From mboxrd@z Thu Jan 1 00:00:00 1970 From: bugzilla-daemon@freedesktop.org Subject: [Bug 109022] ring gfx timeout during particular shader generation on yuzu emulator Date: Wed, 31 Jul 2019 20:16:19 +0000 Message-ID: References: Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============0863257018==" Return-path: Received: from culpepper.freedesktop.org (culpepper.freedesktop.org [131.252.210.165]) by gabe.freedesktop.org (Postfix) with ESMTP id C6F6C6E2EC for ; Wed, 31 Jul 2019 20:16:18 +0000 (UTC) In-Reply-To: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dri-devel-bounces@lists.freedesktop.org Sender: "dri-devel" To: dri-devel@lists.freedesktop.org List-Id: dri-devel@lists.freedesktop.org --===============0863257018== Content-Type: multipart/alternative; boundary="15646041780.d9B62.3382" Content-Transfer-Encoding: 7bit --15646041780.d9B62.3382 Date: Wed, 31 Jul 2019 20:16:18 +0000 MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated https://bugs.freedesktop.org/show_bug.cgi?id=3D109022 --- Comment #23 from e88z4 --- Hi,=20 Is there an update for this ticket? The issue is still easily be replicated= in the latest mesa on RadeonSI driver. System information Linux 5.2.2 Mesa-master=20 Radeon RX 580 Series (POLARIS10, DRM 3.32.0, 5.2.2-htpc, LLVM 10.0.0) I have tried different combination of llvm backend such as llvm-8, llvm-9, = and llvm-10. They produced the same error. Could this be a bug in amd llvm back= end? Previously, I have provided the api trace, I hope the trace can be used to debug this issue. I provided the latest error from today. They are the same error like before. ^[ [Jul31 16:10] amdgpu 0000:23:00.0: GPU fault detected: 147 0x0d6044= 01 for process yuzu pid 15061 thread yuzu:cs0 pid 15068 [ +0.000004] amdgpu 0000:23:00.0: VM_CONTEXT1_PROTECTION_FAULT_ADDR=20=20 0x0C03F7AC [ +0.000001] amdgpu 0000:23:00.0: VM_CONTEXT1_PROTECTION_FAULT_STATUS 0x0E044001 [ +0.000002] amdgpu 0000:23:00.0: VM fault (0x01, vmid 7, pasid 32770) at = page 201586604, read from 'TC5' (0x54433500) (68) [ +5.312442] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting f= or fences timed out or interrupted! [ +4.874028] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout, signaled seq=3D24383, emitted seq=3D24385 [ +0.000066] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process informatio= n: process yuzu pid 15061 thread yuzu:cs0 pid 15068 [ +0.000006] amdgpu 0000:23:00.0: GPU reset begin! [ +0.584424] cp is busy, skip halt cp [ +0.328786] rlc is busy, skip halt rlc [ +0.001031] amdgpu 0000:23:00.0: GPU pci config reset [ +0.463461] amdgpu 0000:23:00.0: GPU reset succeeded, trying to resume [ +0.002152] [drm] PCIE GART of 256M enabled (table at 0x000000F400900000). [ +0.000016] [drm] VRAM is lost due to GPU reset! [ +0.087583] [drm] UVD and UVD ENC initialized successfully. [ +0.099970] [drm] VCE initialized successfully. [ +0.018504] amdgpu 0000:23:00.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERR= OR* IB test failed on uvd (-110). [ +0.003336] amdgpu 0000:23:00.0: ib ring test failed (-110). --=20 You are receiving this mail because: You are the assignee for the bug.= --15646041780.d9B62.3382 Date: Wed, 31 Jul 2019 20:16:18 +0000 MIME-Version: 1.0 Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable X-Bugzilla-URL: http://bugs.freedesktop.org/ Auto-Submitted: auto-generated

Comme= nt # 23 on bug 10902= 2 from e88z4 
Hi,=20

Is there an update for this ticket? The issue is still easily be replicated=
 in
the latest mesa on RadeonSI driver.

System information
Linux 5.2.2
Mesa-master=20
Radeon RX 580 Series (POLARIS10, DRM 3.32.0, 5.2.2-htpc, LLVM 10.0.0)


I have tried different combination of llvm backend such as llvm-8, llvm-9, =
and
llvm-10. They produced the same error. Could this be a bug in amd llvm back=
end?

Previously, I have provided the api trace, I hope the trace can be used to
debug this issue.


I provided the latest error from today. They are the same error like before.

^[      [Jul31 16:10] amdgpu 0000:23:00.0: GPU fault detected: 147 0x0d6044=
01
for process yuzu pid 15061 thread yuzu:cs0 pid 15068
[  +0.000004] amdgpu 0000:23:00.0:   VM_CONTEXT1_PROTECTION_FAULT_ADDR=20=20
0x0C03F7AC
[  +0.000001] amdgpu 0000:23:00.0:   VM_CONTEXT1_PROTECTION_FAULT_STATUS
0x0E044001
[  +0.000002] amdgpu 0000:23:00.0: VM fault (0x01, vmid 7, pasid 32770) at =
page
201586604, read from 'TC5' (0x54433500) (68)
[  +5.312442] [drm:amdgpu_dm_atomic_commit_tail [amdgpu]] *ERROR* Waiting f=
or
fences timed out or interrupted!
[  +4.874028] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* ring gfx timeout,
signaled seq=3D24383, emitted seq=3D24385
[  +0.000066] [drm:amdgpu_job_timedout [amdgpu]] *ERROR* Process informatio=
n:
process yuzu pid 15061 thread yuzu:cs0 pid 15068
[  +0.000006] amdgpu 0000:23:00.0: GPU reset begin!
[  +0.584424] cp is busy, skip halt cp
[  +0.328786] rlc is busy, skip halt rlc
[  +0.001031] amdgpu 0000:23:00.0: GPU pci config reset
[  +0.463461] amdgpu 0000:23:00.0: GPU reset succeeded, trying to resume
[  +0.002152] [drm] PCIE GART of 256M enabled (table at 0x000000F400900000).
[  +0.000016] [drm] VRAM is lost due to GPU reset!
[  +0.087583] [drm] UVD and UVD ENC initialized successfully.
[  +0.099970] [drm] VCE initialized successfully.
[  +0.018504] amdgpu 0000:23:00.0: [drm:amdgpu_ib_ring_tests [amdgpu]] *ERR=
OR*
IB test failed on uvd (-110).
[  +0.003336] amdgpu 0000:23:00.0: ib ring test failed (-110).

You are receiving this mail because:
  • You are the assignee for the bug.
= --15646041780.d9B62.3382-- --===============0863257018== Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: base64 Content-Disposition: inline X19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX19fX18KZHJpLWRldmVs IG1haWxpbmcgbGlzdApkcmktZGV2ZWxAbGlzdHMuZnJlZWRlc2t0b3Aub3JnCmh0dHBzOi8vbGlz dHMuZnJlZWRlc2t0b3Aub3JnL21haWxtYW4vbGlzdGluZm8vZHJpLWRldmVs --===============0863257018==--