[Webkit-unassigned] [Bug 175470] Support the with keyword in DFG

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=175470

Saam Barati <sbarati at apple.com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
 Attachment #318054|review?, commit-queue?      |review+, commit-queue-
              Flags|                            |

--- Comment #10 from Saam Barati <sbarati at apple.com> ---
Comment on attachment 318054
  --> https://bugs.webkit.org/attachment.cgi?id=318054
Patch with test and fixing a bug

View in context: https://bugs.webkit.org/attachment.cgi?id=318054&action=review

r=me still

> Source/JavaScriptCore/dfg/DFGAbstractInterpreterInlines.h:2068
> +        // We cannot use the more precise withScopeStructure() here because it is a LazyProperty and may not yet be allocated.

Style nit: No need for this comment IMO. If you kept, I also wouldn't say "cannot", I'd just say, "we don't", since we could fix how withScopeStructure is allocated during JSGlobalObject creation.

> Source/JavaScriptCore/dfg/DFGAbstractInterpreterInlines.h:2069
> +        forNode(node).setType(m_graph, SpecObject);

You want SpecObjectOther here since that's more precise than SpecObject

> JSTests/stress/with.js:1
> +for (var i = 0; i < 10000; ++i) {

You need to re-run prepare-Changelog so it can modify the Changelog entry inside JSTests/

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20170814/dc521fe1/attachment.html>