linux-toradex.git/mm/compaction.c, branch v4.10

mm, compaction: allow compaction for GFP_NOFS requests

2016-12-15T00:04:07+00:00

compaction has been disabled for GFP_NOFS and GFP_NOIO requests since
the direct compaction was introduced by commit 56de7263fcf3 ("mm:
compaction: direct compact when a high-order allocation fails").  The
main reason is that the migration of page cache pages might recurse back
to fs/io layer and we could potentially deadlock.  This is overly
conservative because all the anonymous memory is migrateable in the
GFP_NOFS context just fine.  This might be a large portion of the memory
in many/most workkloads.

Remove the GFP_NOFS restriction and make sure that we skip all fs pages
(those with a mapping) while isolating pages to be migrated.  We cannot
consider clean fs pages because they might need a metadata update so
only isolate pages without any mapping for nofs requests.

The effect of this patch will be probably very limited in many/most
workloads because higher order GFP_NOFS requests are quite rare,
although different configurations might lead to very different results.
David Chinner has mentioned a heavy metadata workload with 64kB block
which to quote him:

: Unfortunately, there was an era of cargo cult configuration tweaks in the
: Ceph community that has resulted in a large number of production machines
: with XFS filesystems configured this way.  And a lot of them store large
: numbers of small files and run under significant sustained memory
: pressure.
:
: I slowly working towards getting rid of these high order allocations and
: replacing them with the equivalent number of single page allocations, but
: I haven't got that (complex) change working yet.

We can do the following to simulate that workload:
$ mkfs.xfs -f -n size=64k 
$ mount  /mnt/scratch
$ time ./fs_mark  -D  10000  -S0  -n  100000  -s  0  -L  32 \
        -d  /mnt/scratch/0  -d  /mnt/scratch/1 \
        -d  /mnt/scratch/2  -d  /mnt/scratch/3 \
        -d  /mnt/scratch/4  -d  /mnt/scratch/5 \
        -d  /mnt/scratch/6  -d  /mnt/scratch/7 \
        -d  /mnt/scratch/8  -d  /mnt/scratch/9 \
        -d  /mnt/scratch/10  -d  /mnt/scratch/11 \
        -d  /mnt/scratch/12  -d  /mnt/scratch/13 \
        -d  /mnt/scratch/14  -d  /mnt/scratch/15

and indeed is hammers the system with many high order GFP_NOFS requests as
per a simle tracepoint during the load:
$ echo '!(gfp_flags & 0x80) && (gfp_flags &0x400000)' > $TRACE_MNT/events/kmem/mm_page_alloc/filter
I am getting
5287609 order=0
     37 order=1
1594905 order=2
3048439 order=3
6699207 order=4
  66645 order=5

My testing was done in a kvm guest so performance numbers should be
taken with a grain of salt but there seems to be a difference when the
patch is applied:

* Original kernel
FSUse%        Count         Size    Files/sec     App Overhead
     1      1600000            0       4300.1         20745838
     3      3200000            0       4239.9         23849857
     5      4800000            0       4243.4         25939543
     6      6400000            0       4248.4         19514050
     8      8000000            0       4262.1         20796169
     9      9600000            0       4257.6         21288675
    11     11200000            0       4259.7         19375120
    13     12800000            0       4220.7         22734141
    14     14400000            0       4238.5         31936458
    16     16000000            0       4231.5         23409901
    18     17600000            0       4045.3         23577700
    19     19200000            0       2783.4         58299526
    21     20800000            0       2678.2         40616302
    23     22400000            0       2693.5         83973996

and xfs complaining about memory allocation not making progress
[ 2304.372647] XFS: fs_mark(3289) possible memory allocation deadlock size 65624 in kmem_alloc (mode:0x2408240)
[ 2304.443323] XFS: fs_mark(3285) possible memory allocation deadlock size 65728 in kmem_alloc (mode:0x2408240)
[ 4796.772477] XFS: fs_mark(3424) possible memory allocation deadlock size 46936 in kmem_alloc (mode:0x2408240)
[ 4796.775329] XFS: fs_mark(3423) possible memory allocation deadlock size 51416 in kmem_alloc (mode:0x2408240)
[ 4797.388808] XFS: fs_mark(3424) possible memory allocation deadlock size 65728 in kmem_alloc (mode:0x2408240)

