Commit f959492f authored by Filipe Manana's avatar Filipe Manana Committed by Chris Mason

Btrfs: send, fix more issues related to directory renames

This is a continuation of the previous changes titled:

   Btrfs: fix incremental send's decision to delay a dir move/rename
   Btrfs: part 2, fix incremental send's decision to delay a dir move/rename

There's a few more cases where a directory rename/move must be delayed which was
previously overlooked. If our immediate ancestor has a lower inode number than
ours and it doesn't have a delayed rename/move operation associated to it, it
doesn't mean there isn't any non-direct ancestor of our current inode that needs
to be renamed/moved before our current inode (i.e. with a higher inode number
than ours).

So we can't stop the search if our immediate ancestor has a lower inode number than
ours, we need to navigate the directory hierarchy upwards until we hit the root or:

1) find an ancestor with an higher inode number that was renamed/moved in the send
   root too (or already has a pending rename/move registered);
2) find an ancestor that is a new directory (higher inode number than ours and
   exists only in the send root).

Reproducer for case 1)

    $ mkfs.btrfs -f /dev/sdd
    $ mount /dev/sdd /mnt

    $ mkdir -p /mnt/a/b
    $ mkdir -p /mnt/a/c/d
    $ mkdir /mnt/a/b/e
    $ mkdir /mnt/a/c/d/f
    $ mv /mnt/a/b /mnt/a/c/d/2b
    $ mkdir /mnt/a/x
    $ mkdir /mnt/a/y

    $ btrfs subvolume snapshot -r /mnt /mnt/snap1
    $ btrfs send /mnt/snap1 -f /tmp/base.send

    $ mv /mnt/a/x /mnt/a/y
    $ mv /mnt/a/c/d/2b/e /mnt/a/c/d/2b/2e
    $ mv /mnt/a/c/d /mnt/a/h/2d
    $ mv /mnt/a/c /mnt/a/h/2d/2b/2c

    $ btrfs subvolume snapshot -r /mnt /mnt/snap2
    $ btrfs send -p /mnt/snap1 /mnt/snap2 -f /tmp/incremental.send

Simple reproducer for case 2)

    $ mkfs.btrfs -f /dev/sdd
    $ mount /dev/sdd /mnt

    $ mkdir -p /mnt/a/b
    $ mkdir /mnt/a/c
    $ mv /mnt/a/b /mnt/a/c/b2
    $ mkdir /mnt/a/e

    $ btrfs subvolume snapshot -r /mnt /mnt/snap1
    $ btrfs send /mnt/snap1 -f /tmp/base.send

    $ mv /mnt/a/c/b2 /mnt/a/e/b3
    $ mkdir /mnt/a/e/b3/f
    $ mkdir /mnt/a/h
    $ mv /mnt/a/c /mnt/a/e/b3/f/c2
    $ mv /mnt/a/e /mnt/a/h/e2

    $ btrfs subvolume snapshot -r /mnt /mnt/snap2
    $ btrfs send -p /mnt/snap1 /mnt/snap2 -f /tmp/incremental.send

Another simple reproducer for case 2)

    $ mkfs.btrfs -f /dev/sdd
    $ mount /dev/sdd /mnt

    $ mkdir -p /mnt/a/b
    $ mkdir /mnt/a/c
    $ mkdir /mnt/a/b/d
    $ mkdir /mnt/a/c/e

    $ btrfs subvolume snapshot -r /mnt /mnt/snap1
    $ btrfs send /mnt/snap1 -f /tmp/base.send

    $ mkdir /mnt/a/b/d/f
    $ mkdir /mnt/a/b/g
    $ mv /mnt/a/c/e /mnt/a/b/g/e2
    $ mv /mnt/a/c /mnt/a/b/d/f/c2
    $ mv /mnt/a/b/d/f /mnt/a/b/g/e2/f2

    $ btrfs subvolume snapshot -r /mnt /mnt/snap2
    $ btrfs send -p /mnt/snap1 /mnt/snap2 -f /tmp/incremental.send

More complex reproducer for case 2)

