NetBSD-Bugs archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

kern/57059: amdgpu graphics ring test failing



>Number:         57059
>Category:       kern
>Synopsis:       amdgpu graphics ring test failing
>Confidential:   no
>Severity:       serious
>Priority:       medium
>Responsible:    kern-bug-people
>State:          open
>Class:          sw-bug
>Submitter-Id:   net
>Arrival-Date:   Sun Oct 16 03:35:01 +0000 2022
>Originator:     Bruno Melo
>Release:        9.99.100 current
>Organization:
>Environment:
>Description:
So the main problem is that amdgpu runs a test to see whether the
graphics ring is processing commands, and that test fails.  When that
happens, the driver tries to unwind through a series of error
branches, and trips over a secondary problem -- one of the error
branches is buggy.

Most likely the secondary problem is something like
amdgpu_fence_driver_init hasn't been called yet but the error branch
tries to call amdgpu_fence_driver_fini anyway and this hits the
assertion.  But we need to fix the main problem.

Transcribing some lines of the panic:

amdgpu0 {drm:netbsd:amdgpu_ring_test_helper+0x4a} *ERROR* ring gfx test failed -60
{drm:netbsd:amdgpu_device_init+0x1f2a} *ERROR* hw_init of IP block <gfx_v9_0> failed -60
amdgpu0: error: amdgpu_device_ip_init failed
amdgpu0: error: Fatal error during GPU init
[drm] amdgpu: Finishing device
panic: kernel diagnostic assertion "cv_is_valid(cv)" failed: file "/home/bruno/src/sys/kern/kern_condvar.c" line 107
>How-To-Repeat:
Boot on a AMD Ryzen 7 3700U laptop
>Fix:



Home | Main Index | Thread Index | Old Index