Hello,

I am reporting a problem in a fresh install of Debian 13 on ThinkPad P16s Gen 4
with AMD Ryzen AI 9 HX PRO 370 w/ Radeon 890M:

$ uname -r
6.12.57+deb13-amd64

Any attempt to suspend the laptop is unsuccessful, after a short black screen
the laptop immediately wakes again. After disabling each item in
/proc/acpi/wakeup to no avail, I checked dmesg to find these error messages:

[ 7008.878186] mt7925e 0000:c2:00.0: Message 00020007 (seq 10) timeout
[ 7008.878214] mt7925e 0000:c2:00.0: PM: pci_pm_suspend(): mt7925_pci_suspend
[mt7925e] returns -110
[ 7008.878226] mt7925e 0000:c2:00.0: PM: dpm_run_callback(): pci_pm_suspend
returns -110
[ 7008.878236] mt7925e 0000:c2:00.0: PM: failed to suspend async: error -110
[ 7008.878319] PM: Some devices failed to suspend, or early wake event detected

Unloading the module mt7925e indeed resolved the issue.

Thank you in advance for looking into this.

Lubos Boucek

Hi,

There were some bugfixes for mt7925 driver around 6.14-rc1. Can you
please test at least the version in backports (6.17.13-1~bpo13+1) to
see if you can observe the problem with this kernel as well?

Regards,
Salvatore

With this kernel:

$ uname -r
6.17.13+deb13-amd64

The laptop seems to suspend correctly, thanks!

Lubos Boucek

Hi

Ok that are great news.

Can you do further debugging on this issue? I see three major steps,
each in case of debugging "success".

- narrow down more the affected range of Debian versions where the
  issue got fixed.
- Bisect the upstream changes to identify a fixing commit
- Check if the commit can be backported to 6.12.y and resolving the
  issue.

I will outline only the first two first. So between 6.12.57-1 and
6.17.13-1 there are a couple of linux uploads happened:

6.17.13-1
6.17.13-1~bpo13+1
6.17.12-1
6.17.11-1
6.17.10-1
6.17.9-1
6.17.8-1
6.17.8-1~bpo13+1
6.17.7-2
6.17.7-1
6.17.6-1
6.17.5-1~exp1
6.17.2-1~exp1
6.16.12-2
6.16.12-1
6.16.12-1~bpo13+1
6.16.11-1
6.16.10-1
6.16.9-1
6.16.8-1
6.16.7-1
6.16.6-1
6.16.5-1
6.16.3-1
6.16.3-1~bpo13+1
6.16.1-1~exp1
6.16-1~exp1
6.16~rc7-1~exp1
6.15.6-1~exp1
6.15.5-1~exp1
6.15.4-1~exp1
6.15.3-1~exp1
6.15.2-1~exp1
6.15.1-1~exp1
6.15-1~exp1
6.15~rc7-1~exp1
6.14.6-1~exp1
6.14.5-1~exp1
6.14.3-1~exp1
6.13.11-1~exp1
6.13.10-1~exp1
6.13.9-1~exp1
6.13.8-1~exp1
6.13.7-1~exp1
6.13.6-1~exp1
6.13.5-1~exp1
6.13.4-1~exp1
6.13.3-1~exp1
6.13.2-1~exp1
6.13~rc7-1~exp1
6.13~rc6-1~exp1
6.12.63-1
6.12.57-1

From https://snapshot.debian.org/ you can fetch earlier and seen
uploads.so by querying
https://snapshot.debian.org/package/linux-signed-amd64/ you can get
earlier linux-image packages.

The idea would be to now do a "manual" bisect of those changes
identifying at least a major version range where things changes.

Assume you found that, by first searching and testing the major
version bumps that 6.13~rc7-1~exp1 exposes the problem but
6.14.3-1~exp1 will not anymore.

The next step would be to test the equivalent upstream versions and
check that this still holds true and then start the bisection. This
can be done as follows, again remember with the hypotetical above
versions (adapt as needed), and it would involve compiling and testing
a few kernels:

    git clone https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git
    cd linux-stable
    git checkout v6.14.3
    cp /boot/config-$(uname -r) .config
    yes '' | make localmodconfig
    make savedefconfig
    mv defconfig arch/x86/configs/my_defconfig

    # test 6.14.3 to ensure this is "good"
    make my_defconfig
    make -j $(nproc) bindeb-pkg
    ... install the resulting .deb package and confirm problem does not exist

    # test 6.13~rc7 to ensure this is "bad"
    git checkout v6.13-rc7
    make my_defconfig
    make -j $(nproc) bindeb-pkg
    ... install the resulting .deb package and confirm problem exists

With that confirmed, the bisection can start:

    git bisect start --term-new=fixed --term-old=broken
    git bisect fixed v6.14.3
    git bisect broken v6.13-rc7

In each bisection step git checks out a state between the oldest
known-bad and the newest known-good commit. In each step test using:

    make my_defconfig
    make -j $(nproc) bindeb-pkg
    ... install, try to boot / verify if problem exists

and if the problem is hit run:

    git bisect broken

and if the problem doesn't trigger run:

    git bisect fixed

. Please pay attention to always select the just built kernel for
booting, it won't always be the default kernel picked up by grub.

Iterate until git announces to have identified the first 'fixed'
commit.

Then provide the output of

    git bisect log

In the course of the bisection you might have to uninstall previous
kernels again to not exhaust the disk space in /boot. Also in the end
uninstall all self-built kernels again.

Can you please do this bisect?

Regards,
Salvatore

I am going to try this but it will take some time. I will let you know if I encounter any issues.

Lubos