<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=utf-8">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Review request!<br>
<br>
This is a change Ramki did before he left that didn't get checked
in. The basic idea is to move a bunch fields that encapsulate
collector state from
<meta http-equiv="content-type" content="text/html; charset=utf-8">
G1CollectorPolicy into a separate class "
<meta http-equiv="content-type" content="text/html; charset=utf-8">
G1CollectorState".<br>
<br>
Thanks in advance!<br>
<br>
- Derek<br>
<br>
<i>Bug</i>: <a class="moz-txt-link-freetext" href="https://bugs.openjdk.java.net/browse/JDK-7097567">https://bugs.openjdk.java.net/browse/JDK-7097567</a><br>
<i>Webrev</i>:
<a class="moz-txt-link-freetext" href="http://cr.openjdk.java.net/~drwhite/7097567/webrev.00/">http://cr.openjdk.java.net/~drwhite/7097567/webrev.00/</a><br>
<i>Ran</i>:<br>
<ul>
<li>jtreg</li>
<li>jprt</li>
<li>refworkload (looks good, but improvement is unexplained).</li>
</ul>
<blockquote><tt>./compare -r Logs/ref.1/ Logs/cms.1/</tt><tt><br>
</tt><tt>==============================================================================</tt><tt><br>
</tt><tt>Logs/ref.1/:</tt><tt><br>
</tt><tt> Benchmark Samples Mean
Stdev Geomean Weight</tt><tt><br>
</tt><tt> specjbb2000 15 660639.27
8546.73 </tt><tt><br>
</tt><tt> First_Warehouse 15 91712.47
2908.19 </tt><tt><br>
</tt><tt> Last_Warehouse 15 660639.27
8546.73 </tt><tt><br>
</tt><tt> rw.runtime 15 604.00
0.00 </tt><tt><br>
</tt><tt> specjbb2005 15 372212.33
8495.85 </tt><tt><br>
</tt><tt> first 15 44054.38
2190.33 </tt><tt><br>
</tt><tt> interval_average 15 7596.13
173.44 </tt><tt><br>
</tt><tt> last 15 372212.33
8495.85 </tt><tt><br>
</tt><tt> last_warehouse 15 8.00
0.00 </tt><tt><br>
</tt><tt> overall_average 15 342311.21
27049.43 </tt><tt><br>
</tt><tt> peak 15 418062.71
21056.23 </tt><tt><br>
</tt><tt> peak_warehouse 15 2.87
1.25 </tt><tt><br>
</tt><tt> rw.runtime 15 458.40
0.99 </tt><tt><br>
</tt><tt> specjvm98 15 950.96
10.07 </tt><tt><br>
</tt><tt> compress 15 695.97
7.54 </tt><tt><br>
</tt><tt> db 15 361.98
6.56 </tt><tt><br>
</tt><tt> jack 15 1461.28
51.26 </tt><tt><br>
</tt><tt> javac 15 522.44
13.72 </tt><tt><br>
</tt><tt> jess 15 1145.71
15.97 </tt><tt><br>
</tt><tt> mpegaudio 15 1197.56
14.91 </tt><tt><br>
</tt><tt> mtrt 15 2670.94
137.46 </tt><tt><br>
</tt><tt> rw.runtime 15 36.07
0.59 </tt><tt><br>
</tt><tt>==============================================================================</tt><tt><br>
</tt><tt>Logs/cms.1/:</tt><tt><br>
</tt><tt> Benchmark Samples Mean Stdev
%Diff P Significant</tt><tt><br>
</tt><tt> specjbb2000 15 646579.89 5829.25
-2.13 0.000 Yes</tt><tt><br>
</tt><tt> First_Warehouse 15 86084.97 1381.09
-6.14 0.000 Yes</tt><tt><br>
</tt><tt> Last_Warehouse 15 646579.89 5829.25
-2.13 0.000 Yes</tt><tt><br>
</tt><tt> rw.runtime 15 604.33 0.49
-0.06 0.019 *</tt><tt><br>
</tt><tt> specjbb2005 15 372333.48 8449.27
0.03 0.969 *</tt><tt><br>
</tt><tt> first 15 44028.81 591.86
-0.06 0.966 *</tt><tt><br>
</tt><tt> interval_average 15 7598.67 172.40
0.03 0.968 *</tt><tt><br>
</tt><tt> last 15 372333.48 8449.27
0.03 0.969 *</tt><tt><br>
</tt><tt> last_warehouse 15 8.00 0.00
0.00 1.000 *</tt><tt><br>
</tt><tt> overall_average 15 347967.16 8974.17
1.65 0.453 *</tt><tt><br>
</tt><tt> peak 15 427459.90 13498.58
2.25 0.159 *</tt><tt><br>
</tt><tt> peak_warehouse 15 2.27 0.46
20.93 0.097 *</tt><tt><br>
</tt><tt> rw.runtime 15 458.47 0.52
-0.01 0.819 *</tt><tt><br>
</tt><tt> specjvm98 15 891.61 9.95
-6.24 0.000 Yes</tt><tt><br>
</tt><tt> compress 15 696.34 6.57
0.05 0.888 *</tt><tt><br>
</tt><tt> db 15 363.90 5.55
0.53 0.395 *</tt><tt><br>
</tt><tt> jack 15 1402.76 48.98
-4.01 0.003 Yes</tt><tt><br>
</tt><tt> javac 15 453.21 8.57
-13.25 0.000 Yes</tt><tt><br>
</tt><tt> jess 15 1024.87 7.81
-10.55 0.000 Yes</tt><tt><br>
</tt><tt> mpegaudio 15 1146.54 37.59
-4.26 0.000 Yes</tt><tt><br>
</tt><tt> mtrt 15 2369.88 70.39
-11.27 0.000 Yes</tt><tt><br>
</tt><tt> rw.runtime 15 38.00 0.38
-5.36 0.000 Yes</tt><tt><br>
</tt><tt>==============================================================================</tt><tt><br>
</tt><tt> * - Not Significant: A non-zero %Diff for the mean
could be noise. If the</tt><tt><br>
</tt><tt> %Diff is 0, an actual difference may still exist.
In either case, more</tt><tt><br>
</tt><tt> samples would be needed to detect an actual
difference in sample means.</tt><tt><br>
</tt><tt> Alpha for this run: 0.010</tt><tt><br>
</tt></blockquote>
<br>
</body>
</html>