Bug 205335
Summary: | [amdgpu] suspend / screen black after thaw | ||
---|---|---|---|
Product: | Drivers | Reporter: | Arne Woerner (arne_woerner) |
Component: | Video(DRI - non Intel) | Assignee: | drivers_video-dri |
Status: | RESOLVED PATCH_ALREADY_AVAILABLE | ||
Severity: | high | CC: | alexdeucher |
Priority: | P1 | ||
Hardware: | x86-64 | ||
OS: | Linux | ||
Kernel Version: | 5.3.7-arch1-1-ARCH | Subsystem: | |
Regression: | No | Bisected commit-id: | |
Attachments: | some kernel messages regarding the amdgpu driver... |
*** Bug 204953 has been marked as a duplicate of this bug. *** Can you bisect? bisect means compiling a kernel again and again with different patches and testing? in a "nested intervals" fashion? i think that is too much load for my box... i even deleted the 5.1 kernel package from the pacman cache... the old linux-firmware packages, too... furthermore i'm afraid, that there are security issues with old kernels... i think it happened before... https://bugzilla.kernel.org/show_bug.cgi?id=198619 but this time it is worse sometimes: the screen remains dark after thaw... maybe it is the same reason? -arne bisecting is a git feature which allows you to identify what commit caused an issue. There were several suspend/resume issues related to multi-media engines that were fixed recently. Can you try v5.4-rc5? The fixes will be coming back to other kernels via stable. i am using 5.4.0-rc5ARNE now... my first self-compiled kernel since years... giggle OpenGL and btrfs and bcache and sound work fine... it cannot hibernate via s2disk (when i try it, it gets an EIO, immediately after it starts writing the image)... but # echo disk > /sys/power/state works fine so far... is there some kernel config option that i have to enable for s2disk? -arne it thaws nicely now... with 5.4.0-rc5... -arne |
Created attachment 285667 [details] some kernel messages regarding the amdgpu driver... since about kernel-5.2.8 my box does not thaw properly. during suspend it already behaves differently (the screen goes black while the image is being written...). first it was once per 10 thaws... but now it became quite bad (like every second thaw)... the GPU is a RX550: Device: Radeon 550 Series (POLARIS12, DRM 3.33.0, 5.3.7-arch1-1-ARCH, LLVM 9.0.0) (0x699f)