aufs4.x-rcN kbuild patch
diff --git a/fs/Kconfig b/fs/Kconfig
-index b8fcb41..78adefb 100644
+index 7aee6d6..ec92031 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
-@@ -236,6 +236,7 @@ source "fs/pstore/Kconfig"
+@@ -248,6 +248,7 @@ source "fs/pstore/Kconfig"
source "fs/sysv/Kconfig"
source "fs/ufs/Kconfig"
source "fs/exofs/Kconfig"
endif # MISC_FILESYSTEMS
diff --git a/fs/Makefile b/fs/Makefile
-index 85b6e13..e7bb164 100644
+index 7bbaca9..a026491 100644
--- a/fs/Makefile
+++ b/fs/Makefile
@@ -128,3 +128,4 @@ obj-y += exofs/ # Multiple modules
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 8bdae34..65dbd5f 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.x-rcN base patch
diff --git a/MAINTAINERS b/MAINTAINERS
-index 952fd2a..6a8f0f8 100644
+index af0cb69..d360d2e 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
-@@ -2210,6 +2210,19 @@ F: include/linux/audit.h
+@@ -2465,6 +2465,19 @@ F: include/linux/audit.h
F: include/uapi/linux/audit.h
F: kernel/audit*
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
+index 85de673..d44de9d 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
-@@ -712,6 +712,24 @@ static inline int is_loop_device(struct file *file)
+@@ -686,6 +686,24 @@ static inline int is_loop_device(struct file *file)
return i && S_ISBLK(i->i_mode) && MAJOR(i->i_rdev) == LOOP_MAJOR;
}
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
+index f901413..e3719a5 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
-@@ -1202,7 +1202,7 @@ enum d_walk_ret {
+@@ -1197,7 +1197,7 @@ enum d_walk_ret {
*
* The @enter() and @finish() callbacks are called with d_lock held.
*/
void (*finish)(void *))
{
diff --git a/fs/fcntl.c b/fs/fcntl.c
-index 350a2c8..6f42279 100644
+index 448a111..f51c2cf 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
-@@ -29,7 +29,7 @@
+@@ -31,7 +31,7 @@
#define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
{
struct inode * inode = file_inode(filp);
int error = 0;
-@@ -60,6 +60,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
+@@ -62,6 +62,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
if (filp->f_op->check_flags)
error = filp->f_op->check_flags(arg);
return error;
diff --git a/fs/inode.c b/fs/inode.c
-index 4ccbc21..aa6d071 100644
+index d1e35b5..f7800d6 100644
--- a/fs/inode.c
+++ b/fs/inode.c
-@@ -1591,7 +1591,7 @@ EXPORT_SYMBOL(generic_update_time);
+@@ -1655,7 +1655,7 @@ 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.
*/
{
int (*update_time)(struct inode *, struct timespec *, int);
+diff --git a/fs/namespace.c b/fs/namespace.c
+index d18deb4..e5a4a7f 100644
+--- a/fs/namespace.c
++++ b/fs/namespace.c
+@@ -846,6 +846,12 @@ static inline int check_mnt(struct mount *mnt)
+ 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));
++}
++
+ /*
+ * vfsmount lock must be held for write
+ */
diff --git a/fs/read_write.c b/fs/read_write.c
-index 933b53a..2d13282 100644
+index f0d4b16..6aa8c7a 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,
+@@ -483,6 +483,28 @@ ssize_t __vfs_write(struct file *file, const char __user *p, size_t count,
+ return -EINVAL;
}
- EXPORT_SYMBOL(__vfs_write);
+vfs_readf_t vfs_readf(struct file *file)
+{
+ return ERR_PTR(-ENOSYS);
+}
+
- ssize_t __kernel_write(struct file *file, const char *buf, size_t count, loff_t *pos)
+ ssize_t __kernel_write(struct file *file, const void *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
+index f3084cc..eb888c6 100644
--- a/fs/splice.c
+++ b/fs/splice.c
-@@ -1111,8 +1111,8 @@ EXPORT_SYMBOL(generic_splice_sendpage);
+@@ -837,8 +837,8 @@ EXPORT_SYMBOL(generic_splice_sendpage);
/*
* Attempt to initiate a splice from pipe to file.
*/
{
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,
+@@ -854,9 +854,9 @@ static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
/*
* Attempt to initiate a splice from a file to a pipe.
*/
{
ssize_t (*splice_read)(struct file *, loff_t *,
struct pipe_inode_info *, size_t, unsigned int);
+diff --git a/fs/sync.c b/fs/sync.c
+index a576aa2..eb61780 100644
+--- a/fs/sync.c
++++ b/fs/sync.c
+@@ -27,7 +27,7 @@
+ * 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);
diff --git a/include/linux/file.h b/include/linux/file.h
-index 7444f5f..bdac0be 100644
+index 61eb82c..e700888 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,
+ 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 --git a/include/linux/fs.h b/include/linux/fs.h
-index dd28814..b689a48 100644
+index 13dab19..8ab6566 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
-@@ -1306,6 +1306,7 @@ extern void fasync_free(struct fasync_struct *);
+@@ -1264,6 +1264,7 @@ 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 int f_setown(struct file *filp, unsigned long arg, int force);
extern void f_delown(struct file *filp);
-@@ -1690,6 +1691,7 @@ struct file_operations {
+@@ -1710,6 +1711,7 @@ 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 (*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,
+@@ -1780,6 +1782,12 @@ ssize_t rw_copy_check_uvector(int type, const struct iovec __user * uvector,
struct iovec *fast_pointer,
struct iovec **ret_pointer);
+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 ssize_t vfs_write(struct file *, const char __user *, size_t, loff_t *);
+@@ -2182,6 +2190,7 @@ 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);
/* /sys/fs */
extern struct kobject *fs_kobj;
+@@ -2462,6 +2471,7 @@ static inline bool sb_is_blkdev_sb(struct super_block *sb)
+ 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 --git a/include/linux/lockdep.h b/include/linux/lockdep.h
+index bfa8e0b..728d810 100644
+--- a/include/linux/lockdep.h
++++ b/include/linux/lockdep.h
+@@ -405,6 +405,8 @@ static inline int lockdep_match_key(struct lockdep_map *lock,
+ return lock->key == key;
+ }
+
++struct lock_class *lockdep_hlock_class(struct held_lock *hlock);
++
+ /*
+ * Acquire a lock.
+ *
+@@ -529,6 +531,7 @@ 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 --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
+index 12b2ab5..8b810d1 100644
+--- a/include/linux/mnt_namespace.h
++++ b/include/linux/mnt_namespace.h
+@@ -5,11 +5,14 @@
+ 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 --git a/include/linux/splice.h b/include/linux/splice.h
-index da2751d..2e0fca6 100644
+index db42746..12f3a5a 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);
+@@ -86,4 +86,10 @@ extern void splice_shrink_spd(struct splice_pipe_desc *);
extern const struct pipe_buf_operations page_cache_pipe_buf_ops;
+ extern const struct pipe_buf_operations default_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);
+ struct pipe_inode_info *pipe, size_t len,
+ unsigned int flags);
#endif
+diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
+index e36e652..bc97a97 100644
+--- a/kernel/locking/lockdep.c
++++ b/kernel/locking/lockdep.c
+@@ -144,7 +144,7 @@ static struct lock_list list_entries[MAX_LOCKDEP_ENTRIES];
+ unsigned long nr_lock_classes;
+ static struct lock_class lock_classes[MAX_LOCKDEP_KEYS];
+
+-static inline struct lock_class *hlock_class(struct held_lock *hlock)
++inline struct lock_class *lockdep_hlock_class(struct held_lock *hlock)
+ {
+ if (!hlock->class_idx) {
+ /*
+@@ -155,6 +155,7 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock)
+ }
+ return lock_classes + hlock->class_idx - 1;
+ }
++#define hlock_class(hlock) lockdep_hlock_class(hlock)
+
+ #ifdef CONFIG_LOCK_STAT
+ static DEFINE_PER_CPU(struct lock_class_stats[MAX_LOCKDEP_KEYS], cpu_lock_stats);
aufs4.x-rcN mmap patch
diff --git a/fs/proc/base.c b/fs/proc/base.c
-index a11eb71..8f10865 100644
+index ad3b076..ad4a50d 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)
+@@ -1987,7 +1987,7 @@ 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) {
rc = 0;
}
diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
-index f8595e8..cb8eda0 100644
+index 7563437..7c0dc0f 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)
ino = inode->i_ino;
}
diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
-index 4648c7f..061cb85 100644
+index 5589b4b..f60aea2 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)
+@@ -309,7 +309,10 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma, int is_pid)
const char *name = NULL;
if (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)
+@@ -1734,7 +1737,7 @@ 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 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
+index b00b7660..93e8a86 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,
+@@ -155,7 +155,10 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma,
file = vma->vm_file;
if (file) {
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
+index 065d99d..04486c3 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,
+@@ -1348,6 +1348,28 @@ static inline int fixup_user_fault(struct task_struct *tsk,
}
#endif
+#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_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,
- void *buf, int len, int write);
diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
-index ca3e517..10bc491 100644
+index 1861ea8..d85a914 100644
--- a/include/linux/mm_types.h
+++ b/include/linux/mm_types.h
-@@ -274,6 +274,7 @@ struct vm_region {
+@@ -260,6 +260,7 @@ 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 */
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 {
+@@ -334,6 +335,7 @@ 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
+ atomic_long_t swap_readahead_info;
diff --git a/kernel/fork.c b/kernel/fork.c
-index 4a7ec0c..8c8f7ac 100644
+index 07cc743..b1d2b43 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)
+@@ -676,7 +676,7 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
struct inode *inode = file_inode(file);
struct address_space *mapping = file->f_mapping;
atomic_dec(&inode->i_writecount);
i_mmap_lock_write(mapping);
diff --git a/mm/Makefile b/mm/Makefile
-index 78c6f7d..aea4230 100644
+index e3ac3ae..745b26c 100644
--- a/mm/Makefile
+++ b/mm/Makefile
-@@ -37,7 +37,7 @@ obj-y := filemap.o mempool.o oom_kill.o \
+@@ -39,7 +39,7 @@ obj-y := filemap.o mempool.o oom_kill.o \
mm_init.o mmu_context.o percpu.o slab_common.o \
- compaction.o vmacache.o \
+ 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)
obj-y += init-mm.o
diff --git a/mm/filemap.c b/mm/filemap.c
-index 20f3b1f..ee827ce 100644
+index 594d73f..7183aef 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)
+@@ -2590,7 +2590,7 @@ int filemap_page_mkwrite(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);
+- 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 --git a/mm/memory.c b/mm/memory.c
-index cd1f29e..f0c204c 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
+index 680506f..081406a 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)
+@@ -171,7 +171,7 @@ 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)
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);
+@@ -896,7 +896,7 @@ int __vma_adjust(struct vm_area_struct *vma, unsigned long start,
if (remove_next) {
if (file) {
uprobe_munmap(next, next->vm_start, next->vm_end);
}
if (next->anon_vma)
anon_vma_merge(vma, next);
-@@ -1563,8 +1563,8 @@ out:
+@@ -1746,8 +1746,8 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
return addr;
unmap_and_free_vma:
/* 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,
+@@ -2569,7 +2569,7 @@ int __split_vma(struct mm_struct *mm, struct vm_area_struct *vma,
goto out_free_mpol;
if (new->vm_file)
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,
+@@ -2588,7 +2588,7 @@ 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)
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,
+@@ -2750,7 +2750,7 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
struct vm_area_struct *vma;
unsigned long populate = 0;
unsigned long ret = -EINVAL;
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,
+@@ -2825,10 +2825,27 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
}
}
+ file = vma->vm_file;
+ prfile = vma->vm_prfile;
ret = do_mmap_pgoff(vma->vm_file, start, size,
- prot, flags, pgoff, &populate);
+ prot, flags, pgoff, &populate, NULL);
+ if (!IS_ERR_VALUE(ret) && file && prfile) {
+ struct vm_area_struct *new_vma;
+
out:
up_write(&mm->mmap_sem);
if (populate)
-@@ -2873,7 +2890,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
+@@ -3136,7 +3153,7 @@ 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)
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
+index 17c00d9..4bcdf94 100644
--- a/mm/nommu.c
+++ b/mm/nommu.c
-@@ -644,7 +644,7 @@ static void __put_nommu_region(struct vm_region *region)
+@@ -641,7 +641,7 @@ static void __put_nommu_region(struct vm_region *region)
up_write(&nommu_region_sem);
if (region->vm_file)
/* 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)
+@@ -799,7 +799,7 @@ 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)
put_nommu_region(vma->vm_region);
kmem_cache_free(vm_area_cachep, vma);
}
-@@ -1328,7 +1328,7 @@ unsigned long do_mmap(struct file *file,
+@@ -1321,7 +1321,7 @@ unsigned long do_mmap(struct file *file,
goto error_just_free;
}
}
kmem_cache_free(vm_region_jar, region);
region = pregion;
result = start;
-@@ -1403,10 +1403,10 @@ error_just_free:
+@@ -1396,10 +1396,10 @@ unsigned long do_mmap(struct file *file,
up_write(&nommu_region_sem);
error:
if (region->vm_file)
diff --git a/mm/prfile.c b/mm/prfile.c
new file mode 100644
-index 0000000..b323b8a
+index 0000000..1ef053b
--- /dev/null
+++ b/mm/prfile.c
-@@ -0,0 +1,86 @@
+@@ -0,0 +1,85 @@
+/*
-+ * Mainly for aufs which mmap(2) diffrent file and wants to print different path
-+ * in /proc/PID/maps.
++ * 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 Junjro R. Okajima
++ * Copyright (c) 2014-2017 Junjro R. Okajima
+ * Copyright (c) 2014 Ian Campbell
+ */
+
+{
+#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)");
++ pr_info("%s:%d: %s, %pD2\n", func, line, func2, f);
+#endif
+}
+
aufs4.x-rcN standalone patch
diff --git a/fs/dcache.c b/fs/dcache.c
-index c3c0b6d..c99d2d2 100644
+index e3719a5..3203470 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
-@@ -1310,6 +1310,7 @@ rename_retry:
+@@ -1305,6 +1305,7 @@ void d_walk(struct dentry *parent, void *data,
seq = 1;
goto again;
}
+EXPORT_SYMBOL_GPL(d_walk);
- /*
- * Search for at least 1 mount point in the dentry's subdirs.
