Btrfs: fix panic when recovering tree log

A user reported a BUG_ON(ret) that occured during tree log replay.  Ret was
-EAGAIN, so what I think happened is that we removed an extent that covered
a bitmap entry and an extent entry.  We remove the part from the bitmap and
return -EAGAIN and then search for the next piece we want to remove, which
happens to be an entire extent entry, so we just free the sucker and return.
The problem is ret is still set to -EAGAIN so we trip the BUG_ON().  The
user used btrfs-zero-log so I'm not 100% sure this is what happened so I've
added a WARN_ON() to catch the other possibility.  Thanks,

Reported-by: Jan Steffens <jan.steffens@gmail.com>
Signed-off-by: Josef Bacik <jbacik@fusionio.com>
This commit is contained in:
Josef Bacik 2012-12-18 11:39:19 -05:00
parent 201a903894
commit b0175117b9

View File

@ -1862,11 +1862,13 @@ int btrfs_remove_free_space(struct btrfs_block_group_cache *block_group,
{
struct btrfs_free_space_ctl *ctl = block_group->free_space_ctl;
struct btrfs_free_space *info;
int ret = 0;
int ret;
bool re_search = false;
spin_lock(&ctl->tree_lock);
again:
ret = 0;
if (!bytes)
goto out_lock;
@ -1879,17 +1881,17 @@ again:
info = tree_search_offset(ctl, offset_to_bitmap(ctl, offset),
1, 0);
if (!info) {
/* the tree logging code might be calling us before we
* have fully loaded the free space rbtree for this
* block group. So it is possible the entry won't
* be in the rbtree yet at all. The caching code
* will make sure not to put it in the rbtree if
* the logging code has pinned it.
/*
* If we found a partial bit of our free space in a
* bitmap but then couldn't find the other part this may
* be a problem, so WARN about it.
*/
WARN_ON(re_search);
goto out_lock;
}
}
re_search = false;
if (!info->bitmap) {
unlink_free_space(ctl, info);
if (offset == info->offset) {
@ -1935,8 +1937,10 @@ again:
}
ret = remove_from_bitmap(ctl, info, &offset, &bytes);
if (ret == -EAGAIN)
if (ret == -EAGAIN) {
re_search = true;
goto again;
}
BUG_ON(ret); /* logic error */
out_lock:
spin_unlock(&ctl->tree_lock);