-diff -urN linux/fs/aufs/aufs.h linux-aufs/fs/aufs/aufs.h
---- linux/fs/aufs/aufs.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/aufs.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,70 @@
+SPDX-License-Identifier: GPL-2.0
+aufs6.x-rcN kbuild patch
+
+diff --git a/fs/Kconfig b/fs/Kconfig
+index cc07a0cd3172..5ee4f1be48ff 100644
+--- a/fs/Kconfig
++++ b/fs/Kconfig
+@@ -316,6 +316,7 @@ source "fs/sysv/Kconfig"
+ source "fs/ufs/Kconfig"
+ source "fs/erofs/Kconfig"
+ source "fs/vboxsf/Kconfig"
++source "fs/aufs/Kconfig"
+
+ endif # MISC_FILESYSTEMS
+
+diff --git a/fs/Makefile b/fs/Makefile
+index 834f1c3dba46..b5c97dd026cd 100644
+--- a/fs/Makefile
++++ b/fs/Makefile
+@@ -137,3 +137,4 @@ obj-$(CONFIG_EFIVAR_FS) += efivarfs/
+ obj-$(CONFIG_EROFS_FS) += erofs/
+ obj-$(CONFIG_VBOXSF_FS) += vboxsf/
+ obj-$(CONFIG_ZONEFS_FS) += zonefs/
++obj-$(CONFIG_AUFS_FS) += aufs/
+SPDX-License-Identifier: GPL-2.0
+aufs6.x-rcN base patch
+
+diff --git a/MAINTAINERS b/MAINTAINERS
+index e0ad886d3163..a27659e1886b 100644
+--- a/MAINTAINERS
++++ b/MAINTAINERS
+@@ -3380,6 +3380,19 @@ F: include/uapi/linux/audit.h
+ F: kernel/audit*
+ F: lib/*audit.c
+
++AUFS (advanced multi layered unification filesystem) FILESYSTEM
++M: "J. R. Okajima" <hooanon05g@gmail.com>
++L: aufs-users@lists.sourceforge.net (members only)
++L: linux-unionfs@vger.kernel.org
++S: Supported
++W: http://aufs.sourceforge.net
++T: git://github.com/sfjro/aufs4-linux.git
++F: Documentation/ABI/testing/debugfs-aufs
++F: Documentation/ABI/testing/sysfs-aufs
++F: Documentation/filesystems/aufs/
++F: fs/aufs/
++F: include/uapi/linux/aufs_type.h
++
+ AUXILIARY BUS DRIVER
+ M: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
+ R: Dave Ertman <david.m.ertman@intel.com>
+diff --git a/drivers/block/loop.c b/drivers/block/loop.c
+index bc31bb7072a2..131294601819 100644
+--- a/drivers/block/loop.c
++++ b/drivers/block/loop.c
+@@ -645,6 +645,24 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ goto done;
+ }
+
++/*
++ * for AUFS
++ * no get/put for file.
++ */
++struct file *loop_backing_file(struct super_block *sb)
++{
++ struct file *ret;
++ struct loop_device *l;
++
++ ret = NULL;
++ if (MAJOR(sb->s_dev) == LOOP_MAJOR) {
++ l = sb->s_bdev->bd_disk->private_data;
++ ret = l->lo_backing_file;
++ }
++ return ret;
++}
++EXPORT_SYMBOL_GPL(loop_backing_file);
++
+ /* loop sysfs attributes */
+
+ static ssize_t loop_attr_show(struct device *dev, char *page,
+diff --git a/fs/dcache.c b/fs/dcache.c
+index 52e6d5fdab6b..519321f32f95 100644
+--- a/fs/dcache.c
++++ b/fs/dcache.c
+@@ -1345,7 +1345,7 @@ enum d_walk_ret {
+ *
+ * The @enter() callbacks are called with d_lock held.
+ */
+-static void d_walk(struct dentry *parent, void *data,
++void d_walk(struct dentry *parent, void *data,
+ enum d_walk_ret (*enter)(void *, struct dentry *))
+ {
+ struct dentry *this_parent;
+diff --git a/fs/fcntl.c b/fs/fcntl.c
+index b622be119706..9ea58b7bb580 100644
+--- a/fs/fcntl.c
++++ b/fs/fcntl.c
+@@ -34,7 +34,7 @@
+
+ #define SETFL_MASK (O_APPEND | O_NONBLOCK | O_NDELAY | O_DIRECT | O_NOATIME)
+
+-static int setfl(int fd, struct file * filp, unsigned long arg)
++int setfl(int fd, struct file *filp, unsigned long arg)
+ {
+ struct inode * inode = file_inode(filp);
+ int error = 0;
+@@ -64,6 +64,8 @@ static int setfl(int fd, struct file * filp, unsigned long arg)
+
+ if (filp->f_op->check_flags)
+ error = filp->f_op->check_flags(arg);
++ if (!error && filp->f_op->setfl)
++ error = filp->f_op->setfl(filp, arg);
+ if (error)
+ return error;
+
+diff --git a/fs/namespace.c b/fs/namespace.c
+index 54847db5b819..ebb7b4b057ea 100644
+--- a/fs/namespace.c
++++ b/fs/namespace.c
+@@ -849,6 +849,12 @@ 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/splice.c b/fs/splice.c
+index 3e06611d19ae..97a66df75d4c 100644
+--- a/fs/splice.c
++++ b/fs/splice.c
+@@ -865,8 +865,8 @@ static int warn_unsupported(struct file *file, const char *op)
+ /*
+ * Attempt to initiate a splice from pipe to file.
+ */
+-static long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
+- loff_t *ppos, size_t len, unsigned int flags)
++long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
++ loff_t *ppos, size_t len, unsigned int flags)
+ {
+ if (unlikely(!out->f_op->splice_write))
+ return warn_unsupported(out, "write");
+diff --git a/include/linux/fs.h b/include/linux/fs.h
+index 21a981680856..1e3054e5367d 100644
+--- a/include/linux/fs.h
++++ b/include/linux/fs.h
+@@ -1066,6 +1066,7 @@ 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 int f_setown(struct file *filp, unsigned long arg, int force);
+ extern void f_delown(struct file *filp);
+@@ -1793,6 +1794,7 @@ struct file_operations {
+ int (*lock) (struct file *, int, struct file_lock *);
+ unsigned long (*get_unmapped_area)(struct file *, unsigned long, unsigned long, unsigned long, unsigned long);
+ int (*check_flags)(int);
++ int (*setfl)(struct file *, unsigned long);
+ int (*flock) (struct file *, int, struct file_lock *);
+ ssize_t (*splice_write)(struct pipe_inode_info *, struct file *, loff_t *, size_t, unsigned int);
+ ssize_t (*splice_read)(struct file *, loff_t *, struct pipe_inode_info *, size_t, unsigned int);
+diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
+index b32256e9e944..16bd7230f44b 100644
+--- a/include/linux/lockdep.h
++++ b/include/linux/lockdep.h
+@@ -249,6 +249,8 @@ 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.
+ *
+diff --git a/include/linux/mnt_namespace.h b/include/linux/mnt_namespace.h
+index 8f882f5881e8..6b9808f09843 100644
+--- a/include/linux/mnt_namespace.h
++++ b/include/linux/mnt_namespace.h
+@@ -7,12 +7,15 @@ struct mnt_namespace;
+ struct fs_struct;
+ struct user_namespace;
+ struct ns_common;
++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 struct ns_common *from_mnt_ns(struct mnt_namespace *);
+
++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 a55179fd60fc..8e21c53cf883 100644
+--- a/include/linux/splice.h
++++ b/include/linux/splice.h
+@@ -93,4 +93,7 @@ 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);
+ #endif
+diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
+index dcd1d5bfc1e0..85d1f7e0bc07 100644
+--- a/kernel/locking/lockdep.c
++++ b/kernel/locking/lockdep.c
+@@ -218,7 +218,7 @@ unsigned long max_lock_class_idx;
+ struct lock_class lock_classes[MAX_LOCKDEP_KEYS];
+ DECLARE_BITMAP(lock_classes_in_use, 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)
+ {
+ unsigned int class_idx = hlock->class_idx;
+
+@@ -239,6 +239,7 @@ static inline struct lock_class *hlock_class(struct held_lock *hlock)
+ */
+ return lock_classes + class_idx;
+ }
++#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);
+SPDX-License-Identifier: GPL-2.0
+aufs6.x-rcN mmap patch
+
+diff --git a/fs/proc/base.c b/fs/proc/base.c
+index 05452c3b9872..6b92a8430130 100644
+--- a/fs/proc/base.c
++++ b/fs/proc/base.c
+@@ -2218,7 +2218,7 @@ static int map_files_get_link(struct dentry *dentry, struct path *path)
+ rc = -ENOENT;
+ vma = find_exact_vma(mm, vm_start, vm_end);
+ if (vma && vma->vm_file) {
+- *path = vma->vm_file->f_path;
++ *path = vma_pr_or_file(vma)->f_path;
+ path_get(path);
+ rc = 0;
+ }
+diff --git a/fs/proc/nommu.c b/fs/proc/nommu.c
+index 4d3493579458..42edd9a42c78 100644
+--- a/fs/proc/nommu.c
++++ b/fs/proc/nommu.c
+@@ -39,7 +39,10 @@ static int nommu_region_show(struct seq_file *m, struct vm_region *region)
+ file = region->vm_file;
+
+ if (file) {
+- struct inode *inode = file_inode(region->vm_file);
++ struct inode *inode;
++
++ file = vmr_pr_or_file(region);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ }
+diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
+index 420510f6a545..edfc0941806a 100644
+--- a/fs/proc/task_mmu.c
++++ b/fs/proc/task_mmu.c
+@@ -285,7 +285,10 @@ show_map_vma(struct seq_file *m, struct vm_area_struct *vma)
+ const char *name = NULL;
+
+ if (file) {
+- struct inode *inode = file_inode(vma->vm_file);
++ struct inode *inode;
++
++ file = vma_pr_or_file(vma);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ pgoff = ((loff_t)vma->vm_pgoff) << PAGE_SHIFT;
+@@ -1943,7 +1946,7 @@ static int show_numa_map(struct seq_file *m, void *v)
+ struct proc_maps_private *proc_priv = &numa_priv->proc_maps;
+ struct vm_area_struct *vma = v;
+ struct numa_maps *md = &numa_priv->md;
+- struct file *file = vma->vm_file;
++ struct file *file = vma_pr_or_file(vma);
+ struct mm_struct *mm = vma->vm_mm;
+ struct mempolicy *pol;
+ char buffer[64];
+diff --git a/fs/proc/task_nommu.c b/fs/proc/task_nommu.c
+index 0ec35072a8e5..f0f0a03eb2b5 100644
+--- a/fs/proc/task_nommu.c
++++ b/fs/proc/task_nommu.c
+@@ -150,7 +150,10 @@ static int nommu_vma_show(struct seq_file *m, struct vm_area_struct *vma)
+ file = vma->vm_file;
+
+ if (file) {
+- struct inode *inode = file_inode(vma->vm_file);
++ struct inode *inode;
++
++ file = vma_pr_or_file(vma);
++ inode = file_inode(file);
+ dev = inode->i_sb->s_dev;
+ ino = inode->i_ino;
+ pgoff = (loff_t)vma->vm_pgoff << PAGE_SHIFT;
+diff --git a/include/linux/mm.h b/include/linux/mm.h
+index 27ce77080c79..5f07b972242d 100644
+--- a/include/linux/mm.h
++++ b/include/linux/mm.h
+@@ -2353,6 +2353,43 @@ static inline void unmap_shared_mapping_range(struct address_space *mapping,
+ unmap_mapping_range(mapping, holebegin, holelen, 0);
+ }
+
++#if 1 /* IS_ENABLED(CONFIG_AUFS_FS) */
++extern void vma_do_file_update_time(struct vm_area_struct *, const char[], int);
++extern struct file *vma_do_pr_or_file(struct vm_area_struct *, const char[],
++ int);
++extern void vma_do_get_file(struct vm_area_struct *, const char[], int);
++extern void vma_do_fput(struct vm_area_struct *, const char[], int);
++
++#define vma_file_update_time(vma) vma_do_file_update_time(vma, __func__, \
++ __LINE__)
++#define vma_pr_or_file(vma) vma_do_pr_or_file(vma, __func__, \
++ __LINE__)
++#define vma_get_file(vma) vma_do_get_file(vma, __func__, __LINE__)
++#define vma_fput(vma) vma_do_fput(vma, __func__, __LINE__)
++
++#ifndef CONFIG_MMU
++extern struct file *vmr_do_pr_or_file(struct vm_region *, const char[], int);
++extern void vmr_do_fput(struct vm_region *, const char[], int);
++
++#define vmr_pr_or_file(region) vmr_do_pr_or_file(region, __func__, \
++ __LINE__)
++#define vmr_fput(region) vmr_do_fput(region, __func__, __LINE__)
++#endif /* !CONFIG_MMU */
++
++#else
++
++#define vma_file_update_time(vma) file_update_time((vma)->vm_file)
++#define vma_pr_or_file(vma) (vma)->vm_file
++#define vma_get_file(vma) get_file((vma)->vm_file)
++#define vma_fput(vma) fput((vma)->vm_file)
++
++#ifndef CONFIG_MMU
++#define vmr_pr_or_file(region) (region)->vm_file
++#define vmr_fput(region) fput((region)->vm_file)
++#endif /* !CONFIG_MMU */
++
++#endif /* CONFIG_AUFS_FS */
++
+ static inline struct vm_area_struct *vma_lookup(struct mm_struct *mm,
+ unsigned long addr);
+
+diff --git a/include/linux/mm_types.h b/include/linux/mm_types.h
+index 306a3d1a0fa6..56374f0eb825 100644
+--- a/include/linux/mm_types.h
++++ b/include/linux/mm_types.h
+@@ -449,6 +449,9 @@ 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 */
++#if 1 /* IS_ENABLED(CONFIG_AUFS_FS) */
++ struct file *vm_prfile; /* the virtual backing file or NULL */
++#endif
+
+ int vm_usage; /* region usage count (access under nommu_region_sem) */
+ bool vm_icache_flushed : 1; /* true if the icache has been flushed for
+@@ -548,6 +551,9 @@ 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). */
++#if 1 /* IS_ENABLED(CONFIG_AUFS_FS) */
++ struct file *vm_prfile; /* shadow of vm_file */
++#endif
+ void * vm_private_data; /* was vm_pte (shared mem) */
+
+ #ifdef CONFIG_ANON_VMA_NAME
+diff --git a/kernel/fork.c b/kernel/fork.c
+index ed4e01daccaa..e945a1639283 100644
+--- a/kernel/fork.c
++++ b/kernel/fork.c
+@@ -733,7 +733,7 @@ static __latent_entropy int dup_mmap(struct mm_struct *mm,
+ if (file) {
+ struct address_space *mapping = file->f_mapping;
+
+- get_file(file);
++ vma_get_file(tmp);
+ i_mmap_lock_write(mapping);
+ if (tmp->vm_flags & VM_SHARED)
+ mapping_allow_writable(mapping);
+diff --git a/mm/Makefile b/mm/Makefile
+index e29afc890cde..c467601299d4 100644
+--- a/mm/Makefile
++++ b/mm/Makefile
+@@ -137,3 +137,4 @@ obj-$(CONFIG_IO_MAPPING) += io-mapping.o
+ obj-$(CONFIG_HAVE_BOOTMEM_INFO_NODE) += bootmem_info.o
+ obj-$(CONFIG_GENERIC_IOREMAP) += ioremap.o
+ obj-$(CONFIG_SHRINKER_DEBUG) += shrinker_debug.o
++obj-y += prfile.o
+diff --git a/mm/filemap.c b/mm/filemap.c
+index b4c9bd368b7e..ec4f4e5c1a92 100644
+--- a/mm/filemap.c
++++ b/mm/filemap.c
+@@ -3564,7 +3564,7 @@ vm_fault_t filemap_page_mkwrite(struct vm_fault *vmf)
+ vm_fault_t ret = VM_FAULT_LOCKED;
+
+ sb_start_pagefault(mapping->host->i_sb);
+- file_update_time(vmf->vma->vm_file);
++ vma_file_update_time(vmf->vma);
+ folio_lock(folio);
+ if (folio->mapping != mapping) {
+ folio_unlock(folio);
+diff --git a/mm/mmap.c b/mm/mmap.c
+index 13678edaa22c..3b5dd8a8d627 100644
+--- a/mm/mmap.c
++++ b/mm/mmap.c
+@@ -140,7 +140,7 @@ static void remove_vma(struct vm_area_struct *vma, bool unreachable)
+ if (vma->vm_ops && vma->vm_ops->close)
+ vma->vm_ops->close(vma);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ mpol_put(vma_policy(vma));
+ if (unreachable)
+ __vm_area_free(vma);
+@@ -607,7 +607,7 @@ static inline void vma_complete(struct vma_prepare *vp,
+ if (vp->file) {
+ uprobe_munmap(vp->remove, vp->remove->vm_start,
+ vp->remove->vm_end);
+- fput(vp->file);
++ vma_fput(vp->vma);
+ }
+ if (vp->remove->anon_vma)
+ anon_vma_merge(vp->vma, vp->remove);
+@@ -2269,7 +2269,7 @@ int __split_vma(struct vma_iterator *vmi, struct vm_area_struct *vma,
+ goto out_free_mpol;
+
+ if (new->vm_file)
+- get_file(new->vm_file);
++ vma_get_file(new);
+
+ if (new->vm_ops && new->vm_ops->open)
+ new->vm_ops->open(new);
+@@ -2679,7 +2679,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
+ * and cause general protection fault
+ * ultimately.
+ */
+- fput(vma->vm_file);
++ vma_fput(vma);
+ vm_area_free(vma);
+ vma = merge;
+ /* Update vm_flags to pick up the change. */
+@@ -2774,7 +2774,7 @@ unsigned long mmap_region(struct file *file, unsigned long addr,
+
+ if (file || vma->vm_file) {
+ unmap_and_free_vma:
+- fput(vma->vm_file);
++ vma_fput(vma);
+ vma->vm_file = NULL;
+
+ /* Undo any partial mapping done by a device driver. */
+@@ -2843,6 +2843,9 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
+ unsigned long populate = 0;
+ unsigned long ret = -EINVAL;
+ struct file *file;
++#if 1 /* IS_ENABLED(CONFIG_AUFS_FS) */
++ struct file *prfile;
++#endif
+
+ pr_warn_once("%s (%d) uses deprecated remap_file_pages() syscall. See Documentation/mm/remap_file_pages.rst.\n",
+ current->comm, current->pid);
+@@ -2901,10 +2904,34 @@ SYSCALL_DEFINE5(remap_file_pages, unsigned long, start, unsigned long, size,
+ if (vma->vm_flags & VM_LOCKED)
+ flags |= MAP_LOCKED;
+
++#if 1 /* IS_ENABLED(CONFIG_AUFS_FS) */
++ vma_get_file(vma);
++ file = vma->vm_file;
++ prfile = vma->vm_prfile;
++ ret = do_mmap(vma->vm_file, start, size,
++ prot, flags, pgoff, &populate, NULL);
++ if (!IS_ERR_VALUE(ret) && file && prfile) {
++ struct vm_area_struct *new_vma;
++
++ new_vma = find_vma(mm, ret);
++ if (!new_vma->vm_prfile)
++ new_vma->vm_prfile = prfile;
++ if (prfile)
++ get_file(prfile);
++ }
++ /*
++ * two fput()s instead of vma_fput(vma),
++ * coz vma may not be available anymore.
++ */
++ fput(file);
++ if (prfile)
++ fput(prfile);
++#else
+ file = get_file(vma->vm_file);
+ ret = do_mmap(vma->vm_file, start, size,
+ prot, flags, pgoff, &populate, NULL);
+ fput(file);
++#endif /* CONFIG_AUFS_FS */
+ out:
+ mmap_write_unlock(mm);
+ if (populate)
+@@ -3250,7 +3277,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
+ if (anon_vma_clone(new_vma, vma))
+ goto out_free_mempol;
+ if (new_vma->vm_file)
+- get_file(new_vma->vm_file);
++ vma_get_file(new_vma);
+ if (new_vma->vm_ops && new_vma->vm_ops->open)
+ new_vma->vm_ops->open(new_vma);
+ vma_start_write(new_vma);
+@@ -3266,7 +3293,7 @@ struct vm_area_struct *copy_vma(struct vm_area_struct **vmap,
+ new_vma->vm_ops->close(new_vma);
+
+ if (new_vma->vm_file)
+- fput(new_vma->vm_file);
++ vma_fput(new_vma);
+
+ unlink_anon_vmas(new_vma);
+ out_free_mempol:
+diff --git a/mm/nommu.c b/mm/nommu.c
+index f670d9979a26..94d41bc45915 100644
+--- a/mm/nommu.c
++++ b/mm/nommu.c
+@@ -523,7 +523,7 @@ static void __put_nommu_region(struct vm_region *region)
+ up_write(&nommu_region_sem);
+
+ if (region->vm_file)
+- fput(region->vm_file);
++ vmr_fput(region);
+
+ /* IO memory and memory shared directly out of the pagecache
+ * from ramfs/tmpfs mustn't be released here */
+@@ -602,7 +602,7 @@ static void delete_vma(struct mm_struct *mm, struct vm_area_struct *vma)
+ if (vma->vm_ops && vma->vm_ops->close)
+ vma->vm_ops->close(vma);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ put_nommu_region(vma->vm_region);
+ vm_area_free(vma);
+ }
+@@ -1124,7 +1124,7 @@ unsigned long do_mmap(struct file *file,
+ goto error_just_free;
+ }
+ }
+- fput(region->vm_file);
++ vmr_fput(region);
+ kmem_cache_free(vm_region_jar, region);
+ region = pregion;
+ result = start;
+@@ -1206,10 +1206,10 @@ unsigned long do_mmap(struct file *file,
+ error:
+ vma_iter_free(&vmi);
+ if (region->vm_file)
+- fput(region->vm_file);
++ vmr_fput(region);
+ kmem_cache_free(vm_region_jar, region);
+ if (vma->vm_file)
+- fput(vma->vm_file);
++ vma_fput(vma);
+ vm_area_free(vma);
+ return ret;
+
+diff --git a/mm/prfile.c b/mm/prfile.c
+new file mode 100644
+index 000000000000..8f820a235364
+--- /dev/null
++++ b/mm/prfile.c
+@@ -0,0 +1,86 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * 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) 2014-2022 Junjro R. Okajima
++ * Copyright (c) 2014 Ian Campbell
++ */
++
++#include <linux/mm.h>
++#include <linux/file.h>
++#include <linux/fs.h>
++
++/* #define PRFILE_TRACE */
++static inline void prfile_trace(struct file *f, struct file *pr,
++ const char func[], int line, const char func2[])
++{
++#ifdef PRFILE_TRACE
++ if (pr)
++ pr_info("%s:%d: %s, %pD2\n", func, line, func2, f);
++#endif
++}
++
++void vma_do_file_update_time(struct vm_area_struct *vma, const char func[],
++ int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ file_update_time(f);
++ if (f && pr)
++ file_update_time(pr);
++}
++
++struct file *vma_do_pr_or_file(struct vm_area_struct *vma, const char func[],
++ int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ return (f && pr) ? pr : f;
++}
++
++void vma_do_get_file(struct vm_area_struct *vma, const char func[], int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ get_file(f);
++ if (f && pr)
++ get_file(pr);
++}
++
++void vma_do_fput(struct vm_area_struct *vma, const char func[], int line)
++{
++ struct file *f = vma->vm_file, *pr = vma->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ fput(f);
++ if (f && pr)
++ fput(pr);
++}
++
++#ifndef CONFIG_MMU
++struct file *vmr_do_pr_or_file(struct vm_region *region, const char func[],
++ int line)
++{
++ struct file *f = region->vm_file, *pr = region->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ return (f && pr) ? pr : f;
++}
++
++void vmr_do_fput(struct vm_region *region, const char func[], int line)
++{
++ struct file *f = region->vm_file, *pr = region->vm_prfile;
++
++ prfile_trace(f, pr, func, line, __func__);
++ fput(f);
++ if (f && pr)
++ fput(pr);
++}
++#endif /* !CONFIG_MMU */
+SPDX-License-Identifier: GPL-2.0
+aufs6.x-rcN standalone patch
+
+diff --git a/fs/dcache.c b/fs/dcache.c
+index 519321f32f95..267e0c65914a 100644
+--- a/fs/dcache.c
++++ b/fs/dcache.c
+@@ -1450,6 +1450,7 @@ void d_walk(struct dentry *parent, void *data,
+ seq = 1;
+ goto again;
+ }
++EXPORT_SYMBOL_GPL(d_walk);
+
+ struct check_mount {
+ struct vfsmount *mnt;
+@@ -3052,6 +3053,7 @@ 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 a466e797c8e2..c60353b5214f 100644
+--- a/fs/exec.c
++++ b/fs/exec.c
+@@ -112,6 +112,7 @@ bool path_noexec(const struct path *path)
+ return (path->mnt->mnt_flags & MNT_NOEXEC) ||
+ (path->mnt->mnt_sb->s_iflags & SB_I_NOEXEC);
+ }
++EXPORT_SYMBOL_GPL(path_noexec);
+
+ #ifdef CONFIG_USELIB
+ /*
+diff --git a/fs/fcntl.c b/fs/fcntl.c
+index 9ea58b7bb580..99fef189bcd6 100644
+--- a/fs/fcntl.c
++++ b/fs/fcntl.c
+@@ -87,6 +87,7 @@ int setfl(int fd, struct file *filp, unsigned long arg)
+ out:
+ return error;
+ }
++EXPORT_SYMBOL_GPL(setfl);
+
+ static void f_modown(struct file *filp, struct pid *pid, enum pid_type type,
+ int force)
+diff --git a/fs/file_table.c b/fs/file_table.c
+index 372653b92617..69128fe39b6d 100644
+--- a/fs/file_table.c
++++ b/fs/file_table.c
+@@ -199,6 +199,7 @@ struct file *alloc_empty_file(int flags, const struct cred *cred)
+ }
+ return ERR_PTR(-ENFILE);
+ }
++EXPORT_SYMBOL_GPL(alloc_empty_file);
+
+ /*
+ * Variant of alloc_empty_file() that doesn't check and modify nr_files.
+diff --git a/fs/namespace.c b/fs/namespace.c
+index ebb7b4b057ea..6ee47677de2f 100644
+--- a/fs/namespace.c
++++ b/fs/namespace.c
+@@ -457,6 +457,7 @@ void __mnt_drop_write(struct vfsmount *mnt)
+ mnt_dec_writers(real_mount(mnt));
+ preempt_enable();
+ }
++EXPORT_SYMBOL_GPL(__mnt_drop_write);
+
+ /**
+ * mnt_drop_write - give up write access to a mount
+@@ -854,6 +855,7 @@ 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
+@@ -2062,6 +2064,7 @@ int iterate_mounts(int (*f)(struct vfsmount *, void *), void *arg,
+ }
+ return 0;
+ }
++EXPORT_SYMBOL_GPL(iterate_mounts);
+
+ static void lock_mnt_tree(struct mount *mnt)
+ {
+diff --git a/fs/notify/group.c b/fs/notify/group.c
+index 1de6631a3925..3008eb37a18d 100644
+--- a/fs/notify/group.c
++++ b/fs/notify/group.c
+@@ -100,6 +100,7 @@ void fsnotify_get_group(struct fsnotify_group *group)
+ {
+ refcount_inc(&group->refcnt);
+ }
++EXPORT_SYMBOL_GPL(fsnotify_get_group);
+
+ /*
+ * Drop a reference to a group. Free it if it's through.
+diff --git a/fs/open.c b/fs/open.c
+index 4478adcc4f3a..5a984ff90580 100644
+--- a/fs/open.c
++++ b/fs/open.c
+@@ -67,6 +67,7 @@ int do_truncate(struct mnt_idmap *idmap, struct dentry *dentry,
+ inode_unlock(dentry->d_inode);
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(do_truncate);
+
+ long vfs_truncate(const struct path *path, loff_t length)
+ {
+diff --git a/fs/read_write.c b/fs/read_write.c
+index a21ba3be7dbe..ec40061176b3 100644
+--- a/fs/read_write.c
++++ b/fs/read_write.c
+@@ -477,6 +477,7 @@ ssize_t vfs_read(struct file *file, char __user *buf, size_t count, loff_t *pos)
+ inc_syscr(current);
+ 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)
+ {
+@@ -592,6 +593,7 @@ ssize_t vfs_write(struct file *file, const char __user *buf, size_t count, loff_
+ file_end_write(file);
+ return ret;
+ }
++EXPORT_SYMBOL_GPL(vfs_write);
+
+ /* file_ppos returns &file->f_pos or NULL if file is stream */
+ static inline loff_t *file_ppos(struct file *file)
+diff --git a/fs/splice.c b/fs/splice.c
+index 97a66df75d4c..19c8d35f5ebd 100644
+--- a/fs/splice.c
++++ b/fs/splice.c
+@@ -872,6 +872,7 @@ long do_splice_from(struct pipe_inode_info *pipe, struct file *out,
+ return warn_unsupported(out, "write");
+ return out->f_op->splice_write(pipe, out, ppos, len, flags);
+ }
++EXPORT_SYMBOL_GPL(do_splice_from);
+
+ /*
+ * Indicate to the caller that there was a premature EOF when reading from the
+diff --git a/fs/xattr.c b/fs/xattr.c
+index fcf67d80d7f9..174dfb9285fb 100644
+--- a/fs/xattr.c
++++ b/fs/xattr.c
+@@ -406,6 +406,7 @@ vfs_getxattr_alloc(struct mnt_idmap *idmap, struct dentry *dentry,
+ *xattr_value = value;
+ return error;
+ }
++EXPORT_SYMBOL_GPL(vfs_getxattr_alloc);
+
+ ssize_t
+ __vfs_getxattr(struct dentry *dentry, struct inode *inode, const char *name,
+diff --git a/kernel/locking/lockdep.c b/kernel/locking/lockdep.c
+index 85d1f7e0bc07..5e20504eed24 100644
+--- a/kernel/locking/lockdep.c
++++ b/kernel/locking/lockdep.c
+@@ -239,6 +239,7 @@ inline struct lock_class *lockdep_hlock_class(struct held_lock *hlock)
+ */
+ return lock_classes + class_idx;
+ }
++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 065e1ef8fc8d..c623c6f0c645 100644
+--- a/kernel/task_work.c
++++ b/kernel/task_work.c
+@@ -182,3 +182,4 @@ void task_work_run(void)
+ } while (work);
+ }
+ }
++EXPORT_SYMBOL_GPL(task_work_run);
+diff --git a/security/security.c b/security/security.c
+index d5ff7ff45b77..5cca1f5774ef 100644
+--- a/security/security.c
++++ b/security/security.c
+@@ -1711,6 +1711,7 @@ int security_path_rmdir(const struct path *dir, struct dentry *dentry)
+ return 0;
+ return call_int_hook(path_rmdir, 0, dir, dentry);
+ }
++EXPORT_SYMBOL_GPL(security_path_rmdir);
+
+ /**
+ * security_path_unlink() - Check if removing a hard link is allowed
+@@ -1746,6 +1747,7 @@ int security_path_symlink(const struct path *dir, struct dentry *dentry,
+ return 0;
+ return call_int_hook(path_symlink, 0, dir, dentry, old_name);
+ }
++EXPORT_SYMBOL_GPL(security_path_symlink);
+
+ /**
+ * security_path_link - Check if creating a hard link is allowed
+@@ -1764,6 +1766,7 @@ int security_path_link(struct dentry *old_dentry, const struct path *new_dir,
+ return 0;
+ return call_int_hook(path_link, 0, old_dentry, new_dir, new_dentry);
+ }
++EXPORT_SYMBOL_GPL(security_path_link);
+
+ /**
+ * security_path_rename() - Check if renaming a file is allowed
+@@ -1825,6 +1828,7 @@ int security_path_chmod(const struct path *path, umode_t mode)
+ return 0;
+ return call_int_hook(path_chmod, 0, path, mode);
+ }
++EXPORT_SYMBOL_GPL(security_path_chmod);
+
+ /**
+ * security_path_chown() - Check if changing the file's owner/group is allowed
+@@ -1842,6 +1846,7 @@ int security_path_chown(const struct path *path, kuid_t uid, kgid_t gid)
+ return 0;
+ return call_int_hook(path_chown, 0, path, uid, gid);
+ }
++EXPORT_SYMBOL_GPL(security_path_chown);
+
+ /**
+ * security_path_chroot() - Check if changing the root directory is allowed
+@@ -2071,6 +2076,7 @@ int security_inode_permission(struct inode *inode, int mask)
+ return 0;
+ return call_int_hook(inode_permission, 0, inode, mask);
+ }
++EXPORT_SYMBOL_GPL(security_inode_permission);
+
+ /**
+ * security_inode_setattr() - Check if setting file attributes is allowed
+@@ -2549,6 +2555,7 @@ int security_file_permission(struct file *file, int mask)
+
+ return fsnotify_perm(file, mask);
+ }
++EXPORT_SYMBOL_GPL(security_file_permission);
+
+ /**
+ * security_file_alloc() - Allocate and init a file's LSM blob
+@@ -2815,6 +2822,7 @@ int security_file_truncate(struct file *file)
+ {
+ return call_int_hook(file_truncate, 0, file);
+ }
++EXPORT_SYMBOL_GPL(security_file_truncate);
+
+ /**
+ * security_task_alloc() - Allocate a task's LSM blob
+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 2022-11-05 23:02:18.955889283 +0100
+@@ -0,0 +1,55 @@
++What: /debug/aufs/si_<id>/
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ Under /debug/aufs, a directory named si_<id> is created
++ per aufs mount, where <id> is a unique id generated
++ internally.
++
++What: /debug/aufs/si_<id>/plink
++Date: Apr 2013
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It has three lines and shows the information about the
++ pseudo-link. The first line is a single number
++ representing a number of buckets. The second line is a
++ number of pseudo-links per buckets (separated by a
++ blank). The last line is a single number representing a
++ total number of psedo-links.
++ When the aufs mount option 'noplink' is specified, it
++ will show "1\n0\n0\n".
++
++What: /debug/aufs/si_<id>/xib
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the consumed blocks by xib (External Inode Number
++ Bitmap), its block size and file size.
++ When the aufs mount option 'noxino' is specified, it
++ will be empty. About XINO files, see the aufs manual.
++
++What: /debug/aufs/si_<id>/xi<branch-index>
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the consumed blocks by xino (External Inode Number
++ Translation Table), its link count, block size and file
++ size.
++ Due to the file size limit, there may exist multiple
++ xino files per branch. In this case, "-N" is added to
++ the filename and it corresponds to the index of the
++ internal xino array. "-0" is omitted.
++ When the aufs mount option 'noxino' is specified, Those
++ entries won't exist. About XINO files, see the aufs
++ manual.
++
++What: /debug/aufs/si_<id>/xigen
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the consumed blocks by xigen (External Inode
++ Generation Table), its block size and file size.
++ If CONFIG_AUFS_EXPORT is disabled, this entry will not
++ be created.
++ When the aufs mount option 'noxino' is specified, it
++ will be empty. About XINO files, see the aufs manual.
+diff -urN /usr/share/empty/Documentation/ABI/testing/sysfs-aufs linux/Documentation/ABI/testing/sysfs-aufs
+--- /usr/share/empty/Documentation/ABI/testing/sysfs-aufs 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/ABI/testing/sysfs-aufs 2022-11-05 23:02:18.955889283 +0100
+@@ -0,0 +1,31 @@
++What: /sys/fs/aufs/si_<id>/
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ Under /sys/fs/aufs, a directory named si_<id> is created
++ per aufs mount, where <id> is a unique id generated
++ internally.
++
++What: /sys/fs/aufs/si_<id>/br<idx>
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the abolute path of a member directory (which
++ is called branch) in aufs, and its permission.
++
++What: /sys/fs/aufs/si_<id>/brid<idx>
++Date: July 2013
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the id of a member directory (which is called
++ branch) in aufs.
++
++What: /sys/fs/aufs/si_<id>/xi_path
++Date: March 2009
++Contact: J. R. Okajima <hooanon05g@gmail.com>
++Description:
++ It shows the abolute path of XINO (External Inode Number
++ Bitmap, Translation Table and Generation Table) file
++ even if it is the default path.
++ When the aufs mount option 'noxino' is specified, it
++ will be empty. About XINO files, see the aufs manual.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/01intro.txt linux/Documentation/filesystems/aufs/design/01intro.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/01intro.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/01intro.txt 2022-11-05 23:02:18.955889283 +0100
+@@ -0,0 +1,171 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Introduction
++----------------------------------------
++
++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
++ is called as a branch.
++- branch permission and attribute, 'readonly', 'real-readonly',
++ 'readwrite', 'whiteout-able', 'link-able whiteout', etc. and their
++ combination.
++- internal "file copy-on-write".
++- logical deletion, whiteout.
++- dynamic branch manipulation, adding, deleting and changing permission.
++- allow bypassing aufs, user's direct branch access.
++- external inode number translation table and bitmap which maintains the
++ persistent aufs inode number.
++- seekable directory, including NFS readdir.
++- file mapping, mmap and sharing pages.
++- pseudo-link, hardlink over branches.
++- loopback mounted filesystem as a branch.
++- several policies to select one among multiple writable branches.
++- revert a single systemcall when an error occurs in aufs.
++- and more...
++
++
++Multi Layered Stackable Unification Filesystem
++----------------------------------------------------------------------
++Most people already knows what it is.
++It is a filesystem which unifies several directories and provides a
++merged single directory. When users access a file, the access will be
++passed/re-directed/converted (sorry, I am not sure which English word is
++correct) to the real file on the member filesystem. The member
++filesystem is called 'lower filesystem' or 'branch' and has a mode
++'readonly' and 'readwrite.' And the deletion for a file on the lower
++readonly branch is handled by creating 'whiteout' on the upper writable
++branch.
++
++On LKML, there have been discussions about UnionMount (Jan Blunck,
++Bharata B Rao and Valerie Aurora) and Unionfs (Erez Zadok). They took
++different approaches to implement the merged-view.
++The former tries putting it into VFS, and the latter implements as a
++separate filesystem.
++(If I misunderstand about these implementations, please let me know and
++I shall correct it. Because it is a long time ago when I read their
++source files last time).
++
++UnionMount's approach will be able to small, but may be hard to share
++branches between several UnionMount since the whiteout in it is
++implemented in the inode on branch filesystem and always
++shared. According to Bharata's post, readdir does not seems to be
++finished yet.
++There are several missing features known in this implementations such as
++- for users, the inode number may change silently. eg. copy-up.
++- link(2) may break by copy-up.
++- read(2) may get an obsoleted filedata (fstat(2) too).
++- fcntl(F_SETLK) may be broken by copy-up.
++- unnecessary copy-up may happen, for example mmap(MAP_PRIVATE) after
++ open(O_RDWR).
++
++In linux-3.18, "overlay" filesystem (formerly known as "overlayfs") was
++merged into mainline. This is another implementation of UnionMount as a
++separated filesystem. All the limitations and known problems which
++UnionMount are equally inherited to "overlay" filesystem.
++
++Unionfs has a longer history. When I started implementing a stackable
++filesystem (Aug 2005), it already existed. It has virtual super_block,
++inode, dentry and file objects and they have an array pointing lower
++same kind objects. After contributing many patches for Unionfs, I
++re-started my project AUFS (Jun 2006).
++
++In AUFS, the structure of filesystem resembles to Unionfs, but I
++implemented my own ideas, approaches and enhancements and it became
++totally different one.
++
++Comparing DM snapshot and fs based implementation
++- the number of bytes to be copied between devices is much smaller.
++- the type of filesystem must be one and only.
++- the fs must be writable, no readonly fs, even for the lower original
++ device. so the compression fs will not be usable. but if we use
++ loopback mount, we may address this issue.
++ for instance,
++ mount /cdrom/squashfs.img /sq
++ losetup /sq/ext2.img
++ losetup /somewhere/cow
++ dmsetup "snapshot /dev/loop0 /dev/loop1 ..."
++- it will be difficult (or needs more operations) to extract the
++ difference between the original device and COW.
++- DM snapshot-merge may help a lot when users try merging. in the
++ fs-layer union, users will use rsync(1).
++
++You may want to read my old paper "Filesystems in LiveCD"
++(http://aufs.sourceforge.net/aufs2/report/sq/sq.pdf).
++
++
++Several characters/aspects/persona of aufs
++----------------------------------------------------------------------
++
++Aufs has several characters, aspects or persona.
++1. a filesystem, callee of VFS helper
++2. sub-VFS, caller of VFS helper for branches
++3. a virtual filesystem which maintains persistent inode number
++4. reader/writer of files on branches such like an application
++
++1. Callee of VFS Helper
++As an ordinary linux filesystem, aufs is a callee of VFS. For instance,
++unlink(2) from an application reaches sys_unlink() kernel function and
++then vfs_unlink() is called. vfs_unlink() is one of VFS helper and it
++calls filesystem specific unlink operation. Actually aufs implements the
++unlink operation but it behaves like a redirector.
++
++2. Caller of VFS Helper for Branches
++aufs_unlink() passes the unlink request to the branch filesystem as if
++it were called from VFS. So the called unlink operation of the branch
++filesystem acts as usual. As a caller of VFS helper, aufs should handle
++every necessary pre/post operation for the branch filesystem.
++- acquire the lock for the parent dir on a branch
++- lookup in a branch
++- revalidate dentry on a branch
++- mnt_want_write() for a branch
++- vfs_unlink() for a branch
++- mnt_drop_write() for a branch
++- release the lock on a branch
++
++3. Persistent Inode Number
++One of the most important issue for a filesystem is to maintain inode
++numbers. This is particularly important to support exporting a
++filesystem via NFS. Aufs is a virtual filesystem which doesn't have a
++backend block device for its own. But some storage is necessary to
++keep and maintain the inode numbers. It may be a large space and may not
++suit to keep in memory. Aufs rents some space from its first writable
++branch filesystem (by default) and creates file(s) on it. These files
++are created by aufs internally and removed soon (currently) keeping
++opened.
++Note: Because these files are removed, they are totally gone after
++ unmounting aufs. It means the inode numbers are not persistent
++ across unmount or reboot. I have a plan to make them really
++ persistent which will be important for aufs on NFS server.
++
++4. Read/Write Files Internally (copy-on-write)
++Because a branch can be readonly, when you write a file on it, aufs will
++"copy-up" it to the upper writable branch internally. And then write the
++originally requested thing to the file. Generally kernel doesn't
++open/read/write file actively. In aufs, even a single write may cause a
++internal "file copy". This behaviour is very similar to cp(1) command.
++
++Some people may think it is better to pass such work to user space
++helper, instead of doing in kernel space. Actually I am still thinking
++about it. But currently I have implemented it in kernel space.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/02struct.txt linux/Documentation/filesystems/aufs/design/02struct.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/02struct.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/02struct.txt 2022-11-05 23:02:18.955889283 +0100
+@@ -0,0 +1,258 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Basic Aufs Internal Structure
++
++Superblock/Inode/Dentry/File Objects
++----------------------------------------------------------------------
++As like an ordinary filesystem, aufs has its own
++superblock/inode/dentry/file objects. All these objects have a
++dynamically allocated array and store the same kind of pointers to the
++lower filesystem, branch.
++For example, when you build a union with one readwrite branch and one
++readonly, mounted /au, /rw and /ro respectively.
++- /au = /rw + /ro
++- /ro/fileA exists but /rw/fileA
++
++Aufs lookup operation finds /ro/fileA and gets dentry for that. These
++pointers are stored in a aufs dentry. The array in aufs dentry will be,
++- [0] = NULL (because /rw/fileA doesn't exist)
++- [1] = /ro/fileA
++
++This style of an array is essentially same to the aufs
++superblock/inode/dentry/file objects.
++
++Because aufs supports manipulating branches, ie. add/delete/change
++branches dynamically, these objects has its own generation. When
++branches are changed, the generation in aufs superblock is
++incremented. And a generation in other object are compared when it is
++accessed. When a generation in other objects are obsoleted, aufs
++refreshes the internal array.
++
++
++Superblock
++----------------------------------------------------------------------
++Additionally aufs superblock has some data for policies to select one
++among multiple writable branches, XIB files, pseudo-links and kobject.
++See below in detail.
++About the policies which supports copy-down a directory, see
++wbr_policy.txt too.
++
++
++Branch and XINO(External Inode Number Translation Table)
++----------------------------------------------------------------------
++Every branch has its own xino (external inode number translation table)
++file. The xino file is created and unlinked by aufs internally. When two
++members of a union exist on the same filesystem, they share the single
++xino file.
++The struct of a xino file is simple, just a sequence of aufs inode
++numbers which is indexed by the lower inode number.
++In the above sample, assume the inode number of /ro/fileA is i111 and
++aufs assigns the inode number i999 for fileA. Then aufs writes 999 as
++4(8) bytes at 111 * 4(8) bytes offset in the xino file.
++
++When the inode numbers are not contiguous, the xino file will be sparse
++which has a hole in it and doesn't consume as much disk space as it
++might appear. If your branch filesystem consumes disk space for such
++holes, then you should specify 'xino=' option at mounting aufs.
++
++Aufs has a mount option to free the disk blocks for such holes in XINO
++files on tmpfs or ramdisk. But it is not so effective actually. If you
++meet a problem of disk shortage due to XINO files, then you should try
++"tmpfs-ino.patch" (and "vfs-ino.patch" too) in aufs4-standalone.git.
++The patch localizes the assignment inumbers per tmpfs-mount and avoid
++the holes in XINO files.
++
++Also a writable branch has three kinds of "whiteout bases". All these
++are existed when the branch is joined to aufs, and their names are
++whiteout-ed doubly, so that users will never see their names in aufs
++hierarchy.
++1. a regular file which will be hardlinked to all whiteouts.
++2. a directory to store a pseudo-link.
++3. a directory to store an "orphan"-ed file temporary.
++
++1. Whiteout Base
++ When you remove a file on a readonly branch, aufs handles it as a
++ logical deletion and creates a whiteout on the upper writable branch
++ as a hardlink of this file in order not to consume inode on the
++ writable branch.
++2. Pseudo-link Dir
++ See below, Pseudo-link.
++3. Step-Parent Dir
++ When "fileC" exists on the lower readonly branch only and it is
++ opened and removed with its parent dir, and then user writes
++ something into it, then aufs copies-up fileC to this
++ directory. Because there is no other dir to store fileC. After
++ creating a file under this dir, the file is unlinked.
++
++Because aufs supports manipulating branches, ie. add/delete/change
++dynamically, a branch has its own id. When the branch order changes,
++aufs finds the new index by searching the branch id.
++
++
++Pseudo-link
++----------------------------------------------------------------------
++Assume "fileA" exists on the lower readonly branch only and it is
++hardlinked to "fileB" on the branch. When you write something to fileA,
++aufs copies-up it to the upper writable branch. Additionally aufs
++creates a hardlink under the Pseudo-link Directory of the writable
++branch. The inode of a pseudo-link is kept in aufs super_block as a
++simple list. If fileB is read after unlinking fileA, aufs returns
++filedata from the pseudo-link instead of the lower readonly
++branch. Because the pseudo-link is based upon the inode, to keep the
++inode number by xino (see above) is essentially necessary.
++
++All the hardlinks under the Pseudo-link Directory of the writable branch
++should be restored in a proper location later. Aufs provides a utility
++to do this. The userspace helpers executed at remounting and unmounting
++aufs by default.
++During this utility is running, it puts aufs into the pseudo-link
++maintenance mode. In this mode, only the process which began the
++maintenance mode (and its child processes) is allowed to operate in
++aufs. Some other processes which are not related to the pseudo-link will
++be allowed to run too, but the rest have to return an error or wait
++until the maintenance mode ends. If a process already acquires an inode
++mutex (in VFS), it has to return an error.
++
++
++XIB(external inode number bitmap)
++----------------------------------------------------------------------
++Addition to the xino file per a branch, aufs has an external inode number
++bitmap in a superblock object. It is also an internal file such like a
++xino file.
++It is a simple bitmap to mark whether the aufs inode number is in-use or
++not.
++To reduce the file I/O, aufs prepares a single memory page to cache xib.
++
++As well as XINO files, aufs has a feature to truncate/refresh XIB to
++reduce the number of consumed disk blocks for these files.
++
++
++Virtual or Vertical Dir, and Readdir in Userspace
++----------------------------------------------------------------------
++In order to support multiple layers (branches), aufs readdir operation
++constructs a virtual dir block on memory. For readdir, aufs calls
++vfs_readdir() internally for each dir on branches, merges their entries
++with eliminating the whiteout-ed ones, and sets it to file (dir)
++object. So the file object has its entry list until it is closed. The
++entry list will be updated when the file position is zero and becomes
++obsoleted. This decision is made in aufs automatically.
++
++The dynamically allocated memory block for the name of entries has a
++unit of 512 bytes (by default) and stores the names contiguously (no
++padding). Another block for each entry is handled by kmem_cache too.
++During building dir blocks, aufs creates hash list and judging whether
++the entry is whiteouted by its upper branch or already listed.
++The merged result is cached in the corresponding inode object and
++maintained by a customizable life-time option.
++
++Some people may call it can be a security hole or invite DoS attack
++since the opened and once readdir-ed dir (file object) holds its entry
++list and becomes a pressure for system memory. But I'd say it is similar
++to files under /proc or /sys. The virtual files in them also holds a
++memory page (generally) while they are opened. When an idea to reduce
++memory for them is introduced, it will be applied to aufs too.
++For those who really hate this situation, I've developed readdir(3)
++library which operates this merging in userspace. You just need to set
++LD_PRELOAD environment variable, and aufs will not consume no memory in
++kernel space for readdir(3).
++
++
++Workqueue
++----------------------------------------------------------------------
++Aufs sometimes requires privilege access to a branch. For instance,
++in copy-up/down operation. When a user process is going to make changes
++to a file which exists in the lower readonly branch only, and the mode
++of one of ancestor directories may not be writable by a user
++process. Here aufs copy-up the file with its ancestors and they may
++require privilege to set its owner/group/mode/etc.
++This is a typical case of a application character of aufs (see
++Introduction).
++
++Aufs uses workqueue synchronously for this case. It creates its own
++workqueue. The workqueue is a kernel thread and has privilege. Aufs
++passes the request to call mkdir or write (for example), and wait for
++its completion. This approach solves a problem of a signal handler
++simply.
++If aufs didn't adopt the workqueue and changed the privilege of the
++process, then the process may receive the unexpected SIGXFSZ or other
++signals.
++
++Also aufs uses the system global workqueue ("events" kernel thread) too
++for asynchronous tasks, such like handling inotify/fsnotify, re-creating a
++whiteout base and etc. This is unrelated to a privilege.
++Most of aufs operation tries acquiring a rw_semaphore for aufs
++superblock at the beginning, at the same time waits for the completion
++of all queued asynchronous tasks.
++
++
++Whiteout
++----------------------------------------------------------------------
++The whiteout in aufs is very similar to Unionfs's. That is represented
++by its filename. UnionMount takes an approach of a file mode, but I am
++afraid several utilities (find(1) or something) will have to support it.
++
++Basically the whiteout represents "logical deletion" which stops aufs to
++lookup further, but also it represents "dir is opaque" which also stop
++further lookup.
++
++In aufs, rmdir(2) and rename(2) for dir uses whiteout alternatively.
++In order to make several functions in a single systemcall to be
++revertible, aufs adopts an approach to rename a directory to a temporary
++unique whiteouted name.
++For example, in rename(2) dir where the target dir already existed, aufs
++renames the target dir to a temporary unique whiteouted name before the
++actual rename on a branch, and then handles other actions (make it opaque,
++update the attributes, etc). If an error happens in these actions, aufs
++simply renames the whiteouted name back and returns an error. If all are
++succeeded, aufs registers a function to remove the whiteouted unique
++temporary name completely and asynchronously to the system global
++workqueue.
++
++
++Copy-up
++----------------------------------------------------------------------
++It is a well-known feature or concept.
++When user modifies a file on a readonly branch, aufs operate "copy-up"
++internally and makes change to the new file on the upper writable branch.
++When the trigger systemcall does not update the timestamps of the parent
++dir, aufs reverts it after copy-up.
++
++
++Move-down (aufs3.9 and later)
++----------------------------------------------------------------------
++"Copy-up" is one of the essential feature in aufs. It copies a file from
++the lower readonly branch to the upper writable branch when a user
++changes something about the file.
++"Move-down" is an opposite action of copy-up. Basically this action is
++ran manually instead of automatically and internally.
++For desgin and implementation, aufs has to consider these issues.
++- whiteout for the file may exist on the lower branch.
++- ancestor directories may not exist on the lower branch.
++- diropq for the ancestor directories may exist on the upper branch.
++- free space on the lower branch will reduce.
++- another access to the file may happen during moving-down, including
++ UDBA (see "Revalidate Dentry and UDBA").
++- the file should not be hard-linked nor pseudo-linked. they should be
++ handled by auplink utility later.
++
++Sometimes users want to move-down a file from the upper writable branch
++to the lower readonly or writable branch. For instance,
++- the free space of the upper writable branch is going to run out.
++- create a new intermediate branch between the upper and lower branch.
++- etc.
++
++For this purpose, use "aumvdown" command in aufs-util.git.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/03atomic_open.txt linux/Documentation/filesystems/aufs/design/03atomic_open.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/03atomic_open.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/03atomic_open.txt 2022-11-05 23:02:18.955889283 +0100
+@@ -0,0 +1,85 @@
++
++# Copyright (C) 2015-2022 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/>.
++
++Support for a branch who has its ->atomic_open()
++----------------------------------------------------------------------
++The filesystems who implement its ->atomic_open() are not majority. For
++example NFSv4 does, and aufs should call NFSv4 ->atomic_open,
++particularly for open(O_CREAT|O_EXCL, 0400) case. Other than
++->atomic_open(), NFSv4 returns an error for this open(2). While I am not
++sure whether all filesystems who have ->atomic_open() behave like this,
++but NFSv4 surely returns the error.
++
++In order to support ->atomic_open() for aufs, there are a few
++approaches.
++
++A. Introduce aufs_atomic_open()
++ - calls one of VFS:do_last(), lookup_open() or atomic_open() for
++ branch fs.
++B. Introduce aufs_atomic_open() calling create, open and chmod. this is
++ an aufs user Pip Cet's approach
++ - calls aufs_create(), VFS finish_open() and notify_change().
++ - pass fake-mode to finish_open(), and then correct the mode by
++ notify_change().
++C. Extend aufs_open() to call branch fs's ->atomic_open()
++ - no aufs_atomic_open().
++ - aufs_lookup() registers the TID to an aufs internal object.
++ - aufs_create() does nothing when the matching TID is registered, but
++ registers the mode.
++ - aufs_open() calls branch fs's ->atomic_open() when the matching
++ TID is registered.
++D. Extend aufs_open() to re-try branch fs's ->open() with superuser's
++ credential
++ - no aufs_atomic_open().
++ - aufs_create() registers the TID to an internal object. this info
++ represents "this process created this file just now."
++ - when aufs gets EACCES from branch fs's ->open(), then confirm the
++ registered TID and re-try open() with superuser's credential.
++
++Pros and cons for each approach.
++
++A.
++ - straightforward but highly depends upon VFS internal.
++ - the atomic behavaiour is kept.
++ - some of parameters such as nameidata are hard to reproduce for
++ branch fs.
++ - large overhead.
++B.
++ - easy to implement.
++ - the atomic behavaiour is lost.
++C.
++ - the atomic behavaiour is kept.
++ - dirty and tricky.
++ - VFS checks whether the file is created correctly after calling
++ ->create(), which means this approach doesn't work.
++D.
++ - easy to implement.
++ - the atomic behavaiour is lost.
++ - to open a file with superuser's credential and give it to a user
++ process is a bad idea, since the file object keeps the credential
++ in it. It may affect LSM or something. This approach doesn't work
++ either.
++
++The approach A is ideal, but it hard to implement. So here is a
++variation of A, which is to be implemented.
++
++A-1. Introduce aufs_atomic_open()
++ - calls branch fs ->atomic_open() if exists. otherwise calls
++ vfs_create() and finish_open().
++ - the demerit is that the several checks after branch fs
++ ->atomic_open() are lost. in the ordinary case, the checks are
++ done by VFS:do_last(), lookup_open() and atomic_open(). some can
++ be implemented in aufs, but not all I am afraid.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/03lookup.txt linux/Documentation/filesystems/aufs/design/03lookup.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/03lookup.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/03lookup.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,113 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Lookup in a Branch
++----------------------------------------------------------------------
++Since aufs has a character of sub-VFS (see Introduction), it operates
++lookup for branches as VFS does. It may be a heavy work. But almost all
++lookup operation in aufs is the simplest case, ie. lookup only an entry
++directly connected to its parent. Digging down the directory hierarchy
++is unnecessary. VFS has a function lookup_one_len() for that use, and
++aufs calls it.
++
++When a branch is a remote filesystem, aufs basically relies upon its
++->d_revalidate(), also aufs forces the hardest revalidate tests for
++them.
++For d_revalidate, aufs implements three levels of revalidate tests. See
++"Revalidate Dentry and UDBA" in detail.
++
++
++Test Only the Highest One for the Directory Permission (dirperm1 option)
++----------------------------------------------------------------------
++Let's try case study.
++- aufs has two branches, upper readwrite and lower readonly.
++ /au = /rw + /ro
++- "dirA" exists under /ro, but /rw. and its mode is 0700.
++- user invoked "chmod a+rx /au/dirA"
++- the internal copy-up is activated and "/rw/dirA" is created and its
++ permission bits are set to world readable.
++- then "/au/dirA" becomes world readable?
++
++In this case, /ro/dirA is still 0700 since it exists in readonly branch,
++or it may be a natively readonly filesystem. If aufs respects the lower
++branch, it should not respond readdir request from other users. But user
++allowed it by chmod. Should really aufs rejects showing the entries
++under /ro/dirA?
++
++To be honest, I don't have a good solution for this case. So aufs
++implements 'dirperm1' and 'nodirperm1' mount options, and leave it to
++users.
++When dirperm1 is specified, aufs checks only the highest one for the
++directory permission, and shows the entries. Otherwise, as usual, checks
++every dir existing on all branches and rejects the request.
++
++As a side effect, dirperm1 option improves the performance of aufs
++because the number of permission check is reduced when the number of
++branch is many.
++
++
++Revalidate Dentry and UDBA (User's Direct Branch Access)
++----------------------------------------------------------------------
++Generally VFS helpers re-validate a dentry as a part of lookup.
++0. digging down the directory hierarchy.
++1. lock the parent dir by its i_mutex.
++2. lookup the final (child) entry.
++3. revalidate it.
++4. call the actual operation (create, unlink, etc.)
++5. unlock the parent dir
++
++If the filesystem implements its ->d_revalidate() (step 3), then it is
++called. Actually aufs implements it and checks the dentry on a branch is
++still valid.
++But it is not enough. Because aufs has to release the lock for the
++parent dir on a branch at the end of ->lookup() (step 2) and
++->d_revalidate() (step 3) while the i_mutex of the aufs dir is still
++held by VFS.
++If the file on a branch is changed directly, eg. bypassing aufs, after
++aufs released the lock, then the subsequent operation may cause
++something unpleasant result.
++
++This situation is a result of VFS architecture, ->lookup() and
++->d_revalidate() is separated. But I never say it is wrong. It is a good
++design from VFS's point of view. It is just not suitable for sub-VFS
++character in aufs.
++
++Aufs supports such case by three level of revalidation which is
++selectable by user.
++1. Simple Revalidate
++ Addition to the native flow in VFS's, confirm the child-parent
++ relationship on the branch just after locking the parent dir on the
++ branch in the "actual operation" (step 4). When this validation
++ fails, aufs returns EBUSY. ->d_revalidate() (step 3) in aufs still
++ checks the validation of the dentry on branches.
++2. Monitor Changes Internally by Inotify/Fsnotify
++ Addition to above, in the "actual operation" (step 4) aufs re-lookup
++ the dentry on the branch, and returns EBUSY if it finds different
++ dentry.
++ Additionally, aufs sets the inotify/fsnotify watch for every dir on branches
++ during it is in cache. When the event is notified, aufs registers a
++ function to kernel 'events' thread by schedule_work(). And the
++ function sets some special status to the cached aufs dentry and inode
++ private data. If they are not cached, then aufs has nothing to
++ do. When the same file is accessed through aufs (step 0-3) later,
++ aufs will detect the status and refresh all necessary data.
++ In this mode, aufs has to ignore the event which is fired by aufs
++ itself.
++3. No Extra Validation
++ This is the simplest test and doesn't add any additional revalidation
++ test, and skip the revalidation in step 4. It is useful and improves
++ aufs performance when system surely hide the aufs branches from user,
++ by over-mounting something (or another method).
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/04branch.txt linux/Documentation/filesystems/aufs/design/04branch.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/04branch.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/04branch.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,74 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Branch Manipulation
++
++Since aufs supports dynamic branch manipulation, ie. add/remove a branch
++and changing its permission/attribute, there are a lot of works to do.
++
++
++Add a Branch
++----------------------------------------------------------------------
++o Confirm the adding dir exists outside of aufs, including loopback
++ mount, and its various attributes.
++o Initialize the xino file and whiteout bases if necessary.
++ See struct.txt.
++
++o Check the owner/group/mode of the directory
++ When the owner/group/mode of the adding directory differs from the
++ existing branch, aufs issues a warning because it may impose a
++ security risk.
++ For example, when a upper writable branch has a world writable empty
++ top directory, a malicious user can create any files on the writable
++ branch directly, like copy-up and modify manually. If something like
++ /etc/{passwd,shadow} exists on the lower readonly branch but the upper
++ writable branch, and the writable branch is world-writable, then a
++ malicious guy may create /etc/passwd on the writable branch directly
++ and the infected file will be valid in aufs.
++ I am afraid it can be a security issue, but aufs can do nothing except
++ producing a warning.
++
++
++Delete a Branch
++----------------------------------------------------------------------
++o Confirm the deleting branch is not busy
++ To be general, there is one merit to adopt "remount" interface to
++ manipulate branches. It is to discard caches. At deleting a branch,
++ aufs checks the still cached (and connected) dentries and inodes. If
++ there are any, then they are all in-use. An inode without its
++ corresponding dentry can be alive alone (for example, inotify/fsnotify case).
++
++ For the cached one, aufs checks whether the same named entry exists on
++ other branches.
++ If the cached one is a directory, because aufs provides a merged view
++ to users, as long as one dir is left on any branch aufs can show the
++ dir to users. In this case, the branch can be removed from aufs.
++ Otherwise aufs rejects deleting the branch.
++
++ If any file on the deleting branch is opened by aufs, then aufs
++ rejects deleting.
++
++
++Modify the Permission of a Branch
++----------------------------------------------------------------------
++o Re-initialize or remove the xino file and whiteout bases if necessary.
++ See struct.txt.
++
++o rw --> ro: Confirm the modifying branch is not busy
++ Aufs rejects the request if any of these conditions are true.
++ - a file on the branch is mmap-ed.
++ - a regular file on the branch is opened for write and there is no
++ same named entry on the upper branch.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/05wbr_policy.txt linux/Documentation/filesystems/aufs/design/05wbr_policy.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/05wbr_policy.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/05wbr_policy.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,64 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Policies to Select One among Multiple Writable Branches
++----------------------------------------------------------------------
++When the number of writable branch is more than one, aufs has to decide
++the target branch for file creation or copy-up. By default, the highest
++writable branch which has the parent (or ancestor) dir of the target
++file is chosen (top-down-parent policy).
++By user's request, aufs implements some other policies to select the
++writable branch, for file creation several policies, round-robin,
++most-free-space, and other policies. For copy-up, top-down-parent,
++bottom-up-parent, bottom-up and others.
++
++As expected, the round-robin policy selects the branch in circular. When
++you have two writable branches and creates 10 new files, 5 files will be
++created for each branch. mkdir(2) systemcall is an exception. When you
++create 10 new directories, all will be created on the same branch.
++And the most-free-space policy selects the one which has most free
++space among the writable branches. The amount of free space will be
++checked by aufs internally, and users can specify its time interval.
++
++The policies for copy-up is more simple,
++top-down-parent is equivalent to the same named on in create policy,
++bottom-up-parent selects the writable branch where the parent dir
++exists and the nearest upper one from the copyup-source,
++bottom-up selects the nearest upper writable branch from the
++copyup-source, regardless the existence of the parent dir.
++
++There are some rules or exceptions to apply these policies.
++- If there is a readonly branch above the policy-selected branch and
++ the parent dir is marked as opaque (a variation of whiteout), or the
++ target (creating) file is whiteout-ed on the upper readonly branch,
++ then the result of the policy is ignored and the target file will be
++ created on the nearest upper writable branch than the readonly branch.
++- If there is a writable branch above the policy-selected branch and
++ the parent dir is marked as opaque or the target file is whiteouted
++ on the branch, then the result of the policy is ignored and the target
++ file will be created on the highest one among the upper writable
++ branches who has diropq or whiteout. In case of whiteout, aufs removes
++ it as usual.
++- link(2) and rename(2) systemcalls are exceptions in every policy.
++ They try selecting the branch where the source exists as possible
++ since copyup a large file will take long time. If it can't be,
++ ie. the branch where the source exists is readonly, then they will
++ follow the copyup policy.
++- There is an exception for rename(2) when the target exists.
++ If the rename target exists, aufs compares the index of the branches
++ 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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,44 @@
++
++// to view this graph, run dot(1) command in GRAPHVIZ.
++//
++// 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/>.
++
++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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,102 @@
++
++# Copyright (C) 2017-2022 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 referred 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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,118 @@
++
++# Copyright (C) 2011-2022 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/>.
++
++File-based Hierarchical Storage Management (FHSM)
++----------------------------------------------------------------------
++Hierarchical Storage Management (or HSM) is a well-known feature in the
++storage world. Aufs provides this feature as file-based with multiple
++writable branches, based upon the principle of "Colder, the Lower".
++Here the word "colder" means that the less used files, and "lower" means
++that the position in the order of the stacked branches vertically.
++These multiple writable branches are prioritized, ie. the topmost one
++should be the fastest drive and be used heavily.
++
++o Characters in aufs FHSM story
++- aufs itself and a new branch attribute.
++- a new ioctl interface to move-down and to establish a connection with
++ the daemon ("move-down" is a converse of "copy-up").
++- userspace tool and daemon.
++
++The userspace daemon establishes a connection with aufs and waits for
++the notification. The notified information is very similar to struct
++statfs containing the number of consumed blocks and inodes.
++When the consumed blocks/inodes of a branch exceeds the user-specified
++upper watermark, the daemon activates its move-down process until the
++consumed blocks/inodes reaches the user-specified lower watermark.
++
++The actual move-down is done by aufs based upon the request from
++user-space since we need to maintain the inode number and the internal
++pointer arrays in aufs.
++
++Currently aufs FHSM handles the regular files only. Additionally they
++must not be hard-linked nor pseudo-linked.
++
++
++o Cowork of aufs and the user-space daemon
++ During the userspace daemon established the connection, aufs sends a
++ small notification to it whenever aufs writes something into the
++ writable branch. But it may cost high since aufs issues statfs(2)
++ internally. So user can specify a new option to cache the
++ info. Actually the notification is controlled by these factors.
++ + the specified cache time.
++ + classified as "force" by aufs internally.
++ Until the specified time expires, aufs doesn't send the info
++ except the forced cases. When aufs decide forcing, the info is always
++ notified to userspace.
++ For example, the number of free inodes is generally large enough and
++ the shortage of it happens rarely. So aufs doesn't force the
++ notification when creating a new file, directory and others. This is
++ the typical case which aufs doesn't force.
++ When aufs writes the actual filedata and the files consumes any of new
++ blocks, the aufs forces notifying.
++
++
++o Interfaces in aufs
++- New branch attribute.
++ + fhsm
++ Specifies that the branch is managed by FHSM feature. In other word,
++ participant in the FHSM.
++ When nofhsm is set to the branch, it will not be the source/target
++ branch of the move-down operation. This attribute is set
++ independently from coo and moo attributes, and if you want full
++ FHSM, you should specify them as well.
++- New mount option.
++ + fhsm_sec
++ Specifies a second to suppress many less important info to be
++ notified.
++- New ioctl.
++ + AUFS_CTL_FHSM_FD
++ create a new file descriptor which userspace can read the notification
++ (a subset of struct statfs) from aufs.
++- Module parameter 'brs'
++ It has to be set to 1. Otherwise the new mount option 'fhsm' will not
++ be set.
++- mount helpers /sbin/mount.aufs and /sbin/umount.aufs
++ When there are two or more branches with fhsm attributes,
++ /sbin/mount.aufs invokes the user-space daemon and /sbin/umount.aufs
++ terminates it. As a result of remounting and branch-manipulation, the
++ number of branches with fhsm attribute can be one. In this case,
++ /sbin/mount.aufs will terminate the user-space daemon.
++
++
++Finally the operation is done as these steps in kernel-space.
++- make sure that,
++ + no one else is using the file.
++ + the file is not hard-linked.
++ + the file is not pseudo-linked.
++ + the file is a regular file.
++ + the parent dir is not opaqued.
++- find the target writable branch.
++- make sure the file is not whiteout-ed by the upper (than the target)
++ branch.
++- make the parent dir on the target branch.
++- mutex lock the inode on the branch.
++- unlink the whiteout on the target branch (if exists).
++- lookup and create the whiteout-ed temporary name on the target branch.
++- copy the file as the whiteout-ed temporary name on the target branch.
++- rename the whiteout-ed temporary name to the original name.
++- unlink the file on the source branch.
++- maintain the internal pointer array and the external inode number
++ table (XINO).
++- maintain the timestamps and other attributes of the parent dir and the
++ file.
++
++And of course, in every step, an error may happen. So the operation
++should restore the original file state after an error happens.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/06mmap.txt linux/Documentation/filesystems/aufs/design/06mmap.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/06mmap.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/06mmap.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,72 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++mmap(2) -- File Memory Mapping
++----------------------------------------------------------------------
++In aufs, the file-mapped pages are handled by a branch fs directly, no
++interaction with aufs. It means aufs_mmap() calls the branch fs's
++->mmap().
++This approach is simple and good, but there is one problem.
++Under /proc, several entries show the mmapped files by its path (with
++device and inode number), and the printed path will be the path on the
++branch fs's instead of virtual aufs's.
++This is not a problem in most cases, but some utilities lsof(1) (and its
++user) may expect the path on aufs.
++
++To address this issue, aufs adds a new member called vm_prfile in struct
++vm_area_struct (and struct vm_region). The original vm_file points to
++the file on the branch fs in order to handle everything correctly as
++usual. The new vm_prfile points to a virtual file in aufs, and the
++show-functions in procfs refers to vm_prfile if it is set.
++Also we need to maintain several other places where touching vm_file
++such like
++- fork()/clone() copies vma and the reference count of vm_file is
++ incremented.
++- merging vma maintains the ref count too.
++
++This is not a good approach. It just fakes the printed path. But it
++leaves all behaviour around f_mapping unchanged. This is surely an
++advantage.
++Actually aufs had adopted another complicated approach which calls
++generic_file_mmap() and handles struct vm_operations_struct. In this
++approach, aufs met a hard problem and I could not solve it without
++switching the approach.
++
++There may be one more another approach which is
++- bind-mount the branch-root onto the aufs-root internally
++- grab the new vfsmount (ie. struct mount)
++- lazy-umount the branch-root internally
++- in open(2) the aufs-file, open the branch-file with the hidden
++ vfsmount (instead of the original branch's vfsmount)
++- ideally this "bind-mount and lazy-umount" should be done atomically,
++ but it may be possible from userspace by the mount helper.
++
++Adding the internal hidden vfsmount and using it in opening a file, the
++file path under /proc will be printed correctly. This approach looks
++smarter, but is not possible I am afraid.
++- aufs-root may be bind-mount later. when it happens, another hidden
++ vfsmount will be required.
++- it is hard to get the chance to bind-mount and lazy-umount
++ + in kernel-space, FS can have vfsmount in open(2) via
++ file->f_path, and aufs can know its vfsmount. But several locks are
++ already acquired, and if aufs tries to bind-mount and lazy-umount
++ here, then it may cause a deadlock.
++ + in user-space, bind-mount doesn't invoke the mount helper.
++- since /proc shows dev and ino, aufs has to give vma these info. it
++ means a new member vm_prinode will be necessary. this is essentially
++ equivalent to vm_prfile described above.
++
++I have to give up this "looks-smater" approach.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/06xattr.txt linux/Documentation/filesystems/aufs/design/06xattr.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/06xattr.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/06xattr.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,94 @@
++
++# Copyright (C) 2014-2022 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/>.
++
++Listing XATTR/EA and getting the value
++----------------------------------------------------------------------
++For the inode standard attributes (owner, group, timestamps, etc.), aufs
++shows the values from the topmost existing file. This behaviour is good
++for the non-dir entries since the bahaviour exactly matches the shown
++information. But for the directories, aufs considers all the same named
++entries on the lower branches. Which means, if one of the lower entry
++rejects readdir call, then aufs returns an error even if the topmost
++entry allows it. This behaviour is necessary to respect the branch fs's
++security, but can make users confused since the user-visible standard
++attributes don't match the behaviour.
++To address this issue, aufs has a mount option called dirperm1 which
++checks the permission for the topmost entry only, and ignores the lower
++entry's permission.
++
++A similar issue can happen around XATTR.
++getxattr(2) and listxattr(2) families behave as if dirperm1 option is
++always set. Otherwise these very unpleasant situation would happen.
++- listxattr(2) may return the duplicated entries.
++- users may not be able to remove or reset the XATTR forever,
++
++
++XATTR/EA support in the internal (copy,move)-(up,down)
++----------------------------------------------------------------------
++Generally the extended attributes of inode are categorized as these.
++- "security" for LSM and capability.
++- "system" for posix ACL, 'acl' mount option is required for the branch
++ fs generally.
++- "trusted" for userspace, CAP_SYS_ADMIN is required.
++- "user" for userspace, 'user_xattr' mount option is required for the
++ branch fs generally.
++
++Moreover there are some other categories. Aufs handles these rather
++unpopular categories as the ordinary ones, ie. there is no special
++condition nor exception.
++
++In copy-up, the support for XATTR on the dst branch may differ from the
++src branch. In this case, the copy-up operation will get an error and
++the original user operation which triggered the copy-up will fail. It
++can happen that even all copy-up will fail.
++When both of src and dst branches support XATTR and if an error occurs
++during copying XATTR, then the copy-up should fail obviously. That is a
++good reason and aufs should return an error to userspace. But when only
++the src branch support that XATTR, aufs should not return an error.
++For example, the src branch supports ACL but the dst branch doesn't
++because the dst branch may natively un-support it or temporary
++un-support it due to "noacl" mount option. Of course, the dst branch fs
++may NOT return an error even if the XATTR is not supported. It is
++totally up to the branch fs.
++
++Anyway when the aufs internal copy-up gets an error from the dst branch
++fs, then aufs tries removing the just copied entry and returns the error
++to the userspace. The worst case of this situation will be all copy-up
++will fail.
++
++For the copy-up operation, there two basic approaches.
++- copy the specified XATTR only (by category above), and return the
++ error unconditionally if it happens.
++- copy all XATTR, and ignore the error on the specified category only.
++
++In order to support XATTR and to implement the correct behaviour, aufs
++chooses the latter approach and introduces some new branch attributes,
++"icexsec", "icexsys", "icextr", "icexusr", and "icexoth".
++They correspond to the XATTR namespaces (see above). Additionally, to be
++convenient, "icex" is also provided which means all "icex*" attributes
++are set (here the word "icex" stands for "ignore copy-error on XATTR").
++
++The meaning of these attributes is to ignore the error from setting
++XATTR on that branch.
++Note that aufs tries copying all XATTR unconditionally, and ignores the
++error from the dst branch according to the specified attributes.
++
++Some XATTR may have its default value. The default value may come from
++the parent dir or the environment. If the default value is set at the
++file creating-time, it will be overwritten by copy-up.
++Some contradiction may happen I am afraid.
++Do we need another attribute to stop copying XATTR? I am unsure. For
++now, aufs implements the branch attributes to ignore the error.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/07export.txt linux/Documentation/filesystems/aufs/design/07export.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/07export.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/07export.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,58 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Export Aufs via NFS
++----------------------------------------------------------------------
++Here is an approach.
++- like xino/xib, add a new file 'xigen' which stores aufs inode
++ generation.
++- iget_locked(): initialize aufs inode generation for a new inode, and
++ store it in xigen file.
++- destroy_inode(): increment aufs inode generation and store it in xigen
++ file. it is necessary even if it is not unlinked, because any data of
++ inode may be changed by UDBA.
++- encode_fh(): for a root dir, simply return FILEID_ROOT. otherwise
++ build file handle by
++ + branch id (4 bytes)
++ + superblock generation (4 bytes)
++ + inode number (4 or 8 bytes)
++ + parent dir inode number (4 or 8 bytes)
++ + inode generation (4 bytes))
++ + return value of exportfs_encode_fh() for the parent on a branch (4
++ bytes)
++ + file handle for a branch (by exportfs_encode_fh())
++- fh_to_dentry():
++ + find the index of a branch from its id in handle, and check it is
++ still exist in aufs.
++ + 1st level: get the inode number from handle and search it in cache.
++ + 2nd level: if not found in cache, get the parent inode number from
++ the handle and search it in cache. and then open the found parent
++ dir, find the matching inode number by vfs_readdir() and get its
++ name, and call lookup_one_len() for the target dentry.
++ + 3rd level: if the parent dir is not cached, call
++ exportfs_decode_fh() for a branch and get the parent on a branch,
++ build a pathname of it, convert it a pathname in aufs, call
++ path_lookup(). now aufs gets a parent dir dentry, then handle it as
++ the 2nd level.
++ + to open the dir, aufs needs struct vfsmount. aufs keeps vfsmount
++ for every branch, but not itself. to get this, (currently) aufs
++ searches in current->nsproxy->mnt_ns list. it may not be a good
++ idea, but I didn't get other approach.
++ + test the generation of the gotten inode.
++- every inode operation: they may get EBUSY due to UDBA. in this case,
++ convert it into ESTALE for NFSD.
++- readdir(): call lockdep_on/off() because filldir in NFSD calls
++ lookup_one_len(), vfs_getattr(), encode_fh() and others.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/08shwh.txt linux/Documentation/filesystems/aufs/design/08shwh.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/08shwh.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/08shwh.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,52 @@
++
++# Copyright (C) 2005-2022 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/>.
++
++Show Whiteout Mode (shwh)
++----------------------------------------------------------------------
++Generally aufs hides the name of whiteouts. But in some cases, to show
++them is very useful for users. For instance, creating a new middle layer
++(branch) by merging existing layers.
++
++(borrowing aufs1 HOW-TO from a user, Michael Towers)
++When you have three branches,
++- Bottom: 'system', squashfs (underlying base system), read-only
++- Middle: 'mods', squashfs, read-only
++- Top: 'overlay', ram (tmpfs), read-write
++
++The top layer is loaded at boot time and saved at shutdown, to preserve
++the changes made to the system during the session.
++When larger changes have been made, or smaller changes have accumulated,
++the size of the saved top layer data grows. At this point, it would be
++nice to be able to merge the two overlay branches ('mods' and 'overlay')
++and rewrite the 'mods' squashfs, clearing the top layer and thus
++restoring save and load speed.
++
++This merging is simplified by the use of another aufs mount, of just the
++two overlay branches using the 'shwh' option.
++# mount -t aufs -o ro,shwh,br:/livesys/overlay=ro+wh:/livesys/mods=rr+wh \
++ aufs /livesys/merge_union
++
++A merged view of these two branches is then available at
++/livesys/merge_union, and the new feature is that the whiteouts are
++visible!
++Note that in 'shwh' mode the aufs mount must be 'ro', which will disable
++writing to all branches. Also the default mode for all branches is 'ro'.
++It is now possible to save the combined contents of the two overlay
++branches to a new squashfs, e.g.:
++# mksquashfs /livesys/merge_union /path/to/newmods.squash
++
++This new squashfs archive can be stored on the boot device and the
++initramfs will use it to replace the old one at the next boot.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/design/10dynop.txt linux/Documentation/filesystems/aufs/design/10dynop.txt
+--- /usr/share/empty/Documentation/filesystems/aufs/design/10dynop.txt 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/design/10dynop.txt 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,47 @@
++
++# Copyright (C) 2010-2022 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/>.
++
++Dynamically customizable FS operations
++----------------------------------------------------------------------
++Generally FS operations (struct inode_operations, struct
++address_space_operations, struct file_operations, etc.) are defined as
++"static const", but it never means that FS have only one set of
++operation. Some FS have multiple sets of them. For instance, ext2 has
++three sets, one for XIP, for NOBH, and for normal.
++Since aufs overrides and redirects these operations, sometimes aufs has
++to change its behaviour according to the branch FS type. More importantly
++VFS acts differently if a function (member in the struct) is set or
++not. It means aufs should have several sets of operations and select one
++among them according to the branch FS definition.
++
++In order to solve this problem and not to affect the behaviour of VFS,
++aufs defines these operations dynamically. For instance, aufs defines
++dummy direct_IO function for struct address_space_operations, but it may
++not be set to the address_space_operations actually. When the branch FS
++doesn't have it, aufs doesn't set it to its address_space_operations
++while the function definition itself is still alive. So the behaviour
++itself will not change, and it will return an error when direct_IO is
++not set.
++
++The lifetime of these dynamically generated operation object is
++maintained by aufs branch object. When the branch is removed from aufs,
++the reference counter of the object is decremented. When it reaches
++zero, the dynamically generated operation object will be freed.
++
++This approach is designed to support AIO (io_submit), Direct I/O and
++XIP (DAX) mainly.
++Currently this approach is applied to address_space_operations for
++regular files only.
+diff -urN /usr/share/empty/Documentation/filesystems/aufs/README linux/Documentation/filesystems/aufs/README
+--- /usr/share/empty/Documentation/filesystems/aufs/README 1970-01-01 01:00:00.000000000 +0100
++++ linux/Documentation/filesystems/aufs/README 2023-08-28 12:34:41.669969465 +0200
+@@ -0,0 +1,409 @@
++
++Aufs6 -- advanced multi layered unification filesystem version 6.x
++http://aufs.sf.net
++Junjiro R. Okajima
++
++
++0. Introduction
++----------------------------------------
++In the early days, aufs was entirely re-designed and re-implemented
++Unionfs Version 1.x series. Adding many original ideas, approaches,
++improvements and implementations, it became totally different from
++Unionfs while keeping the basic features.
++Later, Unionfs Version 2.x series began taking some of the same
++approaches to aufs1's.
++Unionfs was being developed by Professor Erez Zadok at Stony Brook
++University and his team.
++
++Aufs6 supports linux-v6.0 and later, try aufs6.0 branch in
++aufs-linux.git or aufs-standalone.git.
++If you want older kernel version support,
++- for linux-v5.x series, try aufs-linux.git or aufs-standalone.git
++- for linux-v4.x series, try aufs4-linux.git or aufs4-standalone.git
++- for linux-v3.x series, try aufs3-linux.git or aufs3-standalone.git
++- for linux-v2.6.16 and later, try aufs2-2.6.git, aufs2-standalone.git
++ or aufs1 from CVS on SourceForge.
++
++Note: the name of aufs5-linux.git and aufs5-standalone.git on github
++ were changed. Now they are aufs-linux.git and
++ aufs-standalone.git and they contain aufs5 and later branches.
++
++Note: it becomes clear that "Aufs was rejected. Let's give it up."
++ According to Christoph Hellwig, linux rejects all union-type
++ filesystems but UnionMount.
++<http://marc.info/?l=linux-kernel&m=123938533724484&w=2>
++
++PS. Al Viro seems have a plan to merge aufs as well as overlayfs and
++ UnionMount, and he pointed out an issue around a directory mutex
++ lock and aufs addressed it. But it is still unsure whether aufs will
++ be merged (or any other union solution).
++<http://marc.info/?l=linux-kernel&m=136312705029295&w=1>
++
++
++1. Features
++----------------------------------------
++- unite several directories into a single virtual filesystem. The member
++ directory is called as a branch.
++- you can specify the permission flags to the branch, which are 'readonly',
++ 'readwrite' and 'whiteout-able.'
++- by upper writable branch, internal copyup and whiteout, files/dirs on
++ readonly branch are modifiable logically.
++- dynamic branch manipulation, add, del.
++- etc...
++
++Also there are many enhancements in aufs, such as:
++- test only the highest one for the directory permission (dirperm1)
++- copyup on open (coo=)
++- 'move' policy for copy-up between two writable branches, after
++ checking free space.
++- xattr, acl
++- readdir(3) in userspace.
++- keep inode number by external inode number table
++- keep the timestamps of file/dir in internal copyup operation
++- seekable directory, supporting NFS readdir.
++- whiteout is hardlinked in order to reduce the consumption of inodes
++ on branch
++- do not copyup, nor create a whiteout when it is unnecessary
++- revert a single systemcall when an error occurs in aufs
++- remount interface instead of ioctl
++- maintain /etc/mtab by an external command, /sbin/mount.aufs.
++- loopback mounted filesystem as a branch
++- kernel thread for removing the dir who has a plenty of whiteouts
++- support copyup sparse file (a file which has a 'hole' in it)
++- default permission flags for branches
++- selectable permission flags for ro branch, whether whiteout can
++ exist or not
++- export via NFS.
++- support <sysfs>/fs/aufs and <debugfs>/aufs.
++- support multiple writable branches, some policies to select one
++ among multiple writable branches.
++- a new semantics for link(2) and rename(2) to support multiple
++ writable branches.
++- no glibc changes are required.
++- pseudo hardlink (hardlink over branches)
++- allow a direct access manually to a file on branch, e.g. bypassing aufs.
++ including NFS or remote filesystem branch.
++- userspace wrapper for pathconf(3)/fpathconf(3) with _PC_LINK_MAX.
++- and more...
++
++Currently these features are dropped temporary from aufs6.
++See design/08plan.txt in detail.
++- nested mount, i.e. aufs as readonly no-whiteout branch of another aufs
++ (robr)
++- statistics of aufs thread (/sys/fs/aufs/stat)
++
++Features or just an idea in the future (see also design/*.txt),
++- reorder the branch index without del/re-add.
++- permanent xino files for NFSD
++- an option for refreshing the opened files after add/del branches
++- light version, without branch manipulation. (unnecessary?)
++- copyup in userspace
++- inotify in userspace
++- readv/writev
++
++
++2. Download
++----------------------------------------
++There are three GIT trees for aufs6, aufs-linux.git,
++aufs-standalone.git, and aufs-util.git.
++While the aufs-util is always necessary, you need either of aufs-linux
++or aufs-standalone.
++
++The aufs-linux tree includes the whole linux mainline GIT tree,
++git://git.kernel.org/.../torvalds/linux.git.
++And you cannot select CONFIG_AUFS_FS=m for this version, eg. you cannot
++build aufs6 as an external kernel module.
++Several extra patches are not included in this tree. Only
++aufs-standalone tree contains them. They are described in the later
++section "Configuration and Compilation."
++
++On the other hand, the aufs-standalone tree has only aufs source files
++and necessary patches, and you can select CONFIG_AUFS_FS=m.
++But you need to apply all aufs patches manually.
++
++You will find GIT branches whose name is in form of "aufs6.x" where "x"
++represents the linux kernel version, "linux-6.x". For instance,
++"aufs6.0" is for linux-6.0. For latest "linux-6.x-rcN", use
++"aufs6.x-rcN" branch.
++
++o aufs-linux tree
++$ git clone --reference /your/linux/git/tree \
++ git://github.com/sfjro/aufs-linux.git aufs-linux.git
++- if you don't have linux GIT tree, then remove "--reference ..."
++$ cd aufs-linux.git
++$ git checkout origin/aufs6.0
++
++Or You may want to directly git-pull aufs into your linux GIT tree, and
++leave the patch-work to GIT.
++$ cd /your/linux/git/tree
++$ git remote add aufs git://github.com/sfjro/aufs-linux.git
++$ git fetch aufs
++$ git checkout -b my6.0 v6.0
++$ (add your local change...)
++$ git pull aufs aufs6.0
++- now you have v6.0 + your_changes + aufs6.0 in you my6.0 branch.
++- you may need to solve some conflicts between your_changes and
++ aufs6.0. in this case, git-rerere is recommended so that you can
++ solve the similar conflicts automatically when you upgrade to 6.1 or
++ later in the future.
++
++o aufs-standalone tree
++$ git clone git://github.com/sfjro/aufs-standalone.git aufs-standalone.git
++$ cd aufs-standalone.git
++$ git checkout origin/aufs6.0
++
++o aufs-util tree
++$ git clone git://git.code.sf.net/p/aufs/aufs-util aufs-util.git
++- note that the public aufs-util.git is on SourceForge instead of
++ GitHUB.
++$ cd aufs-util.git
++$ git checkout origin/aufs6.0
++
++Note: The 6.x-rcN branch is to be used with `rc' kernel versions ONLY.
++The minor version number, 'x' in '6.x', of aufs may not always
++follow the minor version number of the kernel.
++Because changes in the kernel that cause the use of a new
++minor version number do not always require changes to aufs-util.
++
++Since aufs-util has its own minor version number, you may not be
++able to find a GIT branch in aufs-util for your kernel's
++exact minor version number.
++In this case, you should git-checkout the branch for the
++nearest lower number.
++
++For (an unreleased) example:
++If you are using "linux-6.10" and the "aufs6.10" branch
++does not exist in aufs-util repository, then "aufs6.9", "aufs6.8"
++or something numerically smaller is the branch for your kernel.
++
++Also you can view all branches by
++ $ git branch -a
++
++
++3. Configuration and Compilation
++----------------------------------------
++Make sure you have git-checkout'ed the correct branch.
++
++For aufs-linux tree,
++- enable CONFIG_AUFS_FS.
++- set other aufs configurations if necessary.
++- for aufs5.13 and later
++ Because aufs is not only an ordinary filesystem (callee of VFS), but
++ also a caller of VFS functions for branch filesystems, subclassing of
++ the internal locks for LOCKDEP is necessary. LOCKDEP is a debugging
++ feature of linux kernel. If you enable CONFIG_LOCKDEP, then you will
++ need to customize some LOCKDEP numbers. Here are what I use on my
++ test environment.
++ CONFIG_LOCKDEP_BITS=21
++ CONFIG_LOCKDEP_CHAINS_BITS=21
++ CONFIG_LOCKDEP_STACK_TRACE_BITS=24
++ Also you will need to expand some constant values in LOCKDEP. Refer
++ to lockdep-debug.patch in aufs-standalone.git.
++
++For aufs-standalone tree,
++There are several ways to build.
++
++1.
++- apply ./aufs6-kbuild.patch to your kernel source files.
++- apply ./aufs6-base.patch too.
++- apply ./aufs6-mmap.patch too.
++- apply ./aufs6-standalone.patch too, if you have a plan to set
++ CONFIG_AUFS_FS=m. otherwise you don't need ./aufs-standalone.patch.
++- copy ./{Documentation,fs,include/uapi/linux/aufs_type.h} files to your
++ kernel source tree. Never copy $PWD/include/uapi/linux/Kbuild.
++- enable CONFIG_AUFS_FS, you can select either
++ =m or =y.
++- and build your kernel as usual.
++- install the built kernel.
++- install the header files too by "make headers_install" to the
++ directory where you specify. By default, it is $PWD/usr.
++ "make help" shows a brief note for headers_install.
++- and reboot your system.
++
++2.
++- module only (CONFIG_AUFS_FS=m).
++- apply ./aufs6-base.patch to your kernel source files.
++- apply ./aufs6-mmap.patch too.
++- apply ./aufs6-standalone.patch too.
++- build your kernel, don't forget "make headers_install", and reboot.
++- edit ./config.mk and set other aufs configurations if necessary.
++ Note: You should read $PWD/fs/aufs/Kconfig carefully which describes
++ every aufs configurations.
++- build the module by simple "make".
++- you can specify ${KDIR} make variable which points to your kernel
++ source tree.
++- install the files
++ + run "make install" to install the aufs module, or copy the built
++ $PWD/aufs.ko to /lib/modules/... and run depmod -a (or reboot simply).
++ + run "make install_headers" (instead of headers_install) to install
++ the modified aufs header file (you can specify DESTDIR which is
++ available in aufs standalone version's Makefile only), or copy
++ $PWD/usr/include/linux/aufs_type.h to /usr/include/linux or wherever
++ you like manually. By default, the target directory is $PWD/usr.
++- no need to apply aufs6-kbuild.patch, nor copying source files to your
++ kernel source tree.
++
++Note: The header file aufs_type.h is necessary to build aufs-util
++ as well as "make headers_install" in the kernel source tree.
++ headers_install is subject to be forgotten, but it is essentially
++ necessary, not only for building aufs-util.
++ You may not meet problems without headers_install in some older
++ version though.
++
++And then,
++- read README in aufs-util, build and install it
++- note that your distribution may contain an obsoleted version of
++ aufs_type.h in /usr/include/linux or something. When you build aufs
++ utilities, make sure that your compiler refers the correct aufs header
++ file which is built by "make headers_install."
++- if you want to use readdir(3) in userspace or pathconf(3) wrapper,
++ then run "make install_ulib" too. And refer to the aufs manual in
++ detail.
++
++There several other patches in aufs-standalone.git. They are all
++optional. When you meet some problems, they will help you.
++- aufs6-loopback.patch
++ Supports a nested loopback mount in a branch-fs. This patch is
++ unnecessary until aufs produces a message like "you may want to try
++ another patch for loopback file".
++- vfs-ino.patch
++ Modifies a system global kernel internal function get_next_ino() in
++ order to stop assigning 0 for an inode-number. Not directly related to
++ aufs, but recommended generally.
++- tmpfs-idr.patch
++ Keeps the tmpfs inode number as the lowest value. Effective to reduce
++ the size of aufs XINO files for tmpfs branch. Also it prevents the
++ duplication of inode number, which is important for backup tools and
++ other utilities. When you find aufs XINO files for tmpfs branch
++ growing too much, try this patch.
++- lockdep-debug.patch
++ Similar to some kernel configurations for LOCKDEP (see the top of
++ this section), you will need expand some constants in LOCKDEP for
++ aufs if you enable CONFIG_LOCKDEP.
++
++
++4. Usage
++----------------------------------------
++At first, make sure aufs-util are installed, and please read the aufs
++manual, aufs.5 in aufs-util.git tree.
++$ man -l aufs.5
++
++And then,
++$ mkdir /tmp/rw /tmp/aufs
++# mount -t aufs -o br=/tmp/rw:${HOME} none /tmp/aufs
++
++Here is another example. The result is equivalent.
++# mount -t aufs -o br=/tmp/rw=rw:${HOME}=ro none /tmp/aufs
++ Or
++# mount -t aufs -o br:/tmp/rw none /tmp/aufs
++# mount -o remount,append:${HOME} /tmp/aufs
++
++Then, you can see whole tree of your home dir through /tmp/aufs. If
++you modify a file under /tmp/aufs, the one on your home directory is
++not affected, instead the same named file will be newly created under
++/tmp/rw. And all of your modification to a file will be applied to
++the one under /tmp/rw. This is called the file based Copy on Write
++(COW) method.
++Aufs mount options are described in aufs.5.
++If you run chroot or something and make your aufs as a root directory,
++then you need to customize the shutdown script. See the aufs manual in
++detail.
++
++Additionally, there are some sample usages of aufs which are a
++diskless system with network booting, and LiveCD over NFS.
++See sample dir in CVS tree on SourceForge.
++
++
++5. Contact
++----------------------------------------
++When you have any problems or strange behaviour in aufs, please let me
++know with:
++- /proc/mounts (instead of the output of mount(8))
++- /sys/module/aufs/*
++- /sys/fs/aufs/* (if you have them)
++- /debug/aufs/* (if you have them)
++- linux kernel version
++ if your kernel is not plain, for example modified by distributor,
++ the url where i can download its source is necessary too.
++- aufs version which was printed at loading the module or booting the
++ system, instead of the date you downloaded.
++- configuration (define/undefine CONFIG_AUFS_xxx)
++- kernel configuration or /proc/config.gz (if you have it)
++- LSM (linux security module, if you are using)
++- behaviour which you think to be incorrect
++- actual operation, reproducible one is better
++- mailto: aufs-users at lists.sourceforge.net
++
++Usually, I don't watch the Public Areas(Bugs, Support Requests, Patches,
++and Feature Requests) on SourceForge. Please join and write to
++aufs-users ML.
++
++
++6. Acknowledgements
++----------------------------------------
++Thanks to everyone who have tried and are using aufs, whoever
++have reported a bug or any feedback.
++
++Especially donators:
++Tomas Matejicek(slax.org) made a donation (much more than once).
++ Since Apr 2010, Tomas M (the author of Slax and Linux Live
++ scripts) is making "doubling" donations.
++ Unfortunately I cannot list all of the donators, but I really
++ appreciate.
++ It ends Aug 2010, but the ordinary donation URL is still available.
++ <http://sourceforge.net/donate/index.php?group_id=167503>
++Dai Itasaka made a donation (2007/8).
++Chuck Smith made a donation (2008/4, 10 and 12).
++Henk Schoneveld made a donation (2008/9).
++Chih-Wei Huang, ASUS, CTC donated Eee PC 4G (2008/10).
++Francois Dupoux made a donation (2008/11).
++Bruno Cesar Ribas and Luis Carlos Erpen de Bona, C3SL serves public
++ aufs2 GIT tree (2009/2).
++William Grant made a donation (2009/3).
++Patrick Lane made a donation (2009/4).
++The Mail Archive (mail-archive.com) made donations (2009/5).
++Nippy Networks (Ed Wildgoose) made a donation (2009/7).
++New Dream Network, LLC (www.dreamhost.com) made a donation (2009/11).
++Pavel Pronskiy made a donation (2011/2).
++Iridium and Inmarsat satellite phone retailer (www.mailasail.com), Nippy
++ Networks (Ed Wildgoose) made a donation for hardware (2011/3).
++Max Lekomcev (DOM-TV project) made a donation (2011/7, 12, 2012/3, 6 and
++11).
++Sam Liddicott made a donation (2011/9).
++Era Scarecrow made a donation (2013/4).
++Bor Ratajc made a donation (2013/4).
++Alessandro Gorreta made a donation (2013/4).
++POIRETTE Marc made a donation (2013/4).
++Alessandro Gorreta made a donation (2013/4).
++lauri kasvandik made a donation (2013/5).
++"pemasu from Finland" made a donation (2013/7).
++The Parted Magic Project made a donation (2013/9 and 11).
++Pavel Barta made a donation (2013/10).
++Nikolay Pertsev made a donation (2014/5).
++James B made a donation (2014/7, 2015/7, and 2021/12).
++Stefano Di Biase made a donation (2014/8).
++Daniel Epellei made a donation (2015/1).
++OmegaPhil made a donation (2016/1, 2018/4).
++Tomasz Szewczyk made a donation (2016/4).
++James Burry made a donation (2016/12).
++Carsten Rose made a donation (2018/9).
++Porteus Kiosk made a donation (2018/10).
++huronOS team: Enya Quetzalli made donations (2022/5, 2023/5 and 8).
++Vasily Mikhaylichenko made a donation (2023/5).
++
++Thank you very much.
++Donations are always, including future donations, very important and
++helpful for me to keep on developing aufs.
++
++
++7.
++----------------------------------------
++If you are an experienced user, no explanation is needed. Aufs is
++just a linux filesystem.
++
++
++Enjoy!
++
++# Local variables: ;
++# mode: text;
++# End: ;
+diff -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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,62 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * main header files
-+ *
-+ * $Id$
++ * all header files
+ */
+
+#ifndef __AUFS_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/version.h>
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 25)
-+#error you got wrong version
-+#endif
-+
-+/* introduced in linux-2.6.27 */
-+#include <linux/bug.h>
-+#ifndef WARN_ONCE
-+#define WARN_ONCE(cond, fmt ...) WARN_ON(cond)
-+#endif
++#define AuStub(type, name, body, ...) \
++ static inline type name(__VA_ARGS__) { body; }
+
-+/* ---------------------------------------------------------------------- */
++#define AuStubVoid(name, ...) \
++ AuStub(void, name, , __VA_ARGS__)
++#define AuStubInt0(name, ...) \
++ AuStub(int, name, return 0, __VA_ARGS__)
+
+#include "debug.h"
+
+#include "branch.h"
+#include "cpup.h"
+#include "dcsub.h"
++#include "dbgaufs.h"
+#include "dentry.h"
+#include "dir.h"
++#include "dirren.h"
++#include "dynop.h"
+#include "file.h"
-+#include "hinode.h"
++#include "fstype.h"
++#include "hbl.h"
+#include "inode.h"
-+#include "misc.h"
++#include "lcnt.h"
++#include "loop.h"
+#include "module.h"
+#include "opts.h"
++#include "rwsem.h"
+#include "super.h"
+#include "sysaufs.h"
+#include "vfsub.h"
+#include "whout.h"
+#include "wkq.h"
-+/* reserved for future use */
-+/* #include "xattr.h" */
-+
-+#ifdef AuNoInlineForStack
-+#undef noinline_for_stack
-+#define noinline_for_stack /* */
-+#endif
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_H__ */
-diff -urN linux/fs/aufs/branch.c linux-aufs/fs/aufs/branch.c
---- linux/fs/aufs/branch.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/branch.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1038 @@
+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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,1427 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * branch management
-+ *
-+ * $Id$
+ */
+
-+#include <linux/iso_fs.h>
-+#include <linux/loop.h>
-+#include <linux/romfs_fs.h>
-+#include <linux/smp_lock.h>
++#include <linux/compat.h>
++#include <linux/statfs.h>
+#include "aufs.h"
+
++/*
++ * free a single branch
++ */
+static void au_br_do_free(struct au_branch *br)
+{
+ int i;
+ struct au_wbr *wbr;
++ struct au_dykey **key;
++
++ au_hnotify_fin_br(br);
++ /* always, regardless the mount option */
++ au_dr_hino_free(&br->br_dirren);
++ au_xino_put(br);
+
-+ AuTraceEnter();
++ AuLCntZero(au_lcnt_read(&br->br_nfiles, /*do_rev*/0));
++ au_lcnt_fin(&br->br_nfiles, /*do_sync*/0);
++ AuLCntZero(au_lcnt_read(&br->br_count, /*do_rev*/0));
++ au_lcnt_fin(&br->br_count, /*do_sync*/0);
+
-+ if (br->br_xino.xi_file)
-+ fput(br->br_xino.xi_file);
-+ mutex_destroy(&br->br_xino.xi_nondir_mtx);
+ wbr = br->br_wbr;
-+ if (wbr)
++ if (wbr) {
+ for (i = 0; i < AuBrWh_Last; i++)
+ dput(wbr->wbr_wh[i]);
-+ /* do not call au_br_nfs_lockdep_off() here */
-+ if (br->br_mnt && !au_test_nfs(br->br_mnt->mnt_sb))
-+ mntput(br->br_mnt);
-+ else {
-+ lockdep_off();
-+ mntput(br->br_mnt);
-+ lockdep_on();
-+ }
-+ AuDebugOn(au_br_count(br));
-+ if (wbr) {
+ AuDebugOn(atomic_read(&wbr->wbr_wh_running));
-+ au_rwsem_destroy(&wbr->wbr_wh_rwsem);
++ AuRwDestroy(&wbr->wbr_wh_rwsem);
++ }
++
++ if (br->br_fhsm) {
++ au_br_fhsm_fin(br->br_fhsm);
++ au_kfree_try_rcu(br->br_fhsm);
+ }
-+ kfree(wbr);
-+ kfree(br);
++
++ key = br->br_dykey;
++ for (i = 0; i < AuBrDynOp; i++, key++)
++ if (*key)
++ au_dy_put(*key);
++ else
++ break;
++
++ /* recursive lock, s_umount of branch's */
++ /* synchronize_rcu(); */ /* why? */
++ lockdep_off();
++ path_put(&br->br_path);
++ lockdep_on();
++ au_kfree_rcu(wbr);
++ au_lcnt_wait_for_fin(&br->br_nfiles);
++ au_lcnt_wait_for_fin(&br->br_count);
++ /* I don't know why, but percpu_refcount requires this */
++ /* synchronize_rcu(); */
++ au_kfree_rcu(br);
+}
+
+/*
+ aufs_bindex_t bmax;
+ struct au_branch **br;
+
-+ AuTraceEnter();
-+ bmax = sbinfo->si_bend + 1;
++ AuRwMustWriteLock(&sbinfo->si_rwsem);
++
++ bmax = sbinfo->si_bbot + 1;
+ br = sbinfo->si_branch;
+ while (bmax--)
+ au_br_do_free(*br++);
+ */
+int au_br_index(struct super_block *sb, aufs_bindex_t br_id)
+{
-+ aufs_bindex_t bindex, bend;
++ aufs_bindex_t bindex, bbot;
+
-+ AuTraceEnter();
-+
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++)
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++)
+ if (au_sbr_id(sb, bindex) == br_id)
+ return bindex;
+ return -1;
+}
+
-+/*
-+ * test if the @h_sb is real-readonly.
-+ */
-+int au_test_def_rr(struct super_block *h_sb)
-+{
-+ switch (h_sb->s_magic) {
-+#ifdef CONFIG_AUFS_RR_SQUASHFS
-+ case SQUASHFS_MAGIC_LZMA:
-+ case SQUASHFS_MAGIC:
-+ case SQUASHFS_MAGIC_LZMA_SWAP:
-+ case SQUASHFS_MAGIC_SWAP:
-+ return 1; /* real readonly */
-+#endif
-+
-+#if defined(CONFIG_ISO9660_FS) || defined(CONFIG_ISO9660_FS_MODULE)
-+ case ISOFS_SUPER_MAGIC:
-+ return 1;
-+#endif
-+
-+#if defined(CONFIG_CRAMFS) || defined(CONFIG_CRAMFS_MODULE)
-+ case CRAMFS_MAGIC:
-+ return 1;
-+#endif
-+
-+#if defined(CONFIG_ROMFS_FS) || defined(CONFIG_ROMFS_FS_MODULE)
-+ case ROMFS_MAGIC:
-+ return 1;
-+#endif
-+
-+ default:
-+ return 0;
-+ }
-+}
-+
+/* ---------------------------------------------------------------------- */
+
+/*
-+ * test if two hidden_dentries have overlapping branches.
++ * add a branch
+ */
-+static int do_test_overlap(struct super_block *sb, struct dentry *h_d1,
-+ struct dentry *h_d2)
-+{
-+ struct dentry *d;
-+
-+ LKTRTrace("%.*s, %.*s\n", AuDLNPair(h_d1), AuDLNPair(h_d2));
-+
-+ d = au_test_subdir(h_d1, h_d2);
-+ if (unlikely(d)) {
-+ AuDbgDentry(h_d1);
-+ AuDbgDentry(h_d2);
-+ }
-+ return !!d;
-+}
-+
-+static int test_overlap_loopback(struct super_block *sb, struct dentry *h_d1,
-+ struct dentry *h_d2)
-+{
-+#if defined(CONFIG_BLK_DEV_LOOP) || defined(CONFIG_BLK_DEV_LOOP_MODULE)
-+ struct inode *h_inode;
-+ struct loop_device *l;
-+
-+ LKTRTrace("%.*s, %.*s\n", AuDLNPair(h_d1), AuDLNPair(h_d2));
-+ AuDbgDentry(h_d1);
-+ AuDbgDentry(h_d2);
-+ AuDbgSb(h_d1->d_sb);
-+ AuDbgSb(h_d2->d_sb);
-+
-+ h_inode = h_d1->d_inode;
-+ if (MAJOR(h_inode->i_sb->s_dev) != LOOP_MAJOR)
-+ return 0;
-+
-+ l = h_inode->i_sb->s_bdev->bd_disk->private_data;
-+ h_d1 = l->lo_backing_file->f_dentry;
-+ /* h_d1 can be local NFS. in this case aufs cannot detect the loop */
-+ AuDbgDentry(h_d1);
-+ AuDbgDentry(h_d2);
-+ AuDbgSb(h_d1->d_sb);
-+ AuDbgSb(h_d2->d_sb);
-+ if (unlikely(h_d1->d_sb == sb))
-+ return 1;
-+ return do_test_overlap(sb, h_d1, h_d2);
-+#else
-+ return 0;
-+#endif
-+}
+
-+static int test_overlap(struct super_block *sb, struct dentry *h_d1,
-+ struct dentry *h_d2)
++static int test_overlap(struct super_block *sb, struct dentry *h_adding,
++ struct dentry *h_root)
+{
-+ LKTRTrace("d1 %.*s, d2 %.*s\n", AuDLNPair(h_d1), AuDLNPair(h_d2));
-+
-+ if (unlikely(h_d1 == h_d2)) {
-+ AuDbgDentry(h_d1);
-+ AuDbgDentry(h_d2);
++ if (unlikely(h_adding == h_root
++ || au_test_loopback_overlap(sb, h_adding)))
+ return 1;
-+ }
-+ return do_test_overlap(sb, h_d1, h_d2)
-+ || do_test_overlap(sb, h_d2, h_d1)
-+ || test_overlap_loopback(sb, h_d1, h_d2)
-+ || test_overlap_loopback(sb, h_d2, h_d1);
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+static int au_br_init_wh(struct super_block *sb, aufs_bindex_t bindex,
-+ struct au_branch *br, int new_perm,
-+ struct dentry *h_root, struct vfsmount *h_mnt)
-+{
-+ int err, old_perm;
-+ struct inode *h_dir;
-+ struct au_wbr *wbr;
-+
-+ LKTRTrace("b%d, new_perm %d\n", bindex, new_perm);
-+ SiMustWriteLock(sb);
-+
-+ wbr = br->br_wbr;
-+ h_dir = h_root->d_inode;
-+ old_perm = br->br_perm;
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ if (wbr)
-+ wbr_wh_write_lock(wbr);
-+ br->br_perm = new_perm;
-+ err = au_wh_init(h_root, br, h_mnt, sb, bindex);
-+ br->br_perm = old_perm;
-+ if (wbr)
-+ wbr_wh_write_unlock(wbr);
-+ mutex_unlock(&h_dir->i_mutex);
-+ if (!err && wbr && !au_br_writable(new_perm)) {
-+ AuDebugOn(wbr->wbr_whbase);
-+ AuDebugOn(wbr->wbr_plink);
-+ AuDebugOn(wbr->wbr_tmp);
-+ kfree(wbr);
-+ br->br_wbr = NULL;
-+ }
-+
-+ AuTraceErr(err);
-+ return err;
++ if (h_adding->d_sb != h_root->d_sb)
++ return 0;
++ return au_test_subdir(h_adding, h_root)
++ || au_test_subdir(h_root, h_adding);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
+/*
+ * returns a newly allocated branch. @new_nbranch is a number of branches
+ * after adding a branch.
+ */
-+static struct au_branch *alloc_addbr(struct super_block *sb, int new_nbranch,
++static struct au_branch *au_br_alloc(struct super_block *sb, int new_nbranch,
+ int perm)
+{
-+ struct au_branch **branchp, *add_branch;
-+ int sz;
-+ void *p;
++ struct au_branch *add_branch;
+ struct dentry *root;
+ struct inode *inode;
-+ struct au_hinode *hinodep;
-+ struct au_hdentry *hdentryp;
-+
-+ LKTRTrace("new_nbranch %d\n", new_nbranch);
-+ SiMustWriteLock(sb);
-+ root = sb->s_root;
-+ DiMustWriteLock(root);
-+ inode = root->d_inode;
-+ IiMustWriteLock(inode);
++ int err;
+
-+ add_branch = kmalloc(sizeof(*add_branch), GFP_NOFS);
++ err = -ENOMEM;
++ add_branch = kzalloc(sizeof(*add_branch), GFP_NOFS);
+ if (unlikely(!add_branch))
+ goto out;
-+ add_branch->br_wbr = NULL;
++ add_branch->br_xino = au_xino_alloc(/*nfile*/1);
++ if (unlikely(!add_branch->br_xino))
++ goto out_br;
++ err = au_hnotify_init_br(add_branch, perm);
++ if (unlikely(err))
++ goto out_xino;
++
+ if (au_br_writable(perm)) {
-+ add_branch->br_wbr = kmalloc(sizeof(*add_branch->br_wbr),
++ /* may be freed separately at changing the branch permission */
++ add_branch->br_wbr = kzalloc(sizeof(*add_branch->br_wbr),
+ GFP_NOFS);
+ if (unlikely(!add_branch->br_wbr))
-+ goto out_br;
++ goto out_hnotify;
+ }
+
-+ sz = sizeof(*branchp) * (new_nbranch - 1);
-+ if (unlikely(!sz))
-+ sz = sizeof(*branchp);
-+ p = au_sbi(sb)->si_branch;
-+ branchp = au_kzrealloc(p, sz, sizeof(*branchp) * new_nbranch, GFP_NOFS);
-+ if (unlikely(!branchp))
-+ goto out_wbr;
-+ au_sbi(sb)->si_branch = branchp;
-+
-+ sz = sizeof(*hdentryp) * (new_nbranch - 1);
-+ if (unlikely(!sz))
-+ sz = sizeof(*hdentryp);
-+ p = au_di(root)->di_hdentry;
-+ hdentryp = au_kzrealloc(p, sz, sizeof(*hdentryp) * new_nbranch,
-+ GFP_NOFS);
-+ if (unlikely(!hdentryp))
-+ goto out_wbr;
-+ au_di(root)->di_hdentry = hdentryp;
-+
-+ sz = sizeof(*hinodep) * (new_nbranch - 1);
-+ if (unlikely(!sz))
-+ sz = sizeof(*hinodep);
-+ p = au_ii(inode)->ii_hinode;
-+ hinodep = au_kzrealloc(p, sz, sizeof(*hinodep) * new_nbranch, GFP_NOFS);
-+ if (unlikely(!hinodep))
-+ goto out_wbr;
-+ au_ii(inode)->ii_hinode = hinodep;
-+ return add_branch; /* success */
++ if (au_br_fhsm(perm)) {
++ err = au_fhsm_br_alloc(add_branch);
++ if (unlikely(err))
++ goto out_wbr;
++ }
+
-+ out_wbr:
-+ kfree(add_branch->br_wbr);
-+ out_br:
-+ kfree(add_branch);
-+ out:
-+ AuTraceErr(-ENOMEM);
-+ return ERR_PTR(-ENOMEM);
++ 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, /*may_shrink*/0);
++ if (!err) {
++ inode = d_inode(root);
++ err = au_hinode_realloc(au_ii(inode), new_nbranch,
++ /*may_shrink*/0);
++ }
++ if (!err)
++ return add_branch; /* success */
++
++out_wbr:
++ au_kfree_rcu(add_branch->br_wbr);
++out_hnotify:
++ au_hnotify_fin_br(add_branch);
++out_xino:
++ au_xino_put(add_branch);
++out_br:
++ au_kfree_rcu(add_branch);
++out:
++ return ERR_PTR(err);
+}
+
+/*
+ * test if the branch permission is legal or not.
+ */
-+static int test_br(struct super_block *sb, struct inode *inode, int brperm,
-+ char *path)
++static int test_br(struct inode *inode, int brperm, char *path)
+{
+ int err;
+
-+ err = 0;
-+ if (unlikely(au_br_writable(brperm) && IS_RDONLY(inode))) {
-+ AuErr("write permission for readonly fs or inode, %s\n", path);
-+ err = -EINVAL;
-+ }
++ err = (au_br_writable(brperm) && IS_RDONLY(inode));
++ if (!err)
++ goto out;
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ err = -EINVAL;
++ pr_err("write permission for readonly mount or inode, %s\n", path);
+
-+static int au_unsupported_fs(struct super_block *sb)
-+{
-+ return sb->s_magic == PROC_SUPER_MAGIC
-+#ifdef SYSFS_MAGIC
-+ || sb->s_magic == SYSFS_MAGIC
-+#endif
-+ || !strcmp(au_sbtype(sb), "unionfs");
++out:
++ return err;
+}
+
+/*
+static int test_add(struct super_block *sb, struct au_opt_add *add, int remount)
+{
+ int err;
-+ struct dentry *root;
++ aufs_bindex_t bbot, bindex;
++ struct dentry *root, *h_dentry;
+ struct inode *inode, *h_inode;
-+ aufs_bindex_t bend, bindex;
-+
-+ LKTRTrace("%s, remo%d\n", add->path, remount);
+
+ root = sb->s_root;
-+ bend = au_sbend(sb);
-+ if (unlikely(bend >= 0
-+ && au_find_dbindex(root, add->nd.path.dentry) >= 0)) {
++ bbot = au_sbbot(sb);
++ if (unlikely(bbot >= 0
++ && au_find_dbindex(root, add->path.dentry) >= 0)) {
+ err = 1;
+ if (!remount) {
+ err = -EINVAL;
-+ AuErr("%s duplicated\n", add->path);
++ pr_err("%s duplicated\n", add->pathname);
+ }
+ goto out;
+ }
+
+ err = -ENOSPC; /* -E2BIG; */
+ if (unlikely(AUFS_BRANCH_MAX <= add->bindex
-+ || AUFS_BRANCH_MAX - 1 <= bend)) {
-+ AuErr("number of branches exceeded %s\n", add->path);
++ || AUFS_BRANCH_MAX - 1 <= bbot)) {
++ pr_err("number of branches exceeded %s\n", add->pathname);
+ goto out;
+ }
+
+ err = -EDOM;
-+ if (unlikely(add->bindex < 0 || bend + 1 < add->bindex)) {
-+ AuErr("bad index %d\n", add->bindex);
++ if (unlikely(add->bindex < 0 || bbot + 1 < add->bindex)) {
++ pr_err("bad index %d\n", add->bindex);
+ goto out;
+ }
+
-+ inode = add->nd.path.dentry->d_inode;
-+ AuDebugOn(!inode || !S_ISDIR(inode->i_mode));
++ inode = d_inode(add->path.dentry);
+ err = -ENOENT;
+ if (unlikely(!inode->i_nlink)) {
-+ AuErr("no existence %s\n", add->path);
++ pr_err("no existence %s\n", add->pathname);
+ goto out;
+ }
+
+ err = -EINVAL;
+ if (unlikely(inode->i_sb == sb)) {
-+ AuErr("%s must be outside\n", add->path);
++ pr_err("%s must be outside\n", add->pathname);
+ goto out;
+ }
+
-+ if (unlikely(au_test_nested(inode->i_sb))) {
-+ AuErr("nested " AUFS_NAME " %s\n", add->path);
++ if (unlikely(au_test_fs_unsuppoted(inode->i_sb))) {
++ pr_err("unsupported filesystem, %s (%s)\n",
++ add->pathname, au_sbtype(inode->i_sb));
+ goto out;
+ }
+
-+ if (unlikely(au_unsupported_fs(inode->i_sb))) {
-+ AuErr("unsupported filesystem, %s\n", add->path);
++ if (unlikely(inode->i_sb->s_stack_depth)) {
++ pr_err("already stacked, %s (%s)\n",
++ add->pathname, au_sbtype(inode->i_sb));
+ goto out;
+ }
+
-+ if (unlikely(au_test_unsupported_nfs(inode->i_sb))) {
-+ AuErr(AuNoNfsBranchMsg " %s\n", add->path);
-+ goto out;
-+ }
-+
-+ if (unlikely(au_test_unsupported_nfs4(inode->i_sb))) {
-+ AuErr(AuNoNfsv4BranchMsg " %s\n", add->path);
-+ goto out;
-+ }
-+
-+ err = test_br(sb, add->nd.path.dentry->d_inode, add->perm, add->path);
++ err = test_br(d_inode(add->path.dentry), add->perm, add->pathname);
+ if (unlikely(err))
+ goto out;
+
-+ if (bend < 0)
++ if (bbot < 0)
+ return 0; /* success */
+
+ err = -EINVAL;
-+ for (bindex = 0; bindex <= bend; bindex++)
-+ if (unlikely(test_overlap(sb, add->nd.path.dentry,
++ for (bindex = 0; bindex <= bbot; bindex++)
++ if (unlikely(test_overlap(sb, add->path.dentry,
+ au_h_dptr(root, bindex)))) {
-+ AuErr("%s is overlapped\n", add->path);
++ pr_err("%s is overlapped\n", add->pathname);
+ goto out;
+ }
+
+ err = 0;
-+ h_inode = au_h_dptr(root, 0)->d_inode;
-+ if (unlikely(au_opt_test(au_mntflags(sb), WARN_PERM)
-+ && ((h_inode->i_mode & S_IALLUGO)
-+ != (inode->i_mode & S_IALLUGO)
-+ || h_inode->i_uid != inode->i_uid
-+ || h_inode->i_gid != inode->i_gid)))
-+ AuWarn("uid/gid/perm %s %u/%u/0%o, %u/%u/0%o\n",
-+ add->path,
-+ inode->i_uid, inode->i_gid, (inode->i_mode & S_IALLUGO),
-+ h_inode->i_uid, h_inode->i_gid,
-+ (h_inode->i_mode & S_IALLUGO));
-+
-+ out:
-+ AuTraceErr(err);
++ if (au_opt_test(au_mntflags(sb), WARN_PERM)) {
++ h_dentry = au_h_dptr(root, 0);
++ h_inode = d_inode(h_dentry);
++ if ((h_inode->i_mode & S_IALLUGO) != (inode->i_mode & S_IALLUGO)
++ || !uid_eq(h_inode->i_uid, inode->i_uid)
++ || !gid_eq(h_inode->i_gid, inode->i_gid))
++ pr_warn("uid/gid/perm %s %u/%u/0%o, %u/%u/0%o\n",
++ add->pathname,
++ i_uid_read(inode), i_gid_read(inode),
++ (inode->i_mode & S_IALLUGO),
++ i_uid_read(h_inode), i_gid_read(h_inode),
++ (h_inode->i_mode & S_IALLUGO));
++ }
++
++out:
++ return err;
++}
++
++/*
++ * initialize or clean the whiteouts for an adding branch
++ */
++static int au_br_init_wh(struct super_block *sb, struct au_branch *br,
++ int new_perm)
++{
++ int err, old_perm;
++ aufs_bindex_t bindex;
++ struct inode *h_inode;
++ struct au_wbr *wbr;
++ struct au_hinode *hdir;
++ struct dentry *h_dentry;
++
++ err = vfsub_mnt_want_write(au_br_mnt(br));
++ if (unlikely(err))
++ goto out;
++
++ wbr = br->br_wbr;
++ old_perm = br->br_perm;
++ br->br_perm = new_perm;
++ hdir = NULL;
++ h_inode = NULL;
++ bindex = au_br_index(sb, br->br_id);
++ if (0 <= bindex) {
++ hdir = au_hi(d_inode(sb->s_root), bindex);
++ au_hn_inode_lock_nested(hdir, AuLsc_I_PARENT);
++ } else {
++ h_dentry = au_br_dentry(br);
++ h_inode = d_inode(h_dentry);
++ inode_lock_nested(h_inode, AuLsc_I_PARENT);
++ }
++ if (!wbr)
++ err = au_wh_init(br, sb);
++ else {
++ wbr_wh_write_lock(wbr);
++ err = au_wh_init(br, sb);
++ wbr_wh_write_unlock(wbr);
++ }
++ if (hdir)
++ au_hn_inode_unlock(hdir);
++ else
++ inode_unlock(h_inode);
++ vfsub_mnt_drop_write(au_br_mnt(br));
++ br->br_perm = old_perm;
++
++ if (!err && wbr && !au_br_writable(new_perm)) {
++ au_kfree_rcu(wbr);
++ br->br_wbr = NULL;
++ }
++
++out:
+ return err;
+}
+
+static int au_wbr_init(struct au_branch *br, struct super_block *sb,
-+ int perm, struct path *path)
++ int perm)
+{
+ int err;
++ struct kstatfs kst;
+ struct au_wbr *wbr;
+
-+ AuTraceEnter();
+ wbr = br->br_wbr;
-+ AuDebugOn(!wbr);
-+
-+ au_rw_init_nolock(&wbr->wbr_wh_rwsem);
-+ memset(wbr->wbr_wh, 0, sizeof(wbr->wbr_wh));
++ au_rw_init(&wbr->wbr_wh_rwsem);
+ atomic_set(&wbr->wbr_wh_running, 0);
-+ wbr->wbr_bytes = 0;
+
-+ err = au_br_init_wh(sb, /*bindex*/-1, br, perm,
-+ path->dentry, path->mnt);
++ /*
++ * a limit for rmdir/rename a dir
++ * cf. AUFS_MAX_NAMELEN in include/uapi/linux/aufs_type.h
++ */
++ err = vfs_statfs(&br->br_path, &kst);
++ if (unlikely(err))
++ goto out;
++ err = -EINVAL;
++ if (kst.f_namelen >= NAME_MAX)
++ err = au_br_init_wh(sb, br, perm);
++ else
++ pr_err("%pd(%s), unsupported namelen %ld\n",
++ au_br_dentry(br),
++ au_sbtype(au_br_dentry(br)->d_sb), kst.f_namelen);
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
++/* initialize a new branch */
+static int au_br_init(struct au_branch *br, struct super_block *sb,
+ struct au_opt_add *add)
+{
+ int err;
-+ unsigned int mnt_flags;
-+
-+ AuTraceEnter();
++ struct au_branch *brbase;
++ struct file *xf;
++ struct inode *h_inode;
+
+ err = 0;
-+ br->br_mnt = NULL;
-+ br->br_xino.xi_file = NULL;
-+ mutex_init(&br->br_xino.xi_nondir_mtx);
-+ atomic_set(&br->br_xino_running, 0);
-+ atomic_set(&br->br_count, 0);
++ br->br_perm = add->perm;
++ br->br_path = add->path; /* set first, path_get() later */
++ spin_lock_init(&br->br_dykey_lock);
++ au_lcnt_init(&br->br_nfiles, /*release*/NULL);
++ au_lcnt_init(&br->br_count, /*release*/NULL);
++ 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, &add->nd.path);
++ err = au_wbr_init(br, sb, add->perm);
+ if (unlikely(err))
-+ goto out;
++ goto out_err;
+ }
+
-+ br->br_mnt = mntget(add->nd.path.mnt);
-+ mnt_flags = au_mntflags(sb);
-+ if (au_opt_test(mnt_flags, XINO)) {
-+ err = au_xino_br(sb, br, add->nd.path.dentry->d_inode->i_ino,
-+ au_sbr(sb, 0)->br_xino.xi_file, /*do_test*/1);
-+ if (unlikely(err)) {
-+ AuDebugOn(br->br_xino.xi_file);
-+ goto out;
-+ }
-+#if 0 /* reserved for future use */
-+ } else if (au_opt_test(mnt_flags, XINODIR)) {
-+ err = au_xinodir_br(sb, br, add->nd.path.dentry->d_inode->i_ino,
-+ /*do_test*/1);
++ if (au_opt_test(au_mntflags(sb), XINO)) {
++ brbase = au_sbr(sb, 0);
++ xf = au_xino_file(brbase->br_xino, /*idx*/-1);
++ AuDebugOn(!xf);
++ h_inode = d_inode(add->path.dentry);
++ err = au_xino_init_br(sb, br, h_inode->i_ino, &xf->f_path);
+ if (unlikely(err)) {
-+ AuDebugOn(br->br_xino.xi_file);
-+ goto out;
++ AuDebugOn(au_xino_file(br->br_xino, /*idx*/-1));
++ goto out_err;
+ }
-+#endif
+ }
+
-+ br->br_id = au_new_br_id(sb);
-+ br->br_perm = add->perm;
-+ br->br_xino_upper = AUFS_XINO_TRUNC_INIT;
+ sysaufs_br_init(br);
-+ br->br_generation = au_sigen(sb);
-+ /* smp_mb(); */ /* atomic_set */
++ path_get(&br->br_path);
++ goto out; /* success */
+
-+ out:
-+ AuTraceErr(err);
++out_err:
++ memset(&br->br_path, 0, sizeof(br->br_path));
++out:
+ return err;
+}
+
++static void au_br_do_add_brp(struct au_sbinfo *sbinfo, aufs_bindex_t bindex,
++ struct au_branch *br, aufs_bindex_t bbot,
++ aufs_bindex_t amount)
++{
++ struct au_branch **brp;
++
++ AuRwMustWriteLock(&sbinfo->si_rwsem);
++
++ brp = sbinfo->si_branch + bindex;
++ memmove(brp + 1, brp, sizeof(*brp) * amount);
++ *brp = br;
++ sbinfo->si_bbot++;
++ if (unlikely(bbot < 0))
++ sbinfo->si_bbot = 0;
++}
++
++static void au_br_do_add_hdp(struct au_dinfo *dinfo, aufs_bindex_t bindex,
++ aufs_bindex_t bbot, aufs_bindex_t amount)
++{
++ struct au_hdentry *hdp;
++
++ AuRwMustWriteLock(&dinfo->di_rwsem);
++
++ hdp = au_hdentry(dinfo, bindex);
++ memmove(hdp + 1, hdp, sizeof(*hdp) * amount);
++ au_h_dentry_init(hdp);
++ dinfo->di_bbot++;
++ if (unlikely(bbot < 0))
++ dinfo->di_btop = 0;
++}
++
++static void au_br_do_add_hip(struct au_iinfo *iinfo, aufs_bindex_t bindex,
++ aufs_bindex_t bbot, aufs_bindex_t amount)
++{
++ struct au_hinode *hip;
++
++ AuRwMustWriteLock(&iinfo->ii_rwsem);
++
++ hip = au_hinode(iinfo, bindex);
++ memmove(hip + 1, hip, sizeof(*hip) * amount);
++ au_hinode_init(hip);
++ iinfo->ii_bbot++;
++ if (unlikely(bbot < 0))
++ iinfo->ii_btop = 0;
++}
++
++static void au_br_do_add(struct super_block *sb, struct au_branch *br,
++ aufs_bindex_t bindex)
++{
++ struct dentry *root, *h_dentry;
++ struct inode *root_inode, *h_inode;
++ aufs_bindex_t bbot, amount;
++
++ root = sb->s_root;
++ root_inode = d_inode(root);
++ bbot = au_sbbot(sb);
++ amount = bbot + 1 - bindex;
++ h_dentry = au_br_dentry(br);
++ au_sbilist_lock();
++ au_br_do_add_brp(au_sbi(sb), bindex, br, bbot, amount);
++ au_br_do_add_hdp(au_di(root), bindex, bbot, amount);
++ au_br_do_add_hip(au_ii(root_inode), bindex, bbot, amount);
++ au_set_h_dptr(root, bindex, dget(h_dentry));
++ h_inode = d_inode(h_dentry);
++ au_set_h_iptr(root_inode, bindex, au_igrab(h_inode), /*flags*/0);
++ au_sbilist_unlock();
++}
++
+int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount)
+{
-+ int err, amount;
-+ aufs_bindex_t bend, add_bindex;
-+ struct dentry *root, *dentry;
-+ struct au_iinfo *iinfo;
-+ struct au_sbinfo *sbinfo;
-+ struct au_dinfo *dinfo;
++ int err;
++ aufs_bindex_t bbot, add_bindex;
++ struct dentry *root, *h_dentry;
+ struct inode *root_inode;
-+ unsigned long long maxb;
-+ struct au_branch **branchp, *add_branch;
-+ struct au_hdentry *hdentryp;
-+ struct au_hinode *hinodep;
-+
-+ dentry = add->nd.path.dentry;
-+ LKTRTrace("b%d, %s, 0x%x, %.*s\n",
-+ add->bindex, add->path, add->perm, AuDLNPair(dentry));
-+ SiMustWriteLock(sb);
++ struct au_branch *add_branch;
++
+ root = sb->s_root;
-+ DiMustWriteLock(root);
-+ root_inode = root->d_inode;
++ root_inode = d_inode(root);
+ IMustLock(root_inode);
+ IiMustWriteLock(root_inode);
-+
+ err = test_add(sb, add, remount);
+ if (unlikely(err < 0))
+ goto out;
+ goto out; /* success */
+ }
+
-+ bend = au_sbend(sb);
-+ add_branch = alloc_addbr(sb, bend + 2, add->perm);
++ bbot = au_sbbot(sb);
++ add_branch = au_br_alloc(sb, bbot + 2, add->perm);
+ err = PTR_ERR(add_branch);
+ if (IS_ERR(add_branch))
+ goto out;
++
+ err = au_br_init(add_branch, sb, add);
+ if (unlikely(err)) {
+ au_br_do_free(add_branch);
+ }
+
+ add_bindex = add->bindex;
-+ if (remount)
-+ sysaufs_brs_del(sb, add_bindex);
++ sysaufs_brs_del(sb, add_bindex); /* remove successors */
++ au_br_do_add(sb, add_branch, add_bindex);
++ sysaufs_brs_add(sb, add_bindex); /* append successors */
++ dbgaufs_brs_add(sb, add_bindex, /*topdown*/0); /* rename successors */
+
-+ sbinfo = au_sbi(sb);
-+ dinfo = au_di(root);
-+ iinfo = au_ii(root_inode);
-+
-+ amount = bend + 1 - add_bindex;
-+ branchp = sbinfo->si_branch + add_bindex;
-+ memmove(branchp + 1, branchp, sizeof(*branchp) * amount);
-+ *branchp = add_branch;
-+ hdentryp = dinfo->di_hdentry + add_bindex;
-+ memmove(hdentryp + 1, hdentryp, sizeof(*hdentryp) * amount);
-+ au_h_dentry_init(hdentryp);
-+ hinodep = iinfo->ii_hinode + add_bindex;
-+ memmove(hinodep + 1, hinodep, sizeof(*hinodep) * amount);
-+ hinodep->hi_inode = NULL;
-+ au_hin_init(hinodep, NULL);
-+
-+ sbinfo->si_bend++;
-+ dinfo->di_bend++;
-+ iinfo->ii_bend++;
-+ if (unlikely(bend < 0)) {
-+ sbinfo->si_bend = 0;
-+ dinfo->di_bstart = 0;
-+ iinfo->ii_bstart = 0;
-+ }
-+ au_set_h_dptr(root, add_bindex, dget(dentry));
-+ au_set_h_iptr(root_inode, add_bindex, au_igrab(dentry->d_inode), 0);
-+ if (remount)
-+ sysaufs_brs_add(sb, add_bindex);
-+
-+ if (!add_bindex)
++ h_dentry = add->path.dentry;
++ if (!add_bindex) {
+ au_cpup_attr_all(root_inode, /*force*/1);
-+ else
-+ au_add_nlink(root_inode, dentry->d_inode);
-+ maxb = dentry->d_sb->s_maxbytes;
-+ if (sb->s_maxbytes < maxb)
-+ sb->s_maxbytes = maxb;
-+
-+ /* safe d_parent reference */
-+ if (!au_xino_def_br(sbinfo)
-+ && add_branch->br_xino.xi_file
-+ && add_branch->br_xino.xi_file->f_dentry->d_parent == dentry)
-+ au_xino_def_br_set(add_branch, sbinfo);
-+
-+ out:
-+ AuTraceErr(err);
++ sb->s_maxbytes = h_dentry->d_sb->s_maxbytes;
++ } else
++ au_add_nlink(root_inode, d_inode(h_dentry));
++
++out:
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+#define AuVerbose(do_info, fmt, args...) do { \
-+ if (!do_info) \
-+ LKTRTrace(fmt, ##args); \
-+ else \
-+ AuInfo(fmt, ##args); \
++static unsigned long long au_farray_cb(struct super_block *sb, void *a,
++ unsigned long long max __maybe_unused,
++ void *arg)
++{
++ unsigned long long n;
++ struct file **p, *f;
++ struct hlist_bl_head *files;
++ struct hlist_bl_node *pos;
++ struct au_finfo *finfo;
++
++ n = 0;
++ p = a;
++ files = &au_sbi(sb)->si_files;
++ 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)) {
++ get_file(f);
++ *p++ = f;
++ n++;
++ AuDebugOn(n > max);
++ }
++ }
++ hlist_bl_unlock(files);
++
++ return n;
++}
++
++static struct file **au_farray_alloc(struct super_block *sb,
++ unsigned long long *max)
++{
++ struct au_sbinfo *sbi;
++
++ sbi = au_sbi(sb);
++ *max = au_lcnt_read(&sbi->si_nfiles, /*do_rev*/1);
++ return au_array_alloc(max, au_farray_cb, sb, /*arg*/NULL);
++}
++
++static void au_farray_free(struct file **a, unsigned long long max)
++{
++ unsigned long long ull;
++
++ for (ull = 0; ull < max; ull++)
++ if (a[ull])
++ fput(a[ull]);
++ kvfree(a);
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * delete a branch
++ */
++
++/* to show the line number, do not make it inlined function */
++#define AuVerbose(do_info, fmt, ...) do { \
++ if (do_info) \
++ pr_info(fmt, ##__VA_ARGS__); \
+} while (0)
+
++static int au_test_ibusy(struct inode *inode, aufs_bindex_t btop,
++ aufs_bindex_t bbot)
++{
++ return (inode && !S_ISDIR(inode->i_mode)) || btop == bbot;
++}
++
++static int au_test_dbusy(struct dentry *dentry, aufs_bindex_t btop,
++ aufs_bindex_t bbot)
++{
++ return au_test_ibusy(d_inode(dentry), btop, bbot);
++}
++
+/*
+ * test if the branch is deletable or not.
+ */
+static int test_dentry_busy(struct dentry *root, aufs_bindex_t bindex,
-+ au_gen_t sigen)
++ unsigned int sigen, const unsigned int verbose)
+{
+ int err, i, j, ndentry;
++ aufs_bindex_t btop, bbot;
+ struct au_dcsub_pages dpages;
+ struct au_dpage *dpage;
+ struct dentry *d;
-+ aufs_bindex_t bstart, bend;
-+ unsigned char verbose;
-+ struct inode *inode;
-+
-+ LKTRTrace("b%d, gen%d\n", bindex, sigen);
-+ SiMustWriteLock(root->d_sb);
+
+ err = au_dpages_init(&dpages, GFP_NOFS);
+ if (unlikely(err))
+ if (unlikely(err))
+ goto out_dpages;
+
-+ verbose = !!au_opt_test(au_mntflags(root->d_sb), VERBOSE);
+ for (i = 0; !err && i < dpages.ndpage; i++) {
+ dpage = dpages.dpages + i;
+ ndentry = dpage->ndentry;
+ for (j = 0; !err && j < ndentry; j++) {
+ d = dpage->dentries[j];
-+ AuDebugOn(!atomic_read(&d->d_count));
-+ inode = d->d_inode;
-+ AuDebugOn(!inode);
-+ if (au_digen(d) == sigen
-+ && au_iigen(inode) == sigen)
++ AuDebugOn(au_dcount(d) <= 0);
++ if (!au_digen_test(d, sigen)) {
+ di_read_lock_child(d, AuLock_IR);
-+ else {
++ if (unlikely(au_dbrange_test(d))) {
++ di_read_unlock(d, AuLock_IR);
++ continue;
++ }
++ } else {
+ di_write_lock_child(d);
++ if (unlikely(au_dbrange_test(d))) {
++ di_write_unlock(d);
++ continue;
++ }
+ err = au_reval_dpath(d, sigen);
+ if (!err)
+ di_downgrade_lock(d, AuLock_IR);
+ }
+ }
+
-+ bstart = au_dbstart(d);
-+ bend = au_dbend(d);
-+ if (bstart <= bindex
-+ && bindex <= bend
++ /* AuDbgDentry(d); */
++ btop = au_dbtop(d);
++ bbot = au_dbbot(d);
++ if (btop <= bindex
++ && bindex <= bbot
+ && au_h_dptr(d, bindex)
-+ && (!S_ISDIR(d->d_inode->i_mode)
-+ || bstart == bend)) {
++ && au_test_dbusy(d, btop, bbot)) {
+ err = -EBUSY;
-+ AuVerbose(verbose, "busy %.*s\n", AuDLNPair(d));
++ AuVerbose(verbose, "busy %pd\n", d);
++ AuDbgDentry(d);
+ }
+ di_read_unlock(d, AuLock_IR);
+ }
+ }
+
-+ out_dpages:
++out_dpages:
+ au_dpages_free(&dpages);
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
+static int test_inode_busy(struct super_block *sb, aufs_bindex_t bindex,
-+ au_gen_t sigen)
++ unsigned int sigen, const unsigned int verbose)
+{
+ int err;
-+ struct inode *i;
-+ aufs_bindex_t bstart, bend;
-+ unsigned char verbose;
++ unsigned long long max, ull;
++ struct inode *i, **array;
++ aufs_bindex_t btop, bbot;
+
-+ LKTRTrace("b%d, gen%d\n", bindex, sigen);
-+ SiMustWriteLock(sb);
++ array = au_iarray_alloc(sb, &max);
++ err = PTR_ERR(array);
++ if (IS_ERR(array))
++ goto out;
+
+ err = 0;
-+ verbose = !!au_opt_test(au_mntflags(sb), VERBOSE);
-+ list_for_each_entry(i, &sb->s_inodes, i_sb_list) {
-+ AuDebugOn(!atomic_read(&i->i_count));
-+ if (!list_empty(&i->i_dentry))
++ AuDbg("b%d\n", bindex);
++ for (ull = 0; !err && ull < max; ull++) {
++ i = array[ull];
++ if (unlikely(!i))
++ break;
++ if (i->i_ino == AUFS_ROOT_INO)
+ continue;
+
-+ if (au_iigen(i) == sigen)
++ /* AuDbgInode(i); */
++ if (au_iigen(i, NULL) == sigen)
+ ii_read_lock_child(i);
+ else {
+ ii_write_lock_child(i);
+ err = au_refresh_hinode_self(i);
++ au_iigen_dec(i);
+ if (!err)
+ ii_downgrade_lock(i);
+ else {
+ }
+ }
+
-+ bstart = au_ibstart(i);
-+ bend = au_ibend(i);
-+ if (bstart <= bindex
-+ && bindex <= bend
++ btop = au_ibtop(i);
++ bbot = au_ibbot(i);
++ if (btop <= bindex
++ && bindex <= bbot
+ && au_h_iptr(i, bindex)
-+ && (!S_ISDIR(i->i_mode) || bstart == bend)) {
++ && au_test_ibusy(i, btop, bbot)) {
+ err = -EBUSY;
+ AuVerbose(verbose, "busy i%lu\n", i->i_ino);
-+ ii_read_unlock(i);
-+ break;
++ AuDbgInode(i);
+ }
+ ii_read_unlock(i);
+ }
++ au_iarray_free(array, max);
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static int test_children_busy(struct dentry *root, aufs_bindex_t bindex)
++static int test_children_busy(struct dentry *root, aufs_bindex_t bindex,
++ const unsigned int verbose)
+{
+ int err;
-+ au_gen_t sigen;
-+
-+ LKTRTrace("b%d\n", bindex);
-+ SiMustWriteLock(root->d_sb);
-+ DiMustWriteLock(root);
-+ /* dont trust BKL */
-+ AuDebugOn(!kernel_locked());
++ unsigned int sigen;
+
+ sigen = au_sigen(root->d_sb);
+ DiMustNoWaiters(root);
-+ IiMustNoWaiters(root->d_inode);
++ IiMustNoWaiters(d_inode(root));
+ di_write_unlock(root);
-+ err = test_dentry_busy(root, bindex, sigen);
++ err = test_dentry_busy(root, bindex, sigen, verbose);
+ if (!err)
-+ err = test_inode_busy(root->d_sb, bindex, sigen);
++ err = test_inode_busy(root->d_sb, bindex, sigen, verbose);
+ di_write_lock_child(root); /* aufs_write_lock() calls ..._child() */
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount)
++static int test_dir_busy(struct file *file, aufs_bindex_t br_id,
++ struct file **to_free, int *idx)
+{
-+ int err, rerr, i;
-+ aufs_bindex_t bindex, bend, br_id;
-+ unsigned char do_wh, verbose;
-+ struct au_sbinfo *sbinfo;
-+ struct au_dinfo *dinfo;
-+ struct au_iinfo *iinfo;
-+ struct au_branch *br, **brp;
-+ struct au_wbr *wbr;
-+ struct au_hdentry *hdp;
-+ struct au_hinode *hip;
-+
-+ LKTRTrace("%s, %.*s\n", del->path, AuDLNPair(del->h_root));
-+ SiMustWriteLock(sb);
-+ DiMustWriteLock(sb->s_root);
-+ IiMustWriteLock(sb->s_root->d_inode);
++ int err;
++ unsigned char matched, root;
++ aufs_bindex_t bindex, bbot;
++ struct au_fidir *fidir;
++ struct au_hfile *hfile;
+
+ err = 0;
-+ bindex = au_find_dbindex(sb->s_root, del->h_root);
-+ if (bindex < 0) {
-+ if (remount)
-+ goto out; /* success */
-+ err = -ENOENT;
-+ AuErr("%s no such branch\n", del->path);
++ root = IS_ROOT(file->f_path.dentry);
++ if (root) {
++ get_file(file);
++ to_free[*idx] = file;
++ (*idx)++;
+ goto out;
+ }
-+ LKTRTrace("bindex b%d\n", bindex);
+
-+ err = -EBUSY;
-+ verbose = !!au_opt_test(au_mntflags(sb), VERBOSE);
-+ bend = au_sbend(sb);
-+ if (unlikely(!bend)) {
-+ AuVerbose(verbose, "no more branches left\n");
-+ goto out;
-+ }
-+ br = au_sbr(sb, bindex);
-+ if (unlikely(au_br_count(br))) {
-+ AuVerbose(verbose, "%d file(s) opened\n", au_br_count(br));
-+ goto out;
-+ }
++ matched = 0;
++ fidir = au_fi(file)->fi_hdir;
++ AuDebugOn(!fidir);
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file); bindex <= bbot; bindex++) {
++ hfile = fidir->fd_hfile + bindex;
++ if (!hfile->hf_file)
++ continue;
+
-+ wbr = br->br_wbr;
-+ do_wh = wbr && (wbr->wbr_whbase || wbr->wbr_plink || wbr->wbr_tmp);
-+ if (do_wh) {
-+#if 0 /* reserved for future use */
-+ /* remove whiteout base */
-+ err = au_br_init_wh(sb, bindex, br, AuBr_RO, del->h_root,
-+ br->br_mnt);
-+ if (unlikely(err))
-+ goto out;
-+#else
-+ for (i = 0; i < AuBrWh_Last; i++) {
-+ dput(wbr->wbr_wh[i]);
-+ wbr->wbr_wh[i] = NULL;
++ if (hfile->hf_br->br_id == br_id) {
++ matched = 1;
++ break;
+ }
-+#endif
+ }
++ if (matched)
++ err = -EBUSY;
+
-+ err = test_children_busy(sb->s_root, bindex);
-+ if (unlikely(err)) {
-+ if (do_wh)
-+ goto out_wh;
-+ goto out;
-+ }
++out:
++ return err;
++}
+
-+ err = 0;
-+ if (remount)
-+ sysaufs_brs_del(sb, bindex);
-+ sbinfo = au_sbi(sb);
-+ dinfo = au_di(sb->s_root);
-+ iinfo = au_ii(sb->s_root->d_inode);
++static int test_file_busy(struct super_block *sb, aufs_bindex_t br_id,
++ struct file **to_free, int opened)
++{
++ int err, idx;
++ unsigned long long ull, max;
++ aufs_bindex_t btop;
++ struct file *file, **array;
++ struct dentry *root;
++ struct au_hfile *hfile;
+
-+ dput(au_h_dptr(sb->s_root, bindex));
-+ au_hiput(iinfo->ii_hinode + bindex);
-+ br_id = br->br_id;
-+ au_br_do_free(br);
++ array = au_farray_alloc(sb, &max);
++ err = PTR_ERR(array);
++ if (IS_ERR(array))
++ goto out;
+
-+ /* todo: realloc and shrink memory? */
-+ if (bindex < bend) {
-+ const aufs_bindex_t n = bend - bindex;
-+
-+ brp = sbinfo->si_branch + bindex;
-+ memmove(brp, brp + 1, sizeof(*brp) * n);
-+ hdp = dinfo->di_hdentry + bindex;
-+ memmove(hdp, hdp + 1, sizeof(*hdp) * n);
-+ hip = iinfo->ii_hinode + bindex;
-+ memmove(hip, hip + 1, sizeof(*hip) * n);
-+ }
-+ sbinfo->si_branch[0 + bend] = NULL;
-+ dinfo->di_hdentry[0 + bend].hd_dentry = NULL;
-+ iinfo->ii_hinode[0 + bend].hi_inode = NULL;
-+ au_hin_init(iinfo->ii_hinode + bend, NULL);
-+
-+ sbinfo->si_bend--;
-+ dinfo->di_bend--;
-+ iinfo->ii_bend--;
-+ if (remount)
-+ sysaufs_brs_add(sb, bindex);
-+
-+ if (!bindex)
-+ au_cpup_attr_all(sb->s_root->d_inode, /*force*/1);
-+ else
-+ au_sub_nlink(sb->s_root->d_inode, del->h_root->d_inode);
-+ if (au_opt_test(au_mntflags(sb), PLINK))
-+ au_plink_half_refresh(sb, br_id);
++ err = 0;
++ idx = 0;
++ root = sb->s_root;
++ di_write_unlock(root);
++ for (ull = 0; ull < max; ull++) {
++ file = array[ull];
++ if (unlikely(!file))
++ break;
+
-+ if (sb->s_maxbytes == del->h_root->d_sb->s_maxbytes) {
-+ bend--;
-+ sb->s_maxbytes = 0;
-+ for (bindex = 0; bindex <= bend; bindex++) {
-+ unsigned long long maxb;
-+ maxb = au_sbr_sb(sb, bindex)->s_maxbytes;
-+ if (sb->s_maxbytes < maxb)
-+ sb->s_maxbytes = maxb;
-+ }
++ /* AuDbg("%pD\n", file); */
++ fi_read_lock(file);
++ btop = au_fbtop(file);
++ if (!d_is_dir(file->f_path.dentry)) {
++ hfile = &au_fi(file)->fi_htop;
++ if (hfile->hf_br->br_id == br_id)
++ err = -EBUSY;
++ } else
++ err = test_dir_busy(file, br_id, to_free, &idx);
++ fi_read_unlock(file);
++ if (unlikely(err))
++ break;
+ }
++ di_write_lock_child(root);
++ au_farray_free(array, max);
++ AuDebugOn(idx > opened);
+
-+ if (au_xino_def_br(sbinfo) == br)
-+ au_xino_def_br_set(NULL, sbinfo);
-+ goto out; /* success */
-+
-+ out_wh:
-+ /* revert */
-+ rerr = au_br_init_wh(sb, bindex, br, br->br_perm, del->h_root,
-+ br->br_mnt);
-+ if (rerr)
-+ AuWarn("failed re-creating base whiteout, %s. (%d)\n",
-+ del->path, rerr);
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static int do_need_sigen_inc(int a, int b)
++static void br_del_file(struct file **to_free, unsigned long long opened,
++ aufs_bindex_t br_id)
+{
-+ return au_br_whable(a) && !au_br_whable(b);
-+}
++ unsigned long long ull;
++ aufs_bindex_t bindex, btop, bbot, bfound;
++ struct file *file;
++ struct au_fidir *fidir;
++ struct au_hfile *hfile;
+
-+static int need_sigen_inc(int old, int new)
-+{
-+ return do_need_sigen_inc(old, new)
-+ || do_need_sigen_inc(new, old);
-+}
++ for (ull = 0; ull < opened; ull++) {
++ file = to_free[ull];
++ if (unlikely(!file))
++ break;
+
-+static int au_br_mod_files_ro(struct super_block *sb, aufs_bindex_t bindex)
++ /* AuDbg("%pD\n", file); */
++ AuDebugOn(!d_is_dir(file->f_path.dentry));
++ bfound = -1;
++ fidir = au_fi(file)->fi_hdir;
++ AuDebugOn(!fidir);
++ fi_write_lock(file);
++ btop = au_fbtop(file);
++ bbot = au_fbbot_dir(file);
++ for (bindex = btop; bindex <= bbot; bindex++) {
++ hfile = fidir->fd_hfile + bindex;
++ if (!hfile->hf_file)
++ continue;
++
++ if (hfile->hf_br->br_id == br_id) {
++ bfound = bindex;
++ break;
++ }
++ }
++ AuDebugOn(bfound < 0);
++ au_set_h_fptr(file, bfound, NULL);
++ if (bfound == btop) {
++ for (btop++; btop <= bbot; btop++)
++ if (au_hf_dir(file, btop)) {
++ au_set_fbtop(file, btop);
++ break;
++ }
++ }
++ fi_write_unlock(file);
++ }
++}
++
++static void au_br_do_del_brp(struct au_sbinfo *sbinfo,
++ const aufs_bindex_t bindex,
++ const aufs_bindex_t bbot)
+{
-+ int err;
-+ struct file *file, *hf;
++ struct au_branch **brp, **p;
++
++ AuRwMustWriteLock(&sbinfo->si_rwsem);
++
++ brp = sbinfo->si_branch + bindex;
++ if (bindex < bbot)
++ memmove(brp, brp + 1, sizeof(*brp) * (bbot - bindex));
++ sbinfo->si_branch[0 + bbot] = NULL;
++ sbinfo->si_bbot--;
++
++ p = au_krealloc(sbinfo->si_branch, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
++ if (p)
++ sbinfo->si_branch = p;
++ /* harmless error */
++}
++
++static void au_br_do_del_hdp(struct au_dinfo *dinfo, const aufs_bindex_t bindex,
++ const aufs_bindex_t bbot)
++{
++ struct au_hdentry *hdp, *p;
++
++ AuRwMustWriteLock(&dinfo->di_rwsem);
++
++ hdp = au_hdentry(dinfo, bindex);
++ if (bindex < bbot)
++ memmove(hdp, hdp + 1, sizeof(*hdp) * (bbot - bindex));
++ /* au_h_dentry_init(au_hdentry(dinfo, bbot); */
++ dinfo->di_bbot--;
++
++ p = au_krealloc(dinfo->di_hdentry, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
++ if (p)
++ dinfo->di_hdentry = p;
++ /* harmless error */
++}
++
++static void au_br_do_del_hip(struct au_iinfo *iinfo, const aufs_bindex_t bindex,
++ const aufs_bindex_t bbot)
++{
++ struct au_hinode *hip, *p;
++
++ AuRwMustWriteLock(&iinfo->ii_rwsem);
++
++ hip = au_hinode(iinfo, bindex);
++ if (bindex < bbot)
++ memmove(hip, hip + 1, sizeof(*hip) * (bbot - bindex));
++ /* au_hinode_init(au_hinode(iinfo, bbot)); */
++ iinfo->ii_bbot--;
++
++ p = au_krealloc(iinfo->ii_hinode, sizeof(*p) * bbot, AuGFP_SBILIST,
++ /*may_shrink*/1);
++ if (p)
++ iinfo->ii_hinode = p;
++ /* harmless error */
++}
++
++static void au_br_do_del(struct super_block *sb, aufs_bindex_t bindex,
++ struct au_branch *br)
++{
++ aufs_bindex_t bbot;
++ struct au_sbinfo *sbinfo;
++ struct dentry *root, *h_root;
++ struct inode *inode, *h_inode;
++ struct au_hinode *hinode;
+
-+ AuTraceEnter();
+ SiMustWriteLock(sb);
+
-+ /* no need file_list_lock() since sbinfo is locked */
++ root = sb->s_root;
++ inode = d_inode(root);
++ sbinfo = au_sbi(sb);
++ bbot = sbinfo->si_bbot;
++
++ h_root = au_h_dptr(root, bindex);
++ hinode = au_hi(inode, bindex);
++ h_inode = au_igrab(hinode->hi_inode);
++ au_hiput(hinode);
++
++ au_sbilist_lock();
++ au_br_do_del_brp(sbinfo, bindex, bbot);
++ au_br_do_del_hdp(au_di(root), bindex, bbot);
++ 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);
++}
++
++static unsigned long long empty_cb(struct super_block *sb, void *array,
++ unsigned long long max, void *arg)
++{
++ return max;
++}
++
++int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount)
++{
++ int err, rerr, i;
++ unsigned long long opened;
++ unsigned int mnt_flags;
++ aufs_bindex_t bindex, bbot, br_id;
++ unsigned char do_wh, verbose;
++ struct au_branch *br;
++ struct au_wbr *wbr;
++ struct dentry *root;
++ struct file **to_free;
++
+ err = 0;
-+ list_for_each_entry(file, &sb->s_files, f_u.fu_list) {
-+ LKTRTrace("%.*s\n", AuDLNPair(file->f_dentry));
-+ if (!au_test_aufs_file(file))
-+ continue;
++ opened = 0;
++ to_free = NULL;
++ root = sb->s_root;
++ bindex = au_find_dbindex(root, del->h_path.dentry);
++ if (bindex < 0) {
++ if (remount)
++ goto out; /* success */
++ err = -ENOENT;
++ pr_err("%s no such branch\n", del->pathname);
++ goto out;
++ }
++ AuDbg("bindex b%d\n", bindex);
+
-+ fi_read_lock(file);
-+ if (!S_ISREG(file->f_dentry->d_inode->i_mode)
-+ || !(file->f_mode & FMODE_WRITE)
-+ || au_fbstart(file) != bindex) {
-+ FiMustNoWaiters(file);
-+ fi_read_unlock(file);
-+ continue;
++ err = -EBUSY;
++ mnt_flags = au_mntflags(sb);
++ verbose = !!au_opt_test(mnt_flags, VERBOSE);
++ bbot = au_sbbot(sb);
++ if (unlikely(!bbot)) {
++ AuVerbose(verbose, "no more branches left\n");
++ goto out;
++ }
++
++ br = au_sbr(sb, bindex);
++ AuDebugOn(!path_equal(&br->br_path, &del->h_path));
++ if (unlikely(au_lcnt_read(&br->br_count, /*do_rev*/1))) {
++ AuVerbose(verbose, "br %pd2 is busy now\n", del->h_path.dentry);
++ goto out;
++ }
++
++ br_id = br->br_id;
++ opened = au_lcnt_read(&br->br_nfiles, /*do_rev*/1);
++ if (unlikely(opened)) {
++ to_free = au_array_alloc(&opened, empty_cb, sb, NULL);
++ err = PTR_ERR(to_free);
++ if (IS_ERR(to_free))
++ goto out;
++
++ err = test_file_busy(sb, br_id, to_free, opened);
++ if (unlikely(err)) {
++ AuVerbose(verbose, "%llu file(s) opened\n", opened);
++ goto out;
++ }
++ }
++
++ wbr = br->br_wbr;
++ do_wh = wbr && (wbr->wbr_whbase || wbr->wbr_plink || wbr->wbr_orph);
++ if (do_wh) {
++ /* instead of WbrWhMustWriteLock(wbr) */
++ SiMustWriteLock(sb);
++ for (i = 0; i < AuBrWh_Last; i++) {
++ dput(wbr->wbr_wh[i]);
++ wbr->wbr_wh[i] = NULL;
++ }
++ }
++
++ err = test_children_busy(root, bindex, verbose);
++ if (unlikely(err)) {
++ if (do_wh)
++ goto out_wh;
++ goto out;
++ }
++
++ err = 0;
++ if (to_free) {
++ /*
++ * now we confirmed the branch is deletable.
++ * let's free the remaining opened dirs on the branch.
++ */
++ di_write_unlock(root);
++ br_del_file(to_free, opened, br_id);
++ di_write_lock_child(root);
++ }
++
++ sysaufs_brs_del(sb, bindex); /* remove successors */
++ dbgaufs_xino_del(br); /* remove one */
++ au_br_do_del(sb, bindex, br);
++ sysaufs_brs_add(sb, bindex); /* append successors */
++ dbgaufs_brs_add(sb, bindex, /*topdown*/1); /* rename successors */
++
++ if (!bindex) {
++ au_cpup_attr_all(d_inode(root), /*force*/1);
++ sb->s_maxbytes = au_sbr_sb(sb, 0)->s_maxbytes;
++ } else
++ au_sub_nlink(d_inode(root), d_inode(del->h_path.dentry));
++ if (au_opt_test(mnt_flags, PLINK))
++ au_plink_half_refresh(sb, br_id);
++
++ goto out; /* success */
++
++out_wh:
++ /* revert */
++ rerr = au_br_init_wh(sb, br, br->br_perm);
++ if (rerr)
++ pr_warn("failed re-creating base whiteout, %s. (%d)\n",
++ del->pathname, rerr);
++out:
++ if (to_free)
++ au_farray_free(to_free, opened);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static int au_ibusy(struct super_block *sb, struct aufs_ibusy __user *arg)
++{
++ int err;
++ aufs_bindex_t btop, bbot;
++ struct aufs_ibusy ibusy;
++ struct inode *inode, *h_inode;
++
++ err = -EPERM;
++ if (unlikely(!capable(CAP_SYS_ADMIN)))
++ goto out;
++
++ err = copy_from_user(&ibusy, arg, sizeof(ibusy));
++ if (!err)
++ /* VERIFY_WRITE */
++ err = !access_ok(&arg->h_ino, sizeof(arg->h_ino));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ goto out;
++ }
++
++ err = -EINVAL;
++ si_read_lock(sb, AuLock_FLUSH);
++ if (unlikely(ibusy.bindex < 0 || ibusy.bindex > au_sbbot(sb)))
++ goto out_unlock;
++
++ err = 0;
++ ibusy.h_ino = 0; /* invalid */
++ inode = ilookup(sb, ibusy.ino);
++ if (!inode
++ || inode->i_ino == AUFS_ROOT_INO
++ || au_is_bad_inode(inode))
++ goto out_unlock;
++
++ ii_read_lock_child(inode);
++ btop = au_ibtop(inode);
++ bbot = au_ibbot(inode);
++ if (btop <= ibusy.bindex && ibusy.bindex <= bbot) {
++ h_inode = au_h_iptr(inode, ibusy.bindex);
++ if (h_inode && au_test_ibusy(inode, btop, bbot))
++ ibusy.h_ino = h_inode->i_ino;
++ }
++ ii_read_unlock(inode);
++ iput(inode);
++
++out_unlock:
++ si_read_unlock(sb);
++ if (!err) {
++ err = __put_user(ibusy.h_ino, &arg->h_ino);
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
+ }
++ }
++out:
++ return err;
++}
++
++long au_ibusy_ioctl(struct file *file, unsigned long arg)
++{
++ return au_ibusy(file->f_path.dentry->d_sb, (void __user *)arg);
++}
++
++#ifdef CONFIG_COMPAT
++long au_ibusy_compat_ioctl(struct file *file, unsigned long arg)
++{
++ return au_ibusy(file->f_path.dentry->d_sb, compat_ptr(arg));
++}
++#endif
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * change a branch permission
++ */
++
++static void au_warn_ima(void)
++{
++#ifdef CONFIG_IMA
++ /* since it doesn't support mark_files_ro() */
++ AuWarn1("RW -> RO makes IMA to produce wrong message\n");
++#endif
++}
++
++static int do_need_sigen_inc(int a, int b)
++{
++ return au_br_whable(a) && !au_br_whable(b);
++}
++
++static int need_sigen_inc(int old, int new)
++{
++ return do_need_sigen_inc(old, new)
++ || do_need_sigen_inc(new, old);
++}
++
++static int au_br_mod_files_ro(struct super_block *sb, aufs_bindex_t bindex)
++{
++ int err, do_warn;
++ unsigned int mnt_flags;
++ unsigned long long ull, max;
++ aufs_bindex_t br_id;
++ unsigned char verbose, writer;
++ struct file *file, *hf, **array;
++ struct au_hfile *hfile;
++ struct inode *h_inode;
++
++ mnt_flags = au_mntflags(sb);
++ verbose = !!au_opt_test(mnt_flags, VERBOSE);
++
++ array = au_farray_alloc(sb, &max);
++ err = PTR_ERR(array);
++ if (IS_ERR(array))
++ goto out;
++
++ do_warn = 0;
++ br_id = au_sbr_id(sb, bindex);
++ for (ull = 0; ull < max; ull++) {
++ file = array[ull];
++ if (unlikely(!file))
++ break;
+
++ /* AuDbg("%pD\n", file); */
++ fi_read_lock(file);
+ if (unlikely(au_test_mmapped(file))) {
+ err = -EBUSY;
++ AuVerbose(verbose, "mmapped %pD\n", file);
++ AuDbgFile(file);
+ FiMustNoWaiters(file);
+ fi_read_unlock(file);
-+ break;
++ goto out_array;
++ }
++
++ hfile = &au_fi(file)->fi_htop;
++ hf = hfile->hf_file;
++ if (!d_is_reg(file->f_path.dentry)
++ || !(file->f_mode & FMODE_WRITE)
++ || hfile->hf_br->br_id != br_id
++ || !(hf->f_mode & FMODE_WRITE))
++ array[ull] = NULL;
++ else {
++ do_warn = 1;
++ get_file(file);
+ }
+
-+ /* todo: already flushed? */
-+ hf = au_h_fptr(file, au_fbstart(file));
-+ hf->f_flags = au_file_roflags(hf->f_flags);
-+ hf->f_mode &= ~FMODE_WRITE;
-+ put_write_access(hf->f_dentry->d_inode);
+ FiMustNoWaiters(file);
+ fi_read_unlock(file);
++ fput(file);
++ }
++
++ err = 0;
++ if (do_warn)
++ au_warn_ima();
++
++ for (ull = 0; ull < max; ull++) {
++ file = array[ull];
++ if (!file)
++ continue;
++
++ /* todo: already flushed? */
++ /*
++ * fs/super.c:mark_files_ro() is gone, but aufs keeps its
++ * approach which resets f_mode and calls mnt_drop_write() and
++ * file_release_write() for each file, because the branch
++ * attribute in aufs world is totally different from the native
++ * fs rw/ro mode.
++ */
++ /* fi_read_lock(file); */
++ hfile = &au_fi(file)->fi_htop;
++ hf = hfile->hf_file;
++ /* fi_read_unlock(file); */
++ spin_lock(&hf->f_lock);
++ writer = !!(hf->f_mode & FMODE_WRITER);
++ hf->f_mode &= ~(FMODE_WRITE | FMODE_WRITER);
++ spin_unlock(&hf->f_lock);
++ if (writer) {
++ h_inode = file_inode(hf);
++ if (hf->f_mode & FMODE_READ)
++ i_readcount_inc(h_inode);
++ put_write_access(h_inode);
++ __mnt_drop_write(hf->f_path.mnt);
++ }
+ }
+
++out_array:
++ au_farray_free(array, max);
++out:
+ AuTraceErr(err);
+ return err;
+}
+
+int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount,
-+ int *do_update)
++ int *do_refresh)
+{
-+ int err;
-+ struct dentry *root;
++ int err, rerr;
+ aufs_bindex_t bindex;
++ struct dentry *root;
+ struct au_branch *br;
-+ struct inode *hidden_dir;
++ struct au_br_fhsm *bf;
+
-+ LKTRTrace("%s, %.*s, 0x%x\n",
-+ mod->path, AuDLNPair(mod->h_root), mod->perm);
-+ SiMustWriteLock(sb);
+ root = sb->s_root;
-+ DiMustWriteLock(root);
-+ IiMustWriteLock(root->d_inode);
-+
+ bindex = au_find_dbindex(root, mod->h_root);
+ if (bindex < 0) {
+ if (remount)
+ return 0; /* success */
+ err = -ENOENT;
-+ AuErr("%s no such branch\n", mod->path);
++ pr_err("%s no such branch\n", mod->path);
+ goto out;
+ }
-+ LKTRTrace("bindex b%d\n", bindex);
++ AuDbg("bindex b%d\n", bindex);
+
-+ hidden_dir = mod->h_root->d_inode;
-+ err = test_br(sb, hidden_dir, mod->perm, mod->path);
++ err = test_br(d_inode(mod->h_root), mod->perm, mod->path);
+ if (unlikely(err))
+ goto out;
+
+ br = au_sbr(sb, bindex);
++ AuDebugOn(mod->h_root != au_br_dentry(br));
+ if (br->br_perm == mod->perm)
+ return 0; /* success */
+
++ /* pre-allocate for non-fhsm --> fhsm */
++ bf = NULL;
++ if (!au_br_fhsm(br->br_perm) && au_br_fhsm(mod->perm)) {
++ err = au_fhsm_br_alloc(br);
++ if (unlikely(err))
++ goto out;
++ bf = br->br_fhsm;
++ br->br_fhsm = NULL;
++ }
++
+ if (au_br_writable(br->br_perm)) {
-+#if 1
+ /* remove whiteout base */
-+ err = au_br_init_wh(sb, bindex, br, mod->perm, mod->h_root,
-+ br->br_mnt);
++ err = au_br_init_wh(sb, br, mod->perm);
+ if (unlikely(err))
-+ goto out;
-+#else /* reserved for future use */
-+ struct au_wbr *wbr;
-+ wbr = br->wbr;
-+ if (wbr)
-+ for (i = 0; i < AuBrWh_Last; i++) {
-+ dput(wbr->wbr_wh[i]);
-+ wbr->wbr_wh[i] = NULL;
-+ }
-+#endif
++ goto out_bf;
+
+ if (!au_br_writable(mod->perm)) {
+ /* rw --> ro, file might be mmapped */
-+
-+#if 1 /* todo: test more? */
+ DiMustNoWaiters(root);
-+ IiMustNoWaiters(root->d_inode);
++ IiMustNoWaiters(d_inode(root));
+ di_write_unlock(root);
+ err = au_br_mod_files_ro(sb, bindex);
+ /* aufs_write_lock() calls ..._child() */
+ di_write_lock_child(root);
-+#endif
++
++ if (unlikely(err)) {
++ rerr = -ENOMEM;
++ br->br_wbr = kzalloc(sizeof(*br->br_wbr),
++ GFP_NOFS);
++ if (br->br_wbr)
++ rerr = au_wbr_init(br, sb, br->br_perm);
++ if (unlikely(rerr)) {
++ AuIOErr("nested error %d (%d)\n",
++ rerr, err);
++ br->br_perm = mod->perm;
++ }
++ }
+ }
+ } else if (au_br_writable(mod->perm)) {
+ /* ro --> rw */
+ err = -ENOMEM;
-+ br->br_wbr = kmalloc(sizeof(*br->br_wbr), GFP_NOFS);
++ br->br_wbr = kzalloc(sizeof(*br->br_wbr), GFP_NOFS);
+ if (br->br_wbr) {
-+ struct path path = {
-+ .mnt = br->br_mnt,
-+ .dentry = mod->h_root
-+ };
-+
-+ err = au_wbr_init(br, sb, mod->perm, &path);
++ err = au_wbr_init(br, sb, mod->perm);
+ if (unlikely(err)) {
-+ kfree(br->br_wbr);
++ au_kfree_rcu(br->br_wbr);
+ br->br_wbr = NULL;
+ }
+ }
+ }
++ if (unlikely(err))
++ goto out_bf;
++
++ if (au_br_fhsm(br->br_perm)) {
++ if (!au_br_fhsm(mod->perm)) {
++ /* fhsm --> non-fhsm */
++ au_br_fhsm_fin(br->br_fhsm);
++ au_kfree_rcu(br->br_fhsm);
++ br->br_fhsm = NULL;
++ }
++ } else if (au_br_fhsm(mod->perm))
++ /* non-fhsm --> fhsm */
++ br->br_fhsm = bf;
++
++ *do_refresh |= need_sigen_inc(br->br_perm, mod->perm);
++ br->br_perm = mod->perm;
++ goto out; /* success */
++
++out_bf:
++ au_kfree_try_rcu(bf);
++out:
++ AuTraceErr(err);
++ return err;
++}
+
++/* ---------------------------------------------------------------------- */
++
++int au_br_stfs(struct au_branch *br, struct aufs_stfs *stfs)
++{
++ int err;
++ struct kstatfs kstfs;
++
++ err = vfs_statfs(&br->br_path, &kstfs);
+ if (!err) {
-+ *do_update |= need_sigen_inc(br->br_perm, mod->perm);
-+ br->br_perm = mod->perm;
++ stfs->f_blocks = kstfs.f_blocks;
++ stfs->f_bavail = kstfs.f_bavail;
++ stfs->f_files = kstfs.f_files;
++ stfs->f_ffree = kstfs.f_ffree;
+ }
+
-+ out:
-+ AuTraceErr(err);
+ return err;
+}
-diff -urN linux/fs/aufs/branch.h linux-aufs/fs/aufs/branch.h
---- linux/fs/aufs/branch.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/branch.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,437 @@
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,375 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * branch filesystems and xino for them
-+ *
-+ * $Id$
+ */
+
+#ifndef __AUFS_BRANCH_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/fs.h>
+#include <linux/mount.h>
-+#include <linux/sysfs.h>
-+#include <linux/aufs_type.h>
-+#include "misc.h"
++#include "dirren.h"
++#include "dynop.h"
++#include "lcnt.h"
++#include "rwsem.h"
+#include "super.h"
+
+/* ---------------------------------------------------------------------- */
+
-+/* an entry in a xino file */
-+struct au_xino_entry {
-+ ino_t ino;
-+ /* __u32 h_gen; */ /* reserved for future use */
-+} __packed;
-+
-+/* reserved for future use */
-+/* #define AuXino_INVALID_HGEN (-1) */
-+
+/* a xino file */
-+struct au_xino_file {
-+ struct file *xi_file;
-+ struct mutex xi_nondir_mtx;
-+
-+ /* reserved for future use */
-+#if 0
++struct au_xino {
+ struct file **xi_file;
++ unsigned int xi_nfile;
+
-+ /* array management */
-+ unsigned long long xi_limit; /* Max xino file size */
-+ unsigned long long xi_size; /* s_maxbytes */
++ struct {
++ spinlock_t spin;
++ ino_t *array;
++ int total;
++ /* reserved for future use */
++ /* unsigned long *bitmap; */
++ wait_queue_head_t wqh;
++ } xi_nondir;
++
++ struct mutex xi_mtx; /* protects xi_file array */
++ struct hlist_bl_head xi_writing;
+
-+ /* truncation */
-+ unsigned long long xi_upper; /* watermark in bytes */
-+ unsigned long long xi_step; /* to next watermark in bytes */
++ atomic_t xi_truncating;
++
++ struct kref xi_kref;
++};
+
-+ /* truncation */
-+ blkcnt_t xi_upper; /* watermark in blocks */
-+ atomic_t xi_running;
++/* File-based Hierarchical Storage Management */
++struct au_br_fhsm {
++#ifdef CONFIG_AUFS_FHSM
++ struct mutex bf_lock;
++ unsigned long bf_jiffy;
++ struct aufs_stfs bf_stfs;
++ int bf_readable;
+#endif
+};
+
+/* members for writable branch only */
-+enum {AuBrWh_BASE, AuBrWh_PLINK, AuBrWh_TMP, AuBrWh_Last};
++enum {AuBrWh_BASE, AuBrWh_PLINK, AuBrWh_ORPH, AuBrWh_Last};
+struct au_wbr {
+ struct au_rwsem wbr_wh_rwsem;
+ struct dentry *wbr_wh[AuBrWh_Last];
-+ atomic_t wbr_wh_running;
++ atomic_t wbr_wh_running;
+#define wbr_whbase wbr_wh[AuBrWh_BASE] /* whiteout base */
+#define wbr_plink wbr_wh[AuBrWh_PLINK] /* pseudo-link dir */
-+#define wbr_tmp wbr_wh[AuBrWh_TMP] /* temporary dir */
++#define wbr_orph wbr_wh[AuBrWh_ORPH] /* dir for orphans */
+
+ /* mfs mode */
+ unsigned long long wbr_bytes;
+};
+
++/* ext2 has 3 types of operations at least, ext3 has 4 */
++#define AuBrDynOp (AuDyLast * 4)
++
++#ifdef CONFIG_AUFS_HFSNOTIFY
++/* support for asynchronous destruction */
++struct au_br_hfsnotify {
++ struct fsnotify_group *hfsn_group;
++};
++#endif
++
++/* sysfs entries */
++struct au_brsysfs {
++ char name[16];
++ struct attribute attr;
++};
++
++enum {
++ AuBrSysfs_BR,
++ AuBrSysfs_BRID,
++ AuBrSysfs_Last
++};
++
+/* protected by superblock rwsem */
+struct au_branch {
-+ struct au_xino_file br_xino;
++ struct au_xino *br_xino;
+
+ aufs_bindex_t br_id;
+
+ int br_perm;
-+ struct vfsmount *br_mnt;
-+ atomic_t br_count;
++ struct path br_path;
++ spinlock_t br_dykey_lock;
++ struct au_dykey *br_dykey[AuBrDynOp];
++ au_lcnt_t br_nfiles; /* opened files */
++ au_lcnt_t br_count; /* in-use for other */
+
+ struct au_wbr *br_wbr;
++ struct au_br_fhsm *br_fhsm;
+
-+#if 1 /* reserved for future use */
-+ /* xino truncation */
-+ blkcnt_t br_xino_upper; /* watermark in blocks */
-+ atomic_t br_xino_running;
++#ifdef CONFIG_AUFS_HFSNOTIFY
++ struct au_br_hfsnotify *br_hfsn;
+#endif
+
+#ifdef CONFIG_SYSFS
-+ /* an entry under sysfs per mount-point */
-+ char br_name[8];
-+ struct attribute br_attr;
++ /* entries under sysfs per mount-point */
++ struct au_brsysfs br_sysfs[AuBrSysfs_Last];
++#endif
++
++#ifdef CONFIG_DEBUG_FS
++ struct dentry *br_dbgaufs; /* xino */
+#endif
+
-+ au_gen_t br_generation;
++ struct au_dr_br br_dirren;
+};
+
+/* ---------------------------------------------------------------------- */
+
-+/* branch permission and attribute */
-+enum {
-+ AuBrPerm_RW, /* writable, linkable wh */
-+ AuBrPerm_RO, /* readonly, no wh */
-+ AuBrPerm_RR, /* natively readonly, no wh */
-+
-+ AuBrPerm_RWNoLinkWH, /* un-linkable whiteouts */
-+
-+ AuBrPerm_ROWH,
-+ AuBrPerm_RRWH, /* whiteout-able */
++static inline struct vfsmount *au_br_mnt(struct au_branch *br)
++{
++ return br->br_path.mnt;
++}
+
-+ AuBrPerm_Last
-+};
++static inline struct dentry *au_br_dentry(struct au_branch *br)
++{
++ return br->br_path.dentry;
++}
+
-+static inline int au_br_writable(int brperm)
++static inline struct mnt_idmap *au_br_idmap(struct au_branch *br)
+{
-+ return brperm == AuBrPerm_RW || brperm == AuBrPerm_RWNoLinkWH;
++ return mnt_idmap(br->br_path.mnt);
+}
+
-+static inline int au_br_whable(int brperm)
++static inline struct super_block *au_br_sb(struct au_branch *br)
+{
-+ return brperm == AuBrPerm_RW
-+ || brperm == AuBrPerm_ROWH
-+ || brperm == AuBrPerm_RRWH;
++ return au_br_mnt(br)->mnt_sb;
+}
+
-+#if 0 /* reserved for future use */
-+static inline int au_br_linkable_wh(int brperm)
++static inline int au_br_rdonly(struct au_branch *br)
+{
-+ return brperm == AuBrPerm_RW;
++ return (sb_rdonly(au_br_sb(br))
++ || !au_br_writable(br->br_perm))
++ ? -EROFS : 0;
+}
-+#endif
+
-+static inline int au_br_hinotifyable(int brperm)
++static inline int au_br_hnotifyable(int brperm __maybe_unused)
+{
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ return brperm != AuBrPerm_RR && brperm != AuBrPerm_RRWH;
++#ifdef CONFIG_AUFS_HNOTIFY
++ return !(brperm & AuBrPerm_RR);
+#else
+ return 0;
+#endif
+}
+
++static inline int au_br_test_oflag(int oflag, struct au_branch *br)
++{
++ int err, exec_flag;
++
++ err = 0;
++ exec_flag = oflag & __FMODE_EXEC;
++ if (unlikely(exec_flag && path_noexec(&br->br_path)))
++ err = -EACCES;
++
++ return err;
++}
++
++static inline void au_xino_get(struct au_branch *br)
++{
++ struct au_xino *xi;
++
++ xi = br->br_xino;
++ if (xi)
++ kref_get(&xi->xi_kref);
++}
++
++static inline int au_xino_count(struct au_branch *br)
++{
++ int v;
++ struct au_xino *xi;
++
++ v = 0;
++ xi = br->br_xino;
++ if (xi)
++ v = kref_read(&xi->xi_kref);
++
++ return v;
++}
++
+/* ---------------------------------------------------------------------- */
+
+/* branch.c */
+struct au_sbinfo;
+void au_br_free(struct au_sbinfo *sinfo);
-+int au_test_def_rr(struct super_block *h_sb);
+int au_br_index(struct super_block *sb, aufs_bindex_t br_id);
+struct au_opt_add;
+int au_br_add(struct super_block *sb, struct au_opt_add *add, int remount);
+struct au_opt_del;
+int au_br_del(struct super_block *sb, struct au_opt_del *del, int remount);
++long au_ibusy_ioctl(struct file *file, unsigned long arg);
++#ifdef CONFIG_COMPAT
++long au_ibusy_compat_ioctl(struct file *file, unsigned long arg);
++#endif
+struct au_opt_mod;
+int au_br_mod(struct super_block *sb, struct au_opt_mod *mod, int remount,
-+ int *do_update);
++ int *do_refresh);
++struct aufs_stfs;
++int au_br_stfs(struct au_branch *br, struct aufs_stfs *stfs);
+
+/* xino.c */
-+#define Au_LOFF_MAX ((loff_t)LLONG_MAX)
-+int au_xib_trunc(struct super_block *sb);
-+ssize_t xino_fread(au_readf_t func, struct file *file, void *buf, size_t size,
-+ loff_t *pos);
-+ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size,
-+ loff_t *pos);
-+struct file *au_xino_create(struct super_block *sb, char *fname, int silent);
-+struct file *au_xino_create2(struct super_block *sb, struct file *base_file,
++static const loff_t au_loff_max = LLONG_MAX;
++
++aufs_bindex_t au_xi_root(struct super_block *sb, struct dentry *dentry);
++struct file *au_xino_create(struct super_block *sb, char *fpath, int silent,
++ int wbrtop);
++struct file *au_xino_create2(struct super_block *sb, struct path *base,
+ struct file *copy_src);
-+ino_t au_xino_new_ino(struct super_block *sb);
-+int au_xino_write0(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ ino_t ino);
-+int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ struct au_xino_entry *xinoe);
++struct au_xi_new {
++ struct au_xino *xi; /* switch between xino and xigen */
++ int idx;
++ struct path *base;
++ struct file *copy_src;
++};
++struct file *au_xi_new(struct super_block *sb, struct au_xi_new *xinew);
++
+int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ struct au_xino_entry *xinoe);
-+int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t hino,
-+ struct file *base_file, int do_test);
-+int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex);
++ ino_t *ino);
++int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ ino_t ino);
++ssize_t xino_fread(struct file *file, void *buf, size_t size, loff_t *pos);
++ssize_t xino_fwrite(struct file *file, void *buf, size_t size, loff_t *pos);
++
++int au_xib_trunc(struct super_block *sb);
++int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex, int idx_begin);
++
++struct au_xino *au_xino_alloc(unsigned int nfile);
++int au_xino_put(struct au_branch *br);
++struct file *au_xino_file1(struct au_xino *xi);
+
+struct au_opt_xino;
-+int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount);
+void au_xino_clr(struct super_block *sb);
++int au_xino_set(struct super_block *sb, struct au_opt_xino *xiopt, int remount);
+struct file *au_xino_def(struct super_block *sb);
++int au_xino_init_br(struct super_block *sb, struct au_branch *br, ino_t hino,
++ struct path *base);
+
-+struct au_opt_xinodir;
-+#if 0 /* def CONFIG_AUFS_EXPORT */ /* reserved for future use */
-+/* export.c */
-+int au_xinodir_br(struct super_block *sb, struct au_branch *br, ino_t hino,
-+ int do_test);
-+int au_xinodir_set(struct super_block *sb, struct au_opt_xinodir *xinodir,
-+ int remount);
-+#else
-+static inline
-+int au_xinodir_br(struct super_block *sb, struct au_branch *br, ino_t hino,
-+ int do_test)
-+{
-+ return 0;
-+}
++ino_t au_xino_new_ino(struct super_block *sb);
++void au_xino_delete_inode(struct inode *inode, const int unlinked);
+
-+static inline
-+int au_xinodir_set(struct super_block *sb, struct au_opt_xinodir *xinodir,
-+ int remount)
-+{
-+ return 0;
-+}
-+#endif
++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);
+
-+/* ---------------------------------------------------------------------- */
++int au_xino_path(struct seq_file *seq, struct file *file);
+
-+/* todo: memory barrier? */
-+static inline int au_br_count(struct au_branch *br)
-+{
-+ return atomic_read(&br->br_count);
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline int au_br_get(struct au_branch *br)
++/* @idx is signed to accept -1 meaning the first file */
++static inline struct file *au_xino_file(struct au_xino *xi, int idx)
+{
-+ return atomic_inc_return(&br->br_count);
-+}
++ struct file *file;
+
-+static inline int au_br_put(struct au_branch *br)
-+{
-+ return atomic_dec_return(&br->br_count);
-+}
++ file = NULL;
++ if (!xi)
++ goto out;
+
-+static inline au_gen_t au_br_gen(struct au_branch *br)
-+{
-+ return br->br_generation;
-+}
++ if (idx >= 0) {
++ if (idx < xi->xi_nfile)
++ file = xi->xi_file[idx];
++ } else
++ file = au_xino_file1(xi);
+
-+/*
-+ * test if the @br is readonly or not.
-+ */
-+static inline int au_br_rdonly(struct au_branch *br)
-+{
-+ return ((br->br_mnt->mnt_sb->s_flags & MS_RDONLY)
-+ || !au_br_writable(br->br_perm))
-+ ? -EROFS : 0;
++out:
++ return file;
+}
+
+/* ---------------------------------------------------------------------- */
+static inline
+struct vfsmount *au_sbr_mnt(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ return au_sbr(sb, bindex)->br_mnt;
++ return au_br_mnt(au_sbr(sb, bindex));
+}
+
+static inline
-+struct super_block *au_sbr_sb(struct super_block *sb, aufs_bindex_t bindex)
-+{
-+ return au_sbr_mnt(sb, bindex)->mnt_sb;
-+}
-+
-+#if 0 /* reserved for future use */
-+static inline int au_sbr_count(struct super_block *sb, aufs_bindex_t bindex)
++struct mnt_idmap *au_sbr_idmap(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ return au_br_count(au_sbr(sb, bindex));
++ return au_br_idmap(au_sbr(sb, bindex));
+}
+
-+static inline void au_sbr_get(struct super_block *sb, aufs_bindex_t bindex)
-+{
-+ au_br_get(au_sbr(sb, bindex));
-+}
-+#endif
-+
-+static inline void au_sbr_put(struct super_block *sb, aufs_bindex_t bindex)
++static inline
++struct super_block *au_sbr_sb(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ au_br_put(au_sbr(sb, bindex));
++ return au_br_sb(au_sbr(sb, bindex));
+}
+
+static inline int au_sbr_perm(struct super_block *sb, aufs_bindex_t bindex)
+ return au_br_whable(au_sbr_perm(sb, bindex));
+}
+
-+static inline int au_br_want_write(struct au_branch *br)
-+{
-+ int err;
-+
-+ AuDebugOn(!au_br_writable(br->br_perm));
-+ err = au_mnt_want_write(br->br_mnt);
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+static inline void au_br_drop_write(struct au_branch *br)
-+{
-+ AuDebugOn(!au_br_writable(br->br_perm));
-+ au_mnt_drop_write(br->br_mnt);
-+}
-+
-+static inline int au_test_trunc_xino(struct super_block *sb)
-+{
-+ return au_test_tmpfs(sb);
-+}
-+
-+/* temporary support for i#1 in cramfs */
-+static inline int au_test_unique_ino(struct dentry *h_dentry, ino_t h_ino)
-+{
-+#if defined(CONFIG_CRAMFS) || defined(CONFIG_CRAMFS_MODULE)
-+ if (unlikely(h_dentry->d_sb->s_magic == CRAMFS_MAGIC))
-+ return h_ino != 1;
-+#endif
-+ return 1;
-+}
-+
-+static inline struct vfsmount *au_do_nfsmnt(struct vfsmount *h_mnt)
-+{
-+ if (!au_test_nfs(h_mnt->mnt_sb)
-+ && !au_test_ecryptfs(h_mnt->mnt_sb))
-+ return NULL;
-+ return h_mnt;
-+}
-+
-+static inline void au_br_nfs_lockdep_off(struct super_block *sb)
-+{
-+ if (au_test_nfs(sb))
-+ lockdep_off();
-+}
-+
-+static inline void au_br_nfs_lockdep_on(struct super_block *sb)
-+{
-+ /* hoping this condition will be optimized... */
-+ if (au_test_nfs(sb))
-+ lockdep_on();
-+}
++/* ---------------------------------------------------------------------- */
+
-+#ifdef CONFIG_AUFS_BR_NFS
-+static inline int au_test_unsupported_nfs(struct super_block *h_sb)
-+{
-+ return 0;
-+}
++#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)
++/*
++#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)
++*/
+
-+/* it doesn't mntget() */
-+static inline
-+struct vfsmount *au_nfsmnt(struct super_block *sb, aufs_bindex_t bindex)
-+{
-+ return au_do_nfsmnt(au_sbr_mnt(sb, bindex));
-+}
++#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 AuNoNfsBranchMsg "dummy"
++#define WbrWhMustNoWaiters(wbr) AuRwMustNoWaiters(&(wbr)->wbr_wh_rwsem)
++#define WbrWhMustAnyLock(wbr) AuRwMustAnyLock(&(wbr)->wbr_wh_rwsem)
++#define WbrWhMustWriteLock(wbr) AuRwMustWriteLock(&(wbr)->wbr_wh_rwsem)
+
-+#else
-+static inline int au_test_unsupported_nfs(struct super_block *h_sb)
-+{
-+ return h_sb->s_magic == NFS_SUPER_MAGIC;
-+}
++/* ---------------------------------------------------------------------- */
+
-+/* it doesn't mntget() */
-+static inline
-+struct vfsmount *au_nfsmnt(struct super_block *sb, aufs_bindex_t bindex)
++#ifdef CONFIG_AUFS_FHSM
++static inline void au_br_fhsm_init(struct au_br_fhsm *brfhsm)
+{
-+ return NULL;
++ mutex_init(&brfhsm->bf_lock);
++ brfhsm->bf_jiffy = 0;
++ brfhsm->bf_readable = 0;
+}
+
-+#define AuNoNfsBranchMsg "NFS branch is not supported" \
-+ ", try some configurations and patches included in aufs source CVS."
-+
-+#endif /* CONFIG_AUFS_BR_NFS */
-+
-+#ifdef CONFIG_AUFS_BR_NFS_V4
-+static inline int au_test_unsupported_nfs4(struct super_block *h_sb)
++static inline void au_br_fhsm_fin(struct au_br_fhsm *brfhsm)
+{
-+ return 0;
++ mutex_destroy(&brfhsm->bf_lock);
+}
-+
-+#define AuNoNfsv4BranchMsg "dummy"
-+
+#else
-+static inline int au_test_unsupported_nfs4(struct super_block *h_sb)
-+{
-+ return h_sb->s_magic == NFS_SUPER_MAGIC
-+ && !strcmp(h_sb->s_type->name, "nfs4");
-+}
-+
-+#define AuNoNfsv4BranchMsg "NFSv4 branch is not supported" \
-+ ", try some configurations and patches included in aufs source CVS."
-+
-+#endif /* CONFIG_AUFS_BR_NFS_v4 */
-+
-+/* support atomic open */
-+static inline int au_test_fs_intent(struct super_block *h_sb)
-+{
-+ return au_test_nfs4(h_sb) /*|| au_test_fuse(h_sb)*/;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * br_wh_read_lock, br_wh_write_lock
-+ * br_wh_read_unlock, br_wh_write_unlock, br_wh_downgrade_lock
-+ */
-+AuSimpleRwsemFuncs(wbr_wh, struct au_wbr *wbr, wbr->wbr_wh_rwsem);
-+
-+/* to debug easier, do not make them inlined functions */
-+#define WbrWhMustReadLock(wbr) do { \
-+ /* SiMustAnyLock(sb); */ \
-+ AuRwMustReadLock(&(wbr)->wbr_wh_rwsem); \
-+} while (0)
-+
-+#define WbrWhMustWriteLock(wbr) do { \
-+ /* SiMustAnyLock(sb); */ \
-+ AuRwMustWriteLock(&(wbr)->wbr_wh_rwsem); \
-+} while (0)
-+
-+#define WbrWhMustAnyLock(br) do { \
-+ /* SiMustAnyLock(sb); */ \
-+ AuRwMustAnyLock(&(wbr)->wbr_wh_rwsem); \
-+} while (0)
++AuStubVoid(au_br_fhsm_init, struct au_br_fhsm *brfhsm)
++AuStubVoid(au_br_fhsm_fin, struct au_br_fhsm *brfhsm)
++#endif
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_BRANCH_H__ */
-diff -urN linux/fs/aufs/br_fuse.c linux-aufs/fs/aufs/br_fuse.c
---- linux/fs/aufs/br_fuse.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/br_fuse.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,80 @@
+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 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,40 @@
++# SPDX-License-Identifier: GPL-2.0
++
++AuConfStr = CONFIG_AUFS_FS=${CONFIG_AUFS_FS}
++
++define AuConf
++ifdef ${1}
++AuConfStr += ${1}=${${1}}
++endif
++endef
++
++AuConfAll = BRANCH_MAX_127 BRANCH_MAX_511 BRANCH_MAX_1023 BRANCH_MAX_32767 \
++ SBILIST \
++ HNOTIFY HFSNOTIFY \
++ EXPORT INO_T_64 \
++ XATTR \
++ FHSM \
++ RDU \
++ DIRREN \
++ SHWH \
++ BR_RAMFS \
++ BR_FUSE POLL \
++ BR_HFSPLUS \
++ BDEV_LOOP \
++ DEBUG MAGIC_SYSRQ
++$(foreach i, ${AuConfAll}, \
++ $(eval $(call AuConf,CONFIG_AUFS_${i})))
++
++AuConfName = ${obj}/conf.str
++${AuConfName}.tmp: FORCE
++ @echo ${AuConfStr} | tr ' ' '\n' | sed -e 's/^/"/' -e 's/$$/\\n"/' > $@
++${AuConfName}: ${AuConfName}.tmp
++ @diff -q $< $@ > /dev/null 2>&1 || { \
++ echo ' GEN ' $@; \
++ cp -p $< $@; \
++ }
++FORCE:
++clean-files += ${AuConfName} ${AuConfName}.tmp
++${obj}/sysfs.o: ${AuConfName}
++
++-include ${srctree}/${src}/conf_priv.mk
+diff -urN /usr/share/empty/fs/aufs/cpup.c linux/fs/aufs/cpup.c
+--- /usr/share/empty/fs/aufs/cpup.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/cpup.c 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,1459 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * special handling for inode attributes on FUSE branch
-+ *
-+ * $Id$
++ * copy-up functions, see wbr_policy.c for copy-down
+ */
+
++#include <linux/fs_stack.h>
++#include <linux/mm.h>
++#include <linux/task_work.h>
+#include "aufs.h"
+
-+/* h_mnt can be NULL, is it safe? */
-+int au_update_fuse_h_inode(struct vfsmount *h_mnt, struct dentry *h_dentry)
++void au_cpup_attr_flags(struct inode *dst, unsigned int iflags)
+{
-+ int err;
-+ struct kstat st;
++ const unsigned int mask = S_DEAD | S_SWAPFILE | S_PRIVATE
++ | S_NOATIME | S_NOCMTIME | S_AUTOMOUNT;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
++ BUILD_BUG_ON(sizeof(iflags) != sizeof(dst->i_flags));
+
-+ err = 0;
-+ if (h_dentry->d_inode
-+ /* && atomic_read(&h_dentry->d_inode->i_count) */
-+ && au_test_fuse(h_dentry->d_sb)) {
-+ err = vfsub_getattr(h_mnt, h_dentry, &st, /*dlgt*/0);
-+ if (unlikely(err)) {
-+ AuDbg("err %d\n", err);
-+ au_debug_on();
-+ AuDbgDentry(h_dentry);
-+ au_debug_off();
-+ WARN_ON(err);
-+ }
-+ }
-+ return err;
++ dst->i_flags |= iflags & ~mask;
++ if (au_test_fs_notime(dst->i_sb))
++ dst->i_flags |= S_NOATIME | S_NOCMTIME;
+}
+
-+#if 0 /* temp */
-+/*
-+ * This function was born after a discussion with the FUSE developer.
-+ * The inode attributes on a filesystem who defines i_op->getattr()
-+ * is unreliable since such fs may not maintain the attributes at lookup.
-+ * This function doesn't want the result of stat, instead wants the side-effect
-+ * which refreshes the attributes.
-+ * Hmm, there seems to be no such filesystem except fuse.
-+ */
-+int vfsub_i_attr(struct vfsmount *mnt, struct dentry *dentry, int dlgt)
++void au_cpup_attr_timesizes(struct inode *inode)
+{
-+ int err;
-+ struct inode *inode;
-+ struct inode_operations *op;
-+ struct kstat st;
-+
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode);
++ struct inode *h_inode;
+
-+ err = 0;
-+ op = inode->i_op;
-+ if (op && op->getattr && !au_test_aufs(dentry->d_sb)) {
-+ err = security_inode_getattr(mnt, dentry);
-+ if (!err)
-+ err = op->getattr(mnt, dentry, &st);
-+ }
-+ AuTraceErr(err);
-+ return err;
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
++ fsstack_copy_attr_times(inode, h_inode);
++ fsstack_copy_inode_size(inode, h_inode);
+}
-+#endif
-diff -urN linux/fs/aufs/br_nfs.c linux-aufs/fs/aufs/br_nfs.c
---- linux/fs/aufs/br_nfs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/br_nfs.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,356 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * lookup functions for NFS branch in linux-2.6.19 and later
-+ *
-+ * $Id$
-+ */
-+
-+#include "aufs.h"
+
-+static struct file *au_find_h_intent(struct au_hdentry *hd, struct file *file)
++void au_cpup_attr_nlink(struct inode *inode, int force)
+{
-+ struct file *h_file, *hf;
-+ struct au_hdintent *hdi, *tmp, *do_free;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(hd->hd_dentry));
++ struct inode *h_inode;
++ struct super_block *sb;
++ aufs_bindex_t bindex, bbot;
+
-+ h_file = NULL;
-+ do_free = NULL;
-+ spin_lock(&hd->hd_lock);
-+ list_for_each_entry_safe(hdi, tmp, hd->hd_intent_list, hdi_list) {
-+ hf = hdi->hdi_file[AuIntent_BRANCH];
-+ if (hdi->hdi_file[AuIntent_AUFS] == file
-+ && hf->f_dentry == hd->hd_dentry) {
-+ h_file = hf;
-+ do_free = hdi;
-+ list_del(&hdi->hdi_list);
-+ break;
-+ }
-+ }
-+ spin_unlock(&hd->hd_lock);
-+ kfree(do_free);
++ sb = inode->i_sb;
++ bindex = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, bindex);
++ if (!force
++ && !S_ISDIR(h_inode->i_mode)
++ && au_opt_test(au_mntflags(sb), PLINK)
++ && au_plink_test(inode))
++ return;
+
-+ return h_file;
-+}
-+
-+struct file *au_h_intent(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *file)
-+{
-+ struct file *h_file;
-+ struct au_hdentry *hd = au_di(dentry)->di_hdentry + bindex;
-+
-+ LKTRTrace("%.*s, b%d, f %p\n", AuDLNPair(dentry), bindex, file);
-+ DiMustAnyLock(dentry);
-+ AuDebugOn(bindex < au_di(dentry)->di_bstart
-+ || bindex > au_di(dentry)->di_bend);
-+
-+ h_file = NULL;
-+ if (!hd->hd_intent_list || !file)
-+ return h_file; /* success */
-+
-+ /* AuDebugOn(au_test_wkq(current)); */
-+ h_file = au_find_h_intent(hd, file);
-+ return h_file;
-+}
-+
-+static int au_set_h_intent(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *file, struct file *h_file)
-+{
-+ int err;
-+ struct au_hdentry *hd = au_di(dentry)->di_hdentry + bindex;
-+ struct au_hdintent *hdi;
-+ struct file *hf;
-+
-+ LKTRTrace("%.*s, b%d, f %p\n", AuDLNPair(dentry), bindex, file);
-+ /* d_revalidate() holds read_lock */
-+ /* DiMustWriteLock(dentry); */
-+ AuDebugOn(bindex < au_di(dentry)->di_bstart
-+ || bindex > au_di(dentry)->di_bend
-+ || !file
-+ || !h_file
-+ /* || au_test_wkq(current) */);
-+
-+ err = -ENOMEM;
-+ if (hd->hd_intent_list) {
-+ while (1) {
-+ hf = au_find_h_intent(hd, file);
-+ if (!hf)
-+ break;
-+ fput(hf);
-+ AuWarn("freed hfile %.*s b%d left\n",
-+ AuDLNPair(dentry), bindex);
-+ }
-+ } else {
-+ spin_lock(&hd->hd_lock);
-+ if (!hd->hd_intent_list) {
-+ hd->hd_intent_list
-+ = kmalloc(sizeof(*hd->hd_intent_list),
-+ GFP_ATOMIC);
-+ if (unlikely(!hd->hd_intent_list)) {
-+ spin_unlock(&hd->hd_lock);
-+ goto out;
-+ }
-+ INIT_LIST_HEAD(hd->hd_intent_list);
-+ }
-+ spin_unlock(&hd->hd_lock);
-+ }
-+
-+ hdi = kmalloc(sizeof(*hdi), GFP_NOFS);
-+ if (unlikely(!hdi))
-+ goto out;
-+
-+ err = 0;
-+ /* hdi->hdi_pid = current->pid; */
-+ hdi->hdi_file[AuIntent_AUFS] = file;
-+ hdi->hdi_file[AuIntent_BRANCH] = h_file;
-+ spin_lock(&hd->hd_lock);
-+ list_add(&hdi->hdi_list, hd->hd_intent_list);
-+ spin_unlock(&hd->hd_lock);
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+int au_br_nfs_h_intent(struct file *nd_file, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct nameidata *nd)
-+{
-+ int err;
-+
-+ AuTraceEnter();
-+
-+ err = 0;
-+ if (!nd_file)
-+ goto out;
-+
-+ AuDebugOn(!nd);
-+ err = au_set_h_intent(dentry, bindex, nd->intent.open.file, nd_file);
-+ if (unlikely(err)) {
-+ fput(nd_file);
-+ au_set_h_dptr(dentry, bindex, NULL);
-+ /* todo: update bstart and bend? */
-+ }
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_hintent_put(struct au_hdentry *hd, int do_free)
-+{
-+ struct au_hdintent *hdi, *tmp;
-+ struct file *hf;
-+
-+ if (hd->hd_intent_list) {
-+ /* no spin lock */
-+ list_for_each_entry_safe(hdi, tmp, hd->hd_intent_list,
-+ hdi_list) {
-+ LKTRTrace("hdi %p\n", hdi);
-+ hf = hdi->hdi_file[AuIntent_BRANCH];
-+ if (hf)
-+ fput(hf);
-+ /* list_del(&hdi->hdi_list); */
-+ kfree(hdi);
-+ }
-+ if (do_free)
-+ kfree(hd->hd_intent_list);
-+ }
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+int au_fake_intent(/* struct au_ndsub *save, */struct nameidata *nd,
-+ struct au_branch *br)
-+{
-+ int err;
-+
-+ LKTRTrace("perm %d\n", br->br_perm);
-+
-+ err = 0;
-+ nd->intent.open.file = NULL;
-+ if (nd->flags & LOOKUP_OPEN) {
-+ if (au_test_fs_intent(br->br_mnt->mnt_sb)) {
-+ err = -ENFILE;
-+ nd->intent.open.file = get_empty_filp();
-+ if (unlikely(!nd->intent.open.file))
-+ goto out;
-+ err = 0;
-+ }
-+ if (!au_br_writable(br->br_perm)) {
-+ nd->intent.open.flags = FMODE_READ
-+ | au_file_roflags(nd->intent.open.flags);
-+ nd->flags &= ~LOOKUP_CREATE;
-+ }
-+ }
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+static void au_put_filp(struct file *file)
-+{
-+#if !defined(CONFIG_AUFS_MODULE) || defined(CONFIG_AUFS_PUT_FILP_PATCH)
-+ if (unlikely(file))
-+ put_filp(file);
-+#else
-+ WARN_ONCE(file, "unexpected put_fillp() call");
-+#endif
-+}
-+
-+int au_hin_after_reval(struct nameidata *nd, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct file *file)
-+{
-+ int err;
-+
-+ LKTRTrace("nd %p, %.*s, b%d, f %d\n",
-+ nd, AuDLNPair(dentry), bindex, !!file);
-+
-+ err = 0;
-+ if ((nd->flags & LOOKUP_OPEN)
-+ && nd->intent.open.file
-+ && !IS_ERR(nd->intent.open.file)) {
-+ if (nd->intent.open.file->f_dentry) {
-+ err = au_set_h_intent(dentry, bindex, file,
-+ nd->intent.open.file);
-+ if (!err)
-+ nd->intent.open.file = NULL;
-+ }
-+ au_put_filp(nd->intent.open.file);
-+ }
-+
-+ return err;
-+}
-+
-+#ifdef CONFIG_AUFS_DLGT
-+struct au_lookup_hash_args {
-+ struct dentry **errp;
-+ struct qstr *name;
-+ struct dentry *base;
-+ struct nameidata *nd;
-+};
-+
-+static void au_call_lookup_hash(void *args)
-+{
-+ struct au_lookup_hash_args *a = args;
-+ *a->errp = vfsub__lookup_hash(a->name, a->base, a->nd);
-+}
-+
-+static struct dentry *
-+au_lkup_hash_dlgt(struct qstr *this, struct dentry *parent,
-+ struct nameidata *nd, unsigned int flags)
-+{
-+ struct dentry *dentry;
-+ int dirperm1;
-+
-+ dirperm1 = au_ftest_ndx(flags, DIRPERM1);
-+ if (!dirperm1 && !au_ftest_ndx(flags, DLGT))
-+ dentry = vfsub__lookup_hash(this, parent, nd);
-+ else {
-+ int wkq_err;
-+ struct au_lookup_hash_args args = {
-+ .errp = &dentry,
-+ .name = this,
-+ .base = parent,
-+ .nd = nd
-+ };
-+ wkq_err = au_wkq_wait(au_call_lookup_hash, &args,
-+ /*dlgt*/!dirperm1);
-+ if (unlikely(wkq_err))
-+ dentry = ERR_PTR(wkq_err);
-+ }
-+
-+ AuTraceErrPtr(dentry);
-+ return dentry;
-+}
-+#else
-+static struct dentry *
-+au_lkup_hash_dlgt(struct qstr *this, struct dentry *parent,
-+ struct nameidata *nd, unsigned int flags)
-+{
-+ return vfsub__lookup_hash(this, parent, nd);
-+}
-+#endif /* CONFIG_AUFS_DLGT */
-+
-+struct dentry *au_lkup_hash(const char *name, struct dentry *parent,
-+ int len, struct au_ndx *ndx)
-+{
-+ struct dentry *dentry;
-+ char *p;
-+ unsigned long hash;
-+ struct qstr this;
-+ unsigned int c;
-+ struct nameidata tmp_nd, *ndo;
-+ int err;
-+
-+ LKTRTrace("%.*s/%.*s\n", AuDLNPair(parent), len, name);
-+
-+ /* todo: export and call __lookup_one_len() in fs/namei.c? */
-+ dentry = ERR_PTR(-EACCES);
-+ this.name = name;
-+ this.len = len;
-+ if (unlikely(!len))
-+ goto out;
-+
-+ p = (void *)name;
-+ hash = init_name_hash();
-+ while (len--) {
-+ c = *p++;
-+ if (unlikely(c == '/' || c == '\0'))
-+ goto out;
-+ hash = partial_name_hash(c, hash);
-+ }
-+ this.hash = end_name_hash(hash);
-+
-+ ndo = ndx->nd;
-+ if (ndo) {
-+ tmp_nd = *ndo;
-+ err = au_fake_intent(&tmp_nd, ndx->br);
-+ dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_intent;
-+ } else
-+ memset(&tmp_nd, 0, sizeof(tmp_nd));
-+
-+ tmp_nd.path.dentry = parent;
-+ tmp_nd.path.mnt = ndx->nfsmnt;
-+ path_get(&tmp_nd.path);
-+ dentry = au_lkup_hash_dlgt(&this, parent, &tmp_nd, ndx->flags);
-+ if (!IS_ERR(dentry)) {
-+ /* why negative dentry for a new dir was unhashed? */
-+ if (unlikely(d_unhashed(dentry)))
-+ d_rehash(dentry);
-+ if (tmp_nd.intent.open.file
-+ && tmp_nd.intent.open.file->f_dentry) {
-+ ndx->nd_file = tmp_nd.intent.open.file;
-+ tmp_nd.intent.open.file = NULL;
-+ /* au_br_get(ndx->br); */
-+ }
-+ }
-+ path_put(&tmp_nd.path);
-+
-+ out_intent:
-+ au_put_filp(tmp_nd.intent.open.file);
-+ out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
-+}
-diff -urN linux/fs/aufs/br_xfs.c linux-aufs/fs/aufs/br_xfs.c
---- linux/fs/aufs/br_xfs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/br_xfs.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,69 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * special handling inode attributes on XFS branch in linux-2.6.24 and later
-+ *
-+ * $Id$
-+ */
-+
-+#include "aufs.h"
-+
-+/* h_mnt can be NULL, is it safe? */
-+dev_t au_h_rdev(struct inode *h_inode, struct vfsmount *h_mnt,
-+ struct dentry *h_dentry)
-+{
-+ dev_t rdev;
-+ int err;
-+ struct kstat st;
-+
-+ LKTRTrace("hi%lu\n", h_inode->i_ino);
-+ if (h_dentry)
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
-+
-+ rdev = h_inode->i_rdev;
-+ if (!rdev || !au_test_xfs(h_inode->i_sb))
-+ goto out;
-+
-+ rdev = 0;
-+ if (!h_dentry) {
-+ err = 0;
-+ h_dentry = d_find_alias(h_inode);
-+ if (unlikely(!h_dentry))
-+ goto failure;
-+ err = PTR_ERR(h_dentry);
-+ if (IS_ERR(h_dentry)) {
-+ h_dentry = NULL;
-+ goto failure;
-+ }
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
-+ } else
-+ dget(h_dentry);
-+
-+ err = vfsub_getattr(h_mnt, h_dentry, &st, /*dlgt*/0);
-+ dput(h_dentry);
-+ if (!err) {
-+ rdev = st.rdev;
-+ goto out; /* success */
-+ }
-+
-+ failure:
-+ AuIOErr("failed rdev for XFS inode, hi%lu, %d\n", h_inode->i_ino, err);
-+ out:
-+ return rdev;
-+}
-diff -urN linux/fs/aufs/cpup.c linux-aufs/fs/aufs/cpup.c
---- linux/fs/aufs/cpup.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/cpup.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1132 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * copy-up functions, see wbr_policy.c for copy-down
-+ *
-+ * $Id$
-+ */
-+
-+#include <linux/fs_stack.h>
-+#include <linux/uaccess.h>
-+#include "aufs.h"
-+
-+/* todo? violent cpup_attr_*() functions don't care inode lock */
-+
-+void au_cpup_attr_timesizes(struct inode *inode)
-+{
-+ struct inode *h_inode;
-+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ /* todo? IMustLock(inode); */
-+ h_inode = au_h_iptr(inode, au_ibstart(inode));
-+ AuDebugOn(!h_inode);
-+ /* todo? IMustLock(!h_inode); */
-+
-+ fsstack_copy_attr_times(inode, h_inode);
-+ vfsub_copy_inode_size(inode, h_inode);
-+}
-+
-+void au_cpup_attr_nlink(struct inode *inode, int force)
-+{
-+ struct inode *h_inode;
-+ struct super_block *sb;
-+ aufs_bindex_t bindex, bend;
-+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ /* todo? IMustLock(inode); */
-+ AuDebugOn(!inode->i_mode);
-+
-+ sb = inode->i_sb;
-+ bindex = au_ibstart(inode);
-+ h_inode = au_h_iptr(inode, bindex);
-+
-+ if (!force
-+ && !S_ISDIR(h_inode->i_mode)
-+ && au_opt_test(au_mntflags(sb), PLINK)
-+ && au_plink_test(sb, inode))
-+ return;
-+
-+ inode->i_nlink = h_inode->i_nlink;
++ /*
++ * 0 can happen in revalidating.
++ * h_inode->i_mutex may not be held here, but it is harmless since once
++ * i_nlink reaches 0, it will never become positive except O_TMPFILE
++ * case.
++ * todo: O_TMPFILE+linkat(AT_SYMLINK_FOLLOW) bypassing aufs may cause
++ * the incorrect link count.
++ */
++ set_nlink(inode, h_inode->i_nlink);
+
+ /*
+ * fewer nlink makes find(1) noisy, but larger nlink doesn't.
+ * it may includes whplink directory.
+ */
+ if (S_ISDIR(h_inode->i_mode)) {
-+ bend = au_ibend(inode);
-+ for (bindex++; bindex <= bend; bindex++) {
++ bbot = au_ibbot(inode);
++ for (bindex++; bindex <= bbot; bindex++) {
+ h_inode = au_h_iptr(inode, bindex);
+ if (h_inode)
+ au_add_nlink(inode, h_inode);
+{
+ struct inode *h_inode;
+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ /* todo? IMustLock(inode); */
-+ h_inode = au_h_iptr(inode, au_ibstart(inode));
-+ AuDebugOn(!h_inode);
-+
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
+ inode->i_mode = h_inode->i_mode;
+ inode->i_uid = h_inode->i_uid;
+ inode->i_gid = h_inode->i_gid;
+ au_cpup_attr_timesizes(inode);
-+ au_cpup_attr_flags(inode, h_inode);
++ au_cpup_attr_flags(inode, h_inode->i_flags);
+}
+
+void au_cpup_igen(struct inode *inode, struct inode *h_inode)
+{
+ struct au_iinfo *iinfo = au_ii(inode);
++
++ IiMustWriteLock(inode);
++
+ iinfo->ii_higen = h_inode->i_generation;
+ iinfo->ii_hsb1 = h_inode->i_sb;
+}
+{
+ struct inode *h_inode;
+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ /* todo? IMustLock(inode); */
-+ h_inode = au_h_iptr(inode, au_ibstart(inode));
-+ AuDebugOn(!h_inode);
-+
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
+ au_cpup_attr_changeable(inode);
+ if (inode->i_nlink > 0)
+ au_cpup_attr_nlink(inode, force);
-+
-+ switch (inode->i_mode & S_IFMT) {
-+ case S_IFBLK:
-+ case S_IFCHR:
-+ inode->i_rdev = au_h_rdev(h_inode, /*h_mnt*/NULL,
-+ /*h_dentry*/NULL);
-+ }
++ inode->i_rdev = h_inode->i_rdev;
+ inode->i_blkbits = h_inode->i_blkbits;
+ au_cpup_igen(inode, h_inode);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+/* Note: dt_dentry and dt_hidden_dentry are not dget/dput-ed */
++/* Note: dt_dentry and dt_h_dentry are not dget/dput-ed */
+
+/* keep the timestamps of the parent dir when cpup */
+void au_dtime_store(struct au_dtime *dt, struct dentry *dentry,
-+ struct dentry *h_dentry, struct au_hinode *hinode,
-+ struct au_hinode *hdir)
++ struct path *h_path)
+{
+ struct inode *h_inode;
+
-+ LKTRTrace("%.*s, hdir %d\n", AuDLNPair(dentry), !!hdir);
-+ AuDebugOn(!dentry || !h_dentry || !h_dentry->d_inode);
-+
+ dt->dt_dentry = dentry;
-+ dt->dt_h_dentry = h_dentry;
-+ dt->dt_hinode = hinode;
-+ dt->dt_hdir = hdir;
-+ h_inode = h_dentry->d_inode;
++ dt->dt_h_path = *h_path;
++ h_inode = d_inode(h_path->dentry);
+ dt->dt_atime = h_inode->i_atime;
+ dt->dt_mtime = h_inode->i_mtime;
+ /* smp_mb(); */
+{
+ struct iattr attr;
+ int err;
-+ struct au_hin_ignore ign[2];
-+ struct vfsub_args vargs;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dt->dt_dentry));
+
+ attr.ia_atime = dt->dt_atime;
+ attr.ia_mtime = dt->dt_mtime;
+ attr.ia_valid = ATTR_FORCE | ATTR_MTIME | ATTR_MTIME_SET
+ | ATTR_ATIME | ATTR_ATIME_SET;
+
-+ vfsub_args_init(&vargs, ign,
-+ au_test_dlgt(au_mntflags(dt->dt_dentry->d_sb)), 0);
-+ /*
-+ * IN_ATTRIB should be divided into
-+ * IN_ATTRIB_ATIME, IN_ATTRIB_MTIME ...,
-+ * and define all ORed new IN_ATTRIB macro.
-+ */
-+ vfsub_ign_hinode(&vargs, IN_ATTRIB, dt->dt_hinode);
-+ vfsub_ign_hinode(&vargs, IN_ATTRIB, dt->dt_hdir);
-+ err = vfsub_notify_change(dt->dt_h_dentry, &attr, &vargs);
++ /* no delegation since this is a directory */
++ err = vfsub_notify_change(&dt->dt_h_path, &attr, /*delegated*/NULL);
+ if (unlikely(err))
-+ AuWarn("restoring timestamps failed(%d). ignored\n", err);
++ pr_warn("restoring timestamps failed(%d). ignored\n", err);
+}
+
+/* ---------------------------------------------------------------------- */
+
++/* internal use only */
++struct au_cpup_reg_attr {
++ int valid;
++ struct kstat st;
++ unsigned int iflags; /* inode->i_flags */
++};
++
+static noinline_for_stack
-+int cpup_iattr(struct dentry *dst, aufs_bindex_t bindex, struct dentry *h_src,
-+ struct au_hinode *hdir, struct vfsub_args *vargs)
++int cpup_iattr(struct dentry *dst, aufs_bindex_t bindex, struct path *h_src,
++ struct au_cpup_reg_attr *h_src_attr)
+{
-+ int err, sbits;
-+ struct dentry *h_dst;
++ int err, sbits, icex;
++ unsigned int mnt_flags;
++ unsigned char verbose;
+ struct iattr ia;
++ struct path h_path;
+ struct inode *h_isrc, *h_idst;
++ struct kstat *h_st;
++ struct au_branch *br;
+
-+ h_dst = au_h_dptr(dst, bindex);
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dst));
-+ h_idst = h_dst->d_inode;
-+ /* todo? IMustLock(h_idst); */
-+ h_isrc = h_src->d_inode;
-+ /* todo? IMustLock(h_isrc); */
-+
-+ ia.ia_valid = ATTR_FORCE | ATTR_MODE | ATTR_UID | ATTR_GID
++ br = au_sbr(dst->d_sb, bindex);
++ h_path.mnt = au_br_mnt(br);
++ h_path.dentry = au_h_dptr(dst, bindex);
++ h_idst = d_inode(h_path.dentry);
++ h_isrc = d_inode(h_src->dentry);
++ ia.ia_valid = ATTR_FORCE | ATTR_UID | ATTR_GID
+ | ATTR_ATIME | ATTR_MTIME
+ | ATTR_ATIME_SET | ATTR_MTIME_SET;
-+ ia.ia_mode = h_isrc->i_mode;
-+ ia.ia_uid = h_isrc->i_uid;
-+ ia.ia_gid = h_isrc->i_gid;
-+ ia.ia_atime = h_isrc->i_atime;
-+ ia.ia_mtime = h_isrc->i_mtime;
-+ sbits = !!(ia.ia_mode & (S_ISUID | S_ISGID));
-+ au_cpup_attr_flags(h_idst, h_isrc);
-+
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_ATTRIB, hdir);
-+ err = vfsub_notify_change(h_dst, &ia, vargs);
++ if (h_src_attr && h_src_attr->valid) {
++ h_st = &h_src_attr->st;
++ ia.ia_uid = h_st->uid;
++ ia.ia_gid = h_st->gid;
++ ia.ia_atime = h_st->atime;
++ ia.ia_mtime = h_st->mtime;
++ if (h_idst->i_mode != h_st->mode
++ && !S_ISLNK(h_idst->i_mode)) {
++ ia.ia_valid |= ATTR_MODE;
++ ia.ia_mode = h_st->mode;
++ }
++ sbits = !!(h_st->mode & (S_ISUID | S_ISGID));
++ au_cpup_attr_flags(h_idst, h_src_attr->iflags);
++ } else {
++ ia.ia_uid = h_isrc->i_uid;
++ ia.ia_gid = h_isrc->i_gid;
++ ia.ia_atime = h_isrc->i_atime;
++ ia.ia_mtime = h_isrc->i_mtime;
++ if (h_idst->i_mode != h_isrc->i_mode
++ && !S_ISLNK(h_idst->i_mode)) {
++ ia.ia_valid |= ATTR_MODE;
++ ia.ia_mode = h_isrc->i_mode;
++ }
++ sbits = !!(h_isrc->i_mode & (S_ISUID | S_ISGID));
++ au_cpup_attr_flags(h_idst, h_isrc->i_flags);
++ }
++ /* no delegation since it is just created */
++ err = vfsub_notify_change(&h_path, &ia, /*delegated*/NULL);
+
+ /* is this nfs only? */
-+ if (!err && sbits && au_test_nfs(h_dst->d_sb)) {
++ if (!err && sbits && au_test_nfs(h_path.dentry->d_sb)) {
+ ia.ia_valid = ATTR_FORCE | ATTR_MODE;
+ ia.ia_mode = h_isrc->i_mode;
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_ATTRIB, hdir);
-+ err = vfsub_notify_change(h_dst, &ia, vargs);
++ err = vfsub_notify_change(&h_path, &ia, /*delegated*/NULL);
++ }
++
++ icex = br->br_perm & AuBrAttr_ICEX;
++ if (!err) {
++ mnt_flags = au_mntflags(dst->d_sb);
++ verbose = !!au_opt_test(mnt_flags, VERBOSE);
++ err = au_cpup_xattr(&h_path, h_src, icex, verbose);
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+/*
-+ * to support a sparse file which is opened with O_APPEND,
-+ * we need to close the file.
++/* ---------------------------------------------------------------------- */
++
++static int au_do_copy_file(struct file *dst, struct file *src, loff_t len,
++ char *buf, unsigned long blksize)
++{
++ int err;
++ size_t sz, rbytes, wbytes;
++ unsigned char all_zero;
++ char *p, *zp;
++ struct inode *h_inode;
++ /* reduce stack usage */
++ struct iattr *ia;
++
++ zp = page_address(ZERO_PAGE(0));
++ if (unlikely(!zp))
++ return -ENOMEM; /* possible? */
++
++ err = 0;
++ all_zero = 0;
++ while (len) {
++ AuDbg("len %lld\n", len);
++ sz = blksize;
++ if (len < blksize)
++ sz = len;
++
++ rbytes = 0;
++ /* todo: signal_pending? */
++ while (!rbytes || err == -EAGAIN || err == -EINTR) {
++ rbytes = vfsub_read_k(src, buf, sz, &src->f_pos);
++ err = rbytes;
++ }
++ if (unlikely(err < 0))
++ break;
++
++ all_zero = 0;
++ if (len >= rbytes && rbytes == blksize)
++ all_zero = !memcmp(buf, zp, rbytes);
++ if (!all_zero) {
++ wbytes = rbytes;
++ p = buf;
++ while (wbytes) {
++ size_t b;
++
++ b = vfsub_write_k(dst, p, wbytes, &dst->f_pos);
++ err = b;
++ /* todo: signal_pending? */
++ if (unlikely(err == -EAGAIN || err == -EINTR))
++ continue;
++ if (unlikely(err < 0))
++ break;
++ wbytes -= b;
++ p += b;
++ }
++ if (unlikely(err < 0))
++ break;
++ } else {
++ loff_t res;
++
++ AuLabel(hole);
++ res = vfsub_llseek(dst, rbytes, SEEK_CUR);
++ err = res;
++ if (unlikely(res < 0))
++ break;
++ }
++ len -= rbytes;
++ err = 0;
++ }
++
++ /* the last block may be a hole */
++ if (!err && all_zero) {
++ AuLabel(last hole);
++
++ err = 1;
++ if (au_test_nfs(dst->f_path.dentry->d_sb)) {
++ /* nfs requires this step to make last hole */
++ /* is this only nfs? */
++ do {
++ /* todo: signal_pending? */
++ err = vfsub_write_k(dst, "\0", 1, &dst->f_pos);
++ } while (err == -EAGAIN || err == -EINTR);
++ if (err == 1)
++ dst->f_pos--;
++ }
++
++ if (err == 1) {
++ ia = (void *)buf;
++ ia->ia_size = dst->f_pos;
++ ia->ia_valid = ATTR_SIZE | ATTR_FILE;
++ ia->ia_file = dst;
++ h_inode = file_inode(dst);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD2);
++ /* no delegation since it is just created */
++ err = vfsub_notify_change(&dst->f_path, ia,
++ /*delegated*/NULL);
++ inode_unlock(h_inode);
++ }
++ }
++
++ return err;
++}
++
++int au_copy_file(struct file *dst, struct file *src, loff_t len)
++{
++ int err;
++ unsigned long blksize;
++ unsigned char do_kfree;
++ char *buf;
++ struct super_block *h_sb;
++
++ err = -ENOMEM;
++ h_sb = file_inode(dst)->i_sb;
++ blksize = h_sb->s_blocksize;
++ if (!blksize || PAGE_SIZE < blksize)
++ blksize = PAGE_SIZE;
++ AuDbg("blksize %lu\n", blksize);
++ do_kfree = (blksize != PAGE_SIZE && blksize >= sizeof(struct iattr *));
++ if (do_kfree)
++ buf = kmalloc(blksize, GFP_NOFS);
++ else
++ buf = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!buf))
++ goto out;
++
++ if (len > (1 << 22))
++ AuDbg("copying a large file %lld\n", (long long)len);
++
++ src->f_pos = 0;
++ dst->f_pos = 0;
++ err = au_do_copy_file(dst, src, len, buf, blksize);
++ if (do_kfree) {
++ AuDebugOn(!au_kfree_do_sz_test(blksize));
++ au_kfree_do_rcu(buf);
++ } else
++ free_page((unsigned long)buf);
++
++out:
++ return err;
++}
++
++static int au_do_copy(struct file *dst, struct file *src, loff_t len)
++{
++ int err;
++ struct super_block *h_src_sb;
++ struct inode *h_src_inode;
++
++ h_src_inode = file_inode(src);
++ h_src_sb = h_src_inode->i_sb;
++
++ /* XFS acquires inode_lock */
++ if (!au_test_xfs(h_src_sb))
++ err = au_copy_file(dst, src, len);
++ else {
++ inode_unlock_shared(h_src_inode);
++ err = au_copy_file(dst, src, len);
++ 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;
++ loff_t lo;
++ 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->remap_file_range) {
++ err = au_do_copy(dst, src, len);
++ goto out;
++ }
++
++ if (!au_test_nfs(h_src_sb)) {
++ inode_unlock_shared(h_src_inode);
++ lo = vfsub_clone_file_range(src, dst, len);
++ inode_lock_shared_nested(h_src_inode, AuLsc_I_CHILD);
++ } else
++ lo = vfsub_clone_file_range(src, dst, len);
++ if (lo == len) {
++ err = 0;
++ goto out; /* success */
++ } else if (lo >= 0)
++ /* todo: possible? */
++ /* paritially succeeded */
++ AuDbg("lo %lld, len %lld. Retrying.\n", lo, len);
++ else if (lo != -EOPNOTSUPP) {
++ /* older XFS has a condition in cloning */
++ err = lo;
++ 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.
+ */
-+static noinline_for_stack
-+int cpup_regular(struct dentry *dentry, aufs_bindex_t bdst, aufs_bindex_t bsrc,
-+ loff_t len, struct au_hinode *hdir, struct vfsub_args *vargs)
++static int au_cp_regular(struct au_cp_generic *cpg)
+{
+ int err, i;
-+ struct super_block *sb;
-+ struct inode *h_inode;
+ enum { SRC, DST };
+ struct {
+ aufs_bindex_t bindex;
+ unsigned int flags;
+ struct dentry *dentry;
++ int force_wr;
+ struct file *file;
-+ void *label, *label_file;
-+ } *h, hidden[] = {
++ } *f, file[] = {
+ {
-+ .bindex = bsrc,
++ .bindex = cpg->bsrc,
+ .flags = O_RDONLY | O_NOATIME | O_LARGEFILE,
-+ .file = NULL,
-+ .label = &&out,
-+ .label_file = &&out_src_file
+ },
+ {
-+ .bindex = bdst,
++ .bindex = cpg->bdst,
+ .flags = O_WRONLY | O_NOATIME | O_LARGEFILE,
-+ .file = NULL,
-+ .label = &&out_src_file,
-+ .label_file = &&out_dst_file
++ .force_wr = !!au_ftest_cpup(cpg->flags, RWDST),
+ }
+ };
++ struct au_branch *br;
++ struct super_block *sb, *h_src_sb;
++ struct inode *h_src_inode;
++ struct task_struct *tsk = current;
+
-+ LKTRTrace("dentry %.*s, bdst %d, bsrc %d, len %lld\n",
-+ AuDLNPair(dentry), bdst, bsrc, len);
-+ AuDebugOn(bsrc <= bdst);
-+ AuDebugOn(!len);
-+ sb = dentry->d_sb;
-+ AuDebugOn(au_test_ro(sb, bdst, dentry->d_inode));
+ /* bsrc branch can be ro/rw. */
++ sb = cpg->dentry->d_sb;
++ f = file;
++ for (i = 0; i < 2; i++, f++) {
++ f->dentry = au_h_dptr(cpg->dentry, f->bindex);
++ f->file = au_h_open(cpg->dentry, f->bindex, f->flags,
++ /*file*/NULL, f->force_wr);
++ if (IS_ERR(f->file)) {
++ err = PTR_ERR(f->file);
++ if (i == SRC)
++ goto out;
++ else
++ goto out_src;
++ }
++ }
+
-+ h = hidden;
-+ for (i = 0; i < 2; i++, h++) {
-+ h->dentry = au_h_dptr(dentry, h->bindex);
-+ AuDebugOn(!h->dentry);
-+ h_inode = h->dentry->d_inode;
-+ AuDebugOn(!h_inode || !S_ISREG(h_inode->i_mode));
-+ h->file = au_h_open(dentry, h->bindex, h->flags, /*file*/NULL);
-+ err = PTR_ERR(h->file);
-+ if (IS_ERR(h->file))
-+ goto *h->label;
-+ err = -EINVAL;
-+ if (unlikely(!h->file->f_op))
-+ goto *h->label_file;
-+ }
-+
-+ /* stop updating while we copyup */
-+ IMustLock(hidden[SRC].dentry->d_inode);
-+ err = au_copy_file(hidden[DST].file, hidden[SRC].file, len, hdir, sb,
-+ vargs);
-+
-+ out_dst_file:
-+ fput(hidden[DST].file);
-+ au_sbr_put(sb, hidden[DST].bindex);
-+ out_src_file:
-+ fput(hidden[SRC].file);
-+ au_sbr_put(sb, hidden[SRC].bindex);
-+ out:
-+ AuTraceErr(err);
++ /* try stopping to update while we copyup */
++ 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)
++ __fput_sync(file[DST].file);
++ else {
++ /* it happened actually */
++ fput(file[DST].file);
++ /*
++ * too bad.
++ * we have to call both since we don't know which place the file
++ * was added to.
++ */
++ task_work_run();
++ flush_delayed_fput();
++ }
++ br = au_sbr(sb, file[DST].bindex);
++ au_lcnt_dec(&br->br_nfiles);
++
++out_src:
++ fput(file[SRC].file);
++ br = au_sbr(sb, file[SRC].bindex);
++ au_lcnt_dec(&br->br_nfiles);
++out:
+ return err;
+}
+
-+static int au_do_cpup_regular(struct dentry *dentry, aufs_bindex_t bdst,
-+ aufs_bindex_t bsrc, loff_t len,
-+ struct au_hinode *hdir, struct dentry *h_dst,
-+ struct vfsub_args *vargs)
++static int au_do_cpup_regular(struct au_cp_generic *cpg,
++ struct au_cpup_reg_attr *h_src_attr)
+{
+ int err, rerr;
+ loff_t l;
-+
-+ AuTraceEnter();
++ struct path h_path;
++ struct inode *h_src_inode, *h_dst_inode;
+
+ err = 0;
-+ l = i_size_read(au_h_iptr(dentry->d_inode, bsrc));
-+ if (len == -1 || l < len)
-+ len = l;
-+ if (len)
-+ err = cpup_regular(dentry, bdst, bsrc, len, hdir, vargs);
-+ if (!err)
-+ goto out; /* success */
-+
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_DELETE, hdir);
-+ rerr = vfsub_unlink(hdir->hi_inode, h_dst, vargs);
-+ if (rerr) {
-+ AuIOErr("failed unlinking cpup-ed %.*s(%d, %d)\n",
-+ AuDLNPair(h_dst), err, rerr);
-+ err = -EIO;
++ h_src_inode = au_h_iptr(d_inode(cpg->dentry), cpg->bsrc);
++ l = i_size_read(h_src_inode);
++ if (cpg->len == -1 || l < cpg->len)
++ cpg->len = l;
++ if (cpg->len) {
++ /* try stopping to update while we are referencing */
++ 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 = vfsub_getattr(&h_path, &h_src_attr->st);
++ else {
++ inode_unlock_shared(h_src_inode);
++ err = vfsub_getattr(&h_path, &h_src_attr->st);
++ inode_lock_shared_nested(h_src_inode, AuLsc_I_CHILD);
++ }
++ if (unlikely(err)) {
++ inode_unlock_shared(h_src_inode);
++ goto out;
++ }
++ h_src_attr->valid = 1;
++ 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;
++ }
++ if (!err && (h_src_inode->i_state & I_LINKABLE)) {
++ h_path.dentry = au_h_dptr(cpg->dentry, cpg->bdst);
++ h_dst_inode = d_inode(h_path.dentry);
++ spin_lock(&h_dst_inode->i_lock);
++ h_dst_inode->i_state |= I_LINKABLE;
++ spin_unlock(&h_dst_inode->i_lock);
+ }
+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static int au_do_cpup_symlink(struct dentry *h_dst, struct dentry *h_src,
-+ struct inode *h_dir, umode_t mode,
-+ struct vfsub_args *vargs)
++static int au_do_cpup_symlink(struct path *h_path, struct dentry *h_src,
++ struct inode *h_dir)
+{
-+ int err, symlen;
-+ char *sym;
-+ mm_segment_t old_fs;
-+
-+ AuTraceEnter();
++ int err;
++ DEFINE_DELAYED_CALL(done);
++ const char *sym;
+
-+ err = -ENOMEM;
-+ sym = __getname();
-+ if (unlikely(!sym))
++ sym = vfs_get_link(h_src, &done);
++ err = PTR_ERR(sym);
++ if (IS_ERR(sym))
+ goto out;
+
-+ old_fs = get_fs();
-+ set_fs(KERNEL_DS);
-+ symlen = h_src->d_inode->i_op->readlink(h_src, (char __user *)sym,
-+ PATH_MAX);
-+ err = symlen;
-+ set_fs(old_fs);
++ err = vfsub_symlink(h_dir, h_path, sym);
+
-+ if (symlen > 0) {
-+ sym[symlen] = 0;
-+ err = vfsub_symlink(h_dir, h_dst, sym, mode, vargs);
-+ }
-+ __putname(sym);
++out:
++ do_delayed_call(&done);
++ return err;
++}
++
++/*
++ * regardless 'acl' option, reset all ACL.
++ * All ACL will be copied up later from the original entry on the lower branch.
++ */
++static int au_reset_acl(struct path *h_path, umode_t mode)
++{
++ int err;
++ struct dentry *h_dentry;
++ /* struct inode *h_inode; */
++ struct mnt_idmap *h_idmap;
++
++ h_idmap = mnt_idmap(h_path->mnt);
++ h_dentry = h_path->dentry;
++ /* h_inode = d_inode(h_dentry); */
++ /* forget_all_cached_acls(h_inode)); */
++ err = vfsub_remove_acl(h_idmap, h_dentry, XATTR_NAME_POSIX_ACL_ACCESS);
++ AuTraceErr(err);
++ if (err == -EOPNOTSUPP)
++ err = 0;
++ if (!err)
++ err = vfsub_acl_chmod(h_idmap, h_dentry, mode);
++
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_do_cpup_dir(struct au_cp_generic *cpg, struct dentry *dst_parent,
++ struct inode *h_dir, struct path *h_path)
++{
++ int err;
++ struct inode *dir, *inode;
++ struct mnt_idmap *h_idmap;
+
-+ out:
++ h_idmap = mnt_idmap(h_path->mnt);
++ err = vfsub_remove_acl(h_idmap, h_path->dentry,
++ XATTR_NAME_POSIX_ACL_DEFAULT);
+ AuTraceErr(err);
++ if (err == -EOPNOTSUPP)
++ err = 0;
++ if (unlikely(err))
++ goto out;
++
++ /*
++ * strange behaviour from the users view,
++ * particularly setattr case
++ */
++ dir = d_inode(dst_parent);
++ if (au_ibtop(dir) == cpg->bdst)
++ au_cpup_attr_nlink(dir, /*force*/1);
++ inode = d_inode(cpg->dentry);
++ au_cpup_attr_nlink(inode, /*force*/1);
++
++out:
+ return err;
+}
+
-+/* return with hidden dst inode is locked */
+static noinline_for_stack
-+int cpup_entry(struct dentry *dentry, aufs_bindex_t bdst, aufs_bindex_t bsrc,
-+ loff_t len, unsigned int flags, struct dentry *dst_parent,
-+ struct vfsub_args *vargs)
++int cpup_entry(struct au_cp_generic *cpg, struct dentry *dst_parent,
++ struct au_cpup_reg_attr *h_src_attr)
+{
+ int err;
-+ unsigned char isdir, hinotify;
-+ struct dentry *h_src, *h_dst, *h_parent, *gparent;
-+ struct inode *h_inode, *h_dir;
-+ struct au_dtime dt;
+ umode_t mode;
-+ struct super_block *sb;
-+ struct au_hinode *hgdir, *hdir;
+ unsigned int mnt_flags;
-+ const int do_dt = au_ftest_cpup(flags, DTIME);
++ unsigned char isdir, isreg, force;
++ const unsigned char do_dt = !!au_ftest_cpup(cpg->flags, DTIME);
++ struct au_dtime dt;
++ struct path h_path;
++ struct dentry *h_src, *h_dst, *h_parent;
++ struct inode *h_inode, *h_dir;
++ struct super_block *sb;
+
-+ LKTRTrace("%.*s, i%lu, bdst %d, bsrc %d, len %lld, dtime %u\n",
-+ AuDLNPair(dentry), dentry->d_inode->i_ino, bdst, bsrc, len,
-+ do_dt);
-+ sb = dentry->d_sb;
-+ AuDebugOn(bdst >= bsrc || au_test_ro(sb, bdst, NULL));
+ /* bsrc branch can be ro/rw. */
-+
-+ h_src = au_h_dptr(dentry, bsrc);
-+ AuDebugOn(!h_src);
-+ h_inode = h_src->d_inode;
-+ AuDebugOn(!h_inode);
-+ AuDebugOn(h_inode != au_h_iptr(dentry->d_inode, bsrc));
-+
-+ /* stop referencing while we are creating */
-+ h_dst = au_h_dptr(dentry, bdst);
-+ AuDebugOn(h_dst && h_dst->d_inode);
++ h_src = au_h_dptr(cpg->dentry, cpg->bsrc);
++ h_inode = d_inode(h_src);
++ AuDebugOn(h_inode != au_h_iptr(d_inode(cpg->dentry), cpg->bsrc));
++
++ /* try stopping to be referenced while we are creating */
++ h_dst = au_h_dptr(cpg->dentry, cpg->bdst);
++ if (au_ftest_cpup(cpg->flags, RENAME))
++ AuDebugOn(strncmp(h_dst->d_name.name, AUFS_WH_PFX,
++ AUFS_WH_PFX_LEN));
+ h_parent = h_dst->d_parent; /* dir inode is locked */
-+ h_dir = h_parent->d_inode;
++ h_dir = d_inode(h_parent);
+ IMustLock(h_dir);
+ AuDebugOn(h_parent != h_dst->d_parent);
+
-+ hdir = NULL;
-+ mnt_flags = au_mntflags(sb);
-+ hinotify = !!au_opt_test(mnt_flags, UDBA_INOTIFY);
-+ if (hinotify) {
-+ hdir = au_hi(dst_parent->d_inode, bdst);
-+ AuDebugOn(hdir->hi_inode != h_dir);
-+ }
-+
++ sb = cpg->dentry->d_sb;
++ h_path.mnt = au_sbr_mnt(sb, cpg->bdst);
+ if (do_dt) {
-+ hgdir = NULL;
-+ if (hinotify && !IS_ROOT(dst_parent)) {
-+ gparent = dget_parent(dst_parent);
-+ hgdir = au_hi(gparent->d_inode, bdst);
-+ IMustLock(hgdir->hi_inode);
-+ dput(gparent);
-+ }
-+ au_dtime_store(&dt, dst_parent, h_parent, hdir, hgdir);
++ h_path.dentry = h_parent;
++ au_dtime_store(&dt, dst_parent, &h_path);
+ }
++ h_path.dentry = h_dst;
+
++ isreg = 0;
+ isdir = 0;
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_CREATE, hdir);
+ mode = h_inode->i_mode;
+ switch (mode & S_IFMT) {
+ case S_IFREG:
-+ /* stop updating while we are referencing */
-+ IMustLock(h_inode);
-+ err = au_h_create(h_dir, h_dst, mode | S_IWUSR, vargs, NULL,
-+ au_nfsmnt(sb, bdst));
++ isreg = 1;
++ err = vfsub_create(h_dir, &h_path, 0600, /*want_excl*/true);
+ if (!err)
-+ err = au_do_cpup_regular(dentry, bdst, bsrc, len,
-+ hdir, h_dst, vargs);
++ err = au_do_cpup_regular(cpg, h_src_attr);
+ break;
+ case S_IFDIR:
+ isdir = 1;
-+ err = vfsub_mkdir(h_dir, h_dst, mode, vargs);
-+ if (!err) {
-+ /* setattr case: dir is not locked */
-+ if (0 && au_ibstart(dst_parent->d_inode) == bdst)
-+ au_cpup_attr_nlink(dst_parent->d_inode,
-+ /*force*/1);
-+ au_cpup_attr_nlink(dentry->d_inode, /*force*/1);
-+ }
++ err = vfsub_mkdir(h_dir, &h_path, mode);
++ if (!err)
++ err = au_do_cpup_dir(cpg, dst_parent, h_dir, &h_path);
+ break;
+ case S_IFLNK:
-+ err = au_do_cpup_symlink(h_dst, h_src, h_dir, mode, vargs);
++ err = au_do_cpup_symlink(&h_path, h_src, h_dir);
+ break;
+ case S_IFCHR:
+ case S_IFBLK:
+ AuDebugOn(!capable(CAP_MKNOD));
-+ /*FALLTHROUGH*/
++ fallthrough;
+ case S_IFIFO:
+ case S_IFSOCK:
-+ err = vfsub_mknod(h_dir, h_dst, mode,
-+ au_h_rdev(h_inode, /*h_mnt*/NULL, h_src),
-+ vargs);
++ err = vfsub_mknod(h_dir, &h_path, mode, h_inode->i_rdev);
+ break;
+ default:
+ AuIOErr("Unknown inode type 0%o\n", mode);
+ err = -EIO;
+ }
++ if (!err)
++ err = au_reset_acl(&h_path, mode);
+
-+ if (hinotify
++ mnt_flags = au_mntflags(sb);
++ if (!au_opt_test(mnt_flags, UDBA_NONE)
+ && !isdir
-+ && au_opt_test_xino(mnt_flags)
-+ && h_inode->i_nlink == 1
++ && au_opt_test(mnt_flags, XINO)
++ && (h_inode->i_nlink == 1
++ || (h_inode->i_state & I_LINKABLE))
+ /* todo: unnecessary? */
-+ /* && dentry->d_inode->i_nlink == 1 */
-+ && bdst < bsrc
-+ && !au_ftest_cpup(flags, KEEPLINO))
-+ au_xino_write0(sb, bsrc, h_inode->i_ino, /*ino*/0);
++ /* && d_inode(cpg->dentry)->i_nlink == 1 */
++ && cpg->bdst < cpg->bsrc
++ && !au_ftest_cpup(cpg->flags, KEEPLINO))
++ au_xino_write(sb, cpg->bsrc, h_inode->i_ino, /*ino*/0);
+ /* ignore this error */
+
++ if (!err) {
++ force = 0;
++ if (isreg) {
++ force = !!cpg->len;
++ if (cpg->len == -1)
++ force = !!i_size_read(h_inode);
++ }
++ au_fhsm_wrote(sb, cpg->bdst, force);
++ }
++
+ if (do_dt)
+ au_dtime_revert(&dt);
-+ AuTraceErr(err);
++ return err;
++}
++
++static int au_do_ren_after_cpup(struct au_cp_generic *cpg, struct path *h_path)
++{
++ int err;
++ struct dentry *dentry, *h_dentry, *h_parent, *parent;
++ struct path h_ppath;
++ struct inode *h_dir;
++ aufs_bindex_t bdst;
++
++ dentry = cpg->dentry;
++ bdst = cpg->bdst;
++ h_ppath.mnt = au_sbr_mnt(dentry->d_sb, bdst);
++ h_dentry = au_h_dptr(dentry, bdst);
++ if (!au_ftest_cpup(cpg->flags, OVERWRITE)) {
++ dget(h_dentry);
++ au_set_h_dptr(dentry, bdst, NULL);
++ err = au_lkup_neg(dentry, bdst, /*wh*/0);
++ if (!err)
++ h_path->dentry = dget(au_h_dptr(dentry, bdst));
++ au_set_h_dptr(dentry, bdst, h_dentry);
++ } else {
++ err = 0;
++ parent = dget_parent(dentry);
++ h_ppath.dentry = au_h_dptr(parent, bdst);
++ dput(parent);
++ h_path->dentry = vfsub_lkup_one(&dentry->d_name, &h_ppath);
++ if (IS_ERR(h_path->dentry))
++ err = PTR_ERR(h_path->dentry);
++ }
++ if (unlikely(err))
++ goto out;
++
++ h_parent = h_dentry->d_parent; /* dir inode is locked */
++ h_dir = d_inode(h_parent);
++ 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,
++ /*flags*/0);
++ dput(h_path->dentry);
++
++out:
+ return err;
+}
+
+/*
+ * copyup the @dentry from @bsrc to @bdst.
-+ * the caller must set the both of hidden dentries.
++ * the caller must set the both of lower dentries.
+ * @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 unnecessary.
