<html>
<head>
<meta http-equiv="content-type" content="text/html; charset=windows-1252">
</head>
<body bgcolor="#FFFFFF" text="#000000">
[Resend to correct email list]<br>
<div class="moz-forward-container"><br>
<br>
-------- Forwarded Message --------
<table class="moz-email-headers-table" border="0" cellpadding="0"
cellspacing="0">
<tbody>
<tr>
<th align="RIGHT" nowrap="nowrap" valign="BASELINE">Subject:
</th>
<td>Re: RFR: 8133051: Concurrent refinement threads may be
activated and deactivated at random</td>
</tr>
<tr>
<th align="RIGHT" nowrap="nowrap" valign="BASELINE">Date: </th>
<td>Fri, 15 Apr 2016 13:12:28 -0400</td>
</tr>
<tr>
<th align="RIGHT" nowrap="nowrap" valign="BASELINE">From: </th>
<td>Derek White <a class="moz-txt-link-rfc2396E" href="mailto:derek.white@oracle.com"><derek.white@oracle.com></a></td>
</tr>
</tbody>
</table>
<br>
<br>
<meta content="text/html; charset=windows-1252"
http-equiv="Content-Type">
<div class="moz-cite-prefix">On 4/12/16 2:31 PM, Kim Barrett
wrote:<br>
</div>
<blockquote
cite="mid:D7DDC4FA-B97E-4B30-B292-F66020925530@oracle.com"
type="cite">
<blockquote type="cite">
<pre wrap="">On Apr 4, 2016, at 2:48 PM, Kim Barrett <a moz-do-not-send="true" class="moz-txt-link-rfc2396E" href="mailto:kim.barrett@oracle.com"><kim.barrett@oracle.com></a> wrote:
Please review this change to the G1 concurrent refinement thread
controller. This change addresses unnecessary activation when there
are many threads and few buffers to be processed. It also addresses
delayed activation due to mis-configuration of the dirty card queue
set's notification mechanism.
</pre>
</blockquote>
<pre wrap="">I had been hoping I might come back from vacation and find some reviews waiting for me to look at.
</pre>
</blockquote>
Hi Kim,<br>
<br>
Looks good.<br>
<br>
My final comment:<br>
<meta http-equiv="content-type" content="text/html;
charset=windows-1252">
<pre><span class="changed">src/share/vm/gc/g1/concurrentG1Refine.cpp:
</span></pre>
<span class="changed"> - In ConcurrentG1Refine::create(), around
line 188, it would be nice to log number "max number of
concurrent refinement threads" (aka thread_num() aka </span><span
class="changed">_n_worker_threads).<br>
<br>
- Derek<br>
</span>
<pre><meta http-equiv="content-type" content="text/html; charset=windows-1252"></pre>
</div>
</body></html>