__alloc_pages_slowpath has traditionally relied on the direct reclaim
and did_some_progress as an indicator that it makes sense to retry
allocation rather than declaring OOM. shrink_zones had to rely on
zone_reclaimable if shrink_zone didn't make any progress to prevent from
a premature OOM killer invocation - the LRU might be full of dirty or
writeback pages and direct reclaim cannot clean those up.
zone_reclaimable allows to rescan the reclaimable lists several times
and restart if a page is freed. This is really subtle behavior and it
might lead to a livelock when a single freed page keeps allocator
looping but the current task will not be able to allocate that single
page. OOM killer would be more appropriate than looping without any
progress for unbounded amount of time.
This patch changes OOM detection logic and pulls it out from shrink_zone
which is too low to be appropriate for any high level decisions such as
OOM which is per zonelist property. It is __alloc_pages_slowpath which
knows how many attempts have been done and what was the progress so far
therefore it is more appropriate to implement this logic.
The new heuristic is implemented in should_reclaim_retry helper called
from __alloc_pages_slowpath. It tries to be more deterministic and
easier to follow. It builds on an assumption that retrying makes sense
only if the currently reclaimable memory + free pages would allow the
current allocation request to succeed (as per __zone_watermark_ok) at
least for one zone in the usable zonelist.
This alone wouldn't be sufficient, though, because the writeback might
get stuck and reclaimable pages might be pinned for a really long time
or even depend on the current allocation context. Therefore there is a
backoff mechanism implemented which reduces the reclaim target after
each reclaim round without any progress. This means that we should
eventually converge to only NR_FREE_PAGES as the target and fail on the
wmark check and proceed to OOM. The backoff is simple and linear with
1/16 of the reclaimable pages for each round without any progress. We
are optimistic and reset counter for successful reclaim rounds.
Costly high order pages mostly preserve their semantic and those without
__GFP_REPEAT fail right away while those which have the flag set will
back off after the amount of reclaimable pages reaches equivalent of the
requested order. The only difference is that if there was no progress
during the reclaim we rely on zone watermark check. This is more
logical thing to do than previous 1<<order attempts which were a result
of zone_reclaimable faking the progress.
[vdavydov@virtuozzo.com: check classzone_idx for shrink_zone]
[hannes@cmpxchg.org: separate the heuristic into should_reclaim_retry]
[rientjes@google.com: use zone_page_state_snapshot for NR_FREE_PAGES]
[rientjes@google.com: shrink_zones doesn't need to return anything]
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <js1304@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Motivation:
As pointed out by Linus [2][3] relying on zone_reclaimable as a way to
communicate the reclaim progress is rater dubious. I tend to agree,
not only it is really obscure, it is not hard to imagine cases where a
single page freed in the loop keeps all the reclaimers looping without
getting any progress because their gfp_mask wouldn't allow to get that
page anyway (e.g. single GFP_ATOMIC alloc and free loop). This is rather
rare so it doesn't happen in the practice but the current logic which we
have is rather obscure and hard to follow a also non-deterministic.
This is an attempt to make the OOM detection more deterministic and
easier to follow because each reclaimer basically tracks its own
progress which is implemented at the page allocator layer rather spread
out between the allocator and the reclaim. The more on the
implementation is described in the first patch.
I have tested several different scenarios but it should be clear that
testing OOM killer is quite hard to be representative. There is usually
a tiny gap between almost OOM and full blown OOM which is often time
sensitive. Anyway, I have tested the following 2 scenarios and I would
appreciate if there are more to test.
Testing environment: a virtual machine with 2G of RAM and 2CPUs without
any swap to make the OOM more deterministic.
1) 2 writers (each doing dd with 4M blocks to an xfs partition with 1G
file size, removes the files and starts over again) running in
parallel for 10s to build up a lot of dirty pages when 100 parallel
mem_eaters (anon private populated mmap which waits until it gets
signal) with 80M each.
This causes an OOM flood of course and I have compared both patched
and unpatched kernels. The test is considered finished after there
are no OOM conditions detected. This should tell us whether there are
any excessive kills or some of them premature (e.g. due to dirty pages):
I have performed two runs this time each after a fresh boot.
* base kernel
$ grep "Out of memory:" base-oom-run1.log | wc -l
78
$ grep "Out of memory:" base-oom-run2.log | wc -l
78
$ grep "Kill process" base-oom-run1.log | tail -n1
[ 91.391203] Out of memory: Kill process 3061 (mem_eater) score 39 or sacrifice child
$ grep "Kill process" base-oom-run2.log | tail -n1
[ 82.141919] Out of memory: Kill process 3086 (mem_eater) score 39 or sacrifice child
$ grep "DMA32 free:" base-oom-run1.log | sed 's@.*free:\([0-9]*\)kB.*@\1@' | calc_min_max.awk
min: 5376.00 max: 6776.00 avg: 5530.75 std: 166.50 nr: 61
$ grep "DMA32 free:" base-oom-run2.log | sed 's@.*free:\([0-9]*\)kB.*@\1@' | calc_min_max.awk
min: 5416.00 max: 5608.00 avg: 5514.15 std: 42.94 nr: 52
$ grep "DMA32.*all_unreclaimable? no" base-oom-run1.log | wc -l
1
$ grep "DMA32.*all_unreclaimable? no" base-oom-run2.log | wc -l
3
* patched kernel
$ grep "Out of memory:" patched-oom-run1.log | wc -l
78
miso@tiehlicka /mnt/share/devel/miso/kvm $ grep "Out of memory:" patched-oom-run2.log | wc -l
77
e grep "Kill process" patched-oom-run1.log | tail -n1
[ 497.317732] Out of memory: Kill process 3108 (mem_eater) score 39 or sacrifice child
$ grep "Kill process" patched-oom-run2.log | tail -n1
[ 316.169920] Out of memory: Kill process 3093 (mem_eater) score 39 or sacrifice child
$ grep "DMA32 free:" patched-oom-run1.log | sed 's@.*free:\([0-9]*\)kB.*@\1@' | calc_min_max.awk
min: 5420.00 max: 5808.00 avg: 5513.90 std: 60.45 nr: 78
$ grep "DMA32 free:" patched-oom-run2.log | sed 's@.*free:\([0-9]*\)kB.*@\1@' | calc_min_max.awk
min: 5380.00 max: 6384.00 avg: 5520.94 std: 136.84 nr: 77
e grep "DMA32.*all_unreclaimable? no" patched-oom-run1.log | wc -l
2
$ grep "DMA32.*all_unreclaimable? no" patched-oom-run2.log | wc -l
3
The patched kernel run noticeably longer while invoking OOM killer same
number of times. This means that the original implementation is much
more aggressive and triggers the OOM killer sooner. free pages stats
show that neither kernels went OOM too early most of the time, though. I
guess the difference is in the backoff when retries without any progress
do sleep for a while if there is memory under writeback or dirty which
is highly likely considering the parallel IO.
Both kernels have seen races where zone wasn't marked unreclaimable
and we still hit the OOM killer. This is most likely a race where
a task managed to exit between the last allocation attempt and the oom
killer invocation.
2) 2 writers again with 10s of run and then 10 mem_eaters to consume as much
memory as possible without triggering the OOM killer. This required a lot
of tuning but I've considered 3 consecutive runs in three different boots
without OOM as a success.
* base kernel
size=$(awk '/MemFree/{printf "%dK", ($2/10)-(16*1024)}' /proc/meminfo)
* patched kernel
size=$(awk '/MemFree/{printf "%dK", ($2/10)-(12*1024)}' /proc/meminfo)
That means 40M more memory was usable without triggering OOM killer. The
base kernel sometimes managed to handle the same as patched but it
wasn't consistent and failed in at least on of the 3 runs. This seems
like a minor improvement.
I was testing also GPF_REPEAT costly requests (hughetlb) with fragmented
memory and under memory pressure. The results are in patch 11 where the
logic is implemented. In short I can see huge improvement there.
I am certainly interested in other usecases as well as well as any
feedback. Especially those which require higher order requests.
This patch (of 14):
While playing with the oom detection rework [1] I have noticed that my
heavy order-9 (hugetlb) load close to OOM ended up in an endless loop
where the reclaim hasn't made any progress but did_some_progress didn't
reflect that and compaction_suitable was backing off because no zone is
above low wmark + 1 << order.