+ */
-+static int au_cpup_single(struct dentry *dentry, aufs_bindex_t bdst,
-+ aufs_bindex_t bsrc, loff_t len, unsigned int flags,
-+ struct dentry *dst_parent, struct vfsub_args *vargs)
++static int au_cpup_single(struct au_cp_generic *cpg, struct dentry *dst_parent)
+{
+ int err, rerr;
-+ unsigned int mnt_flags;
-+ aufs_bindex_t old_ibstart;
-+ unsigned char isdir, plink, hinotify;
-+ struct au_dtime dt;
-+ struct dentry *h_src, *h_dst, *h_parent, *gparent;
-+ struct inode *dst_inode, *h_dir, *inode;
++ aufs_bindex_t old_ibtop;
++ unsigned char isdir, plink;
++ struct dentry *h_src, *h_dst, *h_parent;
++ struct inode *dst_inode, *h_dir, *inode, *delegated, *src_inode;
+ struct super_block *sb;
-+ struct au_hinode *hgdir, *hdir;
++ struct au_branch *br;
++ struct path h_src_path;
++ /* to reduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct path h_path;
++ struct au_cpup_reg_attr h_src_attr;
++ } *a;
++
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++ a->h_src_attr.valid = 0;
++
++ sb = cpg->dentry->d_sb;
++ br = au_sbr(sb, cpg->bdst);
++ a->h_path.mnt = au_br_mnt(br);
++ h_dst = au_h_dptr(cpg->dentry, cpg->bdst);
++ h_parent = h_dst->d_parent; /* dir inode is locked */
++ h_dir = d_inode(h_parent);
++ IMustLock(h_dir);
++
++ h_src = au_h_dptr(cpg->dentry, cpg->bsrc);
++ inode = d_inode(cpg->dentry);
+
-+ LKTRTrace("%.*s, i%lu, bdst %d, bsrc %d, len %lld, flags 0x%x\n",
-+ AuDLNPair(dentry), dentry->d_inode->i_ino, bdst, bsrc, len,
-+ flags);
-+ sb = dentry->d_sb;
-+ AuDebugOn(bsrc <= bdst);
-+ h_dst = au_h_dptr(dentry, bdst);
-+ AuDebugOn(!h_dst || h_dst->d_inode);
-+ h_parent = h_dst->d_parent; /* dir inode is locked */
-+ h_dir = h_parent->d_inode;
-+ IMustLock(h_dir);
-+ h_src = au_h_dptr(dentry, bsrc);
-+ AuDebugOn(!h_src || !h_src->d_inode);
-+ inode = dentry->d_inode;
-+ IiMustWriteLock(inode);
+ if (!dst_parent)
-+ dst_parent = dget_parent(dentry);
++ dst_parent = dget_parent(cpg->dentry);
+ else
+ dget(dst_parent);
+
-+ mnt_flags = au_mntflags(sb);
-+ plink = !!au_opt_test(mnt_flags, PLINK);
-+ hinotify = !!au_opt_test(mnt_flags, UDBA_INOTIFY);
-+ hdir = NULL;
-+ if (hinotify)
-+ hdir = au_hi(dst_parent->d_inode, bdst);
-+ dst_inode = au_h_iptr(inode, bdst);
++ plink = !!au_opt_test(au_mntflags(sb), PLINK);
++ dst_inode = au_h_iptr(inode, cpg->bdst);
+ if (dst_inode) {
+ if (unlikely(!plink)) {
+ err = -EIO;
-+ AuIOErr("i%lu exists on a upper branch "
-+ "but plink is disabled\n", inode->i_ino);
-+ goto out;
++ AuIOErr("hi%lu(i%lu) exists on b%d "
++ "but plink is disabled\n",
++ dst_inode->i_ino, inode->i_ino, cpg->bdst);
++ goto out_parent;
+ }
+
+ if (dst_inode->i_nlink) {
-+ const int do_dt = au_ftest_cpup(flags, DTIME);
++ const int do_dt = au_ftest_cpup(cpg->flags, DTIME);
+
-+ h_src = au_plink_lkup(sb, bdst, inode);
++ h_src = au_plink_lkup(inode, cpg->bdst);
+ err = PTR_ERR(h_src);
+ if (IS_ERR(h_src))
-+ goto out;
-+ if (unlikely(!h_src->d_inode)) {
++ goto out_parent;
++ if (unlikely(d_is_negative(h_src))) {
+ err = -EIO;
-+ AuIOErr("i%lu exists on a upper branch "
-+ "but plink is broken\n", inode->i_ino);
++ AuIOErr("i%lu exists on b%d "
++ "but not pseudo-linked\n",
++ inode->i_ino, cpg->bdst);
+ dput(h_src);
-+ goto out;
++ goto out_parent;
+ }
+
+ if (do_dt) {
-+ hgdir = NULL;
-+ if (hinotify && !IS_ROOT(dst_parent)) {
-+ gparent = dget_parent(dst_parent);
-+ hgdir = au_hi(gparent->d_inode, bdst);
-+ IMustLock(hgdir->hi_inode);
-+ dput(gparent);
-+ }
-+ au_dtime_store(&dt, dst_parent, h_parent, hdir,
-+ hgdir);
++ a->h_path.dentry = h_parent;
++ au_dtime_store(&a->dt, dst_parent, &a->h_path);
+ }
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_CREATE, hdir);
-+ err = vfsub_link(h_src, h_dir, h_dst, vargs);
++
++ a->h_path.dentry = h_dst;
++ delegated = NULL;
++ err = vfsub_link(h_src, h_dir, &a->h_path, &delegated);
++ if (!err && au_ftest_cpup(cpg->flags, RENAME))
++ err = au_do_ren_after_cpup(cpg, &a->h_path);
+ if (do_dt)
-+ au_dtime_revert(&dt);
++ au_dtime_revert(&a->dt);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
+ dput(h_src);
-+ goto out;
++ goto out_parent;
+ } else
+ /* todo: cpup_wh_file? */
+ /* udba work */
-+ au_update_brange(inode, 1);
++ au_update_ibrange(inode, /*do_put_zero*/1);
+ }
+
-+ old_ibstart = au_ibstart(inode);
-+ err = cpup_entry(dentry, bdst, bsrc, len, flags, dst_parent, vargs);
++ isdir = S_ISDIR(inode->i_mode);
++ old_ibtop = au_ibtop(inode);
++ err = cpup_entry(cpg, dst_parent, &a->h_src_attr);
+ if (unlikely(err))
-+ goto out;
-+ dst_inode = h_dst->d_inode;
-+ mutex_lock_nested(&dst_inode->i_mutex, AuLsc_I_CHILD2);
++ goto out_rev;
++ dst_inode = d_inode(h_dst);
++ inode_lock_nested(dst_inode, AuLsc_I_CHILD2);
++ /* todo: necessary? */
++ /* au_pin_hdir_unlock(cpg->pin); */
+
-+ /* todo: test dlgt? */
-+ err = cpup_iattr(dentry, bdst, h_src, hdir, vargs);
-+#if 0 /* reserved for future use */
-+ if (0 && !err)
-+ err = cpup_xattrs(h_src, h_dst);
-+#endif
-+ isdir = S_ISDIR(dst_inode->i_mode);
-+ if (!err) {
-+ if (bdst < old_ibstart)
-+ au_set_ibstart(inode, bdst);
-+ au_set_h_iptr(inode, bdst, au_igrab(dst_inode),
-+ au_hi_flags(inode, isdir));
-+ mutex_unlock(&dst_inode->i_mutex);
-+ if (!isdir
-+ && h_src->d_inode->i_nlink > 1
-+ && plink)
-+ au_plink_append(sb, inode, h_dst, bdst);
-+ goto out; /* success */
++ h_src_path.dentry = h_src;
++ h_src_path.mnt = au_sbr_mnt(sb, cpg->bsrc);
++ err = cpup_iattr(cpg->dentry, cpg->bdst, &h_src_path, &a->h_src_attr);
++ if (unlikely(err)) {
++ /* todo: necessary? */
++ /* au_pin_hdir_relock(cpg->pin); */ /* ignore an error */
++ inode_unlock(dst_inode);
++ goto out_rev;
++ }
++
++ if (cpg->bdst < old_ibtop) {
++ if (S_ISREG(inode->i_mode)) {
++ err = au_dy_iaop(inode, cpg->bdst, dst_inode);
++ if (unlikely(err)) {
++ /* ignore an error */
++ /* au_pin_hdir_relock(cpg->pin); */
++ inode_unlock(dst_inode);
++ goto out_rev;
++ }
++ }
++ au_set_ibtop(inode, cpg->bdst);
++ } else
++ au_set_ibbot(inode, cpg->bdst);
++ au_set_h_iptr(inode, cpg->bdst, au_igrab(dst_inode),
++ au_hi_flags(inode, isdir));
++
++ /* todo: necessary? */
++ /* err = au_pin_hdir_relock(cpg->pin); */
++ inode_unlock(dst_inode);
++ if (unlikely(err))
++ goto out_rev;
++
++ src_inode = d_inode(h_src);
++ if (!isdir
++ && (src_inode->i_nlink > 1
++ || src_inode->i_state & I_LINKABLE)
++ && plink)
++ au_plink_append(inode, cpg->bdst, h_dst);
++
++ if (au_ftest_cpup(cpg->flags, RENAME)) {
++ a->h_path.dentry = h_dst;
++ err = au_do_ren_after_cpup(cpg, &a->h_path);
+ }
++ if (!err)
++ goto out_parent; /* success */
+
+ /* revert */
-+ mutex_unlock(&dst_inode->i_mutex);
-+ hgdir = NULL;
-+ if (au_opt_test(mnt_flags, UDBA_INOTIFY) && !IS_ROOT(dst_parent)) {
-+ gparent = dget_parent(dst_parent);
-+ hgdir = au_hi(gparent->d_inode, bdst);
-+ dput(gparent);
-+ }
-+ au_dtime_store(&dt, dst_parent, h_parent, hdir, hgdir);
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_DELETE, hdir);
-+ if (!isdir)
-+ rerr = vfsub_unlink(h_dir, h_dst, vargs);
-+ else
-+ rerr = vfsub_rmdir(h_dir, h_dst, vargs);
-+ au_dtime_revert(&dt);
++out_rev:
++ a->h_path.dentry = h_parent;
++ au_dtime_store(&a->dt, dst_parent, &a->h_path);
++ a->h_path.dentry = h_dst;
++ rerr = 0;
++ if (d_is_positive(h_dst)) {
++ if (!isdir) {
++ /* no delegation since it is just created */
++ rerr = vfsub_unlink(h_dir, &a->h_path,
++ /*delegated*/NULL, /*force*/0);
++ } else
++ rerr = vfsub_rmdir(h_dir, &a->h_path);
++ }
++ au_dtime_revert(&a->dt);
+ if (rerr) {
+ AuIOErr("failed removing broken entry(%d, %d)\n", err, rerr);
+ err = -EIO;
+ }
-+
-+ out:
++out_parent:
+ dput(dst_parent);
-+ AuTraceErr(err);
++ au_kfree_rcu(a);
++out:
+ return err;
+}
+
++#if 0 /* reserved */
+struct au_cpup_single_args {
+ int *errp;
-+ struct dentry *dentry;
-+ aufs_bindex_t bdst, bsrc;
-+ loff_t len;
-+ unsigned int flags;
++ struct au_cp_generic *cpg;
+ struct dentry *dst_parent;
-+ struct vfsub_args *vargs;
+};
+
+static void au_call_cpup_single(void *args)
+{
+ struct au_cpup_single_args *a = args;
-+ *a->errp = au_cpup_single(a->dentry, a->bdst, a->bsrc, a->len,
-+ a->flags, a->dst_parent, a->vargs);
++
++ au_pin_hdir_acquire_nest(a->cpg->pin);
++ *a->errp = au_cpup_single(a->cpg, a->dst_parent);
++ au_pin_hdir_release(a->cpg->pin);
++}
++#endif
++
++/*
++ * prevent SIGXFSZ in copy-up.
++ * testing CAP_MKNOD is for generic fs,
++ * but CAP_FSETID is for xfs only, currently.
++ */
++static int au_cpup_sio_test(struct au_pin *pin, umode_t mode)
++{
++ int do_sio;
++ struct super_block *sb;
++ struct inode *h_dir;
++
++ do_sio = 0;
++ sb = au_pinned_parent(pin)->d_sb;
++ if (!au_wkq_test()
++ && (!au_sbi(sb)->si_plink_maint_pid
++ || au_plink_maint(sb, AuLock_NOPLM))) {
++ switch (mode & S_IFMT) {
++ case S_IFREG:
++ /* no condition about RLIMIT_FSIZE and the file size */
++ do_sio = 1;
++ break;
++ case S_IFCHR:
++ case S_IFBLK:
++ do_sio = !capable(CAP_MKNOD);
++ break;
++ }
++ if (!do_sio)
++ do_sio = ((mode & (S_ISUID | S_ISGID))
++ && !capable(CAP_FSETID));
++ /* this workaround may be removed in the future */
++ if (!do_sio) {
++ h_dir = au_pinned_h_dir(pin);
++ do_sio = h_dir->i_mode & S_ISVTX;
++ }
++ }
++
++ return do_sio;
+}
+
-+int au_sio_cpup_single(struct dentry *dentry, aufs_bindex_t bdst,
-+ aufs_bindex_t bsrc, loff_t len, unsigned int flags,
-+ struct dentry *dst_parent)
++#if 0 /* reserved */
++int au_sio_cpup_single(struct au_cp_generic *cpg, struct dentry *dst_parent)
+{
+ int err, wkq_err;
+ struct dentry *h_dentry;
-+ umode_t mode;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+
-+ LKTRTrace("%.*s, i%lu, bdst %d, bsrc %d, len %lld, flags 0x%x\n",
-+ AuDLNPair(dentry), dentry->d_inode->i_ino, bdst, bsrc, len,
-+ flags);
-+
-+ vfsub_args_init(&vargs, &ign, au_test_dlgt(au_mntflags(dentry->d_sb)),
-+ /*force_unlink*/0);
-+ h_dentry = au_h_dptr(dentry, bsrc);
-+ mode = h_dentry->d_inode->i_mode & S_IFMT;
-+ if ((mode != S_IFCHR && mode != S_IFBLK)
-+ || capable(CAP_MKNOD))
-+ err = au_cpup_single(dentry, bdst, bsrc, len, flags,
-+ dst_parent, &vargs);
++
++ h_dentry = au_h_dptr(cpg->dentry, cpg->bsrc);
++ if (!au_cpup_sio_test(pin, d_inode(h_dentry)->i_mode))
++ err = au_cpup_single(cpg, dst_parent);
+ else {
+ struct au_cpup_single_args args = {
+ .errp = &err,
-+ .dentry = dentry,
-+ .bdst = bdst,
-+ .bsrc = bsrc,
-+ .len = len,
-+ .flags = flags,
-+ .dst_parent = dst_parent,
-+ .vargs = &vargs
++ .cpg = cpg,
++ .dst_parent = dst_parent
+ };
-+ vfsub_fclr(vargs.flags, DLGT);
-+ wkq_err = au_wkq_wait(au_call_cpup_single, &args, /*dlgt*/0);
++ wkq_err = au_wkq_wait(au_call_cpup_single, &args);
+ if (unlikely(wkq_err))
+ err = wkq_err;
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
++#endif
+
+/*
-+ * copyup the @dentry from the first active hidden branch to @bdst,
++ * copyup the @dentry from the first active lower branch to @bdst,
+ * using au_cpup_single().
+ */
-+static int au_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ unsigned int flags, struct vfsub_args *vargs)
++static int au_cpup_simple(struct au_cp_generic *cpg)
+{
+ int err;
-+ struct inode *inode;
-+ aufs_bindex_t bsrc, bend;
++ unsigned int flags_orig;
++ struct dentry *dentry;
+
-+ LKTRTrace("%.*s, bdst %d, len %lld, flags 0x%x\n",
-+ AuDLNPair(dentry), bdst, len, flags);
-+ inode = dentry->d_inode;
-+ AuDebugOn(!S_ISDIR(inode->i_mode) && au_dbstart(dentry) < bdst);
++ AuDebugOn(cpg->bsrc < 0);
+
-+ bend = au_dbend(dentry);
-+ for (bsrc = bdst + 1; bsrc <= bend; bsrc++)
-+ if (au_h_dptr(dentry, bsrc))
-+ break;
-+ AuDebugOn(!au_h_dptr(dentry, bsrc));
++ dentry = cpg->dentry;
++ DiMustWriteLock(dentry);
+
-+ err = au_lkup_neg(dentry, bdst);
++ err = au_lkup_neg(dentry, cpg->bdst, /*wh*/1);
+ if (!err) {
-+ err = au_cpup_single(dentry, bdst, bsrc, len, flags, NULL,
-+ vargs);
++ flags_orig = cpg->flags;
++ au_fset_cpup(cpg->flags, RENAME);
++ err = au_cpup_single(cpg, NULL);
++ cpg->flags = flags_orig;
+ if (!err)
+ return 0; /* success */
+
+ /* revert */
-+ au_set_h_dptr(dentry, bdst, NULL);
-+ au_set_dbstart(dentry, bsrc);
++ au_set_h_dptr(dentry, cpg->bdst, NULL);
++ au_set_dbtop(dentry, cpg->bsrc);
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
+struct au_cpup_simple_args {
+ int *errp;
-+ struct dentry *dentry;
-+ aufs_bindex_t bdst;
-+ loff_t len;
-+ unsigned int flags;
-+ struct vfsub_args *vargs;
++ struct au_cp_generic *cpg;
+};
+
+static void au_call_cpup_simple(void *args)
+{
+ struct au_cpup_simple_args *a = args;
-+ *a->errp = au_cpup_simple(a->dentry, a->bdst, a->len, a->flags,
-+ a->vargs);
++
++ au_pin_hdir_acquire_nest(a->cpg->pin);
++ *a->errp = au_cpup_simple(a->cpg);
++ au_pin_hdir_release(a->cpg->pin);
+}
+
-+int au_sio_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ unsigned int flags)
++static int au_do_sio_cpup_simple(struct au_cp_generic *cpg)
+{
+ int err, wkq_err;
-+ unsigned char do_sio, dlgt;
-+ struct dentry *parent;
-+ struct inode *h_dir, *dir;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
++ struct dentry *dentry, *parent;
++ struct file *h_file;
++ struct inode *h_dir;
++ struct mnt_idmap *h_idmap;
+
-+ LKTRTrace("%.*s, b%d, len %lld, flags 0x%x\n",
-+ AuDLNPair(dentry), bdst, len, flags);
++ dentry = cpg->dentry;
++ h_file = NULL;
++ if (au_ftest_cpup(cpg->flags, HOPEN)) {
++ AuDebugOn(cpg->bsrc < 0);
++ h_file = au_h_open_pre(dentry, cpg->bsrc, /*force_wr*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++ }
+
+ parent = dget_parent(dentry);
-+ dir = parent->d_inode;
-+ h_dir = au_h_iptr(dir, bdst);
-+ dlgt = !!au_test_dlgt(au_mntflags(dir->i_sb));
-+ do_sio = !!au_test_h_perm_sio(h_dir, MAY_EXEC | MAY_WRITE, dlgt);
-+ if (!do_sio) {
-+ /*
-+ * testing CAP_MKNOD is for generic fs,
-+ * but CAP_FSETID is for xfs only, currently.
-+ */
-+ umode_t mode = dentry->d_inode->i_mode;
-+ do_sio = (((mode & (S_IFCHR | S_IFBLK))
-+ && !capable(CAP_MKNOD))
-+ || ((mode & (S_ISUID | S_ISGID))
-+ && !capable(CAP_FSETID)));
-+ }
-+ vfsub_args_init(&vargs, &ign, dlgt, /*force_unlink*/0);
-+ if (!do_sio)
-+ err = au_cpup_simple(dentry, bdst, len, flags, &vargs);
++ h_dir = au_h_iptr(d_inode(parent), cpg->bdst);
++ h_idmap = au_sbr_idmap(dentry->d_sb, cpg->bdst);
++ if (!au_test_h_perm_sio(h_idmap, h_dir, MAY_EXEC | MAY_WRITE)
++ && !au_cpup_sio_test(cpg->pin, d_inode(dentry)->i_mode))
++ err = au_cpup_simple(cpg);
+ else {
+ struct au_cpup_simple_args args = {
+ .errp = &err,
-+ .dentry = dentry,
-+ .bdst = bdst,
-+ .len = len,
-+ .flags = flags,
-+ .vargs = &vargs
++ .cpg = cpg
+ };
-+ vfsub_fclr(vargs.flags, DLGT);
-+ wkq_err = au_wkq_wait(au_call_cpup_simple, &args, /*dlgt*/0);
++ wkq_err = au_wkq_wait(au_call_cpup_simple, &args);
+ if (unlikely(wkq_err))
+ err = wkq_err;
+ }
+
+ dput(parent);
-+ AuTraceErr(err);
++ if (h_file)
++ au_h_open_post(dentry, cpg->bsrc, h_file);
++
++out:
+ return err;
+}
+
++int au_sio_cpup_simple(struct au_cp_generic *cpg)
++{
++ aufs_bindex_t bsrc, bbot;
++ struct dentry *dentry, *h_dentry;
++
++ if (cpg->bsrc < 0) {
++ dentry = cpg->dentry;
++ bbot = au_dbbot(dentry);
++ for (bsrc = cpg->bdst + 1; bsrc <= bbot; bsrc++) {
++ h_dentry = au_h_dptr(dentry, bsrc);
++ if (h_dentry) {
++ AuDebugOn(d_is_negative(h_dentry));
++ break;
++ }
++ }
++ AuDebugOn(bsrc > bbot);
++ cpg->bsrc = bsrc;
++ }
++ AuDebugOn(cpg->bsrc <= cpg->bdst);
++ return au_do_sio_cpup_simple(cpg);
++}
++
++int au_sio_cpdown_simple(struct au_cp_generic *cpg)
++{
++ AuDebugOn(cpg->bdst <= cpg->bsrc);
++ return au_do_sio_cpup_simple(cpg);
++}
++
+/* ---------------------------------------------------------------------- */
+
-+static int au_do_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst,
-+ struct dentry *wh_dentry, struct file *file,
-+ loff_t len, struct vfsub_args *vargs)
++/*
++ * copyup the deleted file for writing.
++ */
++static int au_do_cpup_wh(struct au_cp_generic *cpg, struct dentry *wh_dentry,
++ struct file *file)
+{
+ int err;
++ unsigned int flags_orig;
++ aufs_bindex_t bsrc_orig;
+ struct au_dinfo *dinfo;
-+ aufs_bindex_t bstart;
-+ struct dentry *h_d_bdst, *h_d_bstart;
++ struct {
++ struct au_hdentry *hd;
++ struct dentry *h_dentry;
++ } hdst, hsrc;
+
-+ AuTraceEnter();
++ dinfo = au_di(cpg->dentry);
++ AuRwMustWriteLock(&dinfo->di_rwsem);
+
-+ dinfo = au_di(dentry);
-+ bstart = dinfo->di_bstart;
-+ h_d_bdst = dinfo->di_hdentry[0 + bdst].hd_dentry;
-+ dinfo->di_bstart = bdst;
-+ dinfo->di_hdentry[0 + bdst].hd_dentry = wh_dentry;
-+ h_d_bstart = dinfo->di_hdentry[0 + bstart].hd_dentry;
-+ if (file)
-+ dinfo->di_hdentry[0 + bstart].hd_dentry
-+ = au_h_fptr(file, au_fbstart(file))->f_dentry;
-+ err = au_cpup_single(dentry, bdst, bstart, len, !AuCpup_DTIME,
-+ /*h_parent*/NULL, vargs);
-+ if (!err && file) {
-+ err = au_reopen_nondir(file);
-+ dinfo->di_hdentry[0 + bstart].hd_dentry = h_d_bstart;
++ bsrc_orig = cpg->bsrc;
++ cpg->bsrc = dinfo->di_btop;
++ hdst.hd = au_hdentry(dinfo, cpg->bdst);
++ hdst.h_dentry = hdst.hd->hd_dentry;
++ hdst.hd->hd_dentry = wh_dentry;
++ dinfo->di_btop = cpg->bdst;
++
++ hsrc.h_dentry = NULL;
++ if (file) {
++ hsrc.hd = au_hdentry(dinfo, cpg->bsrc);
++ hsrc.h_dentry = hsrc.hd->hd_dentry;
++ hsrc.hd->hd_dentry = au_hf_top(file)->f_path.dentry;
++ }
++ flags_orig = cpg->flags;
++ cpg->flags = !AuCpup_DTIME;
++ err = au_cpup_single(cpg, /*h_parent*/NULL);
++ cpg->flags = flags_orig;
++ if (file) {
++ if (!err)
++ err = au_reopen_nondir(file);
++ hsrc.hd->hd_dentry = hsrc.h_dentry;
+ }
-+ dinfo->di_hdentry[0 + bdst].hd_dentry = h_d_bdst;
-+ dinfo->di_bstart = bstart;
++ hdst.hd->hd_dentry = hdst.h_dentry;
++ dinfo->di_btop = cpg->bsrc;
++ cpg->bsrc = bsrc_orig;
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+/*
-+ * copyup the deleted file for writing.
-+ */
-+static int au_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ struct file *file)
++static int au_cpup_wh(struct au_cp_generic *cpg, struct file *file)
+{
+ int err;
-+ unsigned char dlgt;
-+ struct dentry *parent, *h_parent, *wh_dentry;
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
++ aufs_bindex_t bdst;
+ struct au_dtime dt;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_hinode *hgdir, *hdir;
-+ struct au_ndx ndx = {
-+ .nd = NULL,
-+ .flags = 0,
-+ /* .br = NULL */
-+ };
-+
-+ LKTRTrace("%.*s, bdst %d, len %llu\n", AuDLNPair(dentry), bdst, len);
-+ AuDebugOn(S_ISDIR(dentry->d_inode->i_mode)
-+ || (file && !(file->f_mode & FMODE_WRITE)));
-+ DiMustWriteLock(dentry);
++ struct dentry *dentry, *parent, *h_parent, *wh_dentry;
++ struct au_branch *br;
++ struct path h_path;
+
++ dentry = cpg->dentry;
++ bdst = cpg->bdst;
++ br = au_sbr(dentry->d_sb, bdst);
+ parent = dget_parent(dentry);
-+ IiMustAnyLock(parent->d_inode);
+ h_parent = au_h_dptr(parent, bdst);
-+ AuDebugOn(!h_parent);
-+
-+ sb = parent->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = 0;
-+ ndx.nfsmnt = au_nfsmnt(sb, bdst);
-+ if (au_test_dlgt(mnt_flags)) {
-+ dlgt = 1;
-+ au_fset_ndx(ndx.flags, DLGT);
-+ }
-+ wh_dentry = au_whtmp_lkup(h_parent, &dentry->d_name, &ndx);
++ wh_dentry = au_whtmp_lkup(h_parent, br, &dentry->d_name);
+ err = PTR_ERR(wh_dentry);
+ if (IS_ERR(wh_dentry))
+ goto out;
+
-+ hdir = NULL;
-+ hgdir = NULL;
-+ if (au_opt_test(mnt_flags, UDBA_INOTIFY)) {
-+ hdir = au_hi(parent->d_inode, bdst);
-+ if (!IS_ROOT(parent)) {
-+ struct dentry *gparent;
-+ gparent = dget_parent(parent);
-+ hgdir = au_hi(gparent->d_inode, bdst);
-+ dput(gparent);
-+ }
-+ }
-+ au_dtime_store(&dt, parent, h_parent, hdir, hgdir);
-+ vfsub_args_init(&vargs, &ign, dlgt, /*force_unlink*/0);
-+ err = au_do_cpup_wh(dentry, bdst, wh_dentry, file, len, &vargs);
++ h_path.dentry = h_parent;
++ h_path.mnt = au_br_mnt(br);
++ au_dtime_store(&dt, parent, &h_path);
++ err = au_do_cpup_wh(cpg, wh_dentry, file);
+ if (unlikely(err))
+ goto out_wh;
+
-+ AuDebugOn(!d_unhashed(dentry));
-+ /* dget first to force sillyrename on nfs */
+ dget(wh_dentry);
-+ vfsub_args_reinit(&vargs);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, hdir);
-+ err = vfsub_unlink(h_parent->d_inode, wh_dentry, &vargs);
++ h_path.dentry = wh_dentry;
++ if (!d_is_dir(wh_dentry)) {
++ /* no delegation since it is just created */
++ err = vfsub_unlink(d_inode(h_parent), &h_path,
++ /*delegated*/NULL, /*force*/0);
++ } else
++ err = vfsub_rmdir(d_inode(h_parent), &h_path);
+ if (unlikely(err)) {
-+ AuIOErr("failed remove copied-up tmp file %.*s(%d)\n",
-+ AuDLNPair(wh_dentry), err);
++ AuIOErr("failed remove copied-up tmp file %pd(%d)\n",
++ wh_dentry, err);
+ err = -EIO;
+ }
+ au_dtime_revert(&dt);
-+ au_set_hi_wh(dentry->d_inode, bdst, wh_dentry);
++ au_set_hi_wh(d_inode(dentry), bdst, wh_dentry);
+
-+ out_wh:
++out_wh:
+ dput(wh_dentry);
-+ out:
++out:
+ dput(parent);
-+ AuTraceErr(err);
+ return err;
+}
+
+struct au_cpup_wh_args {
+ int *errp;
-+ struct dentry *dentry;
-+ aufs_bindex_t bdst;
-+ loff_t len;
++ struct au_cp_generic *cpg;
+ struct file *file;
+};
+
+static void au_call_cpup_wh(void *args)
+{
+ struct au_cpup_wh_args *a = args;
-+ *a->errp = au_cpup_wh(a->dentry, a->bdst, a->len, a->file);
++
++ au_pin_hdir_acquire_nest(a->cpg->pin);
++ *a->errp = au_cpup_wh(a->cpg, a->file);
++ au_pin_hdir_release(a->cpg->pin);
+}
+
-+int au_sio_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ struct file *file)
++int au_sio_cpup_wh(struct au_cp_generic *cpg, struct file *file)
+{
+ int err, wkq_err;
-+ struct dentry *parent, *h_tmp, *h_parent, *h_dentry;
-+ struct inode *dir, *h_dir, *h_tmpdir, *h_inode;
++ aufs_bindex_t bdst;
++ struct dentry *dentry, *parent, *h_orph, *h_parent;
++ struct inode *dir, *h_dir, *h_tmpdir;
+ struct au_wbr *wbr;
++ struct au_pin wh_pin, *pin_orig;
++ struct mnt_idmap *h_idmap;
+
-+ AuTraceEnter();
++ dentry = cpg->dentry;
++ bdst = cpg->bdst;
+ parent = dget_parent(dentry);
-+ dir = parent->d_inode;
-+ IiMustAnyLock(dir);
-+
-+ h_tmp = NULL;
++ dir = d_inode(parent);
++ h_orph = NULL;
+ h_parent = NULL;
+ h_dir = au_igrab(au_h_iptr(dir, bdst));
+ h_tmpdir = h_dir;
++ pin_orig = NULL;
+ if (!h_dir->i_nlink) {
-+ DiMustWriteLock(parent);
+ wbr = au_sbr(dentry->d_sb, bdst)->br_wbr;
-+ AuDebugOn(!wbr);
-+ h_tmp = wbr->wbr_tmp;
++ h_orph = wbr->wbr_orph;
+
+ h_parent = dget(au_h_dptr(parent, bdst));
-+ au_set_h_dptr(parent, bdst, NULL);
-+ au_set_h_dptr(parent, bdst, dget(h_tmp));
-+ h_tmpdir = h_tmp->d_inode;
-+ au_set_h_iptr(dir, bdst, NULL, 0);
++ au_set_h_dptr(parent, bdst, dget(h_orph));
++ h_tmpdir = d_inode(h_orph);
+ au_set_h_iptr(dir, bdst, au_igrab(h_tmpdir), /*flags*/0);
+
-+ /* this temporary unlock is safe */
-+ if (file)
-+ h_dentry = au_h_fptr(file, au_fbstart(file))->f_dentry;
-+ else
-+ h_dentry = au_h_dptr(dentry, au_dbstart(dentry));
-+ h_inode = h_dentry->d_inode;
-+ IMustLock(h_inode);
-+ mutex_unlock(&h_inode->i_mutex);
-+ mutex_lock_nested(&h_tmpdir->i_mutex, AuLsc_I_PARENT3);
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ }
-+
-+ if (!au_test_h_perm_sio
-+ (h_tmpdir, MAY_EXEC | MAY_WRITE,
-+ au_test_dlgt(au_mntflags(dentry->d_sb))))
-+ err = au_cpup_wh(dentry, bdst, len, file);
++ inode_lock_nested(h_tmpdir, AuLsc_I_PARENT3);
++ /* todo: au_h_open_pre()? */
++
++ pin_orig = cpg->pin;
++ au_pin_init(&wh_pin, dentry, bdst, AuLsc_DI_PARENT,
++ AuLsc_I_PARENT3, cpg->pin->udba, AuPin_DI_LOCKED);
++ cpg->pin = &wh_pin;
++ }
++
++ h_idmap = au_sbr_idmap(dentry->d_sb, bdst);
++ if (!au_test_h_perm_sio(h_idmap, h_tmpdir, MAY_EXEC | MAY_WRITE)
++ && !au_cpup_sio_test(cpg->pin, d_inode(dentry)->i_mode))
++ err = au_cpup_wh(cpg, file);
+ else {
+ struct au_cpup_wh_args args = {
+ .errp = &err,
-+ .dentry = dentry,
-+ .bdst = bdst,
-+ .len = len,
++ .cpg = cpg,
+ .file = file
+ };
-+ wkq_err = au_wkq_wait(au_call_cpup_wh, &args, /*dlgt*/0);
++ wkq_err = au_wkq_wait(au_call_cpup_wh, &args);
+ if (unlikely(wkq_err))
+ err = wkq_err;
+ }
+
-+ /* todo: is this restore safe? */
-+ if (h_tmp) {
-+ mutex_unlock(&h_tmpdir->i_mutex);
-+ au_set_h_iptr(dir, bdst, NULL, 0);
++ if (h_orph) {
++ inode_unlock(h_tmpdir);
++ /* todo: au_h_open_post()? */
+ au_set_h_iptr(dir, bdst, au_igrab(h_dir), /*flags*/0);
-+ au_set_h_dptr(parent, bdst, NULL);
+ au_set_h_dptr(parent, bdst, h_parent);
++ AuDebugOn(!pin_orig);
++ cpg->pin = pin_orig;
+ }
+ iput(h_dir);
+ dput(parent);
+
-+ AuTraceErr(err);
+ return err;
+}
+
+
+/*
+ * generic routine for both of copy-up and copy-down.
-+ * Although I've tried building a path by dcsub, I gave up this approach.
-+ * Since the ancestor directory may be moved/renamed during copy.
+ */
+/* cf. revalidate function in file.c */
+int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst,
+ int (*cp)(struct dentry *dentry, aufs_bindex_t bdst,
++ struct au_pin *pin,
+ struct dentry *h_parent, void *arg),
+ void *arg)
+{
+ int err;
-+ unsigned char pin_flags;
+ struct au_pin pin;
-+ struct super_block *sb;
-+ struct dentry *d, *parent, *h_parent, *real_parent;
++ struct dentry *d, *parent, *h_parent, *real_parent, *h_dentry;
+
-+ LKTRTrace("%.*s, b%d, parent i%lu\n",
-+ AuDLNPair(dentry), bdst, (unsigned long)parent_ino(dentry));
-+ sb = dentry->d_sb;
-+ AuDebugOn(au_test_ro(sb, bdst, NULL));
+ err = 0;
+ parent = dget_parent(dentry);
-+ IiMustWriteLock(parent->d_inode);
+ if (IS_ROOT(parent))
+ goto out;
+
-+ pin_flags = AuPin_MNT_WRITE;
-+ if (au_opt_test(au_mntflags(sb), UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ au_pin_init(&pin, dentry, bdst, AuLsc_DI_PARENT3, AuLsc_I_PARENT2,
-+ pin_flags);
++ au_pin_init(&pin, dentry, bdst, AuLsc_DI_PARENT2, AuLsc_I_PARENT2,
++ au_opt_udba(dentry->d_sb), AuPin_MNT_WRITE);
+
+ /* do not use au_dpage */
+ real_parent = parent;
+ if (h_parent)
+ goto out; /* success */
+
-+ /* find top dir which is needed to cpup */
++ /* find top dir which is necessary to cpup */
+ do {
+ d = parent;
+ dput(parent);
+ di_write_lock_child3(d);
+
+ /* somebody else might create while we were sleeping */
-+ if (!au_h_dptr(d, bdst) || !au_h_dptr(d, bdst)->d_inode) {
-+ if (au_h_dptr(d, bdst))
-+ au_update_dbstart(d);
++ h_dentry = au_h_dptr(d, bdst);
++ if (!h_dentry || d_is_negative(h_dentry)) {
++ if (h_dentry)
++ au_update_dbtop(d);
+
-+ au_pin_do_set_dentry(pin.pin + AuPin_PARENT, d);
-+ err = au_do_pin(pin.pin + AuPin_PARENT,
-+ au_pin_gp(&pin));
++ au_pin_set_dentry(&pin, d);
++ err = au_do_pin(&pin);
+ if (!err) {
-+ err = cp(d, bdst, h_parent, arg);
++ err = cp(d, bdst, &pin, h_parent, arg);
+ au_unpin(&pin);
+ }
+ }
+ break;
+ }
+
-+ out:
++out:
+ dput(parent);
-+ AuTraceErr(err);
+ return err;
+}
+
+static int au_cpup_dir(struct dentry *dentry, aufs_bindex_t bdst,
-+ struct dentry *h_parent, void *arg)
++ struct au_pin *pin,
++ struct dentry *h_parent __maybe_unused,
++ void *arg __maybe_unused)
+{
-+ int err;
-+
-+ err = au_sio_cpup_simple(dentry, bdst, -1, AuCpup_DTIME);
-+
-+ AuTraceErr(err);
-+ return err;
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = bdst,
++ .bsrc = -1,
++ .len = 0,
++ .pin = pin,
++ .flags = AuCpup_DTIME
++ };
++ return au_sio_cpup_simple(&cpg);
+}
+
+int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst)
+{
-+ int err;
-+
-+ err = au_cp_dirs(dentry, bdst, au_cpup_dir, NULL);
-+
-+ AuTraceErr(err);
-+ return err;
++ return au_cp_dirs(dentry, bdst, au_cpup_dir, NULL);
+}
+
+int au_test_and_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst)
+ struct inode *dir;
+
+ parent = dget_parent(dentry);
-+ dir = parent->d_inode;
-+ LKTRTrace("%.*s, b%d, parent i%lu\n",
-+ AuDLNPair(dentry), bdst, dir->i_ino);
-+ DiMustReadLock(parent);
-+ IiMustReadLock(dir);
-+
++ dir = d_inode(parent);
+ err = 0;
+ if (au_h_iptr(dir, bdst))
+ goto out;
+ err = au_cpup_dirs(dentry, bdst);
+ di_downgrade_lock(parent, AuLock_IR);
+
-+ out:
++out:
+ dput(parent);
-+ AuTraceErr(err);
+ return err;
+}
-diff -urN linux/fs/aufs/cpup.h linux-aufs/fs/aufs/cpup.h
---- linux/fs/aufs/cpup.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/cpup.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,86 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,100 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * copy-up/down functions
-+ *
-+ * $Id$
+ */
+
+#ifndef __AUFS_CPUP_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/fs.h>
-+#include <linux/aufs_type.h>
++#include <linux/path.h>
++
++struct inode;
++struct file;
++struct au_pin;
+
++void au_cpup_attr_flags(struct inode *dst, unsigned int iflags);
+void au_cpup_attr_timesizes(struct inode *inode);
+void au_cpup_attr_nlink(struct inode *inode, int force);
+void au_cpup_attr_changeable(struct inode *inode);
+
+/* ---------------------------------------------------------------------- */
+
++struct au_cp_generic {
++ struct dentry *dentry;
++ aufs_bindex_t bdst, bsrc;
++ loff_t len;
++ struct au_pin *pin;
++ unsigned int flags;
++};
++
+/* cpup flags */
-+#define AuCpup_DTIME 1 /* do dtime_store/revert */
-+#define AuCpup_KEEPLINO (1 << 1) /* do not clear the lower xino,
-+ for link(2) */
++#define AuCpup_DTIME 1 /* do dtime_store/revert */
++#define AuCpup_KEEPLINO (1 << 1) /* do not clear the lower xino,
++ for link(2) */
++#define AuCpup_RENAME (1 << 2) /* rename after cpup */
++#define AuCpup_HOPEN (1 << 3) /* call h_open_pre/post() in
++ cpup */
++#define AuCpup_OVERWRITE (1 << 4) /* allow overwriting the
++ existing entry */
++#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) { (flags) |= AuCpup_##name; }
-+#define au_fclr_cpup(flags, name) { (flags) &= ~AuCpup_##name; }
++#define au_fset_cpup(flags, name) \
++ do { (flags) |= AuCpup_##name; } while (0)
++#define au_fclr_cpup(flags, name) \
++ do { (flags) &= ~AuCpup_##name; } while (0)
+
-+int au_sio_cpup_single(struct dentry *dentry, aufs_bindex_t bdst,
-+ aufs_bindex_t bsrc, loff_t len, unsigned int flags,
-+ struct dentry *dst_parent);
-+int au_sio_cpup_simple(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ unsigned int flags);
-+int au_sio_cpup_wh(struct dentry *dentry, aufs_bindex_t bdst, loff_t len,
-+ struct file *file);
++int au_copy_file(struct file *dst, struct file *src, loff_t len);
++int au_sio_cpup_simple(struct au_cp_generic *cpg);
++int au_sio_cpdown_simple(struct au_cp_generic *cpg);
++int au_sio_cpup_wh(struct au_cp_generic *cpg, struct file *file);
+
+int au_cp_dirs(struct dentry *dentry, aufs_bindex_t bdst,
+ int (*cp)(struct dentry *dentry, aufs_bindex_t bdst,
++ struct au_pin *pin,
+ struct dentry *h_parent, void *arg),
+ void *arg);
+int au_cpup_dirs(struct dentry *dentry, aufs_bindex_t bdst);
+
+/* keep timestamps when copyup */
+struct au_dtime {
-+ struct dentry *dt_dentry, *dt_h_dentry;
-+ struct au_hinode *dt_hinode, *dt_hdir;
-+ struct timespec dt_atime, dt_mtime;
++ struct dentry *dt_dentry;
++ struct path dt_h_path;
++ struct timespec64 dt_atime, dt_mtime;
+};
+void au_dtime_store(struct au_dtime *dt, struct dentry *dentry,
-+ struct dentry *h_dentry, struct au_hinode *hinode,
-+ struct au_hinode *hdir);
++ struct path *h_path);
+void au_dtime_revert(struct au_dtime *dt);
+
-+/* ---------------------------------------------------------------------- */
-+
-+static inline void au_cpup_attr_flags(struct inode *inode,
-+ struct inode *h_inode)
-+{
-+ inode->i_flags |= h_inode->i_flags & ~(S_DEAD | S_PRIVATE);
-+}
-+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_CPUP_H__ */
-diff -urN linux/fs/aufs/dcsub.c linux-aufs/fs/aufs/dcsub.c
---- linux/fs/aufs/dcsub.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dcsub.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,251 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,526 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * sub-routines for dentry cache
-+ *
-+ * $Id$
++ * debugfs interface
+ */
+
++#include <linux/debugfs.h>
+#include "aufs.h"
+
-+static void au_dpage_free(struct au_dpage *dpage)
-+{
-+ int i;
++#ifndef CONFIG_SYSFS
++#error DEBUG_FS depends upon SYSFS
++#endif
+
-+ AuTraceEnter();
-+ AuDebugOn(!dpage);
++static struct dentry *dbgaufs;
++static const mode_t dbgaufs_mode = 0444;
+
-+ for (i = 0; i < dpage->ndentry; i++)
-+ dput(dpage->dentries[i]);
-+ free_page((unsigned long)dpage->dentries);
-+}
++/* 20 is max digits length of ulong 64 */
++struct dbgaufs_arg {
++ int n;
++ char a[20 * 4];
++};
+
-+int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp)
++/*
++ * common function for all XINO files
++ */
++static int dbgaufs_xi_release(struct inode *inode __maybe_unused,
++ struct file *file)
+{
-+ int err;
+ void *p;
+
-+ AuTraceEnter();
++ p = file->private_data;
++ if (p) {
++ /* this is struct dbgaufs_arg */
++ AuDebugOn(!au_kfree_sz_test(p));
++ au_kfree_do_rcu(p);
++ }
++ return 0;
++}
++
++static int dbgaufs_xi_open(struct file *xf, struct file *file, int do_fcnt,
++ int cnt)
++{
++ int err;
++ struct kstat st;
++ struct dbgaufs_arg *p;
+
+ err = -ENOMEM;
-+ dpages->dpages = kmalloc(sizeof(*dpages->dpages), gfp);
-+ if (unlikely(!dpages->dpages))
-+ goto out;
-+ p = (void *)__get_free_page(gfp);
++ p = kmalloc(sizeof(*p), GFP_NOFS);
+ if (unlikely(!p))
-+ goto out_dpages;
-+ dpages->dpages[0].ndentry = 0;
-+ dpages->dpages[0].dentries = p;
-+ dpages->ndpage = 1;
-+ return 0; /* success */
++ goto out;
+
-+ out_dpages:
-+ kfree(dpages->dpages);
-+ out:
-+ AuTraceErr(err);
++ err = 0;
++ p->n = 0;
++ file->private_data = p;
++ if (!xf)
++ goto out;
++
++ err = vfsub_getattr(&xf->f_path, &st);
++ if (!err) {
++ if (do_fcnt)
++ p->n = snprintf
++ (p->a, sizeof(p->a), "%d, %llux%u %lld\n",
++ cnt, st.blocks, st.blksize,
++ (long long)st.size);
++ else
++ 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));
++ } else {
++ p->n = snprintf(p->a, sizeof(p->a), "err %d\n", err);
++ err = 0;
++ }
++
++out:
+ return err;
+}
+
-+void au_dpages_free(struct au_dcsub_pages *dpages)
++static ssize_t dbgaufs_xi_read(struct file *file, char __user *buf,
++ size_t count, loff_t *ppos)
+{
-+ int i;
++ struct dbgaufs_arg *p;
+
-+ AuTraceEnter();
-+
-+ for (i = 0; i < dpages->ndpage; i++)
-+ au_dpage_free(dpages->dpages + i);
-+ kfree(dpages->dpages);
++ p = file->private_data;
++ return simple_read_from_buffer(buf, count, ppos, p->a, p->n);
+}
+
-+static int au_dpages_append(struct au_dcsub_pages *dpages,
-+ struct dentry *dentry, gfp_t gfp)
++/* ---------------------------------------------------------------------- */
++
++struct dbgaufs_plink_arg {
++ int n;
++ char a[];
++};
++
++static int dbgaufs_plink_release(struct inode *inode __maybe_unused,
++ struct file *file)
+{
-+ int err, sz;
-+ struct au_dpage *dpage;
-+ void *p;
++ free_page((unsigned long)file->private_data);
++ return 0;
++}
+
-+ /* AuTraceEnter(); */
++static int dbgaufs_plink_open(struct inode *inode, struct file *file)
++{
++ int err, i, limit;
++ unsigned long n, sum;
++ struct dbgaufs_plink_arg *p;
++ struct au_sbinfo *sbinfo;
++ struct super_block *sb;
++ struct hlist_bl_head *hbl;
+
-+ dpage = dpages->dpages + dpages->ndpage - 1;
-+ AuDebugOn(!dpage);
-+ sz = PAGE_SIZE / sizeof(dentry);
-+ if (unlikely(dpage->ndentry >= sz)) {
-+ LKTRLabel(new dpage);
-+ err = -ENOMEM;
-+ sz = dpages->ndpage * sizeof(*dpages->dpages);
-+ p = au_kzrealloc(dpages->dpages, sz,
-+ sz + sizeof(*dpages->dpages), gfp);
-+ if (unlikely(!p))
-+ goto out;
-+ dpages->dpages = p;
-+ dpage = dpages->dpages + dpages->ndpage;
-+ p = (void *)__get_free_page(gfp);
-+ if (unlikely(!p))
-+ goto out;
-+ dpage->ndentry = 0;
-+ dpage->dentries = p;
-+ dpages->ndpage++;
++ err = -ENOMEM;
++ p = (void *)get_zeroed_page(GFP_NOFS);
++ if (unlikely(!p))
++ goto out;
++
++ err = -EFBIG;
++ sbinfo = inode->i_private;
++ sb = sbinfo->si_sb;
++ si_noflush_read_lock(sb);
++ if (au_opt_test(au_mntflags(sb), PLINK)) {
++ limit = PAGE_SIZE - sizeof(p->n);
++
++ /* the number of buckets */
++ n = snprintf(p->a + p->n, limit, "%d\n", AuPlink_NHASH);
++ p->n += n;
++ limit -= n;
++
++ sum = 0;
++ 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);
++ p->n += n;
++ limit -= n;
++ if (unlikely(limit <= 0))
++ goto out_free;
++ }
++ p->a[p->n - 1] = '\n';
++
++ /* the sum of plinks */
++ n = snprintf(p->a + p->n, limit, "%lu\n", sum);
++ p->n += n;
++ limit -= n;
++ if (unlikely(limit <= 0))
++ goto out_free;
++ } else {
++#define str "1\n0\n0\n"
++ p->n = sizeof(str) - 1;
++ strcpy(p->a, str);
++#undef str
+ }
++ si_read_unlock(sb);
+
-+ dpage->dentries[dpage->ndentry++] = dget(dentry);
-+ return 0; /* success */
++ err = 0;
++ file->private_data = p;
++ goto out; /* success */
+
-+ out:
-+ /* AuTraceErr(err); */
++out_free:
++ free_page((unsigned long)p);
++out:
+ return err;
+}
+
-+int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root,
-+ au_dpages_test test, void *arg)
++static ssize_t dbgaufs_plink_read(struct file *file, char __user *buf,
++ size_t count, loff_t *ppos)
++{
++ struct dbgaufs_plink_arg *p;
++
++ p = file->private_data;
++ return simple_read_from_buffer(buf, count, ppos, p->a, p->n);
++}
++
++static const struct file_operations dbgaufs_plink_fop = {
++ .owner = THIS_MODULE,
++ .open = dbgaufs_plink_open,
++ .release = dbgaufs_plink_release,
++ .read = dbgaufs_plink_read
++};
++
++/* ---------------------------------------------------------------------- */
++
++static int dbgaufs_xib_open(struct inode *inode, struct file *file)
+{
+ int err;
-+ struct dentry *this_parent = root;
-+ struct list_head *next;
-+ struct super_block *sb = root->d_sb;
++ struct au_sbinfo *sbinfo;
++ struct super_block *sb;
++
++ sbinfo = inode->i_private;
++ sb = sbinfo->si_sb;
++ si_noflush_read_lock(sb);
++ err = dbgaufs_xi_open(sbinfo->si_xib, file, /*do_fcnt*/0, /*cnt*/0);
++ si_read_unlock(sb);
++ return err;
++}
+
-+ AuTraceEnter();
++static const struct file_operations dbgaufs_xib_fop = {
++ .owner = THIS_MODULE,
++ .open = dbgaufs_xib_open,
++ .release = dbgaufs_xi_release,
++ .read = dbgaufs_xi_read
++};
+
-+ err = 0;
-+ spin_lock(&dcache_lock);
-+ repeat:
-+ next = this_parent->d_subdirs.next;
-+ resume:
-+ if (this_parent->d_sb == sb
-+ && !IS_ROOT(this_parent)
-+ && atomic_read(&this_parent->d_count)
-+ && this_parent->d_inode
-+ && (!test || test(this_parent, arg))) {
-+ err = au_dpages_append(dpages, this_parent, GFP_ATOMIC);
++/* ---------------------------------------------------------------------- */
++
++#define DbgaufsXi_PREFIX "xi"
++
++static int dbgaufs_xino_open(struct inode *inode, struct file *file)
++{
++ int err, idx;
++ long l;
++ aufs_bindex_t bindex;
++ char *p, a[sizeof(DbgaufsXi_PREFIX) + 8];
++ struct au_sbinfo *sbinfo;
++ struct super_block *sb;
++ struct au_xino *xi;
++ struct file *xf;
++ struct qstr *name;
++ struct au_branch *br;
++
++ err = -ENOENT;
++ name = &file->f_path.dentry->d_name;
++ if (unlikely(name->len < sizeof(DbgaufsXi_PREFIX)
++ || memcmp(name->name, DbgaufsXi_PREFIX,
++ sizeof(DbgaufsXi_PREFIX) - 1)))
++ goto out;
++
++ AuDebugOn(name->len >= sizeof(a));
++ memcpy(a, name->name, name->len);
++ a[name->len] = '\0';
++ p = strchr(a, '-');
++ if (p)
++ *p = '\0';
++ err = kstrtol(a + sizeof(DbgaufsXi_PREFIX) - 1, 10, &l);
++ if (unlikely(err))
++ goto out;
++ bindex = l;
++ idx = 0;
++ if (p) {
++ err = kstrtol(p + 1, 10, &l);
+ if (unlikely(err))
+ goto out;
++ idx = l;
+ }
+
-+ while (next != &this_parent->d_subdirs) {
-+ struct list_head *tmp = next;
-+ struct dentry *dentry = list_entry(tmp, struct dentry,
-+ d_u.d_child);
-+ next = tmp->next;
-+ if (/*d_unhashed(dentry) || */!dentry->d_inode)
-+ continue;
-+ if (!list_empty(&dentry->d_subdirs)) {
-+ this_parent = dentry;
-+ goto repeat;
-+ }
-+ if (dentry->d_sb == sb
-+ && atomic_read(&dentry->d_count)
-+ && (!test || test(dentry, arg))) {
-+ err = au_dpages_append(dpages, dentry, GFP_ATOMIC);
-+ if (unlikely(err))
-+ goto out;
-+ }
-+ }
++ err = -ENOENT;
++ sbinfo = inode->i_private;
++ sb = sbinfo->si_sb;
++ si_noflush_read_lock(sb);
++ if (unlikely(bindex < 0 || bindex > au_sbbot(sb)))
++ goto out_si;
++ br = au_sbr(sb, bindex);
++ xi = br->br_xino;
++ if (unlikely(idx >= xi->xi_nfile))
++ goto out_si;
++ xf = au_xino_file(xi, idx);
++ if (xf)
++ err = dbgaufs_xi_open(xf, file, /*do_fcnt*/1,
++ au_xino_count(br));
+
-+ if (this_parent != root) {
-+ next = this_parent->d_u.d_child.next;
-+ this_parent = this_parent->d_parent; /* dcache_lock is locked */
-+ goto resume;
-+ }
-+ out:
-+ spin_unlock(&dcache_lock);
-+#if 0 /* debug */
-+ if (!err) {
-+ int i, j;
-+ j = 0;
-+ for (i = 0; i < dpages->ndpage; i++) {
-+ if ((dpages->dpages + i)->ndentry)
-+ AuDbg("%d: %d\n",
-+ i, (dpages->dpages + i)->ndentry);
-+ j += (dpages->dpages + i)->ndentry;
-+ }
-+ if (j)
-+ AuDbg("ndpage %d, %d\n", dpages->ndpage, j);
-+ }
-+#endif
++out_si:
++ si_read_unlock(sb);
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry,
-+ int do_include, au_dpages_test test, void *arg)
++static const struct file_operations dbgaufs_xino_fop = {
++ .owner = THIS_MODULE,
++ .open = dbgaufs_xino_open,
++ .release = dbgaufs_xi_release,
++ .read = dbgaufs_xi_read
++};
++
++void dbgaufs_xino_del(struct au_branch *br)
+{
-+ int err;
++ struct dentry *dbgaufs;
+
-+ AuTraceEnter();
++ dbgaufs = br->br_dbgaufs;
++ if (!dbgaufs)
++ return;
+
-+ err = 0;
-+ spin_lock(&dcache_lock);
-+ if (do_include && (!test || test(dentry, arg))) {
-+ err = au_dpages_append(dpages, dentry, GFP_ATOMIC);
-+ if (unlikely(err))
-+ goto out;
++ br->br_dbgaufs = NULL;
++ /* debugfs acquires the parent i_mutex */
++ lockdep_off();
++ debugfs_remove(dbgaufs);
++ lockdep_on();
++}
++
++void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex)
++{
++ aufs_bindex_t bbot;
++ struct au_branch *br;
++
++ if (!au_sbi(sb)->si_dbgaufs)
++ return;
++
++ bbot = au_sbbot(sb);
++ for (; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ dbgaufs_xino_del(br);
+ }
-+ while (!IS_ROOT(dentry)) {
-+ dentry = dentry->d_parent; /* dcache_lock is locked */
-+ if (!test || test(dentry, arg)) {
-+ err = au_dpages_append(dpages, dentry, GFP_ATOMIC);
-+ if (unlikely(err))
-+ break;
++}
++
++static void dbgaufs_br_do_add(struct super_block *sb, aufs_bindex_t bindex,
++ unsigned int idx, struct dentry *parent,
++ struct au_sbinfo *sbinfo)
++{
++ struct au_branch *br;
++ struct dentry *d;
++ /* "xi" bindex(5) "-" idx(2) NULL */
++ char name[sizeof(DbgaufsXi_PREFIX) + 8];
++
++ if (!idx)
++ snprintf(name, sizeof(name), DbgaufsXi_PREFIX "%d", bindex);
++ else
++ snprintf(name, sizeof(name), DbgaufsXi_PREFIX "%d-%u",
++ bindex, idx);
++ br = au_sbr(sb, bindex);
++ if (br->br_dbgaufs) {
++ struct qstr qstr = QSTR_INIT(name, strlen(name));
++
++ if (!au_qstreq(&br->br_dbgaufs->d_name, &qstr)) {
++ /* debugfs acquires the parent i_mutex */
++ lockdep_off();
++ d = debugfs_rename(parent, br->br_dbgaufs, parent,
++ name);
++ lockdep_on();
++ if (unlikely(!d))
++ pr_warn("failed renaming %pd/%s, ignored.\n",
++ parent, name);
+ }
++ } else {
++ lockdep_off();
++ br->br_dbgaufs = debugfs_create_file(name, dbgaufs_mode, parent,
++ sbinfo, &dbgaufs_xino_fop);
++ lockdep_on();
++ if (unlikely(!br->br_dbgaufs))
++ pr_warn("failed creating %pd/%s, ignored.\n",
++ parent, name);
+ }
++}
+
-+ out:
-+ spin_unlock(&dcache_lock);
++static void dbgaufs_br_add(struct super_block *sb, aufs_bindex_t bindex,
++ struct dentry *parent, struct au_sbinfo *sbinfo)
++{
++ struct au_branch *br;
++ struct au_xino *xi;
++ unsigned int u;
+
-+ AuTraceErr(err);
++ br = au_sbr(sb, bindex);
++ xi = br->br_xino;
++ for (u = 0; u < xi->xi_nfile; u++)
++ dbgaufs_br_do_add(sb, bindex, u, parent, sbinfo);
++}
++
++void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex, int topdown)
++{
++ struct au_sbinfo *sbinfo;
++ struct dentry *parent;
++ aufs_bindex_t bbot;
++
++ if (!au_opt_test(au_mntflags(sb), XINO))
++ return;
++
++ sbinfo = au_sbi(sb);
++ parent = sbinfo->si_dbgaufs;
++ if (!parent)
++ return;
++
++ bbot = au_sbbot(sb);
++ if (topdown)
++ for (; bindex <= bbot; bindex++)
++ dbgaufs_br_add(sb, bindex, parent, sbinfo);
++ else
++ for (; bbot >= bindex; bbot--)
++ dbgaufs_br_add(sb, bbot, parent, sbinfo);
++}
++
++/* ---------------------------------------------------------------------- */
++
++#ifdef CONFIG_AUFS_EXPORT
++static int dbgaufs_xigen_open(struct inode *inode, struct file *file)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++ struct super_block *sb;
++
++ sbinfo = inode->i_private;
++ sb = sbinfo->si_sb;
++ si_noflush_read_lock(sb);
++ err = dbgaufs_xi_open(sbinfo->si_xigen, file, /*do_fcnt*/0, /*cnt*/0);
++ si_read_unlock(sb);
+ return err;
+}
+
-+struct dentry *au_test_subdir(struct dentry *d1, struct dentry *d2)
++static const struct file_operations dbgaufs_xigen_fop = {
++ .owner = THIS_MODULE,
++ .open = dbgaufs_xigen_open,
++ .release = dbgaufs_xi_release,
++ .read = dbgaufs_xi_read
++};
++
++static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo)
+{
-+ struct dentry *trap, **dentries;
-+ int err, i, j;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
++ int err;
+
-+ LKTRTrace("%.*s, %.*s\n", AuDLNPair(d1), AuDLNPair(d2));
++ /*
++ * This function is a dynamic '__init' function actually,
++ * so the tiny check for si_rwsem is unnecessary.
++ */
++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */
+
-+ trap = ERR_PTR(-ENOMEM);
-+ err = au_dpages_init(&dpages, GFP_NOFS);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_dcsub_pages_rev(&dpages, d1, /*do_include*/1, NULL, NULL);
-+ if (unlikely(err))
-+ goto out_dpages;
++ err = -EIO;
++ sbinfo->si_dbgaufs_xigen = debugfs_create_file
++ ("xigen", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo,
++ &dbgaufs_xigen_fop);
++ if (sbinfo->si_dbgaufs_xigen)
++ err = 0;
+
-+ trap = d1;
-+ for (i = 0; !err && i < dpages.ndpage; i++) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ for (j = 0; !err && j < dpage->ndentry; j++) {
-+ struct dentry *d;
-+ d = dentries[j];
-+ err = (d == d2);
-+ if (!err)
-+ trap = d;
-+ }
++ return err;
++}
++#else
++static int dbgaufs_xigen_init(struct au_sbinfo *sbinfo)
++{
++ return 0;
++}
++#endif /* CONFIG_AUFS_EXPORT */
++
++/* ---------------------------------------------------------------------- */
++
++void dbgaufs_si_fin(struct au_sbinfo *sbinfo)
++{
++ /*
++ * This function is a dynamic '__fin' function actually,
++ * so the tiny check for si_rwsem is unnecessary.
++ */
++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */
++
++ debugfs_remove_recursive(sbinfo->si_dbgaufs);
++ sbinfo->si_dbgaufs = NULL;
++}
++
++int dbgaufs_si_init(struct au_sbinfo *sbinfo)
++{
++ int err;
++ char name[SysaufsSiNameLen];
++
++ /*
++ * This function is a dynamic '__init' function actually,
++ * so the tiny check for si_rwsem is unnecessary.
++ */
++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */
++
++ err = -ENOENT;
++ if (!dbgaufs) {
++ AuErr1("/debug/aufs is uninitialized\n");
++ goto out;
+ }
++
++ err = -EIO;
++ sysaufs_name(sbinfo, name);
++ sbinfo->si_dbgaufs = debugfs_create_dir(name, dbgaufs);
++ if (unlikely(!sbinfo->si_dbgaufs))
++ goto out;
++
++ /* regardless plink/noplink option */
++ sbinfo->si_dbgaufs_plink = debugfs_create_file
++ ("plink", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo,
++ &dbgaufs_plink_fop);
++ if (unlikely(!sbinfo->si_dbgaufs_plink))
++ goto out_dir;
++
++ /* regardless xino/noxino option */
++ sbinfo->si_dbgaufs_xib = debugfs_create_file
++ ("xib", dbgaufs_mode, sbinfo->si_dbgaufs, sbinfo,
++ &dbgaufs_xib_fop);
++ if (unlikely(!sbinfo->si_dbgaufs_xib))
++ goto out_dir;
++
++ err = dbgaufs_xigen_init(sbinfo);
+ if (!err)
-+ trap = NULL;
++ goto out; /* success */
+
-+ out_dpages:
-+ au_dpages_free(&dpages);
-+ out:
-+ AuTraceErrPtr(trap);
-+ return trap;
++out_dir:
++ dbgaufs_si_fin(sbinfo);
++out:
++ if (unlikely(err))
++ pr_err("debugfs/aufs failed\n");
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++void dbgaufs_fin(void)
++{
++ debugfs_remove(dbgaufs);
++}
++
++int __init dbgaufs_init(void)
++{
++ int err;
++
++ err = -EIO;
++ dbgaufs = debugfs_create_dir(AUFS_NAME, NULL);
++ if (dbgaufs)
++ err = 0;
++ return err;
+}
-diff -urN linux/fs/aufs/dcsub.h linux-aufs/fs/aufs/dcsub.h
---- linux/fs/aufs/dcsub.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dcsub.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,54 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,53 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * sub-routines for dentry cache
-+ *
-+ * $Id$
++ * debugfs interface
+ */
+
-+#ifndef __AUFS_DCSUB_H__
-+#define __AUFS_DCSUB_H__
++#ifndef __DBGAUFS_H__
++#define __DBGAUFS_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/dcache.h>
-+
-+struct au_dpage {
-+ int ndentry;
-+ struct dentry **dentries;
-+};
-+
-+struct au_dcsub_pages {
-+ int ndpage;
-+ struct au_dpage *dpages;
-+};
++struct super_block;
++struct au_sbinfo;
++struct au_branch;
+
-+/* ---------------------------------------------------------------------- */
-+
-+int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp);
-+void au_dpages_free(struct au_dcsub_pages *dpages);
-+typedef int (*au_dpages_test)(struct dentry *dentry, void *arg);
-+int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root,
-+ au_dpages_test test, void *arg);
-+int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry,
-+ int do_include, au_dpages_test test, void *arg);
-+struct dentry *au_test_subdir(struct dentry *d1, struct dentry *d2);
++#ifdef CONFIG_DEBUG_FS
++/* dbgaufs.c */
++void dbgaufs_xino_del(struct au_branch *br);
++void dbgaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex);
++void dbgaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex, int topdown);
++void dbgaufs_si_fin(struct au_sbinfo *sbinfo);
++int dbgaufs_si_init(struct au_sbinfo *sbinfo);
++void dbgaufs_fin(void);
++int __init dbgaufs_init(void);
++#else
++AuStubVoid(dbgaufs_xino_del, struct au_branch *br)
++AuStubVoid(dbgaufs_brs_del, struct super_block *sb, aufs_bindex_t bindex)
++AuStubVoid(dbgaufs_brs_add, struct super_block *sb, aufs_bindex_t bindex,
++ int topdown)
++AuStubVoid(dbgaufs_si_fin, struct au_sbinfo *sbinfo)
++AuStubInt0(dbgaufs_si_init, struct au_sbinfo *sbinfo)
++AuStubVoid(dbgaufs_fin, void)
++AuStubInt0(__init dbgaufs_init, void)
++#endif /* CONFIG_DEBUG_FS */
+
+#endif /* __KERNEL__ */
-+#endif /* __AUFS_DCSUB_H__ */
-diff -urN linux/fs/aufs/debug.c linux-aufs/fs/aufs/debug.c
---- linux/fs/aufs/debug.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/debug.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,522 @@
++#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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,225 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * debug print functions
-+ *
-+ * $Id$
++ * sub-routines for dentry cache
+ */
+
+#include "aufs.h"
+
-+atomic_t au_cond = ATOMIC_INIT(0);
-+
-+char *au_plevel = KERN_DEBUG;
-+#define dpri(fmt, arg...) do { \
-+ if (LktrCond) \
-+ printk("%s" fmt, au_plevel, ##arg); \
-+} while (0)
-+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_dpri_whlist(struct au_nhash *whlist)
++static void au_dpage_free(struct au_dpage *dpage)
+{
+ int i;
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos;
++ struct dentry **p;
+
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ head = whlist->heads + i;
-+ hlist_for_each_entry(tpos, pos, head, wh_hash)
-+ dpri("b%d, %.*s, %d\n",
-+ tpos->wh_bindex,
-+ tpos->wh_str.len, tpos->wh_str.name,
-+ tpos->wh_str.len);
-+ }
++ p = dpage->dentries;
++ for (i = 0; i < dpage->ndentry; i++)
++ dput(*p++);
++ free_page((unsigned long)dpage->dentries);
+}
+
-+void au_dpri_vdir(struct au_vdir *vdir)
++int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp)
+{
-+ int i;
-+ union au_vdir_deblk_p p;
-+ unsigned char *o;
-+
-+ if (!vdir || IS_ERR(vdir)) {
-+ dpri("err %ld\n", PTR_ERR(vdir));
-+ return;
-+ }
-+
-+ dpri("nblk %d, deblk %p, last{%d, %p}, ver %lu\n",
-+ vdir->vd_nblk, vdir->vd_deblk,
-+ vdir->vd_last.i, vdir->vd_last.p.p, vdir->vd_version);
-+ for (i = 0; i < vdir->vd_nblk; i++) {
-+ p.deblk = vdir->vd_deblk[i];
-+ o = p.p;
-+ dpri("[%d]: %p\n", i, o);
-+ }
-+}
++ int err;
++ void *p;
+
-+static int do_pri_inode(aufs_bindex_t bindex, struct inode *inode,
-+ struct dentry *wh)
-+{
-+ char *n = NULL;
-+ int l = 0, ntfy = 0;
++ err = -ENOMEM;
++ dpages->dpages = kmalloc(sizeof(*dpages->dpages), gfp);
++ if (unlikely(!dpages->dpages))
++ goto out;
+
-+ if (!inode || IS_ERR(inode)) {
-+ dpri("i%d: err %ld\n", bindex, PTR_ERR(inode));
-+ return -1;
-+ }
++ p = (void *)__get_free_page(gfp);
++ if (unlikely(!p))
++ goto out_dpages;
+
-+ /* the type of i_blocks depends upon CONFIG_LSF */
-+ BUILD_BUG_ON(sizeof(inode->i_blocks) != sizeof(unsigned long)
-+ && sizeof(inode->i_blocks) != sizeof(u64));
-+ if (wh) {
-+ n = (void *)wh->d_name.name;
-+ l = wh->d_name.len;
-+ }
++ dpages->dpages[0].ndentry = 0;
++ dpages->dpages[0].dentries = p;
++ dpages->ndpage = 1;
++ return 0; /* success */
+
-+ ntfy = au_test_inotify(inode);
-+ dpri("i%d: i%lu, %s, cnt %d, nl %u, 0%o, ntfy %d, sz %llu, blk %llu,"
-+ " ct %lld, np %lu, st 0x%lx, f 0x%x, g %x%s%.*s\n",
-+ bindex,
-+ inode->i_ino, inode->i_sb ? au_sbtype(inode->i_sb) : "??",
-+ atomic_read(&inode->i_count), inode->i_nlink, inode->i_mode,
-+ ntfy,
-+ i_size_read(inode), (unsigned long long)inode->i_blocks,
-+ (long long)timespec_to_ns(&inode->i_ctime) & 0x0ffff,
-+ inode->i_mapping ? inode->i_mapping->nrpages : 0,
-+ inode->i_state, inode->i_flags, inode->i_generation,
-+ l ? ", wh " : "", l, n);
-+ return 0;
++out_dpages:
++ au_kfree_try_rcu(dpages->dpages);
++out:
++ return err;
+}
+
-+void au_dpri_inode(struct inode *inode)
++void au_dpages_free(struct au_dcsub_pages *dpages)
+{
-+ struct au_iinfo *iinfo;
-+ aufs_bindex_t bindex;
-+ int err;
-+
-+ err = do_pri_inode(-1, inode, NULL);
-+ if (err || !au_test_aufs(inode->i_sb))
-+ return;
++ int i;
++ struct au_dpage *p;
+
-+ iinfo = au_ii(inode);
-+ if (!iinfo)
-+ return;
-+ dpri("i-1: bstart %d, bend %d, gen %d\n",
-+ iinfo->ii_bstart, iinfo->ii_bend, au_iigen(inode));
-+ if (iinfo->ii_bstart < 0)
-+ return;
-+ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend; bindex++)
-+ do_pri_inode(bindex, iinfo->ii_hinode[0 + bindex].hi_inode,
-+ iinfo->ii_hinode[0 + bindex].hi_whdentry);
++ p = dpages->dpages;
++ for (i = 0; i < dpages->ndpage; i++)
++ au_dpage_free(p++);
++ au_kfree_try_rcu(dpages->dpages);
+}
+
-+static int do_pri_dentry(aufs_bindex_t bindex, struct dentry *dentry,
-+ struct list_head *intent)
++static int au_dpages_append(struct au_dcsub_pages *dpages,
++ struct dentry *dentry, gfp_t gfp)
+{
-+ struct dentry *wh = NULL;
++ int err, sz;
++ struct au_dpage *dpage;
++ void *p;
+
-+ if (!dentry || IS_ERR(dentry)) {
-+ dpri("d%d: err %ld\n", bindex, PTR_ERR(dentry));
-+ return -1;
-+ }
-+ /* do not call dget_parent() here */
-+ dpri("d%d: %.*s?/%.*s, %s, cnt %d, flags 0x%x, intent %d\n",
-+ bindex,
-+ AuDLNPair(dentry->d_parent), AuDLNPair(dentry),
-+ dentry->d_sb ? au_sbtype(dentry->d_sb) : "??",
-+ atomic_read(&dentry->d_count), dentry->d_flags, !!intent);
-+ if (bindex >= 0 && dentry->d_inode && au_test_aufs(dentry->d_sb)) {
-+ struct au_iinfo *iinfo = au_ii(dentry->d_inode);
-+ if (iinfo)
-+ wh = iinfo->ii_hinode[0 + bindex].hi_whdentry;
-+ }
-+ do_pri_inode(bindex, dentry->d_inode, wh);
-+ return 0;
-+}
++ dpage = dpages->dpages + dpages->ndpage - 1;
++ sz = PAGE_SIZE / sizeof(dentry);
++ if (unlikely(dpage->ndentry >= sz)) {
++ AuLabel(new dpage);
++ err = -ENOMEM;
++ sz = dpages->ndpage * sizeof(*dpages->dpages);
++ p = au_kzrealloc(dpages->dpages, sz,
++ sz + sizeof(*dpages->dpages), gfp,
++ /*may_shrink*/0);
++ if (unlikely(!p))
++ goto out;
+
-+static struct list_head *au_dbg_h_intent(struct au_dinfo *dinfo,
-+ aufs_bindex_t bindex)
-+{
-+#ifdef CONFIG_AUFS_BR_NFS
-+ return dinfo->di_hdentry[0 + bindex].hd_intent_list;
-+#else
-+ return NULL;
-+#endif
-+}
++ dpages->dpages = p;
++ dpage = dpages->dpages + dpages->ndpage;
++ p = (void *)__get_free_page(gfp);
++ if (unlikely(!p))
++ goto out;
+
-+void au_dpri_dentry(struct dentry *dentry)
-+{
-+ struct au_dinfo *dinfo;
-+ aufs_bindex_t bindex;
-+ int err;
++ dpage->ndentry = 0;
++ dpage->dentries = p;
++ dpages->ndpage++;
++ }
+
-+ err = do_pri_dentry(-1, dentry, NULL);
-+ if (err || !au_test_aufs(dentry->d_sb))
-+ return;
++ AuDebugOn(au_dcount(dentry) <= 0);
++ dpage->dentries[dpage->ndentry++] = dget_dlock(dentry);
++ return 0; /* success */
+
-+ dinfo = au_di(dentry);
-+ if (!dinfo)
-+ return;
-+ dpri("d-1: bstart %d, bend %d, bwh %d, bdiropq %d, gen %d\n",
-+ dinfo->di_bstart, dinfo->di_bend,
-+ dinfo->di_bwh, dinfo->di_bdiropq, au_digen(dentry));
-+ if (dinfo->di_bstart < 0)
-+ return;
-+ for (bindex = dinfo->di_bstart; bindex <= dinfo->di_bend; bindex++)
-+ do_pri_dentry(bindex, dinfo->di_hdentry[0 + bindex].hd_dentry,
-+ au_dbg_h_intent(dinfo, bindex));
++out:
++ return err;
+}
+
-+static int do_pri_file(aufs_bindex_t bindex, struct file *file)
-+{
-+ char a[32];
++/* todo: BAD approach */
++/* copied from linux/fs/dcache.c */
++enum d_walk_ret {
++ D_WALK_CONTINUE,
++ D_WALK_QUIT,
++ D_WALK_NORETRY,
++ D_WALK_SKIP,
++};
+
-+ if (!file || IS_ERR(file)) {
-+ dpri("f%d: err %ld\n", bindex, PTR_ERR(file));
-+ return -1;
-+ }
-+ a[0] = 0;
-+ if (bindex < 0
-+ && file->f_dentry
-+ && au_test_aufs(file->f_dentry->d_sb)
-+ && au_fi(file))
-+ snprintf(a, sizeof(a), ", mmapped %d", au_test_mmapped(file));
-+ dpri("f%d: mode 0x%x, flags 0%o, cnt %ld, pos %llu%s\n",
-+ bindex, file->f_mode, file->f_flags, (long)file_count(file),
-+ file->f_pos, a);
-+ if (file->f_dentry)
-+ do_pri_dentry(bindex, file->f_dentry, NULL);
-+ return 0;
-+}
++extern void d_walk(struct dentry *parent, void *data,
++ enum d_walk_ret (*enter)(void *, struct dentry *));
+
-+void au_dpri_file(struct file *file)
-+{
-+ struct au_finfo *finfo;
-+ aufs_bindex_t bindex;
++struct ac_dpages_arg {
+ int err;
++ struct au_dcsub_pages *dpages;
++ struct super_block *sb;
++ au_dpages_test test;
++ void *arg;
++};
+
-+ err = do_pri_file(-1, file);
-+ if (err || !file->f_dentry || !au_test_aufs(file->f_dentry->d_sb))
-+ return;
++static enum d_walk_ret au_call_dpages_append(void *_arg, struct dentry *dentry)
++{
++ enum d_walk_ret ret;
++ struct ac_dpages_arg *arg = _arg;
+
-+ finfo = au_fi(file);
-+ if (!finfo)
-+ return;
-+ if (finfo->fi_bstart < 0)
-+ return;
-+ for (bindex = finfo->fi_bstart; bindex <= finfo->fi_bend; bindex++) {
-+ struct au_hfile *hf;
-+ hf = finfo->fi_hfile + bindex;
-+ do_pri_file(bindex, hf ? hf->hf_file : NULL);
++ ret = D_WALK_CONTINUE;
++ if (dentry->d_sb == arg->sb
++ && !IS_ROOT(dentry)
++ && au_dcount(dentry) > 0
++ && au_di(dentry)
++ && (!arg->test || arg->test(dentry, arg->arg))) {
++ arg->err = au_dpages_append(arg->dpages, dentry, GFP_ATOMIC);
++ if (unlikely(arg->err))
++ ret = D_WALK_QUIT;
+ }
++
++ return ret;
+}
+
-+static int do_pri_br(aufs_bindex_t bindex, struct au_branch *br)
++int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root,
++ au_dpages_test test, void *arg)
+{
-+ struct vfsmount *mnt;
-+ struct super_block *sb;
++ struct ac_dpages_arg args = {
++ .err = 0,
++ .dpages = dpages,
++ .sb = root->d_sb,
++ .test = test,
++ .arg = arg
++ };
+
-+ if (!br || IS_ERR(br)
-+ || !(mnt = br->br_mnt) || IS_ERR(mnt)
-+ || !(sb = mnt->mnt_sb) || IS_ERR(sb)) {
-+ dpri("s%d: err %ld\n", bindex, PTR_ERR(br));
-+ return -1;
-+ }
++ d_walk(root, &args, au_call_dpages_append);
+
-+ dpri("s%d: {perm 0x%x, cnt %d, wbr %p}, "
-+ "%s, dev 0x%02x%02x, flags 0x%lx, cnt(BIAS) %d, active %d, "
-+ "xino %d\n",
-+ bindex, br->br_perm, au_br_count(br), br->br_wbr,
-+ au_sbtype(sb), MAJOR(sb->s_dev), MINOR(sb->s_dev),
-+ sb->s_flags, sb->s_count - S_BIAS,
-+ atomic_read(&sb->s_active), !!br->br_xino.xi_file);
-+ return 0;
++ return args.err;
+}
+
-+void au_dpri_sb(struct super_block *sb)
++int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry,
++ int do_include, au_dpages_test test, void *arg)
+{
-+ struct au_sbinfo *sbinfo;
-+ aufs_bindex_t bindex;
+ int err;
-+ /* to reuduce stack size */
-+ struct {
-+ struct vfsmount mnt;
-+ struct au_branch fake;
-+ } *a;
+
-+ /* this function can be called from magic sysrq */
-+ a = kzalloc(sizeof(*a), GFP_ATOMIC);
-+ if (unlikely(!a)) {
-+ dpri("no memory\n");
-+ return;
-+ }
++ err = 0;
++ write_seqlock(&rename_lock);
++ spin_lock(&dentry->d_lock);
++ if (do_include
++ && au_dcount(dentry) > 0
++ && (!test || test(dentry, arg)))
++ err = au_dpages_append(dpages, dentry, GFP_ATOMIC);
++ spin_unlock(&dentry->d_lock);
++ if (unlikely(err))
++ goto out;
+
-+ a->mnt.mnt_sb = sb;
-+ a->fake.br_perm = 0;
-+ a->fake.br_mnt = &a->mnt;
-+ a->fake.br_xino.xi_file = NULL;
-+ atomic_set(&a->fake.br_count, 0);
-+ smp_mb(); /* atomic_set */
-+ err = do_pri_br(-1, &a->fake);
-+ kfree(a);
-+ dpri("dev 0x%x\n", sb->s_dev);
-+ if (err || !au_test_aufs(sb))
-+ return;
++ /*
++ * RCU for vfsmount is unnecessary since this is a traverse in a single
++ * mount
++ */
++ while (!IS_ROOT(dentry)) {
++ dentry = dentry->d_parent; /* rename_lock is locked */
++ spin_lock(&dentry->d_lock);
++ if (au_dcount(dentry) > 0
++ && (!test || test(dentry, arg)))
++ err = au_dpages_append(dpages, dentry, GFP_ATOMIC);
++ spin_unlock(&dentry->d_lock);
++ if (unlikely(err))
++ break;
++ }
+
-+ sbinfo = au_sbi(sb);
-+ if (!sbinfo)
-+ return;
-+ dpri("nw %d, gen %u, kobj %d\n",
-+ atomic_read(&sbinfo->si_nowait.nw_len), sbinfo->si_generation,
-+ atomic_read(&sbinfo->si_kobj.kref.refcount));
-+ for (bindex = 0; bindex <= sbinfo->si_bend; bindex++)
-+ do_pri_br(bindex, sbinfo->si_branch[0 + bindex]);
++out:
++ write_sequnlock(&rename_lock);
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_dbg_sleep(int sec)
++static inline int au_dcsub_dpages_aufs(struct dentry *dentry, void *arg)
+{
-+ static DECLARE_WAIT_QUEUE_HEAD(wq);
-+ wait_event_timeout(wq, 0, sec * HZ);
++ return au_di(dentry) && dentry->d_sb == arg;
+}
+
-+void au_dbg_sleep_jiffy(int jiffy)
++int au_dcsub_pages_rev_aufs(struct au_dcsub_pages *dpages,
++ struct dentry *dentry, int do_include)
+{
-+ static DECLARE_WAIT_QUEUE_HEAD(wq);
-+ wait_event_timeout(wq, 0, jiffy);
++ return au_dcsub_pages_rev(dpages, dentry, do_include,
++ au_dcsub_dpages_aufs, dentry->d_sb);
+}
+
-+void au_dbg_iattr(struct iattr *ia)
++int au_test_subdir(struct dentry *d1, struct dentry *d2)
+{
-+#define AuBit(name) if (ia->ia_valid & ATTR_ ## name) \
-+ dpri(#name "\n")
-+ AuBit(MODE);
-+ AuBit(UID);
-+ AuBit(GID);
-+ AuBit(SIZE);
-+ AuBit(ATIME);
-+ AuBit(MTIME);
-+ AuBit(CTIME);
-+ AuBit(ATIME_SET);
-+ AuBit(MTIME_SET);
-+ AuBit(FORCE);
-+ AuBit(ATTR_FLAG);
-+ AuBit(KILL_SUID);
-+ AuBit(KILL_SGID);
-+ AuBit(FILE);
-+ AuBit(KILL_PRIV);
-+ AuBit(OPEN);
-+ AuBit(TIMES_SET);
-+#undef AuBit
-+ dpri("ia_file %p\n", ia->ia_file);
++ struct path path[2] = {
++ {
++ .dentry = d1
++ },
++ {
++ .dentry = d2
++ }
++ };
++
++ return path_is_under(path + 0, path + 1);
+}
+diff -urN /usr/share/empty/fs/aufs/dcsub.h linux/fs/aufs/dcsub.h
+--- /usr/share/empty/fs/aufs/dcsub.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/dcsub.h 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,137 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * sub-routines for dentry cache
++ */
+
-+void au_debug_sbinfo_init(struct au_sbinfo *sbinfo)
-+{
-+#ifdef ForceInotify
-+ au_opt_set_udba(sbinfo->si_mntflags, UDBA_INOTIFY);
-+#endif
-+#ifdef ForceDlgt
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
-+#endif
-+#ifdef ForceNoPlink
-+ au_opt_clr(sbinfo->si_mntflags, PLINK);
-+#endif
-+#ifdef ForceNoXino
-+ au_opt_clr(sbinfo->si_mntflags, XINO);
-+#endif
-+#ifdef ForceNoRefrof
-+ au_opt_clr(sbinfo->si_mntflags, REFROF);
-+#endif
-+#ifdef ForceShwh
-+ au_opt_set(sbinfo->si_mntflags, SHWH);
-+#endif
++#ifndef __AUFS_DCSUB_H__
++#define __AUFS_DCSUB_H__
+
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+ {
-+ int i;
-+ for (i = 0; i < AuDbgLock_Last; i++)
-+ au_spl_init(sbinfo->si_dbg_lock + i);
-+ }
-+#endif
-+}
++#ifdef __KERNEL__
+
-+int __init au_debug_init(void)
-+{
-+ aufs_bindex_t bindex;
-+ struct au_vdir_destr destr;
++#include <linux/dcache.h>
++#include <linux/fs.h>
+
-+ bindex = -1;
-+ AuDebugOn(bindex >= 0);
++struct au_dpage {
++ int ndentry;
++ struct dentry **dentries;
++};
+
-+ destr.len = -1;
-+ AuDebugOn(destr.len < NAME_MAX);
++struct au_dcsub_pages {
++ int ndpage;
++ struct au_dpage *dpages;
++};
+
-+#ifdef CONFIG_4KSTACKS
-+ AuWarn("CONFIG_4KSTACKS is defined.\n");
-+#endif
++/* ---------------------------------------------------------------------- */
+
-+#ifdef ForceBrs
-+ sysaufs_brs = 1;
-+#endif
++/* dcsub.c */
++int au_dpages_init(struct au_dcsub_pages *dpages, gfp_t gfp);
++void au_dpages_free(struct au_dcsub_pages *dpages);
++typedef int (*au_dpages_test)(struct dentry *dentry, void *arg);
++int au_dcsub_pages(struct au_dcsub_pages *dpages, struct dentry *root,
++ au_dpages_test test, void *arg);
++int au_dcsub_pages_rev(struct au_dcsub_pages *dpages, struct dentry *dentry,
++ int do_include, au_dpages_test test, void *arg);
++int au_dcsub_pages_rev_aufs(struct au_dcsub_pages *dpages,
++ struct dentry *dentry, int do_include);
++int au_test_subdir(struct dentry *d1, struct dentry *d2);
+
-+#if 0 /* verbose debug */
-+ {
-+ union {
-+ struct au_branch *br;
-+ struct au_dinfo *di;
-+ struct au_finfo *fi;
-+ struct au_iinfo *ii;
-+ struct au_hinode *hi;
-+ struct au_sbinfo *si;
-+ struct au_vdir_destr *destr;
-+ struct au_vdir_de *de;
-+ struct au_vdir_wh *wh;
-+ struct au_vdir *vd;
-+ } u;
-+
-+ pr_info("br{"
-+ "xino %d, "
-+ "id %d, perm %d, mnt %d, count %d, "
-+ "wbr %d, "
-+ "xup %d, xrun %d, "
-+ "gen %d, "
-+ "sa %d} %d\n",
-+ offsetof(typeof(*u.br), br_xino),
-+ offsetof(typeof(*u.br), br_id),
-+ offsetof(typeof(*u.br), br_perm),
-+ offsetof(typeof(*u.br), br_mnt),
-+ offsetof(typeof(*u.br), br_count),
-+ offsetof(typeof(*u.br), wbr),
-+ offsetof(typeof(*u.br), br_xino_upper),
-+ offsetof(typeof(*u.br), br_xino_running),
-+ offsetof(typeof(*u.br), br_generation),
-+ offsetof(typeof(*u.br), br_sabr),
-+ sizeof(*u.br));
-+ pr_info("di{gen %d, rwsem %d, bstart %d, bend %d, bwh %d, "
-+ "bdiropq %d, hdentry %d} %d\n",
-+ offsetof(typeof(*u.di), di_generation),
-+ offsetof(typeof(*u.di), di_rwsem),
-+ offsetof(typeof(*u.di), di_bstart),
-+ offsetof(typeof(*u.di), di_bend),
-+ offsetof(typeof(*u.di), di_bwh),
-+ offsetof(typeof(*u.di), di_bdiropq),
-+ offsetof(typeof(*u.di), di_hdentry),
-+ sizeof(*u.di));
-+ pr_info("fi{gen %d, rwsem %d, hfile %d, bstart %d, bend %d, "
-+ "h_vm_ops %d, vdir_cach %d} %d\n",
-+ offsetof(typeof(*u.fi), fi_generation),
-+ offsetof(typeof(*u.fi), fi_rwsem),
-+ offsetof(typeof(*u.fi), fi_hfile),
-+ offsetof(typeof(*u.fi), fi_bstart),
-+ offsetof(typeof(*u.fi), fi_bend),
-+ offsetof(typeof(*u.fi), fi_h_vm_ops),
-+ offsetof(typeof(*u.fi), fi_vdir_cache),
-+ sizeof(*u.fi));
-+ pr_info("ii{gen %d, hsb %d, "
-+ "rwsem %d, bstart %d, bend %d, hinode %d, vdir %d} "
-+ "%d\n",
-+ offsetof(typeof(*u.ii), ii_generation),
-+ offsetof(typeof(*u.ii), ii_hsb1),
-+ offsetof(typeof(*u.ii), ii_rwsem),
-+ offsetof(typeof(*u.ii), ii_bstart),
-+ offsetof(typeof(*u.ii), ii_bend),
-+ offsetof(typeof(*u.ii), ii_hinode),
-+ offsetof(typeof(*u.ii), ii_vdir),
-+ sizeof(*u.ii));
-+ pr_info("hi{inode %d, id %d, notify %d, wh %d} %d\n",
-+ offsetof(typeof(*u.hi), hi_inode),
-+ offsetof(typeof(*u.hi), hi_id),
-+ offsetof(typeof(*u.hi), hi_notify),
-+ offsetof(typeof(*u.hi), hi_whdentry),
-+ sizeof(*u.hi));
-+ pr_info("si{nwt %d, rwsem %d, gen %d, stat %d, "
-+ "bend %d, last id %d, br %d, "
-+ "cpup %d, creat %d, ops %d, ops %d, "
-+ "rr %d, mfs %d, "
-+ "mntflags %d, "
-+ "xread %d, xwrite %d, xib %d, xmtx %d, buf %d, "
-+ "xlast %d, xnext %d, "
-+ "rdcache %d, "
-+ "dirwh %d, "
-+ "pl %d, "
-+ "mnt %d, "
-+ "sys %d, "
-+ /* "lvma_l %d, lvma %d" */
-+ "} %d\n",
-+ offsetof(typeof(*u.si), si_nowait),
-+ offsetof(typeof(*u.si), si_rwsem),
-+ offsetof(typeof(*u.si), si_generation),
-+ offsetof(typeof(*u.si), au_si_status),
-+ offsetof(typeof(*u.si), si_bend),
-+ offsetof(typeof(*u.si), si_last_br_id),
-+ offsetof(typeof(*u.si), si_branch),
-+ offsetof(typeof(*u.si), si_wbr_copyup),
-+ offsetof(typeof(*u.si), si_wbr_create),
-+ offsetof(typeof(*u.si), si_wbr_copyup_ops),
-+ offsetof(typeof(*u.si), si_wbr_create_ops),
-+ offsetof(typeof(*u.si), si_wbr_rr_next),
-+ offsetof(typeof(*u.si), si_wbr_mfs),
-+ offsetof(typeof(*u.si), si_mntflags),
-+ offsetof(typeof(*u.si), si_xread),
-+ offsetof(typeof(*u.si), si_xwrite),
-+ offsetof(typeof(*u.si), si_xib),
-+ offsetof(typeof(*u.si), si_xib_mtx),
-+ offsetof(typeof(*u.si), si_xib_buf),
-+ offsetof(typeof(*u.si), si_xib_last_pindex),
-+ offsetof(typeof(*u.si), si_xib_next_bit),
-+ offsetof(typeof(*u.si), si_rdcache),
-+ offsetof(typeof(*u.si), si_dirwh),
-+ offsetof(typeof(*u.si), si_plink),
-+ offsetof(typeof(*u.si), si_mnt),
-+ offsetof(typeof(*u.si), si_sa),
-+ /*offsetof(typeof(*u.si), si_lvma_lock),
-+ offsetof(typeof(*u.si), si_lvma),*/
-+ sizeof(*u.si));
-+ pr_info("destr{len %d, name %d} %d\n",
-+ offsetof(typeof(*u.destr), len),
-+ offsetof(typeof(*u.destr), name),
-+ sizeof(*u.destr));
-+ pr_info("de{ino %d, type %d, str %d} %d\n",
-+ offsetof(typeof(*u.de), de_ino),
-+ offsetof(typeof(*u.de), de_type),
-+ offsetof(typeof(*u.de), de_str),
-+ sizeof(*u.de));
-+ pr_info("wh{hash %d, bindex %d, str %d} %d\n",
-+ offsetof(typeof(*u.wh), wh_hash),
-+ offsetof(typeof(*u.wh), wh_bindex),
-+ offsetof(typeof(*u.wh), wh_str),
-+ sizeof(*u.wh));
-+ pr_info("vd{deblk %d, nblk %d, last %d, ver %d, jiffy %d} %d\n",
-+ offsetof(typeof(*u.vd), vd_deblk),
-+ offsetof(typeof(*u.vd), vd_nblk),
-+ offsetof(typeof(*u.vd), vd_last),
-+ offsetof(typeof(*u.vd), vd_version),
-+ offsetof(typeof(*u.vd), vd_jiffy),
-+ sizeof(*u.vd));
++/* ---------------------------------------------------------------------- */
++
++/*
++ * todo: in linux-3.13, several similar (but faster) helpers are added to
++ * include/linux/dcache.h. Try them (in the future).
++ */
++
++static inline int au_d_hashed_positive(struct dentry *d)
++{
++ int err;
++ struct inode *inode = d_inode(d);
++
++ err = 0;
++ if (unlikely(d_unhashed(d)
++ || d_is_negative(d)
++ || !inode->i_nlink))
++ err = -ENOENT;
++ return err;
++}
++
++static inline int au_d_linkable(struct dentry *d)
++{
++ int err;
++ struct inode *inode = d_inode(d);
++
++ err = au_d_hashed_positive(d);
++ if (err
++ && d_is_positive(d)
++ && (inode->i_state & I_LINKABLE))
++ err = 0;
++ return err;
++}
++
++static inline int au_d_alive(struct dentry *d)
++{
++ int err;
++ struct inode *inode;
++
++ err = 0;
++ if (!IS_ROOT(d))
++ err = au_d_hashed_positive(d);
++ else {
++ inode = d_inode(d);
++ if (unlikely(d_unlinked(d)
++ || d_is_negative(d)
++ || !inode->i_nlink))
++ err = -ENOENT;
+ }
-+#endif
++ return err;
++}
+
-+ return 0;
++static inline int au_alive_dir(struct dentry *d)
++{
++ int err;
++
++ err = au_d_alive(d);
++ if (unlikely(err || IS_DEADDIR(d_inode(d))))
++ err = -ENOENT;
++ return err;
++}
++
++static inline int au_qstreq(struct qstr *a, struct qstr *b)
++{
++ return a->len == b->len
++ && !memcmp(a->name, b->name, a->len);
+}
-diff -urN linux/fs/aufs/debug.h linux-aufs/fs/aufs/debug.h
---- linux/fs/aufs/debug.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/debug.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,333 @@
++
++/*
++ * by the commit
++ * 360f547 2015-01-25 dcache: let the dentry count go down to zero without
++ * taking d_lock
++ * the type of d_lockref.count became int, but the inlined function d_count()
++ * still returns unsigned int.
++ * I don't know why. Maybe it is for every d_count() users?
++ * Anyway au_dcount() lives on.
++ */
++static inline int au_dcount(struct dentry *d)
++{
++ return (int)d_count(d);
++}
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DCSUB_H__ */
+diff -urN /usr/share/empty/fs/aufs/debug.c linux/fs/aufs/debug.c
+--- /usr/share/empty/fs/aufs/debug.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/debug.c 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,446 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * debug print functions
-+ *
-+ * $Id$
+ */
+
-+#ifndef __AUFS_DEBUG_H__
-+#define __AUFS_DEBUG_H__
-+
-+#ifdef __KERNEL__
-+
-+#include <linux/fs.h>
-+#include <linux/kd.h>
-+#include <linux/vt_kern.h>
-+#include <linux/sysrq.h>
-+
-+/* to debug easier, do not make it an inlined function */
-+#define MtxMustLock(mtx) AuDebugOn(!mutex_is_locked(mtx))
++#include <linux/iversion.h>
++#include "aufs.h"
+
-+#ifdef CONFIG_AUFS_DEBUG
-+/* sparse warns about pointer */
-+#define AuDebugOn(a) BUG_ON(!!(a))
-+extern atomic_t au_cond;
-+#define au_debug_on() atomic_inc_return(&au_cond)
-+#define au_debug_off() atomic_dec_return(&au_cond)
-+static inline int au_debug_test(void)
++/* Returns 0, or -errno. arg is in kp->arg. */
++static int param_atomic_t_set(const char *val, const struct kernel_param *kp)
+{
-+ return atomic_read(&au_cond);
++ int err, n;
++
++ err = kstrtoint(val, 0, &n);
++ if (!err) {
++ if (n > 0)
++ au_debug_on();
++ else
++ au_debug_off();
++ }
++ return err;
+}
-+#else
-+#define AuDebugOn(a) do {} while (0)
-+#define au_debug_on() do {} while (0)
-+#define au_debug_off() do {} while (0)
-+static inline int au_debug_test(void)
++
++/* Returns length written or -errno. Buffer is 4k (ie. be short!) */
++static int param_atomic_t_get(char *buffer, const struct kernel_param *kp)
+{
-+ return 0;
-+}
-+#endif /* CONFIG_AUFS_DEBUG */
++ atomic_t *a;
+
-+/* ---------------------------------------------------------------------- */
++ a = kp->arg;
++ return sprintf(buffer, "%d", atomic_read(a));
++}
+
-+/* debug print */
-+#if defined(CONFIG_LKTR) || defined(CONFIG_LKTR_MODULE)
-+#include <linux/lktr.h>
-+#ifdef CONFIG_AUFS_DEBUG
-+#undef LktrCond
-+#define LktrCond unlikely(au_debug_test() || (lktr_cond && lktr_cond()))
-+#endif
-+#else
-+#define LktrCond au_debug_test()
-+#define LKTRDumpVma(pre, vma, suf) do {} while (0)
-+#define LKTRDumpStack() do {} while (0)
-+#define LKTRTrace(fmt, args...) do { \
-+ if (LktrCond) \
-+ AuDbg(fmt, ##args); \
-+} while (0)
-+#define LKTRLabel(label) LKTRTrace("%s\n", #label)
-+#endif /* CONFIG_LKTR */
++static const struct kernel_param_ops param_ops_atomic_t = {
++ .set = param_atomic_t_set,
++ .get = param_atomic_t_get
++ /* void (*free)(void *arg) */
++};
+
-+#define AuTraceErr(e) do { \
-+ if (unlikely((e) < 0)) \
-+ LKTRTrace("err %d\n", (int)(e)); \
-+} while (0)
++atomic_t aufs_debug = ATOMIC_INIT(0);
++MODULE_PARM_DESC(debug, "debug print");
++module_param_named(debug, aufs_debug, atomic_t, 0664);
+
-+#define AuTraceErrPtr(p) do { \
-+ if (IS_ERR(p)) \
-+ LKTRTrace("err %ld\n", PTR_ERR(p)); \
++DEFINE_MUTEX(au_dbg_mtx); /* just to serialize the dbg msgs */
++char *au_plevel = KERN_DEBUG;
++#define dpri(fmt, ...) do { \
++ if ((au_plevel \
++ && strcmp(au_plevel, KERN_DEBUG)) \
++ || au_debug_test()) \
++ printk("%s" fmt, au_plevel, ##__VA_ARGS__); \
+} while (0)
+
-+#define AuTraceEnter() LKTRLabel(enter)
-+
-+/* dirty macros for debug print, use with "%.*s" and caution */
-+#define AuLNPair(qstr) (qstr)->len, (qstr)->name
-+#define AuDLNPair(d) AuLNPair(&(d)->d_name)
-+
+/* ---------------------------------------------------------------------- */
+
-+#define AuDpri(lvl, fmt, arg...) \
-+ printk(lvl AUFS_NAME " %s:%d:%s[%d]: " fmt, \
-+ __func__, __LINE__, current->comm, current->pid, ##arg)
-+#define AuDbg(fmt, arg...) AuDpri(KERN_DEBUG, fmt, ##arg)
-+#define AuInfo(fmt, arg...) AuDpri(KERN_INFO, fmt, ##arg)
-+#define AuWarn(fmt, arg...) AuDpri(KERN_WARNING, fmt, ##arg)
-+#define AuErr(fmt, arg...) AuDpri(KERN_ERR, fmt, ##arg)
-+#define AuIOErr(fmt, arg...) AuErr("I/O Error, " fmt, ##arg)
-+#define AuIOErrWhck(fmt, arg...) AuErr("I/O Error, try whck. " fmt, ##arg)
-+#define AuWarn1(fmt, arg...) do { \
-+ static unsigned char _c; \
-+ if (!_c++) \
-+ AuWarn(fmt, ##arg); \
-+} while (0)
++void au_dpri_whlist(struct au_nhash *whlist)
++{
++ unsigned long ul, n;
++ struct hlist_head *head;
++ struct au_vdir_wh *pos;
+
-+#define AuErr1(fmt, arg...) do { \
-+ static unsigned char _c; \
-+ if (!_c++) \
-+ AuErr(fmt, ##arg); \
-+} while (0)
++ n = whlist->nh_num;
++ head = whlist->nh_head;
++ for (ul = 0; ul < n; ul++) {
++ hlist_for_each_entry(pos, head, wh_hash)
++ dpri("b%d, %.*s, %d\n",
++ pos->wh_bindex,
++ pos->wh_str.len, pos->wh_str.name,
++ pos->wh_str.len);
++ head++;
++ }
++}
+
-+#define AuIOErr1(fmt, arg...) do { \
-+ static unsigned char _c; \
-+ if (!_c++) \
-+ AuIOErr(fmt, ##arg); \
-+} while (0)
++void au_dpri_vdir(struct au_vdir *vdir)
++{
++ unsigned long ul;
++ union au_vdir_deblk_p p;
++ unsigned char *o;
+
-+#define AuUnsupportMsg "This operation is not supported." \
-+ " Please report this application to aufs-users ML."
-+#define AuUnsupport(fmt, args...) do { \
-+ AuErr(AuUnsupportMsg "\n" fmt, ##args); \
-+ dump_stack(); \
-+} while (0)
++ if (!vdir || IS_ERR(vdir)) {
++ dpri("err %ld\n", PTR_ERR(vdir));
++ return;
++ }
+
-+/* ---------------------------------------------------------------------- */
++ dpri("deblk %u, nblk %lu, deblk %p, last{%lu, %p}, ver %llu\n",
++ vdir->vd_deblk_sz, vdir->vd_nblk, vdir->vd_deblk,
++ vdir->vd_last.ul, vdir->vd_last.p.deblk, vdir->vd_version);
++ for (ul = 0; ul < vdir->vd_nblk; ul++) {
++ p.deblk = vdir->vd_deblk[ul];
++ o = p.deblk;
++ dpri("[%lu]: %p\n", ul, o);
++ }
++}
+
-+struct au_sbinfo;
-+#ifdef CONFIG_AUFS_DEBUG
-+extern char *au_plevel;
-+struct au_nhash;
-+void au_dpri_whlist(struct au_nhash *whlist);
-+struct au_vdir;
-+void au_dpri_vdir(struct au_vdir *vdir);
-+void au_dpri_inode(struct inode *inode);
-+void au_dpri_dentry(struct dentry *dentry);
-+void au_dpri_file(struct file *filp);
-+void au_dpri_sb(struct super_block *sb);
-+void au_dbg_sleep(int sec);
-+void au_dbg_sleep_jiffy(int jiffy);
++static int do_pri_inode(aufs_bindex_t bindex, struct inode *inode, int hn,
++ struct dentry *wh)
++{
++ char *n = NULL;
++ int l = 0;
+
-+#ifndef ATTR_TIMES_SET
-+#define ATTR_TIMES_SET 0
-+#endif
-+void au_dbg_iattr(struct iattr *ia);
-+int __init au_debug_init(void);
-+void au_debug_sbinfo_init(struct au_sbinfo *sbinfo);
-+#define AuDbgWhlist(w) do { \
-+ LKTRTrace(#w "\n"); \
-+ au_dpri_whlist(w); \
-+} while (0)
++ if (!inode || IS_ERR(inode)) {
++ dpri("i%d: err %ld\n", bindex, PTR_ERR(inode));
++ return -1;
++ }
+
-+#define AuDbgVdir(v) do { \
-+ LKTRTrace(#v "\n"); \
-+ au_dpri_vdir(v); \
-+} while (0)
++ /* the type of i_blocks depends upon CONFIG_LBDAF */
++ BUILD_BUG_ON(sizeof(inode->i_blocks) != sizeof(unsigned long)
++ && sizeof(inode->i_blocks) != sizeof(u64));
++ if (wh) {
++ n = (void *)wh->d_name.name;
++ l = wh->d_name.len;
++ }
+
-+#define AuDbgInode(i) do { \
-+ LKTRTrace(#i "\n"); \
-+ au_dpri_inode(i); \
-+} while (0)
++ dpri("i%d: %p, i%lu, %s, cnt %d, nl %u, 0%o, sz %llu, blk %llu,"
++ " acl %p, def_acl %p,"
++ " hn %d, ct %lld, np %lu, st 0x%lx, f 0x%x, v %llu, g %x%s%.*s\n",
++ bindex, inode,
++ inode->i_ino, inode->i_sb ? au_sbtype(inode->i_sb) : "??",
++ atomic_read(&inode->i_count), inode->i_nlink, inode->i_mode,
++ i_size_read(inode), (unsigned long long)inode->i_blocks,
++ inode->i_acl, inode->i_default_acl,
++ hn, (long long)timespec64_to_ns(&inode->i_ctime) & 0x0ffff,
++ inode->i_mapping ? inode->i_mapping->nrpages : 0,
++ inode->i_state, inode->i_flags, inode_peek_iversion(inode),
++ inode->i_generation,
++ l ? ", wh " : "", l, n);
++ return 0;
++}
+
-+#define AuDbgDentry(d) do { \
-+ LKTRTrace(#d "\n"); \
-+ au_dpri_dentry(d); \
-+} while (0)
++void au_dpri_inode(struct inode *inode)
++{
++ struct au_iinfo *iinfo;
++ struct au_hinode *hi;
++ aufs_bindex_t bindex;
++ int err, hn;
+
-+#define AuDbgFile(f) do { \
-+ LKTRTrace(#f "\n"); \
-+ au_dpri_file(f); \
-+} while (0)
++ err = do_pri_inode(-1, inode, -1, NULL);
++ if (err || !au_test_aufs(inode->i_sb) || au_is_bad_inode(inode))
++ return;
+
-+#define AuDbgSb(sb) do { \
-+ LKTRTrace(#sb "\n"); \
-+ au_dpri_sb(sb); \
-+} while (0)
++ iinfo = au_ii(inode);
++ dpri("i-1: btop %d, bbot %d, gen %d\n",
++ iinfo->ii_btop, iinfo->ii_bbot, au_iigen(inode, NULL));
++ if (iinfo->ii_btop < 0)
++ return;
++ hn = 0;
++ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot; bindex++) {
++ hi = au_hinode(iinfo, bindex);
++ hn = !!au_hn(hi);
++ do_pri_inode(bindex, hi->hi_inode, hn, hi->hi_whdentry);
++ }
++}
+
-+#define AuDbgSleep(sec) do { \
-+ AuDbg("sleep %d sec\n", sec); \
-+ au_dbg_sleep(sec); \
-+} while (0)
++void au_dpri_dalias(struct inode *inode)
++{
++ struct dentry *d;
+
-+#define AuDbgSleepJiffy(jiffy) do { \
-+ AuDbg("sleep %d jiffies\n", jiffy); \
-+ au_dbg_sleep_jiffy(jiffy); \
-+} while (0)
++ spin_lock(&inode->i_lock);
++ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias)
++ au_dpri_dentry(d);
++ spin_unlock(&inode->i_lock);
++}
+
-+#define AuDbgIAttr(ia) do { \
-+ AuDbg("ia_valid 0x%x\n", (ia)->ia_valid); \
-+ au_dbg_iattr(ia); \
-+} while (0)
-+#else
-+static inline int au_debug_init(void)
++static int do_pri_dentry(aufs_bindex_t bindex, struct dentry *dentry)
+{
++ struct dentry *wh = NULL;
++ int hn;
++ struct inode *inode;
++ struct au_iinfo *iinfo;
++ struct au_hinode *hi;
++
++ if (!dentry || IS_ERR(dentry)) {
++ dpri("d%d: err %ld\n", bindex, PTR_ERR(dentry));
++ return -1;
++ }
++ /* do not call dget_parent() here */
++ /* note: access d_xxx without d_lock */
++ dpri("d%d: %p, %pd2?, %s, cnt %d, flags 0x%x, %shashed\n",
++ bindex, dentry, dentry,
++ dentry->d_sb ? au_sbtype(dentry->d_sb) : "??",
++ au_dcount(dentry), dentry->d_flags,
++ d_unhashed(dentry) ? "un" : "");
++ hn = -1;
++ inode = NULL;
++ if (d_is_positive(dentry))
++ inode = d_inode(dentry);
++ if (inode
++ && au_test_aufs(dentry->d_sb)
++ && bindex >= 0
++ && !au_is_bad_inode(inode)) {
++ iinfo = au_ii(inode);
++ hi = au_hinode(iinfo, bindex);
++ hn = !!au_hn(hi);
++ wh = hi->hi_whdentry;
++ }
++ do_pri_inode(bindex, inode, hn, wh);
+ return 0;
+}
-+static inline void au_debug_sbinfo_init(struct au_sbinfo *sbinfo)
++
++void au_dpri_dentry(struct dentry *dentry)
+{
-+ /* empty */
-+}
-+#define AuDbgWhlist(w) do {} while (0)
-+#define AuDbgVdir(v) do {} while (0)
-+#define AuDbgInode(i) do {} while (0)
-+#define AuDbgDentry(d) do {} while (0)
-+#define AuDbgFile(f) do {} while (0)
-+#define AuDbgSb(sb) do {} while (0)
-+#define AuDbgSleep(sec) do {} while (0)
-+#define AuDbgSleepJiffy(jiffy) do {} while (0)
-+#define AuDbgIAttr(ia) do {} while (0)
-+#endif /* CONFIG_AUFS_DEBUG */
++ struct au_dinfo *dinfo;
++ aufs_bindex_t bindex;
++ int err;
+
-+#ifdef DbgUdbaRace
-+#define AuDbgSleep_UdbaRace() AuDbgSleep(DbgUdbaRace)
-+#else
-+#define AuDbgSleep_UdbaRace() do {} while (0)
-+#endif
++ err = do_pri_dentry(-1, dentry);
++ if (err || !au_test_aufs(dentry->d_sb))
++ return;
+
-+/* ---------------------------------------------------------------------- */
++ dinfo = au_di(dentry);
++ if (!dinfo)
++ return;
++ dpri("d-1: btop %d, bbot %d, bwh %d, bdiropq %d, gen %d, tmp %d\n",
++ dinfo->di_btop, dinfo->di_bbot,
++ dinfo->di_bwh, dinfo->di_bdiropq, au_digen(dentry),
++ dinfo->di_tmpfile);
++ if (dinfo->di_btop < 0)
++ return;
++ for (bindex = dinfo->di_btop; bindex <= dinfo->di_bbot; bindex++)
++ do_pri_dentry(bindex, au_hdentry(dinfo, bindex)->hd_dentry);
++}
+
-+#ifdef CONFIG_AUFS_MAGIC_SYSRQ
-+int __init au_sysrq_init(void);
-+void au_sysrq_fin(void);
-+
-+#ifdef CONFIG_HW_CONSOLE
-+#define au_dbg_blocked() do { \
-+ WARN_ON(1); \
-+ handle_sysrq('w', vc_cons[fg_console].d->vc_tty); \
-+} while (0)
-+#else
-+#define au_dbg_blocked() do {} while (0)
-+#endif
-+
-+#else
-+static inline int au_sysrq_init(void)
++static int do_pri_file(aufs_bindex_t bindex, struct file *file)
+{
++ char a[32];
++
++ if (!file || IS_ERR(file)) {
++ dpri("f%d: err %ld\n", bindex, PTR_ERR(file));
++ return -1;
++ }
++ a[0] = 0;
++ if (bindex < 0
++ && !IS_ERR_OR_NULL(file->f_path.dentry)
++ && au_test_aufs(file->f_path.dentry->d_sb)
++ && au_fi(file))
++ snprintf(a, sizeof(a), ", gen %d, mmapped %d",
++ au_figen(file), atomic_read(&au_fi(file)->fi_mmapped));
++ dpri("f%d: mode 0x%x, flags 0%o, cnt %ld, v %llu, pos %llu%s\n",
++ bindex, file->f_mode, file->f_flags, (long)file_count(file),
++ file->f_version, file->f_pos, a);
++ if (!IS_ERR_OR_NULL(file->f_path.dentry))
++ do_pri_dentry(bindex, file->f_path.dentry);
+ return 0;
+}
-+#define au_sysrq_fin() do {} while (0)
-+#define au_dbg_blocked() do {} while (0)
-+#endif /* CONFIG_AUFS_MAGIC_SYSRQ */
+
-+enum {
-+ AuDbgLock_SI_LOCKING,
-+ AuDbgLock_SI_LOCKED,
-+ AuDbgLock_DI_LOCKING,
-+ AuDbgLock_DI_LOCKED,
-+ AuDbgLock_II_LOCKING,
-+ AuDbgLock_II_LOCKED,
-+ AuDbgLock_Last
-+};
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+void au_dbg_locking_si_reg(struct super_block *sb, int flags);
-+void au_dbg_locking_si_unreg(struct super_block *sb, int flags);
-+void au_dbg_locked_si_reg(struct super_block *sb, int flags);
-+void au_dbg_locked_si_unreg(struct super_block *sb, int flags);
-+void au_dbg_locking_di_reg(struct dentry *d, int flags, unsigned int lsc);
-+void au_dbg_locking_di_unreg(struct dentry *d, int flags);
-+void au_dbg_locked_di_reg(struct dentry *d, int flags, unsigned int lsc);
-+void au_dbg_locked_di_unreg(struct dentry *d, int flags);
-+void au_dbg_locking_ii_reg(struct inode *i, int flags, unsigned int lsc);
-+void au_dbg_locking_ii_unreg(struct inode *i, int flags);
-+void au_dbg_locked_ii_reg(struct inode *i, int flags, unsigned int lsc);
-+void au_dbg_locked_ii_unreg(struct inode *i, int flags);
-+#else
-+static inline
-+void au_dbg_locking_si_reg(struct super_block *sb, int flags)
-+{
-+ /* empty */
-+}
-+static inline
-+void au_dbg_locking_si_unreg(struct super_block *sb, int flags)
-+{
-+ /* empty */
-+}
-+static inline
-+void au_dbg_locked_si_reg(struct super_block *sb, int flags)
-+{
-+ /* empty */
-+}
-+static inline
-+void au_dbg_locked_si_unreg(struct super_block *sb, int flags)
++void au_dpri_file(struct file *file)
+{
-+ /* empty */
-+}
++ struct au_finfo *finfo;
++ struct au_fidir *fidir;
++ struct au_hfile *hfile;
++ aufs_bindex_t bindex;
++ int err;
+
-+static inline
-+void au_dbg_locking_di_reg(struct dentry *d, int flags, unsigned int lsc)
-+{
-+ /* empty */
-+}
-+static inline
-+void au_dbg_locking_di_unreg(struct dentry *d, int flags)
-+{
-+ /* empty */
++ err = do_pri_file(-1, file);
++ if (err
++ || IS_ERR_OR_NULL(file->f_path.dentry)
++ || !au_test_aufs(file->f_path.dentry->d_sb))
++ return;
++
++ finfo = au_fi(file);
++ if (!finfo)
++ return;
++ if (finfo->fi_btop < 0)
++ return;
++ fidir = finfo->fi_hdir;
++ if (!fidir)
++ do_pri_file(finfo->fi_btop, finfo->fi_htop.hf_file);
++ else
++ for (bindex = finfo->fi_btop;
++ bindex >= 0 && bindex <= fidir->fd_bbot;
++ bindex++) {
++ hfile = fidir->fd_hfile + bindex;
++ do_pri_file(bindex, hfile ? hfile->hf_file : NULL);
++ }
+}
-+static inline
-+void au_dbg_locked_di_reg(struct dentry *d, int flags, unsigned int lsc)
++
++static int do_pri_br(aufs_bindex_t bindex, struct au_branch *br)
+{
-+ /* empty */
++ struct vfsmount *mnt;
++ struct super_block *sb;
++
++ if (!br || IS_ERR(br))
++ goto out;
++ mnt = au_br_mnt(br);
++ if (!mnt || IS_ERR(mnt))
++ goto out;
++ sb = mnt->mnt_sb;
++ if (!sb || IS_ERR(sb))
++ goto out;
++
++ dpri("s%d: {perm 0x%x, id %d, wbr %p}, "
++ "%s, dev 0x%02x%02x, flags 0x%lx, cnt %d, active %d, "
++ "xino %d\n",
++ bindex, br->br_perm, br->br_id, br->br_wbr,
++ au_sbtype(sb), MAJOR(sb->s_dev), MINOR(sb->s_dev),
++ sb->s_flags, sb->s_count,
++ atomic_read(&sb->s_active),
++ !!au_xino_file(br->br_xino, /*idx*/-1));
++ return 0;
++
++out:
++ dpri("s%d: err %ld\n", bindex, PTR_ERR(br));
++ return -1;
+}
-+static inline
-+void au_dbg_locked_di_unreg(struct dentry *d, int flags)
++
++void au_dpri_sb(struct super_block *sb)
+{
-+ /* empty */
++ struct au_sbinfo *sbinfo;
++ aufs_bindex_t bindex;
++ int err;
++ /* to reduce stack size */
++ struct {
++ struct vfsmount mnt;
++ struct au_branch fake;
++ } *a;
++
++ /* this function can be called from magic sysrq */
++ a = kzalloc(sizeof(*a), GFP_ATOMIC);
++ if (unlikely(!a)) {
++ dpri("no memory\n");
++ return;
++ }
++
++ a->mnt.mnt_sb = sb;
++ a->fake.br_path.mnt = &a->mnt;
++ err = do_pri_br(-1, &a->fake);
++ au_kfree_rcu(a);
++ dpri("dev 0x%x\n", sb->s_dev);
++ if (err || !au_test_aufs(sb))
++ return;
++
++ sbinfo = au_sbi(sb);
++ if (!sbinfo)
++ return;
++ 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]);
+}
-+static inline
-+void au_dbg_locking_ii_reg(struct inode *i, int flags, unsigned int lsc)
++
++/* ---------------------------------------------------------------------- */
++
++void __au_dbg_verify_dinode(struct dentry *dentry, const char *func, int line)
+{
-+ /* empty */
++ struct inode *h_inode, *inode = d_inode(dentry);
++ struct dentry *h_dentry;
++ aufs_bindex_t bindex, bbot, bi;
++
++ if (!inode /* || au_di(dentry)->di_lsc == AuLsc_DI_TMP */)
++ return;
++
++ bbot = au_dbbot(dentry);
++ bi = au_ibbot(inode);
++ if (bi < bbot)
++ bbot = bi;
++ bindex = au_dbtop(dentry);
++ bi = au_ibtop(inode);
++ if (bi > bindex)
++ bindex = bi;
++
++ for (; bindex <= bbot; bindex++) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (!h_dentry)
++ continue;
++ h_inode = au_h_iptr(inode, bindex);
++ if (unlikely(h_inode != d_inode(h_dentry))) {
++ au_debug_on();
++ AuDbg("b%d, %s:%d\n", bindex, func, line);
++ AuDbgDentry(dentry);
++ AuDbgInode(inode);
++ au_debug_off();
++ if (au_test_fuse(h_inode->i_sb))
++ WARN_ON_ONCE(1);
++ else
++ BUG();
++ }
++ }
+}
-+static inline
-+void au_dbg_locking_ii_unreg(struct inode *i, int flags)
++
++void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen)
+{
-+ /* empty */
++ int err, i, j;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry **dentries;
++
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ AuDebugOn(err);
++ err = au_dcsub_pages_rev_aufs(&dpages, parent, /*do_include*/1);
++ AuDebugOn(err);
++ for (i = dpages.ndpage - 1; !err && i >= 0; i--) {
++ dpage = dpages.dpages + i;
++ dentries = dpage->dentries;
++ for (j = dpage->ndentry - 1; !err && j >= 0; j--)
++ AuDebugOn(au_digen_test(dentries[j], sigen));
++ }
++ au_dpages_free(&dpages);
+}
-+static inline
-+void au_dbg_locked_ii_reg(struct inode *i, int flags, unsigned int lsc)
++
++void au_dbg_verify_kthread(void)
+{
-+ /* empty */
++ if (au_wkq_test()) {
++ au_dbg_blocked();
++ /*
++ * It may be recursive, but udba=notify between two aufs mounts,
++ * where a single ro branch is shared, is not a problem.
++ */
++ /* WARN_ON(1); */
++ }
+}
-+static inline
-+void au_dbg_locked_ii_unreg(struct inode *i, int flags)
++
++/* ---------------------------------------------------------------------- */
++
++int __init au_debug_init(void)
+{
-+ /* empty */
-+}
-+#endif /* CONFIG_AUFS_DEBUG_LOCK */
++ aufs_bindex_t bindex;
++ struct au_vdir_destr destr;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_DEBUG_H__ */
-diff -urN linux/fs/aufs/dentry.c linux-aufs/fs/aufs/dentry.c
---- linux/fs/aufs/dentry.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dentry.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,972 @@
++ bindex = -1;
++ AuDebugOn(bindex >= 0);
++
++ destr.len = -1;
++ AuDebugOn(destr.len < NAME_MAX);
++
++#ifdef CONFIG_4KSTACKS
++ pr_warn("CONFIG_4KSTACKS is defined.\n");
++#endif
++
++ return 0;
++}
+diff -urN /usr/share/empty/fs/aufs/debug.h linux/fs/aufs/debug.h
+--- /usr/share/empty/fs/aufs/debug.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/debug.h 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,226 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * lookup and dentry operations
-+ *
-+ * $Id$
++ * debug print functions
+ */
+
-+#include "aufs.h"
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * au_lkup_one() is a generic abstract entry function which calls
-+ * lookup_one_len() or __lookup_hash() finally. it is some condisions that makes
-+ * lookup complicated, which are nfs branch, open-intent and dlgt mode.
-+ */
++#ifndef __AUFS_DEBUG_H__
++#define __AUFS_DEBUG_H__
+
-+#if defined(CONFIG_AUFS_BR_NFS) || defined(CONFIG_AUFS_DLGT)
-+/* cf. lookup_one_len() in linux/fs/namei.c */
-+struct dentry *au_lkup_one(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx)
-+{
-+ struct dentry *dentry;
++#ifdef __KERNEL__
+
-+ LKTRTrace("%.*s/%.*s, ndx{%d, 0x%x}\n",
-+ AuDLNPair(parent), len, name, !!ndx->nfsmnt, ndx->flags);
++#include <linux/atomic.h>
++#include <linux/module.h>
++#include <linux/kallsyms.h>
++#include <linux/sysrq.h>
+
-+ ndx->nd_file = NULL;
-+ if (!ndx->nfsmnt)
-+ dentry = au_lkup_one_dlgt(name, parent, len, ndx->flags);
-+ else
-+ dentry = au_lkup_hash(name, parent, len, ndx);
++#ifdef CONFIG_AUFS_DEBUG
++#define AuDebugOn(a) BUG_ON(a)
+
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++/* module parameter */
++extern atomic_t aufs_debug;
++static inline void au_debug_on(void)
++{
++ atomic_inc(&aufs_debug);
++}
++static inline void au_debug_off(void)
++{
++ atomic_dec_if_positive(&aufs_debug);
+}
-+#endif /* CONFIG_AUFS_BR_NFS || CONFIG_AUFS_DLGT */
-+
-+struct au_lkup_one_args {
-+ struct dentry **errp;
-+ const char *name;
-+ struct dentry *parent;
-+ int len;
-+ struct au_ndx *ndx;
-+};
+
-+static void au_call_lkup_one(void *args)
++static inline int au_debug_test(void)
+{
-+ struct au_lkup_one_args *a = args;
-+ *a->errp = au_lkup_one(a->name, a->parent, a->len, a->ndx);
++ return atomic_read(&aufs_debug) > 0;
+}
++#else
++#define AuDebugOn(a) do {} while (0)
++AuStubVoid(au_debug_on, void)
++AuStubVoid(au_debug_off, void)
++AuStubInt0(au_debug_test, void)
++#endif /* CONFIG_AUFS_DEBUG */
+
-+#define AuLkup_ALLOW_NEG 1
-+#define AuLkup_DLGT (1 << 1)
-+#define AuLkup_DIRPERM1 (1 << 2)
-+#define au_ftest_lkup(flags, name) ((flags) & AuLkup_##name)
-+#define au_fset_lkup(flags, name) { (flags) |= AuLkup_##name; }
-+#define au_fclr_lkup(flags, name) { (flags) &= ~AuLkup_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuLkup_DLGT
-+#define AuLkup_DLGT 0
-+#undef AuLkup_DIRPERM1
-+#define AuLkup_DIRPERM1 0
-+#endif
++#define param_check_atomic_t(name, p) __param_check(name, p, atomic_t)
+
-+struct au_do_lookup_args {
-+ unsigned int flags;
-+ mode_t type;
-+ struct nameidata *nd;
-+};
++/* ---------------------------------------------------------------------- */
+
-+/*
-+ * returns positive/negative dentry, NULL or an error.
-+ * NULL means whiteout-ed or not-found.
-+ */
-+static /* noinline_for_stack */
-+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)
-+{
-+ struct dentry *h_dentry;
-+ int err, wh_found, opq;
-+ unsigned char wh_able;
-+ struct inode *h_dir, *h_inode, *inode;
-+ struct qstr *name;
-+ struct super_block *sb;
-+ unsigned int nd_flags;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = args->nd
-+ };
-+ const int allow_neg = au_ftest_lkup(args->flags, ALLOW_NEG);
++/* debug print */
+
-+ LKTRTrace("%.*s/%.*s, b%d, {flags 0x%x, type 0%o, nd %d}\n",
-+ AuDLNPair(h_parent), AuDLNPair(dentry), bindex,
-+ args->flags, args->type, !!args->nd);
-+ if (args->nd)
-+ LKTRTrace("nd{0x%x}\n", args->nd->flags);
-+ AuDebugOn(IS_ROOT(dentry));
-+ h_dir = h_parent->d_inode;
++#define AuDbg(fmt, ...) do { \
++ if (au_debug_test()) \
++ pr_debug("DEBUG: " fmt, ##__VA_ARGS__); \
++} while (0)
++#define AuLabel(l) AuDbg(#l "\n")
++#define AuIOErr(fmt, ...) pr_err("I/O Error, " fmt, ##__VA_ARGS__)
++#define AuWarn1(fmt, ...) do { \
++ static unsigned char _c; \
++ if (!_c++) \
++ pr_warn(fmt, ##__VA_ARGS__); \
++} while (0)
+
-+ nd_flags = 0;
-+ wh_found = 0;
-+ sb = dentry->d_sb;
-+ ndx.nfsmnt = au_nfsmnt(sb, bindex);
-+ if (au_ftest_lkup(args->flags, DLGT))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ if (au_ftest_lkup(args->flags, DIRPERM1))
-+ au_fset_ndx(ndx.flags, DIRPERM1);
-+ LKTRTrace("nfsmnt %p\n", ndx.nfsmnt);
-+ ndx.br = au_sbr(sb, bindex);
-+ wh_able = !!au_br_whable(ndx.br->br_perm);
-+ name = &dentry->d_name;
-+ if (wh_able)
-+ wh_found = au_test_robr_wh(name, h_parent, wh_name,
-+ /*try_sio*/0, &ndx);
-+ h_dentry = ERR_PTR(wh_found);
-+ if (!wh_found)
-+ goto real_lookup;
-+ if (unlikely(wh_found < 0))
-+ goto out;
++#define AuErr1(fmt, ...) do { \
++ static unsigned char _c; \
++ if (!_c++) \
++ pr_err(fmt, ##__VA_ARGS__); \
++} while (0)
+
-+ /* We found a whiteout */
-+ /* au_set_dbend(dentry, bindex); */
-+ au_set_dbwh(dentry, bindex);
-+ if (!allow_neg)
-+ return NULL; /* success */
-+ if (ndx.nd
-+ && au_test_nfs(h_parent->d_sb)
-+ && (ndx.nd->flags & LOOKUP_CREATE)) {
-+ nd_flags = ndx.nd->flags;
-+ ndx.nd->flags &= ~(LOOKUP_OPEN | LOOKUP_CREATE);
-+ }
++#define AuIOErr1(fmt, ...) do { \
++ static unsigned char _c; \
++ if (!_c++) \
++ AuIOErr(fmt, ##__VA_ARGS__); \
++} while (0)
+
-+ real_lookup:
-+ /* do not superio. */
-+ h_dentry = au_lkup_one(name->name, h_parent, name->len, &ndx);
-+ if (IS_ERR(h_dentry))
-+ goto out;
-+ AuDebugOn(d_unhashed(h_dentry));
-+ h_inode = h_dentry->d_inode;
-+ if (!h_inode) {
-+ if (!allow_neg)
-+ goto out_neg;
-+ } else if (wh_found
-+ || (args->type && args->type != (h_inode->i_mode & S_IFMT)))
-+ goto out_neg;
++#define AuUnsupportMsg "This operation is not supported." \
++ " Please report this application to aufs-users ML."
++#define AuUnsupport(fmt, ...) do { \
++ pr_err(AuUnsupportMsg "\n" fmt, ##__VA_ARGS__); \
++ dump_stack(); \
++} while (0)
+
-+ if (au_dbend(dentry) <= bindex)
-+ au_set_dbend(dentry, bindex);
-+ if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry))
-+ au_set_dbstart(dentry, bindex);
-+ au_set_h_dptr(dentry, bindex, h_dentry);
++#define AuTraceErr(e) do { \
++ if (unlikely((e) < 0)) \
++ AuDbg("err %d\n", (int)(e)); \
++} while (0)
+
-+ err = au_br_nfs_h_intent(ndx.nd_file, dentry, bindex, args->nd);
-+ if (unlikely(err)) {
-+ h_dentry = ERR_PTR(err);
++#define AuTraceErrPtr(p) do { \
++ if (IS_ERR(p)) \
++ AuDbg("err %ld\n", PTR_ERR(p)); \
++} while (0)
++
++/* dirty macros for debug print, use with "%.*s" and caution */
++#define AuLNPair(qstr) (qstr)->len, (qstr)->name
++
++/* ---------------------------------------------------------------------- */
++
++struct dentry;
++#ifdef CONFIG_AUFS_DEBUG
++extern struct mutex au_dbg_mtx;
++extern char *au_plevel;
++struct au_nhash;
++void au_dpri_whlist(struct au_nhash *whlist);
++struct au_vdir;
++void au_dpri_vdir(struct au_vdir *vdir);
++struct inode;
++void au_dpri_inode(struct inode *inode);
++void au_dpri_dalias(struct inode *inode);
++void au_dpri_dentry(struct dentry *dentry);
++struct file;
++void au_dpri_file(struct file *filp);
++struct super_block;
++void au_dpri_sb(struct super_block *sb);
++
++#define au_dbg_verify_dinode(d) __au_dbg_verify_dinode(d, __func__, __LINE__)
++void __au_dbg_verify_dinode(struct dentry *dentry, const char *func, int line);
++void au_dbg_verify_gen(struct dentry *parent, unsigned int sigen);
++void au_dbg_verify_kthread(void);
++
++int __init au_debug_init(void);
++
++#define AuDbgWhlist(w) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#w "\n"); \
++ au_dpri_whlist(w); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgVdir(v) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#v "\n"); \
++ au_dpri_vdir(v); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgInode(i) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#i "\n"); \
++ au_dpri_inode(i); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgDAlias(i) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#i "\n"); \
++ au_dpri_dalias(i); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgDentry(d) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#d "\n"); \
++ au_dpri_dentry(d); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgFile(f) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#f "\n"); \
++ au_dpri_file(f); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgSb(sb) do { \
++ mutex_lock(&au_dbg_mtx); \
++ AuDbg(#sb "\n"); \
++ au_dpri_sb(sb); \
++ mutex_unlock(&au_dbg_mtx); \
++} while (0)
++
++#define AuDbgSym(addr) do { \
++ char sym[KSYM_SYMBOL_LEN]; \
++ sprint_symbol(sym, (unsigned long)addr); \
++ AuDbg("%s\n", sym); \
++} while (0)
++#else
++AuStubVoid(au_dbg_verify_dinode, struct dentry *dentry)
++AuStubVoid(au_dbg_verify_gen, struct dentry *parent, unsigned int sigen)
++AuStubVoid(au_dbg_verify_kthread, void)
++AuStubInt0(__init au_debug_init, void)
++
++#define AuDbgWhlist(w) do {} while (0)
++#define AuDbgVdir(v) do {} while (0)
++#define AuDbgInode(i) do {} while (0)
++#define AuDbgDAlias(i) do {} while (0)
++#define AuDbgDentry(d) do {} while (0)
++#define AuDbgFile(f) do {} while (0)
++#define AuDbgSb(sb) do {} while (0)
++#define AuDbgSym(addr) do {} while (0)
++#endif /* CONFIG_AUFS_DEBUG */
++
++/* ---------------------------------------------------------------------- */
++
++#ifdef CONFIG_AUFS_MAGIC_SYSRQ
++int __init au_sysrq_init(void);
++void au_sysrq_fin(void);
++
++#ifdef CONFIG_HW_CONSOLE
++#define au_dbg_blocked() do { \
++ WARN_ON(1); \
++ handle_sysrq('w'); \
++} while (0)
++#else
++AuStubVoid(au_dbg_blocked, void)
++#endif
++
++#else
++AuStubInt0(__init au_sysrq_init, void)
++AuStubVoid(au_sysrq_fin, void)
++AuStubVoid(au_dbg_blocked, void)
++#endif /* CONFIG_AUFS_MAGIC_SYSRQ */
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DEBUG_H__ */
+diff -urN /usr/share/empty/fs/aufs/dentry.c linux/fs/aufs/dentry.c
+--- /usr/share/empty/fs/aufs/dentry.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/dentry.c 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,1168 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * lookup and dentry operations
++ */
++
++#include <linux/iversion.h>
++#include "aufs.h"
++
++/*
++ * 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 au_do_lookup_args *args)
++{
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct au_branch *br;
++ struct mnt_idmap *h_idmap;
++ struct path h_path;
++ int wh_found, opq;
++ unsigned char wh_able;
++ const unsigned char allow_neg = !!au_ftest_lkup(args->flags, ALLOW_NEG);
++ const unsigned char ignore_perm = !!au_ftest_lkup(args->flags,
++ IGNORE_PERM);
++
++ wh_found = 0;
++ br = au_sbr(dentry->d_sb, bindex);
++ h_path.dentry = h_parent;
++ h_path.mnt = au_br_mnt(br);
++ h_idmap = au_br_idmap(br);
++ wh_able = !!au_br_whable(br->br_perm);
++ if (wh_able)
++ wh_found = au_wh_test(h_idmap, &h_path, &args->whname,
++ ignore_perm);
++ h_dentry = ERR_PTR(wh_found);
++ if (!wh_found)
++ goto real_lookup;
++ if (unlikely(wh_found < 0))
++ goto out;
++
++ /* We found a whiteout */
++ /* au_set_dbbot(dentry, bindex); */
++ au_set_dbwh(dentry, bindex);
++ if (!allow_neg)
++ return NULL; /* success */
++
++real_lookup:
++ if (!ignore_perm)
++ h_dentry = vfsub_lkup_one(args->name, &h_path);
++ else
++ h_dentry = au_sio_lkup_one(h_idmap, args->name, &h_path);
++ if (IS_ERR(h_dentry)) {
++ if (PTR_ERR(h_dentry) == -ENAMETOOLONG
++ && !allow_neg)
++ h_dentry = NULL;
+ goto out;
+ }
+
-+ inode = dentry->d_inode;
-+ if (!h_inode || !S_ISDIR(h_inode->i_mode) || !wh_able
-+ || (inode && !S_ISDIR(inode->i_mode)))
++ h_inode = d_inode(h_dentry);
++ if (d_is_negative(h_dentry)) {
++ if (!allow_neg)
++ goto out_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);
++ if (au_dbtop(dentry) < 0 || bindex < au_dbtop(dentry))
++ au_set_dbtop(dentry, bindex);
++ au_set_h_dptr(dentry, bindex, h_dentry);
++
++ if (!d_is_dir(h_dentry)
++ || !wh_able
++ || (d_really_is_positive(dentry) && !d_is_dir(dentry)))
+ goto out; /* success */
+
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ opq = au_diropq_test(h_dentry, &ndx);
-+ mutex_unlock(&h_inode->i_mutex);
++ h_path.dentry = h_dentry;
++ inode_lock_shared_nested(h_inode, AuLsc_I_CHILD);
++ opq = au_diropq_test(h_idmap, &h_path);
++ inode_unlock_shared(h_inode);
+ if (opq > 0)
+ au_set_dbdiropq(dentry, bindex);
+ else if (unlikely(opq < 0)) {
+ }
+ goto out;
+
-+ out_neg:
++out_neg:
+ dput(h_dentry);
+ h_dentry = NULL;
-+ out:
-+ if (nd_flags)
-+ ndx.nd->flags |= (nd_flags & (LOOKUP_OPEN | LOOKUP_CREATE));
-+ AuTraceErrPtr(h_dentry);
++out:
+ return h_dentry;
+}
+
++static int au_test_shwh(struct super_block *sb, const struct qstr *name)
++{
++ if (unlikely(!au_opt_test(au_mntflags(sb), SHWH)
++ && !strncmp(name->name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)))
++ return -EPERM;
++ return 0;
++}
++
+/*
-+ * returns the number of hidden positive dentries,
++ * returns the number of lower positive dentries,
+ * otherwise an error.
+ * can be called at unlinking with @type is zero.
+ */
-+int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type,
-+ struct nameidata *nd)
++int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t btop,
++ unsigned int flags)
+{
+ int npositive, err;
-+ unsigned int mnt_flags;
+ aufs_bindex_t bindex, btail, bdiropq;
-+ unsigned char isdir;
-+ struct qstr whname;
++ unsigned char isdir, dirperm1, dirren;
+ struct au_do_lookup_args args = {
-+ .type = type,
-+ .nd = nd
++ .flags = flags,
++ .name = &dentry->d_name
+ };
-+ const struct qstr *name = &dentry->d_name;
+ struct dentry *parent;
+ struct super_block *sb;
-+ struct inode *inode;
+
-+ LKTRTrace("%.*s, b%d, type 0%o\n", AuLNPair(name), bstart, type);
-+ AuDebugOn(bstart < 0 || IS_ROOT(dentry));
-+
-+ /* dir may not be locked */
-+ parent = dget_parent(dentry);
-+
-+ err = au_test_robr_shwh(dentry->d_sb, name);
++ sb = dentry->d_sb;
++ err = au_test_shwh(sb, args.name);
+ if (unlikely(err))
+ goto out;
+
-+ err = au_wh_name_alloc(name->name, name->len, &whname);
++ err = au_wh_name_alloc(&args.whname, args.name);
+ if (unlikely(err))
+ goto out;
+
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ inode = dentry->d_inode;
-+ isdir = !!(inode && S_ISDIR(inode->i_mode));
-+ args.flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_lkup(args.flags, DLGT);
-+ if (au_test_dirperm1(mnt_flags))
-+ au_fset_lkup(args.flags, DIRPERM1);
-+ if (!type)
-+ au_fset_lkup(args.flags, ALLOW_NEG);
++ 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);
+ btail = au_dbtaildir(parent);
-+ for (bindex = bstart; bindex <= btail; bindex++) {
++ 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_dentry->d_inode)
++ if (d_is_positive(h_dentry))
+ npositive++;
-+ if (type != S_IFDIR)
-+ break;
-+ continue;
++ break;
+ }
+ h_parent = au_h_dptr(parent, bindex);
-+ if (!h_parent)
-+ continue;
-+ h_dir = h_parent->d_inode;
-+ if (!h_dir || !S_ISDIR(h_dir->i_mode))
++ if (!h_parent || !d_is_dir(h_parent))
+ continue;
+
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ h_dentry = au_do_lookup(h_parent, dentry, bindex, &whname,
-+ &args);
-+ mutex_unlock(&h_dir->i_mutex);
++ 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_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_wh;
-+ au_fclr_lkup(args.flags, ALLOW_NEG);
++ goto out_parent;
++ if (h_dentry)
++ au_fclr_lkup(args.flags, ALLOW_NEG);
++ if (dirperm1)
++ au_fset_lkup(args.flags, IGNORE_PERM);
+
-+ if (au_dbwh(dentry) >= 0)
++ if (au_dbwh(dentry) == bindex)
+ break;
+ if (!h_dentry)
+ continue;
-+ h_inode = h_dentry->d_inode;
-+ if (!h_inode)
++ if (d_is_negative(h_dentry))
+ continue;
++ h_inode = d_inode(h_dentry);
+ npositive++;
+ if (!args.type)
+ args.type = h_inode->i_mode & S_IFMT;
+ if (args.type != S_IFDIR)
+ break;
+ else if (isdir) {
-+ /* the type of lowers may be different */
++ /* the type of lower may be different */
+ bdiropq = au_dbdiropq(dentry);
+ 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) {
-+ LKTRLabel(positive);
-+ au_update_dbstart(dentry);
++ AuLabel(positive);
++ au_update_dbtop(dentry);
+ }
+ err = npositive;
-+ if (!au_opt_test(mnt_flags, UDBA_NONE) && au_dbstart(dentry) < 0)
-+ /* both of real entry and whiteout found */
++ if (unlikely(!au_opt_test(au_mntflags(sb), UDBA_NONE)
++ && au_dbtop(dentry) < 0)) {
+ err = -EIO;
++ AuIOErr("both of real entry and whiteout found, %pd, err %d\n",
++ dentry, err);
++ }
+
-+ out_wh:
-+ au_wh_name_free(&whname);
-+ out:
++out_parent:
+ dput(parent);
-+ AuTraceErr(err);
++ au_kfree_try_rcu(args.whname.name);
++ if (dirren)
++ au_dr_lkup_fin(&args);
++out:
+ return err;
+}
+
-+struct dentry *au_sio_lkup_one(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx)
++struct dentry *au_sio_lkup_one(struct mnt_idmap *idmap, struct qstr *name,
++ struct path *ppath)
+{
+ struct dentry *dentry;
+ int wkq_err;
+
-+ LKTRTrace("%.*s/%.*s\n", AuDLNPair(parent), len, name);
-+
-+ if (!au_test_h_perm_sio(parent->d_inode, MAY_EXEC,
-+ au_ftest_ndx(ndx->flags, DLGT)))
-+ dentry = au_lkup_one(name, parent, len, ndx);
++ if (!au_test_h_perm_sio(idmap, d_inode(ppath->dentry), MAY_EXEC))
++ dentry = vfsub_lkup_one(name, ppath);
+ else {
-+ /* todo: ugly? */
-+ unsigned int flags = ndx->flags;
-+ struct au_lkup_one_args args = {
++ struct vfsub_lkup_one_args args = {
+ .errp = &dentry,
+ .name = name,
-+ .parent = parent,
-+ .len = len,
-+ .ndx = ndx
++ .ppath = ppath
+ };
+
-+ au_fclr_ndx(ndx->flags, DLGT);
-+ au_fclr_ndx(ndx->flags, DIRPERM1);
-+ wkq_err = au_wkq_wait(au_call_lkup_one, &args, /*dlgt*/0);
++ wkq_err = au_wkq_wait(vfsub_call_lkup_one, &args);
+ if (unlikely(wkq_err))
+ dentry = ERR_PTR(wkq_err);
-+ ndx->flags = flags;
+ }
+
-+ AuTraceErrPtr(dentry);
+ return dentry;
+}
+
+/*
+ * lookup @dentry on @bindex which should be negative.
+ */
-+int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex)
++int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex, int wh)
+{
+ int err;
-+ struct dentry *parent, *h_parent, *h_dentry;
-+ struct inode *h_dir;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
++ struct dentry *parent, *h_dentry;
++ struct au_branch *br;
++ struct mnt_idmap *h_idmap;
++ struct path h_ppath;
+
-+ LKTRTrace("%.*s, b%d\n", AuDLNPair(dentry), bindex);
-+ /* dir may not be locked */
+ parent = dget_parent(dentry);
-+ AuDebugOn(!parent || !parent->d_inode
-+ || !S_ISDIR(parent->d_inode->i_mode));
-+ h_parent = au_h_dptr(parent, bindex);
-+ AuDebugOn(!h_parent);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!h_dir || !S_ISDIR(h_dir->i_mode));
-+
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ ndx.nfsmnt = au_nfsmnt(sb, bindex);
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ if (au_test_dirperm1(mnt_flags))
-+ au_fset_ndx(ndx.flags, DIRPERM1);
-+ h_dentry = au_sio_lkup_one(dentry->d_name.name, h_parent,
-+ dentry->d_name.len, &ndx);
++ br = au_sbr(dentry->d_sb, bindex);
++ h_ppath.dentry = au_h_dptr(parent, bindex);
++ h_ppath.mnt = au_br_mnt(br);
++ h_idmap = au_br_idmap(br);
++ if (wh)
++ h_dentry = au_whtmp_lkup(h_ppath.dentry, br, &dentry->d_name);
++ else
++ h_dentry = au_sio_lkup_one(h_idmap, &dentry->d_name, &h_ppath);
+ err = PTR_ERR(h_dentry);
+ if (IS_ERR(h_dentry))
+ goto out;
-+ if (unlikely(h_dentry->d_inode)) {
++ if (unlikely(d_is_positive(h_dentry))) {
+ err = -EIO;
-+ AuIOErr("b%d %.*s should be negative.\n",
-+ bindex, AuDLNPair(h_dentry));
++ AuIOErr("%pd should be negative on b%d.\n", h_dentry, bindex);
+ dput(h_dentry);
+ goto out;
+ }
+
-+ if (bindex < au_dbstart(dentry))
-+ au_set_dbstart(dentry, bindex);
-+ if (au_dbend(dentry) < bindex)
-+ au_set_dbend(dentry, bindex);
-+ au_set_h_dptr(dentry, bindex, h_dentry);
+ err = 0;
++ if (bindex < au_dbtop(dentry))
++ au_set_dbtop(dentry, bindex);
++ if (au_dbbot(dentry) < bindex)
++ au_set_dbbot(dentry, bindex);
++ au_set_h_dptr(dentry, bindex, h_dentry);
+
-+ out:
++out:
+ dput(parent);
-+ AuTraceErr(err);
+ return err;
+}
+
++/* ---------------------------------------------------------------------- */
++
++/* subset of struct inode */
++struct au_iattr {
++ unsigned long i_ino;
++ /* unsigned int i_nlink; */
++ kuid_t i_uid;
++ kgid_t i_gid;
++ u64 i_version;
+/*
-+ * returns the number of found hidden positive dentries,
-+ * otherwise an error.
-+ */
-+int au_refresh_hdentry(struct dentry *dentry, mode_t type)
++ loff_t i_size;
++ blkcnt_t i_blocks;
++*/
++ umode_t i_mode;
++};
++
++static void au_iattr_save(struct au_iattr *ia, struct inode *h_inode)
++{
++ ia->i_ino = h_inode->i_ino;
++ /* ia->i_nlink = h_inode->i_nlink; */
++ ia->i_uid = h_inode->i_uid;
++ ia->i_gid = h_inode->i_gid;
++ ia->i_version = inode_query_iversion(h_inode);
++/*
++ ia->i_size = h_inode->i_size;
++ ia->i_blocks = h_inode->i_blocks;
++*/
++ ia->i_mode = (h_inode->i_mode & S_IFMT);
++}
++
++static int au_iattr_test(struct au_iattr *ia, struct inode *h_inode)
++{
++ return ia->i_ino != h_inode->i_ino
++ /* || ia->i_nlink != h_inode->i_nlink */
++ || !uid_eq(ia->i_uid, h_inode->i_uid)
++ || !gid_eq(ia->i_gid, h_inode->i_gid)
++ || !inode_eq_iversion(h_inode, ia->i_version)
++/*
++ || ia->i_size != h_inode->i_size
++ || ia->i_blocks != h_inode->i_blocks
++*/
++ || ia->i_mode != (h_inode->i_mode & S_IFMT);
++}
++
++static int au_h_verify_dentry(struct dentry *h_dentry, struct dentry *h_parent,
++ struct au_branch *br)
++{
++ int err;
++ struct au_iattr ia;
++ struct inode *h_inode;
++ struct dentry *h_d;
++ struct super_block *h_sb;
++ struct path h_ppath;
++
++ err = 0;
++ memset(&ia, -1, sizeof(ia));
++ h_sb = h_dentry->d_sb;
++ h_inode = NULL;
++ if (d_is_positive(h_dentry)) {
++ h_inode = d_inode(h_dentry);
++ au_iattr_save(&ia, h_inode);
++ } else if (au_test_nfs(h_sb) || au_test_fuse(h_sb))
++ /* nfs d_revalidate may return 0 for negative dentry */
++ /* fuse d_revalidate always return 0 for negative dentry */
++ goto out;
++
++ /* main purpose is namei.c:cached_lookup() and d_revalidate */
++ h_ppath.dentry = h_parent;
++ h_ppath.mnt = au_br_mnt(br);
++ h_d = vfsub_lkup_one(&h_dentry->d_name, &h_ppath);
++ err = PTR_ERR(h_d);
++ if (IS_ERR(h_d))
++ goto out;
++
++ err = 0;
++ if (unlikely(h_d != h_dentry
++ || d_inode(h_d) != h_inode
++ || (h_inode && au_iattr_test(&ia, h_inode))))
++ err = au_busy_or_stale();
++ dput(h_d);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir,
++ struct dentry *h_parent, struct au_branch *br)
++{
++ int err;
++
++ err = 0;
++ if (udba == AuOpt_UDBA_REVAL
++ && !au_test_fs_remote(h_dentry->d_sb)) {
++ IMustLock(h_dir);
++ err = (d_inode(h_dentry->d_parent) != h_dir);
++ } else if (udba != AuOpt_UDBA_NONE)
++ err = au_h_verify_dentry(h_dentry, h_parent, br);
++
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static int au_do_refresh_hdentry(struct dentry *dentry, struct dentry *parent)
+{
-+ int npositive, new_sz;
++ int err;
++ aufs_bindex_t new_bindex, bindex, bbot, bwh, bdiropq;
++ struct au_hdentry tmp, *p, *q;
+ struct au_dinfo *dinfo;
+ struct super_block *sb;
-+ struct dentry *parent;
-+ aufs_bindex_t bindex, parent_bend, parent_bstart, bwh, bdiropq, bend;
-+ struct au_hdentry *p;
-+ au_gen_t sgen;
+
-+ LKTRTrace("%.*s, type 0%o\n", AuDLNPair(dentry), type);
+ DiMustWriteLock(dentry);
-+ sb = dentry->d_sb;
-+ AuDebugOn(IS_ROOT(dentry));
-+ sgen = au_sigen(sb);
-+ parent = dget_parent(dentry);
-+ AuDebugOn(au_digen(parent) != sgen
-+ || au_iigen(parent->d_inode) != sgen);
+
-+ npositive = -ENOMEM;
-+ new_sz = sizeof(*dinfo->di_hdentry) * (au_sbend(sb) + 1);
++ sb = dentry->d_sb;
+ dinfo = au_di(dentry);
-+ p = au_kzrealloc(dinfo->di_hdentry, sizeof(*p) * (dinfo->di_bend + 1),
-+ new_sz, GFP_NOFS);
-+ if (unlikely(!p))
-+ goto out;
-+ dinfo->di_hdentry = p;
-+
-+ bend = dinfo->di_bend;
++ bbot = dinfo->di_bbot;
+ bwh = dinfo->di_bwh;
+ bdiropq = dinfo->di_bdiropq;
-+ p += dinfo->di_bstart;
-+ for (bindex = dinfo->di_bstart; bindex <= bend; bindex++, p++) {
-+ struct dentry *hd, *hdp;
-+ struct au_hdentry tmp, *q;
-+ aufs_bindex_t new_bindex;
-+
-+ hd = p->hd_dentry;
-+ if (!hd)
-+ continue;
-+ hdp = dget_parent(hd);
-+ if (hdp == au_h_dptr(parent, bindex)) {
-+ dput(hdp);
++ bindex = dinfo->di_btop;
++ p = au_hdentry(dinfo, bindex);
++ for (; bindex <= bbot; bindex++, p++) {
++ if (!p->hd_dentry)
++ continue;
++
++ new_bindex = au_br_index(sb, p->hd_id);
++ if (new_bindex == bindex)
+ continue;
-+ }
+
-+ new_bindex = au_find_dbindex(parent, hdp);
-+ dput(hdp);
-+ AuDebugOn(new_bindex == bindex);
+ if (dinfo->di_bwh == bindex)
+ bwh = new_bindex;
+ if (dinfo->di_bdiropq == bindex)
+ bdiropq = new_bindex;
-+ /* todo: test more? */
+ if (new_bindex < 0) {
-+ au_hdput(p, /*do_free*/0);
++ au_hdput(p);
+ p->hd_dentry = NULL;
+ continue;
+ }
-+ /* swap two hidden dentries, and loop again */
-+ q = dinfo->di_hdentry + new_bindex;
++
++ /* swap two lower dentries, and loop again */
++ q = au_hdentry(dinfo, new_bindex);
+ tmp = *q;
+ *q = *p;
+ *p = tmp;
+ }
+ }
+
-+ /* todo: test more? */
+ dinfo->di_bwh = -1;
-+ if (bwh >= 0 && bwh <= au_sbend(sb) && au_sbr_whable(sb, bwh))
++ if (bwh >= 0 && bwh <= au_sbbot(sb) && au_sbr_whable(sb, bwh))
+ dinfo->di_bwh = bwh;
++
+ dinfo->di_bdiropq = -1;
+ if (bdiropq >= 0
-+ && bdiropq <= au_sbend(sb)
++ && bdiropq <= au_sbbot(sb)
+ && au_sbr_whable(sb, bdiropq))
+ dinfo->di_bdiropq = bdiropq;
-+ parent_bend = au_dbend(parent);
-+ p = dinfo->di_hdentry;
-+ for (bindex = 0; bindex <= parent_bend; bindex++, p++)
++
++ err = -EIO;
++ dinfo->di_btop = -1;
++ dinfo->di_bbot = -1;
++ bbot = au_dbbot(parent);
++ bindex = 0;
++ p = au_hdentry(dinfo, bindex);
++ for (; bindex <= bbot; bindex++, p++)
+ if (p->hd_dentry) {
-+ dinfo->di_bstart = bindex;
++ dinfo->di_btop = bindex;
+ break;
+ }
-+ p = dinfo->di_hdentry + parent_bend;
-+ for (bindex = parent_bend; bindex >= 0; bindex--, p--)
-+ if (p->hd_dentry) {
-+ dinfo->di_bend = bindex;
-+ break;
++
++ if (dinfo->di_btop >= 0) {
++ bindex = bbot;
++ p = au_hdentry(dinfo, bindex);
++ for (; bindex >= 0; bindex--, p--)
++ if (p->hd_dentry) {
++ dinfo->di_bbot = bindex;
++ err = 0;
++ break;
++ }
++ }
++
++ return err;
++}
++
++static void au_do_hide(struct dentry *dentry)
++{
++ struct inode *inode;
++
++ if (d_really_is_positive(dentry)) {
++ inode = d_inode(dentry);
++ if (!d_is_dir(dentry)) {
++ if (inode->i_nlink && !d_unhashed(dentry))
++ drop_nlink(inode);
++ } else {
++ clear_nlink(inode);
++ /* stop next lookup */
++ inode->i_flags |= S_DEAD;
+ }
++ smp_mb(); /* necessary? */
++ }
++ d_drop(dentry);
++}
+
-+ npositive = 0;
-+ parent_bstart = au_dbstart(parent);
-+ if (type != S_IFDIR && dinfo->di_bstart == parent_bstart)
-+ goto out_dgen; /* success */
++static int au_hide_children(struct dentry *parent)
++{
++ int err, i, j, ndentry;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry *dentry;
+
-+ npositive = au_lkup_dentry(dentry, parent_bstart, type, /*nd*/NULL);
-+ if (npositive < 0)
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ if (unlikely(err))
+ goto out;
-+ if (dinfo->di_bwh >= 0 && dinfo->di_bwh <= dinfo->di_bstart)
-+ d_drop(dentry);
++ err = au_dcsub_pages(&dpages, parent, NULL, NULL);
++ if (unlikely(err))
++ goto out_dpages;
+
-+ out_dgen:
-+ au_update_digen(dentry);
-+ out:
-+ dput(parent);
-+ AuTraceErr(npositive);
-+ return npositive;
++ /* in reverse order */
++ for (i = dpages.ndpage - 1; i >= 0; i--) {
++ dpage = dpages.dpages + i;
++ ndentry = dpage->ndentry;
++ for (j = ndentry - 1; j >= 0; j--) {
++ dentry = dpage->dentries[j];
++ if (dentry != parent)
++ au_do_hide(dentry);
++ }
++ }
++
++out_dpages:
++ au_dpages_free(&dpages);
++out:
++ return err;
+}
+
-+static int au_lock_nd(struct dentry *dentry, struct nameidata *nd)
++static void au_hide(struct dentry *dentry)
+{
-+ int locked = 0;
-+ if (nd && dentry != nd->path.dentry) {
-+ di_read_lock_parent(nd->path.dentry, 0);
-+ locked = 1;
++ int err;
++
++ AuDbgDentry(dentry);
++ if (d_is_dir(dentry)) {
++ /* shrink_dcache_parent(dentry); */
++ err = au_hide_children(dentry);
++ if (unlikely(err))
++ AuIOErr("%pd, failed hiding children, ignored %d\n",
++ dentry, err);
+ }
-+ return locked;
++ au_do_hide(dentry);
+}
+
-+static void au_unlock_nd(int locked, struct nameidata *nd)
++/*
++ * By adding a dirty branch, a cached dentry may be affected in various ways.
++ *
++ * a dirty branch is added
++ * - on the top of layers
++ * - in the middle of layers
++ * - to the bottom of layers
++ *
++ * on the added branch there exists
++ * - a whiteout
++ * - a diropq
++ * - a same named entry
++ * + exist
++ * * negative --> positive
++ * * positive --> positive
++ * - type is unchanged
++ * - type is changed
++ * + doesn't exist
++ * * negative --> negative
++ * * positive --> negative (rejected by au_br_del() for non-dir case)
++ * - none
++ */
++static int au_refresh_by_dinfo(struct dentry *dentry, struct au_dinfo *dinfo,
++ struct au_dinfo *tmp)
+{
-+ if (locked)
-+ di_read_unlock(nd->path.dentry, 0);
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct {
++ struct dentry *dentry;
++ struct inode *inode;
++ mode_t mode;
++ } orig_h, tmp_h = {
++ .dentry = NULL
++ };
++ struct au_hdentry *hd;
++ struct inode *inode, *h_inode;
++ struct dentry *h_dentry;
++
++ err = 0;
++ AuDebugOn(dinfo->di_btop < 0);
++ orig_h.mode = 0;
++ orig_h.dentry = au_hdentry(dinfo, dinfo->di_btop)->hd_dentry;
++ orig_h.inode = NULL;
++ if (d_is_positive(orig_h.dentry)) {
++ orig_h.inode = d_inode(orig_h.dentry);
++ orig_h.mode = orig_h.inode->i_mode & S_IFMT;
++ }
++ if (tmp->di_btop >= 0) {
++ tmp_h.dentry = au_hdentry(tmp, tmp->di_btop)->hd_dentry;
++ if (d_is_positive(tmp_h.dentry)) {
++ tmp_h.inode = d_inode(tmp_h.dentry);
++ tmp_h.mode = tmp_h.inode->i_mode & S_IFMT;
++ }
++ }
++
++ inode = NULL;
++ if (d_really_is_positive(dentry))
++ inode = d_inode(dentry);
++ if (!orig_h.inode) {
++ AuDbg("negative originally\n");
++ if (inode) {
++ au_hide(dentry);
++ goto out;
++ }
++ AuDebugOn(inode);
++ AuDebugOn(dinfo->di_btop != dinfo->di_bbot);
++ AuDebugOn(dinfo->di_bdiropq != -1);
++
++ if (!tmp_h.inode) {
++ AuDbg("negative --> negative\n");
++ /* should have only one negative lower */
++ if (tmp->di_btop >= 0
++ && tmp->di_btop < dinfo->di_btop) {
++ AuDebugOn(tmp->di_btop != tmp->di_bbot);
++ AuDebugOn(dinfo->di_btop != dinfo->di_bbot);
++ au_set_h_dptr(dentry, dinfo->di_btop, NULL);
++ au_di_cp(dinfo, tmp);
++ hd = au_hdentry(tmp, tmp->di_btop);
++ au_set_h_dptr(dentry, tmp->di_btop,
++ dget(hd->hd_dentry));
++ }
++ au_dbg_verify_dinode(dentry);
++ } else {
++ AuDbg("negative --> positive\n");
++ /*
++ * similar to the behaviour of creating with bypassing
++ * aufs.
++ * unhash it in order to force an error in the
++ * succeeding create operation.
++ * we should not set S_DEAD here.
++ */
++ d_drop(dentry);
++ /* au_di_swap(tmp, dinfo); */
++ au_dbg_verify_dinode(dentry);
++ }
++ } else {
++ AuDbg("positive originally\n");
++ /* inode may be NULL */
++ AuDebugOn(inode && (inode->i_mode & S_IFMT) != orig_h.mode);
++ if (!tmp_h.inode) {
++ AuDbg("positive --> negative\n");
++ /* or bypassing aufs */
++ au_hide(dentry);
++ if (tmp->di_bwh >= 0 && tmp->di_bwh <= dinfo->di_btop)
++ dinfo->di_bwh = tmp->di_bwh;
++ if (inode)
++ err = au_refresh_hinode_self(inode);
++ au_dbg_verify_dinode(dentry);
++ } else if (orig_h.mode == tmp_h.mode) {
++ AuDbg("positive --> positive, same type\n");
++ if (!S_ISDIR(orig_h.mode)
++ && dinfo->di_btop > tmp->di_btop) {
++ /*
++ * similar to the behaviour of removing and
++ * creating.
++ */
++ au_hide(dentry);
++ if (inode)
++ err = au_refresh_hinode_self(inode);
++ au_dbg_verify_dinode(dentry);
++ } else {
++ /* fill empty slots */
++ if (dinfo->di_btop > tmp->di_btop)
++ dinfo->di_btop = tmp->di_btop;
++ if (dinfo->di_bbot < tmp->di_bbot)
++ dinfo->di_bbot = tmp->di_bbot;
++ dinfo->di_bwh = tmp->di_bwh;
++ dinfo->di_bdiropq = tmp->di_bdiropq;
++ bbot = dinfo->di_bbot;
++ bindex = tmp->di_btop;
++ hd = au_hdentry(tmp, bindex);
++ for (; bindex <= bbot; bindex++, hd++) {
++ if (au_h_dptr(dentry, bindex))
++ continue;
++ h_dentry = hd->hd_dentry;
++ if (!h_dentry)
++ continue;
++ AuDebugOn(d_is_negative(h_dentry));
++ h_inode = d_inode(h_dentry);
++ AuDebugOn(orig_h.mode
++ != (h_inode->i_mode
++ & S_IFMT));
++ au_set_h_dptr(dentry, bindex,
++ dget(h_dentry));
++ }
++ if (inode)
++ err = au_refresh_hinode(inode, dentry);
++ au_dbg_verify_dinode(dentry);
++ }
++ } else {
++ AuDbg("positive --> positive, different type\n");
++ /* similar to the behaviour of removing and creating */
++ au_hide(dentry);
++ if (inode)
++ err = au_refresh_hinode_self(inode);
++ au_dbg_verify_dinode(dentry);
++ }
++ }
++
++out:
++ return err;
+}
+
-+/* #define TestingFuse */
-+static noinline_for_stack
-+int au_do_h_d_reval(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct nameidata *nd, struct dentry *h_dentry)
++void au_refresh_dop(struct dentry *dentry, int force_reval)
++{
++ const struct dentry_operations *dop
++ = force_reval ? &aufs_dop : dentry->d_sb->s_d_op;
++ static const unsigned int mask
++ = DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE;
++
++ BUILD_BUG_ON(sizeof(mask) != sizeof(dentry->d_flags));
++
++ if (dentry->d_op == dop)
++ return;
++
++ AuDbg("%pd\n", dentry);
++ spin_lock(&dentry->d_lock);
++ if (dop == &aufs_dop)
++ dentry->d_flags |= mask;
++ else
++ dentry->d_flags &= ~mask;
++ dentry->d_op = dop;
++ spin_unlock(&dentry->d_lock);
++}
++
++int au_refresh_dentry(struct dentry *dentry, struct dentry *parent)
+{
-+ int err, valid, e;
-+ int (*reval)(struct dentry *, struct nameidata *);
++ int err, ebrange, nbr;
++ unsigned int sigen;
++ struct au_dinfo *dinfo, *tmp;
+ struct super_block *sb;
-+ struct nameidata fake_nd, *p;
++ struct inode *inode;
+
-+ LKTRTrace("%.*s, b%d, nd %d\n", AuDLNPair(dentry), bindex, !!nd);
++ DiMustWriteLock(dentry);
++ AuDebugOn(IS_ROOT(dentry));
++ AuDebugOn(d_really_is_negative(parent));
+
-+ err = 0;
-+ reval = NULL;
-+ if (h_dentry->d_op)
-+ reval = h_dentry->d_op->d_revalidate;
-+ if (!reval)
++ sb = dentry->d_sb;
++ sigen = au_sigen(sb);
++ err = au_digen_test(parent, sigen);
++ if (unlikely(err))
+ goto out;
+
-+ sb = dentry->d_sb;
-+ if (nd) {
-+ memcpy(&fake_nd, nd, sizeof(*nd));
-+ err = au_fake_intent(&fake_nd, au_sbr(sb, bindex));
-+ if (unlikely(err)) {
-+ err = -EINVAL;
-+ goto out;
++ nbr = au_sbbot(sb) + 1;
++ dinfo = au_di(dentry);
++ err = au_di_realloc(dinfo, nbr, /*may_shrink*/0);
++ if (unlikely(err))
++ goto out;
++ ebrange = au_dbrange_test(dentry);
++ if (!ebrange)
++ ebrange = au_do_refresh_hdentry(dentry, parent);
++
++ if (d_unhashed(dentry) || ebrange /* || dinfo->di_tmpfile */) {
++ AuDebugOn(au_dbtop(dentry) < 0 && au_dbbot(dentry) >= 0);
++ if (d_really_is_positive(dentry)) {
++ inode = d_inode(dentry);
++ err = au_refresh_hinode_self(inode);
+ }
++ au_dbg_verify_dinode(dentry);
++ if (!err)
++ goto out_dgen; /* success */
++ goto out;
++ }
++
++ /* temporary dinfo */
++ AuDbgDentry(dentry);
++ err = -ENOMEM;
++ tmp = au_di_alloc(sb, AuLsc_DI_TMP);
++ if (unlikely(!tmp))
++ goto out;
++ au_di_swap(tmp, dinfo);
++ /* returns the number of positive dentries */
++ /*
++ * if current working dir is removed, it returns an error.
++ * but the dentry is legal.
++ */
++ err = au_lkup_dentry(dentry, /*btop*/0, AuLkup_ALLOW_NEG);
++ AuDbgDentry(dentry);
++ au_di_swap(tmp, dinfo);
++ if (err == -ENOENT)
++ err = 0;
++ if (err >= 0) {
++ /* compare/refresh by dinfo */
++ AuDbgDentry(dentry);
++ err = au_refresh_by_dinfo(dentry, dinfo, tmp);
++ 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))
++ goto out;
++
++out_dgen:
++ au_update_digen(dentry);
++out:
++ if (unlikely(err && !(dentry->d_flags & DCACHE_NFSFS_RENAMED))) {
++ AuIOErr("failed refreshing %pd, %d\n", dentry, err);
++ AuDbgDentry(dentry);
+ }
-+ p = au_fake_dm(&fake_nd, nd, sb, bindex);
-+ AuDebugOn(IS_ERR(p));
-+ AuDebugOn(nd && p != &fake_nd);
-+ LKTRTrace("b%d\n", bindex);
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_do_h_d_reval(struct dentry *h_dentry, unsigned int flags,
++ struct dentry *dentry, aufs_bindex_t bindex)
++{
++ int err, valid;
++
++ err = 0;
++ if (!(h_dentry->d_flags & DCACHE_OP_REVALIDATE))
++ goto out;
+
++ AuDbg("b%d\n", bindex);
++ /*
++ * gave up supporting LOOKUP_CREATE/OPEN for lower fs,
++ * due to whiteout and branch permission.
++ */
++ flags &= ~(/*LOOKUP_PARENT |*/ LOOKUP_OPEN | LOOKUP_CREATE
++ | LOOKUP_FOLLOW | LOOKUP_EXCL);
+ /* it may return tri-state */
-+ valid = reval(h_dentry, p);
++ valid = h_dentry->d_op->d_revalidate(h_dentry, flags);
++
+ if (unlikely(valid < 0))
+ err = valid;
+ else if (!valid)
+ err = -EINVAL;
-+ else
-+ AuDebugOn(err);
-+
-+ if (p) {
-+ AuDebugOn(!nd);
-+ e = au_hin_after_reval(p, dentry, bindex, nd->intent.open.file);
-+#ifndef TestingFuse
-+ au_update_fuse_h_inode(p->path.mnt, h_dentry); /*ignore*/
-+#endif
-+ if (unlikely(e && !err))
-+ err = e;
-+ }
-+#ifndef TestingFuse
-+ else
-+ au_update_fuse_h_inode(NULL, h_dentry); /*ignore*/
-+#endif
-+ au_fake_dm_release(p);
+
-+ out:
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+static noinline_for_stack
-+int h_d_revalidate(struct dentry *dentry, struct inode *inode,
-+ struct nameidata *nd, int do_udba)
++/* todo: remove this */
++static int h_d_revalidate(struct dentry *dentry, struct inode *inode,
++ unsigned int flags, int do_udba, int dirren)
+{
+ int err;
-+ aufs_bindex_t bindex, btail, bstart, ibs, ibe;
-+ unsigned char plus, locked, unhashed, is_root, h_plus;
-+ struct super_block *sb;
-+ struct inode *first, *h_inode, *h_cached_inode;
+ umode_t mode, h_mode;
++ aufs_bindex_t bindex, btail, btop, ibs, ibe;
++ unsigned char plus, unhashed, is_root, h_plus, h_nfs, tmpfile;
++ struct inode *h_inode, *h_cached_inode;
+ struct dentry *h_dentry;
-+ struct qstr *name;
-+
-+ LKTRTrace("%.*s, nd %d\n", AuDLNPair(dentry), !!nd);
-+ AuDebugOn(inode && au_digen(dentry) != au_iigen(inode));
++ struct qstr *name, *h_name;
+
+ err = 0;
-+ sb = dentry->d_sb;
+ plus = 0;
+ mode = 0;
-+ first = NULL;
+ ibs = -1;
+ ibe = -1;
+ unhashed = !!d_unhashed(dentry);
+ is_root = !!IS_ROOT(dentry);
+ name = &dentry->d_name;
++ tmpfile = au_di(dentry)->di_tmpfile;
+
+ /*
-+ * Theoretically, REVAL test should be unnecessary in case of INOTIFY.
-+ * But inotify doesn't fire some necessary events,
++ * Theoretically, REVAL test should be unnecessary in case of
++ * {FS,I}NOTIFY.
++ * But {fs,i}notify doesn't fire some necessary events,
+ * IN_ATTRIB for atime/nlink/pageio
-+ * IN_DELETE for NFS dentry
+ * Let's do REVAL test too.
+ */
+ if (do_udba && inode) {
+ mode = (inode->i_mode & S_IFMT);
+ plus = (inode->i_nlink > 0);
-+ first = au_h_iptr(inode, au_ibstart(inode));
-+ ibs = au_ibstart(inode);
-+ ibe = au_ibend(inode);
++ ibs = au_ibtop(inode);
++ ibe = au_ibbot(inode);
+ }
+
-+ bstart = au_dbstart(dentry);
-+ btail = bstart;
++ btop = au_dbtop(dentry);
++ btail = btop;
+ if (inode && S_ISDIR(inode->i_mode))
+ btail = au_dbtaildir(dentry);
-+ locked = !!au_lock_nd(dentry, nd);
-+ for (bindex = bstart; bindex <= btail; bindex++) {
++ for (bindex = btop; bindex <= btail; bindex++) {
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (!h_dentry)
+ continue;
+
-+ LKTRTrace("b%d, %.*s\n", bindex, AuDLNPair(h_dentry));
-+#ifdef TestingFuse
-+ /* force re-lookup for fuse, in order to update attributes */
-+ if (au_test_fuse(h_dentry->d_sb))
-+ goto err;
-+#endif
-+
++ AuDbg("b%d, %pd\n", bindex, h_dentry);
++ h_nfs = !!au_test_nfs(h_dentry->d_sb);
++ spin_lock(&h_dentry->d_lock);
++ h_name = &h_dentry->d_name;
+ if (unlikely(do_udba
+ && !is_root
-+ && (unhashed != !!d_unhashed(h_dentry)
-+ || name->len != h_dentry->d_name.len
-+ || memcmp(name->name, h_dentry->d_name.name,
-+ name->len)
-+ ))) {
-+ LKTRTrace("unhash 0x%x 0x%x, %.*s %.*s\n",
-+ unhashed, d_unhashed(h_dentry),
-+ AuDLNPair(dentry), AuDLNPair(h_dentry));
++ && ((!h_nfs
++ && (unhashed != !!d_unhashed(h_dentry)
++ || (!tmpfile && !dirren
++ && !au_qstreq(name, h_name))
++ ))
++ || (h_nfs
++ && !(flags & LOOKUP_OPEN)
++ && (h_dentry->d_flags
++ & DCACHE_NFSFS_RENAMED)))
++ )) {
++ int h_unhashed;
++
++ h_unhashed = d_unhashed(h_dentry);
++ spin_unlock(&h_dentry->d_lock);
++ AuDbg("unhash 0x%x 0x%x, %pd %pd\n",
++ unhashed, h_unhashed, dentry, h_dentry);
+ goto err;
+ }
++ spin_unlock(&h_dentry->d_lock);
+
-+ err = au_do_h_d_reval(dentry, bindex, nd, h_dentry);
++ err = au_do_h_d_reval(h_dentry, flags, dentry, bindex);
+ if (unlikely(err))
+ /* do not goto err, to keep the errno */
+ break;
+ continue;
+
+ /* UDBA tests */
-+ h_inode = h_dentry->d_inode;
-+ if (unlikely(!!inode != !!h_inode))
++ if (unlikely(!!inode != d_is_positive(h_dentry)))
+ goto err;
+
++ h_inode = NULL;
++ if (d_is_positive(h_dentry))
++ h_inode = d_inode(h_dentry);
+ h_plus = plus;
+ h_mode = mode;
+ h_cached_inode = h_inode;
+ if (inode && ibs <= bindex && bindex <= ibe)
+ h_cached_inode = au_h_iptr(inode, bindex);
+
-+ LKTRTrace("{%d, 0%o, %p}, h{%d, 0%o, %p}\n",
-+ plus, mode, h_cached_inode,
-+ h_plus, h_mode, h_inode);
-+ if (unlikely(plus != h_plus
-+ || mode != h_mode
++ if (!h_nfs) {
++ if (unlikely(plus != h_plus && !tmpfile))
++ goto err;
++ } else {
++ if (unlikely(!(h_dentry->d_flags & DCACHE_NFSFS_RENAMED)
++ && !is_root
++ && !IS_ROOT(h_dentry)
++ && unhashed != d_unhashed(h_dentry)))
++ goto err;
++ }
++ if (unlikely(mode != h_mode
+ || h_cached_inode != h_inode))
+ goto err;
+ continue;
+
-+ err:
++err:
+ err = -EINVAL;
+ break;
+ }
-+ au_unlock_nd(locked, nd);
-+
-+ /*
-+ * judging by timestamps is meaningless since some filesystem uses
-+ * CURRENT_TIME_SEC instead of CURRENT_TIME.
-+ */
-+ /*
-+ * NFS may stop IN_DELETE because of DCACHE_NFSFS_RENAMED.
-+ */
+
+ AuTraceErr(err);
+ return err;
+}
+
-+static int simple_reval_dpath(struct dentry *dentry, au_gen_t sgen)
++/* todo: consolidate with do_refresh() and au_reval_for_attr() */
++static int simple_reval_dpath(struct dentry *dentry, unsigned int sigen)
+{
+ int err;
-+ mode_t type;
+ struct dentry *parent;
-+ struct inode *inode;
-+
-+ LKTRTrace("%.*s, sgen %d\n", AuDLNPair(dentry), sgen);
-+ SiMustAnyLock(dentry->d_sb);
-+ DiMustWriteLock(dentry);
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode);
+
-+ if (au_digen(dentry) == sgen && au_iigen(inode) == sgen)
++ if (!au_digen_test(dentry, sigen))
+ return 0;
+
+ parent = dget_parent(dentry);
+ di_read_lock_parent(parent, AuLock_IR);
-+ AuDebugOn(au_digen(parent) != sgen
-+ || au_iigen(parent->d_inode) != sgen);
-+#ifdef CONFIG_AUFS_DEBUG
-+ {
-+ int i, j;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
-+ struct dentry **dentries;
-+
-+ err = au_dpages_init(&dpages, GFP_NOFS);
-+ AuDebugOn(err);
-+ err = au_dcsub_pages_rev(&dpages, parent, /*do_include*/1, NULL,
-+ NULL);
-+ AuDebugOn(err);
-+ for (i = dpages.ndpage - 1; !err && i >= 0; i--) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ for (j = dpage->ndentry - 1; !err && j >= 0; j--)
-+ AuDebugOn(au_digen(dentries[j]) != sgen);
-+ }
-+ au_dpages_free(&dpages);
-+ }
-+#endif
-+ type = (inode->i_mode & S_IFMT);
-+ /* returns a number of positive dentries */
-+ err = au_refresh_hdentry(dentry, type);
-+ if (err >= 0)
-+ err = au_refresh_hinode(inode, dentry);
++ AuDebugOn(au_digen_test(parent, sigen));
++ au_dbg_verify_gen(parent, sigen);
++ err = au_refresh_dentry(dentry, parent);
+ di_read_unlock(parent, AuLock_IR);
+ dput(parent);
+ AuTraceErr(err);
+ return err;
+}
+
-+int au_reval_dpath(struct dentry *dentry, au_gen_t sgen)
++int au_reval_dpath(struct dentry *dentry, unsigned int sigen)
+{
+ int err;
+ struct dentry *d, *parent;
-+ struct inode *inode;
-+
-+ LKTRTrace("%.*s, sgen %d\n", AuDLNPair(dentry), sgen);
-+ AuDebugOn(!dentry->d_inode);
-+ DiMustWriteLock(dentry);
+
-+ if (!au_ftest_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIRS))
-+ return simple_reval_dpath(dentry, sgen);
++ if (!au_ftest_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR))
++ return simple_reval_dpath(dentry, sigen);
+
+ /* slow loop, keep it simple and stupid */
+ /* cf: au_cpup_dirs() */
+ err = 0;
+ parent = NULL;
-+ while (au_digen(dentry) != sgen || au_iigen(dentry->d_inode) != sgen) {
++ while (au_digen_test(dentry, sigen)) {
+ d = dentry;
+ while (1) {
+ dput(parent);
+ parent = dget_parent(d);
-+ if (au_digen(parent) == sgen
-+ && au_iigen(parent->d_inode) == sgen)
++ if (!au_digen_test(parent, sigen))
+ break;
+ d = parent;
+ }
+
-+ inode = d->d_inode;
+ if (d != dentry)
-+ di_write_lock_child(d);
++ di_write_lock_child2(d);
+
+ /* someone might update our dentry while we were sleeping */
-+ if (au_digen(d) != sgen || au_iigen(d->d_inode) != sgen) {
++ if (au_digen_test(d, sigen)) {
++ /*
++ * todo: consolidate with simple_reval_dpath(),
++ * do_refresh() and au_reval_for_attr().
++ */
+ di_read_lock_parent(parent, AuLock_IR);
-+ /* returns a number of positive dentries */
-+ err = au_refresh_hdentry(d, inode->i_mode & S_IFMT);
-+ if (err >= 0)
-+ err = au_refresh_hinode(inode, d);
++ err = au_refresh_dentry(d, parent);
+ di_read_unlock(parent, AuLock_IR);
+ }
+
+ break;
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
+/*
-+ * THIS IS A BOOLEAN FUNCTION: returns 1 if valid, 0 otherwise.
-+ * nfsd passes NULL as nameidata.
++ * if valid returns 1, otherwise 0.
+ */
-+static int aufs_d_revalidate(struct dentry *dentry, struct nameidata *nd)
++static int aufs_d_revalidate(struct dentry *dentry, unsigned int flags)
+{
+ int valid, err;
-+ au_gen_t sgen;
-+ unsigned char do_udba;
-+ struct nameidata tmp_nd, *ndp;
++ unsigned int sigen;
++ unsigned char do_udba, dirren;
+ struct super_block *sb;
+ struct inode *inode;
+
-+ LKTRTrace("dentry %.*s\n", AuDLNPair(dentry));
-+ if (nd && nd->path.dentry)
-+ LKTRTrace("nd{%.*s, 0x%x}\n",
-+ AuDLNPair(nd->path.dentry), nd->flags);
-+ /*
-+ * dir case: AuDebugOn(dentry->d_parent != nd->dentry);
-+ * remove failure case:AuDebugOn(!IS_ROOT(dentry)
-+ * && d_unhashed(dentry));
-+ */
-+ AuDebugOn(!dentry->d_fsdata);
++ /* todo: support rcu-walk? */
++ if (flags & LOOKUP_RCU)
++ return -ECHILD;
+
-+ err = -EINVAL;
++ valid = 0;
++ if (unlikely(!au_di(dentry)))
++ goto out;
++
++ valid = 1;
+ sb = dentry->d_sb;
-+ aufs_read_lock(dentry, AuLock_FLUSH | AuLock_DW);
-+ inode = dentry->d_inode;
-+ sgen = au_sigen(sb);
-+ if (au_digen(dentry) != sgen) {
-+ AuDebugOn(IS_ROOT(dentry));
-+#ifdef ForceInotify
-+ AuDbg("UDBA or digen, %.*s\n", AuDLNPair(dentry));
-+#endif
-+ if (inode)
-+ err = au_reval_dpath(dentry, sgen);
-+ if (unlikely(err))
-+ goto out_dgrade;
-+ AuDebugOn(au_digen(dentry) != sgen);
++ /*
++ * todo: very ugly
++ * i_mutex of parent dir may be held,
++ * but we should not return 'invalid' due to busy.
++ */
++ err = aufs_read_lock(dentry, AuLock_FLUSH | AuLock_DW | AuLock_NOPLM);
++ if (unlikely(err)) {
++ valid = err;
++ AuTraceErr(err);
++ goto out;
++ }
++ inode = NULL;
++ if (d_really_is_positive(dentry))
++ inode = d_inode(dentry);
++ if (unlikely(inode && au_is_bad_inode(inode))) {
++ err = -EINVAL;
++ AuTraceErr(err);
++ goto out_dgrade;
+ }
-+ if (inode && au_iigen(inode) != sgen) {
++ if (unlikely(au_dbrange_test(dentry))) {
++ err = -EINVAL;
++ AuTraceErr(err);
++ goto out_dgrade;
++ }
++
++ sigen = au_sigen(sb);
++ if (au_digen_test(dentry, sigen)) {
+ AuDebugOn(IS_ROOT(dentry));
-+#ifdef ForceInotify
-+ AuDbg("UDBA or survived, %.*s\n", AuDLNPair(dentry));
-+#endif
-+ err = au_refresh_hinode(inode, dentry);
-+ if (unlikely(err))
++ err = au_reval_dpath(dentry, sigen);
++ if (unlikely(err)) {
++ AuTraceErr(err);
+ goto out_dgrade;
-+ AuDebugOn(au_iigen(inode) != sgen);
++ }
+ }
+ di_downgrade_lock(dentry, AuLock_IR);
+
-+#if 0 /* todo: support it? */
-+ /* parent dir i_nlink is not updated in the case of setattr */
-+ if (S_ISDIR(inode->i_mode)) {
-+ mutex_lock(&inode->i_mutex);
-+ ii_write_lock(inode);
-+ au_cpup_attr_nlink(inode);
-+ ii_write_unlock(inode);
-+ mutex_unlock(&inode->i_mutex);
++ err = -EINVAL;
++ if (!(flags & (LOOKUP_OPEN | LOOKUP_EMPTY))
++ && inode
++ && !(inode->i_state && I_LINKABLE)
++ && (IS_DEADDIR(inode) || !inode->i_nlink)) {
++ AuTraceErr(err);
++ goto out_inval;
+ }
-+#endif
+
-+ AuDebugOn(au_digen(dentry) != sgen);
-+ AuDebugOn(inode && au_iigen(inode) != sgen);
-+ err = -EINVAL;
+ do_udba = !au_opt_test(au_mntflags(sb), UDBA_NONE);
+ if (do_udba && inode) {
-+ aufs_bindex_t bstart = au_ibstart(inode);
-+ if (bstart >= 0
-+ && au_test_higen(inode, au_h_iptr(inode, bstart)))
-+ goto out;
-+ }
-+ ndp = au_dup_nd(au_sbi(sb), &tmp_nd, nd);
-+#if 0
-+ if (nd) {
-+ path = nd->path;
-+ nd->path.dentry = au_h_dptr(nd->path.dentry, bindex);
-+ if (fake_nd->path.dentry) {
-+ fake_nd->path.mnt = au_sbr_mnt(sb, bindex);
-+ AuDebugOn(!fake_nd->path.mnt);
-+ path_get(&fake_nd->path);
-+ nd->path.
++ aufs_bindex_t btop = au_ibtop(inode);
++ struct inode *h_inode;
++
++ if (btop >= 0) {
++ h_inode = au_h_iptr(inode, btop);
++ if (h_inode && au_test_higen(inode, h_inode)) {
++ AuTraceErr(err);
++ goto out_inval;
+ }
++ }
+ }
-+#endif
-+ err = h_d_revalidate(dentry, inode, ndp, do_udba);
-+ if (unlikely(!err && do_udba && au_dbstart(dentry) < 0))
-+ /* both of real entry and whiteout found */
++
++ 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;
-+ goto out;
++ AuDbg("both of real entry and whiteout found, %p, err %d\n",
++ dentry, err);
++ }
++ goto out_inval;
+
-+ out_dgrade:
++out_dgrade:
+ di_downgrade_lock(dentry, AuLock_IR);
-+ out:
-+ au_store_fmode_exec(nd, inode);
++out_inval:
+ aufs_read_unlock(dentry, AuLock_IR);
+ AuTraceErr(err);
+ valid = !err;
-+ if (!valid)
-+ LKTRTrace("%.*s invalid\n", AuDLNPair(dentry));
++out:
++ if (!valid) {
++ AuDbg("%pd invalid, %d\n", dentry, valid);
++ d_drop(dentry);
++ }
+ return valid;
+}
+
+static void aufs_d_release(struct dentry *dentry)
+{
-+ struct au_dinfo *dinfo;
-+ aufs_bindex_t bend, bindex;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ AuDebugOn(!d_unhashed(dentry));
-+
-+ dinfo = dentry->d_fsdata;
-+ if (!dinfo)
-+ return;
-+
-+ /* dentry may not be revalidated */
-+ bindex = dinfo->di_bstart;
-+ if (bindex >= 0) {
-+ struct au_hdentry *p;
-+ bend = dinfo->di_bend;
-+ AuDebugOn(bend < bindex);
-+ p = dinfo->di_hdentry + bindex;
-+ while (bindex++ <= bend) {
-+ if (p->hd_dentry)
-+ au_hdput(p, /*do_free*/1);
-+ p++;
-+ }
-+ }
-+ kfree(dinfo->di_hdentry);
-+ au_rwsem_destroy(&dinfo->di_rwsem);
-+ au_cache_free_dinfo(dinfo);
-+ au_hin_di_reinit(dentry);
++ if (au_di(dentry)) {
++ au_di_fin(dentry);
++ au_hn_di_reinit(dentry);
++ }
+}
+
-+struct dentry_operations aufs_dop = {
-+ .d_revalidate = aufs_d_revalidate,
-+ .d_release = aufs_d_release,
-+ /* never use d_delete, especially in case of nfs server */
++const struct dentry_operations aufs_dop = {
++ .d_revalidate = aufs_d_revalidate,
++ .d_weak_revalidate = aufs_d_revalidate,
++ .d_release = aufs_d_release
++};
++
++/* aufs_dop without d_revalidate */
++const struct dentry_operations aufs_dop_noreval = {
++ .d_release = aufs_d_release
+};
-diff -urN linux/fs/aufs/dentry.h linux-aufs/fs/aufs/dentry.h
---- linux/fs/aufs/dentry.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dentry.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,396 @@
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,270 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * lookup and dentry operations
-+ *
-+ * $Id$
+ */
+
+#ifndef __AUFS_DENTRY_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/fs.h>
-+#include <linux/namei.h>
-+#include <linux/aufs_type.h>
-+#include "misc.h"
-+#include "super.h"
-+#include "vfsub.h"
-+
-+/* nameidata open_intent */
-+enum {
-+ AuIntent_AUFS,
-+ AuIntent_BRANCH,
-+ AuIntent_Last
-+};
-+
-+struct au_hdintent {
-+ struct list_head hdi_list;
-+ struct file *hdi_file[AuIntent_Last];
-+};
++#include <linux/dcache.h>
++#include "dirren.h"
++#include "rwsem.h"
+
+struct au_hdentry {
+ struct dentry *hd_dentry;
-+
-+#ifdef CONFIG_AUFS_BR_NFS
-+ spinlock_t hd_lock; /* intest_list */
-+ struct list_head *hd_intent_list;
-+#endif
++ aufs_bindex_t hd_id;
+};
+
+struct au_dinfo {
+ atomic_t di_generation;
+
+ struct au_rwsem di_rwsem;
-+ aufs_bindex_t di_bstart, di_bend, di_bwh, di_bdiropq;
++ aufs_bindex_t di_btop, di_bbot, di_bwh, di_bdiropq;
++ unsigned char di_tmpfile; /* to allow the different name */
+ struct au_hdentry *di_hdentry;
-+};
-+
-+/* nameidata extension flags */
-+#define AuNdx_DLGT 1
-+#define AuNdx_DIRPERM1 (1 << 1)
-+#define au_ftest_ndx(flags, name) ((flags) & AuNdx_##name)
-+#define au_fset_ndx(flags, name) { (flags) |= AuNdx_##name; }
-+#define au_fclr_ndx(flags, name) { (flags) &= ~AuNdx_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuNdx_DLGT
-+#define AuNdx_DLGT 0
-+#undef AuNdx_DIRPERM1
-+#define AuNdx_DIRPERM1 0
-+#endif
-+
-+struct au_ndx {
-+ struct vfsmount *nfsmnt;
-+ unsigned int flags;
-+ struct nameidata *nd;
-+ struct au_branch *br;
-+ struct file *nd_file;
-+};
++ struct file *di_htmpfile;
++ struct rcu_head rcu;
++} ____cacheline_aligned_in_smp;
+
+/* ---------------------------------------------------------------------- */
+
-+static inline void au_do_h_dentry_init(struct au_hdentry *hdentry)
-+{
-+ hdentry->hd_dentry = NULL;
-+}
-+
-+#ifdef CONFIG_AUFS_BR_NFS
-+static inline void au_h_dentry_init(struct au_hdentry *hdentry)
-+{
-+ au_do_h_dentry_init(hdentry);
-+ spin_lock_init(&hdentry->hd_lock);
-+}
-+
-+static inline void au_h_dentry_init_all(struct au_hdentry *hdentry, int n)
-+{
-+ while (n--)
-+ spin_lock_init(&hdentry[n].hd_lock);
-+}
-+
-+/* br_nfs.c */
-+struct file *au_h_intent(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *file);
-+int au_br_nfs_h_intent(struct file *nd_file, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct nameidata *nd);
-+void au_hintent_put(struct au_hdentry *hd, int do_free);
-+int au_fake_intent(struct nameidata *nd, struct au_branch *br);
-+int au_hin_after_reval(struct nameidata *nd, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct file *file);
-+struct dentry *au_lkup_hash(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx);
-+#else
-+
-+static inline void au_h_dentry_init(struct au_hdentry *hdentry)
-+{
-+ au_do_h_dentry_init(hdentry);
-+}
-+
-+static inline void au_h_dentry_init_all(struct au_hdentry *hdentry, int n)
-+{
-+ /* nothing */
-+}
-+
-+static inline
-+struct file *au_h_intent(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct file *file)
-+{
-+ /* return ERR_PTR(-ENOSYS); */
-+ return NULL;
-+}
-+
-+static inline
-+int au_br_nfs_h_intent(struct file *nd_file, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct nameidata *nd)
-+{
-+ return 0;
-+}
-+
-+static inline void au_hintent_put(struct au_hdentry *hd, int do_free)
-+{
-+ /* empty */
-+}
++/* 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)
+
-+static inline int au_fake_intent(struct nameidata *nd, struct au_branch *br)
-+{
-+ return 0;
-+}
++#ifndef CONFIG_AUFS_DIRREN
++#undef AuLkup_DIRREN
++#define AuLkup_DIRREN 0
++#endif
+
-+static inline
-+int au_hin_after_reval(struct nameidata *nd, struct dentry *dentry,
-+ aufs_bindex_t bindex, struct file *file)
-+{
-+ return 0;
-+}
++struct au_do_lookup_args {
++ unsigned int flags;
++ mode_t type;
++ struct qstr whname, *name;
++ struct au_dr_lookup dirren;
++};
+
-+#ifdef CONFIG_AUFS_DLGT
-+static inline
-+struct dentry *au_lkup_hash(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx)
-+{
-+ /* return ERR_PTR(-ENOSYS); */
-+ return vfsub_lookup_one_len(name, parent, len);
-+}
-+#endif
-+#endif /* CONFIG_AUFS_BR_NFS */
-+
-+#ifdef CONFIG_AUFS_DLGT
-+/* dlgt.c */
-+struct dentry *au_lkup_one_dlgt(const char *name, struct dentry *parent,
-+ int len, unsigned int flags);
-+#elif defined(CONFIG_AUFS_BR_NFS)
-+/* regardelss kernel version */
-+static inline
-+struct dentry *au_lkup_one_dlgt(const char *name, struct dentry *parent,
-+ int len, unsigned int flags)
-+{
-+ return vfsub_lookup_one_len(name, parent, len);
-+}
-+#endif
++/* ---------------------------------------------------------------------- */
+
+/* dentry.c */
-+extern struct dentry_operations aufs_dop;
-+#if defined(CONFIG_AUFS_BR_NFS) || defined(CONFIG_AUFS_DLGT)
-+struct dentry *au_lkup_one(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx);
-+#else
-+static inline
-+struct dentry *au_lkup_one(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx)
-+{
-+ /* todo? ndx->nd_file = NULL; */
-+ return vfsub_lookup_one_len(name, parent, len);
-+}
-+#endif
-+struct dentry *au_sio_lkup_one(const char *name, struct dentry *parent, int len,
-+ struct au_ndx *ndx);
-+int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t bstart, mode_t type,
-+ struct nameidata *nd);
-+int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex);
-+int au_refresh_hdentry(struct dentry *dentry, mode_t type);
-+int au_reval_dpath(struct dentry *dentry, au_gen_t sgen);
++extern const struct dentry_operations aufs_dop, aufs_dop_noreval;
++struct au_branch;
++struct dentry *au_sio_lkup_one(struct mnt_idmap *idmap, struct qstr *name,
++ struct path *ppath);
++int au_h_verify(struct dentry *h_dentry, unsigned int udba, struct inode *h_dir,
++ struct dentry *h_parent, struct au_branch *br);
++
++int au_lkup_dentry(struct dentry *dentry, aufs_bindex_t btop,
++ unsigned int flags);
++int au_lkup_neg(struct dentry *dentry, aufs_bindex_t bindex, int wh);
++int au_refresh_dentry(struct dentry *dentry, struct dentry *parent);
++int au_reval_dpath(struct dentry *dentry, unsigned int sigen);
++void au_refresh_dop(struct dentry *dentry, int force_reval);
+
+/* dinfo.c */
-+int au_alloc_dinfo(struct dentry *dentry);
-+struct au_dinfo *au_di(struct dentry *dentry);
++void au_di_init_once(void *_di);
++struct au_dinfo *au_di_alloc(struct super_block *sb, unsigned int lsc);
++void au_di_free(struct au_dinfo *dinfo);
++void au_di_swap(struct au_dinfo *a, struct au_dinfo *b);
++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 may_shrink);
+
+void di_read_lock(struct dentry *d, int flags, unsigned int lsc);
+void di_read_unlock(struct dentry *d, int flags);
+void di_write_unlock2(struct dentry *d1, struct dentry *d2);
+
+struct dentry *au_h_dptr(struct dentry *dentry, aufs_bindex_t bindex);
-+
++struct dentry *au_h_d_alias(struct dentry *dentry, aufs_bindex_t bindex);
+aufs_bindex_t au_dbtail(struct dentry *dentry);
+aufs_bindex_t au_dbtaildir(struct dentry *dentry);
-+#if 0 /* reserved for future use */
-+aufs_bindex_t au_dbtail_generic(struct dentry *dentry);
-+#endif
+
-+void au_set_dbdiropq(struct dentry *dentry, aufs_bindex_t bindex);
+void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex,
+ struct dentry *h_dentry);
-+
++int au_digen_test(struct dentry *dentry, unsigned int sigen);
++int au_dbrange_test(struct dentry *dentry);
++void au_update_digen(struct dentry *dentry);
+void au_update_dbrange(struct dentry *dentry, int do_put_zero);
-+void au_update_dbstart(struct dentry *dentry);
-+void au_update_dbend(struct dentry *dentry);
++void au_update_dbtop(struct dentry *dentry);
++void au_update_dbbot(struct dentry *dentry);
+int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry);
+
+/* ---------------------------------------------------------------------- */
+
-+/* todo: memory barrier? */
-+static inline au_gen_t au_digen(struct dentry *d)
-+{
-+ return atomic_read(&au_di(d)->di_generation);
-+}
-+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+static inline au_gen_t au_digen_dec(struct dentry *d)
-+{
-+ return atomic_dec_return(&au_di(d)->di_generation);
-+}
-+
-+static inline void au_hin_di_reinit(struct dentry *d)
++static inline struct au_dinfo *au_di(struct dentry *dentry)
+{
-+ d->d_fsdata = NULL;
-+}
-+#else
-+static inline void au_hin_di_reinit(struct dentry *d)
-+{
-+ /* empty */
++ return dentry->d_fsdata;
+}
-+#endif /* CONFIG_AUFS_HINOTIFY */
+
+/* ---------------------------------------------------------------------- */
+
+/* lock subclass for dinfo */
+enum {
+ AuLsc_DI_CHILD, /* child first */
-+ AuLsc_DI_CHILD2, /* rename(2), link(2), and cpup at hinotify */
++ AuLsc_DI_CHILD2, /* rename(2), link(2), and cpup at hnotify */
+ AuLsc_DI_CHILD3, /* copyup dirs */
+ AuLsc_DI_PARENT,
+ AuLsc_DI_PARENT2,
+ AuLsc_DI_PARENT3,
-+ AuLsc_DI_PARENT4
++ AuLsc_DI_TMP /* temp for replacing dinfo */
+};
+
+/*
+ * di_read_lock_parent, di_write_lock_parent,
+ * di_read_lock_parent2, di_write_lock_parent2,
+ * di_read_lock_parent3, di_write_lock_parent3,
-+ * di_read_lock_parent4, di_write_lock_parent4,
+ */
+#define AuReadLockFunc(name, lsc) \
+static inline void di_read_lock_##name(struct dentry *d, int flags) \
+AuRWLockFuncs(parent, PARENT);
+AuRWLockFuncs(parent2, PARENT2);
+AuRWLockFuncs(parent3, PARENT3);
-+AuRWLockFuncs(parent4, PARENT4);
+
+#undef AuReadLockFunc
+#undef AuWriteLockFunc
+#undef AuRWLockFuncs
+
-+/* to debug easier, do not make them inlined functions */
-+#define DiMustReadLock(d) do { \
-+ SiMustAnyLock((d)->d_sb); \
-+ AuRwMustReadLock(&au_di(d)->di_rwsem); \
-+} while (0)
++#define DiMustNoWaiters(d) AuRwMustNoWaiters(&au_di(d)->di_rwsem)
++#define DiMustAnyLock(d) AuRwMustAnyLock(&au_di(d)->di_rwsem)
++#define DiMustWriteLock(d) AuRwMustWriteLock(&au_di(d)->di_rwsem)
+
-+#define DiMustWriteLock(d) do { \
-+ SiMustAnyLock((d)->d_sb); \
-+ AuRwMustWriteLock(&au_di(d)->di_rwsem); \
-+} while (0)
++/* ---------------------------------------------------------------------- */
+
-+#define DiMustAnyLock(d) do { \
-+ SiMustAnyLock((d)->d_sb); \
-+ AuRwMustAnyLock(&au_di(d)->di_rwsem); \
-+} while (0)
++/* todo: memory barrier? */
++static inline unsigned int au_digen(struct dentry *d)
++{
++ return atomic_read(&au_di(d)->di_generation);
++}
+
-+#define DiMustNoWaiters(d) AuRwMustNoWaiters(&au_di(d)->di_rwsem)
++static inline void au_h_dentry_init(struct au_hdentry *hdentry)
++{
++ hdentry->hd_dentry = NULL;
++}
+
-+/* ---------------------------------------------------------------------- */
++static inline struct au_hdentry *au_hdentry(struct au_dinfo *di,
++ aufs_bindex_t bindex)
++{
++ return di->di_hdentry + bindex;
++}
++
++static inline void au_hdput(struct au_hdentry *hd)
++{
++ if (hd)
++ dput(hd->hd_dentry);
++}
+
-+static inline aufs_bindex_t au_dbstart(struct dentry *dentry)
++static inline aufs_bindex_t au_dbtop(struct dentry *dentry)
+{
+ DiMustAnyLock(dentry);
-+ return au_di(dentry)->di_bstart;
++ return au_di(dentry)->di_btop;
+}
+
-+static inline aufs_bindex_t au_dbend(struct dentry *dentry)
++static inline aufs_bindex_t au_dbbot(struct dentry *dentry)
+{
+ DiMustAnyLock(dentry);
-+ return au_di(dentry)->di_bend;
++ return au_di(dentry)->di_bbot;
+}
+
+static inline aufs_bindex_t au_dbwh(struct dentry *dentry)
+static inline aufs_bindex_t au_dbdiropq(struct dentry *dentry)
+{
+ DiMustAnyLock(dentry);
-+ AuDebugOn(dentry->d_inode
-+ && dentry->d_inode->i_mode
-+ && !S_ISDIR(dentry->d_inode->i_mode));
+ return au_di(dentry)->di_bdiropq;
+}
+
+/* todo: hard/soft set? */
-+static inline void au_set_dbstart(struct dentry *dentry, aufs_bindex_t bindex)
++static inline void au_set_dbtop(struct dentry *dentry, aufs_bindex_t bindex)
+{
+ DiMustWriteLock(dentry);
-+ AuDebugOn(au_sbend(dentry->d_sb) < bindex);
-+ /* */
-+ au_di(dentry)->di_bstart = bindex;
++ au_di(dentry)->di_btop = bindex;
+}
+
-+static inline void au_set_dbend(struct dentry *dentry, aufs_bindex_t bindex)
++static inline void au_set_dbbot(struct dentry *dentry, aufs_bindex_t bindex)
+{
+ DiMustWriteLock(dentry);
-+ AuDebugOn(au_sbend(dentry->d_sb) < bindex
-+ || bindex < au_dbstart(dentry));
-+ au_di(dentry)->di_bend = bindex;
++ au_di(dentry)->di_bbot = bindex;
+}
+
+static inline void au_set_dbwh(struct dentry *dentry, aufs_bindex_t bindex)
+{
+ DiMustWriteLock(dentry);
-+ AuDebugOn(au_sbend(dentry->d_sb) < bindex);
-+ /* dbwh can be outside of bstart - bend range */
++ /* dbwh can be outside of btop - bbot range */
+ au_di(dentry)->di_bwh = bindex;
+}
+
-+static inline void au_hdput(struct au_hdentry *hd, int do_free)
++static inline void au_set_dbdiropq(struct dentry *dentry, aufs_bindex_t bindex)
++{
++ DiMustWriteLock(dentry);
++ au_di(dentry)->di_bdiropq = bindex;
++}
++
++/* ---------------------------------------------------------------------- */
++
++#ifdef CONFIG_AUFS_HNOTIFY
++static inline void au_digen_dec(struct dentry *d)
+{
-+ au_hintent_put(hd, do_free);
-+ dput(hd->hd_dentry);
++ atomic_dec(&au_di(d)->di_generation);
+}
+
-+static inline void au_update_digen(struct dentry *dentry)
++static inline void au_hn_di_reinit(struct dentry *dentry)
+{
-+ AuDebugOn(!dentry->d_sb);
-+ atomic_set(&au_di(dentry)->di_generation, au_sigen(dentry->d_sb));
-+ /* smp_mb(); */ /* atomic_set */
++ dentry->d_fsdata = NULL;
+}
++#else
++AuStubVoid(au_hn_di_reinit, struct dentry *dentry __maybe_unused)
++#endif /* CONFIG_AUFS_HNOTIFY */
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DENTRY_H__ */
-diff -urN linux/fs/aufs/dinfo.c linux-aufs/fs/aufs/dinfo.c
---- linux/fs/aufs/dinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dinfo.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,436 @@
+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 2022-12-17 09:21:34.796521861 +0100
+@@ -0,0 +1,555 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * dentry private data
-+ *
-+ * $Id$
+ */
+
+#include "aufs.h"
+
-+int au_alloc_dinfo(struct dentry *dentry)
++void au_di_init_once(void *_dinfo)
+{
-+ struct au_dinfo *dinfo;
-+ struct super_block *sb;
-+ int nbr;
++ struct au_dinfo *dinfo = _dinfo;
++
++ au_rw_init(&dinfo->di_rwsem);
++}
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ AuDebugOn(dentry->d_fsdata);
++struct au_dinfo *au_di_alloc(struct super_block *sb, unsigned int lsc)
++{
++ struct au_dinfo *dinfo;
++ int nbr, i;
+
+ dinfo = au_cache_alloc_dinfo();
-+ if (dinfo) {
-+ sb = dentry->d_sb;
-+ nbr = au_sbend(sb) + 1;
-+ if (nbr <= 0)
-+ nbr = 1;
-+ dinfo->di_hdentry = kcalloc(nbr, sizeof(*dinfo->di_hdentry),
-+ GFP_NOFS);
-+ if (dinfo->di_hdentry) {
-+ au_h_dentry_init_all(dinfo->di_hdentry, nbr);
-+ atomic_set(&dinfo->di_generation, au_sigen(sb));
-+ /* smp_mb(); */ /* atomic_set */
-+ au_rw_init_wlock_nested(&dinfo->di_rwsem,
-+ AuLsc_DI_CHILD);
-+ au_dbg_locked_di_reg(dentry, AuLock_DW, AuLsc_DI_CHILD);
-+ dinfo->di_bstart = -1;
-+ dinfo->di_bend = -1;
-+ dinfo->di_bwh = -1;
-+ dinfo->di_bdiropq = -1;
-+
-+ dentry->d_fsdata = dinfo;
-+ dentry->d_op = &aufs_dop;
-+ return 0; /* success */
-+ }
-+ au_cache_free_dinfo(dinfo);
++ if (unlikely(!dinfo))
++ goto out;
++
++ nbr = au_sbbot(sb) + 1;
++ if (nbr <= 0)
++ nbr = 1;
++ dinfo->di_hdentry = kcalloc(nbr, sizeof(*dinfo->di_hdentry), GFP_NOFS);
++ if (dinfo->di_hdentry) {
++ au_rw_write_lock_nested(&dinfo->di_rwsem, lsc);
++ dinfo->di_btop = -1;
++ dinfo->di_bbot = -1;
++ dinfo->di_bwh = -1;
++ dinfo->di_bdiropq = -1;
++ dinfo->di_tmpfile = 0;
++ for (i = 0; i < nbr; i++)
++ dinfo->di_hdentry[i].hd_id = -1;
++ dinfo->di_htmpfile = NULL;
++ goto out;
+ }
-+ AuTraceErr(-ENOMEM);
-+ return -ENOMEM;
++
++ au_cache_free_dinfo(dinfo);
++ dinfo = NULL;
++
++out:
++ return dinfo;
+}
+
-+struct au_dinfo *au_di(struct dentry *dentry)
++void au_di_free(struct au_dinfo *dinfo)
+{
-+ struct au_dinfo *dinfo = dentry->d_fsdata;
-+ AuDebugOn(!dinfo
-+ || !dinfo->di_hdentry
-+ /* || au_sbi(dentry->d_sb)->si_bend < dinfo->di_bend */
-+ || dinfo->di_bend < dinfo->di_bstart
-+ /* dbwh can be outside of this range */
-+ || (0 <= dinfo->di_bdiropq
-+ && (dinfo->di_bdiropq < dinfo->di_bstart
-+ /* || dinfo->di_bend < dinfo->di_bdiropq */))
-+ );
-+ return dinfo;
++ struct au_hdentry *p;
++ aufs_bindex_t bbot, bindex;
++
++ /* dentry may not be revalidated */
++ bindex = dinfo->di_btop;
++ if (bindex >= 0) {
++ bbot = dinfo->di_bbot;
++ p = au_hdentry(dinfo, bindex);
++ while (bindex++ <= bbot)
++ au_hdput(p++);
++ }
++ au_kfree_try_rcu(dinfo->di_hdentry);
++ au_cache_free_dinfo(dinfo);
++}
++
++void au_di_swap(struct au_dinfo *a, struct au_dinfo *b)
++{
++ struct au_hdentry *p;
++ aufs_bindex_t bi;
++
++ AuRwMustWriteLock(&a->di_rwsem);
++ AuRwMustWriteLock(&b->di_rwsem);
++
++#define DiSwap(v, name) \
++ do { \
++ v = a->di_##name; \
++ a->di_##name = b->di_##name; \
++ b->di_##name = v; \
++ } while (0)
++
++ DiSwap(p, hdentry);
++ DiSwap(bi, btop);
++ DiSwap(bi, bbot);
++ DiSwap(bi, bwh);
++ DiSwap(bi, bdiropq);
++ /* smp_mb(); */
++
++#undef DiSwap
++}
++
++void au_di_cp(struct au_dinfo *dst, struct au_dinfo *src)
++{
++ AuRwMustWriteLock(&dst->di_rwsem);
++ AuRwMustWriteLock(&src->di_rwsem);
++
++ dst->di_btop = src->di_btop;
++ dst->di_bbot = src->di_bbot;
++ dst->di_bwh = src->di_bwh;
++ dst->di_bdiropq = src->di_bdiropq;
++ /* smp_mb(); */
++}
++
++int au_di_init(struct dentry *dentry)
++{
++ int err;
++ struct super_block *sb;
++ struct au_dinfo *dinfo;
++
++ err = 0;
++ sb = dentry->d_sb;
++ dinfo = au_di_alloc(sb, AuLsc_DI_CHILD);
++ if (dinfo) {
++ atomic_set(&dinfo->di_generation, au_sigen(sb));
++ /* smp_mb(); */ /* atomic_set */
++ dentry->d_fsdata = dinfo;
++ } else
++ err = -ENOMEM;
++
++ return err;
++}
++
++void au_di_fin(struct dentry *dentry)
++{
++ struct au_dinfo *dinfo;
++
++ dinfo = au_di(dentry);
++ AuRwDestroy(&dinfo->di_rwsem);
++ au_di_free(dinfo);
++}
++
++int au_di_realloc(struct au_dinfo *dinfo, int nbr, int may_shrink)
++{
++ int err, sz;
++ struct au_hdentry *hdp;
++
++ AuRwMustWriteLock(&dinfo->di_rwsem);
++
++ err = -ENOMEM;
++ sz = sizeof(*hdp) * (dinfo->di_bbot + 1);
++ if (!sz)
++ sz = sizeof(*hdp);
++ hdp = au_kzrealloc(dinfo->di_hdentry, sz, sizeof(*hdp) * nbr, GFP_NOFS,
++ may_shrink);
++ if (hdp) {
++ dinfo->di_hdentry = hdp;
++ err = 0;
++ }
++
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+ case AuLsc_DI_PARENT3:
+ ii_write_lock_parent3(inode);
+ break;
-+ case AuLsc_DI_PARENT4:
-+ ii_write_lock_parent4(inode);
-+ break;
+ default:
+ BUG();
+ }
+ case AuLsc_DI_PARENT3:
+ ii_read_lock_parent3(inode);
+ break;
-+ case AuLsc_DI_PARENT4:
-+ ii_read_lock_parent4(inode);
-+ break;
+ default:
+ BUG();
+ }
+
+void di_read_lock(struct dentry *d, int flags, unsigned int lsc)
+{
-+ LKTRTrace("%.*s, %u\n", AuDLNPair(d), lsc);
-+ SiMustAnyLock(d->d_sb);
++ struct inode *inode;
+
-+ /* todo: always nested? */
-+ au_dbg_locking_di_reg(d, flags, lsc);
+ au_rw_read_lock_nested(&au_di(d)->di_rwsem, lsc);
-+ au_dbg_locking_di_unreg(d, flags);
-+ au_dbg_locked_di_reg(d, flags, lsc);
-+ if (d->d_inode) {
++ if (d_really_is_positive(d)) {
++ inode = d_inode(d);
+ if (au_ftest_lock(flags, IW))
-+ do_ii_write_lock(d->d_inode, lsc);
++ do_ii_write_lock(inode, lsc);
+ else if (au_ftest_lock(flags, IR))
-+ do_ii_read_lock(d->d_inode, lsc);
++ do_ii_read_lock(inode, lsc);
+ }
+}
+
+void di_read_unlock(struct dentry *d, int flags)
+{
-+ LKTRTrace("%.*s\n", AuDLNPair(d));
-+ SiMustAnyLock(d->d_sb);
++ struct inode *inode;
+
-+ if (d->d_inode) {
-+ if (au_ftest_lock(flags, IW))
-+ ii_write_unlock(d->d_inode);
-+ else if (au_ftest_lock(flags, IR))
-+ ii_read_unlock(d->d_inode);
++ if (d_really_is_positive(d)) {
++ inode = d_inode(d);
++ if (au_ftest_lock(flags, IW)) {
++ au_dbg_verify_dinode(d);
++ ii_write_unlock(inode);
++ } else if (au_ftest_lock(flags, IR)) {
++ au_dbg_verify_dinode(d);
++ ii_read_unlock(inode);
++ }
+ }
+ au_rw_read_unlock(&au_di(d)->di_rwsem);
-+ au_dbg_locked_di_unreg(d, flags);
+}
+
+void di_downgrade_lock(struct dentry *d, int flags)
+{
-+ SiMustAnyLock(d->d_sb);
-+
++ if (d_really_is_positive(d) && au_ftest_lock(flags, IR))
++ ii_downgrade_lock(d_inode(d));
+ au_rw_dgrade_lock(&au_di(d)->di_rwsem);
-+ if (d->d_inode && au_ftest_lock(flags, IR))
-+ ii_downgrade_lock(d->d_inode);
+}
+
+void di_write_lock(struct dentry *d, unsigned int lsc)
+{
-+ LKTRTrace("%.*s, %u\n", AuDLNPair(d), lsc);
-+ SiMustAnyLock(d->d_sb);
-+
-+ /* todo: always nested? */
-+ au_dbg_locking_di_reg(d, AuLock_IW, lsc);
+ au_rw_write_lock_nested(&au_di(d)->di_rwsem, lsc);
-+ au_dbg_locking_di_unreg(d, AuLock_IW);
-+ au_dbg_locked_di_reg(d, AuLock_IW, lsc);
-+ if (d->d_inode)
-+ do_ii_write_lock(d->d_inode, lsc);
++ if (d_really_is_positive(d))
++ do_ii_write_lock(d_inode(d), lsc);
+}
+
+void di_write_unlock(struct dentry *d)
+{
-+ LKTRTrace("%.*s\n", AuDLNPair(d));
-+ SiMustAnyLock(d->d_sb);
-+
-+ if (d->d_inode)
-+ ii_write_unlock(d->d_inode);
++ au_dbg_verify_dinode(d);
++ if (d_really_is_positive(d))
++ ii_write_unlock(d_inode(d));
+ au_rw_write_unlock(&au_di(d)->di_rwsem);
-+ au_dbg_locked_di_unreg(d, AuLock_IW);
+}
+
+void di_write_lock2_child(struct dentry *d1, struct dentry *d2, int isdir)
+{
-+ AuTraceEnter();
+ AuDebugOn(d1 == d2
-+ || d1->d_inode == d2->d_inode
++ || 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);
+ }
+
+void di_write_lock2_parent(struct dentry *d1, struct dentry *d2, int isdir)
+{
-+ AuTraceEnter();
+ AuDebugOn(d1 == d2
-+ || d1->d_inode == d2->d_inode
++ || 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);
+ }
+void di_write_unlock2(struct dentry *d1, struct dentry *d2)
+{
+ di_write_unlock(d1);
-+ if (d1->d_inode == d2->d_inode) {
++ if (d_inode(d1) == d_inode(d2))
+ au_rw_write_unlock(&au_di(d2)->di_rwsem);
-+ au_dbg_locked_di_unreg(d2, AuLock_IW);
-+ } else
++ else
+ di_write_unlock(d2);
+}
+
+ struct dentry *d;
+
+ DiMustAnyLock(dentry);
-+ if (au_dbstart(dentry) < 0 || bindex < au_dbstart(dentry))
++
++ if (au_dbtop(dentry) < 0 || bindex < au_dbtop(dentry))
+ return NULL;
-+ AuDebugOn(bindex < 0
-+ /* || bindex > au_sbend(dentry->d_sb) */);
-+ d = au_di(dentry)->di_hdentry[0 + bindex].hd_dentry;
-+ AuDebugOn(d && (atomic_read(&d->d_count) <= 0));
++ AuDebugOn(bindex < 0);
++ d = au_hdentry(au_di(dentry), bindex)->hd_dentry;
++ AuDebugOn(d && au_dcount(d) <= 0);
+ return d;
+}
+
-+aufs_bindex_t au_dbtail(struct dentry *dentry)
++/*
++ * extended version of au_h_dptr().
++ * returns a hashed and positive (or linkable) h_dentry in bindex, NULL, or
++ * error.
++ */
++struct dentry *au_h_d_alias(struct dentry *dentry, aufs_bindex_t bindex)
+{
-+ aufs_bindex_t bend, bwh;
-+
-+ bend = au_dbend(dentry);
-+ if (0 <= bend) {
-+ bwh = au_dbwh(dentry);
-+ if (!bwh)
-+ return bwh;
-+ if (0 < bwh && bwh < bend)
-+ return bwh - 1;
-+ }
-+ return bend;
-+}
++ struct dentry *h_dentry;
++ struct inode *inode, *h_inode;
+
-+aufs_bindex_t au_dbtaildir(struct dentry *dentry)
-+{
-+ aufs_bindex_t bend, bopq;
++ AuDebugOn(d_really_is_negative(dentry));
+
-+ AuDebugOn(dentry->d_inode
-+ && dentry->d_inode->i_mode
-+ && !S_ISDIR(dentry->d_inode->i_mode));
++ h_dentry = NULL;
++ if (au_dbtop(dentry) <= bindex
++ && bindex <= au_dbbot(dentry))
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry && !au_d_linkable(h_dentry)) {
++ dget(h_dentry);
++ goto out; /* success */
++ }
+
-+ bend = au_dbtail(dentry);
-+ if (0 <= bend) {
-+ bopq = au_dbdiropq(dentry);
-+ AuDebugOn(bend < bopq);
-+ if (0 <= bopq && bopq < bend)
-+ bend = bopq;
++ inode = d_inode(dentry);
++ AuDebugOn(bindex < au_ibtop(inode));
++ AuDebugOn(au_ibbot(inode) < bindex);
++ h_inode = au_h_iptr(inode, bindex);
++ h_dentry = d_find_alias(h_inode);
++ if (h_dentry) {
++ if (!IS_ERR(h_dentry)) {
++ if (!au_d_linkable(h_dentry))
++ goto out; /* success */
++ dput(h_dentry);
++ } else
++ goto out;
+ }
-+ return bend;
-+}
+
-+#if 0 /* reserved for future use */
-+aufs_bindex_t au_dbtail_generic(struct dentry *dentry)
-+{
-+ struct inode *inode;
++ if (au_opt_test(au_mntflags(dentry->d_sb), PLINK)) {
++ h_dentry = au_plink_lkup(inode, bindex);
++ AuDebugOn(!h_dentry);
++ if (!IS_ERR(h_dentry)) {
++ if (!au_d_hashed_positive(h_dentry))
++ goto out; /* success */
++ dput(h_dentry);
++ h_dentry = NULL;
++ }
++ }
+
-+ inode = dentry->d_inode;
-+ if (inode && S_ISDIR(inode->i_mode))
-+ return au_dbtaildir(dentry);
-+ else
-+ return au_dbtail(dentry);
++out:
++ AuDbgDentry(h_dentry);
++ return h_dentry;
+}
-+#endif
+
-+/* ---------------------------------------------------------------------- */
++aufs_bindex_t au_dbtail(struct dentry *dentry)
++{
++ aufs_bindex_t bbot, bwh;
++
++ bbot = au_dbbot(dentry);
++ if (0 <= bbot) {
++ bwh = au_dbwh(dentry);
++ if (!bwh)
++ return bwh;
++ if (0 < bwh && bwh < bbot)
++ return bwh - 1;
++ }
++ return bbot;
++}
+
-+void au_set_dbdiropq(struct dentry *dentry, aufs_bindex_t bindex)
++aufs_bindex_t au_dbtaildir(struct dentry *dentry)
+{
-+ DiMustWriteLock(dentry);
-+ AuDebugOn(au_sbend(dentry->d_sb) < bindex);
-+ AuDebugOn((bindex >= 0
-+ && (bindex < au_dbstart(dentry)
-+ || au_dbend(dentry) < bindex))
-+ || (dentry->d_inode
-+ && dentry->d_inode->i_mode
-+ && !S_ISDIR(dentry->d_inode->i_mode)));
-+ au_di(dentry)->di_bdiropq = bindex;
++ aufs_bindex_t bbot, bopq;
++
++ bbot = au_dbtail(dentry);
++ if (0 <= bbot) {
++ bopq = au_dbdiropq(dentry);
++ if (0 <= bopq && bopq < bbot)
++ bbot = bopq;
++ }
++ return bbot;
+}
+
++/* ---------------------------------------------------------------------- */
++
+void au_set_h_dptr(struct dentry *dentry, aufs_bindex_t bindex,
+ struct dentry *h_dentry)
+{
-+ struct au_hdentry *hd = au_di(dentry)->di_hdentry + bindex;
++ struct au_dinfo *dinfo;
++ struct au_hdentry *hd;
++ struct au_branch *br;
++
+ DiMustWriteLock(dentry);
-+ AuDebugOn(bindex < au_di(dentry)->di_bstart
-+ || bindex > au_di(dentry)->di_bend
-+ || (h_dentry && atomic_read(&h_dentry->d_count) <= 0)
-+ || (h_dentry && hd->hd_dentry)
-+ );
-+ if (hd->hd_dentry)
-+ au_hdput(hd, /*do_free*/0);
++
++ dinfo = au_di(dentry);
++ hd = au_hdentry(dinfo, bindex);
++ au_hdput(hd);
+ hd->hd_dentry = h_dentry;
++ if (h_dentry) {
++ br = au_sbr(dentry->d_sb, bindex);
++ hd->hd_id = br->br_id;
++ }
+}
+
-+/* ---------------------------------------------------------------------- */
++int au_dbrange_test(struct dentry *dentry)
++{
++ int err;
++ aufs_bindex_t btop, bbot;
++
++ err = 0;
++ btop = au_dbtop(dentry);
++ bbot = au_dbbot(dentry);
++ if (btop >= 0)
++ AuDebugOn(bbot < 0 && btop > bbot);
++ else {
++ err = -EIO;
++ AuDebugOn(bbot >= 0);
++ }
++
++ return err;
++}
++
++int au_digen_test(struct dentry *dentry, unsigned int sigen)
++{
++ int err;
++
++ err = 0;
++ if (unlikely(au_digen(dentry) != sigen
++ || au_iigen_test(d_inode(dentry), sigen)))
++ err = -EIO;
++
++ return err;
++}
++
++void au_update_digen(struct dentry *dentry)
++{
++ atomic_set(&au_di(dentry)->di_generation, au_sigen(dentry->d_sb));
++ /* smp_mb(); */ /* atomic_set */
++}
+
+void au_update_dbrange(struct dentry *dentry, int do_put_zero)
+{
+ struct au_dinfo *dinfo;
-+ aufs_bindex_t bindex;
+ struct dentry *h_d;
++ struct au_hdentry *hdp;
++ aufs_bindex_t bindex, bbot;
+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), do_put_zero);
+ DiMustWriteLock(dentry);
+
+ dinfo = au_di(dentry);
-+ if (!dinfo || dinfo->di_bstart < 0)
++ if (!dinfo || dinfo->di_btop < 0)
+ return;
+
+ if (do_put_zero) {
-+ for (bindex = dinfo->di_bstart; bindex <= dinfo->di_bend;
-+ bindex++) {
-+ h_d = dinfo->di_hdentry[0 + bindex].hd_dentry;
-+ if (h_d && !h_d->d_inode)
++ bbot = dinfo->di_bbot;
++ bindex = dinfo->di_btop;
++ hdp = au_hdentry(dinfo, bindex);
++ for (; bindex <= bbot; bindex++, hdp++) {
++ h_d = hdp->hd_dentry;
++ if (h_d && d_is_negative(h_d))
+ au_set_h_dptr(dentry, bindex, NULL);
+ }
+ }
+
-+ dinfo->di_bstart = -1;
-+ while (++dinfo->di_bstart <= dinfo->di_bend)
-+ if (dinfo->di_hdentry[0 + dinfo->di_bstart].hd_dentry)
++ dinfo->di_btop = 0;
++ hdp = au_hdentry(dinfo, dinfo->di_btop);
++ for (; dinfo->di_btop <= dinfo->di_bbot; dinfo->di_btop++, hdp++)
++ if (hdp->hd_dentry)
+ break;
-+ if (dinfo->di_bstart > dinfo->di_bend) {
-+ dinfo->di_bstart = -1;
-+ dinfo->di_bend = -1;
++ if (dinfo->di_btop > dinfo->di_bbot) {
++ dinfo->di_btop = -1;
++ dinfo->di_bbot = -1;
+ return;
+ }
+
-+ dinfo->di_bend++;
-+ while (0 <= --dinfo->di_bend)
-+ if (dinfo->di_hdentry[0 + dinfo->di_bend].hd_dentry)
++ hdp = au_hdentry(dinfo, dinfo->di_bbot);
++ for (; dinfo->di_bbot >= 0; dinfo->di_bbot--, hdp--)
++ if (hdp->hd_dentry)
+ break;
-+ AuDebugOn(dinfo->di_bstart > dinfo->di_bend || dinfo->di_bend < 0);
++ AuDebugOn(dinfo->di_btop > dinfo->di_bbot || dinfo->di_bbot < 0);
+}
+
-+void au_update_dbstart(struct dentry *dentry)
++void au_update_dbtop(struct dentry *dentry)
+{
-+ aufs_bindex_t bindex,
-+ bstart = au_dbstart(dentry),
-+ bend = au_dbend(dentry);
++ aufs_bindex_t bindex, bbot;
+ struct dentry *h_dentry;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ DiMustWriteLock(dentry);
-+
-+ for (bindex = bstart; bindex <= bend; bindex++) {
++ bbot = au_dbbot(dentry);
++ for (bindex = au_dbtop(dentry); bindex <= bbot; bindex++) {
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (!h_dentry)
+ continue;
-+ if (h_dentry->d_inode) {
-+ au_set_dbstart(dentry, bindex);
++ if (d_is_positive(h_dentry)) {
++ au_set_dbtop(dentry, bindex);
+ return;
+ }
+ au_set_h_dptr(dentry, bindex, NULL);
+ }
+}
+
-+void au_update_dbend(struct dentry *dentry)
++void au_update_dbbot(struct dentry *dentry)
+{
-+ aufs_bindex_t bindex,
-+ bstart = au_dbstart(dentry),
-+ bend = au_dbend(dentry);
++ aufs_bindex_t bindex, btop;
+ struct dentry *h_dentry;
+
-+ DiMustWriteLock(dentry);
-+ for (bindex = bend; bindex <= bstart; bindex--) {
++ btop = au_dbtop(dentry);
++ for (bindex = au_dbbot(dentry); bindex >= btop; bindex--) {
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (!h_dentry)
+ continue;
-+ if (h_dentry->d_inode) {
-+ au_set_dbend(dentry, bindex);
++ if (d_is_positive(h_dentry)) {
++ au_set_dbbot(dentry, bindex);
+ return;
+ }
+ au_set_h_dptr(dentry, bindex, NULL);
+
+int au_find_dbindex(struct dentry *dentry, struct dentry *h_dentry)
+{
-+ aufs_bindex_t bindex, bend;
++ aufs_bindex_t bindex, bbot;
+
-+ bend = au_dbend(dentry);
-+ for (bindex = au_dbstart(dentry); bindex <= bend; bindex++)
++ bbot = au_dbbot(dentry);
++ for (bindex = au_dbtop(dentry); bindex <= bbot; bindex++)
+ if (au_h_dptr(dentry, bindex) == h_dentry)
+ return bindex;
+ return -1;
+}
-diff -urN linux/fs/aufs/dir.c linux-aufs/fs/aufs/dir.c
---- linux/fs/aufs/dir.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dir.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,559 @@
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,765 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * directory operations
-+ *
-+ * $Id$
+ */
+
+#include <linux/fs_stack.h>
++#include <linux/iversion.h>
+#include "aufs.h"
+
++void au_add_nlink(struct inode *dir, struct inode *h_dir)
++{
++ unsigned int nlink;
++
++ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode));
++
++ nlink = dir->i_nlink;
++ nlink += h_dir->i_nlink - 2;
++ if (h_dir->i_nlink < 2)
++ nlink += 2;
++ smp_mb(); /* for i_nlink */
++ /* 0 can happen in revaliding */
++ set_nlink(dir, nlink);
++}
++
++void au_sub_nlink(struct inode *dir, struct inode *h_dir)
++{
++ unsigned int nlink;
++
++ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode));
++
++ nlink = dir->i_nlink;
++ nlink -= h_dir->i_nlink - 2;
++ if (h_dir->i_nlink < 2)
++ nlink -= 2;
++ smp_mb(); /* for i_nlink */
++ /* nlink == 0 means the branch-fs is broken */
++ set_nlink(dir, nlink);
++}
++
++loff_t au_dir_size(struct file *file, struct dentry *dentry)
++{
++ loff_t sz;
++ aufs_bindex_t bindex, bbot;
++ struct file *h_file;
++ struct dentry *h_dentry;
++
++ sz = 0;
++ if (file) {
++ AuDebugOn(!d_is_dir(file->f_path.dentry));
++
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file);
++ bindex <= bbot && sz < KMALLOC_MAX_SIZE;
++ bindex++) {
++ h_file = au_hf_dir(file, bindex);
++ if (h_file && file_inode(h_file))
++ sz += vfsub_f_size_read(h_file);
++ }
++ } else {
++ AuDebugOn(!dentry);
++ AuDebugOn(!d_is_dir(dentry));
++
++ bbot = au_dbtaildir(dentry);
++ for (bindex = au_dbtop(dentry);
++ bindex <= bbot && sz < KMALLOC_MAX_SIZE;
++ bindex++) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry && d_is_positive(h_dentry))
++ sz += i_size_read(d_inode(h_dentry));
++ }
++ }
++ if (sz < KMALLOC_MAX_SIZE)
++ sz = roundup_pow_of_two(sz);
++ if (sz > KMALLOC_MAX_SIZE)
++ sz = KMALLOC_MAX_SIZE;
++ else if (sz < NAME_MAX) {
++ BUILD_BUG_ON(AUFS_RDBLK_DEF < NAME_MAX);
++ sz = AUFS_RDBLK_DEF;
++ }
++ return sz;
++}
++
++struct au_dir_ts_arg {
++ struct dentry *dentry;
++ aufs_bindex_t brid;
++};
++
++static void au_do_dir_ts(void *arg)
++{
++ struct au_dir_ts_arg *a = arg;
++ struct au_dtime dt;
++ struct path h_path;
++ struct inode *dir, *h_dir;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_hinode *hdir;
++ int err;
++ aufs_bindex_t btop, bindex;
++
++ sb = a->dentry->d_sb;
++ if (d_really_is_negative(a->dentry))
++ goto out;
++ /* no dir->i_mutex lock */
++ aufs_read_lock(a->dentry, AuLock_DW); /* noflush */
++
++ dir = d_inode(a->dentry);
++ btop = au_ibtop(dir);
++ bindex = au_br_index(sb, a->brid);
++ if (bindex < btop)
++ goto out_unlock;
++
++ br = au_sbr(sb, bindex);
++ h_path.dentry = au_h_dptr(a->dentry, bindex);
++ if (!h_path.dentry)
++ goto out_unlock;
++ h_path.mnt = au_br_mnt(br);
++ au_dtime_store(&dt, a->dentry, &h_path);
++
++ br = au_sbr(sb, btop);
++ if (!au_br_writable(br->br_perm))
++ goto out_unlock;
++ h_path.dentry = au_h_dptr(a->dentry, btop);
++ h_path.mnt = au_br_mnt(br);
++ err = vfsub_mnt_want_write(h_path.mnt);
++ if (err)
++ goto out_unlock;
++ hdir = au_hi(dir, btop);
++ au_hn_inode_lock_nested(hdir, AuLsc_I_PARENT);
++ h_dir = au_h_iptr(dir, btop);
++ if (h_dir->i_nlink
++ && timespec64_compare(&h_dir->i_mtime, &dt.dt_mtime) < 0) {
++ dt.dt_h_path = h_path;
++ au_dtime_revert(&dt);
++ }
++ au_hn_inode_unlock(hdir);
++ vfsub_mnt_drop_write(h_path.mnt);
++ au_cpup_attr_timesizes(dir);
++
++out_unlock:
++ aufs_read_unlock(a->dentry, AuLock_DW);
++out:
++ dput(a->dentry);
++ au_nwt_done(&au_sbi(sb)->si_nowait);
++ au_kfree_try_rcu(arg);
++}
++
++void au_dir_ts(struct inode *dir, aufs_bindex_t bindex)
++{
++ int perm, wkq_err;
++ aufs_bindex_t btop;
++ struct au_dir_ts_arg *arg;
++ struct dentry *dentry;
++ struct super_block *sb;
++
++ IMustLock(dir);
++
++ dentry = d_find_any_alias(dir);
++ AuDebugOn(!dentry);
++ sb = dentry->d_sb;
++ btop = au_ibtop(dir);
++ if (btop == bindex) {
++ au_cpup_attr_timesizes(dir);
++ goto out;
++ }
++
++ perm = au_sbr_perm(sb, btop);
++ if (!au_br_writable(perm))
++ goto out;
++
++ arg = kmalloc(sizeof(*arg), GFP_NOFS);
++ if (!arg)
++ goto out;
++
++ arg->dentry = dget(dentry); /* will be dput-ted by au_do_dir_ts() */
++ arg->brid = au_sbr_id(sb, bindex);
++ wkq_err = au_wkq_nowait(au_do_dir_ts, arg, sb, /*flags*/0);
++ if (unlikely(wkq_err)) {
++ pr_err("wkq %d\n", wkq_err);
++ dput(dentry);
++ au_kfree_try_rcu(arg);
++ }
++
++out:
++ dput(dentry);
++}
++
++/* ---------------------------------------------------------------------- */
++
+static int reopen_dir(struct file *file)
+{
+ int err;
++ unsigned int flags;
++ aufs_bindex_t bindex, btail, btop;
+ struct dentry *dentry, *h_dentry;
-+ aufs_bindex_t bindex, btail, bstart;
+ struct file *h_file;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ AuDebugOn(!S_ISDIR(dentry->d_inode->i_mode));
-+
-+ /* open all hidden dirs */
-+ bstart = au_dbstart(dentry);
-+#if 1 /* todo: necessary? */
-+ for (bindex = au_fbstart(file); bindex < bstart; bindex++)
++ /* open all lower dirs */
++ dentry = file->f_path.dentry;
++ btop = au_dbtop(dentry);
++ for (bindex = au_fbtop(file); bindex < btop; bindex++)
+ au_set_h_fptr(file, bindex, NULL);
-+#endif
-+ au_set_fbstart(file, bstart);
++ au_set_fbtop(file, btop);
++
+ btail = au_dbtaildir(dentry);
-+#if 1 /* todo: necessary? */
-+ for (bindex = au_fbend(file); btail < bindex; bindex--)
++ for (bindex = au_fbbot_dir(file); btail < bindex; bindex--)
+ au_set_h_fptr(file, bindex, NULL);
-+#endif
-+ au_set_fbend(file, btail);
-+ for (bindex = bstart; bindex <= btail; bindex++) {
++ au_set_fbbot_dir(file, btail);
++
++ flags = vfsub_file_flags(file);
++ for (bindex = btop; bindex <= btail; bindex++) {
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (!h_dentry)
+ continue;
-+ h_file = au_h_fptr(file, bindex);
-+ if (h_file) {
-+ AuDebugOn(h_file->f_dentry != h_dentry);
++ h_file = au_hf_dir(file, bindex);
++ if (h_file)
+ continue;
-+ }
+
-+ h_file = au_h_open(dentry, bindex, file->f_flags, file);
++ h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0);
+ err = PTR_ERR(h_file);
+ if (IS_ERR(h_file))
+ goto out; /* close all? */
-+ /* cpup_file_flags(h_file, file); */
+ au_set_h_fptr(file, bindex, h_file);
+ }
+ au_update_figen(file);
+ /* file->f_ra = h_file->f_ra; */
+ err = 0;
+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static int do_open_dir(struct file *file, int flags)
++static int do_open_dir(struct file *file, int flags, struct file *h_file)
+{
+ int err;
+ aufs_bindex_t bindex, btail;
+ struct dentry *dentry, *h_dentry;
-+ struct file *h_file;
++ struct vfsmount *mnt;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, 0x%x\n", AuDLNPair(dentry), flags);
-+ AuDebugOn(!dentry->d_inode || !S_ISDIR(dentry->d_inode->i_mode));
++ FiMustWriteLock(file);
++ AuDebugOn(h_file);
+
+ err = 0;
-+ au_set_fvdir_cache(file, NULL);
-+ file->f_version = dentry->d_inode->i_version;
-+ bindex = au_dbstart(dentry);
-+ au_set_fbstart(file, bindex);
++ mnt = file->f_path.mnt;
++ dentry = file->f_path.dentry;
++ file->f_version = inode_query_iversion(d_inode(dentry));
++ bindex = au_dbtop(dentry);
++ au_set_fbtop(file, bindex);
+ btail = au_dbtaildir(dentry);
-+ au_set_fbend(file, btail);
++ au_set_fbbot_dir(file, btail);
+ for (; !err && bindex <= btail; bindex++) {
+ h_dentry = au_h_dptr(dentry, bindex);
+ if (!h_dentry)
+ continue;
+
-+ h_file = au_h_open(dentry, bindex, flags, file);
++ err = vfsub_test_mntns(mnt, h_dentry->d_sb);
++ if (unlikely(err))
++ break;
++ h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0);
+ if (IS_ERR(h_file)) {
+ err = PTR_ERR(h_file);
+ break;
+ return 0; /* success */
+
+ /* close all */
-+ for (bindex = au_fbstart(file); bindex <= btail; bindex++)
++ for (bindex = au_fbtop(file); bindex <= btail; bindex++)
+ au_set_h_fptr(file, bindex, NULL);
-+ au_set_fbstart(file, -1);
-+ au_set_fbend(file, -1);
++ au_set_fbtop(file, -1);
++ au_set_fbbot_dir(file, -1);
++
+ return err;
+}
+
-+static int aufs_open_dir(struct inode *inode, struct file *file)
++static int aufs_open_dir(struct inode *inode __maybe_unused,
++ struct file *file)
+{
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(file->f_dentry));
++ int err;
++ struct super_block *sb;
++ struct au_fidir *fidir;
+
-+ return au_do_open(inode, file, do_open_dir);
++ err = -ENOMEM;
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ fidir = au_fidir_alloc(sb);
++ if (fidir) {
++ struct au_do_open_args args = {
++ .open = do_open_dir,
++ .fidir = fidir
++ };
++ err = au_do_open(file, &args);
++ if (unlikely(err))
++ au_kfree_rcu(fidir);
++ }
++ si_read_unlock(sb);
++ return err;
+}
+
-+static int aufs_release_dir(struct inode *inode, struct file *file)
++static int aufs_release_dir(struct inode *inode __maybe_unused,
++ struct file *file)
+{
+ struct au_vdir *vdir_cache;
-+ struct super_block *sb;
-+
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(file->f_dentry));
++ struct au_finfo *finfo;
++ struct au_fidir *fidir;
++ struct au_hfile *hf;
++ aufs_bindex_t bindex, bbot;
+
-+ sb = file->f_dentry->d_sb;
-+ si_noflush_read_lock(sb);
-+ fi_write_lock(file);
-+ vdir_cache = au_fvdir_cache(file);
-+ if (vdir_cache)
-+ au_vdir_free(vdir_cache);
-+ fi_write_unlock(file);
++ finfo = au_fi(file);
++ fidir = finfo->fi_hdir;
++ if (fidir) {
++ 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++, hf++)
++ if (hf->hf_file)
++ au_hfput(hf, /*execed*/0);
++ }
++ au_kfree_rcu(fidir);
++ finfo->fi_hdir = NULL;
++ }
+ au_finfo_fin(file);
-+ si_read_unlock(sb);
+ return 0;
+}
+
-+static int fsync_dir(struct dentry *dentry, int datasync)
++/* ---------------------------------------------------------------------- */
++
++static int au_do_flush_dir(struct file *file, fl_owner_t id)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct file *h_file;
++
++ err = 0;
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file); !err && bindex <= bbot; bindex++) {
++ h_file = au_hf_dir(file, bindex);
++ if (h_file)
++ err = vfsub_flush(h_file, id);
++ }
++ return err;
++}
++
++static int aufs_flush_dir(struct file *file, fl_owner_t id)
++{
++ return au_do_flush(file, id, au_do_flush_dir);
++}
++
++/* ---------------------------------------------------------------------- */
++
++static int au_do_fsync_dir_no_file(struct dentry *dentry, int datasync)
+{
+ int err;
++ aufs_bindex_t bbot, bindex;
+ struct inode *inode;
+ struct super_block *sb;
-+ aufs_bindex_t bend, bindex;
+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), datasync);
-+ DiMustAnyLock(dentry);
++ err = 0;
+ sb = dentry->d_sb;
-+ SiMustAnyLock(sb);
-+ inode = dentry->d_inode;
++ inode = d_inode(dentry);
+ IMustLock(inode);
-+ IiMustAnyLock(inode);
-+
-+ err = 0;
-+ bend = au_dbend(dentry);
-+ for (bindex = au_dbstart(dentry); !err && bindex <= bend; bindex++) {
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ struct file_operations *fop;
++ bbot = au_dbbot(dentry);
++ for (bindex = au_dbtop(dentry); !err && bindex <= bbot; bindex++) {
++ struct path h_path;
+
+ if (au_test_ro(sb, bindex, inode))
+ continue;
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (!h_dentry)
++ h_path.dentry = au_h_dptr(dentry, bindex);
++ if (!h_path.dentry)
+ continue;
-+ h_inode = h_dentry->d_inode;
-+ if (!h_inode)
++
++ h_path.mnt = au_sbr_mnt(sb, bindex);
++ err = vfsub_fsync(NULL, &h_path, datasync);
++ }
++
++ return err;
++}
++
++static int au_do_fsync_dir(struct file *file, int datasync)
++{
++ int err;
++ aufs_bindex_t bbot, bindex;
++ struct file *h_file;
++ struct super_block *sb;
++ struct inode *inode;
++
++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1, /*fi_lsc*/0);
++ if (unlikely(err))
++ goto out;
++
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file); !err && bindex <= bbot; bindex++) {
++ h_file = au_hf_dir(file, bindex);
++ if (!h_file || au_test_ro(sb, bindex, inode))
+ continue;
+
-+ /* no mnt_want_write() */
-+ /* cf. fs/nsfd/vfs.c and fs/nfsd/nfs4recover.c */
-+ /* todo: inotiry fired? */
-+ mutex_lock(&h_inode->i_mutex);
-+ fop = (void *)h_inode->i_fop;
-+ err = filemap_fdatawrite(h_inode->i_mapping);
-+ if (!err && fop && fop->fsync)
-+ err = fop->fsync(NULL, h_dentry, datasync);
-+ if (!err)
-+ err = filemap_fdatawrite(h_inode->i_mapping);
-+ if (!err)
-+ au_update_fuse_h_inode(NULL, h_dentry); /*ignore*/
-+ mutex_unlock(&h_inode->i_mutex);
++ err = vfsub_fsync(h_file, &h_file->f_path, datasync);
+ }
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
+/*
+ * @file may be NULL
+ */
-+static int aufs_fsync_dir(struct file *file, struct dentry *dentry,
++static int aufs_fsync_dir(struct file *file, loff_t start, loff_t end,
+ int datasync)
+{
+ int err;
++ struct dentry *dentry;
+ struct inode *inode;
-+ struct file *h_file;
+ struct super_block *sb;
-+ aufs_bindex_t bend, bindex;
-+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), datasync);
-+ inode = dentry->d_inode;
-+ IMustLock(inode);
+
+ err = 0;
++ dentry = file->f_path.dentry;
++ inode = d_inode(dentry);
++ inode_lock(inode);
+ sb = dentry->d_sb;
+ si_noflush_read_lock(sb);
-+ if (file) {
-+ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1,
-+ /*locked*/1);
-+ if (unlikely(err))
-+ goto out;
-+ } else
++ if (file)
++ err = au_do_fsync_dir(file, datasync);
++ else {
+ di_write_lock_child(dentry);
++ err = au_do_fsync_dir_no_file(dentry, datasync);
++ }
++ au_cpup_attr_timesizes(inode);
++ di_write_unlock(dentry);
++ if (file)
++ fi_write_unlock(file);
+
-+ if (file) {
-+ bend = au_fbend(file);
-+ for (bindex = au_fbstart(file); !err && bindex <= bend;
-+ bindex++) {
-+ h_file = au_h_fptr(file, bindex);
-+ if (!h_file || au_test_ro(sb, bindex, inode))
-+ continue;
-+
-+ err = -EINVAL;
-+ if (h_file->f_op && h_file->f_op->fsync) {
-+ /* todo: try do_fsync() in fs/sync.c? */
-+ mutex_lock(&h_file->f_mapping->host->i_mutex);
-+ err = h_file->f_op->fsync
-+ (h_file, h_file->f_dentry, datasync);
-+ if (!err)
-+ au_update_fuse_h_inode
-+ (h_file->f_vfsmnt,
-+ h_file->f_dentry);
-+ /*ignore*/
-+ mutex_unlock(&h_file->f_mapping->host->i_mutex);
-+ }
-+ }
-+ } else
-+ err = fsync_dir(dentry, datasync);
-+ au_cpup_attr_timesizes(inode);
-+ di_write_unlock(dentry);
-+ if (file)
-+ fi_write_unlock(file);
-+
-+ out:
+ si_read_unlock(sb);
-+ AuTraceErr(err);
++ inode_unlock(inode);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int aufs_readdir(struct file *file, void *dirent, filldir_t filldir)
++static int aufs_iterate_shared(struct file *file, struct dir_context *ctx)
+{
+ int err;
+ struct dentry *dentry;
-+ struct inode *inode;
++ struct inode *inode, *h_inode;
+ struct super_block *sb;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, pos %lld\n", AuDLNPair(dentry), file->f_pos);
-+ inode = dentry->d_inode;
++ AuDbg("%pD, ctx{%ps, %llu}\n", file, ctx->actor, ctx->pos);
++
++ dentry = file->f_path.dentry;
++ inode = d_inode(dentry);
+ IMustLock(inode);
+
+ sb = dentry->d_sb;
+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1, /*locked*/1);
++ err = au_reval_and_lock_fdi(file, reopen_dir, /*wlock*/1, /*fi_lsc*/0);
+ if (unlikely(err))
+ goto out;
-+ err = au_vdir_init(file);
++ err = au_alive_dir(dentry);
++ if (!err)
++ err = au_vdir_init(file);
+ di_downgrade_lock(dentry, AuLock_IR);
+ if (unlikely(err))
+ goto out_unlock;
+
-+ if (!au_test_nfsd(current)) {
-+ err = au_vdir_fill_de(file, dirent, filldir);
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
++ if (!au_test_nfsd()) {
++ err = au_vdir_fill_de(file, ctx);
++ fsstack_copy_attr_atime(inode, h_inode);
+ } else {
+ /*
+ * nfsd filldir may call lookup_one_len(), vfs_getattr(),
+ * encode_fh() and others.
+ */
-+ struct inode *h_inode = au_h_iptr(inode, au_ibstart(inode));
-+
++ atomic_inc(&h_inode->i_count);
+ di_read_unlock(dentry, AuLock_IR);
+ si_read_unlock(sb);
-+ lockdep_off();
-+ err = au_vdir_fill_de(file, dirent, filldir);
-+ lockdep_on();
++ err = au_vdir_fill_de(file, ctx);
+ fsstack_copy_attr_atime(inode, h_inode);
+ fi_write_unlock(file);
++ iput(h_inode);
+
+ AuTraceErr(err);
+ return err;
+ }
+
-+ fsstack_copy_attr_atime(inode, au_h_iptr(inode, au_ibstart(inode)));
-+
-+ out_unlock:
++out_unlock:
+ di_read_unlock(dentry, AuLock_IR);
+ fi_write_unlock(file);
-+ out:
++out:
+ si_read_unlock(sb);
-+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
+#define AuTestEmpty_WHONLY 1
-+#define AuTestEmpty_DLGT (1 << 1)
-+#define AuTestEmpty_DIRPERM1 (1 << 2)
-+#define AuTestEmpty_CALLED (1 << 3)
-+#define AuTestEmpty_SHWH (1 << 4)
++#define AuTestEmpty_CALLED (1 << 1)
++#define AuTestEmpty_SHWH (1 << 2)
+#define au_ftest_testempty(flags, name) ((flags) & AuTestEmpty_##name)
-+#define au_fset_testempty(flags, name) { (flags) |= AuTestEmpty_##name; }
-+#define au_fclr_testempty(flags, name) { (flags) &= ~AuTestEmpty_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuTestEmpty_DLGT
-+#define AuTestEmpty_DLGT 0
-+#undef AuTestEmpty_DIRPERM1
-+#define AuTestEmpty_DIRPERM1 0
-+#endif
++#define au_fset_testempty(flags, name) \
++ do { (flags) |= AuTestEmpty_##name; } while (0)
++#define au_fclr_testempty(flags, name) \
++ do { (flags) &= ~AuTestEmpty_##name; } while (0)
++
+#ifndef CONFIG_AUFS_SHWH
+#undef AuTestEmpty_SHWH
+#define AuTestEmpty_SHWH 0
+#endif
+
+struct test_empty_arg {
++ struct dir_context ctx;
+ struct au_nhash *whlist;
+ unsigned int flags;
+ int err;
+ aufs_bindex_t bindex;
+};
+
-+static int test_empty_cb(void *__arg, const char *__name, int namelen,
-+ loff_t offset, u64 ino, unsigned int d_type)
++static bool test_empty_cb(struct dir_context *ctx, const char *__name,
++ int namelen, loff_t offset __maybe_unused, u64 ino,
++ unsigned int d_type)
+{
-+ struct test_empty_arg *arg = __arg;
++ struct test_empty_arg *arg = container_of(ctx, struct test_empty_arg,
++ ctx);
+ char *name = (void *)__name;
+
-+ LKTRTrace("%.*s\n", namelen, name);
-+
+ arg->err = 0;
+ au_fset_testempty(arg->flags, CALLED);
+ /* smp_mb(); */
+ if (name[0] == '.'
