New issue
Advanced search Search tips
Note: Color blocks (like or ) mean that a user may not be available. Tooltip shows the reason.

Issue 606021 link

Starred by 3 users

Issue metadata

Status: Fixed
Owner:
Closed: May 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: All
Pri: 1
Type: Bug

Blocking:
issue 429053



Sign in to add a comment

deqp/functional/gles3/texturefiltering/cube-sizes-*.html flaky on a bunch of platforms

Project Member Reported by zmo@chromium.org, Apr 22 2016

Issue description

I can't reproduce it locally, but it fails on Win AMD, Linux NVidia.

If anyone can reproduce it, please take a shot to see what the problem is.

 
Project Member

Comment 1 by bugdroid1@chromium.org, Apr 22 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/6ea77ea4e7e8379659ec2d6c90e97cdbd88c3a33

commit 6ea77ea4e7e8379659ec2d6c90e97cdbd88c3a33
Author: zmo <zmo@chromium.org>
Date: Fri Apr 22 23:29:37 2016

Roll WebGL 51b33d0..165311e

https://chromium.googlesource.com/external/khronosgroup/webgl.git/+log/51b33d0..165311e

BUG= 606021 , 606074 
TEST=bots
TBR=bajones@chromium.org,qiankun.miao@intel.com
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.win:win_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.linux:linux_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1918463002

Cr-Commit-Position: refs/heads/master@{#389298}

[modify] https://crrev.com/6ea77ea4e7e8379659ec2d6c90e97cdbd88c3a33/DEPS
[modify] https://crrev.com/6ea77ea4e7e8379659ec2d6c90e97cdbd88c3a33/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Project Member

Comment 2 by bugdroid1@chromium.org, Apr 23 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb

commit 34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb
Author: kbr <kbr@chromium.org>
Date: Sat Apr 23 04:15:05 2016

Expand suppressions for texturefiltering* tests.

BUG= 606021 , 606114
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
TBR=zmo@chromium.org, jmadill@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1920613002

Cr-Commit-Position: refs/heads/master@{#389360}

[modify] https://crrev.com/34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 3 by zmo@chromium.org, Apr 25 2016

This test also fails (flaky) on Windows NVidia (both NVidia GPUs)
Project Member

Comment 4 by bugdroid1@chromium.org, Apr 25 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/57fe2f9f0c570b3bd677529b417782ec04cb01b3

commit 57fe2f9f0c570b3bd677529b417782ec04cb01b3
Author: zmo <zmo@chromium.org>
Date: Mon Apr 25 16:50:12 2016

Update WebGL 2 conformance test expectations.

BUG= 606021 
TEST=bots
TBR=bajones@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1921753002

Cr-Commit-Position: refs/heads/master@{#389503}

[modify] https://crrev.com/57fe2f9f0c570b3bd677529b417782ec04cb01b3/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 5 by zmo@chromium.org, Apr 25 2016

OK, this test turns out to be flaky on all windows bots, including AMD, NVidia, and Intel.
Project Member

Comment 6 by bugdroid1@chromium.org, Apr 25 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/db6d20e8689ac07517ac5a946e4107585d6f7b75

commit db6d20e8689ac07517ac5a946e4107585d6f7b75
Author: zmo <zmo@chromium.org>
Date: Mon Apr 25 20:32:51 2016

WebGL 2 conformance test expectations update.

BUG= 606021 
TEST=bots
TBR=bajones@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1913933002

Cr-Commit-Position: refs/heads/master@{#389548}

[modify] https://crrev.com/db6d20e8689ac07517ac5a946e4107585d6f7b75/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Project Member

Comment 7 by bugdroid1@chromium.org, Apr 25 2016

Labels: merge-merged-2716
The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb

commit 34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb
Author: kbr <kbr@chromium.org>
Date: Sat Apr 23 04:15:05 2016

Expand suppressions for texturefiltering* tests.