It turned out that this is in fact an old standing bug in
compaction_ready which ignores the requested_highidx and did the
watermark check for 0 classzone_idx. This succeeds for zone DMA most
of the time as the zone is mostly unused because of lowmem protection.
As a result costly high order allocatios always report a successfull
progress even when there was none. This wasn't a problem so far
because these allocations usually fail quite early or retry only few
times with __GFP_REPEAT but this will change after later patch in this
series so make sure to not lie about the progress and propagate
requested_highidx down to compaction_ready and use it for both the
watermak check and compaction_suitable to fix this issue.
[1] http://lkml.kernel.org/r/1459855533-4600-1-git-send-email-mhocko@kernel.org
[2] https://lkml.org/lkml/2015/10/12/808
[3] https://lkml.org/lkml/2015/10/13/597
Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@suse.de>
Cc: David Rientjes <rientjes@google.com>
Cc: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Cc: Joonsoo Kim <js1304@gmail.com>
Cc: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The inactive file list should still be large enough to contain readahead
windows and freshly written file data, but it no longer is the only
source for detecting multiple accesses to file pages. The workingset
refault measurement code causes recently evicted file pages that get
accessed again after a shorter interval to be promoted directly to the
active list.
With that mechanism in place, we can afford to (on a larger system)
dedicate more memory to the active file list, so we can actually cache
more of the frequently used file pages in memory, and not have them
pushed out by streaming writes, once-used streaming file reads, etc.
This can help things like database workloads, where only half the page
cache can currently be used to cache the database working set. This
patch automatically increases that fraction on larger systems, using the
same ratio that has already been used for anonymous memory.
[hannes@cmpxchg.org: cgroup-awareness]
Signed-off-by: Rik van Riel <riel@redhat.com>
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: Andres Freund <andres@anarazel.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Konstantin Khlebnikov pointed out (nearly four years ago, when lumpy
reclaim was removed) that lru_size can be updated by -nr_taken once per
call to isolate_lru_pages(), instead of page by page.
Update it inside isolate_lru_pages(), or at its two callsites? I chose
to update it at the callsites, rearranging and grouping the updates by
nr_taken and nr_scanned together in both.
With one exception, mem_cgroup_update_lru_size(,lru,) is then used where
__mod_zone_page_state(,NR_LRU_BASE+lru,) is used; and we shall be adding
some more calls in a future commit. Make the code a little smaller and
simpler by incorporating stat update in lru_size update.
The exception was move_active_pages_to_lru(), which aggregated the
pgmoved stat update separately from the individual lru_size updates; but
I still think this a simplification worth making.
However, the __mod_zone_page_state is not peculiar to mem_cgroups: so
better use the name update_lru_size, calls mem_cgroup_update_lru_size
when CONFIG_MEMCG.
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Andres Lagar-Cavilla <andreslc@google.com>
Cc: Yang Shi <yang.shi@linaro.org>
Cc: Ning Qu <quning@gmail.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: Konstantin Khlebnikov <koct9i@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Many developers already know that field for reference count of the
struct page is _count and atomic type. They would try to handle it
directly and this could break the purpose of page reference count
tracepoint. To prevent direct _count modification, this patch rename it
to _refcount and add warning message on the code. After that, developer
who need to handle reference count will find that field should not be
accessed directly.
[akpm@linux-foundation.org: fix comments, per Vlastimil]
[akpm@linux-foundation.org: Documentation/vm/transhuge.txt too]
[sfr@canb.auug.org.au: sync ethernet driver changes]
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Johannes Berg <johannes@sipsolutions.net>
Cc: "David S. Miller" <davem@davemloft.net>
Cc: Sunil Goutham <sgoutham@cavium.com>
Cc: Chris Metcalf <cmetcalf@mellanox.com>
Cc: Manish Chopra <manish.chopra@qlogic.com>
Cc: Yuval Mintz <yuval.mintz@qlogic.com>
Cc: Tariq Toukan <tariqt@mellanox.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
When kswapd goes to sleep it checks if the node is balanced and at first
it sleeps only for HZ/10 time, then rechecks if the node is still
balanced and nobody has woken it during the initial sleep. Only then it
goes fully sleep until an allocation slowpath wakes it up again.
For higher-order allocations, waking up kcompactd is done only before
the full sleep. This turns out to be an issue in case another
high-order allocation fails during the initial sleep. It will wake
kswapd up, however kswapd considers the zone balanced from the order-0
perspective, and will just quickly try to sleep again. So if there's a
longer stream of high-order allocations hitting the slowpath and waking
up kswapd, it might never actually wake up kcompactd, which may be
considered a regression from kswapd-based compaction. In the worst
case, it might be that a single allocation that cannot direct
reclaim/compact itself is waking kswapd in the retry loop and preventing
kcompactd from being woken up and unblocking it.
This patch makes sure kcompactd is woken up in such situations by simply
moving the wakeup before the short initial sleep. More efficient
solution would be to wake kcompactd immediately instead of kswapd if the
node is already order-0 balanced, but in that case we should also move
reset_isolation_suitable() call to kcompactd so it's not adding to the
allocator's latency. Since it's late in the 4.6 cycle, let's go with
the simpler change for now.
Fixes: accf62422b ("mm, kswapd: replace kswapd compaction with waking up kcompactd")
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We have been reclaimed highmem zone if buffer_heads is over limit but
commit 6b4f7799c6 ("mm: vmscan: invoke slab shrinkers from
shrink_zone()") changed the behavior so it doesn't reclaim highmem zone
although buffer_heads is over the limit. This patch restores the logic.
Fixes: 6b4f7799c6 ("mm: vmscan: invoke slab shrinkers from shrink_zone()")
Signed-off-by: Minchan Kim <minchan@kernel.org>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The success of CMA allocation largely depends on the success of
migration and key factor of it is page reference count. Until now, page
reference is manipulated by direct calling atomic functions so we cannot
follow up who and where manipulate it. Then, it is hard to find actual
reason of CMA allocation failure. CMA allocation should be guaranteed
to succeed so finding offending place is really important.
In this patch, call sites where page reference is manipulated are
converted to introduced wrapper function. This is preparation step to
add tracepoint to each page reference manipulation function. With this
facility, we can easily find reason of CMA allocation failure. There is
no functional change in this patch.
In addition, this patch also converts reference read sites. It will
help a second step that renames page._count to something else and
prevents later attempt to direct access to it (Suggested by Andrew).
Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Michal Nazarewicz <mina86@mina86.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Sergey Senozhatsky <sergey.senozhatsky.work@gmail.com>
Cc: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It's just convenient to implement a memcg aware shrinker when you know
that shrink_control->memcg != NULL unless memcg_kmem_enabled() returns
false.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Similarly to direct reclaim/compaction, kswapd attempts to combine
reclaim and compaction to attempt making memory allocation of given
order available.
The details differ from direct reclaim e.g. in having high watermark as
a goal. The code involved in kswapd's reclaim/compaction decisions has
evolved to be quite complex.
Testing reveals that it doesn't actually work in at least one scenario,
and closer inspection suggests that it could be greatly simplified
without compromising on the goal (make high-order page available) or
efficiency (don't reclaim too much). The simplification relieas of
doing all compaction in kcompactd, which is simply woken up when high
watermarks are reached by kswapd's reclaim.
The scenario where kswapd compaction doesn't work was found with mmtests
test stress-highalloc configured to attempt order-9 allocations without
direct reclaim, just waking up kswapd. There was no compaction attempt
from kswapd during the whole test. Some added instrumentation shows
what happens:
- balance_pgdat() sets end_zone to Normal, as it's not balanced
- reclaim is attempted on DMA zone, which sets nr_attempted to 99, but
it cannot reclaim anything, so sc.nr_reclaimed is 0
- for zones DMA32 and Normal, kswapd_shrink_zone uses testorder=0, so
it merely checks if high watermarks were reached for base pages.
