Check libjava completes now, but leaves spurious processes

Alexandre Petit-Bianco apbianco@cygnus.com
Mon Apr 16 16:36:00 GMT 2001


Tom Tromey <tromey@redhat.com> writes:

> I don't have any good way to debug this, since I can't reproduce
> this failure.

I've been seeing this last since early last week, when my nightly
builds started to hang. I did an update this morning: both the
generated executable and the VM hang after completion of the
program. Here's where the native executable hangs:

  #0  0x08100fa6 in __sigsuspend (set=0xbf7ff85c) at ../sysdeps/unix/sysv/linux/sigsuspend.c:45
  45	../sysdeps/unix/sysv/linux/sigsuspend.c: No such file or directory.
  	in ../sysdeps/unix/sysv/linux/sigsuspend.c
  (gdb) up
  #1  0x080ec2e9 in __pthread_wait_for_restart_signal (self=0xbf7ffc00) at pthread.c:926
  926	pthread.c: No such file or directory.
  	in pthread.c
  (gdb) 
  #2  0x080edd09 in __pthread_alt_lock (lock=0x817fae8, self=0x0) at restart.h:34
  34	restart.h: No such file or directory.
  	in restart.h
  (gdb) 
  #3  0x080eaf82 in __pthread_mutex_lock (mutex=0x817fad8) at mutex.c:120
  120	mutex.c: No such file or directory.
  	in mutex.c
  (gdb) 
  #4  0x0808a29c in _Jv_CondNotifyAll(_Jv_ConditionVariable_t*, _Jv_Mutex_t*) (cv=0x8257e80, mu=0x8257e60)
      at /home/apbianco/src/egcs-devel/libjava/posix-threads.cc:243
  243	      pthread_mutex_lock (&target->wait_mutex);

As for the branch, gij segvs on hello world:

  program received signal SIGSEGV, Segmentation fault.
  [Switching to Thread 1024 (LWP 26598)]
  0x4015d384 in _Jv_equalUtf8Consts(_Jv_Utf8Const*, _Jv_Utf8Const*) (a=0x0, b=0x402a0dc2)
      at /home/apbianco/src/gcc-3.0-devel/libjava/prims.cc:149
  149	  if (a->hash != b->hash)
  (gdb) p a
  $1 = (Utf8Const *) 0x0
  (gdb) up
  #1  0x4017741b in _Jv_GetMethodLocal(java::lang::Class*, _Jv_Utf8Const*, _Jv_Utf8Const*) (klass=0x4030d480, name=0x0, 
      signature=0x0) at /home/apbianco/src/gcc-3.0-devel/libjava/java/lang/natClass.cc:800
  800	      if (_Jv_equalUtf8Consts (name, klass->methods[i].name)
  (gdb) up
  #2  0x40177289 in java::lang::Class::initializeClass() (this=0x4030d480)
      at /home/apbianco/src/gcc-3.0-devel/libjava/java/lang/natClass.cc:755
  755	      _Jv_Method *meth = _Jv_GetMethodLocal (this, clinit_name,
  (gdb) p clinit_name
  $2 = (_Jv_Utf8Const *) 0x0

(But this one might be a compiler problem, I was experimenting with a
patch, I'll keep you posted on that one.)

./A



More information about the Gcc mailing list