[Bug 167] JVM crash with CGLIB on a Linux Fedora 9 PPC

bugzilla-daemon at icedtea.classpath.org bugzilla-daemon at icedtea.classpath.org
Fri Jul 11 11:04:21 PDT 2008 

http://icedtea.classpath.org/bugzilla/show_bug.cgi?id=167


mbaudier at argeo.org changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |gbenson at redhat.com,
                   |                            |mbaudier at argeo.org
            Summary|JVM crash with Hibernate /  |JVM crash with CGLIB on a
                   |CGLIB on a Linux Fedora 9   |Linux Fedora 9 PPC
                   |PPC                         |




------- Comment #1 from mbaudier at argeo.org  2008-07-11 18:04 -------
Hi,

I digged further in order to isolate the problem and it is indeed with the
cglib library (http://cglib.sourceforge.net/).

Checking out this project and running the unit tests will lead to the crash
described previously.

#
# Reproducing based on cglib unit tests
#
cvs -z3 -d:pserver:anonymous at cglib.cvs.sourceforge.net:/cvsroot/cglib co cglib
cd cglib
ant

(below a minimal code to reproduce it, assuming cglib 2.2 and asm 3.1 are in
the classpath)

Interestingly, when (java) debugging, the crash occurs at different points. If
a checkpoint is set a the previous location of a crash, it doesn't crash (which
seems to point to a kind of concurrency issue?).

#
# Debugging with gdb
#
I re-tested with the latest version of OpenJdk in the rawhide repository of
Fedora (java-1.6.0-openjdk-1.6.0.0-0.16.b09.fc10.ppc), but I suspect that this
is the same code base. Anyhow, the stack is the same as previously (see
previous comment).

GDB output:

