New issue
Advanced search Search tips

Issue 667094 link

Starred by 1 user

Issue metadata

Status: Duplicate
Merged: issue 667147
Owner:
Closed: Nov 2016
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Linux
Pri: 1
Type: Bug-Regression

Blocking:
issue 651902



Sign in to add a comment

HarfBuzzShaperTest.ResolveCandidateRunsUnicodeVariants failing on Linux

Project Member Reported by kbr@chromium.org, Nov 20 2016

Issue description

HarfBuzzShaperTest.ResolveCandidateRunsUnicodeVariants from blink_platform_unittests is failing reliably on Linux starting in this build:

https://build.chromium.org/p/chromium.webkit/builders/WebKit%20Linux%20Trusty/builds/20123

with the following failure:

Retrying 1 test (retry #1)
[ RUN      ] HarfBuzzShaperTest.ResolveCandidateRunsUnicodeVariants
../../third_party/WebKit/Source/platform/fonts/shaping/HarfBuzzShaperTest.cpp:95: Failure
Value of: testInfo(result)->glyphForTesting(0, 1)
  Actual: 198
Expected: 3u
Which is: 3
Standard Variants of Ideograph
../../third_party/WebKit/Source/platform/fonts/shaping/HarfBuzzShaperTest.cpp:95: Failure
Value of: testInfo(result)->glyphForTesting(0, 1)
  Actual: 198
Expected: 3u
Which is: 3
Ideographic Variants
[  FAILED  ] HarfBuzzShaperTest.ResolveCandidateRunsUnicodeVariants (7 ms)


Based on the regression range I think this CL was the cause:
https://chromium.googlesource.com/chromium/src/+/aacf8d1d11beaff516bd79f88b8571eae0f5d031

It looks like the build dependencies are incomplete, because https://codereview.chromium.org/2370393002 didn't run blink_platform_unittests in its tryjob for linux_chromium_rel_ng. CC'ing dpranke@ and esprehn@ to help investigate the missing dependency.

This is slowing down a lot of CLs going through the commit queue. Either the test should be fixed or the CL should be reverted.

 
Cc: pdr@chromium.org drott@chromium.org
Interesting, it's not obvious to me that that CL would've caused the failure, but it's not obvious to me what else in that range would've, either, unless it was pdr@'s change.

I'll revert it and see what happens. 

Comment 2 by drott@chromium.org, Nov 21 2016

Was this resolved?

Mergedinto: 667147
Status: Duplicate (was: Assigned)
Not really. The test was disabled but we're unsure of the underlying cause. The revert did not affect anything.

 Bug 667147  was filed later for the same problem, and more work went on there, so I'm going to close this as a duplicate.

Sign in to add a comment