P1
Run specjbb using the b69 build as follows (the given switches are the
current preferred set for running specjbb, though these specify a
small heap config).
java -d64 -server -verbose:gc -XX:+PrintGCDetails -Xbatch -XX:+OverrideDefaultLibthread -Xss128k -XX:+UseISM -XX:MaxLiveObjectEvacuationRatio=0 -Xms300m -Xmx300m -XX:NewSize=200m -XX:MaxNewSize=200m spec.jbb.JBBmain -propfile SPECjbb.props
Error occurred during initialization of VM
java.lang.NullPointerException
Java HotSpot(TM) 64-Bit Server VM warning: Unable to remove ISM segment with shmid=701
Running ipcs afterward shows no attached shared memory segment.
The heap size parameters don't matter, while turning off UseISM works fine.
The 32-bit vm doesn't show the problem.
Running under the debug vm, we sucessfully attach the shared mem region
(i.e., stepping through reserve_memory_special shows nothing problematic).
We end up in abort with the stack trace
Error occurred during initialization of VM
java.lang.NullPointerException
Dumping core....
t@1 (l@1) signal ABRT (Abort) in __sigprocmask at 0xffffffff7f21d168
0xffffffff7f21d168: __sigprocmask+0x0008: retl
Current function is os::abort
1067 ::abort(); // dump core (for debugging)
(.../devtools/sparc/SUNWspro/SC6.1/bin/../WS6U1/bin/sparcv9/dbx) where 10
current thread: t@1
[1] __sigprocmask(0x0, 0xffffffff7fffe0b0, 0x0, 0x100107978, 0x0, 0x0), at 0xffffffff7f21d168
[2] _resetsig(0xffffffff7f220200, 0x1, 0x100107988, 0xffffffff7fffe0a0, 0x100107988, 0x0), at 0xffffffff7f211544
[3] _sigon(0xffffffff7f32d700, 0x1001078e0, 0x100107978, 0xffffffff7fffe22c, 0x1001078e0, 0xffffffff7f322000), at 0xffffffff7f210c04
[4] _thrp_kill(0x0, 0x1, 0xffffffff7f322000, 0x1001078e0, 0x0, 0x1), at 0xffffffff7f213fb0
[5] raise(0x6, 0x0, 0x0, 0x0, 0xffffffffffffffff, 0xffffffff7f0b9f88), at 0xffffffff7ef51e04
[6] abort(0xffffffff7f0b0ef8, 0xffffffff7fffe480, 0x0, 0xfffffffffffffff8, 0x4, 0xffffffff7fffe4a1), at 0xffffffff7ef3cc3c
=>[7] os::abort(dump_core = 1), line 1067 in "/net/altair/export/space4/jtg/phh/hs64/src/os/solaris/vm/os_solaris.cpp"
[8] vm_abort(), line 402 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/java.cpp"
[9] vm_exit_during_initialization(exception = CLASS), line 411 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/java.cpp"
[10] ExceptionMark::~ExceptionMark(this = 0xffffffff7fffe910), line 219 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/utilities/exceptions.cpp"
[11] Threads::create_vm(args = 0xffffffff7fffeb78), line 2354 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/thread.cpp"
[12] JNI_CreateJavaVM(vm = 0xffffffff7ffff510, penv = 0xffffffff7ffff508, args = 0xffffffff7fffeb78), line 2194 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/prims/jni.cpp"
[13] InitializeJVM(0xffffffff7ffff510, 0xffffffff7ffff508, 0xffffffff7ffff4c0, 0xffffffff7ffff4f8, 0xffffffff7ffff4d4, 0xffffffff7ffff6e8), at 0x10000432c
[14] main(0xf, 0xffffffff7ffff5e8, 0xffffffff7ffff668, 0x0, 0x0, 0x100000000), at 0x100001a84
The code in ~ExceptionMark believes that there's a pending exception.
