Bug 168239 – REGRESSION (r212197): [ios-simulator] LayoutTest fast/css/target-fragment-match.html is failing

• Status: REOPENED
• Component: New Bugs

[Ryan Haddad]

LayoutTest fast/css/target-fragment-match.html is failing

https://build.webkit.org/results/Apple%20iOS%2010%20Simulator%20Debug%20WK2%20(Tests)/r212197%20(3428)/results.html

https://webkit-test-results.webkit.org/dashboards/flakiness_dashboard.html#showAllRuns=true&tests=fast%2Fcss%2Ftarget-fragment-match.html

--- /Volumes/Data/slave/ios-simulator-10-debug-tests-wk2/build/layout-test-results/fast/css/target-fragment-match-expected.txt
+++ /Volumes/Data/slave/ios-simulator-10-debug-tests-wk2/build/layout-test-results/fast/css/target-fragment-match-actual.txt
@@ -7,3 +7,4 @@
         RenderBlock {P} at (0,0) size 784x20
           RenderText {#text} at (0,0) size 624x19
             text run at (0,0) width 624: "I should be highlighted first because of the anchor, and de-highlighted when there is no fragment."
+scrolled to 0,16

[Ryan Haddad]

Seems to have started with https://trac.webkit.org/changeset/212197

[Simon Fraser (smfr)]

Why didn't EWS show this?

[Ryan Haddad]

(In reply to comment #2)
> Why didn't EWS show this?

This was missed because ios-sim EWS runs release, not debug.

[Ryan Haddad]

If this can't be fixed quickly, can we roll out the change?

[Simon Fraser (smfr)]

The change simply revealed the fact that scroll position restoration on iOS is async, which was being masked before.

[Simon Fraser (smfr)]

Attempt to de-flake in https://trac.webkit.org/r212478

[Ryan Haddad]

It looks like the test now passes on iOS simulator debug, where it was failing on almost every run.

However, it does look like the test has failed once on release since r212478 was landed:

https://build.webkit.org/builders/Apple%20iOS%2010%20Simulator%20Release%20WK2%20(Tests)/builds/3870

[Ryan Haddad]

This test seems to be failing on ios-sim EWS and causing false positives.

[Ryan Haddad]

(In reply to Ryan Haddad from comment #8)
> This test seems to be failing on ios-sim EWS and causing false positives.

See https://bugs.webkit.org/buglist.cgi?quicksearch=fast%2Fcss%2Ftarget-fragment-match.html&list_id=2771578