From: "Eugene W. Stark" <isabelle-users@starkeffect.com>
I just got the error shown below with Isabelle2025-RC1. I had been running this session
for some time, so it certainly does not seem to be something that occurs frequently or
repeatably. It is almost certainly a JRE bug that has nothing specific to do with
Isabelle or JEdit, but I am reporting it in case enough other people observe it that it
might cause reconsideration about the JRE release being used for Isabelle2025.
- Gene Stark
# A fatal error has been detected by the Java Runtime Environment:
# SIGILL (0x4) at pc=0x000076b5755e1dcd, pid=201573, tid=201628
# JRE version: OpenJDK Runtime Environment Zulu21.40+17-CA (21.0.6+7) (build 21.0.6+7-LTS)
Java VM: OpenJDK 64-Bit Server VM Zulu21.40+17-CA (21.0.6+7-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
Problematic frame:
V [libjvm.so+0x7e1dcd] G1ConcurrentMark::try_stealing(unsigned int, G1TaskQueueEntry&)+0x22d
# Core dump will be written. Default location: Core dumps may be processed with "/usr/share/apport/apport -p%p -s%s -c%c -d%d -P%P -u%u -g%g -- %E" (or dumping to /home/gene/Isabelle/core.201573)
# An error report file with more information is saved as:
/home/gene/Isabelle/hs_err_pid201573.log
# If you would like to submit a bug report, please visit:
http://www.azul.com/support/
#
From: Makarius <makarius@sketis.net>
On 11/02/2025 11:52, Eugene W. Stark wrote:
I just got the error shown below with Isabelle2025-RC1. I had been running this session
for some time, so it certainly does not seem to be something that occurs frequently or
repeatably. It is almost certainly a JRE bug that has nothing specific to do with
Isabelle or JEdit, but I am reporting it in case enough other people observe it that it
might cause reconsideration about the JRE release being used for Isabelle2025.
JVM breakdowns do happen occasionally, but are rare. I do have hs_err.log
files lying around all the time.
Java VM: OpenJDK 64-Bit Server VM Zulu21.40+17-CA (21.0.6+7-LTS, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, g1 gc, linux-amd64)
Problematic frame:
V [libjvm.so+0x7e1dcd] G1ConcurrentMark::try_stealing(unsigned int, G1TaskQueueEntry&)+0x22d
That looks specific to the traditional G1 garbage collector.
On my work machine I am actually testing the newer ZGC for a few years. My
settings for ISABELLE_JAVA_SYSTEM_OPTIONS include the follow options:
-XX:+UseZGC -XX:+ZGenerational -XX:SoftMaxHeapSize=2g
That is from some years ago. It might require some further research to see if
these are good or bad options.
I did not use this as default so far, because there used to problems with
older versions of Windows.
Makarius
Last updated: Mar 09 2025 at 12:28 UTC