mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: "Sagatov, Evgeniy" <esagatov@amt.ru>
To: Thorsten Leemhuis <regressions@leemhuis.info>
Cc: "Rafael J. Wysocki" <rafael.j.wysocki@intel.com>,
	"regressions@lists.linux.dev" <regressions@lists.linux.dev>,
	LKML <linux-kernel@vger.kernel.org>,
	Michal Wilczynski <michal.wilczynski@intel.com>
Subject: RE: Pressing the power button causes the device to freeze completely
Date: Mon, 20 Apr 2026 22:20:09 +0000	[thread overview]
Message-ID: <773df6e6f0bd4120b9e39d51559ce6a8@amt.ru> (raw)
In-Reply-To: <34758f5c98864b8281cf89c30602981a@amt.ru>

I'd also like to point out that without this patch, single-core CPU performance does drop.
In my case, it dropped by 30%. I tested it using 7z, stress-ng and sysbench.

-----Original Message-----
From: Sagatov, Evgeniy 
Sent: Monday, April 20, 2026 10:35 PM
To: 'Thorsten Leemhuis' <regressions@leemhuis.info>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>; regressions@lists.linux.dev; LKML <linux-kernel@vger.kernel.org>; Michal Wilczynski <michal.wilczynski@intel.com>
Subject: RE: Pressing the power button causes the device to freeze completely

8c94ccc7cd691472461448f98e2372c75849406c    good
e37617c8e53a1f7fcba6d5e1041f4fd8a2425c27    good
b0d326da462e20285236e11e4cbc32085de9f363    bad

I also built the stock Debian 13 kernel, version 6.12.74, and reverted commit b0d326da462e20285236e11e4cbc32085de9f363.
The PC shuts down without any issues with this kernel when pressing the power button.

----
Sincerely, Evgeny Sagatov

-----Original Message-----
From: Thorsten Leemhuis <regressions@leemhuis.info>
Sent: Monday, April 20, 2026 9:55 AM
To: Sagatov, Evgeniy <esagatov@amt.ru>
Cc: Rafael J. Wysocki <rafael.j.wysocki@intel.com>; regressions@lists.linux.dev; LKML <linux-kernel@vger.kernel.org>; Michal Wilczynski <michal.wilczynski@intel.com>
Subject: Re: Pressing the power button causes the device to freeze completely

On 4/20/26 08:41, Sagatov, Evgeniy wrote:
> Hello Thorsten,
> 
> Thanks for the detailed manual!
> 
> I determined that the problem started between versions 6.7.12 and 6.8-rc1. Then I ran a 'git bisect' according to the manual.
> 
> Result:
> b0d326da462e20285236e11e4cbc32085de9f363 is the first bad commit 
> commit b0d326da462e20285236e11e4cbc32085de9f363
> Merge: 8c94ccc7cd69 e37617c8e53a
> Author: Linus Torvalds <torvalds@linux-foundation.org>
> Date:   Thu Jan 18 11:57:33 2024 -0800
> 
>     Merge tag 'sched-urgent-2024-01-18' of 
> git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip

This is a merge commit. This might be right, but often it is not. Can you please retest commits 8c94ccc7cd691472461448f98e2372c75849406c,
e37617c8e53a1f7fcba6d5e1041f4fd8a2425c27 and
b0d326da462e20285236e11e4cbc32085de9f363 just to be sure and then share your results? tia!