    $ mkfs.btrfs -f /dev/sdd
    $ mount /dev/sdd /mnt

    $ mkdir -p /mnt/a/b
    $ mkdir -p /mnt/a/c/d
    $ mkdir /mnt/a/b/e
    $ mkdir /mnt/a/c/d/f
    $ mv /mnt/a/b /mnt/a/c/d/2b
    $ mkdir /mnt/a/x
    $ mkdir /mnt/a/y

    $ btrfs subvolume snapshot -r /mnt /mnt/snap1
    $ btrfs send /mnt/snap1 -f /tmp/base.send

    $ mv /mnt/a/x /mnt/a/y
    $ mv /mnt/a/c/d/2b/e /mnt/a/c/d/2b/2e
    $ mv /mnt/a/c/d /mnt/a/h/2d
    $ mv /mnt/a/c /mnt/a/h/2d/2b/2c

    $ btrfs subvolume snapshot -r /mnt /mnt/snap2
    $ btrfs send -p /mnt/snap1 /mnt/snap2 -f /tmp/incremental.send

For both cases the incremental send would enter an infinite loop when building
path strings.

While solving these cases, this change also re-implements the code to detect
when directory moves/renames should be delayed. Instead of dealing with several
specific cases separately, it's now more generic handling all cases with a simple
detection algorithm and if when applying a delayed move/rename there's a path loop
detected, it further delays the move/rename registering a new ancestor inode as
the dependency inode (so our rename happens after that ancestor is renamed).

