-+ struct dentry *h_dentry, *hi_wh;
-+ struct au_cp_generic cpg = {
-+ .dentry = file->f_path.dentry,
-+ .bdst = bcpup,
-+ .bsrc = -1,
-+ .len = len,
-+ .pin = pin
++ struct super_block *sb;
++ struct au_do_open_args args = {
++ .open = au_do_open_nondir
+ };
+
-+ au_update_dbtop(cpg.dentry);
-+ inode = d_inode(cpg.dentry);
-+ h_inode = NULL;
-+ if (au_dbtop(cpg.dentry) <= bcpup
-+ && au_dbbot(cpg.dentry) >= bcpup) {
-+ h_dentry = au_h_dptr(cpg.dentry, bcpup);
-+ if (h_dentry && d_is_positive(h_dentry))
-+ h_inode = d_inode(h_dentry);
-+ }
-+ hi_wh = au_hi_wh(inode, bcpup);
-+ if (!hi_wh && !h_inode)
-+ err = au_sio_cpup_wh(&cpg, file);
-+ else
-+ /* already copied-up after unlink */
-+ err = au_reopen_wh(file, bcpup, hi_wh);
-+
-+ if (!err
-+ && (inode->i_nlink > 1
-+ || (inode->i_state & I_LINKABLE))
-+ && au_opt_test(au_mntflags(cpg.dentry->d_sb), PLINK))
-+ au_plink_append(inode, bcpup, au_h_dptr(cpg.dentry, bcpup));
++ AuDbg("%pD, f_flags 0x%x, f_mode 0x%x\n",
++ file, vfsub_file_flags(file), file->f_mode);
+
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ err = au_do_open(file, &args);
++ si_read_unlock(sb);
+ return err;
+}
+
-+/*
-+ * prepare the @file for writing.
-+ */
-+int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin)
++int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file)
+{
-+ int err;
-+ aufs_bindex_t dbtop;
-+ struct dentry *parent;
-+ struct inode *inode;
-+ struct super_block *sb;
-+ struct file *h_file;
-+ struct au_cp_generic cpg = {
-+ .dentry = file->f_path.dentry,
-+ .bdst = -1,
-+ .bsrc = -1,
-+ .len = len,
-+ .pin = pin,
-+ .flags = AuCpup_DTIME
-+ };
-+
-+ sb = cpg.dentry->d_sb;
-+ inode = d_inode(cpg.dentry);
-+ cpg.bsrc = au_fbtop(file);
-+ err = au_test_ro(sb, cpg.bsrc, inode);
-+ if (!err && (au_hf_top(file)->f_mode & FMODE_WRITE)) {
-+ err = au_pin(pin, cpg.dentry, cpg.bsrc, AuOpt_UDBA_NONE,
-+ /*flags*/0);
-+ goto out;
-+ }
-+
-+ /* need to cpup or reopen */
-+ parent = dget_parent(cpg.dentry);
-+ di_write_lock_parent(parent);
-+ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
-+ cpg.bdst = err;
-+ if (unlikely(err < 0))
-+ goto out_dgrade;
-+ err = 0;
-+
-+ if (!d_unhashed(cpg.dentry) && !au_h_dptr(parent, cpg.bdst)) {
-+ err = au_cpup_dirs(cpg.dentry, cpg.bdst);
-+ if (unlikely(err))
-+ goto out_dgrade;
-+ }
-+
-+ err = au_pin(pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ if (unlikely(err))
-+ goto out_dgrade;
++ struct au_finfo *finfo;
++ aufs_bindex_t bindex;
+
-+ dbtop = au_dbtop(cpg.dentry);
-+ if (dbtop <= cpg.bdst)
-+ cpg.bsrc = cpg.bdst;
++ finfo = au_fi(file);
++ au_hbl_del(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
++ bindex = finfo->fi_btop;
++ if (bindex >= 0)
++ au_set_h_fptr(file, bindex, NULL);
+
-+ if (dbtop <= cpg.bdst /* just reopen */
-+ || !d_unhashed(cpg.dentry) /* copyup and reopen */
-+ ) {
-+ h_file = au_h_open_pre(cpg.dentry, cpg.bsrc, /*force_wr*/0);
-+ if (IS_ERR(h_file))
-+ err = PTR_ERR(h_file);
-+ else {
-+ di_downgrade_lock(parent, AuLock_IR);
-+ if (dbtop > cpg.bdst)
-+ err = au_sio_cpup_simple(&cpg);
-+ if (!err)
-+ err = au_reopen_nondir(file);
-+ au_h_open_post(cpg.dentry, cpg.bsrc, h_file);
-+ }
-+ } else { /* copyup as wh and reopen */
-+ /*
-+ * since writable hfsplus branch is not supported,
-+ * h_open_pre/post() are unnecessary.
-+ */
-+ err = au_ready_to_write_wh(file, len, cpg.bdst, pin);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ }
++ au_finfo_fin(file);
++ return 0;
++}
+
-+ if (!err) {
-+ au_pin_set_parent_lflag(pin, /*lflag*/0);
-+ goto out_dput; /* success */
-+ }
-+ au_unpin(pin);
-+ goto out_unlock;
++/* ---------------------------------------------------------------------- */
+
-+out_dgrade:
-+ di_downgrade_lock(parent, AuLock_IR);
-+out_unlock:
-+ di_read_unlock(parent, AuLock_IR);
-+out_dput:
-+ dput(parent);
-+out:
++static int au_do_flush_nondir(struct file *file, fl_owner_t id)
++{
++ int err;
++ struct file *h_file;
++
++ err = 0;
++ h_file = au_hf_top(file);
++ if (h_file)
++ err = vfsub_flush(h_file, id);
+ return err;
+}
+
++static int aufs_flush_nondir(struct file *file, fl_owner_t id)
++{
++ return au_do_flush(file, id, au_do_flush_nondir);
++}
++
+/* ---------------------------------------------------------------------- */
++/*
++ * read and write functions acquire [fdi]_rwsem once, but release before
++ * mmap_sem. This is because to stop a race condition between mmap(2).
++ * Releasing these aufs-rwsem should be safe, no branch-mamagement (by keeping
++ * si_rwsem), no harmful copy-up should happen. Actually copy-up may happen in
++ * read functions after [fdi]_rwsem are released, but it should be harmless.
++ */
+
-+int au_do_flush(struct file *file, fl_owner_t id,
-+ int (*flush)(struct file *file, fl_owner_t id))
++/* Callers should call au_read_post() or fput() in the end */
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc)
+{
++ struct file *h_file;
+ int err;
-+ struct super_block *sb;
-+ struct inode *inode;
+
-+ inode = file_inode(file);
-+ sb = inode->i_sb;
-+ si_noflush_read_lock(sb);
-+ fi_read_lock(file);
-+ ii_read_lock_child(inode);
++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0, lsc);
++ if (!err) {
++ di_read_unlock(file->f_path.dentry, AuLock_IR);
++ h_file = au_hf_top(file);
++ get_file(h_file);
++ if (!keep_fi)
++ fi_read_unlock(file);
++ } else
++ h_file = ERR_PTR(err);
+
-+ err = flush(file, id);
-+ au_cpup_attr_timesizes(inode);
++ return h_file;
++}
+
-+ ii_read_unlock(inode);
-+ fi_read_unlock(file);
-+ si_read_unlock(sb);
-+ return err;
++static void au_read_post(struct inode *inode, struct file *h_file)
++{
++ /* update without lock, I don't think it a problem */
++ fsstack_copy_attr_atime(inode, file_inode(h_file));
++ fput(h_file);
+}
+
-+/* ---------------------------------------------------------------------- */
++struct au_write_pre {
++ /* input */
++ unsigned int lsc;
+
-+static int au_file_refresh_by_inode(struct file *file, int *need_reopen)
++ /* output */
++ blkcnt_t blks;
++ aufs_bindex_t btop;
++};
++
++/*
++ * return with iinfo is write-locked
++ * callers should call au_write_post() or iinfo_write_unlock() + fput() in the
++ * end
++ */
++static struct file *au_write_pre(struct file *file, int do_ready,
++ struct au_write_pre *wpre)
+{
++ struct file *h_file;
++ struct dentry *dentry;
+ int err;
++ unsigned int lsc;
+ struct au_pin pin;
-+ struct au_finfo *finfo;
-+ struct dentry *parent, *hi_wh;
-+ struct inode *inode;
-+ struct super_block *sb;
-+ struct au_cp_generic cpg = {
-+ .dentry = file->f_path.dentry,
-+ .bdst = -1,
-+ .bsrc = -1,
-+ .len = -1,
-+ .pin = &pin,
-+ .flags = AuCpup_DTIME
-+ };
-+
-+ FiMustWriteLock(file);
+
-+ err = 0;
-+ finfo = au_fi(file);
-+ sb = cpg.dentry->d_sb;
-+ inode = d_inode(cpg.dentry);
-+ cpg.bdst = au_ibtop(inode);
-+ if (cpg.bdst == finfo->fi_btop || IS_ROOT(cpg.dentry))
++ lsc = 0;
++ if (wpre)
++ lsc = wpre->lsc;
++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1, lsc);
++ h_file = ERR_PTR(err);
++ if (unlikely(err))
+ goto out;
+
-+ parent = dget_parent(cpg.dentry);
-+ if (au_test_ro(sb, cpg.bdst, inode)) {
-+ di_read_lock_parent(parent, !AuLock_IR);
-+ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
-+ cpg.bdst = err;
-+ di_read_unlock(parent, !AuLock_IR);
-+ if (unlikely(err < 0))
-+ goto out_parent;
-+ err = 0;
-+ }
-+
-+ di_read_lock_parent(parent, AuLock_IR);
-+ hi_wh = au_hi_wh(inode, cpg.bdst);
-+ if (!S_ISDIR(inode->i_mode)
-+ && au_opt_test(au_mntflags(sb), PLINK)
-+ && au_plink_test(inode)
-+ && !d_unhashed(cpg.dentry)
-+ && cpg.bdst < au_dbtop(cpg.dentry)) {
-+ err = au_test_and_cpup_dirs(cpg.dentry, cpg.bdst);
-+ if (unlikely(err))
-+ goto out_unlock;
-+
-+ /* always superio. */
-+ err = au_pin(&pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ if (!err) {
-+ err = au_sio_cpup_simple(&cpg);
-+ au_unpin(&pin);
++ dentry = file->f_path.dentry;
++ if (do_ready) {
++ err = au_ready_to_write(file, -1, &pin);
++ if (unlikely(err)) {
++ h_file = ERR_PTR(err);
++ di_write_unlock(dentry);
++ goto out_fi;
+ }
-+ } else if (hi_wh) {
-+ /* already copied-up after unlink */
-+ err = au_reopen_wh(file, cpg.bdst, hi_wh);
-+ *need_reopen = 0;
+ }
+
-+out_unlock:
-+ di_read_unlock(parent, AuLock_IR);
-+out_parent:
-+ dput(parent);
++ di_downgrade_lock(dentry, /*flags*/0);
++ if (wpre)
++ wpre->btop = au_fbtop(file);
++ h_file = au_hf_top(file);
++ get_file(h_file);
++ if (wpre)
++ wpre->blks = file_inode(h_file)->i_blocks;
++ if (do_ready)
++ au_unpin(&pin);
++ di_read_unlock(dentry, /*flags*/0);
++
++out_fi:
++ fi_write_unlock(file);
+out:
-+ return err;
++ return h_file;
+}
+
-+static void au_do_refresh_dir(struct file *file)
++static void au_write_post(struct inode *inode, struct file *h_file,
++ struct au_write_pre *wpre, ssize_t written)
+{
-+ int execed;
-+ aufs_bindex_t bindex, bbot, new_bindex, brid;
-+ struct au_hfile *p, tmp, *q;
-+ struct au_finfo *finfo;
-+ struct super_block *sb;
-+ struct au_fidir *fidir;
++ struct inode *h_inode;
+
-+ FiMustWriteLock(file);
++ au_cpup_attr_timesizes(inode);
++ AuDebugOn(au_ibtop(inode) != wpre->btop);
++ h_inode = file_inode(h_file);
++ inode->i_mode = h_inode->i_mode;
++ ii_write_unlock(inode);
++ /* AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); */
++ if (written > 0)
++ au_fhsm_wrote(inode->i_sb, wpre->btop,
++ /*force*/h_inode->i_blocks > wpre->blks);
++ fput(h_file);
++}
+
-+ sb = file->f_path.dentry->d_sb;
-+ finfo = au_fi(file);
-+ fidir = finfo->fi_hdir;
-+ AuDebugOn(!fidir);
-+ p = fidir->fd_hfile + finfo->fi_btop;
-+ brid = p->hf_br->br_id;
-+ bbot = fidir->fd_bbot;
-+ for (bindex = finfo->fi_btop; bindex <= bbot; bindex++, p++) {
-+ if (!p->hf_file)
-+ continue;
++static ssize_t aufs_read(struct file *file, char __user *buf, size_t count,
++ loff_t *ppos)
++{
++ ssize_t err;
++ struct inode *inode;
++ struct file *h_file;
++ struct super_block *sb;
+
-+ new_bindex = au_br_index(sb, p->hf_br->br_id);
-+ if (new_bindex == bindex)
-+ continue;
-+ if (new_bindex < 0) {
-+ au_set_h_fptr(file, bindex, NULL);
-+ continue;
-+ }
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ /* swap two lower inode, and loop again */
-+ q = fidir->fd_hfile + new_bindex;
-+ tmp = *q;
-+ *q = *p;
-+ *p = tmp;
-+ if (tmp.hf_file) {
-+ bindex--;
-+ p--;
-+ }
-+ }
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+ execed = vfsub_file_execed(file);
-+ p = fidir->fd_hfile;
-+ if (!au_test_mmapped(file) && !d_unlinked(file->f_path.dentry)) {
-+ bbot = au_sbbot(sb);
-+ for (finfo->fi_btop = 0; finfo->fi_btop <= bbot;
-+ finfo->fi_btop++, p++)
-+ if (p->hf_file) {
-+ if (file_inode(p->hf_file))
-+ break;
-+ au_hfput(p, execed);
-+ }
-+ } else {
-+ bbot = au_br_index(sb, brid);
-+ for (finfo->fi_btop = 0; finfo->fi_btop < bbot;
-+ finfo->fi_btop++, p++)
-+ if (p->hf_file)
-+ au_hfput(p, execed);
-+ bbot = au_sbbot(sb);
-+ }
++ /* filedata may be obsoleted by concurrent copyup, but no problem */
++ err = vfsub_read_u(h_file, buf, count, ppos);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++ au_read_post(inode, h_file);
+
-+ p = fidir->fd_hfile + bbot;
-+ for (fidir->fd_bbot = bbot; fidir->fd_bbot >= finfo->fi_btop;
-+ fidir->fd_bbot--, p--)
-+ if (p->hf_file) {
-+ if (file_inode(p->hf_file))
-+ break;
-+ au_hfput(p, execed);
-+ }
-+ AuDebugOn(fidir->fd_bbot < finfo->fi_btop);
++out:
++ si_read_unlock(sb);
++ return err;
+}
+
+/*
-+ * after branch manipulating, refresh the file.
++ * todo: very ugly
++ * it locks both of i_mutex and si_rwsem for read in safe.
++ * if the plink maintenance mode continues forever (that is the problem),
++ * may loop forever.
+ */
-+static int refresh_file(struct file *file, int (*reopen)(struct file *file))
++static void au_mtx_and_read_lock(struct inode *inode)
+{
-+ int err, need_reopen;
-+ aufs_bindex_t bbot, bindex;
-+ struct dentry *dentry;
-+ struct au_finfo *finfo;
-+ struct au_hfile *hfile;
++ int err;
++ struct super_block *sb = inode->i_sb;
+
-+ dentry = file->f_path.dentry;
-+ finfo = au_fi(file);
-+ if (!finfo->fi_hdir) {
-+ hfile = &finfo->fi_htop;
-+ AuDebugOn(!hfile->hf_file);
-+ bindex = au_br_index(dentry->d_sb, hfile->hf_br->br_id);
-+ AuDebugOn(bindex < 0);
-+ if (bindex != finfo->fi_btop)
-+ au_set_fbtop(file, bindex);
-+ } else {
-+ err = au_fidir_realloc(finfo, au_sbbot(dentry->d_sb) + 1);
-+ if (unlikely(err))
-+ goto out;
-+ au_do_refresh_dir(file);
++ while (1) {
++ inode_lock(inode);
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (!err)
++ break;
++ inode_unlock(inode);
++ si_read_lock(sb, AuLock_NOPLMW);
++ si_read_unlock(sb);
+ }
++}
+
-+ err = 0;
-+ need_reopen = 1;
-+ if (!au_test_mmapped(file))
-+ err = au_file_refresh_by_inode(file, &need_reopen);
-+ if (!err && need_reopen && !d_unlinked(dentry))
-+ err = reopen(file);
-+ if (!err) {
-+ au_update_figen(file);
-+ goto out; /* success */
-+ }
++static ssize_t aufs_write(struct file *file, const char __user *ubuf,
++ size_t count, loff_t *ppos)
++{
++ ssize_t err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
++ char __user *buf = (char __user *)ubuf;
+
-+ /* error, close all lower files */
-+ if (finfo->fi_hdir) {
-+ bbot = au_fbbot_dir(file);
-+ for (bindex = au_fbtop(file); bindex <= bbot; bindex++)
-+ au_set_h_fptr(file, bindex, NULL);
-+ }
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
++
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++
++ err = vfsub_write_u(h_file, buf, count, ppos);
++ au_write_post(inode, h_file, &wpre, err);
+
+out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
+ return err;
+}
+
-+/* common function to regular file and dir */
-+int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
-+ int wlock)
++static ssize_t au_do_iter(struct file *h_file, int rw, struct kiocb *kio,
++ struct iov_iter *iov_iter)
+{
-+ int err;
-+ unsigned int sigen, figen;
-+ aufs_bindex_t btop;
-+ unsigned char pseudo_link;
-+ struct dentry *dentry;
-+ struct inode *inode;
++ ssize_t err;
++ struct file *file;
++ ssize_t (*iter)(struct kiocb *, struct iov_iter *);
+
-+ err = 0;
-+ dentry = file->f_path.dentry;
-+ inode = d_inode(dentry);
-+ sigen = au_sigen(dentry->d_sb);
-+ fi_write_lock(file);
-+ figen = au_figen(file);
-+ di_write_lock_child(dentry);
-+ btop = au_dbtop(dentry);
-+ pseudo_link = (btop != au_ibtop(inode));
-+ if (sigen == figen && !pseudo_link && au_fbtop(file) == btop) {
-+ if (!wlock) {
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ fi_downgrade_lock(file);
-+ }
-+ goto out; /* success */
-+ }
++ err = security_file_permission(h_file, rw);
++ if (unlikely(err))
++ goto out;
+
-+ AuDbg("sigen %d, figen %d\n", sigen, figen);
-+ if (au_digen_test(dentry, sigen)) {
-+ err = au_reval_dpath(dentry, sigen);
-+ AuDebugOn(!err && au_digen_test(dentry, sigen));
-+ }
++ err = -ENOSYS;
++ iter = NULL;
++ if (rw == MAY_READ)
++ iter = h_file->f_op->read_iter;
++ else if (rw == MAY_WRITE)
++ iter = h_file->f_op->write_iter;
+
-+ if (!err)
-+ err = refresh_file(file, reopen);
-+ if (!err) {
-+ if (!wlock) {
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ fi_downgrade_lock(file);
-+ }
-+ } else {
-+ di_write_unlock(dentry);
-+ fi_write_unlock(file);
-+ }
++ file = kio->ki_filp;
++ kio->ki_filp = h_file;
++ if (iter) {
++ lockdep_off();
++ err = iter(kio, iov_iter);
++ lockdep_on();
++ } else
++ /* currently there is no such fs */
++ WARN_ON_ONCE(1);
++ kio->ki_filp = file;
+
+out:
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/* cf. aufs_nopage() */
-+/* for madvise(2) */
-+static int aufs_readpage(struct file *file __maybe_unused, struct page *page)
++static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
+{
-+ unlock_page(page);
-+ return 0;
-+}
-+
-+/* it will never be called, but necessary to support O_DIRECT */
-+static ssize_t aufs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
-+{ BUG(); return 0; }
++ ssize_t err;
++ struct file *file, *h_file;
++ struct inode *inode;
++ struct super_block *sb;
+
-+/* they will never be called. */
-+#ifdef CONFIG_AUFS_DEBUG
-+static int aufs_write_begin(struct file *file, struct address_space *mapping,
-+ loff_t pos, unsigned len, unsigned flags,
-+ struct page **pagep, void **fsdata)
-+{ AuUnsupport(); return 0; }
-+static int aufs_write_end(struct file *file, struct address_space *mapping,
-+ loff_t pos, unsigned len, unsigned copied,
-+ struct page *page, void *fsdata)
-+{ AuUnsupport(); return 0; }
-+static int aufs_writepage(struct page *page, struct writeback_control *wbc)
-+{ AuUnsupport(); return 0; }
++ file = kio->ki_filp;
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+static int aufs_set_page_dirty(struct page *page)
-+{ AuUnsupport(); return 0; }
-+static void aufs_invalidatepage(struct page *page, unsigned int offset,
-+ unsigned int length)
-+{ AuUnsupport(); }
-+static int aufs_releasepage(struct page *page, gfp_t gfp)
-+{ AuUnsupport(); return 0; }
-+#if 0 /* called by memory compaction regardless file */
-+static int aufs_migratepage(struct address_space *mapping, struct page *newpage,
-+ struct page *page, enum migrate_mode mode)
-+{ AuUnsupport(); return 0; }
-+#endif
-+static int aufs_launder_page(struct page *page)
-+{ AuUnsupport(); return 0; }
-+static int aufs_is_partially_uptodate(struct page *page,
-+ unsigned long from,
-+ unsigned long count)
-+{ AuUnsupport(); return 0; }
-+static void aufs_is_dirty_writeback(struct page *page, bool *dirty,
-+ bool *writeback)
-+{ AuUnsupport(); }
-+static int aufs_error_remove_page(struct address_space *mapping,
-+ struct page *page)
-+{ AuUnsupport(); return 0; }
-+static int aufs_swap_activate(struct swap_info_struct *sis, struct file *file,
-+ sector_t *span)
-+{ AuUnsupport(); return 0; }
-+static void aufs_swap_deactivate(struct file *file)
-+{ AuUnsupport(); }
-+#endif /* CONFIG_AUFS_DEBUG */
++ h_file = au_read_pre(file, /*keep_fi*/1, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+const struct address_space_operations aufs_aop = {
-+ .readpage = aufs_readpage,
-+ .direct_IO = aufs_direct_IO,
-+#ifdef CONFIG_AUFS_DEBUG
-+ .writepage = aufs_writepage,
-+ /* no writepages, because of writepage */
-+ .set_page_dirty = aufs_set_page_dirty,
-+ /* no readpages, because of readpage */
-+ .write_begin = aufs_write_begin,
-+ .write_end = aufs_write_end,
-+ /* no bmap, no block device */
-+ .invalidatepage = aufs_invalidatepage,
-+ .releasepage = aufs_releasepage,
-+ /* is fallback_migrate_page ok? */
-+ /* .migratepage = aufs_migratepage, */
-+ .launder_page = aufs_launder_page,
-+ .is_partially_uptodate = aufs_is_partially_uptodate,
-+ .is_dirty_writeback = aufs_is_dirty_writeback,
-+ .error_remove_page = aufs_error_remove_page,
-+ .swap_activate = aufs_swap_activate,
-+ .swap_deactivate = aufs_swap_deactivate
-+#endif /* CONFIG_AUFS_DEBUG */
-+};
-diff -urN /usr/share/empty/fs/aufs/file.h linux/fs/aufs/file.h
---- /usr/share/empty/fs/aufs/file.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/file.h 2016-08-17 18:01:06.135222590 +0200
-@@ -0,0 +1,294 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
++ if (0 && au_test_loopback_kthread()) {
++ au_warn_loopback(h_file->f_path.dentry->d_sb);
++ if (file->f_mapping != h_file->f_mapping) {
++ file->f_mapping = h_file->f_mapping;
++ smp_mb(); /* unnecessary? */
++ }
++ }
++ fi_read_unlock(file);
+
-+/*
-+ * file operations
-+ */
++ err = au_do_iter(h_file, MAY_READ, kio, iov_iter);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++ au_read_post(inode, h_file);
+
-+#ifndef __AUFS_FILE_H__
-+#define __AUFS_FILE_H__
++out:
++ si_read_unlock(sb);
++ return err;
++}
+
-+#ifdef __KERNEL__
++static ssize_t aufs_write_iter(struct kiocb *kio, struct iov_iter *iov_iter)
++{
++ ssize_t err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *file, *h_file;
+
-+#include <linux/file.h>
-+#include <linux/fs.h>
-+#include <linux/poll.h>
-+#include "rwsem.h"
++ file = kio->ki_filp;
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+struct au_branch;
-+struct au_hfile {
-+ struct file *hf_file;
-+ struct au_branch *hf_br;
-+};
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+struct au_vdir;
-+struct au_fidir {
-+ aufs_bindex_t fd_bbot;
-+ aufs_bindex_t fd_nent;
-+ struct au_vdir *fd_vdir_cache;
-+ struct au_hfile fd_hfile[];
-+};
++ err = au_do_iter(h_file, MAY_WRITE, kio, iov_iter);
++ au_write_post(inode, h_file, &wpre, err);
+
-+static inline int au_fidir_sz(int nent)
-+{
-+ AuDebugOn(nent < 0);
-+ return sizeof(struct au_fidir) + sizeof(struct au_hfile) * nent;
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++ return err;
+}
+
-+struct au_finfo {
-+ atomic_t fi_generation;
++static ssize_t aufs_splice_read(struct file *file, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags)
++{
++ ssize_t err;
++ struct file *h_file;
++ struct inode *inode;
++ struct super_block *sb;
+
-+ struct au_rwsem fi_rwsem;
-+ aufs_bindex_t fi_btop;
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ /* do not union them */
-+ struct { /* for non-dir */
-+ struct au_hfile fi_htop;
-+ atomic_t fi_mmapped;
-+ };
-+ struct au_fidir *fi_hdir; /* for dir only */
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+ struct hlist_node fi_hlist;
-+ union {
-+ struct file *fi_file; /* very ugly */
-+ struct llist_node fi_lnode; /* delayed free */
-+ };
-+} ____cacheline_aligned_in_smp;
++ err = vfsub_splice_to(h_file, ppos, pipe, len, flags);
++ /* todo: necessasry? */
++ /* file->f_ra = h_file->f_ra; */
++ au_read_post(inode, h_file);
+
-+/* ---------------------------------------------------------------------- */
++out:
++ si_read_unlock(sb);
++ return err;
++}
+
-+/* file.c */
-+extern const struct address_space_operations aufs_aop;
-+unsigned int au_file_roflags(unsigned int flags);
-+struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
-+ struct file *file, int force_wr);
-+struct au_do_open_args {
-+ int no_lock;
-+ int (*open)(struct file *file, int flags,
-+ struct file *h_file);
-+ struct au_fidir *fidir;
-+ struct file *h_file;
-+};
-+int au_do_open(struct file *file, struct au_do_open_args *args);
-+int au_reopen_nondir(struct file *file);
-+struct au_pin;
-+int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin);
-+int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
-+ int wlock);
-+int au_do_flush(struct file *file, fl_owner_t id,
-+ int (*flush)(struct file *file, fl_owner_t id));
++static ssize_t
++aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos,
++ size_t len, unsigned int flags)
++{
++ ssize_t err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
+
-+/* poll.c */
-+#ifdef CONFIG_AUFS_POLL
-+unsigned int aufs_poll(struct file *file, poll_table *wait);
-+#endif
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+#ifdef CONFIG_AUFS_BR_HFSPLUS
-+/* hfsplus.c */
-+struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
-+ int force_wr);
-+void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *h_file);
-+#else
-+AuStub(struct file *, au_h_open_pre, return NULL, struct dentry *dentry,
-+ aufs_bindex_t bindex, int force_wr)
-+AuStubVoid(au_h_open_post, struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *h_file);
-+#endif
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+/* f_op.c */
-+extern const struct file_operations aufs_file_fop;
-+int au_do_open_nondir(struct file *file, int flags, struct file *h_file);
-+int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file);
-+struct file *au_read_pre(struct file *file, int keep_fi);
++ err = vfsub_splice_from(pipe, h_file, ppos, len, flags);
++ au_write_post(inode, h_file, &wpre, err);
+
-+/* finfo.c */
-+void au_hfput(struct au_hfile *hf, int execed);
-+void au_set_h_fptr(struct file *file, aufs_bindex_t bindex,
-+ struct file *h_file);
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++ return err;
++}
+
-+void au_update_figen(struct file *file);
-+struct au_fidir *au_fidir_alloc(struct super_block *sb);
-+int au_fidir_realloc(struct au_finfo *finfo, int nbr);
++static long aufs_fallocate(struct file *file, int mode, loff_t offset,
++ loff_t len)
++{
++ long err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
+
-+void au_fi_init_once(void *_fi);
-+void au_finfo_fin(struct file *file, int atonce);
-+int au_finfo_init(struct file *file, struct au_fidir *fidir);
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+/* ioctl.c */
-+long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg);
-+#ifdef CONFIG_COMPAT
-+long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
-+ unsigned long arg);
-+long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
-+ unsigned long arg);
-+#endif
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ lockdep_off();
++ err = vfs_fallocate(h_file, mode, offset, len);
++ lockdep_on();
++ au_write_post(inode, h_file, &wpre, /*written*/1);
+
-+static inline struct au_finfo *au_fi(struct file *file)
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++ return err;
++}
++
++static ssize_t aufs_copy_file_range(struct file *src, loff_t src_pos,
++ struct file *dst, loff_t dst_pos,
++ size_t len, unsigned int flags)
+{
-+ return file->private_data;
++ ssize_t err;
++ struct au_write_pre wpre;
++ enum { SRC, DST };
++ struct {
++ struct inode *inode;
++ struct file *h_file;
++ struct super_block *h_sb;
++ } a[2];
++#define a_src a[SRC]
++#define a_dst a[DST]
++
++ err = -EINVAL;
++ a_src.inode = file_inode(src);
++ if (unlikely(!S_ISREG(a_src.inode->i_mode)))
++ goto out;
++ a_dst.inode = file_inode(dst);
++ if (unlikely(!S_ISREG(a_dst.inode->i_mode)))
++ goto out;
++
++ au_mtx_and_read_lock(a_dst.inode);
++ /*
++ * in order to match the order in di_write_lock2_{child,parent}(),
++ * use f_path.dentry for this comparision.
++ */
++ if (src->f_path.dentry < dst->f_path.dentry) {
++ a_src.h_file = au_read_pre(src, /*keep_fi*/1, AuLsc_FI_1);
++ err = PTR_ERR(a_src.h_file);
++ if (IS_ERR(a_src.h_file))
++ goto out_si;
++
++ wpre.lsc = AuLsc_FI_2;
++ a_dst.h_file = au_write_pre(dst, /*do_ready*/1, &wpre);
++ err = PTR_ERR(a_dst.h_file);
++ if (IS_ERR(a_dst.h_file)) {
++ au_read_post(a_src.inode, a_src.h_file);
++ goto out_si;
++ }
++ } else {
++ wpre.lsc = AuLsc_FI_1;
++ a_dst.h_file = au_write_pre(dst, /*do_ready*/1, &wpre);
++ err = PTR_ERR(a_dst.h_file);
++ if (IS_ERR(a_dst.h_file))
++ goto out_si;
++
++ a_src.h_file = au_read_pre(src, /*keep_fi*/1, AuLsc_FI_2);
++ err = PTR_ERR(a_src.h_file);
++ if (IS_ERR(a_src.h_file)) {
++ au_write_post(a_dst.inode, a_dst.h_file, &wpre,
++ /*written*/0);
++ goto out_si;
++ }
++ }
++
++ err = -EXDEV;
++ a_src.h_sb = file_inode(a_src.h_file)->i_sb;
++ a_dst.h_sb = file_inode(a_dst.h_file)->i_sb;
++ if (unlikely(a_src.h_sb != a_dst.h_sb)) {
++ AuDbgFile(src);
++ AuDbgFile(dst);
++ goto out_file;
++ }
++
++ err = vfsub_copy_file_range(a_src.h_file, src_pos, a_dst.h_file,
++ dst_pos, len, flags);
++
++out_file:
++ au_write_post(a_dst.inode, a_dst.h_file, &wpre, err);
++ fi_read_unlock(src);
++ au_read_post(a_src.inode, a_src.h_file);
++out_si:
++ si_read_unlock(a_dst.inode->i_sb);
++ inode_unlock(a_dst.inode);
++out:
++ return err;
++#undef a_src
++#undef a_dst
+}
+
+/* ---------------------------------------------------------------------- */
+
+/*
-+ * fi_read_lock, fi_write_lock,
-+ * fi_read_unlock, fi_write_unlock, fi_downgrade_lock
++ * The locking order around current->mmap_sem.
++ * - in most and regular cases
++ * file I/O syscall -- aufs_read() or something
++ * -- si_rwsem for read -- mmap_sem
++ * (Note that [fdi]i_rwsem are released before mmap_sem).
++ * - in mmap case
++ * mmap(2) -- mmap_sem -- aufs_mmap() -- si_rwsem for read -- [fdi]i_rwsem
++ * This AB-BA order is definitly bad, but is not a problem since "si_rwsem for
++ * read" allows muliple processes to acquire it and [fdi]i_rwsem are not held in
++ * file I/O. Aufs needs to stop lockdep in aufs_mmap() though.
++ * It means that when aufs acquires si_rwsem for write, the process should never
++ * acquire mmap_sem.
++ *
++ * Actually aufs_iterate() holds [fdi]i_rwsem before mmap_sem, but this is not a
++ * problem either since any directory is not able to be mmap-ed.
++ * The similar scenario is applied to aufs_readlink() too.
+ */
-+AuSimpleRwsemFuncs(fi, struct file *f, &au_fi(f)->fi_rwsem);
+
-+#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem)
-+#define FiMustAnyLock(f) AuRwMustAnyLock(&au_fi(f)->fi_rwsem)
-+#define FiMustWriteLock(f) AuRwMustWriteLock(&au_fi(f)->fi_rwsem)
-+
-+/* ---------------------------------------------------------------------- */
++#if 0 /* stop calling security_file_mmap() */
++/* cf. linux/include/linux/mman.h: calc_vm_prot_bits() */
++#define AuConv_VM_PROT(f, b) _calc_vm_trans(f, VM_##b, PROT_##b)
+
-+/* todo: hard/soft set? */
-+static inline aufs_bindex_t au_fbtop(struct file *file)
++static unsigned long au_arch_prot_conv(unsigned long flags)
+{
-+ FiMustAnyLock(file);
-+ return au_fi(file)->fi_btop;
++ /* currently ppc64 only */
++#ifdef CONFIG_PPC64
++ /* cf. linux/arch/powerpc/include/asm/mman.h */
++ AuDebugOn(arch_calc_vm_prot_bits(-1) != VM_SAO);
++ return AuConv_VM_PROT(flags, SAO);
++#else
++ AuDebugOn(arch_calc_vm_prot_bits(-1));
++ return 0;
++#endif
+}
+
-+static inline aufs_bindex_t au_fbbot_dir(struct file *file)
++static unsigned long au_prot_conv(unsigned long flags)
+{
-+ FiMustAnyLock(file);
-+ AuDebugOn(!au_fi(file)->fi_hdir);
-+ return au_fi(file)->fi_hdir->fd_bbot;
++ return AuConv_VM_PROT(flags, READ)
++ | AuConv_VM_PROT(flags, WRITE)
++ | AuConv_VM_PROT(flags, EXEC)
++ | au_arch_prot_conv(flags);
+}
+
-+static inline struct au_vdir *au_fvdir_cache(struct file *file)
-+{
-+ FiMustAnyLock(file);
-+ AuDebugOn(!au_fi(file)->fi_hdir);
-+ return au_fi(file)->fi_hdir->fd_vdir_cache;
-+}
++/* cf. linux/include/linux/mman.h: calc_vm_flag_bits() */
++#define AuConv_VM_MAP(f, b) _calc_vm_trans(f, VM_##b, MAP_##b)
+
-+static inline void au_set_fbtop(struct file *file, aufs_bindex_t bindex)
++static unsigned long au_flag_conv(unsigned long flags)
+{
-+ FiMustWriteLock(file);
-+ au_fi(file)->fi_btop = bindex;
++ return AuConv_VM_MAP(flags, GROWSDOWN)
++ | AuConv_VM_MAP(flags, DENYWRITE)
++ | AuConv_VM_MAP(flags, LOCKED);
+}
++#endif
+
-+static inline void au_set_fbbot_dir(struct file *file, aufs_bindex_t bindex)
++static int aufs_mmap(struct file *file, struct vm_area_struct *vma)
+{
-+ FiMustWriteLock(file);
-+ AuDebugOn(!au_fi(file)->fi_hdir);
-+ au_fi(file)->fi_hdir->fd_bbot = bindex;
-+}
++ int err;
++ const unsigned char wlock
++ = (file->f_mode & FMODE_WRITE) && (vma->vm_flags & VM_SHARED);
++ struct super_block *sb;
++ struct file *h_file;
++ struct inode *inode;
+
-+static inline void au_set_fvdir_cache(struct file *file,
-+ struct au_vdir *vdir_cache)
-+{
-+ FiMustWriteLock(file);
-+ AuDebugOn(!au_fi(file)->fi_hdir);
-+ au_fi(file)->fi_hdir->fd_vdir_cache = vdir_cache;
-+}
++ AuDbgVmRegion(file, vma);
+
-+static inline struct file *au_hf_top(struct file *file)
-+{
-+ FiMustAnyLock(file);
-+ AuDebugOn(au_fi(file)->fi_hdir);
-+ return au_fi(file)->fi_htop.hf_file;
-+}
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ lockdep_off();
++ si_read_lock(sb, AuLock_NOPLMW);
+
-+static inline struct file *au_hf_dir(struct file *file, aufs_bindex_t bindex)
-+{
-+ FiMustAnyLock(file);
-+ AuDebugOn(!au_fi(file)->fi_hdir);
-+ return au_fi(file)->fi_hdir->fd_hfile[0 + bindex].hf_file;
-+}
++ h_file = au_write_pre(file, wlock, /*wpre*/NULL);
++ lockdep_on();
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+/* todo: memory barrier? */
-+static inline unsigned int au_figen(struct file *f)
-+{
-+ return atomic_read(&au_fi(f)->fi_generation);
++ err = 0;
++ au_set_mmapped(file);
++ au_vm_file_reset(vma, h_file);
++ /*
++ * we cannot call security_mmap_file() here since it may acquire
++ * mmap_sem or i_mutex.
++ *
++ * err = security_mmap_file(h_file, au_prot_conv(vma->vm_flags),
++ * au_flag_conv(vma->vm_flags));
++ */
++ if (!err)
++ err = call_mmap(h_file, vma);
++ if (!err) {
++ au_vm_prfile_set(vma, file);
++ fsstack_copy_attr_atime(inode, file_inode(h_file));
++ goto out_fput; /* success */
++ }
++ au_unset_mmapped(file);
++ au_vm_file_reset(vma, file);
++
++out_fput:
++ lockdep_off();
++ ii_write_unlock(inode);
++ lockdep_on();
++ fput(h_file);
++out:
++ lockdep_off();
++ si_read_unlock(sb);
++ lockdep_on();
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline void au_set_mmapped(struct file *f)
++/* ---------------------------------------------------------------------- */
++
++static int aufs_fsync_nondir(struct file *file, loff_t start, loff_t end,
++ int datasync)
+{
-+ if (atomic_inc_return(&au_fi(f)->fi_mmapped))
-+ return;
-+ pr_warn("fi_mmapped wrapped around\n");
-+ while (!atomic_inc_return(&au_fi(f)->fi_mmapped))
-+ ;
++ int err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
++
++ err = 0; /* -EBADF; */ /* posix? */
++ if (unlikely(!(file->f_mode & FMODE_WRITE)))
++ goto out;
++
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
++
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out_unlock;
++
++ err = vfsub_fsync(h_file, &h_file->f_path, datasync);
++ au_write_post(inode, h_file, &wpre, /*written*/0);
++
++out_unlock:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++out:
++ return err;
+}
+
-+static inline void au_unset_mmapped(struct file *f)
++static int aufs_fasync(int fd, struct file *file, int flag)
+{
-+ atomic_dec(&au_fi(f)->fi_mmapped);
++ int err;
++ struct file *h_file;
++ struct super_block *sb;
++
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++
++ if (h_file->f_op->fasync)
++ err = h_file->f_op->fasync(fd, h_file, flag);
++ fput(h_file); /* instead of au_read_post() */
++
++out:
++ si_read_unlock(sb);
++ return err;
+}
+
-+static inline int au_test_mmapped(struct file *f)
++static int aufs_setfl(struct file *file, unsigned long arg)
+{
-+ return atomic_read(&au_fi(f)->fi_mmapped);
-+}
++ int err;
++ struct file *h_file;
++ struct super_block *sb;
+
-+/* customize vma->vm_file */
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+static inline void au_do_vm_file_reset(struct vm_area_struct *vma,
-+ struct file *file)
-+{
-+ struct file *f;
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+ f = vma->vm_file;
-+ get_file(file);
-+ vma->vm_file = file;
-+ fput(f);
++ /* stop calling h_file->fasync */
++ arg |= vfsub_file_flags(file) & FASYNC;
++ err = setfl(/*unused fd*/-1, h_file, arg);
++ fput(h_file); /* instead of au_read_post() */
++
++out:
++ si_read_unlock(sb);
++ return err;
+}
+
-+#ifdef CONFIG_MMU
-+#define AuDbgVmRegion(file, vma) do {} while (0)
++/* ---------------------------------------------------------------------- */
+
-+static inline void au_vm_file_reset(struct vm_area_struct *vma,
-+ struct file *file)
++/* no one supports this operation, currently */
++#if 0
++static ssize_t aufs_sendpage(struct file *file, struct page *page, int offset,
++ size_t len, loff_t *pos, int more)
+{
-+ au_do_vm_file_reset(vma, file);
+}
-+#else
-+#define AuDbgVmRegion(file, vma) \
-+ AuDebugOn((vma)->vm_region && (vma)->vm_region->vm_file != (file))
++#endif
+
-+static inline void au_vm_file_reset(struct vm_area_struct *vma,
-+ struct file *file)
-+{
-+ struct file *f;
++/* ---------------------------------------------------------------------- */
+
-+ au_do_vm_file_reset(vma, file);
-+ f = vma->vm_region->vm_file;
-+ get_file(file);
-+ vma->vm_region->vm_file = file;
-+ fput(f);
-+}
-+#endif /* CONFIG_MMU */
++const struct file_operations aufs_file_fop = {
++ .owner = THIS_MODULE,
+
-+/* handle vma->vm_prfile */
-+static inline void au_vm_prfile_set(struct vm_area_struct *vma,
-+ struct file *file)
-+{
-+ get_file(file);
-+ vma->vm_prfile = file;
-+#ifndef CONFIG_MMU
-+ get_file(file);
-+ vma->vm_region->vm_prfile = file;
-+#endif
-+}
++ .llseek = default_llseek,
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_FILE_H__ */
-diff -urN /usr/share/empty/fs/aufs/finfo.c linux/fs/aufs/finfo.c
---- /usr/share/empty/fs/aufs/finfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/finfo.c 2016-08-17 18:01:06.135222590 +0200
-@@ -0,0 +1,151 @@
++ .read = aufs_read,
++ .write = aufs_write,
++ .read_iter = aufs_read_iter,
++ .write_iter = aufs_write_iter,
++
++#ifdef CONFIG_AUFS_POLL
++ .poll = aufs_poll,
++#endif
++ .unlocked_ioctl = aufs_ioctl_nondir,
++#ifdef CONFIG_COMPAT
++ .compat_ioctl = aufs_compat_ioctl_nondir,
++#endif
++ .mmap = aufs_mmap,
++ .open = aufs_open_nondir,
++ .flush = aufs_flush_nondir,
++ .release = aufs_release_nondir,
++ .fsync = aufs_fsync_nondir,
++ .fasync = aufs_fasync,
++ /* .sendpage = aufs_sendpage, */
++ .setfl = aufs_setfl,
++ .splice_write = aufs_splice_write,
++ .splice_read = aufs_splice_read,
++#if 0
++ .aio_splice_write = aufs_aio_splice_write,
++ .aio_splice_read = aufs_aio_splice_read,
++#endif
++ .fallocate = aufs_fallocate,
++ .copy_file_range = aufs_copy_file_range
++};
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/fhsm.c linux-4.14/fs/aufs/fhsm.c
+--- linux-4.14/fs/aufs/fhsm.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/fhsm.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,426 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2011-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ * along with this program; if not, write to the Free Software
++ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
+ */
+
+/*
-+ * file private data
++ * File-based Hierarchy Storage Management
+ */
+
++#include <linux/anon_inodes.h>
++#include <linux/poll.h>
++#include <linux/seq_file.h>
++#include <linux/statfs.h>
+#include "aufs.h"
+
-+void au_hfput(struct au_hfile *hf, int execed)
++static aufs_bindex_t au_fhsm_bottom(struct super_block *sb)
+{
-+ if (execed)
-+ allow_write_access(hf->hf_file);
-+ fput(hf->hf_file);
-+ hf->hf_file = NULL;
-+ au_br_put(hf->hf_br);
-+ hf->hf_br = NULL;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++
++ SiMustAnyLock(sb);
++
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ AuDebugOn(!fhsm);
++ return fhsm->fhsm_bottom;
+}
+
-+void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val)
++void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ struct au_finfo *finfo = au_fi(file);
-+ struct au_hfile *hf;
-+ struct au_fidir *fidir;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ fidir = finfo->fi_hdir;
-+ if (!fidir) {
-+ AuDebugOn(finfo->fi_btop != bindex);
-+ hf = &finfo->fi_htop;
-+ } else
-+ hf = fidir->fd_hfile + bindex;
++ SiMustWriteLock(sb);
+
-+ if (hf && hf->hf_file)
-+ au_hfput(hf, vfsub_file_execed(file));
-+ if (val) {
-+ FiMustWriteLock(file);
-+ AuDebugOn(IS_ERR_OR_NULL(file->f_path.dentry));
-+ hf->hf_file = val;
-+ hf->hf_br = au_sbr(file->f_path.dentry->d_sb, bindex);
-+ }
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ AuDebugOn(!fhsm);
++ fhsm->fhsm_bottom = bindex;
+}
+
-+void au_update_figen(struct file *file)
++/* ---------------------------------------------------------------------- */
++
++static int au_fhsm_test_jiffy(struct au_sbinfo *sbinfo, struct au_branch *br)
+{
-+ atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_path.dentry));
-+ /* smp_mb(); */ /* atomic_set */
++ struct au_br_fhsm *bf;
++
++ bf = br->br_fhsm;
++ MtxMustLock(&bf->bf_lock);
++
++ return !bf->bf_readable
++ || time_after(jiffies,
++ bf->bf_jiffy + sbinfo->si_fhsm.fhsm_expire);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct au_fidir *au_fidir_alloc(struct super_block *sb)
++static void au_fhsm_notify(struct super_block *sb, int val)
+{
-+ struct au_fidir *fidir;
-+ int nbr;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ nbr = au_sbbot(sb) + 1;
-+ if (nbr < 2)
-+ nbr = 2; /* initial allocate for 2 branches */
-+ fidir = kzalloc(au_fidir_sz(nbr), GFP_NOFS);
-+ if (fidir) {
-+ fidir->fd_bbot = -1;
-+ fidir->fd_nent = nbr;
-+ }
++ SiMustAnyLock(sb);
+
-+ return fidir;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ if (au_fhsm_pid(fhsm)
++ && atomic_read(&fhsm->fhsm_readable) != -1) {
++ atomic_set(&fhsm->fhsm_readable, val);
++ if (val)
++ wake_up(&fhsm->fhsm_wqh);
++ }
+}
+
-+int au_fidir_realloc(struct au_finfo *finfo, int nbr)
++static int au_fhsm_stfs(struct super_block *sb, aufs_bindex_t bindex,
++ struct aufs_stfs *rstfs, int do_lock, int do_notify)
+{
+ int err;
-+ struct au_fidir *fidir, *p;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+ AuRwMustWriteLock(&finfo->fi_rwsem);
-+ fidir = finfo->fi_hdir;
-+ AuDebugOn(!fidir);
++ br = au_sbr(sb, bindex);
++ AuDebugOn(au_br_rdonly(br));
++ bf = br->br_fhsm;
++ AuDebugOn(!bf);
+
-+ err = -ENOMEM;
-+ p = au_kzrealloc(fidir, au_fidir_sz(fidir->fd_nent), au_fidir_sz(nbr),
-+ GFP_NOFS);
-+ if (p) {
-+ p->fd_nent = nbr;
-+ finfo->fi_hdir = p;
-+ err = 0;
++ if (do_lock)
++ mutex_lock(&bf->bf_lock);
++ else
++ MtxMustLock(&bf->bf_lock);
++
++ /* sb->s_root for NFS is unreliable */
++ err = au_br_stfs(br, &bf->bf_stfs);
++ if (unlikely(err)) {
++ AuErr1("FHSM failed (%d), b%d, ignored.\n", bindex, err);
++ goto out;
+ }
+
++ bf->bf_jiffy = jiffies;
++ bf->bf_readable = 1;
++ if (do_notify)
++ au_fhsm_notify(sb, /*val*/1);
++ if (rstfs)
++ *rstfs = bf->bf_stfs;
++
++out:
++ if (do_lock)
++ mutex_unlock(&bf->bf_lock);
++ au_fhsm_notify(sb, /*val*/1);
++
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_finfo_fin(struct file *file, int atonce)
++void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force)
+{
-+ struct au_finfo *finfo;
++ int err;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+ au_nfiles_dec(file->f_path.dentry->d_sb);
++ AuDbg("b%d, force %d\n", bindex, force);
++ SiMustAnyLock(sb);
+
-+ finfo = au_fi(file);
-+ AuDebugOn(finfo->fi_hdir);
-+ AuRwDestroy(&finfo->fi_rwsem);
-+ if (!atonce)
-+ au_cache_dfree_finfo(finfo);
-+ else
-+ au_cache_free_finfo(finfo);
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ if (!au_ftest_si(sbinfo, FHSM)
++ || fhsm->fhsm_bottom == bindex)
++ return;
++
++ br = au_sbr(sb, bindex);
++ bf = br->br_fhsm;
++ AuDebugOn(!bf);
++ mutex_lock(&bf->bf_lock);
++ if (force
++ || au_fhsm_pid(fhsm)
++ || au_fhsm_test_jiffy(sbinfo, br))
++ err = au_fhsm_stfs(sb, bindex, /*rstfs*/NULL, /*do_lock*/0,
++ /*do_notify*/1);
++ mutex_unlock(&bf->bf_lock);
+}
+
-+void au_fi_init_once(void *_finfo)
++void au_fhsm_wrote_all(struct super_block *sb, int force)
+{
-+ struct au_finfo *finfo = _finfo;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
+
-+ au_rw_init(&finfo->fi_rwsem);
++ /* exclude the bottom */
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm))
++ au_fhsm_wrote(sb, bindex, force);
++ }
+}
+
-+int au_finfo_init(struct file *file, struct au_fidir *fidir)
++/* ---------------------------------------------------------------------- */
++
++static unsigned int au_fhsm_poll(struct file *file,
++ struct poll_table_struct *wait)
+{
-+ int err;
-+ struct au_finfo *finfo;
-+ struct dentry *dentry;
++ unsigned int mask;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ err = -ENOMEM;
-+ dentry = file->f_path.dentry;
-+ finfo = au_cache_alloc_finfo();
-+ if (unlikely(!finfo))
-+ goto out;
++ mask = 0;
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++ poll_wait(file, &fhsm->fhsm_wqh, wait);
++ if (atomic_read(&fhsm->fhsm_readable))
++ mask = POLLIN /* | POLLRDNORM */;
+
-+ err = 0;
-+ au_nfiles_inc(dentry->d_sb);
-+ au_rw_write_lock(&finfo->fi_rwsem);
-+ finfo->fi_btop = -1;
-+ finfo->fi_hdir = fidir;
-+ atomic_set(&finfo->fi_generation, au_digen(dentry));
-+ /* smp_mb(); */ /* atomic_set */
++ AuTraceErr((int)mask);
++ return mask;
++}
+
-+ file->private_data = finfo;
++static int au_fhsm_do_read_one(struct aufs_stbr __user *stbr,
++ struct aufs_stfs *stfs, __s16 brid)
++{
++ int err;
++
++ err = copy_to_user(&stbr->stfs, stfs, sizeof(*stfs));
++ if (!err)
++ err = __put_user(brid, &stbr->brid);
++ if (unlikely(err))
++ err = -EFAULT;
+
-+out:
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/f_op.c linux/fs/aufs/f_op.c
---- /usr/share/empty/fs/aufs/f_op.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/f_op.c 2016-08-17 18:01:06.135222590 +0200
-@@ -0,0 +1,772 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
+
-+/*
-+ * file and vm operations
-+ */
++static ssize_t au_fhsm_do_read(struct super_block *sb,
++ struct aufs_stbr __user *stbr, size_t count)
++{
++ ssize_t err;
++ int nstbr;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+#include <linux/aio.h>
-+#include <linux/fs_stack.h>
-+#include <linux/mman.h>
-+#include <linux/security.h>
-+#include "aufs.h"
++ /* except the bottom branch */
++ err = 0;
++ nstbr = 0;
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; !err && bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (!au_br_fhsm(br->br_perm))
++ continue;
+
-+int au_do_open_nondir(struct file *file, int flags, struct file *h_file)
-+{
-+ int err;
-+ aufs_bindex_t bindex;
-+ struct dentry *dentry, *h_dentry;
-+ struct au_finfo *finfo;
-+ struct inode *h_inode;
++ bf = br->br_fhsm;
++ mutex_lock(&bf->bf_lock);
++ if (bf->bf_readable) {
++ err = -EFAULT;
++ if (count >= sizeof(*stbr))
++ err = au_fhsm_do_read_one(stbr++, &bf->bf_stfs,
++ br->br_id);
++ if (!err) {
++ bf->bf_readable = 0;
++ count -= sizeof(*stbr);
++ nstbr++;
++ }
++ }
++ mutex_unlock(&bf->bf_lock);
++ }
++ if (!err)
++ err = sizeof(*stbr) * nstbr;
+
-+ FiMustWriteLock(file);
++ return err;
++}
++
++static ssize_t au_fhsm_read(struct file *file, char __user *buf, size_t count,
++ loff_t *pos)
++{
++ ssize_t err;
++ int readable;
++ aufs_bindex_t nfhsm, bindex, bbot;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ struct au_branch *br;
++ struct super_block *sb;
+
+ err = 0;
-+ dentry = file->f_path.dentry;
-+ AuDebugOn(IS_ERR_OR_NULL(dentry));
-+ finfo = au_fi(file);
-+ memset(&finfo->fi_htop, 0, sizeof(finfo->fi_htop));
-+ atomic_set(&finfo->fi_mmapped, 0);
-+ bindex = au_dbtop(dentry);
-+ if (!h_file) {
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
-+ if (unlikely(err))
-+ goto out;
-+ h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0);
-+ } else {
-+ h_dentry = h_file->f_path.dentry;
-+ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
-+ if (unlikely(err))
-+ goto out;
-+ get_file(h_file);
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++need_data:
++ spin_lock_irq(&fhsm->fhsm_wqh.lock);
++ if (!atomic_read(&fhsm->fhsm_readable)) {
++ if (vfsub_file_flags(file) & O_NONBLOCK)
++ err = -EAGAIN;
++ else
++ err = wait_event_interruptible_locked_irq
++ (fhsm->fhsm_wqh,
++ atomic_read(&fhsm->fhsm_readable));
+ }
-+ if (IS_ERR(h_file))
-+ err = PTR_ERR(h_file);
-+ else {
-+ if ((flags & __O_TMPFILE)
-+ && !(flags & O_EXCL)) {
-+ h_inode = file_inode(h_file);
-+ spin_lock(&h_inode->i_lock);
-+ h_inode->i_state |= I_LINKABLE;
-+ spin_unlock(&h_inode->i_lock);
++ spin_unlock_irq(&fhsm->fhsm_wqh.lock);
++ if (unlikely(err))
++ goto out;
++
++ /* sb may already be dead */
++ au_rw_read_lock(&sbinfo->si_rwsem);
++ readable = atomic_read(&fhsm->fhsm_readable);
++ if (readable > 0) {
++ sb = sbinfo->si_sb;
++ AuDebugOn(!sb);
++ /* exclude the bottom branch */
++ nfhsm = 0;
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm))
++ nfhsm++;
++ }
++ err = -EMSGSIZE;
++ if (nfhsm * sizeof(struct aufs_stbr) <= count) {
++ atomic_set(&fhsm->fhsm_readable, 0);
++ err = au_fhsm_do_read(sbinfo->si_sb, (void __user *)buf,
++ count);
+ }
-+ au_set_fbtop(file, bindex);
-+ au_set_h_fptr(file, bindex, h_file);
-+ au_update_figen(file);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
+ }
++ au_rw_read_unlock(&sbinfo->si_rwsem);
++ if (!readable)
++ goto need_data;
+
+out:
+ return err;
+}
+
-+static int aufs_open_nondir(struct inode *inode __maybe_unused,
-+ struct file *file)
++static int au_fhsm_release(struct inode *inode, struct file *file)
+{
-+ int err;
-+ struct super_block *sb;
-+ struct au_do_open_args args = {
-+ .open = au_do_open_nondir
-+ };
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ AuDbg("%pD, f_flags 0x%x, f_mode 0x%x\n",
-+ file, vfsub_file_flags(file), file->f_mode);
++ /* sb may already be dead */
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock(&fhsm->fhsm_spin);
++ fhsm->fhsm_pid = 0;
++ spin_unlock(&fhsm->fhsm_spin);
++ kobject_put(&sbinfo->si_kobj);
+
-+ sb = file->f_path.dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_do_open(file, &args);
-+ si_read_unlock(sb);
-+ return err;
++ return 0;
+}
+
-+int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file)
++static const struct file_operations au_fhsm_fops = {
++ .owner = THIS_MODULE,
++ .llseek = noop_llseek,
++ .read = au_fhsm_read,
++ .poll = au_fhsm_poll,
++ .release = au_fhsm_release
++};
++
++int au_fhsm_fd(struct super_block *sb, int oflags)
+{
-+ struct au_finfo *finfo;
-+ aufs_bindex_t bindex;
-+ int delayed;
++ int err, fd;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ finfo = au_fi(file);
-+ au_sphl_del(&finfo->fi_hlist,
-+ &au_sbi(file->f_path.dentry->d_sb)->si_files);
-+ bindex = finfo->fi_btop;
-+ if (bindex >= 0)
-+ au_set_h_fptr(file, bindex, NULL);
++ err = -EPERM;
++ if (unlikely(!capable(CAP_SYS_ADMIN)))
++ goto out;
+
-+ delayed = (current->flags & PF_KTHREAD) || in_interrupt();
-+ au_finfo_fin(file, delayed);
-+ return 0;
++ err = -EINVAL;
++ if (unlikely(oflags & ~(O_CLOEXEC | O_NONBLOCK)))
++ goto out;
++
++ err = 0;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock(&fhsm->fhsm_spin);
++ if (!fhsm->fhsm_pid)
++ fhsm->fhsm_pid = current->pid;
++ else
++ err = -EBUSY;
++ spin_unlock(&fhsm->fhsm_spin);
++ if (unlikely(err))
++ goto out;
++
++ oflags |= O_RDONLY;
++ /* oflags |= FMODE_NONOTIFY; */
++ fd = anon_inode_getfd("[aufs_fhsm]", &au_fhsm_fops, sbinfo, oflags);
++ err = fd;
++ if (unlikely(fd < 0))
++ goto out_pid;
++
++ /* succeed reglardless 'fhsm' status */
++ kobject_get(&sbinfo->si_kobj);
++ si_noflush_read_lock(sb);
++ if (au_ftest_si(sbinfo, FHSM))
++ au_fhsm_wrote_all(sb, /*force*/0);
++ si_read_unlock(sb);
++ goto out; /* success */
++
++out_pid:
++ spin_lock(&fhsm->fhsm_spin);
++ fhsm->fhsm_pid = 0;
++ spin_unlock(&fhsm->fhsm_spin);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int au_do_flush_nondir(struct file *file, fl_owner_t id)
++int au_fhsm_br_alloc(struct au_branch *br)
+{
+ int err;
-+ struct file *h_file;
+
+ err = 0;
-+ h_file = au_hf_top(file);
-+ if (h_file)
-+ err = vfsub_flush(h_file, id);
++ br->br_fhsm = kmalloc(sizeof(*br->br_fhsm), GFP_NOFS);
++ if (br->br_fhsm)
++ au_br_fhsm_init(br->br_fhsm);
++ else
++ err = -ENOMEM;
++
+ return err;
+}
+
-+static int aufs_flush_nondir(struct file *file, fl_owner_t id)
++/* ---------------------------------------------------------------------- */
++
++void au_fhsm_fin(struct super_block *sb)
+{
-+ return au_do_flush(file, id, au_do_flush_nondir);
++ au_fhsm_notify(sb, /*val*/-1);
+}
+
-+/* ---------------------------------------------------------------------- */
++void au_fhsm_init(struct au_sbinfo *sbinfo)
++{
++ struct au_fhsm *fhsm;
++
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock_init(&fhsm->fhsm_spin);
++ init_waitqueue_head(&fhsm->fhsm_wqh);
++ atomic_set(&fhsm->fhsm_readable, 0);
++ fhsm->fhsm_expire
++ = msecs_to_jiffies(AUFS_FHSM_CACHE_DEF_SEC * MSEC_PER_SEC);
++ fhsm->fhsm_bottom = -1;
++}
++
++void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec)
++{
++ sbinfo->si_fhsm.fhsm_expire
++ = msecs_to_jiffies(sec * MSEC_PER_SEC);
++}
++
++void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo)
++{
++ unsigned int u;
++
++ if (!au_ftest_si(sbinfo, FHSM))
++ return;
++
++ u = jiffies_to_msecs(sbinfo->si_fhsm.fhsm_expire) / MSEC_PER_SEC;
++ if (u != AUFS_FHSM_CACHE_DEF_SEC)
++ seq_printf(seq, ",fhsm_sec=%u", u);
++}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/file.c linux-4.14/fs/aufs/file.c
+--- linux-4.14/fs/aufs/file.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/file.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,856 @@
+/*
-+ * read and write functions acquire [fdi]_rwsem once, but release before
-+ * mmap_sem. This is because to stop a race condition between mmap(2).
-+ * Releasing these aufs-rwsem should be safe, no branch-mamagement (by keeping
-+ * si_rwsem), no harmful copy-up should happen. Actually copy-up may happen in
-+ * read functions after [fdi]_rwsem are released, but it should be harmless.
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
-+/* Callers should call au_read_post() or fput() in the end */
-+struct file *au_read_pre(struct file *file, int keep_fi)
-+{
-+ struct file *h_file;
-+ int err;
-+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0);
-+ if (!err) {
-+ di_read_unlock(file->f_path.dentry, AuLock_IR);
-+ h_file = au_hf_top(file);
-+ get_file(h_file);
-+ if (!keep_fi)
-+ fi_read_unlock(file);
-+ } else
-+ h_file = ERR_PTR(err);
++/*
++ * handling file/dir, and address_space operation
++ */
+
-+ return h_file;
-+}
++#ifdef CONFIG_AUFS_DEBUG
++#include <linux/migrate.h>
++#endif
++#include <linux/pagemap.h>
++#include "aufs.h"
+
-+static void au_read_post(struct inode *inode, struct file *h_file)
++/* drop flags for writing */
++unsigned int au_file_roflags(unsigned int flags)
+{
-+ /* update without lock, I don't think it a problem */
-+ fsstack_copy_attr_atime(inode, file_inode(h_file));
-+ fput(h_file);
++ flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC);
++ flags |= O_RDONLY | O_NOATIME;
++ return flags;
+}
+
-+struct au_write_pre {
-+ blkcnt_t blks;
-+ aufs_bindex_t btop;
-+};
-+
-+/*
-+ * return with iinfo is write-locked
-+ * callers should call au_write_post() or iinfo_write_unlock() + fput() in the
-+ * end
-+ */
-+static struct file *au_write_pre(struct file *file, int do_ready,
-+ struct au_write_pre *wpre)
++/* common functions to regular file and dir */
++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
++ struct file *file, int force_wr)
+{
+ struct file *h_file;
-+ struct dentry *dentry;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct path h_path;
+ int err;
-+ struct au_pin pin;
+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1);
++ /* a race condition can happen between open and unlink/rmdir */
++ h_file = ERR_PTR(-ENOENT);
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (au_test_nfsd() && (!h_dentry || d_is_negative(h_dentry)))
++ goto out;
++ h_inode = d_inode(h_dentry);
++ spin_lock(&h_dentry->d_lock);
++ err = (!d_unhashed(dentry) && d_unlinked(h_dentry))
++ /* || !d_inode(dentry)->i_nlink */
++ ;
++ spin_unlock(&h_dentry->d_lock);
++ if (unlikely(err))
++ goto out;
++
++ sb = dentry->d_sb;
++ br = au_sbr(sb, bindex);
++ err = au_br_test_oflag(flags, br);
+ h_file = ERR_PTR(err);
+ if (unlikely(err))
+ goto out;
+
-+ dentry = file->f_path.dentry;
-+ if (do_ready) {
-+ err = au_ready_to_write(file, -1, &pin);
++ /* drop flags for writing */
++ if (au_test_ro(sb, bindex, d_inode(dentry))) {
++ if (force_wr && !(flags & O_WRONLY))
++ force_wr = 0;
++ flags = au_file_roflags(flags);
++ if (force_wr) {
++ h_file = ERR_PTR(-EROFS);
++ flags = au_file_roflags(flags);
++ if (unlikely(vfsub_native_ro(h_inode)
++ || IS_APPEND(h_inode)))
++ goto out;
++ flags &= ~O_ACCMODE;
++ flags |= O_WRONLY;
++ }
++ }
++ flags &= ~O_CREAT;
++ au_br_get(br);
++ h_path.dentry = h_dentry;
++ h_path.mnt = au_br_mnt(br);
++ h_file = vfsub_dentry_open(&h_path, flags);
++ if (IS_ERR(h_file))
++ goto out_br;
++
++ if (flags & __FMODE_EXEC) {
++ err = deny_write_access(h_file);
+ if (unlikely(err)) {
++ fput(h_file);
+ h_file = ERR_PTR(err);
-+ di_write_unlock(dentry);
-+ goto out_fi;
++ goto out_br;
+ }
+ }
++ fsnotify_open(h_file);
++ goto out; /* success */
+
-+ di_downgrade_lock(dentry, /*flags*/0);
-+ if (wpre)
-+ wpre->btop = au_fbtop(file);
-+ h_file = au_hf_top(file);
-+ get_file(h_file);
-+ if (wpre)
-+ wpre->blks = file_inode(h_file)->i_blocks;
-+ if (do_ready)
-+ au_unpin(&pin);
-+ di_read_unlock(dentry, /*flags*/0);
-+
-+out_fi:
-+ fi_write_unlock(file);
++out_br:
++ au_br_put(br);
+out:
+ return h_file;
+}
+
-+static void au_write_post(struct inode *inode, struct file *h_file,
-+ struct au_write_pre *wpre, ssize_t written)
-+{
-+ struct inode *h_inode;
-+
-+ au_cpup_attr_timesizes(inode);
-+ AuDebugOn(au_ibtop(inode) != wpre->btop);
-+ h_inode = file_inode(h_file);
-+ inode->i_mode = h_inode->i_mode;
-+ ii_write_unlock(inode);
-+ fput(h_file);
-+
-+ /* AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); */
-+ if (written > 0)
-+ au_fhsm_wrote(inode->i_sb, wpre->btop,
-+ /*force*/h_inode->i_blocks > wpre->blks);
-+}
-+
-+static ssize_t aufs_read(struct file *file, char __user *buf, size_t count,
-+ loff_t *ppos)
++static int au_cmoo(struct dentry *dentry)
+{
-+ ssize_t err;
-+ struct inode *inode;
-+ struct file *h_file;
++ int err, cmoo, matched;
++ unsigned int udba;
++ struct path h_path;
++ struct au_pin pin;
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = -1,
++ .bsrc = -1,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ struct inode *delegated;
+ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ pid_t pid;
++ struct au_branch *br;
++ struct dentry *parent;
++ struct au_hinode *hdir;
+
-+ inode = file_inode(file);
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ DiMustWriteLock(dentry);
++ IiMustWriteLock(d_inode(dentry));
+
-+ h_file = au_read_pre(file, /*keep_fi*/0);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
++ err = 0;
++ if (IS_ROOT(dentry))
++ goto out;
++ cpg.bsrc = au_dbtop(dentry);
++ if (!cpg.bsrc)
+ goto out;
+
-+ /* filedata may be obsoleted by concurrent copyup, but no problem */
-+ err = vfsub_read_u(h_file, buf, count, ppos);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
-+ au_read_post(inode, h_file);
-+
-+out:
-+ si_read_unlock(sb);
-+ return err;
-+}
++ sb = dentry->d_sb;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ pid = au_fhsm_pid(fhsm);
++ rcu_read_lock();
++ matched = (pid
++ && (current->pid == pid
++ || rcu_dereference(current->real_parent)->pid == pid));
++ rcu_read_unlock();
++ if (matched)
++ goto out;
+
-+/*
-+ * todo: very ugly
-+ * it locks both of i_mutex and si_rwsem for read in safe.
-+ * if the plink maintenance mode continues forever (that is the problem),
-+ * may loop forever.
-+ */
-+static void au_mtx_and_read_lock(struct inode *inode)
-+{
-+ int err;
-+ struct super_block *sb = inode->i_sb;
++ br = au_sbr(sb, cpg.bsrc);
++ cmoo = au_br_cmoo(br->br_perm);
++ if (!cmoo)
++ goto out;
++ if (!d_is_reg(dentry))
++ cmoo &= AuBrAttr_COO_ALL;
++ if (!cmoo)
++ goto out;
+
-+ while (1) {
-+ inode_lock(inode);
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ if (!err)
-+ break;
-+ inode_unlock(inode);
-+ si_read_lock(sb, AuLock_NOPLMW);
-+ si_read_unlock(sb);
++ parent = dget_parent(dentry);
++ di_write_lock_parent(parent);
++ err = au_wbr_do_copyup_bu(dentry, cpg.bsrc - 1);
++ cpg.bdst = err;
++ if (unlikely(err < 0)) {
++ err = 0; /* there is no upper writable branch */
++ goto out_dgrade;
+ }
-+}
++ AuDbg("bsrc %d, bdst %d\n", cpg.bsrc, cpg.bdst);
+
-+static ssize_t aufs_write(struct file *file, const char __user *ubuf,
-+ size_t count, loff_t *ppos)
-+{
-+ ssize_t err;
-+ struct au_write_pre wpre;
-+ struct inode *inode;
-+ struct file *h_file;
-+ char __user *buf = (char __user *)ubuf;
++ /* do not respect the coo attrib for the target branch */
++ err = au_cpup_dirs(dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_dgrade;
+
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
++ di_downgrade_lock(parent, AuLock_IR);
++ udba = au_opt_udba(sb);
++ err = au_pin(&pin, dentry, cpg.bdst, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_parent;
+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ if (unlikely(err))
++ goto out_parent;
++ if (!(cmoo & AuBrWAttr_MOO))
++ goto out_parent; /* success */
+
-+ err = vfsub_write_u(h_file, buf, count, ppos);
-+ au_write_post(inode, h_file, &wpre, err);
++ err = au_pin(&pin, dentry, cpg.bsrc, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_parent;
++
++ h_path.mnt = au_br_mnt(br);
++ h_path.dentry = au_h_dptr(dentry, cpg.bsrc);
++ hdir = au_hi(d_inode(parent), cpg.bsrc);
++ delegated = NULL;
++ err = vfsub_unlink(hdir->hi_inode, &h_path, &delegated, /*force*/1);
++ au_unpin(&pin);
++ /* todo: keep h_dentry or not? */
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ if (unlikely(err)) {
++ pr_err("unlink %pd after coo failed (%d), ignored\n",
++ dentry, err);
++ err = 0;
++ }
++ goto out_parent; /* success */
+
++out_dgrade:
++ di_downgrade_lock(parent, AuLock_IR);
++out_parent:
++ di_read_unlock(parent, AuLock_IR);
++ dput(parent);
+out:
-+ si_read_unlock(inode->i_sb);
-+ inode_unlock(inode);
++ AuTraceErr(err);
+ return err;
+}
+
-+static ssize_t au_do_iter(struct file *h_file, int rw, struct kiocb *kio,
-+ struct iov_iter *iov_iter)
++int au_do_open(struct file *file, struct au_do_open_args *args)
+{
-+ ssize_t err;
-+ struct file *file;
-+ ssize_t (*iter)(struct kiocb *, struct iov_iter *);
++ int err, aopen = args->aopen;
++ struct dentry *dentry;
++ struct au_finfo *finfo;
+
-+ err = security_file_permission(h_file, rw);
++ if (!aopen)
++ err = au_finfo_init(file, args->fidir);
++ else {
++ lockdep_off();
++ err = au_finfo_init(file, args->fidir);
++ lockdep_on();
++ }
+ if (unlikely(err))
+ goto out;
+
-+ err = -ENOSYS;
-+ iter = NULL;
-+ if (rw == MAY_READ)
-+ iter = h_file->f_op->read_iter;
-+ else if (rw == MAY_WRITE)
-+ iter = h_file->f_op->write_iter;
++ dentry = file->f_path.dentry;
++ AuDebugOn(IS_ERR_OR_NULL(dentry));
++ di_write_lock_child(dentry);
++ err = au_cmoo(dentry);
++ di_downgrade_lock(dentry, AuLock_IR);
++ if (!err) {
++ if (!aopen)
++ err = args->open(file, vfsub_file_flags(file), NULL);
++ else {
++ lockdep_off();
++ err = args->open(file, vfsub_file_flags(file), NULL);
++ lockdep_on();
++ }
++ }
++ di_read_unlock(dentry, AuLock_IR);
+
-+ file = kio->ki_filp;
-+ kio->ki_filp = h_file;
-+ if (iter) {
++ finfo = au_fi(file);
++ if (!err) {
++ finfo->fi_file = file;
++ au_hbl_add(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
++ }
++ if (!aopen)
++ fi_write_unlock(file);
++ else {
+ lockdep_off();
-+ err = iter(kio, iov_iter);
++ fi_write_unlock(file);
+ lockdep_on();
-+ } else
-+ /* currently there is no such fs */
-+ WARN_ON_ONCE(1);
-+ kio->ki_filp = file;
++ }
++ if (unlikely(err)) {
++ finfo->fi_hdir = NULL;
++ au_finfo_fin(file);
++ }
+
+out:
++ AuTraceErr(err);
+ return err;
+}
+
-+static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
++int au_reopen_nondir(struct file *file)
+{
-+ ssize_t err;
-+ struct file *file, *h_file;
-+ struct inode *inode;
-+ struct super_block *sb;
++ int err;
++ aufs_bindex_t btop;
++ struct dentry *dentry;
++ struct file *h_file, *h_file_tmp;
+
-+ file = kio->ki_filp;
-+ inode = file_inode(file);
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ dentry = file->f_path.dentry;
++ btop = au_dbtop(dentry);
++ h_file_tmp = NULL;
++ if (au_fbtop(file) == btop) {
++ h_file = au_hf_top(file);
++ if (file->f_mode == h_file->f_mode)
++ return 0; /* success */
++ h_file_tmp = h_file;
++ get_file(h_file_tmp);
++ au_set_h_fptr(file, btop, NULL);
++ }
++ AuDebugOn(au_fi(file)->fi_hdir);
++ /*
++ * it can happen
++ * file exists on both of rw and ro
++ * open --> dbtop and fbtop are both 0
++ * prepend a branch as rw, "rw" become ro
++ * remove rw/file
++ * delete the top branch, "rw" becomes rw again
++ * --> dbtop is 1, fbtop is still 0
++ * write --> fbtop is 0 but dbtop is 1
++ */
++ /* AuDebugOn(au_fbtop(file) < btop); */
+
-+ h_file = au_read_pre(file, /*keep_fi*/1);
++ h_file = au_h_open(dentry, btop, vfsub_file_flags(file) & ~O_TRUNC,
++ file, /*force_wr*/0);
+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
-+
-+ if (au_test_loopback_kthread()) {
-+ au_warn_loopback(h_file->f_path.dentry->d_sb);
-+ if (file->f_mapping != h_file->f_mapping) {
-+ file->f_mapping = h_file->f_mapping;
-+ smp_mb(); /* unnecessary? */
++ if (IS_ERR(h_file)) {
++ if (h_file_tmp) {
++ au_sbr_get(dentry->d_sb, btop);
++ au_set_h_fptr(file, btop, h_file_tmp);
++ h_file_tmp = NULL;
+ }
++ goto out; /* todo: close all? */
+ }
-+ fi_read_unlock(file);
+
-+ err = au_do_iter(h_file, MAY_READ, kio, iov_iter);
++ err = 0;
++ au_set_fbtop(file, btop);
++ au_set_h_fptr(file, btop, h_file);
++ au_update_figen(file);
+ /* todo: necessary? */
+ /* file->f_ra = h_file->f_ra; */
-+ au_read_post(inode, h_file);
+
+out:
-+ si_read_unlock(sb);
++ if (h_file_tmp)
++ fput(h_file_tmp);
+ return err;
+}
+
-+static ssize_t aufs_write_iter(struct kiocb *kio, struct iov_iter *iov_iter)
-+{
-+ ssize_t err;
-+ struct au_write_pre wpre;
-+ struct inode *inode;
-+ struct file *file, *h_file;
-+
-+ file = kio->ki_filp;
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
-+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
-+
-+ err = au_do_iter(h_file, MAY_WRITE, kio, iov_iter);
-+ au_write_post(inode, h_file, &wpre, err);
-+
-+out:
-+ si_read_unlock(inode->i_sb);
-+ inode_unlock(inode);
-+ return err;
-+}
++/* ---------------------------------------------------------------------- */
+
-+static ssize_t aufs_splice_read(struct file *file, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags)
++static int au_reopen_wh(struct file *file, aufs_bindex_t btgt,
++ struct dentry *hi_wh)
+{
-+ ssize_t err;
-+ struct file *h_file;
-+ struct inode *inode;
-+ struct super_block *sb;
-+
-+ inode = file_inode(file);
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ int err;
++ aufs_bindex_t btop;
++ struct au_dinfo *dinfo;
++ struct dentry *h_dentry;
++ struct au_hdentry *hdp;
+
-+ h_file = au_read_pre(file, /*keep_fi*/0);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ dinfo = au_di(file->f_path.dentry);
++ AuRwMustWriteLock(&dinfo->di_rwsem);
+
-+ err = vfsub_splice_to(h_file, ppos, pipe, len, flags);
-+ /* todo: necessasry? */
-+ /* file->f_ra = h_file->f_ra; */
-+ au_read_post(inode, h_file);
++ btop = dinfo->di_btop;
++ dinfo->di_btop = btgt;
++ hdp = au_hdentry(dinfo, btgt);
++ h_dentry = hdp->hd_dentry;
++ hdp->hd_dentry = hi_wh;
++ err = au_reopen_nondir(file);
++ hdp->hd_dentry = h_dentry;
++ dinfo->di_btop = btop;
+
-+out:
-+ si_read_unlock(sb);
+ return err;
+}
+
-+static ssize_t
-+aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos,
-+ size_t len, unsigned int flags)
++static int au_ready_to_write_wh(struct file *file, loff_t len,
++ aufs_bindex_t bcpup, struct au_pin *pin)
+{
-+ ssize_t err;
-+ struct au_write_pre wpre;
-+ struct inode *inode;
-+ struct file *h_file;
-+
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
++ int err;
++ struct inode *inode, *h_inode;
++ struct dentry *h_dentry, *hi_wh;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = bcpup,
++ .bsrc = -1,
++ .len = len,
++ .pin = pin
++ };
+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ au_update_dbtop(cpg.dentry);
++ inode = d_inode(cpg.dentry);
++ h_inode = NULL;
++ if (au_dbtop(cpg.dentry) <= bcpup
++ && au_dbbot(cpg.dentry) >= bcpup) {
++ h_dentry = au_h_dptr(cpg.dentry, bcpup);
++ if (h_dentry && d_is_positive(h_dentry))
++ h_inode = d_inode(h_dentry);
++ }
++ hi_wh = au_hi_wh(inode, bcpup);
++ if (!hi_wh && !h_inode)
++ err = au_sio_cpup_wh(&cpg, file);
++ else
++ /* already copied-up after unlink */
++ err = au_reopen_wh(file, bcpup, hi_wh);
+
-+ err = vfsub_splice_from(pipe, h_file, ppos, len, flags);
-+ au_write_post(inode, h_file, &wpre, err);
++ if (!err
++ && (inode->i_nlink > 1
++ || (inode->i_state & I_LINKABLE))
++ && au_opt_test(au_mntflags(cpg.dentry->d_sb), PLINK))
++ au_plink_append(inode, bcpup, au_h_dptr(cpg.dentry, bcpup));
+
-+out:
-+ si_read_unlock(inode->i_sb);
-+ inode_unlock(inode);
+ return err;
+}
+
-+static long aufs_fallocate(struct file *file, int mode, loff_t offset,
-+ loff_t len)
++/*
++ * prepare the @file for writing.
++ */
++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin)
+{
-+ long err;
-+ struct au_write_pre wpre;
++ int err;
++ aufs_bindex_t dbtop;
++ struct dentry *parent;
+ struct inode *inode;
++ struct super_block *sb;
+ struct file *h_file;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = -1,
++ .bsrc = -1,
++ .len = len,
++ .pin = pin,
++ .flags = AuCpup_DTIME
++ };
+
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
-+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
++ sb = cpg.dentry->d_sb;
++ inode = d_inode(cpg.dentry);
++ cpg.bsrc = au_fbtop(file);
++ err = au_test_ro(sb, cpg.bsrc, inode);
++ if (!err && (au_hf_top(file)->f_mode & FMODE_WRITE)) {
++ err = au_pin(pin, cpg.dentry, cpg.bsrc, AuOpt_UDBA_NONE,
++ /*flags*/0);
+ goto out;
++ }
+
-+ lockdep_off();
-+ err = vfs_fallocate(h_file, mode, offset, len);
-+ lockdep_on();
-+ au_write_post(inode, h_file, &wpre, /*written*/1);
++ /* need to cpup or reopen */
++ parent = dget_parent(cpg.dentry);
++ di_write_lock_parent(parent);
++ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
++ cpg.bdst = err;
++ if (unlikely(err < 0))
++ goto out_dgrade;
++ err = 0;
+
-+out:
-+ si_read_unlock(inode->i_sb);
-+ inode_unlock(inode);
-+ return err;
-+}
++ if (!d_unhashed(cpg.dentry) && !au_h_dptr(parent, cpg.bdst)) {
++ err = au_cpup_dirs(cpg.dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_dgrade;
++ }
+
-+/* ---------------------------------------------------------------------- */
++ err = au_pin(pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_dgrade;
+
-+/*
-+ * The locking order around current->mmap_sem.
-+ * - in most and regular cases
-+ * file I/O syscall -- aufs_read() or something
-+ * -- si_rwsem for read -- mmap_sem
-+ * (Note that [fdi]i_rwsem are released before mmap_sem).
-+ * - in mmap case
-+ * mmap(2) -- mmap_sem -- aufs_mmap() -- si_rwsem for read -- [fdi]i_rwsem
-+ * This AB-BA order is definitly bad, but is not a problem since "si_rwsem for
-+ * read" allows muliple processes to acquire it and [fdi]i_rwsem are not held in
-+ * file I/O. Aufs needs to stop lockdep in aufs_mmap() though.
-+ * It means that when aufs acquires si_rwsem for write, the process should never
-+ * acquire mmap_sem.
-+ *
-+ * Actually aufs_iterate() holds [fdi]i_rwsem before mmap_sem, but this is not a
-+ * problem either since any directory is not able to be mmap-ed.
-+ * The similar scenario is applied to aufs_readlink() too.
-+ */
++ dbtop = au_dbtop(cpg.dentry);
++ if (dbtop <= cpg.bdst)
++ cpg.bsrc = cpg.bdst;
+
-+#if 0 /* stop calling security_file_mmap() */
-+/* cf. linux/include/linux/mman.h: calc_vm_prot_bits() */
-+#define AuConv_VM_PROT(f, b) _calc_vm_trans(f, VM_##b, PROT_##b)
++ if (dbtop <= cpg.bdst /* just reopen */
++ || !d_unhashed(cpg.dentry) /* copyup and reopen */
++ ) {
++ h_file = au_h_open_pre(cpg.dentry, cpg.bsrc, /*force_wr*/0);
++ if (IS_ERR(h_file))
++ err = PTR_ERR(h_file);
++ else {
++ di_downgrade_lock(parent, AuLock_IR);
++ if (dbtop > cpg.bdst)
++ err = au_sio_cpup_simple(&cpg);
++ if (!err)
++ err = au_reopen_nondir(file);
++ au_h_open_post(cpg.dentry, cpg.bsrc, h_file);
++ }
++ } else { /* copyup as wh and reopen */
++ /*
++ * since writable hfsplus branch is not supported,
++ * h_open_pre/post() are unnecessary.
