[Bug rtl-optimization/50030] sin optimized in sinf at -O2

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=50030

Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID

--- Comment #3 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-08-10 09:16:24 UTC ---
Because exp -> expf is not always valid but sin -> sinf is (there are no
exceptional output values for sin but there are for exp).

float  sind(float x) {
   return sinl(x);
}

use a long double argument / result.