<html><head><meta http-equiv="Content-Type" content="text/html charset=iso-8859-1"></head><body style="word-wrap: break-word; -webkit-nbsp-mode: space; -webkit-line-break: after-white-space; ">Hi,<div><br></div><div>I've just noticed that in the latest builds of the 1.6.0 and the 1.7.0 that there are new combinations/corruption of ParNew and CMS logs suggesting something has changed in the logging. I didn't see anything in the lists but...</div><div><br></div><div>I've also got this combination of records...</div><div><br></div><div><div>57.535: [GC 57.535: [ParNew</div><div>Desired survivor size 1081344 bytes, new threshold 1 (max 4)</div><div>- age 1: 2154184 bytes, 2154184 total</div><div>: 19136K->2112K(19136K), 0.0291977 secs] 101261K->97883K(126912K), 0.0292664 secs] [Times: user=0.10 sys=0.00, real=0.03 secs] </div><div>57.584: [GC 57.584: [ParNew: 19136K->19136K(19136K), 0.0000203 secs]57.584: [CMS: 95771K->72786K(107776K), 0.2469877 secs] 114907K->72786K(126912K), [CMS Perm : 14265K->14262K(24092K)], 0.2471242 secs] [Times: user=0.24 sys=0.00, real=0.25 secs] </div><div>57.834: [GC [1 CMS-initial-mark: 72786K(107776K)] 73371K(126912K), 0.0013167 secs] [Times: user=0.00 sys=0.00, real=0.01 secs] </div></div><div><br></div><div>Note that the 57.584 is a new corruption that I've not seen before. I assume that the second record is a full gc.</div><div><br></div><div>has there been any changes with logging?</div><div><br></div><div>Regards,</div><div>Kirk</div><div><br></div><div><div><div>On 2012-09-27, at 9:41 PM, Vitaly Davidovich <<a href="mailto:vitalyd@gmail.com">vitalyd@gmail.com</a>> wrote:</div><br class="Apple-interchange-newline"><blockquote type="cite">Thanks Jon -- that blog entry was useful.<div><br></div><div>Vitaly<br><br><div class="gmail_quote">On Thu, Sep 27, 2012 at 12:55 AM, Jon Masamitsu <span dir="ltr"><<a href="mailto:jon.masamitsu@oracle.com" target="_blank">jon.masamitsu@oracle.com</a>></span> wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">Vitaly,<br>
<br>
The current implementation depends on a thread not migrating<br>
between nodes. On solaris that naturally happens. I don't<br>
remember the details but it's something like Solaris sees that<br>
a thread XX is executing on node AA and using memory on AA<br>
so it leaves XX on AA. On linux I'm guessing (really guessing)<br>
that there is a way to create an affinity between XX on AA.<br>
<br>
This has all the things I ever knew about it.<br>
<br>
<a href="https://blogs.oracle.com/jonthecollector/entry/help_for_the_numa_weary" target="_blank">https://blogs.oracle.com/<u></u>jonthecollector/entry/help_<u></u>for_the_numa_weary</a><span class="HOEnZb"><font color="#888888"><br>
<br>
Jon</font></span><div class="HOEnZb"><div class="h5"><br>
<br>
On 9/26/2012 4:09 PM, Vitaly Davidovich wrote:<br>
<blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Hi guys,<br>
<br>
If I understand it correctly, the NUMA allocator splits eden into regions<br>
and tries to ensure that an allocated object is in a region local to the<br>
mutator thread. How does this affect tlabs? Specifically, a tlab will be<br>
handed out to a thread from the current node. If the java thread then<br>
migrates to a different node, its tlab is presumably still on the previous<br>
node, leading to cross-node traffic? Is there a notion of a processor local<br>
tlab? In that case, access to already allocated objects will take a hit but<br>
new allocations will not.<br>
<br>
The way I imagine a processor local tlab working is when a thread migrates,<br>
the previous tlab becomes available for whichever java thread is onproc<br>
there now - that is, tlab ownership changes. The migrated thread then<br>
picks up allocations in the new tlab.<br>
<br>
It can still be a bump the pointer since only one hardware thread can be<br>
running at a time on the processor.<br>
<br>
Is this or something like it already there? If not, what challenges am I<br>
overlooking from my high-level view?<br>
<br>
Thanks<br>
<br>
Sent from my phone<br>
<br>
</blockquote>
</div></div></blockquote></div><br></div>
</blockquote></div><br></div></body></html>