++ */
++ err = au_ready_to_write_wh(file, len, cpg.bdst, pin);
++ di_downgrade_lock(parent, AuLock_IR);
++ }
+
-+static unsigned long au_arch_prot_conv(unsigned long flags)
-+{
-+ /* currently ppc64 only */
-+#ifdef CONFIG_PPC64
-+ /* cf. linux/arch/powerpc/include/asm/mman.h */
-+ AuDebugOn(arch_calc_vm_prot_bits(-1) != VM_SAO);
-+ return AuConv_VM_PROT(flags, SAO);
-+#else
-+ AuDebugOn(arch_calc_vm_prot_bits(-1));
-+ return 0;
-+#endif
-+}
++ if (!err) {
++ au_pin_set_parent_lflag(pin, /*lflag*/0);
++ goto out_dput; /* success */
++ }
++ au_unpin(pin);
++ goto out_unlock;
+
-+static unsigned long au_prot_conv(unsigned long flags)
-+{
-+ return AuConv_VM_PROT(flags, READ)
-+ | AuConv_VM_PROT(flags, WRITE)
-+ | AuConv_VM_PROT(flags, EXEC)
-+ | au_arch_prot_conv(flags);
++out_dgrade:
++ di_downgrade_lock(parent, AuLock_IR);
++out_unlock:
++ di_read_unlock(parent, AuLock_IR);
++out_dput:
++ dput(parent);
++out:
++ return err;
+}
+
-+/* cf. linux/include/linux/mman.h: calc_vm_flag_bits() */
-+#define AuConv_VM_MAP(f, b) _calc_vm_trans(f, VM_##b, MAP_##b)
-+
-+static unsigned long au_flag_conv(unsigned long flags)
-+{
-+ return AuConv_VM_MAP(flags, GROWSDOWN)
-+ | AuConv_VM_MAP(flags, DENYWRITE)
-+ | AuConv_VM_MAP(flags, LOCKED);
-+}
-+#endif
++/* ---------------------------------------------------------------------- */
+
-+static int aufs_mmap(struct file *file, struct vm_area_struct *vma)
++int au_do_flush(struct file *file, fl_owner_t id,
++ int (*flush)(struct file *file, fl_owner_t id))
+{
+ int err;
-+ const unsigned char wlock
-+ = (file->f_mode & FMODE_WRITE) && (vma->vm_flags & VM_SHARED);
+ struct super_block *sb;
-+ struct file *h_file;
+ struct inode *inode;
+
-+ AuDbgVmRegion(file, vma);
-+
+ inode = file_inode(file);
+ sb = inode->i_sb;
-+ lockdep_off();
-+ si_read_lock(sb, AuLock_NOPLMW);
-+
-+ h_file = au_write_pre(file, wlock, /*wpre*/NULL);
-+ lockdep_on();
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ si_noflush_read_lock(sb);
++ fi_read_lock(file);
++ ii_read_lock_child(inode);
+
-+ err = 0;
-+ au_set_mmapped(file);
-+ au_vm_file_reset(vma, h_file);
-+ /*
-+ * we cannot call security_mmap_file() here since it may acquire
-+ * mmap_sem or i_mutex.
-+ *
-+ * err = security_mmap_file(h_file, au_prot_conv(vma->vm_flags),
-+ * au_flag_conv(vma->vm_flags));
-+ */
-+ if (!err)
-+ err = h_file->f_op->mmap(h_file, vma);
-+ if (!err) {
-+ au_vm_prfile_set(vma, file);
-+ fsstack_copy_attr_atime(inode, file_inode(h_file));
-+ goto out_fput; /* success */
-+ }
-+ au_unset_mmapped(file);
-+ au_vm_file_reset(vma, file);
++ err = flush(file, id);
++ au_cpup_attr_timesizes(inode);
+
-+out_fput:
-+ lockdep_off();
-+ ii_write_unlock(inode);
-+ lockdep_on();
-+ fput(h_file);
-+out:
-+ lockdep_off();
++ ii_read_unlock(inode);
++ fi_read_unlock(file);
+ si_read_unlock(sb);
-+ lockdep_on();
-+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int aufs_fsync_nondir(struct file *file, loff_t start, loff_t end,
-+ int datasync)
++static int au_file_refresh_by_inode(struct file *file, int *need_reopen)
+{
+ int err;
-+ struct au_write_pre wpre;
++ struct au_pin pin;
++ struct au_finfo *finfo;
++ struct dentry *parent, *hi_wh;
+ struct inode *inode;
-+ struct file *h_file;
++ struct super_block *sb;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = -1,
++ .bsrc = -1,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME
++ };
+
-+ err = 0; /* -EBADF; */ /* posix? */
-+ if (unlikely(!(file->f_mode & FMODE_WRITE)))
++ FiMustWriteLock(file);
++
++ err = 0;
++ finfo = au_fi(file);
++ sb = cpg.dentry->d_sb;
++ inode = d_inode(cpg.dentry);
++ cpg.bdst = au_ibtop(inode);
++ if (cpg.bdst == finfo->fi_btop || IS_ROOT(cpg.dentry))
+ goto out;
+
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
++ parent = dget_parent(cpg.dentry);
++ if (au_test_ro(sb, cpg.bdst, inode)) {
++ di_read_lock_parent(parent, !AuLock_IR);
++ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
++ cpg.bdst = err;
++ di_read_unlock(parent, !AuLock_IR);
++ if (unlikely(err < 0))
++ goto out_parent;
++ err = 0;
++ }
+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out_unlock;
++ di_read_lock_parent(parent, AuLock_IR);
++ hi_wh = au_hi_wh(inode, cpg.bdst);
++ if (!S_ISDIR(inode->i_mode)
++ && au_opt_test(au_mntflags(sb), PLINK)
++ && au_plink_test(inode)
++ && !d_unhashed(cpg.dentry)
++ && cpg.bdst < au_dbtop(cpg.dentry)) {
++ err = au_test_and_cpup_dirs(cpg.dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_unlock;
+
-+ err = vfsub_fsync(h_file, &h_file->f_path, datasync);
-+ au_write_post(inode, h_file, &wpre, /*written*/0);
++ /* always superio. */
++ err = au_pin(&pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (!err) {
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ }
++ } else if (hi_wh) {
++ /* already copied-up after unlink */
++ err = au_reopen_wh(file, cpg.bdst, hi_wh);
++ *need_reopen = 0;
++ }
+
+out_unlock:
-+ si_read_unlock(inode->i_sb);
-+ inode_unlock(inode);
++ di_read_unlock(parent, AuLock_IR);
++out_parent:
++ dput(parent);
+out:
+ return err;
+}
+
-+/* no one supports this operation, currently */
-+#if 0
-+static int aufs_aio_fsync_nondir(struct kiocb *kio, int datasync)
++static void au_do_refresh_dir(struct file *file)
+{
-+ int err;
-+ struct au_write_pre wpre;
-+ struct inode *inode, *h_inode;
-+ struct file *file, *h_file;
++ aufs_bindex_t bindex, bbot, new_bindex, brid;
++ struct au_hfile *p, tmp, *q;
++ struct au_finfo *finfo;
++ struct super_block *sb;
++ struct au_fidir *fidir;
+
-+ err = 0; /* -EBADF; */ /* posix? */
-+ if (unlikely(!(file->f_mode & FMODE_WRITE)))
-+ goto out;
++ FiMustWriteLock(file);
+
-+ file = kio->ki_filp;
-+ inode = file_inode(file);
-+ au_mtx_and_read_lock(inode);
++ sb = file->f_path.dentry->d_sb;
++ finfo = au_fi(file);
++ fidir = finfo->fi_hdir;
++ AuDebugOn(!fidir);
++ p = fidir->fd_hfile + finfo->fi_btop;
++ brid = p->hf_br->br_id;
++ bbot = fidir->fd_bbot;
++ for (bindex = finfo->fi_btop; bindex <= bbot; bindex++, p++) {
++ if (!p->hf_file)
++ continue;
+
-+ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out_unlock;
++ new_bindex = au_br_index(sb, p->hf_br->br_id);
++ if (new_bindex == bindex)
++ continue;
++ if (new_bindex < 0) {
++ au_set_h_fptr(file, bindex, NULL);
++ continue;
++ }
+
-+ err = -ENOSYS;
-+ h_file = au_hf_top(file);
-+ if (h_file->f_op->aio_fsync) {
-+ h_inode = file_inode(h_file);
-+ if (!is_sync_kiocb(kio)) {
-+ get_file(h_file);
-+ fput(file);
++ /* swap two lower inode, and loop again */
++ q = fidir->fd_hfile + new_bindex;
++ tmp = *q;
++ *q = *p;
++ *p = tmp;
++ if (tmp.hf_file) {
++ bindex--;
++ p--;
+ }
-+ kio->ki_filp = h_file;
-+ err = h_file->f_op->aio_fsync(kio, datasync);
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
-+ if (!err)
-+ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL);
-+ /*ignore*/
-+ inode_unlock(h_inode);
+ }
-+ au_write_post(inode, h_file, &wpre, /*written*/0);
+
-+out_unlock:
-+ si_read_unlock(inode->sb);
-+ inode_unlock(inode);
-+out:
-+ return err;
++ p = fidir->fd_hfile;
++ if (!au_test_mmapped(file) && !d_unlinked(file->f_path.dentry)) {
++ bbot = au_sbbot(sb);
++ for (finfo->fi_btop = 0; finfo->fi_btop <= bbot;
++ finfo->fi_btop++, p++)
++ if (p->hf_file) {
++ if (file_inode(p->hf_file))
++ break;
++ au_hfput(p, /*execed*/0);
++ }
++ } else {
++ bbot = au_br_index(sb, brid);
++ for (finfo->fi_btop = 0; finfo->fi_btop < bbot;
++ finfo->fi_btop++, p++)
++ if (p->hf_file)
++ au_hfput(p, /*execed*/0);
++ bbot = au_sbbot(sb);
++ }
++
++ p = fidir->fd_hfile + bbot;
++ for (fidir->fd_bbot = bbot; fidir->fd_bbot >= finfo->fi_btop;
++ fidir->fd_bbot--, p--)
++ if (p->hf_file) {
++ if (file_inode(p->hf_file))
++ break;
++ au_hfput(p, /*execed*/0);
++ }
++ AuDebugOn(fidir->fd_bbot < finfo->fi_btop);
+}
-+#endif
+
-+static int aufs_fasync(int fd, struct file *file, int flag)
++/*
++ * after branch manipulating, refresh the file.
++ */
++static int refresh_file(struct file *file, int (*reopen)(struct file *file))
+{
-+ int err;
-+ struct file *h_file;
++ int err, need_reopen, nbr;
++ aufs_bindex_t bbot, bindex;
++ struct dentry *dentry;
+ struct super_block *sb;
++ struct au_finfo *finfo;
++ struct au_hfile *hfile;
+
-+ sb = file->f_path.dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ dentry = file->f_path.dentry;
++ sb = dentry->d_sb;
++ nbr = au_sbbot(sb) + 1;
++ finfo = au_fi(file);
++ if (!finfo->fi_hdir) {
++ hfile = &finfo->fi_htop;
++ AuDebugOn(!hfile->hf_file);
++ bindex = au_br_index(sb, hfile->hf_br->br_id);
++ AuDebugOn(bindex < 0);
++ if (bindex != finfo->fi_btop)
++ au_set_fbtop(file, bindex);
++ } else {
++ err = au_fidir_realloc(finfo, nbr, /*may_shrink*/0);
++ if (unlikely(err))
++ goto out;
++ au_do_refresh_dir(file);
++ }
+
-+ h_file = au_read_pre(file, /*keep_fi*/0);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ err = 0;
++ need_reopen = 1;
++ if (!au_test_mmapped(file))
++ err = au_file_refresh_by_inode(file, &need_reopen);
++ if (finfo->fi_hdir)
++ /* harmless if err */
++ au_fidir_realloc(finfo, nbr, /*may_shrink*/1);
++ if (!err && need_reopen && !d_unlinked(dentry))
++ err = reopen(file);
++ if (!err) {
++ au_update_figen(file);
++ goto out; /* success */
++ }
+
-+ if (h_file->f_op->fasync)
-+ err = h_file->f_op->fasync(fd, h_file, flag);
-+ fput(h_file); /* instead of au_read_post() */
++ /* error, close all lower files */
++ if (finfo->fi_hdir) {
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file); bindex <= bbot; bindex++)
++ au_set_h_fptr(file, bindex, NULL);
++ }
+
+out:
-+ si_read_unlock(sb);
+ return err;
+}
+
-+static int aufs_setfl(struct file *file, unsigned long arg)
++/* common function to regular file and dir */
++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
++ int wlock, unsigned int fi_lsc)
+{
+ int err;
-+ struct file *h_file;
-+ struct super_block *sb;
++ unsigned int sigen, figen;
++ aufs_bindex_t btop;
++ unsigned char pseudo_link;
++ struct dentry *dentry;
++ struct inode *inode;
+
-+ sb = file->f_path.dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ err = 0;
++ dentry = file->f_path.dentry;
++ inode = d_inode(dentry);
++ sigen = au_sigen(dentry->d_sb);
++ fi_write_lock_nested(file, fi_lsc);
++ figen = au_figen(file);
++ if (!fi_lsc)
++ di_write_lock_child(dentry);
++ else
++ di_write_lock_child2(dentry);
++ btop = au_dbtop(dentry);
++ pseudo_link = (btop != au_ibtop(inode));
++ if (sigen == figen && !pseudo_link && au_fbtop(file) == btop) {
++ if (!wlock) {
++ di_downgrade_lock(dentry, AuLock_IR);
++ fi_downgrade_lock(file);
++ }
++ goto out; /* success */
++ }
+
-+ h_file = au_read_pre(file, /*keep_fi*/0);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out;
++ AuDbg("sigen %d, figen %d\n", sigen, figen);
++ if (au_digen_test(dentry, sigen)) {
++ err = au_reval_dpath(dentry, sigen);
++ AuDebugOn(!err && au_digen_test(dentry, sigen));
++ }
+
-+ arg |= vfsub_file_flags(file) & FASYNC; /* stop calling h_file->fasync */
-+ err = setfl(/*unused fd*/-1, h_file, arg);
-+ fput(h_file); /* instead of au_read_post() */
++ if (!err)
++ err = refresh_file(file, reopen);
++ if (!err) {
++ if (!wlock) {
++ di_downgrade_lock(dentry, AuLock_IR);
++ fi_downgrade_lock(file);
++ }
++ } else {
++ di_write_unlock(dentry);
++ fi_write_unlock(file);
++ }
+
+out:
-+ si_read_unlock(sb);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+/* no one supports this operation, currently */
-+#if 0
-+static ssize_t aufs_sendpage(struct file *file, struct page *page, int offset,
-+ size_t len, loff_t *pos, int more)
++/* cf. aufs_nopage() */
++/* for madvise(2) */
++static int aufs_readpage(struct file *file __maybe_unused, struct page *page)
+{
++ unlock_page(page);
++ return 0;
+}
-+#endif
-+
-+/* ---------------------------------------------------------------------- */
-+
-+const struct file_operations aufs_file_fop = {
-+ .owner = THIS_MODULE,
-+
-+ .llseek = default_llseek,
-+
-+ .read = aufs_read,
-+ .write = aufs_write,
-+ .read_iter = aufs_read_iter,
-+ .write_iter = aufs_write_iter,
+
-+#ifdef CONFIG_AUFS_POLL
-+ .poll = aufs_poll,
-+#endif
-+ .unlocked_ioctl = aufs_ioctl_nondir,
-+#ifdef CONFIG_COMPAT
-+ .compat_ioctl = aufs_compat_ioctl_nondir,
-+#endif
-+ .mmap = aufs_mmap,
-+ .open = aufs_open_nondir,
-+ .flush = aufs_flush_nondir,
-+ .release = aufs_release_nondir,
-+ .fsync = aufs_fsync_nondir,
-+ /* .aio_fsync = aufs_aio_fsync_nondir, */
-+ .fasync = aufs_fasync,
-+ /* .sendpage = aufs_sendpage, */
-+ .setfl = aufs_setfl,
-+ .splice_write = aufs_splice_write,
-+ .splice_read = aufs_splice_read,
-+#if 0
-+ .aio_splice_write = aufs_aio_splice_write,
-+ .aio_splice_read = aufs_aio_splice_read,
++/* it will never be called, but necessary to support O_DIRECT */
++static ssize_t aufs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
++{ BUG(); return 0; }
++
++/* they will never be called. */
++#ifdef CONFIG_AUFS_DEBUG
++static int aufs_write_begin(struct file *file, struct address_space *mapping,
++ loff_t pos, unsigned len, unsigned flags,
++ struct page **pagep, void **fsdata)
++{ AuUnsupport(); return 0; }
++static int aufs_write_end(struct file *file, struct address_space *mapping,
++ loff_t pos, unsigned len, unsigned copied,
++ struct page *page, void *fsdata)
++{ AuUnsupport(); return 0; }
++static int aufs_writepage(struct page *page, struct writeback_control *wbc)
++{ AuUnsupport(); return 0; }
++
++static int aufs_set_page_dirty(struct page *page)
++{ AuUnsupport(); return 0; }
++static void aufs_invalidatepage(struct page *page, unsigned int offset,
++ unsigned int length)
++{ AuUnsupport(); }
++static int aufs_releasepage(struct page *page, gfp_t gfp)
++{ AuUnsupport(); return 0; }
++#if 0 /* called by memory compaction regardless file */
++static int aufs_migratepage(struct address_space *mapping, struct page *newpage,
++ struct page *page, enum migrate_mode mode)
++{ AuUnsupport(); return 0; }
+#endif
-+ .fallocate = aufs_fallocate
++static bool aufs_isolate_page(struct page *page, isolate_mode_t mode)
++{ AuUnsupport(); return true; }
++static void aufs_putback_page(struct page *page)
++{ AuUnsupport(); }
++static int aufs_launder_page(struct page *page)
++{ AuUnsupport(); return 0; }
++static int aufs_is_partially_uptodate(struct page *page,
++ unsigned long from,
++ unsigned long count)
++{ AuUnsupport(); return 0; }
++static void aufs_is_dirty_writeback(struct page *page, bool *dirty,
++ bool *writeback)
++{ AuUnsupport(); }
++static int aufs_error_remove_page(struct address_space *mapping,
++ struct page *page)
++{ AuUnsupport(); return 0; }
++static int aufs_swap_activate(struct swap_info_struct *sis, struct file *file,
++ sector_t *span)
++{ AuUnsupport(); return 0; }
++static void aufs_swap_deactivate(struct file *file)
++{ AuUnsupport(); }
++#endif /* CONFIG_AUFS_DEBUG */
++
++const struct address_space_operations aufs_aop = {
++ .readpage = aufs_readpage,
++ .direct_IO = aufs_direct_IO,
++#ifdef CONFIG_AUFS_DEBUG
++ .writepage = aufs_writepage,
++ /* no writepages, because of writepage */
++ .set_page_dirty = aufs_set_page_dirty,
++ /* no readpages, because of readpage */
++ .write_begin = aufs_write_begin,
++ .write_end = aufs_write_end,
++ /* no bmap, no block device */
++ .invalidatepage = aufs_invalidatepage,
++ .releasepage = aufs_releasepage,
++ /* is fallback_migrate_page ok? */
++ /* .migratepage = aufs_migratepage, */
++ .isolate_page = aufs_isolate_page,
++ .putback_page = aufs_putback_page,
++ .launder_page = aufs_launder_page,
++ .is_partially_uptodate = aufs_is_partially_uptodate,
++ .is_dirty_writeback = aufs_is_dirty_writeback,
++ .error_remove_page = aufs_error_remove_page,
++ .swap_activate = aufs_swap_activate,
++ .swap_deactivate = aufs_swap_deactivate
++#endif /* CONFIG_AUFS_DEBUG */
+};
-diff -urN /usr/share/empty/fs/aufs/fstype.h linux/fs/aufs/fstype.h
---- /usr/share/empty/fs/aufs/fstype.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/fstype.h 2016-08-17 18:01:06.145222850 +0200
-@@ -0,0 +1,400 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/file.h linux-4.14/fs/aufs/file.h
+--- linux-4.14/fs/aufs/file.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/file.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,340 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * judging filesystem type
++ * file operations
+ */
+
-+#ifndef __AUFS_FSTYPE_H__
-+#define __AUFS_FSTYPE_H__
++#ifndef __AUFS_FILE_H__
++#define __AUFS_FILE_H__
+
+#ifdef __KERNEL__
+
++#include <linux/file.h>
+#include <linux/fs.h>
-+#include <linux/magic.h>
-+#include <linux/nfs_fs.h>
-+#include <linux/romfs_fs.h>
++#include <linux/mm_types.h>
++#include <linux/poll.h>
++#include "rwsem.h"
+
-+static inline int au_test_aufs(struct super_block *sb)
-+{
-+ return sb->s_magic == AUFS_SUPER_MAGIC;
-+}
++struct au_branch;
++struct au_hfile {
++ struct file *hf_file;
++ struct au_branch *hf_br;
++};
+
-+static inline const char *au_sbtype(struct super_block *sb)
-+{
-+ return sb->s_type->name;
-+}
++struct au_vdir;
++struct au_fidir {
++ aufs_bindex_t fd_bbot;
++ aufs_bindex_t fd_nent;
++ struct au_vdir *fd_vdir_cache;
++ struct au_hfile fd_hfile[];
++};
+
-+static inline int au_test_iso9660(struct super_block *sb __maybe_unused)
++static inline int au_fidir_sz(int nent)
+{
-+#if IS_ENABLED(CONFIG_ISO9660_FS)
-+ return sb->s_magic == ISOFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ AuDebugOn(nent < 0);
++ return sizeof(struct au_fidir) + sizeof(struct au_hfile) * nent;
+}
+
-+static inline int au_test_romfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_ROMFS_FS)
-+ return sb->s_magic == ROMFS_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++struct au_finfo {
++ atomic_t fi_generation;
+
-+static inline int au_test_cramfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_CRAMFS)
-+ return sb->s_magic == CRAMFS_MAGIC;
-+#endif
-+ return 0;
-+}
++ struct au_rwsem fi_rwsem;
++ aufs_bindex_t fi_btop;
+
-+static inline int au_test_nfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_NFS_FS)
-+ return sb->s_magic == NFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++ /* do not union them */
++ struct { /* for non-dir */
++ struct au_hfile fi_htop;
++ atomic_t fi_mmapped;
++ };
++ struct au_fidir *fi_hdir; /* for dir only */
+
-+static inline int au_test_fuse(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_FUSE_FS)
-+ return sb->s_magic == FUSE_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++ struct hlist_bl_node fi_hlist;
++ struct file *fi_file; /* very ugly */
++} ____cacheline_aligned_in_smp;
+
-+static inline int au_test_xfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_XFS_FS)
-+ return sb->s_magic == XFS_SB_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline int au_test_tmpfs(struct super_block *sb __maybe_unused)
-+{
-+#ifdef CONFIG_TMPFS
-+ return sb->s_magic == TMPFS_MAGIC;
-+#else
-+ return 0;
++/* file.c */
++extern const struct address_space_operations aufs_aop;
++unsigned int au_file_roflags(unsigned int flags);
++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
++ struct file *file, int force_wr);
++struct au_do_open_args {
++ int aopen;
++ int (*open)(struct file *file, int flags,
++ struct file *h_file);
++ struct au_fidir *fidir;
++ struct file *h_file;
++};
++int au_do_open(struct file *file, struct au_do_open_args *args);
++int au_reopen_nondir(struct file *file);
++struct au_pin;
++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin);
++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
++ int wlock, unsigned int fi_lsc);
++int au_do_flush(struct file *file, fl_owner_t id,
++ int (*flush)(struct file *file, fl_owner_t id));
++
++/* poll.c */
++#ifdef CONFIG_AUFS_POLL
++unsigned int aufs_poll(struct file *file, poll_table *wait);
+#endif
-+}
+
-+static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_ECRYPT_FS)
-+ return !strcmp(au_sbtype(sb), "ecryptfs");
++#ifdef CONFIG_AUFS_BR_HFSPLUS
++/* hfsplus.c */
++struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
++ int force_wr);
++void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file);
+#else
-+ return 0;
++AuStub(struct file *, au_h_open_pre, return NULL, struct dentry *dentry,
++ aufs_bindex_t bindex, int force_wr)
++AuStubVoid(au_h_open_post, struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file);
+#endif
-+}
+
-+static inline int au_test_ramfs(struct super_block *sb)
-+{
-+ return sb->s_magic == RAMFS_MAGIC;
-+}
++/* f_op.c */
++extern const struct file_operations aufs_file_fop;
++int au_do_open_nondir(struct file *file, int flags, struct file *h_file);
++int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file);
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc);
+
-+static inline int au_test_ubifs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_UBIFS_FS)
-+ return sb->s_magic == UBIFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++/* finfo.c */
++void au_hfput(struct au_hfile *hf, int execed);
++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex,
++ struct file *h_file);
+
-+static inline int au_test_procfs(struct super_block *sb __maybe_unused)
-+{
-+#ifdef CONFIG_PROC_FS
-+ return sb->s_magic == PROC_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++void au_update_figen(struct file *file);
++struct au_fidir *au_fidir_alloc(struct super_block *sb);
++int au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink);
+
-+static inline int au_test_sysfs(struct super_block *sb __maybe_unused)
-+{
-+#ifdef CONFIG_SYSFS
-+ return sb->s_magic == SYSFS_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++void au_fi_init_once(void *_fi);
++void au_finfo_fin(struct file *file);
++int au_finfo_init(struct file *file, struct au_fidir *fidir);
+
-+static inline int au_test_configfs(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_CONFIGFS_FS)
-+ return sb->s_magic == CONFIGFS_MAGIC;
-+#else
-+ return 0;
++/* ioctl.c */
++long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg);
++#ifdef CONFIG_COMPAT
++long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
++ unsigned long arg);
++long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
++ unsigned long arg);
+#endif
-+}
+
-+static inline int au_test_minix(struct super_block *sb __maybe_unused)
-+{
-+#if IS_ENABLED(CONFIG_MINIX_FS)
-+ return sb->s_magic == MINIX3_SUPER_MAGIC
-+ || sb->s_magic == MINIX2_SUPER_MAGIC
-+ || sb->s_magic == MINIX2_SUPER_MAGIC2
-+ || sb->s_magic == MINIX_SUPER_MAGIC
-+ || sb->s_magic == MINIX_SUPER_MAGIC2;
-+#else
-+ return 0;
-+#endif
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline int au_test_fat(struct super_block *sb __maybe_unused)
++static inline struct au_finfo *au_fi(struct file *file)
+{
-+#if IS_ENABLED(CONFIG_FAT_FS)
-+ return sb->s_magic == MSDOS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ return file->private_data;
+}
+
-+static inline int au_test_msdos(struct super_block *sb)
-+{
-+ return au_test_fat(sb);
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline int au_test_vfat(struct super_block *sb)
-+{
-+ return au_test_fat(sb);
-+}
++#define fi_read_lock(f) au_rw_read_lock(&au_fi(f)->fi_rwsem)
++#define fi_write_lock(f) au_rw_write_lock(&au_fi(f)->fi_rwsem)
++#define fi_read_trylock(f) au_rw_read_trylock(&au_fi(f)->fi_rwsem)
++#define fi_write_trylock(f) au_rw_write_trylock(&au_fi(f)->fi_rwsem)
++/*
++#define fi_read_trylock_nested(f) \
++ au_rw_read_trylock_nested(&au_fi(f)->fi_rwsem)
++#define fi_write_trylock_nested(f) \
++ au_rw_write_trylock_nested(&au_fi(f)->fi_rwsem)
++*/
+
-+static inline int au_test_securityfs(struct super_block *sb __maybe_unused)
-+{
-+#ifdef CONFIG_SECURITYFS
-+ return sb->s_magic == SECURITYFS_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++#define fi_read_unlock(f) au_rw_read_unlock(&au_fi(f)->fi_rwsem)
++#define fi_write_unlock(f) au_rw_write_unlock(&au_fi(f)->fi_rwsem)
++#define fi_downgrade_lock(f) au_rw_dgrade_lock(&au_fi(f)->fi_rwsem)
+
-+static inline int au_test_squashfs(struct super_block *sb __maybe_unused)
++/* lock subclass for finfo */
++enum {
++ AuLsc_FI_1,
++ AuLsc_FI_2
++};
++
++static inline void fi_read_lock_nested(struct file *f, unsigned int lsc)
+{
-+#if IS_ENABLED(CONFIG_SQUASHFS)
-+ return sb->s_magic == SQUASHFS_MAGIC;
-+#else
-+ return 0;
-+#endif
++ au_rw_read_lock_nested(&au_fi(f)->fi_rwsem, lsc);
+}
+
-+static inline int au_test_btrfs(struct super_block *sb __maybe_unused)
++static inline void fi_write_lock_nested(struct file *f, unsigned int lsc)
+{
-+#if IS_ENABLED(CONFIG_BTRFS_FS)
-+ return sb->s_magic == BTRFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ au_rw_write_lock_nested(&au_fi(f)->fi_rwsem, lsc);
+}
+
-+static inline int au_test_xenfs(struct super_block *sb __maybe_unused)
++/*
++ * fi_read_lock_1, fi_write_lock_1,
++ * fi_read_lock_2, fi_write_lock_2
++ */
++#define AuReadLockFunc(name) \
++static inline void fi_read_lock_##name(struct file *f) \
++{ fi_read_lock_nested(f, AuLsc_FI_##name); }
++
++#define AuWriteLockFunc(name) \
++static inline void fi_write_lock_##name(struct file *f) \
++{ fi_write_lock_nested(f, AuLsc_FI_##name); }
++
++#define AuRWLockFuncs(name) \
++ AuReadLockFunc(name) \
++ AuWriteLockFunc(name)
++
++AuRWLockFuncs(1);
++AuRWLockFuncs(2);
++
++#undef AuReadLockFunc
++#undef AuWriteLockFunc
++#undef AuRWLockFuncs
++
++#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem)
++#define FiMustAnyLock(f) AuRwMustAnyLock(&au_fi(f)->fi_rwsem)
++#define FiMustWriteLock(f) AuRwMustWriteLock(&au_fi(f)->fi_rwsem)
++
++/* ---------------------------------------------------------------------- */
++
++/* todo: hard/soft set? */
++static inline aufs_bindex_t au_fbtop(struct file *file)
+{
-+#if IS_ENABLED(CONFIG_XENFS)
-+ return sb->s_magic == XENFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ FiMustAnyLock(file);
++ return au_fi(file)->fi_btop;
+}
+
-+static inline int au_test_debugfs(struct super_block *sb __maybe_unused)
++static inline aufs_bindex_t au_fbbot_dir(struct file *file)
+{
-+#ifdef CONFIG_DEBUG_FS
-+ return sb->s_magic == DEBUGFS_MAGIC;
-+#else
-+ return 0;
-+#endif
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_bbot;
+}
+
-+static inline int au_test_nilfs(struct super_block *sb __maybe_unused)
++static inline struct au_vdir *au_fvdir_cache(struct file *file)
+{
-+#if IS_ENABLED(CONFIG_NILFS)
-+ return sb->s_magic == NILFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_vdir_cache;
+}
+
-+static inline int au_test_hfsplus(struct super_block *sb __maybe_unused)
++static inline void au_set_fbtop(struct file *file, aufs_bindex_t bindex)
+{
-+#if IS_ENABLED(CONFIG_HFSPLUS_FS)
-+ return sb->s_magic == HFSPLUS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
++ FiMustWriteLock(file);
++ au_fi(file)->fi_btop = bindex;
+}
+
-+/* ---------------------------------------------------------------------- */
-+/*
-+ * they can't be an aufs branch.
-+ */
-+static inline int au_test_fs_unsuppoted(struct super_block *sb)
++static inline void au_set_fbbot_dir(struct file *file, aufs_bindex_t bindex)
+{
-+ return
-+#ifndef CONFIG_AUFS_BR_RAMFS
-+ au_test_ramfs(sb) ||
-+#endif
-+ au_test_procfs(sb)
-+ || au_test_sysfs(sb)
-+ || au_test_configfs(sb)
-+ || au_test_debugfs(sb)
-+ || au_test_securityfs(sb)
-+ || au_test_xenfs(sb)
-+ || au_test_ecryptfs(sb)
-+ /* || !strcmp(au_sbtype(sb), "unionfs") */
-+ || au_test_aufs(sb); /* will be supported in next version */
++ FiMustWriteLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ au_fi(file)->fi_hdir->fd_bbot = bindex;
+}
+
-+static inline int au_test_fs_remote(struct super_block *sb)
++static inline void au_set_fvdir_cache(struct file *file,
++ struct au_vdir *vdir_cache)
+{
-+ return !au_test_tmpfs(sb)
-+#ifdef CONFIG_AUFS_BR_RAMFS
-+ && !au_test_ramfs(sb)
-+#endif
-+ && !(sb->s_type->fs_flags & FS_REQUIRES_DEV);
++ FiMustWriteLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ au_fi(file)->fi_hdir->fd_vdir_cache = vdir_cache;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * Note: these functions (below) are created after reading ->getattr() in all
-+ * filesystems under linux/fs. it means we have to do so in every update...
-+ */
-+
-+/*
-+ * some filesystems require getattr to refresh the inode attributes before
-+ * referencing.
-+ * in most cases, we can rely on the inode attribute in NFS (or every remote fs)
-+ * and leave the work for d_revalidate()
-+ */
-+static inline int au_test_fs_refresh_iattr(struct super_block *sb)
++static inline struct file *au_hf_top(struct file *file)
+{
-+ return au_test_nfs(sb)
-+ || au_test_fuse(sb)
-+ /* || au_test_btrfs(sb) */ /* untested */
-+ ;
++ FiMustAnyLock(file);
++ AuDebugOn(au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_htop.hf_file;
+}
+
-+/*
-+ * filesystems which don't maintain i_size or i_blocks.
-+ */
-+static inline int au_test_fs_bad_iattr_size(struct super_block *sb)
++static inline struct file *au_hf_dir(struct file *file, aufs_bindex_t bindex)
+{
-+ return au_test_xfs(sb)
-+ || au_test_btrfs(sb)
-+ || au_test_ubifs(sb)
-+ || au_test_hfsplus(sb) /* maintained, but incorrect */
-+ /* || au_test_minix(sb) */ /* untested */
-+ ;
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_hfile[0 + bindex].hf_file;
+}
+
-+/*
-+ * filesystems which don't store the correct value in some of their inode
-+ * attributes.
-+ */
-+static inline int au_test_fs_bad_iattr(struct super_block *sb)
++/* todo: memory barrier? */
++static inline unsigned int au_figen(struct file *f)
+{
-+ return au_test_fs_bad_iattr_size(sb)
-+ || au_test_fat(sb)
-+ || au_test_msdos(sb)
-+ || au_test_vfat(sb);
++ return atomic_read(&au_fi(f)->fi_generation);
+}
+
-+/* they don't check i_nlink in link(2) */
-+static inline int au_test_fs_no_limit_nlink(struct super_block *sb)
++static inline void au_set_mmapped(struct file *f)
+{
-+ return au_test_tmpfs(sb)
-+#ifdef CONFIG_AUFS_BR_RAMFS
-+ || au_test_ramfs(sb)
-+#endif
-+ || au_test_ubifs(sb)
-+ || au_test_hfsplus(sb);
++ if (atomic_inc_return(&au_fi(f)->fi_mmapped))
++ return;
++ pr_warn("fi_mmapped wrapped around\n");
++ while (!atomic_inc_return(&au_fi(f)->fi_mmapped))
++ ;
+}
+
-+/*
-+ * filesystems which sets S_NOATIME and S_NOCMTIME.
-+ */
-+static inline int au_test_fs_notime(struct super_block *sb)
++static inline void au_unset_mmapped(struct file *f)
+{
-+ return au_test_nfs(sb)
-+ || au_test_fuse(sb)
-+ || au_test_ubifs(sb)
-+ ;
++ atomic_dec(&au_fi(f)->fi_mmapped);
+}
+
-+/* temporary support for i#1 in cramfs */
-+static inline int au_test_fs_unique_ino(struct inode *inode)
++static inline int au_test_mmapped(struct file *f)
+{
-+ if (au_test_cramfs(inode->i_sb))
-+ return inode->i_ino != 1;
-+ return 1;
++ return atomic_read(&au_fi(f)->fi_mmapped);
+}
+
-+/* ---------------------------------------------------------------------- */
++/* customize vma->vm_file */
+
-+/*
-+ * the filesystem where the xino files placed must support i/o after unlink and
-+ * maintain i_size and i_blocks.
-+ */
-+static inline int au_test_fs_bad_xino(struct super_block *sb)
++static inline void au_do_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
+{
-+ return au_test_fs_remote(sb)
-+ || au_test_fs_bad_iattr_size(sb)
-+ /* don't want unnecessary work for xino */
-+ || au_test_aufs(sb)
-+ || au_test_ecryptfs(sb)
-+ || au_test_nilfs(sb);
++ struct file *f;
++
++ f = vma->vm_file;
++ get_file(file);
++ vma->vm_file = file;
++ fput(f);
+}
+
-+static inline int au_test_fs_trunc_xino(struct super_block *sb)
++#ifdef CONFIG_MMU
++#define AuDbgVmRegion(file, vma) do {} while (0)
++
++static inline void au_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
+{
-+ return au_test_tmpfs(sb)
-+ || au_test_ramfs(sb);
++ au_do_vm_file_reset(vma, file);
+}
++#else
++#define AuDbgVmRegion(file, vma) \
++ AuDebugOn((vma)->vm_region && (vma)->vm_region->vm_file != (file))
+
-+/*
-+ * test if the @sb is real-readonly.
-+ */
-+static inline int au_test_fs_rr(struct super_block *sb)
++static inline void au_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
+{
-+ return au_test_squashfs(sb)
-+ || au_test_iso9660(sb)
-+ || au_test_cramfs(sb)
-+ || au_test_romfs(sb);
++ struct file *f;
++
++ au_do_vm_file_reset(vma, file);
++ f = vma->vm_region->vm_file;
++ get_file(file);
++ vma->vm_region->vm_file = file;
++ fput(f);
+}
++#endif /* CONFIG_MMU */
+
-+/*
-+ * test if the @inode is nfs with 'noacl' option
-+ * NFS always sets MS_POSIXACL regardless its mount option 'noacl.'
-+ */
-+static inline int au_test_nfs_noacl(struct inode *inode)
++/* handle vma->vm_prfile */
++static inline void au_vm_prfile_set(struct vm_area_struct *vma,
++ struct file *file)
+{
-+ return au_test_nfs(inode->i_sb)
-+ /* && IS_POSIXACL(inode) */
-+ && !nfs_server_capable(inode, NFS_CAP_ACLS);
++ get_file(file);
++ vma->vm_prfile = file;
++#ifndef CONFIG_MMU
++ get_file(file);
++ vma->vm_region->vm_prfile = file;
++#endif
+}
+
+#endif /* __KERNEL__ */
-+#endif /* __AUFS_FSTYPE_H__ */
-diff -urN /usr/share/empty/fs/aufs/hfsnotify.c linux/fs/aufs/hfsnotify.c
---- /usr/share/empty/fs/aufs/hfsnotify.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/hfsnotify.c 2016-08-17 18:01:06.145222850 +0200
-@@ -0,0 +1,287 @@
++#endif /* __AUFS_FILE_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/finfo.c linux-4.14/fs/aufs/finfo.c
+--- linux-4.14/fs/aufs/finfo.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/finfo.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,148 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * fsnotify for the lower directories
++ * file private data
+ */
+
+#include "aufs.h"
+
-+/* FS_IN_IGNORED is unnecessary */
-+static const __u32 AuHfsnMask = (FS_MOVED_TO | FS_MOVED_FROM | FS_DELETE
-+ | FS_CREATE | FS_EVENT_ON_CHILD);
-+static DECLARE_WAIT_QUEUE_HEAD(au_hfsn_wq);
-+static __cacheline_aligned_in_smp atomic64_t au_hfsn_ifree = ATOMIC64_INIT(0);
-+
-+static void au_hfsn_free_mark(struct fsnotify_mark *mark)
++void au_hfput(struct au_hfile *hf, int execed)
+{
-+ struct au_hnotify *hn = container_of(mark, struct au_hnotify,
-+ hn_mark);
-+ /* AuDbg("here\n"); */
-+ au_cache_dfree_hnotify(hn);
-+ smp_mb__before_atomic();
-+ if (atomic64_dec_and_test(&au_hfsn_ifree))
-+ wake_up(&au_hfsn_wq);
++ if (execed)
++ allow_write_access(hf->hf_file);
++ fput(hf->hf_file);
++ hf->hf_file = NULL;
++ au_br_put(hf->hf_br);
++ hf->hf_br = NULL;
+}
+
-+static int au_hfsn_alloc(struct au_hinode *hinode)
++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val)
+{
-+ int err;
-+ struct au_hnotify *hn;
-+ struct super_block *sb;
-+ struct au_branch *br;
-+ struct fsnotify_mark *mark;
-+ aufs_bindex_t bindex;
-+
-+ hn = hinode->hi_notify;
-+ sb = hn->hn_aufs_inode->i_sb;
-+ bindex = au_br_index(sb, hinode->hi_id);
-+ br = au_sbr(sb, bindex);
-+ AuDebugOn(!br->br_hfsn);
++ struct au_finfo *finfo = au_fi(file);
++ struct au_hfile *hf;
++ struct au_fidir *fidir;
+
-+ mark = &hn->hn_mark;
-+ fsnotify_init_mark(mark, au_hfsn_free_mark);
-+ mark->mask = AuHfsnMask;
-+ /*
-+ * by udba rename or rmdir, aufs assign a new inode to the known
-+ * h_inode, so specify 1 to allow dups.
-+ */
-+ lockdep_off();
-+ err = fsnotify_add_mark(mark, br->br_hfsn->hfsn_group, hinode->hi_inode,
-+ /*mnt*/NULL, /*allow_dups*/1);
-+ lockdep_on();
++ fidir = finfo->fi_hdir;
++ if (!fidir) {
++ AuDebugOn(finfo->fi_btop != bindex);
++ hf = &finfo->fi_htop;
++ } else
++ hf = fidir->fd_hfile + bindex;
+
-+ return err;
++ if (hf && hf->hf_file)
++ au_hfput(hf, vfsub_file_execed(file));
++ if (val) {
++ FiMustWriteLock(file);
++ AuDebugOn(IS_ERR_OR_NULL(file->f_path.dentry));
++ hf->hf_file = val;
++ hf->hf_br = au_sbr(file->f_path.dentry->d_sb, bindex);
++ }
+}
+
-+static int au_hfsn_free(struct au_hinode *hinode, struct au_hnotify *hn)
++void au_update_figen(struct file *file)
+{
-+ struct fsnotify_mark *mark;
-+ unsigned long long ull;
-+ struct fsnotify_group *group;
-+
-+ ull = atomic64_inc_return(&au_hfsn_ifree);
-+ BUG_ON(!ull);
-+
-+ mark = &hn->hn_mark;
-+ spin_lock(&mark->lock);
-+ group = mark->group;
-+ fsnotify_get_group(group);
-+ spin_unlock(&mark->lock);
-+ lockdep_off();
-+ fsnotify_destroy_mark(mark, group);
-+ fsnotify_put_mark(mark);
-+ fsnotify_put_group(group);
-+ lockdep_on();
-+
-+ /* free hn by myself */
-+ return 0;
++ atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_path.dentry));
++ /* smp_mb(); */ /* atomic_set */
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static void au_hfsn_ctl(struct au_hinode *hinode, int do_set)
++struct au_fidir *au_fidir_alloc(struct super_block *sb)
+{
-+ struct fsnotify_mark *mark;
++ struct au_fidir *fidir;
++ int nbr;
+
-+ mark = &hinode->hi_notify->hn_mark;
-+ spin_lock(&mark->lock);
-+ if (do_set) {
-+ AuDebugOn(mark->mask & AuHfsnMask);
-+ mark->mask |= AuHfsnMask;
-+ } else {
-+ AuDebugOn(!(mark->mask & AuHfsnMask));
-+ mark->mask &= ~AuHfsnMask;
++ nbr = au_sbbot(sb) + 1;
++ if (nbr < 2)
++ nbr = 2; /* initial allocate for 2 branches */
++ fidir = kzalloc(au_fidir_sz(nbr), GFP_NOFS);
++ if (fidir) {
++ fidir->fd_bbot = -1;
++ fidir->fd_nent = nbr;
+ }
-+ spin_unlock(&mark->lock);
-+ /* fsnotify_recalc_inode_mask(hinode->hi_inode); */
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/* #define AuDbgHnotify */
-+#ifdef AuDbgHnotify
-+static char *au_hfsn_name(u32 mask)
-+{
-+#ifdef CONFIG_AUFS_DEBUG
-+#define test_ret(flag) \
-+ do { \
-+ if (mask & flag) \
-+ return #flag; \
-+ } while (0)
-+ test_ret(FS_ACCESS);
-+ test_ret(FS_MODIFY);
-+ test_ret(FS_ATTRIB);
-+ test_ret(FS_CLOSE_WRITE);
-+ test_ret(FS_CLOSE_NOWRITE);
-+ test_ret(FS_OPEN);
-+ test_ret(FS_MOVED_FROM);
-+ test_ret(FS_MOVED_TO);
-+ test_ret(FS_CREATE);
-+ test_ret(FS_DELETE);
-+ test_ret(FS_DELETE_SELF);
-+ test_ret(FS_MOVE_SELF);
-+ test_ret(FS_UNMOUNT);
-+ test_ret(FS_Q_OVERFLOW);
-+ test_ret(FS_IN_IGNORED);
-+ test_ret(FS_ISDIR);
-+ test_ret(FS_IN_ONESHOT);
-+ test_ret(FS_EVENT_ON_CHILD);
-+ return "";
-+#undef test_ret
-+#else
-+ return "??";
-+#endif
-+}
-+#endif
-+
-+/* ---------------------------------------------------------------------- */
-+
-+static void au_hfsn_free_group(struct fsnotify_group *group)
-+{
-+ struct au_br_hfsnotify *hfsn = group->private;
+
-+ /* AuDbg("here\n"); */
-+ au_delayed_kfree(hfsn);
++ return fidir;
+}
+
-+static int au_hfsn_handle_event(struct fsnotify_group *group,
-+ struct inode *inode,
-+ struct fsnotify_mark *inode_mark,
-+ struct fsnotify_mark *vfsmount_mark,
-+ u32 mask, void *data, int data_type,
-+ const unsigned char *file_name, u32 cookie)
++int au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink)
+{
+ int err;
-+ struct au_hnotify *hnotify;
-+ struct inode *h_dir, *h_inode;
-+ struct qstr h_child_qstr = QSTR_INIT(file_name, strlen(file_name));
-+
-+ AuDebugOn(data_type != FSNOTIFY_EVENT_INODE);
++ struct au_fidir *fidir, *p;
+
-+ err = 0;
-+ /* if FS_UNMOUNT happens, there must be another bug */
-+ AuDebugOn(mask & FS_UNMOUNT);
-+ if (mask & (FS_IN_IGNORED | FS_UNMOUNT))
-+ goto out;
++ AuRwMustWriteLock(&finfo->fi_rwsem);
++ fidir = finfo->fi_hdir;
++ AuDebugOn(!fidir);
+
-+ h_dir = inode;
-+ h_inode = NULL;
-+#ifdef AuDbgHnotify
-+ au_debug_on();
-+ if (1 || h_child_qstr.len != sizeof(AUFS_XINO_FNAME) - 1
-+ || strncmp(h_child_qstr.name, AUFS_XINO_FNAME, h_child_qstr.len)) {
-+ AuDbg("i%lu, mask 0x%x %s, hcname %.*s, hi%lu\n",
-+ h_dir->i_ino, mask, au_hfsn_name(mask),
-+ AuLNPair(&h_child_qstr), h_inode ? h_inode->i_ino : 0);
-+ /* WARN_ON(1); */
++ err = -ENOMEM;
++ p = au_kzrealloc(fidir, au_fidir_sz(fidir->fd_nent), au_fidir_sz(nbr),
++ GFP_NOFS, may_shrink);
++ if (p) {
++ p->fd_nent = nbr;
++ finfo->fi_hdir = p;
++ err = 0;
+ }
-+ au_debug_off();
-+#endif
-+
-+ AuDebugOn(!inode_mark);
-+ hnotify = container_of(inode_mark, struct au_hnotify, hn_mark);
-+ err = au_hnotify(h_dir, hnotify, mask, &h_child_qstr, h_inode);
+
-+out:
+ return err;
+}
+
-+static struct fsnotify_ops au_hfsn_ops = {
-+ .handle_event = au_hfsn_handle_event,
-+ .free_group_priv = au_hfsn_free_group
-+};
-+
+/* ---------------------------------------------------------------------- */
+
-+static void au_hfsn_fin_br(struct au_branch *br)
++void au_finfo_fin(struct file *file)
+{
-+ struct au_br_hfsnotify *hfsn;
++ struct au_finfo *finfo;
+
-+ hfsn = br->br_hfsn;
-+ if (hfsn) {
-+ lockdep_off();
-+ fsnotify_put_group(hfsn->hfsn_group);
-+ lockdep_on();
-+ }
++ au_nfiles_dec(file->f_path.dentry->d_sb);
++
++ finfo = au_fi(file);
++ AuDebugOn(finfo->fi_hdir);
++ AuRwDestroy(&finfo->fi_rwsem);
++ au_cache_free_finfo(finfo);
+}
+
-+static int au_hfsn_init_br(struct au_branch *br, int perm)
++void au_fi_init_once(void *_finfo)
+{
-+ int err;
-+ struct fsnotify_group *group;
-+ struct au_br_hfsnotify *hfsn;
++ struct au_finfo *finfo = _finfo;
+
-+ err = 0;
-+ br->br_hfsn = NULL;
-+ if (!au_br_hnotifyable(perm))
-+ goto out;
++ au_rw_init(&finfo->fi_rwsem);
++}
++
++int au_finfo_init(struct file *file, struct au_fidir *fidir)
++{
++ int err;
++ struct au_finfo *finfo;
++ struct dentry *dentry;
+
+ err = -ENOMEM;
-+ hfsn = kmalloc(sizeof(*hfsn), GFP_NOFS);
-+ if (unlikely(!hfsn))
++ dentry = file->f_path.dentry;
++ finfo = au_cache_alloc_finfo();
++ if (unlikely(!finfo))
+ goto out;
+
+ err = 0;
-+ group = fsnotify_alloc_group(&au_hfsn_ops);
-+ if (IS_ERR(group)) {
-+ err = PTR_ERR(group);
-+ pr_err("fsnotify_alloc_group() failed, %d\n", err);
-+ goto out_hfsn;
-+ }
++ au_nfiles_inc(dentry->d_sb);
++ au_rw_write_lock(&finfo->fi_rwsem);
++ finfo->fi_btop = -1;
++ finfo->fi_hdir = fidir;
++ atomic_set(&finfo->fi_generation, au_digen(dentry));
++ /* smp_mb(); */ /* atomic_set */
+
-+ group->private = hfsn;
-+ hfsn->hfsn_group = group;
-+ br->br_hfsn = hfsn;
-+ goto out; /* success */
++ file->private_data = finfo;
+
-+out_hfsn:
-+ au_delayed_kfree(hfsn);
+out:
+ return err;
+}
-+
-+static int au_hfsn_reset_br(unsigned int udba, struct au_branch *br, int perm)
-+{
-+ int err;
-+
-+ err = 0;
-+ if (!br->br_hfsn)
-+ err = au_hfsn_init_br(br, perm);
-+
-+ return err;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+static void au_hfsn_fin(void)
-+{
-+ AuDbg("au_hfsn_ifree %lld\n", (long long)atomic64_read(&au_hfsn_ifree));
-+ wait_event(au_hfsn_wq, !atomic64_read(&au_hfsn_ifree));
-+}
-+
-+const struct au_hnotify_op au_hnotify_op = {
-+ .ctl = au_hfsn_ctl,
-+ .alloc = au_hfsn_alloc,
-+ .free = au_hfsn_free,
-+
-+ .fin = au_hfsn_fin,
-+
-+ .reset_br = au_hfsn_reset_br,
-+ .fin_br = au_hfsn_fin_br,
-+ .init_br = au_hfsn_init_br
-+};
-diff -urN /usr/share/empty/fs/aufs/hfsplus.c linux/fs/aufs/hfsplus.c
---- /usr/share/empty/fs/aufs/hfsplus.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/hfsplus.c 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,56 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/fstype.h linux-4.14/fs/aufs/fstype.h
+--- linux-4.14/fs/aufs/fstype.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/fstype.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,400 @@
+/*
-+ * Copyright (C) 2010-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * special support for filesystems which aqucires an inode mutex
-+ * at final closing a file, eg, hfsplus.
-+ *
-+ * This trick is very simple and stupid, just to open the file before really
-+ * neceeary open to tell hfsplus that this is not the final closing.
-+ * The caller should call au_h_open_pre() after acquiring the inode mutex,
-+ * and au_h_open_post() after releasing it.
++ * judging filesystem type
+ */
+
-+#include "aufs.h"
++#ifndef __AUFS_FSTYPE_H__
++#define __AUFS_FSTYPE_H__
+
-+struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
-+ int force_wr)
++#ifdef __KERNEL__
++
++#include <linux/fs.h>
++#include <linux/magic.h>
++#include <linux/nfs_fs.h>
++#include <linux/romfs_fs.h>
++
++static inline int au_test_aufs(struct super_block *sb)
+{
-+ struct file *h_file;
-+ struct dentry *h_dentry;
++ return sb->s_magic == AUFS_SUPER_MAGIC;
++}
+
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ AuDebugOn(!h_dentry);
-+ AuDebugOn(d_is_negative(h_dentry));
++static inline const char *au_sbtype(struct super_block *sb)
++{
++ return sb->s_type->name;
++}
+
-+ h_file = NULL;
-+ if (au_test_hfsplus(h_dentry->d_sb)
-+ && d_is_reg(h_dentry))
-+ h_file = au_h_open(dentry, bindex,
-+ O_RDONLY | O_NOATIME | O_LARGEFILE,
-+ /*file*/NULL, force_wr);
-+ return h_file;
++static inline int au_test_iso9660(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_ISO9660_FS)
++ return sb->s_magic == ISOFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *h_file)
++static inline int au_test_romfs(struct super_block *sb __maybe_unused)
+{
-+ if (h_file) {
-+ fput(h_file);
-+ au_sbr_put(dentry->d_sb, bindex);
-+ }
++#if IS_ENABLED(CONFIG_ROMFS_FS)
++ return sb->s_magic == ROMFS_MAGIC;
++#else
++ return 0;
++#endif
+}
-diff -urN /usr/share/empty/fs/aufs/hnotify.c linux/fs/aufs/hnotify.c
---- /usr/share/empty/fs/aufs/hnotify.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/hnotify.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,723 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
+
-+/*
-+ * abstraction to notify the direct changes on lower directories
-+ */
++static inline int au_test_cramfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_CRAMFS)
++ return sb->s_magic == CRAMFS_MAGIC;
++#endif
++ return 0;
++}
+
-+#include "aufs.h"
++static inline int au_test_nfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_NFS_FS)
++ return sb->s_magic == NFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+int au_hn_alloc(struct au_hinode *hinode, struct inode *inode)
++static inline int au_test_fuse(struct super_block *sb __maybe_unused)
+{
-+ int err;
-+ struct au_hnotify *hn;
++#if IS_ENABLED(CONFIG_FUSE_FS)
++ return sb->s_magic == FUSE_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ err = -ENOMEM;
-+ hn = au_cache_alloc_hnotify();
-+ if (hn) {
-+ hn->hn_aufs_inode = inode;
-+ hinode->hi_notify = hn;
-+ err = au_hnotify_op.alloc(hinode);
-+ AuTraceErr(err);
-+ if (unlikely(err)) {
-+ hinode->hi_notify = NULL;
-+ au_cache_dfree_hnotify(hn);
-+ /*
-+ * The upper dir was removed by udba, but the same named
-+ * dir left. In this case, aufs assignes a new inode
-+ * number and set the monitor again.
-+ * For the lower dir, the old monitnor is still left.
-+ */
-+ if (err == -EEXIST)
-+ err = 0;
-+ }
-+ }
++static inline int au_test_xfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_XFS_FS)
++ return sb->s_magic == XFS_SB_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ AuTraceErr(err);
-+ return err;
++static inline int au_test_tmpfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_TMPFS
++ return sb->s_magic == TMPFS_MAGIC;
++#else
++ return 0;
++#endif
++}
++
++static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_ECRYPT_FS)
++ return !strcmp(au_sbtype(sb), "ecryptfs");
++#else
++ return 0;
++#endif
++}
++
++static inline int au_test_ramfs(struct super_block *sb)
++{
++ return sb->s_magic == RAMFS_MAGIC;
++}
++
++static inline int au_test_ubifs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_UBIFS_FS)
++ return sb->s_magic == UBIFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
++
++static inline int au_test_procfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_PROC_FS
++ return sb->s_magic == PROC_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
++
++static inline int au_test_sysfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_SYSFS
++ return sb->s_magic == SYSFS_MAGIC;
++#else
++ return 0;
++#endif
++}
++
++static inline int au_test_configfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_CONFIGFS_FS)
++ return sb->s_magic == CONFIGFS_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+void au_hn_free(struct au_hinode *hinode)
++static inline int au_test_minix(struct super_block *sb __maybe_unused)
+{
-+ struct au_hnotify *hn;
-+
-+ hn = hinode->hi_notify;
-+ if (hn) {
-+ hinode->hi_notify = NULL;
-+ if (au_hnotify_op.free(hinode, hn))
-+ au_cache_dfree_hnotify(hn);
-+ }
++#if IS_ENABLED(CONFIG_MINIX_FS)
++ return sb->s_magic == MINIX3_SUPER_MAGIC
++ || sb->s_magic == MINIX2_SUPER_MAGIC
++ || sb->s_magic == MINIX2_SUPER_MAGIC2
++ || sb->s_magic == MINIX_SUPER_MAGIC
++ || sb->s_magic == MINIX_SUPER_MAGIC2;
++#else
++ return 0;
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_hn_ctl(struct au_hinode *hinode, int do_set)
++static inline int au_test_fat(struct super_block *sb __maybe_unused)
+{
-+ if (hinode->hi_notify)
-+ au_hnotify_op.ctl(hinode, do_set);
++#if IS_ENABLED(CONFIG_FAT_FS)
++ return sb->s_magic == MSDOS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+void au_hn_reset(struct inode *inode, unsigned int flags)
++static inline int au_test_msdos(struct super_block *sb)
+{
-+ aufs_bindex_t bindex, bbot;
-+ struct inode *hi;
-+ struct dentry *iwhdentry;
-+
-+ bbot = au_ibbot(inode);
-+ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
-+ hi = au_h_iptr(inode, bindex);
-+ if (!hi)
-+ continue;
-+
-+ /* inode_lock_nested(hi, AuLsc_I_CHILD); */
-+ iwhdentry = au_hi_wh(inode, bindex);
-+ if (iwhdentry)
-+ dget(iwhdentry);
-+ au_igrab(hi);
-+ au_set_h_iptr(inode, bindex, NULL, 0);
-+ au_set_h_iptr(inode, bindex, au_igrab(hi),
-+ flags & ~AuHi_XINO);
-+ iput(hi);
-+ dput(iwhdentry);
-+ /* inode_unlock(hi); */
-+ }
++ return au_test_fat(sb);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static int hn_xino(struct inode *inode, struct inode *h_inode)
++static inline int au_test_vfat(struct super_block *sb)
+{
-+ int err;
-+ aufs_bindex_t bindex, bbot, bfound, btop;
-+ struct inode *h_i;
-+
-+ err = 0;
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ pr_warn("branch root dir was changed\n");
-+ goto out;
-+ }
++ return au_test_fat(sb);
++}
+
-+ bfound = -1;
-+ bbot = au_ibbot(inode);
-+ btop = au_ibtop(inode);
-+#if 0 /* reserved for future use */
-+ if (bindex == bbot) {
-+ /* keep this ino in rename case */
-+ goto out;
-+ }
++static inline int au_test_securityfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_SECURITYFS
++ return sb->s_magic == SECURITYFS_MAGIC;
++#else
++ return 0;
+#endif
-+ for (bindex = btop; bindex <= bbot; bindex++)
-+ if (au_h_iptr(inode, bindex) == h_inode) {
-+ bfound = bindex;
-+ break;
-+ }
-+ if (bfound < 0)
-+ goto out;
-+
-+ for (bindex = btop; bindex <= bbot; bindex++) {
-+ h_i = au_h_iptr(inode, bindex);
-+ if (!h_i)
-+ continue;
-+
-+ err = au_xino_write(inode->i_sb, bindex, h_i->i_ino, /*ino*/0);
-+ /* ignore this error */
-+ /* bad action? */
-+ }
-+
-+ /* children inode number will be broken */
-+
-+out:
-+ AuTraceErr(err);
-+ return err;
+}
+
-+static int hn_gen_tree(struct dentry *dentry)
++static inline int au_test_squashfs(struct super_block *sb __maybe_unused)
+{
-+ int err, i, j, ndentry;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
-+ struct dentry **dentries;
-+
-+ err = au_dpages_init(&dpages, GFP_NOFS);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_dcsub_pages(&dpages, dentry, NULL, NULL);
-+ if (unlikely(err))
-+ goto out_dpages;
++#if IS_ENABLED(CONFIG_SQUASHFS)
++ return sb->s_magic == SQUASHFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ for (i = 0; i < dpages.ndpage; i++) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ ndentry = dpage->ndentry;
-+ for (j = 0; j < ndentry; j++) {
-+ struct dentry *d;
++static inline int au_test_btrfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_BTRFS_FS)
++ return sb->s_magic == BTRFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ d = dentries[j];
-+ if (IS_ROOT(d))
-+ continue;
++static inline int au_test_xenfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_XENFS)
++ return sb->s_magic == XENFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ au_digen_dec(d);
-+ if (d_really_is_positive(d))
-+ /* todo: reset children xino?
-+ cached children only? */
-+ au_iigen_dec(d_inode(d));
-+ }
-+ }
++static inline int au_test_debugfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_DEBUG_FS
++ return sb->s_magic == DEBUGFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+out_dpages:
-+ au_dpages_free(&dpages);
++static inline int au_test_nilfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_NILFS)
++ return sb->s_magic == NILFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+#if 0
-+ /* discard children */
-+ dentry_unhash(dentry);
-+ dput(dentry);
++static inline int au_test_hfsplus(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_HFSPLUS_FS)
++ return sb->s_magic == HFSPLUS_SUPER_MAGIC;
++#else
++ return 0;
+#endif
-+out:
-+ return err;
+}
+
++/* ---------------------------------------------------------------------- */
+/*
-+ * return 0 if processed.
++ * they can't be an aufs branch.
+ */
-+static int hn_gen_by_inode(char *name, unsigned int nlen, struct inode *inode,
-+ const unsigned int isdir)
++static inline int au_test_fs_unsuppoted(struct super_block *sb)
+{
-+ int err;
-+ struct dentry *d;
-+ struct qstr *dname;
-+
-+ err = 1;
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ pr_warn("branch root dir was changed\n");
-+ err = 0;
-+ goto out;
-+ }
-+
-+ if (!isdir) {
-+ AuDebugOn(!name);
-+ au_iigen_dec(inode);
-+ spin_lock(&inode->i_lock);
-+ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
-+ spin_lock(&d->d_lock);
-+ dname = &d->d_name;
-+ if (dname->len != nlen
-+ && memcmp(dname->name, name, nlen)) {
-+ spin_unlock(&d->d_lock);
-+ continue;
-+ }
-+ err = 0;
-+ au_digen_dec(d);
-+ spin_unlock(&d->d_lock);
-+ break;
-+ }
-+ spin_unlock(&inode->i_lock);
-+ } else {
-+ au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIR);
-+ d = d_find_any_alias(inode);
-+ if (!d) {
-+ au_iigen_dec(inode);
-+ goto out;
-+ }
-+
-+ spin_lock(&d->d_lock);
-+ dname = &d->d_name;
-+ if (dname->len == nlen && !memcmp(dname->name, name, nlen)) {
-+ spin_unlock(&d->d_lock);
-+ err = hn_gen_tree(d);
-+ spin_lock(&d->d_lock);
-+ }
-+ spin_unlock(&d->d_lock);
-+ dput(d);
-+ }
-+
-+out:
-+ AuTraceErr(err);
-+ return err;
++ return
++#ifndef CONFIG_AUFS_BR_RAMFS
++ au_test_ramfs(sb) ||
++#endif
++ au_test_procfs(sb)
++ || au_test_sysfs(sb)
++ || au_test_configfs(sb)
++ || au_test_debugfs(sb)
++ || au_test_securityfs(sb)
++ || au_test_xenfs(sb)
++ || au_test_ecryptfs(sb)
++ /* || !strcmp(au_sbtype(sb), "unionfs") */
++ || au_test_aufs(sb); /* will be supported in next version */
+}
+
-+static int hn_gen_by_name(struct dentry *dentry, const unsigned int isdir)
++static inline int au_test_fs_remote(struct super_block *sb)
+{
-+ int err;
-+
-+ if (IS_ROOT(dentry)) {
-+ pr_warn("branch root dir was changed\n");
-+ return 0;
-+ }
-+
-+ err = 0;
-+ if (!isdir) {
-+ au_digen_dec(dentry);
-+ if (d_really_is_positive(dentry))
-+ au_iigen_dec(d_inode(dentry));
-+ } else {
-+ au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR);
-+ if (d_really_is_positive(dentry))
-+ err = hn_gen_tree(dentry);
-+ }
-+
-+ AuTraceErr(err);
-+ return err;
++ return !au_test_tmpfs(sb)
++#ifdef CONFIG_AUFS_BR_RAMFS
++ && !au_test_ramfs(sb)
++#endif
++ && !(sb->s_type->fs_flags & FS_REQUIRES_DEV);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+/* hnotify job flags */
-+#define AuHnJob_XINO0 1
-+#define AuHnJob_GEN (1 << 1)
-+#define AuHnJob_DIRENT (1 << 2)
-+#define AuHnJob_ISDIR (1 << 3)
-+#define AuHnJob_TRYXINO0 (1 << 4)
-+#define AuHnJob_MNTPNT (1 << 5)
-+#define au_ftest_hnjob(flags, name) ((flags) & AuHnJob_##name)
-+#define au_fset_hnjob(flags, name) \
-+ do { (flags) |= AuHnJob_##name; } while (0)
-+#define au_fclr_hnjob(flags, name) \
-+ do { (flags) &= ~AuHnJob_##name; } while (0)
-+
-+enum {
-+ AuHn_CHILD,
-+ AuHn_PARENT,
-+ AuHnLast
-+};
-+
-+struct au_hnotify_args {
-+ struct inode *h_dir, *dir, *h_child_inode;
-+ u32 mask;
-+ unsigned int flags[AuHnLast];
-+ unsigned int h_child_nlen;
-+ char h_child_name[];
-+};
++/*
++ * Note: these functions (below) are created after reading ->getattr() in all
++ * filesystems under linux/fs. it means we have to do so in every update...
++ */
+
-+struct hn_job_args {
-+ unsigned int flags;
-+ struct inode *inode, *h_inode, *dir, *h_dir;
-+ struct dentry *dentry;
-+ char *h_name;
-+ int h_nlen;
-+};
++/*
++ * some filesystems require getattr to refresh the inode attributes before
++ * referencing.
++ * in most cases, we can rely on the inode attribute in NFS (or every remote fs)
++ * and leave the work for d_revalidate()
++ */
++static inline int au_test_fs_refresh_iattr(struct super_block *sb)
++{
++ return au_test_nfs(sb)
++ || au_test_fuse(sb)
++ /* || au_test_btrfs(sb) */ /* untested */
++ ;
++}
+
-+static int hn_job(struct hn_job_args *a)
++/*
++ * filesystems which don't maintain i_size or i_blocks.
++ */
++static inline int au_test_fs_bad_iattr_size(struct super_block *sb)
+{
-+ const unsigned int isdir = au_ftest_hnjob(a->flags, ISDIR);
-+ int e;
++ return au_test_xfs(sb)
++ || au_test_btrfs(sb)
++ || au_test_ubifs(sb)
++ || au_test_hfsplus(sb) /* maintained, but incorrect */
++ /* || au_test_minix(sb) */ /* untested */
++ ;
++}
+
-+ /* reset xino */
-+ if (au_ftest_hnjob(a->flags, XINO0) && a->inode)
-+ hn_xino(a->inode, a->h_inode); /* ignore this error */
++/*
++ * filesystems which don't store the correct value in some of their inode
++ * attributes.
++ */
++static inline int au_test_fs_bad_iattr(struct super_block *sb)
++{
++ return au_test_fs_bad_iattr_size(sb)
++ || au_test_fat(sb)
++ || au_test_msdos(sb)
++ || au_test_vfat(sb);
++}
+
-+ if (au_ftest_hnjob(a->flags, TRYXINO0)
-+ && a->inode
-+ && a->h_inode) {
-+ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
-+ if (!a->h_inode->i_nlink
-+ && !(a->h_inode->i_state & I_LINKABLE))
-+ hn_xino(a->inode, a->h_inode); /* ignore this error */
-+ inode_unlock(a->h_inode);
-+ }
++/* they don't check i_nlink in link(2) */
++static inline int au_test_fs_no_limit_nlink(struct super_block *sb)
++{
++ return au_test_tmpfs(sb)
++#ifdef CONFIG_AUFS_BR_RAMFS
++ || au_test_ramfs(sb)
++#endif
++ || au_test_ubifs(sb)
++ || au_test_hfsplus(sb);
++}
+
-+ /* make the generation obsolete */
-+ if (au_ftest_hnjob(a->flags, GEN)) {
-+ e = -1;
-+ if (a->inode)
-+ e = hn_gen_by_inode(a->h_name, a->h_nlen, a->inode,
-+ isdir);
-+ if (e && a->dentry)
-+ hn_gen_by_name(a->dentry, isdir);
-+ /* ignore this error */
-+ }
++/*
++ * filesystems which sets S_NOATIME and S_NOCMTIME.
++ */
++static inline int au_test_fs_notime(struct super_block *sb)
++{
++ return au_test_nfs(sb)
++ || au_test_fuse(sb)
++ || au_test_ubifs(sb)
++ ;
++}
+
-+ /* make dir entries obsolete */
-+ if (au_ftest_hnjob(a->flags, DIRENT) && a->inode) {
-+ struct au_vdir *vdir;
++/* temporary support for i#1 in cramfs */
++static inline int au_test_fs_unique_ino(struct inode *inode)
++{
++ if (au_test_cramfs(inode->i_sb))
++ return inode->i_ino != 1;
++ return 1;
++}
+
-+ vdir = au_ivdir(a->inode);
-+ if (vdir)
-+ vdir->vd_jiffy = 0;
-+ /* IMustLock(a->inode); */
-+ /* a->inode->i_version++; */
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ /* can do nothing but warn */
-+ if (au_ftest_hnjob(a->flags, MNTPNT)
-+ && a->dentry
-+ && d_mountpoint(a->dentry))
-+ pr_warn("mount-point %pd is removed or renamed\n", a->dentry);
++/*
++ * the filesystem where the xino files placed must support i/o after unlink and
++ * maintain i_size and i_blocks.
++ */
++static inline int au_test_fs_bad_xino(struct super_block *sb)
++{
++ return au_test_fs_remote(sb)
++ || au_test_fs_bad_iattr_size(sb)
++ /* don't want unnecessary work for xino */
++ || au_test_aufs(sb)
++ || au_test_ecryptfs(sb)
++ || au_test_nilfs(sb);
++}
+
-+ return 0;
++static inline int au_test_fs_trunc_xino(struct super_block *sb)
++{
++ return au_test_tmpfs(sb)
++ || au_test_ramfs(sb);
+}
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * test if the @sb is real-readonly.
++ */
++static inline int au_test_fs_rr(struct super_block *sb)
++{
++ return au_test_squashfs(sb)
++ || au_test_iso9660(sb)
++ || au_test_cramfs(sb)
++ || au_test_romfs(sb);
++}
+
-+static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen,
-+ struct inode *dir)
++/*
++ * test if the @inode is nfs with 'noacl' option
++ * NFS always sets MS_POSIXACL regardless its mount option 'noacl.'
++ */
++static inline int au_test_nfs_noacl(struct inode *inode)
+{
-+ struct dentry *dentry, *d, *parent;
-+ struct qstr *dname;
++ return au_test_nfs(inode->i_sb)
++ /* && IS_POSIXACL(inode) */
++ && !nfs_server_capable(inode, NFS_CAP_ACLS);
++}
+
-+ parent = d_find_any_alias(dir);
-+ if (!parent)
-+ return NULL;
++#endif /* __KERNEL__ */
++#endif /* __AUFS_FSTYPE_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/hbl.h linux-4.14/fs/aufs/hbl.h
+--- linux-4.14/fs/aufs/hbl.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/hbl.h 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,64 @@
++/*
++ * Copyright (C) 2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+ dentry = NULL;
-+ spin_lock(&parent->d_lock);
-+ list_for_each_entry(d, &parent->d_subdirs, d_child) {
-+ /* AuDbg("%pd\n", d); */
-+ spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED);
-+ dname = &d->d_name;
-+ if (dname->len != nlen || memcmp(dname->name, name, nlen))
-+ goto cont_unlock;
-+ if (au_di(d))
-+ au_digen_dec(d);
-+ else
-+ goto cont_unlock;
-+ if (au_dcount(d) > 0) {
-+ dentry = dget_dlock(d);
-+ spin_unlock(&d->d_lock);
-+ break;
-+ }
++/*
++ * helpers for hlist_bl.h
++ */
+
-+cont_unlock:
-+ spin_unlock(&d->d_lock);
-+ }
-+ spin_unlock(&parent->d_lock);
-+ dput(parent);
++#ifndef __AUFS_HBL_H__
++#define __AUFS_HBL_H__
+
-+ if (dentry)
-+ di_write_lock_child(dentry);
++#ifdef __KERNEL__
+
-+ return dentry;
++#include <linux/list_bl.h>
++
++static inline void au_hbl_add(struct hlist_bl_node *node,
++ struct hlist_bl_head *hbl)
++{
++ hlist_bl_lock(hbl);
++ hlist_bl_add_head(node, hbl);
++ hlist_bl_unlock(hbl);
+}
+
-+static struct inode *lookup_wlock_by_ino(struct super_block *sb,
-+ aufs_bindex_t bindex, ino_t h_ino)
++static inline void au_hbl_del(struct hlist_bl_node *node,
++ struct hlist_bl_head *hbl)
+{
-+ struct inode *inode;
-+ ino_t ino;
-+ int err;
++ hlist_bl_lock(hbl);
++ hlist_bl_del(node);
++ hlist_bl_unlock(hbl);
++}
+
-+ inode = NULL;
-+ err = au_xino_read(sb, bindex, h_ino, &ino);
-+ if (!err && ino)
-+ inode = ilookup(sb, ino);
-+ if (!inode)
-+ goto out;
++#define au_hbl_for_each(pos, head) \
++ for (pos = hlist_bl_first(head); \
++ pos; \
++ pos = pos->next)
+
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ pr_warn("wrong root branch\n");
-+ iput(inode);
-+ inode = NULL;
-+ goto out;
-+ }
++static inline unsigned long au_hbl_count(struct hlist_bl_head *hbl)
++{
++ unsigned long cnt;
++ struct hlist_bl_node *pos;
+
-+ ii_write_lock_child(inode);
++ cnt = 0;
++ hlist_bl_lock(hbl);
++ au_hbl_for_each(pos, hbl)
++ cnt++;
++ hlist_bl_unlock(hbl);
++ return cnt;
++}
+
-+out:
-+ return inode;
++#endif /* __KERNEL__ */
++#endif /* __AUFS_HBL_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/hfsnotify.c linux-4.14/fs/aufs/hfsnotify.c
+--- linux-4.14/fs/aufs/hfsnotify.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/hfsnotify.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,289 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
++
++/*
++ * fsnotify for the lower directories
++ */
++
++#include "aufs.h"
++
++/* FS_IN_IGNORED is unnecessary */
++static const __u32 AuHfsnMask = (FS_MOVED_TO | FS_MOVED_FROM | FS_DELETE
++ | FS_CREATE | FS_EVENT_ON_CHILD);
++static DECLARE_WAIT_QUEUE_HEAD(au_hfsn_wq);
++static __cacheline_aligned_in_smp atomic64_t au_hfsn_ifree = ATOMIC64_INIT(0);
++
++static void au_hfsn_free_mark(struct fsnotify_mark *mark)
++{
++ struct au_hnotify *hn = container_of(mark, struct au_hnotify,
++ hn_mark);
++ /* AuDbg("here\n"); */
++ au_cache_free_hnotify(hn);
++ smp_mb__before_atomic(); /* for atomic64_dec */
++ if (atomic64_dec_and_test(&au_hfsn_ifree))
++ wake_up(&au_hfsn_wq);
+}
+
-+static void au_hn_bh(void *_args)
++static int au_hfsn_alloc(struct au_hinode *hinode)
+{
-+ struct au_hnotify_args *a = _args;
-+ struct super_block *sb;
-+ aufs_bindex_t bindex, bbot, bfound;
-+ unsigned char xino, try_iput;
+ int err;
-+ struct inode *inode;
-+ ino_t h_ino;
-+ struct hn_job_args args;
-+ struct dentry *dentry;
-+ struct au_sbinfo *sbinfo;
++ struct au_hnotify *hn;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct fsnotify_mark *mark;
++ aufs_bindex_t bindex;
+
-+ AuDebugOn(!_args);
-+ AuDebugOn(!a->h_dir);
-+ AuDebugOn(!a->dir);
-+ AuDebugOn(!a->mask);
-+ AuDbg("mask 0x%x, i%lu, hi%lu, hci%lu\n",
-+ a->mask, a->dir->i_ino, a->h_dir->i_ino,
-+ a->h_child_inode ? a->h_child_inode->i_ino : 0);
++ hn = hinode->hi_notify;
++ sb = hn->hn_aufs_inode->i_sb;
++ bindex = au_br_index(sb, hinode->hi_id);
++ br = au_sbr(sb, bindex);
++ AuDebugOn(!br->br_hfsn);
+
-+ inode = NULL;
-+ dentry = NULL;
++ mark = &hn->hn_mark;
++ fsnotify_init_mark(mark, br->br_hfsn->hfsn_group);
++ mark->mask = AuHfsnMask;
+ /*
-+ * do not lock a->dir->i_mutex here
-+ * because of d_revalidate() may cause a deadlock.
++ * by udba rename or rmdir, aufs assign a new inode to the known
++ * h_inode, so specify 1 to allow dups.
+ */
-+ sb = a->dir->i_sb;
-+ AuDebugOn(!sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!sbinfo);
-+ si_write_lock(sb, AuLock_NOPLMW);
-+
-+ ii_read_lock_parent(a->dir);
-+ bfound = -1;
-+ bbot = au_ibbot(a->dir);
-+ for (bindex = au_ibtop(a->dir); bindex <= bbot; bindex++)
-+ if (au_h_iptr(a->dir, bindex) == a->h_dir) {
-+ bfound = bindex;
-+ break;
-+ }
-+ ii_read_unlock(a->dir);
-+ if (unlikely(bfound < 0))
-+ goto out;
++ lockdep_off();
++ err = fsnotify_add_mark(mark, hinode->hi_inode, /*mnt*/NULL,
++ /*allow_dups*/1);
++ lockdep_on();
+
-+ xino = !!au_opt_test(au_mntflags(sb), XINO);
-+ h_ino = 0;
-+ if (a->h_child_inode)
-+ h_ino = a->h_child_inode->i_ino;
++ return err;
++}
+
-+ if (a->h_child_nlen
-+ && (au_ftest_hnjob(a->flags[AuHn_CHILD], GEN)
-+ || au_ftest_hnjob(a->flags[AuHn_CHILD], MNTPNT)))
-+ dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen,
-+ a->dir);
-+ try_iput = 0;
-+ if (dentry && d_really_is_positive(dentry))
-+ inode = d_inode(dentry);
-+ if (xino && !inode && h_ino
-+ && (au_ftest_hnjob(a->flags[AuHn_CHILD], XINO0)
-+ || au_ftest_hnjob(a->flags[AuHn_CHILD], TRYXINO0)
-+ || au_ftest_hnjob(a->flags[AuHn_CHILD], GEN))) {
-+ inode = lookup_wlock_by_ino(sb, bfound, h_ino);
-+ try_iput = 1;
-+ }
++static int au_hfsn_free(struct au_hinode *hinode, struct au_hnotify *hn)
++{
++ struct fsnotify_mark *mark;
++ unsigned long long ull;
++ struct fsnotify_group *group;
+
-+ args.flags = a->flags[AuHn_CHILD];
-+ args.dentry = dentry;
-+ args.inode = inode;
-+ args.h_inode = a->h_child_inode;
-+ args.dir = a->dir;
-+ args.h_dir = a->h_dir;
-+ args.h_name = a->h_child_name;
-+ args.h_nlen = a->h_child_nlen;
-+ err = hn_job(&args);
-+ if (dentry) {
-+ if (au_di(dentry))
-+ di_write_unlock(dentry);
-+ dput(dentry);
-+ }
-+ if (inode && try_iput) {
-+ ii_write_unlock(inode);
-+ iput(inode);
-+ }
++ ull = atomic64_inc_return(&au_hfsn_ifree);
++ BUG_ON(!ull);
+
-+ ii_write_lock_parent(a->dir);
-+ args.flags = a->flags[AuHn_PARENT];
-+ args.dentry = NULL;
-+ args.inode = a->dir;
-+ args.h_inode = a->h_dir;
-+ args.dir = NULL;
-+ args.h_dir = NULL;
-+ args.h_name = NULL;
-+ args.h_nlen = 0;
-+ err = hn_job(&args);
-+ ii_write_unlock(a->dir);
++ mark = &hn->hn_mark;
++ spin_lock(&mark->lock);
++ group = mark->group;
++ fsnotify_get_group(group);
++ spin_unlock(&mark->lock);
++ lockdep_off();
++ fsnotify_destroy_mark(mark, group);
++ fsnotify_put_mark(mark);
++ fsnotify_put_group(group);
++ lockdep_on();
+
-+out:
-+ iput(a->h_child_inode);
-+ iput(a->h_dir);
-+ iput(a->dir);
-+ si_write_unlock(sb);
-+ au_nwt_done(&sbinfo->si_nowait);
-+ au_delayed_kfree(a);
++ /* free hn by myself */
++ return 0;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
-+ struct qstr *h_child_qstr, struct inode *h_child_inode)
++static void au_hfsn_ctl(struct au_hinode *hinode, int do_set)
+{
-+ int err, len;
-+ unsigned int flags[AuHnLast], f;
-+ unsigned char isdir, isroot, wh;
-+ struct inode *dir;
-+ struct au_hnotify_args *args;
-+ char *p, *h_child_name;
-+
-+ err = 0;
-+ AuDebugOn(!hnotify || !hnotify->hn_aufs_inode);
-+ dir = igrab(hnotify->hn_aufs_inode);
-+ if (!dir)
-+ goto out;
++ struct fsnotify_mark *mark;
+
-+ isroot = (dir->i_ino == AUFS_ROOT_INO);
-+ wh = 0;
-+ h_child_name = (void *)h_child_qstr->name;
-+ len = h_child_qstr->len;
-+ if (h_child_name) {
-+ if (len > AUFS_WH_PFX_LEN
-+ && !memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
-+ h_child_name += AUFS_WH_PFX_LEN;
-+ len -= AUFS_WH_PFX_LEN;
-+ wh = 1;
-+ }
++ mark = &hinode->hi_notify->hn_mark;
++ spin_lock(&mark->lock);
++ if (do_set) {
++ AuDebugOn(mark->mask & AuHfsnMask);
++ mark->mask |= AuHfsnMask;
++ } else {
++ AuDebugOn(!(mark->mask & AuHfsnMask));
++ mark->mask &= ~AuHfsnMask;
+ }
++ spin_unlock(&mark->lock);
++ /* fsnotify_recalc_inode_mask(hinode->hi_inode); */
++}
+
-+ isdir = 0;
-+ if (h_child_inode)
-+ isdir = !!S_ISDIR(h_child_inode->i_mode);
-+ flags[AuHn_PARENT] = AuHnJob_ISDIR;
-+ flags[AuHn_CHILD] = 0;
-+ if (isdir)
-+ flags[AuHn_CHILD] = AuHnJob_ISDIR;
-+ au_fset_hnjob(flags[AuHn_PARENT], DIRENT);
-+ au_fset_hnjob(flags[AuHn_CHILD], GEN);
-+ switch (mask & FS_EVENTS_POSS_ON_CHILD) {
-+ case FS_MOVED_FROM:
-+ case FS_MOVED_TO:
-+ au_fset_hnjob(flags[AuHn_CHILD], XINO0);
-+ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
-+ /*FALLTHROUGH*/
-+ case FS_CREATE:
-+ AuDebugOn(!h_child_name);
-+ break;
++/* ---------------------------------------------------------------------- */
+
-+ case FS_DELETE:
-+ /*
-+ * aufs never be able to get this child inode.
