Bug 102616

Summary:	[CI][SNB,HSW,APL,KBL] igt@gem_eio@in-flight - Failed assertion: __vgem_fence_signal(fd, fence) == 0
Product:	DRI	Reporter:	Martin Peres <martin.peres>
Component:	IGT	Assignee:	Default DRI bug account <dri-devel>
Status:	CLOSED FIXED	QA Contact:
Severity:	critical
Priority:	high	CC:	intel-gfx-bugs
Version:	XOrg git
Hardware:	Other
OS:	All
Whiteboard:	ReadyForDev
i915 platform:	BXT, HSW, KBL, SNB	i915 features:	GEM/Other

Description Martin Peres 2017-09-08 10:13:33 UTC

When running CI_DRM_3057, the machine shard-hsw hit the following assert in igt@gem_eio@in-flight:

(gem_eio:2052) igt-vgem-CRITICAL: Test assertion failure function vgem_fence_signal, file igt_vgem.c:191:
(gem_eio:2052) igt-vgem-CRITICAL: Failed assertion: __vgem_fence_signal(fd, fence) == 0
(gem_eio:2052) igt-vgem-CRITICAL: error: -110 != 0
Subtest in-flight failed.

Full logs: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3057/shard-hsw4/igt@gem_eio@in-flight.html

Comment 1 Martin Peres 2017-09-08 10:14:26 UTC

Also seen on SNB: https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3059/shard-snb5/igt@gem_eio@in-flight.html

Comment 2 Martin Peres 2017-09-08 10:18:55 UTC

Also seen on KBL and APL:
 - https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3059/shard-apl2/igt@gem_eio@in-flight.html
 - https://intel-gfx-ci.01.org/tree/drm-tip/CI_DRM_3059/shard-kbl5/igt@gem_eio@in-flight.html

Comment 3 Chris Wilson 2017-09-08 12:53:24 UTC

https://patchwork.freedesktop.org/series/30022/

Comment 4 Chris Wilson 2017-09-08 13:02:18 UTC

*** Bug 102620 has been marked as a duplicate of this bug. ***

Comment 5 Chris Wilson 2017-09-11 09:52:43 UTC

*** Bug 102652 has been marked as a duplicate of this bug. ***

Comment 6 Chris Wilson 2017-09-11 11:14:30 UTC

Related igt commit 4f082c35d2df545f81d202ae1a08463f6c123552 (upstream/master)
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Sep 8 13:48:05 2017 +0100

    igt/gem_eio: Install an exithandler to unwedge the device after failure
    
    Under normal conditions, we try to repair the damage we inflict to the
    GPU, but if we fail we don't. Make sure that if the test does die, we do
    try to restore normal operation by using an atexit handler.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102616
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Reviewed-by: Arkadiusz Hiler <arkadiusz.hiler@intel.com>

Comment 7 Chris Wilson 2017-09-19 12:17:01 UTC

commit 92e457d0ffe930369d09bc83d3d3ed90c9bd6cb1
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date:   Fri Sep 8 11:33:15 2017 +0100

    igt/gem_eio: inflight wedged requires long plugging
    
    As our hangcheck may exceed 10s to declare the device wedged, we need to
    hold the plugging fence indefinitely. This makes using vgem as our input
    fence unusable, so resort to using sw_sync. At the same time, we can
    then check that the async result is also -EIO.
    
    Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102616
    Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
    Reviewed-by: Michał Winiarski <michal.winiarski@intel.com>

Comment 8 Lakshmi 2018-08-22 07:04:42 UTC

igt@gem_eio@in-flight with "Failed assertion" was not seen recently and there are no open bugs related to failed assertion. So closing this bug.

Comment 9 Lakshmi 2018-08-22 07:06:00 UTC

(In reply to Lakshmi from comment #8)
> igt@gem_eio@in-flight test failure with "Failed assertion" was not seen recently and there are no open bugs related to failed assertion. So closing this bug.

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.