This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: i386 byteswap.h provoking ``may be undefined'' warnings in gcc3
- To: Paul Haahr <paul at paulhaahr dot com>
- Subject: Re: i386 byteswap.h provoking ``may be undefined'' warnings in gcc3
- From: Richard Henderson <rth at redhat dot com>
- Date: Tue, 17 Jul 2001 01:29:10 -0700
- Cc: Andreas Jaeger <aj at suse dot de>, bug-glibc at gnu dot org, gcc-bugs at gcc dot gnu dot org
- References: <8XjdCSp-3@iadd.jivetech.com> <u8d7763f9d.fsf@gromit.moeb> <cXjn9J6COR@iadd.jivetech.com>
On Thu, Jul 12, 2001 at 02:48:29AM -0700, Paul Haahr wrote:
> That's still not clear -- it's hard to argue that, just because some
> code is in a provably untaken branch of an if, it shouldn't be checked
> for warnings. And special casing __builtin_constant_p seems problematic.
It's probably desirable to special case __b_c_p in this way,
as that's part of the point of its existance. However, I'm
not planning on doing anything about this right now.
Feel free to contribute a patch however. ;-)
r~