BUG= 606021 , 606114
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
TBR=zmo@chromium.org, jmadill@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1920613002

Cr-Commit-Position: refs/heads/master@{#389360}

[modify] https://crrev.com/34a7a8cbbf500710c5867a5ecb7cdc6ba369e2cb/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Project Member

Comment 8 by bugdroid1@chromium.org, Apr 25 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/57fe2f9f0c570b3bd677529b417782ec04cb01b3

commit 57fe2f9f0c570b3bd677529b417782ec04cb01b3
Author: zmo <zmo@chromium.org>
Date: Mon Apr 25 16:50:12 2016

Update WebGL 2 conformance test expectations.

BUG= 606021 
TEST=bots
TBR=bajones@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review URL: https://codereview.chromium.org/1921753002

Cr-Commit-Position: refs/heads/master@{#389503}

[modify] https://crrev.com/57fe2f9f0c570b3bd677529b417782ec04cb01b3/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 9 by kbr@chromium.org, Apr 25 2016

Is the test flaky or reliably failing? Flaky failures would actually be more distressing since we wouldn't necessarily know why the test occasionally passes.

Comment 10 by zmo@chromium.org, Apr 25 2016

It's flaky, but pass not that often, so can't mark it as Flaky.

I am trying to reproduce on my Windows.

Comment 11 by kbr@chromium.org, Apr 26 2016

Components: Blink>JavaScript
Summary: deqp/functional/gles3/texturefiltering/cube-sizes-*.html flaky on a bunch of platforms (was: deqp/functional/gles3/texturefiltering/cube-sizes-00.html fails on a bunch of platforms)
Expanding the description of this bug since a flake of WebglConformance.deqp_functional_gles3_texturefiltering_cube_sizes_03 was seen here:

https://build.chromium.org/p/chromium.gpu.fyi/builders/Win7%20Release%20%28NVIDIA%29/builds/24343

The log up to the point of failure:

[3820:640:0426/013021:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_linear: Image verification failed", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Init testcase: filtering", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_nearest", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Test0", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Test1", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Coordinates: 0.8,0.8 -> 1.25,1.2", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Test2", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Coordinates: -1.19,-1.3 -> 1.1,1.35", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Test3", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Coordinates: -1.2,-1.1 -> -0.8,-0.8", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_linear", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Test0", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face -X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Face +X", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "Warning: Verification against high precision requirements failed, trying with lower requirements.", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "ERROR: Verification against lowprecision requirements failed, failing test case.", source:  (11)
[3820:640:0426/013021:INFO:CONSOLE(11)] "FAIL Image verification failed", source:  (11)


I wonder whether optimizations in V8 could be affecting the generation of the reference images?

Marking this test flaky as well.

stdout.txt
1.6 MB View Download
Project Member

Comment 12 by bugdroid1@chromium.org, Apr 26 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/eb697a2f3d13684d56a28951460b64839ed5817c

commit eb697a2f3d13684d56a28951460b64839ed5817c
Author: kbr <kbr@chromium.org>
Date: Tue Apr 26 09:52:15 2016

Mark deqp_functional_gles3_texturefiltering_cube_sizes_03 flaky.

BUG= 606021 
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
TBR=zmo@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1917093003

Cr-Commit-Position: refs/heads/master@{#389737}

[modify] https://crrev.com/eb697a2f3d13684d56a28951460b64839ed5817c/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 13 by kbr@chromium.org, May 17 2016

Components: Blink>WebGL
Labels: -Pri-3 Pri-2
This is affecting try jobs that trigger the optional GPU trybots. See for example https://codereview.chromium.org/1990433002/ and https://build.chromium.org/p/tryserver.chromium.linux/builders/linux_optional_gpu_tests_rel/builds/526 :

