System freeze when navigating in viewport shading when using assets with Cycles/GPU enabled on AMD GPU #98907

New Issue

jean-martin · 2022-06-15T15:21:18+02:00

jean-martin commented

2022-06-15 15:21:18 +02:00

System Information
Operating system: Linux-5.13.0-48-lowlatency-x86_64-with-glibc2.31 64 Bits
Graphics card: AMD Radeon RX 6800 XT (sienna_cichlid, LLVM 13.0.1, DRM 3.41, 5.13.0-48-lowlatency) AMD 4.6 (Core Profile) Mesa 22.0.0-devel

Blender Version
Broken: version: 3.2.0, branch: master, commit date: 2022-06-08 10:22, hash: e05e1e3691
Worked: 3.1.2 (no GPU acceleration of course), no issue on 3.2.0 mac and windows

Short description of error
When importing an asset in a scene, with cycles / GPU enabled the system crashes after navigating a few times with MMB on viewport shading with GPU enabled (no issue on CPU)

Exact steps for others to reproduce the error
[Please describe the exact steps needed to reproduce the issue]
Use the asset.blend file provided as an asset file, import the collection "Pixie_red", setup cycles with GPU enabled and try to navigate on the viewport shading with MMB. Then after a few tries (4-5 I think) the system freezes and display the ugly pixels as shown on the "crash.png" provided file. You can also use the provided blend file named "1.blend". With the same file on mac or windows plateform I have no issue. Everything is fine on Linux with CPU enabled instead of GPU.
As the system crashes I have no crash report.

1.blend
assets.blend

**System Information** Operating system: Linux-5.13.0-48-lowlatency-x86_64-with-glibc2.31 64 Bits Graphics card: AMD Radeon RX 6800 XT (sienna_cichlid, LLVM 13.0.1, DRM 3.41, 5.13.0-48-lowlatency) AMD 4.6 (Core Profile) Mesa 22.0.0-devel **Blender Version** Broken: version: 3.2.0, branch: master, commit date: 2022-06-08 10:22, hash: `e05e1e3691` Worked: 3.1.2 (no GPU acceleration of course), no issue on 3.2.0 mac and windows **Short description of error** When importing an asset in a scene, with cycles / GPU enabled the system crashes after navigating a few times with MMB on viewport shading with GPU enabled (no issue on CPU) **Exact steps for others to reproduce the error** [Please describe the exact steps needed to reproduce the issue] Use the asset.blend file provided as an asset file, import the collection "Pixie_red", setup cycles with GPU enabled and try to navigate on the viewport shading with MMB. Then after a few tries (4-5 I think) the system freezes and display the ugly pixels as shown on the "crash.png" provided file. You can also use the provided blend file named "1.blend". With the same file on mac or windows plateform I have no issue. Everything is fine on Linux with CPU enabled instead of GPU. As the system crashes I have no crash report. ![crash.png](https://archive.blender.org/developer/F13168794/crash.png) [1.blend](https://archive.blender.org/developer/F13168802/1.blend) [assets.blend](https://archive.blender.org/developer/F13168810/assets.blend)

jean-martin commented

2022-06-15 15:21:18 +02:00

Added subscriber: @jean-martin-barbut

jean-martin commented

2022-06-18 16:11:52 +02:00

Nobody has this issue ?

I tried the same process with another file (classroom from blender demo files), same problem. If I use blendluxcore render engine or radeon pro render render engine, everything goes fine.

Nobody has this issue ? I tried the same process with another file (classroom from blender demo files), same problem. If I use blendluxcore render engine or radeon pro render render engine, everything goes fine.

Omar Emara commented

2022-06-20 10:34:42 +02:00

Added subscriber: @OmarEmaraDev

Omar Emara commented

2022-06-20 10:34:42 +02:00

Changed status from 'Needs Triage' to: 'Needs User Info'

Omar Emara commented

2022-06-20 10:34:42 +02:00

Can you attach the crash log created by Blender when it crashes? https://docs.blender.org/manual/en/latest/troubleshooting/crash.html
Can you also attach your system info? Help > Save System Info.