(gdb) run -cp .:../lib/cglib-2.2.jar:../lib/asm-3.1.jar:../lib/asm-util-3.1.jar
crashjvm.CrashJvm
Starting program: /usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0/bin/java -cp
.:../lib/cglib-2.2.jar:../lib/asm-3.1.jar:../lib/asm-util-3.1.jar
crashjvm.CrashJvm
[Thread debugging using libthread_db enabled]
[New process 9767]
Executing new program: /usr/lib/jvm/java-1.6.0-openjdk-1.6.0.0/bin/java
warning: Cannot initialize thread debugging library: generic error
warning: Cannot initialize thread debugging library: generic error
[New process 9767]
[Thread debugging using libthread_db enabled]
[New Thread 0x482bc4b0 (LWP 9768)]
[New Thread 0x6e7904b0 (LWP 9769)]
[New Thread 0x6e8904b0 (LWP 9770)]
[New Thread 0x6e9904b0 (LWP 9771)]
[New Thread 0x6ec904b0 (LWP 9772)]
[New Thread 0x6ed904b0 (LWP 9773)]
[New Thread 0x6ee104b0 (LWP 9774)]
Start test.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x482bc4b0 (LWP 9768)]
constantPoolOopDesc::klass_ref_index_at (this=0x6a5855c0, which=6)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/oops/cpCacheOop.hpp:219
219       int constant_pool_index() const                { return _indices &
0xFFFF; }
Current language:  auto; currently c++
(gdb) bt
#0  constantPoolOopDesc::klass_ref_index_at (this=0x6a5855c0, which=6)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/oops/cpCacheOop.hpp:219
#1  0x0f048d98 in constantPoolOopDesc::klass_ref_at (this=<value optimized
out>, which=<value optimized out>, 
    __the_thread__=<value optimized out>)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/oops/constantPoolOop.cpp:306
#2  0x0f177cb4 in LinkResolver::resolve_klass (result=@0x4823c8f8, pool=Cannot
access memory at address 0x6
)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/interpreter/linkResolver.cpp:111
#3  0x0f17c284 in LinkResolver::resolve_field (result=@0x4823caf8, pool=
        {<Handle> = {_handle = 0x10020918}, <No data fields>}, index=6,
byte=Bytecodes::_getfield, check_only=false, 
    update_pool=<value optimized out>, __the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/interpreter/linkResolver.cpp:404
#4  0x0f17c94c in LinkResolver::resolve_field (result=<value optimized out>,
pool=Cannot access memory at address 0x6
)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/interpreter/linkResolver.cpp:391
#5  0x0f0b1ce4 in InterpreterRuntime::resolve_get_put (thread=0x1001f7a0,
bytecode=Bytecodes::_getfield)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/interpreter/interpreterRuntime.cpp:449
#6  0x0efaebf4 in BytecodeInterpreter::run (istate=0x482bb878)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/interpreter/bytecodeInterpreter.cpp:1693
#7  0x0f04d5d8 in CppInterpreter::normal_entry (method=<value optimized out>,
__the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:74
#8  0x0f04d60c in CppInterpreter::normal_entry (method=<value optimized out>,
__the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:89
#9  0x0f04d60c in CppInterpreter::normal_entry (method=<value optimized out>,
__the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:89
#10 0x0f04d60c in CppInterpreter::normal_entry (method=<value optimized out>,
__the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:89
#11 0x0f04d60c in CppInterpreter::normal_entry (method=<value optimized out>,
__the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/cppInterpreter_zero.cpp:89
#12 0x0f222a78 in StubGenerator::call_stub (call_wrapper=<value optimized out>,
result=0x482bbc78, result_type=T_INT, 
    method=0x6a51d030, entry_point=0xf04d4a0 "\224!��|\b\002�B\237",
parameters=<value optimized out>, 
    parameter_words=<value optimized out>, __the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/stubGenerator_zero.cpp:69
#13 0x0f0b6bfc in JavaCalls::call_helper (result=0x482bbc70, m=<value optimized
out>, args=0x482bbb60, 
    __the_thread__=0x1001f7a0) at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/runtime/javaCalls.cpp:388
#14 0x0f1b95a4 in os::os_exception_wrapper (f=0x24022822, value=<value
optimized out>, method=<value optimized out>, 
    args=<value optimized out>, thread=<value optimized out>)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/os/linux/vm/os_linux.cpp:3785
#15 0x0f0b5874 in JavaCalls::call (result=<value optimized out>, method=Cannot
access memory at address 0xb70011
)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/runtime/javaCalls.cpp:298
#16 0x0f0c6850 in jni_invoke_static (env=0x1001f8b0, result=0x482bbc70,
receiver=<value optimized out>, 
---Type <return> to continue, or q <return> to quit---
    call_type=<value optimized out>, method_id=<value optimized out>,
args=0x482bbc80, __the_thread__=0x1001f7a0)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/prims/jni.cpp:1005
#17 0x0f0d3ad4 in jni_CallStaticVoidMethod (env=0x1001f8b0, cls=<value
optimized out>, methodID=0x10072618)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/prims/jni.cpp:1577
#18 0x10003e70 in JavaMain (_args=<value optimized out>) at
../../../../src/share/bin/java.c:579
#19 0x0f376e3c in start_thread (arg=<value optimized out>) at
pthread_create.c:299
#20 0x48113670 in clone () from /lib/libc.so.6
Backtrace stopped: previous frame inner to this frame (corrupt stack?)
(gdb) finish
Run till exit from #0  constantPoolOopDesc::klass_ref_index_at
(this=0x6a5855c0, which=6)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/share/vm/oops/cpCacheOop.hpp:219
#
# An unexpected error has been detected by Java Runtime Environment:
#
#  Internal Error (os_linux_zero.cpp:144), pid=9767, tid=1210827952
#  Error: caught unhandled signal 11
#
# Java VM: OpenJDK Core VM (1.6.0-b09 interpreted mode linux-)
# An error report file with more information is saved as:
# /home/mbaudier/workspace/crashjvm-cglib/bin/hs_err_pid9767.log

Program received signal SIGSEGV, Segmentation fault.
VMError::report (this=0x4823c198, st=0x4823c118)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/interpreterRT_zero.hpp:42
42          return cif()->nargs;
(gdb) finish
Run till exit from #0  VMError::report (this=0x4823c198, st=0x4823c118)
    at
/usr/src/debug/icedtea6-1.2/openjdk/hotspot/src/cpu/zero/vm/interpreterRT_zero.hpp:42
[Thread 0x6ee104b0 (LWP 9774) exited]
[Thread 0x6ed904b0 (LWP 9773) exited]
[Thread 0x6ec904b0 (LWP 9772) exited]
[Thread 0x6e9904b0 (LWP 9771) exited]
[Thread 0x6e8904b0 (LWP 9770) exited]
[Thread 0x6e7904b0 (LWP 9769) exited]
[Thread 0x482bc4b0 (LWP 9768) exited]

Program terminated with signal SIGSEGV, Segmentation fault.
The program no longer exists.


#
# Crash also with GCJ 1.5.0
#
GCJ 1.5.0 also crashes but later on: the object is enhanced, but the JVM
crashes when it is accessed.

GDB output:

(gdb)  run -cp
.:../lib/cglib-2.2.jar:../lib/asm-3.1.jar:../lib/asm-util-3.1.jar
crashjvm.CrashJvm
Starting program: /usr/lib/jvm/java-1.5.0-gcj-1.5.0.0/bin/java -cp
.:../lib/cglib-2.2.jar:../lib/asm-3.1.jar:../lib/asm-util-3.1.jar
crashjvm.CrashJvm
[Thread debugging using libthread_db enabled]
[New Thread 0x48018580 (LWP 9784)]
[New Thread 0x48ced4b0 (LWP 9791)]
Start test.