* Patched kernel
FSUse%        Count         Size    Files/sec     App Overhead
     1      1600000            0       4289.1         19243934
     3      3200000            0       4241.6         32828865
     5      4800000            0       4248.7         32884693
     6      6400000            0       4314.4         19608921
     8      8000000            0       4269.9         24953292
     9      9600000            0       4270.7         33235572
    11     11200000            0       4346.4         40817101
    13     12800000            0       4285.3         29972397
    14     14400000            0       4297.2         20539765
    16     16000000            0       4219.6         18596767
    18     17600000            0       4273.8         49611187
    19     19200000            0       4300.4         27944451
    21     20800000            0       4270.6         22324585
    22     22400000            0       4317.6         22650382
    24     24000000            0       4065.2         22297964

So the dropdown at Count 19200000 didn't happen and there was only a
single warning about allocation not making progress
[ 3063.815003] XFS: fs_mark(3272) possible memory allocation deadlock size 65624 in kmem_alloc (mode:0x2408240)

This suggests that the patch has helped even though there is not all that
much of anonymous memory as the workload mostly generates fs metadata.  I
assume the success rate would be higher with more anonymous memory which
should be the case in many workloads.

[akpm@linux-foundation.org: fix comment]
Link: http://lkml.kernel.org/r/20161012114721.31853-1-mhocko@kernel.org
Signed-off-by: Michal Hocko 
Acked-by: Vlastimil Babka 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: Dave Chinner 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Merge branch 'akpm' (patches from Andrew)

2016-12-13T04:50:02+00:00

Merge updates from Andrew Morton:

 - various misc bits

 - most of MM (quite a lot of MM material is awaiting the merge of
   linux-next dependencies)

 - kasan

 - printk updates

 - procfs updates

 - MAINTAINERS

 - /lib updates

 - checkpatch updates

* emailed patches from Andrew Morton : (123 commits)
  init: reduce rootwait polling interval time to 5ms
  binfmt_elf: use vmalloc() for allocation of vma_filesz
  checkpatch: don't emit unified-diff error for rename-only patches
  checkpatch: don't check c99 types like uint8_t under tools
  checkpatch: avoid multiple line dereferences
  checkpatch: don't check .pl files, improve absolute path commit log test
  scripts/checkpatch.pl: fix spelling
  checkpatch: don't try to get maintained status when --no-tree is given
  lib/ida: document locking requirements a bit better
  lib/rbtree.c: fix typo in comment of ____rb_erase_color
  lib/Kconfig.debug: make CONFIG_STRICT_DEVMEM depend on CONFIG_DEVMEM
  MAINTAINERS: add drm and drm/i915 irc channels
  MAINTAINERS: add "C:" for URI for chat where developers hang out
  MAINTAINERS: add drm and drm/i915 bug filing info
  MAINTAINERS: add "B:" for URI where to file bugs
  get_maintainer: look for arbitrary letter prefixes in sections
  printk: add Kconfig option to set default console loglevel
  printk/sound: handle more message headers
  printk/btrfs: handle more message headers
  printk/kdb: handle more message headers
  ...

mm, compaction: fix NR_ISOLATED_* stats for pfn based migration

2016-12-13T02:55:07+00:00

Since commit bda807d44454 ("mm: migrate: support non-lru movable page
migration") isolate_migratepages_block) can isolate !PageLRU pages which
would acct_isolated account as NR_ISOLATED_*.  Accounting these non-lru
pages NR_ISOLATED_{ANON,FILE} doesn't make any sense and it can misguide
heuristics based on those counters such as pgdat_reclaimable_pages resp.
too_many_isolated which would lead to unexpected stalls during the
direct reclaim without any good reason.  Note that
__alloc_contig_migrate_range can isolate a lot of pages at once.

