linux/fs/gfs2
Mel Gorman 2457aec637 mm: non-atomically mark page accessed during page cache allocation where possible
aops->write_begin may allocate a new page and make it visible only to have
mark_page_accessed called almost immediately after.  Once the page is
visible the atomic operations are necessary which is noticable overhead
when writing to an in-memory filesystem like tmpfs but should also be
noticable with fast storage.  The objective of the patch is to initialse
the accessed information with non-atomic operations before the page is
visible.

The bulk of filesystems directly or indirectly use
grab_cache_page_write_begin or find_or_create_page for the initial
allocation of a page cache page.  This patch adds an init_page_accessed()
helper which behaves like the first call to mark_page_accessed() but may
called before the page is visible and can be done non-atomically.

The primary APIs of concern in this care are the following and are used
by most filesystems.

	find_get_page
	find_lock_page
	find_or_create_page
	grab_cache_page_nowait
	grab_cache_page_write_begin

All of them are very similar in detail to the patch creates a core helper
pagecache_get_page() which takes a flags parameter that affects its
behavior such as whether the page should be marked accessed or not.  Then
old API is preserved but is basically a thin wrapper around this core
function.

Each of the filesystems are then updated to avoid calling
mark_page_accessed when it is known that the VM interfaces have already
done the job.  There is a slight snag in that the timing of the
mark_page_accessed() has now changed so in rare cases it's possible a page
gets to the end of the LRU as PageReferenced where as previously it might
have been repromoted.  This is expected to be rare but it's worth the
filesystem people thinking about it in case they see a problem with the
timing change.  It is also the case that some filesystems may be marking
pages accessed that previously did not but it makes sense that filesystems
have consistent behaviour in this regard.

The test case used to evaulate this is a simple dd of a large file done
multiple times with the file deleted on each iterations.  The size of the
file is 1/10th physical memory to avoid dirty page balancing.  In the
async case it will be possible that the workload completes without even
hitting the disk and will have variable results but highlight the impact
of mark_page_accessed for async IO.  The sync results are expected to be
more stable.  The exception is tmpfs where the normal case is for the "IO"
to not hit the disk.

The test machine was single socket and UMA to avoid any scheduling or NUMA
artifacts.  Throughput and wall times are presented for sync IO, only wall
times are shown for async as the granularity reported by dd and the
variability is unsuitable for comparison.  As async results were variable
do to writback timings, I'm only reporting the maximum figures.  The sync
results were stable enough to make the mean and stddev uninteresting.

The performance results are reported based on a run with no profiling.
Profile data is based on a separate run with oprofile running.

async dd
                                    3.15.0-rc3            3.15.0-rc3
                                       vanilla           accessed-v2
ext3    Max      elapsed     13.9900 (  0.00%)     11.5900 ( 17.16%)
tmpfs	Max      elapsed      0.5100 (  0.00%)      0.4900 (  3.92%)
btrfs   Max      elapsed     12.8100 (  0.00%)     12.7800 (  0.23%)
ext4	Max      elapsed     18.6000 (  0.00%)     13.3400 ( 28.28%)
xfs	Max      elapsed     12.5600 (  0.00%)      2.0900 ( 83.36%)

The XFS figure is a bit strange as it managed to avoid a worst case by
sheer luck but the average figures looked reasonable.

        samples percentage
ext3       86107    0.9783  vmlinux-3.15.0-rc4-vanilla        mark_page_accessed
ext3       23833    0.2710  vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
ext3        5036    0.0573  vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
ext4       64566    0.8961  vmlinux-3.15.0-rc4-vanilla        mark_page_accessed
ext4        5322    0.0713  vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
ext4        2869    0.0384  vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
xfs        62126    1.7675  vmlinux-3.15.0-rc4-vanilla        mark_page_accessed
xfs         1904    0.0554  vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
xfs          103    0.0030  vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
btrfs      10655    0.1338  vmlinux-3.15.0-rc4-vanilla        mark_page_accessed
btrfs       2020    0.0273  vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
btrfs        587    0.0079  vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed
tmpfs      59562    3.2628  vmlinux-3.15.0-rc4-vanilla        mark_page_accessed
tmpfs       1210    0.0696  vmlinux-3.15.0-rc4-accessed-v3r25 init_page_accessed
tmpfs         94    0.0054  vmlinux-3.15.0-rc4-accessed-v3r25 mark_page_accessed