Can you attach the crash log created by Blender when it crashes? https://docs.blender.org/manual/en/latest/troubleshooting/crash.html Can you also attach your system info? Help > Save System Info.

jean-martin commented

2022-06-22 11:56:01 +02:00

Hello, I tried again this morning and made an append on all the objects of the BMW demo file, craash again. But the whole system freezes, no blender.crash.txt created in /tmp. I join you my system info. I join also another file made by "append" on the objects of the bmw demo file.

system-info.txt

bmw.blend

Hello, I tried again this morning and made an append on all the objects of the BMW demo file, craash again. But the whole system freezes, no blender.crash.txt created in /tmp. I join you my system info. I join also another file made by "append" on the objects of the bmw demo file. [system-info.txt](https://archive.blender.org/developer/F13216344/system-info.txt) [bmw.blend](https://archive.blender.org/developer/F13216345/bmw.blend)

Omar Emara commented

2022-06-22 12:07:03 +02:00

Oh, I thought only Blender crashed. So do you reboot your system after the crash?
If you reboot your system, can you attach the output of journalctl -b -1 just after the reboot?
If you don't reboot your system, can you attach the output of sudo dmesg after the system crashes?

Oh, I thought only Blender crashed. So do you reboot your system after the crash? If you reboot your system, can you attach the output of `journalctl -b -1` just after the reboot? If you don't reboot your system, can you attach the output of `sudo dmesg` after the system crashes?

jean-martin commented

2022-06-22 12:29:20 +02:00

Yes I need to hard reset. For the crash related to the output you asked me for, here is what I did : on a new blend file with cycle GPU compute activated (32 max samples in viewport rendering, 250 for rendering, max bounce 6/3/3/4/0/3) , I append all the objects of the bmw demo file (on blender website) and navigate with MMB on viewport render on cycles. I join you the

journalctl -b -1 output

And the bmw blend file I append the objects from.

There is no crash if I do a render (f12)

journalctl-b-1.txt

bmw27_gpu.blend

Yes I need to hard reset. For the crash related to the output you asked me for, here is what I did : on a new blend file with cycle GPU compute activated (32 max samples in viewport rendering, 250 for rendering, max bounce 6/3/3/4/0/3) , I append all the objects of the bmw demo file (on blender website) and navigate with MMB on viewport render on cycles. I join you the ``` journalctl -b -1 output ``` And the bmw blend file I append the objects from. There is no crash if I do a render (f12) [journalctl-b-1.txt](https://archive.blender.org/developer/F13216375/journalctl-b-1.txt) [bmw27_gpu.blend](https://archive.blender.org/developer/F13216383/bmw27_gpu.blend)

Omar Emara commented

2022-06-22 12:56:08 +02:00

Looks like we are looking at a GPU reset in the kernel. So is not a bug in Blender itself. Can you try to update your kernel version and see if this fixes the issue?
Which version of the user space driver/ROCm are you using?

juin 22 12:12:51 DesktopKubuntu20 kernel: [drm:amdgpu_dm_commit_planes [amdgpu]] *ERROR* Waiting for fences timed out!
juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_dm_commit_planes [amdgpu]] *ERROR* Waiting for fences timed out!
juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=64122, emitted seq=64124
juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process blender pid 5479 thread blender:cs0 pid 5496
juin 22 12:12:56 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset begin!
juin 22 12:12:56 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110)
juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KGQ disable failed
juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110)
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KCQ disable failed
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm:gfx_v10_0_cp_gfx_enable [amdgpu]] *ERROR* failed to halt cp gfx
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] free PSP TMR buffer
juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: MODE1 reset
juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU mode1 reset
juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU smu mode1 reset
juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset succeeded, trying to resume
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] PCIE GART of 512M enabled (table at 0x00000080008CA000).
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] VRAM is lost due to GPU reset!
juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] PSP is resuming...
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu_cs_ioctl: 17 callbacks suppressed
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] psp command (0x22) failed and response status is (0x80000310)
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] reserve 0xa00000 from 0x83fe000000 for PSP TMR
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resuming...
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resumed successfully!
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] DMUB hardware initialized: version=0x02010003
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] kiq ring mec 2 pipe 1 q 0
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] VCN decode and encode initialized successfully(under DPG Mode).
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] JPEG decode initialized successfully.
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring kiq_2.1.0 uses VM inv eng 11 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma2 uses VM inv eng 14 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma3 uses VM inv eng 15 on hub 0
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_dec_1 uses VM inv eng 5 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_1.0 uses VM inv eng 6 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_1.1 uses VM inv eng 7 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 8 on hub 1
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: recover vram bo from shadow start
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: recover vram bo from shadow done
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] Skip scheduling IBs!
...
juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset(2) succeeded!
juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
...
juin 22 12:13:03 DesktopKubuntu20 kernel: amdgpu_cs_ioctl: 17 callbacks suppressed
juin 22 12:13:03 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125!
...

