This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Problem in split_basic_block
- From: Olivier Hainque <hainque at act-europe dot fr>
- To: Richard Henderson <rth at redhat dot com>,Richard Kenner <kenner at vlsi1 dot ultra dot nyu dot edu>, gcc at gcc dot gnu dot org
- Cc: hainque at act-europe dot fr
- Date: Mon, 1 Sep 2003 15:18:14 +0200
- Subject: Re: Problem in split_basic_block
- References: <10308062258.AA03807@vlsi1.ultra.nyu.edu> <20030808211916.GC4310@redhat.com>
Hello,
Richard Henderson wrote:
> [sherman:/scratch/rth/build/vms/gcc/ada] ../gnat1 -O2 [...] a-calend.adb
>
> So was I expecting a crash, or something?
We now understand why you did not see the crash with Ada, and eventually came
up with a tiny C++ testcase exhibiting the issue Richard Kenner raised.
Using the same compiler, could you please try to compile the reproducer below
with -fnon-call-exceptions:
// foo.cc
void hook (void)
{
}
void foo (void)
{
int x, y;
try {
y = 0;
} catch (...) {
hook ();
}
}
We are seeing this:
$ ./cc1plus -fnon-call-exceptions foo.cc
void hook()
void hook()
void foo()
void foo()
foo.cc:15: error: Call edges for non-call insn in bb 9
foo.cc:15: internal compiler error: verify_flow_info failed
Please submit a full bug report,
with preprocessed source if appropriate.
See <URL:http://gcc.gnu.org/bugs.html> for instructions.
Thanks in advance,
Olivier