+ struct check_mount {
+ struct vfsmount *mnt;
+@@ -2894,6 +2895,7 @@ void d_exchange(struct dentry *dentry1, struct dentry *dentry2)
+
+ write_sequnlock(&rename_lock);
+ }
++EXPORT_SYMBOL_GPL(d_exchange);
+
+ /**
+ * d_ancestor - search for an ancestor
diff --git a/fs/exec.c b/fs/exec.c
-index 887c1c9..40e8767 100644
+index 3e14ba2..6818b01 100644
--- a/fs/exec.c
+++ b/fs/exec.c
-@@ -104,6 +104,7 @@ bool path_noexec(const struct path *path)
+@@ -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);
}
#ifdef CONFIG_USELIB
/*
diff --git a/fs/fcntl.c b/fs/fcntl.c
-index 6f42279..04fd33c 100644
+index f51c2cf..58bf222 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
-@@ -82,6 +82,7 @@ int setfl(int fd, struct file * filp, unsigned long arg)
+@@ -84,6 +84,7 @@ int setfl(int fd, struct file * filp, unsigned long arg)
out:
return error;
}
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
+index 61517f5..c6bab39c 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
-@@ -147,6 +147,7 @@ over:
+@@ -148,6 +148,7 @@ struct file *get_empty_filp(void)
}
return ERR_PTR(-ENFILE);
}
+EXPORT_SYMBOL_GPL(put_filp);
void __init files_init(void)
- {
+ {
diff --git a/fs/inode.c b/fs/inode.c
-index aa6d071..f336032 100644
+index f7800d6..f31a6c7 100644
--- a/fs/inode.c
+++ b/fs/inode.c
-@@ -1600,6 +1600,7 @@ int update_time(struct inode *inode, struct timespec *time, int flags)
+@@ -1664,6 +1664,7 @@ int update_time(struct inode *inode, struct timespec *time, int flags)
return update_time(inode, time, flags);
}
/**
* touch_atime - update the access time
diff --git a/fs/namespace.c b/fs/namespace.c
-index 783004a..44abb2d 100644
+index e5a4a7f..6d0c376 100644
--- a/fs/namespace.c
+++ b/fs/namespace.c
-@@ -463,6 +463,7 @@ void __mnt_drop_write(struct vfsmount *mnt)
+@@ -517,6 +517,7 @@ void __mnt_drop_write(struct vfsmount *mnt)
mnt_dec_writers(real_mount(mnt));
preempt_enable();
}
/**
* mnt_drop_write - give up write access to a mount
-@@ -1811,6 +1812,7 @@ int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg,
+@@ -851,6 +852,7 @@ 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
+@@ -1887,6 +1889,7 @@ int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg,
}
return 0;
}
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
+index 3235753..14a2d48 100644
--- a/fs/notify/group.c
+++ b/fs/notify/group.c
@@ -22,6 +22,7 @@
#include <linux/fsnotify_backend.h>
#include "fsnotify.h"
-@@ -81,6 +82,7 @@ void fsnotify_get_group(struct fsnotify_group *group)
+@@ -109,6 +110,7 @@ void fsnotify_get_group(struct fsnotify_group *group)
{
atomic_inc(&group->refcnt);
}
/*
* Drop a reference to a group. Free it if it's through.
-@@ -90,6 +92,7 @@ void fsnotify_put_group(struct fsnotify_group *group)
+@@ -118,6 +120,7 @@ void fsnotify_put_group(struct fsnotify_group *group)
if (atomic_dec_and_test(&group->refcnt))
fsnotify_final_destroy_group(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)
+@@ -147,6 +150,7 @@ struct fsnotify_group *fsnotify_alloc_group(const struct fsnotify_ops *ops)
return 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
+index 9991f88..117042c 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);
- }
+@@ -118,6 +118,7 @@ static bool fsnotify_get_mark_safe(struct fsnotify_mark *mark)
+ {
+ return atomic_inc_not_zero(&mark->refcnt);
}
+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,
+ static void __fsnotify_recalc_mask(struct fsnotify_mark_connector *conn)
+ {
+@@ -395,6 +396,7 @@ 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:
-
+ /*
+ * Sorting function for lists of fsnotify marks.
+@@ -607,6 +609,7 @@ int fsnotify_add_mark_locked(struct fsnotify_mark *mark, struct inode *inode,
+ fsnotify_put_mark(mark);
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;
+ int fsnotify_add_mark(struct fsnotify_mark *mark, struct inode *inode,
+ struct vfsmount *mnt, int allow_dups)
+@@ -742,6 +745,7 @@ void fsnotify_init_mark(struct fsnotify_mark *mark,
+ 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 --git a/fs/open.c b/fs/open.c
-index 93ae3cd..d25b9bd 100644
+index 7ea1184..6e2e241 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,
long vfs_truncate(const struct path *path, loff_t length)
{
-@@ -678,6 +679,7 @@ int open_check_o_direct(struct file *f)
+@@ -691,6 +692,7 @@ int open_check_o_direct(struct file *f)
}
return 0;
}
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
+index 6aa8c7a..b5d392e 100644
--- a/fs/read_write.c
+++ b/fs/read_write.c
-@@ -525,6 +525,7 @@ vfs_readf_t vfs_readf(struct file *file)
+@@ -453,6 +453,7 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
+
+ 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)
+ {
+@@ -493,6 +494,7 @@ vfs_readf_t vfs_readf(struct file *file)
return new_sync_read;
return ERR_PTR(-ENOSYS);
}
vfs_writef_t vfs_writef(struct file *file)
{
-@@ -536,6 +537,7 @@ vfs_writef_t vfs_writef(struct file *file)
+@@ -504,6 +506,7 @@ 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)
+ ssize_t __kernel_write(struct file *file, const void *buf, size_t count, loff_t *pos)
+ {
+@@ -573,6 +576,7 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_
+
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(vfs_write);
+
+ static inline loff_t file_pos_read(struct file *file)
{
diff --git a/fs/splice.c b/fs/splice.c
-index 9326c2a..0606690 100644
+index eb888c6..7ab89d2 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,
+@@ -850,6 +850,7 @@ long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
return splice_write(pipe, out, ppos, len, flags);
}
/*
* 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,
+@@ -879,6 +880,7 @@ long do_splice_to(struct file *in, loff_t *ppos,
return splice_read(in, ppos, pipe, len, flags);
}
/**
* splice_direct_to_actor - splices data directly between two non-pipes
+diff --git a/fs/sync.c b/fs/sync.c
+index eb61780..32c5a05 100644
+--- a/fs/sync.c
++++ b/fs/sync.c
+@@ -38,6 +38,7 @@ int __sync_filesystem(struct super_block *sb, int wait)
+ 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 --git a/fs/xattr.c b/fs/xattr.c
-index 4beafc4..e118715 100644
+index 61cd28b..35570cd 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,
+@@ -297,6 +297,7 @@ 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)
+ __vfs_getxattr(struct dentry *dentry, struct inode *inode, const char *name,
+diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
+index bc97a97..895a1ba 100644
+--- a/kernel/locking/lockdep.c
++++ b/kernel/locking/lockdep.c
+@@ -155,6 +155,7 @@ inline struct lock_class *lockdep_hlock_class(struct held_lock *hlock)
+ }
+ return lock_classes + hlock->class_idx - 1;
+ }
++EXPORT_SYMBOL_GPL(lockdep_hlock_class);
+ #define hlock_class(hlock) lockdep_hlock_class(hlock)
+
+ #ifdef CONFIG_LOCK_STAT
diff --git a/kernel/task_work.c b/kernel/task_work.c
-index 53fa971..bce3211 100644
+index 836a72a..aa00d49 100644
--- a/kernel/task_work.c
+++ b/kernel/task_work.c
-@@ -118,3 +118,4 @@ void task_work_run(void)
+@@ -115,3 +115,4 @@ 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
+index fc46f5b..90543ef 100644
--- a/security/commoncap.c
+++ b/security/commoncap.c
-@@ -1058,12 +1058,14 @@ int cap_mmap_addr(unsigned long addr)
+@@ -1270,12 +1270,14 @@ int cap_mmap_addr(unsigned long addr)
}
return ret;
}
int devcgroup_inode_mknod(int mode, dev_t dev)
{
diff --git a/security/security.c b/security/security.c
-index 7095693..da7fe2c 100644
+index 4bf0f57..b30d1e1 100644
--- a/security/security.c
+++ b/security/security.c
-@@ -434,6 +434,7 @@ int security_path_rmdir(const struct path *dir, struct dentry *dentry)
+@@ -530,6 +530,7 @@ int security_path_rmdir(const struct path *dir, struct dentry *dentry)
return 0;
return call_int_hook(path_rmdir, 0, dir, dentry);
}
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,
+@@ -546,6 +547,7 @@ int security_path_symlink(const struct path *dir, struct dentry *dentry,
return 0;
return call_int_hook(path_symlink, 0, dir, dentry, old_name);
}
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,
+@@ -554,6 +556,7 @@ 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);
}
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)
+@@ -581,6 +584,7 @@ int security_path_truncate(const struct path *path)
return 0;
return call_int_hook(path_truncate, 0, path);
}
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)
+@@ -588,6 +592,7 @@ int security_path_chmod(const struct path *path, umode_t mode)
return 0;
return call_int_hook(path_chmod, 0, path, mode);
}
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)
+@@ -595,6 +600,7 @@ 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);
}
int security_path_chroot(const struct path *path)
{
-@@ -584,6 +590,7 @@ int security_inode_readlink(struct dentry *dentry)
+@@ -680,6 +686,7 @@ int security_inode_readlink(struct dentry *dentry)
return 0;
return call_int_hook(inode_readlink, 0, dentry);
}
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)
+@@ -695,6 +702,7 @@ int security_inode_permission(struct inode *inode, int mask)
return 0;
return call_int_hook(inode_permission, 0, inode, mask);
}
int security_inode_setattr(struct dentry *dentry, struct iattr *attr)
{
-@@ -737,6 +745,7 @@ int security_file_permission(struct file *file, int mask)
+@@ -866,6 +874,7 @@ int security_file_permission(struct file *file, int mask)
return fsnotify_perm(file, mask);
}
int security_file_alloc(struct file *file)
{
-@@ -796,6 +805,7 @@ int security_mmap_file(struct file *file, unsigned long prot,
+@@ -925,6 +934,7 @@ int security_mmap_file(struct file *file, unsigned long prot,
return ret;
return ima_file_mmap(file, prot);
}
{
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
++++ linux/Documentation/ABI/testing/debugfs-aufs 2017-07-29 12:14:25.893041746 +0200
@@ -0,0 +1,50 @@
+What: /debug/aufs/si_<id>/
+Date: March 2009
+ 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
++++ linux/Documentation/ABI/testing/sysfs-aufs 2017-07-29 12:14:25.893041746 +0200
@@ -0,0 +1,31 @@
+What: /sys/fs/aufs/si_<id>/
+Date: March 2009
+ 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 @@
++++ linux/Documentation/filesystems/aufs/design/01intro.txt 2017-11-12 22:24:42.257509799 +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
+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
++++ linux/Documentation/filesystems/aufs/design/02struct.txt 2017-07-29 12:14:25.893041746 +0200
@@ -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
+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
++++ linux/Documentation/filesystems/aufs/design/03atomic_open.txt 2017-07-29 12:14:25.893041746 +0200
@@ -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
+ 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
++++ linux/Documentation/filesystems/aufs/design/03lookup.txt 2017-07-29 12:14:25.893041746 +0200
@@ -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
+ 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
++++ linux/Documentation/filesystems/aufs/design/04branch.txt 2017-07-29 12:14:25.893041746 +0200
@@ -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
+ 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
++++ linux/Documentation/filesystems/aufs/design/05wbr_policy.txt 2017-07-29 12:14:25.893041746 +0200
@@ -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/06dirren.dot linux/Documentation/filesystems/aufs/design/06dirren.dot
+--- /usr/share/empty/Documentation/filesystems/aufs/design/06dirren.dot 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/06dirren.dot 2017-11-12 22:24:44.694244127 +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 -urN /usr/share/empty/Documentation/filesystems/aufs/design/06dirren.txt linux/Documentation/filesystems/aufs/design/06dirren.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/06dirren.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/06dirren.txt 2017-11-12 22:24:44.694244127 +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 -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
++++ linux/Documentation/filesystems/aufs/design/06fhsm.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+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
++++ linux/Documentation/filesystems/aufs/design/06mmap.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+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
++++ linux/Documentation/filesystems/aufs/design/06xattr.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+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
++++ linux/Documentation/filesystems/aufs/design/07export.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+ 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
++++ linux/Documentation/filesystems/aufs/design/08shwh.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+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
++++ linux/Documentation/filesystems/aufs/design/10dynop.txt 2017-07-29 12:14:25.896375188 +0200
@@ -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
+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 @@
++++ linux/Documentation/filesystems/aufs/README 2017-07-29 12:14:25.893041746 +0200
+@@ -0,0 +1,393 @@
+
+Aufs4 -- advanced multi layered unification filesystem version 4.x
+http://aufs.sf.net
+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
+# 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 @@
++++ linux/fs/aufs/aufs.h 2017-11-12 22:24:44.697577553 +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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,1406 @@
++++ linux/fs/aufs/branch.c 2017-11-12 22:24:44.697577553 +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);
+ 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 */
+ 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:
+ kfree(add_branch);
+out:
+ 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);
+}
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 @@
++++ linux/fs/aufs/branch.h 2017-11-12 22:24:44.697577553 +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 /* __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 @@
++++ linux/fs/aufs/conf.mk 2017-11-12 22:24:44.704244405 +0100
+@@ -0,0 +1,39 @@
+
+AuConfStr = CONFIG_AUFS_FS=${CONFIG_AUFS_FS}
+
+ XATTR \
+ FHSM \
+ RDU \
++ DIRREN \
+ SHWH \
+ BR_RAMFS \
+ BR_FUSE POLL \
+-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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,1383 @@
++++ linux/fs/aufs/cpup.c 2017-11-12 22:24:44.704244405 +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
+ 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;
++}
++
+/*
+ * to support a sparse file which is opened with O_APPEND,
+ * we need to close the file.
+ .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);
+
+ 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)
+{
+}
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 @@
++++ linux/fs/aufs/cpup.h 2017-11-12 22:24:44.704244405 +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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,432 @@
++++ linux/fs/aufs/dbgaufs.c 2017-11-12 22:24:44.704244405 +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
+ 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));
+ 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);
+ 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;
+ }
+}
+ br = au_sbr(sb, bindex);
+ xi = &br->br_xino;
+ AuDebugOn(xi->xi_dbgaufs);
++ /* debugfs acquires the parent i_mutex */
++ lockdep_off();
+ xi->xi_dbgaufs = debugfs_create_file(name, dbgaufs_mode, parent,
+ sbinfo, &dbgaufs_xino_fop);
++ lockdep_on();
+ /* ignore an error */
+ if (unlikely(!xi->xi_dbgaufs))
+ AuWarn1("failed %s under debugfs\n", name);
+}
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
++++ linux/fs/aufs/dbgaufs.h 2017-07-29 12:14:25.899708630 +0200
@@ -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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,224 @@
++++ linux/fs/aufs/dcsub.c 2017-11-12 22:24:42.267510077 +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
+ 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;
+
+}
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
++++ linux/fs/aufs/dcsub.h 2017-07-29 12:14:25.899708630 +0200
@@ -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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,441 @@
++++ linux/fs/aufs/debug.c 2017-11-12 22:24:42.267510077 +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
+ sbinfo = au_sbi(sb);
+ if (!sbinfo)
+ return;
-+ dpri("nw %lld, gen %u, kobj %d\n",
-+ percpu_counter_sum(&sbinfo->si_nowait.nw_len),
-+ sbinfo->si_generation,
-+ atomic_read(&sbinfo->si_kobj.kref.refcount));
++ dpri("nw %d, gen %u, kobj %d\n",
++ atomic_read(&sbinfo->si_nowait.nw_len), sbinfo->si_generation,
++ kref_read(&sbinfo->si_kobj.kref));
+ for (bindex = 0; bindex <= sbinfo->si_bbot; bindex++)
+ do_pri_br(bindex, sbinfo->si_branch[0 + bindex]);
+}
+}
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
++++ linux/fs/aufs/debug.h 2017-07-29 12:14:25.899708630 +0200
@@ -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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,1128 @@
++++ linux/fs/aufs/dentry.c 2017-11-12 22:24:44.704244405 +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);
-+ 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",
+};
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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,252 @@
++++ linux/fs/aufs/dentry.h 2017-11-12 22:24:44.704244405 +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 {
+/* 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 /* __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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,552 @@
++++ linux/fs/aufs/dinfo.c 2017-11-12 22:24:42.267510077 +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
+ 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);
+ }
+}
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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,756 @@
++++ linux/fs/aufs/dir.c 2017-11-12 22:24:44.704244405 +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
+ struct au_vdir *vdir_cache;
+ struct au_finfo *finfo;
+ struct au_fidir *fidir;
++ struct au_hfile *hf;
+ aufs_bindex_t bindex, bbot;
+
+ 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);
+
+ bindex = finfo->fi_btop;
+ if (bindex >= 0) {
++ hf = fidir->fd_hfile + bindex;
+ /*
+ * calls fput() instead of filp_close(),
+ * since no dnotify or lock for the lower file.