Program received signal SIGPWR, Power fail/restart.
[Switching to Thread 0x48ced4b0 (LWP 9791)]
0x0f37bf44 in __pthread_cond_wait (cond=<value optimized out>, mutex=<value
optimized out>) at pthread_cond_wait.c:153
153           lll_futex_wait (&cond->__data.__futex, futex_val, pshared);
(gdb) finish
Run till exit from #0  0x0f37bf44 in __pthread_cond_wait (cond=<value optimized
out>, mutex=<value optimized out>)
    at pthread_cond_wait.c:153

Program received signal SIGXCPU, CPU time limit exceeded.
0x0f1b721c in __sigsuspend (set=<value optimized out>) at
../sysdeps/unix/sysv/linux/sigsuspend.c:63
63        return INLINE_SYSCALL (rt_sigsuspend, 2, CHECK_SIGSET (set), _NSIG /
8);
(gdb) finish
Run till exit from #0  0x0f1b721c in __sigsuspend (set=<value optimized out>)
at ../sysdeps/unix/sysv/linux/sigsuspend.c:63
GC_suspend_handler_inner (sig_arg=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:218
218     ../../../boehm-gc/pthread_stop_world.c: No such file or directory.
        in ../../../boehm-gc/pthread_stop_world.c
Value returned is $1 = -1
(gdb) finish
Run till exit from #0  GC_suspend_handler_inner (sig_arg=<value optimized out>)
    at ../../../boehm-gc/pthread_stop_world.c:218
GC_suspend_handler (sig=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:155
155     in ../../../boehm-gc/pthread_stop_world.c
(gdb) finish
Run till exit from #0  GC_suspend_handler (sig=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:155
<signal handler called>
(gdb) finish
Run till exit from #0  <signal handler called>
0x0f37bf40 in __pthread_cond_wait (cond=<value optimized out>, mutex=<value
optimized out>) at pthread_cond_wait.c:153
153           lll_futex_wait (&cond->__data.__futex, futex_val, pshared);
(gdb) finish
Run till exit from #0  0x0f37bf40 in __pthread_cond_wait (cond=<value optimized
out>, mutex=<value optimized out>)
    at pthread_cond_wait.c:153
0x0161331c in _Jv_CondWait (cv=<value optimized out>, mu=<value optimized out>,
millis=<value optimized out>, 
    nanos=<value optimized out>) at ../../../libjava/posix-threads.cc:212
212             r = pthread_cond_wait (&current->wait_cond,
&current->wait_mutex);
Value returned is $2 = 0
Current language:  auto; currently c++
(gdb) finish
Run till exit from #0  0x0161331c in _Jv_CondWait (cv=<value optimized out>,
mu=<value optimized out>, 
    millis=<value optimized out>, nanos=<value optimized out>) at
../../../libjava/posix-threads.cc:212
gnu::gcj::runtime::FinalizerThread::run (this=<value optimized out>)
    at ../../../libjava/gnu/gcj/runtime/natFinalizerThread.cc:58
58            finalizer_ready = false;
Value returned is $3 = 0
(gdb) finish
Run till exit from #0  gnu::gcj::runtime::FinalizerThread::run (this=<value
optimized out>)
    at ../../../libjava/gnu/gcj/runtime/natFinalizerThread.cc:58

Program received signal SIGPWR, Power fail/restart.
0x0f37bf44 in __pthread_cond_wait (cond=<value optimized out>, mutex=<value
optimized out>) at pthread_cond_wait.c:153
153           lll_futex_wait (&cond->__data.__futex, futex_val, pshared);
Current language:  auto; currently c
(gdb) finish
Run till exit from #0  0x0f37bf44 in __pthread_cond_wait (cond=<value optimized
out>, mutex=<value optimized out>)
    at pthread_cond_wait.c:153

