mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Jonathan Morton <chromi@cyberspace.org>
To: Szabolcs Szakacsits <szaka@f-secure.com>
Cc: linux-kernel@vger.kernel.org
Subject: Re: [PATCH] Prevent OOM from killing init
Date: Sun, 25 Mar 2001 16:47:59 +0100	[thread overview]
Message-ID: <l03130320b6e3b6e3feea@[192.168.239.101]> (raw)
In-Reply-To: <Pine.LNX.4.30.0103251549100.13864-100000@fs131-224.f-secure.com>
In-Reply-To: <l03130315b6e242006a4b@[192.168.239.101]>

>> >start your app, wait for malloc to fail, hit enter for the other app and
>> >watch you app to be OOM killed ;)
>>
>> That would only happen if memory_overcommit was turned on, in which case my
>> modification would have zero effect anyway (the overcommit test happens
>> before my code).
>
>Thanks for listening and trying out the above trivial code instead of
>wrong theoretical arguments ;)
>
>So again, Linux *always* overcommit memory, the
>/proc/sys/vm/overcommit_memory controls total overcommit or
>quasi-overcommit [ehen you make your check in vm_enough() the memory is
>already overcommitted].

OK, looks like I got mixed up between *reservation* (malloc) and
*allocation* (access), and we're checking allocated memory when we should
really be checking reserved.  Be patient - I haven't done much of this type
of thing...  but your argument turns out to be correct, and I eventually
figured it out for myself.  I certainly agree that the default should be to
assume that all reserved memory will be used.  Maybe even do little nasty
things like printk(KERN_WARN "root is overcommitting memory!\n"); in
appropriate places, to discourage overcommitting.

>The solution is something like,
>add optional non-overcommit support,
>	http://lwn.net/2000/0406/a/no-overcommit.html

This sounds like a good solution.  Saw the size of the patch, it's big and
touches lots of bits of VM code, but it looks as though parts of my ideas
will also fit in there and be helpful.

Hmm... so we get an adjusted or replaced OOM-kill-selection algorithm, my
out_of_memory() fix and runaway process clamp, and this big(ish)
memory-accounting patch.  Sounds like a good combination to me, fixing all
the problems I've heard about recently.

There are some unrelated performance problems I've encountered during my
testing (eg. kswapd gets incredibly inefficient when swap usage grows
beyond about 500Mb on my 256Mb physical machine, causing swap bandwidth to
fall way below the HDs' capabilities), which I'm going to ignore for now.
Probably whoever takes on the VM balancing problem can look into that, as
it's probably related to that rather than this...

--------------------------------------------------------------
from:     Jonathan "Chromatix" Morton
mail:     chromi@cyberspace.org  (not for attachments)
big-mail: chromatix@penguinpowered.com
uni-mail: j.d.morton@lancaster.ac.uk

The key to knowledge is not to rely on people to teach you it.

Get VNC Server for Macintosh from http://www.chromatix.uklinux.net/vnc/

-----BEGIN GEEK CODE BLOCK-----
Version 3.12
GCS$/E/S dpu(!) s:- a20 C+++ UL++ P L+++ E W+ N- o? K? w--- O-- M++$ V? PS
PE- Y+ PGP++ t- 5- X- R !tv b++ DI+++ D G e+ h+ r++ y+(*)
-----END GEEK CODE BLOCK-----



       reply	other threads:[~2001-03-25 15:49 UTC|newest]