-+ * revalidation should be in d_revalidate()
-+ * by checking i_nlink, i_generation or d_unhashed().
-+ */
-+ AuDebugOn(!h_child_name);
-+ au_fset_hnjob(flags[AuHn_CHILD], TRYXINO0);
-+ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
-+ break;
++/* #define AuDbgHnotify */
++#ifdef AuDbgHnotify
++static char *au_hfsn_name(u32 mask)
++{
++#ifdef CONFIG_AUFS_DEBUG
++#define test_ret(flag) \
++ do { \
++ if (mask & flag) \
++ return #flag; \
++ } while (0)
++ test_ret(FS_ACCESS);
++ test_ret(FS_MODIFY);
++ test_ret(FS_ATTRIB);
++ test_ret(FS_CLOSE_WRITE);
++ test_ret(FS_CLOSE_NOWRITE);
++ test_ret(FS_OPEN);
++ test_ret(FS_MOVED_FROM);
++ test_ret(FS_MOVED_TO);
++ test_ret(FS_CREATE);
++ test_ret(FS_DELETE);
++ test_ret(FS_DELETE_SELF);
++ test_ret(FS_MOVE_SELF);
++ test_ret(FS_UNMOUNT);
++ test_ret(FS_Q_OVERFLOW);
++ test_ret(FS_IN_IGNORED);
++ test_ret(FS_ISDIR);
++ test_ret(FS_IN_ONESHOT);
++ test_ret(FS_EVENT_ON_CHILD);
++ return "";
++#undef test_ret
++#else
++ return "??";
++#endif
++}
++#endif
+
-+ default:
-+ AuDebugOn(1);
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ if (wh)
-+ h_child_inode = NULL;
++static void au_hfsn_free_group(struct fsnotify_group *group)
++{
++ struct au_br_hfsnotify *hfsn = group->private;
+
-+ err = -ENOMEM;
-+ /* iput() and kfree() will be called in au_hnotify() */
-+ args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS);
-+ if (unlikely(!args)) {
-+ AuErr1("no memory\n");
-+ iput(dir);
++ /* AuDbg("here\n"); */
++ kfree(hfsn);
++}
++
++static int au_hfsn_handle_event(struct fsnotify_group *group,
++ struct inode *inode,
++ struct fsnotify_mark *inode_mark,
++ struct fsnotify_mark *vfsmount_mark,
++ u32 mask, const void *data, int data_type,
++ const unsigned char *file_name, u32 cookie,
++ struct fsnotify_iter_info *iter_info)
++{
++ int err;
++ struct au_hnotify *hnotify;
++ struct inode *h_dir, *h_inode;
++ struct qstr h_child_qstr = QSTR_INIT(file_name, strlen(file_name));
++
++ AuDebugOn(data_type != FSNOTIFY_EVENT_INODE);
++
++ err = 0;
++ /* if FS_UNMOUNT happens, there must be another bug */
++ AuDebugOn(mask & FS_UNMOUNT);
++ if (mask & (FS_IN_IGNORED | FS_UNMOUNT))
+ goto out;
-+ }
-+ args->flags[AuHn_PARENT] = flags[AuHn_PARENT];
-+ args->flags[AuHn_CHILD] = flags[AuHn_CHILD];
-+ args->mask = mask;
-+ args->dir = dir;
-+ args->h_dir = igrab(h_dir);
-+ if (h_child_inode)
-+ h_child_inode = igrab(h_child_inode); /* can be NULL */
-+ args->h_child_inode = h_child_inode;
-+ args->h_child_nlen = len;
-+ if (len) {
-+ p = (void *)args;
-+ p += sizeof(*args);
-+ memcpy(p, h_child_name, len);
-+ p[len] = 0;
-+ }
+
-+ /* NFS fires the event for silly-renamed one from kworker */
-+ f = 0;
-+ if (!dir->i_nlink
-+ || (au_test_nfs(h_dir->i_sb) && (mask & FS_DELETE)))
-+ f = AuWkq_NEST;
-+ err = au_wkq_nowait(au_hn_bh, args, dir->i_sb, f);
-+ if (unlikely(err)) {
-+ pr_err("wkq %d\n", err);
-+ iput(args->h_child_inode);
-+ iput(args->h_dir);
-+ iput(args->dir);
-+ au_delayed_kfree(args);
++ h_dir = inode;
++ h_inode = NULL;
++#ifdef AuDbgHnotify
++ au_debug_on();
++ if (1 || h_child_qstr.len != sizeof(AUFS_XINO_FNAME) - 1
++ || strncmp(h_child_qstr.name, AUFS_XINO_FNAME, h_child_qstr.len)) {
++ AuDbg("i%lu, mask 0x%x %s, hcname %.*s, hi%lu\n",
++ h_dir->i_ino, mask, au_hfsn_name(mask),
++ AuLNPair(&h_child_qstr), h_inode ? h_inode->i_ino : 0);
++ /* WARN_ON(1); */
+ }
++ au_debug_off();
++#endif
++
++ AuDebugOn(!inode_mark);
++ hnotify = container_of(inode_mark, struct au_hnotify, hn_mark);
++ err = au_hnotify(h_dir, hnotify, mask, &h_child_qstr, h_inode);
+
+out:
+ return err;
+}
+
++static struct fsnotify_ops au_hfsn_ops = {
++ .handle_event = au_hfsn_handle_event,
++ .free_group_priv = au_hfsn_free_group,
++ .free_mark = au_hfsn_free_mark
++};
++
+/* ---------------------------------------------------------------------- */
+
-+int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm)
++static void au_hfsn_fin_br(struct au_branch *br)
++{
++ struct au_br_hfsnotify *hfsn;
++
++ hfsn = br->br_hfsn;
++ if (hfsn) {
++ lockdep_off();
++ fsnotify_put_group(hfsn->hfsn_group);
++ lockdep_on();
++ }
++}
++
++static int au_hfsn_init_br(struct au_branch *br, int perm)
+{
+ int err;
++ struct fsnotify_group *group;
++ struct au_br_hfsnotify *hfsn;
+
-+ AuDebugOn(!(udba & AuOptMask_UDBA));
++ err = 0;
++ br->br_hfsn = NULL;
++ if (!au_br_hnotifyable(perm))
++ goto out;
++
++ err = -ENOMEM;
++ hfsn = kmalloc(sizeof(*hfsn), GFP_NOFS);
++ if (unlikely(!hfsn))
++ goto out;
+
+ err = 0;
-+ if (au_hnotify_op.reset_br)
-+ err = au_hnotify_op.reset_br(udba, br, perm);
++ group = fsnotify_alloc_group(&au_hfsn_ops);
++ if (IS_ERR(group)) {
++ err = PTR_ERR(group);
++ pr_err("fsnotify_alloc_group() failed, %d\n", err);
++ goto out_hfsn;
++ }
++
++ group->private = hfsn;
++ hfsn->hfsn_group = group;
++ br->br_hfsn = hfsn;
++ goto out; /* success */
+
++out_hfsn:
++ kfree(hfsn);
++out:
+ return err;
+}
+
-+int au_hnotify_init_br(struct au_branch *br, int perm)
++static int au_hfsn_reset_br(unsigned int udba, struct au_branch *br, int perm)
+{
+ int err;
+
+ err = 0;
-+ if (au_hnotify_op.init_br)
-+ err = au_hnotify_op.init_br(br, perm);
++ if (!br->br_hfsn)
++ err = au_hfsn_init_br(br, perm);
+
+ return err;
+}
+
-+void au_hnotify_fin_br(struct au_branch *br)
++/* ---------------------------------------------------------------------- */
++
++static void au_hfsn_fin(void)
+{
-+ if (au_hnotify_op.fin_br)
-+ au_hnotify_op.fin_br(br);
++ AuDbg("au_hfsn_ifree %lld\n", (long long)atomic64_read(&au_hfsn_ifree));
++ wait_event(au_hfsn_wq, !atomic64_read(&au_hfsn_ifree));
+}
+
-+static void au_hn_destroy_cache(void)
-+{
-+ struct au_cache *cp;
++const struct au_hnotify_op au_hnotify_op = {
++ .ctl = au_hfsn_ctl,
++ .alloc = au_hfsn_alloc,
++ .free = au_hfsn_free,
+
-+ flush_delayed_work(&au_dfree.dwork);
-+ cp = au_dfree.cache + AuCache_HNOTIFY;
-+ AuDebugOn(!llist_empty(&cp->llist));
-+ kmem_cache_destroy(cp->cache);
-+ cp->cache = NULL;
-+}
++ .fin = au_hfsn_fin,
+
-+AU_CACHE_DFREE_FUNC(hnotify, HNOTIFY, hn_lnode);
++ .reset_br = au_hfsn_reset_br,
++ .fin_br = au_hfsn_fin_br,
++ .init_br = au_hfsn_init_br
++};
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/hfsplus.c linux-4.14/fs/aufs/hfsplus.c
+--- linux-4.14/fs/aufs/hfsplus.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/hfsplus.c 2021-02-24 21:42:43.441114748 +0100
+@@ -0,0 +1,56 @@
++/*
++ * Copyright (C) 2010-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+int __init au_hnotify_init(void)
-+{
-+ int err;
-+ struct au_cache *cp;
++/*
++ * special support for filesystems which aqucires an inode mutex
++ * at final closing a file, eg, hfsplus.
++ *
++ * This trick is very simple and stupid, just to open the file before really
++ * neceeary open to tell hfsplus that this is not the final closing.
++ * The caller should call au_h_open_pre() after acquiring the inode mutex,
++ * and au_h_open_post() after releasing it.
++ */
+
-+ err = -ENOMEM;
-+ cp = au_dfree.cache + AuCache_HNOTIFY;
-+ cp->cache = AuCache(au_hnotify);
-+ if (cp->cache) {
-+ err = 0;
-+ if (au_hnotify_op.init)
-+ err = au_hnotify_op.init();
-+ if (unlikely(err))
-+ au_hn_destroy_cache();
-+ }
-+ AuTraceErr(err);
-+ return err;
-+}
++#include "aufs.h"
+
-+void au_hnotify_fin(void)
++struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
++ int force_wr)
+{
-+ struct au_cache *cp;
++ struct file *h_file;
++ struct dentry *h_dentry;
+
-+ if (au_hnotify_op.fin)
-+ au_hnotify_op.fin();
++ h_dentry = au_h_dptr(dentry, bindex);
++ AuDebugOn(!h_dentry);
++ AuDebugOn(d_is_negative(h_dentry));
+
-+ /* cf. au_cache_fin() */
-+ cp = au_dfree.cache + AuCache_HNOTIFY;
-+ if (cp->cache)
-+ au_hn_destroy_cache();
++ h_file = NULL;
++ if (au_test_hfsplus(h_dentry->d_sb)
++ && d_is_reg(h_dentry))
++ h_file = au_h_open(dentry, bindex,
++ O_RDONLY | O_NOATIME | O_LARGEFILE,
++ /*file*/NULL, force_wr);
++ return h_file;
++}
++
++void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file)
++{
++ if (h_file) {
++ fput(h_file);
++ au_sbr_put(dentry->d_sb, bindex);
++ }
+}
-diff -urN /usr/share/empty/fs/aufs/iinfo.c linux/fs/aufs/iinfo.c
---- /usr/share/empty/fs/aufs/iinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/iinfo.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,284 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/hnotify.c linux-4.14/fs/aufs/hnotify.c
+--- linux-4.14/fs/aufs/hnotify.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/hnotify.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,719 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * inode private data
++ * abstraction to notify the direct changes on lower directories
+ */
+
+#include "aufs.h"
+
-+struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex)
++int au_hn_alloc(struct au_hinode *hinode, struct inode *inode)
+{
-+ struct inode *h_inode;
-+ struct au_hinode *hinode;
++ int err;
++ struct au_hnotify *hn;
+
-+ IiMustAnyLock(inode);
++ err = -ENOMEM;
++ hn = au_cache_alloc_hnotify();
++ if (hn) {
++ hn->hn_aufs_inode = inode;
++ hinode->hi_notify = hn;
++ err = au_hnotify_op.alloc(hinode);
++ AuTraceErr(err);
++ if (unlikely(err)) {
++ hinode->hi_notify = NULL;
++ au_cache_free_hnotify(hn);
++ /*
++ * The upper dir was removed by udba, but the same named
++ * dir left. In this case, aufs assignes a new inode
++ * number and set the monitor again.
++ * For the lower dir, the old monitnor is still left.
++ */
++ if (err == -EEXIST)
++ err = 0;
++ }
++ }
+
-+ hinode = au_hinode(au_ii(inode), bindex);
-+ h_inode = hinode->hi_inode;
-+ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
-+ return h_inode;
++ AuTraceErr(err);
++ return err;
+}
+
-+/* todo: hard/soft set? */
-+void au_hiput(struct au_hinode *hinode)
++void au_hn_free(struct au_hinode *hinode)
+{
-+ au_hn_free(hinode);
-+ dput(hinode->hi_whdentry);
-+ iput(hinode->hi_inode);
++ struct au_hnotify *hn;
++
++ hn = hinode->hi_notify;
++ if (hn) {
++ hinode->hi_notify = NULL;
++ if (au_hnotify_op.free(hinode, hn))
++ au_cache_free_hnotify(hn);
++ }
+}
+
-+unsigned int au_hi_flags(struct inode *inode, int isdir)
-+{
-+ unsigned int flags;
-+ const unsigned int mnt_flags = au_mntflags(inode->i_sb);
++/* ---------------------------------------------------------------------- */
+
-+ flags = 0;
-+ if (au_opt_test(mnt_flags, XINO))
-+ au_fset_hi(flags, XINO);
-+ if (isdir && au_opt_test(mnt_flags, UDBA_HNOTIFY))
-+ au_fset_hi(flags, HNOTIFY);
-+ return flags;
++void au_hn_ctl(struct au_hinode *hinode, int do_set)
++{
++ if (hinode->hi_notify)
++ au_hnotify_op.ctl(hinode, do_set);
+}
+
-+void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode, unsigned int flags)
++void au_hn_reset(struct inode *inode, unsigned int flags)
+{
-+ struct au_hinode *hinode;
++ aufs_bindex_t bindex, bbot;
+ struct inode *hi;
-+ struct au_iinfo *iinfo = au_ii(inode);
++ struct dentry *iwhdentry;
+
-+ IiMustWriteLock(inode);
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
++ hi = au_h_iptr(inode, bindex);
++ if (!hi)
++ continue;
+
-+ hinode = au_hinode(iinfo, bindex);
-+ hi = hinode->hi_inode;
-+ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
++ /* inode_lock_nested(hi, AuLsc_I_CHILD); */
++ iwhdentry = au_hi_wh(inode, bindex);
++ if (iwhdentry)
++ dget(iwhdentry);
++ au_igrab(hi);
++ au_set_h_iptr(inode, bindex, NULL, 0);
++ au_set_h_iptr(inode, bindex, au_igrab(hi),
++ flags & ~AuHi_XINO);
++ iput(hi);
++ dput(iwhdentry);
++ /* inode_unlock(hi); */
++ }
++}
+
-+ if (hi)
-+ au_hiput(hinode);
-+ hinode->hi_inode = h_inode;
-+ if (h_inode) {
-+ int err;
-+ struct super_block *sb = inode->i_sb;
-+ struct au_branch *br;
++/* ---------------------------------------------------------------------- */
+
-+ AuDebugOn(inode->i_mode
-+ && (h_inode->i_mode & S_IFMT)
-+ != (inode->i_mode & S_IFMT));
-+ if (bindex == iinfo->ii_btop)
-+ au_cpup_igen(inode, h_inode);
-+ br = au_sbr(sb, bindex);
-+ hinode->hi_id = br->br_id;
-+ if (au_ftest_hi(flags, XINO)) {
-+ err = au_xino_write(sb, bindex, h_inode->i_ino,
-+ inode->i_ino);
-+ if (unlikely(err))
-+ AuIOErr1("failed au_xino_write() %d\n", err);
-+ }
++static int hn_xino(struct inode *inode, struct inode *h_inode)
++{
++ int err;
++ aufs_bindex_t bindex, bbot, bfound, btop;
++ struct inode *h_i;
+
-+ if (au_ftest_hi(flags, HNOTIFY)
-+ && au_br_hnotifyable(br->br_perm)) {
-+ err = au_hn_alloc(hinode, inode);
-+ if (unlikely(err))
-+ AuIOErr1("au_hn_alloc() %d\n", err);
-+ }
++ err = 0;
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("branch root dir was changed\n");
++ goto out;
+ }
-+}
+
-+void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_wh)
-+{
-+ struct au_hinode *hinode;
++ bfound = -1;
++ bbot = au_ibbot(inode);
++ btop = au_ibtop(inode);
++#if 0 /* reserved for future use */
++ if (bindex == bbot) {
++ /* keep this ino in rename case */
++ goto out;
++ }
++#endif
++ for (bindex = btop; bindex <= bbot; bindex++)
++ if (au_h_iptr(inode, bindex) == h_inode) {
++ bfound = bindex;
++ break;
++ }
++ if (bfound < 0)
++ goto out;
+
-+ IiMustWriteLock(inode);
++ for (bindex = btop; bindex <= bbot; bindex++) {
++ h_i = au_h_iptr(inode, bindex);
++ if (!h_i)
++ continue;
+
-+ hinode = au_hinode(au_ii(inode), bindex);
-+ AuDebugOn(hinode->hi_whdentry);
-+ hinode->hi_whdentry = h_wh;
-+}
++ err = au_xino_write(inode->i_sb, bindex, h_i->i_ino, /*ino*/0);
++ /* ignore this error */
++ /* bad action? */
++ }
+
-+void au_update_iigen(struct inode *inode, int half)
-+{
-+ struct au_iinfo *iinfo;
-+ struct au_iigen *iigen;
-+ unsigned int sigen;
++ /* children inode number will be broken */
+
-+ sigen = au_sigen(inode->i_sb);
-+ iinfo = au_ii(inode);
-+ iigen = &iinfo->ii_generation;
-+ spin_lock(&iigen->ig_spin);
-+ iigen->ig_generation = sigen;
-+ if (half)
-+ au_ig_fset(iigen->ig_flags, HALF_REFRESHED);
-+ else
-+ au_ig_fclr(iigen->ig_flags, HALF_REFRESHED);
-+ spin_unlock(&iigen->ig_spin);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+/* it may be called at remount time, too */
-+void au_update_ibrange(struct inode *inode, int do_put_zero)
++static int hn_gen_tree(struct dentry *dentry)
+{
-+ struct au_iinfo *iinfo;
-+ aufs_bindex_t bindex, bbot;
++ int err, i, j, ndentry;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry **dentries;
+
-+ AuDebugOn(au_is_bad_inode(inode));
-+ IiMustWriteLock(inode);
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_dcsub_pages(&dpages, dentry, NULL, NULL);
++ if (unlikely(err))
++ goto out_dpages;
+
-+ iinfo = au_ii(inode);
-+ if (do_put_zero && iinfo->ii_btop >= 0) {
-+ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
-+ bindex++) {
-+ struct inode *h_i;
++ for (i = 0; i < dpages.ndpage; i++) {
++ dpage = dpages.dpages + i;
++ dentries = dpage->dentries;
++ ndentry = dpage->ndentry;
++ for (j = 0; j < ndentry; j++) {
++ struct dentry *d;
+
-+ h_i = au_hinode(iinfo, bindex)->hi_inode;
-+ if (h_i
-+ && !h_i->i_nlink
-+ && !(h_i->i_state & I_LINKABLE))
-+ au_set_h_iptr(inode, bindex, NULL, 0);
-+ }
-+ }
++ d = dentries[j];
++ if (IS_ROOT(d))
++ continue;
+
-+ iinfo->ii_btop = -1;
-+ iinfo->ii_bbot = -1;
-+ bbot = au_sbbot(inode->i_sb);
-+ for (bindex = 0; bindex <= bbot; bindex++)
-+ if (au_hinode(iinfo, bindex)->hi_inode) {
-+ iinfo->ii_btop = bindex;
-+ break;
++ au_digen_dec(d);
++ if (d_really_is_positive(d))
++ /* todo: reset children xino?
++ cached children only? */
++ au_iigen_dec(d_inode(d));
+ }
-+ if (iinfo->ii_btop >= 0)
-+ for (bindex = bbot; bindex >= iinfo->ii_btop; bindex--)
-+ if (au_hinode(iinfo, bindex)->hi_inode) {
-+ iinfo->ii_bbot = bindex;
-+ break;
-+ }
-+ AuDebugOn(iinfo->ii_btop > iinfo->ii_bbot);
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_icntnr_init_once(void *_c)
-+{
-+ struct au_icntnr *c = _c;
-+ struct au_iinfo *iinfo = &c->iinfo;
++ }
+
-+ spin_lock_init(&iinfo->ii_generation.ig_spin);
-+ au_rw_init(&iinfo->ii_rwsem);
-+ inode_init_once(&c->vfs_inode);
-+}
++out_dpages:
++ au_dpages_free(&dpages);
+
-+void au_hinode_init(struct au_hinode *hinode)
-+{
-+ hinode->hi_inode = NULL;
-+ hinode->hi_id = -1;
-+ au_hn_init(hinode);
-+ hinode->hi_whdentry = NULL;
++#if 0
++ /* discard children */
++ dentry_unhash(dentry);
++ dput(dentry);
++#endif
++out:
++ return err;
+}
+
-+int au_iinfo_init(struct inode *inode)
++/*
++ * return 0 if processed.
++ */
++static int hn_gen_by_inode(char *name, unsigned int nlen, struct inode *inode,
++ const unsigned int isdir)
+{
-+ struct au_iinfo *iinfo;
-+ struct super_block *sb;
-+ struct au_hinode *hi;
-+ int nbr, i;
++ int err;
++ struct dentry *d;
++ struct qstr *dname;
+
-+ sb = inode->i_sb;
-+ iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
-+ nbr = au_sbbot(sb) + 1;
-+ if (unlikely(nbr <= 0))
-+ nbr = 1;
-+ hi = kmalloc_array(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS);
-+ if (hi) {
-+ au_ninodes_inc(sb);
++ err = 1;
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("branch root dir was changed\n");
++ err = 0;
++ goto out;
++ }
+
-+ iinfo->ii_hinode = hi;
-+ for (i = 0; i < nbr; i++, hi++)
-+ au_hinode_init(hi);
++ if (!isdir) {
++ AuDebugOn(!name);
++ au_iigen_dec(inode);
++ spin_lock(&inode->i_lock);
++ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
++ spin_lock(&d->d_lock);
++ dname = &d->d_name;
++ if (dname->len != nlen
++ && memcmp(dname->name, name, nlen)) {
++ spin_unlock(&d->d_lock);
++ continue;
++ }
++ err = 0;
++ au_digen_dec(d);
++ spin_unlock(&d->d_lock);
++ break;
++ }
++ spin_unlock(&inode->i_lock);
++ } else {
++ au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIR);
++ d = d_find_any_alias(inode);
++ if (!d) {
++ au_iigen_dec(inode);
++ goto out;
++ }
+
-+ iinfo->ii_generation.ig_generation = au_sigen(sb);
-+ iinfo->ii_btop = -1;
-+ iinfo->ii_bbot = -1;
-+ iinfo->ii_vdir = NULL;
-+ return 0;
++ spin_lock(&d->d_lock);
++ dname = &d->d_name;
++ if (dname->len == nlen && !memcmp(dname->name, name, nlen)) {
++ spin_unlock(&d->d_lock);
++ err = hn_gen_tree(d);
++ spin_lock(&d->d_lock);
++ }
++ spin_unlock(&d->d_lock);
++ dput(d);
+ }
-+ return -ENOMEM;
++
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+int au_hinode_realloc(struct au_iinfo *iinfo, int nbr)
++static int hn_gen_by_name(struct dentry *dentry, const unsigned int isdir)
+{
-+ int err, i;
-+ struct au_hinode *hip;
++ int err;
+
-+ AuRwMustWriteLock(&iinfo->ii_rwsem);
++ if (IS_ROOT(dentry)) {
++ pr_warn("branch root dir was changed\n");
++ return 0;
++ }
+
-+ err = -ENOMEM;
-+ hip = krealloc(iinfo->ii_hinode, sizeof(*hip) * nbr, GFP_NOFS);
-+ if (hip) {
-+ iinfo->ii_hinode = hip;
-+ i = iinfo->ii_bbot + 1;
-+ hip += i;
-+ for (; i < nbr; i++, hip++)
-+ au_hinode_init(hip);
-+ err = 0;
++ err = 0;
++ if (!isdir) {
++ au_digen_dec(dentry);
++ if (d_really_is_positive(dentry))
++ au_iigen_dec(d_inode(dentry));
++ } else {
++ au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR);
++ if (d_really_is_positive(dentry))
++ err = hn_gen_tree(dentry);
+ }
+
++ AuTraceErr(err);
+ return err;
+}
+
-+void au_iinfo_fin(struct inode *inode)
++/* ---------------------------------------------------------------------- */
++
++/* hnotify job flags */
++#define AuHnJob_XINO0 1
++#define AuHnJob_GEN (1 << 1)
++#define AuHnJob_DIRENT (1 << 2)
++#define AuHnJob_ISDIR (1 << 3)
++#define AuHnJob_TRYXINO0 (1 << 4)
++#define AuHnJob_MNTPNT (1 << 5)
++#define au_ftest_hnjob(flags, name) ((flags) & AuHnJob_##name)
++#define au_fset_hnjob(flags, name) \
++ do { (flags) |= AuHnJob_##name; } while (0)
++#define au_fclr_hnjob(flags, name) \
++ do { (flags) &= ~AuHnJob_##name; } while (0)
++
++enum {
++ AuHn_CHILD,
++ AuHn_PARENT,
++ AuHnLast
++};
++
++struct au_hnotify_args {
++ struct inode *h_dir, *dir, *h_child_inode;
++ u32 mask;
++ unsigned int flags[AuHnLast];
++ unsigned int h_child_nlen;
++ char h_child_name[];
++};
++
++struct hn_job_args {
++ unsigned int flags;
++ struct inode *inode, *h_inode, *dir, *h_dir;
++ struct dentry *dentry;
++ char *h_name;
++ int h_nlen;
++};
++
++static int hn_job(struct hn_job_args *a)
+{
-+ struct au_iinfo *iinfo;
-+ struct au_hinode *hi;
-+ struct super_block *sb;
-+ aufs_bindex_t bindex, bbot;
-+ const unsigned char unlinked = !inode->i_nlink;
++ const unsigned int isdir = au_ftest_hnjob(a->flags, ISDIR);
++ int e;
+
-+ AuDebugOn(au_is_bad_inode(inode));
++ /* reset xino */
++ if (au_ftest_hnjob(a->flags, XINO0) && a->inode)
++ hn_xino(a->inode, a->h_inode); /* ignore this error */
+
-+ sb = inode->i_sb;
-+ au_ninodes_dec(sb);
-+ if (si_pid_test(sb))
-+ au_xino_delete_inode(inode, unlinked);
-+ else {
-+ /*
-+ * it is safe to hide the dependency between sbinfo and
-+ * sb->s_umount.
-+ */
-+ lockdep_off();
-+ si_noflush_read_lock(sb);
-+ au_xino_delete_inode(inode, unlinked);
-+ si_read_unlock(sb);
-+ lockdep_on();
++ if (au_ftest_hnjob(a->flags, TRYXINO0)
++ && a->inode
++ && a->h_inode) {
++ vfsub_inode_lock_shared_nested(a->h_inode, AuLsc_I_CHILD);
++ if (!a->h_inode->i_nlink
++ && !(a->h_inode->i_state & I_LINKABLE))
++ hn_xino(a->inode, a->h_inode); /* ignore this error */
++ inode_unlock_shared(a->h_inode);
+ }
+
-+ iinfo = au_ii(inode);
-+ if (iinfo->ii_vdir)
-+ au_vdir_free(iinfo->ii_vdir, /*atonce*/0);
++ /* make the generation obsolete */
++ if (au_ftest_hnjob(a->flags, GEN)) {
++ e = -1;
++ if (a->inode)
++ e = hn_gen_by_inode(a->h_name, a->h_nlen, a->inode,
++ isdir);
++ if (e && a->dentry)
++ hn_gen_by_name(a->dentry, isdir);
++ /* ignore this error */
++ }
+
-+ bindex = iinfo->ii_btop;
-+ if (bindex >= 0) {
-+ hi = au_hinode(iinfo, bindex);
-+ bbot = iinfo->ii_bbot;
-+ while (bindex++ <= bbot) {
-+ if (hi->hi_inode)
-+ au_hiput(hi);
-+ hi++;
++ /* make dir entries obsolete */
++ if (au_ftest_hnjob(a->flags, DIRENT) && a->inode) {
++ struct au_vdir *vdir;
++
++ vdir = au_ivdir(a->inode);
++ if (vdir)
++ vdir->vd_jiffy = 0;
++ /* IMustLock(a->inode); */
++ /* a->inode->i_version++; */
++ }
++
++ /* can do nothing but warn */
++ if (au_ftest_hnjob(a->flags, MNTPNT)
++ && a->dentry
++ && d_mountpoint(a->dentry))
++ pr_warn("mount-point %pd is removed or renamed\n", a->dentry);
++
++ return 0;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen,
++ struct inode *dir)
++{
++ struct dentry *dentry, *d, *parent;
++ struct qstr *dname;
++
++ parent = d_find_any_alias(dir);
++ if (!parent)
++ return NULL;
++
++ dentry = NULL;
++ spin_lock(&parent->d_lock);
++ list_for_each_entry(d, &parent->d_subdirs, d_child) {
++ /* AuDbg("%pd\n", d); */
++ spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED);
++ dname = &d->d_name;
++ if (dname->len != nlen || memcmp(dname->name, name, nlen))
++ goto cont_unlock;
++ if (au_di(d))
++ au_digen_dec(d);
++ else
++ goto cont_unlock;
++ if (au_dcount(d) > 0) {
++ dentry = dget_dlock(d);
++ spin_unlock(&d->d_lock);
++ break;
+ }
++
++cont_unlock:
++ spin_unlock(&d->d_lock);
+ }
-+ au_delayed_kfree(iinfo->ii_hinode);
-+ AuRwDestroy(&iinfo->ii_rwsem);
-+}
-diff -urN /usr/share/empty/fs/aufs/inode.c linux/fs/aufs/inode.c
---- /usr/share/empty/fs/aufs/inode.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/inode.c 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,517 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
++ spin_unlock(&parent->d_lock);
++ dput(parent);
+
-+/*
-+ * inode functions
-+ */
++ if (dentry)
++ di_write_lock_child(dentry);
+
-+#include "aufs.h"
++ return dentry;
++}
+
-+struct inode *au_igrab(struct inode *inode)
++static struct inode *lookup_wlock_by_ino(struct super_block *sb,
++ aufs_bindex_t bindex, ino_t h_ino)
+{
-+ if (inode) {
-+ AuDebugOn(!atomic_read(&inode->i_count));
-+ ihold(inode);
++ struct inode *inode;
++ ino_t ino;
++ int err;
++
++ inode = NULL;
++ err = au_xino_read(sb, bindex, h_ino, &ino);
++ if (!err && ino)
++ inode = ilookup(sb, ino);
++ if (!inode)
++ goto out;
++
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("wrong root branch\n");
++ iput(inode);
++ inode = NULL;
++ goto out;
+ }
-+ return inode;
-+}
+
-+static void au_refresh_hinode_attr(struct inode *inode, int do_version)
-+{
-+ au_cpup_attr_all(inode, /*force*/0);
-+ au_update_iigen(inode, /*half*/1);
-+ if (do_version)
-+ inode->i_version++;
++ ii_write_lock_child(inode);
++
++out:
++ return inode;
+}
+
-+static int au_ii_refresh(struct inode *inode, int *update)
++static void au_hn_bh(void *_args)
+{
-+ int err, e;
-+ umode_t type;
-+ aufs_bindex_t bindex, new_bindex;
++ struct au_hnotify_args *a = _args;
+ struct super_block *sb;
-+ struct au_iinfo *iinfo;
-+ struct au_hinode *p, *q, tmp;
-+
-+ AuDebugOn(au_is_bad_inode(inode));
-+ IiMustWriteLock(inode);
-+
-+ *update = 0;
-+ sb = inode->i_sb;
-+ type = inode->i_mode & S_IFMT;
-+ iinfo = au_ii(inode);
-+ err = au_hinode_realloc(iinfo, au_sbbot(sb) + 1);
-+ if (unlikely(err))
-+ goto out;
++ aufs_bindex_t bindex, bbot, bfound;
++ unsigned char xino, try_iput;
++ int err;
++ struct inode *inode;
++ ino_t h_ino;
++ struct hn_job_args args;
++ struct dentry *dentry;
++ struct au_sbinfo *sbinfo;
+
-+ AuDebugOn(iinfo->ii_btop < 0);
-+ p = au_hinode(iinfo, iinfo->ii_btop);
-+ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
-+ bindex++, p++) {
-+ if (!p->hi_inode)
-+ continue;
++ AuDebugOn(!_args);
++ AuDebugOn(!a->h_dir);
++ AuDebugOn(!a->dir);
++ AuDebugOn(!a->mask);
++ AuDbg("mask 0x%x, i%lu, hi%lu, hci%lu\n",
++ a->mask, a->dir->i_ino, a->h_dir->i_ino,
++ a->h_child_inode ? a->h_child_inode->i_ino : 0);
+
-+ AuDebugOn(type != (p->hi_inode->i_mode & S_IFMT));
-+ new_bindex = au_br_index(sb, p->hi_id);
-+ if (new_bindex == bindex)
-+ continue;
++ inode = NULL;
++ dentry = NULL;
++ /*
++ * do not lock a->dir->i_mutex here
++ * because of d_revalidate() may cause a deadlock.
++ */
++ sb = a->dir->i_sb;
++ AuDebugOn(!sb);
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!sbinfo);
++ si_write_lock(sb, AuLock_NOPLMW);
+
-+ if (new_bindex < 0) {
-+ *update = 1;
-+ au_hiput(p);
-+ p->hi_inode = NULL;
-+ continue;
++ if (au_opt_test(sbinfo->si_mntflags, DIRREN))
++ switch (a->mask & FS_EVENTS_POSS_ON_CHILD) {
++ case FS_MOVED_FROM:
++ case FS_MOVED_TO:
++ AuWarn1("DIRREN with UDBA may not work correctly "
++ "for the direct rename(2)\n");
+ }
+
-+ if (new_bindex < iinfo->ii_btop)
-+ iinfo->ii_btop = new_bindex;
-+ if (iinfo->ii_bbot < new_bindex)
-+ iinfo->ii_bbot = new_bindex;
-+ /* swap two lower inode, and loop again */
-+ q = au_hinode(iinfo, new_bindex);
-+ tmp = *q;
-+ *q = *p;
-+ *p = tmp;
-+ if (tmp.hi_inode) {
-+ bindex--;
-+ p--;
++ ii_read_lock_parent(a->dir);
++ bfound = -1;
++ bbot = au_ibbot(a->dir);
++ for (bindex = au_ibtop(a->dir); bindex <= bbot; bindex++)
++ if (au_h_iptr(a->dir, bindex) == a->h_dir) {
++ bfound = bindex;
++ break;
+ }
++ ii_read_unlock(a->dir);
++ if (unlikely(bfound < 0))
++ goto out;
++
++ xino = !!au_opt_test(au_mntflags(sb), XINO);
++ h_ino = 0;
++ if (a->h_child_inode)
++ h_ino = a->h_child_inode->i_ino;
++
++ if (a->h_child_nlen
++ && (au_ftest_hnjob(a->flags[AuHn_CHILD], GEN)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], MNTPNT)))
++ dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen,
++ a->dir);
++ try_iput = 0;
++ if (dentry && d_really_is_positive(dentry))
++ inode = d_inode(dentry);
++ if (xino && !inode && h_ino
++ && (au_ftest_hnjob(a->flags[AuHn_CHILD], XINO0)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], TRYXINO0)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], GEN))) {
++ inode = lookup_wlock_by_ino(sb, bfound, h_ino);
++ try_iput = 1;
+ }
-+ au_update_ibrange(inode, /*do_put_zero*/0);
-+ e = au_dy_irefresh(inode);
-+ if (unlikely(e && !err))
-+ err = e;
++
++ args.flags = a->flags[AuHn_CHILD];
++ args.dentry = dentry;
++ args.inode = inode;
++ args.h_inode = a->h_child_inode;
++ args.dir = a->dir;
++ args.h_dir = a->h_dir;
++ args.h_name = a->h_child_name;
++ args.h_nlen = a->h_child_nlen;
++ err = hn_job(&args);
++ if (dentry) {
++ if (au_di(dentry))
++ di_write_unlock(dentry);
++ dput(dentry);
++ }
++ if (inode && try_iput) {
++ ii_write_unlock(inode);
++ iput(inode);
++ }
++
++ ii_write_lock_parent(a->dir);
++ args.flags = a->flags[AuHn_PARENT];
++ args.dentry = NULL;
++ args.inode = a->dir;
++ args.h_inode = a->h_dir;
++ args.dir = NULL;
++ args.h_dir = NULL;
++ args.h_name = NULL;
++ args.h_nlen = 0;
++ err = hn_job(&args);
++ ii_write_unlock(a->dir);
+
+out:
-+ AuTraceErr(err);
-+ return err;
++ iput(a->h_child_inode);
++ iput(a->h_dir);
++ iput(a->dir);
++ si_write_unlock(sb);
++ au_nwt_done(&sbinfo->si_nowait);
++ kfree(a);
+}
+
-+void au_refresh_iop(struct inode *inode, int force_getattr)
++/* ---------------------------------------------------------------------- */
++
++int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
++ struct qstr *h_child_qstr, struct inode *h_child_inode)
+{
-+ int type;
-+ struct au_sbinfo *sbi = au_sbi(inode->i_sb);
-+ const struct inode_operations *iop
-+ = force_getattr ? aufs_iop : sbi->si_iop_array;
++ int err, len;
++ unsigned int flags[AuHnLast], f;
++ unsigned char isdir, isroot, wh;
++ struct inode *dir;
++ struct au_hnotify_args *args;
++ char *p, *h_child_name;
+
-+ if (inode->i_op == iop)
-+ return;
++ err = 0;
++ AuDebugOn(!hnotify || !hnotify->hn_aufs_inode);
++ dir = igrab(hnotify->hn_aufs_inode);
++ if (!dir)
++ goto out;
+
-+ switch (inode->i_mode & S_IFMT) {
-+ case S_IFDIR:
-+ type = AuIop_DIR;
++ isroot = (dir->i_ino == AUFS_ROOT_INO);
++ wh = 0;
++ h_child_name = (void *)h_child_qstr->name;
++ len = h_child_qstr->len;
++ if (h_child_name) {
++ if (len > AUFS_WH_PFX_LEN
++ && !memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
++ h_child_name += AUFS_WH_PFX_LEN;
++ len -= AUFS_WH_PFX_LEN;
++ wh = 1;
++ }
++ }
++
++ isdir = 0;
++ if (h_child_inode)
++ isdir = !!S_ISDIR(h_child_inode->i_mode);
++ flags[AuHn_PARENT] = AuHnJob_ISDIR;
++ flags[AuHn_CHILD] = 0;
++ if (isdir)
++ flags[AuHn_CHILD] = AuHnJob_ISDIR;
++ au_fset_hnjob(flags[AuHn_PARENT], DIRENT);
++ au_fset_hnjob(flags[AuHn_CHILD], GEN);
++ switch (mask & FS_EVENTS_POSS_ON_CHILD) {
++ case FS_MOVED_FROM:
++ case FS_MOVED_TO:
++ au_fset_hnjob(flags[AuHn_CHILD], XINO0);
++ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
++ /*FALLTHROUGH*/
++ case FS_CREATE:
++ AuDebugOn(!h_child_name);
+ break;
-+ case S_IFLNK:
-+ type = AuIop_SYMLINK;
++
++ case FS_DELETE:
++ /*
++ * aufs never be able to get this child inode.
++ * revalidation should be in d_revalidate()
++ * by checking i_nlink, i_generation or d_unhashed().
++ */
++ AuDebugOn(!h_child_name);
++ au_fset_hnjob(flags[AuHn_CHILD], TRYXINO0);
++ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
+ break;
++
+ default:
-+ type = AuIop_OTHER;
-+ break;
++ AuDebugOn(1);
+ }
+
-+ inode->i_op = iop + type;
-+ /* unnecessary smp_wmb() */
-+}
++ if (wh)
++ h_child_inode = NULL;
+
-+int au_refresh_hinode_self(struct inode *inode)
-+{
-+ int err, update;
++ err = -ENOMEM;
++ /* iput() and kfree() will be called in au_hnotify() */
++ args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS);
++ if (unlikely(!args)) {
++ AuErr1("no memory\n");
++ iput(dir);
++ goto out;
++ }
++ args->flags[AuHn_PARENT] = flags[AuHn_PARENT];
++ args->flags[AuHn_CHILD] = flags[AuHn_CHILD];
++ args->mask = mask;
++ args->dir = dir;
++ args->h_dir = igrab(h_dir);
++ if (h_child_inode)
++ h_child_inode = igrab(h_child_inode); /* can be NULL */
++ args->h_child_inode = h_child_inode;
++ args->h_child_nlen = len;
++ if (len) {
++ p = (void *)args;
++ p += sizeof(*args);
++ memcpy(p, h_child_name, len);
++ p[len] = 0;
++ }
+
-+ err = au_ii_refresh(inode, &update);
-+ if (!err)
-+ au_refresh_hinode_attr(inode, update && S_ISDIR(inode->i_mode));
++ /* NFS fires the event for silly-renamed one from kworker */
++ f = 0;
++ if (!dir->i_nlink
++ || (au_test_nfs(h_dir->i_sb) && (mask & FS_DELETE)))
++ f = AuWkq_NEST;
++ err = au_wkq_nowait(au_hn_bh, args, dir->i_sb, f);
++ if (unlikely(err)) {
++ pr_err("wkq %d\n", err);
++ iput(args->h_child_inode);
++ iput(args->h_dir);
++ iput(args->dir);
++ kfree(args);
++ }
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+int au_refresh_hinode(struct inode *inode, struct dentry *dentry)
++/* ---------------------------------------------------------------------- */
++
++int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm)
+{
-+ int err, e, update;
-+ unsigned int flags;
-+ umode_t mode;
-+ aufs_bindex_t bindex, bbot;
-+ unsigned char isdir;
-+ struct au_hinode *p;
-+ struct au_iinfo *iinfo;
++ int err;
+
-+ err = au_ii_refresh(inode, &update);
-+ if (unlikely(err))
-+ goto out;
++ AuDebugOn(!(udba & AuOptMask_UDBA));
+
-+ update = 0;
-+ iinfo = au_ii(inode);
-+ p = au_hinode(iinfo, iinfo->ii_btop);
-+ mode = (inode->i_mode & S_IFMT);
-+ isdir = S_ISDIR(mode);
-+ flags = au_hi_flags(inode, isdir);
-+ bbot = au_dbbot(dentry);
-+ for (bindex = au_dbtop(dentry); bindex <= bbot; bindex++) {
-+ struct inode *h_i, *h_inode;
-+ struct dentry *h_d;
++ err = 0;
++ if (au_hnotify_op.reset_br)
++ err = au_hnotify_op.reset_br(udba, br, perm);
+
-+ h_d = au_h_dptr(dentry, bindex);
-+ if (!h_d || d_is_negative(h_d))
-+ continue;
++ return err;
++}
+
-+ h_inode = d_inode(h_d);
-+ AuDebugOn(mode != (h_inode->i_mode & S_IFMT));
-+ if (iinfo->ii_btop <= bindex && bindex <= iinfo->ii_bbot) {
-+ h_i = au_h_iptr(inode, bindex);
-+ if (h_i) {
-+ if (h_i == h_inode)
-+ continue;
-+ err = -EIO;
-+ break;
-+ }
-+ }
-+ if (bindex < iinfo->ii_btop)
-+ iinfo->ii_btop = bindex;
-+ if (iinfo->ii_bbot < bindex)
-+ iinfo->ii_bbot = bindex;
-+ au_set_h_iptr(inode, bindex, au_igrab(h_inode), flags);
-+ update = 1;
-+ }
-+ au_update_ibrange(inode, /*do_put_zero*/0);
-+ e = au_dy_irefresh(inode);
-+ if (unlikely(e && !err))
-+ err = e;
-+ if (!err)
-+ au_refresh_hinode_attr(inode, update && isdir);
++int au_hnotify_init_br(struct au_branch *br, int perm)
++{
++ int err;
++
++ err = 0;
++ if (au_hnotify_op.init_br)
++ err = au_hnotify_op.init_br(br, perm);
+
-+out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+static int set_inode(struct inode *inode, struct dentry *dentry)
++void au_hnotify_fin_br(struct au_branch *br)
+{
-+ int err;
-+ unsigned int flags;
-+ umode_t mode;
-+ aufs_bindex_t bindex, btop, btail;
-+ unsigned char isdir;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ struct au_iinfo *iinfo;
-+ struct inode_operations *iop;
++ if (au_hnotify_op.fin_br)
++ au_hnotify_op.fin_br(br);
++}
+
-+ IiMustWriteLock(inode);
++static void au_hn_destroy_cache(void)
++{
++ kmem_cache_destroy(au_cache[AuCache_HNOTIFY]);
++ au_cache[AuCache_HNOTIFY] = NULL;
++}
+
-+ err = 0;
-+ isdir = 0;
-+ iop = au_sbi(inode->i_sb)->si_iop_array;
-+ btop = au_dbtop(dentry);
-+ h_dentry = au_h_dptr(dentry, btop);
-+ h_inode = d_inode(h_dentry);
-+ mode = h_inode->i_mode;
-+ switch (mode & S_IFMT) {
-+ case S_IFREG:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = iop + AuIop_OTHER;
-+ inode->i_fop = &aufs_file_fop;
-+ err = au_dy_iaop(inode, btop, h_inode);
++int __init au_hnotify_init(void)
++{
++ int err;
++
++ err = -ENOMEM;
++ au_cache[AuCache_HNOTIFY] = AuCache(au_hnotify);
++ if (au_cache[AuCache_HNOTIFY]) {
++ err = 0;
++ if (au_hnotify_op.init)
++ err = au_hnotify_op.init();
+ if (unlikely(err))
-+ goto out;
-+ break;
-+ case S_IFDIR:
-+ isdir = 1;
-+ btail = au_dbtaildir(dentry);
-+ inode->i_op = iop + AuIop_DIR;
-+ inode->i_fop = &aufs_dir_fop;
-+ break;
-+ case S_IFLNK:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = iop + AuIop_SYMLINK;
-+ break;
-+ case S_IFBLK:
-+ case S_IFCHR:
-+ case S_IFIFO:
-+ case S_IFSOCK:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = iop + AuIop_OTHER;
-+ init_special_inode(inode, mode, h_inode->i_rdev);
-+ break;
-+ default:
-+ AuIOErr("Unknown file type 0%o\n", mode);
-+ err = -EIO;
-+ goto out;
++ au_hn_destroy_cache();
+ }
++ AuTraceErr(err);
++ return err;
++}
+
-+ /* do not set hnotify for whiteouted dirs (SHWH mode) */
-+ flags = au_hi_flags(inode, isdir);
-+ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)
-+ && au_ftest_hi(flags, HNOTIFY)
-+ && dentry->d_name.len > AUFS_WH_PFX_LEN
-+ && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))
-+ au_fclr_hi(flags, HNOTIFY);
-+ iinfo = au_ii(inode);
-+ iinfo->ii_btop = btop;
-+ iinfo->ii_bbot = btail;
-+ for (bindex = btop; bindex <= btail; bindex++) {
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (h_dentry)
-+ au_set_h_iptr(inode, bindex,
-+ au_igrab(d_inode(h_dentry)), flags);
-+ }
-+ au_cpup_attr_all(inode, /*force*/1);
-+ /*
-+ * to force calling aufs_get_acl() every time,
-+ * do not call cache_no_acl() for aufs inode.
-+ */
++void au_hnotify_fin(void)
++{
++ if (au_hnotify_op.fin)
++ au_hnotify_op.fin();
+
-+out:
-+ return err;
++ /* cf. au_cache_fin() */
++ if (au_cache[AuCache_HNOTIFY])
++ au_hn_destroy_cache();
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/i_op.c linux-4.14/fs/aufs/i_op.c
+--- linux-4.14/fs/aufs/i_op.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/i_op.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,1459 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
+/*
-+ * successful returns with iinfo write_locked
-+ * minus: errno
-+ * zero: success, matched
-+ * plus: no error, but unmatched
++ * inode operations (except add/del/rename)
+ */
-+static int reval_inode(struct inode *inode, struct dentry *dentry)
++
++#include <linux/device_cgroup.h>
++#include <linux/fs_stack.h>
++#include <linux/namei.h>
++#include <linux/security.h>
++#include "aufs.h"
++
++static int h_permission(struct inode *h_inode, int mask,
++ struct path *h_path, int brperm)
+{
+ int err;
-+ unsigned int gen, igflags;
-+ aufs_bindex_t bindex, bbot;
-+ struct inode *h_inode, *h_dinode;
-+ struct dentry *h_dentry;
++ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
+
-+ /*
-+ * before this function, if aufs got any iinfo lock, it must be only
-+ * one, the parent dir.
-+ * it can happen by UDBA and the obsoleted inode number.
-+ */
-+ err = -EIO;
-+ if (unlikely(inode->i_ino == parent_ino(dentry)))
++ err = -EPERM;
++ if (write_mask && IS_IMMUTABLE(h_inode))
+ goto out;
+
-+ err = 1;
-+ ii_write_lock_new_child(inode);
-+ h_dentry = au_h_dptr(dentry, au_dbtop(dentry));
-+ h_dinode = d_inode(h_dentry);
-+ bbot = au_ibbot(inode);
-+ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (!h_inode || h_inode != h_dinode)
-+ continue;
++ err = -EACCES;
++ if (((mask & MAY_EXEC)
++ && S_ISREG(h_inode->i_mode)
++ && (path_noexec(h_path)
++ || !(h_inode->i_mode & S_IXUGO))))
++ goto out;
+
-+ err = 0;
-+ gen = au_iigen(inode, &igflags);
-+ if (gen == au_digen(dentry)
-+ && !au_ig_ftest(igflags, HALF_REFRESHED))
-+ break;
++ /*
++ * - skip the lower fs test in the case of write to ro branch.
++ * - nfs dir permission write check is optimized, but a policy for
++ * link/rename requires a real check.
++ * - nfs always sets MS_POSIXACL regardless its mount option 'noacl.'
++ * in this case, generic_permission() returns -EOPNOTSUPP.
++ */
++ if ((write_mask && !au_br_writable(brperm))
++ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode)
++ && write_mask && !(mask & MAY_READ))
++ || !h_inode->i_op->permission) {
++ /* AuLabel(generic_permission); */
++ /* AuDbg("get_acl %pf\n", h_inode->i_op->get_acl); */
++ err = generic_permission(h_inode, mask);
++ if (err == -EOPNOTSUPP && au_test_nfs_noacl(h_inode))
++ err = h_inode->i_op->permission(h_inode, mask);
++ AuTraceErr(err);
++ } else {
++ /* AuLabel(h_inode->permission); */
++ err = h_inode->i_op->permission(h_inode, mask);
++ AuTraceErr(err);
++ }
+
-+ /* fully refresh inode using dentry */
-+ err = au_refresh_hinode(inode, dentry);
-+ if (!err)
-+ au_update_iigen(inode, /*half*/0);
-+ break;
++ if (!err)
++ err = devcgroup_inode_permission(h_inode, mask);
++ if (!err)
++ err = security_inode_permission(h_inode, mask);
++
++#if 0
++ if (!err) {
++ /* todo: do we need to call ima_path_check()? */
++ struct path h_path = {
++ .dentry =
++ .mnt = h_mnt
++ };
++ err = ima_path_check(&h_path,
++ mask & (MAY_READ | MAY_WRITE | MAY_EXEC),
++ IMA_COUNT_LEAVE);
+ }
++#endif
+
-+ if (unlikely(err))
-+ ii_write_unlock(inode);
+out:
+ return err;
+}
+
-+int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ unsigned int d_type, ino_t *ino)
++static int aufs_permission(struct inode *inode, int mask)
+{
+ int err;
-+ struct mutex *mtx;
++ aufs_bindex_t bindex, bbot;
++ const unsigned char isdir = !!S_ISDIR(inode->i_mode),
++ write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct au_branch *br;
+
-+ /* prevent hardlinked inode number from race condition */
-+ mtx = NULL;
-+ if (d_type != DT_DIR) {
-+ mtx = &au_sbr(sb, bindex)->br_xino.xi_nondir_mtx;
-+ mutex_lock(mtx);
-+ }
-+ err = au_xino_read(sb, bindex, h_ino, ino);
++ /* todo: support rcu-walk? */
++ if (mask & MAY_NOT_BLOCK)
++ return -ECHILD;
++
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ ii_read_lock_child(inode);
++#if 0
++ err = au_iigen_test(inode, au_sigen(sb));
+ if (unlikely(err))
+ goto out;
++#endif
+
-+ if (!*ino) {
-+ err = -EIO;
-+ *ino = au_xino_new_ino(sb);
-+ if (unlikely(!*ino))
-+ goto out;
-+ err = au_xino_write(sb, bindex, h_ino, *ino);
-+ if (unlikely(err))
++ if (!isdir
++ || write_mask
++ || au_opt_test(au_mntflags(sb), DIRPERM1)) {
++ err = au_busy_or_stale();
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
++ if (unlikely(!h_inode
++ || (h_inode->i_mode & S_IFMT)
++ != (inode->i_mode & S_IFMT)))
+ goto out;
++
++ err = 0;
++ bindex = au_ibtop(inode);
++ br = au_sbr(sb, bindex);
++ err = h_permission(h_inode, mask, &br->br_path, br->br_perm);
++ if (write_mask
++ && !err
++ && !special_file(h_inode->i_mode)) {
++ /* test whether the upper writable branch exists */
++ err = -EROFS;
++ for (; bindex >= 0; bindex--)
++ if (!au_br_rdonly(au_sbr(sb, bindex))) {
++ err = 0;
++ break;
++ }
++ }
++ goto out;
++ }
++
++ /* non-write to dir */
++ err = 0;
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); !err && bindex <= bbot; bindex++) {
++ h_inode = au_h_iptr(inode, bindex);
++ if (h_inode) {
++ err = au_busy_or_stale();
++ if (unlikely(!S_ISDIR(h_inode->i_mode)))
++ break;
++
++ br = au_sbr(sb, bindex);
++ err = h_permission(h_inode, mask, &br->br_path,
++ br->br_perm);
++ }
+ }
+
+out:
-+ if (mtx)
-+ mutex_unlock(mtx);
++ ii_read_unlock(inode);
++ si_read_unlock(sb);
+ return err;
+}
+
-+/* successful returns with iinfo write_locked */
-+/* todo: return with unlocked? */
-+struct inode *au_new_inode(struct dentry *dentry, int must_new)
++/* ---------------------------------------------------------------------- */
++
++static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry,
++ unsigned int flags)
+{
-+ struct inode *inode, *h_inode;
-+ struct dentry *h_dentry;
++ struct dentry *ret, *parent;
++ struct inode *inode;
+ struct super_block *sb;
-+ struct mutex *mtx;
-+ ino_t h_ino, ino;
-+ int err;
-+ aufs_bindex_t btop;
++ int err, npositive;
+
-+ sb = dentry->d_sb;
-+ btop = au_dbtop(dentry);
-+ h_dentry = au_h_dptr(dentry, btop);
-+ h_inode = d_inode(h_dentry);
-+ h_ino = h_inode->i_ino;
++ IMustLock(dir);
+
-+ /*
-+ * stop 'race'-ing between hardlinks under different
-+ * parents.
-+ */
-+ mtx = NULL;
-+ if (!d_is_dir(h_dentry))
-+ mtx = &au_sbr(sb, btop)->br_xino.xi_nondir_mtx;
++ /* todo: support rcu-walk? */
++ ret = ERR_PTR(-ECHILD);
++ if (flags & LOOKUP_RCU)
++ goto out;
+
-+new_ino:
-+ if (mtx)
-+ mutex_lock(mtx);
-+ err = au_xino_read(sb, btop, h_ino, &ino);
-+ inode = ERR_PTR(err);
++ ret = ERR_PTR(-ENAMETOOLONG);
++ if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN))
++ goto out;
++
++ sb = dir->i_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ ret = ERR_PTR(err);
+ if (unlikely(err))
+ goto out;
+
-+ if (!ino) {
-+ ino = au_xino_new_ino(sb);
-+ if (unlikely(!ino)) {
-+ inode = ERR_PTR(-EIO);
-+ goto out;
-+ }
-+ }
++ err = au_di_init(dentry);
++ ret = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_si;
+
-+ AuDbg("i%lu\n", (unsigned long)ino);
-+ inode = au_iget_locked(sb, ino);
-+ err = PTR_ERR(inode);
-+ if (IS_ERR(inode))
-+ goto out;
++ inode = NULL;
++ npositive = 0; /* suppress a warning */
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_read_lock_parent(parent, AuLock_IR);
++ err = au_alive_dir(parent);
++ if (!err)
++ err = au_digen_test(parent, au_sigen(sb));
++ if (!err) {
++ /* regardless LOOKUP_CREATE, always ALLOW_NEG */
++ npositive = au_lkup_dentry(dentry, au_dbtop(parent),
++ AuLkup_ALLOW_NEG);
++ err = npositive;
++ }
++ di_read_unlock(parent, AuLock_IR);
++ ret = ERR_PTR(err);
++ if (unlikely(err < 0))
++ goto out_unlock;
+
-+ AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW));
-+ if (inode->i_state & I_NEW) {
-+ ii_write_lock_new_child(inode);
-+ err = set_inode(inode, dentry);
-+ if (!err) {
-+ unlock_new_inode(inode);
-+ goto out; /* success */
++ if (npositive) {
++ inode = au_new_inode(dentry, /*must_new*/0);
++ if (IS_ERR(inode)) {
++ ret = (void *)inode;
++ inode = NULL;
++ goto out_unlock;
+ }
++ }
+
-+ /*
-+ * iget_failed() calls iput(), but we need to call
-+ * ii_write_unlock() after iget_failed(). so dirty hack for
-+ * i_count.
-+ */
++ if (inode)
+ atomic_inc(&inode->i_count);
-+ iget_failed(inode);
-+ ii_write_unlock(inode);
-+ au_xino_write(sb, btop, h_ino, /*ino*/0);
-+ /* ignore this error */
-+ goto out_iput;
-+ } else if (!must_new && !IS_DEADDIR(inode) && inode->i_nlink) {
-+ /*
-+ * horrible race condition between lookup, readdir and copyup
-+ * (or something).
-+ */
-+ if (mtx)
-+ mutex_unlock(mtx);
-+ err = reval_inode(inode, dentry);
-+ if (unlikely(err < 0)) {
-+ mtx = NULL;
-+ goto out_iput;
++ ret = d_splice_alias(inode, dentry);
++#if 0
++ if (unlikely(d_need_lookup(dentry))) {
++ spin_lock(&dentry->d_lock);
++ dentry->d_flags &= ~DCACHE_NEED_LOOKUP;
++ spin_unlock(&dentry->d_lock);
++ } else
++#endif
++ if (inode) {
++ if (!IS_ERR(ret)) {
++ iput(inode);
++ if (ret && ret != dentry)
++ ii_write_unlock(inode);
++ } else {
++ ii_write_unlock(inode);
++ iput(inode);
++ inode = NULL;
+ }
-+
-+ if (!err) {
-+ mtx = NULL;
-+ goto out; /* success */
-+ } else if (mtx)
-+ mutex_lock(mtx);
-+ }
-+
-+ if (unlikely(au_test_fs_unique_ino(h_inode)))
-+ AuWarn1("Warning: Un-notified UDBA or repeatedly renamed dir,"
-+ " b%d, %s, %pd, hi%lu, i%lu.\n",
-+ btop, au_sbtype(h_dentry->d_sb), dentry,
-+ (unsigned long)h_ino, (unsigned long)ino);
-+ ino = 0;
-+ err = au_xino_write(sb, btop, h_ino, /*ino*/0);
-+ if (!err) {
-+ iput(inode);
-+ if (mtx)
-+ mutex_unlock(mtx);
-+ goto new_ino;
+ }
+
-+out_iput:
-+ iput(inode);
-+ inode = ERR_PTR(err);
++out_unlock:
++ di_write_unlock(dentry);
++out_si:
++ si_read_unlock(sb);
+out:
-+ if (mtx)
-+ mutex_unlock(mtx);
-+ return inode;
++ return ret;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode)
-+{
-+ int err;
-+ struct inode *hi;
++struct aopen_node {
++ struct hlist_bl_node hblist;
++ struct file *file, *h_file;
++};
+
-+ err = au_br_rdonly(au_sbr(sb, bindex));
++static int au_do_aopen(struct inode *inode, struct file *file)
++{
++ struct hlist_bl_head *aopen;
++ struct hlist_bl_node *pos;
++ struct aopen_node *node;
++ struct au_do_open_args args = {
++ .aopen = 1,
++ .open = au_do_open_nondir
++ };
+
-+ /* pseudo-link after flushed may happen out of bounds */
-+ if (!err
-+ && inode
-+ && au_ibtop(inode) <= bindex
-+ && bindex <= au_ibbot(inode)) {
-+ /*
-+ * permission check is unnecessary since vfsub routine
-+ * will be called later
-+ */
-+ hi = au_h_iptr(inode, bindex);
-+ if (hi)
-+ err = IS_IMMUTABLE(hi) ? -EROFS : 0;
-+ }
++ aopen = &au_sbi(inode->i_sb)->si_aopen;
++ hlist_bl_lock(aopen);
++ hlist_bl_for_each_entry(node, pos, aopen, hblist)
++ if (node->file == file) {
++ args.h_file = node->h_file;
++ break;
++ }
++ hlist_bl_unlock(aopen);
++ /* AuDebugOn(!args.h_file); */
+
-+ return err;
++ return au_do_open(file, &args);
+}
+
-+int au_test_h_perm(struct inode *h_inode, int mask)
++static int aufs_atomic_open(struct inode *dir, struct dentry *dentry,
++ struct file *file, unsigned int open_flag,
++ umode_t create_mode, int *opened)
+{
-+ if (uid_eq(current_fsuid(), GLOBAL_ROOT_UID))
-+ return 0;
-+ return inode_permission(h_inode, mask);
-+}
++ int err, unlocked, h_opened = *opened;
++ unsigned int lkup_flags;
++ struct dentry *parent, *d;
++ struct hlist_bl_head *aopen;
++ struct vfsub_aopen_args args = {
++ .open_flag = open_flag,
++ .create_mode = create_mode,
++ .opened = &h_opened
++ };
++ struct aopen_node aopen_node = {
++ .file = file
++ };
+
-+int au_test_h_perm_sio(struct inode *h_inode, int mask)
-+{
-+ if (au_test_nfs(h_inode->i_sb)
-+ && (mask & MAY_WRITE)
-+ && S_ISDIR(h_inode->i_mode))
-+ mask |= MAY_READ; /* force permission check */
-+ return au_test_h_perm(h_inode, mask);
-+}
-diff -urN /usr/share/empty/fs/aufs/inode.h linux/fs/aufs/inode.h
---- /usr/share/empty/fs/aufs/inode.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/inode.h 2016-08-17 18:01:06.151889691 +0200
-@@ -0,0 +1,700 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
++ IMustLock(dir);
++ AuDbg("open_flag 0%o\n", open_flag);
++ AuDbgDentry(dentry);
+
-+/*
-+ * inode operations
-+ */
++ err = 0;
++ if (!au_di(dentry)) {
++ lkup_flags = LOOKUP_OPEN;
++ if (open_flag & O_CREAT)
++ lkup_flags |= LOOKUP_CREATE;
++ d = aufs_lookup(dir, dentry, lkup_flags);
++ if (IS_ERR(d)) {
++ err = PTR_ERR(d);
++ AuTraceErr(err);
++ goto out;
++ } else if (d) {
++ /*
++ * obsoleted dentry found.
++ * another error will be returned later.
++ */
++ d_drop(d);
++ AuDbgDentry(d);
++ dput(d);
++ }
++ AuDbgDentry(dentry);
++ }
+
-+#ifndef __AUFS_INODE_H__
-+#define __AUFS_INODE_H__
++ if (d_is_positive(dentry)
++ || d_unhashed(dentry)
++ || d_unlinked(dentry)
++ || !(open_flag & O_CREAT))
++ goto out_no_open;
+
-+#ifdef __KERNEL__
++ unlocked = 0;
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
++ if (unlikely(err))
++ goto out;
+
-+#include <linux/fsnotify.h>
-+#include "rwsem.h"
++ parent = dentry->d_parent; /* dir is locked */
++ di_write_lock_parent(parent);
++ err = au_lkup_dentry(dentry, /*btop*/0, AuLkup_ALLOW_NEG);
++ if (unlikely(err))
++ goto out_unlock;
+
-+struct vfsmount;
++ AuDbgDentry(dentry);
++ if (d_is_positive(dentry))
++ goto out_unlock;
+
-+struct au_hnotify {
-+#ifdef CONFIG_AUFS_HNOTIFY
-+#ifdef CONFIG_AUFS_HFSNOTIFY
-+ /* never use fsnotify_add_vfsmount_mark() */
-+ struct fsnotify_mark hn_mark;
-+#endif
-+ union {
-+ struct inode *hn_aufs_inode; /* no get/put */
-+ struct llist_node hn_lnode; /* delayed free */
-+ };
-+#endif
-+} ____cacheline_aligned_in_smp;
++ args.file = get_empty_filp();
++ err = PTR_ERR(args.file);
++ if (IS_ERR(args.file))
++ goto out_unlock;
+
-+struct au_hinode {
-+ struct inode *hi_inode;
-+ aufs_bindex_t hi_id;
-+#ifdef CONFIG_AUFS_HNOTIFY
-+ struct au_hnotify *hi_notify;
-+#endif
++ args.file->f_flags = file->f_flags;
++ err = au_aopen_or_create(dir, dentry, &args);
++ AuTraceErr(err);
++ AuDbgFile(args.file);
++ if (unlikely(err < 0)) {
++ if (h_opened & FILE_OPENED)
++ fput(args.file);
++ else
++ put_filp(args.file);
++ goto out_unlock;
++ }
++ di_write_unlock(parent);
++ di_write_unlock(dentry);
++ unlocked = 1;
+
-+ /* reference to the copied-up whiteout with get/put */
-+ struct dentry *hi_whdentry;
-+};
++ /* some filesystems don't set FILE_CREATED while succeeded? */
++ *opened |= FILE_CREATED;
++ if (h_opened & FILE_OPENED)
++ aopen_node.h_file = args.file;
++ else {
++ put_filp(args.file);
++ args.file = NULL;
++ }
++ aopen = &au_sbi(dir->i_sb)->si_aopen;
++ au_hbl_add(&aopen_node.hblist, aopen);
++ err = finish_open(file, dentry, au_do_aopen, opened);
++ au_hbl_del(&aopen_node.hblist, aopen);
++ AuTraceErr(err);
++ AuDbgFile(file);
++ if (aopen_node.h_file)
++ fput(aopen_node.h_file);
++
++out_unlock:
++ if (unlocked)
++ si_read_unlock(dentry->d_sb);
++ else {
++ di_write_unlock(parent);
++ aufs_read_unlock(dentry, AuLock_DW);
++ }
++ AuDbgDentry(dentry);
++ if (unlikely(err < 0))
++ goto out;
++out_no_open:
++ if (err >= 0 && !(*opened & FILE_CREATED)) {
++ AuLabel(out_no_open);
++ dget(dentry);
++ err = finish_no_open(file, dentry);
++ }
++out:
++ AuDbg("%pd%s%s\n", dentry,
++ (*opened & FILE_CREATED) ? " created" : "",
++ (*opened & FILE_OPENED) ? " opened" : "");
++ AuTraceErr(err);
++ return err;
++}
++
++
++/* ---------------------------------------------------------------------- */
++
++static int au_wr_dir_cpup(struct dentry *dentry, struct dentry *parent,
++ const unsigned char add_entry, aufs_bindex_t bcpup,
++ aufs_bindex_t btop)
++{
++ int err;
++ struct dentry *h_parent;
++ struct inode *h_dir;
++
++ if (add_entry)
++ IMustLock(d_inode(parent));
++ else
++ di_write_lock_parent(parent);
+
-+/* ig_flags */
-+#define AuIG_HALF_REFRESHED 1
-+#define au_ig_ftest(flags, name) ((flags) & AuIG_##name)
-+#define au_ig_fset(flags, name) \
-+ do { (flags) |= AuIG_##name; } while (0)
-+#define au_ig_fclr(flags, name) \
-+ do { (flags) &= ~AuIG_##name; } while (0)
++ err = 0;
++ if (!au_h_dptr(parent, bcpup)) {
++ if (btop > bcpup)
++ err = au_cpup_dirs(dentry, bcpup);
++ else if (btop < bcpup)
++ err = au_cpdown_dirs(dentry, bcpup);
++ else
++ BUG();
++ }
++ if (!err && add_entry && !au_ftest_wrdir(add_entry, TMPFILE)) {
++ h_parent = au_h_dptr(parent, bcpup);
++ h_dir = d_inode(h_parent);
++ vfsub_inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
++ err = au_lkup_neg(dentry, bcpup, /*wh*/0);
++ /* todo: no unlock here */
++ inode_unlock_shared(h_dir);
+
-+struct au_iigen {
-+ spinlock_t ig_spin;
-+ __u32 ig_generation, ig_flags;
-+};
++ AuDbg("bcpup %d\n", bcpup);
++ if (!err) {
++ if (d_really_is_negative(dentry))
++ au_set_h_dptr(dentry, btop, NULL);
++ au_update_dbrange(dentry, /*do_put_zero*/0);
++ }
++ }
+
-+struct au_vdir;
-+struct au_iinfo {
-+ struct au_iigen ii_generation;
-+ struct super_block *ii_hsb1; /* no get/put */
++ if (!add_entry)
++ di_write_unlock(parent);
++ if (!err)
++ err = bcpup; /* success */
+
-+ struct au_rwsem ii_rwsem;
-+ aufs_bindex_t ii_btop, ii_bbot;
-+ __u32 ii_higen;
-+ struct au_hinode *ii_hinode;
-+ struct au_vdir *ii_vdir;
-+};
++ AuTraceErr(err);
++ return err;
++}
+
-+struct au_icntnr {
-+ struct au_iinfo iinfo;
-+ struct inode vfs_inode;
-+ union {
-+ struct hlist_node plink;
-+ struct llist_node lnode; /* delayed free */
-+ };
-+} ____cacheline_aligned_in_smp;
++/*
++ * decide the branch and the parent dir where we will create a new entry.
++ * returns new bindex or an error.
++ * copyup the parent dir if needed.
++ */
++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
++ struct au_wr_dir_args *args)
++{
++ int err;
++ unsigned int flags;
++ aufs_bindex_t bcpup, btop, src_btop;
++ const unsigned char add_entry
++ = au_ftest_wrdir(args->flags, ADD_ENTRY)
++ | au_ftest_wrdir(args->flags, TMPFILE);
++ struct super_block *sb;
++ struct dentry *parent;
++ struct au_sbinfo *sbinfo;
+
-+/* au_pin flags */
-+#define AuPin_DI_LOCKED 1
-+#define AuPin_MNT_WRITE (1 << 1)
-+#define au_ftest_pin(flags, name) ((flags) & AuPin_##name)
-+#define au_fset_pin(flags, name) \
-+ do { (flags) |= AuPin_##name; } while (0)
-+#define au_fclr_pin(flags, name) \
-+ do { (flags) &= ~AuPin_##name; } while (0)
++ sb = dentry->d_sb;
++ sbinfo = au_sbi(sb);
++ parent = dget_parent(dentry);
++ btop = au_dbtop(dentry);
++ bcpup = btop;
++ if (args->force_btgt < 0) {
++ if (src_dentry) {
++ src_btop = au_dbtop(src_dentry);
++ if (src_btop < btop)
++ bcpup = src_btop;
++ } else if (add_entry) {
++ flags = 0;
++ if (au_ftest_wrdir(args->flags, ISDIR))
++ au_fset_wbr(flags, DIR);
++ err = AuWbrCreate(sbinfo, dentry, flags);
++ bcpup = err;
++ }
+
-+struct au_pin {
-+ /* input */
-+ struct dentry *dentry;
-+ unsigned int udba;
-+ unsigned char lsc_di, lsc_hi, flags;
-+ aufs_bindex_t bindex;
++ if (bcpup < 0 || au_test_ro(sb, bcpup, d_inode(dentry))) {
++ if (add_entry)
++ err = AuWbrCopyup(sbinfo, dentry);
++ else {
++ if (!IS_ROOT(dentry)) {
++ di_read_lock_parent(parent, !AuLock_IR);
++ err = AuWbrCopyup(sbinfo, dentry);
++ di_read_unlock(parent, !AuLock_IR);
++ } else
++ err = AuWbrCopyup(sbinfo, dentry);
++ }
++ bcpup = err;
++ if (unlikely(err < 0))
++ goto out;
++ }
++ } else {
++ bcpup = args->force_btgt;
++ AuDebugOn(au_test_ro(sb, bcpup, d_inode(dentry)));
++ }
+
-+ /* output */
-+ struct dentry *parent;
-+ struct au_hinode *hdir;
-+ struct vfsmount *h_mnt;
++ AuDbg("btop %d, bcpup %d\n", btop, bcpup);
++ err = bcpup;
++ if (bcpup == btop)
++ goto out; /* success */
+
-+ /* temporary unlock/relock for copyup */
-+ struct dentry *h_dentry, *h_parent;
-+ struct au_branch *br;
-+ struct task_struct *task;
-+};
++ /* copyup the new parent into the branch we process */
++ err = au_wr_dir_cpup(dentry, parent, add_entry, bcpup, btop);
++ if (err >= 0) {
++ if (d_really_is_negative(dentry)) {
++ au_set_h_dptr(dentry, btop, NULL);
++ au_set_dbtop(dentry, bcpup);
++ au_set_dbbot(dentry, bcpup);
++ }
++ AuDebugOn(add_entry
++ && !au_ftest_wrdir(args->flags, TMPFILE)
++ && !au_h_dptr(dentry, bcpup));
++ }
+
-+void au_pin_hdir_unlock(struct au_pin *p);
-+int au_pin_hdir_lock(struct au_pin *p);
-+int au_pin_hdir_relock(struct au_pin *p);
-+void au_pin_hdir_acquire_nest(struct au_pin *p);
-+void au_pin_hdir_release(struct au_pin *p);
++out:
++ dput(parent);
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+static inline struct au_iinfo *au_ii(struct inode *inode)
++void au_pin_hdir_unlock(struct au_pin *p)
+{
-+ BUG_ON(is_bad_inode(inode));
-+ return &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
++ if (p->hdir)
++ au_hn_inode_unlock(p->hdir);
+}
+
-+/* ---------------------------------------------------------------------- */
++int au_pin_hdir_lock(struct au_pin *p)
++{
++ int err;
+
-+/* inode.c */
-+struct inode *au_igrab(struct inode *inode);
-+void au_refresh_iop(struct inode *inode, int force_getattr);
-+int au_refresh_hinode_self(struct inode *inode);
-+int au_refresh_hinode(struct inode *inode, struct dentry *dentry);
-+int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ unsigned int d_type, ino_t *ino);
-+struct inode *au_new_inode(struct dentry *dentry, int must_new);
-+int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode);
-+int au_test_h_perm(struct inode *h_inode, int mask);
-+int au_test_h_perm_sio(struct inode *h_inode, int mask);
++ err = 0;
++ if (!p->hdir)
++ goto out;
+
-+static inline int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex,
-+ ino_t h_ino, unsigned int d_type, ino_t *ino)
-+{
-+#ifdef CONFIG_AUFS_SHWH
-+ return au_ino(sb, bindex, h_ino, d_type, ino);
-+#else
-+ return 0;
-+#endif
-+}
++ /* even if an error happens later, keep this lock */
++ au_hn_inode_lock_nested(p->hdir, p->lsc_hi);
+
-+/* i_op.c */
-+enum {
-+ AuIop_SYMLINK,
-+ AuIop_DIR,
-+ AuIop_OTHER,
-+ AuIop_Last
-+};
-+extern struct inode_operations aufs_iop[AuIop_Last],
-+ aufs_iop_nogetattr[AuIop_Last];
++ err = -EBUSY;
++ if (unlikely(p->hdir->hi_inode != d_inode(p->h_parent)))
++ goto out;
+
-+/* au_wr_dir flags */
-+#define AuWrDir_ADD_ENTRY 1
-+#define AuWrDir_ISDIR (1 << 1)
-+#define AuWrDir_TMPFILE (1 << 2)
-+#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name)
-+#define au_fset_wrdir(flags, name) \
-+ do { (flags) |= AuWrDir_##name; } while (0)
-+#define au_fclr_wrdir(flags, name) \
-+ do { (flags) &= ~AuWrDir_##name; } while (0)
++ err = 0;
++ if (p->h_dentry)
++ err = au_h_verify(p->h_dentry, p->udba, p->hdir->hi_inode,
++ p->h_parent, p->br);
+
-+struct au_wr_dir_args {
-+ aufs_bindex_t force_btgt;
-+ unsigned char flags;
-+};
-+int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
-+ struct au_wr_dir_args *args);
++out:
++ return err;
++}
+
-+struct dentry *au_pinned_h_parent(struct au_pin *pin);
-+void au_pin_init(struct au_pin *pin, struct dentry *dentry,
-+ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
-+ unsigned int udba, unsigned char flags);
-+int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned int udba, unsigned char flags) __must_check;
-+int au_do_pin(struct au_pin *pin) __must_check;
-+void au_unpin(struct au_pin *pin);
-+int au_reval_for_attr(struct dentry *dentry, unsigned int sigen);
++int au_pin_hdir_relock(struct au_pin *p)
++{
++ int err, i;
++ struct inode *h_i;
++ struct dentry *h_d[] = {
++ p->h_dentry,
++ p->h_parent
++ };
+
-+#define AuIcpup_DID_CPUP 1
-+#define au_ftest_icpup(flags, name) ((flags) & AuIcpup_##name)
-+#define au_fset_icpup(flags, name) \
-+ do { (flags) |= AuIcpup_##name; } while (0)
-+#define au_fclr_icpup(flags, name) \
-+ do { (flags) &= ~AuIcpup_##name; } while (0)
++ err = au_pin_hdir_lock(p);
++ if (unlikely(err))
++ goto out;
+
-+struct au_icpup_args {
-+ unsigned char flags;
-+ unsigned char pin_flags;
-+ aufs_bindex_t btgt;
-+ unsigned int udba;
-+ struct au_pin pin;
-+ struct path h_path;
-+ struct inode *h_inode;
-+};
++ for (i = 0; !err && i < sizeof(h_d)/sizeof(*h_d); i++) {
++ if (!h_d[i])
++ continue;
++ if (d_is_positive(h_d[i])) {
++ h_i = d_inode(h_d[i]);
++ err = !h_i->i_nlink;
++ }
++ }
+
-+int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
-+ struct au_icpup_args *a);
++out:
++ return err;
++}
+
-+int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path);
++static void au_pin_hdir_set_owner(struct au_pin *p, struct task_struct *task)
++{
++#if !defined(CONFIG_RWSEM_GENERIC_SPINLOCK) && defined(CONFIG_RWSEM_SPIN_ON_OWNER)
++ p->hdir->hi_inode->i_rwsem.owner = task;
++#endif
++}
+
-+/* i_op_add.c */
-+int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir);
-+int aufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode,
-+ dev_t dev);
-+int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname);
-+int aufs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
-+ bool want_excl);
-+struct vfsub_aopen_args;
-+int au_aopen_or_create(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_aopen_args *args);
-+int aufs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode);
-+int aufs_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry);
-+int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode);
++void au_pin_hdir_acquire_nest(struct au_pin *p)
++{
++ if (p->hdir) {
++ rwsem_acquire_nest(&p->hdir->hi_inode->i_rwsem.dep_map,
++ p->lsc_hi, 0, NULL, _RET_IP_);
++ au_pin_hdir_set_owner(p, current);
++ }
++}
+
-+/* i_op_del.c */
-+int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup);
-+int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir);
-+int aufs_unlink(struct inode *dir, struct dentry *dentry);
-+int aufs_rmdir(struct inode *dir, struct dentry *dentry);
++void au_pin_hdir_release(struct au_pin *p)
++{
++ if (p->hdir) {
++ au_pin_hdir_set_owner(p, p->task);
++ rwsem_release(&p->hdir->hi_inode->i_rwsem.dep_map, 1, _RET_IP_);
++ }
++}
+
-+/* i_op_ren.c */
-+int au_wbr(struct dentry *dentry, aufs_bindex_t btgt);
-+int aufs_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry);
++struct dentry *au_pinned_h_parent(struct au_pin *pin)
++{
++ if (pin && pin->parent)
++ return au_h_dptr(pin->parent, pin->bindex);
++ return NULL;
++}
+
-+/* iinfo.c */
-+struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex);
-+void au_hiput(struct au_hinode *hinode);
-+void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_wh);
-+unsigned int au_hi_flags(struct inode *inode, int isdir);
++void au_unpin(struct au_pin *p)
++{
++ if (p->hdir)
++ au_pin_hdir_unlock(p);
++ if (p->h_mnt && au_ftest_pin(p->flags, MNT_WRITE))
++ vfsub_mnt_drop_write(p->h_mnt);
++ if (!p->hdir)
++ return;
+
-+/* hinode flags */
-+#define AuHi_XINO 1
-+#define AuHi_HNOTIFY (1 << 1)
-+#define au_ftest_hi(flags, name) ((flags) & AuHi_##name)
-+#define au_fset_hi(flags, name) \
-+ do { (flags) |= AuHi_##name; } while (0)
-+#define au_fclr_hi(flags, name) \
-+ do { (flags) &= ~AuHi_##name; } while (0)
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ iput(p->hdir->hi_inode);
++ dput(p->parent);
++ p->parent = NULL;
++ p->hdir = NULL;
++ p->h_mnt = NULL;
++ /* do not clear p->task */
++}
+
-+#ifndef CONFIG_AUFS_HNOTIFY
-+#undef AuHi_HNOTIFY
-+#define AuHi_HNOTIFY 0
-+#endif
++int au_do_pin(struct au_pin *p)
++{
++ int err;
++ struct super_block *sb;
++ struct inode *h_dir;
+
-+void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode, unsigned int flags);
++ err = 0;
++ sb = p->dentry->d_sb;
++ p->br = au_sbr(sb, p->bindex);
++ if (IS_ROOT(p->dentry)) {
++ if (au_ftest_pin(p->flags, MNT_WRITE)) {
++ p->h_mnt = au_br_mnt(p->br);
++ err = vfsub_mnt_want_write(p->h_mnt);
++ if (unlikely(err)) {
++ au_fclr_pin(p->flags, MNT_WRITE);
++ goto out_err;
++ }
++ }
++ goto out;
++ }
+
-+void au_update_iigen(struct inode *inode, int half);
-+void au_update_ibrange(struct inode *inode, int do_put_zero);
++ p->h_dentry = NULL;
++ if (p->bindex <= au_dbbot(p->dentry))
++ p->h_dentry = au_h_dptr(p->dentry, p->bindex);
+
-+void au_icntnr_init_once(void *_c);
-+void au_hinode_init(struct au_hinode *hinode);
-+int au_iinfo_init(struct inode *inode);
-+void au_iinfo_fin(struct inode *inode);
-+int au_hinode_realloc(struct au_iinfo *iinfo, int nbr);
++ p->parent = dget_parent(p->dentry);
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_lock(p->parent, AuLock_IR, p->lsc_di);
+
-+#ifdef CONFIG_PROC_FS
-+/* plink.c */
-+int au_plink_maint(struct super_block *sb, int flags);
-+struct au_sbinfo;
-+void au_plink_maint_leave(struct au_sbinfo *sbinfo);
-+int au_plink_maint_enter(struct super_block *sb);
-+#ifdef CONFIG_AUFS_DEBUG
-+void au_plink_list(struct super_block *sb);
-+#else
-+AuStubVoid(au_plink_list, struct super_block *sb)
-+#endif
-+int au_plink_test(struct inode *inode);
-+struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex);
-+void au_plink_append(struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_dentry);
-+void au_plink_put(struct super_block *sb, int verbose);
-+void au_plink_clean(struct super_block *sb, int verbose);
-+void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id);
-+#else
-+AuStubInt0(au_plink_maint, struct super_block *sb, int flags);
-+AuStubVoid(au_plink_maint_leave, struct au_sbinfo *sbinfo);
-+AuStubInt0(au_plink_maint_enter, struct super_block *sb);
-+AuStubVoid(au_plink_list, struct super_block *sb);
-+AuStubInt0(au_plink_test, struct inode *inode);
-+AuStub(struct dentry *, au_plink_lkup, return NULL,
-+ struct inode *inode, aufs_bindex_t bindex);
-+AuStubVoid(au_plink_append, struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_dentry);
-+AuStubVoid(au_plink_put, struct super_block *sb, int verbose);
-+AuStubVoid(au_plink_clean, struct super_block *sb, int verbose);
-+AuStubVoid(au_plink_half_refresh, struct super_block *sb, aufs_bindex_t br_id);
-+#endif /* CONFIG_PROC_FS */
++ h_dir = NULL;
++ p->h_parent = au_h_dptr(p->parent, p->bindex);
++ p->hdir = au_hi(d_inode(p->parent), p->bindex);
++ if (p->hdir)
++ h_dir = p->hdir->hi_inode;
+
-+#ifdef CONFIG_AUFS_XATTR
-+/* xattr.c */
-+int au_cpup_xattr(struct dentry *h_dst, struct dentry *h_src, int ignore_flags,
-+ unsigned int verbose);
-+ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size);
-+ssize_t aufs_getxattr(struct dentry *dentry, struct inode *inode,
-+ const char *name, void *value, size_t size);
-+int aufs_setxattr(struct dentry *dentry, struct inode *inode, const char *name,
-+ const void *value, size_t size, int flags);
-+int aufs_removexattr(struct dentry *dentry, const char *name);
++ /*
++ * udba case, or
++ * if DI_LOCKED is not set, then p->parent may be different
++ * and h_parent can be NULL.