+ */
+ bbot = fidir->fd_bbot;
-+ for (; bindex <= bbot; bindex++)
-+ au_set_h_fptr(file, bindex, NULL);
++ for (; bindex <= bbot; bindex++, hf++)
++ if (hf->hf_file)
++ au_hfput(hf, /*execed*/0);
+ }
+ kfree(fidir);
+ finfo->fi_hdir = NULL;
+ 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 {
+};
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 2016-07-25 19:05:34.811159821 +0200
++++ linux/fs/aufs/dir.h 2017-11-12 22:24:42.267510077 +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
+
+ unsigned long vd_version;
+ unsigned int vd_deblk_sz;
-+ unsigned long vd_jiffy;
++ unsigned long vd_jiffy;
+} ____cacheline_aligned_in_smp;
+
+/* ---------------------------------------------------------------------- */
+
+#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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,369 @@
+diff -urN /usr/share/empty/fs/aufs/dirren.c linux/fs/aufs/dirren.c
+--- /usr/share/empty/fs/aufs/dirren.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/dirren.c 2017-11-12 22:24:44.704244405 +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"
+
-+#define DyPrSym(key) AuDbgSym(key->dk_op.dy_hop)
++static void au_dr_hino_del(struct au_dr_br *dr, struct au_dr_hino *ent)
++{
++ int idx;
+
-+/*
-+ * How large will these lists be?
-+ * Usually just a few elements, 20-30 at most for each, I guess.
-+ */
-+static struct au_splhead dynop[AuDyLast];
++ idx = au_dr_ihash(ent->dr_h_ino);
++ au_hbl_del(&ent->dr_hnode, dr->dr_h_ino + idx);
++}
+
-+static struct au_dykey *dy_gfind_get(struct au_splhead *spl, const void *h_op)
++static int au_dr_hino_test_empty(struct au_dr_br *dr)
+{
-+ struct au_dykey *key, *tmp;
-+ struct list_head *head;
++ int ret, i;
++ struct hlist_bl_head *hbl;
+
-+ key = NULL;
-+ head = &spl->head;
-+ rcu_read_lock();
-+ list_for_each_entry_rcu(tmp, head, dk_list)
-+ if (tmp->dk_op.dy_hop == h_op) {
-+ key = tmp;
-+ kref_get(&key->dk_kref);
-+ break;
-+ }
-+ rcu_read_unlock();
++ 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 key;
++ return ret;
+}
+
-+static struct au_dykey *dy_bradd(struct au_branch *br, struct au_dykey *key)
++static struct au_dr_hino *au_dr_hino_find(struct au_dr_br *dr, ino_t ino)
+{
-+ struct au_dykey **k, *found;
-+ const void *h_op = key->dk_op.dy_hop;
-+ int i;
++ struct au_dr_hino *found, *ent;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ int idx;
+
+ 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
++ 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;
-+ 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 */
-+ }
++ }
++ hlist_bl_unlock(hbl);
+
+ return found;
+}
+
-+/* kref_get() if @key is already added */
-+static struct au_dykey *dy_gadd(struct au_splhead *spl, struct au_dykey *key)
++int au_dr_hino_test_add(struct au_dr_br *dr, ino_t ino,
++ struct au_dr_hino *add_ent)
+{
-+ struct au_dykey *tmp, *found;
-+ struct list_head *head;
-+ const void *h_op = key->dk_op.dy_hop;
++ int found, idx;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dr_hino *ent;
+
-+ found = NULL;
-+ head = &spl->head;
-+ spin_lock(&spl->spin);
-+ list_for_each_entry(tmp, head, dk_list)
-+ if (tmp->dk_op.dy_hop == h_op) {
-+ kref_get(&tmp->dk_kref);
-+ found = tmp;
++ 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)
-+ list_add_rcu(&key->dk_list, head);
-+ spin_unlock(&spl->spin);
++ 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);
+
-+ if (!found)
-+ DyPrSym(key);
+ return found;
+}
+
-+static void dy_free_rcu(struct rcu_head *rcu)
++void au_dr_hino_free(struct au_dr_br *dr)
+{
-+ struct au_dykey *key;
-+
-+ key = container_of(rcu, struct au_dykey, dk_rcu);
-+ DyPrSym(key);
-+ kfree(key);
-+}
++ int i;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
++ struct au_dr_hino *ent;
+
-+static void dy_free(struct kref *kref)
-+{
-+ struct au_dykey *key;
-+ struct au_splhead *spl;
++ /* SiMustWriteLock(sb); */
+
-+ key = container_of(kref, struct au_dykey, dk_kref);
-+ spl = dynop + key->dk_op.dy_type;
-+ au_spl_del_rcu(&key->dk_list, spl);
-+ call_rcu(&key->dk_rcu, dy_free_rcu);
++ 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);
++ }
+}
+
-+void au_dy_put(struct au_dykey *key)
++/* 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)
+{
-+ kref_put(&key->dk_kref, dy_free);
-+}
-+
-+/* ---------------------------------------------------------------------- */
++ 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;
+
-+#define DyDbgSize(cnt, op) AuDebugOn(cnt != sizeof(op)/sizeof(void *))
++ SiMustWriteLock(sb);
++ AuDebugOn(!au_br_writable(br->br_perm));
+
-+#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
++ hinoinode = file_inode(hinofile);
++ oldsize = i_size_read(hinoinode);
+
-+#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)
++ 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;
+
-+#define DySetForce(func, dst, src) do { \
-+ AuDebugOn(!src.func); \
-+ DyDbgInc(cnt); \
-+ dst.func = src.func; \
-+} while (0)
++ /* 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);
++ }
+
-+#define DySetAop(func) \
-+ DySet(func, dyaop->da_op, aufs_aop, h_aop, h_sb)
-+#define DySetAopForce(func) \
-+ DySetForce(func, dyaop->da_op, aufs_aop)
++ AuTraceErr(err);
++ return err;
++}
+
-+static void dy_aop(struct au_dykey *key, const void *h_op,
-+ struct super_block *h_sb __maybe_unused)
++static int au_dr_hino_load(struct au_dr_br *dr, struct file *hinofile)
+{
-+ struct au_dyaop *dyaop = (void *)key;
-+ const struct address_space_operations *h_aop = h_op;
-+ DyDbgDeclare(cnt);
++ 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;
+
-+ AuDbg("%s\n", au_sbtype(h_sb));
++ 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;
++ }
+
-+ 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);
++ 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 */
+
-+ DyDbgSize(cnt, *h_aop);
++out_free:
++ au_dr_hino_free(dr);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static void dy_bug(struct kref *kref)
++/*
++ * @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)
+{
-+ BUG();
-+}
++ 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);
+
-+static struct au_dykey *dy_get(struct au_dynop *op, struct au_branch *br)
-+{
-+ struct au_dykey *key, *old;
-+ struct au_splhead *spl;
-+ 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;
++ AuDebugOn(bindex < 0 && !br);
++ AuDebugOn(bindex >= 0 && br);
+
-+ spl = dynop + op->dy_type;
-+ key = dy_gfind_get(spl, op->dy_hop);
-+ if (key)
-+ goto out_add; /* success */
++ 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;
++ }
+
-+ p = a + op->dy_type;
-+ key = kzalloc(p->sz, GFP_NOFS);
-+ if (unlikely(!key)) {
-+ key = ERR_PTR(-ENOMEM);
-+ 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;
+
-+ 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(spl, key);
-+ if (old) {
-+ kfree(key);
-+ key = old;
++ 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;
+ }
+
-+out_add:
-+ old = dy_bradd(br, key);
-+ if (old)
-+ /* its ref-count should never be zero here */
-+ kref_put(&key->dk_kref, dy_bug);
++ if (load)
++ err = au_dr_hino_load(&br->br_dirren, hinofile);
++ else
++ err = au_dr_hino_store(sb, br, hinofile);
++ fput(hinofile);
++ goto out;
++
++out_dput:
++ dput(hinopath.dentry);
++out_unlock:
++ if (suspend)
++ au_hn_inode_unlock(hdir);
++ else
++ inode_unlock(dir);
+out:
-+ return key;
++ AuTraceErr(err);
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
-+/*
-+ * 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)
++static int au_dr_brid_init(struct au_dr_brid *brid, const struct path *path)
+{
-+ struct au_dyaop *dyaop;
-+ struct au_dynop op;
++ int err;
++ struct kstatfs kstfs;
++ dev_t dev;
++ struct dentry *dentry;
++ struct super_block *sb;
+
-+ op.dy_type = AuDy_AOP;
-+ op.dy_haop = h_aop;
-+ dyaop = (void *)dy_get(&op, br);
-+ if (IS_ERR(dyaop))
++ err = vfs_statfs((void *)path, &kstfs);
++ AuTraceErr(err);
++ if (unlikely(err))
+ goto out;
-+ dy_adx(dyaop, do_dx);
++
++ /* todo: support for UUID */
++
++ 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;
++ }
++ }
+
+out:
-+ return dyaop;
++ return err;
+}
+
-+int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode)
++int au_dr_br_init(struct super_block *sb, struct au_branch *br,
++ const struct path *path)
+{
-+ int err, do_dx;
-+ struct super_block *sb;
-+ struct au_branch *br;
-+ struct au_dyaop *dyaop;
++ int err, i;
++ struct au_dr_br *dr;
++ struct hlist_bl_head *hbl;
+
-+ AuDebugOn(!S_ISREG(h_inode->i_mode));
-+ IiMustWriteLock(inode);
++ dr = &br->br_dirren;
++ hbl = dr->dr_h_ino;
++ for (i = 0; i < AuDirren_NHASH; i++, hbl++)
++ INIT_HLIST_BL_HEAD(hbl);
+
-+ 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() */
++ err = au_dr_brid_init(&dr->dr_brid, path);
++ if (unlikely(err))
+ goto out;
+
-+ err = 0;
-+ inode->i_mapping->a_ops = &dyaop->da_op;
++ if (au_opt_test(au_mntflags(sb), DIRREN))
++ err = au_dr_hino(sb, /*bindex*/-1, br, path);
+
+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)
++int au_dr_br_fin(struct super_block *sb, struct au_branch *br)
+{
+ int err;
-+ aufs_bindex_t btop;
-+ struct inode *h_inode;
+
+ 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);
-+ }
++ 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);
++
+ return err;
+}
+
-+void au_dy_arefresh(int do_dx)
-+{
-+ struct au_splhead *spl;
-+ struct list_head *head;
-+ struct au_dykey *key;
++/* ---------------------------------------------------------------------- */
+
-+ spl = dynop + AuDy_AOP;
-+ head = &spl->head;
-+ spin_lock(&spl->spin);
-+ list_for_each_entry(key, head, dk_list)
-+ dy_adx((void *)key, do_dx);
-+ spin_unlock(&spl->spin);
++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;
++
++ 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;
++
++ return p - buf;
++}
++
++static int au_drinfo_name(struct au_branch *br, char *name, int len)
++{
++ int rlen;
++ struct dentry *br_dentry;
++ struct inode *br_inode;
++
++ 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);
++
++ return rlen;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * 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)
++{
++ 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;
++}
++
++/* callers have to free the return value */
++static struct au_drinfo *au_drinfo_read_k(struct file *file, ino_t h_ino)
++{
++ struct au_drinfo *ret, *drinfo;
++ struct au_drinfo_fdata fdata;
++ int len;
++ loff_t pos;
++ ssize_t ssz;
++
++ 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;
++ }
++
++ 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;
++ }
++
++ drinfo = &fdata.drinfo;
++ len = drinfo->oldnamelen;
++ if (!len) {
++ AuIOErr("broken drinfo %pD2\n", file);
++ goto out;
++ }
++
++ 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 ret;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* in order to be revertible */
++struct au_drinfo_rev_elm {
++ int created;
++ struct dentry *info_dentry;
++ struct au_drinfo *info_last;
++};
++
++struct au_drinfo_rev {
++ unsigned char already;
++ aufs_bindex_t nelm;
++ struct au_drinfo_rev_elm elm[0];
++};
++
++/* 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 */
++};
++
++/* 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, 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(elm
++ && memcmp(elm, page_address(ZERO_PAGE(0)), sizeof(*elm)));
++
++ 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;
++ 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;
++}
++
++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)
++{
++ struct au_call_drinfo_do_store_args *a = args;
++
++ *a->errp = au_drinfo_do_store(a->w, a->elm);
++}
++
++static int au_drinfo_store_sio(struct au_drinfo_store *w,
++ struct au_drinfo_rev_elm *elm)
++{
++ int err, wkq_err;
++
++ 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;
++}
++
++static int au_drinfo_store_work_init(struct au_drinfo_store *w,
++ aufs_bindex_t btgt)
++{
++ int err;
++
++ 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);
++
++ 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;
++}
++
++static void au_drinfo_store_work_fin(struct au_drinfo_store *w)
++{
++ kfree(w->fdata);
++}
++
++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
++ };
++
++ h_dir = d_inode(w->h_ppath.dentry);
++ IMustLock(h_dir);
++
++ 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 */
++ }
++}
++
++/* 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;
++
++ err = au_drinfo_store_work_init(&work, btgt);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out;
++
++ 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;
++
++ 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;
++
++ err = au_drinfo_construct(&work.fdata, work.h_dentry,
++ &work.allocated);
++ AuTraceErr(err);
++ if (unlikely(err))
++ break;
++
++ 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;
++}
++
++/* ---------------------------------------------------------------------- */
++
++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;
++
++ AuDbg("bindex %d\n", bindex);
++
++ err = -ENOMEM;
++ ent = kmalloc(sizeof(*ent), GFP_NOFS);
++ if (unlikely(!ent))
++ goto out;
++
++ 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 */
++ }
++
++ /* revert */
++ if (!already)
++ au_dr_hino_del(dr, ent);
++ kfree(ent);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++void au_dr_rename_fin(struct dentry *src, aufs_bindex_t btgt, void *_rev)
++{
++ struct au_drinfo_rev *rev;
++ struct au_drinfo_rev_elm *elm;
++ int nelm;
++
++ rev = _rev;
++ elm = rev->elm;
++ for (nelm = rev->nelm; nelm > 0; nelm--, elm++) {
++ dput(elm->info_dentry);
++ kfree(elm->info_last);
++ }
++ kfree(rev);
++}
++
++void au_dr_rename_rev(struct dentry *src, aufs_bindex_t btgt, void *_rev)
++{
++ 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;
++
++ 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");
++}
++
++/* ---------------------------------------------------------------------- */
++
++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 void au_call_drinfo_do_load(void *args)
++{
++ struct au_drinfo_do_load_args *a = args;
++
++ *a->drinfop = au_drinfo_do_load(a->h_ppath, a->whname, a->whnamelen,
++ a->info_dentry);
++}
++
++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 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;
++
++ 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);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static void au_dr_lkup_free(struct au_drinfo **drinfo, int n)
++{
++ struct au_drinfo **p = drinfo;
++
++ while (n-- > 0)
++ kfree(*drinfo++);
++ kfree(p);
++}
++
++int au_dr_lkup(struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t btgt)
++{
++ 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;
++
++ AuDbg("%.*s, name %.*s, whname %.*s, b%d\n",
++ AuLNPair(&dentry->d_name), AuLNPair(&lkup->dirren.dr_name),
++ AuLNPair(&lkup->whname), btgt);
++
++ 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;
++ }
++ 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;
++
++ 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;
++}
++
++void au_dr_lkup_fin(struct au_do_lookup_args *lkup)
++{
++ au_dr_lkup_free(lkup->dirren.drinfo, lkup->dirren.ninfo);
++}
++
++int au_dr_lkup_name(struct au_do_lookup_args *lkup, aufs_bindex_t btgt)
++{
++ int err;
++ struct au_drinfo *drinfo;
++
++ err = 0;
++ 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;
++}
++
++int au_dr_lkup_h_ino(struct au_do_lookup_args *lkup, aufs_bindex_t bindex,
++ ino_t h_ino)
++{
++ int match;
++ struct au_drinfo *drinfo;
++
++ match = 1;
++ if (!lkup->dirren.drinfo)
++ goto out;
++ AuDebugOn(lkup->dirren.ninfo < bindex + 1);
++ drinfo = lkup->dirren.drinfo[bindex + 1];
++ if (!drinfo)
++ goto out;
++
++ match = (drinfo->ino == h_ino);
++ AuDbg("match %d\n", match);
++
++out:
++ return match;
++}
++
++/* ---------------------------------------------------------------------- */
++
++int au_dr_opt_set(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);
++ err = au_dr_hino(sb, bindex, /*br*/NULL, &br->br_path);
++ }
++
++ return err;
++}
++
++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);
++ }
++
++ 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:
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/dirren.h linux/fs/aufs/dirren.h
+--- /usr/share/empty/fs/aufs/dirren.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/dirren.h 2017-11-12 22:24:44.704244405 +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
++ */
++
++#ifndef __AUFS_DIRREN_H__
++#define __AUFS_DIRREN_H__
++
++#ifdef __KERNEL__
++
++#include <linux/dcache.h>
++#include <linux/statfs.h>
++#include <linux/uuid.h>
++#include "hbl.h"
++
++#define AuDirren_NHASH 100
++
++#ifdef CONFIG_AUFS_DIRREN
++enum au_brid_type {
++ AuBrid_Unset,
++ AuBrid_UUID,
++ AuBrid_FSID,
++ AuBrid_DEV
++};
++
++struct au_dr_brid {
++ enum au_brid_type type;
++ union {
++ uuid_t uuid; /* unimplemented yet */
++ fsid_t fsid;
++ dev_t dev;
++ };
++};
++
++/* 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_dr_hino {
++ struct hlist_bl_node dr_hnode;
++ ino_t dr_h_ino;
++};
++
++struct au_dr_br {
++ struct hlist_bl_head dr_h_ino[AuDirren_NHASH];
++ struct au_dr_brid dr_brid;
++};
++
++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
++
++/* ---------------------------------------------------------------------- */
++
++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
++
++/* ---------------------------------------------------------------------- */
++
++#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
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DIRREN_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 2017-11-12 22:24:44.704244405 +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/>.
