From: Vlastimil Babka <vbabka@suse.cz>
To: Michal Hocko <mhocko@kernel.org>,
Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>,
Johannes Weiner <hannes@cmpxchg.org>,
Mel Gorman <mgorman@suse.de>,
David Rientjes <rientjes@google.com>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
Joonsoo Kim <js1304@gmail.com>,
Hillf Danton <hillf.zj@alibaba-inc.com>,
linux-mm@kvack.org, LKML <linux-kernel@vger.kernel.org>,
Michal Hocko <mhocko@suse.com>
Subject: Re: [PATCH 14/14] mm, oom, compaction: prevent from should_compact_retry looping for ever for costly orders
Date: Thu, 28 Apr 2016 10:59:22 +0200 [thread overview]
Message-ID: <5721D0EA.3020205@suse.cz> (raw)
In-Reply-To: <1461181647-8039-15-git-send-email-mhocko@kernel.org>
On 04/20/2016 09:47 PM, Michal Hocko wrote:
> From: Michal Hocko <mhocko@suse.com>
>
> "mm: consider compaction feedback also for costly allocation" has
> removed the upper bound for the reclaim/compaction retries based on the
> number of reclaimed pages for costly orders. While this is desirable
> the patch did miss a mis interaction between reclaim, compaction and the
> retry logic.
Hmm perhaps reversing the order of patches 13 and 14 would be a bit
safer wrt future bisections then? Add compaction_zonelist_suitable()
first with the reasoning, and then immediately use it in the other patch.
> The direct reclaim tries to get zones over min watermark
> while compaction backs off and returns COMPACT_SKIPPED when all zones
> are below low watermark + 1<<order gap. If we are getting really close
> to OOM then __compaction_suitable can keep returning COMPACT_SKIPPED a
> high order request (e.g. hugetlb order-9) while the reclaim is not able
> to release enough pages to get us over low watermark. The reclaim is
> still able to make some progress (usually trashing over few remaining
> pages) so we are not able to break out from the loop.
>
> I have seen this happening with the same test described in "mm: consider
> compaction feedback also for costly allocation" on a swapless system.
> The original problem got resolved by "vmscan: consider classzone_idx in
> compaction_ready" but it shows how things might go wrong when we
> approach the oom event horizont.
>
> The reason why compaction requires being over low rather than min
> watermark is not clear to me. This check was there essentially since
> 56de7263fcf3 ("mm: compaction: direct compact when a high-order
> allocation fails"). It is clearly an implementation detail though and we
> shouldn't pull it into the generic retry logic while we should be able
> to cope with such eventuality. The only place in should_compact_retry
> where we retry without any upper bound is for compaction_withdrawn()
> case.
>
> Introduce compaction_zonelist_suitable function which checks the given
> zonelist and returns true only if there is at least one zone which would
> would unblock __compaction_suitable if more memory got reclaimed. In
> this implementation it checks __compaction_suitable with NR_FREE_PAGES
> plus part of the reclaimable memory as the target for the watermark check.
> The reclaimable memory is reduced linearly by the allocation order. The
> idea is that we do not want to reclaim all the remaining memory for a
> single allocation request just unblock __compaction_suitable which
> doesn't guarantee we will make a further progress.
>
> The new helper is then used if compaction_withdrawn() feedback was
> provided so we do not retry if there is no outlook for a further
> progress. !costly requests shouldn't be affected much - e.g. order-2
> pages would require to have at least 64kB on the reclaimable LRUs while
> order-9 would need at least 32M which should be enough to not lock up.