On mobile devices such as 512M ram android Phone, it may use a big zram
swap.  In some cases zram(zsmalloc) uses too many non-lru but
migratedable pages, such as:

      MemTotal: 468148 kB
      Normal free:5620kB
      Free swap:4736kB
      Total swap:409596kB
      ZRAM: 164616kB(zsmalloc non-lru pages)
      active_anon:60700kB
      inactive_anon:60744kB
      active_file:34420kB
      inactive_file:37532kB

Fix this by only accounting lru pages to NR_ISOLATED_* in
isolate_migratepages_block right after they were isolated and we still
know they were on LRU.  Drop acct_isolated because it is called after
the fact and we've lost that information.  Batching per-cpu counter
doesn't make much improvement anyway.  Also make sure that we uncharge
only LRU pages when putting them back on the LRU in
putback_movable_pages resp.  when unmap_and_move migrates the page.

[mhocko@suse.com: replace acct_isolated() with direct counting]
Fixes: bda807d44454 ("mm: migrate: support non-lru movable page migration")
Link: http://lkml.kernel.org/r/20161019080240.9682-1-mhocko@kernel.org
Signed-off-by: Ming Ling 
Signed-off-by: Michal Hocko 
Acked-by: Minchan Kim 
Acked-by: Vlastimil Babka 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm/compaction: Convert to hotplug state machine

2016-12-01T23:52:37+00:00

Install the callbacks via the state machine. Should the hotplug init fail then
no threads are spawned.

Signed-off-by: Anna-Maria Gleixner 
Signed-off-by: Sebastian Andrzej Siewior 
Cc: Michal Hocko 
Cc: Mel Gorman 
Cc: linux-mm@kvack.org
Cc: rt@linutronix.de
Cc: Andrew Morton 
Cc: Vlastimil Babka 
Link: http://lkml.kernel.org/r/20161126231350.10321-15-bigeasy@linutronix.de
Signed-off-by: Thomas Gleixner

mm, compaction: restrict fragindex to costly orders

2016-10-08T01:46:29+00:00

Fragmentation index and the vm.extfrag_threshold sysctl is meant as a
heuristic to prevent excessive compaction for costly orders (i.e.  THP).
It's unlikely to make any difference for non-costly orders, especially
with the default threshold.  But we cannot afford any uncertainty for
the non-costly orders where the only alternative to successful
reclaim/compaction is OOM.  After the recent patches we are guaranteed
maximum effort without heuristics from compaction before deciding OOM,
and fragindex is the last remaining heuristic.  Therefore skip fragindex
altogether for non-costly orders.

Suggested-by: Michal Hocko 
Link: http://lkml.kernel.org/r/20160926162025.21555-5-vbabka@suse.cz
Signed-off-by: Vlastimil Babka 
Acked-by: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, compaction: ignore fragindex from compaction_zonelist_suitable()

2016-10-08T01:46:29+00:00

The compaction_zonelist_suitable() function tries to determine if
compaction will be able to proceed after sufficient reclaim, i.e.
whether there are enough reclaimable pages to provide enough order-0
freepages for compaction.

This addition of reclaimable pages to the free pages works well for the
order-0 watermark check, but in the fragmentation index check we only
consider truly free pages.  Thus we can get fragindex value close to 0
which indicates failure do to lack of memory, and wrongly decide that
compaction won't be suitable even after reclaim.

Instead of trying to somehow adjust fragindex for reclaimable pages,
let's just skip it from compaction_zonelist_suitable().

Link: http://lkml.kernel.org/r/20160926162025.21555-4-vbabka@suse.cz
Signed-off-by: Vlastimil Babka 
Acked-by: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, compaction: make full priority ignore pageblock suitability

2016-10-08T01:46:29+00:00

Several people have reported premature OOMs for order-2 allocations
(stack) due to OOM rework in 4.7.  In the scenario (parallel kernel
build and dd writing to two drives) many pageblocks get marked as
Unmovable and compaction free scanner struggles to isolate free pages.
Joonsoo Kim pointed out that the free scanner skips pageblocks that are
not movable to prevent filling them and forcing non-movable allocations
to fallback to other pageblocks.  Such heuristic makes sense to help
prevent long-term fragmentation, but premature OOMs are relatively more
urgent problem.  As a compromise, this patch disables the heuristic only
for the ultimate compaction priority.

