Bug 1634389 - kernel 4.17, 4.18 does not recognize hardware AMDGPU
Summary: kernel 4.17, 4.18 does not recognize hardware AMDGPU
Keywords:
Status: CLOSED CURRENTRELEASE
Alias: None
Product: Fedora
Classification: Fedora
Component: kernel
Version: 28
Hardware: x86_64
OS: Linux
unspecified
high
Target Milestone: ---
Assignee: Kernel Maintainer List
QA Contact: Fedora Extras Quality Assurance
URL:
Whiteboard:
Depends On:
Blocks:
TreeView+ depends on / blocked
 
Reported: 2018-09-30 13:02 UTC by Dennis Wagelaar
Modified: 2019-01-08 20:24 UTC (History)
17 users (show)

Fixed In Version:
Clone Of:
Environment:
Last Closed: 2019-01-08 20:24:20 UTC
Type: Bug
Embargoed:


Attachments (Terms of Use)
journalctl -xb output for kernel-4.16.16-200.fc27.x86_64 (867.12 KB, text/plain)
2018-09-30 13:02 UTC, Dennis Wagelaar
no flags Details
lspci -nn output (5.03 KB, text/plain)
2018-09-30 13:03 UTC, Dennis Wagelaar
no flags Details
journalctl -xb output for kernel-4.18.9-100.fc27.x86_64 (920.79 KB, text/plain)
2018-09-30 13:12 UTC, Dennis Wagelaar
no flags Details
dmesg output for kernel-4.18.9-100.fc27.x86_64 (77.03 KB, text/plain)
2018-09-30 13:13 UTC, Dennis Wagelaar
no flags Details
dmesg output for kernel-4.16.16-200.fc27.x86_64 (77.80 KB, text/plain)
2018-09-30 13:29 UTC, Dennis Wagelaar
no flags Details
dracut configuration workaround (161 bytes, text/plain)
2018-11-19 15:00 UTC, Dennis Wagelaar
no flags Details

Description Dennis Wagelaar 2018-09-30 13:02:36 UTC
Created attachment 1488571 [details]
journalctl -xb output  for kernel-4.16.16-200.fc27.x86_64

Description of problem:
Boot error messages re. loading AMDGPU kernel modules; no graphical boot screen, no hardware acceleration available.

Version-Release number of selected component (if applicable):
kernel-4.17.11-100.fc27.x86_64
kernel-4.17.12-100.fc27.x86_64
kernel-4.18.9-100.fc27.x86_64

Works on kernel-4.16.16-200.fc27.x86_64

How reproducible:
Boot into broken kernel

Steps to Reproduce:
1. Boot
2.
3.

Actual results:
no graphical boot screen, no hardware acceleration available.

Expected results:
graphical boot screen, 3D hardware acceleration available.

Additional info:
See also bug #1599138

Comment 1 Dennis Wagelaar 2018-09-30 13:03:48 UTC
Created attachment 1488572 [details]
lspci -nn output

Comment 2 Dennis Wagelaar 2018-09-30 13:12:25 UTC
Created attachment 1488573 [details]
journalctl -xb output  for kernel-4.18.9-100.fc27.x86_64

Comment 3 Dennis Wagelaar 2018-09-30 13:13:07 UTC
Created attachment 1488574 [details]
dmesg output for kernel-4.18.9-100.fc27.x86_64

Comment 4 Dennis Wagelaar 2018-09-30 13:29:56 UTC
Created attachment 1488575 [details]
dmesg output for kernel-4.16.16-200.fc27.x86_64

Comment 5 Dennis Wagelaar 2018-10-01 09:15:18 UTC
Created upstream bug at https://bugzilla.kernel.org/show_bug.cgi?id=201295

Comment 6 Dennis Wagelaar 2018-10-01 09:36:30 UTC
From upstream:

"Suspicious section from kernel 4.18.16 dmesg output:

[    2.801753] amdgpu 0000:09:00.0: Direct firmware load for amdgpu/polaris11_mc.bin failed with error -2
[    2.801754] mc: Failed to load firmware "amdgpu/polaris11_mc.bin"
[    2.801852] [drm:gmc_v8_0_sw_init [amdgpu]] *ERROR* Failed to load mc firmware!
[    2.801936] [drm:amdgpu_device_init [amdgpu]] *ERROR* sw_init of IP block <gmc_v8_0> failed -2
[    2.801999] amdgpu 0000:09:00.0: amdgpu_device_ip_init failed
[    2.802061] amdgpu 0000:09:00.0: Fatal error during GPU init
[    2.802117] [drm] amdgpu: finishing device.
[    2.802218] amdgpu: probe of 0000:09:00.0 failed with error -2

Error -2 is -ENOENT, "No such file or directory", so the amdgpu/polaris11_mc.bin file is missing where the amdgpu module is loaded from (could be the initrd).