This is true, so no reclaim is attempted. For DMA, testorder=0
wasn't used, as compaction_suitable() returned COMPACT_SKIPPED
- even though the pgdat_needs_compaction flag wasn't set to false, no
compaction happens due to the condition sc.nr_reclaimed >
nr_attempted being false (as 0 < 99)
- priority-- due to nr_reclaimed being 0, repeat until priority reaches
0 pgdat_balanced() is false as only the small zone DMA appears
balanced (curiously in that check, watermark appears OK and
compaction_suitable() returns COMPACT_PARTIAL, because a lower
classzone_idx is used there)
Now, even if it was decided that reclaim shouldn't be attempted on the
DMA zone, the scenario would be the same, as (sc.nr_reclaimed=0 >
nr_attempted=0) is also false. The condition really should use >= as
the comment suggests. Then there is a mismatch in the check for setting
pgdat_needs_compaction to false using low watermark, while the rest uses
high watermark, and who knows what other subtlety. Hopefully this
demonstrates that this is unsustainable.
Luckily we can simplify this a lot. The reclaim/compaction decisions
make sense for direct reclaim scenario, but in kswapd, our primary goal
is to reach high watermark in order-0 pages. Afterwards we can attempt
compaction just once. Unlike direct reclaim, we don't reclaim extra
pages (over the high watermark), the current code already disallows it
for good reasons.
After this patch, we simply wake up kcompactd to process the pgdat,
after we have either succeeded or failed to reach the high watermarks in
kswapd, which goes to sleep. We pass kswapd's order and classzone_idx,
so kcompactd can apply the same criteria to determine which zones are
worth compacting. Note that we use the classzone_idx from
wakeup_kswapd(), not balanced_classzone_idx which can include higher
zones that kswapd tried to balance too, but didn't consider them in
pgdat_balanced().
Since kswapd now cannot create high-order pages itself, we need to
adjust how it determines the zones to be balanced. The key element here
is adding a "highorder" parameter to zone_balanced, which, when set to
false, makes it consider only order-0 watermark instead of the desired
higher order (this was done previously by kswapd_shrink_zone(), but not
elsewhere). This false is passed for example in pgdat_balanced().
Importantly, wakeup_kswapd() uses true to make sure kswapd and thus
kcompactd are woken up for a high-order allocation failure.
The last thing is to decide what to do with pageblock_skip bitmap
handling. Compaction maintains a pageblock_skip bitmap to record
pageblocks where isolation recently failed. This bitmap can be reset by
three ways:
1) direct compaction is restarting after going through the full deferred cycle
2) kswapd goes to sleep, and some other direct compaction has previously
finished scanning the whole zone and set zone->compact_blockskip_flush.
Note that a successful direct compaction clears this flag.
3) compaction was invoked manually via trigger in /proc
The case 2) is somewhat fuzzy to begin with, but after introducing
kcompactd we should update it. The check for direct compaction in 1),
and to set the flush flag in 2) use current_is_kswapd(), which doesn't
work for kcompactd. Thus, this patch adds bool direct_compaction to
compact_control to use in 2). For the case 1) we remove the check
completely - unlike the former kswapd compaction, kcompactd does use the
deferred compaction functionality, so flushing tied to restarting from
deferred compaction makes sense here.
Note that when kswapd goes to sleep, kcompactd is woken up, so it will
see the flushed pageblock_skip bits. This is different from when the
former kswapd compaction observed the bits and I believe it makes more
sense. Kcompactd can afford to be more thorough than a direct
compaction trying to limit allocation latency, or kswapd whose primary
goal is to reclaim.
For testing, I used stress-highalloc configured to do order-9
allocations with GFP_NOWAIT|__GFP_HIGH|__GFP_COMP, so they relied just
on kswapd/kcompactd reclaim/compaction (the interfering kernel builds in
phases 1 and 2 work as usual):
stress-highalloc
4.5-rc1+before 4.5-rc1+after
-nodirect -nodirect
Success 1 Min 1.00 ( 0.00%) 5.00 (-66.67%)
Success 1 Mean 1.40 ( 0.00%) 6.20 (-55.00%)
Success 1 Max 2.00 ( 0.00%) 7.00 (-16.67%)
Success 2 Min 1.00 ( 0.00%) 5.00 (-66.67%)
Success 2 Mean 1.80 ( 0.00%) 6.40 (-52.38%)
Success 2 Max 3.00 ( 0.00%) 7.00 (-16.67%)
Success 3 Min 34.00 ( 0.00%) 62.00 ( 1.59%)
Success 3 Mean 41.80 ( 0.00%) 63.80 ( 1.24%)
Success 3 Max 53.00 ( 0.00%) 65.00 ( 2.99%)
User 3166.67 3181.09
System 1153.37 1158.25
Elapsed 1768.53 1799.37
4.5-rc1+before 4.5-rc1+after
-nodirect -nodirect
Direct pages scanned 32938 32797
Kswapd pages scanned 2183166 2202613
Kswapd pages reclaimed 2152359 2143524
Direct pages reclaimed 32735 32545
Percentage direct scans 1% 1%
THP fault alloc 579 612
THP collapse alloc 304 316
THP splits 0 0
THP fault fallback 793 778
THP collapse fail 11 16
Compaction stalls 1013 1007
Compaction success 92 67
Compaction failures 920 939
Page migrate success 238457 721374
Page migrate failure 23021 23469
Compaction pages isolated 504695 1479924
Compaction migrate scanned 661390 8812554
Compaction free scanned 13476658 84327916
Compaction cost 262 838
After this patch we see improvements in allocation success rate
(especially for phase 3) along with increased compaction activity. The
compaction stalls (direct compaction) in the interfering kernel builds
(probably THP's) also decreased somewhat thanks to kcompactd activity,
yet THP alloc successes improved a bit.
Note that elapsed and user time isn't so useful for this benchmark,
because of the background interference being unpredictable. It's just
to quickly spot some major unexpected differences. System time is
somewhat more useful and that didn't increase.
Also (after adjusting mmtests' ftrace monitor):
Time kswapd awake 2547781 2269241
Time kcompactd awake 0 119253
Time direct compacting 939937 557649
Time kswapd compacting 0 0
Time kcompactd compacting 0 119099
The decrease of overal time spent compacting appears to not match the
increased compaction stats. I suspect the tasks get rescheduled and
since the ftrace monitor doesn't see that, the reported time is wall
time, not CPU time. But arguably direct compactors care about overall
latency anyway, whether busy compacting or waiting for CPU doesn't
matter. And that latency seems to almost halved.
It's also interesting how much time kswapd spent awake just going
through all the priorities and failing to even try compacting, over and
over.
We can also configure stress-highalloc to perform both direct
reclaim/compaction and wakeup kswapd/kcompactd, by using
GFP_KERNEL|__GFP_HIGH|__GFP_COMP:
stress-highalloc
4.5-rc1+before 4.5-rc1+after
-direct -direct
Success 1 Min 4.00 ( 0.00%) 9.00 (-50.00%)
Success 1 Mean 8.00 ( 0.00%) 10.00 (-19.05%)
Success 1 Max 12.00 ( 0.00%) 11.00 ( 15.38%)
Success 2 Min 4.00 ( 0.00%) 9.00 (-50.00%)
Success 2 Mean 8.20 ( 0.00%) 10.00 (-16.28%)
Success 2 Max 13.00 ( 0.00%) 11.00 ( 8.33%)
Success 3 Min 75.00 ( 0.00%) 74.00 ( 1.33%)
Success 3 Mean 75.60 ( 0.00%) 75.20 ( 0.53%)
Success 3 Max 77.00 ( 0.00%) 76.00 ( 0.00%)
User 3344.73 3246.04
System 1194.24 1172.29
Elapsed 1838.04 1836.76
4.5-rc1+before 4.5-rc1+after
-direct -direct
Direct pages scanned 125146 120966
Kswapd pages scanned 2119757 2135012
Kswapd pages reclaimed 2073183 2108388
Direct pages reclaimed 124909 120577
Percentage direct scans 5% 5%
THP fault alloc 599 652
THP collapse alloc 323 354
THP splits 0 0
THP fault fallback 806 793
THP collapse fail 17 16
Compaction stalls 2457 2025
Compaction success 906 518
Compaction failures 1551 1507
Page migrate success 2031423 2360608
Page migrate failure 32845 40852
Compaction pages isolated 4129761 4802025
Compaction migrate scanned 11996712 21750613
Compaction free scanned 214970969 344372001
Compaction cost 2271 2694
In this scenario, this patch doesn't change the overall success rate as
direct compaction already tries all it can. There's however significant
reduction in direct compaction stalls (that is, the number of
allocations that went into direct compaction). The number of successes
(i.e. direct compaction stalls that ended up with successful
allocation) is reduced by the same number. This means the offload to
kcompactd is working as expected, and direct compaction is reduced
either due to detecting contention, or compaction deferred by kcompactd.
