[webkit-reviews] review granted: [Bug 90924] DFG should have fast virtual calls : [Attachment 151636] the patch

[bugzilla-daemon at webkit.org]

Gavin Barraclough <barraclough at apple.com> has granted Filip Pizlo
<fpizlo at apple.com>'s request for review:
Bug 90924: DFG should have fast virtual calls
https://bugs.webkit.org/show_bug.cgi?id=90924

Attachment 151636: the patch
https://bugs.webkit.org/attachment.cgi?id=151636&action=review

------- Additional Comments from Gavin Barraclough <barraclough at apple.com>
View in context: https://bugs.webkit.org/attachment.cgi?id=151636&action=review


Please revert JSFunction.h & define up the value (1<<29)-1.  Other comments
optional.

> Source/JavaScriptCore/bytecode/CodeOrigin.h:51
> +	   : bytecodeIndex((1 << 29) - 1)

You should really define up a constant for this value, & comment why it is this
value.

> Source/JavaScriptCore/dfg/DFGSpeculativeJIT32_64.cpp:1029
> +    if (calleeTagGPR == GPRInfo::nonArgGPR0) {

Do we not have a helper for this kind of shuffling? - if not, we probably
really should...

> Source/JavaScriptCore/dfg/DFGThunks.cpp:109
> +    // need to remove it and drop it on the floor, since we don't care about
it.

Is it possible to write a test case for this? - when would the C code throw, an
unexpected OOM? - that could be tricky to test for if so... :-/

> Source/JavaScriptCore/runtime/JSFunction.h:157
> +	   

PLease revert this file.