This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: A couple of testcases from GNATS
"Joseph S. Myers" <jsm28@cam.ac.uk> writes:
> +int
> +main (void)
> +{
> + {
> + char s[] = "x";
> + }
> + if (foo ())
> + return 0;
> + else
> + abort ();
> +}
> --- gcc.c-torture/execute/20000801-4.x.orig Thu Jan 1 00:00:00 1970
> +++ gcc.c-torture/execute/20000801-4.x Tue Aug 1 19:05:04 2000
> @@ -0,0 +1,2 @@
> +set torture_execute_xfail "*-*-*"
> +return 0
Hi Joseph,
I already mentioned that you should use exit(0) rather than return 0.
Looking at your testcases more closely, I notice that they don't seem
to guarantee failure. It seems quite likely that on one or more
targets they will pass by accident. In that case, could you mention
in the .x file which hosts are known to fail? Otherwise, I'll see
that these are XPASS on some host, and all I can do is delete the .x
file, because I don't know on which hosts they fail.
--
- Geoffrey Keating <geoffk@cygnus.com>