linux-kernel - Re:Re: [PATCH] kernel/hung_task: Add a whitelist and blacklist mechanism.

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: <AHgAiADIDmen--evL59Deart.3.1618798927815.Hmail.zhouchuangao@vivo.com>
Date:   Mon, 19 Apr 2021 10:22:07 +0800 (GMT+08:00)
From:   周传高 <zhouchuangao@...o.com>
To:     Tetsuo Handa <penguin-kernel@...ove.sakura.ne.jp>
Cc:     Ingo Molnar <mingo@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Juri Lelli <juri.lelli@...hat.com>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Steven Rostedt <rostedt@...dmis.org>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Kees Cook <keescook@...omium.org>,
        Michal Hocko <mhocko@...e.com>,
        "Guilherme G. Piccoli" <gpiccoli@...onical.com>,
        Vlastimil Babka <vbabka@...e.cz>,
        Lukas Bulwahn <lukas.bulwahn@...il.com>,
        linux-kernel@...r.kernel.org
Subject: Re:Re: [PATCH] kernel/hung_task: Add a whitelist and blacklist mechanism.


>On 2021/04/17 23:13, zhouchuangao wrote:
>> The main purpose of this patch is to add a whitelist and blacklist
>> mechanism to the hung task thread.
>
>We stopped using the term 'whitelist'/'blacklist' for new code in Linux kernel,
>and what you are proposing is something like 'ignorelist'/'fatallist'.
>
>I think that matching based on comm name is poor, for comm name is subjected to
>impersonation by malicious user processes.
>
>Moreover, speak of syzkaller testing, most of hang task reports are reaction to
>somebody else consuming too much CPU resources (e.g. printk() flooding, too many
>pending workqueue requests). Even if some process is in 'ignorelist', it is
>possible that some problem that should be reported is already happening. Even if
>some process is in 'fatallist', it is possible that the cause of hang is simply
>somebody else is consuming too much CPU.
>
>By the way, I wish that khungtaskd can report recent top CPU consumers, for it is
>rare that the cause of hung is locking dependency problems / hardware problems.
>

Thank you for your suggestions,

Some SOC vendors' drivers or user-mode processes may be in D state for a long time,
and normally they do not configure HUNG TASK, so we need to ignore these tasks if
we use HUNG TASK. 

By default, ignorelist and fatallist are empty and can be configured by the user
(assuming the user knows what they are doing).

I will try to implement the function of khungtaskd to report top CPU consumers.

Thanks,
zhouchaungao