From: Yan Zheng <zheng.yan@oracle.com>
To: linux-btrfs@vger.kernel.org, Chris Mason <chris.mason@oracle.com>
Cc: yanzheng@21cn.com
Subject: [PATCH 2/4] Fix booked extent race
Date: Mon, 27 Oct 2008 21:36:51 +0800 [thread overview]
Message-ID: <4905C3F3.6060707@oracle.com> (raw)
Hello,
When dropping middle part of an extent, btrfs_drop_extents truncates
the extent at first, then inserts a booked extent. Since truncation
and insertion can't be done atomically, there is a small period that
the booked extent isn't in the tree. This causes problem for functions
that search the tree for file extent item. The way to fix this is lock
the range of the booked extent before truncation.
Regards
Signed-off-by: Yan Zheng <zheng.yan@oracle.com>
---
diff -urp 2/fs/btrfs/extent_io.c 3/fs/btrfs/extent_io.c
--- 2/fs/btrfs/extent_io.c 2008-10-27 10:37:40.000000000 +0800
+++ 3/fs/btrfs/extent_io.c 2008-10-27 16:31:54.000000000 +0800
@@ -945,8 +945,12 @@ int try_lock_extent(struct extent_io_tre
err = set_extent_bit(tree, start, end, EXTENT_LOCKED, 1,
&failed_start, mask);
- if (err == -EEXIST)
+ if (err == -EEXIST) {
+ if (failed_start > start)
+ clear_extent_bit(tree, start, failed_start - 1,
+ EXTENT_LOCKED, 1, 0, mask);
return 0;
+ }
return 1;
}
EXPORT_SYMBOL(try_lock_extent);
diff -urp 2/fs/btrfs/extent-tree.c 3/fs/btrfs/extent-tree.c
--- 2/fs/btrfs/extent-tree.c 2008-10-27 15:32:39.000000000 +0800
+++ 3/fs/btrfs/extent-tree.c 2008-10-27 16:31:54.000000000 +0800
@@ -3373,18 +3373,20 @@ static int noinline relocate_data_extent
struct btrfs_root *root = BTRFS_I(reloc_inode)->root;
struct extent_map_tree *em_tree = &BTRFS_I(reloc_inode)->extent_tree;
struct extent_map *em;
+ u64 start = extent_key->objectid - offset;
+ u64 end = start + extent_key->offset - 1;
em = alloc_extent_map(GFP_NOFS);
BUG_ON(!em || IS_ERR(em));
- em->start = extent_key->objectid - offset;
+ em->start = start;
em->len = extent_key->offset;
em->block_start = extent_key->objectid;
em->bdev = root->fs_info->fs_devices->latest_bdev;
set_bit(EXTENT_FLAG_PINNED, &em->flags);
/* setup extent map to cheat btrfs_readpage */
- mutex_lock(&BTRFS_I(reloc_inode)->extent_mutex);
+ lock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
while (1) {
int ret;
spin_lock(&em_tree->lock);
@@ -3394,13 +3396,11 @@ static int noinline relocate_data_extent
free_extent_map(em);
break;
}
- btrfs_drop_extent_cache(reloc_inode, em->start,
- em->start + em->len - 1, 0);
+ btrfs_drop_extent_cache(reloc_inode, start, end, 0);
}
- mutex_unlock(&BTRFS_I(reloc_inode)->extent_mutex);
+ unlock_extent(&BTRFS_I(reloc_inode)->io_tree, start, end, GFP_NOFS);
- return relocate_inode_pages(reloc_inode, extent_key->objectid - offset,
- extent_key->offset);
+ return relocate_inode_pages(reloc_inode, start, extent_key->offset);
}
struct btrfs_ref_path {
@@ -3815,7 +3815,6 @@ next:
* the file extent item was modified by someone
* before the extent got locked.
*/
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
lock_end, GFP_NOFS);
extent_locked = 0;
@@ -3880,8 +3879,12 @@ next:
lock_start = key.offset;
lock_end = lock_start + num_bytes - 1;
} else {
- BUG_ON(lock_start != key.offset);
- BUG_ON(lock_end - lock_start + 1 < num_bytes);
+ if (lock_start > key.offset ||
+ lock_end + 1 < key.offset + num_bytes) {
+ unlock_extent(&BTRFS_I(inode)->io_tree,
+ lock_start, lock_end, GFP_NOFS);
+ extent_locked = 0;
+ }
}
if (!inode) {
@@ -3935,7 +3938,6 @@ next:
if (ordered)
btrfs_put_ordered_extent(ordered);
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
extent_locked = 1;
continue;
}
@@ -4045,7 +4047,6 @@ next:
}
if (extent_locked) {
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
lock_end, GFP_NOFS);
extent_locked = 0;
@@ -4063,7 +4064,6 @@ out:
if (inode) {
mutex_unlock(&inode->i_mutex);
if (extent_locked) {
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
unlock_extent(&BTRFS_I(inode)->io_tree, lock_start,
lock_end, GFP_NOFS);
}
@@ -4152,10 +4152,8 @@ static int noinline invalidate_extent_ca
lock_extent(&BTRFS_I(inode)->io_tree, key.offset,
key.offset + num_bytes - 1, GFP_NOFS);
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
btrfs_drop_extent_cache(inode, key.offset,
key.offset + num_bytes - 1, 1);
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
unlock_extent(&BTRFS_I(inode)->io_tree, key.offset,
key.offset + num_bytes - 1, GFP_NOFS);
cond_resched();
diff -urp 2/fs/btrfs/file.c 3/fs/btrfs/file.c
--- 2/fs/btrfs/file.c 2008-10-27 15:37:19.000000000 +0800
+++ 3/fs/btrfs/file.c 2008-10-27 16:31:54.000000000 +0800
@@ -316,7 +316,6 @@ static int noinline dirty_and_release_pa
/* step one, delete the existing extents in this range */
aligned_end = (pos + write_bytes + root->sectorsize - 1) &
~((u64)root->sectorsize - 1);
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
err = btrfs_drop_extents(trans, root, inode, start_pos,
aligned_end, aligned_end, &hint_byte);
if (err)
@@ -328,7 +327,6 @@ static int noinline dirty_and_release_pa
inline_size, pages, 0, num_pages);
btrfs_drop_extent_cache(inode, start_pos, aligned_end - 1, 0);
BUG_ON(err);
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
/*
* an ugly way to do all the prop accounting around
@@ -533,15 +531,16 @@ out:
* inline_limit is used to tell this code which offsets in the file to keep
* if they contain inline extents.
