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

Issue 729160 link

Starred by 6 users

Issue metadata

Status: Verified
Owner:
Closed: Jun 2017
Cc:
EstimatedDays: ----
NextAction: ----
OS: Chrome
Pri: 1
Type: Bug



Sign in to add a comment

[Intel only]: graphics_WebGLAquarium is broken starting 9602.0.0

Project Member Reported by laszio@chromium.org, Jun 2 2017

Issue description

Not sure if it works on other platform or not. Also not sure if this is telemetry related but it looks to me that it breaks in telemetry so I'm using the component telemetry temporarily.

$ /usr/bin/test_that --debug --autotest_dir /home/laszio/trunk/src/third_party/autotest/files/ --fast  --board=lumpy chromeos2-row9-rack9-host9.cros graphics_WebGLAquarium

...
Traceback (most recent call last):
  File "/usr/local/autotest/common_lib/test.py", line 818, in _call_test_function
    return func(*args, **dargs)
  File "/usr/local/autotest/common_lib/test.py", line 471, in execute
    dargs)
  File "/usr/local/autotest/common_lib/test.py", line 348, in _call_run_once_with_retry
    postprocess_profiled_run, args, dargs)
  File "/usr/local/autotest/common_lib/test.py", line 381, in _call_run_once
    self.run_once(*args, **dargs)
  File "/usr/local/autotest/tests/graphics_WebGLAquarium/graphics_WebGLAquarium.py", line 314, in run_once
    self.run_fish_test(cr.browser, test_url, n)
  File "/usr/local/autotest/tests/graphics_WebGLAquarium/graphics_WebGLAquarium.py", line 142, in run_fish_test
    avg_fps = tab.EvaluateJavaScript('g_crosFpsCounter.getAvgFps();')
  File "/usr/local/telemetry/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 52, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/browser/web_contents.py", line 167, in EvaluateJavaScript
    return self._inspector_backend.EvaluateJavaScript(*args, **kwargs)
  File "/usr/local/telemetry/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 52, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_backend.py", line 243, in EvaluateJavaScript
    return self._EvaluateJavaScript(expression, context_id, timeout)
  File "/usr/local/telemetry/src/third_party/catapult/common/py_trace_event/py_trace_event/trace_event_impl/decorators.py", line 52, in traced_function
    return func(*args, **kwargs)
  File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_backend.py", line 37, in inner
    return func(inspector_backend, *args, **kwargs)
  File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_backend.py", line 507, in _EvaluateJavaScript
    return self._runtime.Evaluate(expression, context_id, timeout)
  File "/usr/local/telemetry/src/third_party/catapult/telemetry/telemetry/internal/backends/chrome_inspector/inspector_runtime.py", line 58, in Evaluate
    return res['result']['result']['value']
KeyError: 'value'
...

 
Cc: -nednguyen@chromium.org ihf@chromium.org
Components: -Tests>Telemetry
Pretty sure this is not a telemetry issue

Comment 3 by ihf@chromium.org, Jun 2 2017

Cc: pwang@chromium.org
Owner: gurcheta...@chromium.org
Summary: [Intel only]: graphics_WebGLAquarium is broken starting 9602.0.0 (was: autotest: graphics_WebGLAquarium is broken on CrOS)
Gurchetan, could you please take a look? It seems to have happened between 9601 and 9602 but that doesn't show much. So might be between 9600 and 9602 which indeed contains lots of Intel kernel changes
https://crosland.corp.google.com/log/9600.0.0..9602.0.0

https://wmatrix.googleplex.com/unfiltered?hide_missing=True&releases=tot&tests=graphics_WebGLAquarium&days_back=100

Comment 4 by ihf@chromium.org, Jun 2 2017

Issue 728949 has been merged into this issue.
Owner: dcasta...@chromium.org
The culprit is:

https://chromium.googlesource.com/chromiumos/platform2/+/ec0ce0be8b8ba6a82f6a7f06cb21dc91bd372411

Assigning to @dcastagna.
Cc: gurcheta...@chromium.org
That's surprising, that change shouldn't affect Intel at all.
Gurchetan, if you happen to have a setup where you can easily reproduce this, would you mind checking if the parameter "--enable-drm-atomic" is passed to the chrome binary?
Guerchetan confirmed the flag is there on Intel.

