Bug 97555

Summary:	[KBL,SKL,BXT] es2-cts.gtf.gl2extensiontests.egl_create_context.egl_create_context regression
Product:	Mesa	Reporter:	Mark Janes <mark.a.janes>
Component:	Drivers/DRI/i965	Assignee:	Ilia Mirkin <imirkin>
Status:	RESOLVED NOTOURBUG	QA Contact:	Intel 3D Bugs Mailing List <intel-3d-bugs>
Severity:	normal
Priority:	medium
Version:	git
Hardware:	Other
OS:	All
Whiteboard:
i915 platform:		i915 features:

Description Mark Janes 2016-08-31 20:11:26 UTC

bisected to:
a0b1260fe0577e504c8ef31c07e13025324af8f8
i965: enable OES_primitive_bounding_box with the no-op implementation


Standard Output

/tmp/build_root/m64/bin/es/cts/glcts --deqp-case=ES2-CTS.gtf.GL2ExtensionTests.egl_create_context.egl_create_context
dEQP Core GL-CTS-2.0 (0x0052484b) starting..
  target implementation = 'intel-gbm'

Test case 'ES2-CTS.gtf.GL2ExtensionTests.egl_create_context.egl_create_context'..
#+ (Line 313) Check failed: TestVersion(3, 1, EQUAL)
#+ (Line 318) Check failed: success
#+ (Line 365) Check failed: TestOpenGLES3(state)
  Fail (Fail)

DONE!

Test run totals:
  Passed:        0/1 (0.00%)
  Failed:        1/1 (100.00%)
  Not supported: 0/1 (0.00%)
  Warnings:      0/1 (0.00%)

Standard Error


pid: 1625
start time: 1472664007.1461957
end time: 1472664007.1978748

Comment 1 Ilia Mirkin 2016-08-31 20:24:51 UTC

I saw this in the jenkins run when I tested the patch, but assumed it was some sort of random bogosity. I guess it's consistent, but I'm not sure what the issue is.

It might actually be that ES 3.2 is being turned on (because the ver_3_2 requirements are being satisified in version.c) and the CTS test isn't ready for it somehow? Also some of my full ES 3.2 enablement series hasn't landed yet, but I doubt that's particularly relevant to this particular test.

However it'll require someone with CTS access to take a glance as those tests are non-public.

Comment 2 Mark Janes 2016-08-31 20:32:32 UTC

I have a task to update the cts to ES3.2 in the CI.  I'll attempt that now.

Comment 3 Ilia Mirkin 2016-08-31 20:36:03 UTC

Feel free to revert the patch btw - I don't personally care much about it, just figured it was the "nice" thing to do to enable it. Someone on the Intel end can figure out why the CTS test fails and then re-push it.

Comment 4 Matt Turner 2016-10-31 23:05:03 UTC

I believe enabling OES_primitive_bounding_box enabled ES 3.2, and the test wasn't aware 3.2 existed.

Looks like the internal Khronos bug https://cvs.khronos.org/bugzilla/show_bug.cgi?id=14470 fixed this with commit 0fa9ca76f80a52.

I'm not sure if we should mark this resolved, or if Mark wants to keep this open to track some CI change?

Comment 5 Mark Janes 2016-11-01 01:23:52 UTC

updating to the ES3.2 CTS fixed this test.

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.