linux-kernel - Re: schedstat false counting of domain load_balance() tried to move one or more tasks failed

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] [day] [month] [year] [list]

Message-ID: <b6aa59f8-3518-4b05-3f4a-0eab443c6082@bytedance.com>
Date:   Fri, 22 Jul 2022 11:51:26 +0800
From:   Abel Wu <wuyun.abel@...edance.com>
To:     Steven Rostedt <rostedt@...dmis.org>,
        Peter Zijlstra <peterz@...radead.org>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Ingo Molnar <mingo@...nel.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Valentin Schneider <vschneid@...hat.com>,
        Linus Torvalds <torvalds@...ux-foundation.org>
Subject: Re: schedstat false counting of domain load_balance() tried to move
 one or more tasks failed


On 7/19/22 12:42 AM, Steven Rostedt Wrote:
> On Mon, 18 Jul 2022 11:51:26 +0200
> Peter Zijlstra <peterz@...radead.org> wrote:
> 
>>> Do we care? Should it be fixed? Should it be documented?
>>
>> *shrug*, I suppose we can fix. People using this stuff are the sort that
>> are likely to read documentation instead of code.
> 
> Yep.
> 
>>
>> At the same time; I suspect it's been 'broken' like forever, so who
>> knows what people are actually expecting today.
> 
> As you stated, it's used by people that read documentation more than the
> code. My expectation is that they are making wrong decisions because what
> they expect those numbers to mean are not what is actually happening.
> 
> I think it's better to make the functionality match the documentation, and
> if people complain, we can ask them what exactly they expected and why. And
> perhaps they might be complaining that a benchmark isn't behaving as
> expected because they were interpreting the results incorrectly.
> 
> I'll go write up a fix.

Hi Steven, Peter,

There are also some other fields broken I think, such as the imbalance
calculation. It is changed after Vincent's reworking load_balance().
Just for the record if you are planning on a fix. :)

Best regards,
Abel