++ */
++
++/*
++ * dynamically customizable operations for regular files
++ */
++
++#include "aufs.h"
++
++#define DyPrSym(key) AuDbgSym(key->dk_op.dy_hop)
++
++/*
++ * 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 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;
++ }
++ hlist_bl_unlock(hbl);
++
++ return key;
++}
++
++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;
++
++ 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 */
++ }
++
++ return found;
++}
++
++/* kref_get() if @key is already added */
++static struct au_dykey *dy_gadd(struct hlist_bl_head *hbl, struct au_dykey *key)
++{
++ struct au_dykey *tmp, *found;
++ struct hlist_bl_node *pos;
++ const void *h_op = key->dk_op.dy_hop;
++
++ 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);
++
++ if (!found)
++ DyPrSym(key);
++ return found;
++}
++
++static void dy_free_rcu(struct rcu_head *rcu)
++{
++ struct au_dykey *key;
++
++ key = container_of(rcu, struct au_dykey, dk_rcu);
++ DyPrSym(key);
++ kfree(key);
++}
++
++static void dy_free(struct kref *kref)
++{
++ struct au_dykey *key;
++ struct hlist_bl_head *hbl;
++
++ 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);
++}
++
++void au_dy_put(struct au_dykey *key)
++{
++ kref_put(&key->dk_kref, dy_free);
++}
++
++/* ---------------------------------------------------------------------- */
++
++#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)
++
++#define DySetForce(func, dst, src) do { \
++ AuDebugOn(!src.func); \
++ DyDbgInc(cnt); \
++ dst.func = src.func; \
++} while (0)
++
++#define DySetAop(func) \
++ DySet(func, dyaop->da_op, aufs_aop, h_aop, h_sb)
++#define DySetAopForce(func) \
++ DySetForce(func, dyaop->da_op, aufs_aop)
++
++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);
++
++ AuDbg("%s\n", au_sbtype(h_sb));
++
++ 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);
++
++ DyDbgSize(cnt, *h_aop);
++}
++
++/* ---------------------------------------------------------------------- */
++
++static void dy_bug(struct kref *kref)
++{
++ 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;
++
++ hbl = dynop + op->dy_type;
++ key = dy_gfind_get(hbl, op->dy_hop);
++ if (key)
++ goto out_add; /* success */
++
++ p = a + op->dy_type;
++ key = kzalloc(p->sz, GFP_NOFS);
++ if (unlikely(!key)) {
++ key = ERR_PTR(-ENOMEM);
++ 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(hbl, key);
++ if (old) {
++ kfree(key);
++ key = old;
++ }
++
++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;
++}
++
++/* ---------------------------------------------------------------------- */
++/*
++ * 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;
++
++ 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);
++
++out:
++ return dyaop;
++}
++
++int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode)
++{
++ int err, do_dx;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_dyaop *dyaop;
++
++ 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;
++
++ err = 0;
++ inode->i_mapping->a_ops = &dyaop->da_op;
++
++out:
++ 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)
++{
++ int err;
++ aufs_bindex_t btop;
++ struct inode *h_inode;
++
++ 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;
++}
++
++void au_dy_arefresh(int do_dx)
++{
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dykey *key;
++
++ 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);
+}
+
+/* ---------------------------------------------------------------------- */
+ BUILD_BUG_ON(offsetof(struct au_dyaop, da_key));
+
+ for (i = 0; i < AuDyLast; i++)
-+ au_spl_init(dynop + i);
++ INIT_HLIST_BL_HEAD(dynop + i);
+}
+
+void au_dy_fin(void)
+ int i;
+
+ for (i = 0; i < AuDyLast; i++)
-+ WARN_ON(!list_empty(&dynop[i].head));
++ WARN_ON(!hlist_bl_empty(dynop + i));
+}
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 2016-07-25 19:05:34.811159821 +0200
++++ linux/fs/aufs/dynop.h 2017-11-12 22:24:44.704244405 +0100
@@ -0,0 +1,74 @@
+/*
-+ * 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
+
+struct au_dykey {
+ union {
-+ struct list_head dk_list;
++ struct hlist_bl_node dk_hnode;
+ struct rcu_head dk_rcu;
+ };
+ struct au_dynop dk_op;
+#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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,837 @@
++++ linux/fs/aufs/export.c 2017-11-12 22:24:42.267510077 +0100
+@@ -0,0 +1,836 @@
+/*
-+ * 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/nsproxy.h>
+#include <linux/random.h>
+#include <linux/writeback.h>
-+#include "../fs/mount.h"
+#include "aufs.h"
+
+union conv {
+}
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
++++ linux/fs/aufs/fhsm.c 2017-07-29 12:14:25.903042072 +0200
@@ -0,0 +1,426 @@
+/*
-+ * 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
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,843 @@
++++ linux/fs/aufs/file.c 2017-11-12 22:24:44.704244405 +0100
+@@ -0,0 +1,856 @@
+/*
-+ * 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 au_cmoo(struct dentry *dentry)
+{
-+ int err, cmoo;
++ int err, cmoo, matched;
+ unsigned int udba;
+ struct path h_path;
+ struct au_pin pin;
+ sbinfo = au_sbi(sb);
+ fhsm = &sbinfo->si_fhsm;
+ pid = au_fhsm_pid(fhsm);
-+ if (pid
-+ && (current->pid == pid
-+ || current->real_parent->pid == pid))
++ rcu_read_lock();
++ matched = (pid
++ && (current->pid == pid
++ || rcu_dereference(current->real_parent)->pid == pid));
++ rcu_read_unlock();
++ if (matched)
+ goto out;
+
+ br = au_sbr(sb, cpg.bsrc);
+
+int au_do_open(struct file *file, struct au_do_open_args *args)
+{
-+ int err, no_lock = args->no_lock;
++ int err, aopen = args->aopen;
+ struct dentry *dentry;
+ struct au_finfo *finfo;
+
-+ if (!no_lock)
++ if (!aopen)
+ err = au_finfo_init(file, args->fidir);
+ else {
+ lockdep_off();
+
+ 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)
++ 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);
-+ 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);
++ else {
++ lockdep_off();
++ err = args->open(file, vfsub_file_flags(file), NULL);
++ lockdep_on();
++ }
+ }
++ di_read_unlock(dentry, AuLock_IR);
+
+ 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);
++ au_hbl_add(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
+ }
-+ if (!no_lock)
++ if (!aopen)
+ fi_write_unlock(file);
+ else {
+ lockdep_off();
+ }
+
+out:
++ AuTraceErr(err);
+ return err;
+}
+
+ if (p->hf_file) {
+ if (file_inode(p->hf_file))
+ break;
-+ au_hfput(p, file);
++ 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, file);
++ au_hfput(p, /*execed*/0);
+ bbot = au_sbbot(sb);
+ }
+
+ if (p->hf_file) {
+ if (file_inode(p->hf_file))
+ break;
-+ au_hfput(p, file);
++ au_hfput(p, /*execed*/0);
+ }
+ AuDebugOn(fidir->fd_bbot < finfo->fi_btop);
+}
+ */
+static int refresh_file(struct file *file, int (*reopen)(struct file *file))
+{
-+ int err, need_reopen;
++ 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;
+
+ 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(dentry->d_sb, hfile->hf_br->br_id);
++ 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, au_sbbot(dentry->d_sb) + 1);
++ err = au_fidir_realloc(finfo, nbr, /*may_shrink*/0);
+ if (unlikely(err))
+ goto out;
+ au_do_refresh_dir(file);
+ 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) {
+
+/* common function to regular file and dir */
+int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
-+ int wlock)
++ int wlock, unsigned int fi_lsc)
+{
+ int err;
+ unsigned int sigen, figen;
+ dentry = file->f_path.dentry;
+ inode = d_inode(dentry);
+ sigen = au_sigen(dentry->d_sb);
-+ fi_write_lock(file);
++ fi_write_lock_nested(file, fi_lsc);
+ figen = au_figen(file);
-+ di_write_lock_child(dentry);
++ 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) {
+ struct page *page, enum migrate_mode mode)
+{ AuUnsupport(); return 0; }
+#endif
++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,
+ .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,
+};
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,291 @@
++++ linux/fs/aufs/file.h 2017-11-12 22:24:44.704244405 +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
+
+#include <linux/file.h>
+#include <linux/fs.h>
++#include <linux/mm_types.h>
+#include <linux/poll.h>
+#include "rwsem.h"
+
+ };
+ struct au_fidir *fi_hdir; /* for dir only */
+
-+ struct hlist_node fi_hlist;
++ struct hlist_bl_node fi_hlist;
+ struct file *fi_file; /* very ugly */
+} ____cacheline_aligned_in_smp;
+
+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 aopen;
+ int (*open)(struct file *file, int flags,
+ struct file *h_file);
+ struct au_fidir *fidir;
+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 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));
+
+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);
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc);
+
+/* finfo.c */
-+void au_hfput(struct au_hfile *hf, struct file *file);
++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);
+
+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 au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink);
+
+void au_fi_init_once(void *_fi);
+void au_finfo_fin(struct file *file);
+
+/* ---------------------------------------------------------------------- */
+
++#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)
++*/
++
++#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)
++
++/* lock subclass for finfo */
++enum {
++ AuLsc_FI_1,
++ AuLsc_FI_2
++};
++
++static inline void fi_read_lock_nested(struct file *f, unsigned int lsc)
++{
++ au_rw_read_lock_nested(&au_fi(f)->fi_rwsem, lsc);
++}
++
++static inline void fi_write_lock_nested(struct file *f, unsigned int lsc)
++{
++ au_rw_write_lock_nested(&au_fi(f)->fi_rwsem, lsc);
++}
++
+/*
-+ * fi_read_lock, fi_write_lock,
-+ * fi_read_unlock, fi_write_unlock, fi_downgrade_lock
++ * fi_read_lock_1, fi_write_lock_1,
++ * fi_read_lock_2, fi_write_lock_2
+ */
-+AuSimpleRwsemFuncs(fi, struct file *f, &au_fi(f)->fi_rwsem);
++#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)
+#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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,149 @@
++++ linux/fs/aufs/finfo.c 2017-11-12 22:24:42.270843503 +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
+
+#include "aufs.h"
+
-+void au_hfput(struct au_hfile *hf, struct file *file)
++void au_hfput(struct au_hfile *hf, int execed)
+{
-+ /* todo: direct access f_flags */
-+ if (vfsub_file_flags(file) & __FMODE_EXEC)
++ if (execed)
+ allow_write_access(hf->hf_file);
+ fput(hf->hf_file);
+ hf->hf_file = NULL;
+ hf = fidir->fd_hfile + bindex;
+
+ if (hf && hf->hf_file)
-+ au_hfput(hf, file);
++ au_hfput(hf, vfsub_file_execed(file));
+ if (val) {
+ FiMustWriteLock(file);
+ AuDebugOn(IS_ERR_OR_NULL(file->f_path.dentry));
+ return fidir;
+}
+
-+int au_fidir_realloc(struct au_finfo *finfo, int nbr)
++int au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink)
+{
+ int err;
+ struct au_fidir *fidir, *p;
+
+ err = -ENOMEM;
+ p = au_kzrealloc(fidir, au_fidir_sz(fidir->fd_nent), au_fidir_sz(nbr),
-+ GFP_NOFS);
++ GFP_NOFS, may_shrink);
+ if (p) {
+ p->fd_nent = nbr;
+ finfo->fi_hdir = p;
+}
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 2016-07-25 19:05:34.811159821 +0200
-@@ -0,0 +1,770 @@
++++ linux/fs/aufs/f_op.c 2017-11-12 22:24:44.704244405 +0100
+@@ -0,0 +1,817 @@
+/*
-+ * 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
+ aufs_bindex_t bindex;
+
+ finfo = au_fi(file);
-+ 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);
+ bindex = finfo->fi_btop;
+ if (bindex >= 0)
+ au_set_h_fptr(file, bindex, NULL);
+ */
+
+/* Callers should call au_read_post() or fput() in the end */
-+struct file *au_read_pre(struct file *file, int keep_fi)
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc)
+{
+ struct file *h_file;
+ int err;
+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0);
++ 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);
+}
+
+struct au_write_pre {
++ /* input */
++ unsigned int lsc;
++
++ /* output */
+ blkcnt_t blks;
+ aufs_bindex_t btop;
+};
+ struct file *h_file;
+ struct dentry *dentry;
+ int err;
++ unsigned int lsc;
+ struct au_pin pin;
+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1);
++ 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;
+ 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);
++ fput(h_file);
+}
+
+static ssize_t aufs_read(struct file *file, char __user *buf, size_t count,
+ sb = inode->i_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;
+ 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))
+ sb = inode->i_sb;
+ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ h_file = au_read_pre(file, /*keep_fi*/1);
++ h_file = au_read_pre(file, /*keep_fi*/1, /*lsc*/0);
+ err = PTR_ERR(h_file);
+ if (IS_ERR(h_file))
+ 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))