In the previous version of this patchset there was some apparent
reduction of success rate, but the changes in this version (such as
using sync compaction only), new baseline kernel, and/or averaging
results from 5 executions (my bet), made this go away.
Ftrace-based stats seem to roughly agree:
Time kswapd awake 2532984 2326824
Time kcompactd awake 0 257916
Time direct compacting 864839 735130
Time kswapd compacting 0 0
Time kcompactd compacting 0 257585
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
During work on kcompactd integration I have spotted a confusing check of
balance_classzone_idx, which I believe is bogus.
The balanced_classzone_idx is filled by balance_pgdat() as the highest
zone it attempted to balance. This was introduced by commit dc83edd941
("mm: kswapd: use the classzone idx that kswapd was using for
sleeping_prematurely()").
The intention is that (as expressed in today's function names), the
value used for kswapd_shrink_zone() calls in balance_pgdat() is the same
as for the decisions in kswapd_try_to_sleep().
An unwanted side-effect of that commit was breaking the checks in
kswapd() whether there was another kswapd_wakeup with a tighter (=lower)
classzone_idx. Commits 215ddd6664 ("mm: vmscan: only read
new_classzone_idx from pgdat when reclaiming successfully") and
d2ebd0f6b8 ("kswapd: avoid unnecessary rebalance after an unsuccessful
balancing") tried to fixed, but apparently introduced a bogus check that
this patch removes.
Consider zone indexes X < Y < Z, where:
- Z is the value used for the first kswapd wakeup.
- Y is returned as balanced_classzone_idx, which means zones with index higher
than Y (including Z) were found to be unreclaimable.
- X is the value used for the second kswapd wakeup
The new wakeup with value X means that kswapd is now supposed to balance
harder all zones with index <= X. But instead, due to Y < Z, it will go
sleep and won't read the new value X. This is subtly wrong.
The effect of this patch is that kswapd will react better in some
situations, where e.g. the first wakeup is for ZONE_DMA32, the second is
for ZONE_DMA, and due to unreclaimable ZONE_NORMAL. Before this patch,
kswapd would go sleep instead of reclaiming ZONE_DMA harder. I expect
these situations are very rare, and more value is in better
maintainability due to the removal of confusing and bogus check.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Mel Gorman <mgorman@techsingularity.net>
Cc: David Rientjes <rientjes@google.com>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There are several users that nest lock_page_memcg() inside lock_page()
to prevent page->mem_cgroup from changing. But the page lock prevents
pages from moving between cgroups, so that is unnecessary overhead.
Remove lock_page_memcg() in contexts with locked contexts and fix the
debug code in the page stat functions to be okay with the page lock.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that migration doesn't clear page->mem_cgroup of live pages anymore,
it's safe to make lock_page_memcg() and the memcg stat functions take
pages, and spare the callers from memcg objects.
[akpm@linux-foundation.org: fix warnings]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Suggested-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Cache thrash detection (see a528910e12 "mm: thrash detection-based
file cache sizing" for details) currently only works on the system
level, not inside cgroups. Worse, as the refaults are compared to the
global number of active cache, cgroups might wrongfully get all their
refaults activated when their pages are hotter than those of others.
Move the refault machinery from the zone to the lruvec, and then tag
eviction entries with the memcg ID. This makes the thrash detection
work correctly inside cgroups.
[sergey.senozhatsky@gmail.com: do not return from workingset_activation() with locked rcu and page]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Sergey Senozhatsky <sergey.senozhatsky@gmail.com>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
These patches tag the page cache radix tree eviction entries with the
memcg an evicted page belonged to, thus making per-cgroup LRU reclaim
work properly and be as adaptive to new cache workingsets as global
reclaim already is.
This should have been part of the original thrash detection patch
series, but was deferred due to the complexity of those patches.
This patch (of 5):
So far the only sites that needed to exclude charge migration to
stabilize page->mem_cgroup have been per-cgroup page statistics, hence
the name mem_cgroup_begin_page_stat(). But per-cgroup thrash detection
will add another site that needs to ensure page->mem_cgroup lifetime.
Rename these locking functions to the more generic lock_page_memcg() and
unlock_page_memcg(). Since charge migration is a cgroup1 feature only,
we might be able to delete it at some point, and these now easy to
identify locking sites along with it.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Suggested-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
zone_reclaimable_pages() is used in should_reclaim_retry() which uses it
to calculate the target for the watermark check. This means that
precise numbers are important for the correct decision.
zone_reclaimable_pages uses zone_page_state which can contain stale data
with per-cpu diffs not synced yet (the last vmstat_update might have run
1s in the past).
Use zone_page_state_snapshot() in zone_reclaimable_pages() instead.
None of the current callers is in a hot path where getting the precise
value (which involves per-cpu iteration) would cause an unreasonable
overhead.
Signed-off-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Suggested-by: David Rientjes <rientjes@google.com>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Hillf Danton <hillf.zj@alibaba-inc.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently, on shrinker registration we clear SHRINKER_NUMA_AWARE if
there's the only NUMA node present. The comment states that this will
allow us to save some small loop time later. It used to be true when
this code was added (see commit 1d3d4437ea ("vmscan: per-node
deferred work")), but since commit 6b4f7799c6 ("mm: vmscan: invoke
slab shrinkers from shrink_zone()") it doesn't make any difference.
Anyway, running on non-NUMA machine shouldn't make a shrinker NUMA
unaware, so zap this hunk.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Calling isolate_lru_page() is wrong and shouldn't happen, but it not
nessesary fatal: the page just will not be isolated if it's not on LRU.
Let's downgrade the VM_BUG_ON_PAGE() to WARN_RATELIMIT().
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Add support for tracking dirty DAX entries in the struct address_space
radix tree. This tree is already used for dirty page writeback, and it
already supports the use of exceptional (non struct page*) entries.
In order to properly track dirty DAX pages we will insert new
exceptional entries into the radix tree that represent dirty DAX PTE or
PMD pages. These exceptional entries will also contain the writeback
addresses for the PTE or PMD faults that we can use at fsync/msync time.
There are currently two types of exceptional entries (shmem and shadow)
that can be placed into the radix tree, and this adds a third. We rely
on the fact that only one type of exceptional entry can be found in a
given radix tree based on its usage. This happens for free with DAX vs
shmem but we explicitly prevent shadow entries from being added to radix
trees for DAX mappings.
The only shadow entries that would be generated for DAX radix trees
would be to track zero page mappings that were created for holes. These
pages would receive minimal benefit from having shadow entries, and the
choice to have only one type of exceptional entry in a given radix tree
makes the logic simpler both in clear_exceptional_entry() and in the
rest of DAX.
Signed-off-by: Ross Zwisler <ross.zwisler@linux.intel.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andreas Dilger <adilger.kernel@dilger.ca>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jan Kara <jack@suse.com>
Cc: Jeff Layton <jlayton@poochiereds.net>
Cc: Matthew Wilcox <willy@linux.intel.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Matthew Wilcox <matthew.r.wilcox@intel.com>
Cc: Dave Hansen <dave.hansen@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Swap cache pages are freed aggressively if swap is nearly full (>50%
currently), because otherwise we are likely to stop scanning anonymous
when we near the swap limit even if there is plenty of freeable swap cache
pages. We should follow the same trend in case of memory cgroup, which
has its own swap limit.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We don't scan anonymous memory if we ran out of swap, neither should we do
it in case memcg swap limit is hit, because swap out is impossible anyway.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mem_cgroup_lruvec_online() takes lruvec, but it only needs memcg. Since
get_scan_count(), which is the only user of this function, now possesses
pointer to memcg, let's pass memcg directly to mem_cgroup_online() instead
of picking it out of lruvec and rename the function accordingly.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
memcg will come in handy in get_scan_count(). It can already be used for
getting swappiness immediately in get_scan_count() instead of passing it
around. The following patches will add more memcg-related values, which
will be used there.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
On any given memcg, the kmem accounting feature has three separate
states: not initialized, structures allocated, and actively accounting
slab memory. These are represented through a combination of the
kmem_acct_activated and kmem_acct_active flags, which is confusing.
