Btrfs: fix inode caching vs tree log

Currently, with inode cache enabled, we will reuse its inode id immediately
after unlinking file, we may hit something like following:

|->iput inode
|->return inode id into inode cache
|->create dir,fsync
|->power off

An easy way to reproduce this problem is:

mkfs.btrfs -f /dev/sdb
mount /dev/sdb /mnt -o inode_cache,commit=100
dd if=/dev/zero of=/mnt/data bs=1M count=10 oflag=sync
inode_id=`ls -i /mnt/data | awk '{print $1}'`
rm -f /mnt/data

i=1
while [ 1 ]
do
        mkdir /mnt/dir_$i
        test1=`stat /mnt/dir_$i | grep Inode: | awk '{print $4}'`
        if [ $test1 -eq $inode_id ]
        then
		dd if=/dev/zero of=/mnt/dir_$i/data bs=1M count=1 oflag=sync
		echo b > /proc/sysrq-trigger
	fi
	sleep 1
        i=$(($i+1))
done

mount /dev/sdb /mnt
umount /dev/sdb
btrfs check /dev/sdb

We fix this problem by adding unlinked inode's id into pinned tree,
and we can not reuse them until committing transaction.

Cc: stable@vger.kernel.org
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Wang Shilong <wangsl.fnst@cn.fujitsu.com>
Signed-off-by: Chris Mason <clm@fb.com>

diff --git a/fs/btrfs/inode-map.c b/fs/btrfs/inode-map.c
index 8ad529e3e67e3e2413ec2c8e602f10383691a25d..86935f5ae29162b0b3dec1ed30c34df4c93db921 100644 (file)
--- a/fs/btrfs/inode-map.c
+++ b/fs/btrfs/inode-map.c
@@ -209,24 +209,14 @@ again:
 
 void btrfs_return_ino(struct btrfs_root *root, u64 objectid)
 {
-       struct btrfs_free_space_ctl *ctl = root->free_ino_ctl;
        struct btrfs_free_space_ctl *pinned = root->free_ino_pinned;
 
        if (!btrfs_test_opt(root, INODE_MAP_CACHE))
                return;
-
 again:
        if (root->cached == BTRFS_CACHE_FINISHED) {
-               __btrfs_add_free_space(ctl, objectid, 1);
+               __btrfs_add_free_space(pinned, objectid, 1);
        } else {
-               /*
-                * If we are in the process of caching free ino chunks,
-                * to avoid adding the same inode number to the free_ino
-                * tree twice due to cross transaction, we'll leave it
-                * in the pinned tree until a transaction is committed
-                * or the caching work is done.
-                */
-
                down_write(&root->fs_info->commit_root_sem);
                spin_lock(&root->cache_lock);
                if (root->cached == BTRFS_CACHE_FINISHED) {
@@ -238,11 +228,7 @@ again:
 
                start_caching(root);
 
-               if (objectid <= root->cache_progress ||
-                   objectid >= root->highest_objectid)
-                       __btrfs_add_free_space(ctl, objectid, 1);
-               else
-                       __btrfs_add_free_space(pinned, objectid, 1);
+               __btrfs_add_free_space(pinned, objectid, 1);
 
                up_write(&root->fs_info->commit_root_sem);
        }