+ sb = inode->i_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;
+ 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))
+ 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))
+ 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)
++{
++ 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
++}
++
+/* ---------------------------------------------------------------------- */
+
+/*
+ * au_flag_conv(vma->vm_flags));
+ */
+ if (!err)
-+ err = h_file->f_op->mmap(h_file, vma);
++ err = call_mmap(h_file, vma);
+ if (!err) {
+ au_vm_prfile_set(vma, file);
+ fsstack_copy_attr_atime(inode, file_inode(h_file));
+ 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))
+ return err;
+}
+
-+/* no one supports this operation, currently */
-+#if 0
-+static int aufs_aio_fsync_nondir(struct kiocb *kio, int datasync)
-+{
-+ int err;
-+ struct au_write_pre wpre;
-+ struct inode *inode, *h_inode;
-+ struct file *file, *h_file;
-+
-+ err = 0; /* -EBADF; */ /* posix? */
-+ if (unlikely(!(file->f_mode & FMODE_WRITE)))
-+ goto out;
-+
-+ 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_unlock;
-+
-+ 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);
-+ }
-+ 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;
-+}
-+#endif
-+
+static int aufs_fasync(int fd, struct file *file, int flag)
+{
+ int err;
+ 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;
+ 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;
+
-+ arg |= vfsub_file_flags(file) & FASYNC; /* stop calling h_file->fasync */
++ /* 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() */
+
+ .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,
+ .aio_splice_write = aufs_aio_splice_write,
+ .aio_splice_read = aufs_aio_splice_read,
+#endif
-+ .fallocate = aufs_fallocate
++ .fallocate = aufs_fallocate,
++ .copy_file_range = aufs_copy_file_range
+};
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 2016-07-25 19:05:34.814493242 +0200
++++ linux/fs/aufs/fstype.h 2017-07-29 12:14:25.903042072 +0200
@@ -0,0 +1,400 @@
+/*
-+ * 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 inline int au_test_iso9660(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_ISO9660_FS) || defined(CONFIG_ISO9660_FS_MODULE)
++#if IS_ENABLED(CONFIG_ISO9660_FS)
+ return sb->s_magic == ISOFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_romfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_ROMFS_FS) || defined(CONFIG_ROMFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_ROMFS_FS)
+ return sb->s_magic == ROMFS_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_cramfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_CRAMFS) || defined(CONFIG_CRAMFS_MODULE)
++#if IS_ENABLED(CONFIG_CRAMFS)
+ return sb->s_magic == CRAMFS_MAGIC;
+#endif
+ return 0;
+
+static inline int au_test_nfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_NFS_FS) || defined(CONFIG_NFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_NFS_FS)
+ return sb->s_magic == NFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_fuse(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_FUSE_FS) || defined(CONFIG_FUSE_FS_MODULE)
++#if IS_ENABLED(CONFIG_FUSE_FS)
+ return sb->s_magic == FUSE_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_xfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_XFS_FS) || defined(CONFIG_XFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_XFS_FS)
+ return sb->s_magic == XFS_SB_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_ECRYPT_FS) || defined(CONFIG_ECRYPT_FS_MODULE)
++#if IS_ENABLED(CONFIG_ECRYPT_FS)
+ return !strcmp(au_sbtype(sb), "ecryptfs");
+#else
+ return 0;
+
+static inline int au_test_ubifs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_UBIFS_FS) || defined(CONFIG_UBIFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_UBIFS_FS)
+ return sb->s_magic == UBIFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_configfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_CONFIGFS_FS) || defined(CONFIG_CONFIGFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_CONFIGFS_FS)
+ return sb->s_magic == CONFIGFS_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_minix(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_MINIX_FS) || defined(CONFIG_MINIX_FS_MODULE)
++#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
+
+static inline int au_test_fat(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_FAT_FS) || defined(CONFIG_FAT_FS_MODULE)
++#if IS_ENABLED(CONFIG_FAT_FS)
+ return sb->s_magic == MSDOS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_squashfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_SQUASHFS) || defined(CONFIG_SQUASHFS_MODULE)
++#if IS_ENABLED(CONFIG_SQUASHFS)
+ return sb->s_magic == SQUASHFS_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_btrfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_BTRFS_FS) || defined(CONFIG_BTRFS_FS_MODULE)
++#if IS_ENABLED(CONFIG_BTRFS_FS)
+ return sb->s_magic == BTRFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_xenfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_XENFS) || defined(CONFIG_XENFS_MODULE)
++#if IS_ENABLED(CONFIG_XENFS)
+ return sb->s_magic == XENFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_nilfs(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_NILFS) || defined(CONFIG_NILFS_MODULE)
++#if IS_ENABLED(CONFIG_NILFS)
+ return sb->s_magic == NILFS_SUPER_MAGIC;
+#else
+ return 0;
+
+static inline int au_test_hfsplus(struct super_block *sb __maybe_unused)
+{
-+#if defined(CONFIG_HFSPLUS_FS) || defined(CONFIG_HFSPLUS_FS_MODULE)
++#if IS_ENABLED(CONFIG_HFSPLUS_FS)
+ return sb->s_magic == HFSPLUS_SUPER_MAGIC;
+#else
+ return 0;
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_FSTYPE_H__ */
+diff -urN /usr/share/empty/fs/aufs/hbl.h linux/fs/aufs/hbl.h
+--- /usr/share/empty/fs/aufs/hbl.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/hbl.h 2017-11-12 22:24:44.707577830 +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/>.
++ */
++
++/*
++ * helpers for hlist_bl.h
++ */
++
++#ifndef __AUFS_HBL_H__
++#define __AUFS_HBL_H__
++
++#ifdef __KERNEL__
++
++#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 inline void au_hbl_del(struct hlist_bl_node *node,
++ struct hlist_bl_head *hbl)
++{
++ hlist_bl_lock(hbl);
++ hlist_bl_del(node);
++ hlist_bl_unlock(hbl);
++}
++
++#define au_hbl_for_each(pos, head) \
++ for (pos = hlist_bl_first(head); \
++ pos; \
++ pos = pos->next)
++
++static inline unsigned long au_hbl_count(struct hlist_bl_head *hbl)
++{
++ unsigned long cnt;
++ struct hlist_bl_node *pos;
++
++ cnt = 0;
++ hlist_bl_lock(hbl);
++ au_hbl_for_each(pos, hbl)
++ cnt++;
++ hlist_bl_unlock(hbl);
++ return cnt;
++}
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_HBL_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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,287 @@
++++ linux/fs/aufs/hfsnotify.c 2017-11-12 22:24:44.707577830 +0100
+@@ -0,0 +1,289 @@
+/*
-+ * 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
+ hn_mark);
+ /* AuDbg("here\n"); */
+ au_cache_free_hnotify(hn);
-+ smp_mb__before_atomic();
++ smp_mb__before_atomic(); /* for atomic64_dec */
+ if (atomic64_dec_and_test(&au_hfsn_ifree))
+ wake_up(&au_hfsn_wq);
+}
+ AuDebugOn(!br->br_hfsn);
+
+ mark = &hn->hn_mark;
-+ fsnotify_init_mark(mark, au_hfsn_free_mark);
++ fsnotify_init_mark(mark, br->br_hfsn->hfsn_group);
+ 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);
++ err = fsnotify_add_mark(mark, hinode->hi_inode, /*mnt*/NULL,
++ /*allow_dups*/1);
+ lockdep_on();
+
+ return err;
+ 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)
++ 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;
+
+static struct fsnotify_ops au_hfsn_ops = {
+ .handle_event = au_hfsn_handle_event,
-+ .free_group_priv = au_hfsn_free_group
++ .free_group_priv = au_hfsn_free_group,
++ .free_mark = au_hfsn_free_mark
+};
+
+/* ---------------------------------------------------------------------- */
+};
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
++++ linux/fs/aufs/hfsplus.c 2017-07-29 12:14:25.903042072 +0200
@@ -0,0 +1,56 @@
+/*
-+ * 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
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,710 @@
++++ linux/fs/aufs/hnotify.c 2017-11-12 22:24:44.707577830 +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
+ if (au_ftest_hnjob(a->flags, TRYXINO0)
+ && a->inode
+ && a->h_inode) {
-+ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
++ 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(a->h_inode);
++ inode_unlock_shared(a->h_inode);
+ }
+
+ /* make the generation obsolete */
+ AuDebugOn(!sbinfo);
+ si_write_lock(sb, AuLock_NOPLMW);
+
++ 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");
++ }
++
+ ii_read_lock_parent(a->dir);
+ bfound = -1;
+ bbot = au_ibbot(a->dir);
+ || au_ftest_hnjob(a->flags[AuHn_CHILD], GEN))) {
+ inode = lookup_wlock_by_ino(sb, bfound, h_ino);
+ try_iput = 1;
-+ }
++ }
+
+ args.flags = a->flags[AuHn_CHILD];
+ args.dentry = dentry;
+
+static void au_hn_destroy_cache(void)
+{
-+ kmem_cache_destroy(au_cachep[AuCache_HNOTIFY]);
-+ au_cachep[AuCache_HNOTIFY] = NULL;
++ kmem_cache_destroy(au_cache[AuCache_HNOTIFY]);
++ au_cache[AuCache_HNOTIFY] = NULL;
+}
+
+int __init au_hnotify_init(void)
+ int err;
+
+ err = -ENOMEM;
-+ au_cachep[AuCache_HNOTIFY] = AuCache(au_hnotify);
-+ if (au_cachep[AuCache_HNOTIFY]) {
++ 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 (au_hnotify_op.fin)
+ au_hnotify_op.fin();
++
+ /* cf. au_cache_fin() */
-+ if (au_cachep[AuCache_HNOTIFY])
++ if (au_cache[AuCache_HNOTIFY])
+ au_hn_destroy_cache();
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,284 @@
++++ linux/fs/aufs/iinfo.c 2017-11-12 22:24:42.270843503 +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
+ return -ENOMEM;
+}
+
-+int au_hinode_realloc(struct au_iinfo *iinfo, int nbr)
++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 = krealloc(iinfo->ii_hinode, sizeof(*hip) * nbr, GFP_NOFS);
++ hip = au_krealloc(iinfo->ii_hinode, sizeof(*hip) * nbr, GFP_NOFS,
++ may_shrink);
+ if (hip) {
+ iinfo->ii_hinode = hip;
+ i = iinfo->ii_bbot + 1;
+}
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 @@
++++ linux/fs/aufs/inode.c 2017-11-12 22:24:42.270843503 +0100
+@@ -0,0 +1,527 @@
+/*
-+ * 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 au_ii_refresh(struct inode *inode, int *update)
+{
-+ int err, e;
++ int err, e, nbr;
+ umode_t type;
+ aufs_bindex_t bindex, new_bindex;
+ struct super_block *sb;
+
+ *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, au_sbbot(sb) + 1);
++ err = au_hinode_realloc(iinfo, nbr, /*may_shrink*/0);
+ if (unlikely(err))
+ goto out;
+
+ }
+ }
+ 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;
+int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
+ unsigned int d_type, ino_t *ino)
+{
-+ int err;
-+ struct mutex *mtx;
++ int err, idx;
++ const int isnondir = d_type != DT_DIR;
+
+ /* 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);
++ if (isnondir) {
++ err = au_xinondir_enter(sb, bindex, h_ino, &idx);
++ if (unlikely(err))
++ goto out;
+ }
++
+ err = au_xino_read(sb, bindex, h_ino, ino);
+ if (unlikely(err))
-+ goto out;
++ goto out_xinondir;
+
+ if (!*ino) {
+ err = -EIO;
+ *ino = au_xino_new_ino(sb);
+ if (unlikely(!*ino))
-+ goto out;
++ goto out_xinondir;
+ err = au_xino_write(sb, bindex, h_ino, *ino);
+ if (unlikely(err))
-+ goto out;
++ goto out_xinondir;
+ }
+
++out_xinondir:
++ if (isnondir && idx >= 0)
++ au_xinondir_leave(sb, bindex, h_ino, idx);
+out:
-+ if (mtx)
-+ mutex_unlock(mtx);
+ return err;
+}
+
+ struct inode *inode, *h_inode;
+ struct dentry *h_dentry;
+ struct super_block *sb;
-+ struct mutex *mtx;
+ ino_t h_ino, ino;
-+ int err;
++ int err, idx, hlinked;
+ aufs_bindex_t btop;
+
+ sb = dentry->d_sb;
+ 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;
+
++new_ino:
+ /*
+ * 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;
++ if (hlinked) {
++ err = au_xinondir_enter(sb, btop, h_ino, &idx);
++ inode = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
++ }
+
-+new_ino:
-+ if (mtx)
-+ mutex_lock(mtx);
+ err = au_xino_read(sb, btop, h_ino, &ino);
+ inode = ERR_PTR(err);
+ if (unlikely(err))
-+ goto out;
++ goto out_xinondir;
+
+ if (!ino) {
+ ino = au_xino_new_ino(sb);
+ if (unlikely(!ino)) {
+ inode = ERR_PTR(-EIO);
-+ goto out;
++ goto out_xinondir;
+ }
+ }
+
+ inode = au_iget_locked(sb, ino);
+ err = PTR_ERR(inode);
+ if (IS_ERR(inode))
-+ goto out;
++ goto out_xinondir;
+
+ AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW));
+ if (inode->i_state & I_NEW) {
+ err = set_inode(inode, dentry);
+ if (!err) {
+ unlock_new_inode(inode);
-+ goto out; /* success */
++ goto out_xinondir; /* success */
+ }
+
+ /*
+ * horrible race condition between lookup, readdir and copyup
+ * (or something).