Tests for these cases is being added to xfstests too.
Signed-off-by: default avatarFilipe David Borba Manana <fdmanana@gmail.com>
Signed-off-by: default avatarChris Mason <clm@fb.com>
parent a10c4076
...@@ -2940,7 +2940,9 @@ static void free_waiting_dir_move(struct send_ctx *sctx, ...@@ -2940,7 +2940,9 @@ static void free_waiting_dir_move(struct send_ctx *sctx,
static int add_pending_dir_move(struct send_ctx *sctx, static int add_pending_dir_move(struct send_ctx *sctx,
u64 ino, u64 ino,
u64 ino_gen, u64 ino_gen,
u64 parent_ino) u64 parent_ino,
struct list_head *new_refs,
struct list_head *deleted_refs)
{ {
struct rb_node **p = &sctx->pending_dir_moves.rb_node; struct rb_node **p = &sctx->pending_dir_moves.rb_node;
struct rb_node *parent = NULL; struct rb_node *parent = NULL;
...@@ -2972,12 +2974,12 @@ static int add_pending_dir_move(struct send_ctx *sctx, ...@@ -2972,12 +2974,12 @@ static int add_pending_dir_move(struct send_ctx *sctx,
} }
} }
list_for_each_entry(cur, &sctx->deleted_refs, list) { list_for_each_entry(cur, deleted_refs, list) {
ret = dup_ref(cur, &pm->update_refs); ret = dup_ref(cur, &pm->update_refs);
if (ret < 0) if (ret < 0)
goto out; goto out;
} }
list_for_each_entry(cur, &sctx->new_refs, list) { list_for_each_entry(cur, new_refs, list) {
ret = dup_ref(cur, &pm->update_refs); ret = dup_ref(cur, &pm->update_refs);
if (ret < 0) if (ret < 0)
goto out; goto out;
...@@ -3020,6 +3022,48 @@ static struct pending_dir_move *get_pending_dir_moves(struct send_ctx *sctx, ...@@ -3020,6 +3022,48 @@ static struct pending_dir_move *get_pending_dir_moves(struct send_ctx *sctx,
return NULL; return NULL;
} }
static int path_loop(struct send_ctx *sctx, struct fs_path *name,
u64 ino, u64 gen, u64 *ancestor_ino)
{
int ret = 0;
u64 parent_inode = 0;
u64 parent_gen = 0;
u64 start_ino = ino;
*ancestor_ino = 0;
while (ino != BTRFS_FIRST_FREE_OBJECTID) {
fs_path_reset(name);
if (is_waiting_for_rm(sctx, ino))
break;
if (is_waiting_for_move(sctx, ino)) {
if (*ancestor_ino == 0)
*ancestor_ino = ino;
ret = get_first_ref(sctx->parent_root, ino,
&parent_inode, &parent_gen, name);
} else {
ret = __get_cur_name_and_parent(sctx, ino, gen,
&parent_inode,
&parent_gen, name);
if (ret > 0) {
ret = 0;
break;
}
}
if (ret < 0)
break;
if (parent_inode == start_ino) {
ret = 1;
if (*ancestor_ino == 0)
*ancestor_ino = ino;
break;
}
ino = parent_inode;
gen = parent_gen;
}
return ret;
}
static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
{ {
struct fs_path *from_path = NULL; struct fs_path *from_path = NULL;
...@@ -3031,6 +3075,7 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) ...@@ -3031,6 +3075,7 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
struct waiting_dir_move *dm = NULL; struct waiting_dir_move *dm = NULL;
u64 rmdir_ino = 0; u64 rmdir_ino = 0;
int ret; int ret;
u64 ancestor = 0;
name = fs_path_alloc(); name = fs_path_alloc();
from_path = fs_path_alloc(); from_path = fs_path_alloc();
...@@ -3057,11 +3102,25 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm) ...@@ -3057,11 +3102,25 @@ static int apply_dir_move(struct send_ctx *sctx, struct pending_dir_move *pm)
if (ret < 0) if (ret < 0)
goto out; goto out;
sctx->send_progress = sctx->cur_ino + 1;
ret = path_loop(sctx, name, pm->ino, pm->gen, &ancestor);
if (ret) {
LIST_HEAD(deleted_refs);
ASSERT(ancestor > BTRFS_FIRST_FREE_OBJECTID);
ret = add_pending_dir_move(sctx, pm->ino, pm->gen, ancestor,
&pm->update_refs, &deleted_refs);
if (ret < 0)
goto out;
if (rmdir_ino) {
dm = get_waiting_dir_move(sctx, pm->ino);
ASSERT(dm);
dm->rmdir_ino = rmdir_ino;
}
goto out;
}
fs_path_reset(name); fs_path_reset(name);
to_path = name; to_path = name;
name = NULL; name = NULL;
sctx->send_progress = sctx->cur_ino + 1;
ret = get_cur_path(sctx, pm->ino, pm->gen, to_path); ret = get_cur_path(sctx, pm->ino, pm->gen, to_path);
if (ret < 0) if (ret < 0)
goto out; goto out;
...@@ -3185,127 +3244,74 @@ static int apply_children_dir_moves(struct send_ctx *sctx) ...@@ -3185,127 +3244,74 @@ static int apply_children_dir_moves(struct send_ctx *sctx)
static int wait_for_parent_move(struct send_ctx *sctx, static int wait_for_parent_move(struct send_ctx *sctx,
struct recorded_ref *parent_ref) struct recorded_ref *parent_ref)
{ {
int ret; int ret = 0;
u64 ino = parent_ref->dir; u64 ino = parent_ref->dir;
u64 parent_ino_before, parent_ino_after; u64 parent_ino_before, parent_ino_after;
