From: Michal Hocko <mhocko@kernel.org>
To: Chris Edwards <chris.edwards@otago.ac.nz>
Cc: Johannes Weiner <hannes@cmpxchg.org>,
"linux-mm@kvack.org" <linux-mm@kvack.org>,
intel-gfx@lists.freedesktop.org,
Daniel Vetter <daniel.vetter@ffwll.ch>,
Chris Wilson <chris@chris-wilson.co.uk>
Subject: Re: Paging out when free memory is low but not exhausted (and available memory remains high)
Date: Mon, 10 Feb 2020 09:50:32 +0100 [thread overview]
Message-ID: <20200210085032.GA10636@dhcp22.suse.cz> (raw)
In-Reply-To: <1580684216453.76614@otago.ac.nz>
[Let's CC i915 people - the email thread started here:
http://lkml.kernel.org/r/e994b0a76099437b8edbae925e9ccac8@its-mail-p04.registry.otago.ac.nz
Long story short, there is a swap out activity even when there is a
clean page cache available that could be reclaimed and there is a
suspicion that the i915 shrinker might be involved. Could you guys have
a look please?]
On Sun 02-02-20 22:56:56, Chris Edwards wrote:
> ________________________________________
> From: Johannes Weiner <hannes@cmpxchg.org>
> Sent: Thursday, 30 January 2020 05:39
>
> > Chris, could you trace the i915 shrinker?
> >
> > Enable the shrinker trace point:
> >
> > # echo 1 >/sys/kernel/debug/tracing/events/i915/i915_gem_shrink/enable
> >
> > Then watch for events while the swapping is occuring:
> >
> > # cat /sys/kernel/debug/tracing/trace
>
> ________________________________________
>
> Thanks Johannes - that seems to confirm your suspicion: the kernel trace has a lot of contiguous entries relating to kswapd and the i915 shrinker:
>
> kswapd0-90 [001] .... 251812.619380: i915_gem_shrink: dev=0, target=243, flags=17
> kswapd0-90 [001] .... 251813.194977: i915_gem_shrink: dev=0, target=496, flags=3
> kswapd0-90 [001] .... 251813.196205: i915_gem_shrink: dev=0, target=850, flags=3
> kswapd0-90 [001] .... 251813.196207: i915_gem_shrink: dev=0, target=808, flags=17
> kswapd0-90 [001] .... 251813.399428: i915_gem_shrink: dev=0, target=470, flags=3
> kswapd0-90 [001] .... 251813.429971: i915_gem_shrink: dev=0, target=666, flags=3
> kswapd0-90 [001] .... 251813.430039: i915_gem_shrink: dev=0, target=259, flags=17
> kswapd0-90 [001] .... 251813.905045: i915_gem_shrink: dev=0, target=573, flags=3
> kswapd0-90 [001] .... 251813.905085: i915_gem_shrink: dev=0, target=174, flags=17
> kswapd0-90 [001] .N.. 251813.945542: i915_gem_shrink: dev=0, target=670, flags=3
> kswapd0-90 [001] .... 251813.945569: i915_gem_shrink: dev=0, target=474, flags=17
> kswapd0-90 [001] .... 251814.734973: i915_gem_shrink: dev=0, target=611, flags=3
> kswapd0-90 [001] .... 251814.735024: i915_gem_shrink: dev=0, target=417, flags=17
> kswapd0-90 [001] .... 251814.807983: i915_gem_shrink: dev=0, target=990, flags=3
> kswapd0-90 [001] .... 251814.807998: i915_gem_shrink: dev=0, target=962, flags=17
> kswapd0-90 [000] .... 251815.146950: i915_gem_shrink: dev=0, target=860, flags=3
> kswapd0-90 [000] .... 251815.146970: i915_gem_shrink: dev=0, target=758, flags=17
> kswapd0-90 [000] .... 251815.257367: i915_gem_shrink: dev=0, target=924, flags=3
> kswapd0-90 [000] .... 251815.257384: i915_gem_shrink: dev=0, target=832, flags=17
> kswapd0-90 [000] .... 251816.036147: i915_gem_shrink: dev=0, target=576, flags=3
> kswapd0-90 [000] .... 251816.036166: i915_gem_shrink: dev=0, target=140, flags=17
> kswapd0-90 [000] .... 251816.123591: i915_gem_shrink: dev=0, target=403, flags=3
> kswapd0-90 [000] .... 251816.124549: i915_gem_shrink: dev=0, target=546, flags=3
