| Message ID | 2A7D0E8D-951E-4067-B128-2D605CC82233@gmail.com |
|---|---|
| State | New |
| Headers | show |
On 06/09/2010 11:40 AM, FX wrote: > OK to commit and close PR43040? > OK and thanks for checking. Tobias PS: The trunk is now open again for commits according to #gcc. > 2010-06-09 Francois-Xavier Coudert <fxcoudert@gcc.gnu.org> > > PR fortran/43040 > * f95-lang.c (gfc_init_builtin_functions): Remove comment. >
Index: f95-lang.c =================================================================== --- f95-lang.c (revision 160457) +++ f95-lang.c (working copy) @@ -755,8 +755,6 @@ gfc_init_builtin_functions (void) #include "mathbuiltins.def" - /* We define these separately as the fortran versions have different - semantics (they return an integer type) */ gfc_define_builtin ("__builtin_roundl", mfunc_longdouble[0], BUILT_IN_ROUNDL, "roundl", true); gfc_define_builtin ("__builtin_round", mfunc_double[0],
Since I'm currently working on the topic of builtins in the Fortran front-end: To close PR43040 (http://gcc.gnu.org/bugzilla/show_bug.cgi?id=43040), I've checked that the comment below in f95-lang.c is incorrect. We do indeed define the family of round() builtins with the same prototype as in C, so there is not a problem. OK to commit and close PR43040? FX 2010-06-09 Francois-Xavier Coudert <fxcoudert@gcc.gnu.org> PR fortran/43040 * f95-lang.c (gfc_init_builtin_functions): Remove comment.