[ macOS wk2 ] fast/scrolling/mac/rubberband-overflow-in-wheel-region-root-jiggle.html (layout-test) is a flaky timeout

• Component: New Bugs
• Status: NEW ---

[Robert Jenner]

fast/scrolling/mac/rubberband-overflow-in-wheel-region-root-jiggle.html

is a flaky timeout on macOS Catalina and higher on wk2.

HISTORY:
https://results.webkit.org/?suite=layout-tests&test=fast%2Fscrolling%2Fmac%2Frubberband-overflow-in-wheel-region-root-jiggle.html

This test appears to have been flaky timing out for as far back as I could scroll. 

Timeout DIFF:
+#PID UNRESPONSIVE - WebKitTestRunner (pid 69253)
+FAIL: Timed out waiting for notifyDone to be called
 
-Test that rubberbanding inside overflow:scroll in the non-fast scrollable region does not trigger page scrolls
-Waiting for rubberband...
-Waiting for stabilization
-PASS overflowScrollEventCount > 0 is true
-PASS windowScrollEventCount == 0 is true
-PASS successfullyParsed is true
-
-TEST COMPLETE
-
+#EOF
+#EOF

[Robert Jenner]

I was able to reproduce this at Catalina Debug ToT using the following test:

run-webkit-tests fast/scrolling/mac/rubberband-overflow-in-wheel-region-root-jiggle.html --iterations 100 -f

This test appears to have been flaky for a while. It does look like the first timeout occurred at r271646. So there may be a regression point around there. I'm going to attempt to bisect.

[Robert Jenner]

Created attachment 427946 [details]
WebKitTestRunner sample text

Attaching WebKitTestRunner sample.txt from timeout reproduction.

[Robert Jenner]

I have updated the expectations to Pass Timeout here while the test is reviewed:

https://trac.webkit.org/changeset/277138/webkit


I haven't been able to discover a regression point. I was able to reproduce timeouts over 8000 revisions back.

[Radar WebKit Bug Importer]

<rdar://problem/77631567>