<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - Unite op_get_by_id_with_this with op_get_by_id"
href="https://bugs.webkit.org/show_bug.cgi?id=162124#c10">Comment # 10</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - Unite op_get_by_id_with_this with op_get_by_id"
href="https://bugs.webkit.org/show_bug.cgi?id=162124">bug 162124</a>
from <span class="vcard"><a class="email" href="mailto:ticaiolima@gmail.com" title="Caio Lima <ticaiolima@gmail.com>"> <span class="fn">Caio Lima</span></a>
</span></b>
<pre>Created <span class=""><a href="attachment.cgi?id=289331" name="attach_289331" title="Benchmarks results">attachment 289331</a> <a href="attachment.cgi?id=289331&action=edit" title="Benchmarks results">[details]</a></span>
Benchmarks results
It seems that the current Patch keeps benchmarks neutral and over ~20% improvements in super-get-by-id micro benchmarks:
super-get-by-id-with-this-monomorphic 36.5942+-0.7969 ^ 29.8105+-0.3676 ^ definitely 1.2276x faster
super-get-by-id-with-this-polymorphic 35.1846+-2.1380 ^ 28.1864+-0.6060 ^ definitely 1.2483x faster
General results:
Geomean of preferred means:
<scaled-result> 27.6482+-0.1572 27.5743+-0.1447 might be 1.0027x faster
I suspect we can improve it a little more in compilation phase, since it can be possible remove if(m_base->isSuperNode()).</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>