This is the mail archive of the
gcc@gcc.gnu.org
mailing list for the GCC project.
Re: Handling of inlining broken in mainline
- From: Andreas Jaeger <aj at suse dot de>
- To: gcc at gcc dot gnu dot org
- Date: Sat, 05 Oct 2002 22:21:15 +0200
- Subject: Re: Handling of inlining broken in mainline
- References: <u83crldnyp.fsf@gromit.moeb>
Andreas Jaeger <aj@suse.de> writes:
> Currently GCC mainline seems to be broken regarding inlining of
> functions. It defines always __NO_INLINE__ indepenend of the options give:
>
> gromit:~/tmp:[0]$ /opt/gcc/3.3-devel/bin/gcc -O1 -finline -dD -E t.c |grep INLINE
> #define __NO_INLINE__ 1
> gromit:~/tmp:[0]$ /opt/gcc/3.3-devel/bin/gcc -O1 -finline-functions -dD -E t.c |grep INLINE
> #define __NO_INLINE__ 1
> gromit:~/tmp:[0]$ /opt/gcc/3.3-devel/bin/gcc -O3 -dD -E t.c |grep INLINE
> #define __NO_INLINE__ 1
> gromit:~/tmp:[0]$ /opt/gcc/3.3-devel/bin/gcc -O3 -fno-inline -dD -E t.c |grep INLINE
> #define __NO_INLINE__ 1
>
> With gcc 3.2 everything is as expected:
>
> gromit:~/tmp:[0]$ /opt/gcc/3.2-devel/bin/gcc -O3 -dD -E t.c |grep INLINE
> gromit:~/tmp:[1]$ /opt/gcc/3.2-devel/bin/gcc -O3 -fno-inline -dD -E t.c |grep INLINE
> #define __NO_INLINE__ 1
Just for reference: Honza committed a patch for mainline to fix this,
thanks,
Andreas
--
Andreas Jaeger
SuSE Labs aj@suse.de
private aj@arthur.inka.de
http://www.suse.de/~aj