Bug 105454

https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3903/fi-cfl-s2/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

(kms_cursor_legacy:1664) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file kms_cursor_legacy.c:1013:
(kms_cursor_legacy:1664) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:1664) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:1664) CRITICAL: During modeset at least 1 page flip needs to match!
Subtest 2x-nonblocking-modeset-vs-cursor-atomic failed.

Now we have more data from shardlist on BAT machines:
(kms_cursor_legacy:2899) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file ../tests/kms_cursor_legacy.c:1013:
(kms_cursor_legacy:2899) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:2899) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:2899) CRITICAL: During modeset at least 1 page flip needs to match!

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_1/fi-skl-6770hq/igt@kms_cursor_legacy@2x-long-nonblocking-modeset-vs-cursor-atomic.html
https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_1/fi-skl-6770hq/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html
https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_1/fi-kbl-7500u/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_2/fi-skl-6700k2/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

(kms_cursor_legacy:3829) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file ../tests/kms_cursor_legacy.c:1013:
(kms_cursor_legacy:3829) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:3829) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:3829) CRITICAL: During modeset at least 1 page flip needs to match!
Subtest 2x-nonblocking-modeset-vs-cursor-atomic failed.

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_3/fi-skl-guc/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html	

(kms_cursor_legacy:1983) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file ../tests/kms_cursor_legacy.c:1013:
(kms_cursor_legacy:1983) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:1983) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:1983) CRITICAL: During modeset at least 1 page flip needs to match!
Subtest 2x-nonblocking-modeset-vs-cursor-atomic failed.

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_7/fi-cfl-s3/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

(kms_cursor_legacy:2041) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file ../tests/kms_cursor_legacy.c:1013:
(kms_cursor_legacy:2041) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:2041) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:2041) CRITICAL: During modeset at least 1 page flip needs to match!
Subtest 2x-nonblocking-modeset-vs-cursor-atomic failed.

Also seen on KBL and CNL: https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_89/fi-kbl-7500u/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_88/fi-cfl-8109u/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

https://intel-gfx-ci.01.org/tree/drm-tip/drmtip_127/fi-icl-u2/igt@kms_cursor_legacy@2x-nonblocking-modeset-vs-cursor-atomic.html

Starting subtest: 2x-nonblocking-modeset-vs-cursor-atomic
(kms_cursor_legacy:2133) CRITICAL: Test assertion failure function two_screens_flip_vs_cursor, file ../tests/kms_cursor_legacy.c:1014:
(kms_cursor_legacy:2133) CRITICAL: Failed assertion: vblank_matches
(kms_cursor_legacy:2133) CRITICAL: Last errno: 25, Inappropriate ioctl for device
(kms_cursor_legacy:2133) CRITICAL: During modeset at least 1 page flip needs to match!
Subtest 2x-nonblocking-modeset-vs-cursor-atomic failed.

How frequently does this trip?

This change fixes the failure:

--- a/tests/kms_cursor_legacy.c
+++ b/tests/kms_cursor_legacy.c
@@ -976,7 +976,7 @@ static void two_screens_flip_vs_cursor(igt_display_t *display, int nloops, bool
                if (!modeset)
                        igt_assert_eq(vbl.sequence, vblank_start + 1);

-               if (vblank_start && vbl.sequence == vblank_start + 1)
+               if (vblank_start && vbl.sequence >= vblank_start + 1)
                        vblank_matches = true;

The test says: "During modeset at least 1 page flip needs to match!" I am not sure if this is the right fix.

Is patch sent for review?

(In reply to Jani Saarinen from comment #10)
> Is patch sent for review?

Yes, I have posted it for review.

Thanks for driving this bug down! Please include the commit that fixes the bug before closing.

In this case, it appears to be:

commit 09796413394c5490c4adfc5cded5d4344af6af17
Author: Nischala Yelchuri <nischala.yelchuri@intel.com>
Date:   Wed Mar 13 13:16:35 2019 -0700

    tests/kms_cursor_legacy: Fix assertion failure due to vblank mismatch
    
    For the !modeset case, after a pageflip is performed on both crtc's,
    vblank counter is advanced by 1 when compared to its original value.
    
    For the modeset case, when pageflips and modesets are interleaved,
    vblank counter does not always advance by 1 as nonblocking modeset
    could fail, in which case we do not need vblank_matches.
    
    v2: Remove the assert statement instead of changing the condition
    
    Cc: Maarten Lankhorst <maarten.lankhorst@intel.com>
    Signed-off-by: Nischala Yelchuri <nischala.yelchuri@intel.com>
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=105454
    Signed-off-by: Maarten Lankhorst <maarten.lankhorst@linux.intel.com>

No failures happened after this patch landed. Thanks!

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.