FULL PRODUCT VERSION :
java version "1.6.0_14"
Java(TM) SE Runtime Environment (build 1.6.0_14-b08)
Java HotSpot(TM) 64-Bit Server VM (build 14.0-b16, mixed mode)

FULL OS VERSION :
Intel Core i7 920 D0 machine:
Linux renderstationng 2.6.28-11-generic #42-Ubuntu SMP Fri Apr 17 01:58:03 UTC 2009 x86_64 GNU/Linux

Dual AMD Opteron 2376 machines:
Linux dryad06 2.6.18-128.1.1.el5 #1 SMP Mon Jan 26 13:58:24 EST 2009 x86_64 x86_64 x86_64 GNU/Linux

EXTRA RELEVANT SYSTEM CONFIGURATION :
Intel machine: 6GB RAM
AMD machines: 16GB RAM

A DESCRIPTION OF THE PROBLEM :
The problem occurs on a benchmark we are running as part of research project when using the new G1 garbage collector. The benchmark was run multiple times before with the default garbage collector and also with other VMs (OpenJDK, IcedTea) without problems. With JDK6u14 and G1 enabled (VM options used: -XX:+UnlockExperimentalVMOptions -XX:+UseG1GC -Xmx4G) the benchmark crashes with a SIGSEGV on all four AMD and the Intel machine.
This also happens with "-server" added to the VM options.

The benchmark is very memory-intensive and runs (if it runs to completion) for three hours. During these three hours, input load is ramped up linearly. The JVM segfaults somewhen in these three hours, it happened after 10 minutes on one machines up to after 70 minutes on another machine.

THE PROBLEM WAS REPRODUCIBLE WITH -Xint FLAG: Did not try

THE PROBLEM WAS REPRODUCIBLE WITH -server FLAG: Yes

STEPS TO FOLLOW TO REPRODUCE THE PROBLEM :
You will not be able to reproduce this bug without all the code and input data of our benchmark implementation.

EXPECTED VERSUS ACTUAL BEHAVIOR :
Expected: runs to completion
Actual: JVM crashes with SIGSEGV

ERROR MESSAGES/STACK TRACES THAT OCCUR :
Attached seperatly

REPRODUCIBILITY :
This bug can be reproduced always.

---------- BEGIN SOURCE ----------
not available
---------- END SOURCE ----------