Btrfs: hash_lock is no longer needed
Before metadata is written to disk, it is updated to reflect that writeout
has begun. Once this update is done, the block must be cow'd before it
can be modified again.
This update was originally synchronized by using a per-fs spinlock. Today
the buffers for the metadata blocks are locked before writeout begins,
and everyone that tests the flag has the buffer locked as well.
So, the per-fs spinlock (called hash_lock for no good reason) is no
longer required.
Signed-off-by: Chris Mason <chris.mason@oracle.com>
diff --git a/fs/btrfs/ctree.c b/fs/btrfs/ctree.c
index 2603ee5..3b6e35a 100644
--- a/fs/btrfs/ctree.c
+++ b/fs/btrfs/ctree.c
@@ -388,16 +388,14 @@
WARN_ON(1);
}
- spin_lock(&root->fs_info->hash_lock);
if (btrfs_header_generation(buf) == trans->transid &&
btrfs_header_owner(buf) == root->root_key.objectid &&
!btrfs_header_flag(buf, BTRFS_HEADER_FLAG_WRITTEN)) {
*cow_ret = buf;
- spin_unlock(&root->fs_info->hash_lock);
WARN_ON(prealloc_dest);
return 0;
}
- spin_unlock(&root->fs_info->hash_lock);
+
search_start = buf->start & ~((u64)(1024 * 1024 * 1024) - 1);
ret = __btrfs_cow_block(trans, root, buf, parent,
parent_slot, cow_ret, search_start, 0,
@@ -1376,14 +1374,11 @@
int wret;
/* is a cow on this block not required */
- spin_lock(&root->fs_info->hash_lock);
if (btrfs_header_generation(b) == trans->transid &&
btrfs_header_owner(b) == root->root_key.objectid &&
!btrfs_header_flag(b, BTRFS_HEADER_FLAG_WRITTEN)) {
- spin_unlock(&root->fs_info->hash_lock);
goto cow_done;
}
- spin_unlock(&root->fs_info->hash_lock);
/* ok, we have to cow, is our old prealloc the right
* size?