Bug 111897 – Directly plant statically identifiable global resolves

• Component: New Bugs
• Status: NEW

[Oliver Hunt]

Directly plant statically identifiable global resolves

[Oliver Hunt]

Created attachment 192302 [details]
Patch

[Filip Pizlo]

Comment on attachment 192302 [details]
Patch

View in context: https://bugs.webkit.org/attachment.cgi?id=192302&action=review

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:1855
> +    typedef HashMap<RefPtr<StringImpl>, ResolveOperation*, IdentifierRepHash> GlobalResolveMap;
> +    GlobalResolveMap globalResolves;

Can we make it per bytecode so it's nicer to profiler?

> Source/JavaScriptCore/bytecode/CodeBlock.cpp:1988
> +            if (!entry.isNull()) {
> +                instructions[i + 0] = globalData()->interpreter->getOpcode(op_get_global_var);
> +                instructions[i + 3] = &globalObject->registerAt(entry.getIndex());

I can has early break?

[Oliver Hunt]

Created attachment 192305 [details]
Patch

[Build Bot]

Comment on attachment 192305 [details]
Patch

Attachment 192305 [details] did not pass mac-ews (mac):
Output: http://webkit-commit-queue.appspot.com/results/17132034

New failing tests:
fast/regions/named-flow-content-order-2.html
css3/filters/custom/custom-filter-u-tile-size.html
css3/filters/custom/custom-filter-array.html
css3/filters/custom/custom-filter-u-texture-size.html
svg/custom/textPath-insert-path-pattern.svg
css3/filters/custom/custom-filter-u-mesh-box.html
fast/regions/autoheight-dynamic-update.html
fast/regions/listmarker-inside-flowthread.html
fast/js/eval-cache-crash.html
css3/filters/custom/custom-filter-u-mesh-size.html
fast/js/global-resolve-through-eval.html

[Oliver Hunt]

Comment on attachment 192305 [details]
Patch

Causing bizarro crashes, i believe i know the underlying cause but need to verify