> kswapd0-90 [000] .... 251816.125415: i915_gem_shrink: dev=0, target=546, flags=3
> kswapd0-90 [000] .... 251816.126478: i915_gem_shrink: dev=0, target=546, flags=3
> kswapd0-90 [000] .... 251816.127380: i915_gem_shrink: dev=0, target=542, flags=3
> kswapd0-90 [000] .... 251816.128163: i915_gem_shrink: dev=0, target=522, flags=3
> kswapd0-90 [000] .... 251816.128748: i915_gem_shrink: dev=0, target=522, flags=3
> kswapd0-90 [000] .... 251816.129303: i915_gem_shrink: dev=0, target=522, flags=3
> kswapd0-90 [000] .... 251816.131522: i915_gem_shrink: dev=0, target=522, flags=3
> kswapd0-90 [000] .... 251816.133711: i915_gem_shrink: dev=0, target=522, flags=3
> kswapd0-90 [000] .... 251816.134317: i915_gem_shrink: dev=0, target=568, flags=3
> kswapd0-90 [000] .... 251816.134849: i915_gem_shrink: dev=0, target=568, flags=3
> kswapd0-90 [000] .... 251816.135647: i915_gem_shrink: dev=0, target=568, flags=3
> kswapd0-90 [000] .... 251816.233664: i915_gem_shrink: dev=0, target=483, flags=3
> kswapd0-90 [000] .... 251816.234041: i915_gem_shrink: dev=0, target=858, flags=3
> kswapd0-90 [000] .... 251816.234060: i915_gem_shrink: dev=0, target=612, flags=17
> kswapd0-90 [000] .... 251816.818646: i915_gem_shrink: dev=0, target=654, flags=3
> kswapd0-90 [000] .... 251816.818702: i915_gem_shrink: dev=0, target=400, flags=17
> kswapd0-90 [000] .... 251816.865547: i915_gem_shrink: dev=0, target=676, flags=3
> kswapd0-90 [000] .... 251816.865570: i915_gem_shrink: dev=0, target=499, flags=17
> kswapd0-90 [000] .... 251817.299401: i915_gem_shrink: dev=0, target=556, flags=3
> kswapd0-90 [000] .... 251817.299435: i915_gem_shrink: dev=0, target=355, flags=17
> kswapd0-90 [000] .... 251817.301099: i915_gem_shrink: dev=0, target=566, flags=3
> kswapd0-90 [000] .... 251817.301112: i915_gem_shrink: dev=0, target=512, flags=17
> kswapd0-90 [000] .... 251817.369283: i915_gem_shrink: dev=0, target=749, flags=3
> kswapd0-90 [000] .... 251817.369385: i915_gem_shrink: dev=0, target=115, flags=17
> kswapd0-90 [000] .... 251817.911052: i915_gem_shrink: dev=0, target=605, flags=3
> kswapd0-90 [000] .... 251817.911080: i915_gem_shrink: dev=0, target=474, flags=17
> kswapd0-90 [000] .... 251817.964954: i915_gem_shrink: dev=0, target=709, flags=3
> kswapd0-90 [000] .... 251817.964973: i915_gem_shrink: dev=0, target=589, flags=17
> kswapd0-90 [001] .... 251819.005872: i915_gem_shrink: dev=0, target=490, flags=3
>
> Thanks - and Michal, I hope you had a nice holiday. :)
> --
> Chris
--
Michal Hocko
SUSE Labs
prev parent reply other threads:[~2020-02-10 8:50 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-01-22 0:40 Chris Edwards
2020-01-23 12:31 ` Michal Hocko
2020-01-24 5:43 ` Chris Edwards
2020-01-24 10:04 ` Michal Hocko
2020-01-24 10:28 ` Michal Hocko
2020-01-27 10:06 ` Johannes Weiner
2020-01-28 3:22 ` Chris Edwards
2020-01-28 4:58 ` Chris Edwards
2020-01-28 7:19 ` Chris Edwards
2020-01-29 11:36 ` Chris Edwards
2020-01-29 13:39 ` Michal Hocko
2020-01-29 16:39 ` Johannes Weiner
2020-02-02 22:56 ` Chris Edwards
2020-02-10 8:50 ` Michal Hocko [this message]
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=20200210085032.GA10636@dhcp22.suse.cz \
--to=mhocko@kernel.org \
--cc=chris.edwards@otago.ac.nz \
--cc=chris@chris-wilson.co.uk \
--cc=daniel.vetter@ffwll.ch \
--cc=hannes@cmpxchg.org \
--cc=intel-gfx@lists.freedesktop.org \
--cc=linux-mm@kvack.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