[webkit-reviews] review denied: [Bug 31050] [Multipatch] JSVALUE32_64 support on ARM : [Attachment 42365] Proposed patch for JSValue32_64 on ARM

[bugzilla-daemon at webkit.org]

Gavin Barraclough <barraclough at apple.com> has denied Zoltan Herczeg
<zherczeg at inf.u-szeged.hu>'s request for review:
Bug 31050: [Multipatch] JSVALUE32_64 support on ARM
https://bugs.webkit.org/show_bug.cgi?id=31050

Attachment 42365: Proposed patch for JSValue32_64 on ARM
https://bugs.webkit.org/attachment.cgi?id=42365&action=review

------- Additional Comments from Gavin Barraclough <barraclough at apple.com>
As per our conversation on IRC, It will help this change to clean up some of
the existing code in svn first, so let's hold this off for:

https://bugs.webkit.org/show_bug.cgi?id=31104

to land first.	I'm going to r- this patch, since these changes will conflict,
please resubmit after 31104 is in & you've updated.

ECMA says that math is IEEE754 compliant, and IEEE754 specifies that
comparisons between unordered operands (i.e. when either operand is NaN)
evaluates to false.  The MacroAssembler interface in ToT is a mess in its
handling of unordered operands, I've addressed this in the patch on bug 31104
by making the required handling of unordered operands explicit in the
DoubleCondition.

r-, will review a new patch post 31104.