Bug 108718 - Raven Ridge: ring sdma0 timeout on heavy CSS website with Firefox WebRender
Summary: Raven Ridge: ring sdma0 timeout on heavy CSS website with Firefox WebRender
Status: RESOLVED MOVED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/AMDgpu (show other bugs)
Version: XOrg git
Hardware: Other All
: medium normal
Assignee: Default DRI bug account
QA Contact:
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2018-11-12 22:57 UTC by Samantha McVey
Modified: 2019-11-19 09:04 UTC (History)
0 users

See Also:
i915 platform:
i915 features:


Attachments
dmesg log (94.13 KB, text/plain)
2018-11-12 22:57 UTC, Samantha McVey
no flags Details

Description Samantha McVey 2018-11-12 22:57:56 UTC
Created attachment 142451 [details]
dmesg log

This can replicated every time.
System: Lenovo A485 with 2700U APU
Kernel: 4.20.0-rc2

Steps to replicate:
1. Download Firefox Nightly from nightly.mozilla.org
2. Open Firefox Nightly
3. Type about:config in the address bar and press enter
4. Search for gfx.webrender.all in the config search box and double click the "gfx.webrender.all" entry enable it (will show enabled after double clicking)
5. Restart Firefox Nightly
6. Browse to http://diana-adrianne.com/purecss-francine/ in Firefox Nightly
7. Screen should stop responding within a few seconds after zooming/scrolling a bit

Result: Screen will stop responding, though the mouse cursor still moves.

dmesg output:
nov 12 14:42:15 kernel: [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring sdma0 timeout, signaled seq=20166, emitted seq=20168
nov 12 14:42:15 kernel: [drm] GPU recovery disabled.
nov 12 14:42:35 kernel: [TTM] Buffer eviction failed
Comment 1 Haxk20 2019-05-19 18:15:02 UTC
I have the same exact bug. As a workaround i used earlier build of mesa but indeed the latest one is borked and unusable in Firefox.
And without webrenderer firefox is kind of unusable as the color gradient look horrible.
Comment 2 Pierre-Eric Pelloux-Prayer 2019-08-09 14:00:34 UTC
Can you still reproduce this issue?

It seems to work fine here with a recent kernel + mesa configuration.
Comment 3 Matias N. Goldberg 2019-09-04 00:27:30 UTC
I'm on:

OpenGL renderer string: AMD RAVEN (DRM 3.30.0, 5.1.15-050115-generic, LLVM 8.0.1)
OpenGL core profile version string: 4.5 (Core Profile) Mesa 19.2.0-devel (git-8305766 2019-07-11 bionic-oibaf-ppa)

And I can also still repro this issue. Firefox isn't the only way to trigger it, Chromium can trigger it too.

Is it too old already? How recent the kernel and mesa must be?
Comment 4 Pierre-Eric Pelloux-Prayer 2019-09-10 07:55:22 UTC
(In reply to Pierre-Eric Pelloux-Prayer from comment #2)
> Can you still reproduce this issue?
> 
> It seems to work fine here with a recent kernel + mesa configuration.

Correction: I retested and can reproduce the problem partially.

After zooming and scrolling for a while, there are logs like this in dmesg:


[drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* amdgpu_vm_validate_pt_bos() failed.
[drm:amdgpu_cs_ioctl [amdgpu]] *ERROR* Not enough memory for command submission!
[TTM] Failed to find memory space for buffer 0x00000000de5c5282 eviction
[TTM]  No space for 00000000de5c5282 (2224 pages, 8896K, 8M)
[TTM]    placement[0]=0x00060002 (1)
[TTM]      has_type: 1
[TTM]      use_type: 1
[TTM]      flags: 0x0000000A
[TTM]      gpu_offset: 0x00000000
[TTM]      size: 786432
[TTM]      available_caching: 0x00070000
[TTM]      default_caching: 0x00010000
[TTM]  0x0000000000000400-0x0000000000000401: 1: used
[TTM]  0x0000000000000401-0x0000000000000443: 66: used
[TTM]  0x0000000000000443-0x0000000000000445: 2: used
[TTM]  0x0000000000000445-0x0000000000000447: 2: used
[TTM]  0x0000000000000447-0x0000000000000449: 2: used
[TTM]  0x0000000000000449-0x000000000000044b: 2: used
[TTM]  0x000000000000044b-0x000000000000044d: 2: used
[TTM]  0x000000000000044d-0x000000000000044f: 2: used
[TTM]  0x000000000000044f-0x0000000000000451: 2: used
[TTM]  0x0000000000000451-0x0000000000000453: 2: used
[TTM]  0x0000000000000453-0x0000000000000455: 2: used
[TTM]  0x0000000000000455-0x0000000000000456: 1: used
[TTM]  0x0000000000000456-0x0000000000000556: 256: used
[TTM]  0x0000000000000556-0x0000000000000557: 1: used
[TTM]  0x0000000000000557-0x0000000000000558: 1: used
[TTM]  0x0000000000000558-0x0000000000000559: 1: used
[TTM]  0x0000000000000559-0x000000000000055a: 1: used
[TTM]  0x000000000000055a-0x000000000000055b: 1: used
[TTM]  0x000000000000055b-0x000000000000055c: 1: used
[TTM]  0x000000000000055c-0x000000000000055d: 1: used
[TTM]  0x000000000000055d-0x000000000000055e: 1: used
[TTM]  0x000000000000055e-0x0000000000000560: 2: used
[TTM]  0x0000000000000560-0x0000000000000561: 1: used
[TTM]  0x0000000000000561-0x0000000000000562: 1: used
[TTM]  0x0000000000000562-0x0000000000000563: 1: used
[TTM]  0x0000000000000563-0x0000000000000565: 2: used
[TTM]  0x0000000000000565-0x0000000000000665: 256: used
[TTM]  0x0000000000000665-0x000000000000075c: 247: used
[TTM]  0x000000000000075c-0x0000000000000800: 164: free
[TTM]  0x0000000000000800-0x0000000000000900: 256: used
[TTM]  0x0000000000000900-0x0000000000000cca: 970: used
[TTM]  0x0000000000000cca-0x0000000000040000: 258870: free
[TTM]  total: 261120, used 2086 free 259034


The computer doesn't hang or crash though and if I close Firefox's window everything goes back to normal.

(using: Mesa master, LLVM 10, kernel 865b4ca43816 from amd-staging-drm-next)
Comment 5 Pierre-Eric Pelloux-Prayer 2019-09-10 07:56:14 UTC
(In reply to Haxk20 from comment #1)
> I have the same exact bug. As a workaround i used earlier build of mesa but
> indeed the latest one is borked and unusable in Firefox.
> And without webrenderer firefox is kind of unusable as the color gradient
> look horrible.

What is the version of this "earlier build of mesa" that doesn't have this problem?
Comment 6 Martin Peres 2019-11-19 09:04:06 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/607.


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.