java-j9-vame jit coredumps on QNX?

I am porting a large & healthy java app onto qnx. When I run it without the jit all is
good. If I run it with the -jit option the vm coredumps with the following error (tx to gdb)

Program received signal SIGSEGV, Segmentation fault.
[Switching to process 136377136]
0xb83e1018 in ?? () from /opt/vame1.4/ive/bin/libj9jit14.so

I am using vame1.4 and have tested the .jar on NT without any problems.

thoughts? questions? comments? help?

Chris Goebel
cgoebel@tridium.com
804-474-5677

Chris Goebel <cgoebel@tridium.com> wrote:

I am porting a large & healthy java app onto qnx. When I run it without the jit all is
good. If I run it with the -jit option the vm coredumps with the following error (tx to gdb)

Program received signal SIGSEGV, Segmentation fault.
[Switching to process 136377136]
0xb83e1018 in ?? () from /opt/vame1.4/ive/bin/libj9jit14.so

What’s the function it’s dying in? You can objdump -x | sort | less , and then
take a look for the function that the address falls into. That should give you a
clue as to what might be happening. If it’s the same address all the time, you
can start looking at some of the parameters being passed, and perhaps some program
state. But with just a dump, it’s hard to really say, expecially without the library
at hand.

-Adam