[Patch, Fortran] Another floating point speedup
Steve Kargl
sgk@troutmask.apl.washington.edu
Mon Jan 5 06:12:00 GMT 2009
On Sat, Jan 03, 2009 at 08:52:18PM +0100, Daniel Kraft wrote:
> Once again... This patch should finally be rid of the bugs (no
> regressions) and handles exponents of size up to 4 digits (should be
> enough).
>
> In addition, I got rid of the memcpy in convert_real and use
> pointer-casting instead, got another second of speed up.
>
> Ok for 4.5? Or should I first look at the list_read.c floating parsing
> routines (I guess those are for unformatted IO?)
>
Daniel
With your patch, I'm seeing speed improvements without regressions
in my very limited testing on i686-*-freebsd.
4.3.3 trunk
test.f90 40.0 33.1 17.5% (Your test.f90 program)
shell.f 5.06 4.78 5.5% (One of my old codes)
I modified shell.f to use an E28.15 format to get 15 digits.
There are no differences in a comparison of output file for
the 4.3.3 binrary compared to a patched trunk.
--
Steve
More information about the Gcc-patches
mailing list