[webkit-reviews] review denied: [Bug 213372] [JSC][ESNext] Create a new op_code to handle private fields store/define : [Attachment 404563] Patch

[bugzilla-daemon at webkit.org]

Caio Lima <ticaiolima at gmail.com> has denied Caio Lima <ticaiolima at gmail.com>'s
request for review:
Bug 213372: [JSC][ESNext] Create a new op_code to handle private fields
store/define
https://bugs.webkit.org/show_bug.cgi?id=213372

Attachment 404563: Patch

https://bugs.webkit.org/attachment.cgi?id=404563&action=review




--- Comment #11 from Caio Lima <ticaiolima at gmail.com> ---
Comment on attachment 404563
  --> https://bugs.webkit.org/attachment.cgi?id=404563
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=404563&action=review

I've found a bug

>> Source/JavaScriptCore/dfg/DFGByteCodeParser.cpp:6083
>> +			handlePutById(base, identifier, identifierNumber,
value, putByIdStatus, isDirect, nextOpcodeIndex(), ECMAMode::strict());
> 
> nit: there's no reason we can't do this in the block above where
`compiledAsPutById` is set to true

The way we are compiling here is wrong. In a case where we have a megamorphic
callsite, this will compile as PutByID but slow path won't have the proper
semantics for private fields. A test case for it is also needed.