<!DOCTYPE html><html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
Hi Brian,<br>
<br>
As Thomas mentioned you uncovered a problem with THP in JDK 25.<br>
<br>
I guess the reason that you don't see the regression in your
baseline is because you compare JEP 522 against JDK 24 (which is not
affected by the regression), is that correct? <br>
<br>
The workaround for using THP with G1 in JDK 25 is to configure the
THP mode to 'always' instead of 'madvise'. <br>
<br>
Hope this helps,<br>
Stefan<br>
<br>
<div class="moz-cite-prefix">On 2025-09-01 11:26, Thomas Schatzl
wrote:<br>
</div>
<blockquote type="cite" cite="mid:2eb9945b-80e1-4f98-a8f6-073ca221191f@oracle.com">Hi
Brian,
<br>
<br>
On 28.08.25 23:49, Brian S O'Neill wrote:
<br>
<blockquote type="cite">Okay...
<br>
<br>
sysctl -w vm.nr_hugepages=2048
<br>
<br>
java -XX:+UseG1GC --enable-native-access=ALL-UNNAMED
--add-exports java.base/jdk.internal.misc=ALL-UNNAMED
--enable-native-access=ALL- UNNAMED --add-opens
java.base/jdk.internal.misc=ALL-UNNAMED -Xms3g - Xmx3g
-XX:+UseLargePages -XX:-UseTransparentHugePages <main
class>
<br>
<br>
G1GC: 234.7 seconds
<br>
G1GC JEP 522: 230.7 seconds
<br>
<br>
So, it seems that regression is with THP.
<br>
</blockquote>
<br>
given your report about THP issues, a colleague looked into the
THP code, and it seems that THP is somewhat broken for JDK 25.
<br>
<br>
This is currently under investigation in
<a class="moz-txt-link-freetext" href="https://bugs.openjdk.org/browse/JDK-8366434">https://bugs.openjdk.org/browse/JDK-8366434</a>.
<br>
<br>
Not sure if it helps here, because in both baseline and changed
code the error is probably present.
<br>
<br>
Hth,
<br>
Thomas
<br>
<br>
</blockquote>
<br>
</body>
</html>