This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: fix for __builtin_size_t
- From: "Joseph S. Myers" <jsm at polyomino dot org dot uk>
- To: Jan Beulich <JBeulich at novell dot com>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Wed, 7 Jul 2004 12:28:51 +0000 (UTC)
- Subject: Re: fix for __builtin_size_t
- References: <s0ebf63d.090@emea1-mh.id2.novell.com>
On Wed, 7 Jul 2004, Jan Beulich wrote:
> bootstrapped and tested on x86_64-unknown-linux-gnu.
>
> 2004-07-07 Jan Beulich <jbeulich@novell.com>
>
> * c-common.c (c_common_nodes_and_builtins): Use size_type_node
> instead of sizenode for __builtin_size_t.
Why not just get rid of __builtin_size_t and __builtin_ptrdiff_t? The
reason given in <http://gcc.gnu.org/ml/gcc-patches/1999-11n/msg00422.html>
for their introduction no longer applies now that __SIZE_TYPE__ and
__PTRDIFF_TYPE__ are defined within cc1 (and standalone cpp is just a mode
of cc1/c11plus) rather than needing to be defined from specs, and no other
reason for them has ever been stated. They are undocumented, so we've
made no guarantee of them as an interface external software can rely on
staying around, and are used nowhere in GCC (even in the testsuite), or in
glibc. (We also create a huge number of tree nodes for builtins at
startup; getting rid of these two won't make much difference, but they do
seem completely unnecessary.)
--
Joseph S. Myers http://www.srcf.ucam.org/~jsm28/gcc/
jsm@polyomino.org.uk (personal mail)
jsm28@gcc.gnu.org (Bugzilla assignments and CCs)