linux-kernel - Re: [PATCH 0/4] sched: Various reweight

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: <4035fa61-cc90-498b-af84-c9ac4903cd5f@amd.com>
Date: Wed, 11 Feb 2026 11:58:19 +0530
From: K Prateek Nayak <kprateek.nayak@....com>
To: Peter Zijlstra <peterz@...radead.org>
CC: Doug Smythies <dsmythies@...us.net>, <mingo@...nel.org>,
	<juri.lelli@...hat.com>, <vincent.guittot@...aro.org>,
	<dietmar.eggemann@....com>, <rostedt@...dmis.org>, <bsegall@...gle.com>,
	<mgorman@...e.de>, <vschneid@...hat.com>, <linux-kernel@...r.kernel.org>,
	<wangtao554@...wei.com>, <quzicheng@...wei.com>, <wuyun.abel@...edance.com>
Subject: Re: [PATCH 0/4] sched: Various reweight_entity() fixes

On 2/11/2026 1:34 AM, K Prateek Nayak wrote:
> My machine hasn't cracked yet with 30k instances. I'll too will prod
> more tomorrow.

So it did crack after a while and turns out it is not a result of
overflow but some other weirdness. I'll leave the full log below but
when I see pick returning NULL, the cfs_rq looks like:

    (Note: All logs have curr?(1) which is incorrect. The first "se"
     after the "cfs_rq" log is likely the actual "cfs_rq->curr".)

    cfs_rq of overflow:
    cfs_rq: depth(0) weight(2150629376) nr_queued(2051) sum_w_vruntime(5064132716358795264) sum_weight(2149580800) zero_vruntime(18446744071726874277) sum_shift(0) avg_vruntime(18446744071726874277)
    se: weight(1048576) vruntime(18446744069623207599) slice(2800000) deadline(18446744069626004775) curr?(1) task?(0)

    ^^^
    This is likely the curr since I print it first and it has a very
    different vruntime compared to others.

    se: weight(1048576) vruntime(373218199) slice(2800000) deadline(373446702) curr?(1) task?(0)
    se: weight(1048576) vruntime(371876685) slice(2800000) deadline(373815965) curr?(1) task?(0)
    se: weight(1048576) vruntime(372314317) slice(2800000) deadline(373817932) curr?(1) task?(0)
    se: weight(1048576) vruntime(371022075) slice(2800000) deadline(373819151) curr?(1) task?(0)
    ... few thousand more se but they have similar vruntime, slice, deadline


I'll go back and fix my debug patch to give a full view of series of
events that leads to this but in the meantime I've added the full log
below if someone wants to take a look.

-- 
Thanks and Regards,
Prateek

View attachment "crash.log" of type "text/plain" (478751 bytes)