Looks like we are looking at a GPU reset in the kernel. So is not a bug in Blender itself. Can you try to update your kernel version and see if this fixes the issue? Which version of the user space driver/ROCm are you using? ``` juin 22 12:12:51 DesktopKubuntu20 kernel: [drm:amdgpu_dm_commit_planes [amdgpu]] *ERROR* Waiting for fences timed out! juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_dm_commit_planes [amdgpu]] *ERROR* Waiting for fences timed out! juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx_0.0.0 timeout, signaled seq=64122, emitted seq=64124 juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process information: process blender pid 5479 thread blender:cs0 pid 5496 juin 22 12:12:56 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset begin! juin 22 12:12:56 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110) juin 22 12:12:56 DesktopKubuntu20 kernel: [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KGQ disable failed juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: [drm:amdgpu_ring_test_helper [amdgpu]] *ERROR* ring kiq_2.1.0 test failed (-110) juin 22 12:12:57 DesktopKubuntu20 kernel: [drm:gfx_v10_0_hw_fini [amdgpu]] *ERROR* KCQ disable failed juin 22 12:12:57 DesktopKubuntu20 kernel: [drm:gfx_v10_0_cp_gfx_enable [amdgpu]] *ERROR* failed to halt cp gfx juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] free PSP TMR buffer juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: MODE1 reset juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU mode1 reset juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU smu mode1 reset juin 22 12:12:57 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset succeeded, trying to resume juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] PCIE GART of 512M enabled (table at 0x00000080008CA000). juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] VRAM is lost due to GPU reset! juin 22 12:12:57 DesktopKubuntu20 kernel: [drm] PSP is resuming... juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu_cs_ioctl: 17 callbacks suppressed juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125! juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125! juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] psp command (0x22) failed and response status is (0x80000310) juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] reserve 0xa00000 from 0x83fe000000 for PSP TMR juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SECUREDISPLAY: securedisplay ta ucode is not available juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resuming... juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: SMU is resumed successfully! juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] DMUB hardware initialized: version=0x02010003 juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] kiq ring mec 2 pipe 1 q 0 juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] VCN decode and encode initialized successfully(under DPG Mode). juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] JPEG decode initialized successfully. juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring gfx_0.0.0 uses VM inv eng 0 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.0 uses VM inv eng 1 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.0 uses VM inv eng 4 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.0 uses VM inv eng 5 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.0 uses VM inv eng 6 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.0.1 uses VM inv eng 7 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.1.1 uses VM inv eng 8 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.2.1 uses VM inv eng 9 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring comp_1.3.1 uses VM inv eng 10 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring kiq_2.1.0 uses VM inv eng 11 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma0 uses VM inv eng 12 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma1 uses VM inv eng 13 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma2 uses VM inv eng 14 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring sdma3 uses VM inv eng 15 on hub 0 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_dec_0 uses VM inv eng 0 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.0 uses VM inv eng 1 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_0.1 uses VM inv eng 4 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_dec_1 uses VM inv eng 5 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_1.0 uses VM inv eng 6 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring vcn_enc_1.1 uses VM inv eng 7 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: ring jpeg_dec uses VM inv eng 8 on hub 1 juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: recover vram bo from shadow start juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: recover vram bo from shadow done juin 22 12:12:58 DesktopKubuntu20 kernel: [drm] Skip scheduling IBs! ... juin 22 12:12:58 DesktopKubuntu20 kernel: amdgpu 0000:03:00.0: amdgpu: GPU reset(2) succeeded! juin 22 12:12:58 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125! ... juin 22 12:13:03 DesktopKubuntu20 kernel: amdgpu_cs_ioctl: 17 callbacks suppressed juin 22 12:13:03 DesktopKubuntu20 kernel: [drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Failed to initialize parser -125! ... ```

jean-martin commented

2022-06-22 14:09:07 +02:00

This comment was removed by @jean-martin-barbut

*This comment was removed by @jean-martin-barbut*

jean-martin commented

2022-06-22 14:18:26 +02:00

I was afraid it was the reason...

