[webkit-reviews] review granted: [Bug 170102] Air::Liveness shouldn't need HashSets : [Attachment 305435] the patch

[bugzilla-daemon at webkit.org]

Yusuke Suzuki <utatane.tea at gmail.com> has granted Filip Pizlo
<fpizlo at apple.com>'s request for review:
Bug 170102: Air::Liveness shouldn't need HashSets
https://bugs.webkit.org/show_bug.cgi?id=170102

Attachment 305435: the patch

https://bugs.webkit.org/attachment.cgi?id=305435&action=review




--- Comment #9 from Yusuke Suzuki <utatane.tea at gmail.com> ---
Comment on attachment 305435
  --> https://bugs.webkit.org/attachment.cgi?id=305435
the patch

View in context: https://bugs.webkit.org/attachment.cgi?id=305435&action=review

r=me

> Source/JavaScriptCore/ChangeLog:11
> +	   compile time progression on WasmBench.

Cool.

> Source/JavaScriptCore/b3/air/AirLiveness.h:133
>	       dirtyBlocks.set(blockIndex);

Cool.

> Source/JavaScriptCore/b3/air/AirLiveness.h:188
> +			   liveAtTail = m_workset.values();

OK, m_workset.values() is already sorted.

> Source/JavaScriptCore/b3/air/AirLiveness.h:202
> +			   mergeBuffer.resize(0);
> +			   mergeBuffer.reserveCapacity(liveAtTail.size() +
m_workset.size());
> +			   auto iter = mergeDeduplicatedSorted(
> +			       liveAtTail.begin(), liveAtTail.end(),
> +			       m_workset.begin(), m_workset.end(),
> +			       mergeBuffer.begin());
> +			   mergeBuffer.resize(iter - mergeBuffer.begin());
> +			   
> +			   if (mergeBuffer.size() == liveAtTail.size())
> +			       continue;
> +		       
> +			   RELEASE_ASSERT(mergeBuffer.size() >
liveAtTail.size());
> +			   liveAtTail = mergeBuffer;

OK, merging 2 sorted sequence into one sorted mergeBuffer.

> Source/JavaScriptCore/b3/air/AirTmp.h:95
> +

Is this used?

> Source/JavaScriptCore/b3/air/AirTmp.h:149
> +    }

Ditto.