[akpm@linux-foundation.org: don't run init_page_accessed() against an uninitialised pointer]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Jan Kara <jack@suse.cz>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Cc: Theodore Ts'o <tytso@mit.edu>
Cc: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Tested-by: Prabhakar Lad <prabhakar.csengg@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2014-06-04 16:54:10 -07:00
..
Kconfig Finally eradicate CONFIG_HOTPLUG 2013-06-03 14:20:18 -07:00
Makefile GFS2: Rename ops_inode.c to inode.c 2011-05-10 13:12:49 +01:00
acl.c GFS2: inline function gfs2_set_mode 2014-03-19 15:53:52 +00:00
acl.h GFS2: Increase the max number of ACLs 2014-03-19 15:16:24 +00:00
aops.c mm: non-atomically mark page accessed during page cache allocation where possible 2014-06-04 16:54:10 -07:00
bmap.c GFS2: fs/gfs2/bmap.c: kernel-doc warning fixes 2014-05-16 09:34:21 +01:00
bmap.h GFS2: Clean up journal extent mapping 2014-03-03 13:50:12 +00:00
dentry.c gfs2: use check_submounts_and_drop() 2013-09-05 16:23:51 -04:00
dir.c GFS2: Use pr_<level> more consistently 2014-03-07 09:30:51 +00:00
dir.h GFS2: Remember directory insert point 2014-01-06 12:49:43 +00:00
export.c [readdir] constify ->actor 2013-06-29 12:57:05 +04:00
file.c GFS2: fs/gfs2/file.c: kernel-doc warning fixes 2014-05-16 09:34:49 +01:00
gfs2.h [GFS2] Remove remote lock dropping code 2008-06-27 09:39:44 +01:00
glock.c arch: Mass conversion of smp_mb__*() 2014-04-18 14:20:48 +02:00
glock.h GFS2: Don't use ENOBUFS when ENOMEM is the correct error code 2014-01-16 10:31:13 +00:00
glops.c This must be about the smallest merge window patch set ever for GFS2. 2014-06-04 08:30:10 -07:00
glops.h GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
incore.h GFS2: Prevent recovery before the local journal is set 2014-06-02 19:12:06 +01:00
inode.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
inode.h GFS2: Add atomic_open support 2013-06-14 11:17:15 +01:00
lock_dlm.c arch: Mass conversion of smp_mb__*() 2014-04-18 14:20:48 +02:00
log.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
log.h GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
lops.c GFS2: lops.c: replace 0 by NULL for pointers 2014-04-28 09:41:55 +01:00
lops.h GFS2: Move log buffer lists into transaction 2014-02-24 16:54:54 +00:00
main.c GFS2: Use pr_<level> more consistently 2014-03-07 09:30:51 +00:00
meta_io.c mm: non-atomically mark page accessed during page cache allocation where possible 2014-06-04 16:54:10 -07:00
meta_io.h GFS2: Fix address space from page function 2014-03-31 17:48:27 +01:00
ops_fstype.c GFS2: Prevent recovery before the local journal is set 2014-06-02 19:12:06 +01:00
quota.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
quota.h GFS2: Use RCU/hlist_bl based hash for quotas 2014-01-14 19:27:56 +00:00
recovery.c This must be about the smallest merge window patch set ever for GFS2. 2014-06-04 08:30:10 -07:00
recovery.h GFS2: Move recovery variables to journal structure in memory 2014-03-07 09:14:48 +00:00
rgrp.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
rgrp.h GFS2: Don't use ENOBUFS when ENOMEM is the correct error code 2014-01-16 10:31:13 +00:00
super.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
super.h GFS2: Clean up freeze code 2013-01-29 10:29:05 +00:00
sys.c This must be about the smallest merge window patch set ever for GFS2. 2014-06-04 08:30:10 -07:00
sys.h GFS2: dlm based recovery coordination 2012-01-11 09:23:05 +00:00
trace_gfs2.h GFS2: Add origin indicator to glock demote tracing 2013-04-10 10:32:05 +01:00
trans.c GFS2: remove transaction glock 2014-05-14 10:04:34 +01:00
trans.h GFS2: Split gfs2_trans_add_bh() into two 2013-01-29 10:28:04 +00:00
util.c GFS2: Convert gfs2_lm_withdraw to use fs_err 2014-03-07 09:39:18 +00:00
util.h GFS2: Convert gfs2_lm_withdraw to use fs_err 2014-03-07 09:39:18 +00:00
xattr.c gfs2: use generic posix ACL infrastructure 2014-01-25 23:58:22 -05:00
xattr.h sanitize xattr handler prototypes 2009-12-16 12:16:49 -05:00