+ */
-+ if (mtx)
-+ mutex_unlock(mtx);
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
+ err = reval_inode(inode, dentry);
+ if (unlikely(err < 0)) {
-+ mtx = NULL;
++ hlinked = 0;
+ goto out_iput;
+ }
-+
-+ if (!err) {
-+ mtx = NULL;
++ if (!err)
+ goto out; /* success */
-+ } else if (mtx)
-+ mutex_lock(mtx);
++ 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;
++ }
++ }
+ }
+
+ if (unlikely(au_test_fs_unique_ino(h_inode)))
+ err = au_xino_write(sb, btop, h_ino, /*ino*/0);
+ if (!err) {
+ iput(inode);
-+ if (mtx)
-+ mutex_unlock(mtx);
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
+ goto new_ino;
+ }
+
+out_iput:
+ iput(inode);
+ inode = ERR_PTR(err);
++out_xinondir:
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
+out:
-+ if (mtx)
-+ mutex_unlock(mtx);
+ return inode;
+}
+
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,694 @@
++++ linux/fs/aufs/inode.h 2017-11-12 22:24:44.707577830 +0100
+@@ -0,0 +1,695 @@
+/*
-+ * 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
+ /* never use fsnotify_add_vfsmount_mark() */
+ struct fsnotify_mark hn_mark;
+#endif
-+ struct inode *hn_aufs_inode; /* no get/put */
++ struct inode *hn_aufs_inode; /* no get/put */
+#endif
+} ____cacheline_aligned_in_smp;
+
+struct au_icntnr {
+ struct au_iinfo iinfo;
+ struct inode vfs_inode;
-+ struct hlist_node plink;
++ struct hlist_bl_node plink;
+} ____cacheline_aligned_in_smp;
+
+/* au_pin flags */
+int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
+ struct au_icpup_args *a);
+
-+int au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path);
++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,
+/* 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 inode *dir, struct dentry *dentry,
++ unsigned int flags);
+
+/* iinfo.c */
+struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex);
+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 au_hinode_realloc(struct au_iinfo *iinfo, int nbr, int may_shrink);
+
+#ifdef CONFIG_PROC_FS
+/* plink.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);
-+
-+/* void au_xattr_init(struct super_block *sb); */
++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); */
++AuStubVoid(au_xattr_init, struct super_block *sb);
+#endif
+
+#ifdef CONFIG_FS_POSIX_ACL
+#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
+enum {
+ AU_XATTR_SET,
-+ AU_XATTR_REMOVE,
+ AU_ACL_SET
+};
+
-+struct au_srxattr {
++struct au_sxattr {
+ int type;
+ union {
+ struct {
+ 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);
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg);
+#endif
+
+/* ---------------------------------------------------------------------- */
+#undef AuWriteLockFunc
+#undef AuRWLockFuncs
+
-+/*
-+ * ii_read_unlock, ii_write_unlock, ii_downgrade_lock
-+ */
-+AuSimpleUnlockRwsemFuncs(ii, struct inode *i, &au_ii(i)->ii_rwsem);
++#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)
+
+#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem)
+#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem)
+ au_hn_suspend(hdir);
+}
+
++#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
++
+static inline void au_hn_inode_unlock(struct au_hinode *hdir)
+{
+ au_hn_resume(hdir);
+#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
++++ linux/fs/aufs/ioctl.c 2017-07-29 12:14:25.903042072 +0200
@@ -0,0 +1,219 @@
+/*
-+ * 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
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,924 @@
++++ linux/fs/aufs/i_op_add.c 2017-11-12 22:24:42.270843503 +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
+ 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 */)
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,1414 @@
++++ linux/fs/aufs/i_op.c 2017-11-12 22:24:44.707577830 +0100
+@@ -0,0 +1,1459 @@
+/*
-+ * 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 err;
+ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
+
++ err = -EPERM;
++ if (write_mask && IS_IMMUTABLE(h_inode))
++ goto out;
++
+ 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))))
++ if (((mask & MAY_EXEC)
++ && S_ISREG(h_inode->i_mode)
++ && (path_noexec(h_path)
++ || !(h_inode->i_mode & S_IXUGO))))
+ goto out;
+
+ /*
+/* ---------------------------------------------------------------------- */
+
+struct aopen_node {
-+ struct hlist_node hlist;
++ struct hlist_bl_node hblist;
+ struct file *file, *h_file;
+};
+
+static int au_do_aopen(struct inode *inode, struct file *file)
+{
-+ struct au_sphlhead *aopen;
++ struct hlist_bl_head *aopen;
++ struct hlist_bl_node *pos;
+ struct aopen_node *node;
+ struct au_do_open_args args = {
-+ .no_lock = 1,
-+ .open = au_do_open_nondir
++ .aopen = 1,
++ .open = au_do_open_nondir
+ };
+
+ aopen = &au_sbi(inode->i_sb)->si_aopen;
-+ spin_lock(&aopen->spin);
-+ hlist_for_each_entry(node, &aopen->head, hlist)
++ hlist_bl_lock(aopen);
++ hlist_bl_for_each_entry(node, pos, aopen, hblist)
+ if (node->file == file) {
+ args.h_file = node->h_file;
+ break;
+ }
-+ spin_unlock(&aopen->spin);
++ hlist_bl_unlock(aopen);
+ /* AuDebugOn(!args.h_file); */
+
+ return au_do_open(file, &args);
+ struct file *file, unsigned int open_flag,
+ umode_t create_mode, int *opened)
+{
-+ int err, h_opened = *opened;
++ int err, unlocked, h_opened = *opened;
+ unsigned int lkup_flags;
-+ struct dentry *parent;
-+ struct dentry *d;
-+ struct au_sphlhead *aopen;
++ struct dentry *parent, *d;
++ struct hlist_bl_head *aopen;
+ struct vfsub_aopen_args args = {
+ .open_flag = open_flag,
+ .create_mode = create_mode,
+ || !(open_flag & O_CREAT))
+ goto out_no_open;
+
++ unlocked = 0;
+ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
+ if (unlikely(err))
+ goto out;
+ put_filp(args.file);
+ goto out_unlock;
+ }
++ di_write_unlock(parent);
++ di_write_unlock(dentry);
++ unlocked = 1;
+
+ /* some filesystems don't set FILE_CREATED while succeeded? */
+ *opened |= FILE_CREATED;
+ args.file = NULL;
+ }
+ aopen = &au_sbi(dir->i_sb)->si_aopen;
-+ au_sphl_add(&aopen_node.hlist, aopen);
++ au_hbl_add(&aopen_node.hblist, aopen);
+ err = finish_open(file, dentry, au_do_aopen, opened);
-+ au_sphl_del(&aopen_node.hlist, aopen);
++ au_hbl_del(&aopen_node.hblist, aopen);
+ AuTraceErr(err);
+ AuDbgFile(file);
+ if (aopen_node.h_file)
+ fput(aopen_node.h_file);
+
+out_unlock:
-+ di_write_unlock(parent);
-+ aufs_read_unlock(dentry, AuLock_DW);
++ if (unlocked)
++ si_read_unlock(dentry->d_sb);
++ else {
++ di_write_unlock(parent);
++ aufs_read_unlock(dentry, AuLock_DW);
++ }
+ AuDbgDentry(dentry);
-+ if (unlikely(err))
++ if (unlikely(err < 0))
+ goto out;
+out_no_open:
-+ if (!err && !(*opened & FILE_CREATED)) {
++ if (err >= 0 && !(*opened & FILE_CREATED)) {
+ AuLabel(out_no_open);
+ dget(dentry);
+ err = finish_no_open(file, dentry);
+ 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);
++ vfsub_inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
+ err = au_lkup_neg(dentry, bcpup, /*wh*/0);
+ /* todo: no unlock here */
-+ inode_unlock(h_dir);
++ inode_unlock_shared(h_dir);
+
+ AuDbg("bcpup %d\n", bcpup);
+ if (!err) {
+ 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);
++ 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(a->h_inode);
++ inode_unlock_shared(a->h_inode);
+ }
+
+ hi_wh = NULL;
+ inode = d_inode(dentry);
+ IMustLock(inode);
+
++ err = setattr_prepare(dentry, ia);
++ if (unlikely(err))
++ goto out;
++
+ err = -ENOMEM;
+ a = kzalloc(sizeof(*a), GFP_NOFS);
+ if (unlikely(!a))
+ /* 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);
++ 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);
+ return err;
+}
+
-+ssize_t au_srxattr(struct dentry *dentry, struct inode *inode,
-+ struct au_srxattr *arg)
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg)
+{
+ int err;
+ struct path h_path;
+ 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)
++ /* 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);
+}
+
+/*
-+ * common routine for aufs_getattr() and aufs_getxattr().
++ * 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 au_h_path_getattr(struct dentry *dentry, int force, struct path *h_path,
++ int locked)
+{
+ int err;
+ unsigned int mnt_flags, sigen;
+ mnt_flags = au_mntflags(sb);
+ udba_none = !!au_opt_test(mnt_flags, UDBA_NONE);
+
++ if (unlikely(locked))
++ goto body; /* skip locking dinfo */
++
+ /* support fstat(2) */
+ if (!d_unlinked(dentry) && !udba_none) {
+ sigen = au_sigen(sb);
+ } else
+ di_read_lock_child(dentry, AuLock_IR);
+
++body:
+ inode = d_inode(dentry);
+ bindex = au_ibtop(inode);
+ h_path->mnt = au_sbr_mnt(sb, bindex);
+ return err;
+}
+
-+static int aufs_getattr(struct vfsmount *mnt __maybe_unused,
-+ struct dentry *dentry, struct kstat *st)
++static int aufs_getattr(const struct path *path, struct kstat *st,
++ u32 request, unsigned int query)
+{
+ int err;
+ unsigned char positive;
+ struct path h_path;
++ struct dentry *dentry;
+ struct inode *inode;
+ struct super_block *sb;
+
++ 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);
++ err = au_h_path_getattr(dentry, /*force*/0, &h_path, /*locked*/0);
+ if (unlikely(err))
+ goto out_si;
+ if (unlikely(!h_path.dentry))
+
+ positive = d_is_positive(h_path.dentry);
+ if (positive)
-+ err = vfs_getattr(&h_path, st);
++ /* no vfsub version */
++ err = vfs_getattr(&h_path, st, request, query);
+ if (!err) {
+ if (positive)
+ au_refresh_iattr(inode, st,
+ 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);
++ ret = vfs_get_link(h_dentry, done);
+ dput(h_dentry);
+ if (IS_ERR(ret))
+ err = PTR_ERR(ret);
+
+/* ---------------------------------------------------------------------- */
+
++static int au_is_special(struct inode *inode)
++{
++ return (inode->i_mode & (S_IFBLK | S_IFCHR | S_IFIFO | S_IFSOCK));
++}
++
+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;
+
+ sb = inode->i_sb;
++ WARN_ONCE((flags & S_ATIME) && !IS_NOATIME(inode),
++ "unexpected s_flags 0x%lx", sb->s_flags);
++
+ /* 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();
++
++ 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);
++ }
++
+ if (!err)
+ au_cpup_attr_timesizes(inode);
+ ii_write_unlock(inode);
+ .getattr = aufs_getattr,
+
+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
+#endif
+
-+ .readlink = generic_readlink,
+ .get_link = aufs_get_link,
+
+ /* .update_time = aufs_update_time */
+ .getattr = aufs_getattr,
+
+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
+#endif
+
+ .update_time = aufs_update_time,
+ .getattr = aufs_getattr,
+
+#ifdef CONFIG_AUFS_XATTR
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
+#endif
+
+ .update_time = aufs_update_time
+};
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 2016-07-25 19:05:34.814493242 +0200
++++ linux/fs/aufs/i_op_del.c 2017-11-12 22:24:42.270843503 +0100
@@ -0,0 +1,511 @@
+/*
-+ * 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
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,1015 @@
++++ linux/fs/aufs/i_op_ren.c 2017-11-12 22:24:44.707577830 +0100
+@@ -0,0 +1,1246 @@
+/*
-+ * 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
+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 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)
+
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuRen_DIRREN
++#define AuRen_DIRREN 0
++#endif
++
+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_hinode *hdir, *hinode;
+ struct au_dtime dt[AuParentChild];
-+ aufs_bindex_t btop;
++ aufs_bindex_t btop, bdiropq;
+ } sd[AuSrcDst];
+
+#define src_dentry sd[AuSRC].dentry
+#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
+
+#define dst_dentry sd[AuDST].dentry
+#define dst_dir sd[AuDST].dir
+#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 au_hinode *src_hinode;
+ struct path h_path;
+ struct au_nhash whlist;
-+ aufs_bindex_t btgt, src_bwh, src_bdiropq;
++ aufs_bindex_t btgt, src_bwh;
+
-+ unsigned int flags;
++ 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;
+};
+
+/* ---------------------------------------------------------------------- */
+ err = -EIO; \
+} while (0)
+
-+static void au_ren_rev_diropq(int err, struct au_ren_args *a)
++static void au_ren_do_rev_diropq(int err, struct au_ren_args *a, int idx)
+{
+ int rerr;
++ struct dentry *d;
++#define src_or_dst(member) a->sd[idx].member
+
-+ 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);
++ 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", a->src_dentry);
++ RevertFailure("remove diropq %pd", d);
++
++#undef src_or_dst_
++}
++
++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);
+}
+
+static void au_ren_rev_rename(int err, struct au_ren_args *a)
+ 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->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");
+
+ delegated = NULL;
+ rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path,
-+ &delegated);
++ &delegated, a->flags);
+ if (unlikely(rerr == -EWOULDBLOCK)) {
+ pr_warn("cannot retry for NFSv4 delegation"
+ " for an internal rename\n");
+ 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);
++ 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");
+}
+
+/* make it 'opaque' dir. */
-+static int au_ren_diropq(struct au_ren_args *a)
++static int au_ren_do_diropq(struct au_ren_args *a, int idx)
+{
+ int err;
-+ struct dentry *diropq;
++ struct dentry *d, *diropq;
++#define src_or_dst(member) a->sd[idx].member
+
+ 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);
++ 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;
+}
+
-+static int do_rename(struct au_ren_args *a)
++static int au_ren_diropq(struct au_ren_args *a)
+{
+ int err;
-+ struct dentry *d, *h_d;
++ unsigned char always;
++ struct dentry *d;
+
-+ /* 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))
++ 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;
-+ 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;
++ au_fset_ren(a->auren_flags, DIROPQ_SRC);
+ }
++ if (!a->exchange)
++ goto out; /* success */
+
-+ /* rename dentry to tmpwh */
-+ if (a->thargs) {
-+ err = au_whtmp_ren(a->dst_h_dentry, a->br);
++ 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_whdst;
++ goto out_rev_src;
++ au_fset_ren(a->auren_flags, DIROPQ_DST);
++ }
++ goto out; /* success */
+
-+ 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);
++out_rev_src:
++ AuDbg("err %d, reverting src\n", err);
++ au_ren_rev_diropq(err, a);
++out:
++ return err;
++}
++
++static int do_rename(struct au_ren_args *a)
++{
++ int err;
++ struct dentry *d, *h_d;
++
++ 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;
++ }
++
++ /* rename dentry to tmpwh */
++ if (a->thargs) {
++ err = au_whtmp_ren(a->dst_h_dentry, a->br);
++ if (unlikely(err))
++ goto out_whdst;
++
++ 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);
++ }
+ }
+
+ 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
+
+ /* 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;
+
+ /* make dir opaque */
-+ if (au_ftest_ren(a->flags, DIROPQ)) {
-+ err = au_ren_diropq(a);
-+ if (unlikely(err))
-+ goto out_rename;
-+ }
++ err = au_ren_diropq(a);
++ if (unlikely(err))
++ goto out_rename;
+
+ /* 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;
+
-+ /* 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;
-+ }
++ 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;
++ }
+
-+ /* remove whtmp */
-+ if (a->thargs)
-+ au_ren_del_whtmp(a); /* ignore this error */
++ /* remove whtmp */
++ if (a->thargs)
++ au_ren_del_whtmp(a); /* ignore this error */
+
-+ au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0);
++ au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0);
++ }
+ err = 0;
+ goto out_success;
+
+out_diropq:
-+ if (au_ftest_ren(a->flags, DIROPQ))
-+ au_ren_rev_diropq(err, a);
++ au_ren_rev_diropq(err, a);
+out_rename:
+ au_ren_rev_rename(err, a);
+ dput(a->h_dst);
+}
+
+/*
-+ * test if @dentry dir can be rename source or not.