[ RUN      ] WebglConformance.deqp_functional_gles3_texturefiltering_cube_sizes_03
(INFO) 2016-05-17 10:32:10,404 cache_temperature.EnsurePageCacheTemperature:29  PageCacheTemperature: any
[26790:26790:0517/103211:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_linear: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_nearest", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test1", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: 0.8,0.8 -> 1.25,1.2", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test2", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.19,-1.3 -> 1.1,1.35", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test3", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.2,-1.1 -> -0.8,-0.8", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Y", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +Z", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_linear", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "FAIL Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Error: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_nearest_mipmap_nearest", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_nearest_mipmap_nearest: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "FAIL Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Error: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_linear_mipmap_nearest", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_linear_mipmap_nearest: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "FAIL Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Error: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_nearest_mipmap_linear", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_nearest_mipmap_linear: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "FAIL Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Error: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Init testcase: filtering.cube_sizes.7x7_linear_mipmap_linear", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Test0", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Coordinates: -1.25,-1.2 -> 1.2,1.25", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face -X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Face +X", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "filtering.cube_sizes.7x7_linear_mipmap_linear: Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "FAIL Cannot remap cube coordinates", source:  (11)
[26790:26790:0517/103211:INFO:CONSOLE(11)] "Error: Cannot remap cube coordinates", source:  (11)
Traceback (most recent call last):
  File "/tmp/run6X8SYj/third_party/catapult/telemetry/telemetry/internal/story_runner.py", line 84, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File "/tmp/run6X8SYj/content/test/gpu/gpu_tests/gpu_test_base.py", line 122, in RunStory
    RunStoryWithRetries(DesktopGpuSharedPageState, self, results)
  File "/tmp/run6X8SYj/content/test/gpu/gpu_tests/gpu_test_base.py", line 72, in RunStoryWithRetries
    super(cls, shared_page_state).RunStory(results)
  File "/tmp/run6X8SYj/third_party/catapult/telemetry/telemetry/page/shared_page_state.py", line 332, in RunStory
    self._current_page, self._current_tab, results)
  File "/tmp/run6X8SYj/content/test/gpu/gpu_tests/webgl_conformance.py", line 87, in ValidateAndMeasurePage
    raise page_test.Failure(_WebGLTestMessages(tab))
Failure: filtering.cube_sizes.7x7_linear: Cannot remap cube coordinates
Init testcase: filtering
Init testcase: filtering.cube_sizes.7x7_nearest
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
Face -Y
Face +Y
Face -Z
Face +Z
Test1
Coordinates: 0.8,0.8 -> 1.25,1.2
Face -X
Face +X
Face -Y
Face +Y
Face -Z
Face +Z
Test2
Coordinates: -1.19,-1.3 -> 1.1,1.35
Face -X
Face +X
Face -Y
Face +Y
Face -Z
Face +Z
Test3
Coordinates: -1.2,-1.1 -> -0.8,-0.8
Face -X
Face +X
Face -Y
Face +Y
Face -Z
Face +Z
Init testcase: filtering.cube_sizes.7x7_linear
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
FAIL Cannot remap cube coordinates
Error: Cannot remap cube coordinates
Init testcase: filtering.cube_sizes.7x7_nearest_mipmap_nearest
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
filtering.cube_sizes.7x7_nearest_mipmap_nearest: Cannot remap cube coordinates
FAIL Cannot remap cube coordinates
Error: Cannot remap cube coordinates
Init testcase: filtering.cube_sizes.7x7_linear_mipmap_nearest
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
filtering.cube_sizes.7x7_linear_mipmap_nearest: Cannot remap cube coordinates
FAIL Cannot remap cube coordinates
Error: Cannot remap cube coordinates
Init testcase: filtering.cube_sizes.7x7_nearest_mipmap_linear
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
filtering.cube_sizes.7x7_nearest_mipmap_linear: Cannot remap cube coordinates
FAIL Cannot remap cube coordinates
Error: Cannot remap cube coordinates
Init testcase: filtering.cube_sizes.7x7_linear_mipmap_linear
Test0
Coordinates: -1.25,-1.2 -> 1.2,1.25
Face -X
Face +X
filtering.cube_sizes.7x7_linear_mipmap_linear: Cannot remap cube coordinates
FAIL Cannot remap cube coordinates
Error: Cannot remap cube coordinates


[  FAILED  ] WebglConformance.deqp_functional_gles3_texturefiltering_cube_sizes_03 (923 ms)

Project Member

Comment 14 by bugdroid1@chromium.org, May 17 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/ba57588e44bc5fc81985a68927f4c04265bd7ca9

commit ba57588e44bc5fc81985a68927f4c04265bd7ca9
Author: kbr <kbr@chromium.org>
Date: Tue May 17 21:01:03 2016

Mark deqp/functional/gles3/texturefiltering/cube_sizes_03.html flaky on Linux.

