This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: [PATCH] Fold isnan to UNORDERED_EXPR
- From: Roger Sayle <roger at eyesopen dot com>
- To: Paolo Bonzini <paolo dot bonzini at polimi dot it>
- Cc: gcc-patches at gcc dot gnu dot org
- Date: Wed, 30 Jun 2004 13:28:30 -0600 (MDT)
- Subject: Re: [PATCH] Fold isnan to UNORDERED_EXPR
On Wed, 30 Jun 2004, Paolo Bonzini wrote:
> 2004-06-30 Paolo Bonzini <bonzini@gnu.org>
>
> * builtins.c (fold_builtin_classify): New.
> (fold_builtin_unordered_cmp): Fix commentary.
> (fold_builtin): Use it.
> * builtins.def: Define builtins for isnan,
> isnanf, isnanl, finite, finitef, finitel,
> isinf, isinff, isinfl.
This is OK for mainline. Is there any particular reason why isnanf
and isnanl are defined as varargs builtins, but that the float and
long double forms of finite and isinf are fully prototyped? I agree
that the double forms should use varargs, but I'd prefer to be
consistent and use explicit prototypes for foof and fool. (:>)
> 2004-06-30 Paolo Bonzini <bonzini@gnu.org>
>
> * builtins-43.c, builtins-44.c, builtins-45.c: New.
For gcc/testsuite/ChangeLog entries you'll need to include the full
path to the testcase, as we have several subdirectories that contain
builtins-?.c testcases.
This is probably better written as
> 2004-06-30 Paolo Bonzini <bonzini@gnu.org>
>
> * gcc.dg/builtins-43.c: New test case.
> * gcc.dg/builtins-44.c: Likewise.
> * gcc.dg/builtins-45.c: Likewise.
Thanks,
Roger
--