Recent changes to mpn_get_str/mpn_set_str

[Marco Bodrato]

Ciao,

Il Ven, 27 Gennaio 2017 12:27 am, Torbjörn Granlund ha scritto:
> "Marco Bodrato" <bodrato at mail.dm.unipi.it> writes:
>
>   We may also try the division-free algorithm explored by
> Bouvier-Zimmermann
>   in their paper: https://members.loria.fr/PZimmermann/papers/get_str.pdf

> I found the old code implementing the basecase.  This code is very
> likely the basis for the conclusion in the GMP manual where we say this
> is not an improvement.  (See the quote in the paper, section 2.2.)

Yes, they properly reference the source of the idea :-)

> I'll see if I can brush up the code and make use of it for machines with
> pipelined multiply.

If it extracts digits limb by limb, then progressively shorten the buffer
used by the mul_1 phase, I believe it will reach the performances claimed
by the two authors.

Unfortunately I do not have access neither to your old code nor to the
code of the two authors... and I won't write a third implementation from
scratch... not immediately at least :-)

Regards,
m

-- 
http://bodrato.it/