The test vm/concepts/execution/execution056/execution05601/execution05601.html
seg faults intermittently when run on Solaris/x86 systems with 2 CPU.
The script x.sh runs the test until it fails outside of any test harness.
----------------------------
flora% uname -a
SunOS flora 5.6 Generic i86pc i386 i86pc
flora% psrinfo
0 on-line since 01/27/98 16:08:51
1 on-line since 01/27/98 16:08:51
flora% cat x.sh
#!/bin/sh

PATH=$JAVA_HOME/bin:$PATH
CLASSPATH=.
THREADS_FLAG=native

export PATH CLASSPATH THREADS_FLAG

java -version
java -fullversion

javac -d . execution05601c.java
javac -d . execution05601b.java
javac -d . execution05601a.java
javac -d . execution05601.java

while [ "1"="1" ]; do
java -verify javasoft.sqe.tests.vm.execution.execution056.execution05601.execution05601 > /dev/null
STATUS=$?
echo ${STATUS}
if [ ${STATUS} != 95 ]; then
break
fi
done
flora% x.sh
java version "1.1.6"
java full version "JDK1.1.6N"
95
95
95
95
95
95
95
95
95
95
SIGSEGV 11* segmentation violation
    si_signo [11]: SIGSEGV 11* segmentation violation
    si_errno [0]: Error 0
    si_code [1]: SEGV_MAPERR [addr: 0x38]

stackbase=DE24CDB0, stackpointer=DE24C90C

Full thread dump:
    "Thread-6" (TID:0xde703f90, sys_thread_t:0x80a4a20, state:R, thread_t: t@10, sp:0x0 threadID:0xde24cdd0, stack_base:0xde24cdb0, stack_size:0x21000) prio=5 *current thread*
javasoft.sqe.tests.vm.execution.execution056.execution05601.execution05601b.<init>(execution05601b.java:6)
javasoft.sqe.tests.vm.execution.execution056.execution05601.execution05601a.run(execution05601a.java:17)
java.lang.Thread.run(Thread.java)
    "SIGQUIT handler" (TID:0xde700210, sys_thread_t:0x8073d90, state:R, thread_t: t@5, sp:0x0 threadID:0xde2fbdd0, stack_base:0xde2fbdb0, stack_size:0x21000) prio=0
    "Finalizer thread" (TID:0xde700048, sys_thread_t:0x806e9e8, state:CW, thread_t: t@4, sp:0x0 threadID:0xdfc41dd0, stack_base:0xdfc41db0, stack_size:0x21000) prio=1
    "main" (TID:0xde700078, sys_thread_t:0x80653a0, state:CW, thread_t: t@1, sp:0x0 threadID:0x8049a38, stack_base:0x804705c, stack_size:0x848000) prio=5
java.lang.Thread.join(Thread.java)
java.lang.Thread.join(Thread.java)
javasoft.sqe.tests.vm.execution.execution056.execution05601.execution05601.run(execution05601.java:53)
javasoft.sqe.tests.vm.execution.execution056.execution05601.execution05601.main(execution05601.java:72)
Monitor Cache Dump:
    java.lang.Thread@DE703F90/DE74F4A0: <unowned>
Waiting to be notified:
"main" (0x80653a0)
Registered Monitor Dump:
    Thread queue lock: <unowned>
    Verifier lock: <unowned>
    Name and type hash table lock: <unowned>
    String intern lock: <unowned>
    JNI pinning lock: <unowned>
    JNI global reference lock: <unowned>
    BinClass lock: <unowned>
    Class loading lock: <unowned>
    Java stack lock: <unowned>
    Code rewrite lock: <unowned>
    Heap lock: <unowned>
    Has finalization queue lock: <unowned>
    Finalize me queue lock: <unowned>
Waiting to be notified:
"Finalizer thread" (0x806e9e8)
    Monitor registry: owner "Thread-6" (0x80a4a20, 1 entry)
Abort
134
flora%
----------------------------
JVM must never crash. So this bug is against runtime.
The problem was reported by Allan Jacobs from SunSoft.