++ */
++ if (unlikely(!p->hdir || !h_dir || !p->h_parent)) {
++ err = -EBUSY;
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ dput(p->parent);
++ p->parent = NULL;
++ goto out_err;
++ }
+
-+/* void au_xattr_init(struct super_block *sb); */
-+#else
-+AuStubInt0(au_cpup_xattr, struct dentry *h_dst, struct dentry *h_src,
-+ int ignore_flags, unsigned int verbose);
-+/* AuStubVoid(au_xattr_init, struct super_block *sb); */
-+#endif
++ if (au_ftest_pin(p->flags, MNT_WRITE)) {
++ p->h_mnt = au_br_mnt(p->br);
++ err = vfsub_mnt_want_write(p->h_mnt);
++ if (unlikely(err)) {
++ au_fclr_pin(p->flags, MNT_WRITE);
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ dput(p->parent);
++ p->parent = NULL;
++ goto out_err;
++ }
++ }
+
-+#ifdef CONFIG_FS_POSIX_ACL
-+struct posix_acl *aufs_get_acl(struct inode *inode, int type);
-+int aufs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
-+#endif
++ au_igrab(h_dir);
++ err = au_pin_hdir_lock(p);
++ if (!err)
++ goto out; /* success */
+
-+#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
-+enum {
-+ AU_XATTR_SET,
-+ AU_XATTR_REMOVE,
-+ AU_ACL_SET
-+};
++ au_unpin(p);
+
-+struct au_srxattr {
-+ int type;
-+ union {
-+ struct {
-+ const char *name;
-+ const void *value;
-+ size_t size;
-+ int flags;
-+ } set;
-+ struct {
-+ const char *name;
-+ } remove;
-+ struct {
-+ struct posix_acl *acl;
-+ int type;
-+ } acl_set;
-+ } u;
-+};
-+ssize_t au_srxattr(struct dentry *dentry, struct inode *inode,
-+ struct au_srxattr *arg);
-+#endif
++out_err:
++ pr_err("err %d\n", err);
++ err = au_busy_or_stale();
++out:
++ return err;
++}
+
-+/* ---------------------------------------------------------------------- */
++void au_pin_init(struct au_pin *p, struct dentry *dentry,
++ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
++ unsigned int udba, unsigned char flags)
++{
++ p->dentry = dentry;
++ p->udba = udba;
++ p->lsc_di = lsc_di;
++ p->lsc_hi = lsc_hi;
++ p->flags = flags;
++ p->bindex = bindex;
+
-+/* lock subclass for iinfo */
-+enum {
-+ AuLsc_II_CHILD, /* child first */
-+ AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hnotify */
-+ AuLsc_II_CHILD3, /* copyup dirs */
-+ AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */
-+ AuLsc_II_PARENT2,
-+ AuLsc_II_PARENT3, /* copyup dirs */
-+ AuLsc_II_NEW_CHILD
-+};
++ p->parent = NULL;
++ p->hdir = NULL;
++ p->h_mnt = NULL;
+
-+/*
-+ * ii_read_lock_child, ii_write_lock_child,
-+ * ii_read_lock_child2, ii_write_lock_child2,
-+ * ii_read_lock_child3, ii_write_lock_child3,
-+ * ii_read_lock_parent, ii_write_lock_parent,
-+ * ii_read_lock_parent2, ii_write_lock_parent2,
-+ * ii_read_lock_parent3, ii_write_lock_parent3,
-+ * ii_read_lock_new_child, ii_write_lock_new_child,
-+ */
-+#define AuReadLockFunc(name, lsc) \
-+static inline void ii_read_lock_##name(struct inode *i) \
-+{ \
-+ au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
++ p->h_dentry = NULL;
++ p->h_parent = NULL;
++ p->br = NULL;
++ p->task = current;
+}
+
-+#define AuWriteLockFunc(name, lsc) \
-+static inline void ii_write_lock_##name(struct inode *i) \
-+{ \
-+ au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int udba, unsigned char flags)
++{
++ au_pin_init(pin, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2,
++ udba, flags);
++ return au_do_pin(pin);
+}
+
-+#define AuRWLockFuncs(name, lsc) \
-+ AuReadLockFunc(name, lsc) \
-+ AuWriteLockFunc(name, lsc)
-+
-+AuRWLockFuncs(child, CHILD);
-+AuRWLockFuncs(child2, CHILD2);
-+AuRWLockFuncs(child3, CHILD3);
-+AuRWLockFuncs(parent, PARENT);
-+AuRWLockFuncs(parent2, PARENT2);
-+AuRWLockFuncs(parent3, PARENT3);
-+AuRWLockFuncs(new_child, NEW_CHILD);
-+
-+#undef AuReadLockFunc
-+#undef AuWriteLockFunc
-+#undef AuRWLockFuncs
++/* ---------------------------------------------------------------------- */
+
+/*
-+ * ii_read_unlock, ii_write_unlock, ii_downgrade_lock
++ * ->setattr() and ->getattr() are called in various cases.
++ * chmod, stat: dentry is revalidated.
++ * fchmod, fstat: file and dentry are not revalidated, additionally they may be
++ * unhashed.
++ * for ->setattr(), ia->ia_file is passed from ftruncate only.
+ */
-+AuSimpleUnlockRwsemFuncs(ii, struct inode *i, &au_ii(i)->ii_rwsem);
-+
-+#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem)
-+#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem)
-+#define IiMustWriteLock(i) AuRwMustWriteLock(&au_ii(i)->ii_rwsem)
++/* todo: consolidate with do_refresh() and simple_reval_dpath() */
++int au_reval_for_attr(struct dentry *dentry, unsigned int sigen)
++{
++ int err;
++ struct dentry *parent;
+
-+/* ---------------------------------------------------------------------- */
++ err = 0;
++ if (au_digen_test(dentry, sigen)) {
++ parent = dget_parent(dentry);
++ di_read_lock_parent(parent, AuLock_IR);
++ err = au_refresh_dentry(dentry, parent);
++ di_read_unlock(parent, AuLock_IR);
++ dput(parent);
++ }
+
-+static inline void au_icntnr_init(struct au_icntnr *c)
-+{
-+#ifdef CONFIG_AUFS_DEBUG
-+ c->vfs_inode.i_mode = 0;
-+#endif
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline unsigned int au_iigen(struct inode *inode, unsigned int *igflags)
++int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
++ struct au_icpup_args *a)
+{
-+ unsigned int gen;
-+ struct au_iinfo *iinfo;
-+ struct au_iigen *iigen;
++ int err;
++ loff_t sz;
++ aufs_bindex_t btop, ibtop;
++ struct dentry *hi_wh, *parent;
++ struct inode *inode;
++ struct au_wr_dir_args wr_dir_args = {
++ .force_btgt = -1,
++ .flags = 0
++ };
+
-+ iinfo = au_ii(inode);
-+ iigen = &iinfo->ii_generation;
-+ spin_lock(&iigen->ig_spin);
-+ if (igflags)
-+ *igflags = iigen->ig_flags;
-+ gen = iigen->ig_generation;
-+ spin_unlock(&iigen->ig_spin);
++ if (d_is_dir(dentry))
++ au_fset_wrdir(wr_dir_args.flags, ISDIR);
++ /* plink or hi_wh() case */
++ btop = au_dbtop(dentry);
++ inode = d_inode(dentry);
++ ibtop = au_ibtop(inode);
++ if (btop != ibtop && !au_test_ro(inode->i_sb, ibtop, inode))
++ wr_dir_args.force_btgt = ibtop;
++ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args);
++ if (unlikely(err < 0))
++ goto out;
++ a->btgt = err;
++ if (err != btop)
++ au_fset_icpup(a->flags, DID_CPUP);
+
-+ return gen;
-+}
++ err = 0;
++ a->pin_flags = AuPin_MNT_WRITE;
++ parent = NULL;
++ if (!IS_ROOT(dentry)) {
++ au_fset_pin(a->pin_flags, DI_LOCKED);
++ parent = dget_parent(dentry);
++ di_write_lock_parent(parent);
++ }
+
-+/* tiny test for inode number */
-+/* tmpfs generation is too rough */
-+static inline int au_test_higen(struct inode *inode, struct inode *h_inode)
-+{
-+ struct au_iinfo *iinfo;
++ err = au_pin(&a->pin, dentry, a->btgt, a->udba, a->pin_flags);
++ if (unlikely(err))
++ goto out_parent;
+
-+ iinfo = au_ii(inode);
-+ AuRwMustAnyLock(&iinfo->ii_rwsem);
-+ return !(iinfo->ii_hsb1 == h_inode->i_sb
-+ && iinfo->ii_higen == h_inode->i_generation);
-+}
++ sz = -1;
++ a->h_path.dentry = au_h_dptr(dentry, btop);
++ a->h_inode = d_inode(a->h_path.dentry);
++ if (ia && (ia->ia_valid & ATTR_SIZE)) {
++ vfsub_inode_lock_shared_nested(a->h_inode, AuLsc_I_CHILD);
++ if (ia->ia_size < i_size_read(a->h_inode))
++ sz = ia->ia_size;
++ inode_unlock_shared(a->h_inode);
++ }
+
-+static inline void au_iigen_dec(struct inode *inode)
-+{
-+ struct au_iinfo *iinfo;
-+ struct au_iigen *iigen;
++ hi_wh = NULL;
++ if (au_ftest_icpup(a->flags, DID_CPUP) && d_unlinked(dentry)) {
++ hi_wh = au_hi_wh(inode, a->btgt);
++ if (!hi_wh) {
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = a->btgt,
++ .bsrc = -1,
++ .len = sz,
++ .pin = &a->pin
++ };
++ err = au_sio_cpup_wh(&cpg, /*file*/NULL);
++ if (unlikely(err))
++ goto out_unlock;
++ hi_wh = au_hi_wh(inode, a->btgt);
++ /* todo: revalidate hi_wh? */
++ }
++ }
+
-+ iinfo = au_ii(inode);
-+ iigen = &iinfo->ii_generation;
-+ spin_lock(&iigen->ig_spin);
-+ iigen->ig_generation--;
-+ spin_unlock(&iigen->ig_spin);
++ if (parent) {
++ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
++ di_downgrade_lock(parent, AuLock_IR);
++ dput(parent);
++ parent = NULL;
++ }
++ if (!au_ftest_icpup(a->flags, DID_CPUP))
++ goto out; /* success */
++
++ if (!d_unhashed(dentry)) {
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = a->btgt,
++ .bsrc = btop,
++ .len = sz,
++ .pin = &a->pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ err = au_sio_cpup_simple(&cpg);
++ if (!err)
++ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
++ } else if (!hi_wh)
++ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
++ else
++ a->h_path.dentry = hi_wh; /* do not dget here */
++
++out_unlock:
++ a->h_inode = d_inode(a->h_path.dentry);
++ if (!err)
++ goto out; /* success */
++ au_unpin(&a->pin);
++out_parent:
++ if (parent) {
++ di_write_unlock(parent);
++ dput(parent);
++ }
++out:
++ if (!err)
++ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
++ return err;
+}
+
-+static inline int au_iigen_test(struct inode *inode, unsigned int sigen)
++static int aufs_setattr(struct dentry *dentry, struct iattr *ia)
+{
+ int err;
++ struct inode *inode, *delegated;
++ struct super_block *sb;
++ struct file *file;
++ struct au_icpup_args *a;
+
-+ err = 0;
-+ if (unlikely(inode && au_iigen(inode, NULL) != sigen))
-+ err = -EIO;
++ inode = d_inode(dentry);
++ IMustLock(inode);
+
-+ return err;
-+}
++ err = setattr_prepare(dentry, ia);
++ if (unlikely(err))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
+
-+static inline struct au_hinode *au_hinode(struct au_iinfo *iinfo,
-+ aufs_bindex_t bindex)
-+{
-+ return iinfo->ii_hinode + bindex;
-+}
++ if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
++ ia->ia_valid &= ~ATTR_MODE;
+
-+static inline int au_is_bad_inode(struct inode *inode)
-+{
-+ return !!(is_bad_inode(inode) || !au_hinode(au_ii(inode), 0));
-+}
++ file = NULL;
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out_kfree;
+
-+static inline aufs_bindex_t au_ii_br_id(struct inode *inode,
-+ aufs_bindex_t bindex)
-+{
-+ IiMustAnyLock(inode);
-+ return au_hinode(au_ii(inode), bindex)->hi_id;
-+}
++ if (ia->ia_valid & ATTR_FILE) {
++ /* currently ftruncate(2) only */
++ AuDebugOn(!d_is_reg(dentry));
++ file = ia->ia_file;
++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1,
++ /*fi_lsc*/0);
++ if (unlikely(err))
++ goto out_si;
++ ia->ia_file = au_hf_top(file);
++ a->udba = AuOpt_UDBA_NONE;
++ } else {
++ /* fchmod() doesn't pass ia_file */
++ a->udba = au_opt_udba(sb);
++ di_write_lock_child(dentry);
++ /* no d_unlinked(), to set UDBA_NONE for root */
++ if (d_unhashed(dentry))
++ a->udba = AuOpt_UDBA_NONE;
++ if (a->udba != AuOpt_UDBA_NONE) {
++ AuDebugOn(IS_ROOT(dentry));
++ err = au_reval_for_attr(dentry, au_sigen(sb));
++ if (unlikely(err))
++ goto out_dentry;
++ }
++ }
+
-+static inline aufs_bindex_t au_ibtop(struct inode *inode)
-+{
-+ IiMustAnyLock(inode);
-+ return au_ii(inode)->ii_btop;
-+}
++ err = au_pin_and_icpup(dentry, ia, a);
++ if (unlikely(err < 0))
++ goto out_dentry;
++ if (au_ftest_icpup(a->flags, DID_CPUP)) {
++ ia->ia_file = NULL;
++ ia->ia_valid &= ~ATTR_FILE;
++ }
+
-+static inline aufs_bindex_t au_ibbot(struct inode *inode)
-+{
-+ IiMustAnyLock(inode);
-+ return au_ii(inode)->ii_bbot;
-+}
++ a->h_path.mnt = au_sbr_mnt(sb, a->btgt);
++ if ((ia->ia_valid & (ATTR_MODE | ATTR_CTIME))
++ == (ATTR_MODE | ATTR_CTIME)) {
++ err = security_path_chmod(&a->h_path, ia->ia_mode);
++ if (unlikely(err))
++ goto out_unlock;
++ } else if ((ia->ia_valid & (ATTR_UID | ATTR_GID))
++ && (ia->ia_valid & ATTR_CTIME)) {
++ err = security_path_chown(&a->h_path, ia->ia_uid, ia->ia_gid);
++ if (unlikely(err))
++ goto out_unlock;
++ }
+
-+static inline struct au_vdir *au_ivdir(struct inode *inode)
-+{
-+ IiMustAnyLock(inode);
-+ return au_ii(inode)->ii_vdir;
-+}
++ if (ia->ia_valid & ATTR_SIZE) {
++ struct file *f;
+
-+static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ IiMustAnyLock(inode);
-+ return au_hinode(au_ii(inode), bindex)->hi_whdentry;
-+}
++ if (ia->ia_size < i_size_read(inode))
++ /* unmap only */
++ truncate_setsize(inode, ia->ia_size);
+
-+static inline void au_set_ibtop(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ IiMustWriteLock(inode);
-+ au_ii(inode)->ii_btop = bindex;
-+}
++ f = NULL;
++ if (ia->ia_valid & ATTR_FILE)
++ f = ia->ia_file;
++ inode_unlock(a->h_inode);
++ err = vfsub_trunc(&a->h_path, ia->ia_size, ia->ia_valid, f);
++ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
++ } else {
++ delegated = NULL;
++ while (1) {
++ err = vfsub_notify_change(&a->h_path, ia, &delegated);
++ if (delegated) {
++ err = break_deleg_wait(&delegated);
++ if (!err)
++ continue;
++ }
++ break;
++ }
++ }
++ /*
++ * regardless aufs 'acl' option setting.
++ * why don't all acl-aware fs call this func from their ->setattr()?
++ */
++ if (!err && (ia->ia_valid & ATTR_MODE))
++ err = vfsub_acl_chmod(a->h_inode, ia->ia_mode);
++ if (!err)
++ au_cpup_attr_changeable(inode);
+
-+static inline void au_set_ibbot(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ IiMustWriteLock(inode);
-+ au_ii(inode)->ii_bbot = bindex;
++out_unlock:
++ inode_unlock(a->h_inode);
++ au_unpin(&a->pin);
++ if (unlikely(err))
++ au_update_dbtop(dentry);
++out_dentry:
++ di_write_unlock(dentry);
++ if (file) {
++ fi_write_unlock(file);
++ ia->ia_file = file;
++ ia->ia_valid |= ATTR_FILE;
++ }
++out_si:
++ si_read_unlock(sb);
++out_kfree:
++ kfree(a);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir)
++#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
++static int au_h_path_to_set_attr(struct dentry *dentry,
++ struct au_icpup_args *a, struct path *h_path)
+{
-+ IiMustWriteLock(inode);
-+ au_ii(inode)->ii_vdir = vdir;
-+}
++ int err;
++ struct super_block *sb;
+
-+static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ IiMustAnyLock(inode);
-+ return au_hinode(au_ii(inode), bindex);
-+}
++ sb = dentry->d_sb;
++ a->udba = au_opt_udba(sb);
++ /* no d_unlinked(), to set UDBA_NONE for root */
++ if (d_unhashed(dentry))
++ a->udba = AuOpt_UDBA_NONE;
++ if (a->udba != AuOpt_UDBA_NONE) {
++ AuDebugOn(IS_ROOT(dentry));
++ err = au_reval_for_attr(dentry, au_sigen(sb));
++ if (unlikely(err))
++ goto out;
++ }
++ err = au_pin_and_icpup(dentry, /*ia*/NULL, a);
++ if (unlikely(err < 0))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ h_path->dentry = a->h_path.dentry;
++ h_path->mnt = au_sbr_mnt(sb, a->btgt);
+
-+static inline struct dentry *au_pinned_parent(struct au_pin *pin)
-+{
-+ if (pin)
-+ return pin->parent;
-+ return NULL;
++out:
++ return err;
+}
+
-+static inline struct inode *au_pinned_h_dir(struct au_pin *pin)
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg)
+{
-+ if (pin && pin->hdir)
-+ return pin->hdir->hi_inode;
-+ return NULL;
-+}
++ int err;
++ struct path h_path;
++ struct super_block *sb;
++ struct au_icpup_args *a;
++ struct inode *h_inode;
+
-+static inline struct au_hinode *au_pinned_hdir(struct au_pin *pin)
-+{
-+ if (pin)
-+ return pin->hdir;
-+ return NULL;
-+}
++ IMustLock(inode);
+
-+static inline void au_pin_set_dentry(struct au_pin *pin, struct dentry *dentry)
-+{
-+ if (pin)
-+ pin->dentry = dentry;
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out_kfree;
++
++ h_path.dentry = NULL; /* silence gcc */
++ di_write_lock_child(dentry);
++ err = au_h_path_to_set_attr(dentry, a, &h_path);
++ if (unlikely(err))
++ goto out_di;
++
++ inode_unlock(a->h_inode);
++ switch (arg->type) {
++ case AU_XATTR_SET:
++ AuDebugOn(d_is_negative(h_path.dentry));
++ err = vfsub_setxattr(h_path.dentry,
++ arg->u.set.name, arg->u.set.value,
++ arg->u.set.size, arg->u.set.flags);
++ break;
++ case AU_ACL_SET:
++ err = -EOPNOTSUPP;
++ h_inode = d_inode(h_path.dentry);
++ if (h_inode->i_op->set_acl)
++ /* this will call posix_acl_update_mode */
++ err = h_inode->i_op->set_acl(h_inode,
++ arg->u.acl_set.acl,
++ arg->u.acl_set.type);
++ break;
++ }
++ if (!err)
++ au_cpup_attr_timesizes(inode);
++
++ au_unpin(&a->pin);
++ if (unlikely(err))
++ au_update_dbtop(dentry);
++
++out_di:
++ di_write_unlock(dentry);
++ si_read_unlock(sb);
++out_kfree:
++ kfree(a);
++out:
++ AuTraceErr(err);
++ return err;
+}
++#endif
+
-+static inline void au_pin_set_parent_lflag(struct au_pin *pin,
-+ unsigned char lflag)
++static void au_refresh_iattr(struct inode *inode, struct kstat *st,
++ unsigned int nlink)
+{
-+ if (pin) {
-+ if (lflag)
-+ au_fset_pin(pin->flags, DI_LOCKED);
-+ else
-+ au_fclr_pin(pin->flags, DI_LOCKED);
++ unsigned int n;
++
++ inode->i_mode = st->mode;
++ /* don't i_[ug]id_write() here */
++ inode->i_uid = st->uid;
++ inode->i_gid = st->gid;
++ inode->i_atime = st->atime;
++ inode->i_mtime = st->mtime;
++ inode->i_ctime = st->ctime;
++
++ au_cpup_attr_nlink(inode, /*force*/0);
++ if (S_ISDIR(inode->i_mode)) {
++ n = inode->i_nlink;
++ n -= nlink;
++ n += st->nlink;
++ smp_mb(); /* for i_nlink */
++ /* 0 can happen */
++ set_nlink(inode, n);
+ }
-+}
+
-+#if 0 /* reserved */
-+static inline void au_pin_set_parent(struct au_pin *pin, struct dentry *parent)
-+{
-+ if (pin) {
-+ dput(pin->parent);
-+ pin->parent = dget(parent);
-+ }
++ spin_lock(&inode->i_lock);
++ inode->i_blocks = st->blocks;
++ i_size_write(inode, st->size);
++ spin_unlock(&inode->i_lock);
+}
-+#endif
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_branch;
-+#ifdef CONFIG_AUFS_HNOTIFY
-+struct au_hnotify_op {
-+ void (*ctl)(struct au_hinode *hinode, int do_set);
-+ int (*alloc)(struct au_hinode *hinode);
-+
-+ /*
-+ * if it returns true, the the caller should free hinode->hi_notify,
-+ * otherwise ->free() frees it.
-+ */
-+ int (*free)(struct au_hinode *hinode,
-+ struct au_hnotify *hn) __must_check;
+
-+ void (*fin)(void);
-+ int (*init)(void);
++/*
++ * common routine for aufs_getattr() and au_getxattr().
++ * returns zero or negative (an error).
++ * @dentry will be read-locked in success.
++ */
++int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path,
++ int locked)
++{
++ int err;
++ unsigned int mnt_flags, sigen;
++ unsigned char udba_none;
++ aufs_bindex_t bindex;
++ struct super_block *sb, *h_sb;
++ struct inode *inode;
+
-+ int (*reset_br)(unsigned int udba, struct au_branch *br, int perm);
-+ void (*fin_br)(struct au_branch *br);
-+ int (*init_br)(struct au_branch *br, int perm);
-+};
++ h_path->mnt = NULL;
++ h_path->dentry = NULL;
+
-+/* hnotify.c */
-+int au_hn_alloc(struct au_hinode *hinode, struct inode *inode);
-+void au_hn_free(struct au_hinode *hinode);
-+void au_hn_ctl(struct au_hinode *hinode, int do_set);
-+void au_hn_reset(struct inode *inode, unsigned int flags);
-+int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
-+ struct qstr *h_child_qstr, struct inode *h_child_inode);
-+int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm);
-+int au_hnotify_init_br(struct au_branch *br, int perm);
-+void au_hnotify_fin_br(struct au_branch *br);
-+int __init au_hnotify_init(void);
-+void au_hnotify_fin(void);
++ err = 0;
++ sb = dentry->d_sb;
++ mnt_flags = au_mntflags(sb);
++ udba_none = !!au_opt_test(mnt_flags, UDBA_NONE);
+
-+/* hfsnotify.c */
-+extern const struct au_hnotify_op au_hnotify_op;
++ if (unlikely(locked))
++ goto body; /* skip locking dinfo */
+
-+static inline
-+void au_hn_init(struct au_hinode *hinode)
-+{
-+ hinode->hi_notify = NULL;
-+}
++ /* support fstat(2) */
++ if (!d_unlinked(dentry) && !udba_none) {
++ sigen = au_sigen(sb);
++ err = au_digen_test(dentry, sigen);
++ if (!err) {
++ di_read_lock_child(dentry, AuLock_IR);
++ err = au_dbrange_test(dentry);
++ if (unlikely(err)) {
++ di_read_unlock(dentry, AuLock_IR);
++ goto out;
++ }
++ } else {
++ AuDebugOn(IS_ROOT(dentry));
++ di_write_lock_child(dentry);
++ err = au_dbrange_test(dentry);
++ if (!err)
++ err = au_reval_for_attr(dentry, sigen);
++ if (!err)
++ di_downgrade_lock(dentry, AuLock_IR);
++ else {
++ di_write_unlock(dentry);
++ goto out;
++ }
++ }
++ } else
++ di_read_lock_child(dentry, AuLock_IR);
+
-+static inline struct au_hnotify *au_hn(struct au_hinode *hinode)
-+{
-+ return hinode->hi_notify;
-+}
++body:
++ inode = d_inode(dentry);
++ bindex = au_ibtop(inode);
++ h_path->mnt = au_sbr_mnt(sb, bindex);
++ h_sb = h_path->mnt->mnt_sb;
++ if (!force
++ && !au_test_fs_bad_iattr(h_sb)
++ && udba_none)
++ goto out; /* success */
+
-+#else
-+AuStub(int, au_hn_alloc, return -EOPNOTSUPP,
-+ struct au_hinode *hinode __maybe_unused,
-+ struct inode *inode __maybe_unused)
-+AuStub(struct au_hnotify *, au_hn, return NULL, struct au_hinode *hinode)
-+AuStubVoid(au_hn_free, struct au_hinode *hinode __maybe_unused)
-+AuStubVoid(au_hn_ctl, struct au_hinode *hinode __maybe_unused,
-+ int do_set __maybe_unused)
-+AuStubVoid(au_hn_reset, struct inode *inode __maybe_unused,
-+ unsigned int flags __maybe_unused)
-+AuStubInt0(au_hnotify_reset_br, unsigned int udba __maybe_unused,
-+ struct au_branch *br __maybe_unused,
-+ int perm __maybe_unused)
-+AuStubInt0(au_hnotify_init_br, struct au_branch *br __maybe_unused,
-+ int perm __maybe_unused)
-+AuStubVoid(au_hnotify_fin_br, struct au_branch *br __maybe_unused)
-+AuStubInt0(__init au_hnotify_init, void)
-+AuStubVoid(au_hnotify_fin, void)
-+AuStubVoid(au_hn_init, struct au_hinode *hinode __maybe_unused)
-+#endif /* CONFIG_AUFS_HNOTIFY */
++ if (au_dbtop(dentry) == bindex)
++ h_path->dentry = au_h_dptr(dentry, bindex);
++ else if (au_opt_test(mnt_flags, PLINK) && au_plink_test(inode)) {
++ h_path->dentry = au_plink_lkup(inode, bindex);
++ if (IS_ERR(h_path->dentry))
++ /* pretending success */
++ h_path->dentry = NULL;
++ else
++ dput(h_path->dentry);
++ }
+
-+static inline void au_hn_suspend(struct au_hinode *hdir)
-+{
-+ au_hn_ctl(hdir, /*do_set*/0);
++out:
++ return err;
+}
+
-+static inline void au_hn_resume(struct au_hinode *hdir)
++static int aufs_getattr(const struct path *path, struct kstat *st,
++ u32 request, unsigned int query)
+{
-+ au_hn_ctl(hdir, /*do_set*/1);
-+}
++ int err;
++ unsigned char positive;
++ struct path h_path;
++ struct dentry *dentry;
++ struct inode *inode;
++ struct super_block *sb;
+
-+static inline void au_hn_inode_lock(struct au_hinode *hdir)
-+{
-+ inode_lock(hdir->hi_inode);
-+ au_hn_suspend(hdir);
-+}
++ dentry = path->dentry;
++ inode = d_inode(dentry);
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out;
++ err = au_h_path_getattr(dentry, /*force*/0, &h_path, /*locked*/0);
++ if (unlikely(err))
++ goto out_si;
++ if (unlikely(!h_path.dentry))
++ /* illegally overlapped or something */
++ goto out_fill; /* pretending success */
+
-+static inline void au_hn_inode_lock_nested(struct au_hinode *hdir,
-+ unsigned int sc __maybe_unused)
-+{
-+ inode_lock_nested(hdir->hi_inode, sc);
-+ au_hn_suspend(hdir);
-+}
++ positive = d_is_positive(h_path.dentry);
++ if (positive)
++ /* no vfsub version */
++ err = vfs_getattr(&h_path, st, request, query);
++ if (!err) {
++ if (positive)
++ au_refresh_iattr(inode, st,
++ d_inode(h_path.dentry)->i_nlink);
++ goto out_fill; /* success */
++ }
++ AuTraceErr(err);
++ goto out_di;
+
-+static inline void au_hn_inode_unlock(struct au_hinode *hdir)
-+{
-+ au_hn_resume(hdir);
-+ inode_unlock(hdir->hi_inode);
++out_fill:
++ generic_fillattr(inode, st);
++out_di:
++ di_read_unlock(dentry, AuLock_IR);
++out_si:
++ si_read_unlock(sb);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_INODE_H__ */
-diff -urN /usr/share/empty/fs/aufs/ioctl.c linux/fs/aufs/ioctl.c
---- /usr/share/empty/fs/aufs/ioctl.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/ioctl.c 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,219 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
-+
-+/*
-+ * ioctl
-+ * plink-management and readdir in userspace.
-+ * assist the pathconf(3) wrapper library.
-+ * move-down
-+ * File-based Hierarchical Storage Management.
-+ */
-+
-+#include <linux/compat.h>
-+#include <linux/file.h>
-+#include "aufs.h"
++/* ---------------------------------------------------------------------- */
+
-+static int au_wbr_fd(struct path *path, struct aufs_wbr_fd __user *arg)
++static const char *aufs_get_link(struct dentry *dentry, struct inode *inode,
++ struct delayed_call *done)
+{
-+ int err, fd;
-+ aufs_bindex_t wbi, bindex, bbot;
-+ struct file *h_file;
-+ struct super_block *sb;
-+ struct dentry *root;
-+ struct au_branch *br;
-+ struct aufs_wbr_fd wbrfd = {
-+ .oflags = au_dir_roflags,
-+ .brid = -1
-+ };
-+ const int valid = O_RDONLY | O_NONBLOCK | O_LARGEFILE | O_DIRECTORY
-+ | O_NOATIME | O_CLOEXEC;
++ const char *ret;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ int err;
++ aufs_bindex_t bindex;
+
-+ AuDebugOn(wbrfd.oflags & ~valid);
++ ret = NULL; /* suppress a warning */
++ err = -ECHILD;
++ if (!dentry)
++ goto out;
+
-+ if (arg) {
-+ err = copy_from_user(&wbrfd, arg, sizeof(wbrfd));
-+ if (unlikely(err)) {
-+ err = -EFAULT;
-+ goto out;
-+ }
++ err = aufs_read_lock(dentry, AuLock_IR | AuLock_GEN);
++ if (unlikely(err))
++ goto out;
+
-+ err = -EINVAL;
-+ AuDbg("wbrfd{0%o, %d}\n", wbrfd.oflags, wbrfd.brid);
-+ wbrfd.oflags |= au_dir_roflags;
-+ AuDbg("0%o\n", wbrfd.oflags);
-+ if (unlikely(wbrfd.oflags & ~valid))
-+ goto out;
-+ }
++ err = au_d_hashed_positive(dentry);
++ if (unlikely(err))
++ goto out_unlock;
+
-+ fd = get_unused_fd_flags(0);
-+ err = fd;
-+ if (unlikely(fd < 0))
-+ goto out;
++ err = -EINVAL;
++ inode = d_inode(dentry);
++ bindex = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, bindex);
++ if (unlikely(!h_inode->i_op->get_link))
++ goto out_unlock;
+
-+ h_file = ERR_PTR(-EINVAL);
-+ wbi = 0;
-+ br = NULL;
-+ sb = path->dentry->d_sb;
-+ root = sb->s_root;
-+ aufs_read_lock(root, AuLock_IR);
-+ bbot = au_sbbot(sb);
-+ if (wbrfd.brid >= 0) {
-+ wbi = au_br_index(sb, wbrfd.brid);
-+ if (unlikely(wbi < 0 || wbi > bbot))
-+ goto out_unlock;
++ err = -EBUSY;
++ h_dentry = NULL;
++ if (au_dbtop(dentry) <= bindex) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry)
++ dget(h_dentry);
+ }
-+
-+ h_file = ERR_PTR(-ENOENT);
-+ br = au_sbr(sb, wbi);
-+ if (!au_br_writable(br->br_perm)) {
-+ if (arg)
++ if (!h_dentry) {
++ h_dentry = d_find_any_alias(h_inode);
++ if (IS_ERR(h_dentry)) {
++ err = PTR_ERR(h_dentry);
+ goto out_unlock;
-+
-+ bindex = wbi + 1;
-+ wbi = -1;
-+ for (; bindex <= bbot; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ if (au_br_writable(br->br_perm)) {
-+ wbi = bindex;
-+ br = au_sbr(sb, wbi);
-+ break;
-+ }
+ }
+ }
-+ AuDbg("wbi %d\n", wbi);
-+ if (wbi >= 0)
-+ h_file = au_h_open(root, wbi, wbrfd.oflags, NULL,
-+ /*force_wr*/0);
-+
-+out_unlock:
-+ aufs_read_unlock(root, AuLock_IR);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out_fd;
++ if (unlikely(!h_dentry))
++ goto out_unlock;
+
-+ au_br_put(br); /* cf. au_h_open() */
-+ fd_install(fd, h_file);
-+ err = fd;
-+ goto out; /* success */
++ err = 0;
++ AuDbg("%pf\n", h_inode->i_op->get_link);
++ AuDbgDentry(h_dentry);
++ ret = vfs_get_link(h_dentry, done);
++ dput(h_dentry);
++ if (IS_ERR(ret))
++ err = PTR_ERR(ret);
+
-+out_fd:
-+ put_unused_fd(fd);
++out_unlock:
++ aufs_read_unlock(dentry, AuLock_IR);
+out:
-+ AuTraceErr(err);
-+ return err;
++ if (unlikely(err))
++ ret = ERR_PTR(err);
++ AuTraceErrPtr(ret);
++ return ret;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg)
++static int au_is_special(struct inode *inode)
+{
-+ long err;
-+ struct dentry *dentry;
++ return (inode->i_mode & (S_IFBLK | S_IFCHR | S_IFIFO | S_IFSOCK));
++}
+
-+ switch (cmd) {
-+ case AUFS_CTL_RDU:
-+ case AUFS_CTL_RDU_INO:
-+ err = au_rdu_ioctl(file, cmd, arg);
-+ break;
++static int aufs_update_time(struct inode *inode, struct timespec *ts, int flags)
++{
++ int err;
++ aufs_bindex_t bindex;
++ struct super_block *sb;
++ struct inode *h_inode;
++ struct vfsmount *h_mnt;
+
-+ case AUFS_CTL_WBR_FD:
-+ err = au_wbr_fd(&file->f_path, (void __user *)arg);
-+ break;
++ sb = inode->i_sb;
++ WARN_ONCE((flags & S_ATIME) && !IS_NOATIME(inode),
++ "unexpected s_flags 0x%lx", sb->s_flags);
+
-+ case AUFS_CTL_IBUSY:
-+ err = au_ibusy_ioctl(file, arg);
-+ break;
++ /* mmap_sem might be acquired already, cf. aufs_mmap() */
++ lockdep_off();
++ si_read_lock(sb, AuLock_FLUSH);
++ ii_write_lock_child(inode);
+
-+ case AUFS_CTL_BRINFO:
-+ err = au_brinfo_ioctl(file, arg);
-+ break;
++ err = 0;
++ bindex = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, bindex);
++ if (!au_test_ro(sb, bindex, inode)) {
++ h_mnt = au_sbr_mnt(sb, bindex);
++ err = vfsub_mnt_want_write(h_mnt);
++ if (!err) {
++ err = vfsub_update_time(h_inode, ts, flags);
++ vfsub_mnt_drop_write(h_mnt);
++ }
++ } else if (au_is_special(h_inode)) {
++ /*
++ * Never copy-up here.
++ * These special files may already be opened and used for
++ * communicating. If we copied it up, then the communication
++ * would be corrupted.
++ */
++ AuWarn1("timestamps for i%lu are ignored "
++ "since it is on readonly branch (hi%lu).\n",
++ inode->i_ino, h_inode->i_ino);
++ } else if (flags & ~S_ATIME) {
++ err = -EIO;
++ AuIOErr1("unexpected flags 0x%x\n", flags);
++ AuDebugOn(1);
++ }
+
-+ case AUFS_CTL_FHSM_FD:
-+ dentry = file->f_path.dentry;
-+ if (IS_ROOT(dentry))
-+ err = au_fhsm_fd(dentry->d_sb, arg);
-+ else
-+ err = -ENOTTY;
-+ break;
++ if (!err)
++ au_cpup_attr_timesizes(inode);
++ ii_write_unlock(inode);
++ si_read_unlock(sb);
++ lockdep_on();
+
-+ default:
-+ /* do not call the lower */
-+ AuDbg("0x%x\n", cmd);
-+ err = -ENOTTY;
-+ }
++ if (!err && (flags & S_VERSION))
++ inode_inc_iversion(inode);
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg)
-+{
-+ long err;
++/* ---------------------------------------------------------------------- */
+
-+ switch (cmd) {
-+ case AUFS_CTL_MVDOWN:
-+ err = au_mvdown(file->f_path.dentry, (void __user *)arg);
-+ break;
++/* no getattr version will be set by module.c:aufs_init() */
++struct inode_operations aufs_iop_nogetattr[AuIop_Last],
++ aufs_iop[] = {
++ [AuIop_SYMLINK] = {
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl, /* unsupport for symlink? */
++#endif
+
-+ case AUFS_CTL_WBR_FD:
-+ err = au_wbr_fd(&file->f_path, (void __user *)arg);
-+ break;
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
+
-+ default:
-+ /* do not call the lower */
-+ AuDbg("0x%x\n", cmd);
-+ err = -ENOTTY;
-+ }
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
++#endif
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ .get_link = aufs_get_link,
+
-+#ifdef CONFIG_COMPAT
-+long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
-+ unsigned long arg)
-+{
-+ long err;
++ /* .update_time = aufs_update_time */
++ },
++ [AuIop_DIR] = {
++ .create = aufs_create,
++ .lookup = aufs_lookup,
++ .link = aufs_link,
++ .unlink = aufs_unlink,
++ .symlink = aufs_symlink,
++ .mkdir = aufs_mkdir,
++ .rmdir = aufs_rmdir,
++ .mknod = aufs_mknod,
++ .rename = aufs_rename,
+
-+ switch (cmd) {
-+ case AUFS_CTL_RDU:
-+ case AUFS_CTL_RDU_INO:
-+ err = au_rdu_compat_ioctl(file, cmd, arg);
-+ break;
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl,
++#endif
+
-+ case AUFS_CTL_IBUSY:
-+ err = au_ibusy_compat_ioctl(file, arg);
-+ break;
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
+
-+ case AUFS_CTL_BRINFO:
-+ err = au_brinfo_compat_ioctl(file, arg);
-+ break;
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
++#endif
+
-+ default:
-+ err = aufs_ioctl_dir(file, cmd, arg);
-+ }
++ .update_time = aufs_update_time,
++ .atomic_open = aufs_atomic_open,
++ .tmpfile = aufs_tmpfile
++ },
++ [AuIop_OTHER] = {
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl,
++#endif
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
+
-+long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
-+ unsigned long arg)
-+{
-+ return aufs_ioctl_nondir(file, cmd, (unsigned long)compat_ptr(arg));
-+}
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
+#endif
-diff -urN /usr/share/empty/fs/aufs/i_op_add.c linux/fs/aufs/i_op_add.c
---- /usr/share/empty/fs/aufs/i_op_add.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/i_op_add.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,924 @@
++
++ .update_time = aufs_update_time
++ }
++};
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/i_op_add.c linux-4.14/fs/aufs/i_op_add.c
+--- linux-4.14/fs/aufs/i_op_add.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/i_op_add.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,920 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ if (!try_aopen)
+ aufs_read_unlock(dentry, AuLock_DW);
+out_free:
-+ au_delayed_kfree(a);
++ kfree(a);
+out:
+ return err;
+}
+ goto out_parent;
+
+ h_parent = au_h_dptr(parent, bindex);
-+ err = inode_permission(d_inode(h_parent), MAY_WRITE | MAY_EXEC);
-+ if (unlikely(err))
-+ goto out_mnt;
-+
-+ err = -ENOMEM;
-+ h_dentry = d_alloc(h_parent, &dentry->d_name);
-+ if (unlikely(!h_dentry))
++ h_dentry = vfs_tmpfile(h_parent, mode, /*open_flag*/0);
++ if (IS_ERR(h_dentry)) {
++ err = PTR_ERR(h_dentry);
+ goto out_mnt;
-+
-+ err = h_dir->i_op->tmpfile(h_dir, h_dentry, mode);
-+ if (unlikely(err))
-+ goto out_dentry;
++ }
+
+ au_set_dbtop(dentry, bindex);
+ au_set_dbbot(dentry, bindex);
+ if (au_ibtop(dir) == au_dbtop(dentry))
+ au_cpup_attr_timesizes(dir);
+ }
-+
-+out_dentry:
+ dput(h_dentry);
++
+out_mnt:
+ vfsub_mnt_drop_write(h_mnt);
+out_parent:
+ goto out_parent;
+ }
+
++ /*
++ * aufs doesn't touch the credential so
++ * security_dentry_create_files_as() is unnecrssary.
++ */
+ if (au_opt_test(au_mntflags(sb), PLINK)) {
+ if (a->bdst < a->bsrc
+ /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */)
+ }
+ aufs_read_and_write_unlock2(dentry, src_dentry);
+out_kfree:
-+ au_delayed_kfree(a);
++ kfree(a);
+out:
+ AuTraceErr(err);
+ return err;
+}
+
-+int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
++int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode)
++{
++ int err, rerr;
++ aufs_bindex_t bindex;
++ unsigned char diropq;
++ struct path h_path;
++ struct dentry *wh_dentry, *parent, *opq_dentry;
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct {
++ struct au_pin pin;
++ struct au_dtime dt;
++ } *a; /* reduce the stack usage */
++ struct au_wr_dir_args wr_dir_args = {
++ .force_btgt = -1,
++ .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR
++ };
++
++ IMustLock(dir);
++
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
++ if (unlikely(err))
++ goto out_free;
++ err = au_d_may_add(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL,
++ &a->pin, &wr_dir_args);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ sb = dentry->d_sb;
++ bindex = au_dbtop(dentry);
++ h_path.dentry = au_h_dptr(dentry, bindex);
++ h_path.mnt = au_sbr_mnt(sb, bindex);
++ err = vfsub_mkdir(au_pinned_h_dir(&a->pin), &h_path, mode);
++ if (unlikely(err))
++ goto out_unpin;
++
++ /* make the dir opaque */
++ diropq = 0;
++ h_inode = d_inode(h_path.dentry);
++ if (wh_dentry
++ || au_opt_test(au_mntflags(sb), ALWAYS_DIROPQ)) {
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ opq_dentry = au_diropq_create(dentry, bindex);
++ inode_unlock(h_inode);
++ err = PTR_ERR(opq_dentry);
++ if (IS_ERR(opq_dentry))
++ goto out_dir;
++ dput(opq_dentry);
++ diropq = 1;
++ }
++
++ err = epilog(dir, bindex, wh_dentry, dentry);
++ if (!err) {
++ inc_nlink(dir);
++ goto out_unpin; /* success */
++ }
++
++ /* revert */
++ if (diropq) {
++ AuLabel(revert opq);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ rerr = au_diropq_remove(dentry, bindex);
++ inode_unlock(h_inode);
++ if (rerr) {
++ AuIOErr("%pd reverting diropq failed(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ }
++ }
++
++out_dir:
++ AuLabel(revert dir);
++ rerr = vfsub_rmdir(au_pinned_h_dir(&a->pin), &h_path);
++ if (rerr) {
++ AuIOErr("%pd reverting dir failed(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ }
++ au_dtime_revert(&a->dt);
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++out_parent:
++ di_write_unlock(parent);
++out_unlock:
++ if (unlikely(err)) {
++ au_update_dbtop(dentry);
++ d_drop(dentry);
++ }
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ kfree(a);
++out:
++ return err;
++}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/i_op_del.c linux-4.14/fs/aufs/i_op_del.c
+--- linux-4.14/fs/aufs/i_op_del.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/i_op_del.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,511 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
++
++/*
++ * inode operations (del entry)
++ */
++
++#include "aufs.h"
++
++/*
++ * decide if a new whiteout for @dentry is necessary or not.
++ * when it is necessary, prepare the parent dir for the upper branch whose
++ * branch index is @bcpup for creation. the actual creation of the whiteout will
++ * be done by caller.
++ * return value:
++ * 0: wh is unnecessary
++ * plus: wh is necessary
++ * minus: error
++ */
++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup)
++{
++ int need_wh, err;
++ aufs_bindex_t btop;
++ struct super_block *sb;
++
++ sb = dentry->d_sb;
++ btop = au_dbtop(dentry);
++ if (*bcpup < 0) {
++ *bcpup = btop;
++ if (au_test_ro(sb, btop, d_inode(dentry))) {
++ err = AuWbrCopyup(au_sbi(sb), dentry);
++ *bcpup = err;
++ if (unlikely(err < 0))
++ goto out;
++ }
++ } else
++ AuDebugOn(btop < *bcpup
++ || au_test_ro(sb, *bcpup, d_inode(dentry)));
++ AuDbg("bcpup %d, btop %d\n", *bcpup, btop);
++
++ if (*bcpup != btop) {
++ err = au_cpup_dirs(dentry, *bcpup);
++ if (unlikely(err))
++ goto out;
++ need_wh = 1;
++ } else {
++ struct au_dinfo *dinfo, *tmp;
++
++ need_wh = -ENOMEM;
++ dinfo = au_di(dentry);
++ tmp = au_di_alloc(sb, AuLsc_DI_TMP);
++ if (tmp) {
++ au_di_cp(tmp, dinfo);
++ au_di_swap(tmp, dinfo);
++ /* returns the number of positive dentries */
++ need_wh = au_lkup_dentry(dentry, btop + 1,
++ /* AuLkup_IGNORE_PERM */ 0);
++ au_di_swap(tmp, dinfo);
++ au_rw_write_unlock(&tmp->di_rwsem);
++ au_di_free(tmp);
++ }
++ }
++ AuDbg("need_wh %d\n", need_wh);
++ err = need_wh;
++
++out:
++ return err;
++}
++
++/*
++ * simple tests for the del-entry operations.
++ * following the checks in vfs, plus the parent-child relationship.
++ */
++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir)
++{
++ int err;
++ umode_t h_mode;
++ struct dentry *h_dentry, *h_latest;
++ struct inode *h_inode;
++
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (d_really_is_positive(dentry)) {
++ err = -ENOENT;
++ if (unlikely(d_is_negative(h_dentry)))
++ goto out;
++ h_inode = d_inode(h_dentry);
++ if (unlikely(!h_inode->i_nlink))
++ goto out;
++
++ h_mode = h_inode->i_mode;
++ if (!isdir) {
++ err = -EISDIR;
++ if (unlikely(S_ISDIR(h_mode)))
++ goto out;
++ } else if (unlikely(!S_ISDIR(h_mode))) {
++ err = -ENOTDIR;
++ goto out;
++ }
++ } else {
++ /* rename(2) case */
++ err = -EIO;
++ if (unlikely(d_is_positive(h_dentry)))
++ goto out;
++ }
++
++ err = -ENOENT;
++ /* expected parent dir is locked */
++ if (unlikely(h_parent != h_dentry->d_parent))
++ goto out;
++ err = 0;
++
++ /*
++ * rmdir a dir may break the consistency on some filesystem.
++ * let's try heavy test.
++ */
++ err = -EACCES;
++ if (unlikely(!au_opt_test(au_mntflags(dentry->d_sb), DIRPERM1)
++ && au_test_h_perm(d_inode(h_parent),
++ MAY_EXEC | MAY_WRITE)))
++ goto out;
++
++ h_latest = au_sio_lkup_one(&dentry->d_name, h_parent);
++ err = -EIO;
++ if (IS_ERR(h_latest))
++ goto out;
++ if (h_latest == h_dentry)
++ err = 0;
++ dput(h_latest);
++
++out:
++ return err;
++}
++
++/*
++ * decide the branch where we operate for @dentry. the branch index will be set
++ * @rbcpup. after diciding it, 'pin' it and store the timestamps of the parent
++ * dir for reverting.
++ * when a new whiteout is necessary, create it.
++ */
++static struct dentry*
++lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup,
++ struct au_dtime *dt, struct au_pin *pin)
+{
-+ int err, rerr;
-+ aufs_bindex_t bindex;
-+ unsigned char diropq;
-+ struct path h_path;
-+ struct dentry *wh_dentry, *parent, *opq_dentry;
-+ struct inode *h_inode;
++ struct dentry *wh_dentry;
+ struct super_block *sb;
-+ struct {
-+ struct au_pin pin;
-+ struct au_dtime dt;
-+ } *a; /* reduce the stack usage */
-+ struct au_wr_dir_args wr_dir_args = {
-+ .force_btgt = -1,
-+ .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR
-+ };
-+
-+ IMustLock(dir);
++ struct path h_path;
++ int err, need_wh;
++ unsigned int udba;
++ aufs_bindex_t bcpup;
+
-+ err = -ENOMEM;
-+ a = kmalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
++ need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup);
++ wh_dentry = ERR_PTR(need_wh);
++ if (unlikely(need_wh < 0))
+ goto out;
+
-+ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
-+ if (unlikely(err))
-+ goto out_free;
-+ err = au_d_may_add(dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
-+
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_write_lock_parent(parent);
-+ wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL,
-+ &a->pin, &wr_dir_args);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_parent;
-+
+ sb = dentry->d_sb;
-+ bindex = au_dbtop(dentry);
-+ h_path.dentry = au_h_dptr(dentry, bindex);
-+ h_path.mnt = au_sbr_mnt(sb, bindex);
-+ err = vfsub_mkdir(au_pinned_h_dir(&a->pin), &h_path, mode);
++ udba = au_opt_udba(sb);
++ bcpup = *rbcpup;
++ err = au_pin(pin, dentry, bcpup, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ wh_dentry = ERR_PTR(err);
+ if (unlikely(err))
-+ goto out_unpin;
++ goto out;
+
-+ /* make the dir opaque */
-+ diropq = 0;
-+ h_inode = d_inode(h_path.dentry);
-+ if (wh_dentry
-+ || au_opt_test(au_mntflags(sb), ALWAYS_DIROPQ)) {
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
-+ opq_dentry = au_diropq_create(dentry, bindex);
-+ inode_unlock(h_inode);
-+ err = PTR_ERR(opq_dentry);
-+ if (IS_ERR(opq_dentry))
-+ goto out_dir;
-+ dput(opq_dentry);
-+ diropq = 1;
++ h_path.dentry = au_pinned_h_parent(pin);
++ if (udba != AuOpt_UDBA_NONE
++ && au_dbtop(dentry) == bcpup) {
++ err = au_may_del(dentry, bcpup, h_path.dentry, isdir);
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_unpin;
+ }
+
-+ err = epilog(dir, bindex, wh_dentry, dentry);
-+ if (!err) {
-+ inc_nlink(dir);
-+ goto out_unpin; /* success */
-+ }
++ h_path.mnt = au_sbr_mnt(sb, bcpup);
++ au_dtime_store(dt, au_pinned_parent(pin), &h_path);
++ wh_dentry = NULL;
++ if (!need_wh)
++ goto out; /* success, no need to create whiteout */
+
-+ /* revert */
-+ if (diropq) {
-+ AuLabel(revert opq);
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
-+ rerr = au_diropq_remove(dentry, bindex);
-+ inode_unlock(h_inode);
-+ if (rerr) {
-+ AuIOErr("%pd reverting diropq failed(%d, %d)\n",
-+ dentry, err, rerr);
-+ err = -EIO;
-+ }
-+ }
++ wh_dentry = au_wh_create(dentry, bcpup, h_path.dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_unpin;
++
++ /* returns with the parent is locked and wh_dentry is dget-ed */
++ goto out; /* success */
+
-+out_dir:
-+ AuLabel(revert dir);
-+ rerr = vfsub_rmdir(au_pinned_h_dir(&a->pin), &h_path);
-+ if (rerr) {
-+ AuIOErr("%pd reverting dir failed(%d, %d)\n",
-+ dentry, err, rerr);
-+ err = -EIO;
-+ }
-+ au_dtime_revert(&a->dt);
+out_unpin:
-+ au_unpin(&a->pin);
-+ dput(wh_dentry);
-+out_parent:
-+ di_write_unlock(parent);
-+out_unlock:
-+ if (unlikely(err)) {
-+ au_update_dbtop(dentry);
-+ d_drop(dentry);
-+ }
-+ aufs_read_unlock(dentry, AuLock_DW);
-+out_free:
-+ au_delayed_kfree(a);
++ au_unpin(pin);
+out:
-+ return err;
++ return wh_dentry;
+}
-diff -urN /usr/share/empty/fs/aufs/i_op.c linux/fs/aufs/i_op.c
---- /usr/share/empty/fs/aufs/i_op.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/i_op.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,1413 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
+
+/*
-+ * inode operations (except add/del/rename)
++ * when removing a dir, rename it to a unique temporary whiteout-ed name first
++ * in order to be revertible and save time for removing many child whiteouts
++ * under the dir.
++ * returns 1 when there are too many child whiteout and caller should remove
++ * them asynchronously. returns 0 when the number of children is enough small to
++ * remove now or the branch fs is a remote fs.
++ * otherwise return an error.
+ */
-+
-+#include <linux/device_cgroup.h>
-+#include <linux/fs_stack.h>
-+#include <linux/namei.h>
-+#include <linux/security.h>
-+#include "aufs.h"
-+
-+static int h_permission(struct inode *h_inode, int mask,
-+ struct path *h_path, int brperm)
++static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex,
++ struct au_nhash *whlist, struct inode *dir)
+{
-+ int err;
-+ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
++ int rmdir_later, err, dirwh;
++ struct dentry *h_dentry;
++ struct super_block *sb;
++ struct inode *inode;
+
-+ err = -EACCES;
-+ if ((write_mask && IS_IMMUTABLE(h_inode))
-+ || ((mask & MAY_EXEC)
-+ && S_ISREG(h_inode->i_mode)
-+ && (path_noexec(h_path)
-+ || !(h_inode->i_mode & S_IXUGO))))
++ sb = dentry->d_sb;
++ SiMustAnyLock(sb);
++ h_dentry = au_h_dptr(dentry, bindex);
++ err = au_whtmp_ren(h_dentry, au_sbr(sb, bindex));
++ if (unlikely(err))
+ goto out;
+
-+ /*
-+ * - skip the lower fs test in the case of write to ro branch.
-+ * - nfs dir permission write check is optimized, but a policy for
-+ * link/rename requires a real check.
-+ * - nfs always sets MS_POSIXACL regardless its mount option 'noacl.'
-+ * in this case, generic_permission() returns -EOPNOTSUPP.
-+ */
-+ if ((write_mask && !au_br_writable(brperm))
-+ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode)
-+ && write_mask && !(mask & MAY_READ))
-+ || !h_inode->i_op->permission) {
-+ /* AuLabel(generic_permission); */
-+ /* AuDbg("get_acl %pf\n", h_inode->i_op->get_acl); */
-+ err = generic_permission(h_inode, mask);
-+ if (err == -EOPNOTSUPP && au_test_nfs_noacl(h_inode))
-+ err = h_inode->i_op->permission(h_inode, mask);
-+ AuTraceErr(err);
-+ } else {
-+ /* AuLabel(h_inode->permission); */
-+ err = h_inode->i_op->permission(h_inode, mask);
-+ AuTraceErr(err);
-+ }
++ /* stop monitoring */
++ inode = d_inode(dentry);
++ au_hn_free(au_hi(inode, bindex));
+
-+ if (!err)
-+ err = devcgroup_inode_permission(h_inode, mask);
-+ if (!err)
-+ err = security_inode_permission(h_inode, mask);
++ if (!au_test_fs_remote(h_dentry->d_sb)) {
++ dirwh = au_sbi(sb)->si_dirwh;
++ rmdir_later = (dirwh <= 1);
++ if (!rmdir_later)
++ rmdir_later = au_nhash_test_longer_wh(whlist, bindex,
++ dirwh);
++ if (rmdir_later)
++ return rmdir_later;
++ }
+
-+#if 0
-+ if (!err) {
-+ /* todo: do we need to call ima_path_check()? */
-+ struct path h_path = {
-+ .dentry =
-+ .mnt = h_mnt
-+ };
-+ err = ima_path_check(&h_path,
-+ mask & (MAY_READ | MAY_WRITE | MAY_EXEC),
-+ IMA_COUNT_LEAVE);
++ err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist);
++ if (unlikely(err)) {
++ AuIOErr("rmdir %pd, b%d failed, %d. ignored\n",
++ h_dentry, bindex, err);
++ err = 0;
+ }
-+#endif
+
+out:
++ AuTraceErr(err);
+ return err;
+}
+
-+static int aufs_permission(struct inode *inode, int mask)
++/*
++ * final procedure for deleting a entry.
++ * maintain dentry and iattr.
++ */
++static void epilog(struct inode *dir, struct dentry *dentry,
++ aufs_bindex_t bindex)
+{
-+ int err;
-+ aufs_bindex_t bindex, bbot;
-+ const unsigned char isdir = !!S_ISDIR(inode->i_mode),
-+ write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
-+ struct inode *h_inode;
-+ struct super_block *sb;
-+ struct au_branch *br;
-+
-+ /* todo: support rcu-walk? */
-+ if (mask & MAY_NOT_BLOCK)
-+ return -ECHILD;
-+
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ ii_read_lock_child(inode);
-+#if 0
-+ err = au_iigen_test(inode, au_sigen(sb));
-+ if (unlikely(err))
-+ goto out;
-+#endif
++ struct inode *inode;
+
-+ if (!isdir
-+ || write_mask
-+ || au_opt_test(au_mntflags(sb), DIRPERM1)) {
-+ err = au_busy_or_stale();
-+ h_inode = au_h_iptr(inode, au_ibtop(inode));
-+ if (unlikely(!h_inode
-+ || (h_inode->i_mode & S_IFMT)
-+ != (inode->i_mode & S_IFMT)))
-+ goto out;
++ inode = d_inode(dentry);
++ d_drop(dentry);
++ inode->i_ctime = dir->i_ctime;
+
-+ err = 0;
-+ bindex = au_ibtop(inode);
-+ br = au_sbr(sb, bindex);
-+ err = h_permission(h_inode, mask, &br->br_path, br->br_perm);
-+ if (write_mask
-+ && !err
-+ && !special_file(h_inode->i_mode)) {
-+ /* test whether the upper writable branch exists */
-+ err = -EROFS;
-+ for (; bindex >= 0; bindex--)
-+ if (!au_br_rdonly(au_sbr(sb, bindex))) {
-+ err = 0;
-+ break;
-+ }
-+ }
-+ goto out;
-+ }
++ au_dir_ts(dir, bindex);
++ dir->i_version++;
++}
+
-+ /* non-write to dir */
-+ err = 0;
-+ bbot = au_ibbot(inode);
-+ for (bindex = au_ibtop(inode); !err && bindex <= bbot; bindex++) {
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (h_inode) {
-+ err = au_busy_or_stale();
-+ if (unlikely(!S_ISDIR(h_inode->i_mode)))
-+ break;
++/*
++ * when an error happened, remove the created whiteout and revert everything.
++ */
++static int do_revert(int err, struct inode *dir, aufs_bindex_t bindex,
++ aufs_bindex_t bwh, struct dentry *wh_dentry,
++ struct dentry *dentry, struct au_dtime *dt)
++{
++ int rerr;
++ struct path h_path = {
++ .dentry = wh_dentry,
++ .mnt = au_sbr_mnt(dir->i_sb, bindex)
++ };
+
-+ br = au_sbr(sb, bindex);
-+ err = h_permission(h_inode, mask, &br->br_path,
-+ br->br_perm);
-+ }
++ rerr = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, dentry);
++ if (!rerr) {
++ au_set_dbwh(dentry, bwh);
++ au_dtime_revert(dt);
++ return 0;
+ }
+
-+out:
-+ ii_read_unlock(inode);
-+ si_read_unlock(sb);
-+ return err;
++ AuIOErr("%pd reverting whiteout failed(%d, %d)\n", dentry, err, rerr);
++ return -EIO;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry,
-+ unsigned int flags)
++int aufs_unlink(struct inode *dir, struct dentry *dentry)
+{
-+ struct dentry *ret, *parent;
-+ struct inode *inode;
-+ struct super_block *sb;
-+ int err, npositive;
++ int err;
++ aufs_bindex_t bwh, bindex, btop;
++ struct inode *inode, *h_dir, *delegated;
++ struct dentry *parent, *wh_dentry;
++ /* to reuduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct au_pin pin;
++ struct path h_path;
++ } *a;
+
+ IMustLock(dir);
+
-+ /* todo: support rcu-walk? */
-+ ret = ERR_PTR(-ECHILD);
-+ if (flags & LOOKUP_RCU)
-+ goto out;
-+
-+ ret = ERR_PTR(-ENAMETOOLONG);
-+ if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN))
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
+ goto out;
+
-+ sb = dir->i_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ ret = ERR_PTR(err);
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
+ if (unlikely(err))
-+ goto out;
-+
-+ err = au_di_init(dentry);
-+ ret = ERR_PTR(err);
++ goto out_free;
++ err = au_d_hashed_positive(dentry);
+ if (unlikely(err))
-+ goto out_si;
-+
-+ inode = NULL;
-+ npositive = 0; /* suppress a warning */
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_read_lock_parent(parent, AuLock_IR);
-+ err = au_alive_dir(parent);
-+ if (!err)
-+ err = au_digen_test(parent, au_sigen(sb));
-+ if (!err) {
-+ /* regardless LOOKUP_CREATE, always ALLOW_NEG */
-+ npositive = au_lkup_dentry(dentry, au_dbtop(parent),
-+ AuLkup_ALLOW_NEG);
-+ err = npositive;
-+ }
-+ di_read_unlock(parent, AuLock_IR);
-+ ret = ERR_PTR(err);
-+ if (unlikely(err < 0))
+ goto out_unlock;
++ inode = d_inode(dentry);
++ IMustLock(inode);
++ err = -EISDIR;
++ if (unlikely(d_is_dir(dentry)))
++ goto out_unlock; /* possible? */
+
-+ if (npositive) {
-+ inode = au_new_inode(dentry, /*must_new*/0);
-+ if (IS_ERR(inode)) {
-+ ret = (void *)inode;
-+ inode = NULL;
-+ goto out_unlock;
-+ }
-+ }
++ btop = au_dbtop(dentry);
++ bwh = au_dbwh(dentry);
++ bindex = -1;
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &a->dt,
++ &a->pin);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
+
-+ if (inode)
-+ atomic_inc(&inode->i_count);
-+ ret = d_splice_alias(inode, dentry);
-+#if 0
-+ if (unlikely(d_need_lookup(dentry))) {
-+ spin_lock(&dentry->d_lock);
-+ dentry->d_flags &= ~DCACHE_NEED_LOOKUP;
-+ spin_unlock(&dentry->d_lock);
-+ } else
-+#endif
-+ if (inode) {
-+ if (!IS_ERR(ret)) {
-+ iput(inode);
-+ if (ret && ret != dentry)
-+ ii_write_unlock(inode);
-+ } else {
-+ ii_write_unlock(inode);
-+ iput(inode);
-+ inode = NULL;
++ a->h_path.mnt = au_sbr_mnt(dentry->d_sb, btop);
++ a->h_path.dentry = au_h_dptr(dentry, btop);
++ dget(a->h_path.dentry);
++ if (bindex == btop) {
++ h_dir = au_pinned_h_dir(&a->pin);
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, &a->h_path, &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
+ }
++ } else {
++ /* dir inode is locked */
++ h_dir = d_inode(wh_dentry->d_parent);
++ IMustLock(h_dir);
++ err = 0;
+ }
+
-+out_unlock:
-+ di_write_unlock(dentry);
-+out_si:
-+ si_read_unlock(sb);
-+out:
-+ return ret;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct aopen_node {
-+ struct hlist_node hlist;
-+ struct file *file, *h_file;
-+};
++ if (!err) {
++ vfsub_drop_nlink(inode);
++ epilog(dir, dentry, bindex);
+
-+static int au_do_aopen(struct inode *inode, struct file *file)
-+{
-+ struct au_sphlhead *aopen;
-+ struct aopen_node *node;
-+ struct au_do_open_args args = {
-+ .no_lock = 1,
-+ .open = au_do_open_nondir
-+ };
++ /* update target timestamps */
++ if (bindex == btop) {
++ vfsub_update_h_iattr(&a->h_path, /*did*/NULL);
++ /*ignore*/
++ inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
++ } else
++ /* todo: this timestamp may be reverted later */
++ inode->i_ctime = h_dir->i_ctime;
++ goto out_unpin; /* success */
++ }
+
-+ aopen = &au_sbi(inode->i_sb)->si_aopen;
-+ spin_lock(&aopen->spin);
-+ hlist_for_each_entry(node, &aopen->head, hlist)
-+ if (node->file == file) {
-+ args.h_file = node->h_file;
-+ break;
-+ }
-+ spin_unlock(&aopen->spin);
-+ /* AuDebugOn(!args.h_file); */
++ /* revert */
++ if (wh_dentry) {
++ int rerr;
+
-+ return au_do_open(file, &args);
-+}
++ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
++ &a->dt);
++ if (rerr)
++ err = rerr;
++ }
+
-+static int aufs_atomic_open(struct inode *dir, struct dentry *dentry,
-+ struct file *file, unsigned int open_flag,
-+ umode_t create_mode, int *opened)
-+{
-+ int err, h_opened = *opened;
-+ unsigned int lkup_flags;
-+ struct dentry *parent, *d;
-+ struct au_sphlhead *aopen;
-+ struct vfsub_aopen_args args = {
-+ .open_flag = open_flag,
-+ .create_mode = create_mode,
-+ .opened = &h_opened
-+ };
-+ struct aopen_node aopen_node = {
-+ .file = file
-+ };
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++ dput(a->h_path.dentry);
++out_parent:
++ di_write_unlock(parent);
++out_unlock:
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ kfree(a);
++out:
++ return err;
++}
+
-+ IMustLock(dir);
-+ AuDbg("open_flag 0%o\n", open_flag);
-+ AuDbgDentry(dentry);
++int aufs_rmdir(struct inode *dir, struct dentry *dentry)
++{
++ int err, rmdir_later;
++ aufs_bindex_t bwh, bindex, btop;
++ struct inode *inode;
++ struct dentry *parent, *wh_dentry, *h_dentry;
++ struct au_whtmp_rmdir *args;
++ /* to reuduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct au_pin pin;
++ } *a;
+
-+ err = 0;
-+ if (!au_di(dentry)) {
-+ lkup_flags = LOOKUP_OPEN;
-+ if (open_flag & O_CREAT)
-+ lkup_flags |= LOOKUP_CREATE;
-+ d = aufs_lookup(dir, dentry, lkup_flags);
-+ if (IS_ERR(d)) {
-+ err = PTR_ERR(d);
-+ AuTraceErr(err);
-+ goto out;
-+ } else if (d) {
-+ /*
-+ * obsoleted dentry found.
-+ * another error will be returned later.
-+ */
-+ d_drop(d);
-+ AuDbgDentry(d);
-+ dput(d);
-+ }
-+ AuDbgDentry(dentry);
-+ }
++ IMustLock(dir);
+
-+ if (d_is_positive(dentry)
-+ || d_unhashed(dentry)
-+ || d_unlinked(dentry)
-+ || !(open_flag & O_CREAT))
-+ goto out_no_open;
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
+
+ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
+ if (unlikely(err))
-+ goto out;
-+
-+ parent = dentry->d_parent; /* dir is locked */
-+ di_write_lock_parent(parent);
-+ err = au_lkup_dentry(dentry, /*btop*/0, AuLkup_ALLOW_NEG);
++ goto out_free;
++ err = au_alive_dir(dentry);
+ if (unlikely(err))
+ goto out_unlock;
++ inode = d_inode(dentry);
++ IMustLock(inode);
++ err = -ENOTDIR;
++ if (unlikely(!d_is_dir(dentry)))
++ goto out_unlock; /* possible? */
+
-+ AuDbgDentry(dentry);
-+ if (d_is_positive(dentry))
++ err = -ENOMEM;
++ args = au_whtmp_rmdir_alloc(dir->i_sb, GFP_NOFS);
++ if (unlikely(!args))
+ goto out_unlock;
+
-+ args.file = get_empty_filp();
-+ err = PTR_ERR(args.file);
-+ if (IS_ERR(args.file))
-+ goto out_unlock;
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ err = au_test_empty(dentry, &args->whlist);
++ if (unlikely(err))
++ goto out_parent;
+
-+ args.file->f_flags = file->f_flags;
-+ err = au_aopen_or_create(dir, dentry, &args);
-+ AuTraceErr(err);
-+ AuDbgFile(args.file);
-+ if (unlikely(err < 0)) {
-+ if (h_opened & FILE_OPENED)
-+ fput(args.file);
-+ else
-+ put_filp(args.file);
-+ goto out_unlock;
++ btop = au_dbtop(dentry);
++ bwh = au_dbwh(dentry);
++ bindex = -1;
++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &a->dt,
++ &a->pin);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ h_dentry = au_h_dptr(dentry, btop);
++ dget(h_dentry);
++ rmdir_later = 0;
++ if (bindex == btop) {
++ err = renwh_and_rmdir(dentry, btop, &args->whlist, dir);
++ if (err > 0) {
++ rmdir_later = err;
++ err = 0;
++ }
++ } else {
++ /* stop monitoring */
++ au_hn_free(au_hi(inode, btop));
++
++ /* dir inode is locked */
++ IMustLock(d_inode(wh_dentry->d_parent));
++ err = 0;
+ }
+
-+ /* some filesystems don't set FILE_CREATED while succeeded? */
-+ *opened |= FILE_CREATED;
-+ if (h_opened & FILE_OPENED)
-+ aopen_node.h_file = args.file;
-+ else {
-+ put_filp(args.file);
-+ args.file = NULL;
++ if (!err) {
++ vfsub_dead_dir(inode);
++ au_set_dbdiropq(dentry, -1);
++ epilog(dir, dentry, bindex);
++
++ if (rmdir_later) {
++ au_whtmp_kick_rmdir(dir, btop, h_dentry, args);
++ args = NULL;
++ }
++
++ goto out_unpin; /* success */
+ }
-+ aopen = &au_sbi(dir->i_sb)->si_aopen;
-+ au_sphl_add(&aopen_node.hlist, aopen);
-+ err = finish_open(file, dentry, au_do_aopen, opened);
-+ au_sphl_del(&aopen_node.hlist, aopen);
-+ AuTraceErr(err);
-+ AuDbgFile(file);
-+ if (aopen_node.h_file)
-+ fput(aopen_node.h_file);
+
-+out_unlock:
++ /* revert */
++ AuLabel(revert);
++ if (wh_dentry) {
++ int rerr;
++
++ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
++ &a->dt);
++ if (rerr)
++ err = rerr;
++ }
++
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++ dput(h_dentry);
++out_parent:
+ di_write_unlock(parent);
++ if (args)
++ au_whtmp_rmdir_free(args);
++out_unlock:
+ aufs_read_unlock(dentry, AuLock_DW);
-+ AuDbgDentry(dentry);
-+ if (unlikely(err < 0))
-+ goto out;
-+out_no_open:
-+ if (err >= 0 && !(*opened & FILE_CREATED)) {
-+ AuLabel(out_no_open);
-+ dget(dentry);
-+ err = finish_no_open(file, dentry);
-+ }
++out_free:
++ kfree(a);
+out:
-+ AuDbg("%pd%s%s\n", dentry,
-+ (*opened & FILE_CREATED) ? " created" : "",
-+ (*opened & FILE_OPENED) ? " opened" : "");
+ AuTraceErr(err);
+ return err;
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/i_op_ren.c linux-4.14/fs/aufs/i_op_ren.c
+--- linux-4.14/fs/aufs/i_op_ren.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/i_op_ren.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,1246 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
++/*
++ * inode operation (rename entry)
++ * todo: this is crazy monster
++ */
+
-+/* ---------------------------------------------------------------------- */
++#include "aufs.h"
+
-+static int au_wr_dir_cpup(struct dentry *dentry, struct dentry *parent,
-+ const unsigned char add_entry, aufs_bindex_t bcpup,
-+ aufs_bindex_t btop)
-+{
-+ int err;
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
++enum { AuSRC, AuDST, AuSrcDst };
++enum { AuPARENT, AuCHILD, AuParentChild };
+
-+ if (add_entry)
-+ IMustLock(d_inode(parent));
-+ else
-+ di_write_lock_parent(parent);
++#define AuRen_ISDIR_SRC 1
++#define AuRen_ISDIR_DST (1 << 1)
++#define AuRen_ISSAMEDIR (1 << 2)
++#define AuRen_WHSRC (1 << 3)
++#define AuRen_WHDST (1 << 4)
++#define AuRen_MNT_WRITE (1 << 5)
++#define AuRen_DT_DSTDIR (1 << 6)
++#define AuRen_DIROPQ_SRC (1 << 7)
++#define AuRen_DIROPQ_DST (1 << 8)
++#define AuRen_DIRREN (1 << 9)
++#define AuRen_DROPPED_SRC (1 << 10)
++#define AuRen_DROPPED_DST (1 << 11)
++#define au_ftest_ren(flags, name) ((flags) & AuRen_##name)
++#define au_fset_ren(flags, name) \
++ do { (flags) |= AuRen_##name; } while (0)
++#define au_fclr_ren(flags, name) \
++ do { (flags) &= ~AuRen_##name; } while (0)
+
-+ err = 0;
-+ if (!au_h_dptr(parent, bcpup)) {
-+ if (btop > bcpup)
-+ err = au_cpup_dirs(dentry, bcpup);
-+ else if (btop < bcpup)
-+ err = au_cpdown_dirs(dentry, bcpup);
-+ else
-+ BUG();
-+ }
-+ if (!err && add_entry && !au_ftest_wrdir(add_entry, TMPFILE)) {
-+ h_parent = au_h_dptr(parent, bcpup);
-+ h_dir = d_inode(h_parent);
-+ inode_lock_nested(h_dir, AuLsc_I_PARENT);
-+ err = au_lkup_neg(dentry, bcpup, /*wh*/0);
-+ /* todo: no unlock here */
-+ inode_unlock(h_dir);
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuRen_DIRREN
++#define AuRen_DIRREN 0
++#endif
+
-+ AuDbg("bcpup %d\n", bcpup);
-+ if (!err) {
-+ if (d_really_is_negative(dentry))
-+ au_set_h_dptr(dentry, btop, NULL);
-+ au_update_dbrange(dentry, /*do_put_zero*/0);
-+ }
-+ }
++struct au_ren_args {
++ struct {
++ struct dentry *dentry, *h_dentry, *parent, *h_parent,
++ *wh_dentry;
++ struct inode *dir, *inode;
++ struct au_hinode *hdir, *hinode;
++ struct au_dtime dt[AuParentChild];
++ aufs_bindex_t btop, bdiropq;
++ } sd[AuSrcDst];
+
-+ if (!add_entry)
-+ di_write_unlock(parent);
-+ if (!err)
-+ err = bcpup; /* success */
++#define src_dentry sd[AuSRC].dentry
++#define src_dir sd[AuSRC].dir
++#define src_inode sd[AuSRC].inode
++#define src_h_dentry sd[AuSRC].h_dentry
++#define src_parent sd[AuSRC].parent
++#define src_h_parent sd[AuSRC].h_parent
++#define src_wh_dentry sd[AuSRC].wh_dentry
++#define src_hdir sd[AuSRC].hdir
++#define src_hinode sd[AuSRC].hinode
++#define src_h_dir sd[AuSRC].hdir->hi_inode
++#define src_dt sd[AuSRC].dt
++#define src_btop sd[AuSRC].btop
++#define src_bdiropq sd[AuSRC].bdiropq
+
-+ AuTraceErr(err);
-+ return err;
-+}
++#define dst_dentry sd[AuDST].dentry
++#define dst_dir sd[AuDST].dir
++#define dst_inode sd[AuDST].inode
++#define dst_h_dentry sd[AuDST].h_dentry
++#define dst_parent sd[AuDST].parent
++#define dst_h_parent sd[AuDST].h_parent
++#define dst_wh_dentry sd[AuDST].wh_dentry
++#define dst_hdir sd[AuDST].hdir
++#define dst_hinode sd[AuDST].hinode
++#define dst_h_dir sd[AuDST].hdir->hi_inode
++#define dst_dt sd[AuDST].dt
++#define dst_btop sd[AuDST].btop
++#define dst_bdiropq sd[AuDST].bdiropq
++
++ struct dentry *h_trap;
++ struct au_branch *br;
++ struct path h_path;
++ struct au_nhash whlist;
++ aufs_bindex_t btgt, src_bwh;
++
++ struct {
++ unsigned short auren_flags;
++ unsigned char flags; /* syscall parameter */
++ unsigned char exchange;
++ } __packed;
++
++ struct au_whtmp_rmdir *thargs;
++ struct dentry *h_dst;
++ struct au_hinode *h_root;
++};
++
++/* ---------------------------------------------------------------------- */
+
+/*
-+ * decide the branch and the parent dir where we will create a new entry.
-+ * returns new bindex or an error.
-+ * copyup the parent dir if needed.
++ * functions for reverting.
++ * when an error happened in a single rename systemcall, we should revert
++ * everything as if nothing happened.
++ * we don't need to revert the copied-up/down the parent dir since they are
++ * harmless.
+ */
-+int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
-+ struct au_wr_dir_args *args)
++
++#define RevertFailure(fmt, ...) do { \
++ AuIOErr("revert failure: " fmt " (%d, %d)\n", \
++ ##__VA_ARGS__, err, rerr); \
++ err = -EIO; \
++} while (0)
++
++static void au_ren_do_rev_diropq(int err, struct au_ren_args *a, int idx)
+{
-+ int err;
-+ unsigned int flags;
-+ aufs_bindex_t bcpup, btop, src_btop;
-+ const unsigned char add_entry
-+ = au_ftest_wrdir(args->flags, ADD_ENTRY)
-+ | au_ftest_wrdir(args->flags, TMPFILE);
-+ struct super_block *sb;
-+ struct dentry *parent;
-+ struct au_sbinfo *sbinfo;
++ int rerr;
++ struct dentry *d;
++#define src_or_dst(member) a->sd[idx].member
+
-+ sb = dentry->d_sb;
-+ sbinfo = au_sbi(sb);
-+ parent = dget_parent(dentry);
-+ btop = au_dbtop(dentry);
-+ bcpup = btop;
-+ if (args->force_btgt < 0) {
-+ if (src_dentry) {
-+ src_btop = au_dbtop(src_dentry);
-+ if (src_btop < btop)
-+ bcpup = src_btop;
-+ } else if (add_entry) {
-+ flags = 0;
-+ if (au_ftest_wrdir(args->flags, ISDIR))
-+ au_fset_wbr(flags, DIR);
-+ err = AuWbrCreate(sbinfo, dentry, flags);
-+ bcpup = err;
-+ }
++ d = src_or_dst(dentry); /* {src,dst}_dentry */
++ au_hn_inode_lock_nested(src_or_dst(hinode), AuLsc_I_CHILD);
++ rerr = au_diropq_remove(d, a->btgt);
++ au_hn_inode_unlock(src_or_dst(hinode));
++ au_set_dbdiropq(d, src_or_dst(bdiropq));
++ if (rerr)
++ RevertFailure("remove diropq %pd", d);
+
-+ if (bcpup < 0 || au_test_ro(sb, bcpup, d_inode(dentry))) {
-+ if (add_entry)
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ else {
-+ if (!IS_ROOT(dentry)) {
-+ di_read_lock_parent(parent, !AuLock_IR);
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ di_read_unlock(parent, !AuLock_IR);
-+ } else
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ }
-+ bcpup = err;
-+ if (unlikely(err < 0))
-+ goto out;
-+ }
-+ } else {
-+ bcpup = args->force_btgt;
-+ AuDebugOn(au_test_ro(sb, bcpup, d_inode(dentry)));
-+ }
++#undef src_or_dst_
++}
+
-+ AuDbg("btop %d, bcpup %d\n", btop, bcpup);
-+ err = bcpup;
-+ if (bcpup == btop)
-+ goto out; /* success */
++static void au_ren_rev_diropq(int err, struct au_ren_args *a)
++{
++ if (au_ftest_ren(a->auren_flags, DIROPQ_SRC))
++ au_ren_do_rev_diropq(err, a, AuSRC);
++ if (au_ftest_ren(a->auren_flags, DIROPQ_DST))
++ au_ren_do_rev_diropq(err, a, AuDST);
++}
+
-+ /* copyup the new parent into the branch we process */
-+ err = au_wr_dir_cpup(dentry, parent, add_entry, bcpup, btop);
-+ if (err >= 0) {
-+ if (d_really_is_negative(dentry)) {
-+ au_set_h_dptr(dentry, btop, NULL);
-+ au_set_dbtop(dentry, bcpup);
-+ au_set_dbbot(dentry, bcpup);
-+ }
-+ AuDebugOn(add_entry
-+ && !au_ftest_wrdir(args->flags, TMPFILE)
-+ && !au_h_dptr(dentry, bcpup));
++static void au_ren_rev_rename(int err, struct au_ren_args *a)
++{
++ int rerr;
++ struct inode *delegated;
++
++ a->h_path.dentry = vfsub_lkup_one(&a->src_dentry->d_name,
++ a->src_h_parent);
++ rerr = PTR_ERR(a->h_path.dentry);
++ if (IS_ERR(a->h_path.dentry)) {
++ RevertFailure("lkup one %pd", a->src_dentry);
++ return;
+ }
+
-+out:
-+ dput(parent);
-+ return err;
++ delegated = NULL;
++ rerr = vfsub_rename(a->dst_h_dir,
++ au_h_dptr(a->src_dentry, a->btgt),
++ a->src_h_dir, &a->h_path, &delegated, a->flags);
++ if (unlikely(rerr == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */
++ if (rerr)
++ RevertFailure("rename %pd", a->src_dentry);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_pin_hdir_unlock(struct au_pin *p)
++static void au_ren_rev_whtmp(int err, struct au_ren_args *a)
+{
-+ if (p->hdir)
-+ au_hn_inode_unlock(p->hdir);
++ int rerr;
++ struct inode *delegated;
++
++ a->h_path.dentry = vfsub_lkup_one(&a->dst_dentry->d_name,
++ a->dst_h_parent);
++ rerr = PTR_ERR(a->h_path.dentry);
++ if (IS_ERR(a->h_path.dentry)) {
++ RevertFailure("lkup one %pd", a->dst_dentry);
++ return;
++ }
++ if (d_is_positive(a->h_path.dentry)) {
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ return;
++ }
++
++ delegated = NULL;
++ rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path,
++ &delegated, a->flags);
++ if (unlikely(rerr == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ if (!rerr)
++ au_set_h_dptr(a->dst_dentry, a->btgt, dget(a->h_dst));
++ else
++ RevertFailure("rename %pd", a->h_dst);
+}
+
-+int au_pin_hdir_lock(struct au_pin *p)
++static void au_ren_rev_whsrc(int err, struct au_ren_args *a)
+{
-+ int err;
++ int rerr;
+
-+ err = 0;
-+ if (!p->hdir)
-+ goto out;
++ a->h_path.dentry = a->src_wh_dentry;
++ rerr = au_wh_unlink_dentry(a->src_h_dir, &a->h_path, a->src_dentry);
++ au_set_dbwh(a->src_dentry, a->src_bwh);
++ if (rerr)
++ RevertFailure("unlink %pd", a->src_wh_dentry);
++}
++#undef RevertFailure
+
-+ /* even if an error happens later, keep this lock */
-+ au_hn_inode_lock_nested(p->hdir, p->lsc_hi);
++/* ---------------------------------------------------------------------- */
+
-+ err = -EBUSY;
-+ if (unlikely(p->hdir->hi_inode != d_inode(p->h_parent)))
-+ goto out;
++/*
++ * when we have to copyup the renaming entry, do it with the rename-target name
++ * in order to minimize the cost (the later actual rename is unnecessary).