Convert to a kmem_state enum with the states NONE, ALLOCATED, and
ONLINE. Then rename the functions to modify the state accordingly.
This follows the nomenclature of css object states more closely.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Tejun Heo <tj@kernel.org>
Acked-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linux doesn't have an ability to free pages lazy while other OS already
have been supported that named by madvise(MADV_FREE).
The gain is clear that kernel can discard freed pages rather than
swapping out or OOM if memory pressure happens.
Without memory pressure, freed pages would be reused by userspace
without another additional overhead(ex, page fault + allocation +
zeroing).
Jason Evans said:
: Facebook has been using MAP_UNINITIALIZED
: (https://lkml.org/lkml/2012/1/18/308) in some of its applications for
: several years, but there are operational costs to maintaining this
: out-of-tree in our kernel and in jemalloc, and we are anxious to retire it
: in favor of MADV_FREE. When we first enabled MAP_UNINITIALIZED it
: increased throughput for much of our workload by ~5%, and although the
: benefit has decreased using newer hardware and kernels, there is still
: enough benefit that we cannot reasonably retire it without a replacement.
:
: Aside from Facebook operations, there are numerous broadly used
: applications that would benefit from MADV_FREE. The ones that immediately
: come to mind are redis, varnish, and MariaDB. I don't have much insight
: into Android internals and development process, but I would hope to see
: MADV_FREE support eventually end up there as well to benefit applications
: linked with the integrated jemalloc.
:
: jemalloc will use MADV_FREE once it becomes available in the Linux kernel.
: In fact, jemalloc already uses MADV_FREE or equivalent everywhere it's
: available: *BSD, OS X, Windows, and Solaris -- every platform except Linux
: (and AIX, but I'm not sure it even compiles on AIX). The lack of
: MADV_FREE on Linux forced me down a long series of increasingly
: sophisticated heuristics for madvise() volume reduction, and even so this
: remains a common performance issue for people using jemalloc on Linux.
: Please integrate MADV_FREE; many people will benefit substantially.
How it works:
When madvise syscall is called, VM clears dirty bit of ptes of the
range. If memory pressure happens, VM checks dirty bit of page table
and if it found still "clean", it means it's a "lazyfree pages" so VM
could discard the page instead of swapping out. Once there was store
operation for the page before VM peek a page to reclaim, dirty bit is
set so VM can swap out the page instead of discarding.
One thing we should notice is that basically, MADV_FREE relies on dirty
bit in page table entry to decide whether VM allows to discard the page
or not. IOW, if page table entry includes marked dirty bit, VM
shouldn't discard the page.
However, as a example, if swap-in by read fault happens, page table
entry doesn't have dirty bit so MADV_FREE could discard the page
wrongly.
For avoiding the problem, MADV_FREE did more checks with PageDirty and
PageSwapCache. It worked out because swapped-in page lives on swap
cache and since it is evicted from the swap cache, the page has PG_dirty
flag. So both page flags check effectively prevent wrong discarding by
MADV_FREE.
However, a problem in above logic is that swapped-in page has PG_dirty
still after they are removed from swap cache so VM cannot consider the
page as freeable any more even if madvise_free is called in future.
Look at below example for detail.
ptr = malloc();
memset(ptr);
..
..
.. heavy memory pressure so all of pages are swapped out
..
..
var = *ptr; -> a page swapped-in and could be removed from
swapcache. Then, page table doesn't mark
dirty bit and page descriptor includes PG_dirty
..
..
madvise_free(ptr); -> It doesn't clear PG_dirty of the page.
..
..
..
.. heavy memory pressure again.
.. In this time, VM cannot discard the page because the page
.. has *PG_dirty*
To solve the problem, this patch clears PG_dirty if only the page is
owned exclusively by current process when madvise is called because
PG_dirty represents ptes's dirtiness in several processes so we could
clear it only if we own it exclusively.
Firstly, heavy users would be general allocators(ex, jemalloc, tcmalloc
and hope glibc supports it) and jemalloc/tcmalloc already have supported
the feature for other OS(ex, FreeBSD)
barrios@blaptop:~/benchmark/ebizzy$ lscpu
Architecture: x86_64
CPU op-mode(s): 32-bit, 64-bit
Byte Order: Little Endian
CPU(s): 12
On-line CPU(s) list: 0-11
Thread(s) per core: 1
Core(s) per socket: 1
Socket(s): 12
NUMA node(s): 1
Vendor ID: GenuineIntel
CPU family: 6
Model: 2
Stepping: 3
CPU MHz: 3200.185
BogoMIPS: 6400.53
Virtualization: VT-x
Hypervisor vendor: KVM
Virtualization type: full
L1d cache: 32K
L1i cache: 32K
L2 cache: 4096K
NUMA node0 CPU(s): 0-11
ebizzy benchmark(./ebizzy -S 10 -n 512)
Higher avg is better.
vanilla-jemalloc MADV_free-jemalloc
1 thread
records: 10 records: 10
avg: 2961.90 avg: 12069.70
std: 71.96(2.43%) std: 186.68(1.55%)
max: 3070.00 max: 12385.00
min: 2796.00 min: 11746.00
2 thread
records: 10 records: 10
avg: 5020.00 avg: 17827.00
std: 264.87(5.28%) std: 358.52(2.01%)
max: 5244.00 max: 18760.00
min: 4251.00 min: 17382.00
4 thread
records: 10 records: 10
avg: 8988.80 avg: 27930.80
std: 1175.33(13.08%) std: 3317.33(11.88%)
max: 9508.00 max: 30879.00
min: 5477.00 min: 21024.00
8 thread
records: 10 records: 10
avg: 13036.50 avg: 33739.40
std: 170.67(1.31%) std: 5146.22(15.25%)
max: 13371.00 max: 40572.00
min: 12785.00 min: 24088.00
16 thread
records: 10 records: 10
avg: 11092.40 avg: 31424.20
std: 710.60(6.41%) std: 3763.89(11.98%)
max: 12446.00 max: 36635.00
min: 9949.00 min: 25669.00
32 thread
records: 10 records: 10
avg: 11067.00 avg: 34495.80
std: 971.06(8.77%) std: 2721.36(7.89%)
max: 12010.00 max: 38598.00
min: 9002.00 min: 30636.00
In summary, MADV_FREE is about much faster than MADV_DONTNEED.
This patch (of 12):
Add core MADV_FREE implementation.
[akpm@linux-foundation.org: small cleanups]
Signed-off-by: Minchan Kim <minchan@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Mika Penttil <mika.penttila@nextfour.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mgorman@suse.de>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Jason Evans <je@fb.com>
Cc: Daniel Micay <danielmicay@gmail.com>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Shaohua Li <shli@kernel.org>
Cc: <yalin.wang2010@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: "James E.J. Bottomley" <jejb@parisc-linux.org>
Cc: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: "Shaohua Li" <shli@kernel.org>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Chen Gang <gang.chen.5i5j@gmail.com>
Cc: Chris Zankel <chris@zankel.net>
Cc: Darrick J. Wong <darrick.wong@oracle.com>
Cc: David S. Miller <davem@davemloft.net>
Cc: Helge Deller <deller@gmx.de>
Cc: Ivan Kokshaysky <ink@jurassic.park.msu.ru>
Cc: Matt Turner <mattst88@gmail.com>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: Richard Henderson <rth@twiddle.net>
Cc: Roland Dreier <roland@kernel.org>
Cc: Russell King <rmk@arm.linux.org.uk>
Cc: Shaohua Li <shli@kernel.org>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Wu Fengguang <fengguang.wu@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The VM_BUG_ON_PAGE() would catch such cases if any still exists.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Let the networking stack know when a memcg is under reclaim pressure so
that it can clamp its transmit windows accordingly.
Whenever the reclaim efficiency of a cgroup's LRU lists drops low enough
for a MEDIUM or HIGH vmpressure event to occur, assert a pressure state
in the socket and tcp memory code that tells it to curb consumption
growth from sockets associated with said control group.
Traditionally, vmpressure reports for the entire subtree of a memcg
under pressure, which drops useful information on the individual groups
reclaimed. However, it's too late to change the userinterface, so add a
second reporting mode that reports on the level of reclaim instead of at
the level of pressure, and use that report for sockets.
vmpressure events are naturally edge triggered, so for hysteresis assert
socket pressure for a second to allow for subsequent vmpressure events
to occur before letting the socket code return to normal.