+ && (namelen == 1 || (name[1] == '.' && namelen == 2)))
-+ return 0; /* success */
++ goto out; /* success */
+
+ if (namelen <= AUFS_WH_PFX_LEN
+ || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
+ (arg->whlist, name, namelen, ino, d_type, arg->bindex,
+ au_ftest_testempty(arg->flags, SHWH));
+
-+ out:
++out:
+ /* smp_mb(); */
+ AuTraceErr(arg->err);
-+ return arg->err;
++ return !arg->err;
+}
+
+static int do_test_empty(struct dentry *dentry, struct test_empty_arg *arg)
+{
-+ int err, dlgt;
++ int err;
+ struct file *h_file;
-+
-+ LKTRTrace("%.*s, {%p, 0x%x, %d}\n",
-+ AuDLNPair(dentry), arg->whlist, arg->flags, arg->bindex);
++ struct au_branch *br;
+
+ h_file = au_h_open(dentry, arg->bindex,
+ O_RDONLY | O_NONBLOCK | O_DIRECTORY | O_LARGEFILE,
-+ /*file*/NULL);
++ /*file*/NULL, /*force_wr*/0);
+ err = PTR_ERR(h_file);
+ if (IS_ERR(h_file))
+ goto out;
++
+ err = 0;
-+ if (au_opt_test(au_mntflags(dentry->d_sb), UDBA_INOTIFY)
-+ && !h_file->f_dentry->d_inode->i_nlink)
++ if (!au_opt_test(au_mntflags(dentry->d_sb), UDBA_NONE)
++ && !file_inode(h_file)->i_nlink)
+ goto out_put;
+
-+ dlgt = au_ftest_testempty(arg->flags, DLGT);
+ do {
+ arg->err = 0;
+ au_fclr_testempty(arg->flags, CALLED);
+ /* smp_mb(); */
-+ err = vfsub_readdir(h_file, test_empty_cb, arg, dlgt);
++ err = vfsub_iterate_dir(h_file, &arg->ctx);
+ if (err >= 0)
+ err = arg->err;
+ } while (!err && au_ftest_testempty(arg->flags, CALLED));
+
-+ out_put:
++out_put:
+ fput(h_file);
-+ au_sbr_put(dentry->d_sb, arg->bindex);
-+ out:
-+ AuTraceErr(err);
++ br = au_sbr(dentry->d_sb, arg->bindex);
++ au_lcnt_dec(&br->br_nfiles);
++out:
+ return err;
+}
+
+ int err, wkq_err;
+ struct dentry *h_dentry;
+ struct inode *h_inode;
++ struct mnt_idmap *h_idmap;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ h_idmap = au_sbr_idmap(dentry->d_sb, arg->bindex);
+ h_dentry = au_h_dptr(dentry, arg->bindex);
-+ AuDebugOn(!h_dentry);
-+ h_inode = h_dentry->d_inode;
-+ AuDebugOn(!h_inode);
-+
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ err = au_test_h_perm_sio(h_inode, MAY_EXEC | MAY_READ,
-+ au_test_dlgt(au_mntflags(dentry->d_sb)));
-+ mutex_unlock(&h_inode->i_mutex);
++ h_inode = d_inode(h_dentry);
++ /* todo: i_mode changes anytime? */
++ inode_lock_shared_nested(h_inode, AuLsc_I_CHILD);
++ err = au_test_h_perm_sio(h_idmap, h_inode, MAY_EXEC | MAY_READ);
++ inode_unlock_shared(h_inode);
+ if (!err)
+ err = do_test_empty(dentry, arg);
+ else {
+ };
+ unsigned int flags = arg->flags;
+
-+ au_fclr_testempty(arg->flags, DLGT);
-+ au_fclr_testempty(arg->flags, DIRPERM1);
-+ wkq_err = au_wkq_wait(call_do_test_empty, &args, /*dlgt*/0);
++ wkq_err = au_wkq_wait(call_do_test_empty, &args);
+ if (unlikely(wkq_err))
+ err = wkq_err;
+ arg->flags = flags;
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
+int au_test_empty_lower(struct dentry *dentry)
+{
+ int err;
-+ struct inode *inode;
-+ struct test_empty_arg arg;
-+ struct au_nhash *whlist;
-+ aufs_bindex_t bindex, bstart, btail;
-+ unsigned int mnt_flags;
++ unsigned int rdhash;
++ aufs_bindex_t bindex, btop, btail;
++ struct au_nhash whlist;
++ struct test_empty_arg arg = {
++ .ctx = {
++ .actor = test_empty_cb
++ }
++ };
++ int (*test_empty)(struct dentry *dentry, struct test_empty_arg *arg);
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode || !S_ISDIR(inode->i_mode));
++ SiMustAnyLock(dentry->d_sb);
+
-+ whlist = au_nhash_new(GFP_NOFS);
-+ err = PTR_ERR(whlist);
-+ if (IS_ERR(whlist))
++ rdhash = au_sbi(dentry->d_sb)->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, dentry));
++ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
+ goto out;
+
-+ bstart = au_dbstart(dentry);
-+ mnt_flags = au_mntflags(dentry->d_sb);
-+ arg.whlist = whlist;
+ arg.flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_testempty(arg.flags, DLGT);
-+ if (au_opt_test(mnt_flags, SHWH))
++ arg.whlist = &whlist;
++ btop = au_dbtop(dentry);
++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH))
+ au_fset_testempty(arg.flags, SHWH);
-+ arg.bindex = bstart;
-+ err = do_test_empty(dentry, &arg);
++ test_empty = do_test_empty;
++ if (au_opt_test(au_mntflags(dentry->d_sb), DIRPERM1))
++ test_empty = sio_test_empty;
++ arg.bindex = btop;
++ err = test_empty(dentry, &arg);
+ if (unlikely(err))
+ goto out_whlist;
+
+ au_fset_testempty(arg.flags, WHONLY);
-+ if (au_test_dirperm1(mnt_flags))
-+ au_fset_testempty(arg.flags, DIRPERM1);
+ btail = au_dbtaildir(dentry);
-+ for (bindex = bstart + 1; !err && bindex <= btail; bindex++) {
++ for (bindex = btop + 1; !err && bindex <= btail; bindex++) {
+ struct dentry *h_dentry;
++
+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (h_dentry && h_dentry->d_inode) {
++ if (h_dentry && d_is_positive(h_dentry)) {
+ arg.bindex = bindex;
-+ err = do_test_empty(dentry, &arg);
++ err = test_empty(dentry, &arg);
+ }
+ }
+
-+ out_whlist:
-+ au_nhash_del(whlist);
-+ out:
-+ AuTraceErr(err);
++out_whlist:
++ au_nhash_wh_free(&whlist);
++out:
+ return err;
+}
+
+int au_test_empty(struct dentry *dentry, struct au_nhash *whlist)
+{
+ int err;
-+ struct inode *inode;
-+ struct test_empty_arg arg;
++ struct test_empty_arg arg = {
++ .ctx = {
++ .actor = test_empty_cb
++ }
++ };
+ aufs_bindex_t bindex, btail;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode || !S_ISDIR(inode->i_mode));
-+
+ err = 0;
+ arg.whlist = whlist;
+ arg.flags = AuTestEmpty_WHONLY;
+ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH))
+ au_fset_testempty(arg.flags, SHWH);
+ btail = au_dbtaildir(dentry);
-+ for (bindex = au_dbstart(dentry); !err && bindex <= btail; bindex++) {
++ for (bindex = au_dbtop(dentry); !err && bindex <= btail; bindex++) {
+ struct dentry *h_dentry;
++
+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (h_dentry && h_dentry->d_inode) {
++ if (h_dentry && d_is_positive(h_dentry)) {
+ arg.bindex = bindex;
+ err = sio_test_empty(dentry, &arg);
+ }
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct file_operations aufs_dir_fop = {
++const struct file_operations aufs_dir_fop = {
++ .owner = THIS_MODULE,
++ .llseek = default_llseek,
+ .read = generic_read_dir,
-+ .readdir = aufs_readdir,
-+ /* .unlocked_ioctl = aufs_ioctl_dir, */
++ .iterate_shared = aufs_iterate_shared,
++ .unlocked_ioctl = aufs_ioctl_dir,
++#ifdef CONFIG_COMPAT
++ .compat_ioctl = aufs_compat_ioctl_dir,
++#endif
+ .open = aufs_open_dir,
+ .release = aufs_release_dir,
-+ .flush = aufs_flush,
-+ .fsync = aufs_fsync_dir,
++ .flush = aufs_flush_dir,
++ .fsync = aufs_fsync_dir
+};
-diff -urN linux/fs/aufs/dir.h linux-aufs/fs/aufs/dir.h
---- linux/fs/aufs/dir.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dir.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,147 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,134 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * directory operations
-+ *
-+ * $Id$
+ */
+
+#ifndef __AUFS_DIR_H__
+#ifdef __KERNEL__
+
+#include <linux/fs.h>
-+#include <linux/aufs_type.h>
+
+/* ---------------------------------------------------------------------- */
+
+/* need to be faster and smaller */
+
-+/* todo: changeable? */
-+#define AuSize_DEBLK 512
-+#define AuSize_NHASH 32
-+#if AuSize_DEBLK < NAME_MAX || PAGE_SIZE < AuSize_DEBLK
-+#error invalid size AuSize_DEBLK
-+#endif
-+
-+typedef char au_vdir_deblk_t[AuSize_DEBLK];
-+
+struct au_nhash {
-+ struct hlist_head heads[AuSize_NHASH];
++ unsigned int nh_num;
++ struct hlist_head *nh_head;
+};
+
+struct au_vdir_destr {
+ unsigned char len;
-+ char name[0];
++ unsigned char name[];
+} __packed;
+
+struct au_vdir_dehstr {
+ struct hlist_node hash;
+ struct au_vdir_destr *str;
-+};
++ struct rcu_head rcu;
++} ____cacheline_aligned_in_smp;
+
+struct au_vdir_de {
+ ino_t de_ino;
+
+struct au_vdir_wh {
+ struct hlist_node wh_hash;
-+ aufs_bindex_t wh_bindex;
+#ifdef CONFIG_AUFS_SHWH
+ ino_t wh_ino;
++ aufs_bindex_t wh_bindex;
+ unsigned char wh_type;
-+ /* caution: packed */
++#else
++ aufs_bindex_t wh_bindex;
+#endif
++ /* caution: packed */
+ struct au_vdir_destr wh_str;
+} __packed;
+
+union au_vdir_deblk_p {
-+ unsigned char *p;
-+ au_vdir_deblk_t *deblk;
++ unsigned char *deblk;
+ struct au_vdir_de *de;
+};
+
+struct au_vdir {
-+ au_vdir_deblk_t **vd_deblk;
-+ int vd_nblk;
++ unsigned char **vd_deblk;
++ unsigned long vd_nblk;
+ struct {
-+ int i;
++ unsigned long ul;
+ union au_vdir_deblk_p p;
+ } vd_last;
+
-+ unsigned long vd_version;
++ u64 vd_version;
++ unsigned int vd_deblk_sz;
+ unsigned long vd_jiffy;
-+};
++ struct rcu_head rcu;
++} ____cacheline_aligned_in_smp;
+
+/* ---------------------------------------------------------------------- */
+
+/* dir.c */
-+extern struct file_operations aufs_dir_fop;
++extern const struct file_operations aufs_dir_fop;
++void au_add_nlink(struct inode *dir, struct inode *h_dir);
++void au_sub_nlink(struct inode *dir, struct inode *h_dir);
++loff_t au_dir_size(struct file *file, struct dentry *dentry);
++void au_dir_ts(struct inode *dir, aufs_bindex_t bsrc);
+int au_test_empty_lower(struct dentry *dentry);
+int au_test_empty(struct dentry *dentry, struct au_nhash *whlist);
+
+/* vdir.c */
-+struct au_nhash *au_nhash_new(gfp_t gfp);
-+void au_nhash_del(struct au_nhash *nhash);
-+void au_nhash_init(struct au_nhash *nhash);
-+void au_nhash_move(struct au_nhash *dst, struct au_nhash *src);
-+void au_nhash_fin(struct au_nhash *nhash);
++unsigned int au_rdhash_est(loff_t sz);
++int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp);
++void au_nhash_wh_free(struct au_nhash *whlist);
+int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt,
+ int limit);
-+int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int namelen);
-+int au_nhash_append_wh(struct au_nhash *whlist, char *name, int namelen,
-+ ino_t ino, unsigned int d_type, aufs_bindex_t bindex,
++int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen);
++int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino,
++ unsigned int d_type, aufs_bindex_t bindex,
+ unsigned char shwh);
+void au_vdir_free(struct au_vdir *vdir);
+int au_vdir_init(struct file *file);
-+int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir);
++int au_vdir_fill_de(struct file *file, struct dir_context *ctx);
+
-+/* ---------------------------------------------------------------------- */
++/* ioctl.c */
++long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg);
+
-+static inline
-+void au_shwh_init_wh(struct au_vdir_wh *wh, ino_t ino, unsigned char d_type)
-+{
-+#ifdef CONFIG_AUFS_SHWH
-+ wh->wh_ino = ino;
-+ wh->wh_type = d_type;
++#ifdef CONFIG_AUFS_RDU
++/* rdu.c */
++long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
++#ifdef CONFIG_COMPAT
++long au_rdu_compat_ioctl(struct file *file, unsigned int cmd,
++ unsigned long arg);
++#endif
++#else
++AuStub(long, au_rdu_ioctl, return -EINVAL, struct file *file,
++ unsigned int cmd, unsigned long arg)
++#ifdef CONFIG_COMPAT
++AuStub(long, au_rdu_compat_ioctl, return -EINVAL, struct file *file,
++ unsigned int cmd, unsigned long arg)
++#endif
+#endif
-+}
-+
-+static inline void au_add_nlink(struct inode *dir, struct inode *h_dir)
-+{
-+ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode));
-+ dir->i_nlink += h_dir->i_nlink - 2;
-+ if (h_dir->i_nlink < 2)
-+ dir->i_nlink += 2;
-+}
-+
-+static inline void au_sub_nlink(struct inode *dir, struct inode *h_dir)
-+{
-+ AuDebugOn(!S_ISDIR(dir->i_mode) || !S_ISDIR(h_dir->i_mode));
-+ dir->i_nlink -= h_dir->i_nlink - 2;
-+ if (h_dir->i_nlink < 2)
-+ dir->i_nlink -= 2;
-+}
+
+#endif /* __KERNEL__ */
+#endif /* __AUFS_DIR_H__ */
-diff -urN linux/fs/aufs/dlgt.c linux-aufs/fs/aufs/dlgt.c
---- linux/fs/aufs/dlgt.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/dlgt.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,113 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,1315 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2017-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * lookup functions in 'delegate' mode
-+ *
-+ * $Id$
++ * special handling in renaming a directory
++ * in order to support looking-up the before-renamed name on the lower readonly
++ * branches
+ */
+
++#include <linux/byteorder/generic.h>
+#include "aufs.h"
+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_lookup_one_len_args {
-+ struct dentry **errp;
-+ const char *name;
-+ struct dentry *parent;
-+ int len;
-+};
-+
-+static void au_call_lookup_one_len(void *args)
++static void au_dr_hino_del(struct au_dr_br *dr, struct au_dr_hino *ent)
+{
-+ struct au_lookup_one_len_args *a = args;
-+ *a->errp = vfsub_lookup_one_len(a->name, a->parent, a->len);
++ int idx;
++
++ idx = au_dr_ihash(ent->dr_h_ino);
++ au_hbl_del(&ent->dr_hnode, dr->dr_h_ino + idx);
+}
+
-+struct dentry *au_lkup_one_dlgt(const char *name, struct dentry *parent,
-+ int len, unsigned int flags)
++static int au_dr_hino_test_empty(struct au_dr_br *dr)
+{
-+ struct dentry *dentry;
-+ int dirperm1;
-+
-+ LKTRTrace("%.*s/%.*s, 0x%x\n", AuDLNPair(parent), len, name, flags);
++ int ret, i;
++ struct hlist_bl_head *hbl;
+
-+ dirperm1 = au_ftest_ndx(flags, DIRPERM1);
-+ if (!dirperm1 && !au_ftest_ndx(flags, DLGT))
-+ dentry = vfsub_lookup_one_len(name, parent, len);
-+ else {
-+ int wkq_err;
-+ struct au_lookup_one_len_args args = {
-+ .errp = &dentry,
-+ .name = name,
-+ .parent = parent,
-+ .len = len
-+ };
-+ wkq_err = au_wkq_wait(au_call_lookup_one_len, &args,
-+ /*dlgt*/!dirperm1);
-+ if (unlikely(wkq_err))
-+ dentry = ERR_PTR(wkq_err);
++ 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);
+ }
+
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return ret;
+}
+
-+/* ---------------------------------------------------------------------- */
++static struct au_dr_hino *au_dr_hino_find(struct au_dr_br *dr, ino_t ino)
++{
++ struct au_dr_hino *found, *ent;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ int idx;
+
-+struct security_inode_permission_args {
-+ int *errp;
-+ struct inode *h_inode;
-+ int mask;
-+ struct nameidata *fake_nd;
-+};
++ found = NULL;
++ idx = au_dr_ihash(ino);
++ hbl = dr->dr_h_ino + idx;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(ent, pos, hbl, dr_hnode)
++ if (ent->dr_h_ino == ino) {
++ found = ent;
++ break;
++ }
++ hlist_bl_unlock(hbl);
+
-+static void call_security_inode_permission(void *args)
-+{
-+ struct security_inode_permission_args *a = args;
-+ LKTRTrace("fsuid %d\n", current->fsuid);
-+ *a->errp = vfsub_security_inode_permission(a->h_inode, a->mask,
-+ a->fake_nd);
++ return found;
+}
+
-+int au_security_inode_permission(struct inode *h_inode, int mask,
-+ struct nameidata *fake_nd, int dlgt)
++int au_dr_hino_test_add(struct au_dr_br *dr, ino_t ino,
++ struct au_dr_hino *add_ent)
+{
-+ int err;
++ int found, idx;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dr_hino *ent;
+
-+ AuTraceEnter();
++ found = 0;
++ idx = au_dr_ihash(ino);
++ hbl = dr->dr_h_ino + idx;
++#if 0 /* debug print */
++ {
++ struct hlist_bl_node *tmp;
+
-+ if (!dlgt)
-+ err = vfsub_security_inode_permission(h_inode, mask, fake_nd);
-+ else {
-+ int wkq_err;
-+ struct security_inode_permission_args args = {
-+ .errp = &err,
-+ .h_inode = h_inode,
-+ .mask = mask,
-+ .fake_nd = fake_nd
-+ };
-+ wkq_err = au_wkq_wait(call_security_inode_permission, &args,
-+ /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ hlist_bl_for_each_entry_safe(ent, pos, tmp, hbl, dr_hnode)
++ AuDbg("hi%llu\n", (unsigned long long)ent->dr_h_ino);
+ }
-+
-+ AuTraceErr(err);
-+ return err;
-+}
-diff -urN linux/fs/aufs/export.c linux-aufs/fs/aufs/export.c
---- linux/fs/aufs/export.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/export.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,826 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * export via nfs
-+ *
-+ * $Id$
-+ */
-+
-+#include <linux/exportfs.h>
-+#include <linux/mnt_namespace.h>
-+#include <linux/random.h>
-+#include "aufs.h"
-+
-+union conv {
-+#ifdef CONFIG_AUFS_INO_T_64
-+ __u32 a[2];
-+#else
-+ __u32 a[1];
+#endif
-+ ino_t ino;
-+};
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(ent, pos, hbl, dr_hnode)
++ if (ent->dr_h_ino == ino) {
++ found = 1;
++ break;
++ }
++ if (!found && add_ent)
++ hlist_bl_add_head(&add_ent->dr_hnode, hbl);
++ hlist_bl_unlock(hbl);
+
-+static ino_t decode_ino(__u32 *a)
-+{
-+ union conv u;
++ if (!found && add_ent)
++ AuDbg("i%llu added\n", (unsigned long long)add_ent->dr_h_ino);
+
-+ BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a));
-+ u.a[0] = a[0];
-+#ifdef CONFIG_AUFS_INO_T_64
-+ u.a[1] = a[1];
-+#endif
-+ return u.ino;
++ return found;
+}
+
-+static void encode_ino(__u32 *a, ino_t ino)
++void au_dr_hino_free(struct au_dr_br *dr)
+{
-+ union conv u;
-+
-+ u.ino = ino;
-+ a[0] = u.a[0];
-+#ifdef CONFIG_AUFS_INO_T_64
-+ a[1] = u.a[1];
-+#endif
-+}
-+
-+/* NFS file handle */
-+enum {
-+ Fh_br_id,
-+ Fh_sigen,
-+#ifdef CONFIG_AUFS_INO_T_64
-+ /* support 64bit inode number */
-+ Fh_ino1,
-+ Fh_ino2,
-+ Fh_dir_ino1,
-+ Fh_dir_ino2,
-+#else
-+ Fh_ino1,
-+ Fh_dir_ino1,
-+#endif
-+ Fh_igen,
-+ Fh_h_type,
-+ Fh_tail,
++ int i;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
++ struct au_dr_hino *ent;
+
-+ Fh_ino = Fh_ino1,
-+ Fh_dir_ino = Fh_dir_ino1
-+};
++ /* SiMustWriteLock(sb); */
+
-+static int au_test_anon(struct dentry *dentry)
-+{
-+ return !!(dentry->d_flags & DCACHE_DISCONNECTED);
++ 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)
++ au_kfree_rcu(ent);
++ INIT_HLIST_BL_HEAD(hbl);
++ }
+}
+
-+/* ---------------------------------------------------------------------- */
-+/* inode generation external table */
-+
-+int au_xigen_inc(struct inode *inode)
++/* returns the number of inodes or an error */
++static int au_dr_hino_store(struct super_block *sb, struct au_branch *br,
++ struct file *hinofile)
+{
-+ int err;
-+ loff_t pos;
-+ ssize_t sz;
-+ __u32 igen;
-+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
++ 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;
+
-+ LKTRTrace("i%lu\n", (unsigned long)inode->i_ino);
++ SiMustWriteLock(sb);
++ AuDebugOn(!au_br_writable(br->br_perm));
+
-+ err = 0;
-+ sb = inode->i_sb;
-+ if (unlikely(!au_opt_test_xino(au_mntflags(sb))))
-+ goto out;
++ hinoinode = file_inode(hinofile);
++ oldsize = i_size_read(hinoinode);
+
-+ pos = inode->i_ino;
-+ pos *= sizeof(igen);
-+ igen = inode->i_generation + 1;
-+ sbinfo = au_sbi(sb);
-+ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xigen, &igen,
-+ sizeof(igen), &pos);
-+ if (sz == sizeof(igen))
-+ goto out; /* success */
++ 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;
+
-+ err = sz;
-+ if (unlikely(sz >= 0)) {
-+ err = -EIO;
-+ AuIOErr("xigen error (%zd)\n", sz);
++ /* 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);
+ }
+
-+ out:
+ AuTraceErr(err);
+ return err;
+}
+
-+int au_xigen_new(struct inode *inode)
++static int au_dr_hino_load(struct au_dr_br *dr, struct file *hinofile)
+{
-+ int err;
++ int err, hidx;
++ ssize_t ssz;
++ size_t sz, n;
+ loff_t pos;
-+ ssize_t sz;
-+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file;
-+
-+ LKTRTrace("i%lu\n", inode->i_ino);
++ uint64_t u64;
++ struct au_dr_hino *ent;
++ struct inode *hinoinode;
++ struct hlist_bl_head *hbl;
+
+ err = 0;
-+ /* todo: dirty, at mount time */
-+ if (inode->i_ino == AUFS_ROOT_INO)
-+ goto out;
-+ sb = inode->i_sb;
-+ if (unlikely(!au_opt_test_xino(au_mntflags(sb))))
-+ goto out;
++ 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;
++ }
+
-+ err = -EFBIG;
-+ pos = inode->i_ino;
-+ if (unlikely(Au_LOFF_MAX / sizeof(inode->i_generation) - 1 < pos)) {
-+ AuIOErr1("too large i%lld\n", pos);
-+ goto out;
++ 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);
+ }
-+ pos *= sizeof(inode->i_generation);
++ goto out; /* success */
+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ file = sbinfo->si_xigen;
-+ BUG_ON(!file);
++out_free:
++ au_dr_hino_free(dr);
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ if (i_size_read(file->f_dentry->d_inode)
-+ < pos + sizeof(inode->i_generation)) {
-+ inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next);
-+ sz = xino_fwrite(sbinfo->si_xwrite, file, &inode->i_generation,
-+ sizeof(inode->i_generation), &pos);
-+ } else
-+ sz = xino_fread(sbinfo->si_xread, file, &inode->i_generation,
-+ sizeof(inode->i_generation), &pos);
-+ if (sz == sizeof(inode->i_generation))
-+ goto out; /* success */
++/*
++ * @bindex/@br is a switch to distinguish whether suspending hnotify or not.
++ * @path is a switch to distinguish load and store.
++ */
++static int au_dr_hino(struct super_block *sb, aufs_bindex_t bindex,
++ struct au_branch *br, const struct path *path)
++{
++ int err, flags;
++ unsigned char load, suspend;
++ struct file *hinofile;
++ struct au_hinode *hdir;
++ struct inode *dir, *delegated;
++ struct path hinopath;
++ struct qstr hinoname = QSTR_INIT(AUFS_WH_DR_BRHINO,
++ sizeof(AUFS_WH_DR_BRHINO) - 1);
+
-+ err = sz;
-+ if (unlikely(sz >= 0)) {
-+ err = -EIO;
-+ AuIOErr("xigen error (%zd)\n", sz);
++ AuDebugOn(bindex < 0 && !br);
++ AuDebugOn(bindex >= 0 && br);
++
++ err = -EINVAL;
++ suspend = !br;
++ if (suspend)
++ br = au_sbr(sb, bindex);
++ load = !!path;
++ if (!load) {
++ path = &br->br_path;
++ AuDebugOn(!au_br_writable(br->br_perm));
++ if (unlikely(!au_br_writable(br->br_perm)))
++ goto out;
++ }
++
++ hdir = NULL;
++ if (suspend) {
++ dir = d_inode(sb->s_root);
++ hdir = au_hinode(au_ii(dir), bindex);
++ dir = hdir->hi_inode;
++ au_hn_inode_lock_nested(hdir, AuLsc_I_CHILD);
++ } else {
++ dir = d_inode(path->dentry);
++ inode_lock_nested(dir, AuLsc_I_CHILD);
++ }
++ hinopath.mnt = path->mnt;
++ hinopath.dentry = vfsub_lkup_one(&hinoname, (struct path *)path);
++ err = PTR_ERR(hinopath.dentry);
++ if (IS_ERR(hinopath.dentry))
++ goto out_unlock;
++
++ err = 0;
++ flags = O_RDONLY;
++ if (load) {
++ if (d_is_negative(hinopath.dentry))
++ goto out_dput; /* success */
++ } else {
++ if (au_dr_hino_test_empty(&br->br_dirren)) {
++ if (d_is_positive(hinopath.dentry)) {
++ delegated = NULL;
++ err = vfsub_unlink(dir, &hinopath, &delegated,
++ /*force*/0);
++ AuTraceErr(err);
++ if (unlikely(err))
++ pr_err("ignored err %d, %pd2\n",
++ err, hinopath.dentry);
++ if (unlikely(err == -EWOULDBLOCK))
++ iput(delegated);
++ err = 0;
++ }
++ goto out_dput;
++ } else if (!d_is_positive(hinopath.dentry)) {
++ err = vfsub_create(dir, &hinopath, 0600,
++ /*want_excl*/false);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out_dput;
++ }
++ flags = O_WRONLY;
+ }
++ 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;
++ }
++
++ 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:
++out_dput:
++ dput(hinopath.dentry);
++out_unlock:
++ if (suspend)
++ au_hn_inode_unlock(hdir);
++ else
++ inode_unlock(dir);
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+int au_xigen_set(struct super_block *sb, struct file *base)
++/* ---------------------------------------------------------------------- */
++
++static int au_dr_brid_init(struct au_dr_brid *brid, const struct path *path)
+{
+ int err;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file;
++ struct kstatfs kstfs;
++ dev_t dev;
++ struct dentry *dentry;
++ struct super_block *sb;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(base->f_dentry));
-+ SiMustWriteLock(sb);
++ err = vfs_statfs((void *)path, &kstfs);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out;
+
-+ sbinfo = au_sbi(sb);
-+ file = au_xino_create2(sb, base, sbinfo->si_xigen);
-+ err = PTR_ERR(file);
-+ if (IS_ERR(file))
++ /* 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 err;
++}
++
++int au_dr_br_init(struct super_block *sb, struct au_branch *br,
++ const struct path *path)
++{
++ int err, i;
++ struct au_dr_br *dr;
++ struct hlist_bl_head *hbl;
++
++ dr = &br->br_dirren;
++ hbl = dr->dr_h_ino;
++ for (i = 0; i < AuDirren_NHASH; i++, hbl++)
++ INIT_HLIST_BL_HEAD(hbl);
++
++ err = au_dr_brid_init(&dr->dr_brid, path);
++ if (unlikely(err))
+ goto out;
-+ err = 0;
-+ if (sbinfo->si_xigen)
-+ fput(sbinfo->si_xigen);
-+ sbinfo->si_xigen = file;
+
-+ out:
++ if (au_opt_test(au_mntflags(sb), DIRREN))
++ err = au_dr_hino(sb, /*bindex*/-1, br, path);
++
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+void au_xigen_clr(struct super_block *sb)
++int au_dr_br_fin(struct super_block *sb, struct au_branch *br)
+{
-+ struct au_sbinfo *sbinfo;
++ int err;
+
-+ sbinfo = au_sbi(sb);
-+ if (sbinfo->si_xigen) {
-+ fput(sbinfo->si_xigen);
-+ sbinfo->si_xigen = NULL;
-+ }
++ err = 0;
++ if (au_br_writable(br->br_perm))
++ err = au_dr_hino(sb, /*bindex*/-1, br, /*path*/NULL);
++ if (!err)
++ au_dr_hino_free(&br->br_dirren);
++
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino,
-+ ino_t dir_ino)
++static int au_brid_str(struct au_dr_brid *brid, struct inode *h_inode,
++ char *buf, size_t sz)
+{
-+ struct dentry *dentry, *d;
-+ struct inode *inode;
-+ au_gen_t sigen;
++ int err;
++ unsigned int major, minor;
++ char *p;
+
-+ LKTRTrace("i%lu, diri%lu\n",
-+ (unsigned long)ino, (unsigned long)dir_ino);
++ 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;
+
-+ dentry = NULL;
-+ inode = ilookup(sb, ino);
-+ if (!inode)
-+ goto out;
++ return p - buf;
++}
+
-+ dentry = ERR_PTR(-ESTALE);
-+ sigen = au_sigen(sb);
-+ if (unlikely(is_bad_inode(inode)
-+ || IS_DEADDIR(inode)
-+ || sigen != au_iigen(inode)))
-+ goto out_iput;
++static int au_drinfo_name(struct au_branch *br, char *name, int len)
++{
++ int rlen;
++ struct dentry *br_dentry;
++ struct inode *br_inode;
+
-+ dentry = NULL;
-+ if (!dir_ino || S_ISDIR(inode->i_mode))
-+ dentry = d_find_alias(inode);
-+ else {
-+ spin_lock(&dcache_lock);
-+ list_for_each_entry(d, &inode->i_dentry, d_alias)
-+ if (!au_test_anon(d)
-+ && d->d_parent->d_inode->i_ino == dir_ino) {
-+ dentry = dget_locked(d);
-+ break;
-+ }
-+ spin_unlock(&dcache_lock);
-+ }
-+ if (unlikely(dentry && sigen != au_digen(dentry))) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ }
++ 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);
+
-+ out_iput:
-+ iput(inode);
-+ out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return rlen;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+/* todo: dirty? */
+/*
-+ * when you mntput() for the return value of this function,
-+ * you have to store it to your local var.
-+ * ie. never mntput si_mntcache directly.
++ * 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 struct vfsmount *au_do_mnt_get(struct super_block *sb)
++static int au_drinfo_construct(struct au_drinfo_fdata **fdata,
++ struct dentry *h_dentry,
++ unsigned char *allocated)
+{
-+ struct mnt_namespace *ns;
-+ struct vfsmount *pos, *mnt;
-+
-+ AuTraceEnter();
++ int err, v;
++ struct au_drinfo_fdata *f, *p;
++ struct au_drinfo *drinfo;
++ struct inode *h_inode;
++ struct qstr *qname;
+
-+ /* vfsmount_lock is not exported */
-+ /* no get/put ?? */
-+ AuDebugOn(!current->nsproxy);
-+ ns = current->nsproxy->mnt_ns;
-+ AuDebugOn(!ns);
-+ mnt = NULL;
-+ /* the order (reverse) will not be a problem */
-+ list_for_each_entry(pos, &ns->list, mnt_list)
-+ if (pos->mnt_sb == sb) {
-+ mnt = pos;
-+ break;
++ 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;
+ }
-+ AuDebugOn(!mnt);
++ 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);
+
-+ return mntget(mnt);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+static struct vfsmount *au_mnt_get(struct super_block *sb)
++/* callers have to free the return value */
++static struct au_drinfo *au_drinfo_read_k(struct file *file, ino_t h_ino)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct vfsmount *mnt;
++ struct au_drinfo *ret, *drinfo;
++ struct au_drinfo_fdata fdata;
++ int len;
++ loff_t pos;
++ ssize_t ssz;
+
-+ sbinfo = au_sbi(sb);
-+ spin_lock(&sbinfo->si_mntcache_lock);
-+ if (sbinfo->si_mntcache)
-+ mnt = mntget(sbinfo->si_mntcache);
-+ else {
-+ sbinfo->si_mntcache = au_do_mnt_get(sb);
-+ mnt = sbinfo->si_mntcache;
++ 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;
+ }
-+ spin_unlock(&sbinfo->si_mntcache_lock);
-+ return mnt;
-+}
-+#else
-+static struct vfsmount *au_mnt_get(struct super_block *sb)
-+{
-+ return au_do_mnt_get(sb);
-+}
-+#endif
-+
-+struct au_nfsd_si_lock {
-+ const au_gen_t sigen;
-+ const aufs_bindex_t br_id;
-+ unsigned char force_lock;
-+};
+
-+static aufs_bindex_t si_nfsd_read_lock(struct super_block *sb,
-+ struct au_nfsd_si_lock *nsi_lock)
-+{
-+ aufs_bindex_t bindex;
++ 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;
++ }
+
-+ si_read_lock(sb, AuLock_FLUSH);
++ drinfo = &fdata.drinfo;
++ len = drinfo->oldnamelen;
++ if (!len) {
++ AuIOErr("broken drinfo %pD2\n", file);
++ goto out;
++ }
+
-+ /* branch id may be wrapped around */
-+ bindex = au_br_index(sb, nsi_lock->br_id);
-+ LKTRTrace("b%d\n", bindex);
-+ if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb))
++ 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 */
++ }
+
-+ if (!nsi_lock->force_lock)
-+ si_read_unlock(sb);
-+ bindex = -1;
++ ret = kmalloc(sizeof(*ret) + len, GFP_NOFS);
++ if (unlikely(!ret)) {
++ ret = ERR_PTR(-ENOMEM);
++ AuTraceErrPtr(ret);
++ goto out;
++ }
+
-+ out:
-+ return bindex;
++ *ret = *drinfo;
++ ssz = vfsub_read_k(file, (void *)ret->oldname, len, &pos);
++ if (unlikely(ssz != len)) {
++ au_kfree_rcu(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;
+}
+
-+struct find_name_by_ino {
-+ int called, found;
-+ ino_t ino;
-+ char *name;
-+ int namelen;
-+};
++/* ---------------------------------------------------------------------- */
+
-+static int
-+find_name_by_ino(void *arg, const char *name, int namelen, loff_t offset,
-+ u64 ino, unsigned int d_type)
-+{
-+ struct find_name_by_ino *a = arg;
++/* in order to be revertible */
++struct au_drinfo_rev_elm {
++ int created;
++ struct dentry *info_dentry;
++ struct au_drinfo *info_last;
++};
+
-+ a->called++;
-+ if (a->ino != ino)
-+ return 0;
++struct au_drinfo_rev {
++ unsigned char already;
++ aufs_bindex_t nelm;
++ struct au_drinfo_rev_elm elm[];
++};
+
-+ memcpy(a->name, name, namelen);
-+ a->namelen = namelen;
-+ a->found = 1;
-+ return 1;
-+}
++/* 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 generated name */
++ renameback; /* renamed back */
++};
+
-+static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino,
-+ struct au_nfsd_si_lock *nsi_lock)
++/* 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)
+{
-+ struct dentry *dentry, *parent;
-+ struct file *file;
-+ struct inode *dir;
-+ struct find_name_by_ino arg;
-+ int err;
++ 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,
++ w->whnamelen);
++ AuTraceErrPtr(infopath.dentry);
++ if (IS_ERR(infopath.dentry)) {
++ err = PTR_ERR(infopath.dentry);
++ goto out;
++ }
+
-+ parent = path->dentry;
-+ LKTRTrace("%.*s, i%lu\n", AuDLNPair(parent), (unsigned long)ino);
++ 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);
++ }
+
-+ if (nsi_lock)
-+ si_read_unlock(parent->d_sb);
-+ path_get(path);
-+ file = dentry_open(parent, path->mnt, au_dir_roflags);
-+ dentry = (void *)file;
-+ if (IS_ERR(file))
-+ goto out;
++ infofile = vfsub_dentry_open(&infopath, O_RDWR);
++ AuTraceErrPtr(infofile);
++ if (IS_ERR(infofile)) {
++ err = PTR_ERR(infofile);
++ goto out_dput;
++ }
+
-+ dentry = ERR_PTR(-ENOMEM);
-+ arg.name = __getname();
-+ if (unlikely(!arg.name))
-+ goto out_file;
-+ arg.ino = ino;
-+ arg.found = 0;
-+ do {
-+ arg.called = 0;
-+ /* smp_mb(); */
-+ err = vfsub_readdir(file, find_name_by_ino, &arg, /*dlgt*/0);
-+ } while (!err && !arg.found && arg.called);
-+ dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_name;
-+ dentry = ERR_PTR(-ENOENT);
-+ if (!arg.found)
-+ goto out_name;
++ 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;
++ }
++ }
+
-+ /* do not call au_lkup_one(), nor dlgt */
-+ dir = parent->d_inode;
-+ mutex_lock(&dir->i_mutex);
-+ dentry = vfsub_lookup_one_len(arg.name, parent, arg.namelen);
-+ mutex_unlock(&dir->i_mutex);
-+ AuTraceErrPtr(dentry);
-+ if (IS_ERR(dentry))
-+ goto out_name;
-+ AuDebugOn(au_test_anon(dentry));
-+ if (unlikely(!dentry->d_inode)) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ENOENT);
++ 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;
+ }
+
-+ out_name:
-+ __putname(arg.name);
-+ out_file:
-+ fput(file);
-+ out:
-+ if (unlikely(nsi_lock
-+ && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0))
-+ if (!IS_ERR(dentry)) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ }
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ 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;
+}
+
-+static /* noinline_for_stack */
-+struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino,
-+ ino_t dir_ino, struct au_nfsd_si_lock *nsi_lock)
++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 dentry *dentry, *parent;
-+ struct path path;
++ struct au_call_drinfo_do_store_args *a = args;
+
-+ LKTRTrace("i%lu, diri%lu\n",
-+ (unsigned long)ino, (unsigned long)dir_ino);
++ *a->errp = au_drinfo_do_store(a->w, a->elm);
++}
+
-+ parent = sb->s_root;
-+ if (dir_ino != AUFS_ROOT_INO) {
-+ parent = decode_by_ino(sb, dir_ino, 0);
-+ dentry = parent;
-+ if (!parent)
-+ goto out;
-+ if (IS_ERR(parent))
-+ goto out;
-+ AuDebugOn(au_test_anon(parent));
-+ } else
-+ dget(parent);
++static int au_drinfo_store_sio(struct au_drinfo_store *w,
++ struct au_drinfo_rev_elm *elm)
++{
++ int err, wkq_err;
+
-+ path.dentry = parent;
-+ path.mnt = au_mnt_get(sb);
-+ dentry = au_lkup_by_ino(&path, ino, nsi_lock);
-+ path_put(&path);
++ 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);
+
-+ out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++static int au_drinfo_store_work_init(struct au_drinfo_store *w,
++ aufs_bindex_t btgt)
++{
++ int err;
+
-+static int h_acceptable(void *expv, struct dentry *dentry)
++ 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)
+{
-+ return 1;
++ au_kfree_rcu(w->fdata);
+}
+
-+static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath,
-+ char *buf, int len, struct super_block *sb)
++static void au_drinfo_store_rev(struct au_drinfo_rev *rev,
++ struct au_drinfo_store *w)
+{
-+ char *p;
-+ int n;
-+ struct path path;
++ struct au_drinfo_rev_elm *elm;
++ struct inode *h_dir, *delegated;
++ int err, nelm;
++ struct path infopath = {
++ .mnt = w->h_ppath.mnt
++ };
+
-+ AuTraceEnter();
++ h_dir = d_inode(w->h_ppath.dentry);
++ IMustLock(h_dir);
+
-+ p = d_path(h_rootpath, buf, len);
-+ if (IS_ERR(p))
-+ goto out;
-+ n = strlen(p);
++ 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);
++ au_kfree_rcu(elm->info_last);
++ }
++ if (unlikely(err))
++ AuIOErr("%d, %s\n", err, w->whname);
++ /* go on even if err */
++ }
++}
+
-+ path.mnt = h_rootpath->mnt;
-+ path.dentry = h_parent;
-+ p = d_path(&path, buf, len);
-+ if (IS_ERR(p))
-+ goto out;
-+ LKTRTrace("%s\n", p);
-+ if (n != 1)
-+ p += n;
-+ LKTRTrace("%p, %s, %ld\n",
-+ p, p, (long)(p - buf));
++/* 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;
+
-+ path.mnt = au_mnt_get(sb);
-+ path.dentry = sb->s_root;
-+ p = d_path(&path, buf, len - strlen(p));
-+ mntput(path.mnt);
-+ if (IS_ERR(p))
++ err = au_drinfo_store_work_init(&work, btgt);
++ AuTraceErr(err);
++ if (unlikely(err))
+ goto out;
-+ if (n != 1)
-+ p[strlen(p)] = '/';
-+ LKTRTrace("%s\n", p);
+
-+ out:
-+ AuTraceErrPtr(p);
-+ return p;
++ 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);
++ au_kfree_try_rcu(rev);
++ *p = NULL;
++ }
++ au_hn_inode_unlock(hdir);
++
++out_args:
++ au_drinfo_store_work_fin(&work);
++out:
++ return err;
+}
+
-+static noinline_for_stack
-+struct dentry *decode_by_path(struct super_block *sb, aufs_bindex_t bindex,
-+ ino_t ino, __u32 *fh, int fh_len,
-+ struct au_nfsd_si_lock *nsi_lock)
++/* ---------------------------------------------------------------------- */
++
++int au_dr_rename(struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev)
+{
-+ struct dentry *dentry, *h_parent, *root;
-+ struct super_block *h_sb;
-+ char *pathname, *p;
-+ struct vfsmount *h_mnt;
++ int err, already;
++ ino_t ino;
++ struct super_block *sb;
+ struct au_branch *br;
-+ int err;
-+ struct nameidata nd;
++ struct au_dr_br *dr;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct au_dr_hino *ent;
++ struct au_drinfo_rev *rev, **p;
+
-+ LKTRTrace("b%d\n", bindex);
-+ SiMustAnyLock(sb);
++ AuDbg("bindex %d\n", bindex);
+
-+ br = au_sbr(sb, bindex);
-+ /* au_br_get(br); */
-+ h_mnt = br->br_mnt;
-+ h_sb = h_mnt->mnt_sb;
-+ LKTRTrace("%s, h_decode_fh\n", au_sbtype(h_sb));
-+ /* in linux-2.6.24, it takes struct fid * as file handle */
-+ /* todo: call lower fh_to_dentry()? fh_to_parent()? */
-+ h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail),
-+ fh_len - Fh_tail, fh[Fh_h_type],
-+ h_acceptable, /*context*/NULL);
-+ dentry = h_parent;
-+ if (unlikely(!h_parent || IS_ERR(h_parent))) {
-+ AuWarn1("%s decode_fh failed, %ld\n",
-+ au_sbtype(h_sb), PTR_ERR(h_parent));
++ err = -ENOMEM;
++ ent = kmalloc(sizeof(*ent), GFP_NOFS);
++ if (unlikely(!ent))
+ goto out;
-+ }
-+ dentry = NULL;
-+ if (unlikely(au_test_anon(h_parent))) {
-+ AuWarn1("%s decode_fh returned a disconnected dentry\n",
-+ au_sbtype(h_sb));
-+ goto out_h_parent;
++
++ 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 */
+ }
+
-+ dentry = ERR_PTR(-ENOMEM);
-+ pathname = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!pathname))
-+ goto out_h_parent;
++ /* revert */
++ if (!already)
++ au_dr_hino_del(dr, ent);
++ au_kfree_rcu(ent);
+
-+ root = sb->s_root;
-+ nd.path.mnt = h_mnt;
-+ di_read_lock_parent(root, !AuLock_IR);
-+ nd.path.dentry = au_h_dptr(root, bindex);
-+ di_read_unlock(root, !AuLock_IR);
-+ p = au_build_path(h_parent, &nd.path, pathname, PAGE_SIZE, sb);
-+ dentry = (void *)p;
-+ if (IS_ERR(p))
-+ goto out_pathname;
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ LKTRTrace("%s\n", p);
-+ si_read_unlock(sb);
-+ err = vfsub_path_lookup(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &nd);
-+ dentry = ERR_PTR(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);
++ au_kfree_rcu(elm->info_last);
++ }
++ au_kfree_try_rcu(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_relock;
++ goto out;
+
-+ dentry = ERR_PTR(-ENOENT);
-+ AuDebugOn(au_test_anon(nd.path.dentry));
-+ if (unlikely(!nd.path.dentry->d_inode))
-+ goto out_nd;
++ 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;
+
-+ if (ino != nd.path.dentry->d_inode->i_ino)
-+ dentry = au_lkup_by_ino(&nd.path, ino, /*nsi_lock*/NULL);
-+ else
-+ dentry = dget(nd.path.dentry);
++ 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);
++ au_kfree_rcu(ent);
+
-+ out_nd:
-+ path_put(&nd.path);
-+ out_relock:
-+ if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0))
-+ if (!IS_ERR(dentry)) {
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ }
-+ out_pathname:
-+ free_page((unsigned long)pathname);
-+ out_h_parent:
-+ dput(h_parent);
-+ out:
-+ /* au_br_put(br); */
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++out:
++ au_kfree_try_rcu(rev);
++ if (unlikely(err))
++ pr_err("failed to remove dirren info\n");
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *
-+aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len,
-+ int fh_type)
++static struct au_drinfo *au_drinfo_do_load(struct path *h_ppath,
++ char *whname, int whnamelen,
++ struct dentry **info_dentry)
+{
-+ struct dentry *dentry;
-+ __u32 *fh = fid->raw;
-+ ino_t ino, dir_ino;
-+ aufs_bindex_t bindex;
-+ struct au_nfsd_si_lock nsi_lock = {
-+ .sigen = fh[Fh_sigen],
-+ .br_id = fh[Fh_br_id],
-+ .force_lock = 0
-+ };
++ 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);
++ inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
++ infopath.dentry = vfsub_lookup_one_len(whname, h_ppath, whnamelen);
++ if (IS_ERR(infopath.dentry)) {
++ drinfo = (void *)infopath.dentry;
++ goto out;
++ }
+
-+ LKTRTrace("%d, fh{br_id %u, sigen %u, i%u, diri%u, g%u}\n",
-+ fh_type, fh[Fh_br_id], fh[Fh_sigen], fh[Fh_ino],
-+ fh[Fh_dir_ino], fh[Fh_igen]);
-+ AuDebugOn(fh_len < Fh_tail);
++ if (d_is_negative(infopath.dentry))
++ goto out_dput; /* success */
+
-+ dentry = ERR_PTR(-ESTALE);
-+ /* branch id may be wrapped around */
-+ bindex = si_nfsd_read_lock(sb, &nsi_lock);
-+ if (unlikely(bindex < 0))
-+ goto out;
-+ nsi_lock.force_lock = 1;
++ 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;
++ }
+
-+ /* is this inode still cached? */
-+ ino = decode_ino(fh + Fh_ino);
-+ AuDebugOn(ino == AUFS_ROOT_INO);
-+ dir_ino = decode_ino(fh + Fh_dir_ino);
-+ dentry = decode_by_ino(sb, ino, dir_ino);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (dentry)
-+ goto accept;
++ drinfo = au_drinfo_read_k(f, /*h_ino*/0);
++ if (IS_ERR_OR_NULL(drinfo))
++ goto out_fput;
+
-+ /* is the parent dir cached? */
-+ dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (dentry)
-+ goto accept;
++ AuDbg("oldname %.*s\n", drinfo->oldnamelen, drinfo->oldname);
++ *info_dentry = dget(infopath.dentry); /* keep it alive */
+
-+ /* lookup path */
-+ dentry = decode_by_path(sb, bindex, ino, fh, fh_len, &nsi_lock);
-+ if (IS_ERR(dentry))
-+ goto out_unlock;
-+ if (unlikely(!dentry))
-+ /* todo?: make it ESTALE */
-+ goto out_unlock;
++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;
++};
+
-+ accept:
-+ LKTRLabel(accept);
-+ if (dentry->d_inode->i_generation == fh[Fh_igen])
-+ goto out_unlock; /* success */
++static void au_call_drinfo_do_load(void *args)
++{
++ struct au_drinfo_do_load_args *a = args;
+
-+ LKTRLabel(stale);
-+ dput(dentry);
-+ dentry = ERR_PTR(-ESTALE);
-+ out_unlock:
-+ LKTRLabel(out_unlock);
-+ si_read_unlock(sb);
-+ out:
-+ LKTRLabel(out);
-+ if (0 && IS_ERR(dentry))
-+ dentry = ERR_PTR(-ESTALE);
-+ AuTraceErrPtr(dentry);
-+ return dentry;
++ *a->drinfop = au_drinfo_do_load(a->h_ppath, a->whname, a->whnamelen,
++ a->info_dentry);
+}
+
-+#if 0 /* reserved for future use */
-+/* support subtreecheck option */
-+static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid,
-+ int fh_len, int fh_type)
-+{
-+ struct dentry *parent;
-+ __u32 *fh = fid->raw;
-+ ino_t dir_ino;
++struct au_drinfo_load {
++ struct path h_ppath;
++ struct qstr *qname;
++ unsigned char no_sio;
+
-+ dir_ino = decode_ino(fh + Fh_dir_ino);
-+ parent = decode_by_ino(sb, dir_ino, 0);
-+ if (IS_ERR(parent))
++ 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;
-+ if (!parent)
-+ parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]),
-+ dir_ino, fh, fh_len);
+
-+ out:
-+ AuTraceErrPtr(parent);
-+ return parent;
++ err = 0;
++ oldname.len = drinfo->oldnamelen;
++ oldname.name = drinfo->oldname;
++ if (au_qstreq(w->qname, &oldname)) {
++ /* the name is renamed back */
++ au_kfree_rcu(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);
++ }
++ au_kfree_rcu(w->drinfo[bindex]);
++ w->drinfo[bindex] = drinfo;
++ dput(info_dentry);
++
++out:
++ AuTraceErr(err);
++ return err;
+}
-+#endif
+
+/* ---------------------------------------------------------------------- */
+
-+static int aufs_encode_fh(struct dentry *dentry, __u32 *fh, int *max_len,
-+ int connectable)
++static void au_dr_lkup_free(struct au_drinfo **drinfo, int n)
+{
-+ int err;
-+ aufs_bindex_t bindex, bend;
-+ struct super_block *sb, *h_sb;
-+ struct inode *inode;
-+ struct dentry *parent, *h_parent;
-+ struct au_branch *br;
++ struct au_drinfo **p = drinfo;
+
-+ LKTRTrace("%.*s, max %d, conn %d\n",
-+ AuDLNPair(dentry), *max_len, connectable);
-+ AuDebugOn(au_test_anon(dentry));
++ while (n-- > 0)
++ au_kfree_rcu(*drinfo++);
++ au_kfree_try_rcu(p);
++}
+
-+ parent = NULL;
-+ err = -ENOSPC;
-+ if (unlikely(*max_len <= Fh_tail)) {
-+ AuWarn1("NFSv2 client (max_len %d)?\n", *max_len);
-+ goto out;
-+ }
++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;
+
-+ err = FILEID_ROOT;
-+ if (IS_ROOT(dentry)) {
-+ AuDebugOn(dentry->d_inode->i_ino != AUFS_ROOT_INO);
-+ goto out;
-+ }
++ AuDbg("%.*s, name %.*s, whname %.*s, b%d\n",
++ AuLNPair(&dentry->d_name), AuLNPair(&lkup->dirren.dr_name),
++ AuLNPair(&lkup->whname), btgt);
+
-+ err = -EIO;
-+ h_parent = NULL;
+ sb = dentry->d_sb;
-+ aufs_read_lock(dentry, AuLock_FLUSH | AuLock_IR);
-+ parent = dget_parent(dentry);
-+ di_read_lock_parent(parent, !AuLock_IR);
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode);
-+#ifdef CONFIG_AUFS_DEBUG
-+ {
-+ unsigned int mnt_flags = au_mntflags(sb);
-+
-+ if (unlikely(!au_opt_test_xino(mnt_flags)))
-+ AuWarn1("NFS-exporting requires xino\n");
-+ if (unlikely(0 && !au_opt_test(mnt_flags, UDBA_INOTIFY)))
-+ AuWarn1("udba=inotify is recommended "
-+ "for NFS-exporting\n");
-+ }
-+#endif
-+
-+ bend = au_dbtaildir(parent);
-+ for (bindex = au_dbstart(parent); bindex <= bend; bindex++) {
-+ h_parent = au_h_dptr(parent, bindex);
-+ if (h_parent) {
-+ dget(h_parent);
-+ break;
++ 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;
+ }
-+ if (unlikely(!h_parent))
-+ goto out_unlock;
-+ LKTRTrace("b%d\n", bindex);
++ w.drinfo = lkup->dirren.drinfo;
++ w.no_sio = !!uid_eq(current_fsuid(), GLOBAL_ROOT_UID);
++ w.h_ppath.dentry = au_h_dptr(dentry, btgt);
++ AuDebugOn(!w.h_ppath.dentry);
++ w.h_ppath.mnt = au_sbr_mnt(sb, btgt);
++ w.qname = &dentry->d_name;
+
-+ err = -EPERM;
-+ br = au_sbr(sb, bindex);
-+ h_sb = br->br_mnt->mnt_sb;
-+ if (unlikely(!h_sb->s_export_op)) {
-+ AuErr1("%s branch is not exportable\n", au_sbtype(h_sb));
-+ goto out_dput;
++ 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);
++ au_kfree_rcu(ent);
+ }
++ goto out; /* success */
+
-+ fh[Fh_br_id] = br->br_id;
-+ fh[Fh_sigen] = au_sigen(sb);
-+ encode_ino(fh + Fh_ino, inode->i_ino);
-+ encode_ino(fh + Fh_dir_ino, parent->d_inode->i_ino);
-+ fh[Fh_igen] = inode->i_generation;
++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;
++}
+
-+ *max_len -= Fh_tail;
-+ /* in linux-2.6.24, it takes struct fid * as file handle */
-+ fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail),
-+ max_len,
-+ /*connectable or subtreecheck*/0);
-+ err = fh[Fh_h_type];
-+ *max_len += Fh_tail;
-+ /* todo: macros? */
-+ if (err != 255)
-+ err = 99;
-+ else
-+ AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb));
++void au_dr_lkup_fin(struct au_do_lookup_args *lkup)
++{
++ au_dr_lkup_free(lkup->dirren.drinfo, lkup->dirren.ninfo);
++}
+
-+ out_dput:
-+ dput(h_parent);
-+ out_unlock:
-+ di_read_unlock(parent, !AuLock_IR);
-+ dput(parent);
-+ aufs_read_unlock(dentry, AuLock_IR);
-+ out:
++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);
++ drinfo = lkup->dirren.drinfo[btgt];
++ if (!drinfo)
++ goto out;
++
++ au_kfree_try_rcu(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);
-+ if (unlikely(err < 0))
-+ err = 255;
+ 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);
++ drinfo = lkup->dirren.drinfo[bindex];
++ if (!drinfo)
++ goto out;
++
++ match = (drinfo->ino == h_ino);
++ AuDbg("match %d\n", match);
++
++out:
++ return match;
++}
++
+/* ---------------------------------------------------------------------- */
+
-+static struct export_operations aufs_export_op = {
-+ .fh_to_dentry = aufs_fh_to_dentry,
-+ //.fh_to_parent = aufs_fh_to_parent,
-+ .encode_fh = aufs_encode_fh
-+};
++int au_dr_opt_set(struct super_block *sb)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
+
-+void au_export_init(struct super_block *sb)
++ 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)
+{
-+ struct au_sbinfo *sbinfo;
-+ __u32 u;
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
++ 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);
++ }
+
-+ sb->s_export_op = &aufs_export_op;
-+ sbinfo = au_sbi(sb);
-+ sbinfo->si_xigen = NULL;
-+ get_random_bytes(&u, sizeof(u));
-+ BUILD_BUG_ON(sizeof(u) != sizeof(int));
-+ atomic_set(&sbinfo->si_xigen_next, u);
-+ memset(&sbinfo->si_xinodir, 0, sizeof(struct path));
++ 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 linux/fs/aufs/file.c linux-aufs/fs/aufs/file.c
---- linux/fs/aufs/file.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/file.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,777 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,140 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2017-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * handling file/dir, and address_space operation
-+ *
-+ * $Id$
++ * renamed dir info
+ */
+
-+#include <linux/pagemap.h>
-+#include "aufs.h"
-+
-+/*
-+ * a dirty trick for handling FMODE_EXEC and deny_write_access().
-+ * because FMODE_EXEC flag is not passed to f_op->open(),
-+ * set it to file->private_data temporary.
-+ */
-+#if !defined(CONFIG_AUFS_MODULE) || defined(CONFIG_AUFS_DENY_WRITE_ACCESS_PATCH)
-+int au_store_fmode_exec(struct nameidata *nd, struct inode *inode)
-+{
-+ int err;
-+ union {
-+ void *p;
-+ unsigned long ul;
-+ } u;
++#ifndef __AUFS_DIRREN_H__
++#define __AUFS_DIRREN_H__
+
-+ err = 0;
-+ if (nd
-+ && (nd->flags & LOOKUP_OPEN)
-+ && nd->intent.open.file
-+ && (nd->intent.open.flags & FMODE_EXEC)
-+ && inode
-+ && S_ISREG(inode->i_mode)) {
-+ u.ul = nd->intent.open.flags;
-+ nd->intent.open.file->private_data = u.p;
-+ /* smp_mb(); */
-+ err = 1;
-+ }
++#ifdef __KERNEL__
+
-+ return err;
-+}
-+#endif
++#include <linux/dcache.h>
++#include <linux/statfs.h>
++#include <linux/uuid.h>
++#include "hbl.h"
+
-+/* drop flags for writing */
-+unsigned int au_file_roflags(unsigned int flags)
-+{
-+ flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC);
-+ flags |= O_RDONLY | O_NOATIME;
-+ return flags;
-+}
++#define AuDirren_NHASH 100
+
-+/* common functions to regular file and dir */
-+struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
-+ struct file *file)
-+{
-+ struct file *h_file;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ struct super_block *sb;
-+ struct au_branch *br;
-+ int err;
++#ifdef CONFIG_AUFS_DIRREN
++enum au_brid_type {
++ AuBrid_Unset,
++ AuBrid_UUID,
++ AuBrid_FSID,
++ AuBrid_DEV
++};
+
-+ LKTRTrace("%.*s, b%d, flags 0%o, f %d\n",
-+ AuDLNPair(dentry), bindex, flags, !!file);
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ AuDebugOn(!h_dentry);
-+ h_inode = h_dentry->d_inode;
++struct au_dr_brid {
++ enum au_brid_type type;
++ union {
++ uuid_t uuid; /* unimplemented yet */
++ fsid_t fsid;
++ dev_t dev;
++ };
++};
+
-+ /* a race condition can happen between open and unlink/rmdir */
-+ h_file = ERR_PTR(-ENOENT);
-+ if (unlikely((!d_unhashed(dentry) && d_unhashed(h_dentry))
-+ || !h_inode))
-+ goto out;
++/* 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)
+
-+ sb = dentry->d_sb;
-+ br = au_sbr(sb, bindex);
-+ au_br_get(br);
-+ /* drop flags for writing */
-+ if (au_test_ro(sb, bindex, dentry->d_inode))
-+ flags = au_file_roflags(flags);
-+ flags &= ~O_CREAT;
++struct au_dr_hino {
++ struct hlist_bl_node dr_hnode;
++ ino_t dr_h_ino;
++};
+
-+ h_file = NULL;
-+ if (file && au_test_nfs(h_dentry->d_sb))
-+ h_file = au_h_intent(dentry, bindex, file);
-+ if (!h_file)
-+ h_file = dentry_open(dget(h_dentry), mntget(br->br_mnt), flags);
++struct au_dr_br {
++ struct hlist_bl_head dr_h_ino[AuDirren_NHASH];
++ struct au_dr_brid dr_brid;
++};
+
-+ /*
-+ * a dirty trick for handling FMODE_EXEC and deny_write_access().
-+ */
-+ if (file && (file->f_mode & FMODE_EXEC)) {
-+ h_file->f_mode |= FMODE_EXEC;
-+ smp_mb(); /* flush f_mode */
-+ err = au_deny_write_access(h_file);
-+ if (unlikely(err)) {
-+ fput(h_file);
-+ h_file = ERR_PTR(err);
-+ }
-+ }
-+ if (IS_ERR(h_file))
-+ au_br_put(br);
++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
+
-+out:
-+ AuTraceErrPtr(h_file);
-+ return h_file;
-+}
++/* ---------------------------------------------------------------------- */
+
-+static int do_coo(struct dentry *dentry, aufs_bindex_t bstart)
-+{
-+ int err;
-+ aufs_bindex_t bcpup;
-+ unsigned char pin_flags;
-+ struct au_pin pin;
-+ struct dentry *parent;
-+ struct mutex *h_mtx;
-+ struct super_block *sb;
++struct au_branch;
++struct au_do_lookup_args;
++struct au_hinode;
++#ifdef CONFIG_AUFS_DIRREN
++int au_dr_hino_test_add(struct au_dr_br *dr, ino_t h_ino,
++ struct au_dr_hino *add_ent);
++void au_dr_hino_free(struct au_dr_br *dr);
++int au_dr_br_init(struct super_block *sb, struct au_branch *br,
++ const struct path *path);
++int au_dr_br_fin(struct super_block *sb, struct au_branch *br);
++int au_dr_rename(struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev);
++void au_dr_rename_fin(struct dentry *src, aufs_bindex_t btgt, void *rev);
++void au_dr_rename_rev(struct dentry *src, aufs_bindex_t bindex, void *rev);
++int au_dr_lkup(struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t bindex);
++int au_dr_lkup_name(struct au_do_lookup_args *lkup, aufs_bindex_t btgt);
++int au_dr_lkup_h_ino(struct au_do_lookup_args *lkup, aufs_bindex_t bindex,
++ ino_t h_ino);
++void au_dr_lkup_fin(struct au_do_lookup_args *lkup);
++int au_dr_opt_set(struct super_block *sb);
++int au_dr_opt_flush(struct super_block *sb);
++int au_dr_opt_clr(struct super_block *sb, int no_flush);
++#else
++AuStubInt0(au_dr_hino_test_add, struct au_dr_br *dr, ino_t h_ino,
++ struct au_dr_hino *add_ent);
++AuStubVoid(au_dr_hino_free, struct au_dr_br *dr);
++AuStubInt0(au_dr_br_init, struct super_block *sb, struct au_branch *br,
++ const struct path *path);
++AuStubInt0(au_dr_br_fin, struct super_block *sb, struct au_branch *br);
++AuStubInt0(au_dr_rename, struct dentry *src, aufs_bindex_t bindex,
++ struct qstr *dst_name, void *_rev);
++AuStubVoid(au_dr_rename_fin, struct dentry *src, aufs_bindex_t btgt, void *rev);
++AuStubVoid(au_dr_rename_rev, struct dentry *src, aufs_bindex_t bindex,
++ void *rev);
++AuStubInt0(au_dr_lkup, struct au_do_lookup_args *lkup, struct dentry *dentry,
++ aufs_bindex_t bindex);
++AuStubInt0(au_dr_lkup_name, struct au_do_lookup_args *lkup, aufs_bindex_t btgt);
++AuStubInt0(au_dr_lkup_h_ino, struct au_do_lookup_args *lkup,
++ aufs_bindex_t bindex, ino_t h_ino);
++AuStubVoid(au_dr_lkup_fin, struct au_do_lookup_args *lkup);
++AuStubInt0(au_dr_opt_set, struct super_block *sb);
++AuStubInt0(au_dr_opt_flush, struct super_block *sb);
++AuStubInt0(au_dr_opt_clr, struct super_block *sb, int no_flush);
++#endif
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ AuDebugOn(IS_ROOT(dentry));
-+ DiMustWriteLock(dentry);
++/* ---------------------------------------------------------------------- */
+
-+ parent = dget_parent(dentry);
-+ di_write_lock_parent(parent);
-+ sb = dentry->d_sb;
-+ err = AuWbrCopyup(au_sbi(sb), dentry);
-+ bcpup = err;
-+ if (err < 0) {
-+ err = 0; /* stop copyup, it is not an error */
-+ goto out_dgrade;
-+ }
-+ err = 0;
++#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
+
-+ if (!au_h_dptr(parent, bcpup)) {
-+ err = au_cpup_dirs(dentry, bcpup);
-+ if (unlikely(err))
-+ goto out_dgrade;
-+ }
++#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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,366 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2010-2022 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/>.
++ */
+
-+ di_downgrade_lock(parent, AuLock_IR);
-+ pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (au_opt_test(au_mntflags(sb), UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ err = au_pin(&pin, dentry, bcpup, pin_flags);
-+ if (unlikely(err))
-+ goto out;
-+ h_mtx = &au_h_dptr(dentry, bstart)->d_inode->i_mutex;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ AuDebugOn(au_h_dptr(dentry, bcpup));
-+ err = au_sio_cpup_simple(dentry, bcpup, -1, AuCpup_DTIME);
-+ AuTraceErr(err);
-+ mutex_unlock(h_mtx);
-+ au_unpin(&pin);
-+ goto out;
++/*
++ * dynamically customizable operations for regular files
++ */
+
-+ out_dgrade:
-+ di_downgrade_lock(parent, AuLock_IR);
-+ out:
-+ di_read_unlock(parent, AuLock_IR);
-+ dput(parent);
-+ AuTraceErr(err);
-+ return err;
-+}
++#include "aufs.h"
+
-+int au_do_open(struct inode *inode, struct file *file,
-+ int (*open)(struct file *file, int flags))
-+{
-+ int err;
-+ struct dentry *dentry;
-+ struct super_block *sb;
-+ aufs_bindex_t bstart;
-+ unsigned char coo;
++#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];
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(dentry));
++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;
+
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ coo = 0;
-+ if (!(sb->s_flags & MS_RDONLY))
-+ switch (au_mntflags(sb) & AuOptMask_COO) {
-+ case AuOpt_COO_LEAF:
-+ coo = !S_ISDIR(inode->i_mode);
-+ break;
-+ case AuOpt_COO_ALL:
-+ coo = 1;
++ key = NULL;
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(tmp, pos, hbl, dk_hnode)
++ if (tmp->dk_op.dy_hop == h_op) {
++ if (kref_get_unless_zero(&tmp->dk_kref))
++ key = tmp;
+ break;
+ }
-+ err = au_finfo_init(file);
-+ if (unlikely(err))
-+ goto out;
++ hlist_bl_unlock(hbl);
+
-+ if (!coo)
-+ di_read_lock_child(dentry, AuLock_IR);
-+ else {
-+ di_write_lock_child(dentry);
-+ bstart = au_dbstart(dentry);
-+ if (au_test_ro(sb, bstart, dentry->d_inode)) {
-+ err = do_coo(dentry, bstart);
-+ if (err) {
-+ di_write_unlock(dentry);
-+ goto out_finfo;
++ 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;
+ }
-+ }
-+ di_downgrade_lock(dentry, AuLock_IR);
++ } 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 */
+ }
+
-+ err = open(file, file->f_flags);
-+ di_read_unlock(dentry, AuLock_IR);
++ return found;
++}
+
-+ out_finfo:
-+ fi_write_unlock(file);
-+ if (unlikely(err))
-+ au_finfo_fin(file);
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ return err;
++/* 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) {
++ if (kref_get_unless_zero(&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;
+}
+
-+int au_reopen_nondir(struct file *file)
++static void dy_free_rcu(struct rcu_head *rcu)
+{
-+ int err;
-+ aufs_bindex_t bstart, bindex, bend;
-+ struct dentry *dentry;
-+ struct file *h_file, *h_file_tmp;
++ struct au_dykey *key;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ bstart = au_dbstart(dentry);
-+ //bstart = au_ibstart(inode);
-+ AuDebugOn(S_ISDIR(dentry->d_inode->i_mode)
-+ || !au_h_dptr(dentry, bstart)->d_inode);
++ key = container_of(rcu, struct au_dykey, dk_rcu);
++ DyPrSym(key);
++ kfree(key);
++}
+
-+ h_file_tmp = NULL;
-+ if (au_fbstart(file) == bstart) {
-+ h_file = au_h_fptr(file, bstart);
-+ if (file->f_mode == h_file->f_mode)
-+ return 0; /* success */
-+ h_file_tmp = h_file;
-+ get_file(h_file_tmp);
-+ au_set_h_fptr(file, bstart, NULL);
-+ }
-+ AuDebugOn(au_fbstart(file) < bstart
-+ || au_fi(file)->fi_hfile[0 + bstart].hf_file);
++static void dy_free(struct kref *kref)
++{
++ struct au_dykey *key;
++ struct hlist_bl_head *hbl;
+
-+ h_file = au_h_open(dentry, bstart, file->f_flags & ~O_TRUNC, file);
-+ err = PTR_ERR(h_file);
-+ if (IS_ERR(h_file))
-+ goto out; /* todo: close all? */
-+ err = 0;
-+ /* cpup_file_flags(h_file, file); */
-+ au_set_fbstart(file, bstart);
-+ au_set_h_fptr(file, bstart, h_file);
-+ au_update_figen(file);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
++ 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);
++}
+
-+ /* close lower files */
-+ bend = au_fbend(file);
-+ for (bindex = bstart + 1; bindex <= bend; bindex++)
-+ au_set_h_fptr(file, bindex, NULL);
-+ au_set_fbend(file, bstart);
++void au_dy_put(struct au_dykey *key)
++{
++ kref_put(&key->dk_kref, dy_free);
++}
+
-+ out:
-+ if (h_file_tmp)
-+ fput(h_file_tmp);
-+ AuTraceErr(err);
-+ return err;
++/* ---------------------------------------------------------------------- */
++
++#define DyDbgSize(cnt, op) AuDebugOn(cnt != sizeof(op)/sizeof(void *))
++
++#ifdef CONFIG_AUFS_DEBUG
++#define DyDbgDeclare(cnt) unsigned int cnt = 0
++#define DyDbgInc(cnt) do { cnt++; } while (0)
++#else
++#define DyDbgDeclare(cnt) do {} while (0)
++#define DyDbgInc(cnt) do {} while (0)
++#endif
++
++#define DySet(func, dst, src, h_op, h_sb) do { \
++ DyDbgInc(cnt); \
++ if (h_op->func) { \
++ if (src.func) \
++ dst.func = src.func; \
++ else \
++ AuDbg("%s %s\n", au_sbtype(h_sb), #func); \
++ } \
++} while (0)
++
++#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(read_folio); /* force */
++ DySetAop(writepages);
++ DySetAop(dirty_folio);
++ DySetAop(invalidate_folio);
++ DySetAop(readahead);
++ DySetAop(write_begin);
++ DySetAop(write_end);
++ DySetAop(bmap);
++ DySetAop(release_folio);
++ DySetAop(free_folio);
++ /* this one will be changed according to an aufs mount option */
++ DySetAop(direct_IO);
++ DySetAop(migrate_folio);
++ DySetAop(launder_folio);
++ DySetAop(is_partially_uptodate);
++ DySetAop(is_dirty_writeback);
++ DySetAop(error_remove_page);
++ DySetAop(swap_activate);
++ DySetAop(swap_deactivate);
++ DySetAop(swap_rw);
++
++ DyDbgSize(cnt, *h_aop);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int au_ready_to_write_wh(struct file *file, loff_t len,
-+ aufs_bindex_t bcpup)
++static void dy_bug(struct kref *kref)
+{
-+ int err;
-+ aufs_bindex_t old_bstart;
-+ struct inode *inode;
-+ struct dentry *dentry, *hi_wh, *old_h_dentry;
-+ struct au_dinfo *dinfo;
-+ struct super_block *sb;
++ 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;
+
-+ AuTraceEnter();
++ hbl = dynop + op->dy_type;
++ key = dy_gfind_get(hbl, op->dy_hop);
++ if (key)
++ goto out_add; /* success */
+
-+ dentry = file->f_dentry;
-+ inode = dentry->d_inode;
-+ hi_wh = au_hi_wh(inode, bcpup);
-+ if (!hi_wh)
-+ err = au_sio_cpup_wh(dentry, bcpup, len, file);
-+ else {
-+ /* already copied-up after unlink */
-+ dinfo = au_di(dentry);
-+ old_bstart = dinfo->di_bstart;
-+ dinfo->di_bstart = bcpup;
-+ old_h_dentry = dinfo->di_hdentry[0 + bcpup].hd_dentry;
-+ dinfo->di_hdentry[0 + bcpup].hd_dentry = hi_wh;
-+ err = au_reopen_nondir(file);
-+ dinfo->di_hdentry[0 + bcpup].hd_dentry = old_h_dentry;
-+ dinfo->di_bstart = old_bstart;
++ p = a + op->dy_type;
++ key = kzalloc(p->sz, GFP_NOFS);
++ if (unlikely(!key)) {
++ key = ERR_PTR(-ENOMEM);
++ goto out;
+ }
+
-+ sb = dentry->d_sb;
-+ if (!err && inode->i_nlink > 1 && au_opt_test(au_mntflags(sb), PLINK))
-+ au_plink_append(sb, inode, au_h_dptr(dentry, bcpup), bcpup);
++ 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) {
++ au_kfree_rcu(key);
++ key = old;
++ }
+
-+ AuTraceErr(err);
-+ return err;
++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;
+}
+
++/* ---------------------------------------------------------------------- */
+/*
-+ * prepare the @file for writing.
++ * Aufs prohibits O_DIRECT by default 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.
+ */
-+int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin)
++static void dy_adx(struct au_dyaop *dyaop, int do_dx)
+{
-+ int err;
-+ unsigned char pin_flags;
-+ aufs_bindex_t bstart, bcpup;
-+ struct dentry *dentry, *parent, *h_dentry;
-+ struct inode *h_inode, *inode;
-+ struct super_block *sb;
++ if (!do_dx)
++ dyaop->da_op.direct_IO = NULL;
++ else
++ dyaop->da_op.direct_IO = aufs_aop.direct_IO;
++}
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, len %lld\n", AuDLNPair(dentry), len);
-+ FiMustWriteLock(file);
++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;
+
-+ sb = dentry->d_sb;
-+ bstart = au_fbstart(file);
-+ AuDebugOn(au_fbr(file, bstart) != au_sbr(sb, bstart));
++ 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;
+
-+ inode = dentry->d_inode;
-+ AuDebugOn(S_ISDIR(inode->i_mode));
-+ LKTRTrace("rdonly %d, bstart %d\n",
-+ au_test_ro(sb, bstart, inode), bstart);
++ AuDebugOn(!S_ISREG(h_inode->i_mode));
++ IiMustWriteLock(inode);
+
-+ err = au_test_ro(sb, bstart, inode);
-+ if (!err && (au_h_fptr(file, bstart)->f_mode & FMODE_WRITE)) {
-+ err = au_pin(pin, dentry, bstart, /*flags*/0);
++ 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;
-+ }
+
-+ /* need to cpup */
-+ parent = dget_parent(dentry);
-+ di_write_lock_parent(parent);
-+ err = AuWbrCopyup(au_sbi(sb), dentry);
-+ bcpup = err;
-+ if (unlikely(err < 0))
-+ goto out_dgrade;
+ err = 0;
++ inode->i_mapping->a_ops = &dyaop->da_op;
+
-+ if (!au_h_dptr(parent, bcpup)) {
-+ err = au_cpup_dirs(dentry, bcpup);
-+ if (unlikely(err))
-+ goto out_dgrade;
-+ }
++out:
++ return err;
++}
+
-+ pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (au_opt_test(au_mntflags(sb), UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ err = au_pin(pin, dentry, bcpup, pin_flags);
-+ if (unlikely(err))
-+ goto out_dgrade;
++/*
++ * 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;
+
-+ AuDebugOn(au_fbstart(file) != bstart);
-+ h_dentry = au_h_fptr(file, bstart)->f_dentry;
-+ h_inode = h_dentry->d_inode;
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ if (d_unhashed(dentry) /* || d_unhashed(h_dentry) */
-+ /* || !h_inode->i_nlink */) {
-+ err = au_ready_to_write_wh(file, len, bcpup);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ } else {
-+ di_downgrade_lock(parent, AuLock_IR);
-+ if (!au_h_dptr(dentry, bcpup))
-+ err = au_sio_cpup_simple(dentry, bcpup, len,
-+ AuCpup_DTIME);
-+ AuTraceErr(err);
-+ if (!err)
-+ err = au_reopen_nondir(file);
-+ AuTraceErr(err);
++ 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);
+ }
-+ mutex_unlock(&h_inode->i_mutex);
++ return err;
++}
+
-+ if (!err) {
-+ au_unpin_gp(pin);
-+ au_pin_set_parent_lflag(pin, /*lflag*/0);
-+ goto out_dput; /* success */
-+ }
-+ au_unpin(pin);
-+ goto out_unlock;
++void au_dy_arefresh(int do_dx)
++{
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_dykey *key;
+
-+ out_dgrade:
-+ di_downgrade_lock(parent, AuLock_IR);
-+ out_unlock:
-+ di_read_unlock(parent, AuLock_IR);
-+ out_dput:
-+ dput(parent);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ 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);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int au_file_refresh_by_inode(struct file *file, int *need_reopen)
++void __init au_dy_init(void)
+{
-+ int err;
-+ unsigned int mnt_flags;
-+ unsigned char pin_flags;
-+ aufs_bindex_t bstart, new_bstart, old_bstart;
-+ struct au_pin pin;
-+ struct au_finfo *finfo;
-+ struct dentry *dentry, *parent, *old_h_dentry, *hi_wh;
-+ struct inode *inode, *dir;
-+ struct super_block *sb;
-+ struct au_dinfo *dinfo;
++ int i;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ FiMustWriteLock(file);
++ for (i = 0; i < AuDyLast; i++)
++ INIT_HLIST_BL_HEAD(dynop + i);
++}
+
-+ err = 0;
-+ finfo = au_fi(file);
-+ inode = dentry->d_inode;
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (au_opt_test(mnt_flags, UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ again:
-+ bstart = au_ibstart(inode);
-+ if (bstart == finfo->fi_bstart)
-+ goto out;
++void au_dy_fin(void)
++{
++ int i;
+
-+ new_bstart = bstart;
-+ parent = dget_parent(dentry);
-+ dir = parent->d_inode;
-+ if (au_test_ro(sb, bstart, inode)) {
-+ di_read_lock_parent(parent, !AuLock_IR);
-+ err = AuWbrCopyup(au_sbi(sb), dentry);
-+ new_bstart = err;
-+ di_read_unlock(parent, !AuLock_IR);
-+ if (unlikely(err < 0))
-+ goto out_dput;
-+ err = 0;
-+ }
-+ /* someone else might change our inode while we were sleeping */
-+ /* todo: test more? */
-+ if (bstart != au_ibstart(inode)) {
-+ err = 0;
-+ dput(parent);
-+ goto again;
-+ }
-+ di_read_lock_parent(parent, AuLock_IR);
-+ bstart = new_bstart;
-+
-+ hi_wh = au_hi_wh(inode, bstart);
-+ if (au_opt_test(mnt_flags, PLINK)
-+ && au_plink_test(sb, inode)
-+ && !d_unhashed(dentry)) {
-+ err = au_test_and_cpup_dirs(dentry, bstart);
-+ if (unlikely(err))
-+ goto out_unlock;
-+
-+ /* always superio. */
-+#if 1
-+ err = au_pin(&pin, dentry, bstart, pin_flags);
-+ if (!err)
-+ err = au_sio_cpup_simple(dentry, bstart, -1,
-+ AuCpup_DTIME);
-+ au_unpin(&pin);
-+#else /* reserved for future use */
-+ if (!au_test_wkq(current)) {
-+ int wkq_err;
-+ struct cpup_pseudo_link_args args = {
-+ .errp = &err,
-+ .dentry = dentry,
-+ .bdst = bstart,
-+ .do_lock = 1
-+ };
-+ wkq_err = au_wkq_wait(call_cpup_pseudo_link, &args);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ } else
-+ err = cpup_pseudo_link(dentry, bstart, /*do_lock*/1);
-+#endif
-+ } else if (hi_wh) {
-+ /* already copied-up after unlink */
-+ dinfo = au_di(dentry);
-+ old_bstart = dinfo->di_bstart;
-+ dinfo->di_bstart = bstart;
-+ old_h_dentry = dinfo->di_hdentry[0 + bstart].hd_dentry;
-+ dinfo->di_hdentry[0 + bstart].hd_dentry = hi_wh;
-+ err = au_reopen_nondir(file);
-+ dinfo->di_hdentry[0 + bstart].hd_dentry = old_h_dentry;
-+ dinfo->di_bstart = old_bstart;
-+ *need_reopen = 0;
-+ }
-+
-+ out_unlock:
-+ di_read_unlock(parent, AuLock_IR);
-+ out_dput:
-+ dput(parent);
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+/*
-+ * after branch manipulating, refresh the file.
-+ */
-+static int refresh_file(struct file *file, int (*reopen)(struct file *file))
-+{
-+ int err, new_sz, need_reopen;
-+ struct dentry *dentry;
-+ aufs_bindex_t bend, bindex, brid;
-+ struct au_hfile *p;
-+ struct au_finfo *finfo;
-+ struct super_block *sb;
-+ struct inode *inode;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ FiMustWriteLock(file);
-+ DiMustAnyLock(dentry);
-+ inode = dentry->d_inode;
-+ IiMustAnyLock(inode);
-+
-+ err = -ENOMEM;
-+ sb = dentry->d_sb;
-+ finfo = au_fi(file);
-+ new_sz = sizeof(*finfo->fi_hfile) * (au_sbend(sb) + 1);
-+ p = au_kzrealloc(finfo->fi_hfile, sizeof(*p) * (finfo->fi_bend + 1),
-+ new_sz, GFP_NOFS);
-+ if (unlikely(!p))
-+ goto out;
-+ finfo->fi_hfile = p;
-+
-+ p += finfo->fi_bstart;
-+ brid = p->hf_br->br_id;
-+ bend = finfo->fi_bend;
-+ for (bindex = finfo->fi_bstart; bindex <= bend; bindex++, p++) {
-+ struct au_hfile tmp, *q;
-+ aufs_bindex_t new_bindex;
-+
-+ if (!p->hf_file)
-+ continue;
-+ new_bindex = au_find_bindex(sb, p->hf_br);
-+ if (new_bindex == bindex)
-+ continue;
-+ /* todo: test more? */
-+ if (new_bindex < 0) {
-+ au_set_h_fptr(file, bindex, NULL);
-+ continue;
-+ }
-+
-+ /* swap two hidden inode, and loop again */
-+ q = finfo->fi_hfile + new_bindex;
-+ tmp = *q;
-+ *q = *p;
-+ *p = tmp;
-+ if (tmp.hf_file) {
-+ bindex--;
-+ p--;
-+ }
-+ }
-+ {
-+ aufs_bindex_t s = finfo->fi_bstart, e = finfo->fi_bend;
-+ finfo->fi_bstart = 0;
-+ finfo->fi_bend = au_sbend(sb);
-+ finfo->fi_bstart = s;
-+ finfo->fi_bend = e;
-+ }
-+
-+ p = finfo->fi_hfile;
-+ if (!au_test_mmapped(file) && !d_unhashed(dentry)) {
-+ bend = au_sbend(sb);
-+ for (finfo->fi_bstart = 0; finfo->fi_bstart <= bend;
-+ finfo->fi_bstart++, p++)
-+ if (p->hf_file) {
-+ if (p->hf_file->f_dentry
-+ && p->hf_file->f_dentry->d_inode)
-+ break;
-+ else
-+ au_hfput(p);
-+ }
-+ } else {
-+ bend = au_br_index(sb, brid);
-+ for (finfo->fi_bstart = 0; finfo->fi_bstart < bend;
-+ finfo->fi_bstart++, p++)
-+ if (p->hf_file)
-+ au_hfput(p);
-+ bend = au_sbend(sb);
-+ }
-+
-+ p = finfo->fi_hfile + bend;
-+ for (finfo->fi_bend = bend; finfo->fi_bend >= finfo->fi_bstart;
-+ finfo->fi_bend--, p--)
-+ if (p->hf_file) {
-+ if (p->hf_file->f_dentry
-+ && p->hf_file->f_dentry->d_inode)
-+ break;
-+ else
-+ au_hfput(p);
-+ }
-+ AuDebugOn(finfo->fi_bend < finfo->fi_bstart);
-+
-+ err = 0;
-+ need_reopen = 1;
-+ if (!au_test_mmapped(file))
-+ err = au_file_refresh_by_inode(file, &need_reopen);
-+ if (!err && need_reopen && !d_unhashed(dentry))
-+ err = reopen(file);
-+ if (!err) {
-+ au_update_figen(file);
-+ return 0; /* success */
-+ }
-+
-+ /* error, close all hidden files */
-+ bend = au_fbend(file);
-+ for (bindex = au_fbstart(file); bindex <= bend; bindex++)
-+ au_set_h_fptr(file, bindex, NULL);
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+/* common function to regular file and dir */
-+int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
-+ int wlock, int locked)
-+{
-+ int err;
-+ struct dentry *dentry;
-+ struct super_block *sb;
-+ aufs_bindex_t bstart;
-+ unsigned char pseudo_link;
-+ au_gen_t sgen, fgen;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, w %d, l %d\n", AuDLNPair(dentry), wlock, locked);
-+ sb = dentry->d_sb;
-+ SiMustAnyLock(sb);
-+
-+ err = 0;
-+ sgen = au_sigen(sb);
-+ fi_write_lock(file);
-+ fgen = au_figen(file);
-+ di_write_lock_child(dentry);
-+ bstart = au_dbstart(dentry);
-+ pseudo_link = (bstart != au_ibstart(dentry->d_inode));
-+ if (sgen == fgen && !pseudo_link && au_fbstart(file) == bstart) {
-+ if (!wlock) {
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ fi_downgrade_lock(file);
-+ }
-+ goto out; /* success */
-+ }
-+
-+ LKTRTrace("sgen %d, fgen %d\n", sgen, fgen);
-+ if (sgen != au_digen(dentry)
-+ || sgen != au_iigen(dentry->d_inode)) {
-+ /*
-+ * d_path() and path_lookup() is a simple and good approach
-+ * to revalidate. but si_rwsem in DEBUG_RWSEM will cause a
-+ * deadlock. removed the code.
-+ */
-+ err = au_reval_dpath(dentry, sgen);
-+ if (unlikely(err < 0))
-+ goto out;
-+ AuDebugOn(au_digen(dentry) != sgen
-+ || au_iigen(dentry->d_inode) != sgen);
-+ }
-+
-+ err = refresh_file(file, reopen
-+ /* , au_opt_test(au_mnt_flags(sb), REFROF) */);
-+ if (!err) {
-+ if (!wlock) {
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ fi_downgrade_lock(file);
-+ }
-+ } else {
-+ di_write_unlock(dentry);
-+ fi_write_unlock(file);
-+ }
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/* cf. aufs_nopage() */
-+/* for madvise(2) */
-+static int aufs_readpage(struct file *file, struct page *page)
-+{
-+ AuTraceEnter();
-+ unlock_page(page);
-+ return 0;
++ for (i = 0; i < AuDyLast; i++)
++ WARN_ON(!hlist_bl_empty(dynop + i));
+}
-+
-+/* they will never be called. */
-+#ifdef CONFIG_AUFS_DEBUG
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 28)
-+static int aufs_prepare_write(struct file *file, struct page *page,
-+ unsigned from, unsigned to)
-+{ AuUnsupport(); return 0; }
-+static int aufs_commit_write(struct file *file, struct page *page,
-+ unsigned from, unsigned to)
-+{ AuUnsupport(); return 0; }
-+#endif
-+static int aufs_write_begin(struct file *file, struct address_space *mapping,
-+ loff_t pos, unsigned len, unsigned flags,
-+ struct page **pagep, void **fsdata)
-+{ AuUnsupport(); return 0; }
-+static int aufs_write_end(struct file *file, struct address_space *mapping,
-+ loff_t pos, unsigned len, unsigned copied,
-+ struct page *page, void *fsdata)
-+{ AuUnsupport(); return 0; }
-+static int aufs_writepage(struct page *page, struct writeback_control *wbc)
-+{ AuUnsupport(); return 0; }
-+static void aufs_sync_page(struct page *page)
-+{ AuUnsupport(); }
-+
-+static int aufs_set_page_dirty(struct page *page)
-+{ AuUnsupport(); return 0; }
-+static void aufs_invalidatepage(struct page *page, unsigned long offset)
-+{ AuUnsupport(); }
-+static int aufs_releasepage(struct page *page, gfp_t gfp)
-+{ AuUnsupport(); return 0; }
-+static ssize_t aufs_direct_IO(int rw, struct kiocb *iocb,
-+ const struct iovec *iov, loff_t offset,
-+ unsigned long nr_segs)
-+{ AuUnsupport(); return 0; }
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+static struct page *aufs_get_xip_page(struct address_space *mapping,
-+ sector_t offset, int create)
-+{ AuUnsupport(); return NULL; }
-+#endif
-+#endif /* CONFIG_AUFS_DEBUG */
-+
-+struct address_space_operations aufs_aop = {
-+ .readpage = aufs_readpage,
-+#ifdef CONFIG_AUFS_DEBUG
-+ .writepage = aufs_writepage,
-+ .sync_page = aufs_sync_page,
-+ .set_page_dirty = aufs_set_page_dirty,
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 28)
-+ .prepare_write = aufs_prepare_write,
-+ .commit_write = aufs_commit_write,
-+#endif
-+ .write_begin = aufs_write_begin,
-+ .write_end = aufs_write_end,
-+ .invalidatepage = aufs_invalidatepage,
-+ .releasepage = aufs_releasepage,
-+ .direct_IO = aufs_direct_IO,
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+ .get_xip_page = aufs_get_xip_page,
-+#endif
-+#endif /* CONFIG_AUFS_DEBUG */
-+};
-diff -urN linux/fs/aufs/file.h linux-aufs/fs/aufs/file.h
---- linux/fs/aufs/file.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/file.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,257 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,77 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2010-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * file operations
-+ *
-+ * $Id$
++ * dynamically customizable operations (for regular files only)
+ */
+
-+#ifndef __AUFS_FILE_H__
-+#define __AUFS_FILE_H__
++#ifndef __AUFS_DYNOP_H__
++#define __AUFS_DYNOP_H__
+
+#ifdef __KERNEL__
+
-+#include <linux/file.h>
+#include <linux/fs.h>
-+#include <linux/mm.h>
-+#include <linux/aufs_type.h>
-+#include "dentry.h"
-+#include "misc.h"
-+#include "super.h"
++#include <linux/kref.h>
+
-+/* ---------------------------------------------------------------------- */
++enum {AuDy_AOP, AuDyLast};
+
-+struct au_branch;
-+struct au_hfile {
-+ struct file *hf_file;
-+ struct au_branch *hf_br;
++struct au_dynop {
++ int dy_type;
++ union {
++ const void *dy_hop;
++ const struct address_space_operations *dy_haop;
++ };
+};
+
-+struct au_vdir;
-+struct au_finfo {
-+ atomic_t fi_generation;
-+
-+ struct au_rwsem fi_rwsem;
-+ struct au_hfile *fi_hfile;
-+ aufs_bindex_t fi_bstart, fi_bend;
-+
++struct au_dykey {
+ union {
-+ struct vm_operations_struct *fi_h_vm_ops;
-+ struct au_vdir *fi_vdir_cache;
++ struct hlist_bl_node dk_hnode;
++ struct rcu_head dk_rcu;
+ };
++ struct au_dynop dk_op;
++
++ /*
++ * during I am in the branch local array, kref is gotten. when the
++ * branch is removed, kref is put.
++ */
++ struct kref dk_kref;
++};
++
++/* stop unioning since their sizes are very different from each other */
++struct au_dyaop {
++ struct au_dykey da_key;
++ struct address_space_operations da_op; /* not const */
+};
++/* make sure that 'struct au_dykey *' can be any type */
++static_assert(!offsetof(struct au_dyaop, da_key));
+
+/* ---------------------------------------------------------------------- */
+
-+/* file.c */
-+extern struct address_space_operations aufs_aop;
-+unsigned int au_file_roflags(unsigned int flags);
-+struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
-+ struct file *file);
-+int au_do_open(struct inode *inode, struct file *file,
-+ int (*open)(struct file *file, int flags));
-+int au_reopen_nondir(struct file *file);
-+struct au_pin;
-+int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin);
-+int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
-+ int wlock, int locked);
++/* dynop.c */
++struct au_branch;
++void au_dy_put(struct au_dykey *key);
++int au_dy_iaop(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode);
++int au_dy_irefresh(struct inode *inode);
++void au_dy_arefresh(int do_dio);
+
-+/* f_op.c */
-+extern struct file_operations aufs_file_fop;
-+int aufs_flush(struct file *file, fl_owner_t id);
++void __init au_dy_init(void);
++void au_dy_fin(void);
+
-+/* finfo.c */
-+struct au_finfo *au_fi(struct file *file);
-+struct au_branch *au_fbr(struct file *file, aufs_bindex_t bindex);
-+struct file *au_h_fptr(struct file *file, aufs_bindex_t bindex);
++#endif /* __KERNEL__ */
++#endif /* __AUFS_DYNOP_H__ */
+diff -urN /usr/share/empty/fs/aufs/export.c linux/fs/aufs/export.c
+--- /usr/share/empty/fs/aufs/export.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/export.c 2022-12-17 09:21:34.796521861 +0100
+@@ -0,0 +1,830 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+void au_hfput(struct au_hfile *hf);
-+void au_set_h_fptr(struct file *file, aufs_bindex_t bindex,
-+ struct file *h_file);
++/*
++ * export via nfs
++ */
+
-+void au_finfo_fin(struct file *file);
-+int au_finfo_init(struct file *file);
++#include <linux/exportfs.h>
++#include <linux/fs_struct.h>
++#include <linux/nsproxy.h>
++#include <linux/random.h>
++#include <linux/writeback.h>
++#include "aufs.h"
+
-+#ifdef CONFIG_AUFS_ROBR
-+/* robr.c */
-+struct file *au_robr_safe_file(struct vm_area_struct *vma);
-+void au_robr_reset_file(struct vm_area_struct *vma, struct file *file);
++union conv {
++#ifdef CONFIG_AUFS_INO_T_64
++ __u32 a[2];
+#else
-+static inline struct file *au_robr_safe_file(struct vm_area_struct *vma)
++ __u32 a[1];
++#endif
++ ino_t ino;
++};
++
++static ino_t decode_ino(__u32 *a)
+{
-+ struct file *file;
++ union conv u;
+
-+ file = vma->vm_file;
-+ if (file->private_data && au_test_aufs(file->f_dentry->d_sb))
-+ return file;
-+ return NULL;
++ BUILD_BUG_ON(sizeof(u.ino) != sizeof(u.a));
++ u.a[0] = a[0];
++#ifdef CONFIG_AUFS_INO_T_64
++ u.a[1] = a[1];
++#endif
++ return u.ino;
+}
+
-+static inline
-+void au_robr_reset_file(struct vm_area_struct *vma, struct file *file)
++static void encode_ino(__u32 *a, ino_t ino)
+{
-+ vma->vm_file = file;
-+ /* smp_mb(); */ /* flush vm_file */
-+}
-+#endif /* CONFIG_AUFS_ROBR */
++ union conv u;
+
-+#if 0 /* reserved for future use */
-+/* ioctl.c */
-+long aufs_ioctl(struct file *file, unsigned int cmd, unsigned long arg);
-+long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg);
++ u.ino = ino;
++ a[0] = u.a[0];
++#ifdef CONFIG_AUFS_INO_T_64
++ a[1] = u.a[1];
+#endif
++}
+
-+/* ---------------------------------------------------------------------- */
++/* NFS file handle */
++enum {
++ Fh_br_id,
++ Fh_sigen,
++#ifdef CONFIG_AUFS_INO_T_64
++ /* support 64bit inode number */
++ Fh_ino1,
++ Fh_ino2,
++ Fh_dir_ino1,
++ Fh_dir_ino2,
++#else
++ Fh_ino1,
++ Fh_dir_ino1,
++#endif
++ Fh_igen,
++ Fh_h_type,
++ Fh_tail,
+
-+/* todo: memory barrier? */
-+static inline au_gen_t au_figen(struct file *f)
-+{
-+ return atomic_read(&au_fi(f)->fi_generation);
-+}
++ Fh_ino = Fh_ino1,
++ Fh_dir_ino = Fh_dir_ino1
++};
+
-+static inline int au_test_mmapped(struct file *f)
++static int au_test_anon(struct dentry *dentry)
+{
-+ return !!(au_fi(f)->fi_h_vm_ops);
++ /* note: read d_flags without d_lock */
++ return !!(dentry->d_flags & DCACHE_DISCONNECTED);
+}
+
-+static inline int au_test_aufs_file(struct file *f)
++int au_test_nfsd(void)
+{
-+ return !(f->f_dentry->d_inode->i_mode
-+ & (S_IFCHR | S_IFBLK | S_IFIFO | S_IFSOCK));
++ int ret;
++ struct task_struct *tsk = current;
++ char comm[sizeof(tsk->comm)];
++
++ ret = 0;
++ if (tsk->flags & PF_KTHREAD) {
++ get_task_comm(comm, tsk);
++ ret = !strcmp(comm, "nfsd");
++ }
++
++ return ret;
+}
+
+/* ---------------------------------------------------------------------- */
++/* inode generation external table */
+
-+#if !defined(CONFIG_AUFS_MODULE) || defined(CONFIG_AUFS_DENY_WRITE_ACCESS_PATCH)
-+int au_store_fmode_exec(struct nameidata *nd, struct inode *inode);
-+
-+static inline int au_deny_write_access(struct file *h_file)
++void au_xigen_inc(struct inode *inode)
+{
-+ LKTRTrace("%.*s\n", AuDLNPair(h_file->f_dentry));
-+ return deny_write_access(h_file);
-+}
++ loff_t pos;
++ ssize_t sz;
++ __u32 igen;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
+
-+static inline void au_allow_write_access(struct file *h_file)
-+{
-+ allow_write_access(h_file);
-+}
++ sb = inode->i_sb;
++ AuDebugOn(!au_opt_test(au_mntflags(sb), XINO));
+
-+#else
++ sbinfo = au_sbi(sb);
++ pos = inode->i_ino;
++ pos *= sizeof(igen);
++ igen = inode->i_generation + 1;
++ sz = xino_fwrite(sbinfo->si_xigen, &igen, sizeof(igen), &pos);
++ if (sz == sizeof(igen))
++ return; /* success */
+
-+static inline int au_store_fmode_exec(struct nameidata *nd, struct inode *inode)
-+{
-+ /* nothing */
-+ return 0;
++ if (unlikely(sz >= 0))
++ AuIOErr("xigen error (%zd)\n", sz);
+}
+
-+static inline int au_deny_write_access(struct file *h_file)
++int au_xigen_new(struct inode *inode)
+{
-+ /* nothing */
-+ return 0;
-+}
++ int err;
++ loff_t pos;
++ ssize_t sz;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
+
-+static inline void au_allow_write_access(struct file *h_file)
-+{
-+ /* nothing */
-+}
-+#endif /* CONFIG_AUFS_DENY_WRITE_ACCESS_PATCH */
++ err = 0;
++ /* todo: dirty, at mount time */
++ if (inode->i_ino == AUFS_ROOT_INO)
++ goto out;
++ sb = inode->i_sb;
++ SiMustAnyLock(sb);
++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ err = -EFBIG;
++ pos = inode->i_ino;
++ if (unlikely(au_loff_max / sizeof(inode->i_generation) - 1 < pos)) {
++ AuIOErr1("too large i%lld\n", pos);
++ goto out;
++ }
++ pos *= sizeof(inode->i_generation);
+
-+/*
-+ * fi_read_lock, fi_write_lock,
-+ * fi_read_unlock, fi_write_unlock, fi_downgrade_lock
-+ */
-+AuSimpleRwsemFuncs(fi, struct file *f, au_fi(f)->fi_rwsem);
++ err = 0;
++ sbinfo = au_sbi(sb);
++ file = sbinfo->si_xigen;
++ BUG_ON(!file);
+
-+/* to debug easier, do not make them inlined functions */
-+#define FiMustReadLock(f) do { \
-+ SiMustAnyLock((f)->f_dentry->d_sb); \
-+ AuRwMustReadLock(&au_fi(f)->fi_rwsem); \
-+} while (0)
++ if (vfsub_f_size_read(file)
++ < pos + sizeof(inode->i_generation)) {
++ inode->i_generation = atomic_inc_return(&sbinfo->si_xigen_next);
++ sz = xino_fwrite(file, &inode->i_generation,
++ sizeof(inode->i_generation), &pos);
++ } else
++ sz = xino_fread(file, &inode->i_generation,
++ sizeof(inode->i_generation), &pos);
++ if (sz == sizeof(inode->i_generation))
++ goto out; /* success */
+
-+#define FiMustWriteLock(f) do { \
-+ SiMustAnyLock((f)->f_dentry->d_sb); \
-+ AuRwMustWriteLock(&au_fi(f)->fi_rwsem); \
-+} while (0)
++ err = sz;
++ if (unlikely(sz >= 0)) {
++ err = -EIO;
++ AuIOErr("xigen error (%zd)\n", sz);
++ }
+
-+#define FiMustAnyLock(f) do { \
-+ SiMustAnyLock((f)->f_dentry->d_sb); \
-+ AuRwMustAnyLock(&au_fi(f)->fi_rwsem); \
-+} while (0)
++out:
++ return err;
++}
+
-+#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem)
++int au_xigen_set(struct super_block *sb, struct path *path)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
+
-+/* ---------------------------------------------------------------------- */
++ SiMustWriteLock(sb);
+
-+/* todo: hard/soft set? */
-+static inline aufs_bindex_t au_fbstart(struct file *file)
-+{
-+ FiMustAnyLock(file);
-+ return au_fi(file)->fi_bstart;
-+}
++ sbinfo = au_sbi(sb);
++ file = au_xino_create2(sb, path, sbinfo->si_xigen);
++ err = PTR_ERR(file);
++ if (IS_ERR(file))
++ goto out;
++ err = 0;
++ if (sbinfo->si_xigen)
++ fput(sbinfo->si_xigen);
++ sbinfo->si_xigen = file;
+
-+static inline aufs_bindex_t au_fbend(struct file *file)
-+{
-+ FiMustAnyLock(file);
-+ return au_fi(file)->fi_bend;
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline struct au_vdir *au_fvdir_cache(struct file *file)
++void au_xigen_clr(struct super_block *sb)
+{
-+#ifdef CONFIG_AUFS_EXPORT
-+ AuRwMustAnyLock(&au_fi(file)->fi_rwsem);
-+#else
-+ FiMustAnyLock(file);
-+#endif
-+ return au_fi(file)->fi_vdir_cache;
-+}
++ struct au_sbinfo *sbinfo;
+
-+static inline void au_set_fbstart(struct file *file, aufs_bindex_t bindex)
-+{
-+ FiMustWriteLock(file);
-+ AuDebugOn(au_sbend(file->f_dentry->d_sb) < bindex);
-+ au_fi(file)->fi_bstart = bindex;
-+}
++ SiMustWriteLock(sb);
+
-+static inline void au_set_fbend(struct file *file, aufs_bindex_t bindex)
-+{
-+ FiMustWriteLock(file);
-+ AuDebugOn(au_sbend(file->f_dentry->d_sb) < bindex
-+ || bindex < au_fbstart(file));
-+ au_fi(file)->fi_bend = bindex;
++ sbinfo = au_sbi(sb);
++ if (sbinfo->si_xigen) {
++ fput(sbinfo->si_xigen);
++ sbinfo->si_xigen = NULL;
++ }
+}
+
-+static inline void au_set_fvdir_cache(struct file *file,
-+ struct au_vdir *vdir_cache)
-+{
-+ FiMustWriteLock(file);
-+ AuDebugOn(!S_ISDIR(file->f_dentry->d_inode->i_mode)
-+ || (au_fi(file)->fi_vdir_cache && vdir_cache));
-+ au_fi(file)->fi_vdir_cache = vdir_cache;
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline void au_update_figen(struct file *file)
++static struct dentry *decode_by_ino(struct super_block *sb, ino_t ino,
++ ino_t dir_ino)
+{
-+ atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_dentry));
-+ /* smp_mb(); */ /* atomic_set */
-+}
++ struct dentry *dentry, *d;
++ struct inode *inode;
++ unsigned int sigen;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_FILE_H__ */
-diff -urN linux/fs/aufs/finfo.c linux-aufs/fs/aufs/finfo.c
---- linux/fs/aufs/finfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/finfo.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,186 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ dentry = NULL;
++ inode = ilookup(sb, ino);
++ if (!inode)
++ goto out;
+
-+/*
-+ * file private data
-+ *
-+ * $Id$
-+ */
++ dentry = ERR_PTR(-ESTALE);
++ sigen = au_sigen(sb);
++ if (unlikely(au_is_bad_inode(inode)
++ || IS_DEADDIR(inode)
++ || sigen != au_iigen(inode, NULL)))
++ goto out_iput;
+
-+#include "aufs.h"
++ dentry = NULL;
++ if (!dir_ino || S_ISDIR(inode->i_mode))
++ dentry = d_find_alias(inode);
++ else {
++ spin_lock(&inode->i_lock);
++ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
++ spin_lock(&d->d_lock);
++ if (!au_test_anon(d)
++ && d_inode(d->d_parent)->i_ino == dir_ino) {
++ dentry = dget_dlock(d);
++ spin_unlock(&d->d_lock);
++ break;
++ }
++ spin_unlock(&d->d_lock);
++ }
++ spin_unlock(&inode->i_lock);
++ }
++ if (unlikely(dentry && au_digen_test(dentry, sigen))) {
++ /* need to refresh */
++ dput(dentry);
++ dentry = NULL;
++ }
+
-+struct au_finfo *au_fi(struct file *file)
-+{
-+ struct au_finfo *finfo = file->private_data;
-+ AuDebugOn(!finfo
-+ || !finfo->fi_hfile
-+ || (0 < finfo->fi_bend
-+ && (/* au_sbi(file->f_dentry->d_sb)->si_bend
-+ < finfo->fi_bend
-+ || */ finfo->fi_bend < finfo->fi_bstart)));
-+ return finfo;
++out_iput:
++ iput(inode);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+struct au_branch *au_fbr(struct file *file, aufs_bindex_t bindex)
-+{
-+ struct au_finfo *finfo = au_fi(file);
-+ struct au_hfile *hf;
++/* ---------------------------------------------------------------------- */
+
-+ FiMustAnyLock(file);
-+ AuDebugOn(!finfo
-+ || finfo->fi_bstart < 0
-+ || bindex < finfo->fi_bstart
-+ || finfo->fi_bend < bindex);
-+ hf = finfo->fi_hfile + bindex;
-+ AuDebugOn(hf->hf_br && au_br_count(hf->hf_br) <= 0);
-+ return hf->hf_br;
-+}
++/* todo: dirty? */
++/* if exportfs_decode_fh() passed vfsmount*, we could be happy */
+
-+struct file *au_h_fptr(struct file *file, aufs_bindex_t bindex)
-+{
-+ struct au_finfo *finfo = au_fi(file);
-+ struct au_hfile *hf;
++struct au_compare_mnt_args {
++ /* input */
++ struct super_block *sb;
+
-+ FiMustAnyLock(file);
-+ AuDebugOn(!finfo
-+ || finfo->fi_bstart < 0
-+ || bindex < finfo->fi_bstart
-+ || finfo->fi_bend < bindex);
-+ hf = finfo->fi_hfile + bindex;
-+ AuDebugOn(hf->hf_file
-+ && file_count(hf->hf_file) <= 0
-+ && au_br_count(hf->hf_br) <= 0);
-+ return hf->hf_file;
-+}
++ /* output */
++ struct vfsmount *mnt;
++};
+
-+void au_hfput(struct au_hfile *hf)
++static int au_compare_mnt(struct vfsmount *mnt, void *arg)
+{
-+ if (hf->hf_file->f_mode & FMODE_EXEC)
-+ au_allow_write_access(hf->hf_file);
-+ fput(hf->hf_file);
-+ hf->hf_file = NULL;
-+ AuDebugOn(!hf->hf_br);
-+ au_br_put(hf->hf_br);
-+ hf->hf_br = NULL;
++ struct au_compare_mnt_args *a = arg;
++
++ if (mnt->mnt_sb != a->sb)
++ return 0;
++ a->mnt = mntget(mnt);
++ return 1;
+}
+
-+void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val)
++static struct vfsmount *au_mnt_get(struct super_block *sb)
+{
-+ struct au_finfo *finfo = au_fi(file);
-+ struct au_hfile *hf;
++ int err;
++ struct path root;
++ struct au_compare_mnt_args args = {
++ .sb = sb
++ };
+
-+ FiMustWriteLock(file);
-+ AuDebugOn(!finfo
-+ || finfo->fi_bstart < 0
-+ || bindex < finfo->fi_bstart
-+ || finfo->fi_bend < bindex);
-+ AuDebugOn(val && file_count(val) <= 0);
-+ hf = finfo->fi_hfile + bindex;
-+ AuDebugOn(val && hf->hf_file);
-+ if (hf->hf_file)
-+ au_hfput(hf);
-+ if (val) {
-+ hf->hf_file = val;
-+ hf->hf_br = au_sbr(file->f_dentry->d_sb, bindex);
-+ }
++ get_fs_root(current->fs, &root);
++ rcu_read_lock();
++ err = iterate_mounts(au_compare_mnt, &args, root.mnt);
++ rcu_read_unlock();
++ path_put(&root);
++ AuDebugOn(!err);
++ AuDebugOn(!args.mnt);
++ return args.mnt;
+}
+
-+void au_finfo_fin(struct file *file)
++struct au_nfsd_si_lock {
++ unsigned int sigen;
++ aufs_bindex_t bindex, br_id;
++ unsigned char force_lock;
++};
++
++static int si_nfsd_read_lock(struct super_block *sb,
++ struct au_nfsd_si_lock *nsi_lock)
+{
-+ struct au_finfo *finfo;
-+ struct dentry *dentry;
-+ aufs_bindex_t bindex, bend;
++ int err;
++ aufs_bindex_t bindex;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ SiMustAnyLock(dentry->d_sb);
++ si_read_lock(sb, AuLock_FLUSH);
+
-+ fi_write_lock(file);
-+ bend = au_fbend(file);
-+ bindex = au_fbstart(file);
-+ if (bindex >= 0)
-+ for (; bindex <= bend; bindex++)
-+ au_set_h_fptr(file, bindex, NULL);
++ /* branch id may be wrapped around */
++ err = 0;
++ bindex = au_br_index(sb, nsi_lock->br_id);
++ if (bindex >= 0 && nsi_lock->sigen + AUFS_BRANCH_MAX > au_sigen(sb))
++ goto out; /* success */
+
-+ finfo = au_fi(file);
-+#ifdef CONFIG_AUFS_DEBUG
-+ if (finfo->fi_bstart >= 0) {
-+ bend = au_fbend(file);
-+ for (bindex = finfo->fi_bstart; bindex <= bend; bindex++) {
-+ struct au_hfile *hf;
-+ hf = finfo->fi_hfile + bindex;
-+ AuDebugOn(hf->hf_file || hf->hf_br);
-+ }
-+ }
-+#endif
++ err = -ESTALE;
++ bindex = -1;
++ if (!nsi_lock->force_lock)
++ si_read_unlock(sb);
+
-+ kfree(finfo->fi_hfile);
-+ fi_write_unlock(file);
-+ au_rwsem_destroy(&finfo->fi_rwsem);
-+ au_cache_free_finfo(finfo);
++out:
++ nsi_lock->bindex = bindex;
++ return err;
+}
+
-+int au_finfo_init(struct file *file)
-+{
-+ struct au_finfo *finfo;
-+ struct dentry *dentry;
-+ union {
-+ void *p;
-+ unsigned long ul;
-+ } u;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ AuDebugOn(!dentry->d_inode);
-+
-+ finfo = au_cache_alloc_finfo();
-+ if (finfo) {
-+ finfo->fi_hfile = kcalloc(au_sbend(dentry->d_sb) + 1,
-+ sizeof(*finfo->fi_hfile), GFP_NOFS);
-+ if (finfo->fi_hfile) {
-+ au_rw_init_wlock(&finfo->fi_rwsem);
-+ //au_dbg_lock_fi_reg(file);
-+ finfo->fi_bstart = -1;
-+ finfo->fi_bend = -1;
-+ atomic_set(&finfo->fi_generation, au_digen(dentry));
-+ /* smp_mb(); */ /* atomic_set */
++struct find_name_by_ino {
++ struct dir_context ctx;
++ int called, found;
++ ino_t ino;
++ char *name;
++ int namelen;
++};
+
-+ /*
-+ * a dirty trick for handling FMODE_EXEC and
-+ * deny_write_access().
-+ * because FMODE_EXEC flag is not passed to
-+ * f_op->open(),
-+ * aufs set it to file->private_data temporary in lookup
-+ * or dentry revalidation operations.
-+ * restore the flag to f_mode here.
-+ */
-+ u.p = file->private_data;
-+ if (u.ul & FMODE_EXEC) {
-+ file->f_mode |= FMODE_EXEC;
-+ smp_mb(); /* flush f_mode */
-+ }
++static bool
++find_name_by_ino(struct dir_context *ctx, const char *name, int namelen,
++ loff_t offset, u64 ino, unsigned int d_type)
++{
++ struct find_name_by_ino *a = container_of(ctx, struct find_name_by_ino,
++ ctx);
+
-+ file->private_data = finfo;
-+ return 0; /* success */
-+ }
-+ au_cache_free_finfo(finfo);
-+ }
++ a->called++;
++ if (a->ino != ino)
++ return true;
+
-+ AuTraceErr(-ENOMEM);
-+ return -ENOMEM;
++ memcpy(a->name, name, namelen);
++ a->namelen = namelen;
++ a->found = 1;
++ return false;
+}
-diff -urN linux/fs/aufs/f_op.c linux-aufs/fs/aufs/f_op.c
---- linux/fs/aufs/f_op.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/f_op.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,673 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * file and vm operations
-+ *
-+ * $Id$
-+ */
-+
-+#include <linux/fs_stack.h>
-+#include <linux/poll.h>
-+#include "aufs.h"
+
-+/* common function to regular file and dir */
-+int aufs_flush(struct file *file, fl_owner_t id)
++static struct dentry *au_lkup_by_ino(struct path *path, ino_t ino,
++ struct au_nfsd_si_lock *nsi_lock)
+{
++ struct dentry *dentry, *parent;
++ struct file *file;
++ struct inode *dir;
++ struct find_name_by_ino arg = {
++ .ctx = {
++ .actor = find_name_by_ino
++ }
++ };
+ int err;
-+ struct dentry *dentry;
-+ aufs_bindex_t bindex, bend;
-+ struct file *h_file;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ parent = path->dentry;
++ if (nsi_lock)
++ si_read_unlock(parent->d_sb);
++ file = vfsub_dentry_open(path, au_dir_roflags);
++ dentry = (void *)file;
++ if (IS_ERR(file))
++ goto out;
+
-+ si_noflush_read_lock(dentry->d_sb);
-+ fi_read_lock(file);
-+ di_read_lock_child(dentry, AuLock_IW);
++ dentry = ERR_PTR(-ENOMEM);
++ arg.name = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!arg.name))
++ goto out_file;
++ arg.ino = ino;
++ arg.found = 0;
++ do {
++ arg.called = 0;
++ /* smp_mb(); */
++ err = vfsub_iterate_dir(file, &arg.ctx);
++ } while (!err && !arg.found && arg.called);
++ dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_name;
++ /* instead of ENOENT */
++ dentry = ERR_PTR(-ESTALE);
++ if (!arg.found)
++ goto out_name;
+
-+ err = 0;
-+ bend = au_fbend(file);
-+ for (bindex = au_fbstart(file); !err && bindex <= bend; bindex++) {
-+ h_file = au_h_fptr(file, bindex);
-+ if (h_file && h_file->f_op && h_file->f_op->flush) {
-+ err = h_file->f_op->flush(h_file, id);
-+ if (!err)
-+ au_update_fuse_h_inode
-+ (h_file->f_vfsmnt, h_file->f_dentry);
-+ /*ignore*/
-+ }
++ /* do not call vfsub_lkup_one() */
++ dir = d_inode(parent);
++ dentry = vfsub_lookup_one_len_unlocked(arg.name, path, arg.namelen);
++ AuTraceErrPtr(dentry);
++ if (IS_ERR(dentry))
++ goto out_name;
++ AuDebugOn(au_test_anon(dentry));
++ if (unlikely(d_really_is_negative(dentry))) {
++ dput(dentry);
++ dentry = ERR_PTR(-ENOENT);
+ }
-+ au_cpup_attr_timesizes(dentry->d_inode);
+
-+ di_read_unlock(dentry, AuLock_IW);
-+ fi_read_unlock(file);
-+ si_read_unlock(dentry->d_sb);
-+ AuTraceErr(err);
-+ return err;
++out_name:
++ free_page((unsigned long)arg.name);
++out_file:
++ fput(file);
++out:
++ if (unlikely(nsi_lock
++ && si_nfsd_read_lock(parent->d_sb, nsi_lock) < 0))
++ if (!IS_ERR(dentry)) {
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++ }
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static int do_open_nondir(struct file *file, int flags)
++static struct dentry *decode_by_dir_ino(struct super_block *sb, ino_t ino,
++ ino_t dir_ino,
++ struct au_nfsd_si_lock *nsi_lock)
+{
-+ int err;
-+ aufs_bindex_t bindex;
-+ struct super_block *sb;
-+ struct file *h_file;
+ struct dentry *dentry;
-+ struct inode *inode;
-+ struct au_finfo *finfo;
++ struct path path;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, flags 0%o\n", AuDLNPair(dentry), flags);
-+ FiMustWriteLock(file);
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode || S_ISDIR(inode->i_mode));
++ if (dir_ino != AUFS_ROOT_INO) {
++ path.dentry = decode_by_ino(sb, dir_ino, 0);
++ dentry = path.dentry;
++ if (!path.dentry || IS_ERR(path.dentry))
++ goto out;
++ AuDebugOn(au_test_anon(path.dentry));
++ } else
++ path.dentry = dget(sb->s_root);
+
-+ err = 0;
-+ finfo = au_fi(file);
-+ finfo->fi_h_vm_ops = NULL;
-+ sb = dentry->d_sb;
-+ bindex = au_dbstart(dentry);
-+ AuDebugOn(!au_h_dptr(dentry, bindex)->d_inode);
-+ /* O_TRUNC is processed already */
-+ BUG_ON(au_test_ro(sb, bindex, inode) && (flags & O_TRUNC));
++ path.mnt = au_mnt_get(sb);
++ dentry = au_lkup_by_ino(&path, ino, nsi_lock);
++ path_put(&path);
+
-+ h_file = au_h_open(dentry, bindex, flags, file);
-+ if (IS_ERR(h_file))
-+ err = PTR_ERR(h_file);
-+ else {
-+ au_set_fbstart(file, bindex);
-+ au_set_fbend(file, bindex);
-+ au_set_h_fptr(file, bindex, h_file);
-+ au_update_figen(file);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
-+ err = 0;
-+ }
-+ AuTraceErr(err);
-+ return err;
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+static int aufs_open_nondir(struct inode *inode, struct file *file)
-+{
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(file->f_dentry));
-+
-+ return au_do_open(inode, file, do_open_nondir);
-+}
++/* ---------------------------------------------------------------------- */
+
-+static int aufs_release_nondir(struct inode *inode, struct file *file)
++static int h_acceptable(void *expv, struct dentry *dentry)
+{
-+ struct super_block *sb = file->f_dentry->d_sb;
-+
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(file->f_dentry));
-+
-+ si_noflush_read_lock(sb);
-+ au_finfo_fin(file);
-+ si_read_unlock(sb);
-+ return 0;
++ return 1;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static ssize_t aufs_read(struct file *file, char __user *buf, size_t count,
-+ loff_t *ppos)
++static char *au_build_path(struct dentry *h_parent, struct path *h_rootpath,
++ char *buf, int len, struct super_block *sb)
+{
-+ ssize_t err;
-+ struct dentry *dentry;
-+ struct file *h_file;
-+ struct super_block *sb;
-+ struct inode *h_inode;
++ char *p;
++ int n;
++ struct path path;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, cnt %zu, pos %lld\n", AuDLNPair(dentry), count, *ppos);
++ p = d_path(h_rootpath, buf, len);
++ if (IS_ERR(p))
++ goto out;
++ n = strlen(p);
+
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0,
-+ /*locked*/0);
-+ if (unlikely(err))
++ path.mnt = h_rootpath->mnt;
++ path.dentry = h_parent;
++ p = d_path(&path, buf, len);
++ if (IS_ERR(p))
+ goto out;
++ if (n != 1)
++ p += n;
+
-+ /* support LSM and notify */
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ h_inode = h_file->f_dentry->d_inode;
-+ err = vfsub_read_u(h_file, buf, count, ppos,
-+ au_test_dlgt(au_mntflags(sb)));
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
-+ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode);
++ path.mnt = au_mnt_get(sb);
++ path.dentry = sb->s_root;
++ p = d_path(&path, buf, len - strlen(p));
++ mntput(path.mnt);
++ if (IS_ERR(p))
++ goto out;
++ if (n != 1)
++ p[strlen(p)] = '/';
+
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_read_unlock(file);
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ return err;
++out:
++ AuTraceErrPtr(p);
++ return p;
+}
+
-+static ssize_t aufs_write(struct file *file, const char __user *ubuf,
-+ size_t count, loff_t *ppos)
++static
++struct dentry *decode_by_path(struct super_block *sb, ino_t ino, __u32 *fh,
++ int fh_len, struct au_nfsd_si_lock *nsi_lock)
+{
-+ ssize_t err;
-+ struct dentry *dentry;
-+ struct inode *inode;
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ struct file *h_file;
-+ char __user *buf = (char __user *)ubuf;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ aufs_bindex_t bstart;
-+ int hinotify;
-+ struct au_pin pin;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, cnt %zu, pos %lld\n", AuDLNPair(dentry), count, *ppos);
-+
-+ inode = dentry->d_inode;
-+ mutex_lock(&inode->i_mutex);
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ mnt_flags = au_mntflags(sb);
-+ hinotify = !!au_opt_test(mnt_flags, UDBA_INOTIFY);
-+ vfsub_args_init(&vargs, &ign, au_test_dlgt(mnt_flags), 0);
++ struct dentry *dentry, *h_parent, *root;
++ struct super_block *h_sb;
++ char *pathname, *p;
++ struct vfsmount *h_mnt;
++ struct au_branch *br;
++ int err;
++ struct path path;
+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1,
-+ /*locked*/1);
-+ if (unlikely(err))
++ br = au_sbr(sb, nsi_lock->bindex);
++ h_mnt = au_br_mnt(br);
++ h_sb = h_mnt->mnt_sb;
++ /* todo: call lower fh_to_dentry()? fh_to_parent()? */
++ lockdep_off();
++ h_parent = exportfs_decode_fh(h_mnt, (void *)(fh + Fh_tail),
++ fh_len - Fh_tail, fh[Fh_h_type],
++ h_acceptable, /*context*/NULL);
++ lockdep_on();
++ dentry = h_parent;
++ if (unlikely(!h_parent || IS_ERR(h_parent))) {
++ AuWarn1("%s decode_fh failed, %ld\n",
++ au_sbtype(h_sb), PTR_ERR(h_parent));
+ goto out;
-+ err = au_ready_to_write(file, -1, &pin);
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ if (unlikely(err))
-+ goto out_unlock;
-+
-+ bstart = au_fbstart(file);
-+ h_file = au_h_fptr(file, bstart);
-+ if (!hinotify) {
-+ au_unpin(&pin);
-+ err = vfsub_write_u(h_file, buf, count, ppos, &vargs);
-+ } else {
-+ vfsub_ign_hinode(&vargs, IN_MODIFY, au_pinned_hdir(&pin));
-+ err = vfsub_write_u(h_file, buf, count, ppos, &vargs);
-+ au_unpin(&pin);
+ }
-+ au_cpup_attr_timesizes(inode);
++ dentry = NULL;
++ if (unlikely(au_test_anon(h_parent))) {
++ AuWarn1("%s decode_fh returned a disconnected dentry\n",
++ au_sbtype(h_sb));
++ goto out_h_parent;
++ }
++
++ dentry = ERR_PTR(-ENOMEM);
++ pathname = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!pathname))
++ goto out_h_parent;
++
++ root = sb->s_root;
++ path.mnt = h_mnt;
++ di_read_lock_parent(root, !AuLock_IR);
++ path.dentry = au_h_dptr(root, nsi_lock->bindex);
++ di_read_unlock(root, !AuLock_IR);
++ p = au_build_path(h_parent, &path, pathname, PAGE_SIZE, sb);
++ dentry = (void *)p;
++ if (IS_ERR(p))
++ goto out_pathname;
+
-+ out_unlock:
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_write_unlock(file);
-+ out:
+ si_read_unlock(sb);
-+ mutex_unlock(&inode->i_mutex);
-+ AuTraceErr(err);
-+ return err;
-+}
++ err = vfsub_kern_path(p, LOOKUP_FOLLOW | LOOKUP_DIRECTORY, &path);
++ dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_relock;
+
-+#ifdef CONFIG_AUFS_SPLICE_PATCH
-+static int au_test_loopback(void)
-+{
-+ const char c = current->comm[4];
-+ /* true if a kernel thread named 'loop[0-9].*' accesses a file */
-+ const int loopback = (current->mm == NULL
-+ && '0' <= c && c <= '9'
-+ && strncmp(current->comm, "loop", 4) == 0);
-+ return loopback;
++ dentry = ERR_PTR(-ENOENT);
++ AuDebugOn(au_test_anon(path.dentry));
++ if (unlikely(d_really_is_negative(path.dentry)))
++ goto out_path;
++
++ if (ino != d_inode(path.dentry)->i_ino)
++ dentry = au_lkup_by_ino(&path, ino, /*nsi_lock*/NULL);
++ else
++ dentry = dget(path.dentry);
++
++out_path:
++ path_put(&path);
++out_relock:
++ if (unlikely(si_nfsd_read_lock(sb, nsi_lock) < 0))
++ if (!IS_ERR(dentry)) {
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++ }
++out_pathname:
++ free_page((unsigned long)pathname);
++out_h_parent:
++ dput(h_parent);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+static ssize_t aufs_splice_read(struct file *file, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags)
++/* ---------------------------------------------------------------------- */
++
++static struct dentry *
++aufs_fh_to_dentry(struct super_block *sb, struct fid *fid, int fh_len,
++ int fh_type)
+{
-+ ssize_t err;
-+ struct file *h_file;
+ struct dentry *dentry;
-+ struct super_block *sb;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, pos %lld, len %zu\n", AuDLNPair(dentry), *ppos, len);
++ __u32 *fh = fid->raw;
++ struct au_branch *br;
++ ino_t ino, dir_ino;
++ struct au_nfsd_si_lock nsi_lock = {
++ .force_lock = 0
++ };
+
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0,
-+ /*locked*/0);
-+ if (unlikely(err))
++ dentry = ERR_PTR(-ESTALE);
++ /* it should never happen, but the file handle is unreliable */
++ if (unlikely(fh_len < Fh_tail))
+ goto out;
++ nsi_lock.sigen = fh[Fh_sigen];
++ nsi_lock.br_id = fh[Fh_br_id];
+
-+ err = -EINVAL;
-+ /* support LSM and notify */
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ if (au_test_loopback()) {
-+ file->f_mapping = h_file->f_mapping;
-+ smp_mb(); /* unnecessary? */
-+ }
-+ err = vfsub_splice_to(h_file, ppos, pipe, len, flags,
-+ au_test_dlgt(au_mntflags(sb)));
-+ /* todo: necessasry? */
-+ /* file->f_ra = h_file->f_ra; */
-+ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode);
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_read_unlock(file);
-+
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ return err;
-+}
++ /* branch id may be wrapped around */
++ br = NULL;
++ if (unlikely(si_nfsd_read_lock(sb, &nsi_lock)))
++ goto out;
++ nsi_lock.force_lock = 1;
+
-+static ssize_t
-+aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos,
-+ size_t len, unsigned int flags)
-+{
-+ ssize_t err;
-+ struct dentry *dentry;
-+ struct inode *inode, *h_inode;
-+ struct super_block *sb;
-+ struct file *h_file;
-+ /* struct au_hin_ignore ign; */
-+ struct vfsub_args vargs;
-+ unsigned int mnt_flags;
-+ struct au_pin pin;
++ /* is this inode still cached? */
++ ino = decode_ino(fh + Fh_ino);
++ /* it should never happen */
++ if (unlikely(ino == AUFS_ROOT_INO))
++ goto out_unlock;
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, len %zu, pos %lld\n", AuDLNPair(dentry), len, *ppos);
++ dir_ino = decode_ino(fh + Fh_dir_ino);
++ dentry = decode_by_ino(sb, ino, dir_ino);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (dentry)
++ goto accept;
+
-+ inode = dentry->d_inode;
-+ mutex_lock(&inode->i_mutex);
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ mnt_flags = au_mntflags(sb);
-+ vfsub_args_init(&vargs, /*&ign*/NULL, au_test_dlgt(mnt_flags), 0);
++ /* is the parent dir cached? */
++ br = au_sbr(sb, nsi_lock.bindex);
++ au_lcnt_inc(&br->br_nfiles);
++ dentry = decode_by_dir_ino(sb, ino, dir_ino, &nsi_lock);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (dentry)
++ goto accept;
+
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1,
-+ /*locked*/1);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_ready_to_write(file, -1, &pin);
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ if (unlikely(err))
++ /* lookup path */
++ dentry = decode_by_path(sb, ino, fh, fh_len, &nsi_lock);
++ if (IS_ERR(dentry))
++ goto out_unlock;
++ if (unlikely(!dentry))
++ /* todo?: make it ESTALE */
+ goto out_unlock;
+
-+ /* support LSM and notify */
-+ /* current vfs_splice_from() doesn't fire up the inotify event */
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ h_inode = h_file->f_dentry->d_inode;
-+ if (1 || !au_opt_test(mnt_flags, UDBA_INOTIFY)) {
-+ au_unpin(&pin);
-+ err = vfsub_splice_from(pipe, h_file, ppos, len, flags, &vargs);
-+ }
-+#if 0 /* reserved for future use */
-+ else {
-+ struct dentry *parent = dget_parent(dentry);
-+ vfsub_ign_hinode(&vargs, IN_MODIFY, au_pinned_hdir(&pin));
-+ err = vfsub_splice_from(pipe, h_file, ppos, len, flags, &vargs);
-+ au_unpin(&pin);
-+ }
-+#endif
-+ au_cpup_attr_timesizes(inode);
++accept:
++ if (!au_digen_test(dentry, au_sigen(sb))
++ && d_inode(dentry)->i_generation == fh[Fh_igen])
++ goto out_unlock; /* success */
+
-+ out_unlock:
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_write_unlock(file);
-+ out:
++ dput(dentry);
++ dentry = ERR_PTR(-ESTALE);
++out_unlock:
++ if (br)
++ au_lcnt_dec(&br->br_nfiles);
+ si_read_unlock(sb);
-+ mutex_unlock(&inode->i_mutex);
-+ AuTraceErr(err);
-+ return err;
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
-+#endif /* CONFIG_AUFS_SPLICE_PATCH */
+
-+/* ---------------------------------------------------------------------- */
-+
-+static int aufs_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
++#if 0 /* reserved for future use */
++/* support subtreecheck option */
++static struct dentry *aufs_fh_to_parent(struct super_block *sb, struct fid *fid,
++ int fh_len, int fh_type)
+{
-+ int err;
-+ struct dentry *dentry;
-+ struct file *file, *h_file;
-+ struct inode *inode;
-+ static DECLARE_WAIT_QUEUE_HEAD(wq);
-+ struct au_finfo *finfo;
++ struct dentry *parent;
++ __u32 *fh = fid->raw;
++ ino_t dir_ino;
+
-+ AuTraceEnter();
-+ AuDebugOn(!vma || !vma->vm_file);
-+ /* todo: non-robr mode, user vm_file as it is? */
-+ wait_event(wq, (file = au_robr_safe_file(vma)));
-+ AuDebugOn(!au_test_aufs(file->f_dentry->d_sb));
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ inode = dentry->d_inode;
-+ AuDebugOn(!S_ISREG(inode->i_mode));
-+
-+ /* do not revalidate, no si lock */
-+ finfo = au_fi(file);
-+ h_file = finfo->fi_hfile[0 + finfo->fi_bstart].hf_file;
-+ AuDebugOn(!h_file || !au_test_mmapped(file));
-+ fi_write_lock(file);
-+ vma->vm_file = h_file;
-+ err = finfo->fi_h_vm_ops->fault(vma, vmf);
-+ /* todo: necessary? */
-+ /* file->f_ra = h_file->f_ra; */
-+ au_robr_reset_file(vma, file);
-+ fi_write_unlock(file);
-+#if 0 /* def CONFIG_SMP */
-+ /* wake_up_nr(&wq, online_cpu - 1); */
-+ wake_up_all(&wq);
-+#else
-+ wake_up(&wq);
-+#endif
++ dir_ino = decode_ino(fh + Fh_dir_ino);
++ parent = decode_by_ino(sb, dir_ino, 0);
++ if (IS_ERR(parent))
++ goto out;
++ if (!parent)
++ parent = decode_by_path(sb, au_br_index(sb, fh[Fh_br_id]),
++ dir_ino, fh, fh_len);
+
-+ if (!(err & VM_FAULT_ERROR)) {
-+#if 0 /* debug */
-+ struct page *page;
-+ page = vmf->page;
-+ AuDbg("%p, %d\n", page, page_mapcount(page));
-+
-+ page->mapping = file->f_mapping;
-+ get_page(page);
-+ file->f_mapping = h_file->f_mapping;
-+ touch_atime(NULL, dentry);
-+ inode->i_atime = h_file->f_dentry->d_inode->i_atime;
-+#endif
-+ }
-+ AuTraceErr(err);
-+ return err;
++out:
++ AuTraceErrPtr(parent);
++ return parent;
+}
-+
-+static struct vm_operations_struct aufs_vm_ops = {
-+ .fault = aufs_fault,
-+#if 0 /* reserved for future use */
-+ .open = aufs_vmaopen,
-+ .close = aufs_vmaclose,
-+ unsigned long (*nopfn)(struct vm_area_struct *area,
-+ unsigned long address);
-+ page_mkwrite(struct vm_area_struct *vma, struct page *page)
+#endif
-+};
+
+/* ---------------------------------------------------------------------- */
+
-+static struct vm_operations_struct *au_vm_ops(struct file *h_file,
-+ struct vm_area_struct *vma)
++static int aufs_encode_fh(struct inode *inode, __u32 *fh, int *max_len,
++ struct inode *dir)
+{
-+ struct vm_operations_struct *vm_ops;
+ int err;
++ aufs_bindex_t bindex;
++ struct super_block *sb, *h_sb;
++ struct dentry *dentry, *parent, *h_parent;
++ struct inode *h_dir;
++ struct au_branch *br;
+
-+ AuTraceEnter();
-+
-+ au_br_nfs_lockdep_off(h_file->f_vfsmnt->mnt_sb);
-+ err = h_file->f_op->mmap(h_file, vma);
-+ au_br_nfs_lockdep_on(h_file->f_vfsmnt->mnt_sb);
-+ vm_ops = ERR_PTR(err);
-+ if (unlikely(err))
++ err = -ENOSPC;
++ if (unlikely(*max_len <= Fh_tail)) {
++ AuWarn1("NFSv2 client (max_len %d)?\n", *max_len);
+ goto out;
-+ vm_ops = vma->vm_ops;
-+ err = do_munmap(current->mm, vma->vm_start,
-+ vma->vm_end - vma->vm_start);
-+ if (unlikely(err)) {
-+ AuIOErr("failed internal unmapping %.*s, %d\n",
-+ AuDLNPair(h_file->f_dentry), err);
-+ vm_ops = ERR_PTR(-EIO);
+ }
+
-+ out:
-+ AuTraceErrPtr(vm_ops);
-+ return vm_ops;
-+}
-+
-+static int aufs_mmap(struct file *file, struct vm_area_struct *vma)
-+{
-+ int err;
-+ unsigned char wlock, mmapped;
-+ struct dentry *dentry;
-+ struct super_block *sb;
-+ struct file *h_file;
-+ struct vm_operations_struct *vm_ops;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, %lx, len %lu\n",
-+ AuDLNPair(dentry), vma->vm_start,
-+ vma->vm_end - vma->vm_start);
-+ AuDebugOn(!S_ISREG(dentry->d_inode->i_mode));
-+ AuDebugOn(down_write_trylock(&vma->vm_mm->mmap_sem));
++ err = FILEID_ROOT;
++ if (inode->i_ino == AUFS_ROOT_INO) {
++ AuDebugOn(inode->i_ino != AUFS_ROOT_INO);
++ goto out;
++ }
+
-+ mmapped = au_test_mmapped(file); /* can be harmless race condition */
-+ wlock = !!(file->f_mode & FMODE_WRITE);
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, wlock | !mmapped,
-+ /*locked*/0);
++ h_parent = NULL;
++ sb = inode->i_sb;
++ err = si_read_lock(sb, AuLock_FLUSH);
+ if (unlikely(err))
+ goto out;
+
-+ if (wlock) {
-+ struct au_pin pin;
-+
-+ err = au_ready_to_write(file, -1, &pin);
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ if (unlikely(err))
++#ifdef CONFIG_AUFS_DEBUG
++ if (unlikely(!au_opt_test(au_mntflags(sb), XINO)))
++ AuWarn1("NFS-exporting requires xino\n");
++#endif
++ err = -EIO;
++ parent = NULL;
++ ii_read_lock_child(inode);
++ bindex = au_ibtop(inode);
++ if (!dir) {
++ dentry = d_find_any_alias(inode);
++ if (unlikely(!dentry))
+ goto out_unlock;
-+ au_unpin(&pin);
-+ } else if (!mmapped)
-+ di_downgrade_lock(dentry, AuLock_IR);
-+
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ if (au_test_fuse(h_file->f_dentry->d_sb)) {
-+ /*
-+ * by this assignment, f_mapping will differs from aufs inode
-+ * i_mapping.
-+ * if someone else mixes the use of f_dentry->d_inode and
-+ * f_mapping->host, then a problem may arise.
-+ */
-+ file->f_mapping = h_file->f_mapping;
++ AuDebugOn(au_test_anon(dentry));
++ parent = dget_parent(dentry);
++ dput(dentry);
++ if (unlikely(!parent))
++ goto out_unlock;
++ if (d_really_is_positive(parent))
++ dir = d_inode(parent);
+ }
+
-+ if (0 && h_file->f_op->mmap == generic_file_mmap) {
-+ err = generic_file_mmap(file, vma); /* instead of h_file */
-+ if (unlikely(err))
-+ goto out_unlock;
-+ au_fi(file)->fi_h_vm_ops = vma->vm_ops;
-+ } else {
-+ vm_ops = NULL;
-+ if (!mmapped) {
-+ vm_ops = au_vm_ops(h_file, vma);
-+ err = PTR_ERR(vm_ops);
-+ if (IS_ERR(vm_ops))
-+ goto out_unlock;
-+ }
++ ii_read_lock_parent(dir);
++ h_dir = au_h_iptr(dir, bindex);
++ ii_read_unlock(dir);
++ if (unlikely(!h_dir))
++ goto out_parent;
++ h_parent = d_find_any_alias(h_dir);
++ if (unlikely(!h_parent))
++ goto out_hparent;
+
-+ err = generic_file_mmap(file, vma);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ vma->vm_ops = &aufs_vm_ops;
-+ /* test again */
-+ if (!au_test_mmapped(file)) {
-+ FiMustWriteLock(file);
-+ au_fi(file)->fi_h_vm_ops = vm_ops;
-+ }
++ err = -EPERM;
++ br = au_sbr(sb, bindex);
++ h_sb = au_br_sb(br);
++ if (unlikely(!h_sb->s_export_op)) {
++ AuErr1("%s branch is not exportable\n", au_sbtype(h_sb));
++ goto out_hparent;
+ }
+
-+ file_accessed(h_file);
-+ au_update_fuse_h_inode(h_file->f_vfsmnt, h_file->f_dentry); /*ignore*/
-+ fsstack_copy_attr_atime(dentry->d_inode, h_file->f_dentry->d_inode);
++ fh[Fh_br_id] = br->br_id;
++ fh[Fh_sigen] = au_sigen(sb);
++ encode_ino(fh + Fh_ino, inode->i_ino);
++ encode_ino(fh + Fh_dir_ino, dir->i_ino);
++ fh[Fh_igen] = inode->i_generation;
+
-+ out_unlock:
-+ di_read_unlock(dentry, AuLock_IR);
-+ if (!wlock && mmapped)
-+ fi_read_unlock(file);
++ *max_len -= Fh_tail;
++ fh[Fh_h_type] = exportfs_encode_fh(h_parent, (void *)(fh + Fh_tail),
++ max_len,
++ /*connectable or subtreecheck*/0);
++ err = fh[Fh_h_type];
++ *max_len += Fh_tail;
++ /* todo: macros? */
++ if (err != FILEID_INVALID)
++ err = 99;
+ else
-+ fi_write_unlock(file);
-+ out:
++ AuWarn1("%s encode_fh failed\n", au_sbtype(h_sb));
++
++out_hparent:
++ dput(h_parent);
++out_parent:
++ dput(parent);
++out_unlock:
++ ii_read_unlock(inode);
+ si_read_unlock(sb);
-+ AuTraceErr(err);
++out:
++ if (unlikely(err < 0))
++ err = FILEID_INVALID;
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static unsigned int aufs_poll(struct file *file, poll_table *wait)
++static int aufs_commit_metadata(struct inode *inode)
+{
-+ unsigned int mask;
-+ struct file *h_file;
+ int err;
-+ struct dentry *dentry;
++ aufs_bindex_t bindex;
+ struct super_block *sb;
++ struct inode *h_inode;
++ int (*f)(struct inode *inode);
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, wait %p\n", AuDLNPair(dentry), wait);
-+ AuDebugOn(S_ISDIR(dentry->d_inode->i_mode));
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++ ii_write_lock_child(inode);
++ bindex = au_ibtop(inode);
++ AuDebugOn(bindex < 0);
++ h_inode = au_h_iptr(inode, bindex);
+
-+ /* We should pretend an error happened. */
-+ mask = POLLERR /* | POLLIN | POLLOUT */;
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0,
-+ /*locked*/0);
-+ if (unlikely(err))
-+ goto out;
-+
-+ /* it is not an error of hidden_file has no operation */
-+ mask = DEFAULT_POLLMASK;
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ if (h_file->f_op && h_file->f_op->poll)
-+ mask = h_file->f_op->poll(h_file, wait);
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_read_unlock(file);
-+
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErr((int)mask);
-+ return mask;
-+}
-+
-+static int aufs_fsync_nondir(struct file *file, struct dentry *dentry,
-+ int datasync)
-+{
-+ int err;
-+ struct inode *inode;
-+ struct file *h_file;
-+ struct super_block *sb;
-+ struct au_pin pin;
-+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), datasync);
-+ inode = dentry->d_inode;
-+
-+ IMustLock(file->f_mapping->host);
-+ if (inode != file->f_mapping->host) {
-+ mutex_unlock(&file->f_mapping->host->i_mutex);
-+ mutex_lock(&inode->i_mutex);
-+ }
-+ IMustLock(inode);
-+
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = 0; /* -EBADF; */ /* posix? */
-+ if (unlikely(!(file->f_mode & FMODE_WRITE)))
-+ goto out;
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1,
-+ /*locked*/1);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_ready_to_write(file, -1, &pin);
-+ di_downgrade_lock(dentry, AuLock_IR);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ au_unpin(&pin);
-+
-+ err = -EINVAL;
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ if (h_file->f_op && h_file->f_op->fsync) {
-+ struct mutex *h_mtx = &h_file->f_dentry->d_inode->i_mutex;
-+
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ err = h_file->f_op->fsync(h_file, h_file->f_dentry, datasync);
-+ if (!err)
-+ au_update_fuse_h_inode(h_file->f_vfsmnt,
-+ h_file->f_dentry);
-+ au_cpup_attr_timesizes(inode);
-+ mutex_unlock(h_mtx);
-+ }
++ f = h_inode->i_sb->s_export_op->commit_metadata;
++ if (f)
++ err = f(h_inode);
++ else
++ err = sync_inode_metadata(h_inode, /*wait*/1);
+
-+ out_unlock:
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_write_unlock(file);
-+ out:
++ au_cpup_attr_timesizes(inode);
++ ii_write_unlock(inode);
+ si_read_unlock(sb);
-+ if (inode != file->f_mapping->host) {
-+ mutex_unlock(&inode->i_mutex);
-+ mutex_lock(&file->f_mapping->host->i_mutex);
-+ }
-+ AuTraceErr(err);
+ return err;
+}
+
-+static int aufs_fasync(int fd, struct file *file, int flag)
-+{
-+ int err;
-+ struct file *h_file;
-+ struct dentry *dentry;
-+ struct super_block *sb;
++/* ---------------------------------------------------------------------- */
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), flag);
++static struct export_operations aufs_export_op = {
++ .fh_to_dentry = aufs_fh_to_dentry,
++ /* .fh_to_parent = aufs_fh_to_parent, */
++ .encode_fh = aufs_encode_fh,
++ .commit_metadata = aufs_commit_metadata
++};
+
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0,
-+ /*locked*/0);
-+ if (unlikely(err))
-+ goto out;
++void au_export_init(struct super_block *sb)
++{
++ struct au_sbinfo *sbinfo;
++ __u32 u;
+
-+ h_file = au_h_fptr(file, au_fbstart(file));
-+ if (h_file->f_op && h_file->f_op->fasync)
-+ err = h_file->f_op->fasync(fd, h_file, flag);
-+ di_read_unlock(dentry, AuLock_IR);
-+ fi_read_unlock(file);
++ BUILD_BUG_ON_MSG(IS_BUILTIN(CONFIG_AUFS_FS)
++ && IS_MODULE(CONFIG_EXPORTFS),
++ AUFS_NAME ": unsupported configuration "
++ "CONFIG_EXPORTFS=m and CONFIG_AUFS_FS=y");
+
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ return err;
++ sb->s_export_op = &aufs_export_op;
++ sbinfo = au_sbi(sb);
++ sbinfo->si_xigen = NULL;
++ get_random_bytes(&u, sizeof(u));
++ BUILD_BUG_ON(sizeof(u) != sizeof(int));
++ atomic_set(&sbinfo->si_xigen_next, u);
+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct file_operations aufs_file_fop = {
-+ /*
-+ * while generic_file_llseek/_unlocked() don't use BKL,
-+ * don't use it since it operates file->f_mapping->host.
-+ * in aufs, it may be a real file and may confuse users by UDBA.
-+ */
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 27)
-+ /* .llseek = generic_file_llseek, */
-+#endif
-+ .read = aufs_read,
-+ .write = aufs_write,
-+ .poll = aufs_poll,
-+ /* .unlocked_ioctl = aufs_ioctl, */
-+ .mmap = aufs_mmap,
-+ .open = aufs_open_nondir,
-+ .flush = aufs_flush,
-+ .release = aufs_release_nondir,
-+ .fsync = aufs_fsync_nondir,
-+ .fasync = aufs_fasync,
-+#ifdef CONFIG_AUFS_SPLICE_PATCH
-+ .splice_write = aufs_splice_write,
-+ .splice_read = aufs_splice_read,
-+#endif
-+};
-diff -urN linux/fs/aufs/getattr.c linux-aufs/fs/aufs/getattr.c
---- linux/fs/aufs/getattr.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/getattr.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,133 @@
+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 2022-11-05 23:02:18.962555950 +0100
+@@ -0,0 +1,426 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2011-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode attributes on NFS/FUSE branch or HINOTIFY
-+ *
-+ * $Id$
++ * File-based Hierarchy Storage Management
+ */
+
++#include <linux/anon_inodes.h>
++#include <linux/poll.h>
++#include <linux/seq_file.h>
++#include <linux/statfs.h>
+#include "aufs.h"
+
-+static struct dentry *
-+au_h_dget_any(struct dentry *dentry, aufs_bindex_t *bindex)
++static aufs_bindex_t au_fhsm_bottom(struct super_block *sb)
+{
-+ struct dentry *h_dentry;
-+ struct inode *inode, *h_inode;
-+ struct super_block *sb;
-+ aufs_bindex_t ib, db;
-+
-+ /* must be positive dentry */
-+ inode = dentry->d_inode;
-+ LKTRTrace("%.*s, i%lu\n", AuDLNPair(dentry), inode->i_ino);
-+
-+ sb = dentry->d_sb;
-+ db = au_dbstart(dentry);
-+ ib = au_ibstart(inode);
-+ if (db == ib) {
-+ *bindex = db;
-+ h_dentry = dget(au_h_dptr(dentry, db));
-+ if (h_dentry)
-+ goto out; /* success */
-+ }
-+
-+ *bindex = ib;
-+ h_inode = au_h_iptr(inode, ib);
-+ h_dentry = d_find_alias(h_inode);
-+ if (h_dentry)
-+ goto out; /* success */
-+
-+#if 0
-+ if (au_opt_test(au_mntflags(sb), PLINK)
-+ && au_plink_test(sb, inode)) {
-+ h_dentry = au_plink_lkup(sb, ib, inode);
-+ if (IS_ERR(h_dentry))
-+ goto out;
-+ AuDebugOn(!h_dentry->d_inode);
-+ goto out; /* success */
-+ }
-+#endif
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ h_dentry = dget(au_hi_wh(inode, ib));
++ SiMustAnyLock(sb);
+
-+ out:
-+ AuTraceErrPtr(h_dentry);
-+ return h_dentry;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ AuDebugOn(!fhsm);
++ return fhsm->fhsm_bottom;
+}
+
-+int aufs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *st)
++void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ int err;
-+ unsigned int mnt_flags;
-+ aufs_bindex_t bindex;
-+ unsigned char did_lock;
-+ struct inode *inode;
-+ struct dentry *h_dentry;
-+ struct super_block *sb, *h_sb;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ SiMustWriteLock(sb);
+
-+ err = 0;
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ mnt_flags = au_mntflags(sb);
-+ if (dentry != sb->s_root) {
-+ di_read_lock_parent(dentry, AuLock_IR);
-+ inode = dentry->d_inode;
-+ did_lock = 1;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ AuDebugOn(!fhsm);
++ fhsm->fhsm_bottom = bindex;
++}
+
-+ /* todo: test bit inotify option too? */
-+ bindex = au_ibstart(inode);
-+ h_sb = au_sbr_sb(sb, bindex);
-+ /* todo: fix this condition */
-+ if ((au_opt_test(mnt_flags, PLINK) && au_plink_test(sb, inode))
-+ /* au_iigen(inode) == au_sigen(sb) */
-+ || (!au_test_fuse(h_sb) && !au_test_nfs(h_sb)))
-+ goto fill;
++/* ---------------------------------------------------------------------- */
+
-+ h_dentry = au_h_dget_any(dentry, &bindex);
-+ err = PTR_ERR(h_dentry);
-+ if (IS_ERR(h_dentry))
-+ goto out;
-+ } else {
-+ /* lock free root dinfo */
-+ did_lock = 0;
-+ bindex = 0;
-+ inode = dentry->d_inode;
-+ h_dentry = dget(au_di(dentry)->di_hdentry->hd_dentry);
-+ }
++static int au_fhsm_test_jiffy(struct au_sbinfo *sbinfo, struct au_branch *br)
++{
++ struct au_br_fhsm *bf;
+
-+ err = -EIO;
-+ if (h_dentry && h_dentry->d_inode)
-+ err = vfsub_getattr(au_sbr_mnt(sb, bindex), h_dentry, st,
-+ au_test_dlgt(mnt_flags));
-+ dput(h_dentry);
-+ if (!err) {
-+ /* todo: I don't like this approach */
-+ au_cpup_attr_all(inode, /*force*/0);
-+ fill:
-+ generic_fillattr(inode, st);
-+ }
++ bf = br->br_fhsm;
++ MtxMustLock(&bf->bf_lock);
+
-+ out:
-+ if (did_lock)
-+ di_read_unlock(dentry, AuLock_IR);
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ return err;
++ return !bf->bf_readable
++ || time_after(jiffies,
++ bf->bf_jiffy + sbinfo->si_fhsm.fhsm_expire);
+}
-diff -urN linux/fs/aufs/hinode.h linux-aufs/fs/aufs/hinode.h
---- linux/fs/aufs/hinode.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/hinode.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,175 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
+
-+/*
-+ * lower (branch filesystem) inode and setting inotify
-+ *
-+ * $Id$
-+ */
-+
-+#ifndef __AUFS_HINODE_H__
-+#define __AUFS_HINODE_H__
-+
-+#ifdef __KERNEL__
++/* ---------------------------------------------------------------------- */
+
-+#include <linux/fs.h>
-+#include <linux/inotify.h>
-+#include <linux/aufs_type.h>
-+#include "super.h"
-+#include "vfsub.h"
++static void au_fhsm_notify(struct super_block *sb, int val)
++{
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+/* ---------------------------------------------------------------------- */
++ SiMustAnyLock(sb);
+
-+struct au_hinotify {
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ struct au_splhead hin_ignore;
-+ struct inotify_watch hin_watch;
-+ struct inode *hin_aufs_inode; /* no get/put */
-+#endif
-+};
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ if (au_fhsm_pid(fhsm)
++ && atomic_read(&fhsm->fhsm_readable) != -1) {
++ atomic_set(&fhsm->fhsm_readable, val);
++ if (val)
++ wake_up(&fhsm->fhsm_wqh);
++ }
++}
+
-+struct au_hinode {
-+ struct inode *hi_inode;
-+ aufs_bindex_t hi_id;
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ struct au_hinotify *hi_notify;
-+#endif
++static int au_fhsm_stfs(struct super_block *sb, aufs_bindex_t bindex,
++ struct aufs_stfs *rstfs, int do_lock, int do_notify)
++{
++ int err;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+ /* reference to the copied-up whiteout with get/put */
-+ struct dentry *hi_whdentry;
-+};
++ br = au_sbr(sb, bindex);
++ AuDebugOn(au_br_rdonly(br));
++ bf = br->br_fhsm;
++ AuDebugOn(!bf);
+
-+struct au_hin_ignore {
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ struct list_head ign_list;
++ if (do_lock)
++ mutex_lock(&bf->bf_lock);
++ else
++ MtxMustLock(&bf->bf_lock);
+
-+ pid_t ign_pid;
-+ __u32 ign_events, ign_handled;
-+ struct au_hinode *ign_hinode;
-+#endif
-+};
++ /* sb->s_root for NFS is unreliable */
++ err = au_br_stfs(br, &bf->bf_stfs);
++ if (unlikely(err)) {
++ AuErr1("FHSM failed (%d), b%d, ignored.\n", bindex, err);
++ goto out;
++ }
+
-+/* ---------------------------------------------------------------------- */
++ bf->bf_jiffy = jiffies;
++ bf->bf_readable = 1;
++ if (do_notify)
++ au_fhsm_notify(sb, /*val*/1);
++ if (rstfs)
++ *rstfs = bf->bf_stfs;
+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+/* inotify events */
-+static const __u32 AuInMask = (IN_MOVE | IN_DELETE | IN_CREATE
-+ /* | IN_ACCESS */
-+ | IN_MODIFY | IN_ATTRIB
-+ /* | IN_DELETE_SELF | IN_MOVE_SELF */
-+ );
++out:
++ if (do_lock)
++ mutex_unlock(&bf->bf_lock);
++ au_fhsm_notify(sb, /*val*/1);
+
-+static inline
-+void au_hin_init(struct au_hinode *hinode, struct au_hinotify *val)
-+{
-+ hinode->hi_notify = val;
++ return err;
+}
+
-+/* hinotify.c */
-+int au_hin_alloc(struct au_hinode *hinode, struct inode *inode,
-+ struct inode *h_inode);
-+void au_hin_free(struct au_hinode *hinode);
-+void au_hin_ctl(struct au_hinode *hinode, const __u32 mask);
-+void au_reset_hinotify(struct inode *inode, unsigned int flags);
++void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+int au_hin_verify_gen(struct dentry *dentry);
++ AuDbg("b%d, force %d\n", bindex, force);
++ SiMustAnyLock(sb);
+
-+int __init au_inotify_init(void);
-+void au_inotify_fin(void);
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ if (!au_ftest_si(sbinfo, FHSM)
++ || fhsm->fhsm_bottom == bindex)
++ return;
+
-+static inline void au_hin_suspend(struct au_hinode *hinode)
-+{
-+ au_hin_ctl(hinode, 0);
++ br = au_sbr(sb, bindex);
++ bf = br->br_fhsm;
++ AuDebugOn(!bf);
++ mutex_lock(&bf->bf_lock);
++ if (force
++ || au_fhsm_pid(fhsm)
++ || au_fhsm_test_jiffy(sbinfo, br))
++ err = au_fhsm_stfs(sb, bindex, /*rstfs*/NULL, /*do_lock*/0,
++ /*do_notify*/1);
++ mutex_unlock(&bf->bf_lock);
+}
+
-+static inline void au_hin_resume(struct au_hinode *hinode)
++void au_fhsm_wrote_all(struct super_block *sb, int force)
+{
-+ au_hin_ctl(hinode, AuInMask);
-+}
-+
-+#else
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
+
-+static inline
-+void au_hin_init(struct au_hinode *hinode, struct au_hinotify *val)
-+{
-+ /* empty */
++ /* exclude the bottom */
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm))
++ au_fhsm_wrote(sb, bindex, force);
++ }
+}
+
-+static inline
-+int au_hin_alloc(struct au_hinode *hinode, struct inode *inode,
-+ struct inode *h_inode)
-+{
-+ return -EOPNOTSUPP;
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline void au_hin_free(struct au_hinode *hinode)
++static __poll_t au_fhsm_poll(struct file *file, struct poll_table_struct *wait)
+{
-+ /* nothing */
-+}
++ __poll_t mask;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+static inline void au_reset_hinotify(struct inode *inode, unsigned int flags)
-+{
-+ /* nothing */
-+}
++ mask = 0;
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++ poll_wait(file, &fhsm->fhsm_wqh, wait);
++ if (atomic_read(&fhsm->fhsm_readable))
++ mask = EPOLLIN /* | EPOLLRDNORM */;
+
-+static inline int au_hin_verify_gen(struct dentry *dentry)
-+{
-+ return 0;
++ if (!mask)
++ AuDbg("mask 0x%x\n", mask);
++ return mask;
+}
+
-+static inline int au_inotify_init(void)
++static int au_fhsm_do_read_one(struct aufs_stbr __user *stbr,
++ struct aufs_stfs *stfs, __s16 brid)
+{
-+ return 0;
-+}
++ int err;
+
-+#define au_inotify_fin() do {} while (0)
++ err = copy_to_user(&stbr->stfs, stfs, sizeof(*stfs));
++ if (!err)
++ err = __put_user(brid, &stbr->brid);
++ if (unlikely(err))
++ err = -EFAULT;
+
-+static inline void au_hin_suspend(struct au_hinode *hinode)
-+{
-+ /* empty */
++ return err;
+}
+
-+static inline void au_hin_resume(struct au_hinode *hinode)
++static ssize_t au_fhsm_do_read(struct super_block *sb,
++ struct aufs_stbr __user *stbr, size_t count)
+{
-+ /* empty */
-+}
-+#endif /* CONFIG_AUFS_HINOTIFY */
++ ssize_t err;
++ int nstbr;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++ struct au_br_fhsm *bf;
+
-+#if defined(CONFIG_AUFS_HINOTIFY) && defined(CONFIG_AUFS_DEBUG)
-+static inline void au_hin_list_del(struct list_head *e)
-+{
-+ list_del_init(e);
-+}
++ /* except the bottom branch */
++ err = 0;
++ nstbr = 0;
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; !err && bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (!au_br_fhsm(br->br_perm))
++ continue;
+
-+void au_dbg_hin_list(struct vfsub_args *vargs);
-+#else
-+static inline void au_hin_list_del(struct list_head *e)
-+{
-+ list_del(e);
++ bf = br->br_fhsm;
++ mutex_lock(&bf->bf_lock);
++ if (bf->bf_readable) {
++ err = -EFAULT;
++ if (count >= sizeof(*stbr))
++ err = au_fhsm_do_read_one(stbr++, &bf->bf_stfs,
++ br->br_id);
++ if (!err) {
++ bf->bf_readable = 0;
++ count -= sizeof(*stbr);
++ nstbr++;
++ }
++ }
++ mutex_unlock(&bf->bf_lock);
++ }
++ if (!err)
++ err = sizeof(*stbr) * nstbr;
++
++ return err;
+}
+
-+static inline void au_dbg_hin_list(struct vfsub_args *vargs)
++static ssize_t au_fhsm_read(struct file *file, char __user *buf, size_t count,
++ loff_t *pos)
+{
-+ /* empty */
-+}
-+#endif /* CONFIG_AUFS_DEBUG */
-+
-+/* ---------------------------------------------------------------------- */
++ ssize_t err;
++ int readable;
++ aufs_bindex_t nfhsm, bindex, bbot;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ struct au_branch *br;
++ struct super_block *sb;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_HINODE_H__ */
-diff -urN linux/fs/aufs/hin_or_dlgt.c linux-aufs/fs/aufs/hin_or_dlgt.c
---- linux/fs/aufs/hin_or_dlgt.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/hin_or_dlgt.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,736 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ err = 0;
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++need_data:
++ spin_lock_irq(&fhsm->fhsm_wqh.lock);
++ if (!atomic_read(&fhsm->fhsm_readable)) {
++ if (vfsub_file_flags(file) & O_NONBLOCK)
++ err = -EAGAIN;
++ else
++ err = wait_event_interruptible_locked_irq
++ (fhsm->fhsm_wqh,
++ atomic_read(&fhsm->fhsm_readable));
++ }
++ spin_unlock_irq(&fhsm->fhsm_wqh.lock);
++ if (unlikely(err))
++ goto out;
+
-+/*
-+ * sub-routines for vfs in hinotify or dlgt mode
-+ *
-+ * $Id$
-+ */
++ /* sb may already be dead */
++ au_rw_read_lock(&sbinfo->si_rwsem);
++ readable = atomic_read(&fhsm->fhsm_readable);
++ if (readable > 0) {
++ sb = sbinfo->si_sb;
++ AuDebugOn(!sb);
++ /* exclude the bottom branch */
++ nfhsm = 0;
++ bbot = au_fhsm_bottom(sb);
++ for (bindex = 0; bindex < bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm))
++ nfhsm++;
++ }
++ err = -EMSGSIZE;
++ if (nfhsm * sizeof(struct aufs_stbr) <= count) {
++ atomic_set(&fhsm->fhsm_readable, 0);
++ err = au_fhsm_do_read(sbinfo->si_sb, (void __user *)buf,
++ count);
++ }
++ }
++ au_rw_read_unlock(&sbinfo->si_rwsem);
++ if (!readable)
++ goto need_data;
+
-+#include <linux/uaccess.h>
-+#include "aufs.h"
++out:
++ return err;
++}
+
-+#if !defined(CONFIG_AUFS_HINOTIFY) && !defined(CONFIG_AUFS_DLGT)
-+#error mis-configuraion or Makefile
-+#endif
++static int au_fhsm_release(struct inode *inode, struct file *file)
++{
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++
++ /* sb may already be dead */
++ sbinfo = file->private_data;
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock(&fhsm->fhsm_spin);
++ fhsm->fhsm_pid = 0;
++ spin_unlock(&fhsm->fhsm_spin);
++ kobject_put(&sbinfo->si_kobj);
+
-+/* ---------------------------------------------------------------------- */
++ return 0;
++}
+
-+struct permission_args {
-+ int *errp;
-+ struct inode *inode;
-+ int mask;
-+ struct nameidata *nd;
++static const struct file_operations au_fhsm_fops = {
++ .owner = THIS_MODULE,
++ .llseek = noop_llseek,
++ .read = au_fhsm_read,
++ .poll = au_fhsm_poll,
++ .release = au_fhsm_release
+};
+
-+static void call_permission(void *args)
++int au_fhsm_fd(struct super_block *sb, int oflags)
+{
-+ struct permission_args *a = args;
-+ *a->errp = do_vfsub_permission(a->inode, a->mask, a->nd);
-+}
++ int err, fd;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
+
-+int vfsub_permission(struct inode *inode, int mask, struct nameidata *nd,
-+ int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_permission(inode, mask, nd);
-+ else {
-+ int err, wkq_err;
-+ struct permission_args args = {
-+ .errp = &err,
-+ .inode = inode,
-+ .mask = mask,
-+ .nd = nd
-+ };
-+ wkq_err = au_wkq_wait(call_permission, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ return err;
-+ }
-+}
++ err = -EPERM;
++ if (unlikely(!capable(CAP_SYS_ADMIN)))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ err = -EINVAL;
++ if (unlikely(oflags & ~(O_CLOEXEC | O_NONBLOCK)))
++ goto out;
+
-+struct create_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ int mode;
-+ struct nameidata *nd;
-+ struct vfsub_args *vargs;
-+};
++ err = 0;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock(&fhsm->fhsm_spin);
++ if (!fhsm->fhsm_pid)
++ fhsm->fhsm_pid = current->pid;
++ else
++ err = -EBUSY;
++ spin_unlock(&fhsm->fhsm_spin);
++ if (unlikely(err))
++ goto out;
+
-+static void call_create(void *args)
-+{
-+ struct create_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_create(a->dir, a->dentry, a->mode, a->nd);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
-+}
++ oflags |= O_RDONLY;
++ /* oflags |= FMODE_NONOTIFY; */
++ fd = anon_inode_getfd("[aufs_fhsm]", &au_fhsm_fops, sbinfo, oflags);
++ err = fd;
++ if (unlikely(fd < 0))
++ goto out_pid;
+
-+int vfsub_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd, struct vfsub_args *vargs)
-+{
-+ int err;
-+ struct create_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .mode = mode,
-+ .nd = nd,
-+ .vargs = vargs
-+ };
++ /* succeed regardless 'fhsm' status */
++ kobject_get(&sbinfo->si_kobj);
++ si_noflush_read_lock(sb);
++ if (au_ftest_si(sbinfo, FHSM))
++ au_fhsm_wrote_all(sb, /*force*/0);
++ si_read_unlock(sb);
++ goto out; /* success */
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_create(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_create, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
++out_pid:
++ spin_lock(&fhsm->fhsm_spin);
++ fhsm->fhsm_pid = 0;
++ spin_unlock(&fhsm->fhsm_spin);
++out:
++ AuTraceErr(err);
+ return err;
+}
+
-+struct symlink_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ const char *symname;
-+ int mode;
-+ struct vfsub_args *vargs;
-+};
-+
-+static void call_symlink(void *args)
-+{
-+ struct symlink_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_symlink(a->dir, a->dentry, a->symname, a->mode);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
-+}
++/* ---------------------------------------------------------------------- */
+
-+int vfsub_symlink(struct inode *dir, struct dentry *dentry, const char *symname,
-+ int mode, struct vfsub_args *vargs)
++int au_fhsm_br_alloc(struct au_branch *br)
+{
+ int err;
-+ struct symlink_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .symname = symname,
-+ .mode = mode,
-+ .vargs = vargs
-+ };
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_symlink(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_symlink, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
++ err = 0;
++ br->br_fhsm = kmalloc(sizeof(*br->br_fhsm), GFP_NOFS);
++ if (br->br_fhsm)
++ au_br_fhsm_init(br->br_fhsm);
++ else
++ err = -ENOMEM;
++
+ return err;
+}
+
-+struct mknod_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ int mode;
-+ dev_t dev;
-+ struct vfsub_args *vargs;
-+};
++/* ---------------------------------------------------------------------- */
+
-+static void call_mknod(void *args)
++void au_fhsm_fin(struct super_block *sb)
+{
-+ struct mknod_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_mknod(a->dir, a->dentry, a->mode, a->dev);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++ au_fhsm_notify(sb, /*val*/-1);
+}
+
-+int vfsub_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev,
-+ struct vfsub_args *vargs)
++void au_fhsm_init(struct au_sbinfo *sbinfo)
+{
-+ int err;
-+ struct mknod_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .mode = mode,
-+ .dev = dev,
-+ .vargs = vargs
-+ };
++ struct au_fhsm *fhsm;
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_mknod(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_mknod, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
-+ return err;
++ fhsm = &sbinfo->si_fhsm;
++ spin_lock_init(&fhsm->fhsm_spin);
++ init_waitqueue_head(&fhsm->fhsm_wqh);
++ atomic_set(&fhsm->fhsm_readable, 0);
++ fhsm->fhsm_expire
++ = msecs_to_jiffies(AUFS_FHSM_CACHE_DEF_SEC * MSEC_PER_SEC);
++ fhsm->fhsm_bottom = -1;
+}
+
-+struct mkdir_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ int mode;
-+ struct vfsub_args *vargs;
-+};
-+
-+static void call_mkdir(void *args)
++void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec)
+{
-+ struct mkdir_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_mkdir(a->dir, a->dentry, a->mode);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++ sbinfo->si_fhsm.fhsm_expire
++ = msecs_to_jiffies(sec * MSEC_PER_SEC);
+}
+
-+int vfsub_mkdir(struct inode *dir, struct dentry *dentry, int mode,
-+ struct vfsub_args *vargs)
++void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo)
+{
-+ int err;
-+ struct mkdir_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .mode = mode,
-+ .vargs = vargs
-+ };
++ unsigned int u;
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_mkdir(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_mkdir, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
-+ return err;
++ if (!au_ftest_si(sbinfo, FHSM))
++ return;
++
++ u = jiffies_to_msecs(sbinfo->si_fhsm.fhsm_expire) / MSEC_PER_SEC;
++ if (u != AUFS_FHSM_CACHE_DEF_SEC)
++ seq_printf(seq, ",fhsm_sec=%u", u);
+}
+diff -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 2022-12-17 09:21:34.799855195 +0100
+@@ -0,0 +1,860 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * handling file/dir, and address_space operation
++ */
+
-+struct link_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *src_dentry, *dentry;
-+ struct vfsub_args *vargs;
-+};
++#ifdef CONFIG_AUFS_DEBUG
++#include <linux/migrate.h>
++#endif
++#include <linux/pagemap.h>
++#include "aufs.h"
+
-+static void call_link(void *args)
++/* drop flags for writing */
++unsigned int au_file_roflags(unsigned int flags)
+{
-+ struct link_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_link(a->src_dentry, a->dir, a->dentry);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++ flags &= ~(O_WRONLY | O_RDWR | O_APPEND | O_CREAT | O_TRUNC);
++ flags |= O_RDONLY | O_NOATIME;
++ return flags;
+}
+
-+int vfsub_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry, struct vfsub_args *vargs)
++/* common functions to regular file and dir */
++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
++ struct file *file, int force_wr)
+{
++ struct file *h_file;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct path h_path;
+ int err;
-+ struct link_args args = {
-+ .errp = &err,
-+ .src_dentry = src_dentry,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .vargs = vargs
-+ };
-+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_link(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_link, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
-+ return err;
-+}
-+
-+struct rename_args {
-+ int *errp;
-+ struct inode *src_dir, *dir;
-+ struct dentry *src_dentry, *dentry;
-+ struct vfsub_args *vargs;
-+};
+
-+static void call_rename(void *args)
-+{
-+ struct rename_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_rename(a->src_dir, a->src_dentry, a->dir,
-+ a->dentry);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
-+}
++ /* a race condition can happen between open and unlink/rmdir */
++ h_file = ERR_PTR(-ENOENT);
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (au_test_nfsd() && (!h_dentry || d_is_negative(h_dentry)))
++ goto out;
++ h_inode = d_inode(h_dentry);
++ spin_lock(&h_dentry->d_lock);
++ err = (!d_unhashed(dentry) && d_unlinked(h_dentry))
++ /* || !d_inode(dentry)->i_nlink */
++ ;
++ spin_unlock(&h_dentry->d_lock);
++ if (unlikely(err))
++ goto out;
+
-+int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs)
-+{
-+ int err;
-+ struct rename_args args = {
-+ .errp = &err,
-+ .src_dir = src_dir,
-+ .src_dentry = src_dentry,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .vargs = vargs
-+ };
++ sb = dentry->d_sb;
++ br = au_sbr(sb, bindex);
++ err = au_br_test_oflag(flags, br);
++ h_file = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_rename(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_rename, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ /* drop flags for writing */
++ if (au_test_ro(sb, bindex, d_inode(dentry))) {
++ if (force_wr && !(flags & O_WRONLY))
++ force_wr = 0;
++ flags = au_file_roflags(flags);
++ if (force_wr) {
++ h_file = ERR_PTR(-EROFS);
++ flags = au_file_roflags(flags);
++ if (unlikely(vfsub_native_ro(h_inode)
++ || IS_APPEND(h_inode)))
++ goto out;
++ flags &= ~O_ACCMODE;
++ flags |= O_WRONLY;
++ }
+ }
-+ return err;
-+}
++ flags &= ~O_CREAT;
++ au_lcnt_inc(&br->br_nfiles);
++ h_path.dentry = h_dentry;
++ h_path.mnt = au_br_mnt(br);
++ h_file = vfsub_dentry_open(&h_path, flags);
++ if (IS_ERR(h_file))
++ goto out_br;
+
-+struct rmdir_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ struct vfsub_args *vargs;
-+};
++ if (flags & __FMODE_EXEC) {
++ err = deny_write_access(h_file);
++ if (unlikely(err)) {
++ fput(h_file);
++ h_file = ERR_PTR(err);
++ goto out_br;
++ }
++ }
++ fsnotify_open(h_file);
++ goto out; /* success */
+
-+static void call_rmdir(void *args)
-+{
-+ struct rmdir_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_rmdir(a->dir, a->dentry);
-+ if (unlikely(*a->errp))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++out_br:
++ au_lcnt_dec(&br->br_nfiles);
++out:
++ return h_file;
+}
+
-+int vfsub_rmdir(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs)
++static int au_cmoo(struct dentry *dentry)
+{
-+ int err;
-+ struct rmdir_args args = {
-+ .errp = &err,
-+ .dir = dir,
++ int err, cmoo, matched;
++ unsigned int udba;
++ struct path h_path;
++ struct au_pin pin;
++ struct au_cp_generic cpg = {
+ .dentry = dentry,
-+ .vargs = vargs
++ .bdst = -1,
++ .bsrc = -1,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
+ };
++ struct inode *delegated;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct au_fhsm *fhsm;
++ pid_t pid;
++ struct au_branch *br;
++ struct dentry *parent;
++ struct au_hinode *hdir;
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_rmdir(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_rmdir, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
-+ return err;
-+}
++ DiMustWriteLock(dentry);
++ IiMustWriteLock(d_inode(dentry));
+
-+/* ---------------------------------------------------------------------- */
++ err = 0;
++ if (IS_ROOT(dentry))
++ goto out;
++ cpg.bsrc = au_dbtop(dentry);
++ if (!cpg.bsrc)
++ goto out;
+
-+struct read_args {
-+ ssize_t *errp;
-+ struct file *file;
-+ union {
-+ void *kbuf;
-+ char __user *ubuf;
-+ };
-+ size_t count;
-+ loff_t *ppos;
-+};
++ sb = dentry->d_sb;
++ sbinfo = au_sbi(sb);
++ fhsm = &sbinfo->si_fhsm;
++ pid = au_fhsm_pid(fhsm);
++ rcu_read_lock();
++ matched = (pid
++ && (current->pid == pid
++ || rcu_dereference(current->real_parent)->pid == pid));
++ rcu_read_unlock();
++ if (matched)
++ goto out;
+
-+static void call_read_k(void *args)
-+{
-+ struct read_args *a = args;
-+ LKTRTrace("%.*s, cnt %zu, pos %lld\n",
-+ AuDLNPair(a->file->f_dentry), a->count, *a->ppos);
-+ *a->errp = do_vfsub_read_k(a->file, a->kbuf, a->count, a->ppos);
-+}
++ br = au_sbr(sb, cpg.bsrc);
++ cmoo = au_br_cmoo(br->br_perm);
++ if (!cmoo)
++ goto out;
++ if (!d_is_reg(dentry))
++ cmoo &= AuBrAttr_COO_ALL;
++ if (!cmoo)
++ goto out;
+
-+ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
-+ loff_t *ppos, int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_read_u(file, ubuf, count, ppos);
-+ else {
-+ int wkq_err;
-+ ssize_t err, read;
-+ struct read_args args = {
-+ .errp = &err,
-+ .file = file,
-+ .count = count,
-+ .ppos = ppos
-+ };
++ parent = dget_parent(dentry);
++ di_write_lock_parent(parent);
++ err = au_wbr_do_copyup_bu(dentry, cpg.bsrc - 1);
++ cpg.bdst = err;
++ if (unlikely(err < 0)) {
++ err = 0; /* there is no upper writable branch */
++ goto out_dgrade;
++ }
++ AuDbg("bsrc %d, bdst %d\n", cpg.bsrc, cpg.bdst);
+
-+ if (!count)
-+ return 0;
++ /* do not respect the coo attrib for the target branch */
++ err = au_cpup_dirs(dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_dgrade;
+
-+ /*
-+ * workaround an application bug.
-+ * generally, read(2) or write(2) may return the value shorter
-+ * than requested. But many applications don't support it,
-+ * for example bash.
-+ */
-+ err = -ENOMEM;
-+ if (args.count > PAGE_SIZE)
-+ args.count = PAGE_SIZE;
-+ args.kbuf = kmalloc(args.count, GFP_NOFS);
-+ if (unlikely(!args.kbuf))
-+ goto out;
++ di_downgrade_lock(parent, AuLock_IR);
++ udba = au_opt_udba(sb);
++ err = au_pin(&pin, dentry, cpg.bdst, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_parent;
+
-+ read = 0;
-+ do {
-+ wkq_err = au_wkq_wait(call_read_k, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ if (unlikely(err > 0
-+ && copy_to_user(ubuf, args.kbuf, err))) {
-+ err = -EFAULT;
-+ goto out_free;
-+ } else if (!err)
-+ break;
-+ else if (unlikely(err < 0))
-+ goto out_free;
-+ count -= err;
-+ /* do not read too much because of file i/o pointer */
-+ if (count < args.count)
-+ args.count = count;
-+ ubuf += err;
-+ read += err;
-+ } while (count);
-+ smp_mb(); /* flush ubuf */
-+ err = read;
-+
-+ out_free:
-+ kfree(args.kbuf);
-+ out:
-+ return err;
-+ }
-+}
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ if (unlikely(err))
++ goto out_parent;
++ if (!(cmoo & AuBrWAttr_MOO))
++ goto out_parent; /* success */
+
-+ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_read_k(file, kbuf, count, ppos);
-+ else {
-+ ssize_t err;
-+ int wkq_err;
-+ struct read_args args = {
-+ .errp = &err,
-+ .file = file,
-+ .count = count,
-+ .ppos = ppos
-+ };
-+ args.kbuf = kbuf;
-+ wkq_err = au_wkq_wait(call_read_k, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ return err;
-+ }
-+}
++ err = au_pin(&pin, dentry, cpg.bsrc, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_parent;
+
-+struct write_args {
-+ ssize_t *errp;
-+ struct file *file;
-+ union {
-+ void *kbuf;
-+ const char __user *ubuf;
-+ };
-+ size_t count;
-+ loff_t *ppos;
-+ struct vfsub_args *vargs;
-+};
++ h_path.mnt = au_br_mnt(br);
++ h_path.dentry = au_h_dptr(dentry, cpg.bsrc);
++ hdir = au_hi(d_inode(parent), cpg.bsrc);
++ delegated = NULL;
++ err = vfsub_unlink(hdir->hi_inode, &h_path, &delegated, /*force*/1);
++ au_unpin(&pin);
++ /* todo: keep h_dentry or not? */
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ if (unlikely(err)) {
++ pr_err("unlink %pd after coo failed (%d), ignored\n",
++ dentry, err);
++ err = 0;
++ }
++ goto out_parent; /* success */
+
-+static void call_write_k(void *args)
-+{
-+ struct write_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_write_k(a->file, a->kbuf, a->count, a->ppos);
-+ if (unlikely(*a->errp <= 0))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++out_dgrade:
++ di_downgrade_lock(parent, AuLock_IR);
++out_parent:
++ di_read_unlock(parent, AuLock_IR);
++ dput(parent);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count,
-+ loff_t *ppos, struct vfsub_args *vargs)
++int au_do_open(struct file *file, struct au_do_open_args *args)
+{
-+ ssize_t err;
-+
-+ if (!vfsub_ftest(vargs->flags, DLGT)) {
-+ vfsub_ignore(vargs);
-+ err = do_vfsub_write_u(file, ubuf, count, ppos);
-+ if (unlikely(err <= 0))
-+ vfsub_unignore(vargs);
-+ au_dbg_hin_list(vargs);
-+ } else {
-+ ssize_t written;
-+ int wkq_err;
-+ struct write_args args = {
-+ .errp = &err,
-+ .file = file,
-+ .count = count,
-+ .ppos = ppos,
-+ .vargs = vargs
-+ };
-+
-+ if (!count)
-+ return 0;
-+
-+ /*
-+ * workaround an application bug.
-+ * generally, read(2) or write(2) may return the value shorter
-+ * than requested. But many applications don't support it,
-+ * for example bash.
-+ */
-+ err = -ENOMEM;
-+ if (args.count > PAGE_SIZE)
-+ args.count = PAGE_SIZE;
-+ args.kbuf = kmalloc(args.count, GFP_NOFS);
-+ if (unlikely(!args.kbuf))
-+ goto out;
++ int err, aopen = args->aopen;
++ struct dentry *dentry;
++ struct au_finfo *finfo;
+
-+ written = 0;
-+ do {
-+ if (unlikely(copy_from_user(args.kbuf, ubuf,
-+ args.count))) {
-+ err = -EFAULT;
-+ goto out_free;
-+ }
++ if (!aopen)
++ err = au_finfo_init(file, args->fidir);
++ else {
++ lockdep_off();
++ err = au_finfo_init(file, args->fidir);
++ lockdep_on();
++ }
++ if (unlikely(err))
++ goto out;
+
-+ wkq_err = au_wkq_wait(call_write_k, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ if (err > 0) {
-+ count -= err;
-+ if (count < args.count)
-+ args.count = count;
-+ ubuf += err;
-+ written += err;
-+ } else if (!err)
-+ break;
-+ else if (unlikely(err < 0))
-+ goto out_free;
-+ } while (count);
-+ err = written;
++ dentry = file->f_path.dentry;
++ AuDebugOn(IS_ERR_OR_NULL(dentry));
++ di_write_lock_child(dentry);
++ err = au_cmoo(dentry);
++ if (!err) {
++ if (!aopen) {
++ err = args->open(file, vfsub_file_flags(file),
++ au_di(dentry)->di_htmpfile);
++ di_write_unlock(dentry);
++ } else {
++ di_downgrade_lock(dentry, AuLock_IR);
++ lockdep_off();
++ err = args->open(file, vfsub_file_flags(file),
++ args->h_file);
++ lockdep_on();
++ di_read_unlock(dentry, AuLock_IR);
++ }
++ }
+
-+ out_free:
-+ kfree(args.kbuf);
++ finfo = au_fi(file);
++ if (!err) {
++ finfo->fi_file = file;
++ au_hbl_add(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
++ }
++ if (!aopen)
++ fi_write_unlock(file);
++ else {
++ lockdep_off();
++ fi_write_unlock(file);
++ lockdep_on();
++ }
++ if (unlikely(err)) {
++ finfo->fi_hdir = NULL;
++ au_finfo_fin(file);
+ }
-+ out:
++
++out:
++ AuTraceErr(err);
+ return err;
+}
+
-+ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ struct vfsub_args *vargs)
++int au_reopen_nondir(struct file *file)
+{
-+ ssize_t err;
-+ struct write_args args = {
-+ .errp = &err,
-+ .file = file,
-+ .count = count,
-+ .ppos = ppos,
-+ .vargs = vargs
-+ };
++ int err;
++ aufs_bindex_t btop;
++ struct dentry *dentry;
++ struct au_branch *br;
++ struct file *h_file, *h_file_tmp;
+
-+ args.kbuf = kbuf;
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_write_k(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_write_k, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ dentry = file->f_path.dentry;
++ btop = au_dbtop(dentry);
++ br = au_sbr(dentry->d_sb, btop);
++ h_file_tmp = NULL;
++ if (au_fbtop(file) == btop) {
++ h_file = au_hf_top(file);
++ if (file->f_mode == h_file->f_mode)
++ return 0; /* success */
++ h_file_tmp = h_file;
++ get_file(h_file_tmp);
++ au_lcnt_inc(&br->br_nfiles);
++ au_set_h_fptr(file, btop, NULL);
+ }
-+ return err;
-+}
++ AuDebugOn(au_fi(file)->fi_hdir);
++ /*
++ * it can happen
++ * file exists on both of rw and ro
++ * open --> dbtop and fbtop are both 0
++ * prepend a branch as rw, "rw" become ro
++ * remove rw/file
++ * delete the top branch, "rw" becomes rw again
++ * --> dbtop is 1, fbtop is still 0
++ * write --> fbtop is 0 but dbtop is 1
++ */
++ /* AuDebugOn(au_fbtop(file) < btop); */
+
-+struct readdir_args {
-+ int *errp;
-+ struct file *file;
-+ filldir_t filldir;
-+ void *arg;
-+};
++ h_file = au_h_open(dentry, btop, vfsub_file_flags(file) & ~O_TRUNC,
++ file, /*force_wr*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file)) {
++ if (h_file_tmp) {
++ /* revert */
++ au_set_h_fptr(file, btop, h_file_tmp);
++ h_file_tmp = NULL;
++ }
++ goto out; /* todo: close all? */
++ }
+
-+static void call_readdir(void *args)
-+{
-+ struct readdir_args *a = args;
-+ *a->errp = do_vfsub_readdir(a->file, a->filldir, a->arg);
-+}
++ err = 0;
++ au_set_fbtop(file, btop);
++ au_set_h_fptr(file, btop, h_file);
++ au_update_figen(file);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
+
-+int vfsub_readdir(struct file *file, filldir_t filldir, void *arg, int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_readdir(file, filldir, arg);
-+ else {
-+ int err, wkq_err;
-+ struct readdir_args args = {
-+ .errp = &err,
-+ .file = file,
-+ .filldir = filldir,
-+ .arg = arg
-+ };
-+ wkq_err = au_wkq_wait(call_readdir, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ return err;
++out:
++ if (h_file_tmp) {
++ fput(h_file_tmp);
++ au_lcnt_dec(&br->br_nfiles);
+ }
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct splice_to_args {
-+ long *errp;
-+ struct file *in;
-+ loff_t *ppos;
-+ struct pipe_inode_info *pipe;
-+ size_t len;
-+ unsigned int flags;
-+};
-+
-+static void call_splice_to(void *args)
++static int au_reopen_wh(struct file *file, aufs_bindex_t btgt,
++ struct dentry *hi_wh)
+{
-+ struct splice_to_args *a = args;
-+ *a->errp = do_vfsub_splice_to(a->in, a->ppos, a->pipe, a->len,
-+ a->flags);
-+}
++ int err;
++ aufs_bindex_t btop;
++ struct au_dinfo *dinfo;
++ struct dentry *h_dentry;
++ struct au_hdentry *hdp;
+
-+long vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags, int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_splice_to(in, ppos, pipe, len, flags);
-+ else {
-+ long err;
-+ int wkq_err;
-+ struct splice_to_args args = {
-+ .errp = &err,
-+ .in = in,
-+ .ppos = ppos,
-+ .pipe = pipe,
-+ .len = len,
-+ .flags = flags
-+ };
-+ wkq_err = au_wkq_wait(call_splice_to, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ return err;
-+ }
-+}
++ dinfo = au_di(file->f_path.dentry);
++ AuRwMustWriteLock(&dinfo->di_rwsem);
+
-+struct splice_from_args {
-+ long *errp;
-+ struct pipe_inode_info *pipe;
-+ struct file *out;
-+ loff_t *ppos;
-+ size_t len;
-+ unsigned int flags;
-+ struct vfsub_args *vargs;
-+};
++ btop = dinfo->di_btop;
++ dinfo->di_btop = btgt;
++ hdp = au_hdentry(dinfo, btgt);
++ h_dentry = hdp->hd_dentry;
++ hdp->hd_dentry = hi_wh;
++ err = au_reopen_nondir(file);
++ hdp->hd_dentry = h_dentry;
++ dinfo->di_btop = btop;
+
-+static void call_splice_from(void *args)
-+{
-+ struct splice_from_args *a = args;
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_splice_from(a->pipe, a->out, a->ppos, a->len,
-+ a->flags);
-+ if (unlikely(*a->errp < 0))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
++ return err;
+}
+
-+long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags,
-+ struct vfsub_args *vargs)
++static int au_ready_to_write_wh(struct file *file, loff_t len,
++ aufs_bindex_t bcpup, struct au_pin *pin)
+{
-+ long err;
-+ struct splice_from_args args = {
-+ .errp = &err,
-+ .pipe = pipe,
-+ .out = out,
-+ .ppos = ppos,
++ int err;
++ struct inode *inode, *h_inode;
++ struct dentry *h_dentry, *hi_wh;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = bcpup,
++ .bsrc = -1,
+ .len = len,
-+ .flags = flags,
-+ .vargs = vargs
++ .pin = pin
+ };
+
-+ if (!vfsub_ftest(vargs->flags, DLGT))
-+ call_splice_from(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_splice_from, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ au_update_dbtop(cpg.dentry);
++ inode = d_inode(cpg.dentry);
++ h_inode = NULL;
++ if (au_dbtop(cpg.dentry) <= bcpup
++ && au_dbbot(cpg.dentry) >= bcpup) {
++ h_dentry = au_h_dptr(cpg.dentry, bcpup);
++ if (h_dentry && d_is_positive(h_dentry))
++ h_inode = d_inode(h_dentry);
+ }
++ hi_wh = au_hi_wh(inode, bcpup);
++ if (!hi_wh && !h_inode)
++ err = au_sio_cpup_wh(&cpg, file);
++ else
++ /* already copied-up after unlink */
++ err = au_reopen_wh(file, bcpup, hi_wh);
++
++ if (!err
++ && (inode->i_nlink > 1
++ || (inode->i_state & I_LINKABLE))
++ && au_opt_test(au_mntflags(cpg.dentry->d_sb), PLINK))
++ au_plink_append(inode, bcpup, au_h_dptr(cpg.dentry, bcpup));
++
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * prepare the @file for writing.
++ */
++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin)
++{
++ int err;
++ aufs_bindex_t dbtop;
++ struct dentry *parent;
++ struct inode *inode;
++ struct super_block *sb;
++ struct file *h_file;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = -1,
++ .bsrc = -1,
++ .len = len,
++ .pin = pin,
++ .flags = AuCpup_DTIME
++ };
+
-+struct getattr_args {
-+ int *errp;
-+ struct vfsmount *mnt;
-+ struct dentry *dentry;
-+ struct kstat *st;
-+};
++ sb = cpg.dentry->d_sb;
++ inode = d_inode(cpg.dentry);
++ cpg.bsrc = au_fbtop(file);
++ err = au_test_ro(sb, cpg.bsrc, inode);
++ if (!err && (au_hf_top(file)->f_mode & FMODE_WRITE)) {
++ err = au_pin(pin, cpg.dentry, cpg.bsrc, AuOpt_UDBA_NONE,
++ /*flags*/0);
++ goto out;
++ }
+
-+static void call_getattr(void *args)
-+{
-+ struct getattr_args *a = args;
-+ *a->errp = do_vfsub_getattr(a->mnt, a->dentry, a->st);
-+}
++ /* need to cpup or reopen */
++ parent = dget_parent(cpg.dentry);
++ di_write_lock_parent(parent);
++ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
++ cpg.bdst = err;
++ if (unlikely(err < 0))
++ goto out_dgrade;
++ err = 0;
+
-+int vfsub_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *st,
-+ int dlgt)
-+{
-+ if (!dlgt)
-+ return do_vfsub_getattr(mnt, dentry, st);
-+ else {
-+ int err, wkq_err;
-+ struct getattr_args args = {
-+ .errp = &err,
-+ .mnt = mnt,
-+ .dentry = dentry,
-+ .st = st
-+ };
-+ wkq_err = au_wkq_wait(call_getattr, &args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ return err;
++ if (!d_unhashed(cpg.dentry) && !au_h_dptr(parent, cpg.bdst)) {
++ err = au_cpup_dirs(cpg.dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_dgrade;
+ }
-+}
-diff -urN linux/fs/aufs/hinotify.c linux-aufs/fs/aufs/hinotify.c
---- linux/fs/aufs/hinotify.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/hinotify.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1146 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * internal/hidden inotify handler
-+ *
-+ * $Id$
-+ */
-+
-+#include "aufs.h"
-+
-+/*
-+#ifdef DbgInotify
-+#define AuDbgHin(args...) AuDbg(##args)
-+#else
-+#define AuDbgHin(args...) do {} while ()
-+#endif
-+*/
-+
-+static struct inotify_handle *in_handle;
-+
-+AuCacheFuncs(hinotify, AuCache_HINOTIFY);
+
-+int au_hin_alloc(struct au_hinode *hinode, struct inode *inode,
-+ struct inode *h_inode)
-+{
-+ int err;
-+ struct au_hinotify *hin;
-+ s32 wd;
++ err = au_pin(pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_dgrade;
+
-+ LKTRTrace("i%lu, hi%lu\n", inode->i_ino, h_inode->i_ino);
++ dbtop = au_dbtop(cpg.dentry);
++ if (dbtop <= cpg.bdst)
++ cpg.bsrc = cpg.bdst;
+
-+ err = -ENOMEM;
-+ hin = au_cache_alloc_hinotify();
-+ if (hin) {
-+ AuDebugOn(hinode->hi_notify);
-+ hinode->hi_notify = hin;
-+ au_spl_init(&hin->hin_ignore);
-+ hin->hin_aufs_inode = inode;
-+
-+ inotify_init_watch(&hin->hin_watch);
-+ wd = inotify_add_watch(in_handle, &hin->hin_watch, h_inode,
-+ AuInMask);
-+ if (wd >= 0)
-+ return 0; /* success */
++ if (dbtop <= cpg.bdst /* just reopen */
++ || !d_unhashed(cpg.dentry) /* copyup and reopen */
++ ) {
++ h_file = au_h_open_pre(cpg.dentry, cpg.bsrc, /*force_wr*/0);
++ if (IS_ERR(h_file))
++ err = PTR_ERR(h_file);
++ else {
++ di_downgrade_lock(parent, AuLock_IR);
++ if (dbtop > cpg.bdst)
++ err = au_sio_cpup_simple(&cpg);
++ if (!err)
++ err = au_reopen_nondir(file);
++ au_h_open_post(cpg.dentry, cpg.bsrc, h_file);
++ }
++ } else { /* copyup as wh and reopen */
++ /*
++ * since writable hfsplus branch is not supported,
++ * h_open_pre/post() are unnecessary.
++ */
++ err = au_ready_to_write_wh(file, len, cpg.bdst, pin);
++ di_downgrade_lock(parent, AuLock_IR);
++ }
+
-+ err = wd;
-+ put_inotify_watch(&hin->hin_watch);
-+ au_cache_free_hinotify(hin);
-+ hinode->hi_notify = NULL;
++ if (!err) {
++ au_pin_set_parent_lflag(pin, /*lflag*/0);
++ goto out_dput; /* success */
+ }
++ au_unpin(pin);
++ goto out_unlock;
+
-+ AuTraceErr(err);
++out_dgrade:
++ di_downgrade_lock(parent, AuLock_IR);
++out_unlock:
++ di_read_unlock(parent, AuLock_IR);
++out_dput:
++ dput(parent);
++out:
+ return err;
+}
+
-+void au_hin_free(struct au_hinode *hinode)
-+{
-+ int err;
-+ struct au_hinotify *hin;
-+
-+ AuTraceEnter();
-+
-+ hin = hinode->hi_notify;
-+ if (hin) {
-+ err = 0;
-+ if (atomic_read(&hin->hin_watch.count))
-+ err = inotify_rm_watch(in_handle, &hin->hin_watch);
-+ if (unlikely(err))
-+ /* it means the watch is already removed */
-+ LKTRTrace("failed inotify_rm_watch() %d\n", err);
-+ au_cache_free_hinotify(hin);
-+ hinode->hi_notify = NULL;
-+ }
-+}
-+
+/* ---------------------------------------------------------------------- */
+
-+void au_hin_ctl(struct au_hinode *hinode, const __u32 mask)
++int au_do_flush(struct file *file, fl_owner_t id,
++ int (*flush)(struct file *file, fl_owner_t id))
+{
-+ struct inode *h_inode;
-+ struct inotify_watch *watch;
-+
-+ h_inode = hinode->hi_inode;
-+ LKTRTrace("hi%lu, sb %p, 0x%x\n", h_inode->i_ino, h_inode->i_sb, mask);
-+ IMustLock(h_inode);
-+ if (!hinode->hi_notify)
-+ return;
-+
-+ watch = &hinode->hi_notify->hin_watch;
-+#if 0 /* reserved for future use */
-+ {
-+ u32 wd;
-+ wd = inotify_find_update_watch(in_handle, h_inode, mask);
-+ AuTraceErr(wd);
-+ /* ignore an err; */
-+ }
-+#else
-+ /* struct inotify_handle is hidden */
-+ mutex_lock(&h_inode->inotify_mutex);
-+ /* mutex_lock(&watch->ih->mutex); */
-+ watch->mask = mask;
-+ /* mutex_unlock(&watch->ih->mutex); */
-+ mutex_unlock(&h_inode->inotify_mutex);
-+#endif
-+ LKTRTrace("watch %p, mask %u\n", watch, watch->mask);
-+}
++ int err;
++ struct super_block *sb;
++ struct inode *inode;
+
-+void au_reset_hinotify(struct inode *inode, unsigned int flags)
-+{
-+ aufs_bindex_t bindex, bend;
-+ struct inode *hi;
-+ struct dentry *iwhdentry;
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ si_noflush_read_lock(sb);
++ fi_read_lock(file);
++ ii_read_lock_child(inode);
+
-+ LKTRTrace("i%lu, 0x%x\n", inode->i_ino, flags);
++ err = flush(file, id);
++ au_cpup_attr_timesizes(inode);
+
-+ bend = au_ibend(inode);
-+ for (bindex = au_ibstart(inode); bindex <= bend; bindex++) {
-+ hi = au_h_iptr(inode, bindex);
-+ if (hi) {
-+ /* mutex_lock_nested(&hi->i_mutex, AuLsc_I_CHILD); */
-+ iwhdentry = au_hi_wh(inode, bindex);
-+ if (iwhdentry)
-+ dget(iwhdentry);
-+ au_igrab(hi);
-+ au_set_h_iptr(inode, bindex, NULL, 0);
-+ au_set_h_iptr(inode, bindex, au_igrab(hi),
-+ flags & ~AuHi_XINO);
-+ iput(hi);
-+ dput(iwhdentry);
-+ /* mutex_unlock(&hi->i_mutex); */
-+ }
-+ }
++ ii_read_unlock(inode);
++ fi_read_unlock(file);
++ si_read_unlock(sb);
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+void au_unpin_gp(struct au_pin *args)
-+{
-+ struct au_pin1 *gp;
-+
-+ gp = au_pin_gp(args);
-+ AuDebugOn(!gp);
-+ if (gp->dentry)
-+ LKTRTrace("%.*s\n", AuDLNPair(gp->dentry));
-+ else
-+ AuTraceEnter();
-+
-+ au_do_unpin(gp, NULL);
-+}
-+
-+int au_hin_verify_gen(struct dentry *dentry)
++static int au_file_refresh_by_inode(struct file *file, int *need_reopen)
+{
-+ struct super_block *sb = dentry->d_sb;
-+ au_gen_t sigen;
++ int err;
++ struct au_pin pin;
++ struct au_finfo *finfo;
++ struct dentry *parent, *hi_wh;
+ struct inode *inode;
++ struct super_block *sb;
++ struct au_cp_generic cpg = {
++ .dentry = file->f_path.dentry,
++ .bdst = -1,
++ .bsrc = -1,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME
++ };
+
-+ if (!au_opt_test(au_mntflags(sb), UDBA_INOTIFY))
-+ return 0;
++ FiMustWriteLock(file);
+
-+ sigen = au_sigen(dentry->d_sb);
-+ inode = dentry->d_inode;
-+ return au_digen(dentry) != sigen
-+ || (inode && au_iigen(inode) != sigen);
-+}
++ err = 0;
++ finfo = au_fi(file);
++ sb = cpg.dentry->d_sb;
++ inode = d_inode(cpg.dentry);
++ cpg.bdst = au_ibtop(inode);
++ if (cpg.bdst == finfo->fi_btop || IS_ROOT(cpg.dentry))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ parent = dget_parent(cpg.dentry);
++ if (au_test_ro(sb, cpg.bdst, inode)) {
++ di_read_lock_parent(parent, !AuLock_IR);
++ err = AuWbrCopyup(au_sbi(sb), cpg.dentry);
++ cpg.bdst = err;
++ di_read_unlock(parent, !AuLock_IR);
++ if (unlikely(err < 0))
++ goto out_parent;
++ err = 0;
++ }
+
-+/* cf. fsnotify_change() */
-+__u32 vfsub_events_notify_change(struct iattr *ia)
-+{
-+ __u32 events;
-+ const unsigned int amtime = (ATTR_ATIME | ATTR_MTIME);
++ di_read_lock_parent(parent, AuLock_IR);
++ hi_wh = au_hi_wh(inode, cpg.bdst);
++ if (!S_ISDIR(inode->i_mode)
++ && au_opt_test(au_mntflags(sb), PLINK)
++ && au_plink_test(inode)
++ && !d_unhashed(cpg.dentry)
++ && cpg.bdst < au_dbtop(cpg.dentry)) {
++ err = au_test_and_cpup_dirs(cpg.dentry, cpg.bdst);
++ if (unlikely(err))
++ goto out_unlock;
++
++ /* always superio. */
++ err = au_pin(&pin, cpg.dentry, cpg.bdst, AuOpt_UDBA_NONE,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (!err) {
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ }
++ } else if (hi_wh) {
++ /* already copied-up after unlink */
++ err = au_reopen_wh(file, cpg.bdst, hi_wh);
++ *need_reopen = 0;
++ }
+
-+ events = 0;
-+ if ((ia->ia_valid & (ATTR_UID | ATTR_GID | ATTR_MODE))
-+ || (ia->ia_valid & amtime) == amtime)
-+ events |= IN_ATTRIB;
-+ if ((ia->ia_valid & ATTR_SIZE)
-+ || (ia->ia_valid & amtime) == ATTR_MTIME)
-+ events |= IN_MODIFY;
-+ return events;
++out_unlock:
++ di_read_unlock(parent, AuLock_IR);
++out_parent:
++ dput(parent);
++out:
++ return err;
+}
+
-+void vfsub_ign_hinode(struct vfsub_args *vargs, __u32 events,
-+ struct au_hinode *hinode)
++static void au_do_refresh_dir(struct file *file)
+{
-+ struct au_hinotify *hin;
++ aufs_bindex_t bindex, bbot, new_bindex, brid;
++ struct au_hfile *p, tmp, *q;
++ struct au_finfo *finfo;
+ struct super_block *sb;
-+ struct au_hin_ignore *ign;
++ struct au_fidir *fidir;
+
-+ if (!hinode)
-+ return;
++ FiMustWriteLock(file);
+
-+ hin = hinode->hi_notify;
-+ if (!hin || !hin->hin_watch.mask)
-+ return;
++ sb = file->f_path.dentry->d_sb;
++ finfo = au_fi(file);
++ fidir = finfo->fi_hdir;
++ AuDebugOn(!fidir);
++ p = fidir->fd_hfile + finfo->fi_btop;
++ brid = p->hf_br->br_id;
++ bbot = fidir->fd_bbot;
++ for (bindex = finfo->fi_btop; bindex <= bbot; bindex++, p++) {
++ if (!p->hf_file)
++ continue;
+
-+ sb = hin->hin_aufs_inode->i_sb;
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), UDBA_INOTIFY));
++ new_bindex = au_br_index(sb, p->hf_br->br_id);
++ if (new_bindex == bindex)
++ continue;
++ if (new_bindex < 0) {
++ au_set_h_fptr(file, bindex, NULL);
++ continue;
++ }
+
-+ ign = vargs->ignore + vargs->nignore++;
-+ ign->ign_events = events;
-+ ign->ign_handled = 0;
-+ ign->ign_hinode = hinode;
++ /* swap two lower inode, and loop again */
++ q = fidir->fd_hfile + new_bindex;
++ tmp = *q;
++ *q = *p;
++ *p = tmp;
++ if (tmp.hf_file) {
++ bindex--;
++ p--;
++ }
++ }
+
-+ {
-+ struct inode *h_inode;
-+ h_inode = hinode->hi_inode;
-+ if (!mutex_is_locked(&h_inode->i_mutex))
-+ au_dbg_blocked();
-+ IMustLock(h_inode);
++ p = fidir->fd_hfile;
++ if (!au_test_mmapped(file) && !d_unlinked(file->f_path.dentry)) {
++ bbot = au_sbbot(sb);
++ for (finfo->fi_btop = 0; finfo->fi_btop <= bbot;
++ finfo->fi_btop++, p++)
++ if (p->hf_file) {
++ if (file_inode(p->hf_file))
++ break;
++ au_hfput(p, /*execed*/0);
++ }
++ } else {
++ bbot = au_br_index(sb, brid);
++ for (finfo->fi_btop = 0; finfo->fi_btop < bbot;
++ finfo->fi_btop++, p++)
++ if (p->hf_file)
++ au_hfput(p, /*execed*/0);
++ bbot = au_sbbot(sb);
+ }
++
++ p = fidir->fd_hfile + bbot;
++ for (fidir->fd_bbot = bbot; fidir->fd_bbot >= finfo->fi_btop;
++ fidir->fd_bbot--, p--)
++ if (p->hf_file) {
++ if (file_inode(p->hf_file))
++ break;
++ au_hfput(p, /*execed*/0);
++ }
++ AuDebugOn(fidir->fd_bbot < finfo->fi_btop);
+}
+
-+static void au_hin_ignore(struct au_hin_ignore *ign)
++/*
++ * after branch manipulating, refresh the file.
++ */
++static int refresh_file(struct file *file, int (*reopen)(struct file *file))
+{
-+ struct au_hinode *hinode;
-+ __u32 events;
-+ struct au_hinotify *hin;
-+ struct inode *h_inode;
++ 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;
+
-+ hinode = ign->ign_hinode;
-+ events = ign->ign_events;
-+ LKTRTrace("0x%x\n", events);
-+ AuDebugOn(!hinode || !events);
++ dentry = file->f_path.dentry;
++ sb = dentry->d_sb;
++ nbr = au_sbbot(sb) + 1;
++ finfo = au_fi(file);
++ if (!finfo->fi_hdir) {
++ hfile = &finfo->fi_htop;
++ AuDebugOn(!hfile->hf_file);
++ bindex = au_br_index(sb, hfile->hf_br->br_id);
++ AuDebugOn(bindex < 0);
++ if (bindex != finfo->fi_btop)
++ au_set_fbtop(file, bindex);
++ } else {
++ err = au_fidir_realloc(finfo, nbr, /*may_shrink*/0);
++ if (unlikely(err))
++ goto out;
++ au_do_refresh_dir(file);
++ }
+
-+ hin = hinode->hi_notify;
-+ h_inode = hinode->hi_inode;
-+ if (h_inode && hin) {
-+ LKTRTrace("hi%lu\n", h_inode->i_ino);
-+#ifdef DbgInotify
-+ AuDbg("hi%lu, 0x%x\n", h_inode->i_ino, events);
-+#endif
++ err = 0;
++ need_reopen = 1;
++ if (!au_test_mmapped(file))
++ err = au_file_refresh_by_inode(file, &need_reopen);
++ if (finfo->fi_hdir)
++ /* harmless if err */
++ au_fidir_realloc(finfo, nbr, /*may_shrink*/1);
++ if (!err && need_reopen && !d_unlinked(dentry))
++ err = reopen(file);
++ if (!err) {
++ au_update_figen(file);
++ goto out; /* success */
++ }
+
-+ au_spl_add(&ign->ign_list, &hin->hin_ignore);
-+ /* AuDbg("list_add %p, 0x%x\n", ign, events); */
++ /* error, close all lower files */
++ if (finfo->fi_hdir) {
++ bbot = au_fbbot_dir(file);
++ for (bindex = au_fbtop(file); bindex <= bbot; bindex++)
++ au_set_h_fptr(file, bindex, NULL);
+ }
-+#if 1 /* todo: test dlgt */
-+ else
-+ /*
-+ * it may happen by this scenario.
-+ * - a file and its parent dir exist on two branches
-+ * - a file on the upper branch is opened
-+ * - the parent dir and the file are removed by udba
-+ * - the parent is re-accessed, and new dentry/inode in
-+ * aufs is generated for it, based upon the one on the lower
-+ * branch
-+ * - the opened file is re-accessed, re-validated, and it may be
-+ * re-connected to the new parent dentry
-+ * it means the file in aufs cannot get the actual removed
-+ * parent dir on the branch.
-+ */
-+ INIT_LIST_HEAD(&ign->ign_list);
-+#endif
++
++out:
++ return err;
+}
+
-+static void au_hin_unignore(struct au_hin_ignore *ign)
++/* common function to regular file and dir */
++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
++ int wlock, unsigned int fi_lsc)
+{
-+ struct au_hinode *hinode;
-+ __u32 events;
-+ struct au_hinotify *hin;
-+ struct inode *h_inode;
-+
-+ hinode = ign->ign_hinode;
-+ events = ign->ign_events;
-+ LKTRTrace("0x%x\n", events);
-+ /* AuDbg("0x%x\n", events); */
-+ AuDebugOn(!hinode || !events);
-+
-+ hin = hinode->hi_notify;
-+ h_inode = hinode->hi_inode;
-+ if (!h_inode || !hin)
-+ return;
-+ LKTRTrace("hi%lu\n", h_inode->i_ino);
-+#ifdef DbgInotify
-+ AuDbg("hi%lu, 0x%x\n", h_inode->i_ino, events);
-+#endif
++ int err;
++ unsigned int sigen, figen;
++ aufs_bindex_t btop;
++ unsigned char pseudo_link;
++ struct dentry *dentry;
++ struct inode *inode;
+
-+ spin_lock(&hin->hin_ignore.spin);
-+ au_hin_list_del(&ign->ign_list);
-+ spin_unlock(&hin->hin_ignore.spin);
-+ /* AuDbg("list_del %p, 0x%x\n", ign, events); */
-+}
++ err = 0;
++ dentry = file->f_path.dentry;
++ inode = d_inode(dentry);
++ sigen = au_sigen(dentry->d_sb);
++ fi_write_lock_nested(file, fi_lsc);
++ figen = au_figen(file);
++ if (!fi_lsc)
++ di_write_lock_child(dentry);
++ else
++ di_write_lock_child2(dentry);
++ btop = au_dbtop(dentry);
++ pseudo_link = (btop != au_ibtop(inode));
++ if (sigen == figen && !pseudo_link && au_fbtop(file) == btop) {
++ if (!wlock) {
++ di_downgrade_lock(dentry, AuLock_IR);
++ fi_downgrade_lock(file);
++ }
++ goto out; /* success */
++ }
+
-+static int au_hin_test_ignore(u32 mask, struct au_hinotify *hin)
-+{
-+ int do_ignore;
-+ u32 events;
-+ struct au_hin_ignore *ign, *tmp;
-+ struct list_head *head;
++ AuDbg("sigen %d, figen %d\n", sigen, figen);
++ if (au_digen_test(dentry, sigen)) {
++ err = au_reval_dpath(dentry, sigen);
++ AuDebugOn(!err && au_digen_test(dentry, sigen));
++ }
+
-+ do_ignore = 0;
-+ head = &hin->hin_ignore.head;
-+ spin_lock(&hin->hin_ignore.spin);
-+ list_for_each_entry_safe(ign, tmp, head, ign_list) {
-+ /* AuDbg("ign %p\n", ign); */
-+ if (ign->ign_pid == current->pid) {
-+ events = (mask & ign->ign_events);
-+ if (events) {
-+ do_ignore = 1;
-+ ign->ign_handled |= events;
-+ if (ign->ign_events == ign->ign_handled) {
-+ list_del_init(&ign->ign_list);
-+ /*
-+ AuDbg("list_del %p, 0x%x\n",
-+ ign, events);
-+ */
-+ }
-+ break;
-+ }
++ if (!err)
++ err = refresh_file(file, reopen);
++ if (!err) {
++ if (!wlock) {
++ di_downgrade_lock(dentry, AuLock_IR);
++ fi_downgrade_lock(file);
+ }
++ } else {
++ di_write_unlock(dentry);
++ fi_write_unlock(file);
+ }
-+ spin_unlock(&hin->hin_ignore.spin);
+
-+ return do_ignore;
++out:
++ return err;
+}
+
-+void vfsub_ignore(struct vfsub_args *vargs)
++/* ---------------------------------------------------------------------- */
++
++/* cf. aufs_nopage() */
++/* for madvise(2) */
++static int aufs_read_folio(struct file *file __maybe_unused, struct folio *folio)
+{
-+ int n;
-+ struct au_hin_ignore *ign;
-+ struct super_block *sb;
-+ struct au_hinode *hinode;
-+ struct inode *h_inode;
++ folio_unlock(folio);
++ return 0;
++}
+
-+ n = vargs->nignore;
-+ if (!n)
-+ return;
++/* it will never be called, but necessary to support O_DIRECT */
++static ssize_t aufs_direct_IO(struct kiocb *iocb, struct iov_iter *iter)
++{ BUG(); return 0; }
+
-+ ign = vargs->ignore;
-+ hinode = ign->ign_hinode;
-+ sb = hinode->hi_notify->hin_aufs_inode->i_sb;
-+ h_inode = hinode->hi_inode;
-+ if (au_opt_test(au_mntflags(sb), UDBA_INOTIFY)) {
-+ if (!mutex_is_locked(&h_inode->i_mutex))
-+ au_dbg_blocked();
-+ IMustLock(h_inode);
-+ }
-+ while (n-- > 0) {
-+ ign->ign_pid = current->pid;
-+ au_hin_ignore(ign++);
-+ }
-+}
++/* they will never be called. */
++#ifdef CONFIG_AUFS_DEBUG
++static int aufs_write_begin(struct file *file, struct address_space *mapping,
++ loff_t pos, unsigned len,
++ struct page **pagep, void **fsdata)
++{ AuUnsupport(); return 0; }
++static int aufs_write_end(struct file *file, struct address_space *mapping,
++ loff_t pos, unsigned len, unsigned copied,
++ struct page *page, void *fsdata)
++{ AuUnsupport(); return 0; }
++static int aufs_writepage(struct page *page, struct writeback_control *wbc)
++{ AuUnsupport(); return 0; }
+
-+void vfsub_unignore(struct vfsub_args *vargs)
-+{
-+ int n;
-+ struct au_hin_ignore *ign;
++static bool aufs_dirty_folio(struct address_space *mapping, struct folio *folio)
++{ AuUnsupport(); return true; }
++static void aufs_invalidate_folio(struct folio *folio, size_t offset, size_t len)
++{ AuUnsupport(); }
++static bool aufs_release_folio(struct folio *folio, gfp_t gfp)
++{ AuUnsupport(); return true; }
++#if 0 /* called by memory compaction regardless file */
++static int aufs_migrate_folio(struct address_space *mapping, struct folio *dst,
++ struct folio *src, enum migrate_mode mode)
++{ AuUnsupport(); return 0; }
++#endif
++static int aufs_launder_folio(struct folio *folio)
++{ AuUnsupport(); return 0; }
++static bool aufs_is_partially_uptodate(struct folio *folio, size_t from,
++ size_t count)
++{ AuUnsupport(); return true; }
++static void aufs_is_dirty_writeback(struct folio *folio, bool *dirty,
++ bool *writeback)
++{ AuUnsupport(); }
++static int aufs_error_remove_page(struct address_space *mapping,
++ struct page *page)
++{ AuUnsupport(); return 0; }
++static int aufs_swap_activate(struct swap_info_struct *sis, struct file *file,
++ sector_t *span)
++{ AuUnsupport(); return 0; }
++static void aufs_swap_deactivate(struct file *file)
++{ AuUnsupport(); }
++static int aufs_swap_rw(struct kiocb *iocb, struct iov_iter *iter)
++{ AuUnsupport(); return 0; }
++#endif /* CONFIG_AUFS_DEBUG */
+
-+ n = vargs->nignore;
-+ if (!n)
-+ return;
++const struct address_space_operations aufs_aop = {
++ .read_folio = aufs_read_folio,
++ .direct_IO = aufs_direct_IO,
++#ifdef CONFIG_AUFS_DEBUG
++ .writepage = aufs_writepage,
++ /* no writepages, because of writepage */
++ .dirty_folio = aufs_dirty_folio,
++ /* no readpages, because of readpage */
++ .write_begin = aufs_write_begin,
++ .write_end = aufs_write_end,
++ /* no bmap, no block device */
++ .invalidate_folio = aufs_invalidate_folio,
++ .release_folio = aufs_release_folio,
++ /* is fallback_migrate_page ok? */
++ /* .migrate_folio = aufs_migrate_folio, */
++ .launder_folio = aufs_launder_folio,
++ .is_partially_uptodate = aufs_is_partially_uptodate,
++ .is_dirty_writeback = aufs_is_dirty_writeback,
++ .error_remove_page = aufs_error_remove_page,
++ .swap_activate = aufs_swap_activate,
++ .swap_deactivate = aufs_swap_deactivate,
++ .swap_rw = aufs_swap_rw
++#endif /* CONFIG_AUFS_DEBUG */
++};
+diff -urN /usr/share/empty/fs/aufs/file.h linux/fs/aufs/file.h
+--- /usr/share/empty/fs/aufs/file.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/file.h 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,342 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+ ign = vargs->ignore;
-+ while (n-- > 0)
-+ au_hin_unignore(ign++);
-+}
++/*
++ * file operations
++ */
+
-+#ifdef CONFIG_AUFS_DEBUG
-+void au_dbg_hin_list(struct vfsub_args *vargs)
-+{
-+ int n;
-+ struct au_hin_ignore *ign;
++#ifndef __AUFS_FILE_H__
++#define __AUFS_FILE_H__
+
-+ n = vargs->nignore;
-+ if (!n)
-+ return;
++#ifdef __KERNEL__
+
-+ ign = vargs->ignore;
-+ while (n-- > 0) {
-+ /* AuDebugOn(!list_empty(&ign++->ign_list)); */
-+ if (list_empty(&ign++->ign_list))
-+ continue;
-+ ign--;
-+ AuDbg("%d: pid %d, 0x%x\n",
-+ n + 1, ign->ign_pid, ign->ign_events);
-+ au_hin_unignore(ign);
-+ ign++;
-+ au_dbg_blocked();
-+ }
-+}
-+#endif
++#include <linux/file.h>
++#include <linux/fs.h>
++#include <linux/mm_types.h>
++#include <linux/poll.h>
++#include "rwsem.h"
+
-+/* ---------------------------------------------------------------------- */
++struct au_branch;
++struct au_hfile {
++ struct file *hf_file;
++ struct au_branch *hf_br;
++};
++
++struct au_vdir;
++struct au_fidir {
++ aufs_bindex_t fd_bbot;
++ aufs_bindex_t fd_nent;
++ struct au_vdir *fd_vdir_cache;
++ struct au_hfile fd_hfile[];
++};
+
-+static char *in_name(u32 mask)
++static inline int au_fidir_sz(int nent)
+{
-+#ifdef CONFIG_AUFS_DEBUG
-+#define test_ret(flag) if (mask & flag) \
-+ return #flag;
-+ test_ret(IN_ACCESS);
-+ test_ret(IN_MODIFY);
-+ test_ret(IN_ATTRIB);
-+ test_ret(IN_CLOSE_WRITE);
-+ test_ret(IN_CLOSE_NOWRITE);
-+ test_ret(IN_OPEN);
-+ test_ret(IN_MOVED_FROM);
-+ test_ret(IN_MOVED_TO);
-+ test_ret(IN_CREATE);
-+ test_ret(IN_DELETE);
-+ test_ret(IN_DELETE_SELF);
-+ test_ret(IN_MOVE_SELF);
-+ test_ret(IN_UNMOUNT);
-+ test_ret(IN_Q_OVERFLOW);
-+ test_ret(IN_IGNORED);
-+ return "";
-+#undef test_ret
-+#else
-+ return "??";
-+#endif
++ AuDebugOn(nent < 0);
++ return sizeof(struct au_fidir) + sizeof(struct au_hfile) * nent;
+}
+
++struct au_finfo {
++ atomic_t fi_generation;
++
++ struct au_rwsem fi_rwsem;
++ aufs_bindex_t fi_btop;
++
++ /* do not union them */
++ struct { /* for non-dir */
++ struct au_hfile fi_htop;
++ atomic_t fi_mmapped;
++ };
++ struct au_fidir *fi_hdir; /* for dir only */
++
++ struct hlist_bl_node fi_hlist;
++ struct file *fi_file; /* very ugly */
++ struct rcu_head rcu;
++} ____cacheline_aligned_in_smp;
++
+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen,
-+ struct inode *dir)
-+{
-+ struct dentry *dentry, *d, *parent;
-+ struct qstr *dname;
++/* file.c */
++extern const struct address_space_operations aufs_aop;
++unsigned int au_file_roflags(unsigned int flags);
++struct file *au_h_open(struct dentry *dentry, aufs_bindex_t bindex, int flags,
++ struct file *file, int force_wr);
++struct au_do_open_args {
++ int aopen;
++ int (*open)(struct file *file, int flags,
++ struct file *h_file);
++ struct au_fidir *fidir;
++ struct file *h_file;
++};
++int au_do_open(struct file *file, struct au_do_open_args *args);
++int au_reopen_nondir(struct file *file);
++struct au_pin;
++int au_ready_to_write(struct file *file, loff_t len, struct au_pin *pin);
++int au_reval_and_lock_fdi(struct file *file, int (*reopen)(struct file *file),
++ int wlock, unsigned int fi_lsc);
++int au_do_flush(struct file *file, fl_owner_t id,
++ int (*flush)(struct file *file, fl_owner_t id));
+
-+ LKTRTrace("%.*s, dir%lu\n", nlen, name, dir->i_ino);
++/* poll.c */
++#ifdef CONFIG_AUFS_POLL
++__poll_t aufs_poll(struct file *file, struct poll_table_struct *pt);
++#endif
+
-+ parent = d_find_alias(dir);
-+ if (!parent)
-+ return NULL;
++#ifdef CONFIG_AUFS_BR_HFSPLUS
++/* hfsplus.c */
++struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
++ int force_wr);
++void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file);
++#else
++AuStub(struct file *, au_h_open_pre, return NULL, struct dentry *dentry,
++ aufs_bindex_t bindex, int force_wr)
++AuStubVoid(au_h_open_post, struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file);
++#endif
+
-+ dentry = NULL;
-+ spin_lock(&dcache_lock);
-+ list_for_each_entry(d, &parent->d_subdirs, d_u.d_child) {
-+ LKTRTrace("%.*s\n", AuDLNPair(d));
-+ dname = &d->d_name;
-+ if (dname->len != nlen || memcmp(dname->name, name, nlen))
-+ continue;
-+ if (!atomic_read(&d->d_count) || !d->d_fsdata) {
-+ spin_lock(&d->d_lock);
-+ __d_drop(d);
-+ spin_unlock(&d->d_lock);
-+ continue;
-+ }
++/* f_op.c */
++extern const struct file_operations aufs_file_fop;
++int au_do_open_nondir(struct file *file, int flags, struct file *h_file);
++int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file);
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc);
+
-+ dentry = dget(d);
-+ break;
-+ }
-+ spin_unlock(&dcache_lock);
-+ dput(parent);
++/* finfo.c */
++void au_hfput(struct au_hfile *hf, int execed);
++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex,
++ struct file *h_file);
+
-+ if (dentry) {
-+#if 0
-+ lktr_set_pid(current->pid, LktrArrayPid);
-+ AuDbgDentry(dentry);
-+ lktr_clear_pid(current->pid, LktrArrayPid);
++void au_update_figen(struct file *file);
++struct au_fidir *au_fidir_alloc(struct super_block *sb);
++int au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink);
++
++void au_fi_init_once(void *_fi);
++void au_finfo_fin(struct file *file);
++int au_finfo_init(struct file *file, struct au_fidir *fidir);
++
++/* ioctl.c */
++long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg);
++#ifdef CONFIG_COMPAT
++long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
++ unsigned long arg);
++long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
++ unsigned long arg);
+#endif
-+ di_write_lock_child(dentry);
-+ }
-+ return dentry;
-+}
+
-+static struct inode *lookup_wlock_by_ino(struct super_block *sb,
-+ aufs_bindex_t bindex, ino_t h_ino)
++/* ---------------------------------------------------------------------- */
++
++static inline struct au_finfo *au_fi(struct file *file)
+{
-+ struct inode *inode;
-+ struct au_xino_entry xinoe;
-+ int err;
++ return file->private_data;
++}
+
-+ LKTRTrace("b%d, hi%lu\n", bindex, (unsigned long)h_ino);
-+ AuDebugOn(!au_opt_test_xino(au_mntflags(sb)));
++/* ---------------------------------------------------------------------- */
+
-+ inode = NULL;
-+ err = au_xino_read(sb, bindex, h_ino, &xinoe);
-+ if (!err && xinoe.ino)
-+ inode = ilookup(sb, xinoe.ino);
-+ if (!inode)
-+ goto out;
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ AuWarn("wrong root branch\n");
-+ iput(inode);
-+ inode = NULL;
-+ goto out;
-+ }
++#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)
++*/
+
-+ ii_write_lock_child(inode);
++#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)
+
-+ out:
-+ return inode;
++/* 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 int hin_xino(struct inode *inode, struct inode *h_inode)
++static inline void fi_write_lock_nested(struct file *f, unsigned int lsc)
+{
-+ int err;
-+ aufs_bindex_t bindex, bend, bfound, bstart;
-+ struct inode *h_i;
++ au_rw_write_lock_nested(&au_fi(f)->fi_rwsem, lsc);
++}
+
-+ LKTRTrace("i%lu, hi%lu\n", inode->i_ino, h_inode->i_ino);
++/*
++ * fi_read_lock_1, fi_write_lock_1,
++ * fi_read_lock_2, fi_write_lock_2
++ */
++#define AuReadLockFunc(name) \
++static inline void fi_read_lock_##name(struct file *f) \
++{ fi_read_lock_nested(f, AuLsc_FI_##name); }
+
-+ err = 0;
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ AuWarn("branch root dir was changed\n");
-+ goto out;
-+ }
++#define AuWriteLockFunc(name) \
++static inline void fi_write_lock_##name(struct file *f) \
++{ fi_write_lock_nested(f, AuLsc_FI_##name); }
+
-+ bfound = -1;
-+ bend = au_ibend(inode);
-+ bstart = au_ibstart(inode);
-+#if 0 /* reserved for future use */
-+ if (bindex == bend) {
-+ /* keep this ino in rename case */
-+ goto out;
-+ }
-+#endif
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ if (au_h_iptr(inode, bindex) == h_inode) {
-+ bfound = bindex;
-+ break;
-+ }
-+ }
-+ if (bfound < 0)
-+ goto out;
++#define AuRWLockFuncs(name) \
++ AuReadLockFunc(name) \
++ AuWriteLockFunc(name)
+
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ h_i = au_h_iptr(inode, bindex);
-+ if (h_i)
-+ err = au_xino_write0(inode->i_sb, bindex, h_i->i_ino,
-+ 0);
-+ /* ignore this error */
-+ /* bad action? */
-+ }
++AuRWLockFuncs(1);
++AuRWLockFuncs(2);
+
-+ /* children inode number will be broken */
++#undef AuReadLockFunc
++#undef AuWriteLockFunc
++#undef AuRWLockFuncs
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++#define FiMustNoWaiters(f) AuRwMustNoWaiters(&au_fi(f)->fi_rwsem)
++#define FiMustAnyLock(f) AuRwMustAnyLock(&au_fi(f)->fi_rwsem)
++#define FiMustWriteLock(f) AuRwMustWriteLock(&au_fi(f)->fi_rwsem)
+
-+static int hin_gen_tree(struct dentry *dentry)
-+{
-+ int err, i, j, ndentry;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
-+ struct dentry **dentries;
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++/* todo: hard/soft set? */
++static inline aufs_bindex_t au_fbtop(struct file *file)
++{
++ FiMustAnyLock(file);
++ return au_fi(file)->fi_btop;
++}
+
-+ err = au_dpages_init(&dpages, GFP_NOFS);
-+ if (unlikely(err))
-+ goto out;
-+ err = au_dcsub_pages(&dpages, dentry, NULL, NULL);
-+ if (unlikely(err))
-+ goto out_dpages;
++static inline aufs_bindex_t au_fbbot_dir(struct file *file)
++{
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_bbot;
++}
+
-+ for (i = 0; i < dpages.ndpage; i++) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ ndentry = dpage->ndentry;
-+ for (j = 0; j < ndentry; j++) {
-+ struct dentry *d;
-+ d = dentries[j];
-+ LKTRTrace("%.*s\n", AuDLNPair(d));
-+ if (IS_ROOT(d))
-+ continue;
++static inline struct au_vdir *au_fvdir_cache(struct file *file)
++{
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_vdir_cache;
++}
+
-+ d_drop(d);
-+ au_digen_dec(d);
-+ if (d->d_inode)
-+ /* todo: reset children xino?
-+ cached children only? */
-+ au_iigen_dec(d->d_inode);
-+ }
-+ }
++static inline void au_set_fbtop(struct file *file, aufs_bindex_t bindex)
++{
++ FiMustWriteLock(file);
++ au_fi(file)->fi_btop = bindex;
++}
+
-+ out_dpages:
-+ au_dpages_free(&dpages);
++static inline void au_set_fbbot_dir(struct file *file, aufs_bindex_t bindex)
++{
++ FiMustWriteLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ au_fi(file)->fi_hdir->fd_bbot = bindex;
++}
+
-+ /* discard children */
-+ dentry_unhash(dentry);
-+ dput(dentry);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++static inline void au_set_fvdir_cache(struct file *file,
++ struct au_vdir *vdir_cache)
++{
++ FiMustWriteLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ au_fi(file)->fi_hdir->fd_vdir_cache = vdir_cache;
+}
+
-+/*
-+ * return 0 if processed.
-+ */
-+static int hin_gen_by_inode(char *name, unsigned int nlen, struct inode *inode,
-+ const unsigned int isdir)
++static inline struct file *au_hf_top(struct file *file)
+{
-+ int err;
-+ struct dentry *d;
-+ struct qstr *dname;
++ FiMustAnyLock(file);
++ AuDebugOn(au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_htop.hf_file;
++}
+
-+ LKTRTrace("%.*s, i%lu\n", nlen, name, inode->i_ino);
++static inline struct file *au_hf_dir(struct file *file, aufs_bindex_t bindex)
++{
++ FiMustAnyLock(file);
++ AuDebugOn(!au_fi(file)->fi_hdir);
++ return au_fi(file)->fi_hdir->fd_hfile[0 + bindex].hf_file;
++}
+
-+ err = 1;
-+ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
-+ AuWarn("branch root dir was changed\n");
-+ err = 0;
-+ goto out;
-+ }
-+
-+ if (!isdir) {
-+ AuDebugOn(!name);
-+ au_iigen_dec(inode);
-+ spin_lock(&dcache_lock);
-+ list_for_each_entry(d, &inode->i_dentry, d_alias) {
-+ dname = &d->d_name;
-+ if (dname->len != nlen
-+ && memcmp(dname->name, name, nlen))
-+ continue;
-+ err = 0;
-+ spin_lock(&d->d_lock);
-+ __d_drop(d);
-+ au_digen_dec(d);
-+ spin_unlock(&d->d_lock);
-+ break;
-+ }
-+ spin_unlock(&dcache_lock);
-+ } else {
-+ au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIRS);
-+ d = d_find_alias(inode);
-+ if (!d) {
-+ au_iigen_dec(inode);
-+ goto out;
-+ }
-+
-+ dname = &d->d_name;
-+ if (dname->len == nlen && !memcmp(dname->name, name, nlen))
-+ err = hin_gen_tree(d);
-+ dput(d);
-+ }
++/* todo: memory barrier? */
++static inline unsigned int au_figen(struct file *f)
++{
++ return atomic_read(&au_fi(f)->fi_generation);
++}
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++static inline void au_set_mmapped(struct file *f)
++{
++ if (atomic_inc_return(&au_fi(f)->fi_mmapped))
++ return;
++ pr_warn("fi_mmapped wrapped around\n");
++ while (!atomic_inc_return(&au_fi(f)->fi_mmapped))
++ ;
+}
+
-+static int hin_gen_by_name(struct dentry *dentry, const unsigned int isdir)
++static inline void au_unset_mmapped(struct file *f)
+{
-+ int err;
-+ struct inode *inode;
++ atomic_dec(&au_fi(f)->fi_mmapped);
++}
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++static inline int au_test_mmapped(struct file *f)
++{
++ return atomic_read(&au_fi(f)->fi_mmapped);
++}
+
-+ inode = dentry->d_inode;
-+ if (IS_ROOT(dentry)
-+ /* || (inode && inode->i_ino == AUFS_ROOT_INO) */
-+ ) {
-+ AuWarn("branch root dir was changed\n");
-+ return 0;
-+ }
++/* customize vma->vm_file */
+
-+ err = 0;
-+ if (!isdir) {
-+ d_drop(dentry);
-+ au_digen_dec(dentry);
-+ if (inode)
-+ au_iigen_dec(inode);
-+ } else {
-+ au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIRS);
-+ if (inode)
-+ err = hin_gen_tree(dentry);
-+ }
++static inline void au_do_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
++{
++ struct file *f;
+
-+ AuTraceErr(err);
-+ return err;
++ f = vma->vm_file;
++ get_file(file);
++ vma->vm_file = file;
++ fput(f);
+}
+
-+static void hin_attr(struct inode *inode, struct inode *h_inode)
-+{
-+ struct dentry *h_dentry;
++#ifdef CONFIG_MMU
++#define AuDbgVmRegion(file, vma) do {} while (0)
+
-+ LKTRTrace("i%lu, hi%lu\n", inode->i_ino, h_inode->i_ino);
++static inline void au_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
++{
++ au_do_vm_file_reset(vma, file);
++}
++#else
++#define AuDbgVmRegion(file, vma) \
++ AuDebugOn((vma)->vm_region && (vma)->vm_region->vm_file != (file))
+
-+ if (au_h_iptr(inode, au_ibstart(inode)) != h_inode)
-+ return;
++static inline void au_vm_file_reset(struct vm_area_struct *vma,
++ struct file *file)
++{
++ struct file *f;
+
-+ h_dentry = d_find_alias(h_inode);
-+ if (h_dentry) {
-+ au_update_fuse_h_inode(NULL, h_dentry);
-+ /* ignore an error*/
-+ dput(h_dentry);
-+ }
++ au_do_vm_file_reset(vma, file);
++ f = vma->vm_region->vm_file;
++ get_file(file);
++ vma->vm_region->vm_file = file;
++ fput(f);
++}
++#endif /* CONFIG_MMU */
+
-+ au_cpup_attr_all(inode, /*force*/1);
++/* handle vma->vm_prfile */
++static inline void au_vm_prfile_set(struct vm_area_struct *vma,
++ struct file *file)
++{
++ get_file(file);
++ vma->vm_prfile = file;
++#ifndef CONFIG_MMU
++ get_file(file);
++ vma->vm_region->vm_prfile = file;
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
++#endif /* __KERNEL__ */
++#endif /* __AUFS_FILE_H__ */
+diff -urN /usr/share/empty/fs/aufs/finfo.c linux/fs/aufs/finfo.c
+--- /usr/share/empty/fs/aufs/finfo.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/finfo.c 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,149 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+/* hinotify job flags */
-+#define AuHinJob_XINO0 1
-+#define AuHinJob_GEN (1 << 1)
-+#define AuHinJob_DIRENT (1 << 2)
-+#define AuHinJob_ATTR (1 << 3)
-+#define AuHinJob_ISDIR (1 << 4)
-+#define AuHinJob_TRYXINO0 (1 << 5)
-+#define AuHinJob_MNTPNT (1 << 6)
-+#define au_ftest_hinjob(flags, name) ((flags) & AuHinJob_##name)
-+#define au_fset_hinjob(flags, name) { (flags) |= AuHinJob_##name; }
-+#define au_fclr_hinjob(flags, name) { (flags) &= ~AuHinJob_##name; }
++/*
++ * file private data
++ */
+
-+struct hin_job_args {
-+ unsigned int flags;
-+ struct inode *inode, *h_inode, *dir, *h_dir;
-+ struct dentry *dentry;
-+ char *h_name;
-+ int h_nlen;
-+};
++#include "aufs.h"
+
-+static int hin_job(struct hin_job_args *a)
++void au_hfput(struct au_hfile *hf, int execed)
+{
-+ const unsigned int isdir = au_ftest_hinjob(a->flags, ISDIR);
-+
-+ /* reset xino */
-+ if (au_ftest_hinjob(a->flags, XINO0) && a->inode)
-+ hin_xino(a->inode, a->h_inode);
-+ /* ignore this error */
-+
-+ if (au_ftest_hinjob(a->flags, TRYXINO0)
-+ && a->inode
-+ && a->h_inode) {
-+ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD);
-+ if (!a->h_inode->i_nlink)
-+ hin_xino(a->inode, a->h_inode);
-+ /* ignore this error */
-+ mutex_unlock(&a->h_inode->i_mutex);
-+ }
++ if (execed)
++ allow_write_access(hf->hf_file);
++ fput(hf->hf_file);
++ hf->hf_file = NULL;
++ au_lcnt_dec(&hf->hf_br->br_nfiles);
++ hf->hf_br = NULL;
++}
+
-+ /* make the generation obsolete */
-+ if (au_ftest_hinjob(a->flags, GEN)) {
-+ int err = -1;
-+ if (a->inode)
-+ err = hin_gen_by_inode(a->h_name, a->h_nlen, a->inode,
-+ isdir);
-+ if (err && a->dentry)
-+ hin_gen_by_name(a->dentry, isdir);
-+ /* ignore this error */
-+ }
++void au_set_h_fptr(struct file *file, aufs_bindex_t bindex, struct file *val)
++{
++ struct au_finfo *finfo = au_fi(file);
++ struct au_hfile *hf;
++ struct au_fidir *fidir;
+
-+ /* make dir entries obsolete */
-+ if (au_ftest_hinjob(a->flags, DIRENT) && a->inode) {
-+ struct au_vdir *vdir;
-+ IiMustWriteLock(a->inode);
-+ vdir = au_ivdir(a->inode);
-+ if (vdir)
-+ vdir->vd_jiffy = 0;
-+ /* IMustLock(a->inode); */
-+ /* a->inode->i_version++; */
-+ }
++ fidir = finfo->fi_hdir;
++ if (!fidir) {
++ AuDebugOn(finfo->fi_btop != bindex);
++ hf = &finfo->fi_htop;
++ } else
++ hf = fidir->fd_hfile + bindex;
+
-+ /* update the attr */
-+ if (au_ftest_hinjob(a->flags, ATTR) && a->inode && a->h_inode) {
-+ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD);
-+ hin_attr(a->inode, a->h_inode);
-+ mutex_unlock(&a->h_inode->i_mutex);
++ if (hf && hf->hf_file)
++ au_hfput(hf, vfsub_file_execed(file));
++ if (val) {
++ FiMustWriteLock(file);
++ AuDebugOn(IS_ERR_OR_NULL(file->f_path.dentry));
++ hf->hf_file = val;
++ hf->hf_br = au_sbr(file->f_path.dentry->d_sb, bindex);
+ }
++}
+
-+ /* can do nothing but warn */
-+ if (au_ftest_hinjob(a->flags, MNTPNT)
-+ && a->dentry
-+ && d_mountpoint(a->dentry))
-+ AuWarn("mount-point %.*s is removed or renamed\n",
-+ AuDLNPair(a->dentry));
-+
-+ return 0;
++void au_update_figen(struct file *file)
++{
++ atomic_set(&au_fi(file)->fi_generation, au_digen(file->f_path.dentry));
++ /* smp_mb(); */ /* atomic_set */
+}
+
+/* ---------------------------------------------------------------------- */
+
-+enum { CHILD, PARENT };
-+struct postproc_args {
-+ struct inode *h_dir, *dir, *h_child_inode;
-+ u32 mask;
-+ unsigned int flags[2];
-+ unsigned int h_child_nlen;
-+ char h_child_name[];
-+};
-+
-+static void postproc(void *_args)
++struct au_fidir *au_fidir_alloc(struct super_block *sb)
+{
-+ struct postproc_args *a = _args;
-+ struct super_block *sb;
-+ aufs_bindex_t bindex, bend, bfound;
-+ unsigned char xino, try_iput;
-+ int err;
-+ struct inode *inode;
-+ ino_t h_ino;
-+ struct hin_job_args args;
-+ struct dentry *dentry;
-+ struct au_sbinfo *sbinfo;
++ struct au_fidir *fidir;
++ int nbr;
+
-+ AuDebugOn(!_args);
-+ AuDebugOn(!a->h_dir);
-+ AuDebugOn(!a->dir);
-+ AuDebugOn(!a->mask);
-+ LKTRTrace("mask 0x%x %s, i%lu, hi%lu, hci%lu\n",
-+ a->mask, in_name(a->mask), a->dir->i_ino, a->h_dir->i_ino,
-+ a->h_child_inode ? a->h_child_inode->i_ino : 0);
++ nbr = au_sbbot(sb) + 1;
++ if (nbr < 2)
++ nbr = 2; /* initial allocate for 2 branches */
++ fidir = kzalloc(au_fidir_sz(nbr), GFP_NOFS);
++ if (fidir) {
++ fidir->fd_bbot = -1;
++ fidir->fd_nent = nbr;
++ }
+
-+ inode = NULL;
-+ dentry = NULL;
-+ /*
-+ * do not lock a->dir->i_mutex here
-+ * because of d_revalidate() may cause a deadlock.
-+ */
-+#if 0
-+ /*
-+ * just wait for the dir becoming non-busy.
-+ * for instance, prevent NFSD lookup from "nlink == 0" message.
-+ * but it is not guranteed.
-+ */
-+ mutex_lock(&a->dir->i_mutex);
-+ mutex_unlock(&a->dir->i_mutex);
-+#endif
++ return fidir;
++}
+
-+ sb = a->dir->i_sb;
-+ AuDebugOn(!sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!sbinfo);
-+ /* big aufs lock */
-+ si_noflush_write_lock(sb);
++int au_fidir_realloc(struct au_finfo *finfo, int nbr, int may_shrink)
++{
++ int err;
++ struct au_fidir *fidir, *p;
+
-+ ii_read_lock_parent(a->dir);
-+ bfound = -1;
-+ bend = au_ibend(a->dir);
-+ for (bindex = au_ibstart(a->dir); bindex <= bend; bindex++)
-+ if (au_h_iptr(a->dir, bindex) == a->h_dir) {
-+ bfound = bindex;
-+ break;
-+ }
-+ ii_read_unlock(a->dir);
-+ if (unlikely(bfound < 0))
-+ goto out;
++ AuRwMustWriteLock(&finfo->fi_rwsem);
++ fidir = finfo->fi_hdir;
++ AuDebugOn(!fidir);
+
-+ xino = !!au_opt_test_xino(au_mntflags(sb));
-+ h_ino = 0;
-+ if (a->h_child_inode)
-+ h_ino = a->h_child_inode->i_ino;
++ err = -ENOMEM;
++ p = au_kzrealloc(fidir, au_fidir_sz(fidir->fd_nent), au_fidir_sz(nbr),
++ GFP_NOFS, may_shrink);
++ if (p) {
++ p->fd_nent = nbr;
++ finfo->fi_hdir = p;
++ err = 0;
++ }
+
-+ if (a->h_child_nlen
-+ && (au_ftest_hinjob(a->flags[CHILD], GEN)
-+ || au_ftest_hinjob(a->flags[CHILD], MNTPNT)))
-+ dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen,
-+ a->dir);
-+ try_iput = 0;
-+ if (dentry)
-+ inode = dentry->d_inode;
-+ if (xino && !inode && h_ino
-+ && (au_ftest_hinjob(a->flags[CHILD], XINO0)
-+ || au_ftest_hinjob(a->flags[CHILD], TRYXINO0)
-+ || au_ftest_hinjob(a->flags[CHILD], GEN)
-+ || au_ftest_hinjob(a->flags[CHILD], ATTR))) {
-+ inode = lookup_wlock_by_ino(sb, bfound, h_ino);
-+ try_iput = 1;
-+ }
++ return err;
++}
+
-+ args.flags = a->flags[CHILD];
-+ args.dentry = dentry;
-+ args.inode = inode;
-+ args.h_inode = a->h_child_inode;
-+ args.dir = a->dir;
-+ args.h_dir = a->h_dir;
-+ args.h_name = a->h_child_name;
-+ args.h_nlen = a->h_child_nlen;
-+ err = hin_job(&args);
-+ if (dentry) {
-+ if (dentry->d_fsdata)
-+ di_write_unlock(dentry);
-+ dput(dentry);
-+ }
-+ if (inode && try_iput) {
-+ ii_write_unlock(inode);
-+ iput(inode);
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ ii_write_lock_parent(a->dir);
-+ args.flags = a->flags[PARENT];
-+ args.dentry = NULL;
-+ args.inode = a->dir;
-+ args.h_inode = a->h_dir;
-+ args.dir = NULL;
-+ args.h_dir = NULL;
-+ args.h_name = NULL;
-+ args.h_nlen = 0;
-+ err = hin_job(&args);
-+ ii_write_unlock(a->dir);
++void au_finfo_fin(struct file *file)
++{
++ struct au_finfo *finfo;
+
-+ out:
-+ au_nwt_done(&sbinfo->si_nowait);
-+ si_write_unlock(sb);
++ au_lcnt_dec(&au_sbi(file->f_path.dentry->d_sb)->si_nfiles);
+
-+ iput(a->h_child_inode);
-+ iput(a->h_dir);
-+ iput(a->dir);
-+ kfree(a);
++ finfo = au_fi(file);
++ AuDebugOn(finfo->fi_hdir);
++ AuRwDestroy(&finfo->fi_rwsem);
++ au_cache_free_finfo(finfo);
+}
+
-+static void aufs_inotify(struct inotify_watch *watch, u32 wd, u32 mask,
-+ u32 cookie, const char *h_child_name,
-+ struct inode *h_child_inode)
++void au_fi_init_once(void *_finfo)
+{
-+ struct au_hinotify *hinotify;
-+ struct postproc_args *args;
-+ int len, wkq_err;
-+ unsigned char isdir, isroot, wh;
-+ char *p;
-+ struct inode *dir;
-+ unsigned int flags[2];
++ struct au_finfo *finfo = _finfo;
+
-+ LKTRTrace("i%lu, wd %d, mask 0x%x %s, cookie 0x%x, hcname %s, hi%lu\n",
-+ watch->inode->i_ino, wd, mask, in_name(mask), cookie,
-+ h_child_name ? h_child_name : "",
-+ h_child_inode ? h_child_inode->i_ino : 0);
++ au_rw_init(&finfo->fi_rwsem);
++}
+
-+ /* if IN_UNMOUNT happens, there must be another bug */
-+ if (mask & (IN_IGNORED | IN_UNMOUNT)) {
-+ put_inotify_watch(watch);
-+ return;
-+ }
++int au_finfo_init(struct file *file, struct au_fidir *fidir)
++{
++ int err;
++ struct au_finfo *finfo;
++ struct dentry *dentry;
+
-+#ifdef DbgInotify
-+ if (!h_child_name || strcmp(h_child_name, AUFS_XINO_FNAME))
-+ AuDbg("i%lu, wd %d, mask 0x%x %s, cookie 0x%x, hcname %s,"
-+ " hi%lu\n",
-+ watch->inode->i_ino, wd, mask, in_name(mask), cookie,
-+ h_child_name ? h_child_name : "",
-+ h_child_inode ? h_child_inode->i_ino : 0);
-+#endif
++ err = -ENOMEM;
++ dentry = file->f_path.dentry;
++ finfo = au_cache_alloc_finfo();
++ if (unlikely(!finfo))
++ goto out;
+
-+ hinotify = container_of(watch, struct au_hinotify, hin_watch);
-+ AuDebugOn(!hinotify || !hinotify->hin_aufs_inode);
-+ if (au_hin_test_ignore(mask, hinotify)) {
-+#ifdef DbgInotify
-+ AuDbg("ignored\n");
-+#endif
-+ return;
-+ }
-+#if 0 /* tmp debug */
-+ if (h_child_name && !strcmp(h_child_name, AUFS_XINO_FNAME)) {
-+ AuDbg("i%lu, wd %d, mask 0x%x %s, cookie 0x%x, hcname %s, hi%lu\n",
-+ watch->inode->i_ino, wd, mask, in_name(mask), cookie,
-+ h_child_name ? h_child_name : "",
-+ h_child_inode ? h_child_inode->i_ino : 0);
-+ //au_dbg_blocked();
-+ }
-+#endif
++ err = 0;
++ au_lcnt_inc(&au_sbi(dentry->d_sb)->si_nfiles);
++ au_rw_write_lock(&finfo->fi_rwsem);
++ finfo->fi_btop = -1;
++ finfo->fi_hdir = fidir;
++ atomic_set(&finfo->fi_generation, au_digen(dentry));
++ /* smp_mb(); */ /* atomic_set */
+
-+ dir = igrab(hinotify->hin_aufs_inode);
-+ if (!dir)
-+ return;
-+ isroot = (dir->i_ino == AUFS_ROOT_INO);
-+ len = 0;
-+ wh = 0;
-+ if (h_child_name) {
-+ len = strlen(h_child_name);
-+ if (!memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
-+ h_child_name += AUFS_WH_PFX_LEN;
-+ len -= AUFS_WH_PFX_LEN;
-+ wh = 1;
-+ }
-+ }
-+
-+ isdir = 0;
-+ if (h_child_inode)
-+ isdir = !!S_ISDIR(h_child_inode->i_mode);
-+ flags[PARENT] = AuHinJob_ISDIR;
-+ flags[CHILD] = 0;
-+ if (isdir)
-+ flags[CHILD] = AuHinJob_ISDIR;
-+ switch (mask & IN_ALL_EVENTS) {
-+ case IN_MODIFY:
-+ /*FALLTHROUGH*/
-+ case IN_ATTRIB:
-+ if (h_child_inode) {
-+ if (!wh)
-+ au_fset_hinjob(flags[CHILD], ATTR);
-+ } else
-+ au_fset_hinjob(flags[PARENT], ATTR);
-+ break;
-+
-+ /* IN_MOVED_FROM is the first event in rename(2) */
-+ case IN_MOVED_FROM:
-+ case IN_MOVED_TO:
-+ AuDebugOn(!h_child_name || !h_child_inode);
-+ au_fset_hinjob(flags[CHILD], GEN);
-+ au_fset_hinjob(flags[CHILD], ATTR);
-+ if (1 || isdir)
-+ au_fset_hinjob(flags[CHILD], XINO0);
-+ au_fset_hinjob(flags[CHILD], MNTPNT);
-+
-+ au_fset_hinjob(flags[PARENT], ATTR);
-+ au_fset_hinjob(flags[PARENT], DIRENT);
-+ break;
-+
-+ case IN_CREATE:
-+ AuDebugOn(!h_child_name || !h_child_inode);
-+ au_fset_hinjob(flags[PARENT], ATTR);
-+ au_fset_hinjob(flags[PARENT], DIRENT);
-+ au_fset_hinjob(flags[CHILD], GEN);
-+ /* hard link */
-+ if (!isdir && h_child_inode->i_nlink > 1)
-+ au_fset_hinjob(flags[CHILD], ATTR);
-+ break;
-+
-+ case IN_DELETE:
-+ /*
-+ * aufs never be able to get this child inode.
-+ * revalidation should be in d_revalidate()
-+ * by checking i_nlink, i_generation or d_unhashed().
-+ */
-+ AuDebugOn(!h_child_name);
-+ au_fset_hinjob(flags[PARENT], ATTR);
-+ au_fset_hinjob(flags[PARENT], DIRENT);
-+ au_fset_hinjob(flags[CHILD], GEN);
-+ au_fset_hinjob(flags[CHILD], TRYXINO0);
-+ au_fset_hinjob(flags[CHILD], MNTPNT);
-+ break;
-+
-+ case IN_DELETE_SELF:
-+#if 0
-+ if (!isroot)
-+ au_fset_hinjob(flags[PARENT], GEN);
-+ /*FALLTHROUGH*/
-+#endif
-+
-+ case IN_MOVE_SELF:
-+#if 0
-+ /*
-+ * when an inotify is set to an aufs inode,
-+ * such inode can be isolated and this event can be fired
-+ * solely.
