diff options
| author | Filipe Manana <fdmanana@suse.com> | 2016-04-26 15:36:38 +0100 |
|---|---|---|
| committer | Filipe Manana <fdmanana@suse.com> | 2016-05-13 01:59:14 +0100 |
| commit | 578def7c50f236432ba140d35bb7ca4ef0a1b20b (patch) | |
| tree | c6b2e052c0ba5b95d1659f0e95651445951426e1 /fs/btrfs/relocation.c | |
| parent | 3f9749f6e9edcf8ec569fb542efc3be35e06e84a (diff) | |
Btrfs: don't wait for unrelated IO to finish before relocation
Before the relocation process of a block group starts, it sets the block
group to readonly mode, then flushes all delalloc writes and then finally
it waits for all ordered extents to complete. This last step includes
waiting for ordered extents destinated at extents allocated in other block
groups, making us waste unecessary time.
So improve this by waiting only for ordered extents that fall into the
block group's range.
Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Josef Bacik <jbacik@fb.com>
Reviewed-by: Liu Bo <bo.li.liu@oracle.com>
Diffstat (limited to 'fs/btrfs/relocation.c')
| -rw-r--r-- | fs/btrfs/relocation.c | 4 |
1 files changed, 3 insertions, 1 deletions
diff --git a/fs/btrfs/relocation.c b/fs/btrfs/relocation.c index 08ef890deca6..30f77ed60133 100644 --- a/fs/btrfs/relocation.c +++ b/fs/btrfs/relocation.c @@ -4259,7 +4259,9 @@ int btrfs_relocate_block_group(struct btrfs_root *extent_root, u64 group_start) err = ret; goto out; } - btrfs_wait_ordered_roots(fs_info, -1); + btrfs_wait_ordered_roots(fs_info, -1, + rc->block_group->key.objectid, + rc->block_group->key.offset); while (1) { mutex_lock(&fs_info->cleaner_mutex); |