This will likely need finetuning for a wider variety of workloads, but
for now stick to the vmpressure presets and keep hysteresis simple.
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: David S. Miller <davem@davemloft.net>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
list_to_page() in readahead.c is the same as lru_to_page() in vmscan.c.
So I move lru_to_page to internal.h and drop list_to_page().
Signed-off-by: Geliang Tang <geliangtang@163.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
zone_reclaimable_pages counts how many pages are reclaimable in the
given zone. This currently includes all pages on file lrus and anon
lrus if there is an available swap storage. We do not consider
NR_ISOLATED_{ANON,FILE} counters though which is not correct because
these counters reflect temporarily isolated pages which are still
reclaimable because they either get back to their LRU or get freed
either by the page reclaim or page migration.
The number of these pages might be sufficiently high to confuse users of
zone_reclaimable_pages (e.g. mbind can migrate large ranges of memory
at once).
Signed-off-by: Michal Hocko <mhocko@suse.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We assume there is enough inactive page cache if the size of inactive
file lru is greater than the size of active file lru, in which case we
force-scan file lru ignoring anonymous pages. While this logic works
fine when there are plenty of page cache pages, it fails if the size of
file lru is small (several MB): in this case (lru_size >> prio) will be
0 for normal scan priorities, as a result, if inactive file lru happens
to be larger than active file lru, anonymous pages of a cgroup will
never get evicted unless the system experiences severe memory pressure,
even if there are gigabytes of unused anonymous memory there, which is
unfair in respect to other cgroups, whose workloads might be page cache
oriented.
This patch attempts to fix this by elaborating the "enough inactive page
cache" check: it makes it not only check that inactive lru size > active
lru size, but also that we will scan something from the cgroup at the
current scan priority. If these conditions do not hold, we proceed to
SCAN_FRACT as usual.
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move node_id zone_idx shrink flags into trace function, so thay we don't
need caculate these args if the trace is disabled, and will make this
function have less arguments.
Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Move trace_reclaim_flags() into trace function, so that we don't need
caculate these flags if the trace is disabled.
Signed-off-by: yalin wang <yalin.wang2010@gmail.com>
Reviewed-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
__GFP_WAIT has been used to identify atomic context in callers that hold
spinlocks or are in interrupts. They are expected to be high priority and
have access one of two watermarks lower than "min" which can be referred
to as the "atomic reserve". __GFP_HIGH users get access to the first
lower watermark and can be called the "high priority reserve".
Over time, callers had a requirement to not block when fallback options
were available. Some have abused __GFP_WAIT leading to a situation where
an optimisitic allocation with a fallback option can access atomic
reserves.
This patch uses __GFP_ATOMIC to identify callers that are truely atomic,
cannot sleep and have no alternative. High priority users continue to use
__GFP_HIGH. __GFP_DIRECT_RECLAIM identifies callers that can sleep and
are willing to enter direct reclaim. __GFP_KSWAPD_RECLAIM to identify
callers that want to wake kswapd for background reclaim. __GFP_WAIT is
redefined as a caller that is willing to enter direct reclaim and wake
kswapd for background reclaim.
This patch then converts a number of sites
o __GFP_ATOMIC is used by callers that are high priority and have memory
pools for those requests. GFP_ATOMIC uses this flag.
o Callers that have a limited mempool to guarantee forward progress clear
__GFP_DIRECT_RECLAIM but keep __GFP_KSWAPD_RECLAIM. bio allocations fall
into this category where kswapd will still be woken but atomic reserves
are not used as there is a one-entry mempool to guarantee progress.
o Callers that are checking if they are non-blocking should use the
helper gfpflags_allow_blocking() where possible. This is because
checking for __GFP_WAIT as was done historically now can trigger false
positives. Some exceptions like dm-crypt.c exist where the code intent
is clearer if __GFP_DIRECT_RECLAIM is used instead of the helper due to
flag manipulations.
o Callers that built their own GFP flags instead of starting with GFP_KERNEL
and friends now also need to specify __GFP_KSWAPD_RECLAIM.
The first key hazard to watch out for is callers that removed __GFP_WAIT
and was depending on access to atomic reserves for inconspicuous reasons.
In some cases it may be appropriate for them to use __GFP_HIGH.
The second key hazard is callers that assembled their own combination of
GFP flags instead of starting with something like GFP_KERNEL. They may
now wish to specify __GFP_KSWAPD_RECLAIM. It's almost certainly harmless
if it's missed in most cases as other activity will wake kswapd.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Christoph Lameter <cl@linux.com>
Cc: David Rientjes <rientjes@google.com>
Cc: Vitaly Wool <vitalywool@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Overall, the intent of this series is to remove the zonelist cache which
was introduced to avoid high overhead in the page allocator. Once this is
done, it is necessary to reduce the cost of watermark checks.
The series starts with minor micro-optimisations.
Next it notes that GFP flags that affect watermark checks are abused.
__GFP_WAIT historically identified callers that could not sleep and could
access reserves. This was later abused to identify callers that simply
prefer to avoid sleeping and have other options. A patch distinguishes
between atomic callers, high-priority callers and those that simply wish
to avoid sleep.
The zonelist cache has been around for a long time but it is of dubious
merit with a lot of complexity and some issues that are explained. The
most important issue is that a failed THP allocation can cause a zone to
be treated as "full". This potentially causes unnecessary stalls, reclaim
activity or remote fallbacks. The issues could be fixed but it's not
worth it. The series places a small number of other micro-optimisations
on top before examining GFP flags watermarks.
High-order watermarks enforcement can cause high-order allocations to fail
even though pages are free. The watermark checks both protect high-order
atomic allocations and make kswapd aware of high-order pages but there is
a much better way that can be handled using migrate types. This series
uses page grouping by mobility to reserve pageblocks for high-order
allocations with the size of the reservation depending on demand. kswapd
awareness is maintained by examining the free lists. By patch 12 in this
series, there are no high-order watermark checks while preserving the
properties that motivated the introduction of the watermark checks.
This patch (of 10):
No user of zone_watermark_ok_safe() specifies alloc_flags. This patch
removes the unnecessary parameter.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Acked-by: David Rientjes <rientjes@google.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Acked-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: Christoph Lameter <cl@linux.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Merge patch-bomb from Andrew Morton:
- inotify tweaks
- some ocfs2 updates (many more are awaiting review)
- various misc bits
- kernel/watchdog.c updates
- Some of mm. I have a huge number of MM patches this time and quite a
lot of it is quite difficult and much will be held over to next time.
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (162 commits)
selftests: vm: add tests for lock on fault
mm: mlock: add mlock flags to enable VM_LOCKONFAULT usage
mm: introduce VM_LOCKONFAULT
mm: mlock: add new mlock system call
mm: mlock: refactor mlock, munlock, and munlockall code
kasan: always taint kernel on report
mm, slub, kasan: enable user tracking by default with KASAN=y
kasan: use IS_ALIGNED in memory_is_poisoned_8()
kasan: Fix a type conversion error
lib: test_kasan: add some testcases
kasan: update reference to kasan prototype repo
kasan: move KASAN_SANITIZE in arch/x86/boot/Makefile
kasan: various fixes in documentation
kasan: update log messages
kasan: accurately determine the type of the bad access
kasan: update reported bug types for kernel memory accesses
kasan: update reported bug types for not user nor kernel memory accesses
mm/kasan: prevent deadlock in kasan reporting
mm/kasan: don't use kasan shadow pointer in generic functions
mm/kasan: MODULE_VADDR is not available on all archs
...
In zone_reclaimable_pages(), `nr' is returned by a function which is
declared as returning "unsigned long", so declare it such. Negative
values are meaningless here.
In zone_pagecache_reclaimable() we should also declare `delta' and
`nr_pagecache_reclaimable' as being unsigned longs because they're used to
store the values returned by zone_page_state() and
zone_unmapped_file_pages() which also happen to return unsigned integers.
[akpm@linux-foundation.org: make zone_pagecache_reclaimable() return ulong rather than long]
Signed-off-by: Alexandru Moise <00moses.alexander00@gmail.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Make inactive_anon/file_is_low return bool due to these particular
functions only using either one or zero as their return value.
