This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
__NO_INLINE__
- From: Jan Hubicka <jh at suse dot cz>
- To: Andreas Jaeger <aj at suse dot de>, gcc-patches at gcc dot gnu dot org
- Cc: Jan Hubicka <jh at suse dot cz>
- Date: Sat, 5 Oct 2002 19:44:37 +0200
- Subject: __NO_INLINE__
- References: <20021004233620.B906E1B9D5@monteverdi.suse.de> <20021004234825.GA15355@kam.mff.cuni.cz> <u8bs69dokj.fsf@gromit.moeb>
> Jan Hubicka <jh@suse.cz> writes:
>
> > Hi,
> > this looks bogus. There should be change at least in mesa binary.
>
> The problem is that -O3 -ffast-math defines:
> #define __NO_INLINE__ 1
>
> And in this case mathinlines.h is not used at all. What's going on
> with GCC 3.3 that it is defining this? Can you look into it, please?
Hi, I've just commited the attached fix as obivous.
Bootstrapped i586/Linux
Sat Oct 5 19:42:20 CEST 2002 Jan Hubicka <jh@suse.cz>
* c-common.c (cb_register_builtins): Use really_no_inline.
Index: c-common.c
===================================================================
RCS file: /cvs/gcc/egcs/gcc/c-common.c,v
retrieving revision 1.385
diff -c -3 -p -r1.385 c-common.c
*** c-common.c 27 Sep 2002 13:30:05 -0000 1.385
--- c-common.c 5 Oct 2002 17:42:13 -0000
*************** cb_register_builtins (pfile)
*** 4951,4957 ****
if (fast_math_flags_set_p ())
cpp_define (pfile, "__FAST_MATH__");
! if (flag_no_inline)
cpp_define (pfile, "__NO_INLINE__");
if (flag_signaling_nans)
cpp_define (pfile, "__SUPPORT_SNAN__");
--- 4951,4957 ----
if (fast_math_flags_set_p ())
cpp_define (pfile, "__FAST_MATH__");
! if (flag_really_no_inline)
cpp_define (pfile, "__NO_INLINE__");
if (flag_signaling_nans)
cpp_define (pfile, "__SUPPORT_SNAN__");