Bug 78877 - [BDW bisected]Piglit spec_ARB_sample_shading_builtin-gl-sample-id_2 fails
Summary: [BDW bisected]Piglit spec_ARB_sample_shading_builtin-gl-sample-id_2 fails
Status: CLOSED FIXED
Alias: None
Product: DRI
Classification: Unclassified
Component: DRM/Intel (show other bugs)
Version: unspecified
Hardware: All Linux (All)
: highest major
Assignee: Mika Kuoppala
QA Contact: Intel GFX Bugs mailing list
URL:
Whiteboard:
Keywords:
Depends on:
Blocks:
 
Reported: 2014-05-19 05:16 UTC by lu hua
Modified: 2016-10-13 06:43 UTC (History)
6 users (show)

See Also:
i915 platform:
i915 features:


Attachments
dmesg (81.40 KB, text/plain)
2014-05-19 05:19 UTC, lu hua
no flags Details

Description lu hua 2014-05-19 05:16:57 UTC
System Environment:
--------------------------
Platform: Broadwell
Libdrm:	(master)libdrm-2.4.54-8-g305478ce02ebd908a75c9830ecea15f6e2469b42
Mesa:	(master)5646319f25c7880b3706bb7590e24c84fd8de0fc
Xserver:(master)xorg-server-1.15.99.902-91-g01e18af17f8dc91451fbd0902049045afd1cea7e
Xf86_video_intel:(master)2.99.911-173-gdf297e9fafe1a7de1036f3151f93de67661c2a4e
Libva:	(staging)968ade9411de9c5ae2eead0a7e8755747a32a3a2
Libva_intel_driver:		(staging)1c4d3468229797e787f4b99b0729baf90a115a1d
Kernel:	(drm-intel-nightly)f79ba79cf037eea9ee757ad37730b00f43d5ef80

Bug detailed description:
-----------------------------
It fails on Broadwell with -queued and -nightly kernel, It works well on -fixes kernel.
Following cases also fail:
spec_ARB_sample_shading_builtin-gl-sample-mask_0
spec_ARB_sample_shading_builtin-gl-sample-mask_2
spec_ARB_sample_shading_builtin-gl-sample-mask_4
spec_ARB_sample_shading_builtin-gl-sample-mask_6
spec_ARB_sample_shading_builtin-gl-sample-mask_8
spec_ARB_sample_shading_builtin-gl-sample-position_2
spec_EXT_framebuffer_multisample_multisample-blit_2_color
spec_EXT_framebuffer_multisample_multisample-blit_2_color_linear
spec_EXT_framebuffer_multisample_multisample-blit_2_depth
spec_EXT_framebuffer_multisample_no-color_2_depth-computed_combined
spec_EXT_framebuffer_multisample_no-color_2_depth-computed_separate
spec_EXT_framebuffer_multisample_no-color_2_depth-computed_single
spec_EXT_framebuffer_multisample_no-color_2_depth_combined
spec_EXT_framebuffer_multisample_no-color_2_depth_separate
spec_EXT_framebuffer_multisample_no-color_2_depth_single
spec_EXT_framebuffer_multisample_unaligned-blit_2_color_msaa
spec_EXT_framebuffer_multisample_unaligned-blit_2_depth_msaa

The latest known good commit: b7c0d9df97c10ec5693a838df2fd53058f8e9e96
The latest known bad commit: d3b448d9917a3d6531e499d88bfb13ea5e31e4ad

output:
Probe color at (124,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 1.000000 0.000000 0.000000 1.000000
FBO attachment = TEXTURE, result = fail
Probe color at (2,0)
  Expected: 0.000000 1.000000 0.000000 1.000000
  Observed: 1.000000 0.000000 0.000000 1.000000
FBO attachment = RENDERBUFFER, result = fail
PIGLIT: {'result': 'fail' }


Reproduce steps:
---------------------------- 
1. xinit
2. bin/arb_sample_shading-builtin-gl-sample-id 2 -auto
Comment 1 lu hua 2014-05-19 05:19:09 UTC
Created attachment 99294 [details]
dmesg
Comment 2 Daniel Vetter 2014-05-19 08:50:59 UTC
The bisect, please.
Comment 3 lu hua 2014-05-20 07:31:23 UTC
Bisect shows: 9d0a6fa6c5e618bd978d625a215dc4a240ba3b3c is the first bad commit
commit 9d0a6fa6c5e618bd978d625a215dc4a240ba3b3c
Author:     Mika Kuoppala <mika.kuoppala@linux.intel.com>
AuthorDate: Wed May 14 17:02:16 2014 +0300
Commit:     Daniel Vetter <daniel.vetter@ffwll.ch>
CommitDate: Wed May 14 19:16:13 2014 +0200

    drm/i915: add render state initialization

    HW guys say that it is not a cool idea to let device
    go into rc6 without proper 3d pipeline state.

    For each new uninitialized context, generate a
    valid null render state to be run on context
    creation.

    This patch introduces a skeleton with empty states.

    v2: - No need to vmap (Chris Wilson)
        - use .c files for state (Daniel Vetter)
        - no need to flush as i915_add_request does it
        - remove parameter for batch alloc size
        - don't wait for the init (Ben Widawsky)

    v3: - move to cpu/gpu (Chris Wilson)

    Tested-by: Kristen Carlson Accardi <kristen@linux.intel.com> (v1)
    Tested-by: Oscar Mateo <oscar.mateo@intel.com>
    Reviewed-by: Damien Lespiau <damien.lespiau@intel.com>
    Signed-off-by: Mika Kuoppala <mika.kuoppala@intel.com>
    Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Comment 4 Chris Wilson 2014-05-20 07:48:48 UTC
Amusing.
Comment 5 Daniel Vetter 2014-05-20 09:15:10 UTC
Adding Ken. Big wtf indeed.
Comment 6 Kenneth Graunke 2014-05-20 09:29:03 UTC
Odd.  Does this work if you set INTEL_DEBUG=no16 before running the tests?
Comment 7 Daniel Vetter 2014-05-22 13:37:54 UTC
(In reply to comment #6)
> Odd.  Does this work if you set INTEL_DEBUG=no16 before running the tests?

Please test.
Comment 8 lu hua 2014-05-23 04:57:29 UTC
(In reply to comment #6)
> Odd.  Does this work if you set INTEL_DEBUG=no16 before running the tests?

It still fails.
Comment 9 Ben Widawsky 2014-06-03 23:10:11 UTC
Please test:
http://patchwork.freedesktop.org/patch/26884/
Comment 10 lu hua 2014-06-05 07:23:50 UTC
(In reply to comment #9)
> Please test:
> http://patchwork.freedesktop.org/patch/26884/

patching file drivers/gpu/drm/i915/i915_gem_context.c
Hunk #1 succeeded at 598 (offset -43 lines).
Hunk #2 FAILED at 740.
1 out of 2 hunks FAILED -- saving rejects to file drivers/gpu/drm/i915/i915_gem_context.c.rej
Comment 11 Ben Widawsky 2014-06-25 00:06:45 UTC
Okay. Pleasre retest with the late drm-intel-nightly. It should have all the relevant fixes for the null context. If that doesn't work, we'll have to think of more ideas.
Comment 12 lu hua 2014-06-25 06:05:09 UTC
Works well on latest -nightly kernel.
Comment 13 lu hua 2014-06-25 06:05:22 UTC
Verified.Fixed.
Comment 14 Jari Tahvanainen 2016-10-13 06:43:26 UTC
Closing verified+fixed (commit 967ab6b).


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.