I've now seen a couple of JPRT test jobs that have crashed on OSX when doing GC:
#
# A fatal error has been detected by the Java Runtime Environment:
#
# SIGSEGV (0xb) at pc=0x000000010d8b0685, pid=31483, tid=14595
#
# JRE version: Java(TM) SE Runtime Environment (8.0-b83) (build 1.8.0-ea-b83)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.0-b26-internal-201304042344.daholme.hotspot-rt mixed mode bsd-amd64 compressed oops)
# Problematic frame:
# V [libjvm.dylib+0x1b3685]
#
# Core dump written. Default location: /cores/core or core.31483
#
# If you would like to submit a bug report, please visit:
# http://bugreport.sun.com/bugreport/crash.jsp
#
--------------- T H R E A D ---------------
Current thread (0x00007fe7b8815800): GCTaskThread [stack: 0x00000001304c8000,0x00000001305c8000] [id=14595]
#
# A fatal error has been detected by the Java Runtime Environment:
#
# Internal Error (/opt/jprt/T/P1/131255.hseigel/s/src/share/vm/gc_implementation/parallelScavenge/parMarkBitMap.hpp:433), pid=72049, tid=15619
# assert(addr <= region_start() + region_size()) failed: addr too big
#
# JRE version: Java(TM) SE Runtime Environment (8.0-b82) (build 1.8.0-ea-b82)
# Java VM: Java HotSpot(TM) 64-Bit Server VM (25.0-b25-internal-201303251312.hseigel.bug_7154889-nojdk-fastdebug mixed mode bsd-amd64 compressed oops)
# Core dump written. Default location: /cores/core or core.72049
#
# If you would like to submit a bug report, please visit:
# http://bugreport.sun.com/bugreport/crash.jsp
#
--------------- T H R E A D ---------------
Current thread (0x00007fcfb3028800): GCTaskThread [stack: 0x000000011e416000,0x000000011e516000] [id=15619]
Stack: [0x000000011e416000,0x000000011e516000], sp=0x000000011e5152f0, free space=1020k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V [libjvm.dylib+0x7b2c76] VMError::report_and_die()+0x622
V [libjvm.dylib+0x2d320d] report_vm_error(char const*, int, char const*, char const*)+0x63
V [libjvm.dylib+0x63b4be] ParMarkBitMap::mark_obj(HeapWord*, unsigned long)+0x1ce
V [libjvm.dylib+0x3e76a9] void PSParallelCompact::mark_and_push<unsigned int>(ParCompactionManager*, unsigned int*)+0x14b
V [libjvm.dylib+0x3de888] InstanceKlass::oop_follow_contents(ParCompactionManager*, oopDesc*)+0x1d8
V [libjvm.dylib+0x659ac5] oopDesc::follow_contents(ParCompactionManager*)+0x18d
V [libjvm.dylib+0x66c493] ParCompactionManager::follow_marking_stacks()+0x7f
V [libjvm.dylib+0x659077] StealMarkingTask::do_it(GCTaskManager*, unsigned int)+0xe7
V [libjvm.dylib+0x3825e0] GCTaskThread::run()+0x1fe
V [libjvm.dylib+0x62a228] java_start(Thread*)+0xe4
C [libsystem_c.dylib+0x14742] _pthread_start+0x147
C [libsystem_c.dylib+0x1181] thread_start+0xd
and there may be others scattered through the JPRT archives.
I can't tell if these are actually related of course.