[webkit-reviews] review granted: [Bug 184467] Introduce remote variants of Frame / DOMWindow classes : [Attachment 337738] Patch

[bugzilla-daemon at webkit.org]

Ryosuke Niwa <rniwa at webkit.org> has granted Chris Dumez <cdumez at apple.com>'s
request for review:
Bug 184467: Introduce remote variants of Frame / DOMWindow classes
https://bugs.webkit.org/show_bug.cgi?id=184467

Attachment 337738: Patch

https://bugs.webkit.org/attachment.cgi?id=337738&action=review




--- Comment #13 from Ryosuke Niwa <rniwa at webkit.org> ---
Comment on attachment 337738
  --> https://bugs.webkit.org/attachment.cgi?id=337738
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=337738&action=review

> Source/WebCore/page/DOMWindow.cpp:404
> +    : AbstractDOMWindow(GlobalWindowIdentifier { Process::identifier(),
generateObjectIdentifier<WindowIdentifierType>() })

Process::identifier() grabs a lock... Can we cache the result in a static local
variable for the main thread?

> Source/WebCore/page/GlobalFrameIdentifier.h:32
> +// Frame identifier that is unique across all WebContent processes.

If this is the case, can't we just make DOMWindow identifier generated from
frame ID & some monotonically increasing number?
We can't have a DOMWindow without a frame, right?

> Source/WebCore/page/GlobalWindowIdentifier.h:39
> +    ProcessIdentifier processIdentifier;

Can't we just use pageID here too?