Bug 162517 - LayoutTest imported/w3c/web-platform-tests/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-support.htm is a flaky timeout
Summary: LayoutTest imported/w3c/web-platform-tests/html/the-xhtml-syntax/parsing-xhtm...
Status: RESOLVED FIXED
Alias: None
Product: WebKit
Classification: Unclassified
Component: New Bugs (show other bugs)
Version: WebKit Nightly Build
Hardware: Unspecified Unspecified
: P2 Normal
Assignee: Chris Dumez
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2016-09-23 16:00 PDT by Ryan Haddad
Modified: 2016-09-25 21:00 PDT (History)
3 users (show)

See Also:


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Ryan Haddad 2016-09-23 16:00:20 PDT
LayoutTest imported/w3c/web-platform-tests/html/the-xhtml-syntax/parsing-xhtml-documents/xhtml-mathml-dtd-entity-support.htm is a flaky timeout

This test frequently times out on  Yosemite Debug WK2 and iOS 10 Simulator Debug WK1

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=imported%2Fw3c%2Fweb-platform-tests%2Fhtml%2Fthe-xhtml-syntax%2Fparsing-xhtml-documents%2Fxhtml-mathml-dtd-entity-support.htm
Comment 1 Ryan Haddad 2016-09-23 16:00:51 PDT
Test was added with https://trac.webkit.org/changeset/206118
Comment 2 Alexey Proskuryakov 2016-09-25 14:44:13 PDT
Chris, could you please take a look? The test is super slow on some of debug bots, running in ~30 seconds even when not timing out.

Strangely, it looks like not all of them are affected - Mac only fails with WK2, but iOS and Windows fail with WK1.
Comment 3 Chris Dumez 2016-09-25 15:48:15 PDT
(In reply to comment #2)
> Chris, could you please take a look? The test is super slow on some of debug
> bots, running in ~30 seconds even when not timing out.
> 
> Strangely, it looks like not all of them are affected - Mac only fails with
> WK2, but iOS and Windows fail with WK1.

Will do tomorrow.
Comment 4 Chris Dumez 2016-09-25 20:56:20 PDT
It seems this test is indeed slow (it tests a lot of things) and generates 21358 lines out of output. Would it be OK to skip it on Debug builds like we do already for the slower web platform tests?
Comment 5 Chris Dumez 2016-09-25 21:00:28 PDT
(In reply to comment #4)
> It seems this test is indeed slow (it tests a lot of things) and generates
> 21358 lines out of output. Would it be OK to skip it on Debug builds like we
> do already for the slower web platform tests?

Done so in <http://trac.webkit.org/changeset/206366>.