diff options
| author | Josef Bacik <jbacik@fb.com> | 2016-04-12 00:37:40 +0300 |
|---|---|---|
| committer | Chris Mason <clm@fb.com> | 2016-06-23 03:54:18 +0300 |
| commit | 31b9655f439a26856edca0f3f8daa368a61f16d5 (patch) | |
| tree | ac17462f6647fa2740ba9854ffeff250917db675 /fs/btrfs/inode.c | |
| parent | de18c165509edb36bde42ca6ff930854f85e21ea (diff) | |
| download | linux-31b9655f439a26856edca0f3f8daa368a61f16d5.tar.xz | |
Btrfs: track transid for delayed ref flushing
Using the offwakecputime bpf script I noticed most of our time was spent waiting
on the delayed ref throttling. This is what is supposed to happen, but
sometimes the transaction can commit and then we're waiting for throttling that
doesn't matter anymore. So change this stuff to be a little smarter by tracking
the transid we were in when we initiated the throttling. If the transaction we
get is different then we can just bail out. This resulted in a 50% speedup in
my fs_mark test, and reduced the amount of time spent throttling by 60 seconds
over the entire run (which is about 30 minutes). Thanks,
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Chris Mason <clm@fb.com>
Diffstat (limited to 'fs/btrfs/inode.c')
| -rw-r--r-- | fs/btrfs/inode.c | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index bb62418b8023..78582e339f33 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -4558,6 +4558,7 @@ delete: BUG_ON(ret); if (btrfs_should_throttle_delayed_refs(trans, root)) btrfs_async_run_delayed_refs(root, + trans->transid, trans->delayed_ref_updates * 2, 0); if (be_nice) { if (truncate_space_check(trans, root, |