Issue

When the Concurrent Mark Sweep (CMS) garbage collector (GC) is used as a GC Policy, if ConcGCThreads count is more than vm core count, the jvm instance crashes. (-XX:ConcGCThreads)

# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGSEGV (0xb) at pc=0x00007f55ac1a515f, pid=29830, tid=140005673514752
#
# JRE version: 6.0_22-b22
# Java VM: OpenJDK 64-Bit Server VM (20.0-b11 mixed mode linux-amd64 compressed oops)
# Derivative: IcedTea6 1.10.8
# Distribution: Fedora release 15 (Lovelock), package fedora-65.1.10.8.fc15-x86_64
# Problematic frame:
# V  [libjvm.so+0x49115f]  AsyncGetCallTrace+0x8271f
.. .. ..
---------------  T H R E A D  ---------------

Current thread (0x00007f55a411d800):  VMThread [stack: 0x00007f559c5fb000,0x00007f559c6fc000] [id=29847]
.. .. ..
Stack: [0x00007f559c5fb000,0x00007f559c6fc000],  sp=0x00007f559c6fa620,  free space=1021k
Native frames: (J=compiled Java code, j=interpreted, Vv=VM code, C=native code)
V  [libjvm.so+0x49115f]  AsyncGetCallTrace+0x8271f
V  [libjvm.so+0x36a73d]  ftell+0x36a73d
V  [libjvm.so+0x447eb9]  AsyncGetCallTrace+0x39479
V  [libjvm.so+0x43d8ff]  AsyncGetCallTrace+0x2eebf
V  [libjvm.so+0x43f2ed]  AsyncGetCallTrace+0x308ad
V  [libjvm.so+0x43f7b7]  AsyncGetCallTrace+0x30d77
V  [libjvm.so+0x3a3184]  ftell+0x3a3184
V  [libjvm.so+0x3a6411]  ftell+0x3a6411

JVM parameters

-XX:+CMSClassUnloadingEnabled -XX:+CMSPermGenSweepingEnabled -XX:+UseConcMarkSweepGC -XX:+CMSConcurrentMTEnabled -XX:ConcGCThreads=4