This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
RE: gcc throwing an internal error only on HP-UX ?
- To: "'Harri Pasanen'" <harri dot pasanen at trema dot com>, <gcc at gcc dot gnu dot org>
- Subject: RE: gcc throwing an internal error only on HP-UX ?
- From: "Jessica Han" <jessica at cup dot hp dot com>
- Date: Tue, 2 Oct 2001 10:34:53 -0700
- Reply-To: <jessica at cup dot hp dot com>
first_reachable is 0, and cfun->eh->region_array[first_reachable] is null.
Is it possible the problem is fixed in 3.0.2?
-----Original Message-----
From: Harri Pasanen [mailto:harri.pasanen@trema.com]
Sent: Tuesday, October 02, 2001 7:11 AM
To: gcc@gcc.gnu.org
Subject: Re: gcc throwing an internal error only on HP-UX ?
took a look with gdb to see where about it is crashing, hopefully someone
can guess what might be wrong:
(gdb) set args -quiet poa.ii
(gdb) r
Starting program: /usr/gnu/lib/gcc-lib/hppa2.0w-hp-hpux11.00/3.0.2/cc1plus
-quiet poa.ii
Program received signal SIGSEGV, Segmentation fault.
0xf6934 in sjlj_emit_dispatch_table (dispatch_label=0x40204938,
lp_info=0x40204938) at ../../gcc/gcc/except.c:2324
2324 emit_insns_before (seq, (cfun->eh->region_array[first_reachable]
(gdb) bt
#0 0xf6934 in sjlj_emit_dispatch_table (dispatch_label=0x40204938,
lp_info=0x40204938) at ../../gcc/gcc/except.c:2324
#1 0xf6a24 in sjlj_build_landing_pads () at ../../gcc/gcc/except.c:2349
#2 0xf6b08 in finish_eh_generation () at ../../gcc/gcc/except.c:2387
#3 0xbe934 in rest_of_compilation (decl=0x7a74c9c0)
at ../../gcc/gcc/toplev.c:2936
#4 0x9cbe4 in genrtl_finish_function (fn=0x7a6b1660)
at ../../gcc/gcc/cp/semantics.c:2719
#5 0x9c778 in expand_body (fn=0x7a74c9c0) at
../../gcc/gcc/cp/semantics.c:2505
#6 0xf6938 in sjlj_emit_dispatch_table (dispatch_label=0x4002d3c8,
lp_info=0x64c) at ../../gcc/gcc/except.c:2324
(gdb) list
2319 }
2320
2321 seq = get_insns ();
2322 end_sequence ();
2323
2324 emit_insns_before (seq, (cfun->eh->region_array[first_reachable]
2325 ->post_landing_pad));
2326 }
2327
Unfortunately I can't examine cfun with gdb, but my guess is that some
pointer starting from cfun above is invalid.
Regards,
Harri