linux-kernel - Re: help needed in debugging missed wakeup(?)

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: <20200323092754.GD14300@localhost.localdomain>
Date:   Mon, 23 Mar 2020 10:27:54 +0100
From:   Juri Lelli <juri.lelli@...hat.com>
To:     Lucas Stach <l.stach@...gutronix.de>
Cc:     Steven Rostedt <rostedt@...dmis.org>, linux-kernel@...r.kernel.org,
        Ingo Molnar <mingo@...hat.com>,
        Peter Zijlstra <peterz@...radead.org>,
        Vincent Guittot <vincent.guittot@...aro.org>,
        Dietmar Eggemann <dietmar.eggemann@....com>,
        Ben Segall <bsegall@...gle.com>, Mel Gorman <mgorman@...e.de>
Subject: Re: help needed in debugging missed wakeup(?)

Hi,

On 20/03/20 20:04, Lucas Stach wrote:

[...]

> My hope was that someone would could guide me a little with an educated
> guess, to make those trace_printks more helpful. The lockup is hard to
> trigger and there's other unrelated system activity going on, even when
> the relevant user task is stuck, so randomly sprinkling debug
> statements will generate lots of output to filter down.

I'd start by enabling sched:sched_pi_setprio (or adding trace_printks
around there to confirm your suspicion). Once that is confirmed, I'd try
to look at update_curr_dl::throttle and verify if the misbehaving task
was indeed throttled at some point; it shouldn't be, because dl_boosted
flag should be set for it at this point (maybe check this condition as
well).

Best,

Juri