-+ */
-+ AuDebugOn(h_child_name || h_child_inode);
-+ if (unlikely(isroot)) {
-+ AuWarn("root branch was moved\n");
-+ iput(dir);
-+ return;
-+ }
-+ au_fset_hinjob(flags[PARENT], XINO0);
-+ au_fset_hinjob(flags[PARENT], GEN);
-+ au_fset_hinjob(flags[PARENT], ATTR);
-+ au_fset_hinjob(flags[PARENT], DIRENT);
-+ /* au_fset_hinjob(flags[PARENT], MNTPNT); */
-+ break;
-+#endif
-+
-+ case IN_ACCESS:
-+ default:
-+ AuDebugOn(1);
-+ }
-+
-+ if (wh)
-+ h_child_inode = NULL;
-+
-+ /* iput() and kfree() will be called in postproc() */
-+ /*
-+ * inotify_mutex is already acquired and kmalloc/prune_icache may lock
-+ * iprune_mutex. strange.
-+ */
-+ lockdep_off();
-+ args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS);
-+ lockdep_on();
-+ if (unlikely(!args)) {
-+ AuErr1("no memory\n");
-+ iput(dir);
-+ return;
-+ }
-+ args->flags[PARENT] = flags[PARENT];
-+ args->flags[CHILD] = flags[CHILD];
-+ args->mask = mask;
-+ args->dir = dir;
-+ args->h_dir = igrab(watch->inode);
-+ if (h_child_inode)
-+ h_child_inode = igrab(h_child_inode); /* can be NULL */
-+ args->h_child_inode = h_child_inode;
-+ args->h_child_nlen = len;
-+ if (len) {
-+ p = (void *)args;
-+ p += sizeof(*args);
-+ memcpy(p, h_child_name, len + 1);
-+ }
-+
-+ lockdep_off();
-+ wkq_err = au_wkq_nowait(postproc, args, dir->i_sb, /*dlgt*/0);
-+ lockdep_on();
-+ if (unlikely(wkq_err))
-+ AuErr("wkq %d\n", wkq_err);
-+}
-+
-+static void aufs_inotify_destroy(struct inotify_watch *watch)
-+{
-+ return;
-+}
-+
-+static struct inotify_operations aufs_inotify_ops = {
-+ .handle_event = aufs_inotify,
-+ .destroy_watch = aufs_inotify_destroy
-+};
-+
-+/* ---------------------------------------------------------------------- */
-+
-+static void au_hin_destroy_cache(void)
-+{
-+ kmem_cache_destroy(au_cachep[AuCache_HINOTIFY]);
-+ au_cachep[AuCache_HINOTIFY] = NULL;
-+}
-+
-+int __init au_inotify_init(void)
-+{
-+ int err;
++ file->private_data = finfo;
+
-+ err = -ENOMEM;
-+ au_cachep[AuCache_HINOTIFY] = AuCache(au_hinotify);
-+ if (au_cachep[AuCache_HINOTIFY]) {
-+ err = 0;
-+ in_handle = inotify_init(&aufs_inotify_ops);
-+ if (IS_ERR(in_handle)) {
-+ err = PTR_ERR(in_handle);
-+ au_hin_destroy_cache();
-+ }
-+ }
-+ AuTraceErr(err);
++out:
+ return err;
+}
-+
-+void au_inotify_fin(void)
-+{
-+ inotify_destroy(in_handle);
-+ if (au_cachep[AuCache_HINOTIFY])
-+ au_hin_destroy_cache();
-+}
-diff -urN linux/fs/aufs/iinfo.c linux-aufs/fs/aufs/iinfo.c
---- linux/fs/aufs/iinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/iinfo.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,284 @@
+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 2023-02-20 21:05:51.959693785 +0100
+@@ -0,0 +1,780 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode private data
-+ *
-+ * $Id$
++ * file and vm operations
+ */
+
++#include <linux/aio.h>
++#include <linux/fs_stack.h>
++#include <linux/mman.h>
++#include <linux/security.h>
+#include "aufs.h"
+
-+struct au_iinfo *au_ii(struct inode *inode)
++int au_do_open_nondir(struct file *file, int flags, struct file *h_file)
+{
-+ struct au_iinfo *iinfo;
-+
-+ iinfo = &(container_of(inode, struct aufs_icntnr, vfs_inode)->iinfo);
-+ /* bad_inode case */
-+ if (unlikely(!iinfo->ii_hinode))
-+ return NULL;
-+ AuDebugOn(!iinfo->ii_hinode
-+ /* || au_sbi(inode->i_sb)->si_bend < iinfo->ii_bend */
-+ || iinfo->ii_bend < iinfo->ii_bstart);
-+ return iinfo;
-+}
++ int err;
++ aufs_bindex_t bindex;
++ struct dentry *dentry, *h_dentry;
++ struct au_finfo *finfo;
++ struct inode *h_inode;
+
-+struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ struct inode *hidden_inode;
++ FiMustWriteLock(file);
+
-+ /* lock free root dinfo/inode */
-+ if (inode->i_ino != AUFS_ROOT_INO) {
-+ IiMustAnyLock(inode);
-+ AuDebugOn(bindex < 0 || au_ibend(inode) < bindex);
++ err = 0;
++ dentry = file->f_path.dentry;
++ AuDebugOn(IS_ERR_OR_NULL(dentry));
++ finfo = au_fi(file);
++ memset(&finfo->fi_htop, 0, sizeof(finfo->fi_htop));
++ atomic_set(&finfo->fi_mmapped, 0);
++ bindex = au_dbtop(dentry);
++ if (!h_file) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
++ if (unlikely(err))
++ goto out;
++ h_file = au_h_open(dentry, bindex, flags, file, /*force_wr*/0);
++ if (IS_ERR(h_file)) {
++ err = PTR_ERR(h_file);
++ goto out;
++ }
+ } else {
-+ SiMustAnyLock(inode->i_sb);
-+ AuDebugOn(bindex < 0 || au_sbend(inode->i_sb) < bindex);
++ h_dentry = h_file->f_path.dentry;
++ err = vfsub_test_mntns(file->f_path.mnt, h_dentry->d_sb);
++ if (unlikely(err))
++ goto out;
++ /* br ref is already inc-ed */
++ }
++
++ if (flags & __O_TMPFILE) {
++ AuDebugOn(!h_file);
++ AuDebugOn(h_file != au_di(dentry)->di_htmpfile);
++ au_di(dentry)->di_htmpfile = NULL;
++
++ if (!(flags & O_EXCL)) {
++ h_inode = file_inode(h_file);
++ spin_lock(&h_inode->i_lock);
++ h_inode->i_state |= I_LINKABLE;
++ spin_unlock(&h_inode->i_lock);
++ }
+ }
-+ hidden_inode = au_ii(inode)->ii_hinode[0 + bindex].hi_inode;
-+ AuDebugOn(hidden_inode && atomic_read(&hidden_inode->i_count) <= 0);
-+ return hidden_inode;
++ au_set_fbtop(file, bindex);
++ au_set_h_fptr(file, bindex, h_file);
++ au_update_figen(file);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++
++out:
++ return err;
+}
+
-+aufs_bindex_t au_ii_br_id(struct inode *inode, aufs_bindex_t bindex)
++static int aufs_open_nondir(struct inode *inode __maybe_unused,
++ struct file *file)
+{
-+ IiMustAnyLock(inode);
-+ AuDebugOn(bindex < 0
-+ || au_ibend(inode) < bindex
-+ || !au_ii(inode)->ii_hinode[0 + bindex].hi_inode);
-+ return au_ii(inode)->ii_hinode[0 + bindex].hi_id;
++ int err;
++ struct super_block *sb;
++ struct au_do_open_args args = {
++ .open = au_do_open_nondir
++ };
++
++ AuDbg("%pD, f_flags 0x%x, f_mode 0x%x\n",
++ file, vfsub_file_flags(file), file->f_mode);
++
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ err = au_do_open(file, &args);
++ si_read_unlock(sb);
++ return err;
+}
+
-+/* todo: hard/soft set? */
-+void au_set_ibstart(struct inode *inode, aufs_bindex_t bindex)
++int aufs_release_nondir(struct inode *inode __maybe_unused, struct file *file)
+{
-+ struct au_iinfo *iinfo = au_ii(inode);
-+ struct inode *h_inode;
++ struct au_finfo *finfo;
++ aufs_bindex_t bindex;
+
-+ IiMustWriteLock(inode);
-+ AuDebugOn(au_sbend(inode->i_sb) < bindex);
-+ iinfo->ii_bstart = bindex;
-+ h_inode = iinfo->ii_hinode[bindex + 0].hi_inode;
-+ if (h_inode)
-+ au_cpup_igen(inode, h_inode);
++ finfo = au_fi(file);
++ au_hbl_del(&finfo->fi_hlist,
++ &au_sbi(file->f_path.dentry->d_sb)->si_files);
++ bindex = finfo->fi_btop;
++ if (bindex >= 0)
++ au_set_h_fptr(file, bindex, NULL);
++
++ au_finfo_fin(file);
++ return 0;
+}
+
-+unsigned int au_hi_flags(struct inode *inode, int isdir)
++/* ---------------------------------------------------------------------- */
++
++static int au_do_flush_nondir(struct file *file, fl_owner_t id)
+{
-+ unsigned int flags;
-+ const unsigned int mnt_flags = au_mntflags(inode->i_sb);
++ int err;
++ struct file *h_file;
+
-+ flags = 0;
-+ if (au_opt_test_xino(mnt_flags))
-+ au_fset_hi(flags, XINO);
-+ if (isdir && au_opt_test(mnt_flags, UDBA_INOTIFY))
-+ au_fset_hi(flags, NOTIFY);
-+ return flags;
++ err = 0;
++ h_file = au_hf_top(file);
++ if (h_file)
++ err = vfsub_flush(h_file, id);
++ return err;
+}
+
-+void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode, unsigned int flags)
++static int aufs_flush_nondir(struct file *file, fl_owner_t id)
+{
-+ struct au_hinode *hinode;
-+ struct inode *hi;
-+ struct au_iinfo *iinfo = au_ii(inode);
++ return au_do_flush(file, id, au_do_flush_nondir);
++}
+
-+ LKTRTrace("i%lu, b%d, hi%lu, flags 0x%x\n",
-+ inode->i_ino, bindex, h_inode ? h_inode->i_ino : 0, flags);
-+ IiMustWriteLock(inode);
-+ hinode = iinfo->ii_hinode + bindex;
-+ hi = hinode->hi_inode;
-+ AuDebugOn(bindex < au_ibstart(inode) || au_ibend(inode) < bindex);
-+ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
-+ AuDebugOn(h_inode && hi);
++/* ---------------------------------------------------------------------- */
++/*
++ * read and write functions acquire [fdi]_rwsem once, but release before
++ * mmap_sem. This is because to stop a race condition between mmap(2).
++ * Releasing these aufs-rwsem should be safe, no branch-management (by keeping
++ * si_rwsem), no harmful copy-up should happen. Actually copy-up may happen in
++ * read functions after [fdi]_rwsem are released, but it should be harmless.
++ */
+
-+ if (hi)
-+ au_hiput(hinode);
-+ hinode->hi_inode = h_inode;
-+ if (h_inode) {
-+ int err;
-+ struct super_block *sb = inode->i_sb;
++/* Callers should call au_read_post() or fput() in the end */
++struct file *au_read_pre(struct file *file, int keep_fi, unsigned int lsc)
++{
++ struct file *h_file;
++ int err;
+
-+ if (bindex == iinfo->ii_bstart)
-+ au_cpup_igen(inode, h_inode);
-+ hinode->hi_id = au_sbr_id(sb, bindex);
-+ if (au_ftest_hi(flags, XINO)) {
-+ struct au_xino_entry xinoe = {
-+ .ino = inode->i_ino,
-+ /* .h_gen = h_inode->i_generation */
-+ };
-+ err = au_xino_write(sb, bindex, h_inode->i_ino, &xinoe);
-+ if (unlikely(err))
-+ AuIOErr1("failed au_xino_write() %d\n", err);
-+ }
++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/0, lsc);
++ if (!err) {
++ di_read_unlock(file->f_path.dentry, AuLock_IR);
++ h_file = au_hf_top(file);
++ get_file(h_file);
++ if (!keep_fi)
++ fi_read_unlock(file);
++ } else
++ h_file = ERR_PTR(err);
+
-+ if (au_ftest_hi(flags, NOTIFY)
-+ && au_br_hinotifyable(au_sbr_perm(sb, bindex))) {
-+ err = au_hin_alloc(hinode, inode, h_inode);
-+ if (unlikely(err))
-+ AuIOErr1("au_hin_alloc() %d\n", err);
-+ }
-+ }
++ return h_file;
+}
+
-+void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_wh)
++static void au_read_post(struct inode *inode, struct file *h_file)
+{
-+ struct au_hinode *hinode;
-+
-+ IiMustWriteLock(inode);
-+ hinode = au_ii(inode)->ii_hinode + bindex;
-+ AuDebugOn(hinode->hi_whdentry);
-+ hinode->hi_whdentry = h_wh;
++ /* update without lock, I don't think it a problem */
++ fsstack_copy_attr_atime(inode, file_inode(h_file));
++ fput(h_file);
+}
+
-+void au_update_iigen(struct inode *inode)
-+{
-+ AuDebugOn(!inode->i_sb);
-+ atomic_set(&au_ii(inode)->ii_generation, au_sigen(inode->i_sb));
-+ /* smp_mb(); */ /* atomic_set */
-+}
++struct au_write_pre {
++ /* input */
++ unsigned int lsc;
+
-+/* it may be called at remount time, too */
-+void au_update_brange(struct inode *inode, int do_put_zero)
-+{
-+ struct au_iinfo *iinfo;
++ /* output */
++ blkcnt_t blks;
++ aufs_bindex_t btop;
++};
+
-+ LKTRTrace("i%lu, %d\n", inode->i_ino, do_put_zero);
-+ IiMustWriteLock(inode);
++/*
++ * return with iinfo is write-locked
++ * callers should call au_write_post() or iinfo_write_unlock() + fput() in the
++ * end
++ */
++static struct file *au_write_pre(struct file *file, int do_ready,
++ struct au_write_pre *wpre)
++{
++ struct file *h_file;
++ struct dentry *dentry;
++ int err;
++ unsigned int lsc;
++ struct au_pin pin;
+
-+ iinfo = au_ii(inode);
-+ if (!iinfo || iinfo->ii_bstart < 0)
-+ return;
++ 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;
+
-+ if (do_put_zero) {
-+ aufs_bindex_t bindex;
-+ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend;
-+ bindex++) {
-+ struct inode *h_i;
-+ h_i = iinfo->ii_hinode[0 + bindex].hi_inode;
-+ if (h_i && !h_i->i_nlink)
-+ au_set_h_iptr(inode, bindex, NULL, 0);
++ dentry = file->f_path.dentry;
++ if (do_ready) {
++ err = au_ready_to_write(file, -1, &pin);
++ if (unlikely(err)) {
++ h_file = ERR_PTR(err);
++ di_write_unlock(dentry);
++ goto out_fi;
+ }
+ }
+
-+ iinfo->ii_bstart = -1;
-+ while (++iinfo->ii_bstart <= iinfo->ii_bend)
-+ if (iinfo->ii_hinode[0 + iinfo->ii_bstart].hi_inode)
-+ break;
-+ if (iinfo->ii_bstart > iinfo->ii_bend) {
-+ iinfo->ii_bstart = -1;
-+ iinfo->ii_bend = -1;
-+ return;
-+ }
++ di_downgrade_lock(dentry, /*flags*/0);
++ if (wpre)
++ wpre->btop = au_fbtop(file);
++ h_file = au_hf_top(file);
++ get_file(h_file);
++ if (wpre)
++ wpre->blks = file_inode(h_file)->i_blocks;
++ if (do_ready)
++ au_unpin(&pin);
++ di_read_unlock(dentry, /*flags*/0);
+
-+ iinfo->ii_bend++;
-+ while (0 <= --iinfo->ii_bend)
-+ if (iinfo->ii_hinode[0 + iinfo->ii_bend].hi_inode)
-+ break;
-+ AuDebugOn(iinfo->ii_bstart > iinfo->ii_bend || iinfo->ii_bend < 0);
++out_fi:
++ fi_write_unlock(file);
++out:
++ return h_file;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+int au_iinfo_init(struct inode *inode)
++static void au_write_post(struct inode *inode, struct file *h_file,
++ struct au_write_pre *wpre, ssize_t written)
+{
-+ struct au_iinfo *iinfo;
-+ struct super_block *sb;
-+ int nbr, i;
++ struct inode *h_inode;
+
-+ sb = inode->i_sb;
-+ AuDebugOn(!sb);
-+ iinfo = &(container_of(inode, struct aufs_icntnr, vfs_inode)->iinfo);
-+ AuDebugOn(iinfo->ii_hinode);
-+ nbr = au_sbend(sb) + 1;
-+ if (nbr <= 0)
-+ nbr = 1;
-+ iinfo->ii_hinode = kcalloc(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS);
-+ if (iinfo->ii_hinode) {
-+ for (i = 0; i < nbr; i++)
-+ iinfo->ii_hinode[i].hi_id = -1;
-+ atomic_set(&iinfo->ii_generation, au_sigen(sb));
-+ /* smp_mb(); */ /* atomic_set */
-+ au_rw_init_nolock(&iinfo->ii_rwsem);
-+ iinfo->ii_bstart = -1;
-+ iinfo->ii_bend = -1;
-+ iinfo->ii_vdir = NULL;
-+ return 0;
-+ }
-+ return -ENOMEM;
++ au_cpup_attr_timesizes(inode);
++ AuDebugOn(au_ibtop(inode) != wpre->btop);
++ h_inode = file_inode(h_file);
++ inode->i_mode = h_inode->i_mode;
++ ii_write_unlock(inode);
++ /* AuDbg("blks %llu, %llu\n", (u64)blks, (u64)h_inode->i_blocks); */
++ if (written > 0)
++ au_fhsm_wrote(inode->i_sb, wpre->btop,
++ /*force*/h_inode->i_blocks > wpre->blks);
++ fput(h_file);
+}
+
-+static int au_iinfo_write0(struct super_block *sb, struct au_hinode *hinode,
-+ ino_t ino)
++/*
++ * todo: very ugly
++ * it locks both of i_mutex and si_rwsem for read in safe.
++ * if the plink maintenance mode continues forever (that is the problem),
++ * may loop forever.
++ */
++static void au_mtx_and_read_lock(struct inode *inode)
+{
-+ int err, locked;
-+ aufs_bindex_t bindex;
++ int err;
++ struct super_block *sb = inode->i_sb;
+
-+ err = 0;
-+ locked = si_noflush_read_trylock(sb); /* crucio! */
-+ bindex = au_br_index(sb, hinode->hi_id);
-+ if (bindex >= 0)
-+ err = au_xino_write0(sb, bindex, hinode->hi_inode->i_ino, ino);
-+ /* error action? */
-+ if (locked)
++ while (1) {
++ inode_lock(inode);
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (!err)
++ break;
++ inode_unlock(inode);
++ si_read_lock(sb, AuLock_NOPLMW);
+ si_read_unlock(sb);
-+ return err;
++ }
+}
+
-+void au_iinfo_fin(struct inode *inode)
++static ssize_t au_do_iter(struct file *h_file, int rw, struct kiocb *kio,
++ struct iov_iter *iov_iter)
+{
-+ struct au_iinfo *iinfo;
-+ aufs_bindex_t bend;
-+ unsigned char unlinked;
-+ struct au_hinode *hi;
-+ struct super_block *sb;
-+ ino_t ino;
++ ssize_t err;
++ struct file *file;
++ ssize_t (*iter)(struct kiocb *, struct iov_iter *);
+
-+ iinfo = au_ii(inode);
-+ /* bad_inode case */
-+ if (!iinfo)
-+ return;
++ err = security_file_permission(h_file, rw);
++ if (unlikely(err))
++ goto out;
+
-+ if (iinfo->ii_vdir)
-+ au_vdir_free(iinfo->ii_vdir);
++ err = -ENOSYS; /* the branch doesn't have its ->(read|write)_iter() */
++ iter = NULL;
++ if (rw == MAY_READ)
++ iter = h_file->f_op->read_iter;
++ else if (rw == MAY_WRITE)
++ iter = h_file->f_op->write_iter;
+
-+ if (iinfo->ii_bstart >= 0) {
-+ sb = inode->i_sb;
-+ unlinked = !inode->i_nlink;
-+ ino = 0;
-+ if (unlinked)
-+ ino = inode->i_ino;
-+ hi = iinfo->ii_hinode + iinfo->ii_bstart;
-+ bend = iinfo->ii_bend;
-+ while (iinfo->ii_bstart++ <= bend) {
-+ if (hi->hi_inode) {
-+ if (unlinked || !hi->hi_inode->i_nlink) {
-+ au_iinfo_write0(sb, hi, ino);
-+ /* ignore this error */
-+ ino = 0;
-+ }
-+ au_hiput(hi);
-+ }
-+ hi++;
-+ }
-+ }
++ file = kio->ki_filp;
++ kio->ki_filp = h_file;
++ if (iter) {
++ lockdep_off();
++ err = iter(kio, iov_iter);
++ lockdep_on();
++ } else
++ /* currently there is no such fs */
++ WARN_ON_ONCE(1);
++ kio->ki_filp = file;
+
-+ kfree(iinfo->ii_hinode);
-+ au_rwsem_destroy(&iinfo->ii_rwsem);
++out:
++ return err;
+}
-diff -urN linux/fs/aufs/inode.c linux-aufs/fs/aufs/inode.c
---- linux/fs/aufs/inode.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/inode.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,421 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * inode functions
-+ *
-+ * $Id$
-+ */
-+
-+#include "aufs.h"
+
-+int au_refresh_hinode_self(struct inode *inode)
++static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
+{
-+ int err, new_sz, update;
-+ struct inode *first;
-+ struct au_hinode *p, *q, tmp;
++ ssize_t err;
++ struct file *file, *h_file;
++ struct inode *inode;
+ struct super_block *sb;
-+ struct au_iinfo *iinfo;
-+ aufs_bindex_t bindex, bend, new_bindex;
-+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ IiMustWriteLock(inode);
+
-+ err = -ENOMEM;
-+ update = 0;
++ file = kio->ki_filp;
++ inode = file_inode(file);
+ sb = inode->i_sb;
-+ bend = au_sbend(sb);
-+ new_sz = sizeof(*iinfo->ii_hinode) * (bend + 1);
-+ iinfo = au_ii(inode);
-+ p = au_kzrealloc(iinfo->ii_hinode, sizeof(*p) * (iinfo->ii_bend + 1),
-+ new_sz, GFP_NOFS);
-+ if (unlikely(!p))
-+ goto out;
-+
-+ iinfo->ii_hinode = p;
-+ p = iinfo->ii_hinode + iinfo->ii_bstart;
-+ first = p->hi_inode;
-+ err = 0;
-+ for (bindex = iinfo->ii_bstart; bindex <= iinfo->ii_bend;
-+ bindex++, p++) {
-+ if (!p->hi_inode)
-+ continue;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ new_bindex = au_br_index(sb, p->hi_id);
-+ if (new_bindex == bindex)
-+ continue;
-+ if (new_bindex < 0) {
-+ update++;
-+ au_hiput(p);
-+ p->hi_inode = NULL;
-+ continue;
-+ }
++ h_file = au_read_pre(file, /*keep_fi*/1, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
-+ if (new_bindex < iinfo->ii_bstart)
-+ iinfo->ii_bstart = new_bindex;
-+ if (iinfo->ii_bend < new_bindex)
-+ iinfo->ii_bend = new_bindex;
-+ /* swap two hidden inode, and loop again */
-+ q = iinfo->ii_hinode + new_bindex;
-+ tmp = *q;
-+ *q = *p;
-+ *p = tmp;
-+ if (tmp.hi_inode) {
-+ bindex--;
-+ p--;
++ if (au_test_loopback_kthread()) {
++ au_warn_loopback(h_file->f_path.dentry->d_sb);
++ if (file->f_mapping != h_file->f_mapping) {
++ file->f_mapping = h_file->f_mapping;
++ smp_mb(); /* unnecessary? */
+ }
+ }
-+ au_update_brange(inode, /*do_put_zero*/0);
-+
-+ if (unlikely(err))
-+ goto out;
++ fi_read_unlock(file);
+
-+ if (1 || first != au_h_iptr(inode, iinfo->ii_bstart))
-+ au_cpup_attr_all(inode, /*force*/0);
-+ if (update && S_ISDIR(inode->i_mode))
-+ inode->i_version++;
-+ au_update_iigen(inode);
++ err = au_do_iter(h_file, MAY_READ, kio, iov_iter);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++ au_read_post(inode, h_file);
+
-+ out:
-+ AuTraceErr(err);
++out:
++ si_read_unlock(sb);
+ return err;
+}
+
-+int au_refresh_hinode(struct inode *inode, struct dentry *dentry)
++static ssize_t aufs_write_iter(struct kiocb *kio, struct iov_iter *iov_iter)
+{
-+ int err, update;
-+ struct inode *first;
-+ struct au_hinode *p;
-+ struct super_block *sb;
-+ struct au_iinfo *iinfo;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char isdir;
-+ unsigned int flags;
++ ssize_t err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *file, *h_file;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ IiMustWriteLock(inode);
++ file = kio->ki_filp;
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+ err = au_refresh_hinode_self(inode);
-+ if (unlikely(err))
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
+ goto out;
+
-+ sb = dentry->d_sb;
-+ bend = au_sbend(sb);
-+ iinfo = au_ii(inode);
-+ update = 0;
-+ p = iinfo->ii_hinode + iinfo->ii_bstart;
-+ first = p->hi_inode;
-+ isdir = S_ISDIR(inode->i_mode);
-+ flags = au_hi_flags(inode, isdir);
-+ bend = au_dbend(dentry);
-+ for (bindex = au_dbstart(dentry); bindex <= bend; bindex++) {
-+ struct inode *hi;
-+ struct dentry *hd;
++ err = au_do_iter(h_file, MAY_WRITE, kio, iov_iter);
++ au_write_post(inode, h_file, &wpre, err);
+
-+ hd = au_h_dptr(dentry, bindex);
-+ if (!hd || !hd->d_inode)
-+ continue;
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++ return err;
++}
+
-+ if (iinfo->ii_bstart <= bindex && bindex <= iinfo->ii_bend) {
-+ hi = au_h_iptr(inode, bindex);
-+ if (hi) {
-+ if (hi == hd->d_inode)
-+ continue;
-+ err = -ESTALE;
-+ break;
-+ }
-+ }
-+ if (bindex < iinfo->ii_bstart)
-+ iinfo->ii_bstart = bindex;
-+ if (iinfo->ii_bend < bindex)
-+ iinfo->ii_bend = bindex;
-+ au_set_h_iptr(inode, bindex, au_igrab(hd->d_inode), flags);
-+ update++;
-+ }
-+ au_update_brange(inode, /*do_put_zero*/0);
++/*
++ * We may be able to remove aufs_splice_{read,write}() since almost all FSes
++ * don't have their own .splice_{read,write} implimentations, and they use
++ * generic_file_splice_read() and iter_file_splice_write() who can act like the
++ * simple converters to f_op->iter_read() and ->iter_write().
++ * But we keep our own implementations because some non-mainlined FSes may have
++ * their own .splice_{read,write} implimentations and aufs doesn't want to take
++ * away an opportunity to co-work with aufs from them.
++ */
++static ssize_t aufs_splice_read(struct file *file, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags)
++{
++ ssize_t err;
++ struct file *h_file;
++ struct inode *inode;
++ struct super_block *sb;
+
-+ if (unlikely(err))
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
+ goto out;
+
-+ if (1 || first != au_h_iptr(inode, iinfo->ii_bstart))
-+ au_cpup_attr_all(inode, /*force*/0);
-+ if (update && isdir)
-+ inode->i_version++;
-+ au_update_iigen(inode);
++ err = vfsub_splice_to(h_file, ppos, pipe, len, flags);
++ /* todo: necessary? */
++ /* file->f_ra = h_file->f_ra; */
++ au_read_post(inode, h_file);
+
-+ out:
-+ AuTraceErr(err);
++out:
++ si_read_unlock(sb);
+ return err;
+}
+
-+static int set_inode(struct inode *inode, struct dentry *dentry)
++static ssize_t
++aufs_splice_write(struct pipe_inode_info *pipe, struct file *file, loff_t *ppos,
++ size_t len, unsigned int flags)
+{
-+ int err;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ umode_t mode;
-+ aufs_bindex_t bindex, bstart, btail;
-+ unsigned char isdir;
-+ struct au_iinfo *iinfo;
-+ unsigned int flags;
++ ssize_t err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
+
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(dentry));
-+ AuDebugOn(!(inode->i_state & I_NEW));
-+ IiMustWriteLock(inode);
-+ bstart = au_dbstart(dentry);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ AuDebugOn(!h_dentry);
-+ h_inode = h_dentry->d_inode;
-+ AuDebugOn(!h_inode);
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+ err = 0;
-+ isdir = 0;
-+ mode = h_inode->i_mode;
-+ switch (mode & S_IFMT) {
-+ case S_IFREG:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = &aufs_iop;
-+ inode->i_fop = &aufs_file_fop;
-+ inode->i_mapping->a_ops = &aufs_aop;
-+ break;
-+ case S_IFDIR:
-+ isdir = 1;
-+ btail = au_dbtaildir(dentry);
-+ inode->i_op = &aufs_dir_iop;
-+ inode->i_fop = &aufs_dir_fop;
-+ break;
-+ case S_IFLNK:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = &aufs_symlink_iop;
-+ /* inode->i_fop = &aufs_file_fop; */
-+ break;
-+ case S_IFBLK:
-+ case S_IFCHR:
-+ case S_IFIFO:
-+ case S_IFSOCK:
-+ btail = au_dbtail(dentry);
-+ inode->i_op = &aufs_iop;
-+ init_special_inode(inode, mode,
-+ au_h_rdev(h_inode, /*h_mnt*/NULL, h_dentry));
-+ break;
-+ default:
-+ AuIOErr("Unknown file type 0%o\n", mode);
-+ err = -EIO;
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
+ goto out;
-+ }
+
-+ /* do not set inotify for whiteouted dirs (SHWH mode) */
-+ flags = au_hi_flags(inode, isdir);
-+ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)
-+ && au_ftest_hi(flags, NOTIFY)
-+ && dentry->d_name.len > AUFS_WH_PFX_LEN
-+ && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))
-+ au_fclr_hi(flags, NOTIFY);
-+ iinfo = au_ii(inode);
-+ iinfo->ii_bstart = bstart;
-+ iinfo->ii_bend = btail;
-+ for (bindex = bstart; bindex <= btail; bindex++) {
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (!h_dentry)
-+ continue;
-+ AuDebugOn(!h_dentry->d_inode);
-+ au_set_h_iptr(inode, bindex, au_igrab(h_dentry->d_inode),
-+ flags);
-+ }
-+ au_cpup_attr_all(inode, /*force*/1);
++ err = vfsub_splice_from(pipe, h_file, ppos, len, flags);
++ au_write_post(inode, h_file, &wpre, err);
+
-+ out:
-+ AuTraceErr(err);
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
+ return err;
+}
+
-+/* successful returns with iinfo write_locked */
-+/* todo: return with unlocked? */
-+static int reval_inode(struct inode *inode, struct dentry *dentry, int *matched)
++static long aufs_fallocate(struct file *file, int mode, loff_t offset,
++ loff_t len)
+{
-+ int err;
-+ struct inode *h_inode, *h_dinode;
-+ aufs_bindex_t bindex, bend;
++ long err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
+
-+ LKTRTrace("i%lu, %.*s\n", inode->i_ino, AuDLNPair(dentry));
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
+
-+ *matched = 0;
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
+
++ lockdep_off();
++ err = vfs_fallocate(h_file, mode, offset, len);
++ lockdep_on();
+ /*
-+ * before this function, if aufs got any iinfo lock, it must be only
-+ * one, the parent dir.
-+ * it can happen by UDBA and the obsoleted inode number.
++ * we don't need to call file_modifed() here since au_write_post()
++ * is equivalent and copies-up all timestamps and permission bits.
+ */
-+ err = -EIO;
-+ if (unlikely(inode->i_ino == parent_ino(dentry)))
-+ goto out;
-+
-+ h_dinode = au_h_dptr(dentry, au_dbstart(dentry))->d_inode;
-+ /* mutex_lock_nested(&inode->i_mutex, AuLsc_I_CHILD); */
-+ ii_write_lock_new_child(inode);
-+ /* it happend */
-+ if (unlikely(IS_DEADDIR(inode)))
-+ goto out_unlock;
-+
-+ err = 0;
-+ bend = au_ibend(inode);
-+ for (bindex = au_ibstart(inode); bindex <= bend; bindex++) {
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (h_inode && h_inode == h_dinode) {
-+ /* && (ibs != bstart
-+ || !au_test_higen(inode, h_inode))); */
-+ *matched = 1;
-+ err = 0;
-+ if (au_iigen(inode) != au_digen(dentry))
-+ err = au_refresh_hinode(inode, dentry);
-+ break;
-+ }
-+ }
++ au_write_post(inode, h_file, &wpre, /*written*/1);
+
-+ out_unlock:
-+ if (unlikely(err))
-+ ii_write_unlock(inode);
-+ /* mutex_unlock(&inode->i_mutex); */
-+ out:
-+ AuTraceErr(err);
++out:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
+ return err;
+}
+
-+/* successful returns with iinfo write_locked */
-+/* todo: return with unlocked? */
-+struct inode *au_new_inode(struct dentry *dentry, int must_new)
++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)
+{
-+ struct inode *inode, *h_inode;
-+ struct dentry *h_dentry;
-+ struct super_block *sb;
-+ ino_t h_ino;
-+ int err, match;
-+ aufs_bindex_t bstart;
-+ struct au_xino_entry xinoe;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ sb = dentry->d_sb;
-+ bstart = au_dbstart(dentry);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ AuDebugOn(!h_dentry);
-+ h_inode = h_dentry->d_inode;
-+ AuDebugOn(!h_inode);
++ 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]
+
-+ h_ino = h_inode->i_ino;
-+ err = au_xino_read(sb, bstart, h_ino, &xinoe);
-+ inode = ERR_PTR(err);
-+ if (unlikely(err))
++ err = -EINVAL;
++ a_src.inode = file_inode(src);
++ if (unlikely(!S_ISREG(a_src.inode->i_mode)))
+ goto out;
-+ new_ino:
-+ if (!xinoe.ino) {
-+ xinoe.ino = au_xino_new_ino(sb);
-+ if (!xinoe.ino) {
-+ inode = ERR_PTR(-EIO);
-+ goto out;
-+ }
-+ }
-+
-+ LKTRTrace("i%lu\n", (unsigned long)xinoe.ino);
-+ inode = au_iget_locked(sb, xinoe.ino);
-+ err = PTR_ERR(inode);
-+ if (IS_ERR(inode))
++ a_dst.inode = file_inode(dst);
++ if (unlikely(!S_ISREG(a_dst.inode->i_mode)))
+ goto out;
+
-+ LKTRTrace("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW));
-+ if (inode->i_state & I_NEW) {
-+ ii_write_lock_new_child(inode);
-+ err = set_inode(inode, dentry);
-+ unlock_new_inode(inode);
-+ if (!err)
-+ goto out; /* success */
-+ iget_failed(inode);
-+ ii_write_unlock(inode);
-+ goto out_iput;
-+ } else if (!must_new) {
-+ AuDebugOn(inode->i_state & I_LOCK);
-+ err = reval_inode(inode, dentry, &match);
-+ if (!err)
-+ goto out; /* success */
-+ else if (match)
-+ goto out_iput;
++ 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 comparison.
++ */
++ 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;
++ }
+ }
+
-+ if (unlikely(au_test_unique_ino(h_dentry, h_ino)))
-+ AuWarn1("Un-notified UDBA or repeatedly renamed dir,"
-+ " b%d, %s, %.*s, hi%lu, i%lu.\n",
-+ bstart, au_sbtype(h_dentry->d_sb), AuDLNPair(dentry),
-+ (unsigned long)h_ino, (unsigned long)xinoe.ino);
-+ xinoe.ino = 0;
-+ err = au_xino_write0(sb, bstart, h_ino, 0);
-+ if (!err) {
-+ iput(inode);
-+ goto new_ino;
++ 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;
+ }
-+ /* force noxino? */
+
-+ out_iput:
-+ iput(inode);
-+ inode = ERR_PTR(err);
-+ out:
-+ AuTraceErrPtr(inode);
-+ return inode;
++ 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
+}
+
+/* ---------------------------------------------------------------------- */
+
-+int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode)
++/*
++ * The locking order around current->mmap_sem.
++ * - in most and regular cases
++ * file I/O syscall -- aufs_read() or something
++ * -- si_rwsem for read -- mmap_sem
++ * (Note that [fdi]i_rwsem are released before mmap_sem).
++ * - in mmap case
++ * mmap(2) -- mmap_sem -- aufs_mmap() -- si_rwsem for read -- [fdi]i_rwsem
++ * This AB-BA order is definitely bad, but is not a problem since "si_rwsem for
++ * read" allows multiple processes to acquire it and [fdi]i_rwsem are not held
++ * in file I/O. Aufs needs to stop lockdep in aufs_mmap() though.
++ * It means that when aufs acquires si_rwsem for write, the process should never
++ * acquire mmap_sem.
++ *
++ * Actually aufs_iterate() holds [fdi]i_rwsem before mmap_sem, but this is not a
++ * problem either since any directory is not able to be mmap-ed.
++ * The similar scenario is applied to aufs_readlink() too.
++ */
++
++#if 0 /* stop calling security_file_mmap() */
++/* cf. linux/include/linux/mman.h: calc_vm_prot_bits() */
++#define AuConv_VM_PROT(f, b) _calc_vm_trans(f, VM_##b, PROT_##b)
++
++static unsigned long au_arch_prot_conv(unsigned long flags)
++{
++ /* currently ppc64 only */
++#ifdef CONFIG_PPC64
++ /* cf. linux/arch/powerpc/include/asm/mman.h */
++ AuDebugOn(arch_calc_vm_prot_bits(-1) != VM_SAO);
++ return AuConv_VM_PROT(flags, SAO);
++#else
++ AuDebugOn(arch_calc_vm_prot_bits(-1));
++ return 0;
++#endif
++}
++
++static unsigned long au_prot_conv(unsigned long flags)
++{
++ return AuConv_VM_PROT(flags, READ)
++ | AuConv_VM_PROT(flags, WRITE)
++ | AuConv_VM_PROT(flags, EXEC)
++ | au_arch_prot_conv(flags);
++}
++
++/* cf. linux/include/linux/mman.h: calc_vm_flag_bits() */
++#define AuConv_VM_MAP(f, b) _calc_vm_trans(f, VM_##b, MAP_##b)
++
++static unsigned long au_flag_conv(unsigned long flags)
++{
++ return AuConv_VM_MAP(flags, GROWSDOWN)
++ | AuConv_VM_MAP(flags, DENYWRITE)
++ | AuConv_VM_MAP(flags, LOCKED);
++}
++#endif
++
++static int aufs_mmap(struct file *file, struct vm_area_struct *vma)
+{
+ int err;
++ const unsigned char wlock
++ = (file->f_mode & FMODE_WRITE) && (vma->vm_flags & VM_SHARED);
++ struct super_block *sb;
++ struct file *h_file;
++ struct inode *inode;
+
-+ err = au_br_rdonly(au_sbr(sb, bindex));
++ AuDbgVmRegion(file, vma);
+
-+ /* pseudo-link after flushed may out of bounds */
-+ if (!err
-+ && inode
-+ && au_ibstart(inode) <= bindex
-+ && bindex <= au_ibend(inode)) {
-+ /*
-+ * permission check is unnecessary since vfsub routine
-+ * will be called later
-+ */
-+ struct inode *hi = au_h_iptr(inode, bindex);
-+ if (hi)
-+ err = IS_IMMUTABLE(hi) ? -EROFS : 0;
++ inode = file_inode(file);
++ sb = inode->i_sb;
++ lockdep_off();
++ si_read_lock(sb, AuLock_NOPLMW);
++
++ h_file = au_write_pre(file, wlock, /*wpre*/NULL);
++ lockdep_on();
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++
++ err = 0;
++ au_set_mmapped(file);
++ au_vm_file_reset(vma, h_file);
++ /*
++ * we cannot call security_mmap_file() here since it may acquire
++ * mmap_sem or i_mutex.
++ *
++ * err = security_mmap_file(h_file, au_prot_conv(vma->vm_flags),
++ * au_flag_conv(vma->vm_flags));
++ */
++ if (!err)
++ err = call_mmap(h_file, vma);
++ if (!err) {
++ au_vm_prfile_set(vma, file);
++ fsstack_copy_attr_atime(inode, file_inode(h_file));
++ goto out_fput; /* success */
+ }
++ au_unset_mmapped(file);
++ au_vm_file_reset(vma, file);
+
++out_fput:
++ lockdep_off();
++ ii_write_unlock(inode);
++ lockdep_on();
++ fput(h_file);
++out:
++ lockdep_off();
++ si_read_unlock(sb);
++ lockdep_on();
+ AuTraceErr(err);
+ return err;
+}
+
-+int au_test_h_perm(struct inode *h_inode, int mask, int dlgt)
++/* ---------------------------------------------------------------------- */
++
++static int aufs_fsync_nondir(struct file *file, loff_t start, loff_t end,
++ int datasync)
+{
-+ if (!current->fsuid)
-+ return 0;
-+ /* todo: fake nameidata? */
-+ return vfsub_permission(h_inode, mask, NULL, dlgt);
++ int err;
++ struct au_write_pre wpre;
++ struct inode *inode;
++ struct file *h_file;
++
++ err = 0; /* -EBADF; */ /* posix? */
++ if (unlikely(!(file->f_mode & FMODE_WRITE)))
++ goto out;
++
++ inode = file_inode(file);
++ au_mtx_and_read_lock(inode);
++
++ wpre.lsc = 0;
++ h_file = au_write_pre(file, /*do_ready*/1, &wpre);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out_unlock;
++
++ err = vfsub_fsync(h_file, &h_file->f_path, datasync);
++ au_write_post(inode, h_file, &wpre, /*written*/0);
++
++out_unlock:
++ si_read_unlock(inode->i_sb);
++ inode_unlock(inode);
++out:
++ return err;
+}
+
-+int au_test_h_perm_sio(struct inode *h_inode, int mask, int dlgt)
++static int aufs_fasync(int fd, struct file *file, int flag)
++{
++ int err;
++ struct file *h_file;
++ struct super_block *sb;
++
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++
++ if (h_file->f_op->fasync)
++ err = h_file->f_op->fasync(fd, h_file, flag);
++ fput(h_file); /* instead of au_read_post() */
++
++out:
++ si_read_unlock(sb);
++ return err;
++}
++
++static int aufs_setfl(struct file *file, unsigned long arg)
++{
++ int err;
++ struct file *h_file;
++ struct super_block *sb;
++
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
++
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out;
++
++ /* stop calling h_file->fasync */
++ arg |= vfsub_file_flags(file) & FASYNC;
++ err = setfl(/*unused fd*/-1, h_file, arg);
++ fput(h_file); /* instead of au_read_post() */
++
++out:
++ si_read_unlock(sb);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* no one supports this operation, currently */
++#if 0 /* reserved for future use */
++static ssize_t aufs_sendpage(struct file *file, struct page *page, int offset,
++ size_t len, loff_t *pos, int more)
+{
-+ if (au_test_nfs(h_inode->i_sb)
-+ && (mask & MAY_WRITE)
-+ && S_ISDIR(h_inode->i_mode))
-+ mask |= MAY_READ; /* force permission check */
-+ return au_test_h_perm(h_inode, mask, dlgt);
+}
-diff -urN linux/fs/aufs/inode.h linux-aufs/fs/aufs/inode.h
---- linux/fs/aufs/inode.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/inode.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,595 @@
++#endif
++
++/* ---------------------------------------------------------------------- */
++
++const struct file_operations aufs_file_fop = {
++ .owner = THIS_MODULE,
++
++ .llseek = default_llseek,
++
++ .read_iter = aufs_read_iter,
++ .write_iter = aufs_write_iter,
++
++#ifdef CONFIG_AUFS_POLL
++ .poll = aufs_poll,
++#endif
++ .unlocked_ioctl = aufs_ioctl_nondir,
++#ifdef CONFIG_COMPAT
++ .compat_ioctl = aufs_compat_ioctl_nondir,
++#endif
++ .mmap = aufs_mmap,
++ .open = aufs_open_nondir,
++ .flush = aufs_flush_nondir,
++ .release = aufs_release_nondir,
++ .fsync = aufs_fsync_nondir,
++ .fasync = aufs_fasync,
++ /* .sendpage = aufs_sendpage, */
++ .setfl = aufs_setfl,
++ .splice_write = aufs_splice_write,
++ .splice_read = aufs_splice_read,
++#if 0 /* reserved for future use */
++ .aio_splice_write = aufs_aio_splice_write,
++ .aio_splice_read = aufs_aio_splice_read,
++#endif
++ .fallocate = aufs_fallocate,
++ .copy_file_range = aufs_copy_file_range
++};
+diff -urN /usr/share/empty/fs/aufs/fsctx.c linux/fs/aufs/fsctx.c
+--- /usr/share/empty/fs/aufs/fsctx.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/fsctx.c 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,1242 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode operations
-+ *
-+ * $Id$
++ * fs context, aka new mount api
+ */
+
-+#ifndef __AUFS_INODE_H__
-+#define __AUFS_INODE_H__
-+
-+#ifdef __KERNEL__
++#include <linux/fs_context.h>
++#include "aufs.h"
+
-+#include <linux/fs.h>
-+#include <linux/namei.h>
-+#include <linux/security.h>
-+#include <linux/aufs_type.h>
-+#include "hinode.h"
-+#include "misc.h"
-+#include "super.h"
++struct au_fsctx_opts {
++ aufs_bindex_t bindex;
++ unsigned char skipped;
++ struct au_opt *opt, *opt_tail;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct au_opts opts;
++};
+
-+struct au_hinode;
-+struct au_vdir;
-+struct au_iinfo {
-+ atomic_t ii_generation;
-+ struct super_block *ii_hsb1; /* no get/put */
++/* stop extra interpretation of errno in mount(8), and strange error messages */
++static int cvt_err(int err)
++{
++ AuTraceErr(err);
+
-+ struct au_rwsem ii_rwsem;
-+ aufs_bindex_t ii_bstart, ii_bend;
-+ __u32 ii_higen;
-+ struct au_hinode *ii_hinode;
-+ struct au_vdir *ii_vdir;
-+};
++ switch (err) {
++ case -ENOENT:
++ case -ENOTDIR:
++ case -EEXIST:
++ case -EIO:
++ err = -EINVAL;
++ }
++ return err;
++}
+
-+struct aufs_icntnr {
-+ struct au_iinfo iinfo;
-+ struct inode vfs_inode;
-+};
++static int au_fsctx_reconfigure(struct fs_context *fc)
++{
++ int err, do_dx;
++ unsigned int mntflags;
++ struct dentry *root;
++ struct super_block *sb;
++ struct inode *inode;
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+/* au_pin flags */
-+#define AuPin_DI_LOCKED 1
-+#define AuPin_DO_GPARENT (1 << 1)
-+#define AuPin_MNT_WRITE (1 << 2)
-+#define AuPin_VFS_RENAME (1 << 3)
-+/* will be set automatically */
-+#define AuPin_VERIFY (1 << 4)
-+#define au_ftest_pin(flags, name) ((flags) & AuPin_##name)
-+#define au_fset_pin(flags, name) { (flags) |= AuPin_##name; }
-+#define au_fclr_pin(flags, name) { (flags) &= ~AuPin_##name; }
++ AuDbg("fc %p\n", fc);
+
-+struct au_pin1 {
-+ /* input */
-+ struct dentry *dentry;
-+ unsigned char lsc_di, lsc_hi, flags;
-+ aufs_bindex_t bindex;
++ root = fc->root;
++ sb = root->d_sb;
++ err = si_write_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (!err) {
++ di_write_lock_child(root);
++ err = au_opts_verify(sb, fc->sb_flags, /*pending*/0);
++ aufs_write_unlock(root);
++ }
+
-+ /* output */
-+ struct dentry *parent;
-+ struct inode *h_dir;
-+};
++ inode = d_inode(root);
++ inode_lock(inode);
++ err = si_write_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out;
++ di_write_lock_child(root);
+
-+enum {AuPin_PARENT, AuPin_GPARENT};
-+struct au_pin {
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ struct au_pin1 pin[2];
-+#else
-+ struct au_pin1 pin[1]; /* no grand parent */
-+#endif
-+};
++ /* au_opts_remount() may return an error */
++ err = au_opts_remount(sb, &a->opts);
+
-+/* ---------------------------------------------------------------------- */
++ if (au_ftest_opts(a->opts.flags, REFRESH))
++ au_remount_refresh(sb, au_ftest_opts(a->opts.flags,
++ REFRESH_IDOP));
+
-+/* inode.c */
-+int au_refresh_hinode_self(struct inode *inode);
-+int au_refresh_hinode(struct inode *inode, struct dentry *dentry);
-+struct inode *au_new_inode(struct dentry *dentry, int must_new);
-+int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode);
-+int au_test_h_perm(struct inode *h_inode, int mask, int dlgt);
-+int au_test_h_perm_sio(struct inode *h_inode, int mask, int dlgt);
++ if (au_ftest_opts(a->opts.flags, REFRESH_DYAOP)) {
++ mntflags = au_mntflags(sb);
++ do_dx = !!au_opt_test(mntflags, DIO);
++ au_dy_arefresh(do_dx);
++ }
+
-+/* i_op.c */
-+extern struct inode_operations aufs_iop, aufs_symlink_iop, aufs_dir_iop;
++ au_fhsm_wrote_all(sb, /*force*/1); /* ?? */
++ aufs_write_unlock(root);
+
-+/* au_wr_dir flags */
-+#define AuWrDir_ADD_ENTRY 1
-+#define AuWrDir_ISDIR (1 << 1)
-+#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name)
-+#define au_fset_wrdir(flags, name) { (flags) |= AuWrDir_##name; }
-+#define au_fclr_wrdir(flags, name) { (flags) &= ~AuWrDir_##name; }
++out:
++ inode_unlock(inode);
++ err = cvt_err(err);
++ AuTraceErr(err);
+
-+struct au_wr_dir_args {
-+ aufs_bindex_t force_btgt;
-+ unsigned char flags;
-+};
-+int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
-+ struct au_wr_dir_args *args);
++ return err;
++}
+
-+void au_pin_init(struct au_pin *args, struct dentry *dentry,
-+ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
-+ unsigned char flags);
-+int au_pin(struct au_pin *args, struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned char flags) __must_check;
-+int au_do_pin(struct au_pin1 *p, struct au_pin1 *gp) __must_check;
-+void au_do_unpin(struct au_pin1 *p, struct au_pin1 *gp);
++/* ---------------------------------------------------------------------- */
+
-+/* i_op_add.c */
-+struct au_ndx;
-+int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir, struct au_ndx *ndx);
-+int aufs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev);
-+int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname);
-+int aufs_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd);
-+int aufs_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry);
-+int aufs_mkdir(struct inode *dir, struct dentry *dentry, int mode);
++static int au_fsctx_fill_super(struct super_block *sb, struct fs_context *fc)
++{
++ int err;
++ struct au_fsctx_opts *a = fc->fs_private;
++ struct au_sbinfo *sbinfo = a->sbinfo;
++ struct dentry *root;
++ struct inode *inode;
+
-+/* i_op_del.c */
-+int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup);
-+int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir, struct au_ndx *ndx);
-+int aufs_unlink(struct inode *dir, struct dentry *dentry);
-+int aufs_rmdir(struct inode *dir, struct dentry *dentry);
++ sbinfo->si_sb = sb;
++ sb->s_fs_info = sbinfo;
++ kobject_get(&sbinfo->si_kobj);
+
-+/* 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);
++ __si_write_lock(sb);
++ si_pid_set(sb);
++ au_sbilist_add(sb);
+
-+#ifdef CONFIG_AUFS_DLGT
-+/* dlgt.c */
-+int au_security_inode_permission(struct inode *h_inode, int mask,
-+ struct nameidata *fake_nd, int dlgt);
-+#else
-+static inline
-+int au_security_inode_permission(struct inode *h_inode, int mask,
-+ struct nameidata *fake_nd, int dlgt)
++ /* all timestamps always follow the ones on the branch */
++ sb->s_flags |= SB_NOATIME | SB_NODIRATIME;
++ sb->s_flags |= SB_I_VERSION; /* do we really need this? */
++ sb->s_op = &aufs_sop;
++ sb->s_d_op = &aufs_dop;
++ sb->s_magic = AUFS_SUPER_MAGIC;
++ sb->s_maxbytes = 0;
++ sb->s_stack_depth = 1;
++ au_export_init(sb);
++ au_xattr_init(sb);
++
++ err = au_alloc_root(sb);
++ if (unlikely(err)) {
++ si_write_unlock(sb);
++ goto out;
++ }
++ root = sb->s_root;
++ inode = d_inode(root);
++ ii_write_lock_parent(inode);
++ aufs_write_unlock(root);
++
++ /* lock vfs_inode first, then aufs. */
++ inode_lock(inode);
++ aufs_write_lock(root);
++ err = au_opts_mount(sb, &a->opts);
++ AuTraceErr(err);
++ if (!err && au_ftest_si(sbinfo, NO_DREVAL)) {
++ sb->s_d_op = &aufs_dop_noreval;
++ /* infofc(fc, "%ps", sb->s_d_op); */
++ pr_info("%ps\n", sb->s_d_op);
++ au_refresh_dop(root, /*force_reval*/0);
++ sbinfo->si_iop_array = aufs_iop_nogetattr;
++ au_refresh_iop(inode, /*force_getattr*/0);
++ }
++ aufs_write_unlock(root);
++ inode_unlock(inode);
++ if (!err)
++ goto out; /* success */
++
++ dput(root);
++ sb->s_root = NULL;
++
++out:
++ if (unlikely(err))
++ kobject_put(&sbinfo->si_kobj);
++ AuTraceErr(err);
++ err = cvt_err(err);
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_fsctx_get_tree(struct fs_context *fc)
+{
-+ return vfsub_security_inode_permission(h_inode, mask, fake_nd);
++ int err;
++
++ AuDbg("fc %p\n", fc);
++ err = get_tree_nodev(fc, au_fsctx_fill_super);
++
++ AuTraceErr(err);
++ return err;
+}
-+#endif /* CONFIG_AUFS_DLGT */
+
-+#ifdef CONFIG_AUFS_GETATTR
-+/* getattr.c */
-+int aufs_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *st);
-+#endif
++/* ---------------------------------------------------------------------- */
+
-+#if 0 /* reserved for future use */
-+/* xattr.c */
-+int aufs_setxattr(struct dentry *dentry, const char *name, const void *value,
-+ size_t sz, int flags);
-+ssize_t aufs_getxattr(struct dentry *dentry, const char *name, void *value,
-+ size_t sz);
-+ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t sz);
-+int aufs_removexattr(struct dentry *dentry, const char *name);
-+#endif
++static void au_fsctx_dump(struct au_opts *opts)
++{
++#ifdef CONFIG_AUFS_DEBUG
++ /* reduce stack space */
++ union {
++ struct au_opt_add *add;
++ struct au_opt_del *del;
++ struct au_opt_mod *mod;
++ struct au_opt_xino *xino;
++ struct au_opt_xino_itrunc *xino_itrunc;
++ struct au_opt_wbr_create *create;
++ } u;
++ struct au_opt *opt;
+
-+/* iinfo.c */
-+struct au_iinfo *au_ii(struct inode *inode);
-+struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex);
-+aufs_bindex_t au_ii_br_id(struct inode *inode, aufs_bindex_t bindex);
++ opt = opts->opt;
++ while (opt->type != Opt_tail) {
++ switch (opt->type) {
++ case Opt_add:
++ u.add = &opt->add;
++ AuDbg("add {b%d, %s, 0x%x, %p}\n",
++ u.add->bindex, u.add->pathname, u.add->perm,
++ u.add->path.dentry);
++ break;
++ case Opt_del:
++ fallthrough;
++ case Opt_idel:
++ u.del = &opt->del;
++ AuDbg("del {%s, %p}\n",
++ u.del->pathname, u.del->h_path.dentry);
++ break;
++ case Opt_mod:
++ fallthrough;
++ case Opt_imod:
++ u.mod = &opt->mod;
++ AuDbg("mod {%s, 0x%x, %p}\n",
++ u.mod->path, u.mod->perm, u.mod->h_root);
++ break;
++ case Opt_append:
++ u.add = &opt->add;
++ AuDbg("append {b%d, %s, 0x%x, %p}\n",
++ u.add->bindex, u.add->pathname, u.add->perm,
++ u.add->path.dentry);
++ break;
++ case Opt_prepend:
++ u.add = &opt->add;
++ AuDbg("prepend {b%d, %s, 0x%x, %p}\n",
++ u.add->bindex, u.add->pathname, u.add->perm,
++ u.add->path.dentry);
++ break;
+
-+void au_set_ibstart(struct inode *inode, aufs_bindex_t bindex);
-+void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
-+ struct dentry *h_wh);
-+unsigned int au_hi_flags(struct inode *inode, int isdir);
++ case Opt_dirwh:
++ AuDbg("dirwh %d\n", opt->dirwh);
++ break;
++ case Opt_rdcache:
++ AuDbg("rdcache %d\n", opt->rdcache);
++ break;
++ case Opt_rdblk:
++ AuDbg("rdblk %d\n", opt->rdblk);
++ break;
++ case Opt_rdhash:
++ AuDbg("rdhash %u\n", opt->rdhash);
++ break;
+
-+/* hinode flags */
-+#define AuHi_XINO 1
-+#define AuHi_NOTIFY (1 << 1)
-+#define au_ftest_hi(flags, name) ((flags) & AuHi_##name)
-+#define au_fset_hi(flags, name) { (flags) |= AuHi_##name; }
-+#define au_fclr_hi(flags, name) { (flags) &= ~AuHi_##name; }
-+#ifndef CONFIG_AUFS_HINOTIFY
-+#undef AuHi_NOTIFY
-+#define AuHi_NOTIFY 0
-+#endif
++ case Opt_xino:
++ u.xino = &opt->xino;
++ AuDbg("xino {%s %pD}\n", u.xino->path, u.xino->file);
++ break;
+
-+void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
-+ struct inode *h_inode, unsigned int flags);
++#define au_fsctx_TF(name) \
++ case Opt_##name: \
++ if (opt->tf) \
++ AuLabel(name); \
++ else \
++ AuLabel(no##name); \
++ break;
+
-+void au_update_iigen(struct inode *inode);
-+void au_update_brange(struct inode *inode, int do_put_zero);
++ /* simple true/false flag */
++ au_fsctx_TF(trunc_xino);
++ au_fsctx_TF(trunc_xib);
++ au_fsctx_TF(dirperm1);
++ au_fsctx_TF(plink);
++ au_fsctx_TF(shwh);
++ au_fsctx_TF(dio);
++ au_fsctx_TF(warn_perm);
++ au_fsctx_TF(verbose);
++ au_fsctx_TF(sum);
++ au_fsctx_TF(dirren);
++ au_fsctx_TF(acl);
++#undef au_fsctx_TF
+
-+int au_iinfo_init(struct inode *inode);
-+void au_iinfo_fin(struct inode *inode);
++ case Opt_trunc_xino_path:
++ fallthrough;
++ case Opt_itrunc_xino:
++ u.xino_itrunc = &opt->xino_itrunc;
++ AuDbg("trunc_xino %d\n", u.xino_itrunc->bindex);
++ break;
++ case Opt_noxino:
++ AuLabel(noxino);
++ break;
+
-+/* plink.c */
-+#ifdef CONFIG_AUFS_DEBUG
-+void au_plink_list(struct super_block *sb);
-+#else
-+static inline void au_plink_list(struct super_block *sb)
-+{
-+ /* nothing */
-+}
++ case Opt_list_plink:
++ AuLabel(list_plink);
++ break;
++ case Opt_udba:
++ AuDbg("udba %d, %s\n",
++ opt->udba, au_optstr_udba(opt->udba));
++ break;
++ case Opt_diropq_a:
++ AuLabel(diropq_a);
++ break;
++ case Opt_diropq_w:
++ AuLabel(diropq_w);
++ break;
++ case Opt_wsum:
++ AuLabel(wsum);
++ break;
++ case Opt_wbr_create:
++ u.create = &opt->wbr_create;
++ AuDbg("create %d, %s\n", u.create->wbr_create,
++ au_optstr_wbr_create(u.create->wbr_create));
++ switch (u.create->wbr_create) {
++ case AuWbrCreate_MFSV:
++ fallthrough;
++ case AuWbrCreate_PMFSV:
++ AuDbg("%d sec\n", u.create->mfs_second);
++ break;
++ case AuWbrCreate_MFSRR:
++ fallthrough;
++ case AuWbrCreate_TDMFS:
++ AuDbg("%llu watermark\n",
++ u.create->mfsrr_watermark);
++ break;
++ case AuWbrCreate_MFSRRV:
++ fallthrough;
++ case AuWbrCreate_TDMFSV:
++ fallthrough;
++ case AuWbrCreate_PMFSRRV:
++ AuDbg("%llu watermark, %d sec\n",
++ u.create->mfsrr_watermark,
++ u.create->mfs_second);
++ break;
++ }
++ break;
++ case Opt_wbr_copyup:
++ AuDbg("copyup %d, %s\n", opt->wbr_copyup,
++ au_optstr_wbr_copyup(opt->wbr_copyup));
++ break;
++ case Opt_fhsm_sec:
++ AuDbg("fhsm_sec %u\n", opt->fhsm_second);
++ break;
++
++ default:
++ AuDbg("type %d\n", opt->type);
++ BUG();
++ }
++ opt++;
++ }
+#endif
-+int au_plink_test(struct super_block *sb, struct inode *inode);
-+struct dentry *au_plink_lkup(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode);
-+void au_plink_append(struct super_block *sb, struct inode *inode,
-+ struct dentry *h_dentry, aufs_bindex_t bindex);
-+void au_plink_put(struct super_block *sb);
-+void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id);
++}
+
+/* ---------------------------------------------------------------------- */
+
-+/* lock subclass for iinfo */
-+enum {
-+ AuLsc_II_CHILD, /* child first */
-+ AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hinotify */
-+ AuLsc_II_CHILD3, /* copyup dirs */
-+ AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */
-+ AuLsc_II_PARENT2,
-+ AuLsc_II_PARENT3,
-+ AuLsc_II_PARENT4,
-+ AuLsc_II_NEW_CHILD,
-+};
-+
+/*
-+ * ii_read_lock_child, ii_write_lock_child,
-+ * ii_read_lock_child2, ii_write_lock_child2,
-+ * ii_read_lock_child3, ii_write_lock_child3,
-+ * ii_read_lock_parent, ii_write_lock_parent,
-+ * ii_read_lock_parent2, ii_write_lock_parent2,
-+ * ii_read_lock_parent3, ii_write_lock_parent3,
-+ * ii_read_lock_parent4, ii_write_lock_parent4,
-+ * ii_read_lock_new_child, ii_write_lock_new_child,
++ * For conditionally compiled mount options.
++ * Instead of fsparam_flag_no(), use this macro to distinguish ignore_silent.
+ */
-+#define AuReadLockFunc(name, lsc) \
-+static inline void ii_read_lock_##name(struct inode *i) \
-+{ \
-+ au_dbg_locking_ii_reg(i, 0, AuLsc_II_##lsc); \
-+ au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
-+ au_dbg_locking_ii_unreg(i, 0); \
-+ au_dbg_locked_ii_reg(i, 0, AuLsc_II_##lsc); \
-+}
++#define au_ignore_flag(name, action) \
++ fsparam_flag(name, action), \
++ fsparam_flag("no" name, Opt_ignore_silent)
++
++const struct fs_parameter_spec aufs_fsctx_paramspec[] = {
++ fsparam_string("br", Opt_br),
++
++ /* "add=%d:%s" or "ins=%d:%s" */
++ fsparam_string("add", Opt_add),
++ fsparam_string("ins", Opt_add),
++ fsparam_path("append", Opt_append),
++ fsparam_path("prepend", Opt_prepend),
++
++ fsparam_path("del", Opt_del),
++ /* fsparam_s32("idel", Opt_idel), */
++ fsparam_path("mod", Opt_mod),
++ /* fsparam_string("imod", Opt_imod), */
++
++ fsparam_s32("dirwh", Opt_dirwh),
++
++ fsparam_path("xino", Opt_xino),
++ fsparam_flag("noxino", Opt_noxino),
++ fsparam_flag_no("trunc_xino", Opt_trunc_xino),
++ /* "trunc_xino_v=%d:%d" */
++ /* fsparam_string("trunc_xino_v", Opt_trunc_xino_v), */
++ fsparam_path("trunc_xino", Opt_trunc_xino_path),
++ fsparam_s32("itrunc_xino", Opt_itrunc_xino),
++ /* fsparam_path("zxino", Opt_zxino), */
++ fsparam_flag_no("trunc_xib", Opt_trunc_xib),
++
++#ifdef CONFIG_PROC_FS
++ fsparam_flag_no("plink", Opt_plink),
++#else
++ au_ignore_flag("plink", Opt_ignore),
++#endif
+
-+#define AuWriteLockFunc(name, lsc) \
-+static inline void ii_write_lock_##name(struct inode *i) \
-+{ \
-+ au_dbg_locking_ii_reg(i, 1, AuLsc_II_##lsc); \
-+ au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
-+ au_dbg_locking_ii_unreg(i, 1); \
-+ au_dbg_locked_ii_reg(i, 1, AuLsc_II_##lsc); \
-+}
++#ifdef CONFIG_AUFS_DEBUG
++ fsparam_flag("list_plink", Opt_list_plink),
++#endif
+
-+#define AuRWLockFuncs(name, lsc) \
-+ AuReadLockFunc(name, lsc) \
-+ AuWriteLockFunc(name, lsc)
++ fsparam_string("udba", Opt_udba),
+
-+AuRWLockFuncs(child, CHILD);
-+AuRWLockFuncs(child2, CHILD2);
-+AuRWLockFuncs(child3, CHILD3);
-+AuRWLockFuncs(parent, PARENT);
-+AuRWLockFuncs(parent2, PARENT2);
-+AuRWLockFuncs(parent3, PARENT3);
-+AuRWLockFuncs(parent4, PARENT4);
-+AuRWLockFuncs(new_child, NEW_CHILD);
++ fsparam_flag_no("dio", Opt_dio),
+
-+#undef AuReadLockFunc
-+#undef AuWriteLockFunc
-+#undef AuRWLockFuncs
++#ifdef CONFIG_AUFS_DIRREN
++ fsparam_flag_no("dirren", Opt_dirren),
++#else
++ au_ignore_flag("dirren", Opt_ignore),
++#endif
+
-+/*
-+ * ii_read_unlock, ii_write_unlock, ii_downgrade_lock
-+ */
-+AuSimpleUnlockRwsemFuncs(ii_do, struct inode *i, au_ii(i)->ii_rwsem);
-+static inline void ii_read_unlock(struct inode *i)
-+{
-+ ii_do_read_unlock(i);
-+ au_dbg_locked_ii_unreg(i, 0);
-+}
++#ifdef CONFIG_AUFS_FHSM
++ fsparam_s32("fhsm_sec", Opt_fhsm_sec),
++#else
++ fsparam_s32("fhsm_sec", Opt_ignore),
++#endif
+
-+static inline void ii_write_unlock(struct inode *i)
-+{
-+ ii_do_write_unlock(i);
-+ au_dbg_locked_ii_unreg(i, 0);
-+}
++ /* always | a | whiteouted | w */
++ fsparam_string("diropq", Opt_diropq),
+
-+static inline void ii_downgrade_lock(struct inode *i)
-+{
-+ ii_do_downgrade_lock(i);
-+}
++ fsparam_flag_no("warn_perm", Opt_warn_perm),
+
-+/* to debug easier, do not make them inlined functions */
-+#define IiMustReadLock(i) do { \
-+ SiMustAnyLock((i)->i_sb); \
-+ AuRwMustReadLock(&au_ii(i)->ii_rwsem); \
-+} while (0)
++#ifdef CONFIG_AUFS_SHWH
++ fsparam_flag_no("shwh", Opt_shwh),
++#else
++ au_ignore_flag("shwh", Opt_err),
++#endif
+
-+#define IiMustWriteLock(i) do { \
-+ SiMustAnyLock((i)->i_sb); \
-+ AuRwMustWriteLock(&au_ii(i)->ii_rwsem); \
-+} while (0)
++ fsparam_flag_no("dirperm1", Opt_dirperm1),
+
-+#define IiMustAnyLock(i) do { \
-+ SiMustAnyLock((i)->i_sb); \
-+ AuRwMustAnyLock(&au_ii(i)->ii_rwsem); \
-+} while (0)
++ fsparam_flag_no("verbose", Opt_verbose),
++ fsparam_flag("v", Opt_verbose),
++ fsparam_flag("quiet", Opt_noverbose),
++ fsparam_flag("q", Opt_noverbose),
++ /* user-space may handle this */
++ fsparam_flag("silent", Opt_noverbose),
+
-+#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem)
++ fsparam_flag_no("sum", Opt_sum),
++ fsparam_flag("wsum", Opt_wsum),
+
-+/* ---------------------------------------------------------------------- */
++ fsparam_s32("rdcache", Opt_rdcache),
++ /* "def" or s32 */
++ fsparam_string("rdblk", Opt_rdblk),
++ /* "def" or s32 */
++ fsparam_string("rdhash", Opt_rdhash),
+
-+static inline struct inode *au_igrab(struct inode *inode)
-+{
-+ if (inode) {
-+ AuDebugOn(!atomic_read(&inode->i_count));
-+ atomic_inc_return(&inode->i_count);
-+ }
-+ return inode;
-+}
++ fsparam_string("create", Opt_wbr_create),
++ fsparam_string("create_policy", Opt_wbr_create),
++ fsparam_string("cpup", Opt_wbr_copyup),
++ fsparam_string("copyup", Opt_wbr_copyup),
++ fsparam_string("copyup_policy", Opt_wbr_copyup),
+
-+/* ---------------------------------------------------------------------- */
++ /* generic VFS flag */
++#ifdef CONFIG_FS_POSIX_ACL
++ fsparam_flag_no("acl", Opt_acl),
++#else
++ au_ignore_flag("acl", Opt_ignore),
++#endif
+
-+static inline aufs_bindex_t au_ibstart(struct inode *inode)
-+{
-+ /* lock free root dinfo/inode */
-+ if (inode->i_ino != AUFS_ROOT_INO)
-+ IiMustAnyLock(inode);
-+ else
-+ SiMustAnyLock(inode->i_sb);
-+ return au_ii(inode)->ii_bstart;
-+}
++ /* internal use for the scripts */
++ fsparam_string("si", Opt_ignore_silent),
+
-+static inline aufs_bindex_t au_ibend(struct inode *inode)
-+{
-+ /* lock free root dinfo/inode */
-+ if (inode->i_ino != AUFS_ROOT_INO)
-+ IiMustAnyLock(inode);
-+ else
-+ SiMustAnyLock(inode->i_sb);
-+ return au_ii(inode)->ii_bend;
-+}
++ /* obsoleted, keep them temporary */
++ fsparam_flag("nodlgt", Opt_ignore_silent),
++ fsparam_flag("clean_plink", Opt_ignore),
++ fsparam_string("dirs", Opt_br),
++ fsparam_u32("debug", Opt_ignore),
++ /* "whiteout" or "all" */
++ fsparam_string("delete", Opt_ignore),
++ fsparam_string("imap", Opt_ignore),
+
-+static inline struct au_vdir *au_ivdir(struct inode *inode)
-+{
-+ IiMustAnyLock(inode);
-+ AuDebugOn(!S_ISDIR(inode->i_mode));
-+ return au_ii(inode)->ii_vdir;
-+}
++ /* temporary workaround, due to old mount(8)? */
++ fsparam_flag("relatime", Opt_ignore_silent),
+
-+static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ struct au_hinode *hinode;
-+ IiMustAnyLock(inode);
-+ hinode = au_ii(inode)->ii_hinode + bindex;
-+ return hinode->hi_whdentry;
-+}
++ {}
++};
+
-+static inline void au_set_ibend(struct inode *inode, aufs_bindex_t bindex)
++static int au_fsctx_parse_do_add(struct fs_context *fc, struct au_opt *opt,
++ char *brspec, size_t speclen,
++ aufs_bindex_t bindex)
+{
-+ IiMustWriteLock(inode);
-+ AuDebugOn(au_sbend(inode->i_sb) < bindex || bindex < au_ibstart(inode));
-+ au_ii(inode)->ii_bend = bindex;
-+}
++ int err;
++ char *p;
+
-+static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir)
-+{
-+ IiMustWriteLock(inode);
-+ AuDebugOn(!S_ISDIR(inode->i_mode) || (au_ii(inode)->ii_vdir && vdir));
-+ au_ii(inode)->ii_vdir = vdir;
-+}
++ AuDbg("brspec %s\n", brspec);
+
-+static inline void au_hiput(struct au_hinode *hinode)
-+{
-+ au_hin_free(hinode);
-+ dput(hinode->hi_whdentry);
-+ iput(hinode->hi_inode);
-+}
++ err = -ENOMEM;
++ if (!speclen)
++ speclen = strlen(brspec);
++ /* will be freed by au_fsctx_free() */
++ p = kmemdup_nul(brspec, speclen, GFP_NOFS);
++ if (unlikely(!p)) {
++ errorfc(fc, "failed in %s", brspec);
++ goto out;
++ }
++ err = au_opt_add(opt, p, fc->sb_flags, bindex);
+
-+static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex)
-+{
-+ /* todo: this lock check causes some unnecessary locks in callers. */
-+ IiMustAnyLock(inode);
-+ return au_ii(inode)->ii_hinode + bindex;
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+/* tiny test for inode number */
-+/* tmpfs generation is too rough */
-+static inline int au_test_higen(struct inode *inode, struct inode *h_inode)
++static int au_fsctx_parse_br(struct fs_context *fc, char *brspec)
+{
-+ struct au_iinfo *iinfo;
++ int err;
++ char *p;
++ struct au_fsctx_opts *a = fc->fs_private;
++ struct au_opt *opt = a->opt;
++ aufs_bindex_t bindex = a->bindex;
+
-+ IiMustAnyLock(inode);
++ AuDbg("brspec %s\n", brspec);
+
-+ iinfo = au_ii(inode);
-+ return !(iinfo->ii_hsb1 == h_inode->i_sb
-+ && iinfo->ii_higen == h_inode->i_generation);
-+}
++ err = -EINVAL;
++ while ((p = strsep(&brspec, ":")) && *p) {
++ err = au_fsctx_parse_do_add(fc, opt, p, /*len*/0, bindex);
++ AuTraceErr(err);
++ if (unlikely(err))
++ break;
++ bindex++;
++ opt++;
++ if (unlikely(opt > a->opt_tail)) {
++ err = -E2BIG;
++ bindex--;
++ opt--;
++ break;
++ }
++ opt->type = Opt_tail;
++ a->skipped = 1;
++ }
++ a->bindex = bindex;
++ a->opt = opt;
+
-+static inline au_gen_t au_iigen(struct inode *inode)
-+{
-+ return atomic_read(&au_ii(inode)->ii_generation);
++ AuTraceErr(err);
++ return err;
+}
+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+static inline au_gen_t au_iigen_dec(struct inode *inode)
++static int au_fsctx_parse_add(struct fs_context *fc, char *addspec)
+{
-+ /* AuDbg("i%lu\n", inode->i_ino); */
-+ return atomic_dec_return(&au_ii(inode)->ii_generation);
-+}
-+#endif
++ int err, n;
++ char *p;
++ struct au_fsctx_opts *a = fc->fs_private;
++ struct au_opt *opt = a->opt;
+
-+/* ---------------------------------------------------------------------- */
++ err = -EINVAL;
++ p = strchr(addspec, ':');
++ if (unlikely(!p)) {
++ errorfc(fc, "bad arg in %s", addspec);
++ goto out;
++ }
++ *p++ = '\0';
++ err = kstrtoint(addspec, 0, &n);
++ if (unlikely(err)) {
++ errorfc(fc, "bad integer in %s", addspec);
++ goto out;
++ }
++ AuDbg("n %d\n", n);
++ err = au_fsctx_parse_do_add(fc, opt, p, /*len*/0, n);
+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+static inline struct au_pin1 *au_pin_gp(struct au_pin *args)
-+{
-+ return args->pin + AuPin_GPARENT;
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+/* hinotify.c */
-+void au_unpin_gp(struct au_pin *args);
++static int au_fsctx_parse_del(struct fs_context *fc, struct au_opt_del *del,
++ struct fs_parameter *param)
++{
++ int err;
+
-+#else
++ err = -ENOMEM;
++ /* will be freed by au_fsctx_free() */
++ del->pathname = kmemdup_nul(param->string, param->size, GFP_NOFS);
++ if (unlikely(!del->pathname))
++ goto out;
++ AuDbg("del %s\n", del->pathname);
++ err = vfsub_kern_path(del->pathname, AuOpt_LkupDirFlags, &del->h_path);
++ if (unlikely(err))
++ errorfc(fc, "lookup failed %s (%d)", del->pathname, err);
+
-+static inline struct au_pin1 *au_pin_gp(struct au_pin *args)
-+{
-+ return NULL;
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline void au_unpin_gp(struct au_pin *args)
++#if 0 /* reserved for future use */
++static int au_fsctx_parse_idel(struct fs_context *fc, struct au_opt_del *del,
++ aufs_bindex_t bindex)
+{
-+ /* empty */
-+}
-+#endif /* HINOTIFY */
++ int err;
++ struct super_block *sb;
++ struct dentry *root;
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+static inline void au_unpin(struct au_pin *args)
-+{
-+ au_do_unpin(args->pin + AuPin_PARENT, au_pin_gp(args));
-+}
++ sb = a->sb;
++ AuDebugOn(!sb);
+
-+static inline
-+struct au_hinode *au_do_pinned_hdir(struct au_pin1 *pin)
-+{
-+ if (pin && pin->parent)
-+ return au_hi(pin->parent->d_inode, pin->bindex);
-+ return NULL;
-+}
++ err = -EINVAL;
++ root = sb->s_root;
++ aufs_read_lock(root, AuLock_FLUSH);
++ if (bindex < 0 || au_sbbot(sb) < bindex) {
++ errorfc(fc, "out of bounds, %d", bindex);
++ goto out;
++ }
+
-+struct dentry *au_do_pinned_h_parent(struct au_pin1 *pin);
++ err = 0;
++ del->h_path.dentry = dget(au_h_dptr(root, bindex));
++ del->h_path.mnt = mntget(au_sbr_mnt(sb, bindex));
+
-+static inline struct dentry *au_do_pinned_parent(struct au_pin1 *pin)
-+{
-+ if (pin)
-+ return pin->parent;
-+ return NULL;
++out:
++ aufs_read_unlock(root, !AuLock_IR);
++ AuTraceErr(err);
++ return err;
+}
++#endif
+
-+static inline struct inode *au_do_pinned_h_dir(struct au_pin1 *pin)
++static int au_fsctx_parse_mod(struct fs_context *fc, struct au_opt_mod *mod,
++ struct fs_parameter *param)
+{
-+ if (pin)
-+ return pin->h_dir;
-+ return NULL;
-+}
++ int err;
++ struct path path;
++ char *p;
+
-+static inline
-+void au_pin_do_set_dentry(struct au_pin1 *pin, struct dentry *dentry)
-+{
-+ if (pin)
-+ pin->dentry = dentry;
-+}
++ err = -ENOMEM;
++ /* will be freed by au_fsctx_free() */
++ mod->path = kmemdup_nul(param->string, param->size, GFP_NOFS);
++ if (unlikely(!mod->path))
++ goto out;
+
-+static inline
-+void au_pin_do_set_parent(struct au_pin1 *pin, struct dentry *parent)
-+{
-+ if (pin) {
-+ dput(pin->parent);
-+ pin->parent = dget(parent);
++ err = -EINVAL;
++ p = strchr(mod->path, '=');
++ if (unlikely(!p)) {
++ errorfc(fc, "no permission %s", mod->path);
++ goto out;
+ }
-+}
+
-+static inline void au_pin_do_set_h_dir(struct au_pin1 *pin, struct inode *h_dir)
-+{
-+ if (pin) {
-+ iput(pin->h_dir);
-+ pin->h_dir = au_igrab(h_dir);
++ *p++ = 0;
++ err = vfsub_kern_path(mod->path, AuOpt_LkupDirFlags, &path);
++ if (unlikely(err)) {
++ errorfc(fc, "lookup failed %s (%d)", mod->path, err);
++ goto out;
+ }
-+}
+
-+static inline
-+void au_pin_do_set_parent_lflag(struct au_pin1 *pin, unsigned char lflag)
-+{
-+ if (pin) {
-+ /* dirty macros require brackets */
-+ if (lflag) {
-+ au_fset_pin(pin->flags, DI_LOCKED);
-+ } else {
-+ au_fclr_pin(pin->flags, DI_LOCKED);
-+ }
-+ }
-+}
++ mod->perm = au_br_perm_val(p);
++ AuDbg("mod path %s, perm 0x%x, %s", mod->path, mod->perm, p);
++ mod->h_root = dget(path.dentry);
++ path_put(&path);
+
-+static inline
-+struct au_hinode *au_pinned_hdir(struct au_pin *args)
-+{
-+ return au_do_pinned_hdir(args->pin + AuPin_PARENT);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline
-+struct au_hinode *au_pinned_hgdir(struct au_pin *args)
++#if 0 /* reserved for future use */
++static int au_fsctx_parse_imod(struct fs_context *fc, struct au_opt_mod *mod,
++ char *ibrspec)
+{
-+ return au_do_pinned_hdir(au_pin_gp(args));
-+}
++ int err, n;
++ char *p;
++ struct super_block *sb;
++ struct dentry *root;
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+static inline
-+struct dentry *au_pinned_h_parent(struct au_pin *args)
-+{
-+ return au_do_pinned_h_parent(args->pin + AuPin_PARENT);
-+}
++ sb = a->sb;
++ AuDebugOn(!sb);
+
-+#if 0 /* reserved for future use */
-+static inline
-+struct dentry *au_pinned_h_gparent(struct au_pin *args)
-+{
-+ return au_do_pinned_h_parent(au_pin_gp(args));
++ err = -EINVAL;
++ p = strchr(ibrspec, ':');
++ if (unlikely(!p)) {
++ errorfc(fc, "no index, %s", ibrspec);
++ goto out;
++ }
++ *p++ = '\0';
++ err = kstrtoint(ibrspec, 0, &n);
++ if (unlikely(err)) {
++ errorfc(fc, "bad integer in %s", ibrspec);
++ goto out;
++ }
++ AuDbg("n %d\n", n);
++
++ root = sb->s_root;
++ aufs_read_lock(root, AuLock_FLUSH);
++ if (n < 0 || au_sbbot(sb) < n) {
++ errorfc(fc, "out of bounds, %d", bindex);
++ goto out_root;
++ }
++
++ err = 0;
++ mod->perm = au_br_perm_val(p);
++ AuDbg("mod path %s, perm 0x%x, %s\n",
++ mod->path, mod->perm, p);
++ mod->h_root = dget(au_h_dptr(root, bindex));
++
++out_root:
++ aufs_read_unlock(root, !AuLock_IR);
++out:
++ AuTraceErr(err);
++ return err;
+}
+#endif
+
-+static inline
-+struct dentry *au_pinned_parent(struct au_pin *args)
++static int au_fsctx_parse_xino(struct fs_context *fc,
++ struct au_opt_xino *xino,
++ struct fs_parameter *param)
+{
-+ return au_do_pinned_parent(args->pin + AuPin_PARENT);
-+}
++ int err;
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+static inline
-+struct dentry *au_pinned_gparent(struct au_pin *args)
-+{
-+ return au_do_pinned_parent(au_pin_gp(args));
-+}
++ err = -ENOMEM;
++ /* will be freed by au_opts_free() */
++ xino->path = kmemdup_nul(param->string, param->size, GFP_NOFS);
++ if (unlikely(!xino->path))
++ goto out;
++ AuDbg("path %s\n", xino->path);
+
-+static inline
-+struct inode *au_pinned_h_dir(struct au_pin *args)
-+{
-+ return au_do_pinned_h_dir(args->pin + AuPin_PARENT);
-+}
++ xino->file = au_xino_create(a->sb, xino->path, /*silent*/0,
++ /*wbrtop*/0);
++ err = PTR_ERR(xino->file);
++ if (IS_ERR(xino->file)) {
++ xino->file = NULL;
++ goto out;
++ }
+
-+static inline
-+struct inode *au_pinned_h_gdir(struct au_pin *args)
-+{
-+ return au_do_pinned_h_dir(au_pin_gp(args));
-+}
++ err = 0;
++ if (unlikely(a->sb && xino->file->f_path.dentry->d_sb == a->sb)) {
++ err = -EINVAL;
++ errorfc(fc, "%s must be outside", xino->path);
++ }
+
-+static inline void au_pin_set_parent(struct au_pin *args, struct dentry *d)
-+{
-+ au_pin_do_set_parent(args->pin + AuPin_PARENT, d);
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline void au_pin_set_gparent(struct au_pin *args, struct dentry *d)
++static
++int au_fsctx_parse_xino_itrunc_path(struct fs_context *fc,
++ struct au_opt_xino_itrunc *xino_itrunc,
++ char *pathname)
+{
-+ au_pin_do_set_parent(au_pin_gp(args), d);
-+}
++ int err;
++ aufs_bindex_t bbot, bindex;
++ struct path path;
++ struct dentry *root;
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+static inline void au_pin_set_h_dir(struct au_pin *args, struct inode *h_dir)
-+{
-+ au_pin_do_set_h_dir(args->pin + AuPin_PARENT, h_dir);
-+}
++ AuDebugOn(!a->sb);
+
-+static inline void au_pin_set_h_gdir(struct au_pin *args, struct inode *h_dir)
-+{
-+ au_pin_do_set_h_dir(au_pin_gp(args), h_dir);
++ err = vfsub_kern_path(pathname, AuOpt_LkupDirFlags, &path);
++ if (unlikely(err)) {
++ errorfc(fc, "lookup failed %s (%d)", pathname, err);
++ goto out;
++ }
++
++ xino_itrunc->bindex = -1;
++ root = a->sb->s_root;
++ aufs_read_lock(root, AuLock_FLUSH);
++ bbot = au_sbbot(a->sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ if (au_h_dptr(root, bindex) == path.dentry) {
++ xino_itrunc->bindex = bindex;
++ break;
++ }
++ }
++ aufs_read_unlock(root, !AuLock_IR);
++ path_put(&path);
++
++ if (unlikely(xino_itrunc->bindex < 0)) {
++ err = -EINVAL;
++ errorfc(fc, "no such branch %s", pathname);
++ }
++
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline
-+void au_pin_set_parent_lflag(struct au_pin *args, unsigned char lflag)
++static int au_fsctx_parse_xino_itrunc(struct fs_context *fc,
++ struct au_opt_xino_itrunc *xino_itrunc,
++ unsigned int bindex)
+{
-+ au_pin_do_set_parent_lflag(args->pin + AuPin_PARENT, lflag);
++ int err;
++ aufs_bindex_t bbot;
++ struct super_block *sb;
++ struct au_fsctx_opts *a = fc->fs_private;
++
++ sb = a->sb;
++ AuDebugOn(!sb);
++
++ err = 0;
++ si_noflush_read_lock(sb);
++ bbot = au_sbbot(sb);
++ si_read_unlock(sb);
++ if (bindex <= bbot)
++ xino_itrunc->bindex = bindex;
++ else {
++ err = -EINVAL;
++ errorfc(fc, "out of bounds, %u", bindex);
++ }
++
++ AuTraceErr(err);
++ return err;
+}
+
-+static inline
-+void au_pin_set_gparent_lflag(struct au_pin *args, unsigned char lflag)
++static int au_fsctx_parse_param(struct fs_context *fc, struct fs_parameter *param)
+{
-+ au_pin_do_set_parent_lflag(au_pin_gp(args), lflag);
-+}
++ int err, token;
++ struct fs_parse_result result;
++ struct au_fsctx_opts *a = fc->fs_private;
++ struct au_opt *opt = a->opt;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_INODE_H__ */
-diff -urN linux/fs/aufs/i_op_add.c linux-aufs/fs/aufs/i_op_add.c
---- linux/fs/aufs/i_op_add.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/i_op_add.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,757 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ AuDbg("fc %p, param {key %s, string %s}\n",
++ fc, param->key, param->string);
++ err = fs_parse(fc, aufs_fsctx_paramspec, param, &result);
++ if (unlikely(err < 0))
++ goto out;
++ token = err;
++ AuDbg("token %d, res{negated %d, uint64 %llu}\n",
++ token, result.negated, result.uint_64);
+
-+/*
-+ * inode operations (add entry)
-+ *
-+ * $Id$
-+ */
++ err = -EINVAL;
++ a->skipped = 0;
++ switch (token) {
++ case Opt_br:
++ err = au_fsctx_parse_br(fc, param->string);
++ break;
++ case Opt_add:
++ err = au_fsctx_parse_add(fc, param->string);
++ break;
++ case Opt_append:
++ err = au_fsctx_parse_do_add(fc, opt, param->string, param->size,
++ /*dummy bindex*/1);
++ break;
++ case Opt_prepend:
++ err = au_fsctx_parse_do_add(fc, opt, param->string, param->size,
++ /*bindex*/0);
++ break;
+
-+#include "aufs.h"
++ case Opt_del:
++ err = au_fsctx_parse_del(fc, &opt->del, param);
++ break;
++#if 0 /* reserved for future use */
++ case Opt_idel:
++ if (!a->sb) {
++ err = 0;
++ a->skipped = 1;
++ break;
++ }
++ del->pathname = "(indexed)";
++ err = au_opts_parse_idel(fc, &opt->del, result.uint_32);
++ break;
++#endif
+
-+/*
-+ * final procedure of adding a new entry, except link(2).
-+ * remove whiteout, instantiate, copyup the parent dir's times and size
-+ * and update version.
-+ * if it failed, re-create the removed whiteout.
-+ */
-+static int epilog(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *wh_dentry, struct dentry *dentry)
-+{
-+ int err, rerr;
-+ aufs_bindex_t bwh;
-+ struct inode *inode, *h_dir;
-+ struct dentry *wh;
-+ struct au_ndx ndx;
-+ struct super_block *sb;
++ case Opt_mod:
++ err = au_fsctx_parse_mod(fc, &opt->mod, param);
++ break;
++#ifdef IMOD /* reserved for future use */
++ case Opt_imod:
++ if (!a->sb) {
++ err = 0;
++ a->skipped = 1;
++ break;
++ }
++ u.mod->path = "(indexed)";
++ err = au_opts_parse_imod(fc, &opt->mod, param->string);
++ break;
++#endif
+
-+ LKTRTrace("wh %p, %.*s\n", wh_dentry, AuDLNPair(dentry));
++ case Opt_xino:
++ err = au_fsctx_parse_xino(fc, &opt->xino, param);
++ break;
++ case Opt_trunc_xino_path:
++ if (!a->sb) {
++ errorfc(fc, "no such branch %s", param->string);
++ break;
++ }
++ err = au_fsctx_parse_xino_itrunc_path(fc, &opt->xino_itrunc,
++ param->string);
++ break;
++#if 0
++ case Opt_trunc_xino_v:
++ if (!a->sb) {
++ err = 0;
++ a->skipped = 1;
++ break;
++ }
++ err = au_fsctx_parse_xino_itrunc_path(fc, &opt->xino_itrunc,
++ param->string);
++ break;
++#endif
++ case Opt_itrunc_xino:
++ if (!a->sb) {
++ errorfc(fc, "out of bounds %s", param->string);
++ break;
++ }
++ err = au_fsctx_parse_xino_itrunc(fc, &opt->xino_itrunc,
++ result.int_32);
++ break;
+
-+ sb = dentry->d_sb;
-+ bwh = -1;
-+ if (wh_dentry) {
-+ h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */
-+ IMustLock(h_dir);
-+ AuDebugOn(au_h_iptr(dir, bindex) != h_dir);
-+ bwh = au_dbwh(dentry);
-+ err = au_wh_unlink_dentry(au_hi(dir, bindex), wh_dentry, dentry,
-+ /*dlgt*/0);
-+ if (unlikely(err))
-+ goto out;
-+ }
++ case Opt_dirwh:
++ err = 0;
++ opt->dirwh = result.int_32;
++ break;
+
-+ inode = au_new_inode(dentry, /*must_new*/1);
-+ if (!IS_ERR(inode)) {
-+ d_instantiate(dentry, inode);
-+ dir = dentry->d_parent->d_inode; /* dir inode is locked */
-+ IMustLock(dir);
-+ /* or always cpup dir mtime? */
-+ if (au_ibstart(dir) == au_dbstart(dentry))
-+ au_cpup_attr_timesizes(dir);
-+ dir->i_version++;
-+ return 0; /* success */
-+ }
++ case Opt_rdcache:
++ if (unlikely(result.int_32 > AUFS_RDCACHE_MAX)) {
++ errorfc(fc, "rdcache must be smaller than %d",
++ AUFS_RDCACHE_MAX);
++ break;
++ }
++ err = 0;
++ opt->rdcache = result.int_32;
++ break;
+
-+ err = PTR_ERR(inode);
-+ if (!wh_dentry)
-+ goto out;
++ case Opt_rdblk:
++ err = 0;
++ opt->rdblk = AUFS_RDBLK_DEF;
++ if (!strcmp(param->string, "def"))
++ break;
+
-+ /* revert */
-+ ndx.flags = 0;
-+ if (au_test_dlgt(au_mntflags(sb)))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nfsmnt = au_nfsmnt(sb, bwh);
-+ ndx.nd = NULL;
-+ /* ndx.br = NULL; */
-+ /* dir inode is locked */
-+ wh = au_wh_create(dentry, bwh, wh_dentry->d_parent, &ndx);
-+ rerr = PTR_ERR(wh);
-+ if (IS_ERR(wh)) {
-+ AuIOErr("%.*s reverting whiteout failed(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ err = -EIO;
-+ } else
-+ dput(wh);
++ err = kstrtoint(param->string, 0, &result.int_32);
++ if (unlikely(err)) {
++ errorfc(fc, "bad value in %s", param->key);
++ break;
++ }
++ err = -EINVAL;
++ if (unlikely(result.int_32 < 0
++ || result.int_32 > KMALLOC_MAX_SIZE)) {
++ errorfc(fc, "bad value in %s", param->key);
++ break;
++ }
++ if (unlikely(result.int_32 && result.int_32 < NAME_MAX)) {
++ errorfc(fc, "rdblk must be larger than %d", NAME_MAX);
++ break;
++ }
++ err = 0;
++ opt->rdblk = result.int_32;
++ break;
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++ case Opt_rdhash:
++ err = 0;
++ opt->rdhash = AUFS_RDHASH_DEF;
++ if (!strcmp(param->string, "def"))
++ break;
+
-+/*
-+ * simple tests for the adding inode operations.
-+ * following the checks in vfs, plus the parent-child relationship.
-+ */
-+int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir, struct au_ndx *ndx)
-+{
-+ int err, exist;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ umode_t h_mode;
++ err = kstrtoint(param->string, 0, &result.int_32);
++ if (unlikely(err)) {
++ errorfc(fc, "bad value in %s", param->key);
++ break;
++ }
++ /* how about zero? */
++ if (result.int_32 < 0
++ || result.int_32 * sizeof(struct hlist_head)
++ > KMALLOC_MAX_SIZE) {
++ err = -EINVAL;
++ errorfc(fc, "bad integer in %s", param->key);
++ break;
++ }
++ opt->rdhash = result.int_32;
++ break;
+
-+ LKTRTrace("%.*s/%.*s, b%d, dir %d\n",
-+ AuDLNPair(h_parent), AuDLNPair(dentry), bindex, isdir);
++ case Opt_diropq:
++ /*
++ * As other options, fs/aufs/opts.c can handle these strings by
++ * match_token(). But "diropq=" is deprecated now and will
++ * never have other value. So simple strcmp() is enough here.
++ */
++ if (!strcmp(param->string, "a") ||
++ !strcmp(param->string, "always")) {
++ err = 0;
++ opt->type = Opt_diropq_a;
++ } else if (!strcmp(param->string, "w") ||
++ !strcmp(param->string, "whiteouted")) {
++ err = 0;
++ opt->type = Opt_diropq_w;
++ } else
++ errorfc(fc, "unknown value %s", param->string);
++ break;
+
-+ exist = !!dentry->d_inode;
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ h_inode = h_dentry->d_inode;
-+ if (!exist) {
-+ err = -EEXIST;
-+ if (unlikely(h_inode))
-+ goto out;
-+ } else {
-+ /* rename(2) case */
-+ err = -EIO;
-+ if (unlikely(!h_inode || !h_inode->i_nlink))
-+ goto out;
++ case Opt_udba:
++ opt->udba = au_udba_val(param->string);
++ if (opt->udba >= 0)
++ err = 0;
++ else
++ errorf(fc, "wrong value, %s", param->string);
++ break;
+
-+ h_mode = h_inode->i_mode;
-+ if (!isdir) {
-+ err = -EISDIR;
-+ if (unlikely(S_ISDIR(h_mode)))
-+ goto out;
-+ } else if (unlikely(!S_ISDIR(h_mode))) {
-+ err = -ENOTDIR;
-+ goto out;
++ case Opt_wbr_create:
++ opt->wbr_create.wbr_create
++ = au_wbr_create_val(param->string, &opt->wbr_create);
++ if (opt->wbr_create.wbr_create >= 0)
++ err = 0;
++ else
++ errorf(fc, "wrong value, %s", param->key);
++ break;
++
++ case Opt_wbr_copyup:
++ opt->wbr_copyup = au_wbr_copyup_val(param->string);
++ if (opt->wbr_copyup >= 0)
++ err = 0;
++ else
++ errorfc(fc, "wrong value, %s", param->key);
++ break;
++
++ case Opt_fhsm_sec:
++ if (unlikely(result.int_32 < 0)) {
++ errorfc(fc, "bad integer in %s\n", param->key);
++ break;
+ }
-+ }
++ err = 0;
++ if (sysaufs_brs)
++ opt->fhsm_second = result.int_32;
++ else
++ warnfc(fc, "ignored %s %s", param->key, param->string);
++ break;
+
-+ err = -EIO;
-+ /* expected parent dir is locked */
-+ if (unlikely(h_parent != h_dentry->d_parent))
-+ goto out;
-+ err = 0;
++ /* simple true/false flag */
++#define au_fsctx_TF(name) \
++ case Opt_##name: \
++ err = 0; \
++ opt->tf = !result.negated; \
++ break
++ au_fsctx_TF(trunc_xino);
++ au_fsctx_TF(trunc_xib);
++ au_fsctx_TF(dirperm1);
++ au_fsctx_TF(plink);
++ au_fsctx_TF(shwh);
++ au_fsctx_TF(dio);
++ au_fsctx_TF(warn_perm);
++ au_fsctx_TF(verbose);
++ au_fsctx_TF(sum);
++ au_fsctx_TF(dirren);
++ au_fsctx_TF(acl);
++#undef au_fsctx_TF
+
-+ if (au_opt_test(au_mntflags(dentry->d_sb), UDBA_INOTIFY)) {
-+ struct dentry *h_latest;
-+ struct qstr *qstr = &dentry->d_name;
++ case Opt_noverbose:
++ err = 0;
++ opt->type = Opt_verbose;
++ opt->tf = false;
++ break;
+
-+ err = -EACCES;
-+ if (unlikely(au_test_h_perm
-+ (h_parent->d_inode, MAY_EXEC | MAY_WRITE,
-+ au_ftest_ndx(ndx->flags, DLGT))))
-+ goto out;
++ case Opt_noxino:
++ fallthrough;
++ case Opt_list_plink:
++ fallthrough;
++ case Opt_wsum:
++ err = 0;
++ break;
+
-+ h_latest = au_sio_lkup_one(qstr->name, h_parent, qstr->len,
-+ ndx);
-+ err = PTR_ERR(h_latest);
-+ if (IS_ERR(h_latest))
-+ goto out;
-+ err = -EIO;
-+ dput(h_latest);
-+ /* fuse d_revalidate always return 0 for negative dentries */
-+ if (h_latest == h_dentry || au_test_fuse(h_dentry->d_sb))
-+ err = 0;
++ case Opt_ignore:
++ warnfc(fc, "ignored %s", param->key);
++ fallthrough;
++ case Opt_ignore_silent:
++ a->skipped = 1;
++ err = 0;
++ break;
++ default:
++ a->skipped = 1;
++ err = -ENOPARAM;
++ break;
+ }
++ if (unlikely(err))
++ goto out;
++ if (a->skipped)
++ goto out;
+
-+ out:
-+ AuTraceErr(err);
++ switch (token) {
++ case Opt_br:
++ fallthrough;
++ case Opt_noverbose:
++ fallthrough;
++ case Opt_diropq:
++ break;
++ default:
++ opt->type = token;
++ break;
++ }
++ opt++;
++ if (unlikely(opt > a->opt_tail)) {
++ err = -E2BIG;
++ opt--;
++ }
++ opt->type = Opt_tail;
++ a->opt = opt;
++
++out:
+ return err;
+}
+
+/*
-+ * initial procedure of adding a new entry.
-+ * prepare writable branch and the parent dir, lock it,
-+ * lookup whiteout for the new entry.
++ * these options accept both 'name=val' and 'name:val' form.
++ * some accept optional '=' in its value.
++ * eg. br:/br1=rw:/br2=ro and br=/br1=rw:/br2=ro
+ */
-+static struct dentry*
-+lock_hdir_lkup_wh(struct dentry *dentry, struct au_dtime *dt,
-+ struct dentry *src_dentry, struct au_pin *pin,
-+ struct au_wr_dir_args *wr_dir_args)
++static inline unsigned int is_colonopt(char *str)
++{
++#define do_test(name) \
++ if (!strncmp(str, name ":", sizeof(name))) \
++ return sizeof(name) - 1
++ do_test("br");
++ do_test("add");
++ do_test("ins");
++ do_test("append");
++ do_test("prepend");
++ do_test("del");
++ /* do_test("idel"); */
++ do_test("mod");
++ /* do_test("imod"); */
++#undef do_test
++
++ return 0;
++}
++
++static int au_fsctx_parse_monolithic(struct fs_context *fc, void *data)
+{
-+ struct dentry *wh_dentry, *h_parent;
-+ struct super_block *sb;
+ int err;
-+ unsigned int mnt_flags;
-+ unsigned char pin_flags;
-+ aufs_bindex_t bstart, bcpup;
-+ struct au_ndx ndx;
-+
-+ LKTRTrace("%.*s, src %p\n", AuDLNPair(dentry), src_dentry);
++ unsigned int u;
++ char *str;
++ struct au_fsctx_opts *a = fc->fs_private;
++
++ str = data;
++ AuDbg("str %s\n", str);
++ while (str) {
++ u = is_colonopt(str);
++ if (u)
++ str[u] = '=';
++ str = strchr(str, ',');
++ if (!str)
++ break;
++ str++;
++ }
++ str = data;
++ AuDbg("str %s\n", str);
+
-+ bstart = au_dbstart(dentry);
-+ err = au_wr_dir(dentry, src_dentry, wr_dir_args);
-+ bcpup = err;
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err < 0))
-+ goto out;
++ err = generic_parse_monolithic(fc, str);
++ AuTraceErr(err);
++ au_fsctx_dump(&a->opts);
+
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (dt && au_opt_test(mnt_flags, UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ err = au_pin(pin, dentry, bcpup, pin_flags);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out;
++ return err;
++}
+
-+ ndx.nfsmnt = au_nfsmnt(sb, bcpup);
-+ ndx.flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nd = NULL;
-+ /* ndx.br = NULL; */
-+ /* ndx.nd_file = NULL; */
++/* ---------------------------------------------------------------------- */
+
-+ h_parent = au_pinned_h_parent(pin);
-+ if (!au_opt_test(mnt_flags, UDBA_NONE) && au_dbstart(dentry) == bcpup) {
-+ struct nameidata nd;
++static void au_fsctx_opts_free(struct au_opts *opts)
++{
++ struct au_opt *opt;
+
-+ if (ndx.nfsmnt) {
-+ /* todo: dirty? */
-+ ndx.nd = &nd;
-+ ndx.br = au_sbr(sb, bcpup);
-+ memset(&nd, 0, sizeof(nd));
-+ nd.flags = LOOKUP_CREATE;
-+ nd.intent.open.flags = O_EXCL;
-+ }
-+ err = au_may_add(dentry, bcpup, h_parent,
-+ au_ftest_wrdir(wr_dir_args->flags, ISDIR),
-+ &ndx);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_unpin;
-+ ndx.nd = NULL;
-+ ndx.br = NULL;
++ opt = opts->opt;
++ while (opt->type != Opt_tail) {
++ switch (opt->type) {
++ case Opt_add:
++ fallthrough;
++ case Opt_append:
++ fallthrough;
++ case Opt_prepend:
++ kfree(opt->add.pathname);
++ path_put(&opt->add.path);
++ break;
++ case Opt_del:
++ kfree(opt->del.pathname);
++ fallthrough;
++ case Opt_idel:
++ path_put(&opt->del.h_path);
++ break;
++ case Opt_mod:
++ kfree(opt->mod.path);
++ fallthrough;
++ case Opt_imod:
++ dput(opt->mod.h_root);
++ break;
++ case Opt_xino:
++ kfree(opt->xino.path);
++ fput(opt->xino.file);
++ break;
++ }
++ opt++;
+ }
++}
+
-+ if (dt)
-+ au_dtime_store(dt, au_pinned_parent(pin), h_parent,
-+ au_pinned_hdir(pin), au_pinned_hgdir(pin));
-+
-+ wh_dentry = NULL;
-+ if (/* bcpup != bstart || */ bcpup != au_dbwh(dentry))
-+ goto out; /* success */
++static void au_fsctx_free(struct fs_context *fc)
++{
++ struct au_fsctx_opts *a = fc->fs_private;
+
-+ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, &ndx);
++ /* fs_type=%p, root=%pD */
++ AuDbg("fc %p{sb_flags 0x%x, sb_flags_mask 0x%x, purpose %u\n",
++ fc, fc->sb_flags, fc->sb_flags_mask, fc->purpose);
+
-+ out_unpin:
-+ if (IS_ERR(wh_dentry))
-+ au_unpin(pin);
-+ out:
-+ AuTraceErrPtr(wh_dentry);
-+ return wh_dentry;
++ kobject_put(&a->sbinfo->si_kobj);
++ au_fsctx_opts_free(&a->opts);
++ free_page((unsigned long)a->opts.opt);
++ au_kfree_rcu(a);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+enum { Mknod, Symlink, Creat };
-+struct simple_arg {
-+ int type;
-+ union {
-+ struct {
-+ int mode;
-+ struct nameidata *nd;
-+ } c;
-+ struct {
-+ const char *symname;
-+ } s;
-+ struct {
-+ int mode;
-+ dev_t dev;
-+ } m;
-+ } u;
++static const struct fs_context_operations au_fsctx_ops = {
++ .free = au_fsctx_free,
++ .parse_param = au_fsctx_parse_param,
++ .parse_monolithic = au_fsctx_parse_monolithic,
++ .get_tree = au_fsctx_get_tree,
++ .reconfigure = au_fsctx_reconfigure
++ /*
++ * nfs4 requires ->dup()? No.
++ * I don't know what is this ->dup() for.
++ */
+};
+
-+static int add_simple(struct inode *dir, struct dentry *dentry,
-+ struct simple_arg *arg)
++int aufs_fsctx_init(struct fs_context *fc)
+{
+ int err;
-+ struct dentry *h_dentry, *wh_dentry, *parent;
-+ struct inode *h_dir;
-+ struct au_dtime dt;
-+ struct vfsub_args vargs;
-+ struct super_block *sb;
-+ aufs_bindex_t bstart;
-+ unsigned char created;
-+ struct au_hin_ignore ign;
-+ struct au_pin pin;
-+ struct au_wr_dir_args wr_dir_args = {
-+ .force_btgt = -1,
-+ .flags = AuWrDir_ADD_ENTRY
-+ };
++ struct au_fsctx_opts *a;
+
-+ LKTRTrace("type %d, %.*s\n", arg->type, AuDLNPair(dentry));
-+ IMustLock(dir);
++ /* fs_type=%p, root=%pD */
++ AuDbg("fc %p{sb_flags 0x%x, sb_flags_mask 0x%x, purpose %u\n",
++ fc, fc->sb_flags, fc->sb_flags_mask, fc->purpose);
+
-+ sb = dir->i_sb;
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ aufs_read_lock(dentry, AuLock_DW);
-+ vfsub_args_init(&vargs, &ign, !!au_test_dlgt(au_mntflags(sb)), 0);
-+ di_write_lock_parent(parent);
-+ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, /*src_dentry*/NULL, &pin,
-+ &wr_dir_args);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
++ /* they will be freed by au_fsctx_free() */
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
+ goto out;
-+
-+ bstart = au_dbstart(dentry);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ h_dir = au_pinned_h_dir(&pin);
-+ vfsub_ign_hinode(&vargs, IN_CREATE, au_pinned_hdir(&pin));
-+
-+ switch (arg->type) {
-+ case Creat:
-+ AuDebugOn(au_test_nfs(h_dir->i_sb) && !arg->u.c.nd);
-+ err = au_h_create(h_dir, h_dentry, arg->u.c.mode, &vargs,
-+ arg->u.c.nd, au_nfsmnt(sb, bstart));
-+ break;
-+ case Symlink:
-+ err = vfsub_symlink(h_dir, h_dentry, arg->u.s.symname,
-+ S_IALLUGO, &vargs);
-+ break;
-+ case Mknod:
-+ err = vfsub_mknod(h_dir, h_dentry, arg->u.m.mode, arg->u.m.dev,
-+ &vargs);
-+ break;
-+ default:
-+ BUG();
-+ }
-+ created = !err;
-+ if (!err)
-+ err = epilog(dir, bstart, wh_dentry, dentry);
-+
-+ /* revert */
-+ if (unlikely(created && err && h_dentry->d_inode)) {
-+ int rerr;
-+ vfsub_args_reinit(&vargs);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, au_pinned_hdir(&pin));
-+ rerr = vfsub_unlink(h_dir, h_dentry, &vargs);
-+ if (rerr) {
-+ AuIOErr("%.*s revert failure(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ err = -EIO;
-+ }
-+ /* todo: inotify will be fired to the grand parent dir? */
-+ au_dtime_revert(&dt);
-+ d_drop(dentry);
++ a->bindex = 0;
++ a->opts.opt = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!a->opts.opt))
++ goto out_a;
++ a->opt = a->opts.opt;
++ a->opt->type = Opt_tail;
++ a->opts.max_opt = PAGE_SIZE / sizeof(*a->opts.opt);
++ a->opt_tail = a->opt + a->opts.max_opt - 1;
++ a->opts.sb_flags = fc->sb_flags;
++
++ a->sb = NULL;
++ if (fc->root) {
++ AuDebugOn(fc->purpose != FS_CONTEXT_FOR_RECONFIGURE);
++ a->opts.flags = AuOpts_REMOUNT;
++ a->sb = fc->root->d_sb;
++ a->sbinfo = au_sbi(a->sb);
++ kobject_get(&a->sbinfo->si_kobj);
++ } else {
++ a->sbinfo = au_si_alloc(a->sb);
++ AuDebugOn(!a->sbinfo);
++ err = PTR_ERR(a->sbinfo);
++ if (IS_ERR(a->sbinfo))
++ goto out_opt;
++ au_rw_write_unlock(&a->sbinfo->si_rwsem);
+ }
+
-+ au_unpin(&pin);
-+ dput(wh_dentry);
++ err = 0;
++ fc->fs_private = a;
++ fc->ops = &au_fsctx_ops;
++ goto out; /* success */
+
-+ out:
-+ if (unlikely(err)) {
-+ au_update_dbstart(dentry);
-+ d_drop(dentry);
-+ }
-+ di_write_unlock(parent);
-+ aufs_read_unlock(dentry, AuLock_DW);
++out_opt:
++ free_page((unsigned long)a->opts.opt);
++out_a:
++ au_kfree_rcu(a);
++out:
+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
+ return err;
+}
+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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,401 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * judging filesystem type
++ */
++
++#ifndef __AUFS_FSTYPE_H__
++#define __AUFS_FSTYPE_H__
++
++#ifdef __KERNEL__
++
++#include <linux/fs.h>
++#include <linux/magic.h>
++#include <linux/nfs_fs.h>
++#include <linux/romfs_fs.h>
+
-+int aufs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev)
++static inline int au_test_aufs(struct super_block *sb)
+{
-+ struct simple_arg arg = {
-+ .type = Mknod,
-+ .u.m = {
-+ .mode = mode,
-+ .dev = dev
-+ }
-+ };
-+ return add_simple(dir, dentry, &arg);
++ return sb->s_magic == AUFS_SUPER_MAGIC;
+}
+
-+int aufs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
++static inline const char *au_sbtype(struct super_block *sb)
+{
-+ struct simple_arg arg = {
-+ .type = Symlink,
-+ .u.s.symname = symname
-+ };
-+ return add_simple(dir, dentry, &arg);
++ return sb->s_type->name;
+}
+
-+int aufs_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd)
++static inline int au_test_iso9660(struct super_block *sb __maybe_unused)
+{
-+ struct simple_arg arg = {
-+ .type = Creat,
-+ .u.c = {
-+ .mode = mode,
-+ .nd = nd
-+ }
-+ };
-+ return add_simple(dir, dentry, &arg);
++#if IS_ENABLED(CONFIG_ISO9660_FS)
++ return sb->s_magic == ISOFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
++static inline int au_test_romfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_ROMFS_FS)
++ return sb->s_magic == ROMFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+struct au_link_args {
-+ unsigned int mnt_flags;
-+ unsigned char pin_flags;
-+ aufs_bindex_t bdst, bsrc;
-+ struct vfsub_args vargs;
-+ struct au_pin pin;
-+ struct au_hin_ignore ign;
-+ struct dentry *h_dentry;
-+ struct dentry *src_parent, *parent;
-+};
++static inline int au_test_cramfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_CRAMFS)
++ return sb->s_magic == CRAMFS_MAGIC;
++#endif
++ return 0;
++}
+
-+static int au_cpup_before_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry, struct au_link_args *a)
++static inline int au_test_nfs(struct super_block *sb __maybe_unused)
+{
-+ int err;
-+ const int hinotify = au_opt_test(a->mnt_flags, UDBA_INOTIFY);
-+ struct mutex *h_mtx;
++#if IS_ENABLED(CONFIG_NFS_FS)
++ return sb->s_magic == NFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ LKTRTrace("src %.*s, i%lu, dst %.*s\n",
-+ AuDLNPair(src_dentry), dir->i_ino, AuDLNPair(dentry));
++static inline int au_test_fuse(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_FUSE_FS)
++ return sb->s_magic == FUSE_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ di_read_lock_parent(a->src_parent, AuLock_IR);
-+ err = au_test_and_cpup_dirs(src_dentry, a->bdst);
-+ if (unlikely(err))
-+ goto out;
++static inline int au_test_xfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_XFS_FS)
++ return sb->s_magic == XFS_SB_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ AuDebugOn(au_dbstart(src_dentry) != a->bsrc);
-+ h_mtx = &au_h_dptr(src_dentry, a->bsrc)->d_inode->i_mutex;
-+ a->pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (hinotify)
-+ au_fset_pin(a->pin_flags, DO_GPARENT);
-+ err = au_pin(&a->pin, src_dentry, a->bdst, a->pin_flags);
-+ if (unlikely(err))
-+ goto out;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ /* todo: no KEEPLINO because of noplink? */
-+ err = au_sio_cpup_simple(src_dentry, a->bdst, -1,
-+ AuCpup_DTIME /* | AuCpup_KEEPLINO */);
-+ mutex_unlock(h_mtx);
-+ au_unpin(&a->pin);
++static inline int au_test_tmpfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_TMPFS
++ return sb->s_magic == TMPFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ out:
-+ di_read_unlock(a->src_parent, AuLock_IR);
-+ AuTraceErr(err);
-+ return err;
++static inline int au_test_ecryptfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_ECRYPT_FS)
++ return !strcmp(au_sbtype(sb), "ecryptfs");
++#else
++ return 0;
++#endif
+}
+
-+static int au_cpup_or_link(struct dentry *src_dentry, struct au_link_args *a)
++static inline int au_test_ramfs(struct super_block *sb)
+{
-+ int err;
-+ struct inode *h_inode;
-+ aufs_bindex_t bstart;
-+ struct dentry *h_src_dentry;
++ return sb->s_magic == RAMFS_MAGIC;
++}
+
-+ AuTraceEnter();
-+ AuDebugOn(au_dbstart(src_dentry) != a->bsrc);
++static inline int au_test_ubifs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_UBIFS_FS)
++ return sb->s_magic == UBIFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ bstart = au_ibstart(src_dentry->d_inode);
-+ h_inode = NULL;
-+ if (bstart <= a->bdst)
-+ h_inode = au_h_iptr(src_dentry->d_inode, a->bdst);
-+ if (!h_inode || !h_inode->i_nlink) {
-+ /* copyup src_dentry as the name of dentry. */
-+ au_set_dbstart(src_dentry, a->bdst);
-+ au_set_h_dptr(src_dentry, a->bdst, dget(a->h_dentry));
-+ h_inode = au_h_dptr(src_dentry, a->bsrc)->d_inode;
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ err = au_sio_cpup_single(src_dentry, a->bdst, a->bsrc, -1,
-+ AuCpup_KEEPLINO, a->parent);
-+ mutex_unlock(&h_inode->i_mutex);
-+ au_set_h_dptr(src_dentry, a->bdst, NULL);
-+ au_set_dbstart(src_dentry, a->bsrc);
-+ } else {
-+ /* the inode of src_dentry already exists on a.bdst branch */
-+ h_src_dentry = d_find_alias(h_inode);
-+ if (h_src_dentry) {
-+ /* vfsub_args_reinit(&a->vargs); */
-+ vfsub_ign_hinode(&a->vargs, IN_CREATE,
-+ au_pinned_hdir(&a->pin));
-+ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin),
-+ a->h_dentry, &a->vargs);
-+ dput(h_src_dentry);
-+ } else {
-+ AuIOErr("no dentry found for i%lu on b%d\n",
-+ h_inode->i_ino, a->bdst);
-+ err = -EIO;
-+ }
-+ }
++static inline int au_test_procfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_PROC_FS
++ return sb->s_magic == PROC_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ if (!err)
-+ au_plink_append(src_dentry->d_sb, src_dentry->d_inode,
-+ a->h_dentry, a->bdst);
++static inline int au_test_sysfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_SYSFS
++ return sb->s_magic == SYSFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ AuTraceErr(err);
-+ return err;
++static inline int au_test_configfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_CONFIGFS_FS)
++ return sb->s_magic == CONFIGFS_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+int aufs_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry)
++static inline int au_test_minix(struct super_block *sb __maybe_unused)
+{
-+ int err, rerr;
-+ struct au_dtime dt;
-+ struct super_block *sb;
-+ struct au_link_args *a;
-+ struct au_wr_dir_args wr_dir_args = {
-+ /* .force_btgt = -1, */
-+ .flags = AuWrDir_ADD_ENTRY
-+ };
-+ struct dentry *wh_dentry, *h_src_dentry;
-+ struct inode *inode;
++#if IS_ENABLED(CONFIG_MINIX_FS)
++ return sb->s_magic == MINIX3_SUPER_MAGIC
++ || sb->s_magic == MINIX2_SUPER_MAGIC
++ || sb->s_magic == MINIX2_SUPER_MAGIC2
++ || sb->s_magic == MINIX_SUPER_MAGIC
++ || sb->s_magic == MINIX_SUPER_MAGIC2;
++#else
++ return 0;
++#endif
++}
+
-+ LKTRTrace("src %.*s, i%lu, dst %.*s\n",
-+ AuDLNPair(src_dentry), dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
-+ inode = src_dentry->d_inode;
-+ IMustLock(inode);
-+ AuDebugOn(S_ISDIR(inode->i_mode));
++static inline int au_test_fat(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_FAT_FS)
++ return sb->s_magic == MSDOS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ err = -ENOMEM;
-+ a = kzalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++static inline int au_test_msdos(struct super_block *sb)
++{
++ return au_test_fat(sb);
++}
+
-+ sb = dentry->d_sb;
-+ a->parent = dentry->d_parent; /* dir inode is locked */
-+ aufs_read_and_write_lock2(dentry, src_dentry, /*AuLock_FLUSH*/0);
-+ a->src_parent = dget_parent(src_dentry);
-+ wr_dir_args.force_btgt = au_dbstart(src_dentry);
-+ a->mnt_flags = au_mntflags(sb);
-+ vfsub_args_init(&a->vargs, &a->ign, au_test_dlgt(a->mnt_flags), 0);
++static inline int au_test_vfat(struct super_block *sb)
++{
++ return au_test_fat(sb);
++}
+
-+ di_write_lock_parent(a->parent);
-+ wr_dir_args.force_btgt = au_wbr(dentry, wr_dir_args.force_btgt);
-+ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, src_dentry, &a->pin,
-+ &wr_dir_args);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_unlock;
-+ err = 0;
++static inline int au_test_securityfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_SECURITYFS
++ return sb->s_magic == SECURITYFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ a->bdst = au_dbstart(dentry);
-+ a->h_dentry = au_h_dptr(dentry, a->bdst);
++static inline int au_test_squashfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_SQUASHFS)
++ return sb->s_magic == SQUASHFS_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ /* todo: minor optimize,
-+ their sb may be same while their bindex differs? */
-+ a->bsrc = au_dbstart(src_dentry);
-+ if (au_opt_test(a->mnt_flags, PLINK)) {
-+ if (a->bdst < a->bsrc
-+ /* && h_src_dentry->d_sb != a->h_dentry->d_sb */)
-+ err = au_cpup_or_link(src_dentry, a);
-+ else {
-+ h_src_dentry = au_h_dptr(src_dentry, a->bdst);
-+ AuDebugOn(!h_src_dentry);
-+ AuDebugOn(!h_src_dentry->d_inode);
-+ vfsub_ign_hinode(&a->vargs, IN_CREATE,
-+ au_pinned_hdir(&a->pin));
-+ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin),
-+ a->h_dentry, &a->vargs);
-+ }
-+ } else {
-+ /*
-+ * copyup src_dentry to the branch we process,
-+ * and then link(2) to it.
-+ */
-+ if (a->bdst < a->bsrc
-+ /* && h_src_dentry->d_sb != a->h_dentry->d_sb */) {
-+ au_unpin(&a->pin);
-+ di_write_unlock(a->parent);
-+ err = au_cpup_before_link(src_dentry, dir, dentry, a);
-+ if (!err) {
-+ a->pin_flags
-+ = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (au_opt_test(a->mnt_flags, UDBA_INOTIFY))
-+ au_fset_pin(a->pin_flags, DO_GPARENT);
-+ di_write_lock_parent(a->parent);
-+ err = au_pin(&a->pin, dentry, a->bdst,
-+ a->pin_flags);
-+ if (unlikely(err))
-+ goto out_wh;
-+ }
-+ }
-+ if (!err) {
-+ /* vfsub_args_reinit(&a->vargs); */
-+ vfsub_ign_hinode(&a->vargs, IN_CREATE,
-+ au_pinned_hdir(&a->pin));
-+ h_src_dentry = au_h_dptr(src_dentry, a->bdst);
-+ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin),
-+ a->h_dentry, &a->vargs);
-+ }
-+ }
-+ if (unlikely(err))
-+ goto out_unpin;
++static inline int au_test_btrfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_BTRFS_FS)
++ return sb->s_magic == BTRFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ if (wh_dentry) {
-+ err = au_wh_unlink_dentry(au_pinned_hdir(&a->pin), wh_dentry,
-+ dentry, /*dlgt*/0);
-+ if (unlikely(err))
-+ goto out_revert;
-+ }
++static inline int au_test_xenfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_XENFS)
++ return sb->s_magic == XENFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+#if 0 /* cannot support it */
-+ /* fuse has different memory inode for the same inode number */
-+ if (au_test_fuse(a->h_dentry->d_sb)) {
-+ LKTRLabel(here);
-+ d_drop(a->h_dentry);
-+ /*d_drop(h_src_dentry);
-+ d_drop(src_dentry);*/
-+ inc_nlink(a->inode);
-+ a->inode->i_ctime = dir->i_ctime;
-+ }
++static inline int au_test_debugfs(struct super_block *sb __maybe_unused)
++{
++#ifdef CONFIG_DEBUG_FS
++ return sb->s_magic == DEBUGFS_MAGIC;
++#else
++ return 0;
+#endif
++}
+
-+ dir->i_version++;
-+ if (au_ibstart(dir) == au_dbstart(dentry))
-+ au_cpup_attr_timesizes(dir);
-+ if (!d_unhashed(a->h_dentry)
-+ /* || h_old_inode->i_nlink <= nlink */
-+ /* || SB_NFS(h_src_dentry->d_sb) */) {
-+ d_instantiate(dentry, au_igrab(inode));
-+ inc_nlink(inode);
-+ inode->i_ctime = dir->i_ctime;
-+ } else
-+ /* nfs case (< 2.6.15) */
-+ d_drop(dentry);
-+ goto out_unpin; /* success */
++static inline int au_test_nilfs(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_NILFS)
++ return sb->s_magic == NILFS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
++}
+
-+ out_revert:
-+ vfsub_args_reinit(&a->vargs);
-+ vfsub_ign_hinode(&a->vargs, IN_DELETE, au_pinned_hdir(&a->pin));
-+ rerr = vfsub_unlink(au_pinned_h_dir(&a->pin), a->h_dentry, &a->vargs);
-+ if (!rerr)
-+ goto out_dt;
-+ AuIOErr("%.*s reverting failed(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ err = -EIO;
-+ out_dt:
-+ d_drop(dentry);
-+ au_dtime_revert(&dt);
-+ out_unpin:
-+ au_unpin(&a->pin);
-+ out_wh:
-+ dput(wh_dentry);
-+ out_unlock:
-+ if (unlikely(err)) {
-+ au_update_dbstart(dentry);
-+ d_drop(dentry);
-+ }
-+ di_write_unlock(a->parent);
-+ dput(a->src_parent);
-+ aufs_read_and_write_unlock2(dentry, src_dentry);
-+ kfree(a);
-+ out:
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++static inline int au_test_hfsplus(struct super_block *sb __maybe_unused)
++{
++#if IS_ENABLED(CONFIG_HFSPLUS_FS)
++ return sb->s_magic == HFSPLUS_SUPER_MAGIC;
++#else
++ return 0;
++#endif
+}
+
-+int aufs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
++/* ---------------------------------------------------------------------- */
++/*
++ * they can't be an aufs branch.
++ */
++static inline int au_test_fs_unsuppoted(struct super_block *sb)
+{
-+ int err, rerr;
-+ struct dentry *h_dentry, *wh_dentry, *parent, *opq_dentry;
-+ struct mutex *h_mtx;
-+ struct au_dtime dt;
-+ aufs_bindex_t bindex;
-+ unsigned char diropq, dlgt;
-+ unsigned int mnt_flags;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_pin pin;
-+ struct au_wr_dir_args wr_dir_args = {
-+ .force_btgt = -1,
-+ .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR
-+ };
++ return
++#ifndef CONFIG_AUFS_BR_RAMFS
++ au_test_ramfs(sb) ||
++#endif
++ au_test_procfs(sb)
++ || au_test_sysfs(sb)
++ || au_test_configfs(sb)
++ || au_test_debugfs(sb)
++ || au_test_securityfs(sb)
++ || au_test_xenfs(sb)
++ || au_test_ecryptfs(sb)
++ /* || !strcmp(au_sbtype(sb), "unionfs") */
++ || au_test_aufs(sb); /* will be supported in next version */
++}
+
-+ LKTRTrace("i%lu, %.*s, mode 0%o\n",
-+ dir->i_ino, AuDLNPair(dentry), mode);
-+ IMustLock(dir);
++static inline int au_test_fs_remote(struct super_block *sb)
++{
++ return !au_test_tmpfs(sb)
++#ifdef CONFIG_AUFS_BR_RAMFS
++ && !au_test_ramfs(sb)
++#endif
++ && !(sb->s_type->fs_flags & FS_REQUIRES_DEV);
++}
+
-+ aufs_read_lock(dentry, AuLock_DW);
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ mnt_flags = au_mntflags(dentry->d_sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
++/* ---------------------------------------------------------------------- */
+
-+ di_write_lock_parent(parent);
-+ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, /*src_dentry*/NULL, &pin,
-+ &wr_dir_args);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out;
++/*
++ * Note: these functions (below) are created after reading ->getattr() in all
++ * filesystems under linux/fs. it means we have to do so in every update...
++ */
+
-+ bindex = au_dbstart(dentry);
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ vfsub_ign_hinode(&vargs, IN_CREATE, au_pinned_hdir(&pin));
-+ err = vfsub_mkdir(au_pinned_h_dir(&pin), h_dentry, mode, &vargs);
-+ if (unlikely(err))
-+ goto out_unlock;
++/*
++ * some filesystems require getattr to refresh the inode attributes before
++ * referencing.
++ * in most cases, we can rely on the inode attribute in NFS (or every remote fs)
++ * and leave the work for d_revalidate()
++ */
++static inline int au_test_fs_refresh_iattr(struct super_block *sb)
++{
++ return au_test_nfs(sb)
++ || au_test_fuse(sb)
++ /* || au_test_btrfs(sb) */ /* untested */
++ ;
++}
+
-+ /* make the dir opaque */
-+ diropq = 0;
-+ h_mtx = &h_dentry->d_inode->i_mutex;
-+ if (wh_dentry || au_opt_test(mnt_flags, ALWAYS_DIROPQ)) {
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ opq_dentry = au_diropq_create(dentry, bindex, /*dlgt*/0);
-+ mutex_unlock(h_mtx);
-+ err = PTR_ERR(opq_dentry);
-+ if (IS_ERR(opq_dentry))
-+ goto out_dir;
-+ dput(opq_dentry);
-+ diropq = 1;
-+ }
++/*
++ * filesystems which don't maintain i_size or i_blocks.
++ */
++static inline int au_test_fs_bad_iattr_size(struct super_block *sb)
++{
++ return au_test_xfs(sb)
++ || au_test_btrfs(sb)
++ || au_test_ubifs(sb)
++ || au_test_hfsplus(sb) /* maintained, but incorrect */
++ /* || au_test_minix(sb) */ /* untested */
++ ;
++}
+
-+ err = epilog(dir, bindex, wh_dentry, dentry);
-+ if (!err) {
-+ inc_nlink(dir);
-+ goto out_unlock; /* success */
-+ }
++/*
++ * filesystems which don't store the correct value in some of their inode
++ * attributes.
++ */
++static inline int au_test_fs_bad_iattr(struct super_block *sb)
++{
++ return au_test_fs_bad_iattr_size(sb)
++ || au_test_fat(sb)
++ || au_test_msdos(sb)
++ || au_test_vfat(sb);
++}
+
-+ /* revert */
-+ if (diropq) {
-+ LKTRLabel(revert opq);
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ rerr = au_diropq_remove(dentry, bindex, dlgt);
-+ mutex_unlock(h_mtx);
-+ if (rerr) {
-+ AuIOErr("%.*s reverting diropq failed(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ err = -EIO;
-+ }
-+ }
++/* they don't check i_nlink in link(2) */
++static inline int au_test_fs_no_limit_nlink(struct super_block *sb)
++{
++ return au_test_tmpfs(sb)
++#ifdef CONFIG_AUFS_BR_RAMFS
++ || au_test_ramfs(sb)
++#endif
++ || au_test_ubifs(sb)
++ || au_test_hfsplus(sb);
++}
+
-+ out_dir:
-+ LKTRLabel(revert dir);
-+ vfsub_args_reinit(&vargs);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, au_pinned_hdir(&pin));
-+ rerr = vfsub_rmdir(au_pinned_h_dir(&pin), h_dentry, &vargs);
-+ if (rerr) {
-+ AuIOErr("%.*s reverting dir failed(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ err = -EIO;
-+ }
-+ d_drop(dentry);
-+ au_dtime_revert(&dt);
-+ out_unlock:
-+ au_unpin(&pin);
-+ dput(wh_dentry);
-+ out:
-+ if (unlikely(err)) {
-+ au_update_dbstart(dentry);
-+ d_drop(dentry);
-+ }
-+ di_write_unlock(parent);
-+ aufs_read_unlock(dentry, AuLock_DW);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++/*
++ * filesystems which sets S_NOATIME and S_NOCMTIME.
++ */
++static inline int au_test_fs_notime(struct super_block *sb)
++{
++ return au_test_nfs(sb)
++ || au_test_fuse(sb)
++ || au_test_ubifs(sb)
++ ;
++}
++
++/* temporary support for i#1 in cramfs */
++static inline int au_test_fs_unique_ino(struct inode *inode)
++{
++ if (au_test_cramfs(inode->i_sb))
++ return inode->i_ino != 1;
++ return 1;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * the filesystem where the xino files placed must support i/o after unlink and
++ * maintain i_size and i_blocks.
++ */
++static inline int au_test_fs_bad_xino(struct super_block *sb)
++{
++ return au_test_fs_remote(sb)
++ || au_test_fs_bad_iattr_size(sb)
++ /* don't want unnecessary work for xino */
++ || au_test_aufs(sb)
++ || au_test_ecryptfs(sb)
++ || au_test_nilfs(sb);
++}
++
++static inline int au_test_fs_trunc_xino(struct super_block *sb)
++{
++ return au_test_tmpfs(sb)
++ || au_test_ramfs(sb);
++}
++
++/*
++ * test if the @sb is real-readonly.
++ */
++static inline int au_test_fs_rr(struct super_block *sb)
++{
++ return au_test_squashfs(sb)
++ || au_test_iso9660(sb)
++ || au_test_cramfs(sb)
++ || au_test_romfs(sb);
++}
++
++/*
++ * test if the @inode is nfs with 'noacl' option
++ * NFS always sets SB_POSIXACL regardless its mount option 'noacl.'
++ */
++static inline int au_test_nfs_noacl(struct inode *inode)
++{
++ return au_test_nfs(inode->i_sb)
++ /* && IS_POSIXACL(inode) */
++ && !nfs_server_capable(inode, NFS_CAP_ACLS);
+}
-diff -urN linux/fs/aufs/i_op.c linux-aufs/fs/aufs/i_op.c
---- linux/fs/aufs/i_op.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/i_op.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1049 @@
++
++#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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,65 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2017-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode operations (except add/del/rename)
-+ *
-+ * $Id$
++ * helpers for hlist_bl.h
+ */
+
-+#include <linux/fs_stack.h>
-+#include <linux/uaccess.h>
-+#include "aufs.h"
++#ifndef __AUFS_HBL_H__
++#define __AUFS_HBL_H__
+
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 27)
-+static int h_permission(struct inode *h_inode, int mask,
-+ struct vfsmount *h_mnt, int brperm, int dlgt)
-+{
-+ int err, submask;
-+ const int write_mask = (mask & (MAY_WRITE | MAY_APPEND));
++#ifdef __KERNEL__
+
-+ LKTRTrace("ino %lu, mask 0x%x, brperm 0x%x\n",
-+ h_inode->i_ino, mask, brperm);
++#include <linux/list_bl.h>
+
-+ err = -EACCES;
-+ if ((write_mask && IS_IMMUTABLE(h_inode))
-+ || ((mask & MAY_EXEC) && S_ISREG(h_inode->i_mode)
-+ && (h_mnt->mnt_flags & MNT_NOEXEC)))
-+ goto out;
++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);
++}
+
-+ /*
-+ * - skip hidden fs test in the case of write to ro branch.
-+ * - nfs dir permission write check is optimized, but a policy for
-+ * link/rename requires a real check.
-+ */
-+ submask = mask & ~MAY_APPEND;
-+ if ((write_mask && !au_br_writable(brperm))
-+ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode)
-+ && write_mask && !(mask & MAY_READ))
-+ || !h_inode->i_op
-+ || !h_inode->i_op->permission) {
-+ /* LKTRLabel(generic_permission); */
-+ err = generic_permission(h_inode, submask, NULL);
-+ } else {
-+ /* LKTRLabel(h_inode->permission); */
-+ err = h_inode->i_op->permission(h_inode, submask);
-+ AuTraceErr(err);
-+ }
++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);
++}
+
-+#if 1 /* todo: export? */
-+ if (!err)
-+ err = au_security_inode_permission(h_inode, mask, NULL,
-+ dlgt);
-+#endif
++#define au_hbl_for_each(pos, head) \
++ for (pos = hlist_bl_first(head); \
++ pos; \
++ pos = pos->next)
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++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;
+}
+
-+static int aufs_permission(struct inode *inode, int mask)
++#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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,290 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * fsnotify for the lower directories
++ */
++
++#include "aufs.h"
++
++/* FS_IN_IGNORED is unnecessary */
++static const __u32 AuHfsnMask = (FS_MOVED_TO | FS_MOVED_FROM | FS_DELETE
++ | FS_CREATE | FS_EVENT_ON_CHILD);
++static DECLARE_WAIT_QUEUE_HEAD(au_hfsn_wq);
++static __cacheline_aligned_in_smp atomic64_t au_hfsn_ifree = ATOMIC64_INIT(0);
++
++static void au_hfsn_free_mark(struct fsnotify_mark *mark)
++{
++ struct au_hnotify *hn = container_of(mark, struct au_hnotify,
++ hn_mark);
++ /* AuDbg("here\n"); */
++ au_cache_free_hnotify(hn);
++ smp_mb__before_atomic(); /* for atomic64_dec */
++ if (atomic64_dec_and_test(&au_hfsn_ifree))
++ wake_up(&au_hfsn_wq);
++}
++
++static int au_hfsn_alloc(struct au_hinode *hinode)
+{
+ int err;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char dlgt;
-+ const unsigned char isdir = S_ISDIR(inode->i_mode);
-+ struct inode *h_inode;
++ struct au_hnotify *hn;
+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ const int write_mask = (mask & (MAY_WRITE | MAY_APPEND));
++ struct au_branch *br;
++ struct fsnotify_mark *mark;
++ aufs_bindex_t bindex;
+
-+ LKTRTrace("ino %lu, mask 0x%x, isdir %d, write_mask %d\n",
-+ inode->i_ino, mask, isdir, write_mask);
++ hn = hinode->hi_notify;
++ sb = hn->hn_aufs_inode->i_sb;
++ bindex = au_br_index(sb, hinode->hi_id);
++ br = au_sbr(sb, bindex);
++ AuDebugOn(!br->br_hfsn);
+
-+ sb = inode->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ ii_read_lock_child(inode);
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
++ mark = &hn->hn_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_inode_mark(mark, hinode->hi_inode, /*allow_dups*/1);
++ lockdep_on();
+
-+ if (!isdir || write_mask || au_test_dirperm1(mnt_flags)) {
-+ h_inode = au_h_iptr(inode, au_ibstart(inode));
-+ AuDebugOn(!h_inode
-+ || ((h_inode->i_mode & S_IFMT)
-+ != (inode->i_mode & S_IFMT)));
-+ err = 0;
-+ bindex = au_ibstart(inode);
-+ LKTRTrace("b%d\n", bindex);
-+ err = h_permission(h_inode, mask, au_sbr_mnt(sb, bindex),
-+ au_sbr_perm(sb, bindex), dlgt);
++ return err;
++}
+
-+ if (write_mask && !err) {
-+ /* test whether the upper writable branch exists */
-+ err = -EROFS;
-+ for (; bindex >= 0; bindex--)
-+ if (!au_br_rdonly(au_sbr(sb, bindex))) {
-+ err = 0;
-+ break;
-+ }
-+ }
-+ goto out;
-+ }
++static int au_hfsn_free(struct au_hinode *hinode, struct au_hnotify *hn)
++{
++ struct fsnotify_mark *mark;
++ unsigned long long ull;
++ struct fsnotify_group *group;
+
-+ /* non-write to dir */
-+ err = 0;
-+ bend = au_ibend(inode);
-+ for (bindex = au_ibstart(inode); !err && bindex <= bend; bindex++) {
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (!h_inode)
-+ continue;
-+ AuDebugOn(!S_ISDIR(h_inode->i_mode));
++ ull = atomic64_inc_return(&au_hfsn_ifree);
++ BUG_ON(!ull);
+
-+ LKTRTrace("b%d\n", bindex);
-+ err = h_permission(h_inode, mask, au_sbr_mnt(sb, bindex),
-+ au_sbr_perm(sb, bindex), dlgt);
-+ }
++ mark = &hn->hn_mark;
++ spin_lock(&mark->lock);
++ group = mark->group;
++ fsnotify_get_group(group);
++ spin_unlock(&mark->lock);
++ lockdep_off();
++ fsnotify_destroy_mark(mark, group);
++ fsnotify_put_mark(mark);
++ fsnotify_put_group(group);
++ lockdep_on();
+
-+ out:
-+ ii_read_unlock(inode);
-+ si_read_unlock(sb);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++ /* free hn by myself */
++ return 0;
+}
-+#else
+
-+static int silly_lock(struct inode *inode, struct nameidata *nd)
-+{
-+ int locked = 0;
-+ struct super_block *sb = inode->i_sb;
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("i%lu, nd %p\n", inode->i_ino, nd);
++static void au_hfsn_ctl(struct au_hinode *hinode, int do_set)
++{
++ struct fsnotify_mark *mark;
+
-+ if (!nd || !nd->path.dentry) {
-+ si_read_lock(sb, AuLock_FLUSH);
-+ ii_read_lock_child(inode);
-+ } else if (nd->path.dentry->d_inode != inode) {
-+ locked = 1;
-+ /* lock child first, then parent */
-+ si_read_lock(sb, AuLock_FLUSH);
-+ ii_read_lock_child(inode);
-+ di_read_lock_parent(nd->path.dentry, 0);
++ mark = &hinode->hi_notify->hn_mark;
++ spin_lock(&mark->lock);
++ if (do_set) {
++ AuDebugOn(mark->mask & AuHfsnMask);
++ mark->mask |= AuHfsnMask;
+ } else {
-+ locked = 2;
-+ aufs_read_lock(nd->path.dentry, AuLock_FLUSH | AuLock_IR);
++ AuDebugOn(!(mark->mask & AuHfsnMask));
++ mark->mask &= ~AuHfsnMask;
+ }
-+ return locked;
++ spin_unlock(&mark->lock);
++ /* fsnotify_recalc_inode_mask(hinode->hi_inode); */
+}
+
-+static void silly_unlock(int locked, struct inode *inode, struct nameidata *nd)
++/* ---------------------------------------------------------------------- */
++
++/* #define AuDbgHnotify */
++#ifdef AuDbgHnotify
++static char *au_hfsn_name(u32 mask)
+{
-+ struct super_block *sb = inode->i_sb;
++#ifdef CONFIG_AUFS_DEBUG
++#define test_ret(flag) \
++ do { \
++ if (mask & flag) \
++ return #flag; \
++ } while (0)
++ test_ret(FS_ACCESS);
++ test_ret(FS_MODIFY);
++ test_ret(FS_ATTRIB);
++ test_ret(FS_CLOSE_WRITE);
++ test_ret(FS_CLOSE_NOWRITE);
++ test_ret(FS_OPEN);
++ test_ret(FS_MOVED_FROM);
++ test_ret(FS_MOVED_TO);
++ test_ret(FS_CREATE);
++ test_ret(FS_DELETE);
++ test_ret(FS_DELETE_SELF);
++ test_ret(FS_MOVE_SELF);
++ test_ret(FS_UNMOUNT);
++ test_ret(FS_Q_OVERFLOW);
++ test_ret(FS_IN_IGNORED);
++ test_ret(FS_ISDIR);
++ test_ret(FS_IN_ONESHOT);
++ test_ret(FS_EVENT_ON_CHILD);
++ return "";
++#undef test_ret
++#else
++ return "??";
++#endif
++}
++#endif
+
-+ LKTRTrace("locked %d, i%lu, nd %p\n", locked, inode->i_ino, nd);
++/* ---------------------------------------------------------------------- */
+
-+ switch (locked) {
-+ case 0:
-+ ii_read_unlock(inode);
-+ si_read_unlock(sb);
-+ break;
-+ case 1:
-+ di_read_unlock(nd->path.dentry, 0);
-+ ii_read_unlock(inode);
-+ si_read_unlock(sb);
-+ break;
-+ case 2:
-+ aufs_read_unlock(nd->path.dentry, AuLock_IR);
-+ break;
-+ default:
-+ BUG();
-+ }
++static void au_hfsn_free_group(struct fsnotify_group *group)
++{
++ struct au_br_hfsnotify *hfsn = group->private;
++
++ /* AuDbg("here\n"); */
++ au_kfree_try_rcu(hfsn);
+}
+
-+static int h_permission(struct inode *h_inode, int mask,
-+ struct nameidata *fake_nd, int brperm, int dlgt)
++static int au_hfsn_handle_event(struct fsnotify_group *group,
++ u32 mask, const void *data, int data_type,
++ struct inode *dir,
++ const struct qstr *file_name, u32 cookie,
++ struct fsnotify_iter_info *iter_info)
+{
-+ int err, submask;
-+ const int write_mask = (mask & (MAY_WRITE | MAY_APPEND));
++ int err;
++ struct au_hnotify *hnotify;
++ struct inode *h_dir, *h_inode;
++ struct fsnotify_mark *inode_mark;
+
-+ LKTRTrace("ino %lu, mask 0x%x, brperm 0x%x\n",
-+ h_inode->i_ino, mask, brperm);
++ AuDebugOn(!(data_type == FSNOTIFY_EVENT_INODE
++ || data_type == FSNOTIFY_EVENT_DENTRY));
+
-+ err = -EACCES;
-+ if ((write_mask && IS_IMMUTABLE(h_inode))
-+ || ((mask & MAY_EXEC) && S_ISREG(h_inode->i_mode)
-+ && fake_nd && fake_nd->path.mnt
-+ && (fake_nd->path.mnt->mnt_flags & MNT_NOEXEC)))
++ err = 0;
++ /* if FS_UNMOUNT happens, there must be another bug */
++ AuDebugOn(mask & FS_UNMOUNT);
++ if (mask & (FS_IN_IGNORED | FS_UNMOUNT))
+ goto out;
+
-+ /*
-+ * - skip hidden fs test in the case of write to ro branch.
-+ * - nfs dir permission write check is optimized, but a policy for
-+ * link/rename requires a real check.
-+ */
-+ submask = mask & ~MAY_APPEND;
-+ if ((write_mask && !au_br_writable(brperm))
-+ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode)
-+ && write_mask && !(mask & MAY_READ))
-+ || !h_inode->i_op
-+ || !h_inode->i_op->permission) {
-+ /* LKTRLabel(generic_permission); */
-+ err = generic_permission(h_inode, submask, NULL);
-+ } else {
-+ /* LKTRLabel(h_inode->permission); */
-+ err = h_inode->i_op->permission(h_inode, submask, fake_nd);
-+ AuTraceErr(err);
++ h_dir = dir;
++ h_inode = NULL;
++#ifdef AuDbgHnotify
++ au_debug_on();
++ if (1 || file_name.len != sizeof(AUFS_XINO_FNAME) - 1
++ || strncmp(file_name.name, AUFS_XINO_FNAME, file_name.len)) {
++ AuDbg("i%lu, mask 0x%x %s, hcname %.*s, hi%lu\n",
++ h_dir->i_ino, mask, au_hfsn_name(mask),
++ AuLNPair(file_name), h_inode ? h_inode->i_ino : 0);
++ /* WARN_ON(1); */
+ }
-+
-+#if 1 /* todo: export? */
-+ if (!err)
-+ err = au_security_inode_permission(h_inode, mask, fake_nd,
-+ dlgt);
++ au_debug_off();
+#endif
+
-+ out:
-+ AuTraceErr(err);
++ inode_mark = fsnotify_iter_inode_mark(iter_info);
++ AuDebugOn(!inode_mark);
++ hnotify = container_of(inode_mark, struct au_hnotify, hn_mark);
++ err = au_hnotify(h_dir, hnotify, mask, file_name, h_inode);
++
++out:
+ return err;
+}
+
-+static int aufs_permission(struct inode *inode, int mask, struct nameidata *nd)
++static struct fsnotify_ops au_hfsn_ops = {
++ .handle_event = au_hfsn_handle_event,
++ .free_group_priv = au_hfsn_free_group,
++ .free_mark = au_hfsn_free_mark
++};
++
++/* ---------------------------------------------------------------------- */
++
++static void au_hfsn_fin_br(struct au_branch *br)
+{
-+ int err;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char locked, dlgt, do_nd;
-+ const unsigned char isdir = S_ISDIR(inode->i_mode);
-+ struct inode *h_inode;
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ struct path path;
-+ const int write_mask = (mask & (MAY_WRITE | MAY_APPEND));
++ struct au_br_hfsnotify *hfsn;
+
-+ LKTRTrace("ino %lu, mask 0x%x, isdir %d, write_mask %d, "
-+ "nd %d{%d, %d}\n",
-+ inode->i_ino, mask, isdir, write_mask,
-+ !!nd, nd ? !!nd->path.dentry : 0, nd ? !!nd->path.mnt : 0);
++ hfsn = br->br_hfsn;
++ if (hfsn) {
++ lockdep_off();
++ fsnotify_put_group(hfsn->hfsn_group);
++ lockdep_on();
++ }
++}
+
-+ sb = inode->i_sb;
-+ locked = silly_lock(inode, nd);
-+ do_nd = (nd && locked >= 1);
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
++static int au_hfsn_init_br(struct au_branch *br, int perm)
++{
++ int err;
++ struct fsnotify_group *group;
++ struct au_br_hfsnotify *hfsn;
+
-+ if (!isdir || write_mask || au_test_dirperm1(mnt_flags)) {
-+ h_inode = au_h_iptr(inode, au_ibstart(inode));
-+ AuDebugOn(!h_inode
-+ || ((h_inode->i_mode & S_IFMT)
-+ != (inode->i_mode & S_IFMT)));
-+ err = 0;
-+ bindex = au_ibstart(inode);
-+ LKTRTrace("b%d\n", bindex);
-+ if (do_nd) {
-+ path = nd->path;
-+ nd->path.mnt = au_sbr_mnt(sb, bindex);
-+ nd->path.dentry = au_h_dptr(nd->path.dentry, bindex);
-+ path_get(&nd->path);
-+ err = h_permission(h_inode, mask, nd,
-+ au_sbr_perm(sb, bindex), dlgt);
-+ path_put(&nd->path);
-+ nd->path = path;
-+ } else {
-+ AuDebugOn(nd && nd->path.mnt);
-+ err = h_permission(h_inode, mask, nd,
-+ au_sbr_perm(sb, bindex), dlgt);
-+ }
++ err = 0;
++ br->br_hfsn = NULL;
++ if (!au_br_hnotifyable(perm))
++ goto out;
+
-+ if (write_mask && !err) {
-+ /* test whether the upper writable branch exists */
-+ err = -EROFS;
-+ for (; bindex >= 0; bindex--)
-+ if (!au_br_rdonly(au_sbr(sb, bindex))) {
-+ err = 0;
-+ break;
-+ }
-+ }
++ err = -ENOMEM;
++ hfsn = kmalloc(sizeof(*hfsn), GFP_NOFS);
++ if (unlikely(!hfsn))
+ goto out;
-+ }
+
-+ /* non-write to dir */
-+ if (do_nd)
-+ path = nd->path;
-+ else {
-+ path.mnt = NULL;
-+ path.dentry = NULL;
-+ }
+ err = 0;
-+ bend = au_ibend(inode);
-+ for (bindex = au_ibstart(inode); !err && bindex <= bend; bindex++) {
-+ h_inode = au_h_iptr(inode, bindex);
-+ if (!h_inode)
-+ continue;
-+ AuDebugOn(!S_ISDIR(h_inode->i_mode));
-+
-+ LKTRTrace("b%d\n", bindex);
-+ if (do_nd) {
-+ nd->path.mnt = au_sbr_mnt(sb, bindex);
-+ nd->path.dentry = au_h_dptr(path.dentry, bindex);
-+ path_get(&nd->path);
-+ err = h_permission(h_inode, mask, nd,
-+ au_sbr_perm(sb, bindex), dlgt);
-+ path_put(&nd->path);
-+ } else {
-+ AuDebugOn(nd && nd->path.mnt);
-+ err = h_permission(h_inode, mask, nd,
-+ au_sbr_perm(sb, bindex), dlgt);
-+ }
++ group = fsnotify_alloc_group(&au_hfsn_ops,
++ /*flags - not for userspace*/0);
++ if (IS_ERR(group)) {
++ err = PTR_ERR(group);
++ pr_err("fsnotify_alloc_group() failed, %d\n", err);
++ goto out_hfsn;
+ }
-+ if (do_nd)
-+ nd->path = path;
+
-+ out:
-+ silly_unlock(locked, inode, nd);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
++ group->private = hfsn;
++ hfsn->hfsn_group = group;
++ br->br_hfsn = hfsn;
++ goto out; /* success */
++
++out_hfsn:
++ au_kfree_try_rcu(hfsn);
++out:
+ return err;
+}
-+#endif /* KERNEL_VERSION(2, 6, 27) */
-+
-+/* ---------------------------------------------------------------------- */
+
-+static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry,
-+ struct nameidata *nd)
++static int au_hfsn_reset_br(unsigned int udba, struct au_branch *br, int perm)
+{
-+ struct dentry *ret, *parent;
-+ int err, npositive;
-+ struct inode *inode, *h_inode;
-+ struct nameidata tmp_nd, *ndp;
-+ aufs_bindex_t bstart;
-+ struct mutex *mtx;
-+ struct super_block *sb;
-+
-+ LKTRTrace("dir %lu, %.*s, nd{0x%x}\n",
-+ dir->i_ino, AuDLNPair(dentry), nd ? nd->flags : 0);
-+ AuDebugOn(IS_ROOT(dentry));
-+ IMustLock(dir);
++ int err;
+
-+ sb = dir->i_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ err = au_alloc_dinfo(dentry);
-+ ret = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out;
++ err = 0;
++ if (!br->br_hfsn)
++ err = au_hfsn_init_br(br, perm);
+
-+ /* nd can be NULL */
-+ ndp = au_dup_nd(au_sbi(sb), &tmp_nd, nd);
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_read_lock_parent(parent, AuLock_IR);
-+ npositive = au_lkup_dentry(dentry, au_dbstart(parent), /*type*/0, ndp);
-+ di_read_unlock(parent, AuLock_IR);
-+ err = npositive;
-+ ret = ERR_PTR(err);
-+ if (unlikely(err < 0))
-+ goto out_unlock;
++ return err;
++}
+
-+ inode = NULL;
-+ if (npositive) {
-+ bstart = au_dbstart(dentry);
-+ h_inode = au_h_dptr(dentry, bstart)->d_inode;
-+ AuDebugOn(!h_inode);
-+ if (!S_ISDIR(h_inode->i_mode)) {
-+ /*
-+ * stop 'race'-ing between hardlinks under different
-+ * parents.
-+ */
-+ mtx = &au_sbr(sb, bstart)->br_xino.xi_nondir_mtx;
-+ mutex_lock(mtx);
-+ inode = au_new_inode(dentry, /*must_new*/0);
-+ mutex_unlock(mtx);
-+ } else
-+ inode = au_new_inode(dentry, /*must_new*/0);
-+ ret = (void *)inode;
-+ }
-+ if (!IS_ERR(inode)) {
-+ ret = d_splice_alias(inode, dentry);
-+ if (unlikely(IS_ERR(ret) && inode))
-+ ii_write_unlock(inode);
-+ AuDebugOn(nd
-+ && (nd->flags & LOOKUP_OPEN)
-+ && nd->intent.open.file
-+ && nd->intent.open.file->f_dentry);
-+ au_store_fmode_exec(nd, inode);
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ out_unlock:
-+ di_write_unlock(dentry);
-+ out:
-+ si_read_unlock(sb);
-+ AuTraceErrPtr(ret);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return ret;
++static void au_hfsn_fin(void)
++{
++ AuDbg("au_hfsn_ifree %lld\n", (long long)atomic64_read(&au_hfsn_ifree));
++ wait_event(au_hfsn_wq, !atomic64_read(&au_hfsn_ifree));
+}
+
-+/* ---------------------------------------------------------------------- */
++const struct au_hnotify_op au_hnotify_op = {
++ .ctl = au_hfsn_ctl,
++ .alloc = au_hfsn_alloc,
++ .free = au_hfsn_free,
++
++ .fin = au_hfsn_fin,
+
++ .reset_br = au_hfsn_reset_br,
++ .fin_br = au_hfsn_fin_br,
++ .init_br = au_hfsn_init_br
++};
+diff -urN /usr/share/empty/fs/aufs/hfsplus.c linux/fs/aufs/hfsplus.c
+--- /usr/share/empty/fs/aufs/hfsplus.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/hfsplus.c 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,60 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * decide the branch and the parent dir where we will create a new entry.
-+ * returns new bindex or an error.
-+ * copyup the parent dir if needed.
++ * Copyright (C) 2010-2022 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/>.
+ */
-+int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
-+ struct au_wr_dir_args *args)
++
++/*
++ * special support for filesystems which acquires an inode mutex
++ * at final closing a file, eg, hfsplus.
++ *
++ * This trick is very simple and stupid, just to open the file before really
++ * necessary open to tell hfsplus that this is not the final closing.
++ * The caller should call au_h_open_pre() after acquiring the inode mutex,
++ * and au_h_open_post() after releasing it.
++ */
++
++#include "aufs.h"
++
++struct file *au_h_open_pre(struct dentry *dentry, aufs_bindex_t bindex,
++ int force_wr)
+{
-+ int err;
-+ aufs_bindex_t bcpup, bstart, src_bstart;
-+ struct super_block *sb;
-+ struct dentry *parent;
-+ struct au_sbinfo *sbinfo;
-+ const int add_entry = au_ftest_wrdir(args->flags, ADD_ENTRY);
++ struct file *h_file;
++ struct dentry *h_dentry;
+
-+ LKTRTrace("%.*s, src %p, {%d, 0x%x}\n",
-+ AuDLNPair(dentry), src_dentry, args->force_btgt, args->flags);
++ h_dentry = au_h_dptr(dentry, bindex);
++ AuDebugOn(!h_dentry);
++ AuDebugOn(d_is_negative(h_dentry));
+
-+ sb = dentry->d_sb;
-+ sbinfo = au_sbi(sb);
-+ parent = dget_parent(dentry);
-+ bstart = au_dbstart(dentry);
-+ bcpup = bstart;
-+ if (args->force_btgt < 0) {
-+ if (src_dentry) {
-+ src_bstart = au_dbstart(src_dentry);
-+ if (src_bstart < bstart)
-+ bcpup = src_bstart;
-+ } else if (add_entry) {
-+ err = AuWbrCreate(sbinfo, dentry,
-+ au_ftest_wrdir(args->flags, ISDIR));
-+ bcpup = err;
-+ }
++ h_file = NULL;
++ if (au_test_hfsplus(h_dentry->d_sb)
++ && d_is_reg(h_dentry))
++ h_file = au_h_open(dentry, bindex,
++ O_RDONLY | O_NOATIME | O_LARGEFILE,
++ /*file*/NULL, force_wr);
++ return h_file;
++}
+
-+ if (bcpup < 0 || au_test_ro(sb, bcpup, dentry->d_inode)) {
-+ if (add_entry)
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ else {
-+ if (!IS_ROOT(dentry)) {
-+ di_read_lock_parent(parent, !AuLock_IR);
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ di_read_unlock(parent, !AuLock_IR);
-+ } else
-+ err = AuWbrCopyup(sbinfo, dentry);
-+ }
-+ bcpup = err;
-+ if (unlikely(err < 0))
-+ goto out;
-+ }
-+ } else {
-+ bcpup = args->force_btgt;
-+ AuDebugOn(au_test_ro(sb, bcpup, dentry->d_inode));
++void au_h_open_post(struct dentry *dentry, aufs_bindex_t bindex,
++ struct file *h_file)
++{
++ struct au_branch *br;
++
++ if (h_file) {
++ fput(h_file);
++ br = au_sbr(dentry->d_sb, bindex);
++ au_lcnt_dec(&br->br_nfiles);
+ }
-+ LKTRTrace("bstart %d, bcpup %d\n", bstart, bcpup);
-+ if (bstart < bcpup)
-+ au_update_dbrange(dentry, /*do_put_zero*/1);
++}
+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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,715 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+ err = bcpup;
-+ if (bcpup == bstart)
-+ goto out; /* success */
++/*
++ * abstraction to notify the direct changes on lower directories
++ */
+
-+ /* copyup the new parent into the branch we process */
-+ if (add_entry) {
-+ au_update_dbstart(dentry);
-+ IMustLock(parent->d_inode);
-+ DiMustWriteLock(parent);
-+ IiMustWriteLock(parent->d_inode);
-+ } else
-+ di_write_lock_parent(parent);
++/* #include <linux/iversion.h> */
++#include "aufs.h"
+
-+ err = 0;
-+ if (!au_h_dptr(parent, bcpup)) {
-+ if (bstart < bcpup)
-+ err = au_cpdown_dirs(dentry, bcpup);
-+ else
-+ err = au_cpup_dirs(dentry, bcpup);
-+ }
-+ if (!err && add_entry) {
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
++int au_hn_alloc(struct au_hinode *hinode, struct inode *inode)
++{
++ int err;
++ struct au_hnotify *hn;
+
-+ h_parent = au_h_dptr(parent, bcpup);
-+ AuDebugOn(!h_parent);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!h_dir);
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ err = au_lkup_neg(dentry, bcpup);
-+ mutex_unlock(&h_dir->i_mutex);
-+ if (bstart < bcpup && au_dbstart(dentry) < 0) {
-+ au_set_dbstart(dentry, 0);
-+ au_update_dbrange(dentry, /*do_put_zero*/0);
++ err = -ENOMEM;
++ hn = au_cache_alloc_hnotify();
++ if (hn) {
++ hn->hn_aufs_inode = inode;
++ hinode->hi_notify = hn;
++ err = au_hnotify_op.alloc(hinode);
++ AuTraceErr(err);
++ if (unlikely(err)) {
++ hinode->hi_notify = NULL;
++ au_cache_free_hnotify(hn);
++ /*
++ * The upper dir was removed by udba, but the same named
++ * dir left. In this case, aufs assigns a new inode
++ * number and set the monitor again.
++ * For the lower dir, the old monitor is still left.
++ */
++ if (err == -EEXIST)
++ err = 0;
+ }
+ }
+
-+ if (!add_entry)
-+ di_write_unlock(parent);
-+ if (!err)
-+ err = bcpup; /* success */
-+ out:
-+ dput(parent);
-+ LKTRTrace("err %d\n", err);
+ AuTraceErr(err);
+ return err;
+}
+
++void au_hn_free(struct au_hinode *hinode)
++{
++ struct au_hnotify *hn;
++
++ hn = hinode->hi_notify;
++ if (hn) {
++ hinode->hi_notify = NULL;
++ if (au_hnotify_op.free(hinode, hn))
++ au_cache_free_hnotify(hn);
++ }
++}
++
+/* ---------------------------------------------------------------------- */
+
-+struct dentry *au_do_pinned_h_parent(struct au_pin1 *pin)
++void au_hn_ctl(struct au_hinode *hinode, int do_set)
+{
-+ if (pin && pin->parent)
-+ return au_h_dptr(pin->parent, pin->bindex);
-+ return NULL;
++ if (hinode->hi_notify)
++ au_hnotify_op.ctl(hinode, do_set);
+}
+
-+void au_do_unpin(struct au_pin1 *p, struct au_pin1 *gp)
++void au_hn_reset(struct inode *inode, unsigned int flags)
+{
-+ if (p->dentry)
-+ LKTRTrace("%.*s\n", AuDLNPair(p->dentry));
-+ LKTRTrace("p{%d, %d, %d, %d, 0x%x}, gp %p\n",
-+ p->lsc_di, p->lsc_hi, !!p->parent, !!p->h_dir, p->flags, gp);
++ aufs_bindex_t bindex, bbot;
++ struct inode *hi;
++ struct dentry *iwhdentry;
+
-+ if (au_ftest_pin(p->flags, MNT_WRITE))
-+ au_br_drop_write(au_sbr(p->dentry->d_sb, p->bindex));
-+ if (au_ftest_pin(p->flags, VFS_RENAME))
-+ vfsub_unlock_rename_mutex(au_sbr_sb(p->dentry->d_sb,
-+ p->bindex));
-+ if (!p->h_dir)
-+ return;
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
++ hi = au_h_iptr(inode, bindex);
++ if (!hi)
++ continue;
+
-+ mutex_unlock(&p->h_dir->i_mutex);
-+ if (gp)
-+ au_do_unpin(gp, NULL);
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_unlock(p->parent, AuLock_IR);
-+ iput(p->h_dir);
-+ dput(p->parent);
-+ p->parent = NULL;
-+ p->h_dir = NULL;
++ /* inode_lock_nested(hi, AuLsc_I_CHILD); */
++ iwhdentry = au_hi_wh(inode, bindex);
++ if (iwhdentry)
++ dget(iwhdentry);
++ au_igrab(hi);
++ au_set_h_iptr(inode, bindex, NULL, 0);
++ au_set_h_iptr(inode, bindex, au_igrab(hi),
++ flags & ~AuHi_XINO);
++ iput(hi);
++ dput(iwhdentry);
++ /* inode_unlock(hi); */
++ }
+}
+
-+int au_do_pin(struct au_pin1 *p, struct au_pin1 *gp)
++/* ---------------------------------------------------------------------- */
++
++static int hn_xino(struct inode *inode, struct inode *h_inode)
+{
+ int err;
-+ struct dentry *h_dentry;
-+
-+ LKTRTrace("%.*s, %d, b%d, 0x%x\n",
-+ AuDLNPair(p->dentry), !!gp, p->bindex, p->flags);
-+ AuDebugOn(au_ftest_pin(p->flags, DO_GPARENT) && !gp);
-+ /* AuDebugOn(!do_gp && gp); */
++ aufs_bindex_t bindex, bbot, bfound, btop;
++ struct inode *h_i;
+
+ err = 0;
-+ if (IS_ROOT(p->dentry)) {
-+ if (au_ftest_pin(p->flags, VFS_RENAME))
-+ vfsub_lock_rename_mutex(au_sbr_sb(p->dentry->d_sb,
-+ p->bindex));
-+ if (au_ftest_pin(p->flags, MNT_WRITE)) {
-+ err = au_br_want_write(au_sbr(p->dentry->d_sb,
-+ p->bindex));
-+ if (unlikely(err))
-+ au_fclr_pin(p->flags, MNT_WRITE);
-+ }
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("branch root dir was changed\n");
+ goto out;
+ }
+
-+ h_dentry = NULL;
-+ if (p->bindex <= au_dbend(p->dentry))
-+ h_dentry = au_h_dptr(p->dentry, p->bindex);
-+
-+ p->parent = dget_parent(p->dentry);
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_lock(p->parent, AuLock_IR, p->lsc_di);
-+ else
-+ DiMustAnyLock(p->parent);
-+ AuDebugOn(!p->parent->d_inode);
-+ p->h_dir = au_igrab(au_h_iptr(p->parent->d_inode, p->bindex));
-+ /* udba case */
-+ if (unlikely(au_ftest_pin(p->flags, VERIFY) && !p->h_dir)) {
-+ err = -EIO;
-+ if (!au_ftest_pin(p->flags, DI_LOCKED))
-+ di_read_unlock(p->parent, AuLock_IR);
-+ dput(p->parent);
-+ p->parent = NULL;
++ bfound = -1;
++ bbot = au_ibbot(inode);
++ btop = au_ibtop(inode);
++#if 0 /* reserved for future use */
++ if (bindex == bbot) {
++ /* keep this ino in rename case */
+ goto out;
+ }
-+
-+ if (au_ftest_pin(p->flags, DO_GPARENT)) {
-+ gp->dentry = p->parent;
-+ err = au_do_pin(gp, NULL);
-+ if (unlikely(err))
-+ gp->dentry = NULL;
-+ }
-+ if (au_ftest_pin(p->flags, VFS_RENAME))
-+ vfsub_lock_rename_mutex(p->h_dir->i_sb);
-+ mutex_lock_nested(&p->h_dir->i_mutex, p->lsc_hi);
-+ if (!err) {
-+ /* todo: call d_revalidate() here? */
-+#if 0
-+ if (h_dentry && h_dentry->d_op && h_dentry->d_op->d_revalidate
-+ && !h_dentry->d_op->d_revalidate(h_dentry, NULL))
-+ err = -EIO;
+#endif
-+ if (!h_dentry
-+ || !au_ftest_pin(p->flags, VERIFY)
-+ || !au_verify_parent(h_dentry, p->h_dir)) {
-+ if (au_ftest_pin(p->flags, MNT_WRITE)) {
-+ err = au_br_want_write(au_sbr(p->dentry->d_sb,
-+ p->bindex));
-+ if (unlikely(err))
-+ au_fclr_pin(p->flags, MNT_WRITE);
-+ }
-+ if (!err)
-+ goto out; /* success */
-+ } else
-+ err = -EIO;
-+ }
++ for (bindex = btop; bindex <= bbot; bindex++)
++ if (au_h_iptr(inode, bindex) == h_inode) {
++ bfound = bindex;
++ break;
++ }
++ if (bfound < 0)
++ goto out;
+
-+ AuDbgDentry(p->dentry);
-+ AuDbgDentry(h_dentry);
-+ AuDbgDentry(p->parent);
-+ AuDbgInode(p->h_dir);
-+ if (h_dentry)
-+ AuDbgDentry(h_dentry->d_parent);
++ for (bindex = btop; bindex <= bbot; bindex++) {
++ h_i = au_h_iptr(inode, bindex);
++ if (!h_i)
++ continue;
+
-+ au_do_unpin(p, gp);
-+ if (au_ftest_pin(p->flags, DO_GPARENT))
-+ gp->dentry = NULL;
++ err = au_xino_write(inode->i_sb, bindex, h_i->i_ino, /*ino*/0);
++ /* ignore this error */
++ /* bad action? */
++ }
++
++ /* children inode number will be broken */
+
-+ out:
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+void au_pin_init(struct au_pin *args, struct dentry *dentry,
-+ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
-+ unsigned char flags)
++static int hn_gen_tree(struct dentry *dentry)
+{
-+ struct au_pin1 *p;
-+ unsigned char f;
++ int err, i, j, ndentry;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry **dentries;
+
-+ AuTraceEnter();
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_dcsub_pages(&dpages, dentry, NULL, NULL);
++ if (unlikely(err))
++ goto out_dpages;
+
-+ memset(args, 0, sizeof(*args));
-+ p = args->pin + AuPin_PARENT;
-+ p->dentry = dentry;
-+ p->lsc_di = lsc_di;
-+ p->lsc_hi = lsc_hi;
-+ p->flags = flags;
-+ p->bindex = bindex;
-+ if (!au_opt_test(au_mntflags(dentry->d_sb), UDBA_NONE))
-+ au_fset_pin(p->flags, VERIFY);
-+ if (!au_ftest_pin(flags, DO_GPARENT))
-+ return;
++ for (i = 0; i < dpages.ndpage; i++) {
++ dpage = dpages.dpages + i;
++ dentries = dpage->dentries;
++ ndentry = dpage->ndentry;
++ for (j = 0; j < ndentry; j++) {
++ struct dentry *d;
+
-+ f = p->flags;
-+ p = au_pin_gp(args);
-+ if (p) {
-+ p->lsc_di = lsc_di + 1; /* child first */
-+ p->lsc_hi = lsc_hi - 1; /* parent first */
-+ p->bindex = bindex;
-+ p->flags = f & ~(AuPin_MNT_WRITE
-+ | AuPin_DO_GPARENT
-+ | AuPin_DI_LOCKED);
-+ }
-+}
++ d = dentries[j];
++ if (IS_ROOT(d))
++ continue;
+
-+int au_pin(struct au_pin *args, struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned char flags)
-+{
-+ LKTRTrace("%.*s, b%d, 0x%x\n",
-+ AuDLNPair(dentry), bindex, flags);
++ au_digen_dec(d);
++ if (d_really_is_positive(d))
++ /* todo: reset children xino?
++ cached children only? */
++ au_iigen_dec(d_inode(d));
++ }
++ }
+
-+ au_pin_init(args, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2,
-+ flags);
-+ return au_do_pin(args->pin + AuPin_PARENT, au_pin_gp(args));
++out_dpages:
++ au_dpages_free(&dpages);
++out:
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_icpup_args {
-+ unsigned char isdir, hinotify, did_cpup; /* flags */
-+ unsigned char pin_flags;
-+ aufs_bindex_t btgt;
-+ struct au_pin pin;
-+ struct au_hin_ignore ign[2];
-+ struct vfsub_args vargs;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+};
-+
-+/* todo: refine it */
-+static int au_lock_and_icpup(struct dentry *dentry, loff_t sz,
-+ struct au_icpup_args *a)
++/*
++ * return 0 if processed.
++ */
++static int hn_gen_by_inode(char *name, unsigned int nlen, struct inode *inode,
++ const unsigned int isdir)
+{
+ int err;
-+ aufs_bindex_t bstart;
-+ struct super_block *sb;
-+ struct dentry *hi_wh, *parent;
-+ struct inode *inode;
-+ struct au_wr_dir_args wr_dir_args = {
-+ .force_btgt = -1,
-+ .flags = 0
-+ };
-+
-+ LKTRTrace("%.*s, %lld\n", AuDLNPair(dentry), sz);
-+
-+ di_write_lock_child(dentry);
-+ bstart = au_dbstart(dentry);
-+ sb = dentry->d_sb;
-+ inode = dentry->d_inode;
-+ a->isdir = !!S_ISDIR(inode->i_mode);
-+ if (a->isdir)
-+ au_fset_wrdir(wr_dir_args.flags, ISDIR);
-+ /* plink or hi_wh() */
-+ if (bstart != au_ibstart(inode))
-+ wr_dir_args.force_btgt = au_ibstart(inode);
-+ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args);
-+ if (unlikely(err < 0))
-+ goto out_dentry;
-+ a->btgt = err;
-+ a->did_cpup = (err != bstart);
-+ err = 0;
++ struct dentry *d;
++ struct qstr *dname;
+
-+ /* crazy udba locks */
-+ a->pin_flags = AuPin_MNT_WRITE;
-+ a->hinotify = !!au_opt_test(au_mntflags(sb), UDBA_INOTIFY);
-+ if (a->hinotify)
-+ au_fset_pin(a->pin_flags, DO_GPARENT);
-+ parent = NULL;
-+ if (!IS_ROOT(dentry)) {
-+ au_fset_pin(a->pin_flags, DI_LOCKED);
-+ parent = dget_parent(dentry);
-+ di_write_lock_parent(parent);
-+ }
-+ err = au_pin(&a->pin, dentry, a->btgt, a->pin_flags);
-+ if (unlikely(err)) {
-+ if (parent) {
-+ di_write_unlock(parent);
-+ dput(parent);
-+ }
-+ goto out_dentry;
-+ }
-+ a->h_dentry = au_h_dptr(dentry, bstart);
-+ a->h_inode = a->h_dentry->d_inode;
-+ AuDebugOn(!a->h_inode);
-+ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD);
-+ if (!a->did_cpup) {
-+ au_unpin_gp(&a->pin);
-+ if (parent) {
-+ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ dput(parent);
-+ }
-+ goto out; /* success */
++ err = 1;
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("branch root dir was changed\n");
++ err = 0;
++ goto out;
+ }
+
-+ hi_wh = NULL;
-+ if (!d_unhashed(dentry)) {
-+ if (parent) {
-+ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ dput(parent);
++ if (!isdir) {
++ AuDebugOn(!name);
++ au_iigen_dec(inode);
++ spin_lock(&inode->i_lock);
++ hlist_for_each_entry(d, &inode->i_dentry, d_u.d_alias) {
++ spin_lock(&d->d_lock);
++ dname = &d->d_name;
++ if (dname->len != nlen
++ && memcmp(dname->name, name, nlen)) {
++ spin_unlock(&d->d_lock);
++ continue;
++ }
++ err = 0;
++ au_digen_dec(d);
++ spin_unlock(&d->d_lock);
++ break;
+ }
-+ err = au_sio_cpup_simple(dentry, a->btgt, sz, AuCpup_DTIME);
-+ if (!err)
-+ a->h_dentry = au_h_dptr(dentry, a->btgt);
++ spin_unlock(&inode->i_lock);
+ } else {
-+ hi_wh = au_hi_wh(inode, a->btgt);
-+ if (!hi_wh) {
-+ err = au_sio_cpup_wh(dentry, a->btgt, sz,
-+ /*file*/NULL);
-+ if (!err)
-+ hi_wh = au_hi_wh(inode, a->btgt);
-+ /* todo: revalidate hi_wh? */
-+ }
-+ if (parent) {
-+ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
-+ di_downgrade_lock(parent, AuLock_IR);
-+ dput(parent);
++ au_fset_si(au_sbi(inode->i_sb), FAILED_REFRESH_DIR);
++ d = d_find_any_alias(inode);
++ if (!d) {
++ au_iigen_dec(inode);
++ goto out;
+ }
-+ if (!hi_wh)
-+ a->h_dentry = au_h_dptr(dentry, a->btgt);
-+ else
-+ a->h_dentry = hi_wh; /* do not dget here */
-+ }
+
-+ mutex_unlock(&a->h_inode->i_mutex);
-+ a->h_inode = a->h_dentry->d_inode;
-+ AuDebugOn(!a->h_inode);
-+ if (!err) {
-+ mutex_lock_nested(&a->h_inode->i_mutex, AuLsc_I_CHILD);
-+ au_unpin_gp(&a->pin);
-+ goto out; /* success */
++ spin_lock(&d->d_lock);
++ dname = &d->d_name;
++ if (dname->len == nlen && !memcmp(dname->name, name, nlen)) {
++ spin_unlock(&d->d_lock);
++ err = hn_gen_tree(d);
++ spin_lock(&d->d_lock);
++ }
++ spin_unlock(&d->d_lock);
++ dput(d);
+ }
+
-+ au_unpin(&a->pin);
-+
-+ out_dentry:
-+ di_write_unlock(dentry);
-+ out:
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+static int aufs_setattr(struct dentry *dentry, struct iattr *ia)
++static int hn_gen_by_name(struct dentry *dentry, const unsigned int isdir)
+{
+ int err;
-+ struct inode *inode;
-+ struct super_block *sb;
-+ __u32 events;
-+ struct file *file;
-+ loff_t sz;
-+ struct au_icpup_args *a;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ inode = dentry->d_inode;
-+ IMustLock(inode);
++ if (IS_ROOT(dentry)) {
++ pr_warn("branch root dir was changed\n");
++ return 0;
++ }
+
-+ err = -ENOMEM;
-+ a = kzalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++ err = 0;
++ if (!isdir) {
++ au_digen_dec(dentry);
++ if (d_really_is_positive(dentry))
++ au_iigen_dec(d_inode(dentry));
++ } else {
++ au_fset_si(au_sbi(dentry->d_sb), FAILED_REFRESH_DIR);
++ if (d_really_is_positive(dentry))
++ err = hn_gen_tree(dentry);
++ }
+
-+ file = NULL;
-+ sb = dentry->d_sb;
-+ si_read_lock(sb, AuLock_FLUSH);
-+ vfsub_args_init(&a->vargs, a->ign, au_test_dlgt(au_mntflags(sb)), 0);
++ AuTraceErr(err);
++ return err;
++}
+
-+ if (ia->ia_valid & ATTR_FILE) {
-+ /* currently ftruncate(2) only */
-+ file = ia->ia_file;
-+ fi_write_lock(file);
-+ ia->ia_file = au_h_fptr(file, au_fbstart(file));
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ sz = -1;
-+ if ((ia->ia_valid & ATTR_SIZE)
-+ && ia->ia_size < i_size_read(inode))
-+ sz = ia->ia_size;
-+ err = au_lock_and_icpup(dentry, sz, a);
-+ if (unlikely(err < 0))
-+ goto out_si;
-+ if (a->did_cpup) {
-+ ia->ia_file = NULL;
-+ ia->ia_valid &= ~ATTR_FILE;
++/* hnotify job flags */
++#define AuHnJob_XINO0 1
++#define AuHnJob_GEN (1 << 1)
++#define AuHnJob_DIRENT (1 << 2)
++#define AuHnJob_ISDIR (1 << 3)
++#define AuHnJob_TRYXINO0 (1 << 4)
++#define AuHnJob_MNTPNT (1 << 5)
++#define au_ftest_hnjob(flags, name) ((flags) & AuHnJob_##name)
++#define au_fset_hnjob(flags, name) \
++ do { (flags) |= AuHnJob_##name; } while (0)
++#define au_fclr_hnjob(flags, name) \
++ do { (flags) &= ~AuHnJob_##name; } while (0)
++
++enum {
++ AuHn_CHILD,
++ AuHn_PARENT,
++ AuHnLast
++};
++
++struct au_hnotify_args {
++ struct inode *h_dir, *dir, *h_child_inode;
++ u32 mask;
++ unsigned int flags[AuHnLast];
++ unsigned int h_child_nlen;
++ char h_child_name[];
++};
++
++struct hn_job_args {
++ unsigned int flags;
++ struct inode *inode, *h_inode, *dir, *h_dir;
++ struct dentry *dentry;
++ char *h_name;
++ int h_nlen;
++};
++
++static int hn_job(struct hn_job_args *a)
++{
++ const unsigned int isdir = au_ftest_hnjob(a->flags, ISDIR);
++ int e;
++
++ /* reset xino */
++ if (au_ftest_hnjob(a->flags, XINO0) && a->inode)
++ hn_xino(a->inode, a->h_inode); /* ignore this error */
++
++ if (au_ftest_hnjob(a->flags, TRYXINO0)
++ && a->inode
++ && a->h_inode) {
++ inode_lock_shared_nested(a->h_inode, AuLsc_I_CHILD);
++ if (!a->h_inode->i_nlink
++ && !(a->h_inode->i_state & I_LINKABLE))
++ hn_xino(a->inode, a->h_inode); /* ignore this error */
++ inode_unlock_shared(a->h_inode);
+ }
+
-+ if ((ia->ia_valid & ATTR_SIZE)
-+ && ia->ia_size < i_size_read(inode)) {
-+ err = vmtruncate(inode, ia->ia_size);
-+ if (unlikely(err))
-+ goto out_unlock;
++ /* make the generation obsolete */
++ if (au_ftest_hnjob(a->flags, GEN)) {
++ e = -1;
++ if (a->inode)
++ e = hn_gen_by_inode(a->h_name, a->h_nlen, a->inode,
++ isdir);
++ if (e && a->dentry)
++ hn_gen_by_name(a->dentry, isdir);
++ /* ignore this error */
+ }
+
-+ if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
-+ ia->ia_valid &= ~ATTR_MODE;
++ /* make dir entries obsolete */
++ if (au_ftest_hnjob(a->flags, DIRENT) && a->inode) {
++ struct au_vdir *vdir;
+
-+ events = 0;
-+ if (a->hinotify) {
-+ events = vfsub_events_notify_change(ia);
-+ if (events) {
-+ if (a->isdir)
-+ vfsub_ign_hinode(&a->vargs, events,
-+ au_hi(inode, a->btgt));
-+ vfsub_ign_hinode(&a->vargs, events,
-+ au_pinned_hdir(&a->pin));
-+ }
++ vdir = au_ivdir(a->inode);
++ if (vdir)
++ vdir->vd_jiffy = 0;
++ /* IMustLock(a->inode); */
++ /* inode_inc_iversion(a->inode); */
+ }
-+ err = vfsub_notify_change(a->h_dentry, ia, &a->vargs);
-+ if (!err)
-+ au_cpup_attr_changeable(inode);
+
-+ out_unlock:
-+ mutex_unlock(&a->h_inode->i_mutex);
-+ au_unpin(&a->pin);
-+ di_write_unlock(dentry);
-+ out_si:
-+ if (file) {
-+ fi_write_unlock(file);
-+ ia->ia_file = file;
-+ ia->ia_valid |= ATTR_FILE;
-+ }
-+ si_read_unlock(sb);
-+ kfree(a);
-+ out:
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++ /* can do nothing but warn */
++ if (au_ftest_hnjob(a->flags, MNTPNT)
++ && a->dentry
++ && d_mountpoint(a->dentry))
++ pr_warn("mount-point %pd is removed or renamed\n", a->dentry);
++
++ return 0;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static int h_readlink(struct dentry *dentry, int bindex, char __user *buf,
-+ int bufsiz)
++static struct dentry *lookup_wlock_by_name(char *name, unsigned int nlen,
++ struct inode *dir)
+{
-+ struct super_block *sb;
-+ struct dentry *h_dentry;
++ struct dentry *dentry, *d, *parent;
++ struct qstr *dname;
+
-+ LKTRTrace("%.*s, b%d, %d\n", AuDLNPair(dentry), bindex, bufsiz);
++ parent = d_find_any_alias(dir);
++ if (!parent)
++ return NULL;
+
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (unlikely(/* !h_dentry
-+ || !h_dentry->d_inode
-+ || */
-+ !h_dentry->d_inode->i_op
-+ || !h_dentry->d_inode->i_op->readlink))
-+ return -EINVAL;
++ dentry = NULL;
++ spin_lock(&parent->d_lock);
++ list_for_each_entry(d, &parent->d_subdirs, d_child) {
++ /* AuDbg("%pd\n", d); */
++ spin_lock_nested(&d->d_lock, DENTRY_D_LOCK_NESTED);
++ dname = &d->d_name;
++ if (dname->len != nlen || memcmp(dname->name, name, nlen))
++ goto cont_unlock;
++ if (au_di(d))
++ au_digen_dec(d);
++ else
++ goto cont_unlock;
++ if (au_dcount(d) > 0) {
++ dentry = dget_dlock(d);
++ spin_unlock(&d->d_lock);
++ break;
++ }
+
-+ sb = dentry->d_sb;
-+ if (!au_test_ro(sb, bindex, dentry->d_inode)) {
-+ touch_atime(au_sbr_mnt(sb, bindex), h_dentry);
-+ au_update_fuse_h_inode(NULL, h_dentry); /*ignore*/
-+ fsstack_copy_attr_atime(dentry->d_inode, h_dentry->d_inode);
++cont_unlock:
++ spin_unlock(&d->d_lock);
+ }
-+ return h_dentry->d_inode->i_op->readlink(h_dentry, buf, bufsiz);
++ spin_unlock(&parent->d_lock);
++ dput(parent);
++
++ if (dentry)
++ di_write_lock_child(dentry);
++
++ return dentry;
+}
+
-+static int aufs_readlink(struct dentry *dentry, char __user *buf, int bufsiz)
++static struct inode *lookup_wlock_by_ino(struct super_block *sb,
++ aufs_bindex_t bindex, ino_t h_ino)
+{
++ struct inode *inode;
++ ino_t ino;
+ int err;
+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), bufsiz);
++ inode = NULL;
++ err = au_xino_read(sb, bindex, h_ino, &ino);
++ if (!err && ino)
++ inode = ilookup(sb, ino);
++ if (!inode)
++ goto out;
+
-+ aufs_read_lock(dentry, AuLock_IR);
-+ err = h_readlink(dentry, au_dbstart(dentry), buf, bufsiz);
-+ aufs_read_unlock(dentry, AuLock_IR);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++ if (unlikely(inode->i_ino == AUFS_ROOT_INO)) {
++ pr_warn("wrong root branch\n");
++ iput(inode);
++ inode = NULL;
++ goto out;
++ }
++
++ ii_write_lock_child(inode);
++
++out:
++ return inode;
+}
+
-+static void *aufs_follow_link(struct dentry *dentry, struct nameidata *nd)
++static void au_hn_bh(void *_args)
+{
++ struct au_hnotify_args *a = _args;
++ struct super_block *sb;
++ aufs_bindex_t bindex, bbot, bfound;
++ unsigned char xino, try_iput;
+ int err;
-+ char *buf;
-+ mm_segment_t old_fs;
++ struct inode *inode;
++ ino_t h_ino;
++ struct hn_job_args args;
++ struct dentry *dentry;
++ struct au_sbinfo *sbinfo;
+
-+ LKTRTrace("%.*s, nd %.*s\n",
-+ AuDLNPair(dentry), AuDLNPair(nd->path.dentry));
++ AuDebugOn(!_args);
++ AuDebugOn(!a->h_dir);
++ AuDebugOn(!a->dir);
++ AuDebugOn(!a->mask);
++ AuDbg("mask 0x%x, i%lu, hi%lu, hci%lu\n",
++ a->mask, a->dir->i_ino, a->h_dir->i_ino,
++ a->h_child_inode ? a->h_child_inode->i_ino : 0);
+
-+ err = -ENOMEM;
-+ buf = __getname();
-+ if (unlikely(!buf))
++ inode = NULL;
++ dentry = NULL;
++ /*
++ * do not lock a->dir->i_mutex here
++ * because of d_revalidate() may cause a deadlock.
++ */
++ sb = a->dir->i_sb;
++ AuDebugOn(!sb);
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!sbinfo);
++ si_write_lock(sb, AuLock_NOPLMW);
++
++ if (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);
++ for (bindex = au_ibtop(a->dir); bindex <= bbot; bindex++)
++ if (au_h_iptr(a->dir, bindex) == a->h_dir) {
++ bfound = bindex;
++ break;
++ }
++ ii_read_unlock(a->dir);
++ if (unlikely(bfound < 0))
+ goto out;
+
-+ aufs_read_lock(dentry, AuLock_IR);
-+ old_fs = get_fs();
-+ set_fs(KERNEL_DS);
-+ err = h_readlink(dentry, au_dbstart(dentry), (char __user *)buf,
-+ PATH_MAX);
-+ set_fs(old_fs);
-+ aufs_read_unlock(dentry, AuLock_IR);
++ xino = !!au_opt_test(au_mntflags(sb), XINO);
++ h_ino = 0;
++ if (a->h_child_inode)
++ h_ino = a->h_child_inode->i_ino;
+
-+ if (err >= 0) {
-+ buf[err] = 0;
-+ /* will be freed by put_link */
-+ nd_set_link(nd, buf);
-+ return NULL; /* success */
++ if (a->h_child_nlen
++ && (au_ftest_hnjob(a->flags[AuHn_CHILD], GEN)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], MNTPNT)))
++ dentry = lookup_wlock_by_name(a->h_child_name, a->h_child_nlen,
++ a->dir);
++ try_iput = 0;
++ if (dentry && d_really_is_positive(dentry))
++ inode = d_inode(dentry);
++ if (xino && !inode && h_ino
++ && (au_ftest_hnjob(a->flags[AuHn_CHILD], XINO0)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], TRYXINO0)
++ || au_ftest_hnjob(a->flags[AuHn_CHILD], GEN))) {
++ inode = lookup_wlock_by_ino(sb, bfound, h_ino);
++ try_iput = 1;
+ }
-+ __putname(buf);
+
-+ out:
-+ path_put(&nd->path);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return ERR_PTR(err);
-+}
++ args.flags = a->flags[AuHn_CHILD];
++ args.dentry = dentry;
++ args.inode = inode;
++ args.h_inode = a->h_child_inode;
++ args.dir = a->dir;
++ args.h_dir = a->h_dir;
++ args.h_name = a->h_child_name;
++ args.h_nlen = a->h_child_nlen;
++ err = hn_job(&args);
++ if (dentry) {
++ if (au_di(dentry))
++ di_write_unlock(dentry);
++ dput(dentry);
++ }
++ if (inode && try_iput) {
++ ii_write_unlock(inode);
++ iput(inode);
++ }
+
-+static void aufs_put_link(struct dentry *dentry, struct nameidata *nd,
-+ void *cookie)
-+{
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ __putname(nd_get_link(nd));
++ ii_write_lock_parent(a->dir);
++ args.flags = a->flags[AuHn_PARENT];
++ args.dentry = NULL;
++ args.inode = a->dir;
++ args.h_inode = a->h_dir;
++ args.dir = NULL;
++ args.h_dir = NULL;
++ args.h_name = NULL;
++ args.h_nlen = 0;
++ err = hn_job(&args);
++ ii_write_unlock(a->dir);
++
++out:
++ iput(a->h_child_inode);
++ iput(a->h_dir);
++ iput(a->dir);
++ si_write_unlock(sb);
++ au_nwt_done(&sbinfo->si_nowait);
++ au_kfree_rcu(a);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static void aufs_truncate_range(struct inode *inode, loff_t start, loff_t end)
++int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
++ const struct qstr *h_child_qstr, struct inode *h_child_inode)
+{
-+ AuUnsupport();
-+}
++ int err, len;
++ unsigned int flags[AuHnLast], f;
++ unsigned char isdir, isroot, wh;
++ struct inode *dir;
++ struct au_hnotify_args *args;
++ char *p, *h_child_name;
+
-+/* ---------------------------------------------------------------------- */
++ err = 0;
++ AuDebugOn(!hnotify || !hnotify->hn_aufs_inode);
++ dir = igrab(hnotify->hn_aufs_inode);
++ if (!dir)
++ goto out;
+
-+struct inode_operations aufs_symlink_iop = {
-+ .permission = aufs_permission,
-+ .setattr = aufs_setattr,
-+#ifdef CONFIG_AUFS_GETATTR
-+ .getattr = aufs_getattr,
-+#endif
++ isroot = (dir->i_ino == AUFS_ROOT_INO);
++ wh = 0;
++ h_child_name = (void *)h_child_qstr->name;
++ len = h_child_qstr->len;
++ if (h_child_name) {
++ if (len > AUFS_WH_PFX_LEN
++ && !memcmp(h_child_name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
++ h_child_name += AUFS_WH_PFX_LEN;
++ len -= AUFS_WH_PFX_LEN;
++ wh = 1;
++ }
++ }
+
-+ .readlink = aufs_readlink,
-+ .follow_link = aufs_follow_link,
-+ .put_link = aufs_put_link
-+};
++ isdir = 0;
++ if (h_child_inode)
++ isdir = !!S_ISDIR(h_child_inode->i_mode);
++ flags[AuHn_PARENT] = AuHnJob_ISDIR;
++ flags[AuHn_CHILD] = 0;
++ if (isdir)
++ flags[AuHn_CHILD] = AuHnJob_ISDIR;
++ au_fset_hnjob(flags[AuHn_PARENT], DIRENT);
++ au_fset_hnjob(flags[AuHn_CHILD], GEN);
++ switch (mask & ALL_FSNOTIFY_DIRENT_EVENTS) {
++ case FS_MOVED_FROM:
++ case FS_MOVED_TO:
++ au_fset_hnjob(flags[AuHn_CHILD], XINO0);
++ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
++ fallthrough;
++ case FS_CREATE:
++ AuDebugOn(!h_child_name);
++ break;
+
-+struct inode_operations aufs_dir_iop = {
-+ .create = aufs_create,
-+ .lookup = aufs_lookup,
-+ .link = aufs_link,
-+ .unlink = aufs_unlink,
-+ .symlink = aufs_symlink,
-+ .mkdir = aufs_mkdir,
-+ .rmdir = aufs_rmdir,
-+ .mknod = aufs_mknod,
-+ .rename = aufs_rename,
-+
-+ .permission = aufs_permission,
-+ .setattr = aufs_setattr,
-+#ifdef CONFIG_AUFS_GETATTR
-+ .getattr = aufs_getattr,
-+#endif
++ case FS_DELETE:
++ /*
++ * aufs never be able to get this child inode.
++ * revalidation should be in d_revalidate()
++ * by checking i_nlink, i_generation or d_unhashed().
++ */
++ AuDebugOn(!h_child_name);
++ au_fset_hnjob(flags[AuHn_CHILD], TRYXINO0);
++ au_fset_hnjob(flags[AuHn_CHILD], MNTPNT);
++ break;
+
-+#if 0 /* reserved for future use */
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
-+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr
-+#endif
-+};
++ default:
++ AuDebugOn(1);
++ }
+
-+struct inode_operations aufs_iop = {
-+ .permission = aufs_permission,
-+ .setattr = aufs_setattr,
-+#ifdef CONFIG_AUFS_GETATTR
-+ .getattr = aufs_getattr,
-+#endif
++ if (wh)
++ h_child_inode = NULL;
+
-+#if 0 /* reserved for future use */
-+ .setxattr = aufs_setxattr,
-+ .getxattr = aufs_getxattr,
-+ .listxattr = aufs_listxattr,
-+ .removexattr = aufs_removexattr,
-+#endif
++ err = -ENOMEM;
++ /* iput() and kfree() will be called in au_hnotify() */
++ args = kmalloc(sizeof(*args) + len + 1, GFP_NOFS);
++ if (unlikely(!args)) {
++ AuErr1("no memory\n");
++ iput(dir);
++ goto out;
++ }
++ args->flags[AuHn_PARENT] = flags[AuHn_PARENT];
++ args->flags[AuHn_CHILD] = flags[AuHn_CHILD];
++ args->mask = mask;
++ args->dir = dir;
++ args->h_dir = igrab(h_dir);
++ if (h_child_inode)
++ h_child_inode = igrab(h_child_inode); /* can be NULL */
++ args->h_child_inode = h_child_inode;
++ args->h_child_nlen = len;
++ if (len) {
++ p = (void *)args;
++ p += sizeof(*args);
++ memcpy(p, h_child_name, len);
++ p[len] = 0;
++ }
+
-+ .truncate_range = aufs_truncate_range
-+};
-diff -urN linux/fs/aufs/i_op_del.c linux-aufs/fs/aufs/i_op_del.c
---- linux/fs/aufs/i_op_del.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/i_op_del.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,566 @@
++ /* NFS fires the event for silly-renamed one from kworker */
++ f = 0;
++ if (!dir->i_nlink
++ || (au_test_nfs(h_dir->i_sb) && (mask & FS_DELETE)))
++ f = AuWkq_NEST;
++ err = au_wkq_nowait(au_hn_bh, args, dir->i_sb, f);
++ if (unlikely(err)) {
++ pr_err("wkq %d\n", err);
++ iput(args->h_child_inode);
++ iput(args->h_dir);
++ iput(args->dir);
++ au_kfree_rcu(args);
++ }
++
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm)
++{
++ int err;
++
++ AuDebugOn(!(udba & AuOptMask_UDBA));
++
++ err = 0;
++ if (au_hnotify_op.reset_br)
++ err = au_hnotify_op.reset_br(udba, br, perm);
++
++ return err;
++}
++
++int au_hnotify_init_br(struct au_branch *br, int perm)
++{
++ int err;
++
++ err = 0;
++ if (au_hnotify_op.init_br)
++ err = au_hnotify_op.init_br(br, perm);
++
++ return err;
++}
++
++void au_hnotify_fin_br(struct au_branch *br)
++{
++ if (au_hnotify_op.fin_br)
++ au_hnotify_op.fin_br(br);
++}
++
++static void au_hn_destroy_cache(void)
++{
++ kmem_cache_destroy(au_cache[AuCache_HNOTIFY]);
++ au_cache[AuCache_HNOTIFY] = NULL;
++}
++
++int __init au_hnotify_init(void)
++{
++ int err;
++
++ err = -ENOMEM;
++ au_cache[AuCache_HNOTIFY] = AuCache(au_hnotify);
++ if (au_cache[AuCache_HNOTIFY]) {
++ err = 0;
++ if (au_hnotify_op.init)
++ err = au_hnotify_op.init();
++ if (unlikely(err))
++ au_hn_destroy_cache();
++ }
++ AuTraceErr(err);
++ return err;
++}
++
++void au_hnotify_fin(void)
++{
++ if (au_hnotify_op.fin)
++ au_hnotify_op.fin();
++
++ /* cf. au_cache_fin() */
++ 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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,286 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode operations (del entry)
-+ *
-+ * $Id$
++ * inode private data
+ */
+
+#include "aufs.h"
+
-+/* returns,
-+ * 0: wh is unnecessary
-+ * plus: wh is necessary
-+ * minus: error
-+ */
-+int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup)
++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex)
+{
-+ int need_wh, err;
-+ aufs_bindex_t bstart;
-+ struct dentry *h_dentry;
-+ struct super_block *sb;
++ struct inode *h_inode;
++ struct au_hinode *hinode;
+
-+ LKTRTrace("%.*s, isdir %d, *bcpup %d\n",
-+ AuDLNPair(dentry), isdir, *bcpup);
-+ sb = dentry->d_sb;
++ IiMustAnyLock(inode);
+
-+ bstart = au_dbstart(dentry);
-+ LKTRTrace("bcpup %d, bstart %d\n", *bcpup, bstart);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ if (*bcpup < 0) {
-+ *bcpup = bstart;
-+ if (au_test_ro(sb, bstart, dentry->d_inode)) {
-+ err = AuWbrCopyup(au_sbi(sb), dentry);
-+ *bcpup = err;
-+ if (unlikely(err < 0))
-+ goto out;
-+ }
-+ } else
-+ AuDebugOn(bstart < *bcpup
-+ || au_test_ro(sb, *bcpup, dentry->d_inode));
-+ LKTRTrace("bcpup %d, bstart %d\n", *bcpup, bstart);
++ hinode = au_hinode(au_ii(inode), bindex);
++ h_inode = hinode->hi_inode;
++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
++ return h_inode;
++}
+
-+ if (*bcpup != bstart) {
-+ err = au_cpup_dirs(dentry, *bcpup);
-+ if (unlikely(err))
-+ goto out;
-+ need_wh = 1;
-+ } else {
-+ aufs_bindex_t old_bend, new_bend, bdiropq = -1;
-+ old_bend = au_dbend(dentry);
-+ if (isdir) {
-+ bdiropq = au_dbdiropq(dentry);
-+ au_set_dbdiropq(dentry, -1);
-+ }
-+ need_wh = au_lkup_dentry(dentry, bstart + 1, /*type*/0,
-+ /*nd*/NULL);
-+ err = need_wh;
-+ if (isdir)
-+ au_set_dbdiropq(dentry, bdiropq);
-+ if (unlikely(err < 0))
-+ goto out;
-+ new_bend = au_dbend(dentry);
-+ if (!need_wh && old_bend != new_bend) {
-+ au_set_h_dptr(dentry, new_bend, NULL);
-+ au_set_dbend(dentry, old_bend);
-+#if 0 /* todo: remove this? */
-+ } else if (!au_h_dptr(dentry, new_bend)->d_inode) {
-+ LKTRTrace("negative\n");
-+ au_set_h_dptr(dentry, new_bend, NULL);
-+ au_set_dbend(dentry, old_bend);
-+ need_wh = 0;
-+#endif
-+ }
-+ }
-+ LKTRTrace("need_wh %d\n", need_wh);
-+ err = need_wh;
++/* todo: hard/soft set? */
++void au_hiput(struct au_hinode *hinode)
++{
++ au_hn_free(hinode);
++ dput(hinode->hi_whdentry);
++ iput(hinode->hi_inode);
++}
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++unsigned int au_hi_flags(struct inode *inode, int isdir)
++{
++ unsigned int flags;
++ const unsigned int mnt_flags = au_mntflags(inode->i_sb);
++
++ flags = 0;
++ if (au_opt_test(mnt_flags, XINO))
++ au_fset_hi(flags, XINO);
++ if (isdir && au_opt_test(mnt_flags, UDBA_HNOTIFY))
++ au_fset_hi(flags, HNOTIFY);
++ return flags;
+}
+
-+/*
-+ * simple tests for the removal inode operations.
-+ * following the checks in vfs, plus the parent-child relationship.
-+ */
-+int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, int isdir, struct au_ndx *ndx)
++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode, unsigned int flags)
+{
-+ int err, exist;
-+ struct super_block *sb;
-+ struct dentry *h_dentry;
-+ struct inode *h_inode;
-+ umode_t h_mode;
++ struct au_hinode *hinode;
++ struct inode *hi;
++ struct au_iinfo *iinfo = au_ii(inode);
+
-+ LKTRTrace("%.*s/%.*s, b%d, dir %d\n",
-+ AuDLNPair(h_parent), AuDLNPair(dentry), bindex, isdir);
++ IiMustWriteLock(inode);
+
-+ sb = dentry->d_sb;
-+ exist = !!dentry->d_inode;
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ h_inode = h_dentry->d_inode;
-+ if (exist) {
-+ err = -ENOENT;
-+ if (unlikely(!h_inode || !h_inode->i_nlink))
-+ goto out;
++ hinode = au_hinode(iinfo, bindex);
++ hi = hinode->hi_inode;
++ AuDebugOn(h_inode && atomic_read(&h_inode->i_count) <= 0);
+
-+ h_mode = h_inode->i_mode;
-+ if (!isdir) {
-+ err = -EISDIR;
-+ if (unlikely(S_ISDIR(h_mode)))
-+ goto out;
-+ } else if (unlikely(!S_ISDIR(h_mode))) {
-+ err = -ENOTDIR;
-+ goto out;
++ if (hi)
++ au_hiput(hinode);
++ hinode->hi_inode = h_inode;
++ if (h_inode) {
++ int err;
++ struct super_block *sb = inode->i_sb;
++ struct au_branch *br;
++
++ AuDebugOn(inode->i_mode
++ && (h_inode->i_mode & S_IFMT)
++ != (inode->i_mode & S_IFMT));
++ if (bindex == iinfo->ii_btop)
++ au_cpup_igen(inode, h_inode);
++ br = au_sbr(sb, bindex);
++ hinode->hi_id = br->br_id;
++ if (au_ftest_hi(flags, XINO)) {
++ err = au_xino_write(sb, bindex, h_inode->i_ino,
++ inode->i_ino);
++ if (unlikely(err))
++ AuIOErr1("failed au_xino_write() %d\n", err);
++ }
++
++ if (au_ftest_hi(flags, HNOTIFY)
++ && au_br_hnotifyable(br->br_perm)) {
++ err = au_hn_alloc(hinode, inode);
++ if (unlikely(err))
++ AuIOErr1("au_hn_alloc() %d\n", err);
+ }
-+ } else {
-+ /* rename(2) case */
-+ err = -EIO;
-+ if (unlikely(h_inode))
-+ goto out;
+ }
++}
+
-+ err = -ENOENT;
-+ /* expected parent dir is locked */
-+ if (unlikely(h_parent != h_dentry->d_parent))
-+ goto out;
-+ err = 0;
++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_wh)
++{
++ struct au_hinode *hinode;
+
-+ /*
-+ * some filesystem may unlink a dir and corrupt its consistency.
-+ * so let's try heavy test.
-+ */
-+ if (1 /*au_opt_test(au_mntflags(sb), UDBA_INOTIFY)*/) {
-+ struct dentry *h_latest;
-+ struct qstr *qstr = &dentry->d_name;
++ IiMustWriteLock(inode);
+
-+ err = -EACCES;
-+ if (unlikely(au_test_h_perm(h_parent->d_inode,
-+ MAY_EXEC | MAY_WRITE,
-+ au_ftest_ndx(ndx->flags, DLGT))))
-+ goto out;
++ hinode = au_hinode(au_ii(inode), bindex);
++ AuDebugOn(hinode->hi_whdentry);
++ hinode->hi_whdentry = h_wh;
++}
+
-+ h_latest = au_sio_lkup_one(qstr->name, h_parent, qstr->len,
-+ ndx);
-+ err = -EIO;
-+ if (IS_ERR(h_latest))
-+ goto out;
-+ dput(h_latest);
-+ if (h_latest == h_dentry)
-+ err = 0;
-+ }
++void au_update_iigen(struct inode *inode, int half)
++{
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
++ unsigned int sigen;
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ sigen = au_sigen(inode->i_sb);
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ iigen->ig_generation = sigen;
++ if (half)
++ au_ig_fset(iigen->ig_flags, HALF_REFRESHED);
++ else
++ au_ig_fclr(iigen->ig_flags, HALF_REFRESHED);
++ spin_unlock(&iigen->ig_spin);
+}
+
-+static struct dentry *
-+lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup,
-+ struct au_dtime *dt, struct au_pin *pin)
++/* it may be called at remount time, too */
++void au_update_ibrange(struct inode *inode, int do_put_zero)
+{
-+ struct dentry *wh_dentry, *h_parent;
-+ struct super_block *sb;
-+ int err, need_wh;
-+ unsigned int mnt_flags;
-+ unsigned char pin_flags;
-+ aufs_bindex_t bcpup;
-+ struct au_ndx ndx;
++ struct au_iinfo *iinfo;
++ aufs_bindex_t bindex, bbot;
+
-+ LKTRTrace("%.*s, isdir %d\n", AuDLNPair(dentry), isdir);
++ AuDebugOn(au_is_bad_inode(inode));
++ IiMustWriteLock(inode);
+
-+ need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup);
-+ err = need_wh;
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err < 0))
-+ goto out;
++ iinfo = au_ii(inode);
++ if (do_put_zero && iinfo->ii_btop >= 0) {
++ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
++ bindex++) {
++ struct inode *h_i;
+
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ bcpup = *rbcpup;
-+ pin_flags = AuPin_DI_LOCKED | AuPin_MNT_WRITE;
-+ if (au_opt_test(mnt_flags, UDBA_INOTIFY))
-+ au_fset_pin(pin_flags, DO_GPARENT);
-+ err = au_pin(pin, dentry, bcpup, pin_flags);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out;
-+ h_parent = au_pinned_h_parent(pin);
-+ if (!au_opt_test(mnt_flags, UDBA_NONE) && au_dbstart(dentry) == bcpup) {
-+ ndx.nfsmnt = au_nfsmnt(sb, bcpup);
-+ ndx.flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nd = NULL;
-+ /* ndx.br = au_sbr(sb, bcpup); */
-+ /* ndx.nd_file = NULL; */
-+ err = au_may_del(dentry, bcpup, h_parent, isdir, &ndx);
-+ wh_dentry = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out_unpin;
++ h_i = au_hinode(iinfo, bindex)->hi_inode;
++ if (h_i
++ && !h_i->i_nlink
++ && !(h_i->i_state & I_LINKABLE))
++ au_set_h_iptr(inode, bindex, NULL, 0);
++ }
+ }
+
-+ au_dtime_store(dt, au_pinned_parent(pin), h_parent, au_pinned_hdir(pin),
-+ au_pinned_hgdir(pin));
-+ wh_dentry = NULL;
-+ if (!need_wh)
-+ goto out; /* success, no need to create whiteout */
++ iinfo->ii_btop = -1;
++ iinfo->ii_bbot = -1;
++ bbot = au_sbbot(inode->i_sb);
++ for (bindex = 0; bindex <= bbot; bindex++)
++ if (au_hinode(iinfo, bindex)->hi_inode) {
++ iinfo->ii_btop = bindex;
++ break;
++ }
++ if (iinfo->ii_btop >= 0)
++ for (bindex = bbot; bindex >= iinfo->ii_btop; bindex--)
++ if (au_hinode(iinfo, bindex)->hi_inode) {
++ iinfo->ii_bbot = bindex;
++ break;
++ }
++ AuDebugOn(iinfo->ii_btop > iinfo->ii_bbot);
++}
+
-+ ndx.nfsmnt = au_nfsmnt(sb, bcpup);
-+ ndx.flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nd = NULL;
-+ /* ndx.br = NULL; */
-+ wh_dentry = au_wh_create(dentry, bcpup, h_parent, &ndx);
-+ if (!IS_ERR(wh_dentry))
-+ goto out; /* success */
-+ /* returns with the parent is locked and wh_dentry is DGETed */
++/* ---------------------------------------------------------------------- */
+
-+ out_unpin:
-+ au_unpin(pin);
-+ out:
-+ AuTraceErrPtr(wh_dentry);
-+ return wh_dentry;
++void au_icntnr_init_once(void *_c)
++{
++ struct au_icntnr *c = _c;
++ struct au_iinfo *iinfo = &c->iinfo;
++
++ spin_lock_init(&iinfo->ii_generation.ig_spin);
++ au_rw_init(&iinfo->ii_rwsem);
++ inode_init_once(&c->vfs_inode);
+}
+
-+static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct au_nhash *whlist, struct inode *dir)
++void au_hinode_init(struct au_hinode *hinode)
+{
-+ int rmdir_later, err;
-+ struct dentry *h_dentry;
-+ struct inode *inode, *h_inode;
++ hinode->hi_inode = NULL;
++ hinode->hi_id = -1;
++ au_hn_init(hinode);
++ hinode->hi_whdentry = NULL;
++}
++
++int au_iinfo_init(struct inode *inode)
++{
++ struct au_iinfo *iinfo;
+ struct super_block *sb;
++ struct au_hinode *hi;
++ int nbr, i;
+
-+ LKTRTrace("%.*s, b%d\n", AuDLNPair(dentry), bindex);
-+
-+ inode = NULL;
-+ h_inode = NULL;
-+ sb = dentry->d_sb;
-+ if (au_opt_test(au_mntflags(sb), UDBA_INOTIFY)) {
-+ inode = dentry->d_inode;
-+ h_inode = au_h_iptr(inode, bindex);
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ }
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ err = au_whtmp_ren(dir, bindex, h_dentry);
-+ if (inode) {
-+ /* todo: bad approach? */
-+ if (!err)
-+ au_hin_suspend(au_hi(inode, bindex));
-+ mutex_unlock(&h_inode->i_mutex);
-+ }
-+ if (unlikely(err))
-+ goto out;
-+
-+ if (!au_test_nfs(h_dentry->d_sb)) {
-+ const int dirwh = au_sbi(sb)->si_dirwh;
-+ rmdir_later = (dirwh <= 1);
-+ if (!rmdir_later)
-+ rmdir_later = au_nhash_test_longer_wh(whlist, bindex,
-+ dirwh);
-+ if (rmdir_later)
-+ return rmdir_later;
-+ }
-+
-+ err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist);
-+ if (unlikely(err)) {
-+ AuIOErr("rmdir %.*s, b%d failed, %d. ignored\n",
-+ AuDLNPair(h_dentry), bindex, err);
-+ /* we do not revert the inotify watch */
-+ err = 0;
-+ }
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+static void epilog(struct inode *dir, struct dentry *dentry,
-+ aufs_bindex_t bindex)
-+{
-+ /* todo: unnecessary? */
-+ d_drop(dentry);
-+ dentry->d_inode->i_ctime = dir->i_ctime;
-+
-+ if (atomic_read(&dentry->d_count) == 1) {
-+ au_set_h_dptr(dentry, au_dbstart(dentry), NULL);
-+ au_update_dbstart(dentry);
-+ }
-+ if (au_ibstart(dir) == bindex)
-+ au_cpup_attr_timesizes(dir);
-+ dir->i_version++;
-+}
-+
-+/* revert flags */
-+#define AuRev_DLGT 1
-+#define au_ftest_rev(flags, name) ((flags) & AuRev_##name)
-+#define au_fset_rev(flags, name) { (flags) |= AuRev_##name; }
-+#define au_fclr_rev(flags, name) { (flags) &= ~AuRev_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuRev_DLGT
-+#define AuRev_DLGT 0
-+#endif
++ sb = inode->i_sb;
++ iinfo = &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
++ nbr = au_sbbot(sb) + 1;
++ if (unlikely(nbr <= 0))
++ nbr = 1;
++ hi = kmalloc_array(nbr, sizeof(*iinfo->ii_hinode), GFP_NOFS);
++ if (hi) {
++ au_lcnt_inc(&au_sbi(sb)->si_ninodes);
+
-+static int do_revert(int err, struct inode *dir, aufs_bindex_t bwh,
-+ struct dentry *wh_dentry, struct dentry *dentry,
-+ struct au_dtime *dt, unsigned int flags)
-+{
-+ int rerr;
++ iinfo->ii_hinode = hi;
++ for (i = 0; i < nbr; i++, hi++)
++ au_hinode_init(hi);
+
-+ rerr = au_wh_unlink_dentry(au_hi(dir, bwh), wh_dentry, dentry,
-+ au_ftest_rev(flags, DLGT));
-+ if (!rerr) {
-+ au_set_dbwh(dentry, bwh);
-+ au_dtime_revert(dt);
++ iinfo->ii_generation.ig_generation = au_sigen(sb);
++ iinfo->ii_btop = -1;
++ iinfo->ii_bbot = -1;
++ iinfo->ii_vdir = NULL;
+ return 0;
+ }
-+
-+ AuIOErr("%.*s reverting whiteout failed(%d, %d)\n",
-+ AuDLNPair(dentry), err, rerr);
-+ return -EIO;
++ return -ENOMEM;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+int aufs_unlink(struct inode *dir, struct dentry *dentry)
++int au_hinode_realloc(struct au_iinfo *iinfo, int nbr, int may_shrink)
+{
-+ int err;
-+ struct inode *inode, *h_dir;
-+ struct dentry *parent, *wh_dentry, *h_dentry;
-+ struct au_dtime dt;
-+ aufs_bindex_t bwh, bindex, bstart;
-+ unsigned char dlgt;
-+ struct super_block *sb;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_pin pin;
-+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
-+ inode = dentry->d_inode;
-+ if (unlikely(!inode))
-+ return -ENOENT; /* possible? */
-+ IMustLock(inode);
-+
-+ aufs_read_lock(dentry, AuLock_DW);
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_write_lock_parent(parent);
-+
-+ bstart = au_dbstart(dentry);
-+ bwh = au_dbwh(dentry);
-+ bindex = -1;
-+ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &dt, &pin);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out;
++ int err, i;
++ struct au_hinode *hip;
+
-+ sb = dir->i_sb;
-+ dlgt = !!au_test_dlgt(au_mntflags(sb));
-+ AuDebugOn(au_dbstart(dentry) != bstart);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ dget(h_dentry);
++ AuRwMustWriteLock(&iinfo->ii_rwsem);
+
-+ if (bindex == bstart) {
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, au_pinned_hdir(&pin));
-+ h_dir = au_pinned_h_dir(&pin);
-+ err = vfsub_unlink(h_dir, h_dentry, &vargs);
-+ } else {
-+ /* dir inode is locked */
-+ AuDebugOn(!wh_dentry
-+ || wh_dentry->d_parent != au_h_dptr(parent, bindex));
-+ h_dir = wh_dentry->d_parent->d_inode;
-+ IMustLock(h_dir);
++ err = -ENOMEM;
++ hip = au_krealloc(iinfo->ii_hinode, sizeof(*hip) * nbr, GFP_NOFS,
++ may_shrink);
++ if (hip) {
++ iinfo->ii_hinode = hip;
++ i = iinfo->ii_bbot + 1;
++ hip += i;
++ for (; i < nbr; i++, hip++)
++ au_hinode_init(hip);
+ err = 0;
+ }
+
-+ if (!err) {
-+ drop_nlink(inode);
-+#if 0 /* todo: update plink? */
-+ if (!inode->i_nlink
-+ && au_opt_test(p->a.mnt_flags, PLINK)
-+ && au_plink_test(sb, inode)
-+ /* && atomic_read(&inode->i_count) == 2) */) {
-+ au_debug_on();
-+ DbgInode(inode);
-+ au_debug_off();
-+ }
-+#endif
-+ /*
-+ * although this is not a dir,
-+ * set it here since we need to detect the dead inode.
-+ */
-+ if (!inode->i_nlink)
-+ inode->i_flags |= S_DEAD;
-+ epilog(dir, dentry, bindex);
-+
-+ /* update target timestamps */
-+ if (bindex == bstart) {
-+ au_update_fuse_h_inode(NULL, h_dentry); /*ignore*/
-+ inode->i_ctime = h_dentry->d_inode->i_ctime;
-+ } else
-+ /* todo: this timestamp may be reverted later */
-+ inode->i_ctime = h_dir->i_ctime;
-+ goto out_unlock; /* success */
-+ }
-+
-+ /* revert */
-+ if (wh_dentry) {
-+ int rerr;
-+ unsigned int rev_flags;
-+
-+ rev_flags = 0;
-+ if (dlgt)
-+ au_fset_rev(rev_flags, DLGT);
-+ rerr = do_revert(err, dir, bwh, wh_dentry, dentry, &dt,
-+ rev_flags);
-+ if (rerr)
-+ err = rerr;
-+ }
-+
-+ out_unlock:
-+ au_unpin(&pin);
-+ dput(wh_dentry);
-+ dput(h_dentry);
-+ out:
-+ di_write_unlock(parent);
-+ aufs_read_unlock(dentry, AuLock_DW);
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
+ return err;
+}
+
-+int aufs_rmdir(struct inode *dir, struct dentry *dentry)
++void au_iinfo_fin(struct inode *inode)
+{
-+ int err, rmdir_later;
-+ struct inode *inode;
-+ struct dentry *parent, *wh_dentry, *h_dentry;
-+ struct au_dtime dt;
-+ aufs_bindex_t bwh, bindex, bstart;
-+ struct au_whtmp_rmdir_args *args;
-+ struct au_nhash *whlist;
++ struct au_iinfo *iinfo;
++ struct au_hinode *hi;
+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ struct au_pin pin;
-+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
-+ inode = dentry->d_inode;
-+ err = -ENOENT; /* possible? */
-+ if (unlikely(!inode))
-+ goto out;
-+ IMustLock(inode);
-+
-+ whlist = au_nhash_new(GFP_NOFS);
-+ err = PTR_ERR(whlist);
-+ if (IS_ERR(whlist))
-+ goto out;
++ aufs_bindex_t bindex, bbot;
++ const unsigned char unlinked = !inode->i_nlink;
+
-+ err = -ENOMEM;
-+ args = kmalloc(sizeof(*args), GFP_NOFS);
-+ if (unlikely(!args))
-+ goto out_whlist;
-+
-+ aufs_read_lock(dentry, AuLock_DW);
-+ parent = dentry->d_parent; /* dir inode is locked */
-+ di_write_lock_parent(parent);
-+ err = au_test_empty(dentry, whlist);
-+ if (unlikely(err))
-+ goto out_args;
-+
-+ bstart = au_dbstart(dentry);
-+ bwh = au_dbwh(dentry);
-+ bindex = -1;
-+ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &dt, &pin);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_args;
-+
-+ AuDebugOn(au_dbstart(dentry) != bstart);
-+ h_dentry = au_h_dptr(dentry, bstart);
-+ dget(h_dentry);
++ AuDebugOn(au_is_bad_inode(inode));
+
-+ rmdir_later = 0;
-+ if (bindex == bstart) {
-+ err = renwh_and_rmdir(dentry, bstart, whlist, dir);
-+ if (err > 0) {
-+ rmdir_later = err;
-+ err = 0;
-+ }
-+ } else {
-+ /* dir inode is locked */
-+ AuDebugOn(!wh_dentry
-+ || wh_dentry->d_parent != au_h_dptr(parent, bindex));
-+ IMustLock(wh_dentry->d_parent->d_inode);
-+ err = 0;
++ sb = inode->i_sb;
++ au_lcnt_dec(&au_sbi(sb)->si_ninodes);
++ if (si_pid_test(sb))
++ au_xino_delete_inode(inode, unlinked);
++ else {
++ /*
++ * it is safe to hide the dependency between sbinfo and
++ * sb->s_umount.
++ */
++ lockdep_off();
++ si_noflush_read_lock(sb);
++ au_xino_delete_inode(inode, unlinked);
++ si_read_unlock(sb);
++ lockdep_on();
+ }
+
-+ sb = dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ if (!err) {
-+ if (au_opt_test(mnt_flags, UDBA_INOTIFY) && rmdir_later)
-+ au_reset_hinotify(inode, /*flags*/0);
-+ clear_nlink(inode);
-+ inode->i_flags |= S_DEAD;
-+ au_set_dbdiropq(dentry, -1);
-+ epilog(dir, dentry, bindex);
++ iinfo = au_ii(inode);
++ if (iinfo->ii_vdir)
++ au_vdir_free(iinfo->ii_vdir);
+
-+ if (rmdir_later) {
-+ au_whtmp_kick_rmdir(dir, bstart, h_dentry, whlist,
-+ args);
-+ args = NULL;
++ bindex = iinfo->ii_btop;
++ if (bindex >= 0) {
++ hi = au_hinode(iinfo, bindex);
++ bbot = iinfo->ii_bbot;
++ while (bindex++ <= bbot) {
++ if (hi->hi_inode)
++ au_hiput(hi);
++ hi++;
+ }
-+
-+ goto out_unlock; /* success */
-+ }
-+
-+ /* revert */
-+ LKTRLabel(revert);
-+ if (wh_dentry) {
-+ int rerr;
-+ unsigned int rev_flags;
-+
-+ rev_flags = 0;
-+ if (au_test_dlgt(mnt_flags))
-+ au_fset_rev(rev_flags, DLGT);
-+ rerr = do_revert(err, dir, bwh, wh_dentry, dentry, &dt,
-+ rev_flags);
-+ if (rerr)
-+ err = rerr;
+ }
-+
-+ out_unlock:
-+ au_unpin(&pin);
-+ dput(wh_dentry);
-+ dput(h_dentry);
-+ out_args:
-+ di_write_unlock(parent);
-+ aufs_read_unlock(dentry, AuLock_DW);
-+ kfree(args);
-+ out_whlist:
-+ au_nhash_del(whlist);
-+ out:
-+ AuTraceErr(err);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++ au_kfree_rcu(iinfo->ii_hinode);
++ AuRwDestroy(&iinfo->ii_rwsem);
+}
-diff -urN linux/fs/aufs/i_op_ren.c linux-aufs/fs/aufs/i_op_ren.c
---- linux/fs/aufs/i_op_ren.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/i_op_ren.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1261 @@
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,531 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * inode operation (rename entry)
-+ * todo: this is crazy monster
-+ *
-+ * $Id$
++ * inode functions
+ */
+
++#include <linux/iversion.h>
+#include "aufs.h"
+
-+enum { SRC, DST };
++struct inode *au_igrab(struct inode *inode)
++{
++ if (inode) {
++ AuDebugOn(!atomic_read(&inode->i_count));
++ ihold(inode);
++ }
++ return inode;
++}
+
-+#define AuRen_ISDIR 1
-+#define AuRen_ISSAMEDIR (1 << 1)
-+#define AuRen_WHSRC (1 << 2)
-+#define AuRen_WHDST (1 << 3)
-+#define AuRen_DLGT (1 << 4)
-+#define AuRen_VFSLOCK (1 << 5)
-+#define AuRen_PINNED (1 << 6)
-+#define AuRen_MNT_WRITE (1 << 7)
-+#define au_ftest_ren(flags, name) ((flags) & AuRen_##name)
-+#define au_fset_ren(flags, name) { (flags) |= AuRen_##name; }
-+#define au_fclr_ren(flags, name) { (flags) &= ~AuRen_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuRen_DLGT
-+#define AuRen_DLGT 0
-+#endif
++static void au_refresh_hinode_attr(struct inode *inode, int do_version)
++{
++ au_cpup_attr_all(inode, /*force*/0);
++ au_update_iigen(inode, /*half*/1);
++ if (do_version)
++ inode_inc_iversion(inode);
++}
+
-+struct au_ren_args {
-+ /* original args */
-+ struct dentry *src_dentry, *dentry;
-+ struct inode *src_dir, *dir;
-+
-+ struct dentry *h_dentry[2], *h_parent[2], *h_trap, *h_locked[2];
-+ /* todo: remove them */
-+ struct dentry *parent[2], *gparent[2];
-+ struct au_pin pin[2];
-+ struct au_nhash whlist;
-+ aufs_bindex_t btgt, bstart[2];
-+ /* do_rename() only */
-+ unsigned char need_diropq, bycpup;
++static int au_ii_refresh(struct inode *inode, int *update)
++{
++ int err, e, nbr;
++ umode_t type;
++ aufs_bindex_t bindex, new_bindex;
+ struct super_block *sb;
-+ unsigned int flags;
-+ unsigned int mnt_flags;
-+ struct au_ndx ndx;
++ struct au_iinfo *iinfo;
++ struct au_hinode *p, *q, tmp;
+
-+ /* do_rename() only */
-+#ifdef CONFIG_AUFS_BR_NFS
-+ struct au_hin_ignore ign[3];
-+#else
-+ struct au_hin_ignore ign[2];
-+#endif
-+ struct vfsub_args vargs;
-+ struct au_whtmp_rmdir_args *thargs;
-+ struct dentry *wh_dentry[2], *h_dst, *h_src;
-+};
++ AuDebugOn(au_is_bad_inode(inode));
++ IiMustWriteLock(inode);
+
-+/* ---------------------------------------------------------------------- */
++ *update = 0;
++ sb = inode->i_sb;
++ nbr = au_sbbot(sb) + 1;
++ type = inode->i_mode & S_IFMT;
++ iinfo = au_ii(inode);
++ err = au_hinode_realloc(iinfo, nbr, /*may_shrink*/0);
++ if (unlikely(err))
++ goto out;
+
-+#define RevertFailure(fmt, args...) do { \
-+ AuIOErrWhck("revert failure: " fmt " (%d, %d)\n", \
-+ ##args, err, rerr); \
-+ err = -EIO; \
-+ } while (0)
++ AuDebugOn(iinfo->ii_btop < 0);
++ p = au_hinode(iinfo, iinfo->ii_btop);
++ for (bindex = iinfo->ii_btop; bindex <= iinfo->ii_bbot;
++ bindex++, p++) {
++ if (!p->hi_inode)
++ continue;
+
-+static noinline_for_stack
-+void au_ren_rev_diropq(int err, struct au_ren_args *a)
-+{
-+ int rerr;
-+ struct mutex *h_mtx;
-+
-+ /* lock inode simply since inotify is not set to h_inode. */
-+ h_mtx = &au_h_dptr(a->src_dentry, a->btgt)->d_inode->i_mutex;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ rerr = au_diropq_remove(a->src_dentry, a->btgt,
-+ au_ftest_ren(a->flags, DLGT));
-+ mutex_unlock(h_mtx);
-+ if (rerr)
-+ RevertFailure("remove diropq %.*s", AuDLNPair(a->src_dentry));
++ AuDebugOn(type != (p->hi_inode->i_mode & S_IFMT));
++ new_bindex = au_br_index(sb, p->hi_id);
++ if (new_bindex == bindex)
++ continue;
++
++ if (new_bindex < 0) {
++ *update = 1;
++ au_hiput(p);
++ p->hi_inode = NULL;
++ continue;
++ }
++
++ if (new_bindex < iinfo->ii_btop)
++ iinfo->ii_btop = new_bindex;
++ if (iinfo->ii_bbot < new_bindex)
++ iinfo->ii_bbot = new_bindex;
++ /* swap two lower inode, and loop again */
++ q = au_hinode(iinfo, new_bindex);
++ tmp = *q;
++ *q = *p;
++ *p = tmp;
++ if (tmp.hi_inode) {
++ bindex--;
++ p--;
++ }
++ }
++ 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;
++
++out:
++ AuTraceErr(err);
++ return err;
+}
+
-+static noinline_for_stack
-+void au_ren_rev_rename(int err, struct au_ren_args *a)
++void au_refresh_iop(struct inode *inode, int force_getattr)
+{
-+ int rerr;
-+ struct dentry *h_d;
-+ struct qstr *name = &a->src_dentry->d_name;
++ int type;
++ struct au_sbinfo *sbi = au_sbi(inode->i_sb);
++ const struct inode_operations *iop
++ = force_getattr ? aufs_iop : sbi->si_iop_array;
+
-+ h_d = au_lkup_one(name->name, a->h_parent[SRC], name->len, &a->ndx);
-+ rerr = PTR_ERR(h_d);
-+ if (IS_ERR(h_d)) {
-+ RevertFailure("au_lkup_one %.*s", AuDLNPair(a->src_dentry));
++ if (inode->i_op == iop)
+ return;
-+ }
+
-+ AuDebugOn(h_d->d_inode);
-+ vfsub_args_reinit(&a->vargs);
-+ vfsub_ign_hinode(&a->vargs, IN_MOVED_FROM,
-+ au_pinned_hdir(a->pin + DST));
-+ vfsub_ign_hinode(&a->vargs, IN_MOVED_TO, au_pinned_hdir(a->pin + SRC));
-+ rerr = vfsub_rename(au_pinned_h_dir(a->pin + DST),
-+ au_h_dptr(a->src_dentry, a->btgt),
-+ au_pinned_h_dir(a->pin + SRC), h_d, &a->vargs);
-+ d_drop(h_d);
-+ dput(h_d);
-+ /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */
-+ if (rerr)
-+ RevertFailure("rename %.*s", AuDLNPair(a->src_dentry));
-+}
-+
-+static noinline_for_stack
-+void au_ren_rev_cpup(int err, struct au_ren_args *a)
-+{
-+ int rerr;
++ switch (inode->i_mode & S_IFMT) {
++ case S_IFDIR:
++ type = AuIop_DIR;
++ break;
++ case S_IFLNK:
++ type = AuIop_SYMLINK;
++ break;
++ default:
++ type = AuIop_OTHER;
++ break;
++ }
+
-+ vfsub_args_reinit(&a->vargs);
-+ vfsub_ign_hinode(&a->vargs, IN_DELETE, au_pinned_hdir(a->pin + DST));
-+ rerr = vfsub_unlink(au_pinned_h_dir(a->pin + DST), a->h_dentry[DST],
-+ &a->vargs);
-+ au_set_h_dptr(a->src_dentry, a->btgt, NULL);
-+ au_set_dbstart(a->src_dentry, a->bstart[SRC]);
-+ if (rerr)
-+ RevertFailure("unlink %.*s", AuDLNPair(a->h_dentry[DST]));
++ inode->i_op = iop + type;
++ /* unnecessary smp_wmb() */
+}
+
-+static noinline_for_stack
-+void au_ren_rev_whtmp(int err, struct au_ren_args *a)
++int au_refresh_hinode_self(struct inode *inode)
+{
-+ int rerr;
-+ struct dentry *h_d;
-+ struct mutex *h_mtx;
-+ struct qstr *name = &a->dentry->d_name;
++ int err, update;
+
-+ h_d = au_lkup_one(name->name, a->h_parent[DST], name->len, &a->ndx);
-+ rerr = PTR_ERR(h_d);
-+ if (IS_ERR(h_d)) {
-+ RevertFailure("lookup %.*s", AuLNPair(name));
-+ return;
-+ }
-+ if (h_d->d_inode) {
-+ d_drop(h_d);
-+ dput(h_d);
-+ return;
-+ }
++ err = au_ii_refresh(inode, &update);
++ if (!err)
++ au_refresh_hinode_attr(inode, update && S_ISDIR(inode->i_mode));
+
-+ h_mtx = &a->h_dst->d_inode->i_mutex;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ au_hin_resume(au_hi(a->src_dentry->d_inode, a->btgt));
-+ mutex_unlock(h_mtx);
-+ vfsub_args_reinit(&a->vargs);
-+ vfsub_ign_hinode(&a->vargs, IN_MOVED_TO | IN_MOVED_FROM,
-+ au_pinned_hdir(a->pin + DST));
-+ rerr = vfsub_rename(au_pinned_h_dir(a->pin + DST), a->h_dst,
-+ au_pinned_h_dir(a->pin + DST), h_d, &a->vargs);
-+ d_drop(h_d);
-+ dput(h_d);
-+ if (!rerr) {
-+ au_set_h_dptr(a->dentry, a->btgt, NULL);
-+ au_set_h_dptr(a->dentry, a->btgt, dget(a->h_dst));
-+ } else
-+ RevertFailure("rename %.*s", AuDLNPair(a->h_dst));
++ AuTraceErr(err);
++ return err;
+}
+
-+static noinline_for_stack
-+void au_ren_rev_whsrc(int err, struct au_ren_args *a)
++int au_refresh_hinode(struct inode *inode, struct dentry *dentry)
+{
-+ int rerr;
-+
-+ rerr = au_wh_unlink_dentry(au_pinned_hdir(a->pin + SRC),
-+ a->wh_dentry[SRC], a->src_dentry, /*dlgt*/0);
-+ if (rerr)
-+ RevertFailure("unlink %.*s", AuDLNPair(a->wh_dentry[SRC]));
-+}
-+#undef RevertFailure
++ int err, e, update;
++ unsigned int flags;
++ umode_t mode;
++ aufs_bindex_t bindex, bbot;
++ unsigned char isdir;
++ struct au_hinode *p;
++ struct au_iinfo *iinfo;
+
-+/* ---------------------------------------------------------------------- */
++ err = au_ii_refresh(inode, &update);
++ if (unlikely(err))
++ goto out;
+
-+static /* noinline_for_stack */
-+int au_ren_or_cpup(struct au_ren_args *a)
-+{
-+ int err;
++ update = 0;
++ iinfo = au_ii(inode);
++ p = au_hinode(iinfo, iinfo->ii_btop);
++ mode = (inode->i_mode & S_IFMT);
++ isdir = S_ISDIR(mode);
++ flags = au_hi_flags(inode, isdir);
++ bbot = au_dbbot(dentry);
++ for (bindex = au_dbtop(dentry); bindex <= bbot; bindex++) {
++ struct inode *h_i, *h_inode;
++ struct dentry *h_d;
+
-+ AuTraceEnter();
++ h_d = au_h_dptr(dentry, bindex);
++ if (!h_d || d_is_negative(h_d))
++ continue;
+
-+ if (au_dbstart(a->src_dentry) == a->btgt) {
-+ if (a->need_diropq && au_dbdiropq(a->src_dentry) == a->btgt)
-+ a->need_diropq = 0;
-+ vfsub_ign_hinode(&a->vargs, IN_MOVED_FROM,
-+ au_pinned_hdir(a->pin + SRC));
-+ vfsub_ign_hinode(&a->vargs, IN_MOVED_TO,
-+ au_pinned_hdir(a->pin + DST));
-+ /* nfs_rename() calls d_delete() */
-+ if (au_test_nfs(au_pinned_h_dir(a->pin + DST)->i_sb)
-+ && a->h_dentry[DST]->d_inode
-+ && (S_ISDIR(a->h_dentry[DST]->d_inode->i_mode)
-+ || atomic_read(&a->h_dentry[DST]->d_count) <= 2))
-+ vfsub_ign_hinode(&a->vargs, IN_DELETE,
-+ au_pinned_hdir(a->pin + DST));
-+ AuDebugOn(au_dbstart(a->src_dentry) != a->btgt);
-+ err = vfsub_rename(au_pinned_h_dir(a->pin + SRC),
-+ au_h_dptr(a->src_dentry, a->btgt),
-+ au_pinned_h_dir(a->pin + DST),
-+ a->h_dentry[DST], &a->vargs);
-+ } else {
-+ struct mutex *h_mtx = &a->h_dentry[SRC]->d_inode->i_mutex;
-+
-+ a->bycpup = 1;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ au_set_dbstart(a->src_dentry, a->btgt);
-+ au_set_h_dptr(a->src_dentry, a->btgt, dget(a->h_dentry[DST]));
-+ err = au_sio_cpup_single(a->src_dentry, a->btgt, a->bstart[SRC],
-+ -1, !AuCpup_DTIME, a->parent[DST]);
-+ if (unlikely(err)) {
-+ au_set_h_dptr(a->src_dentry, a->btgt, NULL);
-+ au_set_dbstart(a->src_dentry, a->bstart[SRC]);
++ h_inode = d_inode(h_d);
++ AuDebugOn(mode != (h_inode->i_mode & S_IFMT));
++ if (iinfo->ii_btop <= bindex && bindex <= iinfo->ii_bbot) {
++ h_i = au_h_iptr(inode, bindex);
++ if (h_i) {
++ if (h_i == h_inode)
++ continue;
++ err = -EIO;
++ break;
++ }
+ }
-+ mutex_unlock(h_mtx);
-+ }
++ if (bindex < iinfo->ii_btop)
++ iinfo->ii_btop = bindex;
++ if (iinfo->ii_bbot < bindex)
++ iinfo->ii_bbot = bindex;
++ au_set_h_iptr(inode, bindex, au_igrab(h_inode), flags);
++ update = 1;
++ }
++ au_update_ibrange(inode, /*do_put_zero*/0);
++ e = au_dy_irefresh(inode);
++ if (unlikely(e && !err))
++ err = e;
++ if (!err)
++ au_refresh_hinode_attr(inode, update && isdir);
+
++out:
++ AuTraceErr(err);
+ return err;
+}
+
-+static /* noinline_for_stack */
-+int au_ren_del_whtmp(struct au_ren_args *a)
++static int set_inode(struct inode *inode, struct dentry *dentry)
+{
+ int err;
++ unsigned int flags;
++ umode_t mode;
++ aufs_bindex_t bindex, btop, btail;
++ unsigned char isdir;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ struct au_iinfo *iinfo;
++ const struct inode_operations *iop;
+
-+ AuTraceEnter();
++ IiMustWriteLock(inode);
+
-+ if (au_test_nfs(a->h_dst->d_sb)
-+ || !au_nhash_test_longer_wh(&a->whlist, a->btgt,
-+ au_sbi(a->sb)->si_dirwh)) {
-+ err = au_whtmp_rmdir(a->dir, a->btgt, a->h_dst, &a->whlist);
++ err = 0;
++ isdir = 0;
++ iop = au_sbi(inode->i_sb)->si_iop_array;
++ btop = au_dbtop(dentry);
++ h_dentry = au_h_dptr(dentry, btop);
++ h_inode = d_inode(h_dentry);
++ mode = h_inode->i_mode;
++ switch (mode & S_IFMT) {
++ case S_IFREG:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_OTHER;
++ inode->i_fop = &aufs_file_fop;
++ err = au_dy_iaop(inode, btop, h_inode);
+ if (unlikely(err))
-+ AuWarn("failed removing whtmp dir %.*s (%d), "
-+ "ignored.\n", AuDLNPair(a->h_dst), err);
-+ } else {
-+ au_whtmp_kick_rmdir(a->dir, a->btgt, a->h_dst, &a->whlist,
-+ a->thargs);
-+ dput(a->h_dst);
-+ a->thargs = NULL;
++ goto out;
++ break;
++ case S_IFDIR:
++ isdir = 1;
++ btail = au_dbtaildir(dentry);
++ inode->i_op = iop + AuIop_DIR;
++ inode->i_fop = &aufs_dir_fop;
++ break;
++ case S_IFLNK:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_SYMLINK;
++ break;
++ case S_IFBLK:
++ case S_IFCHR:
++ case S_IFIFO:
++ case S_IFSOCK:
++ btail = au_dbtail(dentry);
++ inode->i_op = iop + AuIop_OTHER;
++ init_special_inode(inode, mode, h_inode->i_rdev);
++ break;
++ default:
++ AuIOErr("Unknown file type 0%o\n", mode);
++ err = -EIO;
++ goto out;
+ }
+
-+ return 0;
++ /* do not set hnotify for whiteouted dirs (SHWH mode) */
++ flags = au_hi_flags(inode, isdir);
++ if (au_opt_test(au_mntflags(dentry->d_sb), SHWH)
++ && au_ftest_hi(flags, HNOTIFY)
++ && dentry->d_name.len > AUFS_WH_PFX_LEN
++ && !memcmp(dentry->d_name.name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))
++ au_fclr_hi(flags, HNOTIFY);
++ iinfo = au_ii(inode);
++ iinfo->ii_btop = btop;
++ iinfo->ii_bbot = btail;
++ for (bindex = btop; bindex <= btail; bindex++) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry)
++ au_set_h_iptr(inode, bindex,
++ au_igrab(d_inode(h_dentry)), flags);
++ }
++ au_cpup_attr_all(inode, /*force*/1);
++ /*
++ * to force calling aufs_get_inode_acl() every time,
++ * do not call cache_no_acl() for aufs inode.
++ */
++
++out:
++ return err;
+}
+
-+static /* noinline_for_stack */
-+int au_ren_diropq(struct au_ren_args *a)
++/*
++ * successful returns with iinfo write_locked
++ * minus: errno
++ * zero: success, matched
++ * plus: no error, but unmatched
++ */
++static int reval_inode(struct inode *inode, struct dentry *dentry)
+{
+ int err;
-+ struct dentry *diropq;
-+ struct mutex *h_mtx;
++ unsigned int gen, igflags;
++ aufs_bindex_t bindex, bbot;
++ struct inode *h_inode, *h_dinode;
++ struct dentry *h_dentry;
++
++ /*
++ * before this function, if aufs got any iinfo lock, it must be only
++ * one, the parent dir.
++ * it can happen by UDBA and the obsoleted inode number.
++ */
++ err = -EIO;
++ if (unlikely(inode->i_ino == parent_ino(dentry)))
++ goto out;
+
-+ AuTraceEnter();
++ err = 1;
++ ii_write_lock_new_child(inode);
++ h_dentry = au_h_dptr(dentry, au_dbtop(dentry));
++ h_dinode = d_inode(h_dentry);
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); bindex <= bbot; bindex++) {
++ h_inode = au_h_iptr(inode, bindex);
++ if (!h_inode || h_inode != h_dinode)
++ continue;
+
-+ err = 0;
-+ h_mtx = &au_h_dptr(a->src_dentry, a->btgt)->d_inode->i_mutex;
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ diropq = au_diropq_create(a->src_dentry, a->btgt,
-+ au_ftest_ren(a->flags, DLGT));
-+ mutex_unlock(h_mtx);
-+ if (IS_ERR(diropq))
-+ err = PTR_ERR(diropq);
-+ dput(diropq);
++ err = 0;
++ gen = au_iigen(inode, &igflags);
++ if (gen == au_digen(dentry)
++ && !au_ig_ftest(igflags, HALF_REFRESHED))
++ break;
++
++ /* fully refresh inode using dentry */
++ err = au_refresh_hinode(inode, dentry);
++ if (!err)
++ au_update_iigen(inode, /*half*/0);
++ break;
++ }
+
++ if (unlikely(err))
++ ii_write_unlock(inode);
++out:
+ return err;
+}
+
-+static /* noinline_for_stack */
-+int do_rename(struct au_ren_args *a)
++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ unsigned int d_type, ino_t *ino)
+{
-+ int err;
-+ aufs_bindex_t bindex, bend;
-+ struct dentry *h_d;
++ int err, idx;
++ const int isnondir = d_type != DT_DIR;
+
-+ LKTRTrace("%.*s/%.*s, %.*s/%.*s, "
-+ "hd{%p, %p}, hp{%p, %p}, wh %p, btgt %d, bstart{%d, %d}, "
-+ "flags 0x%x\n",
-+ AuDLNPair(a->parent[SRC]), AuDLNPair(a->src_dentry),
-+ AuDLNPair(a->parent[DST]), AuDLNPair(a->dentry),
-+ a->h_dentry[SRC], a->h_dentry[DST],
-+ a->h_parent[SRC], a->h_parent[DST],
-+ &a->whlist, a->btgt,
-+ a->bstart[SRC], a->bstart[DST],
-+ a->flags);
-+
-+ /* prepare workqueue args */
-+ if (au_ftest_ren(a->flags, ISDIR) && a->h_dentry[DST]->d_inode) {
-+ err = -ENOMEM;
-+ a->thargs = kmalloc(sizeof(*a->thargs), GFP_NOFS);
-+ if (unlikely(!a->thargs))
++ /* prevent hardlinked inode number from race condition */
++ if (isnondir) {
++ err = au_xinondir_enter(sb, bindex, h_ino, &idx);
++ if (unlikely(err))
+ goto out;
-+ a->h_dst = dget(a->h_dentry[DST]);
-+ }
-+
-+ a->ndx.nfsmnt = au_nfsmnt(a->sb, a->btgt);
-+ if (au_ftest_ren(a->flags, DLGT))
-+ au_fset_ndx(a->ndx.flags, DLGT);
-+
-+ /* create whiteout for src_dentry */
-+ if (au_ftest_ren(a->flags, WHSRC)) {
-+ a->wh_dentry[SRC] = au_wh_create(a->src_dentry, a->btgt,
-+ a->h_parent[SRC], &a->ndx);
-+ err = PTR_ERR(a->wh_dentry[SRC]);
-+ if (IS_ERR(a->wh_dentry[SRC]))
-+ goto out_thargs;
-+ }
-+
-+ /* lookup whiteout for dentry */
-+ if (au_ftest_ren(a->flags, WHDST)) {
-+ h_d = au_wh_lkup(a->h_parent[DST], &a->dentry->d_name, &a->ndx);
-+ err = PTR_ERR(h_d);
-+ if (IS_ERR(h_d))
-+ goto out_whsrc;
-+ if (!h_d->d_inode)
-+ dput(h_d);
-+ else
-+ a->wh_dentry[DST] = h_d;
+ }
+
-+ /* rename dentry to tmpwh */
-+ if (a->thargs) {
-+ struct au_hinode *hinode;
++ err = au_xino_read(sb, bindex, h_ino, ino);
++ if (unlikely(err))
++ goto out_xinondir;
+
-+ AuDbgDentry(a->h_dentry[DST]);
-+ err = au_whtmp_ren(a->dir, a->btgt, a->h_dentry[DST]);
-+ if (unlikely(err))
-+ goto out_whdst;
-+ AuDbgDentry(a->h_dentry[DST]);
-+ hinode = au_hi(a->dentry->d_inode, a->btgt);
-+ /* todo: bad approach? */
-+ mutex_lock_nested(&hinode->hi_inode->i_mutex, AuLsc_I_CHILD);
-+ au_hin_suspend(hinode);
-+ mutex_unlock(&hinode->hi_inode->i_mutex);
-+ au_set_h_dptr(a->dentry, a->btgt, NULL);
-+ AuDbgDentry(a->h_dentry[DST]);
-+ err = au_lkup_neg(a->dentry, a->btgt);
++ if (!*ino) {
++ err = -EIO;
++ *ino = au_xino_new_ino(sb);
++ if (unlikely(!*ino))
++ goto out_xinondir;
++ err = au_xino_write(sb, bindex, h_ino, *ino);
+ if (unlikely(err))
-+ goto out_whtmp;
-+ a->h_dentry[DST] = au_h_dptr(a->dentry, a->btgt);
++ goto out_xinondir;
+ }
+
-+ /* cpup src */
-+ if (a->h_dentry[DST]->d_inode && a->bstart[SRC] != a->btgt) {
-+ struct mutex *h_mtx = &a->h_dentry[SRC]->d_inode->i_mutex;
++out_xinondir:
++ if (isnondir && idx >= 0)
++ au_xinondir_leave(sb, bindex, h_ino, idx);
++out:
++ return err;
++}
+
-+ mutex_lock_nested(h_mtx, AuLsc_I_CHILD);
-+ err = au_sio_cpup_simple(a->src_dentry, a->btgt, -1,
-+ !AuCpup_DTIME);
-+ mutex_unlock(h_mtx);
-+ if (unlikely(err))
-+ goto out_whtmp;
-+ }
++/* successful returns with iinfo write_locked */
++/* todo: return with unlocked? */
++struct inode *au_new_inode(struct dentry *dentry, int must_new)
++{
++ struct inode *inode, *h_inode;
++ struct dentry *h_dentry;
++ struct super_block *sb;
++ ino_t h_ino, ino;
++ int err, idx, hlinked;
++ aufs_bindex_t btop;
+
-+ /* rename by vfs_rename or cpup */
-+ a->need_diropq = au_ftest_ren(a->flags, ISDIR)
-+ && (a->wh_dentry[DST]
-+ || au_dbdiropq(a->dentry) == a->btgt
-+ /* hide the lower to keep xino */
-+ || a->btgt < au_dbend(a->dentry)
-+ || au_opt_test(a->mnt_flags, ALWAYS_DIROPQ));
-+ a->bycpup = 0;
-+ vfsub_args_init(&a->vargs, a->ign, au_ftest_ren(a->flags, DLGT), 0);
-+ err = au_ren_or_cpup(a);
-+ if (unlikely(err))
-+ goto out_whtmp;
++ sb = dentry->d_sb;
++ btop = au_dbtop(dentry);
++ h_dentry = au_h_dptr(dentry, btop);
++ h_inode = d_inode(h_dentry);
++ h_ino = h_inode->i_ino;
++ hlinked = !d_is_dir(h_dentry) && h_inode->i_nlink > 1;
+
-+ /* make dir opaque */
-+ if (a->need_diropq) {
-+ err = au_ren_diropq(a);
++new_ino:
++ /*
++ * stop 'race'-ing between hardlinks under different
++ * parents.
