This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Inline DOT_PRODUCT revisited
- From: Steven Bosscher <stevenb dot gcc at gmail dot com>
- To: gcc-patches at gcc dot gnu dot org
- Cc: Paul Thomas <paulthomas2 at wanadoo dot fr>, "'fortran at gcc dot gnu dot org'" <fortran at gcc dot gnu dot org>, Dominique Dhumieres <dominiq at lps dot ens dot fr>
- Date: Sat, 25 Feb 2006 23:47:57 +0100
- Subject: Re: Inline DOT_PRODUCT revisited
- References: <43730120.7030808@wanadoo.fr> <200602252337.39067.steven@gcc.gnu.org>
On Saturday 25 February 2006 23:37, Steven Bosscher wrote:
> I would _very_ much like to have this patch in GCC 4.1.1
...the motivation being, of course, that the almost industry-standard
Fortran benchmark (Polyhedron) benefits from this patch tremendeously,
as does a lot of real-world code because DOT_PRODUCT is a very popular
intrinsic.
Timings before the patch (the numbers are proeduced by the test case):
DOT_PRODUCT test
array length time(ns)
4 11700.00
8 10992.00
16 10589.00
32 9268.00
64 9734.00
128 10751.00
256 12929.00
Timings _with_ the patch applied:
DOT_PRODUCT test
array length time(ns)
4 2472.00
8 2513.00
16 2541.00
32 2691.00
64 2920.00
128 3315.00
256 4240.00
Gr.
Steven