[webkit-changes] [51457] trunk/JavaScriptCore

[Sam Weinig]

Can these ASSERTS be STATIC_ASSERTS?  Did you test the performance?   
The comment indicates the malloc was needed but there is no  
information in the changelog to indicate that is no longer true.

- Sam

On Nov 28, 2009, at 5:31 AM, zoltan at webkit.org wrote:

> Revision
> 51457
> Author
> zoltan at webkit.org
> Date
> 2009-11-28 02:31:18 -0800 (Sat, 28 Nov 2009)
> Log Message
>
> 2009-11-28  Zoltan Herczeg  <zherczeg at inf.u-szeged.hu>
>
>         Reviewed by Gavin Barraclough.
>
>         https://bugs.webkit.org/show_bug.cgi?id=31930
>
>         Seems a typo. We don't need ~270k memory to determine the  
> vptrs.
>
>         * runtime/JSGlobalData.cpp:
>         (JSC::VPtrSet::VPtrSet):
> Modified Paths
>
> trunk/JavaScriptCore/ChangeLog
> trunk/JavaScriptCore/runtime/JSGlobalData.cpp
> Diff
>
> Modified: trunk/JavaScriptCore/ChangeLog (51456 => 51457)
>
> --- trunk/JavaScriptCore/ChangeLog	2009-11-28 07:39:12 UTC (rev 51456)
> +++ trunk/JavaScriptCore/ChangeLog	2009-11-28 10:31:18 UTC (rev 51457)
> @@ -1,3 +1,14 @@
> +2009-11-28  Zoltan Herczeg  <zherczeg at inf.u-szeged.hu>
> +
> +        Reviewed by Gavin Barraclough.
> +
> +        https://bugs.webkit.org/show_bug.cgi?id=31930
> +
> +        Seems a typo. We don't need ~270k memory to determine the  
> vptrs.
> +
> +        * runtime/JSGlobalData.cpp:
> +        (JSC::VPtrSet::VPtrSet):
> +
>  2009-11-27  Shinichiro Hamaji  <hamaji at chromium.org>
>
>          Unreviewed.
> Modified: trunk/JavaScriptCore/runtime/JSGlobalData.cpp (51456 =>  
> 51457)
>
> --- trunk/JavaScriptCore/runtime/JSGlobalData.cpp	2009-11-28  
> 07:39:12 UTC (rev 51456)
> +++ trunk/JavaScriptCore/runtime/JSGlobalData.cpp	2009-11-28  
> 10:31:18 UTC (rev 51457)
> @@ -90,26 +90,28 @@
>
>  VPtrSet::VPtrSet()
>  {
> -    // Bizarrely, calling fastMalloc here is faster than allocating  
> space on the stack.
> -    void* storage = fastMalloc(sizeof(CollectorBlock));
> +    CollectorCell cell;
> +    void* storage = &cell;
>
> +    ASSERT(sizeof(JSArray) <= sizeof(CollectorCell));
>      JSCell* jsArray = new (storage) JSArray(JSArray::createStructure 
> (jsNull()));
>      jsArrayVPtr = jsArray->vptr();
>      jsArray->~JSCell();
>
> +    ASSERT(sizeof(JSByteArray) <= sizeof(CollectorCell));
>      JSCell* jsByteArray = new (storage) JSByteArray 
> (JSByteArray::VPtrStealingHack);
>      jsByteArrayVPtr = jsByteArray->vptr();
>      jsByteArray->~JSCell();
>
> +    ASSERT(sizeof(JSString) <= sizeof(CollectorCell));
>      JSCell* jsString = new (storage) JSString 
> (JSString::VPtrStealingHack);
>      jsStringVPtr = jsString->vptr();
>      jsString->~JSCell();
>
> +    ASSERT(sizeof(JSFunction) <= sizeof(CollectorCell));
>      JSCell* jsFunction = new (storage) JSFunction 
> (JSFunction::createStructure(jsNull()));
>      jsFunctionVPtr = jsFunction->vptr();
>      jsFunction->~JSCell();
> -
> -    fastFree(storage);
>  }
>
>  JSGlobalData::JSGlobalData(bool isShared, const VPtrSet& vptrSet)
> _______________________________________________
> webkit-changes mailing list
> webkit-changes at lists.webkit.org
> http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-changes/attachments/20091128/a43ceb56/attachment.html>