<html><head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
</head>
<body>
On 04/01/2023 04:59, robert engels wrote:<br>
<blockquote type="cite" cite="mid:96631C23-9C63-4848-B310-C0521FB094E9@ix.netcom.com">
<div dir="auto" style="word-wrap: break-word; -webkit-nbsp-mode:
space; line-break: after-white-space;" class="">
<div style="word-wrap: break-word; -webkit-nbsp-mode: space;
line-break: after-white-space;" class="">:
<div style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class=""><br class="">
</div>
<div style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class="">Here is a async profile capture of 3b:</div>
<div style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class=""><br class="">
</div>
<div style="caret-color: rgb(0, 0, 0); color: rgb(0, 0, 0);" class=""><img apple-inline="yes" id="DA5D9975-772E-4B78-AA2A-1C9A62D9AE78" src="cid:part1.cOQdsm4E.0ZQolXID@oracle.com" class="" width="640" height="208"></div>
<div class=""><br class="">
</div>
<div class="">Notice that the vast majority of the time is
used in internal context switching.</div>
<br>
</div>
</div>
</blockquote>
I can't immediately tell if there is a Heisenberg effect here but
the notifyJvmtiMountXXX methods suggest that the profiler has a
JVMTI environment. There is work in progress to address scalability
issues with JVMTI in exactly this area. In the mean-time, it would
be useful if you could re-run the profiler with the test running
with:<br>
<br>
-XX:+UnlockExperimentalVMOptions
-XX:+DoJVMTIVirtualThreadTransitions <br>
<br>
-Alan<br>
</body>
</html>