++ * otherwise rename it on the target branch.
++ */
++static int au_ren_or_cpup(struct au_ren_args *a)
++{
++ int err;
++ struct dentry *d;
++ struct inode *delegated;
+
-+ err = 0;
-+ if (p->h_dentry)
-+ err = au_h_verify(p->h_dentry, p->udba, p->hdir->hi_inode,
-+ p->h_parent, p->br);
++ d = a->src_dentry;
++ if (au_dbtop(d) == a->btgt) {
++ a->h_path.dentry = a->dst_h_dentry;
++ AuDebugOn(au_dbtop(d) != a->btgt);
++ delegated = NULL;
++ err = vfsub_rename(a->src_h_dir, au_h_dptr(d, a->btgt),
++ a->dst_h_dir, &a->h_path, &delegated,
++ a->flags);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ } else
++ BUG();
++
++ if (!err && a->h_dst)
++ /* it will be set to dinfo later */
++ dget(a->h_dst);
+
-+out:
+ return err;
+}
+
-+int au_pin_hdir_relock(struct au_pin *p)
++/* cf. aufs_rmdir() */
++static int au_ren_del_whtmp(struct au_ren_args *a)
+{
-+ int err, i;
-+ struct inode *h_i;
-+ struct dentry *h_d[] = {
-+ p->h_dentry,
-+ p->h_parent
-+ };
-+
-+ err = au_pin_hdir_lock(p);
-+ if (unlikely(err))
-+ goto out;
++ int err;
++ struct inode *dir;
+
-+ for (i = 0; !err && i < sizeof(h_d)/sizeof(*h_d); i++) {
-+ if (!h_d[i])
-+ continue;
-+ if (d_is_positive(h_d[i])) {
-+ h_i = d_inode(h_d[i]);
-+ err = !h_i->i_nlink;
-+ }
++ dir = a->dst_dir;
++ SiMustAnyLock(dir->i_sb);
++ if (!au_nhash_test_longer_wh(&a->whlist, a->btgt,
++ au_sbi(dir->i_sb)->si_dirwh)
++ || au_test_fs_remote(a->h_dst->d_sb)) {
++ err = au_whtmp_rmdir(dir, a->btgt, a->h_dst, &a->whlist);
++ if (unlikely(err))
++ pr_warn("failed removing whtmp dir %pd (%d), "
++ "ignored.\n", a->h_dst, err);
++ } else {
++ au_nhash_wh_free(&a->thargs->whlist);
++ a->thargs->whlist = a->whlist;
++ a->whlist.nh_num = 0;
++ au_whtmp_kick_rmdir(dir, a->btgt, a->h_dst, a->thargs);
++ dput(a->h_dst);
++ a->thargs = NULL;
+ }
+
-+out:
-+ return err;
++ return 0;
+}
+
-+static void au_pin_hdir_set_owner(struct au_pin *p, struct task_struct *task)
++/* make it 'opaque' dir. */
++static int au_ren_do_diropq(struct au_ren_args *a, int idx)
+{
-+#if !defined(CONFIG_RWSEM_GENERIC_SPINLOCK) && defined(CONFIG_RWSEM_SPIN_ON_OWNER)
-+ p->hdir->hi_inode->i_rwsem.owner = task;
-+#endif
++ int err;
++ struct dentry *d, *diropq;
++#define src_or_dst(member) a->sd[idx].member
++
++ err = 0;
++ d = src_or_dst(dentry); /* {src,dst}_dentry */
++ src_or_dst(bdiropq) = au_dbdiropq(d);
++ src_or_dst(hinode) = au_hi(src_or_dst(inode), a->btgt);
++ au_hn_inode_lock_nested(src_or_dst(hinode), AuLsc_I_CHILD);
++ diropq = au_diropq_create(d, a->btgt);
++ au_hn_inode_unlock(src_or_dst(hinode));
++ if (IS_ERR(diropq))
++ err = PTR_ERR(diropq);
++ else
++ dput(diropq);
++
++#undef src_or_dst_
++ return err;
+}
+
-+void au_pin_hdir_acquire_nest(struct au_pin *p)
++static int au_ren_diropq(struct au_ren_args *a)
+{
-+ if (p->hdir) {
-+ rwsem_acquire_nest(&p->hdir->hi_inode->i_rwsem.dep_map,
-+ p->lsc_hi, 0, NULL, _RET_IP_);
-+ au_pin_hdir_set_owner(p, current);
++ int err;
++ unsigned char always;
++ struct dentry *d;
++
++ err = 0;
++ d = a->dst_dentry; /* already renamed on the branch */
++ always = !!au_opt_test(au_mntflags(d->d_sb), ALWAYS_DIROPQ);
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ && !au_ftest_ren(a->auren_flags, DIRREN)
++ && a->btgt != au_dbdiropq(a->src_dentry)
++ && (a->dst_wh_dentry
++ || a->btgt <= au_dbdiropq(d)
++ /* hide the lower to keep xino */
++ /* the lowers may not be a dir, but we hide them anyway */
++ || a->btgt < au_dbbot(d)
++ || always)) {
++ AuDbg("here\n");
++ err = au_ren_do_diropq(a, AuSRC);
++ if (unlikely(err))
++ goto out;
++ au_fset_ren(a->auren_flags, DIROPQ_SRC);
+ }
-+}
++ if (!a->exchange)
++ goto out; /* success */
+
-+void au_pin_hdir_release(struct au_pin *p)
-+{
-+ if (p->hdir) {
-+ au_pin_hdir_set_owner(p, p->task);
-+ rwsem_release(&p->hdir->hi_inode->i_rwsem.dep_map, 1, _RET_IP_);
++ d = a->src_dentry; /* already renamed on the branch */
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST)
++ && a->btgt != au_dbdiropq(a->dst_dentry)
++ && (a->btgt < au_dbdiropq(d)
++ || a->btgt < au_dbbot(d)
++ || always)) {
++ AuDbgDentry(a->src_dentry);
++ AuDbgDentry(a->dst_dentry);
++ err = au_ren_do_diropq(a, AuDST);
++ if (unlikely(err))
++ goto out_rev_src;
++ au_fset_ren(a->auren_flags, DIROPQ_DST);
+ }
-+}
++ goto out; /* success */
+
-+struct dentry *au_pinned_h_parent(struct au_pin *pin)
-+{
-+ if (pin && pin->parent)
-+ return au_h_dptr(pin->parent, pin->bindex);
-+ return NULL;
++out_rev_src:
++ AuDbg("err %d, reverting src\n", err);
++ au_ren_rev_diropq(err, a);
++out:
++ return err;
+}
+
-+void au_unpin(struct au_pin *p)
++static int do_rename(struct au_ren_args *a)
+{
-+ if (p->hdir)
-+ au_pin_hdir_unlock(p);
-+ if (p->h_mnt && au_ftest_pin(p->flags, MNT_WRITE))
-+ vfsub_mnt_drop_write(p->h_mnt);
-+ if (!p->hdir)
-+ return;
++ int err;
++ struct dentry *d, *h_d;
+
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_unlock(p->parent, AuLock_IR);
-+ iput(p->hdir->hi_inode);
-+ dput(p->parent);
-+ p->parent = NULL;
-+ p->hdir = NULL;
-+ p->h_mnt = NULL;
-+ /* do not clear p->task */
-+}
++ if (!a->exchange) {
++ /* prepare workqueue args for asynchronous rmdir */
++ h_d = a->dst_h_dentry;
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST)
++ /* && !au_ftest_ren(a->auren_flags, DIRREN) */
++ && d_is_positive(h_d)) {
++ err = -ENOMEM;
++ a->thargs = au_whtmp_rmdir_alloc(a->src_dentry->d_sb,
++ GFP_NOFS);
++ if (unlikely(!a->thargs))
++ goto out;
++ a->h_dst = dget(h_d);
++ }
++
++ /* create whiteout for src_dentry */
++ if (au_ftest_ren(a->auren_flags, WHSRC)) {
++ a->src_bwh = au_dbwh(a->src_dentry);
++ AuDebugOn(a->src_bwh >= 0);
++ a->src_wh_dentry = au_wh_create(a->src_dentry, a->btgt,
++ a->src_h_parent);
++ err = PTR_ERR(a->src_wh_dentry);
++ if (IS_ERR(a->src_wh_dentry))
++ goto out_thargs;
++ }
++
++ /* lookup whiteout for dentry */
++ if (au_ftest_ren(a->auren_flags, WHDST)) {
++ h_d = au_wh_lkup(a->dst_h_parent,
++ &a->dst_dentry->d_name, a->br);
++ err = PTR_ERR(h_d);
++ if (IS_ERR(h_d))
++ goto out_whsrc;
++ if (d_is_negative(h_d))
++ dput(h_d);
++ else
++ a->dst_wh_dentry = h_d;
++ }
+
-+int au_do_pin(struct au_pin *p)
-+{
-+ int err;
-+ struct super_block *sb;
-+ struct inode *h_dir;
++ /* rename dentry to tmpwh */
++ if (a->thargs) {
++ err = au_whtmp_ren(a->dst_h_dentry, a->br);
++ if (unlikely(err))
++ goto out_whdst;
+
-+ err = 0;
-+ sb = p->dentry->d_sb;
-+ p->br = au_sbr(sb, p->bindex);
-+ if (IS_ROOT(p->dentry)) {
-+ if (au_ftest_pin(p->flags, MNT_WRITE)) {
-+ p->h_mnt = au_br_mnt(p->br);
-+ err = vfsub_mnt_want_write(p->h_mnt);
-+ if (unlikely(err)) {
-+ au_fclr_pin(p->flags, MNT_WRITE);
-+ goto out_err;
-+ }
++ d = a->dst_dentry;
++ au_set_h_dptr(d, a->btgt, NULL);
++ err = au_lkup_neg(d, a->btgt, /*wh*/0);
++ if (unlikely(err))
++ goto out_whtmp;
++ a->dst_h_dentry = au_h_dptr(d, a->btgt);
+ }
-+ goto out;
+ }
+
-+ p->h_dentry = NULL;
-+ if (p->bindex <= au_dbbot(p->dentry))
-+ p->h_dentry = au_h_dptr(p->dentry, p->bindex);
++ BUG_ON(d_is_positive(a->dst_h_dentry) && a->src_btop != a->btgt);
++#if 0
++ BUG_ON(!au_ftest_ren(a->auren_flags, DIRREN)
++ && d_is_positive(a->dst_h_dentry)
++ && a->src_btop != a->btgt);
++#endif
+
-+ p->parent = dget_parent(p->dentry);
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_lock(p->parent, AuLock_IR, p->lsc_di);
++ /* rename by vfs_rename or cpup */
++ err = au_ren_or_cpup(a);
++ if (unlikely(err))
++ /* leave the copied-up one */
++ goto out_whtmp;
+
-+ h_dir = NULL;
-+ p->h_parent = au_h_dptr(p->parent, p->bindex);
-+ p->hdir = au_hi(d_inode(p->parent), p->bindex);
-+ if (p->hdir)
-+ h_dir = p->hdir->hi_inode;
++ /* make dir opaque */
++ err = au_ren_diropq(a);
++ if (unlikely(err))
++ goto out_rename;
+
-+ /*
-+ * udba case, or
-+ * if DI_LOCKED is not set, then p->parent may be different
-+ * and h_parent can be NULL.
-+ */
-+ if (unlikely(!p->hdir || !h_dir || !p->h_parent)) {
-+ err = -EBUSY;
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_unlock(p->parent, AuLock_IR);
-+ dput(p->parent);
-+ p->parent = NULL;
-+ goto out_err;
++ /* update target timestamps */
++ if (a->exchange) {
++ AuDebugOn(au_dbtop(a->dst_dentry) != a->btgt);
++ a->h_path.dentry = au_h_dptr(a->dst_dentry, a->btgt);
++ vfsub_update_h_iattr(&a->h_path, /*did*/NULL); /*ignore*/
++ a->dst_inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
+ }
++ AuDebugOn(au_dbtop(a->src_dentry) != a->btgt);
++ a->h_path.dentry = au_h_dptr(a->src_dentry, a->btgt);
++ vfsub_update_h_iattr(&a->h_path, /*did*/NULL); /*ignore*/
++ a->src_inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
+
-+ if (au_ftest_pin(p->flags, MNT_WRITE)) {
-+ p->h_mnt = au_br_mnt(p->br);
-+ err = vfsub_mnt_want_write(p->h_mnt);
-+ if (unlikely(err)) {
-+ au_fclr_pin(p->flags, MNT_WRITE);
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_unlock(p->parent, AuLock_IR);
-+ dput(p->parent);
-+ p->parent = NULL;
-+ goto out_err;
++ if (!a->exchange) {
++ /* remove whiteout for dentry */
++ if (a->dst_wh_dentry) {
++ a->h_path.dentry = a->dst_wh_dentry;
++ err = au_wh_unlink_dentry(a->dst_h_dir, &a->h_path,
++ a->dst_dentry);
++ if (unlikely(err))
++ goto out_diropq;
+ }
-+ }
+
-+ au_igrab(h_dir);
-+ err = au_pin_hdir_lock(p);
-+ if (!err)
-+ goto out; /* success */
++ /* remove whtmp */
++ if (a->thargs)
++ au_ren_del_whtmp(a); /* ignore this error */
+
-+ au_unpin(p);
++ au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0);
++ }
++ err = 0;
++ goto out_success;
+
-+out_err:
-+ pr_err("err %d\n", err);
-+ err = au_busy_or_stale();
++out_diropq:
++ au_ren_rev_diropq(err, a);
++out_rename:
++ au_ren_rev_rename(err, a);
++ dput(a->h_dst);
++out_whtmp:
++ if (a->thargs)
++ au_ren_rev_whtmp(err, a);
++out_whdst:
++ dput(a->dst_wh_dentry);
++ a->dst_wh_dentry = NULL;
++out_whsrc:
++ if (a->src_wh_dentry)
++ au_ren_rev_whsrc(err, a);
++out_success:
++ dput(a->src_wh_dentry);
++ dput(a->dst_wh_dentry);
++out_thargs:
++ if (a->thargs) {
++ dput(a->h_dst);
++ au_whtmp_rmdir_free(a->thargs);
++ a->thargs = NULL;
++ }
+out:
+ return err;
+}
+
-+void au_pin_init(struct au_pin *p, struct dentry *dentry,
-+ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
-+ unsigned int udba, unsigned char flags)
-+{
-+ p->dentry = dentry;
-+ p->udba = udba;
-+ p->lsc_di = lsc_di;
-+ p->lsc_hi = lsc_hi;
-+ p->flags = flags;
-+ p->bindex = bindex;
-+
-+ p->parent = NULL;
-+ p->hdir = NULL;
-+ p->h_mnt = NULL;
-+
-+ p->h_dentry = NULL;
-+ p->h_parent = NULL;
-+ p->br = NULL;
-+ p->task = current;
-+}
++/* ---------------------------------------------------------------------- */
+
-+int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned int udba, unsigned char flags)
++/*
++ * test if @dentry dir can be rename destination or not.
++ * success means, it is a logically empty dir.
++ */
++static int may_rename_dstdir(struct dentry *dentry, struct au_nhash *whlist)
+{
-+ au_pin_init(pin, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2,
-+ udba, flags);
-+ return au_do_pin(pin);
++ return au_test_empty(dentry, whlist);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
+/*
-+ * ->setattr() and ->getattr() are called in various cases.
-+ * chmod, stat: dentry is revalidated.
-+ * fchmod, fstat: file and dentry are not revalidated, additionally they may be
-+ * unhashed.
-+ * for ->setattr(), ia->ia_file is passed from ftruncate only.
++ * test if @a->src_dentry dir can be rename source or not.
++ * if it can, return 0.
++ * success means,
++ * - it is a logically empty dir.
++ * - or, it exists on writable branch and has no children including whiteouts
++ * on the lower branch unless DIRREN is on.
+ */
-+/* todo: consolidate with do_refresh() and simple_reval_dpath() */
-+int au_reval_for_attr(struct dentry *dentry, unsigned int sigen)
++static int may_rename_srcdir(struct au_ren_args *a)
+{
+ int err;
-+ struct dentry *parent;
++ unsigned int rdhash;
++ aufs_bindex_t btop, btgt;
++ struct dentry *dentry;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
+
-+ err = 0;
-+ if (au_digen_test(dentry, sigen)) {
-+ parent = dget_parent(dentry);
-+ di_read_lock_parent(parent, AuLock_IR);
-+ err = au_refresh_dentry(dentry, parent);
-+ di_read_unlock(parent, AuLock_IR);
-+ dput(parent);
++ dentry = a->src_dentry;
++ sb = dentry->d_sb;
++ sbinfo = au_sbi(sb);
++ if (au_opt_test(sbinfo->si_mntflags, DIRREN))
++ au_fset_ren(a->auren_flags, DIRREN);
++
++ btgt = a->btgt;
++ btop = au_dbtop(dentry);
++ if (btop != btgt) {
++ struct au_nhash whlist;
++
++ SiMustAnyLock(sb);
++ rdhash = sbinfo->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL,
++ dentry));
++ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_test_empty(dentry, &whlist);
++ au_nhash_wh_free(&whlist);
++ goto out;
+ }
+
-+ AuTraceErr(err);
++ if (btop == au_dbtaildir(dentry))
++ return 0; /* success */
++
++ err = au_test_empty_lower(dentry);
++
++out:
++ if (err == -ENOTEMPTY) {
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ err = 0;
++ } else {
++ AuWarn1("renaming dir who has child(ren) on multiple "
++ "branches, is not supported\n");
++ err = -EXDEV;
++ }
++ }
+ return err;
+}
+
-+int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
-+ struct au_icpup_args *a)
++/* side effect: sets whlist and h_dentry */
++static int au_ren_may_dir(struct au_ren_args *a)
+{
+ int err;
-+ loff_t sz;
-+ aufs_bindex_t btop, ibtop;
-+ struct dentry *hi_wh, *parent;
-+ struct inode *inode;
-+ struct au_wr_dir_args wr_dir_args = {
-+ .force_btgt = -1,
-+ .flags = 0
-+ };
++ unsigned int rdhash;
++ struct dentry *d;
+
-+ if (d_is_dir(dentry))
-+ au_fset_wrdir(wr_dir_args.flags, ISDIR);
-+ /* plink or hi_wh() case */
-+ btop = au_dbtop(dentry);
-+ inode = d_inode(dentry);
-+ ibtop = au_ibtop(inode);
-+ if (btop != ibtop && !au_test_ro(inode->i_sb, ibtop, inode))
-+ wr_dir_args.force_btgt = ibtop;
-+ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args);
-+ if (unlikely(err < 0))
-+ goto out;
-+ a->btgt = err;
-+ if (err != btop)
-+ au_fset_icpup(a->flags, DID_CPUP);
++ d = a->dst_dentry;
++ SiMustAnyLock(d->d_sb);
+
+ err = 0;
-+ a->pin_flags = AuPin_MNT_WRITE;
-+ parent = NULL;
-+ if (!IS_ROOT(dentry)) {
-+ au_fset_pin(a->pin_flags, DI_LOCKED);
-+ parent = dget_parent(dentry);
-+ di_write_lock_parent(parent);
-+ }
-+
-+ err = au_pin(&a->pin, dentry, a->btgt, a->udba, a->pin_flags);
-+ if (unlikely(err))
-+ goto out_parent;
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST) && a->dst_inode) {
++ rdhash = au_sbi(d->d_sb)->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, d));
++ err = au_nhash_alloc(&a->whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
+
-+ sz = -1;
-+ a->h_path.dentry = au_h_dptr(dentry, btop);
-+ a->h_inode = d_inode(a->h_path.dentry);
-+ if (ia && (ia->ia_valid & ATTR_SIZE)) {
-+ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
-+ if (ia->ia_size < i_size_read(a->h_inode))
-+ sz = ia->ia_size;
-+ inode_unlock(a->h_inode);
++ if (!a->exchange) {
++ au_set_dbtop(d, a->dst_btop);
++ err = may_rename_dstdir(d, &a->whlist);
++ au_set_dbtop(d, a->btgt);
++ } else
++ err = may_rename_srcdir(a);
+ }
++ a->dst_h_dentry = au_h_dptr(d, au_dbtop(d));
++ if (unlikely(err))
++ goto out;
+
-+ hi_wh = NULL;
-+ if (au_ftest_icpup(a->flags, DID_CPUP) && d_unlinked(dentry)) {
-+ hi_wh = au_hi_wh(inode, a->btgt);
-+ if (!hi_wh) {
-+ struct au_cp_generic cpg = {
-+ .dentry = dentry,
-+ .bdst = a->btgt,
-+ .bsrc = -1,
-+ .len = sz,
-+ .pin = &a->pin
-+ };
-+ err = au_sio_cpup_wh(&cpg, /*file*/NULL);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ hi_wh = au_hi_wh(inode, a->btgt);
-+ /* todo: revalidate hi_wh? */
++ d = a->src_dentry;
++ a->src_h_dentry = au_h_dptr(d, au_dbtop(d));
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)) {
++ err = may_rename_srcdir(a);
++ if (unlikely(err)) {
++ au_nhash_wh_free(&a->whlist);
++ a->whlist.nh_num = 0;
+ }
+ }
-+
-+ if (parent) {
-+ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ dput(parent);
-+ parent = NULL;
-+ }
-+ if (!au_ftest_icpup(a->flags, DID_CPUP))
-+ goto out; /* success */
-+
-+ if (!d_unhashed(dentry)) {
-+ struct au_cp_generic cpg = {
-+ .dentry = dentry,
-+ .bdst = a->btgt,
-+ .bsrc = btop,
-+ .len = sz,
-+ .pin = &a->pin,
-+ .flags = AuCpup_DTIME | AuCpup_HOPEN
-+ };
-+ err = au_sio_cpup_simple(&cpg);
-+ if (!err)
-+ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
-+ } else if (!hi_wh)
-+ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
-+ else
-+ a->h_path.dentry = hi_wh; /* do not dget here */
-+
-+out_unlock:
-+ a->h_inode = d_inode(a->h_path.dentry);
-+ if (!err)
-+ goto out; /* success */
-+ au_unpin(&a->pin);
-+out_parent:
-+ if (parent) {
-+ di_write_unlock(parent);
-+ dput(parent);
-+ }
+out:
-+ if (!err)
-+ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
+ return err;
+}
+
-+static int aufs_setattr(struct dentry *dentry, struct iattr *ia)
++/* ---------------------------------------------------------------------- */
++
++/*
++ * simple tests for rename.
++ * following the checks in vfs, plus the parent-child relationship.
++ */
++static int au_may_ren(struct au_ren_args *a)
+{
-+ int err;
-+ struct inode *inode, *delegated;
-+ struct super_block *sb;
-+ struct file *file;
-+ struct au_icpup_args *a;
++ int err, isdir;
++ struct inode *h_inode;
+
-+ inode = d_inode(dentry);
-+ IMustLock(inode);
++ if (a->src_btop == a->btgt) {
++ err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent,
++ au_ftest_ren(a->auren_flags, ISDIR_SRC));
++ if (unlikely(err))
++ goto out;
++ err = -EINVAL;
++ if (unlikely(a->src_h_dentry == a->h_trap))
++ goto out;
++ }
+
-+ err = -ENOMEM;
-+ a = kzalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
++ err = 0;
++ if (a->dst_btop != a->btgt)
+ goto out;
+
-+ if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
-+ ia->ia_valid &= ~ATTR_MODE;
-+
-+ file = NULL;
-+ sb = dentry->d_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ if (unlikely(err))
-+ goto out_kfree;
++ err = -ENOTEMPTY;
++ if (unlikely(a->dst_h_dentry == a->h_trap))
++ goto out;
+
-+ if (ia->ia_valid & ATTR_FILE) {
-+ /* currently ftruncate(2) only */
-+ AuDebugOn(!d_is_reg(dentry));
-+ file = ia->ia_file;
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1);
-+ if (unlikely(err))
-+ goto out_si;
-+ ia->ia_file = au_hf_top(file);
-+ a->udba = AuOpt_UDBA_NONE;
++ err = -EIO;
++ isdir = !!au_ftest_ren(a->auren_flags, ISDIR_DST);
++ if (d_really_is_negative(a->dst_dentry)) {
++ if (d_is_negative(a->dst_h_dentry))
++ err = au_may_add(a->dst_dentry, a->btgt,
++ a->dst_h_parent, isdir);
+ } else {
-+ /* fchmod() doesn't pass ia_file */
-+ a->udba = au_opt_udba(sb);
-+ di_write_lock_child(dentry);
-+ /* no d_unlinked(), to set UDBA_NONE for root */
-+ if (d_unhashed(dentry))
-+ a->udba = AuOpt_UDBA_NONE;
-+ if (a->udba != AuOpt_UDBA_NONE) {
-+ AuDebugOn(IS_ROOT(dentry));
-+ err = au_reval_for_attr(dentry, au_sigen(sb));
-+ if (unlikely(err))
-+ goto out_dentry;
-+ }
++ if (unlikely(d_is_negative(a->dst_h_dentry)))
++ goto out;
++ h_inode = d_inode(a->dst_h_dentry);
++ if (h_inode->i_nlink)
++ err = au_may_del(a->dst_dentry, a->btgt,
++ a->dst_h_parent, isdir);
+ }
+
-+ err = au_pin_and_icpup(dentry, ia, a);
-+ if (unlikely(err < 0))
-+ goto out_dentry;
-+ if (au_ftest_icpup(a->flags, DID_CPUP)) {
-+ ia->ia_file = NULL;
-+ ia->ia_valid &= ~ATTR_FILE;
-+ }
++out:
++ if (unlikely(err == -ENOENT || err == -EEXIST))
++ err = -EIO;
++ AuTraceErr(err);
++ return err;
++}
+
-+ a->h_path.mnt = au_sbr_mnt(sb, a->btgt);
-+ if ((ia->ia_valid & (ATTR_MODE | ATTR_CTIME))
-+ == (ATTR_MODE | ATTR_CTIME)) {
-+ err = security_path_chmod(&a->h_path, ia->ia_mode);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ } else if ((ia->ia_valid & (ATTR_UID | ATTR_GID))
-+ && (ia->ia_valid & ATTR_CTIME)) {
-+ err = security_path_chown(&a->h_path, ia->ia_uid, ia->ia_gid);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ }
++/* ---------------------------------------------------------------------- */
++
++/*
++ * locking order
++ * (VFS)
++ * - src_dir and dir by lock_rename()
++ * - inode if exitsts
++ * (aufs)
++ * - lock all
++ * + src_dentry and dentry by aufs_read_and_write_lock2() which calls,
++ * + si_read_lock
++ * + di_write_lock2_child()
++ * + di_write_lock_child()
++ * + ii_write_lock_child()
++ * + di_write_lock_child2()
++ * + ii_write_lock_child2()
++ * + src_parent and parent
++ * + di_write_lock_parent()
++ * + ii_write_lock_parent()
++ * + di_write_lock_parent2()
++ * + ii_write_lock_parent2()
++ * + lower src_dir and dir by vfsub_lock_rename()
++ * + verify the every relationships between child and parent. if any
++ * of them failed, unlock all and return -EBUSY.
++ */
++static void au_ren_unlock(struct au_ren_args *a)
++{
++ vfsub_unlock_rename(a->src_h_parent, a->src_hdir,
++ a->dst_h_parent, a->dst_hdir);
++ if (au_ftest_ren(a->auren_flags, DIRREN)
++ && a->h_root)
++ au_hn_inode_unlock(a->h_root);
++ if (au_ftest_ren(a->auren_flags, MNT_WRITE))
++ vfsub_mnt_drop_write(au_br_mnt(a->br));
++}
++
++static int au_ren_lock(struct au_ren_args *a)
++{
++ int err;
++ unsigned int udba;
+
-+ if (ia->ia_valid & ATTR_SIZE) {
-+ struct file *f;
++ err = 0;
++ a->src_h_parent = au_h_dptr(a->src_parent, a->btgt);
++ a->src_hdir = au_hi(a->src_dir, a->btgt);
++ a->dst_h_parent = au_h_dptr(a->dst_parent, a->btgt);
++ a->dst_hdir = au_hi(a->dst_dir, a->btgt);
+
-+ if (ia->ia_size < i_size_read(inode))
-+ /* unmap only */
-+ truncate_setsize(inode, ia->ia_size);
++ err = vfsub_mnt_want_write(au_br_mnt(a->br));
++ if (unlikely(err))
++ goto out;
++ au_fset_ren(a->auren_flags, MNT_WRITE);
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ struct dentry *root;
++ struct inode *dir;
+
-+ f = NULL;
-+ if (ia->ia_valid & ATTR_FILE)
-+ f = ia->ia_file;
-+ inode_unlock(a->h_inode);
-+ err = vfsub_trunc(&a->h_path, ia->ia_size, ia->ia_valid, f);
-+ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
-+ } else {
-+ delegated = NULL;
-+ while (1) {
-+ err = vfsub_notify_change(&a->h_path, ia, &delegated);
-+ if (delegated) {
-+ err = break_deleg_wait(&delegated);
-+ if (!err)
-+ continue;
-+ }
-+ break;
++ /*
++ * sbinfo is already locked, so this ii_read_lock is
++ * unnecessary. but our debugging feature checks it.
++ */
++ root = a->src_inode->i_sb->s_root;
++ if (root != a->src_parent && root != a->dst_parent) {
++ dir = d_inode(root);
++ ii_read_lock_parent3(dir);
++ a->h_root = au_hi(dir, a->btgt);
++ ii_read_unlock(dir);
++ au_hn_inode_lock_nested(a->h_root, AuLsc_I_PARENT3);
+ }
+ }
-+ /*
-+ * regardless aufs 'acl' option setting.
-+ * why don't all acl-aware fs call this func from their ->setattr()?
-+ */
-+ if (!err && (ia->ia_valid & ATTR_MODE))
-+ err = vfsub_acl_chmod(a->h_inode, ia->ia_mode);
++ a->h_trap = vfsub_lock_rename(a->src_h_parent, a->src_hdir,
++ a->dst_h_parent, a->dst_hdir);
++ udba = au_opt_udba(a->src_dentry->d_sb);
++ if (unlikely(a->src_hdir->hi_inode != d_inode(a->src_h_parent)
++ || a->dst_hdir->hi_inode != d_inode(a->dst_h_parent)))
++ err = au_busy_or_stale();
++ if (!err && au_dbtop(a->src_dentry) == a->btgt)
++ err = au_h_verify(a->src_h_dentry, udba,
++ d_inode(a->src_h_parent), a->src_h_parent,
++ a->br);
++ if (!err && au_dbtop(a->dst_dentry) == a->btgt)
++ err = au_h_verify(a->dst_h_dentry, udba,
++ d_inode(a->dst_h_parent), a->dst_h_parent,
++ a->br);
+ if (!err)
-+ au_cpup_attr_changeable(inode);
++ goto out; /* success */
++
++ err = au_busy_or_stale();
++ au_ren_unlock(a);
+
-+out_unlock:
-+ inode_unlock(a->h_inode);
-+ au_unpin(&a->pin);
-+ if (unlikely(err))
-+ au_update_dbtop(dentry);
-+out_dentry:
-+ di_write_unlock(dentry);
-+ if (file) {
-+ fi_write_unlock(file);
-+ ia->ia_file = file;
-+ ia->ia_valid |= ATTR_FILE;
-+ }
-+out_si:
-+ si_read_unlock(sb);
-+out_kfree:
-+ au_delayed_kfree(a);
+out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
-+static int au_h_path_to_set_attr(struct dentry *dentry,
-+ struct au_icpup_args *a, struct path *h_path)
++/* ---------------------------------------------------------------------- */
++
++static void au_ren_refresh_dir(struct au_ren_args *a)
+{
-+ int err;
-+ struct super_block *sb;
++ struct inode *dir;
+
-+ sb = dentry->d_sb;
-+ a->udba = au_opt_udba(sb);
-+ /* no d_unlinked(), to set UDBA_NONE for root */
-+ if (d_unhashed(dentry))
-+ a->udba = AuOpt_UDBA_NONE;
-+ if (a->udba != AuOpt_UDBA_NONE) {
-+ AuDebugOn(IS_ROOT(dentry));
-+ err = au_reval_for_attr(dentry, au_sigen(sb));
-+ if (unlikely(err))
-+ goto out;
++ dir = a->dst_dir;
++ dir->i_version++;
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)) {
++ /* is this updating defined in POSIX? */
++ au_cpup_attr_timesizes(a->src_inode);
++ au_cpup_attr_nlink(dir, /*force*/1);
+ }
-+ err = au_pin_and_icpup(dentry, /*ia*/NULL, a);
-+ if (unlikely(err < 0))
-+ goto out;
++ au_dir_ts(dir, a->btgt);
+
-+ h_path->dentry = a->h_path.dentry;
-+ h_path->mnt = au_sbr_mnt(sb, a->btgt);
++ if (a->exchange) {
++ dir = a->src_dir;
++ dir->i_version++;
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST)) {
++ /* is this updating defined in POSIX? */
++ au_cpup_attr_timesizes(a->dst_inode);
++ au_cpup_attr_nlink(dir, /*force*/1);
++ }
++ au_dir_ts(dir, a->btgt);
++ }
+
-+out:
-+ return err;
++ if (au_ftest_ren(a->auren_flags, ISSAMEDIR))
++ return;
++
++ dir = a->src_dir;
++ dir->i_version++;
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC))
++ au_cpup_attr_nlink(dir, /*force*/1);
++ au_dir_ts(dir, a->btgt);
+}
+
-+ssize_t au_srxattr(struct dentry *dentry, struct inode *inode,
-+ struct au_srxattr *arg)
++static void au_ren_refresh(struct au_ren_args *a)
+{
-+ int err;
-+ struct path h_path;
++ aufs_bindex_t bbot, bindex;
++ struct dentry *d, *h_d;
++ struct inode *i, *h_i;
+ struct super_block *sb;
-+ struct au_icpup_args *a;
-+ struct inode *h_inode;
+
-+ IMustLock(inode);
-+
-+ err = -ENOMEM;
-+ a = kzalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++ d = a->dst_dentry;
++ d_drop(d);
++ if (a->h_dst)
++ /* already dget-ed by au_ren_or_cpup() */
++ au_set_h_dptr(d, a->btgt, a->h_dst);
+
-+ sb = dentry->d_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ if (unlikely(err))
-+ goto out_kfree;
++ i = a->dst_inode;
++ if (i) {
++ if (!a->exchange) {
++ if (!au_ftest_ren(a->auren_flags, ISDIR_DST))
++ vfsub_drop_nlink(i);
++ else {
++ vfsub_dead_dir(i);
++ au_cpup_attr_timesizes(i);
++ }
++ au_update_dbrange(d, /*do_put_zero*/1);
++ } else
++ au_cpup_attr_nlink(i, /*force*/1);
++ } else {
++ bbot = a->btgt;
++ for (bindex = au_dbtop(d); bindex < bbot; bindex++)
++ au_set_h_dptr(d, bindex, NULL);
++ bbot = au_dbbot(d);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++)
++ au_set_h_dptr(d, bindex, NULL);
++ au_update_dbrange(d, /*do_put_zero*/0);
++ }
+
-+ h_path.dentry = NULL; /* silence gcc */
-+ di_write_lock_child(dentry);
-+ err = au_h_path_to_set_attr(dentry, a, &h_path);
-+ if (unlikely(err))
-+ goto out_di;
++ if (a->exchange
++ || au_ftest_ren(a->auren_flags, DIRREN)) {
++ d_drop(a->src_dentry);
++ if (au_ftest_ren(a->auren_flags, DIRREN))
++ au_set_dbwh(a->src_dentry, -1);
++ return;
++ }
+
-+ inode_unlock(a->h_inode);
-+ switch (arg->type) {
-+ case AU_XATTR_SET:
-+ AuDebugOn(d_is_negative(h_path.dentry));
-+ err = vfsub_setxattr(h_path.dentry,
-+ arg->u.set.name, arg->u.set.value,
-+ arg->u.set.size, arg->u.set.flags);
-+ break;
-+ case AU_XATTR_REMOVE:
-+ err = vfsub_removexattr(h_path.dentry, arg->u.remove.name);
-+ break;
-+ case AU_ACL_SET:
-+ err = -EOPNOTSUPP;
-+ h_inode = d_inode(h_path.dentry);
-+ if (h_inode->i_op->set_acl)
-+ err = h_inode->i_op->set_acl(h_inode,
-+ arg->u.acl_set.acl,
-+ arg->u.acl_set.type);
-+ break;
++ d = a->src_dentry;
++ au_set_dbwh(d, -1);
++ bbot = au_dbbot(d);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
++ h_d = au_h_dptr(d, bindex);
++ if (h_d)
++ au_set_h_dptr(d, bindex, NULL);
+ }
-+ if (!err)
-+ au_cpup_attr_timesizes(inode);
++ au_set_dbbot(d, a->btgt);
+
-+ au_unpin(&a->pin);
-+ if (unlikely(err))
-+ au_update_dbtop(dentry);
++ sb = d->d_sb;
++ i = a->src_inode;
++ if (au_opt_test(au_mntflags(sb), PLINK) && au_plink_test(i))
++ return; /* success */
+
-+out_di:
-+ di_write_unlock(dentry);
-+ si_read_unlock(sb);
-+out_kfree:
-+ au_delayed_kfree(a);
-+out:
-+ AuTraceErr(err);
-+ return err;
++ bbot = au_ibbot(i);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
++ h_i = au_h_iptr(i, bindex);
++ if (h_i) {
++ au_xino_write(sb, bindex, h_i->i_ino, /*ino*/0);
++ /* ignore this error */
++ au_set_h_iptr(i, bindex, NULL, 0);
++ }
++ }
++ au_set_ibbot(i, a->btgt);
+}
-+#endif
+
-+static void au_refresh_iattr(struct inode *inode, struct kstat *st,
-+ unsigned int nlink)
++/* ---------------------------------------------------------------------- */
++
++/* mainly for link(2) and rename(2) */
++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt)
+{
-+ unsigned int n;
++ aufs_bindex_t bdiropq, bwh;
++ struct dentry *parent;
++ struct au_branch *br;
+
-+ inode->i_mode = st->mode;
-+ /* don't i_[ug]id_write() here */
-+ inode->i_uid = st->uid;
-+ inode->i_gid = st->gid;
-+ inode->i_atime = st->atime;
-+ inode->i_mtime = st->mtime;
-+ inode->i_ctime = st->ctime;
++ parent = dentry->d_parent;
++ IMustLock(d_inode(parent)); /* dir is locked */
+
-+ au_cpup_attr_nlink(inode, /*force*/0);
-+ if (S_ISDIR(inode->i_mode)) {
-+ n = inode->i_nlink;
-+ n -= nlink;
-+ n += st->nlink;
-+ smp_mb(); /* for i_nlink */
-+ /* 0 can happen */
-+ set_nlink(inode, n);
-+ }
++ bdiropq = au_dbdiropq(parent);
++ bwh = au_dbwh(dentry);
++ br = au_sbr(dentry->d_sb, btgt);
++ if (au_br_rdonly(br)
++ || (0 <= bdiropq && bdiropq < btgt)
++ || (0 <= bwh && bwh < btgt))
++ btgt = -1;
+
-+ spin_lock(&inode->i_lock);
-+ inode->i_blocks = st->blocks;
-+ i_size_write(inode, st->size);
-+ spin_unlock(&inode->i_lock);
++ AuDbg("btgt %d\n", btgt);
++ return btgt;
+}
+
-+/*
-+ * common routine for aufs_getattr() and aufs_getxattr().
-+ * returns zero or negative (an error).
-+ * @dentry will be read-locked in success.
-+ */
-+int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path)
++/* sets src_btop, dst_btop and btgt */
++static int au_ren_wbr(struct au_ren_args *a)
+{
+ int err;
-+ unsigned int mnt_flags, sigen;
-+ unsigned char udba_none;
-+ aufs_bindex_t bindex;
-+ struct super_block *sb, *h_sb;
-+ struct inode *inode;
++ struct au_wr_dir_args wr_dir_args = {
++ /* .force_btgt = -1, */
++ .flags = AuWrDir_ADD_ENTRY
++ };
+
-+ h_path->mnt = NULL;
-+ h_path->dentry = NULL;
++ a->src_btop = au_dbtop(a->src_dentry);
++ a->dst_btop = au_dbtop(a->dst_dentry);
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ || au_ftest_ren(a->auren_flags, ISDIR_DST))
++ au_fset_wrdir(wr_dir_args.flags, ISDIR);
++ wr_dir_args.force_btgt = a->src_btop;
++ if (a->dst_inode && a->dst_btop < a->src_btop)
++ wr_dir_args.force_btgt = a->dst_btop;
++ wr_dir_args.force_btgt = au_wbr(a->dst_dentry, wr_dir_args.force_btgt);
++ err = au_wr_dir(a->dst_dentry, a->src_dentry, &wr_dir_args);
++ a->btgt = err;
++ if (a->exchange)
++ au_update_dbtop(a->dst_dentry);
+
-+ err = 0;
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ udba_none = !!au_opt_test(mnt_flags, UDBA_NONE);
++ return err;
++}
+
-+ /* support fstat(2) */
-+ if (!d_unlinked(dentry) && !udba_none) {
-+ sigen = au_sigen(sb);
-+ err = au_digen_test(dentry, sigen);
-+ if (!err) {
-+ di_read_lock_child(dentry, AuLock_IR);
-+ err = au_dbrange_test(dentry);
-+ if (unlikely(err)) {
-+ di_read_unlock(dentry, AuLock_IR);
-+ goto out;
-+ }
-+ } else {
-+ AuDebugOn(IS_ROOT(dentry));
-+ di_write_lock_child(dentry);
-+ err = au_dbrange_test(dentry);
-+ if (!err)
-+ err = au_reval_for_attr(dentry, sigen);
-+ if (!err)
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ else {
-+ di_write_unlock(dentry);
-+ goto out;
-+ }
-+ }
-+ } else
-+ di_read_lock_child(dentry, AuLock_IR);
++static void au_ren_dt(struct au_ren_args *a)
++{
++ a->h_path.dentry = a->src_h_parent;
++ au_dtime_store(a->src_dt + AuPARENT, a->src_parent, &a->h_path);
++ if (!au_ftest_ren(a->auren_flags, ISSAMEDIR)) {
++ a->h_path.dentry = a->dst_h_parent;
++ au_dtime_store(a->dst_dt + AuPARENT, a->dst_parent, &a->h_path);
++ }
+
-+ inode = d_inode(dentry);
-+ bindex = au_ibtop(inode);
-+ h_path->mnt = au_sbr_mnt(sb, bindex);
-+ h_sb = h_path->mnt->mnt_sb;
-+ if (!force
-+ && !au_test_fs_bad_iattr(h_sb)
-+ && udba_none)
-+ goto out; /* success */
++ au_fclr_ren(a->auren_flags, DT_DSTDIR);
++ if (!au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ && !a->exchange)
++ return;
+
-+ if (au_dbtop(dentry) == bindex)
-+ h_path->dentry = au_h_dptr(dentry, bindex);
-+ else if (au_opt_test(mnt_flags, PLINK) && au_plink_test(inode)) {
-+ h_path->dentry = au_plink_lkup(inode, bindex);
-+ if (IS_ERR(h_path->dentry))
-+ /* pretending success */
-+ h_path->dentry = NULL;
-+ else
-+ dput(h_path->dentry);
++ a->h_path.dentry = a->src_h_dentry;
++ au_dtime_store(a->src_dt + AuCHILD, a->src_dentry, &a->h_path);
++ if (d_is_positive(a->dst_h_dentry)) {
++ au_fset_ren(a->auren_flags, DT_DSTDIR);
++ a->h_path.dentry = a->dst_h_dentry;
++ au_dtime_store(a->dst_dt + AuCHILD, a->dst_dentry, &a->h_path);
+ }
-+
-+out:
-+ return err;
+}
+
-+static int aufs_getattr(struct vfsmount *mnt __maybe_unused,
-+ struct dentry *dentry, struct kstat *st)
++static void au_ren_rev_dt(int err, struct au_ren_args *a)
+{
-+ int err;
-+ unsigned char positive;
-+ struct path h_path;
-+ struct inode *inode;
-+ struct super_block *sb;
++ struct dentry *h_d;
++ struct inode *h_inode;
+
-+ inode = d_inode(dentry);
-+ sb = dentry->d_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_h_path_getattr(dentry, /*force*/0, &h_path);
-+ if (unlikely(err))
-+ goto out_si;
-+ if (unlikely(!h_path.dentry))
-+ /* illegally overlapped or something */
-+ goto out_fill; /* pretending success */
++ au_dtime_revert(a->src_dt + AuPARENT);
++ if (!au_ftest_ren(a->auren_flags, ISSAMEDIR))
++ au_dtime_revert(a->dst_dt + AuPARENT);
+
-+ positive = d_is_positive(h_path.dentry);
-+ if (positive)
-+ err = vfs_getattr(&h_path, st);
-+ if (!err) {
-+ if (positive)
-+ au_refresh_iattr(inode, st,
-+ d_inode(h_path.dentry)->i_nlink);
-+ goto out_fill; /* success */
-+ }
-+ AuTraceErr(err);
-+ goto out_di;
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC) && err != -EIO) {
++ h_d = a->src_dt[AuCHILD].dt_h_path.dentry;
++ h_inode = d_inode(h_d);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ au_dtime_revert(a->src_dt + AuCHILD);
++ inode_unlock(h_inode);
+
-+out_fill:
-+ generic_fillattr(inode, st);
-+out_di:
-+ di_read_unlock(dentry, AuLock_IR);
-+out_si:
-+ si_read_unlock(sb);
-+out:
-+ AuTraceErr(err);
-+ return err;
++ if (au_ftest_ren(a->auren_flags, DT_DSTDIR)) {
++ h_d = a->dst_dt[AuCHILD].dt_h_path.dentry;
++ h_inode = d_inode(h_d);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ au_dtime_revert(a->dst_dt + AuCHILD);
++ inode_unlock(h_inode);
++ }
++ }
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static const char *aufs_get_link(struct dentry *dentry, struct inode *inode,
-+ struct delayed_call *done)
++int aufs_rename(struct inode *_src_dir, struct dentry *_src_dentry,
++ struct inode *_dst_dir, struct dentry *_dst_dentry,
++ unsigned int _flags)
+{
-+ const char *ret;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ int err;
-+ aufs_bindex_t bindex;
++ int err, lock_flags;
++ void *rev;
++ /* reduce stack space */
++ struct au_ren_args *a;
++ struct au_pin pin;
+
-+ ret = NULL; /* suppress a warning */
-+ err = -ECHILD;
-+ if (!dentry)
-+ goto out;
++ AuDbg("%pd, %pd, 0x%x\n", _src_dentry, _dst_dentry, _flags);
++ IMustLock(_src_dir);
++ IMustLock(_dst_dir);
+
-+ err = aufs_read_lock(dentry, AuLock_IR | AuLock_GEN);
-+ if (unlikely(err))
++ err = -EINVAL;
++ if (unlikely(_flags & RENAME_WHITEOUT))
+ goto out;
+
-+ err = au_d_hashed_positive(dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
++ err = -ENOMEM;
++ BUILD_BUG_ON(sizeof(*a) > PAGE_SIZE);
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
+
-+ err = -EINVAL;
-+ inode = d_inode(dentry);
-+ bindex = au_ibtop(inode);
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (unlikely(!h_inode->i_op->get_link))
-+ goto out_unlock;
++ a->flags = _flags;
++ a->exchange = _flags & RENAME_EXCHANGE;
++ a->src_dir = _src_dir;
++ a->src_dentry = _src_dentry;
++ a->src_inode = NULL;
++ if (d_really_is_positive(a->src_dentry))
++ a->src_inode = d_inode(a->src_dentry);
++ a->src_parent = a->src_dentry->d_parent; /* dir inode is locked */
++ a->dst_dir = _dst_dir;
++ a->dst_dentry = _dst_dentry;
++ a->dst_inode = NULL;
++ if (d_really_is_positive(a->dst_dentry))
++ a->dst_inode = d_inode(a->dst_dentry);
++ a->dst_parent = a->dst_dentry->d_parent; /* dir inode is locked */
++ if (a->dst_inode) {
++ /*
++ * if EXCHANGE && src is non-dir && dst is dir,
++ * dst is not locked.
++ */
++ /* IMustLock(a->dst_inode); */
++ au_igrab(a->dst_inode);
++ }
+
-+ err = -EBUSY;
-+ h_dentry = NULL;
-+ if (au_dbtop(dentry) <= bindex) {
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (h_dentry)
-+ dget(h_dentry);
++ err = -ENOTDIR;
++ lock_flags = AuLock_FLUSH | AuLock_NOPLM | AuLock_GEN;
++ if (d_is_dir(a->src_dentry)) {
++ au_fset_ren(a->auren_flags, ISDIR_SRC);
++ if (unlikely(!a->exchange
++ && d_really_is_positive(a->dst_dentry)
++ && !d_is_dir(a->dst_dentry)))
++ goto out_free;
++ lock_flags |= AuLock_DIRS;
+ }
-+ if (!h_dentry) {
-+ h_dentry = d_find_any_alias(h_inode);
-+ if (IS_ERR(h_dentry)) {
-+ err = PTR_ERR(h_dentry);
-+ goto out_unlock;
-+ }
++ if (a->dst_inode && d_is_dir(a->dst_dentry)) {
++ au_fset_ren(a->auren_flags, ISDIR_DST);
++ if (unlikely(!a->exchange
++ && d_really_is_positive(a->src_dentry)
++ && !d_is_dir(a->src_dentry)))
++ goto out_free;
++ lock_flags |= AuLock_DIRS;
+ }
-+ if (unlikely(!h_dentry))
-+ goto out_unlock;
-+
-+ err = 0;
-+ AuDbg("%pf\n", h_inode->i_op->get_link);
-+ AuDbgDentry(h_dentry);
-+ ret = h_inode->i_op->get_link(h_dentry, h_inode, done);
-+ dput(h_dentry);
-+ if (IS_ERR(ret))
-+ err = PTR_ERR(ret);
-+
-+out_unlock:
-+ aufs_read_unlock(dentry, AuLock_IR);
-+out:
++ err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry,
++ lock_flags);
+ if (unlikely(err))
-+ ret = ERR_PTR(err);
-+ AuTraceErrPtr(ret);
-+ return ret;
-+}
++ goto out_free;
+
-+/* ---------------------------------------------------------------------- */
++ err = au_d_hashed_positive(a->src_dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ err = -ENOENT;
++ if (a->dst_inode) {
++ /*
++ * If it is a dir, VFS unhash it before this
++ * function. It means we cannot rely upon d_unhashed().
++ */
++ if (unlikely(!a->dst_inode->i_nlink))
++ goto out_unlock;
++ if (!au_ftest_ren(a->auren_flags, ISDIR_DST)) {
++ err = au_d_hashed_positive(a->dst_dentry);
++ if (unlikely(err && !a->exchange))
++ goto out_unlock;
++ } else if (unlikely(IS_DEADDIR(a->dst_inode)))
++ goto out_unlock;
++ } else if (unlikely(d_unhashed(a->dst_dentry)))
++ goto out_unlock;
+
-+static int aufs_update_time(struct inode *inode, struct timespec *ts, int flags)
-+{
-+ int err;
-+ struct super_block *sb;
-+ struct inode *h_inode;
++ /*
++ * is it possible?
++ * yes, it happened (in linux-3.3-rcN) but I don't know why.
++ * there may exist a problem somewhere else.
++ */
++ err = -EINVAL;
++ if (unlikely(d_inode(a->dst_parent) == d_inode(a->src_dentry)))
++ goto out_unlock;
+
-+ sb = inode->i_sb;
-+ /* mmap_sem might be acquired already, cf. aufs_mmap() */
-+ lockdep_off();
-+ si_read_lock(sb, AuLock_FLUSH);
-+ ii_write_lock_child(inode);
-+ lockdep_on();
-+ h_inode = au_h_iptr(inode, au_ibtop(inode));
-+ err = vfsub_update_time(h_inode, ts, flags);
-+ lockdep_off();
-+ if (!err)
-+ au_cpup_attr_timesizes(inode);
-+ ii_write_unlock(inode);
-+ si_read_unlock(sb);
-+ lockdep_on();
++ au_fset_ren(a->auren_flags, ISSAMEDIR); /* temporary */
++ di_write_lock_parent(a->dst_parent);
+
-+ if (!err && (flags & S_VERSION))
-+ inode_inc_iversion(inode);
++ /* which branch we process */
++ err = au_ren_wbr(a);
++ if (unlikely(err < 0))
++ goto out_parent;
++ a->br = au_sbr(a->dst_dentry->d_sb, a->btgt);
++ a->h_path.mnt = au_br_mnt(a->br);
+
-+ return err;
-+}
++ /* are they available to be renamed */
++ err = au_ren_may_dir(a);
++ if (unlikely(err))
++ goto out_children;
+
-+/* ---------------------------------------------------------------------- */
++ /* prepare the writable parent dir on the same branch */
++ if (a->dst_btop == a->btgt) {
++ au_fset_ren(a->auren_flags, WHDST);
++ } else {
++ err = au_cpup_dirs(a->dst_dentry, a->btgt);
++ if (unlikely(err))
++ goto out_children;
++ }
+
-+/* no getattr version will be set by module.c:aufs_init() */
-+struct inode_operations aufs_iop_nogetattr[AuIop_Last],
-+ aufs_iop[] = {
-+ [AuIop_SYMLINK] = {
-+ .permission = aufs_permission,
-+#ifdef CONFIG_FS_POSIX_ACL
-+ .get_acl = aufs_get_acl,
-+ .set_acl = aufs_set_acl, /* unsupport for symlink? */
-+#endif
++ err = 0;
++ if (!a->exchange) {
++ if (a->src_dir != a->dst_dir) {
++ /*
++ * this temporary unlock is safe,
++ * because both dir->i_mutex are locked.
++ */
++ di_write_unlock(a->dst_parent);
++ di_write_lock_parent(a->src_parent);
++ err = au_wr_dir_need_wh(a->src_dentry,
++ au_ftest_ren(a->auren_flags,
++ ISDIR_SRC),
++ &a->btgt);
++ di_write_unlock(a->src_parent);
++ di_write_lock2_parent(a->src_parent, a->dst_parent,
++ /*isdir*/1);
++ au_fclr_ren(a->auren_flags, ISSAMEDIR);
++ } else
++ err = au_wr_dir_need_wh(a->src_dentry,
++ au_ftest_ren(a->auren_flags,
++ ISDIR_SRC),
++ &a->btgt);
++ }
++ if (unlikely(err < 0))
++ goto out_children;
++ if (err)
++ au_fset_ren(a->auren_flags, WHSRC);
+
-+ .setattr = aufs_setattr,
-+ .getattr = aufs_getattr,
++ /* cpup src */
++ if (a->src_btop != a->btgt) {
++ err = au_pin(&pin, a->src_dentry, a->btgt,
++ au_opt_udba(a->src_dentry->d_sb),
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (!err) {
++ struct au_cp_generic cpg = {
++ .dentry = a->src_dentry,
++ .bdst = a->btgt,
++ .bsrc = a->src_btop,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ AuDebugOn(au_dbtop(a->src_dentry) != a->src_btop);
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ }
++ if (unlikely(err))
++ goto out_children;
++ a->src_btop = a->btgt;
++ a->src_h_dentry = au_h_dptr(a->src_dentry, a->btgt);
++ if (!a->exchange)
++ au_fset_ren(a->auren_flags, WHSRC);
++ }
+
-+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
-+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
-+#endif
++ /* cpup dst */
++ if (a->exchange && a->dst_inode
++ && a->dst_btop != a->btgt) {
++ err = au_pin(&pin, a->dst_dentry, a->btgt,
++ au_opt_udba(a->dst_dentry->d_sb),
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (!err) {
++ struct au_cp_generic cpg = {
++ .dentry = a->dst_dentry,
++ .bdst = a->btgt,
++ .bsrc = a->dst_btop,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ }
++ if (unlikely(err))
++ goto out_children;
++ a->dst_btop = a->btgt;
++ a->dst_h_dentry = au_h_dptr(a->dst_dentry, a->btgt);
++ }
+
-+ .readlink = generic_readlink,
-+ .get_link = aufs_get_link,
++ /* lock them all */
++ err = au_ren_lock(a);
++ if (unlikely(err))
++ /* leave the copied-up one */
++ goto out_children;
+
-+ /* .update_time = aufs_update_time */
-+ },
-+ [AuIop_DIR] = {
-+ .create = aufs_create,
-+ .lookup = aufs_lookup,
-+ .link = aufs_link,
-+ .unlink = aufs_unlink,
-+ .symlink = aufs_symlink,
-+ .mkdir = aufs_mkdir,
-+ .rmdir = aufs_rmdir,
-+ .mknod = aufs_mknod,
-+ .rename = aufs_rename,
++ if (!a->exchange) {
++ if (!au_opt_test(au_mntflags(a->dst_dir->i_sb), UDBA_NONE))
++ err = au_may_ren(a);
++ else if (unlikely(a->dst_dentry->d_name.len > AUFS_MAX_NAMELEN))
++ err = -ENAMETOOLONG;
++ if (unlikely(err))
++ goto out_hdir;
++ }
+
-+ .permission = aufs_permission,
-+#ifdef CONFIG_FS_POSIX_ACL
-+ .get_acl = aufs_get_acl,
-+ .set_acl = aufs_set_acl,
-+#endif
++ /* store timestamps to be revertible */
++ au_ren_dt(a);
+
-+ .setattr = aufs_setattr,
-+ .getattr = aufs_getattr,
++ /* store dirren info */
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ err = au_dr_rename(a->src_dentry, a->btgt,
++ &a->dst_dentry->d_name, &rev);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out_dt;
++ }
+
-+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
-+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
-+#endif
++ /* here we go */
++ err = do_rename(a);
++ if (unlikely(err))
++ goto out_dirren;
+
-+ .update_time = aufs_update_time,
-+ .atomic_open = aufs_atomic_open,
-+ .tmpfile = aufs_tmpfile
-+ },
-+ [AuIop_OTHER] = {
-+ .permission = aufs_permission,
-+#ifdef CONFIG_FS_POSIX_ACL
-+ .get_acl = aufs_get_acl,
-+ .set_acl = aufs_set_acl,
-+#endif
++ if (au_ftest_ren(a->auren_flags, DIRREN))
++ au_dr_rename_fin(a->src_dentry, a->btgt, rev);
+
-+ .setattr = aufs_setattr,
-+ .getattr = aufs_getattr,
++ /* update dir attributes */
++ au_ren_refresh_dir(a);
+
-+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
-+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
-+#endif
++ /* dput/iput all lower dentries */
++ au_ren_refresh(a);
+
-+ .update_time = aufs_update_time
++ goto out_hdir; /* success */
++
++out_dirren:
++ if (au_ftest_ren(a->auren_flags, DIRREN))
++ au_dr_rename_rev(a->src_dentry, a->btgt, rev);
++out_dt:
++ au_ren_rev_dt(err, a);
++out_hdir:
++ au_ren_unlock(a);
++out_children:
++ au_nhash_wh_free(&a->whlist);
++ if (err && a->dst_inode && a->dst_btop != a->btgt) {
++ AuDbg("btop %d, btgt %d\n", a->dst_btop, a->btgt);
++ au_set_h_dptr(a->dst_dentry, a->btgt, NULL);
++ au_set_dbtop(a->dst_dentry, a->dst_btop);
+ }
-+};
-diff -urN /usr/share/empty/fs/aufs/i_op_del.c linux/fs/aufs/i_op_del.c
---- /usr/share/empty/fs/aufs/i_op_del.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/i_op_del.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,511 @@
++out_parent:
++ if (!err) {
++ if (d_unhashed(a->src_dentry))
++ au_fset_ren(a->auren_flags, DROPPED_SRC);
++ if (d_unhashed(a->dst_dentry))
++ au_fset_ren(a->auren_flags, DROPPED_DST);
++ if (!a->exchange)
++ d_move(a->src_dentry, a->dst_dentry);
++ else {
++ d_exchange(a->src_dentry, a->dst_dentry);
++ if (au_ftest_ren(a->auren_flags, DROPPED_DST))
++ d_drop(a->dst_dentry);
++ }
++ if (au_ftest_ren(a->auren_flags, DROPPED_SRC))
++ d_drop(a->src_dentry);
++ } else {
++ au_update_dbtop(a->dst_dentry);
++ if (!a->dst_inode)
++ d_drop(a->dst_dentry);
++ }
++ if (au_ftest_ren(a->auren_flags, ISSAMEDIR))
++ di_write_unlock(a->dst_parent);
++ else
++ di_write_unlock2(a->src_parent, a->dst_parent);
++out_unlock:
++ aufs_read_and_write_unlock2(a->dst_dentry, a->src_dentry);
++out_free:
++ iput(a->dst_inode);
++ if (a->thargs)
++ au_whtmp_rmdir_free(a->thargs);
++ kfree(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/iinfo.c linux-4.14/fs/aufs/iinfo.c
+--- linux-4.14/fs/aufs/iinfo.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/iinfo.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,285 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ */
+
+/*
-+ * inode operations (del entry)
++ * inode private data
+ */
+
+#include "aufs.h"
+
-+/*
-+ * decide if a new whiteout for @dentry is necessary or not.
-+ * when it is necessary, prepare the parent dir for the upper branch whose
-+ * branch index is @bcpup for creation. the actual creation of the whiteout will
-+ * be done by caller.
-+ * return value:
-+ * 0: wh is unnecessary
-+ * plus: wh is necessary
-+ * minus: error
-+ */
-+int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup)
++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex)
+{
-+ int need_wh, err;
-+ aufs_bindex_t btop;
-+ struct super_block *sb;
++ struct inode *h_inode;
++ struct au_hinode *hinode;
+
-+ sb = dentry->d_sb;
-+ btop = au_dbtop(dentry);
-+ if (*bcpup < 0) {
-+ *bcpup = btop;
-+ if (au_test_ro(sb, btop, d_inode(dentry))) {
-+ err = AuWbrCopyup(au_sbi(sb), dentry);
-+ *bcpup = err;
-+ if (unlikely(err < 0))
-+ goto out;
-+ }
-+ } else
-+ AuDebugOn(btop < *bcpup
-+ || au_test_ro(sb, *bcpup, d_inode(dentry)));
-+ AuDbg("bcpup %d, btop %d\n", *bcpup, btop);
++ IiMustAnyLock(inode);
+
-+ if (*bcpup != btop) {
-+ err = au_cpup_dirs(dentry, *bcpup);
-+ if (unlikely(err))
-+ goto out;
-+ need_wh = 1;
-+ } else {
-+ struct au_dinfo *dinfo, *tmp;
++ hinode = au_hinode(au_ii(inode), bindex);
++ h_inode = hinode->hi_inode;
++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
++ return h_inode;
++}
+
-+ need_wh = -ENOMEM;
-+ dinfo = au_di(dentry);
-+ tmp = au_di_alloc(sb, AuLsc_DI_TMP);
-+ if (tmp) {
-+ au_di_cp(tmp, dinfo);
-+ au_di_swap(tmp, dinfo);
-+ /* returns the number of positive dentries */
-+ need_wh = au_lkup_dentry(dentry, btop + 1,
-+ /* AuLkup_IGNORE_PERM */ 0);
-+ au_di_swap(tmp, dinfo);
-+ au_rw_write_unlock(&tmp->di_rwsem);
-+ au_di_free(tmp);
-+ }
-+ }
-+ AuDbg("need_wh %d\n", need_wh);
-+ err = need_wh;
++/* todo: hard/soft set? */
++void au_hiput(struct au_hinode *hinode)
++{
++ au_hn_free(hinode);
++ dput(hinode->hi_whdentry);
++ iput(hinode->hi_inode);
++}
+
-+out:
-+ return err;
++unsigned int au_hi_flags(struct inode *inode, int isdir)
++{
++ unsigned int flags;
++ const unsigned int mnt_flags = au_mntflags(inode->i_sb);
++
++ flags = 0;
++ if (au_opt_test(mnt_flags, XINO))
++ au_fset_hi(flags, XINO);
++ if (isdir && au_opt_test(mnt_flags, UDBA_HNOTIFY))
++ au_fset_hi(flags, HNOTIFY);
++ return flags;
+}
+
-+/*
-+ * simple tests for the del-entry operations.
-+ * following the checks in vfs, plus the parent-child relationship.
-+ */
-+int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir)
++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode, unsigned int flags)
+{
-+ int err;
-+ umode_t h_mode;
-+ struct dentry *h_dentry, *h_latest;
-+ struct inode *h_inode;
++ struct au_hinode *hinode;
++ struct inode *hi;
++ struct au_iinfo *iinfo = au_ii(inode);
+
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (d_really_is_positive(dentry)) {
-+ err = -ENOENT;
-+ if (unlikely(d_is_negative(h_dentry)))
-+ goto out;
-+ h_inode = d_inode(h_dentry);
-+ if (unlikely(!h_inode->i_nlink))
-+ goto out;
++ IiMustWriteLock(inode);
+
-+ h_mode = h_inode->i_mode;
-+ if (!isdir) {
-+ err = -EISDIR;
-+ if (unlikely(S_ISDIR(h_mode)))
-+ goto out;
-+ } else if (unlikely(!S_ISDIR(h_mode))) {
-+ err = -ENOTDIR;
-+ goto out;
++ hinode = au_hinode(iinfo, bindex);
++ hi = hinode->hi_inode;
++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
++
++ if (hi)
++ au_hiput(hinode);
++ hinode->hi_inode = h_inode;
++ if (h_inode) {
++ int err;
++ struct super_block *sb = inode->i_sb;
++ struct au_branch *br;
++
++ AuDebugOn(inode->i_mode
++ && (h_inode->i_mode & S_IFMT)
++ != (inode->i_mode & S_IFMT));
++ if (bindex == iinfo->ii_btop)
++ au_cpup_igen(inode, h_inode);
++ br = au_sbr(sb, bindex);
++ hinode->hi_id = br->br_id;
++ if (au_ftest_hi(flags, XINO)) {
++ err = au_xino_write(sb, bindex, h_inode->i_ino,
++ inode->i_ino);
++ if (unlikely(err))
++ AuIOErr1("failed au_xino_write() %d\n", err);
++ }
++
++ if (au_ftest_hi(flags, HNOTIFY)
++ && au_br_hnotifyable(br->br_perm)) {
++ err = au_hn_alloc(hinode, inode);
++ if (unlikely(err))
++ AuIOErr1("au_hn_alloc() %d\n", err);
+ }
-+ } else {
-+ /* rename(2) case */
-+ err = -EIO;
-+ if (unlikely(d_is_positive(h_dentry)))
-+ goto out;
+ }
++}
+
-+ err = -ENOENT;
-+ /* expected parent dir is locked */
-+ if (unlikely(h_parent != h_dentry->d_parent))
-+ goto out;
-+ err = 0;
++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_wh)
++{
++ struct au_hinode *hinode;
+
-+ /*
-+ * rmdir a dir may break the consistency on some filesystem.
-+ * let's try heavy test.
-+ */
-+ err = -EACCES;
-+ if (unlikely(!au_opt_test(au_mntflags(dentry->d_sb), DIRPERM1)
-+ && au_test_h_perm(d_inode(h_parent),
-+ MAY_EXEC | MAY_WRITE)))
-+ goto out;
++ IiMustWriteLock(inode);
+
-+ h_latest = au_sio_lkup_one(&dentry->d_name, h_parent);
-+ err = -EIO;
-+ if (IS_ERR(h_latest))
-+ goto out;
-+ if (h_latest == h_dentry)
-+ err = 0;
-+ dput(h_latest);
++ hinode = au_hinode(au_ii(inode), bindex);
++ AuDebugOn(hinode->hi_whdentry);
++ hinode->hi_whdentry = h_wh;
++}
+
-+out:
-+ return err;
++void au_update_iigen(struct inode *inode, int half)
++{
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
++ unsigned int sigen;
++
++ sigen = au_sigen(inode->i_sb);
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ iigen->ig_generation = sigen;
++ if (half)
++ au_ig_fset(iigen->ig_flags, HALF_REFRESHED);
++ else
++ au_ig_fclr(iigen->ig_flags, HALF_REFRESHED);
++ spin_unlock(&iigen->ig_spin);
+}
+
-+/*
-+ * decide the branch where we operate for @dentry. the branch index will be set
-+ * @rbcpup. after diciding it, 'pin' it and store the timestamps of the parent
-+ * dir for reverting.
-+ * when a new whiteout is necessary, create it.
-+ */
-+static struct dentry*
-+lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup,
-+ struct au_dtime *dt, struct au_pin *pin)
++/* it may be called at remount time, too */
++void au_update_ibrange(struct inode *inode, int do_put_zero)
+{
-+ struct dentry *wh_dentry;
-+ struct super_block *sb;
-+ struct path h_path;
-+ int err, need_wh;
-+ unsigned int udba;
-+ aufs_bindex_t bcpup;
++ struct au_iinfo *iinfo;
++ aufs_bindex_t bindex, bbot;
+
-+ need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup);
-+ wh_dentry = ERR_PTR(need_wh);
-+ if (unlikely(need_wh < 0))
-+ goto out;
++ AuDebugOn(au_is_bad_inode(inode));
++ IiMustWriteLock(inode);
+
-+ sb = dentry->d_sb;
-+ udba = au_opt_udba(sb);
-+ bcpup = *rbcpup;
-+ err = au_pin(pin, dentry, bcpup, udba,
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out;
++ iinfo = au_ii(inode);
++ if (do_put_zero && iinfo->ii_btop >= 0) {
++ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
++ bindex++) {
++ struct inode *h_i;
+
-+ h_path.dentry = au_pinned_h_parent(pin);
-+ if (udba != AuOpt_UDBA_NONE
-+ && au_dbtop(dentry) == bcpup) {
-+ err = au_may_del(dentry, bcpup, h_path.dentry, isdir);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_unpin;
++ h_i = au_hinode(iinfo, bindex)->hi_inode;
++ if (h_i
++ && !h_i->i_nlink
++ && !(h_i->i_state & I_LINKABLE))
++ au_set_h_iptr(inode, bindex, NULL, 0);
++ }
+ }
+
-+ h_path.mnt = au_sbr_mnt(sb, bcpup);
-+ au_dtime_store(dt, au_pinned_parent(pin), &h_path);
-+ wh_dentry = NULL;
-+ if (!need_wh)
-+ goto out; /* success, no need to create whiteout */
++ iinfo->ii_btop = -1;
++ iinfo->ii_bbot = -1;
++ bbot = au_sbbot(inode->i_sb);
++ for (bindex = 0; bindex <= bbot; bindex++)
++ if (au_hinode(iinfo, bindex)->hi_inode) {
++ iinfo->ii_btop = bindex;
++ break;
++ }
++ if (iinfo->ii_btop >= 0)
++ for (bindex = bbot; bindex >= iinfo->ii_btop; bindex--)
++ if (au_hinode(iinfo, bindex)->hi_inode) {
++ iinfo->ii_bbot = bindex;
++ break;
++ }
++ AuDebugOn(iinfo->ii_btop > iinfo->ii_bbot);
++}
+
-+ wh_dentry = au_wh_create(dentry, bcpup, h_path.dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_unpin;
++/* ---------------------------------------------------------------------- */
+
-+ /* returns with the parent is locked and wh_dentry is dget-ed */
-+ goto out; /* success */
++void au_icntnr_init_once(void *_c)
++{
++ struct au_icntnr *c = _c;
++ struct au_iinfo *iinfo = &c->iinfo;
+
-+out_unpin:
-+ au_unpin(pin);
-+out:
-+ return wh_dentry;
++ spin_lock_init(&iinfo->ii_generation.ig_spin);
++ au_rw_init(&iinfo->ii_rwsem);
++ inode_init_once(&c->vfs_inode);
+}
+
-+/*
-+ * when removing a dir, rename it to a unique temporary whiteout-ed name first
-+ * in order to be revertible and save time for removing many child whiteouts
-+ * under the dir.
-+ * returns 1 when there are too many child whiteout and caller should remove
-+ * them asynchronously. returns 0 when the number of children is enough small to
-+ * remove now or the branch fs is a remote fs.
-+ * otherwise return an error.
-+ */
-+static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct au_nhash *whlist, struct inode *dir)
++void au_hinode_init(struct au_hinode *hinode)
+{
-+ int rmdir_later, err, dirwh;
-+ struct dentry *h_dentry;
++ hinode->hi_inode = NULL;
++ hinode->hi_id = -1;
++ au_hn_init(hinode);
++ hinode->hi_whdentry = NULL;
++}
++
++int au_iinfo_init(struct inode *inode)
++{
++ struct au_iinfo *iinfo;
+ struct super_block *sb;
-+ struct inode *inode;
++ struct au_hinode *hi;
++ int nbr, i;
+
-+ sb = dentry->d_sb;
-+ SiMustAnyLock(sb);
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ err = au_whtmp_ren(h_dentry, au_sbr(sb, bindex));
-+ if (unlikely(err))
-+ goto out;
++ sb = inode->i_sb;
++ iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
++ nbr = au_sbbot(sb) + 1;
++ if (unlikely(nbr <= 0))
++ nbr = 1;
++ hi = kmalloc_array(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS);
++ if (hi) {
++ au_ninodes_inc(sb);
+
-+ /* stop monitoring */
-+ inode = d_inode(dentry);
-+ au_hn_free(au_hi(inode, bindex));
++ iinfo->ii_hinode = hi;
++ for (i = 0; i < nbr; i++, hi++)
++ au_hinode_init(hi);
+
-+ if (!au_test_fs_remote(h_dentry->d_sb)) {
-+ dirwh = au_sbi(sb)->si_dirwh;
-+ rmdir_later = (dirwh <= 1);
-+ if (!rmdir_later)
-+ rmdir_later = au_nhash_test_longer_wh(whlist, bindex,
-+ dirwh);
-+ if (rmdir_later)
-+ return rmdir_later;
++ iinfo->ii_generation.ig_generation = au_sigen(sb);
++ iinfo->ii_btop = -1;
++ iinfo->ii_bbot = -1;
++ iinfo->ii_vdir = NULL;
++ return 0;
+ }
++ return -ENOMEM;
++}
+
-+ err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist);
-+ if (unlikely(err)) {
-+ AuIOErr("rmdir %pd, b%d failed, %d. ignored\n",
-+ h_dentry, bindex, err);
++int au_hinode_realloc(struct au_iinfo *iinfo, int nbr, int may_shrink)
++{
++ int err, i;
++ struct au_hinode *hip;
++
++ AuRwMustWriteLock(&iinfo->ii_rwsem);
++
++ err = -ENOMEM;
++ hip = au_krealloc(iinfo->ii_hinode, sizeof(*hip) * nbr, GFP_NOFS,
++ may_shrink);
++ if (hip) {
++ iinfo->ii_hinode = hip;
++ i = iinfo->ii_bbot + 1;
++ hip += i;
++ for (; i < nbr; i++, hip++)
++ au_hinode_init(hip);
+ err = 0;
+ }
+
-+out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+/*
-+ * final procedure for deleting a entry.
-+ * maintain dentry and iattr.
-+ */
-+static void epilog(struct inode *dir, struct dentry *dentry,
-+ aufs_bindex_t bindex)
++void au_iinfo_fin(struct inode *inode)
+{
-+ struct inode *inode;
++ struct au_iinfo *iinfo;
++ struct au_hinode *hi;
++ struct super_block *sb;
++ aufs_bindex_t bindex, bbot;
++ const unsigned char unlinked = !inode->i_nlink;
+
-+ inode = d_inode(dentry);
-+ d_drop(dentry);
-+ inode->i_ctime = dir->i_ctime;
++ AuDebugOn(au_is_bad_inode(inode));
+
-+ au_dir_ts(dir, bindex);
-+ dir->i_version++;
++ sb = inode->i_sb;
++ au_ninodes_dec(sb);
++ if (si_pid_test(sb))
++ au_xino_delete_inode(inode, unlinked);
++ else {
++ /*
++ * it is safe to hide the dependency between sbinfo and
++ * sb->s_umount.
++ */
++ lockdep_off();
++ si_noflush_read_lock(sb);
++ au_xino_delete_inode(inode, unlinked);
++ si_read_unlock(sb);
++ lockdep_on();
++ }
++
++ iinfo = au_ii(inode);
++ if (iinfo->ii_vdir)
++ au_vdir_free(iinfo->ii_vdir);
++
++ bindex = iinfo->ii_btop;
++ if (bindex >= 0) {
++ hi = au_hinode(iinfo, bindex);
++ bbot = iinfo->ii_bbot;
++ while (bindex++ <= bbot) {
++ if (hi->hi_inode)
++ au_hiput(hi);
++ hi++;
++ }
++ }
++ kfree(iinfo->ii_hinode);
++ AuRwDestroy(&iinfo->ii_rwsem);
+}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/inode.c linux-4.14/fs/aufs/inode.c
+--- linux-4.14/fs/aufs/inode.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/inode.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,527 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
+/*
-+ * when an error happened, remove the created whiteout and revert everything.
++ * inode functions
+ */
-+static int do_revert(int err, struct inode *dir, aufs_bindex_t bindex,
-+ aufs_bindex_t bwh, struct dentry *wh_dentry,
-+ struct dentry *dentry, struct au_dtime *dt)
-+{
-+ int rerr;
-+ struct path h_path = {
-+ .dentry = wh_dentry,
-+ .mnt = au_sbr_mnt(dir->i_sb, bindex)
-+ };
+
-+ rerr = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, dentry);
-+ if (!rerr) {
-+ au_set_dbwh(dentry, bwh);
-+ au_dtime_revert(dt);
-+ return 0;
-+ }
++#include "aufs.h"
+
-+ AuIOErr("%pd reverting whiteout failed(%d, %d)\n", dentry, err, rerr);
-+ return -EIO;
++struct inode *au_igrab(struct inode *inode)
++{
++ if (inode) {
++ AuDebugOn(!atomic_read(&inode->i_count));
++ ihold(inode);
++ }
++ return inode;
+}
+
-+/* ---------------------------------------------------------------------- */
++static void au_refresh_hinode_attr(struct inode *inode, int do_version)
++{
++ au_cpup_attr_all(inode, /*force*/0);
++ au_update_iigen(inode, /*half*/1);
++ if (do_version)
++ inode->i_version++;
++}
+
-+int aufs_unlink(struct inode *dir, struct dentry *dentry)
++static int au_ii_refresh(struct inode *inode, int *update)
+{
-+ int err;
-+ aufs_bindex_t bwh, bindex, btop;
-+ struct inode *inode, *h_dir, *delegated;
-+ struct dentry *parent, *wh_dentry;
-+ /* to reuduce stack size */
-+ struct {
-+ struct au_dtime dt;
-+ struct au_pin pin;
-+ struct path h_path;
-+ } *a;
++ int err, e, nbr;
++ umode_t type;
++ aufs_bindex_t bindex, new_bindex;
++ struct super_block *sb;
++ struct au_iinfo *iinfo;
++ struct au_hinode *p, *q, tmp;
+
-+ IMustLock(dir);
++ AuDebugOn(au_is_bad_inode(inode));
++ IiMustWriteLock(inode);
+
-+ err = -ENOMEM;
-+ a = kmalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
++ *update = 0;
++ sb = inode->i_sb;
++ nbr = au_sbbot(sb) + 1;
++ type = inode->i_mode & S_IFMT;
++ iinfo = au_ii(inode);
++ err = au_hinode_realloc(iinfo, nbr, /*may_shrink*/0);
++ if (unlikely(err))
+ goto out;
+
-+ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
-+ if (unlikely(err))
-+ goto out_free;
-+ err = au_d_hashed_positive(dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ inode = d_inode(dentry);
-+ IMustLock(inode);
-+ err = -EISDIR;
-+ if (unlikely(d_is_dir(dentry)))
-+ goto out_unlock; /* possible? */
++ AuDebugOn(iinfo->ii_btop < 0);
++ p = au_hinode(iinfo, iinfo->ii_btop);
++ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
++ bindex++, p++) {
++ if (!p->hi_inode)
++ continue;
+
-+ btop = au_dbtop(dentry);
-+ bwh = au_dbwh(dentry);
-+ bindex = -1;
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_write_lock_parent(parent);
-+ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &a->dt,
-+ &a->pin);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_parent;
++ AuDebugOn(type != (p->hi_inode->i_mode & S_IFMT));
++ new_bindex = au_br_index(sb, p->hi_id);
++ if (new_bindex == bindex)
++ continue;
+
-+ a->h_path.mnt = au_sbr_mnt(dentry->d_sb, btop);
-+ a->h_path.dentry = au_h_dptr(dentry, btop);
-+ dget(a->h_path.dentry);
-+ if (bindex == btop) {
-+ h_dir = au_pinned_h_dir(&a->pin);
-+ delegated = NULL;
-+ err = vfsub_unlink(h_dir, &a->h_path, &delegated, /*force*/0);
-+ if (unlikely(err == -EWOULDBLOCK)) {
-+ pr_warn("cannot retry for NFSv4 delegation"
-+ " for an internal unlink\n");
-+ iput(delegated);
++ if (new_bindex < 0) {
++ *update = 1;
++ au_hiput(p);
++ p->hi_inode = NULL;
++ continue;
++ }
++
++ if (new_bindex < iinfo->ii_btop)
++ iinfo->ii_btop = new_bindex;
++ if (iinfo->ii_bbot < new_bindex)
++ iinfo->ii_bbot = new_bindex;
++ /* swap two lower inode, and loop again */
++ q = au_hinode(iinfo, new_bindex);
++ tmp = *q;
++ *q = *p;
++ *p = tmp;
++ if (tmp.hi_inode) {
++ bindex--;
++ p--;
+ }
-+ } else {
-+ /* dir inode is locked */
-+ h_dir = d_inode(wh_dentry->d_parent);
-+ IMustLock(h_dir);
-+ err = 0;
+ }
++ au_update_ibrange(inode, /*do_put_zero*/0);
++ au_hinode_realloc(iinfo, nbr, /*may_shrink*/1); /* harmless if err */
++ e = au_dy_irefresh(inode);
++ if (unlikely(e && !err))
++ err = e;
+
-+ if (!err) {
-+ vfsub_drop_nlink(inode);
-+ epilog(dir, dentry, bindex);
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ /* update target timestamps */
-+ if (bindex == btop) {
-+ vfsub_update_h_iattr(&a->h_path, /*did*/NULL);
-+ /*ignore*/
-+ inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
-+ } else
-+ /* todo: this timestamp may be reverted later */
-+ inode->i_ctime = h_dir->i_ctime;
-+ goto out_unpin; /* success */
-+ }
++void au_refresh_iop(struct inode *inode, int force_getattr)
++{
++ int type;
++ struct au_sbinfo *sbi = au_sbi(inode->i_sb);
++ const struct inode_operations *iop
++ = force_getattr ? aufs_iop : sbi->si_iop_array;
+
-+ /* revert */
-+ if (wh_dentry) {
-+ int rerr;
++ if (inode->i_op == iop)
++ return;
+
-+ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
-+ &a->dt);
-+ if (rerr)
-+ err = rerr;
++ switch (inode->i_mode & S_IFMT) {
++ case S_IFDIR:
++ type = AuIop_DIR;
++ break;
++ case S_IFLNK:
++ type = AuIop_SYMLINK;
++ break;
++ default:
++ type = AuIop_OTHER;
++ break;
+ }
+
-+out_unpin:
-+ au_unpin(&a->pin);
-+ dput(wh_dentry);
-+ dput(a->h_path.dentry);
-+out_parent:
-+ di_write_unlock(parent);
-+out_unlock:
-+ aufs_read_unlock(dentry, AuLock_DW);
-+out_free:
-+ au_delayed_kfree(a);
-+out:
-+ return err;
++ inode->i_op = iop + type;
++ /* unnecessary smp_wmb() */
+}
+
-+int aufs_rmdir(struct inode *dir, struct dentry *dentry)
++int au_refresh_hinode_self(struct inode *inode)
+{
-+ int err, rmdir_later;
-+ aufs_bindex_t bwh, bindex, btop;
-+ struct inode *inode;
-+ struct dentry *parent, *wh_dentry, *h_dentry;
-+ struct au_whtmp_rmdir *args;
-+ /* to reuduce stack size */
-+ struct {
-+ struct au_dtime dt;
-+ struct au_pin pin;
-+ } *a;
-+
-+ IMustLock(dir);
++ int err, update;
+
-+ err = -ENOMEM;
-+ a = kmalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++ err = au_ii_refresh(inode, &update);
++ if (!err)
++ au_refresh_hinode_attr(inode, update && S_ISDIR(inode->i_mode));
+
-+ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
-+ if (unlikely(err))
-+ goto out_free;
-+ err = au_alive_dir(dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ inode = d_inode(dentry);
-+ IMustLock(inode);
-+ err = -ENOTDIR;
-+ if (unlikely(!d_is_dir(dentry)))
-+ goto out_unlock; /* possible? */
++ AuTraceErr(err);
++ return err;
++}
+
-+ err = -ENOMEM;
-+ args = au_whtmp_rmdir_alloc(dir->i_sb, GFP_NOFS);
-+ if (unlikely(!args))
-+ goto out_unlock;
++int au_refresh_hinode(struct inode *inode, struct dentry *dentry)
++{
++ int err, e, update;
++ unsigned int flags;
++ umode_t mode;
++ aufs_bindex_t bindex, bbot;
++ unsigned char isdir;
++ struct au_hinode *p;
++ struct au_iinfo *iinfo;
+
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_write_lock_parent(parent);
-+ err = au_test_empty(dentry, &args->whlist);
++ err = au_ii_refresh(inode, &update);
+ if (unlikely(err))
-+ goto out_parent;
++ goto out;
+
-+ btop = au_dbtop(dentry);
-+ bwh = au_dbwh(dentry);
-+ bindex = -1;
-+ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &a->dt,
-+ &a->pin);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_parent;
++ update = 0;
++ iinfo = au_ii(inode);
++ p = au_hinode(iinfo, iinfo->ii_btop);
++ mode = (inode->i_mode & S_IFMT);
++ isdir = S_ISDIR(mode);
++ flags = au_hi_flags(inode, isdir);
++ bbot = au_dbbot(dentry);
++ for (bindex = au_dbtop(dentry); bindex <= bbot; bindex++) {
++ struct inode *h_i, *h_inode;
++ struct dentry *h_d;
+
-+ h_dentry = au_h_dptr(dentry, btop);
-+ dget(h_dentry);
-+ rmdir_later = 0;
-+ if (bindex == btop) {
-+ err = renwh_and_rmdir(dentry, btop, &args->whlist, dir);
-+ if (err > 0) {
-+ rmdir_later = err;
-+ err = 0;
-+ }
-+ } else {
-+ /* stop monitoring */
-+ au_hn_free(au_hi(inode, btop));
++ h_d = au_h_dptr(dentry, bindex);
++ if (!h_d || d_is_negative(h_d))
++ continue;
+
-+ /* dir inode is locked */
-+ IMustLock(d_inode(wh_dentry->d_parent));
-+ err = 0;
++ h_inode = d_inode(h_d);
++ AuDebugOn(mode != (h_inode->i_mode & S_IFMT));
++ if (iinfo->ii_btop <= bindex && bindex <= iinfo->ii_bbot) {
++ h_i = au_h_iptr(inode, bindex);
++ if (h_i) {
++ if (h_i == h_inode)
++ continue;
++ err = -EIO;
++ break;
++ }
++ }
++ if (bindex < iinfo->ii_btop)
++ iinfo->ii_btop = bindex;
++ if (iinfo->ii_bbot < bindex)
++ iinfo->ii_bbot = bindex;
++ au_set_h_iptr(inode, bindex, au_igrab(h_inode), flags);
++ update = 1;
+ }
++ au_update_ibrange(inode, /*do_put_zero*/0);
++ e = au_dy_irefresh(inode);
++ if (unlikely(e && !err))
++ err = e;
++ if (!err)
++ au_refresh_hinode_attr(inode, update && isdir);
+
-+ if (!err) {
-+ vfsub_dead_dir(inode);
-+ au_set_dbdiropq(dentry, -1);
-+ epilog(dir, dentry, bindex);
-+
-+ if (rmdir_later) {
-+ au_whtmp_kick_rmdir(dir, btop, h_dentry, args);
-+ args = NULL;
-+ }
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ goto out_unpin; /* success */
-+ }
++static int set_inode(struct inode *inode, struct dentry *dentry)
++{
++ int err;
++ unsigned int flags;
++ umode_t mode;
++ aufs_bindex_t bindex, btop, btail;
++ unsigned char isdir;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct au_iinfo *iinfo;
++ struct inode_operations *iop;
+
-+ /* revert */
-+ AuLabel(revert);
-+ if (wh_dentry) {
-+ int rerr;
++ IiMustWriteLock(inode);
+
-+ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
-+ &a->dt);
-+ if (rerr)
-+ err = rerr;
++ err = 0;
++ isdir = 0;
++ iop = au_sbi(inode->i_sb)->si_iop_array;
++ btop = au_dbtop(dentry);
++ h_dentry = au_h_dptr(dentry, btop);
++ h_inode = d_inode(h_dentry);
++ mode = h_inode->i_mode;
++ switch (mode & S_IFMT) {
++ case S_IFREG:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_OTHER;
++ inode->i_fop = &aufs_file_fop;
++ err = au_dy_iaop(inode, btop, h_inode);
++ if (unlikely(err))
++ goto out;
++ break;
++ case S_IFDIR:
++ isdir = 1;
++ btail = au_dbtaildir(dentry);
++ inode->i_op = iop + AuIop_DIR;
++ inode->i_fop = &aufs_dir_fop;
++ break;
++ case S_IFLNK:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_SYMLINK;
++ break;
++ case S_IFBLK:
++ case S_IFCHR:
++ case S_IFIFO:
++ case S_IFSOCK:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_OTHER;
++ init_special_inode(inode, mode, h_inode->i_rdev);
++ break;
++ default:
++ AuIOErr("Unknown file type 0%o\n", mode);
++ err = -EIO;
++ goto out;
+ }
+
-+out_unpin:
-+ au_unpin(&a->pin);
-+ dput(wh_dentry);
-+ dput(h_dentry);
-+out_parent:
-+ di_write_unlock(parent);
-+ if (args)
-+ au_whtmp_rmdir_free(args);
-+out_unlock:
-+ aufs_read_unlock(dentry, AuLock_DW);
-+out_free:
-+ au_delayed_kfree(a);
++ /* do not set hnotify for whiteouted dirs (SHWH mode) */
++ flags = au_hi_flags(inode, isdir);
++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)
++ && au_ftest_hi(flags, HNOTIFY)
++ && dentry->d_name.len > AUFS_WH_PFX_LEN
++ && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))
++ au_fclr_hi(flags, HNOTIFY);
++ iinfo = au_ii(inode);
++ iinfo->ii_btop = btop;
++ iinfo->ii_bbot = btail;
++ for (bindex = btop; bindex <= btail; bindex++) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry)
++ au_set_h_iptr(inode, bindex,
++ au_igrab(d_inode(h_dentry)), flags);
++ }
++ au_cpup_attr_all(inode, /*force*/1);
++ /*
++ * to force calling aufs_get_acl() every time,
++ * do not call cache_no_acl() for aufs inode.
