Bug 102405

Summary:	[CI][SNB] igt@gem_sync@basic-store-each: Failed assertion: intel_detect_and_clear_missed_interrupts(fd) == 0
Product:	DRI	Reporter:	Martin Peres <martin.peres>
Component:	DRM/Intel	Assignee:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Status:	CLOSED DUPLICATE	QA Contact:	Intel GFX Bugs mailing list <intel-gfx-bugs>
Severity:	normal
Priority:	medium	CC:	intel-gfx-bugs
Version:	XOrg git
Hardware:	Other
OS:	All
Whiteboard:	ReadyForDev
i915 platform:	SNB	i915 features:	display/Other

Description Martin Peres 2017-08-25 11:09:54 UTC

On CI_DRM_3000, the machine shard-snb4 hit the following assertion while running igt@gem_sync@basic-store-each:

(gem_sync:1544) CRITICAL: Test assertion failure function store_ring, file gem_sync.c:277:
(gem_sync:1544) CRITICAL: Failed assertion: intel_detect_and_clear_missed_interrupts(fd) == 0
(gem_sync:1544) CRITICAL: error: 1 != 0
Subtest basic-store-each failed.

Full logs: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3000/shard-snb4/igt@gem_sync@basic-store-each.html

Comment 1 Chris Wilson 2017-08-25 11:14:44 UTC

Yup, for exactly the same reason as ilk, snb, ivb, byt: the seqno write is not ordered wrt the interrupt. Our w/a for this hw issue is to add a delay following the interrupt, but that's not a guarantee (and there is no way we (at least that we know of) to flush that write from the GPU into memory).

*** This bug has been marked as a duplicate of bug 10007 ***

Comment 2 Martin Peres 2017-08-25 11:15:45 UTC

Thanks for the info, Chris!

Comment 3 Martin Peres 2017-08-25 11:19:06 UTC


*** This bug has been marked as a duplicate of bug 100007 ***

Comment 4 Martin Peres 2017-08-25 11:21:39 UTC

Fix the platform tag. Sorry, getting tired!

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.