Bug 225667 – [ Catalina+ wk2 ] http/tests/webAPIStatistics/screen-functions-accessed-data-collection.html (layout-test) is a flaky timeout

• Status: NEW
• Component: New Bugs

[Robert Jenner]

http/tests/webAPIStatistics/screen-functions-accessed-data-collection.html

is a flaky timeout on BigSur wk2 on Apple Silicon Macs only. 

HISTORY:
https://results.webkit.org/?suite=layout-tests&test=http%2Ftests%2FwebAPIStatistics%2Fscreen-functions-accessed-data-collection.html

TIMEOUT TEXT:
-Tests for screen functions accessed data collection in ResourceLoadStatistics plist by querying for all the screen properties and dumping the entire resource load statistics map.
+#PID UNRESPONSIVE - WebKitTestRunner (pid 28959)
+FAIL: Timed out waiting for notifyDone to be called
+
 Resource load statistics:
 
 Registrable domain: 127.0.0.1
@@ -9,3 +11,5 @@
     isPrevalentResource: No
     isVeryPrevalentResource: No
     dataRecordsRemoved: 0
+#EOF
+#EOF

[Robert Jenner]

After looking at the history a little more thoroughly, I have determined that this is also timing out on Catalina wk2 Release on Intel. Though, the timeouts are substantially more frequent on arm64 platforms. Timeouts do not appear to have occurred on Intel Macs running Big Sur. I will attempt reproduction on my Catalina Mac, but cannot reproduce on Apple Silicon as I don't have access to said system type.

[Radar WebKit Bug Importer]

<rdar://problem/77856848>

[Robert Jenner]

With a few other tests that were flaky timeouts only on arm64, there is a theory that it may be bot configuration related, and not Apple Silicon related. While this test has had a history of timing out on other configurations, it has occurred most frequently on BigSur wk2 Release arm64. 

For this, we are going to mark this test as slow, and see if conditions improve. I did run the test on my local Intel Mac, and it did take 18 seconds for the test to run. So this may just be a slow test. 

Expectation updated to Slow here:
https://trac.webkit.org/changeset/277405/webkit

[Robert Jenner]

It appears I broke this for wk1 by accidentally putting the expectation under all of Mac, instead of just Mac wk2. I have fixed my error here:

https://trac.webkit.org/changeset/277441/webkit

It looks like the test is still timing out (Very flaky).