Bug 192619 – [BigInt] Implement DFG/FTL typeof for BigInt

• Status: RESOLVED
• Resolution: FIXED
• Component: New Bugs

[Yusuke Suzuki]

[BigInt] Implement DFG/FTL typeof for BigInt

[Yusuke Suzuki]

Created attachment 357114 [details]
Patch

[Caio Lima]

Comment on attachment 357114 [details]
Patch

LGTM, informal r+

[Keith Miller]

Comment on attachment 357114 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=357114&action=review

r=me.

> Source/JavaScriptCore/ftl/FTLLowerDFGToB3.cpp:14220
> +        //     } else if (is bigint) {

Do we think that bigints are going to be more or less common than symbol? 

Also, when we add a value representation for "small big ints" I would expect symbols would be more common. Can we add a FIXME: here to switch the ordering when that happens?

> Source/JavaScriptCore/jit/AssemblyHelpers.h:1658
> +        //     } else if (is bigint) {
> +        //         return bigint

Ditto

[Yusuke Suzuki]

Comment on attachment 357114 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=357114&action=review

Thank you!

>> Source/JavaScriptCore/ftl/FTLLowerDFGToB3.cpp:14220
>> +        //     } else if (is bigint) {
> 
> Do we think that bigints are going to be more or less common than symbol? 
> 
> Also, when we add a value representation for "small big ints" I would expect symbols would be more common. Can we add a FIXME: here to switch the ordering when that happens?

Make sense. Added FIXME here.

>> Source/JavaScriptCore/jit/AssemblyHelpers.h:1658
>> +        //         return bigint
> 
> Ditto

Fixed.

[Yusuke Suzuki]

Committed r239141: <https://trac.webkit.org/changeset/239141>

[Radar WebKit Bug Importer]

<rdar://problem/46683130>