<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<font face="Times New Roman, Times, serif">Thomas,<br>
<br>
Change looks good.<br>
<br>
For the sake of future generations maybe a comment such as<br>
</font><br>
<pre> 611 class G1ClearBitMapTask : public AbstractGangTask {
<span class="new"> 612 public:
</span>// A chunk size of 1M was determined empirically to be a reasonable size.
<span class="new">613 static size_t chunk_size() { return M; }</span></pre>
Jon<br>
<br>
<div class="moz-cite-prefix">On 04/28/2016 04:15 AM, Thomas Schatzl
wrote:<br>
</div>
<blockquote cite="mid:1461842149.2484.23.camel@oracle.com"
type="cite">
<pre wrap="">Hi all,
during testing of 8077144 I found that mark bitmap clearing can be
significantly sped up by not using that many threads.
A good conservative estimate (tends to use too many threads) that still
cuts mark bitmap clear time by >50% is to give each thread at least 1
MB of data to clear.
CR:
<a class="moz-txt-link-freetext" href="https://bugs.openjdk.java.net/browse/JDK-8155230">https://bugs.openjdk.java.net/browse/JDK-8155230</a>
Webrev:
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~tschatzl/8155230/webrev/">http://cr.openjdk.java.net/~tschatzl/8155230/webrev/</a>
Testing:
jprt, webrev, various performance tests
Depends on 8155232.
Thanks,
Thomas
</pre>
</blockquote>
<br>
</body>
</html>