Bug 111123

Summary: Display issues AMD RAVEN Ryzen 5 3400G APU
Product: DRI Reporter: Redsandro <freedesktop>
Component: DRM/AMDgpuAssignee: Default DRI bug account <dri-devel>
Status: RESOLVED MOVED QA Contact:
Severity: normal    
Priority: medium CC: nicholas.kazlauskas
Version: unspecified   
Hardware: x86-64 (AMD64)   
OS: Linux (All)   
i915 platform: i915 features:
Description Flags
dmesg none

Description Redsandro 2019-07-13 20:35:44 UTC
I'm having issues using hardware accelerated graphics on my AMD Ryzen 5 3400G with A300M-STX mainboard and A300 chipset by ASRock.

That's the AMD Ryzen 5 3400G with Radeon Vega Graphics (family: 0x17, model: 0x18, stepping: 0x1)

When attaching an UHD (4k) monitor over the DisplayPort, everything works as expected as long as I am running an older kernel that doesn't support the APU video hardware. E.g. 4.15 or 4.18.

When I use 5.0, 5.1 or 5.2 kernel, the _same display connection that worked fine before_ doesn't work anymore. The image keeps disappearing, shaking, distorting and creating green pixel noise, as if the cable is bad quality.

When starting those same kernels with `nomodeset` from GRUB, the image is stable again, but the hardware acceleration is disabled.

HDMI works fine on a HD display (1920x1080) but also no picture on UHD.

HWE Kernel does have hardware rendering, but shows the following:

[    1.296083] kfd kfd: DID 15d8 is missing in supported_devices
[    1.296084] kfd kfd: kgd2kfd_probe failed

So I used Ukuu to install 5.2.0-RC7.

$ uname -a
Linux abu 5.2.0-050200rc7-generic #201906300430 SMP Sun Jun 30 04:32:31 UTC 2019 x86_64 x86_64 x86_64 GNU/Linux

$ lspci | grep -i vga
04:00.0 VGA compatible controller: Advanced Micro Devices, Inc. [AMD/ATI] Picasso (rev c8)

$ glxinfo  | grep Extended -A11
Extended renderer info (GLX_MESA_query_renderer):
    Vendor: X.Org (0x1002)
    Device: AMD RAVEN (DRM 3.32.0, 5.2.0-050200rc7-generic, LLVM 8.0.1) (0x15d8)
    Version: 19.2.0
    Accelerated: yes
    Video memory: 2048MB
    Unified memory: no
    Preferred profile: core (0x1)
    Max core profile version: 4.5
    Max compat profile version: 4.5
    Max GLES1 profile version: 1.1
    Max GLES[23] profile version: 3.2

$ dmesg | grep -i amdgpu | grep -v ring
[    1.202002] [drm] amdgpu kernel modesetting enabled.
[    1.202142] amdgpu 0000:04:00.0: remove_conflicting_pci_framebuffers: bar 0: 0xe0000000 -> 0xefffffff
[    1.202143] amdgpu 0000:04:00.0: remove_conflicting_pci_framebuffers: bar 2: 0xf0000000 -> 0xf01fffff
[    1.202144] amdgpu 0000:04:00.0: remove_conflicting_pci_framebuffers: bar 5: 0xfcc00000 -> 0xfcc7ffff
[    1.202146] fb0: switching to amdgpudrmfb from EFI VGA
[    1.202220] amdgpu 0000:04:00.0: vgaarb: deactivate vga console
[    1.202253] amdgpu 0000:04:00.0: enabling device (0006 -> 0007)
[    1.202474] amdgpu 0000:04:00.0: VRAM: 2048M 0x000000F400000000 - 0x000000F47FFFFFFF (2048M used)
[    1.202475] amdgpu 0000:04:00.0: GART: 1024M 0x0000000000000000 - 0x000000003FFFFFFF
[    1.202476] amdgpu 0000:04:00.0: AGP: 267419648M 0x000000F800000000 - 0x0000FFFFFFFFFFFF
[    1.202577] [drm] amdgpu: 2048M of VRAM memory ready
[    1.202578] [drm] amdgpu: 3072M of GTT memory ready.
[    1.509190] fbcon: amdgpudrmfb (fb0) is primary device
[    1.559664] amdgpu 0000:04:00.0: fb0: amdgpudrmfb frame buffer device
[    1.584215] [drm] Initialized amdgpu 3.32.0 20150101 for 0000:04:00.0 on minor 0

Tested on Linux Mint 19.1.
Comment 1 Michel Dänzer 2019-07-15 10:47:26 UTC
Please attach the full output of dmesg corresponding to the problem.
Comment 2 Redsandro 2019-07-15 15:01:56 UTC
Created attachment 144790 [details]
Comment 3 Brian Schott 2019-07-21 02:29:04 UTC
I have this problem as well with my 3400G. It's very similar to the problem that my 2700u laptop has with 5.2: https://bugs.freedesktop.org/show_bug.cgi?id=111122
Comment 4 Leonardo da Silva 2019-11-13 16:26:33 UTC
i have the same problem, I bought 2x 3400g with A320-S2H, one machine running ubuntu 19.10 and the other Arch Linux, both with the same problem described above when using integrated graphics.
Comment 5 Martin Peres 2019-11-19 09:32:33 UTC
-- GitLab Migration Automatic Message --

This bug has been migrated to freedesktop.org's GitLab instance and has been closed from further activity.

You can subscribe and participate further through the new bug through this link to our GitLab instance: https://gitlab.freedesktop.org/drm/amd/issues/845.

Use of freedesktop.org services, including Bugzilla, is subject to our Code of Conduct. How we collect and use information is described in our Privacy Policy.