linux-kernel - Re: mm: 5.16 regression: reclaim_throttle leads to stall in near-OOM conditions

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <20211128043736.5a6dcc39@mail.inbox.lv>
Date:   Sun, 28 Nov 2021 04:37:36 +0900
From:   Alexey Avramov <hakavlad@...ox.lv>
To:     Mel Gorman <mgorman@...hsingularity.net>
Cc:     linux-mm@...ck.org, linux-kernel@...r.kernel.org, mhocko@...e.com,
        vbabka@...e.cz, neilb@...e.de, akpm@...ux-foundation.org,
        corbet@....net, riel@...riel.com, hannes@...xchg.org,
        david@...morbit.com, willy@...radead.org, hdanton@...a.com,
        penguin-kernel@...ove.sakura.ne.jp, oleksandr@...alenko.name,
        kernel@...mod.org, michael@...haellarabel.com, aros@....com,
        hakavlad@...il.com
Subject: Re: mm: 5.16 regression: reclaim_throttle leads to stall in
 near-OOM conditions

> I think there might be an unwritten mm law now that someone is always
> unhappy with OOM behaviour :(

It's okay if someone isn't happy with the default values.

It is not okay if there is no way to get the desired behavior using
tunables. And the problem is quite solvable [1].

[1] https://lore.kernel.org/linux-mm/20101028191523.GA14972@google.com/