FS#52227 - [linux] amdgpu_init / hw_init of IP block <gfx_v8_0> failed → black screen
Attached to Project:
Arch Linux
Opened by Luca Steinke (phre4k) - Wednesday, 21 December 2016, 10:12 GMT
Last edited by Laurent Carlier (lordheavy) - Monday, 21 May 2018, 06:44 GMT
Opened by Luca Steinke (phre4k) - Wednesday, 21 December 2016, 10:12 GMT
Last edited by Laurent Carlier (lordheavy) - Monday, 21 May 2018, 06:44 GMT
|
Details
Description:
Boot fails with kernel errors related to amdgpu. Hardware: Powercolor Radeon RX480 Red Devil. Additional info: * linux 4.8-13-1-ARCH journalctl -b | grep kernel attached. Maybe relevant line: [drm:amdgpu_device_init [amdgpu]] *ERROR* hw_init of IP block <gfx_v8_0> failed -22 Steps to reproduce: Have a Powercolor RX480 and a recent kernel, I guess? Can't test with other hardware :/ |
This task depends upon
Closed by Laurent Carlier (lordheavy)
Monday, 21 May 2018, 06:44 GMT
Reason for closing: No response
Monday, 21 May 2018, 06:44 GMT
Reason for closing: No response
Comment by
Laurent Carlier (lordheavy) -
Wednesday, 21 December 2016, 19:08 GMT
Comment by Luca Steinke (phre4k) -
Wednesday, 21 December 2016, 22:36 GMT
Comment by Luca Steinke (phre4k) -
Wednesday, 21 December 2016, 22:39 GMT
Comment by Waoi Chun (parcho) -
Tuesday, 24 January 2017, 15:30 GMT
Comment by
Laurent Carlier (lordheavy) -
Saturday, 21 October 2017, 04:33 GMT
Your card fail to report the amount of VRAM available. Is it the
original BIOS or have you flashed your card with an alternative
one?
It's the original "silent" BIOS. When I noticed the error, I
updated to the newest BIOS via AtiWinFlash successfully, but
nothing changed.
Logs for Kernel 4.7.6, which is working fine, attached.
I am seeing the same problem with kernel version 4.9.5 when
running under KVM virtualization. I can run the same kernel from
within the VM and from without the VM. This feature (using the
card with VFIO) used to work.
Is it still an issue?