u64 old_gen;
struct fs_path *path_before = NULL; struct fs_path *path_before = NULL;
struct fs_path *path_after = NULL; struct fs_path *path_after = NULL;
int len1, len2; int len1, len2;
int register_upper_dirs;
u64 gen;
if (is_waiting_for_move(sctx, ino))
return 1;
if (parent_ref->dir <= sctx->cur_ino)
return 0;
ret = get_inode_info(sctx->parent_root, ino, NULL, &old_gen,
NULL, NULL, NULL, NULL);
if (ret == -ENOENT)
return 0;
else if (ret < 0)
return ret;
if (parent_ref->dir_gen != old_gen)
return 0;
path_before = fs_path_alloc();
if (!path_before)
return -ENOMEM;
ret = get_first_ref(sctx->parent_root, ino, &parent_ino_before,
NULL, path_before);
if (ret == -ENOENT) {
ret = 0;
goto out;
} else if (ret < 0) {
goto out;
}
path_after = fs_path_alloc(); path_after = fs_path_alloc();
if (!path_after) { path_before = fs_path_alloc();
if (!path_after || !path_before) {
ret = -ENOMEM; ret = -ENOMEM;
goto out; goto out;
} }
ret = get_first_ref(sctx->send_root, ino, &parent_ino_after,
&gen, path_after);
if (ret == -ENOENT) {
ret = 0;
goto out;
} else if (ret < 0) {
goto out;
}
len1 = fs_path_len(path_before);
len2 = fs_path_len(path_after);
if (parent_ino_before != parent_ino_after || len1 != len2 ||
memcmp(path_before->start, path_after->start, len1)) {
ret = 1;
goto out;
}
ret = 0;
/* /*
* Ok, our new most direct ancestor has a higher inode number but * Our current directory inode may not yet be renamed/moved because some
* wasn't moved/renamed. So maybe some of the new ancestors higher in * ancestor (immediate or not) has to be renamed/moved first. So find if
* the hierarchy have an higher inode number too *and* were renamed * such ancestor exists and make sure our own rename/move happens after
* or moved - in this case we need to wait for the ancestor's rename * that ancestor is processed.
* or move operation before we can do the move/rename for the current
* inode.
*/ */
register_upper_dirs = 0; while (ino > BTRFS_FIRST_FREE_OBJECTID) {
ino = parent_ino_after; if (is_waiting_for_move(sctx, ino)) {
again: ret = 1;
while ((ret == 0 || register_upper_dirs) && ino > sctx->cur_ino) { break;
u64 parent_gen; }
fs_path_reset(path_before); fs_path_reset(path_before);
fs_path_reset(path_after); fs_path_reset(path_after);
ret = get_first_ref(sctx->send_root, ino, &parent_ino_after, ret = get_first_ref(sctx->send_root, ino, &parent_ino_after,
&parent_gen, path_after); NULL, path_after);
if (ret < 0) if (ret < 0)
goto out; goto out;
ret = get_first_ref(sctx->parent_root, ino, &parent_ino_before, ret = get_first_ref(sctx->parent_root, ino, &parent_ino_before,
NULL, path_before); NULL, path_before);
if (ret == -ENOENT) { if (ret < 0 && ret != -ENOENT) {
ret = 0;
break;
} else if (ret < 0) {
goto out; goto out;
} else if (ret == -ENOENT) {
ret = 1;
break;
} }
len1 = fs_path_len(path_before); len1 = fs_path_len(path_before);
len2 = fs_path_len(path_after); len2 = fs_path_len(path_after);
if (parent_ino_before != parent_ino_after || len1 != len2 || if (ino > sctx->cur_ino &&
memcmp(path_before->start, path_after->start, len1)) { (parent_ino_before != parent_ino_after || len1 != len2 ||
memcmp(path_before->start, path_after->start, len1))) {
ret = 1; ret = 1;
if (register_upper_dirs) {
break; break;
} else {
register_upper_dirs = 1;
ino = parent_ref->dir;
gen = parent_ref->dir_gen;
goto again;
} }
} else if (register_upper_dirs) {
ret = add_pending_dir_move(sctx, ino, gen,
parent_ino_after);
if (ret < 0 && ret != -EEXIST)
goto out;
}
ino = parent_ino_after; ino = parent_ino_after;
gen = parent_gen;
} }
out: out:
fs_path_free(path_before); fs_path_free(path_before);
fs_path_free(path_after); fs_path_free(path_after);
if (ret == 1) {
ret = add_pending_dir_move(sctx,
sctx->cur_ino,
sctx->cur_inode_gen,
ino,
&sctx->new_refs,
&sctx->deleted_refs);
if (!ret)
ret = 1;
}
return ret; return ret;
} }
...@@ -3466,10 +3472,6 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino); ...@@ -3466,10 +3472,6 @@ verbose_printk("btrfs: process_recorded_refs %llu\n", sctx->cur_ino);
if (ret < 0) if (ret < 0)
goto out; goto out;
if (ret) { if (ret) {
ret = add_pending_dir_move(sctx,
sctx->cur_ino,
sctx->cur_inode_gen,
cur->dir);
*pending_move = 1; *pending_move = 1;
} else { } else {
ret = send_rename(sctx, valid_path, ret = send_rename(sctx, valid_path,
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment