| Summary: | [CI][DRMTIP] igt@kms_flip_tiling@flip-to-y-tiled / igt@kms_rotation_crc@* - fail - Exceeded timeout (\d+ms) while waiting for a pageflip | ||
|---|---|---|---|
| Product: | DRI | Reporter: | Lakshmi <lakshminarayana.vudum> |
| Component: | DRM/Intel | Assignee: | Juha-Pekka Heikkilä <juhapekka.heikkila> |
| Status: | CLOSED FIXED | QA Contact: | Intel GFX Bugs mailing list <intel-gfx-bugs> |
| Severity: | major | ||
| Priority: | medium | CC: | arkadiusz.hiler, intel-gfx-bugs, stanislav.lisovskiy |
| Version: | XOrg git | ||
| Hardware: | Other | ||
| OS: | All | ||
| Whiteboard: | ReadyForDev | ||
| i915 platform: | CNL, ICL | i915 features: | |
|
Description
Lakshmi
2018-11-26 15:39:43 UTC
On ICL https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_176/fi-icl-u3/igt@kms_rotation_crc@primary-y-tiled-reflect-x-270.html Starting subtest: primary-y-tiled-reflect-x-270 (kms_rotation_crc:1201) igt_kms-CRITICAL: Test assertion failure function kmstest_wait_for_pageflip, file ../lib/igt_kms.c:1501: (kms_rotation_crc:1201) igt_kms-CRITICAL: Failed assertion: ret == 1 (kms_rotation_crc:1201) igt_kms-CRITICAL: Last errno: 25, Inappropriate ioctl for device (kms_rotation_crc:1201) igt_kms-CRITICAL: error: 0 != 1 Subtest primary-y-tiled-reflect-x-270 failed. This has been seen only 2 times and last time on drmtip_176. A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-y-tiled-reflect-x-270 - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25, -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-y-tiled-reflect-x-270|90 - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25, +}
New failures caught by the filter:
* https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_183/fi-icl-u3/igt@kms_rotation_crc@primary-y-tiled-reflect-x-90.html
A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-y-tiled-reflect-x-270|90 - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25, -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-y-tiled-reflect-x-270|90|180 - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 +}
New failures caught by the filter:
* https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_190/fi-icl-u3/igt@kms_rotation_crc@primary-y-tiled-reflect-x-180.html
A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-y-tiled-reflect-x-270|90|180 - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 +}
New failures caught by the filter:
* https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_232/fi-icl-y/igt@kms_rotation_crc@primary-x-tiled-reflect-x-180.html
A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@primary-* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 +}
New failures caught by the filter:
* https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_232/fi-icl-y/igt@kms_rotation_crc@sprite-rotation-90.html
The issue is strange. Noways it is seen only on ICL-y. ICL-u seems fine. The failure is us waiting for pageflip to happen (select()) with a timeout of 50ms, but we exceed that timeout. The errno we see in the error is random noise and not relevant to the issue. I'll work on improved logging here. There were issues with IRQ storms, which seems to be solved now by a BIOS update. Let's see if we still can reproduce this with drmtip_248. If the issue is sill here then we may try to wait longer to see if a pageflip is comming at all, as it may take some time to reconfigure the HW. But 50ms should be plenty anyway... Customer impact assessment will be added after looking at drmtip_248 results. This commit improves logging for this issue: https://patchwork.freedesktop.org/patch/296260/ A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 / Exceeded timeout (\d+ms) while waiting for a pageflip +}
No new failures caught with the new filter
JP did you saw this issue still? Also same question to Stan? (In reply to Jani Saarinen from comment #13) > Also same question to Stan? Yes, timeout while waiting for page flip is still visible. Even though it doesn't hang anymore. A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled|igt@kms_rotation_crc@* - fail - Failed assertion: ret == 1\n[^\n]+Last errno: 25 / Exceeded timeout (\d+ms) while waiting for a pageflip -}
{+ CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled / igt@kms_rotation_crc@* - fail - Exceeded timeout (\d+ms) while waiting for a pageflip +}
No new failures caught with the new filter
on u2/u3 and shards BIOS updates made, this only seen on ICL-Y On ICL-y there were i2c changes made and why issue might not been seen also on ICL-Y anymore. Now this was seen on every build before and after change nothing, I would consider this fixed by making i2c IRQ change. Lowering priority Was happening on every builds and stopped on seeing after drmtip_252 Resolving. A CI Bug Log filter associated to this bug has been updated:
{- CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled / igt@kms_rotation_crc@* - fail - Exceeded timeout (\d+ms) while waiting for a pageflip -}
{+ SKL CNL ICL: igt@kms_flip_tiling@flip-to-y-tiled / igt@kms_rotation_crc@* - fail - Exceeded timeout (\d+ms) while waiting for a pageflip +}
New failures caught by the filter:
* https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_6546/shard-skl10/igt@kms_rotation_crc@sprite-rotation-90.html
Last seen drmtip_252 (7 months, 1 week old) not seen in the last 97 runs, so closing and archiving this The CI Bug Log issue associated to this bug has been archived. New failures matching the above filters will not be associated to this bug anymore. |
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.