[webkit-reviews] review denied: [Bug 82660] Frequent crashes in JSC::parse in a worker thread when running regression tests : [Attachment 135202] Updated patch

bugzilla-daemon at webkit.org bugzilla-daemon at webkit.org
Mon Apr 2 19:06:45 PDT 2012


Geoffrey Garen <ggaren at apple.com> has denied Michael Saboff
<msaboff at apple.com>'s request for review:
Bug 82660: Frequent crashes in JSC::parse in a worker thread when running
regression tests
https://bugs.webkit.org/show_bug.cgi?id=82660

Attachment 135202: Updated patch
https://bugs.webkit.org/attachment.cgi?id=135202&action=review

------- Additional Comments from Geoffrey Garen <ggaren at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=135202&action=review


>>> Source/WebCore/workers/WorkerThread.cpp:153
>>> +	 if (!m_startupData->m_sourceCode.isNull())
>>> +	     script->evaluate(ScriptSourceCode(m_startupData->m_sourceCode,
m_startupData->m_scriptURL));
>> 
>> With this, what will show up in the Inspector? My concern is it would show
something fake starting for the current context instead of ignoring the
previous context.
> 
> With this, what will show up in the Inspector? My concern is it would show
something fake starting for the current context instead of ignoring the
previous context.

A better bottleneck for the NULL check is Worker::notifyFinished(). We should
never start a worker thread with a NULL script. Once the worker has started, it
should be able to assume that it has valid JavaScript to execute.

Do you know why we end up with a NULL script and yet Worker::m_scriptLoader
doesn't report that it failed to load?


More information about the webkit-reviews mailing list