Bug 217543 – [ macOS ] webaudio/OfflineAudioContext/onstatechange.html is a flaky failure

• Component: Web Audio
• Status: RESOLVED FIXED

[Hector Lopez]

webaudio/OfflineAudioContext/onstatechange.html


Test is a flaky failure according to history on macOS. First occurrence coincides with import at r267307

History:

https://results.webkit.org/?suite=layout-tests&test=webaudio%2FOfflineAudioContext%2Fonstatechange.html&platform=mac&limit=50000

Diff:

--- /Volumes/Data/slave/catalina-debug-tests-wk2/build/layout-test-results/webaudio/OfflineAudioContext/onstatechange-expected.txt
+++ /Volumes/Data/slave/catalina-debug-tests-wk2/build/layout-test-results/webaudio/OfflineAudioContext/onstatechange-actual.txt
@@ -3,11 +3,11 @@
 PASS Executing "test"
 PASS Audit report
 PASS > [test] Signaling of statechange event
-PASS   context.state is equal to running.
+FAIL X context.state is not equal to running. Got closed. assert_true: expected true got false
+PASS   Context rendering resolved correctly.
 PASS   context.state is equal to closed.
-PASS   Context rendering resolved correctly.
 PASS   stateChangeCount is equal to 2.
 PASS   After rendering context.state is equal to closed.
-PASS < [test] All assertions passed. (total 5 assertions)
-PASS # AUDIT TASK RUNNER FINISHED: 1 tasks ran successfully.
+FAIL < [test] 1 out of 5 assertions were failed. assert_true: expected true got false
+FAIL # AUDIT TASK RUNNER FINISHED: 1 out of 1 tasks were failed. assert_true: expected true got false

[Radar WebKit Bug Importer]

<rdar://problem/70160122>

[Hector Lopez]

Test expectation while investigated:

https://trac.webkit.org/changeset/268303/webkit

[Chris Dumez]

Created attachment 412960 [details]
Patch

[Darin Adler]

Comment on attachment 412960 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=412960&action=review

> Source/WebCore/Modules/webaudio/OfflineAudioDestinationNode.cpp:105
> +        callOnMainThread([this, result, currentSampleFrame = m_currentSampleFrame, protectedThis = WTFMove(protectedThis)]() mutable {
> +            context().postTask([this, protectedThis = WTFMove(protectedThis), result, currentSampleFrame]() mutable {

Doesn’t seem like we need callOnMainThread if calling postTask. It already does callOnMainThread.

[Darin Adler]

Oops, didn’t mean to change this to darin/review?

Don’t know why bugs.webkit.org does that.

[Chris Dumez]

(In reply to Darin Adler from comment #4)
> Comment on attachment 412960 [details]
> Patch
> 
> View in context:
> https://bugs.webkit.org/attachment.cgi?id=412960&action=review
> 
> > Source/WebCore/Modules/webaudio/OfflineAudioDestinationNode.cpp:105
> > +        callOnMainThread([this, result, currentSampleFrame = m_currentSampleFrame, protectedThis = WTFMove(protectedThis)]() mutable {
> > +            context().postTask([this, protectedThis = WTFMove(protectedThis), result, currentSampleFrame]() mutable {
> 
> Doesn’t seem like we need callOnMainThread if calling postTask. It already
> does callOnMainThread.

I think you are looking at the wrong postTask?

void BaseAudioContext::postTask(WTF::Function<void()>&& task)
{
    ASSERT(isMainThread());
    if (m_isStopScheduled)
        return;

    queueTaskKeepingObjectAlive(*this, TaskSource::MediaElement, WTFMove(task));
}

[Darin Adler]

OK.

[Darin Adler]

Alex said review+. Alex, want to do it again so bugs.webkit.org has the record?

[Chris Dumez]

Created attachment 412961 [details]
Patch

[EWS]

Committed r269271: <https://trac.webkit.org/changeset/269271>

All reviewed patches have been landed. Closing bug and clearing flags on attachment 412961 [details].