From: Andrew Morton <akpm@linux-foundation.org>
To: Michal Hocko <mhocko@kernel.org>
Cc: Mel Gorman <mgorman@suse.de>,
Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>,
Rik van Riel <riel@redhat.com>,
Johannes Weiner <hannes@cmpxchg.org>,
Vlastimil Babka <vbabka@suse.cz>, <linux-mm@kvack.org>,
LKML <linux-kernel@vger.kernel.org>,
Michal Hocko <mhocko@suse.com>
Subject: Re: [PATCH] mm, vmscan: do not loop on too_many_isolated for ever
Date: Wed, 19 Jul 2017 15:20:14 -0700 [thread overview]
Message-ID: <20170719152014.53a861c57bcb636d6cd9d002@linux-foundation.org> (raw)
In-Reply-To: <20170710074842.23175-1-mhocko@kernel.org>
On Mon, 10 Jul 2017 09:48:42 +0200 Michal Hocko <mhocko@kernel.org> wrote:
> From: Michal Hocko <mhocko@suse.com>
>
> Tetsuo Handa has reported [1][2][3]that direct reclaimers might get stuck
> in too_many_isolated loop basically for ever because the last few pages
> on the LRU lists are isolated by the kswapd which is stuck on fs locks
> when doing the pageout or slab reclaim. This in turn means that there is
> nobody to actually trigger the oom killer and the system is basically
> unusable.
>
> too_many_isolated has been introduced by 35cd78156c49 ("vmscan: throttle
> direct reclaim when too many pages are isolated already") to prevent
> from pre-mature oom killer invocations because back then no reclaim
> progress could indeed trigger the OOM killer too early. But since the
> oom detection rework 0a0337e0d1d1 ("mm, oom: rework oom detection")
> the allocation/reclaim retry loop considers all the reclaimable pages
> and throttles the allocation at that layer so we can loosen the direct
> reclaim throttling.
>
> Make shrink_inactive_list loop over too_many_isolated bounded and returns
> immediately when the situation hasn't resolved after the first sleep.
> Replace congestion_wait by a simple schedule_timeout_interruptible because
> we are not really waiting on the IO congestion in this path.
>
> Please note that this patch can theoretically cause the OOM killer to
> trigger earlier while there are many pages isolated for the reclaim
> which makes progress only very slowly. This would be obvious from the oom
> report as the number of isolated pages are printed there. If we ever hit
> this should_reclaim_retry should consider those numbers in the evaluation
> in one way or another.
Need to figure out which kernels to patch. Maybe just 4.13-rc after a
week or two?
> --- a/mm/vmscan.c
> +++ b/mm/vmscan.c
> @@ -1713,9 +1713,15 @@ shrink_inactive_list(unsigned long nr_to_scan, struct lruvec *lruvec,
> int file = is_file_lru(lru);
> struct pglist_data *pgdat = lruvec_pgdat(lruvec);
> struct zone_reclaim_stat *reclaim_stat = &lruvec->reclaim_stat;
> + bool stalled = false;
>
> while (unlikely(too_many_isolated(pgdat, file, sc))) {
> - congestion_wait(BLK_RW_ASYNC, HZ/10);
> + if (stalled)
> + return 0;
> +
> + /* wait a bit for the reclaimer. */
> + schedule_timeout_interruptible(HZ/10);
a) if this task has signal_pending(), this falls straight through
and I suspect the code breaks?
b) replacing congestion_wait() with schedule_timeout_interruptible()
means this task no longer contributes to load average here and it's
a (slightly) user-visible change.
c) msleep_interruptible() is nicer
d) IOW, methinks we should be using msleep() here?
> + stalled = true;
>
> /* We are about to die and free our memory. Return now. */
> if (fatal_signal_pending(current))
(Gets distracted by the thought that we should do
s/msleep/msleep_uninterruptible/g)
next prev parent reply other threads:[~2017-07-19 22:20 UTC|newest]
Thread overview: 41+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-07-10 7:48 [PATCH] mm, vmscan: do not loop on too_many_isolated for ever Michal Hocko
2017-07-10 13:16 ` Vlastimil Babka
2017-07-10 13:58 ` Rik van Riel
2017-07-10 16:58 ` Johannes Weiner
2017-07-10 17:09 ` Michal Hocko
2017-07-19 22:20 ` Andrew Morton [this message]
2017-07-20 6:56 ` Michal Hocko
2017-07-21 23:01 ` Andrew Morton
2017-07-24 6:50 ` Michal Hocko
2017-07-20 1:54 ` Hugh Dickins
2017-07-20 10:44 ` Tetsuo Handa
2017-07-24 7:01 ` Hugh Dickins
2017-07-24 11:12 ` Tetsuo Handa
2017-07-20 13:22 ` Michal Hocko
2017-07-24 7:03 ` Hugh Dickins
-- strict thread matches above, loose matches on Subject: below --
2017-03-07 13:30 Michal Hocko
2017-03-07 19:52 ` Rik van Riel
2017-03-08 9:21 ` Michal Hocko
2017-03-08 15:54 ` Rik van Riel
2017-03-09 9:12 ` Michal Hocko
2017-03-09 14:16 ` Rik van Riel
2017-03-09 14:59 ` Michal Hocko
2017-03-09 18:05 ` Johannes Weiner
2017-03-09 22:18 ` Rik van Riel
2017-03-10 10:27 ` Michal Hocko
2017-03-10 10:20 ` Michal Hocko
2017-03-10 11:44 ` Tetsuo Handa
2017-03-21 10:37 ` Tetsuo Handa
2017-04-23 10:24 ` Tetsuo Handa
2017-04-24 12:39 ` Stanislaw Gruszka
2017-04-24 13:06 ` Tetsuo Handa
2017-04-25 6:33 ` Stanislaw Gruszka
2017-06-30 0:14 ` Tetsuo Handa
2017-06-30 13:32 ` Michal Hocko
2017-06-30 15:59 ` Tetsuo Handa
2017-06-30 16:19 ` Michal Hocko
2017-07-01 11:43 ` Tetsuo Handa
2017-07-05 8:19 ` Michal Hocko
2017-07-05 8:20 ` Michal Hocko
2017-07-06 10:48 ` Tetsuo Handa
2017-03-09 14:31 ` Mel Gorman
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=20170719152014.53a861c57bcb636d6cd9d002@linux-foundation.org \
--to=akpm@linux-foundation.org \
--cc=hannes@cmpxchg.org \
--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=riel@redhat.com \
--cc=vbabka@suse.cz \
/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).