No functional change.
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Delete unnecessary if to let inactive_anon_is_low_global return
directly.
No functional changes.
Signed-off-by: Yaowei Bai <bywxiaobai@163.com>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cgroup updates from Tejun Heo:
"The cgroup core saw several significant updates this cycle:
- percpu_rwsem for threadgroup locking is reinstated. This was
temporarily dropped due to down_write latency issues. Oleg's
rework of percpu_rwsem which is scheduled to be merged in this
merge window resolves the issue.
- On the v2 hierarchy, when controllers are enabled and disabled, all
operations are atomic and can fail and revert cleanly. This allows
->can_attach() failure which is necessary for cpu RT slices.
- Tasks now stay associated with the original cgroups after exit
until released. This allows tracking resources held by zombies
(e.g. pids) and makes it easy to find out where zombies came from
on the v2 hierarchy. The pids controller was broken before these
changes as zombies escaped the limits; unfortunately, updating this
behavior required too many invasive changes and I don't think it's
a good idea to backport them, so the pids controller on 4.3, the
first version which included the pids controller, will stay broken
at least until I'm sure about the cgroup core changes.
- Optimization of a couple common tests using static_key"
* 'for-4.4' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup: (38 commits)
cgroup: fix race condition around termination check in css_task_iter_next()
blkcg: don't create "io.stat" on the root cgroup
cgroup: drop cgroup__DEVEL__legacy_files_on_dfl
cgroup: replace error handling in cgroup_init() with WARN_ON()s
cgroup: add cgroup_subsys->free() method and use it to fix pids controller
cgroup: keep zombies associated with their original cgroups
cgroup: make css_set_rwsem a spinlock and rename it to css_set_lock
cgroup: don't hold css_set_rwsem across css task iteration
cgroup: reorganize css_task_iter functions
cgroup: factor out css_set_move_task()
cgroup: keep css_set and task lists in chronological order
cgroup: make cgroup_destroy_locked() test cgroup_is_populated()
cgroup: make css_sets pin the associated cgroups
cgroup: relocate cgroup_[try]get/put()
cgroup: move check_for_release() invocation
cgroup: replace cgroup_has_tasks() with cgroup_is_populated()
cgroup: make cgroup->nr_populated count the number of populated css_sets
cgroup: remove an unused parameter from cgroup_task_migrate()
cgroup: fix too early usage of static_branch_disable()
cgroup: make cgroup_update_dfl_csses() migrate all target processes atomically
...
The sane_reclaim() helper is supposed to return false for memcg reclaim
if the legacy hierarchy is used, because the latter lacks dirty
throttling mechanism, and so it did before it was accidentally broken by
commit 33398cf2f3 ("memcg: export struct mem_cgroup"). Fix it.
Fixes: 33398cf2f3 ("memcg: export struct mem_cgroup")
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: Tejun Heo <tj@kernel.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This is merely a politeness: I've not found that shrink_page_list()
leads to deadlock with the page it holds locked across
wait_on_page_writeback(); but nevertheless, why hold others off by
keeping the page locked there?
And while we're at it: remove the mistaken "not " from the commentary on
this Case 3 (and a distracting blank line from Case 2, if I may).
Signed-off-by: Hugh Dickins <hughd@google.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
reclaim_clean_pages_from_list() assumes that shrink_page_list() returns
number of pages removed from the candidate list. But shrink_page_list()
puts back mlocked pages without passing it to caller and without
counting as nr_reclaimed. This increases nr_isolated.
To fix this, this patch changes shrink_page_list() to pass unevictable
pages back to caller. Caller will take care those pages.
Minchan said:
It fixes two issues.
1. With unevictable page, cma_alloc will be successful.
Exactly speaking, cma_alloc of current kernel will fail due to
unevictable pages.
2. fix leaking of NR_ISOLATED counter of vmstat
With it, too_many_isolated works. Otherwise, it could make hang until
the process get SIGKILL.
Signed-off-by: Jaewon Kim <jaewon31.kim@samsung.com>
Acked-by: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If transparent huge pages are enabled, we can isolate many more pages
than we actually need to scan, because we count both single and huge
pages equally in isolate_lru_pages().
Since commit 5bc7b8aca9 ("mm: thp: add split tail pages to shrink
page list in page reclaim"), we scan all the tail pages immediately
after a huge page split (see shrink_page_list()). As a result, we can
reclaim up to SWAP_CLUSTER_MAX * HPAGE_PMD_NR (512 MB) in one run!
This is easy to catch on memcg reclaim with zswap enabled. The latter
makes swapout instant so that if we happen to scan an unreferenced huge
page we will evict both its head and tail pages immediately, which is
likely to result in excessive reclaim.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Michal Hocko <mhocko@suse.cz>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Minchan Kim <minchan@kernel.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mem_cgroup structure is defined in mm/memcontrol.c currently which means
that the code outside of this file has to use external API even for
trivial access stuff.
This patch exports mm_struct with its dependencies and makes some of the
exported functions inlines. This even helps to reduce the code size a bit
(make defconfig + CONFIG_MEMCG=y)
text data bss dec hex filename
12355346 1823792 1089536 15268674 e8fb42 vmlinux.before
12354970 1823792 1089536 15268298 e8f9ca vmlinux.after
This is not much (370B) but better than nothing.
We also save a function call in some hot paths like callers of
mem_cgroup_count_vm_event which is used for accounting.
The patch doesn't introduce any functional changes.
[vdavykov@parallels.com: inline memcg_kmem_is_active]
[vdavykov@parallels.com: do not expose type outside of CONFIG_MEMCG]
[akpm@linux-foundation.org: memcontrol.h needs eventfd.h for eventfd_ctx]
[akpm@linux-foundation.org: export mem_cgroup_from_task() to modules]
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Reviewed-by: Vladimir Davydov <vdavydov@parallels.com>
Suggested-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If a PTE is unmapped and it's dirty then it was writable recently. Due to
deferred TLB flushing, it's best to assume a writable TLB cache entry
exists. With that assumption, the TLB must be flushed before any IO can
start or the page is freed to avoid lost writes or data corruption. This
patch defers flushing of potentially writable TLBs as long as possible.
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
An IPI is sent to flush remote TLBs when a page is unmapped that was
potentially accesssed by other CPUs. There are many circumstances where
this happens but the obvious one is kswapd reclaiming pages belonging to a
running process as kswapd and the task are likely running on separate
CPUs.
On small machines, this is not a significant problem but as machine gets
larger with more cores and more memory, the cost of these IPIs can be
high. This patch uses a simple structure that tracks CPUs that
potentially have TLB entries for pages being unmapped. When the unmapping
is complete, the full TLB is flushed on the assumption that a refill cost
is lower than flushing individual entries.
Architectures wishing to do this must give the following guarantee.
If a clean page is unmapped and not immediately flushed, the
architecture must guarantee that a write to that linear address
from a CPU with a cached TLB entry will trap a page fault.
This is essentially what the kernel already depends on but the window is
much larger with this patch applied and is worth highlighting. The
architecture should consider whether the cost of the full TLB flush is
higher than sending an IPI to flush each individual entry. An additional
architecture helper called flush_tlb_local is required. It's a trivial
wrapper with some accounting in the x86 case.
The impact of this patch depends on the workload as measuring any benefit
requires both mapped pages co-located on the LRU and memory pressure. The
case with the biggest impact is multiple processes reading mapped pages
taken from the vm-scalability test suite. The test case uses NR_CPU
readers of mapped files that consume 10*RAM.
Linear mapped reader on a 4-node machine with 64G RAM and 48 CPUs
4.2.0-rc1 4.2.0-rc1
vanilla flushfull-v7
Ops lru-file-mmap-read-elapsed 159.62 ( 0.00%) 120.68 ( 24.40%)
Ops lru-file-mmap-read-time_range 30.59 ( 0.00%) 2.80 ( 90.85%)
Ops lru-file-mmap-read-time_stddv 6.70 ( 0.00%) 0.64 ( 90.38%)
4.2.0-rc1 4.2.0-rc1
vanilla flushfull-v7
User 581.00 611.43
System 5804.93 4111.76
Elapsed 161.03 122.12
This is showing that the readers completed 24.40% faster with 29% less
system CPU time. From vmstats, it is known that the vanilla kernel was
interrupted roughly 900K times per second during the steady phase of the
test and the patched kernel was interrupts 180K times per second.
