Btrfs: make btrfs_search_forward return with nodes unlocked
authorFilipe Manana <fdmanana@suse.com>
Mon, 4 Aug 2014 18:37:21 +0000 (19:37 +0100)
committerChris Mason <clm@fb.com>
Wed, 17 Sep 2014 20:38:02 +0000 (13:38 -0700)
commitf98de9b9c07485f7e21edfd5b2b20c89d662af3c
treecf83b225c48c5f3fa32be73c25ad4c20b087d580
parent79aec2b80db1089f983d60e4a22cb95d8b939844
Btrfs: make btrfs_search_forward return with nodes unlocked

None of the uses of btrfs_search_forward() need to have the path
nodes (level >= 1) read locked, only the leaf needs to be locked
while the caller processes it. Therefore make it return a path
with all nodes unlocked, except for the leaf.

This change is motivated by the observation that during a file
fsync we repeatdly call btrfs_search_forward() and process the
returned leaf while upper nodes of the returned path (level >= 1)
are read locked, which unnecessarily blocks other tasks that want
to write to the same fs/subvol btree.
Therefore instead of modifying the fsync code to unlock all nodes
with level >= 1 immediately after calling btrfs_search_forward(),
change btrfs_search_forward() to do it, so that it benefits all
callers.

Signed-off-by: Filipe Manana <fdmanana@suse.com>
Signed-off-by: Chris Mason <clm@fb.com>
fs/btrfs/ctree.c
fs/btrfs/ioctl.c
fs/btrfs/tree-log.c
fs/btrfs/uuid-tree.c
fs/btrfs/volumes.c