There is something wrong with the 'drm_atomic' USE flag we just added that should not be defined on Intel.

I'll revert https://chromium.googlesource.com/chromiumos/platform2/+/ec0ce0be8b8ba6a82f6a7f06cb21dc91bd372411 temporarily.
Cc: yang...@intel.com hoegsberg@chromium.org kainino@chromium.org marc...@chromium.org xinghua....@intel.com
 Issue 727962  has been merged into this issue.
Status: Assigned (was: Verified)
This bug wasn't solved for all boards. Now we have the same issues, but with other boards. So, for example, there was no problem with daisy, kevin and peach_pi, but now graphics_WebGLAquarium is failing with these boards. 
Status: Verified (was: Assigned)
Thanks, filed  crbug.com/732613  since it looks like a different ARM based regression.

Comment 13 by ihf@chromium.org, Jun 13 2017

I filed  issue 732616  for the new regression.

Comment 14 Deleted

Project Member

Comment 15 by bugdroid1@chromium.org, Jul 11 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/overlays/board-overlays/+/371491d3929a89c48754701dd691454ea0f48da8

commit 371491d3929a89c48754701dd691454ea0f48da8
Author: Daniele Castagna <dcastagna@chromium.org>
Date: Tue Jul 11 07:39:02 2017

Remove unused old 'drm_atomic' flag.

'drm_atomic' USE flag was initially added to all the boards and
was used to include libdrm atomic patches.

With https://chromium-review.googlesource.com/c/302068/ we stopped
checking the flag, but the flag still remeained set for all the boards.

Recently we re-introduced 'drm_atomic' USE flag, to enable atomic
pageflipping, without knowing it was already set on all the boards.
Unexpectedly, we enabled atomic pageflipping on all the boards:
 crbug.com/729160 

This CL removes 'drm_atomic' flag from all the boards since it's
about to have a new meaning and it'll be enabled only on the boards
where atomic pageflipping works.

BUG= chromium:729160 
TEST=Built libchromeos-ui, grep drm_atomic /build/eve/etc/ui_use_flags.txt

Change-Id: I120141307d748ca00c29bbbcd680e370856f512f
Reviewed-on: https://chromium-review.googlesource.com/565635
Commit-Ready: Daniele Castagna <dcastagna@chromium.org>
Tested-by: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: Dan Erat <derat@chromium.org>

[modify] https://crrev.com/371491d3929a89c48754701dd691454ea0f48da8/project-freon/profiles/base/make.defaults

Project Member

Comment 16 by bugdroid1@chromium.org, Jul 13 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromiumos/platform2/+/28945febb98536abcdedf386c27819365ccf3d83

commit 28945febb98536abcdedf386c27819365ccf3d83
Author: Daniele Castagna <dcastagna@chromium.org>
Date: Thu Jul 13 18:44:38 2017

platform2: Use drm_atomic flag instead of special case board.

drm_atomic use flag has been added. We used to check the board to
determine when to enable drm-atomic.

This CL uses the USE flag 'drm_atomic' instead.

BUG= chromium:729160 
TEST=None

Change-Id: I10785a747c47345aa1784504b72823a5b44e90ad
Reviewed-on: https://chromium-review.googlesource.com/567281
Commit-Ready: Daniele Castagna <dcastagna@chromium.org>
Tested-by: Daniele Castagna <dcastagna@chromium.org>
Reviewed-by: Dan Erat <derat@chromium.org>
Reviewed-by: Daniele Castagna <dcastagna@chromium.org>

[modify] https://crrev.com/28945febb98536abcdedf386c27819365ccf3d83/login_manager/chrome_setup.cc
[modify] https://crrev.com/28945febb98536abcdedf386c27819365ccf3d83/libchromeos-ui/chromeos/ui/chromium_command_builder.cc

Sign in to add a comment