Program received signal SIGXCPU, CPU time limit exceeded.
0x0f1b721c in __sigsuspend (set=<value optimized out>) at
../sysdeps/unix/sysv/linux/sigsuspend.c:63
63        return INLINE_SYSCALL (rt_sigsuspend, 2, CHECK_SIGSET (set), _NSIG /
8);
(gdb) finish
Run till exit from #0  0x0f1b721c in __sigsuspend (set=<value optimized out>)
at ../sysdeps/unix/sysv/linux/sigsuspend.c:63
GC_suspend_handler_inner (sig_arg=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:218
218     ../../../boehm-gc/pthread_stop_world.c: No such file or directory.
        in ../../../boehm-gc/pthread_stop_world.c
Value returned is $4 = -1
(gdb) finish
Run till exit from #0  GC_suspend_handler_inner (sig_arg=<value optimized out>)
    at ../../../boehm-gc/pthread_stop_world.c:218
GC_suspend_handler (sig=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:155
155     in ../../../boehm-gc/pthread_stop_world.c
(gdb) finish
Run till exit from #0  GC_suspend_handler (sig=<value optimized out>) at
../../../boehm-gc/pthread_stop_world.c:155
<signal handler called>
(gdb) finish
Run till exit from #0  <signal handler called>
0x0f37bf40 in __pthread_cond_wait (cond=<value optimized out>, mutex=<value
optimized out>) at pthread_cond_wait.c:153
153           lll_futex_wait (&cond->__data.__futex, futex_val, pshared);
(gdb) finish
Run till exit from #0  0x0f37bf40 in __pthread_cond_wait (cond=<value optimized
out>, mutex=<value optimized out>)
    at pthread_cond_wait.c:153
0x0161331c in _Jv_CondWait (cv=<value optimized out>, mu=<value optimized out>,
millis=<value optimized out>, 
    nanos=<value optimized out>) at ../../../libjava/posix-threads.cc:212
212             r = pthread_cond_wait (&current->wait_cond,
&current->wait_mutex);
Value returned is $5 = 0
Current language:  auto; currently c++
(gdb) finish
Run till exit from #0  0x0161331c in _Jv_CondWait (cv=<value optimized out>,
mu=<value optimized out>, 
    millis=<value optimized out>, nanos=<value optimized out>) at
../../../libjava/posix-threads.cc:212
gnu::gcj::runtime::FinalizerThread::run (this=<value optimized out>)
    at ../../../libjava/gnu/gcj/runtime/natFinalizerThread.cc:58
58            finalizer_ready = false;
Value returned is $6 = 0
(gdb) finish
Run till exit from #0  gnu::gcj::runtime::FinalizerThread::run (this=<value
optimized out>)
    at ../../../libjava/gnu/gcj/runtime/natFinalizerThread.cc:58
Enhanced object.

Program received signal SIGSEGV, Segmentation fault.
[Switching to Thread 0x48018580 (LWP 9784)]
ffi_call_SYSV () at ../../../libffi/src/powerpc/sysv.S:41
41              stwux   %r1,%r1,%r4
Current language:  auto; currently asm
(gdb) finish
Run till exit from #0  ffi_call_SYSV () at
../../../libffi/src/powerpc/sysv.S:41

Program received signal SIGSEGV, Segmentation fault.
ffi_call_SYSV () at ../../../libffi/src/powerpc/sysv.S:41
41              stwux   %r1,%r1,%r4
(gdb) finish
Run till exit from #0  ffi_call_SYSV () at
../../../libffi/src/powerpc/sysv.S:41
[Thread 0x48ced4b0 (LWP 9791) exited]

Program terminated with signal SIGSEGV, Segmentation fault.
The program no longer exists.

#
# Trying to build IcedTea
#
I gave another try to compile the latest version of the icedtea6 tree, but with
no more success. Here were my commands (again I am on an up-to-date Fedora Core
9):
./configure --with-gcj-home=/usr/lib/jvm/java-1.5.0-gcj-1.5.0.0
--with-libgcj-jar=/usr/share/java/libgcj-4.3.0.jar
--prefix=/usr/local/icedtea-head --enable-zero
make
(also tried with the patches indicated here:
http://icedtea.classpath.org/wiki/DebianBuildingInstructions#Building_Icedtea_on_Debian_.28Lenny.29_PowerPC)

But I keep errors (after plenty of conversion warnings when building in
icedtea6/openjdk-ecj/control/build/linux-ppc/hotspot/outputdir/linux_zero_core/product)

Please tell me if it useful that I keep trying building it, so that I can send
a message to the mailing-list for support. (I don't want to clutter this bug
with build issues)

# Reproducing with a minimal code
package crashjvm;

import java.lang.reflect.Method;

import net.sf.cglib.proxy.Enhancer;
import net.sf.cglib.proxy.MethodInterceptor;
import net.sf.cglib.proxy.MethodProxy;

public class CrashJvm {

    public static void main(String[] args) {
        Object obj = Enhancer.create(Object.class, new MethodInterceptor() {
            public Object intercept(Object obj, Method method, Object[] args,
                    MethodProxy proxy) throws Throwable {
                return obj;
            }
        });

    }

}


-- 
Configure bugmail: http://icedtea.classpath.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are the assignee for the bug, or are watching the assignee.

More information about the distro-pkg-dev mailing list