linux/fs/gfs2
Darrick J. Wong 1d1d1a7672 mm: only enforce stable page writes if the backing device requires it
Create a helper function to check if a backing device requires stable
page writes and, if so, performs the necessary wait.  Then, make it so
that all points in the memory manager that handle making pages writable
use the helper function.  This should provide stable page write support
to most filesystems, while eliminating unnecessary waiting for devices
that don't require the feature.

Before this patchset, all filesystems would block, regardless of whether
or not it was necessary.  ext3 would wait, but still generate occasional
checksum errors.  The network filesystems were left to do their own
thing, so they'd wait too.

After this patchset, all the disk filesystems except ext3 and btrfs will
wait only if the hardware requires it.  ext3 (if necessary) snapshots
pages instead of blocking, and btrfs provides its own bdi so the mm will
never wait.  Network filesystems haven't been touched, so either they
provide their own stable page guarantees or they don't block at all.
The blocking behavior is back to what it was before 3.0 if you don't
have a disk requiring stable page writes.

Here's the result of using dbench to test latency on ext2:

3.8.0-rc3:
 Operation      Count    AvgLat    MaxLat
 ----------------------------------------
 WriteX        109347     0.028    59.817
 ReadX         347180     0.004     3.391
 Flush          15514    29.828   287.283

Throughput 57.429 MB/sec  4 clients  4 procs  max_latency=287.290 ms

3.8.0-rc3 + patches:
 WriteX        105556     0.029     4.273
 ReadX         335004     0.005     4.112
 Flush          14982    30.540   298.634

Throughput 55.4496 MB/sec  4 clients  4 procs  max_latency=298.650 ms

As you can see, the maximum write latency drops considerably with this
patch enabled.  The other filesystems (ext3/ext4/xfs/btrfs) behave
similarly, but see the cover letter for those results.

Signed-off-by: Darrick J. Wong <darrick.wong@oracle.com>
Acked-by: Steven Whitehouse <swhiteho@redhat.com>
Reviewed-by: Jan Kara <jack@suse.cz>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Artem Bityutskiy <dedekind1@gmail.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Mark Fasheh <mfasheh@suse.com>
Cc: Jens Axboe <axboe@kernel.dk>
Cc: Eric Van Hensbergen <ericvh@gmail.com>
Cc: Ron Minnich <rminnich@sandia.gov>
Cc: Latchesar Ionkov <lucho@ionkov.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2013-02-21 17:22:19 -08:00
..
Kconfig GFS2: use depends instead of select in kconfig 2012-03-26 09:18:02 +01:00
Makefile GFS2: Rename ops_inode.c to inode.c 2011-05-10 13:12:49 +01:00
acl.c userns: Pass a userns parameter into posix_acl_to_xattr and posix_acl_from_xattr 2012-09-18 01:01:35 -07:00
acl.h fs: take the ACL checks to common code 2011-07-25 14:30:23 -04:00
aops.c GFS2: Use ->writepages for ordered writes 2013-01-29 10:29:17 +00:00
bmap.c GFS2: Get a block reservation before resizing a file 2013-02-01 20:37:33 +00:00
bmap.h GFS2: New truncate sequence 2010-09-20 11:18:16 +01:00
dentry.c stop passing nameidata * to ->d_revalidate() 2012-07-14 16:34:14 +04:00
dir.c GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
dir.h GFS2: Add readahead to sequential directory traversal 2011-11-08 09:52:12 +00:00
export.c tmpfs,ceph,gfs2,isofs,reiserfs,xfs: fix fh_len checking 2012-10-09 23:33:55 -04:00
file.c mm: only enforce stable page writes if the backing device requires it 2013-02-21 17:22:19 -08:00
gfs2.h
glock.c GFS2: Split glock lru processing into two parts 2013-02-01 20:36:03 +00:00
glock.h GFS2: Review bug traps in glops.c 2012-11-07 13:31:07 +00:00
glops.c GFS2: only use lvb on glocks that need it 2012-11-15 10:16:59 +00:00
glops.h GFS2: Fix AIL flush issue during fsync 2011-10-21 12:39:41 +01:00
incore.h GFS2: Reinstate withdraw ack system 2013-02-13 12:21:40 +00:00
inode.c GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
inode.h GFS2: Remove unused argument from gfs2_internal_read 2012-04-24 16:44:37 +01:00
lock_dlm.c GFS2: fix skip unlock condition 2013-01-28 09:49:15 +00:00
log.c GFS2: Use ->writepages for ordered writes 2013-01-29 10:29:17 +00:00
log.h GFS2: Use ->writepages for ordered writes 2013-01-29 10:29:17 +00:00
lops.c GFS2: Copy gfs2_trans_add_bh into new data/meta functions 2013-01-29 10:28:28 +00:00
lops.h GFS2: Merge gfs2_attach_bufdata() into trans.c 2013-01-29 10:28:44 +00:00
main.c GFS2: Fold quota data into the reservations struct 2012-06-06 11:20:22 +01:00
meta_io.c GFS2: Merge gfs2_attach_bufdata() into trans.c 2013-01-29 10:28:44 +00:00
meta_io.h GFS2: Merge gfs2_attach_bufdata() into trans.c 2013-01-29 10:28:44 +00:00
ops_fstype.c GFS2: Reinstate withdraw ack system 2013-02-13 12:21:40 +00:00
quota.c GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
quota.h quota: Split dquot_quota_sync() to writeback and cache flushing part 2012-07-22 23:58:19 +04:00
recovery.c GFS2: fail mount if journal recovery fails 2012-01-11 09:24:48 +00:00
recovery.h
rgrp.c GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
rgrp.h GFS2: Add Orlov allocator 2012-11-07 13:33:17 +00:00
super.c GFS2: Use ->writepages for ordered writes 2013-01-29 10:29:17 +00:00
super.h GFS2: Clean up freeze code 2013-01-29 10:29:05 +00:00
sys.c GFS2: Reinstate withdraw ack system 2013-02-13 12:21:40 +00:00
sys.h GFS2: dlm based recovery coordination 2012-01-11 09:23:05 +00:00
trace_gfs2.h GFS2: don't reference inode's glock during block allocation trace 2012-11-16 14:21:48 +00:00
trans.c GFS2: Use ->writepages for ordered writes 2013-01-29 10:29:17 +00:00
trans.h GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
util.c GFS2: Reinstate withdraw ack system 2013-02-13 12:21:40 +00:00
util.h GFS2: Fix error handling when reading an invalid block from the journal 2012-06-06 11:27:49 +01:00
xattr.c GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
xattr.h