++ */
++ if (hlinked) {
++ err = au_xinondir_enter(sb, btop, h_ino, &idx);
++ inode = ERR_PTR(err);
+ if (unlikely(err))
-+ goto out_rename;
++ goto out;
+ }
+
-+ /* update target timestamps */
-+ AuDebugOn(au_dbstart(a->src_dentry) != a->btgt);
-+ a->h_src = au_h_dptr(a->src_dentry, a->btgt);
-+ au_update_fuse_h_inode(NULL, a->h_src); /*ignore*/
-+ /*
-+ * fsstack_copy_attr_atime(a->src_dentry->d_inode, a->h_src->d_inode);
-+ */
-+ a->src_dentry->d_inode->i_ctime = a->h_src->d_inode->i_ctime;
++ err = au_xino_read(sb, btop, h_ino, &ino);
++ inode = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_xinondir;
+
-+ /* remove whiteout for dentry */
-+ if (a->wh_dentry[DST]) {
-+ err = au_wh_unlink_dentry(au_pinned_hdir(a->pin + DST),
-+ a->wh_dentry[DST], a->dentry,
-+ /*dlgt*/0);
-+ if (unlikely(err))
-+ goto out_diropq;
++ if (!ino) {
++ ino = au_xino_new_ino(sb);
++ if (unlikely(!ino)) {
++ inode = ERR_PTR(-EIO);
++ goto out_xinondir;
++ }
+ }
+
-+ /* remove whtmp */
-+ if (a->thargs)
-+ /* ignore this error */
-+ au_ren_del_whtmp(a);
++ AuDbg("i%lu\n", (unsigned long)ino);
++ inode = au_iget_locked(sb, ino);
++ err = PTR_ERR(inode);
++ if (IS_ERR(inode))
++ goto out_xinondir;
+
-+ err = 0;
-+ goto out_success;
++ AuDbg("%lx, new %d\n", inode->i_state, !!(inode->i_state & I_NEW));
++ if (inode->i_state & I_NEW) {
++ ii_write_lock_new_child(inode);
++ err = set_inode(inode, dentry);
++ if (!err) {
++ unlock_new_inode(inode);
++ goto out_xinondir; /* success */
++ }
+
-+ out_diropq:
-+ if (a->need_diropq)
-+ au_ren_rev_diropq(err, a);
-+ out_rename:
-+ if (!a->bycpup)
-+ au_ren_rev_rename(err, a);
-+ else
-+ au_ren_rev_cpup(err, a);
-+ out_whtmp:
-+ if (a->thargs)
-+ au_ren_rev_whtmp(err, a);
-+ out_whdst:
-+ dput(a->wh_dentry[DST]);
-+ a->wh_dentry[DST] = NULL;
-+ out_whsrc:
-+ if (a->wh_dentry[SRC])
-+ au_ren_rev_whsrc(err, a);
-+ d_drop(a->src_dentry);
-+ bend = au_dbend(a->src_dentry);
-+ for (bindex = au_dbstart(a->src_dentry); bindex <= bend; bindex++) {
-+ h_d = au_h_dptr(a->src_dentry, bindex);
-+ if (h_d)
-+ d_drop(h_d);
-+ }
-+ d_drop(a->dentry);
-+ bend = au_dbend(a->dentry);
-+ for (bindex = au_dbstart(a->dentry); bindex <= bend; bindex++) {
-+ h_d = au_h_dptr(a->dentry, bindex);
-+ if (h_d)
-+ d_drop(h_d);
-+ }
-+ au_update_dbstart(a->dentry);
-+ if (a->thargs)
-+ d_drop(a->h_dst);
-+ out_success:
-+ dput(a->wh_dentry[SRC]);
-+ dput(a->wh_dentry[DST]);
-+ out_thargs:
-+ if (a->thargs) {
-+ dput(a->h_dst);
-+ kfree(a->thargs);
++ /*
++ * iget_failed() calls iput(), but we need to call
++ * ii_write_unlock() after iget_failed(). so dirty hack for
++ * i_count.
++ */
++ atomic_inc(&inode->i_count);
++ iget_failed(inode);
++ ii_write_unlock(inode);
++ au_xino_write(sb, btop, h_ino, /*ino*/0);
++ /* ignore this error */
++ goto out_iput;
++ } else if (!must_new && !IS_DEADDIR(inode) && inode->i_nlink) {
++ /*
++ * horrible race condition between lookup, readdir and copyup
++ * (or something).
++ */
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++ err = reval_inode(inode, dentry);
++ if (unlikely(err < 0)) {
++ hlinked = 0;
++ goto out_iput;
++ }
++ if (!err)
++ goto out; /* success */
++ else if (hlinked && idx >= 0) {
++ err = au_xinondir_enter(sb, btop, h_ino, &idx);
++ if (unlikely(err)) {
++ iput(inode);
++ inode = ERR_PTR(err);
++ goto out;
++ }
++ }
+ }
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
+
-+/*
-+ * test if @dentry dir can be rename destination or not.
-+ * success means, it is a logically empty dir.
-+ */
-+static int may_rename_dstdir(struct dentry *dentry, aufs_bindex_t btgt,
-+ struct au_nhash *whlist)
-+{
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ if (unlikely(au_test_fs_unique_ino(h_inode)))
++ AuWarn1("Warning: Un-notified UDBA or repeatedly renamed dir,"
++ " b%d, %s, %pd, hi%lu, i%lu.\n",
++ btop, au_sbtype(h_dentry->d_sb), dentry,
++ (unsigned long)h_ino, (unsigned long)ino);
++ ino = 0;
++ err = au_xino_write(sb, btop, h_ino, /*ino*/0);
++ if (!err) {
++ iput(inode);
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++ goto new_ino;
++ }
+
-+ return au_test_empty(dentry, whlist);
++out_iput:
++ iput(inode);
++ inode = ERR_PTR(err);
++out_xinondir:
++ if (hlinked && idx >= 0)
++ au_xinondir_leave(sb, btop, h_ino, idx);
++out:
++ return inode;
+}
+
-+/*
-+ * test if @dentry dir can be rename source or not.
-+ * if it can, return 0 and @children is filled.
-+ * success means,
-+ * - or, it is a logically empty dir.
-+ * - or, it exists on writable branch and has no children including whiteouts
-+ * on the lower branch.
-+ */
-+static int may_rename_srcdir(struct dentry *dentry, aufs_bindex_t btgt)
++/* ---------------------------------------------------------------------- */
++
++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
++ struct inode *inode)
+{
+ int err;
-+ aufs_bindex_t bstart;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ struct inode *hi;
+
-+ bstart = au_dbstart(dentry);
-+ if (bstart != btgt) {
-+ struct au_nhash *whlist;
++ err = au_br_rdonly(au_sbr(sb, bindex));
+
-+ whlist = au_nhash_new(GFP_NOFS);
-+ err = PTR_ERR(whlist);
-+ if (IS_ERR(whlist))
-+ goto out;
-+ err = au_test_empty(dentry, whlist);
-+ au_nhash_del(whlist);
-+ goto out;
++ /* pseudo-link after flushed may happen out of bounds */
++ if (!err
++ && inode
++ && au_ibtop(inode) <= bindex
++ && bindex <= au_ibbot(inode)) {
++ /*
++ * permission check is unnecessary since vfsub routine
++ * will be called later
++ */
++ hi = au_h_iptr(inode, bindex);
++ if (hi)
++ err = IS_IMMUTABLE(hi) ? -EROFS : 0;
+ }
+
-+ if (bstart == au_dbtaildir(dentry))
-+ return 0; /* success */
-+
-+ err = au_test_empty_lower(dentry);
-+
-+ out:
-+ if (/* unlikely */(err == -ENOTEMPTY)) {
-+ AuWarn1("renaming dir who has child(ren) on multiple branches,"
-+ " is not supported\n");
-+ err = -EXDEV;
-+ }
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* mainly for link(2) and rename(2) */
-+int au_wbr(struct dentry *dentry, aufs_bindex_t btgt)
++int au_test_h_perm(struct mnt_idmap *h_idmap, struct inode *h_inode,
++ int mask)
+{
-+ aufs_bindex_t bdiropq, bwh;
-+ struct dentry *parent;
-+
-+ LKTRTrace("%.*s, b%d\n", AuDLNPair(dentry), btgt);
-+ parent = dentry->d_parent;
-+ IMustLock(parent->d_inode); /* dir is locked */
-+
-+ bdiropq = au_dbdiropq(parent);
-+ bwh = au_dbwh(dentry);
-+ if (au_br_rdonly(au_sbr(dentry->d_sb, btgt))
-+ || (0 <= bdiropq && bdiropq < btgt)
-+ || (0 <= bwh && bwh < btgt))
-+ btgt = -1;
-+
-+ LKTRTrace("btgt %d\n", btgt);
-+ return btgt;
++ if (uid_eq(current_fsuid(), GLOBAL_ROOT_UID))
++ return 0;
++ return inode_permission(h_idmap, h_inode, mask);
+}
+
-+/*
-+ * simple tests for rename.
-+ * following the checks in vfs, plus the parent-child relationship.
-+ */
-+static int au_may_ren(struct au_ren_args *a)
++int au_test_h_perm_sio(struct mnt_idmap *h_idmap, struct inode *h_inode,
++ int mask)
+{
-+ int err;
-+ struct inode *h_inode;
-+
-+ AuTraceEnter();
-+
-+ if (a->bstart[SRC] == a->btgt) {
-+ err = au_may_del(a->src_dentry, a->btgt, a->h_parent[SRC],
-+ au_ftest_ren(a->flags, ISDIR), &a->ndx);
-+ if (unlikely(err))
-+ goto out;
-+ err = -EINVAL;
-+ if (unlikely(a->h_dentry[SRC] == a->h_trap))
-+ goto out;
-+ }
-+
-+ err = 0;
-+ if (a->bstart[DST] != a->btgt)
-+ goto out;
-+
-+ err = -EIO;
-+ h_inode = a->h_dentry[DST]->d_inode;
-+ if (!a->dentry->d_inode) {
-+ if (unlikely(h_inode))
-+ goto out;
-+ err = au_may_add(a->dentry, a->btgt, a->h_parent[DST],
-+ au_ftest_ren(a->flags, ISDIR), &a->ndx);
-+ } else {
-+ if (unlikely(!h_inode || !h_inode->i_nlink))
-+ goto out;
-+ err = au_may_del(a->dentry, a->btgt, a->h_parent[DST],
-+ au_ftest_ren(a->flags, ISDIR), &a->ndx);
-+ if (unlikely(err))
-+ goto out;
-+ err = -ENOTEMPTY;
-+ if (unlikely(a->h_dentry[DST] == a->h_trap))
-+ goto out;
-+ err = 0;
-+ }
-+
-+ out:
-+ if (unlikely(err == -ENOENT || err == -EEXIST))
-+ err = -EIO;
-+ AuTraceErr(err);
-+ return err;
++ if (au_test_nfs(h_inode->i_sb)
++ && (mask & MAY_WRITE)
++ && S_ISDIR(h_inode->i_mode))
++ mask |= MAY_READ; /* force permission check */
++ return au_test_h_perm(h_idmap, h_inode, mask);
+}
-+
+diff -urN /usr/share/empty/fs/aufs/inode.h linux/fs/aufs/inode.h
+--- /usr/share/empty/fs/aufs/inode.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/inode.h 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,707 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * locking order
-+ * (VFS)
-+ * - src_dir and dir by lock_rename()
-+ * - inode if exitsts
-+ * (aufs)
-+ * - lock all
-+ * + src_dentry and dentry by aufs_read_and_write_lock2() which calls,
-+ * + si_read_lock
-+ * + di_write_lock2_child()
-+ * + di_write_lock_child()
-+ * + ii_write_lock_child()
-+ * + di_write_lock_child2()
-+ * + ii_write_lock_child2()
-+ * + src_parent and parent
-+ * + di_write_lock_parent()
-+ * + ii_write_lock_parent()
-+ * + di_write_lock_parent2()
-+ * + ii_write_lock_parent2()
-+ * + if udab=inotify is specified, lock grand parents (crazy)
-+ * + di_read_lock_gparent()
-+ * + ii_read_lock_gparent()
-+ * + di_read_lock_gparent2()
-+ * + ii_read_lock_gparent2()
-+ * + mutex_lock(s_vfs_rename_mutex)
-+ * + mutex_lock_gparent()
-+ * + mutex_lock_gparent2()
-+ * + mutex_lock_parent()
-+ * + mutex_lock_parent2()
-+ * + else lower src_dir and dir by vfsub_lock_rename()
-+ * + verify the every relations between child, parent and grand parent. if any
-+ * of them failed, unlock all and return -EBUSY.
++ * Copyright (C) 2005-2022 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/>.
+ */
-+static void au_ren_pin_init(struct au_pin *first, struct dentry *d1,
-+ struct au_pin *next, struct dentry *d2,
-+ aufs_bindex_t bindex)
-+{
-+ AuTraceEnter();
+
-+ /* AuLsc_DI_PARENT3 is for higher gparent initially */
-+ au_pin_init(first, d1, bindex, AuLsc_DI_PARENT2, AuLsc_I_PARENT2,
-+ AuPin_DI_LOCKED | AuPin_DO_GPARENT);
-+ /* AuLsc_DI_PARENT4 is for lower gparent initially */
-+ au_pin_init(next, d2, bindex, AuLsc_DI_PARENT3, AuLsc_I_PARENT4,
-+ AuPin_DI_LOCKED | AuPin_DO_GPARENT);
-+}
++/*
++ * inode operations
++ */
+
-+static void au_ren_fake_pin(struct au_ren_args *a)
-+{
-+ int i;
-+ struct au_pin1 *p;
-+ struct inode *h_i;
++#ifndef __AUFS_INODE_H__
++#define __AUFS_INODE_H__
+
-+ AuTraceEnter();
++#ifdef __KERNEL__
+
-+ /* they increment the ref counter */
-+ for (i = 0; i < 2; i++) {
-+ p = a->pin[i].pin + AuPin_PARENT;
-+ au_pin_set_parent(a->pin + i, a->parent[i]);
-+ dput(a->parent[i]);
-+ h_i = a->h_parent[i]->d_inode;
-+ au_pin_set_h_dir(a->pin + i, h_i);
-+ iput(h_i);
++#include <linux/fsnotify.h>
++#include "rwsem.h"
+
-+ if (!a->gparent[i]) {
-+ au_pin_set_gparent(a->pin + i, NULL);
-+ au_pin_set_h_gdir(a->pin + i, NULL);
-+ } else {
-+ au_pin_set_gparent(a->pin + i, a->gparent[i]);
-+ dput(a->gparent[i]);
-+ h_i = au_h_iptr(a->gparent[i]->d_inode, a->btgt);
-+ au_pin_set_h_gdir(a->pin + i, h_i);
-+ iput(h_i);
-+ }
-+ }
-+}
++struct vfsmount;
+
-+/* crazy */
-+/* cf. i_op.c: au_do_pin() */
-+static int au_ren_pin4(int higher, int lower, struct au_ren_args *a)
-+{
-+ int err, i, lsc;
-+ struct au_pin *p;
-+ struct au_pin1 *p4[4];
-+ struct inode *h_dir;
++struct au_hnotify {
++#ifdef CONFIG_AUFS_HNOTIFY
++#ifdef CONFIG_AUFS_HFSNOTIFY
++ /* never use fsnotify_add_vfsmount_mark() */
++ struct fsnotify_mark hn_mark;
++#endif
++ struct inode *hn_aufs_inode; /* no get/put */
++ struct rcu_head rcu;
++#endif
++} ____cacheline_aligned_in_smp;
+
-+#if 0
-+ lktr_set_pid(current->pid, LktrArrayPid);
-+ LKTRTrace("i%lu, %.*s, i%lu, %.*s\n",
-+ a->src_dir->i_ino, AuDLNPair(a->src_dentry),
-+ a->dir->i_ino, AuDLNPair(a->dentry));
-+ lktr_clear_pid(current->pid, LktrArrayPid);
++struct au_hinode {
++ struct inode *hi_inode;
++ aufs_bindex_t hi_id;
++#ifdef CONFIG_AUFS_HNOTIFY
++ struct au_hnotify *hi_notify;
+#endif
-+ LKTRTrace("%d, %d\n", higher, lower);
+
-+ err = 0;
-+ p = a->pin + higher;
-+ p4[0] = au_pin_gp(p); /* highest */
-+ p4[1] = p->pin + AuPin_PARENT;
-+ p = a->pin + lower;
-+ p4[2] = au_pin_gp(p);
-+ p4[3] = p->pin + AuPin_PARENT;
++ /* reference to the copied-up whiteout with get/put */
++ struct dentry *hi_whdentry;
++};
+
-+ if (a->gparent[higher]) {
-+ au_pin_do_set_parent(p4[0], a->gparent[higher]);
-+ au_pin_do_set_dentry(p4[0], a->parent[higher]);
-+ }
-+ au_pin_do_set_parent(p4[1], a->parent[higher]);
-+ if (a->gparent[lower]) {
-+ au_pin_do_set_parent(p4[2], a->gparent[lower]);
-+ au_pin_do_set_dentry(p4[2], a->parent[lower]);
-+ }
-+ au_pin_do_set_parent(p4[3], a->parent[lower]);
++/* ig_flags */
++#define AuIG_HALF_REFRESHED 1
++#define au_ig_ftest(flags, name) ((flags) & AuIG_##name)
++#define au_ig_fset(flags, name) \
++ do { (flags) |= AuIG_##name; } while (0)
++#define au_ig_fclr(flags, name) \
++ do { (flags) &= ~AuIG_##name; } while (0)
++
++struct au_iigen {
++ spinlock_t ig_spin;
++ __u32 ig_generation, ig_flags;
++};
+
-+#if 0
-+ lktr_set_pid(current->pid, LktrArrayPid);
-+ for (i = 0; i < 4; i++)
-+ AuDbgDentry(p4[i]->parent);
-+#endif
-+ DiMustWriteLock(p4[3]->parent);
-+ di_write_unlock(p4[1]->parent);
-+ if (p4[2]->parent)
-+ di_read_lock_parent2(p4[2]->parent, AuLock_IR);
-+ di_write_lock_parent3(p4[1]->parent);
-+ if (p4[0]->parent)
-+ di_read_lock_parent4(p4[0]->parent, AuLock_IR);
-+ //lktr_clear_pid(current->pid, LktrArrayPid);
++struct au_vdir;
++struct au_iinfo {
++ struct au_iigen ii_generation;
++ struct super_block *ii_hsb1; /* no get/put */
+
-+ vfsub_lock_rename_mutex(au_sbr_sb(a->dentry->d_sb, a->btgt));
-+ au_fset_ren(a->flags, VFSLOCK);
++ struct au_rwsem ii_rwsem;
++ aufs_bindex_t ii_btop, ii_bbot;
++ __u32 ii_higen;
++ struct au_hinode *ii_hinode;
++ struct au_vdir *ii_vdir;
++};
+
-+ lsc = AuLsc_I_PARENT;
-+ for (i = 0; i < 4; i++, lsc++) {
-+ if (p4[i]->parent) {
-+ h_dir = au_h_iptr(p4[i]->parent->d_inode, a->btgt);
-+ au_pin_do_set_h_dir(p4[i], h_dir);
-+ mutex_lock_nested(&h_dir->i_mutex, lsc);
-+ }
-+ }
++struct au_icntnr {
++ struct au_iinfo iinfo;
++ struct inode vfs_inode;
++ struct hlist_bl_node plink;
++ struct rcu_head rcu;
++} ____cacheline_aligned_in_smp;
+
-+ AuTraceErr(err);
-+ return err;
-+}
++/* au_pin flags */
++#define AuPin_DI_LOCKED 1
++#define AuPin_MNT_WRITE (1 << 1)
++#define au_ftest_pin(flags, name) ((flags) & AuPin_##name)
++#define au_fset_pin(flags, name) \
++ do { (flags) |= AuPin_##name; } while (0)
++#define au_fclr_pin(flags, name) \
++ do { (flags) &= ~AuPin_##name; } while (0)
+
-+static struct dentry *au_ren_pin3(int higher, int lower, struct au_ren_args *a)
-+{
-+ struct dentry *h_trap;
-+ struct au_pin *p;
-+ int err;
++struct au_pin {
++ /* input */
++ struct dentry *dentry;
++ unsigned int udba;
++ unsigned char lsc_di, lsc_hi, flags;
++ aufs_bindex_t bindex;
+
-+ LKTRTrace("%d, %d\n", higher, lower);
++ /* output */
++ struct dentry *parent;
++ struct au_hinode *hdir;
++ struct vfsmount *h_mnt;
+
-+ p = a->pin + higher;
-+ AuDebugOn(!au_pin_gp(p));
-+ au_fset_pin(au_pin_gp(p)->flags, VFS_RENAME);
-+ err = au_do_pin(p->pin + AuPin_PARENT, au_pin_gp(p));
-+ h_trap = ERR_PTR(err);
-+ if (unlikely(err))
-+ goto out;
-+ p = a->pin + lower;
-+ au_fclr_pin(p->pin[AuPin_PARENT].flags, DO_GPARENT);
-+ err = au_do_pin(p->pin + AuPin_PARENT, NULL);
-+ h_trap = ERR_PTR(err);
-+ if (unlikely(err)) {
-+ p = a->pin + higher;
-+ au_do_unpin(p->pin + AuPin_PARENT, au_pin_gp(p));
-+ goto out;
-+ }
-+ h_trap = au_pinned_h_parent(p);
++ /* temporary unlock/relock for copyup */
++ struct dentry *h_dentry, *h_parent;
++ struct au_branch *br;
++ struct task_struct *task;
++};
+
-+ out:
-+ AuTraceErrPtr(h_trap);
-+ return h_trap;
-+}
++void au_pin_hdir_unlock(struct au_pin *p);
++int au_pin_hdir_lock(struct au_pin *p);
++int au_pin_hdir_relock(struct au_pin *p);
++void au_pin_hdir_acquire_nest(struct au_pin *p);
++void au_pin_hdir_release(struct au_pin *p);
++
++/* ---------------------------------------------------------------------- */
+
-+static struct dentry *au_ren_pin(struct au_ren_args *a)
++static inline struct au_iinfo *au_ii(struct inode *inode)
+{
-+ struct dentry *h_trap;
-+ struct inode *h_gdir;
-+ int err, i, same_gp;
++ BUG_ON(is_bad_inode(inode));
++ return &(container_of(inode, struct au_icntnr, vfs_inode)->iinfo);
++}
+
-+ //lktr_set_pid(current->pid, LktrArrayPid);
-+ AuTraceEnter();
-+ AuDebugOn(!au_opt_test(a->mnt_flags, UDBA_INOTIFY));
++/* ---------------------------------------------------------------------- */
+
-+ /* gdir is not locked */
-+ same_gp = 0;
-+ if (!IS_ROOT(a->parent[SRC]))
-+ a->gparent[SRC] = dget_parent(a->parent[SRC]);
-+ if (!IS_ROOT(a->parent[DST])) {
-+ a->gparent[DST] = dget_parent(a->parent[DST]);
-+ same_gp = (a->gparent[SRC] == a->gparent[DST]);
-+ }
++/* inode.c */
++struct inode *au_igrab(struct inode *inode);
++void au_refresh_iop(struct inode *inode, int force_getattr);
++int au_refresh_hinode_self(struct inode *inode);
++int au_refresh_hinode(struct inode *inode, struct dentry *dentry);
++int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ unsigned int d_type, ino_t *ino);
++struct inode *au_new_inode(struct dentry *dentry, int must_new);
++int au_test_ro(struct super_block *sb, aufs_bindex_t bindex,
++ struct inode *inode);
++int au_test_h_perm(struct mnt_idmap *h_idmap, struct inode *h_inode,
++ int mask);
++int au_test_h_perm_sio(struct mnt_idmap *h_idmap, struct inode *h_inode,
++ int mask);
+
-+ /*
-+ * patterns
-+ * - gparent[SRC] is parent[DST]
-+ * - parent[SRC] is gparent[DST]
-+ * - gparent[SRC] is gparent[DST]
-+ * - gparent[SRC] is a descendant of parent[DST]
-+ * - parent[SRC] is an ancestor of gparent[DST]
-+ * - not within grand parent range
-+ */
-+ err = 0;
-+ h_trap = ERR_PTR(-EBUSY);
-+ if (a->gparent[SRC] == a->parent[DST]) {
-+ //AuDbg("here\n");
-+ LKTRLabel(here);
-+ au_ren_pin_init(a->pin + DST, a->dentry, a->pin + SRC,
-+ a->src_dentry, a->btgt);
-+ h_trap = au_ren_pin3(DST, SRC, a);
-+ if (!IS_ERR(h_trap)) {
-+ h_gdir = au_pinned_h_dir(a->pin + DST);
-+ err = au_verify_parent(a->h_parent[SRC], h_gdir);
-+ if (unlikely(err))
-+ h_trap = ERR_PTR(-EBUSY);
-+ }
-+ } else if (a->parent[SRC] == a->gparent[DST] || same_gp) {
-+ //AuDbg("here\n");
-+ LKTRLabel(here);
-+ au_ren_pin_init(a->pin + SRC, a->src_dentry, a->pin + DST,
-+ a->dentry, a->btgt);
-+ h_trap = au_ren_pin3(SRC, DST, a);
-+ if (!IS_ERR(h_trap)) {
-+ if (!same_gp)
-+ h_gdir = au_pinned_h_dir(a->pin + SRC);
-+ else
-+ h_gdir = au_pinned_h_gdir(a->pin + SRC);
-+ err = au_verify_parent(a->h_parent[DST], h_gdir);
-+ if (unlikely(err))
-+ h_trap = ERR_PTR(-EBUSY);
-+ }
-+ } else if (a->gparent[SRC]
-+ && (h_trap = au_test_subdir(a->gparent[SRC],
-+ a->parent[DST]))) {
-+ //AuDbg("here\n");
-+ LKTRLabel(here);
-+ au_ren_pin_init(a->pin + DST, a->dentry, a->pin + SRC,
-+ a->src_dentry, a->btgt);
-+ if (a->gparent[DST]) {
-+ err = au_ren_pin4(DST, SRC, a);
-+ if (unlikely(err))
-+ h_trap = ERR_PTR(err);
-+ } else {
-+ struct dentry *t;
-+ t = au_ren_pin3(DST, SRC, a);
-+ AuDebugOn(t == h_trap);
-+ }
-+#if 1
-+ } else if (a->gparent[DST]
-+ && (h_trap = au_test_subdir(a->gparent[DST],
-+ a->parent[SRC]))) {
-+ /* todo: is this really necessary? */
-+ //AuDbg("here\n");
-+ LKTRLabel(here);
-+ au_ren_pin_init(a->pin + SRC, a->src_dentry, a->pin + DST,
-+ a->dentry, a->btgt);
-+ if (a->gparent[SRC]) {
-+ err = au_ren_pin4(SRC, DST, a);
-+ if (unlikely(err))
-+ h_trap = ERR_PTR(err);
-+ } else {
-+ struct dentry *t;
-+ t = au_ren_pin3(SRC, DST, a);
-+ AuDebugOn(t == h_trap);
-+ }
++static inline int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex,
++ ino_t h_ino, unsigned int d_type, ino_t *ino)
++{
++#ifdef CONFIG_AUFS_SHWH
++ return au_ino(sb, bindex, h_ino, d_type, ino);
++#else
++ return 0;
+#endif
-+ } else /* if (a->gparent[DST]
-+ && (h_trap = au_test_subdir(a->gparent[DST],
-+ a->parent[SRC]))) */ {
-+ //AuDbg("here\n");
-+ LKTRLabel(here);
-+ h_trap = NULL;
-+ if (a->gparent[DST])
-+ h_trap = au_test_subdir(a->gparent[DST],
-+ a->parent[SRC]);
-+ au_ren_pin_init(a->pin + SRC, a->src_dentry, a->pin + DST,
-+ a->dentry, a->btgt);
-+ err = au_ren_pin4(SRC, DST, a);
-+ if (unlikely(err))
-+ h_trap = ERR_PTR(err);
-+ }
-+ au_fset_ren(a->flags, PINNED);
++}
+
-+ if (!IS_ERR(h_trap)) {
-+ err = 0;
-+ for (i = 0; !err && i < 2; i++) {
-+ h_gdir = au_pinned_h_gdir(a->pin + i);
-+ if (h_gdir)
-+ err = au_verify_parent(a->h_parent[i], h_gdir);
-+ }
-+ if (unlikely(err)) {
-+ h_trap = ERR_PTR(err);
-+ //AuDbg("here\n");
-+ }
-+ }
++/* i_op.c */
++enum {
++ AuIop_SYMLINK,
++ AuIop_DIR,
++ AuIop_OTHER,
++ AuIop_Last
++};
++extern struct inode_operations aufs_iop[AuIop_Last], /* not const */
++ aufs_iop_nogetattr[AuIop_Last];
+
-+ dput(a->gparent[SRC]);
-+ dput(a->gparent[DST]);
-+ memset(a->gparent, 0, sizeof(a->gparent));
-+ AuTraceErrPtr(h_trap);
-+ //lktr_clear_pid(current->pid, LktrArrayPid);
-+ return h_trap;
-+}
++/* au_wr_dir flags */
++#define AuWrDir_ADD_ENTRY 1
++#define AuWrDir_ISDIR (1 << 1)
++#define AuWrDir_TMPFILE (1 << 2)
++#define au_ftest_wrdir(flags, name) ((flags) & AuWrDir_##name)
++#define au_fset_wrdir(flags, name) \
++ do { (flags) |= AuWrDir_##name; } while (0)
++#define au_fclr_wrdir(flags, name) \
++ do { (flags) &= ~AuWrDir_##name; } while (0)
+
-+static void au_ren_unlock(struct au_ren_args *a)
-+{
-+ int i;
++struct au_wr_dir_args {
++ aufs_bindex_t force_btgt;
++ unsigned char flags;
++};
++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
++ struct au_wr_dir_args *args);
+
-+ AuTraceEnter();
++struct dentry *au_pinned_h_parent(struct au_pin *pin);
++void au_pin_init(struct au_pin *pin, struct dentry *dentry,
++ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
++ unsigned int udba, unsigned char flags);
++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int udba, unsigned char flags) __must_check;
++int au_do_pin(struct au_pin *pin) __must_check;
++void au_unpin(struct au_pin *pin);
++int au_reval_for_attr(struct dentry *dentry, unsigned int sigen);
++
++#define AuIcpup_DID_CPUP 1
++#define au_ftest_icpup(flags, name) ((flags) & AuIcpup_##name)
++#define au_fset_icpup(flags, name) \
++ do { (flags) |= AuIcpup_##name; } while (0)
++#define au_fclr_icpup(flags, name) \
++ do { (flags) &= ~AuIcpup_##name; } while (0)
+
-+ if (au_ftest_ren(a->flags, MNT_WRITE))
-+ au_mnt_drop_write(au_sbr_mnt(a->dentry->d_sb, a->btgt));
-+ if (a->h_locked[0])
-+ vfsub_unlock_rename(a->h_locked[0], a->h_locked[1]);
-+ if (au_ftest_ren(a->flags, PINNED)) {
-+ au_unpin(a->pin + SRC);
-+ au_unpin(a->pin + DST);
-+ memset(a->gparent, 0, sizeof(a->gparent));
-+ }
-+ if (au_ftest_ren(a->flags, VFSLOCK))
-+ vfsub_unlock_rename_mutex(au_sbr_sb(a->dentry->d_sb, a->btgt));
-+ for (i = 0; i < 2; i++)
-+ if (a->gparent[i]) {
-+ di_read_unlock(a->gparent[i], AuLock_IR);
-+ dput(a->gparent[i]);
-+ }
-+}
++struct au_icpup_args {
++ unsigned char flags;
++ unsigned char pin_flags;
++ aufs_bindex_t btgt;
++ unsigned int udba;
++ struct au_pin pin;
++ struct path h_path;
++ struct inode *h_inode;
++};
+
-+static int au_ren_lock(struct au_ren_args *a)
-+{
-+ int err;
-+ const int hinotify = au_opt_test(a->mnt_flags, UDBA_INOTIFY);
++int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
++ struct au_icpup_args *a);
+
-+ AuTraceEnter();
++int au_h_path_getattr(struct dentry *dentry, struct inode *inode, int force,
++ struct path *h_path, int locked);
+
-+ err = 0;
-+ if (!hinotify
-+ || (au_ftest_ren(a->flags, ISSAMEDIR) && IS_ROOT(a->parent[SRC]))) {
-+ au_ren_pin_init(a->pin + SRC, a->src_dentry, a->pin + DST,
-+ a->dentry, a->btgt);
-+ LKTRLabel(here);
-+ a->h_locked[0] = a->h_parent[SRC];
-+ a->h_locked[1] = a->h_parent[DST];
-+ a->h_trap = vfsub_lock_rename(a->h_locked[0], a->h_locked[1]);
-+ au_ren_fake_pin(a);
-+ } else if (au_ftest_ren(a->flags, ISSAMEDIR)
-+ && !IS_ROOT(a->parent[SRC])) {
-+ /* this and next block should not be compiled when
-+ hinotify is disabled */
-+ /* irregular/tricky rename lock */
-+ LKTRLabel(here);
-+ au_ren_pin_init(a->pin + SRC, a->src_dentry, a->pin + DST,
-+ a->dentry, a->btgt);
-+ a->gparent[SRC] = dget_parent(a->parent[SRC]);
-+ di_read_lock_parent2(a->gparent[SRC], AuLock_IR);
-+ a->h_locked[0] = a->h_parent[SRC];
-+ a->h_locked[1] = dget_parent(a->h_parent[SRC]);
-+ a->h_trap = vfsub_lock_rename(a->h_locked[0], a->h_locked[1]);
-+ err = au_verify_parent(a->h_parent[SRC],
-+ a->h_locked[1]->d_inode);
-+ dput(a->h_locked[1]);
-+ if (!err)
-+ au_ren_fake_pin(a);
-+ } else {
-+ /* 3 or 4 dir locks. crazy */
-+ LKTRLabel(here);
-+ a->h_trap = au_ren_pin(a);
-+ if (IS_ERR(a->h_trap))
-+ err = PTR_ERR(a->h_trap);
-+ }
-+
-+ if (!err && au_dbstart(a->src_dentry) == a->btgt)
-+ err = au_verify_parent(a->h_dentry[SRC],
-+ a->h_parent[SRC]->d_inode);
-+ if (!err && au_dbstart(a->dentry) == a->btgt)
-+ err = au_verify_parent(a->h_dentry[DST],
-+ a->h_parent[DST]->d_inode);
-+ if (!err) {
-+ err = au_br_want_write(au_sbr(a->dentry->d_sb, a->btgt));
-+ if (unlikely(err))
-+ goto out_unlock;
-+ au_fset_ren(a->flags, MNT_WRITE);
-+ goto out; /* success */
-+ }
++/* i_op_add.c */
++int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir);
++int aufs_mknod(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode, dev_t dev);
++int aufs_symlink(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, const char *symname);
++int aufs_create(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode, bool want_excl);
++struct vfsub_aopen_args;
++int au_aopen_or_create(struct inode *dir, struct dentry *dentry,
++ struct vfsub_aopen_args *args);
++int aufs_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
++ struct file *file, umode_t mode);
++int aufs_link(struct dentry *src_dentry, struct inode *dir,
++ struct dentry *dentry);
++int aufs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode);
+
-+ err = -EBUSY;
++/* i_op_del.c */
++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup);
++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir);
++int aufs_unlink(struct inode *dir, struct dentry *dentry);
++int aufs_rmdir(struct inode *dir, struct dentry *dentry);
+
-+ out_unlock:
-+ au_ren_unlock(a);
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++/* i_op_ren.c */
++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt);
++int aufs_rename(struct mnt_idmap *idmap,
++ struct inode *_src_dir, struct dentry *_src_dentry,
++ struct inode *_dst_dir, struct dentry *_dst_dentry,
++ unsigned int _flags);
+
-+int aufs_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry)
-+{
-+ int err;
-+ aufs_bindex_t bend, bindex;
-+ unsigned char do_dt_dstdir, hinotify;
-+ struct inode *inode[2];
-+ enum { PARENT, CHILD };
-+ /* reduce stack space */
-+ struct {
-+ struct au_ren_args a;
-+ struct au_dtime dt[2][2];
-+ } *p;
-+ struct au_wr_dir_args wr_dir_args = {
-+ /* .force_btgt = -1, */
-+ .flags = AuWrDir_ADD_ENTRY
-+ };
-+
-+ LKTRTrace("i%lu, %.*s, i%lu, %.*s\n",
-+ src_dir->i_ino, AuDLNPair(src_dentry),
-+ dir->i_ino, AuDLNPair(dentry));
-+ AuDebugOn(IS_ROOT(src_dentry) || IS_ROOT(dentry));
-+ IMustLock(src_dir);
-+ IMustLock(dir);
-+ inode[DST] = dentry->d_inode;
-+ if (inode[DST]) {
-+ IMustLock(inode[DST]);
-+ au_igrab(inode[DST]);
-+ }
-+
-+ err = -ENOMEM;
-+ BUILD_BUG_ON(sizeof(*p) > PAGE_SIZE);
-+ p = kzalloc(sizeof(*p), GFP_NOFS);
-+ if (unlikely(!p))
-+ goto out;
-+
-+ err = -ENOTDIR;
-+ p->a.src_dir = src_dir;
-+ p->a.src_dentry = src_dentry;
-+ p->a.dir = dir;
-+ p->a.dentry = dentry;
-+ p->a.sb = src_dentry->d_sb;
-+ inode[SRC] = src_dentry->d_inode;
-+ p->a.flags = 0;
-+ if (S_ISDIR(inode[SRC]->i_mode)) {
-+ au_fset_ren(p->a.flags, ISDIR);
-+ if (unlikely(inode[DST] && !S_ISDIR(inode[DST]->i_mode)))
-+ goto out_free;
-+ aufs_read_and_write_lock2(dentry, src_dentry, AuLock_DIR);
-+ } else
-+ aufs_read_and_write_lock2(dentry, src_dentry, 0);
++/* iinfo.c */
++struct inode *au_h_iptr(struct inode *inode, aufs_bindex_t bindex);
++void au_hiput(struct au_hinode *hinode);
++void au_set_hi_wh(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_wh);
++unsigned int au_hi_flags(struct inode *inode, int isdir);
+
-+ p->a.mnt_flags = au_mntflags(p->a.sb);
-+ if (au_test_dlgt(p->a.mnt_flags))
-+ au_fset_ren(p->a.flags, DLGT);
-+ p->a.parent[SRC] = src_dentry->d_parent; /* dir inode is locked */
-+ p->a.parent[DST] = dentry->d_parent; /* dir inode is locked */
-+ au_fset_ren(p->a.flags, ISSAMEDIR); /* temporary */
-+ di_write_lock_parent(p->a.parent[DST]);
++/* hinode flags */
++#define AuHi_XINO 1
++#define AuHi_HNOTIFY (1 << 1)
++#define au_ftest_hi(flags, name) ((flags) & AuHi_##name)
++#define au_fset_hi(flags, name) \
++ do { (flags) |= AuHi_##name; } while (0)
++#define au_fclr_hi(flags, name) \
++ do { (flags) &= ~AuHi_##name; } while (0)
+
-+ /* which branch we process */
-+ p->a.bstart[SRC] = au_dbstart(src_dentry);
-+ p->a.bstart[DST] = au_dbstart(dentry);
-+ if (au_ftest_ren(p->a.flags, ISDIR))
-+ au_fset_wrdir(wr_dir_args.flags, ISDIR);
-+ wr_dir_args.force_btgt = p->a.bstart[SRC];
-+ if (dentry->d_inode && p->a.bstart[DST] < p->a.bstart[SRC])
-+ wr_dir_args.force_btgt = p->a.bstart[DST];
-+ wr_dir_args.force_btgt = au_wbr(dentry, wr_dir_args.force_btgt);
-+ err = au_wr_dir(dentry, src_dentry, &wr_dir_args);
-+ p->a.btgt = err;
-+ if (unlikely(err < 0))
-+ goto out_unlock;
++#ifndef CONFIG_AUFS_HNOTIFY
++#undef AuHi_HNOTIFY
++#define AuHi_HNOTIFY 0
++#endif
+
-+ /* are they available to be renamed */
-+ err = 0;
-+ au_nhash_init(&p->a.whlist);
-+ if (au_ftest_ren(p->a.flags, ISDIR) && inode[DST]) {
-+ au_set_dbstart(dentry, p->a.bstart[DST]);
-+ err = may_rename_dstdir(dentry, p->a.btgt, &p->a.whlist);
-+ au_set_dbstart(dentry, p->a.btgt);
-+ }
-+ p->a.h_dentry[DST] = au_h_dptr(dentry, au_dbstart(dentry));
-+ if (unlikely(err))
-+ goto out_unlock;
-+ /* todo: minor optimize,
-+ their sb may be same while their bindex differs? */
-+ p->a.h_dentry[SRC] = au_h_dptr(src_dentry, au_dbstart(src_dentry));
-+ if (au_ftest_ren(p->a.flags, ISDIR)) {
-+ err = may_rename_srcdir(src_dentry, p->a.btgt);
-+ if (unlikely(err))
-+ goto out_children;
-+ }
++void au_set_h_iptr(struct inode *inode, aufs_bindex_t bindex,
++ struct inode *h_inode, unsigned int flags);
+
-+ /* prepare the writable parent dir on the same branch */
-+ if (p->a.bstart[DST] == p->a.btgt) {
-+ au_fset_ren(p->a.flags, WHDST);
-+ } else {
-+ err = au_cpup_dirs(dentry, p->a.btgt);
-+ if (unlikely(err))
-+ goto out_children;
-+ }
++void au_update_iigen(struct inode *inode, int half);
++void au_update_ibrange(struct inode *inode, int do_put_zero);
+
-+ if (src_dir != dir) {
-+ /*
-+ * this temporary unlock is safe,
-+ * because both dir->i_mutex are locked.
-+ */
-+ di_write_unlock(p->a.parent[DST]);
-+ di_write_lock_parent(p->a.parent[SRC]);
-+ err = au_wr_dir_need_wh
-+ (src_dentry, au_ftest_ren(p->a.flags, ISDIR),
-+ &p->a.btgt);
-+ di_write_unlock(p->a.parent[SRC]);
-+ di_write_lock2_parent(p->a.parent[SRC], p->a.parent[DST],
-+ /*isdir*/1);
-+ au_fclr_ren(p->a.flags, ISSAMEDIR);
-+ } else
-+ err = au_wr_dir_need_wh
-+ (src_dentry, au_ftest_ren(p->a.flags, ISDIR),
-+ &p->a.btgt);
-+ if (unlikely(err < 0))
-+ goto out_children;
-+ if (err)
-+ au_fset_ren(p->a.flags, WHSRC);
++void au_icntnr_init_once(void *_c);
++void au_hinode_init(struct au_hinode *hinode);
++int au_iinfo_init(struct inode *inode);
++void au_iinfo_fin(struct inode *inode);
++int au_hinode_realloc(struct au_iinfo *iinfo, int nbr, int may_shrink);
+
-+ hinotify = au_opt_test(p->a.mnt_flags, UDBA_INOTIFY);
-+ p->a.h_parent[SRC] = au_h_dptr(p->a.parent[SRC], p->a.btgt);
-+ p->a.h_parent[DST] = au_h_dptr(p->a.parent[DST], p->a.btgt);
-+ err = au_ren_lock(&p->a);
-+ if (unlikely(err))
-+ goto out_children;
++#ifdef CONFIG_PROC_FS
++/* plink.c */
++int au_plink_maint(struct super_block *sb, int flags);
++struct au_sbinfo;
++void au_plink_maint_leave(struct au_sbinfo *sbinfo);
++int au_plink_maint_enter(struct super_block *sb);
++#ifdef CONFIG_AUFS_DEBUG
++void au_plink_list(struct super_block *sb);
++#else
++AuStubVoid(au_plink_list, struct super_block *sb)
++#endif
++int au_plink_test(struct inode *inode);
++struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex);
++void au_plink_append(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_dentry);
++void au_plink_put(struct super_block *sb, int verbose);
++void au_plink_clean(struct super_block *sb, int verbose);
++void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id);
++#else
++AuStubInt0(au_plink_maint, struct super_block *sb, int flags);
++AuStubVoid(au_plink_maint_leave, struct au_sbinfo *sbinfo);
++AuStubInt0(au_plink_maint_enter, struct super_block *sb);
++AuStubVoid(au_plink_list, struct super_block *sb);
++AuStubInt0(au_plink_test, struct inode *inode);
++AuStub(struct dentry *, au_plink_lkup, return NULL,
++ struct inode *inode, aufs_bindex_t bindex);
++AuStubVoid(au_plink_append, struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_dentry);
++AuStubVoid(au_plink_put, struct super_block *sb, int verbose);
++AuStubVoid(au_plink_clean, struct super_block *sb, int verbose);
++AuStubVoid(au_plink_half_refresh, struct super_block *sb, aufs_bindex_t br_id);
++#endif /* CONFIG_PROC_FS */
++
++#ifdef CONFIG_AUFS_XATTR
++/* xattr.c */
++int au_cpup_xattr(struct path *h_dst, struct path *h_src, int ignore_flags,
++ unsigned int verbose);
++ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size);
++void au_xattr_init(struct super_block *sb);
++#else
++AuStubInt0(au_cpup_xattr, struct path *h_dst, struct path *h_src,
++ int ignore_flags, unsigned int verbose);
++AuStubVoid(au_xattr_init, struct super_block *sb);
++#endif
+
-+ if (!au_opt_test(p->a.mnt_flags, UDBA_NONE)) {
-+ p->a.ndx.nfsmnt = au_nfsmnt(p->a.sb, p->a.btgt);
-+ if (au_ftest_ren(p->a.flags, DLGT))
-+ au_fset_ndx(p->a.ndx.flags, DLGT);
-+ err = au_may_ren(&p->a);
-+ if (unlikely(err))
-+ goto out_hdir;
-+ memset(&p->a.ndx, 0, sizeof(p->a.ndx));
-+ }
++#ifdef CONFIG_FS_POSIX_ACL
++struct posix_acl *aufs_get_inode_acl(struct inode *inode, int type, bool rcu);
++struct posix_acl *aufs_get_acl(struct mnt_idmap *idmap,
++ struct dentry *dentry, int type);
++int aufs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
++ struct posix_acl *acl, int type);
++#endif
+
-+ /* store timestamps to be revertible */
-+ au_dtime_store(p->dt[PARENT] + SRC, p->a.parent[SRC],
-+ p->a.h_parent[SRC], au_pinned_hdir(p->a.pin + SRC),
-+ au_pinned_hgdir(p->a.pin + SRC)
-+ /* hgdir[SRC] */);
-+ if (!au_ftest_ren(p->a.flags, ISSAMEDIR))
-+ au_dtime_store(p->dt[PARENT] + DST, p->a.parent[DST],
-+ p->a.h_parent[DST],
-+ au_pinned_hdir(p->a.pin + DST),
-+ au_pinned_hgdir(p->a.pin + DST)
-+ /* hgdir[DST] */);
-+ do_dt_dstdir = 0;
-+ if (au_ftest_ren(p->a.flags, ISDIR)) {
-+ au_dtime_store(p->dt[CHILD] + SRC, src_dentry,
-+ p->a.h_dentry[SRC], au_hi(inode[SRC], p->a.btgt),
-+ au_pinned_hdir(p->a.pin + SRC));
-+ if (p->a.h_dentry[DST]->d_inode) {
-+ do_dt_dstdir = 1;
-+ au_dtime_store(p->dt[CHILD] + DST, dentry,
-+ p->a.h_dentry[DST],
-+ au_hi(inode[DST], p->a.btgt),
-+ au_pinned_hdir(p->a.pin + DST));
-+ }
-+ }
-+
-+ err = do_rename(&p->a);
-+ if (unlikely(err))
-+ goto out_dt;
++#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
++enum {
++ AU_XATTR_SET,
++ AU_ACL_SET
++};
+
-+ /* update dir attributes */
-+ dir->i_version++;
-+ if (au_ftest_ren(p->a.flags, ISDIR)) {
-+ /* is this updating defined in POSIX? */
-+ /* mutex_lock(&inode[SRC]->i_mutex); */
-+ au_cpup_attr_timesizes(inode[SRC]);
-+ /* mutex_unlock(&inode[SRC]->i_mutex); */
++struct au_sxattr {
++ int type;
++ union {
++ struct {
++ const char *name;
++ const void *value;
++ size_t size;
++ int flags;
++ } set;
++ struct {
++ struct posix_acl *acl;
++ int type;
++ } acl_set;
++ } u;
++};
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg);
++#endif
+
-+ au_cpup_attr_nlink(dir, /*force*/1);
-+ if (inode[DST]) {
-+ clear_nlink(inode[DST]);
-+ au_cpup_attr_timesizes(inode[DST]);
-+ }
-+ }
-+ if (au_ibstart(dir) == p->a.btgt)
-+ au_cpup_attr_timesizes(dir);
++/* ---------------------------------------------------------------------- */
+
-+ if (!au_ftest_ren(p->a.flags, ISSAMEDIR)) {
-+ src_dir->i_version++;
-+ if (au_ftest_ren(p->a.flags, ISDIR))
-+ au_cpup_attr_nlink(src_dir, /*force*/1);
-+ if (au_ibstart(src_dir) == p->a.btgt)
-+ au_cpup_attr_timesizes(src_dir);
-+ }
++/* lock subclass for iinfo */
++enum {
++ AuLsc_II_CHILD, /* child first */
++ AuLsc_II_CHILD2, /* rename(2), link(2), and cpup at hnotify */
++ AuLsc_II_CHILD3, /* copyup dirs */
++ AuLsc_II_PARENT, /* see AuLsc_I_PARENT in vfsub.h */
++ AuLsc_II_PARENT2,
++ AuLsc_II_PARENT3, /* copyup dirs */
++ AuLsc_II_NEW_CHILD
++};
+
-+ /* todo: simple d_drop(src_dentry) is not enough? */
-+ /* dput/iput all lower dentries */
-+ au_set_dbwh(src_dentry, -1);
-+ bend = au_dbend(src_dentry);
-+ for (bindex = p->a.btgt + 1; bindex <= bend; bindex++) {
-+ struct dentry *hd;
-+ hd = au_h_dptr(src_dentry, bindex);
-+ if (hd)
-+ au_set_h_dptr(src_dentry, bindex, NULL);
-+ }
-+ au_set_dbend(src_dentry, p->a.btgt);
-+
-+ if (au_opt_test(p->a.mnt_flags, PLINK)
-+ && au_plink_test(src_dentry->d_sb, inode[SRC]))
-+ goto out_hdir; /* success */
-+ bend = au_ibend(inode[SRC]);
-+ for (bindex = p->a.btgt + 1; bindex <= bend; bindex++) {
-+ struct inode *hi;
-+ hi = au_h_iptr(inode[SRC], bindex);
-+ if (hi) {
-+ au_xino_write0(p->a.sb, bindex, hi->i_ino, 0);
-+ /* ignore this error */
-+ au_set_h_iptr(inode[SRC], bindex, NULL, 0);
-+ }
-+ }
-+ au_set_ibend(inode[SRC], p->a.btgt);
-+ goto out_hdir; /* success */
++/*
++ * ii_read_lock_child, ii_write_lock_child,
++ * ii_read_lock_child2, ii_write_lock_child2,
++ * ii_read_lock_child3, ii_write_lock_child3,
++ * ii_read_lock_parent, ii_write_lock_parent,
++ * ii_read_lock_parent2, ii_write_lock_parent2,
++ * ii_read_lock_parent3, ii_write_lock_parent3,
++ * ii_read_lock_new_child, ii_write_lock_new_child,
++ */
++#define AuReadLockFunc(name, lsc) \
++static inline void ii_read_lock_##name(struct inode *i) \
++{ \
++ au_rw_read_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
++}
+
-+ out_dt:
-+ au_dtime_revert(p->dt[PARENT] + SRC);
-+ if (!au_ftest_ren(p->a.flags, ISSAMEDIR))
-+ au_dtime_revert(p->dt[PARENT] + DST);
-+ if (au_ftest_ren(p->a.flags, ISDIR) && err != -EIO) {
-+ struct dentry *hd;
-+
-+ hd = p->dt[CHILD][SRC].dt_h_dentry;
-+ mutex_lock_nested(&hd->d_inode->i_mutex, AuLsc_I_CHILD);
-+ au_dtime_revert(p->dt[CHILD] + SRC);
-+ mutex_unlock(&hd->d_inode->i_mutex);
-+ if (do_dt_dstdir) {
-+ hd = p->dt[CHILD][DST].dt_h_dentry;
-+ mutex_lock_nested(&hd->d_inode->i_mutex, AuLsc_I_CHILD);
-+ au_dtime_revert(p->dt[CHILD] + DST);
-+ mutex_unlock(&hd->d_inode->i_mutex);
-+ }
-+ }
-+ out_hdir:
-+ au_ren_unlock(&p->a);
-+ out_children:
-+ au_nhash_fin(&p->a.whlist);
-+ out_unlock:
-+ if (unlikely(err && au_ftest_ren(p->a.flags, ISDIR))) {
-+ au_update_dbstart(dentry);
-+ d_drop(dentry);
-+ }
-+ if (!err) {
-+ d_move(src_dentry, dentry);
-+ if (inode[DST]
-+ && (inode[DST]->i_nlink == 1
-+ || au_ftest_ren(p->a.flags, ISDIR)))
-+ inode[DST]->i_flags |= S_DEAD;
-+ }
-+ if (au_ftest_ren(p->a.flags, ISSAMEDIR))
-+ di_write_unlock(p->a.parent[DST]);
-+ else
-+ di_write_unlock2(p->a.parent[SRC], p->a.parent[DST]);
-+ aufs_read_and_write_unlock2(dentry, src_dentry);
-+ out_free:
-+ kfree(p);
-+ out:
-+ iput(inode[DST]);
-+ AuTraceErr(err);
-+ //lktr_clear_pid(current->pid, LktrArrayPid);
-+ if (unlikely(err == -EBUSY && au_test_nfsd(current)))
-+ err = -ESTALE;
-+ return err;
++#define AuWriteLockFunc(name, lsc) \
++static inline void ii_write_lock_##name(struct inode *i) \
++{ \
++ au_rw_write_lock_nested(&au_ii(i)->ii_rwsem, AuLsc_II_##lsc); \
+}
-diff -urN linux/fs/aufs/Makefile linux-aufs/fs/aufs/Makefile
---- linux/fs/aufs/Makefile 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/Makefile 2008-10-20 03:47:24.000000000 +0200
-@@ -0,0 +1,74 @@
-+# AUFS Makefile for the Linux 2.6.25 and later
-+# $Id$
+
-+# the environment variables are not inherited since 2.6.23
-+ifdef AUFS_EXTRA_CFLAGS
-+ccflags-y += ${AUFS_EXTRA_CFLAGS}
-+endif
++#define AuRWLockFuncs(name, lsc) \
++ AuReadLockFunc(name, lsc) \
++ AuWriteLockFunc(name, lsc)
+
-+########################################
++AuRWLockFuncs(child, CHILD);
++AuRWLockFuncs(child2, CHILD2);
++AuRWLockFuncs(child3, CHILD3);
++AuRWLockFuncs(parent, PARENT);
++AuRWLockFuncs(parent2, PARENT2);
++AuRWLockFuncs(parent3, PARENT3);
++AuRWLockFuncs(new_child, NEW_CHILD);
+
-+ifdef CONFIG_AUFS_RR_SQUASHFS
-+# cf. squashfs3.2-r2 and sqlzma patch.
-+ccflags-y += -DSQUASHFS_MAGIC=0x73717368
-+ccflags-y += -DSQUASHFS_MAGIC_SWAP=0x68737173
-+ccflags-y += -DSQUASHFS_MAGIC_LZMA=0x71736873
-+ccflags-y += -DSQUASHFS_MAGIC_LZMA_SWAP=0x73687371
-+endif
++#undef AuReadLockFunc
++#undef AuWriteLockFunc
++#undef AuRWLockFuncs
+
-+# defined in ${srctree}/fs/fuse/inode.c
-+ccflags-$(CONFIG_AUFS_WORKAROUND_FUSE) += -DFUSE_SUPER_MAGIC=0x65735546
++#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)
+
-+# defined in ${srctree}/fs/xfs/xfs_sb.h
-+# tristate
-+ifdef CONFIG_XFS_FS
-+ccflags-y += -DXFS_SB_MAGIC=0x58465342
-+endif
++#define IiMustNoWaiters(i) AuRwMustNoWaiters(&au_ii(i)->ii_rwsem)
++#define IiMustAnyLock(i) AuRwMustAnyLock(&au_ii(i)->ii_rwsem)
++#define IiMustWriteLock(i) AuRwMustWriteLock(&au_ii(i)->ii_rwsem)
+
-+# defined in ${srctree}/mm/shmem.c
-+# tristate
-+ifdef CONFIG_TMPFS
-+ccflags-y += -DTMPFS_MAGIC=0x01021994
-+endif
++/* ---------------------------------------------------------------------- */
+
-+# defined in ${srctree}fs/sysfs/mount.c
-+# bool
-+ccflags-$(CONFIG_SYSFS) += -DSYSFS_MAGIC=0x62656572
++static inline void au_icntnr_init(struct au_icntnr *c)
++{
++#ifdef CONFIG_AUFS_DEBUG
++ c->vfs_inode.i_mode = 0;
++#endif
++}
+
-+ifndef EXTRAVERSION
-+EXTRAVERSION = $(shell echo ${KERNELVERSION} | cut -f3- -d. | cut -f2- -d-)
-+endif
-+# for -mm tree, support the latest version only
-+ifneq ($(strip $(shell echo ${EXTRAVERSION} | fgrep -- mm)),)
-+ccflags-y += -DCONFIG_AUFS_UNIONFS22_PATCH -DCONFIG_AUFS_UNIONFS23_PATCH
-+endif
++static inline unsigned int au_iigen(struct inode *inode, unsigned int *igflags)
++{
++ unsigned int gen;
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
+
-+-include $(dir $(lastword ${MAKEFILE_LIST}))priv.mk
-+#$(warning ${ccflags-y})
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ if (igflags)
++ *igflags = iigen->ig_flags;
++ gen = iigen->ig_generation;
++ spin_unlock(&iigen->ig_spin);
+
-+########################################
++ return gen;
++}
+
-+obj-$(CONFIG_AUFS) += aufs.o
-+aufs-y := module.o super.o sbinfo.o branch.o xino.o sysaufs.o opts.o \
-+ wkq.o vfsub.o dcsub.o \
-+ cpup.o whout.o plink.o wbr_policy.o \
-+ dentry.o dinfo.o \
-+ file.o f_op.o finfo.o \
-+ dir.o vdir.o \
-+ inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o iinfo.o \
-+ misc.o
++/* tiny test for inode number */
++/* tmpfs generation is too rough */
++static inline int au_test_higen(struct inode *inode, struct inode *h_inode)
++{
++ struct au_iinfo *iinfo;
+
-+aufs-$(CONFIG_SYSFS) += sysfs.o
-+aufs-$(CONFIG_AUFS_BR_NFS) += br_nfs.o
-+aufs-$(CONFIG_AUFS_BR_XFS) += br_xfs.o
-+aufs-$(CONFIG_AUFS_WORKAROUND_FUSE) += br_fuse.o
-+aufs-$(CONFIG_AUFS_DLGT) += dlgt.o
-+aufs-$(CONFIG_AUFS_HINOTIFY) += hinotify.o
-+aufs-$(CONFIG_AUFS_HIN_OR_DLGT) += hin_or_dlgt.o
-+aufs-$(CONFIG_AUFS_GETATTR) += getattr.o
-+aufs-$(CONFIG_AUFS_EXPORT) += export.o
-+aufs-$(CONFIG_AUFS_ROBR) += robr.o
-+# reserved for future use
-+#aufs-$(CONFIG_AUFS_XATTR) += xattr.o
-+aufs-$(CONFIG_AUFS_DEBUG) += debug.o
-+aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o
-diff -urN linux/fs/aufs/misc.c linux-aufs/fs/aufs/misc.c
---- linux/fs/aufs/misc.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/misc.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,308 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ iinfo = au_ii(inode);
++ AuRwMustAnyLock(&iinfo->ii_rwsem);
++ return !(iinfo->ii_hsb1 == h_inode->i_sb
++ && iinfo->ii_higen == h_inode->i_generation);
++}
+
-+/*
-+ * $Id$
-+ */
++static inline void au_iigen_dec(struct inode *inode)
++{
++ struct au_iinfo *iinfo;
++ struct au_iigen *iigen;
+
-+#include "aufs.h"
++ iinfo = au_ii(inode);
++ iigen = &iinfo->ii_generation;
++ spin_lock(&iigen->ig_spin);
++ iigen->ig_generation--;
++ spin_unlock(&iigen->ig_spin);
++}
+
-+void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp)
++static inline int au_iigen_test(struct inode *inode, unsigned int sigen)
+{
-+ void *q;
++ int err;
+
-+ LKTRTrace("p %p, nused %d, sz %d\n", p, nused, new_sz);
-+ AuDebugOn(new_sz <= 0);
-+ if (new_sz <= nused)
-+ return p;
++ err = 0;
++ if (unlikely(inode && au_iigen(inode, NULL) != sigen))
++ err = -EIO;
+
-+ q = krealloc(p, new_sz, gfp);
-+ if (q)
-+ memset(q + nused, 0, new_sz - nused);
-+ return q;
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct nameidata *au_dup_nd(struct au_sbinfo *sbinfo, struct nameidata *dst,
-+ struct nameidata *src)
++static inline struct au_hinode *au_hinode(struct au_iinfo *iinfo,
++ aufs_bindex_t bindex)
+{
-+ LKTRTrace("src %p\n", src);
-+
-+ if (src) {
-+ *dst = *src;
-+ dst->flags &= ~LOOKUP_PARENT;
-+ if (sbinfo->si_wbr_create == AuWbrCreate_TDP) {
-+ if ((dst->flags & LOOKUP_CREATE)
-+ && !(dst->intent.open.flags & O_CREAT))
-+ dst->flags &= ~LOOKUP_CREATE;
-+ } else {
-+ dst->flags &= ~LOOKUP_CREATE;
-+ dst->intent.open.flags &= ~O_CREAT;
-+ }
-+ } else
-+ dst = NULL;
-+
-+ return dst;
++ return iinfo->ii_hinode + bindex;
+}
+
-+struct nameidata *au_fake_dm(struct nameidata *fake_nd, struct nameidata *nd,
-+ struct super_block *sb, aufs_bindex_t bindex)
++static inline int au_is_bad_inode(struct inode *inode)
+{
-+ LKTRTrace("nd %p, b%d\n", nd, bindex);
-+
-+ if (!nd)
-+ return NULL;
-+
-+ DiMustAnyLock(nd->path.dentry);
-+
-+ fake_nd->path.dentry = NULL;
-+ fake_nd->path.mnt = NULL;
-+
-+ if (bindex <= au_dbend(nd->path.dentry))
-+ fake_nd->path.dentry = au_h_dptr(nd->path.dentry, bindex);
-+ if (fake_nd->path.dentry) {
-+ fake_nd->path.mnt = au_sbr_mnt(sb, bindex);
-+ AuDebugOn(!fake_nd->path.mnt);
-+ path_get(&fake_nd->path);
-+ } else
-+ fake_nd = ERR_PTR(-ENOENT);
-+
-+ AuTraceErrPtr(fake_nd);
-+ return fake_nd;
++ return !!(is_bad_inode(inode) || !au_hinode(au_ii(inode), 0));
+}
+
-+void au_fake_dm_release(struct nameidata *fake_nd)
++static inline aufs_bindex_t au_ii_br_id(struct inode *inode,
++ aufs_bindex_t bindex)
+{
-+ if (fake_nd)
-+ path_put(&fake_nd->path);
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex)->hi_id;
+}
+
-+int au_h_create(struct inode *h_dir, struct dentry *h_dentry, int mode,
-+ struct vfsub_args *vargs, struct nameidata *nd,
-+ struct vfsmount *nfsmnt)
++static inline aufs_bindex_t au_ibtop(struct inode *inode)
+{
-+ int err;
-+
-+ LKTRTrace("hi%lu, %.*s, 0%o, nd %d, nfsmnt %d\n",
-+ h_dir->i_ino, AuDLNPair(h_dentry), mode, !!nd, !!nfsmnt);
-+
-+ err = -ENOSYS;
-+ if (!nfsmnt)
-+ err = vfsub_create(h_dir, h_dentry, mode, /*nd*/NULL, vargs);
-+ else {
-+ struct nameidata fake_nd;
-+
-+ if (nd)
-+ fake_nd = *nd;
-+ else
-+ memset(&fake_nd, 0, sizeof(fake_nd));
-+ fake_nd.path.dentry = h_dentry;
-+ fake_nd.path.mnt = nfsmnt;
-+ path_get(&fake_nd.path);
-+ fake_nd.flags = LOOKUP_CREATE;
-+ fake_nd.intent.open.flags = O_CREAT | FMODE_READ;
-+ fake_nd.intent.open.create_mode = mode;
-+
-+ err = vfsub_create(h_dir, h_dentry, mode, &fake_nd, vargs);
-+ path_put(&fake_nd.path);
-+ }
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_btop;
++}
+
-+ AuTraceErr(err);
-+ return err;
++static inline aufs_bindex_t au_ibbot(struct inode *inode)
++{
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_bbot;
+}
+
-+/* ---------------------------------------------------------------------- */
++static inline struct au_vdir *au_ivdir(struct inode *inode)
++{
++ IiMustAnyLock(inode);
++ return au_ii(inode)->ii_vdir;
++}
+
-+/* empty_zero_page is not exported on PPC before 2.6.26 */
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26) \
-+ && (defined(CONFIG_PPC) || defined(CONFIG_PPC64))
-+static char *au_zp_alloc(void)
++static inline struct dentry *au_hi_wh(struct inode *inode, aufs_bindex_t bindex)
+{
-+ return (void *)get_zeroed_page(GFP_NOFS);
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex)->hi_whdentry;
+}
+
-+static void au_zp_free(char *p)
++static inline void au_set_ibtop(struct inode *inode, aufs_bindex_t bindex)
+{
-+ if (p)
-+ free_page((unsigned long)p);
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_btop = bindex;
+}
-+#else
-+static char *au_zp_alloc(void)
++
++static inline void au_set_ibbot(struct inode *inode, aufs_bindex_t bindex)
+{
-+ return page_address(ZERO_PAGE(0));
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_bbot = bindex;
+}
+
-+static void au_zp_free(char *p)
++static inline void au_set_ivdir(struct inode *inode, struct au_vdir *vdir)
+{
-+ /* empty */
++ IiMustWriteLock(inode);
++ au_ii(inode)->ii_vdir = vdir;
+}
-+#endif
+
-+int au_copy_file(struct file *dst, struct file *src, loff_t len,
-+ struct au_hinode *hdir, struct super_block *sb,
-+ struct vfsub_args *vargs)
++static inline struct au_hinode *au_hi(struct inode *inode, aufs_bindex_t bindex)
+{
-+ int err, all_zero, do_kfree;
-+ unsigned long blksize;
-+ char *buf, *zp;
-+ /* reduce stack usage */
-+ struct iattr *ia;
-+
-+ LKTRTrace("%.*s, %.*s\n",
-+ AuDLNPair(dst->f_dentry), AuDLNPair(src->f_dentry));
-+ AuDebugOn(!(dst->f_mode & FMODE_WRITE));
-+#ifdef CONFIG_AUFS_DEBUG
-+ {
-+ struct dentry *parent;
-+ parent = dget_parent(dst->f_dentry);
-+ IMustLock(parent->d_inode);
-+ dput(parent);
-+ }
-+#endif
-+
-+ err = -ENOMEM;
-+ zp = au_zp_alloc();
-+ if (unlikely(!zp))
-+ goto out;
-+ blksize = dst->f_dentry->d_sb->s_blocksize;
-+ if (!blksize || PAGE_SIZE < blksize)
-+ blksize = PAGE_SIZE;
-+ LKTRTrace("blksize %lu\n", blksize);
-+ do_kfree = (blksize != PAGE_SIZE && blksize >= sizeof(*ia));
-+ if (do_kfree)
-+ buf = kmalloc(blksize, GFP_NOFS);
-+ else
-+ buf = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!buf))
-+ goto out;
-+
-+#ifdef CONFIG_AUFS_DEBUG
-+ if (len > (1 << 22))
-+ AuWarn("copying a large file %lld\n", (long long)len);
-+#endif
-+ err = 0;
-+ all_zero = 0;
-+ src->f_pos = 0;
-+ dst->f_pos = 0;
-+ while (len) {
-+ size_t sz, rbytes, wbytes;
-+ char *p;
-+
-+ LKTRTrace("len %lld\n", len);
-+ sz = blksize;
-+ if (len < blksize)
-+ sz = len;
-+
-+ /* support LSM and notify */
-+ rbytes = 0;
-+ /* todo: signal_pending? */
-+ while (!rbytes || err == -EAGAIN || err == -EINTR) {
-+ rbytes = vfsub_read_k(src, buf, sz, &src->f_pos,
-+ vfsub_ftest(vargs->flags, DLGT));
-+ err = rbytes;
-+ }
-+ if (unlikely(err < 0))
-+ break;
-+
-+ all_zero = 0;
-+ if (len >= rbytes && rbytes == blksize) {
-+#if 1
-+ all_zero = !memcmp(buf, zp, rbytes);
-+#else /* reserved for future use */
-+ unsigned long long *ullp;
-+ size_t n, i;
-+
-+ all_zero = 1;
-+ ullp = (void *)buf;
-+ n = rbytes / sizeof(*ullp);
-+ i = n;
-+ while (n-- > 0 && all_zero)
-+ all_zero = !*ullp++;
-+ p = (void *)ullp;
-+ i *= sizeof(*ullp);
-+ for (; all_zero && i < rbytes; i++)
-+ all_zero = !*p++;
-+#endif
-+ }
-+ if (!all_zero) {
-+ wbytes = rbytes;
-+ p = buf;
-+ while (wbytes) {
-+ size_t b;
-+ /* support LSM and notify */
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_MODIFY, hdir);
-+ b = vfsub_write_k(dst, p, wbytes, &dst->f_pos,
-+ vargs);
-+ err = b;
-+ /* todo: signal_pending? */
-+ if (unlikely(err == -EAGAIN || err == -EINTR))
-+ continue;
-+ if (unlikely(err < 0))
-+ break;
-+ wbytes -= b;
-+ p += b;
-+ }
-+ } else {
-+ loff_t res;
-+ LKTRLabel(hole);
-+ res = vfsub_llseek(dst, rbytes, SEEK_CUR);
-+ err = res;
-+ if (unlikely(res < 0))
-+ break;
-+ }
-+ len -= rbytes;
-+ err = 0;
-+ }
-+
-+ /* the last block may be a hole */
-+ if (!err && all_zero) {
-+ struct dentry *h_d = dst->f_dentry;
-+ struct inode *h_i = h_d->d_inode;
-+
-+ LKTRLabel(last hole);
-+ do {
-+ /* todo: signal_pending? */
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_MODIFY, hdir);
-+ err = vfsub_write_k(dst, "\0", 1, &dst->f_pos, vargs);
-+ } while (err == -EAGAIN || err == -EINTR);
-+ if (err == 1) {
-+ ia = (void *)buf;
-+ ia->ia_size = dst->f_pos;
-+ ia->ia_valid = ATTR_SIZE | ATTR_FILE;
-+ ia->ia_file = dst;
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, vfsub_events_notify_change(ia),
-+ hdir);
-+ mutex_lock_nested(&h_i->i_mutex, AuLsc_I_CHILD2);
-+ err = vfsub_notify_change(h_d, ia, vargs);
-+ mutex_unlock(&h_i->i_mutex);
-+ }
-+ }
-+ if (do_kfree)
-+ kfree(buf);
-+ else
-+ free_page((unsigned long)buf);
-+
-+ out:
-+ au_zp_free(zp);
-+ AuTraceErr(err);
-+ return err;
-+}
-diff -urN linux/fs/aufs/misc.h linux-aufs/fs/aufs/misc.h
---- linux/fs/aufs/misc.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/misc.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,244 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
-+
-+/*
-+ * $Id$
-+ */
-+
-+#ifndef __AUFS_MISC_H__
-+#define __AUFS_MISC_H__
-+
-+#ifdef __KERNEL__
-+
-+#include <linux/fs.h>
-+#include <linux/namei.h>
-+#include <linux/aufs_type.h>
++ IiMustAnyLock(inode);
++ return au_hinode(au_ii(inode), bindex);
++}
+
+/* ---------------------------------------------------------------------- */
+
-+typedef unsigned int au_gen_t;
-+/* see linux/include/linux/jiffies.h */
-+#define AuGenYounger(a, b) ((int)(b) - (int)(a) < 0)
-+#define AuGenOlder(a, b) AufsGenYounger(b, a)
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_splhead {
-+ spinlock_t spin;
-+ struct list_head head;
-+};
++static inline struct dentry *au_pinned_parent(struct au_pin *pin)
++{
++ if (pin)
++ return pin->parent;
++ return NULL;
++}
+
-+static inline void au_spl_init(struct au_splhead *spl)
++static inline struct inode *au_pinned_h_dir(struct au_pin *pin)
+{
-+ spin_lock_init(&spl->spin);
-+ INIT_LIST_HEAD(&spl->head);
++ if (pin && pin->hdir)
++ return pin->hdir->hi_inode;
++ return NULL;
+}
+
-+static inline void au_spl_add(struct list_head *list, struct au_splhead *spl)
++static inline struct au_hinode *au_pinned_hdir(struct au_pin *pin)
+{
-+ spin_lock(&spl->spin);
-+ list_add(list, &spl->head);
-+ spin_unlock(&spl->spin);
++ if (pin)
++ return pin->hdir;
++ return NULL;
+}
+
-+static inline void au_spl_del(struct list_head *list, struct au_splhead *spl)
++static inline void au_pin_set_dentry(struct au_pin *pin, struct dentry *dentry)
+{
-+ spin_lock(&spl->spin);
-+ list_del(list);
-+ spin_unlock(&spl->spin);
++ if (pin)
++ pin->dentry = dentry;
+}
+
-+/* ---------------------------------------------------------------------- */
++static inline void au_pin_set_parent_lflag(struct au_pin *pin,
++ unsigned char lflag)
++{
++ if (pin) {
++ if (lflag)
++ au_fset_pin(pin->flags, DI_LOCKED);
++ else
++ au_fclr_pin(pin->flags, DI_LOCKED);
++ }
++}
+
-+struct au_rwsem {
-+ struct rw_semaphore rwsem;
-+#ifdef CONFIG_AUFS_DEBUG
-+ atomic_t rcnt;
++#if 0 /* reserved */
++static inline void au_pin_set_parent(struct au_pin *pin, struct dentry *parent)
++{
++ if (pin) {
++ dput(pin->parent);
++ pin->parent = dget(parent);
++ }
++}
+#endif
-+};
+
-+#ifdef CONFIG_AUFS_DEBUG
-+#define AuDbgRcntInit(rw) do { \
-+ atomic_set(&(rw)->rcnt, 0); \
-+ smp_mb(); /* atomic set */ \
-+} while (0)
++/* ---------------------------------------------------------------------- */
+
-+#define AuDbgRcntInc(rw) atomic_inc_return(&(rw)->rcnt)
-+#define AuDbgRcntDec(rw) WARN_ON(atomic_dec_return(&(rw)->rcnt) < 0)
-+#else
-+#define AuDbgRcntInit(rw) do {} while (0)
-+#define AuDbgRcntInc(rw) do {} while (0)
-+#define AuDbgRcntDec(rw) do {} while (0)
-+#endif /* CONFIG_AUFS_DEBUG */
++struct au_branch;
++#ifdef CONFIG_AUFS_HNOTIFY
++struct au_hnotify_op {
++ void (*ctl)(struct au_hinode *hinode, int do_set);
++ int (*alloc)(struct au_hinode *hinode);
+
-+#define au_rwsem_destroy(rw) AuDebugOn(rwsem_is_locked(&(rw)->rwsem))
++ /*
++ * if it returns true, the caller should free hinode->hi_notify,
++ * otherwise ->free() frees it.
++ */
++ int (*free)(struct au_hinode *hinode,
++ struct au_hnotify *hn) __must_check;
+
-+static inline void au_rw_init_nolock(struct au_rwsem *rw)
-+{
-+ AuDbgRcntInit(rw);
-+ init_rwsem(&rw->rwsem);
-+}
++ void (*fin)(void);
++ int (*init)(void);
+
-+static inline void au_rw_init_wlock(struct au_rwsem *rw)
-+{
-+ au_rw_init_nolock(rw);
-+ down_write(&rw->rwsem);
-+}
++ int (*reset_br)(unsigned int udba, struct au_branch *br, int perm);
++ void (*fin_br)(struct au_branch *br);
++ int (*init_br)(struct au_branch *br, int perm);
++};
+
-+static inline void au_rw_init_wlock_nested(struct au_rwsem *rw,
-+ unsigned int lsc)
-+{
-+ au_rw_init_nolock(rw);
-+ down_write_nested(&rw->rwsem, lsc);
-+}
++/* hnotify.c */
++int au_hn_alloc(struct au_hinode *hinode, struct inode *inode);
++void au_hn_free(struct au_hinode *hinode);
++void au_hn_ctl(struct au_hinode *hinode, int do_set);
++void au_hn_reset(struct inode *inode, unsigned int flags);
++int au_hnotify(struct inode *h_dir, struct au_hnotify *hnotify, u32 mask,
++ const struct qstr *h_child_qstr, struct inode *h_child_inode);
++int au_hnotify_reset_br(unsigned int udba, struct au_branch *br, int perm);
++int au_hnotify_init_br(struct au_branch *br, int perm);
++void au_hnotify_fin_br(struct au_branch *br);
++int __init au_hnotify_init(void);
++void au_hnotify_fin(void);
++
++/* hfsnotify.c */
++extern const struct au_hnotify_op au_hnotify_op;
+
-+static inline void au_rw_read_lock(struct au_rwsem *rw)
++static inline
++void au_hn_init(struct au_hinode *hinode)
+{
-+ down_read(&rw->rwsem);
-+ AuDbgRcntInc(rw);
++ hinode->hi_notify = NULL;
+}
+
-+static inline void au_rw_read_lock_nested(struct au_rwsem *rw, unsigned int lsc)
++static inline struct au_hnotify *au_hn(struct au_hinode *hinode)
+{
-+ down_read_nested(&rw->rwsem, lsc);
-+ AuDbgRcntInc(rw);
++ return hinode->hi_notify;
+}
+
-+static inline void au_rw_read_unlock(struct au_rwsem *rw)
-+{
-+ AuDbgRcntDec(rw);
-+ up_read(&rw->rwsem);
-+}
++#else
++AuStub(int, au_hn_alloc, return -EOPNOTSUPP,
++ struct au_hinode *hinode __maybe_unused,
++ struct inode *inode __maybe_unused)
++AuStub(struct au_hnotify *, au_hn, return NULL, struct au_hinode *hinode)
++AuStubVoid(au_hn_free, struct au_hinode *hinode __maybe_unused)
++AuStubVoid(au_hn_ctl, struct au_hinode *hinode __maybe_unused,
++ int do_set __maybe_unused)
++AuStubVoid(au_hn_reset, struct inode *inode __maybe_unused,
++ unsigned int flags __maybe_unused)
++AuStubInt0(au_hnotify_reset_br, unsigned int udba __maybe_unused,
++ struct au_branch *br __maybe_unused,
++ int perm __maybe_unused)
++AuStubInt0(au_hnotify_init_br, struct au_branch *br __maybe_unused,
++ int perm __maybe_unused)
++AuStubVoid(au_hnotify_fin_br, struct au_branch *br __maybe_unused)
++AuStubInt0(__init au_hnotify_init, void)
++AuStubVoid(au_hnotify_fin, void)
++AuStubVoid(au_hn_init, struct au_hinode *hinode __maybe_unused)
++#endif /* CONFIG_AUFS_HNOTIFY */
+
-+static inline void au_rw_dgrade_lock(struct au_rwsem *rw)
++static inline void au_hn_suspend(struct au_hinode *hdir)
+{
-+ AuDbgRcntInc(rw);
-+ downgrade_write(&rw->rwsem);
++ au_hn_ctl(hdir, /*do_set*/0);
+}
+
-+static inline void au_rw_write_lock(struct au_rwsem *rw)
++static inline void au_hn_resume(struct au_hinode *hdir)
+{
-+ down_write(&rw->rwsem);
++ au_hn_ctl(hdir, /*do_set*/1);
+}
+
-+static inline void au_rw_write_lock_nested(struct au_rwsem *rw,
-+ unsigned int lsc)
++static inline void au_hn_inode_lock(struct au_hinode *hdir)
+{
-+ down_write_nested(&rw->rwsem, lsc);
++ inode_lock(hdir->hi_inode);
++ au_hn_suspend(hdir);
+}
+
-+static inline void au_rw_write_unlock(struct au_rwsem *rw)
++static inline void au_hn_inode_lock_nested(struct au_hinode *hdir,
++ unsigned int sc __maybe_unused)
+{
-+ up_write(&rw->rwsem);
++ inode_lock_nested(hdir->hi_inode, sc);
++ au_hn_suspend(hdir);
+}
+
-+/* why is not _nested version defined */
-+static inline int au_rw_read_trylock(struct au_rwsem *rw)
++#if 0 /* unused */
++#include "vfsub.h"
++static inline void au_hn_inode_lock_shared_nested(struct au_hinode *hdir,
++ unsigned int sc)
+{
-+ int ret = down_read_trylock(&rw->rwsem);
-+ if (ret)
-+ AuDbgRcntInc(rw);
-+ return ret;
++ inode_lock_shared_nested(hdir->hi_inode, sc);
++ au_hn_suspend(hdir);
+}
++#endif
+
-+static inline int au_rw_write_trylock(struct au_rwsem *rw)
++static inline void au_hn_inode_unlock(struct au_hinode *hdir)
+{
-+ return down_write_trylock(&rw->rwsem);
++ au_hn_resume(hdir);
++ inode_unlock(hdir->hi_inode);
+}
+
-+#undef AuDbgRcntInit
-+#undef AuDbgRcntInc
-+#undef AuDbgRcntDec
-+
-+/* to debug easier, do not make them inlined functions */
-+#define AuRwMustNoWaiters(rw) AuDebugOn(!list_empty(&(rw)->rwsem.wait_list))
-+#define AuRwMustAnyLock(rw) AuDebugOn(down_write_trylock(&(rw)->rwsem))
-+#ifdef CONFIG_AUFS_DEBUG
-+#define AuRwMustReadLock(rw) do { \
-+ AuRwMustAnyLock(rw); \
-+ AuDebugOn(!atomic_read(&(rw)->rcnt)); \
-+} while (0)
-+
-+#define AuRwMustWriteLock(rw) do { \
-+ AuRwMustAnyLock(rw); \
-+ AuDebugOn(atomic_read(&(rw)->rcnt)); \
-+} while (0)
-+#else
-+#define AuRwMustReadLock(rw) AuRwMustAnyLock(rw)
-+#define AuRwMustWriteLock(rw) AuRwMustAnyLock(rw)
-+#endif /* CONFIG_AUFS_DEBUG */
-+
-+#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)); }
-+/* 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), nested);} */
-+
-+#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)
-+
-+/* ---------------------------------------------------------------------- */
-+
-+void *au_kzrealloc(void *p, unsigned int nused, unsigned int new_sz, gfp_t gfp);
-+
-+struct au_nd_store {
-+ unsigned int flags;
-+ struct path path;
-+ struct open_intent intent;
-+};
-+struct au_sbinfo;
-+void au_nd_store(struct au_nd_store *store, struct nameidata *nd,
-+ struct au_sbinfo *sbinfo);
-+void au_nd_revert(struct au_nd_store *store, struct nameidata *nd,
-+ struct au_sbinfo *sbinfo);
-+
-+struct nameidata *au_dup_nd(struct au_sbinfo *sbinfo, struct nameidata *dst,
-+ struct nameidata *src);
-+
-+struct nameidata *au_fake_dm(struct nameidata *fake_nd, struct nameidata *nd,
-+ struct super_block *sb, aufs_bindex_t bindex);
-+void au_fake_dm_release(struct nameidata *fake_nd);
-+struct vfsub_args;
-+int au_h_create(struct inode *h_dir, struct dentry *h_dentry, int mode,
-+ struct vfsub_args *vargs, struct nameidata *nd,
-+ struct vfsmount *nfsmnt);
-+
-+struct au_hinode;
-+int au_copy_file(struct file *dst, struct file *src, loff_t len,
-+ struct au_hinode *hdir, struct super_block *sb,
-+ struct vfsub_args *vargs);
-+
+#endif /* __KERNEL__ */
-+#endif /* __AUFS_MISC_H__ */
-diff -urN linux/fs/aufs/module.c linux-aufs/fs/aufs/module.c
---- linux/fs/aufs/module.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/module.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,269 @@
++#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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,220 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * module global variables and operations
-+ *
-+ * $Id$
++ * ioctl
++ * plink-management and readdir in userspace.
++ * assist the pathconf(3) wrapper library.
++ * move-down
++ * File-based Hierarchical Storage Management.
+ */
+
-+#include <linux/module.h>
++#include <linux/compat.h>
++#include <linux/file.h>
+#include "aufs.h"
+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * aufs caches
-+ */
-+struct kmem_cache *au_cachep[AuCache_Last];
-+static int __init create_cache(void)
-+{
-+ au_cachep[AuCache_DINFO] = AuCache(au_dinfo);
-+ if (au_cachep[AuCache_DINFO])
-+ au_cachep[AuCache_ICNTNR] = AuCache(aufs_icntnr);
-+ if (au_cachep[AuCache_ICNTNR])
-+ au_cachep[AuCache_FINFO] = AuCache(au_finfo);
-+ 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])
-+ return 0;
++static int au_wbr_fd(struct path *path, struct aufs_wbr_fd __user *arg)
++{
++ int err, fd;
++ aufs_bindex_t wbi, bindex, bbot;
++ struct file *h_file;
++ struct super_block *sb;
++ struct dentry *root;
++ struct au_branch *br;
++ struct aufs_wbr_fd wbrfd = {
++ .oflags = au_dir_roflags,
++ .brid = -1
++ };
++ const int valid = O_RDONLY | O_NONBLOCK | O_LARGEFILE | O_DIRECTORY
++ | O_NOATIME | O_CLOEXEC;
+
-+ return -ENOMEM;
-+}
++ AuDebugOn(wbrfd.oflags & ~valid);
+
-+static void destroy_cache(void)
-+{
-+ int i;
-+ for (i = 0; i < AuCache_Last; i++)
-+ if (au_cachep[i]) {
-+ kmem_cache_destroy(au_cachep[i]);
-+ au_cachep[i] = NULL;
++ if (arg) {
++ err = copy_from_user(&wbrfd, arg, sizeof(wbrfd));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ goto out;
+ }
-+}
+
-+/* ---------------------------------------------------------------------- */
++ err = -EINVAL;
++ AuDbg("wbrfd{0%o, %d}\n", wbrfd.oflags, wbrfd.brid);
++ wbrfd.oflags |= au_dir_roflags;
++ AuDbg("0%o\n", wbrfd.oflags);
++ if (unlikely(wbrfd.oflags & ~valid))
++ goto out;
++ }
+
-+char au_esc_chars[0x20 + 3]; /* 0x01-0x20, backslash, del, and NULL */
-+int au_dir_roflags;
++ fd = get_unused_fd_flags(0);
++ err = fd;
++ if (unlikely(fd < 0))
++ goto out;
+
-+/*
-+ * functions for module interface.
-+ */
-+MODULE_LICENSE("GPL");
-+/* MODULE_LICENSE("GPL v2"); */
-+MODULE_AUTHOR("Junjiro Okajima");
-+MODULE_DESCRIPTION(AUFS_NAME " -- Another unionfs");
-+MODULE_VERSION(AUFS_VERSION);
++ h_file = ERR_PTR(-EINVAL);
++ wbi = 0;
++ br = NULL;
++ sb = path->dentry->d_sb;
++ root = sb->s_root;
++ aufs_read_lock(root, AuLock_IR);
++ bbot = au_sbbot(sb);
++ if (wbrfd.brid >= 0) {
++ wbi = au_br_index(sb, wbrfd.brid);
++ if (unlikely(wbi < 0 || wbi > bbot))
++ goto out_unlock;
++ }
++
++ h_file = ERR_PTR(-ENOENT);
++ br = au_sbr(sb, wbi);
++ if (!au_br_writable(br->br_perm)) {
++ if (arg)
++ goto out_unlock;
+
-+/* it should be 'byte', but param_set_byte() prints it by "%c" */
-+short aufs_nwkq = AUFS_NWKQ_DEF;
-+MODULE_PARM_DESC(nwkq, "the number of workqueue thread, " AUFS_WKQ_NAME);
-+module_param_named(nwkq, aufs_nwkq, short, S_IRUGO);
++ bindex = wbi + 1;
++ wbi = -1;
++ for (; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_writable(br->br_perm)) {
++ wbi = bindex;
++ br = au_sbr(sb, wbi);
++ break;
++ }
++ }
++ }
++ AuDbg("wbi %d\n", wbi);
++ if (wbi >= 0)
++ h_file = au_h_open(root, wbi, wbrfd.oflags, NULL,
++ /*force_wr*/0);
+
-+int sysaufs_brs;
-+MODULE_PARM_DESC(brs, "use <sysfs>/fs/aufs/si_*/brN");
-+module_param_named(brs, sysaufs_brs, int, S_IRUGO);
++out_unlock:
++ aufs_read_unlock(root, AuLock_IR);
++ err = PTR_ERR(h_file);
++ if (IS_ERR(h_file))
++ goto out_fd;
++
++ au_lcnt_dec(&br->br_nfiles); /* cf. au_h_open() */
++ fd_install(fd, h_file);
++ err = fd;
++ goto out; /* success */
++
++out_fd:
++ put_unused_fd(fd);
++out:
++ AuTraceErr(err);
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+static int __init aufs_init(void)
++long aufs_ioctl_dir(struct file *file, unsigned int cmd, unsigned long arg)
+{
-+ int err, i;
-+ char *p;
++ long err;
++ struct dentry *dentry;
+
-+ au_debug_init();
-+#ifdef CONFIG_AUFS_INO_T_64
-+ BUILD_BUG_ON(sizeof(ino_t) != sizeof(long long));
-+#else
-+ BUILD_BUG_ON(sizeof(ino_t) != sizeof(int));
-+#endif
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_ioctl(file, cmd, arg);
++ break;
+
-+ p = au_esc_chars;
-+ for (i = 1; i <= ' '; i++)
-+ *p++ = i;
-+ *p++ = '\\';
-+ *p++ = '\x7f';
-+ *p = 0;
++ case AUFS_CTL_WBR_FD:
++ err = au_wbr_fd(&file->f_path, (void __user *)arg);
++ break;
+
-+ au_dir_roflags = au_file_roflags(O_DIRECTORY | O_LARGEFILE);
++ case AUFS_CTL_IBUSY:
++ err = au_ibusy_ioctl(file, arg);
++ break;
+
-+ err = -EINVAL;
-+ if (unlikely(aufs_nwkq <= 0))
-+ goto out;
++ case AUFS_CTL_BRINFO:
++ err = au_brinfo_ioctl(file, arg);
++ break;
+
-+ err = sysaufs_init();
-+ if (unlikely(err))
-+ goto out;
-+ err = au_wkq_init();
-+ if (unlikely(err))
-+ goto out_sysaufs;
-+ err = au_inotify_init();
-+ if (unlikely(err))
-+ goto out_wkq;
-+ err = au_sysrq_init();
-+ if (unlikely(err))
-+ goto out_inotify;
-+
-+ err = create_cache();
-+ if (unlikely(err))
-+ goto out_sysrq;
++ case AUFS_CTL_FHSM_FD:
++ dentry = file->f_path.dentry;
++ if (IS_ROOT(dentry))
++ err = au_fhsm_fd(dentry->d_sb, arg);
++ else
++ err = -ENOTTY;
++ break;
+
-+ err = register_filesystem(&aufs_fs_type);
-+ if (unlikely(err))
-+ goto out_cache;
-+ pr_info(AUFS_NAME " " AUFS_VERSION "\n");
-+ return 0; /* success */
++ default:
++ /* do not call the lower */
++ AuDbg("0x%x\n", cmd);
++ err = -ENOTTY;
++ }
+
-+ out_cache:
-+ destroy_cache();
-+ out_sysrq:
-+ au_sysrq_fin();
-+ out_inotify:
-+ au_inotify_fin();
-+ out_wkq:
-+ au_wkq_fin();
-+ out_sysaufs:
-+ sysaufs_fin();
-+ out:
+ AuTraceErr(err);
+ return err;
+}
+
-+static void __exit aufs_exit(void)
++long aufs_ioctl_nondir(struct file *file, unsigned int cmd, unsigned long arg)
+{
-+ unregister_filesystem(&aufs_fs_type);
-+ destroy_cache();
-+
-+ au_sysrq_fin();
-+ au_inotify_fin();
-+ au_wkq_fin();
-+ sysaufs_fin();
-+}
-+
-+module_init(aufs_init);
-+module_exit(aufs_exit);
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/* fake Kconfig */
-+#if 1
-+
-+#if AUFS_BRANCH_MAX > 511 && PAGE_SIZE > 4096
-+#warning pagesize is larger than 4kb, \
-+ CONFIG_AUFS_BRANCH_MAX_511 or smaller is recommended.
-+#endif
-+
-+#ifdef CONFIG_AUFS_STAT
-+#ifndef CONFIG_SYSFS
-+#error CONFIG_AUFS_STAT requires CONFIG_SYSFS.
-+#endif
-+#endif /* CONFIG_AUFS_STAT */
-+
-+#ifdef CONFIG_AUFS_SYSAUFS
-+#warning CONFIG_AUFS_SYSAUFS is unnecessary for linux-2.6.25 and later.
-+#endif
-+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+#ifndef CONFIG_INOTIFY
-+#error enable CONFIG_INOTIFY to use CONFIG_AUFS_HINOTIFY.
-+#endif
-+#endif /* CONFIG_AUFS_HINOTIFY */
-+
-+#ifdef CONFIG_AUFS_EXPORT
-+#if !defined(CONFIG_EXPORTFS) && !defined(CONFIG_EXPORTFS_MODULE)
-+#error CONFIG_AUFS_EXPORT requires CONFIG_EXPORTFS
-+#endif
-+#if defined(CONFIG_EXPORTFS_MODULE) && defined(CONFIG_AUFS)
-+#error need CONFIG_EXPORTFS = y to link aufs statically with CONFIG_AUFS_EXPORT.
-+#endif
-+#endif /* CONFIG_AUFS_EXPORT */
-+
-+#ifdef CONFIG_AUFS_SEC_PERM_PATCH
-+#ifndef CONFIG_SECURITY
-+#warning AUFS_SEC_PERM_PATCH is unnecessary since SECURITY is disabled.
-+#endif
-+#ifdef CONFIG_AUFS
-+#warning AUFS_SEC_PERM_PATCH is unnecessary since AUFS is not a module.
-+#endif
-+#endif
++ long err;
+
-+#ifdef CONFIG_AUFS_LHASH_PATCH
-+#if !defined(CONFIG_NFS_FS) && !defined(CONFIG_NFS_FS_MODULE)
-+#warning CONFIG_AUFS_LHASH_PATCH is unnecessary since CONFIG_NFS_FS is disabled.
-+#endif
-+#endif
++ switch (cmd) {
++ case AUFS_CTL_MVDOWN:
++ err = au_mvdown(file->f_path.dentry, (void __user *)arg);
++ break;
+
-+#ifdef CONFIG_AUFS_PUT_FILP_PATCH
-+#ifndef CONFIG_NFS_V4
-+#warning AUFS_PUT_FILP_PATCH is unnecessary since NFS_V4 is disabled.
-+#endif
-+#ifdef CONFIG_AUFS
-+#warning AUFS_PUT_FILP_PATCH is unnecessary since AUFS is not a module.
-+#endif
-+#endif /* CONFIG_AUFS_PUT_FILP_PATCH */
++ case AUFS_CTL_WBR_FD:
++ err = au_wbr_fd(&file->f_path, (void __user *)arg);
++ break;
+
-+#ifdef CONFIG_AUFS_FSYNC_SUPER_PATCH
-+#ifdef CONFIG_AUFS
-+#warning AUFS_FSYNC_SUPER_PATCH is unnecessary since AUFS is not a module.
-+#endif
-+#endif
++ default:
++ /* do not call the lower */
++ AuDbg("0x%x\n", cmd);
++ err = -ENOTTY;
++ }
+
-+#ifdef CONFIG_AUFS_DENY_WRITE_ACCESS_PATCH
-+#ifdef CONFIG_AUFS
-+#warning AUFS_DENY_WRITE_ACCESS_PATCH is unnecessary since AUFS is not a module.
-+#endif
-+#endif
++ AuTraceErr(err);
++ return err;
++}
+
-+#ifdef CONFIG_AUFS_KSIZE_PATCH
-+#warning CONFIG_AUFS_KSIZE_PATCH is unnecessary for linux-2.6.22 and later.
-+#endif
++#ifdef CONFIG_COMPAT
++long aufs_compat_ioctl_dir(struct file *file, unsigned int cmd,
++ unsigned long arg)
++{
++ long err;
+
-+#ifdef CONFIG_AUFS_WORKAROUND_FUSE
-+#if !defined(CONFIG_FUSE_FS) && !defined(CONFIG_FUSE_FS_MODULE)
-+#warning CONFIG_AUFS_WORKAROUND_FUSE is enabled while FUSE is disabled.
-+#endif
-+#endif
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_compat_ioctl(file, cmd, arg);
++ break;
+
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+#ifndef CONFIG_AUFS_MAGIC_SYSRQ
-+#warning CONFIG_AUFS_DEBUG_LOCK is enabled but CONFIG_AUFS_MAGIC_SYSRQ.
-+#endif
-+#endif
++ case AUFS_CTL_IBUSY:
++ err = au_ibusy_compat_ioctl(file, arg);
++ break;
+
-+#ifdef CONFIG_AUFS_COMPAT
-+#warning CONFIG_AUFS_COMPAT will be removed in the near future.
-+#endif
++ case AUFS_CTL_BRINFO:
++ err = au_brinfo_compat_ioctl(file, arg);
++ break;
+
-+#ifdef CONFIG_AUFS_UNIONFS23_PATCH
-+#ifndef CONFIG_AUFS_UNIONFS22_PATCH
-+#error mis-config. AUFS_UNIONFS23_PATCH is enabled but AUFS_UNIONFS22_PATCH.
-+#endif
-+#ifndef CONFIG_AUFS_SPLICE_PATCH
-+#error mis-config. AUFS_UNIONFS23_PATCH is enabled but AUFS_SPLICE_PATCH.
-+#endif
-+#endif
++ default:
++ err = aufs_ioctl_dir(file, cmd, arg);
++ }
+
-+#ifdef CONFIG_DEBUG_PROVE_LOCKING
-+#if MAX_LOCKDEP_SUBCLASSES < AuLsc_I_End
-+#warning lockdep will not work since aufs uses deeper locks.
-+#endif
-+#endif
++ AuTraceErr(err);
++ return err;
++}
+
++long aufs_compat_ioctl_nondir(struct file *file, unsigned int cmd,
++ unsigned long arg)
++{
++ return aufs_ioctl_nondir(file, cmd, (unsigned long)compat_ptr(arg));
++}
+#endif
-diff -urN linux/fs/aufs/module.h linux-aufs/fs/aufs/module.h
---- linux/fs/aufs/module.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/module.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,73 @@
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,972 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * module initialization and module-global
-+ *
-+ * $Id$
++ * inode operations (add entry)
+ */
+
-+#ifndef __AUFS_MODULE_H__
-+#define __AUFS_MODULE_H__
-+
-+#ifdef __KERNEL__
-+
-+/* module parameters */
-+extern short aufs_nwkq;
-+extern int sysaufs_brs;
-+
-+/* ---------------------------------------------------------------------- */
++#include <linux/iversion.h>
++#include "aufs.h"
+
-+extern char au_esc_chars[];
-+extern int au_dir_roflags;
++/*
++ * final procedure of adding a new entry, except link(2).
++ * remove whiteout, instantiate, copyup the parent dir's times and size
++ * and update version.
++ * if it failed, re-create the removed whiteout.
++ */
++static int epilog(struct inode *dir, aufs_bindex_t bindex,
++ struct dentry *wh_dentry, struct dentry *dentry)
++{
++ int err, rerr;
++ aufs_bindex_t bwh;
++ struct path h_path;
++ struct super_block *sb;
++ struct inode *inode, *h_dir;
++ struct dentry *wh;
+
-+/* kmem cache */
-+enum {
-+ AuCache_DINFO,
-+ AuCache_ICNTNR,
-+ AuCache_FINFO,
-+ AuCache_VDIR,
-+ AuCache_DEHSTR,
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ AuCache_HINOTIFY,
-+#endif
-+ AuCache_Last
-+};
++ bwh = -1;
++ sb = dir->i_sb;
++ if (wh_dentry) {
++ h_dir = d_inode(wh_dentry->d_parent); /* dir inode is locked */
++ IMustLock(h_dir);
++ AuDebugOn(au_h_iptr(dir, bindex) != h_dir);
++ bwh = au_dbwh(dentry);
++ h_path.dentry = wh_dentry;
++ h_path.mnt = au_sbr_mnt(sb, bindex);
++ err = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path,
++ dentry);
++ if (unlikely(err))
++ goto out;
++ }
+
-+extern struct kmem_cache *au_cachep[];
++ inode = au_new_inode(dentry, /*must_new*/1);
++ if (!IS_ERR(inode)) {
++ d_instantiate(dentry, inode);
++ dir = d_inode(dentry->d_parent); /* dir inode is locked */
++ IMustLock(dir);
++ au_dir_ts(dir, bindex);
++ inode_inc_iversion(dir);
++ au_fhsm_wrote(sb, bindex, /*force*/0);
++ return 0; /* success */
++ }
+
-+#define AuCacheArgs(type, sz) (type), (sz), 0, SLAB_RECLAIM_ACCOUNT, NULL
-+#define AuCache(type) \
-+ kmem_cache_create(AuCacheArgs(#type, sizeof(struct type)))
++ err = PTR_ERR(inode);
++ if (!wh_dentry)
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ /* revert */
++ /* dir inode is locked */
++ wh = au_wh_create(dentry, bwh, wh_dentry->d_parent);
++ rerr = PTR_ERR(wh);
++ if (IS_ERR(wh)) {
++ AuIOErr("%pd reverting whiteout failed(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ } else
++ dput(wh);
+
-+#define AuCacheFuncs(name, index) \
-+static inline void *au_cache_alloc_##name(void) \
-+{ return kmem_cache_alloc(au_cachep[index], GFP_NOFS); } \
-+static inline void au_cache_free_##name(void *p) \
-+{ kmem_cache_free(au_cachep[index], p); }
++out:
++ return err;
++}
+
-+AuCacheFuncs(dinfo, AuCache_DINFO);
-+AuCacheFuncs(icntnr, AuCache_ICNTNR);
-+AuCacheFuncs(finfo, AuCache_FINFO);
-+AuCacheFuncs(vdir, AuCache_VDIR);
-+AuCacheFuncs(dehstr, AuCache_DEHSTR);
++static int au_d_may_add(struct dentry *dentry)
++{
++ int err;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_MODULE_H__ */
-diff -urN linux/fs/aufs/opts.c linux-aufs/fs/aufs/opts.c
---- linux/fs/aufs/opts.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/opts.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1613 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ err = 0;
++ if (unlikely(d_unhashed(dentry)))
++ err = -ENOENT;
++ if (unlikely(d_really_is_positive(dentry)))
++ err = -EEXIST;
++ return err;
++}
+
+/*
-+ * mount options/flags
-+ *
-+ * $Id$
++ * simple tests for the adding inode operations.
++ * following the checks in vfs, plus the parent-child relationship.
+ */
++int au_may_add(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir)
++{
++ int err;
++ umode_t h_mode;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
+
-+#include <linux/types.h> /* a distribution requires */
-+#include <linux/parser.h>
-+#include "aufs.h"
++ err = -ENAMETOOLONG;
++ if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN))
++ goto out;
+
-+/* ---------------------------------------------------------------------- */
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (d_really_is_negative(dentry)) {
++ err = -EEXIST;
++ if (unlikely(d_is_positive(h_dentry)))
++ goto out;
++ } else {
++ /* rename(2) case */
++ err = -EIO;
++ if (unlikely(d_is_negative(h_dentry)))
++ goto out;
++ h_inode = d_inode(h_dentry);
++ if (unlikely(!h_inode->i_nlink))
++ goto out;
+
-+enum {
-+ Opt_br,
-+ Opt_add, Opt_del, Opt_mod, Opt_reorder, Opt_append, Opt_prepend,
-+ Opt_idel, Opt_imod, Opt_ireorder,
-+ Opt_dirwh, Opt_rdcache, Opt_deblk, Opt_nhash, Opt_rendir,
-+ Opt_xino, Opt_zxino, Opt_noxino,
-+ Opt_trunc_xino, Opt_trunc_xino_v, Opt_notrunc_xino,
-+ Opt_trunc_xino_path, Opt_itrunc_xino,
-+ Opt_xinodir,
-+ Opt_trunc_xib, Opt_notrunc_xib,
-+ Opt_dirperm1, Opt_nodirperm1,
-+ Opt_shwh, Opt_noshwh,
-+ Opt_plink, Opt_noplink, Opt_list_plink, Opt_clean_plink,
-+ Opt_udba,
-+ /* Opt_lock, Opt_unlock, */
-+ Opt_cmd, Opt_cmd_args,
-+ Opt_diropq_a, Opt_diropq_w,
-+ Opt_warn_perm, Opt_nowarn_perm,
-+ Opt_wbr_copyup, Opt_wbr_create,
-+ Opt_coo,
-+ Opt_dlgt, Opt_nodlgt,
-+ Opt_refrof, Opt_norefrof,
-+ Opt_verbose, Opt_noverbose,
-+ Opt_sum, Opt_nosum, Opt_wsum,
-+ Opt_tail, Opt_ignore, Opt_ignore_silent, Opt_err
-+};
++ h_mode = h_inode->i_mode;
++ if (!isdir) {
++ err = -EISDIR;
++ if (unlikely(S_ISDIR(h_mode)))
++ goto out;
++ } else if (unlikely(!S_ISDIR(h_mode))) {
++ err = -ENOTDIR;
++ goto out;
++ }
++ }
+
-+static match_table_t options = {
-+ {Opt_br, "br=%s"},
-+ {Opt_br, "br:%s"},
-+
-+ {Opt_add, "add=%d:%s"},
-+ {Opt_add, "add:%d:%s"},
-+ {Opt_add, "ins=%d:%s"},
-+ {Opt_add, "ins:%d:%s"},
-+ {Opt_append, "append=%s"},
-+ {Opt_append, "append:%s"},
-+ {Opt_prepend, "prepend=%s"},
-+ {Opt_prepend, "prepend:%s"},
-+
-+ {Opt_del, "del=%s"},
-+ {Opt_del, "del:%s"},
-+ /* {Opt_idel, "idel:%d"}, */
-+ {Opt_mod, "mod=%s"},
-+ {Opt_mod, "mod:%s"},
-+ {Opt_imod, "imod:%d:%s"},
-+
-+ {Opt_dirwh, "dirwh=%d"},
-+ {Opt_dirwh, "dirwh:%d"},
-+
-+ {Opt_xino, "xino=%s"},
-+ {Opt_xino, "xino:%s"},
-+#if 0 /* def CONFIG_AUFS_EXPORT */ /* reserved for futur use */
-+ {Opt_xinodir, "xinodir=%s"},
-+ {Opt_xinodir, "xinodir:%s"},
-+#endif
-+ {Opt_noxino, "noxino"},
-+ {Opt_trunc_xino, "trunc_xino"},
-+ {Opt_trunc_xino_v, "trunc_xino_v=%d:%d"},
-+ {Opt_notrunc_xino, "notrunc_xino"},
-+ {Opt_trunc_xino_path, "trunc_xino=%s"},
-+ {Opt_trunc_xino_path, "trunc_xino:%s"},
-+ {Opt_itrunc_xino, "itrunc_xino=%d"},
-+ {Opt_itrunc_xino, "itrunc_xino:%d"},
-+ /* {Opt_zxino, "zxino=%s"}, */
-+ {Opt_trunc_xib, "trunc_xib"},
-+ {Opt_notrunc_xib, "notrunc_xib"},
-+
-+ {Opt_plink, "plink"},
-+ {Opt_noplink, "noplink"},
-+#ifdef CONFIG_AUFS_DEBUG
-+ {Opt_list_plink, "list_plink"},
-+#endif
-+ {Opt_clean_plink, "clean_plink"},
++ err = 0;
++ /* expected parent dir is locked */
++ if (unlikely(h_parent != h_dentry->d_parent))
++ err = -EIO;
+
-+ {Opt_udba, "udba=%s"},
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ {Opt_diropq_a, "diropq=always"},
-+ {Opt_diropq_a, "diropq=a"},
-+ {Opt_diropq_w, "diropq=whiteouted"},
-+ {Opt_diropq_w, "diropq=w"},
++/*
++ * initial procedure of adding a new entry.
++ * prepare writable branch and the parent dir, lock it,
++ * and lookup whiteout for the new entry.
++ */
++static struct dentry*
++lock_hdir_lkup_wh(struct dentry *dentry, struct au_dtime *dt,
++ struct dentry *src_dentry, struct au_pin *pin,
++ struct au_wr_dir_args *wr_dir_args)
++{
++ struct dentry *wh_dentry, *h_parent;
++ struct super_block *sb;
++ struct au_branch *br;
++ int err;
++ unsigned int udba;
++ aufs_bindex_t bcpup;
+
-+ {Opt_warn_perm, "warn_perm"},
-+ {Opt_nowarn_perm, "nowarn_perm"},
++ AuDbg("%pd\n", dentry);
+
-+#ifdef CONFIG_AUFS_DLGT
-+ {Opt_dlgt, "dlgt"},
-+ {Opt_dirperm1, "dirperm1"},
-+#endif
-+ {Opt_nodlgt, "nodlgt"},
-+ {Opt_nodirperm1, "nodirperm1"},
++ err = au_wr_dir(dentry, src_dentry, wr_dir_args);
++ bcpup = err;
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err < 0))
++ goto out;
+
-+#ifdef CONFIG_AUFS_SHWH
-+ {Opt_shwh, "shwh"},
-+#endif
-+ {Opt_noshwh, "noshwh"},
-+
-+ {Opt_rendir, "rendir=%d"},
-+ {Opt_rendir, "rendir:%d"},
-+
-+ {Opt_refrof, "refrof"},
-+ {Opt_norefrof, "norefrof"},
-+
-+ {Opt_verbose, "verbose"},
-+ {Opt_verbose, "v"},
-+ {Opt_noverbose, "noverbose"},
-+ {Opt_noverbose, "quiet"},
-+ {Opt_noverbose, "q"},
-+ {Opt_noverbose, "silent"},
-+
-+ {Opt_sum, "sum"},
-+ {Opt_nosum, "nosum"},
-+ {Opt_wsum, "wsum"},
++ sb = dentry->d_sb;
++ udba = au_opt_udba(sb);
++ err = au_pin(pin, dentry, bcpup, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
+
-+ {Opt_rdcache, "rdcache=%d"},
-+ {Opt_rdcache, "rdcache:%d"},
++ h_parent = au_pinned_h_parent(pin);
++ if (udba != AuOpt_UDBA_NONE
++ && au_dbtop(dentry) == bcpup)
++ err = au_may_add(dentry, bcpup, h_parent,
++ au_ftest_wrdir(wr_dir_args->flags, ISDIR));
++ else if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN))
++ err = -ENAMETOOLONG;
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_unpin;
+
-+ {Opt_coo, "coo=%s"},
++ br = au_sbr(sb, bcpup);
++ if (dt) {
++ struct path tmp = {
++ .dentry = h_parent,
++ .mnt = au_br_mnt(br)
++ };
++ au_dtime_store(dt, au_pinned_parent(pin), &tmp);
++ }
+
-+ {Opt_wbr_create, "create=%s"},
-+ {Opt_wbr_create, "create:%s"},
-+ {Opt_wbr_create, "create_policy=%s"},
-+ {Opt_wbr_create, "create_policy:%s"},
-+ {Opt_wbr_copyup, "cpup=%s"},
-+ {Opt_wbr_copyup, "cpup:%s"},
-+ {Opt_wbr_copyup, "copyup=%s"},
-+ {Opt_wbr_copyup, "copyup:%s"},
-+ {Opt_wbr_copyup, "copyup_policy=%s"},
-+ {Opt_wbr_copyup, "copyup_policy:%s"},
++ wh_dentry = NULL;
++ if (bcpup != au_dbwh(dentry))
++ goto out; /* success */
+
-+ /* internal use for the scripts */
-+ {Opt_ignore_silent, "si=%s"},
++ /*
++ * ENAMETOOLONG here means that if we allowed create such name, then it
++ * would not be able to removed in the future. So we don't allow such
++ * name here and we don't handle ENAMETOOLONG differently here.
++ */
++ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, br);
+
-+#if 0 /* reserved for future use */
-+ {Opt_deblk, "deblk=%d"},
-+ {Opt_deblk, "deblk:%d"},
-+ {Opt_nhash, "nhash=%d"},
-+ {Opt_nhash, "nhash:%d"},
-+#endif
++out_unpin:
++ if (IS_ERR(wh_dentry))
++ au_unpin(pin);
++out:
++ return wh_dentry;
++}
+
-+ {Opt_br, "dirs=%s"},
-+ {Opt_ignore, "debug=%d"},
-+ {Opt_ignore, "delete=whiteout"},
-+ {Opt_ignore, "delete=all"},
-+ {Opt_ignore, "imap=%s"},
++/* ---------------------------------------------------------------------- */
+
-+ {Opt_err, NULL}
++enum { Mknod, Symlink, Creat };
++struct simple_arg {
++ int type;
++ union {
++ struct {
++ umode_t mode;
++ bool want_excl;
++ bool try_aopen;
++ struct vfsub_aopen_args *aopen;
++ } c;
++ struct {
++ const char *symname;
++ } s;
++ struct {
++ umode_t mode;
++ dev_t dev;
++ } m;
++ } u;
+};
+
-+/* ---------------------------------------------------------------------- */
-+
-+static const char *au_parser_pattern(int val, struct match_token *token)
++static int add_simple(struct inode *dir, struct dentry *dentry,
++ struct simple_arg *arg)
+{
-+ while (token->pattern) {
-+ if (token->token == val)
-+ return token->pattern;
-+ token++;
-+ }
-+ BUG();
-+ return "??";
-+}
++ int err, rerr;
++ aufs_bindex_t btop;
++ unsigned char created;
++ const unsigned char try_aopen
++ = (arg->type == Creat && arg->u.c.try_aopen);
++ struct vfsub_aopen_args *aopen = arg->u.c.aopen;
++ struct dentry *wh_dentry, *parent;
++ struct inode *h_dir;
++ struct super_block *sb;
++ struct au_branch *br;
++ /* to reduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct au_pin pin;
++ struct path h_path;
++ struct au_wr_dir_args wr_dir_args;
++ } *a;
+
-+/* ---------------------------------------------------------------------- */
++ AuDbg("%pd\n", dentry);
++ IMustLock(dir);
+
-+#define RW "rw"
-+#define RO "ro"
-+#define WH "wh"
-+#define RR "rr"
-+#define NoLinkWH "nolwh"
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++ a->wr_dir_args.force_btgt = -1;
++ a->wr_dir_args.flags = AuWrDir_ADD_ENTRY;
+
-+static match_table_t brperms = {
-+ {AuBrPerm_RR, RR},
-+ {AuBrPerm_RO, RO},
-+ {AuBrPerm_RW, RW},
++ parent = dentry->d_parent; /* dir inode is locked */
++ if (!try_aopen) {
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
++ if (unlikely(err))
++ goto out_free;
++ }
++ err = au_d_may_add(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ if (!try_aopen)
++ di_write_lock_parent(parent);
++ wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL,
++ &a->pin, &a->wr_dir_args);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
+
-+ {AuBrPerm_RRWH, RR "+" WH},
-+ {AuBrPerm_ROWH, RO "+" WH},
-+ {AuBrPerm_RWNoLinkWH, RW "+" NoLinkWH},
++ btop = au_dbtop(dentry);
++ sb = dentry->d_sb;
++ br = au_sbr(sb, btop);
++ a->h_path.dentry = au_h_dptr(dentry, btop);
++ a->h_path.mnt = au_br_mnt(br);
++ h_dir = au_pinned_h_dir(&a->pin);
++ switch (arg->type) {
++ case Creat:
++ if (!try_aopen || !h_dir->i_op->atomic_open) {
++ err = vfsub_create(h_dir, &a->h_path, arg->u.c.mode,
++ arg->u.c.want_excl);
++ created = !err;
++ if (!err && try_aopen)
++ aopen->file->f_mode |= FMODE_CREATED;
++ } else {
++ aopen->br = br;
++ err = vfsub_atomic_open(h_dir, a->h_path.dentry, aopen);
++ AuDbg("err %d\n", err);
++ AuDbgFile(aopen->file);
++ created = err >= 0
++ && !!(aopen->file->f_mode & FMODE_CREATED);
++ }
++ break;
++ case Symlink:
++ err = vfsub_symlink(h_dir, &a->h_path, arg->u.s.symname);
++ created = !err;
++ break;
++ case Mknod:
++ err = vfsub_mknod(h_dir, &a->h_path, arg->u.m.mode,
++ arg->u.m.dev);
++ created = !err;
++ break;
++ default:
++ BUG();
++ }
++ if (unlikely(err < 0))
++ goto out_unpin;
+
-+ {AuBrPerm_ROWH, "nfsro"},
-+ {AuBrPerm_RO, NULL}
-+};
++ err = epilog(dir, btop, wh_dentry, dentry);
++ if (!err)
++ goto out_unpin; /* success */
+
-+static noinline_for_stack int br_perm_val(char *perm)
-+{
-+ int val;
-+ substring_t args[MAX_OPT_ARGS];
++ /* revert */
++ if (created /* && d_is_positive(a->h_path.dentry) */) {
++ /* no delegation since it is just created */
++ rerr = vfsub_unlink(h_dir, &a->h_path, /*delegated*/NULL,
++ /*force*/0);
++ if (rerr) {
++ AuIOErr("%pd revert failure(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ }
++ au_dtime_revert(&a->dt);
++ }
++ if (try_aopen && h_dir->i_op->atomic_open
++ && (aopen->file->f_mode & FMODE_OPENED))
++ /* aopen->file is still opened */
++ au_lcnt_dec(&aopen->br->br_nfiles);
+
-+ AuDebugOn(!perm || !*perm);
-+ LKTRTrace("perm %s\n", perm);
-+ val = match_token(perm, brperms, args);
-+ AuTraceErr(val);
-+ return val;
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++out_parent:
++ if (!try_aopen)
++ di_write_unlock(parent);
++out_unlock:
++ if (unlikely(err)) {
++ au_update_dbtop(dentry);
++ d_drop(dentry);
++ }
++ if (!try_aopen)
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ au_kfree_rcu(a);
++out:
++ return err;
+}
+
-+const char *au_optstr_br_perm(int brperm)
++int aufs_mknod(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode, dev_t dev)
+{
-+ return au_parser_pattern(brperm, (void *)brperms);
++ struct simple_arg arg = {
++ .type = Mknod,
++ .u.m = {
++ .mode = mode,
++ .dev = dev
++ }
++ };
++ return add_simple(dir, dentry, &arg);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static match_table_t udbalevel = {
-+ {AuOpt_UDBA_REVAL, "reval"},
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ {AuOpt_UDBA_INOTIFY, "inotify"},
-+#endif
-+ {AuOpt_UDBA_NONE, "none"},
-+ {-1, NULL}
-+};
-+
-+static noinline_for_stack int udba_val(char *str)
++int aufs_symlink(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, const char *symname)
+{
-+ substring_t args[MAX_OPT_ARGS];
-+ return match_token(str, udbalevel, args);
++ struct simple_arg arg = {
++ .type = Symlink,
++ .u.s.symname = symname
++ };
++ return add_simple(dir, dentry, &arg);
+}
+
-+const char *au_optstr_udba(int udba)
++int aufs_create(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode, bool want_excl)
+{
-+ return au_parser_pattern(udba, (void *)udbalevel);
++ struct simple_arg arg = {
++ .type = Creat,
++ .u.c = {
++ .mode = mode,
++ .want_excl = want_excl
++ }
++ };
++ return add_simple(dir, dentry, &arg);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static match_table_t coolevel = {
-+ {AuOpt_COO_NONE, "none"},
-+ {AuOpt_COO_LEAF, "leaf"},
-+ {AuOpt_COO_ALL, "all"},
-+ {-1, NULL}
-+};
-+
-+static noinline_for_stack int coo_val(char *str)
++int au_aopen_or_create(struct inode *dir, struct dentry *dentry,
++ struct vfsub_aopen_args *aopen_args)
+{
-+ substring_t args[MAX_OPT_ARGS];
-+ return match_token(str, coolevel, args);
++ struct simple_arg arg = {
++ .type = Creat,
++ .u.c = {
++ .mode = aopen_args->create_mode,
++ .want_excl = aopen_args->open_flag & O_EXCL,
++ .try_aopen = true,
++ .aopen = aopen_args
++ }
++ };
++ return add_simple(dir, dentry, &arg);
+}
+
-+const char *au_optstr_coo(int coo)
++int aufs_tmpfile(struct mnt_idmap *idmap, struct inode *dir,
++ struct file *file, umode_t mode)
+{
-+ return au_parser_pattern(coo, (void *)coolevel);
-+}
++ int err;
++ aufs_bindex_t bindex;
++ struct path h_ppath;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct dentry *dentry, *parent, *h_parent, *h_dentry;
++ struct inode *h_dir, *inode;
++ struct vfsmount *h_mnt;
++ struct mnt_idmap *h_idmap;
++ struct file *h_file;
++ struct au_wr_dir_args wr_dir_args = {
++ .force_btgt = -1,
++ .flags = AuWrDir_TMPFILE
++ };
+
-+/* ---------------------------------------------------------------------- */
++ /* copy-up may happen */
++ inode_lock(dir);
+
-+static match_table_t au_wbr_create_policy = {
-+ {AuWbrCreate_TDP, "tdp"},
-+ {AuWbrCreate_TDP, "top-down-parent"},
-+ {AuWbrCreate_RR, "rr"},
-+ {AuWbrCreate_RR, "round-robin"},
-+ {AuWbrCreate_MFS, "mfs"},
-+ {AuWbrCreate_MFS, "most-free-space"},
-+ {AuWbrCreate_MFSV, "mfs:%d"},
-+ {AuWbrCreate_MFSV, "most-free-space:%d"},
++ h_file = NULL;
++ sb = dir->i_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out;
+
-+ {AuWbrCreate_MFSRR, "mfsrr:%d"},
-+ {AuWbrCreate_MFSRRV, "mfsrr:%d:%d"},
-+ {AuWbrCreate_PMFS, "pmfs"},
-+ {AuWbrCreate_PMFSV, "pmfs:%d"},
++ dentry = file->f_path.dentry;
++ err = au_di_init(dentry);
++ if (unlikely(err))
++ goto out_si;
+
-+ {-1, NULL}
-+};
++ err = -EBUSY;
++ parent = d_find_any_alias(dir);
++ AuDebugOn(!parent);
++ di_write_lock_parent(parent);
++ if (unlikely(d_inode(parent) != dir))
++ goto out_parent;
+
-+/* cf. linux/lib/parser.c */
-+static int au_match_ull(substring_t *s, unsigned long long *result)
-+{
-+ int err;
-+ unsigned int len;
-+ char a[32];
++ err = au_digen_test(parent, au_sigen(sb));
++ if (unlikely(err))
++ goto out_parent;
+
-+ err = -ERANGE;
-+ len = s->to - s->from;
-+ if (len + 1 <= sizeof(a)) {
-+ memcpy(a, s->from, len);
-+ a[len] = '\0';
-+#if 1
-+ err = strict_strtoull(a, 0, result);
-+#else
-+ char *next;
-+ *result = memparse(a, &next);
-+ err = *result;
-+ if (!IS_ERR((void *)err))
-+ err = 0;
-+#endif
-+ }
-+ return err;
-+}
++ bindex = au_dbtop(parent);
++ au_set_dbtop(dentry, bindex);
++ au_set_dbbot(dentry, bindex);
++ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args);
++ bindex = err;
++ if (unlikely(err < 0))
++ goto out_parent;
+
-+static int au_wbr_mfs_wmark(substring_t *arg, char *str,
-+ struct au_opt_wbr_create *create)
-+{
-+ int err;
-+ unsigned long long ull;
++ err = -EOPNOTSUPP;
++ h_dir = au_h_iptr(dir, bindex);
++ if (unlikely(!h_dir->i_op->tmpfile))
++ goto out_parent;
+
-+ err = 0;
-+ if (!au_match_ull(arg, &ull))
-+ create->mfsrr_watermark = ull;
-+ else {
-+ AuErr("bad integer in %s\n", str);
-+ err = -EINVAL;
++ br = au_sbr(sb, bindex);
++ h_mnt = au_br_mnt(br);
++ err = vfsub_mnt_want_write(h_mnt);
++ if (unlikely(err))
++ goto out_parent;
++
++ h_idmap = mnt_idmap(h_mnt);
++ h_parent = au_h_dptr(parent, bindex);
++ h_ppath.mnt = h_mnt;
++ h_ppath.dentry = h_parent;
++ h_file = vfs_tmpfile_open(h_idmap, &h_ppath, mode, /*open_flag*/0,
++ current_cred());
++ if (IS_ERR(h_file)) {
++ err = PTR_ERR(h_file);
++ h_file = NULL;
++ goto out_mnt;
+ }
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ h_dentry = h_file->f_path.dentry;
++ au_set_dbtop(dentry, bindex);
++ au_set_dbbot(dentry, bindex);
++ au_set_h_dptr(dentry, bindex, dget(h_dentry));
++ inode = au_new_inode(dentry, /*must_new*/1);
++ if (IS_ERR(inode)) {
++ err = PTR_ERR(inode);
++ au_set_h_dptr(dentry, bindex, NULL);
++ au_set_dbtop(dentry, -1);
++ au_set_dbbot(dentry, -1);
++ goto out_h_file;
++ }
+
-+static int au_wbr_mfs_sec(substring_t *arg, char *str,
-+ struct au_opt_wbr_create *create)
-+{
-+ int n, err;
++ if (!inode->i_nlink)
++ set_nlink(inode, 1);
++ d_tmpfile(file, inode);
++ au_di(dentry)->di_tmpfile = 1;
++ get_file(h_file);
++ au_di(dentry)->di_htmpfile = h_file;
+
-+ err = 0;
-+ if (!match_int(arg, &n) && 0 <= n)
-+ create->mfs_second = n;
-+ else {
-+ AuErr("bad integer in %s\n", str);
-+ err = -EINVAL;
-+ }
++ /* update without i_mutex */
++ if (au_ibtop(dir) == au_dbtop(dentry))
++ au_cpup_attr_timesizes(dir);
+
++out_h_file:
++ fput(h_file);
++out_mnt:
++ vfsub_mnt_drop_write(h_mnt);
++out_parent:
++ di_write_unlock(parent);
++ dput(parent);
++ di_write_unlock(dentry);
++ if (!err)
++ goto out_si;
++ if (h_file)
++ fput(h_file);
++ au_di(dentry)->di_htmpfile = NULL;
++ au_di_fin(dentry);
++ dentry->d_fsdata = NULL;
++out_si:
++ si_read_unlock(sb);
++ if (!err && h_file) {
++ /* finally... */
++ err = finish_open_simple(file, err);
++ if (!err)
++ au_lcnt_inc(&br->br_nfiles);
++ else {
++ fput(h_file);
++ au_di(dentry)->di_htmpfile = NULL;
++ au_di_fin(dentry);
++ dentry->d_fsdata = NULL;
++ }
++ }
++out:
++ inode_unlock(dir);
+ AuTraceErr(err);
+ return err;
+}
+
-+static noinline_for_stack
-+int au_wbr_create_val(char *str, struct au_opt_wbr_create *create)
++/* ---------------------------------------------------------------------- */
++
++struct au_link_args {
++ aufs_bindex_t bdst, bsrc;
++ struct au_pin pin;
++ struct path h_path;
++ struct dentry *src_parent, *parent;
++};
++
++static int au_cpup_before_link(struct dentry *src_dentry,
++ struct au_link_args *a)
+{
-+ int err, e;
-+ substring_t args[MAX_OPT_ARGS];
++ int err;
++ struct dentry *h_src_dentry;
++ struct au_cp_generic cpg = {
++ .dentry = src_dentry,
++ .bdst = a->bdst,
++ .bsrc = a->bsrc,
++ .len = -1,
++ .pin = &a->pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN /* | AuCpup_KEEPLINO */
++ };
+
-+ err = match_token(str, au_wbr_create_policy, args);
-+ create->wbr_create = err;
++ di_read_lock_parent(a->src_parent, AuLock_IR);
++ err = au_test_and_cpup_dirs(src_dentry, a->bdst);
++ if (unlikely(err))
++ goto out;
++
++ h_src_dentry = au_h_dptr(src_dentry, a->bsrc);
++ err = au_pin(&a->pin, src_dentry, a->bdst,
++ au_opt_udba(src_dentry->d_sb),
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out;
++
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&a->pin);
++
++out:
++ di_read_unlock(a->src_parent, AuLock_IR);
++ return err;
++}
++
++static int au_cpup_or_link(struct dentry *src_dentry, struct dentry *dentry,
++ struct au_link_args *a)
++{
++ int err;
++ unsigned char plink;
++ aufs_bindex_t bbot;
++ struct dentry *h_src_dentry;
++ struct inode *h_inode, *inode, *delegated;
++ struct super_block *sb;
++ struct file *h_file;
++
++ plink = 0;
++ h_inode = NULL;
++ sb = src_dentry->d_sb;
++ inode = d_inode(src_dentry);
++ if (au_ibtop(inode) <= a->bdst)
++ h_inode = au_h_iptr(inode, a->bdst);
++ if (!h_inode || !h_inode->i_nlink) {
++ /* copyup src_dentry as the name of dentry. */
++ bbot = au_dbbot(dentry);
++ if (bbot < a->bsrc)
++ au_set_dbbot(dentry, a->bsrc);
++ au_set_h_dptr(dentry, a->bsrc,
++ dget(au_h_dptr(src_dentry, a->bsrc)));
++ dget(a->h_path.dentry);
++ au_set_h_dptr(dentry, a->bdst, NULL);
++ AuDbg("temporary d_inode...\n");
++ spin_lock(&dentry->d_lock);
++ dentry->d_inode = d_inode(src_dentry); /* tmp */
++ spin_unlock(&dentry->d_lock);
++ h_file = au_h_open_pre(dentry, a->bsrc, /*force_wr*/0);
++ if (IS_ERR(h_file))
++ err = PTR_ERR(h_file);
++ else {
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = a->bdst,
++ .bsrc = -1,
++ .len = -1,
++ .pin = &a->pin,
++ .flags = AuCpup_KEEPLINO
++ };
++ err = au_sio_cpup_simple(&cpg);
++ au_h_open_post(dentry, a->bsrc, h_file);
++ if (!err) {
++ dput(a->h_path.dentry);
++ a->h_path.dentry = au_h_dptr(dentry, a->bdst);
++ } else
++ au_set_h_dptr(dentry, a->bdst,
++ a->h_path.dentry);
++ }
++ spin_lock(&dentry->d_lock);
++ dentry->d_inode = NULL; /* restore */
++ spin_unlock(&dentry->d_lock);
++ AuDbg("temporary d_inode...done\n");
++ au_set_h_dptr(dentry, a->bsrc, NULL);
++ au_set_dbbot(dentry, bbot);
++ } else {
++ /* the inode of src_dentry already exists on a.bdst branch */
++ h_src_dentry = d_find_alias(h_inode);
++ if (!h_src_dentry && au_plink_test(inode)) {
++ plink = 1;
++ h_src_dentry = au_plink_lkup(inode, a->bdst);
++ err = PTR_ERR(h_src_dentry);
++ if (IS_ERR(h_src_dentry))
++ goto out;
++
++ if (unlikely(d_is_negative(h_src_dentry))) {
++ dput(h_src_dentry);
++ h_src_dentry = NULL;
++ }
++
++ }
++ if (h_src_dentry) {
++ delegated = NULL;
++ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin),
++ &a->h_path, &delegated);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
++ dput(h_src_dentry);
++ } else {
++ AuIOErr("no dentry found for hi%lu on b%d\n",
++ h_inode->i_ino, a->bdst);
++ err = -EIO;
++ }
++ }
++
++ if (!err && !plink)
++ au_plink_append(inode, a->bdst, a->h_path.dentry);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++int aufs_link(struct dentry *src_dentry, struct inode *dir,
++ struct dentry *dentry)
++{
++ int err, rerr;
++ struct au_dtime dt;
++ struct au_link_args *a;
++ struct dentry *wh_dentry, *h_src_dentry;
++ struct inode *inode, *delegated;
++ struct super_block *sb;
++ struct au_wr_dir_args wr_dir_args = {
++ /* .force_btgt = -1, */
++ .flags = AuWrDir_ADD_ENTRY
++ };
++
++ IMustLock(dir);
++ inode = d_inode(src_dentry);
++ IMustLock(inode);
++
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ a->parent = dentry->d_parent; /* dir inode is locked */
++ err = aufs_read_and_write_lock2(dentry, src_dentry,
++ AuLock_NOPLM | AuLock_GEN);
++ if (unlikely(err))
++ goto out_kfree;
++ err = au_d_linkable(src_dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ err = au_d_may_add(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++
++ a->src_parent = dget_parent(src_dentry);
++ wr_dir_args.force_btgt = au_ibtop(inode);
++
++ di_write_lock_parent(a->parent);
++ wr_dir_args.force_btgt = au_wbr(dentry, wr_dir_args.force_btgt);
++ wh_dentry = lock_hdir_lkup_wh(dentry, &dt, src_dentry, &a->pin,
++ &wr_dir_args);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ err = 0;
++ sb = dentry->d_sb;
++ a->bdst = au_dbtop(dentry);
++ a->h_path.dentry = au_h_dptr(dentry, a->bdst);
++ a->h_path.mnt = au_sbr_mnt(sb, a->bdst);
++ a->bsrc = au_ibtop(inode);
++ h_src_dentry = au_h_d_alias(src_dentry, a->bsrc);
++ if (!h_src_dentry && au_di(src_dentry)->di_tmpfile)
++ h_src_dentry = dget(au_hi_wh(inode, a->bsrc));
++ if (!h_src_dentry) {
++ a->bsrc = au_dbtop(src_dentry);
++ h_src_dentry = au_h_d_alias(src_dentry, a->bsrc);
++ AuDebugOn(!h_src_dentry);
++ } else if (IS_ERR(h_src_dentry)) {
++ err = PTR_ERR(h_src_dentry);
++ goto out_parent;
++ }
++
++ /*
++ * aufs doesn't touch the credential so
++ * security_dentry_create_files_as() is unnecessary.
++ */
++ if (au_opt_test(au_mntflags(sb), PLINK)) {
++ if (a->bdst < a->bsrc
++ /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */)
++ err = au_cpup_or_link(src_dentry, dentry, a);
++ else {
++ delegated = NULL;
++ err = vfsub_link(h_src_dentry, au_pinned_h_dir(&a->pin),
++ &a->h_path, &delegated);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
++ }
++ dput(h_src_dentry);
++ } else {
++ /*
++ * copyup src_dentry to the branch we process,
++ * and then link(2) to it.
++ */
++ dput(h_src_dentry);
++ if (a->bdst < a->bsrc
++ /* && h_src_dentry->d_sb != a->h_path.dentry->d_sb */) {
++ au_unpin(&a->pin);
++ di_write_unlock(a->parent);
++ err = au_cpup_before_link(src_dentry, a);
++ di_write_lock_parent(a->parent);
++ if (!err)
++ err = au_pin(&a->pin, dentry, a->bdst,
++ au_opt_udba(sb),
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (unlikely(err))
++ goto out_wh;
++ }
++ if (!err) {
++ h_src_dentry = au_h_dptr(src_dentry, a->bdst);
++ err = -ENOENT;
++ if (h_src_dentry && d_is_positive(h_src_dentry)) {
++ delegated = NULL;
++ err = vfsub_link(h_src_dentry,
++ au_pinned_h_dir(&a->pin),
++ &a->h_path, &delegated);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry"
++ " for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
++ }
++ }
++ }
++ if (unlikely(err))
++ goto out_unpin;
++
++ if (wh_dentry) {
++ a->h_path.dentry = wh_dentry;
++ err = au_wh_unlink_dentry(au_pinned_h_dir(&a->pin), &a->h_path,
++ dentry);
++ if (unlikely(err))
++ goto out_revert;
++ }
++
++ au_dir_ts(dir, a->bdst);
++ inode_inc_iversion(dir);
++ inc_nlink(inode);
++ inode->i_ctime = dir->i_ctime;
++ d_instantiate(dentry, au_igrab(inode));
++ if (d_unhashed(a->h_path.dentry))
++ /* some filesystem calls d_drop() */
++ d_drop(dentry);
++ /* some filesystems consume an inode even hardlink */
++ au_fhsm_wrote(sb, a->bdst, /*force*/0);
++ goto out_unpin; /* success */
++
++out_revert:
++ /* no delegation since it is just created */
++ rerr = vfsub_unlink(au_pinned_h_dir(&a->pin), &a->h_path,
++ /*delegated*/NULL, /*force*/0);
++ if (unlikely(rerr)) {
++ AuIOErr("%pd reverting failed(%d, %d)\n", dentry, err, rerr);
++ err = -EIO;
++ }
++ au_dtime_revert(&dt);
++out_unpin:
++ au_unpin(&a->pin);
++out_wh:
++ dput(wh_dentry);
++out_parent:
++ di_write_unlock(a->parent);
++ dput(a->src_parent);
++out_unlock:
++ if (unlikely(err)) {
++ au_update_dbtop(dentry);
++ d_drop(dentry);
++ }
++ aufs_read_and_write_unlock2(dentry, src_dentry);
++out_kfree:
++ au_kfree_rcu(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++int aufs_mkdir(struct mnt_idmap *idmap, struct inode *dir,
++ struct dentry *dentry, umode_t mode)
++{
++ int err, rerr;
++ aufs_bindex_t bindex;
++ unsigned char diropq;
++ struct path h_path;
++ struct dentry *wh_dentry, *parent, *opq_dentry;
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct {
++ struct au_pin pin;
++ struct au_dtime dt;
++ } *a; /* reduce the stack usage */
++ struct au_wr_dir_args wr_dir_args = {
++ .force_btgt = -1,
++ .flags = AuWrDir_ADD_ENTRY | AuWrDir_ISDIR
++ };
++
++ IMustLock(dir);
++
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
++ if (unlikely(err))
++ goto out_free;
++ err = au_d_may_add(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ wh_dentry = lock_hdir_lkup_wh(dentry, &a->dt, /*src_dentry*/NULL,
++ &a->pin, &wr_dir_args);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ sb = dentry->d_sb;
++ bindex = au_dbtop(dentry);
++ h_path.dentry = au_h_dptr(dentry, bindex);
++ h_path.mnt = au_sbr_mnt(sb, bindex);
++ err = vfsub_mkdir(au_pinned_h_dir(&a->pin), &h_path, mode);
++ if (unlikely(err))
++ goto out_unpin;
++
++ /* make the dir opaque */
++ diropq = 0;
++ h_inode = d_inode(h_path.dentry);
++ if (wh_dentry
++ || au_opt_test(au_mntflags(sb), ALWAYS_DIROPQ)) {
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ opq_dentry = au_diropq_create(dentry, bindex);
++ inode_unlock(h_inode);
++ err = PTR_ERR(opq_dentry);
++ if (IS_ERR(opq_dentry))
++ goto out_dir;
++ dput(opq_dentry);
++ diropq = 1;
++ }
++
++ err = epilog(dir, bindex, wh_dentry, dentry);
++ if (!err) {
++ inc_nlink(dir);
++ goto out_unpin; /* success */
++ }
++
++ /* revert */
++ if (diropq) {
++ AuLabel(revert opq);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ rerr = au_diropq_remove(dentry, bindex);
++ inode_unlock(h_inode);
++ if (rerr) {
++ AuIOErr("%pd reverting diropq failed(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ }
++ }
++
++out_dir:
++ AuLabel(revert dir);
++ rerr = vfsub_rmdir(au_pinned_h_dir(&a->pin), &h_path);
++ if (rerr) {
++ AuIOErr("%pd reverting dir failed(%d, %d)\n",
++ dentry, err, rerr);
++ err = -EIO;
++ }
++ au_dtime_revert(&a->dt);
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++out_parent:
++ di_write_unlock(parent);
++out_unlock:
++ if (unlikely(err)) {
++ au_update_dbtop(dentry);
++ d_drop(dentry);
++ }
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ au_kfree_rcu(a);
++out:
++ return err;
++}
+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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,1517 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * inode operations (except add/del/rename)
++ */
++
++#include <linux/device_cgroup.h>
++#include <linux/filelock.h>
++#include <linux/fs_stack.h>
++#include <linux/iversion.h>
++#include <linux/security.h>
++#include "aufs.h"
++
++static int h_permission(struct inode *h_inode, int mask,
++ struct path *h_path, int brperm)
++{
++ int err;
++ const unsigned char write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
++ struct mnt_idmap *h_idmap;
++
++ err = -EPERM;
++ if (write_mask && IS_IMMUTABLE(h_inode))
++ goto out;
++
++ err = -EACCES;
++ if (((mask & MAY_EXEC)
++ && S_ISREG(h_inode->i_mode)
++ && (path_noexec(h_path)
++ || !(h_inode->i_mode & 0111))))
++ goto out;
++
++ /*
++ * - skip the lower fs test in the case of write to ro branch.
++ * - nfs dir permission write check is optimized, but a policy for
++ * link/rename requires a real check.
++ * - nfs always sets SB_POSIXACL regardless its mount option 'noacl.'
++ * in this case, generic_permission() returns -EOPNOTSUPP.
++ */
++ h_idmap = mnt_idmap(h_path->mnt);
++ if ((write_mask && !au_br_writable(brperm))
++ || (au_test_nfs(h_inode->i_sb) && S_ISDIR(h_inode->i_mode)
++ && write_mask && !(mask & MAY_READ))
++ || !h_inode->i_op->permission) {
++ /* AuLabel(generic_permission); */
++ /* AuDbg("get_inode_acl %ps\n",
++ h_inode->i_op->get_inode_acl); */
++ err = generic_permission(h_idmap, h_inode, mask);
++ if (err == -EOPNOTSUPP && au_test_nfs_noacl(h_inode))
++ err = h_inode->i_op->permission(h_idmap, h_inode,
++ mask);
++ AuTraceErr(err);
++ } else {
++ /* AuLabel(h_inode->permission); */
++ err = h_inode->i_op->permission(h_idmap, h_inode, mask);
++ AuTraceErr(err);
++ }
++
++ if (!err)
++ err = devcgroup_inode_permission(h_inode, mask);
++ if (!err)
++ err = security_inode_permission(h_inode, mask);
++
++out:
++ return err;
++}
++
++static int aufs_permission(struct mnt_idmap *idmap, struct inode *inode,
++ int mask)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ const unsigned char isdir = !!S_ISDIR(inode->i_mode),
++ write_mask = !!(mask & (MAY_WRITE | MAY_APPEND));
++ struct inode *h_inode;
++ struct super_block *sb;
++ struct au_branch *br;
++
++ /* todo: support rcu-walk? */
++ if (mask & MAY_NOT_BLOCK)
++ return -ECHILD;
++
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ ii_read_lock_child(inode);
++#if 0 /* reserved for future use */
++ /*
++ * This test may be rather 'too much' since the test is essentially done
++ * in the aufs_lookup(). Theoretically it is possible that the inode
++ * generation doesn't match to the superblock's here. But it isn't a
++ * big deal I suppose.
++ */
++ err = au_iigen_test(inode, au_sigen(sb));
++ if (unlikely(err))
++ goto out;
++#endif
++
++ if (!isdir
++ || write_mask
++ || au_opt_test(au_mntflags(sb), DIRPERM1)) {
++ err = au_busy_or_stale();
++ h_inode = au_h_iptr(inode, au_ibtop(inode));
++ if (unlikely(!h_inode
++ || (h_inode->i_mode & S_IFMT)
++ != (inode->i_mode & S_IFMT)))
++ goto out;
++
++ err = 0;
++ bindex = au_ibtop(inode);
++ br = au_sbr(sb, bindex);
++ err = h_permission(h_inode, mask, &br->br_path, br->br_perm);
++ if (write_mask
++ && !err
++ && !special_file(h_inode->i_mode)) {
++ /* test whether the upper writable branch exists */
++ err = -EROFS;
++ for (; bindex >= 0; bindex--)
++ if (!au_br_rdonly(au_sbr(sb, bindex))) {
++ err = 0;
++ break;
++ }
++ }
++ goto out;
++ }
++
++ /* non-write to dir */
++ err = 0;
++ bbot = au_ibbot(inode);
++ for (bindex = au_ibtop(inode); !err && bindex <= bbot; bindex++) {
++ h_inode = au_h_iptr(inode, bindex);
++ if (h_inode) {
++ err = au_busy_or_stale();
++ if (unlikely(!S_ISDIR(h_inode->i_mode)))
++ break;
++
++ br = au_sbr(sb, bindex);
++ err = h_permission(h_inode, mask, &br->br_path,
++ br->br_perm);
++ }
++ }
++
++out:
++ ii_read_unlock(inode);
++ si_read_unlock(sb);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static struct dentry *aufs_lookup(struct inode *dir, struct dentry *dentry,
++ unsigned int flags)
++{
++ struct dentry *ret, *parent;
++ struct inode *inode;
++ struct super_block *sb;
++ int err, npositive;
++
++ IMustLock(dir);
++
++ /* todo: support rcu-walk? */
++ ret = ERR_PTR(-ECHILD);
++ if (flags & LOOKUP_RCU)
++ goto out;
++
++ ret = ERR_PTR(-ENAMETOOLONG);
++ if (unlikely(dentry->d_name.len > AUFS_MAX_NAMELEN))
++ goto out;
++
++ sb = dir->i_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ ret = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
++
++ err = au_di_init(dentry);
++ ret = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_si;
++
++ inode = NULL;
++ npositive = 0; /* suppress a warning */
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_read_lock_parent(parent, AuLock_IR);
++ err = au_alive_dir(parent);
++ if (!err)
++ err = au_digen_test(parent, au_sigen(sb));
++ if (!err) {
++ /* regardless LOOKUP_CREATE, always ALLOW_NEG */
++ npositive = au_lkup_dentry(dentry, au_dbtop(parent),
++ AuLkup_ALLOW_NEG);
++ err = npositive;
++ }
++ di_read_unlock(parent, AuLock_IR);
++ ret = ERR_PTR(err);
++ if (unlikely(err < 0))
++ goto out_unlock;
++
++ if (npositive) {
++ inode = au_new_inode(dentry, /*must_new*/0);
++ if (IS_ERR(inode)) {
++ ret = (void *)inode;
++ inode = NULL;
++ goto out_unlock;
++ }
++ }
++
++ if (inode)
++ atomic_inc(&inode->i_count);
++ ret = d_splice_alias(inode, dentry);
++#if 0 /* reserved for future use */
++ if (unlikely(d_need_lookup(dentry))) {
++ spin_lock(&dentry->d_lock);
++ dentry->d_flags &= ~DCACHE_NEED_LOOKUP;
++ spin_unlock(&dentry->d_lock);
++ } else
++#endif
++ if (inode) {
++ if (!IS_ERR(ret)) {
++ iput(inode);
++ if (ret && ret != dentry)
++ ii_write_unlock(inode);
++ } else {
++ ii_write_unlock(inode);
++ iput(inode);
++ inode = NULL;
++ }
++ }
++
++out_unlock:
++ di_write_unlock(dentry);
++out_si:
++ si_read_unlock(sb);
++out:
++ return ret;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * very dirty and complicated aufs ->atomic_open().
++ * aufs_atomic_open()
++ * + au_aopen_or_create()
++ * + add_simple()
++ * + vfsub_atomic_open()
++ * + branch fs ->atomic_open()
++ * may call the actual 'open' for h_file
++ * + inc br_nfiles only if opened
++ * + au_aopen_no_open() or au_aopen_do_open()
++ *
++ * au_aopen_do_open()
++ * + finish_open()
++ * + au_do_aopen()
++ * + au_do_open() the body of all 'open'
++ * + au_do_open_nondir()
++ * set the passed h_file
++ *
++ * au_aopen_no_open()
++ * + finish_no_open()
++ */
++
++struct aopen_node {
++ struct hlist_bl_node hblist;
++ struct file *file, *h_file;
++};
++
++static int au_do_aopen(struct inode *inode, struct file *file)
++{
++ struct hlist_bl_head *aopen;
++ struct hlist_bl_node *pos;
++ struct aopen_node *node;
++ struct au_do_open_args args = {
++ .aopen = 1,
++ .open = au_do_open_nondir
++ };
++
++ aopen = &au_sbi(inode->i_sb)->si_aopen;
++ hlist_bl_lock(aopen);
++ hlist_bl_for_each_entry(node, pos, aopen, hblist)
++ if (node->file == file) {
++ args.h_file = node->h_file;
++ break;
++ }
++ hlist_bl_unlock(aopen);
++ /* AuDebugOn(!args.h_file); */
++
++ return au_do_open(file, &args);
++}
++
++static int au_aopen_do_open(struct file *file, struct dentry *dentry,
++ struct aopen_node *aopen_node)
++{
++ int err;
++ struct hlist_bl_head *aopen;
++
++ AuLabel(here);
++ aopen = &au_sbi(dentry->d_sb)->si_aopen;
++ au_hbl_add(&aopen_node->hblist, aopen);
++ err = finish_open(file, dentry, au_do_aopen);
++ au_hbl_del(&aopen_node->hblist, aopen);
++ /* AuDbgFile(file); */
++ AuDbg("%pd%s%s\n", dentry,
++ (file->f_mode & FMODE_CREATED) ? " created" : "",
++ (file->f_mode & FMODE_OPENED) ? " opened" : "");
++
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_aopen_no_open(struct file *file, struct dentry *dentry)
++{
++ int err;
++
++ AuLabel(here);
++ dget(dentry);
++ err = finish_no_open(file, dentry);
++
++ AuTraceErr(err);
++ return err;
++}
++
++static int aufs_atomic_open(struct inode *dir, struct dentry *dentry,
++ struct file *file, unsigned int open_flag,
++ umode_t create_mode)
++{
++ int err, did_open;
++ unsigned int lkup_flags;
++ aufs_bindex_t bindex;
++ struct super_block *sb;
++ struct dentry *parent, *d;
++ struct vfsub_aopen_args args = {
++ .open_flag = open_flag,
++ .create_mode = create_mode
++ };
++ struct aopen_node aopen_node = {
++ .file = file
++ };
++
++ IMustLock(dir);
++ AuDbg("open_flag 0%o\n", open_flag);
++ AuDbgDentry(dentry);
++
++ err = 0;
++ if (!au_di(dentry)) {
++ lkup_flags = LOOKUP_OPEN;
++ if (open_flag & O_CREAT)
++ lkup_flags |= LOOKUP_CREATE;
++ d = aufs_lookup(dir, dentry, lkup_flags);
++ if (IS_ERR(d)) {
++ err = PTR_ERR(d);
++ AuTraceErr(err);
++ goto out;
++ } else if (d) {
++ /*
++ * obsoleted dentry found.
++ * another error will be returned later.
++ */
++ d_drop(d);
++ AuDbgDentry(d);
++ dput(d);
++ }
++ AuDbgDentry(dentry);
++ }
++
++ if (d_is_positive(dentry)
++ || d_unhashed(dentry)
++ || d_unlinked(dentry)
++ || !(open_flag & O_CREAT)) {
++ err = au_aopen_no_open(file, dentry);
++ goto out; /* success */
++ }
++
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
++ if (unlikely(err))
++ goto out;
++
++ sb = dentry->d_sb;
++ parent = dentry->d_parent; /* dir is locked */
++ di_write_lock_parent(parent);
++ err = au_lkup_dentry(dentry, /*btop*/0, AuLkup_ALLOW_NEG);
++ if (unlikely(err < 0))
++ goto out_parent;
++
++ AuDbgDentry(dentry);
++ if (d_is_positive(dentry)) {
++ err = au_aopen_no_open(file, dentry);
++ goto out_parent; /* success */
++ }
++
++ args.file = alloc_empty_file(file->f_flags, current_cred());
++ err = PTR_ERR(args.file);
++ if (IS_ERR(args.file))
++ goto out_parent;
++
++ bindex = au_dbtop(dentry);
++ err = au_aopen_or_create(dir, dentry, &args);
++ AuTraceErr(err);
++ AuDbgFile(args.file);
++ file->f_mode = args.file->f_mode & ~FMODE_OPENED;
++ did_open = !!(args.file->f_mode & FMODE_OPENED);
++ if (!did_open) {
++ fput(args.file);
++ args.file = NULL;
++ }
++ di_write_unlock(parent);
++ di_write_unlock(dentry);
++ if (unlikely(err < 0)) {
++ if (args.file)
++ fput(args.file);
++ goto out_sb;
++ }
++
++ if (!did_open)
++ err = au_aopen_no_open(file, dentry);
++ else {
++ aopen_node.h_file = args.file;
++ err = au_aopen_do_open(file, dentry, &aopen_node);
++ }
++ if (unlikely(err < 0)) {
++ if (args.file)
++ fput(args.file);
++ if (did_open)
++ au_lcnt_dec(&args.br->br_nfiles);
++ }
++ goto out_sb; /* success */
++
++out_parent:
++ di_write_unlock(parent);
++ di_write_unlock(dentry);
++out_sb:
++ si_read_unlock(sb);
++out:
++ AuTraceErr(err);
++ AuDbgFile(file);
++ return err;
++}
++
++
++/* ---------------------------------------------------------------------- */
++
++static int au_wr_dir_cpup(struct dentry *dentry, struct dentry *parent,
++ const unsigned char add_entry, aufs_bindex_t bcpup,
++ aufs_bindex_t btop)
++{
++ int err;
++ struct dentry *h_parent;
++ struct inode *h_dir;
++
++ if (add_entry)
++ IMustLock(d_inode(parent));
++ else
++ di_write_lock_parent(parent);
++
++ err = 0;
++ if (!au_h_dptr(parent, bcpup)) {
++ if (btop > bcpup)
++ err = au_cpup_dirs(dentry, bcpup);
++ else if (btop < bcpup)
++ err = au_cpdown_dirs(dentry, bcpup);
++ else
++ BUG();
++ }
++ if (!err && add_entry && !au_ftest_wrdir(add_entry, TMPFILE)) {
++ h_parent = au_h_dptr(parent, bcpup);
++ h_dir = d_inode(h_parent);
++ inode_lock_shared_nested(h_dir, AuLsc_I_PARENT);
++ err = au_lkup_neg(dentry, bcpup, /*wh*/0);
++ /* todo: no unlock here */
++ inode_unlock_shared(h_dir);
++
++ AuDbg("bcpup %d\n", bcpup);
++ if (!err) {
++ if (d_really_is_negative(dentry))
++ au_set_h_dptr(dentry, btop, NULL);
++ au_update_dbrange(dentry, /*do_put_zero*/0);
++ }
++ }
++
++ if (!add_entry)
++ di_write_unlock(parent);
++ if (!err)
++ err = bcpup; /* success */
++
++ AuTraceErr(err);
++ return err;
++}
++
++/*
++ * decide the branch and the parent dir where we will create a new entry.
++ * returns new bindex or an error.
++ * copyup the parent dir if needed.
++ */
++int au_wr_dir(struct dentry *dentry, struct dentry *src_dentry,
++ struct au_wr_dir_args *args)
++{
++ int err;
++ unsigned int flags;
++ aufs_bindex_t bcpup, btop, src_btop;
++ const unsigned char add_entry
++ = au_ftest_wrdir(args->flags, ADD_ENTRY)
++ | au_ftest_wrdir(args->flags, TMPFILE);
++ struct super_block *sb;
++ struct dentry *parent;
++ struct au_sbinfo *sbinfo;
++
++ sb = dentry->d_sb;
++ sbinfo = au_sbi(sb);
++ parent = dget_parent(dentry);
++ btop = au_dbtop(dentry);
++ bcpup = btop;
++ if (args->force_btgt < 0) {
++ if (src_dentry) {
++ src_btop = au_dbtop(src_dentry);
++ if (src_btop < btop)
++ bcpup = src_btop;
++ } else if (add_entry) {
++ flags = 0;
++ if (au_ftest_wrdir(args->flags, ISDIR))
++ au_fset_wbr(flags, DIR);
++ err = AuWbrCreate(sbinfo, dentry, flags);
++ bcpup = err;
++ }
++
++ if (bcpup < 0 || au_test_ro(sb, bcpup, d_inode(dentry))) {
++ if (add_entry)
++ err = AuWbrCopyup(sbinfo, dentry);
++ else {
++ if (!IS_ROOT(dentry)) {
++ di_read_lock_parent(parent, !AuLock_IR);
++ err = AuWbrCopyup(sbinfo, dentry);
++ di_read_unlock(parent, !AuLock_IR);
++ } else
++ err = AuWbrCopyup(sbinfo, dentry);
++ }
++ bcpup = err;
++ if (unlikely(err < 0))
++ goto out;
++ }
++ } else {
++ bcpup = args->force_btgt;
++ AuDebugOn(au_test_ro(sb, bcpup, d_inode(dentry)));
++ }
++
++ AuDbg("btop %d, bcpup %d\n", btop, bcpup);
++ err = bcpup;
++ if (bcpup == btop)
++ goto out; /* success */
++
++ /* copyup the new parent into the branch we process */
++ err = au_wr_dir_cpup(dentry, parent, add_entry, bcpup, btop);
++ if (err >= 0) {
++ if (d_really_is_negative(dentry)) {
++ au_set_h_dptr(dentry, btop, NULL);
++ au_set_dbtop(dentry, bcpup);
++ au_set_dbbot(dentry, bcpup);
++ }
++ AuDebugOn(add_entry
++ && !au_ftest_wrdir(args->flags, TMPFILE)
++ && !au_h_dptr(dentry, bcpup));
++ }
++
++out:
++ dput(parent);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++void au_pin_hdir_unlock(struct au_pin *p)
++{
++ if (p->hdir)
++ au_hn_inode_unlock(p->hdir);
++}
++
++int au_pin_hdir_lock(struct au_pin *p)
++{
++ int err;
++
++ err = 0;
++ if (!p->hdir)
++ goto out;
++
++ /* even if an error happens later, keep this lock */
++ au_hn_inode_lock_nested(p->hdir, p->lsc_hi);
++
++ err = -EBUSY;
++ if (unlikely(p->hdir->hi_inode != d_inode(p->h_parent)))
++ goto out;
++
++ err = 0;
++ if (p->h_dentry)
++ err = au_h_verify(p->h_dentry, p->udba, p->hdir->hi_inode,
++ p->h_parent, p->br);
++
++out:
++ return err;
++}
++
++int au_pin_hdir_relock(struct au_pin *p)
++{
++ int err, i;
++ struct inode *h_i;
++ struct dentry *h_d[] = {
++ p->h_dentry,
++ p->h_parent
++ };
++
++ err = au_pin_hdir_lock(p);
++ if (unlikely(err))
++ goto out;
++
++ for (i = 0; !err && i < sizeof(h_d)/sizeof(*h_d); i++) {
++ if (!h_d[i])
++ continue;
++ if (d_is_positive(h_d[i])) {
++ h_i = d_inode(h_d[i]);
++ err = !h_i->i_nlink;
++ }
++ }
++
++out:
++ return err;
++}
++
++static void au_pin_hdir_set_owner(struct au_pin *p, struct task_struct *task)
++{
++ atomic_long_set(&p->hdir->hi_inode->i_rwsem.owner, (long)task);
++}
++
++void au_pin_hdir_acquire_nest(struct au_pin *p)
++{
++ if (p->hdir) {
++ rwsem_acquire_nest(&p->hdir->hi_inode->i_rwsem.dep_map,
++ p->lsc_hi, 0, NULL, _RET_IP_);
++ au_pin_hdir_set_owner(p, current);
++ }
++}
++
++void au_pin_hdir_release(struct au_pin *p)
++{
++ if (p->hdir) {
++ au_pin_hdir_set_owner(p, p->task);
++ rwsem_release(&p->hdir->hi_inode->i_rwsem.dep_map, _RET_IP_);
++ }
++}
++
++struct dentry *au_pinned_h_parent(struct au_pin *pin)
++{
++ if (pin && pin->parent)
++ return au_h_dptr(pin->parent, pin->bindex);
++ return NULL;
++}
++
++void au_unpin(struct au_pin *p)
++{
++ if (p->hdir)
++ au_pin_hdir_unlock(p);
++ if (p->h_mnt && au_ftest_pin(p->flags, MNT_WRITE))
++ vfsub_mnt_drop_write(p->h_mnt);
++ if (!p->hdir)
++ return;
++
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ iput(p->hdir->hi_inode);
++ dput(p->parent);
++ p->parent = NULL;
++ p->hdir = NULL;
++ p->h_mnt = NULL;
++ /* do not clear p->task */
++}
++
++int au_do_pin(struct au_pin *p)
++{
++ int err;
++ struct super_block *sb;
++ struct inode *h_dir;
++
++ err = 0;
++ sb = p->dentry->d_sb;
++ p->br = au_sbr(sb, p->bindex);
++ if (IS_ROOT(p->dentry)) {
++ if (au_ftest_pin(p->flags, MNT_WRITE)) {
++ p->h_mnt = au_br_mnt(p->br);
++ err = vfsub_mnt_want_write(p->h_mnt);
++ if (unlikely(err)) {
++ au_fclr_pin(p->flags, MNT_WRITE);
++ goto out_err;
++ }
++ }
++ goto out;
++ }
++
++ p->h_dentry = NULL;
++ if (p->bindex <= au_dbbot(p->dentry))
++ p->h_dentry = au_h_dptr(p->dentry, p->bindex);
++
++ p->parent = dget_parent(p->dentry);
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_lock(p->parent, AuLock_IR, p->lsc_di);
++
++ h_dir = NULL;
++ p->h_parent = au_h_dptr(p->parent, p->bindex);
++ p->hdir = au_hi(d_inode(p->parent), p->bindex);
++ if (p->hdir)
++ h_dir = p->hdir->hi_inode;
++
++ /*
++ * udba case, or
++ * if DI_LOCKED is not set, then p->parent may be different
++ * and h_parent can be NULL.
++ */
++ if (unlikely(!p->hdir || !h_dir || !p->h_parent)) {
++ err = -EBUSY;
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ dput(p->parent);
++ p->parent = NULL;
++ goto out_err;
++ }
++
++ if (au_ftest_pin(p->flags, MNT_WRITE)) {
++ p->h_mnt = au_br_mnt(p->br);
++ err = vfsub_mnt_want_write(p->h_mnt);
++ if (unlikely(err)) {
++ au_fclr_pin(p->flags, MNT_WRITE);
++ if (!au_ftest_pin(p->flags, DI_LOCKED))
++ di_read_unlock(p->parent, AuLock_IR);
++ dput(p->parent);
++ p->parent = NULL;
++ goto out_err;
++ }
++ }
++
++ au_igrab(h_dir);
++ err = au_pin_hdir_lock(p);
++ if (!err)
++ goto out; /* success */
++
++ au_unpin(p);
++
++out_err:
++ pr_err("err %d\n", err);
++ err = au_busy_or_stale();
++out:
++ return err;
++}
++
++void au_pin_init(struct au_pin *p, struct dentry *dentry,
++ aufs_bindex_t bindex, int lsc_di, int lsc_hi,
++ unsigned int udba, unsigned char flags)
++{
++ p->dentry = dentry;
++ p->udba = udba;
++ p->lsc_di = lsc_di;
++ p->lsc_hi = lsc_hi;
++ p->flags = flags;
++ p->bindex = bindex;
++
++ p->parent = NULL;
++ p->hdir = NULL;
++ p->h_mnt = NULL;
++
++ p->h_dentry = NULL;
++ p->h_parent = NULL;
++ p->br = NULL;
++ p->task = current;
++}
++
++int au_pin(struct au_pin *pin, struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int udba, unsigned char flags)
++{
++ au_pin_init(pin, dentry, bindex, AuLsc_DI_PARENT, AuLsc_I_PARENT2,
++ udba, flags);
++ return au_do_pin(pin);
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * ->setattr() and ->getattr() are called in various cases.
++ * chmod, stat: dentry is revalidated.
++ * fchmod, fstat: file and dentry are not revalidated, additionally they may be
++ * unhashed.
++ * for ->setattr(), ia->ia_file is passed from ftruncate only.
++ */
++/* todo: consolidate with do_refresh() and simple_reval_dpath() */
++int au_reval_for_attr(struct dentry *dentry, unsigned int sigen)
++{
++ int err;
++ struct dentry *parent;
++
++ err = 0;
++ if (au_digen_test(dentry, sigen)) {
++ parent = dget_parent(dentry);
++ di_read_lock_parent(parent, AuLock_IR);
++ err = au_refresh_dentry(dentry, parent);
++ di_read_unlock(parent, AuLock_IR);
++ dput(parent);
++ }
++
++ AuTraceErr(err);
++ return err;
++}
++
++int au_pin_and_icpup(struct dentry *dentry, struct iattr *ia,
++ struct au_icpup_args *a)
++{
++ int err;
++ loff_t sz;
++ aufs_bindex_t btop, ibtop;
++ struct dentry *hi_wh, *parent;
++ struct inode *inode;
++ struct au_wr_dir_args wr_dir_args = {
++ .force_btgt = -1,
++ .flags = 0
++ };
++
++ if (d_is_dir(dentry))
++ au_fset_wrdir(wr_dir_args.flags, ISDIR);
++ /* plink or hi_wh() case */
++ btop = au_dbtop(dentry);
++ inode = d_inode(dentry);
++ ibtop = au_ibtop(inode);
++ if (btop != ibtop && !au_test_ro(inode->i_sb, ibtop, inode))
++ wr_dir_args.force_btgt = ibtop;
++ err = au_wr_dir(dentry, /*src_dentry*/NULL, &wr_dir_args);
++ if (unlikely(err < 0))
++ goto out;
++ a->btgt = err;
++ if (err != btop)
++ au_fset_icpup(a->flags, DID_CPUP);
++
++ err = 0;
++ a->pin_flags = AuPin_MNT_WRITE;
++ parent = NULL;
++ if (!IS_ROOT(dentry)) {
++ au_fset_pin(a->pin_flags, DI_LOCKED);
++ parent = dget_parent(dentry);
++ di_write_lock_parent(parent);
++ }
++
++ err = au_pin(&a->pin, dentry, a->btgt, a->udba, a->pin_flags);
++ if (unlikely(err))
++ goto out_parent;
++
++ sz = -1;
++ a->h_path.dentry = au_h_dptr(dentry, btop);
++ a->h_inode = d_inode(a->h_path.dentry);
++ if (ia && (ia->ia_valid & ATTR_SIZE)) {
++ inode_lock_shared_nested(a->h_inode, AuLsc_I_CHILD);
++ if (ia->ia_size < i_size_read(a->h_inode))
++ sz = ia->ia_size;
++ inode_unlock_shared(a->h_inode);
++ }
++
++ hi_wh = NULL;
++ if (au_ftest_icpup(a->flags, DID_CPUP) && d_unlinked(dentry)) {
++ hi_wh = au_hi_wh(inode, a->btgt);
++ if (!hi_wh) {
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = a->btgt,
++ .bsrc = -1,
++ .len = sz,
++ .pin = &a->pin
++ };
++ err = au_sio_cpup_wh(&cpg, /*file*/NULL);
++ if (unlikely(err))
++ goto out_unlock;
++ hi_wh = au_hi_wh(inode, a->btgt);
++ /* todo: revalidate hi_wh? */
++ }
++ }
++
++ if (parent) {
++ au_pin_set_parent_lflag(&a->pin, /*lflag*/0);
++ di_downgrade_lock(parent, AuLock_IR);
++ dput(parent);
++ parent = NULL;
++ }
++ if (!au_ftest_icpup(a->flags, DID_CPUP))
++ goto out; /* success */
++
++ if (!d_unhashed(dentry)) {
++ struct au_cp_generic cpg = {
++ .dentry = dentry,
++ .bdst = a->btgt,
++ .bsrc = btop,
++ .len = sz,
++ .pin = &a->pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ err = au_sio_cpup_simple(&cpg);
++ if (!err)
++ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
++ } else if (!hi_wh)
++ a->h_path.dentry = au_h_dptr(dentry, a->btgt);
++ else
++ a->h_path.dentry = hi_wh; /* do not dget here */
++
++out_unlock:
++ a->h_inode = d_inode(a->h_path.dentry);
++ if (!err)
++ goto out; /* success */
++ au_unpin(&a->pin);
++out_parent:
++ if (parent) {
++ di_write_unlock(parent);
++ dput(parent);
++ }
++out:
++ if (!err)
++ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
++ return err;
++}
++
++static int aufs_setattr(struct mnt_idmap *idmap, struct dentry *dentry,
++ struct iattr *ia)
++{
++ int err;
++ struct inode *inode, *delegated;
++ struct super_block *sb;
++ struct file *file;
++ struct au_icpup_args *a;
++ struct mnt_idmap *h_idmap;
++
++ inode = d_inode(dentry);
++ IMustLock(inode);
++
++ err = setattr_prepare(idmap, dentry, ia);
++ if (unlikely(err))
++ goto out;
++
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ if (ia->ia_valid & (ATTR_KILL_SUID | ATTR_KILL_SGID))
++ ia->ia_valid &= ~ATTR_MODE;
++
++ file = NULL;
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out_kfree;
++
++ if (ia->ia_valid & ATTR_FILE) {
++ /* currently ftruncate(2) only */
++ AuDebugOn(!d_is_reg(dentry));
++ file = ia->ia_file;
++ err = au_reval_and_lock_fdi(file, au_reopen_nondir, /*wlock*/1,
++ /*fi_lsc*/0);
++ if (unlikely(err))
++ goto out_si;
++ ia->ia_file = au_hf_top(file);
++ a->udba = AuOpt_UDBA_NONE;
++ } else {
++ /* fchmod() doesn't pass ia_file */
++ a->udba = au_opt_udba(sb);
++ di_write_lock_child(dentry);
++ /* no d_unlinked(), to set UDBA_NONE for root */
++ if (d_unhashed(dentry))
++ a->udba = AuOpt_UDBA_NONE;
++ if (a->udba != AuOpt_UDBA_NONE) {
++ AuDebugOn(IS_ROOT(dentry));
++ err = au_reval_for_attr(dentry, au_sigen(sb));
++ if (unlikely(err))
++ goto out_dentry;
++ }
++ }
++
++ err = au_pin_and_icpup(dentry, ia, a);
++ if (unlikely(err < 0))
++ goto out_dentry;
++ if (au_ftest_icpup(a->flags, DID_CPUP)) {
++ ia->ia_file = NULL;
++ ia->ia_valid &= ~ATTR_FILE;
++ }
++
++ a->h_path.mnt = au_sbr_mnt(sb, a->btgt);
++ if ((ia->ia_valid & (ATTR_MODE | ATTR_CTIME))
++ == (ATTR_MODE | ATTR_CTIME)) {
++ err = security_path_chmod(&a->h_path, ia->ia_mode);
++ if (unlikely(err))
++ goto out_unlock;
++ } else if ((ia->ia_valid & (ATTR_UID | ATTR_GID))
++ && (ia->ia_valid & ATTR_CTIME)) {
++ err = security_path_chown(&a->h_path, ia->ia_uid, ia->ia_gid);
++ if (unlikely(err))
++ goto out_unlock;
++ }
++
++ if (ia->ia_valid & ATTR_SIZE) {
++ struct file *f;
++
++ if (ia->ia_size < i_size_read(inode))
++ /* unmap only */
++ truncate_setsize(inode, ia->ia_size);
++
++ f = NULL;
++ if (ia->ia_valid & ATTR_FILE)
++ f = ia->ia_file;
++ inode_unlock(a->h_inode);
++ err = vfsub_trunc(&a->h_path, ia->ia_size, ia->ia_valid, f);
++ inode_lock_nested(a->h_inode, AuLsc_I_CHILD);
++ } else {
++ delegated = NULL;
++ while (1) {
++ err = vfsub_notify_change(&a->h_path, ia, &delegated);
++ if (delegated) {
++ err = break_deleg_wait(&delegated);
++ if (!err)
++ continue;
++ }
++ break;
++ }
++ }
++ /*
++ * regardless aufs 'acl' option setting.
++ * why don't all acl-aware fs call this func from their ->setattr()?
++ */
++ if (!err && (ia->ia_valid & ATTR_MODE)) {
++ h_idmap = mnt_idmap(a->h_path.mnt);
++ err = vfsub_acl_chmod(h_idmap, a->h_path.dentry, ia->ia_mode);
++ }
++ if (!err)
++ au_cpup_attr_changeable(inode);
++
++out_unlock:
++ inode_unlock(a->h_inode);
++ au_unpin(&a->pin);
++ if (unlikely(err))
++ au_update_dbtop(dentry);
++out_dentry:
++ di_write_unlock(dentry);
++ if (file) {
++ fi_write_unlock(file);
++ ia->ia_file = file;
++ ia->ia_valid |= ATTR_FILE;
++ }
++out_si:
++ si_read_unlock(sb);
++out_kfree:
++ au_kfree_rcu(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++#if IS_ENABLED(CONFIG_AUFS_XATTR) || IS_ENABLED(CONFIG_FS_POSIX_ACL)
++static int au_h_path_to_set_attr(struct dentry *dentry,
++ struct au_icpup_args *a, struct path *h_path)
++{
++ int err;
++ struct super_block *sb;
++
++ sb = dentry->d_sb;
++ a->udba = au_opt_udba(sb);
++ /* no d_unlinked(), to set UDBA_NONE for root */
++ if (d_unhashed(dentry))
++ a->udba = AuOpt_UDBA_NONE;
++ if (a->udba != AuOpt_UDBA_NONE) {
++ AuDebugOn(IS_ROOT(dentry));
++ err = au_reval_for_attr(dentry, au_sigen(sb));
++ if (unlikely(err))
++ goto out;
++ }
++ err = au_pin_and_icpup(dentry, /*ia*/NULL, a);
++ if (unlikely(err < 0))
++ goto out;
++
++ h_path->dentry = a->h_path.dentry;
++ h_path->mnt = au_sbr_mnt(sb, a->btgt);
++
++out:
++ return err;
++}
++
++ssize_t au_sxattr(struct dentry *dentry, struct inode *inode,
++ struct au_sxattr *arg)
++{
++ int err;
++ struct path h_path;
++ struct super_block *sb;
++ struct au_icpup_args *a;
++ struct inode *h_inode;
++ struct mnt_idmap *h_idmap;
++
++ IMustLock(inode);
++
++ err = -ENOMEM;
++ a = kzalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out_kfree;
++
++ h_path.dentry = NULL; /* silence gcc */
++ di_write_lock_child(dentry);
++ err = au_h_path_to_set_attr(dentry, a, &h_path);
++ if (unlikely(err))
++ goto out_di;
++ h_idmap = mnt_idmap(h_path.mnt);
++
++ inode_unlock(a->h_inode);
++ switch (arg->type) {
++ case AU_XATTR_SET:
++ AuDebugOn(d_is_negative(h_path.dentry));
++ err = vfsub_setxattr(h_idmap, h_path.dentry,
++ arg->u.set.name, arg->u.set.value,
++ arg->u.set.size, arg->u.set.flags);
++ break;
++ case AU_ACL_SET:
++ err = -EOPNOTSUPP;
++ h_inode = d_inode(h_path.dentry);
++ if (h_inode->i_op->set_acl) {
++ /* this will call posix_acl_update_mode */
++ err = h_inode->i_op->set_acl(h_idmap, h_path.dentry,
++ arg->u.acl_set.acl,
++ arg->u.acl_set.type);
++ }
++ break;
++ }
++ if (!err)
++ au_cpup_attr_timesizes(inode);
++
++ au_unpin(&a->pin);
++ if (unlikely(err))
++ au_update_dbtop(dentry);
++
++out_di:
++ di_write_unlock(dentry);
++ si_read_unlock(sb);
++out_kfree:
++ au_kfree_rcu(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
++#endif
++
++static void au_refresh_iattr(struct inode *inode, struct kstat *st,
++ unsigned int nlink)
++{
++ unsigned int n;
++
++ inode->i_mode = st->mode;
++ /* don't i_[ug]id_write() here */
++ inode->i_uid = st->uid;
++ inode->i_gid = st->gid;
++ inode->i_atime = st->atime;
++ inode->i_mtime = st->mtime;
++ inode->i_ctime = st->ctime;
++
++ au_cpup_attr_nlink(inode, /*force*/0);
++ if (S_ISDIR(inode->i_mode)) {
++ n = inode->i_nlink;
++ n -= nlink;
++ n += st->nlink;
++ smp_mb(); /* for i_nlink */
++ /* 0 can happen */
++ set_nlink(inode, n);
++ }
++
++ spin_lock(&inode->i_lock);
++ inode->i_blocks = st->blocks;
++ i_size_write(inode, st->size);
++ spin_unlock(&inode->i_lock);
++}
++
++/*
++ * 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, struct inode *inode, int force,
++ struct path *h_path, int locked)
++{
++ int err;
++ unsigned int mnt_flags, sigen;
++ unsigned char udba_none;
++ aufs_bindex_t bindex;
++ struct super_block *sb, *h_sb;
++
++ h_path->mnt = NULL;
++ h_path->dentry = NULL;
++
++ err = 0;
++ sb = dentry->d_sb;
++ 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);
++ err = au_digen_test(dentry, sigen);
++ if (!err) {
++ di_read_lock_child(dentry, AuLock_IR);
++ err = au_dbrange_test(dentry);
++ if (unlikely(err)) {
++ di_read_unlock(dentry, AuLock_IR);
++ goto out;
++ }
++ } else {
++ AuDebugOn(IS_ROOT(dentry));
++ di_write_lock_child(dentry);
++ err = au_dbrange_test(dentry);
++ if (!err)
++ err = au_reval_for_attr(dentry, sigen);
++ if (!err)
++ di_downgrade_lock(dentry, AuLock_IR);
++ else {
++ di_write_unlock(dentry);
++ goto out;
++ }
++ }
++ } else
++ di_read_lock_child(dentry, AuLock_IR);
++
++body:
++ if (!inode) {
++ inode = d_inode(dentry);
++ if (unlikely(!inode))
++ goto out;
++ }
++ bindex = au_ibtop(inode);
++ h_path->mnt = au_sbr_mnt(sb, bindex);
++ h_sb = h_path->mnt->mnt_sb;
++ if (!force
++ && !au_test_fs_bad_iattr(h_sb)
++ && udba_none)
++ goto out; /* success */
++
++ if (au_dbtop(dentry) == bindex)
++ h_path->dentry = au_h_dptr(dentry, bindex);
++ else if (au_opt_test(mnt_flags, PLINK) && au_plink_test(inode)) {
++ h_path->dentry = au_plink_lkup(inode, bindex);
++ if (IS_ERR(h_path->dentry))
++ /* pretending success */
++ h_path->dentry = NULL;
++ else
++ dput(h_path->dentry);
++ }
++
++out:
++ return err;
++}
++
++static int aufs_getattr(struct mnt_idmap *idmap, 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, /*inode*/NULL, /*force*/0, &h_path,
++ /*locked*/0);
++ if (unlikely(err))
++ goto out_si;
++ if (unlikely(!h_path.dentry))
++ /* illegally overlapped or something */
++ goto out_fill; /* pretending success */
++
++ positive = d_is_positive(h_path.dentry);
++ if (positive)
++ /* no vfsub version */
++ err = vfs_getattr(&h_path, st, request, query);
++ if (!err) {
++ if (positive)
++ au_refresh_iattr(inode, st,
++ d_inode(h_path.dentry)->i_nlink);
++ goto out_fill; /* success */
++ }
++ AuTraceErr(err);
++ goto out_di;
++
++out_fill:
++ generic_fillattr(idmap, inode, st);
++out_di:
++ di_read_unlock(dentry, AuLock_IR);
++out_si:
++ si_read_unlock(sb);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static const char *aufs_get_link(struct dentry *dentry, struct inode *inode,
++ struct delayed_call *done)
++{
++ const char *ret;
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++ int err;
++ aufs_bindex_t bindex;
++
++ ret = NULL; /* suppress a warning */
++ err = -ECHILD;
++ if (!dentry)
++ goto out;
++
++ err = aufs_read_lock(dentry, AuLock_IR | AuLock_GEN);
++ if (unlikely(err))
++ goto out;
++
++ err = au_d_hashed_positive(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++
++ err = -EINVAL;
++ inode = d_inode(dentry);
++ bindex = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, bindex);
++ if (unlikely(!h_inode->i_op->get_link))
++ goto out_unlock;
++
++ err = -EBUSY;
++ h_dentry = NULL;
++ if (au_dbtop(dentry) <= bindex) {
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (h_dentry)
++ dget(h_dentry);
++ }
++ if (!h_dentry) {
++ h_dentry = d_find_any_alias(h_inode);
++ if (IS_ERR(h_dentry)) {
++ err = PTR_ERR(h_dentry);
++ goto out_unlock;
++ }
++ }
++ if (unlikely(!h_dentry))
++ goto out_unlock;
++
++ err = 0;
++ AuDbg("%ps\n", h_inode->i_op->get_link);
++ AuDbgDentry(h_dentry);
++ ret = vfs_get_link(h_dentry, done);
++ dput(h_dentry);
++ if (IS_ERR(ret))
++ err = PTR_ERR(ret);
++
++out_unlock:
++ aufs_read_unlock(dentry, AuLock_IR);
++out:
++ if (unlikely(err))
++ ret = ERR_PTR(err);
++ AuTraceErrPtr(ret);
++ return 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 timespec64 *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);
++
++ 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);
++ si_read_unlock(sb);
++ lockdep_on();
++
++ if (!err && (flags & S_VERSION))
++ inode_inc_iversion(inode);
++
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* no getattr version will be set by module.c:aufs_init() */
++struct inode_operations aufs_iop_nogetattr[AuIop_Last],
++ aufs_iop[] = {
++ [AuIop_SYMLINK] = {
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_inode_acl = aufs_get_inode_acl,
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl, /* unsupport for symlink? */
++#endif
++
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
++
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
++#endif
++
++ .get_link = aufs_get_link
++
++ /* .update_time = aufs_update_time */
++ },
++ [AuIop_DIR] = {
++ .create = aufs_create,
++ .lookup = aufs_lookup,
++ .link = aufs_link,
++ .unlink = aufs_unlink,
++ .symlink = aufs_symlink,
++ .mkdir = aufs_mkdir,
++ .rmdir = aufs_rmdir,
++ .mknod = aufs_mknod,
++ .rename = aufs_rename,
++
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_inode_acl = aufs_get_inode_acl,
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl,
++#endif
++
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
++
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
++#endif
++
++ .update_time = aufs_update_time,
++ .atomic_open = aufs_atomic_open,
++ .tmpfile = aufs_tmpfile
++ },
++ [AuIop_OTHER] = {
++ .permission = aufs_permission,
++#ifdef CONFIG_FS_POSIX_ACL
++ .get_inode_acl = aufs_get_inode_acl,
++ .get_acl = aufs_get_acl,
++ .set_acl = aufs_set_acl,
++#endif
++
++ .setattr = aufs_setattr,
++ .getattr = aufs_getattr,
++
++#ifdef CONFIG_AUFS_XATTR
++ .listxattr = aufs_listxattr,
++#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 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,522 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * inode operations (del entry)
++ */
++
++#include <linux/iversion.h>
++#include "aufs.h"
++
++/*
++ * decide if a new whiteout for @dentry is necessary or not.
++ * when it is necessary, prepare the parent dir for the upper branch whose
++ * branch index is @bcpup for creation. the actual creation of the whiteout will
++ * be done by caller.
++ * return value:
++ * 0: wh is unnecessary
++ * plus: wh is necessary
++ * minus: error
++ */
++int au_wr_dir_need_wh(struct dentry *dentry, int isdir, aufs_bindex_t *bcpup)
++{
++ int need_wh, err;
++ aufs_bindex_t btop;
++ struct super_block *sb;
++
++ sb = dentry->d_sb;
++ btop = au_dbtop(dentry);
++ if (*bcpup < 0) {
++ *bcpup = btop;
++ if (au_test_ro(sb, btop, d_inode(dentry))) {
++ err = AuWbrCopyup(au_sbi(sb), dentry);
++ *bcpup = err;
++ if (unlikely(err < 0))
++ goto out;
++ }
++ } else
++ AuDebugOn(btop < *bcpup
++ || au_test_ro(sb, *bcpup, d_inode(dentry)));
++ AuDbg("bcpup %d, btop %d\n", *bcpup, btop);
++
++ if (*bcpup != btop) {
++ err = au_cpup_dirs(dentry, *bcpup);
++ if (unlikely(err))
++ goto out;
++ need_wh = 1;
++ } else {
++ struct au_dinfo *dinfo, *tmp;
++
++ need_wh = -ENOMEM;
++ dinfo = au_di(dentry);
++ tmp = au_di_alloc(sb, AuLsc_DI_TMP);
++ if (tmp) {
++ au_di_cp(tmp, dinfo);
++ au_di_swap(tmp, dinfo);
++ /* returns the number of positive dentries */
++ need_wh = au_lkup_dentry(dentry, btop + 1,
++ /* AuLkup_IGNORE_PERM */ 0);
++ au_di_swap(tmp, dinfo);
++ au_rw_write_unlock(&tmp->di_rwsem);
++ au_di_free(tmp);
++ }
++ }
++ AuDbg("need_wh %d\n", need_wh);
++ err = need_wh;
++
++out:
++ return err;
++}
++
++/*
++ * simple tests for the del-entry operations.
++ * following the checks in vfs, plus the parent-child relationship.
++ */
++int au_may_del(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent, int isdir)
++{
++ int err;
++ umode_t h_mode;
++ struct dentry *h_dentry, *h_latest;
++ struct inode *h_inode;
++ struct path h_ppath;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct mnt_idmap *h_idmap;
++
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (d_really_is_positive(dentry)) {
++ err = -ENOENT;
++ if (unlikely(d_is_negative(h_dentry)))
++ goto out;
++ h_inode = d_inode(h_dentry);
++ if (unlikely(!h_inode->i_nlink))
++ goto out;
++
++ h_mode = h_inode->i_mode;
++ if (!isdir) {
++ err = -EISDIR;
++ if (unlikely(S_ISDIR(h_mode)))
++ goto out;
++ } else if (unlikely(!S_ISDIR(h_mode))) {
++ err = -ENOTDIR;
++ goto out;
++ }
++ } else {
++ /* rename(2) case */
++ err = -EIO;
++ if (unlikely(d_is_positive(h_dentry)))
++ goto out;
++ }
++
++ err = -ENOENT;
++ /* expected parent dir is locked */
++ if (unlikely(h_parent != h_dentry->d_parent))
++ goto out;
++ err = 0;
++
++ /*
++ * rmdir a dir may break the consistency on some filesystem.
++ * let's try heavy test.
++ */
++ err = -EACCES;
++ sb = dentry->d_sb;
++ br = au_sbr(sb, bindex);
++ h_idmap = au_br_idmap(br);
++ if (unlikely(!au_opt_test(au_mntflags(sb), DIRPERM1)
++ && au_test_h_perm(h_idmap, d_inode(h_parent),
++ MAY_EXEC | MAY_WRITE)))
++ goto out;
++
++ h_ppath.dentry = h_parent;
++ h_ppath.mnt = au_br_mnt(br);
++ h_latest = au_sio_lkup_one(h_idmap, &dentry->d_name, &h_ppath);
++ err = -EIO;
++ if (IS_ERR(h_latest))
++ goto out;
++ if (h_latest == h_dentry)
++ err = 0;
++ dput(h_latest);
++
++out:
++ return err;
++}
++
++/*
++ * decide the branch where we operate for @dentry. the branch index will be set
++ * @rbcpup. after deciding it, 'pin' it and store the timestamps of the parent
++ * dir for reverting.
++ * when a new whiteout is necessary, create it.
++ */
++static struct dentry*
++lock_hdir_create_wh(struct dentry *dentry, int isdir, aufs_bindex_t *rbcpup,
++ struct au_dtime *dt, struct au_pin *pin)
++{
++ struct dentry *wh_dentry;
++ struct super_block *sb;
++ struct path h_path;
++ int err, need_wh;
++ unsigned int udba;
++ aufs_bindex_t bcpup;
++
++ need_wh = au_wr_dir_need_wh(dentry, isdir, rbcpup);
++ wh_dentry = ERR_PTR(need_wh);
++ if (unlikely(need_wh < 0))
++ goto out;
++
++ sb = dentry->d_sb;
++ udba = au_opt_udba(sb);
++ bcpup = *rbcpup;
++ err = au_pin(pin, dentry, bcpup, udba,
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out;
++
++ h_path.dentry = au_pinned_h_parent(pin);
++ if (udba != AuOpt_UDBA_NONE
++ && au_dbtop(dentry) == bcpup) {
++ err = au_may_del(dentry, bcpup, h_path.dentry, isdir);
++ wh_dentry = ERR_PTR(err);
++ if (unlikely(err))
++ goto out_unpin;
++ }
++
++ h_path.mnt = au_sbr_mnt(sb, bcpup);
++ au_dtime_store(dt, au_pinned_parent(pin), &h_path);
++ wh_dentry = NULL;
++ if (!need_wh)
++ goto out; /* success, no need to create whiteout */
++
++ wh_dentry = au_wh_create(dentry, bcpup, h_path.dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_unpin;
++
++ /* returns with the parent is locked and wh_dentry is dget-ed */
++ goto out; /* success */
++
++out_unpin:
++ au_unpin(pin);
++out:
++ return wh_dentry;
++}
++
++/*
++ * when removing a dir, rename it to a unique temporary whiteout-ed name first
++ * in order to be revertible and save time for removing many child whiteouts
++ * under the dir.
++ * returns 1 when there are too many child whiteout and caller should remove
++ * them asynchronously. returns 0 when the number of children is enough small to
++ * remove now or the branch fs is a remote fs.
++ * otherwise return an error.
++ */
++static int renwh_and_rmdir(struct dentry *dentry, aufs_bindex_t bindex,
++ struct au_nhash *whlist, struct inode *dir)
++{
++ int rmdir_later, err, dirwh;
++ struct dentry *h_dentry;
++ struct super_block *sb;
++ struct inode *inode;
++
++ sb = dentry->d_sb;
++ SiMustAnyLock(sb);
++ h_dentry = au_h_dptr(dentry, bindex);
++ err = au_whtmp_ren(h_dentry, au_sbr(sb, bindex));
++ if (unlikely(err))
++ goto out;
++
++ /* stop monitoring */
++ inode = d_inode(dentry);
++ au_hn_free(au_hi(inode, bindex));
++
++ if (!au_test_fs_remote(h_dentry->d_sb)) {
++ dirwh = au_sbi(sb)->si_dirwh;
++ rmdir_later = (dirwh <= 1);
++ if (!rmdir_later)
++ rmdir_later = au_nhash_test_longer_wh(whlist, bindex,
++ dirwh);
++ if (rmdir_later)
++ return rmdir_later;
++ }
++
++ err = au_whtmp_rmdir(dir, bindex, h_dentry, whlist);
++ if (unlikely(err)) {
++ AuIOErr("rmdir %pd, b%d failed, %d. ignored\n",
++ h_dentry, bindex, err);
++ err = 0;
++ }
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/*
++ * final procedure for deleting a entry.
++ * maintain dentry and iattr.
++ */
++static void epilog(struct inode *dir, struct dentry *dentry,
++ aufs_bindex_t bindex)
++{
++ struct inode *inode;
++
++ inode = d_inode(dentry);
++ d_drop(dentry);
++ inode->i_ctime = dir->i_ctime;
++
++ au_dir_ts(dir, bindex);
++ inode_inc_iversion(dir);
++}
++
++/*
++ * when an error happened, remove the created whiteout and revert everything.
++ */
++static int do_revert(int err, struct inode *dir, aufs_bindex_t bindex,
++ aufs_bindex_t bwh, struct dentry *wh_dentry,
++ struct dentry *dentry, struct au_dtime *dt)
++{
++ int rerr;
++ struct path h_path = {
++ .dentry = wh_dentry,
++ .mnt = au_sbr_mnt(dir->i_sb, bindex)
++ };
++
++ rerr = au_wh_unlink_dentry(au_h_iptr(dir, bindex), &h_path, dentry);
++ if (!rerr) {
++ au_set_dbwh(dentry, bwh);
++ au_dtime_revert(dt);
++ return 0;
++ }
++
++ AuIOErr("%pd reverting whiteout failed(%d, %d)\n", dentry, err, rerr);
++ return -EIO;
++}
++
++/* ---------------------------------------------------------------------- */
++
++int aufs_unlink(struct inode *dir, struct dentry *dentry)
++{
++ int err;
++ aufs_bindex_t bwh, bindex, btop;
++ struct inode *inode, *h_dir, *delegated;
++ struct dentry *parent, *wh_dentry;
++ /* to reduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct au_pin pin;
++ struct path h_path;
++ } *a;
++
++ IMustLock(dir);
++
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_GEN);
++ if (unlikely(err))
++ goto out_free;
++ err = au_d_hashed_positive(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ inode = d_inode(dentry);
++ IMustLock(inode);
++ err = -EISDIR;
++ if (unlikely(d_is_dir(dentry)))
++ goto out_unlock; /* possible? */
++
++ btop = au_dbtop(dentry);
++ bwh = au_dbwh(dentry);
++ bindex = -1;
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/0, &bindex, &a->dt,
++ &a->pin);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ a->h_path.mnt = au_sbr_mnt(dentry->d_sb, btop);
++ a->h_path.dentry = au_h_dptr(dentry, btop);
++ dget(a->h_path.dentry);
++ if (bindex == btop) {
++ h_dir = au_pinned_h_dir(&a->pin);
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, &a->h_path, &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ } else {
++ /* dir inode is locked */
++ h_dir = d_inode(wh_dentry->d_parent);
++ IMustLock(h_dir);
++ err = 0;
++ }
++
++ if (!err) {
++ vfsub_drop_nlink(inode);
++ epilog(dir, dentry, bindex);
++
++ /* update target timestamps */
++ if (bindex == btop) {
++ vfsub_update_h_iattr(&a->h_path, /*did*/NULL);
++ /*ignore*/
++ inode->i_ctime = d_inode(a->h_path.dentry)->i_ctime;
++ } else
++ /* todo: this timestamp may be reverted later */
++ inode->i_ctime = h_dir->i_ctime;
++ goto out_unpin; /* success */
++ }
++
++ /* revert */
++ if (wh_dentry) {
++ int rerr;
++
++ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
++ &a->dt);
++ if (rerr)
++ err = rerr;
++ }
++
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++ dput(a->h_path.dentry);
++out_parent:
++ di_write_unlock(parent);
++out_unlock:
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ au_kfree_rcu(a);
++out:
++ return err;
++}
++
++int aufs_rmdir(struct inode *dir, struct dentry *dentry)
++{
++ int err, rmdir_later;
++ aufs_bindex_t bwh, bindex, btop;
++ struct inode *inode;
++ struct dentry *parent, *wh_dentry, *h_dentry;
++ struct au_whtmp_rmdir *args;
++ /* to reduce stack size */
++ struct {
++ struct au_dtime dt;
++ struct au_pin pin;
++ } *a;
++
++ IMustLock(dir);
++
++ err = -ENOMEM;
++ a = kmalloc(sizeof(*a), GFP_NOFS);
++ if (unlikely(!a))
++ goto out;
++
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_GEN);
++ if (unlikely(err))
++ goto out_free;
++ err = au_alive_dir(dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ inode = d_inode(dentry);
++ IMustLock(inode);
++ err = -ENOTDIR;
++ if (unlikely(!d_is_dir(dentry)))
++ goto out_unlock; /* possible? */
++
++ err = -ENOMEM;
++ args = au_whtmp_rmdir_alloc(dir->i_sb, GFP_NOFS);
++ if (unlikely(!args))
++ goto out_unlock;
++
++ parent = dentry->d_parent; /* dir inode is locked */
++ di_write_lock_parent(parent);
++ err = au_test_empty(dentry, &args->whlist);
++ if (unlikely(err))
++ goto out_parent;
++
++ btop = au_dbtop(dentry);
++ bwh = au_dbwh(dentry);
++ bindex = -1;
++ wh_dentry = lock_hdir_create_wh(dentry, /*isdir*/1, &bindex, &a->dt,
++ &a->pin);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry))
++ goto out_parent;
++
++ h_dentry = au_h_dptr(dentry, btop);
++ dget(h_dentry);
++ rmdir_later = 0;
++ if (bindex == btop) {
++ err = renwh_and_rmdir(dentry, btop, &args->whlist, dir);
++ if (err > 0) {
++ rmdir_later = err;
++ err = 0;
++ }
++ } else {
++ /* stop monitoring */
++ au_hn_free(au_hi(inode, btop));
++
++ /* dir inode is locked */
++ IMustLock(d_inode(wh_dentry->d_parent));
++ err = 0;
++ }
++
++ if (!err) {
++ vfsub_dead_dir(inode);
++ au_set_dbdiropq(dentry, -1);
++ epilog(dir, dentry, bindex);
++
++ if (rmdir_later) {
++ au_whtmp_kick_rmdir(dir, btop, h_dentry, args);
++ args = NULL;
++ }
++
++ goto out_unpin; /* success */
++ }
++
++ /* revert */
++ AuLabel(revert);
++ if (wh_dentry) {
++ int rerr;
++
++ rerr = do_revert(err, dir, bindex, bwh, wh_dentry, dentry,
++ &a->dt);
++ if (rerr)
++ err = rerr;
++ }
++
++out_unpin:
++ au_unpin(&a->pin);
++ dput(wh_dentry);
++ dput(h_dentry);
++out_parent:
++ di_write_unlock(parent);
++ if (args)
++ au_whtmp_rmdir_free(args);
++out_unlock:
++ aufs_read_unlock(dentry, AuLock_DW);
++out_free:
++ au_kfree_rcu(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/i_op_ren.c linux/fs/aufs/i_op_ren.c
+--- /usr/share/empty/fs/aufs/i_op_ren.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/i_op_ren.c 2023-08-28 12:34:39.956636132 +0200
+@@ -0,0 +1,1257 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * inode operation (rename entry)
++ * todo: this is crazy monster
++ */
++
++#include <linux/iversion.h>
++#include "aufs.h"
++
++enum { AuSRC, AuDST, AuSrcDst };
++enum { AuPARENT, AuCHILD, AuParentChild };
++
++#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, *hinode;
++ struct au_dtime dt[AuParentChild];
++ aufs_bindex_t btop, bdiropq;
++ } sd[AuSrcDst];
++
++#define src_dentry sd[AuSRC].dentry
++#define src_dir sd[AuSRC].dir
++#define src_inode sd[AuSRC].inode
++#define src_h_dentry sd[AuSRC].h_dentry
++#define src_parent sd[AuSRC].parent
++#define src_h_parent sd[AuSRC].h_parent
++#define src_wh_dentry sd[AuSRC].wh_dentry
++#define src_hdir sd[AuSRC].hdir
++#define src_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_inode sd[AuDST].inode
++#define dst_h_dentry sd[AuDST].h_dentry
++#define dst_parent sd[AuDST].parent
++#define dst_h_parent sd[AuDST].h_parent
++#define dst_wh_dentry sd[AuDST].wh_dentry
++#define dst_hdir sd[AuDST].hdir
++#define dst_hinode sd[AuDST].hinode
++#define dst_h_dir sd[AuDST].hdir->hi_inode
++#define dst_dt sd[AuDST].dt
++#define dst_btop sd[AuDST].btop
++#define dst_bdiropq sd[AuDST].bdiropq
++
++ struct dentry *h_trap;
++ struct au_branch *br;
++ struct path h_path;
++ struct au_nhash whlist;
++ aufs_bindex_t btgt, src_bwh;
++
++ struct {
++ unsigned short auren_flags;
++ unsigned char flags; /* syscall parameter */
++ unsigned char exchange;
++ } __packed;
++
++ struct au_whtmp_rmdir *thargs;
++ struct dentry *h_dst;
++ struct au_hinode *h_root;
++};
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * functions for reverting.
++ * when an error happened in a single rename systemcall, we should revert
++ * everything as if nothing happened.
++ * we don't need to revert the copied-up/down the parent dir since they are
++ * harmless.
++ */
++
++#define RevertFailure(fmt, ...) do { \
++ AuIOErr("revert failure: " fmt " (%d, %d)\n", \
++ ##__VA_ARGS__, err, rerr); \
++ err = -EIO; \
++} while (0)
++
++static void au_ren_do_rev_diropq(int err, struct au_ren_args *a, int idx)
++{
++ int rerr;
++ struct dentry *d;
++#define src_or_dst(member) a->sd[idx].member
++
++ d = src_or_dst(dentry); /* {src,dst}_dentry */
++ au_hn_inode_lock_nested(src_or_dst(hinode), AuLsc_I_CHILD);
++ rerr = au_diropq_remove(d, a->btgt);
++ au_hn_inode_unlock(src_or_dst(hinode));
++ au_set_dbdiropq(d, src_or_dst(bdiropq));
++ if (rerr)
++ RevertFailure("remove diropq %pd", d);
++
++#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)
++{
++ int rerr;
++ struct inode *delegated;
++ struct path h_ppath = {
++ .dentry = a->src_h_parent,
++ .mnt = a->h_path.mnt
++ };
++
++ a->h_path.dentry = vfsub_lkup_one(&a->src_dentry->d_name, &h_ppath);
++ rerr = PTR_ERR(a->h_path.dentry);
++ if (IS_ERR(a->h_path.dentry)) {
++ RevertFailure("lkup one %pd", a->src_dentry);
++ return;
++ }
++
++ delegated = NULL;
++ rerr = vfsub_rename(a->dst_h_dir,
++ au_h_dptr(a->src_dentry, a->btgt),
++ a->src_h_dir, &a->h_path, &delegated, a->flags);
++ if (unlikely(rerr == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ /* au_set_h_dptr(a->src_dentry, a->btgt, NULL); */
++ if (rerr)
++ RevertFailure("rename %pd", a->src_dentry);
++}
++
++static void au_ren_rev_whtmp(int err, struct au_ren_args *a)
++{
++ int rerr;
++ struct inode *delegated;
++ struct path h_ppath = {
++ .dentry = a->dst_h_parent,
++ .mnt = a->h_path.mnt
++ };
++
++ a->h_path.dentry = vfsub_lkup_one(&a->dst_dentry->d_name, &h_ppath);
++ rerr = PTR_ERR(a->h_path.dentry);
++ if (IS_ERR(a->h_path.dentry)) {
++ RevertFailure("lkup one %pd", a->dst_dentry);
++ return;
++ }
++ if (d_is_positive(a->h_path.dentry)) {
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ return;
++ }
++
++ delegated = NULL;
++ rerr = vfsub_rename(a->dst_h_dir, a->h_dst, a->dst_h_dir, &a->h_path,
++ &delegated, a->flags);
++ if (unlikely(rerr == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ d_drop(a->h_path.dentry);
++ dput(a->h_path.dentry);
++ if (!rerr)
++ au_set_h_dptr(a->dst_dentry, a->btgt, dget(a->h_dst));
++ else
++ RevertFailure("rename %pd", a->h_dst);
++}
++
++static void au_ren_rev_whsrc(int err, struct au_ren_args *a)
++{
++ int rerr;
++
++ a->h_path.dentry = a->src_wh_dentry;
++ rerr = au_wh_unlink_dentry(a->src_h_dir, &a->h_path, a->src_dentry);
++ au_set_dbwh(a->src_dentry, a->src_bwh);
++ if (rerr)
++ RevertFailure("unlink %pd", a->src_wh_dentry);
++}
++#undef RevertFailure
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * when we have to copyup the renaming entry, do it with the rename-target name
++ * in order to minimize the cost (the later actual rename is unnecessary).
++ * otherwise rename it on the target branch.
++ */
++static int au_ren_or_cpup(struct au_ren_args *a)
++{
++ int err;
++ struct dentry *d;
++ struct inode *delegated;
++
++ d = a->src_dentry;
++ if (au_dbtop(d) == a->btgt) {
++ a->h_path.dentry = a->dst_h_dentry;
++ AuDebugOn(au_dbtop(d) != a->btgt);
++ delegated = NULL;
++ err = vfsub_rename(a->src_h_dir, au_h_dptr(d, a->btgt),
++ a->dst_h_dir, &a->h_path, &delegated,
++ a->flags);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ } else
++ BUG();
++
++ if (!err && a->h_dst)
++ /* it will be set to dinfo later */
++ dget(a->h_dst);
++
++ return err;
++}
++
++/* cf. aufs_rmdir() */
++static int au_ren_del_whtmp(struct au_ren_args *a)
++{
++ int err;
++ struct inode *dir;
++
++ dir = a->dst_dir;
++ SiMustAnyLock(dir->i_sb);
++ if (!au_nhash_test_longer_wh(&a->whlist, a->btgt,
++ au_sbi(dir->i_sb)->si_dirwh)
++ || au_test_fs_remote(a->h_dst->d_sb)) {
++ err = au_whtmp_rmdir(dir, a->btgt, a->h_dst, &a->whlist);
++ if (unlikely(err))
++ pr_warn("failed removing whtmp dir %pd (%d), "
++ "ignored.\n", a->h_dst, err);
++ } else {
++ au_nhash_wh_free(&a->thargs->whlist);
++ a->thargs->whlist = a->whlist;
++ a->whlist.nh_num = 0;
++ au_whtmp_kick_rmdir(dir, a->btgt, a->h_dst, a->thargs);
++ dput(a->h_dst);
++ a->thargs = NULL;
++ }
++
++ return 0;
++}
++
++/* make it 'opaque' dir. */
++static int au_ren_do_diropq(struct au_ren_args *a, int idx)
++{
++ int err;
++ struct dentry *d, *diropq;
++#define src_or_dst(member) a->sd[idx].member
++
++ err = 0;
++ d = src_or_dst(dentry); /* {src,dst}_dentry */
++ src_or_dst(bdiropq) = au_dbdiropq(d);
++ src_or_dst(hinode) = au_hi(src_or_dst(inode), a->btgt);
++ au_hn_inode_lock_nested(src_or_dst(hinode), AuLsc_I_CHILD);
++ diropq = au_diropq_create(d, a->btgt);
++ au_hn_inode_unlock(src_or_dst(hinode));
++ if (IS_ERR(diropq))
++ err = PTR_ERR(diropq);
++ else
++ dput(diropq);
++
++#undef src_or_dst_
++ return err;
++}
++
++static int au_ren_diropq(struct au_ren_args *a)
++{
++ int err;
++ unsigned char always;
++ struct dentry *d;
++
++ err = 0;
++ d = a->dst_dentry; /* already renamed on the branch */
++ always = !!au_opt_test(au_mntflags(d->d_sb), ALWAYS_DIROPQ);
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ && !au_ftest_ren(a->auren_flags, DIRREN)
++ && a->btgt != au_dbdiropq(a->src_dentry)
++ && (a->dst_wh_dentry
++ || a->btgt <= au_dbdiropq(d)
++ /* hide the lower to keep xino */
++ /* the lowers may not be a dir, but we hide them anyway */
++ || a->btgt < au_dbbot(d)
++ || always)) {
++ AuDbg("here\n");
++ err = au_ren_do_diropq(a, AuSRC);
++ if (unlikely(err))
++ goto out;
++ au_fset_ren(a->auren_flags, DIROPQ_SRC);
++ }
++ if (!a->exchange)
++ goto out; /* success */
++
++ d = a->src_dentry; /* already renamed on the branch */
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST)
++ && a->btgt != au_dbdiropq(a->dst_dentry)
++ && (a->btgt < au_dbdiropq(d)
++ || a->btgt < au_dbbot(d)
++ || always)) {
++ AuDbgDentry(a->src_dentry);
++ AuDbgDentry(a->dst_dentry);
++ err = au_ren_do_diropq(a, AuDST);
++ if (unlikely(err))
++ goto out_rev_src;
++ au_fset_ren(a->auren_flags, DIROPQ_DST);
++ }
++ goto out; /* success */
++
++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 /* debugging */
++ 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 */
++ err = au_ren_or_cpup(a);
++ if (unlikely(err))
++ /* leave the copied-up one */
++ goto out_whtmp;
++
++ /* make dir opaque */
++ 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;
++
++ 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 */
++
++ au_fhsm_wrote(a->src_dentry->d_sb, a->btgt, /*force*/0);
++ }
++ err = 0;
++ goto out_success;
++
++out_diropq:
++ au_ren_rev_diropq(err, a);
++out_rename:
++ au_ren_rev_rename(err, a);
++ dput(a->h_dst);
++out_whtmp:
++ if (a->thargs)
++ au_ren_rev_whtmp(err, a);
++out_whdst:
++ dput(a->dst_wh_dentry);
++ a->dst_wh_dentry = NULL;
++out_whsrc:
++ if (a->src_wh_dentry)
++ au_ren_rev_whsrc(err, a);
++out_success:
++ dput(a->src_wh_dentry);
++ dput(a->dst_wh_dentry);
++out_thargs:
++ if (a->thargs) {
++ dput(a->h_dst);
++ au_whtmp_rmdir_free(a->thargs);
++ a->thargs = NULL;
++ }
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * test if @dentry dir can be rename destination or not.
++ * success means, it is a logically empty dir.
++ */
++static int may_rename_dstdir(struct dentry *dentry, struct au_nhash *whlist)
++{
++ return au_test_empty(dentry, whlist);
++}
++
++/*
++ * test if @a->src_dentry dir can be rename source or not.
++ * if it can, return 0.
++ * success means,
++ * - it is a logically empty dir.
++ * - or, it exists on writable branch and has no children including whiteouts
++ * on the lower branch unless DIRREN is on.
++ */
++static int may_rename_srcdir(struct au_ren_args *a)
++{
++ int err;
++ unsigned int rdhash;
++ 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(sb);
++ rdhash = sbinfo->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL,
++ dentry));
++ err = au_nhash_alloc(&whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_test_empty(dentry, &whlist);
++ au_nhash_wh_free(&whlist);
++ goto out;
++ }
++
++ if (btop == au_dbtaildir(dentry))
++ return 0; /* success */
++
++ err = au_test_empty_lower(dentry);
++
++out:
++ if (err == -ENOTEMPTY) {
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ err = 0;
++ } else {
++ AuWarn1("renaming dir who has child(ren) on multiple "
++ "branches, is not supported\n");
++ err = -EXDEV;
++ }
++ }
++ return err;
++}
++
++/* side effect: sets whlist and h_dentry */
++static int au_ren_may_dir(struct au_ren_args *a)
++{
++ int err;
++ unsigned int rdhash;
++ struct dentry *d;
++
++ d = a->dst_dentry;
++ SiMustAnyLock(d->d_sb);
++
++ err = 0;
++ if (au_ftest_ren(a->auren_flags, ISDIR_DST) && a->dst_inode) {
++ rdhash = au_sbi(d->d_sb)->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(/*file*/NULL, d));
++ err = au_nhash_alloc(&a->whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++
++ if (!a->exchange) {
++ au_set_dbtop(d, a->dst_btop);
++ err = may_rename_dstdir(d, &a->whlist);
++ au_set_dbtop(d, a->btgt);
++ } else
++ err = may_rename_srcdir(a);
++ }
++ a->dst_h_dentry = au_h_dptr(d, au_dbtop(d));
++ if (unlikely(err))
++ goto out;
++
++ d = a->src_dentry;
++ a->src_h_dentry = au_h_dptr(d, au_dbtop(d));
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)) {
++ err = may_rename_srcdir(a);
++ if (unlikely(err)) {
++ au_nhash_wh_free(&a->whlist);
++ a->whlist.nh_num = 0;
++ }
++ }
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * simple tests for rename.
++ * following the checks in vfs, plus the parent-child relationship.
++ */
++static int au_may_ren(struct au_ren_args *a)
++{
++ int err, isdir;
++ struct inode *h_inode;
++
++ if (a->src_btop == a->btgt) {
++ err = au_may_del(a->src_dentry, a->btgt, a->src_h_parent,
++ au_ftest_ren(a->auren_flags, ISDIR_SRC));
++ if (unlikely(err))
++ goto out;
++ err = -EINVAL;
++ if (unlikely(a->src_h_dentry == a->h_trap))
++ goto out;
++ }
++
++ err = 0;
++ if (a->dst_btop != a->btgt)
++ goto out;
++
++ err = -ENOTEMPTY;
++ if (unlikely(a->dst_h_dentry == a->h_trap))
++ goto out;
++
++ err = -EIO;
++ isdir = !!au_ftest_ren(a->auren_flags, ISDIR_DST);
++ if (d_really_is_negative(a->dst_dentry)) {
++ if (d_is_negative(a->dst_h_dentry))
++ err = au_may_add(a->dst_dentry, a->btgt,
++ a->dst_h_parent, isdir);
++ } else {
++ if (unlikely(d_is_negative(a->dst_h_dentry)))
++ goto out;
++ h_inode = d_inode(a->dst_h_dentry);
++ if (h_inode->i_nlink)
++ err = au_may_del(a->dst_dentry, a->btgt,
++ a->dst_h_parent, isdir);
++ }
++
++out:
++ if (unlikely(err == -ENOENT || err == -EEXIST))
++ err = -EIO;
++ AuTraceErr(err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * locking order
++ * (VFS)
++ * - src_dir and dir by lock_rename()
++ * - inode if exists
++ * (aufs)
++ * - lock all
++ * + src_dentry and dentry by aufs_read_and_write_lock2() which calls,
++ * + si_read_lock
++ * + di_write_lock2_child()
++ * + di_write_lock_child()
++ * + ii_write_lock_child()
++ * + di_write_lock_child2()
++ * + ii_write_lock_child2()
++ * + src_parent and parent
++ * + di_write_lock_parent()
++ * + ii_write_lock_parent()
++ * + di_write_lock_parent2()
++ * + ii_write_lock_parent2()
++ * + lower src_dir and dir by vfsub_lock_rename()
++ * + verify the every relationships between child and parent. if any
++ * of them failed, unlock all and return -EBUSY.
++ */
++static void au_ren_unlock(struct au_ren_args *a)
++{
++ vfsub_unlock_rename(a->src_h_parent, a->src_hdir,
++ a->dst_h_parent, a->dst_hdir);
++ if (au_ftest_ren(a->auren_flags, DIRREN)
++ && a->h_root)
++ au_hn_inode_unlock(a->h_root);
++ if (au_ftest_ren(a->auren_flags, MNT_WRITE))
++ vfsub_mnt_drop_write(au_br_mnt(a->br));
++}
++
++static int au_ren_lock(struct au_ren_args *a)
++{
++ int err;
++ unsigned int udba;
++
++ err = 0;
++ a->src_h_parent = au_h_dptr(a->src_parent, a->btgt);
++ a->src_hdir = au_hi(a->src_dir, a->btgt);
++ a->dst_h_parent = au_h_dptr(a->dst_parent, a->btgt);
++ a->dst_hdir = au_hi(a->dst_dir, a->btgt);
++
++ err = vfsub_mnt_want_write(au_br_mnt(a->br));
++ if (unlikely(err))
++ goto out;
++ au_fset_ren(a->auren_flags, MNT_WRITE);
++ if (au_ftest_ren(a->auren_flags, DIRREN)) {
++ struct dentry *root;
++ struct inode *dir;
++
++ /*
++ * 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);
++ if (unlikely(a->src_hdir->hi_inode != d_inode(a->src_h_parent)
++ || a->dst_hdir->hi_inode != d_inode(a->dst_h_parent)))
++ err = au_busy_or_stale();
++ if (!err && au_dbtop(a->src_dentry) == a->btgt)
++ err = au_h_verify(a->src_h_dentry, udba,
++ d_inode(a->src_h_parent), a->src_h_parent,
++ a->br);
++ if (!err && au_dbtop(a->dst_dentry) == a->btgt)
++ err = au_h_verify(a->dst_h_dentry, udba,
++ d_inode(a->dst_h_parent), a->dst_h_parent,
++ a->br);
++ if (!err)
++ goto out; /* success */
++
++ err = au_busy_or_stale();
++ au_ren_unlock(a);
++
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static void au_ren_refresh_dir(struct au_ren_args *a)
++{
++ struct inode *dir;
++
++ dir = a->dst_dir;
++ inode_inc_iversion(dir);
++ 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 (a->exchange) {
++ dir = a->src_dir;
++ inode_inc_iversion(dir);
++ 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;
++ inode_inc_iversion(dir);
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC))
++ au_cpup_attr_nlink(dir, /*force*/1);
++ au_dir_ts(dir, a->btgt);
++}
++
++static void au_ren_refresh(struct au_ren_args *a)
++{
++ aufs_bindex_t bbot, bindex;
++ struct dentry *d, *h_d;
++ struct inode *i, *h_i;
++ struct super_block *sb;
++
++ d = a->dst_dentry;
++ d_drop(d);
++ if (a->h_dst)
++ /* already dget-ed by au_ren_or_cpup() */
++ au_set_h_dptr(d, a->btgt, a->h_dst);
++
++ i = a->dst_inode;
++ if (i) {
++ if (!a->exchange) {
++ if (!au_ftest_ren(a->auren_flags, ISDIR_DST))
++ vfsub_drop_nlink(i);
++ else {
++ vfsub_dead_dir(i);
++ au_cpup_attr_timesizes(i);
++ }
++ au_update_dbrange(d, /*do_put_zero*/1);
++ } else
++ au_cpup_attr_nlink(i, /*force*/1);
++ } else {
++ bbot = a->btgt;
++ for (bindex = au_dbtop(d); bindex < bbot; bindex++)
++ au_set_h_dptr(d, bindex, NULL);
++ bbot = au_dbbot(d);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++)
++ au_set_h_dptr(d, bindex, NULL);
++ au_update_dbrange(d, /*do_put_zero*/0);
++ }
++
++ 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);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
++ h_d = au_h_dptr(d, bindex);
++ if (h_d)
++ au_set_h_dptr(d, bindex, NULL);
++ }
++ au_set_dbbot(d, a->btgt);
++
++ sb = d->d_sb;
++ i = a->src_inode;
++ if (au_opt_test(au_mntflags(sb), PLINK) && au_plink_test(i))
++ return; /* success */
++
++ bbot = au_ibbot(i);
++ for (bindex = a->btgt + 1; bindex <= bbot; bindex++) {
++ h_i = au_h_iptr(i, bindex);
++ if (h_i) {
++ au_xino_write(sb, bindex, h_i->i_ino, /*ino*/0);
++ /* ignore this error */
++ au_set_h_iptr(i, bindex, NULL, 0);
++ }
++ }
++ au_set_ibbot(i, a->btgt);
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* mainly for link(2) and rename(2) */
++int au_wbr(struct dentry *dentry, aufs_bindex_t btgt)
++{
++ aufs_bindex_t bdiropq, bwh;
++ struct dentry *parent;
++ struct au_branch *br;
++
++ parent = dentry->d_parent;
++ IMustLock(d_inode(parent)); /* dir is locked */
++
++ bdiropq = au_dbdiropq(parent);
++ bwh = au_dbwh(dentry);
++ br = au_sbr(dentry->d_sb, btgt);
++ if (au_br_rdonly(br)
++ || (0 <= bdiropq && bdiropq < btgt)
++ || (0 <= bwh && bwh < btgt))
++ btgt = -1;
++
++ AuDbg("btgt %d\n", btgt);
++ return btgt;
++}
++
++/* sets src_btop, dst_btop and btgt */
++static int au_ren_wbr(struct au_ren_args *a)
++{
++ int err;
++ struct au_wr_dir_args wr_dir_args = {
++ /* .force_btgt = -1, */
++ .flags = AuWrDir_ADD_ENTRY
++ };
++
++ a->src_btop = au_dbtop(a->src_dentry);
++ a->dst_btop = au_dbtop(a->dst_dentry);
++ if (au_ftest_ren(a->auren_flags, ISDIR_SRC)
++ || au_ftest_ren(a->auren_flags, ISDIR_DST))
++ au_fset_wrdir(wr_dir_args.flags, ISDIR);
++ wr_dir_args.force_btgt = a->src_btop;
++ if (a->dst_inode && a->dst_btop < a->src_btop)
++ wr_dir_args.force_btgt = a->dst_btop;
++ wr_dir_args.force_btgt = au_wbr(a->dst_dentry, wr_dir_args.force_btgt);
++ err = au_wr_dir(a->dst_dentry, a->src_dentry, &wr_dir_args);
++ a->btgt = err;
++ if (a->exchange)
++ au_update_dbtop(a->dst_dentry);
++
++ return err;
++}
++
++static void au_ren_dt(struct au_ren_args *a)
++{
++ a->h_path.dentry = a->src_h_parent;
++ au_dtime_store(a->src_dt + AuPARENT, a->src_parent, &a->h_path);
++ if (!au_ftest_ren(a->auren_flags, ISSAMEDIR)) {
++ a->h_path.dentry = a->dst_h_parent;
++ au_dtime_store(a->dst_dt + AuPARENT, a->dst_parent, &a->h_path);
++ }
++
++ 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->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);
++ }
++}
++
++static void au_ren_rev_dt(int err, struct au_ren_args *a)
++{
++ struct dentry *h_d;
++ struct inode *h_inode;
++
++ au_dtime_revert(a->src_dt + AuPARENT);
++ if (!au_ftest_ren(a->auren_flags, ISSAMEDIR))
++ au_dtime_revert(a->dst_dt + AuPARENT);
++
++ 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->auren_flags, DT_DSTDIR)) {
++ h_d = a->dst_dt[AuCHILD].dt_h_path.dentry;
++ h_inode = d_inode(h_d);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ au_dtime_revert(a->dst_dt + AuCHILD);
++ inode_unlock(h_inode);
++ }
++ }
++}
++
++/* ---------------------------------------------------------------------- */
++
++int aufs_rename(struct mnt_idmap *idmap,
++ struct inode *_src_dir, struct dentry *_src_dentry,
++ struct inode *_dst_dir, struct dentry *_dst_dentry,
++ unsigned int _flags)
++{
++ int err, lock_flags;
++ void *rev;
++ /* reduce stack space */
++ struct au_ren_args *a;
++ struct au_pin pin;
++
++ 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;
++ BUILD_BUG_ON(sizeof(a->exchange) == sizeof(u8)
++ && RENAME_EXCHANGE > U8_MAX);
++ a->exchange = _flags & RENAME_EXCHANGE;
++ a->src_dir = _src_dir;
++ a->src_dentry = _src_dentry;
++ a->src_inode = NULL;
++ if (d_really_is_positive(a->src_dentry))
++ a->src_inode = d_inode(a->src_dentry);
++ a->src_parent = a->src_dentry->d_parent; /* dir inode is locked */
++ a->dst_dir = _dst_dir;
++ a->dst_dentry = _dst_dentry;
++ a->dst_inode = NULL;
++ if (d_really_is_positive(a->dst_dentry))
++ a->dst_inode = d_inode(a->dst_dentry);
++ a->dst_parent = a->dst_dentry->d_parent; /* dir inode is locked */
++ if (a->dst_inode) {
++ /*
++ * if EXCHANGE && src is non-dir && dst is dir,
++ * dst is not locked.
++ */
++ /* IMustLock(a->dst_inode); */
++ au_igrab(a->dst_inode);
++ }
++
++ err = -ENOTDIR;
++ lock_flags = AuLock_FLUSH | AuLock_NOPLM | AuLock_GEN;
++ if (d_is_dir(a->src_dentry)) {
++ au_fset_ren(a->auren_flags, ISDIR_SRC);
++ if (unlikely(!a->exchange
++ && d_really_is_positive(a->dst_dentry)
++ && !d_is_dir(a->dst_dentry)))
++ goto out_free;
++ lock_flags |= AuLock_DIRS;
++ }
++ if (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,
++ lock_flags);
++ if (unlikely(err))
++ goto out_free;
++
++ err = au_d_hashed_positive(a->src_dentry);
++ if (unlikely(err))
++ goto out_unlock;
++ err = -ENOENT;
++ if (a->dst_inode) {
++ /*
++ * If it is a dir, VFS unhash it before this
++ * function. It means we cannot rely upon d_unhashed().
++ */
++ if (unlikely(!a->dst_inode->i_nlink))
++ goto out_unlock;
++ if (!au_ftest_ren(a->auren_flags, ISDIR_DST)) {
++ err = au_d_hashed_positive(a->dst_dentry);
++ if (unlikely(err && !a->exchange))
++ goto out_unlock;
++ } else if (unlikely(IS_DEADDIR(a->dst_inode)))
++ goto out_unlock;
++ } else if (unlikely(d_unhashed(a->dst_dentry)))
++ goto out_unlock;
++
++ /*
++ * is it possible?
++ * yes, it happened (in linux-3.3-rcN) but I don't know why.
++ * there may exist a problem somewhere else.
++ */
++ err = -EINVAL;
++ if (unlikely(d_inode(a->dst_parent) == d_inode(a->src_dentry)))
++ goto out_unlock;
++
++ au_fset_ren(a->auren_flags, ISSAMEDIR); /* temporary */
++ di_write_lock_parent(a->dst_parent);
++
++ /* which branch we process */
++ err = au_ren_wbr(a);
++ if (unlikely(err < 0))
++ goto out_parent;
++ a->br = au_sbr(a->dst_dentry->d_sb, a->btgt);
++ a->h_path.mnt = au_br_mnt(a->br);
++
++ /* are they available to be renamed */
++ err = au_ren_may_dir(a);
++ if (unlikely(err))
++ goto out_children;
++
++ /* prepare the writable parent dir on the same branch */
++ if (a->dst_btop == a->btgt) {
++ au_fset_ren(a->auren_flags, WHDST);
++ } else {
++ err = au_cpup_dirs(a->dst_dentry, a->btgt);
++ if (unlikely(err))
++ goto out_children;
++ }
++
++ err = 0;
++ if (!a->exchange) {
++ if (a->src_dir != a->dst_dir) {
++ /*
++ * this temporary unlock is safe,
++ * because both dir->i_mutex are locked.
++ */
++ di_write_unlock(a->dst_parent);
++ di_write_lock_parent(a->src_parent);
++ err = au_wr_dir_need_wh(a->src_dentry,
++ au_ftest_ren(a->auren_flags,
++ ISDIR_SRC),
++ &a->btgt);
++ di_write_unlock(a->src_parent);
++ di_write_lock2_parent(a->src_parent, a->dst_parent,
++ /*isdir*/1);
++ au_fclr_ren(a->auren_flags, ISSAMEDIR);
++ } else
++ err = au_wr_dir_need_wh(a->src_dentry,
++ au_ftest_ren(a->auren_flags,
++ ISDIR_SRC),
++ &a->btgt);
++ }
++ if (unlikely(err < 0))
++ goto out_children;
++ if (err)
++ au_fset_ren(a->auren_flags, WHSRC);
++
++ /* cpup src */
++ if (a->src_btop != a->btgt) {
++ err = au_pin(&pin, a->src_dentry, a->btgt,
++ au_opt_udba(a->src_dentry->d_sb),
++ AuPin_DI_LOCKED | AuPin_MNT_WRITE);
++ if (!err) {
++ struct au_cp_generic cpg = {
++ .dentry = a->src_dentry,
++ .bdst = a->btgt,
++ .bsrc = a->src_btop,
++ .len = -1,
++ .pin = &pin,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++ AuDebugOn(au_dbtop(a->src_dentry) != a->src_btop);
++ err = au_sio_cpup_simple(&cpg);
++ au_unpin(&pin);
++ }
++ if (unlikely(err))
++ goto out_children;
++ a->src_btop = a->btgt;
++ a->src_h_dentry = au_h_dptr(a->src_dentry, a->btgt);
++ if (!a->exchange)
++ au_fset_ren(a->auren_flags, WHSRC);
++ }
++
++ /* 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 */
++ err = au_ren_lock(a);
++ if (unlikely(err))
++ /* leave the copied-up one */
++ goto out_children;
++
++ 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_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);
++
++ /* dput/iput all lower dentries */
++ au_ren_refresh(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_ren_unlock(a);
++out_children:
++ au_nhash_wh_free(&a->whlist);
++ if (err && a->dst_inode && a->dst_btop != a->btgt) {
++ AuDbg("btop %d, btgt %d\n", a->dst_btop, a->btgt);
++ au_set_h_dptr(a->dst_dentry, a->btgt, NULL);
++ au_set_dbtop(a->dst_dentry, a->dst_btop);
++ }
++out_parent:
++ if (!err) {
++ if (d_unhashed(a->src_dentry))
++ au_fset_ren(a->auren_flags, DROPPED_SRC);
++ if (d_unhashed(a->dst_dentry))
++ au_fset_ren(a->auren_flags, DROPPED_DST);
++ if (!a->exchange)
++ d_move(a->src_dentry, a->dst_dentry);
++ else {
++ d_exchange(a->src_dentry, a->dst_dentry);
++ if (au_ftest_ren(a->auren_flags, DROPPED_DST))
++ d_drop(a->dst_dentry);
++ }
++ if (au_ftest_ren(a->auren_flags, DROPPED_SRC))
++ d_drop(a->src_dentry);
++ } else {
++ au_update_dbtop(a->dst_dentry);
++ if (!a->dst_inode)
++ d_drop(a->dst_dentry);
++ }
++ if (au_ftest_ren(a->auren_flags, ISSAMEDIR))
++ di_write_unlock(a->dst_parent);
++ else
++ di_write_unlock2(a->src_parent, a->dst_parent);
++out_unlock:
++ aufs_read_and_write_unlock2(a->dst_dentry, a->src_dentry);
++out_free:
++ iput(a->dst_inode);
++ if (a->thargs)
++ au_whtmp_rmdir_free(a->thargs);
++ au_kfree_rcu(a);
++out:
++ AuTraceErr(err);
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/Kconfig linux/fs/aufs/Kconfig
+--- /usr/share/empty/fs/aufs/Kconfig 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/Kconfig 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,199 @@
++# SPDX-License-Identifier: GPL-2.0
++config AUFS_FS
++ tristate "Aufs (Advanced multi layered unification filesystem) support"
++ help
++ Aufs is a stackable unification filesystem such as Unionfs,
++ which unifies several directories and provides a merged single
++ directory.
++ In the early days, aufs was entirely re-designed and
++ re-implemented Unionfs Version 1.x series. Introducing many
++ original ideas, approaches and improvements, it becomes totally
++ different from Unionfs while keeping the basic features.
++
++if AUFS_FS
++choice
++ prompt "Maximum number of branches"
++ default AUFS_BRANCH_MAX_127
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_127
++ bool "127"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_511
++ bool "511"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_1023
++ bool "1023"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++config AUFS_BRANCH_MAX_32767
++ bool "32767"
++ help
++ Specifies the maximum number of branches (or member directories)
++ in a single aufs. The larger value consumes more system
++ resources and has a minor impact to performance.
++endchoice
++
++config AUFS_SBILIST
++ bool
++ depends on AUFS_MAGIC_SYSRQ || PROC_FS
++ default y
++ help
++ Automatic configuration for internal use.
++ When aufs supports Magic SysRq or /proc, enabled automatically.
++
++config AUFS_HNOTIFY
++ bool "Detect direct branch access (bypassing aufs)"
++ help
++ If you want to modify files on branches directly, eg. bypassing aufs,
++ and want aufs to detect the changes of them fully, then enable this
++ option and use 'udba=notify' mount option.
++ Currently there is only one available configuration, "fsnotify".
++ It will have a negative impact to the performance.
++ See detail in aufs.5.
++
++choice
++ prompt "method" if AUFS_HNOTIFY
++ default AUFS_HFSNOTIFY
++config AUFS_HFSNOTIFY
++ bool "fsnotify"
++ select FSNOTIFY
++endchoice
++
++config AUFS_EXPORT
++ bool "NFS-exportable aufs"
++ depends on EXPORTFS
++ help
++ If you want to export your mounted aufs via NFS, then enable this
++ option. There are several requirements for this configuration.
++ See detail in aufs.5.
++
++config AUFS_INO_T_64
++ bool
++ depends on AUFS_EXPORT
++ depends on 64BIT && !(ALPHA || S390)
++ default y
++ help
++ Automatic configuration for internal use.
++ /* typedef unsigned long/int __kernel_ino_t */
++ /* alpha and s390x are int */
++
++config AUFS_XATTR
++ bool "support for XATTR/EA (including Security Labels)"
++ help
++ If your branch fs supports XATTR/EA and you want to make them
++ available in aufs too, then enable this opsion and specify the
++ branch attributes for EA.
++ See detail in aufs.5.
++
++config AUFS_FHSM
++ bool "File-based Hierarchical Storage Management"
++ help
++ Hierarchical Storage Management (or HSM) is a well-known feature
++ in the storage world. Aufs provides this feature as file-based.
++ with multiple branches.
++ These multiple branches are prioritized, ie. the topmost one
++ should be the fastest drive and be used heavily.
++
++config AUFS_RDU
++ bool "Readdir in userspace"
++ help
++ Aufs has two methods to provide a merged view for a directory,
++ by a user-space library and by kernel-space natively. The latter
++ is always enabled but sometimes large and slow.
++ If you enable this option, install the library in aufs2-util
++ package, and set some environment variables for your readdir(3),
++ then the work will be handled in user-space which generally
++ shows better performance in most cases.
++ See detail in aufs.5.
++
++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 information on the writable
++ branch. Obviously this costs high (and I don't like it).
++ To use this feature, you need to enable this configuration AND
++ to specify the mount option `dirren.'
++ See details in aufs.5 and the design documents.
++
++config AUFS_SHWH
++ bool "Show whiteouts"
++ help
++ If you want to make the whiteouts in aufs visible, then enable
++ this option and specify 'shwh' mount option. Although it may
++ sounds like philosophy or something, but in technically it
++ simply shows the name of whiteout with keeping its behaviour.
++
++config AUFS_BR_RAMFS
++ bool "Ramfs (initramfs/rootfs) as an aufs branch"
++ help
++ If you want to use ramfs as an aufs branch fs, then enable this
++ option. Generally tmpfs is recommended.
++ Aufs prohibited them to be a branch fs by default, because
++ initramfs becomes unusable after switch_root or something
++ generally. If you sets initramfs as an aufs branch and boot your
++ system by switch_root, you will meet a problem easily since the
++ files in initramfs may be inaccessible.
++ Unless you are going to use ramfs as an aufs branch fs without
++ switch_root or something, leave it N.
++
++config AUFS_BR_FUSE
++ bool "Fuse fs as an aufs branch"
++ depends on FUSE_FS
++ select AUFS_POLL
++ help
++ If you want to use fuse-based userspace filesystem as an aufs
++ branch fs, then enable this option.
++ It implements the internal poll(2) operation which is
++ implemented by fuse only (curretnly).
++
++config AUFS_POLL
++ bool
++ help
++ Automatic configuration for internal use.
++
++config AUFS_BR_HFSPLUS
++ bool "Hfsplus as an aufs branch"
++ depends on HFSPLUS_FS
++ default y
++ help
++ If you want to use hfsplus fs as an aufs branch fs, then enable
++ this option. This option introduces a small overhead at
++ copying-up a file on hfsplus.
++
++config AUFS_BDEV_LOOP
++ bool
++ depends on BLK_DEV_LOOP
++ default y
++ help
++ Automatic configuration for internal use.
++ Convert =[ym] into =y.
++
++config AUFS_DEBUG
++ bool "Debug aufs"
++ help
++ Enable this to compile aufs internal debug code.
++ It will have a negative impact to the performance.
++
++config AUFS_MAGIC_SYSRQ
++ bool
++ depends on AUFS_DEBUG && MAGIC_SYSRQ
++ default y
++ help
++ Automatic configuration for internal use.
++ When aufs supports Magic SysRq, enabled automatically.
++endif
+diff -urN /usr/share/empty/fs/aufs/lcnt.h linux/fs/aufs/lcnt.h
+--- /usr/share/empty/fs/aufs/lcnt.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/lcnt.h 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,186 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2018-2022 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/>.
++ */
++
++/*
++ * simple long counter wrapper
++ */
++
++#ifndef __AUFS_LCNT_H__
++#define __AUFS_LCNT_H__
++
++#ifdef __KERNEL__
++
++#include "debug.h"
++
++#define AuLCntATOMIC 1
++#define AuLCntPCPUCNT 2
++/*
++ * why does percpu_refcount require extra synchronize_rcu()s in
++ * au_br_do_free()
++ */
++#define AuLCntPCPUREF 3
++
++/* #define AuLCntChosen AuLCntATOMIC */
++#define AuLCntChosen AuLCntPCPUCNT
++/* #define AuLCntChosen AuLCntPCPUREF */
++
++#if AuLCntChosen == AuLCntATOMIC
++#include <linux/atomic.h>
++
++typedef atomic_long_t au_lcnt_t;
++
++static inline int au_lcnt_init(au_lcnt_t *cnt, void *release __maybe_unused)
++{
++ atomic_long_set(cnt, 0);
++ return 0;
++}
++
++static inline void au_lcnt_wait_for_fin(au_lcnt_t *cnt __maybe_unused)
++{
++ /* empty */
++}
++
++static inline void au_lcnt_fin(au_lcnt_t *cnt __maybe_unused,
++ int do_sync __maybe_unused)
++{
++ /* empty */
++}
++
++static inline void au_lcnt_inc(au_lcnt_t *cnt)
++{
++ atomic_long_inc(cnt);
++}
++
++static inline void au_lcnt_dec(au_lcnt_t *cnt)
++{
++ atomic_long_dec(cnt);
++}
++
++static inline long au_lcnt_read(au_lcnt_t *cnt, int do_rev __maybe_unused)
++{
++ return atomic_long_read(cnt);
++}
++#endif
++
++#if AuLCntChosen == AuLCntPCPUCNT
++#include <linux/percpu_counter.h>
++
++typedef struct percpu_counter au_lcnt_t;
++
++static inline int au_lcnt_init(au_lcnt_t *cnt, void *release __maybe_unused)
++{
++ return percpu_counter_init(cnt, 0, GFP_NOFS);
++}
++
++static inline void au_lcnt_wait_for_fin(au_lcnt_t *cnt __maybe_unused)
++{
++ /* empty */
++}
++
++static inline void au_lcnt_fin(au_lcnt_t *cnt, int do_sync __maybe_unused)
++{
++ percpu_counter_destroy(cnt);
++}
++
++static inline void au_lcnt_inc(au_lcnt_t *cnt)
++{
++ percpu_counter_inc(cnt);
++}
++
++static inline void au_lcnt_dec(au_lcnt_t *cnt)
++{
++ percpu_counter_dec(cnt);
++}
++
++static inline long au_lcnt_read(au_lcnt_t *cnt, int do_rev __maybe_unused)
++{
++ s64 n;
++
++ n = percpu_counter_sum(cnt);
++ BUG_ON(n < 0);
++ if (LONG_MAX != LLONG_MAX
++ && n > LONG_MAX)
++ AuWarn1("%s\n", "wrap-around");
++
++ return n;
++}
++#endif
++
++#if AuLCntChosen == AuLCntPCPUREF
++#include <linux/percpu-refcount.h>
++
++typedef struct percpu_ref au_lcnt_t;
++
++static inline int au_lcnt_init(au_lcnt_t *cnt, percpu_ref_func_t *release)
++{
++ if (!release)
++ release = percpu_ref_exit;
++ return percpu_ref_init(cnt, release, /*percpu mode*/0, GFP_NOFS);
++}
++
++static inline void au_lcnt_wait_for_fin(au_lcnt_t *cnt __maybe_unused)
++{
++ synchronize_rcu();
++}
++
++static inline void au_lcnt_fin(au_lcnt_t *cnt, int do_sync)
++{
++ percpu_ref_kill(cnt);
++ if (do_sync)
++ au_lcnt_wait_for_fin(cnt);
++}
++
++static inline void au_lcnt_inc(au_lcnt_t *cnt)
++{
++ percpu_ref_get(cnt);
++}
++
++static inline void au_lcnt_dec(au_lcnt_t *cnt)
++{
++ percpu_ref_put(cnt);
++}
++
++/*
++ * avoid calling this func as possible.
++ */
++static inline long au_lcnt_read(au_lcnt_t *cnt, int do_rev)
++{
++ long l;
++
++ percpu_ref_switch_to_atomic_sync(cnt);
++ l = atomic_long_read(&cnt->count);
++ if (do_rev)
++ percpu_ref_switch_to_percpu(cnt);
++
++ /* percpu_ref is initialized by 1 instead of 0 */
++ return l - 1;
++}
++#endif
++
++#ifdef CONFIG_AUFS_DEBUG
++#define AuLCntZero(val) do { \
++ long l = val; \
++ if (l) \
++ AuDbg("%s = %ld\n", #val, l); \
++} while (0)
++#else
++#define AuLCntZero(val) do {} while (0)
++#endif
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_LCNT_H__ */
+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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,148 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * support for loopback block device as a branch
++ */
++
++#include "aufs.h"
++
++/* added into drivers/block/loop.c */
++static struct file *(*backing_file_func)(struct super_block *sb);
++
++/*
++ * test if two lower dentries have overlapping branches.
++ */
++int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_adding)
++{
++ struct super_block *h_sb;
++ struct file *backing_file;
++
++ if (unlikely(!backing_file_func)) {
++ /* don't load "loop" module here */
++ backing_file_func = symbol_get(loop_backing_file);
++ if (unlikely(!backing_file_func))
++ /* "loop" module is not loaded */
++ return 0;
++ }
++
++ h_sb = h_adding->d_sb;
++ backing_file = backing_file_func(h_sb);
++ if (!backing_file)
++ return 0;
++
++ h_adding = backing_file->f_path.dentry;
++ /*
++ * h_adding can be local NFS.
++ * in this case aufs cannot detect the loop.
++ */
++ if (unlikely(h_adding->d_sb == sb))
++ return 1;
++ return !!au_test_subdir(h_adding, sb->s_root);
++}
++
++/* true if a kernel thread named 'loop[0-9].*' accesses a file */
++int au_test_loopback_kthread(void)
++{
++ int ret;
++ struct task_struct *tsk = current;
++ char c, comm[sizeof(tsk->comm)];
++
++ ret = 0;
++ if (tsk->flags & PF_KTHREAD) {
++ get_task_comm(comm, tsk);
++ c = comm[4];
++ ret = ('0' <= c && c <= '9'
++ && !strncmp(comm, "loop", 4));
++ }
++
++ return ret;
++}
++
++/* ---------------------------------------------------------------------- */
++
++#define au_warn_loopback_step 16
++static int au_warn_loopback_nelem = au_warn_loopback_step;
++static unsigned long *au_warn_loopback_array;
++
++void au_warn_loopback(struct super_block *h_sb)
++{
++ int i, new_nelem;
++ unsigned long *a, magic;
++ static DEFINE_SPINLOCK(spin);
++
++ magic = h_sb->s_magic;
++ spin_lock(&spin);
++ a = au_warn_loopback_array;
++ for (i = 0; i < au_warn_loopback_nelem && *a; i++)
++ if (a[i] == magic) {
++ spin_unlock(&spin);
++ return;
++ }
++
++ /* h_sb is new to us, print it */
++ if (i < au_warn_loopback_nelem) {
++ a[i] = magic;
++ goto pr;
++ }
++
++ /* expand the array */
++ 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,
++ /*may_shrink*/0);
++ if (a) {
++ au_warn_loopback_nelem = new_nelem;
++ au_warn_loopback_array = a;
++ a[i] = magic;
++ goto pr;
++ }
++
++ spin_unlock(&spin);
++ AuWarn1("realloc failed, ignored\n");
++ return;
++
++pr:
++ spin_unlock(&spin);
++ pr_warn("you may want to try another patch for loopback file "
++ "on %s(0x%lx) branch\n", au_sbtype(h_sb), magic);
++}
++
++int au_loopback_init(void)
++{
++ int err;
++ struct super_block *sb __maybe_unused;
++
++ BUILD_BUG_ON(sizeof(sb->s_magic) != sizeof(*au_warn_loopback_array));
++
++ err = 0;
++ au_warn_loopback_array = kcalloc(au_warn_loopback_step,
++ sizeof(unsigned long), GFP_NOFS);
++ if (unlikely(!au_warn_loopback_array))
++ err = -ENOMEM;
++
++ return err;
++}
++
++void au_loopback_fin(void)
++{
++ if (backing_file_func)
++ symbol_put(loop_backing_file);
++ au_kfree_try_rcu(au_warn_loopback_array);
++}
+diff -urN /usr/share/empty/fs/aufs/loop.h linux/fs/aufs/loop.h
+--- /usr/share/empty/fs/aufs/loop.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/loop.h 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,55 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * support for loopback mount as a branch
++ */
++
++#ifndef __AUFS_LOOP_H__
++#define __AUFS_LOOP_H__
++
++#ifdef __KERNEL__
++
++struct dentry;
++struct super_block;
++
++#ifdef CONFIG_AUFS_BDEV_LOOP
++/* drivers/block/loop.c */
++struct file *loop_backing_file(struct super_block *sb);
++
++/* loop.c */
++int au_test_loopback_overlap(struct super_block *sb, struct dentry *h_adding);
++int au_test_loopback_kthread(void);
++void au_warn_loopback(struct super_block *h_sb);
++
++int au_loopback_init(void);
++void au_loopback_fin(void);
++#else
++AuStub(struct file *, loop_backing_file, return NULL, struct super_block *sb)
++
++AuStubInt0(au_test_loopback_overlap, struct super_block *sb,
++ struct dentry *h_adding)
++AuStubInt0(au_test_loopback_kthread, void)
++AuStubVoid(au_warn_loopback, struct super_block *h_sb)
++
++AuStubInt0(au_loopback_init, void)
++AuStubVoid(au_loopback_fin, void)
++#endif /* BLK_DEV_LOOP */
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_LOOP_H__ */
+diff -urN /usr/share/empty/fs/aufs/magic.mk linux/fs/aufs/magic.mk
+--- /usr/share/empty/fs/aufs/magic.mk 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/magic.mk 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,31 @@
++# SPDX-License-Identifier: GPL-2.0
++
++# defined in ${srctree}/fs/fuse/inode.c
++# tristate
++ifdef CONFIG_FUSE_FS
++ccflags-y += -DFUSE_SUPER_MAGIC=0x65735546
++endif
++
++# defined in ${srctree}/fs/xfs/xfs_sb.h
++# tristate
++ifdef CONFIG_XFS_FS
++ccflags-y += -DXFS_SB_MAGIC=0x58465342
++endif
++
++# defined in ${srctree}/fs/configfs/mount.c
++# tristate
++ifdef CONFIG_CONFIGFS_FS
++ccflags-y += -DCONFIGFS_MAGIC=0x62656570
++endif
++
++# defined in ${srctree}/fs/ubifs/ubifs.h
++# tristate
++ifdef CONFIG_UBIFS_FS
++ccflags-y += -DUBIFS_SUPER_MAGIC=0x24051905
++endif
++
++# defined in ${srctree}/fs/hfsplus/hfsplus_raw.h
++# tristate
++ifdef CONFIG_HFSPLUS_FS
++ccflags-y += -DHFSPLUS_SUPER_MAGIC=0x482b
++endif
+diff -urN /usr/share/empty/fs/aufs/Makefile linux/fs/aufs/Makefile
+--- /usr/share/empty/fs/aufs/Makefile 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/Makefile 2022-11-05 23:02:18.959222617 +0100
+@@ -0,0 +1,46 @@
++# SPDX-License-Identifier: GPL-2.0
++
++include ${src}/magic.mk
++ifeq (${CONFIG_AUFS_FS},m)
++include ${src}/conf.mk
++endif
++-include ${src}/priv_def.mk
++
++# cf. include/linux/kernel.h
++# enable pr_debug
++ccflags-y += -DDEBUG
++# sparse requires the full pathname
++ifdef M
++ccflags-y += -include ${M}/../../include/uapi/linux/aufs_type.h
++else
++ccflags-y += -include ${srctree}/include/uapi/linux/aufs_type.h
++endif
++
++obj-$(CONFIG_AUFS_FS) += aufs.o
++aufs-y := module.o sbinfo.o super.o branch.o xino.o sysaufs.o opts.o fsctx.o \
++ wkq.o vfsub.o dcsub.o \
++ cpup.o whout.o wbr_policy.o \
++ dinfo.o dentry.o \
++ dynop.o \
++ finfo.o file.o f_op.o \
++ dir.o vdir.o \
++ iinfo.o inode.o i_op.o i_op_add.o i_op_del.o i_op_ren.o \
++ mvdown.o ioctl.o
++
++# all are boolean
++aufs-$(CONFIG_PROC_FS) += procfs.o plink.o
++aufs-$(CONFIG_SYSFS) += sysfs.o
++aufs-$(CONFIG_DEBUG_FS) += dbgaufs.o
++aufs-$(CONFIG_AUFS_BDEV_LOOP) += loop.o
++aufs-$(CONFIG_AUFS_HNOTIFY) += hnotify.o
++aufs-$(CONFIG_AUFS_HFSNOTIFY) += hfsnotify.o
++aufs-$(CONFIG_AUFS_EXPORT) += export.o
++aufs-$(CONFIG_AUFS_XATTR) += xattr.o
++aufs-$(CONFIG_FS_POSIX_ACL) += posix_acl.o
++aufs-$(CONFIG_AUFS_DIRREN) += dirren.o
++aufs-$(CONFIG_AUFS_FHSM) += fhsm.o
++aufs-$(CONFIG_AUFS_POLL) += poll.o
++aufs-$(CONFIG_AUFS_RDU) += rdu.o
++aufs-$(CONFIG_AUFS_BR_HFSPLUS) += hfsplus.o
++aufs-$(CONFIG_AUFS_DEBUG) += debug.o
++aufs-$(CONFIG_AUFS_MAGIC_SYSRQ) += sysrq.o
+diff -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 2022-11-05 23:02:18.965889284 +0100
+@@ -0,0 +1,273 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * module global variables and operations
++ */
++
++#include <linux/module.h>
++#include <linux/seq_file.h>
++#include "aufs.h"
++
++/* shrinkable realloc */
++void *au_krealloc(void *p, unsigned int new_sz, gfp_t gfp, int may_shrink)
++{
++ size_t sz;
++ int diff;
++
++ sz = 0;
++ diff = -1;
++ if (p) {
++#if 0 /* unused */
++ if (!new_sz) {
++ au_kfree_rcu(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);
++ au_kfree_try_rcu(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_cache[AuCache_Last];
++
++static void au_cache_fin(void)
++{
++ int i;
++
++ /*
++ * Make sure all delayed rcu free inodes are flushed before we
++ * destroy cache.
++ */
++ rcu_barrier();
++
++ /* excluding AuCache_HNOTIFY */
++ BUILD_BUG_ON(AuCache_HNOTIFY + 1 != AuCache_Last);
++ for (i = 0; i < AuCache_HNOTIFY; i++) {
++ kmem_cache_destroy(au_cache[i]);
++ au_cache[i] = NULL;
++ }
++}
++
++static int __init au_cache_init(void)
++{
++ au_cache[AuCache_DINFO] = AuCacheCtor(au_dinfo, au_di_init_once);
++ if (au_cache[AuCache_DINFO])
++ /* SLAB_DESTROY_BY_RCU */
++ 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();
++ return -ENOMEM;
++}
++
++/* ---------------------------------------------------------------------- */
++
++int au_dir_roflags;
++
++#ifdef CONFIG_AUFS_SBILIST
++/*
++ * iterate_supers_type() doesn't protect us from
++ * remounting (branch management)
++ */
++struct hlist_bl_head au_sbilist;
++#endif
++
++/*
++ * functions for module interface.
