Bug 24450 - kernel-desktop-4.14.89 and later hangs with radeon RX550 after "fb: switching to amdgpudrmfb from EFI VGA"
Summary: kernel-desktop-4.14.89 and later hangs with radeon RX550 after "fb: switching...
Status: RESOLVED FIXED
Alias: None
Product: Mageia
Classification: Unclassified
Component: RPM Packages (show other bugs)
Version: 6
Hardware: x86_64 Linux
Priority: Normal normal
Target Milestone: ---
Assignee: QA Team
QA Contact:
URL:
Whiteboard: MGA6-64-OK
Keywords: advisory, validated_update
Depends on:
Blocks:
 
Reported: 2019-02-28 20:39 CET by R Ghetta
Modified: 2019-03-14 22:41 CET (History)
3 users (show)

See Also:
Source RPM: radeon-firmware
CVE:
Status comment:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Description R Ghetta 2019-02-28 20:39:31 CET 
Description of problem:
when booting with kernel-desktop-4.14.89 or 4.14.100 the boot process stops after giving "fb: switching to amdgpudrmfb from EFI VGA"
Googling around I found https://bugzilla.redhat.com/show_bug.cgi?id=1659810 hinting to a obsolete firmware package.

4.14.78 works perfectly
My video card is a Rx550, identified as                 \
description: VGA compatible controller
product: Baffin [Radeon RX 550 640SP / RX 560/560X]
configuration: driver=amdgpu latency=0

booting with nomodeset works, but then x cannot start
Comment 1 R Ghetta 2019-02-28 21:09:02 CET 
btw, the boot process continues even after the "fb: switching ..." message, only the graphics card is affected.
looking at dmesg via ssh I see

[    1.413458] [drm] GPU post is not needed
[    1.413562] [drm] vm size is 64 GB, block size is 13-bit, fragment size is 4-bit
[    1.413574] amdgpu 0000:01:00.0: Direct firmware load for amdgpu/polaris11_k_mc.bin failed with error -2
[    1.413576] mc: Failed to load firmware "amdgpu/polaris11_k_mc.bin"
[    1.413600] [drm:gmc_v8_0_sw_init [amdgpu]] *ERROR* Failed to load mc firmware!
[    1.413616] [drm:amdgpu_device_init [amdgpu]] *ERROR* sw_init of IP block <gmc_v8_0> failed -2
[    1.413618] amdgpu 0000:01:00.0: amdgpu_init failed
[    1.413620] amdgpu 0000:01:00.0: Fatal error during GPU init
[    1.413621] [drm] amdgpu: finishing device.
[    1.413622] [TTM] Memory type 2 has not been initialized
[    1.433237] amdgpu: probe of 0000:01:00.0 failed with error -2
[    1.434587] dracut: Starting plymouth daemon

"mc: Failed to load firmware "amdgpu/polaris11_k_mc.bin" is the same message noted in the redhat bugzilla linked above.
In their instance they pushed a new firmware package containing https://git.kernel.org/pub/scm/linux/kernel/git/firmware/linux-firmware.git/tree/amdgpu/polaris11_k_mc.bin  (see comment 2)
Comment 2 Thomas Backlund 2019-02-28 22:08:17 CET 
Yeah, Amd fixed up their firmwares and spiltted some firmwares depending on hw revision...


Updated firmware pushed to nonfree updates testing
 

Advisory:
Upddated amdgpu firmwares

This radeon-firmware update provides new and updated firmware for Amd
GPUs using the amdgpu kernel driver.


(S)RPM:
radeon-firmware-20190221-1.mga6



For QA / Testers:
To get the new firmwares used:
1. install the package.
2. recreate the initrd with: dracut -f
3. reboot

CC: (none) => tmb
Assignee: bugsquad => qa-bugs

Thomas Backlund 2019-02-28 22:11:49 CET

Source RPM: kernel-4.14.89-1.mga6.src.rpm => radeon-firmware

Comment 3 R Ghetta 2019-03-03 09:25:13 CET 
With the new firmware package both 4.14.89 and 4.14.100 work.
The same with 4.18.20 from backports.
Thanks
Comment 4 Dave Hodgins 2019-03-14 22:00:40 CET 
Oking based on comment 3. Advisory committed to svn. Validating the update.

Whiteboard: (none) => MGA6-64-OK
Keywords: (none) => advisory, validated_update
CC: (none) => davidwhodgins, sysadmin-bugs

Comment 5 Mageia Robot 2019-03-14 22:41:07 CET 
An update for this issue has been pushed to the Mageia Updates repository.

https://advisories.mageia.org/MGAA-2019-0021.html

Resolution: (none) => FIXED
Status: NEW => RESOLVED
Note You need to log in before you can comment on or make changes to this bug.