<html>
<head>
<meta content="text/html; charset=utf-8" http-equiv="Content-Type">
</head>
<body bgcolor="#FFFFFF" text="#000000">
<div class="moz-cite-prefix">On 1/27/15 6:03 PM, Claes Redestad
wrote:<br>
</div>
<blockquote cite="mid:54C81953.7080003@oracle.com" type="cite">Hi,
<br>
<br>
Assuming ref.1 is the baseline and cms.1 is with the patch applied
(running G1?),
<br>
it actually looks like all significant results are regressions,
not improvements. Some
<br>
regressions even appear pretty severe.
<br>
<br>
</blockquote>
<br>
Thanks Claes,<br>
<br>
I made some beginner mistakes here. Yes, ref is the baseline, and I
read +/- backwards. But I didn't specify g1 (or any options at all)
in the refworkload property file. So the changed code should not
have run at all, which really leaves the differences unexplained.<br>
<br>
In any case, running baseline and changed code again with g1
specified.<br>
<blockquote cite="mid:54C81953.7080003@oracle.com" type="cite">Grouping
state changed often(?) together could be increasing overall false
sharing
<br>
costs and the added indirections could have some effect. I think
it'd be good with a
<br>
more thorough performance analysis of this.
<br>
</blockquote>
<br>
The fields were moved to another class, but it's more like a struct
inlined within
<meta http-equiv="content-type" content="text/html; charset=utf-8">
G1CollectorPolicy. There is not an actually separately allocated
object, so in theory the compiler can generate the exact same code
as before. If I still see a regression I'll have to look at the
generated code to see if the compiler gave up at some point.<br>
<br>
Also, what more thorough performance analysis would you recommend
(if it's needed).<br>
<br>
Thanks!<br>
<br>
- Derek<br>
<br>
<br>
</body>
</html>