>
> [vbabka@suse.cz: fix classzone_idx vs. high_zoneidx usage in
> compaction_zonelist_suitable]
> Signed-off-by: Michal Hocko <mhocko@suse.com>
Acked-by: Vlastimil Babka <vbabka@suse.cz>
next prev parent reply other threads:[~2016-04-28 8:59 UTC|newest]
Thread overview: 60+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-20 19:47 [PATCH 0.14] oom detection rework v6 Michal Hocko
2016-04-20 19:47 ` [PATCH 01/14] vmscan: consider classzone_idx in compaction_ready Michal Hocko
2016-04-21 3:32 ` Hillf Danton
2016-05-04 13:56 ` Michal Hocko
2016-04-20 19:47 ` [PATCH 02/14] mm, compaction: change COMPACT_ constants into enum Michal Hocko
2016-04-20 19:47 ` [PATCH 03/14] mm, compaction: cover all compaction mode in compact_zone Michal Hocko
2016-04-20 19:47 ` [PATCH 04/14] mm, compaction: distinguish COMPACT_DEFERRED from COMPACT_SKIPPED Michal Hocko
2016-04-21 7:08 ` Hillf Danton
2016-04-20 19:47 ` [PATCH 05/14] mm, compaction: distinguish between full and partial COMPACT_COMPLETE Michal Hocko
2016-04-21 6:39 ` Hillf Danton
2016-04-20 19:47 ` [PATCH 06/14] mm, compaction: Update compaction_result ordering Michal Hocko
2016-04-21 6:45 ` Hillf Danton
2016-04-20 19:47 ` [PATCH 07/14] mm, compaction: Simplify __alloc_pages_direct_compact feedback interface Michal Hocko
2016-04-21 6:50 ` Hillf Danton
2016-04-20 19:47 ` [PATCH 08/14] mm, compaction: Abstract compaction feedback to helpers Michal Hocko
2016-04-21 6:57 ` Hillf Danton
2016-04-28 8:47 ` Vlastimil Babka
2016-04-20 19:47 ` [PATCH 09/14] mm: use compaction feedback for thp backoff conditions Michal Hocko
2016-04-21 7:05 ` Hillf Danton
2016-04-28 8:53 ` Vlastimil Babka
2016-04-28 12:35 ` Michal Hocko
2016-04-29 9:16 ` Vlastimil Babka
2016-04-29 9:28 ` Michal Hocko
2016-04-20 19:47 ` [PATCH 10/14] mm, oom: rework oom detection Michal Hocko
2016-04-20 19:47 ` [PATCH 11/14] mm: throttle on IO only when there are too many dirty and writeback pages Michal Hocko
2016-04-20 19:47 ` [PATCH 12/14] mm, oom: protect !costly allocations some more Michal Hocko
2016-04-21 8:03 ` Hillf Danton
2016-05-04 6:01 ` Joonsoo Kim
2016-05-04 6:31 ` Joonsoo Kim
2016-05-04 8:56 ` Michal Hocko
2016-05-04 14:57 ` Joonsoo Kim
2016-05-04 18:19 ` Michal Hocko
2016-05-04 8:53 ` Michal Hocko
2016-05-04 14:39 ` Joonsoo Kim
2016-05-04 18:20 ` Michal Hocko
2016-04-20 19:47 ` [PATCH 13/14] mm: consider compaction feedback also for costly allocation Michal Hocko
2016-04-21 8:13 ` Hillf Danton
2016-04-20 19:47 ` [PATCH 14/14] mm, oom, compaction: prevent from should_compact_retry looping for ever for costly orders Michal Hocko
2016-04-21 8:24 ` Hillf Danton
2016-04-28 8:59 ` Vlastimil Babka [this message]
2016-04-28 12:39 ` Michal Hocko
2016-05-04 6:27 ` Joonsoo Kim
2016-05-04 9:04 ` Michal Hocko
2016-05-04 15:14 ` Joonsoo Kim
2016-05-04 19:22 ` Michal Hocko
2016-05-04 5:45 ` [PATCH 0.14] oom detection rework v6 Joonsoo Kim
2016-05-04 8:12 ` Vlastimil Babka
2016-05-04 8:32 ` Joonsoo Kim
2016-05-04 8:50 ` Michal Hocko
2016-05-04 8:47 ` Michal Hocko
2016-05-04 14:32 ` Joonsoo Kim
2016-05-04 18:16 ` Michal Hocko
2016-05-10 6:41 ` Joonsoo Kim
2016-05-10 7:09 ` Vlastimil Babka
2016-05-10 8:00 ` Joonsoo Kim
2016-05-10 9:44 ` Michal Hocko
2016-05-10 9:43 ` Michal Hocko
2016-05-12 2:23 ` Joonsoo Kim
2016-05-12 5:19 ` Joonsoo Kim
2016-05-12 10:59 ` Michal Hocko
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=5721D0EA.3020205@suse.cz \
--to=vbabka@suse.cz \
--cc=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--cc=hillf.zj@alibaba-inc.com \
--cc=js1304@gmail.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@suse.de \
--cc=mhocko@kernel.org \
--cc=mhocko@suse.com \
--cc=penguin-kernel@I-love.SAKURA.ne.jp \
--cc=rientjes@google.com \
--cc=torvalds@linux-foundation.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).