[CI] igt@kms_cursor_crc@*| igt@kms_draw_crc@* - fail - CRC mismatch

Submitted by Marta Löfstedt @marta

Assigned to Mika Kahola

Link to original bug (#103232)

Description

CI_DRM_3220 KBL-shards fail igt@kms_cursor_crc@cursor-64x64-suspend

(kms_cursor_crc:2465) igt-debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file igt_debugfs.c:346:
(kms_cursor_crc:2465) igt-debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-64x64-suspend failed.

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3220/shard-kbl4/igt@kms_cursor_crc@cursor-64x64-suspend.html

Blocking

• Bug 105978

added CI feature: display/Other platform: BDW platform: BSW/CHT platform: BXT platform: BYT platform: CFL platform: CML platform: CNL platform: G33 platform: G45 platform: GLK platform: GM45 platform: HSW platform: I830 platform: I845G platform: I85X platform: I865G platform: I915G platform: I915GM platform: I945G platform: I945GM platform: I965G platform: I965GM platform: ICL platform: ILK platform: IVB platform: KBL platform: PNV platform: SKL platform: SNB priority::medium severity::normal + 1 deleted label

Marta Löfstedt @marta said:

CI_DRM_3287 shard-kbl1 igt@kms_cursor_crc@cursor-64x64-random

(kms_cursor_crc:2955) igt-debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file igt_debugfs.c:346:
(kms_cursor_crc:2955) igt-debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-64x64-random failed.

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3287/shard-kbl1/igt@kms_cursor_crc@cursor-64x64-random.html

Marta Löfstedt @marta said:

CI_DRM_3292 shard-kbl2 igt@kms_cursor_crc@cursor-256x256-random

(kms_cursor_crc:1879) igt-debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file igt_debugfs.c:346:
(kms_cursor_crc:1879) igt-debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-256x256-random failed.

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3292/shard-kbl2/igt@kms_cursor_crc@cursor-256x256-random.html

Marta Löfstedt @marta said:

https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4016/shard-apl3/igt@kms_cursor_crc@cursor-64x64-dpms.html

Marta Löfstedt @marta said:

This issue is no longer reproducible. I will close and archive

Marta Löfstedt @marta said:

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3910/shard-apl8/igt@kms_cursor_crc@cursor-64x64-suspend.html

(kms_cursor_crc:1681) igt_debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file ../lib/igt_debugfs.c:370:
(kms_cursor_crc:1681) igt_debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-64x64-suspend failed.

Marta Löfstedt @marta said:

https://intel-gfx-ci.01.org/tree/drm-tip/IGT_4376/shard-apl4/igt@kms_cursor_crc@cursor-256x256-dpms.html


(kms_cursor_crc:1377) igt_debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file ../lib/igt_debugfs.c:370:
(kms_cursor_crc:1377) igt_debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-256x256-dpms failed.

Marta Löfstedt @marta said:

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_4001/shard-apl7/igt@kms_cursor_crc@cursor-128x128-dpms.html

(kms_cursor_crc:1383) igt_debugfs-CRITICAL: Test assertion failure function igt_assert_crc_equal, file ../lib/igt_debugfs.c:370:
(kms_cursor_crc:1383) igt_debugfs-CRITICAL: Failed assertion: !mismatch
Subtest cursor-128x128-dpms failed.

James Ausmus said:

Not seeing this issue anymore on the shards, marking as WORKSFORME

Martin Peres @mupuf said:

(In reply to James Ausmus from comment 8)

Not seeing this issue anymore on the shards, marking as WORKSFORME

You are right, it works pretty well now, since CI_DRM_4453_full (1 week, 3 days ago). However, there is still fi-gdg-551 which fails every single run:

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_89/fi-gdg-551/igt@kms_cursor_crc@cursor-256x256-random.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_89/fi-gdg-551/igt@kms_cursor_crc@cursor-128x128-sliding.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_89/fi-gdg-551/igt@kms_cursor_crc@cursor-256x256-onscreen.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_89/fi-gdg-551/igt@kms_cursor_crc@cursor-256x256-dpms.html

I dropped the priority to reflect that though. The next step would be to check if the CRC support works at all on this platform or if there is just an issue related to the cursor.

LAKSHMINARAYANA VUDUM @l4kshmi said:

Mika, any updates here. Here is the latest results.
https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_5004/shard-skl5/igt@kms_cursor_crc@cursor-size-change.html

LAKSHMINARAYANA VUDUM @l4kshmi said:

This failure happens to occur mostly with Shards and drmtip. Setting the priority to High.

Mika Kahola said:

I sent a patch series that fixes how we read crcs in a test

https://patchwork.freedesktop.org/series/54944/

Clinton Taylor said:

Uma Shankars series https://patchwork.freedesktop.org/series/51408/ appears to resolve many of the crc failures in kms_cursor_crc.

Mika Kahola said:

It seems that Uma's patch series improves the crc mismatch situation. There's still one test on shard runs that fails with !mismatch error

igt@kms_cursor_crc@cursor-64x21-sliding

CI Bug Log said:

The CI Bug Log issue associated to this bug has been updated.

New filters associated

* ICL: igt@runner@aborted - Fail - Previous test: kms_cursor_crc (cursor-256x256-suspend)
- https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_218/fi-icl-u3/igt@runner@aborted.html

Mika Kahola said:

Some findings today on ICL-Y. If I change the test so that it uses overlay plane instead of cursor plane and test the plane only with fully onscreen i.e. 1 and 2 pixels off screen are neglected the test passes with flying colors. I wonder with the question why cursor plane behaves differently?

Mika Kahola said:

The patch

https://patchwork.freedesktop.org/series/57619/

seems to fix crc mismatch errors

Maarten Lankhorst said:

commit 26eeea1506838c77524fa90c8e6f1cc246762a4a (HEAD -> drm-intel-next-queued, drm-intel/drm-intel-next-queued)
Author: Aditya Swarup aditya.swarup@intel.com
Date: Wed Mar 6 18:14:12 2019 -0800

drm/i915/icl: Fix CRC mismatch error for DP link layer compliance

Should hopefully fix this on icl, is this still an issue on other platforms?

Arkadiusz Hiler @ivyl said:

(In reply to Maarten Lankhorst from comment 18)

commit 26eeea1506838c77524fa90c8e6f1cc246762a4a (HEAD ->
drm-intel-next-queued, drm-intel/drm-intel-next-queued)
Author: Aditya Swarup aditya.swarup@intel.com
Date: Wed Mar 6 18:14:12 2019 -0800

drm/i915/icl: Fix CRC mismatch error for DP link layer compliance

Should hopefully fix this on icl, is this still an issue on other platforms?

The bug is still seen on all platforms, including ICL.

E.g.: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_5869/shard-iclb2/igt@kms_cursor_crc@cursor-256x256-onscreen.html