The impact is lower on a single socket machine.
4.2.0-rc1 4.2.0-rc1
vanilla flushfull-v7
Ops lru-file-mmap-read-elapsed 25.33 ( 0.00%) 20.38 ( 19.54%)
Ops lru-file-mmap-read-time_range 0.91 ( 0.00%) 1.44 (-58.24%)
Ops lru-file-mmap-read-time_stddv 0.28 ( 0.00%) 0.47 (-65.34%)
4.2.0-rc1 4.2.0-rc1
vanilla flushfull-v7
User 58.09 57.64
System 111.82 76.56
Elapsed 27.29 22.55
It's still a noticeable improvement with vmstat showing interrupts went
from roughly 500K per second to 45K per second.
The patch will have no impact on workloads with no memory pressure or have
relatively few mapped pages. It will have an unpredictable impact on the
workload running on the CPU being flushed as it'll depend on how many TLB
entries need to be refilled and how long that takes. Worst case, the TLB
will be completely cleared of active entries when the target PFNs were not
resident at all.
[sasha.levin@oracle.com: trace tlb flush after disabling preemption in try_to_unmap_flush]
Signed-off-by: Mel Gorman <mgorman@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Dave Hansen <dave.hansen@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Nikolay has reported a hang when a memcg reclaim got stuck with the
following backtrace:
PID: 18308 TASK: ffff883d7c9b0a30 CPU: 1 COMMAND: "rsync"
#0 __schedule at ffffffff815ab152
#1 schedule at ffffffff815ab76e
#2 schedule_timeout at ffffffff815ae5e5
#3 io_schedule_timeout at ffffffff815aad6a
#4 bit_wait_io at ffffffff815abfc6
#5 __wait_on_bit at ffffffff815abda5
#6 wait_on_page_bit at ffffffff8111fd4f
#7 shrink_page_list at ffffffff81135445
#8 shrink_inactive_list at ffffffff81135845
#9 shrink_lruvec at ffffffff81135ead
#10 shrink_zone at ffffffff811360c3
#11 shrink_zones at ffffffff81136eff
#12 do_try_to_free_pages at ffffffff8113712f
#13 try_to_free_mem_cgroup_pages at ffffffff811372be
#14 try_charge at ffffffff81189423
#15 mem_cgroup_try_charge at ffffffff8118c6f5
#16 __add_to_page_cache_locked at ffffffff8112137d
#17 add_to_page_cache_lru at ffffffff81121618
#18 pagecache_get_page at ffffffff8112170b
#19 grow_dev_page at ffffffff811c8297
#20 __getblk_slow at ffffffff811c91d6
#21 __getblk_gfp at ffffffff811c92c1
#22 ext4_ext_grow_indepth at ffffffff8124565c
#23 ext4_ext_create_new_leaf at ffffffff81246ca8
#24 ext4_ext_insert_extent at ffffffff81246f09
#25 ext4_ext_map_blocks at ffffffff8124a848
#26 ext4_map_blocks at ffffffff8121a5b7
#27 mpage_map_one_extent at ffffffff8121b1fa
#28 mpage_map_and_submit_extent at ffffffff8121f07b
#29 ext4_writepages at ffffffff8121f6d5
#30 do_writepages at ffffffff8112c490
#31 __filemap_fdatawrite_range at ffffffff81120199
#32 filemap_flush at ffffffff8112041c
#33 ext4_alloc_da_blocks at ffffffff81219da1
#34 ext4_rename at ffffffff81229b91
#35 ext4_rename2 at ffffffff81229e32
#36 vfs_rename at ffffffff811a08a5
#37 SYSC_renameat2 at ffffffff811a3ffc
#38 sys_renameat2 at ffffffff811a408e
#39 sys_rename at ffffffff8119e51e
#40 system_call_fastpath at ffffffff815afa89
Dave Chinner has properly pointed out that this is a deadlock in the
reclaim code because ext4 doesn't submit pages which are marked by
PG_writeback right away.
The heuristic was introduced by commit e62e384e9d ("memcg: prevent OOM
with too many dirty pages") and it was applied only when may_enter_fs
was specified. The code has been changed by c3b94f44fc ("memcg:
further prevent OOM with too many dirty pages") which has removed the
__GFP_FS restriction with a reasoning that we do not get into the fs
code. But this is not sufficient apparently because the fs doesn't
necessarily submit pages marked PG_writeback for IO right away.
ext4_bio_write_page calls io_submit_add_bh but that doesn't necessarily
submit the bio. Instead it tries to map more pages into the bio and
mpage_map_one_extent might trigger memcg charge which might end up
waiting on a page which is marked PG_writeback but hasn't been submitted
yet so we would end up waiting for something that never finishes.
Fix this issue by replacing __GFP_IO by may_enter_fs check (for case 2)
before we go to wait on the writeback. The page fault path, which is
the only path that triggers memcg oom killer since 3.12, shouldn't
require GFP_NOFS and so we shouldn't reintroduce the premature OOM
killer issue which was originally addressed by the heuristic.
As per David Chinner the xfs is doing similar thing since 2.6.15 already
so ext4 is not the only affected filesystem. Moreover he notes:
: For example: IO completion might require unwritten extent conversion
: which executes filesystem transactions and GFP_NOFS allocations. The
: writeback flag on the pages can not be cleared until unwritten
: extent conversion completes. Hence memory reclaim cannot wait on
: page writeback to complete in GFP_NOFS context because it is not
: safe to do so, memcg reclaim or otherwise.
Cc: stable@vger.kernel.org # 3.9+
[tytso@mit.edu: corrected the control flow]
Fixes: c3b94f44fc ("memcg: further prevent OOM with too many dirty pages")
Reported-by: Nikolay Borisov <kernel@kyup.com>
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull cgroup writeback support from Jens Axboe:
"This is the big pull request for adding cgroup writeback support.
This code has been in development for a long time, and it has been
simmering in for-next for a good chunk of this cycle too. This is one
of those problems that has been talked about for at least half a
decade, finally there's a solution and code to go with it.
Also see last weeks writeup on LWN:
http://lwn.net/Articles/648292/"
* 'for-4.2/writeback' of git://git.kernel.dk/linux-block: (85 commits)
writeback, blkio: add documentation for cgroup writeback support
vfs, writeback: replace FS_CGROUP_WRITEBACK with SB_I_CGROUPWB
writeback: do foreign inode detection iff cgroup writeback is enabled
v9fs: fix error handling in v9fs_session_init()
bdi: fix wrong error return value in cgwb_create()
buffer: remove unusued 'ret' variable
writeback: disassociate inodes from dying bdi_writebacks
writeback: implement foreign cgroup inode bdi_writeback switching
writeback: add lockdep annotation to inode_to_wb()
writeback: use unlocked_inode_to_wb transaction in inode_congested()
writeback: implement unlocked_inode_to_wb transaction and use it for stat updates
writeback: implement [locked_]inode_to_wb_and_lock_list()
writeback: implement foreign cgroup inode detection
writeback: make writeback_control track the inode being written back
writeback: relocate wb[_try]_get(), wb_put(), inode_{attach|detach}_wb()
mm: vmscan: disable memcg direct reclaim stalling if cgroup writeback support is in use
writeback: implement memcg writeback domain based throttling
writeback: reset wb_domain->dirty_limit[_tstmp] when memcg domain size changes
writeback: implement memcg wb_domain
writeback: update wb_over_bg_thresh() to use wb_domain aware operations
...
The name SWAP implies that we are dealing with anonymous pages only. In
fact, the original patch that introduced the min_unmapped_ratio logic
was to fix an issue related to file pages. Rename it to RECLAIM_UNMAP
to match what does.
Historically, commit a6dc60f897 ("vmscan: rename sc.may_swap to
may_unmap") renamed .may_swap to .may_unmap, leaving RECLAIM_SWAP
behind. commit 2e2e425989 ("vmscan,memcg: reintroduce sc->may_swap")
reintroduced .may_swap for memory controller.
Signed-off-by: Zhihui Zhang <zzhsuny@gmail.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Cc: Michal Hocko <mhocko@suse.cz>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Michal Hocko