This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: c/6815: [3.2/3.3 regression] ICE with longjmp if -fprofile-arcs -O is specified
- From: Richard Henderson <rth at redhat dot com>
- To: Christian Ehrhardt <ehrhardt at mathematik dot uni-ulm dot de>
- Cc: gcc-gnats at gcc dot gnu dot org, gcc-prs at gcc dot gnu dot org, gcc-bugs at gcc dot gnu dot org, nobody at gcc dot gnu dot org, dtucker at zip dot com dot au
- Date: Thu, 12 Dec 2002 10:08:35 -0800
- Subject: Re: c/6815: [3.2/3.3 regression] ICE with longjmp if -fprofile-arcs -O is specified
- References: <20021212123726.14013.qmail@theseus.mathematik.uni-ulm.de>
On Thu, Dec 12, 2002 at 01:37:26PM +0100, Christian Ehrhardt wrote:
> where e is NULL. insert_insn_on_edge in turn is called from
> flow_call_edges_add in cfganal.c:325 with this piece of code:
>
> | for (e = bb->succ; e; e = e->succ_next)
> | if (e->dest == EXIT_BLOCK_PTR)
> | break;
> |
> | insert_insn_on_edge (gen_rtx_USE (VOIDmode, const0_rtx), e);
> | commit_edge_insertions ();
>
> which looks a bit strange if we keep in mind that insert_insn_on_edge
> will crash if the second parameter is NULL.
Indeed, this piece of code cannot handle a noreturn function
in the last block. The following should fix it.
r~
Index: cfganal.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/cfganal.c,v
retrieving revision 1.30
diff -c -p -d -u -r1.30 cfganal.c
--- cfganal.c 11 Oct 2002 21:09:59 -0000 1.30
+++ cfganal.c 12 Dec 2002 18:07:41 -0000
@@ -320,10 +320,11 @@ flow_call_edges_add (blocks)
for (e = bb->succ; e; e = e->succ_next)
if (e->dest == EXIT_BLOCK_PTR)
- break;
-
- insert_insn_on_edge (gen_rtx_USE (VOIDmode, const0_rtx), e);
- commit_edge_insertions ();
+ {
+ insert_insn_on_edge (gen_rtx_USE (VOIDmode, const0_rtx), e);
+ commit_edge_insertions ();
+ break;
+ }
}
}