java/6092: sparc-sun-solaris2.7 has hundreds of libjava failu res with -m64
Boehm, Hans
hans_boehm@hp.com
Thu Apr 11 11:46:00 GMT 2002
The following reply was made to PR java/6092; it has been noted by GNATS.
From: "Boehm, Hans" <hans_boehm@hp.com>
To: "'tromey@redhat.com'" <tromey@redhat.com>,
"Kaveh R. Ghazi"
<ghazi@caip.rutgers.edu>
Cc: gcc-gnats@gcc.gnu.org, "Boehm, Hans" <hans_boehm@hp.com>
Subject: RE: java/6092: sparc-sun-solaris2.7 has hundreds of libjava failu
res with -m64
Date: Thu, 11 Apr 2002 11:28:17 -0700
Tom -
That line in GC_stopped_mark calls GC_mark_some. Thus I would conclude that
this version of gdb is only partially functional, and it actually dies
somewhere in GC_mark_some().
A possible cause of that is confusion about where the roots are. If you can
check GC_stackbottom, check that the collector and libgcj configuration
agree on threads, get GC_dump() output, and check the root locations against
the nm output that might identify something. Otherwise we need better debug
information.
Hans
> -----Original Message-----
> From: Tom Tromey [mailto:tromey@redhat.com]
>
> I found a gdb here that seems to know how to debug on this machine.
> It gives me the appended stack trace.
>
> Hans, any clues you can give me on debugging this would be helpful.
>
> Tom
>
>
> (gdb) bt
> #0 0xfffffeab7ed6fd50 in ?? ()
> #1 0xffffffff7ed68f20 in GC_stopped_mark (
> stop_func=0xffffffff7ed68598 <GC_never_stop_func>)
> at ../../../../gcc/boehm-gc/alloc.c:489
> #2 0xffffffff7ed68b88 in GC_try_to_collect_inner (
> stop_func=0xffffffff7ed68598 <GC_never_stop_func>)
> at ../../../../gcc/boehm-gc/alloc.c:350
> #3 0xffffffff7ed72de0 in GC_init_inner ()
> at ../../../../gcc/boehm-gc/misc.c:673
> #4 0xffffffff7ed72a60 in GC_init () at
> ../../../../gcc/boehm-gc/misc.c:455
> #5 0xffffffff7ed6cce4 in GC_init_gcj_malloc (mp_index=32752,
> mp=0xffffffff7ed655e4) at ../../../../gcc/boehm-gc/gcj_mlc.c:62
> #6 0xffffffff7ed6636c in _Jv_InitGC() ()
> at ../../../../gcc/libjava/boehm.cc:465
> #7 0xffffffff7ebe6520 in _Jv_CreateJavaVM(void*) ()
> at ../../../../gcc/libjava/prims.cc:892
> #8 0xffffffff7ebe69b0 in _Jv_RunMain(java::lang::Class*,
> char const*, int, char const**, bool) (klass=0x100102518,
> name=0x0, argc=1, argv=0xffffffff7ffff718,
> is_jar=false) at ../../../../gcc/libjava/prims.cc:982
> #9 0x100001820 in main ()
>
More information about the Gcc-prs
mailing list