This is the mail archive of the
fortran@gcc.gnu.org
mailing list for the GNU Fortran project.
Re: [committed] speed up GC for fortran
- From: Geoffrey Keating <geoffk at geoffk dot org>
- To: Andrew Pinski <pinskia at physics dot uc dot edu>
- Cc: gfortran <fortran at gcc dot gnu dot org>, gcc-patches at gcc dot gnu dot org
- Date: 30 Oct 2004 13:06:06 -0700
- Subject: Re: [committed] speed up GC for fortran
- References: <649DB2EB-2A83-11D9-9310-000A95D692F4@physics.uc.edu>
Andrew Pinski <pinskia@physics.uc.edu> writes:
> This is the patch which should/will speed up GC for the fortran
> front-end,
Have you actually timed it?
When I created chain_next in the first place, I found that the
compiler went slower, not faster. It's there to avoid excessive stack
use, not for speed.
> corresponding code has been in the other front-ends for a long time. It
> also should get rid of some of the large function call overhead for the
> GC in the fortran front-end. This also should fix the ICE which Paul
> was seeing when compiling one of the SPEC benchmark at -O1.
>
> Committed as pre-approved by Steven B.
>
> Thanks,
> Andrew Pinski
>
> ChangeLog:
>
> * f95-lang.c (lang_tree_node): Add chain_next to be the TREE_CHAIN.