++ */
++
+out:
-+ AuTraceErr(err);
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/i_op_ren.c linux/fs/aufs/i_op_ren.c
---- /usr/share/empty/fs/aufs/i_op_ren.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/i_op_ren.c 2016-08-17 18:01:06.148556271 +0200
-@@ -0,0 +1,1015 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
+
+/*
-+ * inode operation (rename entry)
-+ * todo: this is crazy monster
++ * successful returns with iinfo write_locked
++ * minus: errno
++ * zero: success, matched
++ * plus: no error, but unmatched
+ */
++static int reval_inode(struct inode *inode, struct dentry *dentry)
++{
++ int err;
++ unsigned int gen, igflags;
++ aufs_bindex_t bindex, bbot;
++ struct inode *h_inode, *h_dinode;
++ struct dentry *h_dentry;
+
-+#include "aufs.h"
-+
-+enum { AuSRC, AuDST, AuSrcDst };
-+enum { AuPARENT, AuCHILD, AuParentChild };
-+
-+#define AuRen_ISDIR 1
-+#define AuRen_ISSAMEDIR (1 << 1)
-+#define AuRen_WHSRC (1 << 2)
-+#define AuRen_WHDST (1 << 3)
-+#define AuRen_MNT_WRITE (1 << 4)
-+#define AuRen_DT_DSTDIR (1 << 5)
-+#define AuRen_DIROPQ (1 << 6)
-+#define au_ftest_ren(flags, name) ((flags) & AuRen_##name)
-+#define au_fset_ren(flags, name) \
-+ do { (flags) |= AuRen_##name; } while (0)
-+#define au_fclr_ren(flags, name) \
-+ do { (flags) &= ~AuRen_##name; } while (0)
-+
-+struct au_ren_args {
-+ struct {
-+ struct dentry *dentry, *h_dentry, *parent, *h_parent,
-+ *wh_dentry;
-+ struct inode *dir, *inode;
-+ struct au_hinode *hdir;
-+ struct au_dtime dt[AuParentChild];
-+ aufs_bindex_t btop;
-+ } sd[AuSrcDst];
-+
-+#define src_dentry sd[AuSRC].dentry
-+#define src_dir sd[AuSRC].dir
-+#define src_inode sd[AuSRC].inode
-+#define src_h_dentry sd[AuSRC].h_dentry
-+#define src_parent sd[AuSRC].parent
-+#define src_h_parent sd[AuSRC].h_parent
-+#define src_wh_dentry sd[AuSRC].wh_dentry
-+#define src_hdir sd[AuSRC].hdir
-+#define src_h_dir sd[AuSRC].hdir->hi_inode
-+#define src_dt sd[AuSRC].dt
-+#define src_btop sd[AuSRC].btop
++ /*
++ * before this function, if aufs got any iinfo lock, it must be only
++ * one, the parent dir.
++ * it can happen by UDBA and the obsoleted inode number.
++ */
++ err = -EIO;
++ if (unlikely(inode->i_ino == parent_ino(dentry)))
++ goto out;
+
-+#define dst_dentry sd[AuDST].dentry
-+#define dst_dir sd[AuDST].dir
-+#define dst_inode sd[AuDST].inode
-+#define dst_h_dentry sd[AuDST].h_dentry
-+#define dst_parent sd[AuDST].parent
-+#define dst_h_parent sd[AuDST].h_parent
-+#define dst_wh_dentry sd[AuDST].wh_dentry
-+#define dst_hdir sd[AuDST].hdir
-+#define dst_h_dir sd[AuDST].hdir->hi_inode
-+#define dst_dt sd[AuDST].dt
-+#define dst_btop sd[AuDST].btop
++ err = 1;
++ ii_write_lock_new_child(inode);
++ h_dentry = au_h_dptr(dentry, au_dbtop(dentry));
++ h_dinode = d_inode(h_dentry);
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
++ h_inode = au_h_iptr(inode, bindex);
++ if (!h_inode || h_inode != h_dinode)
++ continue;
+
-+ struct dentry *h_trap;
-+ struct au_branch *br;
-+ struct au_hinode *src_hinode;
-+ struct path h_path;
-+ struct au_nhash whlist;
-+ aufs_bindex_t btgt, src_bwh, src_bdiropq;
++ err = 0;
++ gen = au_iigen(inode, &igflags);
++ if (gen == au_digen(dentry)
++ && !au_ig_ftest(igflags, HALF_REFRESHED))
++ break;
+
-+ unsigned int flags;
++ /* fully refresh inode using dentry */
++ err = au_refresh_hinode(inode, dentry);
++ if (!err)
++ au_update_iigen(inode, /*half*/0);
++ break;
++ }
+
-+ struct au_whtmp_rmdir *thargs;
-+ struct dentry *h_dst;
-+};
++ if (unlikely(err))
++ ii_write_unlock(inode);
++out:
++ return err;
++}
+
-+/* ---------------------------------------------------------------------- */
++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ unsigned int d_type, ino_t *ino)
++{
++ int err, idx;
++ const int isnondir = d_type != DT_DIR;
+
-+/*
-+ * functions for reverting.
-+ * when an error happened in a single rename systemcall, we should revert
-+ * everything as if nothing happened.
-+ * we don't need to revert the copied-up/down the parent dir since they are
-+ * harmless.
-+ */
++ /* prevent hardlinked inode number from race condition */
++ if (isnondir) {
++ err = au_xinondir_enter(sb, bindex, h_ino, &idx);
++ if (unlikely(err))
++ goto out;
++ }
+
-+#define RevertFailure(fmt, ...) do { \
-+ AuIOErr("revert failure: " fmt " (%d, %d)\n", \
-+ ##__VA_ARGS__, err, rerr); \
-+ err = -EIO; \
-+} while (0)
++ err = au_xino_read(sb, bindex, h_ino, ino);
++ if (unlikely(err))
++ goto out_xinondir;
+
-+static void au_ren_rev_diropq(int err, struct au_ren_args *a)
-+{
-+ int rerr;
++ if (!*ino) {
++ err = -EIO;
++ *ino = au_xino_new_ino(sb);
++ if (unlikely(!*ino))
++ goto out_xinondir;
++ err = au_xino_write(sb, bindex, h_ino, *ino);
++ if (unlikely(err))
++ goto out_xinondir;
++ }
+
-+ au_hn_inode_lock_nested(a->src_hinode, AuLsc_I_CHILD);
-+ rerr = au_diropq_remove(a->src_dentry, a->btgt);
-+ au_hn_inode_unlock(a->src_hinode);
-+ au_set_dbdiropq(a->src_dentry, a->src_bdiropq);
-+ if (rerr)
-+ RevertFailure("remove diropq %pd", a->src_dentry);
++out_xinondir:
++ if (isnondir && idx >= 0)
++ au_xinondir_leave(sb, bindex, h_ino, idx);
++out:
++ return err;
+}
+
-+static void au_ren_rev_rename(int err, struct au_ren_args *a)
++/* successful returns with iinfo write_locked */
++/* todo: return with unlocked? */
++struct inode *au_new_inode(struct dentry *dentry, int must_new)
+{
-+ int rerr;
-+ struct inode *delegated;
++ struct inode *inode, *h_inode;
++ struct dentry *h_dentry;
++ struct super_block *sb;
++ ino_t h_ino, ino;
++ int err, idx, hlinked;
++ aufs_bindex_t btop;
+
-+ a->h_path.dentry = vfsub_lkup_one(&a->src_dentry->d_name,
-+ a->src_h_parent);
-+ rerr = PTR_ERR(a->h_path.dentry);
-+ if (IS_ERR(a->h_path.dentry)) {
-+ RevertFailure("lkup one %pd", a->src_dentry);
-+ return;
-+ }
++ sb = dentry->d_sb;
++ btop = au_dbtop(dentry);
++ h_dentry = au_h_dptr(dentry, btop);
++ h_inode = d_inode(h_dentry);
++ h_ino = h_inode->i_ino;
++ hlinked = !d_is_dir(h_dentry) && h_inode->i_nlink > 1;
+
-+ delegated = NULL;
-+ rerr = vfsub_rename(a->dst_h_dir,
-+ au_h_dptr(a->src_dentry, a->btgt),
-+ a->src_h_dir, &a->h_path, &delegated);
-+ if (unlikely(rerr == -EWOULDBLOCK)) {
-+ pr_warn("cannot retry for NFSv4 delegation"
-+ " for an internal rename\n");
-+ iput(delegated);
++new_ino:
++ /*
++ * stop 'race'-ing between hardlinks under different
++ * parents.
++ */
++ if (hlinked) {
++ err = au_xinondir_enter(sb, btop, h_ino, &idx);
++ inode = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
+ }
-+ d_drop(a->h_path.dentry);
-+ dput(a->h_path.dentry);
-+ /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */
-+ if (rerr)
-+ RevertFailure("rename %pd", a->src_dentry);
-+}
+
-+static void au_ren_rev_whtmp(int err, struct au_ren_args *a)
-+{
-+ int rerr;
-+ struct inode *delegated;
++ err = au_xino_read(sb, btop, h_ino, &ino);
++ inode = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_xinondir;
+
-+ a->h_path.dentry = vfsub_lkup_one(&a->dst_dentry->d_name,
-+ a->dst_h_parent);
-+ rerr = PTR_ERR(a->h_path.dentry);
-+ if (IS_ERR(a->h_path.dentry)) {
-+ RevertFailure("lkup one %pd", a->dst_dentry);
-+ return;
-+ }
-+ if (d_is_positive(a->h_path.dentry)) {
-+ d_drop(a->h_path.dentry);
-+ dput(a->h_path.dentry);
-+ return;
++ if (!ino) {
++ ino = au_xino_new_ino(sb);
++ if (unlikely(!ino)) {
++ inode = ERR_PTR(-EIO);
++ goto out_xinondir;
++ }
+ }
+
-+ delegated = NULL;
-+ rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path,
-+ &delegated);
-+ if (unlikely(rerr == -EWOULDBLOCK)) {
-+ pr_warn("cannot retry for NFSv4 delegation"
-+ " for an internal rename\n");
-+ iput(delegated);
++ AuDbg("i%lu\n", (unsigned long)ino);
++ inode = au_iget_locked(sb, ino);
++ err = PTR_ERR(inode);
++ if (IS_ERR(inode))
++ goto out_xinondir;
++
++ AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW));
++ if (inode->i_state & I_NEW) {
++ ii_write_lock_new_child(inode);
++ err = set_inode(inode, dentry);
++ if (!err) {
++ unlock_new_inode(inode);
++ goto out_xinondir; /* success */
++ }
++
++ /*
++ * iget_failed() calls iput(), but we need to call
++ * ii_write_unlock() after iget_failed(). so dirty hack for
++ * i_count.
++ */
++ atomic_inc(&inode->i_count);
++ iget_failed(inode);
++ ii_write_unlock(inode);
++ au_xino_write(sb, btop, h_ino, /*ino*/0);
++ /* ignore this error */
++ goto out_iput;
++ } else if (!must_new && !IS_DEADDIR(inode) && inode->i_nlink) {
++ /*
++ * horrible race condition between lookup, readdir and copyup
++ * (or something).
++ */
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++ err = reval_inode(inode, dentry);
++ if (unlikely(err < 0)) {
++ hlinked = 0;
++ goto out_iput;
++ }
++ if (!err)
++ goto out; /* success */
++ else if (hlinked && idx >= 0) {
++ err = au_xinondir_enter(sb, btop, h_ino, &idx);
++ if (unlikely(err)) {
++ iput(inode);
++ inode = ERR_PTR(err);
++ goto out;
++ }
++ }
+ }
-+ d_drop(a->h_path.dentry);
-+ dput(a->h_path.dentry);
-+ if (!rerr)
-+ au_set_h_dptr(a->dst_dentry, a->btgt, dget(a->h_dst));
-+ else
-+ RevertFailure("rename %pd", a->h_dst);
-+}
+
-+static void au_ren_rev_whsrc(int err, struct au_ren_args *a)
-+{
-+ int rerr;
++ if (unlikely(au_test_fs_unique_ino(h_inode)))
++ AuWarn1("Warning: Un-notified UDBA or repeatedly renamed dir,"
++ " b%d, %s, %pd, hi%lu, i%lu.\n",
++ btop, au_sbtype(h_dentry->d_sb), dentry,
++ (unsigned long)h_ino, (unsigned long)ino);
++ ino = 0;
++ err = au_xino_write(sb, btop, h_ino, /*ino*/0);
++ if (!err) {
++ iput(inode);
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++ goto new_ino;
++ }
+
-+ a->h_path.dentry = a->src_wh_dentry;
-+ rerr = au_wh_unlink_dentry(a->src_h_dir, &a->h_path, a->src_dentry);
-+ au_set_dbwh(a->src_dentry, a->src_bwh);
-+ if (rerr)
-+ RevertFailure("unlink %pd", a->src_wh_dentry);
++out_iput:
++ iput(inode);
++ inode = ERR_PTR(err);
++out_xinondir:
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++out:
++ return inode;
+}
-+#undef RevertFailure
+
+/* ---------------------------------------------------------------------- */
+
-+/*
-+ * when we have to copyup the renaming entry, do it with the rename-target name
-+ * in order to minimize the cost (the later actual rename is unnecessary).
-+ * otherwise rename it on the target branch.
-+ */
-+static int au_ren_or_cpup(struct au_ren_args *a)
++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
++ struct inode *inode)
+{
+ int err;
-+ struct dentry *d;
-+ struct inode *delegated;
++ struct inode *hi;
+
-+ d = a->src_dentry;
-+ if (au_dbtop(d) == a->btgt) {
-+ a->h_path.dentry = a->dst_h_dentry;
-+ if (au_ftest_ren(a->flags, DIROPQ)
-+ && au_dbdiropq(d) == a->btgt)
-+ au_fclr_ren(a->flags, DIROPQ);
-+ AuDebugOn(au_dbtop(d) != a->btgt);
-+ delegated = NULL;
-+ err = vfsub_rename(a->src_h_dir, au_h_dptr(d, a->btgt),
-+ a->dst_h_dir, &a->h_path, &delegated);
-+ if (unlikely(err == -EWOULDBLOCK)) {
-+ pr_warn("cannot retry for NFSv4 delegation"
-+ " for an internal rename\n");
-+ iput(delegated);
-+ }
-+ } else
-+ BUG();
++ err = au_br_rdonly(au_sbr(sb, bindex));
+
-+ if (!err && a->h_dst)
-+ /* it will be set to dinfo later */
-+ dget(a->h_dst);
++ /* pseudo-link after flushed may happen out of bounds */
++ if (!err
++ && inode
++ && au_ibtop(inode) <= bindex
++ && bindex <= au_ibbot(inode)) {
++ /*
++ * permission check is unnecessary since vfsub routine
++ * will be called later
++ */
++ hi = au_h_iptr(inode, bindex);
++ if (hi)
++ err = IS_IMMUTABLE(hi) ? -EROFS : 0;
++ }
+
+ return err;
+}
+
-+/* cf. aufs_rmdir() */
-+static int au_ren_del_whtmp(struct au_ren_args *a)
++int au_test_h_perm(struct inode *h_inode, int mask)
+{
-+ int err;
-+ struct inode *dir;
-+
-+ dir = a->dst_dir;
-+ SiMustAnyLock(dir->i_sb);
-+ if (!au_nhash_test_longer_wh(&a->whlist, a->btgt,
-+ au_sbi(dir->i_sb)->si_dirwh)
-+ || au_test_fs_remote(a->h_dst->d_sb)) {
-+ err = au_whtmp_rmdir(dir, a->btgt, a->h_dst, &a->whlist);
-+ if (unlikely(err))
-+ pr_warn("failed removing whtmp dir %pd (%d), "
-+ "ignored.\n", a->h_dst, err);
-+ } else {
-+ au_nhash_wh_free(&a->thargs->whlist);
-+ a->thargs->whlist = a->whlist;
-+ a->whlist.nh_num = 0;
-+ au_whtmp_kick_rmdir(dir, a->btgt, a->h_dst, a->thargs);
-+ dput(a->h_dst);
-+ a->thargs = NULL;
-+ }
-+
-+ return 0;
++ if (uid_eq(current_fsuid(), GLOBAL_ROOT_UID))
++ return 0;
++ return inode_permission(h_inode, mask);
+}
+
-+/* make it 'opaque' dir. */
-+static int au_ren_diropq(struct au_ren_args *a)
++int au_test_h_perm_sio(struct inode *h_inode, int mask)
+{
-+ int err;
-+ struct dentry *diropq;
++ if (au_test_nfs(h_inode->i_sb)
++ && (mask & MAY_WRITE)
++ && S_ISDIR(h_inode->i_mode))
++ mask |= MAY_READ; /* force permission check */
++ return au_test_h_perm(h_inode, mask);
++}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/inode.h linux-4.14/fs/aufs/inode.h
+--- linux-4.14/fs/aufs/inode.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/inode.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,695 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+ err = 0;
-+ a->src_bdiropq = au_dbdiropq(a->src_dentry);
-+ a->src_hinode = au_hi(a->src_inode, a->btgt);
-+ au_hn_inode_lock_nested(a->src_hinode, AuLsc_I_CHILD);
-+ diropq = au_diropq_create(a->src_dentry, a->btgt);
-+ au_hn_inode_unlock(a->src_hinode);
-+ if (IS_ERR(diropq))
-+ err = PTR_ERR(diropq);
-+ else
-+ dput(diropq);
++/*
++ * inode operations
++ */
+
-+ return err;
-+}
++#ifndef __AUFS_INODE_H__
++#define __AUFS_INODE_H__
+
-+static int do_rename(struct au_ren_args *a)
-+{
-+ int err;
-+ struct dentry *d, *h_d;
++#ifdef __KERNEL__
+
-+ /* prepare workqueue args for asynchronous rmdir */
-+ h_d = a->dst_h_dentry;
-+ if (au_ftest_ren(a->flags, ISDIR) && d_is_positive(h_d)) {
-+ err = -ENOMEM;
-+ a->thargs = au_whtmp_rmdir_alloc(a->src_dentry->d_sb, GFP_NOFS);
-+ if (unlikely(!a->thargs))
-+ goto out;
-+ a->h_dst = dget(h_d);
-+ }
-+
-+ /* create whiteout for src_dentry */
-+ if (au_ftest_ren(a->flags, WHSRC)) {
-+ a->src_bwh = au_dbwh(a->src_dentry);
-+ AuDebugOn(a->src_bwh >= 0);
-+ a->src_wh_dentry
-+ = au_wh_create(a->src_dentry, a->btgt, a->src_h_parent);
-+ err = PTR_ERR(a->src_wh_dentry);
-+ if (IS_ERR(a->src_wh_dentry))
-+ goto out_thargs;
-+ }
-+
-+ /* lookup whiteout for dentry */
-+ if (au_ftest_ren(a->flags, WHDST)) {
-+ h_d = au_wh_lkup(a->dst_h_parent, &a->dst_dentry->d_name,
-+ a->br);
-+ err = PTR_ERR(h_d);
-+ if (IS_ERR(h_d))
-+ goto out_whsrc;
-+ if (d_is_negative(h_d))
-+ dput(h_d);
-+ else
-+ a->dst_wh_dentry = h_d;
-+ }
++#include <linux/fsnotify.h>
++#include "rwsem.h"
+
-+ /* rename dentry to tmpwh */
-+ if (a->thargs) {
-+ err = au_whtmp_ren(a->dst_h_dentry, a->br);
-+ if (unlikely(err))
-+ goto out_whdst;
++struct vfsmount;
+
-+ d = a->dst_dentry;
-+ au_set_h_dptr(d, a->btgt, NULL);
-+ err = au_lkup_neg(d, a->btgt, /*wh*/0);
-+ if (unlikely(err))
-+ goto out_whtmp;
-+ a->dst_h_dentry = au_h_dptr(d, a->btgt);
-+ }
++struct au_hnotify {
++#ifdef CONFIG_AUFS_HNOTIFY
++#ifdef CONFIG_AUFS_HFSNOTIFY
++ /* never use fsnotify_add_vfsmount_mark() */
++ struct fsnotify_mark hn_mark;
++#endif
++ struct inode *hn_aufs_inode; /* no get/put */
++#endif
++} ____cacheline_aligned_in_smp;
+
-+ BUG_ON(d_is_positive(a->dst_h_dentry) && a->src_btop != a->btgt);
++struct au_hinode {
++ struct inode *hi_inode;
++ aufs_bindex_t hi_id;
++#ifdef CONFIG_AUFS_HNOTIFY
++ struct au_hnotify *hi_notify;
++#endif
+
-+ /* rename by vfs_rename or cpup */
-+ d = a->dst_dentry;
-+ if (au_ftest_ren(a->flags, ISDIR)
-+ && (a->dst_wh_dentry
-+ || au_dbdiropq(d) == a->btgt
-+ /* hide the lower to keep xino */
-+ || a->btgt < au_dbbot(d)
-+ || au_opt_test(au_mntflags(d->d_sb), ALWAYS_DIROPQ)))
-+ au_fset_ren(a->flags, DIROPQ);
-+ err = au_ren_or_cpup(a);
-+ if (unlikely(err))
-+ /* leave the copied-up one */
-+ goto out_whtmp;
++ /* reference to the copied-up whiteout with get/put */
++ struct dentry *hi_whdentry;
++};
+
-+ /* make dir opaque */
-+ if (au_ftest_ren(a->flags, DIROPQ)) {
-+ err = au_ren_diropq(a);
-+ if (unlikely(err))
-+ goto out_rename;
-+ }
++/* ig_flags */
++#define AuIG_HALF_REFRESHED 1
++#define au_ig_ftest(flags, name) ((flags) & AuIG_##name)
++#define au_ig_fset(flags, name) \
++ do { (flags) |= AuIG_##name; } while (0)
++#define au_ig_fclr(flags, name) \
++ do { (flags) &= ~AuIG_##name; } while (0)
+
-+ /* update target timestamps */
-+ AuDebugOn(au_dbtop(a->src_dentry) != a->btgt);
-+ a->h_path.dentry = au_h_dptr(a->src_dentry, a->btgt);
-+ vfsub_update_h_iattr(&a->h_path, /*did*/NULL); /*ignore*/
-+ a->src_inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
++struct au_iigen {
++ spinlock_t ig_spin;
++ __u32 ig_generation, ig_flags;
++};
+
-+ /* remove whiteout for dentry */
-+ if (a->dst_wh_dentry) {
-+ a->h_path.dentry = a->dst_wh_dentry;
-+ err = au_wh_unlink_dentry(a->dst_h_dir, &a->h_path,
-+ a->dst_dentry);
-+ if (unlikely(err))
-+ goto out_diropq;
-+ }
++struct au_vdir;
++struct au_iinfo {
++ struct au_iigen ii_generation;
++ struct super_block *ii_hsb1; /* no get/put */
+
-+ /* remove whtmp */
-+ if (a->thargs)
-+ au_ren_del_whtmp(a); /* ignore this error */
++ struct au_rwsem ii_rwsem;
++ aufs_bindex_t ii_btop, ii_bbot;
++ __u32 ii_higen;
++ struct au_hinode *ii_hinode;
++ struct au_vdir *ii_vdir;
++};
+
-+ au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0);
-+ err = 0;
-+ goto out_success;
++struct au_icntnr {
++ struct au_iinfo iinfo;
++ struct inode vfs_inode;
++ struct hlist_bl_node plink;
++} ____cacheline_aligned_in_smp;
++
++/* au_pin flags */
++#define AuPin_DI_LOCKED 1
++#define AuPin_MNT_WRITE (1 << 1)
++#define au_ftest_pin(flags, name) ((flags) & AuPin_##name)
++#define au_fset_pin(flags, name) \
++ do { (flags) |= AuPin_##name; } while (0)
++#define au_fclr_pin(flags, name) \
++ do { (flags) &= ~AuPin_##name; } while (0)
++
++struct au_pin {
++ /* input */
++ struct dentry *dentry;
++ unsigned int udba;
++ unsigned char lsc_di, lsc_hi, flags;
++ aufs_bindex_t bindex;
++
++ /* output */
++ struct dentry *parent;
++ struct au_hinode *hdir;
++ struct vfsmount *h_mnt;
++
++ /* temporary unlock/relock for copyup */
++ struct dentry *h_dentry, *h_parent;
++ struct au_branch *br;
++ struct task_struct *task;
++};
+
-+out_diropq:
-+ if (au_ftest_ren(a->flags, DIROPQ))
-+ au_ren_rev_diropq(err, a);
-+out_rename:
-+ au_ren_rev_rename(err, a);
-+ dput(a->h_dst);
-+out_whtmp:
-+ if (a->thargs)
-+ au_ren_rev_whtmp(err, a);
-+out_whdst:
-+ dput(a->dst_wh_dentry);
-+ a->dst_wh_dentry = NULL;
-+out_whsrc:
-+ if (a->src_wh_dentry)
-+ au_ren_rev_whsrc(err, a);
-+out_success:
-+ dput(a->src_wh_dentry);
-+ dput(a->dst_wh_dentry);
-+out_thargs:
-+ if (a->thargs) {
-+ dput(a->h_dst);
-+ au_whtmp_rmdir_free(a->thargs);
-+ a->thargs = NULL;
-+ }
-+out:
-+ return err;
-+}
++void au_pin_hdir_unlock(struct au_pin *p);
++int au_pin_hdir_lock(struct au_pin *p);
++int au_pin_hdir_relock(struct au_pin *p);
++void au_pin_hdir_acquire_nest(struct au_pin *p);
++void au_pin_hdir_release(struct au_pin *p);
+
+/* ---------------------------------------------------------------------- */
+
-+/*
-+ * test if @dentry dir can be rename destination or not.
-+ * success means, it is a logically empty dir.
-+ */
-+static int may_rename_dstdir(struct dentry *dentry, struct au_nhash *whlist)
++static inline struct au_iinfo *au_ii(struct inode *inode)
+{
-+ return au_test_empty(dentry, whlist);
++ BUG_ON(is_bad_inode(inode));
++ return &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
+}
+
-+/*
-+ * test if @dentry dir can be rename source or not.
-+ * if it can, return 0 and @children is filled.
-+ * success means,
-+ * - it is a logically empty dir.
-+ * - or, it exists on writable branch and has no children including whiteouts
-+ * on the lower branch.
-+ */
-+static int may_rename_srcdir(struct dentry *dentry, aufs_bindex_t btgt)
-+{
-+ int err;
-+ unsigned int rdhash;
-+ aufs_bindex_t btop;
++/* ---------------------------------------------------------------------- */
+
-+ btop = au_dbtop(dentry);
-+ if (btop != btgt) {
-+ struct au_nhash whlist;
++/* inode.c */
++struct inode *au_igrab(struct inode *inode);
++void au_refresh_iop(struct inode *inode, int force_getattr);
++int au_refresh_hinode_self(struct inode *inode);
++int au_refresh_hinode(struct inode *inode, struct dentry *dentry);
++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ unsigned int d_type, ino_t *ino);
++struct inode *au_new_inode(struct dentry *dentry, int must_new);
++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
++ struct inode *inode);
++int au_test_h_perm(struct inode *h_inode, int mask);
++int au_test_h_perm_sio(struct inode *h_inode, int mask);
+
-+ SiMustAnyLock(dentry->d_sb);
-+ rdhash = au_sbi(dentry->d_sb)->si_rdhash;
-+ if (!rdhash)
-+ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL,
-+ dentry));
-+ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_test_empty(dentry, &whlist);
-+ au_nhash_wh_free(&whlist);
-+ goto out;
-+ }
++static inline int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex,
++ ino_t h_ino, unsigned int d_type, ino_t *ino)
++{
++#ifdef CONFIG_AUFS_SHWH
++ return au_ino(sb, bindex, h_ino, d_type, ino);
++#else
++ return 0;
++#endif
++}
+
-+ if (btop == au_dbtaildir(dentry))
-+ return 0; /* success */
++/* i_op.c */
++enum {
++ AuIop_SYMLINK,
++ AuIop_DIR,
++ AuIop_OTHER,
++ AuIop_Last
++};
++extern struct inode_operations aufs_iop[AuIop_Last],
++ aufs_iop_nogetattr[AuIop_Last];
+
-+ err = au_test_empty_lower(dentry);
++/* au_wr_dir flags */
++#define AuWrDir_ADD_ENTRY 1
++#define AuWrDir_ISDIR (1 << 1)
++#define AuWrDir_TMPFILE (1 << 2)
++#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name)
++#define au_fset_wrdir(flags, name) \
++ do { (flags) |= AuWrDir_##name; } while (0)
++#define au_fclr_wrdir(flags, name) \
++ do { (flags) &= ~AuWrDir_##name; } while (0)
+
-+out:
-+ if (err == -ENOTEMPTY) {
-+ AuWarn1("renaming dir who has child(ren) on multiple branches,"
-+ " is not supported\n");
-+ err = -EXDEV;
-+ }
-+ return err;
-+}
++struct au_wr_dir_args {
++ aufs_bindex_t force_btgt;
++ unsigned char flags;
++};
++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
++ struct au_wr_dir_args *args);
+
-+/* side effect: sets whlist and h_dentry */
-+static int au_ren_may_dir(struct au_ren_args *a)
-+{
-+ int err;
-+ unsigned int rdhash;
-+ struct dentry *d;
++struct dentry *au_pinned_h_parent(struct au_pin *pin);
++void au_pin_init(struct au_pin *pin, struct dentry *dentry,
++ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
++ unsigned int udba, unsigned char flags);
++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int udba, unsigned char flags) __must_check;
++int au_do_pin(struct au_pin *pin) __must_check;
++void au_unpin(struct au_pin *pin);
++int au_reval_for_attr(struct dentry *dentry, unsigned int sigen);
+
-+ d = a->dst_dentry;
-+ SiMustAnyLock(d->d_sb);
++#define AuIcpup_DID_CPUP 1
++#define au_ftest_icpup(flags, name) ((flags) & AuIcpup_##name)
++#define au_fset_icpup(flags, name) \
++ do { (flags) |= AuIcpup_##name; } while (0)
++#define au_fclr_icpup(flags, name) \
++ do { (flags) &= ~AuIcpup_##name; } while (0)
+
-+ err = 0;
-+ if (au_ftest_ren(a->flags, ISDIR) && a->dst_inode) {
-+ rdhash = au_sbi(d->d_sb)->si_rdhash;
-+ if (!rdhash)
-+ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, d));
-+ err = au_nhash_alloc(&a->whlist, rdhash, GFP_NOFS);
-+ if (unlikely(err))
-+ goto out;
++struct au_icpup_args {
++ unsigned char flags;
++ unsigned char pin_flags;
++ aufs_bindex_t btgt;
++ unsigned int udba;
++ struct au_pin pin;
++ struct path h_path;
++ struct inode *h_inode;
++};
+
-+ au_set_dbtop(d, a->dst_btop);
-+ err = may_rename_dstdir(d, &a->whlist);
-+ au_set_dbtop(d, a->btgt);
-+ }
-+ a->dst_h_dentry = au_h_dptr(d, au_dbtop(d));
-+ if (unlikely(err))
-+ goto out;
++int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
++ struct au_icpup_args *a);
+
-+ d = a->src_dentry;
-+ a->src_h_dentry = au_h_dptr(d, au_dbtop(d));
-+ if (au_ftest_ren(a->flags, ISDIR)) {
-+ err = may_rename_srcdir(d, a->btgt);
-+ if (unlikely(err)) {
-+ au_nhash_wh_free(&a->whlist);
-+ a->whlist.nh_num = 0;
-+ }
-+ }
-+out:
-+ return err;
-+}
++int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path,
++ int locked);
+
-+/* ---------------------------------------------------------------------- */
++/* i_op_add.c */
++int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir);
++int aufs_mknod(struct inode *dir, struct dentry *dentry, umode_t mode,
++ dev_t dev);
++int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname);
++int aufs_create(struct inode *dir, struct dentry *dentry, umode_t mode,
++ bool want_excl);
++struct vfsub_aopen_args;
++int au_aopen_or_create(struct inode *dir, struct dentry *dentry,
++ struct vfsub_aopen_args *args);
++int aufs_tmpfile(struct inode *dir, struct dentry *dentry, umode_t mode);
++int aufs_link(struct dentry *src_dentry, struct inode *dir,
++ struct dentry *dentry);
++int aufs_mkdir(struct inode *dir, struct dentry *dentry, umode_t mode);
+
-+/*
-+ * simple tests for rename.
-+ * following the checks in vfs, plus the parent-child relationship.
-+ */
-+static int au_may_ren(struct au_ren_args *a)
-+{
-+ int err, isdir;
-+ struct inode *h_inode;
++/* i_op_del.c */
++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup);
++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir);
++int aufs_unlink(struct inode *dir, struct dentry *dentry);
++int aufs_rmdir(struct inode *dir, struct dentry *dentry);
+
-+ if (a->src_btop == a->btgt) {
-+ err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent,
-+ au_ftest_ren(a->flags, ISDIR));
-+ if (unlikely(err))
-+ goto out;
-+ err = -EINVAL;
-+ if (unlikely(a->src_h_dentry == a->h_trap))
-+ goto out;
-+ }
++/* i_op_ren.c */
++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt);
++int aufs_rename(struct inode *src_dir, struct dentry *src_dentry,
++ struct inode *dir, struct dentry *dentry,
++ unsigned int flags);
+
-+ err = 0;
-+ if (a->dst_btop != a->btgt)
-+ goto out;
++/* iinfo.c */
++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex);
++void au_hiput(struct au_hinode *hinode);
++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_wh);
++unsigned int au_hi_flags(struct inode *inode, int isdir);
+
-+ err = -ENOTEMPTY;
-+ if (unlikely(a->dst_h_dentry == a->h_trap))
-+ goto out;
++/* hinode flags */
++#define AuHi_XINO 1
++#define AuHi_HNOTIFY (1 << 1)
++#define au_ftest_hi(flags, name) ((flags) & AuHi_##name)
++#define au_fset_hi(flags, name) \
++ do { (flags) |= AuHi_##name; } while (0)
++#define au_fclr_hi(flags, name) \
++ do { (flags) &= ~AuHi_##name; } while (0)
+
-+ err = -EIO;
-+ isdir = !!au_ftest_ren(a->flags, ISDIR);
-+ if (d_really_is_negative(a->dst_dentry)) {
-+ if (d_is_negative(a->dst_h_dentry))
-+ err = au_may_add(a->dst_dentry, a->btgt,
-+ a->dst_h_parent, isdir);
-+ } else {
-+ if (unlikely(d_is_negative(a->dst_h_dentry)))
-+ goto out;
-+ h_inode = d_inode(a->dst_h_dentry);
-+ if (h_inode->i_nlink)
-+ err = au_may_del(a->dst_dentry, a->btgt,
-+ a->dst_h_parent, isdir);
-+ }
++#ifndef CONFIG_AUFS_HNOTIFY
++#undef AuHi_HNOTIFY
++#define AuHi_HNOTIFY 0
++#endif
+
-+out:
-+ if (unlikely(err == -ENOENT || err == -EEXIST))
-+ err = -EIO;
-+ AuTraceErr(err);
-+ return err;
-+}
++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode, unsigned int flags);
+
-+/* ---------------------------------------------------------------------- */
++void au_update_iigen(struct inode *inode, int half);
++void au_update_ibrange(struct inode *inode, int do_put_zero);
+
-+/*
-+ * locking order
-+ * (VFS)
-+ * - src_dir and dir by lock_rename()
-+ * - inode if exitsts
-+ * (aufs)
-+ * - lock all
-+ * + src_dentry and dentry by aufs_read_and_write_lock2() which calls,
-+ * + si_read_lock
-+ * + di_write_lock2_child()
-+ * + di_write_lock_child()
-+ * + ii_write_lock_child()
-+ * + di_write_lock_child2()
-+ * + ii_write_lock_child2()
-+ * + src_parent and parent
-+ * + di_write_lock_parent()
-+ * + ii_write_lock_parent()
-+ * + di_write_lock_parent2()
-+ * + ii_write_lock_parent2()
-+ * + lower src_dir and dir by vfsub_lock_rename()
-+ * + verify the every relationships between child and parent. if any
-+ * of them failed, unlock all and return -EBUSY.
-+ */
-+static void au_ren_unlock(struct au_ren_args *a)
-+{
-+ vfsub_unlock_rename(a->src_h_parent, a->src_hdir,
-+ a->dst_h_parent, a->dst_hdir);
-+ if (au_ftest_ren(a->flags, MNT_WRITE))
-+ vfsub_mnt_drop_write(au_br_mnt(a->br));
-+}
++void au_icntnr_init_once(void *_c);
++void au_hinode_init(struct au_hinode *hinode);
++int au_iinfo_init(struct inode *inode);
++void au_iinfo_fin(struct inode *inode);
++int au_hinode_realloc(struct au_iinfo *iinfo, int nbr, int may_shrink);
+
-+static int au_ren_lock(struct au_ren_args *a)
-+{
-+ int err;
-+ unsigned int udba;
++#ifdef CONFIG_PROC_FS
++/* plink.c */
++int au_plink_maint(struct super_block *sb, int flags);
++struct au_sbinfo;
++void au_plink_maint_leave(struct au_sbinfo *sbinfo);
++int au_plink_maint_enter(struct super_block *sb);
++#ifdef CONFIG_AUFS_DEBUG
++void au_plink_list(struct super_block *sb);
++#else
++AuStubVoid(au_plink_list, struct super_block *sb)
++#endif
++int au_plink_test(struct inode *inode);
++struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex);
++void au_plink_append(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_dentry);
++void au_plink_put(struct super_block *sb, int verbose);
++void au_plink_clean(struct super_block *sb, int verbose);
++void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id);
++#else
++AuStubInt0(au_plink_maint, struct super_block *sb, int flags);
++AuStubVoid(au_plink_maint_leave, struct au_sbinfo *sbinfo);
++AuStubInt0(au_plink_maint_enter, struct super_block *sb);
++AuStubVoid(au_plink_list, struct super_block *sb);
++AuStubInt0(au_plink_test, struct inode *inode);
++AuStub(struct dentry *, au_plink_lkup, return NULL,
++ struct inode *inode, aufs_bindex_t bindex);
++AuStubVoid(au_plink_append, struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_dentry);
++AuStubVoid(au_plink_put, struct super_block *sb, int verbose);
++AuStubVoid(au_plink_clean, struct super_block *sb, int verbose);
++AuStubVoid(au_plink_half_refresh, struct super_block *sb, aufs_bindex_t br_id);
++#endif /* CONFIG_PROC_FS */
+
-+ err = 0;
-+ a->src_h_parent = au_h_dptr(a->src_parent, a->btgt);
-+ a->src_hdir = au_hi(a->src_dir, a->btgt);
-+ a->dst_h_parent = au_h_dptr(a->dst_parent, a->btgt);
-+ a->dst_hdir = au_hi(a->dst_dir, a->btgt);
++#ifdef CONFIG_AUFS_XATTR
++/* xattr.c */
++int au_cpup_xattr(struct dentry *h_dst, struct dentry *h_src, int ignore_flags,
++ unsigned int verbose);
++ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size);
++void au_xattr_init(struct super_block *sb);
++#else
++AuStubInt0(au_cpup_xattr, struct dentry *h_dst, struct dentry *h_src,
++ int ignore_flags, unsigned int verbose);
++AuStubVoid(au_xattr_init, struct super_block *sb);
++#endif
+
-+ err = vfsub_mnt_want_write(au_br_mnt(a->br));
-+ if (unlikely(err))
-+ goto out;
-+ au_fset_ren(a->flags, MNT_WRITE);
-+ a->h_trap = vfsub_lock_rename(a->src_h_parent, a->src_hdir,
-+ a->dst_h_parent, a->dst_hdir);
-+ udba = au_opt_udba(a->src_dentry->d_sb);
-+ if (unlikely(a->src_hdir->hi_inode != d_inode(a->src_h_parent)
-+ || a->dst_hdir->hi_inode != d_inode(a->dst_h_parent)))
-+ err = au_busy_or_stale();
-+ if (!err && au_dbtop(a->src_dentry) == a->btgt)
-+ err = au_h_verify(a->src_h_dentry, udba,
-+ d_inode(a->src_h_parent), a->src_h_parent,
-+ a->br);
-+ if (!err && au_dbtop(a->dst_dentry) == a->btgt)
-+ err = au_h_verify(a->dst_h_dentry, udba,
-+ d_inode(a->dst_h_parent), a->dst_h_parent,
-+ a->br);
-+ if (!err)
-+ goto out; /* success */
++#ifdef CONFIG_FS_POSIX_ACL
++struct posix_acl *aufs_get_acl(struct inode *inode, int type);
++int aufs_set_acl(struct inode *inode, struct posix_acl *acl, int type);
++#endif
+
-+ err = au_busy_or_stale();
-+ au_ren_unlock(a);
++#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
++enum {
++ AU_XATTR_SET,
++ AU_ACL_SET
++};
+
-+out:
-+ return err;
-+}
++struct au_sxattr {
++ int type;
++ union {
++ struct {
++ const char *name;
++ const void *value;
++ size_t size;
++ int flags;
++ } set;
++ struct {
++ struct posix_acl *acl;
++ int type;
++ } acl_set;
++ } u;
++};
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg);
++#endif
+
+/* ---------------------------------------------------------------------- */
+
-+static void au_ren_refresh_dir(struct au_ren_args *a)
-+{
-+ struct inode *dir;
-+
-+ dir = a->dst_dir;
-+ dir->i_version++;
-+ if (au_ftest_ren(a->flags, ISDIR)) {
-+ /* is this updating defined in POSIX? */
-+ au_cpup_attr_timesizes(a->src_inode);
-+ au_cpup_attr_nlink(dir, /*force*/1);
-+ }
-+
-+ au_dir_ts(dir, a->btgt);
++/* lock subclass for iinfo */
++enum {
++ AuLsc_II_CHILD, /* child first */
++ AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hnotify */
++ AuLsc_II_CHILD3, /* copyup dirs */
++ AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */
++ AuLsc_II_PARENT2,
++ AuLsc_II_PARENT3, /* copyup dirs */
++ AuLsc_II_NEW_CHILD
++};
+
-+ if (au_ftest_ren(a->flags, ISSAMEDIR))
-+ return;
++/*
++ * ii_read_lock_child, ii_write_lock_child,
++ * ii_read_lock_child2, ii_write_lock_child2,
++ * ii_read_lock_child3, ii_write_lock_child3,
++ * ii_read_lock_parent, ii_write_lock_parent,
++ * ii_read_lock_parent2, ii_write_lock_parent2,
++ * ii_read_lock_parent3, ii_write_lock_parent3,
++ * ii_read_lock_new_child, ii_write_lock_new_child,
++ */
++#define AuReadLockFunc(name, lsc) \
++static inline void ii_read_lock_##name(struct inode *i) \
++{ \
++ au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
++}
+
-+ dir = a->src_dir;
-+ dir->i_version++;
-+ if (au_ftest_ren(a->flags, ISDIR))
-+ au_cpup_attr_nlink(dir, /*force*/1);
-+ au_dir_ts(dir, a->btgt);
++#define AuWriteLockFunc(name, lsc) \
++static inline void ii_write_lock_##name(struct inode *i) \
++{ \
++ au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
+}
+
-+static void au_ren_refresh(struct au_ren_args *a)
-+{
-+ aufs_bindex_t bbot, bindex;
-+ struct dentry *d, *h_d;
-+ struct inode *i, *h_i;
-+ struct super_block *sb;
++#define AuRWLockFuncs(name, lsc) \
++ AuReadLockFunc(name, lsc) \
++ AuWriteLockFunc(name, lsc)
+
-+ d = a->dst_dentry;
-+ d_drop(d);
-+ if (a->h_dst)
-+ /* already dget-ed by au_ren_or_cpup() */
-+ au_set_h_dptr(d, a->btgt, a->h_dst);
++AuRWLockFuncs(child, CHILD);
++AuRWLockFuncs(child2, CHILD2);
++AuRWLockFuncs(child3, CHILD3);
++AuRWLockFuncs(parent, PARENT);
++AuRWLockFuncs(parent2, PARENT2);
++AuRWLockFuncs(parent3, PARENT3);
++AuRWLockFuncs(new_child, NEW_CHILD);
+
-+ i = a->dst_inode;
-+ if (i) {
-+ if (!au_ftest_ren(a->flags, ISDIR))
-+ vfsub_drop_nlink(i);
-+ else {
-+ vfsub_dead_dir(i);
-+ au_cpup_attr_timesizes(i);
-+ }
-+ au_update_dbrange(d, /*do_put_zero*/1);
-+ } else {
-+ bbot = a->btgt;
-+ for (bindex = au_dbtop(d); bindex < bbot; bindex++)
-+ au_set_h_dptr(d, bindex, NULL);
-+ bbot = au_dbbot(d);
-+ for (bindex = a->btgt + 1; bindex <= bbot; bindex++)
-+ au_set_h_dptr(d, bindex, NULL);
-+ au_update_dbrange(d, /*do_put_zero*/0);
-+ }
++#undef AuReadLockFunc
++#undef AuWriteLockFunc
++#undef AuRWLockFuncs
+
-+ d = a->src_dentry;
-+ au_set_dbwh(d, -1);
-+ bbot = au_dbbot(d);
-+ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
-+ h_d = au_h_dptr(d, bindex);
-+ if (h_d)
-+ au_set_h_dptr(d, bindex, NULL);
-+ }
-+ au_set_dbbot(d, a->btgt);
++#define ii_read_unlock(i) au_rw_read_unlock(&au_ii(i)->ii_rwsem)
++#define ii_write_unlock(i) au_rw_write_unlock(&au_ii(i)->ii_rwsem)
++#define ii_downgrade_lock(i) au_rw_dgrade_lock(&au_ii(i)->ii_rwsem)
+
-+ sb = d->d_sb;
-+ i = a->src_inode;
-+ if (au_opt_test(au_mntflags(sb), PLINK) && au_plink_test(i))
-+ return; /* success */
++#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem)
++#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem)
++#define IiMustWriteLock(i) AuRwMustWriteLock(&au_ii(i)->ii_rwsem)
+
-+ bbot = au_ibbot(i);
-+ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
-+ h_i = au_h_iptr(i, bindex);
-+ if (h_i) {
-+ au_xino_write(sb, bindex, h_i->i_ino, /*ino*/0);
-+ /* ignore this error */
-+ au_set_h_iptr(i, bindex, NULL, 0);
-+ }
-+ }
-+ au_set_ibbot(i, a->btgt);
++/* ---------------------------------------------------------------------- */
++
++static inline void au_icntnr_init(struct au_icntnr *c)
++{
++#ifdef CONFIG_AUFS_DEBUG
++ c->vfs_inode.i_mode = 0;
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
++static inline unsigned int au_iigen(struct inode *inode, unsigned int *igflags)
++{
++ unsigned int gen;
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
++
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ if (igflags)
++ *igflags = iigen->ig_flags;
++ gen = iigen->ig_generation;
++ spin_unlock(&iigen->ig_spin);
+
-+/* mainly for link(2) and rename(2) */
-+int au_wbr(struct dentry *dentry, aufs_bindex_t btgt)
++ return gen;
++}
++
++/* tiny test for inode number */
++/* tmpfs generation is too rough */
++static inline int au_test_higen(struct inode *inode, struct inode *h_inode)
+{
-+ aufs_bindex_t bdiropq, bwh;
-+ struct dentry *parent;
-+ struct au_branch *br;
++ struct au_iinfo *iinfo;
+
-+ parent = dentry->d_parent;
-+ IMustLock(d_inode(parent)); /* dir is locked */
++ iinfo = au_ii(inode);
++ AuRwMustAnyLock(&iinfo->ii_rwsem);
++ return !(iinfo->ii_hsb1 == h_inode->i_sb
++ && iinfo->ii_higen == h_inode->i_generation);
++}
+
-+ bdiropq = au_dbdiropq(parent);
-+ bwh = au_dbwh(dentry);
-+ br = au_sbr(dentry->d_sb, btgt);
-+ if (au_br_rdonly(br)
-+ || (0 <= bdiropq && bdiropq < btgt)
-+ || (0 <= bwh && bwh < btgt))
-+ btgt = -1;
++static inline void au_iigen_dec(struct inode *inode)
++{
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
+
-+ AuDbg("btgt %d\n", btgt);
-+ return btgt;
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ iigen->ig_generation--;
++ spin_unlock(&iigen->ig_spin);
+}
+
-+/* sets src_btop, dst_btop and btgt */
-+static int au_ren_wbr(struct au_ren_args *a)
++static inline int au_iigen_test(struct inode *inode, unsigned int sigen)
+{
+ int err;
-+ struct au_wr_dir_args wr_dir_args = {
-+ /* .force_btgt = -1, */
-+ .flags = AuWrDir_ADD_ENTRY
-+ };
+
-+ a->src_btop = au_dbtop(a->src_dentry);
-+ a->dst_btop = au_dbtop(a->dst_dentry);
-+ if (au_ftest_ren(a->flags, ISDIR))
-+ au_fset_wrdir(wr_dir_args.flags, ISDIR);
-+ wr_dir_args.force_btgt = a->src_btop;
-+ if (a->dst_inode && a->dst_btop < a->src_btop)
-+ wr_dir_args.force_btgt = a->dst_btop;
-+ wr_dir_args.force_btgt = au_wbr(a->dst_dentry, wr_dir_args.force_btgt);
-+ err = au_wr_dir(a->dst_dentry, a->src_dentry, &wr_dir_args);
-+ a->btgt = err;
++ err = 0;
++ if (unlikely(inode && au_iigen(inode, NULL) != sigen))
++ err = -EIO;
+
+ return err;
+}
+
-+static void au_ren_dt(struct au_ren_args *a)
++/* ---------------------------------------------------------------------- */
++
++static inline struct au_hinode *au_hinode(struct au_iinfo *iinfo,
++ aufs_bindex_t bindex)
+{
-+ a->h_path.dentry = a->src_h_parent;
-+ au_dtime_store(a->src_dt + AuPARENT, a->src_parent, &a->h_path);
-+ if (!au_ftest_ren(a->flags, ISSAMEDIR)) {
-+ a->h_path.dentry = a->dst_h_parent;
-+ au_dtime_store(a->dst_dt + AuPARENT, a->dst_parent, &a->h_path);
-+ }
++ return iinfo->ii_hinode + bindex;
++}
+
-+ au_fclr_ren(a->flags, DT_DSTDIR);
-+ if (!au_ftest_ren(a->flags, ISDIR))
-+ return;
++static inline int au_is_bad_inode(struct inode *inode)
++{
++ return !!(is_bad_inode(inode) || !au_hinode(au_ii(inode), 0));
++}
+
-+ a->h_path.dentry = a->src_h_dentry;
-+ au_dtime_store(a->src_dt + AuCHILD, a->src_dentry, &a->h_path);
-+ if (d_is_positive(a->dst_h_dentry)) {
-+ au_fset_ren(a->flags, DT_DSTDIR);
-+ a->h_path.dentry = a->dst_h_dentry;
-+ au_dtime_store(a->dst_dt + AuCHILD, a->dst_dentry, &a->h_path);
-+ }
++static inline aufs_bindex_t au_ii_br_id(struct inode *inode,
++ aufs_bindex_t bindex)
++{
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex)->hi_id;
+}
+
-+static void au_ren_rev_dt(int err, struct au_ren_args *a)
++static inline aufs_bindex_t au_ibtop(struct inode *inode)
+{
-+ struct dentry *h_d;
-+ struct inode *h_inode;
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_btop;
++}
+
-+ au_dtime_revert(a->src_dt + AuPARENT);
-+ if (!au_ftest_ren(a->flags, ISSAMEDIR))
-+ au_dtime_revert(a->dst_dt + AuPARENT);
++static inline aufs_bindex_t au_ibbot(struct inode *inode)
++{
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_bbot;
++}
+
-+ if (au_ftest_ren(a->flags, ISDIR) && err != -EIO) {
-+ h_d = a->src_dt[AuCHILD].dt_h_path.dentry;
-+ h_inode = d_inode(h_d);
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
-+ au_dtime_revert(a->src_dt + AuCHILD);
-+ inode_unlock(h_inode);
++static inline struct au_vdir *au_ivdir(struct inode *inode)
++{
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_vdir;
++}
+
-+ if (au_ftest_ren(a->flags, DT_DSTDIR)) {
-+ h_d = a->dst_dt[AuCHILD].dt_h_path.dentry;
-+ h_inode = d_inode(h_d);
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD);
-+ au_dtime_revert(a->dst_dt + AuCHILD);
-+ inode_unlock(h_inode);
-+ }
-+ }
++static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex)
++{
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex)->hi_whdentry;
++}
++
++static inline void au_set_ibtop(struct inode *inode, aufs_bindex_t bindex)
++{
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_btop = bindex;
++}
++
++static inline void au_set_ibbot(struct inode *inode, aufs_bindex_t bindex)
++{
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_bbot = bindex;
++}
++
++static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir)
++{
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_vdir = vdir;
++}
++
++static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex)
++{
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+int aufs_rename(struct inode *_src_dir, struct dentry *_src_dentry,
-+ struct inode *_dst_dir, struct dentry *_dst_dentry)
++static inline struct dentry *au_pinned_parent(struct au_pin *pin)
+{
-+ int err, flags;
-+ /* reduce stack space */
-+ struct au_ren_args *a;
++ if (pin)
++ return pin->parent;
++ return NULL;
++}
+
-+ AuDbg("%pd, %pd\n", _src_dentry, _dst_dentry);
-+ IMustLock(_src_dir);
-+ IMustLock(_dst_dir);
++static inline struct inode *au_pinned_h_dir(struct au_pin *pin)
++{
++ if (pin && pin->hdir)
++ return pin->hdir->hi_inode;
++ return NULL;
++}
+
-+ err = -ENOMEM;
-+ BUILD_BUG_ON(sizeof(*a) > PAGE_SIZE);
-+ a = kzalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++static inline struct au_hinode *au_pinned_hdir(struct au_pin *pin)
++{
++ if (pin)
++ return pin->hdir;
++ return NULL;
++}
+
-+ a->src_dir = _src_dir;
-+ a->src_dentry = _src_dentry;
-+ a->src_inode = NULL;
-+ if (d_really_is_positive(a->src_dentry))
-+ a->src_inode = d_inode(a->src_dentry);
-+ a->src_parent = a->src_dentry->d_parent; /* dir inode is locked */
-+ a->dst_dir = _dst_dir;
-+ a->dst_dentry = _dst_dentry;
-+ a->dst_inode = NULL;
-+ if (d_really_is_positive(a->dst_dentry))
-+ a->dst_inode = d_inode(a->dst_dentry);
-+ a->dst_parent = a->dst_dentry->d_parent; /* dir inode is locked */
-+ if (a->dst_inode) {
-+ IMustLock(a->dst_inode);
-+ au_igrab(a->dst_inode);
++static inline void au_pin_set_dentry(struct au_pin *pin, struct dentry *dentry)
++{
++ if (pin)
++ pin->dentry = dentry;
++}
++
++static inline void au_pin_set_parent_lflag(struct au_pin *pin,
++ unsigned char lflag)
++{
++ if (pin) {
++ if (lflag)
++ au_fset_pin(pin->flags, DI_LOCKED);
++ else
++ au_fclr_pin(pin->flags, DI_LOCKED);
+ }
++}
+
-+ err = -ENOTDIR;
-+ flags = AuLock_FLUSH | AuLock_NOPLM | AuLock_GEN;
-+ if (d_is_dir(a->src_dentry)) {
-+ au_fset_ren(a->flags, ISDIR);
-+ if (unlikely(d_really_is_positive(a->dst_dentry)
-+ && !d_is_dir(a->dst_dentry)))
-+ goto out_free;
-+ flags |= AuLock_DIRS;
++#if 0 /* reserved */
++static inline void au_pin_set_parent(struct au_pin *pin, struct dentry *parent)
++{
++ if (pin) {
++ dput(pin->parent);
++ pin->parent = dget(parent);
+ }
-+ err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, flags);
-+ if (unlikely(err))
-+ goto out_free;
++}
++#endif
+
-+ err = au_d_hashed_positive(a->src_dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ err = -ENOENT;
-+ if (a->dst_inode) {
-+ /*
-+ * If it is a dir, VFS unhash dst_dentry before this
-+ * function. It means we cannot rely upon d_unhashed().
-+ */
-+ if (unlikely(!a->dst_inode->i_nlink))
-+ goto out_unlock;
-+ if (!S_ISDIR(a->dst_inode->i_mode)) {
-+ err = au_d_hashed_positive(a->dst_dentry);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ } else if (unlikely(IS_DEADDIR(a->dst_inode)))
-+ goto out_unlock;
-+ } else if (unlikely(d_unhashed(a->dst_dentry)))
-+ goto out_unlock;
++/* ---------------------------------------------------------------------- */
++
++struct au_branch;
++#ifdef CONFIG_AUFS_HNOTIFY
++struct au_hnotify_op {
++ void (*ctl)(struct au_hinode *hinode, int do_set);
++ int (*alloc)(struct au_hinode *hinode);
+
+ /*
-+ * is it possible?
-+ * yes, it happened (in linux-3.3-rcN) but I don't know why.
-+ * there may exist a problem somewhere else.
++ * if it returns true, the the caller should free hinode->hi_notify,
++ * otherwise ->free() frees it.
+ */
-+ err = -EINVAL;
-+ if (unlikely(d_inode(a->dst_parent) == d_inode(a->src_dentry)))
-+ goto out_unlock;
++ int (*free)(struct au_hinode *hinode,
++ struct au_hnotify *hn) __must_check;
+
-+ au_fset_ren(a->flags, ISSAMEDIR); /* temporary */
-+ di_write_lock_parent(a->dst_parent);
++ void (*fin)(void);
++ int (*init)(void);
+
-+ /* which branch we process */
-+ err = au_ren_wbr(a);
-+ if (unlikely(err < 0))
-+ goto out_parent;
-+ a->br = au_sbr(a->dst_dentry->d_sb, a->btgt);
-+ a->h_path.mnt = au_br_mnt(a->br);
++ int (*reset_br)(unsigned int udba, struct au_branch *br, int perm);
++ void (*fin_br)(struct au_branch *br);
++ int (*init_br)(struct au_branch *br, int perm);
++};
+
-+ /* are they available to be renamed */
-+ err = au_ren_may_dir(a);
-+ if (unlikely(err))
-+ goto out_children;
++/* hnotify.c */
++int au_hn_alloc(struct au_hinode *hinode, struct inode *inode);
++void au_hn_free(struct au_hinode *hinode);
++void au_hn_ctl(struct au_hinode *hinode, int do_set);
++void au_hn_reset(struct inode *inode, unsigned int flags);
++int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
++ struct qstr *h_child_qstr, struct inode *h_child_inode);
++int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm);
++int au_hnotify_init_br(struct au_branch *br, int perm);
++void au_hnotify_fin_br(struct au_branch *br);
++int __init au_hnotify_init(void);
++void au_hnotify_fin(void);
+
-+ /* prepare the writable parent dir on the same branch */
-+ if (a->dst_btop == a->btgt) {
-+ au_fset_ren(a->flags, WHDST);
-+ } else {
-+ err = au_cpup_dirs(a->dst_dentry, a->btgt);
-+ if (unlikely(err))
-+ goto out_children;
-+ }
++/* hfsnotify.c */
++extern const struct au_hnotify_op au_hnotify_op;
+
-+ if (a->src_dir != a->dst_dir) {
-+ /*
-+ * this temporary unlock is safe,
-+ * because both dir->i_mutex are locked.
-+ */
-+ di_write_unlock(a->dst_parent);
-+ di_write_lock_parent(a->src_parent);
-+ err = au_wr_dir_need_wh(a->src_dentry,
-+ au_ftest_ren(a->flags, ISDIR),
-+ &a->btgt);
-+ di_write_unlock(a->src_parent);
-+ di_write_lock2_parent(a->src_parent, a->dst_parent, /*isdir*/1);
-+ au_fclr_ren(a->flags, ISSAMEDIR);
-+ } else
-+ err = au_wr_dir_need_wh(a->src_dentry,
-+ au_ftest_ren(a->flags, ISDIR),
-+ &a->btgt);
-+ if (unlikely(err < 0))
-+ goto out_children;
-+ if (err)
-+ au_fset_ren(a->flags, WHSRC);
++static inline
++void au_hn_init(struct au_hinode *hinode)
++{
++ hinode->hi_notify = NULL;
++}
+
-+ /* cpup src */
-+ if (a->src_btop != a->btgt) {
-+ struct au_pin pin;
++static inline struct au_hnotify *au_hn(struct au_hinode *hinode)
++{
++ return hinode->hi_notify;
++}
+
-+ err = au_pin(&pin, a->src_dentry, a->btgt,
-+ au_opt_udba(a->src_dentry->d_sb),
-+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
-+ if (!err) {
-+ struct au_cp_generic cpg = {
-+ .dentry = a->src_dentry,
-+ .bdst = a->btgt,
-+ .bsrc = a->src_btop,
-+ .len = -1,
-+ .pin = &pin,
-+ .flags = AuCpup_DTIME | AuCpup_HOPEN
-+ };
-+ AuDebugOn(au_dbtop(a->src_dentry) != a->src_btop);
-+ err = au_sio_cpup_simple(&cpg);
-+ au_unpin(&pin);
-+ }
-+ if (unlikely(err))
-+ goto out_children;
-+ a->src_btop = a->btgt;
-+ a->src_h_dentry = au_h_dptr(a->src_dentry, a->btgt);
-+ au_fset_ren(a->flags, WHSRC);
-+ }
++#else
++AuStub(int, au_hn_alloc, return -EOPNOTSUPP,
++ struct au_hinode *hinode __maybe_unused,
++ struct inode *inode __maybe_unused)
++AuStub(struct au_hnotify *, au_hn, return NULL, struct au_hinode *hinode)
++AuStubVoid(au_hn_free, struct au_hinode *hinode __maybe_unused)
++AuStubVoid(au_hn_ctl, struct au_hinode *hinode __maybe_unused,
++ int do_set __maybe_unused)
++AuStubVoid(au_hn_reset, struct inode *inode __maybe_unused,
++ unsigned int flags __maybe_unused)
++AuStubInt0(au_hnotify_reset_br, unsigned int udba __maybe_unused,
++ struct au_branch *br __maybe_unused,
++ int perm __maybe_unused)
++AuStubInt0(au_hnotify_init_br, struct au_branch *br __maybe_unused,
++ int perm __maybe_unused)
++AuStubVoid(au_hnotify_fin_br, struct au_branch *br __maybe_unused)
++AuStubInt0(__init au_hnotify_init, void)
++AuStubVoid(au_hnotify_fin, void)
++AuStubVoid(au_hn_init, struct au_hinode *hinode __maybe_unused)
++#endif /* CONFIG_AUFS_HNOTIFY */
+
-+ /* lock them all */
-+ err = au_ren_lock(a);
-+ if (unlikely(err))
-+ /* leave the copied-up one */
-+ goto out_children;
++static inline void au_hn_suspend(struct au_hinode *hdir)
++{
++ au_hn_ctl(hdir, /*do_set*/0);
++}
+
-+ if (!au_opt_test(au_mntflags(a->dst_dir->i_sb), UDBA_NONE))
-+ err = au_may_ren(a);
-+ else if (unlikely(a->dst_dentry->d_name.len > AUFS_MAX_NAMELEN))
-+ err = -ENAMETOOLONG;
-+ if (unlikely(err))
-+ goto out_hdir;
++static inline void au_hn_resume(struct au_hinode *hdir)
++{
++ au_hn_ctl(hdir, /*do_set*/1);
++}
+
-+ /* store timestamps to be revertible */
-+ au_ren_dt(a);
++static inline void au_hn_inode_lock(struct au_hinode *hdir)
++{
++ inode_lock(hdir->hi_inode);
++ au_hn_suspend(hdir);
++}
+
-+ /* here we go */
-+ err = do_rename(a);
-+ if (unlikely(err))
-+ goto out_dt;
++static inline void au_hn_inode_lock_nested(struct au_hinode *hdir,
++ unsigned int sc __maybe_unused)
++{
++ inode_lock_nested(hdir->hi_inode, sc);
++ au_hn_suspend(hdir);
++}
+
-+ /* update dir attributes */
-+ au_ren_refresh_dir(a);
++#if 0 /* unused */
++#include "vfsub.h"
++static inline void au_hn_inode_lock_shared_nested(struct au_hinode *hdir,
++ unsigned int sc)
++{
++ vfsub_inode_lock_shared_nested(hdir->hi_inode, sc);
++ au_hn_suspend(hdir);
++}
++#endif
+
-+ /* dput/iput all lower dentries */
-+ au_ren_refresh(a);
++static inline void au_hn_inode_unlock(struct au_hinode *hdir)
++{
++ au_hn_resume(hdir);
++ inode_unlock(hdir->hi_inode);
++}
+
-+ goto out_hdir; /* success */
++#endif /* __KERNEL__ */
++#endif /* __AUFS_INODE_H__ */
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/ioctl.c linux-4.14/fs/aufs/ioctl.c
+--- linux-4.14/fs/aufs/ioctl.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/ioctl.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,219 @@
++/*
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
++ *
++ * This program, aufs is free software; you can redistribute it and/or modify
++ * it under the terms of the GNU General Public License as published by
++ * the Free Software Foundation; either version 2 of the License, or
++ * (at your option) any later version.
++ *
++ * This program is distributed in the hope that it will be useful,
++ * but WITHOUT ANY WARRANTY; without even the implied warranty of
++ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
++ * GNU General Public License for more details.
++ *
++ * You should have received a copy of the GNU General Public License
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
++ */
+
-+out_dt:
-+ au_ren_rev_dt(err, a);
-+out_hdir:
-+ au_ren_unlock(a);
-+out_children:
-+ au_nhash_wh_free(&a->whlist);
-+ if (err && a->dst_inode && a->dst_btop != a->btgt) {
-+ AuDbg("btop %d, btgt %d\n", a->dst_btop, a->btgt);
-+ au_set_h_dptr(a->dst_dentry, a->btgt, NULL);
-+ au_set_dbtop(a->dst_dentry, a->dst_btop);
++/*
++ * ioctl
++ * plink-management and readdir in userspace.
++ * assist the pathconf(3) wrapper library.
++ * move-down
++ * File-based Hierarchical Storage Management.
++ */
++
++#include <linux/compat.h>
++#include <linux/file.h>
++#include "aufs.h"
++
++static int au_wbr_fd(struct path *path, struct aufs_wbr_fd __user *arg)
++{
++ int err, fd;
++ aufs_bindex_t wbi, bindex, bbot;
++ struct file *h_file;
++ struct super_block *sb;
++ struct dentry *root;
++ struct au_branch *br;
++ struct aufs_wbr_fd wbrfd = {
++ .oflags = au_dir_roflags,
++ .brid = -1
++ };
++ const int valid = O_RDONLY | O_NONBLOCK | O_LARGEFILE | O_DIRECTORY
++ | O_NOATIME | O_CLOEXEC;
++
++ AuDebugOn(wbrfd.oflags & ~valid);
++
++ if (arg) {
++ err = copy_from_user(&wbrfd, arg, sizeof(wbrfd));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ goto out;
++ }
++
++ err = -EINVAL;
++ AuDbg("wbrfd{0%o, %d}\n", wbrfd.oflags, wbrfd.brid);
++ wbrfd.oflags |= au_dir_roflags;
++ AuDbg("0%o\n", wbrfd.oflags);
++ if (unlikely(wbrfd.oflags & ~valid))
++ goto out;
+ }
-+out_parent:
-+ if (!err)
-+ d_move(a->src_dentry, a->dst_dentry);
-+ else {
-+ au_update_dbtop(a->dst_dentry);
-+ if (!a->dst_inode)
-+ d_drop(a->dst_dentry);
++
++ fd = get_unused_fd_flags(0);
++ err = fd;
++ if (unlikely(fd < 0))
++ goto out;
++
++ h_file = ERR_PTR(-EINVAL);
++ wbi = 0;
++ br = NULL;
++ sb = path->dentry->d_sb;
++ root = sb->s_root;
++ aufs_read_lock(root, AuLock_IR);
++ bbot = au_sbbot(sb);
++ if (wbrfd.brid >= 0) {
++ wbi = au_br_index(sb, wbrfd.brid);
++ if (unlikely(wbi < 0 || wbi > bbot))
++ goto out_unlock;
+ }
-+ if (au_ftest_ren(a->flags, ISSAMEDIR))
-+ di_write_unlock(a->dst_parent);
-+ else
-+ di_write_unlock2(a->src_parent, a->dst_parent);
++
++ h_file = ERR_PTR(-ENOENT);
++ br = au_sbr(sb, wbi);
++ if (!au_br_writable(br->br_perm)) {
++ if (arg)
++ goto out_unlock;
++
++ bindex = wbi + 1;
++ wbi = -1;
++ for (; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_writable(br->br_perm)) {
++ wbi = bindex;
++ br = au_sbr(sb, wbi);
++ break;
++ }
++ }
++ }
++ AuDbg("wbi %d\n", wbi);
++ if (wbi >= 0)
++ h_file = au_h_open(root, wbi, wbrfd.oflags, NULL,
++ /*force_wr*/0);
++
+out_unlock:
-+ aufs_read_and_write_unlock2(a->dst_dentry, a->src_dentry);
-+out_free:
-+ iput(a->dst_inode);
-+ if (a->thargs)
-+ au_whtmp_rmdir_free(a->thargs);
-+ au_delayed_kfree(a);
++ aufs_read_unlock(root, AuLock_IR);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out_fd;
++
++ au_br_put(br); /* cf. au_h_open() */
++ fd_install(fd, h_file);
++ err = fd;
++ goto out; /* success */
++
++out_fd:
++ put_unused_fd(fd);
+out:
+ AuTraceErr(err);
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/Kconfig linux/fs/aufs/Kconfig
---- /usr/share/empty/fs/aufs/Kconfig 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/Kconfig 2015-09-24 10:47:58.248052907 +0200
-@@ -0,0 +1,185 @@
-+config AUFS_FS
-+ tristate "Aufs (Advanced multi layered unification filesystem) support"
-+ help
-+ Aufs is a stackable unification filesystem such as Unionfs,
-+ which unifies several directories and provides a merged single
-+ directory.
-+ In the early days, aufs was entirely re-designed and
-+ re-implemented Unionfs Version 1.x series. Introducing many
-+ original ideas, approaches and improvements, it becomes totally
-+ different from Unionfs while keeping the basic features.
+
-+if AUFS_FS
-+choice
-+ prompt "Maximum number of branches"
-+ default AUFS_BRANCH_MAX_127
-+ help
-+ Specifies the maximum number of branches (or member directories)
-+ in a single aufs. The larger value consumes more system
-+ resources and has a minor impact to performance.
-+config AUFS_BRANCH_MAX_127
-+ bool "127"
-+ help
-+ Specifies the maximum number of branches (or member directories)
-+ in a single aufs. The larger value consumes more system
-+ resources and has a minor impact to performance.
-+config AUFS_BRANCH_MAX_511
-+ bool "511"
-+ help
-+ Specifies the maximum number of branches (or member directories)
-+ in a single aufs. The larger value consumes more system
-+ resources and has a minor impact to performance.
-+config AUFS_BRANCH_MAX_1023
-+ bool "1023"
-+ help
-+ Specifies the maximum number of branches (or member directories)
-+ in a single aufs. The larger value consumes more system
-+ resources and has a minor impact to performance.
-+config AUFS_BRANCH_MAX_32767
-+ bool "32767"
-+ help
-+ Specifies the maximum number of branches (or member directories)
-+ in a single aufs. The larger value consumes more system
-+ resources and has a minor impact to performance.
-+endchoice
++/* ---------------------------------------------------------------------- */
+
-+config AUFS_SBILIST
-+ bool
-+ depends on AUFS_MAGIC_SYSRQ || PROC_FS
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports Magic SysRq or /proc, enabled automatically.
++long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg)
++{
++ long err;
++ struct dentry *dentry;
+
-+config AUFS_HNOTIFY
-+ bool "Detect direct branch access (bypassing aufs)"
-+ help
-+ If you want to modify files on branches directly, eg. bypassing aufs,
-+ and want aufs to detect the changes of them fully, then enable this
-+ option and use 'udba=notify' mount option.
-+ Currently there is only one available configuration, "fsnotify".
-+ It will have a negative impact to the performance.
-+ See detail in aufs.5.
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_ioctl(file, cmd, arg);
++ break;
+
-+choice
-+ prompt "method" if AUFS_HNOTIFY
-+ default AUFS_HFSNOTIFY
-+config AUFS_HFSNOTIFY
-+ bool "fsnotify"
-+ select FSNOTIFY
-+endchoice
++ case AUFS_CTL_WBR_FD:
++ err = au_wbr_fd(&file->f_path, (void __user *)arg);
++ break;
+
-+config AUFS_EXPORT
-+ bool "NFS-exportable aufs"
-+ depends on EXPORTFS
-+ help
-+ If you want to export your mounted aufs via NFS, then enable this
-+ option. There are several requirements for this configuration.
-+ See detail in aufs.5.
++ case AUFS_CTL_IBUSY:
++ err = au_ibusy_ioctl(file, arg);
++ break;
+
-+config AUFS_INO_T_64
-+ bool
-+ depends on AUFS_EXPORT
-+ depends on 64BIT && !(ALPHA || S390)
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ /* typedef unsigned long/int __kernel_ino_t */
-+ /* alpha and s390x are int */
++ case AUFS_CTL_BRINFO:
++ err = au_brinfo_ioctl(file, arg);
++ break;
+
-+config AUFS_XATTR
-+ bool "support for XATTR/EA (including Security Labels)"
-+ help
-+ If your branch fs supports XATTR/EA and you want to make them
-+ available in aufs too, then enable this opsion and specify the
-+ branch attributes for EA.
-+ See detail in aufs.5.
++ case AUFS_CTL_FHSM_FD:
++ dentry = file->f_path.dentry;
++ if (IS_ROOT(dentry))
++ err = au_fhsm_fd(dentry->d_sb, arg);
++ else
++ err = -ENOTTY;
++ break;
+
-+config AUFS_FHSM
-+ bool "File-based Hierarchical Storage Management"
-+ help
-+ Hierarchical Storage Management (or HSM) is a well-known feature
-+ in the storage world. Aufs provides this feature as file-based.
-+ with multiple branches.
-+ These multiple branches are prioritized, ie. the topmost one
-+ should be the fastest drive and be used heavily.
++ default:
++ /* do not call the lower */
++ AuDbg("0x%x\n", cmd);
++ err = -ENOTTY;
++ }
+
-+config AUFS_RDU
-+ bool "Readdir in userspace"
-+ help
-+ Aufs has two methods to provide a merged view for a directory,
-+ by a user-space library and by kernel-space natively. The latter
-+ is always enabled but sometimes large and slow.
-+ If you enable this option, install the library in aufs2-util
-+ package, and set some environment variables for your readdir(3),
-+ then the work will be handled in user-space which generally
-+ shows better performance in most cases.
-+ See detail in aufs.5.
++ AuTraceErr(err);
++ return err;
++}
+
-+config AUFS_SHWH
-+ bool "Show whiteouts"
-+ help
-+ If you want to make the whiteouts in aufs visible, then enable
-+ this option and specify 'shwh' mount option. Although it may
-+ sounds like philosophy or something, but in technically it
-+ simply shows the name of whiteout with keeping its behaviour.
++long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg)
++{
++ long err;
+
-+config AUFS_BR_RAMFS
-+ bool "Ramfs (initramfs/rootfs) as an aufs branch"
-+ help
-+ If you want to use ramfs as an aufs branch fs, then enable this
-+ option. Generally tmpfs is recommended.
-+ Aufs prohibited them to be a branch fs by default, because
-+ initramfs becomes unusable after switch_root or something
-+ generally. If you sets initramfs as an aufs branch and boot your
-+ system by switch_root, you will meet a problem easily since the
-+ files in initramfs may be inaccessible.
-+ Unless you are going to use ramfs as an aufs branch fs without
-+ switch_root or something, leave it N.
++ switch (cmd) {
++ case AUFS_CTL_MVDOWN:
++ err = au_mvdown(file->f_path.dentry, (void __user *)arg);
++ break;
+
-+config AUFS_BR_FUSE
-+ bool "Fuse fs as an aufs branch"
-+ depends on FUSE_FS
-+ select AUFS_POLL
-+ help
-+ If you want to use fuse-based userspace filesystem as an aufs
-+ branch fs, then enable this option.
-+ It implements the internal poll(2) operation which is
-+ implemented by fuse only (curretnly).
++ case AUFS_CTL_WBR_FD:
++ err = au_wbr_fd(&file->f_path, (void __user *)arg);
++ break;
+
-+config AUFS_POLL
-+ bool
-+ help
-+ Automatic configuration for internal use.
++ default:
++ /* do not call the lower */
++ AuDbg("0x%x\n", cmd);
++ err = -ENOTTY;
++ }
+
-+config AUFS_BR_HFSPLUS
-+ bool "Hfsplus as an aufs branch"
-+ depends on HFSPLUS_FS
-+ default y
-+ help
-+ If you want to use hfsplus fs as an aufs branch fs, then enable
-+ this option. This option introduces a small overhead at
-+ copying-up a file on hfsplus.
++ AuTraceErr(err);
++ return err;
++}
+
-+config AUFS_BDEV_LOOP
-+ bool
-+ depends on BLK_DEV_LOOP
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ Convert =[ym] into =y.
++#ifdef CONFIG_COMPAT
++long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
++ unsigned long arg)
++{
++ long err;
+
-+config AUFS_DEBUG
-+ bool "Debug aufs"
-+ help
-+ Enable this to compile aufs internal debug code.
-+ It will have a negative impact to the performance.
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_compat_ioctl(file, cmd, arg);
++ break;
+
-+config AUFS_MAGIC_SYSRQ
-+ bool
-+ depends on AUFS_DEBUG && MAGIC_SYSRQ
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports Magic SysRq, enabled automatically.
