[Gc] segfault with CACAO on OpenSolaris
hans.boehm at hp.com
Mon Aug 18 13:38:35 PDT 2008
In the GC7.1 tar file, line 179 in thread_local_alloc.c is a GC_err_printf call that isn't normally compiled in for a standard build. Can you get the source context, and some information about which access is faulting?
> -----Original Message-----
> From: gc-bounces at napali.hpl.hp.com
> [mailto:gc-bounces at napali.hpl.hp.com] On Behalf Of Christian Thalinger
> Sent: Friday, August 15, 2008 4:45 AM
> To: gc ml
> Subject: [Gc] segfault with CACAO on OpenSolaris
> I tried to get CACAO, which uses Boehm-GC 7.1, running on my
> x86_64 OpenSolaris box and sometimes the GC crashes like this:
> ---- called from signal handler with signal 11 (SIGSEGV) ------
>  GC_malloc(bytes = 32U), line 179 in "thread_local_alloc.c"
>  heap_alloc(size = 32U, references = 524288, finalizer
> = (nil), collect = true), line 153 in "gc-boehm.cpp"
>  builtin_new(c = 0x48cb00), line 879 in "builtin.c"
>  jni_NewObject(env = 0x413e60, clazz = 0x48cb00,
> methodID = 0x9bad40, ...), line 1301 in "jni.cpp"
>  Java_java_lang_VMProcess_nativeSpawn(env =
> 0xfffffd7ffee22200, this = 0x575060, cmdArray = 0x594a80,
> envArray = (nil), dirFile = 0xb, redirect = '\0'), line 235
> in "java_lang_VMProcess.c"
> A lot of program work though. It seems it's somehow related
> to JNI calls, but that's just a guess. The same programs
> work on x86_64 GNU/Linux compiled from the same source code base.
> Any idea what could be the problem?
> - twisti
> Gc mailing list
> Gc at linux.hpl.hp.com
More information about the Gc