linux-kernel - Re: BUG: bisected: thermald regression (MEMLEAK) in commit c7ff29763989bd09c433f73fae3c1e1c15d9cda4

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]

Date:   Thu, 27 Oct 2022 19:44:30 +0200
From:   Mirsad Goran Todorovac <mirsad.todorovac@....unizg.hr>
To:     srinivas pandruvada <srinivas.pandruvada@...ux.intel.com>,
        LKML <linux-kernel@...r.kernel.org>, rjw@...ysocki.net
Cc:     regressions@...ts.linux.dev, regressions@...mhuis.info,
        "linux-acpi@...r.kernel.org" <linux-acpi@...r.kernel.org>,
        Robert Moore <robert.moore@...el.com>, devel@...ica.org
Subject: Re: BUG: bisected: thermald regression (MEMLEAK) in commit
 c7ff29763989bd09c433f73fae3c1e1c15d9cda4

On 27. 10. 2022. 00:48, srinivas pandruvada wrote:

> On Wed, 2022-10-26 at 19:52 +0200, Mirsad Goran Todorovac wrote:
>> Dear all,
>>
>> On 24. 10. 2022. 20:56, Mirsad Goran Todorovac wrote:
>>> On 24. 10. 2022. 20:39, srinivas pandruvada wrote:
>>>
>>>>> Thank you for the patch. Unfortunately, when applied to v6.0.3
>>>>> it
>>>>> didn't
>>>>> fix the issue.
>>>> Thanks for the test. I copied to acpi and acpica mailing list.
>>>> Someone
>>>> can tell us what is this call doing wrong here.
> Of course it needs to be debugged. To start/stop systemctl service you
> need root access. If you have root access, there are other worse things
> can be done.

Indeed.

About fixing the bug, I tried to look at the source, but indeed it is 
above my means.
I don't have exactly experience in debugging kernel drivers.

However, I've noticed something else.

A command:

for a in {1..100}; do
     echo $a
     systemctl stop thermald
     sleep 1
     systemctl start thermald
     sleep 1
done

... it produces 180 memory leaks, attached in the attachment 1. 
Apparetnly, most of the processes
created 2 leaks.

This is different from the test of loop of 1000 iterations, but without 
sleep command, which
created only a handful of memleaks.

Sounds like a racing condition, rather than a deterministic bug.

It happens with the bisect done on Lenovo Ideapad 3 | Intel Core i5 | 
Ubuntu 22.04. Output of lshw
might be useful, so I have attached it as attachment 2.

This came as I was driving my bike home, but other than this I ran out 
of ideas how to help you in
debugging.

Thanks.

--
Mirsad Goran Todorovac
Sistem inženjer
Grafički fakultet | Akademija likovnih umjetnosti
Sveučilište u Zagrebu
-- 
System engineer
Faculty of Graphic Arts | Academy of Fine Arts
University of Zagreb, Republic of Croatia
The European Union

View attachment "thermald-memleak.wsleep1.txt" of type "text/plain" (222187 bytes)

View attachment "thermald-memleak.lshw.txt" of type "text/plain" (27653 bytes)