Ciao Thorsten

>     Pull scheduler fix from Ingo Molnar:
>      "Fix a cpufreq related performance regression on certain systems, where
>       the CPU would remain at the lowest frequency, degrading performance
>       substantially"
> 
>     * tag 'sched-urgent-2024-01-18' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
>       sched/fair: Fix frequency selection for non-invariant case
> 
>  kernel/sched/cpufreq_schedutil.c | 6 +++++-
>  1 file changed, 5 insertions(+), 1 deletion(-)
> 
> ----
> Sincerely, Evgeny Sagatov
> 
>  ----------------
> 16.04.2026, 20:35, Thorsten Leemhuis (mailto:regressions@leemhuis.info):
> To: Evgeny Sagatov (mailto:sagatov@ya.ru), Rafael J. Wysocki 
> (mailto:rafael.j.wysocki@intel.com);
> Copy: mailto:regressions@lists.linux.dev, LKML 
> (mailto:linux-kernel@vger.kernel.org), Michal Wilczynski 
> (mailto:michal.wilczynski@intel.com);
> Topic: Pressing the power button causes the device to freeze 
> completely;
>  
> On 4/16/26 19:22, Evgeny Sagatov wrote:
>  Could you please try 7.0 as well?
>  I compiled 7.0.0 kernel from sources.
>  Unfortunately, this kernel also freezes when the power button is pressed.
> 
> Thx for testing.
>   
>  How did you find that this causes the problem? Did you do a proper
>  bisection and thus checked if the parent of that commit really works?
>   
>  I don't think I did it correctly.
>  Because the previous commits didn't shut down the PC or cause it to
>  freeze. I think they just weren't working.
>   
>  I also tried a very simple module tity-power-button. Pressing the 
> power
>  button when using this module also causes a full freeze.
>  I think the problem lies with the ACPI functionality, but none of 
> kernel
>  options helped fix it.
>   
>  I'm confused.
> 
> Then you likely want to properly bisect the bug. Might be best to do 
> that on a Debian 12 install (where everything used to be working) by 
> following this guide:
> 
> https://docs.kernel.org/admin-guide/verify-bugs-and-bisect-regressions
> .html
> 
> But be warned: the regression is old. Even if the culprit is 
> identified the usual rules wrt to "no regression" might not apply, as 
> fixing the regression could cause regressions for other users.
> 
> Ciao, Thorsten
>  
>  ----------------
>  To: Evgeny Sagatov (mailto:sagatov@ya.ru), Rafael J. Wysocki
>  (mailto:rafael.j.wysocki@intel.com);
>  Copy: mailto:regressions@lists.linux.dev, LKML 
> (mailto:linux-kernel@vger.kernel.org),
>  Michal Wilczynski (mailto:michal.wilczynski@intel.com);
>  Topic: Pressing the power button causes the device to freeze 
> completely;
>  16.04.2026, 11:23, "Thorsten Leemhuis" <mailto:regressions@leemhuis.info>:
>  
>      [removing the stable list, this is not something for them]
>  
>      On 4/13/26 18:46, Evgeny Sagatov wrote:
>  
>           I have a PC with a motherboard GA-EP45T-UD3LR (Rev. 1.0) 
> <https://
>           https://www.gigabyte.com/Motherboard/GA-EP45T-UD3LR-rev-10
> <https://
>          https://www.gigabyte.com/Motherboard/GA-EP45T-UD3LR-rev-10>>
> and
>          processor Intel
>           Core 2 Quad Q9550.
>           Previously, I had Debian 12 with kernel 6.1 installed, and
>          pressing the
>           power button would shut down the PC as usual.
>           I updated to Debian 13 with kernel 6.12.74, and now pressing
>          the power
>           button causes the PC to freeze completely.
>            
>           I see that systemd doesn't even begin to shut down. It 
> freezes
>           immediately after pressing the button. There's no error 
> message in
>           console. There's no error message in the logs either.
>          Netconsole doesn't
>           report any errors, and the kernel doesn't panic. It just
>          freezes completely.
>            
>           I checked that Live CD with Debian 12 continue to shut down
>          normally,
>           but Live CD with Debian 13 freeze.
>           I tried updating and resetting the BIOS. I tried various 
> kernel
>           parameters related to ACPI settings. I also tried kernels
> 6.18
>          and 6.19.
>           Nothing fixed the problem.
>  
>  
>      Could you please try 7.0 as well? It contains a fix for the 
> commit you
>      mentioned below.
>       
>  
>            
>           If I unload the "button" module, the system doesn't freeze, 
> but
>          it also
>           doesn't shut down.
>           I built "button" module by rolling back its version to 
> commit
>          1a20d40
>           <https://github.com/torvalds/linux/ <https://github.com/
>          torvalds/linux/>
>           commit/1a20d409c874255086e2f42a729826d490294c91>, which
>          corresponds to
>           kernel version 6.1. This module does not freeze, but does 
> not
>          turn off
>           the PC either.
>            
>           I've found that the freezes have been occurring since commit
>          0d51157
>           <https://github.com/torvalds/linux/ <https://github.com/
>          torvalds/linux/>
>           commit/0d51157dfaac05ea66616d8a250dce04bef49a4f>.
>  
>  
>      That is 0d51157dfaac05 ("ACPI: button: Eliminate the driver 
> notify
>      callback") [v6.5-rc1] from Rafael. But that is a very old commit; 
> and
>      you seem to be switching distro versions and kernel at the same 
> time,
>      which could interfere.
>  
>      How did you find that this causes the problem? Did you do a 
> proper
>      bisection and thus checked if the parent of that commit really works?
>       
>  
>           Unfortunately, my skills are not enough to understand the
>          reason for the
>           freeze.
>            
>           I'm attaching dmesg and lspci logs.
>  
>      Ciao, Thorsten
> 



