From: Michal Hocko <mhocko@kernel.org>
To: Alkis Georgopoulos <alkisg@gmail.com>
Cc: Andrew Morton <akpm@linux-foundation.org>,
linux-mm@kvack.org, bugzilla-daemon@bugzilla.kernel.org,
Mel Gorman <mgorman@techsingularity.net>,
Johannes Weiner <hannes@cmpxchg.org>
Subject: Re: [Bug 196157] New: 100+ times slower disk writes on 4.x+/i386/16+RAM, compared to 3.x
Date: Mon, 26 Jun 2017 11:12:54 +0200 [thread overview]
Message-ID: <20170626091254.GG11534@dhcp22.suse.cz> (raw)
In-Reply-To: <7b78db49-e0d8-9ace-bada-a48c9392a8ca@gmail.com>
On Mon 26-06-17 10:02:23, Alkis Georgopoulos wrote:
> IGBPI?I1I? 26/06/2017 08:46 I?I 1/4 , I? Michal Hocko I-I3I?I+-I?Iu:
> >Unfortunatelly, this is not something that can be applied in general.
> >This can lead to a premature OOM killer invocations. E.g. a direct write
> >to the block device cannot use highmem, yet there won't be anything to
> >throttle those writes properly. Unfortunately, our documentation is
> >silent about this setting. I will post a patch later.
>
>
> I should also note that highmem_is_dirtyable was 0 in all the 3.x kernel
> tests that I did; yet they didn't have the "slow disk writes" issue.
Yes this is possible. There were some changes in the dirty memory
throttling that could lead to visible behavior changes. I remember that
ab8fabd46f81 ("mm: exclude reserved pages from dirtyable memory") had
noticeable effect. The patch is something that we really want and it is
unnfortunate it has eaten some more from the dirtyable lowmem.
> I.e. I think that setting highmem_is_dirtyable=1 works around the issue, but
> is not the exact point which caused the regression that we see in 4.x
> kernels...
yes as I've said this is a workaround for for something that is an
inherent 32b lowmem/highmem issue.
--
Michal Hocko
SUSE Labs
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2017-06-26 9:12 UTC|newest]
Thread overview: 18+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <bug-196157-27@https.bugzilla.kernel.org/>
2017-06-22 19:37 ` [Bug 196157] New: 100+ times slower disk writes on 4.x+/i386/16+RAM, compared to 3.x Andrew Morton
2017-06-22 20:58 ` Alkis Georgopoulos
2017-06-23 7:13 ` Michal Hocko
2017-06-23 7:44 ` Alkis Georgopoulos
2017-06-23 11:38 ` Michal Hocko
2017-06-26 5:28 ` Alkis Georgopoulos
2017-06-26 5:46 ` Michal Hocko
2017-06-26 7:02 ` Alkis Georgopoulos
2017-06-26 9:12 ` Michal Hocko [this message]
2017-06-29 6:14 ` Alkis Georgopoulos
2017-06-29 7:16 ` Michal Hocko
2017-06-29 8:02 ` Alkis Georgopoulos
2018-04-19 20:36 ` Andrew Morton
[not found] <234273956.1792577.1524210929280.ref@mail.yahoo.com>
2018-04-20 7:55 ` Thierry
[not found] <328204943.8183321.1534496501208.ref@mail.yahoo.com>
2018-08-17 9:01 ` Thierry
2018-08-17 9:29 ` Michal Hocko
[not found] <1978465524.8206495.1534505385491.ref@mail.yahoo.com>
2018-08-17 11:29 ` Thierry
2018-08-17 14:46 ` 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=20170626091254.GG11534@dhcp22.suse.cz \
--to=mhocko@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=alkisg@gmail.com \
--cc=bugzilla-daemon@bugzilla.kernel.org \
--cc=hannes@cmpxchg.org \
--cc=linux-mm@kvack.org \
--cc=mgorman@techsingularity.net \
/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).