Thread overview: 153+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <Pine.LNX.4.30.0103251549100.13864-100000@fs131-224.f-secure.com>
     [not found] ` <l03130315b6e242006a4b@[192.168.239.101]>
2001-03-25 15:47   ` Jonathan Morton [this message]
2001-03-24 23:41 Benoit Garnier
2001-03-25  5:45 ` Stephen Satchell
2001-03-25  6:58   ` Stephen Clouse
2001-03-25 14:37   ` Martin Dalecki
2001-03-25 14:32 ` Martin Dalecki
  -- strict thread matches above, loose matches on Subject: below --
2001-03-24 10:18 Andries.Brouwer
2001-03-24  2:30 Andreas Franck
2001-03-24  1:38 Jonathan Morton
2001-03-24  1:11 Andries.Brouwer
2001-03-23 23:15 Andries.Brouwer
2001-03-23 23:17 ` Martin Dalecki
2001-03-24  0:13 ` Jonathan Morton
2001-03-24  6:58   ` Rik van Riel
2001-03-24 12:38   ` Jonathan Morton
2001-03-24 13:12   ` Jonathan Morton
2001-03-24  1:59 ` Paul Jakma
2001-03-23 19:33 Stephen Satchell
2001-03-23 18:29 Andries.Brouwer
2001-03-23 18:38 ` Alan Cox
2001-03-24  0:46   ` Tim Wright
2001-03-24 16:48   ` Jesse Pollard
2001-03-25 16:12     ` Szabolcs Szakacsits
2001-03-25 16:39     ` Jonathan Morton
2001-03-23 18:43 ` nick
2001-03-23 19:01   ` Martin Dalecki
2001-03-23 19:23     ` nick
2001-03-23 22:12     ` Alan Cox
2001-03-23 23:23       ` Stephen E. Clark
2001-03-24 10:40         ` Gérard Roudier
2001-03-23 21:14 ` Jonathan Morton
2001-03-25 14:56   ` Marco Colombo
2001-03-23  9:48 Only 10 MB/sec with via 82c686b - FIXED Alan Cox
2001-03-23 17:00 ` [PATCH] Prevent OOM from killing init SodaPop
2001-03-23 18:42   ` Martin Dalecki
2001-03-23 20:25     ` SodaPop
2001-03-23 20:33       ` Martin Dalecki
2001-03-23 19:19   ` Jonathan Morton
2001-03-23  9:28 Heusden, Folkert van
2001-03-23  0:09 Mikael Pettersson
2001-03-23  0:27 ` Andrew Morton
2001-03-23 12:29 ` Mikael Pettersson
2001-03-23 16:24 ` Horst von Brand
2001-03-23 16:49   ` Guest section DW
2001-03-23 17:04     ` Alan Cox
2001-03-22 23:35 Mikael Pettersson
2001-03-22 23:43 ` Alan Cox
2001-03-27  7:58   ` Helge Hafting
     [not found] <4605B269DB001E4299157DD1569079D2809930@EXCHANGE03.plaza.ds.adp.com>
2001-03-22 16:29 ` Rik van Riel
2001-03-22 18:32   ` Christian Bodmer
2001-03-23 15:08     ` Horst von Brand
2001-03-24  7:48       ` Doug Ledford
2001-03-24 10:21         ` Mike Galbraith
2001-03-24 18:19           ` Doug Ledford
2001-03-24 22:47             ` Mike Galbraith
2001-03-24 23:35             ` Jonathan Morton
2001-03-25 18:35               ` Jonathan Morton
2001-03-26  4:40                 ` Horst von Brand
2001-03-26  8:36                 ` Mike Galbraith
2001-03-26 10:01                 ` Jonathan Morton
2001-03-26 14:48                   ` Rik van Riel
2001-03-25 19:07               ` Mike Galbraith
2001-03-24 20:04           ` Jonathan Morton
2001-03-24 20:59           ` Jonathan Morton
2001-03-24 22:11             ` Rik van Riel
2001-03-24 23:36             ` Jonathan Morton
2001-03-25 14:30             ` Martin Dalecki
2001-03-25 14:13           ` Martin Dalecki
2001-03-24 12:42         ` Jonathan Morton
2001-03-24 15:06           ` Mike Galbraith
2001-03-25 14:10         ` Martin Dalecki
2001-03-22 11:08 Heusden, Folkert van
2001-03-21 23:41 Leif Sawyer
2001-03-22  0:32 ` Kevin Buhr
2001-03-21 22:54 Patrick O'Rourke
2001-03-21 23:11 ` Eli Carter
2001-03-21 23:40   ` Patrick O'Rourke
2001-03-21 23:48 ` Rik van Riel
2001-03-22  8:14   ` Eric W. Biederman
2001-03-22  9:24     ` Rik van Riel
2001-03-22 19:29     ` Philipp Rumpf