My kernel is the latest updated by the system : 5.13.0-51-lowlatency

As i'm using jackaudio a lot, I need lowlatency. I run ubuntustudio on KDE plasma (kubuntu).

I installed the rocm stack with this command :

sudo amdgpu-install --usecase=rocm --no-dkms

no-dkms because in the past I noticed (with my former RX580 GPUs ) that dkms was an issue with the kernels. Ans I have the latests kisak mesa drivers (updated this morning).

I'm using those radeon repos :

https://repo.radeon.com/rocm/apt/5.1.3 ubuntu main
https://repo.radeon.com/amdgpu/22.10.3/ubuntu focal main

I was afraid it was the reason... My kernel is the latest updated by the system : 5.13.0-51-lowlatency As i'm using jackaudio a lot, I need lowlatency. I run ubuntustudio on KDE plasma (kubuntu). I installed the rocm stack with this command : ``` sudo amdgpu-install --usecase=rocm --no-dkms ``` - no-dkms because in the past I noticed (with my former RX580 GPUs ) that dkms was an issue with the kernels. Ans I have the latests kisak mesa drivers (updated this morning). I'm using those radeon repos : ``` https://repo.radeon.com/rocm/apt/5.1.3 ubuntu main https://repo.radeon.com/amdgpu/22.10.3/ubuntu focal main ```

Omar Emara commented

2022-06-22 14:46:48 +02:00

Changed status from 'Needs User Info' to: 'Archived'

Omar Emara closed this issue

2022-06-22 14:46:48 +02:00

Omar Emara commented

2022-06-22 14:46:48 +02:00

I see, your drivers seems to be up to date. I am not sure there is anything we can do from our end. I would suggest you try to report that to the AMDGPU DRM module in the Linux kernel.
Since this is not a bug in Blender itself, I will archive this for now and reopen later if needed.

I see, your drivers seems to be up to date. I am not sure there is anything we can do from our end. I would suggest you try to report that to the AMDGPU DRM module in the Linux kernel. Since this is not a bug in Blender itself, I will archive this for now and reopen later if needed.

jean-martin commented

2022-06-22 14:57:47 +02:00

By any chance do you know where to report to the AMDGPU DRM module in the Linux kernel ?

Omar Emara commented

2022-06-22 15:11:10 +02:00

Create an issue here https://gitlab.freedesktop.org/drm/amd/-/issues.
Maybe check the README beforehand https://gitlab.freedesktop.org/drm/amd.

Create an issue here https://gitlab.freedesktop.org/drm/amd/-/issues. Maybe check the README beforehand https://gitlab.freedesktop.org/drm/amd.

jean-martin commented

2022-06-25 17:42:39 +02:00

In #98907#1378554, @OmarEmaraDev wrote:
Create an issue here https://gitlab.freedesktop.org/drm/amd/-/issues.
Maybe check the README beforehand https://gitlab.freedesktop.org/drm/amd.

Thanks for the links.

I don't know if it is relevant or not but there is no issue if I use BlendLuxcore or RadeonProRrender render engine. It seems to be related with cycles GPU compute in viewport render.

> In #98907#1378554, @OmarEmaraDev wrote: > Create an issue here https://gitlab.freedesktop.org/drm/amd/-/issues. > Maybe check the README beforehand https://gitlab.freedesktop.org/drm/amd. Thanks for the links. I don't know if it is relevant or not but there is no issue if I use BlendLuxcore or RadeonProRrender render engine. It seems to be related with cycles GPU compute in viewport render.

Omar Emara commented

2022-06-27 08:35:29 +02:00

BlendLuxcore and RadeonProRrender are built using a different technology, so it is expected that any issue that appears when using Blender may not necessarily appear when using either of them.

jean-martin commented

2022-07-21 17:47:30 +02:00

This comment was removed by @jean-martin-barbut

*This comment was removed by @jean-martin-barbut*

Sign in to join this conversation.

No Label

Download

What's New

Blender Studio

Manual

Developers Blog

Documentation

Benchmark

Blender Conference

Development Fund

One-time Donations

System freeze when navigating in viewport shading when using assets with Cycles/GPU enabled on AMD GPU #98907