[webkit-reviews] review granted: [Bug 170494] REGRESSION r214905 added an invalid assertion : [Attachment 306268] patch

[bugzilla-daemon at webkit.org]

Yusuke Suzuki <utatane.tea at gmail.com> has granted Saam Barati
<sbarati at apple.com>'s request for review:
Bug 170494: REGRESSION r214905 added an invalid assertion
https://bugs.webkit.org/show_bug.cgi?id=170494

Attachment 306268: patch

https://bugs.webkit.org/attachment.cgi?id=306268&action=review




--- Comment #4 from Yusuke Suzuki <utatane.tea at gmail.com> ---
Comment on attachment 306268
  --> https://bugs.webkit.org/attachment.cgi?id=306268
patch

View in context: https://bugs.webkit.org/attachment.cgi?id=306268&action=review

This patch looks good. But I have a question about Wasm::Callee.

> Source/JavaScriptCore/ChangeLog:20
> +	   the function a scarier name.

I think we still have a room to fill a pointer to JSValue's NaN boxing bit
patterns.
Since Wasm::Callee is not gc-managed, dirty pointer in JSValue is OK without
drastically changing the GC marking.
Is there any problem when assigning a new tag to Wasm::Callee in JSValue? For
example, we can choose the tag `0b11` (TagBitTypeOther | true).