-+endif
-diff -urN /usr/share/empty/fs/aufs/loop.c linux/fs/aufs/loop.c
---- /usr/share/empty/fs/aufs/loop.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/loop.c 2016-08-17 18:01:06.151889691 +0200
-@@ -0,0 +1,146 @@
++ case AUFS_CTL_IBUSY:
++ err = au_ibusy_compat_ioctl(file, arg);
++ break;
++
++ case AUFS_CTL_BRINFO:
++ err = au_brinfo_compat_ioctl(file, arg);
++ break;
++
++ default:
++ err = aufs_ioctl_dir(file, cmd, arg);
++ }
++
++ AuTraceErr(err);
++ return err;
++}
++
++long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
++ unsigned long arg)
++{
++ return aufs_ioctl_nondir(file, cmd, (unsigned long)compat_ptr(arg));
++}
++#endif
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/loop.c linux-4.14/fs/aufs/loop.c
+--- linux-4.14/fs/aufs/loop.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/loop.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,163 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ new_nelem = au_warn_loopback_nelem + au_warn_loopback_step;
+ a = au_kzrealloc(au_warn_loopback_array,
+ au_warn_loopback_nelem * sizeof(unsigned long),
-+ new_nelem * sizeof(unsigned long), GFP_ATOMIC);
++ new_nelem * sizeof(unsigned long), GFP_ATOMIC,
++ /*may_shrink*/0);
+ if (a) {
+ au_warn_loopback_nelem = new_nelem;
+ au_warn_loopback_array = a;
+{
+ if (backing_file_func)
+ symbol_put(loop_backing_file);
-+ au_delayed_kfree(au_warn_loopback_array);
++ kfree(au_warn_loopback_array);
+}
-diff -urN /usr/share/empty/fs/aufs/loop.h linux/fs/aufs/loop.h
---- /usr/share/empty/fs/aufs/loop.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/loop.h 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,52 @@
++
++/* ---------------------------------------------------------------------- */
++
++/* support the loopback block device insude aufs */
++
++struct file *aufs_real_loop(struct file *file)
++{
++ struct file *f;
++
++ BUG_ON(!au_test_aufs(file->f_path.dentry->d_sb));
++ fi_read_lock(file);
++ f = au_hf_top(file);
++ fi_read_unlock(file);
++ AuDebugOn(!f);
++ return f;
++}
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/loop.h linux-4.14/fs/aufs/loop.h
+--- linux-4.14/fs/aufs/loop.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/loop.h 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,58 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+int au_loopback_init(void);
+void au_loopback_fin(void);
++
++struct file *aufs_real_loop(struct file *file);
+#else
++AuStub(struct file *, loop_backing_file, return NULL)
++
+AuStubInt0(au_test_loopback_overlap, struct super_block *sb,
+ struct dentry *h_adding)
+AuStubInt0(au_test_loopback_kthread, void)
+
+AuStubInt0(au_loopback_init, void)
+AuStubVoid(au_loopback_fin, void)
++
++AuStub(struct file *, aufs_real_loop, return NULL, struct file *file)
+#endif /* BLK_DEV_LOOP */
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_LOOP_H__ */
-diff -urN /usr/share/empty/fs/aufs/magic.mk linux/fs/aufs/magic.mk
---- /usr/share/empty/fs/aufs/magic.mk 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/magic.mk 2015-09-24 10:47:58.254719746 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/magic.mk linux-4.14/fs/aufs/magic.mk
+--- linux-4.14/fs/aufs/magic.mk 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/magic.mk 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,30 @@
+
+# defined in ${srctree}/fs/fuse/inode.c
+ifdef CONFIG_HFSPLUS_FS
+ccflags-y += -DHFSPLUS_SUPER_MAGIC=0x482b
+endif
-diff -urN /usr/share/empty/fs/aufs/Makefile linux/fs/aufs/Makefile
---- /usr/share/empty/fs/aufs/Makefile 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/Makefile 2015-09-24 10:47:58.248052907 +0200
-@@ -0,0 +1,44 @@
-+
-+include ${src}/magic.mk
-+ifeq (${CONFIG_AUFS_FS},m)
-+include ${src}/conf.mk
-+endif
-+-include ${src}/priv_def.mk
-+
-+# cf. include/linux/kernel.h
-+# enable pr_debug
-+ccflags-y += -DDEBUG
-+# sparse requires the full pathname
-+ifdef M
-+ccflags-y += -include ${M}/../../include/uapi/linux/aufs_type.h
-+else
-+ccflags-y += -include ${srctree}/include/uapi/linux/aufs_type.h
-+endif
-+
-+obj-$(CONFIG_AUFS_FS) += aufs.o
-+aufs-y := module.o sbinfo.o super.o branch.o xino.o sysaufs.o opts.o \
-+ wkq.o vfsub.o dcsub.o \
-+ cpup.o whout.o wbr_policy.o \
-+ dinfo.o dentry.o \
-+ dynop.o \
-+ finfo.o file.o f_op.o \
-+ dir.o vdir.o \
-+ iinfo.o inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o \
-+ mvdown.o ioctl.o
-+
-+# all are boolean
-+aufs-$(CONFIG_PROC_FS) += procfs.o plink.o
-+aufs-$(CONFIG_SYSFS) += sysfs.o
-+aufs-$(CONFIG_DEBUG_FS) += dbgaufs.o
-+aufs-$(CONFIG_AUFS_BDEV_LOOP) += loop.o
-+aufs-$(CONFIG_AUFS_HNOTIFY) += hnotify.o
-+aufs-$(CONFIG_AUFS_HFSNOTIFY) += hfsnotify.o
-+aufs-$(CONFIG_AUFS_EXPORT) += export.o
-+aufs-$(CONFIG_AUFS_XATTR) += xattr.o
-+aufs-$(CONFIG_FS_POSIX_ACL) += posix_acl.o
-+aufs-$(CONFIG_AUFS_FHSM) += fhsm.o
-+aufs-$(CONFIG_AUFS_POLL) += poll.o
-+aufs-$(CONFIG_AUFS_RDU) += rdu.o
-+aufs-$(CONFIG_AUFS_BR_HFSPLUS) += hfsplus.o
-+aufs-$(CONFIG_AUFS_DEBUG) += debug.o
-+aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o
-diff -urN /usr/share/empty/fs/aufs/module.c linux/fs/aufs/module.c
---- /usr/share/empty/fs/aufs/module.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/module.c 2016-08-17 18:01:06.151889691 +0200
-@@ -0,0 +1,289 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/module.c linux-4.14/fs/aufs/module.c
+--- linux-4.14/fs/aufs/module.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/module.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,266 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#include <linux/seq_file.h>
+#include "aufs.h"
+
-+void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp)
++/* shrinkable realloc */
++void *au_krealloc(void *p, unsigned int new_sz, gfp_t gfp, int may_shrink)
+{
-+ if (new_sz <= nused)
-+ return p;
++ size_t sz;
++ int diff;
+
-+ p = krealloc(p, new_sz, gfp);
-+ if (p)
++ sz = 0;
++ diff = -1;
++ if (p) {
++#if 0 /* unused */
++ if (!new_sz) {
++ kfree(p);
++ p = NULL;
++ goto out;
++ }
++#else
++ AuDebugOn(!new_sz);
++#endif
++ sz = ksize(p);
++ diff = au_kmidx_sub(sz, new_sz);
++ }
++ if (sz && !diff)
++ goto out;
++
++ if (sz < new_sz)
++ /* expand or SLOB */
++ p = krealloc(p, new_sz, gfp);
++ else if (new_sz < sz && may_shrink) {
++ /* shrink */
++ void *q;
++
++ q = kmalloc(new_sz, gfp);
++ if (q) {
++ if (p) {
++ memcpy(q, p, new_sz);
++ kfree(p);
++ }
++ p = q;
++ } else
++ p = NULL;
++ }
++
++out:
++ return p;
++}
++
++void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp,
++ int may_shrink)
++{
++ p = au_krealloc(p, new_sz, gfp, may_shrink);
++ if (p && new_sz > nused)
+ memset(p + nused, 0, new_sz - nused);
+ return p;
+}
+/*
+ * aufs caches
+ */
-+
-+struct au_dfree au_dfree;
-+
-+/* delayed free */
-+static void au_do_dfree(struct work_struct *work __maybe_unused)
-+{
-+ struct llist_head *head;
-+ struct llist_node *node, *next;
-+
-+#define AU_CACHE_DFREE_DO_BODY(name, idx, lnode) do { \
-+ head = &au_dfree.cache[AuCache_##idx].llist; \
-+ node = llist_del_all(head); \
-+ for (; node; node = next) { \
-+ struct au_##name *p = \
-+ p = llist_entry(node, struct au_##name, \
-+ lnode); \
-+ next = llist_next(node); \
-+ au_cache_free_##name(p); \
-+ } \
-+ } while (0)
-+
-+ AU_CACHE_DFREE_DO_BODY(dinfo, DINFO, di_lnode);
-+ AU_CACHE_DFREE_DO_BODY(icntnr, ICNTNR, lnode);
-+ AU_CACHE_DFREE_DO_BODY(finfo, FINFO, fi_lnode);
-+ AU_CACHE_DFREE_DO_BODY(vdir, VDIR, vd_lnode);
-+ AU_CACHE_DFREE_DO_BODY(vdir_dehstr, DEHSTR, lnode);
-+#ifdef CONFIG_AUFS_HNOTIFY
-+ AU_CACHE_DFREE_DO_BODY(hnotify, HNOTIFY, hn_lnode);
-+#endif
-+
-+#define AU_DFREE_DO_BODY(llist, func) do { \
-+ node = llist_del_all(llist); \
-+ for (; node; node = next) { \
-+ next = llist_next(node); \
-+ func(node); \
-+ } \
-+ } while (0)
-+
-+ AU_DFREE_DO_BODY(au_dfree.llist + AU_DFREE_KFREE, kfree);
-+ AU_DFREE_DO_BODY(au_dfree.llist + AU_DFREE_FREE_PAGE, au_free_page);
-+
-+#undef AU_CACHE_DFREE_DO_BODY
-+#undef AU_DFREE_DO_BODY
-+}
-+
-+AU_CACHE_DFREE_FUNC(dinfo, DINFO, di_lnode);
-+AU_CACHE_DFREE_FUNC(icntnr, ICNTNR, lnode);
-+AU_CACHE_DFREE_FUNC(finfo, FINFO, fi_lnode);
-+AU_CACHE_DFREE_FUNC(vdir, VDIR, vd_lnode);
-+AU_CACHE_DFREE_FUNC(vdir_dehstr, DEHSTR, lnode);
++struct kmem_cache *au_cache[AuCache_Last];
+
+static void au_cache_fin(void)
+{
+ int i;
-+ struct au_cache *cp;
+
+ /*
+ * Make sure all delayed rcu free inodes are flushed before we
+
+ /* excluding AuCache_HNOTIFY */
+ BUILD_BUG_ON(AuCache_HNOTIFY + 1 != AuCache_Last);
-+ flush_delayed_work(&au_dfree.dwork);
+ for (i = 0; i < AuCache_HNOTIFY; i++) {
-+ cp = au_dfree.cache + i;
-+ AuDebugOn(!llist_empty(&cp->llist));
-+ kmem_cache_destroy(cp->cache);
-+ cp->cache = NULL;
++ kmem_cache_destroy(au_cache[i]);
++ au_cache[i] = NULL;
+ }
+}
+
+static int __init au_cache_init(void)
+{
-+ struct au_cache *cp;
-+
-+ cp = au_dfree.cache;
-+ cp[AuCache_DINFO].cache = AuCacheCtor(au_dinfo, au_di_init_once);
-+ if (cp[AuCache_DINFO].cache)
++ au_cache[AuCache_DINFO] = AuCacheCtor(au_dinfo, au_di_init_once);
++ if (au_cache[AuCache_DINFO])
+ /* SLAB_DESTROY_BY_RCU */
-+ cp[AuCache_ICNTNR].cache = AuCacheCtor(au_icntnr,
++ au_cache[AuCache_ICNTNR] = AuCacheCtor(au_icntnr,
+ au_icntnr_init_once);
-+ if (cp[AuCache_ICNTNR].cache)
-+ cp[AuCache_FINFO].cache = AuCacheCtor(au_finfo,
++ if (au_cache[AuCache_ICNTNR])
++ au_cache[AuCache_FINFO] = AuCacheCtor(au_finfo,
+ au_fi_init_once);
-+ if (cp[AuCache_FINFO].cache)
-+ cp[AuCache_VDIR].cache = AuCache(au_vdir);
-+ if (cp[AuCache_VDIR].cache)
-+ cp[AuCache_DEHSTR].cache = AuCache(au_vdir_dehstr);
-+ if (cp[AuCache_DEHSTR].cache)
++ if (au_cache[AuCache_FINFO])
++ au_cache[AuCache_VDIR] = AuCache(au_vdir);
++ if (au_cache[AuCache_VDIR])
++ au_cache[AuCache_DEHSTR] = AuCache(au_vdir_dehstr);
++ if (au_cache[AuCache_DEHSTR])
+ return 0;
+
+ au_cache_fin();
+ * iterate_supers_type() doesn't protect us from
+ * remounting (branch management)
+ */
-+struct au_sphlhead au_sbilist;
++struct hlist_bl_head au_sbilist;
+#endif
+
+/*
+ int err;
+
+ err = seq_path(seq, path, au_esc_chars);
-+ if (err > 0)
++ if (err >= 0)
+ err = 0;
-+ else if (err < 0)
++ else
+ err = -ENOMEM;
+
+ return err;
+{
+ int err, i;
+ char *p;
-+ struct au_cache *cp;
+
+ p = au_esc_chars;
+ for (i = 1; i <= ' '; i++)
+ for (i = 0; i < AuIop_Last; i++)
+ aufs_iop_nogetattr[i].getattr = NULL;
+
-+ /* First, initialize au_dfree */
-+ for (i = 0; i < AuCache_Last; i++) { /* including hnotify */
-+ cp = au_dfree.cache + i;
-+ cp->cache = NULL;
-+ init_llist_head(&cp->llist);
-+ }
-+ for (i = 0; i < AU_DFREE_Last; i++)
-+ init_llist_head(au_dfree.llist + i);
-+ INIT_DELAYED_WORK(&au_dfree.dwork, au_do_dfree);
++ memset(au_cache, 0, sizeof(au_cache)); /* including hnotify */
+
+ au_sbilist_init();
+ sysaufs_brs_init();
+out_sysaufs:
+ sysaufs_fin();
+ au_dy_fin();
-+ flush_delayed_work(&au_dfree.dwork);
+out:
+ return err;
+}
+ au_procfs_fin();
+ sysaufs_fin();
+ au_dy_fin();
-+ flush_delayed_work(&au_dfree.dwork);
+}
+
+module_init(aufs_init);
+module_exit(aufs_exit);
-diff -urN /usr/share/empty/fs/aufs/module.h linux/fs/aufs/module.h
---- /usr/share/empty/fs/aufs/module.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/module.h 2016-08-17 18:01:06.158556531 +0200
-@@ -0,0 +1,144 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/module.h linux-4.14/fs/aufs/module.h
+--- linux-4.14/fs/aufs/module.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/module.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,101 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#ifdef __KERNEL__
+
+#include <linux/slab.h>
-+#include "debug.h"
+
+struct path;
+struct seq_file;
+
+extern int au_dir_roflags;
+
-+void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp);
++void *au_krealloc(void *p, unsigned int new_sz, gfp_t gfp, int may_shrink);
++void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp,
++ int may_shrink);
++
++static inline int au_kmidx_sub(size_t sz, size_t new_sz)
++{
++#ifndef CONFIG_SLOB
++ return kmalloc_index(sz) - kmalloc_index(new_sz);
++#else
++ return -1; /* SLOB is untested */
++#endif
++}
++
+int au_seq_path(struct seq_file *seq, struct path *path);
+
+#ifdef CONFIG_PROC_FS
+
+/* ---------------------------------------------------------------------- */
+
-+/* kmem cache and delayed free */
++/* kmem cache */
+enum {
+ AuCache_DINFO,
+ AuCache_ICNTNR,
+ AuCache_Last
+};
+
-+enum {
-+ AU_DFREE_KFREE,
-+ AU_DFREE_FREE_PAGE,
-+ AU_DFREE_Last
-+};
-+
-+struct au_cache {
-+ struct kmem_cache *cache;
-+ struct llist_head llist; /* delayed free */
-+};
-+
-+/*
-+ * in order to reduce the cost of the internal timer, consolidate all the
-+ * delayed free works into a single delayed_work.
-+ */
-+struct au_dfree {
-+ struct au_cache cache[AuCache_Last];
-+ struct llist_head llist[AU_DFREE_Last];
-+ struct delayed_work dwork;
-+};
-+
-+extern struct au_dfree au_dfree;
++extern struct kmem_cache *au_cache[AuCache_Last];
+
+#define AuCacheFlags (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD)
+#define AuCache(type) KMEM_CACHE(type, AuCacheFlags)
+ kmem_cache_create(#type, sizeof(struct type), \
+ __alignof__(struct type), AuCacheFlags, ctor)
+
-+#define AU_DFREE_DELAY msecs_to_jiffies(10)
-+#define AU_DFREE_BODY(lnode, llist) do { \
-+ if (llist_add(lnode, llist)) \
-+ schedule_delayed_work(&au_dfree.dwork, \
-+ AU_DFREE_DELAY); \
-+ } while (0)
-+#define AU_CACHE_DFREE_FUNC(name, idx, lnode) \
-+ void au_cache_dfree_##name(struct au_##name *p) \
-+ { \
-+ struct au_cache *cp = au_dfree.cache + AuCache_##idx; \
-+ AU_DFREE_BODY(&p->lnode, &cp->llist); \
-+ }
-+
+#define AuCacheFuncs(name, index) \
+static inline struct au_##name *au_cache_alloc_##name(void) \
-+{ return kmem_cache_alloc(au_dfree.cache[AuCache_##index].cache, GFP_NOFS); } \
++{ return kmem_cache_alloc(au_cache[AuCache_##index], GFP_NOFS); } \
+static inline void au_cache_free_##name(struct au_##name *p) \
-+{ kmem_cache_free(au_dfree.cache[AuCache_##index].cache, p); } \
-+void au_cache_dfree_##name(struct au_##name *p)
++{ kmem_cache_free(au_cache[AuCache_##index], p); }
+
+AuCacheFuncs(dinfo, DINFO);
+AuCacheFuncs(icntnr, ICNTNR);
+AuCacheFuncs(hnotify, HNOTIFY);
+#endif
+
-+static inline void au_delayed_kfree(const void *p)
-+{
-+ AuDebugOn(!p);
-+ AuDebugOn(ksize(p) < sizeof(struct llist_node));
-+
-+ AU_DFREE_BODY((void *)p, au_dfree.llist + AU_DFREE_KFREE);
-+}
-+
-+/* cast only */
-+static inline void au_free_page(void *p)
-+{
-+ free_page((unsigned long)p);
-+}
-+
-+static inline void au_delayed_free_page(unsigned long addr)
-+{
-+ AU_DFREE_BODY((void *)addr, au_dfree.llist + AU_DFREE_FREE_PAGE);
-+}
-+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_MODULE_H__ */
-diff -urN /usr/share/empty/fs/aufs/mvdown.c linux/fs/aufs/mvdown.c
---- /usr/share/empty/fs/aufs/mvdown.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/mvdown.c 2016-08-17 18:01:06.158556531 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/mvdown.c linux-4.14/fs/aufs/mvdown.c
+--- linux-4.14/fs/aufs/mvdown.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/mvdown.c 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,704 @@
+/*
-+ * Copyright (C) 2011-2016 Junjiro R. Okajima
++ * Copyright (C) 2011-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ for (bindex++; bindex <= bbot; bindex++) {
+ br = au_sbr(sb, bindex);
+ if (au_br_fhsm(br->br_perm)
-+ && (!(au_br_sb(br)->s_flags & MS_RDONLY)))
++ && !sb_rdonly(au_br_sb(br)))
+ return bindex;
+ }
+ else if (!(a->mvdown.flags & AUFS_MVDOWN_ROLOWER))
+ else
+ for (bindex++; bindex <= bbot; bindex++) {
+ br = au_sbr(sb, bindex);
-+ if (!(au_br_sb(br)->s_flags & MS_RDONLY)) {
++ if (!sb_rdonly(au_br_sb(br))) {
+ if (au_br_rdonly(br))
+ a->mvdown.flags
+ |= AUFS_MVDOWN_ROLOWER_R;
+ e = copy_to_user(uarg, &args->mvdown, sizeof(args->mvdown));
+ if (unlikely(e))
+ err = -EFAULT;
-+ au_delayed_kfree(args);
++ kfree(args);
+out:
+ AuTraceErr(err);
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/opts.c linux/fs/aufs/opts.c
---- /usr/share/empty/fs/aufs/opts.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/opts.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,1860 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/opts.c linux-4.14/fs/aufs/opts.c
+--- linux-4.14/fs/aufs/opts.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/opts.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,1891 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ Opt_verbose, Opt_noverbose,
+ Opt_sum, Opt_nosum, Opt_wsum,
+ Opt_dirperm1, Opt_nodirperm1,
++ Opt_dirren, Opt_nodirren,
+ Opt_acl, Opt_noacl,
+ Opt_tail, Opt_ignore, Opt_ignore_silent, Opt_err
+};
+ {Opt_dio, "dio"},
+ {Opt_nodio, "nodio"},
+
++#ifdef CONFIG_AUFS_DIRREN
++ {Opt_dirren, "dirren"},
++ {Opt_nodirren, "nodirren"},
++#else
++ {Opt_ignore, "dirren"},
++ {Opt_ignore_silent, "nodirren"},
++#endif
++
+#ifdef CONFIG_AUFS_FHSM
+ {Opt_fhsm_sec, "fhsm_sec=%d"},
+#else
-+ {Opt_ignore_silent, "fhsm_sec=%d"},
++ {Opt_ignore, "fhsm_sec=%d"},
+#endif
+
+ {Opt_diropq_a, "diropq=always"},
+
+ /* keep them temporary */
+ {Opt_ignore_silent, "nodlgt"},
-+ {Opt_ignore_silent, "clean_plink"},
++ {Opt_ignore, "clean_plink"},
+
+#ifdef CONFIG_AUFS_SHWH
+ {Opt_shwh, "shwh"},
+ {Opt_acl, "acl"},
+ {Opt_noacl, "noacl"},
+#else
-+ {Opt_ignore_silent, "acl"},
++ {Opt_ignore, "acl"},
+ {Opt_ignore_silent, "noacl"},
+#endif
+
+ {AuWbrCreate_MFSV, "mfs:%d"},
+ {AuWbrCreate_MFSV, "most-free-space:%d"},
+
++ /* top-down regardless the parent, and then mfs */
++ {AuWbrCreate_TDMFS, "tdmfs:%d"},
++ {AuWbrCreate_TDMFSV, "tdmfs:%d:%d"},
++
+ {AuWbrCreate_MFSRR, "mfsrr:%d"},
+ {AuWbrCreate_MFSRRV, "mfsrr:%d:%d"},
+ {AuWbrCreate_PMFS, "pmfs"},
+ {-1, NULL}
+};
+
-+/*
-+ * cf. linux/lib/parser.c and cmdline.c
-+ * gave up calling memparse() since it uses simple_strtoull() instead of
-+ * kstrto...().
-+ */
-+static int noinline_for_stack
-+au_match_ull(substring_t *s, unsigned long long *result)
-+{
-+ int err;
-+ unsigned int len;
-+ char a[32];
-+
-+ err = -ERANGE;
-+ len = s->to - s->from;
-+ if (len + 1 <= sizeof(a)) {
-+ memcpy(a, s->from, len);
-+ a[len] = '\0';
-+ err = kstrtoull(a, 0, result);
-+ }
-+ return err;
-+}
-+
+static int au_wbr_mfs_wmark(substring_t *arg, char *str,
+ struct au_opt_wbr_create *create)
+{
+ unsigned long long ull;
+
+ err = 0;
-+ if (!au_match_ull(arg, &ull))
++ if (!match_u64(arg, &ull))
+ create->mfsrr_watermark = ull;
+ else {
+ pr_err("bad integer in %s\n", str);
+ create->wbr_create = err;
+ switch (err) {
+ case AuWbrCreate_MFSRRV:
++ case AuWbrCreate_TDMFSV:
+ case AuWbrCreate_PMFSRRV:
+ e = au_wbr_mfs_wmark(&args[0], str, create);
+ if (!e)
+ err = e;
+ break;
+ case AuWbrCreate_MFSRR:
++ case AuWbrCreate_TDMFS:
+ case AuWbrCreate_PMFSRR:
+ e = au_wbr_mfs_wmark(&args[0], str, create);
+ if (unlikely(e)) {
+ AuDbg("%d sec\n", u.create->mfs_second);
+ break;
+ case AuWbrCreate_MFSRR:
++ case AuWbrCreate_TDMFS:
+ AuDbg("%llu watermark\n",
+ u.create->mfsrr_watermark);
+ break;
+ case AuWbrCreate_MFSRRV:
++ case AuWbrCreate_TDMFSV:
+ case AuWbrCreate_PMFSRRV:
+ AuDbg("%llu watermark, %d sec\n",
+ u.create->mfsrr_watermark,
+ case Opt_fhsm_sec:
+ AuDbg("fhsm_sec %u\n", opt->fhsm_second);
+ break;
++ case Opt_dirren:
++ AuLabel(dirren);
++ break;
++ case Opt_nodirren:
++ AuLabel(nodirren);
++ break;
+ case Opt_acl:
+ AuLabel(acl);
+ break;
+ case Opt_wsum:
+ case Opt_rdblk_def:
+ case Opt_rdhash_def:
++ case Opt_dirren:
++ case Opt_nodirren:
+ case Opt_acl:
+ case Opt_noacl:
+ err = 0;
+ }
+ }
+
-+ au_delayed_kfree(a);
++ kfree(a);
+ dump_opts(opts);
+ if (unlikely(err))
+ au_opts_free(opts);
+ switch (create->wbr_create) {
+ case AuWbrCreate_MFSRRV:
+ case AuWbrCreate_MFSRR:
++ case AuWbrCreate_TDMFS:
++ case AuWbrCreate_TDMFSV:
+ case AuWbrCreate_PMFSRR:
+ case AuWbrCreate_PMFSRRV:
+ sbinfo->si_wbr_mfs.mfsrr_watermark = create->mfsrr_watermark;
+ au_fclr_opts(opts->flags, TRUNC_XIB);
+ break;
+
++ case Opt_dirren:
++ err = 1;
++ if (!au_opt_test(sbinfo->si_mntflags, DIRREN)) {
++ err = au_dr_opt_set(sb);
++ if (!err)
++ err = 1;
++ }
++ if (err == 1)
++ au_opt_set(sbinfo->si_mntflags, DIRREN);
++ break;
++ case Opt_nodirren:
++ err = 1;
++ if (au_opt_test(sbinfo->si_mntflags, DIRREN)) {
++ err = au_dr_opt_clr(sb, au_ftest_opts(opts->flags,
++ DR_FLUSHED));
++ if (!err)
++ err = 1;
++ }
++ if (err == 1)
++ au_opt_clr(sbinfo->si_mntflags, DIRREN);
++ break;
++
+ case Opt_acl:
+ sb->s_flags |= MS_POSIXACL;
+ break;
+ }
+ break;
+ }
-+
+ return err;
+}
+
+ au_hn_inode_unlock(hdir);
+
+ if (!err && do_free) {
-+ if (wbr)
-+ au_delayed_kfree(wbr);
++ kfree(wbr);
+ br->br_wbr = NULL;
+ }
+ }
+
+ SiMustWriteLock(sb);
+
-+ err = 0;
++ err = au_dr_opt_flush(sb);
++ if (unlikely(err))
++ goto out;
++ au_fset_opts(opts->flags, DR_FLUSHED);
++
+ dir = d_inode(sb->s_root);
+ sbinfo = au_sbi(sb);
+ opt_xino = NULL;
+ au_fset_opts(opts->flags, REFRESH);
+
+ AuDbg("status 0x%x\n", opts->flags);
++
++out:
+ return err;
+}
+
+{
+ return au_mntflags(sb) & AuOptMask_UDBA;
+}
-diff -urN /usr/share/empty/fs/aufs/opts.h linux/fs/aufs/opts.h
---- /usr/share/empty/fs/aufs/opts.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/opts.h 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,211 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/opts.h linux-4.14/fs/aufs/opts.h
+--- linux-4.14/fs/aufs/opts.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/opts.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,224 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#include <linux/path.h>
+
+struct file;
-+struct super_block;
+
+/* ---------------------------------------------------------------------- */
+
+#define AuOpt_WARN_PERM (1 << 12) /* warn when add-branch */
+#define AuOpt_VERBOSE (1 << 13) /* busy inode when del-branch */
+#define AuOpt_DIO (1 << 14) /* direct io */
++#define AuOpt_DIRREN (1 << 15) /* directory rename */
+
+#ifndef CONFIG_AUFS_HNOTIFY
+#undef AuOpt_UDBA_HNOTIFY
+#define AuOpt_UDBA_HNOTIFY 0
+#endif
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuOpt_DIRREN
++#define AuOpt_DIRREN 0
++#endif
+#ifndef CONFIG_AUFS_SHWH
+#undef AuOpt_SHWH
+#define AuOpt_SHWH 0
+ AuWbrCreate_MFSV, /* mfs with seconds */
+ AuWbrCreate_MFSRR, /* mfs then rr */
+ AuWbrCreate_MFSRRV, /* mfs then rr with seconds */
++ AuWbrCreate_TDMFS, /* top down regardless parent and mfs */
++ AuWbrCreate_TDMFSV, /* top down regardless parent and mfs */
+ AuWbrCreate_PMFS, /* parent and mfs */
+ AuWbrCreate_PMFSV, /* parent and mfs with seconds */
+ AuWbrCreate_PMFSRR, /* parent, mfs and round-robin */
+#define AuOpts_TRUNC_XIB (1 << 2)
+#define AuOpts_REFRESH_DYAOP (1 << 3)
+#define AuOpts_REFRESH_IDOP (1 << 4)
++#define AuOpts_DR_FLUSHED (1 << 5)
+#define au_ftest_opts(flags, name) ((flags) & AuOpts_##name)
+#define au_fset_opts(flags, name) \
+ do { (flags) |= AuOpts_##name; } while (0)
+#define au_fclr_opts(flags, name) \
+ do { (flags) &= ~AuOpts_##name; } while (0)
+
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuOpts_DR_FLUSHED
++#define AuOpts_DR_FLUSHED 0
++#endif
++
+struct au_opts {
+ struct au_opt *opt;
+ int max_opt;
+const char *au_optstr_wbr_create(int wbr_create);
+
+void au_opts_free(struct au_opts *opts);
++struct super_block;
+int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts);
+int au_opts_verify(struct super_block *sb, unsigned long sb_flags,
+ unsigned int pending);
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_OPTS_H__ */
-diff -urN /usr/share/empty/fs/aufs/plink.c linux/fs/aufs/plink.c
---- /usr/share/empty/fs/aufs/plink.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/plink.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,514 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/plink.c linux-4.14/fs/aufs/plink.c
+--- linux-4.14/fs/aufs/plink.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/plink.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,515 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+{
+ int i;
+ struct au_sbinfo *sbinfo;
-+ struct hlist_head *plink_hlist;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
+ struct au_icntnr *icntnr;
+
+ SiMustAnyLock(sb);
+ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
+
+ for (i = 0; i < AuPlink_NHASH; i++) {
-+ plink_hlist = &sbinfo->si_plink[i].head;
-+ rcu_read_lock();
-+ hlist_for_each_entry_rcu(icntnr, plink_hlist, plink)
++ hbl = sbinfo->si_plink + i;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(icntnr, pos, hbl, plink)
+ AuDbg("%lu\n", icntnr->vfs_inode.i_ino);
-+ rcu_read_unlock();
++ hlist_bl_unlock(hbl);
+ }
+}
+#endif
+{
+ int found, i;
+ struct au_sbinfo *sbinfo;
-+ struct hlist_head *plink_hlist;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
+ struct au_icntnr *icntnr;
+
+ sbinfo = au_sbi(inode->i_sb);
+
+ found = 0;
+ i = au_plink_hash(inode->i_ino);
-+ plink_hlist = &sbinfo->si_plink[i].head;
-+ rcu_read_lock();
-+ hlist_for_each_entry_rcu(icntnr, plink_hlist, plink)
++ hbl = sbinfo->si_plink + i;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(icntnr, pos, hbl, plink)
+ if (&icntnr->vfs_inode == inode) {
+ found = 1;
+ break;
+ }
-+ rcu_read_unlock();
++ hlist_bl_unlock(hbl);
+ return found;
+}
+
+ struct inode *h_inode;
+
+ h_inode = d_inode(h_parent);
-+ inode_lock_nested(h_inode, AuLsc_I_CHILD2);
++ vfsub_inode_lock_shared_nested(h_inode, AuLsc_I_CHILD2);
+ h_dentry = vfsub_lkup_one(tgtname, h_parent);
-+ inode_unlock(h_inode);
++ inode_unlock_shared(h_inode);
+ return h_dentry;
+}
+
+{
+ struct super_block *sb;
+ struct au_sbinfo *sbinfo;
-+ struct hlist_head *plink_hlist;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
+ struct au_icntnr *icntnr;
-+ struct au_sphlhead *sphl;
+ int found, err, cnt, i;
+
+ sb = inode->i_sb;
+ return;
+
+ i = au_plink_hash(inode->i_ino);
-+ sphl = sbinfo->si_plink + i;
-+ plink_hlist = &sphl->head;
++ hbl = sbinfo->si_plink + i;
+ au_igrab(inode);
+
-+ spin_lock(&sphl->spin);
-+ hlist_for_each_entry(icntnr, plink_hlist, plink) {
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(icntnr, pos, hbl, plink) {
+ if (&icntnr->vfs_inode == inode) {
+ found = 1;
+ break;
+ }
+ if (!found) {
+ icntnr = container_of(inode, struct au_icntnr, vfs_inode);
-+ hlist_add_head_rcu(&icntnr->plink, plink_hlist);
++ hlist_bl_add_head(&icntnr->plink, hbl);
+ }
-+ spin_unlock(&sphl->spin);
++ hlist_bl_unlock(hbl);
+ if (!found) {
-+ cnt = au_sphl_count(sphl);
++ cnt = au_hbl_count(hbl);
+#define msg "unexpectedly unblanced or too many pseudo-links"
+ if (cnt > AUFS_PLINK_WARN)
+ AuWarn1(msg ", %d\n", cnt);
+ err = whplink(h_dentry, inode, bindex, au_sbr(sb, bindex));
+ if (unlikely(err)) {
+ pr_warn("err %d, damaged pseudo link.\n", err);
-+ au_sphl_del_rcu(&icntnr->plink, sphl);
++ au_hbl_del(&icntnr->plink, hbl);
+ iput(&icntnr->vfs_inode);
+ }
+ } else
+{
+ int i, warned;
+ struct au_sbinfo *sbinfo;
-+ struct hlist_head *plink_hlist;
-+ struct hlist_node *tmp;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
+ struct au_icntnr *icntnr;
+
+ SiMustWriteLock(sb);
+ /* no spin_lock since sbinfo is write-locked */
+ warned = 0;
+ for (i = 0; i < AuPlink_NHASH; i++) {
-+ plink_hlist = &sbinfo->si_plink[i].head;
-+ if (!warned && verbose && !hlist_empty(plink_hlist)) {
++ hbl = sbinfo->si_plink + i;
++ if (!warned && verbose && !hlist_bl_empty(hbl)) {
+ pr_warn("pseudo-link is not flushed");
+ warned = 1;
+ }
-+ hlist_for_each_entry_safe(icntnr, tmp, plink_hlist, plink)
++ hlist_bl_for_each_entry_safe(icntnr, pos, tmp, hbl, plink)
+ iput(&icntnr->vfs_inode);
-+ INIT_HLIST_HEAD(plink_hlist);
++ INIT_HLIST_BL_HEAD(hbl);
+ }
+}
+
+void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id)
+{
+ struct au_sbinfo *sbinfo;
-+ struct hlist_head *plink_hlist;
-+ struct hlist_node *tmp;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
+ struct au_icntnr *icntnr;
+ struct inode *inode;
+ int i, do_put;
+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
+ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
+
-+ /* no spin_lock since sbinfo is write-locked */
++ /* no bit_lock since sbinfo is write-locked */
+ for (i = 0; i < AuPlink_NHASH; i++) {
-+ plink_hlist = &sbinfo->si_plink[i].head;
-+ hlist_for_each_entry_safe(icntnr, tmp, plink_hlist, plink) {
++ hbl = sbinfo->si_plink + i;
++ hlist_bl_for_each_entry_safe(icntnr, pos, tmp, hbl, plink) {
+ inode = au_igrab(&icntnr->vfs_inode);
+ ii_write_lock_child(inode);
+ do_put = au_plink_do_half_refresh(inode, br_id);
+ if (do_put) {
-+ hlist_del(&icntnr->plink);
++ hlist_bl_del(&icntnr->plink);
+ iput(inode);
+ }
+ ii_write_unlock(inode);
+ }
+ }
+}
-diff -urN /usr/share/empty/fs/aufs/poll.c linux/fs/aufs/poll.c
---- /usr/share/empty/fs/aufs/poll.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/poll.c 2016-07-25 19:05:34.817826663 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/poll.c linux-4.14/fs/aufs/poll.c
+--- linux-4.14/fs/aufs/poll.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/poll.c 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,52 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ sb = file->f_path.dentry->d_sb;
+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ h_file = au_read_pre(file, /*keep_fi*/0);
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
+ err = PTR_ERR(h_file);
+ if (IS_ERR(h_file))
+ goto out;
+ AuTraceErr((int)mask);
+ return mask;
+}
-diff -urN /usr/share/empty/fs/aufs/posix_acl.c linux/fs/aufs/posix_acl.c
---- /usr/share/empty/fs/aufs/posix_acl.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/posix_acl.c 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,98 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/posix_acl.c linux-4.14/fs/aufs/posix_acl.c
+--- linux-4.14/fs/aufs/posix_acl.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/posix_acl.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,102 @@
+/*
-+ * Copyright (C) 2014-2016 Junjiro R. Okajima
++ * Copyright (C) 2014-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+ /* always topmost only */
+ acl = get_acl(h_inode, type);
++ if (!IS_ERR_OR_NULL(acl))
++ set_cached_acl(inode, type, acl);
+
+out:
+ ii_read_unlock(inode);
+ int err;
+ ssize_t ssz;
+ struct dentry *dentry;
-+ struct au_srxattr arg = {
++ struct au_sxattr arg = {
+ .type = AU_ACL_SET,
+ .u.acl_set = {
+ .acl = acl,
+ }
+ }
+
-+ ssz = au_srxattr(dentry, inode, &arg);
++ ssz = au_sxattr(dentry, inode, &arg);
+ dput(dentry);
+ err = ssz;
-+ if (ssz >= 0)
++ if (ssz >= 0) {
+ err = 0;
++ set_cached_acl(inode, type, acl);
++ }
+
+out:
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/procfs.c linux/fs/aufs/procfs.c
---- /usr/share/empty/fs/aufs/procfs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/procfs.c 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,169 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/procfs.c linux-4.14/fs/aufs/procfs.c
+--- linux-4.14/fs/aufs/procfs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/procfs.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,170 @@
+/*
-+ * Copyright (C) 2010-2016 Junjiro R. Okajima
++ * Copyright (C) 2010-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ int err;
+ struct super_block *sb;
+ struct au_sbinfo *sbinfo;
++ struct hlist_bl_node *pos;
+
+ err = -EBUSY;
+ if (unlikely(file->private_data))
+
+ sb = NULL;
+ /* don't use au_sbilist_lock() here */
-+ spin_lock(&au_sbilist.spin);
-+ hlist_for_each_entry(sbinfo, &au_sbilist.head, si_list)
++ hlist_bl_lock(&au_sbilist);
++ hlist_bl_for_each_entry(sbinfo, pos, &au_sbilist, si_list)
+ if (id == sysaufs_si_id(sbinfo)) {
+ kobject_get(&sbinfo->si_kobj);
+ sb = sbinfo->si_sb;
+ break;
+ }
-+ spin_unlock(&au_sbilist.spin);
++ hlist_bl_unlock(&au_sbilist);
+
+ err = -EINVAL;
+ if (unlikely(!sb))
+out:
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/rdu.c linux/fs/aufs/rdu.c
---- /usr/share/empty/fs/aufs/rdu.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/rdu.c 2016-07-25 19:05:34.817826663 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/rdu.c linux-4.14/fs/aufs/rdu.c
+--- linux-4.14/fs/aufs/rdu.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/rdu.c 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,381 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ return err;
+}
+#endif
-diff -urN /usr/share/empty/fs/aufs/rwsem.h linux/fs/aufs/rwsem.h
---- /usr/share/empty/fs/aufs/rwsem.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/rwsem.h 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,198 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/rwsem.h linux-4.14/fs/aufs/rwsem.h
+--- linux-4.14/fs/aufs/rwsem.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/rwsem.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,72 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#include "debug.h"
+
-+struct au_rwsem {
-+ struct rw_semaphore rwsem;
-+#ifdef CONFIG_AUFS_DEBUG
-+ /* just for debugging, not almighty counter */
-+ atomic_t rcnt, wcnt;
-+#endif
-+};
-+
-+#ifdef CONFIG_LOCKDEP
-+#define au_lockdep_set_name(rw) \
-+ lockdep_set_class_and_name(&(rw)->rwsem, \
-+ /*original key*/(rw)->rwsem.dep_map.key, \
-+ /*name*/#rw)
-+#else
-+#define au_lockdep_set_name(rw) do {} while (0)
-+#endif
-+
-+#ifdef CONFIG_AUFS_DEBUG
-+#define AuDbgCntInit(rw) do { \
-+ atomic_set(&(rw)->rcnt, 0); \
-+ atomic_set(&(rw)->wcnt, 0); \
-+ smp_mb(); /* atomic set */ \
-+} while (0)
-+
-+#define AuDbgCnt(rw, cnt) atomic_read(&(rw)->cnt)
-+#define AuDbgCntInc(rw, cnt) atomic_inc(&(rw)->cnt)
-+#define AuDbgCntDec(rw, cnt) WARN_ON(atomic_dec_return(&(rw)->cnt) < 0)
-+#define AuDbgRcntInc(rw) AuDbgCntInc(rw, rcnt)
-+#define AuDbgRcntDec(rw) AuDbgCntDec(rw, rcnt)
-+#define AuDbgWcntInc(rw) AuDbgCntInc(rw, wcnt)
-+#define AuDbgWcntDec(rw) AuDbgCntDec(rw, wcnt)
-+#else
-+#define AuDbgCnt(rw, cnt) 0
-+#define AuDbgCntInit(rw) do {} while (0)
-+#define AuDbgRcntInc(rw) do {} while (0)
-+#define AuDbgRcntDec(rw) do {} while (0)
-+#define AuDbgWcntInc(rw) do {} while (0)
-+#define AuDbgWcntDec(rw) do {} while (0)
-+#endif /* CONFIG_AUFS_DEBUG */
++/* in the futre, the name 'au_rwsem' will be totally gone */
++#define au_rwsem rw_semaphore
+
+/* to debug easier, do not make them inlined functions */
-+#define AuRwMustNoWaiters(rw) AuDebugOn(rwsem_is_contended(&(rw)->rwsem))
++#define AuRwMustNoWaiters(rw) AuDebugOn(rwsem_is_contended(rw))
+/* rwsem_is_locked() is unusable */
-+#define AuRwMustReadLock(rw) AuDebugOn(AuDbgCnt(rw, rcnt) <= 0)
-+#define AuRwMustWriteLock(rw) AuDebugOn(AuDbgCnt(rw, wcnt) <= 0)
-+#define AuRwMustAnyLock(rw) AuDebugOn(AuDbgCnt(rw, rcnt) <= 0 \
-+ && AuDbgCnt(rw, wcnt) <= 0)
-+#define AuRwDestroy(rw) AuDebugOn(AuDbgCnt(rw, rcnt) \
-+ || AuDbgCnt(rw, wcnt))
-+
-+#define au_rw_init(rw) do { \
-+ AuDbgCntInit(rw); \
-+ init_rwsem(&(rw)->rwsem); \
-+ au_lockdep_set_name(rw); \
-+ } while (0)
++#define AuRwMustReadLock(rw) AuDebugOn(!lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held_type(rw, 1))
++#define AuRwMustWriteLock(rw) AuDebugOn(!lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held_type(rw, 0))
++#define AuRwMustAnyLock(rw) AuDebugOn(!lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held(rw))
++#define AuRwDestroy(rw) AuDebugOn(!lockdep_recursing(current) \
++ && debug_locks \
++ && lockdep_is_held(rw))
++
++#define au_rw_init(rw) init_rwsem(rw)
+
+#define au_rw_init_wlock(rw) do { \
+ au_rw_init(rw); \
-+ down_write(&(rw)->rwsem); \
-+ AuDbgWcntInc(rw); \
++ down_write(rw); \
+ } while (0)
+
-+#define au_rw_init_wlock_nested(rw, lsc) do { \
-+ au_rw_init(rw); \
-+ down_write_nested(&(rw)->rwsem, lsc); \
-+ AuDbgWcntInc(rw); \
++#define au_rw_init_wlock_nested(rw, lsc) do { \
++ au_rw_init(rw); \
++ down_write_nested(rw, lsc); \
+ } while (0)
+
-+static inline void au_rw_read_lock(struct au_rwsem *rw)
-+{
-+ down_read(&rw->rwsem);
-+ AuDbgRcntInc(rw);
-+}
-+
-+static inline void au_rw_read_lock_nested(struct au_rwsem *rw, unsigned int lsc)
-+{
-+ down_read_nested(&rw->rwsem, lsc);
-+ AuDbgRcntInc(rw);
-+}
-+
-+static inline void au_rw_read_unlock(struct au_rwsem *rw)
-+{
-+ AuRwMustReadLock(rw);
-+ AuDbgRcntDec(rw);
-+ up_read(&rw->rwsem);
-+}
-+
-+static inline void au_rw_dgrade_lock(struct au_rwsem *rw)
-+{
-+ AuRwMustWriteLock(rw);
-+ AuDbgRcntInc(rw);
-+ AuDbgWcntDec(rw);
-+ downgrade_write(&rw->rwsem);
-+}
-+
-+static inline void au_rw_write_lock(struct au_rwsem *rw)
-+{
-+ down_write(&rw->rwsem);
-+ AuDbgWcntInc(rw);
-+}
-+
-+static inline void au_rw_write_lock_nested(struct au_rwsem *rw,
-+ unsigned int lsc)
-+{
-+ down_write_nested(&rw->rwsem, lsc);
-+ AuDbgWcntInc(rw);
-+}
-+
-+static inline void au_rw_write_unlock(struct au_rwsem *rw)
-+{
-+ AuRwMustWriteLock(rw);
-+ AuDbgWcntDec(rw);
-+ up_write(&rw->rwsem);
-+}
-+
-+/* why is not _nested version defined */
-+static inline int au_rw_read_trylock(struct au_rwsem *rw)
-+{
-+ int ret;
-+
-+ ret = down_read_trylock(&rw->rwsem);
-+ if (ret)
-+ AuDbgRcntInc(rw);
-+ return ret;
-+}
-+
-+static inline int au_rw_write_trylock(struct au_rwsem *rw)
-+{
-+ int ret;
-+
-+ ret = down_write_trylock(&rw->rwsem);
-+ if (ret)
-+ AuDbgWcntInc(rw);
-+ return ret;
-+}
-+
-+#undef AuDbgCntDec
-+#undef AuDbgRcntInc
-+#undef AuDbgRcntDec
-+#undef AuDbgWcntDec
-+
-+#define AuSimpleLockRwsemFuncs(prefix, param, rwsem) \
-+static inline void prefix##_read_lock(param) \
-+{ au_rw_read_lock(rwsem); } \
-+static inline void prefix##_write_lock(param) \
-+{ au_rw_write_lock(rwsem); } \
-+static inline int prefix##_read_trylock(param) \
-+{ return au_rw_read_trylock(rwsem); } \
-+static inline int prefix##_write_trylock(param) \
-+{ return au_rw_write_trylock(rwsem); }
-+/* why is not _nested version defined */
-+/* static inline void prefix##_read_trylock_nested(param, lsc)
-+{ au_rw_read_trylock_nested(rwsem, lsc)); }
-+static inline void prefix##_write_trylock_nestd(param, lsc)
-+{ au_rw_write_trylock_nested(rwsem, lsc); } */
-+
-+#define AuSimpleUnlockRwsemFuncs(prefix, param, rwsem) \
-+static inline void prefix##_read_unlock(param) \
-+{ au_rw_read_unlock(rwsem); } \
-+static inline void prefix##_write_unlock(param) \
-+{ au_rw_write_unlock(rwsem); } \
-+static inline void prefix##_downgrade_lock(param) \
-+{ au_rw_dgrade_lock(rwsem); }
-+
-+#define AuSimpleRwsemFuncs(prefix, param, rwsem) \
-+ AuSimpleLockRwsemFuncs(prefix, param, rwsem) \
-+ AuSimpleUnlockRwsemFuncs(prefix, param, rwsem)
++#define au_rw_read_lock(rw) down_read(rw)
++#define au_rw_read_lock_nested(rw, lsc) down_read_nested(rw, lsc)
++#define au_rw_read_unlock(rw) up_read(rw)
++#define au_rw_dgrade_lock(rw) downgrade_write(rw)
++#define au_rw_write_lock(rw) down_write(rw)
++#define au_rw_write_lock_nested(rw, lsc) down_write_nested(rw, lsc)
++#define au_rw_write_unlock(rw) up_write(rw)
++/* why is not _nested version defined? */
++#define au_rw_read_trylock(rw) down_read_trylock(rw)
++#define au_rw_write_trylock(rw) down_write_trylock(rw)
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_RWSEM_H__ */
-diff -urN /usr/share/empty/fs/aufs/sbinfo.c linux/fs/aufs/sbinfo.c
---- /usr/share/empty/fs/aufs/sbinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/sbinfo.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,354 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/sbinfo.c linux-4.14/fs/aufs/sbinfo.c
+--- linux-4.14/fs/aufs/sbinfo.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/sbinfo.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,304 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
+ for (i = 0; i < AuPlink_NHASH; i++)
-+ AuDebugOn(!hlist_empty(&sbinfo->si_plink[i].head));
++ AuDebugOn(!hlist_bl_empty(sbinfo->si_plink + i));
+ AuDebugOn(atomic_read(&sbinfo->si_nowait.nw_len));
+
+ AuDebugOn(percpu_counter_sum(&sbinfo->si_ninodes));
+ au_br_free(sbinfo);
+ au_rw_write_unlock(&sbinfo->si_rwsem);
+
-+ au_delayed_kfree(sbinfo->si_branch);
-+ for (i = 0; i < AU_NPIDMAP; i++)
-+ if (sbinfo->au_si_pid.pid_bitmap[i])
-+ au_delayed_kfree(sbinfo->au_si_pid.pid_bitmap[i]);
-+ mutex_destroy(&sbinfo->au_si_pid.pid_mtx);
++ kfree(sbinfo->si_branch);
+ mutex_destroy(&sbinfo->si_xib_mtx);
+ AuRwDestroy(&sbinfo->si_rwsem);
+
-+ au_delayed_kfree(sbinfo);
++ kfree(sbinfo);
+}
+
+int au_si_alloc(struct super_block *sb)
+
+ au_nwt_init(&sbinfo->si_nowait);
+ au_rw_init_wlock(&sbinfo->si_rwsem);
-+ mutex_init(&sbinfo->au_si_pid.pid_mtx);
+
+ percpu_counter_init(&sbinfo->si_ninodes, 0, GFP_NOFS);
+ percpu_counter_init(&sbinfo->si_nfiles, 0, GFP_NOFS);
+ sbinfo->si_xino_brid = -1;
+ /* leave si_xib_last_pindex and si_xib_next_bit */
+
-+ au_sphl_init(&sbinfo->si_aopen);
++ INIT_HLIST_BL_HEAD(&sbinfo->si_aopen);
+
+ sbinfo->si_rdcache = msecs_to_jiffies(AUFS_RDCACHE_DEF * MSEC_PER_SEC);
+ sbinfo->si_rdblk = AUFS_RDBLK_DEF;
+ sbinfo->si_dirwh = AUFS_DIRWH_DEF;
+
+ for (i = 0; i < AuPlink_NHASH; i++)
-+ au_sphl_init(sbinfo->si_plink + i);
++ INIT_HLIST_BL_HEAD(sbinfo->si_plink + i);
+ init_waitqueue_head(&sbinfo->si_plink_wq);
+ spin_lock_init(&sbinfo->si_plink_maint_lock);
+
-+ au_sphl_init(&sbinfo->si_files);
++ INIT_HLIST_BL_HEAD(&sbinfo->si_files);
+
+ /* with getattr by default */
+ sbinfo->si_iop_array = aufs_iop;
+ return 0; /* success */
+
+out_br:
-+ au_delayed_kfree(sbinfo->si_branch);
++ kfree(sbinfo->si_branch);
+out_sbinfo:
-+ au_delayed_kfree(sbinfo);
++ kfree(sbinfo);
+out:
+ return err;
+}
+
-+int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr)
++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr, int may_shrink)
+{
+ int err, sz;
+ struct au_branch **brp;
+ sz = sizeof(*brp) * (sbinfo->si_bbot + 1);
+ if (unlikely(!sz))
+ sz = sizeof(*brp);
-+ brp = au_kzrealloc(sbinfo->si_branch, sz, sizeof(*brp) * nbr, GFP_NOFS);
++ brp = au_kzrealloc(sbinfo->si_branch, sz, sizeof(*brp) * nbr, GFP_NOFS,
++ may_shrink);
+ if (brp) {
+ sbinfo->si_branch = brp;
+ err = 0;
+ di_write_unlock2(d1, d2);
+ si_read_unlock(d1->d_sb);
+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+static void si_pid_alloc(struct au_si_pid *au_si_pid, int idx)
-+{
-+ unsigned long *p;
-+
-+ BUILD_BUG_ON(sizeof(unsigned long) !=
-+ sizeof(*au_si_pid->pid_bitmap));
-+
-+ mutex_lock(&au_si_pid->pid_mtx);
-+ p = au_si_pid->pid_bitmap[idx];
-+ while (!p) {
-+ /*
-+ * bad approach.
-+ * but keeping 'si_pid_set()' void is more important.
-+ */
-+ p = kcalloc(BITS_TO_LONGS(AU_PIDSTEP),
-+ sizeof(*au_si_pid->pid_bitmap),
-+ GFP_NOFS);
-+ if (p)
-+ break;
-+ cond_resched();
-+ }
-+ au_si_pid->pid_bitmap[idx] = p;
-+ mutex_unlock(&au_si_pid->pid_mtx);
-+}
-+
-+void si_pid_set(struct super_block *sb)
-+{
-+ pid_t bit;
-+ int idx;
-+ unsigned long *bitmap;
-+ struct au_si_pid *au_si_pid;
-+
-+ si_pid_idx_bit(&idx, &bit);
-+ au_si_pid = &au_sbi(sb)->au_si_pid;
-+ bitmap = au_si_pid->pid_bitmap[idx];
-+ if (!bitmap) {
-+ si_pid_alloc(au_si_pid, idx);
-+ bitmap = au_si_pid->pid_bitmap[idx];
-+ }
-+ AuDebugOn(test_bit(bit, bitmap));
-+ set_bit(bit, bitmap);
-+ /* smp_mb(); */
-+}
-diff -urN /usr/share/empty/fs/aufs/spl.h linux/fs/aufs/spl.h
---- /usr/share/empty/fs/aufs/spl.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/spl.h 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,113 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/super.c linux-4.14/fs/aufs/super.c
+--- linux-4.14/fs/aufs/super.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/super.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,1049 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
-+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
-+ * it under the terms of the GNU General Public License as published by
-+ * the Free Software Foundation; either version 2 of the License, or
-+ * (at your option) any later version.
-+ *
-+ * This program is distributed in the hope that it will be useful,
-+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
-+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
-+ * GNU General Public License for more details.
-+ *
-+ * You should have received a copy of the GNU General Public License
-+ * along with this program. If not, see <http://www.gnu.org/licenses/>.
-+ */
-+
-+/*
-+ * simple list protected by a spinlock
-+ */
-+
-+#ifndef __AUFS_SPL_H__
-+#define __AUFS_SPL_H__
-+
-+#ifdef __KERNEL__
-+
-+#if 0
-+struct au_splhead {
-+ spinlock_t spin;
-+ struct list_head head;
-+};
-+
-+static inline void au_spl_init(struct au_splhead *spl)
-+{
-+ spin_lock_init(&spl->spin);
-+ INIT_LIST_HEAD(&spl->head);
-+}
-+
-+static inline void au_spl_add(struct list_head *list, struct au_splhead *spl)
-+{
-+ spin_lock(&spl->spin);
-+ list_add(list, &spl->head);
-+ spin_unlock(&spl->spin);
-+}
-+
-+static inline void au_spl_del(struct list_head *list, struct au_splhead *spl)
-+{
-+ spin_lock(&spl->spin);
-+ list_del(list);
-+ spin_unlock(&spl->spin);
-+}
-+
-+static inline void au_spl_del_rcu(struct list_head *list,
-+ struct au_splhead *spl)
-+{
-+ spin_lock(&spl->spin);
-+ list_del_rcu(list);
-+ spin_unlock(&spl->spin);
-+}
-+#endif
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_sphlhead {
-+ spinlock_t spin;
-+ struct hlist_head head;
-+};
-+
-+static inline void au_sphl_init(struct au_sphlhead *sphl)
-+{
-+ spin_lock_init(&sphl->spin);
-+ INIT_HLIST_HEAD(&sphl->head);
-+}
-+
-+static inline void au_sphl_add(struct hlist_node *hlist,
-+ struct au_sphlhead *sphl)
-+{
-+ spin_lock(&sphl->spin);
-+ hlist_add_head(hlist, &sphl->head);
-+ spin_unlock(&sphl->spin);
-+}
-+
-+static inline void au_sphl_del(struct hlist_node *hlist,
-+ struct au_sphlhead *sphl)
-+{
-+ spin_lock(&sphl->spin);
-+ hlist_del(hlist);
-+ spin_unlock(&sphl->spin);
-+}
-+
-+static inline void au_sphl_del_rcu(struct hlist_node *hlist,
-+ struct au_sphlhead *sphl)
-+{
-+ spin_lock(&sphl->spin);
-+ hlist_del_rcu(hlist);
-+ spin_unlock(&sphl->spin);
-+}
-+
-+static inline unsigned long au_sphl_count(struct au_sphlhead *sphl)
-+{
-+ unsigned long cnt;
-+ struct hlist_node *pos;
-+
-+ cnt = 0;
-+ spin_lock(&sphl->spin);
-+ hlist_for_each(pos, &sphl->head)
-+ cnt++;
-+ spin_unlock(&sphl->spin);
-+ return cnt;
-+}
-+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_SPL_H__ */
-diff -urN /usr/share/empty/fs/aufs/super.c linux/fs/aufs/super.c
---- /usr/share/empty/fs/aufs/super.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/super.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,1038 @@
-+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+{
+ struct inode *inode = container_of(head, struct inode, i_rcu);
+
-+ au_cache_dfree_icntnr(container_of(inode, struct au_icntnr, vfs_inode));
++ au_cache_free_icntnr(container_of(inode, struct au_icntnr, vfs_inode));
+}
+
+static void aufs_destroy_inode(struct inode *inode)
+ return err;
+}
+
++static void au_gen_fmt(char *fmt, int len __maybe_unused, const char *pat,
++ const char *append)
++{
++ char *p;
++
++ p = fmt;
++ while (*pat != ':')
++ *p++ = *pat++;
++ *p++ = *pat++;
++ strcpy(p, append);
++ AuDebugOn(strlen(fmt) >= len);
++}
++
+static void au_show_wbr_create(struct seq_file *m, int v,
+ struct au_sbinfo *sbinfo)
+{
+ const char *pat;
++ char fmt[32];
++ struct au_wbr_mfs *mfs;
+
+ AuRwMustAnyLock(&sbinfo->si_rwsem);
+
+ seq_puts(m, ",create=");
+ pat = au_optstr_wbr_create(v);
++ mfs = &sbinfo->si_wbr_mfs;
+ switch (v) {
+ case AuWbrCreate_TDP:
+ case AuWbrCreate_RR:
+ case AuWbrCreate_PMFS:
+ seq_puts(m, pat);
+ break;
-+ case AuWbrCreate_MFSV:
-+ seq_printf(m, /*pat*/"mfs:%lu",
-+ jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire)
-+ / MSEC_PER_SEC);
++ case AuWbrCreate_MFSRR:
++ case AuWbrCreate_TDMFS:
++ case AuWbrCreate_PMFSRR:
++ au_gen_fmt(fmt, sizeof(fmt), pat, "%llu");
++ seq_printf(m, fmt, mfs->mfsrr_watermark);
+ break;
++ case AuWbrCreate_MFSV:
+ case AuWbrCreate_PMFSV:
-+ seq_printf(m, /*pat*/"pmfs:%lu",
-+ jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire)
++ au_gen_fmt(fmt, sizeof(fmt), pat, "%lu");
++ seq_printf(m, fmt,
++ jiffies_to_msecs(mfs->mfs_expire)
+ / MSEC_PER_SEC);
+ break;
-+ case AuWbrCreate_MFSRR:
-+ seq_printf(m, /*pat*/"mfsrr:%llu",
-+ sbinfo->si_wbr_mfs.mfsrr_watermark);
-+ break;
+ case AuWbrCreate_MFSRRV:
-+ seq_printf(m, /*pat*/"mfsrr:%llu:%lu",
-+ sbinfo->si_wbr_mfs.mfsrr_watermark,
-+ jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire)
-+ / MSEC_PER_SEC);
-+ break;
-+ case AuWbrCreate_PMFSRR:
-+ seq_printf(m, /*pat*/"pmfsrr:%llu",
-+ sbinfo->si_wbr_mfs.mfsrr_watermark);
-+ break;
++ case AuWbrCreate_TDMFSV:
+ case AuWbrCreate_PMFSRRV:
-+ seq_printf(m, /*pat*/"pmfsrr:%llu:%lu",
-+ sbinfo->si_wbr_mfs.mfsrr_watermark,
-+ jiffies_to_msecs(sbinfo->si_wbr_mfs.mfs_expire)
-+ / MSEC_PER_SEC);
++ au_gen_fmt(fmt, sizeof(fmt), pat, "%llu:%lu");
++ seq_printf(m, fmt, mfs->mfsrr_watermark,
++ jiffies_to_msecs(mfs->mfs_expire) / MSEC_PER_SEC);
+ break;
++ default:
++ BUG();
+ }
+}
+
+
+ au_fhsm_show(m, sbinfo);
+
++ AuBool(DIRREN, dirren);
+ AuBool(SUM, sum);
+ /* AuBool(SUM_W, wsum); */
+ AuBool(WARN_PERM, warn_perm);
+ continue;
+
+ h_sb = au_sbr_sb(sb, bindex);
-+ if (h_sb->s_op->sync_fs) {
-+ e = h_sb->s_op->sync_fs(h_sb, wait);
-+ if (unlikely(e && !err))
-+ err = e;
-+ /* go on even if an error happens */
-+ }
++ e = vfsub_sync_filesystem(h_sb, wait);
++ if (unlikely(e && !err))
++ err = e;
++ /* go on even if an error happens */
+ }
+ si_read_unlock(sb);
+
+out_mtx:
+ inode_unlock(inode);
+out_opts:
-+ au_delayed_free_page((unsigned long)opts.opt);
++ free_page((unsigned long)opts.opt);
+out:
+ err = cvt_err(err);
+ AuTraceErr(err);
+ .statfs = aufs_statfs,
+ .put_super = aufs_put_super,
+ .sync_fs = aufs_sync_fs,
-+ .remount_fs = aufs_remount_fs
++ .remount_fs = aufs_remount_fs,
++#ifdef CONFIG_AUFS_BDEV_LOOP
++ .real_loop = aufs_real_loop
++#endif
+};
+
+/* ---------------------------------------------------------------------- */
+ sb->s_maxbytes = 0;
+ sb->s_stack_depth = 1;
+ au_export_init(sb);
-+ /* au_xattr_init(sb); */
++ au_xattr_init(sb);
+
+ err = alloc_root(sb);
+ if (unlikely(err)) {
+ kobject_put(&sbinfo->si_kobj);
+ sb->s_fs_info = NULL;
+out_opts:
-+ au_delayed_free_page((unsigned long)opts.opt);
++ free_page((unsigned long)opts.opt);
+out:
+ AuTraceErr(err);
+ err = cvt_err(err);
+ if (au_opt_test(sbinfo->si_mntflags, PLINK))
+ au_plink_put(sb, /*verbose*/1);
+ au_xino_clr(sb);
++ au_dr_opt_flush(sb);
+ sbinfo->si_sb = NULL;
+ aufs_write_unlock(sb->s_root);
+ au_nwt_flush(&sbinfo->si_nowait);
+ /* no need to __module_get() and module_put(). */
+ .owner = THIS_MODULE,
+};
-diff -urN /usr/share/empty/fs/aufs/super.h linux/fs/aufs/super.h
---- /usr/share/empty/fs/aufs/super.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/super.h 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,638 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/super.h linux-4.14/fs/aufs/super.h
+--- linux-4.14/fs/aufs/super.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/super.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,626 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#include <linux/fs.h>
+#include <linux/kobject.h>
++#include "hbl.h"
+#include "rwsem.h"
-+#include "spl.h"
+#include "wkq.h"
+
+/* policies to select one among multiple writable branches */
+#endif
+};
+
-+#define AU_PIDSTEP (int)(BITS_TO_LONGS(PID_MAX_DEFAULT) * BITS_PER_LONG)
-+#define AU_NPIDMAP (int)DIV_ROUND_UP(PID_MAX_LIMIT, AU_PIDSTEP)
-+struct au_si_pid {
-+ unsigned long *pid_bitmap[AU_NPIDMAP];
-+ struct mutex pid_mtx;
-+};
-+
+struct au_branch;
+struct au_sbinfo {
+ /* nowait tasks in the system-wide workqueue */
+ */
+ struct au_rwsem si_rwsem;
+
-+ /* prevent recursive locking in deleting inode */
-+ struct au_si_pid au_si_pid;
-+
+ /*
+ * dirty approach to protect sb->sb_inodes and ->s_files (gone) from
+ * remount.
+#endif
+
+ /* dirty trick to suppoer atomic_open */
-+ struct au_sphlhead si_aopen;
++ struct hlist_bl_head si_aopen;
+
+ /* vdir parameters */
+ unsigned long si_rdcache; /* max cache time in jiffies */
+ unsigned int si_dirwh;
+
+ /* pseudo_link list */
-+ struct au_sphlhead si_plink[AuPlink_NHASH];
++ struct hlist_bl_head si_plink[AuPlink_NHASH];
+ wait_queue_head_t si_plink_wq;
+ spinlock_t si_plink_maint_lock;
+ pid_t si_plink_maint_pid;
+
+ /* file list */
-+ struct au_sphlhead si_files;
++ struct hlist_bl_head si_files;
+
+ /* with/without getattr, brother of sb->s_d_op */
+ struct inode_operations *si_iop_array;
+#endif
+
+#ifdef CONFIG_AUFS_SBILIST
-+ struct hlist_node si_list;
++ struct hlist_bl_node si_list;
+#endif
+
+ /* dirty, necessary for unmounting, sysfs and sysrq */
+#define AuLock_IW (1 << 2) /* write-lock inode */
+#define AuLock_FLUSH (1 << 3) /* wait for 'nowait' tasks */
+#define AuLock_DIRS (1 << 4) /* target is a pair of dirs */
++ /* except RENAME_EXCHANGE */
+#define AuLock_NOPLM (1 << 5) /* return err in plm mode */
+#define AuLock_NOPLMW (1 << 6) /* wait for plm mode ends */
+#define AuLock_GEN (1 << 7) /* test digen/iigen */
+/* sbinfo.c */
+void au_si_free(struct kobject *kobj);
+int au_si_alloc(struct super_block *sb);
-+int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr);
++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr, int may_shrink);
+
+unsigned int au_sigen_inc(struct super_block *sb);
+aufs_bindex_t au_new_br_id(struct super_block *sb);
+
+#ifdef CONFIG_AUFS_SBILIST
+/* module.c */
-+extern struct au_sphlhead au_sbilist;
++extern struct hlist_bl_head au_sbilist;
+
+static inline void au_sbilist_init(void)
+{
-+ au_sphl_init(&au_sbilist);
++ INIT_HLIST_BL_HEAD(&au_sbilist);
+}
+
+static inline void au_sbilist_add(struct super_block *sb)
+{
-+ au_sphl_add(&au_sbi(sb)->si_list, &au_sbilist);
++ au_hbl_add(&au_sbi(sb)->si_list, &au_sbilist);
+}
+
+static inline void au_sbilist_del(struct super_block *sb)
+{
-+ au_sphl_del(&au_sbi(sb)->si_list, &au_sbilist);
++ au_hbl_del(&au_sbi(sb)->si_list, &au_sbilist);
+}
+
+#ifdef CONFIG_AUFS_MAGIC_SYSRQ
+static inline void au_sbilist_lock(void)
+{
-+ spin_lock(&au_sbilist.spin);
++ hlist_bl_lock(&au_sbilist);
+}
+
+static inline void au_sbilist_unlock(void)
+{
-+ spin_unlock(&au_sbilist.spin);
++ hlist_bl_unlock(&au_sbilist);
+}
+#define AuGFP_SBILIST GFP_ATOMIC
+#else
+
+/* ---------------------------------------------------------------------- */
+
-+static inline void si_pid_idx_bit(int *idx, pid_t *bit)
-+{
-+ /* the origin of pid is 1, but the bitmap's is 0 */
-+ *bit = current->pid - 1;
-+ *idx = *bit / AU_PIDSTEP;
-+ *bit %= AU_PIDSTEP;
-+}
++/* current->atomic_flags */
++/* this value should never corrupt the ones defined in linux/sched.h */
++#define PFA_AUFS 7
++
++TASK_PFA_TEST(AUFS, test_aufs) /* task_test_aufs */
++TASK_PFA_SET(AUFS, aufs) /* task_set_aufs */
++TASK_PFA_CLEAR(AUFS, aufs) /* task_clear_aufs */
+
+static inline int si_pid_test(struct super_block *sb)
+{
-+ pid_t bit;
-+ int idx;
-+ unsigned long *bitmap;
-+
-+ si_pid_idx_bit(&idx, &bit);
-+ bitmap = au_sbi(sb)->au_si_pid.pid_bitmap[idx];
-+ if (bitmap)
-+ return test_bit(bit, bitmap);
-+ return 0;
++ return !!task_test_aufs(current);
+}
+
+static inline void si_pid_clr(struct super_block *sb)
+{
-+ pid_t bit;
-+ int idx;
-+ unsigned long *bitmap;
-+
-+ si_pid_idx_bit(&idx, &bit);
-+ bitmap = au_sbi(sb)->au_si_pid.pid_bitmap[idx];
-+ BUG_ON(!bitmap);
-+ AuDebugOn(!test_bit(bit, bitmap));
-+ clear_bit(bit, bitmap);
-+ /* smp_mb(); */
++ AuDebugOn(!task_test_aufs(current));
++ task_clear_aufs(current);
+}
+
-+void si_pid_set(struct super_block *sb);
++static inline void si_pid_set(struct super_block *sb)
++{
++ AuDebugOn(task_test_aufs(current));
++ task_set_aufs(current);
++}
+
+/* ---------------------------------------------------------------------- */
+
+/* lock superblock. mainly for entry point functions */
-+/*
-+ * __si_read_lock, __si_write_lock,
-+ * __si_read_unlock, __si_write_unlock, __si_downgrade_lock
-+ */
-+AuSimpleRwsemFuncs(__si, struct super_block *sb, &au_sbi(sb)->si_rwsem);
++#define __si_read_lock(sb) au_rw_read_lock(&au_sbi(sb)->si_rwsem)
++#define __si_write_lock(sb) au_rw_write_lock(&au_sbi(sb)->si_rwsem)
++#define __si_read_trylock(sb) au_rw_read_trylock(&au_sbi(sb)->si_rwsem)
++#define __si_write_trylock(sb) au_rw_write_trylock(&au_sbi(sb)->si_rwsem)
++/*
++#define __si_read_trylock_nested(sb) \
++ au_rw_read_trylock_nested(&au_sbi(sb)->si_rwsem)
++#define __si_write_trylock_nested(sb) \
++ au_rw_write_trylock_nested(&au_sbi(sb)->si_rwsem)
++*/
++
++#define __si_read_unlock(sb) au_rw_read_unlock(&au_sbi(sb)->si_rwsem)
++#define __si_write_unlock(sb) au_rw_write_unlock(&au_sbi(sb)->si_rwsem)
++#define __si_downgrade_lock(sb) au_rw_dgrade_lock(&au_sbi(sb)->si_rwsem)
+
+#define SiMustNoWaiters(sb) AuRwMustNoWaiters(&au_sbi(sb)->si_rwsem)
+#define SiMustAnyLock(sb) AuRwMustAnyLock(&au_sbi(sb)->si_rwsem)
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_SUPER_H__ */
-diff -urN /usr/share/empty/fs/aufs/sysaufs.c linux/fs/aufs/sysaufs.c
---- /usr/share/empty/fs/aufs/sysaufs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/sysaufs.c 2016-07-25 19:05:34.817826663 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/sysaufs.c linux-4.14/fs/aufs/sysaufs.c
+--- linux-4.14/fs/aufs/sysaufs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/sysaufs.c 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,104 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+out:
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/sysaufs.h linux/fs/aufs/sysaufs.h
---- /usr/share/empty/fs/aufs/sysaufs.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/sysaufs.h 2016-07-25 19:05:34.817826663 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/sysaufs.h linux-4.14/fs/aufs/sysaufs.h
+--- linux-4.14/fs/aufs/sysaufs.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/sysaufs.h 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,101 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#endif /* __KERNEL__ */
+#endif /* __SYSAUFS_H__ */
-diff -urN /usr/share/empty/fs/aufs/sysfs.c linux/fs/aufs/sysfs.c
---- /usr/share/empty/fs/aufs/sysfs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/sysfs.c 2016-08-17 18:01:06.161889951 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/sysfs.c linux-4.14/fs/aufs/sysfs.c
+--- linux-4.14/fs/aufs/sysfs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/sysfs.c 2021-02-24 21:42:43.444448075 +0100
@@ -0,0 +1,376 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ if (unlikely(err == PAGE_SIZE))
+ err = -EFBIG;
+ }
-+ au_delayed_kfree(seq);
++ kfree(seq);
+out_unlock:
+ si_read_unlock(sb);
+out:
+ err = -EFAULT;
+
+out_seq:
-+ au_delayed_kfree(seq);
++ kfree(seq);
+out_buf:
-+ au_delayed_free_page((unsigned long)buf);
++ free_page((unsigned long)buf);
+out:
+ si_read_unlock(sb);
+ return err;
+ }
+ }
+}
-diff -urN /usr/share/empty/fs/aufs/sysrq.c linux/fs/aufs/sysrq.c
---- /usr/share/empty/fs/aufs/sysrq.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/sysrq.c 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,157 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/sysrq.c linux-4.14/fs/aufs/sysrq.c
+--- linux-4.14/fs/aufs/sysrq.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/sysrq.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,159 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ char *plevel;
+ struct au_sbinfo *sbinfo;
+ struct file *file;
-+ struct au_sphlhead *files;
++ struct hlist_bl_head *files;
++ struct hlist_bl_node *pos;
+ struct au_finfo *finfo;
+
+ plevel = au_plevel;
+#endif
+ pr("files\n");
+ files = &au_sbi(sb)->si_files;
-+ spin_lock(&files->spin);
-+ hlist_for_each_entry(finfo, &files->head, fi_hlist) {
++ hlist_bl_lock(files);
++ hlist_bl_for_each_entry(finfo, pos, files, fi_hlist) {
+ umode_t mode;
+
+ file = finfo->fi_file;
+ if (!special_file(mode))
+ au_dpri_file(file);
+ }
-+ spin_unlock(&files->spin);
++ hlist_bl_unlock(files);
+ pr("done\n");
+
+#undef pr
+static void au_sysrq(int key __maybe_unused)
+{
+ struct au_sbinfo *sbinfo;
++ struct hlist_bl_node *pos;
+
+ lockdep_off();
+ au_sbilist_lock();
-+ hlist_for_each_entry(sbinfo, &au_sbilist.head, si_list)
++ hlist_bl_for_each_entry(sbinfo, pos, &au_sbilist, si_list)
+ sysrq_sb(sbinfo->si_sb);
+ au_sbilist_unlock();
+ lockdep_on();
+ if (unlikely(err))
+ pr_err("err %d (ignored)\n", err);
+}
-diff -urN /usr/share/empty/fs/aufs/vdir.c linux/fs/aufs/vdir.c
---- /usr/share/empty/fs/aufs/vdir.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/vdir.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,899 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/vdir.c linux-4.14/fs/aufs/vdir.c
+--- linux-4.14/fs/aufs/vdir.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/vdir.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,892 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ struct hlist_node *node;
+
+ hlist_for_each_entry_safe(pos, node, head, wh_hash)
-+ au_delayed_kfree(pos);
++ kfree(pos);
+}
+
+static void au_nhash_de_do_free(struct hlist_head *head)
+ struct hlist_node *node;
+
+ hlist_for_each_entry_safe(pos, node, head, hash)
-+ au_cache_dfree_vdir_dehstr(pos);
++ au_cache_free_vdir_dehstr(pos);
+}
+
+static void au_nhash_do_free(struct au_nhash *nhash,
+ nhash_count(head);
+ free(head++);
+ }
-+ au_delayed_kfree(nhash->nh_head);
++ kfree(nhash->nh_head);
+}
+
+void au_nhash_wh_free(struct au_nhash *whlist)
+ unsigned char **o;
+
+ err = -ENOMEM;
-+ o = krealloc(vdir->vd_deblk, sizeof(*o) * (vdir->vd_nblk + 1),
-+ GFP_NOFS);
++ o = au_krealloc(vdir->vd_deblk, sizeof(*o) * (vdir->vd_nblk + 1),
++ GFP_NOFS, /*may_shrink*/0);
+ if (unlikely(!o))
+ goto out;
+
+
+/* ---------------------------------------------------------------------- */
+
-+void au_vdir_free(struct au_vdir *vdir, int atonce)
++void au_vdir_free(struct au_vdir *vdir)
+{
+ unsigned char **deblk;
+
+ deblk = vdir->vd_deblk;
-+ if (!atonce) {
-+ while (vdir->vd_nblk--)
-+ au_delayed_kfree(*deblk++);
-+ au_delayed_kfree(vdir->vd_deblk);
-+ au_cache_dfree_vdir(vdir);
-+ } else {
-+ /* not delayed */
-+ while (vdir->vd_nblk--)
-+ kfree(*deblk++);
-+ kfree(vdir->vd_deblk);
-+ au_cache_free_vdir(vdir);
-+ }
++ while (vdir->vd_nblk--)
++ kfree(*deblk++);
++ kfree(vdir->vd_deblk);
++ au_cache_free_vdir(vdir);
+}
+
+static struct au_vdir *alloc_vdir(struct file *file)
+ if (!err)
+ return vdir; /* success */
+
-+ au_delayed_kfree(vdir->vd_deblk);
++ kfree(vdir->vd_deblk);
+
+out_free:
-+ au_cache_dfree_vdir(vdir);
++ au_cache_free_vdir(vdir);
+out:
+ vdir = ERR_PTR(err);
+ return vdir;
+ union au_vdir_deblk_p p, deblk_end;
+
+ while (vdir->vd_nblk > 1) {
-+ au_delayed_kfree(vdir->vd_deblk[vdir->vd_nblk - 1]);
++ kfree(vdir->vd_deblk[vdir->vd_nblk - 1]);
+ /* vdir->vd_deblk[vdir->vd_nblk - 1] = NULL; */
+ vdir->vd_nblk--;
+ }
+ }
+ }
+
-+ au_delayed_free_page((unsigned long)o);
++ free_page((unsigned long)o);
+
+out:
+ AuTraceErr(err);
+ if (allocated)
+ au_set_ivdir(inode, allocated);
+ } else if (allocated)
-+ au_vdir_free(allocated, /*atonce*/0);
++ au_vdir_free(allocated);
+
+out:
+ return err;
+ if (tgt->vd_nblk < src->vd_nblk) {
+ unsigned char **p;
+
-+ p = krealloc(tgt->vd_deblk, sizeof(*p) * src->vd_nblk,
-+ GFP_NOFS);
++ p = au_krealloc(tgt->vd_deblk, sizeof(*p) * src->vd_nblk,
++ GFP_NOFS, /*may_shrink*/0);
+ if (unlikely(!p))
+ goto out;
+ tgt->vd_deblk = p;
+ unsigned char *p;
+
+ tgt->vd_deblk_sz = deblk_sz;
-+ p = krealloc(tgt->vd_deblk[0], deblk_sz, GFP_NOFS);
++ p = au_krealloc(tgt->vd_deblk[0], deblk_sz, GFP_NOFS,
++ /*may_shrink*/1);
+ if (unlikely(!p))
+ goto out;
+ tgt->vd_deblk[0] = p;
+ if (allocated)
+ au_set_fvdir_cache(file, allocated);
+ } else if (allocated)
-+ au_vdir_free(allocated, /*atonce*/0);
++ au_vdir_free(allocated);
+
+out:
+ return err;
+ /* smp_mb(); */
+ return 0;
+}
-diff -urN /usr/share/empty/fs/aufs/vfsub.c linux/fs/aufs/vfsub.c
---- /usr/share/empty/fs/aufs/vfsub.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/vfsub.c 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,884 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/vfsub.c linux-4.14/fs/aufs/vfsub.c
+--- linux-4.14/fs/aufs/vfsub.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/vfsub.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,894 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * sub-routines for VFS
+ */
+
++#include <linux/mnt_namespace.h>
+#include <linux/namei.h>
+#include <linux/nsproxy.h>
+#include <linux/security.h>
+#include <linux/splice.h>
-+#include "../fs/mount.h"
+#include "aufs.h"
+
+#ifdef CONFIG_AUFS_BR_FUSE
+int vfsub_test_mntns(struct vfsmount *mnt, struct super_block *h_sb)
+{
-+ struct nsproxy *ns;
-+
+ if (!au_test_fuse(h_sb) || !au_userns)
+ return 0;
+
-+ ns = current->nsproxy;
-+ /* no {get,put}_nsproxy(ns) */
-+ return real_mount(mnt)->mnt_ns == ns->mnt_ns ? 0 : -EACCES;
++ return is_current_mnt_ns(mnt) ? 0 : -EACCES;
+}
+#endif
+
++int vfsub_sync_filesystem(struct super_block *h_sb, int wait)
++{
++ int err;
++
++ lockdep_off();
++ down_read(&h_sb->s_umount);
++ err = __sync_filesystem(h_sb, wait);
++ up_read(&h_sb->s_umount);
++ lockdep_on();
++
++ return err;
++}
++
+/* ---------------------------------------------------------------------- */
+
+int vfsub_update_h_iattr(struct path *h_path, int *did)
+ h_sb = h_path->dentry->d_sb;
+ *did = (!au_test_fs_remote(h_sb) && au_test_fs_refresh_iattr(h_sb));
+ if (*did)
-+ err = vfs_getattr(h_path, &st);
++ err = vfsub_getattr(h_path, &st);
+
+ return err;
+}
+
+int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
+ struct inode *dir, struct path *path,
-+ struct inode **delegated_inode)
++ struct inode **delegated_inode, unsigned int flags)
+{
+ int err;
+ struct path tmp = {
+
+ lockdep_off();
+ err = vfs_rename(src_dir, src_dentry, dir, path->dentry,
-+ delegated_inode, /*flags*/0);
++ delegated_inode, flags);
+ lockdep_on();
+ if (!err) {
+ int did;
+ lockdep_on();
+ if (err >= 0)
+ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
++
+ return err;
+}
+
+
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/vfsub.h linux/fs/aufs/vfsub.h
---- /usr/share/empty/fs/aufs/vfsub.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/vfsub.h 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,316 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/vfsub.h linux-4.14/fs/aufs/vfsub.h
+--- linux-4.14/fs/aufs/vfsub.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/vfsub.h 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,360 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+#define MtxMustLock(mtx) AuDebugOn(!mutex_is_locked(mtx))
+#define IMustLock(i) AuDebugOn(!inode_is_locked(i))
+
++/* why VFS doesn't define it? */
++static inline
++void vfsub_inode_lock_shared_nested(struct inode *inode, unsigned int sc)
++{
++ down_read_nested(&inode->i_rwsem, sc);
++}
++
+/* ---------------------------------------------------------------------- */
+
+static inline void vfsub_drop_nlink(struct inode *inode)
+
+static inline int vfsub_native_ro(struct inode *inode)
+{
-+ return (inode->i_sb->s_flags & MS_RDONLY)
++ return sb_rdonly(inode->i_sb)
+ || IS_RDONLY(inode)
+ /* || IS_APPEND(inode) */
+ || IS_IMMUTABLE(inode);
+AuStubInt0(vfsub_test_mntns, struct vfsmount *mnt, struct super_block *h_sb);
+#endif
+
++int vfsub_sync_filesystem(struct super_block *h_sb, int wait);
++
+/* ---------------------------------------------------------------------- */
+
+int vfsub_update_h_iattr(struct path *h_path, int *did);
+ struct path *path, struct inode **delegated_inode);
+int vfsub_rename(struct inode *src_hdir, struct dentry *src_dentry,
+ struct inode *hdir, struct path *path,
-+ struct inode **delegated_inode);
++ struct inode **delegated_inode, unsigned int flags);
+int vfsub_mkdir(struct inode *dir, struct path *path, int mode);
+int vfsub_rmdir(struct inode *dir, struct path *path);
+
+ struct file *h_file);
+int vfsub_fsync(struct file *file, struct path *path, int datasync);
+
++/*
++ * re-use branch fs's ioctl(FICLONE) while aufs itself doesn't support such
++ * ioctl.
