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

Issue 745881 link

Starred by 1 user

Issue metadata

Status: Fixed
Owner:
Closed: Mar 2018
Cc:
Components:
EstimatedDays: ----
NextAction: ----
OS: Windows
Pri: 2
Type: Bug



Sign in to add a comment

fast/text/ellipsis-in-relative-inline.html failing on Site Isolation Win FYI bot

Project Member Reported by alex...@chromium.org, Jul 18 2017

Issue description

First failure: https://build.chromium.org/p/chromium.fyi/builders/Site%20Isolation%20Win/builds/19894

The failure appears to be caused by an image diff due to a slight difference in how the letter "C" is rendered: https://storage.googleapis.com/chromium-layout-test-archives/Site_Isolation_Win/19894/layout-test-results/fast/text/ellipsis-in-relative-inline-diff.png

It seems the culprit CL is https://chromium-review.googlesource.com/c/563625/.

I'm not sure yet why --site-per-process would affect the rendering here.  It's also possible there's something special about our bot config, and the test could be failing on that config without --site-per-process as well.

robhogan@ or eae@: do you know what such an image diff means?

Also adding lukasza@ who has recently looked at similar pixel diff issues in  issue 740583 , though there aren't any OOPIFs or even frames in this test.

Repro steps:
run-webkit-tests -t gn fast/text/ellipsis-in-relative-inline.html --no-retry --additional-drt-flag=--site-per-process --time-out-ms=30000
 

Comment 1 by e...@chromium.org, Jul 18 2017

Cc: drott@chromium.org
I'd guess that this has more to do with  issue 740583  and 657646 than the change in question.

My guess would be that we reuse a cached resource that has a slightly different alignment.
Owner: d...@chromium.org
Status: Assigned (was: Available)
Thanks for the pointer to issue 657646.  dba@: would you be able to check if your fix from https://crbug.com/657646#c83 would help our bot, which is using vm976-m1?  And/or, perhaps it needs to be reimaged?


Project Member

Comment 3 by bugdroid1@chromium.org, Jul 18 2017

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/c0b70166181fab966daa8497273c9df40648b0e9

commit c0b70166181fab966daa8497273c9df40648b0e9
Author: Alex Moshchuk <alexmos@chromium.org>
Date: Tue Jul 18 23:22:11 2017

Disable fast/text/ellipsis-in-relative-inline.html on Site Isolation Win

Bug:  745881 
Change-Id: I0e36c92b66e2715081589206e58d9a669c712d2a
Reviewed-on: https://chromium-review.googlesource.com/575258
Reviewed-by: Lukasz Anforowicz <lukasza@chromium.org>
Commit-Queue: Alex Moshchuk <alexmos@chromium.org>
Cr-Commit-Position: refs/heads/master@{#487662}
[modify] https://crrev.com/c0b70166181fab966daa8497273c9df40648b0e9/third_party/WebKit/LayoutTests/FlagExpectations/site-per-process

Comment 4 by d...@chromium.org, Jul 21 2017

Components: Infra>Labs
Sorry, didnt notice this assigned to me. In the future you can simply just drop requests like this in Infra>Labs queue for the person on-duty to pick up.

I believe I've fixed the settings on vm976-m1.
Project Member

Comment 5 by bugdroid1@chromium.org, Mar 5 2018

The following revision refers to this bug:
  https://chromium.googlesource.com/chromium/src.git/+/16295fe01a1264e488dc0f7837f53e6f01e74743

commit 16295fe01a1264e488dc0f7837f53e6f01e74743
Author: Lukasz Anforowicz <lukasza@chromium.org>
Date: Mon Mar 05 22:18:14 2018

Remove test exceptions for tests that have "healed" themselves

Bug: 477150,  789781 ,  769508 ,  661725 
Bug: 611232,  745881 ,  801992 
Change-Id: If90fb04e97513b99716afd844a2a77ca0905ab3d
Reviewed-on: https://chromium-review.googlesource.com/942316
Reviewed-by: Alex Moshchuk <alexmos@chromium.org>
Reviewed-by: Charlie Reis <creis@chromium.org>
Commit-Queue: Ɓukasz Anforowicz <lukasza@chromium.org>
Cr-Commit-Position: refs/heads/master@{#540958}
[modify] https://crrev.com/16295fe01a1264e488dc0f7837f53e6f01e74743/third_party/WebKit/LayoutTests/FlagExpectations/site-per-process

Status: Fixed (was: Assigned)
I think there is no other action that needs to be taken here - let's just mark this bug as fixed.

Sign in to add a comment