Ubuntu 20.04 AMD GPU with Intel process stuck on 1024x768(4:3) Resolution

10,052

My situation is a little different since I'm using a headless server for crypto mining, but I was getting the same error message regarding amdgpu-dkms. I did manage to get 20.45 driver working on Ubuntu Server 20.04 using the following steps:

./amdgpu-pro-install --opencl=legacy,pal,rocr --headless -y

omit --headless if using desktop gui. errors will still be present

get mainline kernel update script

wget https://raw.githubusercontent.com/pimlie/ubuntu-mainline-kernel.sh/master/ubuntu-mainline-kernel.sh

install kernel update script

sudo install ubuntu-mainline-kernel.sh /usr/local/bin/

run kernel update script

sudo ubuntu-mainline-kernel.sh -i

script will display most current kernel available and ask for confirmation to install. In my case it was 5.10.1-051001-generic

after reboot, mining software still didn't work. amdgpu didn't show up in lsmod so I did:

sudo modprobe amdgpu

and suddenly my mining software worked. haven't tried rebooting because I'm just so relieved that it started working.

I also installed kernel 5.6.19 at one point and ran update-initramfs -u

I don't think that had any effect but just mentioning it for completeness.

Share:
10,052
ALTAMASH
Author by

ALTAMASH

Updated on September 18, 2022

Comments

  • ALTAMASH
    ALTAMASH over 1 year

    I've installed Ubuntu with single boot option. I wasn't able to install Ubuntu straight away, I've had to use safe graphics mode option to install Ubuntu 20.04.After successful installation, I was presented with a black screen. Then I started Ubuntu again in recovery mode and tried to install AMD latest Ubuntu 20.04 drivers and I got the below error at around 97%. Command used to run AMD driver is below:

    ./amdgpu-install -y

    Error Message I got is this

    Building for 5.4.0-47-generic
    Building for architecture x86_64
    Building initial module for 5.4.0-47-generic
    Error! Bad return status for module build on kernel: 5.4.0-47-generic (x86_64)
    Consult /var/lib/dkms/amdgpu/5.6.0.15-1098277/build/make.log for more information.
    dpkg: error processing package amdgpu-dkms (--configure):
     installed amdgpu-dkms package post-installation script subprocess returned error exit status 10
    dpkg: dependency problems prevent configuration of amdgpu:
     amdgpu depends on amdgpu-dkms (= 1:5.6.0.15-1098277); however:
      Package amdgpu-dkms is not configured yet.
    
    dpkg: error processing package amdgpu (--configure):
     dependency problems - leaving unconfigured
    Setting up libegl1-amdgpu-mesa-drivers:i386 (1:20.0.5-1098277) ...
    Setting up g++-9 (9.3.0-10ubuntu2) ...
    No apport report written because the error message indicates its a followup error from a previous failur
    e.
      Setting up g++ (4:9.3.0-1ubuntu2) ...
    update-alternatives: using /usr/bin/g++ to provide /usr/bin/c++ (c++) in auto mode
    Setting up build-essential (12.8ubuntu1) ...
    Setting up amdgpu-lib32 (20.20-1098277) ...
    Processing triggers for man-db (2.9.1-1) ...
    Processing triggers for libc-bin (2.31-0ubuntu9) ...
    Errors were encountered while processing:
     amdgpu-dkms
     amdgpu
    E: Sub-process /usr/bin/dpkg returned an error code (1)
    

    I've tried the above method twice and have uninstalled the drivers. The below result is after uninstalling amdgpu driver from AMD site.

    The interesting point is the result from the command:

    sudo lshw -c video
    *-display UNCLAIMED       
           description: VGA compatible controller
           product: Tonga XT / Amethyst XT [Radeon R9 380X / R9 M295X]
           vendor: Advanced Micro Devices, Inc. [AMD/ATI]
           physical id: 0
           bus info: pci@0000:01:00.0
           version: f1
           width: 64 bits
           clock: 33MHz
           capabilities: pm pciexpress msi vga_controller bus_master cap_list
           configuration: latency=0
           resources: memory:c0000000-cfffffff memory:d0000000-d01fffff ioport:e000(size=256) memory:dfe00000-dfe3ffff memory:c0000-dffff
    

    Thanks in advance!

    • ALTAMASH
      ALTAMASH over 3 years
      I've tried the latest version till date and that is 17-Sep-2020 which is present on AMD site and the AMD driver version is 20.30 and problem still exists. The good news is it didn't gave any error and installed it. One positive sign there, it may have helped someone with this update, but not me. Thanks!
    • r0ng
      r0ng over 3 years
      I just tried to install 20.45, for 5700XT on Ubuntu 20.04, had the exact same problem
  • matinict
    matinict over 3 years
    it may be crush os