[Octave-bug-tracker] [bug #53733] Panic/segfault when calling Java method after forge:im[read|write]

[Adrian]

Follow-up Comment #51, bug #53733 (project octave):

Submit bug report to OpenJDK ?

The Java issues that were cited here consider the Segfault at JVM creation
resolved since basically summer 2017, so the issues are closed. Notably 
https://bugs.openjdk.java.net/browse/JDK-8183188
observed the bug on a Xeon X5675 processor, where indeed the simple program
now executes flawlessly. Wouldn't it be worthwhile re-opening the bug or
submitting a new issue on the fact that the test program still fails under gdb
on certain processors (possibly for a different reason)? This seriously
hampers debugging of programs using jvm-functionnality, as here, and also
indicates that something is still rotten in the land of JNI. However it would
probably be better if someone more knowledgeable than me in Java/JNI-matters
and generally in programming reported the bug. :-)

[Never mind the question in my previous comment: I imagine that if it were
easy to link octave with a non-openjdk JVM you'd have tried that.]


    _______________________________________________________

Reply to this item at:

  <http://savannah.gnu.org/bugs/?53733>

_______________________________________________
  Message sent via Savannah
  https://savannah.gnu.org/