[webkit-reviews] review granted: [Bug 69181] [chromium] Add WebWidget API for accessing the current WebCompositor : [Attachment 109364] Patch

[bugzilla-daemon at webkit.org]

Darin Fisher (:fishd, Google) <fishd at chromium.org> has granted James Robinson
<jamesr at chromium.org>'s request for review:
Bug 69181: [chromium] Add WebWidget API for accessing the current WebCompositor
https://bugs.webkit.org/show_bug.cgi?id=69181

Attachment 109364: Patch
https://bugs.webkit.org/attachment.cgi?id=109364&action=review

------- Additional Comments from Darin Fisher (:fishd, Google)
<fishd at chromium.org>
View in context: https://bugs.webkit.org/attachment.cgi?id=109364&action=review


> Source/WebKit/chromium/public/WebWidget.h:176
> +    // The WebCompositor may only be accessed and is destroyed from the
compositor thread.

nit: This almost sounds like it is saying that you need to call this
WebWidget::compositor()
method on the compositor thread.  Instead, maybe tweak the text slightly like
so:

  // The WebCompositor's methods may only be called on the compositor thread.

I'm concerned there may still be a race-condition at setup time.  Could the
WebCompositor
get destroyed, on the compositor thread, before the consumer has a chance to
set the
WebCompositorClient pointer on the WebCompositor?

Hmm...