[webkit-reviews] review granted: [Bug 76361] DFG should be able to do JS and custom getter caching : [Attachment 122678] the patch

[bugzilla-daemon at webkit.org]

Geoffrey Garen <ggaren at apple.com> has granted Filip Pizlo <fpizlo at apple.com>'s
request for review:
Bug 76361: DFG should be able to do JS and custom getter caching
https://bugs.webkit.org/show_bug.cgi?id=76361

Attachment 122678: the patch
https://bugs.webkit.org/attachment.cgi?id=122678&action=review

------- Additional Comments from Geoffrey Garen <ggaren at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=122678&action=review


r=me

Qt build not happy.

> Source/JavaScriptCore/dfg/DFGJITCompiler.cpp:131
> +	       // FIXME: this is absurdly wrong.

(1) Yikes.

(2) Why don't you file a bug about what's wrong, and remove the FIXME.

> Source/JavaScriptCore/dfg/DFGOperations.cpp:835
> +    CodeOrigin codeOrigin = stubInfo->codeOrigin;
> +    while (codeOrigin.inlineCallFrame)
> +	   codeOrigin = codeOrigin.inlineCallFrame->caller;

Is this loop also absurdly wrong, as the loop above is?

> Source/JavaScriptCore/dfg/DFGRepatch.cpp:283
> +	   // Need extra checks.

A "why" comment would be better here, explaining that non-flushing gets must be
trivial values, because non-trivial values might call to C++ code and stomp on
our non-flushed registers.