Bug 219492
| Summary: | amdgpu: failed to write reg 28b4 wait reg 28c6 and amdgpu: failed to write reg 1a6f4 wait reg 1a706 | ||
|---|---|---|---|
| Product: | Drivers | Reporter: | Shubhra Prakash Nandi (email2shubhra) |
| Component: | Video(DRI - non Intel) | Assignee: | drivers_video-dri |
| Status: | RESOLVED ANSWERED | ||
| Severity: | normal | CC: | oyvinds |
| Priority: | P3 | ||
| Hardware: | AMD | ||
| OS: | Linux | ||
| Kernel Version: | 6.6.x and 6.10.x | Subsystem: | |
| Regression: | No | Bisected commit-id: | |
| Attachments: | Kernel log for hardware | ||
Please first test 6.11.7 and if it's still reproducible, report here: https://gitlab.freedesktop.org/drm/amd/-/issues This still happens with 6.12.11 and a AMD Ryzen 5 PRO 4650U. [174713.452341] amdgpu 0000:06:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 [174721.488476] amdgpu 0000:06:00.0: amdgpu: failed to write reg 28b4 wait reg 28c6 [174721.494879] amdgpu 0000:06:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 [174721.501252] amdgpu 0000:06:00.0: amdgpu: failed to write reg 28b4 wait reg 28c6 [174721.507537] amdgpu 0000:06:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 [174780.330661] amdgpu 0000:06:00.0: amdgpu: failed to write reg 28b4 wait reg 28c6 [174780.337011] amdgpu 0000:06:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 [174780.360568] amdgpu 0000:06:00.0: amdgpu: failed to write reg 28b4 wait reg 28c6 [174780.367548] amdgpu 0000:06:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 It seems to be related to playing video with VAAPI in Firefox after suspend/resume. |
Created attachment 307209 [details] Kernel log for hardware I get the below two kernel errors in log when I resume from hibernate and while using DRI probably. I have latest amdgpu non-free firmware and mesa v24 installed. amdgpu 0000:04:00.0: amdgpu: failed to write reg 28b4 wait reg 28c6 amdgpu 0000:04:00.0: amdgpu: failed to write reg 1a6f4 wait reg 1a706 I have the following kernel parameter set to try and mitigate the above issue but still the issue happens. iommu=soft The above issue does not occur during s2idle sleep/wakeup but only after resuming from hibernate. Attaching some kernel log if that would be of any help. I found this issue with kernel version 6.6.x and 6.10.x