++ */
++MODULE_LICENSE("GPL");
++/* MODULE_LICENSE("GPL v2"); */
++MODULE_AUTHOR("Junjiro R. Okajima <aufs-users@lists.sourceforge.net>");
++MODULE_DESCRIPTION(AUFS_NAME
++ " -- Advanced multi layered unification filesystem");
++MODULE_VERSION(AUFS_VERSION);
++MODULE_ALIAS_FS(AUFS_NAME);
++
++/* this module parameter has no meaning when SYSFS is disabled */
++int sysaufs_brs = 1;
++MODULE_PARM_DESC(brs, "use <sysfs>/fs/aufs/si_*/brN");
++module_param_named(brs, sysaufs_brs, int, 0444);
++
++/* this module parameter has no meaning when USER_NS is disabled */
++bool au_userns;
++MODULE_PARM_DESC(allow_userns, "allow unprivileged to mount under userns");
++module_param_named(allow_userns, au_userns, bool, 0444);
++
++/* ---------------------------------------------------------------------- */
++
++static char au_esc_chars[0x20 + 3]; /* 0x01-0x20, backslash, del, and NULL */
++
++int au_seq_path(struct seq_file *seq, struct path *path)
++{
++ int err;
++
++ err = seq_path(seq, path, au_esc_chars);
++ if (err >= 0)
++ err = 0;
++ else
++ err = -ENOMEM;
++
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static int __init aufs_init(void)
++{
++ int err, i;
++ char *p;
++
++ p = au_esc_chars;
++ for (i = 1; i <= ' '; i++)
++ *p++ = i;
++ *p++ = '\\';
++ *p++ = '\x7f';
++ *p = 0;
++
++ au_dir_roflags = au_file_roflags(O_DIRECTORY | O_LARGEFILE);
++
++ memcpy(aufs_iop_nogetattr, aufs_iop, sizeof(aufs_iop));
++ 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();
++ au_dy_init();
++ err = sysaufs_init();
++ if (unlikely(err))
++ goto out;
++ err = dbgaufs_init();
++ if (unlikely(err))
++ goto out_sysaufs;
++ err = au_procfs_init();
++ if (unlikely(err))
++ goto out_dbgaufs;
++ err = au_wkq_init();
++ if (unlikely(err))
++ goto out_procfs;
++ err = au_loopback_init();
++ if (unlikely(err))
++ goto out_wkq;
++ err = au_hnotify_init();
++ if (unlikely(err))
++ goto out_loopback;
++ err = au_sysrq_init();
++ if (unlikely(err))
++ goto out_hin;
++ err = au_cache_init();
++ if (unlikely(err))
++ goto out_sysrq;
++
++ aufs_fs_type.fs_flags |= au_userns ? FS_USERNS_MOUNT : 0;
++ err = register_filesystem(&aufs_fs_type);
++ if (unlikely(err))
++ goto out_cache;
++
++ /* since we define pr_fmt, call printk directly */
++ printk(KERN_INFO AUFS_NAME " " AUFS_VERSION "\n");
++ goto out; /* success */
++
++out_cache:
++ au_cache_fin();
++out_sysrq:
++ au_sysrq_fin();
++out_hin:
++ au_hnotify_fin();
++out_loopback:
++ au_loopback_fin();
++out_wkq:
++ au_wkq_fin();
++out_procfs:
++ au_procfs_fin();
++out_dbgaufs:
++ dbgaufs_fin();
++out_sysaufs:
++ sysaufs_fin();
++ au_dy_fin();
++out:
++ return err;
++}
++
++static void __exit aufs_exit(void)
++{
++ unregister_filesystem(&aufs_fs_type);
++ au_cache_fin();
++ au_sysrq_fin();
++ au_hnotify_fin();
++ au_loopback_fin();
++ au_wkq_fin();
++ au_procfs_fin();
++ dbgaufs_fin();
++ sysaufs_fin();
++ au_dy_fin();
++}
++
++module_init(aufs_init);
++module_exit(aufs_exit);
+diff -urN /usr/share/empty/fs/aufs/module.h linux/fs/aufs/module.h
+--- /usr/share/empty/fs/aufs/module.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/module.h 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,180 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * module initialization and module-global
++ */
++
++#ifndef __AUFS_MODULE_H__
++#define __AUFS_MODULE_H__
++
++#ifdef __KERNEL__
++
++#include <linux/slab.h>
++#include "debug.h"
++#include "dentry.h"
++#include "dir.h"
++#include "file.h"
++#include "inode.h"
++
++struct path;
++struct seq_file;
++
++/* module parameters */
++extern int sysaufs_brs;
++extern bool au_userns;
++
++/* ---------------------------------------------------------------------- */
++
++extern int au_dir_roflags;
++
++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);
++
++/*
++ * Comparing the size of the object with sizeof(struct rcu_head)
++ * case 1: object is always larger
++ * --> au_kfree_rcu() or au_kfree_do_rcu()
++ * case 2: object is always smaller
++ * --> au_kfree_small()
++ * case 3: object can be any size
++ * --> au_kfree_try_rcu()
++ */
++
++static inline void au_kfree_do_rcu(const void *p)
++{
++ struct {
++ struct rcu_head rcu;
++ } *a = (void *)p;
++
++ kfree_rcu(a, rcu);
++}
++
++#define au_kfree_rcu(_p) do { \
++ typeof(_p) p = (_p); \
++ BUILD_BUG_ON(sizeof(*p) < sizeof(struct rcu_head)); \
++ if (p) \
++ au_kfree_do_rcu(p); \
++ } while (0)
++
++#define au_kfree_do_sz_test(sz) (sz >= sizeof(struct rcu_head))
++#define au_kfree_sz_test(p) (p && au_kfree_do_sz_test(ksize(p)))
++
++static inline void au_kfree_try_rcu(const void *p)
++{
++ if (!p)
++ return;
++ if (au_kfree_sz_test(p))
++ au_kfree_do_rcu(p);
++ else
++ kfree(p);
++}
++
++static inline void au_kfree_small(const void *p)
++{
++ if (!p)
++ return;
++ AuDebugOn(au_kfree_sz_test(p));
++ kfree(p);
++}
++
++static inline int au_kmidx_sub(size_t sz, size_t new_sz)
++{
++#ifndef CONFIG_SLOB
++ return __kmalloc_index(sz, false) - __kmalloc_index(new_sz, false);
++#else
++ return -1; /* SLOB is untested */
++#endif
++}
++
++int au_seq_path(struct seq_file *seq, struct path *path);
++
++#ifdef CONFIG_PROC_FS
++/* procfs.c */
++int __init au_procfs_init(void);
++void au_procfs_fin(void);
++#else
++AuStubInt0(au_procfs_init, void);
++AuStubVoid(au_procfs_fin, void);
++#endif
++
++/* ---------------------------------------------------------------------- */
++
++/* kmem cache */
++enum {
++ AuCache_DINFO,
++ AuCache_ICNTNR,
++ AuCache_FINFO,
++ AuCache_VDIR,
++ AuCache_DEHSTR,
++ AuCache_HNOTIFY, /* must be last */
++ 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)
++
++#define AuCacheFuncAlloc(name, index) \
++ static inline struct au_##name *au_cache_alloc_##name(void) \
++ { return kmem_cache_alloc(au_cache[AuCache_##index], GFP_NOFS); }
++
++#define AuCacheFuncs(name, index) \
++ static inline void au_cache_free_##name##_norcu(struct au_##name *p) \
++ { kmem_cache_free(au_cache[AuCache_##index], p); } \
++ \
++ static inline void au_cache_free_##name##_rcu_cb(struct rcu_head *rcu) \
++ { void *p = rcu; \
++ p -= offsetof(struct au_##name, rcu); \
++ kmem_cache_free(au_cache[AuCache_##index], p); } \
++ static inline void au_cache_free_##name##_rcu(struct au_##name *p) \
++ { BUILD_BUG_ON(sizeof(struct au_##name) < sizeof(struct rcu_head)); \
++ call_rcu(&p->rcu, au_cache_free_##name##_rcu_cb); } \
++ \
++ static inline void au_cache_free_##name(struct au_##name *p) \
++ { /* au_cache_free_##name##_norcu(p); */ \
++ au_cache_free_##name##_rcu(p); }
++
++AuCacheFuncs(dinfo, DINFO);
++AuCacheFuncAlloc(dinfo, DINFO);
++
++AuCacheFuncs(icntnr, ICNTNR);
++static inline struct au_icntnr *au_cache_alloc_icntnr(struct super_block *sb)
++{ return alloc_inode_sb(sb, au_cache[AuCache_ICNTNR], GFP_NOFS); }
++
++AuCacheFuncs(finfo, FINFO);
++AuCacheFuncAlloc(finfo, FINFO);
++
++AuCacheFuncs(vdir, VDIR);
++AuCacheFuncAlloc(vdir, VDIR);
++
++AuCacheFuncs(vdir_dehstr, DEHSTR);
++AuCacheFuncAlloc(vdir_dehstr, DEHSTR);
++
++#ifdef CONFIG_AUFS_HNOTIFY
++AuCacheFuncs(hnotify, HNOTIFY);
++AuCacheFuncAlloc(hnotify, HNOTIFY);
++#endif
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_MODULE_H__ */
+diff -urN /usr/share/empty/fs/aufs/mvdown.c linux/fs/aufs/mvdown.c
+--- /usr/share/empty/fs/aufs/mvdown.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/mvdown.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,706 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2011-2022 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/>.
++ */
++
++/*
++ * move-down, opposite of copy-up
++ */
++
++#include "aufs.h"
++
++struct au_mvd_args {
++ struct {
++ struct super_block *h_sb;
++ struct dentry *h_parent;
++ struct au_hinode *hdir;
++ struct inode *h_dir, *h_inode;
++ struct au_pin pin;
++ } info[AUFS_MVDOWN_NARRAY];
++
++ struct aufs_mvdown mvdown;
++ struct dentry *dentry, *parent;
++ struct inode *inode, *dir;
++ struct super_block *sb;
++ aufs_bindex_t bopq, bwh, bfound;
++ unsigned char rename_lock;
++};
++
++#define mvd_errno mvdown.au_errno
++#define mvd_bsrc mvdown.stbr[AUFS_MVDOWN_UPPER].bindex
++#define mvd_src_brid mvdown.stbr[AUFS_MVDOWN_UPPER].brid
++#define mvd_bdst mvdown.stbr[AUFS_MVDOWN_LOWER].bindex
++#define mvd_dst_brid mvdown.stbr[AUFS_MVDOWN_LOWER].brid
++
++#define mvd_h_src_sb info[AUFS_MVDOWN_UPPER].h_sb
++#define mvd_h_src_parent info[AUFS_MVDOWN_UPPER].h_parent
++#define mvd_hdir_src info[AUFS_MVDOWN_UPPER].hdir
++#define mvd_h_src_dir info[AUFS_MVDOWN_UPPER].h_dir
++#define mvd_h_src_inode info[AUFS_MVDOWN_UPPER].h_inode
++#define mvd_pin_src info[AUFS_MVDOWN_UPPER].pin
++
++#define mvd_h_dst_sb info[AUFS_MVDOWN_LOWER].h_sb
++#define mvd_h_dst_parent info[AUFS_MVDOWN_LOWER].h_parent
++#define mvd_hdir_dst info[AUFS_MVDOWN_LOWER].hdir
++#define mvd_h_dst_dir info[AUFS_MVDOWN_LOWER].h_dir
++#define mvd_h_dst_inode info[AUFS_MVDOWN_LOWER].h_inode
++#define mvd_pin_dst info[AUFS_MVDOWN_LOWER].pin
++
++#define AU_MVD_PR(flag, ...) do { \
++ if (flag) \
++ pr_err(__VA_ARGS__); \
++ } while (0)
++
++static int find_lower_writable(struct au_mvd_args *a)
++{
++ struct super_block *sb;
++ aufs_bindex_t bindex, bbot;
++ struct au_branch *br;
++
++ sb = a->sb;
++ bindex = a->mvd_bsrc;
++ bbot = au_sbbot(sb);
++ if (a->mvdown.flags & AUFS_MVDOWN_FHSM_LOWER)
++ for (bindex++; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm)
++ && !sb_rdonly(au_br_sb(br)))
++ return bindex;
++ }
++ else if (!(a->mvdown.flags & AUFS_MVDOWN_ROLOWER))
++ for (bindex++; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (!au_br_rdonly(br))
++ return bindex;
++ }
++ else
++ for (bindex++; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (!sb_rdonly(au_br_sb(br))) {
++ if (au_br_rdonly(br))
++ a->mvdown.flags
++ |= AUFS_MVDOWN_ROLOWER_R;
++ return bindex;
++ }
++ }
++
++ return -1;
++}
++
++/* make the parent dir on bdst */
++static int au_do_mkdir(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++
++ err = 0;
++ a->mvd_hdir_src = au_hi(a->dir, a->mvd_bsrc);
++ a->mvd_hdir_dst = au_hi(a->dir, a->mvd_bdst);
++ a->mvd_h_src_parent = au_h_dptr(a->parent, a->mvd_bsrc);
++ a->mvd_h_dst_parent = NULL;
++ if (au_dbbot(a->parent) >= a->mvd_bdst)
++ a->mvd_h_dst_parent = au_h_dptr(a->parent, a->mvd_bdst);
++ if (!a->mvd_h_dst_parent) {
++ err = au_cpdown_dirs(a->dentry, a->mvd_bdst);
++ if (unlikely(err)) {
++ AU_MVD_PR(dmsg, "cpdown_dirs failed\n");
++ goto out;
++ }
++ a->mvd_h_dst_parent = au_h_dptr(a->parent, a->mvd_bdst);
++ }
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/* lock them all */
++static int au_do_lock(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct dentry *h_trap;
++
++ a->mvd_h_src_sb = au_sbr_sb(a->sb, a->mvd_bsrc);
++ a->mvd_h_dst_sb = au_sbr_sb(a->sb, a->mvd_bdst);
++ err = au_pin(&a->mvd_pin_dst, a->dentry, a->mvd_bdst,
++ au_opt_udba(a->sb),
++ AuPin_MNT_WRITE | AuPin_DI_LOCKED);
++ AuTraceErr(err);
++ if (unlikely(err)) {
++ AU_MVD_PR(dmsg, "pin_dst failed\n");
++ goto out;
++ }
++
++ if (a->mvd_h_src_sb != a->mvd_h_dst_sb) {
++ a->rename_lock = 0;
++ au_pin_init(&a->mvd_pin_src, a->dentry, a->mvd_bsrc,
++ AuLsc_DI_PARENT, AuLsc_I_PARENT3,
++ au_opt_udba(a->sb),
++ AuPin_MNT_WRITE | AuPin_DI_LOCKED);
++ err = au_do_pin(&a->mvd_pin_src);
++ AuTraceErr(err);
++ a->mvd_h_src_dir = d_inode(a->mvd_h_src_parent);
++ if (unlikely(err)) {
++ AU_MVD_PR(dmsg, "pin_src failed\n");
++ goto out_dst;
++ }
++ goto out; /* success */
++ }
++
++ a->rename_lock = 1;
++ au_pin_hdir_unlock(&a->mvd_pin_dst);
++ err = au_pin(&a->mvd_pin_src, a->dentry, a->mvd_bsrc,
++ au_opt_udba(a->sb),
++ AuPin_MNT_WRITE | AuPin_DI_LOCKED);
++ AuTraceErr(err);
++ a->mvd_h_src_dir = d_inode(a->mvd_h_src_parent);
++ if (unlikely(err)) {
++ AU_MVD_PR(dmsg, "pin_src failed\n");
++ au_pin_hdir_lock(&a->mvd_pin_dst);
++ goto out_dst;
++ }
++ au_pin_hdir_unlock(&a->mvd_pin_src);
++ h_trap = vfsub_lock_rename(a->mvd_h_src_parent, a->mvd_hdir_src,
++ a->mvd_h_dst_parent, a->mvd_hdir_dst);
++ if (h_trap) {
++ err = (h_trap != a->mvd_h_src_parent);
++ if (err)
++ err = (h_trap != a->mvd_h_dst_parent);
++ }
++ BUG_ON(err); /* it should never happen */
++ if (unlikely(a->mvd_h_src_dir != au_pinned_h_dir(&a->mvd_pin_src))) {
++ err = -EBUSY;
++ AuTraceErr(err);
++ vfsub_unlock_rename(a->mvd_h_src_parent, a->mvd_hdir_src,
++ a->mvd_h_dst_parent, a->mvd_hdir_dst);
++ au_pin_hdir_lock(&a->mvd_pin_src);
++ au_unpin(&a->mvd_pin_src);
++ au_pin_hdir_lock(&a->mvd_pin_dst);
++ goto out_dst;
++ }
++ goto out; /* success */
++
++out_dst:
++ au_unpin(&a->mvd_pin_dst);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++static void au_do_unlock(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ if (!a->rename_lock)
++ au_unpin(&a->mvd_pin_src);
++ else {
++ vfsub_unlock_rename(a->mvd_h_src_parent, a->mvd_hdir_src,
++ a->mvd_h_dst_parent, a->mvd_hdir_dst);
++ au_pin_hdir_lock(&a->mvd_pin_src);
++ au_unpin(&a->mvd_pin_src);
++ au_pin_hdir_lock(&a->mvd_pin_dst);
++ }
++ au_unpin(&a->mvd_pin_dst);
++}
++
++/* copy-down the file */
++static int au_do_cpdown(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct au_cp_generic cpg = {
++ .dentry = a->dentry,
++ .bdst = a->mvd_bdst,
++ .bsrc = a->mvd_bsrc,
++ .len = -1,
++ .pin = &a->mvd_pin_dst,
++ .flags = AuCpup_DTIME | AuCpup_HOPEN
++ };
++
++ AuDbg("b%d, b%d\n", cpg.bsrc, cpg.bdst);
++ if (a->mvdown.flags & AUFS_MVDOWN_OWLOWER)
++ au_fset_cpup(cpg.flags, OVERWRITE);
++ if (a->mvdown.flags & AUFS_MVDOWN_ROLOWER)
++ au_fset_cpup(cpg.flags, RWDST);
++ err = au_sio_cpdown_simple(&cpg);
++ if (unlikely(err))
++ AU_MVD_PR(dmsg, "cpdown failed\n");
++
++ AuTraceErr(err);
++ return err;
++}
++
++/*
++ * unlink the whiteout on bdst if exist which may be created by UDBA while we
++ * were sleeping
++ */
++static int au_do_unlink_wh(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct path h_path;
++ struct au_branch *br;
++ struct inode *delegated;
++
++ br = au_sbr(a->sb, a->mvd_bdst);
++ h_path.dentry = au_wh_lkup(a->mvd_h_dst_parent, &a->dentry->d_name, br);
++ err = PTR_ERR(h_path.dentry);
++ if (IS_ERR(h_path.dentry)) {
++ AU_MVD_PR(dmsg, "wh_lkup failed\n");
++ goto out;
++ }
++
++ err = 0;
++ if (d_is_positive(h_path.dentry)) {
++ h_path.mnt = au_br_mnt(br);
++ delegated = NULL;
++ err = vfsub_unlink(d_inode(a->mvd_h_dst_parent), &h_path,
++ &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ if (unlikely(err))
++ AU_MVD_PR(dmsg, "wh_unlink failed\n");
++ }
++ dput(h_path.dentry);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/*
++ * unlink the topmost h_dentry
++ */
++static int au_do_unlink(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct path h_path;
++ struct inode *delegated;
++
++ h_path.mnt = au_sbr_mnt(a->sb, a->mvd_bsrc);
++ h_path.dentry = au_h_dptr(a->dentry, a->mvd_bsrc);
++ delegated = NULL;
++ err = vfsub_unlink(a->mvd_h_src_dir, &h_path, &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ if (unlikely(err))
++ AU_MVD_PR(dmsg, "unlink failed\n");
++
++ AuTraceErr(err);
++ return err;
++}
++
++/* Since mvdown succeeded, we ignore an error of this function */
++static void au_do_stfs(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct au_branch *br;
++
++ a->mvdown.flags |= AUFS_MVDOWN_STFS_FAILED;
++ br = au_sbr(a->sb, a->mvd_bsrc);
++ err = au_br_stfs(br, &a->mvdown.stbr[AUFS_MVDOWN_UPPER].stfs);
++ if (!err) {
++ br = au_sbr(a->sb, a->mvd_bdst);
++ a->mvdown.stbr[AUFS_MVDOWN_LOWER].brid = br->br_id;
++ err = au_br_stfs(br, &a->mvdown.stbr[AUFS_MVDOWN_LOWER].stfs);
++ }
++ if (!err)
++ a->mvdown.flags &= ~AUFS_MVDOWN_STFS_FAILED;
++ else
++ AU_MVD_PR(dmsg, "statfs failed (%d), ignored\n", err);
++}
++
++/*
++ * copy-down the file and unlink the bsrc file.
++ * - unlink the bdst whout if exist
++ * - copy-down the file (with whtmp name and rename)
++ * - unlink the bsrc file
++ */
++static int au_do_mvdown(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++
++ err = au_do_mkdir(dmsg, a);
++ if (!err)
++ err = au_do_lock(dmsg, a);
++ if (unlikely(err))
++ goto out;
++
++ /*
++ * do not revert the activities we made on bdst since they should be
++ * harmless in aufs.
++ */
++
++ err = au_do_cpdown(dmsg, a);
++ if (!err)
++ err = au_do_unlink_wh(dmsg, a);
++ if (!err && !(a->mvdown.flags & AUFS_MVDOWN_KUPPER))
++ err = au_do_unlink(dmsg, a);
++ if (unlikely(err))
++ goto out_unlock;
++
++ AuDbg("%pd2, 0x%x, %d --> %d\n",
++ a->dentry, a->mvdown.flags, a->mvd_bsrc, a->mvd_bdst);
++ if (find_lower_writable(a) < 0)
++ a->mvdown.flags |= AUFS_MVDOWN_BOTTOM;
++
++ if (a->mvdown.flags & AUFS_MVDOWN_STFS)
++ au_do_stfs(dmsg, a);
++
++ /* maintain internal array */
++ if (!(a->mvdown.flags & AUFS_MVDOWN_KUPPER)) {
++ au_set_h_dptr(a->dentry, a->mvd_bsrc, NULL);
++ au_set_dbtop(a->dentry, a->mvd_bdst);
++ au_set_h_iptr(a->inode, a->mvd_bsrc, NULL, /*flags*/0);
++ au_set_ibtop(a->inode, a->mvd_bdst);
++ } else {
++ /* hide the lower */
++ au_set_h_dptr(a->dentry, a->mvd_bdst, NULL);
++ au_set_dbbot(a->dentry, a->mvd_bsrc);
++ au_set_h_iptr(a->inode, a->mvd_bdst, NULL, /*flags*/0);
++ au_set_ibbot(a->inode, a->mvd_bsrc);
++ }
++ if (au_dbbot(a->dentry) < a->mvd_bdst)
++ au_set_dbbot(a->dentry, a->mvd_bdst);
++ if (au_ibbot(a->inode) < a->mvd_bdst)
++ au_set_ibbot(a->inode, a->mvd_bdst);
++
++out_unlock:
++ au_do_unlock(dmsg, a);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* make sure the file is idle */
++static int au_mvd_args_busy(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err, plinked;
++
++ err = 0;
++ plinked = !!au_opt_test(au_mntflags(a->sb), PLINK);
++ if (au_dbtop(a->dentry) == a->mvd_bsrc
++ && au_dcount(a->dentry) == 1
++ && atomic_read(&a->inode->i_count) == 1
++ /* && a->mvd_h_src_inode->i_nlink == 1 */
++ && (!plinked || !au_plink_test(a->inode))
++ && a->inode->i_nlink == 1)
++ goto out;
++
++ err = -EBUSY;
++ AU_MVD_PR(dmsg,
++ "b%d, d{b%d, c%d?}, i{c%d?, l%u}, hi{l%u}, p{%d, %d}\n",
++ a->mvd_bsrc, au_dbtop(a->dentry), au_dcount(a->dentry),
++ atomic_read(&a->inode->i_count), a->inode->i_nlink,
++ a->mvd_h_src_inode->i_nlink,
++ plinked, plinked ? au_plink_test(a->inode) : 0);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++/* make sure the parent dir is fine */
++static int au_mvd_args_parent(const unsigned char dmsg,
++ struct au_mvd_args *a)
++{
++ int err;
++ aufs_bindex_t bindex;
++
++ err = 0;
++ if (unlikely(au_alive_dir(a->parent))) {
++ err = -ENOENT;
++ AU_MVD_PR(dmsg, "parent dir is dead\n");
++ goto out;
++ }
++
++ a->bopq = au_dbdiropq(a->parent);
++ bindex = au_wbr_nonopq(a->dentry, a->mvd_bdst);
++ AuDbg("b%d\n", bindex);
++ if (unlikely((bindex >= 0 && bindex < a->mvd_bdst)
++ || (a->bopq != -1 && a->bopq < a->mvd_bdst))) {
++ err = -EINVAL;
++ a->mvd_errno = EAU_MVDOWN_OPAQUE;
++ AU_MVD_PR(dmsg, "ancestor is opaque b%d, b%d\n",
++ a->bopq, a->mvd_bdst);
++ }
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_mvd_args_intermediate(const unsigned char dmsg,
++ struct au_mvd_args *a)
++{
++ int err;
++ struct au_dinfo *dinfo, *tmp;
++
++ /* lookup the next lower positive entry */
++ err = -ENOMEM;
++ tmp = au_di_alloc(a->sb, AuLsc_DI_TMP);
++ if (unlikely(!tmp))
++ goto out;
++
++ a->bfound = -1;
++ a->bwh = -1;
++ dinfo = au_di(a->dentry);
++ au_di_cp(tmp, dinfo);
++ au_di_swap(tmp, dinfo);
++
++ /* returns the number of positive dentries */
++ err = au_lkup_dentry(a->dentry, a->mvd_bsrc + 1,
++ /* AuLkup_IGNORE_PERM */ 0);
++ if (!err)
++ a->bwh = au_dbwh(a->dentry);
++ else if (err > 0)
++ a->bfound = au_dbtop(a->dentry);
++
++ au_di_swap(tmp, dinfo);
++ au_rw_write_unlock(&tmp->di_rwsem);
++ au_di_free(tmp);
++ if (unlikely(err < 0))
++ AU_MVD_PR(dmsg, "failed look-up lower\n");
++
++ /*
++ * here, we have these cases.
++ * bfound == -1
++ * no positive dentry under bsrc. there are more sub-cases.
++ * bwh < 0
++ * there no whiteout, we can safely move-down.
++ * bwh <= bsrc
++ * impossible
++ * bsrc < bwh && bwh < bdst
++ * there is a whiteout on RO branch. cannot proceed.
++ * bwh == bdst
++ * there is a whiteout on the RW target branch. it should
++ * be removed.
++ * bdst < bwh
++ * there is a whiteout somewhere unrelated branch.
++ * -1 < bfound && bfound <= bsrc
++ * impossible.
++ * bfound < bdst
++ * found, but it is on RO branch between bsrc and bdst. cannot
++ * proceed.
++ * bfound == bdst
++ * found, replace it if AUFS_MVDOWN_FORCE is set. otherwise return
++ * error.
++ * bdst < bfound
++ * found, after we create the file on bdst, it will be hidden.
++ */
++
++ AuDebugOn(a->bfound == -1
++ && a->bwh != -1
++ && a->bwh <= a->mvd_bsrc);
++ AuDebugOn(-1 < a->bfound
++ && a->bfound <= a->mvd_bsrc);
++
++ err = -EINVAL;
++ if (a->bfound == -1
++ && a->mvd_bsrc < a->bwh
++ && a->bwh != -1
++ && a->bwh < a->mvd_bdst) {
++ a->mvd_errno = EAU_MVDOWN_WHITEOUT;
++ AU_MVD_PR(dmsg, "bsrc %d, bdst %d, bfound %d, bwh %d\n",
++ a->mvd_bsrc, a->mvd_bdst, a->bfound, a->bwh);
++ goto out;
++ } else if (a->bfound != -1 && a->bfound < a->mvd_bdst) {
++ a->mvd_errno = EAU_MVDOWN_UPPER;
++ AU_MVD_PR(dmsg, "bdst %d, bfound %d\n",
++ a->mvd_bdst, a->bfound);
++ goto out;
++ }
++
++ err = 0; /* success */
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_mvd_args_exist(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++
++ err = 0;
++ if (!(a->mvdown.flags & AUFS_MVDOWN_OWLOWER)
++ && a->bfound == a->mvd_bdst)
++ err = -EEXIST;
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_mvd_args(const unsigned char dmsg, struct au_mvd_args *a)
++{
++ int err;
++ struct au_branch *br;
++
++ err = -EISDIR;
++ if (unlikely(S_ISDIR(a->inode->i_mode)))
++ goto out;
++
++ err = -EINVAL;
++ if (!(a->mvdown.flags & AUFS_MVDOWN_BRID_UPPER))
++ a->mvd_bsrc = au_ibtop(a->inode);
++ else {
++ a->mvd_bsrc = au_br_index(a->sb, a->mvd_src_brid);
++ if (unlikely(a->mvd_bsrc < 0
++ || (a->mvd_bsrc < au_dbtop(a->dentry)
++ || au_dbbot(a->dentry) < a->mvd_bsrc
++ || !au_h_dptr(a->dentry, a->mvd_bsrc))
++ || (a->mvd_bsrc < au_ibtop(a->inode)
++ || au_ibbot(a->inode) < a->mvd_bsrc
++ || !au_h_iptr(a->inode, a->mvd_bsrc)))) {
++ a->mvd_errno = EAU_MVDOWN_NOUPPER;
++ AU_MVD_PR(dmsg, "no upper\n");
++ goto out;
++ }
++ }
++ if (unlikely(a->mvd_bsrc == au_sbbot(a->sb))) {
++ a->mvd_errno = EAU_MVDOWN_BOTTOM;
++ AU_MVD_PR(dmsg, "on the bottom\n");
++ goto out;
++ }
++ a->mvd_h_src_inode = au_h_iptr(a->inode, a->mvd_bsrc);
++ br = au_sbr(a->sb, a->mvd_bsrc);
++ err = au_br_rdonly(br);
++ if (!(a->mvdown.flags & AUFS_MVDOWN_ROUPPER)) {
++ if (unlikely(err))
++ goto out;
++ } else if (!(vfsub_native_ro(a->mvd_h_src_inode)
++ || IS_APPEND(a->mvd_h_src_inode))) {
++ if (err)
++ a->mvdown.flags |= AUFS_MVDOWN_ROUPPER_R;
++ /* go on */
++ } else
++ goto out;
++
++ err = -EINVAL;
++ if (!(a->mvdown.flags & AUFS_MVDOWN_BRID_LOWER)) {
++ a->mvd_bdst = find_lower_writable(a);
++ if (unlikely(a->mvd_bdst < 0)) {
++ a->mvd_errno = EAU_MVDOWN_BOTTOM;
++ AU_MVD_PR(dmsg, "no writable lower branch\n");
++ goto out;
++ }
++ } else {
++ a->mvd_bdst = au_br_index(a->sb, a->mvd_dst_brid);
++ if (unlikely(a->mvd_bdst < 0
++ || au_sbbot(a->sb) < a->mvd_bdst)) {
++ a->mvd_errno = EAU_MVDOWN_NOLOWERBR;
++ AU_MVD_PR(dmsg, "no lower brid\n");
++ goto out;
++ }
++ }
++
++ err = au_mvd_args_busy(dmsg, a);
++ if (!err)
++ err = au_mvd_args_parent(dmsg, a);
++ if (!err)
++ err = au_mvd_args_intermediate(dmsg, a);
++ if (!err)
++ err = au_mvd_args_exist(dmsg, a);
++ if (!err)
++ AuDbg("b%d, b%d\n", a->mvd_bsrc, a->mvd_bdst);
++
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++int au_mvdown(struct dentry *dentry, struct aufs_mvdown __user *uarg)
++{
++ int err, e;
++ unsigned char dmsg;
++ struct au_mvd_args *args;
++ struct inode *inode;
++
++ inode = d_inode(dentry);
++ err = -EPERM;
++ if (unlikely(!capable(CAP_SYS_ADMIN)))
++ goto out;
++
++ err = -ENOMEM;
++ args = kmalloc(sizeof(*args), GFP_NOFS);
++ if (unlikely(!args))
++ goto out;
++
++ err = copy_from_user(&args->mvdown, uarg, sizeof(args->mvdown));
++ if (!err)
++ /* VERIFY_WRITE */
++ err = !access_ok(uarg, sizeof(*uarg));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ goto out_free;
++ }
++ AuDbg("flags 0x%x\n", args->mvdown.flags);
++ args->mvdown.flags &= ~(AUFS_MVDOWN_ROLOWER_R | AUFS_MVDOWN_ROUPPER_R);
++ args->mvdown.au_errno = 0;
++ args->dentry = dentry;
++ args->inode = inode;
++ args->sb = dentry->d_sb;
++
++ err = -ENOENT;
++ dmsg = !!(args->mvdown.flags & AUFS_MVDOWN_DMSG);
++ args->parent = dget_parent(dentry);
++ args->dir = d_inode(args->parent);
++ inode_lock_nested(args->dir, I_MUTEX_PARENT);
++ dput(args->parent);
++ if (unlikely(args->parent != dentry->d_parent)) {
++ AU_MVD_PR(dmsg, "parent dir is moved\n");
++ goto out_dir;
++ }
++
++ inode_lock_nested(inode, I_MUTEX_CHILD);
++ err = aufs_read_lock(dentry, AuLock_DW | AuLock_FLUSH | AuLock_NOPLMW);
++ if (unlikely(err))
++ goto out_inode;
++
++ di_write_lock_parent(args->parent);
++ err = au_mvd_args(dmsg, args);
++ if (unlikely(err))
++ goto out_parent;
++
++ err = au_do_mvdown(dmsg, args);
++ if (unlikely(err))
++ goto out_parent;
++
++ au_cpup_attr_timesizes(args->dir);
++ au_cpup_attr_timesizes(inode);
++ if (!(args->mvdown.flags & AUFS_MVDOWN_KUPPER))
++ au_cpup_igen(inode, au_h_iptr(inode, args->mvd_bdst));
++ /* au_digen_dec(dentry); */
++
++out_parent:
++ di_write_unlock(args->parent);
++ aufs_read_unlock(dentry, AuLock_DW);
++out_inode:
++ inode_unlock(inode);
++out_dir:
++ inode_unlock(args->dir);
++out_free:
++ e = copy_to_user(uarg, &args->mvdown, sizeof(args->mvdown));
++ if (unlikely(e))
++ err = -EFAULT;
++ au_kfree_rcu(args);
++out:
++ AuTraceErr(err);
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/opts.c linux/fs/aufs/opts.c
+--- /usr/share/empty/fs/aufs/opts.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/opts.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,1032 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * mount options/flags
++ */
++
++#include <linux/types.h> /* a distribution requires */
++#include <linux/parser.h>
++#include "aufs.h"
++
++/* ---------------------------------------------------------------------- */
++
++static const char *au_parser_pattern(int val, match_table_t tbl)
++{
++ struct match_token *p;
++
++ p = tbl;
++ while (p->pattern) {
++ if (p->token == val)
++ return p->pattern;
++ p++;
++ }
++ BUG();
++ return "??";
++}
++
++static const char *au_optstr(int *val, match_table_t tbl)
++{
++ struct match_token *p;
++ int v;
++
++ v = *val;
++ if (!v)
++ goto out;
++ p = tbl;
++ while (p->pattern) {
++ if (p->token
++ && (v & p->token) == p->token) {
++ *val &= ~p->token;
++ return p->pattern;
++ }
++ p++;
++ }
++
++out:
++ return NULL;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static match_table_t brperm = {
++ {AuBrPerm_RO, AUFS_BRPERM_RO},
++ {AuBrPerm_RR, AUFS_BRPERM_RR},
++ {AuBrPerm_RW, AUFS_BRPERM_RW},
++ {0, NULL}
++};
++
++static match_table_t brattr = {
++ /* general */
++ {AuBrAttr_COO_REG, AUFS_BRATTR_COO_REG},
++ {AuBrAttr_COO_ALL, AUFS_BRATTR_COO_ALL},
++ /* 'unpin' attrib is meaningless since linux-3.18-rc1 */
++ {AuBrAttr_UNPIN, AUFS_BRATTR_UNPIN},
++#ifdef CONFIG_AUFS_FHSM
++ {AuBrAttr_FHSM, AUFS_BRATTR_FHSM},
++#endif
++#ifdef CONFIG_AUFS_XATTR
++ {AuBrAttr_ICEX, AUFS_BRATTR_ICEX},
++ {AuBrAttr_ICEX_SEC, AUFS_BRATTR_ICEX_SEC},
++ {AuBrAttr_ICEX_SYS, AUFS_BRATTR_ICEX_SYS},
++ {AuBrAttr_ICEX_TR, AUFS_BRATTR_ICEX_TR},
++ {AuBrAttr_ICEX_USR, AUFS_BRATTR_ICEX_USR},
++ {AuBrAttr_ICEX_OTH, AUFS_BRATTR_ICEX_OTH},
++#endif
++
++ /* ro/rr branch */
++ {AuBrRAttr_WH, AUFS_BRRATTR_WH},
++
++ /* rw branch */
++ {AuBrWAttr_MOO, AUFS_BRWATTR_MOO},
++ {AuBrWAttr_NoLinkWH, AUFS_BRWATTR_NLWH},
++
++ {0, NULL}
++};
++
++static int br_attr_val(char *str, match_table_t table, substring_t args[])
++{
++ int attr, v;
++ char *p;
++
++ attr = 0;
++ do {
++ p = strchr(str, '+');
++ if (p)
++ *p = 0;
++ v = match_token(str, table, args);
++ if (v) {
++ if (v & AuBrAttr_CMOO_Mask)
++ attr &= ~AuBrAttr_CMOO_Mask;
++ attr |= v;
++ } else {
++ if (p)
++ *p = '+';
++ pr_warn("ignored branch attribute %s\n", str);
++ break;
++ }
++ if (p)
++ str = p + 1;
++ } while (p);
++
++ return attr;
++}
++
++static int au_do_optstr_br_attr(au_br_perm_str_t *str, int perm)
++{
++ int sz;
++ const char *p;
++ char *q;
++
++ q = str->a;
++ *q = 0;
++ p = au_optstr(&perm, brattr);
++ if (p) {
++ sz = strlen(p);
++ memcpy(q, p, sz + 1);
++ q += sz;
++ } else
++ goto out;
++
++ do {
++ p = au_optstr(&perm, brattr);
++ if (p) {
++ *q++ = '+';
++ sz = strlen(p);
++ memcpy(q, p, sz + 1);
++ q += sz;
++ }
++ } while (p);
++
++out:
++ return q - str->a;
++}
++
++int au_br_perm_val(char *perm)
++{
++ int val, bad, sz;
++ char *p;
++ substring_t args[MAX_OPT_ARGS];
++ au_br_perm_str_t attr;
++
++ p = strchr(perm, '+');
++ if (p)
++ *p = 0;
++ val = match_token(perm, brperm, args);
++ if (!val) {
++ if (p)
++ *p = '+';
++ pr_warn("ignored branch permission %s\n", perm);
++ val = AuBrPerm_RO;
++ goto out;
++ }
++ if (!p)
++ goto out;
++
++ val |= br_attr_val(p + 1, brattr, args);
++
++ bad = 0;
++ switch (val & AuBrPerm_Mask) {
++ case AuBrPerm_RO:
++ case AuBrPerm_RR:
++ bad = val & AuBrWAttr_Mask;
++ val &= ~AuBrWAttr_Mask;
++ break;
++ case AuBrPerm_RW:
++ bad = val & AuBrRAttr_Mask;
++ val &= ~AuBrRAttr_Mask;
++ break;
++ }
++
++ /*
++ * 'unpin' attrib becomes meaningless since linux-3.18-rc1, but aufs
++ * does not treat it as an error, just warning.
++ * this is a tiny guard for the user operation.
++ */
++ if (val & AuBrAttr_UNPIN) {
++ bad |= AuBrAttr_UNPIN;
++ val &= ~AuBrAttr_UNPIN;
++ }
++
++ if (unlikely(bad)) {
++ sz = au_do_optstr_br_attr(&attr, bad);
++ AuDebugOn(!sz);
++ pr_warn("ignored branch attribute %s\n", attr.a);
++ }
++
++out:
++ return val;
++}
++
++void au_optstr_br_perm(au_br_perm_str_t *str, int perm)
++{
++ au_br_perm_str_t attr;
++ const char *p;
++ char *q;
++ int sz;
++
++ q = str->a;
++ p = au_optstr(&perm, brperm);
++ AuDebugOn(!p || !*p);
++ sz = strlen(p);
++ memcpy(q, p, sz + 1);
++ q += sz;
++
++ sz = au_do_optstr_br_attr(&attr, perm);
++ if (sz) {
++ *q++ = '+';
++ memcpy(q, attr.a, sz + 1);
++ }
++
++ AuDebugOn(strlen(str->a) >= sizeof(str->a));
++}
++
++/* ---------------------------------------------------------------------- */
++
++static match_table_t udbalevel = {
++ {AuOpt_UDBA_REVAL, "reval"},
++ {AuOpt_UDBA_NONE, "none"},
++#ifdef CONFIG_AUFS_HNOTIFY
++ {AuOpt_UDBA_HNOTIFY, "notify"}, /* abstraction */
++#ifdef CONFIG_AUFS_HFSNOTIFY
++ {AuOpt_UDBA_HNOTIFY, "fsnotify"},
++#endif
++#endif
++ {-1, NULL}
++};
++
++int au_udba_val(char *str)
++{
++ substring_t args[MAX_OPT_ARGS];
++
++ return match_token(str, udbalevel, args);
++}
++
++const char *au_optstr_udba(int udba)
++{
++ return au_parser_pattern(udba, udbalevel);
++}
++
++/* ---------------------------------------------------------------------- */
++
++static match_table_t au_wbr_create_policy = {
++ {AuWbrCreate_TDP, "tdp"},
++ {AuWbrCreate_TDP, "top-down-parent"},
++ {AuWbrCreate_RR, "rr"},
++ {AuWbrCreate_RR, "round-robin"},
++ {AuWbrCreate_MFS, "mfs"},
++ {AuWbrCreate_MFS, "most-free-space"},
++ {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"},
++ {AuWbrCreate_PMFSV, "pmfs:%d"},
++ {AuWbrCreate_PMFSRR, "pmfsrr:%d"},
++ {AuWbrCreate_PMFSRRV, "pmfsrr:%d:%d"},
++
++ {-1, NULL}
++};
++
++static int au_wbr_mfs_wmark(substring_t *arg, char *str,
++ struct au_opt_wbr_create *create)
++{
++ int err;
++ unsigned long long ull;
++
++ err = 0;
++ if (!match_u64(arg, &ull))
++ create->mfsrr_watermark = ull;
++ else {
++ pr_err("bad integer in %s\n", str);
++ err = -EINVAL;
++ }
++
++ return err;
++}
++
++static int au_wbr_mfs_sec(substring_t *arg, char *str,
++ struct au_opt_wbr_create *create)
++{
++ int n, err;
++
++ err = 0;
++ if (!match_int(arg, &n) && 0 <= n && n <= AUFS_MFS_MAX_SEC)
++ create->mfs_second = n;
++ else {
++ pr_err("bad integer in %s\n", str);
++ err = -EINVAL;
++ }
++
++ return err;
++}
++
++int au_wbr_create_val(char *str, struct au_opt_wbr_create *create)
++{
++ int err, e;
++ substring_t args[MAX_OPT_ARGS];
++
++ err = match_token(str, au_wbr_create_policy, args);
++ 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)
+ e = au_wbr_mfs_sec(&args[1], str, create);
+ if (unlikely(e))
+ err = e;
+ break;
-+ case AuWbrCreate_MFSRR:
-+ e = au_wbr_mfs_wmark(&args[0], str, create);
-+ if (unlikely(e)) {
-+ err = e;
++ case AuWbrCreate_MFSRR:
++ case AuWbrCreate_TDMFS:
++ case AuWbrCreate_PMFSRR:
++ e = au_wbr_mfs_wmark(&args[0], str, create);
++ if (unlikely(e)) {
++ err = e;
++ break;
++ }
++ fallthrough;
++ case AuWbrCreate_MFS:
++ case AuWbrCreate_PMFS:
++ create->mfs_second = AUFS_MFS_DEF_SEC;
++ break;
++ case AuWbrCreate_MFSV:
++ case AuWbrCreate_PMFSV:
++ e = au_wbr_mfs_sec(&args[0], str, create);
++ if (unlikely(e))
++ err = e;
++ break;
++ }
++
++ return err;
++}
++
++const char *au_optstr_wbr_create(int wbr_create)
++{
++ return au_parser_pattern(wbr_create, au_wbr_create_policy);
++}
++
++static match_table_t au_wbr_copyup_policy = {
++ {AuWbrCopyup_TDP, "tdp"},
++ {AuWbrCopyup_TDP, "top-down-parent"},
++ {AuWbrCopyup_BUP, "bup"},
++ {AuWbrCopyup_BUP, "bottom-up-parent"},
++ {AuWbrCopyup_BU, "bu"},
++ {AuWbrCopyup_BU, "bottom-up"},
++ {-1, NULL}
++};
++
++int au_wbr_copyup_val(char *str)
++{
++ substring_t args[MAX_OPT_ARGS];
++
++ return match_token(str, au_wbr_copyup_policy, args);
++}
++
++const char *au_optstr_wbr_copyup(int wbr_copyup)
++{
++ return au_parser_pattern(wbr_copyup, au_wbr_copyup_policy);
++}
++
++/* ---------------------------------------------------------------------- */
++
++int au_opt_add(struct au_opt *opt, char *opt_str, unsigned long sb_flags,
++ aufs_bindex_t bindex)
++{
++ int err;
++ struct au_opt_add *add = &opt->add;
++ char *p;
++
++ add->bindex = bindex;
++ add->perm = AuBrPerm_RO;
++ add->pathname = opt_str;
++ p = strchr(opt_str, '=');
++ if (p) {
++ *p++ = 0;
++ if (*p)
++ add->perm = au_br_perm_val(p);
++ }
++
++ err = vfsub_kern_path(add->pathname, AuOpt_LkupDirFlags, &add->path);
++ if (!err) {
++ if (!p) {
++ add->perm = AuBrPerm_RO;
++ if (au_test_fs_rr(add->path.dentry->d_sb))
++ add->perm = AuBrPerm_RR;
++ else if (!bindex && !(sb_flags & SB_RDONLY))
++ add->perm = AuBrPerm_RW;
++ }
++ opt->type = Opt_add;
++ goto out;
++ }
++ pr_err("lookup failed %s (%d)\n", add->pathname, err);
++ err = -EINVAL;
++
++out:
++ return err;
++}
++
++static int au_opt_wbr_create(struct super_block *sb,
++ struct au_opt_wbr_create *create)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++
++ SiMustWriteLock(sb);
++
++ err = 1; /* handled */
++ sbinfo = au_sbi(sb);
++ if (sbinfo->si_wbr_create_ops->fin) {
++ err = sbinfo->si_wbr_create_ops->fin(sb);
++ if (!err)
++ err = 1;
++ }
++
++ sbinfo->si_wbr_create = create->wbr_create;
++ sbinfo->si_wbr_create_ops = au_wbr_create_ops + create->wbr_create;
++ 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;
++ fallthrough;
++ case AuWbrCreate_MFS:
++ case AuWbrCreate_MFSV:
++ case AuWbrCreate_PMFS:
++ case AuWbrCreate_PMFSV:
++ sbinfo->si_wbr_mfs.mfs_expire
++ = msecs_to_jiffies(create->mfs_second * MSEC_PER_SEC);
++ break;
++ }
++
++ if (sbinfo->si_wbr_create_ops->init)
++ sbinfo->si_wbr_create_ops->init(sb); /* ignore */
++
++ return err;
++}
++
++/*
++ * returns,
++ * plus: processed without an error
++ * zero: unprocessed
++ */
++static int au_opt_simple(struct super_block *sb, struct au_opt *opt,
++ struct au_opts *opts)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++
++ SiMustWriteLock(sb);
++
++ err = 1; /* handled */
++ sbinfo = au_sbi(sb);
++ switch (opt->type) {
++ case Opt_udba:
++ sbinfo->si_mntflags &= ~AuOptMask_UDBA;
++ sbinfo->si_mntflags |= opt->udba;
++ opts->given_udba |= opt->udba;
++ break;
++
++ case Opt_plink:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, PLINK);
++ else {
++ if (au_opt_test(sbinfo->si_mntflags, PLINK))
++ au_plink_put(sb, /*verbose*/1);
++ au_opt_clr(sbinfo->si_mntflags, PLINK);
++ }
++ break;
++ case Opt_list_plink:
++ if (au_opt_test(sbinfo->si_mntflags, PLINK))
++ au_plink_list(sb);
++ break;
++
++ case Opt_dio:
++ if (opt->tf) {
++ au_opt_set(sbinfo->si_mntflags, DIO);
++ au_fset_opts(opts->flags, REFRESH_DYAOP);
++ } else {
++ au_opt_clr(sbinfo->si_mntflags, DIO);
++ au_fset_opts(opts->flags, REFRESH_DYAOP);
++ }
++ break;
++
++ case Opt_fhsm_sec:
++ au_fhsm_set(sbinfo, opt->fhsm_second);
++ break;
++
++ case Opt_diropq_a:
++ au_opt_set(sbinfo->si_mntflags, ALWAYS_DIROPQ);
++ break;
++ case Opt_diropq_w:
++ au_opt_clr(sbinfo->si_mntflags, ALWAYS_DIROPQ);
++ break;
++
++ case Opt_warn_perm:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, WARN_PERM);
++ else
++ au_opt_clr(sbinfo->si_mntflags, WARN_PERM);
++ break;
++
++ case Opt_verbose:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, VERBOSE);
++ else
++ au_opt_clr(sbinfo->si_mntflags, VERBOSE);
++ break;
++
++ case Opt_sum:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, SUM);
++ else {
++ au_opt_clr(sbinfo->si_mntflags, SUM);
++ au_opt_clr(sbinfo->si_mntflags, SUM_W);
++ }
++ break;
++ case Opt_wsum:
++ au_opt_clr(sbinfo->si_mntflags, SUM);
++ au_opt_set(sbinfo->si_mntflags, SUM_W);
++ break;
++
++ case Opt_wbr_create:
++ err = au_opt_wbr_create(sb, &opt->wbr_create);
++ break;
++ case Opt_wbr_copyup:
++ sbinfo->si_wbr_copyup = opt->wbr_copyup;
++ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + opt->wbr_copyup;
++ break;
++
++ case Opt_dirwh:
++ sbinfo->si_dirwh = opt->dirwh;
++ break;
++
++ case Opt_rdcache:
++ sbinfo->si_rdcache
++ = msecs_to_jiffies(opt->rdcache * MSEC_PER_SEC);
++ break;
++ case Opt_rdblk:
++ sbinfo->si_rdblk = opt->rdblk;
++ break;
++ case Opt_rdhash:
++ sbinfo->si_rdhash = opt->rdhash;
++ break;
++
++ case Opt_shwh:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, SHWH);
++ else
++ au_opt_clr(sbinfo->si_mntflags, SHWH);
++ break;
++
++ case Opt_dirperm1:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, DIRPERM1);
++ else
++ au_opt_clr(sbinfo->si_mntflags, DIRPERM1);
++ break;
++
++ case Opt_trunc_xino:
++ if (opt->tf)
++ au_opt_set(sbinfo->si_mntflags, TRUNC_XINO);
++ else
++ au_opt_clr(sbinfo->si_mntflags, TRUNC_XINO);
++ break;
++
++ case Opt_trunc_xino_path:
++ case Opt_itrunc_xino:
++ err = au_xino_trunc(sb, opt->xino_itrunc.bindex,
++ /*idx_begin*/0);
++ if (!err)
++ err = 1;
++ break;
++
++ case Opt_trunc_xib:
++ if (opt->tf)
++ au_fset_opts(opts->flags, TRUNC_XIB);
++ else
++ au_fclr_opts(opts->flags, TRUNC_XIB);
++ break;
++
++ case Opt_dirren:
++ err = 1;
++ if (opt->tf) {
++ 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);
++ } else {
++ 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:
++ if (opt->tf)
++ sb->s_flags |= SB_POSIXACL;
++ else
++ sb->s_flags &= ~SB_POSIXACL;
++ break;
++
++ default:
++ err = 0;
++ break;
++ }
++
++ return err;
++}
++
++/*
++ * returns tri-state.
++ * plus: processed without an error
++ * zero: unprocessed
++ * minus: error
++ */
++static int au_opt_br(struct super_block *sb, struct au_opt *opt,
++ struct au_opts *opts)
++{
++ int err, do_refresh;
++
++ err = 0;
++ switch (opt->type) {
++ case Opt_append:
++ opt->add.bindex = au_sbbot(sb) + 1;
++ if (opt->add.bindex < 0)
++ opt->add.bindex = 0;
++ goto add;
++ /* Always goto add, not fallthrough */
++ case Opt_prepend:
++ opt->add.bindex = 0;
++ fallthrough;
++ add: /* indented label */
++ case Opt_add:
++ err = au_br_add(sb, &opt->add,
++ au_ftest_opts(opts->flags, REMOUNT));
++ if (!err) {
++ err = 1;
++ au_fset_opts(opts->flags, REFRESH);
++ }
++ break;
++
++ case Opt_del:
++ case Opt_idel:
++ err = au_br_del(sb, &opt->del,
++ au_ftest_opts(opts->flags, REMOUNT));
++ if (!err) {
++ err = 1;
++ au_fset_opts(opts->flags, TRUNC_XIB);
++ au_fset_opts(opts->flags, REFRESH);
++ }
++ break;
++
++ case Opt_mod:
++ case Opt_imod:
++ err = au_br_mod(sb, &opt->mod,
++ au_ftest_opts(opts->flags, REMOUNT),
++ &do_refresh);
++ if (!err) {
++ err = 1;
++ if (do_refresh)
++ au_fset_opts(opts->flags, REFRESH);
++ }
++ break;
++ }
++ return err;
++}
++
++static int au_opt_xino(struct super_block *sb, struct au_opt *opt,
++ struct au_opt_xino **opt_xino,
++ struct au_opts *opts)
++{
++ int err;
++
++ err = 0;
++ switch (opt->type) {
++ case Opt_xino:
++ err = au_xino_set(sb, &opt->xino,
++ !!au_ftest_opts(opts->flags, REMOUNT));
++ if (!err)
++ *opt_xino = &opt->xino;
++ break;
++ case Opt_noxino:
++ au_xino_clr(sb);
++ *opt_xino = (void *)-1;
++ break;
++ }
++
++ return err;
++}
++
++int au_opts_verify(struct super_block *sb, unsigned long sb_flags,
++ unsigned int pending)
++{
++ int err, fhsm;
++ aufs_bindex_t bindex, bbot;
++ unsigned char do_plink, skip, do_free, can_no_dreval;
++ struct au_branch *br;
++ struct au_wbr *wbr;
++ struct dentry *root, *dentry;
++ struct inode *dir, *h_dir;
++ struct au_sbinfo *sbinfo;
++ struct au_hinode *hdir;
++
++ SiMustAnyLock(sb);
++
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!(sbinfo->si_mntflags & AuOptMask_UDBA));
++
++ if (!(sb_flags & SB_RDONLY)) {
++ if (unlikely(!au_br_writable(au_sbr_perm(sb, 0))))
++ pr_warn("first branch should be rw\n");
++ if (unlikely(au_opt_test(sbinfo->si_mntflags, SHWH)))
++ pr_warn_once("shwh should be used with ro\n");
++ }
++
++ if (au_opt_test((sbinfo->si_mntflags | pending), UDBA_HNOTIFY)
++ && !au_opt_test(sbinfo->si_mntflags, XINO))
++ pr_warn_once("udba=*notify requires xino\n");
++
++ if (au_opt_test(sbinfo->si_mntflags, DIRPERM1))
++ pr_warn_once("dirperm1 breaks the protection"
++ " by the permission bits on the lower branch\n");
++
++ err = 0;
++ fhsm = 0;
++ root = sb->s_root;
++ dir = d_inode(root);
++ do_plink = !!au_opt_test(sbinfo->si_mntflags, PLINK);
++ can_no_dreval = !!au_opt_test((sbinfo->si_mntflags | pending),
++ UDBA_NONE);
++ bbot = au_sbbot(sb);
++ for (bindex = 0; !err && bindex <= bbot; bindex++) {
++ skip = 0;
++ h_dir = au_h_iptr(dir, bindex);
++ br = au_sbr(sb, bindex);
++
++ if ((br->br_perm & AuBrAttr_ICEX)
++ && !h_dir->i_op->listxattr)
++ br->br_perm &= ~AuBrAttr_ICEX;
++#if 0 /* untested */
++ if ((br->br_perm & AuBrAttr_ICEX_SEC)
++ && (au_br_sb(br)->s_flags & SB_NOSEC))
++ br->br_perm &= ~AuBrAttr_ICEX_SEC;
++#endif
++
++ do_free = 0;
++ wbr = br->br_wbr;
++ if (wbr)
++ wbr_wh_read_lock(wbr);
++
++ if (!au_br_writable(br->br_perm)) {
++ do_free = !!wbr;
++ skip = (!wbr
++ || (!wbr->wbr_whbase
++ && !wbr->wbr_plink
++ && !wbr->wbr_orph));
++ } else if (!au_br_wh_linkable(br->br_perm)) {
++ /* skip = (!br->br_whbase && !br->br_orph); */
++ skip = (!wbr || !wbr->wbr_whbase);
++ if (skip && wbr) {
++ if (do_plink)
++ skip = !!wbr->wbr_plink;
++ else
++ skip = !wbr->wbr_plink;
++ }
++ } else {
++ /* skip = (br->br_whbase && br->br_ohph); */
++ skip = (wbr && wbr->wbr_whbase);
++ if (skip) {
++ if (do_plink)
++ skip = !!wbr->wbr_plink;
++ else
++ skip = !wbr->wbr_plink;
++ }
++ }
++ if (wbr)
++ wbr_wh_read_unlock(wbr);
++
++ if (can_no_dreval) {
++ dentry = br->br_path.dentry;
++ spin_lock(&dentry->d_lock);
++ if (dentry->d_flags &
++ (DCACHE_OP_REVALIDATE | DCACHE_OP_WEAK_REVALIDATE))
++ can_no_dreval = 0;
++ spin_unlock(&dentry->d_lock);
++ }
++
++ if (au_br_fhsm(br->br_perm)) {
++ fhsm++;
++ AuDebugOn(!br->br_fhsm);
++ }
++
++ if (skip)
++ continue;
++
++ hdir = au_hi(dir, bindex);
++ au_hn_inode_lock_nested(hdir, AuLsc_I_PARENT);
++ if (wbr)
++ wbr_wh_write_lock(wbr);
++ err = au_wh_init(br, sb);
++ if (wbr)
++ wbr_wh_write_unlock(wbr);
++ au_hn_inode_unlock(hdir);
++
++ if (!err && do_free) {
++ au_kfree_rcu(wbr);
++ br->br_wbr = NULL;
++ }
++ }
++
++ if (can_no_dreval)
++ au_fset_si(sbinfo, NO_DREVAL);
++ else
++ au_fclr_si(sbinfo, NO_DREVAL);
++
++ if (fhsm >= 2) {
++ au_fset_si(sbinfo, FHSM);
++ for (bindex = bbot; bindex >= 0; bindex--) {
++ br = au_sbr(sb, bindex);
++ if (au_br_fhsm(br->br_perm)) {
++ au_fhsm_set_bottom(sb, bindex);
++ break;
++ }
++ }
++ } else {
++ au_fclr_si(sbinfo, FHSM);
++ au_fhsm_set_bottom(sb, -1);
++ }
++
++ return err;
++}
++
++int au_opts_mount(struct super_block *sb, struct au_opts *opts)
++{
++ int err;
++ unsigned int tmp;
++ aufs_bindex_t bindex, bbot;
++ struct au_opt *opt;
++ struct au_opt_xino *opt_xino, xino;
++ struct au_sbinfo *sbinfo;
++ struct au_branch *br;
++ struct inode *dir;
++
++ SiMustWriteLock(sb);
++
++ err = 0;
++ opt_xino = NULL;
++ opt = opts->opt;
++ while (err >= 0 && opt->type != Opt_tail)
++ err = au_opt_simple(sb, opt++, opts);
++ if (err > 0)
++ err = 0;
++ else if (unlikely(err < 0))
++ goto out;
++
++ /* disable xino and udba temporary */
++ sbinfo = au_sbi(sb);
++ tmp = sbinfo->si_mntflags;
++ au_opt_clr(sbinfo->si_mntflags, XINO);
++ au_opt_set_udba(sbinfo->si_mntflags, UDBA_REVAL);
++
++ opt = opts->opt;
++ while (err >= 0 && opt->type != Opt_tail)
++ err = au_opt_br(sb, opt++, opts);
++ if (err > 0)
++ err = 0;
++ else if (unlikely(err < 0))
++ goto out;
++
++ bbot = au_sbbot(sb);
++ if (unlikely(bbot < 0)) {
++ err = -EINVAL;
++ pr_err("no branches\n");
++ goto out;
++ }
++
++ if (au_opt_test(tmp, XINO))
++ au_opt_set(sbinfo->si_mntflags, XINO);
++ opt = opts->opt;
++ while (!err && opt->type != Opt_tail)
++ err = au_opt_xino(sb, opt++, &opt_xino, opts);
++ if (unlikely(err))
++ goto out;
++
++ err = au_opts_verify(sb, sb->s_flags, tmp);
++ if (unlikely(err))
++ goto out;
++
++ /* restore xino */
++ if (au_opt_test(tmp, XINO) && !opt_xino) {
++ xino.file = au_xino_def(sb);
++ err = PTR_ERR(xino.file);
++ if (IS_ERR(xino.file))
++ goto out;
++
++ err = au_xino_set(sb, &xino, /*remount*/0);
++ fput(xino.file);
++ if (unlikely(err))
++ goto out;
++ }
++
++ /* restore udba */
++ tmp &= AuOptMask_UDBA;
++ sbinfo->si_mntflags &= ~AuOptMask_UDBA;
++ sbinfo->si_mntflags |= tmp;
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ err = au_hnotify_reset_br(tmp, br, br->br_perm);
++ if (unlikely(err))
++ AuIOErr("hnotify failed on br %d, %d, ignored\n",
++ bindex, err);
++ /* go on even if err */
++ }
++ if (au_opt_test(tmp, UDBA_HNOTIFY)) {
++ dir = d_inode(sb->s_root);
++ au_hn_reset(dir, au_hi_flags(dir, /*isdir*/1) & ~AuHi_XINO);
++ }
++
++out:
++ return err;
++}
++
++int au_opts_remount(struct super_block *sb, struct au_opts *opts)
++{
++ int err, rerr;
++ unsigned char no_dreval;
++ struct inode *dir;
++ struct au_opt_xino *opt_xino;
++ struct au_opt *opt;
++ struct au_sbinfo *sbinfo;
++
++ SiMustWriteLock(sb);
++
++ 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;
++ opt = opts->opt;
++ while (err >= 0 && opt->type != Opt_tail) {
++ err = au_opt_simple(sb, opt, opts);
++ if (!err)
++ err = au_opt_br(sb, opt, opts);
++ if (!err)
++ err = au_opt_xino(sb, opt, &opt_xino, opts);
++ opt++;
++ }
++ if (err > 0)
++ err = 0;
++ AuTraceErr(err);
++ /* go on even err */
++
++ no_dreval = !!au_ftest_si(sbinfo, NO_DREVAL);
++ rerr = au_opts_verify(sb, opts->sb_flags, /*pending*/0);
++ if (unlikely(rerr && !err))
++ err = rerr;
++
++ if (no_dreval != !!au_ftest_si(sbinfo, NO_DREVAL))
++ au_fset_opts(opts->flags, REFRESH_IDOP);
++
++ if (au_ftest_opts(opts->flags, TRUNC_XIB)) {
++ rerr = au_xib_trunc(sb);
++ if (unlikely(rerr && !err))
++ err = rerr;
++ }
++
++ /* will be handled by the caller */
++ if (!au_ftest_opts(opts->flags, REFRESH)
++ && (opts->given_udba
++ || au_opt_test(sbinfo->si_mntflags, XINO)
++ || au_ftest_opts(opts->flags, REFRESH_IDOP)
++ ))
++ au_fset_opts(opts->flags, REFRESH);
++
++ AuDbg("status 0x%x\n", opts->flags);
++
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++unsigned int au_opt_udba(struct super_block *sb)
++{
++ return au_mntflags(sb) & AuOptMask_UDBA;
++}
+diff -urN /usr/share/empty/fs/aufs/opts.h linux/fs/aufs/opts.h
+--- /usr/share/empty/fs/aufs/opts.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/opts.h 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,263 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * mount options/flags
++ */
++
++#ifndef __AUFS_OPTS_H__
++#define __AUFS_OPTS_H__
++
++#ifdef __KERNEL__
++
++#include <linux/fs_parser.h>
++#include <linux/namei.h>
++#include <linux/path.h>
++
++enum {
++ Opt_br,
++ Opt_add, Opt_del, Opt_mod, Opt_append, Opt_prepend,
++ Opt_idel, Opt_imod,
++ Opt_dirwh, Opt_rdcache, Opt_rdblk, Opt_rdhash,
++ Opt_xino, Opt_noxino,
++ Opt_trunc_xino, Opt_trunc_xino_v,
++ Opt_trunc_xino_path, Opt_itrunc_xino,
++ Opt_trunc_xib,
++ Opt_shwh,
++ Opt_plink, Opt_list_plink,
++ Opt_udba,
++ Opt_dio,
++ Opt_diropq, Opt_diropq_a, Opt_diropq_w,
++ Opt_warn_perm,
++ Opt_wbr_copyup, Opt_wbr_create,
++ Opt_fhsm_sec,
++ Opt_verbose, Opt_noverbose,
++ Opt_sum, Opt_wsum,
++ Opt_dirperm1,
++ Opt_dirren,
++ Opt_acl,
++ Opt_tail, Opt_ignore, Opt_ignore_silent, Opt_err
++};
++
++/* ---------------------------------------------------------------------- */
++
++/* mount flags */
++#define AuOpt_XINO 1 /* external inode number bitmap
++ and translation table */
++#define AuOpt_TRUNC_XINO (1 << 1) /* truncate xino files */
++#define AuOpt_UDBA_NONE (1 << 2) /* users direct branch access */
++#define AuOpt_UDBA_REVAL (1 << 3)
++#define AuOpt_UDBA_HNOTIFY (1 << 4)
++#define AuOpt_SHWH (1 << 5) /* show whiteout */
++#define AuOpt_PLINK (1 << 6) /* pseudo-link */
++#define AuOpt_DIRPERM1 (1 << 7) /* ignore the lower dir's perm
++ bits */
++#define AuOpt_ALWAYS_DIROPQ (1 << 9) /* policy to creating diropq */
++#define AuOpt_SUM (1 << 10) /* summation for statfs(2) */
++#define AuOpt_SUM_W (1 << 11) /* unimplemented */
++#define AuOpt_WARN_PERM (1 << 12) /* warn when add-branch */
++#define AuOpt_VERBOSE (1 << 13) /* print the cause of error */
++#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
++#endif
++
++#define AuOpt_Def (AuOpt_XINO \
++ | AuOpt_UDBA_REVAL \
++ | AuOpt_PLINK \
++ /* | AuOpt_DIRPERM1 */ \
++ | AuOpt_WARN_PERM)
++#define AuOptMask_UDBA (AuOpt_UDBA_NONE \
++ | AuOpt_UDBA_REVAL \
++ | AuOpt_UDBA_HNOTIFY)
++
++#define AuOpt_LkupDirFlags (LOOKUP_FOLLOW | LOOKUP_DIRECTORY)
++
++#define au_opt_test(flags, name) (flags & AuOpt_##name)
++#define au_opt_set(flags, name) do { \
++ BUILD_BUG_ON(AuOpt_##name & AuOptMask_UDBA); \
++ ((flags) |= AuOpt_##name); \
++} while (0)
++#define au_opt_set_udba(flags, name) do { \
++ (flags) &= ~AuOptMask_UDBA; \
++ ((flags) |= AuOpt_##name); \
++} while (0)
++#define au_opt_clr(flags, name) do { \
++ ((flags) &= ~AuOpt_##name); \
++} while (0)
++
++static inline unsigned int au_opts_plink(unsigned int mntflags)
++{
++#ifdef CONFIG_PROC_FS
++ return mntflags;
++#else
++ return mntflags & ~AuOpt_PLINK;
++#endif
++}
++
++/* ---------------------------------------------------------------------- */
++
++/* policies to select one among multiple writable branches */
++enum {
++ AuWbrCreate_TDP, /* top down parent */
++ AuWbrCreate_RR, /* round robin */
++ AuWbrCreate_MFS, /* most free space */
++ 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 */
++ AuWbrCreate_PMFSRRV, /* plus seconds */
++
++ AuWbrCreate_Def = AuWbrCreate_TDP
++};
++
++enum {
++ AuWbrCopyup_TDP, /* top down parent */
++ AuWbrCopyup_BUP, /* bottom up parent */
++ AuWbrCopyup_BU, /* bottom up */
++
++ AuWbrCopyup_Def = AuWbrCopyup_TDP
++};
++
++/* ---------------------------------------------------------------------- */
++
++struct file;
++
++struct au_opt_add {
++ aufs_bindex_t bindex;
++ char *pathname;
++ int perm;
++ struct path path;
++};
++
++struct au_opt_del {
++ char *pathname;
++ struct path h_path;
++};
++
++struct au_opt_mod {
++ char *path;
++ int perm;
++ struct dentry *h_root;
++};
++
++struct au_opt_xino {
++ char *path;
++ struct file *file;
++};
++
++struct au_opt_xino_itrunc {
++ aufs_bindex_t bindex;
++};
++
++struct au_opt_wbr_create {
++ int wbr_create;
++ int mfs_second;
++ unsigned long long mfsrr_watermark;
++};
++
++struct au_opt {
++ int type;
++ union {
++ struct au_opt_xino xino;
++ struct au_opt_xino_itrunc xino_itrunc;
++ struct au_opt_add add;
++ struct au_opt_del del;
++ struct au_opt_mod mod;
++ int dirwh;
++ int rdcache;
++ unsigned int rdblk;
++ unsigned int rdhash;
++ int udba;
++ struct au_opt_wbr_create wbr_create;
++ int wbr_copyup;
++ unsigned int fhsm_second;
++ bool tf; /* generic flag, true or false */
++ };
++};
++
++/* opts flags */
++#define AuOpts_REMOUNT 1
++#define AuOpts_REFRESH (1 << 1)
++#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;
++
++ unsigned int given_udba;
++ unsigned int flags;
++ unsigned long sb_flags;
++};
++
++/* ---------------------------------------------------------------------- */
++
++/* opts.c */
++int au_br_perm_val(char *perm);
++void au_optstr_br_perm(au_br_perm_str_t *str, int perm);
++int au_udba_val(char *str);
++const char *au_optstr_udba(int udba);
++int au_wbr_create_val(char *str, struct au_opt_wbr_create *create);
++const char *au_optstr_wbr_create(int wbr_create);
++int au_wbr_copyup_val(char *str);
++const char *au_optstr_wbr_copyup(int wbr_copyup);
++
++int au_opt_add(struct au_opt *opt, char *opt_str, unsigned long sb_flags,
++ aufs_bindex_t bindex);
++struct super_block;
++int au_opts_verify(struct super_block *sb, unsigned long sb_flags,
++ unsigned int pending);
++int au_opts_mount(struct super_block *sb, struct au_opts *opts);
++int au_opts_remount(struct super_block *sb, struct au_opts *opts);
++
++unsigned int au_opt_udba(struct super_block *sb);
++
++/* fsctx.c */
++int aufs_fsctx_init(struct fs_context *fc);
++extern const struct fs_parameter_spec aufs_fsctx_paramspec[];
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_OPTS_H__ */
+diff -urN /usr/share/empty/fs/aufs/plink.c linux/fs/aufs/plink.c
+--- /usr/share/empty/fs/aufs/plink.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/plink.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,516 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * pseudo-link
++ */
++
++#include "aufs.h"
++
++/*
++ * the pseudo-link maintenance mode.
++ * during a user process maintains the pseudo-links,
++ * prohibit adding a new plink and branch manipulation.
++ *
++ * Flags
++ * NOPLM:
++ * For entry functions which will handle plink, and i_mutex is already held
++ * in VFS.
++ * They cannot wait and should return an error at once.
++ * Callers has to check the error.
++ * NOPLMW:
++ * For entry functions which will handle plink, but i_mutex is not held
++ * in VFS.
++ * They can wait the plink maintenance mode to finish.
++ *
++ * They behave like F_SETLK and F_SETLKW.
++ * If the caller never handle plink, then both flags are unnecessary.
++ */
++
++int au_plink_maint(struct super_block *sb, int flags)
++{
++ int err;
++ pid_t pid, ppid;
++ struct task_struct *parent, *prev;
++ struct au_sbinfo *sbi;
++
++ SiMustAnyLock(sb);
++
++ err = 0;
++ if (!au_opt_test(au_mntflags(sb), PLINK))
++ goto out;
++
++ sbi = au_sbi(sb);
++ pid = sbi->si_plink_maint_pid;
++ if (!pid || pid == current->pid)
++ goto out;
++
++ /* todo: it highly depends upon /sbin/mount.aufs */
++ prev = NULL;
++ parent = current;
++ ppid = 0;
++ rcu_read_lock();
++ 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;
+ }
-+ /*FALLTHROUGH*/
-+ case AuWbrCreate_MFS:
-+ case AuWbrCreate_PMFS:
-+ create->mfs_second = AUFS_MFS_SECOND_DEF;
-+ break;
-+ case AuWbrCreate_MFSV:
-+ case AuWbrCreate_PMFSV:
-+ e = au_wbr_mfs_sec(&args[0], str, create);
-+ if (unlikely(e))
-+ err = e;
-+ break;
++ prev = parent;
++ }
++ rcu_read_unlock();
++
++ if (au_ftest_lock(flags, NOPLMW)) {
++ /* if there is no i_mutex lock in VFS, we don't need to wait */
++ /* AuDebugOn(!lockdep_depth(current)); */
++ while (sbi->si_plink_maint_pid) {
++ si_read_unlock(sb);
++ /* gave up wake_up_bit() */
++ wait_event(sbi->si_plink_wq, !sbi->si_plink_maint_pid);
++
++ if (au_ftest_lock(flags, FLUSH))
++ au_nwt_flush(&sbi->si_nowait);
++ si_noflush_read_lock(sb);
++ }
++ } else if (au_ftest_lock(flags, NOPLM)) {
++ AuDbg("ppid %d, pid %d\n", ppid, pid);
++ err = -EAGAIN;
++ }
++
++out:
++ return err;
++}
++
++void au_plink_maint_leave(struct au_sbinfo *sbinfo)
++{
++ spin_lock(&sbinfo->si_plink_maint_lock);
++ sbinfo->si_plink_maint_pid = 0;
++ spin_unlock(&sbinfo->si_plink_maint_lock);
++ wake_up_all(&sbinfo->si_plink_wq);
++}
++
++int au_plink_maint_enter(struct super_block *sb)
++{
++ int err;
++ struct au_sbinfo *sbinfo;
++
++ err = 0;
++ sbinfo = au_sbi(sb);
++ /* make sure i am the only one in this fs */
++ si_write_lock(sb, AuLock_FLUSH);
++ if (au_opt_test(au_mntflags(sb), PLINK)) {
++ spin_lock(&sbinfo->si_plink_maint_lock);
++ if (!sbinfo->si_plink_maint_pid)
++ sbinfo->si_plink_maint_pid = current->pid;
++ else
++ err = -EBUSY;
++ spin_unlock(&sbinfo->si_plink_maint_lock);
++ }
++ si_write_unlock(sb);
++
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++#ifdef CONFIG_AUFS_DEBUG
++void au_plink_list(struct super_block *sb)
++{
++ int i;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_icntnr *icntnr;
++
++ SiMustAnyLock(sb);
++
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
++
++ for (i = 0; i < AuPlink_NHASH; i++) {
++ 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);
++ hlist_bl_unlock(hbl);
+ }
++}
++#endif
+
-+ return err;
++/* is the inode pseudo-linked? */
++int au_plink_test(struct inode *inode)
++{
++ int found, i;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_icntnr *icntnr;
++
++ sbinfo = au_sbi(inode->i_sb);
++ AuRwMustAnyLock(&sbinfo->si_rwsem);
++ AuDebugOn(!au_opt_test(au_mntflags(inode->i_sb), PLINK));
++ AuDebugOn(au_plink_maint(inode->i_sb, AuLock_NOPLM));
++
++ found = 0;
++ i = au_plink_hash(inode->i_ino);
++ 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;
++ }
++ hlist_bl_unlock(hbl);
++ return found;
+}
+
-+const char *au_optstr_wbr_create(int wbr_create)
++/* ---------------------------------------------------------------------- */
++
++/*
++ * generate a name for plink.
++ * the file will be stored under AUFS_WH_PLINKDIR.
++ */
++/* 20 is max digits length of ulong 64 */
++#define PLINK_NAME_LEN ((20 + 1) * 2)
++
++static int plink_name(char *name, int len, struct inode *inode,
++ aufs_bindex_t bindex)
+{
-+ return au_parser_pattern(wbr_create, (void *)au_wbr_create_policy);
++ int rlen;
++ struct inode *h_inode;
++
++ h_inode = au_h_iptr(inode, bindex);
++ rlen = snprintf(name, len, "%lu.%lu", inode->i_ino, h_inode->i_ino);
++ return rlen;
+}
+
-+static match_table_t au_wbr_copyup_policy = {
-+ {AuWbrCopyup_TDP, "tdp"},
-+ {AuWbrCopyup_TDP, "top-down-parent"},
-+ {AuWbrCopyup_BUP, "bup"},
-+ {AuWbrCopyup_BUP, "bottom-up-parent"},
-+ {AuWbrCopyup_BU, "bu"},
-+ {AuWbrCopyup_BU, "bottom-up"},
-+ {-1, NULL}
++struct au_do_plink_lkup_args {
++ struct dentry **errp;
++ struct qstr *tgtname;
++ struct path *h_ppath;
+};
+
-+static noinline_for_stack int au_wbr_copyup_val(char *str)
++static struct dentry *au_do_plink_lkup(struct qstr *tgtname,
++ struct path *h_ppath)
+{
-+ substring_t args[MAX_OPT_ARGS];
-+ return match_token(str, au_wbr_copyup_policy, args);
++ struct dentry *h_dentry;
++ struct inode *h_inode;
++
++ h_inode = d_inode(h_ppath->dentry);
++ inode_lock_shared_nested(h_inode, AuLsc_I_CHILD2);
++ h_dentry = vfsub_lkup_one(tgtname, h_ppath);
++ inode_unlock_shared(h_inode);
++
++ return h_dentry;
+}
+
-+const char *au_optstr_wbr_copyup(int wbr_copyup)
++static void au_call_do_plink_lkup(void *args)
+{
-+ return au_parser_pattern(wbr_copyup, (void *)au_wbr_copyup_policy);
++ struct au_do_plink_lkup_args *a = args;
++ *a->errp = au_do_plink_lkup(a->tgtname, a->h_ppath);
+}
+
-+/* ---------------------------------------------------------------------- */
++/* lookup the plink-ed @inode under the branch at @bindex */
++struct dentry *au_plink_lkup(struct inode *inode, aufs_bindex_t bindex)
++{
++ struct dentry *h_dentry;
++ struct au_branch *br;
++ struct path h_ppath;
++ int wkq_err;
++ char a[PLINK_NAME_LEN];
++ struct qstr tgtname = QSTR_INIT(a, 0);
++
++ AuDebugOn(au_plink_maint(inode->i_sb, AuLock_NOPLM));
++
++ br = au_sbr(inode->i_sb, bindex);
++ h_ppath.dentry = br->br_wbr->wbr_plink;
++ h_ppath.mnt = au_br_mnt(br);
++ tgtname.len = plink_name(a, sizeof(a), inode, bindex);
++
++ if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID)) {
++ struct au_do_plink_lkup_args args = {
++ .errp = &h_dentry,
++ .tgtname = &tgtname,
++ .h_ppath = &h_ppath
++ };
++
++ wkq_err = au_wkq_wait(au_call_do_plink_lkup, &args);
++ if (unlikely(wkq_err))
++ h_dentry = ERR_PTR(wkq_err);
++ } else
++ h_dentry = au_do_plink_lkup(&tgtname, &h_ppath);
+
-+static const int lkup_dirflags = LOOKUP_FOLLOW | LOOKUP_DIRECTORY;
++ return h_dentry;
++}
+
-+static void dump_opts(struct au_opts *opts)
++/* create a pseudo-link */
++static int do_whplink(struct qstr *tgt, struct path *h_ppath,
++ struct dentry *h_dentry)
+{
-+#ifdef CONFIG_AUFS_DEBUG
-+ /* reduce stack space */
-+ union {
-+ struct au_opt_add *add;
-+ struct au_opt_del *del;
-+ struct au_opt_mod *mod;
-+ struct au_opt_xino *xino;
-+ struct au_opt_xinodir *xinodir;
-+ struct au_opt_xino_itrunc *xino_itrunc;
-+ struct au_opt_wbr_create *create;
-+ } u;
-+ struct au_opt *opt;
++ int err;
++ struct path h_path;
++ struct inode *h_dir, *delegated;
++
++ h_dir = d_inode(h_ppath->dentry);
++ inode_lock_nested(h_dir, AuLsc_I_CHILD2);
++ h_path.mnt = h_ppath->mnt;
++again:
++ h_path.dentry = vfsub_lkup_one(tgt, h_ppath);
++ err = PTR_ERR(h_path.dentry);
++ if (IS_ERR(h_path.dentry))
++ goto out;
+
-+ AuTraceEnter();
++ err = 0;
++ /* wh.plink dir is not monitored */
++ /* todo: is it really safe? */
++ if (d_is_positive(h_path.dentry)
++ && d_inode(h_path.dentry) != d_inode(h_dentry)) {
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, &h_path, &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ dput(h_path.dentry);
++ h_path.dentry = NULL;
++ if (!err)
++ goto again;
++ }
++ if (!err && d_is_negative(h_path.dentry)) {
++ delegated = NULL;
++ err = vfsub_link(h_dentry, h_dir, &h_path, &delegated);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
++ }
++ dput(h_path.dentry);
+
-+ opt = opts->opt;
-+ while (/* opt < opts_tail && */ opt->type != Opt_tail) {
-+ switch (opt->type) {
-+ case Opt_add:
-+ u.add = &opt->add;
-+ LKTRTrace("add {b%d, %s, 0x%x, %p}\n",
-+ u.add->bindex, u.add->path, u.add->perm,
-+ u.add->nd.path.dentry);
-+ break;
-+ case Opt_del:
-+ case Opt_idel:
-+ u.del = &opt->del;
-+ LKTRTrace("del {%s, %p}\n", u.del->path, u.del->h_root);
-+ break;
-+ case Opt_mod:
-+ case Opt_imod:
-+ u.mod = &opt->mod;
-+ LKTRTrace("mod {%s, 0x%x, %p}\n",
-+ u.mod->path, u.mod->perm, u.mod->h_root);
-+ break;
-+ case Opt_append:
-+ u.add = &opt->add;
-+ LKTRTrace("append {b%d, %s, 0x%x, %p}\n",
-+ u.add->bindex, u.add->path, u.add->perm,
-+ u.add->nd.path.dentry);
-+ break;
-+ case Opt_prepend:
-+ u.add = &opt->add;
-+ LKTRTrace("prepend {b%d, %s, 0x%x, %p}\n",
-+ u.add->bindex, u.add->path, u.add->perm,
-+ u.add->nd.path.dentry);
-+ break;
-+ case Opt_dirwh:
-+ LKTRTrace("dirwh %d\n", opt->dirwh);
-+ break;
-+ case Opt_rdcache:
-+ LKTRTrace("rdcache %d\n", opt->rdcache);
-+ break;
-+ case Opt_xino:
-+ u.xino = &opt->xino;
-+ LKTRTrace("xino {%s %.*s}\n",
-+ u.xino->path,
-+ AuDLNPair(u.xino->file->f_dentry));
-+ break;
-+ case Opt_xinodir:
-+ u.xinodir = &opt->xinodir;
-+ LKTRTrace("xinodir {%s %.*s}\n",
-+ u.xinodir->name,
-+ AuDLNPair(u.xinodir->path.dentry));
-+ break;
-+ case Opt_trunc_xino:
-+ LKTRLabel(trunc_xino);
-+ break;
-+ case Opt_notrunc_xino:
-+ LKTRLabel(notrunc_xino);
-+ break;
-+ case Opt_trunc_xino_path:
-+ case Opt_itrunc_xino:
-+ u.xino_itrunc = &opt->xino_itrunc;
-+ LKTRTrace("trunc_xino %d\n", u.xino_itrunc->bindex);
-+ break;
++out:
++ inode_unlock(h_dir);
++ return err;
++}
+
-+ case Opt_noxino:
-+ LKTRLabel(noxino);
-+ break;
-+ case Opt_trunc_xib:
-+ LKTRLabel(trunc_xib);
-+ break;
-+ case Opt_notrunc_xib:
-+ LKTRLabel(notrunc_xib);
-+ break;
-+ case Opt_dirperm1:
-+ LKTRLabel(dirperm1);
-+ break;
-+ case Opt_nodirperm1:
-+ LKTRLabel(nodirperm1);
-+ break;
-+ case Opt_shwh:
-+ LKTRLabel(shwh);
-+ break;
-+ case Opt_noshwh:
-+ LKTRLabel(noshwh);
-+ break;
-+ case Opt_plink:
-+ LKTRLabel(plink);
-+ break;
-+ case Opt_noplink:
-+ LKTRLabel(noplink);
-+ break;
-+ case Opt_list_plink:
-+ LKTRLabel(list_plink);
-+ break;
-+ case Opt_clean_plink:
-+ LKTRLabel(clean_plink);
-+ break;
-+ case Opt_udba:
-+ LKTRTrace("udba %d, %s\n",
-+ opt->udba, au_optstr_udba(opt->udba));
-+ break;
-+ case Opt_diropq_a:
-+ LKTRLabel(diropq_a);
-+ break;
-+ case Opt_diropq_w:
-+ LKTRLabel(diropq_w);
-+ break;
-+ case Opt_warn_perm:
-+ LKTRLabel(warn_perm);
-+ break;
-+ case Opt_nowarn_perm:
-+ LKTRLabel(nowarn_perm);
-+ break;
-+ case Opt_dlgt:
-+ LKTRLabel(dlgt);
-+ break;
-+ case Opt_nodlgt:
-+ LKTRLabel(nodlgt);
-+ break;
-+ case Opt_refrof:
-+ LKTRLabel(refrof);
-+ break;
-+ case Opt_norefrof:
-+ LKTRLabel(norefrof);
-+ break;
-+ case Opt_verbose:
-+ LKTRLabel(verbose);
-+ break;
-+ case Opt_noverbose:
-+ LKTRLabel(noverbose);
-+ break;
-+ case Opt_sum:
-+ LKTRLabel(sum);
-+ break;
-+ case Opt_nosum:
-+ LKTRLabel(nosum);
-+ break;
-+ case Opt_wsum:
-+ LKTRLabel(wsum);
++struct do_whplink_args {
++ int *errp;
++ struct qstr *tgt;
++ struct path *h_ppath;
++ struct dentry *h_dentry;
++};
++
++static void call_do_whplink(void *args)
++{
++ struct do_whplink_args *a = args;
++ *a->errp = do_whplink(a->tgt, a->h_ppath, a->h_dentry);
++}
++
++static int whplink(struct dentry *h_dentry, struct inode *inode,
++ aufs_bindex_t bindex)
++{
++ int err, wkq_err;
++ struct au_branch *br;
++ struct au_wbr *wbr;
++ struct path h_ppath;
++ char a[PLINK_NAME_LEN];
++ struct qstr tgtname = QSTR_INIT(a, 0);
++
++ br = au_sbr(inode->i_sb, bindex);
++ wbr = br->br_wbr;
++ h_ppath.dentry = wbr->wbr_plink;
++ h_ppath.mnt = au_br_mnt(br);
++ tgtname.len = plink_name(a, sizeof(a), inode, bindex);
++
++ /* always superio. */
++ if (!uid_eq(current_fsuid(), GLOBAL_ROOT_UID)) {
++ struct do_whplink_args args = {
++ .errp = &err,
++ .tgt = &tgtname,
++ .h_ppath = &h_ppath,
++ .h_dentry = h_dentry
++ };
++ wkq_err = au_wkq_wait(call_do_whplink, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
++ } else
++ err = do_whplink(&tgtname, &h_ppath, h_dentry);
++
++ return err;
++}
++
++/*
++ * create a new pseudo-link for @h_dentry on @bindex.
++ * the linked inode is held in aufs @inode.
++ */
++void au_plink_append(struct inode *inode, aufs_bindex_t bindex,
++ struct dentry *h_dentry)
++{
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_icntnr *icntnr;
++ int found, err, cnt, i;
++
++ sb = inode->i_sb;
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
++
++ found = au_plink_test(inode);
++ if (found)
++ return;
++
++ i = au_plink_hash(inode->i_ino);
++ hbl = sbinfo->si_plink + i;
++ au_igrab(inode);
++
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry(icntnr, pos, hbl, plink) {
++ if (&icntnr->vfs_inode == inode) {
++ found = 1;
+ break;
-+ case Opt_coo:
-+ LKTRTrace("coo %d, %s\n",
-+ opt->coo, au_optstr_coo(opt->coo));
++ }
++ }
++ if (!found) {
++ icntnr = container_of(inode, struct au_icntnr, vfs_inode);
++ hlist_bl_add_head(&icntnr->plink, hbl);
++ }
++ hlist_bl_unlock(hbl);
++ if (!found) {
++ cnt = au_hbl_count(hbl);
++#define msg "unexpectedly unbalanced or too many pseudo-links"
++ if (cnt > AUFS_PLINK_WARN)
++ AuWarn1(msg ", %d\n", cnt);
++#undef msg
++ err = whplink(h_dentry, inode, bindex);
++ if (unlikely(err)) {
++ pr_warn("err %d, damaged pseudo link.\n", err);
++ au_hbl_del(&icntnr->plink, hbl);
++ iput(&icntnr->vfs_inode);
++ }
++ } else
++ iput(&icntnr->vfs_inode);
++}
++
++/* free all plinks */
++void au_plink_put(struct super_block *sb, int verbose)
++{
++ int i, warned;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
++ struct au_icntnr *icntnr;
++
++ SiMustWriteLock(sb);
++
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
++
++ /* no spin_lock since sbinfo is write-locked */
++ warned = 0;
++ for (i = 0; i < AuPlink_NHASH; i++) {
++ hbl = sbinfo->si_plink + i;
++ if (!warned && verbose && !hlist_bl_empty(hbl)) {
++ pr_warn("pseudo-link is not flushed");
++ warned = 1;
++ }
++ hlist_bl_for_each_entry_safe(icntnr, pos, tmp, hbl, plink)
++ iput(&icntnr->vfs_inode);
++ INIT_HLIST_BL_HEAD(hbl);
++ }
++}
++
++void au_plink_clean(struct super_block *sb, int verbose)
++{
++ struct dentry *root;
++
++ root = sb->s_root;
++ aufs_write_lock(root);
++ if (au_opt_test(au_mntflags(sb), PLINK))
++ au_plink_put(sb, verbose);
++ aufs_write_unlock(root);
++}
++
++static int au_plink_do_half_refresh(struct inode *inode, aufs_bindex_t br_id)
++{
++ int do_put;
++ aufs_bindex_t btop, bbot, bindex;
++
++ do_put = 0;
++ btop = au_ibtop(inode);
++ bbot = au_ibbot(inode);
++ if (btop >= 0) {
++ for (bindex = btop; bindex <= bbot; bindex++) {
++ if (!au_h_iptr(inode, bindex)
++ || au_ii_br_id(inode, bindex) != br_id)
++ continue;
++ au_set_h_iptr(inode, bindex, NULL, 0);
++ do_put = 1;
+ break;
-+ case Opt_wbr_create:
-+ u.create = &opt->wbr_create;
-+ LKTRTrace("create %d, %s\n", u.create->wbr_create,
-+ au_optstr_wbr_create(u.create->wbr_create));
-+ switch (u.create->wbr_create) {
-+ case AuWbrCreate_MFSV:
-+ case AuWbrCreate_PMFSV:
-+ LKTRTrace("%d sec\n", u.create->mfs_second);
-+ break;
-+ case AuWbrCreate_MFSRR:
-+ LKTRTrace("%llu watermark\n",
-+ u.create->mfsrr_watermark);
-+ break;
-+ case AuWbrCreate_MFSRRV:
-+ LKTRTrace("%llu watermark, %d sec\n",
-+ u.create->mfsrr_watermark,
-+ u.create->mfs_second);
-+ break;
++ }
++ if (do_put)
++ for (bindex = btop; bindex <= bbot; bindex++)
++ if (au_h_iptr(inode, bindex)) {
++ do_put = 0;
++ break;
++ }
++ } else
++ do_put = 1;
++
++ return do_put;
++}
++
++/* free the plinks on a branch specified by @br_id */
++void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id)
++{
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *tmp;
++ struct au_icntnr *icntnr;
++ struct inode *inode;
++ int i, do_put;
++
++ SiMustWriteLock(sb);
++
++ sbinfo = au_sbi(sb);
++ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ AuDebugOn(au_plink_maint(sb, AuLock_NOPLM));
++
++ /* no bit_lock since sbinfo is write-locked */
++ for (i = 0; i < AuPlink_NHASH; i++) {
++ 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_bl_del(&icntnr->plink);
++ iput(inode);
+ }
-+ break;
-+ case Opt_wbr_copyup:
-+ LKTRTrace("copyup %d, %s\n", opt->wbr_copyup,
-+ au_optstr_wbr_copyup(opt->wbr_copyup));
-+ break;
-+ default:
-+ BUG();
++ ii_write_unlock(inode);
++ iput(inode);
+ }
-+ opt++;
+ }
-+#endif
+}
+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 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,51 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
++
++/*
++ * poll operation
++ * There is only one filesystem which implements ->poll operation, currently.
++ */
+
-+void au_opts_free(struct au_opts *opts)
++#include "aufs.h"
++
++__poll_t aufs_poll(struct file *file, struct poll_table_struct *pt)
+{
-+ struct au_opt *opt;
++ __poll_t mask;
++ struct file *h_file;
++ struct super_block *sb;
+
-+ AuTraceEnter();
++ /* We should pretend an error happened. */
++ mask = EPOLLERR /* | EPOLLIN | EPOLLOUT */;
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH | AuLock_NOPLMW);
+
-+ opt = opts->opt;
-+ while (opt->type != Opt_tail) {
-+ switch (opt->type) {
-+ case Opt_add:
-+ case Opt_append:
-+ case Opt_prepend:
-+ path_put(&opt->add.nd.path);
-+ break;
-+ case Opt_del:
-+ case Opt_idel:
-+ dput(opt->del.h_root);
-+ break;
-+ case Opt_mod:
-+ case Opt_imod:
-+ dput(opt->mod.h_root);
-+ break;
-+ case Opt_xino:
-+ fput(opt->xino.file);
-+ break;
-+ case Opt_xinodir:
-+ path_put(&opt->xinodir.path);
-+ break;
-+ }
-+ opt++;
++ h_file = au_read_pre(file, /*keep_fi*/0, /*lsc*/0);
++ if (IS_ERR(h_file)) {
++ AuDbg("h_file %ld\n", PTR_ERR(h_file));
++ goto out;
+ }
++
++ mask = vfs_poll(h_file, pt);
++ fput(h_file); /* instead of au_read_post() */
++
++out:
++ si_read_unlock(sb);
++ if (mask & EPOLLERR)
++ AuDbg("mask 0x%x\n", mask);
++ return mask;
+}
+diff -urN /usr/share/empty/fs/aufs/posix_acl.c linux/fs/aufs/posix_acl.c
+--- /usr/share/empty/fs/aufs/posix_acl.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/posix_acl.c 2023-08-28 12:34:39.959969465 +0200
+@@ -0,0 +1,108 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2014-2022 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/>.
++ */
++
++/*
++ * posix acl operations
++ */
+
-+static int opt_add(struct au_opt *opt, char *opt_str, unsigned long sb_flags,
-+ struct super_block *sb, aufs_bindex_t bindex)
++#include <linux/fs.h>
++#include "aufs.h"
++
++struct posix_acl *aufs_get_inode_acl(struct inode *inode, int type, bool rcu)
+{
++ struct posix_acl *acl;
+ int err;
-+ struct au_opt_add *add = &opt->add;
-+ char *p;
++ aufs_bindex_t bindex;
++ struct inode *h_inode;
++ struct super_block *sb;
+
-+ LKTRTrace("%s, b%d\n", opt_str, bindex);
++ acl = ERR_PTR(-ECHILD);
++ if (rcu)
++ goto out;
+
-+ add->bindex = bindex;
-+ add->perm = AuBrPerm_Last;
-+ add->path = opt_str;
-+ p = strchr(opt_str, '=');
-+ if (p) {
-+ *p++ = 0;
-+ if (*p)
-+ add->perm = br_perm_val(p);
-+ }
++ acl = NULL;
++ sb = inode->i_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ ii_read_lock_child(inode);
++ if (!(sb->s_flags & SB_POSIXACL))
++ goto unlock;
+
-+ /* LSM may detect it */
-+ /* do not superio. */
-+ err = vfsub_path_lookup(add->path, lkup_dirflags, &add->nd);
-+ if (!err) {
-+ if (!p /* && add->perm == AuBrPerm_Last */) {
-+ add->perm = AuBrPerm_RO;
-+ if (au_test_def_rr(add->nd.path.dentry->d_sb))
-+ add->perm = AuBrPerm_RR;
-+ if (!bindex && !(sb_flags & MS_RDONLY))
-+ add->perm = AuBrPerm_RW;
-+#ifdef CONFIG_AUFS_COMPAT
-+ add->perm = AuBrPerm_RW;
-+#endif
-+ }
-+ opt->type = Opt_add;
-+ goto out;
-+ }
-+ AuErr("lookup failed %s (%d)\n", add->path, err);
-+ err = -EINVAL;
++ bindex = au_ibtop(inode);
++ h_inode = au_h_iptr(inode, bindex);
++ if (unlikely(!h_inode
++ || ((h_inode->i_mode & S_IFMT)
++ != (inode->i_mode & S_IFMT)))) {
++ err = au_busy_or_stale();
++ acl = ERR_PTR(err);
++ goto unlock;
++ }
++
++ /* always topmost only */
++ acl = get_inode_acl(h_inode, type);
++ if (IS_ERR(acl))
++ forget_cached_acl(inode, type);
++ else
++ set_cached_acl(inode, type, acl);
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++unlock:
++ ii_read_unlock(inode);
++ si_read_unlock(sb);
++
++out:
++ AuTraceErrPtr(acl);
++ return acl;
+}
+
-+/* called without aufs lock */
-+int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts)
++struct posix_acl *aufs_get_acl(struct mnt_idmap *idmap,
++ struct dentry *dentry, int type)
+{
-+ int err, n, token;
-+ struct dentry *root;
-+ struct au_opt *opt, *opt_tail;
-+ char *opt_str, *p;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char skipped;
-+ union {
-+ struct au_opt_del *del;
-+ struct au_opt_mod *mod;
-+ struct au_opt_xino *xino;
-+ struct au_opt_xinodir *xinodir;
-+ struct au_opt_xino_itrunc *xino_itrunc;
-+ struct au_opt_wbr_create *create;
-+ } u;
-+ struct file *file;
-+ /* reduce the stack space */
-+ struct {
-+ substring_t args[MAX_OPT_ARGS];
-+ struct nameidata nd;
-+ } *a;
-+
-+ LKTRTrace("%s, nopts %d\n", str, opts->max_opt);
++ struct posix_acl *acl;
++ struct inode *inode;
+
-+ err = -ENOMEM;
-+ a = kmalloc(sizeof(*a), GFP_NOFS);
-+ if (unlikely(!a))
-+ goto out;
++ inode = d_inode(dentry);
++ acl = aufs_get_inode_acl(inode, type, /*rcu*/false);
+
-+ root = sb->s_root;
-+ err = 0;
-+ bindex = 0;
-+ opt = opts->opt;
-+ opt_tail = opt + opts->max_opt - 1;
-+ opt->type = Opt_tail;
-+ while (!err && (opt_str = strsep(&str, ",")) && *opt_str) {
-+ err = -EINVAL;
-+ token = match_token(opt_str, options, a->args);
-+ LKTRTrace("%s, token %d, a->args[0]{%p, %p}\n",
-+ opt_str, token, a->args[0].from, a->args[0].to);
++ return acl;
++}
+
-+ skipped = 0;
-+ switch (token) {
-+ case Opt_br:
-+ err = 0;
-+ while (!err && (opt_str = strsep(&a->args[0].from, ":"))
-+ && *opt_str) {
-+ err = opt_add(opt, opt_str, opts->sb_flags, sb,
-+ bindex++);
-+ if (unlikely(!err && ++opt > opt_tail)) {
-+ err = -E2BIG;
-+ break;
-+ }
-+ opt->type = Opt_tail;
-+ skipped = 1;
-+ }
-+ break;
-+ case Opt_add:
-+ if (unlikely(match_int(&a->args[0], &n))) {
-+ AuErr("bad integer in %s\n", opt_str);
-+ break;
-+ }
-+ bindex = n;
-+ err = opt_add(opt, a->args[1].from, opts->sb_flags, sb,
-+ bindex);
-+ break;
-+ case Opt_append:
-+ err = opt_add(opt, a->args[0].from, opts->sb_flags, sb,
-+ /*dummy bindex*/1);
-+ if (!err)
-+ opt->type = token;
-+ break;
-+ case Opt_prepend:
-+ err = opt_add(opt, a->args[0].from, opts->sb_flags, sb,
-+ /*bindex*/0);
-+ if (!err)
-+ opt->type = token;
-+ break;
-+ case Opt_del:
-+ u.del = &opt->del;
-+ u.del->path = a->args[0].from;
-+ LKTRTrace("del path %s\n", u.del->path);
-+ /* LSM may detect it */
-+ /* do not superio. */
-+ err = vfsub_path_lookup(u.del->path, lkup_dirflags,
-+ &a->nd);
-+ if (unlikely(err)) {
-+ AuErr("lookup failed %s (%d)\n",
-+ u.del->path, err);
-+ break;
-+ }
-+ u.del->h_root = dget(a->nd.path.dentry);
-+ path_put(&a->nd.path);
-+ opt->type = token;
-+ break;
-+#if 0 /* reserved for future use */
-+ case Opt_idel:
-+ u.del = &opt->del;
-+ u.del->path = "(indexed)";
-+ if (unlikely(match_int(&a->args[0], &n))) {
-+ AuErr("bad integer in %s\n", opt_str);
-+ break;
-+ }
-+ bindex = n;
-+ aufs_read_lock(root, AuLock_FLUSH);
-+ if (bindex < 0 || au_sbend(sb) < bindex) {
-+ AuErr("out of bounds, %d\n", bindex);
-+ aufs_read_unlock(root, !AuLock_IR);
-+ break;
-+ }
-+ err = 0;
-+ u.del->h_root = dget(au_h_dptr(root, bindex));
-+ opt->type = token;
-+ aufs_read_unlock(root, !AuLock_IR);
-+ break;
-+#endif
-+ case Opt_mod:
-+ u.mod = &opt->mod;
-+ u.mod->path = a->args[0].from;
-+ p = strchr(u.mod->path, '=');
-+ if (unlikely(!p)) {
-+ AuErr("no permssion %s\n", opt_str);
-+ break;
-+ }
-+ *p++ = 0;
-+ u.mod->perm = br_perm_val(p);
-+ LKTRTrace("mod path %s, perm 0x%x, %s\n",
-+ u.mod->path, u.mod->perm, p);
-+ /* LSM may detect it */
-+ /* do not superio. */
-+ err = vfsub_path_lookup(u.mod->path, lkup_dirflags,
-+ &a->nd);
-+ if (unlikely(err)) {
-+ AuErr("lookup failed %s (%d)\n",
-+ u.mod->path, err);
-+ break;
-+ }
-+ u.mod->h_root = dget(a->nd.path.dentry);
-+ path_put(&a->nd.path);
-+ opt->type = token;
-+ break;
-+#ifdef IMOD /* reserved for future use */
-+ case Opt_imod:
-+ u.mod = &opt->mod;
-+ u.mod->path = "(indexed)";
-+ if (unlikely(match_int(&a->args[0], &n))) {
-+ AuErr("bad integer in %s\n", opt_str);
-+ break;
-+ }
-+ bindex = n;
-+ aufs_read_lock(root, AuLock_FLUSH);
-+ if (bindex < 0 || au_sbend(sb) < bindex) {
-+ AuErr("out of bounds, %d\n", bindex);
-+ aufs_read_unlock(root, !AuLock_IR);
-+ break;
-+ }
-+ u.mod->perm = br_perm_val(a->args[1].from);
-+ LKTRTrace("mod path %s, perm 0x%x, %s\n",
-+ u.mod->path, u.mod->perm, a->args[1].from);
-+ err = 0;
-+ u.mod->h_root = dget(au_h_dptr(root, bindex));
-+ opt->type = token;
-+ aufs_read_unlock(root, !AuLock_IR);
-+ break;
-+#endif
-+ case Opt_xino:
-+ u.xino = &opt->xino;
-+ file = au_xino_create(sb, a->args[0].from, /*silent*/0);
-+ err = PTR_ERR(file);
-+ if (IS_ERR(file))
-+ break;
-+ err = -EINVAL;
-+ if (unlikely(file->f_dentry->d_sb == sb)) {
-+ fput(file);
-+ AuErr("%s must be outside\n", a->args[0].from);
-+ break;
-+ }
-+ err = 0;
-+ u.xino->file = file;
-+ u.xino->path = a->args[0].from;
-+ opt->type = token;
-+ break;
++int aufs_set_acl(struct mnt_idmap *idmap, struct dentry *dentry,
++ struct posix_acl *acl, int type)
++{
++ int err;
++ ssize_t ssz;
++ struct inode *inode;
++ struct au_sxattr arg = {
++ .type = AU_ACL_SET,
++ .u.acl_set = {
++ .acl = acl,
++ .type = type
++ },
++ };
+
-+#if 0 /* def CONFIG_AUFS_EXPORT */ /* reserved for futur use */
-+ case Opt_xinodir:
-+ u.xinodir = &opt->xinodir;
-+ u.xinodir->name = a->args[0].from;
-+ err = vfsub_path_lookup(u.xinodir->name, lkup_dirflags,
-+ &a->nd);
-+ if (unlikely(err)) {
-+ AuErr("lookup failed %s (%d)\n",
-+ u.xinodir->name, err);
-+ break;
-+ }
-+ u.xinodir->path = a->nd.path;
-+ /* do not path_put() */
-+ opt->type = token;
-+ break;
-+#endif
++ inode = d_inode(dentry);
++ IMustLock(inode);
+
-+ case Opt_trunc_xino_path:
-+ u.xino_itrunc = &opt->xino_itrunc;
-+ p = a->args[0].from;
-+ LKTRTrace("trunc_xino path %s\n", p);
-+ /* LSM may detect it */
-+ /* do not superio. */
-+ err = vfsub_path_lookup(p, lkup_dirflags, &a->nd);
-+ if (unlikely(err)) {
-+ AuErr("lookup failed %s (%d)\n", p , err);
-+ break;
-+ }
-+ u.xino_itrunc->bindex = -1;
-+ aufs_read_lock(root, AuLock_FLUSH);
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++) {
-+ if (au_h_dptr(root, bindex)
-+ == a->nd.path.dentry) {
-+ u.xino_itrunc->bindex = bindex;
-+ break;
-+ }
-+ }
-+ aufs_read_unlock(root, !AuLock_IR);
-+ path_put(&a->nd.path);
-+ if (unlikely(u.xino_itrunc->bindex < 0)) {
-+ AuErr("no such branch %s\n", p);
-+ err = -EINVAL;
-+ break;
-+ }
-+ opt->type = token;
-+ break;
++ ssz = au_sxattr(dentry, inode, &arg);
++ /* forget even it if succeeds since the branch might set differently */
++ forget_cached_acl(inode, type);
++ err = ssz;
++ if (ssz >= 0)
++ err = 0;
+
-+ case Opt_itrunc_xino:
-+ u.xino_itrunc = &opt->xino_itrunc;
-+ if (unlikely(match_int(&a->args[0], &n))) {
-+ AuErr("bad integer in %s\n", opt_str);
-+ break;
-+ }
-+ u.xino_itrunc->bindex = n;
-+ aufs_read_lock(root, AuLock_FLUSH);
-+ if (n < 0 || au_sbend(sb) < n) {
-+ AuErr("out of bounds, %d\n", n);
-+ aufs_read_unlock(root, !AuLock_IR);
-+ break;
-+ }
-+ aufs_read_unlock(root, !AuLock_IR);
-+ err = 0;
-+ opt->type = token;
-+ break;
++ 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 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,170 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2010-2022 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/>.
++ */
+
-+ case Opt_dirwh:
-+ if (unlikely(match_int(&a->args[0], &opt->dirwh)))
-+ break;
-+ err = 0;
-+ opt->type = token;
-+ break;
++/*
++ * procfs interfaces
++ */
+
-+ case Opt_rdcache:
-+ if (unlikely(match_int(&a->args[0], &opt->rdcache)))
-+ break;
-+ err = 0;
-+ opt->type = token;
-+ break;
++#include <linux/proc_fs.h>
++#include "aufs.h"
+
-+ case Opt_shwh:
-+ if (opts->sb_flags & MS_RDONLY) {
-+ err = 0;
-+ opt->type = token;
-+ } else
-+ AuErr("shwh requires ro\n");
-+ break;
++static int au_procfs_plm_release(struct inode *inode, struct file *file)
++{
++ struct au_sbinfo *sbinfo;
+
-+ case Opt_trunc_xino:
-+ case Opt_notrunc_xino:
-+ case Opt_noxino:
-+ case Opt_trunc_xib:
-+ case Opt_notrunc_xib:
-+ case Opt_dirperm1:
-+ case Opt_nodirperm1:
-+ case Opt_noshwh:
-+ case Opt_plink:
-+ case Opt_noplink:
-+ case Opt_list_plink:
-+ case Opt_clean_plink:
-+ case Opt_diropq_a:
-+ case Opt_diropq_w:
-+ case Opt_warn_perm:
-+ case Opt_nowarn_perm:
-+ case Opt_dlgt:
-+ case Opt_nodlgt:
-+ case Opt_refrof:
-+ case Opt_norefrof:
-+ case Opt_verbose:
-+ case Opt_noverbose:
-+ case Opt_sum:
-+ case Opt_nosum:
-+ case Opt_wsum:
-+ err = 0;
-+ opt->type = token;
-+ break;
++ sbinfo = file->private_data;
++ if (sbinfo) {
++ au_plink_maint_leave(sbinfo);
++ kobject_put(&sbinfo->si_kobj);
++ }
+
-+ case Opt_udba:
-+ opt->udba = udba_val(a->args[0].from);
-+ if (opt->udba >= 0) {
-+ err = 0;
-+ opt->type = token;
-+ } else
-+ AuErr("wrong value, %s\n", opt_str);
-+ break;
++ return 0;
++}
+
-+ case Opt_wbr_create:
-+ u.create = &opt->wbr_create;
-+ u.create->wbr_create
-+ = au_wbr_create_val(a->args[0].from, u.create);
-+ if (u.create->wbr_create >= 0) {
-+ err = 0;
-+ opt->type = token;
-+ } else
-+ AuErr("wrong value, %s\n", opt_str);
-+ break;
-+ case Opt_wbr_copyup:
-+ opt->wbr_copyup = au_wbr_copyup_val(a->args[0].from);
-+ if (opt->wbr_copyup >= 0) {
-+ err = 0;
-+ opt->type = token;
-+ } else
-+ AuErr("wrong value, %s\n", opt_str);
-+ break;
++static void au_procfs_plm_write_clean(struct file *file)
++{
++ struct au_sbinfo *sbinfo;
+
-+ case Opt_coo:
-+ opt->coo = coo_val(a->args[0].from);
-+ if (opt->coo >= 0) {
-+ if (opt->coo == AuOpt_COO_NONE
-+ || !(opts->sb_flags & MS_RDONLY)) {
-+ err = 0;
-+ opt->type = token;
-+ } else
-+ AuErr("bad %s for readonly mount\n",
-+ opt_str);
-+ } else
-+ AuErr("wrong value, %s\n", opt_str);
-+ break;
++ sbinfo = file->private_data;
++ if (sbinfo)
++ au_plink_clean(sbinfo->si_sb, /*verbose*/0);
++}
+
-+ case Opt_ignore:
-+#ifndef CONFIG_AUFS_COMPAT
-+ AuWarn("ignored %s\n", opt_str);
-+#endif
-+ case Opt_ignore_silent:
-+ skipped = 1;
-+ err = 0;
-+ break;
-+ case Opt_err:
-+ AuErr("unknown option %s\n", opt_str);
-+ break;
-+ }
++static int au_procfs_plm_write_si(struct file *file, unsigned long id)
++{
++ int err;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_node *pos;
+
-+ if (!err && !skipped) {
-+ if (unlikely(++opt > opt_tail)) {
-+ err = -E2BIG;
-+ opt--;
-+ opt->type = Opt_tail;
-+ break;
-+ }
-+ opt->type = Opt_tail;
++ err = -EBUSY;
++ if (unlikely(file->private_data))
++ goto out;
++
++ sb = NULL;
++ /* don't use au_sbilist_lock() here */
++ hlist_bl_lock(&au_sbilist);
++ hlist_bl_for_each_entry(sbinfo, pos, &au_sbilist, si_list)
++ if (id == sysaufs_si_id(sbinfo)) {
++ if (kobject_get_unless_zero(&sbinfo->si_kobj))
++ sb = sbinfo->si_sb;
++ break;
+ }
-+ }
++ hlist_bl_unlock(&au_sbilist);
+
-+ kfree(a);
-+ dump_opts(opts);
-+ if (unlikely(err))
-+ au_opts_free(opts);
++ err = -EINVAL;
++ if (unlikely(!sb))
++ goto out;
+
-+ out:
-+ AuTraceErr(err);
++ err = au_plink_maint_enter(sb);
++ if (!err)
++ /* keep kobject_get() */
++ file->private_data = sbinfo;
++ else
++ kobject_put(&sbinfo->si_kobj);
++out:
+ return err;
+}
+
+/*
-+ * returns,
-+ * plus: processed without an error
-+ * zero: unprocessed
++ * Accept a valid "si=xxxx" only.
++ * Once it is accepted successfully, accept "clean" too.
+ */
-+static int au_opt_simple(struct super_block *sb, struct au_opt *opt,
-+ struct au_opts *opts)
++static ssize_t au_procfs_plm_write(struct file *file, const char __user *ubuf,
++ size_t count, loff_t *ppos)
+{
-+ int err;
-+ struct au_sbinfo *sbinfo;
-+ struct au_opt_wbr_create *create;
-+
-+ AuTraceEnter();
-+
-+ err = 1; /* handled */
-+ sbinfo = au_sbi(sb);
-+ switch (opt->type) {
-+ case Opt_udba:
-+ sbinfo->si_mntflags &= ~AuOptMask_UDBA;
-+ sbinfo->si_mntflags |= opt->udba;
-+ opts->given_udba |= opt->udba;
-+ break;
-+
-+ case Opt_plink:
-+ au_opt_set(sbinfo->si_mntflags, PLINK);
-+ break;
-+ case Opt_noplink:
-+ if (au_opt_test(sbinfo->si_mntflags, PLINK))
-+ au_plink_put(sb);
-+ au_opt_clr(sbinfo->si_mntflags, PLINK);
-+ break;
-+ case Opt_list_plink:
-+ if (au_opt_test(sbinfo->si_mntflags, PLINK))
-+ au_plink_list(sb);
-+ break;
-+ case Opt_clean_plink:
-+ if (au_opt_test(sbinfo->si_mntflags, PLINK))
-+ au_plink_put(sb);
-+ break;
++ ssize_t err;
++ unsigned long id;
++ /* last newline is allowed */
++ char buf[3 + sizeof(unsigned long) * 2 + 1];
+
-+ case Opt_diropq_a:
-+ au_opt_set(sbinfo->si_mntflags, ALWAYS_DIROPQ);
-+ break;
-+ case Opt_diropq_w:
-+ au_opt_clr(sbinfo->si_mntflags, ALWAYS_DIROPQ);
-+ break;
++ err = -EACCES;
++ if (unlikely(!capable(CAP_SYS_ADMIN)))
++ goto out;
+
-+ case Opt_dlgt:
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
-+ break;
-+ case Opt_nodlgt:
-+ au_opt_clr(sbinfo->si_mntflags, DLGT);
-+ break;
++ err = -EINVAL;
++ if (unlikely(count > sizeof(buf)))
++ goto out;
+
-+ case Opt_warn_perm:
-+ au_opt_set(sbinfo->si_mntflags, WARN_PERM);
-+ break;
-+ case Opt_nowarn_perm:
-+ au_opt_clr(sbinfo->si_mntflags, WARN_PERM);
-+ break;
++ err = copy_from_user(buf, ubuf, count);
++ if (unlikely(err)) {
++ err = -EFAULT;
++ goto out;
++ }
++ buf[count] = 0;
+
-+ case Opt_refrof:
-+ au_opt_set(sbinfo->si_mntflags, REFROF);
-+ break;
-+ case Opt_norefrof:
-+ /* au_opt_set(sbinfo->si_mntflags, COO_LEAF); */
-+ au_opt_clr(sbinfo->si_mntflags, REFROF);
-+ break;
++ err = -EINVAL;
++ if (!strcmp("clean", buf)) {
++ au_procfs_plm_write_clean(file);
++ goto out_success;
++ } else if (unlikely(strncmp("si=", buf, 3)))
++ goto out;
+
-+ case Opt_verbose:
-+ au_opt_set(sbinfo->si_mntflags, VERBOSE);
-+ break;
-+ case Opt_noverbose:
-+ au_opt_clr(sbinfo->si_mntflags, VERBOSE);
-+ break;
++ err = kstrtoul(buf + 3, 16, &id);
++ if (unlikely(err))
++ goto out;
+
-+ case Opt_sum:
-+ au_opt_set(sbinfo->si_mntflags, SUM);
-+ break;
-+ case Opt_wsum:
-+ au_opt_clr(sbinfo->si_mntflags, SUM);
-+ au_opt_set(sbinfo->si_mntflags, SUM_W);
-+ case Opt_nosum:
-+ au_opt_clr(sbinfo->si_mntflags, SUM);
-+ au_opt_clr(sbinfo->si_mntflags, SUM_W);
-+ break;
++ err = au_procfs_plm_write_si(file, id);
++ if (unlikely(err))
++ goto out;
+
-+ case Opt_wbr_create:
-+ create = &opt->wbr_create;
-+ if (sbinfo->si_wbr_create_ops->fin) {
-+ err = sbinfo->si_wbr_create_ops->fin(sb);
-+ if (!err)
-+ err = 1;
-+ }
-+ sbinfo->si_wbr_create = create->wbr_create;
-+ sbinfo->si_wbr_create_ops
-+ = au_wbr_create_ops + create->wbr_create;
-+ switch (create->wbr_create) {
-+ case AuWbrCreate_MFSRRV:
-+ case AuWbrCreate_MFSRR:
-+ sbinfo->si_wbr_mfs.mfsrr_watermark
-+ = create->mfsrr_watermark;
-+ /*FALLTHROUGH*/
-+ case AuWbrCreate_MFS:
-+ case AuWbrCreate_MFSV:
-+ case AuWbrCreate_PMFS:
-+ case AuWbrCreate_PMFSV:
-+ sbinfo->si_wbr_mfs.mfs_expire = create->mfs_second * HZ;
-+ break;
-+ }
-+ if (sbinfo->si_wbr_create_ops->init)
-+ sbinfo->si_wbr_create_ops->init(sb); /* ignore */
-+ break;
-+ case Opt_wbr_copyup:
-+ sbinfo->si_wbr_copyup = opt->wbr_copyup;
-+ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + opt->wbr_copyup;
-+ break;
++out_success:
++ err = count; /* success */
++out:
++ return err;
++}
+
-+ case Opt_coo:
-+ sbinfo->si_mntflags &= ~AuOptMask_COO;
-+ sbinfo->si_mntflags |= opt->coo;
-+ break;
++static const struct proc_ops au_procfs_plm_op = {
++ .proc_write = au_procfs_plm_write,
++ .proc_release = au_procfs_plm_release
++};
+
-+ case Opt_dirwh:
-+ sbinfo->si_dirwh = opt->dirwh;
-+ break;
++/* ---------------------------------------------------------------------- */
+
-+ case Opt_rdcache:
-+ sbinfo->si_rdcache = opt->rdcache * HZ;
-+ break;
++static struct proc_dir_entry *au_procfs_dir;
+
-+ case Opt_trunc_xino:
-+ au_opt_set(sbinfo->si_mntflags, TRUNC_XINO);
-+ break;
-+ case Opt_notrunc_xino:
-+ au_opt_clr(sbinfo->si_mntflags, TRUNC_XINO);
-+ break;
++void au_procfs_fin(void)
++{
++ remove_proc_entry(AUFS_PLINK_MAINT_NAME, au_procfs_dir);
++ remove_proc_entry(AUFS_PLINK_MAINT_DIR, NULL);
++}
+
-+ case Opt_dirperm1:
-+ au_opt_set(sbinfo->si_mntflags, DIRPERM1);
-+ break;
-+ case Opt_nodirperm1:
-+ au_opt_clr(sbinfo->si_mntflags, DIRPERM1);
-+ break;
++int __init au_procfs_init(void)
++{
++ int err;
++ struct proc_dir_entry *entry;
+
-+ case Opt_shwh:
-+ au_opt_set(sbinfo->si_mntflags, SHWH);
-+ break;
-+ case Opt_noshwh:
-+ au_opt_clr(sbinfo->si_mntflags, SHWH);
-+ break;
++ err = -ENOMEM;
++ au_procfs_dir = proc_mkdir(AUFS_PLINK_MAINT_DIR, NULL);
++ if (unlikely(!au_procfs_dir))
++ goto out;
+
-+ case Opt_trunc_xino_path:
-+ case Opt_itrunc_xino:
-+ err = au_xino_trunc(sb, opt->xino_itrunc.bindex);
-+ if (!err)
-+ err = 1;
-+ break;
++ entry = proc_create(AUFS_PLINK_MAINT_NAME, S_IFREG | 0200,
++ au_procfs_dir, &au_procfs_plm_op);
++ if (unlikely(!entry))
++ goto out_dir;
+
-+ case Opt_trunc_xib:
-+ au_fset_opts(opts->flags, TRUNC_XIB);
-+ break;
-+ case Opt_notrunc_xib:
-+ au_fclr_opts(opts->flags, TRUNC_XIB);
-+ break;
++ err = 0;
++ goto out; /* success */
+
-+ default:
-+ err = 0;
-+ break;
-+ }
+
-+ AuTraceErr(err);
++out_dir:
++ remove_proc_entry(AUFS_PLINK_MAINT_DIR, NULL);
++out:
+ return err;
+}
+diff -urN /usr/share/empty/fs/aufs/rdu.c linux/fs/aufs/rdu.c
+--- /usr/share/empty/fs/aufs/rdu.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/rdu.c 2022-12-17 09:21:34.799855195 +0100
+@@ -0,0 +1,384 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
+/*
-+ * returns tri-state.
-+ * plus: processed without an error
-+ * zero: unprocessed
-+ * minus: error
++ * readdir in userspace.
+ */
-+static int au_opt_br(struct super_block *sb, struct au_opt *opt,
-+ struct au_opts *opts)
-+{
-+ int err, do_refresh;
+
-+ AuTraceEnter();
++#include <linux/compat.h>
++#include <linux/fs_stack.h>
++#include <linux/security.h>
++#include "aufs.h"
+
-+ err = 0;
-+ switch (opt->type) {
-+ case Opt_append:
-+ opt->add.bindex = au_sbend(sb) + 1;
-+ if (opt->add.bindex < 0)
-+ opt->add.bindex = 0;
-+ goto add;
-+ case Opt_prepend:
-+ opt->add.bindex = 0;
-+ add:
-+ case Opt_add:
-+ err = au_br_add(sb, &opt->add,
-+ au_ftest_opts(opts->flags, REMOUNT));
-+ if (!err) {
-+ err = 1;
-+ au_fset_opts(opts->flags, REFRESH_DIR);
-+ if (au_br_whable(opt->add.perm))
-+ au_fset_opts(opts->flags, REFRESH_NONDIR);
-+ }
-+ break;
++/* bits for struct aufs_rdu.flags */
++#define AuRdu_CALLED 1
++#define AuRdu_CONT (1 << 1)
++#define AuRdu_FULL (1 << 2)
++#define au_ftest_rdu(flags, name) ((flags) & AuRdu_##name)
++#define au_fset_rdu(flags, name) \
++ do { (flags) |= AuRdu_##name; } while (0)
++#define au_fclr_rdu(flags, name) \
++ do { (flags) &= ~AuRdu_##name; } while (0)
++
++struct au_rdu_arg {
++ struct dir_context ctx;
++ struct aufs_rdu *rdu;
++ union au_rdu_ent_ul ent;
++ unsigned long end;
++
++ struct super_block *sb;
++ int err;
++};
+
-+ case Opt_del:
-+ case Opt_idel:
-+ err = au_br_del(sb, &opt->del,
-+ au_ftest_opts(opts->flags, REMOUNT));
-+ if (!err) {
-+ err = 1;
-+ au_fset_opts(opts->flags, TRUNC_XIB);
-+ au_fset_opts(opts->flags, REFRESH_DIR);
-+ au_fset_opts(opts->flags, REFRESH_NONDIR);
-+ }
-+ break;
++static bool au_rdu_fill(struct dir_context *ctx, const char *name, int nlen,
++ loff_t offset, u64 h_ino, unsigned int d_type)
++{
++ int err, len;
++ struct au_rdu_arg *arg = container_of(ctx, struct au_rdu_arg, ctx);
++ struct aufs_rdu *rdu = arg->rdu;
++ struct au_rdu_ent ent;
+
-+ case Opt_mod:
-+ case Opt_imod:
-+ err = au_br_mod(sb, &opt->mod,
-+ au_ftest_opts(opts->flags, REMOUNT),
-+ &do_refresh);
-+ if (!err) {
-+ err = 1;
-+ if (do_refresh) {
-+ au_fset_opts(opts->flags, REFRESH_DIR);
-+ au_fset_opts(opts->flags, REFRESH_NONDIR);
-+ }
-+ }
-+ break;
++ err = 0;
++ arg->err = 0;
++ au_fset_rdu(rdu->cookie.flags, CALLED);
++ len = au_rdu_len(nlen);
++ if (arg->ent.ul + len < arg->end) {
++ ent.ino = h_ino;
++ ent.bindex = rdu->cookie.bindex;
++ ent.type = d_type;
++ ent.nlen = nlen;
++ if (unlikely(nlen > AUFS_MAX_NAMELEN))
++ ent.type = DT_UNKNOWN;
++
++ /* unnecessary to support mmap_sem since this is a dir */
++ err = -EFAULT;
++ if (copy_to_user(arg->ent.e, &ent, sizeof(ent)))
++ goto out;
++ if (copy_to_user(arg->ent.e->name, name, nlen))
++ goto out;
++ /* the terminating NULL */
++ if (__put_user(0, arg->ent.e->name + nlen))
++ goto out;
++ err = 0;
++ /* AuDbg("%p, %.*s\n", arg->ent.p, nlen, name); */
++ arg->ent.ul += len;
++ rdu->rent++;
++ } else {
++ err = -EFAULT;
++ au_fset_rdu(rdu->cookie.flags, FULL);
++ rdu->full = 1;
++ rdu->tail = arg->ent;
+ }
+
++out:
++ /* AuTraceErr(err); */
++ return !err;
++}
++
++static int au_rdu_do(struct file *h_file, struct au_rdu_arg *arg)
++{
++ int err;
++ loff_t offset;
++ struct au_rdu_cookie *cookie = &arg->rdu->cookie;
++
++ /* we don't have to care (FMODE_32BITHASH | FMODE_64BITHASH) for ext4 */
++ offset = vfsub_llseek(h_file, cookie->h_pos, SEEK_SET);
++ err = offset;
++ if (unlikely(offset != cookie->h_pos))
++ goto out;
++
++ err = 0;
++ do {
++ arg->err = 0;
++ au_fclr_rdu(cookie->flags, CALLED);
++ /* smp_mb(); */
++ err = vfsub_iterate_dir(h_file, &arg->ctx);
++ if (err >= 0)
++ err = arg->err;
++ } while (!err
++ && au_ftest_rdu(cookie->flags, CALLED)
++ && !au_ftest_rdu(cookie->flags, FULL));
++ cookie->h_pos = h_file->f_pos;
++
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+static int au_opt_xino(struct super_block *sb, struct au_opt *opt,
-+ struct au_opt_xino **opt_xino,
-+ struct au_opt_xinodir **opt_xinodir,
-+ struct au_opts *opts)
++static int au_rdu(struct file *file, struct aufs_rdu *rdu)
+{
+ int err;
-+ const int remount = !!au_ftest_opts(opts->flags, REMOUNT);
++ aufs_bindex_t bbot;
++ struct au_rdu_arg arg = {
++ .ctx = {
++ .actor = au_rdu_fill
++ }
++ };
++ struct dentry *dentry;
++ struct inode *inode;
++ struct file *h_file;
++ struct au_rdu_cookie *cookie = &rdu->cookie;
++
++ /* VERIFY_WRITE */
++ err = !access_ok(rdu->ent.e, rdu->sz);
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ goto out;
++ }
++ rdu->rent = 0;
++ rdu->tail = rdu->ent;
++ rdu->full = 0;
++ arg.rdu = rdu;
++ arg.ent = rdu->ent;
++ arg.end = arg.ent.ul;
++ arg.end += rdu->sz;
++
++ err = -ENOTDIR;
++ if (unlikely(!file->f_op->iterate && !file->f_op->iterate_shared))
++ goto out;
++
++ err = security_file_permission(file, MAY_READ);
++ AuTraceErr(err);
++ if (unlikely(err))
++ goto out;
++
++ dentry = file->f_path.dentry;
++ inode = d_inode(dentry);
++ inode_lock_shared(inode);
++
++ arg.sb = inode->i_sb;
++ err = si_read_lock(arg.sb, AuLock_FLUSH | AuLock_NOPLM);
++ if (unlikely(err))
++ goto out_mtx;
++ err = au_alive_dir(dentry);
++ if (unlikely(err))
++ goto out_si;
++ /* todo: reval? */
++ fi_read_lock(file);
+
-+ AuTraceEnter();
++ err = -EAGAIN;
++ if (unlikely(au_ftest_rdu(cookie->flags, CONT)
++ && cookie->generation != au_figen(file)))
++ goto out_unlock;
+
+ err = 0;
-+ switch (opt->type) {
-+ case Opt_xino:
-+ err = au_xino_set(sb, &opt->xino, remount);
-+ if (!err)
-+ *opt_xino = &opt->xino;
-+ break;
-+#if 0 /* def CONFIG_AUFS_EXPORT */ /* reserved for futur use */
-+ case Opt_xinodir:
-+ err = au_xinodir_set(sb, &opt->xinodir, remount);
-+ if (!err)
-+ *opt_xinodir = &opt->xinodir;
-+ break;
-+#endif
-+ case Opt_noxino:
-+ au_xino_clr(sb);
-+ *opt_xino = (void *)-1;
-+ break;
++ if (!rdu->blk) {
++ rdu->blk = au_sbi(arg.sb)->si_rdblk;
++ if (!rdu->blk)
++ rdu->blk = au_dir_size(file, /*dentry*/NULL);
++ }
++ bbot = au_fbtop(file);
++ if (cookie->bindex < bbot)
++ cookie->bindex = bbot;
++ bbot = au_fbbot_dir(file);
++ /* AuDbg("b%d, b%d\n", cookie->bindex, bbot); */
++ for (; !err && cookie->bindex <= bbot;
++ cookie->bindex++, cookie->h_pos = 0) {
++ h_file = au_hf_dir(file, cookie->bindex);
++ if (!h_file)
++ continue;
++
++ au_fclr_rdu(cookie->flags, FULL);
++ err = au_rdu_do(h_file, &arg);
++ AuTraceErr(err);
++ if (unlikely(au_ftest_rdu(cookie->flags, FULL) || err))
++ break;
+ }
++ AuDbg("rent %llu\n", rdu->rent);
++
++ if (!err && !au_ftest_rdu(cookie->flags, CONT)) {
++ rdu->shwh = !!au_opt_test(au_sbi(arg.sb)->si_mntflags, SHWH);
++ au_fset_rdu(cookie->flags, CONT);
++ cookie->generation = au_figen(file);
++ }
++
++ ii_read_lock_child(inode);
++ fsstack_copy_attr_atime(inode, au_h_iptr(inode, au_ibtop(inode)));
++ ii_read_unlock(inode);
+
++out_unlock:
++ fi_read_unlock(file);
++out_si:
++ si_read_unlock(arg.sb);
++out_mtx:
++ inode_unlock_shared(inode);
++out:
+ AuTraceErr(err);
+ return err;
+}
+
-+int verify_opts(struct super_block *sb, unsigned long sb_flags,
-+ unsigned int pending, int remount)
++static int au_rdu_ino(struct file *file, struct aufs_rdu *rdu)
+{
+ int err;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char do_plink, skip, do_free;
-+ struct au_branch *br;
-+ struct au_wbr *wbr;
-+ struct dentry *root;
-+ struct inode *dir, *h_dir;
-+ unsigned int mnt_flags;
-+
-+ AuTraceEnter();
-+ mnt_flags = au_mntflags(sb);
-+ AuDebugOn(!(mnt_flags & AuOptMask_COO));
-+ AuDebugOn(!(mnt_flags & AuOptMask_UDBA));
-+
-+ if (!(sb_flags & MS_RDONLY)) {
-+ if (unlikely(!au_br_writable(au_sbr_perm(sb, 0))))
-+ AuWarn("first branch should be rw\n");
-+ if (unlikely(au_opt_test(mnt_flags, SHWH)))
-+ AuWarn("shwh should be used with ro\n");
-+ } else if (unlikely(!au_opt_test(mnt_flags, COO_NONE))) {
-+ AuWarn("resetting coo for readonly mount\n");
-+ au_opt_set_coo(au_sbi(sb)->si_mntflags, COO_NONE);
-+ }
-+
-+ if (unlikely(au_opt_test((mnt_flags | pending), UDBA_INOTIFY)
-+ && !au_opt_test_xino(mnt_flags)))
-+ AuWarn("udba=inotify requires xino\n");
++ ino_t ino;
++ unsigned long long nent;
++ union au_rdu_ent_ul *u;
++ struct au_rdu_ent ent;
++ struct super_block *sb;
+
+ err = 0;
-+ root = sb->s_root;
-+ dir = sb->s_root->d_inode;
-+ do_plink = !!au_opt_test(mnt_flags, PLINK);
-+ bend = au_sbend(sb);
-+ for (bindex = 0; !err && bindex <= bend; bindex++) {
-+ skip = 0;
-+ h_dir = au_h_iptr(dir, bindex);
-+ br = au_sbr(sb, bindex);
-+ do_free = 0;
-+ wbr = br->br_wbr;
-+ if (wbr)
-+ wbr_wh_read_lock(wbr);
-+ switch (br->br_perm) {
-+ case AuBrPerm_RR:
-+ case AuBrPerm_RO:
-+ case AuBrPerm_RRWH:
-+ case AuBrPerm_ROWH:
-+ do_free = !!wbr;
-+ skip = (!wbr
-+ || (!wbr->wbr_whbase
-+ && !wbr->wbr_plink
-+ && !wbr->wbr_tmp));
-+ break;
-+
-+ case AuBrPerm_RWNoLinkWH:
-+ /* skip = (!br->br_whbase && !br->br_tmp); */
-+ skip = (!wbr || !wbr->wbr_whbase);
-+ if (skip && wbr) {
-+ if (do_plink)
-+ skip = !!wbr->wbr_plink;
-+ else
-+ skip = !wbr->wbr_plink;
-+ }
++ nent = rdu->nent;
++ u = &rdu->ent;
++ sb = file->f_path.dentry->d_sb;
++ si_read_lock(sb, AuLock_FLUSH);
++ while (nent-- > 0) {
++ /* unnecessary to support mmap_sem since this is a dir */
++ err = copy_from_user(&ent, u->e, sizeof(ent));
++ if (!err)
++ /* VERIFY_WRITE */
++ err = !access_ok(&u->e->ino, sizeof(ino));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
+ break;
++ }
+
-+ case AuBrPerm_RW:
-+ /* skip = (br->br_whbase && br->br_tmp); */
-+ skip = (wbr && wbr->wbr_whbase);
-+ if (skip) {
-+ if (do_plink)
-+ skip = !!wbr->wbr_plink;
-+ else
-+ skip = !wbr->wbr_plink;
-+ }
++ /* AuDbg("b%d, i%llu\n", ent.bindex, ent.ino); */
++ if (!ent.wh)
++ err = au_ino(sb, ent.bindex, ent.ino, ent.type, &ino);
++ else
++ err = au_wh_ino(sb, ent.bindex, ent.ino, ent.type,
++ &ino);
++ if (unlikely(err)) {
++ AuTraceErr(err);
+ break;
-+
-+ default:
-+ BUG();
+ }
-+ if (wbr)
-+ wbr_wh_read_unlock(wbr);
-+
-+ if (skip)
-+ continue;
-+
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ if (wbr)
-+ wbr_wh_write_lock(wbr);
-+ err = au_wh_init(au_h_dptr(root, bindex), br, br->br_mnt, sb,
-+ bindex);
-+ if (wbr)
-+ wbr_wh_write_unlock(wbr);
-+ mutex_unlock(&h_dir->i_mutex);
+
-+ if (!err && do_free) {
-+ kfree(wbr);
-+ br->br_wbr = NULL;
++ err = __put_user(ino, &u->e->ino);
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ break;
+ }
++ u->ul += au_rdu_len(ent.nlen);
+ }
++ si_read_unlock(sb);
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+int au_opts_mount(struct super_block *sb, struct au_opts *opts)
++/* ---------------------------------------------------------------------- */
++
++static int au_rdu_verify(struct aufs_rdu *rdu)
+{
-+ int err;
-+ struct inode *dir;
-+ struct au_opt *opt;
-+ struct au_opt_xino *opt_xino, xino;
-+ struct au_opt_xinodir *opt_xinodir;
-+ aufs_bindex_t bend;
-+ struct au_sbinfo *sbinfo;
-+ unsigned int tmp;
-+ struct au_branch *br;
++ AuDbg("rdu{%llu, %p, %u | %u | %llu, %u, %u | "
++ "%llu, b%d, 0x%x, g%u}\n",
++ rdu->sz, rdu->ent.e, rdu->verify[AufsCtlRduV_SZ],
++ rdu->blk,
++ rdu->rent, rdu->shwh, rdu->full,
++ rdu->cookie.h_pos, rdu->cookie.bindex, rdu->cookie.flags,
++ rdu->cookie.generation);
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+ DiMustWriteLock(sb->s_root);
-+ dir = sb->s_root->d_inode;
-+ IiMustWriteLock(dir);
++ if (rdu->verify[AufsCtlRduV_SZ] == sizeof(*rdu))
++ return 0;
+
-+ err = 0;
-+ opt_xino = NULL;
-+ opt_xinodir = NULL;
-+ opt = opts->opt;
-+ while (err >= 0 && opt->type != Opt_tail)
-+ err = au_opt_simple(sb, opt++, opts);
-+ if (err > 0)
-+ err = 0;
-+ else if (unlikely(err < 0))
-+ goto out;
++ AuDbg("%u:%u\n",
++ rdu->verify[AufsCtlRduV_SZ], (unsigned int)sizeof(*rdu));
++ return -EINVAL;
++}
+
-+ /* disable xino, xinodir, hinotify, dlgt temporary */
-+ sbinfo = au_sbi(sb);
-+ tmp = sbinfo->si_mntflags;
-+ au_opt_clr(sbinfo->si_mntflags, XINO);
-+ au_opt_clr(sbinfo->si_mntflags, XINODIR);
-+ au_opt_clr(sbinfo->si_mntflags, DLGT);
-+ au_opt_set_udba(sbinfo->si_mntflags, UDBA_REVAL);
++long au_rdu_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
++{
++ long err, e;
++ struct aufs_rdu rdu;
++ void __user *p = (void __user *)arg;
+
-+ opt = opts->opt;
-+ while (err >= 0 && opt->type != Opt_tail)
-+ err = au_opt_br(sb, opt++, opts);
-+ if (err > 0)
-+ err = 0;
-+ else if (unlikely(err < 0))
++ err = copy_from_user(&rdu, p, sizeof(rdu));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
+ goto out;
++ }
++ err = au_rdu_verify(&rdu);
++ if (unlikely(err))
++ goto out;
++
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ err = au_rdu(file, &rdu);
++ if (unlikely(err))
++ break;
++
++ e = copy_to_user(p, &rdu, sizeof(rdu));
++ if (unlikely(e)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ }
++ break;
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_ino(file, &rdu);
++ break;
+
-+ bend = au_sbend(sb);
-+ if (unlikely(bend < 0)) {
++ default:
++ /* err = -ENOTTY; */
+ err = -EINVAL;
-+ AuErr("no branches\n");
-+ goto out;
+ }
+
-+ if (au_opt_test(tmp, XINO))
-+ au_opt_set(sbinfo->si_mntflags, XINO);
-+ else if (au_opt_test(tmp, XINODIR))
-+ au_opt_set(sbinfo->si_mntflags, XINODIR);
-+ opt = opts->opt;
-+ while (!err && opt->type != Opt_tail)
-+ err = au_opt_xino(sb, opt++, &opt_xino, &opt_xinodir, opts);
-+ if (unlikely(err))
-+ goto out;
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++#ifdef CONFIG_COMPAT
++long au_rdu_compat_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
++{
++ long err, e;
++ struct aufs_rdu rdu;
++ void __user *p = compat_ptr(arg);
+
-+ /* todo: test this error case? */
-+ err = verify_opts(sb, sb->s_flags, tmp, /*remount*/0);
++ /* todo: get_user()? */
++ err = copy_from_user(&rdu, p, sizeof(rdu));
++ if (unlikely(err)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ goto out;
++ }
++ rdu.ent.e = compat_ptr(rdu.ent.ul);
++ err = au_rdu_verify(&rdu);
+ if (unlikely(err))
+ goto out;
+
-+ /* enable xino */
-+ if (au_opt_test(tmp, XINO) && !opt_xino) {
-+ xino.file = au_xino_def(sb);
-+ err = PTR_ERR(xino.file);
-+ if (IS_ERR(xino.file))
-+ goto out;
-+
-+ br = au_xino_def_br(sbinfo);
-+ err = au_xino_set(sb, &xino, /*remount*/0);
-+ fput(xino.file);
++ switch (cmd) {
++ case AUFS_CTL_RDU:
++ err = au_rdu(file, &rdu);
+ if (unlikely(err))
-+ goto out;
-+ au_xino_def_br_set(br, sbinfo);
-+ }
++ break;
+
-+ /* restore hinotify */
-+ sbinfo->si_mntflags &= ~AuOptMask_UDBA;
-+ sbinfo->si_mntflags |= (tmp & AuOptMask_UDBA);
-+ if (au_opt_test(tmp, UDBA_INOTIFY))
-+ au_reset_hinotify(dir, au_hi_flags(dir, 1) & ~AuHi_XINO);
++ rdu.ent.ul = ptr_to_compat(rdu.ent.e);
++ rdu.tail.ul = ptr_to_compat(rdu.tail.e);
++ e = copy_to_user(p, &rdu, sizeof(rdu));
++ if (unlikely(e)) {
++ err = -EFAULT;
++ AuTraceErr(err);
++ }
++ break;
++ case AUFS_CTL_RDU_INO:
++ err = au_rdu_ino(file, &rdu);
++ break;
+
-+ /* restore dlgt */
-+ if (au_opt_test(tmp, DLGT))
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
++ default:
++ /* err = -ENOTTY; */
++ err = -EINVAL;
++ }
+
-+ out:
++out:
+ AuTraceErr(err);
+ return err;
+}
++#endif
+diff -urN /usr/share/empty/fs/aufs/rwsem.h linux/fs/aufs/rwsem.h
+--- /usr/share/empty/fs/aufs/rwsem.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/rwsem.h 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,85 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+int au_opts_remount(struct super_block *sb, struct au_opts *opts)
-+{
-+ int err, rerr;
-+ struct inode *dir;
-+ struct au_opt_xino *opt_xino;
-+ struct au_opt_xinodir *opt_xinodir;
-+ struct au_opt *opt;
-+ unsigned char dlgt;
-+ struct au_sbinfo *sbinfo;
++/*
++ * simple read-write semaphore wrappers
++ */
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+ DiMustWriteLock(sb->s_root);
-+ dir = sb->s_root->d_inode;
-+ IiMustWriteLock(dir);
-+ sbinfo = au_sbi(sb);
++#ifndef __AUFS_RWSEM_H__
++#define __AUFS_RWSEM_H__
+
-+ err = 0;
-+ dlgt = !!au_opt_test(sbinfo->si_mntflags, DLGT);
-+ opt_xino = NULL;
-+ opt_xinodir = NULL;
-+ opt = opts->opt;
-+ while (err >= 0 && opt->type != Opt_tail) {
-+ err = au_opt_simple(sb, opt, opts);
++#ifdef __KERNEL__
+
-+ /* disable it temporary */
-+ dlgt = !!au_opt_test(sbinfo->si_mntflags, DLGT);
-+ au_opt_clr(sbinfo->si_mntflags, DLGT);
++#include "debug.h"
+
-+ if (!err)
-+ err = au_opt_br(sb, opt, opts);
-+ if (!err)
-+ err = au_opt_xino(sb, opt, &opt_xino, &opt_xinodir,
-+ opts);
++/* in the future, the name 'au_rwsem' will be totally gone */
++#define au_rwsem rw_semaphore
+
-+ /* restore it */
-+ if (dlgt)
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
-+ opt++;
-+ }
-+ if (err > 0)
-+ err = 0;
-+ AuTraceErr(err);
++/* to debug easier, do not make them inlined functions */
++#define AuRwMustNoWaiters(rw) AuDebugOn(rwsem_is_contended(rw))
++
++#ifdef CONFIG_LOCKDEP
++/* rwsem_is_locked() is unusable */
++#define AuRwMustReadLock(rw) AuDebugOn(IS_ENABLED(CONFIG_LOCKDEP) \
++ && !lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held_type(rw, 1))
++#define AuRwMustWriteLock(rw) AuDebugOn(IS_ENABLED(CONFIG_LOCKDEP) \
++ && !lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held_type(rw, 0))
++#define AuRwMustAnyLock(rw) AuDebugOn(IS_ENABLED(CONFIG_LOCKDEP) \
++ && !lockdep_recursing(current) \
++ && debug_locks \
++ && !lockdep_is_held(rw))
++#define AuRwDestroy(rw) AuDebugOn(IS_ENABLED(CONFIG_LOCKDEP) \
++ && !lockdep_recursing(current) \
++ && debug_locks \
++ && lockdep_is_held(rw))
++#else
++#define AuRwMustReadLock(rw) do {} while (0)
++#define AuRwMustWriteLock(rw) do {} while (0)
++#define AuRwMustAnyLock(rw) do {} while (0)
++#define AuRwDestroy(rw) do {} while (0)
++#endif
+
-+ /* go on even err */
++#define au_rw_init(rw) init_rwsem(rw)
+
-+ /* todo: test this error case? */
-+ au_opt_clr(sbinfo->si_mntflags, DLGT);
-+ rerr = verify_opts(sb, opts->sb_flags, sbinfo->si_mntflags,
-+ /*remount*/1);
-+ if (unlikely(dlgt))
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
-+ if (unlikely(rerr && !err))
-+ err = rerr;
++#define au_rw_init_wlock(rw) do { \
++ au_rw_init(rw); \
++ down_write(rw); \
++ } while (0)
+
-+ if (au_ftest_opts(opts->flags, TRUNC_XIB)) {
-+ rerr = au_xib_trunc(sb);
-+ if (unlikely(rerr && !err))
-+ err = rerr;
-+ }
++#define au_rw_init_wlock_nested(rw, lsc) do { \
++ au_rw_init(rw); \
++ down_write_nested(rw, lsc); \
++ } while (0)
+
-+ /* they are handled by the caller */
-+ if (!au_ftest_opts(opts->flags, REFRESH_DIR)
-+ && (opts->given_udba || au_opt_test_xino(sbinfo->si_mntflags)))
-+ au_fset_opts(opts->flags, REFRESH_DIR);
++#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)
+
-+ LKTRTrace("status 0x%x\n", opts->flags);
-+ AuTraceErr(err);
-+ return err;
-+}
-diff -urN linux/fs/aufs/opts.h linux-aufs/fs/aufs/opts.h
---- linux/fs/aufs/opts.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/opts.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,255 @@
++#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 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,316 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * mount options/flags
-+ *
-+ * $Id$
++ * superblock private data
+ */
+
-+#ifndef __AUFS_OPTS_H__
-+#define __AUFS_OPTS_H__
++#include <linux/iversion.h>
++#include "aufs.h"
+
-+#ifdef __KERNEL__
++/*
++ * they are necessary regardless sysfs is disabled.
++ */
++void au_si_free(struct kobject *kobj)
++{
++ int i;
++ struct au_sbinfo *sbinfo;
++ char *locked __maybe_unused; /* debug only */
+
-+#include <linux/fs.h>
-+#include <linux/namei.h>
-+#include <linux/aufs_type.h>
-+#include "wkq.h"
++ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
++ for (i = 0; i < AuPlink_NHASH; i++)
++ AuDebugOn(!hlist_bl_empty(sbinfo->si_plink + i));
++ AuDebugOn(atomic_read(&sbinfo->si_nowait.nw_len));
+
-+/* ---------------------------------------------------------------------- */
-+/* mount flags */
++ AuLCntZero(au_lcnt_read(&sbinfo->si_ninodes, /*do_rev*/0));
++ au_lcnt_fin(&sbinfo->si_ninodes, /*do_sync*/0);
++ AuLCntZero(au_lcnt_read(&sbinfo->si_nfiles, /*do_rev*/0));
++ au_lcnt_fin(&sbinfo->si_nfiles, /*do_sync*/0);
+
-+/* external inode number bitmap and translation table */
-+#define AuOpt_XINO 1
-+#define AuOpt_XINODIR (1 << 1)
-+#define AuOpt_TRUNC_XINO (1 << 2)
-+#define AuOpt_UDBA_NONE (1 << 3) /* users direct branch access */
-+#define AuOpt_UDBA_REVAL (1 << 4)
-+#define AuOpt_UDBA_INOTIFY (1 << 5)
-+#define AuOpt_SHWH (1 << 6)
-+#define AuOpt_PLINK (1 << 7)
-+#define AuOpt_WARN_PERM (1 << 8)
-+#define AuOpt_DIRPERM1 (1 << 9)
-+#define AuOpt_DLGT (1 << 10)
-+#define AuOpt_COO_NONE (1 << 11) /* copyup on open */
-+#define AuOpt_COO_LEAF (1 << 12)
-+#define AuOpt_COO_ALL (1 << 13)
-+#define AuOpt_ALWAYS_DIROPQ (1 << 14)
-+#define AuOpt_REFROF (1 << 15)
-+#define AuOpt_VERBOSE (1 << 16)
-+#define AuOpt_SUM (1 << 17)
-+#define AuOpt_SUM_W (1 << 18) /* unimplemented */
-+
-+#if 1 /* ndef CONFIG_AUFS_EXPORT */ /* reserved for future use */
-+#undef AuOpt_XINODIR
-+#define AuOpt_XINODIR 0
-+#endif
-+#ifndef CONFIG_AUFS_HINOTIFY
-+#undef AuOpt_UDBA_INOTIFY
-+#define AuOpt_UDBA_INOTIFY 0
-+#endif
-+#ifndef CONFIG_AUFS_SHWH
-+#undef AuOpt_SHWH
-+#define AuOpt_SHWH 0
-+#endif
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuOpt_DIRPERM1
-+#define AuOpt_DIRPERM1 0
-+#undef AuOpt_DLGT
-+#define AuOpt_DLGT 0
-+#endif
++ dbgaufs_si_fin(sbinfo);
++ au_rw_write_lock(&sbinfo->si_rwsem);
++ au_br_free(sbinfo);
++ au_rw_write_unlock(&sbinfo->si_rwsem);
+
-+/* policies to select one among multiple writable branches */
-+enum {
-+ AuWbrCreate_TDP, /* top down parent */
-+ AuWbrCreate_RR, /* round robin */
-+ AuWbrCreate_MFS, /* most free space */
-+ AuWbrCreate_MFSV, /* mfs with seconds */
-+ AuWbrCreate_MFSRR, /* mfs then rr */
-+ AuWbrCreate_MFSRRV, /* mfs then rr with seconds */
-+ AuWbrCreate_PMFS, /* parent and mfs */
-+ AuWbrCreate_PMFSV, /* parent and mfs with seconds */
++ au_kfree_try_rcu(sbinfo->si_branch);
++ mutex_destroy(&sbinfo->si_xib_mtx);
++ AuRwDestroy(&sbinfo->si_rwsem);
+
-+ AuWbrCreate_Def = AuWbrCreate_TDP
-+};
++ au_lcnt_wait_for_fin(&sbinfo->si_ninodes);
++ /* si_nfiles is waited too */
++ au_kfree_rcu(sbinfo);
++}
+
-+enum {
-+ AuWbrCopyup_TDP, /* top down parent */
-+ AuWbrCopyup_BUP, /* bottom up parent */
-+ AuWbrCopyup_BU, /* bottom up */
++struct au_sbinfo *au_si_alloc(struct super_block *sb)
++{
++ struct au_sbinfo *sbinfo;
++ int err, i;
+
-+ AuWbrCopyup_Def = AuWbrCopyup_TDP
-+};
++ err = -ENOMEM;
++ sbinfo = kzalloc(sizeof(*sbinfo), GFP_NOFS);
++ if (unlikely(!sbinfo))
++ goto out;
+
-+#define AuOptMask_COO (AuOpt_COO_NONE \
-+ | AuOpt_COO_LEAF \
-+ | AuOpt_COO_ALL)
-+#define AuOptMask_UDBA (AuOpt_UDBA_NONE \
-+ | AuOpt_UDBA_REVAL \
-+ | AuOpt_UDBA_INOTIFY)
++ /* will be reallocated separately */
++ sbinfo->si_branch = kzalloc(sizeof(*sbinfo->si_branch), GFP_NOFS);
++ if (unlikely(!sbinfo->si_branch))
++ goto out_sbinfo;
+
-+#ifdef CONFIG_AUFS_COMPAT
-+#define AuOpt_DefExtra1 AuOpt_ALWAYS_DIROPQ
-+#else
-+#define AuOpt_DefExtra1 0
-+#endif
++ err = sysaufs_si_init(sbinfo);
++ if (!err) {
++ dbgaufs_si_null(sbinfo);
++ err = dbgaufs_si_init(sbinfo);
++ if (unlikely(err))
++ kobject_put(&sbinfo->si_kobj);
++ }
++ if (unlikely(err))
++ goto out_br;
+
-+#define AuOpt_Def (AuOpt_XINO \
-+ | AuOpt_UDBA_REVAL \
-+ | AuOpt_WARN_PERM \
-+ | AuOpt_COO_NONE \
-+ | AuOpt_PLINK \
-+ | AuOpt_DefExtra1)
++ au_nwt_init(&sbinfo->si_nowait);
++ au_rw_init_wlock(&sbinfo->si_rwsem);
+
-+/* ---------------------------------------------------------------------- */
++ au_lcnt_init(&sbinfo->si_ninodes, /*release*/NULL);
++ au_lcnt_init(&sbinfo->si_nfiles, /*release*/NULL);
+
-+struct au_opt_add {
-+ aufs_bindex_t bindex;
-+ char *path;
-+ int perm;
-+ struct nameidata nd;
-+};
++ sbinfo->si_bbot = -1;
++ sbinfo->si_last_br_id = AUFS_BRANCH_MAX / 2;
+
-+struct au_opt_del {
-+ char *path;
-+ struct dentry *h_root;
-+};
++ sbinfo->si_wbr_copyup = AuWbrCopyup_Def;
++ sbinfo->si_wbr_create = AuWbrCreate_Def;
++ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + sbinfo->si_wbr_copyup;
++ sbinfo->si_wbr_create_ops = au_wbr_create_ops + sbinfo->si_wbr_create;
+
-+struct au_opt_mod {
-+ char *path;
-+ int perm;
-+ struct dentry *h_root;
-+};
++ au_fhsm_init(sbinfo);
+
-+struct au_opt_xino {
-+ char *path;
-+ struct file *file;
-+};
++ sbinfo->si_mntflags = au_opts_plink(AuOpt_Def);
+
-+struct au_opt_xinodir {
-+ char *name;
-+ struct path path;
-+};
++ sbinfo->si_xino_jiffy = jiffies;
++ sbinfo->si_xino_expire
++ = msecs_to_jiffies(AUFS_XINO_DEF_SEC * MSEC_PER_SEC);
++ mutex_init(&sbinfo->si_xib_mtx);
++ /* leave si_xib_last_pindex and si_xib_next_bit */
+
-+struct au_opt_xino_itrunc {
-+ aufs_bindex_t bindex;
-+};
++ INIT_HLIST_BL_HEAD(&sbinfo->si_aopen);
+
-+struct au_opt_xino_trunc_v {
-+ unsigned long long upper;
-+ int step;
-+};
++ sbinfo->si_rdcache = msecs_to_jiffies(AUFS_RDCACHE_DEF * MSEC_PER_SEC);
++ sbinfo->si_rdblk = AUFS_RDBLK_DEF;
++ sbinfo->si_rdhash = AUFS_RDHASH_DEF;
++ sbinfo->si_dirwh = AUFS_DIRWH_DEF;
+
-+struct au_opt_wbr_create {
-+ int wbr_create;
-+ int mfs_second;
-+ unsigned long long mfsrr_watermark;
-+};
++ for (i = 0; i < AuPlink_NHASH; i++)
++ INIT_HLIST_BL_HEAD(sbinfo->si_plink + i);
++ init_waitqueue_head(&sbinfo->si_plink_wq);
++ spin_lock_init(&sbinfo->si_plink_maint_lock);
+
-+struct au_opt {
-+ int type;
-+ union {
-+ struct au_opt_xino xino;
-+ struct au_opt_xinodir xinodir;
-+ struct au_opt_xino_itrunc xino_itrunc;
-+ struct au_opt_add add;
-+ struct au_opt_del del;
-+ struct au_opt_mod mod;
-+ int dirwh;
-+ int rdcache;
-+ int deblk;
-+ int nhash;
-+ int udba;
-+ int coo;
-+ struct au_opt_wbr_create wbr_create;
-+ int wbr_copyup;
-+ };
-+};
++ INIT_HLIST_BL_HEAD(&sbinfo->si_files);
+
-+/* opts flags */
-+#define AuOpts_REMOUNT 1
-+#define AuOpts_REFRESH_DIR (1 << 1)
-+#define AuOpts_REFRESH_NONDIR (1 << 2)
-+#define AuOpts_TRUNC_XIB (1 << 3)
-+#define au_ftest_opts(flags, name) ((flags) & AuOpts_##name)
-+#define au_fset_opts(flags, name) { (flags) |= AuOpts_##name; }
-+#define au_fclr_opts(flags, name) { (flags) &= ~AuOpts_##name; }
++ /* with getattr by default */
++ sbinfo->si_iop_array = aufs_iop;
+
-+struct au_opts {
-+ struct au_opt *opt;
-+ int max_opt;
++ /* leave other members for sysaufs and si_mnt. */
++ sbinfo->si_sb = sb;
++ if (sb) {
++ sb->s_fs_info = sbinfo;
++ si_pid_set(sb);
++ }
++ return sbinfo; /* success */
+
-+ unsigned int given_udba;
-+ unsigned int flags;
-+ unsigned long sb_flags;
-+};
++out_br:
++ au_kfree_try_rcu(sbinfo->si_branch);
++out_sbinfo:
++ au_kfree_rcu(sbinfo);
++out:
++ return ERR_PTR(err);
++}
++
++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr, int may_shrink)
++{
++ int err, sz;
++ struct au_branch **brp;
++
++ AuRwMustWriteLock(&sbinfo->si_rwsem);
++
++ err = -ENOMEM;
++ 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,
++ may_shrink);
++ if (brp) {
++ sbinfo->si_branch = brp;
++ err = 0;
++ }
++
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+const char *au_optstr_br_perm(int brperm);
-+const char *au_optstr_udba(int udba);
-+const char *au_optstr_coo(int coo);
-+const char *au_optstr_wbr_copyup(int wbr_copyup);
-+const char *au_optstr_wbr_create(int wbr_create);
++unsigned int au_sigen_inc(struct super_block *sb)
++{
++ unsigned int gen;
++ struct inode *inode;
+
-+void au_opts_free(struct au_opts *opts);
-+int au_opts_parse(struct super_block *sb, char *str, struct au_opts *opts);
-+int verify_opts(struct super_block *sb, unsigned long sb_flags,
-+ unsigned int pending, int remount);
-+int au_opts_mount(struct super_block *sb, struct au_opts *opts);
-+int au_opts_remount(struct super_block *sb, struct au_opts *opts);
++ SiMustWriteLock(sb);
++
++ gen = ++au_sbi(sb)->si_generation;
++ au_update_digen(sb->s_root);
++ inode = d_inode(sb->s_root);
++ au_update_iigen(inode, /*half*/0);
++ inode_inc_iversion(inode);
++ return gen;
++}
++
++aufs_bindex_t au_new_br_id(struct super_block *sb)
++{
++ aufs_bindex_t br_id;
++ int i;
++ struct au_sbinfo *sbinfo;
++
++ SiMustWriteLock(sb);
++
++ sbinfo = au_sbi(sb);
++ for (i = 0; i <= AUFS_BRANCH_MAX; i++) {
++ br_id = ++sbinfo->si_last_br_id;
++ AuDebugOn(br_id < 0);
++ if (br_id && au_br_index(sb, br_id) < 0)
++ return br_id;
++ }
++
++ return -1;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+#define au_opt_test(flags, name) (flags & AuOpt_##name)
++/* it is ok that new 'nwt' tasks are appended while we are sleeping */
++int si_read_lock(struct super_block *sb, int flags)
++{
++ int err;
++
++ err = 0;
++ if (au_ftest_lock(flags, FLUSH))
++ au_nwt_flush(&au_sbi(sb)->si_nowait);
+
-+static inline int au_opt_test_xino(unsigned int flags)
++ si_noflush_read_lock(sb);
++ err = au_plink_maint(sb, flags);
++ if (unlikely(err))
++ si_read_unlock(sb);
++
++ return err;
++}
++
++int si_write_lock(struct super_block *sb, int flags)
+{
-+ return flags & (AuOpt_XINO | AuOpt_XINODIR);
++ int err;
++
++ if (au_ftest_lock(flags, FLUSH))
++ au_nwt_flush(&au_sbi(sb)->si_nowait);
++
++ si_noflush_write_lock(sb);
++ err = au_plink_maint(sb, flags);
++ if (unlikely(err))
++ si_write_unlock(sb);
++
++ return err;
+}
+
-+#define au_opt_set(flags, name) do { \
-+ BUILD_BUG_ON(AuOpt_##name & (AuOptMask_COO | AuOptMask_UDBA)); \
-+ ((flags) |= AuOpt_##name); \
-+} while (0)
++/* dentry and super_block lock. call at entry point */
++int aufs_read_lock(struct dentry *dentry, int flags)
++{
++ int err;
++ struct super_block *sb;
+
-+#define au_opt_set_coo(flags, name) do { \
-+ (flags) &= ~AuOptMask_COO; \
-+ ((flags) |= AuOpt_##name); \
-+} while (0)
++ sb = dentry->d_sb;
++ err = si_read_lock(sb, flags);
++ if (unlikely(err))
++ goto out;
++
++ if (au_ftest_lock(flags, DW))
++ di_write_lock_child(dentry);
++ else
++ di_read_lock_child(dentry, flags);
++
++ if (au_ftest_lock(flags, GEN)) {
++ err = au_digen_test(dentry, au_sigen(sb));
++ if (!au_opt_test(au_mntflags(sb), UDBA_NONE))
++ AuDebugOn(!err && au_dbrange_test(dentry));
++ else if (!err)
++ err = au_dbrange_test(dentry);
++ if (unlikely(err))
++ aufs_read_unlock(dentry, flags);
++ }
++
++out:
++ return err;
++}
++
++void aufs_read_unlock(struct dentry *dentry, int flags)
++{
++ if (au_ftest_lock(flags, DW))
++ di_write_unlock(dentry);
++ else
++ di_read_unlock(dentry, flags);
++ si_read_unlock(dentry->d_sb);
++}
++
++void aufs_write_lock(struct dentry *dentry)
++{
++ si_write_lock(dentry->d_sb, AuLock_FLUSH | AuLock_NOPLMW);
++ di_write_lock_child(dentry);
++}
++
++void aufs_write_unlock(struct dentry *dentry)
++{
++ di_write_unlock(dentry);
++ si_write_unlock(dentry->d_sb);
++}
++
++int aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags)
++{
++ int err;
++ unsigned int sigen;
++ struct super_block *sb;
+
-+#define au_opt_set_udba(flags, name) do { \
-+ (flags) &= ~AuOptMask_UDBA; \
-+ ((flags) |= AuOpt_##name); \
-+} while (0)
++ sb = d1->d_sb;
++ err = si_read_lock(sb, flags);
++ if (unlikely(err))
++ goto out;
+
-+#define au_opt_clr(flags, name) { ((flags) &= ~AuOpt_##name); }
++ di_write_lock2_child(d1, d2, au_ftest_lock(flags, DIRS));
+
-+static inline int au_test_dlgt(unsigned int flags)
-+{
-+ return au_opt_test(flags, DLGT) && !au_test_wkq(current);
++ if (au_ftest_lock(flags, GEN)) {
++ sigen = au_sigen(sb);
++ err = au_digen_test(d1, sigen);
++ AuDebugOn(!err && au_dbrange_test(d1));
++ if (!err) {
++ err = au_digen_test(d2, sigen);
++ AuDebugOn(!err && au_dbrange_test(d2));
++ }
++ if (unlikely(err))
++ aufs_read_and_write_unlock2(d1, d2);
++ }
++
++out:
++ return err;
+}
+
-+static inline int au_test_dirperm1(unsigned int flags)
++void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2)
+{
-+ return au_opt_test(flags, DIRPERM1) && !au_test_wkq(current);
++ di_write_unlock2(d1, d2);
++ si_read_unlock(d1->d_sb);
+}
-+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_OPTS_H__ */
-diff -urN linux/fs/aufs/plink.c linux-aufs/fs/aufs/plink.c
---- linux/fs/aufs/plink.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/plink.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,364 @@
+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 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,871 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * pseudo-link
-+ *
-+ * $Id$
++ * mount and super_block operations
+ */
+
++#include <linux/iversion.h>
++#include <linux/mm.h>
++#include <linux/seq_file.h>
++#include <linux/statfs.h>
++#include <linux/vmalloc.h>
+#include "aufs.h"
+
-+struct pseudo_link {
-+ struct list_head list;
-+ struct inode *inode;
-+};
-+
-+#ifdef CONFIG_AUFS_DEBUG
-+void au_plink_list(struct super_block *sb)
-+{
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *plink_list;
-+ struct pseudo_link *plink;
-+
-+ AuTraceEnter();
-+ SiMustAnyLock(sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
-+
-+ plink_list = &sbinfo->si_plink.head;
-+ spin_lock(&sbinfo->si_plink.spin);
-+ list_for_each_entry(plink, plink_list, list)
-+ AuDbg("%lu\n", plink->inode->i_ino);
-+ spin_unlock(&sbinfo->si_plink.spin);
-+}
-+#endif
-+
-+int au_plink_test(struct super_block *sb, struct inode *inode)
++/*
++ * super_operations
++ */
++static struct inode *aufs_alloc_inode(struct super_block *sb __maybe_unused)
+{
-+ int found;
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *plink_list;
-+ struct pseudo_link *plink;
-+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ SiMustAnyLock(sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ struct au_icntnr *c;
+
-+ found = 0;
-+ plink_list = &sbinfo->si_plink.head;
-+ spin_lock(&sbinfo->si_plink.spin);
-+ list_for_each_entry(plink, plink_list, list)
-+ if (plink->inode == inode) {
-+ found = 1;
-+ break;
-+ }
-+ spin_unlock(&sbinfo->si_plink.spin);
-+ return found;
++ c = au_cache_alloc_icntnr(sb);
++ if (c) {
++ au_icntnr_init(c);
++ inode_set_iversion(&c->vfs_inode, 1); /* sigen(sb); */
++ c->iinfo.ii_hinode = NULL;
++ return &c->vfs_inode;
++ }
++ return NULL;
+}
+
-+/* 20 is max digits length of ulong 64 */
-+#define PLINK_NAME_LEN ((20 + 1) * 2)
-+
-+static int plink_name(char *name, int len, struct inode *inode,
-+ aufs_bindex_t bindex)
++static void aufs_destroy_inode(struct inode *inode)
+{
-+ int rlen;
-+ struct inode *h_inode;
-+
-+ LKTRTrace("i%lu, b%d\n", inode->i_ino, bindex);
-+ AuDebugOn(len != PLINK_NAME_LEN);
-+ h_inode = au_h_iptr(inode, bindex);
-+ AuDebugOn(!h_inode);
-+ rlen = snprintf(name, len, "%lu.%lu", inode->i_ino, h_inode->i_ino);
-+ AuDebugOn(rlen >= len);
-+ return rlen;
++ if (!au_is_bad_inode(inode))
++ au_iinfo_fin(inode);
+}
+
-+struct dentry *au_plink_lkup(struct super_block *sb, aufs_bindex_t bindex,
-+ struct inode *inode)
++static void aufs_free_inode(struct inode *inode)
+{
-+ struct dentry *h_dentry, *h_parent;
-+ struct au_branch *br;
-+ struct au_wbr *wbr;
-+ struct inode *h_dir;
-+ char tgtname[PLINK_NAME_LEN];
-+ int len;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+
-+ LKTRTrace("b%d, i%lu\n", bindex, inode->i_ino);
-+ br = au_sbr(sb, bindex);
-+ wbr = br->br_wbr;
-+ AuDebugOn(!wbr);
-+ h_parent = wbr->wbr_plink;
-+ AuDebugOn(!h_parent);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!h_dir);
-+
-+ len = plink_name(tgtname, sizeof(tgtname), inode, bindex);
-+
-+ /* always superio. */
-+ ndx.nfsmnt = au_do_nfsmnt(br->br_mnt);
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_CHILD2);
-+ h_dentry = au_sio_lkup_one(tgtname, h_parent, len, &ndx);
-+ mutex_unlock(&h_dir->i_mutex);
-+ return h_dentry;
++ au_cache_free_icntnr(container_of(inode, struct au_icntnr, vfs_inode));
+}
+
-+static int do_whplink(char *tgt, int len, struct dentry *h_parent,
-+ struct dentry *h_dentry, struct vfsmount *nfsmnt,
-+ struct super_block *sb)
++struct inode *au_iget_locked(struct super_block *sb, ino_t ino)
+{
-+ int err, dlgt;
-+ struct dentry *h_tgt;
-+ struct inode *h_dir;
-+ struct vfsub_args vargs;
-+ struct au_ndx ndx = {
-+ .nfsmnt = nfsmnt,
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+
-+ AuTraceEnter();
++ struct inode *inode;
++ int err;
+
-+ dlgt = !!au_test_dlgt(au_mntflags(sb));
-+ if (dlgt)
-+ au_fset_ndx(ndx.flags, DLGT);
-+ h_tgt = au_lkup_one(tgt, h_parent, len, &ndx);
-+ err = PTR_ERR(h_tgt);
-+ if (IS_ERR(h_tgt))
++ inode = iget_locked(sb, ino);
++ if (unlikely(!inode)) {
++ inode = ERR_PTR(-ENOMEM);
+ goto out;
-+
-+ err = 0;
-+ vfsub_args_init(&vargs, NULL, dlgt, 0);
-+ /* wh.plink dir is not monitored */
-+ h_dir = h_parent->d_inode;
-+ if (h_tgt->d_inode && h_tgt->d_inode != h_dentry->d_inode)
-+ err = vfsub_unlink(h_dir, h_tgt, &vargs);
-+ if (!err && !h_tgt->d_inode) {
-+ err = vfsub_link(h_dentry, h_dir, h_tgt, &vargs);
-+ /* todo: unnecessary? */
-+ /* inc_nlink(inode); */
+ }
-+ dput(h_tgt);
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++ if (!(inode->i_state & I_NEW))
++ goto out;
+
-+struct do_whplink_args {
-+ int *errp;
-+ char *tgt;
-+ int len;
-+ struct dentry *h_parent;
-+ struct dentry *h_dentry;
-+ struct vfsmount *nfsmnt;
-+ struct super_block *sb;
-+};
++ err = au_xigen_new(inode);
++ if (!err)
++ err = au_iinfo_init(inode);
++ if (!err)
++ inode_inc_iversion(inode);
++ else {
++ iget_failed(inode);
++ inode = ERR_PTR(err);
++ }
+
-+static void call_do_whplink(void *args)
-+{
-+ struct do_whplink_args *a = args;
-+ *a->errp = do_whplink(a->tgt, a->len, a->h_parent, a->h_dentry,
-+ a->nfsmnt, a->sb);
++out:
++ /* never return NULL */
++ AuDebugOn(!inode);
++ AuTraceErrPtr(inode);
++ return inode;
+}
+
-+static int whplink(struct dentry *h_dentry, struct inode *inode,
-+ aufs_bindex_t bindex, struct super_block *sb)
++/* lock free root dinfo */
++static int au_show_brs(struct seq_file *seq, struct super_block *sb)
+{
-+ int err, len, wkq_err;
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct path path;
++ struct au_hdentry *hdp;
+ struct au_branch *br;
-+ struct au_wbr *wbr;
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
-+ char tgtname[PLINK_NAME_LEN];
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
-+ br = au_sbr(inode->i_sb, bindex);
-+ wbr = br->br_wbr;
-+ AuDebugOn(!wbr);
-+ h_parent = wbr->wbr_plink;
-+ AuDebugOn(!h_parent);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!h_dir);
++ au_br_perm_str_t perm;
+
-+ len = plink_name(tgtname, sizeof(tgtname), inode, bindex);
-+
-+ /* always superio. */
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_CHILD2);
-+ if (!au_test_wkq(current)) {
-+ struct do_whplink_args args = {
-+ .errp = &err,
-+ .tgt = tgtname,
-+ .len = len,
-+ .h_parent = h_parent,
-+ .h_dentry = h_dentry,
-+ .nfsmnt = au_do_nfsmnt(br->br_mnt),
-+ .sb = sb
-+ };
-+ wkq_err = au_wkq_wait(call_do_whplink, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ } else
-+ err = do_whplink(tgtname, len, h_parent, h_dentry,
-+ au_do_nfsmnt(br->br_mnt), sb);
-+ mutex_unlock(&h_dir->i_mutex);
++ err = 0;
++ bbot = au_sbbot(sb);
++ bindex = 0;
++ hdp = au_hdentry(au_di(sb->s_root), bindex);
++ for (; !err && bindex <= bbot; bindex++, hdp++) {
++ br = au_sbr(sb, bindex);
++ path.mnt = au_br_mnt(br);
++ path.dentry = hdp->hd_dentry;
++ err = au_seq_path(seq, &path);
++ if (!err) {
++ au_optstr_br_perm(&perm, br->br_perm);
++ seq_printf(seq, "=%s", perm.a);
++ if (bindex != bbot)
++ seq_putc(seq, ':');
++ }
++ }
++ if (unlikely(err || seq_has_overflowed(seq)))
++ err = -E2BIG;
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+static void do_put_plink(struct pseudo_link *plink, int do_del)
++static void au_gen_fmt(char *fmt, int len __maybe_unused, const char *pat,
++ const char *append)
+{
-+ AuTraceEnter();
++ char *p;
+
-+ iput(plink->inode);
-+ if (do_del)
-+ list_del(&plink->list);
-+ kfree(plink);
++ p = fmt;
++ while (*pat != ':')
++ *p++ = *pat++;
++ *p++ = *pat++;
++ strcpy(p, append);
++ AuDebugOn(strlen(fmt) >= len);
+}
+
-+void au_plink_append(struct super_block *sb, struct inode *inode,
-+ struct dentry *h_dentry, aufs_bindex_t bindex)
++static void au_show_wbr_create(struct seq_file *m, int v,
++ struct au_sbinfo *sbinfo)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *plink_list;
-+ struct pseudo_link *plink;
-+ int found, err, cnt;
++ const char *pat;
++ char fmt[32];
++ struct au_wbr_mfs *mfs;
+
-+ LKTRTrace("i%lu\n", inode->i_ino);
-+ SiMustAnyLock(sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ AuRwMustAnyLock(&sbinfo->si_rwsem);
+
-+ cnt = 0;
-+ found = 0;
-+ plink_list = &sbinfo->si_plink.head;
-+ spin_lock(&sbinfo->si_plink.spin);
-+ list_for_each_entry(plink, plink_list, list) {
-+ cnt++;
-+ if (plink->inode == inode) {
-+ found = 1;
-+ break;
-+ }
++ 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_MFS:
++ case AuWbrCreate_PMFS:
++ seq_puts(m, pat);
++ break;
++ 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:
++ au_gen_fmt(fmt, sizeof(fmt), pat, "%lu");
++ seq_printf(m, fmt,
++ jiffies_to_msecs(mfs->mfs_expire)
++ / MSEC_PER_SEC);
++ break;
++ case AuWbrCreate_MFSRRV:
++ case AuWbrCreate_TDMFSV:
++ case AuWbrCreate_PMFSRRV:
++ 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();
+ }
++}
+
-+ err = 0;
-+ plink = NULL;
-+ if (!found) {
-+ plink = kmalloc(sizeof(*plink), GFP_ATOMIC);
-+ if (plink) {
-+ plink->inode = au_igrab(inode);
-+ list_add(&plink->list, plink_list);
-+ cnt++;
-+ } else
-+ err = -ENOMEM;
-+ }
-+ spin_unlock(&sbinfo->si_plink.spin);
++static int au_show_xino(struct seq_file *seq, struct super_block *sb)
++{
++#ifdef CONFIG_SYSFS
++ return 0;
++#else
++ int err;
++ const int len = sizeof(AUFS_XINO_FNAME) - 1;
++ aufs_bindex_t bindex, brid;
++ struct qstr *name;
++ struct file *f;
++ struct dentry *d, *h_root;
++ struct au_branch *br;
+
-+#if 0 /* todo: test here */
-+ if (found)
-+ return; /* success */
-+#endif
++ AuRwMustAnyLock(&sbinfo->si_rwsem);
+
-+ if (!err)
-+ err = whplink(h_dentry, inode, bindex, sb);
++ err = 0;
++ f = au_sbi(sb)->si_xib;
++ if (!f)
++ goto out;
+
-+ if (unlikely(cnt > AUFS_PLINK_WARN))
-+ AuWarn1("unexpectedly many pseudo links, %d\n", cnt);
-+ if (unlikely(err)) {
-+ AuWarn("err %d, damaged pseudo link. ignored.\n", err);
-+ if (!found && plink)
-+ do_put_plink(plink, /*do_del*/1);
++ /* stop printing the default xino path on the first writable branch */
++ h_root = NULL;
++ bindex = au_xi_root(sb, f->f_path.dentry);
++ if (bindex >= 0) {
++ br = au_sbr_sb(sb, bindex);
++ h_root = au_br_dentry(br);
+ }
-+}
+
-+void au_plink_put(struct super_block *sb)
-+{
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *plink_list;
-+ struct pseudo_link *plink, *tmp;
++ d = f->f_path.dentry;
++ name = &d->d_name;
++ /* safe ->d_parent because the file is unlinked */
++ if (d->d_parent == h_root
++ && name->len == len
++ && !memcmp(name->name, AUFS_XINO_FNAME, len))
++ goto out;
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
++ seq_puts(seq, ",xino=");
++ err = au_xino_path(seq, f);
+
-+ plink_list = &sbinfo->si_plink.head;
-+ /* spin_lock(&sbinfo->si_plink.spin); */
-+ list_for_each_entry_safe(plink, tmp, plink_list, list)
-+ do_put_plink(plink, 0);
-+ INIT_LIST_HEAD(plink_list);
-+ /* spin_unlock(&sbinfo->si_plink.head); */
++out:
++ return err;
++#endif
+}
+
-+void au_plink_half_refresh(struct super_block *sb, aufs_bindex_t br_id)
++/* seq_file will re-call me in case of too long string */
++static int aufs_show_options(struct seq_file *m, struct dentry *dentry)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *plink_list;
-+ struct pseudo_link *plink, *tmp;
-+ struct inode *inode;
-+ aufs_bindex_t bstart, bend, bindex;
-+ int do_put;
-+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+ sbinfo = au_sbi(sb);
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), PLINK));
-+
-+ plink_list = &sbinfo->si_plink.head;
-+ /* spin_lock(&sbinfo->si_plink.spin); */
-+ list_for_each_entry_safe(plink, tmp, plink_list, list) {
-+ do_put = 0;
-+ inode = au_igrab(plink->inode);
-+ ii_write_lock_child(inode);
-+ bstart = au_ibstart(inode);
-+ bend = au_ibend(inode);
-+ if (bstart >= 0) {
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ if (!au_h_iptr(inode, bindex)
-+ || au_ii_br_id(inode, bindex) != br_id)
-+ continue;
-+ au_set_h_iptr(inode, bindex, NULL, 0);
-+ do_put = 1;
-+ break;
-+ }
-+ } else
-+ do_put_plink(plink, 1);
-+
-+ if (do_put) {
-+ for (bindex = bstart; bindex <= bend; bindex++)
-+ if (au_h_iptr(inode, bindex)) {
-+ do_put = 0;
-+ break;
-+ }
-+ if (do_put)
-+ do_put_plink(plink, 1);
-+ }
-+ ii_write_unlock(inode);
-+ iput(inode);
-+ }
-+ /* spin_unlock(&sbinfo->si_plink.spin); */
-+}
-diff -urN linux/fs/aufs/robr.c linux-aufs/fs/aufs/robr.c
---- linux/fs/aufs/robr.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/robr.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,117 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ int err;
++ unsigned int mnt_flags, v;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
+
-+/*
-+ * 'robr', aufs as readonly branch of another aufs
-+ *
-+ * $Id$
-+ */
++#define AuBool(name, str) do { \
++ v = au_opt_test(mnt_flags, name); \
++ if (v != au_opt_test(AuOpt_Def, name)) \
++ seq_printf(m, ",%s" #str, v ? "" : "no"); \
++} while (0)
+
-+#include "aufs.h"
++#define AuStr(name, str) do { \
++ v = mnt_flags & AuOptMask_##name; \
++ if (v != (AuOpt_Def & AuOptMask_##name)) \
++ seq_printf(m, "," #str "=%s", au_optstr_##str(v)); \
++} while (0)
+
-+/* ---------------------------------------------------------------------- */
++#define AuUInt(name, str, val) do { \
++ if (val != AUFS_##name##_DEF) \
++ seq_printf(m, "," #str "=%u", val); \
++} while (0)
+
-+int au_test_robr_wh(struct qstr *name, struct dentry *h_parent,
-+ struct qstr *wh_name, int try_sio, struct au_ndx *ndx)
-+{
-+ if (strncmp(name->name, AUFS_WH_PFX, AUFS_WH_PFX_LEN))
-+ return au_wh_test(h_parent, wh_name, try_sio, ndx);
-+ return -EPERM;
-+}
++ sb = dentry->d_sb;
++ if (sb->s_flags & SB_POSIXACL)
++ seq_puts(m, ",acl");
++#if 0 /* reserved for future use */
++ if (sb->s_flags & SB_I_VERSION)
++ seq_puts(m, ",i_version");
++#endif
+
-+int au_test_robr_shwh(struct super_block *sb, const struct qstr *name)
-+{
-+ return 0;
-+}
++ /* lock free root dinfo */
++ si_noflush_read_lock(sb);
++ sbinfo = au_sbi(sb);
++ seq_printf(m, ",si=%lx", sysaufs_si_id(sbinfo));
+
-+/* ---------------------------------------------------------------------- */
++ mnt_flags = au_mntflags(sb);
++ if (au_opt_test(mnt_flags, XINO)) {
++ err = au_show_xino(m, sb);
++ if (unlikely(err))
++ goto out;
++ } else
++ seq_puts(m, ",noxino");
+
-+struct au_robr_lvma {
-+ struct list_head list;
-+ struct vm_area_struct *vma;
-+};
++ AuBool(TRUNC_XINO, trunc_xino);
++ AuStr(UDBA, udba);
++ AuBool(SHWH, shwh);
++ AuBool(PLINK, plink);
++ AuBool(DIO, dio);
++ AuBool(DIRPERM1, dirperm1);
+
-+struct file *au_robr_safe_file(struct vm_area_struct *vma)
-+{
-+ struct file *file = vma->vm_file;
-+ struct super_block *sb = file->f_dentry->d_sb;
-+ struct au_robr_lvma *lvma, *entry;
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *head;
-+ unsigned char found, warn;
++ v = sbinfo->si_wbr_create;
++ if (v != AuWbrCreate_Def)
++ au_show_wbr_create(m, v, sbinfo);
+
-+ AuTraceEnter();
++ v = sbinfo->si_wbr_copyup;
++ if (v != AuWbrCopyup_Def)
++ seq_printf(m, ",cpup=%s", au_optstr_wbr_copyup(v));
+
-+ if (!file->private_data || !au_test_aufs(sb))
-+ return NULL;
++ v = au_opt_test(mnt_flags, ALWAYS_DIROPQ);
++ if (v != au_opt_test(AuOpt_Def, ALWAYS_DIROPQ))
++ seq_printf(m, ",diropq=%c", v ? 'a' : 'w');
+
-+ warn = 0;
-+ found = 0;
-+ sbinfo = au_sbi(sb);
-+ head = &sbinfo->si_lvma.head;
-+ spin_lock(&sbinfo->si_lvma.spin);
-+ list_for_each_entry(entry, head, list) {
-+ found = (entry->vma == vma);
-+ if (unlikely(found))
-+ break;
-+ }
-+ if (!found) {
-+ lvma = kmalloc(sizeof(*lvma), GFP_ATOMIC);
-+ if (lvma) {
-+ lvma->vma = vma;
-+ list_add(&lvma->list, head);
-+ } else {
-+ warn = 1;
-+ file = NULL;
-+ }
-+ } else
-+ file = NULL;
-+ spin_unlock(&sbinfo->si_lvma.spin);
++ AuUInt(DIRWH, dirwh, sbinfo->si_dirwh);
+
-+ if (unlikely(warn))
-+ AuWarn1("no memory for lvma\n");
-+ return file;
-+}
++ v = jiffies_to_msecs(sbinfo->si_rdcache) / MSEC_PER_SEC;
++ AuUInt(RDCACHE, rdcache, v);
+
-+void au_robr_reset_file(struct vm_area_struct *vma, struct file *file)
-+{
-+ struct super_block *sb = file->f_dentry->d_sb;
-+ struct au_robr_lvma *entry, *found;
-+ struct au_sbinfo *sbinfo;
-+ struct list_head *head;
++ AuUInt(RDBLK, rdblk, sbinfo->si_rdblk);
++ AuUInt(RDHASH, rdhash, sbinfo->si_rdhash);
+
-+ AuTraceEnter();
-+ AuDebugOn(!au_test_aufs(sb));
++ au_fhsm_show(m, sbinfo);
+
-+ vma->vm_file = file;
-+ /* smp_mb(); */ /* flush vm_file */
++ AuBool(DIRREN, dirren);
++ AuBool(SUM, sum);
++ /* AuBool(SUM_W, wsum); */
++ AuBool(WARN_PERM, warn_perm);
++ AuBool(VERBOSE, verbose);
+
-+ found = NULL;
-+ sbinfo = au_sbi(sb);
-+ head = &sbinfo->si_lvma.head;
-+ spin_lock(&sbinfo->si_lvma.spin);
-+ list_for_each_entry(entry, head, list)
-+ if (entry->vma == vma) {
-+ found = entry;
-+ break;
-+ }
-+ AuDebugOn(!found);
-+ list_del(&found->list);
-+ spin_unlock(&sbinfo->si_lvma.spin);
-+ kfree(found);
++out:
++ /* be sure to print "br:" last */
++ if (!sysaufs_brs) {
++ seq_puts(m, ",br:");
++ au_show_brs(m, sb);
++ }
++ si_read_unlock(sb);
++ return 0;
++
++#undef AuBool
++#undef AuStr
++#undef AuUInt
+}
-diff -urN linux/fs/aufs/sbinfo.c linux-aufs/fs/aufs/sbinfo.c
---- linux/fs/aufs/sbinfo.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/sbinfo.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,221 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
+
-+/*
-+ * superblock private data
-+ *
-+ * $Id$
-+ */
++/* ---------------------------------------------------------------------- */
+
-+#include <linux/smp_lock.h>
-+#include "aufs.h"
++/* sum mode which returns the summation for statfs(2) */
+
-+/*
-+ * they are necessary regardless sysfs is disabled.
-+ */
-+void au_si_free(struct kobject *kobj)
++static u64 au_add_till_max(u64 a, u64 b)
+{
-+ struct au_sbinfo *sbinfo;
-+ struct super_block *sb;
++ u64 old;
+
-+ LKTRTrace("kobj %p\n", kobj);
-+ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
-+ LKTRTrace("sbinfo %p\n", sbinfo);
-+ AuDebugOn(!list_empty(&sbinfo->si_plink.head));
++ old = a;
++ a += b;
++ if (old <= a)
++ return a;
++ return ULLONG_MAX;
++}
+
-+ sb = sbinfo->si_sb;
-+ si_write_lock(sb);
-+ au_xino_clr(sb);
-+ au_br_free(sbinfo);
-+ kfree(sbinfo->si_branch);
-+ au_export_put(sbinfo);
-+ mutex_destroy(&sbinfo->si_xib_mtx);
-+ si_write_unlock(sb);
-+ au_rwsem_destroy(&sbinfo->si_rwsem);
++static u64 au_mul_till_max(u64 a, long mul)
++{
++ u64 old;
+
-+ kfree(sbinfo);
++ old = a;
++ a *= mul;
++ if (old <= a)
++ return a;
++ return ULLONG_MAX;
+}
+
-+int au_si_alloc(struct super_block *sb)
++static int au_statfs_sum(struct super_block *sb, struct kstatfs *buf)
+{
+ int err;
-+ struct au_sbinfo *sbinfo;
++ long bsize, factor;
++ u64 blocks, bfree, bavail, files, ffree;
++ aufs_bindex_t bbot, bindex, i;
++ unsigned char shared;
++ struct path h_path;
++ struct super_block *h_sb;
+
-+ AuTraceEnter();
++ err = 0;
++ bsize = LONG_MAX;
++ files = 0;
++ ffree = 0;
++ blocks = 0;
++ bfree = 0;
++ bavail = 0;
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ h_path.mnt = au_sbr_mnt(sb, bindex);
++ h_sb = h_path.mnt->mnt_sb;
++ shared = 0;
++ for (i = 0; !shared && i < bindex; i++)
++ shared = (au_sbr_sb(sb, i) == h_sb);
++ if (shared)
++ continue;
+
-+ err = -ENOMEM;
-+ sbinfo = kmalloc(sizeof(*sbinfo), GFP_NOFS);
-+ if (unlikely(!sbinfo))
-+ goto out;
-+ sbinfo->si_branch = kzalloc(sizeof(*sbinfo->si_branch), GFP_NOFS);
-+ if (unlikely(!sbinfo->si_branch))
-+ goto out_sbinfo;
++ /* sb->s_root for NFS is unreliable */
++ h_path.dentry = h_path.mnt->mnt_root;
++ err = vfs_statfs(&h_path, buf);
++ if (unlikely(err))
++ goto out;
+
-+ memset(&sbinfo->si_kobj, 0, sizeof(sbinfo->si_kobj));
-+ err = sysaufs_si_init(sbinfo);
-+ if (unlikely(err))
-+ goto out_br;
++ if (bsize > buf->f_bsize) {
++ /*
++ * we will reduce bsize, so we have to expand blocks
++ * etc. to match them again
++ */
++ factor = (bsize / buf->f_bsize);
++ blocks = au_mul_till_max(blocks, factor);
++ bfree = au_mul_till_max(bfree, factor);
++ bavail = au_mul_till_max(bavail, factor);
++ bsize = buf->f_bsize;
++ }
+
-+ au_rw_init_wlock(&sbinfo->si_rwsem);
-+ //au_dbg_locked_si_reg(sb, 1);
-+ sbinfo->si_generation = 0;
-+ sbinfo->au_si_status = 0;
-+ sbinfo->si_bend = -1;
-+ sbinfo->si_last_br_id = 0;
++ factor = (buf->f_bsize / bsize);
++ blocks = au_add_till_max(blocks,
++ au_mul_till_max(buf->f_blocks, factor));
++ bfree = au_add_till_max(bfree,
++ au_mul_till_max(buf->f_bfree, factor));
++ bavail = au_add_till_max(bavail,
++ au_mul_till_max(buf->f_bavail, factor));
++ files = au_add_till_max(files, buf->f_files);
++ ffree = au_add_till_max(ffree, buf->f_ffree);
++ }
+
-+ sbinfo->si_wbr_copyup = AuWbrCopyup_Def;
-+ sbinfo->si_wbr_create = AuWbrCreate_Def;
-+ sbinfo->si_wbr_copyup_ops = au_wbr_copyup_ops + AuWbrCopyup_Def;
-+ sbinfo->si_wbr_create_ops = au_wbr_create_ops + AuWbrCreate_Def;
++ buf->f_bsize = bsize;
++ buf->f_blocks = blocks;
++ buf->f_bfree = bfree;
++ buf->f_bavail = bavail;
++ buf->f_files = files;
++ buf->f_ffree = ffree;
++ buf->f_frsize = 0;
+
-+ sbinfo->si_mntflags = AuOpt_Def;
++out:
++ return err;
++}
+
-+ sbinfo->si_xread = NULL;
-+ sbinfo->si_xwrite = NULL;
-+ sbinfo->si_xib = NULL;
-+ mutex_init(&sbinfo->si_xib_mtx);
-+ sbinfo->si_xib_buf = NULL;
-+ au_xino_def_br_set(NULL, sbinfo);
-+ /* leave si_xib_last_pindex and si_xib_next_bit */
++static int aufs_statfs(struct dentry *dentry, struct kstatfs *buf)
++{
++ int err;
++ struct path h_path;
++ struct super_block *sb;
+
-+ au_nwt_init(&sbinfo->si_nowait);
++ /* lock free root dinfo */
++ sb = dentry->d_sb;
++ si_noflush_read_lock(sb);
++ if (!au_opt_test(au_mntflags(sb), SUM)) {
++ /* sb->s_root for NFS is unreliable */
++ h_path.mnt = au_sbr_mnt(sb, 0);
++ h_path.dentry = h_path.mnt->mnt_root;
++ err = vfs_statfs(&h_path, buf);
++ } else
++ err = au_statfs_sum(sb, buf);
++ si_read_unlock(sb);
+
-+ sbinfo->si_rdcache = AUFS_RDCACHE_DEF * HZ;
-+ sbinfo->si_dirwh = AUFS_DIRWH_DEF;
++ if (!err) {
++ buf->f_type = AUFS_SUPER_MAGIC;
++ buf->f_namelen = AUFS_MAX_NAMELEN;
++ memset(&buf->f_fsid, 0, sizeof(buf->f_fsid));
++ }
++ /* buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; */
+
-+ au_spl_init(&sbinfo->si_plink);
++ return err;
++}
+
-+ au_robr_lvma_init(sbinfo);
++/* ---------------------------------------------------------------------- */
+
-+ /* leave other members for sysaufs and si_mnt. */
-+ sbinfo->si_sb = sb;
++static int aufs_sync_fs(struct super_block *sb, int wait)
++{
++ int err, e;
++ aufs_bindex_t bbot, bindex;
++ struct au_branch *br;
++ struct super_block *h_sb;
+
-+ sb->s_fs_info = sbinfo;
++ err = 0;
++ si_noflush_read_lock(sb);
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (!au_br_writable(br->br_perm))
++ continue;
+
-+ au_debug_sbinfo_init(sbinfo);
-+ return 0; /* success */
++ h_sb = au_sbr_sb(sb, bindex);
++ e = vfsub_sync_filesystem(h_sb);
++ if (unlikely(e && !err))
++ err = e;
++ /* go on even if an error happens */
++ }
++ si_read_unlock(sb);
+
-+ out_br:
-+ kfree(sbinfo->si_branch);
-+ out_sbinfo:
-+ kfree(sbinfo);
-+ out:
-+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct au_branch *au_sbr(struct super_block *sb, aufs_bindex_t bindex)
++/* final actions when unmounting a file system */
++static void aufs_put_super(struct super_block *sb)
+{
-+ struct au_branch *br;
++ struct au_sbinfo *sbinfo;
+
-+ SiMustAnyLock(sb);
-+ AuDebugOn(bindex < 0 || au_sbend(sb) < bindex);
-+ br = au_sbi(sb)->si_branch[0 + bindex];
-+ AuDebugOn(!br);
-+ return br;
++ sbinfo = au_sbi(sb);
++ if (sbinfo)
++ kobject_put(&sbinfo->si_kobj);
+}
+
-+au_gen_t au_sigen_inc(struct super_block *sb)
++/* ---------------------------------------------------------------------- */
++
++void *au_array_alloc(unsigned long long *hint, au_arraycb_t cb,
++ struct super_block *sb, void *arg)
+{
-+ au_gen_t gen;
++ void *array;
++ unsigned long long n, sz;
+
-+ SiMustWriteLock(sb);
-+ gen = ++au_sbi(sb)->si_generation;
-+ au_update_digen(sb->s_root);
-+ au_update_iigen(sb->s_root->d_inode);
-+ sb->s_root->d_inode->i_version++;
-+ return gen;
-+}
++ array = NULL;
++ n = 0;
++ if (!*hint)
++ goto out;
+
-+int au_find_bindex(struct super_block *sb, struct au_branch *br)
-+{
-+ aufs_bindex_t bindex, bend;
++ if (*hint > ULLONG_MAX / sizeof(array)) {
++ array = ERR_PTR(-EMFILE);
++ pr_err("hint %llu\n", *hint);
++ goto out;
++ }
+
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++)
-+ if (au_sbr(sb, bindex) == br)
-+ return bindex;
-+ return -1;
-+}
++ sz = sizeof(array) * *hint;
++ array = kzalloc(sz, GFP_NOFS);
++ if (unlikely(!array))
++ array = vzalloc(sz);
++ if (unlikely(!array)) {
++ array = ERR_PTR(-ENOMEM);
++ goto out;
++ }
+
-+/* ---------------------------------------------------------------------- */
++ n = cb(sb, array, *hint, arg);
++ AuDebugOn(n > *hint);
+
-+/* dentry and super_block lock. call at entry point */
-+void aufs_read_lock(struct dentry *dentry, int flags)
-+{
-+ si_read_lock(dentry->d_sb, flags);
-+ if (au_ftest_lock(flags, DW))
-+ di_write_lock_child(dentry);
-+ else
-+ di_read_lock_child(dentry, flags);
++out:
++ *hint = n;
++ return array;
+}
+
-+void aufs_read_unlock(struct dentry *dentry, int flags)
++static unsigned long long au_iarray_cb(struct super_block *sb, void *a,
++ unsigned long long max __maybe_unused,
++ void *arg)
+{
-+ if (au_ftest_lock(flags, DW))
-+ di_write_unlock(dentry);
-+ else
-+ di_read_unlock(dentry, flags);
-+ si_read_unlock(dentry->d_sb);
-+}
++ unsigned long long n;
++ struct inode **p, *inode;
++ struct list_head *head;
+
-+void aufs_write_lock(struct dentry *dentry)
-+{
-+ si_write_lock(dentry->d_sb);
-+ di_write_lock_child(dentry);
-+}
++ n = 0;
++ p = a;
++ head = arg;
++ spin_lock(&sb->s_inode_list_lock);
++ list_for_each_entry(inode, head, i_sb_list) {
++ if (!au_is_bad_inode(inode)
++ && au_ii(inode)->ii_btop >= 0) {
++ spin_lock(&inode->i_lock);
++ if (atomic_read(&inode->i_count)) {
++ au_igrab(inode);
++ *p++ = inode;
++ n++;
++ AuDebugOn(n > max);
++ }
++ spin_unlock(&inode->i_lock);
++ }
++ }
++ spin_unlock(&sb->s_inode_list_lock);
+
-+void aufs_write_unlock(struct dentry *dentry)
-+{
-+ di_write_unlock(dentry);
-+ si_write_unlock(dentry->d_sb);
++ return n;
+}
+
-+void aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags)
++struct inode **au_iarray_alloc(struct super_block *sb, unsigned long long *max)
+{
-+ AuDebugOn(d1 == d2 || d1->d_sb != d2->d_sb);
-+ si_read_lock(d1->d_sb, flags);
-+ di_write_lock2_child(d1, d2, au_ftest_lock(flags, DIR));
++ struct au_sbinfo *sbi;
++
++ sbi = au_sbi(sb);
++ *max = au_lcnt_read(&sbi->si_ninodes, /*do_rev*/1);
++ return au_array_alloc(max, au_iarray_cb, sb, &sb->s_inodes);
+}
+
-+void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2)
++void au_iarray_free(struct inode **a, unsigned long long max)
+{
-+ AuDebugOn(d1 == d2 || d1->d_sb != d2->d_sb);
-+ di_write_unlock2(d1, d2);
-+ si_read_unlock(d1->d_sb);
++ unsigned long long ull;
++
++ for (ull = 0; ull < max; ull++)
++ iput(a[ull]);
++ kvfree(a);
+}
+
+/* ---------------------------------------------------------------------- */
+
-+aufs_bindex_t au_new_br_id(struct super_block *sb)
++/*
++ * refresh dentry and inode at remount time.
++ */
++/* todo: consolidate with simple_reval_dpath() and au_reval_for_attr() */
++static int au_do_refresh(struct dentry *dentry, unsigned int dir_flags,
++ struct dentry *parent)
+{
-+ aufs_bindex_t br_id;
-+ struct au_sbinfo *sbinfo;
++ int err;
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
++ di_write_lock_child(dentry);
++ di_read_lock_parent(parent, AuLock_IR);
++ err = au_refresh_dentry(dentry, parent);
++ if (!err && dir_flags)
++ au_hn_reset(d_inode(dentry), dir_flags);
++ di_read_unlock(parent, AuLock_IR);
++ di_write_unlock(dentry);
+
-+ sbinfo = au_sbi(sb);
-+ while (1) {
-+ br_id = ++sbinfo->si_last_br_id;
-+ if (br_id && au_br_index(sb, br_id) < 0)
-+ return br_id;
-+ }
++ return err;
+}
-diff -urN linux/fs/aufs/super.c linux-aufs/fs/aufs/super.c
---- linux/fs/aufs/super.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/super.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,939 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
+
-+/*
-+ * mount and super_block operations
-+ *
-+ * $Id$
-+ */
++static int au_do_refresh_d(struct dentry *dentry, unsigned int sigen,
++ struct au_sbinfo *sbinfo,
++ const unsigned int dir_flags, unsigned int do_idop)
++{
++ int err;
++ struct dentry *parent;
+
-+#include <linux/module.h>
-+#include <linux/buffer_head.h>
-+#include <linux/seq_file.h>
-+#include <linux/smp_lock.h>
-+#include <linux/statfs.h>
++ err = 0;
++ parent = dget_parent(dentry);
++ if (!au_digen_test(parent, sigen) && au_digen_test(dentry, sigen)) {
++ if (d_really_is_positive(dentry)) {
++ if (!d_is_dir(dentry))
++ err = au_do_refresh(dentry, /*dir_flags*/0,
++ parent);
++ else {
++ err = au_do_refresh(dentry, dir_flags, parent);
++ if (unlikely(err))
++ au_fset_si(sbinfo, FAILED_REFRESH_DIR);
++ }
++ } else
++ err = au_do_refresh(dentry, /*dir_flags*/0, parent);
++ AuDbgDentry(dentry);
++ }
++ dput(parent);
+
-+#include "aufs.h"
++ if (!err) {
++ if (do_idop)
++ au_refresh_dop(dentry, /*force_reval*/0);
++ } else
++ au_refresh_dop(dentry, /*force_reval*/1);
+
-+/*
-+ * super_operations
-+ */
-+static struct inode *aufs_alloc_inode(struct super_block *sb)
++ AuTraceErr(err);
++ return err;
++}
++
++static int au_refresh_d(struct super_block *sb, unsigned int do_idop)
+{
-+ struct aufs_icntnr *c;
++ int err, i, j, ndentry, e;
++ unsigned int sigen;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry **dentries, *d;
++ struct au_sbinfo *sbinfo;
++ struct dentry *root = sb->s_root;
++ const unsigned int dir_flags = au_hi_flags(d_inode(root), /*isdir*/1);
+
-+ AuTraceEnter();
++ if (do_idop)
++ au_refresh_dop(root, /*force_reval*/0);
+
-+ c = au_cache_alloc_icntnr();
-+ if (c) {
-+ inode_init_once(&c->vfs_inode);
-+ c->vfs_inode.i_version = 1; /* sigen(sb); */
-+ c->iinfo.ii_hinode = NULL;
-+ return &c->vfs_inode;
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_dcsub_pages(&dpages, root, NULL, NULL);
++ if (unlikely(err))
++ goto out_dpages;
++
++ sigen = au_sigen(sb);
++ sbinfo = au_sbi(sb);
++ for (i = 0; i < dpages.ndpage; i++) {
++ dpage = dpages.dpages + i;
++ dentries = dpage->dentries;
++ ndentry = dpage->ndentry;
++ for (j = 0; j < ndentry; j++) {
++ d = dentries[j];
++ e = au_do_refresh_d(d, sigen, sbinfo, dir_flags,
++ do_idop);
++ if (unlikely(e && !err))
++ err = e;
++ /* go on even err */
++ }
+ }
-+ return NULL;
++
++out_dpages:
++ au_dpages_free(&dpages);
++out:
++ return err;
+}
+
-+static void aufs_destroy_inode(struct inode *inode)
++static int au_refresh_i(struct super_block *sb, unsigned int do_idop)
+{
-+ int err;
++ int err, e;
++ unsigned int sigen;
++ unsigned long long max, ull;
++ struct inode *inode, **array;
+
-+ LKTRTrace("i%lu\n", inode->i_ino);
++ array = au_iarray_alloc(sb, &max);
++ err = PTR_ERR(array);
++ if (IS_ERR(array))
++ goto out;
+
-+ if (!inode->i_nlink || IS_DEADDIR(inode)) {
-+ /* todo: move this lock into xigen_inc() */
-+ /* in nowait task or remount, sbi is write-locked */
-+ struct super_block *sb = inode->i_sb;
-+ const int locked = si_noflush_read_trylock(sb);
++ err = 0;
++ sigen = au_sigen(sb);
++ for (ull = 0; ull < max; ull++) {
++ inode = array[ull];
++ if (unlikely(!inode))
++ break;
+
-+ err = au_xigen_inc(inode);
-+ if (unlikely(err))
-+ AuWarn1("failed resetting i_generation, %d\n", err);
-+ if (locked)
-+ si_read_unlock(sb);
++ e = 0;
++ ii_write_lock_child(inode);
++ if (au_iigen(inode, NULL) != sigen) {
++ e = au_refresh_hinode_self(inode);
++ if (unlikely(e)) {
++ au_refresh_iop(inode, /*force_getattr*/1);
++ pr_err("error %d, i%lu\n", e, inode->i_ino);
++ if (!err)
++ err = e;
++ /* go on even if err */
++ }
++ }
++ if (!e && do_idop)
++ au_refresh_iop(inode, /*force_getattr*/0);
++ ii_write_unlock(inode);
+ }
+
-+ au_iinfo_fin(inode);
-+ au_cache_free_icntnr(container_of(inode, struct aufs_icntnr,
-+ vfs_inode));
++ au_iarray_free(array, max);
++
++out:
++ return err;
+}
+
-+struct inode *au_iget_locked(struct super_block *sb, ino_t ino)
++void au_remount_refresh(struct super_block *sb, unsigned int do_idop)
+{
++ int err, e;
++ unsigned int udba;
++ aufs_bindex_t bindex, bbot;
++ struct dentry *root;
+ struct inode *inode;
-+ int err;
++ struct au_branch *br;
++ struct au_sbinfo *sbi;
+
-+ LKTRTrace("i%lu\n", (unsigned long)ino);
++ au_sigen_inc(sb);
++ sbi = au_sbi(sb);
++ au_fclr_si(sbi, FAILED_REFRESH_DIR);
+
-+ inode = iget_locked(sb, ino);
-+ if (unlikely(!inode)) {
-+ inode = ERR_PTR(-ENOMEM);
-+ goto out;
-+ }
-+ AuDebugOn(IS_ERR(inode));
-+ if (!(inode->i_state & I_NEW))
-+ goto out;
++ root = sb->s_root;
++ DiMustNoWaiters(root);
++ inode = d_inode(root);
++ IiMustNoWaiters(inode);
+
-+ err = au_xigen_new(inode);
-+ if (!err)
-+ err = au_iinfo_init(inode);
-+ if (!err)
-+ inode->i_version++;
-+ else {
-+ iget_failed(inode);
-+ inode = ERR_PTR(err);
++ udba = au_opt_udba(sb);
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ err = au_hnotify_reset_br(udba, br, br->br_perm);
++ if (unlikely(err))
++ AuIOErr("hnotify failed on br %d, %d, ignored\n",
++ bindex, err);
++ /* go on even if err */
++ }
++ au_hn_reset(inode, au_hi_flags(inode, /*isdir*/1));
++
++ if (do_idop) {
++ if (au_ftest_si(sbi, NO_DREVAL)) {
++ AuDebugOn(sb->s_d_op == &aufs_dop_noreval);
++ sb->s_d_op = &aufs_dop_noreval;
++ AuDebugOn(sbi->si_iop_array == aufs_iop_nogetattr);
++ sbi->si_iop_array = aufs_iop_nogetattr;
++ } else {
++ AuDebugOn(sb->s_d_op == &aufs_dop);
++ sb->s_d_op = &aufs_dop;
++ AuDebugOn(sbi->si_iop_array == aufs_iop);
++ sbi->si_iop_array = aufs_iop;
++ }
++ pr_info("reset to %ps and %ps\n",
++ sb->s_d_op, sbi->si_iop_array);
+ }
+
-+ out:
-+ /* never return NULL */
-+ AuDebugOn(!inode);
-+ AuTraceErrPtr(inode);
-+ return inode;
++ di_write_unlock(root);
++ err = au_refresh_d(sb, do_idop);
++ e = au_refresh_i(sb, do_idop);
++ if (unlikely(e && !err))
++ err = e;
++ /* aufs_write_lock() calls ..._child() */
++ di_write_lock_child(root);
++
++ au_cpup_attr_all(inode, /*force*/1);
++
++ if (unlikely(err))
++ AuIOErr("refresh failed, ignored, %d\n", err);
+}
+
-+/* lock free root dinfo */
-+static int au_show_brs(struct seq_file *seq, struct super_block *sb)
++const struct super_operations aufs_sop = {
++ .alloc_inode = aufs_alloc_inode,
++ .destroy_inode = aufs_destroy_inode,
++ .free_inode = aufs_free_inode,
++ /* always deleting, no clearing */
++ .drop_inode = generic_delete_inode,
++ .show_options = aufs_show_options,
++ .statfs = aufs_statfs,
++ .put_super = aufs_put_super,
++ .sync_fs = aufs_sync_fs
++};
++
++/* ---------------------------------------------------------------------- */
++
++int au_alloc_root(struct super_block *sb)
+{
+ int err;
-+ aufs_bindex_t bindex, bend;
-+ struct path path;
-+ struct au_hdentry *hd;
++ struct inode *inode;
++ struct dentry *root;
++
++ err = -ENOMEM;
++ inode = au_iget_locked(sb, AUFS_ROOT_INO);
++ err = PTR_ERR(inode);
++ if (IS_ERR(inode))
++ goto out;
+
-+ AuTraceEnter();
++ inode->i_op = aufs_iop + AuIop_DIR; /* with getattr by default */
++ inode->i_fop = &aufs_dir_fop;
++ inode->i_mode = S_IFDIR;
++ set_nlink(inode, 2);
++ unlock_new_inode(inode);
+
-+ err = 0;
-+ bend = au_sbend(sb);
-+ hd = au_di(sb->s_root)->di_hdentry;
-+ for (bindex = 0; !err && bindex <= bend; bindex++) {
-+ path.mnt = au_sbr_mnt(sb, bindex);
-+ path.dentry = hd[bindex].hd_dentry;
-+ err = seq_path(seq, &path, au_esc_chars);
-+ if (err > 0) {
-+ const char *p = au_optstr_br_perm(au_sbr_perm(sb,
-+ bindex));
-+ err = seq_printf(seq, "=%s", p);
-+ }
-+ if (!err && bindex != bend)
-+ err = seq_putc(seq, ':');
++ root = d_make_root(inode);
++ if (unlikely(!root))
++ goto out;
++ err = PTR_ERR(root);
++ if (IS_ERR(root))
++ goto out;
++
++ err = au_di_init(root);
++ if (!err) {
++ sb->s_root = root;
++ return 0; /* success */
+ }
++ dput(root);
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static void au_show_wbr_create(struct seq_file *m, int v,
-+ struct au_sbinfo *sbinfo)
++/* ---------------------------------------------------------------------- */
++
++static void aufs_kill_sb(struct super_block *sb)
+{
-+ const char *pat;
++ struct au_sbinfo *sbinfo;
++ struct dentry *root;
+
-+ AuDebugOn(v == AuWbrCreate_Def);
++ sbinfo = au_sbi(sb);
++ if (!sbinfo)
++ goto out;
+
-+ seq_printf(m, ",create=");
-+ pat = au_optstr_wbr_create(v);
-+ switch (v) {
-+ case AuWbrCreate_TDP:
-+ case AuWbrCreate_RR:
-+ case AuWbrCreate_MFS:
-+ case AuWbrCreate_PMFS:
-+ seq_printf(m, pat);
-+ break;
-+ case AuWbrCreate_MFSV:
-+ seq_printf(m, /*pat*/"mfs:%lu",
-+ sbinfo->si_wbr_mfs.mfs_expire / HZ);
-+ break;
-+ case AuWbrCreate_PMFSV:
-+ seq_printf(m, /*pat*/"pmfs:%lu",
-+ sbinfo->si_wbr_mfs.mfs_expire / HZ);
-+ 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,
-+ sbinfo->si_wbr_mfs.mfs_expire / HZ);
-+ break;
++ au_sbilist_del(sb);
++
++ root = sb->s_root;
++ if (root)
++ aufs_write_lock(root);
++ else
++ __si_write_lock(sb);
++
++ au_fhsm_fin(sb);
++ if (sbinfo->si_wbr_create_ops->fin)
++ sbinfo->si_wbr_create_ops->fin(sb);
++ if (au_opt_test(sbinfo->si_mntflags, UDBA_HNOTIFY)) {
++ au_opt_set_udba(sbinfo->si_mntflags, UDBA_NONE);
++ au_remount_refresh(sb, /*do_idop*/0);
+ }
-+}
++ if (au_opt_test(sbinfo->si_mntflags, PLINK))
++ au_plink_put(sb, /*verbose*/1);
++ au_xino_clr(sb);
++ if (root)
++ au_dr_opt_flush(sb);
+
-+/* seq_file will re-call me in case of too long string */
-+static int aufs_show_options(struct seq_file *m, struct vfsmount *mnt)
-+{
-+ int err, n;
-+ unsigned int mnt_flags, v;
-+ struct path path;
-+ struct super_block *sb;
-+ struct au_sbinfo *sbinfo;
-+ struct file *xino;
++ if (root)
++ aufs_write_unlock(root);
++ else
++ __si_write_unlock(sb);
+
-+ AuTraceEnter();
++ sbinfo->si_sb = NULL;
++ au_nwt_flush(&sbinfo->si_nowait);
+
-+ sb = mnt->mnt_sb;
-+ /* lock free root dinfo */
-+ si_noflush_read_lock(sb);
-+ sbinfo = au_sbi(sb);
-+ seq_printf(m, ",si=%lx", au_si_mask ^ (unsigned long)sbinfo);
-+ mnt_flags = au_mntflags(sb);
-+ if (au_opt_test(mnt_flags, XINO)) {
-+ seq_puts(m, ",xino=");
-+ xino = sbinfo->si_xib;
-+ path.mnt = xino->f_vfsmnt;
-+ path.dentry = xino->f_dentry;
-+ err = seq_path(m, &path, au_esc_chars);
-+ if (unlikely(err <= 0))
-+ goto out;
-+ err = 0;
-+#define Deleted "\\040(deleted)"
-+ m->count -= sizeof(Deleted) - 1;
-+ AuDebugOn(memcmp(m->buf + m->count, Deleted,
-+ sizeof(Deleted) - 1));
-+#undef Deleted
-+#ifdef CONFIG_AUFS_EXPORT /* reserved for future use */
-+ } else if (au_opt_test(mnt_flags, XINODIR)) {
-+ seq_puts(m, ",xinodir=");
-+ seq_path(m, &sbinfo->si_xinodir, au_esc_chars);
-+#endif
-+ } else
-+ seq_puts(m, ",noxino");
++out:
++ kill_anon_super(sb);
++}
++
++struct file_system_type aufs_fs_type = {
++ .name = AUFS_FSTYPE,
++ /* a race between rename and others */
++ .fs_flags = FS_RENAME_DOES_D_MOVE
++ /* untested */
++ /*| FS_ALLOW_IDMAP*/
++ ,
++ .init_fs_context = aufs_fsctx_init,
++ .parameters = aufs_fsctx_paramspec,
++ .kill_sb = aufs_kill_sb,
++ /* no need to __module_get() and module_put(). */
++ .owner = THIS_MODULE,
++};
+diff -urN /usr/share/empty/fs/aufs/super.h linux/fs/aufs/super.h
+--- /usr/share/empty/fs/aufs/super.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/super.h 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,592 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+#define AuBool(name, str) do { \
-+ v = au_opt_test(mnt_flags, name); \
-+ if (v != au_opt_test(AuOpt_Def, name)) \
-+ seq_printf(m, ",%s" #str, v ? "" : "no"); \
-+} while (0)
++/*
++ * super_block operations
++ */
+
-+#define AuStr(name, str) do { \
-+ v = mnt_flags & AuOptMask_##name; \
-+ if (v != (AuOpt_Def & AuOptMask_##name)) \
-+ seq_printf(m, "," #str "=%s", au_optstr_##str(v)); \
-+} while (0)
++#ifndef __AUFS_SUPER_H__
++#define __AUFS_SUPER_H__
+
-+#ifdef CONFIG_AUFS_COMPAT
-+#define AuStr_BrOpt "dirs="
-+#else
-+#define AuStr_BrOpt "br:"
-+#endif
++#ifdef __KERNEL__
+
-+ AuBool(TRUNC_XINO, trunc_xino);
-+ AuBool(DIRPERM1, dirperm1);
-+ AuBool(SHWH, shwh);
-+ AuBool(PLINK, plink);
-+ AuStr(UDBA, udba);
++#include <linux/fs.h>
++#include <linux/kobject.h>
++#include "hbl.h"
++#include "lcnt.h"
++#include "rwsem.h"
++#include "wkq.h"
+
-+ v = sbinfo->si_wbr_create;
-+ if (v != AuWbrCreate_Def)
-+ au_show_wbr_create(m, v, sbinfo);
++/* policies to select one among multiple writable branches */
++struct au_wbr_copyup_operations {
++ int (*copyup)(struct dentry *dentry);
++};
+
-+ v = sbinfo->si_wbr_copyup;
-+ if (v != AuWbrCopyup_Def)
-+ seq_printf(m, ",cpup=%s", au_optstr_wbr_copyup(v));
++#define AuWbr_DIR 1 /* target is a dir */
++#define AuWbr_PARENT (1 << 1) /* always require a parent */
+
-+ v = au_opt_test(mnt_flags, ALWAYS_DIROPQ);
-+ if (v != au_opt_test(AuOpt_Def, ALWAYS_DIROPQ))
-+ seq_printf(m, ",diropq=%c", v ? 'a' : 'w');
-+ AuBool(REFROF, refrof);
-+ AuBool(DLGT, dlgt);
-+ AuBool(WARN_PERM, warn_perm);
-+ AuBool(VERBOSE, verbose);
-+ AuBool(SUM, sum);
-+ /* AuBool(SUM_W, wsum); */
++#define au_ftest_wbr(flags, name) ((flags) & AuWbr_##name)
++#define au_fset_wbr(flags, name) { (flags) |= AuWbr_##name; }
++#define au_fclr_wbr(flags, name) { (flags) &= ~AuWbr_##name; }
+
-+ n = sbinfo->si_dirwh;
-+ if (n != AUFS_DIRWH_DEF)
-+ seq_printf(m, ",dirwh=%d", n);
-+ n = sbinfo->si_rdcache / HZ;
-+ if (n != AUFS_RDCACHE_DEF)
-+ seq_printf(m, ",rdcache=%d", n);
++struct au_wbr_create_operations {
++ int (*create)(struct dentry *dentry, unsigned int flags);
++ int (*init)(struct super_block *sb);
++ int (*fin)(struct super_block *sb);
++};
+
-+ AuStr(COO, coo);
++struct au_wbr_mfs {
++ struct mutex mfs_lock; /* protect this structure */
++ unsigned long mfs_jiffy;
++ unsigned long mfs_expire;
++ aufs_bindex_t mfs_bindex;
+
-+ out:
-+ if (!sysaufs_brs) {
-+ seq_puts(m, "," AuStr_BrOpt);
-+ au_show_brs(m, sb);
-+ }
-+ si_read_unlock(sb);
-+ return 0;
++ unsigned long long mfsrr_bytes;
++ unsigned long long mfsrr_watermark;
++};
+
-+#undef AuBool
-+#undef AuStr
-+#undef AuStr_BrOpt
++#define AuPlink_NHASH 100
++static inline int au_plink_hash(ino_t ino)
++{
++ return ino % AuPlink_NHASH;
+}
+
-+static u64 au_add_till_max(u64 a, u64 b)
-+{
-+ u64 old;
++/* File-based Hierarchical Storage Management */
++struct au_fhsm {
++#ifdef CONFIG_AUFS_FHSM
++ /* allow only one process who can receive the notification */
++ spinlock_t fhsm_spin;
++ pid_t fhsm_pid;
++ wait_queue_head_t fhsm_wqh;
++ atomic_t fhsm_readable;
+
-+ old = a;
-+ a += b;
-+ if (old < a)
-+ return a;
-+ return ULLONG_MAX;
-+}
++ /* these are protected by si_rwsem */
++ unsigned long fhsm_expire;
++ aufs_bindex_t fhsm_bottom;
++#endif
++};
+
-+static int au_statfs_sum(struct super_block *sb, struct kstatfs *buf, int dlgt)
-+{
-+ int err;
-+ aufs_bindex_t bend, bindex, i;
-+ unsigned char shared;
-+ u64 blocks, bfree, bavail, files, ffree;
-+ struct super_block *h_sb;
++struct au_branch;
++struct au_sbinfo {
++ /* nowait tasks in the system-wide workqueue */
++ struct au_nowait_tasks si_nowait;
+
-+ AuTraceEnter();
++ /*
++ * tried sb->s_umount, but failed due to the dependency between i_mutex.
++ * rwsem for au_sbinfo is necessary.
++ */
++ struct au_rwsem si_rwsem;
+
-+ blocks = 0;
-+ bfree = 0;
-+ bavail = 0;
-+ files = 0;
-+ ffree = 0;
++ /*
++ * dirty approach to protect sb->sb_inodes and ->s_files (gone) from
++ * remount.
++ */
++ au_lcnt_t si_ninodes, si_nfiles;
+
-+ err = 0;
-+ bend = au_sbend(sb);
-+ for (bindex = bend; bindex >= 0; bindex--) {
-+ h_sb = au_sbr_sb(sb, bindex);
-+ shared = 0;
-+ for (i = bindex + 1; !shared && i <= bend; i++)
-+ shared = au_sbr_sb(sb, i) == h_sb;
-+ if (shared)
-+ continue;
++ /* branch management */
++ unsigned int si_generation;
+
-+ err = vfsub_statfs(h_sb->s_root, buf, dlgt);
-+ if (unlikely(err))
-+ goto out;
++ /* see AuSi_ flags */
++ unsigned char au_si_status;
+
-+ blocks = au_add_till_max(blocks, buf->f_blocks);
-+ bfree = au_add_till_max(bfree, buf->f_bfree);
-+ bavail = au_add_till_max(bavail, buf->f_bavail);
-+ files = au_add_till_max(files, buf->f_files);
-+ ffree = au_add_till_max(ffree, buf->f_ffree);
-+ }
++ aufs_bindex_t si_bbot;
+
-+ buf->f_blocks = blocks;
-+ buf->f_bfree = bfree;
-+ buf->f_bavail = bavail;
-+ buf->f_files = files;
-+ buf->f_ffree = ffree;
++ /* dirty trick to keep br_id plus */
++ unsigned int si_last_br_id :
++ sizeof(aufs_bindex_t) * BITS_PER_BYTE - 1;
++ struct au_branch **si_branch;
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++ /* policy to select a writable branch */
++ unsigned char si_wbr_copyup;
++ unsigned char si_wbr_create;
++ struct au_wbr_copyup_operations *si_wbr_copyup_ops;
++ struct au_wbr_create_operations *si_wbr_create_ops;
+
-+/* todo: in case of round-robin policy, return the sum of all rw branches? */
-+static int aufs_statfs(struct dentry *dentry, struct kstatfs *buf)
-+{
-+ int err;
-+ unsigned int mnt_flags;
-+ unsigned char dlgt;
-+ struct super_block *sb;
++ /* round robin */
++ atomic_t si_wbr_rr_next;
+
-+ AuTraceEnter();
++ /* most free space */
++ struct au_wbr_mfs si_wbr_mfs;
+
-+ /* lock free root dinfo */
-+ sb = dentry->d_sb;
-+ si_noflush_read_lock(sb);
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ if (!au_opt_test(mnt_flags, SUM))
-+ err = vfsub_statfs(au_sbr_sb(sb, 0)->s_root, buf, dlgt);
-+ else
-+ err = au_statfs_sum(sb, buf, dlgt);
-+ si_read_unlock(sb);
-+ if (!err) {
-+ buf->f_type = AUFS_SUPER_MAGIC;
-+ buf->f_namelen -= AUFS_WH_PFX_LEN;
-+ memset(&buf->f_fsid, 0, sizeof(buf->f_fsid));
-+ }
-+ /* buf->f_bsize = buf->f_blocks = buf->f_bfree = buf->f_bavail = -1; */
++ /* File-based Hierarchical Storage Management */
++ struct au_fhsm si_fhsm;
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ /* mount flags */
++ /* include/asm-ia64/siginfo.h defines a macro named si_flags */
++ unsigned int si_mntflags;
+
-+static void au_fsync_br(struct super_block *sb)
-+{
-+#ifdef CONFIG_AUFS_FSYNC_SUPER_PATCH
-+ aufs_bindex_t bend, bindex;
-+ int brperm;
-+ struct au_branch *br;
-+ struct super_block *h_sb;
++ /* external inode number (bitmap and translation table) */
++ loff_t si_ximaxent; /* max entries in a xino */
+
-+ AuTraceEnter();
++ struct file *si_xib;
++ struct mutex si_xib_mtx; /* protect xib members */
++ unsigned long *si_xib_buf;
++ unsigned long si_xib_last_pindex;
++ int si_xib_next_bit;
+
-+ si_write_lock(sb);
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex < bend; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ brperm = br->br_perm;
-+ if (brperm == AuBrPerm_RR || brperm == AuBrPerm_RRWH)
-+ continue;
-+ h_sb = br->br_mnt->mnt_sb;
-+ if (bdev_read_only(h_sb->s_bdev))
-+ continue;
++ unsigned long si_xino_jiffy;
++ unsigned long si_xino_expire;
++ /* reserved for future use */
++ /* unsigned long long si_xib_limit; */ /* Max xib file size */
+
-+ lockdep_off();
-+ down_write(&h_sb->s_umount);
-+ shrink_dcache_sb(h_sb);
-+ fsync_super(h_sb);
-+ up_write(&h_sb->s_umount);
-+ lockdep_on();
-+ }
-+ si_write_unlock(sb);
++#ifdef CONFIG_AUFS_EXPORT
++ /* i_generation */
++ /* todo: make xigen file an array to support many inode numbers */
++ struct file *si_xigen;
++ atomic_t si_xigen_next;
+#endif
-+}
+
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 26)
-+/* this IS NOT for super_operations */
-+static void aufs_umount_begin(struct super_block *arg)
-+#define AuUmountBeginSb(arg) (arg)
-+#else
-+/* this IS for super_operations */
-+static void aufs_umount_begin(struct vfsmount *arg, int flags)
-+#define AuUmountBeginSb(arg) ((arg)->mnt_sb)
-+#endif
-+{
-+ struct super_block *sb = AuUmountBeginSb(arg);
-+ struct au_sbinfo *sbinfo;
++ /* dirty trick to support atomic_open */
++ struct hlist_bl_head si_aopen;
+
-+ AuTraceEnter();
-+ /* dont trust BKL */
-+ AuDebugOn(!kernel_locked());
++ /* vdir parameters */
++ unsigned long si_rdcache; /* max cache time in jiffies */
++ unsigned int si_rdblk; /* deblk size */
++ unsigned int si_rdhash; /* hash size */
+
-+ sbinfo = au_sbi(sb);
-+ if (!sbinfo)
-+ return;
++ /*
++ * If the number of whiteouts are larger than si_dirwh, leave all of
++ * them after au_whtmp_ren to reduce the cost of rmdir(2).
++ * future fsck.aufs or kernel thread will remove them later.
++ * Otherwise, remove all whiteouts and the dir in rmdir(2).
++ */
++ unsigned int si_dirwh;
+
-+ au_fsync_br(sb);
++ /* pseudo_link list */
++ 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;
+
-+ si_write_lock(sb);
-+ if (au_opt_test(au_mntflags(sb), PLINK))
-+ au_plink_put(sb);
-+ au_mnt_reset(sbinfo);
-+#if 0 /* reserved for future use */
-+ if (sbinfo->si_wbr_create_ops->fin)
-+ sbinfo->si_wbr_create_ops->fin(sb);
-+#endif
-+ si_write_unlock(sb);
-+}
++ /* file list */
++ struct hlist_bl_head si_files;
+
-+/* final actions when unmounting a file system */
-+static void aufs_put_super(struct super_block *sb)
-+{
-+ struct au_sbinfo *sbinfo;
++ /* with/without getattr, brother of sb->s_d_op */
++ const struct inode_operations *si_iop_array;
+
-+ AuTraceEnter();
++ /*
++ * sysfs and lifetime management.
++ * this is not a small structure and it may be a waste of memory in case
++ * of sysfs is disabled, particularly when many aufs-es are mounted.
++ * but using sysfs is majority.
++ */
++ struct kobject si_kobj;
++#ifdef CONFIG_DEBUG_FS
++ struct dentry *si_dbgaufs;
++ struct dentry *si_dbgaufs_plink;
++ struct dentry *si_dbgaufs_xib;
++#ifdef CONFIG_AUFS_EXPORT
++ struct dentry *si_dbgaufs_xigen;
++#endif
++#endif
+
-+ sbinfo = au_sbi(sb);
-+ if (!sbinfo)
-+ return;
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 26)
-+ aufs_umount_begin(sb);
++#ifdef CONFIG_AUFS_SBILIST
++ struct hlist_bl_node si_list;
+#endif
-+ kobject_put(&sbinfo->si_kobj);
-+}
+
-+/* ---------------------------------------------------------------------- */
++ /* dirty, necessary for unmounting, sysfs and sysrq */
++ struct super_block *si_sb;
++};
+
++/* sbinfo status flags */
+/*
-+ * refresh dentry and inode at remount time.
++ * set true when refresh_dirs() failed at remount time.
++ * then try refreshing dirs at access time again.
++ * if it is false, refreshing dirs at access time is unnecessary
+ */
-+static int do_refresh(struct dentry *dentry, mode_t type,
-+ unsigned int dir_flags)
-+{
-+ int err;
-+ struct dentry *parent;
-+ struct inode *inode;
-+
-+ LKTRTrace("%.*s, 0%o\n", AuDLNPair(dentry), type);
-+ inode = dentry->d_inode;
-+ AuDebugOn(!inode);
++#define AuSi_FAILED_REFRESH_DIR 1
++#define AuSi_FHSM (1 << 1) /* fhsm is active now */
++#define AuSi_NO_DREVAL (1 << 2) /* disable all d_revalidate */
+
-+ di_write_lock_child(dentry);
-+ parent = dget_parent(dentry);
-+ di_read_lock_parent(parent, AuLock_IR);
-+ /* returns a number of positive dentries */
-+ err = au_refresh_hdentry(dentry, type);
-+ if (err >= 0) {
-+ err = au_refresh_hinode(inode, dentry);
-+ if (!err && type == S_IFDIR)
-+ au_reset_hinotify(inode, dir_flags);
-+ }
-+ if (unlikely(err))
-+ AuErr("unrecoverable error %d, %.*s\n", err, AuDLNPair(dentry));
-+ di_read_unlock(parent, AuLock_IR);
-+ dput(parent);
-+ di_write_unlock(dentry);
++#ifndef CONFIG_AUFS_FHSM
++#undef AuSi_FHSM
++#define AuSi_FHSM 0
++#endif
+
-+ AuTraceErr(err);
-+ return err;
++static inline unsigned char au_do_ftest_si(struct au_sbinfo *sbi,
++ unsigned int flag)
++{
++ AuRwMustAnyLock(&sbi->si_rwsem);
++ return sbi->au_si_status & flag;
+}
++#define au_ftest_si(sbinfo, name) au_do_ftest_si(sbinfo, AuSi_##name)
++#define au_fset_si(sbinfo, name) do { \
++ AuRwMustWriteLock(&(sbinfo)->si_rwsem); \
++ (sbinfo)->au_si_status |= AuSi_##name; \
++} while (0)
++#define au_fclr_si(sbinfo, name) do { \
++ AuRwMustWriteLock(&(sbinfo)->si_rwsem); \
++ (sbinfo)->au_si_status &= ~AuSi_##name; \
++} while (0)
++
++/* ---------------------------------------------------------------------- */
++
++/* policy to select one among writable branches */
++#define AuWbrCopyup(sbinfo, ...) \
++ ((sbinfo)->si_wbr_copyup_ops->copyup(__VA_ARGS__))
++#define AuWbrCreate(sbinfo, ...) \
++ ((sbinfo)->si_wbr_create_ops->create(__VA_ARGS__))
+
-+static int test_dir(struct dentry *dentry, void *arg)
-+{
-+ return S_ISDIR(dentry->d_inode->i_mode);
-+}
++/* flags for si_read_lock()/aufs_read_lock()/di_read_lock() */
++#define AuLock_DW 1 /* write-lock dentry */
++#define AuLock_IR (1 << 1) /* read-lock inode */
++#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 */
++#define au_ftest_lock(flags, name) ((flags) & AuLock_##name)
++#define au_fset_lock(flags, name) \
++ do { (flags) |= AuLock_##name; } while (0)
++#define au_fclr_lock(flags, name) \
++ do { (flags) &= ~AuLock_##name; } while (0)
+
-+/* todo: merge with refresh_nondir()? */
-+static int refresh_dir(struct dentry *root, au_gen_t sgen)
-+{
-+ int err, i, j, ndentry, e;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
-+ struct dentry **dentries;
-+ struct inode *inode;
-+ const unsigned int flags = au_hi_flags(root->d_inode, /*isdir*/1);
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("sgen %d\n", sgen);
-+ SiMustWriteLock(root->d_sb);
-+ /* dont trust BKL */
-+ AuDebugOn(au_digen(root) != sgen || !kernel_locked());
++/* super.c */
++struct inode *au_iget_locked(struct super_block *sb, ino_t ino);
+
-+ err = 0;
-+ list_for_each_entry(inode, &root->d_sb->s_inodes, i_sb_list)
-+ if (S_ISDIR(inode->i_mode) && au_iigen(inode) != sgen) {
-+ ii_write_lock_child(inode);
-+ e = au_refresh_hinode_self(inode);
-+ ii_write_unlock(inode);
-+ if (unlikely(e)) {
-+ LKTRTrace("e %d, i%lu\n", e, inode->i_ino);
-+ if (!err)
-+ err = e;
-+ /* go on even if err */
-+ }
-+ }
++typedef unsigned long long (*au_arraycb_t)(struct super_block *sb, void *array,
++ unsigned long long max, void *arg);
++void *au_array_alloc(unsigned long long *hint, au_arraycb_t cb,
++ struct super_block *sb, void *arg);
++struct inode **au_iarray_alloc(struct super_block *sb, unsigned long long *max);
++void au_iarray_free(struct inode **a, unsigned long long max);
+
-+ e = au_dpages_init(&dpages, GFP_NOFS);
-+ if (unlikely(e)) {
-+ if (!err)
-+ err = e;
-+ goto out;
-+ }
-+ e = au_dcsub_pages(&dpages, root, test_dir, NULL);
-+ if (unlikely(e)) {
-+ if (!err)
-+ err = e;
-+ goto out_dpages;
-+ }
++void au_remount_refresh(struct super_block *sb, unsigned int do_idop);
++extern const struct super_operations aufs_sop;
++int au_alloc_root(struct super_block *sb);
++extern struct file_system_type aufs_fs_type;
+
-+ for (i = 0; !e && i < dpages.ndpage; i++) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ ndentry = dpage->ndentry;
-+ for (j = 0; !e && j < ndentry; j++) {
-+ struct dentry *d;
-+ d = dentries[j];
-+#ifdef CONFIG_AUFS_DEBUG
-+ {
-+ struct dentry *parent;
-+ parent = dget_parent(d);
-+ AuDebugOn(!S_ISDIR(d->d_inode->i_mode)
-+ || IS_ROOT(d)
-+ || au_digen(parent) != sgen);
-+ dput(parent);
-+ }
-+#endif
-+ if (au_digen(d) != sgen) {
-+ e = do_refresh(d, S_IFDIR, flags);
-+ if (unlikely(e && !err))
-+ err = e;
-+ /* break on err */
-+ }
-+ }
-+ }
++/* sbinfo.c */
++void au_si_free(struct kobject *kobj);
++struct au_sbinfo *au_si_alloc(struct super_block *sb);
++int au_sbr_realloc(struct au_sbinfo *sbinfo, int nbr, int may_shrink);
+
-+ out_dpages:
-+ au_dpages_free(&dpages);
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++unsigned int au_sigen_inc(struct super_block *sb);
++aufs_bindex_t au_new_br_id(struct super_block *sb);
+
-+static int test_nondir(struct dentry *dentry, void *arg)
-+{
-+ return !S_ISDIR(dentry->d_inode->i_mode);
-+}
++int si_read_lock(struct super_block *sb, int flags);
++int si_write_lock(struct super_block *sb, int flags);
++int aufs_read_lock(struct dentry *dentry, int flags);
++void aufs_read_unlock(struct dentry *dentry, int flags);
++void aufs_write_lock(struct dentry *dentry);
++void aufs_write_unlock(struct dentry *dentry);
++int aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int flags);
++void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2);
+
-+static int refresh_nondir(struct dentry *root, au_gen_t sgen, int do_dentry)
-+{
-+ int err, i, j, ndentry, e;
-+ struct au_dcsub_pages dpages;
-+ struct au_dpage *dpage;
-+ struct dentry **dentries;
-+ struct inode *inode;
++/* wbr_policy.c */
++extern struct au_wbr_copyup_operations au_wbr_copyup_ops[];
++extern struct au_wbr_create_operations au_wbr_create_ops[];
++int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst);
++int au_wbr_nonopq(struct dentry *dentry, aufs_bindex_t bindex);
++int au_wbr_do_copyup_bu(struct dentry *dentry, aufs_bindex_t btop);
+
-+ LKTRTrace("sgen %d\n", sgen);
-+ SiMustWriteLock(root->d_sb);
-+ /* dont trust BKL */
-+ AuDebugOn(au_digen(root) != sgen || !kernel_locked());
++/* mvdown.c */
++int au_mvdown(struct dentry *dentry, struct aufs_mvdown __user *arg);
+
-+ err = 0;
-+ list_for_each_entry(inode, &root->d_sb->s_inodes, i_sb_list)
-+ if (!S_ISDIR(inode->i_mode) && au_iigen(inode) != sgen) {
-+ ii_write_lock_child(inode);
-+ e = au_refresh_hinode_self(inode);
-+ ii_write_unlock(inode);
-+ if (unlikely(e)) {
-+ LKTRTrace("e %d, i%lu\n", e, inode->i_ino);
-+ if (!err)
-+ err = e;
-+ /* go on even if err */
-+ }
-+ }
++#ifdef CONFIG_AUFS_FHSM
++/* fhsm.c */
+
-+ if (!do_dentry)
-+ goto out;
++static inline pid_t au_fhsm_pid(struct au_fhsm *fhsm)
++{
++ pid_t pid;
+
-+ e = au_dpages_init(&dpages, GFP_NOFS);
-+ if (unlikely(e)) {
-+ if (!err)
-+ err = e;
-+ goto out;
-+ }
-+ e = au_dcsub_pages(&dpages, root, test_nondir, NULL);
-+ if (unlikely(e)) {
-+ if (!err)
-+ err = e;
-+ goto out_dpages;
-+ }
++ spin_lock(&fhsm->fhsm_spin);
++ pid = fhsm->fhsm_pid;
++ spin_unlock(&fhsm->fhsm_spin);
+
-+ for (i = 0; i < dpages.ndpage; i++) {
-+ dpage = dpages.dpages + i;
-+ dentries = dpage->dentries;
-+ ndentry = dpage->ndentry;
-+ for (j = 0; j < ndentry; j++) {
-+ struct dentry *d;
-+ d = dentries[j];
-+#ifdef CONFIG_AUFS_DEBUG
-+ {
-+ struct dentry *parent;
-+ parent = dget_parent(d);
-+ AuDebugOn(S_ISDIR(d->d_inode->i_mode)
-+ || au_digen(parent) != sgen);
-+ dput(parent);
-+ }
++ return pid;
++}
++
++void au_fhsm_wrote(struct super_block *sb, aufs_bindex_t bindex, int force);
++void au_fhsm_wrote_all(struct super_block *sb, int force);
++int au_fhsm_fd(struct super_block *sb, int oflags);
++int au_fhsm_br_alloc(struct au_branch *br);
++void au_fhsm_set_bottom(struct super_block *sb, aufs_bindex_t bindex);
++void au_fhsm_fin(struct super_block *sb);
++void au_fhsm_init(struct au_sbinfo *sbinfo);
++void au_fhsm_set(struct au_sbinfo *sbinfo, unsigned int sec);
++void au_fhsm_show(struct seq_file *seq, struct au_sbinfo *sbinfo);
++#else
++AuStubVoid(au_fhsm_wrote, struct super_block *sb, aufs_bindex_t bindex,
++ int force)
++AuStubVoid(au_fhsm_wrote_all, struct super_block *sb, int force)
++AuStub(int, au_fhsm_fd, return -EOPNOTSUPP, struct super_block *sb, int oflags)
++AuStub(pid_t, au_fhsm_pid, return 0, struct au_fhsm *fhsm)
++AuStubInt0(au_fhsm_br_alloc, struct au_branch *br)
++AuStubVoid(au_fhsm_set_bottom, struct super_block *sb, aufs_bindex_t bindex)
++AuStubVoid(au_fhsm_fin, struct super_block *sb)
++AuStubVoid(au_fhsm_init, struct au_sbinfo *sbinfo)
++AuStubVoid(au_fhsm_set, struct au_sbinfo *sbinfo, unsigned int sec)
++AuStubVoid(au_fhsm_show, struct seq_file *seq, struct au_sbinfo *sbinfo)
+#endif
-+ inode = d->d_inode;
-+ if (inode && au_digen(d) != sgen) {
-+ e = do_refresh(d, inode->i_mode & S_IFMT, 0);
-+ if (unlikely(e && !err))
-+ err = e;
-+ /* go on even err */
-+ }
-+ }
-+ }
+
-+ out_dpages:
-+ au_dpages_free(&dpages);
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++/* ---------------------------------------------------------------------- */
+
-+/* stop extra interpretation of errno in mount(8), and strange error messages */
-+static int cvt_err(int err)
++static inline struct au_sbinfo *au_sbi(struct super_block *sb)
+{
-+ AuTraceErr(err);
-+
-+ switch (err) {
-+ case -ENOENT:
-+ case -ENOTDIR:
-+ case -EEXIST:
-+ case -EIO:
-+ err = -EINVAL;
-+ }
-+ return err;
++ return sb->s_fs_info;
+}
+
-+/* protected by s_umount */
-+static int aufs_remount_fs(struct super_block *sb, int *flags, char *data)
-+{
-+ int err, rerr;
-+ au_gen_t sigen;
-+ struct dentry *root;
-+ struct inode *inode;
-+ struct au_opts opts;
-+ struct au_sbinfo *sbinfo;
-+ unsigned char dlgt;
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("flags 0x%x, data %s, len %lu\n",
-+ *flags, data ? data : "NULL",
-+ (unsigned long)(data ? strlen(data) : 0));
++#ifdef CONFIG_AUFS_EXPORT
++int au_test_nfsd(void);
++void au_export_init(struct super_block *sb);
++void au_xigen_inc(struct inode *inode);
++int au_xigen_new(struct inode *inode);
++int au_xigen_set(struct super_block *sb, struct path *path);
++void au_xigen_clr(struct super_block *sb);
+
-+ au_fsync_br(sb);
++static inline int au_busy_or_stale(void)
++{
++ if (!au_test_nfsd())
++ return -EBUSY;
++ return -ESTALE;
++}
++#else
++AuStubInt0(au_test_nfsd, void)
++AuStubVoid(au_export_init, struct super_block *sb)
++AuStubVoid(au_xigen_inc, struct inode *inode)
++AuStubInt0(au_xigen_new, struct inode *inode)
++AuStubInt0(au_xigen_set, struct super_block *sb, struct path *path)
++AuStubVoid(au_xigen_clr, struct super_block *sb)
++AuStub(int, au_busy_or_stale, return -EBUSY, void)
++#endif /* CONFIG_AUFS_EXPORT */
+
-+ err = 0;
-+ root = sb->s_root;
-+ if (!data || !*data) {
-+ aufs_write_lock(root);
-+ err = verify_opts(sb, *flags, /*pending*/0, /*remount*/1);
-+ aufs_write_unlock(root);
-+ goto out; /* success */
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ err = -ENOMEM;
-+ memset(&opts, 0, sizeof(opts));
-+ opts.opt = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!opts.opt))
-+ goto out;
-+ opts.max_opt = PAGE_SIZE / sizeof(*opts.opt);
-+ opts.flags = AuOpts_REMOUNT;
-+ opts.sb_flags = *flags;
++#ifdef CONFIG_AUFS_SBILIST
++/* module.c */
++extern struct hlist_bl_head au_sbilist;
+
-+ /* parse it before aufs lock */
-+ err = au_opts_parse(sb, data, &opts);
-+ if (unlikely(err))
-+ goto out_opts;
++static inline void au_sbilist_init(void)
++{
++ INIT_HLIST_BL_HEAD(&au_sbilist);
++}
+
-+ sbinfo = au_sbi(sb);
-+ inode = root->d_inode;
-+ mutex_lock(&inode->i_mutex);
-+ aufs_write_lock(root);
++static inline void au_sbilist_add(struct super_block *sb)
++{
++ au_hbl_add(&au_sbi(sb)->si_list, &au_sbilist);
++}
+
-+ /* au_do_opts() may return an error */
-+ err = au_opts_remount(sb, &opts);
-+ au_opts_free(&opts);
++static inline void au_sbilist_del(struct super_block *sb)
++{
++ au_hbl_del(&au_sbi(sb)->si_list, &au_sbilist);
++}
+
-+ if (au_ftest_opts(opts.flags, REFRESH_DIR)
-+ || au_ftest_opts(opts.flags, REFRESH_NONDIR)) {
-+ dlgt = !!au_opt_test(sbinfo->si_mntflags, DLGT);
-+ au_opt_clr(sbinfo->si_mntflags, DLGT);
-+ au_sigen_inc(sb);
-+ au_reset_hinotify(inode, au_hi_flags(inode, /*isdir*/1));
-+ sigen = au_sigen(sb);
-+ au_fclr_si(sbinfo, FAILED_REFRESH_DIRS);
++#ifdef CONFIG_AUFS_MAGIC_SYSRQ
++static inline void au_sbilist_lock(void)
++{
++ hlist_bl_lock(&au_sbilist);
++}
+
-+ DiMustNoWaiters(root);
-+ IiMustNoWaiters(root->d_inode);
-+ di_write_unlock(root);
++static inline void au_sbilist_unlock(void)
++{
++ hlist_bl_unlock(&au_sbilist);
++}
++#define AuGFP_SBILIST GFP_ATOMIC
++#else
++AuStubVoid(au_sbilist_lock, void)
++AuStubVoid(au_sbilist_unlock, void)
++#define AuGFP_SBILIST GFP_NOFS
++#endif /* CONFIG_AUFS_MAGIC_SYSRQ */
++#else
++AuStubVoid(au_sbilist_init, void)
++AuStubVoid(au_sbilist_add, struct super_block *sb)
++AuStubVoid(au_sbilist_del, struct super_block *sb)
++AuStubVoid(au_sbilist_lock, void)
++AuStubVoid(au_sbilist_unlock, void)
++#define AuGFP_SBILIST GFP_NOFS
++#endif
+
-+ rerr = refresh_dir(root, sigen);
-+ if (unlikely(rerr)) {
-+ au_fset_si(sbinfo, FAILED_REFRESH_DIRS);
-+ AuWarn("Refreshing directories failed, ignores (%d)\n",
-+ rerr);
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ if (au_ftest_opts(opts.flags, REFRESH_NONDIR)) {
-+ rerr = refresh_nondir(root, sigen, !rerr);
-+ if (unlikely(rerr))
-+ AuWarn("Refreshing non-directories failed,"
-+ " ignores (%d)\n", rerr);
-+ }
++static inline void dbgaufs_si_null(struct au_sbinfo *sbinfo)
++{
++ /*
++ * This function is a dynamic '__init' function actually,
++ * so the tiny check for si_rwsem is unnecessary.