BUG= 606021 
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel
NOTRY=true
TBR=cwallez@chromium.org,zmo@chromium.org

Review-Url: https://codereview.chromium.org/1985213002
Cr-Commit-Position: refs/heads/master@{#394225}

[modify] https://crrev.com/ba57588e44bc5fc81985a68927f4c04265bd7ca9/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 15 by zmo@chromium.org, May 17 2016

Owner: zmo@chromium.org
Status: Started (was: Available)
I am looking this flakiness.

Comment 16 by zmo@chromium.org, May 18 2016

This looks like a bug in v8.

Steps to reproduce:

1) On a windows machine, build and run chrome with --enable-unsafe-es3-apis --allow-file-access-to-files
   * Note that the bug also manifests on other platforms, but it's easiest to reproduce on Windows: almost every 10 runs, I can catch one failure on my win bot.
2) open third_party/webgl/src/sdk/tests/deqp/functional/gles3/texturefiltering/cube_sizes_00.html
3) refresh a few times, and hopefully see a failure

Comment 17 by zmo@chromium.org, May 18 2016

Further looking into third_party/webgl/src/sdk/tests/deqp/framework/common/tcuTexture.js

The function remapCubeEdgeCoords() is where the failure happens.

Change the last line of this function to

  throw new Error('Cannot remap cube coordinates: coords = ' + coords +
                  ', canonizedCoords = ' + canonizedCoords +
                  ', origCoords = {' + origCoords.face + ', (' + origCoords.s + ',' + origCoords.t +
                  ')}, size = ' + size);

when this happens, canonizedCoords looks like junk data.

Comment 18 by kbr@chromium.org, May 18 2016

Labels: -merge-merged-2716

Comment 19 by kbr@chromium.org, May 18 2016

Cc: ishell@chromium.org bmeu...@chromium.org
Labels: -Pri-2 Pri-1
Owner: jkummerow@chromium.org
Status: Assigned (was: Started)
Jakob, Igor, Benedikt: I'm not sure in which backend this bug might be (Crankshaft, Turbofan, etc.) but it seems to be an intermittent bug in the code generator. Could one of you please triage it and reassign as necessary? This bug is high priority for our team, as it is causing random test failures affecting the commit queue. There are other random failures in the same test suite that are probably related, but this is the most obvious one, and if it's fixed it might fix the others.

Reproducible on 52.0.2739.0 (Official Build) canary (64-bit). about:gpu attached.

To reiterate, to reproduce:

1) Install Canary on Windows
2) "C:\Users\[username]\AppData\Local\Google\Chrome SxS\Application\chrome.exe" --user-data-dir=C:\Users\[username]\tmp\t1 --enable-unsafe-es3-apis --allow-file-access-from-files
3) Navigate to e.g. file:///D:/src/chrome/src/third_party/webgl/src/sdk/tests/deqp/functional/gles3/texturefiltering/cube_sizes_00.html (obviously, fix up the path)
(Alternatively, navigate to https://www.khronos.org/registry/webgl/sdk/tests/deqp/functional/gles3/texturefiltering/cube_sizes_00.html , though you won't be able to modify the sources)

Refresh a few times. Within a few iterations you'll see something like:

"FAIL filtering.cube_sizes.8x8_linear: Cannot remap cube coordinates"

This is the bug.

about-gpu.txt
5.4 KB View Download

Comment 20 Deleted

I also did some investigations on failure of deqp/functional/gles3/texturefiltering/cube_sizes_03.html. I tested it on Linux with nVidia grahpics card. It should be caused by a V8 roll, r386417. Before this roll, the test always passed. After this roll, it fails in remapCubeEdgeCoords() the same as comments#17 sometimes.
Cc: jkummerow@chromium.org
Owner: mvstan...@chromium.org
"canonizedCoords" (see #17) is initialized from either of a set of array literals:

    switch (origCoords.face) {
        case tcuTexture.CubeFace.CUBEFACE_NEGATIVE_X: canonizedCoords = [0, size - 1 - coords[1], coords[0]]; break;
        case tcuTexture.CubeFace.CUBEFACE_POSITIVE_X: canonizedCoords = [size - 1, size - 1 - coords[1], size - 1 - coords[0]]; break;
        case tcuTexture.CubeFace.CUBEFACE_NEGATIVE_Y: canonizedCoords = [coords[0], 0, size - 1 - coords[1]]; break;
        case tcuTexture.CubeFace.CUBEFACE_POSITIVE_Y: canonizedCoords = [coords[0], size - 1, coords[1]]; break;
        case tcuTexture.CubeFace.CUBEFACE_NEGATIVE_Z: canonizedCoords = [size - 1 - coords[0], size - 1 - coords[1], 0]; break;
        case tcuTexture.CubeFace.CUBEFACE_POSITIVE_Z: canonizedCoords = [coords[0], size - 1 - coords[1], size - 1]; break;
        default: throw new Error('Invalid cube face:' + origCoords.face);
    }

In the regression range, there's exactly one CL that says that it's dealing with literals: https://codereview.chromium.org/1865013002
Michael, can you please check if there's any corner case that may still be going wrong? Wild guess: maybe if one of the case-branches is uninitialized?

Comment 23 by kbr@chromium.org, May 19 2016

Thanks for tracking this down Jakob and Qiankun.

It is strange that the test works most of the time. There must be something non-deterministic in compilation.

Comment 24 by kbr@chromium.org, May 19 2016

 Issue 612378  has been merged into this issue.
Status: Started (was: Assigned)
Checking in a fix...thanks for the investigation and apologies for the effort involved!
Project Member

Comment 27 by bugdroid1@chromium.org, May 20 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/v8/v8.git/+/b71f1cc2f4fe7ff0779cc3cf95f104c2bd2c5259

commit b71f1cc2f4fe7ff0779cc3cf95f104c2bd2c5259
Author: mvstanton <mvstanton@chromium.org>
Date: Fri May 20 13:07:06 2016

Bugfix: Crankshaft array literals with incorrect values.

In VisitArrayLiteral(), we stopped creating boilerplates during
 compilation. This created a bug, because we'd emit an HStoreKeyed with an
 ElementsKind based on the initial elements kind of the boilerplate. Since
 boilerplates may transition as part of elements transition feedback, this can
 lead to incorrect values (storing a smi in a double array).

BUG= chromium:606021 

Review-Url: https://codereview.chromium.org/2000673002
Cr-Commit-Position: refs/heads/master@{#36408}

[modify] https://crrev.com/b71f1cc2f4fe7ff0779cc3cf95f104c2bd2c5259/src/crankshaft/hydrogen.cc
[add] https://crrev.com/b71f1cc2f4fe7ff0779cc3cf95f104c2bd2c5259/test/mjsunit/regress/regress-606021.js

Labels: Merge-Request-52 Merge-Request-51
Status: Fixed (was: Started)

Comment 29 by tin...@google.com, May 20 2016

Labels: -Merge-Request-51 Merge-Review-51 Hotlist-Merge-Review
[Automated comment] DEPS changes referenced in bugdroid comments, needs manual review.

Comment 30 by tin...@google.com, May 20 2016

Labels: -Merge-Request-52 Merge-Review-52
[Automated comment] DEPS changes referenced in bugdroid comments, needs manual review.

Comment 31 by zmo@chromium.org, May 20 2016

Thanks for the quick turnaround. Let us know when the fix is rolled into chromium, so we can remove our suppression.
Will this also fix cube_combinations? These flaky tests are still plaguing the ANGLE CQ. 

Comment 33 by zmo@chromium.org, May 20 2016

Yes, the same root cause.
Awesome.
Is this bug applicable to specific OS or all os?

Also before we approve merge to M51, Could you please confirm whether this change is baked/verified in Canary and safe to merge?
Cc: sshruthi@chromium.org

Comment 37 by zmo@chromium.org, May 20 2016

It applies to all OS.
Labels: OS-All
mvstanton@, before we approve merge to M51, Could you please confirm whether this change is baked/verified in Canary and safe to merge?
#31: The fix is in Chromium, you can remove your suppression(s).