Link: http://lkml.kernel.org/r/20160906135258.18335-5-vbabka@suse.cz
Reported-by: Ralf-Peter Rohbeck 
Reported-by: Arkadiusz Miskiewicz 
Reported-by: Olaf Hering 
Suggested-by: Joonsoo Kim 
Signed-off-by: Vlastimil Babka 
Acked-by: Michal Hocko 
Cc: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Cc: Tetsuo Handa 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, compaction: require only min watermarks for non-costly orders

2016-10-08T01:46:27+00:00

The __compaction_suitable() function checks the low watermark plus a
compact_gap() gap to decide if there's enough free memory to perform
compaction.  Then __isolate_free_page uses low watermark check to decide
if particular free page can be isolated.  In the latter case, using low
watermark is needlessly pessimistic, as the free page isolations are
only temporary.  For __compaction_suitable() the higher watermark makes
sense for high-order allocations where more freepages increase the
chance of success, and we can typically fail with some order-0 fallback
when the system is struggling to reach that watermark.  But for
low-order allocation, forming the page should not be that hard.  So
using low watermark here might just prevent compaction from even trying,
and eventually lead to OOM killer even if we are above min watermarks.

So after this patch, we use min watermark for non-costly orders in
__compaction_suitable(), and for all orders in __isolate_free_page().

[vbabka@suse.cz: clarify __isolate_free_page() comment]
 Link: http://lkml.kernel.org/r/7ae4baec-4eca-e70b-2a69-94bea4fb19fa@suse.cz
Link: http://lkml.kernel.org/r/20160810091226.6709-11-vbabka@suse.cz
Signed-off-by: Vlastimil Babka 
Tested-by: Lorenzo Stoakes 
Acked-by: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Signed-off-by: Vlastimil Babka 
Tested-by: Lorenzo Stoakes 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, compaction: use proper alloc_flags in __compaction_suitable()

2016-10-08T01:46:27+00:00

The __compaction_suitable() function checks the low watermark plus a
compact_gap() gap to decide if there's enough free memory to perform
compaction.  This check uses direct compactor's alloc_flags, but that's
wrong, since these flags are not applicable for freepage isolation.

For example, alloc_flags may indicate access to memory reserves, making
compaction proceed, and then fail watermark check during the isolation.

A similar problem exists for ALLOC_CMA, which may be part of
alloc_flags, but not during freepage isolation.  In this case however it
makes sense to use ALLOC_CMA both in __compaction_suitable() and
__isolate_free_page(), since there's actually nothing preventing the
freepage scanner to isolate from CMA pageblocks, with the assumption
that a page that could be migrated once by compaction can be migrated
also later by CMA allocation.  Thus we should count pages in CMA
pageblocks when considering compaction suitability and when isolating
freepages.

To sum up, this patch should remove some false positives from
__compaction_suitable(), and allow compaction to proceed when free pages
required for compaction reside in the CMA pageblocks.

Link: http://lkml.kernel.org/r/20160810091226.6709-10-vbabka@suse.cz
Signed-off-by: Vlastimil Babka 
Tested-by: Lorenzo Stoakes 
Cc: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

mm, compaction: create compact_gap wrapper

2016-10-08T01:46:27+00:00

Compaction uses a watermark gap of (2UL << order) pages at various
places and it's not immediately obvious why.  Abstract it through a
compact_gap() wrapper to create a single place with a thorough
explanation.

[vbabka@suse.cz: clarify the comment of compact_gap()]
 Link: http://lkml.kernel.org/r/7b6aed1f-fdf8-2063-9ff4-bbe4de712d37@suse.cz
Link: http://lkml.kernel.org/r/20160810091226.6709-9-vbabka@suse.cz
Signed-off-by: Vlastimil Babka 
Tested-by: Lorenzo Stoakes 
Acked-by: Michal Hocko 
Cc: Mel Gorman 
Cc: Joonsoo Kim 
Cc: David Rientjes 
Cc: Rik van Riel 
Signed-off-by: Vlastimil Babka 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds