Summary: | Navi10 null pointer dereference in kernel 5.4 while running Vulkan CTS | ||||||
---|---|---|---|---|---|---|---|
Product: | DRI | Reporter: | Timur Kristóf <venemo> | ||||
Component: | DRM/AMDgpu | Assignee: | Default DRI bug account <dri-devel> | ||||
Status: | RESOLVED MOVED | QA Contact: | |||||
Severity: | normal | ||||||
Priority: | not set | CC: | christian.koenig | ||||
Version: | DRI git | ||||||
Hardware: | All | ||||||
OS: | All | ||||||
Whiteboard: | |||||||
i915 platform: | i915 features: | ||||||
Attachments: |
|
I think that's a bug in dma-buf fixed with this patch: https://cgit.freedesktop.org/drm/drm-misc/commit/?id=7fbd0782bf94d46993a7e5c08c0feda8dcab1f76 FWIW that patch fixes the mentioned stacktrace for me. Seems like the patch is only enqueued for drm-misc-next, shouldn't this be in drm-misc-fixes? Christian was going to commit it to the fixes branch as well. This patch is already on the fixes branch for a while. -- GitLab Migration Automatic Message -- This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity. You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/933. |
Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.
Created attachment 145698 [details] dmesg log with kernel 5.4-rc2 that shows the null pointer dereference While running the Vulkan CTS, I encountered the following: Running the cases dEQP-VK.memory.allocation.* will trigger a kernel NULL pointer dereference in amdgpu on 5.4-rc2, see the attached dmesg log. After this happens, the GPU can hang too, but doesn't always hang. Hardware: AMD Radeon RX 5700XT (reference card, made by Sapphire) Dell U2718Q (two of them: one through HDMI, other through DisplayPort) AMD Ryzen 7 3700X MSI B450i Software: Fedora Workstation 31 x86_64 LLVM 9 Mesa master (what will be 19.3)