New issue
Advanced search Search tips

Issue 670069 link

Starred by 2 users

Issue metadata

Status: Fixed
Owner:
Closed: Dec 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux , Windows
Pri: 2
Type: Bug-Regression



Sign in to add a comment

blink_perf.bindings failure on chromium.perf at 434623 - 434718

Project Member Reported by martiniss@chromium.org, Nov 30 2016

Issue description

Revision range first seen:

434623 - 434718 -- https://uberchromegw.corp.google.com/i/chromium.perf/builders/Linux%20Perf
435082 - 435082 -- https://uberchromegw.corp.google.com/i/chromium.perf/builders/Win%207%20x64%20Perf
434297 - 434311 -- https://uberchromegw.corp.google.com/i/chromium.perf/builders/Mac%2010.10%20Perf
435082 - 435082 -- https://uberchromegw.corp.google.com/i/chromium.perf/builders/Win%207%20Intel%20GPU%20Perf
435082 - 435082 -- https://uberchromegw.corp.google.com/i/chromium.perf/builders/Win%207%20Nvidia%20GPU%20Perf

Unclear what the actual revision range is. Looks like an actual failure though.


Link to failing step log:
https://uberchromegw.corp.google.com/i/chromium.perf/builders/Mac%2010.10%20Perf/builds/112/steps/blink_perf.bindings%20on%20Intel%20GPU%20on%20Mac%20on%20Mac-10.10/logs/stdio
Just one of the failing builders. I think this is the same failure on multiple 

relevant part of step log is this, I believe:

INFO:root:*************** BROWSER STANDARD OUTPUT ***************
INFO:root:2016-11-30 07:12:05.505 Google Chrome[24592:179749] NSWindow warning: adding an unknown subview: <FullSizeContentView: 0x7f8123da08d0>
2016-11-30 07:12:05.506 Google Chrome[24592:179749] Call stack:
(
    "+callStackSymbols disabled for performance reasons"
)

INFO:root:*********** END OF BROWSER STANDARD OUTPUT ************
INFO:root:********************* BROWSER LOG *********************
INFO:root:No log file
INFO:root:***************** END OF BROWSER LOG ******************
FATAL: Got an exception while running test.run with name=RangeError, message=Maximum call stack size exceeded
Traceback (most recent call last):
  File "/b/s/w/irH38iuQ/third_party/catapult/telemetry/telemetry/internal/story_runner.py", line 87, in _RunStoryAndProcessErrorIfNeeded
    state.RunStory(results)
  File "/b/s/w/irH38iuQ/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 "/b/s/w/irH38iuQ/third_party/catapult/telemetry/telemetry/page/shared_page_state.py", line 301, in RunStory
    self._current_page, self._current_tab, results)
  File "/b/s/w/irH38iuQ/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 "/b/s/w/irH38iuQ/tools/perf/benchmarks/blink_perf.py", line 115, in ValidateAndMeasurePage
    results.current_page, metric, units, values))
  File "/b/s/w/irH38iuQ/third_party/catapult/telemetry/telemetry/value/list_of_scalar_values.py", line 83, in __init__
    assert len(values) > 0
AssertionError

If the test is disabled, please downgrade to Pri-2.

cc-ed current test owners.

 
Summary: blink_perf.bindings failure on chromium.perf at 434623 - 434718 (was: failure on chromium.perf at <revisionrange>)
Cc: eyaich@chromium.org
Buildbot blame is missing, because of the conversion to swarming, so I'll have to just disable I think.

cc-ing eyaich@ who did the swarming conversion for FYI
Owner: martiniss@chromium.org
Status: Started (was: Available)
Disabling the tests...
Disabling the tests.
Project Member

Comment 5 by bugdroid1@chromium.org, Dec 1 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/eb080bd2b9753c6956b69a48cdb67fc085bc9d5c

commit eb080bd2b9753c6956b69a48cdb67fc085bc9d5c
Author: martiniss <martiniss@chromium.org>
Date: Thu Dec 01 01:26:24 2016

Disable blink_perf.bindings on all bots

This test is failing on all bots. Disable for now to turn the bots green
again.

BUG= 670069 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq
TBR=sullivan,yukishiino

Review-Url: https://codereview.chromium.org/2543783002
Cr-Commit-Position: refs/heads/master@{#435517}

[modify] https://crrev.com/eb080bd2b9753c6956b69a48cdb67fc085bc9d5c/tools/perf/benchmarks/blink_perf.py

Owner: yukishiino@chromium.org
Status: Assigned (was: Started)
Assigning to test owner to re-enable the test.
Components: Blink>Bindings
Is there progress on this? It'd be nice to have these tests running again.
Unfortunately no progress.  I'm behind of work...  X(
Status: Started (was: Assigned)
Labels: -Pri-1 Pri-2
Project Member

Comment 12 by bugdroid1@chromium.org, Dec 20 2016

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/c344e57998af6ad9ec3778ae510bf3ce635d1614

commit c344e57998af6ad9ec3778ae510bf3ce635d1614
Author: yukishiino <yukishiino@chromium.org>
Date: Tue Dec 20 19:03:36 2016

binding: Re-enables blink_perf.bindings on all platforms.

serialize-nested-array.html in blink_perf.bindings was failing
because of too much of nest level in test code, which was
causing stack overflow in V8.  The failure was triggered with
https://crrev.com/2517813002 .

Discussed with the test owner peria@ and reached to a conclusion
that we don't need such a huge nest level for this test.  Thus,
this CL reduces the nest level of the test.

BUG= 670069 
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.perf:linux_perf_cq;master.tryserver.chromium.perf:mac_retina_perf_cq;master.tryserver.chromium.perf:winx64_10_perf_cq

Review-Url: https://codereview.chromium.org/2553993002
Cr-Commit-Position: refs/heads/master@{#439859}

[modify] https://crrev.com/c344e57998af6ad9ec3778ae510bf3ce635d1614/third_party/WebKit/PerformanceTests/Bindings/serialize-nested-array.html
[modify] https://crrev.com/c344e57998af6ad9ec3778ae510bf3ce635d1614/tools/perf/benchmarks/blink_perf.py

Status: Fixed (was: Started)
Status: Assigned (was: Fixed)
This is still failing on 8 windows bots: 

https://uberchromegw.corp.google.com/i/chromium.perf/waterfall

I think we should disable again until it is actually fixed.
Labels: -Pri-2 Pri-1
Labels: -Pri-1 -OS-Mac Pri-2
Status: Fixed (was: Assigned)
If it's failing on only Windows bots, then I think it should be a dup of  Issue 668164 .  That's a separate issue from this one.

mlippautz@ has already been working on the issue, and the bots look green now.

Sign in to add a comment