Toon Moene wrote:
I can measure the contribution of the cbrt effect in isolation, though
(given the above change of HIRLAM source code).
Well, the effect of the cbrt change (X ** (1./3.) -> cbrt(X)) is close
to zero.
Some other change must have diminished the number of pow[f] calls
substantially. I think Andrew P's patch to declare the Fortran pow
functions constant might be a good candidate.
[ I did this exercise for g77 in 2000 - I'd never thought that that
wouldn't be the basis for people to work on gfortran ... ]
Cheers,
--
Toon Moene - e-mail: [EMAIL PROTECTED] - phone: +31 346 214290
Saturnushof 14, 3738 XG Maartensdijk, The Netherlands
A maintainer of GNU Fortran: http://gcc.gnu.org/fortran/
Who's working on GNU Fortran:
http://gcc.gnu.org/ml/gcc/2006-01/msg00000.html