<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
Michail,<br>
<br>
thanks for explanation. <br>
Your fix looks good to me now.<br>
<br>
-- Dima<br>
<br>
<div class="moz-cite-prefix">On 11/10/2016 2:25 PM, Michail Chernov
wrote:<br>
</div>
<blockquote
cite="mid:8b32dc3d-f1fe-8e13-c5bf-5499d5ef8b47@oracle.com"
type="cite">
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
<font size="+1"><tt>Hi Dima,</tt><tt><br>
</tt><tt><br>
</tt><tt>Thanks for your answer.
ExplicitGCInvokesConcurrent=true is used in CMS and G1
([1][2]). CMS ends with concurrent phases, G1 ends with STW
phases [3][4]. Performance counters are updated in
TraceMemoryManagerStats::~TraceMemoryManagerStats() which is
invoked in G1CollectedHeap::do_full_collection() and
CMSCollector::sweep(). So I don't see any reason why we should
don't execute this test with G1 and
ExplicitGCInvokesConcurrent=true. I'd like to keep requires as
in this change. Furthermore I have a change for WhiteBox to
retreive CMS state (is in idling or not)[5], which can allow
this test run with CMS and ExplicitGCInvokesConcurrent, but it
targeted to 10.</tt><tt><br>
</tt><tt><br>
</tt><tt>[1]
hotspot/src/share/vm/gc/shared/genCollectedHeap.cpp:309Â bool
GenCollectedHeap::should_do_concurrent_full_gc(GCCause::Cause
cause)</tt><tt><br>
</tt><tt>[2] hotspot/src/share/vm/gc/g1/g1CollectedHeap.cpp:2172
boolÂ
G1CollectedHeap::is_user_requested_concurrent_full_gc(GCCause::Cause
cause)</tt><tt><br>
</tt><tt>[3] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="http://www.oracle.com/technetwork/tutorials/tutorials-1876574.html">http://www.oracle.com/technetwork/tutorials/tutorials-1876574.html</a></tt><tt><br>
</tt><tt>[4] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://blogs.oracle.com/jonthecollector/entry/hey_joe_phases_of_cms">https://blogs.oracle.com/jonthecollector/entry/hey_joe_phases_of_cms</a></tt><tt><br>
</tt><tt>[5] <a moz-do-not-send="true"
class="moz-txt-link-freetext"
href="https://bugs.openjdk.java.net/browse/JDK-8169510">https://bugs.openjdk.java.net/browse/JDK-8169510</a></tt></font><br>
<br>
<div class="moz-cite-prefix">On 10/11/2016 12:47, Dmitry
Fazunenenko wrote:<br>
</div>
<blockquote
cite="mid:b218241b-eb17-e043-2ede-bde19280e23a@oracle.com"
type="cite">Hi Michail, <br>
<br>
I think this test is not designed to work in
ExplicitGCInvokesConcurrent=true. <br>
So, I would require: <br>
 * @requires vm.opt.ExplicitGCInvokesConcurrent != true <br>
<br>
Thanks, <br>
Dima <br>
<br>
On 11/9/2016 7:44 PM, Michail Chernov wrote: <br>
<blockquote type="cite">Hi, <br>
<br>
Could I have a reviews for this small change which excludes
test run with CMS and -XX:+ExplicitGCInvokesConcurrent,
please? <br>
<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="https://bugs.openjdk.java.net/browse/JDK-8165955">https://bugs.openjdk.java.net/browse/JDK-8165955</a>
<br>
<a moz-do-not-send="true" class="moz-txt-link-freetext"
href="http://cr.openjdk.java.net/%7Emchernov/8165955/webrev.00/">http://cr.openjdk.java.net/~mchernov/8165955/webrev.00/</a>
<br>
<br>
Test fails because CMS with -XX:+ExplicitGCInvokesConcurrent
causes to java's threads unlocking before
CMSCollector::sweep() has been done. So the GC counters can be
updated after test gets their value. <br>
<br>
Thanks, <br>
Michail <br>
</blockquote>
<br>
<br>
</blockquote>
<br>
</blockquote>
<br>
</body>
</html>