*/
-int noinline btrfs_drop_extents(struct btrfs_trans_handle *trans,
+int btrfs_drop_extents(struct btrfs_trans_handle *trans,
struct btrfs_root *root, struct inode *inode,
u64 start, u64 end, u64 inline_limit, u64 *hint_byte)
{
- u64 extent_end = 0;
+ u64 extent_end;
u64 search_start = start;
u64 leaf_start;
u64 root_gen;
u64 root_owner;
+ u64 locked_end = end;
struct extent_buffer *leaf;
struct btrfs_file_extent_item *extent;
struct btrfs_path *path;
@@ -580,6 +579,7 @@ next_slot:
bookend = 0;
found_extent = 0;
found_inline = 0;
+ extent_end = 0;
leaf_start = 0;
root_gen = 0;
root_owner = 0;
@@ -642,11 +642,6 @@ next_slot:
goto next_slot;
}
- if (found_inline) {
- u64 mask = root->sectorsize - 1;
- search_start = (extent_end + mask) & ~mask;
- } else
- search_start = extent_end;
if (end <= extent_end && start >= key.offset && found_inline) {
*hint_byte = EXTENT_MAP_INLINE;
goto out;
@@ -665,6 +660,26 @@ next_slot:
if (found_inline && start <= key.offset)
keep = 1;
}
+
+ if (bookend && found_extent && locked_end < extent_end) {
+ ret = try_lock_extent(&BTRFS_I(inode)->io_tree,
+ locked_end, extent_end - 1, GFP_NOFS);
+ if (!ret) {
+ btrfs_release_path(root, path);
+ lock_extent(&BTRFS_I(inode)->io_tree,
+ locked_end, extent_end - 1, GFP_NOFS);
+ locked_end = extent_end;
+ continue;
+ }
+ locked_end = extent_end;
+ }
+
+ if (found_inline) {
+ u64 mask = root->sectorsize - 1;
+ search_start = (extent_end + mask) & ~mask;
+ } else
+ search_start = extent_end;
+
/* truncate existing extent */
if (start > key.offset) {
u64 new_num;
@@ -787,6 +802,10 @@ next_slot:
}
out:
btrfs_free_path(path);
+ if (locked_end > end) {
+ unlock_extent(&BTRFS_I(inode)->io_tree, end, locked_end - 1,
+ GFP_NOFS);
+ }
btrfs_check_file(root, inode);
return ret;
}
diff -urp 2/fs/btrfs/inode.c 3/fs/btrfs/inode.c
--- 2/fs/btrfs/inode.c 2008-10-27 21:22:39.000000000 +0800
+++ 3/fs/btrfs/inode.c 2008-10-27 21:25:18.000000000 +0800
@@ -145,9 +145,7 @@ static int cow_file_range(struct inode *
goto out;
BUG_ON(num_bytes > btrfs_super_total_bytes(&root->fs_info->super_copy));
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
btrfs_drop_extent_cache(inode, start, start + num_bytes - 1, 0);
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
while(num_bytes > 0) {
cur_alloc_size = min(num_bytes, root->fs_info->max_extent);
@@ -163,7 +161,6 @@ static int cow_file_range(struct inode *
em->len = ins.offset;
em->block_start = ins.objectid;
em->bdev = root->fs_info->fs_devices->latest_bdev;
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
set_bit(EXTENT_FLAG_PINNED, &em->flags);
while(1) {
spin_lock(&em_tree->lock);
@@ -176,7 +173,6 @@ static int cow_file_range(struct inode *
btrfs_drop_extent_cache(inode, start,
start + ins.offset - 1, 0);
}
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
cur_alloc_size = ins.offset;
ret = btrfs_add_ordered_extent(inode, start, ins.objectid,
@@ -609,8 +605,6 @@ static int btrfs_finish_ordered_io(struc
INIT_LIST_HEAD(&list);
- mutex_lock(&BTRFS_I(inode)->extent_mutex);
-
ret = btrfs_drop_extents(trans, root, inode,
ordered_extent->file_offset,
ordered_extent->file_offset +
@@ -641,7 +635,6 @@ static int btrfs_finish_ordered_io(struc
btrfs_drop_extent_cache(inode, ordered_extent->file_offset,
ordered_extent->file_offset +
ordered_extent->len - 1, 0);
- mutex_unlock(&BTRFS_I(inode)->extent_mutex);
ins.objectid = ordered_extent->start;
ins.offset = ordered_extent->len;
reply other threads:[~2008-10-27 13:36 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=4905C3F3.6060707@oracle.com \
--to=zheng.yan@oracle.com \
--cc=chris.mason@oracle.com \
--cc=linux-btrfs@vger.kernel.org \
--cc=yanzheng@21cn.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.