++ */
++ /* AuRwMustWriteLock(&sbinfo->si_rwsem); */
++#ifdef CONFIG_DEBUG_FS
++ sbinfo->si_dbgaufs = NULL;
++ sbinfo->si_dbgaufs_plink = NULL;
++ sbinfo->si_dbgaufs_xib = NULL;
++#ifdef CONFIG_AUFS_EXPORT
++ sbinfo->si_dbgaufs_xigen = NULL;
++#endif
++#endif
++}
+
-+ /* aufs_write_lock() calls ..._child() */
-+ di_write_lock_child(root);
++/* ---------------------------------------------------------------------- */
+
-+ au_cpup_attr_all(inode, /*force*/1);
-+ if (dlgt)
-+ au_opt_set(sbinfo->si_mntflags, DLGT);
-+ }
++/* current->atomic_flags */
++/* this value should never corrupt the ones defined in linux/sched.h */
++#define PFA_AUFS 0x10
+
-+ aufs_write_unlock(root);
-+ mutex_unlock(&inode->i_mutex);
++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 */
+
-+ out_opts:
-+ free_page((unsigned long)opts.opt);
-+ out:
-+ err = cvt_err(err);
-+ AuTraceErr(err);
-+ return err;
++static inline int si_pid_test(struct super_block *sb)
++{
++ return !!task_test_aufs(current);
+}
+
-+static struct super_operations aufs_sop = {
-+ .alloc_inode = aufs_alloc_inode,
-+ .destroy_inode = aufs_destroy_inode,
-+ .drop_inode = generic_delete_inode,
-+
-+ .show_options = aufs_show_options,
-+ .statfs = aufs_statfs,
++static inline void si_pid_clr(struct super_block *sb)
++{
++ AuDebugOn(!task_test_aufs(current));
++ task_clear_aufs(current);
++}
+
-+ .put_super = aufs_put_super,
-+ .remount_fs = aufs_remount_fs,
-+#if LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+ .umount_begin = aufs_umount_begin
-+#endif
-+};
++static inline void si_pid_set(struct super_block *sb)
++{
++ AuDebugOn(task_test_aufs(current));
++ task_set_aufs(current);
++}
+
+/* ---------------------------------------------------------------------- */
+
-+static int alloc_root(struct super_block *sb)
-+{
-+ int err;
-+ struct inode *inode;
-+ struct dentry *root;
++/* lock superblock. mainly for entry point functions */
++#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)
++*/
+
-+ AuTraceEnter();
++#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)
+
-+ err = -ENOMEM;
-+ inode = au_iget_locked(sb, AUFS_ROOT_INO);
-+ err = PTR_ERR(inode);
-+ if (IS_ERR(inode))
-+ goto out;
-+ inode->i_op = &aufs_dir_iop;
-+ inode->i_fop = &aufs_dir_fop;
-+ inode->i_mode = S_IFDIR;
-+ unlock_new_inode(inode);
-+ root = d_alloc_root(inode);
-+ if (unlikely(!root))
-+ goto out_iput;
-+ err = PTR_ERR(root);
-+ if (IS_ERR(root))
-+ goto out_iput;
++#define SiMustNoWaiters(sb) AuRwMustNoWaiters(&au_sbi(sb)->si_rwsem)
++#define SiMustAnyLock(sb) AuRwMustAnyLock(&au_sbi(sb)->si_rwsem)
++#define SiMustWriteLock(sb) AuRwMustWriteLock(&au_sbi(sb)->si_rwsem)
+
-+ err = au_alloc_dinfo(root);
-+ if (!err) {
-+ sb->s_root = root;
-+ return 0; /* success */
-+ }
-+ dput(root);
-+ goto out; /* do not iput */
++static inline void si_noflush_read_lock(struct super_block *sb)
++{
++ __si_read_lock(sb);
++ si_pid_set(sb);
++}
+
-+ out_iput:
-+ iget_failed(inode);
-+ iput(inode);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++static inline int si_noflush_read_trylock(struct super_block *sb)
++{
++ int locked;
+
++ locked = __si_read_trylock(sb);
++ if (locked)
++ si_pid_set(sb);
++ return locked;
+}
+
-+static int aufs_fill_super(struct super_block *sb, void *raw_data, int silent)
++static inline void si_noflush_write_lock(struct super_block *sb)
+{
-+ int err;
-+ struct dentry *root;
-+ struct inode *inode;
-+ struct au_opts opts;
-+ char *arg = raw_data;
++ __si_write_lock(sb);
++ si_pid_set(sb);
++}
+
-+ if (unlikely(!arg || !*arg)) {
-+ err = -EINVAL;
-+ AuErr("no arg\n");
-+ goto out;
-+ }
-+ LKTRTrace("%s, silent %d\n", arg, silent);
++static inline int si_noflush_write_trylock(struct super_block *sb)
++{
++ int locked;
+
-+ err = -ENOMEM;
-+ memset(&opts, 0, sizeof(opts));
-+ opts.opt = (void *)__get_free_page(GFP_NOFS);
-+ if (unlikely(!opts.opt))
-+ goto out;
-+ opts.max_opt = PAGE_SIZE / sizeof(*opts.opt);
-+ opts.sb_flags = sb->s_flags;
++ locked = __si_write_trylock(sb);
++ if (locked)
++ si_pid_set(sb);
++ return locked;
++}
+
-+ err = au_si_alloc(sb);
-+ if (unlikely(err))
-+ goto out_opts;
-+ SiMustWriteLock(sb);
++#if 0 /* reserved */
++static inline int si_read_trylock(struct super_block *sb, int flags)
++{
++ if (au_ftest_lock(flags, FLUSH))
++ au_nwt_flush(&au_sbi(sb)->si_nowait);
++ return si_noflush_read_trylock(sb);
++}
++#endif
+
-+ /* all timestamps always follow the ones on the branch */
-+ sb->s_flags |= MS_NOATIME | MS_NODIRATIME;
-+ sb->s_op = &aufs_sop;
-+ sb->s_magic = AUFS_SUPER_MAGIC;
-+ au_export_init(sb);
++static inline void si_read_unlock(struct super_block *sb)
++{
++ si_pid_clr(sb);
++ __si_read_unlock(sb);
++}
+
-+ err = alloc_root(sb);
-+ if (unlikely(err)) {
-+ AuDebugOn(sb->s_root);
-+ si_write_unlock(sb);
-+ goto out_info;
-+ }
-+ root = sb->s_root;
-+ DiMustWriteLock(root);
-+ inode = root->d_inode;
-+ inode->i_nlink = 2;
++#if 0 /* reserved */
++static inline int si_write_trylock(struct super_block *sb, int flags)
++{
++ if (au_ftest_lock(flags, FLUSH))
++ au_nwt_flush(&au_sbi(sb)->si_nowait);
++ return si_noflush_write_trylock(sb);
++}
++#endif
+
-+ /*
-+ * actually we can parse options regardless aufs lock here.
-+ * but at remount time, parsing must be done before aufs lock.
-+ * so we follow the same rule.
-+ */
-+ ii_write_lock_parent(inode);
-+ au_dbg_locked_si_reg(sb, 1);
-+ aufs_write_unlock(root);
-+ err = au_opts_parse(sb, arg, &opts);
-+ if (unlikely(err))
-+ goto out_root;
++static inline void si_write_unlock(struct super_block *sb)
++{
++ si_pid_clr(sb);
++ __si_write_unlock(sb);
++}
+
-+ /* lock vfs_inode first, then aufs. */
-+ mutex_lock(&inode->i_mutex);
-+ inode->i_op = &aufs_dir_iop;
-+ inode->i_fop = &aufs_dir_fop;
-+ aufs_write_lock(root);
++#if 0 /* reserved */
++static inline void si_downgrade_lock(struct super_block *sb)
++{
++ __si_downgrade_lock(sb);
++}
++#endif
+
-+ sb->s_maxbytes = 0;
-+ err = au_opts_mount(sb, &opts);
-+ au_opts_free(&opts);
-+ if (unlikely(err))
-+ goto out_unlock;
-+ AuDebugOn(!sb->s_maxbytes);
++/* ---------------------------------------------------------------------- */
+
-+ aufs_write_unlock(root);
-+ mutex_unlock(&inode->i_mutex);
-+ goto out_opts; /* success */
++static inline aufs_bindex_t au_sbbot(struct super_block *sb)
++{
++ SiMustAnyLock(sb);
++ return au_sbi(sb)->si_bbot;
++}
+
-+ out_unlock:
-+ aufs_write_unlock(root);
-+ mutex_unlock(&inode->i_mutex);
-+ out_root:
-+ dput(root);
-+ sb->s_root = NULL;
-+ out_info:
-+ kobject_put(&au_sbi(sb)->si_kobj);
-+ sb->s_fs_info = NULL;
-+ out_opts:
-+ free_page((unsigned long)opts.opt);
-+ out:
-+ AuTraceErr(err);
-+ err = cvt_err(err);
-+ AuTraceErr(err);
-+ return err;
++static inline unsigned int au_mntflags(struct super_block *sb)
++{
++ SiMustAnyLock(sb);
++ return au_sbi(sb)->si_mntflags;
+}
+
-+/* ---------------------------------------------------------------------- */
++static inline unsigned int au_sigen(struct super_block *sb)
++{
++ SiMustAnyLock(sb);
++ return au_sbi(sb)->si_generation;
++}
+
-+static int aufs_get_sb(struct file_system_type *fs_type, int flags,
-+ const char *dev_name, void *raw_data,
-+ struct vfsmount *mnt)
++static inline struct au_branch *au_sbr(struct super_block *sb,
++ aufs_bindex_t bindex)
+{
-+ int err;
-+ struct super_block *sb;
++ SiMustAnyLock(sb);
++ return au_sbi(sb)->si_branch[0 + bindex];
++}
+
-+ /* all timestamps always follow the ones on the branch */
-+ /* mnt->mnt_flags |= MNT_NOATIME | MNT_NODIRATIME; */
-+ err = get_sb_nodev(fs_type, flags, raw_data, aufs_fill_super, mnt);
-+ if (!err) {
-+ sb = mnt->mnt_sb;
-+ au_mnt_init(au_sbi(sb), mnt);
-+ si_write_lock(sb);
-+ sysaufs_brs_add(sb, 0);
-+ si_write_unlock(sb);
-+ }
-+ return err;
++static inline loff_t au_xi_maxent(struct super_block *sb)
++{
++ SiMustAnyLock(sb);
++ return au_sbi(sb)->si_ximaxent;
+}
+
-+struct file_system_type aufs_fs_type = {
-+ .name = AUFS_FSTYPE,
-+ .fs_flags =
-+ FS_RENAME_DOES_D_MOVE /* a race between rename and others*/
-+ | FS_REVAL_DOT, /* for NFS branch and udba */
-+ .get_sb = aufs_get_sb,
-+ .kill_sb = generic_shutdown_super,
-+ /* no need to __module_get() and module_put(). */
-+ .owner = THIS_MODULE,
-+};
-diff -urN linux/fs/aufs/super.h linux-aufs/fs/aufs/super.h
---- linux/fs/aufs/super.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/super.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,579 @@
++#endif /* __KERNEL__ */
++#endif /* __AUFS_SUPER_H__ */
+diff -urN /usr/share/empty/fs/aufs/sysaufs.c linux/fs/aufs/sysaufs.c
+--- /usr/share/empty/fs/aufs/sysaufs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/sysaufs.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,94 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * super_block operations
-+ *
-+ * $Id$
++ * sysfs interface and lifetime management
++ * they are necessary regardless sysfs is disabled.
+ */
+
-+#ifndef __AUFS_SUPER_H__
-+#define __AUFS_SUPER_H__
-+
-+#ifdef __KERNEL__
++#include <linux/random.h>
++#include "aufs.h"
+
-+#include <linux/fs.h>
-+#include <linux/cramfs_fs.h>
-+#include <linux/kobject.h>
-+#include <linux/magic.h>
-+#include <linux/mount.h>
-+#include <linux/aufs_type.h>
-+#include "misc.h"
-+#include "wkq.h"
++unsigned long sysaufs_si_mask;
++struct kset *sysaufs_kset;
+
-+typedef ssize_t (*au_readf_t)(struct file *, char __user *, size_t, loff_t *);
-+typedef ssize_t (*au_writef_t)(struct file *, const char __user *, size_t,
-+ loff_t *);
++#define AuSiAttr(_name) { \
++ .attr = { .name = __stringify(_name), .mode = 0444 }, \
++ .show = sysaufs_si_##_name, \
++}
+
-+struct au_wbr_copyup_operations {
-+ int (*copyup)(struct dentry *dentry);
++static struct sysaufs_si_attr sysaufs_si_attr_xi_path = AuSiAttr(xi_path);
++struct attribute *sysaufs_si_attrs[] = {
++ &sysaufs_si_attr_xi_path.attr,
++ NULL,
+};
++ATTRIBUTE_GROUPS(sysaufs_si);
+
-+struct au_wbr_create_operations {
-+ int (*create)(struct dentry *dentry, int isdir);
-+ int (*init)(struct super_block *sb);
-+ int (*fin)(struct super_block *sb);
++static const struct sysfs_ops au_sbi_ops = {
++ .show = sysaufs_si_show
+};
+
-+struct au_wbr_mfs {
-+ struct mutex mfs_lock; /* protect this structure */
-+ unsigned long mfs_jiffy;
-+ unsigned long mfs_expire;
-+ aufs_bindex_t mfs_bindex;
-+
-+ unsigned long long mfsrr_bytes;
-+ unsigned long long mfsrr_watermark;
++static struct kobj_type au_sbi_ktype = {
++ .release = au_si_free,
++ .sysfs_ops = &au_sbi_ops,
++ .default_groups = sysaufs_si_groups
+};
+
-+/* sbinfo status flags */
-+/*
-+ * set true when refresh_dirs() failed at remount time.
-+ * then try refreshing dirs at access time again.
-+ * if it is false, refreshing dirs at access time is unnecesary
-+ */
-+#define AuSi_FAILED_REFRESH_DIRS 1
-+#define au_ftest_si(sbinfo, name) ((sbinfo)->au_si_status & AuSi_##name)
-+#define au_fset_si(sbinfo, name) \
-+ { (sbinfo)->au_si_status |= AuSi_##name; }
-+#define au_fclr_si(sbinfo, name) \
-+ { (sbinfo)->au_si_status &= ~AuSi_##name; }
-+
-+struct au_branch;
-+struct au_sbinfo {
-+ /* nowait tasks in the system-wide workqueue */
-+ struct au_nowait_tasks si_nowait;
-+
-+ struct au_rwsem si_rwsem;
-+
-+ /* branch management */
-+ au_gen_t si_generation;
-+
-+ /* see above flags */
-+ unsigned char au_si_status;
-+
-+ aufs_bindex_t si_bend;
-+ aufs_bindex_t si_last_br_id;
-+ struct au_branch **si_branch;
-+
-+ /* policy to select a writable branch */
-+ unsigned char si_wbr_copyup;
-+ unsigned char si_wbr_create;
-+ struct au_wbr_copyup_operations *si_wbr_copyup_ops;
-+ struct au_wbr_create_operations *si_wbr_create_ops;
-+
-+ /* round robin */
-+ atomic_t si_wbr_rr_next;
-+
-+ /* most free space */
-+ struct au_wbr_mfs si_wbr_mfs;
-+
-+ /* mount flags */
-+ /* include/asm-ia64/siginfo.h defines a macro named si_flags */
-+ unsigned int si_mntflags;
-+
-+ /* external inode number (bitmap and translation table) */
-+ au_readf_t si_xread;
-+ au_writef_t si_xwrite;
-+ struct file *si_xib;
-+ struct mutex si_xib_mtx; /* protect xib members */
-+ unsigned long *si_xib_buf;
-+ unsigned long si_xib_last_pindex;
-+ int si_xib_next_bit;
-+ /* reserved for future use */
-+ /* unsigned long long si_xib_limit; */ /* Max xib file size */
-+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ struct au_branch *si_xino_def_br;
-+#endif
-+
-+#ifdef CONFIG_AUFS_EXPORT
-+ /* i_generation */
-+ struct file *si_xigen;
-+ atomic_t si_xigen_next;
-+#endif
-+
-+ /* readdir cache time, max, in HZ */
-+ unsigned long si_rdcache;
-+
-+ /*
-+ * If the number of whiteouts are larger than si_dirwh, leave all of
-+ * them after au_whtmp_ren to reduce the cost of rmdir(2).
-+ * future fsck.aufs or kernel thread will remove them later.
-+ * Otherwise, remove all whiteouts and the dir in rmdir(2).
-+ */
-+ unsigned int si_dirwh;
-+
-+ /*
-+ * rename(2) a directory with all children.
-+ */
-+ /* reserved for future use */
-+ /* int si_rendir; */
++/* ---------------------------------------------------------------------- */
+
-+ /* pseudo_link list */ /* todo: dirty? */
-+ struct au_splhead si_plink;
++int sysaufs_si_init(struct au_sbinfo *sbinfo)
++{
++ int err;
+
-+#if defined(CONFIG_AUFS_EXPORT) && LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+ /* dirty, for export, async ops, and sysfs */
-+ spinlock_t si_mntcache_lock;
-+ struct vfsmount *si_mntcache; /* no get/put */
-+#endif
++ sbinfo->si_kobj.kset = sysaufs_kset;
++ /* cf. sysaufs_name() */
++ err = kobject_init_and_add
++ (&sbinfo->si_kobj, &au_sbi_ktype, /*&sysaufs_kset->kobj*/NULL,
++ SysaufsSiNamePrefix "%lx", sysaufs_si_id(sbinfo));
+
-+ /*
-+ * sysfs and lifetime management.
-+ * this is not a small structure and it may be a waste of memory in case
-+ * of sysfs is disabled, particulary when many aufs-es are mounted.
-+ */
-+ struct kobject si_kobj;
++ return err;
++}
+
-+#ifdef CONFIG_AUFS_ROBR
-+ /* locked vma list for mmap() */ /* todo: dirty? */
-+ struct au_splhead si_lvma;
-+#endif
++void sysaufs_fin(void)
++{
++ sysfs_remove_group(&sysaufs_kset->kobj, sysaufs_attr_group);
++ kset_unregister(sysaufs_kset);
++}
+
-+#ifdef CONFIG_AUFS_EXPORT /* reserved for future use */
-+ struct path si_xinodir;
-+#endif
++int __init sysaufs_init(void)
++{
++ int err;
+
-+ /* dirty, necessary for unmounting, sysfs and sysrq */
-+ struct super_block *si_sb;
++ do {
++ get_random_bytes(&sysaufs_si_mask, sizeof(sysaufs_si_mask));
++ } while (!sysaufs_si_mask);
+
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+ struct au_splhead si_dbg_lock[AuDbgLock_Last];
-+#endif
-+};
++ err = -EINVAL;
++ sysaufs_kset = kset_create_and_add(AUFS_NAME, NULL, fs_kobj);
++ if (unlikely(!sysaufs_kset))
++ goto out;
++ err = PTR_ERR(sysaufs_kset);
++ if (IS_ERR(sysaufs_kset))
++ goto out;
++ err = sysfs_create_group(&sysaufs_kset->kobj, sysaufs_attr_group);
++ if (unlikely(err))
++ kset_unregister(sysaufs_kset);
+
-+/* ---------------------------------------------------------------------- */
++out:
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/sysaufs.h linux/fs/aufs/sysaufs.h
+--- /usr/share/empty/fs/aufs/sysaufs.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/sysaufs.h 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,102 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+/* policy to select one among writable branches */
-+#define AuWbrCopyup(sbinfo, args...) \
-+ ((sbinfo)->si_wbr_copyup_ops->copyup(args))
-+#define AuWbrCreate(sbinfo, args...) \
-+ ((sbinfo)->si_wbr_create_ops->create(args))
++/*
++ * sysfs interface and mount lifetime management
++ */
+
-+/* flags for si_read_lock()/aufs_read_lock()/di_read_lock() */
-+#define AuLock_DW 1 /* write-lock dentry */
-+#define AuLock_IR (1 << 1) /* read-lock inode */
-+#define AuLock_IW (1 << 2) /* write-lock inode */
-+#define AuLock_FLUSH (1 << 3) /* wait for 'nowait' tasks */
-+#define AuLock_DIR (1 << 4) /* target is a dir */
-+#define au_ftest_lock(flags, name) ((flags) & AuLock_##name)
-+#define au_fset_lock(flags, name) { (flags) |= AuLock_##name; }
-+#define au_fclr_lock(flags, name) { (flags) &= ~AuLock_##name; }
++#ifndef __SYSAUFS_H__
++#define __SYSAUFS_H__
+
-+/* ---------------------------------------------------------------------- */
++#ifdef __KERNEL__
+
-+/* super.c */
-+extern struct file_system_type aufs_fs_type;
-+struct inode *au_iget_locked(struct super_block *sb, ino_t ino);
++#include <linux/sysfs.h>
++#include "module.h"
+
-+/* sbinfo.c */
-+void au_si_free(struct kobject *kobj);
-+int au_si_alloc(struct super_block *sb);
-+struct au_branch *au_sbr(struct super_block *sb, aufs_bindex_t bindex);
-+au_gen_t au_sigen_inc(struct super_block *sb);
-+int au_find_bindex(struct super_block *sb, struct au_branch *br);
++struct super_block;
++struct au_sbinfo;
+
-+void aufs_read_lock(struct dentry *dentry, int flags);
-+void aufs_read_unlock(struct dentry *dentry, int flags);
-+void aufs_write_lock(struct dentry *dentry);
-+void aufs_write_unlock(struct dentry *dentry);
-+void aufs_read_and_write_lock2(struct dentry *d1, struct dentry *d2, int isdir);
-+void aufs_read_and_write_unlock2(struct dentry *d1, struct dentry *d2);
++struct sysaufs_si_attr {
++ struct attribute attr;
++ int (*show)(struct seq_file *seq, struct super_block *sb);
++};
+
-+aufs_bindex_t au_new_br_id(struct super_block *sb);
++/* ---------------------------------------------------------------------- */
+
-+/* wbr_policy.c */
-+extern struct au_wbr_copyup_operations au_wbr_copyup_ops[];
-+extern struct au_wbr_create_operations au_wbr_create_ops[];
-+int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst);
++/* sysaufs.c */
++extern unsigned long sysaufs_si_mask;
++extern struct kset *sysaufs_kset;
++extern struct attribute *sysaufs_si_attrs[];
++int sysaufs_si_init(struct au_sbinfo *sbinfo);
++int __init sysaufs_init(void);
++void sysaufs_fin(void);
+
+/* ---------------------------------------------------------------------- */
+
-+#if defined(CONFIG_AUFS_EXPORT) && LINUX_VERSION_CODE < KERNEL_VERSION(2, 6, 26)
-+static inline void au_mnt_init(struct au_sbinfo *sbinfo, struct vfsmount *mnt)
++/* some people doesn't like to show a pointer in kernel */
++static inline unsigned long sysaufs_si_id(struct au_sbinfo *sbinfo)
+{
-+ spin_lock_init(&sbinfo->si_mntcache_lock);
-+ sbinfo->si_mntcache = mnt;
++ return sysaufs_si_mask ^ (unsigned long)sbinfo;
+}
+
-+static inline void au_mnt_reset(struct au_sbinfo *sbinfo)
-+{
-+ spin_lock(&sbinfo->si_mntcache_lock);
-+ sbinfo->si_mntcache = NULL;
-+ spin_unlock(&sbinfo->si_mntcache_lock);
-+}
-+#else
-+static inline void au_mnt_init(struct au_sbinfo *sbinfo, struct vfsmount *mnt)
++#define SysaufsSiNamePrefix "si_"
++#define SysaufsSiNameLen (sizeof(SysaufsSiNamePrefix) + 16)
++static inline void sysaufs_name(struct au_sbinfo *sbinfo, char *name)
+{
-+ /* emptr */
++ snprintf(name, SysaufsSiNameLen, SysaufsSiNamePrefix "%lx",
++ sysaufs_si_id(sbinfo));
+}
+
-+static inline void au_mnt_reset(struct au_sbinfo *sbinfo)
-+{
-+ /* emptr */
-+}
-+#endif /* EXPORT && < 2.6.26 */
++struct au_branch;
++#ifdef CONFIG_SYSFS
++/* sysfs.c */
++extern struct attribute_group *sysaufs_attr_group;
+
-+/* ---------------------------------------------------------------------- */
++int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb);
++ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr,
++ char *buf);
++long au_brinfo_ioctl(struct file *file, unsigned long arg);
++#ifdef CONFIG_COMPAT
++long au_brinfo_compat_ioctl(struct file *file, unsigned long arg);
++#endif
+
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 26)
-+static inline int au_mnt_want_write(struct vfsmount *h_mnt)
-+{
-+ return mnt_want_write(h_mnt);
-+}
++void sysaufs_br_init(struct au_branch *br);
++void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex);
++void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex);
++
++#define sysaufs_brs_init() do {} while (0)
+
-+static inline void au_mnt_drop_write(struct vfsmount *h_mnt)
-+{
-+ mnt_drop_write(h_mnt);
-+}
+#else
-+static inline int au_mnt_want_write(struct vfsmount *h_mnt)
-+{
-+ return 0;
-+}
++#define sysaufs_attr_group NULL
++
++AuStubInt0(sysaufs_si_xi_path, struct seq_file *seq, struct super_block *sb)
++AuStub(ssize_t, sysaufs_si_show, return 0, struct kobject *kobj,
++ struct attribute *attr, char *buf)
++AuStubVoid(sysaufs_br_init, struct au_branch *br)
++AuStubVoid(sysaufs_brs_add, struct super_block *sb, aufs_bindex_t bindex)
++AuStubVoid(sysaufs_brs_del, struct super_block *sb, aufs_bindex_t bindex)
+
-+static inline void au_mnt_drop_write(struct vfsmount *h_mnt)
++static inline void sysaufs_brs_init(void)
+{
-+ /* empty */
++ sysaufs_brs = 0;
+}
-+#endif
+
-+/* ---------------------------------------------------------------------- */
++#endif /* CONFIG_SYSFS */
+
-+static inline struct au_sbinfo *au_sbi(struct super_block *sb)
-+{
-+ return sb->s_fs_info;
-+}
++#endif /* __KERNEL__ */
++#endif /* __SYSAUFS_H__ */
+diff -urN /usr/share/empty/fs/aufs/sysfs.c linux/fs/aufs/sysfs.c
+--- /usr/share/empty/fs/aufs/sysfs.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/sysfs.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,374 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+static inline const char *au_sbtype(struct super_block *sb)
-+{
-+ return sb->s_type->name;
-+}
++/*
++ * sysfs interface
++ */
+
-+static inline int au_test_aufs(struct super_block *sb)
-+{
-+ return sb->s_magic == AUFS_SUPER_MAGIC;
-+}
++#include <linux/compat.h>
++#include <linux/seq_file.h>
++#include "aufs.h"
+
-+static inline int au_test_nfs(struct super_block *sb)
++#ifdef CONFIG_AUFS_FS_MODULE
++/* this entry violates the "one line per file" policy of sysfs */
++static ssize_t config_show(struct kobject *kobj, struct kobj_attribute *attr,
++ char *buf)
+{
-+#ifdef CONFIG_AUFS_BR_NFS
-+ return sb->s_magic == NFS_SUPER_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++ ssize_t err;
++ static char *conf =
++/* this file is generated at compiling */
++#include "conf.str"
++ ;
+
-+static inline int au_test_nfs4(struct super_block *sb)
-+{
-+#ifdef CONFIG_AUFS_BR_NFS_V4
-+ return au_test_nfs(sb) && !strcmp(sb->s_type->name, "nfs4");
-+#else
-+ return 0;
-+#endif
++ err = snprintf(buf, PAGE_SIZE, conf);
++ if (unlikely(err >= PAGE_SIZE))
++ err = -EFBIG;
++ return err;
+}
+
-+static inline int au_test_fuse(struct super_block *sb)
-+{
-+#ifdef CONFIG_AUFS_WORKAROUND_FUSE
-+ return sb->s_magic == FUSE_SUPER_MAGIC;
-+#else
-+ return 0;
++static struct kobj_attribute au_config_attr = __ATTR_RO(config);
+#endif
-+}
+
-+static inline int au_test_xfs(struct super_block *sb)
-+{
-+#ifdef CONFIG_AUFS_BR_XFS
-+ return sb->s_magic == XFS_SB_MAGIC;
-+#else
-+ return 0;
++static struct attribute *au_attr[] = {
++#ifdef CONFIG_AUFS_FS_MODULE
++ &au_config_attr.attr,
+#endif
-+}
++ NULL, /* need to NULL terminate the list of attributes */
++};
+
-+static inline int au_test_tmpfs(struct super_block *sb)
-+{
-+#ifdef CONFIG_TMPFS
-+ return sb->s_magic == TMPFS_MAGIC;
-+#else
-+ return 0;
-+#endif
-+}
++static struct attribute_group sysaufs_attr_group_body = {
++ .attrs = au_attr
++};
+
-+static inline int au_test_ecryptfs(struct super_block *sb)
-+{
-+#if defined(CONFIG_ECRYPT_FS) || defined(CONFIG_ECRYPT_FS_MODULE)
-+ /* why don't they use s_magic? */
-+ return !strcmp(sb->s_type->name, "ecryptfs");
-+#else
-+ return 0;
-+#endif
-+}
++struct attribute_group *sysaufs_attr_group = &sysaufs_attr_group_body;
+
+/* ---------------------------------------------------------------------- */
+
-+#ifdef CONFIG_AUFS_HINOTIFY
-+static inline void au_xino_def_br_set(struct au_branch *br,
-+ struct au_sbinfo *sbinfo)
++int sysaufs_si_xi_path(struct seq_file *seq, struct super_block *sb)
+{
-+ sbinfo->si_xino_def_br = br;
-+}
++ int err;
+
-+static inline struct au_branch *au_xino_def_br(struct au_sbinfo *sbinfo)
-+{
-+ return sbinfo->si_xino_def_br;
-+}
-+#else
-+static inline void au_xino_def_br_set(struct au_branch *br,
-+ struct au_sbinfo *sbinfo)
-+{
-+ /* empty */
++ SiMustAnyLock(sb);
++
++ err = 0;
++ if (au_opt_test(au_mntflags(sb), XINO)) {
++ err = au_xino_path(seq, au_sbi(sb)->si_xib);
++ seq_putc(seq, '\n');
++ }
++ return err;
+}
+
-+static inline struct au_branch *au_xino_def_br(struct au_sbinfo *sbinfo)
++/*
++ * the lifetime of branch is independent from the entry under sysfs.
++ * sysfs handles the lifetime of the entry, and never call ->show() after it is
++ * unlinked.
++ */
++static int sysaufs_si_br(struct seq_file *seq, struct super_block *sb,
++ aufs_bindex_t bindex, int idx)
+{
-+ return NULL;
-+}
-+#endif
++ int err;
++ struct path path;
++ struct dentry *root;
++ struct au_branch *br;
++ au_br_perm_str_t perm;
+
-+/* ---------------------------------------------------------------------- */
++ AuDbg("b%d\n", bindex);
+
-+#ifdef CONFIG_AUFS_EXPORT
-+void au_export_init(struct super_block *sb);
++ err = 0;
++ root = sb->s_root;
++ di_read_lock_parent(root, !AuLock_IR);
++ br = au_sbr(sb, bindex);
+
-+static inline int au_test_nfsd(struct task_struct *tsk)
-+{
-+ return !tsk->mm && !strcmp(tsk->comm, "nfsd");
-+}
++ switch (idx) {
++ case AuBrSysfs_BR:
++ path.mnt = au_br_mnt(br);
++ path.dentry = au_h_dptr(root, bindex);
++ err = au_seq_path(seq, &path);
++ if (!err) {
++ au_optstr_br_perm(&perm, br->br_perm);
++ seq_printf(seq, "=%s\n", perm.a);
++ }
++ break;
++ case AuBrSysfs_BRID:
++ seq_printf(seq, "%d\n", br->br_id);
++ break;
++ }
++ di_read_unlock(root, !AuLock_IR);
++ if (unlikely(err || seq_has_overflowed(seq)))
++ err = -E2BIG;
+
-+static inline void au_export_put(struct au_sbinfo *sbinfo)
-+{
-+ path_put(&sbinfo->si_xinodir);
++ return err;
+}
+
-+int au_xigen_inc(struct inode *inode);
-+int au_xigen_new(struct inode *inode);
-+int au_xigen_set(struct super_block *sb, struct file *base);
-+void au_xigen_clr(struct super_block *sb);
-+#else
-+static inline void au_export_init(struct super_block *sb)
-+{
-+ /* nothing */
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline int au_test_nfsd(struct task_struct *tsk)
++static struct seq_file *au_seq(char *p, ssize_t len)
+{
-+ return 0;
-+}
++ struct seq_file *seq;
+
-+static inline void au_export_put(struct au_sbinfo *sbinfo)
-+{
-+ /* nothing */
-+}
++ seq = kzalloc(sizeof(*seq), GFP_NOFS);
++ if (seq) {
++ /* mutex_init(&seq.lock); */
++ seq->buf = p;
++ seq->size = len;
++ return seq; /* success */
++ }
+
-+static inline int au_xigen_inc(struct inode *inode)
-+{
-+ return 0;
++ seq = ERR_PTR(-ENOMEM);
++ return seq;
+}
+
-+static inline int au_xigen_new(struct inode *inode)
-+{
-+ return 0;
-+}
++#define SysaufsBr_PREFIX "br"
++#define SysaufsBrid_PREFIX "brid"
+
-+static inline int au_xigen_set(struct super_block *sb, struct file *base)
++/* todo: file size may exceed PAGE_SIZE */
++ssize_t sysaufs_si_show(struct kobject *kobj, struct attribute *attr,
++ char *buf)
+{
-+ return 0;
-+}
++ ssize_t err;
++ int idx;
++ long l;
++ aufs_bindex_t bbot;
++ struct au_sbinfo *sbinfo;
++ struct super_block *sb;
++ struct seq_file *seq;
++ char *name;
++ struct attribute **cattr;
+
-+static inline void au_xigen_clr(struct super_block *sb)
-+{
-+ /* empty */
-+}
-+#endif /* CONFIG_AUFS_EXPORT */
++ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
++ sb = sbinfo->si_sb;
+
-+#ifdef CONFIG_AUFS_ROBR
-+static inline int au_test_nested(struct super_block *h_sb)
-+{
-+ return 0;
-+}
++ /*
++ * prevent a race condition between sysfs and aufs.
++ * for instance, sysfs_file_read() calls sysfs_get_active_two() which
++ * prohibits maintaining the sysfs entries.
++ * hew we acquire read lock after sysfs_get_active_two().
++ * on the other hand, the remount process may maintain the sysfs/aufs
++ * entries after acquiring write lock.
++ * it can cause a deadlock.
++ * simply we gave up processing read here.
++ */
++ err = -EBUSY;
++ if (unlikely(!si_noflush_read_trylock(sb)))
++ goto out;
+
-+static inline void au_robr_lvma_init(struct au_sbinfo *sbinfo)
-+{
-+ au_spl_init(&sbinfo->si_lvma);
-+}
-+#else
-+static inline int au_test_nested(struct super_block *h_sb)
-+{
-+ int err = 0;
-+ if (unlikely(au_test_aufs(h_sb))) {
-+ err = -EINVAL;
-+ AuTraceErr(err);
++ seq = au_seq(buf, PAGE_SIZE);
++ err = PTR_ERR(seq);
++ if (IS_ERR(seq))
++ goto out_unlock;
++
++ name = (void *)attr->name;
++ cattr = sysaufs_si_attrs;
++ while (*cattr) {
++ if (!strcmp(name, (*cattr)->name)) {
++ err = container_of(*cattr, struct sysaufs_si_attr, attr)
++ ->show(seq, sb);
++ goto out_seq;
++ }
++ cattr++;
+ }
-+ return err;
-+}
+
-+static inline void au_robr_lvma_init(struct au_sbinfo *sbinfo)
-+{
-+ /* empty */
++ if (!strncmp(name, SysaufsBrid_PREFIX,
++ sizeof(SysaufsBrid_PREFIX) - 1)) {
++ idx = AuBrSysfs_BRID;
++ name += sizeof(SysaufsBrid_PREFIX) - 1;
++ } else if (!strncmp(name, SysaufsBr_PREFIX,
++ sizeof(SysaufsBr_PREFIX) - 1)) {
++ idx = AuBrSysfs_BR;
++ name += sizeof(SysaufsBr_PREFIX) - 1;
++ } else
++ BUG();
++
++ err = kstrtol(name, 10, &l);
++ if (!err) {
++ bbot = au_sbbot(sb);
++ if (l <= bbot)
++ err = sysaufs_si_br(seq, sb, (aufs_bindex_t)l, idx);
++ else
++ err = -ENOENT;
++ }
++
++out_seq:
++ if (!err) {
++ err = seq->count;
++ /* sysfs limit */
++ if (unlikely(err == PAGE_SIZE))
++ err = -EFBIG;
++ }
++ au_kfree_rcu(seq);
++out_unlock:
++ si_read_unlock(sb);
++out:
++ return err;
+}
-+#endif /* CONFIG_AUFS_ROBR */
+
+/* ---------------------------------------------------------------------- */
+
-+/* lock superblock. mainly for entry point functions */
-+/*
-+ * si_do_noflush_read_lock, si_do_noflush_write_lock,
-+ * si_do_read_unlock, si_do_write_unlock, si_do_downgrade_lock
-+ */
-+AuSimpleLockRwsemFuncs(si_do_noflush, struct super_block *sb,
-+ au_sbi(sb)->si_rwsem);
-+AuSimpleUnlockRwsemFuncs(si_do, struct super_block *sb, au_sbi(sb)->si_rwsem);
-+
-+static inline void si_noflush_read_lock(struct super_block *sb)
++static int au_brinfo(struct super_block *sb, union aufs_brinfo __user *arg)
+{
-+ au_dbg_locking_si_reg(sb, 0);
-+ si_do_noflush_read_lock(sb);
-+ au_dbg_locking_si_unreg(sb, 0);
-+ au_dbg_locked_si_reg(sb, 0);
-+}
++ int err;
++ int16_t brid;
++ aufs_bindex_t bindex, bbot;
++ size_t sz;
++ char *buf;
++ struct seq_file *seq;
++ struct au_branch *br;
++
++ si_read_lock(sb, AuLock_FLUSH);
++ bbot = au_sbbot(sb);
++ err = bbot + 1;
++ if (!arg)
++ goto out;
++
++ err = -ENOMEM;
++ buf = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!buf))
++ goto out;
+
-+static inline void si_noflush_write_lock(struct super_block *sb)
-+{
-+ au_dbg_locking_si_reg(sb, 1);
-+ si_do_noflush_write_lock(sb);
-+ au_dbg_locking_si_unreg(sb, 1);
-+ au_dbg_locked_si_reg(sb, 1);
-+}
++ seq = au_seq(buf, PAGE_SIZE);
++ err = PTR_ERR(seq);
++ if (IS_ERR(seq))
++ goto out_buf;
+
-+static inline int si_noflush_read_trylock(struct super_block *sb)
-+{
-+ int locked;
++ sz = sizeof(*arg) - offsetof(union aufs_brinfo, path);
++ for (bindex = 0; bindex <= bbot; bindex++, arg++) {
++ /* VERIFY_WRITE */
++ err = !access_ok(arg, sizeof(*arg));
++ if (unlikely(err))
++ break;
+
-+ au_dbg_locking_si_reg(sb, 0);
-+ locked = si_do_noflush_read_trylock(sb);
-+ au_dbg_locking_si_unreg(sb, 0);
-+ if (locked)
-+ au_dbg_locked_si_reg(sb, 0);
-+ return locked;
-+}
++ br = au_sbr(sb, bindex);
++ brid = br->br_id;
++ BUILD_BUG_ON(sizeof(brid) != sizeof(arg->id));
++ err = __put_user(brid, &arg->id);
++ if (unlikely(err))
++ break;
+
-+static inline int si_noflush_write_trylock(struct super_block *sb)
-+{
-+ int locked;
++ BUILD_BUG_ON(sizeof(br->br_perm) != sizeof(arg->perm));
++ err = __put_user(br->br_perm, &arg->perm);
++ if (unlikely(err))
++ break;
+
-+ au_dbg_locking_si_reg(sb, 1);
-+ locked = si_do_noflush_write_trylock(sb);
-+ au_dbg_locking_si_unreg(sb, 1);
-+ if (locked)
-+ au_dbg_locked_si_reg(sb, 1);
-+ return locked;
-+}
++ err = au_seq_path(seq, &br->br_path);
++ if (unlikely(err))
++ break;
++ seq_putc(seq, '\0');
++ if (!seq_has_overflowed(seq)) {
++ err = copy_to_user(arg->path, seq->buf, seq->count);
++ seq->count = 0;
++ if (unlikely(err))
++ break;
++ } else {
++ err = -E2BIG;
++ goto out_seq;
++ }
++ }
++ if (unlikely(err))
++ err = -EFAULT;
+
-+static inline void si_read_unlock(struct super_block *sb)
-+{
-+ si_do_read_unlock(sb);
-+ au_dbg_locked_si_unreg(sb, 0);
++out_seq:
++ au_kfree_rcu(seq);
++out_buf:
++ free_page((unsigned long)buf);
++out:
++ si_read_unlock(sb);
++ return err;
+}
+
-+static inline void si_write_unlock(struct super_block *sb)
++long au_brinfo_ioctl(struct file *file, unsigned long arg)
+{
-+ si_do_write_unlock(sb);
-+ au_dbg_locked_si_unreg(sb, 1);
++ return au_brinfo(file->f_path.dentry->d_sb, (void __user *)arg);
+}
+
-+static inline void si_downgrade_lock(struct super_block *sb)
++#ifdef CONFIG_COMPAT
++long au_brinfo_compat_ioctl(struct file *file, unsigned long arg)
+{
-+ si_do_downgrade_lock(sb);
++ return au_brinfo(file->f_path.dentry->d_sb, compat_ptr(arg));
+}
++#endif
+
-+static inline void si_read_lock(struct super_block *sb, int flags)
-+{
-+ if (/* !au_test_nfsd(current) && */au_ftest_lock(flags, FLUSH))
-+ au_nwt_flush(&au_sbi(sb)->si_nowait);
-+ si_noflush_read_lock(sb);
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline void si_write_lock(struct super_block *sb)
++void sysaufs_br_init(struct au_branch *br)
+{
-+ //WARN_ON(au_test_nfsd(current));
-+ au_nwt_flush(&au_sbi(sb)->si_nowait);
-+ si_noflush_write_lock(sb);
-+}
++ int i;
++ struct au_brsysfs *br_sysfs;
++ struct attribute *attr;
+
-+static inline int si_read_trylock(struct super_block *sb, int flags)
-+{
-+ if (/* !au_test_nfsd(current) && */au_ftest_lock(flags, FLUSH))
-+ au_nwt_flush(&au_sbi(sb)->si_nowait);
-+ return si_noflush_read_trylock(sb);
++ br_sysfs = br->br_sysfs;
++ for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) {
++ attr = &br_sysfs->attr;
++ sysfs_attr_init(attr);
++ attr->name = br_sysfs->name;
++ attr->mode = 0444;
++ br_sysfs++;
++ }
+}
+
-+static inline int si_write_trylock(struct super_block *sb, int flags)
++void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ if (/* !au_test_nfsd(current) && */au_ftest_lock(flags, FLUSH))
-+ au_nwt_flush(&au_sbi(sb)->si_nowait);
-+ return si_noflush_write_trylock(sb);
-+}
-+
-+/* to debug easier, do not make them inlined functions */
-+#define SiMustReadLock(sb) AuRwMustReadLock(&au_sbi(sb)->si_rwsem)
-+#define SiMustWriteLock(sb) AuRwMustWriteLock(&au_sbi(sb)->si_rwsem)
-+#define SiMustAnyLock(sb) AuRwMustAnyLock(&au_sbi(sb)->si_rwsem)
++ struct au_branch *br;
++ struct kobject *kobj;
++ struct au_brsysfs *br_sysfs;
++ int i;
++ aufs_bindex_t bbot;
+
-+/* ---------------------------------------------------------------------- */
++ if (!sysaufs_brs)
++ return;
+
-+static inline aufs_bindex_t au_sbend(struct super_block *sb)
-+{
-+ SiMustAnyLock(sb);
-+ return au_sbi(sb)->si_bend;
++ kobj = &au_sbi(sb)->si_kobj;
++ bbot = au_sbbot(sb);
++ for (; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ br_sysfs = br->br_sysfs;
++ for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) {
++ sysfs_remove_file(kobj, &br_sysfs->attr);
++ br_sysfs++;
++ }
++ }
+}
+
-+static inline unsigned int au_mntflags(struct super_block *sb)
++void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ SiMustAnyLock(sb);
-+ return au_sbi(sb)->si_mntflags;
-+}
++ int err, i;
++ aufs_bindex_t bbot;
++ struct kobject *kobj;
++ struct au_branch *br;
++ struct au_brsysfs *br_sysfs;
+
-+static inline au_gen_t au_sigen(struct super_block *sb)
-+{
-+ SiMustAnyLock(sb);
-+ return au_sbi(sb)->si_generation;
-+}
++ if (!sysaufs_brs)
++ return;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_SUPER_H__ */
-diff -urN linux/fs/aufs/sysaufs.c linux-aufs/fs/aufs/sysaufs.c
---- linux/fs/aufs/sysaufs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/sysaufs.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,109 @@
++ kobj = &au_sbi(sb)->si_kobj;
++ bbot = au_sbbot(sb);
++ for (; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ br_sysfs = br->br_sysfs;
++ snprintf(br_sysfs[AuBrSysfs_BR].name, sizeof(br_sysfs->name),
++ SysaufsBr_PREFIX "%d", bindex);
++ snprintf(br_sysfs[AuBrSysfs_BRID].name, sizeof(br_sysfs->name),
++ SysaufsBrid_PREFIX "%d", bindex);
++ for (i = 0; i < ARRAY_SIZE(br->br_sysfs); i++) {
++ err = sysfs_create_file(kobj, &br_sysfs->attr);
++ if (unlikely(err))
++ pr_warn("failed %s under sysfs(%d)\n",
++ br_sysfs->name, err);
++ br_sysfs++;
++ }
++ }
++}
+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 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,149 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * sysfs interface and lifetime management
-+ * they are necessary regardless sysfs is disabled.
-+ *
-+ * $Id$
++ * magic sysrq handler
+ */
+
-+#include <linux/fs.h>
-+#include <linux/module.h>
-+#include <linux/random.h>
-+#include <linux/sysfs.h>
++/* #include <linux/sysrq.h> */
++#include <linux/writeback.h>
+#include "aufs.h"
+
+/* ---------------------------------------------------------------------- */
+
-+unsigned long au_si_mask;
++static void sysrq_sb(struct super_block *sb)
++{
++ char *plevel;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
++ struct hlist_bl_head *files;
++ struct hlist_bl_node *pos;
++ struct au_finfo *finfo;
++ struct inode *i;
+
-+/* ---------------------------------------------------------------------- */
++ plevel = au_plevel;
++ au_plevel = KERN_WARNING;
+
-+struct kset *au_kset;
++ /* since we define pr_fmt, call printk directly */
++#define pr(str) printk(KERN_WARNING AUFS_NAME ": " str)
+
-+#define AuSbiAttr(_name) { \
-+ .attr = { .name = __stringify(_name), .mode = 0444 }, \
-+ .show = sysaufs_sbi_##_name, \
-+}
++ sbinfo = au_sbi(sb);
++ printk(KERN_WARNING "si=%lx\n", sysaufs_si_id(sbinfo));
++ pr("superblock\n");
++ au_dpri_sb(sb);
+
-+static struct au_sbi_attr au_sbi_attr_xino = AuSbiAttr(xino);
-+#ifdef CONFIG_AUFS_EXPORT
-+static struct au_sbi_attr au_sbi_attr_xigen = AuSbiAttr(xigen);
-+#endif
-+struct attribute *au_sbi_attrs[] = {
-+ &au_sbi_attr_xino.attr,
-+#ifdef CONFIG_AUFS_EXPORT
-+ &au_sbi_attr_xigen.attr,
++#if 0 /* reserved */
++ do {
++ int err, i, j, ndentry;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++
++ err = au_dpages_init(&dpages, GFP_ATOMIC);
++ if (unlikely(err))
++ break;
++ err = au_dcsub_pages(&dpages, sb->s_root, NULL, NULL);
++ if (!err)
++ for (i = 0; i < dpages.ndpage; i++) {
++ dpage = dpages.dpages + i;
++ ndentry = dpage->ndentry;
++ for (j = 0; j < ndentry; j++)
++ au_dpri_dentry(dpage->dentries[j]);
++ }
++ au_dpages_free(&dpages);
++ } while (0);
+#endif
-+ NULL,
-+};
+
-+static struct sysfs_ops au_sbi_ops = {
-+ .show = sysaufs_sbi_show
-+};
++ pr("isolated inode\n");
++ spin_lock(&sb->s_inode_list_lock);
++ list_for_each_entry(i, &sb->s_inodes, i_sb_list) {
++ spin_lock(&i->i_lock);
++ if (hlist_empty(&i->i_dentry))
++ au_dpri_inode(i);
++ spin_unlock(&i->i_lock);
++ }
++ spin_unlock(&sb->s_inode_list_lock);
+
-+static struct kobj_type au_sbi_ktype = {
-+ .release = au_si_free,
-+ .sysfs_ops = &au_sbi_ops,
-+ .default_attrs = au_sbi_attrs
-+};
++ pr("files\n");
++ files = &au_sbi(sb)->si_files;
++ hlist_bl_lock(files);
++ hlist_bl_for_each_entry(finfo, pos, files, fi_hlist) {
++ umode_t mode;
++
++ file = finfo->fi_file;
++ mode = file_inode(file)->i_mode;
++ if (!special_file(mode))
++ au_dpri_file(file);
++ }
++ hlist_bl_unlock(files);
++ pr("done\n");
++
++#undef pr
++ au_plevel = plevel;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+int sysaufs_si_init(struct au_sbinfo *sbinfo)
++/* module parameter */
++static char *aufs_sysrq_key = "a";
++module_param_named(sysrq, aufs_sysrq_key, charp, 0444);
++MODULE_PARM_DESC(sysrq, "MagicSysRq key for " AUFS_NAME);
++
++static void au_sysrq(int key __maybe_unused)
+{
-+ int err;
++ struct au_sbinfo *sbinfo;
++ struct hlist_bl_node *pos;
+
-+ sbinfo->si_kobj.kset = au_kset;
-+ /* some people doesn't like to show a pointer in kernel */
-+ err = kobject_init_and_add(&sbinfo->si_kobj, &au_sbi_ktype,
-+ NULL/*&au_kset->kobj*/,
-+ SysaufsSb_PREFIX "%lx",
-+ au_si_mask ^ (unsigned long)sbinfo);
-+ AuTraceErr(err);
-+ return err;
++ lockdep_off();
++ au_sbilist_lock();
++ hlist_bl_for_each_entry(sbinfo, pos, &au_sbilist, si_list)
++ sysrq_sb(sbinfo->si_sb);
++ au_sbilist_unlock();
++ lockdep_on();
+}
+
++static struct sysrq_key_op au_sysrq_op = {
++ .handler = au_sysrq,
++ .help_msg = "Aufs",
++ .action_msg = "Aufs",
++ .enable_mask = SYSRQ_ENABLE_DUMP
++};
+
+/* ---------------------------------------------------------------------- */
+
-+void sysaufs_fin(void)
++int __init au_sysrq_init(void)
+{
-+ sysfs_remove_group(&au_kset->kobj, au_attr_group);
-+ kset_unregister(au_kset);
++ int err;
++ char key;
++
++ err = -1;
++ key = *aufs_sysrq_key;
++ if ('a' <= key && key <= 'z')
++ err = register_sysrq_key(key, &au_sysrq_op);
++ if (unlikely(err))
++ pr_err("err %d, sysrq=%c\n", err, key);
++ return err;
+}
+
-+int __init sysaufs_init(void)
++void au_sysrq_fin(void)
+{
+ int err;
+
-+ get_random_bytes(&au_si_mask, sizeof(au_si_mask));
-+
-+ au_kset = kset_create_and_add(AUFS_NAME, NULL, fs_kobj);
-+ err = PTR_ERR(au_kset);
-+ if (IS_ERR(au_kset))
-+ goto out;
-+ err = sysfs_create_group(&au_kset->kobj, au_attr_group);
++ err = unregister_sysrq_key(*aufs_sysrq_key, &au_sysrq_op);
+ if (unlikely(err))
-+ kset_unregister(au_kset);
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ pr_err("err %d (ignored)\n", err);
+}
-diff -urN linux/fs/aufs/sysaufs.h linux-aufs/fs/aufs/sysaufs.h
---- linux/fs/aufs/sysaufs.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/sysaufs.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,119 @@
+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 2022-12-17 09:21:34.799855195 +0100
+@@ -0,0 +1,896 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * sysfs interface and lifetime management
-+ *
-+ * $Id$
++ * virtual or vertical directory
+ */
+
-+#ifndef __SYSAUFS_H__
-+#define __SYSAUFS_H__
-+
-+#ifdef __KERNEL__
-+
-+#include <linux/fs.h>
-+#include <linux/sysfs.h>
-+#include "module.h"
-+#include "super.h"
-+
-+#define SysaufsSb_PREFIX "si_" /* followed by %p */
-+
-+struct au_sbi_attr {
-+ struct attribute attr;
-+ int (*show)(struct seq_file *seq, struct super_block *sb);
-+};
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/* sysaufs.c */
-+extern unsigned long au_si_mask;
-+extern struct kset *au_kset;
-+extern struct attribute *au_sbi_attrs[];
-+int sysaufs_si_init(struct au_sbinfo *sbinfo);
-+int __init sysaufs_init(void);
-+void sysaufs_fin(void);
-+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_branch;
-+#ifdef CONFIG_SYSFS
-+/* sysfs.c */
-+extern struct attribute_group *au_attr_group;
-+extern struct kobj_type *au_ktype;
-+
-+int sysaufs_sbi_xino(struct seq_file *seq, struct super_block *sb);
-+#ifdef CONFIG_AUFS_EXPORT
-+int sysaufs_sbi_xigen(struct seq_file *seq, struct super_block *sb);
-+#endif
-+int sysaufs_sbi_mntpnt1(struct seq_file *seq, struct super_block *sb);
-+ssize_t sysaufs_sbi_show(struct kobject *kobj, struct attribute *attr,
-+ char *buf);
-+
-+void sysaufs_br_init(struct au_branch *br);
-+void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex);
-+void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex);
-+#else
-+#define au_attr_group NULL
-+#define au_ktype NULL
-+
-+static inline
-+int sysaufs_sbi_xino(struct seq_file *seq, struct super_block *sb)
-+{
-+ return 0;
-+}
-+
-+#ifdef CONFIG_AUFS_EXPORT
-+static inline
-+int sysaufs_sbi_xigen(struct seq_file *seq, struct super_block *sb)
-+{
-+ return 0;
-+}
-+#endif
++#include <linux/iversion.h>
++#include "aufs.h"
+
-+static inline
-+int sysaufs_sbi_mntpnt1(struct seq_file *seq, struct super_block *sb)
++static unsigned int calc_size(int nlen)
+{
-+ return 0;
++ return ALIGN(sizeof(struct au_vdir_de) + nlen, sizeof(ino_t));
+}
+
-+static inline
-+ssize_t sysaufs_sbi_show(struct kobject *kobj, struct attribute *attr,
-+ char *buf)
++static int set_deblk_end(union au_vdir_deblk_p *p,
++ union au_vdir_deblk_p *deblk_end)
+{
-+ return 0;
++ if (calc_size(0) <= deblk_end->deblk - p->deblk) {
++ p->de->de_str.len = 0;
++ /* smp_mb(); */
++ return 0;
++ }
++ return -1; /* error */
+}
+
-+static inline void sysaufs_br_init(struct au_branch *br)
++/* returns true or false */
++static int is_deblk_end(union au_vdir_deblk_p *p,
++ union au_vdir_deblk_p *deblk_end)
+{
-+ /* empty */
++ if (calc_size(0) <= deblk_end->deblk - p->deblk)
++ return !p->de->de_str.len;
++ return 1;
+}
+
-+static inline void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex)
++static unsigned char *last_deblk(struct au_vdir *vdir)
+{
-+ /* nothing */
++ return vdir->vd_deblk[vdir->vd_nblk - 1];
+}
+
-+static inline void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex)
++/* ---------------------------------------------------------------------- */
++
++/* estimate the appropriate size for name hash table */
++unsigned int au_rdhash_est(loff_t sz)
+{
-+ /* nothing */
-+}
-+#endif /* CONFIG_SYSFS */
++ unsigned int n;
+
-+#endif /* __KERNEL__ */
-+#endif /* __SYSAUFS_H__ */
-diff -urN linux/fs/aufs/sysfs.c linux-aufs/fs/aufs/sysfs.c
---- linux/fs/aufs/sysfs.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/sysfs.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,533 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ n = UINT_MAX;
++ sz >>= 10;
++ if (sz < n)
++ n = sz;
++ if (sz < AUFS_RDHASH_DEF)
++ n = AUFS_RDHASH_DEF;
++ /* pr_info("n %u\n", n); */
++ return n;
++}
+
+/*
-+ * sysfs interface
-+ *
-+ * $Id$
++ * the allocated memory has to be freed by
++ * au_nhash_wh_free() or au_nhash_de_free().
+ */
++int au_nhash_alloc(struct au_nhash *nhash, unsigned int num_hash, gfp_t gfp)
++{
++ struct hlist_head *head;
++ unsigned int u;
++ size_t sz;
++
++ sz = sizeof(*nhash->nh_head) * num_hash;
++ head = kmalloc(sz, gfp);
++ if (head) {
++ nhash->nh_num = num_hash;
++ nhash->nh_head = head;
++ for (u = 0; u < num_hash; u++)
++ INIT_HLIST_HEAD(head++);
++ return 0; /* success */
++ }
+
-+#include <linux/fs.h>
-+#include <linux/module.h>
-+#include <linux/seq_file.h>
-+#include <linux/sysfs.h>
-+#include "aufs.h"
-+
++ return -ENOMEM;
++}
+
-+#ifdef CONFIG_AUFS_LOCAL
-+static ssize_t config_show(struct kobject *kobj, struct kobj_attribute *attr,
-+ char *buf)
++static void nhash_count(struct hlist_head *head)
+{
-+#define conf_bool(name) "CONFIG_AUFS_" #name "=y\n"
-+ static const char opt[] =
-+#ifdef CONFIG_AUFS
-+ "CONFIG_AUFS=y\n"
-+#else
-+ "CONFIG_AUFS=m\n"
-+#endif
-+#ifdef CONFIG_AUFS_BRANCH_MAX_127
-+ conf_bool(BRANCH_MAX_127)
-+#elif defined(CONFIG_AUFS_BRANCH_MAX_511)
-+ conf_bool(BRANCH_MAX_511)
-+#elif defined(CONFIG_AUFS_BRANCH_MAX_1023)
-+ conf_bool(BRANCH_MAX_1023)
-+#elif defined(CONFIG_AUFS_BRANCH_MAX_32767)
-+ conf_bool(BRANCH_MAX_32767)
-+#endif
-+#ifdef CONFIG_AUFS_STAT
-+ conf_bool(STAT)
-+#endif
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ conf_bool(HINOTIFY)
-+#endif
-+#ifdef CONFIG_AUFS_EXPORT
-+ conf_bool(EXPORT)
-+#endif
-+#ifdef CONFIG_AUFS_INO_T_64
-+ conf_bool(INO_T_64)
-+#endif
-+#ifdef CONFIG_AUFS_ROBR
-+ conf_bool(ROBR)
-+#endif
-+#ifdef CONFIG_AUFS_DLGT
-+ conf_bool(DLGT)
-+#endif
-+#ifdef CONFIG_AUFS_HIN_OR_DLGT
-+ conf_bool(HIN_OR_DLGT)
-+#endif
-+#ifdef CONFIG_AUFS_SHWH
-+ conf_bool(SHWH)
-+#endif
-+#ifdef CONFIG_AUFS_RR_SQUASHFS
-+ conf_bool(RR_SQUASHFS)
-+#endif
-+#ifdef CONFIG_AUFS_SEC_PERM_PATCH
-+ conf_bool(SEC_PERM_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_SPLICE_PATCH
-+ conf_bool(SPLICE_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_LHASH_PATCH
-+ conf_bool(LHASH_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_PUT_FILP_PATCH
-+ conf_bool(PUT_FILP_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_BR_NFS
-+ conf_bool(BR_NFS)
-+#endif
-+#ifdef CONFIG_AUFS_BR_NFS_V4
-+ conf_bool(BR_NFS_V4)
-+#endif
-+#ifdef CONFIG_AUFS_BR_XFS
-+ conf_bool(BR_XFS)
-+#endif
-+#ifdef CONFIG_AUFS_FSYNC_SUPER_PATCH
-+ conf_bool(FSYNC_SUPER_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_DENY_WRITE_ACCESS_PATCH
-+ conf_bool(DENY_WRITE_ACCESS_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_KSIZE_PATCH
-+ conf_bool(KSIZE_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_WORKAROUND_FUSE
-+ conf_bool(WORKAROUND_FUSE)
-+#endif
-+#ifdef CONFIG_AUFS_GETATTR
-+ conf_bool(GETATTR)
-+#endif
-+#ifdef CONFIG_AUFS_DEBUG
-+ conf_bool(DEBUG)
-+#endif
-+#ifdef CONFIG_AUFS_MAGIC_SYSRQ
-+ conf_bool(MAGIC_SYSRQ)
-+#endif
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+ conf_bool(DEBUG_LOCK)
-+#endif
-+#ifdef CONFIG_AUFS_COMPAT
-+ conf_bool(COMPAT)
-+#endif
-+#ifdef CONFIG_AUFS_UNIONFS22_PATCH
-+ conf_bool(UNIONFS22_PATCH)
-+#endif
-+#ifdef CONFIG_AUFS_UNIONFS23_PATCH
-+ conf_bool(UNIONFS23_PATCH)
++#if 0 /* debugging */
++ unsigned long n;
++ struct hlist_node *pos;
++
++ n = 0;
++ hlist_for_each(pos, head)
++ n++;
++ pr_info("%lu\n", n);
+#endif
-+ ;
-+#undef conf_bool
++}
+
-+ char *p = buf;
-+ const char *end = buf + PAGE_SIZE;
++static void au_nhash_wh_do_free(struct hlist_head *head)
++{
++ struct au_vdir_wh *pos;
++ struct hlist_node *node;
+
-+ p += snprintf(p, end - p, "%s", opt);
-+#ifdef DbgUdbaRace
-+ if (p < end)
-+ p += snprintf(p, end - p, "DbgUdbaRace=%d\n", DbgUdbaRace);
-+#endif
-+ if (p < end)
-+ return p - buf;
-+ else
-+ return -EFBIG;
++ hlist_for_each_entry_safe(pos, node, head, wh_hash)
++ au_kfree_rcu(pos);
+}
+
-+static struct kobj_attribute au_config_attr = __ATTR_RO(config);
-+#endif
++static void au_nhash_de_do_free(struct hlist_head *head)
++{
++ struct au_vdir_dehstr *pos;
++ struct hlist_node *node;
++
++ hlist_for_each_entry_safe(pos, node, head, hash)
++ au_cache_free_vdir_dehstr(pos);
++}
+
-+#ifdef CONFIG_AUFS_STAT
-+static ssize_t stat_show(struct kobject *kobj, struct kobj_attribute *attr,
-+ char *buf)
++static void au_nhash_do_free(struct au_nhash *nhash,
++ void (*free)(struct hlist_head *head))
+{
-+ char *p = buf;
-+ const char *end = buf + PAGE_SIZE;
-+ int i;
++ unsigned int n;
++ struct hlist_head *head;
+
-+ p += snprintf(p, end - p, "wkq max_busy:");
-+ for (i = 0; p < end && i < aufs_nwkq; i++)
-+ p += snprintf(p, end - p, " %u", au_wkq[i].max_busy);
-+ if (p < end)
-+ p += snprintf(p, end - p, ", %u(generic)\n",
-+ au_wkq[aufs_nwkq].max_busy);
++ n = nhash->nh_num;
++ if (!n)
++ return;
+
-+ if (p < end)
-+ return p - buf;
-+ else
-+ return -EFBIG;
++ head = nhash->nh_head;
++ while (n-- > 0) {
++ nhash_count(head);
++ free(head++);
++ }
++ au_kfree_try_rcu(nhash->nh_head);
+}
+
-+static struct kobj_attribute au_stat_attr = __ATTR_RO(stat);
-+#endif
-+
-+#ifdef CONFIG_AUFS_DEBUG
-+static ssize_t debug_show(struct kobject *kobj, struct kobj_attribute *attr,
-+ char *buf)
++void au_nhash_wh_free(struct au_nhash *whlist)
+{
-+ return sprintf(buf, "%d\n", au_debug_test());
++ au_nhash_do_free(whlist, au_nhash_wh_do_free);
+}
+
-+static ssize_t debug_store(struct kobject *kobj, struct kobj_attribute *attr,
-+ const char *buf, size_t sz)
++static void au_nhash_de_free(struct au_nhash *delist)
+{
-+ LKTRTrace("%.*s\n", (unsigned int)sz, buf);
++ au_nhash_do_free(delist, au_nhash_de_do_free);
++}
+
-+ if (unlikely(!sz || (*buf != '0' && *buf != '1')))
-+ return -EOPNOTSUPP;
++/* ---------------------------------------------------------------------- */
+
-+ if (*buf == '0')
-+ au_debug_off();
-+ else if (*buf == '1')
-+ au_debug_on();
-+ return sz;
++int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt,
++ int limit)
++{
++ int num;
++ unsigned int u, n;
++ struct hlist_head *head;
++ struct au_vdir_wh *pos;
++
++ num = 0;
++ n = whlist->nh_num;
++ head = whlist->nh_head;
++ for (u = 0; u < n; u++, head++)
++ hlist_for_each_entry(pos, head, wh_hash)
++ if (pos->wh_bindex == btgt && ++num > limit)
++ return 1;
++ return 0;
+}
+
-+static struct kobj_attribute au_debug_attr = __ATTR(debug, S_IRUGO | S_IWUSR,
-+ debug_show, debug_store);
-+#endif
-+
-+static struct attribute *au_attr[] = {
-+#ifdef CONFIG_AUFS_LOCAL
-+ &au_config_attr.attr,
-+#endif
-+#ifdef CONFIG_AUFS_STAT
-+ &au_stat_attr.attr,
-+#endif
-+#ifdef CONFIG_AUFS_DEBUG
-+ &au_debug_attr.attr,
-+#endif
-+ NULL, /* need to NULL terminate the list of attributes */
-+};
++static struct hlist_head *au_name_hash(struct au_nhash *nhash,
++ unsigned char *name,
++ unsigned int len)
++{
++ unsigned int v;
++ /* const unsigned int magic_bit = 12; */
+
-+static struct attribute_group au_attr_group_body = {
-+ .attrs = au_attr
-+};
++ AuDebugOn(!nhash->nh_num || !nhash->nh_head);
+
-+struct attribute_group *au_attr_group = &au_attr_group_body;
++ v = 0;
++ if (len > 8)
++ len = 8;
++ while (len--)
++ v += *name++;
++ /* v = hash_long(v, magic_bit); */
++ v %= nhash->nh_num;
++ return nhash->nh_head + v;
++}
+
-+/* ---------------------------------------------------------------------- */
++static int au_nhash_test_name(struct au_vdir_destr *str, const char *name,
++ int nlen)
++{
++ return str->len == nlen && !memcmp(str->name, name, nlen);
++}
+
-+/*
-+ * they are copied from linux/lib/kobject.c,
-+ * and will be exported in the future.
-+ */
-+static ssize_t au_attr_show(struct kobject *kobj, struct attribute *attr,
-+ char *buf)
++/* returns found or not */
++int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int nlen)
+{
-+ struct kobj_attribute *kattr;
-+ ssize_t ret = -EIO;
++ struct hlist_head *head;
++ struct au_vdir_wh *pos;
++ struct au_vdir_destr *str;
+
-+ kattr = container_of(attr, struct kobj_attribute, attr);
-+ if (kattr->show)
-+ ret = kattr->show(kobj, kattr, buf);
-+ return ret;
++ head = au_name_hash(whlist, name, nlen);
++ hlist_for_each_entry(pos, head, wh_hash) {
++ str = &pos->wh_str;
++ AuDbg("%.*s\n", str->len, str->name);
++ if (au_nhash_test_name(str, name, nlen))
++ return 1;
++ }
++ return 0;
+}
+
-+#ifdef CONFIG_AUFS_DEBUG
-+static ssize_t au_attr_store(struct kobject *kobj, struct attribute *attr,
-+ const char *buf, size_t count)
++/* returns found(true) or not */
++static int test_known(struct au_nhash *delist, char *name, int nlen)
+{
-+ struct kobj_attribute *kattr;
-+ ssize_t ret = -EIO;
++ struct hlist_head *head;
++ struct au_vdir_dehstr *pos;
++ struct au_vdir_destr *str;
+
-+ kattr = container_of(attr, struct kobj_attribute, attr);
-+ if (kattr->store)
-+ ret = kattr->store(kobj, kattr, buf, count);
-+ return ret;
++ head = au_name_hash(delist, name, nlen);
++ hlist_for_each_entry(pos, head, hash) {
++ str = pos->str;
++ AuDbg("%.*s\n", str->len, str->name);
++ if (au_nhash_test_name(str, name, nlen))
++ return 1;
++ }
++ return 0;
+}
-+#endif
+
-+static struct sysfs_ops sysaufs_ops = {
-+ .show = au_attr_show,
-+#ifdef CONFIG_AUFS_DEBUG
-+ .store = au_attr_store
++static void au_shwh_init_wh(struct au_vdir_wh *wh, ino_t ino,
++ unsigned char d_type)
++{
++#ifdef CONFIG_AUFS_SHWH
++ wh->wh_ino = ino;
++ wh->wh_type = d_type;
+#endif
-+};
-+
-+static struct kobj_type au_ktype_body = {
-+ .sysfs_ops = &sysaufs_ops
-+};
-+struct kobj_type *au_ktype = &au_ktype_body;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+static noinline_for_stack
-+int sysaufs_sbi_xi(struct seq_file *seq, struct file *xf, int dlgt,
-+ int print_path)
++int au_nhash_append_wh(struct au_nhash *whlist, char *name, int nlen, ino_t ino,
++ unsigned int d_type, aufs_bindex_t bindex,
++ unsigned char shwh)
+{
+ int err;
-+ struct kstat st;
-+ struct path path;
++ struct au_vdir_destr *str;
++ struct au_vdir_wh *wh;
+
-+ err = vfsub_getattr(xf->f_vfsmnt, xf->f_dentry, &st, dlgt);
-+ if (!err) {
-+ seq_printf(seq, "%llux%lu %lld",
-+ st.blocks, st.blksize, (long long)st.size);
-+ if (print_path) {
-+ path.dentry = xf->f_dentry;
-+ path.mnt = xf->f_vfsmnt;
-+ seq_putc(seq, ' ');
-+ seq_path(seq, &path, au_esc_chars);
-+ }
-+ seq_putc(seq, '\n');
-+ } else
-+ seq_printf(seq, "err %d\n", err);
++ AuDbg("%.*s\n", nlen, name);
++ AuDebugOn(!whlist->nh_num || !whlist->nh_head);
+
-+ AuTraceErr(err);
++ err = -ENOMEM;
++ wh = kmalloc(sizeof(*wh) + nlen, GFP_NOFS);
++ if (unlikely(!wh))
++ goto out;
++
++ err = 0;
++ wh->wh_bindex = bindex;
++ if (shwh)
++ au_shwh_init_wh(wh, ino, d_type);
++ str = &wh->wh_str;
++ str->len = nlen;
++ memcpy(str->name, name, nlen);
++ hlist_add_head(&wh->wh_hash, au_name_hash(whlist, name, nlen));
++ /* smp_mb(); */
++
++out:
+ return err;
+}
+
-+int sysaufs_sbi_xino(struct seq_file *seq, struct super_block *sb)
++static int append_deblk(struct au_vdir *vdir)
+{
+ int err;
-+ unsigned int mnt_flags;
-+ aufs_bindex_t bend, bindex;
-+ unsigned char dlgt, xinodir;
-+ struct kstat st;
-+ struct path path;
-+ struct au_sbinfo *sbinfo;
-+ struct file *xf;
++ unsigned long ul;
++ const unsigned int deblk_sz = vdir->vd_deblk_sz;
++ union au_vdir_deblk_p p, deblk_end;
++ unsigned char **o;
+
-+ AuTraceEnter();
++ err = -ENOMEM;
++ o = au_krealloc(vdir->vd_deblk, sizeof(*o) * (vdir->vd_nblk + 1),
++ GFP_NOFS, /*may_shrink*/0);
++ if (unlikely(!o))
++ goto out;
+
-+ sbinfo = au_sbi(sb);
-+ mnt_flags = au_mntflags(sb);
-+ xinodir = !!au_opt_test(mnt_flags, XINODIR);
-+ if (!au_opt_test_xino(mnt_flags)) {
-+#ifdef CONFIG_AUFS_DEBUG
-+ AuDebugOn(sbinfo->si_xib);
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++)
-+ AuDebugOn(au_sbr(sb, bindex)->br_xino.xi_file);
-+#endif
-+ err = 0;
-+ goto out; /* success */
++ vdir->vd_deblk = o;
++ p.deblk = kmalloc(deblk_sz, GFP_NOFS);
++ if (p.deblk) {
++ ul = vdir->vd_nblk++;
++ vdir->vd_deblk[ul] = p.deblk;
++ vdir->vd_last.ul = ul;
++ vdir->vd_last.p.deblk = p.deblk;
++ deblk_end.deblk = p.deblk + deblk_sz;
++ err = set_deblk_end(&p, &deblk_end);
+ }
+
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ err = sysaufs_sbi_xi(seq, sbinfo->si_xib, dlgt, xinodir);
++out:
++ return err;
++}
+
-+ bend = au_sbend(sb);
-+ for (bindex = 0; !err && bindex <= bend; bindex++) {
-+ xf = au_sbr(sb, bindex)->br_xino.xi_file;
-+ if (!xf)
-+ continue;
-+ seq_printf(seq, "%d: ", bindex);
-+ err = vfsub_getattr(xf->f_vfsmnt, xf->f_dentry, &st, dlgt);
-+ if (!err) {
-+ seq_printf(seq, "%ld, %llux%lu %lld",
-+ (long)file_count(xf), st.blocks, st.blksize,
-+ (long long)st.size);
-+ if (xinodir) {
-+ path.dentry = xf->f_dentry;
-+ path.mnt = xf->f_vfsmnt;
-+ seq_putc(seq, ' ');
-+ seq_path(seq, &path, au_esc_chars);
-+ }
-+ seq_putc(seq, '\n');
-+ } else
-+ seq_printf(seq, "err %d\n", err);
++static int append_de(struct au_vdir *vdir, char *name, int nlen, ino_t ino,
++ unsigned int d_type, struct au_nhash *delist)
++{
++ int err;
++ unsigned int sz;
++ const unsigned int deblk_sz = vdir->vd_deblk_sz;
++ union au_vdir_deblk_p p, *room, deblk_end;
++ struct au_vdir_dehstr *dehstr;
++
++ p.deblk = last_deblk(vdir);
++ deblk_end.deblk = p.deblk + deblk_sz;
++ room = &vdir->vd_last.p;
++ AuDebugOn(room->deblk < p.deblk || deblk_end.deblk <= room->deblk
++ || !is_deblk_end(room, &deblk_end));
++
++ sz = calc_size(nlen);
++ if (unlikely(sz > deblk_end.deblk - room->deblk)) {
++ err = append_deblk(vdir);
++ if (unlikely(err))
++ goto out;
++
++ p.deblk = last_deblk(vdir);
++ deblk_end.deblk = p.deblk + deblk_sz;
++ /* smp_mb(); */
++ AuDebugOn(room->deblk != p.deblk);
+ }
+
-+ out:
-+ AuTraceErr(err);
++ err = -ENOMEM;
++ dehstr = au_cache_alloc_vdir_dehstr();
++ if (unlikely(!dehstr))
++ goto out;
++
++ dehstr->str = &room->de->de_str;
++ hlist_add_head(&dehstr->hash, au_name_hash(delist, name, nlen));
++ room->de->de_ino = ino;
++ room->de->de_type = d_type;
++ room->de->de_str.len = nlen;
++ memcpy(room->de->de_str.name, name, nlen);
++
++ err = 0;
++ room->deblk += sz;
++ if (unlikely(set_deblk_end(room, &deblk_end)))
++ err = append_deblk(vdir);
++ /* smp_mb(); */
++
++out:
+ return err;
+}
+
-+#ifdef CONFIG_AUFS_EXPORT
-+int sysaufs_sbi_xigen(struct seq_file *seq, struct super_block *sb)
++/* ---------------------------------------------------------------------- */
++
++void au_vdir_free(struct au_vdir *vdir)
++{
++ unsigned char **deblk;
++
++ deblk = vdir->vd_deblk;
++ while (vdir->vd_nblk--)
++ au_kfree_try_rcu(*deblk++);
++ au_kfree_try_rcu(vdir->vd_deblk);
++ au_cache_free_vdir(vdir);
++}
++
++static struct au_vdir *alloc_vdir(struct file *file)
+{
++ struct au_vdir *vdir;
++ struct super_block *sb;
+ int err;
-+ unsigned int mnt_flags;
-+ struct au_sbinfo *sbinfo;
+
-+ AuTraceEnter();
++ sb = file->f_path.dentry->d_sb;
++ SiMustAnyLock(sb);
++
++ err = -ENOMEM;
++ vdir = au_cache_alloc_vdir();
++ if (unlikely(!vdir))
++ goto out;
++
++ vdir->vd_deblk = kzalloc(sizeof(*vdir->vd_deblk), GFP_NOFS);
++ if (unlikely(!vdir->vd_deblk))
++ goto out_free;
++
++ vdir->vd_deblk_sz = au_sbi(sb)->si_rdblk;
++ if (!vdir->vd_deblk_sz) {
++ /* estimate the appropriate size for deblk */
++ vdir->vd_deblk_sz = au_dir_size(file, /*dentry*/NULL);
++ /* pr_info("vd_deblk_sz %u\n", vdir->vd_deblk_sz); */
++ }
++ vdir->vd_nblk = 0;
++ vdir->vd_version = 0;
++ vdir->vd_jiffy = 0;
++ err = append_deblk(vdir);
++ if (!err)
++ return vdir; /* success */
+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ mnt_flags = au_mntflags(sb);
-+ if (au_opt_test_xino(mnt_flags))
-+ err = sysaufs_sbi_xi(seq, sbinfo->si_xigen,
-+ !!au_opt_test(mnt_flags, DLGT),
-+ !!au_opt_test(mnt_flags, XINODIR));
++ au_kfree_try_rcu(vdir->vd_deblk);
+
-+ AuTraceErr(err);
-+ return err;
++out_free:
++ au_cache_free_vdir(vdir);
++out:
++ vdir = ERR_PTR(err);
++ return vdir;
+}
-+#endif
+
-+/*
-+ * the lifetime of branch is independent from the entry under sysfs.
-+ * sysfs handles the lifetime of the entry, and never call ->show() after it is
-+ * unlinked.
-+ */
-+#define SysaufsBr_PREFIX "br"
-+static int sysaufs_sbi_br(struct seq_file *seq, struct super_block *sb,
-+ aufs_bindex_t bindex)
++static int reinit_vdir(struct au_vdir *vdir)
+{
+ int err;
-+ struct dentry *root;
-+ struct au_branch *br;
-+ struct path path;
-+
-+ LKTRTrace("b%d\n", bindex);
-+
-+ err = -ENOENT;
-+ if (unlikely(au_sbend(sb) < bindex))
-+ goto out;
-+
-+ err = 0;
-+ root = sb->s_root;
-+ di_read_lock_parent(root, !AuLock_IR);
-+ br = au_sbr(sb, bindex);
-+ path.mnt = br->br_mnt;
-+ path.dentry = au_h_dptr(root, bindex);
-+ seq_path(seq, &path, au_esc_chars);
-+ di_read_unlock(root, !AuLock_IR);
-+ seq_printf(seq, "=%s\n", au_optstr_br_perm(br->br_perm));
++ union au_vdir_deblk_p p, deblk_end;
+
-+ out:
-+ AuTraceErr(err);
++ while (vdir->vd_nblk > 1) {
++ au_kfree_try_rcu(vdir->vd_deblk[vdir->vd_nblk - 1]);
++ /* vdir->vd_deblk[vdir->vd_nblk - 1] = NULL; */
++ vdir->vd_nblk--;
++ }
++ p.deblk = vdir->vd_deblk[0];
++ deblk_end.deblk = p.deblk + vdir->vd_deblk_sz;
++ err = set_deblk_end(&p, &deblk_end);
++ /* keep vd_dblk_sz */
++ vdir->vd_last.ul = 0;
++ vdir->vd_last.p.deblk = vdir->vd_deblk[0];
++ vdir->vd_version = 0;
++ vdir->vd_jiffy = 0;
++ /* smp_mb(); */
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+static struct seq_file *au_seq(char *p, ssize_t len)
++#define AuFillVdir_CALLED 1
++#define AuFillVdir_WHABLE (1 << 1)
++#define AuFillVdir_SHWH (1 << 2)
++#define au_ftest_fillvdir(flags, name) ((flags) & AuFillVdir_##name)
++#define au_fset_fillvdir(flags, name) \
++ do { (flags) |= AuFillVdir_##name; } while (0)
++#define au_fclr_fillvdir(flags, name) \
++ do { (flags) &= ~AuFillVdir_##name; } while (0)
++
++#ifndef CONFIG_AUFS_SHWH
++#undef AuFillVdir_SHWH
++#define AuFillVdir_SHWH 0
++#endif
++
++struct fillvdir_arg {
++ struct dir_context ctx;
++ struct file *file;
++ struct au_vdir *vdir;
++ struct au_nhash delist;
++ struct au_nhash whlist;
++ aufs_bindex_t bindex;
++ unsigned int flags;
++ int err;
++};
++
++static bool fillvdir(struct dir_context *ctx, const char *__name, int nlen,
++ loff_t offset __maybe_unused, u64 h_ino,
++ unsigned int d_type)
+{
-+ struct seq_file *seq;
++ struct fillvdir_arg *arg = container_of(ctx, struct fillvdir_arg, ctx);
++ char *name = (void *)__name;
++ struct super_block *sb;
++ ino_t ino;
++ const unsigned char shwh = !!au_ftest_fillvdir(arg->flags, SHWH);
+
-+ seq = kzalloc(sizeof(*seq), GFP_NOFS);
-+ if (seq) {
-+ /* todo: necessary? */
-+ /* mutex_init(&seq.lock); */
-+ seq->buf = p;
-+ seq->size = len;
-+ return seq; /* success */
++ arg->err = 0;
++ sb = arg->file->f_path.dentry->d_sb;
++ au_fset_fillvdir(arg->flags, CALLED);
++ /* smp_mb(); */
++ if (nlen <= AUFS_WH_PFX_LEN
++ || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
++ if (test_known(&arg->delist, name, nlen)
++ || au_nhash_test_known_wh(&arg->whlist, name, nlen))
++ goto out; /* already exists or whiteouted */
++
++ arg->err = au_ino(sb, arg->bindex, h_ino, d_type, &ino);
++ if (!arg->err) {
++ if (unlikely(nlen > AUFS_MAX_NAMELEN))
++ d_type = DT_UNKNOWN;
++ arg->err = append_de(arg->vdir, name, nlen, ino,
++ d_type, &arg->delist);
++ }
++ } else if (au_ftest_fillvdir(arg->flags, WHABLE)) {
++ name += AUFS_WH_PFX_LEN;
++ nlen -= AUFS_WH_PFX_LEN;
++ if (au_nhash_test_known_wh(&arg->whlist, name, nlen))
++ goto out; /* already whiteouted */
++
++ ino = 0; /* just to suppress a warning */
++ if (shwh)
++ arg->err = au_wh_ino(sb, arg->bindex, h_ino, d_type,
++ &ino);
++ if (!arg->err) {
++ if (nlen <= AUFS_MAX_NAMELEN + AUFS_WH_PFX_LEN)
++ d_type = DT_UNKNOWN;
++ arg->err = au_nhash_append_wh
++ (&arg->whlist, name, nlen, ino, d_type,
++ arg->bindex, shwh);
++ }
+ }
+
-+ seq = ERR_PTR(-ENOMEM);
-+ AuTraceErrPtr(seq);
-+ return seq;
++out:
++ if (!arg->err)
++ arg->vdir->vd_jiffy = jiffies;
++ /* smp_mb(); */
++ AuTraceErr(arg->err);
++ return !arg->err;
+}
+
-+/* todo: file size may exceed PAGE_SIZE */
-+ssize_t sysaufs_sbi_show(struct kobject *kobj, struct attribute *attr,
-+ char *buf)
++static int au_handle_shwh(struct super_block *sb, struct au_vdir *vdir,
++ struct au_nhash *whlist, struct au_nhash *delist)
+{
-+ ssize_t err;
-+ long l;
-+ struct au_sbinfo *sbinfo;
-+ struct super_block *sb;
-+ struct seq_file *seq;
-+ char *name;
-+ struct attribute **cattr;
-+
-+ LKTRTrace("%s/%s\n", kobject_name(kobj), attr->name);
++#ifdef CONFIG_AUFS_SHWH
++ int err;
++ unsigned int nh, u;
++ struct hlist_head *head;
++ struct au_vdir_wh *pos;
++ struct hlist_node *n;
++ char *p, *o;
++ struct au_vdir_destr *destr;
+
-+ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
-+ sb = sbinfo->si_sb;
-+ si_noflush_read_lock(sb);
++ AuDebugOn(!au_opt_test(au_mntflags(sb), SHWH));
+
-+ seq = au_seq(buf, PAGE_SIZE);
-+ err = PTR_ERR(seq);
-+ if (IS_ERR(seq))
++ err = -ENOMEM;
++ o = p = (void *)__get_free_page(GFP_NOFS);
++ if (unlikely(!p))
+ goto out;
+
-+ name = (void *)attr->name;
-+ cattr = au_sbi_attrs;
-+ while (*cattr) {
-+ if (!strcmp(name, (*cattr)->name)) {
-+ err = container_of(*cattr, struct au_sbi_attr, attr)
-+ ->show(seq, sb);
-+ goto out_seq;
++ err = 0;
++ nh = whlist->nh_num;
++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
++ p += AUFS_WH_PFX_LEN;
++ for (u = 0; u < nh; u++) {
++ head = whlist->nh_head + u;
++ hlist_for_each_entry_safe(pos, n, head, wh_hash) {
++ destr = &pos->wh_str;
++ memcpy(p, destr->name, destr->len);
++ err = append_de(vdir, o, destr->len + AUFS_WH_PFX_LEN,
++ pos->wh_ino, pos->wh_type, delist);
++ if (unlikely(err))
++ break;
+ }
-+ cattr++;
+ }
+
-+ if (!strncmp(name, SysaufsBr_PREFIX, sizeof(SysaufsBr_PREFIX) - 1)) {
-+ name += sizeof(SysaufsBr_PREFIX) - 1;
-+ err = strict_strtol(name, 10, &l);
-+ if (!err)
-+ err = sysaufs_sbi_br(seq, sb, (aufs_bindex_t)l);
-+ goto out_seq;
-+ }
-+ BUG();
++ free_page((unsigned long)o);
+
-+ out_seq:
-+ if (!err) {
-+ err = seq->count;
-+ /* sysfs limit */
-+ if (unlikely(err == PAGE_SIZE))
-+ err = -EFBIG;
-+ }
-+ kfree(seq);
-+ out:
-+ si_read_unlock(sb);
++out:
+ AuTraceErr(err);
+ return err;
++#else
++ return 0;
++#endif
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void sysaufs_br_init(struct au_branch *br)
++static int au_do_read_vdir(struct fillvdir_arg *arg)
+{
-+ br->br_attr.name = br->br_name;
-+ br->br_attr.mode = S_IRUGO;
-+ br->br_attr.owner = THIS_MODULE;
-+}
++ int err;
++ unsigned int rdhash;
++ loff_t offset;
++ aufs_bindex_t bbot, bindex, btop;
++ unsigned char shwh;
++ struct file *hf, *file;
++ struct super_block *sb;
+
-+void sysaufs_brs_del(struct super_block *sb, aufs_bindex_t bindex)
-+{
-+ struct au_sbinfo *sbinfo;
-+ aufs_bindex_t bend;
++ file = arg->file;
++ sb = file->f_path.dentry->d_sb;
++ SiMustAnyLock(sb);
+
-+ LKTRTrace("b%d\n", bindex);
++ rdhash = au_sbi(sb)->si_rdhash;
++ if (!rdhash)
++ rdhash = au_rdhash_est(au_dir_size(file, /*dentry*/NULL));
++ err = au_nhash_alloc(&arg->delist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ err = au_nhash_alloc(&arg->whlist, rdhash, GFP_NOFS);
++ if (unlikely(err))
++ goto out_delist;
+
-+ if (!sysaufs_brs)
-+ return;
++ err = 0;
++ arg->flags = 0;
++ shwh = 0;
++ if (au_opt_test(au_mntflags(sb), SHWH)) {
++ shwh = 1;
++ au_fset_fillvdir(arg->flags, SHWH);
++ }
++ btop = au_fbtop(file);
++ bbot = au_fbbot_dir(file);
++ for (bindex = btop; !err && bindex <= bbot; bindex++) {
++ hf = au_hf_dir(file, bindex);
++ if (!hf)
++ continue;
+
-+ sbinfo = au_sbi(sb);
-+ bend = au_sbend(sb);
-+ for (; bindex <= bend; bindex++)
-+ sysfs_remove_file(&sbinfo->si_kobj,
-+ &au_sbr(sb, bindex)->br_attr);
++ offset = vfsub_llseek(hf, 0, SEEK_SET);
++ err = offset;
++ if (unlikely(offset))
++ break;
++
++ arg->bindex = bindex;
++ au_fclr_fillvdir(arg->flags, WHABLE);
++ if (shwh
++ || (bindex != bbot
++ && au_br_whable(au_sbr_perm(sb, bindex))))
++ au_fset_fillvdir(arg->flags, WHABLE);
++ do {
++ arg->err = 0;
++ au_fclr_fillvdir(arg->flags, CALLED);
++ /* smp_mb(); */
++ err = vfsub_iterate_dir(hf, &arg->ctx);
++ if (err >= 0)
++ err = arg->err;
++ } while (!err && au_ftest_fillvdir(arg->flags, CALLED));
++
++ /*
++ * dir_relax() may be good for concurrency, but aufs should not
++ * use it since it will cause a lockdep problem.
++ */
++ }
++
++ if (!err && shwh)
++ err = au_handle_shwh(sb, arg->vdir, &arg->whlist, &arg->delist);
++
++ au_nhash_wh_free(&arg->whlist);
++
++out_delist:
++ au_nhash_de_free(&arg->delist);
++out:
++ return err;
+}
+
-+void sysaufs_brs_add(struct super_block *sb, aufs_bindex_t bindex)
++static int read_vdir(struct file *file, int may_read)
+{
+ int err;
-+ struct kobject *kobj;
-+ aufs_bindex_t bend;
-+ struct au_branch *br;
-+
-+ LKTRTrace("b%d\n", bindex);
++ unsigned long expire;
++ unsigned char do_read;
++ struct fillvdir_arg arg = {
++ .ctx = {
++ .actor = fillvdir
++ }
++ };
++ struct inode *inode;
++ struct au_vdir *vdir, *allocated;
+
-+ if (!sysaufs_brs)
-+ return;
++ err = 0;
++ inode = file_inode(file);
++ IMustLock(inode);
++ IiMustWriteLock(inode);
++ SiMustAnyLock(inode->i_sb);
+
-+ kobj = &au_sbi(sb)->si_kobj;
-+ bend = au_sbend(sb);
-+ for (; bindex <= bend; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ snprintf(br->br_name, sizeof(br->br_name),
-+ SysaufsBr_PREFIX "%d", bindex);
-+ err = sysfs_create_file(kobj, &br->br_attr);
++ allocated = NULL;
++ do_read = 0;
++ expire = au_sbi(inode->i_sb)->si_rdcache;
++ vdir = au_ivdir(inode);
++ if (!vdir) {
++ do_read = 1;
++ vdir = alloc_vdir(file);
++ err = PTR_ERR(vdir);
++ if (IS_ERR(vdir))
++ goto out;
++ err = 0;
++ allocated = vdir;
++ } else if (may_read
++ && (!inode_eq_iversion(inode, vdir->vd_version)
++ || time_after(jiffies, vdir->vd_jiffy + expire))) {
++ do_read = 1;
++ err = reinit_vdir(vdir);
+ if (unlikely(err))
-+ AuWarn("failed %s under sysfs(%d)\n", br->br_name, err);
++ goto out;
+ }
-+}
-diff -urN linux/fs/aufs/sysrq.c linux-aufs/fs/aufs/sysrq.c
---- linux/fs/aufs/sysrq.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/sysrq.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,280 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
+
-+/*
-+ * magic sysrq hanlder
-+ *
-+ * $Id$
-+ */
-+
-+#include <linux/fs.h>
-+#include <linux/module.h>
-+#include <linux/moduleparam.h>
-+/* #include <linux/sysrq.h> */
-+#include "aufs.h"
-+
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+struct au_dbg_lock {
-+ struct list_head list;
-+ union {
-+ struct super_block *sb;
-+ struct dentry *dentry;
-+ struct inode *inode;
-+ void *any;
-+ };
-+ int flags;
-+ unsigned int lsc;
-+ pid_t pid;
-+};
++ if (!do_read)
++ return 0; /* success */
+
-+static void au_dbg_reg(struct au_splhead *spl, void *any, int flags,
-+ unsigned int lsc)
-+{
-+ struct au_dbg_lock *p = kmalloc(sizeof(*p), GFP_NOFS);
++ arg.file = file;
++ arg.vdir = vdir;
++ err = au_do_read_vdir(&arg);
++ if (!err) {
++ /* file->f_pos = 0; */ /* todo: ctx->pos? */
++ vdir->vd_version = inode_query_iversion(inode);
++ vdir->vd_last.ul = 0;
++ vdir->vd_last.p.deblk = vdir->vd_deblk[0];
++ if (allocated)
++ au_set_ivdir(inode, allocated);
++ } else if (allocated)
++ au_vdir_free(allocated);
+
-+ if (p) {
-+ p->any = any;
-+ p->flags = flags;
-+ p->lsc = lsc;
-+ p->pid = current->pid;
-+ spin_lock(&spl->spin);
-+ list_add(&p->list, &spl->head);
-+ spin_unlock(&spl->spin);
-+ }
-+ WARN_ON(!p);
++out:
++ return err;
+}
+
-+static void au_dbg_unreg(struct au_splhead *spl, void *any, int flags)
++static int copy_vdir(struct au_vdir *tgt, struct au_vdir *src)
+{
-+ struct au_dbg_lock *p, *tmp, *found;
-+ struct list_head *head = &spl->head;
-+ const pid_t pid = current->pid;
++ int err, rerr;
++ unsigned long ul, n;
++ const unsigned int deblk_sz = src->vd_deblk_sz;
+
-+ found = NULL;
-+ spin_lock(&spl->spin);
-+ list_for_each_entry_safe(p, tmp, head, list)
-+ if (p->any == any && p->pid == pid) {
-+ list_del(&p->list);
-+ found = p;
-+ break;
-+ }
-+ spin_unlock(&spl->spin);
-+ kfree(found);
-+ WARN_ON(!found);
-+}
++ AuDebugOn(tgt->vd_nblk != 1);
+
-+/* ---------------------------------------------------------------------- */
++ err = -ENOMEM;
++ if (tgt->vd_nblk < src->vd_nblk) {
++ unsigned char **p;
+
-+void au_dbg_locking_si_reg(struct super_block *sb, int flags)
-+{
-+ au_dbg_reg(au_sbi(sb)->si_dbg_lock + AuDbgLock_SI_LOCKING, sb, flags,
-+ -1);
-+}
++ 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;
++ }
+
-+void au_dbg_locking_si_unreg(struct super_block *sb, int flags)
-+{
-+ au_dbg_unreg(au_sbi(sb)->si_dbg_lock + AuDbgLock_SI_LOCKING, sb, flags);
-+}
++ if (tgt->vd_deblk_sz != deblk_sz) {
++ unsigned char *p;
+
-+void au_dbg_locked_si_reg(struct super_block *sb, int flags)
-+{
-+ au_dbg_reg(au_sbi(sb)->si_dbg_lock + AuDbgLock_SI_LOCKED, sb, flags,
-+ -1);
-+}
++ tgt->vd_deblk_sz = deblk_sz;
++ p = au_krealloc(tgt->vd_deblk[0], deblk_sz, GFP_NOFS,
++ /*may_shrink*/1);
++ if (unlikely(!p))
++ goto out;
++ tgt->vd_deblk[0] = p;
++ }
++ memcpy(tgt->vd_deblk[0], src->vd_deblk[0], deblk_sz);
++ tgt->vd_version = src->vd_version;
++ tgt->vd_jiffy = src->vd_jiffy;
+
-+void au_dbg_locked_si_unreg(struct super_block *sb, int flags)
-+{
-+ au_dbg_unreg(au_sbi(sb)->si_dbg_lock + AuDbgLock_SI_LOCKED, sb, flags);
-+}
++ n = src->vd_nblk;
++ for (ul = 1; ul < n; ul++) {
++ tgt->vd_deblk[ul] = kmemdup(src->vd_deblk[ul], deblk_sz,
++ GFP_NOFS);
++ if (unlikely(!tgt->vd_deblk[ul]))
++ goto out;
++ tgt->vd_nblk++;
++ }
++ tgt->vd_nblk = n;
++ tgt->vd_last.ul = tgt->vd_last.ul;
++ tgt->vd_last.p.deblk = tgt->vd_deblk[tgt->vd_last.ul];
++ tgt->vd_last.p.deblk += src->vd_last.p.deblk
++ - src->vd_deblk[src->vd_last.ul];
++ /* smp_mb(); */
++ return 0; /* success */
+
-+void au_dbg_locking_di_reg(struct dentry *d, int flags, unsigned int lsc)
-+{
-+ au_dbg_reg(au_sbi(d->d_sb)->si_dbg_lock + AuDbgLock_DI_LOCKING, d,
-+ flags, lsc);
++out:
++ rerr = reinit_vdir(tgt);
++ BUG_ON(rerr);
++ return err;
+}
+
-+void au_dbg_locking_di_unreg(struct dentry *d, int flags)
++int au_vdir_init(struct file *file)
+{
-+ au_dbg_unreg(au_sbi(d->d_sb)->si_dbg_lock + AuDbgLock_DI_LOCKING, d,
-+ flags);
-+}
++ int err;
++ struct inode *inode;
++ struct au_vdir *vdir_cache, *allocated;
+
-+void au_dbg_locked_di_reg(struct dentry *d, int flags, unsigned int lsc)
-+{
-+ au_dbg_reg(au_sbi(d->d_sb)->si_dbg_lock + AuDbgLock_DI_LOCKED, d, flags,
-+ lsc);
-+}
++ /* test file->f_pos here instead of ctx->pos */
++ err = read_vdir(file, !file->f_pos);
++ if (unlikely(err))
++ goto out;
+
-+void au_dbg_locked_di_unreg(struct dentry *d, int flags)
-+{
-+ au_dbg_unreg(au_sbi(d->d_sb)->si_dbg_lock + AuDbgLock_DI_LOCKED, d,
-+ flags);
-+}
++ allocated = NULL;
++ vdir_cache = au_fvdir_cache(file);
++ if (!vdir_cache) {
++ vdir_cache = alloc_vdir(file);
++ err = PTR_ERR(vdir_cache);
++ if (IS_ERR(vdir_cache))
++ goto out;
++ allocated = vdir_cache;
++ } else if (!file->f_pos && vdir_cache->vd_version != file->f_version) {
++ /* test file->f_pos here instead of ctx->pos */
++ err = reinit_vdir(vdir_cache);
++ if (unlikely(err))
++ goto out;
++ } else
++ return 0; /* success */
+
-+void au_dbg_locking_ii_reg(struct inode *i, int flags, unsigned int lsc)
-+{
-+ au_dbg_reg(au_sbi(i->i_sb)->si_dbg_lock + AuDbgLock_II_LOCKING, i,
-+ flags, lsc);
-+}
++ inode = file_inode(file);
++ err = copy_vdir(vdir_cache, au_ivdir(inode));
++ if (!err) {
++ file->f_version = inode_query_iversion(inode);
++ if (allocated)
++ au_set_fvdir_cache(file, allocated);
++ } else if (allocated)
++ au_vdir_free(allocated);
+
-+void au_dbg_locking_ii_unreg(struct inode *i, int flags)
-+{
-+ au_dbg_unreg(au_sbi(i->i_sb)->si_dbg_lock + AuDbgLock_II_LOCKING, i,
-+ flags);
++out:
++ return err;
+}
+
-+void au_dbg_locked_ii_reg(struct inode *i, int flags, unsigned int lsc)
++static loff_t calc_offset(struct au_vdir *vdir)
+{
-+ au_dbg_reg(au_sbi(i->i_sb)->si_dbg_lock + AuDbgLock_II_LOCKED, i, flags,
-+ lsc);
++ loff_t offset;
++ union au_vdir_deblk_p p;
++
++ p.deblk = vdir->vd_deblk[vdir->vd_last.ul];
++ offset = vdir->vd_last.p.deblk - p.deblk;
++ offset += vdir->vd_deblk_sz * vdir->vd_last.ul;
++ return offset;
+}
+
-+void au_dbg_locked_ii_unreg(struct inode *i, int flags)
++/* returns true or false */
++static int seek_vdir(struct file *file, struct dir_context *ctx)
+{
-+ au_dbg_unreg(au_sbi(i->i_sb)->si_dbg_lock + AuDbgLock_II_LOCKED, i,
-+ flags);
-+}
-+#endif /* CONFIG_AUFS_DEBUG_LOCK */
++ int valid;
++ unsigned int deblk_sz;
++ unsigned long ul, n;
++ loff_t offset;
++ union au_vdir_deblk_p p, deblk_end;
++ struct au_vdir *vdir_cache;
+
-+/* ---------------------------------------------------------------------- */
++ valid = 1;
++ vdir_cache = au_fvdir_cache(file);
++ offset = calc_offset(vdir_cache);
++ AuDbg("offset %lld\n", offset);
++ if (ctx->pos == offset)
++ goto out;
+
-+static void sysrq_sb(struct super_block *sb)
-+{
-+ char *plevel;
-+ struct inode *i;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file;
++ vdir_cache->vd_last.ul = 0;
++ vdir_cache->vd_last.p.deblk = vdir_cache->vd_deblk[0];
++ if (!ctx->pos)
++ goto out;
+
-+ plevel = au_plevel;
-+ au_plevel = KERN_WARNING;
-+ au_debug_on();
++ valid = 0;
++ deblk_sz = vdir_cache->vd_deblk_sz;
++ ul = div64_u64(ctx->pos, deblk_sz);
++ AuDbg("ul %lu\n", ul);
++ if (ul >= vdir_cache->vd_nblk)
++ goto out;
+
-+ sbinfo = au_sbi(sb);
-+ pr_warning("si=%lx\n", au_si_mask ^ (unsigned long)sbinfo);
-+ pr_warning(AUFS_NAME ": superblock\n");
-+ au_dpri_sb(sb);
-+ pr_warning(AUFS_NAME ": root dentry\n");
-+ au_dpri_dentry(sb->s_root);
-+ pr_warning(AUFS_NAME ": root inode\n");
-+ au_dpri_inode(sb->s_root->d_inode);
-+ pr_warning(AUFS_NAME ": isolated inode\n");
-+ list_for_each_entry(i, &sb->s_inodes, i_sb_list)
-+ if (list_empty(&i->i_dentry))
-+ au_dpri_inode(i);
-+ pr_warning(AUFS_NAME ": files\n");
-+ list_for_each_entry(file, &sb->s_files, f_u.fu_list)
-+ if (au_test_aufs_file(file))
-+ au_dpri_file(file);
++ n = vdir_cache->vd_nblk;
++ for (; ul < n; ul++) {
++ p.deblk = vdir_cache->vd_deblk[ul];
++ deblk_end.deblk = p.deblk + deblk_sz;
++ offset = ul;
++ offset *= deblk_sz;
++ while (!is_deblk_end(&p, &deblk_end) && offset < ctx->pos) {
++ unsigned int l;
+
-+#ifdef CONFIG_AUFS_DEBUG_LOCK
-+ {
-+ struct au_dbg_lock *p;
-+ struct list_head *head;
-+
-+ pr_warning(AUFS_NAME ": locking si\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_SI_LOCKING].head;
-+ list_for_each_entry(p, head, list)
-+ pr_warning("pid: %d, 0x%x\n", p->pid, p->flags);
-+ pr_warning(AUFS_NAME ": locked si\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_SI_LOCKED].head;
-+ list_for_each_entry(p, head, list)
-+ pr_warning("pid: %d, 0x%x\n", p->pid, p->flags);
-+ pr_warning(AUFS_NAME ": locking di\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_DI_LOCKING].head;
-+ list_for_each_entry(p, head, list) {
-+ pr_warning("pid: %d, 0x%x, %d\n",
-+ p->pid, p->flags, p->lsc);
-+ au_dpri_dentry(p->dentry);
-+ }
-+ pr_warning(AUFS_NAME ": locked di\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_DI_LOCKED].head;
-+ list_for_each_entry(p, head, list) {
-+ pr_warning("pid: %d, 0x%x, %d\n",
-+ p->pid, p->flags, p->lsc);
-+ au_dpri_dentry(p->dentry);
-+ }
-+ pr_warning(AUFS_NAME ": locking ii\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_II_LOCKING].head;
-+ list_for_each_entry(p, head, list) {
-+ pr_warning("pid: %d, %d\n", p->pid, p->lsc);
-+ au_dpri_inode(p->inode);
-+ }
-+ pr_warning(AUFS_NAME ": locked ii\n");
-+ head = &sbinfo->si_dbg_lock[AuDbgLock_II_LOCKED].head;
-+ list_for_each_entry(p, head, list) {
-+ pr_warning("pid: %d, %d\n", p->pid, p->lsc);
-+ au_dpri_inode(p->inode);
++ l = calc_size(p.de->de_str.len);
++ offset += l;
++ p.deblk += l;
++ }
++ if (!is_deblk_end(&p, &deblk_end)) {
++ valid = 1;
++ vdir_cache->vd_last.ul = ul;
++ vdir_cache->vd_last.p = p;
++ break;
+ }
+ }
-+#endif
+
-+ au_plevel = plevel;
-+ au_debug_off();
++out:
++ /* smp_mb(); */
++ if (!valid)
++ AuDbg("valid %d\n", !valid);
++ return valid;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/* module parameter */
-+static char *aufs_sysrq_key = "a";
-+module_param_named(sysrq, aufs_sysrq_key, charp, S_IRUGO);
-+MODULE_PARM_DESC(sysrq, "MagicSysRq key for " AUFS_NAME);
-+
-+static void au_sysrq(int key, struct tty_struct *tty)
++int au_vdir_fill_de(struct file *file, struct dir_context *ctx)
+{
-+ struct kobject *kobj;
-+ struct au_sbinfo *sbinfo;
-+
-+ /* spin_lock(&au_kset->list_lock); */
-+ list_for_each_entry(kobj, &au_kset->list, entry) {
-+ sbinfo = container_of(kobj, struct au_sbinfo, si_kobj);
-+ sysrq_sb(sbinfo->si_sb);
-+ }
-+ /* spin_unlock(&au_kset->list_lock); */
-+}
-+
-+static struct sysrq_key_op au_sysrq_op = {
-+ .handler = au_sysrq,
-+ .help_msg = "Aufs",
-+ .action_msg = "Aufs",
-+ /* todo: test mask? */
-+ .enable_mask = SYSRQ_ENABLE_DUMP
-+};
++ unsigned int l, deblk_sz;
++ union au_vdir_deblk_p deblk_end;
++ struct au_vdir *vdir_cache;
++ struct au_vdir_de *de;
+
-+/* ---------------------------------------------------------------------- */
++ if (!seek_vdir(file, ctx))
++ return 0;
+
-+int __init au_sysrq_init(void)
-+{
-+ int err;
-+ char key;
++ vdir_cache = au_fvdir_cache(file);
++ deblk_sz = vdir_cache->vd_deblk_sz;
++ while (1) {
++ deblk_end.deblk = vdir_cache->vd_deblk[vdir_cache->vd_last.ul];
++ deblk_end.deblk += deblk_sz;
++ while (!is_deblk_end(&vdir_cache->vd_last.p, &deblk_end)) {
++ de = vdir_cache->vd_last.p.de;
++ AuDbg("%.*s, off%lld, i%lu, dt%d\n",
++ de->de_str.len, de->de_str.name, ctx->pos,
++ (unsigned long)de->de_ino, de->de_type);
++ if (unlikely(!dir_emit(ctx, de->de_str.name,
++ de->de_str.len, de->de_ino,
++ de->de_type))) {
++ /* todo: ignore the error caused by udba? */
++ /* return err; */
++ return 0;
++ }
+
-+ err = -1;
-+ key = *aufs_sysrq_key;
-+ if ('a' <= key && key <= 'z')
-+ err = register_sysrq_key(key, &au_sysrq_op);
-+ if (unlikely(err))
-+ AuErr("err %d, sysrq=%c\n", err, key);
-+ return err;
-+}
++ l = calc_size(de->de_str.len);
++ vdir_cache->vd_last.p.deblk += l;
++ ctx->pos += l;
++ }
++ if (vdir_cache->vd_last.ul < vdir_cache->vd_nblk - 1) {
++ vdir_cache->vd_last.ul++;
++ vdir_cache->vd_last.p.deblk
++ = vdir_cache->vd_deblk[vdir_cache->vd_last.ul];
++ ctx->pos = deblk_sz * vdir_cache->vd_last.ul;
++ continue;
++ }
++ break;
++ }
+
-+void au_sysrq_fin(void)
-+{
-+ int err;
-+ err = unregister_sysrq_key(*aufs_sysrq_key, &au_sysrq_op);
-+ if (unlikely(err))
-+ AuErr("err %d (ignored)\n", err);
++ /* smp_mb(); */
++ return 0;
+}
-diff -urN linux/fs/aufs/vdir.c linux-aufs/fs/aufs/vdir.c
---- linux/fs/aufs/vdir.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/vdir.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,941 @@
+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 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,918 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * virtual or vertical directory
-+ *
-+ * $Id$
++ * sub-routines for VFS
+ */
+
++#include <linux/mnt_namespace.h>
++#include <linux/nsproxy.h>
++#include <linux/security.h>
++#include <linux/splice.h>
+#include "aufs.h"
+
-+static int calc_size(int namelen)
++#ifdef CONFIG_AUFS_BR_FUSE
++int vfsub_test_mntns(struct vfsmount *mnt, struct super_block *h_sb)
+{
-+ int sz;
++ if (!au_test_fuse(h_sb) || !au_userns)
++ return 0;
+
-+ sz = sizeof(struct au_vdir_de) + namelen;
-+ if (sizeof(ino_t) == sizeof(long)) {
-+ const int mask = sizeof(ino_t) - 1;
-+ if (sz & mask) {
-+ sz += sizeof(ino_t);
-+ sz &= ~mask;
-+ }
-+ }
++ return is_current_mnt_ns(mnt) ? 0 : -EACCES;
++}
++#endif
+
-+ AuDebugOn(sz % sizeof(ino_t));
-+ return sz;
++int vfsub_sync_filesystem(struct super_block *h_sb)
++{
++ int err;
++
++ lockdep_off();
++ down_read(&h_sb->s_umount);
++ err = sync_filesystem(h_sb);
++ up_read(&h_sb->s_umount);
++ lockdep_on();
++
++ return err;
+}
+
-+static int set_deblk_end(union au_vdir_deblk_p *p,
-+ union au_vdir_deblk_p *deblk_end)
++/* ---------------------------------------------------------------------- */
++
++int vfsub_update_h_iattr(struct path *h_path, int *did)
+{
-+ if (calc_size(0) <= deblk_end->p - p->p) {
-+ p->de->de_str.len = 0;
-+ /* smp_mb(); */
-+ return 0;
-+ }
-+ return -1; /* error */
++ int err;
++ struct kstat st;
++ struct super_block *h_sb;
++
++ /*
++ * Always needs h_path->mnt for LSM or FUSE branch.
++ */
++ AuDebugOn(!h_path->mnt);
++
++ /* for remote fs, leave work for its getattr or d_revalidate */
++ /* for bad i_attr fs, handle them in aufs_getattr() */
++ /* still some fs may acquire i_mutex. we need to skip them */
++ err = 0;
++ if (!did)
++ did = &err;
++ h_sb = h_path->dentry->d_sb;
++ *did = (!au_test_fs_remote(h_sb) && au_test_fs_refresh_iattr(h_sb));
++ if (*did)
++ err = vfsub_getattr(h_path, &st);
++
++ return err;
+}
+
-+/* returns true or false */
-+static int is_deblk_end(union au_vdir_deblk_p *p,
-+ union au_vdir_deblk_p *deblk_end)
++/* ---------------------------------------------------------------------- */
++
++struct file *vfsub_dentry_open(struct path *path, int flags)
+{
-+ if (calc_size(0) <= deblk_end->p - p->p)
-+ return !p->de->de_str.len;
-+ return 1;
++ return dentry_open(path, flags /* | __FMODE_NONOTIFY */,
++ current_cred());
+}
+
-+static au_vdir_deblk_t *last_deblk(struct au_vdir *vdir)
++struct file *vfsub_filp_open(const char *path, int oflags, int mode)
+{
-+ return vdir->vd_deblk[vdir->vd_nblk - 1];
++ struct file *file;
++
++ lockdep_off();
++ file = filp_open(path,
++ oflags /* | __FMODE_NONOTIFY */,
++ mode);
++ lockdep_on();
++ if (IS_ERR(file))
++ goto out;
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
++
++out:
++ return file;
+}
+
-+void au_nhash_init(struct au_nhash *nhash)
-+{
-+ int i;
-+ for (i = 0; i < AuSize_NHASH; i++)
-+ INIT_HLIST_HEAD(nhash->heads + i);
-+}
++/*
++ * Ideally this function should call VFS:do_last() in order to keep all its
++ * checkings. But it is very hard for aufs to regenerate several VFS internal
++ * structure such as nameidata. This is a second (or third) best approach.
++ * cf. linux/fs/namei.c:do_last(), lookup_open() and atomic_open().
++ */
++int vfsub_atomic_open(struct inode *dir, struct dentry *dentry,
++ struct vfsub_aopen_args *args)
++{
++ int err;
++ struct au_branch *br = args->br;
++ struct file *file = args->file;
++ /* copied from linux/fs/namei.c:atomic_open() */
++ struct dentry *const DENTRY_NOT_SET = (void *)-1UL;
++
++ IMustLock(dir);
++ AuDebugOn(!dir->i_op->atomic_open);
++
++ err = au_br_test_oflag(args->open_flag, br);
++ if (unlikely(err))
++ goto out;
+
-+struct au_nhash *au_nhash_new(gfp_t gfp)
-+{
-+ struct au_nhash *nhash;
++ au_lcnt_inc(&br->br_nfiles);
++ file->f_path.dentry = DENTRY_NOT_SET;
++ file->f_path.mnt = au_br_mnt(br);
++ AuDbg("%ps\n", dir->i_op->atomic_open);
++ err = dir->i_op->atomic_open(dir, dentry, file, args->open_flag,
++ args->create_mode);
++ if (unlikely(err < 0)) {
++ au_lcnt_dec(&br->br_nfiles);
++ goto out;
++ }
++
++ /* temporary workaround for nfsv4 branch */
++ if (au_test_nfs(dir->i_sb))
++ nfs_mark_for_revalidate(dir);
+
-+ nhash = kmalloc(sizeof(*nhash), gfp);
-+ if (nhash) {
-+ au_nhash_init(nhash);
-+ return nhash;
++ if (file->f_mode & FMODE_CREATED)
++ fsnotify_create(dir, dentry);
++ if (!(file->f_mode & FMODE_OPENED)) {
++ au_lcnt_dec(&br->br_nfiles);
++ goto out;
+ }
-+ return ERR_PTR(-ENOMEM);
-+}
+
-+void au_nhash_del(struct au_nhash *nhash)
-+{
-+ au_nhash_fin(nhash);
-+ kfree(nhash);
++ /* todo: call VFS:may_open() here */
++ /* todo: ima_file_check() too? */
++ if (!err && (args->open_flag & __FMODE_EXEC))
++ err = deny_write_access(file);
++ if (!err)
++ fsnotify_open(file);
++ else
++ au_lcnt_dec(&br->br_nfiles);
++ /* note that the file is created and still opened */
++
++out:
++ return err;
+}
+
-+void au_nhash_move(struct au_nhash *dst, struct au_nhash *src)
++int vfsub_kern_path(const char *name, unsigned int flags, struct path *path)
+{
-+ int i;
-+
-+ AuTraceEnter();
++ int err;
+
-+ *dst = *src;
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ struct hlist_head *h;
-+ h = dst->heads + i;
-+ if (h->first)
-+ h->first->pprev = &h->first;
-+ INIT_HLIST_HEAD(src->heads + i);
-+ }
-+ /* smp_mb(); */
++ err = kern_path(name, flags, path);
++ if (!err && d_is_positive(path->dentry))
++ vfsub_update_h_iattr(path, /*did*/NULL); /*ignore*/
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+void au_nhash_fin(struct au_nhash *whlist)
++struct dentry *vfsub_lookup_one_len_unlocked(const char *name,
++ struct path *ppath, int len)
+{
-+ int i;
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos, *n;
-+
-+ AuTraceEnter();
++ struct path path;
+
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ head = whlist->heads + i;
-+ hlist_for_each_entry_safe(tpos, pos, n, head, wh_hash) {
-+ /* hlist_del(pos); */
-+ kfree(tpos);
-+ }
++ path.dentry = lookup_one_len_unlocked(name, ppath->dentry, len);
++ if (IS_ERR(path.dentry))
++ goto out;
++ if (d_is_positive(path.dentry)) {
++ path.mnt = ppath->mnt;
++ vfsub_update_h_iattr(&path, /*did*/NULL); /*ignore*/
+ }
++
++out:
++ AuTraceErrPtr(path.dentry);
++ return path.dentry;
+}
+
-+int au_nhash_test_longer_wh(struct au_nhash *whlist, aufs_bindex_t btgt,
-+ int limit)
++struct dentry *vfsub_lookup_one_len(const char *name, struct path *ppath,
++ int len)
+{
-+ int n, i;
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos;
++ struct path path;
+
-+ LKTRTrace("limit %d\n", limit);
++ /* VFS checks it too, but by WARN_ON_ONCE() */
++ IMustLock(d_inode(ppath->dentry));
+
-+ n = 0;
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ head = whlist->heads + i;
-+ hlist_for_each_entry(tpos, pos, head, wh_hash)
-+ if (tpos->wh_bindex == btgt && ++n > limit)
-+ return 1;
++ path.dentry = lookup_one_len(name, ppath->dentry, len);
++ if (IS_ERR(path.dentry))
++ goto out;
++ if (d_is_positive(path.dentry)) {
++ path.mnt = ppath->mnt;
++ vfsub_update_h_iattr(&path, /*did*/NULL); /*ignore*/
+ }
-+ return 0;
++
++out:
++ AuTraceErrPtr(path.dentry);
++ return path.dentry;
+}
+
-+static unsigned int au_name_hash(const unsigned char *name, unsigned int len)
++void vfsub_call_lkup_one(void *args)
+{
-+ return full_name_hash(name, len) % AuSize_NHASH;
++ struct vfsub_lkup_one_args *a = args;
++ *a->errp = vfsub_lkup_one(a->name, a->ppath);
+}
+
-+/* returns found(true) or not */
-+int au_nhash_test_known_wh(struct au_nhash *whlist, char *name, int namelen)
++/* ---------------------------------------------------------------------- */
++
++struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1,
++ struct dentry *d2, struct au_hinode *hdir2)
+{
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos;
-+ struct au_vdir_destr *str;
++ struct dentry *d;
+
-+ LKTRTrace("%.*s\n", namelen, name);
++ lockdep_off();
++ d = lock_rename(d1, d2);
++ lockdep_on();
++ au_hn_suspend(hdir1);
++ if (hdir1 != hdir2)
++ au_hn_suspend(hdir2);
+
-+ head = whlist->heads + au_name_hash(name, namelen);
-+ hlist_for_each_entry(tpos, pos, head, wh_hash) {
-+ str = &tpos->wh_str;
-+ LKTRTrace("%.*s\n", str->len, str->name);
-+ if (str->len == namelen && !memcmp(str->name, name, namelen))
-+ return 1;
-+ }
-+ return 0;
++ return d;
+}
+
-+int au_nhash_append_wh(struct au_nhash *whlist, char *name, int namelen,
-+ ino_t ino, unsigned int d_type, aufs_bindex_t bindex,
-+ unsigned char shwh)
++void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1,
++ struct dentry *d2, struct au_hinode *hdir2)
+{
-+ int err;
-+ struct au_vdir_destr *str;
-+ struct au_vdir_wh *wh;
-+
-+ LKTRTrace("%.*s\n", namelen, name);
-+
-+ err = -ENOMEM;
-+ wh = kmalloc(sizeof(*wh) + namelen, GFP_NOFS);
-+ if (unlikely(!wh))
-+ goto out;
-+ err = 0;
-+ wh->wh_bindex = bindex;
-+ if (shwh)
-+ au_shwh_init_wh(wh, ino, d_type);
-+ str = &wh->wh_str;
-+ str->len = namelen;
-+ memcpy(str->name, name, namelen);
-+ hlist_add_head(&wh->wh_hash,
-+ whlist->heads + au_name_hash(name, namelen));
-+ /* smp_mb(); */
-+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ au_hn_resume(hdir1);
++ if (hdir1 != hdir2)
++ au_hn_resume(hdir2);
++ lockdep_off();
++ unlock_rename(d1, d2);
++ lockdep_on();
+}
+
+/* ---------------------------------------------------------------------- */
+
-+void au_vdir_free(struct au_vdir *vdir)
++int vfsub_create(struct inode *dir, struct path *path, int mode, bool want_excl)
+{
-+ au_vdir_deblk_t **deblk;
++ int err;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
+
-+ AuTraceEnter();
++ IMustLock(dir);
+
-+ deblk = vdir->vd_deblk;
-+ while (vdir->vd_nblk--) {
-+ kfree(*deblk);
-+ deblk++;
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_mknod(path, d, mode, 0);
++ path->dentry = d;
++ if (unlikely(err))
++ goto out;
++ idmap = mnt_idmap(path->mnt);
++
++ lockdep_off();
++ err = vfs_create(idmap, dir, path->dentry, mode, want_excl);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = *path;
++ int did;
++
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = path->dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ }
++ /*ignore*/
+ }
-+ kfree(vdir->vd_deblk);
-+ au_cache_free_vdir(vdir);
++
++out:
++ return err;
+}
+
-+static int append_deblk(struct au_vdir *vdir)
++int vfsub_symlink(struct inode *dir, struct path *path, const char *symname)
+{
-+ int err, sz, i;
-+ au_vdir_deblk_t **o;
-+ union au_vdir_deblk_p p, deblk_end;
++ int err;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
+
-+ AuTraceEnter();
++ IMustLock(dir);
+
-+ err = -ENOMEM;
-+ sz = sizeof(*o) * vdir->vd_nblk;
-+ o = au_kzrealloc(vdir->vd_deblk, sz, sz + sizeof(*o), GFP_NOFS);
-+ if (unlikely(!o))
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_symlink(path, d, symname);
++ path->dentry = d;
++ if (unlikely(err))
+ goto out;
-+ vdir->vd_deblk = o;
-+ p.deblk = kmalloc(sizeof(*p.deblk), GFP_NOFS);
-+ if (p.deblk) {
-+ i = vdir->vd_nblk++;
-+ vdir->vd_deblk[i] = p.deblk;
-+ vdir->vd_last.i = i;
-+ vdir->vd_last.p.p = p.p;
-+ deblk_end.deblk = p.deblk + 1;
-+ err = set_deblk_end(&p, &deblk_end);
-+ AuDebugOn(err);
++ idmap = mnt_idmap(path->mnt);
++
++ lockdep_off();
++ err = vfs_symlink(idmap, dir, path->dentry, symname);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = *path;
++ int did;
++
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = path->dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ }
++ /*ignore*/
+ }
+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+static struct au_vdir *alloc_vdir(void)
++int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev)
+{
-+ struct au_vdir *vdir;
+ int err;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
+
-+ AuTraceEnter();
++ IMustLock(dir);
+
-+ err = -ENOMEM;
-+ vdir = au_cache_alloc_vdir();
-+ if (unlikely(!vdir))
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_mknod(path, d, mode, new_encode_dev(dev));
++ path->dentry = d;
++ if (unlikely(err))
+ goto out;
-+ vdir->vd_deblk = kzalloc(sizeof(*vdir->vd_deblk), GFP_NOFS);
-+ if (unlikely(!vdir->vd_deblk))
-+ goto out_free;
++ idmap = mnt_idmap(path->mnt);
+
-+ vdir->vd_nblk = 0;
-+ vdir->vd_version = 0;
-+ vdir->vd_jiffy = 0;
-+ err = append_deblk(vdir);
-+ if (!err)
-+ return vdir; /* success */
++ lockdep_off();
++ err = vfs_mknod(idmap, dir, path->dentry, mode, dev);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = *path;
++ int did;
+
-+ kfree(vdir->vd_deblk);
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = path->dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ }
++ /*ignore*/
++ }
+
-+ out_free:
-+ au_cache_free_vdir(vdir);
-+ out:
-+ vdir = ERR_PTR(err);
-+ AuTraceErrPtr(vdir);
-+ return vdir;
++out:
++ return err;
+}
+
-+static int reinit_vdir(struct au_vdir *vdir)
++static int au_test_nlink(struct inode *inode)
++{
++ const unsigned int link_max = UINT_MAX >> 1; /* rough margin */
++
++ if (!au_test_fs_no_limit_nlink(inode->i_sb)
++ || inode->i_nlink < link_max)
++ return 0;
++ return -EMLINK;
++}
++
++int vfsub_link(struct dentry *src_dentry, struct inode *dir, struct path *path,
++ struct inode **delegated_inode)
+{
+ int err;
-+ union au_vdir_deblk_p p, deblk_end;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
++
++ IMustLock(dir);
+
-+ AuTraceEnter();
++ err = au_test_nlink(d_inode(src_dentry));
++ if (unlikely(err))
++ return err;
+
-+ while (vdir->vd_nblk > 1) {
-+ kfree(vdir->vd_deblk[vdir->vd_nblk - 1]);
-+ vdir->vd_deblk[vdir->vd_nblk - 1] = NULL;
-+ vdir->vd_nblk--;
++ /* we don't call may_linkat() */
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_link(src_dentry, path, d);
++ path->dentry = d;
++ if (unlikely(err))
++ goto out;
++ idmap = mnt_idmap(path->mnt);
++
++ lockdep_off();
++ err = vfs_link(src_dentry, idmap, dir, path->dentry, delegated_inode);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = *path;
++ int did;
++
++ /* fuse has different memory inode for the same inumber */
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = path->dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ tmp.dentry = src_dentry;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ }
++ /*ignore*/
+ }
-+ p.deblk = vdir->vd_deblk[0];
-+ deblk_end.deblk = p.deblk + 1;
-+ err = set_deblk_end(&p, &deblk_end);
-+ AuDebugOn(err);
-+ vdir->vd_version = 0;
-+ vdir->vd_jiffy = 0;
-+ vdir->vd_last.i = 0;
-+ vdir->vd_last.p.deblk = vdir->vd_deblk[0];
-+ /* smp_mb(); */
++
++out:
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static void free_dehlist(struct au_nhash *dehlist)
++int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
++ struct inode *dir, struct path *path,
++ struct inode **delegated_inode, unsigned int flags)
+{
-+ int i;
-+ struct hlist_head *head;
-+ struct au_vdir_dehstr *tpos;
-+ struct hlist_node *pos, *n;
++ int err;
++ struct renamedata rd;
++ struct path tmp = {
++ .mnt = path->mnt
++ };
++ struct dentry *d;
++
++ IMustLock(dir);
++ IMustLock(src_dir);
+
-+ AuTraceEnter();
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ tmp.dentry = src_dentry->d_parent;
++ err = security_path_rename(&tmp, src_dentry, path, d, /*flags*/0);
++ path->dentry = d;
++ if (unlikely(err))
++ goto out;
+
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ head = dehlist->heads + i;
-+ hlist_for_each_entry_safe(tpos, pos, n, head, hash) {
-+ /* hlist_del(pos); */
-+ au_cache_free_dehstr(tpos);
++ rd.old_mnt_idmap = mnt_idmap(path->mnt);
++ rd.old_dir = src_dir;
++ rd.old_dentry = src_dentry;
++ rd.new_mnt_idmap = rd.old_mnt_idmap;
++ rd.new_dir = dir;
++ rd.new_dentry = path->dentry;
++ rd.delegated_inode = delegated_inode;
++ rd.flags = flags;
++ lockdep_off();
++ err = vfs_rename(&rd);
++ lockdep_on();
++ if (!err) {
++ int did;
++
++ tmp.dentry = d->d_parent;
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = src_dentry;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ tmp.dentry = src_dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
+ }
++ /*ignore*/
+ }
++
++out:
++ return err;
+}
+
-+/* returns found(true) or not */
-+static int test_known(struct au_nhash *delist, char *name, int namelen)
++int vfsub_mkdir(struct inode *dir, struct path *path, int mode)
+{
-+ struct hlist_head *head;
-+ struct au_vdir_dehstr *tpos;
-+ struct hlist_node *pos;
-+ struct au_vdir_destr *str;
++ int err;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
+
-+ LKTRTrace("%.*s\n", namelen, name);
++ IMustLock(dir);
+
-+ head = delist->heads + au_name_hash(name, namelen);
-+ hlist_for_each_entry(tpos, pos, head, hash) {
-+ str = tpos->str;
-+ LKTRTrace("%.*s\n", str->len, str->name);
-+ if (str->len == namelen && !memcmp(str->name, name, namelen))
-+ return 1;
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_mkdir(path, d, mode);
++ path->dentry = d;
++ if (unlikely(err))
++ goto out;
++ idmap = mnt_idmap(path->mnt);
++
++ lockdep_off();
++ err = vfs_mkdir(idmap, dir, path->dentry, mode);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = *path;
++ int did;
++
++ vfsub_update_h_iattr(&tmp, &did);
++ if (did) {
++ tmp.dentry = path->dentry->d_parent;
++ vfsub_update_h_iattr(&tmp, /*did*/NULL);
++ }
++ /*ignore*/
+ }
-+ return 0;
+
++out:
++ return err;
+}
+
-+static int append_de(struct au_vdir *vdir, char *name, int namelen, ino_t ino,
-+ unsigned int d_type, struct au_nhash *delist)
++int vfsub_rmdir(struct inode *dir, struct path *path)
+{
-+ int err, sz;
-+ union au_vdir_deblk_p p, *room, deblk_end;
-+ struct au_vdir_dehstr *dehstr;
++ int err;
++ struct dentry *d;
++ struct mnt_idmap *idmap;
+
-+ LKTRTrace("%.*s %d, i%lu, dt%u\n",
-+ namelen, name, namelen, (unsigned long)ino, d_type);
++ IMustLock(dir);
+
-+ p.deblk = last_deblk(vdir);
-+ deblk_end.deblk = p.deblk + 1;
-+ room = &vdir->vd_last.p;
-+ AuDebugOn(room->p < p.p || deblk_end.p <= room->p
-+ || !is_deblk_end(room, &deblk_end));
++ d = path->dentry;
++ path->dentry = d->d_parent;
++ err = security_path_rmdir(path, d);
++ path->dentry = d;
++ if (unlikely(err))
++ goto out;
++ idmap = mnt_idmap(path->mnt);
+
-+ sz = calc_size(namelen);
-+ if (unlikely(sz > deblk_end.p - room->p)) {
-+ err = append_deblk(vdir);
-+ if (unlikely(err))
-+ goto out;
-+ p.deblk = last_deblk(vdir);
-+ deblk_end.deblk = p.deblk + 1;
-+ /* smp_mb(); */
-+ AuDebugOn(room->p != p.p);
++ lockdep_off();
++ err = vfs_rmdir(idmap, dir, path->dentry);
++ lockdep_on();
++ if (!err) {
++ struct path tmp = {
++ .dentry = path->dentry->d_parent,
++ .mnt = path->mnt
++ };
++
++ vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/
+ }
+
-+ err = -ENOMEM;
-+ dehstr = au_cache_alloc_dehstr();
-+ if (unlikely(!dehstr))
-+ goto out;
-+ dehstr->str = &room->de->de_str;
-+ hlist_add_head(&dehstr->hash,
-+ delist->heads + au_name_hash(name, namelen));
++out:
++ return err;
++}
+
-+ room->de->de_ino = ino;
-+ room->de->de_type = d_type;
-+ room->de->de_str.len = namelen;
-+ memcpy(room->de->de_str.name, name, namelen);
++/* ---------------------------------------------------------------------- */
++
++/* todo: support mmap_sem? */
++ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
++ loff_t *ppos)
++{
++ ssize_t err;
++
++ lockdep_off();
++ err = vfs_read(file, ubuf, count, ppos);
++ lockdep_on();
++ if (err >= 0)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
++ return err;
++}
+
-+ err = 0;
-+ room->p += sz;
-+ if (unlikely(set_deblk_end(room, &deblk_end)))
-+ err = append_deblk(vdir);
-+ /* smp_mb(); */
++ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count,
++ loff_t *ppos)
++{
++ ssize_t err;
+
-+ out:
++ lockdep_off();
++ err = kernel_read(file, kbuf, count, ppos);
++ lockdep_on();
+ AuTraceErr(err);
++ if (err >= 0)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count,
++ loff_t *ppos)
++{
++ ssize_t err;
++
++ lockdep_off();
++ err = vfs_write(file, ubuf, count, ppos);
++ lockdep_on();
++ if (err >= 0)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
++ return err;
++}
+
-+static int au_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ unsigned int d_type, ino_t *ino)
++ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos)
+{
-+ int err;
-+ struct au_xino_entry xinoe;
-+ struct mutex *mtx;
-+ const int isdir = (d_type == DT_DIR);
++ ssize_t err;
+
-+ /* prevent hardlinks from race condition */
-+ mtx = NULL;
-+ if (!isdir) {
-+ mtx = &au_sbr(sb, bindex)->br_xino.xi_nondir_mtx;
-+ mutex_lock(mtx);
-+ }
-+ err = au_xino_read(sb, bindex, h_ino, &xinoe);
-+ if (unlikely(err))
-+ goto out;
++ lockdep_off();
++ err = kernel_write(file, kbuf, count, ppos);
++ lockdep_on();
++ if (err >= 0)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
++ return err;
++}
+
-+ if (!xinoe.ino) {
-+ err = -EIO;
-+ xinoe.ino = au_xino_new_ino(sb);
-+ if (unlikely(!xinoe.ino))
-+ goto out;
++int vfsub_flush(struct file *file, fl_owner_t id)
++{
++ int err;
+
-+#if 0 /* reserved for future use */
-+ struct inode *h_inode;
-+ xinoe.h_gen = AuXino_INVALID_HGEN;
-+ h_inode = ilookup(au_sbr_sb(sb, bindex), h_ino);
-+ if (h_inode) {
-+ if (!is_bad_inode(h_inode)) {
-+ xinoe.h_gen = h_inode->i_generation;
-+ WARN_ON(xinoe.h_gen == AuXino_INVALID_HGEN);
-+ }
-+ iput(h_inode);
++ err = 0;
++ if (file->f_op->flush) {
++ if (!au_test_nfs(file->f_path.dentry->d_sb))
++ err = file->f_op->flush(file, id);
++ else {
++ lockdep_off();
++ err = file->f_op->flush(file, id);
++ lockdep_on();
+ }
-+#endif
-+ err = au_xino_write(sb, bindex, h_ino, &xinoe);
-+ if (unlikely(err))
-+ goto out;
++ if (!err)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL);
++ /*ignore*/
+ }
-+
-+ *ino = xinoe.ino;
-+
-+ out:
-+ if (!isdir)
-+ mutex_unlock(mtx);
-+ AuTraceErr(err);
+ return err;
+}
+
-+static int au_wh_ino(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ unsigned int d_type, ino_t *ino)
++int vfsub_iterate_dir(struct file *file, struct dir_context *ctx)
+{
-+#ifdef CONFIG_AUFS_SHWH
-+ return au_ino(sb, bindex, h_ino, d_type, ino);
-+#else
-+ return 0;
-+#endif
-+}
-+
-+#define AuFillVdir_CALLED 1
-+#define AuFillVdir_SHWH (1 << 1)
-+#define au_ftest_fillvdir(flags, name) ((flags) & AuFillVdir_##name)
-+#define au_fset_fillvdir(flags, name) { (flags) |= AuFillVdir_##name; }
-+#define au_fclr_fillvdir(flags, name) { (flags) &= ~AuFillVdir_##name; }
-+#ifndef CONFIG_AUFS_SHWH
-+#undef AuFillVdir_SHWH
-+#define AuFillVdir_SHWH 0
-+#endif
++ int err;
+
-+struct fillvdir_arg {
-+ struct file *file;
-+ struct au_vdir *vdir;
-+ struct au_nhash *delist;
-+ struct au_nhash *whlist;
-+ aufs_bindex_t bindex;
-+ unsigned int flags;
-+ int err;
-+};
++ AuDbg("%pD, ctx{%ps, %llu}\n", file, ctx->actor, ctx->pos);
+
-+static int fillvdir(void *__arg, const char *__name, int namelen, loff_t offset,
-+ u64 h_ino, unsigned int d_type)
-+{
-+ struct fillvdir_arg *arg = __arg;
-+ char *name = (void *)__name;
-+ aufs_bindex_t bindex, bend;
-+ struct super_block *sb;
-+ ino_t ino;
++ lockdep_off();
++ err = iterate_dir(file, ctx);
++ lockdep_on();
++ if (err >= 0)
++ vfsub_update_h_iattr(&file->f_path, /*did*/NULL); /*ignore*/
+
-+ LKTRTrace("%.*s, namelen %d, i%llu, dt%u\n",
-+ namelen, name, namelen, (unsigned long long)h_ino, d_type);
++ return err;
++}
+
-+ sb = arg->file->f_dentry->d_sb;
-+ bend = arg->bindex;
-+ arg->err = 0;
-+ au_fset_fillvdir(arg->flags, CALLED);
-+ /* smp_mb(); */
-+ if (namelen <= AUFS_WH_PFX_LEN
-+ || memcmp(name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)) {
-+ for (bindex = 0; bindex < bend; bindex++)
-+ if (test_known(arg->delist + bindex, name, namelen)
-+ || au_nhash_test_known_wh(arg->whlist + bindex,
-+ name, namelen))
-+ goto out; /* already exists or whiteouted */
-+
-+ ino = 1; /* why does gcc warns? */
-+ arg->err = au_ino(sb, bend, h_ino, d_type, &ino);
-+ if (!arg->err)
-+ arg->err = append_de(arg->vdir, name, namelen, ino,
-+ d_type, arg->delist + bend);
-+ } else {
-+ name += AUFS_WH_PFX_LEN;
-+ namelen -= AUFS_WH_PFX_LEN;
-+ for (bindex = 0; bindex < bend; bindex++)
-+ if (au_nhash_test_known_wh(arg->whlist + bend, name,
-+ namelen))
-+ goto out; /* already whiteouted */
-+
-+ ino = 1; /* dummy */
-+ if (au_ftest_fillvdir(arg->flags, SHWH))
-+ arg->err = au_wh_ino(sb, bend, h_ino, d_type, &ino);
-+ if (!arg->err)
-+ arg->err = au_nhash_append_wh
-+ (arg->whlist + bend, name, namelen, ino, d_type,
-+ bend, au_ftest_fillvdir(arg->flags, SHWH));
-+ }
++long vfsub_splice_to(struct file *in, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags)
++{
++ long err;
+
-+ out:
-+ if (!arg->err)
-+ arg->vdir->vd_jiffy = jiffies;
-+ /* smp_mb(); */
-+ AuTraceErr(arg->err);
-+ return arg->err;
++ lockdep_off();
++ err = vfs_splice_read(in, ppos, pipe, len, flags);
++ lockdep_on();
++ file_accessed(in);
++ if (err >= 0)
++ vfsub_update_h_iattr(&in->f_path, /*did*/NULL); /*ignore*/
++ return err;
+}
+
-+static int au_handle_shwh(struct super_block *sb, struct au_vdir *vdir,
-+ aufs_bindex_t bstart, aufs_bindex_t bend,
-+ struct au_nhash *_whlist, struct au_nhash *_delist)
++long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
++ loff_t *ppos, size_t len, unsigned int flags)
+{
-+#ifdef CONFIG_AUFS_SHWH
-+ int err, i;
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos, *n;
-+ char *p, *o;
-+ struct au_nhash *whlist, *delist;
-+ struct au_vdir_destr *destr;
-+ aufs_bindex_t bindex;
++ long err;
+
-+ AuTraceEnter();
-+ AuDebugOn(!au_opt_test(au_mntflags(sb), SHWH));
++ lockdep_off();
++ err = do_splice_from(pipe, out, ppos, len, flags);
++ lockdep_on();
++ if (err >= 0)
++ vfsub_update_h_iattr(&out->f_path, /*did*/NULL); /*ignore*/
++ return err;
++}
+
-+ err = -ENOMEM;
-+ o = p = __getname();
-+ if (unlikely(!p))
-+ goto out;
++int vfsub_fsync(struct file *file, struct path *path, int datasync)
++{
++ int err;
+
-+ err = 0;
-+ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
-+ p += AUFS_WH_PFX_LEN;
-+ for (bindex = bstart; !err && bindex <= bend; bindex++) {
-+ whlist = _whlist + bindex;
-+ delist = _delist + bindex;
-+
-+ for (i = 0; i < AuSize_NHASH; i++) {
-+ head = whlist->heads + i;
-+ hlist_for_each_entry_safe(tpos, pos, n, head, wh_hash) {
-+ destr = &tpos->wh_str;
-+ memcpy(p, destr->name, destr->len);
-+ err = append_de(vdir, o,
-+ destr->len + AUFS_WH_PFX_LEN,
-+ tpos->wh_ino, tpos->wh_type,
-+ delist);
-+ if (unlikely(err))
-+ break;
-+ }
++ /* file can be NULL */
++ lockdep_off();
++ err = vfs_fsync(file, datasync);
++ lockdep_on();
++ if (!err) {
++ if (!path) {
++ AuDebugOn(!file);
++ path = &file->f_path;
+ }
++ vfsub_update_h_iattr(path, /*did*/NULL); /*ignore*/
+ }
-+
-+ __putname(o);
-+
-+ out:
-+ AuTraceErr(err);
+ return err;
-+#else
-+ return 0;
-+#endif
+}
+
-+static int au_do_read_vdir(struct fillvdir_arg *arg)
++/* cf. open.c:do_sys_truncate() and do_sys_ftruncate() */
++int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr,
++ struct file *h_file)
+{
+ int err;
-+ unsigned int mnt_flags;
-+ loff_t offset;
-+ aufs_bindex_t bend, bindex, bstart;
-+ unsigned char dlgt, shwh;
-+ struct super_block *sb;
-+ struct file *hf;
-+
-+ AuTraceEnter();
++ struct inode *h_inode;
++ struct super_block *h_sb;
++ struct mnt_idmap *h_idmap;
+
-+ err = -ENOMEM;
-+ bend = au_fbend(arg->file);
-+ arg->delist = kmalloc(sizeof(*arg->delist) * (bend + 1), GFP_NOFS);
-+ if (unlikely(!arg->delist))
++ if (!h_file) {
++ err = vfsub_truncate(h_path, length);
+ goto out;
-+ arg->whlist = kmalloc(sizeof(*arg->whlist) * (bend + 1), GFP_NOFS);
-+ if (unlikely(!arg->whlist))
-+ goto out_delist;
-+ err = 0;
-+ for (bindex = 0; bindex <= bend; bindex++) {
-+ au_nhash_init(arg->delist + bindex);
-+ au_nhash_init(arg->whlist + bindex);
+ }
+
-+ sb = arg->file->f_dentry->d_sb;
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ arg->flags = 0;
-+ shwh = 0;
-+ if (au_opt_test(mnt_flags, SHWH)) {
-+ shwh = 1;
-+ au_fset_fillvdir(arg->flags, SHWH);
++ h_inode = d_inode(h_path->dentry);
++ h_sb = h_inode->i_sb;
++ lockdep_off();
++ sb_start_write(h_sb);
++ lockdep_on();
++ err = security_file_truncate(h_file);
++ if (!err) {
++ h_idmap = mnt_idmap(h_path->mnt);
++ lockdep_off();
++ err = do_truncate(h_idmap, h_path->dentry, length, attr,
++ h_file);
++ lockdep_on();
+ }
-+ bstart = au_fbstart(arg->file);
-+ for (bindex = bstart; !err && bindex <= bend; bindex++) {
-+ hf = au_h_fptr(arg->file, bindex);
-+ if (!hf)
-+ continue;
++ lockdep_off();
++ sb_end_write(h_sb);
++ lockdep_on();
+
-+ offset = vfsub_llseek(hf, 0, SEEK_SET);
-+ err = offset;
-+ if (unlikely(offset))
-+ break;
-+ arg->bindex = bindex;
-+ do {
-+ arg->err = 0;
-+ au_fclr_fillvdir(arg->flags, CALLED);
-+ /* smp_mb(); */
-+ err = vfsub_readdir(hf, fillvdir, arg, dlgt);
-+ if (err >= 0)
-+ err = arg->err;
-+ } while (!err && au_ftest_fillvdir(arg->flags, CALLED));
-+ }
++out:
++ return err;
++}
+
-+ if (!err && shwh)
-+ err = au_handle_shwh(sb, arg->vdir, bstart, bend, arg->whlist,
-+ arg->delist);
++/* ---------------------------------------------------------------------- */
+
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ free_dehlist(arg->delist + bindex);
-+ au_nhash_fin(arg->whlist + bindex);
-+ }
-+ kfree(arg->whlist);
++struct au_vfsub_mkdir_args {
++ int *errp;
++ struct inode *dir;
++ struct path *path;
++ int mode;
++};
+
-+ out_delist:
-+ kfree(arg->delist);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++static void au_call_vfsub_mkdir(void *args)
++{
++ struct au_vfsub_mkdir_args *a = args;
++ *a->errp = vfsub_mkdir(a->dir, a->path, a->mode);
+}
+
-+static int read_vdir(struct file *file, int may_read)
++int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode)
+{
-+ int err;
-+ unsigned long expire;
-+ struct fillvdir_arg arg;
-+ unsigned char do_read;
-+ struct dentry *dentry;
-+ struct inode *inode;
-+ struct au_vdir *vdir, *allocated;
-+ struct super_block *sb;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, may %d\n", AuDLNPair(dentry), may_read);
-+ FiMustWriteLock(file);
-+ inode = dentry->d_inode;
-+ IMustLock(inode);
-+ IiMustWriteLock(inode);
-+ AuDebugOn(!S_ISDIR(inode->i_mode));
++ int err, do_sio, wkq_err;
++ struct mnt_idmap *idmap;
+
-+ err = 0;
-+ allocated = NULL;
-+ do_read = 0;
-+ sb = inode->i_sb;
-+ expire = au_sbi(sb)->si_rdcache;
-+ vdir = au_ivdir(inode);
-+ if (!vdir) {
-+ AuDebugOn(au_fvdir_cache(file));
-+ do_read = 1;
-+ vdir = alloc_vdir();
-+ err = PTR_ERR(vdir);
-+ if (IS_ERR(vdir))
-+ goto out;
-+ err = 0;
-+ allocated = vdir;
-+ } else if (may_read
-+ && (inode->i_version != vdir->vd_version
-+ || time_after(jiffies, vdir->vd_jiffy + expire))) {
-+ LKTRTrace("iver %llu, vdver %lu, exp %lu\n",
-+ (unsigned long long)inode->i_version,
-+ vdir->vd_version, vdir->vd_jiffy + expire);
-+ do_read = 1;
-+ err = reinit_vdir(vdir);
-+ if (unlikely(err))
-+ goto out;
++ idmap = mnt_idmap(path->mnt);
++ do_sio = au_test_h_perm_sio(idmap, dir, MAY_EXEC | MAY_WRITE);
++ if (!do_sio) {
++ lockdep_off();
++ err = vfsub_mkdir(dir, path, mode);
++ lockdep_on();
++ } else {
++ struct au_vfsub_mkdir_args args = {
++ .errp = &err,
++ .dir = dir,
++ .path = path,
++ .mode = mode
++ };
++ wkq_err = au_wkq_wait(au_call_vfsub_mkdir, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
+ }
+
-+ if (!do_read)
-+ return 0; /* success */
++ return err;
++}
+
-+ arg.file = file;
-+ arg.vdir = vdir;
-+ err = au_do_read_vdir(&arg);
-+ if (!err) {
-+ /* todo: necessary? */
-+ /* file->f_pos = 0; */
-+ vdir->vd_version = inode->i_version;
-+ vdir->vd_last.i = 0;
-+ vdir->vd_last.p.deblk = vdir->vd_deblk[0];
-+ if (allocated)
-+ au_set_ivdir(inode, allocated);
-+ } else if (allocated)
-+ au_vdir_free(allocated);
++struct au_vfsub_rmdir_args {
++ int *errp;
++ struct inode *dir;
++ struct path *path;
++};
++
++static void au_call_vfsub_rmdir(void *args)
++{
++ struct au_vfsub_rmdir_args *a = args;
++ *a->errp = vfsub_rmdir(a->dir, a->path);
++}
++
++int vfsub_sio_rmdir(struct inode *dir, struct path *path)
++{
++ int err, do_sio, wkq_err;
++ struct mnt_idmap *idmap;
++
++ idmap = mnt_idmap(path->mnt);
++ do_sio = au_test_h_perm_sio(idmap, dir, MAY_EXEC | MAY_WRITE);
++ if (!do_sio) {
++ lockdep_off();
++ err = vfsub_rmdir(dir, path);
++ lockdep_on();
++ } else {
++ struct au_vfsub_rmdir_args args = {
++ .errp = &err,
++ .dir = dir,
++ .path = path
++ };
++ wkq_err = au_wkq_wait(au_call_vfsub_rmdir, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
++ }
+
-+ out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+static int copy_vdir(struct au_vdir *tgt, struct au_vdir *src)
-+{
-+ int err, i, rerr, n;
++/* ---------------------------------------------------------------------- */
+
-+ AuTraceEnter();
-+ AuDebugOn(tgt->vd_nblk != 1);
++struct notify_change_args {
++ int *errp;
++ struct path *path;
++ struct iattr *ia;
++ struct inode **delegated_inode;
++};
+
-+ err = -ENOMEM;
-+ if (tgt->vd_nblk < src->vd_nblk) {
-+ au_vdir_deblk_t **p;
-+ p = au_kzrealloc(tgt->vd_deblk, sizeof(*p) * tgt->vd_nblk,
-+ sizeof(*p) * src->vd_nblk, GFP_NOFS);
-+ if (unlikely(!p))
-+ goto out;
-+ tgt->vd_deblk = p;
-+ }
++static void call_notify_change(void *args)
++{
++ struct notify_change_args *a = args;
++ struct inode *h_inode;
++ struct mnt_idmap *idmap;
+
-+ tgt->vd_nblk = src->vd_nblk;
-+ n = src->vd_nblk;
-+ memcpy(tgt->vd_deblk[0], src->vd_deblk[0], AuSize_DEBLK);
-+ /* tgt->vd_last.i = 0; */
-+ /* tgt->vd_last.p.deblk = tgt->vd_deblk[0]; */
-+ tgt->vd_version = src->vd_version;
-+ tgt->vd_jiffy = src->vd_jiffy;
++ h_inode = d_inode(a->path->dentry);
++ IMustLock(h_inode);
+
-+ for (i = 1; i < n; i++) {
-+ tgt->vd_deblk[i] = kmalloc(AuSize_DEBLK, GFP_NOFS);
-+ if (tgt->vd_deblk[i])
-+ memcpy(tgt->vd_deblk[i], src->vd_deblk[i],
-+ AuSize_DEBLK);
-+ else
-+ goto out;
++ *a->errp = -EPERM;
++ if (!IS_IMMUTABLE(h_inode) && !IS_APPEND(h_inode)) {
++ idmap = mnt_idmap(a->path->mnt);
++ lockdep_off();
++ *a->errp = notify_change(idmap, a->path->dentry, a->ia,
++ a->delegated_inode);
++ lockdep_on();
++ if (!*a->errp)
++ vfsub_update_h_iattr(a->path, /*did*/NULL); /*ignore*/
+ }
-+ /* smp_mb(); */
-+ return 0; /* success */
-+
-+ out:
-+ rerr = reinit_vdir(tgt);
-+ BUG_ON(rerr);
-+ AuTraceErr(err);
-+ return err;
++ AuTraceErr(*a->errp);
+}
+
-+int au_vdir_init(struct file *file)
++int vfsub_notify_change(struct path *path, struct iattr *ia,
++ struct inode **delegated_inode)
+{
+ int err;
-+ struct dentry *dentry;
-+ struct inode *inode;
-+ struct au_vdir *vdir_cache, *allocated;
++ struct notify_change_args args = {
++ .errp = &err,
++ .path = path,
++ .ia = ia,
++ .delegated_inode = delegated_inode
++ };
+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, pos %lld\n", AuDLNPair(dentry), file->f_pos);
-+ FiMustWriteLock(file);
-+ inode = dentry->d_inode;
-+ IiMustWriteLock(inode);
-+ AuDebugOn(!S_ISDIR(inode->i_mode));
++ call_notify_change(&args);
+
-+ err = read_vdir(file, !file->f_pos);
-+ if (unlikely(err))
-+ goto out;
++ return err;
++}
+
-+ allocated = NULL;
-+ vdir_cache = au_fvdir_cache(file);
-+ if (!vdir_cache) {
-+ vdir_cache = alloc_vdir();
-+ err = PTR_ERR(vdir_cache);
-+ if (IS_ERR(vdir_cache))
-+ goto out;
-+ allocated = vdir_cache;
-+ } else if (!file->f_pos && vdir_cache->vd_version != file->f_version) {
-+ err = reinit_vdir(vdir_cache);
-+ if (unlikely(err))
-+ goto out;
-+ } else
-+ return 0; /* success */
++int vfsub_sio_notify_change(struct path *path, struct iattr *ia,
++ struct inode **delegated_inode)
++{
++ int err, wkq_err;
++ struct notify_change_args args = {
++ .errp = &err,
++ .path = path,
++ .ia = ia,
++ .delegated_inode = delegated_inode
++ };
+
-+ err = copy_vdir(vdir_cache, au_ivdir(inode));
-+ if (!err) {
-+ file->f_version = inode->i_version;
-+ if (allocated)
-+ au_set_fvdir_cache(file, allocated);
-+ } else if (allocated)
-+ au_vdir_free(allocated);
++ wkq_err = au_wkq_wait(call_notify_change, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
+
-+ out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+static loff_t calc_offset(struct au_vdir *vdir)
-+{
-+ loff_t offset;
-+ union au_vdir_deblk_p p;
++/* ---------------------------------------------------------------------- */
+
-+ p.deblk = vdir->vd_deblk[vdir->vd_last.i];
-+ offset = vdir->vd_last.p.p - p.p;
-+ offset += sizeof(*p.deblk) * vdir->vd_last.i;
-+ return offset;
-+}
++struct unlink_args {
++ int *errp;
++ struct inode *dir;
++ struct path *path;
++ struct inode **delegated_inode;
++};
+
-+/* returns true or false */
-+static int seek_vdir(struct file *file)
++static void call_unlink(void *args)
+{
-+ int valid, i, n;
-+ struct dentry *dentry;
-+ struct au_vdir *vdir_cache;
-+ loff_t offset;
-+ union au_vdir_deblk_p p, deblk_end;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, pos %lld\n", AuDLNPair(dentry), file->f_pos);
-+ vdir_cache = au_fvdir_cache(file);
-+ AuDebugOn(!vdir_cache);
++ struct unlink_args *a = args;
++ struct dentry *d = a->path->dentry;
++ struct inode *h_inode;
++ struct mnt_idmap *idmap;
++ const int stop_sillyrename = (au_test_nfs(d->d_sb)
++ && au_dcount(d) == 1);
+
-+ valid = 1;
-+ offset = calc_offset(vdir_cache);
-+ LKTRTrace("offset %lld\n", offset);
-+ if (file->f_pos == offset)
-+ goto out;
++ IMustLock(a->dir);
+
-+ vdir_cache->vd_last.i = 0;
-+ vdir_cache->vd_last.p.deblk = vdir_cache->vd_deblk[0];
-+ if (!file->f_pos)
-+ goto out;
++ a->path->dentry = d->d_parent;
++ *a->errp = security_path_unlink(a->path, d);
++ a->path->dentry = d;
++ if (unlikely(*a->errp))
++ return;
+
-+ valid = 0;
-+ i = file->f_pos / AuSize_DEBLK;
-+ LKTRTrace("i %d\n", i);
-+ if (i >= vdir_cache->vd_nblk)
-+ goto out;
++ if (!stop_sillyrename)
++ dget(d);
++ h_inode = NULL;
++ if (d_is_positive(d)) {
++ h_inode = d_inode(d);
++ ihold(h_inode);
++ }
+
-+ n = vdir_cache->vd_nblk;
-+ for (; i < n; i++) {
-+ p.deblk = vdir_cache->vd_deblk[i];
-+ deblk_end.deblk = p.deblk + 1;
-+ offset = i;
-+ offset *= AuSize_DEBLK;
-+ while (!is_deblk_end(&p, &deblk_end) && offset < file->f_pos) {
-+ int l;
-+ l = calc_size(p.de->de_str.len);
-+ offset += l;
-+ p.p += l;
-+ }
-+ if (!is_deblk_end(&p, &deblk_end)) {
-+ valid = 1;
-+ vdir_cache->vd_last.i = i;
-+ vdir_cache->vd_last.p = p;
-+ break;
-+ }
++ idmap = mnt_idmap(a->path->mnt);
++ lockdep_off();
++ *a->errp = vfs_unlink(idmap, a->dir, d, a->delegated_inode);
++ lockdep_on();
++ if (!*a->errp) {
++ struct path tmp = {
++ .dentry = d->d_parent,
++ .mnt = a->path->mnt
++ };
++ vfsub_update_h_iattr(&tmp, /*did*/NULL); /*ignore*/
+ }
+
-+ out:
-+ /* smp_mb(); */
-+ AuTraceErr(!valid);
-+ return valid;
++ if (!stop_sillyrename)
++ dput(d);
++ if (h_inode)
++ iput(h_inode);
++
++ AuTraceErr(*a->errp);
+}
+
-+int au_vdir_fill_de(struct file *file, void *dirent, filldir_t filldir)
++/*
++ * @dir: must be locked.
++ * @dentry: target dentry.
++ */
++int vfsub_unlink(struct inode *dir, struct path *path,
++ struct inode **delegated_inode, int force)
+{
-+ int err, l;
-+ struct dentry *dentry;
-+ struct au_vdir *vdir_cache;
-+ struct au_vdir_de *de;
-+ union au_vdir_deblk_p deblk_end;
-+
-+ dentry = file->f_dentry;
-+ LKTRTrace("%.*s, pos %lld\n", AuDLNPair(dentry), file->f_pos);
-+ vdir_cache = au_fvdir_cache(file);
-+ AuDebugOn(!vdir_cache);
-+
-+ if (!seek_vdir(file))
-+ return 0;
++ int err;
++ struct unlink_args args = {
++ .errp = &err,
++ .dir = dir,
++ .path = path,
++ .delegated_inode = delegated_inode
++ };
+
-+ while (1) {
-+ deblk_end.deblk
-+ = vdir_cache->vd_deblk[vdir_cache->vd_last.i] + 1;
-+ while (!is_deblk_end(&vdir_cache->vd_last.p, &deblk_end)) {
-+ de = vdir_cache->vd_last.p.de;
-+ LKTRTrace("%.*s, off%lld, i%lu, dt%d\n",
-+ de->de_str.len, de->de_str.name,
-+ file->f_pos, (unsigned long)de->de_ino,
-+ de->de_type);
-+ err = filldir(dirent, de->de_str.name, de->de_str.len,
-+ file->f_pos, de->de_ino, de->de_type);
-+ if (unlikely(err)) {
-+ AuTraceErr(err);
-+ /* todo: ignore the error caused by udba? */
-+ /* return err; */
-+ return 0;
-+ }
++ if (!force)
++ call_unlink(&args);
++ else {
++ int wkq_err;
+
-+ l = calc_size(de->de_str.len);
-+ vdir_cache->vd_last.p.p += l;
-+ file->f_pos += l;
-+ }
-+ if (vdir_cache->vd_last.i < vdir_cache->vd_nblk - 1) {
-+ vdir_cache->vd_last.i++;
-+ vdir_cache->vd_last.p.deblk
-+ = vdir_cache->vd_deblk[vdir_cache->vd_last.i];
-+ file->f_pos = sizeof(*vdir_cache->vd_last.p.deblk)
-+ * vdir_cache->vd_last.i;
-+ continue;
-+ }
-+ break;
++ wkq_err = au_wkq_wait(call_unlink, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
+ }
+
-+ /* smp_mb(); */
-+ return 0;
++ return err;
+}
-diff -urN linux/fs/aufs/vfsub.c linux-aufs/fs/aufs/vfsub.c
---- linux/fs/aufs/vfsub.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/vfsub.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,686 @@
+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 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,390 @@
++/* SPDX-License-Identifier: GPL-2.0 */
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
+ * sub-routines for VFS
-+ *
-+ * $Id$
+ */
+
-+#include <linux/uaccess.h>
-+#include "aufs.h"
++#ifndef __AUFS_VFSUB_H__
++#define __AUFS_VFSUB_H__
++
++#ifdef __KERNEL__
++
++#include <linux/fs.h>
++#include <linux/mount.h>
++#include <linux/posix_acl.h>
++#include <linux/xattr.h>
++#include "debug.h"
++
++/* copied from linux/fs/internal.h */
++/* todo: BAD approach!! */
++extern void __mnt_drop_write(struct vfsmount *);
++extern struct file *alloc_empty_file(int, const struct cred *);
++
++/* ---------------------------------------------------------------------- */
++
++/* lock subclass for lower inode */
++/* default MAX_LOCKDEP_SUBCLASSES(8) is not enough */
++/* reduce? gave up. */
++enum {
++ AuLsc_I_Begin = I_MUTEX_PARENT2, /* 5 */
++ AuLsc_I_PARENT, /* lower inode, parent first */
++ AuLsc_I_PARENT2, /* copyup dirs */
++ AuLsc_I_PARENT3, /* copyup wh */
++ AuLsc_I_CHILD,
++ AuLsc_I_CHILD2,
++ AuLsc_I_End
++};
++
++/* to debug easier, do not make them inlined functions */
++#define MtxMustLock(mtx) AuDebugOn(!mutex_is_locked(mtx))
++#define IMustLock(i) AuDebugOn(!inode_is_locked(i))
+
+/* ---------------------------------------------------------------------- */
+
-+void vfsub_args_init(struct vfsub_args *vargs, struct au_hin_ignore *ign,
-+ int dlgt, int force_unlink)
++static inline void vfsub_drop_nlink(struct inode *inode)
++{
++ AuDebugOn(!inode->i_nlink);
++ drop_nlink(inode);
++}
++
++static inline void vfsub_dead_dir(struct inode *inode)
+{
-+ do_vfsub_args_reinit(vargs, ign);
-+ vargs->flags = 0;
-+ if (dlgt)
-+ vfsub_fset(vargs->flags, DLGT);
-+ if (force_unlink)
-+ vfsub_fset(vargs->flags, FORCE_UNLINK);
++ AuDebugOn(!S_ISDIR(inode->i_mode));
++ inode->i_flags |= S_DEAD;
++ clear_nlink(inode);
++}
++
++static inline int vfsub_native_ro(struct inode *inode)
++{
++ return sb_rdonly(inode->i_sb)
++ || IS_RDONLY(inode)
++ /* || IS_APPEND(inode) */
++ || IS_IMMUTABLE(inode);
+}
+
++#ifdef CONFIG_AUFS_BR_FUSE
++int vfsub_test_mntns(struct vfsmount *mnt, struct super_block *h_sb);
++#else
++AuStubInt0(vfsub_test_mntns, struct vfsmount *mnt, struct super_block *h_sb);
++#endif
++
++int vfsub_sync_filesystem(struct super_block *h_sb);
++
+/* ---------------------------------------------------------------------- */
+
-+struct file *vfsub_filp_open(const char *path, int oflags, int mode)
++int vfsub_update_h_iattr(struct path *h_path, int *did);
++struct file *vfsub_dentry_open(struct path *path, int flags);
++struct file *vfsub_filp_open(const char *path, int oflags, int mode);
++struct au_branch;
++struct vfsub_aopen_args {
++ struct file *file;
++ unsigned int open_flag;
++ umode_t create_mode;
++ struct au_branch *br;
++};
++int vfsub_atomic_open(struct inode *dir, struct dentry *dentry,
++ struct vfsub_aopen_args *args);
++int vfsub_kern_path(const char *name, unsigned int flags, struct path *path);
++
++struct dentry *vfsub_lookup_one_len_unlocked(const char *name,
++ struct path *ppath, int len);
++struct dentry *vfsub_lookup_one_len(const char *name, struct path *ppath,
++ int len);
++
++struct vfsub_lkup_one_args {
++ struct dentry **errp;
++ struct qstr *name;
++ struct path *ppath;
++};
++
++static inline struct dentry *vfsub_lkup_one(struct qstr *name,
++ struct path *ppath)
+{
-+ struct file *err;
++ return vfsub_lookup_one_len(name->name, ppath, name->len);
++}
++
++void vfsub_call_lkup_one(void *args);
++
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("%s\n", path);
++static inline int vfsub_mnt_want_write(struct vfsmount *mnt)
++{
++ int err;
+
+ lockdep_off();
-+ err = filp_open(path, oflags, mode);
++ err = mnt_want_write(mnt);
+ lockdep_on();
-+ if (!IS_ERR(err))
-+ au_update_fuse_h_inode(err->f_vfsmnt, err->f_dentry); /*ignore*/
+ return err;
+}
+
-+int vfsub_path_lookup(const char *name, unsigned int flags,
-+ struct nameidata *nd)
++static inline void vfsub_mnt_drop_write(struct vfsmount *mnt)
++{
++ lockdep_off();
++ mnt_drop_write(mnt);
++ lockdep_on();
++}
++
++#if 0 /* reserved */
++static inline void vfsub_mnt_drop_write_file(struct file *file)
++{
++ lockdep_off();
++ mnt_drop_write_file(file);
++ lockdep_on();
++}
++#endif
++
++/* ---------------------------------------------------------------------- */
++
++struct au_hinode;
++struct dentry *vfsub_lock_rename(struct dentry *d1, struct au_hinode *hdir1,
++ struct dentry *d2, struct au_hinode *hdir2);
++void vfsub_unlock_rename(struct dentry *d1, struct au_hinode *hdir1,
++ struct dentry *d2, struct au_hinode *hdir2);
++
++int vfsub_create(struct inode *dir, struct path *path, int mode,
++ bool want_excl);
++int vfsub_symlink(struct inode *dir, struct path *path,
++ const char *symname);
++int vfsub_mknod(struct inode *dir, struct path *path, int mode, dev_t dev);
++int vfsub_link(struct dentry *src_dentry, struct inode *dir,
++ 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, unsigned int flags);
++int vfsub_mkdir(struct inode *dir, struct path *path, int mode);
++int vfsub_rmdir(struct inode *dir, struct path *path);
++
++/* ---------------------------------------------------------------------- */
++
++ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
++ loff_t *ppos);
++ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count,
++ loff_t *ppos);
++ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count,
++ loff_t *ppos);
++ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count,
++ loff_t *ppos);
++int vfsub_flush(struct file *file, fl_owner_t id);
++int vfsub_iterate_dir(struct file *file, struct dir_context *ctx);
++
++static inline loff_t vfsub_f_size_read(struct file *file)
+{
-+ int err;
-+
-+ LKTRTrace("%s\n", name);
-+
-+ /* lockdep_off(); */
-+ err = path_lookup(name, flags, nd);
-+ /* lockdep_on(); */
-+ if (!err)
-+ au_update_fuse_h_inode(nd->path.mnt, nd->path.dentry);
-+ /*ignore*/
-+ return err;
++ return i_size_read(file_inode(file));
+}
+
-+struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent,
-+ int len)
++static inline unsigned int vfsub_file_flags(struct file *file)
+{
-+ struct dentry *d;
++ unsigned int flags;
+
-+ LKTRTrace("%.*s/%.*s\n", AuDLNPair(parent), len, name);
-+ IMustLock(parent->d_inode);
++ spin_lock(&file->f_lock);
++ flags = file->f_flags;
++ spin_unlock(&file->f_lock);
+
-+ d = lookup_one_len(name, parent, len);
-+ if (!IS_ERR(d))
-+ au_update_fuse_h_inode(NULL, d); /*ignore*/
-+ return d;
++ return flags;
+}
+
-+#ifdef CONFIG_AUFS_LHASH_PATCH
-+struct dentry *vfsub__lookup_hash(struct qstr *name, struct dentry *parent,
-+ struct nameidata *nd)
++static inline int vfsub_file_execed(struct file *file)
+{
-+ struct dentry *d;
-+
-+ LKTRTrace("%.*s/%.*s, nd %d\n",
-+ AuDLNPair(parent), AuLNPair(name), !!nd);
-+ if (nd)
-+ LKTRTrace("nd{0x%x}\n", nd->flags);
-+ IMustLock(parent->d_inode);
++ /* todo: direct access f_flags */
++ return !!(vfsub_file_flags(file) & __FMODE_EXEC);
++}
+
-+ d = __lookup_hash(name, parent, nd);
-+ if (!IS_ERR(d))
-+ au_update_fuse_h_inode(NULL, d); /*ignore*/
-+ return d;
++#if 0 /* reserved */
++static inline void vfsub_file_accessed(struct file *h_file)
++{
++ file_accessed(h_file);
++ vfsub_update_h_iattr(&h_file->f_path, /*did*/NULL); /*ignore*/
+}
+#endif
+
-+/* ---------------------------------------------------------------------- */
-+
-+int do_vfsub_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd)
++#if 0 /* reserved */
++static inline void vfsub_touch_atime(struct vfsmount *h_mnt,
++ struct dentry *h_dentry)
+{
-+ int err;
-+ struct vfsmount *mnt;
-+
-+ LKTRTrace("i%lu, %.*s, 0x%x\n", dir->i_ino, AuDLNPair(dentry), mode);
-+ IMustLock(dir);
-+
-+ err = vfs_create(dir, dentry, mode, nd);
-+ if (!err) {
-+ mnt = NULL;
-+ if (nd)
-+ mnt = nd->path.mnt;
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(mnt, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(mnt, dentry); /*ignore*/
-+ }
-+ return err;
++ struct path h_path = {
++ .dentry = h_dentry,
++ .mnt = h_mnt
++ };
++ touch_atime(&h_path);
++ vfsub_update_h_iattr(&h_path, /*did*/NULL); /*ignore*/
+}
-+
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 27)
-+#define VfsubSymlinkArgs dir, dentry, symname
-+#else
-+#define VfsubSymlinkArgs dir, dentry, symname, mode
+#endif
+
-+int do_vfsub_symlink(struct inode *dir, struct dentry *dentry,
-+ const char *symname, int mode)
++static inline int vfsub_update_time(struct inode *h_inode,
++ struct timespec64 *ts, int flags)
+{
-+ int err;
-+
-+ LKTRTrace("i%lu, %.*s, %s, 0x%x\n",
-+ dir->i_ino, AuDLNPair(dentry), symname, mode);
-+ IMustLock(dir);
-+
-+ err = vfs_symlink(VfsubSymlinkArgs);
-+ if (!err) {
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, dentry); /*ignore*/
-+ }
-+ return err;
++ return inode_update_time(h_inode, ts, flags);
++ /* no vfsub_update_h_iattr() since we don't have struct path */
+}
+
-+int do_vfsub_mknod(struct inode *dir, struct dentry *dentry, int mode,
-+ dev_t dev)
++#ifdef CONFIG_FS_POSIX_ACL
++static inline int vfsub_acl_chmod(struct mnt_idmap *h_idmap,
++ struct dentry *h_dentry, umode_t h_mode)
+{
+ int err;
+
-+ LKTRTrace("i%lu, %.*s, 0x%x\n", dir->i_ino, AuDLNPair(dentry), mode);
-+ IMustLock(dir);
-+
-+ err = vfs_mknod(dir, dentry, mode, dev);
-+ if (!err) {
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, dentry); /*ignore*/
-+ }
++ err = posix_acl_chmod(h_idmap, h_dentry, h_mode);
++ if (err == -EOPNOTSUPP)
++ err = 0;
+ return err;
+}
++#else
++AuStubInt0(vfsub_acl_chmod, struct mnt_idmap *h_idmap,
++ struct dentry *h_dentry, umode_t h_mode);
++#endif
+
-+int do_vfsub_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry)
-+{
-+ int err;
++long vfsub_splice_to(struct file *in, loff_t *ppos,
++ struct pipe_inode_info *pipe, size_t len,
++ unsigned int flags);
++long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
++ loff_t *ppos, size_t len, unsigned int flags);
+
-+ LKTRTrace("%.*s, i%lu, %.*s\n",
-+ AuDLNPair(src_dentry), dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
++static inline long vfsub_truncate(struct path *path, loff_t length)
++{
++ long err;
+
+ lockdep_off();
-+ err = vfs_link(src_dentry, dir, dentry);
++ err = vfs_truncate(path, length);
+ lockdep_on();
-+ if (!err) {
-+ LKTRTrace("src_i %p, dst_i %p\n",
-+ src_dentry->d_inode, dentry->d_inode);
-+ /* fuse has different memory inode for the same inumber */
-+ au_update_fuse_h_inode(NULL, src_dentry); /*ignore*/
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, dentry); /*ignore*/
-+ }
+ return err;
+}
+
-+int do_vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry)
-+{
-+ int err;
++int vfsub_trunc(struct path *h_path, loff_t length, unsigned int attr,
++ struct file *h_file);
++int vfsub_fsync(struct file *file, struct path *path, int datasync);
+
-+ LKTRTrace("i%lu, %.*s, i%lu, %.*s\n",
-+ src_dir->i_ino, AuDLNPair(src_dentry),
-+ dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
-+ IMustLock(src_dir);
-+ AuDebugOn(src_dir != dir && !vfsub_is_rename_mutex_locked(dir->i_sb));
++/*
++ * re-use branch fs's ioctl(FICLONE) while aufs itself doesn't support such
++ * ioctl.
++ */
++static inline loff_t vfsub_clone_file_range(struct file *src, struct file *dst,
++ loff_t len)
++{
++ loff_t err;
+
+ lockdep_off();
-+ err = vfs_rename(src_dir, src_dentry, dir, dentry);
++ err = vfs_clone_file_range(src, 0, dst, 0, len, /*remap_flags*/0);
+ lockdep_on();
-+ if (!err) {
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, src_dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, src_dentry); /*ignore*/
-+ }
-+ return err;
-+}
-+
-+int do_vfsub_mkdir(struct inode *dir, struct dentry *dentry, int mode)
-+{
-+ int err;
-+
-+ LKTRTrace("i%lu, %.*s, 0x%x\n", dir->i_ino, AuDLNPair(dentry), mode);
-+ IMustLock(dir);
+
-+ err = vfs_mkdir(dir, dentry, mode);
-+ if (!err) {
-+ /* dir inode is locked */
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ au_update_fuse_h_inode(NULL, dentry); /*ignore*/
-+ }
+ return err;
+}
+
-+int do_vfsub_rmdir(struct inode *dir, struct dentry *dentry)
++/* 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)
+{
-+ int err;
-+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
++ ssize_t ssz;
+
+ lockdep_off();
-+ err = vfs_rmdir(dir, dentry);
++ ssz = vfs_copy_file_range(src, src_pos, dst, dst_pos, len, flags);
+ lockdep_on();
-+ /* dir inode is locked */
-+ if (!err)
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
-+ return err;
++
++ return ssz;
+}
+
-+int do_vfsub_unlink(struct inode *dir, struct dentry *dentry)
-+{
-+ int err;
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
-+ IMustLock(dir);
++static inline loff_t vfsub_llseek(struct file *file, loff_t offset, int origin)
++{
++ loff_t err;
+
-+ /* vfs_unlink() locks inode */
+ lockdep_off();
-+ err = vfs_unlink(dir, dentry);
++ err = vfs_llseek(file, offset, origin);
+ lockdep_on();
-+ /* dir inode is locked */
-+ if (!err)
-+ au_update_fuse_h_inode(NULL, dentry->d_parent); /*ignore*/
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+ssize_t do_vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
-+ loff_t *ppos)
-+{
-+ ssize_t err;
-+
-+ LKTRTrace("%.*s, cnt %zu, pos %lld\n",
-+ AuDLNPair(file->f_dentry), count, *ppos);
-+
-+ /* todo: always off, regardless nfs branch? */
-+ au_br_nfs_lockdep_off(file->f_vfsmnt->mnt_sb);
-+ err = vfs_read(file, ubuf, count, ppos);
-+ au_br_nfs_lockdep_on(file->f_vfsmnt->mnt_sb);
-+ if (err >= 0)
-+ au_update_fuse_h_inode(file->f_vfsmnt, file->f_dentry);
-+ /*ignore*/
-+ return err;
-+}
++int vfsub_sio_mkdir(struct inode *dir, struct path *path, int mode);
++int vfsub_sio_rmdir(struct inode *dir, struct path *path);
++int vfsub_sio_notify_change(struct path *path, struct iattr *ia,
++ struct inode **delegated_inode);
++int vfsub_notify_change(struct path *path, struct iattr *ia,
++ struct inode **delegated_inode);
++int vfsub_unlink(struct inode *dir, struct path *path,
++ struct inode **delegated_inode, int force);
+
-+/* todo: kernel_read()? */
-+ssize_t do_vfsub_read_k(struct file *file, void *kbuf, size_t count,
-+ loff_t *ppos)
++static inline int vfsub_getattr(const struct path *path, struct kstat *st)
+{
-+ ssize_t err;
-+ mm_segment_t oldfs;
-+
-+ oldfs = get_fs();
-+ set_fs(KERNEL_DS);
-+ err = do_vfsub_read_u(file, (char __user *)kbuf, count, ppos);
-+ set_fs(oldfs);
-+ return err;
++ return vfs_getattr(path, st, STATX_BASIC_STATS, AT_STATX_SYNC_AS_STAT);
+}
+
-+ssize_t do_vfsub_write_u(struct file *file, const char __user *ubuf,
-+ size_t count, loff_t *ppos)
-+{
-+ ssize_t err;
++/* ---------------------------------------------------------------------- */
+
-+ LKTRTrace("%.*s, cnt %zu, pos %lld\n",
-+ AuDLNPair(file->f_dentry), count, *ppos);
++static inline int vfsub_setxattr(struct mnt_idmap *idmap,
++ struct dentry *dentry, const char *name,
++ const void *value, size_t size, int flags)
++{
++ int err;
+
+ lockdep_off();
-+ err = vfs_write(file, ubuf, count, ppos);
++ err = vfs_setxattr(idmap, dentry, name, value, size, flags);
+ lockdep_on();
-+ if (err >= 0)
-+ au_update_fuse_h_inode(file->f_vfsmnt, file->f_dentry);
-+ /*ignore*/
-+ return err;
-+}
+
-+ssize_t do_vfsub_write_k(struct file *file, void *kbuf, size_t count,
-+ loff_t *ppos)
-+{
-+ ssize_t err;
-+ mm_segment_t oldfs;
-+
-+ oldfs = get_fs();
-+ set_fs(KERNEL_DS);
-+ err = do_vfsub_write_u(file, (const char __user *)kbuf, count, ppos);
-+ set_fs(oldfs);
+ return err;
+}
+
-+int do_vfsub_readdir(struct file *file, filldir_t filldir, void *arg)
++static inline int vfsub_removexattr(struct mnt_idmap *idmap,
++ struct dentry *dentry, const char *name)
+{
+ int err;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(file->f_dentry));
-+
+ lockdep_off();
-+ err = vfs_readdir(file, filldir, arg);
++ err = vfs_removexattr(idmap, dentry, name);
+ lockdep_on();
-+ if (err >= 0)
-+ au_update_fuse_h_inode(file->f_vfsmnt, file->f_dentry);
-+ /*ignore*/
++
+ return err;
+}
+
-+#ifdef CONFIG_AUFS_SPLICE_PATCH
-+long do_vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags)
++#ifdef CONFIG_FS_POSIX_ACL
++static inline int vfsub_set_acl(struct mnt_idmap *idmap,
++ struct dentry *dentry, const char *name,
++ struct posix_acl *acl)
+{
-+ long err;
-+
-+ LKTRTrace("%.*s, pos %lld, len %zu, 0x%x\n",
-+ AuDLNPair(in->f_dentry), *ppos, len, flags);
++ int err;
+
+ lockdep_off();
-+ err = vfs_splice_to(in, ppos, pipe, len, flags);
++ err = vfs_set_acl(idmap, dentry, name, acl);
+ lockdep_on();
-+ if (err >= 0)
-+ au_update_fuse_h_inode(in->f_vfsmnt, in->f_dentry); /*ignore*/
++
+ return err;
+}
+
-+long do_vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags)
++static inline int vfsub_remove_acl(struct mnt_idmap *idmap,
++ struct dentry *dentry, const char *name)
+{
-+ long err;
-+
-+ LKTRTrace("%.*s, pos %lld, len %zu, 0x%x\n",
-+ AuDLNPair(out->f_dentry), *ppos, len, flags);
++ int err;
+
+ lockdep_off();
-+ err = vfs_splice_from(pipe, out, ppos, len, flags);
++ err = vfs_remove_acl(idmap, dentry, name);
+ lockdep_on();
-+ if (err >= 0)
-+ au_update_fuse_h_inode(out->f_vfsmnt, out->f_dentry); /*ignore*/
++
+ return err;
+}
++#else
++AuStubInt0(vfsub_set_acl, struct mnt_idmap *idmap, struct dentry *dentry,
++ const char *name, struct posix_acl *acl);
++AuStubInt0(vfsub_remove_acl, struct mnt_idmap *idmap,
++ struct dentry *dentry, const char *name);
+#endif
+
-+/* ---------------------------------------------------------------------- */
-+
-+struct au_vfsub_mkdir_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ int mode;
-+ struct vfsub_args *vargs;
-+};
-+
-+static void au_call_vfsub_mkdir(void *args)
-+{
-+ struct au_vfsub_mkdir_args *a = args;
-+ *a->errp = vfsub_mkdir(a->dir, a->dentry, a->mode, a->vargs);
-+}
-+
-+int vfsub_sio_mkdir(struct au_hinode *hdir, struct dentry *dentry, int mode,
-+ int dlgt)
-+{
-+ int err, do_sio, wkq_err;
-+ struct inode *dir = hdir->hi_inode;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
-+
-+ vfsub_args_init(&vargs, &ign, dlgt, /*force_unlink*/0);
-+ vfsub_ign_hinode(&vargs, IN_CREATE, hdir);
-+ do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE, dlgt);
-+ if (!do_sio)
-+ err = vfsub_mkdir(dir, dentry, mode, &vargs);
-+ else {
-+ struct au_vfsub_mkdir_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .mode = mode,
-+ .vargs = &vargs
-+ };
-+ vfsub_fclr(vargs.flags, DLGT);
-+ wkq_err = au_wkq_wait(au_call_vfsub_mkdir, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
-+
-+ AuTraceErr(err);
-+ return err;
-+}
++#endif /* __KERNEL__ */
++#endif /* __AUFS_VFSUB_H__ */
+diff -urN /usr/share/empty/fs/aufs/wbr_policy.c linux/fs/aufs/wbr_policy.c
+--- /usr/share/empty/fs/aufs/wbr_policy.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/wbr_policy.c 2022-11-05 23:02:18.969222617 +0100
+@@ -0,0 +1,830 @@
++// SPDX-License-Identifier: GPL-2.0
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+struct au_vfsub_rmdir_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ struct vfsub_args *vargs;
-+};
++/*
++ * policies for selecting one among multiple writable branches
++ */
+
-+static void au_call_vfsub_rmdir(void *args)
-+{
-+ struct au_vfsub_rmdir_args *a = args;
-+ *a->errp = vfsub_rmdir(a->dir, a->dentry, a->vargs);
-+}
++#include <linux/statfs.h>
++#include "aufs.h"
+
-+int vfsub_sio_rmdir(struct au_hinode *hdir, struct dentry *dentry, int dlgt)
++/* subset of cpup_attr() */
++static noinline_for_stack
++int au_cpdown_attr(struct path *h_path, struct dentry *h_src)
+{
-+ int err, do_sio, wkq_err;
-+ struct inode *dir = hdir->hi_inode;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
++ int err, sbits;
++ struct iattr ia;
++ struct inode *h_isrc;
+
-+ LKTRTrace("i%lu, %.*s\n", dir->i_ino, AuDLNPair(dentry));
++ h_isrc = d_inode(h_src);
++ ia.ia_valid = ATTR_FORCE | ATTR_MODE | ATTR_UID | ATTR_GID;
++ ia.ia_mode = h_isrc->i_mode;
++ ia.ia_uid = h_isrc->i_uid;
++ ia.ia_gid = h_isrc->i_gid;
++ sbits = !!(ia.ia_mode & (S_ISUID | S_ISGID));
++ au_cpup_attr_flags(d_inode(h_path->dentry), h_isrc->i_flags);
++ /* no delegation since it is just created */
++ err = vfsub_sio_notify_change(h_path, &ia, /*delegated*/NULL);
+
-+ vfsub_args_init(&vargs, &ign, dlgt, /*force_unlink*/0);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, hdir);
-+ do_sio = au_test_h_perm_sio(dir, MAY_EXEC | MAY_WRITE, dlgt);
-+ if (!do_sio)
-+ err = vfsub_rmdir(dir, dentry, &vargs);
-+ else {
-+ struct au_vfsub_rmdir_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .vargs = &vargs
-+ };
-+ vfsub_fclr(vargs.flags, DLGT);
-+ wkq_err = au_wkq_wait(au_call_vfsub_rmdir, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ /* is this nfs only? */
++ if (!err && sbits && au_test_nfs(h_path->dentry->d_sb)) {
++ ia.ia_valid = ATTR_FORCE | ATTR_MODE;
++ ia.ia_mode = h_isrc->i_mode;
++ err = vfsub_sio_notify_change(h_path, &ia, /*delegated*/NULL);
+ }
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+struct notify_change_args {
-+ int *errp;
-+ struct dentry *h_dentry;
-+ struct iattr *ia;
-+ struct vfsub_args *vargs;
-+};
-+
-+static void call_notify_change(void *args)
-+{
-+ struct notify_change_args *a = args;
-+ struct inode *h_inode;
-+
-+ LKTRTrace("%.*s, ia_valid 0x%x\n",
-+ AuDLNPair(a->h_dentry), a->ia->ia_valid);
-+ h_inode = a->h_dentry->d_inode;
-+ IMustLock(h_inode);
-+
-+ *a->errp = -EPERM;
-+ if (!IS_IMMUTABLE(h_inode) && !IS_APPEND(h_inode)) {
-+ vfsub_ignore(a->vargs);
-+ lockdep_off();
-+ *a->errp = notify_change(a->h_dentry, a->ia);
-+ lockdep_on();
-+ if (!*a->errp)
-+ au_update_fuse_h_inode(NULL, a->h_dentry); /*ignore*/
-+ else
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
-+ }
-+ AuTraceErr(*a->errp);
-+}
-+
-+#ifdef CONFIG_AUFS_DLGT
-+static void vfsub_notify_change_dlgt(struct notify_change_args *args,
-+ unsigned int flags)
-+{
-+ if (!vfsub_ftest(flags, DLGT))
-+ call_notify_change(args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_notify_change, args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ *args->errp = wkq_err;
-+ }
-+}
-+#else
-+static void vfsub_notify_change_dlgt(struct notify_change_args *args,
-+ unsigned int flags)
-+{
-+ call_notify_change(args);
-+}
-+#endif
++#define AuCpdown_PARENT_OPQ 1
++#define AuCpdown_WHED (1 << 1)
++#define AuCpdown_MADE_DIR (1 << 2)
++#define AuCpdown_DIROPQ (1 << 3)
++#define au_ftest_cpdown(flags, name) ((flags) & AuCpdown_##name)
++#define au_fset_cpdown(flags, name) \
++ do { (flags) |= AuCpdown_##name; } while (0)
++#define au_fclr_cpdown(flags, name) \
++ do { (flags) &= ~AuCpdown_##name; } while (0)
+
-+int vfsub_notify_change(struct dentry *dentry, struct iattr *ia,
-+ struct vfsub_args *vargs)
++static int au_cpdown_dir_opq(struct dentry *dentry, aufs_bindex_t bdst,
++ unsigned int *flags)
+{
+ int err;
-+ struct notify_change_args args = {
-+ .errp = &err,
-+ .h_dentry = dentry,
-+ .ia = ia,
-+ .vargs = vargs
-+ };
++ struct dentry *opq_dentry;
+
-+ vfsub_notify_change_dlgt(&args, vargs->flags);
++ opq_dentry = au_diropq_create(dentry, bdst);
++ err = PTR_ERR(opq_dentry);
++ if (IS_ERR(opq_dentry))
++ goto out;
++ dput(opq_dentry);
++ au_fset_cpdown(*flags, DIROPQ);
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+int vfsub_sio_notify_change(struct au_hinode *hdir, struct dentry *dentry,
-+ struct iattr *ia)
++static int au_cpdown_dir_wh(struct dentry *dentry, struct dentry *h_parent,
++ struct inode *dir, aufs_bindex_t bdst)
+{
-+ int err, wkq_err;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ __u32 events;
-+ struct notify_change_args args = {
-+ .errp = &err,
-+ .h_dentry = dentry,
-+ .ia = ia,
-+ .vargs = &vargs
-+ };
++ int err;
++ struct path h_path;
++ struct au_branch *br;
+
-+ LKTRTrace("%.*s, 0x%x\n", AuDLNPair(dentry), ia->ia_valid);
++ br = au_sbr(dentry->d_sb, bdst);
++ h_path.dentry = au_wh_lkup(h_parent, &dentry->d_name, br);
++ err = PTR_ERR(h_path.dentry);
++ if (IS_ERR(h_path.dentry))
++ goto out;
+
-+ vfsub_args_init(&vargs, &ign, /*dlgt*/0, /*force_unlink*/0);
-+ events = vfsub_events_notify_change(ia);
-+ if (events)
-+ vfsub_ign_hinode(&vargs, events, hdir);
-+ wkq_err = au_wkq_wait(call_notify_change, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
++ err = 0;
++ if (d_is_positive(h_path.dentry)) {
++ h_path.mnt = au_br_mnt(br);
++ err = au_wh_unlink_dentry(au_h_iptr(dir, bdst), &h_path,
++ dentry);
++ }
++ dput(h_path.dentry);
+
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++static int au_cpdown_dir(struct dentry *dentry, aufs_bindex_t bdst,
++ struct au_pin *pin,
++ struct dentry *h_parent, void *arg)
++{
++ int err, rerr;
++ aufs_bindex_t bopq, btop;
++ struct path h_path;
++ struct dentry *parent;
++ struct inode *h_dir, *h_inode, *inode, *dir;
++ unsigned int *flags = arg;
+
-+struct unlink_args {
-+ int *errp;
-+ struct inode *dir;
-+ struct dentry *dentry;
-+ struct vfsub_args *vargs;
-+};
++ btop = au_dbtop(dentry);
++ /* dentry is di-locked */
++ parent = dget_parent(dentry);
++ dir = d_inode(parent);
++ h_dir = d_inode(h_parent);
++ AuDebugOn(h_dir != au_h_iptr(dir, bdst));
++ IMustLock(h_dir);
+
-+static void call_unlink(void *args)
-+{
-+ struct unlink_args *a = args;
-+ struct inode *h_inode;
-+ const int stop_sillyrename = (au_test_nfs(a->dentry->d_sb)
-+ && atomic_read(&a->dentry->d_count) == 1);
++ err = au_lkup_neg(dentry, bdst, /*wh*/0);
++ if (unlikely(err < 0))
++ goto out;
++ h_path.dentry = au_h_dptr(dentry, bdst);
++ h_path.mnt = au_sbr_mnt(dentry->d_sb, bdst);
++ err = vfsub_sio_mkdir(au_h_iptr(dir, bdst), &h_path, 0755);
++ if (unlikely(err))
++ goto out_put;
++ au_fset_cpdown(*flags, MADE_DIR);
+
-+ LKTRTrace("%.*s, stop_silly %d, cnt %d\n",
-+ AuDLNPair(a->dentry), stop_sillyrename,
-+ atomic_read(&a->dentry->d_count));
++ bopq = au_dbdiropq(dentry);
++ au_fclr_cpdown(*flags, WHED);
++ au_fclr_cpdown(*flags, DIROPQ);
++ if (au_dbwh(dentry) == bdst)
++ au_fset_cpdown(*flags, WHED);
++ if (!au_ftest_cpdown(*flags, PARENT_OPQ) && bopq <= bdst)
++ au_fset_cpdown(*flags, PARENT_OPQ);
++ h_inode = d_inode(h_path.dentry);
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ if (au_ftest_cpdown(*flags, WHED)) {
++ err = au_cpdown_dir_opq(dentry, bdst, flags);
++ if (unlikely(err)) {
++ inode_unlock(h_inode);
++ goto out_dir;
++ }
++ }
+
-+ if (!stop_sillyrename)
-+ dget(a->dentry);
-+ h_inode = a->dentry->d_inode;
-+ if (h_inode)
-+ atomic_inc_return(&h_inode->i_count);
-+ vfsub_ignore(a->vargs);
-+ *a->errp = do_vfsub_unlink(a->dir, a->dentry);
-+ if (unlikely(*a->errp || (a->dentry->d_flags & DCACHE_NFSFS_RENAMED)))
-+ vfsub_unignore(a->vargs);
-+ au_dbg_hin_list(a->vargs);
-+ if (!stop_sillyrename)
-+ dput(a->dentry);
-+ if (h_inode)
-+ iput(h_inode);
++ err = au_cpdown_attr(&h_path, au_h_dptr(dentry, btop));
++ inode_unlock(h_inode);
++ if (unlikely(err))
++ goto out_opq;
+
-+ AuTraceErr(*a->errp);
++ if (au_ftest_cpdown(*flags, WHED)) {
++ err = au_cpdown_dir_wh(dentry, h_parent, dir, bdst);
++ if (unlikely(err))
++ goto out_opq;
++ }
++
++ inode = d_inode(dentry);
++ if (au_ibbot(inode) < bdst)
++ au_set_ibbot(inode, bdst);
++ au_set_h_iptr(inode, bdst, au_igrab(h_inode),
++ au_hi_flags(inode, /*isdir*/1));
++ au_fhsm_wrote(dentry->d_sb, bdst, /*force*/0);
++ goto out; /* success */
++
++ /* revert */
++out_opq:
++ if (au_ftest_cpdown(*flags, DIROPQ)) {
++ inode_lock_nested(h_inode, AuLsc_I_CHILD);
++ rerr = au_diropq_remove(dentry, bdst);
++ inode_unlock(h_inode);
++ if (unlikely(rerr)) {
++ AuIOErr("failed removing diropq for %pd b%d (%d)\n",
++ dentry, bdst, rerr);
++ err = -EIO;
++ goto out;
++ }
++ }
++out_dir:
++ if (au_ftest_cpdown(*flags, MADE_DIR)) {
++ rerr = vfsub_sio_rmdir(au_h_iptr(dir, bdst), &h_path);
++ if (unlikely(rerr)) {
++ AuIOErr("failed removing %pd b%d (%d)\n",
++ dentry, bdst, rerr);
++ err = -EIO;
++ }
++ }
++out_put:
++ au_set_h_dptr(dentry, bdst, NULL);
++ if (au_dbbot(dentry) == bdst)
++ au_update_dbbot(dentry);
++out:
++ dput(parent);
++ return err;
+}
+
-+/*
-+ * @dir: must be locked.
-+ * @dentry: target dentry.
-+ */
-+int vfsub_unlink(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs)
++int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst)
+{
+ int err;
-+ struct unlink_args args = {
-+ .errp = &err,
-+ .dir = dir,
-+ .dentry = dentry,
-+ .vargs = vargs
-+ };
++ unsigned int flags;
+
-+ if (!vfsub_ftest(vargs->flags, DLGT)
-+ && !vfsub_ftest(vargs->flags, FORCE_UNLINK))
-+ call_unlink(&args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_unlink, &args,
-+ vfsub_ftest(vargs->flags, DLGT));
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ }
++ flags = 0;
++ err = au_cp_dirs(dentry, bdst, au_cpdown_dir, &flags);
+
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct statfs_args {
-+ int *errp;
-+ void *arg;
-+ struct kstatfs *buf;
-+};
++/* policies for create */
+
-+static void call_statfs(void *args)
++int au_wbr_nonopq(struct dentry *dentry, aufs_bindex_t bindex)
+{
-+ struct statfs_args *a = args;
-+ *a->errp = vfs_statfs(a->arg, a->buf);
-+}
++ int err, i, j, ndentry;
++ aufs_bindex_t bopq;
++ struct au_dcsub_pages dpages;
++ struct au_dpage *dpage;
++ struct dentry **dentries, *parent, *d;
+
-+#ifdef CONFIG_AUFS_DLGT
-+static void vfsub_statfs_dlgt(struct statfs_args *args, int dlgt)
-+{
-+ if (!dlgt)
-+ call_statfs(args);
-+ else {
-+ int wkq_err;
-+ wkq_err = au_wkq_wait(call_statfs, args, /*dlgt*/1);
-+ if (unlikely(wkq_err))
-+ *args->errp = wkq_err;
++ err = au_dpages_init(&dpages, GFP_NOFS);
++ if (unlikely(err))
++ goto out;
++ parent = dget_parent(dentry);
++ err = au_dcsub_pages_rev_aufs(&dpages, parent, /*do_include*/0);
++ if (unlikely(err))
++ goto out_free;
++
++ err = bindex;
++ for (i = 0; i < dpages.ndpage; i++) {
++ dpage = dpages.dpages + i;
++ dentries = dpage->dentries;
++ ndentry = dpage->ndentry;
++ for (j = 0; j < ndentry; j++) {
++ d = dentries[j];
++ di_read_lock_parent2(d, !AuLock_IR);
++ bopq = au_dbdiropq(d);
++ di_read_unlock(d, !AuLock_IR);
++ if (bopq >= 0 && bopq < err)
++ err = bopq;
++ }
+ }
++
++out_free:
++ dput(parent);
++ au_dpages_free(&dpages);
++out:
++ return err;
+}
-+#else
-+static void vfsub_statfs_dlgt(struct statfs_args *args, int dlgt)
++
++static int au_wbr_bu(struct super_block *sb, aufs_bindex_t bindex)
+{
-+ call_statfs(args);
++ for (; bindex >= 0; bindex--)
++ if (!au_br_rdonly(au_sbr(sb, bindex)))
++ return bindex;
++ return -EROFS;
+}
-+#endif
+
-+int vfsub_statfs(void *arg, struct kstatfs *buf, int dlgt)
++/* top down parent */
++static int au_wbr_create_tdp(struct dentry *dentry,
++ unsigned int flags __maybe_unused)
+{
+ int err;
-+ struct statfs_args args = {
-+ .errp = &err,
-+ .arg = arg,
-+ .buf = buf
-+ };
-+
-+ vfsub_statfs_dlgt(&args, dlgt);
++ aufs_bindex_t btop, bindex;
++ struct super_block *sb;
++ struct dentry *parent, *h_parent;
+
-+ return err;
-+}
-diff -urN linux/fs/aufs/vfsub.h linux-aufs/fs/aufs/vfsub.h
---- linux/fs/aufs/vfsub.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/vfsub.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,530 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ sb = dentry->d_sb;
++ btop = au_dbtop(dentry);
++ err = btop;
++ if (!au_br_rdonly(au_sbr(sb, btop)))
++ goto out;
+
-+/*
-+ * sub-routines for VFS
-+ *
-+ * $Id$
-+ */
++ err = -EROFS;
++ parent = dget_parent(dentry);
++ for (bindex = au_dbtop(parent); bindex < btop; bindex++) {
++ h_parent = au_h_dptr(parent, bindex);
++ if (!h_parent || d_is_negative(h_parent))
++ continue;
+
-+#ifndef __AUFS_VFSUB_H__
-+#define __AUFS_VFSUB_H__
++ if (!au_br_rdonly(au_sbr(sb, bindex))) {
++ err = bindex;
++ break;
++ }
++ }
++ dput(parent);
+
-+#ifdef __KERNEL__
++ /* bottom up here */
++ if (unlikely(err < 0)) {
++ err = au_wbr_bu(sb, btop - 1);
++ if (err >= 0)
++ err = au_wbr_nonopq(dentry, err);
++ }
+
-+#include <linux/fs.h>
-+#include <linux/fs_stack.h>
-+#include <linux/inotify.h>
-+#include <linux/namei.h>
-+#include <linux/security.h>
-+#include <linux/splice.h>
++out:
++ AuDbg("b%d\n", err);
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+/* vfsub flags */
-+#define Vfsub_DLGT 1 /* operation with delegation */
-+#define Vfsub_FORCE_UNLINK (1 << 1) /* force unlinking */
-+#define vfsub_ftest(flags, name) ((flags) & Vfsub_##name)
-+#define vfsub_fset(flags, name) { (flags) |= Vfsub_##name; }
-+#define vfsub_fclr(flags, name) { (flags) &= ~Vfsub_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef Vfsub_DLGT
-+#define Vfsub_DLGT 0
-+#endif
++/* an exception for the policy other than tdp */
++static int au_wbr_create_exp(struct dentry *dentry)
++{
++ int err;
++ aufs_bindex_t bwh, bdiropq;
++ struct dentry *parent;
+
-+struct au_hin_ignore;
-+struct vfsub_args {
-+#ifdef CONFIG_AUFS_HINOTIFY
-+ /* inotify events to be ignored */
-+ int nignore;
-+ struct au_hin_ignore *ignore;
-+#endif
++ err = -1;
++ bwh = au_dbwh(dentry);
++ parent = dget_parent(dentry);
++ bdiropq = au_dbdiropq(parent);
++ if (bwh >= 0) {
++ if (bdiropq >= 0)
++ err = min(bdiropq, bwh);
++ else
++ err = bwh;
++ AuDbg("%d\n", err);
++ } else if (bdiropq >= 0) {
++ err = bdiropq;
++ AuDbg("%d\n", err);
++ }
++ dput(parent);
+
-+ unsigned int flags;
-+};
++ if (err >= 0)
++ err = au_wbr_nonopq(dentry, err);
+
-+struct au_hinode;
-+#ifdef CONFIG_AUFS_HINOTIFY
-+static inline
-+void do_vfsub_args_reinit(struct vfsub_args *vargs, struct au_hin_ignore *ign)
-+{
-+ vargs->nignore = 0;
-+ vargs->ignore = ign;
-+}
++ if (err >= 0 && au_br_rdonly(au_sbr(dentry->d_sb, err)))
++ err = -1;
+
-+static inline void vfsub_args_reinit(struct vfsub_args *vargs)
-+{
-+ vargs->nignore = 0;
++ AuDbg("%d\n", err);
++ return err;
+}
+
-+__u32 vfsub_events_notify_change(struct iattr *ia);
-+void vfsub_ign_hinode(struct vfsub_args *vargs, __u32 events,
-+ struct au_hinode *hinode);
-+void vfsub_ignore(struct vfsub_args *vargs);
-+void vfsub_unignore(struct vfsub_args *vargs);
-+#else
-+static inline
-+void do_vfsub_args_reinit(struct vfsub_args *vargs, struct au_hin_ignore *ign)
-+{
-+ /* empty */
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline void vfsub_args_reinit(struct vfsub_args *vargs)
++/* round robin */
++static int au_wbr_create_init_rr(struct super_block *sb)
+{
-+ /* empty */
-+}
++ int err;
+
-+static inline __u32 vfsub_events_notify_change(struct iattr *ia)
-+{
-+ return 0;
-+}
++ err = au_wbr_bu(sb, au_sbbot(sb));
++ atomic_set(&au_sbi(sb)->si_wbr_rr_next, -err); /* less important */
++ /* smp_mb(); */
+
-+static inline void vfsub_ign_hinode(struct vfsub_args *vargs, __u32 events,
-+ struct au_hinode *hinode)
-+{
-+ /* empty */
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline void vfsub_ignore(struct vfsub_args *vargs)
++static int au_wbr_create_rr(struct dentry *dentry, unsigned int flags)
+{
-+ /* empty */
-+}
++ int err, nbr;
++ unsigned int u;
++ aufs_bindex_t bindex, bbot;
++ struct super_block *sb;
++ atomic_t *next;
+
-+static inline void vfsub_unignore(struct vfsub_args *vargs)
-+{
-+ /* empty */
++ err = au_wbr_create_exp(dentry);
++ if (err >= 0)
++ goto out;
++
++ sb = dentry->d_sb;
++ next = &au_sbi(sb)->si_wbr_rr_next;
++ bbot = au_sbbot(sb);
++ nbr = bbot + 1;
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ if (!au_ftest_wbr(flags, DIR)) {
++ err = atomic_dec_return(next) + 1;
++ /* modulo for 0 is meaningless */
++ if (unlikely(!err))
++ err = atomic_dec_return(next) + 1;
++ } else
++ err = atomic_read(next);
++ AuDbg("%d\n", err);
++ u = err;
++ err = u % nbr;
++ AuDbg("%d\n", err);
++ if (!au_br_rdonly(au_sbr(sb, err)))
++ break;
++ err = -EROFS;
++ }
++
++ if (err >= 0)
++ err = au_wbr_nonopq(dentry, err);
++
++out:
++ AuDbg("%d\n", err);
++ return err;
+}
-+#endif /* CONFIG_AUFS_HINOTIFY */
-+
-+void vfsub_args_init(struct vfsub_args *vargs, struct au_hin_ignore *ign,
-+ int dlgt, int force_unlink);
+
+/* ---------------------------------------------------------------------- */
+
-+/* inotify_inode_watched() is not exported */
-+static inline int au_test_inotify(struct inode *inode)
++/* most free space */
++static void au_mfs(struct dentry *dentry, struct dentry *parent)
+{
-+#ifdef CONFIG_INOTIFY
-+ return !list_empty(&inode->inotify_watches);
-+#endif
-+ return 0;
-+}
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_wbr_mfs *mfs;
++ struct dentry *h_parent;
++ aufs_bindex_t bindex, bbot;
++ int err;
++ unsigned long long b, bavail;
++ struct path h_path;
++ /* reduce the stack usage */
++ struct kstatfs *st;
+
-+/* ---------------------------------------------------------------------- */
++ st = kmalloc(sizeof(*st), GFP_NOFS);
++ if (unlikely(!st)) {
++ AuWarn1("failed updating mfs(%d), ignored\n", -ENOMEM);
++ return;
++ }
+
-+/* lock subclass for hidden inode */
-+/* default MAX_LOCKDEP_SUBCLASSES(8) is not enough */
-+/* reduce? gave up. */
-+enum {
-+ AuLsc_I_Begin = I_MUTEX_QUOTA, /* 4 */
-+ AuLsc_I_PARENT, /* hidden inode, parent first */
-+ AuLsc_I_PARENT2, /* copyup dirs */
-+ AuLsc_I_PARENT3, /* rename with hinotify */
-+ AuLsc_I_PARENT4, /* ditto */
-+ AuLsc_I_CHILD,
-+ AuLsc_I_CHILD2,
-+ AuLsc_I_End
-+};
++ bavail = 0;
++ sb = dentry->d_sb;
++ mfs = &au_sbi(sb)->si_wbr_mfs;
++ MtxMustLock(&mfs->mfs_lock);
++ mfs->mfs_bindex = -EROFS;
++ mfs->mfsrr_bytes = 0;
++ if (!parent) {
++ bindex = 0;
++ bbot = au_sbbot(sb);
++ } else {
++ bindex = au_dbtop(parent);
++ bbot = au_dbtaildir(parent);
++ }
++
++ for (; bindex <= bbot; bindex++) {
++ if (parent) {
++ h_parent = au_h_dptr(parent, bindex);
++ if (!h_parent || d_is_negative(h_parent))
++ continue;
++ }
++ br = au_sbr(sb, bindex);
++ if (au_br_rdonly(br))
++ continue;
+
-+#define IMustLock(i) MtxMustLock(&(i)->i_mutex)
++ /* sb->s_root for NFS is unreliable */
++ h_path.mnt = au_br_mnt(br);
++ h_path.dentry = h_path.mnt->mnt_root;
++ err = vfs_statfs(&h_path, st);
++ if (unlikely(err)) {
++ AuWarn1("failed statfs, b%d, %d\n", bindex, err);
++ continue;
++ }
+
-+static inline void vfsub_lock_rename_mutex(struct super_block *sb)
-+{
-+ lockdep_off();
-+ mutex_lock(&sb->s_vfs_rename_mutex);
-+ lockdep_on();
-+}
++ /* when the available size is equal, select the lower one */
++ BUILD_BUG_ON(sizeof(b) < sizeof(st->f_bavail)
++ || sizeof(b) < sizeof(st->f_bsize));
++ b = st->f_bavail * st->f_bsize;
++ br->br_wbr->wbr_bytes = b;
++ if (b >= bavail) {
++ bavail = b;
++ mfs->mfs_bindex = bindex;
++ mfs->mfs_jiffy = jiffies;
++ }
++ }
+
-+static inline void vfsub_unlock_rename_mutex(struct super_block *sb)
-+{
-+ lockdep_off();
-+ mutex_unlock(&sb->s_vfs_rename_mutex);
-+ lockdep_on();
++ mfs->mfsrr_bytes = bavail;
++ AuDbg("b%d\n", mfs->mfs_bindex);
++ au_kfree_rcu(st);
+}
+
-+static inline int vfsub_is_rename_mutex_locked(struct super_block *sb)
++static int au_wbr_create_mfs(struct dentry *dentry, unsigned int flags)
+{
-+ int res;
-+
-+ lockdep_off();
-+ res = mutex_is_locked(&sb->s_vfs_rename_mutex);
-+ lockdep_on();
++ int err;
++ struct dentry *parent;
++ struct super_block *sb;
++ struct au_wbr_mfs *mfs;
+
-+ return res;
-+}
++ err = au_wbr_create_exp(dentry);
++ if (err >= 0)
++ goto out;
+
-+static inline
-+struct dentry *vfsub_lock_rename(struct dentry *d1, struct dentry *d2)
-+{
-+ struct dentry *d;
++ sb = dentry->d_sb;
++ parent = NULL;
++ if (au_ftest_wbr(flags, PARENT))
++ parent = dget_parent(dentry);
++ 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_br_rdonly(au_sbr(sb, mfs->mfs_bindex)))
++ au_mfs(dentry, parent);
++ mutex_unlock(&mfs->mfs_lock);
++ err = mfs->mfs_bindex;
++ dput(parent);
+
-+ lockdep_off();
-+ d = lock_rename(d1, d2);
-+ lockdep_on();
-+ return d;
-+}
++ if (err >= 0)
++ err = au_wbr_nonopq(dentry, err);
+
-+static inline void vfsub_unlock_rename(struct dentry *d1, struct dentry *d2)
-+{
-+ lockdep_off();
-+ unlock_rename(d1, d2);
-+ lockdep_on();
++out:
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline int au_verify_parent(struct dentry *dentry, struct inode *dir)
++static int au_wbr_create_init_mfs(struct super_block *sb)
+{
-+ IMustLock(dir);
-+ return /* !dir->i_nlink || */ dentry->d_parent->d_inode != dir;
-+}
++ struct au_wbr_mfs *mfs;
+
-+/* ---------------------------------------------------------------------- */
++ mfs = &au_sbi(sb)->si_wbr_mfs;
++ mutex_init(&mfs->mfs_lock);
++ mfs->mfs_jiffy = 0;
++ mfs->mfs_bindex = -EROFS;
+
-+#ifdef CONFIG_AUFS_WORKAROUND_FUSE
-+/* br_fuse.c */
-+int au_update_fuse_h_inode(struct vfsmount *h_mnt, struct dentry *h_dentry);
-+#else
-+static inline
-+int au_update_fuse_h_inode(struct vfsmount *h_mnt, struct dentry *h_dentry)
-+{
+ return 0;
+}
-+#endif
-+
-+#ifdef CONFIG_AUFS_BR_XFS
-+/* br_xfs.c */
-+dev_t au_h_rdev(struct inode *h_inode, struct vfsmount *h_mnt,
-+ struct dentry *h_dentry);
-+#else
-+static inline
-+dev_t au_h_rdev(struct inode *h_inode, struct vfsmount *h_mnt,
-+ struct dentry *h_dentry)
-+{
-+ return h_inode->i_rdev;
-+}
-+#endif
-+
-+/* simple abstractions, for future use */
-+static inline
-+int do_vfsub_permission(struct inode *inode, int mask, struct nameidata *nd)
-+{
-+ LKTRTrace("i%lu, mask 0x%x, nd %d\n", inode->i_ino, mask, !!nd);
-+ IMustLock(inode);
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 27)
-+ return inode_permission(inode, mask);
-+#else
-+ return permission(inode, mask, nd);
-+#endif
-+}
+
-+static inline
-+int vfsub_security_inode_permission(struct inode *inode, int mask,
-+ struct nameidata *nd)
++static int au_wbr_create_fin_mfs(struct super_block *sb __maybe_unused)
+{
-+#if LINUX_VERSION_CODE >= KERNEL_VERSION(2, 6, 27)
-+ mask &= MAY_READ | MAY_WRITE | MAY_EXEC | MAY_APPEND;
-+ return security_inode_permission(inode, mask);
-+#else
-+ return security_inode_permission(inode, mask, nd);
-+#endif
++ mutex_destroy(&au_sbi(sb)->si_wbr_mfs.mfs_lock);
++ return 0;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct file *vfsub_filp_open(const char *path, int oflags, int mode);
-+int vfsub_path_lookup(const char *name, unsigned int flags,
-+ struct nameidata *nd);
-+struct dentry *vfsub_lookup_one_len(const char *name, struct dentry *parent,
-+ int len);
-+
-+#ifdef CONFIG_AUFS_LHASH_PATCH
-+struct dentry *vfsub__lookup_hash(struct qstr *name, struct dentry *parent,
-+ struct nameidata *nd);
-+#endif
++/* 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);
+
-+int do_vfsub_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd);
-+int do_vfsub_symlink(struct inode *dir, struct dentry *dentry,
-+ const char *symname, int mode);
-+int do_vfsub_mknod(struct inode *dir, struct dentry *dentry, int mode,
-+ dev_t dev);
-+int do_vfsub_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry);
-+int do_vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry);
-+int do_vfsub_mkdir(struct inode *dir, struct dentry *dentry, int mode);
-+int do_vfsub_rmdir(struct inode *dir, struct dentry *dentry);
-+int do_vfsub_unlink(struct inode *dir, struct dentry *dentry);
++ /* 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;
+
-+ssize_t do_vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
-+ loff_t *ppos);
-+/* todo: kernel_read()? */
-+ssize_t do_vfsub_read_k(struct file *file, void *kbuf, size_t count,
-+ loff_t *ppos);
-+ssize_t do_vfsub_write_u(struct file *file, const char __user *ubuf,
-+ size_t count, loff_t *ppos);
-+ssize_t do_vfsub_write_k(struct file *file, void *kbuf, size_t count,
-+ loff_t *ppos);
-+int do_vfsub_readdir(struct file *file, filldir_t filldir, void *arg);
++out:
++ dput(parent);
++ AuDbg("b%d\n", err);
++ return err;
++}
+
+/* ---------------------------------------------------------------------- */
+
-+#ifndef CONFIG_AUFS_UNIONFS22_PATCH
-+static inline void vfsub_copy_inode_size(struct inode *inode,
-+ struct inode *h_inode)
-+{
-+ spin_lock(&inode->i_lock);
-+ fsstack_copy_inode_size(inode, h_inode);
-+ spin_unlock(&inode->i_lock);
-+}
-+#else
-+static inline void vfsub_copy_inode_size(struct inode *inode,
-+ struct inode *h_inode)
++/* most free space and then round robin */
++static int au_wbr_create_mfsrr(struct dentry *dentry, unsigned int flags)
+{
-+ fsstack_copy_inode_size(inode, h_inode);
-+}
-+#endif
++ int err;
++ struct au_wbr_mfs *mfs;
+
-+#ifndef CONFIG_AUFS_UNIONFS23_PATCH
-+#define vfs_splice_to do_splice_to
-+#define vfs_splice_from do_splice_from
-+#endif
++ err = au_wbr_create_mfs(dentry, flags);
++ if (err >= 0) {
++ mfs = &au_sbi(dentry->d_sb)->si_wbr_mfs;
++ mutex_lock(&mfs->mfs_lock);
++ if (mfs->mfsrr_bytes < mfs->mfsrr_watermark)
++ err = au_wbr_create_rr(dentry, flags);
++ mutex_unlock(&mfs->mfs_lock);
++ }
+
-+#ifdef CONFIG_AUFS_SPLICE_PATCH
-+long do_vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags);
-+long do_vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags);
-+#else
-+static inline
-+long do_vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags)
-+{
-+ return -ENOSYS;
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline
-+long do_vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags)
++static int au_wbr_create_init_mfsrr(struct super_block *sb)
+{
-+ return -ENOSYS;
++ int err;
++
++ au_wbr_create_init_mfs(sb); /* ignore */
++ err = au_wbr_create_init_rr(sb);
++
++ return err;
+}
-+#endif /* CONFIG_AUFS_SPLICE_PATCH */
+
+/* ---------------------------------------------------------------------- */
+
-+static inline loff_t vfsub_llseek(struct file *file, loff_t offset, int origin)
++/* top down parent and most free space */
++static int au_wbr_create_pmfs(struct dentry *dentry, unsigned int flags)
+{
-+ loff_t err;
++ int err, e2;
++ unsigned long long b;
++ aufs_bindex_t bindex, btop, bbot;
++ struct super_block *sb;
++ struct dentry *parent, *h_parent;
++ struct au_branch *br;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(file->f_dentry));
++ err = au_wbr_create_tdp(dentry, flags);
++ if (unlikely(err < 0))
++ goto out;
++ parent = dget_parent(dentry);
++ btop = au_dbtop(parent);
++ bbot = au_dbtaildir(parent);
++ if (btop == bbot)
++ goto out_parent; /* success */
+
-+ lockdep_off();
-+ err = vfs_llseek(file, offset, origin);
-+ lockdep_on();
-+ return err;
-+}
++ e2 = au_wbr_create_mfs(dentry, flags);
++ if (e2 < 0)
++ goto out_parent; /* success */
+
-+static inline int do_vfsub_getattr(struct vfsmount *mnt, struct dentry *dentry,
-+ struct kstat *st)
-+{
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+ return vfs_getattr(mnt, dentry, st);
-+}
++ /* when the available size is equal, select upper one */
++ sb = dentry->d_sb;
++ br = au_sbr(sb, err);
++ b = br->br_wbr->wbr_bytes;
++ AuDbg("b%d, %llu\n", err, b);
+
-+/* ---------------------------------------------------------------------- */
++ for (bindex = btop; bindex <= bbot; bindex++) {
++ h_parent = au_h_dptr(parent, bindex);
++ if (!h_parent || d_is_negative(h_parent))
++ continue;
+
-+#ifdef CONFIG_AUFS_HIN_OR_DLGT
-+/* hin_or_dlgt.c */
-+int vfsub_permission(struct inode *inode, int mask, struct nameidata *nd,
-+ int dlgt);
++ br = au_sbr(sb, bindex);
++ if (!au_br_rdonly(br) && br->br_wbr->wbr_bytes > b) {
++ b = br->br_wbr->wbr_bytes;
++ err = bindex;
++ AuDbg("b%d, %llu\n", err, b);
++ }
++ }
+
-+int vfsub_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd, struct vfsub_args *vargs);
-+int vfsub_symlink(struct inode *dir, struct dentry *dentry, const char *symname,
-+ int mode, struct vfsub_args *vargs);
-+int vfsub_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev,
-+ struct vfsub_args *vargs);
-+int vfsub_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry, struct vfsub_args *vargs);
-+int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs);
-+int vfsub_mkdir(struct inode *dir, struct dentry *dentry, int mode,
-+ struct vfsub_args *vargs);
-+int vfsub_rmdir(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs);
++ if (err >= 0)
++ err = au_wbr_nonopq(dentry, err);
+
-+ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
-+ loff_t *ppos, int dlgt);
-+ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ int dlgt);
-+ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count,
-+ loff_t *ppos, struct vfsub_args *vargs);
-+ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ struct vfsub_args *vargs);
-+int vfsub_readdir(struct file *file, filldir_t filldir, void *arg, int dlgt);
-+long vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags, int dlgt);
-+long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags,
-+ struct vfsub_args *vargs);
++out_parent:
++ dput(parent);
++out:
++ AuDbg("b%d\n", err);
++ return err;
++}
+
-+int vfsub_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *st,
-+ int dlgt);
-+#else
++/* ---------------------------------------------------------------------- */
+
-+static inline
-+int vfsub_permission(struct inode *inode, int mask, struct nameidata *nd,
-+ int dlgt)
++/*
++ * - top down parent
++ * - most free space with parent
++ * - most free space round-robin regardless parent
++ */
++static int au_wbr_create_pmfsrr(struct dentry *dentry, unsigned int flags)
+{
-+ return do_vfsub_permission(inode, mask, nd);
-+}
++ int err;
++ unsigned long long watermark;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct au_wbr_mfs *mfs;
+
-+static inline
-+int vfsub_create(struct inode *dir, struct dentry *dentry, int mode,
-+ struct nameidata *nd, struct vfsub_args *vargs)
-+{
-+ return do_vfsub_create(dir, dentry, mode, nd);
-+}
++ err = au_wbr_create_pmfs(dentry, flags | AuWbr_PARENT);
++ if (unlikely(err < 0))
++ goto out;
+
-+static inline
-+int vfsub_symlink(struct inode *dir, struct dentry *dentry, const char *symname,
-+ int mode, struct vfsub_args *vargs)
-+{
-+ return do_vfsub_symlink(dir, dentry, symname, mode);
++ sb = dentry->d_sb;
++ br = au_sbr(sb, err);
++ mfs = &au_sbi(sb)->si_wbr_mfs;
++ mutex_lock(&mfs->mfs_lock);
++ watermark = mfs->mfsrr_watermark;
++ mutex_unlock(&mfs->mfs_lock);
++ if (br->br_wbr->wbr_bytes < watermark)
++ /* regardless the parent dir */
++ err = au_wbr_create_mfsrr(dentry, flags);
++
++out:
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline
-+int vfsub_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t dev,
-+ struct vfsub_args *vargs)
-+{
-+ return do_vfsub_mknod(dir, dentry, mode, dev);
-+}
++/* ---------------------------------------------------------------------- */
+
-+static inline
-+int vfsub_link(struct dentry *src_dentry, struct inode *dir,
-+ struct dentry *dentry, struct vfsub_args *vargs)
-+{
-+ return do_vfsub_link(src_dentry, dir, dentry);
-+}
++/* policies for copyup */
+
-+static inline
-+int vfsub_rename(struct inode *src_dir, struct dentry *src_dentry,
-+ struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs)
++/* top down parent */
++static int au_wbr_copyup_tdp(struct dentry *dentry)
+{
-+ return do_vfsub_rename(src_dir, src_dentry, dir, dentry);
++ return au_wbr_create_tdp(dentry, /*flags, anything is ok*/0);
+}
+
-+static inline
-+int vfsub_mkdir(struct inode *dir, struct dentry *dentry, int mode,
-+ struct vfsub_args *vargs)
++/* bottom up parent */
++static int au_wbr_copyup_bup(struct dentry *dentry)
+{
-+ return do_vfsub_mkdir(dir, dentry, mode);
-+}
++ int err;
++ aufs_bindex_t bindex, btop;
++ struct dentry *parent, *h_parent;
++ struct super_block *sb;
+
-+static inline
-+int vfsub_rmdir(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs)
-+{
-+ return do_vfsub_rmdir(dir, dentry);
-+}
++ err = -EROFS;
++ sb = dentry->d_sb;
++ parent = dget_parent(dentry);
++ btop = au_dbtop(parent);
++ for (bindex = au_dbtop(dentry); bindex >= btop; bindex--) {
++ h_parent = au_h_dptr(parent, bindex);
++ if (!h_parent || d_is_negative(h_parent))
++ continue;
+
-+static inline
-+ssize_t vfsub_read_u(struct file *file, char __user *ubuf, size_t count,
-+ loff_t *ppos, int dlgt)
-+{
-+ return do_vfsub_read_u(file, ubuf, count, ppos);
-+}
++ if (!au_br_rdonly(au_sbr(sb, bindex))) {
++ err = bindex;
++ break;
++ }
++ }
++ dput(parent);
+
-+static inline
-+ssize_t vfsub_read_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ int dlgt)
-+{
-+ return do_vfsub_read_k(file, kbuf, count, ppos);
-+}
++ /* bottom up here */
++ if (unlikely(err < 0))
++ err = au_wbr_bu(sb, btop - 1);
+
-+static inline
-+ssize_t vfsub_write_u(struct file *file, const char __user *ubuf, size_t count,
-+ loff_t *ppos, struct vfsub_args *vargs)
-+{
-+ return do_vfsub_write_u(file, ubuf, count, ppos);
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline
-+ssize_t vfsub_write_k(struct file *file, void *kbuf, size_t count, loff_t *ppos,
-+ struct vfsub_args *vargs)
++/* bottom up */
++int au_wbr_do_copyup_bu(struct dentry *dentry, aufs_bindex_t btop)
+{
-+ return do_vfsub_write_k(file, kbuf, count, ppos);
-+}
++ int err;
+
-+static inline
-+int vfsub_readdir(struct file *file, filldir_t filldir, void *arg, int dlgt)
-+{
-+ return do_vfsub_readdir(file, filldir, arg);
-+}
++ err = au_wbr_bu(dentry->d_sb, btop);
++ AuDbg("b%d\n", err);
++ if (err > btop)
++ err = au_wbr_nonopq(dentry, err);
+
-+static inline
-+long vfsub_splice_to(struct file *in, loff_t *ppos,
-+ struct pipe_inode_info *pipe, size_t len,
-+ unsigned int flags, int dlgt)
-+{
-+ return do_vfsub_splice_to(in, ppos, pipe, len, flags);
++ AuDbg("b%d\n", err);
++ return err;
+}
+
-+static inline
-+long vfsub_splice_from(struct pipe_inode_info *pipe, struct file *out,
-+ loff_t *ppos, size_t len, unsigned int flags,
-+ struct vfsub_args *vargs)
++static int au_wbr_copyup_bu(struct dentry *dentry)
+{
-+ return do_vfsub_splice_from(pipe, out, ppos, len, flags);
-+}
++ int err;
++ aufs_bindex_t btop;
+
-+static inline
-+int vfsub_getattr(struct vfsmount *mnt, struct dentry *dentry, struct kstat *st,
-+ int dlgt)
-+{
-+ return do_vfsub_getattr(mnt, dentry, st);
++ btop = au_dbtop(dentry);
++ err = au_wbr_do_copyup_bu(dentry, btop);
++ return err;
+}
-+#endif /* HIN_OR_DLGT */
-+
-+/* ---------------------------------------------------------------------- */
-+
-+int vfsub_sio_mkdir(struct au_hinode *hdir, struct dentry *dentry, int mode,
-+ int dlgt);
-+int vfsub_sio_rmdir(struct au_hinode *hdir, struct dentry *dentry, int dlgt);
-+int vfsub_sio_notify_change(struct au_hinode *hdir, struct dentry *dentry,
-+ struct iattr *ia);
+
+/* ---------------------------------------------------------------------- */
+
-+int vfsub_notify_change(struct dentry *dentry, struct iattr *ia,
-+ struct vfsub_args *vargs);
-+int vfsub_unlink(struct inode *dir, struct dentry *dentry,
-+ struct vfsub_args *vargs);
-+int vfsub_statfs(void *arg, struct kstatfs *buf, int dlgt);
++struct au_wbr_copyup_operations au_wbr_copyup_ops[] = {
++ [AuWbrCopyup_TDP] = {
++ .copyup = au_wbr_copyup_tdp
++ },
++ [AuWbrCopyup_BUP] = {
++ .copyup = au_wbr_copyup_bup
++ },
++ [AuWbrCopyup_BU] = {
++ .copyup = au_wbr_copyup_bu
++ }
++};
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_VFSUB_H__ */
-diff -urN linux/fs/aufs/wbr_policy.c linux-aufs/fs/aufs/wbr_policy.c
---- linux/fs/aufs/wbr_policy.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/wbr_policy.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,696 @@
++struct au_wbr_create_operations au_wbr_create_ops[] = {
++ [AuWbrCreate_TDP] = {
++ .create = au_wbr_create_tdp
++ },
++ [AuWbrCreate_RR] = {
++ .create = au_wbr_create_rr,
++ .init = au_wbr_create_init_rr
++ },
++ [AuWbrCreate_MFS] = {
++ .create = au_wbr_create_mfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_MFSV] = {
++ .create = au_wbr_create_mfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_MFSRR] = {
++ .create = au_wbr_create_mfsrr,
++ .init = au_wbr_create_init_mfsrr,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_MFSRRV] = {
++ .create = au_wbr_create_mfsrr,
++ .init = au_wbr_create_init_mfsrr,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_TDMFS] = {
++ .create = au_wbr_create_tdmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_TDMFSV] = {
++ .create = au_wbr_create_tdmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_PMFS] = {
++ .create = au_wbr_create_pmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_PMFSV] = {
++ .create = au_wbr_create_pmfs,
++ .init = au_wbr_create_init_mfs,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_PMFSRR] = {
++ .create = au_wbr_create_pmfsrr,
++ .init = au_wbr_create_init_mfsrr,
++ .fin = au_wbr_create_fin_mfs
++ },
++ [AuWbrCreate_PMFSRRV] = {
++ .create = au_wbr_create_pmfsrr,
++ .init = au_wbr_create_init_mfsrr,
++ .fin = au_wbr_create_fin_mfs
++ }
++};
+diff -urN /usr/share/empty/fs/aufs/whout.c linux/fs/aufs/whout.c
+--- /usr/share/empty/fs/aufs/whout.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/whout.c 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,1072 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * policies for selecting one among multiple writable branches
-+ *
-+ * $Id$
++ * whiteout for logical deletion and opaque directory
+ */
+
-+#include <linux/statfs.h>
+#include "aufs.h"
+
-+static int au_cpdown_attr(struct au_hinode *hdir, struct dentry *h_dst,
-+ struct dentry *h_src)
-+{
-+ int err, sbits;
-+ struct iattr ia;
-+ struct inode *h_idst, *h_isrc;
++#define WH_MASK 0444
+
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dst));
-+ h_idst = h_dst->d_inode;
-+ /* todo? IMustLock(h_idst); */
-+ h_isrc = h_src->d_inode;
-+ /* todo? IMustLock(h_isrc); */
++/*
++ * If a directory contains this file, then it is opaque. We start with the
++ * .wh. flag so that it is blocked by lookup.
++ */
++static struct qstr diropq_name = QSTR_INIT(AUFS_WH_DIROPQ,
++ sizeof(AUFS_WH_DIROPQ) - 1);
+
-+ ia.ia_valid = ATTR_FORCE | ATTR_MODE | ATTR_UID | ATTR_GID;
-+ ia.ia_mode = h_isrc->i_mode;
-+ ia.ia_uid = h_isrc->i_uid;
-+ ia.ia_gid = h_isrc->i_gid;
-+ sbits = !!(ia.ia_mode & (S_ISUID | S_ISGID));
-+ au_cpup_attr_flags(h_idst, h_isrc);
-+ err = vfsub_sio_notify_change(hdir, h_dst, &ia);
++/*
++ * generate whiteout name, which is NOT terminated by NULL.
++ * @name: original d_name.name
++ * @len: original d_name.len
++ * @wh: whiteout qstr
++ * returns zero when succeeds, otherwise error.
++ * succeeded value as wh->name should be freed by kfree().
++ */
++int au_wh_name_alloc(struct qstr *wh, const struct qstr *name)
++{
++ char *p;
+
-+ /* is this nfs only? */
-+ if (!err && sbits && au_test_nfs(h_dst->d_sb)) {
-+ ia.ia_valid = ATTR_FORCE | ATTR_MODE;
-+ ia.ia_mode = h_isrc->i_mode;
-+ err = vfsub_sio_notify_change(hdir, h_dst, &ia);
-+ }
++ if (unlikely(name->len > PATH_MAX - AUFS_WH_PFX_LEN))
++ return -ENAMETOOLONG;
+
-+ AuTraceErr(err);
-+ return err;
++ wh->len = name->len + AUFS_WH_PFX_LEN;
++ p = kmalloc(wh->len, GFP_NOFS);
++ wh->name = p;
++ if (p) {
++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
++ memcpy(p + AUFS_WH_PFX_LEN, name->name, name->len);
++ /* smp_mb(); */
++ return 0;
++ }
++ return -ENOMEM;
+}
+
-+struct au_cpdown_dir_args {
-+ struct dentry *parent;
-+ unsigned int parent_opq; /* bit-flags */
-+};
++/* ---------------------------------------------------------------------- */
+
-+static int au_cpdown_dir(struct dentry *dentry, aufs_bindex_t bdst,
-+ struct dentry *h_parent, void *arg)
++/*
++ * test if the @wh_name exists under @h_ppath.
++ * @try_sio specifies the necessary of super-io.
++ */
++int au_wh_test(struct mnt_idmap *h_idmap, struct path *h_ppath,
++ struct qstr *wh_name, int try_sio)
+{
-+ int err, rerr;
-+ struct au_cpdown_dir_args *args = arg;
-+ aufs_bindex_t bend, bopq, bstart;
-+ unsigned char parent_opq, whed, dlgt, do_opq, made_dir, diropq;
-+ struct dentry *h_dentry, *opq_dentry, *wh_dentry, *parent;
-+ struct inode *h_dir, *h_inode, *inode, *dir;
-+
-+ LKTRTrace("%.*s, b%d\n", AuDLNPair(dentry), bdst);
-+ bstart = au_dbstart(dentry);
-+ AuDebugOn(bstart <= bdst
-+ && bdst <= au_dbend(dentry)
-+ && au_h_dptr(dentry, bdst));
-+ AuDebugOn(!h_parent);
-+ /* todo: safe? */
-+ parent = dget_parent(dentry);
-+ dir = parent->d_inode;
-+ dput(parent);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!h_dir);
-+ AuDebugOn(h_dir != au_h_iptr(dir, bdst));
-+ IMustLock(h_dir);
++ int err;
++ struct dentry *wh_dentry;
+
-+ err = au_lkup_neg(dentry, bdst);
-+ if (unlikely(err < 0))
++ if (!try_sio)
++ wh_dentry = vfsub_lkup_one(wh_name, h_ppath);
++ else
++ wh_dentry = au_sio_lkup_one(h_idmap, wh_name, h_ppath);
++ err = PTR_ERR(wh_dentry);
++ if (IS_ERR(wh_dentry)) {
++ if (err == -ENAMETOOLONG)
++ err = 0;
+ goto out;
-+ h_dentry = au_h_dptr(dentry, bdst);
-+ dlgt = !!au_test_dlgt(au_mntflags(dentry->d_sb));
-+ err = vfsub_sio_mkdir(au_hi(dir, bdst), h_dentry,
-+ S_IRWXU | S_IRUGO | S_IXUGO, dlgt);
-+ if (unlikely(err))
-+ goto out_put;
-+
-+ made_dir = 1;
-+ bend = au_dbend(dentry);
-+ bopq = au_dbdiropq(dentry);
-+ whed = (au_dbwh(dentry) == bdst);
-+ if (!args->parent_opq)
-+ args->parent_opq |= (bopq <= bdst);
-+ parent_opq = (args->parent_opq && args->parent == dentry);
-+ do_opq = 0;
-+ diropq = 0;
-+ h_inode = h_dentry->d_inode;
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ if (whed || (parent_opq && do_opq)) {
-+ opq_dentry = au_diropq_create(dentry, bdst, dlgt);
-+ err = PTR_ERR(opq_dentry);
-+ if (IS_ERR(opq_dentry)) {
-+ mutex_unlock(&h_inode->i_mutex);
-+ goto out_dir;
-+ }
-+ dput(opq_dentry);
-+ diropq = 1;
+ }
+
-+ err = au_cpdown_attr(au_hi(dir, bdst), h_dentry,
-+ au_h_dptr(dentry, bstart));
-+ mutex_unlock(&h_inode->i_mutex);
-+ if (unlikely(err))
-+ goto out_opq;
++ err = 0;
++ if (d_is_negative(wh_dentry))
++ goto out_wh; /* success */
+
-+ wh_dentry = NULL;
-+ if (whed) {
-+ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, /*ndx*/NULL);
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out_opq;
-+ err = 0;
-+ if (wh_dentry->d_inode)
-+ err = au_wh_unlink_dentry(au_hi(dir, bdst), wh_dentry,
-+ dentry, dlgt);
-+ dput(wh_dentry);
-+ if (unlikely(err))
-+ goto out_opq;
-+ }
++ err = 1;
++ if (d_is_reg(wh_dentry))
++ goto out_wh; /* success */
+
-+ inode = dentry->d_inode;
-+ if (au_ibend(inode) < bdst)
-+ au_set_ibend(inode, bdst);
-+ au_set_h_iptr(inode, bdst, au_igrab(h_inode), au_hi_flags(inode, 1));
-+ goto out; /* success */
++ err = -EIO;
++ AuIOErr("%pd Invalid whiteout entry type 0%o.\n",
++ wh_dentry, d_inode(wh_dentry)->i_mode);
+
-+ /* revert */
-+ out_opq:
-+ if (diropq) {
-+ mutex_lock_nested(&h_inode->i_mutex, AuLsc_I_CHILD);
-+ rerr = au_diropq_remove(dentry, bdst, dlgt);
-+ mutex_unlock(&h_inode->i_mutex);
-+ if (unlikely(rerr)) {
-+ AuIOErr("failed removing diropq for %.*s b%d (%d)\n",
-+ AuDLNPair(dentry), bdst, rerr);
-+ err = -EIO;
++out_wh:
++ dput(wh_dentry);
++out:
++ return err;
++}
++
++/*
++ * test if the @h_path->dentry sets opaque or not.
++ */
++int au_diropq_test(struct mnt_idmap *h_idmap, struct path *h_path)
++{
++ int err;
++ struct inode *h_dir;
++
++ h_dir = d_inode(h_path->dentry);
++ err = au_wh_test(h_idmap, h_path, &diropq_name,
++ au_test_h_perm_sio(h_idmap, h_dir, MAY_EXEC));
++ return err;
++}
++
++/*
++ * returns a negative dentry whose name is unique and temporary.
++ */
++struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br,
++ struct qstr *prefix)
++{
++ struct dentry *dentry;
++ int i;
++ char defname[NAME_MAX - AUFS_MAX_NAMELEN + DNAME_INLINE_LEN + 1],
++ *name, *p;
++ /* strict atomic_t is unnecessary here */
++ static unsigned short cnt;
++ struct qstr qs;
++ struct path h_ppath;
++ struct mnt_idmap *h_idmap;
++
++ BUILD_BUG_ON(sizeof(cnt) * 2 > AUFS_WH_TMP_LEN);
++
++ name = defname;
++ qs.len = sizeof(defname) - DNAME_INLINE_LEN + prefix->len - 1;
++ if (unlikely(prefix->len > DNAME_INLINE_LEN)) {
++ dentry = ERR_PTR(-ENAMETOOLONG);
++ if (unlikely(qs.len > NAME_MAX))
++ goto out;
++ dentry = ERR_PTR(-ENOMEM);
++ name = kmalloc(qs.len + 1, GFP_NOFS);
++ if (unlikely(!name))
+ goto out;
-+ }
+ }
-+ out_dir:
-+ if (made_dir) {
-+ rerr = vfsub_sio_rmdir(au_hi(dir, bdst), h_dentry, dlgt);
-+ if (unlikely(rerr)) {
-+ AuIOErr("failed removing %.*s b%d (%d)\n",
-+ AuDLNPair(dentry), bdst, rerr);
-+ err = -EIO;
-+ }
++
++ /* doubly whiteout-ed */
++ memcpy(name, AUFS_WH_PFX AUFS_WH_PFX, AUFS_WH_PFX_LEN * 2);
++ p = name + AUFS_WH_PFX_LEN * 2;
++ memcpy(p, prefix->name, prefix->len);
++ p += prefix->len;
++ *p++ = '.';
++ AuDebugOn(name + qs.len + 1 - p <= AUFS_WH_TMP_LEN);
++
++ h_ppath.dentry = h_parent;
++ h_ppath.mnt = au_br_mnt(br);
++ h_idmap = au_br_idmap(br);
++ qs.name = name;
++ for (i = 0; i < 3; i++) {
++ sprintf(p, "%.*x", AUFS_WH_TMP_LEN, cnt++);
++ dentry = au_sio_lkup_one(h_idmap, &qs, &h_ppath);
++ if (IS_ERR(dentry) || d_is_negative(dentry))
++ goto out_name;
++ dput(dentry);
+ }
-+ out_put:
-+ au_set_h_dptr(dentry, bdst, NULL);
-+ if (au_dbend(dentry) == bdst)
-+ au_update_dbend(dentry);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ /* pr_warn("could not get random name\n"); */
++ dentry = ERR_PTR(-EEXIST);
++ AuDbg("%.*s\n", AuLNPair(&qs));
++ BUG();
++
++out_name:
++ if (name != defname)
++ au_kfree_try_rcu(name);
++out:
++ AuTraceErrPtr(dentry);
++ return dentry;
+}
+
-+int au_cpdown_dirs(struct dentry *dentry, aufs_bindex_t bdst)
++/*
++ * rename the @h_dentry on @br to the whiteouted temporary name.
++ */
++int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br)
+{
+ int err;
-+ struct au_cpdown_dir_args args = {
-+ .parent = dget_parent(dentry),
-+ .parent_opq = 0
++ struct path h_path = {
++ .mnt = au_br_mnt(br)
+ };
++ struct inode *h_dir, *delegated;
++ struct dentry *h_parent;
+
-+ LKTRTrace("%.*s, b%d\n", AuDLNPair(dentry), bdst);
++ h_parent = h_dentry->d_parent; /* dir inode is locked */
++ h_dir = d_inode(h_parent);
++ IMustLock(h_dir);
++
++ h_path.dentry = au_whtmp_lkup(h_parent, br, &h_dentry->d_name);
++ err = PTR_ERR(h_path.dentry);
++ if (IS_ERR(h_path.dentry))
++ goto out;
+
-+ err = au_cp_dirs(dentry, bdst, au_cpdown_dir, &args);
-+ dput(args.parent);
++ /* under the same dir, no need to lock_rename() */
++ delegated = NULL;
++ 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"
++ " for an internal rename\n");
++ iput(delegated);
++ }
++ dput(h_path.dentry);
+
++out:
+ AuTraceErr(err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
++/*
++ * functions for removing a whiteout
++ */
+
-+/* policies for create */
-+
-+static int au_wbr_bu(struct super_block *sb, aufs_bindex_t bindex)
++static int do_unlink_wh(struct inode *h_dir, struct path *h_path)
+{
-+ for (; bindex >= 0; bindex--)
-+ if (!au_br_rdonly(au_sbr(sb, bindex)))
-+ return bindex;
-+ return -EROFS;
++ int err, force;
++ struct inode *delegated;
++
++ /*
++ * forces superio when the dir has a sticky bit.
++ * this may be a violation of unix fs semantics.
++ */
++ force = (h_dir->i_mode & S_ISVTX)
++ && !uid_eq(current_fsuid(), d_inode(h_path->dentry)->i_uid);
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, h_path, &delegated, force);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ return err;
+}
+
-+/* top down parent */
-+static int au_wbr_create_tdp(struct dentry *dentry, int isdir)
++int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path,
++ struct dentry *dentry)
+{
+ int err;
-+ struct super_block *sb;
-+ aufs_bindex_t bstart, bindex;
-+ unsigned char dirperm1;
-+ struct dentry *parent, *h_parent;
-+ struct inode *h_dir;
+
-+ LKTRTrace("%.*s, dir %d\n", AuDLNPair(dentry), isdir);
++ err = do_unlink_wh(h_dir, h_path);
++ if (!err && dentry)
++ au_set_dbwh(dentry, -1);
+
-+ sb = dentry->d_sb;
-+ dirperm1 = !!au_test_dirperm1(au_mntflags(sb));
-+ bstart = au_dbstart(dentry);
-+ AuDebugOn(bstart < 0);
-+ err = bstart;
-+ /* todo: can 'err' be an illegal? */
-+ if (/* err >= 0 && */ !au_br_rdonly(au_sbr(sb, bstart)))
-+ goto out;
++ return err;
++}
+
-+ err = -EROFS;
-+ parent = dget_parent(dentry);
-+ for (bindex = au_dbstart(parent); bindex < bstart; bindex++) {
-+ h_parent = au_h_dptr(parent, bindex);
-+ if (!h_parent)
-+ continue;
-+ h_dir = h_parent->d_inode;
-+ if (!h_dir)
-+ continue;
++static int unlink_wh_name(struct path *h_ppath, struct qstr *wh)
++{
++ int err;
++ struct path h_path;
+
-+ if (!au_br_rdonly(au_sbr(sb, bindex))
-+ && (!dirperm1
-+ || au_test_h_perm(h_dir, MAY_WRITE | MAY_EXEC,
-+ /*dlgt*/0))) {
-+ err = bindex;
-+ break;
++ err = 0;
++ h_path.dentry = vfsub_lkup_one(wh, h_ppath);
++ if (IS_ERR(h_path.dentry))
++ err = PTR_ERR(h_path.dentry);
++ else {
++ if (d_is_reg(h_path.dentry)) {
++ h_path.mnt = h_ppath->mnt;
++ err = do_unlink_wh(d_inode(h_ppath->dentry), &h_path);
+ }
++ dput(h_path.dentry);
+ }
-+ dput(parent);
-+
-+ /* bottom up here */
-+ if (unlikely(err < 0))
-+ err = au_wbr_bu(sb, bstart - 1);
+
-+ out:
-+ LKTRTrace("b%d\n", err);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
++/*
++ * initialize/clean whiteout for a branch
++ */
+
-+/* an exception for the policy other than tdp */
-+static int au_wbr_create_exp(struct dentry *dentry)
++static void au_wh_clean(struct inode *h_dir, struct path *whpath,
++ const int isdir)
+{
+ int err;
-+ struct dentry *parent;
-+ aufs_bindex_t bwh, bdiropq;
++ struct inode *delegated;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ if (d_is_negative(whpath->dentry))
++ return;
+
-+ err = -1;
-+ bwh = au_dbwh(dentry);
-+ parent = dget_parent(dentry);
-+ bdiropq = au_dbdiropq(parent);
-+ if (bwh >= 0) {
-+ if (bdiropq >= 0)
-+ err = min(bdiropq, bwh);
-+ else
-+ err = bwh;
-+ LKTRTrace("%d\n", err);
-+ } else if (bdiropq >= 0) {
-+ err = bdiropq;
-+ LKTRTrace("%d\n", err);
++ if (isdir)
++ err = vfsub_rmdir(h_dir, whpath);
++ else {
++ delegated = NULL;
++ err = vfsub_unlink(h_dir, whpath, &delegated, /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
+ }
-+ dput(parent);
-+
-+ if (err >= 0 && au_br_rdonly(au_sbr(dentry->d_sb, err)))
-+ err = -1;
-+
-+ LKTRTrace("%d\n", err);
-+ return err;
++ if (unlikely(err))
++ pr_warn("failed removing %pd (%d), ignored.\n",
++ whpath->dentry, err);
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/* round robin */
-+static int au_wbr_create_init_rr(struct super_block *sb)
++static int test_linkable(struct dentry *h_root)
+{
-+ int err;
++ struct inode *h_dir = d_inode(h_root);
+
-+ err = au_wbr_bu(sb, au_sbend(sb));
-+ atomic_set(&au_sbi(sb)->si_wbr_rr_next, -err); /* less important */
++ if (h_dir->i_op->link)
++ return 0;
+
-+ LKTRTrace("b%d\n", err);
-+ return err;
++ pr_err("%pd (%s) doesn't support link(2), use noplink and rw+nolwh\n",
++ h_root, au_sbtype(h_root->d_sb));
++ return -ENOSYS; /* the branch doesn't have its ->link() */
+}
+
-+static int au_wbr_create_rr(struct dentry *dentry, int isdir)
++/* todo: should this mkdir be done in /sbin/mount.aufs helper? */
++static int au_whdir(struct inode *h_dir, struct path *path)
+{
-+ int err, nbr;
-+ struct super_block *sb;
-+ atomic_t *next;
-+ unsigned int u;
-+ aufs_bindex_t bindex, bend;
-+
-+ LKTRTrace("%.*s, dir %d\n", AuDLNPair(dentry), isdir);
++ int err;
+
-+ sb = dentry->d_sb;
-+ next = NULL;
-+ err = au_wbr_create_exp(dentry);
-+ if (err >= 0)
-+ goto out;
++ err = -EEXIST;
++ if (d_is_negative(path->dentry)) {
++ int mode = 0700;
+
-+ next = &au_sbi(sb)->si_wbr_rr_next;
-+ bend = au_sbend(sb);
-+ nbr = bend + 1;
-+ for (bindex = 0; bindex <= bend; bindex++) {
-+ if (!isdir) {
-+ err = atomic_dec_return(next) + 1;
-+ /* modulo for 0 is meaningless */
-+ if (unlikely(!err))
-+ err = atomic_dec_return(next) + 1;
-+ } else
-+ err = atomic_read(next);
-+ LKTRTrace("%d\n", err);
-+ u = err;
-+ err = u % nbr;
-+ LKTRTrace("%d\n", err);
-+ if (!au_br_rdonly(au_sbr(sb, err)))
-+ break;
-+ err = -EROFS;
-+ }
++ if (au_test_nfs(path->dentry->d_sb))
++ mode |= 0111;
++ err = vfsub_mkdir(h_dir, path, mode);
++ } else if (d_is_dir(path->dentry))
++ err = 0;
++ else
++ pr_err("unknown %pd exists\n", path->dentry);
+
-+ out:
-+ LKTRTrace("%d\n", err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++struct au_wh_base {
++ const struct qstr *name;
++ struct dentry *dentry;
++};
+
-+/* most free space */
-+static void *au_wbr_statfs_arg(struct au_branch *br, struct super_block *sb,
-+ aufs_bindex_t bindex)
++static void au_wh_init_ro(struct inode *h_dir, struct au_wh_base base[],
++ struct path *h_path)
+{
-+ struct super_block *h_sb;
-+
-+ h_sb = br->br_mnt->mnt_sb;
-+
-+ if (!au_test_nfs(h_sb))
-+ return h_sb->s_root;
-+
-+ /* sigh,,, why nfs s_root has wrong inode? */
-+ return au_di(sb->s_root)->di_hdentry[0 + bindex].hd_dentry;
++ h_path->dentry = base[AuBrWh_BASE].dentry;
++ au_wh_clean(h_dir, h_path, /*isdir*/0);
++ h_path->dentry = base[AuBrWh_PLINK].dentry;
++ au_wh_clean(h_dir, h_path, /*isdir*/1);
++ h_path->dentry = base[AuBrWh_ORPH].dentry;
++ au_wh_clean(h_dir, h_path, /*isdir*/1);
+}
+
-+static void au_mfs(struct dentry *dentry)
++/*
++ * returns tri-state,
++ * minus: error, caller should print the message
++ * zero: success
++ * plus: error, caller should NOT print the message
++ */
++static int au_wh_init_rw_nolink(struct dentry *h_root, struct au_wbr *wbr,
++ int do_plink, struct au_wh_base base[],
++ struct path *h_path)
+{
-+ struct super_block *sb;
-+ aufs_bindex_t bindex, bend;
-+ unsigned char dlgt;
+ int err;
-+ struct kstatfs st;
-+ unsigned long long b, bavail;
-+ void *arg;
-+ struct au_branch *br;
-+ struct au_wbr_mfs *mfs;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
-+
-+ bavail = 0;
-+ sb = dentry->d_sb;
-+ mfs = &au_sbi(sb)->si_wbr_mfs;
-+ mfs->mfs_bindex = -EROFS;
-+ mfs->mfsrr_bytes = 0;
-+ dlgt = !!au_test_dlgt(au_mntflags(sb));
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++) {
-+ br = au_sbr(sb, bindex);
-+ if (au_br_rdonly(br))
-+ continue;
-+ AuDebugOn(!br->br_wbr);
-+ arg = au_wbr_statfs_arg(br, sb, bindex);
-+ if (!arg)
-+ continue;
++ struct inode *h_dir;
+
-+ err = vfsub_statfs(arg, &st, dlgt);
-+ LKTRTrace("b%d, %d, %llu\n",
-+ bindex, err, (unsigned long long)st.f_bavail);
++ h_dir = d_inode(h_root);
++ h_path->dentry = base[AuBrWh_BASE].dentry;
++ au_wh_clean(h_dir, h_path, /*isdir*/0);
++ h_path->dentry = base[AuBrWh_PLINK].dentry;
++ if (do_plink) {
++ err = test_linkable(h_root);
+ if (unlikely(err)) {
-+ AuWarn1("failed statfs, b%d, %d\n", bindex, err);
-+ continue;
++ err = 1;
++ goto out;
+ }
+
-+ /* when the available size is equal, select lower one */
-+ b = st.f_bavail * st.f_bsize;
-+ br->br_wbr->wbr_bytes = b;
-+ if (b >= bavail) {
-+ bavail = b;
-+ mfs->mfs_bindex = bindex;
-+ mfs->mfs_jiffy = jiffies;
-+ }
-+ }
++ err = au_whdir(h_dir, h_path);
++ if (unlikely(err))
++ goto out;
++ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry);
++ } else
++ au_wh_clean(h_dir, h_path, /*isdir*/1);
++ h_path->dentry = base[AuBrWh_ORPH].dentry;
++ err = au_whdir(h_dir, h_path);
++ if (unlikely(err))
++ goto out;
++ wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry);
+
-+ mfs->mfsrr_bytes = bavail;
-+ LKTRTrace("b%d\n", mfs->mfs_bindex);
++out:
++ return err;
+}
+
-+static int au_wbr_create_mfs(struct dentry *dentry, int isdir)
++/*
++ * for the moment, aufs supports the branch filesystem which does not support
++ * link(2). testing on FAT which does not support i_op->setattr() fully either,
++ * copyup failed. finally, such filesystem will not be used as the writable
++ * branch.
++ *
++ * returns tri-state, see above.
++ */
++static int au_wh_init_rw(struct dentry *h_root, struct au_wbr *wbr,
++ int do_plink, struct au_wh_base base[],
++ struct path *h_path)
+{
+ int err;
-+ struct super_block *sb;
-+ struct au_wbr_mfs *mfs;
++ struct inode *h_dir;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ WbrWhMustWriteLock(wbr);
+
-+ sb = dentry->d_sb;
-+ err = au_wbr_create_exp(dentry);
-+ if (err >= 0)
++ err = test_linkable(h_root);
++ if (unlikely(err)) {
++ err = 1;
+ goto out;
++ }
+
-+ mfs = &au_sbi(sb)->si_wbr_mfs;
-+ mutex_lock(&mfs->mfs_lock);
-+ if (unlikely(time_after(jiffies, mfs->mfs_jiffy + mfs->mfs_expire)
-+ || mfs->mfs_bindex < 0
-+ || au_br_rdonly(au_sbr(sb, mfs->mfs_bindex))))
-+ au_mfs(dentry);
-+ mutex_unlock(&mfs->mfs_lock);
-+ err = mfs->mfs_bindex;
-+
-+ out:
-+ LKTRTrace("b%d\n", err);
-+ return err;
-+}
-+
-+static int au_wbr_create_init_mfs(struct super_block *sb)
-+{
-+ struct au_wbr_mfs *mfs;
++ /*
++ * todo: should this create be done in /sbin/mount.aufs helper?
++ */
++ err = -EEXIST;
++ h_dir = d_inode(h_root);
++ if (d_is_negative(base[AuBrWh_BASE].dentry)) {
++ h_path->dentry = base[AuBrWh_BASE].dentry;
++ err = vfsub_create(h_dir, h_path, WH_MASK, /*want_excl*/true);
++ } else if (d_is_reg(base[AuBrWh_BASE].dentry))
++ err = 0;
++ else
++ pr_err("unknown %pd2 exists\n", base[AuBrWh_BASE].dentry);
++ if (unlikely(err))
++ goto out;
+
-+ mfs = &au_sbi(sb)->si_wbr_mfs;
-+ LKTRTrace("expire %lu\n", mfs->mfs_expire);
++ h_path->dentry = base[AuBrWh_PLINK].dentry;
++ if (do_plink) {
++ err = au_whdir(h_dir, h_path);
++ if (unlikely(err))
++ goto out;
++ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry);
++ } else
++ au_wh_clean(h_dir, h_path, /*isdir*/1);
++ wbr->wbr_whbase = dget(base[AuBrWh_BASE].dentry);
+
-+ mutex_init(&mfs->mfs_lock);
-+ mfs->mfs_jiffy = 0;
-+ mfs->mfs_bindex = -EROFS;
++ h_path->dentry = base[AuBrWh_ORPH].dentry;
++ err = au_whdir(h_dir, h_path);
++ if (unlikely(err))
++ goto out;
++ wbr->wbr_orph = dget(base[AuBrWh_ORPH].dentry);
+
-+ return 0;
++out:
++ return err;
+}
+
-+static int au_wbr_create_fin_mfs(struct super_block *sb)
++/*
++ * initialize the whiteout base file/dir for @br.
++ */
++int au_wh_init(struct au_branch *br, struct super_block *sb)
+{
-+ AuTraceEnter();
-+ mutex_destroy(&au_sbi(sb)->si_wbr_mfs.mfs_lock);
-+ return 0;
-+}
-+
-+/* ---------------------------------------------------------------------- */
++ int err, i;
++ const unsigned char do_plink
++ = !!au_opt_test(au_mntflags(sb), PLINK);
++ struct inode *h_dir;
++ struct path path = br->br_path;
++ struct dentry *h_root = path.dentry;
++ struct au_wbr *wbr = br->br_wbr;
++ static const struct qstr base_name[] = {
++ [AuBrWh_BASE] = QSTR_INIT(AUFS_BASE_NAME,
++ sizeof(AUFS_BASE_NAME) - 1),
++ [AuBrWh_PLINK] = QSTR_INIT(AUFS_PLINKDIR_NAME,
++ sizeof(AUFS_PLINKDIR_NAME) - 1),
++ [AuBrWh_ORPH] = QSTR_INIT(AUFS_ORPHDIR_NAME,
++ sizeof(AUFS_ORPHDIR_NAME) - 1)
++ };
++ struct au_wh_base base[] = {
++ [AuBrWh_BASE] = {
++ .name = base_name + AuBrWh_BASE,
++ .dentry = NULL
++ },
++ [AuBrWh_PLINK] = {
++ .name = base_name + AuBrWh_PLINK,
++ .dentry = NULL
++ },
++ [AuBrWh_ORPH] = {
++ .name = base_name + AuBrWh_ORPH,
++ .dentry = NULL
++ }
++ };
+
-+/* most free space and then round robin */
-+static int au_wbr_create_mfsrr(struct dentry *dentry, int isdir)
-+{
-+ int err;
-+ struct au_wbr_mfs *mfs;
++ if (wbr)
++ WbrWhMustWriteLock(wbr);
+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), isdir);
++ for (i = 0; i < AuBrWh_Last; i++) {
++ /* doubly whiteouted */
++ struct dentry *d;
+
-+ err = au_wbr_create_mfs(dentry, isdir);
-+ if (err >= 0) {
-+ mfs = &au_sbi(dentry->d_sb)->si_wbr_mfs;
-+ LKTRTrace("%llu bytes, %llu wmark\n",
-+ mfs->mfsrr_bytes, mfs->mfsrr_watermark);
-+ if (mfs->mfsrr_bytes < mfs->mfsrr_watermark)
-+ err = au_wbr_create_rr(dentry, isdir);
-+ }
++ d = au_wh_lkup(h_root, (void *)base[i].name, br);
++ err = PTR_ERR(d);
++ if (IS_ERR(d))
++ goto out;
+
-+ LKTRTrace("b%d\n", err);
-+ return err;
-+}
++ base[i].dentry = d;
++ AuDebugOn(wbr
++ && wbr->wbr_wh[i]
++ && wbr->wbr_wh[i] != base[i].dentry);
++ }
+
-+static int au_wbr_create_init_mfsrr(struct super_block *sb)
-+{
-+ int err;
++ if (wbr)
++ for (i = 0; i < AuBrWh_Last; i++) {
++ dput(wbr->wbr_wh[i]);
++ wbr->wbr_wh[i] = NULL;
++ }
+
-+ au_wbr_create_init_mfs(sb); /* ignore */
-+ err = au_wbr_create_init_rr(sb);
++ err = 0;
++ if (!au_br_writable(br->br_perm)) {
++ h_dir = d_inode(h_root);
++ au_wh_init_ro(h_dir, base, &path);
++ } else if (!au_br_wh_linkable(br->br_perm)) {
++ err = au_wh_init_rw_nolink(h_root, wbr, do_plink, base, &path);
++ if (err > 0)
++ goto out;
++ else if (err)
++ goto out_err;
++ } else {
++ err = au_wh_init_rw(h_root, wbr, do_plink, base, &path);
++ if (err > 0)
++ goto out;
++ else if (err)
++ goto out_err;
++ }
++ goto out; /* success */
+
-+ AuTraceErr(err);
++out_err:
++ pr_err("an error(%d) on the writable branch %pd(%s)\n",
++ err, h_root, au_sbtype(h_root->d_sb));
++out:
++ for (i = 0; i < AuBrWh_Last; i++)
++ dput(base[i].dentry);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
++/*
++ * whiteouts are all hard-linked usually.
++ * when its link count reaches a ceiling, we create a new whiteout base
++ * asynchronously.
++ */
+
-+/* top down parent and most free space */
-+static int au_wbr_create_pmfs(struct dentry *dentry, int isdir)
-+{
-+ int err, e2;
++struct reinit_br_wh {
+ struct super_block *sb;
-+ struct dentry *parent, *h_parent;
-+ aufs_bindex_t bindex, bstart, bend;
-+ unsigned char dirperm1;
+ struct au_branch *br;
-+ unsigned long long b;
-+ struct inode *h_dir;
++};
+
-+ LKTRTrace("%.*s, %d\n", AuDLNPair(dentry), isdir);
++static void reinit_br_wh(void *arg)
++{
++ int err;
++ aufs_bindex_t bindex;
++ struct path h_path;
++ struct reinit_br_wh *a = arg;
++ struct au_wbr *wbr;
++ struct inode *dir, *delegated;
++ struct dentry *h_root;
++ struct au_hinode *hdir;
+
-+ err = au_wbr_create_tdp(dentry, isdir);
-+ if (unlikely(err < 0))
++ err = 0;
++ wbr = a->br->br_wbr;
++ /* big aufs lock */
++ si_noflush_write_lock(a->sb);
++ if (!au_br_writable(a->br->br_perm))
++ goto out;
++ bindex = au_br_index(a->sb, a->br->br_id);
++ if (unlikely(bindex < 0))
+ goto out;
-+ parent = dget_parent(dentry);
-+ bstart = au_dbstart(parent);
-+ bend = au_dbtaildir(parent);
-+ if (bstart == bend)
-+ goto out_parent; /* success */
-+
-+ e2 = au_wbr_create_mfs(dentry, isdir);
-+ if (e2 < 0)
-+ goto out_parent; /* success */
-+
-+ /* when the available size is equal, select upper one */
-+ sb = dentry->d_sb;
-+ br = au_sbr(sb, err);
-+ AuDebugOn(!br->br_wbr);
-+ dirperm1 = !!au_test_dirperm1(au_mntflags(sb));
-+ b = br->br_wbr->wbr_bytes;
-+ LKTRTrace("b%d, %llu\n", err, b);
-+
-+ if (dirperm1) {
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ h_parent = au_h_dptr(parent, bindex);
-+ if (!h_parent)
-+ continue;
-+ h_dir = h_parent->d_inode;
-+ if (!h_dir)
-+ continue;
+
-+ br = au_sbr(sb, bindex);
-+ if (!au_br_rdonly(br)
-+ && au_test_h_perm(h_dir, MAY_WRITE | MAY_EXEC,
-+ /*dlgt*/0)
-+ && br->br_wbr->wbr_bytes > b) {
-+ b = br->br_wbr->wbr_bytes;
-+ err = bindex;
-+ LKTRTrace("b%d, %llu\n", err, b);
-+ }
-+ }
-+ if (err >= 0)
-+ goto out_parent;
-+ }
-+ for (bindex = bstart; bindex <= bend; bindex++) {
-+ h_parent = au_h_dptr(parent, bindex);
-+ if (!h_parent || !h_parent->d_inode)
-+ continue;
++ di_read_lock_parent(a->sb->s_root, AuLock_IR);
++ dir = d_inode(a->sb->s_root);
++ hdir = au_hi(dir, bindex);
++ h_root = au_h_dptr(a->sb->s_root, bindex);
++ AuDebugOn(h_root != au_br_dentry(a->br));
+
-+ br = au_sbr(sb, bindex);
-+ if (!au_br_rdonly(br) && br->br_wbr->wbr_bytes > b) {
-+ b = br->br_wbr->wbr_bytes;
-+ err = bindex;
-+ LKTRTrace("b%d, %llu\n", err, b);
++ au_hn_inode_lock_nested(hdir, AuLsc_I_PARENT);
++ wbr_wh_write_lock(wbr);
++ err = au_h_verify(wbr->wbr_whbase, au_opt_udba(a->sb), hdir->hi_inode,
++ h_root, a->br);
++ if (!err) {
++ h_path.dentry = wbr->wbr_whbase;
++ h_path.mnt = au_br_mnt(a->br);
++ delegated = NULL;
++ err = vfsub_unlink(hdir->hi_inode, &h_path, &delegated,
++ /*force*/0);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
+ }
++ } else {
++ pr_warn("%pd is moved, ignored\n", wbr->wbr_whbase);
++ err = 0;
+ }
++ dput(wbr->wbr_whbase);
++ wbr->wbr_whbase = NULL;
++ if (!err)
++ err = au_wh_init(a->br, a->sb);
++ wbr_wh_write_unlock(wbr);
++ au_hn_inode_unlock(hdir);
++ di_read_unlock(a->sb->s_root, AuLock_IR);
++ if (!err)
++ au_fhsm_wrote(a->sb, bindex, /*force*/0);
+
-+ out_parent:
-+ dput(parent);
-+ out:
-+ LKTRTrace("b%d\n", err);
-+ return err;
-+}
-+
-+/* ---------------------------------------------------------------------- */
-+
-+/* policies for copyup */
-+
-+/* top down parent */
-+static int au_wbr_copyup_tdp(struct dentry *dentry)
-+{
-+ return au_wbr_create_tdp(dentry, /*isdir, anything is ok*/0);
++out:
++ if (wbr)
++ atomic_dec(&wbr->wbr_wh_running);
++ au_lcnt_dec(&a->br->br_count);
++ si_write_unlock(a->sb);
++ au_nwt_done(&au_sbi(a->sb)->si_nowait);
++ au_kfree_rcu(a);
++ if (unlikely(err))
++ AuIOErr("err %d\n", err);
+}
+
-+/* bottom up parent */
-+static int au_wbr_copyup_bup(struct dentry *dentry)
++static void kick_reinit_br_wh(struct super_block *sb, struct au_branch *br)
+{
-+ int err;
-+ struct dentry *parent, *h_parent;
-+ aufs_bindex_t bindex, bstart;
-+ unsigned char dirperm1;
-+ struct super_block *sb;
-+ struct inode *h_dir;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ int do_dec, wkq_err;
++ struct reinit_br_wh *arg;
+
-+ err = -EROFS;
-+ sb = dentry->d_sb;
-+ dirperm1 = !!au_test_dirperm1(au_mntflags(sb));
-+ parent = dget_parent(dentry);
-+ bstart = au_dbstart(parent);
-+ for (bindex = au_dbstart(dentry); bindex >= bstart; bindex--) {
-+ h_parent = au_h_dptr(parent, bindex);
-+ if (!h_parent)
-+ continue;
-+ h_dir = h_parent->d_inode;
-+ if (!h_dir)
-+ continue;
++ do_dec = 1;
++ if (atomic_inc_return(&br->br_wbr->wbr_wh_running) != 1)
++ goto out;
+
-+ if (!au_br_rdonly(au_sbr(sb, bindex))
-+ && (!dirperm1
-+ || au_test_h_perm(h_dir, MAY_WRITE | MAY_EXEC,
-+ /*dlgt*/0))) {
-+ err = bindex;
-+ break;
++ /* ignore ENOMEM */
++ arg = kmalloc(sizeof(*arg), GFP_NOFS);
++ if (arg) {
++ /*
++ * dec(wh_running), kfree(arg) and dec(br_count)
++ * in reinit function
++ */
++ arg->sb = sb;
++ arg->br = br;
++ au_lcnt_inc(&br->br_count);
++ wkq_err = au_wkq_nowait(reinit_br_wh, arg, sb, /*flags*/0);
++ if (unlikely(wkq_err)) {
++ atomic_dec(&br->br_wbr->wbr_wh_running);
++ au_lcnt_dec(&br->br_count);
++ au_kfree_rcu(arg);
+ }
++ do_dec = 0;
+ }
-+ dput(parent);
-+
-+ /* bottom up here */
-+ if (unlikely(err < 0))
-+ err = au_wbr_bu(sb, bstart - 1);
+
-+ LKTRTrace("b%d\n", err);
-+ return err;
++out:
++ if (do_dec)
++ atomic_dec(&br->br_wbr->wbr_wh_running);
+}
+
-+/* bottom up */
-+static int au_wbr_copyup_bu(struct dentry *dentry)
++/* ---------------------------------------------------------------------- */
++
++/*
++ * create the whiteout @wh.
++ */
++static int link_or_create_wh(struct super_block *sb, aufs_bindex_t bindex,
++ struct dentry *wh)
+{
+ int err;
++ struct path h_path = {
++ .dentry = wh
++ };
++ struct au_branch *br;
++ struct au_wbr *wbr;
++ struct dentry *h_parent;
++ struct inode *h_dir, *delegated;
++
++ h_parent = wh->d_parent; /* dir inode is locked */
++ h_dir = d_inode(h_parent);
++ IMustLock(h_dir);
++
++ br = au_sbr(sb, bindex);
++ h_path.mnt = au_br_mnt(br);
++ wbr = br->br_wbr;
++ wbr_wh_read_lock(wbr);
++ if (wbr->wbr_whbase) {
++ delegated = NULL;
++ err = vfsub_link(wbr->wbr_whbase, h_dir, &h_path, &delegated);
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal link\n");
++ iput(delegated);
++ }
++ if (!err || err != -EMLINK)
++ goto out;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(dentry));
++ /* link count full. re-initialize br_whbase. */
++ kick_reinit_br_wh(sb, br);
++ }
+
-+ err = au_wbr_bu(dentry->d_sb, au_dbstart(dentry));
++ /* return this error in this context */
++ err = vfsub_create(h_dir, &h_path, WH_MASK, /*want_excl*/true);
++ if (!err)
++ au_fhsm_wrote(sb, bindex, /*force*/0);
+
-+ LKTRTrace("b%d\n", err);
++out:
++ wbr_wh_read_unlock(wbr);
+ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+struct au_wbr_copyup_operations au_wbr_copyup_ops[] = {
-+ [AuWbrCopyup_TDP] = {
-+ .copyup = au_wbr_copyup_tdp
-+ },
-+ [AuWbrCopyup_BUP] = {
-+ .copyup = au_wbr_copyup_bup
-+ },
-+ [AuWbrCopyup_BU] = {
-+ .copyup = au_wbr_copyup_bu
-+ }
-+};
-+
-+struct au_wbr_create_operations au_wbr_create_ops[] = {
-+ [AuWbrCreate_TDP] = {
-+ .create = au_wbr_create_tdp
-+ },
-+ [AuWbrCreate_RR] = {
-+ .create = au_wbr_create_rr,
-+ .init = au_wbr_create_init_rr
-+ },
-+ [AuWbrCreate_MFS] = {
-+ .create = au_wbr_create_mfs,
-+ .init = au_wbr_create_init_mfs,
-+ .fin = au_wbr_create_fin_mfs
-+ },
-+ [AuWbrCreate_MFSV] = {
-+ .create = au_wbr_create_mfs,
-+ .init = au_wbr_create_init_mfs,
-+ .fin = au_wbr_create_fin_mfs
-+ },
-+ [AuWbrCreate_MFSRR] = {
-+ .create = au_wbr_create_mfsrr,
-+ .init = au_wbr_create_init_mfsrr,
-+ .fin = au_wbr_create_fin_mfs
-+ },
-+ [AuWbrCreate_MFSRRV] = {
-+ .create = au_wbr_create_mfsrr,
-+ .init = au_wbr_create_init_mfsrr,
-+ .fin = au_wbr_create_fin_mfs
-+ },
-+ [AuWbrCreate_PMFS] = {
-+ .create = au_wbr_create_pmfs,
-+ .init = au_wbr_create_init_mfs,
-+ .fin = au_wbr_create_fin_mfs
-+ },
-+ [AuWbrCreate_PMFSV] = {
-+ .create = au_wbr_create_pmfs,
-+ .init = au_wbr_create_init_mfs,
-+ .fin = au_wbr_create_fin_mfs
-+ }
-+};
-diff -urN linux/fs/aufs/whout.c linux-aufs/fs/aufs/whout.c
---- linux/fs/aufs/whout.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/whout.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1160 @@
+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * create or remove the diropq.
+ */
++static struct dentry *do_diropq(struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int flags)
++{
++ struct dentry *opq_dentry;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct path h_path;
++ int err;
+
-+/*
-+ * whiteout for logical deletion and opaque directory
-+ *
-+ * $Id$
-+ */
++ sb = dentry->d_sb;
++ br = au_sbr(sb, bindex);
++ h_path.dentry = au_h_dptr(dentry, bindex);
++ h_path.mnt = au_br_mnt(br);
++ opq_dentry = vfsub_lkup_one(&diropq_name, &h_path);
++ if (IS_ERR(opq_dentry))
++ goto out;
+
-+#include <linux/fs.h>
-+#include <linux/namei.h>
-+#include "aufs.h"
++ if (au_ftest_diropq(flags, CREATE)) {
++ err = link_or_create_wh(sb, bindex, opq_dentry);
++ if (!err) {
++ au_set_dbdiropq(dentry, bindex);
++ goto out; /* success */
++ }
++ } else {
++ h_path.dentry = opq_dentry;
++ err = do_unlink_wh(au_h_iptr(d_inode(dentry), bindex), &h_path);
++ if (!err)
++ au_set_dbdiropq(dentry, -1);
++ }
++ dput(opq_dentry);
++ opq_dentry = ERR_PTR(err);
+
-+#define WH_MASK S_IRUGO
++out:
++ return opq_dentry;
++}
+
-+/* If a directory contains this file, then it is opaque. We start with the
-+ * .wh. flag so that it is blocked by lookup.
-+ */
-+static struct qstr diropq_name = {
-+ .name = AUFS_WH_DIROPQ,
-+ .len = sizeof(AUFS_WH_DIROPQ) - 1
++struct do_diropq_args {
++ struct dentry **errp;
++ struct dentry *dentry;
++ aufs_bindex_t bindex;
++ unsigned int flags;
+};
+
-+/*
-+ * generate whiteout name, which is NOT terminated by NULL.
-+ * @name: original d_name.name
-+ * @len: original d_name.len
-+ * @wh: whiteout qstr
-+ * returns zero when succeeds, otherwise error.
-+ * succeeded value as wh->name should be freed by au_wh_name_free().
-+ */
-+int au_wh_name_alloc(const char *name, int len, struct qstr *wh)
++static void call_do_diropq(void *args)
+{
-+ char *p;
++ struct do_diropq_args *a = args;
++ *a->errp = do_diropq(a->dentry, a->bindex, a->flags);
++}
+
-+ AuDebugOn(!name || !len || !wh);
++struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int flags)
++{
++ struct dentry *diropq, *h_dentry;
++ struct mnt_idmap *h_idmap;
+
-+ if (unlikely(len > PATH_MAX - AUFS_WH_PFX_LEN))
-+ return -ENAMETOOLONG;
++ h_idmap = au_sbr_idmap(dentry->d_sb, bindex);
++ h_dentry = au_h_dptr(dentry, bindex);
++ if (!au_test_h_perm_sio(h_idmap, d_inode(h_dentry),
++ MAY_EXEC | MAY_WRITE))
++ diropq = do_diropq(dentry, bindex, flags);
++ else {
++ int wkq_err;
++ struct do_diropq_args args = {
++ .errp = &diropq,
++ .dentry = dentry,
++ .bindex = bindex,
++ .flags = flags
++ };
+
-+ wh->len = len + AUFS_WH_PFX_LEN;
-+ p = kmalloc(wh->len, GFP_NOFS);
-+ wh->name = p;
-+ if (p) {
-+ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
-+ memcpy(p + AUFS_WH_PFX_LEN, name, len);
-+ /* smp_mb(); */
-+ return 0;
++ wkq_err = au_wkq_wait(call_do_diropq, &args);
++ if (unlikely(wkq_err))
++ diropq = ERR_PTR(wkq_err);
+ }
-+ return -ENOMEM;
-+}
+
-+void au_wh_name_free(struct qstr *wh)
-+{
-+ AuDebugOn(!wh || !wh->name);
-+ kfree(wh->name);
++ return diropq;
+}
+
+/* ---------------------------------------------------------------------- */
+
+/*
-+ * test if the @wh_name exists under @h_parent.
-+ * @try_sio specifies the necessary of super-io.
++ * lookup whiteout dentry.
++ * @h_parent: lower parent dentry which must exist and be locked
++ * @base_name: name of dentry which will be whiteouted
++ * returns dentry for whiteout.
+ */
-+int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, int try_sio,
-+ struct au_ndx *ndx)
++struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name,
++ struct au_branch *br)
+{
+ int err;
++ struct qstr wh_name;
+ struct dentry *wh_dentry;
-+ struct inode *h_dir;
-+ unsigned int flags;
-+
-+ LKTRTrace("%.*s/%.*s, ndx{%p, 0x%x}\n", AuDLNPair(h_parent),
-+ wh_name->len, wh_name->name, ndx->nfsmnt, ndx->flags);
-+ h_dir = h_parent->d_inode;
-+ AuDebugOn(!S_ISDIR(h_dir->i_mode));
++ struct path h_path;
+
-+ flags = 0;
-+ if (ndx && ndx->nd) {
-+ flags = ndx->nd->flags;
-+ ndx->nd->flags &= ~(LOOKUP_OPEN | LOOKUP_CREATE);
++ err = au_wh_name_alloc(&wh_name, base_name);
++ wh_dentry = ERR_PTR(err);
++ if (!err) {
++ h_path.dentry = h_parent;
++ h_path.mnt = au_br_mnt(br);
++ wh_dentry = vfsub_lkup_one(&wh_name, &h_path);
++ au_kfree_try_rcu(wh_name.name);
+ }
-+
-+ if (!try_sio)
-+ wh_dentry = au_lkup_one(wh_name->name, h_parent,
-+ wh_name->len, ndx);
-+ else
-+ wh_dentry = au_sio_lkup_one(wh_name->name, h_parent,
-+ wh_name->len, ndx);
-+ if (flags)
-+ ndx->nd->flags = flags;
-+ err = PTR_ERR(wh_dentry);
-+ if (IS_ERR(wh_dentry))
-+ goto out;
-+
-+ err = 0;
-+ if (!wh_dentry->d_inode)
-+ goto out_wh; /* success */
-+
-+ err = 1;
-+ if (S_ISREG(wh_dentry->d_inode->i_mode))
-+ goto out_wh; /* success */
-+
-+ err = -EIO;
-+ AuIOErr("%.*s Invalid whiteout entry type 0%o.\n",
-+ AuDLNPair(wh_dentry), wh_dentry->d_inode->i_mode);
-+
-+ out_wh:
-+ dput(wh_dentry);
-+ out:
-+ AuTraceErr(err);
-+ return err;
++ return wh_dentry;
+}
+
+/*
-+ * test if the @h_dentry sets opaque or not.
++ * link/create a whiteout for @dentry on @bindex.
+ */
-+int au_diropq_test(struct dentry *h_dentry, struct au_ndx *ndx)
++struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent)
+{
-+ int err, try_sio;
-+ struct inode *h_dir;
++ struct dentry *wh_dentry;
++ struct super_block *sb;
++ int err;
+
-+ LKTRTrace("dentry %.*s\n", AuDLNPair(h_dentry));
-+ h_dir = h_dentry->d_inode;
-+ AuDebugOn(!S_ISDIR(h_dir->i_mode));
++ sb = dentry->d_sb;
++ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, au_sbr(sb, bindex));
++ if (!IS_ERR(wh_dentry) && d_is_negative(wh_dentry)) {
++ err = link_or_create_wh(sb, bindex, wh_dentry);
++ if (!err) {
++ au_set_dbwh(dentry, bindex);
++ au_fhsm_wrote(sb, bindex, /*force*/0);
++ } else {
++ dput(wh_dentry);
++ wh_dentry = ERR_PTR(err);
++ }
++ }
+
-+ try_sio = au_test_h_perm_sio(h_dir, MAY_EXEC,
-+ au_ftest_ndx(ndx->flags, DLGT));
-+ err = au_wh_test(h_dentry, &diropq_name, try_sio, ndx);
-+ AuTraceErr(err);
-+ return err;
++ return wh_dentry;
+}
+
-+/*
-+ * returns a negative dentry whose name is unique and temporary.
-+ */
-+struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct qstr *prefix,
-+ struct au_ndx *ndx)
++/* ---------------------------------------------------------------------- */
++
++/* Delete all whiteouts in this directory on branch bindex. */
++static int del_wh_children(struct path *h_path, struct au_nhash *whlist,
++ aufs_bindex_t bindex)
+{
-+#define HEX_LEN 4
-+ struct dentry *dentry;
-+ int len, i;
-+ char defname[AUFS_WH_PFX_LEN * 2 + DNAME_INLINE_LEN_MIN + 1
-+ + HEX_LEN + 1], *name, *p;
-+ static unsigned char cnt;
++ int err;
++ unsigned long ul, n;
++ struct qstr wh_name;
++ char *p;
++ struct hlist_head *head;
++ struct au_vdir_wh *pos;
++ struct au_vdir_destr *str;
++
++ err = -ENOMEM;
++ p = (void *)__get_free_page(GFP_NOFS);
++ wh_name.name = p;
++ if (unlikely(!wh_name.name))
++ goto out;
+
-+ LKTRTrace("hp %.*s, prefix %.*s\n",
-+ AuDLNPair(h_parent), prefix->len, prefix->name);
-+ AuDebugOn(!h_parent->d_inode);
++ err = 0;
++ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
++ p += AUFS_WH_PFX_LEN;
++ n = whlist->nh_num;
++ head = whlist->nh_head;
++ for (ul = 0; !err && ul < n; ul++, head++) {
++ hlist_for_each_entry(pos, head, wh_hash) {
++ if (pos->wh_bindex != bindex)
++ continue;
+
-+ name = defname;
-+ len = sizeof(defname) - DNAME_INLINE_LEN_MIN + prefix->len - 1;
-+ if (unlikely(prefix->len > DNAME_INLINE_LEN_MIN)) {
-+ dentry = ERR_PTR(-ENAMETOOLONG);
-+ if (unlikely(len >= PATH_MAX))
-+ goto out;
-+ dentry = ERR_PTR(-ENOMEM);
-+ name = kmalloc(len + 1, GFP_NOFS);
-+ if (unlikely(!name))
-+ goto out;
++ str = &pos->wh_str;
++ if (str->len + AUFS_WH_PFX_LEN <= PATH_MAX) {
++ memcpy(p, str->name, str->len);
++ wh_name.len = AUFS_WH_PFX_LEN + str->len;
++ err = unlink_wh_name(h_path, &wh_name);
++ if (!err)
++ continue;
++ break;
++ }
++ AuIOErr("whiteout name too long %.*s\n",
++ str->len, str->name);
++ err = -EIO;
++ break;
++ }
+ }
++ free_page((unsigned long)wh_name.name);
+
-+ /* doubly whiteout-ed */
-+ memcpy(name, AUFS_WH_PFX AUFS_WH_PFX, AUFS_WH_PFX_LEN * 2);
-+ p = name + AUFS_WH_PFX_LEN * 2;
-+ memcpy(p, prefix->name, prefix->len);
-+ p += prefix->len;
-+ *p++ = '.';
-+ AuDebugOn(name + len + 1 - p <= HEX_LEN);
++out:
++ return err;
++}
+
-+ for (i = 0; i < 3; i++) {
-+ sprintf(p, "%.*d", HEX_LEN, cnt++);
-+ dentry = au_sio_lkup_one(name, h_parent, len, ndx);
-+ if (IS_ERR(dentry) || !dentry->d_inode)
-+ goto out_name;
-+ dput(dentry);
-+ }
-+ /* AuWarn("could not get random name\n"); */
-+ dentry = ERR_PTR(-EEXIST);
-+ AuDbg("%.*s\n", len, name);
-+ BUG();
++struct del_wh_children_args {
++ int *errp;
++ struct path *h_path;
++ struct au_nhash *whlist;
++ aufs_bindex_t bindex;
++};
+
-+ out_name:
-+ if (name != defname)
-+ kfree(name);
-+ out:
-+ AuTraceErrPtr(dentry);
-+ return dentry;
-+#undef HEX_LEN
++static void call_del_wh_children(void *args)
++{
++ struct del_wh_children_args *a = args;
++ *a->errp = del_wh_children(a->h_path, a->whlist, a->bindex);
+}
+
-+/*
-+ * rename the @dentry of @bindex to the whiteouted temporary name.
-+ */
-+int au_whtmp_ren(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *h_dentry)
++/* ---------------------------------------------------------------------- */
++
++struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp)
+{
-+ int err, dlgt;
-+ struct inode *h_dir;
-+ struct dentry *h_parent, *tmp_dentry;
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
++ struct au_whtmp_rmdir *whtmp;
++ int err;
++ unsigned int rdhash;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
-+ AuDebugOn(!h_dentry->d_inode);
-+ h_parent = h_dentry->d_parent; /* dir inode is locked */
-+ h_dir = h_parent->d_inode;
-+ IMustLock(h_dir);
++ SiMustAnyLock(sb);
+
-+ sb = dir->i_sb;
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ if (dlgt)
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nfsmnt = au_nfsmnt(sb, bindex);
-+ tmp_dentry = au_whtmp_lkup(h_parent, &h_dentry->d_name, &ndx);
-+ err = PTR_ERR(tmp_dentry);
-+ if (IS_ERR(tmp_dentry))
++ whtmp = kzalloc(sizeof(*whtmp), gfp);
++ if (unlikely(!whtmp)) {
++ whtmp = ERR_PTR(-ENOMEM);
+ goto out;
++ }
+
-+ /* under the same dir, no need to lock_rename() */
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
-+ AuDebugOn(!S_ISDIR(h_dentry->d_inode->i_mode));
-+ vfsub_ign_hinode(&vargs, IN_MOVED_FROM | IN_MOVED_TO,
-+ au_hi(dir, bindex));
-+ err = vfsub_rename(h_dir, h_dentry, h_dir, tmp_dentry, &vargs);
-+ AuTraceErr(err);
-+ dput(tmp_dentry);
++ /* no estimation for dir size */
++ rdhash = au_sbi(sb)->si_rdhash;
++ if (!rdhash)
++ rdhash = AUFS_RDHASH_DEF;
++ err = au_nhash_alloc(&whtmp->whlist, rdhash, gfp);
++ if (unlikely(err)) {
++ au_kfree_rcu(whtmp);
++ whtmp = ERR_PTR(err);
++ }
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
++out:
++ return whtmp;
+}
+
-+/* ---------------------------------------------------------------------- */
++void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp)
++{
++ if (whtmp->br)
++ au_lcnt_dec(&whtmp->br->br_count);
++ dput(whtmp->wh_dentry);
++ iput(whtmp->dir);
++ au_nhash_wh_free(&whtmp->whlist);
++ au_kfree_rcu(whtmp);
++}
+
-+static int do_unlink_wh(struct au_hinode *hdir, struct inode *h_dir,
-+ struct dentry *wh_dentry, const int dlgt)
++/*
++ * rmdir the whiteouted temporary named dir @h_dentry.
++ * @whlist: whiteouted children.
++ */
++int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex,
++ struct dentry *wh_dentry, struct au_nhash *whlist)
+{
+ int err;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
++ unsigned int h_nlink;
++ struct path wh_path;
++ struct inode *wh_inode, *h_dir;
++ struct au_branch *br;
++ struct mnt_idmap *h_idmap;
++
++ h_dir = d_inode(wh_dentry->d_parent); /* dir inode is locked */
++ IMustLock(h_dir);
+
-+ AuDebugOn(hdir && h_dir);
-+ AuDebugOn(!hdir && !h_dir);
-+ if (!h_dir)
-+ h_dir = hdir->hi_inode;
-+ LKTRTrace("hi%lu, wh %.*s\n", h_dir->i_ino, AuDLNPair(wh_dentry));
-+ AuDebugOn(!wh_dentry->d_inode || !S_ISREG(wh_dentry->d_inode->i_mode));
++ br = au_sbr(dir->i_sb, bindex);
++ wh_path.dentry = wh_dentry;
++ wh_path.mnt = au_br_mnt(br);
++ h_idmap = au_br_idmap(br);
++ wh_inode = d_inode(wh_dentry);
++ inode_lock_nested(wh_inode, AuLsc_I_CHILD);
+
+ /*
-+ * forces superio when the dir has a sticky bit.
-+ * this may be a violation of unix fs semantics.
++ * someone else might change some whiteouts while we were sleeping.
++ * it means this whlist may have an obsoleted entry.
+ */
-+ vfsub_args_init(&vargs, &ign, dlgt,
-+ (h_dir->i_mode & S_ISVTX)
-+ && wh_dentry->d_inode->i_uid != current->fsuid);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, hdir);
-+ err = vfsub_unlink(h_dir, wh_dentry, &vargs);
-+ AuTraceErr(err);
++ if (!au_test_h_perm_sio(h_idmap, wh_inode, MAY_EXEC | MAY_WRITE))
++ err = del_wh_children(&wh_path, whlist, bindex);
++ else {
++ int wkq_err;
++ struct del_wh_children_args args = {
++ .errp = &err,
++ .h_path = &wh_path,
++ .whlist = whlist,
++ .bindex = bindex
++ };
++
++ wkq_err = au_wkq_wait(call_del_wh_children, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
++ }
++ inode_unlock(wh_inode);
++
++ if (!err) {
++ h_nlink = h_dir->i_nlink;
++ err = vfsub_rmdir(h_dir, &wh_path);
++ /* some fs doesn't change the parent nlink in some cases */
++ h_nlink -= h_dir->i_nlink;
++ }
++
++ if (!err) {
++ if (au_ibtop(dir) == bindex) {
++ /* todo: dir->i_mutex is necessary */
++ au_cpup_attr_timesizes(dir);
++ if (h_nlink)
++ vfsub_drop_nlink(dir);
++ }
++ return 0; /* success */
++ }
++
++ pr_warn("failed removing %pd(%d), ignored\n", wh_dentry, err);
+ return err;
+}
+
-+int au_wh_unlink_dentry(struct au_hinode *hdir, struct dentry *wh_dentry,
-+ struct dentry *dentry, int dlgt)
++static void call_rmdir_whtmp(void *args)
+{
+ int err;
++ aufs_bindex_t bindex;
++ struct au_whtmp_rmdir *a = args;
++ struct super_block *sb;
++ struct dentry *h_parent;
++ struct inode *h_dir;
++ struct au_hinode *hdir;
+
-+ LKTRTrace("i%lu, wh %.*s, d %p\n",
-+ hdir->hi_inode->i_ino, AuDLNPair(wh_dentry), dentry);
-+ AuDebugOn((dentry && au_dbwh(dentry) < 0)
-+ || !wh_dentry->d_inode
-+ || !S_ISREG(wh_dentry->d_inode->i_mode));
++ /* rmdir by nfsd may cause deadlock with this i_mutex */
++ /* inode_lock(a->dir); */
++ err = -EROFS;
++ sb = a->dir->i_sb;
++ si_read_lock(sb, !AuLock_FLUSH);
++ if (!au_br_writable(a->br->br_perm))
++ goto out;
++ bindex = au_br_index(sb, a->br->br_id);
++ if (unlikely(bindex < 0))
++ goto out;
+
-+ err = do_unlink_wh(hdir, /*h_dir*/NULL, wh_dentry, dlgt);
-+ if (!err && dentry)
-+ au_set_dbwh(dentry, -1);
++ err = -EIO;
++ ii_write_lock_parent(a->dir);
++ h_parent = dget_parent(a->wh_dentry);
++ h_dir = d_inode(h_parent);
++ hdir = au_hi(a->dir, bindex);
++ err = vfsub_mnt_want_write(au_br_mnt(a->br));
++ if (unlikely(err))
++ goto out_mnt;
++ au_hn_inode_lock_nested(hdir, AuLsc_I_PARENT);
++ err = au_h_verify(a->wh_dentry, au_opt_udba(sb), h_dir, h_parent,
++ a->br);
++ if (!err)
++ err = au_whtmp_rmdir(a->dir, bindex, a->wh_dentry, &a->whlist);
++ au_hn_inode_unlock(hdir);
++ vfsub_mnt_drop_write(au_br_mnt(a->br));
+
-+ AuTraceErr(err);
-+ return err;
++out_mnt:
++ dput(h_parent);
++ ii_write_unlock(a->dir);
++out:
++ /* inode_unlock(a->dir); */
++ au_whtmp_rmdir_free(a);
++ si_read_unlock(sb);
++ au_nwt_done(&au_sbi(sb)->si_nowait);
++ if (unlikely(err))
++ AuIOErr("err %d\n", err);
+}
+
-+static int unlink_wh_name(struct dentry *h_parent, struct qstr *wh,
-+ struct au_ndx *ndx)
++void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex,
++ struct dentry *wh_dentry, struct au_whtmp_rmdir *args)
+{
-+ int err;
-+ struct dentry *wh_dentry;
++ int wkq_err;
++ struct super_block *sb;
+
-+ LKTRTrace("%.*s/%.*s\n", AuDLNPair(h_parent), AuLNPair(wh));
++ IMustLock(dir);
+
-+ /* au_test_h_perm() is already done */
-+ wh_dentry = au_lkup_one(wh->name, h_parent, wh->len, ndx);
-+ if (IS_ERR(wh_dentry))
-+ err = PTR_ERR(wh_dentry);
-+ else {
-+ err = 0;
-+ if (wh_dentry->d_inode && S_ISREG(wh_dentry->d_inode->i_mode))
-+ err = do_unlink_wh(/*hdir*/NULL, h_parent->d_inode,
-+ wh_dentry,
-+ au_ftest_ndx(ndx->flags, DLGT));
-+ dput(wh_dentry);
++ /* all post-process will be done in do_rmdir_whtmp(). */
++ sb = dir->i_sb;
++ args->dir = au_igrab(dir);
++ args->br = au_sbr(sb, bindex);
++ au_lcnt_inc(&args->br->br_count);
++ args->wh_dentry = dget(wh_dentry);
++ wkq_err = au_wkq_nowait(call_rmdir_whtmp, args, sb, /*flags*/0);
++ if (unlikely(wkq_err)) {
++ pr_warn("rmdir error %pd (%d), ignored\n", wh_dentry, wkq_err);
++ au_whtmp_rmdir_free(args);
+ }
-+
-+ AuTraceErr(err);
-+ return err;
+}
+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 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,87 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+/* ---------------------------------------------------------------------- */
++/*
++ * whiteout for logical deletion and opaque directory
++ */
+
-+static void clean_wh(struct inode *h_dir, struct path *whpath,
-+ struct au_hinode *hdir, struct vfsub_args *vargs)
-+{
-+ int err;
++#ifndef __AUFS_WHOUT_H__
++#define __AUFS_WHOUT_H__
+
-+ AuTraceEnter();
++#ifdef __KERNEL__
+
-+ if (!whpath->dentry->d_inode)
-+ return;
++#include "dir.h"
+
-+ err = au_mnt_want_write(whpath->mnt);
-+ if (!err) {
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_DELETE, hdir);
-+ err = vfsub_unlink(h_dir, whpath->dentry, vargs);
-+ au_mnt_drop_write(whpath->mnt);
-+ }
-+ if (unlikely(err))
-+ AuWarn("failed unlink %.*s (%d), ignored.\n",
-+ AuDLNPair(whpath->dentry), err);
-+}
++/* whout.c */
++int au_wh_name_alloc(struct qstr *wh, const struct qstr *name);
++int au_wh_test(struct mnt_idmap *h_idmap, struct path *h_ppath,
++ struct qstr *wh_name, int try_sio);
++int au_diropq_test(struct mnt_idmap *h_idmap, struct path *h_path);
++struct au_branch;
++struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct au_branch *br,
++ struct qstr *prefix);
++int au_whtmp_ren(struct dentry *h_dentry, struct au_branch *br);
++int au_wh_unlink_dentry(struct inode *h_dir, struct path *h_path,
++ struct dentry *dentry);
++int au_wh_init(struct au_branch *br, struct super_block *sb);
+
-+static void au_whdir_clean(struct inode *h_dir, struct path *whpath,
-+ struct au_hinode *hdir, struct vfsub_args *vargs)
-+{
-+ int err;
++/* diropq flags */
++#define AuDiropq_CREATE 1
++#define au_ftest_diropq(flags, name) ((flags) & AuDiropq_##name)
++#define au_fset_diropq(flags, name) \
++ do { (flags) |= AuDiropq_##name; } while (0)
++#define au_fclr_diropq(flags, name) \
++ do { (flags) &= ~AuDiropq_##name; } while (0)
+
-+ AuTraceEnter();
++struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex,
++ unsigned int flags);
++struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name,
++ struct au_branch *br);
++struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex,
++ struct dentry *h_parent);
+
-+ if (!whpath->dentry->d_inode)
-+ return;
++/* real rmdir for the whiteout-ed dir */
++struct au_whtmp_rmdir {
++ struct inode *dir;
++ struct au_branch *br;
++ struct dentry *wh_dentry;
++ struct au_nhash whlist;
++};
+
-+ err = au_mnt_want_write(whpath->mnt);
-+ if (!err) {
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_DELETE, hdir);
-+ err = vfsub_rmdir(h_dir, whpath->dentry, vargs);
-+ au_mnt_drop_write(whpath->mnt);
-+ }
-+ if (unlikely(err))
-+ AuWarn("failed rmdir %.*s (%d), ignored.\n",
-+ AuDLNPair(whpath->dentry), err);
-+}
++struct au_whtmp_rmdir *au_whtmp_rmdir_alloc(struct super_block *sb, gfp_t gfp);
++void au_whtmp_rmdir_free(struct au_whtmp_rmdir *whtmp);
++int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex,
++ struct dentry *wh_dentry, struct au_nhash *whlist);
++void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex,
++ struct dentry *wh_dentry, struct au_whtmp_rmdir *args);
++
++/* ---------------------------------------------------------------------- */
+
-+static int test_linkable(struct inode *h_dir)
++static inline struct dentry *au_diropq_create(struct dentry *dentry,
++ aufs_bindex_t bindex)
+{
-+ if (h_dir->i_op && h_dir->i_op->link)
-+ return 0;
-+ return -ENOSYS;
++ return au_diropq_sio(dentry, bindex, AuDiropq_CREATE);
+}
+
-+/* todo: should this mkdir be done in /sbin/mount.aufs script? */
-+static int au_whdir(struct inode *h_dir, struct path *path,
-+ struct au_hinode *hdir, struct vfsub_args *vargs)
++static inline int au_diropq_remove(struct dentry *dentry, aufs_bindex_t bindex)
+{
-+ int err;
-+
-+ err = -EEXIST;
-+ if (!path->dentry->d_inode) {
-+ int mode = S_IRWXU;
-+ if (au_test_nfs(path->dentry->d_sb))
-+ mode |= S_IXUGO;
-+ err = au_mnt_want_write(path->mnt);
-+ if (!err) {
-+ vfsub_args_reinit(vargs);
-+ vfsub_ign_hinode(vargs, IN_CREATE, hdir);
-+ err = vfsub_mkdir(h_dir, path->dentry, mode, vargs);
-+ au_mnt_drop_write(path->mnt);
-+ }
-+ } else if (S_ISDIR(path->dentry->d_inode->i_mode))
-+ err = 0;
-+ else
-+ AuErr("unknown %.*s exists\n", AuDLNPair(path->dentry));
-+
-+ return err;
++ return PTR_ERR(au_diropq_sio(dentry, bindex, !AuDiropq_CREATE));
+}
+
++#endif /* __KERNEL__ */
++#endif /* __AUFS_WHOUT_H__ */
+diff -urN /usr/share/empty/fs/aufs/wkq.c linux/fs/aufs/wkq.c
+--- /usr/share/empty/fs/aufs/wkq.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/wkq.c 2022-11-05 23:02:18.972555950 +0100
+@@ -0,0 +1,372 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * initialize the whiteout base file/dir for @br.
++ * Copyright (C) 2005-2022 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/>.
+ */
-+int au_wh_init(struct dentry *h_root, struct au_branch *br,
-+ struct vfsmount *h_mnt, struct super_block *sb,
-+ aufs_bindex_t bindex)
-+{
-+ int err, i;
-+ const unsigned int mnt_flags = au_mntflags(sb);
-+ const unsigned char do_plink = !!au_opt_test(mnt_flags, PLINK),
-+ do_hinotify = au_opt_test(mnt_flags, UDBA_INOTIFY);
-+ struct path path = {
-+ .mnt = h_mnt
-+ };
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct inode *h_dir;
-+ struct au_hinode *hdir;
-+ struct au_wbr *wbr = br->br_wbr;
-+ struct vfsmount *nfsmnt = au_do_nfsmnt(h_mnt);
-+ static const struct qstr base_name[] = {
-+ [AuBrWh_BASE] = {
-+ .name = AUFS_BASE_NAME,
-+ .len = sizeof(AUFS_BASE_NAME) - 1
-+ },
-+ [AuBrWh_PLINK] = {
-+ .name = AUFS_PLINKDIR_NAME,
-+ .len = sizeof(AUFS_PLINKDIR_NAME) - 1
-+ },
-+ [AuBrWh_TMP] = {
-+ .name = AUFS_TMPDIR_NAME,
-+ .len = sizeof(AUFS_TMPDIR_NAME) - 1
-+ }
-+ };
-+ struct {
-+ const struct qstr *name;
-+ struct dentry *dentry;
-+ } base[] = {
-+ [AuBrWh_BASE] = {
-+ .name = base_name + AuBrWh_BASE,
-+ .dentry = NULL
-+ },
-+ [AuBrWh_PLINK] = {
-+ .name = base_name + AuBrWh_PLINK,
-+ .dentry = NULL
-+ },
-+ [AuBrWh_TMP] = {
-+ .name = base_name + AuBrWh_TMP,
-+ .dentry = NULL
-+ }
-+ };
-+ struct au_ndx ndx = {
-+ .nfsmnt = nfsmnt,
-+ .flags = 0, /* always no dlgt */
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+
-+ LKTRTrace("nfsmnt %p\n", nfsmnt);
-+ WbrWhMustWriteLock(wbr);
-+ SiMustWriteLock(sb);
-+ h_dir = h_root->d_inode;
-+
-+ for (i = 0; i < AuBrWh_Last; i++) {
-+ /* doubly whiteouted */
-+ struct dentry *d;
-+ d = au_wh_lkup(h_root, (void *)base[i].name, &ndx);
-+ err = PTR_ERR(d);
-+ if (IS_ERR(d))
-+ goto out;
-+ base[i].dentry = d;
-+ if (!au_test_ecryptfs(d->d_sb))
-+ AuDebugOn(wbr
-+ && wbr->wbr_wh[i]
-+ && wbr->wbr_wh[i] != base[i].dentry);
-+ else
-+ /* ecryptfs problem?: it returns different dentry */
-+ AuDebugOn(wbr
-+ && wbr->wbr_wh[i]
-+ && !!wbr->wbr_wh[i] != !!base[i].dentry);
-+ }
-+
-+ if (wbr)
-+ for (i = 0; i < AuBrWh_Last; i++) {
-+ dput(wbr->wbr_wh[i]);
-+ wbr->wbr_wh[i] = NULL;
-+ }
-+
-+ err = 0;
-+ hdir = NULL;
-+ if (bindex >= 0 && do_hinotify)
-+ hdir = au_hi(sb->s_root->d_inode, bindex);
-+ vfsub_args_init(&vargs, &ign, au_test_dlgt(mnt_flags), 0);
-+
-+ switch (br->br_perm) {
-+ case AuBrPerm_RR:
-+ case AuBrPerm_RO:
-+ case AuBrPerm_RRWH:
-+ case AuBrPerm_ROWH:
-+ path.dentry = base[AuBrWh_BASE].dentry;
-+ clean_wh(h_dir, &path, hdir, &vargs);
-+ path.dentry = base[AuBrWh_PLINK].dentry;
-+ au_whdir_clean(h_dir, &path, hdir, &vargs);
-+ path.dentry = base[AuBrWh_TMP].dentry;
-+ au_whdir_clean(h_dir, &path, hdir, &vargs);
-+ break;
+
-+ case AuBrPerm_RWNoLinkWH:
-+ path.dentry = base[AuBrWh_BASE].dentry;
-+ clean_wh(h_dir, &path, hdir, &vargs);
-+ path.dentry = base[AuBrWh_PLINK].dentry;
-+ if (do_plink) {
-+ err = test_linkable(h_dir);
-+ if (unlikely(err))
-+ goto out_nolink;
++/*
++ * workqueue for asynchronous/super-io operations
++ * todo: try new credential scheme
++ */
+
-+ err = au_whdir(h_dir, &path, hdir, &vargs);
-+ if (unlikely(err))
-+ goto out_err;
-+ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry);
-+ } else
-+ au_whdir_clean(h_dir, &path, hdir, &vargs);
-+ path.dentry = base[AuBrWh_TMP].dentry;
-+ err = au_whdir(h_dir, &path, hdir, &vargs);
-+ if (unlikely(err))
-+ goto out_err;
-+ wbr->wbr_tmp = dget(base[AuBrWh_TMP].dentry);
-+ break;
++#include <linux/module.h>
++#include "aufs.h"
+
-+ case AuBrPerm_RW:
-+ /*
-+ * for the moment, aufs supports the branch filesystem
-+ * which does not support link(2).
-+ * testing on FAT which does not support i_op->setattr() fully
-+ * either, copyup failed.
-+ * finally, such filesystem will not be used as the writable
-+ * branch.
-+ */
-+ err = test_linkable(h_dir);
-+ if (unlikely(err))
-+ goto out_nolink;
++/* internal workqueue named AUFS_WKQ_NAME */
+
-+ err = -EEXIST;
-+ /*
-+ * todo: should this create be done
-+ * in /sbin/mount.aufs script?
-+ */
-+ if (!base[AuBrWh_BASE].dentry->d_inode) {
-+ err = au_mnt_want_write(h_mnt);
-+ if (!err) {
-+ vfsub_args_reinit(&vargs);
-+ vfsub_ign_hinode(&vargs, IN_CREATE, hdir);
-+ err = au_h_create
-+ (h_dir, base[AuBrWh_BASE].dentry,
-+ WH_MASK, &vargs, /*nd*/NULL, nfsmnt);
-+ au_mnt_drop_write(h_mnt);
-+ }
-+ } else if (S_ISREG(base[AuBrWh_BASE].dentry->d_inode->i_mode))
-+ err = 0;
-+ else
-+ AuErr("unknown %.*s/%.*s exists\n",
-+ AuDLNPair(h_root),
-+ AuDLNPair(base[AuBrWh_BASE].dentry));
-+ if (unlikely(err))
-+ goto out_err;
++static struct workqueue_struct *au_wkq;
+
-+ path.dentry = base[AuBrWh_PLINK].dentry;
-+ if (do_plink) {
-+ err = au_whdir(h_dir, &path, hdir, &vargs);
-+ if (unlikely(err))
-+ goto out_err;
-+ wbr->wbr_plink = dget(base[AuBrWh_PLINK].dentry);
-+ } else
-+ au_whdir_clean(h_dir, &path, hdir, &vargs);
-+ wbr->wbr_whbase = dget(base[AuBrWh_BASE].dentry);
++struct au_wkinfo {
++ struct work_struct wk;
++ struct kobject *kobj;
+
-+ path.dentry = base[AuBrWh_TMP].dentry;
-+ err = au_whdir(h_dir, &path, hdir, &vargs);
-+ if (unlikely(err))
-+ goto out_err;
-+ wbr->wbr_tmp = dget(base[AuBrWh_TMP].dentry);
-+ break;
++ unsigned int flags; /* see wkq.h */
+
-+ default:
-+ BUG();
-+ }
++ au_wkq_func_t func;
++ void *args;
+
-+ out:
-+ for (i = 0; i < AuBrWh_Last; i++)
-+ dput(base[i].dentry);
-+ AuTraceErr(err);
-+ return err;
-+ out_nolink:
-+ AuErr("%.*s doesn't support link(2), use noplink and rw+nolwh\n",
-+ AuDLNPair(h_root));
-+ goto out;
-+ out_err:
-+ AuErr("an error(%d) on the writable branch %.*s(%s)\n",
-+ err, AuDLNPair(h_root), au_sbtype(h_root->d_sb));
-+ goto out;
-+}
++#ifdef CONFIG_LOCKDEP
++ int dont_check;
++ struct held_lock **hlock;
++#endif
+
-+struct reinit_br_wh {
-+ struct super_block *sb;
-+ struct au_branch *br;
++ struct completion *comp;
+};
+
-+static void reinit_br_wh(void *arg)
-+{
-+ int err;
-+ struct reinit_br_wh *a = arg;
-+ struct au_wbr *wbr;
-+ struct inode *h_dir, *dir;
-+ struct dentry *h_root;
-+ aufs_bindex_t bindex;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
++/* ---------------------------------------------------------------------- */
++/*
++ * 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.
++ */
+
-+ AuTraceEnter();
-+ AuDebugOn(current->fsuid);
++#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;
++}
+
-+ err = 0;
-+ wbr = a->br->br_wbr;
-+ /* big aufs lock */
-+ si_noflush_write_lock(a->sb);
-+ if (!au_br_writable(a->br->br_perm))
-+ goto out;
-+ bindex = au_br_index(a->sb, a->br->br_id);
-+ if (unlikely(bindex < 0))
-+ goto out;
++/*
++ * 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;
+
-+ AuDebugOn(!wbr);
-+ AuDebugOn(!wbr->wbr_whbase || !wbr->wbr_whbase->d_inode);
++ /* 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;
++ }
+
-+ dir = a->sb->s_root->d_inode;
-+ ii_read_lock_parent(dir);
-+ h_root = dget_parent(wbr->wbr_whbase);
-+ h_dir = h_root->d_inode;
-+ AuDebugOn(!h_dir->i_op || !h_dir->i_op->link);
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ wbr_wh_write_lock(wbr);
-+ if (!au_verify_parent(wbr->wbr_whbase, h_dir)) {
-+ err = au_br_want_write(a->br);
-+ if (!err) {
-+ vfsub_args_init(&vargs, &ign, /*dlgt*/0, 0);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, au_hi(dir, bindex));
-+ err = vfsub_unlink(h_dir, wbr->wbr_whbase, &vargs);
-+ au_br_drop_write(a->br);
++ 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;
+ }
-+ } else {
-+ AuWarn("%.*s is moved, ignored\n", AuDLNPair(wbr->wbr_whbase));
-+ err = 0;
-+ }
-+ dput(wbr->wbr_whbase);
-+ wbr->wbr_whbase = NULL;
-+ if (!err)
-+ err = au_wh_init(h_root, a->br, a->br->br_mnt, a->sb, bindex);
-+ wbr_wh_write_unlock(wbr);
-+ mutex_unlock(&h_dir->i_mutex);
-+ dput(h_root);
-+ ii_read_unlock(dir);
++ 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;
+
-+ out:
-+ if (wbr)
-+ atomic_dec_return(&wbr->wbr_wh_running);
-+ au_br_put(a->br);
-+ au_nwt_done(&au_sbi(a->sb)->si_nowait);
-+ si_write_unlock(a->sb);
-+ kfree(arg);
-+ if (unlikely(err))
-+ AuIOErr("err %d\n", err);
++match:
++ return 1;
++unmatch:
++ return 0;
+}
+
-+static void kick_reinit_br_wh(struct super_block *sb, struct au_branch *br)
++static int au_wkq_lockdep_alloc(struct au_wkinfo *wkinfo)
+{
-+ int do_dec, wkq_err;
-+ struct reinit_br_wh *arg;
++ int err, n;
++ struct task_struct *curr;
++ struct held_lock **hl, *held_locks, *p;
+
-+ AuTraceEnter();
-+ AuDebugOn(!br->br_wbr);
++ err = 0;
++ curr = current;
++ wkinfo->dont_check = lockdep_recursing(curr);
++ if (wkinfo->dont_check)
++ goto out;
++ n = curr->lockdep_depth;
++ if (!n)
++ goto out;
+
-+ do_dec = 1;
-+ if (atomic_inc_return(&br->br_wbr->wbr_wh_running) != 1)
++ err = -ENOMEM;
++ wkinfo->hlock = kmalloc_array(n + 1, sizeof(*wkinfo->hlock), GFP_NOFS);
++ if (unlikely(!wkinfo->hlock))
+ goto out;
+
-+ /* ignore ENOMEM */
-+ arg = kmalloc(sizeof(*arg), GFP_NOFS);
-+ if (arg) {
-+ /*
-+ * dec(wh_running), kfree(arg) and au_br_put()
-+ * in reinit function
-+ */
-+ arg->sb = sb;
-+ arg->br = br;
-+ au_br_get(br);
-+ wkq_err = au_wkq_nowait(reinit_br_wh, arg, sb, /*dlgt*/0);
-+ if (unlikely(wkq_err)) {
-+ atomic_dec_return(&br->br_wbr->wbr_wh_running);
-+ au_br_put(br);
-+ kfree(arg);
-+ }
-+ do_dec = 0;
++ err = 0;
++#if 0 /* left for debugging */
++ if (0 && au_debug_test())
++ 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:
-+ if (do_dec)
-+ atomic_dec_return(&br->br_wbr->wbr_wh_running);
++out:
++ return err;
+}
+
-+/*
-+ * create the whiteout @wh.
-+ */
-+static int link_or_create_wh(struct super_block *sb, aufs_bindex_t bindex,
-+ struct dentry *wh, struct inode *dir)
++static void au_wkq_lockdep_free(struct au_wkinfo *wkinfo)
+{
-+ int err, dlgt;
-+ struct au_branch *br;
-+ struct au_wbr *wbr;
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(wh));
-+ h_parent = wh->d_parent; /* dir inode is locked */
-+ h_dir = h_parent->d_inode;
-+ IMustLock(h_dir);
-+ br = au_sbr(sb, bindex);
-+ wbr = br->br_wbr;
-+ AuDebugOn(!wbr);
++ au_kfree_try_rcu(wkinfo->hlock);
++}
+
-+ dlgt = !!au_test_dlgt(au_mntflags(sb));
-+ wbr_wh_read_lock(wbr);
-+ if (wbr->wbr_whbase) {
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
-+ if (dir)
-+ vfsub_ign_hinode(&vargs, IN_CREATE, au_hi(dir, bindex));
-+ err = vfsub_link(wbr->wbr_whbase, h_dir, wh, &vargs);
-+ if (!err || err != -EMLINK)
-+ goto out;
++static void au_wkq_lockdep_pre(struct au_wkinfo *wkinfo)
++{
++ struct held_lock *p, **hl = wkinfo->hlock;
++ int subclass;
+
-+ /* link count full. re-initialize br_whbase. */
-+ kick_reinit_br_wh(sb, br);
++ 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_);
+ }
++}
+
-+ /* return this error in this context */
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
-+ if (dir)
-+ vfsub_ign_hinode(&vargs, IN_CREATE, au_hi(dir, bindex));
-+ err = au_h_create(h_dir, wh, WH_MASK, &vargs, /*nd*/NULL,
-+ au_do_nfsmnt(br->br_mnt));
++static void au_wkq_lockdep_post(struct au_wkinfo *wkinfo)
++{
++ struct held_lock *p, **hl = wkinfo->hlock;
+
-+ out:
-+ wbr_wh_read_unlock(wbr);
-+ AuTraceErr(err);
-+ return err;
++ if (wkinfo->dont_check)
++ lockdep_on();
++ if (!hl)
++ return;
++ while ((p = *hl++)) /* assignment */
++ rwsem_release(p->instance, /*p->acquire_ip*/_RET_IP_);
+}
++#endif
+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * create or remove the diropq.
-+ */
-+static struct dentry *do_diropq(struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned int flags)
++static void wkq_func(struct work_struct *wk)
+{
-+ struct dentry *opq_dentry, *h_dentry;
-+ struct inode *h_dir;
-+ int err, dlgt;
-+ struct super_block *sb;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+
-+ LKTRTrace("%.*s, bindex %d, flags 0x%x\n",
-+ AuDLNPair(dentry), bindex, flags);
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ AuDebugOn(!h_dentry);
-+ h_dir = h_dentry->d_inode;
-+ AuDebugOn(!h_dir || !S_ISDIR(h_dir->i_mode));
++ struct au_wkinfo *wkinfo = container_of(wk, struct au_wkinfo, wk);
+
-+ /* already checked by au_test_h_perm(). */
-+ sb = dentry->d_sb;
-+ ndx.nfsmnt = au_nfsmnt(sb, bindex);
-+ dlgt = 0;
-+ if (au_ftest_diropq(flags, DLGT)) {
-+ dlgt = 1;
-+ au_fset_ndx(ndx.flags, DLGT);
-+ }
-+ opq_dentry = au_lkup_one(diropq_name.name, h_dentry, diropq_name.len,
-+ &ndx);
-+ if (IS_ERR(opq_dentry))
-+ goto out;
++ AuDebugOn(!uid_eq(current_fsuid(), GLOBAL_ROOT_UID));
++ AuDebugOn(rlimit(RLIMIT_FSIZE) != RLIM_INFINITY);
+
-+ if (au_ftest_diropq(flags, CREATE)) {
-+ AuDebugOn(opq_dentry->d_inode);
-+ err = link_or_create_wh(dentry->d_sb, bindex, opq_dentry,
-+ dentry->d_inode);
-+ if (!err) {
-+ au_set_dbdiropq(dentry, bindex);
-+ goto out; /* success */
-+ }
-+ } else {
-+ AuDebugOn(/* !S_ISDIR(dentry->d_inode->i_mode)
-+ * || */!opq_dentry->d_inode);
-+ err = do_unlink_wh(au_hi(dentry->d_inode, bindex),
-+ /*h_dir*/NULL, opq_dentry, dlgt);
-+ if (!err)
-+ au_set_dbdiropq(dentry, -1);
++ au_wkq_lockdep_pre(wkinfo);
++ wkinfo->func(wkinfo->args);
++ au_wkq_lockdep_post(wkinfo);
++ if (au_ftest_wkq(wkinfo->flags, WAIT))
++ complete(wkinfo->comp);
++ else {
++ kobject_put(wkinfo->kobj);
++ module_put(THIS_MODULE); /* todo: ?? */
++ au_kfree_rcu(wkinfo);
+ }
-+ dput(opq_dentry);
-+ opq_dentry = ERR_PTR(err);
-+
-+ out:
-+ AuTraceErrPtr(opq_dentry);
-+ return opq_dentry;
+}
+
-+struct do_diropq_args {
-+ struct dentry **errp;
-+ struct dentry *dentry;
-+ aufs_bindex_t bindex;
-+ unsigned int flags;
-+};
++/*
++ * Since struct completion is large, try allocating it dynamically.
++ */
++#define AuWkqCompDeclare(name) struct completion *comp = NULL
+
-+static void call_do_diropq(void *args)
++static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp)
+{
-+ struct do_diropq_args *a = args;
-+ *a->errp = do_diropq(a->dentry, a->bindex, a->flags);
++ *comp = kmalloc(sizeof(**comp), GFP_NOFS);
++ if (*comp) {
++ init_completion(*comp);
++ wkinfo->comp = *comp;
++ return 0;
++ }
++ return -ENOMEM;
+}
+
-+struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned int flags)
++static void au_wkq_comp_free(struct completion *comp)
+{
-+ struct dentry *diropq, *h_dentry;
++ au_kfree_rcu(comp);
++}
+
-+ LKTRTrace("%.*s, bindex %d, flags 0x%x\n",
-+ AuDLNPair(dentry), bindex, flags);
++static void au_wkq_run(struct au_wkinfo *wkinfo)
++{
++ if (au_ftest_wkq(wkinfo->flags, NEST)) {
++ if (au_wkq_test()) {
++ AuWarn1("wkq from wkq, unless silly-rename on NFS,"
++ " due to a dead dir by UDBA,"
++ " or async xino write?\n");
++ AuDebugOn(au_ftest_wkq(wkinfo->flags, WAIT));
++ }
++ } else
++ au_dbg_verify_kthread();
+
-+ h_dentry = au_h_dptr(dentry, bindex);
-+ if (!au_test_h_perm_sio(h_dentry->d_inode, MAY_EXEC | MAY_WRITE,
-+ au_ftest_diropq(flags, DLGT)))
-+ diropq = do_diropq(dentry, bindex, flags);
-+ else {
-+ int wkq_err;
-+ struct do_diropq_args args = {
-+ .errp = &diropq,
-+ .dentry = dentry,
-+ .bindex = bindex,
-+ .flags = flags
-+ };
-+ wkq_err = au_wkq_wait(call_do_diropq, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ diropq = ERR_PTR(wkq_err);
++ if (au_ftest_wkq(wkinfo->flags, WAIT)) {
++ INIT_WORK_ONSTACK(&wkinfo->wk, wkq_func);
++ queue_work(au_wkq, &wkinfo->wk);
++ } else {
++ INIT_WORK(&wkinfo->wk, wkq_func);
++ schedule_work(&wkinfo->wk);
+ }
-+
-+ AuTraceErrPtr(diropq);
-+ return diropq;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
+/*
-+ * lookup whiteout dentry.
-+ * @h_parent: hidden parent dentry which must exist and be locked
-+ * @base_name: name of dentry which will be whiteouted
-+ * returns dentry for whiteout.
++ * Be careful. It is easy to make deadlock happen.
++ * processA: lock, wkq and wait
++ * processB: wkq and wait, lock in wkq
++ * --> deadlock
+ */
-+struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name,
-+ struct au_ndx *ndx)
++int au_wkq_do_wait(unsigned int flags, au_wkq_func_t func, void *args)
+{
+ int err;
-+ struct qstr wh_name;
-+ struct dentry *wh_dentry;
-+
-+ LKTRTrace("%.*s/%.*s\n", AuDLNPair(h_parent), AuLNPair(base_name));
++ AuWkqCompDeclare(comp);
++ struct au_wkinfo wkinfo = {
++ .flags = flags,
++ .func = func,
++ .args = args
++ };
+
-+ err = au_wh_name_alloc(base_name->name, base_name->len, &wh_name);
-+ wh_dentry = ERR_PTR(err);
++ 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) {
-+ /* do not superio. */
-+ wh_dentry = au_lkup_one(wh_name.name, h_parent,
-+ wh_name.len, ndx);
-+ au_wh_name_free(&wh_name);
++ au_wkq_run(&wkinfo);
++ /* no timeout, no interrupt */
++ wait_for_completion(wkinfo.comp);
+ }
-+ AuTraceErrPtr(wh_dentry);
-+ return wh_dentry;
++ au_wkq_lockdep_free(&wkinfo);
++
++out_comp:
++ au_wkq_comp_free(comp);
++out:
++ destroy_work_on_stack(&wkinfo.wk);
++ return err;
+}
+
+/*
-+ * link/create a whiteout for @dentry on @bindex.
++ * Note: dget/dput() in func for aufs dentries are not supported. It will be a
++ * problem in a concurrent umounting.
+ */
-+struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, struct au_ndx *ndx)
++int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb,
++ unsigned int flags)
+{
-+ struct dentry *wh_dentry;
-+ struct inode *dir;
+ int err;
-+ struct super_block *sb;
++ struct au_wkinfo *wkinfo;
+
-+ LKTRTrace("%.*s/%.*s on b%d\n", AuDLNPair(h_parent),
-+ AuDLNPair(dentry), bindex);
++ atomic_inc(&au_sbi(sb)->si_nowait.nw_len);
+
-+ sb = dentry->d_sb;
-+ wh_dentry = au_wh_lkup(h_parent, &dentry->d_name, ndx);
-+ if (!IS_ERR(wh_dentry) && !wh_dentry->d_inode) {
-+ dir = dentry->d_parent->d_inode; /* dir is locked */
-+ IMustLock(dir);
-+ err = link_or_create_wh(sb, bindex, wh_dentry, dir);
-+ if (!err)
-+ au_set_dbwh(dentry, bindex);
-+ else {
-+ dput(wh_dentry);
-+ wh_dentry = ERR_PTR(err);
-+ }
++ /*
++ * wkq_func() must free this wkinfo.
++ * it highly depends upon the implementation of workqueue.
++ */
++ err = 0;
++ wkinfo = kmalloc(sizeof(*wkinfo), GFP_NOFS);
++ if (wkinfo) {
++ wkinfo->kobj = &au_sbi(sb)->si_kobj;
++ wkinfo->flags = flags & ~AuWkq_WAIT;
++ wkinfo->func = func;
++ wkinfo->args = args;
++ wkinfo->comp = NULL;
++ au_wkq_lockdep_init(wkinfo);
++ kobject_get(wkinfo->kobj);
++ __module_get(THIS_MODULE); /* todo: ?? */
++
++ au_wkq_run(wkinfo);
++ } else {
++ err = -ENOMEM;
++ au_nwt_done(&au_sbi(sb)->si_nowait);
+ }
+
-+ AuTraceErrPtr(wh_dentry);
-+ return wh_dentry;
++ return err;
+}
+
+/* ---------------------------------------------------------------------- */
+
-+/* Delete all whiteouts in this directory on branch bindex. */
-+static int del_wh_children(struct dentry *h_dentry, struct au_nhash *whlist,
-+ aufs_bindex_t bindex, struct au_ndx *ndx)
++void au_nwt_init(struct au_nowait_tasks *nwt)
+{
-+ int err, i;
-+ struct qstr wh_name;
-+ char *p;
-+ struct inode *h_inode;
-+ struct hlist_head *head;
-+ struct au_vdir_wh *tpos;
-+ struct hlist_node *pos;
-+ struct au_vdir_destr *str;
++ atomic_set(&nwt->nw_len, 0);
++ /* smp_mb(); */ /* atomic_set */
++ init_waitqueue_head(&nwt->nw_wq);
++}
+
-+ LKTRTrace("%.*s\n", AuDLNPair(h_dentry));
-+ h_inode = h_dentry->d_inode;
-+ AuDebugOn(IS_RDONLY(h_inode));
++void au_wkq_fin(void)
++{
++ destroy_workqueue(au_wkq);
++}
+
-+ err = -ENOMEM;
-+ p = __getname();
-+ wh_name.name = p;
-+ if (unlikely(!wh_name.name))
-+ goto out;
-+ memcpy(p, AUFS_WH_PFX, AUFS_WH_PFX_LEN);
-+ p += AUFS_WH_PFX_LEN;
++int __init au_wkq_init(void)
++{
++ int err;
+
-+ /* already checked by au_test_h_perm(). */
+ err = 0;
-+ for (i = 0; !err && i < AuSize_NHASH; i++) {
-+ head = whlist->heads + i;
-+ hlist_for_each_entry(tpos, pos, head, wh_hash) {
-+ if (tpos->wh_bindex != bindex)
-+ continue;
-+ str = &tpos->wh_str;
-+ if (str->len + AUFS_WH_PFX_LEN <= PATH_MAX) {
-+ memcpy(p, str->name, str->len);
-+ wh_name.len = AUFS_WH_PFX_LEN + str->len;
-+ err = unlink_wh_name(h_dentry, &wh_name, ndx);
-+ if (!err)
-+ continue;
-+ break;
-+ }
-+ AuIOErr("whiteout name too long %.*s\n",
-+ str->len, str->name);
-+ err = -EIO;
-+ break;
-+ }
-+ }
-+ __putname(wh_name.name);
++ au_wkq = alloc_workqueue(AUFS_WKQ_NAME, 0, WQ_DFL_ACTIVE);
++ if (IS_ERR(au_wkq))
++ err = PTR_ERR(au_wkq);
++ else if (!au_wkq)
++ err = -ENOMEM;
++
++ return err;
++}
+diff -urN /usr/share/empty/fs/aufs/wkq.h linux/fs/aufs/wkq.h
+--- /usr/share/empty/fs/aufs/wkq.h 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/wkq.h 2022-11-05 23:02:18.972555950 +0100
+@@ -0,0 +1,89 @@
++/* SPDX-License-Identifier: GPL-2.0 */
++/*
++ * Copyright (C) 2005-2022 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/>.
++ */
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++/*
++ * workqueue for asynchronous/super-io operations
++ * todo: try new credentials management scheme
++ */
+
-+struct del_wh_children_args {
-+ int *errp;
-+ struct dentry *h_dentry;
-+ struct au_nhash *whlist;
-+ aufs_bindex_t bindex;
-+ struct au_ndx *ndx;
-+};
++#ifndef __AUFS_WKQ_H__
++#define __AUFS_WKQ_H__
+
-+static void call_del_wh_children(void *args)
-+{
-+ struct del_wh_children_args *a = args;
-+ *a->errp = del_wh_children(a->h_dentry, a->whlist, a->bindex, a->ndx);
-+}
++#ifdef __KERNEL__
++
++#include <linux/wait.h>
++
++struct super_block;
+
+/* ---------------------------------------------------------------------- */
+
+/*
-+ * rmdir the whiteouted temporary named dir @h_dentry.
-+ * @whlist: whiteouted children.
++ * in the next operation, wait for the 'nowait' tasks in system-wide workqueue
+ */
-+int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *wh_dentry, struct au_nhash *whlist)
-+{
-+ int err, dlgt;
-+ struct inode *wh_inode, *h_dir;
-+ struct super_block *sb;
-+ unsigned int mnt_flags;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_ndx ndx = {
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
-+
-+ LKTRTrace("i%lu, %.*s, b%d\n",
-+ dir->i_ino, AuDLNPair(wh_dentry), bindex);
-+ /* IMustLock(dir); */
-+ IiMustAnyLock(dir);
-+ h_dir = wh_dentry->d_parent->d_inode; /* dir inode is locked */
-+ IMustLock(h_dir);
++struct au_nowait_tasks {
++ atomic_t nw_len;
++ wait_queue_head_t nw_wq;
++};
+
-+ sb = dir->i_sb;
-+ mnt_flags = au_mntflags(sb);
-+ dlgt = !!au_test_dlgt(mnt_flags);
-+ if (dlgt)
-+ au_fset_ndx(ndx.flags, DLGT);
-+ ndx.nfsmnt = au_nfsmnt(sb, bindex);
-+ wh_inode = wh_dentry->d_inode;
-+ mutex_lock_nested(&wh_inode->i_mutex, AuLsc_I_CHILD);
++/* ---------------------------------------------------------------------- */
+
-+ /*
-+ * someone else might change some whiteouts while we were sleeping.
-+ * it means this whlist may have an obsoleted entry.
-+ */
-+ if (!au_test_h_perm_sio(wh_inode, MAY_EXEC | MAY_WRITE, dlgt))
-+ err = del_wh_children(wh_dentry, whlist, bindex, &ndx);
-+ else {
-+ int wkq_err;
-+ /* ugly */
-+ unsigned int flags = ndx.flags;
-+ struct del_wh_children_args args = {
-+ .errp = &err,
-+ .h_dentry = wh_dentry,
-+ .whlist = whlist,
-+ .bindex = bindex,
-+ .ndx = &ndx
-+ };
++typedef void (*au_wkq_func_t)(void *args);
+
-+ ndx.flags = 0;
-+ wkq_err = au_wkq_wait(call_del_wh_children, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ ndx.flags = flags;
-+ }
-+ mutex_unlock(&wh_inode->i_mutex);
++/* wkq flags */
++#define AuWkq_WAIT 1
++#define AuWkq_NEST (1 << 1)
++#define au_ftest_wkq(flags, name) ((flags) & AuWkq_##name)
++#define au_fset_wkq(flags, name) \
++ do { (flags) |= AuWkq_##name; } while (0)
++#define au_fclr_wkq(flags, name) \
++ do { (flags) &= ~AuWkq_##name; } while (0)
+
-+ if (!err) {
-+ vfsub_args_init(&vargs, &ign, dlgt, 0);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, au_hi(dir, bindex));
-+ err = vfsub_rmdir(h_dir, wh_dentry, &vargs);
-+ /* d_drop(h_dentry); */
-+ }
++/* wkq.c */
++int au_wkq_do_wait(unsigned int flags, au_wkq_func_t func, void *args);
++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);
++int __init au_wkq_init(void);
++void au_wkq_fin(void);
+
-+ if (!err) {
-+ if (au_ibstart(dir) == bindex) {
-+ au_cpup_attr_timesizes(dir);
-+ /* au_cpup_attr_nlink(dir); */
-+ drop_nlink(dir);
-+ }
-+ return 0; /* success */
-+ }
++/* ---------------------------------------------------------------------- */
+
-+ AuWarn("failed removing %.*s(%d), ignored\n",
-+ AuDLNPair(wh_dentry), err);
-+ return err;
++static inline int au_wkq_test(void)
++{
++ return current->flags & PF_WQ_WORKER;
+}
+
-+static void au_whtmp_rmdir_free_args(struct au_whtmp_rmdir_args *args)
++static inline int au_wkq_wait(au_wkq_func_t func, void *args)
+{
-+ au_nhash_fin(&args->whlist);
-+ dput(args->wh_dentry);
-+ iput(args->dir);
-+ kfree(args);
++ return au_wkq_do_wait(AuWkq_WAIT, func, args);
+}
+
-+static void call_rmdir_whtmp(void *args)
++static inline void au_nwt_done(struct au_nowait_tasks *nwt)
+{
-+ int err;
-+ struct au_whtmp_rmdir_args *a = args;
-+ struct super_block *sb;
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
-+
-+ LKTRTrace("%.*s, b%d, dir i%lu\n",
-+ AuDLNPair(a->wh_dentry), a->bindex, a->dir->i_ino);
-+
-+ /* rmdir by nfsd may cause deadlock with this i_mutex */
-+ /* mutex_lock(&a->dir->i_mutex); */
-+ sb = a->dir->i_sb;
-+ si_noflush_read_lock(sb);
-+ err = au_test_ro(sb, a->bindex, NULL);
-+ if (unlikely(err))
-+ goto out;
-+
-+ err = -EIO;
-+ ii_write_lock_parent(a->dir);
-+ h_parent = dget_parent(a->wh_dentry);
-+ h_dir = h_parent->d_inode;
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ if (!au_verify_parent(a->wh_dentry, h_dir)) {
-+ err = au_br_want_write(au_sbr(sb, a->bindex));
-+ if (!err) {
-+ err = au_whtmp_rmdir(a->dir, a->bindex, a->wh_dentry,
-+ &a->whlist);
-+ au_br_drop_write(au_sbr(sb, a->bindex));
-+ }
-+ }
-+ mutex_unlock(&h_dir->i_mutex);
-+ dput(h_parent);
-+ ii_write_unlock(a->dir);
-+
-+ out:
-+ /* mutex_unlock(&a->dir->i_mutex); */
-+ au_nwt_done(&au_sbi(sb)->si_nowait);
-+ si_read_unlock(sb);
-+ au_whtmp_rmdir_free_args(a);
-+ if (unlikely(err))
-+ AuIOErr("err %d\n", err);
++ if (atomic_dec_and_test(&nwt->nw_len))
++ wake_up_all(&nwt->nw_wq);
+}
+
-+void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *wh_dentry, struct au_nhash *whlist,
-+ struct au_whtmp_rmdir_args *args)
++static inline int au_nwt_flush(struct au_nowait_tasks *nwt)
+{
-+ int wkq_err;
-+
-+ LKTRTrace("%.*s\n", AuDLNPair(wh_dentry));
-+ IMustLock(dir);
-+
-+ /* all post-process will be done in do_rmdir_whtmp(). */
-+ args->dir = au_igrab(dir);
-+ args->bindex = bindex;
-+ args->wh_dentry = dget(wh_dentry);
-+ au_nhash_init(&args->whlist);
-+ au_nhash_move(&args->whlist, whlist);
-+ wkq_err = au_wkq_nowait(call_rmdir_whtmp, args, dir->i_sb, /*dlgt*/0);
-+ if (unlikely(wkq_err)) {
-+ AuWarn("rmdir error %.*s (%d), ignored\n",
-+ AuDLNPair(wh_dentry), wkq_err);
-+ au_whtmp_rmdir_free_args(args);
-+ }
++ wait_event(nwt->nw_wq, !atomic_read(&nwt->nw_len));
++ return 0;
+}
-diff -urN linux/fs/aufs/whout.h linux-aufs/fs/aufs/whout.h
---- linux/fs/aufs/whout.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/whout.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,137 @@
++
++#endif /* __KERNEL__ */
++#endif /* __AUFS_WKQ_H__ */
+diff -urN /usr/share/empty/fs/aufs/xattr.c linux/fs/aufs/xattr.c
+--- /usr/share/empty/fs/aufs/xattr.c 1970-01-01 01:00:00.000000000 +0100
++++ linux/fs/aufs/xattr.c 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,360 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2014-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * whiteout for logical deletion and opaque directory
-+ *
-+ * $Id$
++ * handling xattr functions
+ */
+
-+#ifndef __AUFS_WHOUT_H__
-+#define __AUFS_WHOUT_H__
++#include <linux/fs.h>
++#include <linux/xattr.h>
++#include "aufs.h"
+
-+#ifdef __KERNEL__
++static int au_xattr_ignore(int err, char *name, unsigned int ignore_flags)
++{
++ if (!ignore_flags)
++ goto out;
++ switch (err) {
++ case -ENOMEM:
++ case -EDQUOT:
++ goto out;
++ }
+
-+#include <linux/fs.h>
-+#include <linux/aufs_type.h>
-+#include "dir.h"
-+#include "opts.h"
-+#include "super.h"
++ if ((ignore_flags & AuBrAttr_ICEX) == AuBrAttr_ICEX) {
++ err = 0;
++ goto out;
++ }
+
-+int au_wh_name_alloc(const char *name, int len, struct qstr *wh);
-+void au_wh_name_free(struct qstr *wh);
++#define cmp(brattr, prefix) do { \
++ if (!strncmp(name, XATTR_##prefix##_PREFIX, \
++ XATTR_##prefix##_PREFIX_LEN)) { \
++ if (ignore_flags & AuBrAttr_ICEX_##brattr) \
++ err = 0; \
++ goto out; \
++ } \
++ } while (0)
+
-+struct au_ndx;
-+int au_wh_test(struct dentry *h_parent, struct qstr *wh_name, int try_sio,
-+ struct au_ndx *ndx);
-+int au_diropq_test(struct dentry *h_dentry, struct au_ndx *ndx);
++ cmp(SEC, SECURITY);
++ cmp(SYS, SYSTEM);
++ cmp(TR, TRUSTED);
++ cmp(USR, USER);
++#undef cmp
+
-+struct dentry *au_whtmp_lkup(struct dentry *h_parent, struct qstr *prefix,
-+ struct au_ndx *ndx);
-+int au_whtmp_ren(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *h_dentry);
-+int au_wh_unlink_dentry(struct au_hinode *dir, struct dentry *wh_dentry,
-+ struct dentry *dentry, int dlgt);
++ if (ignore_flags & AuBrAttr_ICEX_OTH)
++ err = 0;
+
-+struct au_branch;
-+int au_wh_init(struct dentry *h_parent, struct au_branch *br,
-+ struct vfsmount *nfsmnt, struct super_block *sb,
-+ aufs_bindex_t bindex);
++out:
++ return err;
++}
+
-+/* diropq flags */
-+#define AuDiropq_CREATE 1
-+#define AuDiropq_DLGT (1 << 1)
-+#define au_ftest_diropq(flags, name) ((flags) & AuDiropq_##name)
-+#define au_fset_diropq(flags, name) { (flags) |= AuDiropq_##name; }
-+#define au_fclr_diropq(flags, name) { (flags) &= ~AuDiropq_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuDiropq_DLGT
-+#define AuDiropq_DLGT 0
-+#endif
++static const int au_xattr_out_of_list = AuBrAttr_ICEX_OTH << 1;
++
++static int au_do_cpup_xattr(struct path *h_dst, struct path *h_src,
++ char *name, char **buf, unsigned int ignore_flags,
++ unsigned int verbose)
++{
++ int err, is_acl;
++ ssize_t ssz;
++ struct inode *h_idst;
++ struct dentry *h_dst_dentry, *h_src_dentry;
++ struct mnt_idmap *h_dst_idmap, *h_src_idmap;
++ struct posix_acl *acl;
++
++ is_acl = !!is_posix_acl_xattr(name);
++ h_src_idmap = mnt_idmap(h_src->mnt);
++ h_src_dentry = h_src->dentry;
++ if (is_acl) {
++ acl = vfs_get_acl(h_src_idmap, h_src_dentry, name);
++ AuDebugOn(!acl);
++ if (unlikely(IS_ERR(acl))) {
++ err = PTR_ERR(acl);
++ if (err == -ENODATA)
++ err = 0;
++ else if (err == -EOPNOTSUPP
++ && au_test_nfs_noacl(d_inode(h_src_dentry)))
++ err = 0;
++ else if (verbose || au_debug_test())
++ pr_err("%s, err %d\n", name, err);
++ goto out;
++ }
++ } else {
++ ssz = vfs_getxattr_alloc(h_src_idmap, h_src_dentry, name, buf,
++ 0, GFP_NOFS);
++ if (unlikely(ssz <= 0)) {
++ err = ssz;
++ if (err == -ENODATA)
++ err = 0;
++ else if (err == -EOPNOTSUPP
++ && (ignore_flags & au_xattr_out_of_list))
++ err = 0;
++ else if (err && (verbose || au_debug_test()))
++ pr_err("%s, err %d\n", name, err);
++ goto out;
++ }
++ }
+
-+struct dentry *au_diropq_sio(struct dentry *dentry, aufs_bindex_t bindex,
-+ unsigned int flags);
++ /* unlock it temporary */
++ h_dst_idmap = mnt_idmap(h_dst->mnt);
++ h_dst_dentry = h_dst->dentry;
++ h_idst = d_inode(h_dst_dentry);
++ inode_unlock(h_idst);
++ if (is_acl) {
++ err = vfsub_set_acl(h_dst_idmap, h_dst_dentry, name, acl);
++ posix_acl_release(acl);
++ } else
++ err = vfsub_setxattr(h_dst_idmap, h_dst_dentry, name, *buf,
++ ssz, /*flags*/0);
++ inode_lock_nested(h_idst, AuLsc_I_CHILD2);
++ if (unlikely(err)) {
++ if (verbose || au_debug_test())
++ pr_err("%s, err %d\n", name, err);
++ err = au_xattr_ignore(err, name, ignore_flags);
++ }
+
-+struct dentry *au_wh_lkup(struct dentry *h_parent, struct qstr *base_name,
-+ struct au_ndx *ndx);
-+struct dentry *au_wh_create(struct dentry *dentry, aufs_bindex_t bindex,
-+ struct dentry *h_parent, struct au_ndx *ndx);
++out:
++ return err;
++}
+
-+/* real rmdir the whiteout-ed dir */
-+struct au_whtmp_rmdir_args {
-+ struct inode *dir;
-+ aufs_bindex_t bindex;
-+ struct dentry *wh_dentry;
-+ struct au_nhash whlist;
++int au_cpup_xattr(struct path *h_dst, struct path *h_src, int ignore_flags,
++ unsigned int verbose)
++{
++ int err, unlocked;
++ ssize_t ssz;
++ struct dentry *h_dst_dentry, *h_src_dentry;
++ struct inode *h_isrc, *h_idst;
++ char *value, *p, *o, *e;
++
++ /* try stopping to update the source inode while we are referencing */
++ /* there should not be the parent-child relationship between them */
++ h_dst_dentry = h_dst->dentry;
++ h_idst = d_inode(h_dst_dentry);
++ h_src_dentry = h_src->dentry;
++ h_isrc = d_inode(h_src_dentry);
++ inode_unlock(h_idst);
++ inode_lock_shared_nested(h_isrc, AuLsc_I_CHILD);
++ inode_lock_nested(h_idst, AuLsc_I_CHILD2);
++ unlocked = 0;
++
++ /* some filesystems don't list POSIX ACL, for example tmpfs */
++ ssz = vfs_listxattr(h_src_dentry, NULL, 0);
++ err = ssz;
++ if (unlikely(err < 0)) {
++ AuTraceErr(err);
++ if (err == -ENODATA
++ || err == -EOPNOTSUPP)
++ err = 0; /* ignore */
++ goto out;
++ }
++
++ err = 0;
++ p = NULL;
++ o = NULL;
++ if (ssz) {
++ err = -ENOMEM;
++ p = kmalloc(ssz, GFP_NOFS);
++ o = p;
++ if (unlikely(!p))
++ goto out;
++ err = vfs_listxattr(h_src_dentry, p, ssz);
++ }
++ inode_unlock_shared(h_isrc);
++ unlocked = 1;
++ AuDbg("err %d, ssz %zd\n", err, ssz);
++ if (unlikely(err < 0))
++ goto out_free;
++
++ err = 0;
++ e = p + ssz;
++ value = NULL;
++ ignore_flags |= au_xattr_out_of_list;
++ while (!err && p < e) {
++ err = au_do_cpup_xattr(h_dst, h_src, p, &value, ignore_flags,
++ verbose);
++ p += strlen(p) + 1;
++ }
++ au_kfree_try_rcu(value);
++
++out_free:
++ au_kfree_try_rcu(o);
++out:
++ if (!unlocked)
++ inode_unlock_shared(h_isrc);
++ AuTraceErr(err);
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++static int au_smack_reentering(struct super_block *sb)
++{
++#if IS_ENABLED(CONFIG_SECURITY_SMACK) || IS_ENABLED(CONFIG_SECURITY_SELINUX)
++ /*
++ * 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
+};
+
-+struct au_nhash;
-+int au_whtmp_rmdir(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *wh_dentry, struct au_nhash *whlist);
-+void au_whtmp_kick_rmdir(struct inode *dir, aufs_bindex_t bindex,
-+ struct dentry *wh_dentry, struct au_nhash *whlist,
-+ struct au_whtmp_rmdir_args *args);
++struct au_lgxattr {
++ int type;
++ union {
++ struct {
++ char *list;
++ size_t size;
++ } list;
++ struct {
++ const char *name;
++ void *value;
++ size_t size;
++ } get;
++ } u;
++};
++
++static ssize_t au_lgxattr(struct dentry *dentry, struct inode *inode,
++ struct au_lgxattr *arg)
++{
++ ssize_t err;
++ int reenter;
++ struct path h_path;
++ struct super_block *sb;
++
++ sb = dentry->d_sb;
++ 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, inode, /*force*/1, &h_path, reenter);
++ if (unlikely(err))
++ goto out_si;
++ if (unlikely(!h_path.dentry))
++ /* illegally overlapped or something */
++ goto out_di; /* pretending success */
++
++ /* always topmost entry only */
++ switch (arg->type) {
++ case AU_XATTR_LIST:
++ err = vfs_listxattr(h_path.dentry,
++ arg->u.list.list, arg->u.list.size);
++ break;
++ case AU_XATTR_GET:
++ AuDebugOn(d_is_negative(h_path.dentry));
++ err = vfs_getxattr(mnt_idmap(h_path.mnt), h_path.dentry,
++ arg->u.get.name, arg->u.get.value,
++ arg->u.get.size);
++ break;
++ }
++
++out_di:
++ if (!reenter)
++ di_read_unlock(dentry, AuLock_IR);
++out_si:
++ if (!reenter)
++ si_read_unlock(sb);
++out:
++ AuTraceErr(err);
++ return err;
++}
++
++ssize_t aufs_listxattr(struct dentry *dentry, char *list, size_t size)
++{
++ struct au_lgxattr arg = {
++ .type = AU_XATTR_LIST,
++ .u.list = {
++ .list = list,
++ .size = size
++ },
++ };
++
++ return au_lgxattr(dentry, /*inode*/NULL, &arg);
++}
++
++static ssize_t au_getxattr(struct dentry *dentry, struct inode *inode,
++ const char *name, void *value, size_t size)
++{
++ struct au_lgxattr arg = {
++ .type = AU_XATTR_GET,
++ .u.get = {
++ .name = name,
++ .value = value,
++ .size = size
++ },
++ };
++
++ return au_lgxattr(dentry, inode, &arg);
++}
++
++static int au_setxattr(struct dentry *dentry, struct inode *inode,
++ const char *name, const void *value, size_t size,
++ int flags)
++{
++ struct au_sxattr arg = {
++ .type = AU_XATTR_SET,
++ .u.set = {
++ .name = name,
++ .value = value,
++ .size = size,
++ .flags = flags
++ },
++ };
++
++ return au_sxattr(dentry, inode, &arg);
++}
+
+/* ---------------------------------------------------------------------- */
+
-+static inline
-+struct dentry *au_diropq_create(struct dentry *dentry, aufs_bindex_t bindex,
-+ int dlgt)
++static int au_xattr_get(const struct xattr_handler *handler,
++ struct dentry *dentry, struct inode *inode,
++ const char *name, void *buffer, size_t size)
+{
-+ unsigned int flags = AuDiropq_CREATE;
-+ if (dlgt)
-+ au_fset_diropq(flags, DLGT);
-+ return au_diropq_sio(dentry, bindex, flags);
++ return au_getxattr(dentry, inode, name, buffer, size);
+}
+
-+static inline
-+int au_diropq_remove(struct dentry *dentry, aufs_bindex_t bindex, int dlgt)
++static int au_xattr_set(const struct xattr_handler *handler,
++ struct mnt_idmap *idmap,
++ struct dentry *dentry, struct inode *inode,
++ const char *name, const void *value, size_t size,
++ int flags)
+{
-+ unsigned int flags = !AuDiropq_CREATE;
-+ if (dlgt)
-+ au_fset_diropq(flags, DLGT);
-+ return PTR_ERR(au_diropq_sio(dentry, bindex, flags));
++ return au_setxattr(dentry, inode, name, value, size, flags);
+}
+
-+/* ---------------------------------------------------------------------- */
++static const struct xattr_handler au_xattr_handler = {
++ .name = "",
++ .prefix = "",
++ .get = au_xattr_get,
++ .set = au_xattr_set
++};
+
-+#ifdef CONFIG_AUFS_ROBR
-+/* robr.c */
-+int au_test_robr_wh(struct qstr *name, struct dentry *h_parent,
-+ struct qstr *wh_name, int try_sio, struct au_ndx *ndx);
-+int au_test_robr_shwh(struct super_block *sb, const struct qstr *name);
-+#else
-+static inline
-+int au_test_robr_wh(struct qstr *name, struct dentry *h_parent,
-+ struct qstr *wh_name, int try_sio, struct au_ndx *ndx)
-+{
-+ return au_wh_test(h_parent, wh_name, try_sio, ndx);
-+}
++static const struct xattr_handler *au_xattr_handlers[] = {
++ &au_xattr_handler, /* must be last */
++ NULL
++};
+
-+static inline
-+int au_test_robr_shwh(struct super_block *sb, const struct qstr *name)
++void au_xattr_init(struct super_block *sb)
+{
-+ if (unlikely(!au_opt_test(au_mntflags(sb), SHWH)
-+ && !strncmp(name->name, AUFS_WH_PFX, AUFS_WH_PFX_LEN)))
-+ return -EPERM;
-+ return 0;
++ sb->s_xattr = au_xattr_handlers;
+}
-+#endif /* CONFIG_AUFS_ROBR */
-+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_WHOUT_H__ */
-diff -urN linux/fs/aufs/wkq.c linux-aufs/fs/aufs/wkq.c
---- linux/fs/aufs/wkq.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/wkq.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,366 @@
+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 2023-08-28 12:34:39.999969465 +0200
+@@ -0,0 +1,1926 @@
++// SPDX-License-Identifier: GPL-2.0
+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
+/*
-+ * workqueue for asynchronous/super-io/delegated operations
++ * external inode number translation table and bitmap
+ *
-+ * $Id$
++ * things to consider
++ * - the lifetime
++ * + au_xino object
++ * + XINO files (xino, xib, xigen)
++ * + dynamic debugfs entries (xiN)
++ * + static debugfs entries (xib, xigen)
++ * + static sysfs entry (xi_path)
++ * - several entry points to handle them.
++ * + mount(2) without xino option (default)
++ * + mount(2) with xino option
++ * + mount(2) with noxino option
++ * + umount(2)
++ * + remount with add/del branches
++ * + remount with xino/noxino options
+ */
+
-+#include <linux/module.h>
++#include <linux/seq_file.h>
++#include <linux/statfs.h>
+#include "aufs.h"
+
-+struct au_wkq *au_wkq;
-+
-+struct au_cred {
-+#ifdef CONFIG_AUFS_DLGT
-+ int umask;
-+ uid_t fsuid;
-+ gid_t fsgid;
-+ kernel_cap_t cap_effective, cap_inheritable, cap_permitted;
-+#if 0 /* reserved for future use */
-+ unsigned keep_capabilities:1;
-+ struct user_struct *user;
-+ struct fs_struct *fs;
-+ struct nsproxy *nsproxy;
-+#endif
-+#endif
-+};
-+
-+struct au_wkinfo {
-+ struct work_struct wk;
-+ struct super_block *sb;
-+
-+ unsigned int flags;
-+ struct au_cred cred;
-+
-+ au_wkq_func_t func;
-+ void *args;
-+
-+ atomic_t *busyp;
-+ struct completion *comp;
-+};
-+
-+/* ---------------------------------------------------------------------- */
-+
-+#ifdef CONFIG_AUFS_DLGT
-+static void cred_store(struct au_cred *cred)
-+{
-+ cred->umask = current->fs->umask;
-+ cred->fsuid = current->fsuid;
-+ cred->fsgid = current->fsgid;
-+ cred->cap_effective = current->cap_effective;
-+ cred->cap_inheritable = current->cap_inheritable;
-+ cred->cap_permitted = current->cap_permitted;
-+}
-+
-+static void cred_revert(struct au_cred *cred)
++static aufs_bindex_t sbr_find_shared(struct super_block *sb, aufs_bindex_t btop,
++ aufs_bindex_t bbot,
++ struct super_block *h_sb)
+{
-+ AuDebugOn(!au_test_wkq(current));
-+ current->fs->umask = cred->umask;
-+ current->fsuid = cred->fsuid;
-+ current->fsgid = cred->fsgid;
-+ current->cap_effective = cred->cap_effective;
-+ current->cap_inheritable = cred->cap_inheritable;
-+ current->cap_permitted = cred->cap_permitted;
-+}
-+
-+static void cred_switch(struct au_cred *old, struct au_cred *new)
-+{
-+ cred_store(old);
-+ cred_revert(new);
++ /* todo: try binary-search if the branches are many */
++ for (; btop <= bbot; btop++)
++ if (h_sb == au_sbr_sb(sb, btop))
++ return btop;
++ return -1;
+}
+
-+static void dlgt_cred_store(unsigned int flags, struct au_wkinfo *wkinfo)
++/*
++ * find another branch who is on the same filesystem of the specified
++ * branch{@btgt}. search until @bbot.
++ */
++static aufs_bindex_t is_sb_shared(struct super_block *sb, aufs_bindex_t btgt,
++ aufs_bindex_t bbot)
+{
-+ if (au_ftest_wkq(flags, DLGT))
-+ cred_store(&wkinfo->cred);
-+}
++ aufs_bindex_t bindex;
++ struct super_block *tgt_sb;
+
-+static void dlgt_func(struct au_wkinfo *wkinfo)
-+{
-+ if (!au_ftest_wkq(wkinfo->flags, DLGT))
-+ wkinfo->func(wkinfo->args);
-+ else {
-+ struct au_cred cred;
-+ cred_switch(&cred, &wkinfo->cred);
-+ wkinfo->func(wkinfo->args);
-+ cred_revert(&cred);
-+ }
-+}
-+#else
-+static void dlgt_cred_store(unsigned int flags, struct au_wkinfo *wkinfo)
-+{
-+ /* empty */
-+}
++ tgt_sb = au_sbr_sb(sb, btgt);
++ bindex = sbr_find_shared(sb, /*btop*/0, btgt - 1, tgt_sb);
++ if (bindex < 0)
++ bindex = sbr_find_shared(sb, btgt + 1, bbot, tgt_sb);
+
-+static void dlgt_func(struct au_wkinfo *wkinfo)
-+{
-+ wkinfo->func(wkinfo->args);
++ return bindex;
+}
-+#endif /* CONFIG_AUFS_DLGT */
+
+/* ---------------------------------------------------------------------- */
+
-+static void update_busy(struct au_wkq *wkq, struct au_wkinfo *wkinfo)
++/*
++ * stop unnecessary notify events at creating xino files
++ */
++
++aufs_bindex_t au_xi_root(struct super_block *sb, struct dentry *dentry)
+{
-+#ifdef CONFIG_AUFS_STAT
-+ unsigned int new, old;
++ aufs_bindex_t bfound, bindex, bbot;
++ struct dentry *parent;
++ struct au_branch *br;
+
-+ do {
-+ new = atomic_read(wkinfo->busyp);
-+ old = wkq->max_busy;
-+ if (new <= old)
++ bfound = -1;
++ parent = dentry->d_parent; /* safe d_parent access */
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ br = au_sbr(sb, bindex);
++ if (au_br_dentry(br) == parent) {
++ bfound = bindex;
+ break;
-+ } while (cmpxchg(&wkq->max_busy, old, new) == old);
-+#endif
-+}
-+
-+static int enqueue(struct au_wkq *wkq, struct au_wkinfo *wkinfo)
-+{
-+ AuTraceEnter();
++ }
++ }
+
-+ wkinfo->busyp = &wkq->busy;
-+ update_busy(wkq, wkinfo);
-+ if (au_ftest_wkq(wkinfo->flags, WAIT))
-+ return !queue_work(wkq->q, &wkinfo->wk);
-+ else
-+ return !schedule_work(&wkinfo->wk);
++ AuDbg("bfound b%d\n", bfound);
++ return bfound;
+}
+
-+static void do_wkq(struct au_wkinfo *wkinfo)
-+{
-+ unsigned int idle, n;
-+ int i, idle_idx;
-+
-+ AuTraceEnter();
-+
-+ while (1) {
-+ if (au_ftest_wkq(wkinfo->flags, WAIT)) {
-+ idle_idx = 0;
-+ idle = UINT_MAX;
-+ for (i = 0; i < aufs_nwkq; i++) {
-+ n = atomic_inc_return(&au_wkq[i].busy);
-+ if (n == 1 && !enqueue(au_wkq + i, wkinfo))
-+ return; /* success */
-+
-+ if (n < idle) {
-+ idle_idx = i;
-+ idle = n;
-+ }
-+ atomic_dec_return(&au_wkq[i].busy);
-+ }
-+ } else
-+ idle_idx = aufs_nwkq;
-+
-+ atomic_inc_return(&au_wkq[idle_idx].busy);
-+ if (!enqueue(au_wkq + idle_idx, wkinfo))
-+ return; /* success */
-+
-+ /* impossible? */
-+ AuWarn1("failed to queue_work()\n");
-+ yield();
-+ }
-+}
++struct au_xino_lock_dir {
++ struct au_hinode *hdir;
++ struct dentry *parent;
++ struct inode *dir;
++};
+
-+static void wkq_func(struct work_struct *wk)
++static struct dentry *au_dget_parent_lock(struct dentry *dentry,
++ unsigned int lsc)
+{
-+ struct au_wkinfo *wkinfo = container_of(wk, struct au_wkinfo, wk);
-+
-+ LKTRTrace("wkinfo{0x%x, %p, %p, %p}\n",
-+ wkinfo->flags, wkinfo->func, wkinfo->busyp, wkinfo->comp);
++ struct dentry *parent;
++ struct inode *dir;
+
-+ dlgt_func(wkinfo);
-+ atomic_dec_return(wkinfo->busyp);
-+ if (au_ftest_wkq(wkinfo->flags, WAIT))
-+ complete(wkinfo->comp);
-+ else {
-+ kobject_put(&au_sbi(wkinfo->sb)->si_kobj);
-+ module_put(THIS_MODULE);
-+ kfree(wkinfo);
++ parent = dget_parent(dentry);
++ dir = d_inode(parent);
++ inode_lock_nested(dir, lsc);
++#if 0 /* it should not happen */
++ spin_lock(&dentry->d_lock);
++ if (unlikely(dentry->d_parent != parent)) {
++ spin_unlock(&dentry->d_lock);
++ inode_unlock(dir);
++ dput(parent);
++ parent = NULL;
++ goto out;
+ }
-+}
-+
-+#if defined(CONFIG_4KSTACKS) || defined(Test4KSTACKS)
-+#define AuWkqCompDeclare(name) struct completion *comp = NULL
++ spin_unlock(&dentry->d_lock);
+
-+static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp)
-+{
-+ *comp = kmalloc(sizeof(**comp), GFP_NOFS);
-+ if (*comp) {
-+ init_completion(*comp);
-+ wkinfo->comp = *comp;
-+ return 0;
-+ }
-+ return -ENOMEM;
++out:
++#endif
++ return parent;
+}
+
-+static void au_wkq_comp_free(struct completion *comp)
++static void au_xino_lock_dir(struct super_block *sb, struct path *xipath,
++ struct au_xino_lock_dir *ldir)
+{
-+ kfree(comp);
-+}
-+
-+#else
-+
-+/* no braces */
-+#define AuWkqCompDeclare(name) \
-+ DECLARE_COMPLETION_ONSTACK(_ ## name); \
-+ struct completion *comp = &_ ## name
++ aufs_bindex_t bindex;
+
-+static int au_wkq_comp_alloc(struct au_wkinfo *wkinfo, struct completion **comp)
-+{
-+ wkinfo->comp = *comp;
-+ return 0;
++ ldir->hdir = NULL;
++ bindex = au_xi_root(sb, xipath->dentry);
++ if (bindex >= 0) {
++ /* rw branch root */
++ ldir->hdir = au_hi(d_inode(sb->s_root), bindex);
++ au_hn_inode_lock_nested(ldir->hdir, AuLsc_I_PARENT);
++ } else {
++ /* other */
++ ldir->parent = au_dget_parent_lock(xipath->dentry,
++ AuLsc_I_PARENT);
++ ldir->dir = d_inode(ldir->parent);
++ }
+}
+
-+static void au_wkq_comp_free(struct completion *comp)
++static void au_xino_unlock_dir(struct au_xino_lock_dir *ldir)
+{
-+ /* empty */
++ if (ldir->hdir)
++ au_hn_inode_unlock(ldir->hdir);
++ else {
++ inode_unlock(ldir->dir);
++ dput(ldir->parent);
++ }
+}
-+#endif /* 4KSTACKS */
-+
-+static void au_wkq_run(struct au_wkinfo *wkinfo)
-+{
-+#if 1 /* tmp debug */
-+ if (au_test_wkq(current))
-+ au_dbg_blocked();
-+#endif
-+ AuDebugOn(au_test_wkq(current));
+
-+ INIT_WORK(&wkinfo->wk, wkq_func);
-+ dlgt_cred_store(wkinfo->flags, wkinfo);
-+ do_wkq(wkinfo);
-+}
++/* ---------------------------------------------------------------------- */
+
-+int au_wkq_wait(au_wkq_func_t func, void *args, int dlgt)
++/*
++ * create and set a new xino file
++ */
++struct file *au_xino_create(struct super_block *sb, char *fpath, int silent,
++ int wbrtop)
+{
++ struct file *file;
++ struct dentry *h_parent, *d;
++ struct inode *h_dir, *inode;
+ int err;
-+ AuWkqCompDeclare(comp);
-+ struct au_wkinfo wkinfo = {
-+ .flags = AuWkq_WAIT,
-+ .func = func,
-+ .args = args
-+ };
++ static DEFINE_MUTEX(mtx);
+
-+ LKTRTrace("dlgt %d\n", dlgt);
++ /*
++ * at mount-time, and the xino file is the default path,
++ * hnotify is disabled so we have no notify events to ignore.
++ * when a user specified the xino, we cannot get au_hdir to be ignored.
++ */
++ if (!wbrtop)
++ mutex_lock(&mtx);
++ file = vfsub_filp_open(fpath, O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE
++ /* | __FMODE_NONOTIFY */,
++ 0666);
++ if (IS_ERR(file)) {
++ if (!wbrtop)
++ mutex_unlock(&mtx);
++ if (!silent)
++ pr_err("open %s(%ld)\n", fpath, PTR_ERR(file));
++ return file;
++ }
+
-+ err = au_wkq_comp_alloc(&wkinfo, &comp);
-+ if (!err) {
-+ if (dlgt)
-+ au_fset_wkq(wkinfo.flags, DLGT);
-+ au_wkq_run(&wkinfo);
-+ /* no timeout, no interrupt */
-+ wait_for_completion(wkinfo.comp);
-+ au_wkq_comp_free(comp);
++ /* keep file count */
++ err = 0;
++ d = file->f_path.dentry;
++ h_parent = au_dget_parent_lock(d, AuLsc_I_PARENT);
++ if (!wbrtop)
++ mutex_unlock(&mtx);
++ /* mnt_want_write() is unnecessary here */
++ h_dir = d_inode(h_parent);
++ inode = file_inode(file);
++ /* no delegation since it is just created */
++ if (inode->i_nlink)
++ err = vfsub_unlink(h_dir, &file->f_path, /*delegated*/NULL,
++ /*force*/0);
++ inode_unlock(h_dir);
++ dput(h_parent);
++ if (unlikely(err)) {
++ if (!silent)
++ pr_err("unlink %s(%d)\n", fpath, err);
++ goto out;
+ }
+
-+ AuTraceErr(err);
-+ return err;
++ err = -EINVAL;
++ if (unlikely(sb && sb == d->d_sb)) {
++ if (!silent)
++ pr_err("%s must be outside\n", fpath);
++ goto out;
++ }
++ if (unlikely(au_test_fs_bad_xino(d->d_sb))) {
++ if (!silent)
++ pr_err("xino doesn't support %s(%s)\n",
++ fpath, au_sbtype(d->d_sb));
++ goto out;
++ }
++ return file; /* success */
+
++out:
++ fput(file);
++ file = ERR_PTR(err);
++ return file;
+}
+
-+int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb,
-+ int dlgt)
++/*
++ * create a new xinofile at the same place/path as @base.
++ */
++struct file *au_xino_create2(struct super_block *sb, struct path *base,
++ struct file *copy_src)
+{
-+ int err;
-+ struct au_wkinfo *wkinfo;
++ struct file *file;
++ struct dentry *dentry;
++ struct inode *dir, *delegated;
++ struct qstr *name;
++ struct path ppath, path;
++ int err, do_unlock;
++ struct au_xino_lock_dir ldir;
++
++ do_unlock = 1;
++ au_xino_lock_dir(sb, base, &ldir);
++ dentry = base->dentry;
++ ppath.dentry = dentry->d_parent; /* dir inode is locked */
++ ppath.mnt = base->mnt;
++ dir = d_inode(ppath.dentry);
++ IMustLock(dir);
+
-+ LKTRTrace("dlgt %d\n", dlgt);
-+ AuDebugOn(!sb);
++ name = &dentry->d_name;
++ path.dentry = vfsub_lookup_one_len(name->name, &ppath, name->len);
++ if (IS_ERR(path.dentry)) {
++ file = (void *)path.dentry;
++ pr_err("%pd lookup err %ld\n", dentry, PTR_ERR(path.dentry));
++ goto out;
++ }
+
-+ atomic_inc_return(&au_sbi(sb)->si_nowait.nw_len);
++ /* no need to mnt_want_write() since we call dentry_open() later */
++ err = vfs_create(mnt_idmap(base->mnt), dir, path.dentry, 0666, NULL);
++ if (unlikely(err)) {
++ file = ERR_PTR(err);
++ pr_err("%pd create err %d\n", dentry, err);
++ goto out_dput;
++ }
+
-+ /*
-+ * wkq_func() must free this wkinfo.
-+ * it highly depends upon the implementation of workqueue.
-+ */
-+ err = 0;
-+ wkinfo = kmalloc(sizeof(*wkinfo), GFP_NOFS);
-+ if (wkinfo) {
-+ wkinfo->sb = sb;
-+ wkinfo->flags = !AuWkq_WAIT;
-+ wkinfo->func = func;
-+ wkinfo->args = args;
-+ wkinfo->comp = NULL;
-+ if (dlgt)
-+ au_fset_wkq(wkinfo->flags, DLGT);
-+ kobject_get(&au_sbi(sb)->si_kobj);
-+ __module_get(THIS_MODULE);
++ path.mnt = base->mnt;
++ file = vfsub_dentry_open(&path,
++ O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE
++ /* | __FMODE_NONOTIFY */);
++ if (IS_ERR(file)) {
++ pr_err("%pd open err %ld\n", dentry, PTR_ERR(file));
++ goto out_dput;
++ }
+
-+ au_wkq_run(wkinfo);
-+ } else {
-+ err = -ENOMEM;
-+ atomic_dec_return(&au_sbi(sb)->si_nowait.nw_len);
++ delegated = NULL;
++ err = vfsub_unlink(dir, &file->f_path, &delegated, /*force*/0);
++ au_xino_unlock_dir(&ldir);
++ do_unlock = 0;
++ if (unlikely(err == -EWOULDBLOCK)) {
++ pr_warn("cannot retry for NFSv4 delegation"
++ " for an internal unlink\n");
++ iput(delegated);
++ }
++ if (unlikely(err)) {
++ pr_err("%pd unlink err %d\n", dentry, err);
++ goto out_fput;
+ }
+
-+ AuTraceErr(err);
-+ return err;
-+}
++ if (copy_src) {
++ /* no one can touch copy_src xino */
++ err = au_copy_file(file, copy_src, vfsub_f_size_read(copy_src));
++ if (unlikely(err)) {
++ pr_err("%pd copy err %d\n", dentry, err);
++ goto out_fput;
++ }
++ }
++ goto out_dput; /* success */
+
-+/* ---------------------------------------------------------------------- */
++out_fput:
++ fput(file);
++ file = ERR_PTR(err);
++out_dput:
++ dput(path.dentry);
++out:
++ if (do_unlock)
++ au_xino_unlock_dir(&ldir);
++ return file;
++}
+
-+void au_wkq_fin(void)
++struct file *au_xino_file1(struct au_xino *xi)
+{
-+ int i;
++ struct file *file;
++ unsigned int u, nfile;
+
-+ AuTraceEnter();
++ file = NULL;
++ nfile = xi->xi_nfile;
++ for (u = 0; u < nfile; u++) {
++ file = xi->xi_file[u];
++ if (file)
++ break;
++ }
+
-+ for (i = 0; i < aufs_nwkq; i++)
-+ if (au_wkq[i].q && !IS_ERR(au_wkq[i].q))
-+ destroy_workqueue(au_wkq[i].q);
-+ kfree(au_wkq);
++ return file;
+}
+
-+int __init au_wkq_init(void)
++static int au_xino_file_set(struct au_xino *xi, int idx, struct file *file)
+{
-+ int err, i;
-+ struct au_wkq *nowaitq;
-+
-+ LKTRTrace("%d\n", aufs_nwkq);
++ int err;
++ struct file *f;
++ void *p;
+
-+ /* '+1' is for accounting of nowait queue */
-+ err = -ENOMEM;
-+ au_wkq = kcalloc(aufs_nwkq + 1, sizeof(*au_wkq), GFP_NOFS);
-+ if (unlikely(!au_wkq))
-+ goto out;
++ if (file)
++ get_file(file);
+
+ err = 0;
-+ for (i = 0; i < aufs_nwkq; i++) {
-+ au_wkq[i].q = create_singlethread_workqueue(AUFS_WKQ_NAME);
-+ if (au_wkq[i].q && !IS_ERR(au_wkq[i].q)) {
-+ atomic_set(&au_wkq[i].busy, 0);
-+ au_wkq_max_busy_init(au_wkq + i);
-+ continue;
++ f = NULL;
++ if (idx < xi->xi_nfile) {
++ f = xi->xi_file[idx];
++ if (f)
++ fput(f);
++ } else {
++ p = au_kzrealloc(xi->xi_file,
++ sizeof(*xi->xi_file) * xi->xi_nfile,
++ sizeof(*xi->xi_file) * (idx + 1),
++ GFP_NOFS, /*may_shrink*/0);
++ if (p) {
++ MtxMustLock(&xi->xi_mtx);
++ xi->xi_file = p;
++ xi->xi_nfile = idx + 1;
++ } else {
++ err = -ENOMEM;
++ if (file)
++ fput(file);
++ goto out;
+ }
-+
-+ err = PTR_ERR(au_wkq[i].q);
-+ au_wkq_fin();
-+ break;
+ }
++ xi->xi_file[idx] = file;
+
-+ /* nowait accounting */
-+ nowaitq = au_wkq + aufs_nwkq;
-+ atomic_set(&nowaitq->busy, 0);
-+ au_wkq_max_busy_init(nowaitq);
-+ nowaitq->q = NULL;
-+ /* smp_mb(); */ /* atomic_set */
-+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
-diff -urN linux/fs/aufs/wkq.h linux-aufs/fs/aufs/wkq.h
---- linux/fs/aufs/wkq.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/wkq.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,128 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
+
+/*
-+ * workqueue for asynchronous/super-io/delegated operations
-+ *
-+ * $Id$
++ * if @xinew->xi is not set, then create new xigen file.
+ */
++struct file *au_xi_new(struct super_block *sb, struct au_xi_new *xinew)
++{
++ struct file *file;
++ int err;
+
-+#ifndef __AUFS_WKQ_H__
-+#define __AUFS_WKQ_H__
-+
-+#ifdef __KERNEL__
-+
-+#include <linux/fs.h>
-+#include <linux/sched.h>
-+#include <linux/workqueue.h>
-+#include <linux/aufs_type.h>
++ SiMustAnyLock(sb);
+
-+/* ---------------------------------------------------------------------- */
++ file = au_xino_create2(sb, xinew->base, xinew->copy_src);
++ if (IS_ERR(file)) {
++ err = PTR_ERR(file);
++ pr_err("%s[%d], err %d\n",
++ xinew->xi ? "xino" : "xigen",
++ xinew->idx, err);
++ goto out;
++ }
+
-+/* internal workqueue named AUFS_WKQ_NAME */
-+struct au_wkq {
-+ struct workqueue_struct *q;
++ if (xinew->xi)
++ err = au_xino_file_set(xinew->xi, xinew->idx, file);
++ else {
++ BUG();
++ /* todo: make xigen file an array */
++ /* err = au_xigen_file_set(sb, xinew->idx, file); */
++ }
++ fput(file);
++ if (unlikely(err))
++ file = ERR_PTR(err);
+
-+ /* balancing */
-+ atomic_t busy;
++out:
++ return file;
++}
+
-+ /* accounting */
-+#ifdef CONFIG_AUFS_STAT
-+ unsigned int max_busy;
-+#endif
-+};
++/* ---------------------------------------------------------------------- */
+
+/*
-+ * in the next operation, wait for the 'nowait' tasks in system-wide workqueue
++ * truncate xino files
+ */
-+struct au_nowait_tasks {
-+ atomic_t nw_len;
-+ wait_queue_head_t nw_wq;
-+};
++static int au_xino_do_trunc(struct super_block *sb, aufs_bindex_t bindex,
++ int idx, struct kstatfs *st)
++{
++ int err;
++ blkcnt_t blocks;
++ struct file *file, *new_xino;
++ struct au_xi_new xinew = {
++ .idx = idx
++ };
+
-+/* ---------------------------------------------------------------------- */
++ err = 0;
++ xinew.xi = au_sbr(sb, bindex)->br_xino;
++ file = au_xino_file(xinew.xi, idx);
++ if (!file)
++ goto out;
+
-+extern struct au_wkq *au_wkq;
-+typedef void (*au_wkq_func_t)(void *args);
++ xinew.base = &file->f_path;
++ err = vfs_statfs(xinew.base, st);
++ if (unlikely(err)) {
++ AuErr1("statfs err %d, ignored\n", err);
++ err = 0;
++ goto out;
++ }
+
-+/* wkq flags */
-+#define AuWkq_WAIT 1
-+#define AuWkq_DLGT (1 << 1)
-+#define au_ftest_wkq(flags, name) ((flags) & AuWkq_##name)
-+#define au_fset_wkq(flags, name) { (flags) |= AuWkq_##name; }
-+#define au_fclr_wkq(flags, name) { (flags) &= ~AuWkq_##name; }
-+#ifndef CONFIG_AUFS_DLGT
-+#undef AuWkq_DLGT
-+#define AuWkq_DLGT 0
-+#endif
++ blocks = file_inode(file)->i_blocks;
++ pr_info("begin truncating xino(b%d-%d), ib%llu, %llu/%llu free blks\n",
++ bindex, idx, (u64)blocks, st->f_bfree, st->f_blocks);
+
-+int au_wkq_wait(au_wkq_func_t func, void *args, int dlgt);
-+int au_wkq_nowait(au_wkq_func_t func, void *args, struct super_block *sb,
-+ int dlgt);
-+int __init au_wkq_init(void);
-+void au_wkq_fin(void);
++ xinew.copy_src = file;
++ new_xino = au_xi_new(sb, &xinew);
++ if (IS_ERR(new_xino)) {
++ err = PTR_ERR(new_xino);
++ pr_err("xino(b%d-%d), err %d, ignored\n", bindex, idx, err);
++ goto out;
++ }
+
-+/* ---------------------------------------------------------------------- */
++ err = vfs_statfs(&new_xino->f_path, st);
++ if (!err)
++ pr_info("end truncating xino(b%d-%d), ib%llu, %llu/%llu free blks\n",
++ bindex, idx, (u64)file_inode(new_xino)->i_blocks,
++ st->f_bfree, st->f_blocks);
++ else {
++ AuErr1("statfs err %d, ignored\n", err);
++ err = 0;
++ }
+
-+static inline int au_test_nowait_wkq(struct task_struct *tsk)
-+{
-+ static const char *p = "events";
-+ return !tsk->mm && !strncmp(tsk->comm, p, strlen(p));
++out:
++ return err;
+}
+
-+static inline int au_test_wkq(struct task_struct *tsk)
++int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex, int idx_begin)
+{
-+ return !tsk->mm && !strcmp(tsk->comm, AUFS_WKQ_NAME);
-+#if 0 /* reserved for future use, per-cpu workqueue */
-+ return !tsk->mm
-+ && !memcmp(tsk->comm, AUFS_WKQ_NAME "/",
-+ sizeof(AUFS_WKQ_NAME));
-+#endif
-+}
++ int err, i;
++ unsigned long jiffy;
++ aufs_bindex_t bbot;
++ struct kstatfs *st;
++ struct au_branch *br;
++ struct au_xino *xi;
+
-+static inline void au_wkq_max_busy_init(struct au_wkq *wkq)
-+{
-+#ifdef CONFIG_AUFS_STAT
-+ wkq->max_busy = 0;
-+#endif
-+}
++ err = -ENOMEM;
++ st = kmalloc(sizeof(*st), GFP_NOFS);
++ if (unlikely(!st))
++ goto out;
+
-+/* todo: memory barrier? */
-+static inline void au_nwt_init(struct au_nowait_tasks *nwt)
-+{
-+ atomic_set(&nwt->nw_len, 0);
-+ smp_mb(); /* atomic_set */
-+ init_waitqueue_head(&nwt->nw_wq);
-+}
++ err = -EINVAL;
++ bbot = au_sbbot(sb);
++ if (unlikely(bindex < 0 || bbot < bindex))
++ goto out_st;
+
-+static inline void au_nwt_done(struct au_nowait_tasks *nwt)
-+{
-+ AuTraceEnter();
++ err = 0;
++ jiffy = jiffies;
++ br = au_sbr(sb, bindex);
++ xi = br->br_xino;
++ for (i = idx_begin; !err && i < xi->xi_nfile; i++)
++ err = au_xino_do_trunc(sb, bindex, i, st);
++ if (!err)
++ au_sbi(sb)->si_xino_jiffy = jiffy;
+
-+ if (!atomic_dec_return(&nwt->nw_len))
-+ wake_up_all(&nwt->nw_wq);
++out_st:
++ au_kfree_rcu(st);
++out:
++ return err;
+}
+
-+static inline int au_nwt_flush(struct au_nowait_tasks *nwt)
++struct xino_do_trunc_args {
++ struct super_block *sb;
++ struct au_branch *br;
++ int idx;
++};
++
++static void xino_do_trunc(void *_args)
+{
-+ wait_event(nwt->nw_wq, !atomic_read(&nwt->nw_len));
-+ return 0;
-+}
++ struct xino_do_trunc_args *args = _args;
++ struct super_block *sb;
++ struct au_branch *br;
++ struct inode *dir;
++ int err, idx;
++ aufs_bindex_t bindex;
+
-+#endif /* __KERNEL__ */
-+#endif /* __AUFS_WKQ_H__ */
-diff -urN linux/fs/aufs/xino.c linux-aufs/fs/aufs/xino.c
---- linux/fs/aufs/xino.c 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/fs/aufs/xino.c 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,1268 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro 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, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
-+ */
++ err = 0;
++ sb = args->sb;
++ dir = d_inode(sb->s_root);
++ br = args->br;
++ idx = args->idx;
++
++ si_noflush_write_lock(sb);
++ ii_read_lock_parent(dir);
++ bindex = au_br_index(sb, br->br_id);
++ err = au_xino_trunc(sb, bindex, idx);
++ ii_read_unlock(dir);
++ if (unlikely(err))
++ pr_warn("err b%d, (%d)\n", bindex, err);
++ atomic_dec(&br->br_xino->xi_truncating);
++ au_lcnt_dec(&br->br_count);
++ si_write_unlock(sb);
++ au_nwt_done(&au_sbi(sb)->si_nowait);
++ au_kfree_rcu(args);
++}
+
+/*
-+ * external inode number translation table and bitmap
-+ *
-+ * $Id$
++ * returns the index in the xi_file array whose corresponding file is necessary
++ * to truncate, or -1 which means no need to truncate.
+ */
-+
-+#include <linux/uaccess.h>
-+#include "aufs.h"
-+
-+/* ---------------------------------------------------------------------- */
-+
-+ssize_t xino_fread(au_readf_t func, struct file *file, void *buf, size_t size,
-+ loff_t *pos)
++static int xino_trunc_test(struct super_block *sb, struct au_branch *br)
+{
-+ ssize_t err;
-+ mm_segment_t oldfs;
++ int err;
++ unsigned int u;
++ struct kstatfs st;
++ struct au_sbinfo *sbinfo;
++ struct au_xino *xi;
++ struct file *file;
+
-+ LKTRTrace("%.*s, sz %zu, *pos %lld\n",
-+ AuDLNPair(file->f_dentry), size, *pos);
++ /* todo: si_xino_expire and the ratio should be customizable */
++ sbinfo = au_sbi(sb);
++ if (time_before(jiffies,
++ sbinfo->si_xino_jiffy + sbinfo->si_xino_expire))
++ return -1;
+
-+ oldfs = get_fs();
-+ set_fs(KERNEL_DS);
-+ do {
-+ /* todo: signal_pending? */
-+ err = func(file, (char __user *)buf, size, pos);
-+ } while (err == -EAGAIN || err == -EINTR);
-+ set_fs(oldfs);
++ /* truncation border */
++ xi = br->br_xino;
++ for (u = 0; u < xi->xi_nfile; u++) {
++ file = au_xino_file(xi, u);
++ if (!file)
++ continue;
+
-+#if 0 /* reserved for future use */
-+ if (err > 0)
-+ fsnotify_access(file->f_dentry);
-+#endif
++ err = vfs_statfs(&file->f_path, &st);
++ if (unlikely(err)) {
++ AuErr1("statfs err %d, ignored\n", err);
++ return -1;
++ }
++ if (div64_u64(st.f_bfree * 100, st.f_blocks)
++ >= AUFS_XINO_DEF_TRUNC)
++ return u;
++ }
+
-+ AuTraceErr(err);
-+ return err;
++ return -1;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static ssize_t do_xino_fwrite(au_writef_t func, struct file *file, void *buf,
-+ size_t size, loff_t *pos)
++static void xino_try_trunc(struct super_block *sb, struct au_branch *br)
+{
-+ ssize_t err;
-+ mm_segment_t oldfs;
++ int idx;
++ struct xino_do_trunc_args *args;
++ int wkq_err;
+
-+ lockdep_off();
-+ oldfs = get_fs();
-+ set_fs(KERNEL_DS);
-+ do {
-+ /* todo: signal_pending? */
-+ err = func(file, (const char __user *)buf, size, pos);
-+ } while (err == -EAGAIN || err == -EINTR);
-+ set_fs(oldfs);
-+ lockdep_on();
++ idx = xino_trunc_test(sb, br);
++ if (idx < 0)
++ return;
+
-+ if (err >= 0)
-+ au_update_fuse_h_inode(file->f_vfsmnt, file->f_dentry);
-+ /*ignore*/
++ if (atomic_inc_return(&br->br_xino->xi_truncating) > 1)
++ goto out;
+
-+#if 0 /* reserved for future use */
-+ if (err > 0)
-+ fsnotify_modify(file->f_dentry);
-+#endif
++ /* lock and kfree() will be called in trunc_xino() */
++ args = kmalloc(sizeof(*args), GFP_NOFS);
++ if (unlikely(!args)) {
++ AuErr1("no memory\n");
++ goto out;
++ }
+
-+ AuTraceErr(err);
-+ return err;
++ au_lcnt_inc(&br->br_count);
++ args->sb = sb;
++ args->br = br;
++ args->idx = idx;
++ wkq_err = au_wkq_nowait(xino_do_trunc, args, sb, /*flags*/0);
++ if (!wkq_err)
++ return; /* success */
++
++ pr_err("wkq %d\n", wkq_err);
++ au_lcnt_dec(&br->br_count);
++ au_kfree_rcu(args);
++
++out:
++ atomic_dec(&br->br_xino->xi_truncating);
+}
+
-+struct do_xino_fwrite_args {
-+ ssize_t *errp;
-+ au_writef_t func;
-+ struct file *file;
-+ void *buf;
-+ size_t size;
-+ loff_t *pos;
++/* ---------------------------------------------------------------------- */
++
++struct au_xi_calc {
++ int idx;
++ loff_t pos;
+};
+
-+static void call_do_xino_fwrite(void *args)
++static void au_xi_calc(struct super_block *sb, ino_t h_ino,
++ struct au_xi_calc *calc)
+{
-+ struct do_xino_fwrite_args *a = args;
-+ *a->errp = do_xino_fwrite(a->func, a->file, a->buf, a->size, a->pos);
++ loff_t maxent;
++
++ maxent = au_xi_maxent(sb);
++ calc->idx = div64_u64_rem(h_ino, maxent, &calc->pos);
++ calc->pos *= sizeof(ino_t);
+}
+
-+ssize_t xino_fwrite(au_writef_t func, struct file *file, void *buf, size_t size,
-+ loff_t *pos)
++static int au_xino_do_new_async(struct super_block *sb, struct au_branch *br,
++ struct au_xi_calc *calc)
+{
-+ ssize_t err;
++ int err;
++ struct file *file;
++ struct au_xino *xi = br->br_xino;
++ struct au_xi_new xinew = {
++ .xi = xi
++ };
+
-+ LKTRTrace("%.*s, sz %zu, *pos %lld\n",
-+ AuDLNPair(file->f_dentry), size, *pos);
++ SiMustAnyLock(sb);
+
-+ /* todo: signal block and no wkq? */
-+ /*
-+ * it breaks RLIMIT_FSIZE and normal user's limit,
-+ * users should care about quota and real 'filesystem full.'
-+ */
-+ if (!au_test_wkq(current)) {
-+ int wkq_err;
-+ struct do_xino_fwrite_args args = {
-+ .errp = &err,
-+ .func = func,
-+ .file = file,
-+ .buf = buf,
-+ .size = size,
-+ .pos = pos
-+ };
-+ wkq_err = au_wkq_wait(call_do_xino_fwrite, &args, /*dlgt*/0);
-+ if (unlikely(wkq_err))
-+ err = wkq_err;
-+ } else
-+ err = do_xino_fwrite(func, file, buf, size, pos);
++ err = 0;
++ if (!xi)
++ goto out;
+
-+ AuTraceErr(err);
++ mutex_lock(&xi->xi_mtx);
++ file = au_xino_file(xi, calc->idx);
++ if (file)
++ goto out_mtx;
++
++ file = au_xino_file(xi, /*idx*/-1);
++ AuDebugOn(!file);
++ xinew.idx = calc->idx;
++ xinew.base = &file->f_path;
++ /* xinew.copy_src = NULL; */
++ file = au_xi_new(sb, &xinew);
++ if (IS_ERR(file))
++ err = PTR_ERR(file);
++
++out_mtx:
++ mutex_unlock(&xi->xi_mtx);
++out:
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+struct xino_do_trunc_args {
++struct au_xino_do_new_async_args {
+ struct super_block *sb;
+ struct au_branch *br;
++ struct au_xi_calc calc;
++ ino_t ino;
++};
++
++struct au_xi_writing {
++ struct hlist_bl_node node;
++ ino_t h_ino, ino;
+};
+
-+static void xino_do_trunc(void *_args)
++static int au_xino_do_write(struct file *file, struct au_xi_calc *calc,
++ ino_t ino);
++
++static void au_xino_call_do_new_async(void *args)
+{
-+ struct xino_do_trunc_args *args = _args;
++ struct au_xino_do_new_async_args *a = args;
++ struct au_branch *br;
+ struct super_block *sb;
-+ aufs_bindex_t bindex;
-+ int err;
++ struct au_sbinfo *sbi;
++ struct inode *root;
+ struct file *file;
-+ struct inode *dir;
-+ struct au_sbinfo *sbinfo;
++ struct au_xi_writing *del, *p;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ int err;
+
-+ err = 0;
-+ sb = args->sb;
-+ dir = sb->s_root->d_inode;
-+ si_noflush_write_lock(sb);
-+ ii_read_lock_parent(dir);
-+ bindex = au_br_index(sb, args->br->br_id);
-+ AuDebugOn(bindex < 0);
-+ err = au_xino_trunc(sb, bindex);
-+ if (unlikely(err))
++ br = a->br;
++ sb = a->sb;
++ sbi = au_sbi(sb);
++ si_noflush_read_lock(sb);
++ root = d_inode(sb->s_root);
++ ii_read_lock_child(root);
++ err = au_xino_do_new_async(sb, br, &a->calc);
++ if (unlikely(err)) {
++ AuIOErr("err %d\n", err);
+ goto out;
++ }
+
-+ file = args->br->br_xino.xi_file;
-+ au_update_fuse_h_inode(args->br->br_mnt, file->f_dentry); /*ignore*/
-+ if (file->f_dentry->d_inode->i_blocks >= args->br->br_xino_upper)
-+ args->br->br_xino_upper += AUFS_XINO_TRUNC_STEP;
++ file = au_xino_file(br->br_xino, a->calc.idx);
++ AuDebugOn(!file);
++ err = au_xino_do_write(file, &a->calc, a->ino);
++ if (unlikely(err)) {
++ AuIOErr("err %d\n", err);
++ goto out;
++ }
+
-+ out:
-+ ii_read_unlock(dir);
-+ if (unlikely(err))
-+ AuWarn("err b%d, (%d)\n", bindex, err);
-+ atomic_dec_return(&args->br->br_xino_running);
-+ au_br_put(args->br);
-+ sbinfo = au_sbi(sb);
-+ au_nwt_done(&sbinfo->si_nowait);
-+ si_write_unlock(sb);
-+ kfree(args);
++ del = NULL;
++ hbl = &br->br_xino->xi_writing;
++ hlist_bl_lock(hbl);
++ au_hbl_for_each(pos, hbl) {
++ p = container_of(pos, struct au_xi_writing, node);
++ if (p->ino == a->ino) {
++ del = p;
++ hlist_bl_del(&p->node);
++ break;
++ }
++ }
++ hlist_bl_unlock(hbl);
++ au_kfree_rcu(del);
++
++out:
++ au_lcnt_dec(&br->br_count);
++ ii_read_unlock(root);
++ si_read_unlock(sb);
++ au_nwt_done(&sbi->si_nowait);
++ au_kfree_rcu(a);
+}
+
-+static void xino_try_trunc(struct super_block *sb, struct au_branch *br)
++/*
++ * create a new xino file asynchronously
++ */
++static int au_xino_new_async(struct super_block *sb, struct au_branch *br,
++ struct au_xi_calc *calc, ino_t ino)
+{
-+ struct xino_do_trunc_args *args;
-+ struct au_sbinfo *sbinfo;
-+ struct file *file = br->br_xino.xi_file;
-+ int wkq_err;
++ int err;
++ struct au_xino_do_new_async_args *arg;
+
-+ au_update_fuse_h_inode(br->br_mnt, file->f_dentry); /*ignore*/
-+ if (file->f_dentry->d_inode->i_blocks < br->br_xino_upper)
-+ return;
-+ if (atomic_inc_return(&br->br_xino_running) > 1)
++ err = -ENOMEM;
++ arg = kmalloc(sizeof(*arg), GFP_NOFS);
++ if (unlikely(!arg))
+ goto out;
+
-+ /* lock and kfree() will be called in trunc_xino() */
-+ args = kmalloc(sizeof(*args), GFP_NOFS);
-+ if (unlikely(!args)) {
-+ AuErr1("no memory\n");
-+ goto out_args;
++ arg->sb = sb;
++ arg->br = br;
++ arg->calc = *calc;
++ arg->ino = ino;
++ au_lcnt_inc(&br->br_count);
++ err = au_wkq_nowait(au_xino_call_do_new_async, arg, sb, AuWkq_NEST);
++ if (unlikely(err)) {
++ pr_err("wkq %d\n", err);
++ au_lcnt_dec(&br->br_count);
++ au_kfree_rcu(arg);
+ }
+
-+ sbinfo = au_sbi(sb);
-+ au_br_get(br);
-+ args->sb = sb;
-+ args->br = br;
-+ wkq_err = au_wkq_nowait(xino_do_trunc, args, sb, /*dlgt*/0);
-+ if (!wkq_err)
-+ return; /* success */
-+
-+ AuErr("wkq %d\n", wkq_err);
-+ au_br_put(br);
-+
-+ out_args:
-+ kfree(args);
-+ out:
-+ atomic_dec_return(&br->br_xino_running);
++out:
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static int au_xino_do_write(au_writef_t write, struct file *file,
-+ ino_t h_ino, struct au_xino_entry *xinoe)
++/*
++ * read @ino from xinofile for the specified branch{@sb, @bindex}
++ * at the position of @h_ino.
++ */
++int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
++ ino_t *ino)
+{
-+ loff_t pos;
++ int err;
+ ssize_t sz;
++ struct au_xi_calc calc;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
++ struct au_xino *xi;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos;
++ struct au_xi_writing *p;
+
-+ AuTraceEnter();
++ *ino = 0;
++ if (!au_opt_test(au_mntflags(sb), XINO))
++ return 0; /* no xino */
++
++ err = 0;
++ au_xi_calc(sb, h_ino, &calc);
++ xi = au_sbr(sb, bindex)->br_xino;
++ file = au_xino_file(xi, calc.idx);
++ if (!file) {
++ hbl = &xi->xi_writing;
++ hlist_bl_lock(hbl);
++ au_hbl_for_each(pos, hbl) {
++ p = container_of(pos, struct au_xi_writing, node);
++ if (p->h_ino == h_ino) {
++ AuDbg("hi%llu, i%llu, found\n",
++ (u64)p->h_ino, (u64)p->ino);
++ *ino = p->ino;
++ break;
++ }
++ }
++ hlist_bl_unlock(hbl);
++ return 0;
++ } else if (vfsub_f_size_read(file) < calc.pos + sizeof(*ino))
++ return 0; /* no xino */
++
++ sbinfo = au_sbi(sb);
++ sz = xino_fread(file, ino, sizeof(*ino), &calc.pos);
++ if (sz == sizeof(*ino))
++ return 0; /* success */
+
-+ pos = h_ino;
-+ if (unlikely(Au_LOFF_MAX / sizeof(*xinoe) - 1 < pos)) {
-+ AuIOErr1("too large hi%lu\n", (unsigned long)h_ino);
-+ return -EFBIG;
++ err = sz;
++ if (unlikely(sz >= 0)) {
++ err = -EIO;
++ AuIOErr("xino read error (%zd)\n", sz);
+ }
-+ pos *= sizeof(*xinoe);
-+ sz = xino_fwrite(write, file, xinoe, sizeof(*xinoe), &pos);
-+ if (sz == sizeof(*xinoe))
++ return err;
++}
++
++static int au_xino_do_write(struct file *file, struct au_xi_calc *calc,
++ ino_t ino)
++{
++ ssize_t sz;
++
++ sz = xino_fwrite(file, &ino, sizeof(ino), &calc->pos);
++ if (sz == sizeof(ino))
+ return 0; /* success */
+
+ AuIOErr("write failed (%zd)\n", sz);
+
+/*
+ * write @ino to the xinofile for the specified branch{@sb, @bindex}
-+ * at the position of @_ino.
-+ * when @ino is zero, it is written to the xinofile and means no entry.
++ * at the position of @h_ino.
++ * even if @ino is zero, it is written to the xinofile and means no entry.
++ * if the size of the xino file on a specific filesystem exceeds the watermark,
++ * try truncating it.
+ */
+int au_xino_write(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ struct au_xino_entry *xinoe)
++ ino_t ino)
+{
+ int err;
++ unsigned int mnt_flags;
++ struct au_xi_calc calc;
+ struct file *file;
+ struct au_branch *br;
-+ unsigned int mnt_flags;
++ struct au_xino *xi;
++ struct au_xi_writing *p;
+
-+ LKTRTrace("b%d, hi%lu, i%lu\n",
-+ bindex, (unsigned long)h_ino, (unsigned long)xinoe->ino);
-+ BUILD_BUG_ON(sizeof(long long) != sizeof(Au_LOFF_MAX)
-+ || ((loff_t)-1) > 0);
++ SiMustAnyLock(sb);
+
+ mnt_flags = au_mntflags(sb);
-+ if (!au_opt_test_xino(mnt_flags))
++ if (!au_opt_test(mnt_flags, XINO))
+ return 0;
+
++ au_xi_calc(sb, h_ino, &calc);
+ br = au_sbr(sb, bindex);
-+ file = br->br_xino.xi_file;
-+ AuDebugOn(!file);
++ xi = br->br_xino;
++ file = au_xino_file(xi, calc.idx);
++ if (!file) {
++ /* store the inum pair into the list */
++ p = kmalloc(sizeof(*p), GFP_NOFS | __GFP_NOFAIL);
++ p->h_ino = h_ino;
++ p->ino = ino;
++ au_hbl_add(&p->node, &xi->xi_writing);
++
++ /* create and write a new xino file asynchronously */
++ err = au_xino_new_async(sb, br, &calc, ino);
++ if (!err)
++ return 0; /* success */
++ goto out;
++ }
+
-+ err = au_xino_do_write(au_sbi(sb)->si_xwrite, file, h_ino, xinoe);
++ err = au_xino_do_write(file, &calc, ino);
+ if (!err) {
++ br = au_sbr(sb, bindex);
+ if (au_opt_test(mnt_flags, TRUNC_XINO)
-+ && au_test_trunc_xino(br->br_mnt->mnt_sb))
++ && au_test_fs_trunc_xino(au_br_sb(br)))
+ xino_try_trunc(sb, br);
+ return 0; /* success */
+ }
+
++out:
+ AuIOErr("write failed (%d)\n", err);
+ return -EIO;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+static const int page_bits = (int)PAGE_SIZE * BITS_PER_BYTE;
-+static ino_t xib_calc_ino(unsigned long pindex, int bit)
-+{
-+ ino_t ino;
-+
-+ AuDebugOn(bit < 0 || page_bits <= bit);
-+ ino = AUFS_FIRST_INO + pindex * page_bits + bit;
-+ return ino;
-+}
-+
-+static void xib_calc_bit(ino_t ino, unsigned long *pindex, int *bit)
-+{
-+ AuDebugOn(ino < AUFS_FIRST_INO);
-+ ino -= AUFS_FIRST_INO;
-+ *pindex = ino / page_bits;
-+ *bit = ino % page_bits;
-+}
++static ssize_t xino_fread_wkq(struct file *file, void *buf, size_t size,
++ loff_t *pos);
+
-+static int xib_pindex(struct super_block *sb, unsigned long pindex)
++/* todo: unnecessary to support mmap_sem since kernel-space? */
++ssize_t xino_fread(struct file *file, void *kbuf, size_t size, loff_t *pos)
+{
-+ int err;
-+ struct au_sbinfo *sbinfo;
-+ loff_t pos;
-+ ssize_t sz;
-+ struct file *xib;
-+ unsigned long *p;
-+
-+ LKTRTrace("pindex %lu\n", pindex);
-+ sbinfo = au_sbi(sb);
-+ MtxMustLock(&sbinfo->si_xib_mtx);
-+ AuDebugOn(pindex > ULONG_MAX / PAGE_SIZE
-+ || !au_opt_test_xino(sbinfo->si_mntflags));
-+
-+ if (pindex == sbinfo->si_xib_last_pindex)
-+ return 0;
++ ssize_t err;
++ int i;
++ const int prevent_endless = 10;
+
-+ xib = sbinfo->si_xib;
-+ p = sbinfo->si_xib_buf;
-+ pos = sbinfo->si_xib_last_pindex;
-+ pos *= PAGE_SIZE;
-+ sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos);
-+ if (unlikely(sz != PAGE_SIZE))
-+ goto out;
++ i = 0;
++ do {
++ err = vfsub_read_k(file, kbuf, size, pos);
++ if (err == -EINTR
++ && !au_wkq_test()
++ && fatal_signal_pending(current)) {
++ err = xino_fread_wkq(file, kbuf, size, pos);
++ BUG_ON(err == -EINTR);
++ }
++ } while (i++ < prevent_endless
++ && (err == -EAGAIN || err == -EINTR));
+
-+ pos = pindex;
-+ pos *= PAGE_SIZE;
-+ if (i_size_read(xib->f_dentry->d_inode) >= pos + PAGE_SIZE)
-+ sz = xino_fread(sbinfo->si_xread, xib, p, PAGE_SIZE, &pos);
-+ else {
-+ memset(p, 0, PAGE_SIZE);
-+ sz = xino_fwrite(sbinfo->si_xwrite, xib, p, PAGE_SIZE, &pos);
-+ }
-+ if (sz == PAGE_SIZE) {
-+ sbinfo->si_xib_last_pindex = pindex;
-+ return 0; /* success */
-+ }
++#if 0 /* reserved for future use */
++ if (err > 0)
++ fsnotify_access(file->f_path.dentry);
++#endif
+
-+ out:
-+ AuIOErr1("write failed (%zd)\n", sz);
-+ err = sz;
-+ if (sz >= 0)
-+ err = -EIO;
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++struct xino_fread_args {
++ ssize_t *errp;
++ struct file *file;
++ void *buf;
++ size_t size;
++ loff_t *pos;
++};
+
-+int au_xino_write0(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ ino_t ino)
++static void call_xino_fread(void *args)
+{
-+ int err, bit;
-+ unsigned long pindex;
-+ struct au_sbinfo *sbinfo;
-+ struct au_xino_entry xinoe = {
-+ .ino = 0
-+ };
-+
-+ LKTRTrace("b%d, hi%lu, i%lu\n",
-+ bindex, (unsigned long)h_ino, (unsigned long)ino);
-+
-+ if (!au_opt_test_xino(au_mntflags(sb)))
-+ return 0;
-+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ if (ino) {
-+ AuDebugOn(ino < AUFS_FIRST_INO);
-+ xib_calc_bit(ino, &pindex, &bit);
-+ AuDebugOn(page_bits <= bit);
-+ mutex_lock(&sbinfo->si_xib_mtx);
-+ err = xib_pindex(sb, pindex);
-+ if (!err) {
-+ clear_bit(bit, sbinfo->si_xib_buf);
-+ sbinfo->si_xib_next_bit = bit;
-+ }
-+ mutex_unlock(&sbinfo->si_xib_mtx);
-+ }
-+
-+ if (!err)
-+ err = au_xino_write(sb, bindex, h_ino, &xinoe);
-+ return err;
++ struct xino_fread_args *a = args;
++ *a->errp = xino_fread(a->file, a->buf, a->size, a->pos);
+}
+
-+ino_t au_xino_new_ino(struct super_block *sb)
++static ssize_t xino_fread_wkq(struct file *file, void *buf, size_t size,
++ loff_t *pos)
+{
-+ ino_t ino;
-+ struct au_sbinfo *sbinfo;
-+ int free_bit, err;
-+ unsigned long *p, pindex, ul, pend;
-+ struct file *file;
-+
-+ AuTraceEnter();
-+
-+ if (!au_opt_test_xino(au_mntflags(sb)))
-+ return iunique(sb, AUFS_FIRST_INO);
-+
-+ sbinfo = au_sbi(sb);
-+ mutex_lock(&sbinfo->si_xib_mtx);
-+ p = sbinfo->si_xib_buf;
-+ free_bit = sbinfo->si_xib_next_bit;
-+ if (free_bit < page_bits && !test_bit(free_bit, p))
-+ goto out; /* success */
-+ free_bit = find_first_zero_bit(p, page_bits);
-+ if (free_bit < page_bits)
-+ goto out; /* success */
-+
-+ pindex = sbinfo->si_xib_last_pindex;
-+ for (ul = pindex - 1; ul < ULONG_MAX; ul--) {
-+ err = xib_pindex(sb, ul);
-+ if (unlikely(err))
-+ goto out_err;
-+ free_bit = find_first_zero_bit(p, page_bits);
-+ if (free_bit < page_bits)
-+ goto out; /* success */
-+ }
++ ssize_t err;
++ int wkq_err;
++ struct xino_fread_args args = {
++ .errp = &err,
++ .file = file,
++ .buf = buf,
++ .size = size,
++ .pos = pos
++ };
+
-+ file = sbinfo->si_xib;
-+ pend = i_size_read(file->f_dentry->d_inode) / PAGE_SIZE;
-+ for (ul = pindex + 1; ul <= pend; ul++) {
-+ err = xib_pindex(sb, ul);
-+ if (unlikely(err))
-+ goto out_err;
-+ free_bit = find_first_zero_bit(p, page_bits);
-+ if (free_bit < page_bits)
-+ goto out; /* success */
-+ }
-+ BUG();
++ wkq_err = au_wkq_wait(call_xino_fread, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
+
-+ out:
-+ set_bit(free_bit, p);
-+ sbinfo->si_xib_next_bit++;
-+ pindex = sbinfo->si_xib_last_pindex;
-+ mutex_unlock(&sbinfo->si_xib_mtx);
-+ ino = xib_calc_ino(pindex, free_bit);
-+ LKTRTrace("i%lu\n", (unsigned long)ino);
-+ return ino;
-+ out_err:
-+ mutex_unlock(&sbinfo->si_xib_mtx);
-+ LKTRTrace("i0\n");
-+ return 0;
++ return err;
+}
+
-+/*
-+ * read @ino from xinofile for the specified branch{@sb, @bindex}
-+ * at the position of @h_ino.
-+ * if @ino does not exist and @do_new is true, get new one.
-+ */
-+int au_xino_read(struct super_block *sb, aufs_bindex_t bindex, ino_t h_ino,
-+ struct au_xino_entry *xinoe)
-+{
-+ int err;
-+ struct file *file;
-+ loff_t pos;
-+ ssize_t sz;
-+ struct au_sbinfo *sbinfo;
-+
-+ LKTRTrace("b%d, hi%lu\n", bindex, (unsigned long)h_ino);
-+
-+ xinoe->ino = 0;
-+ if (!au_opt_test_xino(au_mntflags(sb)))
-+ return 0; /* no ino */
-+
-+ err = 0;
-+ sbinfo = au_sbi(sb);
-+ pos = h_ino;
-+ if (unlikely(Au_LOFF_MAX / sizeof(*xinoe) - 1 < pos)) {
-+ AuIOErr1("too large hi%lu\n", (unsigned long)h_ino);
-+ return -EFBIG;
-+ }
-+ pos *= sizeof(*xinoe);
++static ssize_t xino_fwrite_wkq(struct file *file, void *buf, size_t size,
++ loff_t *pos);
+
-+ file = au_sbr(sb, bindex)->br_xino.xi_file;
-+ AuDebugOn(!file);
-+ if (i_size_read(file->f_dentry->d_inode) < pos + sizeof(*xinoe))
-+ return 0; /* no ino */
++static ssize_t do_xino_fwrite(struct file *file, void *kbuf, size_t size,
++ loff_t *pos)
++{
++ ssize_t err;
++ int i;
++ const int prevent_endless = 10;
+
-+ sz = xino_fread(sbinfo->si_xread, file, xinoe, sizeof(*xinoe), &pos);
-+ if (sz == sizeof(*xinoe))
-+ return 0; /* success */
++ i = 0;
++ do {
++ err = vfsub_write_k(file, kbuf, size, pos);
++ if (err == -EINTR
++ && !au_wkq_test()
++ && fatal_signal_pending(current)) {
++ err = xino_fwrite_wkq(file, kbuf, size, pos);
++ BUG_ON(err == -EINTR);
++ }
++ } while (i++ < prevent_endless
++ && (err == -EAGAIN || err == -EINTR));
+
-+ err = sz;
-+ if (unlikely(sz >= 0)) {
-+ err = -EIO;
-+ AuIOErr("xino read error (%zd)\n", sz);
-+ }
++#if 0 /* reserved for future use */
++ if (err > 0)
++ fsnotify_modify(file->f_path.dentry);
++#endif
+
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
++struct do_xino_fwrite_args {
++ ssize_t *errp;
++ struct file *file;
++ void *buf;
++ size_t size;
++ loff_t *pos;
++};
+
-+struct file *au_xino_create(struct super_block *sb, char *fname, int silent)
++static void call_do_xino_fwrite(void *args)
+{
-+ struct file *file;
-+ int err;
-+ struct dentry *h_parent;
-+ struct inode *h_dir;
-+ struct vfsub_args vargs;
++ struct do_xino_fwrite_args *a = args;
++ *a->errp = do_xino_fwrite(a->file, a->buf, a->size, a->pos);
++}
+
-+ LKTRTrace("%s\n", fname);
++static ssize_t xino_fwrite_wkq(struct file *file, void *buf, size_t size,
++ loff_t *pos)
++{
++ ssize_t err;
++ int wkq_err;
++ struct do_xino_fwrite_args args = {
++ .errp = &err,
++ .file = file,
++ .buf = buf,
++ .size = size,
++ .pos = pos
++ };
+
+ /*
-+ * at mount-time, and the xino file is the default path,
-+ * hinotify is disabled so we have no inotify events to ignore.
-+ * when a user specified the xino, we cannot get au_hdir to be ignored.
++ * it breaks RLIMIT_FSIZE and normal user's limit,
++ * users should care about quota and real 'filesystem full.'
+ */
-+ vfsub_args_init(&vargs, /*ign*/NULL, /*dlgt*/0, 0);
-+ file = vfsub_filp_open(fname, O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE,
-+ S_IRUGO | S_IWUGO);
-+ if (IS_ERR(file)) {
-+ if (!silent)
-+ AuErr("open %s(%ld)\n", fname, PTR_ERR(file));
-+ return file;
-+ }
++ wkq_err = au_wkq_wait(call_do_xino_fwrite, &args);
++ if (unlikely(wkq_err))
++ err = wkq_err;
+
-+ /* keep file count */
-+ h_parent = dget_parent(file->f_dentry);
-+ h_dir = h_parent->d_inode;
-+ mutex_lock_nested(&h_dir->i_mutex, AuLsc_I_PARENT);
-+ /* mnt_want_write() is unnecessary here */
-+ err = vfsub_unlink(h_dir, file->f_dentry, &vargs);
-+ mutex_unlock(&h_dir->i_mutex);
-+ dput(h_parent);
-+ if (unlikely(err)) {
-+ if (!silent)
-+ AuErr("unlink %s(%d)\n", fname, err);
-+ goto out;
-+ }
++ return err;
++}
+
-+ if (sb != file->f_dentry->d_sb)
-+ return file; /* success */
++ssize_t xino_fwrite(struct file *file, void *buf, size_t size, loff_t *pos)
++{
++ ssize_t err;
+
-+ if (!silent)
-+ AuErr("%s must be outside\n", fname);
-+ err = -EINVAL;
++ if (rlimit(RLIMIT_FSIZE) == RLIM_INFINITY) {
++ lockdep_off();
++ err = do_xino_fwrite(file, buf, size, pos);
++ lockdep_on();
++ } else {
++ lockdep_off();
++ err = xino_fwrite_wkq(file, buf, size, pos);
++ lockdep_on();
++ }
+
-+ out:
-+ fput(file);
-+ file = ERR_PTR(err);
-+ return file;
++ return err;
+}
+
++/* ---------------------------------------------------------------------- */
++
+/*
-+ * find another branch who is on the same filesystem of the specified
-+ * branch{@btgt}. search until @bend.
++ * inode number bitmap
+ */
-+static int is_sb_shared(struct super_block *sb, aufs_bindex_t btgt,
-+ aufs_bindex_t bend)
++static const int page_bits = (int)PAGE_SIZE * BITS_PER_BYTE;
++static ino_t xib_calc_ino(unsigned long pindex, int bit)
+{
-+ aufs_bindex_t bindex;
-+ struct super_block *tgt_sb = au_sbr_sb(sb, btgt);
++ ino_t ino;
+
-+ for (bindex = 0; bindex < btgt; bindex++)
-+ if (unlikely(tgt_sb == au_sbr_sb(sb, bindex)))
-+ return bindex;
-+ for (bindex++; bindex <= bend; bindex++)
-+ if (unlikely(tgt_sb == au_sbr_sb(sb, bindex)))
-+ return bindex;
-+ return -1;
++ AuDebugOn(bit < 0 || page_bits <= bit);
++ ino = AUFS_FIRST_INO + pindex * page_bits + bit;
++ return ino;
+}
+
-+/*
-+ * create a new xinofile at the same place/path as @base_file.
-+ */
-+struct file *au_xino_create2(struct super_block *sb, struct file *base_file,
-+ struct file *copy_src)
++static void xib_calc_bit(ino_t ino, unsigned long *pindex, int *bit)
++{
++ AuDebugOn(ino < AUFS_FIRST_INO);
++ ino -= AUFS_FIRST_INO;
++ *pindex = ino / page_bits;
++ *bit = ino % page_bits;
++}
++
++static int xib_pindex(struct super_block *sb, unsigned long pindex)
+{
-+ struct file *file;
+ int err;
-+ struct dentry *base, *dentry, *parent;
-+ struct inode *dir, *inode;
-+ struct qstr *name;
-+ struct au_hinode *hdir;
-+ struct au_branch *br;
-+ aufs_bindex_t bindex;
-+ struct au_hin_ignore ign;
-+ struct vfsub_args vargs;
-+ struct au_ndx ndx = {
-+ .nfsmnt = NULL,
-+ .flags = 0,
-+ .nd = NULL,
-+ /* .br = NULL */
-+ };
++ loff_t pos;
++ ssize_t sz;
++ struct au_sbinfo *sbinfo;
++ struct file *xib;
++ unsigned long *p;
+
-+ base = base_file->f_dentry;
-+ LKTRTrace("%.*s\n", AuDLNPair(base));
-+ parent = base->d_parent; /* dir inode is locked */
-+ dir = parent->d_inode;
-+ IMustLock(dir);
++ sbinfo = au_sbi(sb);
++ MtxMustLock(&sbinfo->si_xib_mtx);
++ AuDebugOn(pindex > ULONG_MAX / PAGE_SIZE
++ || !au_opt_test(sbinfo->si_mntflags, XINO));
+
-+ file = ERR_PTR(-EINVAL);
-+ if (unlikely(au_test_nfs(parent->d_sb)
-+ || au_test_ecryptfs(parent->d_sb)))
-+ goto out;
++ if (pindex == sbinfo->si_xib_last_pindex)
++ return 0;
+
-+ /* do not superio, nor NFS. */
-+ name = &base->d_name;
-+ dentry = au_lkup_one(name->name, parent, name->len, &ndx);
-+ if (IS_ERR(dentry)) {
-+ file = (void *)dentry;
-+ AuErr("%.*s lookup err %ld\n", AuLNPair(name), PTR_ERR(dentry));
++ xib = sbinfo->si_xib;
++ p = sbinfo->si_xib_buf;
++ pos = sbinfo->si_xib_last_pindex;
++ pos *= PAGE_SIZE;
++ sz = xino_fwrite(xib, p, PAGE_SIZE, &pos);
++ if (unlikely(sz != PAGE_SIZE))
+ goto out;
-+ }
+
-+ hdir = NULL;
-+ br = au_xino_def_br(au_sbi(sb));
-+ if (br) {
-+ bindex = au_find_bindex(sb, br);
-+ if (bindex >= 0)
-+ hdir = au_hi(sb->s_root->d_inode, bindex);
-+ }
-+ vfsub_args_init(&vargs, &ign, 0, 0);
-+ vfsub_ign_hinode(&vargs, IN_CREATE, hdir);
-+ /* no need to mnt_want_write() since we call dentry_open() later */
-+ err = vfsub_create(dir, dentry, S_IRUGO | S_IWUGO, NULL, &vargs);
-+ if (unlikely(err)) {
-+ file = ERR_PTR(err);
-+ AuErr("%.*s create err %d\n", AuLNPair(name), err);
-+ goto out_dput;
-+ }
-+ file = dentry_open(dget(dentry), mntget(base_file->f_vfsmnt),
-+ O_RDWR | O_CREAT | O_EXCL | O_LARGEFILE);
-+ if (IS_ERR(file)) {
-+ AuErr("%.*s open err %ld\n", AuLNPair(name), PTR_ERR(file));
-+ goto out_dput;
-+ }
-+ vfsub_args_reinit(&vargs);
-+ vfsub_ign_hinode(&vargs, IN_DELETE, hdir);
-+ err = vfsub_unlink(dir, dentry, &vargs);
-+ if (unlikely(err)) {
-+ AuErr("%.*s unlink err %d\n", AuLNPair(name), err);
-+ goto out_fput;
++ pos = pindex;
++ pos *= PAGE_SIZE;
++ if (vfsub_f_size_read(xib) >= pos + PAGE_SIZE)
++ sz = xino_fread(xib, p, PAGE_SIZE, &pos);
++ else {
++ memset(p, 0, PAGE_SIZE);
++ sz = xino_fwrite(xib, p, PAGE_SIZE, &pos);
+ }
-+
-+ if (copy_src) {
-+ inode = copy_src->f_dentry->d_inode;
-+ err = au_copy_file(file, copy_src, i_size_read(inode),
-+ hdir, sb, &vargs);
-+ if (unlikely(err)) {
-+ AuErr("%.*s copy err %d\n", AuLNPair(name), err);
-+ goto out_fput;
-+ }
++ if (sz == PAGE_SIZE) {
++ sbinfo->si_xib_last_pindex = pindex;
++ return 0; /* success */
+ }
-+ goto out_dput; /* success */
+
-+ out_fput:
-+ fput(file);
-+ file = ERR_PTR(err);
-+ out_dput:
-+ dput(dentry);
-+ out:
-+ AuTraceErrPtr(file);
-+ return file;
++out:
++ AuIOErr1("write failed (%zd)\n", sz);
++ err = sz;
++ if (sz >= 0)
++ err = -EIO;
++ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * initialize the xinofile for the specified branch{@sb, @bindex}
-+ * at the place/path where @base_file indicates.
-+ * test whether another branch is on the same filesystem or not,
-+ * if @do_test is true.
-+ */
-+int au_xino_br(struct super_block *sb, struct au_branch *br, ino_t h_ino,
-+ struct file *base_file, int do_test)
++static void au_xib_clear_bit(struct inode *inode)
+{
-+ int err;
-+ struct au_branch *shared_br;
-+ aufs_bindex_t bshared, bend, bindex;
-+ unsigned char do_create;
-+ struct inode *dir;
-+ struct au_xino_entry xinoe;
-+ struct dentry *parent;
-+ struct file *file;
-+ struct super_block *tgt_sb;
++ int err, bit;
++ unsigned long pindex;
++ struct super_block *sb;
++ struct au_sbinfo *sbinfo;
+
-+ LKTRTrace("base_file %p, do_test %d\n", base_file, do_test);
-+ SiMustWriteLock(sb);
-+ AuDebugOn(!au_opt_test_xino(au_mntflags(sb)));
-+ AuDebugOn(br->br_xino.xi_file);
-+
-+ do_create = 1;
-+ bshared = -1;
-+ shared_br = NULL;
-+ bend = au_sbend(sb);
-+ if (do_test) {
-+ tgt_sb = br->br_mnt->mnt_sb;
-+ for (bindex = 0; bindex <= bend; bindex++)
-+ if (tgt_sb == au_sbr_sb(sb, bindex)) {
-+ bshared = bindex;
-+ break;
-+ }
-+ }
-+ if (bshared >= 0) {
-+ shared_br = au_sbr(sb, bshared);
-+ do_create = !shared_br->br_xino.xi_file;
-+ }
++ AuDebugOn(inode->i_nlink);
+
-+ if (do_create) {
-+ parent = dget_parent(base_file->f_dentry);
-+ dir = parent->d_inode;
-+ mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT);
-+ /* mnt_want_write() is unnecessary here */
-+ file = au_xino_create2(sb, base_file, NULL);
-+ mutex_unlock(&dir->i_mutex);
-+ dput(parent);
-+ err = PTR_ERR(file);
-+ if (IS_ERR(file))
-+ goto out;
-+ br->br_xino.xi_file = file;
-+ } else {
-+ br->br_xino.xi_file = shared_br->br_xino.xi_file;
-+ get_file(br->br_xino.xi_file);
++ sb = inode->i_sb;
++ xib_calc_bit(inode->i_ino, &pindex, &bit);
++ AuDebugOn(page_bits <= bit);
++ sbinfo = au_sbi(sb);
++ mutex_lock(&sbinfo->si_xib_mtx);
++ err = xib_pindex(sb, pindex);
++ if (!err) {
++ clear_bit(bit, sbinfo->si_xib_buf);
++ sbinfo->si_xib_next_bit = bit;
+ }
++ mutex_unlock(&sbinfo->si_xib_mtx);
++}
+
-+ xinoe.ino = AUFS_ROOT_INO;
-+#if 0 /* reserved for future use */
-+ xinoe.h_gen = h_inode->i_generation;
-+ WARN_ON(xinoe.h_gen == AuXino_INVALID_HGEN);
-+#endif
-+ err = au_xino_do_write(au_sbi(sb)->si_xwrite, br->br_xino.xi_file,
-+ h_ino, &xinoe);
-+ if (!err)
-+ return 0; /* success */
-+
++/* ---------------------------------------------------------------------- */
+
-+ out:
-+ AuTraceErr(err);
-+ return err;
-+}
++/*
++ * truncate a xino bitmap file
++ */
+
-+/* too slow */
++/* todo: slow */
+static int do_xib_restore(struct super_block *sb, struct file *file, void *page)
+{
+ int err, bit;
-+ struct au_sbinfo *sbinfo;
-+ au_readf_t func;
-+ loff_t pos, pend;
+ ssize_t sz;
-+ struct au_xino_entry *xinoe;
+ unsigned long pindex;
-+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
++ loff_t pos, pend;
++ struct au_sbinfo *sbinfo;
++ ino_t *ino;
++ unsigned long *p;
+
+ err = 0;
+ sbinfo = au_sbi(sb);
-+ func = sbinfo->si_xread;
-+ pend = i_size_read(file->f_dentry->d_inode);
-+#ifdef CONFIG_AUFS_DEBUG
-+ if (unlikely(pend > (1 << 22)))
-+ AuWarn("testing a large xino file %lld\n", (long long)pend);
-+#endif
++ MtxMustLock(&sbinfo->si_xib_mtx);
++ p = sbinfo->si_xib_buf;
++ pend = vfsub_f_size_read(file);
+ pos = 0;
+ while (pos < pend) {
-+ sz = xino_fread(func, file, page, PAGE_SIZE, &pos);
++ sz = xino_fread(file, page, PAGE_SIZE, &pos);
+ err = sz;
+ if (unlikely(sz <= 0))
+ goto out;
+
+ err = 0;
-+ for (xinoe = page; sz > 0; xinoe++, sz -= sizeof(xinoe)) {
-+ if (unlikely(xinoe->ino < AUFS_FIRST_INO))
++ for (ino = page; sz > 0; ino++, sz -= sizeof(ino)) {
++ if (unlikely(*ino < AUFS_FIRST_INO))
+ continue;
+
-+ xib_calc_bit(xinoe->ino, &pindex, &bit);
++ xib_calc_bit(*ino, &pindex, &bit);
+ AuDebugOn(page_bits <= bit);
+ err = xib_pindex(sb, pindex);
+ if (!err)
-+ set_bit(bit, sbinfo->si_xib_buf);
++ set_bit(bit, p);
+ else
+ goto out;
+ }
+ }
+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
+static int xib_restore(struct super_block *sb)
+{
-+ int err;
-+ aufs_bindex_t bindex, bend;
++ int err, i;
++ unsigned int nfile;
++ aufs_bindex_t bindex, bbot;
+ void *page;
-+
-+ AuTraceEnter();
++ struct au_branch *br;
++ struct au_xino *xi;
++ struct file *file;
+
+ err = -ENOMEM;
+ page = (void *)__get_free_page(GFP_NOFS);
+ goto out;
+
+ err = 0;
-+ bend = au_sbend(sb);
-+ for (bindex = 0; !err && bindex <= bend; bindex++)
-+ if (!bindex || is_sb_shared(sb, bindex, bindex - 1) < 0)
-+ err = do_xib_restore
-+ (sb, au_sbr(sb, bindex)->br_xino.xi_file, page);
-+ else
-+ LKTRTrace("b%d\n", bindex);
++ bbot = au_sbbot(sb);
++ for (bindex = 0; !err && bindex <= bbot; bindex++)
++ if (!bindex || is_sb_shared(sb, bindex, bindex - 1) < 0) {
++ br = au_sbr(sb, bindex);
++ xi = br->br_xino;
++ nfile = xi->xi_nfile;
++ for (i = 0; i < nfile; i++) {
++ file = au_xino_file(xi, i);
++ if (file)
++ err = do_xib_restore(sb, file, page);
++ }
++ } else
++ AuDbg("skip shared b%d\n", bindex);
+ free_page((unsigned long)page);
+
-+ out:
-+ AuTraceErr(err);
++out:
+ return err;
+}
+
+int au_xib_trunc(struct super_block *sb)
+{
+ int err;
-+ unsigned int mnt_flags;
+ ssize_t sz;
+ loff_t pos;
+ struct au_sbinfo *sbinfo;
+ unsigned long *p;
-+ struct dentry *parent;
-+ struct inode *dir;
+ struct file *file;
+
-+ AuTraceEnter();
+ SiMustWriteLock(sb);
+
+ err = 0;
-+ mnt_flags = au_mntflags(sb);
-+ if (!au_opt_test_xino(mnt_flags))
++ sbinfo = au_sbi(sb);
++ if (!au_opt_test(sbinfo->si_mntflags, XINO))
+ goto out;
+
-+ sbinfo = au_sbi(sb);
-+ if (i_size_read(sbinfo->si_xib->f_dentry->d_inode) <= PAGE_SIZE)
++ file = sbinfo->si_xib;
++ if (vfsub_f_size_read(file) <= PAGE_SIZE)
+ goto out;
-+ parent = dget_parent(sbinfo->si_xib->f_dentry);
-+ dir = parent->d_inode;
-+ mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT);
-+ /* mnt_want_write() is unnecessary here */
-+ file = au_xino_create2(sb, sbinfo->si_xib, NULL);
-+ mutex_unlock(&dir->i_mutex);
-+ dput(parent);
++
++ file = au_xino_create2(sb, &sbinfo->si_xib->f_path, NULL);
+ err = PTR_ERR(file);
+ if (IS_ERR(file))
+ goto out;
-+ fput(sbinfo->si_xib);
-+ sbinfo->si_xib = file;
++ fput(sbinfo->si_xib);
++ sbinfo->si_xib = file;
++
++ p = sbinfo->si_xib_buf;
++ memset(p, 0, PAGE_SIZE);
++ pos = 0;
++ sz = xino_fwrite(sbinfo->si_xib, p, PAGE_SIZE, &pos);
++ if (unlikely(sz != PAGE_SIZE)) {
++ err = sz;
++ AuIOErr("err %d\n", err);
++ if (sz >= 0)
++ err = -EIO;
++ goto out;
++ }
++
++ mutex_lock(&sbinfo->si_xib_mtx);
++ /* mnt_want_write() is unnecessary here */
++ err = xib_restore(sb);
++ mutex_unlock(&sbinfo->si_xib_mtx);
++
++out:
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++struct au_xino *au_xino_alloc(unsigned int nfile)
++{
++ struct au_xino *xi;
++
++ xi = kzalloc(sizeof(*xi), GFP_NOFS);
++ if (unlikely(!xi))
++ goto out;
++ xi->xi_nfile = nfile;
++ xi->xi_file = kcalloc(nfile, sizeof(*xi->xi_file), GFP_NOFS);
++ if (unlikely(!xi->xi_file))
++ goto out_free;
+
-+ p = sbinfo->si_xib_buf;
-+ memset(p, 0, PAGE_SIZE);
-+ pos = 0;
-+ sz = xino_fwrite(sbinfo->si_xwrite, sbinfo->si_xib, p, PAGE_SIZE, &pos);
-+ if (unlikely(sz != PAGE_SIZE)) {
-+ err = sz;
-+ AuIOErr("err %d\n", err);
-+ if (sz >= 0)
-+ err = -EIO;
++ xi->xi_nondir.total = 8; /* initial size */
++ xi->xi_nondir.array = kcalloc(xi->xi_nondir.total, sizeof(ino_t),
++ GFP_NOFS);
++ if (unlikely(!xi->xi_nondir.array))
++ goto out_file;
++
++ spin_lock_init(&xi->xi_nondir.spin);
++ init_waitqueue_head(&xi->xi_nondir.wqh);
++ mutex_init(&xi->xi_mtx);
++ INIT_HLIST_BL_HEAD(&xi->xi_writing);
++ atomic_set(&xi->xi_truncating, 0);
++ kref_init(&xi->xi_kref);
++ goto out; /* success */
++
++out_file:
++ au_kfree_try_rcu(xi->xi_file);
++out_free:
++ au_kfree_rcu(xi);
++ xi = NULL;
++out:
++ return xi;
++}
++
++static int au_xino_init(struct au_branch *br, int idx, struct file *file)
++{
++ int err;
++ struct au_xino *xi;
++
++ err = 0;
++ xi = au_xino_alloc(idx + 1);
++ if (unlikely(!xi)) {
++ err = -ENOMEM;
+ goto out;
+ }
+
-+ if (au_opt_test_xino(mnt_flags)) {
-+ mutex_lock(&sbinfo->si_xib_mtx);
-+ /* mnt_want_write() is unnecessary here */
-+ err = xib_restore(sb);
-+ mutex_unlock(&sbinfo->si_xib_mtx);
-+#if 0 /* reserved for future use */
-+ } else {
-+ /* is it really safe? */
-+ /* dont trust BKL */
-+ AuDebugOn(!kernel_locked());
-+ ino = AUFS_FIRST_INO;
-+ list_for_each_entry(inode, &sb->s_inodes, i_sb_list)
-+ if (ino < inode->i_ino)
-+ ino = inode->i_ino;
-+
-+ /* make iunique to return larger than active max inode number */
-+ iunique(sb, ino);
-+ err = 0;
-+#endif
-+ }
++ if (file)
++ get_file(file);
++ xi->xi_file[idx] = file;
++ AuDebugOn(br->br_xino);
++ br->br_xino = xi;
+
+out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
-+/*
-+ * xino mount option handlers
-+ */
-+static au_readf_t find_readf(struct file *h_file)
++static void au_xino_release(struct kref *kref)
+{
-+ const struct file_operations *fop = h_file->f_op;
-+
-+ if (fop) {
-+ if (fop->read)
-+ return fop->read;
-+ if (fop->aio_read)
-+ return do_sync_read;
++ struct au_xino *xi;
++ int i;
++ unsigned long ul;
++ struct hlist_bl_head *hbl;
++ struct hlist_bl_node *pos, *n;
++ struct au_xi_writing *p;
++
++ xi = container_of(kref, struct au_xino, xi_kref);
++ for (i = 0; i < xi->xi_nfile; i++)
++ if (xi->xi_file[i])
++ fput(xi->xi_file[i]);
++ for (i = xi->xi_nondir.total - 1; i >= 0; i--)
++ AuDebugOn(xi->xi_nondir.array[i]);
++ mutex_destroy(&xi->xi_mtx);
++ hbl = &xi->xi_writing;
++ ul = au_hbl_count(hbl);
++ if (unlikely(ul)) {
++ pr_warn("xi_writing %lu\n", ul);
++ hlist_bl_lock(hbl);
++ hlist_bl_for_each_entry_safe(p, pos, n, hbl, node) {
++ hlist_bl_del(&p->node);
++ /* kmemleak reported au_kfree_rcu() doesn't free it */
++ kfree(p);
++ }
++ hlist_bl_unlock(hbl);
+ }
-+ return ERR_PTR(-ENOSYS);
++ au_kfree_try_rcu(xi->xi_file);
++ au_kfree_try_rcu(xi->xi_nondir.array);
++ au_kfree_rcu(xi);
+}
+
-+static au_writef_t find_writef(struct file *h_file)
++int au_xino_put(struct au_branch *br)
+{
-+ const struct file_operations *fop = h_file->f_op;
++ int ret;
++ struct au_xino *xi;
+
-+ if (fop) {
-+ if (fop->write)
-+ return fop->write;
-+ if (fop->aio_write)
-+ return do_sync_write;
++ ret = 0;
++ xi = br->br_xino;
++ if (xi) {
++ br->br_xino = NULL;
++ ret = kref_put(&xi->xi_kref, au_xino_release);
+ }
-+ return ERR_PTR(-ENOSYS);
++
++ return ret;
+}
+
++/* ---------------------------------------------------------------------- */
++
++/*
++ * xino mount option handlers
++ */
++
+/* xino bitmap */
+static void xino_clear_xib(struct super_block *sb)
+{
+ struct au_sbinfo *sbinfo;
+
-+ AuTraceEnter();
+ SiMustWriteLock(sb);
+
+ sbinfo = au_sbi(sb);
-+ sbinfo->si_xread = NULL;
-+ sbinfo->si_xwrite = NULL;
+ 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;
+}
+
-+static int au_xino_set_xib(struct super_block *sb, struct file *base)
++static int au_xino_set_xib(struct super_block *sb, struct path *path)
+{
+ int err;
++ loff_t pos;
+ struct au_sbinfo *sbinfo;
+ struct file *file;
-+ loff_t pos;
++ struct super_block *xi_sb;
+
-+ LKTRTrace("%.*s\n", AuDLNPair(base->f_dentry));
+ SiMustWriteLock(sb);
+
+ sbinfo = au_sbi(sb);
-+ file = au_xino_create2(sb, base, sbinfo->si_xib);
++ file = au_xino_create2(sb, path, sbinfo->si_xib);
+ err = PTR_ERR(file);
+ if (IS_ERR(file))
+ goto out;
+ if (sbinfo->si_xib)
+ fput(sbinfo->si_xib);
+ sbinfo->si_xib = file;
-+ sbinfo->si_xread = find_readf(file);
-+ AuDebugOn(IS_ERR(sbinfo->si_xread));
-+ sbinfo->si_xwrite = find_writef(file);
-+ AuDebugOn(IS_ERR(sbinfo->si_xwrite));
++ xi_sb = file_inode(file)->i_sb;
++ sbinfo->si_ximaxent = xi_sb->s_maxbytes;
++ if (unlikely(sbinfo->si_ximaxent < PAGE_SIZE)) {
++ err = -EIO;
++ pr_err("s_maxbytes(%llu) on %s is too small\n",
++ (u64)sbinfo->si_ximaxent, au_sbtype(xi_sb));
++ goto out_unset;
++ }
++ sbinfo->si_ximaxent /= sizeof(ino_t);
+
+ err = -ENOMEM;
+ if (!sbinfo->si_xib_buf)
+
+ sbinfo->si_xib_last_pindex = 0;
+ sbinfo->si_xib_next_bit = 0;
-+
-+ /* no need to lock for i_size_read() */
-+ if (i_size_read(file->f_dentry->d_inode) < PAGE_SIZE) {
++ if (vfsub_f_size_read(file) < PAGE_SIZE) {
+ pos = 0;
-+ err = xino_fwrite(sbinfo->si_xwrite, file, sbinfo->si_xib_buf,
-+ PAGE_SIZE, &pos);
++ err = xino_fwrite(file, sbinfo->si_xib_buf, PAGE_SIZE, &pos);
+ if (unlikely(err != PAGE_SIZE))
+ goto out_free;
+ }
+ err = 0;
+ goto out; /* success */
+
-+ out_free:
-+ free_page((unsigned long)sbinfo->si_xib_buf);
++out_free:
++ if (sbinfo->si_xib_buf)
++ free_page((unsigned long)sbinfo->si_xib_buf);
+ sbinfo->si_xib_buf = NULL;
+ if (err >= 0)
+ err = -EIO;
-+ out_unset:
++out_unset:
+ fput(sbinfo->si_xib);
+ sbinfo->si_xib = NULL;
-+ sbinfo->si_xread = NULL;
-+ sbinfo->si_xwrite = NULL;
-+ out:
++out:
+ AuTraceErr(err);
+ return err;
+}
+/* xino for each branch */
+static void xino_clear_br(struct super_block *sb)
+{
-+ aufs_bindex_t bindex, bend;
++ aufs_bindex_t bindex, bbot;
+ struct au_branch *br;
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+
-+ bend = au_sbend(sb);
-+ for (bindex = 0; bindex <= bend; bindex++) {
++ bbot = au_sbbot(sb);
++ for (bindex = 0; bindex <= bbot; bindex++) {
+ br = au_sbr(sb, bindex);
-+ if (!br || !br->br_xino.xi_file)
-+ continue;
++ AuDebugOn(!br);
++ au_xino_put(br);
++ }
++}
++
++static void au_xino_set_br_shared(struct super_block *sb, struct au_branch *br,
++ aufs_bindex_t bshared)
++{
++ struct au_branch *brshared;
+
-+ fput(br->br_xino.xi_file);
-+ br->br_xino.xi_file = NULL;
++ brshared = au_sbr(sb, bshared);
++ AuDebugOn(!brshared->br_xino);
++ AuDebugOn(!brshared->br_xino->xi_file);
++ if (br->br_xino != brshared->br_xino) {
++ au_xino_get(brshared);
++ au_xino_put(br);
++ br->br_xino = brshared->br_xino;
+ }
+}
+
-+static int au_xino_set_br(struct super_block *sb, struct file *base)
++struct au_xino_do_set_br {
++ struct au_branch *br;
++ ino_t h_ino;
++ aufs_bindex_t bshared;
++};
++
++static int au_xino_do_set_br(struct super_block *sb, struct path *path,
++ struct au_xino_do_set_br *args)
+{
+ int err;
-+ aufs_bindex_t bindex, bend, bshared;
-+ struct {
-+ struct file *old, *new;
-+ } *fpair, *p;
++ struct au_xi_calc calc;
++ struct file *file;
+ struct au_branch *br;
-+ struct au_xino_entry xinoe;
-+ struct inode *inode;
-+ au_writef_t writef;
++ struct au_xi_new xinew = {
++ .base = path
++ };
+
-+ LKTRTrace("%.*s\n", AuDLNPair(base->f_dentry));
-+ SiMustWriteLock(sb);
++ br = args->br;
++ xinew.xi = br->br_xino;
++ au_xi_calc(sb, args->h_ino, &calc);
++ xinew.copy_src = au_xino_file(xinew.xi, calc.idx);
++ if (args->bshared >= 0)
++ /* shared xino */
++ au_xino_set_br_shared(sb, br, args->bshared);
++ else if (!xinew.xi) {
++ /* new xino */
++ err = au_xino_init(br, calc.idx, xinew.copy_src);
++ if (unlikely(err))
++ goto out;
++ }
+
-+ err = -ENOMEM;
-+ bend = au_sbend(sb);
-+ fpair = kcalloc(bend + 1, sizeof(*fpair), GFP_NOFS);
-+ if (unlikely(!fpair))
++ /* force re-creating */
++ xinew.xi = br->br_xino;
++ xinew.idx = calc.idx;
++ mutex_lock(&xinew.xi->xi_mtx);
++ file = au_xi_new(sb, &xinew);
++ mutex_unlock(&xinew.xi->xi_mtx);
++ err = PTR_ERR(file);
++ if (IS_ERR(file))
+ goto out;
++ AuDebugOn(!file);
+
-+ inode = sb->s_root->d_inode;
-+ xinoe.ino = AUFS_ROOT_INO;
-+ writef = au_sbi(sb)->si_xwrite;
-+ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) {
-+ br = au_sbr(sb, bindex);
-+ bshared = is_sb_shared(sb, bindex, bindex - 1);
-+ if (bshared >= 0) {
-+ /* shared xino */
-+ *p = fpair[bshared];
-+ get_file(p->new);
-+ }
-+
-+ if (!p->new) {
-+ /* new xino */
-+ p->old = br->br_xino.xi_file;
-+ p->new = au_xino_create2(sb, base, br->br_xino.xi_file);
-+ err = PTR_ERR(p->new);
-+ if (IS_ERR(p->new)) {
-+ p->new = NULL;
-+ goto out_pair;
-+ }
-+ }
++ err = au_xino_do_write(file, &calc, AUFS_ROOT_INO);
++ if (unlikely(err))
++ au_xino_put(br);
+
-+ err = au_xino_do_write(writef, p->new,
-+ au_h_iptr(inode, bindex)->i_ino, &xinoe);
-+ if (unlikely(err))
-+ goto out_pair;
-+ }
++out:
++ AuTraceErr(err);
++ return err;
++}
+
-+ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++) {
-+ br = au_sbr(sb, bindex);
-+ AuDebugOn(p->old != br->br_xino.xi_file);
-+ if (br->br_xino.xi_file)
-+ fput(br->br_xino.xi_file);
-+ get_file(p->new);
-+ br->br_xino.xi_file = p->new;
-+ }
++static int au_xino_set_br(struct super_block *sb, struct path *path)
++{
++ int err;
++ aufs_bindex_t bindex, bbot;
++ struct au_xino_do_set_br args;
++ struct inode *inode;
+
-+ out_pair:
-+ for (bindex = 0, p = fpair; bindex <= bend; bindex++, p++)
-+ if (p->new)
-+ fput(p->new);
-+ else
++ SiMustWriteLock(sb);
++
++ bbot = au_sbbot(sb);
++ inode = d_inode(sb->s_root);
++ for (bindex = 0; bindex <= bbot; bindex++) {
++ args.h_ino = au_h_iptr(inode, bindex)->i_ino;
++ args.br = au_sbr(sb, bindex);
++ args.bshared = is_sb_shared(sb, bindex, bindex - 1);
++ err = au_xino_do_set_br(sb, path, &args);
++ if (unlikely(err))
+ break;
-+ kfree(fpair);
-+ out:
++ }
++
+ AuTraceErr(err);
+ return err;
+}
+{
+ struct au_sbinfo *sbinfo;
+
-+ AuTraceEnter();
-+ SiMustWriteLock(sb);
-+
+ au_xigen_clr(sb);
+ xino_clear_xib(sb);
+ xino_clear_br(sb);
++ dbgaufs_brs_del(sb, 0);
+ sbinfo = au_sbi(sb);
+ /* lvalue, do not call au_mntflags() */
+ au_opt_clr(sbinfo->si_mntflags, XINO);
-+ au_xino_def_br_set(NULL, sbinfo);
+}
+
-+int au_xino_set(struct super_block *sb, struct au_opt_xino *xino, int remount)
++int au_xino_set(struct super_block *sb, struct au_opt_xino *xiopt, int remount)
+{
+ int err, skip;
-+ struct dentry *parent, *cur_parent;
++ struct dentry *dentry, *parent, *cur_dentry, *cur_parent;
+ struct qstr *dname, *cur_name;
+ struct file *cur_xino;
-+ struct inode *dir;
+ struct au_sbinfo *sbinfo;
++ struct path *path, *cur_path;
+
-+ LKTRTrace("remount %d\n", remount);
+ SiMustWriteLock(sb);
+
+ err = 0;
+ sbinfo = au_sbi(sb);
-+ parent = dget_parent(xino->file->f_dentry);
++ path = &xiopt->file->f_path;
++ dentry = path->dentry;
++ parent = dget_parent(dentry);
+ if (remount) {
+ skip = 0;
-+ dname = &xino->file->f_dentry->d_name;
+ cur_xino = sbinfo->si_xib;
+ if (cur_xino) {
-+ cur_parent = dget_parent(cur_xino->f_dentry);
-+ cur_name = &cur_xino->f_dentry->d_name;
++ cur_path = &cur_xino->f_path;
++ cur_dentry = cur_path->dentry;
++ cur_parent = dget_parent(cur_dentry);
++ cur_name = &cur_dentry->d_name;
++ dname = &dentry->d_name;
+ skip = (cur_parent == parent
-+ && dname->len == cur_name->len
-+ && !memcmp(dname->name, cur_name->name,
-+ dname->len));
++ && au_qstreq(dname, cur_name));
+ dput(cur_parent);
+ }
+ if (skip)
+ }
+
+ au_opt_set(sbinfo->si_mntflags, XINO);
-+ au_xino_def_br_set(NULL, sbinfo);
-+ dir = parent->d_inode;
-+ mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT);
-+ /* mnt_want_write() is unnecessary here */
-+ err = au_xino_set_xib(sb, xino->file);
++ err = au_xino_set_xib(sb, path);
++ /* si_x{read,write} are set */
+ if (!err)
-+ err = au_xigen_set(sb, xino->file);
-+ if (!err)
-+ err = au_xino_set_br(sb, xino->file);
-+ mutex_unlock(&dir->i_mutex);
++ err = au_xigen_set(sb, path);
+ if (!err)
++ err = au_xino_set_br(sb, path);
++ if (!err) {
++ dbgaufs_brs_add(sb, 0, /*topdown*/1);
+ goto out; /* success */
++ }
+
+ /* reset all */
-+ AuIOErr("failed creating xino(%d).\n", err);
-+
-+ out:
-+ dput(parent);
-+ AuTraceErr(err);
-+ return err;
-+}
-+
-+int au_xino_trunc(struct super_block *sb, aufs_bindex_t bindex)
-+{
-+ int err;
-+ struct au_branch *br;
-+ struct file *new_xino;
-+ struct super_block *h_sb;
-+ aufs_bindex_t bi, bend;
-+ struct dentry *parent;
-+ struct inode *dir;
-+
-+ LKTRTrace("b%d\n", bindex);
-+ SiMustWriteLock(sb);
-+
-+ err = -EINVAL;
-+ bend = au_sbend(sb);
-+ if (unlikely(bindex < 0 || bend < bindex))
-+ goto out;
-+ br = au_sbr(sb, bindex);
-+ if (!br->br_xino.xi_file)
-+ goto out;
++ AuIOErr("failed setting xino(%d).\n", err);
++ au_xino_clr(sb);
+
-+ parent = dget_parent(br->br_xino.xi_file->f_dentry);
-+ dir = parent->d_inode;
-+ mutex_lock_nested(&dir->i_mutex, AuLsc_I_PARENT);
-+ /* mnt_want_write() is unnecessary here */
-+ new_xino = au_xino_create2(sb, br->br_xino.xi_file,
-+ br->br_xino.xi_file);
-+ mutex_unlock(&dir->i_mutex);
++out:
+ dput(parent);
-+ err = PTR_ERR(new_xino);
-+ if (IS_ERR(new_xino))
-+ goto out;
-+ err = 0;
-+ fput(br->br_xino.xi_file);
-+ br->br_xino.xi_file = new_xino;
-+
-+ h_sb = br->br_mnt->mnt_sb;
-+ for (bi = 0; bi <= bend; bi++) {
-+ if (unlikely(bi == bindex))
-+ continue;
-+ br = au_sbr(sb, bi);
-+ if (br->br_mnt->mnt_sb != h_sb)
-+ continue;
-+
-+ fput(br->br_xino.xi_file);
-+ br->br_xino.xi_file = new_xino;
-+ get_file(new_xino);
-+ }
-+
-+ out:
-+ AuTraceErr(err);
+ return err;
+}
+
-+/* ---------------------------------------------------------------------- */
-+
+/*
+ * create a xinofile at the default place/path.
+ */
+struct file *au_xino_def(struct super_block *sb)
+{
+ struct file *file;
-+ aufs_bindex_t bend, bindex, bwr;
+ char *page, *p;
-+ struct path path;
+ struct au_branch *br;
++ struct super_block *h_sb;
++ struct path path;
++ aufs_bindex_t bbot, bindex, bwr;
+
-+ AuTraceEnter();
-+
-+ bend = au_sbend(sb);
++ br = NULL;
++ bbot = au_sbbot(sb);
+ bwr = -1;
-+ for (bindex = 0; bindex <= bend; bindex++) {
++ for (bindex = 0; bindex <= bbot; bindex++) {
+ br = au_sbr(sb, bindex);
+ if (au_br_writable(br->br_perm)
-+ && !au_test_nfs(br->br_mnt->mnt_sb)
-+ && !au_test_ecryptfs(br->br_mnt->mnt_sb)) {
++ && !au_test_fs_bad_xino(au_br_sb(br))) {
+ bwr = bindex;
+ break;
+ }
+
+ if (bwr >= 0) {
+ file = ERR_PTR(-ENOMEM);
-+ page = __getname();
++ page = (void *)__get_free_page(GFP_NOFS);
+ if (unlikely(!page))
+ goto out;
-+ path.mnt = au_sbr_mnt(sb, bwr);
++ path.mnt = au_br_mnt(br);
+ path.dentry = au_h_dptr(sb->s_root, bwr);
+ p = d_path(&path, page, PATH_MAX - sizeof(AUFS_XINO_FNAME));
+ file = (void *)p;
+ if (!IS_ERR(p)) {
+ strcat(p, "/" AUFS_XINO_FNAME);
-+ LKTRTrace("%s\n", p);
-+ file = au_xino_create(sb, p, /*silent*/0);
-+ if (!IS_ERR(file))
-+ au_xino_def_br_set(au_sbr(sb, bwr), au_sbi(sb));
++ AuDbg("%s\n", p);
++ file = au_xino_create(sb, p, /*silent*/0, /*wbrtop*/1);
+ }
-+ __putname(page);
++ free_page((unsigned long)page);
+ } else {
-+ file = au_xino_create(sb, AUFS_XINO_DEFPATH, /*silent*/0);
-+ if (unlikely(au_test_nfs(file->f_dentry->d_sb)
-+ || au_test_ecryptfs(file->f_dentry->d_sb))) {
-+ AuErr("xino or noxino option is required "
-+ "since %s is NFS\n", AUFS_XINO_DEFPATH);
++ file = au_xino_create(sb, AUFS_XINO_DEFPATH, /*silent*/0,
++ /*wbrtop*/0);
++ if (IS_ERR(file))
++ goto out;
++ h_sb = file->f_path.dentry->d_sb;
++ if (unlikely(au_test_fs_bad_xino(h_sb))) {
++ pr_err("xino doesn't support %s(%s)\n",
++ AUFS_XINO_DEFPATH, au_sbtype(h_sb));
+ fput(file);
+ file = ERR_PTR(-EINVAL);
+ }
-+ if (!IS_ERR(file))
-+ au_xino_def_br_set(NULL, au_sbi(sb));
+ }
+
-+ out:
-+ AuTraceErrPtr(file);
++out:
+ return file;
+}
-diff -urN linux/fs/Kconfig linux-aufs/fs/Kconfig
---- linux/fs/Kconfig 2009-01-05 11:42:15.000000000 +0100
-+++ linux-aufs/fs/Kconfig 2009-03-12 17:02:25.753047602 +0100
-@@ -1022,6 +1022,350 @@
- help
- If you say Y here, you can turn on debugging output from Unionfs.
-
-+config AUFS
-+ tristate "Another unionfs"
-+ help
-+ Aufs is a stackable unification filesystem such as Unionfs,
-+ which unifies several directories and provides a merged single
-+ directory.
-+ In the early days, aufs was entirely re-designed and
-+ re-implemented Unionfs Version 1.x series. After many original
-+ ideas, approaches and improvements, it becomes totally
-+ different from Unionfs while keeping the basic features.
-+ See Unionfs for the basic features.
+
-+if AUFS
-+comment "These options are for "#UTS_RELEASE
++/* ---------------------------------------------------------------------- */
+
-+choice
-+ prompt "Maximum number of branches"
-+ default AUFS_BRANCH_MAX_127
-+ help
-+ Specifies the maximum number of branches (or member directories) in a
-+ single aufs. The larger value consumes more system resources and has
-+ an impact to performance.
-+config AUFS_BRANCH_MAX_127
-+ bool "127"
-+ help
-+ Specifies the maximum number of branches (or member directories) in a
-+ single aufs. The larger value consumes more system resources and has
-+ an impact to performance.
-+config AUFS_BRANCH_MAX_511
-+ bool "511"
-+ help
-+ Specifies the maximum number of branches (or member directories) in a
-+ single aufs. The larger value consumes more system resources and has
-+ an impact to performance.
-+config AUFS_BRANCH_MAX_1023
-+ bool "1023"
-+ help
-+ Specifies the maximum number of branches (or member directories) in a
-+ single aufs. The larger value consumes more system resources and has
-+ an impact to performance.
-+# 128kb is available for kmalloc?
-+config AUFS_BRANCH_MAX_32767
-+ bool "32767"
-+ help
-+ Specifies the maximum number of branches (or member directories) in a
-+ single aufs. The larger value consumes more system resources and has
-+ an impact to performance.
-+endchoice
++/*
++ * initialize the xinofile for the specified branch @br
++ * at the place/path where @base_file indicates.
++ * test whether another branch is on the same filesystem or not,
++ * if found then share the xinofile with another branch.
++ */
++int au_xino_init_br(struct super_block *sb, struct au_branch *br, ino_t h_ino,
++ struct path *base)
++{
++ int err;
++ struct au_xino_do_set_br args = {
++ .h_ino = h_ino,
++ .br = br
++ };
++
++ args.bshared = sbr_find_shared(sb, /*btop*/0, au_sbbot(sb),
++ au_br_sb(br));
++ err = au_xino_do_set_br(sb, base, &args);
++ if (unlikely(err))
++ au_xino_put(br);
++
++ return err;
++}
++
++/* ---------------------------------------------------------------------- */
++
++/*
++ * get an unused inode number from bitmap
++ */
++ino_t au_xino_new_ino(struct super_block *sb)
++{
++ ino_t ino;
++ unsigned long *p, pindex, ul, pend;
++ struct au_sbinfo *sbinfo;
++ struct file *file;
++ int free_bit, err;
++
++ if (!au_opt_test(au_mntflags(sb), XINO))
++ return iunique(sb, AUFS_FIRST_INO);
++
++ sbinfo = au_sbi(sb);
++ mutex_lock(&sbinfo->si_xib_mtx);
++ p = sbinfo->si_xib_buf;
++ free_bit = sbinfo->si_xib_next_bit;
++ if (free_bit < page_bits && !test_bit(free_bit, p))
++ goto out; /* success */
++ free_bit = find_first_zero_bit(p, page_bits);
++ if (free_bit < page_bits)
++ goto out; /* success */
++
++ pindex = sbinfo->si_xib_last_pindex;
++ for (ul = pindex - 1; ul < ULONG_MAX; ul--) {
++ err = xib_pindex(sb, ul);
++ if (unlikely(err))
++ goto out_err;
++ free_bit = find_first_zero_bit(p, page_bits);
++ if (free_bit < page_bits)
++ goto out; /* success */
++ }
++
++ file = sbinfo->si_xib;
++ pend = vfsub_f_size_read(file) / PAGE_SIZE;
++ for (ul = pindex + 1; ul <= pend; ul++) {
++ err = xib_pindex(sb, ul);
++ if (unlikely(err))
++ goto out_err;
++ free_bit = find_first_zero_bit(p, page_bits);
++ if (free_bit < page_bits)
++ goto out; /* success */
++ }
++ BUG();
++
++out:
++ set_bit(free_bit, p);
++ sbinfo->si_xib_next_bit = free_bit + 1;
++ pindex = sbinfo->si_xib_last_pindex;
++ mutex_unlock(&sbinfo->si_xib_mtx);
++ ino = xib_calc_ino(pindex, free_bit);
++ AuDbg("i%lu\n", (unsigned long)ino);
++ return ino;
++out_err:
++ mutex_unlock(&sbinfo->si_xib_mtx);
++ AuDbg("i0\n");
++ return 0;
++}
++
++/* for s_op->delete_inode() */
++void au_xino_delete_inode(struct inode *inode, const int unlinked)
++{
++ int err;
++ unsigned int mnt_flags;
++ aufs_bindex_t bindex, bbot, bi;
++ unsigned char try_trunc;
++ struct au_iinfo *iinfo;
++ struct super_block *sb;
++ struct au_hinode *hi;
++ struct inode *h_inode;
++ struct au_branch *br;
++ struct au_xi_calc calc;
++ struct file *file;
++
++ AuDebugOn(au_is_bad_inode(inode));
++
++ sb = inode->i_sb;
++ mnt_flags = au_mntflags(sb);
++ if (!au_opt_test(mnt_flags, XINO)
++ || inode->i_ino == AUFS_ROOT_INO)
++ return;
++
++ if (unlinked) {
++ au_xigen_inc(inode);
++ au_xib_clear_bit(inode);
++ }
++
++ iinfo = au_ii(inode);
++ bindex = iinfo->ii_btop;
++ if (bindex < 0)
++ return;
+
-+config AUFS_STAT
-+ bool "Use <sysfs>/fs/aufs/stat"
-+ depends on SYSFS
-+ help
-+ Shows some statistic data via <sysfs>/fs/aufs/stat.
-+ See detail in aufs.5.
++ try_trunc = !!au_opt_test(mnt_flags, TRUNC_XINO);
++ hi = au_hinode(iinfo, bindex);
++ bbot = iinfo->ii_bbot;
++ for (; bindex <= bbot; bindex++, hi++) {
++ h_inode = hi->hi_inode;
++ if (!h_inode
++ || (!unlinked && h_inode->i_nlink))
++ continue;
+
-+comment "SYSFS and AUFS_STAT are disabled"
-+ depends on SYSFS = n
++ /* inode may not be revalidated */
++ bi = au_br_index(sb, hi->hi_id);
++ if (bi < 0)
++ continue;
+
-+config AUFS_HINOTIFY
-+ bool "Use inotify to detect actions on a branch"
-+ depends on INOTIFY
-+ help
-+ If you want to modify files on branches directly, eg. bypassing aufs,
-+ and want aufs to detect the changes of them fully, then enable this
-+ option and use 'udba=inotify' mount option.
-+ It will have a negative impact to the performance.
-+ See detail in aufs.5.
++ br = au_sbr(sb, bi);
++ au_xi_calc(sb, h_inode->i_ino, &calc);
++ file = au_xino_file(br->br_xino, calc.idx);
++ if (IS_ERR_OR_NULL(file))
++ continue;
+
-+comment "INOTIFY and AUFS_HINOTIFY are disabled"
-+ depends on INOTIFY = n
++ err = au_xino_do_write(file, &calc, /*ino*/0);
++ if (!err && try_trunc
++ && au_test_fs_trunc_xino(au_br_sb(br)))
++ xino_try_trunc(sb, br);
++ }
++}
+
-+config AUFS_EXPORT
-+ bool "NFS-exportable aufs"
-+ depends on (AUFS = y && EXPORTFS = y) || (AUFS = m && EXPORTFS)
-+ help
-+ If you want to export your mounted aufs, then enable this
-+ option. There are several requirements for this configuration.
-+ See detail in aufs.5.
++/* ---------------------------------------------------------------------- */
+
-+comment "EXPORTFS and AUFS_EXPORT are disabled"
-+ depends on EXPORTFS = n
++static int au_xinondir_find(struct au_xino *xi, ino_t h_ino)
++{
++ int found, total, i;
+
-+comment "AUFS_EXPORT is disabled since EXPORTFS is a module but AUFS"
-+ depends on EXPORTFS = m && AUFS = y
++ found = -1;
++ total = xi->xi_nondir.total;
++ for (i = 0; i < total; i++) {
++ if (xi->xi_nondir.array[i] != h_ino)
++ continue;
++ found = i;
++ break;
++ }
+
-+# automatic configuration for internal use
-+config AUFS_INO_T_64
-+ bool # never appear on the menu
-+ # typedef unsigned long/int __kernel_ino_t
-+ # alpha and s390x are int
-+ depends on 64BIT && !(ALPHA || S390)
-+ default y
++ return found;
++}
+
-+config AUFS_ROBR
-+ bool "Aufs as an readonly branch of another aufs mount"
-+ help
-+ If you want make your aufs to be a part of another aufs, then
-+ enable this option. In other words, you can specify your aufs
-+ path in 'br:' mount option for another aufs, but cannot
-+ specify 'rw' as the branch permission.
-+ It will have a negative impact to the performance.
-+ See detail in aufs.5.
++static int au_xinondir_expand(struct au_xino *xi)
++{
++ int err, sz;
++ ino_t *p;
+
-+config AUFS_DLGT
-+ bool "Delegate the internal branch access the kernel thread"
-+ help
-+ If you want aufs to delegate
-+ the internal access to the branches which is made by aufs, to
-+ the kernel thread, in order to hide the access issued by your
-+ application from your LSM or something or make your
-+ application to be traced strictly by the task I/O accounting,
-+ then enable this option and use 'dlgt' mount option.
-+ It will have a negative impact to the performance.
-+ See detail in aufs.5.
++ BUILD_BUG_ON(KMALLOC_MAX_SIZE > INT_MAX);
+
-+# automatic configuration for internal use
-+config AUFS_HIN_OR_DLGT
-+ bool # never appear on the menu
-+ depends on AUFS_HINOTIFY || AUFS_DLGT
-+ default y
-+ help
-+ Automatic configuration for internal use.
++ err = -ENOMEM;
++ sz = xi->xi_nondir.total * sizeof(ino_t);
++ if (unlikely(sz > KMALLOC_MAX_SIZE / 2))
++ goto out;
++ p = au_kzrealloc(xi->xi_nondir.array, sz, sz << 1, GFP_ATOMIC,
++ /*may_shrink*/0);
++ if (p) {
++ xi->xi_nondir.array = p;
++ xi->xi_nondir.total <<= 1;
++ AuDbg("xi_nondir.total %d\n", xi->xi_nondir.total);
++ err = 0;
++ }
+
-+config AUFS_SHWH
-+ bool "Show whiteouts"
-+ help
-+ If you want to make the whiteouts in aufs visible, then enable
-+ this option and specify 'shwh' mount option. Although it may
-+ sounds like philosophy or something, but in technically it
-+ simply shows the name of whiteout with keeping its behaviour.
++out:
++ return err;
++}
+
-+# this is not in mainstream kernel
-+config AUFS_RR_SQUASHFS
-+ bool "Make squashfs branch RR (real readonly) by default"
-+ default y
-+ help
-+ If you use squashfs or LZMA patched squashfs as an aufs branch
-+ and want to set '=rr' to it by default, then enable this
-+ configuration.
-+ 'rr' stands for real readonly and it optimizes some aspects of
-+ 'ro.'
-+ See detail in aufs.5.
++void au_xinondir_leave(struct super_block *sb, aufs_bindex_t bindex,
++ ino_t h_ino, int idx)
++{
++ struct au_xino *xi;
+
-+config AUFS_SEC_PERM_PATCH
-+ bool "sec_perm-2.6.24.patch was applied or not"
-+ depends on AUFS = m
-+ depends on SECURITY
-+ help
-+ If you build aufs as a module and enabled CONFIG_SECURITY,
-+ then you need to apply the patch
-+ 'CVS_TREE/aufs/patch/sec_perm-2.6.24.patch' to your kernel
-+ source, and enable this configuration.
-+ The sec_perm-2.6.24.patch exports a kernel function
-+ security_inode_permission() to modules.
-+
-+comment "SECURITY and AUFS_SEC_PERM_PATCH are disabled"
-+ depends on SECURITY = n
-+
-+config AUFS_SPLICE_PATCH
-+ bool "splice.patch for sendfile(2) and splice(2)"
-+ help
-+ If you use 'loopback mount' on a fs-image file, or use
-+ splice(2) or sendfile(2) systemcall in aufs, then you need to
-+ apply the patch 'CVS_TREE/aufs/patch/splice.patch' to your
-+ kernel source, and enable this configuration.
-+ The splice.patch makes the kernel function do_splice_to/from()
-+ global and exports them to modules.
-+
-+config AUFS_LHASH_PATCH
-+ bool "lhash.patch for NFS branch"
-+ depends on NFS_FS
-+ help
-+ If you use mounted NFS as an aufs branch filesystem, then you
-+ need to apply the patch 'CVS_TREE/aufs/patch/lhash.patch' (or
-+ lhash-2.6.22.patch for linux-2.6.22 and later) to your kernel
-+ source, and enable this configuration.
-+ The patch file makes the kernel function __lookup_hash() global
-+ and exports it to modules.
-+
-+comment "NFS_FS and AUFS_LHASH_PATCH are disabled"
-+ depends on NFS_FS = n
-+
-+config AUFS_PUT_FILP_PATCH
-+ bool "put_filp.patch for 'atomic open'"
-+ depends on AUFS = m && NFS_V4
-+ help
-+ If you build aufs as a module and use a filesystem which
-+ operates 'atomic open' (for instance NFSv4) as an aufs branch
-+ filesystem, then you need to apply the patch
-+ 'CVS_TREE/aufs/patch/put_filp.patch' to your kernel source,
-+ and enable this configuration.
-+ The put_filp.patch exports a kernel function put_filp() to
-+ modules.
-+
-+comment "NFS_V4 and AUFS_PUT_FILP_PATCH are disabled"
-+ depends on NFS_V4 = n
-+
-+# automatic configuration for internal use
-+config AUFS_BR_NFS
-+ bool # never appear on the menu
-+ depends on NFS_FS
-+ depends on AUFS_LHASH_PATCH
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports NFS branch, enabled automatically.
++ AuDebugOn(!au_opt_test(au_mntflags(sb), XINO));
++ xi = au_sbr(sb, bindex)->br_xino;
++ AuDebugOn(idx < 0 || xi->xi_nondir.total <= idx);
+
-+config AUFS_BR_NFS_V4
-+ bool # never appear on the menu
-+ depends on NFS_V4 && AUFS_BR_NFS
-+ depends on AUFS = y || (AUFS = m && AUFS_PUT_FILP_PATCH)
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports a branch filesystem which operates
-+ 'atomic_open', for instance NFSv4, this configuration is enabled
-+ automatically.
++ spin_lock(&xi->xi_nondir.spin);
++ AuDebugOn(xi->xi_nondir.array[idx] != h_ino);
++ xi->xi_nondir.array[idx] = 0;
++ spin_unlock(&xi->xi_nondir.spin);
++ wake_up_all(&xi->xi_nondir.wqh);
++}
+
-+config AUFS_BR_XFS
-+ bool # never appear on the menu
-+ depends on XFS_FS
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports XFS branch, enabled automatically.
++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 *xi;
+
-+config AUFS_FSYNC_SUPER_PATCH
-+ bool "fsync_super-2.6.xx.patch was applied or not"
-+ depends on AUFS = m
-+ help
-+ If you build aufs as a module and want to flush everything for
-+ branch filesystems which are not marked as 'rr' nor 'rr+wh' at
-+ umount or remount time, then you need to apply the patch
-+ 'CVS_TREE/aufs/patch/fsync_super-2.6.16.patch' or
-+ '...-2.6.19.patch' to your kernel source, and enable this
-+ configuration.
-+ It may be helpful at shutdown time in case of your aufs is a
-+ root filesystem. But this behaviour will not guarantee the
-+ consistency of branch filesystems. To guarantee it, try the
-+ approach described in the aufs manual, and do not forget
-+ installing auplink script.
-+ The fsync_super-2.6.xx.patch does nothing but exports a kernel
-+ function fsync_super() to modules.
-+
-+config AUFS_DENY_WRITE_ACCESS_PATCH
-+ bool "deny_write_access.patch was applied or not"
-+ depends on AUFS = m
-+ help
-+ A security enhancement to deny writing to a running executable
-+ which exists on an aufs branch filesystem and executed through
-+ aufs. If you applied
-+ 'CVS_TREE/aufs/patch/deny_write_access.patch' to your kernel
-+ and you are compiling aufs as a module, then enable this
-+ option.
-+ The write_deny_access.patch does nothing but export the
-+ function.
-+
-+config AUFS_WORKAROUND_FUSE
-+ bool "Special handling for FUSE-based filesystem"
-+ depends on FUSE_FS
-+ help
-+ A FUSE-based filesystem may not initialize its inode
-+ attributes and the FUSE developer thinks the inode attributes
-+ in a positive dentry which is returned by VFS lookup operation
-+ are not reliable.
-+ If you use a FUSE-based filesystem as an aufs branch, and it
-+ customizes the inode attribute on it without overriding
-+ fuse_lowlevel_ops.lookup, probably you need to enable this
-+ configuration.
-+ If you enable this configuration, aufs calls getattr operation
-+ in every lookup and revalidate operation for the FUSE-based
-+ filesystem branch.
-+ It will have a negative impact to the performance even if you do
-+ not use a FUSE-based filesystem branch.
-+
-+# automatic configuration for internal use
-+config AUFS_GETATTR
-+ bool # never appear on the menu
-+ depends on AUFS_HINOTIFY || AUFS_WORKAROUND_FUSE || AUFS_BR_NFS
-+ default y
-+ help
-+ Automatic configuration for internal use.
++ err = 0;
++ *idx = -1;
++ if (!au_opt_test(au_mntflags(sb), XINO))
++ goto out; /* no xino */
++
++ xi = au_sbr(sb, bindex)->br_xino;
++
++again:
++ spin_lock(&xi->xi_nondir.spin);
++ found = au_xinondir_find(xi, h_ino);
++ if (found == -1) {
++ empty = au_xinondir_find(xi, /*h_ino*/0);
++ if (empty == -1) {
++ empty = xi->xi_nondir.total;
++ err = au_xinondir_expand(xi);
++ if (unlikely(err))
++ goto out_unlock;
++ }
++ xi->xi_nondir.array[empty] = h_ino;
++ *idx = empty;
++ } else {
++ spin_unlock(&xi->xi_nondir.spin);
++ wait_event(xi->xi_nondir.wqh,
++ xi->xi_nondir.array[found] != h_ino);
++ goto again;
++ }
+
-+config AUFS_DEBUG
-+ bool "Debug aufs"
-+ default y
-+ help
-+ Enable this to compile aufs internal debug code.
-+ It will have a negative impact to the performance.
++out_unlock:
++ spin_unlock(&xi->xi_nondir.spin);
++out:
++ return err;
++}
+
-+# automatic configuration for internal use
-+config AUFS_MAGIC_SYSRQ
-+ bool # never appear on the menu
-+ depends on AUFS_DEBUG && MAGIC_SYSRQ
-+ default y
-+ help
-+ Automatic configuration for internal use.
-+ When aufs supports Magic SysRq, enabled automatically.
++/* ---------------------------------------------------------------------- */
+
-+config AUFS_DEBUG_LOCK
-+ bool "Show lock status in Magic SysRq"
-+ depends on AUFS_MAGIC_SYSRQ
-+ help
-+ For developers only. Sometimes it is useful when a deadlock
-+ occurs in aufs. Trace every aufs lock acquire and release, and
-+ print them when Magic SysRq-<AUFS> key is pressed.
-+ It will have a huge negative impact to the performance.
++int au_xino_path(struct seq_file *seq, struct file *file)
++{
++ int err;
+
-+config AUFS_COMPAT
-+ bool "Compatibility with Unionfs (obsolete)"
-+ help
-+ This makes aufs compatible with unionfs-style mount options and some
-+ behaviours.
-+ The dirs= mount option and =nfsro branch permission flag are always
-+ interpreted as br: mount option and =ro flag respectively. The
-+ 'debug', 'delete' and 'imap' mount options are ignored.
-+ If you disable this option, you will get,
-+ - aufs issues a warning about the ignored mount options
-+ - the default branch permission flag is set. RW for the first branch,
-+ and RO for the rests.
-+ - the name of a internal file which represents the directory is
-+ 'opaque', becomes '.wh..wh..opq'
-+ - the 'diropq=w' mount option is set by default
-+
-+config AUFS_UNIONFS22_PATCH
-+ bool "Unionfs-2.2 or later patch is applied or not (obsolete)"
-+ help
-+ Unionfs version 2.2 (and later) patch introduces some changes in VFS
-+ layer which has an impact to aufs. If you have applied such
-+ patch to your kernel, you need to enable this configuration even
-+ if you disabled CONFIG_UNIONFS.
-+
-+config AUFS_UNIONFS23_PATCH
-+ bool "Unionfs-2.3 or later patch is applied or not (obsolete)"
-+ select AUFS_SPLICE_PATCH
-+ select AUFS_UNIONFS22_PATCH
-+ help
-+ Unionfs version 2.3 (and later) patch introduces some changes in VFS
-+ layer which has an impact to aufs. If you have applied such
-+ patch to your kernel, you need to enable this configuration even
-+ if you disabled CONFIG_UNIONFS.
-+endif
++ err = au_seq_path(seq, &file->f_path);
++ if (unlikely(err))
++ goto out;
+
- endmenu
-
- menu "Miscellaneous filesystems"
-diff -urN linux/fs/Makefile linux-aufs/fs/Makefile
---- linux/fs/Makefile 2009-03-12 11:25:26.000000000 +0100
-+++ linux-aufs/fs/Makefile 2009-03-12 17:03:22.897115295 +0100
-@@ -89,6 +89,7 @@
- obj-$(CONFIG_HFS_FS) += hfs/
- obj-$(CONFIG_ECRYPT_FS) += ecryptfs/
- obj-$(CONFIG_UNION_FS) += unionfs/
-+obj-$(CONFIG_AUFS) += aufs/
- obj-$(CONFIG_VXFS_FS) += freevxfs/
- obj-$(CONFIG_NFS_FS) += nfs/
- obj-$(CONFIG_EXPORTFS) += exportfs/
-diff -urN linux/include/linux/aufs_type.h linux-aufs/include/linux/aufs_type.h
---- linux/include/linux/aufs_type.h 1970-01-01 01:00:00.000000000 +0100
-+++ linux-aufs/include/linux/aufs_type.h 2009-03-12 16:18:04.000000000 +0100
-@@ -0,0 +1,127 @@
-+/*
-+ * Copyright (C) 2005-2009 Junjiro Okajima
++#define Deleted "\\040(deleted)"
++ seq->count -= sizeof(Deleted) - 1;
++ AuDebugOn(memcmp(seq->buf + seq->count, Deleted,
++ sizeof(Deleted) - 1));
++#undef Deleted
++
++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 2023-08-28 12:34:41.669969465 +0200
+@@ -0,0 +1,452 @@
++/* SPDX-License-Identifier: GPL-2.0 WITH Linux-syscall-note */
++/*
++ * Copyright (C) 2005-2022 Junjiro R. Okajima
+ *
-+ * This program, aufs is free software; you can redistribute it and/or modify
++ * 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.
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
-+ * along with this program; if not, write to the Free Software
-+ * Foundation, Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
++ * along with this program. If not, see <http://www.gnu.org/licenses/>.
+ */
+
-+/* $Id$ */
-+
-+#include <linux/ioctl.h>
-+
+#ifndef __AUFS_TYPE_H__
+#define __AUFS_TYPE_H__
+
-+#define AUFS_VERSION "20090126"
++#define AUFS_NAME "aufs"
++
++#ifdef __KERNEL__
++/*
++ * define it before including all other headers.
++ * sched.h may use pr_* macros before defining "current", so define the
++ * no-current version first, and re-define later.
++ */
++#define pr_fmt(fmt) AUFS_NAME " %s:%d: " fmt, __func__, __LINE__
++#include <linux/sched.h>
++#undef pr_fmt
++#define pr_fmt(fmt) \
++ AUFS_NAME " %s:%d:%.*s[%d]: " fmt, __func__, __LINE__, \
++ (int)sizeof(current->comm), current->comm, current->pid
++#include <linux/limits.h>
++#else
++#include <stdint.h>
++#include <sys/types.h>
++#include <limits.h>
++#endif /* __KERNEL__ */
++
++#define AUFS_VERSION "6.x-rcN-20230828"
+
-+/* move this to linux-2.6.19/include/magic.h */
++/* todo? move this to linux-2.6.19/include/magic.h */
+#define AUFS_SUPER_MAGIC ('a' << 24 | 'u' << 16 | 'f' << 8 | 's')
+
+/* ---------------------------------------------------------------------- */
+
++#ifdef __KERNEL__
+#ifdef CONFIG_AUFS_BRANCH_MAX_127
-+/* some environments treat 'char' as 'unsigned char' by default */
-+typedef signed char aufs_bindex_t;
++typedef int8_t aufs_bindex_t;
+#define AUFS_BRANCH_MAX 127
+#else
-+typedef short aufs_bindex_t;
++typedef int16_t aufs_bindex_t;
+#ifdef CONFIG_AUFS_BRANCH_MAX_511
+#define AUFS_BRANCH_MAX 511
+#elif defined(CONFIG_AUFS_BRANCH_MAX_1023)
+#define AUFS_BRANCH_MAX 1023
+#elif defined(CONFIG_AUFS_BRANCH_MAX_32767)
+#define AUFS_BRANCH_MAX 32767
-+#else
-+#error unknown CONFIG_AUFS_BRANCH_MAX value
+#endif
+#endif
+
-+#define AUFS_NAME "aufs"
++#ifndef AUFS_BRANCH_MAX
++#error unknown CONFIG_AUFS_BRANCH_MAX value
++#endif
++#endif /* __KERNEL__ */
++
++/* ---------------------------------------------------------------------- */
++
+#define AUFS_FSTYPE AUFS_NAME
+
+#define AUFS_ROOT_INO 2
+
+#define AUFS_WH_PFX ".wh."
+#define AUFS_WH_PFX_LEN ((int)sizeof(AUFS_WH_PFX) - 1)
++#define AUFS_WH_TMP_LEN 4
++/* a limit for rmdir/rename a dir and copyup */
++#define AUFS_MAX_NAMELEN (NAME_MAX \
++ - AUFS_WH_PFX_LEN * 2 /* doubly whiteouted */\
++ - 1 /* dot */\
++ - AUFS_WH_TMP_LEN) /* hex */
+#define AUFS_XINO_FNAME "." AUFS_NAME ".xino"
+#define AUFS_XINO_DEFPATH "/tmp/" AUFS_XINO_FNAME
-+#define AUFS_XINO_TRUNC_INIT 64 /* blocks */
-+#define AUFS_XINO_TRUNC_STEP 4 /* blocks */
++#define AUFS_XINO_DEF_SEC 30 /* seconds */
++#define AUFS_XINO_DEF_TRUNC 45 /* percentage */
+#define AUFS_DIRWH_DEF 3
+#define AUFS_RDCACHE_DEF 10 /* seconds */
++#define AUFS_RDCACHE_MAX 3600 /* seconds */
++#define AUFS_RDBLK_DEF 512 /* bytes */
++#define AUFS_RDHASH_DEF 32
+#define AUFS_WKQ_NAME AUFS_NAME "d"
-+#define AUFS_NWKQ_DEF 4
-+#define AUFS_MFS_SECOND_DEF 30 /* seconds */
-+#define AUFS_PLINK_WARN 100 /* number of plinks */
++#define AUFS_MFS_DEF_SEC 30 /* seconds */
++#define AUFS_MFS_MAX_SEC 3600 /* seconds */
++#define AUFS_FHSM_CACHE_DEF_SEC 30 /* seconds */
++#define AUFS_PLINK_WARN 50 /* number of plinks in a single bucket */
++
++/* pseudo-link maintenace under /proc */
++#define AUFS_PLINK_MAINT_NAME "plink_maint"
++#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
+
-+#ifdef CONFIG_AUFS_COMPAT
-+#define AUFS_DIROPQ_NAME "__dir_opaque"
-+#else
+#define AUFS_DIROPQ_NAME AUFS_WH_PFX ".opq" /* whiteouted doubly */
-+#endif
+#define AUFS_WH_DIROPQ AUFS_WH_PFX AUFS_DIROPQ_NAME
+
+#define AUFS_BASE_NAME AUFS_WH_PFX AUFS_NAME
+#define AUFS_PLINKDIR_NAME AUFS_WH_PFX "plnk"
-+#define AUFS_TMPDIR_NAME AUFS_WH_PFX ".tmp"
++#define AUFS_ORPHDIR_NAME AUFS_WH_PFX "orph"
+
+/* doubly whiteouted */
+#define AUFS_WH_BASE AUFS_WH_PFX AUFS_BASE_NAME
+#define AUFS_WH_PLINKDIR AUFS_WH_PFX AUFS_PLINKDIR_NAME
-+#define AUFS_WH_TMPDIR AUFS_WH_PFX AUFS_TMPDIR_NAME
++#define AUFS_WH_ORPHDIR AUFS_WH_PFX AUFS_ORPHDIR_NAME
++
++/* branch permissions and attributes */
++#define AUFS_BRPERM_RW "rw"
++#define AUFS_BRPERM_RO "ro"
++#define AUFS_BRPERM_RR "rr"
++#define AUFS_BRATTR_COO_REG "coo_reg"
++#define AUFS_BRATTR_COO_ALL "coo_all"
++#define AUFS_BRATTR_FHSM "fhsm"
++#define AUFS_BRATTR_UNPIN "unpin"
++#define AUFS_BRATTR_ICEX "icex"
++#define AUFS_BRATTR_ICEX_SEC "icexsec"
++#define AUFS_BRATTR_ICEX_SYS "icexsys"
++#define AUFS_BRATTR_ICEX_TR "icextr"
++#define AUFS_BRATTR_ICEX_USR "icexusr"
++#define AUFS_BRATTR_ICEX_OTH "icexoth"
++#define AUFS_BRRATTR_WH "wh"
++#define AUFS_BRWATTR_NLWH "nolwh"
++#define AUFS_BRWATTR_MOO "moo"
++
++#define AuBrPerm_RW 1 /* writable, hardlinkable wh */
++#define AuBrPerm_RO (1 << 1) /* readonly */
++#define AuBrPerm_RR (1 << 2) /* natively readonly */
++#define AuBrPerm_Mask (AuBrPerm_RW | AuBrPerm_RO | AuBrPerm_RR)
++
++#define AuBrAttr_COO_REG (1 << 3) /* copy-up on open */
++#define AuBrAttr_COO_ALL (1 << 4)
++#define AuBrAttr_COO_Mask (AuBrAttr_COO_REG | AuBrAttr_COO_ALL)
++
++#define AuBrAttr_FHSM (1 << 5) /* file-based hsm */
++#define AuBrAttr_UNPIN (1 << 6) /* rename-able top dir of
++ branch. meaningless since
++ linux-3.18-rc1 */
++
++/* ignore error in copying XATTR */
++#define AuBrAttr_ICEX_SEC (1 << 7)
++#define AuBrAttr_ICEX_SYS (1 << 8)
++#define AuBrAttr_ICEX_TR (1 << 9)
++#define AuBrAttr_ICEX_USR (1 << 10)
++#define AuBrAttr_ICEX_OTH (1 << 11)
++#define AuBrAttr_ICEX (AuBrAttr_ICEX_SEC \
++ | AuBrAttr_ICEX_SYS \
++ | AuBrAttr_ICEX_TR \
++ | AuBrAttr_ICEX_USR \
++ | AuBrAttr_ICEX_OTH)
++
++#define AuBrRAttr_WH (1 << 12) /* whiteout-able */
++#define AuBrRAttr_Mask AuBrRAttr_WH
++
++#define AuBrWAttr_NoLinkWH (1 << 13) /* un-hardlinkable whiteouts */
++#define AuBrWAttr_MOO (1 << 14) /* move-up on open */
++#define AuBrWAttr_Mask (AuBrWAttr_NoLinkWH | AuBrWAttr_MOO)
++
++#define AuBrAttr_CMOO_Mask (AuBrAttr_COO_Mask | AuBrWAttr_MOO)
++
++/* #warning test userspace */
++#ifdef __KERNEL__
++#ifndef CONFIG_AUFS_FHSM
++#undef AuBrAttr_FHSM
++#define AuBrAttr_FHSM 0
++#endif
++#ifndef CONFIG_AUFS_XATTR
++#undef AuBrAttr_ICEX
++#define AuBrAttr_ICEX 0
++#undef AuBrAttr_ICEX_SEC
++#define AuBrAttr_ICEX_SEC 0
++#undef AuBrAttr_ICEX_SYS
++#define AuBrAttr_ICEX_SYS 0
++#undef AuBrAttr_ICEX_TR
++#define AuBrAttr_ICEX_TR 0
++#undef AuBrAttr_ICEX_USR
++#define AuBrAttr_ICEX_USR 0
++#undef AuBrAttr_ICEX_OTH
++#define AuBrAttr_ICEX_OTH 0
++#endif
++#endif
++
++/* the longest combination */
++/* AUFS_BRATTR_ICEX and AUFS_BRATTR_ICEX_TR don't affect here */
++#define AuBrPermStrSz sizeof(AUFS_BRPERM_RW \
++ "+" AUFS_BRATTR_COO_REG \
++ "+" AUFS_BRATTR_FHSM \
++ "+" AUFS_BRATTR_UNPIN \
++ "+" AUFS_BRATTR_ICEX_SEC \
++ "+" AUFS_BRATTR_ICEX_SYS \
++ "+" AUFS_BRATTR_ICEX_USR \
++ "+" AUFS_BRATTR_ICEX_OTH \
++ "+" AUFS_BRWATTR_NLWH)
++
++typedef struct {
++ char a[AuBrPermStrSz];
++} au_br_perm_str_t;
++
++static inline int au_br_writable(int brperm)
++{
++ return brperm & AuBrPerm_RW;
++}
++
++static inline int au_br_whable(int brperm)
++{
++ return brperm & (AuBrPerm_RW | AuBrRAttr_WH);
++}
++
++static inline int au_br_wh_linkable(int brperm)
++{
++ return !(brperm & AuBrWAttr_NoLinkWH);
++}
++
++static inline int au_br_cmoo(int brperm)
++{
++ return brperm & AuBrAttr_CMOO_Mask;
++}
++
++static inline int au_br_fhsm(int brperm)
++{
++ return brperm & AuBrAttr_FHSM;
++}
+
+/* ---------------------------------------------------------------------- */
+
+/* ioctl */
-+#if 0 /* reserved for future use */
+enum {
-+ AuCtlErr,
-+ AuCtlErr_Last
++ /* readdir in userspace */
++ AuCtl_RDU,
++ AuCtl_RDU_INO,
++
++ AuCtl_WBR_FD, /* pathconf wrapper */
++ AuCtl_IBUSY, /* busy inode */
++ AuCtl_MVDOWN, /* move-down */
++ AuCtl_BR, /* info about branches */
++ AuCtl_FHSM_FD /* connection for fhsm */
++};
++
++/* borrowed from linux/include/linux/kernel.h */
++#ifndef ALIGN
++#ifdef _GNU_SOURCE
++#define ALIGN(x, a) __ALIGN_MASK(x, (typeof(x))(a)-1)
++#else
++#define ALIGN(x, a) (((x) + (a) - 1) & ~((a) - 1))
++#endif
++#define __ALIGN_MASK(x, mask) (((x)+(mask))&~(mask))
++#endif
++
++/* borrowed from linux/include/linux/compiler-gcc3.h */
++#ifndef __aligned
++#define __aligned(x) __attribute__((aligned(x)))
++#endif
++
++#ifdef __KERNEL__
++#ifndef __packed
++#define __packed __attribute__((packed))
++#endif
++#endif
++
++struct au_rdu_cookie {
++ uint64_t h_pos;
++ int16_t bindex;
++ uint8_t flags;
++ uint8_t pad;
++ uint32_t generation;
++} __aligned(8);
++
++struct au_rdu_ent {
++ uint64_t ino;
++ int16_t bindex;
++ uint8_t type;
++ uint8_t nlen;
++ uint8_t wh;
++ char name[];
++} __aligned(8);
++
++static inline int au_rdu_len(int nlen)
++{
++ /* include the terminating NULL */
++ return ALIGN(sizeof(struct au_rdu_ent) + nlen + 1,
++ sizeof(uint64_t));
++}
++
++union au_rdu_ent_ul {
++ struct au_rdu_ent __user *e;
++ uint64_t ul;
+};
++
+enum {
-+ AuCtl_DIROPQ_GET, AuCtl_DIROPQ_SET,
-+ AuCtl_MOVE,
-+ AuCtl_MVDOWN,
-+
-+ /* unimplmented */
-+ AuCtl_REFRESH, AuCtl_REFRESHV,
-+ AuCtl_FLUSH_PLINK,
-+ AuCtl_CPUP,
-+ AuCtl_CPDOWN
++ AufsCtlRduV_SZ,
++ AufsCtlRduV_End
+};
+
-+struct aufs_ctl {
-+ int err;
-+ aufs_bindex_t bsrc, bdst;
-+ char *path;
++struct aufs_rdu {
++ /* input */
++ union {
++ uint64_t sz; /* AuCtl_RDU */
++ uint64_t nent; /* AuCtl_RDU_INO */
++ };
++ union au_rdu_ent_ul ent;
++ uint16_t verify[AufsCtlRduV_End];
++
++ /* input/output */
++ uint32_t blk;
++
++ /* output */
++ union au_rdu_ent_ul tail;
++ /* number of entries which were added in a single call */
++ uint64_t rent;
++ uint8_t full;
++ uint8_t shwh;
++
++ struct au_rdu_cookie cookie;
++} __aligned(8);
++
++/* ---------------------------------------------------------------------- */
++
++/* 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[];
++} __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;
++} __aligned(8);
++
++/* ---------------------------------------------------------------------- */
++
++struct aufs_ibusy {
++ uint64_t ino, h_ino;
++ int16_t bindex;
++} __aligned(8);
++
++/* ---------------------------------------------------------------------- */
++
++/* error code for move-down */
++/* the actual message strings are implemented in aufs-util.git */
++enum {
++ EAU_MVDOWN_OPAQUE = 1,
++ EAU_MVDOWN_WHITEOUT,
++ EAU_MVDOWN_UPPER,
++ EAU_MVDOWN_BOTTOM,
++ EAU_MVDOWN_NOUPPER,
++ EAU_MVDOWN_NOLOWERBR,
++ EAU_Last
++};
++
++/* flags for move-down */
++#define AUFS_MVDOWN_DMSG 1
++#define AUFS_MVDOWN_OWLOWER (1 << 1) /* overwrite lower */
++#define AUFS_MVDOWN_KUPPER (1 << 2) /* keep upper */
++#define AUFS_MVDOWN_ROLOWER (1 << 3) /* do even if lower is RO */
++#define AUFS_MVDOWN_ROLOWER_R (1 << 4) /* did on lower RO */
++#define AUFS_MVDOWN_ROUPPER (1 << 5) /* do even if upper is RO */
++#define AUFS_MVDOWN_ROUPPER_R (1 << 6) /* did on upper RO */
++#define AUFS_MVDOWN_BRID_UPPER (1 << 7) /* upper brid */
++#define AUFS_MVDOWN_BRID_LOWER (1 << 8) /* lower brid */
++#define AUFS_MVDOWN_FHSM_LOWER (1 << 9) /* find fhsm attr for lower */
++#define AUFS_MVDOWN_STFS (1 << 10) /* req. stfs */
++#define AUFS_MVDOWN_STFS_FAILED (1 << 11) /* output: stfs is unusable */
++#define AUFS_MVDOWN_BOTTOM (1 << 12) /* output: no more lowers */
++
++/* index for move-down */
++enum {
++ AUFS_MVDOWN_UPPER,
++ AUFS_MVDOWN_LOWER,
++ AUFS_MVDOWN_NARRAY
++};
++
++/*
++ * additional info of move-down
++ * number of free blocks and inodes.
++ * subset of struct kstatfs, but smaller and always 64bit.
++ */
++struct aufs_stfs {
++ uint64_t f_blocks;
++ uint64_t f_bavail;
++ uint64_t f_files;
++ uint64_t f_ffree;
+};
+
++struct aufs_stbr {
++ int16_t brid; /* optional input */
++ int16_t bindex; /* output */
++ struct aufs_stfs stfs; /* output when AUFS_MVDOWN_STFS set */
++} __aligned(8);
++
++struct aufs_mvdown {
++ uint32_t flags; /* input/output */
++ struct aufs_stbr stbr[AUFS_MVDOWN_NARRAY]; /* input/output */
++ int8_t au_errno; /* output */
++} __aligned(8);
++
++/* ---------------------------------------------------------------------- */
++
++union aufs_brinfo {
++ /* PATH_MAX may differ between kernel-space and user-space */
++ char _spacer[4096];
++ struct {
++ int16_t id;
++ int perm;
++ char path[];
++ };
++} __aligned(8);
++
++/* ---------------------------------------------------------------------- */
++
+#define AuCtlType 'A'
-+#define AUFS_CTL_DIROPQ_GET _IO(AuCtlType, AuCtl_DIROPQ_GET)
-+#define AUFS_CTL_DIROPQ_SET _IOW(AuCtlType, AuCtl_DIROPQ_SET, aufs_bindex_t)
-+#define AUFS_CTL_MOVE \
-+ _IOW(AuCtlType, AuCtl_MVDOWN, aufs_bindex_t)
-+#define AUFS_CTL_MVDOWN \
-+ _IOWR(AuCtlType, AuCtl_MVDOWN, struct aufs_ctl)
-+
-+#define AUFS_CTL_REFRESH _IO(AuCtlType, AuCtl_REFRESH)
-+#define AUFS_CTL_REFRESHV _IO(AuCtlType, AuCtl_REFRESHV)
-+#define AUFS_CTL_FLUSH_PLINK _IOR(AuCtlType, AuCtl_FLUSH_PLINK)
-+#define AUFS_CTL_CPUP _IOWR(AuCtlType, AuCtl_CPUP, struct aufs_ctl)
-+#define AUFS_CTL_CPDOWN \
-+ _IOWR(AuCtlType, AuCtl_CPDOWN, struct aufs_ctl_cp)
-+#endif
++#define AUFS_CTL_RDU _IOWR(AuCtlType, AuCtl_RDU, struct aufs_rdu)
++#define AUFS_CTL_RDU_INO _IOWR(AuCtlType, AuCtl_RDU_INO, struct aufs_rdu)
++#define AUFS_CTL_WBR_FD _IOW(AuCtlType, AuCtl_WBR_FD, \
++ struct aufs_wbr_fd)
++#define AUFS_CTL_IBUSY _IOWR(AuCtlType, AuCtl_IBUSY, struct aufs_ibusy)
++#define AUFS_CTL_MVDOWN _IOWR(AuCtlType, AuCtl_MVDOWN, \
++ struct aufs_mvdown)
++#define AUFS_CTL_BRINFO _IOW(AuCtlType, AuCtl_BR, union aufs_brinfo)
++#define AUFS_CTL_FHSM_FD _IOW(AuCtlType, AuCtl_FHSM_FD, int)
+
+#endif /* __AUFS_TYPE_H__ */
+SPDX-License-Identifier: GPL-2.0
+aufs6.x-rcN loopback patch
+
+diff --git a/drivers/block/loop.c b/drivers/block/loop.c
+index 131294601819..56016fc9c9d8 100644
+--- a/drivers/block/loop.c
++++ b/drivers/block/loop.c
+@@ -54,7 +54,7 @@ struct loop_device {
+ int lo_flags;
+ char lo_file_name[LO_NAME_SIZE];
+
+- struct file * lo_backing_file;
++ struct file *lo_backing_file, *lo_backing_virt_file;
+ struct block_device *lo_device;
+
+ gfp_t old_gfp_mask;
+@@ -510,6 +510,15 @@ static inline void loop_update_dio(struct loop_device *lo)
+ lo->use_dio);
+ }
+
++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)
+ {
+ int rc;
+@@ -567,6 +576,7 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ {
+ struct file *file = fget(arg);
+ struct file *old_file;
++ struct file *f, *virt_file = NULL, *old_virt_file;
+ int error;
+ bool partscan;
+ bool is_loop;
+@@ -590,11 +600,19 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ if (!(lo->lo_flags & LO_FLAGS_READ_ONLY))
+ goto out_err;
+
++ f = loop_real_file(file);
++ if (f) {
++ virt_file = file;
++ file = f;
++ get_file(file);
++ }
++
+ error = loop_validate_file(file, bdev);
+ if (error)
+ goto out_err;
+
+ old_file = lo->lo_backing_file;
++ old_virt_file = lo->lo_backing_virt_file;
+
+ error = -EINVAL;
+
+@@ -607,6 +625,7 @@ 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));
+@@ -629,6 +648,8 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ * dependency.
+ */
+ fput(old_file);
++ if (old_virt_file)
++ fput(old_virt_file);
+ if (partscan)
+ loop_reread_partitions(lo);
+
+@@ -642,6 +663,8 @@ static int loop_change_fd(struct loop_device *lo, struct block_device *bdev,
+ loop_global_unlock(lo, is_loop);
+ out_putf:
+ fput(file);
++ if (virt_file)
++ fput(virt_file);
+ goto done;
+ }
+
+@@ -1013,6 +1036,7 @@ static int loop_configure(struct loop_device *lo, fmode_t mode,
+ const struct loop_config *config)
+ {
+ struct file *file = fget(config->fd);
++ struct file *f, *virt_file = NULL;
+ struct inode *inode;
+ struct address_space *mapping;
+ int error;
+@@ -1028,6 +1052,13 @@ static int loop_configure(struct loop_device *lo, fmode_t mode,
+ /* This is safe, since we have a reference from open(). */
+ __module_get(THIS_MODULE);
+
++ f = loop_real_file(file);
++ if (f) {
++ virt_file = file;
++ file = f;
++ get_file(file);
++ }
++
+ /*
+ * If we don't hold exclusive handle for the device, upgrade to it
+ * here to avoid changing device under exclusive owner.
+@@ -1091,6 +1122,7 @@ static int loop_configure(struct loop_device *lo, fmode_t mode,
+ lo->use_dio = lo->lo_flags & LO_FLAGS_DIRECT_IO;
+ lo->lo_device = bdev;
+ lo->lo_backing_file = file;
++ lo->lo_backing_virt_file = virt_file;
+ lo->old_gfp_mask = mapping_gfp_mask(mapping);
+ mapping_set_gfp_mask(mapping, lo->old_gfp_mask & ~(__GFP_IO|__GFP_FS));
+
+@@ -1146,6 +1178,8 @@ static int loop_configure(struct loop_device *lo, fmode_t mode,
+ bd_abort_claiming(bdev, loop_configure);
+ out_putf:
+ fput(file);
++ if (virt_file)
++ fput(virt_file);
+ /* This is safe: open() is still holding a reference. */
+ module_put(THIS_MODULE);
+ return error;
+@@ -1154,6 +1188,7 @@ static int loop_configure(struct loop_device *lo, fmode_t mode,
+ static void __loop_clr_fd(struct loop_device *lo, bool release)
+ {
+ struct file *filp;
++ struct file *virt_filp = lo->lo_backing_virt_file;
+ gfp_t gfp = lo->old_gfp_mask;
+
+ if (test_bit(QUEUE_FLAG_WC, &lo->lo_queue->queue_flags))
+@@ -1170,6 +1205,7 @@ static void __loop_clr_fd(struct loop_device *lo, bool release)
+ spin_lock_irq(&lo->lo_lock);
+ filp = lo->lo_backing_file;
+ lo->lo_backing_file = NULL;
++ lo->lo_backing_virt_file = NULL;
+ spin_unlock_irq(&lo->lo_lock);
+
+ lo->lo_device = NULL;
+@@ -1232,6 +1268,8 @@ static void __loop_clr_fd(struct loop_device *lo, bool release)
+ * fput can take open_mutex which is usually taken before lo_mutex.
+ */
+ fput(filp);
++ if (virt_filp)
++ fput(virt_filp);
+ }
+
+ static int loop_clr_fd(struct loop_device *lo)
+diff --git a/fs/aufs/f_op.c b/fs/aufs/f_op.c
+index 1ed300047a41..ea63191d6473 100644
+--- a/fs/aufs/f_op.c
++++ b/fs/aufs/f_op.c
+@@ -309,7 +309,7 @@ static ssize_t aufs_read_iter(struct kiocb *kio, struct iov_iter *iov_iter)
+ if (IS_ERR(h_file))
+ goto out;
+
+- if (au_test_loopback_kthread()) {
++ if (0 && au_test_loopback_kthread()) {
+ au_warn_loopback(h_file->f_path.dentry->d_sb);
+ if (file->f_mapping != h_file->f_mapping) {
+ file->f_mapping = h_file->f_mapping;
+diff --git a/fs/aufs/loop.c b/fs/aufs/loop.c
+index 58043e31e5f3..e2bfae6f9d59 100644
+--- a/fs/aufs/loop.c
++++ b/fs/aufs/loop.c
+@@ -133,3 +133,19 @@ void au_loopback_fin(void)
+ symbol_put(loop_backing_file);
+ au_kfree_try_rcu(au_warn_loopback_array);
+ }
++
++/* ---------------------------------------------------------------------- */
++
++/* support the loopback block device insude aufs */
++
++struct file *aufs_real_loop(struct file *file)
++{
++ struct file *f;
++
++ BUG_ON(!au_test_aufs(file->f_path.dentry->d_sb));
++ fi_read_lock(file);
++ f = au_hf_top(file);
++ fi_read_unlock(file);
++ AuDebugOn(!f);
++ return f;
++}
+diff --git a/fs/aufs/loop.h b/fs/aufs/loop.h
+index 03d4908a6c03..34d356e181d5 100644
+--- a/fs/aufs/loop.h
++++ b/fs/aufs/loop.h
+@@ -26,6 +26,8 @@ void au_warn_loopback(struct super_block *h_sb);
+
+ int au_loopback_init(void);
+ void au_loopback_fin(void);
++
++struct file *aufs_real_loop(struct file *file);
+ #else
+ AuStub(struct file *, loop_backing_file, return NULL, struct super_block *sb)
+
+@@ -36,6 +38,8 @@ AuStubVoid(au_warn_loopback, struct super_block *h_sb)
+
+ AuStubInt0(au_loopback_init, void)
+ AuStubVoid(au_loopback_fin, void)
++
++AuStub(struct file *, aufs_real_loop, return NULL, struct file *file)
+ #endif /* BLK_DEV_LOOP */
+
+ #endif /* __KERNEL__ */
+diff --git a/fs/aufs/super.c b/fs/aufs/super.c
+index 81922d4faf54..c8a62c267d72 100644
+--- a/fs/aufs/super.c
++++ b/fs/aufs/super.c
+@@ -758,7 +758,10 @@ const struct super_operations aufs_sop = {
+ .show_options = aufs_show_options,
+ .statfs = aufs_statfs,
+ .put_super = aufs_put_super,
+- .sync_fs = aufs_sync_fs
++ .sync_fs = aufs_sync_fs,
++#ifdef CONFIG_AUFS_BDEV_LOOP
++ .real_loop = aufs_real_loop
++#endif
+ };
+
+ /* ---------------------------------------------------------------------- */
+diff --git a/include/linux/fs.h b/include/linux/fs.h
+index 1e3054e5367d..2a780fcbdc75 100644
+--- a/include/linux/fs.h
++++ b/include/linux/fs.h
+@@ -1934,6 +1934,10 @@ struct super_operations {
+ long (*free_cached_objects)(struct super_block *,
+ struct shrink_control *);
+ void (*shutdown)(struct super_block *sb);
++#if IS_ENABLED(CONFIG_BLK_DEV_LOOP) || IS_ENABLED(CONFIG_BLK_DEV_LOOP_MODULE)
++ /* and aufs */
++ struct file *(*real_loop)(struct file *);
++#endif
+ };
+
+ /*