-+ * if it can, return 0 and @children is filled.
++ * 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.
++ * on the lower branch unless DIRREN is on.
+ */
-+static int may_rename_srcdir(struct dentry *dentry, aufs_bindex_t btgt)
++static int may_rename_srcdir(struct au_ren_args *a)
+{
+ int err;
+ unsigned int rdhash;
-+ aufs_bindex_t btop;
++ aufs_bindex_t btop, btgt;
++ struct dentry *dentry;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++
++ 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(dentry->d_sb);
-+ rdhash = au_sbi(dentry->d_sb)->si_rdhash;
++ SiMustAnyLock(sb);
++ rdhash = sbinfo->si_rdhash;
+ if (!rdhash)
+ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL,
+ dentry));
+
+out:
+ if (err == -ENOTEMPTY) {
-+ AuWarn1("renaming dir who has child(ren) on multiple branches,"
-+ " is not supported\n");
-+ err = -EXDEV;
++ 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;
+}
+ SiMustAnyLock(d->d_sb);
+
+ err = 0;
-+ if (au_ftest_ren(a->flags, ISDIR) && a->dst_inode) {
++ 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));
+ if (unlikely(err))
+ goto out;
+
-+ au_set_dbtop(d, a->dst_btop);
-+ err = may_rename_dstdir(d, &a->whlist);
-+ au_set_dbtop(d, a->btgt);
++ 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))
+
+ 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 (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 (a->src_btop == a->btgt) {
+ err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent,
-+ au_ftest_ren(a->flags, ISDIR));
++ au_ftest_ren(a->auren_flags, ISDIR_SRC));
+ if (unlikely(err))
+ goto out;
+ err = -EINVAL;
+ goto out;
+
+ err = -EIO;
-+ isdir = !!au_ftest_ren(a->flags, ISDIR);
++ 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,
+{
+ 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))
++ 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));
+}
+
+ err = vfsub_mnt_want_write(au_br_mnt(a->br));
+ if (unlikely(err))
+ goto out;
-+ au_fset_ren(a->flags, MNT_WRITE);
++ au_fset_ren(a->auren_flags, MNT_WRITE);
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ struct dentry *root;
++ struct inode *dir;
++
++ /*
++ * 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);
++ }
++ }
+ 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);
+
+ dir = a->dst_dir;
+ dir->i_version++;
-+ if (au_ftest_ren(a->flags, ISDIR)) {
++ 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);
+ }
-+
+ au_dir_ts(dir, a->btgt);
+
-+ if (au_ftest_ren(a->flags, ISSAMEDIR))
++ 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);
++ }
++
++ if (au_ftest_ren(a->auren_flags, ISSAMEDIR))
+ return;
+
+ dir = a->src_dir;
+ dir->i_version++;
-+ if (au_ftest_ren(a->flags, ISDIR))
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC))
+ au_cpup_attr_nlink(dir, /*force*/1);
+ au_dir_ts(dir, a->btgt);
+}
+
+ 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);
++ 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_update_dbrange(d, /*do_put_zero*/0);
+ }
+
++ 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;
++ }
++
+ d = a->src_dentry;
+ au_set_dbwh(d, -1);
+ bbot = au_dbbot(d);
+
+ a->src_btop = au_dbtop(a->src_dentry);
+ a->dst_btop = au_dbtop(a->dst_dentry);
-+ if (au_ftest_ren(a->flags, ISDIR))
++ 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 = 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);
+
+ return err;
+}
+{
+ 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)) {
++ 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);
+ }
+
-+ au_fclr_ren(a->flags, DT_DSTDIR);
-+ if (!au_ftest_ren(a->flags, ISDIR))
++ au_fclr_ren(a->auren_flags, DT_DSTDIR);
++ if (!au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ && !a->exchange)
+ return;
+
+ 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);
++ 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);
+ }
+ struct inode *h_inode;
+
+ au_dtime_revert(a->src_dt + AuPARENT);
-+ if (!au_ftest_ren(a->flags, ISSAMEDIR))
++ if (!au_ftest_ren(a->auren_flags, ISSAMEDIR))
+ au_dtime_revert(a->dst_dt + AuPARENT);
+
-+ if (au_ftest_ren(a->flags, ISDIR) && err != -EIO) {
++ 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);
+
-+ if (au_ftest_ren(a->flags, DT_DSTDIR)) {
++ 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);
+/* ---------------------------------------------------------------------- */
+
+int aufs_rename(struct inode *_src_dir, struct dentry *_src_dentry,
-+ struct inode *_dst_dir, struct dentry *_dst_dentry)
++ struct inode *_dst_dir, struct dentry *_dst_dentry,
++ unsigned int _flags)
+{
-+ int err, flags;
++ int err, lock_flags;
++ void *rev;
+ /* reduce stack space */
+ struct au_ren_args *a;
++ struct au_pin pin;
+
-+ AuDbg("%pd, %pd\n", _src_dentry, _dst_dentry);
++ AuDbg("%pd, %pd, 0x%x\n", _src_dentry, _dst_dentry, _flags);
+ IMustLock(_src_dir);
+ IMustLock(_dst_dir);
+
++ err = -EINVAL;
++ if (unlikely(_flags & RENAME_WHITEOUT))
++ goto out;
++
+ err = -ENOMEM;
+ BUILD_BUG_ON(sizeof(*a) > PAGE_SIZE);
+ a = kzalloc(sizeof(*a), GFP_NOFS);
+ if (unlikely(!a))
+ goto out;
+
++ a->flags = _flags;
++ a->exchange = _flags & RENAME_EXCHANGE;
+ a->src_dir = _src_dir;
+ a->src_dentry = _src_dentry;
+ a->src_inode = NULL;
+ 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);
++ /*
++ * if EXCHANGE && src is non-dir && dst is dir,
++ * dst is not locked.
++ */
++ /* IMustLock(a->dst_inode); */
+ au_igrab(a->dst_inode);
+ }
+
+ err = -ENOTDIR;
-+ flags = AuLock_FLUSH | AuLock_NOPLM | AuLock_GEN;
++ lock_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)
++ 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;
-+ flags |= AuLock_DIRS;
++ lock_flags |= AuLock_DIRS;
++ }
++ 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;
+ }
-+ err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry, flags);
++ err = aufs_read_and_write_lock2(a->dst_dentry, a->src_dentry,
++ lock_flags);
+ if (unlikely(err))
+ goto out_free;
+
+ err = -ENOENT;
+ if (a->dst_inode) {
+ /*
-+ * If it is a dir, VFS unhash dst_dentry before this
++ * 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 (!S_ISDIR(a->dst_inode->i_mode)) {
++ if (!au_ftest_ren(a->auren_flags, ISDIR_DST)) {
+ err = au_d_hashed_positive(a->dst_dentry);
-+ if (unlikely(err))
++ if (unlikely(err && !a->exchange))
+ goto out_unlock;
+ } else if (unlikely(IS_DEADDIR(a->dst_inode)))
+ goto out_unlock;
+ if (unlikely(d_inode(a->dst_parent) == d_inode(a->src_dentry)))
+ goto out_unlock;
+
-+ au_fset_ren(a->flags, ISSAMEDIR); /* temporary */
++ au_fset_ren(a->auren_flags, ISSAMEDIR); /* temporary */
+ di_write_lock_parent(a->dst_parent);
+
+ /* which branch we process */
+
+ /* prepare the writable parent dir on the same branch */
+ if (a->dst_btop == a->btgt) {
-+ au_fset_ren(a->flags, WHDST);
++ au_fset_ren(a->auren_flags, WHDST);
+ } else {
+ err = au_cpup_dirs(a->dst_dentry, a->btgt);
+ if (unlikely(err))
+ goto out_children;
+ }
+
-+ 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);
++ 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->flags, WHSRC);
++ au_fset_ren(a->auren_flags, WHSRC);
+
+ /* cpup src */
+ if (a->src_btop != a->btgt) {
-+ struct au_pin pin;
-+
+ err = au_pin(&pin, a->src_dentry, a->btgt,
+ au_opt_udba(a->src_dentry->d_sb),
+ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
+ 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);
++ if (!a->exchange)
++ au_fset_ren(a->auren_flags, WHSRC);
++ }
++
++ /* 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);
+ }
+
+ /* lock them all */
+ /* leave the copied-up one */
+ goto out_children;
+
-+ 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;
++ 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;
++ }
+
+ /* store timestamps to be revertible */
+ au_ren_dt(a);
+
++ /* 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;
++ }
++
+ /* here we go */
+ err = do_rename(a);
+ if (unlikely(err))
-+ goto out_dt;
++ goto out_dirren;
++
++ if (au_ftest_ren(a->auren_flags, DIRREN))
++ au_dr_rename_fin(a->src_dentry, a->btgt, rev);
+
+ /* update dir attributes */
+ au_ren_refresh_dir(a);
+
+ 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_set_dbtop(a->dst_dentry, a->dst_btop);
+ }
+out_parent:
-+ if (!err)
-+ d_move(a->src_dentry, a->dst_dentry);
-+ else {
++ 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->flags, ISSAMEDIR))
++ if (au_ftest_ren(a->auren_flags, ISSAMEDIR))
+ di_write_unlock(a->dst_parent);
+ else
+ di_write_unlock2(a->src_parent, a->dst_parent);
+}
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 @@
++++ linux/fs/aufs/Kconfig 2017-11-12 22:24:44.697577553 +0100
+@@ -0,0 +1,198 @@
+config AUFS_FS
+ tristate "Aufs (Advanced multi layered unification filesystem) support"
+ help
+ shows better performance in most cases.
+ See detail in aufs.5.