++ */
++static inline int vfsub_clone_file_range(struct file *src, struct file *dst,
++ u64 len)
++{
++ int err;
++
++ lockdep_off();
++ err = vfs_clone_file_range(src, 0, dst, 0, len);
++ lockdep_on();
++
++ return err;
++}
++
++/* copy_file_range(2) is a systemcall */
++static inline ssize_t vfsub_copy_file_range(struct file *src, loff_t src_pos,
++ struct file *dst, loff_t dst_pos,
++ size_t len, unsigned int flags)
++{
++ ssize_t ssz;
++
++ lockdep_off();
++ ssz = vfs_copy_file_range(src, src_pos, dst, dst_pos, len, flags);
++ lockdep_on();
++
++ return ssz;
++}
++
+/* ---------------------------------------------------------------------- */
+
+static inline loff_t vfsub_llseek(struct file *file, loff_t offset, int origin)
+int vfsub_unlink(struct inode *dir, struct path *path,
+ struct inode **delegated_inode, int force);
+
++static inline int vfsub_getattr(const struct path *path, struct kstat *st)
++{
++ return vfs_getattr(path, st, STATX_BASIC_STATS, AT_STATX_SYNC_AS_STAT);
++}
++
+/* ---------------------------------------------------------------------- */
+
+static inline int vfsub_setxattr(struct dentry *dentry, const char *name,
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_VFSUB_H__ */
-diff -urN /usr/share/empty/fs/aufs/wbr_policy.c linux/fs/aufs/wbr_policy.c
---- /usr/share/empty/fs/aufs/wbr_policy.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/wbr_policy.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,765 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/wbr_policy.c linux-4.14/fs/aufs/wbr_policy.c
+--- linux-4.14/fs/aufs/wbr_policy.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/wbr_policy.c 2021-02-24 21:42:43.444448075 +0100
+@@ -0,0 +1,830 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+ mfs->mfsrr_bytes = bavail;
+ AuDbg("b%d\n", mfs->mfs_bindex);
-+ au_delayed_kfree(st);
++ kfree(st);
+}
+
+static int au_wbr_create_mfs(struct dentry *dentry, unsigned int flags)
+
+/* ---------------------------------------------------------------------- */
+
++/* top down regardless parent, and then mfs */
++static int au_wbr_create_tdmfs(struct dentry *dentry,
++ unsigned int flags __maybe_unused)
++{
++ int err;
++ aufs_bindex_t bwh, btail, bindex, bfound, bmfs;
++ unsigned long long watermark;
++ struct super_block *sb;
++ struct au_wbr_mfs *mfs;
++ struct au_branch *br;
++ struct dentry *parent;
++
++ sb = dentry->d_sb;
++ mfs = &au_sbi(sb)->si_wbr_mfs;
++ mutex_lock(&mfs->mfs_lock);
++ if (time_after(jiffies, mfs->mfs_jiffy + mfs->mfs_expire)
++ || mfs->mfs_bindex < 0)
++ au_mfs(dentry, /*parent*/NULL);
++ watermark = mfs->mfsrr_watermark;
++ bmfs = mfs->mfs_bindex;
++ mutex_unlock(&mfs->mfs_lock);
++
++ /* another style of au_wbr_create_exp() */
++ bwh = au_dbwh(dentry);
++ parent = dget_parent(dentry);
++ btail = au_dbtaildir(parent);
++ if (bwh >= 0 && bwh < btail)
++ btail = bwh;
++
++ err = au_wbr_nonopq(dentry, btail);
++ if (unlikely(err < 0))
++ goto out;
++ btail = err;
++ bfound = -1;
++ for (bindex = 0; bindex <= btail; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_rdonly(br))
++ continue;
++ if (br->br_wbr->wbr_bytes > watermark) {
++ bfound = bindex;
++ break;
++ }
++ }
++ err = bfound;
++ if (err < 0)
++ err = bmfs;
++
++out:
++ dput(parent);
++ AuDbg("b%d\n", err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
+/* most free space and then round robin */
+static int au_wbr_create_mfsrr(struct dentry *dentry, unsigned int flags)
+{
+ .init = au_wbr_create_init_mfsrr,
+ .fin = au_wbr_create_fin_mfs
+ },
++ [AuWbrCreate_TDMFS] = {
++ .create = au_wbr_create_tdmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_TDMFSV] = {
++ .create = au_wbr_create_tdmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
+ [AuWbrCreate_PMFS] = {
+ .create = au_wbr_create_pmfs,
+ .init = au_wbr_create_init_mfs,
+ .fin = au_wbr_create_fin_mfs
+ }
+};
-diff -urN /usr/share/empty/fs/aufs/whout.c linux/fs/aufs/whout.c
---- /usr/share/empty/fs/aufs/whout.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/whout.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,1060 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/whout.c linux-4.14/fs/aufs/whout.c
+--- linux-4.14/fs/aufs/whout.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/whout.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,1061 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+out_name:
+ if (name != defname)
-+ au_delayed_kfree(name);
++ kfree(name);
+out:
+ AuTraceErrPtr(dentry);
+ return dentry;
+
+ /* under the same dir, no need to lock_rename() */
+ delegated = NULL;
-+ err = vfsub_rename(h_dir, h_dentry, h_dir, &h_path, &delegated);
++ err = vfsub_rename(h_dir, h_dentry, h_dir, &h_path, &delegated,
++ /*flags*/0);
+ AuTraceErr(err);
+ if (unlikely(err == -EWOULDBLOCK)) {
+ pr_warn("cannot retry for NFSv4 delegation"
+ au_br_put(a->br);
+ si_write_unlock(a->sb);
+ au_nwt_done(&au_sbi(a->sb)->si_nowait);
-+ au_delayed_kfree(arg);
++ kfree(arg);
+ if (unlikely(err))
+ AuIOErr("err %d\n", err);
+}
+ if (unlikely(wkq_err)) {
+ atomic_dec(&br->br_wbr->wbr_wh_running);
+ au_br_put(br);
-+ au_delayed_kfree(arg);
++ kfree(arg);
+ }
+ do_dec = 0;
+ }
+ wh_dentry = ERR_PTR(err);
+ if (!err) {
+ wh_dentry = vfsub_lkup_one(&wh_name, h_parent);
-+ au_delayed_kfree(wh_name.name);
++ kfree(wh_name.name);
+ }
+ return wh_dentry;
+}
+ break;
+ }
+ }
-+ au_delayed_free_page((unsigned long)wh_name.name);
++ free_page((unsigned long)wh_name.name);
+
+out:
+ return err;
+ rdhash = AUFS_RDHASH_DEF;
+ err = au_nhash_alloc(&whtmp->whlist, rdhash, gfp);
+ if (unlikely(err)) {
-+ au_delayed_kfree(whtmp);
++ kfree(whtmp);
+ whtmp = ERR_PTR(err);
+ }
+
+ dput(whtmp->wh_dentry);
+ iput(whtmp->dir);
+ au_nhash_wh_free(&whtmp->whlist);
-+ au_delayed_kfree(whtmp);
++ kfree(whtmp);
+}
+
+/*
+ au_whtmp_rmdir_free(args);
+ }
+}
-diff -urN /usr/share/empty/fs/aufs/whout.h linux/fs/aufs/whout.h
---- /usr/share/empty/fs/aufs/whout.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/whout.h 2016-07-25 19:05:34.817826663 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/whout.h linux-4.14/fs/aufs/whout.h
+--- linux-4.14/fs/aufs/whout.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/whout.h 2021-02-24 21:42:43.447781402 +0100
@@ -0,0 +1,85 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_WHOUT_H__ */
-diff -urN /usr/share/empty/fs/aufs/wkq.c linux/fs/aufs/wkq.c
---- /usr/share/empty/fs/aufs/wkq.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/wkq.c 2016-08-17 18:01:06.161889951 +0200
-@@ -0,0 +1,213 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/wkq.c linux-4.14/fs/aufs/wkq.c
+--- linux-4.14/fs/aufs/wkq.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/wkq.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,390 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ au_wkq_func_t func;
+ void *args;
+
++#ifdef CONFIG_LOCKDEP
++ int dont_check;
++ struct held_lock **hlock;
++#endif
++
+ struct completion *comp;
+};
+
+/* ---------------------------------------------------------------------- */
++/*
++ * Aufs passes some operations to the workqueue such as the internal copyup.
++ * This scheme looks rather unnatural for LOCKDEP debugging feature, since the
++ * job run by workqueue depends upon the locks acquired in the other task.
++ * Delegating a small operation to the workqueue, aufs passes its lockdep
++ * information too. And the job in the workqueue restores the info in order to
++ * pretend as if it acquired those locks. This is just to make LOCKDEP work
++ * correctly and expectedly.
++ */
++
++#ifndef CONFIG_LOCKDEP
++AuStubInt0(au_wkq_lockdep_alloc, struct au_wkinfo *wkinfo);
++AuStubVoid(au_wkq_lockdep_free, struct au_wkinfo *wkinfo);
++AuStubVoid(au_wkq_lockdep_pre, struct au_wkinfo *wkinfo);
++AuStubVoid(au_wkq_lockdep_post, struct au_wkinfo *wkinfo);
++AuStubVoid(au_wkq_lockdep_init, struct au_wkinfo *wkinfo);
++#else
++static void au_wkq_lockdep_init(struct au_wkinfo *wkinfo)
++{
++ wkinfo->hlock = NULL;
++ wkinfo->dont_check = 0;
++}
++
++/*
++ * 1: matched
++ * 0: unmatched
++ */
++static int au_wkq_lockdep_test(struct lock_class_key *key, const char *name)
++{
++ static DEFINE_SPINLOCK(spin);
++ static struct {
++ char *name;
++ struct lock_class_key *key;
++ } a[] = {
++ { .name = "&sbinfo->si_rwsem" },
++ { .name = "&finfo->fi_rwsem" },
++ { .name = "&dinfo->di_rwsem" },
++ { .name = "&iinfo->ii_rwsem" }
++ };
++ static int set;
++ int i;
++
++ /* lockless read from 'set.' see below */
++ if (set == ARRAY_SIZE(a)) {
++ for (i = 0; i < ARRAY_SIZE(a); i++)
++ if (a[i].key == key)
++ goto match;
++ goto unmatch;
++ }
++
++ spin_lock(&spin);
++ if (set)
++ for (i = 0; i < ARRAY_SIZE(a); i++)
++ if (a[i].key == key) {
++ spin_unlock(&spin);
++ goto match;
++ }
++ for (i = 0; i < ARRAY_SIZE(a); i++) {
++ if (a[i].key) {
++ if (unlikely(a[i].key == key)) { /* rare but possible */
++ spin_unlock(&spin);
++ goto match;
++ } else
++ continue;
++ }
++ if (strstr(a[i].name, name)) {
++ /*
++ * the order of these three lines is important for the
++ * lockless read above.
++ */
++ a[i].key = key;
++ spin_unlock(&spin);
++ set++;
++ /* AuDbg("%d, %s\n", set, name); */
++ goto match;
++ }
++ }
++ spin_unlock(&spin);
++ goto unmatch;
++
++match:
++ return 1;
++unmatch:
++ return 0;
++}
++
++static int au_wkq_lockdep_alloc(struct au_wkinfo *wkinfo)
++{
++ int err, n;
++ struct task_struct *curr;
++ struct held_lock **hl, *held_locks, *p;
++
++ err = 0;
++ curr = current;
++ wkinfo->dont_check = lockdep_recursing(curr);
++ if (wkinfo->dont_check)
++ goto out;
++ n = curr->lockdep_depth;
++ if (!n)
++ goto out;
++
++ err = -ENOMEM;
++ wkinfo->hlock = kmalloc_array(n + 1, sizeof(*wkinfo->hlock), GFP_NOFS);
++ if (unlikely(!wkinfo->hlock))
++ goto out;
++
++ err = 0;
++#if 0
++ if (0 && au_debug_test()) /* left for debugging */
++ lockdep_print_held_locks(curr);
++#endif
++ held_locks = curr->held_locks;
++ hl = wkinfo->hlock;
++ while (n--) {
++ p = held_locks++;
++ if (au_wkq_lockdep_test(p->instance->key, p->instance->name))
++ *hl++ = p;
++ }
++ *hl = NULL;
++
++out:
++ return err;
++}
++
++static void au_wkq_lockdep_free(struct au_wkinfo *wkinfo)
++{
++ kfree(wkinfo->hlock);
++}
++
++static void au_wkq_lockdep_pre(struct au_wkinfo *wkinfo)
++{
++ struct held_lock *p, **hl = wkinfo->hlock;
++ int subclass;
++
++ if (wkinfo->dont_check)
++ lockdep_off();
++ if (!hl)
++ return;
++ while ((p = *hl++)) { /* assignment */
++ subclass = lockdep_hlock_class(p)->subclass;
++ /* AuDbg("%s, %d\n", p->instance->name, subclass); */
++ if (p->read)
++ rwsem_acquire_read(p->instance, subclass, 0,
++ /*p->acquire_ip*/_RET_IP_);
++ else
++ rwsem_acquire(p->instance, subclass, 0,
++ /*p->acquire_ip*/_RET_IP_);
++ }
++}
++
++static void au_wkq_lockdep_post(struct au_wkinfo *wkinfo)
++{
++ struct held_lock *p, **hl = wkinfo->hlock;
++
++ if (wkinfo->dont_check)
++ lockdep_on();
++ if (!hl)
++ return;
++ while ((p = *hl++)) /* assignment */
++ rwsem_release(p->instance, 0, /*p->acquire_ip*/_RET_IP_);
++}
++#endif
+
+static void wkq_func(struct work_struct *wk)
+{
+ AuDebugOn(!uid_eq(current_fsuid(), GLOBAL_ROOT_UID));
+ AuDebugOn(rlimit(RLIMIT_FSIZE) != RLIM_INFINITY);
+
++ au_wkq_lockdep_pre(wkinfo);
+ wkinfo->func(wkinfo->args);
++ au_wkq_lockdep_post(wkinfo);
+ if (au_ftest_wkq(wkinfo->flags, WAIT))
+ complete(wkinfo->comp);
+ else {
+ kobject_put(wkinfo->kobj);
+ module_put(THIS_MODULE); /* todo: ?? */
-+ au_delayed_kfree(wkinfo);
++ kfree(wkinfo);
+ }
+}
+
+
+static void au_wkq_comp_free(struct completion *comp)
+{
-+ au_delayed_kfree(comp);
++ kfree(comp);
+}
+
+#else
+ };
+
+ err = au_wkq_comp_alloc(&wkinfo, &comp);
++ if (unlikely(err))
++ goto out;
++ err = au_wkq_lockdep_alloc(&wkinfo);
++ if (unlikely(err))
++ goto out_comp;
+ if (!err) {
+ au_wkq_run(&wkinfo);
+ /* no timeout, no interrupt */
+ wait_for_completion(wkinfo.comp);
-+ au_wkq_comp_free(comp);
-+ destroy_work_on_stack(&wkinfo.wk);
+ }
++ au_wkq_lockdep_free(&wkinfo);
+
++out_comp:
++ au_wkq_comp_free(comp);
++out:
++ destroy_work_on_stack(&wkinfo.wk);
+ return err;
-+
+}
+
+/*
+ wkinfo->func = func;
+ wkinfo->args = args;
+ wkinfo->comp = NULL;
++ au_wkq_lockdep_init(wkinfo);
+ kobject_get(wkinfo->kobj);
+ __module_get(THIS_MODULE); /* todo: ?? */
+
+
+ return err;
+}
-diff -urN /usr/share/empty/fs/aufs/wkq.h linux/fs/aufs/wkq.h
---- /usr/share/empty/fs/aufs/wkq.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/wkq.h 2016-08-17 18:01:06.161889951 +0200
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/wkq.h linux-4.14/fs/aufs/wkq.h
+--- linux-4.14/fs/aufs/wkq.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/wkq.h 2021-02-24 21:42:43.447781402 +0100
@@ -0,0 +1,93 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#ifdef __KERNEL__
+
-+#include <linux/percpu_counter.h>
++#include <linux/wait.h>
+
+struct super_block;
+
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_WKQ_H__ */
-diff -urN /usr/share/empty/fs/aufs/xattr.c linux/fs/aufs/xattr.c
---- /usr/share/empty/fs/aufs/xattr.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/xattr.c 2016-08-17 18:01:06.165223371 +0200
-@@ -0,0 +1,347 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/xattr.c linux-4.14/fs/aufs/xattr.c
+--- linux-4.14/fs/aufs/xattr.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/xattr.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,355 @@
+/*
-+ * Copyright (C) 2014-2016 Junjiro R. Okajima
++ * Copyright (C) 2014-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * handling xattr functions
+ */
+
++#include <linux/fs.h>
++#include <linux/posix_acl_xattr.h>
+#include <linux/xattr.h>
+#include "aufs.h"
+
+ h_isrc = d_inode(h_src);
+ h_idst = d_inode(h_dst);
+ inode_unlock(h_idst);
-+ inode_lock_nested(h_isrc, AuLsc_I_CHILD);
++ vfsub_inode_lock_shared_nested(h_isrc, AuLsc_I_CHILD);
+ inode_lock_nested(h_idst, AuLsc_I_CHILD2);
+ unlocked = 0;
+
+ goto out;
+ err = vfs_listxattr(h_src, p, ssz);
+ }
-+ inode_unlock(h_isrc);
++ inode_unlock_shared(h_isrc);
+ unlocked = 1;
+ AuDbg("err %d, ssz %zd\n", err, ssz);
+ if (unlikely(err < 0))
+ AuTraceErr(err);
+ }
+
-+ if (value)
-+ au_delayed_kfree(value);
++ kfree(value);
+
+out_free:
-+ if (o)
-+ au_delayed_kfree(o);
++ kfree(o);
+out:
+ if (!unlocked)
-+ inode_unlock(h_isrc);
++ inode_unlock_shared(h_isrc);
+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
++static int au_smack_reentering(struct super_block *sb)
++{
++#if IS_ENABLED(CONFIG_SECURITY_SMACK)
++ /*
++ * as a part of lookup, smack_d_instantiate() is called, and it calls
++ * i_op->getxattr(). ouch.
++ */
++ return si_pid_test(sb);
++#else
++ return 0;
++#endif
++}
++
+enum {
+ AU_XATTR_LIST,
+ AU_XATTR_GET
+static ssize_t au_lgxattr(struct dentry *dentry, struct au_lgxattr *arg)
+{
+ ssize_t err;
++ int reenter;
+ struct path h_path;
+ struct super_block *sb;
+
+ sb = dentry->d_sb;
-+ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_h_path_getattr(dentry, /*force*/1, &h_path);
++ reenter = au_smack_reentering(sb);
++ if (!reenter) {
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out;
++ }
++ err = au_h_path_getattr(dentry, /*force*/1, &h_path, reenter);
+ if (unlikely(err))
+ goto out_si;
+ if (unlikely(!h_path.dentry))
+ }
+
+out_di:
-+ di_read_unlock(dentry, AuLock_IR);
++ if (!reenter)
++ di_read_unlock(dentry, AuLock_IR);
+out_si:
-+ si_read_unlock(sb);
++ if (!reenter)
++ si_read_unlock(sb);
+out:
+ AuTraceErr(err);
+ return err;
+ return au_lgxattr(dentry, &arg);
+}
+
-+ssize_t aufs_getxattr(struct dentry *dentry, struct inode *inode __maybe_unused,
-+ const char *name, void *value, size_t size)
++static ssize_t au_getxattr(struct dentry *dentry,
++ struct inode *inode __maybe_unused,
++ const char *name, void *value, size_t size)
+{
+ struct au_lgxattr arg = {
+ .type = AU_XATTR_GET,
+ return au_lgxattr(dentry, &arg);
+}
+
-+int aufs_setxattr(struct dentry *dentry, struct inode *inode, const char *name,
-+ const void *value, size_t size, int flags)
++static int au_setxattr(struct dentry *dentry, struct inode *inode,
++ const char *name, const void *value, size_t size,
++ int flags)
+{
-+ struct au_srxattr arg = {
++ struct au_sxattr arg = {
+ .type = AU_XATTR_SET,
+ .u.set = {
+ .name = name,
+ },
+ };
+
-+ return au_srxattr(dentry, inode, &arg);
-+}
-+
-+int aufs_removexattr(struct dentry *dentry, const char *name)
-+{
-+ struct au_srxattr arg = {
-+ .type = AU_XATTR_REMOVE,
-+ .u.remove = {
-+ .name = name
-+ },
-+ };
-+
-+ return au_srxattr(dentry, d_inode(dentry), &arg);
++ return au_sxattr(dentry, inode, &arg);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+#if 0
-+static size_t au_xattr_list(struct dentry *dentry, char *list, size_t list_size,
-+ const char *name, size_t name_len, int type)
-+{
-+ return aufs_listxattr(dentry, list, list_size);
-+}
-+
-+static int au_xattr_get(struct dentry *dentry, const char *name, void *buffer,
-+ size_t size, int type)
++static int au_xattr_get(const struct xattr_handler *handler,
++ struct dentry *dentry, struct inode *inode,
++ const char *name, void *buffer, size_t size)
+{
-+ return aufs_getxattr(dentry, name, buffer, size);
++ return au_getxattr(dentry, inode, name, buffer, size);
+}
+
-+static int au_xattr_set(struct dentry *dentry, const char *name,
-+ const void *value, size_t size, int flags, int type)
++static int au_xattr_set(const struct xattr_handler *handler,
++ struct dentry *dentry, struct inode *inode,
++ const char *name, const void *value, size_t size,
++ int flags)
+{
-+ return aufs_setxattr(dentry, name, value, size, flags);
++ return au_setxattr(dentry, inode, name, value, size, flags);
+}
+
+static const struct xattr_handler au_xattr_handler = {
-+ /* no prefix, no flags */
-+ .list = au_xattr_list,
++ .name = "",
++ .prefix = "",
+ .get = au_xattr_get,
+ .set = au_xattr_set
-+ /* why no remove? */
+};
+
+static const struct xattr_handler *au_xattr_handlers[] = {
-+ &au_xattr_handler
++#ifdef CONFIG_FS_POSIX_ACL
++ &posix_acl_access_xattr_handler,
++ &posix_acl_default_xattr_handler,
++#endif
++ &au_xattr_handler, /* must be last */
++ NULL
+};
+
+void au_xattr_init(struct super_block *sb)
+{
-+ /* sb->s_xattr = au_xattr_handlers; */
++ sb->s_xattr = au_xattr_handlers;
+}
-+#endif
-diff -urN /usr/share/empty/fs/aufs/xino.c linux/fs/aufs/xino.c
---- /usr/share/empty/fs/aufs/xino.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux/fs/aufs/xino.c 2016-08-17 18:01:06.165223371 +0200
-@@ -0,0 +1,1318 @@
+diff -urNp -x '*.orig' linux-4.14/fs/aufs/xino.c linux-4.14/fs/aufs/xino.c
+--- linux-4.14/fs/aufs/xino.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/fs/aufs/xino.c 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,1418 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ lockdep_off();
+ err = do_xino_fwrite(func, file, buf, size, pos);
+ lockdep_on();
-+ } else
++ } else {
++ lockdep_off();
+ err = xino_fwrite_wkq(func, file, buf, size, pos);
++ lockdep_on();
++ }
+
+ return err;
+}
+ AuErr1("statfs err %d, ignored\n", err);
+
+out_st:
-+ au_delayed_kfree(st);
++ kfree(st);
+out:
+ return err;
+}
+ au_br_put(br);
+ si_write_unlock(sb);
+ au_nwt_done(&au_sbi(sb)->si_nowait);
-+ au_delayed_kfree(args);
++ kfree(args);
+}
+
+static int xino_trunc_test(struct super_block *sb, struct au_branch *br)
+
+ pr_err("wkq %d\n", wkq_err);
+ au_br_put(br);
-+ au_delayed_kfree(args);
++ kfree(args);
+
+out:
+ atomic_dec(&br->br_xino_running);
+ (sb, au_sbr(sb, bindex)->br_xino.xi_file, page);
+ else
+ AuDbg("b%d\n", bindex);
-+ au_delayed_free_page((unsigned long)page);
++ free_page((unsigned long)page);
+
+out:
+ return err;
+ fput(sbinfo->si_xib);
+ sbinfo->si_xib = NULL;
+ if (sbinfo->si_xib_buf)
-+ au_delayed_free_page((unsigned long)sbinfo->si_xib_buf);
++ free_page((unsigned long)sbinfo->si_xib_buf);
+ sbinfo->si_xib_buf = NULL;
+}
+
+
+out_free:
+ if (sbinfo->si_xib_buf)
-+ au_delayed_free_page((unsigned long)sbinfo->si_xib_buf);
++ free_page((unsigned long)sbinfo->si_xib_buf);
+ sbinfo->si_xib_buf = NULL;
+ if (err >= 0)
+ err = -EIO;
+ fput(p->new);
+ else
+ break;
-+ au_delayed_kfree(fpair);
++ kfree(fpair);
+out:
+ return err;
+}
+ if (!IS_ERR(file))
+ au_xino_brid_set(sb, br->br_id);
+ }
-+ au_delayed_free_page((unsigned long)page);
++ free_page((unsigned long)page);
+ } else {
+ file = au_xino_create(sb, AUFS_XINO_DEFPATH, /*silent*/0);
+ if (IS_ERR(file))
+ return file;
+}
+
-+/* ---------------------------------------------------------------------- */
++/* ---------------------------------------------------------------------- */
++
++int au_xino_path(struct seq_file *seq, struct file *file)
++{
++ int err;
++
++ err = au_seq_path(seq, &file->f_path);
++ if (unlikely(err))
++ goto out;
++
++#define Deleted "\\040(deleted)"
++ seq->count -= sizeof(Deleted) - 1;
++ AuDebugOn(memcmp(seq->buf + seq->count, Deleted,
++ sizeof(Deleted) - 1));
++#undef Deleted
++
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++void au_xinondir_leave(struct super_block *sb, aufs_bindex_t bindex,
++ ino_t h_ino, int idx)
++{
++ struct au_xino_file *xino;
++
++ AuDebugOn(!au_opt_test(au_mntflags(sb), XINO));
++ xino = &au_sbr(sb, bindex)->br_xino;
++ AuDebugOn(idx < 0 || xino->xi_nondir.total <= idx);
++
++ spin_lock(&xino->xi_nondir.spin);
++ AuDebugOn(xino->xi_nondir.array[idx] != h_ino);
++ xino->xi_nondir.array[idx] = 0;
++ spin_unlock(&xino->xi_nondir.spin);
++ wake_up_all(&xino->xi_nondir.wqh);
++}
++
++static int au_xinondir_find(struct au_xino_file *xino, ino_t h_ino)
++{
++ int found, total, i;
++
++ found = -1;
++ total = xino->xi_nondir.total;
++ for (i = 0; i < total; i++) {
++ if (xino->xi_nondir.array[i] != h_ino)
++ continue;
++ found = i;
++ break;
++ }
++
++ return found;
++}
++
++static int au_xinondir_expand(struct au_xino_file *xino)
++{
++ int err, sz;
++ ino_t *p;
++
++ BUILD_BUG_ON(KMALLOC_MAX_SIZE > INT_MAX);
++
++ err = -ENOMEM;
++ sz = xino->xi_nondir.total * sizeof(ino_t);
++ if (unlikely(sz > KMALLOC_MAX_SIZE / 2))
++ goto out;
++ p = au_kzrealloc(xino->xi_nondir.array, sz, sz << 1, GFP_ATOMIC,
++ /*may_shrink*/0);
++ if (p) {
++ xino->xi_nondir.array = p;
++ xino->xi_nondir.total <<= 1;
++ AuDbg("xi_nondir.total %d\n", xino->xi_nondir.total);
++ err = 0;
++ }
++
++out:
++ return err;
++}
++
++int au_xinondir_enter(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ int *idx)
++{
++ int err, found, empty;
++ struct au_xino_file *xino;
++
++ err = 0;
++ *idx = -1;
++ if (!au_opt_test(au_mntflags(sb), XINO))
++ goto out; /* no xino */
++
++ xino = &au_sbr(sb, bindex)->br_xino;
++
++again:
++ spin_lock(&xino->xi_nondir.spin);
++ found = au_xinondir_find(xino, h_ino);
++ if (found == -1) {
++ empty = au_xinondir_find(xino, /*h_ino*/0);
++ if (empty == -1) {
++ empty = xino->xi_nondir.total;
++ err = au_xinondir_expand(xino);
++ if (unlikely(err))
++ goto out_unlock;
++ }
++ xino->xi_nondir.array[empty] = h_ino;
++ *idx = empty;
++ } else {
++ spin_unlock(&xino->xi_nondir.spin);
++ wait_event(xino->xi_nondir.wqh,
++ xino->xi_nondir.array[found] != h_ino);
++ goto again;
++ }
++
++out_unlock:
++ spin_unlock(&xino->xi_nondir.spin);
++out:
++ return err;
++}
+diff -urNp -x '*.orig' linux-4.14/fs/dcache.c linux-4.14/fs/dcache.c
+--- linux-4.14/fs/dcache.c 2021-02-24 21:42:35.464465471 +0100
++++ linux-4.14/fs/dcache.c 2021-02-24 21:42:43.437781422 +0100
+@@ -1217,7 +1217,7 @@ enum d_walk_ret {
+ *
+ * The @enter() and @finish() callbacks are called with d_lock held.
+ */
+-static void d_walk(struct dentry *parent, void *data,
++void d_walk(struct dentry *parent, void *data,
+ enum d_walk_ret (*enter)(void *, struct dentry *),
+ void (*finish)(void *))
+ {
+@@ -1325,6 +1325,7 @@ rename_retry:
+ seq = 1;
+ goto again;
+ }
++EXPORT_SYMBOL_GPL(d_walk);
+
+ struct check_mount {
+ struct vfsmount *mnt;
+@@ -2950,6 +2951,7 @@ void d_exchange(struct dentry *dentry1,
+
+ write_sequnlock(&rename_lock);
+ }
++EXPORT_SYMBOL_GPL(d_exchange);
+
+ /**
+ * d_ancestor - search for an ancestor
+diff -urNp -x '*.orig' linux-4.14/fs/exec.c linux-4.14/fs/exec.c
+--- linux-4.14/fs/exec.c 2021-02-24 21:42:35.467798797 +0100
++++ linux-4.14/fs/exec.c 2021-02-24 21:42:43.437781422 +0100
+@@ -109,6 +109,7 @@ bool path_noexec(const struct path *path
+ return (path->mnt->mnt_flags & MNT_NOEXEC) ||
+ (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC);
+ }
++EXPORT_SYMBOL_GPL(path_noexec);
+
+ #ifdef CONFIG_USELIB
+ /*
+diff -urNp -x '*.orig' linux-4.14/fs/fcntl.c linux-4.14/fs/fcntl.c
+--- linux-4.14/fs/fcntl.c 2021-02-24 21:42:35.474465447 +0100
++++ linux-4.14/fs/fcntl.c 2021-02-24 21:42:43.437781422 +0100
+@@ -32,7 +32,7 @@
+
+ #define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
+
+-static int setfl(int fd, struct file * filp, unsigned long arg)
++int setfl(int fd, struct file * filp, unsigned long arg)
+ {
+ struct inode * inode = file_inode(filp);
+ int error = 0;
+@@ -63,6 +63,8 @@ static int setfl(int fd, struct file * f
+
+ if (filp->f_op->check_flags)
+ error = filp->f_op->check_flags(arg);
++ if (!error && filp->f_op->setfl)
++ error = filp->f_op->setfl(filp, arg);
+ if (error)
+ return error;
+
+@@ -83,6 +85,7 @@ static int setfl(int fd, struct file * f
+ out:
+ return error;
+ }
++EXPORT_SYMBOL_GPL(setfl);
+
+ static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
+ int force)
+diff -urNp -x '*.orig' linux-4.14/fs/file_table.c linux-4.14/fs/file_table.c
+--- linux-4.14/fs/file_table.c 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/file_table.c 2021-02-24 21:42:43.437781422 +0100
+@@ -148,6 +148,7 @@ over:
+ }
+ return ERR_PTR(-ENFILE);
+ }
++EXPORT_SYMBOL_GPL(get_empty_filp);
+
+ /**
+ * alloc_file - allocate and initialize a 'struct file'
+@@ -258,6 +259,7 @@ void flush_delayed_fput(void)
+ {
+ delayed_fput(NULL);
+ }
++EXPORT_SYMBOL_GPL(flush_delayed_fput);
+
+ static DECLARE_DELAYED_WORK(delayed_fput_work, delayed_fput);
+
+@@ -300,6 +302,7 @@ void __fput_sync(struct file *file)
+ }
+
+ EXPORT_SYMBOL(fput);
++EXPORT_SYMBOL_GPL(__fput_sync);
+
+ void put_filp(struct file *file)
+ {
+@@ -308,6 +311,7 @@ void put_filp(struct file *file)
+ file_free(file);
+ }
+ }
++EXPORT_SYMBOL_GPL(put_filp);
+
+ void __init files_init(void)
+ {
+diff -urNp -x '*.orig' linux-4.14/fs/inode.c linux-4.14/fs/inode.c
+--- linux-4.14/fs/inode.c 2021-02-24 21:42:35.481132097 +0100
++++ linux-4.14/fs/inode.c 2021-02-24 21:42:43.437781422 +0100
+@@ -1664,7 +1664,7 @@ EXPORT_SYMBOL(generic_update_time);
+ * This does the actual work of updating an inodes time or version. Must have
+ * had called mnt_want_write() before calling this.
+ */
+-static int update_time(struct inode *inode, struct timespec *time, int flags)
++int update_time(struct inode *inode, struct timespec *time, int flags)
+ {
+ int (*update_time)(struct inode *, struct timespec *, int);
+
+@@ -1673,6 +1673,7 @@ static int update_time(struct inode *ino
+
+ return update_time(inode, time, flags);
+ }
++EXPORT_SYMBOL_GPL(update_time);
+
+ /**
+ * touch_atime - update the access time
+diff -urNp -x '*.orig' linux-4.14/fs/namespace.c linux-4.14/fs/namespace.c
+--- linux-4.14/fs/namespace.c 2021-02-24 21:42:35.481132097 +0100
++++ linux-4.14/fs/namespace.c 2021-02-24 21:42:43.437781422 +0100
+@@ -517,6 +517,7 @@ void __mnt_drop_write(struct vfsmount *m
+ mnt_dec_writers(real_mount(mnt));
+ preempt_enable();
+ }
++EXPORT_SYMBOL_GPL(__mnt_drop_write);
+
+ /**
+ * mnt_drop_write - give up write access to a mount
+@@ -855,6 +856,13 @@ static inline int check_mnt(struct mount
+ return mnt->mnt_ns == current->nsproxy->mnt_ns;
+ }
+
++/* for aufs, CONFIG_AUFS_BR_FUSE */
++int is_current_mnt_ns(struct vfsmount *mnt)
++{
++ return check_mnt(real_mount(mnt));
++}
++EXPORT_SYMBOL_GPL(is_current_mnt_ns);
++
+ /*
+ * vfsmount lock must be held for write
+ */
+@@ -1918,6 +1926,7 @@ int iterate_mounts(int (*f)(struct vfsmo
+ }
+ return 0;
+ }
++EXPORT_SYMBOL_GPL(iterate_mounts);
+
+ static void cleanup_group_ids(struct mount *mnt, struct mount *end)
+ {
+diff -urNp -x '*.orig' linux-4.14/fs/notify/group.c linux-4.14/fs/notify/group.c
+--- linux-4.14/fs/notify/group.c 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/notify/group.c 2021-02-24 21:42:43.437781422 +0100
+@@ -22,6 +22,7 @@
+ #include <linux/srcu.h>
+ #include <linux/rculist.h>
+ #include <linux/wait.h>
++#include <linux/module.h>
+
+ #include <linux/fsnotify_backend.h>
+ #include "fsnotify.h"
+@@ -109,6 +110,7 @@ void fsnotify_get_group(struct fsnotify_
+ {
+ atomic_inc(&group->refcnt);
+ }
++EXPORT_SYMBOL_GPL(fsnotify_get_group);
+
+ /*
+ * Drop a reference to a group. Free it if it's through.
+@@ -118,6 +120,7 @@ void fsnotify_put_group(struct fsnotify_
+ if (atomic_dec_and_test(&group->refcnt))
+ fsnotify_final_destroy_group(group);
+ }
++EXPORT_SYMBOL_GPL(fsnotify_put_group);
+
+ /*
+ * Create a new fsnotify_group and hold a reference for the group returned.
+@@ -147,6 +150,7 @@ struct fsnotify_group *fsnotify_alloc_gr
+
+ return group;
+ }
++EXPORT_SYMBOL_GPL(fsnotify_alloc_group);
+
+ int fsnotify_fasync(int fd, struct file *file, int on)
+ {
+diff -urNp -x '*.orig' linux-4.14/fs/notify/mark.c linux-4.14/fs/notify/mark.c
+--- linux-4.14/fs/notify/mark.c 2021-02-24 21:42:35.491132073 +0100
++++ linux-4.14/fs/notify/mark.c 2021-02-24 21:42:43.437781422 +0100
+@@ -124,6 +124,7 @@ static void __fsnotify_recalc_mask(struc
+ else if (conn->flags & FSNOTIFY_OBJ_TYPE_VFSMOUNT)
+ real_mount(conn->mnt)->mnt_fsnotify_mask = new_mask;
+ }
++EXPORT_SYMBOL_GPL(fsnotify_put_mark);
+
+ /*
+ * Calculate mask of events for a list of marks. The caller must make sure
+@@ -392,6 +393,7 @@ void fsnotify_destroy_mark(struct fsnoti
+ mutex_unlock(&group->mark_mutex);
+ fsnotify_free_mark(mark);
+ }
++EXPORT_SYMBOL_GPL(fsnotify_destroy_mark);
+
+ /*
+ * Sorting function for lists of fsnotify marks.
+@@ -604,6 +606,7 @@ err:
+ fsnotify_put_mark(mark);
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(fsnotify_add_mark);
+
+ int fsnotify_add_mark(struct fsnotify_mark *mark, struct inode *inode,
+ struct vfsmount *mnt, int allow_dups)
+@@ -739,6 +742,7 @@ void fsnotify_init_mark(struct fsnotify_
+ fsnotify_get_group(group);
+ mark->group = group;
+ }
++EXPORT_SYMBOL_GPL(fsnotify_init_mark);
+
+ /*
+ * Destroy all marks in destroy_list, waits for SRCU period to finish before
+diff -urNp -x '*.orig' linux-4.14/fs/open.c linux-4.14/fs/open.c
+--- linux-4.14/fs/open.c 2021-02-24 21:42:35.494465399 +0100
++++ linux-4.14/fs/open.c 2021-02-24 21:42:43.437781422 +0100
+@@ -64,6 +64,7 @@ int do_truncate(struct dentry *dentry, l
+ inode_unlock(dentry->d_inode);
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(do_truncate);
+
+ long vfs_truncate(const struct path *path, loff_t length)
+ {
+@@ -710,6 +711,7 @@ int open_check_o_direct(struct file *f)
+ }
+ return 0;
+ }
++EXPORT_SYMBOL_GPL(open_check_o_direct);
+
+ static int do_dentry_open(struct file *f,
+ struct inode *inode,
+diff -urNp -x '*.orig' linux-4.14/fs/proc/base.c linux-4.14/fs/proc/base.c
+--- linux-4.14/fs/proc/base.c 2021-02-24 21:42:35.494465399 +0100
++++ linux-4.14/fs/proc/base.c 2021-02-24 21:42:43.434448095 +0100
+@@ -2030,7 +2030,7 @@ static int map_files_get_link(struct den
+ down_read(&mm->mmap_sem);
+ vma = find_exact_vma(mm, vm_start, vm_end);
+ if (vma && vma->vm_file) {
+- *path = vma->vm_file->f_path;
++ *path = vma_pr_or_file(vma)->f_path;
+ path_get(path);
+ rc = 0;
+ }
+diff -urNp -x '*.orig' linux-4.14/fs/proc/nommu.c linux-4.14/fs/proc/nommu.c
+--- linux-4.14/fs/proc/nommu.c 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/proc/nommu.c 2021-02-24 21:42:43.434448095 +0100
+@@ -45,7 +45,10 @@ static int nommu_region_show(struct seq_
+ file = region->vm_file;
+
+ if (file) {
+- struct inode *inode = file_inode(region->vm_file);
++ struct inode *inode;
++
++ file = vmr_pr_or_file(region);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ }
+diff -urNp -x '*.orig' linux-4.14/fs/proc/task_mmu.c linux-4.14/fs/proc/task_mmu.c
+--- linux-4.14/fs/proc/task_mmu.c 2021-02-24 21:42:35.497798724 +0100
++++ linux-4.14/fs/proc/task_mmu.c 2021-02-24 21:42:43.434448095 +0100
+@@ -310,7 +310,10 @@ show_map_vma(struct seq_file *m, struct
+ const char *name = NULL;
+
+ if (file) {
+- struct inode *inode = file_inode(vma->vm_file);
++ struct inode *inode;
++
++ file = vma_pr_or_file(vma);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT;
+@@ -1775,7 +1778,7 @@ static int show_numa_map(struct seq_file
+ struct proc_maps_private *proc_priv = &numa_priv->proc_maps;
+ struct vm_area_struct *vma = v;
+ struct numa_maps *md = &numa_priv->md;
+- struct file *file = vma->vm_file;
++ struct file *file = vma_pr_or_file(vma);
+ struct mm_struct *mm = vma->vm_mm;
+ struct mm_walk walk = {
+ .hugetlb_entry = gather_hugetlb_stats,
+diff -urNp -x '*.orig' linux-4.14/fs/proc/task_nommu.c linux-4.14/fs/proc/task_nommu.c
+--- linux-4.14/fs/proc/task_nommu.c 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/proc/task_nommu.c 2021-02-24 21:42:43.434448095 +0100
+@@ -156,7 +156,10 @@ static int nommu_vma_show(struct seq_fil
+ file = vma->vm_file;
+
+ if (file) {
+- struct inode *inode = file_inode(vma->vm_file);
++ struct inode *inode;
++
++ file = vma_pr_or_file(vma);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
+diff -urNp -x '*.orig' linux-4.14/fs/read_write.c linux-4.14/fs/read_write.c
+--- linux-4.14/fs/read_write.c 2021-02-24 21:42:35.497798724 +0100
++++ linux-4.14/fs/read_write.c 2021-02-24 21:42:43.437781422 +0100
+@@ -454,6 +454,7 @@ ssize_t vfs_read(struct file *file, char
+
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(vfs_read);
+
+ static ssize_t new_sync_write(struct file *filp, const char __user *buf, size_t len, loff_t *ppos)
+ {
+@@ -484,6 +485,30 @@ ssize_t __vfs_write(struct file *file, c
+ return -EINVAL;
+ }
+
++vfs_readf_t vfs_readf(struct file *file)
++{
++ const struct file_operations *fop = file->f_op;
++
++ if (fop->read)
++ return fop->read;
++ if (fop->read_iter)
++ return new_sync_read;
++ return ERR_PTR(-ENOSYS);
++}
++EXPORT_SYMBOL_GPL(vfs_readf);
++
++vfs_writef_t vfs_writef(struct file *file)
++{
++ const struct file_operations *fop = file->f_op;
++
++ if (fop->write)
++ return fop->write;
++ if (fop->write_iter)
++ return new_sync_write;
++ return ERR_PTR(-ENOSYS);
++}
++EXPORT_SYMBOL_GPL(vfs_writef);
++
+ ssize_t __kernel_write(struct file *file, const void *buf, size_t count, loff_t *pos)
+ {
+ mm_segment_t old_fs;
+@@ -552,6 +577,7 @@ ssize_t vfs_write(struct file *file, con
+
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(vfs_write);
+
+ static inline loff_t file_pos_read(struct file *file)
+ {
+diff -urNp -x '*.orig' linux-4.14/fs/splice.c linux-4.14/fs/splice.c
+--- linux-4.14/fs/splice.c 2021-02-24 21:42:35.497798724 +0100
++++ linux-4.14/fs/splice.c 2021-02-24 21:42:43.437781422 +0100
+@@ -837,8 +837,8 @@ EXPORT_SYMBOL(generic_splice_sendpage);
+ /*
+ * Attempt to initiate a splice from pipe to file.
+ */
+-static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
+- loff_t *ppos, size_t len, unsigned int flags)
++long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
++ loff_t *ppos, size_t len, unsigned int flags)
+ {
+ ssize_t (*splice_write)(struct pipe_inode_info *, struct file *,
+ loff_t *, size_t, unsigned int);
+@@ -850,13 +850,14 @@ static long do_splice_from(struct pipe_i
+
+ return splice_write(pipe, out, ppos, len, flags);
+ }
++EXPORT_SYMBOL_GPL(do_splice_from);
+
+ /*
+ * Attempt to initiate a splice from a file to a pipe.
+ */
+-static long do_splice_to(struct file *in, loff_t *ppos,
+- struct pipe_inode_info *pipe, size_t len,
+- unsigned int flags)
++long do_splice_to(struct file *in, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags)
+ {
+ ssize_t (*splice_read)(struct file *, loff_t *,
+ struct pipe_inode_info *, size_t, unsigned int);
+@@ -879,6 +880,7 @@ static long do_splice_to(struct file *in
+
+ return splice_read(in, ppos, pipe, len, flags);
+ }
++EXPORT_SYMBOL_GPL(do_splice_to);
+
+ /**
+ * splice_direct_to_actor - splices data directly between two non-pipes
+diff -urNp -x '*.orig' linux-4.14/fs/sync.c linux-4.14/fs/sync.c
+--- linux-4.14/fs/sync.c 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/fs/sync.c 2021-02-24 21:42:43.437781422 +0100
+@@ -28,7 +28,7 @@
+ * wait == 1 case since in that case write_inode() functions do
+ * sync_dirty_buffer() and thus effectively write one block at a time.
+ */
+-static int __sync_filesystem(struct super_block *sb, int wait)
++int __sync_filesystem(struct super_block *sb, int wait)
+ {
+ if (wait)
+ sync_inodes_sb(sb);
+@@ -39,6 +39,7 @@ static int __sync_filesystem(struct supe
+ sb->s_op->sync_fs(sb, wait);
+ return __sync_blockdev(sb->s_bdev, wait);
+ }
++EXPORT_SYMBOL_GPL(__sync_filesystem);
+
+ /*
+ * Write out and wait upon all dirty data associated with this
+diff -urNp -x '*.orig' linux-4.14/fs/xattr.c linux-4.14/fs/xattr.c
+--- linux-4.14/fs/xattr.c 2021-02-24 21:42:35.501132049 +0100
++++ linux-4.14/fs/xattr.c 2021-02-24 21:42:43.437781422 +0100
+@@ -334,6 +334,7 @@ vfs_getxattr_alloc(struct dentry *dentry
+ *xattr_value = value;
+ return error;
+ }
++EXPORT_SYMBOL_GPL(vfs_getxattr_alloc);
+
+ ssize_t
+ __vfs_getxattr(struct dentry *dentry, struct inode *inode, const char *name,
+diff -urNp -x '*.orig' linux-4.14/include/linux/file.h linux-4.14/include/linux/file.h
+--- linux-4.14/include/linux/file.h 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/include/linux/file.h 2021-02-24 21:42:43.434448095 +0100
+@@ -20,6 +20,7 @@ struct dentry;
+ struct path;
+ extern struct file *alloc_file(const struct path *, fmode_t mode,
+ const struct file_operations *fop);
++extern struct file *get_empty_filp(void);
+
+ static inline void fput_light(struct file *file, int fput_needed)
+ {
+diff -urNp -x '*.orig' linux-4.14/include/linux/fs.h linux-4.14/include/linux/fs.h
+--- linux-4.14/include/linux/fs.h 2021-02-24 21:42:35.511132025 +0100
++++ linux-4.14/include/linux/fs.h 2021-02-24 21:42:43.447781402 +0100
+@@ -1271,6 +1271,7 @@ extern void fasync_free(struct fasync_st
+ /* can be called from interrupts */
+ extern void kill_fasync(struct fasync_struct **, int, int);
+
++extern int setfl(int fd, struct file * filp, unsigned long arg);
+ extern void __f_setown(struct file *filp, struct pid *, enum pid_type, int force);
+ extern int f_setown(struct file *filp, unsigned long arg, int force);
+ extern void f_delown(struct file *filp);
+@@ -1718,6 +1719,7 @@ struct file_operations {
+ ssize_t (*sendpage) (struct file *, struct page *, int, size_t, loff_t *, int);
+ unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
+ int (*check_flags)(int);
++ int (*setfl)(struct file *, unsigned long);
+ int (*flock) (struct file *, int, struct file_lock *);
+ ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
+ ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
+@@ -1788,6 +1790,12 @@ ssize_t rw_copy_check_uvector(int type,
+ struct iovec *fast_pointer,
+ struct iovec **ret_pointer);
+
++typedef ssize_t (*vfs_readf_t)(struct file *, char __user *, size_t, loff_t *);
++typedef ssize_t (*vfs_writef_t)(struct file *, const char __user *, size_t,
++ loff_t *);
++vfs_readf_t vfs_readf(struct file *file);
++vfs_writef_t vfs_writef(struct file *file);
++
+ extern ssize_t __vfs_read(struct file *, char __user *, size_t, loff_t *);
+ extern ssize_t vfs_read(struct file *, char __user *, size_t, loff_t *);
+ extern ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *);
+@@ -1840,6 +1848,10 @@ struct super_operations {
+ struct shrink_control *);
+ long (*free_cached_objects)(struct super_block *,
+ struct shrink_control *);
++#if defined(CONFIG_BLK_DEV_LOOP) || defined(CONFIG_BLK_DEV_LOOP_MODULE)
++ /* and aufs */
++ struct file *(*real_loop)(struct file *);
++#endif
+ };
+
+ /*
+@@ -2196,6 +2208,7 @@ extern int current_umask(void);
+ extern void ihold(struct inode * inode);
+ extern void iput(struct inode *);
+ extern int generic_update_time(struct inode *, struct timespec *, int);
++extern int update_time(struct inode *, struct timespec *, int);
+
+ /* /sys/fs */
+ extern struct kobject *fs_kobj;
+@@ -2476,6 +2489,7 @@ static inline bool sb_is_blkdev_sb(struc
+ return false;
+ }
+ #endif
++extern int __sync_filesystem(struct super_block *, int);
+ extern int sync_filesystem(struct super_block *);
+ extern const struct file_operations def_blk_fops;
+ extern const struct file_operations def_chr_fops;
+diff -urNp -x '*.orig' linux-4.14/include/linux/lockdep.h linux-4.14/include/linux/lockdep.h
+--- linux-4.14/include/linux/lockdep.h 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/include/linux/lockdep.h 2021-02-24 21:42:43.434448095 +0100
+@@ -406,6 +406,8 @@ static inline int lockdep_match_key(stru
+ return lock->key == key;
+ }
+
++struct lock_class *lockdep_hlock_class(struct held_lock *hlock);
++
+ /*
+ * Acquire a lock.
+ *
+@@ -530,6 +532,7 @@ struct lock_class_key { };
+
+ #define lockdep_depth(tsk) (0)
+
++#define lockdep_is_held(lock) (1)
+ #define lockdep_is_held_type(l, r) (1)
+
+ #define lockdep_assert_held(l) do { (void)(l); } while (0)
+diff -urNp -x '*.orig' linux-4.14/include/linux/mm.h linux-4.14/include/linux/mm.h
+--- linux-4.14/include/linux/mm.h 2021-02-24 21:42:35.517798675 +0100
++++ linux-4.14/include/linux/mm.h 2021-02-24 21:42:43.434448095 +0100
+@@ -1370,6 +1370,28 @@ static inline int fixup_user_fault(struc
+ }
+ #endif
+
++extern void vma_do_file_update_time(struct vm_area_struct *, const char[], int);
++extern struct file *vma_do_pr_or_file(struct vm_area_struct *, const char[],
++ int);
++extern void vma_do_get_file(struct vm_area_struct *, const char[], int);
++extern void vma_do_fput(struct vm_area_struct *, const char[], int);
++
++#define vma_file_update_time(vma) vma_do_file_update_time(vma, __func__, \
++ __LINE__)
++#define vma_pr_or_file(vma) vma_do_pr_or_file(vma, __func__, \
++ __LINE__)
++#define vma_get_file(vma) vma_do_get_file(vma, __func__, __LINE__)
++#define vma_fput(vma) vma_do_fput(vma, __func__, __LINE__)
+
-+int au_xino_path(struct seq_file *seq, struct file *file)
-+{
-+ int err;
++#ifndef CONFIG_MMU
++extern struct file *vmr_do_pr_or_file(struct vm_region *, const char[], int);
++extern void vmr_do_fput(struct vm_region *, const char[], int);
+
-+ err = au_seq_path(seq, &file->f_path);
-+ if (unlikely(err))
-+ goto out;
++#define vmr_pr_or_file(region) vmr_do_pr_or_file(region, __func__, \
++ __LINE__)
++#define vmr_fput(region) vmr_do_fput(region, __func__, __LINE__)
++#endif /* !CONFIG_MMU */
+
-+#define Deleted "\\040(deleted)"
-+ seq->count -= sizeof(Deleted) - 1;
-+ AuDebugOn(memcmp(seq->buf + seq->count, Deleted,
-+ sizeof(Deleted) - 1));
-+#undef Deleted
+ extern int access_process_vm(struct task_struct *tsk, unsigned long addr, void *buf, int len,
+ unsigned int gup_flags);
+ extern int access_remote_vm(struct mm_struct *mm, unsigned long addr,
+diff -urNp -x '*.orig' linux-4.14/include/linux/mm_types.h linux-4.14/include/linux/mm_types.h
+--- linux-4.14/include/linux/mm_types.h 2021-02-24 21:42:35.517798675 +0100
++++ linux-4.14/include/linux/mm_types.h 2021-02-24 21:42:43.434448095 +0100
+@@ -258,6 +258,7 @@ struct vm_region {
+ unsigned long vm_top; /* region allocated to here */
+ unsigned long vm_pgoff; /* the offset in vm_file corresponding to vm_start */
+ struct file *vm_file; /* the backing file or NULL */
++ struct file *vm_prfile; /* the virtual backing file or NULL */
+
+ int vm_usage; /* region usage count (access under nommu_region_sem) */
+ bool vm_icache_flushed : 1; /* true if the icache has been flushed for
+@@ -332,6 +333,7 @@ struct vm_area_struct {
+ unsigned long vm_pgoff; /* Offset (within vm_file) in PAGE_SIZE
+ units */
+ struct file * vm_file; /* File we map to (can be NULL). */
++ struct file *vm_prfile; /* shadow of vm_file */
+ void * vm_private_data; /* was vm_pte (shared mem) */
+
+ atomic_long_t swap_readahead_info;
+diff -urNp -x '*.orig' linux-4.14/include/linux/mnt_namespace.h linux-4.14/include/linux/mnt_namespace.h
+--- linux-4.14/include/linux/mnt_namespace.h 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/include/linux/mnt_namespace.h 2021-02-24 21:42:43.434448095 +0100
+@@ -6,11 +6,14 @@
+ struct mnt_namespace;
+ struct fs_struct;
+ struct user_namespace;
++struct vfsmount;
+
+ extern struct mnt_namespace *copy_mnt_ns(unsigned long, struct mnt_namespace *,
+ struct user_namespace *, struct fs_struct *);
+ extern void put_mnt_ns(struct mnt_namespace *ns);
+
++extern int is_current_mnt_ns(struct vfsmount *mnt);
++
+ extern const struct file_operations proc_mounts_operations;
+ extern const struct file_operations proc_mountinfo_operations;
+ extern const struct file_operations proc_mountstats_operations;
+diff -urNp -x '*.orig' linux-4.14/include/linux/splice.h linux-4.14/include/linux/splice.h
+--- linux-4.14/include/linux/splice.h 2017-11-12 19:46:13.000000000 +0100
++++ linux-4.14/include/linux/splice.h 2021-02-24 21:42:43.434448095 +0100
+@@ -87,4 +87,10 @@ extern void splice_shrink_spd(struct spl
+
+ extern const struct pipe_buf_operations page_cache_pipe_buf_ops;
+ extern const struct pipe_buf_operations default_pipe_buf_ops;
+
-+out:
-+ return err;
-+}
-diff -urN /usr/share/empty/include/uapi/linux/aufs_type.h linux/include/uapi/linux/aufs_type.h
---- /usr/share/empty/include/uapi/linux/aufs_type.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux/include/uapi/linux/aufs_type.h 2016-08-17 18:01:21.295617591 +0200
-@@ -0,0 +1,419 @@
++extern long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
++ loff_t *ppos, size_t len, unsigned int flags);
++extern long do_splice_to(struct file *in, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags);
+ #endif
+diff -urNp -x '*.orig' linux-4.14/include/uapi/linux/aufs_type.h linux-4.14/include/uapi/linux/aufs_type.h
+--- linux-4.14/include/uapi/linux/aufs_type.h 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/include/uapi/linux/aufs_type.h 2021-02-24 21:42:43.447781402 +0100
+@@ -0,0 +1,447 @@
+/*
-+ * Copyright (C) 2005-2016 Junjiro R. Okajima
++ * Copyright (C) 2005-2017 Junjiro R. Okajima
+ *
+ * This program, aufs is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+
+#include <linux/limits.h>
+
-+#define AUFS_VERSION "4.7-20160815"
++#define AUFS_VERSION "4.x-rcN-20171106"
+
+/* todo? move this to linux-2.6.19/include/magic.h */
+#define AUFS_SUPER_MAGIC ('a' << 24 | 'u' << 16 | 'f' << 8 | 's')
+#define AUFS_PLINK_MAINT_DIR "fs/" AUFS_NAME
+#define AUFS_PLINK_MAINT_PATH AUFS_PLINK_MAINT_DIR "/" AUFS_PLINK_MAINT_NAME
+
++/* dirren, renamed dir */
++#define AUFS_DR_INFO_PFX AUFS_WH_PFX ".dr."
++#define AUFS_DR_BRHINO_NAME AUFS_WH_PFX "hino"
++/* whiteouted doubly */
++#define AUFS_WH_DR_INFO_PFX AUFS_WH_PFX AUFS_DR_INFO_PFX
++#define AUFS_WH_DR_BRHINO AUFS_WH_PFX AUFS_DR_BRHINO_NAME
++
+#define AUFS_DIROPQ_NAME AUFS_WH_PFX ".opq" /* whiteouted doubly */
+#define AUFS_WH_DIROPQ AUFS_WH_PFX AUFS_DIROPQ_NAME
+
+
+/* ---------------------------------------------------------------------- */
+
++/* dirren. the branch is identified by the filename who contains this */
++struct au_drinfo {
++ uint64_t ino;
++ union {
++ uint8_t oldnamelen;
++ uint64_t _padding;
++ };
++ uint8_t oldname[0];
++} __aligned(8);
++
++struct au_drinfo_fdata {
++ uint32_t magic;
++ struct au_drinfo drinfo;
++} __aligned(8);
++
++#define AUFS_DRINFO_MAGIC_V1 ('a' << 24 | 'd' << 16 | 'r' << 8 | 0x01)
++/* future */
++#define AUFS_DRINFO_MAGIC_V2 ('a' << 24 | 'd' << 16 | 'r' << 8 | 0x02)
++
++/* ---------------------------------------------------------------------- */
++
+struct aufs_wbr_fd {
+ uint32_t oflags;
+ int16_t brid;
+#define AUFS_CTL_FHSM_FD _IOW(AuCtlType, AuCtl_FHSM_FD, int)
+
+#endif /* __AUFS_TYPE_H__ */
-aufs4.7 loopback patch
-
-diff --git a/drivers/block/loop.c b/drivers/block/loop.c
-index 7339e65..76e5da4 100644
---- a/drivers/block/loop.c
-+++ b/drivers/block/loop.c
-@@ -562,7 +562,7 @@ static int do_req_filebacked(struct loop_device *lo, struct request *rq)
- }
-
- struct switch_request {
-- struct file *file;
-+ struct file *file, *virt_file;
- struct completion wait;
- };
-
-@@ -588,6 +588,7 @@ static void do_loop_switch(struct loop_device *lo, struct switch_request *p)
- mapping = file->f_mapping;
- mapping_set_gfp_mask(old_file->f_mapping, lo->old_gfp_mask);
- lo->lo_backing_file = file;
-+ lo->lo_backing_virt_file = p->virt_file;
- lo->lo_blocksize = S_ISBLK(mapping->host->i_mode) ?
- mapping->host->i_bdev->bd_block_size : PAGE_SIZE;
- lo->old_gfp_mask = mapping_gfp_mask(mapping);
-@@ -600,11 +601,13 @@ static void do_loop_switch(struct loop_device *lo, struct switch_request *p)
- * First it needs to flush existing IO, it does this by sending a magic
- * BIO down the pipe. The completion of this BIO does the actual switch.
- */
--static int loop_switch(struct loop_device *lo, struct file *file)
-+static int loop_switch(struct loop_device *lo, struct file *file,
-+ struct file *virt_file)
- {
- struct switch_request w;
+diff -urNp -x '*.orig' linux-4.14/kernel/fork.c linux-4.14/kernel/fork.c
+--- linux-4.14/kernel/fork.c 2021-02-24 21:42:35.541131952 +0100
++++ linux-4.14/kernel/fork.c 2021-02-24 21:42:43.434448095 +0100
+@@ -670,7 +670,7 @@ static __latent_entropy int dup_mmap(str
+ struct inode *inode = file_inode(file);
+ struct address_space *mapping = file->f_mapping;
- w.file = file;
-+ w.virt_file = virt_file;
+- get_file(file);
++ vma_get_file(tmp);
+ if (tmp->vm_flags & VM_DENYWRITE)
+ atomic_dec(&inode->i_writecount);
+ i_mmap_lock_write(mapping);
+diff -urNp -x '*.orig' linux-4.14/kernel/locking/lockdep.c linux-4.14/kernel/locking/lockdep.c
+--- linux-4.14/kernel/locking/lockdep.c 2021-02-24 21:42:35.544465277 +0100
++++ linux-4.14/kernel/locking/lockdep.c 2021-02-24 21:42:43.437781422 +0100
+@@ -143,7 +143,7 @@ static struct lock_list list_entries[MAX
+ unsigned long nr_lock_classes;
+ static struct lock_class lock_classes[MAX_LOCKDEP_KEYS];
- /* freeze queue and wait for completion of scheduled requests */
- blk_mq_freeze_queue(lo->lo_queue);
-@@ -623,7 +626,16 @@ static int loop_switch(struct loop_device *lo, struct file *file)
- */
- static int loop_flush(struct loop_device *lo)
+-static inline struct lock_class *hlock_class(struct held_lock *hlock)
++inline struct lock_class *lockdep_hlock_class(struct held_lock *hlock)
{
-- return loop_switch(lo, NULL);
-+ return loop_switch(lo, NULL, NULL);
-+}
-+
-+static struct file *loop_real_file(struct file *file)
-+{
-+ struct file *f = NULL;
-+
-+ if (file->f_path.dentry->d_sb->s_op->real_loop)
-+ f = file->f_path.dentry->d_sb->s_op->real_loop(file);
-+ return f;
+ if (!hlock->class_idx) {
+ /*
+@@ -154,6 +154,8 @@ static inline struct lock_class *hlock_c
+ }
+ return lock_classes + hlock->class_idx - 1;
}
++EXPORT_SYMBOL_GPL(lockdep_hlock_class);
++#define hlock_class(hlock) lockdep_hlock_class(hlock)
- static void loop_reread_partitions(struct loop_device *lo,
-@@ -660,6 +672,7 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
- unsigned int arg)
- {
- struct file *file, *old_file;
-+ struct file *f, *virt_file = NULL, *old_virt_file;
- struct inode *inode;
- int error;
+ #ifdef CONFIG_LOCK_STAT
+ static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], cpu_lock_stats);
+diff -urNp -x '*.orig' linux-4.14/kernel/task_work.c linux-4.14/kernel/task_work.c
+--- linux-4.14/kernel/task_work.c 2021-02-24 21:42:35.551131928 +0100
++++ linux-4.14/kernel/task_work.c 2021-02-24 21:42:43.437781422 +0100
+@@ -116,3 +116,4 @@ void task_work_run(void)
+ } while (work);
+ }
+ }
++EXPORT_SYMBOL_GPL(task_work_run);
+diff -urNp -x '*.orig' linux-4.14/mm/Makefile linux-4.14/mm/Makefile
+--- linux-4.14/mm/Makefile 2021-02-24 21:42:35.564465229 +0100
++++ linux-4.14/mm/Makefile 2021-02-24 21:42:43.434448095 +0100
+@@ -39,7 +39,7 @@ obj-y := filemap.o mempool.o oom_kill.
+ mm_init.o mmu_context.o percpu.o slab_common.o \
+ compaction.o vmacache.o swap_slots.o \
+ interval_tree.o list_lru.o workingset.o \
+- debug.o $(mmu-y)
++ prfile.o debug.o $(mmu-y)
-@@ -676,9 +689,16 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
- file = fget(arg);
- if (!file)
- goto out;
-+ f = loop_real_file(file);
-+ if (f) {
-+ virt_file = file;
-+ file = f;
-+ get_file(file);
-+ }
+ obj-y += init-mm.o
- inode = file->f_mapping->host;
- old_file = lo->lo_backing_file;
-+ old_virt_file = lo->lo_backing_virt_file;
+diff -urNp -x '*.orig' linux-4.14/mm/filemap.c linux-4.14/mm/filemap.c
+--- linux-4.14/mm/filemap.c 2021-02-24 21:42:35.564465229 +0100
++++ linux-4.14/mm/filemap.c 2021-02-24 21:42:43.434448095 +0100
+@@ -2612,7 +2612,7 @@ int filemap_page_mkwrite(struct vm_fault
+ int ret = VM_FAULT_LOCKED;
- error = -EINVAL;
+ sb_start_pagefault(inode->i_sb);
+- file_update_time(vmf->vma->vm_file);
++ vma_file_update_time(vmf->vma);
+ lock_page(page);
+ if (page->mapping != inode->i_mapping) {
+ unlock_page(page);
+diff -urNp -x '*.orig' linux-4.14/mm/mmap.c linux-4.14/mm/mmap.c
+--- linux-4.14/mm/mmap.c 2021-02-24 21:42:35.567798554 +0100
++++ linux-4.14/mm/mmap.c 2021-02-24 21:42:43.434448095 +0100
+@@ -166,7 +166,7 @@ static struct vm_area_struct *remove_vma
+ if (vma->vm_ops && vma->vm_ops->close)
+ vma->vm_ops->close(vma);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ mpol_put(vma_policy(vma));
+ kmem_cache_free(vm_area_cachep, vma);
+ return next;
+@@ -891,7 +891,7 @@ again:
+ if (remove_next) {
+ if (file) {
+ uprobe_munmap(next, next->vm_start, next->vm_end);
+- fput(file);
++ vma_fput(vma);
+ }
+ if (next->anon_vma)
+ anon_vma_merge(vma, next);
+@@ -1773,8 +1773,8 @@ out:
+ return addr;
-@@ -690,17 +710,21 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
- goto out_putf;
+ unmap_and_free_vma:
++ vma_fput(vma);
+ vma->vm_file = NULL;
+- fput(file);
- /* and ... switch */
-- error = loop_switch(lo, file);
-+ error = loop_switch(lo, file, virt_file);
- if (error)
- goto out_putf;
+ /* Undo any partial mapping done by a device driver. */
+ unmap_region(mm, vma, prev, vma->vm_start, vma->vm_end);
+@@ -2603,7 +2603,7 @@ int __split_vma(struct mm_struct *mm, st
+ goto out_free_mpol;
- fput(old_file);
-+ if (old_virt_file)
-+ fput(old_virt_file);
- if (lo->lo_flags & LO_FLAGS_PARTSCAN)
- loop_reread_partitions(lo, bdev);
- return 0;
+ if (new->vm_file)
+- get_file(new->vm_file);
++ vma_get_file(new);
- out_putf:
- fput(file);
-+ if (virt_file)
-+ fput(virt_file);
- out:
- return error;
- }
-@@ -887,7 +911,7 @@ static int loop_prepare_queue(struct loop_device *lo)
- static int loop_set_fd(struct loop_device *lo, fmode_t mode,
- struct block_device *bdev, unsigned int arg)
- {
-- struct file *file, *f;
-+ struct file *file, *f, *virt_file = NULL;
- struct inode *inode;
- struct address_space *mapping;
- unsigned lo_blocksize;
-@@ -902,6 +926,12 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
- file = fget(arg);
- if (!file)
- goto out;
-+ f = loop_real_file(file);
-+ if (f) {
-+ virt_file = file;
-+ file = f;
-+ get_file(file);
-+ }
+ if (new->vm_ops && new->vm_ops->open)
+ new->vm_ops->open(new);
+@@ -2622,7 +2622,7 @@ int __split_vma(struct mm_struct *mm, st
+ if (new->vm_ops && new->vm_ops->close)
+ new->vm_ops->close(new);
+ if (new->vm_file)
+- fput(new->vm_file);
++ vma_fput(new);
+ unlink_anon_vmas(new);
+ out_free_mpol:
+ mpol_put(vma_policy(new));
+@@ -2784,7 +2784,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsign
+ struct vm_area_struct *vma;
+ unsigned long populate = 0;
+ unsigned long ret = -EINVAL;
+- struct file *file;
++ struct file *file, *prfile;
- error = -EBUSY;
- if (lo->lo_state != Lo_unbound)
-@@ -954,6 +984,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
- lo->lo_device = bdev;
- lo->lo_flags = lo_flags;
- lo->lo_backing_file = file;
-+ lo->lo_backing_virt_file = virt_file;
- lo->transfer = NULL;
- lo->ioctl = NULL;
- lo->lo_sizelimit = 0;
-@@ -986,6 +1017,8 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
+ pr_warn_once("%s (%d) uses deprecated remap_file_pages() syscall. See Documentation/vm/remap_file_pages.txt.\n",
+ current->comm, current->pid);
+@@ -2859,10 +2859,27 @@ SYSCALL_DEFINE5(remap_file_pages, unsign
+ }
+ }
- out_putf:
+- file = get_file(vma->vm_file);
++ vma_get_file(vma);
++ file = vma->vm_file;
++ prfile = vma->vm_prfile;
+ ret = do_mmap_pgoff(vma->vm_file, start, size,
+ prot, flags, pgoff, &populate, NULL);
++ if (!IS_ERR_VALUE(ret) && file && prfile) {
++ struct vm_area_struct *new_vma;
++
++ new_vma = find_vma(mm, ret);
++ if (!new_vma->vm_prfile)
++ new_vma->vm_prfile = prfile;
++ if (new_vma != vma)
++ get_file(prfile);
++ }
++ /*
++ * two fput()s instead of vma_fput(vma),
++ * coz vma may not be available anymore.
++ */
fput(file);
-+ if (virt_file)
-+ fput(virt_file);
- out:
- /* This is safe: open() is still holding a reference. */
- module_put(THIS_MODULE);
-@@ -1032,6 +1065,7 @@ loop_init_xfer(struct loop_device *lo, struct loop_func_table *xfer,
- static int loop_clr_fd(struct loop_device *lo)
- {
- struct file *filp = lo->lo_backing_file;
-+ struct file *virt_filp = lo->lo_backing_virt_file;
- gfp_t gfp = lo->old_gfp_mask;
- struct block_device *bdev = lo->lo_device;
++ if (prfile)
++ fput(prfile);
+ out:
+ up_write(&mm->mmap_sem);
+ if (populate)
+@@ -3174,7 +3191,7 @@ struct vm_area_struct *copy_vma(struct v
+ if (anon_vma_clone(new_vma, vma))
+ goto out_free_mempol;
+ if (new_vma->vm_file)
+- get_file(new_vma->vm_file);
++ vma_get_file(new_vma);
+ if (new_vma->vm_ops && new_vma->vm_ops->open)
+ new_vma->vm_ops->open(new_vma);
+ vma_link(mm, new_vma, prev, rb_link, rb_parent);
+diff -urNp -x '*.orig' linux-4.14/mm/nommu.c linux-4.14/mm/nommu.c
+--- linux-4.14/mm/nommu.c 2021-02-24 21:42:35.567798554 +0100
++++ linux-4.14/mm/nommu.c 2021-02-24 21:42:43.434448095 +0100
+@@ -645,7 +645,7 @@ static void __put_nommu_region(struct vm
+ up_write(&nommu_region_sem);
-@@ -1063,6 +1097,7 @@ static int loop_clr_fd(struct loop_device *lo)
- spin_lock_irq(&lo->lo_lock);
- lo->lo_state = Lo_rundown;
- lo->lo_backing_file = NULL;
-+ lo->lo_backing_virt_file = NULL;
- spin_unlock_irq(&lo->lo_lock);
+ if (region->vm_file)
+- fput(region->vm_file);
++ vmr_fput(region);
- loop_release_xfer(lo);
-@@ -1107,6 +1142,8 @@ static int loop_clr_fd(struct loop_device *lo)
- * bd_mutex which is usually taken before lo_ctl_mutex.
- */
- fput(filp);
-+ if (virt_filp)
-+ fput(virt_filp);
- return 0;
+ /* IO memory and memory shared directly out of the pagecache
+ * from ramfs/tmpfs mustn't be released here */
+@@ -803,7 +803,7 @@ static void delete_vma(struct mm_struct
+ if (vma->vm_ops && vma->vm_ops->close)
+ vma->vm_ops->close(vma);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ put_nommu_region(vma->vm_region);
+ kmem_cache_free(vm_area_cachep, vma);
}
+@@ -1325,7 +1325,7 @@ unsigned long do_mmap(struct file *file,
+ goto error_just_free;
+ }
+ }
+- fput(region->vm_file);
++ vmr_fput(region);
+ kmem_cache_free(vm_region_jar, region);
+ region = pregion;
+ result = start;
+@@ -1400,10 +1400,10 @@ error_just_free:
+ up_write(&nommu_region_sem);
+ error:
+ if (region->vm_file)
+- fput(region->vm_file);
++ vmr_fput(region);
+ kmem_cache_free(vm_region_jar, region);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ kmem_cache_free(vm_area_cachep, vma);
+ return ret;
-diff --git a/drivers/block/loop.h b/drivers/block/loop.h
-index fb2237c..c3888c5 100644
---- a/drivers/block/loop.h
-+++ b/drivers/block/loop.h
-@@ -46,7 +46,7 @@ struct loop_device {
- int (*ioctl)(struct loop_device *, int cmd,
- unsigned long arg);
-
-- struct file * lo_backing_file;
-+ struct file * lo_backing_file, *lo_backing_virt_file;
- struct block_device *lo_device;
- unsigned lo_blocksize;
- void *key_data;
-diff --git a/fs/aufs/f_op.c b/fs/aufs/f_op.c
-index 00475fb..01390e1 100644
---- a/fs/aufs/f_op.c
-+++ b/fs/aufs/f_op.c
-@@ -348,7 +348,7 @@ static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
- if (IS_ERR(h_file))
- goto out;
-
-- if (au_test_loopback_kthread()) {
-+ if (0 && au_test_loopback_kthread()) {
- au_warn_loopback(h_file->f_path.dentry->d_sb);
- if (file->f_mapping != h_file->f_mapping) {
- file->f_mapping = h_file->f_mapping;
-diff --git a/fs/aufs/loop.c b/fs/aufs/loop.c
-index e92a345..35f4d48 100644
---- a/fs/aufs/loop.c
-+++ b/fs/aufs/loop.c
-@@ -131,3 +131,19 @@ void au_loopback_fin(void)
- symbol_put(loop_backing_file);
- au_delayed_kfree(au_warn_loopback_array);
- }
+diff -urNp -x '*.orig' linux-4.14/mm/prfile.c linux-4.14/mm/prfile.c
+--- linux-4.14/mm/prfile.c 1970-01-01 01:00:00.000000000 +0100
++++ linux-4.14/mm/prfile.c 2021-02-24 21:42:43.434448095 +0100
+@@ -0,0 +1,85 @@
++/*
++ * Mainly for aufs which mmap(2) different file and wants to print different
++ * path in /proc/PID/maps.
++ * Call these functions via macros defined in linux/mm.h.
++ *
++ * See Documentation/filesystems/aufs/design/06mmap.txt
++ *
++ * Copyright (c) 2014-2017 Junjro R. Okajima
++ * Copyright (c) 2014 Ian Campbell
++ */
+
-+/* ---------------------------------------------------------------------- */
++#include <linux/mm.h>
++#include <linux/file.h>
++#include <linux/fs.h>
+
-+/* support the loopback block device insude aufs */
++/* #define PRFILE_TRACE */
++static inline void prfile_trace(struct file *f, struct file *pr,
++ const char func[], int line, const char func2[])
++{
++#ifdef PRFILE_TRACE
++ if (pr)
++ pr_info("%s:%d: %s, %pD2\n", func, line, func2, f);
++#endif
++}
+
-+struct file *aufs_real_loop(struct file *file)
++void vma_do_file_update_time(struct vm_area_struct *vma, const char func[],
++ int line)
+{
-+ struct file *f;
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
+
-+ BUG_ON(!au_test_aufs(file->f_path.dentry->d_sb));
-+ fi_read_lock(file);
-+ f = au_hf_top(file);
-+ fi_read_unlock(file);
-+ AuDebugOn(!f);
-+ return f;
++ prfile_trace(f, pr, func, line, __func__);
++ file_update_time(f);
++ if (f && pr)
++ file_update_time(pr);
+}
-diff --git a/fs/aufs/loop.h b/fs/aufs/loop.h
-index 48bf070..66afec7 100644
---- a/fs/aufs/loop.h
-+++ b/fs/aufs/loop.h
-@@ -25,7 +25,11 @@ void au_warn_loopback(struct super_block *h_sb);
-
- int au_loopback_init(void);
- void au_loopback_fin(void);
+
-+struct file *aufs_real_loop(struct file *file);
- #else
-+AuStub(struct file *, loop_backing_file, return NULL)
++struct file *vma_do_pr_or_file(struct vm_area_struct *vma, const char func[],
++ int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
+
- AuStubInt0(au_test_loopback_overlap, struct super_block *sb,
- struct dentry *h_adding)
- AuStubInt0(au_test_loopback_kthread, void)
-@@ -33,6 +37,8 @@ AuStubVoid(au_warn_loopback, struct super_block *h_sb)
-
- AuStubInt0(au_loopback_init, void)
- AuStubVoid(au_loopback_fin, void)
++ prfile_trace(f, pr, func, line, __func__);
++ return (f && pr) ? pr : f;
++}
+
-+AuStub(struct file *, aufs_real_loop, return NULL, struct file *file)
- #endif /* BLK_DEV_LOOP */
++void vma_do_get_file(struct vm_area_struct *vma, const char func[], int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ get_file(f);
++ if (f && pr)
++ get_file(pr);
++}
++
++void vma_do_fput(struct vm_area_struct *vma, const char func[], int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ fput(f);
++ if (f && pr)
++ fput(pr);
++}
++
++#ifndef CONFIG_MMU
++struct file *vmr_do_pr_or_file(struct vm_region *region, const char func[],
++ int line)
++{
++ struct file *f = region->vm_file, *pr = region->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ return (f && pr) ? pr : f;
++}
++
++void vmr_do_fput(struct vm_region *region, const char func[], int line)
++{
++ struct file *f = region->vm_file, *pr = region->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ fput(f);
++ if (f && pr)
++ fput(pr);
++}
++#endif /* !CONFIG_MMU */
+diff -urNp -x '*.orig' linux-4.14/security/commoncap.c linux-4.14/security/commoncap.c
+--- linux-4.14/security/commoncap.c 2021-02-24 21:42:35.641131710 +0100
++++ linux-4.14/security/commoncap.c 2021-02-24 21:42:43.437781422 +0100
+@@ -1270,12 +1270,14 @@ int cap_mmap_addr(unsigned long addr)
+ }
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(cap_mmap_addr);
- #endif /* __KERNEL__ */
-diff --git a/fs/aufs/super.c b/fs/aufs/super.c
-index 58a773c..75f212c 100644
---- a/fs/aufs/super.c
-+++ b/fs/aufs/super.c
-@@ -831,7 +831,10 @@ static const struct super_operations aufs_sop = {
- .statfs = aufs_statfs,
- .put_super = aufs_put_super,
- .sync_fs = aufs_sync_fs,
-- .remount_fs = aufs_remount_fs
-+ .remount_fs = aufs_remount_fs,
-+#ifdef CONFIG_AUFS_BDEV_LOOP
-+ .real_loop = aufs_real_loop
-+#endif
- };
+ int cap_mmap_file(struct file *file, unsigned long reqprot,
+ unsigned long prot, unsigned long flags)
+ {
+ return 0;
+ }
++EXPORT_SYMBOL_GPL(cap_mmap_file);
- /* ---------------------------------------------------------------------- */
-diff --git a/include/linux/fs.h b/include/linux/fs.h
-index 33c6953..f9b06a2 100644
---- a/include/linux/fs.h
-+++ b/include/linux/fs.h
-@@ -1805,6 +1805,10 @@ struct super_operations {
- struct shrink_control *);
- long (*free_cached_objects)(struct super_block *,
- struct shrink_control *);
-+#if defined(CONFIG_BLK_DEV_LOOP) || defined(CONFIG_BLK_DEV_LOOP_MODULE)
-+ /* and aufs */
-+ struct file *(*real_loop)(struct file *);
-+#endif
- };
+ #ifdef CONFIG_SECURITY
- /*
+diff -urNp -x '*.orig' linux-4.14/security/device_cgroup.c linux-4.14/security/device_cgroup.c
+--- linux-4.14/security/device_cgroup.c 2021-02-24 21:42:35.641131710 +0100
++++ linux-4.14/security/device_cgroup.c 2021-02-24 21:42:43.437781422 +0100
+@@ -8,6 +8,7 @@
+ #include <linux/device_cgroup.h>
+ #include <linux/cgroup.h>
+ #include <linux/ctype.h>
++#include <linux/export.h>
+ #include <linux/list.h>
+ #include <linux/uaccess.h>
+ #include <linux/seq_file.h>
+@@ -850,6 +851,7 @@ int __devcgroup_inode_permission(struct
+ return __devcgroup_check_permission(type, imajor(inode), iminor(inode),
+ access);
+ }
++EXPORT_SYMBOL_GPL(__devcgroup_inode_permission);
+
+ int devcgroup_inode_mknod(int mode, dev_t dev)
+ {
+diff -urNp -x '*.orig' linux-4.14/security/security.c linux-4.14/security/security.c
+--- linux-4.14/security/security.c 2021-02-24 21:42:35.641131710 +0100
++++ linux-4.14/security/security.c 2021-02-24 21:42:43.437781422 +0100
+@@ -532,6 +532,7 @@ int security_path_rmdir(const struct pat
+ return 0;
+ return call_int_hook(path_rmdir, 0, dir, dentry);
+ }
++EXPORT_SYMBOL_GPL(security_path_rmdir);
+
+ int security_path_unlink(const struct path *dir, struct dentry *dentry)
+ {
+@@ -548,6 +549,7 @@ int security_path_symlink(const struct p
+ return 0;
+ return call_int_hook(path_symlink, 0, dir, dentry, old_name);
+ }
++EXPORT_SYMBOL_GPL(security_path_symlink);
+
+ int security_path_link(struct dentry *old_dentry, const struct path *new_dir,
+ struct dentry *new_dentry)
+@@ -556,6 +558,7 @@ int security_path_link(struct dentry *ol
+ return 0;
+ return call_int_hook(path_link, 0, old_dentry, new_dir, new_dentry);
+ }
++EXPORT_SYMBOL_GPL(security_path_link);
+
+ int security_path_rename(const struct path *old_dir, struct dentry *old_dentry,
+ const struct path *new_dir, struct dentry *new_dentry,
+@@ -583,6 +586,7 @@ int security_path_truncate(const struct
+ return 0;
+ return call_int_hook(path_truncate, 0, path);
+ }
++EXPORT_SYMBOL_GPL(security_path_truncate);
+
+ int security_path_chmod(const struct path *path, umode_t mode)
+ {
+@@ -590,6 +594,7 @@ int security_path_chmod(const struct pat
+ return 0;
+ return call_int_hook(path_chmod, 0, path, mode);
+ }
++EXPORT_SYMBOL_GPL(security_path_chmod);
+
+ int security_path_chown(const struct path *path, kuid_t uid, kgid_t gid)
+ {
+@@ -597,6 +602,7 @@ int security_path_chown(const struct pat
+ return 0;
+ return call_int_hook(path_chown, 0, path, uid, gid);
+ }
++EXPORT_SYMBOL_GPL(security_path_chown);
+
+ int security_path_chroot(const struct path *path)
+ {
+@@ -682,6 +688,7 @@ int security_inode_readlink(struct dentr
+ return 0;
+ return call_int_hook(inode_readlink, 0, dentry);
+ }
++EXPORT_SYMBOL_GPL(security_inode_readlink);
+
+ int security_inode_follow_link(struct dentry *dentry, struct inode *inode,
+ bool rcu)
+@@ -697,6 +704,7 @@ int security_inode_permission(struct ino
+ return 0;
+ return call_int_hook(inode_permission, 0, inode, mask);
+ }
++EXPORT_SYMBOL_GPL(security_inode_permission);
+
+ int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
+ {
+@@ -868,6 +876,7 @@ int security_file_permission(struct file
+
+ return fsnotify_perm(file, mask);
+ }
++EXPORT_SYMBOL_GPL(security_file_permission);
+
+ int security_file_alloc(struct file *file)
+ {
+@@ -927,6 +936,7 @@ int security_mmap_file(struct file *file
+ return ret;
+ return ima_file_mmap(file, prot, prot_adj, flags);
+ }
++EXPORT_SYMBOL_GPL(security_mmap_file);
+
+ int security_mmap_addr(unsigned long addr)
+ {