Hello,
One person on the Gitlab issue have potentially bisected the commit causing the kernel freeze: https://gitlab.freedesktop.org/drm/amd/-/issues/3787#note_2741901 https://gitlab.freedesktop.org/drm/kernel/-/commit/de848da12f752170c2ebe114804a985314fd5a6a Also I apologize Alex I think I sent you an email without cc the mailing list. On Tuesday, January 14th, 2025 at 3:02 PM, SyntheticBird <syntheticbird...@protonmail.com> wrote: > > > > > > > > On Tuesday, January 14th, 2025 at 2:44 PM, Alex Deucher alexdeuc...@gmail.com > wrote: > > > On Tue, Jan 14, 2025 at 9:05 AM SyntheticBird > > syntheticbird...@protonmail.com wrote: > > > > > Hello and happy new year to all members of this list. > > > > > > I know that mailing lists aren't meant for begging for support, but after > > > discussing this in another distribution channel, it seems to me like it > > > is the only way for me and other users to grab the attention of a > > > maintainer. > > > > > > Since 6.12.1, several users have reported a kernel freeze issue with > > > AMDGPU, including a complete stack trace. It corresponds to the following > > > issues on GitLab: > > > > > > https://gitlab.freedesktop.org/drm/amd/-/issues/3787https://gitlab.freedesktop.org/drm/amd/-/issues/3866 > > > > > > I also suspect this issue may have happened to one user on the Level1Tech > > > forum: > > > https://forum.level1techs.com/t/help-with-troubleshooting-freezing-with-linux-kernel-6-12-on-arch/221340/22 > > > > > > Two months have passed since the first report of this bug, and no one in > > > either of the two issues has ever reached out to us. I've checked the > > > mailing list archive, and it seems like no maintainer has ever discussed > > > it. We're simply left alone with only hope. > > > > > > I perfectly understand that maintainers are busy, and I do not blame > > > them, but we as users and owners of AMDGPU have no other choice but to > > > try reaching out to you, to at least be sure that this issue hasn't been > > > ignored in the depths of the GitLab issue list. > > > > > > This issue is impacting versions 6.12.1 to the latest 6.13 rc. Several > > > users are still stuck on either 6.6 LTS or 6.11.9-10. For users relying > > > on features that have been introduced between 6.6 and 6.12, they cannot > > > downgrade to 6.6 and also cannot upgrade to 6.12, leaving them without > > > vulnerabilities being patched on mainline and LTS. > > > > Per my comments on the bug report, did you also update mesa or GPU > > firmware? If so, can you narrow down if any of those other components > > had an impact? Alternatively, can you bisect the kernel? If you have > > known good and bad kernels, it would be really helpful if you could > > bisect to identify what commit caused the regression for you. > > > > Alex > > > Hi, > > Thanks Alex for reaching out. Also per my comment on the bug report, I can't > right now but will be able to bisect during this week-end. > I do have two versions of the kernel one good (6.11.10) and one bad (6.12.1). > I checked and both mesa (24.3.3) and firmware (20241210) are up to date. > > I will update here whenever I find the commit.
