mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Michal Hocko <mhocko@kernel.org>
To: Mel Gorman <mgorman@techsingularity.net>
Cc: Trevor Cordes <trevor@tecnopolis.ca>,
	linux-kernel@vger.kernel.org,
	Joonsoo Kim <iamjoonsoo.kim@lge.com>,
	Minchan Kim <minchan@kernel.org>, Rik van Riel <riel@surriel.com>,
	Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Subject: Re: mm, vmscan: commit makes PAE kernel crash nightly (bisected)
Date: Wed, 25 Jan 2017 10:46:37 +0100	[thread overview]
Message-ID: <20170125094637.GE32377@dhcp22.suse.cz> (raw)
In-Reply-To: <20170123110412.z5e372pabnnezdnu@techsingularity.net>

On Mon 23-01-17 11:04:12, Mel Gorman wrote:
[...]
> 1. In should_reclaim_retry, account for SLAB_RECLAIMABLE as available
>    pages when deciding to retry reclaim

I am pretty sure I have considered this but then decided to not go that
way. I do not remember details so I will think about this some more. It
might have been just "let's wait for the real issue here". Anyway we can
give it a try and it would be as simple as
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 94ebd30d0f09..87221491be84 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -3566,7 +3566,7 @@ should_reclaim_retry(gfp_t gfp_mask, unsigned order,
 		unsigned long min_wmark = min_wmark_pages(zone);
 		bool wmark;
 
-		available = reclaimable = zone_reclaimable_pages(zone);
+		available = reclaimable = zone_reclaimable_pages(zone) + zone_page_state_snapshot(zone, NR_SLAB_RECLAIMABLE);
 		available -= DIV_ROUND_UP((*no_progress_loops) * available,
 					  MAX_RECLAIM_RETRIES);
 		available += zone_page_state_snapshot(zone, NR_FREE_PAGES);

I am not sure it would really help much on its own without further
changes to how we scale LRU->slab scanning. Could you give this a try
on top of the mmotm or linux-next tree?

> 2. Stall in should_reclaim_retry for __GFP_NOFAIL|__GFP_NOFS with a
>    comment stating that the intent is to allow kswapd make progress
>    with the shrinker

The current mmotm tree doesnt need this because we no longer trigger the
oom killer for this combinations of flags.

> 3. Stall __GFP_NOFS in direct reclaimer on a workqueue when it's
>    failing to make progress to allow kswapd to do some work. This
>    may be impaired if kswapd is locked up waiting for a lock held
>    by the direct reclaimer
> 4. Schedule the system workqueue to drain slab for
>    __GFP_NOFS|__GFP_NOFAIL.
> 
> 3 and 4 are extremely heavy handed so we should try them one at a time.

I am not even sure they are really necessary.
-- 
Michal Hocko
SUSE Labs

  reply	other threads:[~2017-01-25  9:46 UTC|newest]

Thread overview: 40+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-01-11 10:32 Trevor Cordes
2017-01-11 12:11 ` Mel Gorman
2017-01-11 12:14   ` Mel Gorman
2017-01-11 22:52     ` Trevor Cordes
2017-01-12  9:36       ` Michal Hocko
2017-01-15  6:27         ` Trevor Cordes
2017-01-16 11:09           ` Mel Gorman
2017-01-17 13:52             ` Michal Hocko
2017-01-17 14:21               ` Mel Gorman
2017-01-17 14:54                 ` Michal Hocko
2017-01-18  7:25                   ` Trevor Cordes
2017-01-18 17:48                   ` Mel Gorman
2017-01-18 18:07                   ` Mel Gorman
2017-01-19  9:48                   ` Trevor Cordes
2017-01-19 11:37                     ` Michal Hocko
2017-01-20  6:35                       ` Trevor Cordes
2017-01-20 11:02                         ` Mel Gorman
2017-01-20 15:55                           ` Mel Gorman
2017-01-23  0:45                             ` Trevor Cordes
2017-01-23 10:48                               ` Mel Gorman
2017-01-23 11:04                                 ` Mel Gorman
2017-01-25  9:46                                   ` Michal Hocko [this message]
2017-01-24 12:59                                 ` Michal Hocko
2017-01-25 10:02                                 ` Trevor Cordes
2017-01-25 12:04                                   ` Michal Hocko
2017-01-29 22:50                                     ` Trevor Cordes
2017-01-30  7:51                                       ` Michal Hocko
2017-02-01  9:29                                         ` Trevor Cordes
2017-02-01 10:14                                           ` Michal Hocko
2017-02-04  0:36                                             ` Trevor Cordes
2017-02-04 20:05                                               ` Rik van Riel
2017-02-05 10:03                                               ` Michal Hocko
2017-02-05 22:53                                                 ` Trevor Cordes
2017-01-30  9:10                                       ` Mel Gorman
2017-01-24 12:54                               ` Michal Hocko
2017-01-26 23:18                                 ` Trevor Cordes
2017-01-27  7:36                                   ` Michal Hocko
2017-01-24 12:51                         ` Michal Hocko
2017-01-18  6:52             ` Trevor Cordes
2017-01-17 13:45           ` 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=20170125094637.GE32377@dhcp22.suse.cz \
    --to=mhocko@kernel.org \
    --cc=iamjoonsoo.kim@lge.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mgorman@techsingularity.net \
    --cc=minchan@kernel.org \
    --cc=riel@surriel.com \
    --cc=srikar@linux.vnet.ibm.com \
    --cc=trevor@tecnopolis.ca \
    /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

all inboxes | Powered by JetHome®