Bug 102402

Summary: [BAT][HSW] igt@kms_cursor_legacy@basic-flip-(before|after)-cursor-varying-size - Failed assertion: get_vblank(display->drm_fd, pipe, 0) == vblank_start
Product: DRI Reporter: Martin Peres <martin.peres>
Component: DRM/IntelAssignee: Maarten Lankhorst <bugs>
Status: CLOSED FIXED QA Contact: Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity: critical    
Priority: highest CC: intel-gfx-bugs
Version: XOrg git   
Hardware: Other   
OS: All   
Whiteboard: ReadyForDev
i915 platform: HSW i915 features: display/Other
Attachments:
Description Flags
dmesg snippet pre-merge
none
dmesg snippet CI none

Description Martin Peres 2017-08-25 10:04:07 UTC
On the fi-hsw-4770 machine, we have been getting the following error message quite often in pre-merge:

(kms_cursor_legacy:4154) CRITICAL: Test assertion failure function basic_flip_cursor, file kms_cursor_legacy.c:612:
(kms_cursor_legacy:4154) CRITICAL: Failed assertion: get_vblank(display->drm_fd, pipe, 0) == vblank_start
(kms_cursor_legacy:4154) CRITICAL: error: 16071 != 16070

Full logs:
 - https://intel-gfx-ci.01.org/tree/drm-tip/Trybot_1085/fi-hsw-4770/igt@kms_cursor_legacy@basic-flip-before-cursor-varying-size.html
 - https://intel-gfx-ci.01.org/tree/drm-tip/Trybot_1085/fi-hsw-4770/igt@kms_cursor_legacy@basic-flip-after-cursor-varying-size.html
Comment 1 Marta Löfstedt 2017-08-28 13:18:15 UTC
Created attachment 133838 [details]
dmesg snippet pre-merge

I removed timing so it is easier to diff
Comment 2 Marta Löfstedt 2017-08-28 13:19:24 UTC
Created attachment 133839 [details]
dmesg snippet CI

I removed timing info to make it easier to diff with the pre-merge log
Comment 3 Marta Löfstedt 2017-08-28 13:20:05 UTC
If I diff the logs it appear as if there are a lot more prints on the pre-merge.
Comment 5 Maarten Lankhorst 2017-10-04 11:32:42 UTC
Considering this was always failing before, maybe this fixed it?

commit 1043c09ccbcba8e5c2ec5f2a358a442346348bd8
Author: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
Date:   Tue Sep 19 13:31:13 2017 +0200

    tests/kms_cursor_legacy: Do not start collecting CRC after making FB busy
    
    Collecting CRC may force a modeset, which is a bad idea after we just
    forced a hang. The hang is intended to make sure the page flip doesn't
    complete before the cursor, making sure that works.
Comment 6 Martin Peres 2017-10-04 12:45:58 UTC
(In reply to Maarten Lankhorst from comment #5)
> Considering this was always failing before, maybe this fixed it?
> 
> commit 1043c09ccbcba8e5c2ec5f2a358a442346348bd8
> Author: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>
> Date:   Tue Sep 19 13:31:13 2017 +0200
> 
>     tests/kms_cursor_legacy: Do not start collecting CRC after making FB busy
>     
>     Collecting CRC may force a modeset, which is a bad idea after we just
>     forced a hang. The hang is intended to make sure the page flip doesn't
>     complete before the cursor, making sure that works.

Looks like it, indeed! I'll close the bug now and archive the issue.

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.