  reply	other threads:[~2026-04-20 22:20 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <184091776095270@mail.yandex.ru>
2026-04-16  8:22 ` Thorsten Leemhuis
     [not found]   ` <186071776359023@mail.yandex.ru>
2026-04-16 17:35     ` Thorsten Leemhuis
     [not found]       ` <26521776666243@mail.yandex.ru>
2026-04-20  6:41         ` Sagatov, Evgeniy
2026-04-20  6:54           ` Thorsten Leemhuis
2026-04-20 19:34             ` Sagatov, Evgeniy
2026-04-20 22:20               ` Sagatov, Evgeniy [this message]
2026-04-21  5:01                 ` Thorsten Leemhuis
2026-04-21  9:28                   ` Sagatov, Evgeniy
2026-04-21 15:11 ` Wysocki, Rafael J
2026-04-21 20:41   ` Rafael J. Wysocki
     [not found]     ` <168591776860440@mail.yandex.ru>
2026-04-22 12:36       ` Evgeny Sagatov
2026-04-22 13:26         ` Rafael J. Wysocki
2026-04-22 13:34           ` Evgeny Sagatov
2026-04-22 14:04             ` Rafael J. Wysocki
2026-04-22 14:48               ` Evgeny Sagatov
2026-04-23 13:17                 ` Rafael J. Wysocki
2026-04-23 14:51                   ` Evgeny Sagatov
2026-04-23 18:21                     ` Rafael J. Wysocki
2026-04-23 20:07                       ` Evgeny Sagatov
2026-04-24 14:40                         ` Rafael J. Wysocki
2026-04-24 17:06                           ` Evgeny Sagatov
2026-04-24 19:45                             ` Rafael J. Wysocki
2026-04-24 21:18                               ` Evgeny Sagatov
2026-04-26 14:50                                 ` Rafael J. Wysocki
2026-04-26 14:54                                   ` Rafael J. Wysocki
2026-04-26 20:45                                     ` Evgeny Sagatov
2026-04-27 18:48                                       ` Rafael J. Wysocki
2026-04-27 20:12                                         ` Evgeny Sagatov
2026-04-27 20:31                                           ` Rafael J. Wysocki
2026-04-27 21:49                                             ` Evgeny Sagatov
2026-04-28 17:40                                               ` Rafael J. Wysocki
2026-04-28 19:11                                                 ` Evgeny Sagatov
2026-04-28 19:58                                                   ` Rafael J. Wysocki
2026-04-28 21:05                                                     ` Evgeny Sagatov
2026-04-29 18:24                                                       ` Pressing the power button causes the device to freeze completely (schedutil involved) Rafael J. Wysocki
2026-04-29 20:22                                                         ` Rafael J. Wysocki
2026-04-29 21:16                                                           ` Evgeny Sagatov
2026-04-30 10:40                                                             ` Rafael J. Wysocki
2026-04-30 10:53                                                               ` Rafael J. Wysocki
2026-04-30 11:41                                                                 ` Evgeny Sagatov
2026-04-30 11:57                                                                   ` Evgeny Sagatov
2026-04-30 14:10                                                                     ` Rafael J. Wysocki
2026-04-30 16:04                                                                       ` Evgeny Sagatov
2026-04-30 13:57                                                                   ` Rafael J. Wysocki
2026-04-30 14:42                                                                     ` Rafael J. Wysocki
2026-04-30 23:05                                                                       ` Evgeny Sagatov
2026-04-30 23:17                                                                       ` Evgeny Sagatov
2026-05-01 12:00                                                                         ` Rafael J. Wysocki

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=773df6e6f0bd4120b9e39d51559ce6a8@amt.ru \
    --to=esagatov@amt.ru \
    --cc=linux-kernel@vger.kernel.org \
    --cc=michal.wilczynski@intel.com \
    --cc=rafael.j.wysocki@intel.com \
    --cc=regressions@leemhuis.info \
    --cc=regressions@lists.linux.dev \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox

all inboxes | Powered by JetHome®