linux-kernel - Re: [PATCH v3] sched/debug: Use sched_debug_lock to serialize use of cgroup

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]

Date:   Mon, 5 Apr 2021 11:59:18 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Waiman Long <longman@...hat.com>
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>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>,
        Daniel Bristot de Oliveira <bristot@...hat.com>,
        Bharata B Rao <bharata@...ux.vnet.ibm.com>,
        Phil Auld <pauld@...hat.com>,
        Daniel Thompson <daniel.thompson@...aro.org>,
        linux-kernel@...r.kernel.org
Subject: Re: [PATCH v3] sched/debug: Use sched_debug_lock to serialize use
 of cgroup_path[] only

On Sun, 4 Apr 2021 21:27:00 -0400
Waiman Long <longman@...hat.com> wrote:

> Thanks for the suggestion, but it also sound complicated.

It's not that complicated. Similar tricks have been used elsewhere in the
kernel.

> 
> I think we can fix this lockup problem if we are willing to lose some 

It's not a lockup problem, it's just a delay.

> information in case of contention. As you have suggested, a trylock will 
> be used to acquire sched_debug_lock. If succeeded, all is good. 
> Otherwise, a shorter stack buffer will be used for cgroup path. The path 
> may be truncated in this case. If we detect that the full length of the 
> buffer is used, we assume truncation and add, e.g. "...", to indicate 
> there is more to the actual path.
> 
> Do you think this is an acceptable comprise?

I guess that needs to be decided by those that use this information.

-- Steve