This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Gcc 3.0, printf() and exim-3.30
- To: Russ Allbery <rra at stanford dot edu>
- Subject: Re: Gcc 3.0, printf() and exim-3.30
- From: Nix <nix at esperi dot demon dot co dot uk>
- Date: 21 Jun 2001 22:48:10 +0100
- Cc: Vadim Vygonets <vadik at vygo dot net>, gcc at gcc dot gnu dot org
- References: <20010619161158.F9339@vygo.net> <87bsnihitd.fsf@loki.wkstn.nix><ylzob2b4xw.fsf@windlord.stanford.edu>
On 21 Jun 2001, Russ Allbery said:
> (Out of curiosity, is that little aid to GCC as braindead as it looks?
> I'd think that if GCC optimized fprintf, it would pick up printf too....)
Looks to me like it does indeed do that:
,----[ gcc/c-common.c:c_common_nodes_and_builtins() ]
| builtin_function_2 ("__builtin_printf", "printf",
| printf_ftype, printf_ftype,
| BUILT_IN_PRINTF, BUILT_IN_FRONTEND, 1, 0, 0);
| builtin_function_2 ("__builtin_fprintf", "fprintf",
| fprintf_ftype, fprintf_ftype,
| BUILT_IN_FPRINTF, BUILT_IN_FRONTEND, 1, 0, 0);
`----
(But, of course, on the evidence of the last week, my observational
ability right now is slightly less than that of a drowned gnat...
so I may be wrong.)
--
`This will immediately become a flamewar.' --- Mark Mitchell