+
++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
+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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,146 @@
++++ linux/fs/aufs/loop.c 2017-11-12 22:24:42.270843503 +0100
+@@ -0,0 +1,147 @@
+/*
-+ * 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;
+}
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
++++ linux/fs/aufs/loop.h 2017-07-29 12:14:25.903042072 +0200
@@ -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
+#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
++++ linux/fs/aufs/magic.mk 2017-07-29 12:14:25.903042072 +0200
@@ -0,0 +1,30 @@
+
+# defined in ${srctree}/fs/fuse/inode.c
+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 @@
++++ linux/fs/aufs/Makefile 2017-11-12 22:24:44.697577553 +0100
+@@ -0,0 +1,45 @@
+
+include ${src}/magic.mk
+ifeq (${CONFIG_AUFS_FS},m)
+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_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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,223 @@
++++ linux/fs/aufs/module.c 2017-11-12 22:24:44.707577830 +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 kmem_cache *au_cachep[AuCache_Last] = {
-+ [0] = NULL
-+};
++struct kmem_cache *au_cache[AuCache_Last];
+
+static void au_cache_fin(void)
+{
+ /* excluding AuCache_HNOTIFY */
+ BUILD_BUG_ON(AuCache_HNOTIFY + 1 != AuCache_Last);
+ for (i = 0; i < AuCache_HNOTIFY; i++) {
-+ kmem_cache_destroy(au_cachep[i]);
-+ au_cachep[i] = NULL;
++ kmem_cache_destroy(au_cache[i]);
++ au_cache[i] = NULL;
+ }
+}
+
+static int __init au_cache_init(void)
+{
-+ au_cachep[AuCache_DINFO] = AuCacheCtor(au_dinfo, au_di_init_once);
-+ if (au_cachep[AuCache_DINFO])
++ au_cache[AuCache_DINFO] = AuCacheCtor(au_dinfo, au_di_init_once);
++ if (au_cache[AuCache_DINFO])
+ /* SLAB_DESTROY_BY_RCU */
-+ au_cachep[AuCache_ICNTNR] = AuCacheCtor(au_icntnr,
-+ au_icntnr_init_once);
-+ if (au_cachep[AuCache_ICNTNR])
-+ au_cachep[AuCache_FINFO] = AuCacheCtor(au_finfo,
-+ au_fi_init_once);
-+ if (au_cachep[AuCache_FINFO])
-+ au_cachep[AuCache_VDIR] = AuCache(au_vdir);
-+ if (au_cachep[AuCache_VDIR])
-+ au_cachep[AuCache_DEHSTR] = AuCache(au_vdir_dehstr);
-+ if (au_cachep[AuCache_DEHSTR])
++ au_cache[AuCache_ICNTNR] = AuCacheCtor(au_icntnr,
++ au_icntnr_init_once);
++ if (au_cache[AuCache_ICNTNR])
++ au_cache[AuCache_FINFO] = AuCacheCtor(au_finfo,
++ au_fi_init_once);
++ 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;
+ for (i = 0; i < AuIop_Last; i++)
+ aufs_iop_nogetattr[i].getattr = NULL;
+
++ memset(au_cache, 0, sizeof(au_cache)); /* including hnotify */
++
+ au_sbilist_init();
+ sysaufs_brs_init();
+ au_debug_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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,89 @@
++++ linux/fs/aufs/module.h 2017-11-12 22:24:42.270843503 +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
+
+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
+ AuCache_Last
+};
+
++extern struct kmem_cache *au_cache[AuCache_Last];
++
+#define AuCacheFlags (SLAB_RECLAIM_ACCOUNT | SLAB_MEM_SPREAD)
+#define AuCache(type) KMEM_CACHE(type, AuCacheFlags)
+#define AuCacheCtor(type, ctor) \
+ kmem_cache_create(#type, sizeof(struct type), \
+ __alignof__(struct type), AuCacheFlags, ctor)
+
-+extern struct kmem_cache *au_cachep[];
-+
+#define AuCacheFuncs(name, index) \
+static inline struct au_##name *au_cache_alloc_##name(void) \
-+{ return kmem_cache_alloc(au_cachep[AuCache_##index], 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_cachep[AuCache_##index], p); }
++{ kmem_cache_free(au_cache[AuCache_##index], p); }
+
+AuCacheFuncs(dinfo, DINFO);
+AuCacheFuncs(icntnr, ICNTNR);
+#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 2016-07-25 19:05:34.814493242 +0200
++++ linux/fs/aufs/mvdown.c 2017-11-12 22:24:44.707577830 +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;
+}
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 2016-07-25 19:05:34.814493242 +0200
-@@ -0,0 +1,1859 @@
++++ linux/fs/aufs/opts.c 2017-11-12 22:24:44.707577830 +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;
+ 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;
+}
+
+
+ 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;
+}
+
+}
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 @@
++++ linux/fs/aufs/opts.h 2017-11-12 22:24:44.707577830 +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 /* __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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,502 @@
++++ linux/fs/aufs/plink.c 2017-11-12 22:24:44.707577830 +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 err;
+ pid_t pid, ppid;
++ struct task_struct *parent, *prev;
+ struct au_sbinfo *sbi;
+
+ SiMustAnyLock(sb);
+ goto out;
+
+ /* todo: it highly depends upon /sbin/mount.aufs */
++ prev = NULL;
++ parent = current;
++ ppid = 0;
+ rcu_read_lock();
-+ ppid = task_pid_vnr(rcu_dereference(current->real_parent));
++ while (1) {
++ parent = rcu_dereference(parent->real_parent);
++ if (parent == prev)
++ break;
++ ppid = task_pid_vnr(parent);
++ if (pid == ppid) {
++ rcu_read_unlock();
++ goto out;
++ }
++ prev = parent;
++ }
+ rcu_read_unlock();
-+ if (pid == ppid)
-+ goto out;
+
+ if (au_ftest_lock(flags, NOPLMW)) {
+ /* if there is no i_mutex lock in VFS, we don't need to wait */
+{
+ 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
++++ linux/fs/aufs/poll.c 2017-11-12 22:24:42.274176929 +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;
+}
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 @@
++++ linux/fs/aufs/posix_acl.c 2017-11-12 22:24:42.274176929 +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 @@
++++ linux/fs/aufs/procfs.c 2017-11-12 22:24:44.707577830 +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))
+}
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
++++ linux/fs/aufs/rdu.c 2017-11-12 22:24:42.274176929 +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
+#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 @@
++++ linux/fs/aufs/rwsem.h 2017-11-12 22:24:44.707577830 +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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,353 @@
++++ linux/fs/aufs/sbinfo.c 2017-11-12 22:24:44.707577830 +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));
-+ au_nwt_fin(&sbinfo->si_nowait);
++ AuDebugOn(!hlist_bl_empty(sbinfo->si_plink + i));
++ AuDebugOn(atomic_read(&sbinfo->si_nowait.nw_len));
+
+ AuDebugOn(percpu_counter_sum(&sbinfo->si_ninodes));
+ percpu_counter_destroy(&sbinfo->si_ninodes);
+ au_rw_write_unlock(&sbinfo->si_rwsem);
+
+ kfree(sbinfo->si_branch);
-+ for (i = 0; i < AU_NPIDMAP; i++)
-+ kfree(sbinfo->au_si_pid.pid_bitmap[i]);
-+ mutex_destroy(&sbinfo->au_si_pid.pid_mtx);
+ mutex_destroy(&sbinfo->si_xib_mtx);
+ AuRwDestroy(&sbinfo->si_rwsem);
+
+
+ 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 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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,111 @@
-+/*
-+ * 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__
-+
-+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);
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,1039 @@
++++ linux/fs/aufs/super.c 2017-11-12 22:24:44.707577830 +0100
+@@ -0,0 +1,1046 @@
+/*
-+ * 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);
+
-+ INIT_HLIST_HEAD(&inode->i_dentry);
+ au_cache_free_icntnr(container_of(inode, struct au_icntnr, vfs_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);
+
+ 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)) {
+ 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);
+};
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 @@
++++ linux/fs/aufs/super.h 2017-11-12 22:24:44.707577830 +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 /* __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
++++ linux/fs/aufs/sysaufs.c 2017-07-29 12:14:25.906375514 +0200
@@ -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
+}
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
++++ linux/fs/aufs/sysaufs.h 2017-07-29 12:14:25.906375514 +0200
@@ -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 /* __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 2016-07-25 19:05:34.817826663 +0200
++++ linux/fs/aufs/sysfs.c 2017-11-12 22:24:42.274176929 +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
+}
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 @@
++++ linux/fs/aufs/sysrq.c 2017-11-12 22:24:44.707577830 +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();
+}
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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,889 @@
++++ linux/fs/aufs/vdir.c 2017-11-12 22:24:42.274176929 +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
+ AuDebugOn(!nhash->nh_num || !nhash->nh_head);
+
+ v = 0;
++ if (len > 8)
++ len = 8;
+ while (len--)
+ v += *name++;
+ /* v = hash_long(v, magic_bit); */
+ 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;
+
+ 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;
+}
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 @@
++++ linux/fs/aufs/vfsub.c 2017-11-12 22:24:44.707577830 +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;
+}
+
+}
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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,310 @@
++++ linux/fs/aufs/vfsub.h 2017-11-12 22:24:44.707577830 +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);
+
+ return flags;
+}
+
++static inline int vfsub_file_execed(struct file *file)
++{
++ /* todo: direct access f_flags */
++ return !!(vfsub_file_flags(file) & __FMODE_EXEC);
++}
++
+#if 0 /* reserved */
+static inline void vfsub_file_accessed(struct file *h_file)
+{
+ 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 /* __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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,765 @@
++++ linux/fs/aufs/wbr_policy.c 2017-11-12 22:24:42.274176929 +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
+
+/* ---------------------------------------------------------------------- */
+
++/* 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,
+};
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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,1060 @@
++++ linux/fs/aufs/whout.c 2017-11-12 22:24:42.274176929 +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
+
+ /* 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"
+}
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
++++ linux/fs/aufs/whout.h 2017-09-05 10:42:11.058755349 +0200
@@ -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 /* __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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,218 @@
++++ linux/fs/aufs/wkq.c 2017-11-12 22:24:44.707577830 +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 {
+ };
+
+ 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;
-+
+}
+
+/*
+ int err;
+ struct au_wkinfo *wkinfo;
+
-+ percpu_counter_inc(&au_sbi(sb)->si_nowait.nw_len);
++ atomic_inc(&au_sbi(sb)->si_nowait.nw_len);
+
+ /*
+ * wkq_func() must free this wkinfo.
+ wkinfo->func = func;
+ wkinfo->args = args;
+ wkinfo->comp = NULL;
++ au_wkq_lockdep_init(wkinfo);
+ kobject_get(wkinfo->kobj);
+ __module_get(THIS_MODULE); /* todo: ?? */
+
+
+void au_nwt_init(struct au_nowait_tasks *nwt)
+{
-+ percpu_counter_init(&nwt->nw_len, 0, GFP_NOFS);
++ atomic_set(&nwt->nw_len, 0);
++ /* smp_mb(); */ /* atomic_set */
+ init_waitqueue_head(&nwt->nw_wq);
+}
+
-+void au_nwt_fin(struct au_nowait_tasks *nwt)
-+{
-+ AuDebugOn(percpu_counter_sum(&nwt->nw_len));
-+ percpu_counter_destroy(&nwt->nw_len);
-+}
-+
+void au_wkq_fin(void)
+{
+ destroy_workqueue(au_wkq);
+}
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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,95 @@
++++ linux/fs/aufs/wkq.h 2017-11-12 22:24:44.707577830 +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;
+
+ * in the next operation, wait for the 'nowait' tasks in system-wide workqueue
+ */
+struct au_nowait_tasks {
-+ struct percpu_counter nw_len;
++ atomic_t nw_len;
+ wait_queue_head_t nw_wq;
+};
+
+int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb,
+ unsigned int flags);
+void au_nwt_init(struct au_nowait_tasks *nwt);
-+void au_nwt_fin(struct au_nowait_tasks *nwt);
+int __init au_wkq_init(void);
+void au_wkq_fin(void);
+
+
+static inline void au_nwt_done(struct au_nowait_tasks *nwt)
+{
-+ percpu_counter_dec(&nwt->nw_len);
-+ if (!percpu_counter_sum(&nwt->nw_len))
++ if (atomic_dec_and_test(&nwt->nw_len))
+ wake_up_all(&nwt->nw_wq);
+}
+
+static inline int au_nwt_flush(struct au_nowait_tasks *nwt)
+{
-+ wait_event(nwt->nw_wq, !percpu_counter_sum(&nwt->nw_len));
++ wait_event(nwt->nw_wq, !atomic_read(&nwt->nw_len));
+ return 0;
+}
+
+#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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,345 @@
++++ linux/fs/aufs/xattr.c 2017-11-12 22:24:44.707577830 +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))
+ 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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,1317 @@
++++ linux/fs/aufs/xino.c 2017-11-12 22:24:44.710911257 +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;
+}
+ args = kmalloc(sizeof(*args), GFP_NOFS);
+ if (unlikely(!args)) {
+ AuErr1("no memory\n");
-+ goto out_args;
++ goto out;
+ }
+
+ au_br_get(br);
+
+ pr_err("wkq %d\n", wkq_err);
+ au_br_put(br);
-+
-+out_args:
+ kfree(args);
++
+out:
+ atomic_dec(&br->br_xino_running);
+}
+ if (sbinfo->si_xib)
+ fput(sbinfo->si_xib);
+ sbinfo->si_xib = NULL;
-+ free_page((unsigned long)sbinfo->si_xib_buf);
++ if (sbinfo->si_xib_buf)
++ free_page((unsigned long)sbinfo->si_xib_buf);
+ sbinfo->si_xib_buf = NULL;
+}
+
+ goto out; /* success */
+
+out_free:
-+ free_page((unsigned long)sbinfo->si_xib_buf);
++ if (sbinfo->si_xib_buf)
++ free_page((unsigned long)sbinfo->si_xib_buf);
+ sbinfo->si_xib_buf = NULL;
+ if (err >= 0)
+ err = -EIO;
+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 -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 2016-07-25 19:05:34.817826663 +0200
-@@ -0,0 +1,419 @@
++++ linux/include/uapi/linux/aufs_type.h 2017-11-12 22:24:44.710911257 +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.x-rcN-20160704"
++#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;
aufs4.x-rcN loopback patch
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
-index 7339e65..76e5da4 100644
+index d44de9d..095672b 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)
+@@ -595,6 +595,15 @@ static inline void loop_update_dio(struct loop_device *lo)
+ lo->use_dio);
}
- 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;
-
- w.file = file;
-+ w.virt_file = virt_file;
-
- /* 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)
- {
-- 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;
- }
-
++}
++
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,
+ struct block_device *bdev)
+ {
+@@ -629,6 +638,7 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
unsigned int arg)
{
struct file *file, *old_file;
struct inode *inode;
int error;
-@@ -676,9 +689,16 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+@@ -645,13 +655,20 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
file = fget(arg);
if (!file)
goto out;
+ get_file(file);
+ }
+ error = loop_validate_file(file, bdev);
+ if (error)
+ goto out_putf;
+
inode = file->f_mapping->host;
old_file = lo->lo_backing_file;
+ old_virt_file = lo->lo_backing_virt_file;
error = -EINVAL;
-@@ -690,17 +710,21 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
- goto out_putf;
-
- /* and ... switch */
-- error = loop_switch(lo, file);
-+ error = loop_switch(lo, file, virt_file);
- if (error)
- goto out_putf;
+@@ -662,6 +679,7 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ 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));
+@@ -669,12 +687,16 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ blk_mq_unfreeze_queue(lo->lo_queue);
fput(old_file);
+ if (old_virt_file)
out:
return error;
}
-@@ -887,7 +911,7 @@ static int loop_prepare_queue(struct loop_device *lo)
+@@ -868,7 +890,7 @@ 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;
+ 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,
+ int lo_flags = 0;
+@@ -882,6 +904,12 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
file = fget(arg);
if (!file)
goto out;
error = -EBUSY;
if (lo->lo_state != Lo_unbound)
-@@ -954,6 +984,7 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
+@@ -930,6 +958,7 @@ 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->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,
+@@ -963,6 +992,8 @@ static int loop_set_fd(struct loop_device *lo, fmode_t mode,
out_putf:
fput(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,
+@@ -1009,6 +1040,7 @@ 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;
gfp_t gfp = lo->old_gfp_mask;
struct block_device *bdev = lo->lo_device;
-@@ -1063,6 +1097,7 @@ static int loop_clr_fd(struct loop_device *lo)
+@@ -1040,6 +1072,7 @@ 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;
spin_unlock_irq(&lo->lo_lock);
loop_release_xfer(lo);
-@@ -1107,6 +1142,8 @@ static int loop_clr_fd(struct loop_device *lo)
+@@ -1087,6 +1120,8 @@ static int loop_clr_fd(struct loop_device *lo)
* bd_mutex which is usually taken before lo_ctl_mutex.
*/
fput(filp);
}
diff --git a/drivers/block/loop.h b/drivers/block/loop.h
-index fb2237c..c3888c5 100644
+index 1f39567..128b137 100644
--- a/drivers/block/loop.h
+++ b/drivers/block/loop.h
@@ -46,7 +46,7 @@ struct loop_device {
- 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 504b767..09426ad 100644
+index 8aff060..e853272 100644
--- a/fs/aufs/f_op.c
+++ b/fs/aufs/f_op.c
-@@ -346,7 +346,7 @@ static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
+@@ -357,7 +357,7 @@ static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
if (IS_ERR(h_file))
goto out;
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 5711e7a..9df5d16 100644
+index e35f015..b37f1ae 100644
--- a/fs/aufs/loop.c
+++ b/fs/aufs/loop.c
-@@ -131,3 +131,19 @@ void au_loopback_fin(void)
+@@ -132,3 +132,19 @@ void au_loopback_fin(void)
symbol_put(loop_backing_file);
kfree(au_warn_loopback_array);
}
+ return f;
+}
diff --git a/fs/aufs/loop.h b/fs/aufs/loop.h
-index 48bf070..66afec7 100644
+index e2df495..36e5052 100644
--- a/fs/aufs/loop.h
+++ b/fs/aufs/loop.h
@@ -25,7 +25,11 @@ void au_warn_loopback(struct super_block *h_sb);
#endif /* __KERNEL__ */
diff --git a/fs/aufs/super.c b/fs/aufs/super.c
-index 8bd2d9c..26581c0 100644
+index 3c300125..128d790 100644
--- a/fs/aufs/super.c
+++ b/fs/aufs/super.c
-@@ -832,7 +832,10 @@ static const struct super_operations aufs_sop = {
+@@ -838,7 +838,10 @@ static const struct super_operations aufs_sop = {
.statfs = aufs_statfs,
.put_super = aufs_put_super,
.sync_fs = aufs_sync_fs,
/* ---------------------------------------------------------------------- */
diff --git a/include/linux/fs.h b/include/linux/fs.h
-index 33c6953..f9b06a2 100644
+index 8ab6566..8dbaa52 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
-@@ -1805,6 +1805,10 @@ struct super_operations {
+@@ -1838,6 +1838,10 @@ struct super_operations {
struct shrink_control *);
long (*free_cached_objects)(struct super_block *,
struct shrink_control *);