summaryrefslogtreecommitdiffstats
path: root/fs/xfs/linux-2.6/xfs_super.c
diff options
context:
space:
mode:
authorDave Chinner <dchinner@redhat.com>2011-04-08 12:45:07 +1000
committerDave Chinner <david@fromorbit.com>2011-04-08 12:45:07 +1000
commit89e4cb550a492cfca038a555fcc1bdac58822ec3 (patch)
treeab688a1849d6361c92b9f60ae0586045908010da /fs/xfs/linux-2.6/xfs_super.c
parentc6d09b666de11eb272326a6eb6cd3246da571014 (diff)
downloadlinux-89e4cb550a492cfca038a555fcc1bdac58822ec3.tar.bz2
xfs: convert ENOSPC inode flushing to use new syncd workqueue
On of the problems with the current inode flush at ENOSPC is that we queue a flush per ENOSPC event, regardless of how many are already queued. Thi can result in hundreds of queued flushes, most of which simply burn CPU scanned and do no real work. This simply slows down allocation at ENOSPC. We really only need one active flush at a time, and we can easily implement that via the new xfs_syncd_wq. All we need to do is queue a flush if one is not already active, then block waiting for the currently active flush to complete. The result is that we only ever have a single ENOSPC inode flush active at a time and this greatly reduces the overhead of ENOSPC processing. On my 2p test machine, this results in tests exercising ENOSPC conditions running significantly faster - 042 halves execution time, 083 drops from 60s to 5s, etc - while not introducing test regressions. This allows us to remove the old xfssyncd threads and infrastructure as they are no longer used. Signed-off-by: Dave Chinner <dchinner@redhat.com> Reviewed-by: Christoph Hellwig <hch@lst.de> Reviewed-by: Alex Elder <aelder@sgi.com>
Diffstat (limited to 'fs/xfs/linux-2.6/xfs_super.c')
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c2
1 files changed, 0 insertions, 2 deletions
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index c71b6ed45e41..ee0e981aa9d1 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -1480,8 +1480,6 @@ xfs_fs_fill_super(
spin_lock_init(&mp->m_sb_lock);
mutex_init(&mp->m_growlock);
atomic_set(&mp->m_active_trans, 0);
- INIT_LIST_HEAD(&mp->m_sync_list);
- spin_lock_init(&mp->m_sync_lock);
mp->m_super = sb;
sb->s_fs_info = mp;