Shared toom evaluation functions
Paul.Zimmermann at loria.fr
Thu Nov 19 16:05:23 CET 2009
> nisse at lysator.liu.se (Niels Möller) writes:
> Torbjorn Granlund <tg at gmplib.org> writes:
> > I think we might want to make our lives in toom more difficult:
> > We should add, to gmp-mparam.h, cycle counts for available mpn functions
> > as first degree polynomials. Measurements for that can easily be added
> > to tuneup.c.
> I was about to suggest something similar, but a little simpler, with
> only a cycles/limb value for each of the O(n) functions. Do you think
> the constant overhead is important here?
> Perhaps not here, but it will be useful elsewhere. It is not hard to
> measure one paramater when we measure the other.
I personally like this idea of measuring the cycles as a+bn instead of
just bn. We might extend it to a+bn+cn^2 for say mpn_mul_n_basecase.
More information about the gmp-devel