#39: The fix is in Canary as of 2745 and is looking good there.
Cc: geoffl...@chromium.org zmo@chromium.org cwallez@chromium.org jmad...@chromium.org
 Issue 607393  has been merged into this issue.
+ sshruthi@ (M51 Desktop TPM), could you please PTAL and approve if you think it is ok to do so as I see multiple cls are listed here. As per comment #40, The fix is in Canary as of 2745 and is looking good there.
Labels: -Merge-Review-51 Merge-Approved-51
Merge approved for change in c#27 to M51 (branch 2704). Please merge in the change ASAP, we are cutting a stable candidate at 5PM PST today. 

Comment 44 by zmo@chromium.org, May 23 2016

Status: Assigned (was: Fixed)
Actually I am rolling back v8 due to a different failure on Linux GPU bots.

https://codereview.chromium.org/2005983003/

Comment 45 by zmo@chromium.org, May 23 2016

Status: Fixed (was: Assigned)
Never mind.  The roll back does not affect this fix.
Project Member

Comment 46 by bugdroid1@chromium.org, May 23 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/aa4c6dde834eb922f7b3ba1198d2daa94e77d124

commit aa4c6dde834eb922f7b3ba1198d2daa94e77d124
Author: zmo <zmo@chromium.org>
Date: Mon May 23 22:33:33 2016

Remove flaky texturefiltering tests after v8 fix being rolled in.

BUG= 606021 
TEST=GPU bots
TBR=jmadill@chromium.org
NOTRY=true
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_optional_gpu_tests_rel;tryserver.chromium.mac:mac_optional_gpu_tests_rel;tryserver.chromium.win:win_optional_gpu_tests_rel

Review-Url: https://codereview.chromium.org/2001223002
Cr-Commit-Position: refs/heads/master@{#395439}

[modify] https://crrev.com/aa4c6dde834eb922f7b3ba1198d2daa94e77d124/content/test/gpu/gpu_tests/webgl2_conformance_expectations.py

Comment 47 Deleted

Update on comment #43:
Merge approved for change in c#27 to M51 (branch 2704). Please merge in the change ASAP,latest by 12:00 PM PST, noon tomorrow (Tuesday). 
Hi, my apologies here...M51 doesn't need the fix. I mistakenly thought it did!

However, M52 does need it.

Thank you, I'll make no M51 changes.
Cc: -bmeu...@chromium.org
Labels: -Merge-Approved-51 -Merge-Review-52 Merge-Approved-52 Merge-rejected-5.1
Project Member

Comment 52 by bugdroid1@chromium.org, May 24 2016

Labels: merge-merged-5.2
The following revision refers to this bug:
  https://chromium.googlesource.com/v8/v8.git/+/38dfd6c521fe216c95d3eec55104184dc9adabe0

commit 38dfd6c521fe216c95d3eec55104184dc9adabe0
Author: Michael Stanton <mvstanton@chromium.org>
Date: Tue May 24 12:50:50 2016

Version 5.2.361.7 (cherry-pick)

Merged b71f1cc2f4fe7ff0779cc3cf95f104c2bd2c5259

Bugfix: Crankshaft array literals with incorrect values.

BUG= chromium:606021 
LOG=N
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/2010503002 .

Cr-Commit-Position: refs/branch-heads/5.2@{#11}
Cr-Branched-From: 2cd36d6d0439ddfbe84cd90e112dced85084ec95-refs/heads/5.2.361@{#1}
Cr-Branched-From: 3fef34e02388e07d46067c516320f1ff12304c8e-refs/heads/master@{#36332}

[modify] https://crrev.com/38dfd6c521fe216c95d3eec55104184dc9adabe0/src/crankshaft/hydrogen.cc
[add] https://crrev.com/38dfd6c521fe216c95d3eec55104184dc9adabe0/test/mjsunit/regress/regress-606021.js

Comment 53 Deleted

Cc: -sshruthi@chromium.org
Please merge your change to M52 branch 2743 before 4:00 PM PST tomorrow, Wednesday (05/25).So we can take it for this week last M52 Dev release on Thursday (05/26).Thank you.
Labels: -Merge-Approved-52
Components: -Internals>GPU>WebGL

Sign in to add a comment