2001-03-22 11:47   ` Guest section DW
2001-03-22 15:01     ` Rik van Riel
2001-03-22 19:04       ` Guest section DW
2001-03-22 16:41     ` Eric W. Biederman
2001-03-22 20:28     ` Stephen Clouse
2001-03-22 21:01       ` Ingo Oeser
2001-03-22 21:23       ` Alan Cox
2001-03-22 22:00         ` Guest section DW
2001-03-22 22:12           ` Ed Tomlinson
2001-03-22 22:52           ` Alan Cox
2001-03-22 23:27             ` Guest section DW
2001-03-22 23:37               ` Rik van Riel
2001-03-26 19:04                 ` James Antill
2001-03-26 20:05                   ` Rik van Riel
2001-03-22 23:40               ` Alan Cox
2001-03-23 20:09                 ` Szabolcs Szakacsits
2001-03-23 22:21                   ` Alan Cox
2001-03-23 22:37                     ` Szabolcs Szakacsits
2001-03-23 19:57           ` Szabolcs Szakacsits
2001-03-22 22:10         ` Doug Ledford
2001-03-22 22:53           ` Alan Cox
2001-03-22 23:30             ` Doug Ledford
2001-03-22 23:40               ` Alan Cox
2001-03-22 23:43         ` Stephen Clouse
2001-03-23 19:26         ` Szabolcs Szakacsits
2001-03-23 20:41           ` Paul Jakma
2001-03-23 21:58             ` george anzinger
2001-03-24  5:55               ` Rik van Riel
2001-03-23 22:18             ` Szabolcs Szakacsits
2001-03-24  2:08               ` Paul Jakma
2001-03-27 15:05       ` Anthony de Boer - USEnet
2002-03-23  0:33       ` Martin Dalecki
2001-03-22 23:53         ` Rik van Riel
2002-03-23  1:21           ` Martin Dalecki
2001-03-23  0:20         ` Stephen Clouse
2002-03-23  1:30           ` Martin Dalecki
2001-03-23  1:37             ` Rik van Riel
2001-03-23 10:48               ` Martin Dalecki
2001-03-23 14:56                 ` Rik van Riel
2001-03-23 16:43                   ` Guest section DW
2001-03-24  5:57                     ` Rik van Riel
2001-03-25 16:35                       ` Guest section DW
2001-03-23 20:20                   ` Tom Diehl
2001-03-23 23:56                     ` Tim Wright
2001-03-24  0:21                       ` Tom Diehl
2001-03-23  1:31     ` Michael Peddemors
2001-03-23  7:04       ` Rik van Riel
2001-03-23 11:28         ` Guest section DW
2001-03-23 14:50           ` Eric W. Biederman
2001-03-23 17:21             ` Guest section DW
2001-03-23 20:18               ` Paul Jakma
2001-03-24 20:19                 ` Jesse Pollard
2001-03-23 23:48               ` Eric W. Biederman
2001-03-23 21:11           ` José Luis Domingo López
2001-03-23 17:26     ` James A. Sutherland
2001-03-23 17:32       ` Alan Cox
2001-03-23 18:58         ` Martin Dalecki
2001-03-23 19:45         ` Jonathan Morton
2001-03-23 23:26           ` Eric W. Biederman
2001-03-25 15:30         ` Martin Dalecki
2001-03-25 20:47         ` Stephen Satchell
2001-03-24  0:03       ` Guest section DW
2001-03-24  7:52       ` Doug Ledford
2001-03-25  0:32       ` Kurt Garloff
2001-03-25 15:02         ` Sandy Harris
2001-03-25 18:07         ` Guest section DW
2001-03-22 14:53   ` Patrick O'Rourke
2001-03-22 19:24   ` Philipp Rumpf
2001-03-22 22:20   ` James A. Sutherland
2001-03-23 17:31   ` Szabolcs Szakacsits
2001-03-24  5:54     ` Rik van Riel
2001-03-24  6:55       ` Juha Saarinen
2001-03-27  8:31       ` Roger Gammans

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='l03130320b6e3b6e3feea@[192.168.239.101]' \
    --to=chromi@cyberspace.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=szaka@f-secure.com \
    /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®