The b68 build works fine.
paul.hohensee@East 2001-06-22
Putbacks to the b68 vm thru 20010619081056.sgoldman.strictfp2 (the one
just before Ken's) work. Ken's putback installed in the b69 jdk fails.
paul.hohensee@East 2001-06-22
current preferred set for running specjbb, though these specify a
small heap config).
java -d64 -server -verbose:gc -XX:+PrintGCDetails -Xbatch -XX:+OverrideDefaultLibthread -Xss128k -XX:+UseISM -XX:MaxLiveObjectEvacuationRatio=0 -Xms300m -Xmx300m -XX:NewSize=200m -XX:MaxNewSize=200m spec.jbb.JBBmain -propfile SPECjbb.props
Error occurred during initialization of VM
java.lang.NullPointerException
Java HotSpot(TM) 64-Bit Server VM warning: Unable to remove ISM segment with shmid=701
Running ipcs afterward shows no attached shared memory segment.
The heap size parameters don't matter, while turning off UseISM works fine.
The 32-bit vm doesn't show the problem.
Running under the debug vm, we sucessfully attach the shared mem region
(i.e., stepping through reserve_memory_special shows nothing problematic).
We end up in abort with the stack trace
Error occurred during initialization of VM
java.lang.NullPointerException
Dumping core....
t@1 (l@1) signal ABRT (Abort) in __sigprocmask at 0xffffffff7f21d168
0xffffffff7f21d168: __sigprocmask+0x0008: retl
Current function is os::abort
1067 ::abort(); // dump core (for debugging)
(.../devtools/sparc/SUNWspro/SC6.1/bin/../WS6U1/bin/sparcv9/dbx) where 10
current thread: t@1
[1] __sigprocmask(0x0, 0xffffffff7fffe0b0, 0x0, 0x100107978, 0x0, 0x0), at 0xffffffff7f21d168
[2] _resetsig(0xffffffff7f220200, 0x1, 0x100107988, 0xffffffff7fffe0a0, 0x100107988, 0x0), at 0xffffffff7f211544
[3] _sigon(0xffffffff7f32d700, 0x1001078e0, 0x100107978, 0xffffffff7fffe22c, 0x1001078e0, 0xffffffff7f322000), at 0xffffffff7f210c04
[4] _thrp_kill(0x0, 0x1, 0xffffffff7f322000, 0x1001078e0, 0x0, 0x1), at 0xffffffff7f213fb0
[5] raise(0x6, 0x0, 0x0, 0x0, 0xffffffffffffffff, 0xffffffff7f0b9f88), at 0xffffffff7ef51e04
[6] abort(0xffffffff7f0b0ef8, 0xffffffff7fffe480, 0x0, 0xfffffffffffffff8, 0x4, 0xffffffff7fffe4a1), at 0xffffffff7ef3cc3c
=>[7] os::abort(dump_core = 1), line 1067 in "/net/altair/export/space4/jtg/phh/hs64/src/os/solaris/vm/os_solaris.cpp"
[8] vm_abort(), line 402 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/java.cpp"
[9] vm_exit_during_initialization(exception = CLASS), line 411 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/java.cpp"
[10] ExceptionMark::~ExceptionMark(this = 0xffffffff7fffe910), line 219 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/utilities/exceptions.cpp"
[11] Threads::create_vm(args = 0xffffffff7fffeb78), line 2354 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/runtime/thread.cpp"
[12] JNI_CreateJavaVM(vm = 0xffffffff7ffff510, penv = 0xffffffff7ffff508, args = 0xffffffff7fffeb78), line 2194 in "/net/altair/export/space4/jtg/phh/hs64/src/share/vm/prims/jni.cpp"
[13] InitializeJVM(0xffffffff7ffff510, 0xffffffff7ffff508, 0xffffffff7ffff4c0, 0xffffffff7ffff4f8, 0xffffffff7ffff4d4, 0xffffffff7ffff6e8), at 0x10000432c
[14] main(0xf, 0xffffffff7ffff5e8, 0xffffffff7ffff668, 0x0, 0x0, 0x100000000), at 0x100001a84
The code in ~ExceptionMark believes that there's a pending exception.
The b68 build works fine.
paul.hohensee@East 2001-06-22
Putbacks to the b68 vm thru 20010619081056.sgoldman.strictfp2 (the one
just before Ken's) work. Ken's putback installed in the b69 jdk fails.
paul.hohensee@East 2001-06-22