JFR integration

[Erik Gahlin]

Hi Jonathan,

There is an open bug related to missing threads in JFR [1]. It could be the issue, or it might something new for VT. We will take a look. Thanks for reporting. Today we record OS thread in the thread struct[2], maybe we could add the carrier/platform thread as well. It would allow JMC to render VTs in the thread view.

Erik

[1] https://bugs.openjdk.org/browse/JDK-8304033

[2] https://github.com/openjdk/jdk/blob/master/src/hotspot/share/jfr/metadata/metadata.xml#L1207
[https://opengraph.githubassets.com/a5d3f73dbf47f40527cd2c8e76db552e72032595270113a4cd1943e26cc905ca/openjdk/jdk]<https://github.com/openjdk/jdk/blob/master/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
jdk/metadata.xml at master · openjdk/jdk<https://github.com/openjdk/jdk/blob/master/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
JDK main-line development https://openjdk.org/projects/jdk - jdk/metadata.xml at master · openjdk/jdk
github.com

[https://opengraph.githubassets.com/a5d3f73dbf47f40527cd2c8e76db552e72032595270113a4cd1943e26cc905ca/openjdk/jdk]<https://github.com/openjdk/jdk/blob/e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
jdk/metadata.xml at e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1 · openjdk/jdk<https://github.com/openjdk/jdk/blob/e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
JDK main-line development https://openjdk.org/projects/jdk - jdk/metadata.xml at e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1 · openjdk/jdk
github.com

[https://opengraph.githubassets.com/a5d3f73dbf47f40527cd2c8e76db552e72032595270113a4cd1943e26cc905ca/openjdk/jdk]<https://github.com/openjdk/jdk/blob/e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
jdk/metadata.xml at e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1 · openjdk/jdk<https://github.com/openjdk/jdk/blob/e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1/src/hotspot/share/jfr/metadata/metadata.xml#L1207>
JDK main-line development https://openjdk.org/projects/jdk - jdk/metadata.xml at e73411a2354cf266ab7a5ddadfb6ea98d7eb4cd1 · openjdk/jdk
github.com


Erik
________________________________
From: hotspot-jfr-dev <hotspot-jfr-dev-retn at openjdk.org> on behalf of Alan Bateman <Alan.Bateman at oracle.com>
Sent: Wednesday, March 22, 2023 9:21 PM
To: Jonathan Ross <jonathan.ross at cjug.org>
Cc: hotspot-jfr-dev <hotspot-jfr-dev at openjdk.java.net>; loom-dev at openjdk.org <loom-dev at openjdk.org>
Subject: Re: JFR integration

On 22/03/2023 19:16, Jonathan Ross wrote:
> Okay, it must be something else causing the jfr problem, I'll contact
> jfr-dev as you suggest, thanks for that suggestion.
>
> Can you comment on jfr events not including information about the
> carrier thread?  Is there any particular reason not to (want to)
> include platform thread information?  I, for one, would like to be
> able to see the underlying fork-join scheduler at work in flight
> recordings.
The scheduler threads are distinct and you may see some events committed
on those threads. What would you do with the linkage if it was in the
recording?


>
> A second, somewhat related question: is it possible to use multiple
> distinct carrier thread pools, with different parallelism? (Perhaps
> there is some magic around creating virtual threads from within a
> forkjoin task?)
>
Not at this time. It will take time for the eco system to get used to
virtual threads and in the future we can evaluate if are compelling
reasons to allow for custom schedulers.

-Alan

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://mail.openjdk.org/pipermail/loom-dev/attachments/20230322/5a81c239/attachment.htm>