<html>
<head>
<meta content="text/html; charset=UTF-8" http-equiv="Content-Type">
</head>
<body text="#000000" bgcolor="#FFFFFF">
Hi Everyone,<br>
<br>
Here's a new webrev for this change:
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~johnc/8015237/webrev.1">http://cr.openjdk.java.net/~johnc/8015237/webrev.1</a><br>
<br>
Changes from before:<br>
* Refactored the code that loops over the buckets into its own
routine.<br>
* Removed the commented out instrumentation (oops).<br>
* Changed the types to int to be consistent with the rest of
symbolTable and allow removal of the casts.<br>
* Increase the number of buckets per claim to 32 based upon a verbal
comment from John Coomes.<br>
* Removed the additional worker ID parameter for the sake of peace
and harmony.<br>
<br>
Testing: jprt.<br>
<br>
Thanks,<br>
<br>
JohnC<br>
<br>
<div class="moz-cite-prefix">On 5/24/2013 3:19 PM, John Cuthbertson
wrote:<br>
</div>
<blockquote cite="mid:519FE787.70408@oracle.com" type="cite">
<meta http-equiv="content-type" content="text/html; charset=UTF-8">
Hi Everyone,<br>
<br>
Can I have a couple of reviewers look over these changes - the
webrev is: <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="http://cr.openjdk.java.net/%7Ejohnc/8015237/webrev.0/">http://cr.openjdk.java.net/~johnc/8015237/webrev.0/</a><br>
<br>
Summary:<br>
On some workloads we are seeing that the scan of the intern string
table (among others) can sometimes take quite a while. This showed
up on some FMW workloads with G1 where the scan of the string
table dominated the pause time for some pauses. G1 was
particularly affected since it doesn't do class unloading (and
hence pruning of the string table) except at full GCs. The
solution was to change the string table from being considered a
single root task and treat similarly to the Java thread stacks:
each GC worker claims a given number of buckets and scans the
entries in those buckets.<br>
<br>
Testing<br>
Kitchensink; jprt; GC test suite. With all collectors.<br>
<br>
Overhead:<br>
Not real performance numbers but I did some measurement of the
synchronization overhead of using 1 GC worker thread. They are
summarized here:<br>
<br>
<table height="104" width="485" border="1" cellpadding="2"
cellspacing="2">
<tbody>
<tr>
<td valign="top"><br>
</td>
<td valign="top">0-threads (ms)<br>
</td>
<td valign="top">1-thread-chunked (ms)<br>
</td>
</tr>
<tr>
<td valign="top">Min</td>
<td valign="top">0.200<br>
</td>
<td valign="top">0.300<br>
</td>
</tr>
<tr>
<td valign="top">Max</td>
<td valign="top">6.900<br>
</td>
<td valign="top">8.800<br>
</td>
</tr>
<tr>
<td valign="top">Avg</td>
<td valign="top">0.658<br>
</td>
<td valign="top">0.794<br>
</td>
</tr>
</tbody>
</table>
<br>
These were from 1 hour long runs of Kitchensink with around ~2800
GCs in each run.<br>
<br>
Thanks,<br>
<br>
JohnC<br>
</blockquote>
<br>
</body>
</html>