This is a configuration error, not a bug."

Can you check that the kernel config includes all necessary bits in the initrd image?

Comment 8 Dennis Wagelaar 2018-10-08 10:10:24 UTC
Workaround:
Boot into latest kernel, and run:

  dracut --force -v -I "/usr/lib/firmware/amdgpu/*"

Comment 9 Denis 2018-10-08 21:41:28 UTC
Thanks

Workaround works

Comment 10 Dennis Wagelaar 2018-11-19 15:00:23 UTC
Created attachment 1507284 [details]
dracut configuration workaround

The attached '99-amdgpu.conf' file goes in /etc/dracut.conf.d/, and is meant to be a more permanent workaround. It forces every initrd image to always contain the amdgpu and radeon firmware.

Comment 11 Ben Cotton 2018-11-27 13:32:13 UTC
This message is a reminder that Fedora 27 is nearing its end of life.
On 2018-Nov-30  Fedora will stop maintaining and issuing updates for
Fedora 27. It is Fedora's policy to close all bug reports from releases
that are no longer maintained. At that time this bug will be closed as
EOL if it remains open with a Fedora  'version' of '27'.

Package Maintainer: If you wish for this bug to remain open because you
plan to fix it in a currently maintained version, simply change the 'version' 
to a later Fedora version.

Thank you for reporting this issue and we are sorry that we were not 
able to fix it before Fedora 27 is end of life. If you would still like 
to see this bug fixed and are able to reproduce it against a later version 
of Fedora, you are encouraged  change the 'version' to a later Fedora 
version prior this bug is closed as described in the policy above.

Although we aim to fix as many bugs as possible during every release's 
lifetime, sometimes those efforts are overtaken by events. Often a 
more recent Fedora release includes newer upstream software that fixes 
bugs or makes them obsolete.

Comment 12 Ben Cotton 2018-11-30 23:45:58 UTC
Fedora 27 changed to end-of-life (EOL) status on 2018-11-30. Fedora 27 is
no longer maintained, which means that it will not receive any further
security or bug fix updates. As a result we are closing this bug.

If you can reproduce this bug against a currently maintained version of
Fedora please feel free to reopen this bug against that version. If you
are unable to reopen this bug, please file a new report against the
current release. If you experience problems, please add a comment to this
bug.

Thank you for reporting this bug and we are sorry it could not be fixed.

Comment 13 Dennis Wagelaar 2018-12-23 21:13:01 UTC
Since kernel-4.19.10-200.fc28.x86_64 this changed to:

[    3.098710] amdgpu 0000:09:00.0: No more image in the PCI ROM
[    3.098729] ATOM BIOS: 113-C98121-H01
[    3.098759] [drm] vm size is 64 GB, 2 levels, block size is 10-bit, fragment size is 9-bit
[    3.098777] amdgpu 0000:09:00.0: Direct firmware load for amdgpu/polaris11_k_mc.bin failed with error -2
[    3.098779] mc: Failed to load firmware "amdgpu/polaris11_k_mc.bin"
[    3.098848] [drm:gmc_v8_0_sw_init [amdgpu]] *ERROR* Failed to load mc firmware!
[    3.098908] [drm:amdgpu_device_init.cold.28 [amdgpu]] *ERROR* sw_init of IP block <gmc_v8_0> failed -2
[    3.098910] amdgpu 0000:09:00.0: amdgpu_device_ip_init failed
[    3.098911] amdgpu 0000:09:00.0: Fatal error during GPU init
[    3.098911] [drm] amdgpu: finishing device.
[    3.099050] amdgpu: probe of 0000:09:00.0 failed with error -2

in spite of the attached '99-amdgpu.conf' file.

Comment 14 Dennis Wagelaar 2018-12-23 21:19:52 UTC
This is https://bugzilla.kernel.org/show_bug.cgi?id=201985

New firmware should arrive from upstream

Comment 15 Dennis Wagelaar 2019-01-08 20:07:28 UTC
(In reply to Dennis Wagelaar from comment #14)
> This is https://bugzilla.kernel.org/show_bug.cgi?id=201985
> 
> New firmware should arrive from upstream

Firmware arrived in linux-firmware-20181219-89.git0f22c852.fc28.noarch

Also, dracut.conf.d files have started appearing again with each kernel update (e.g. amdgpu-4.19.13-200.fc28.x86_64.conf).

I will test if things work again without the '99-amdgpu.conf' file workaround...

Comment 16 Dennis Wagelaar 2019-01-08 20:24:20 UTC
Works on current kernel without workarounds (kernel-4.19.13-200.fc28.x86_64)


Note You need to log in before you can comment on or make changes to this bug.