hgcd1/2

[Niels Möller]

tg at gmplib.org (Torbjörn Granlund) writes:

> tg at gmplib.org (Torbjörn Granlund) writes:
>
>   I cooked a modern alternative:
>
> I went ahead and committed that version, replacing the old
> HGCD2_METHOD=2.  I expect it is be the fastest method on some platform.

Will be interesting to see results on thresholds.

> (We might want to arrange for longlong.h to use lzcnt instead of bsr for
> modern AMD processors; the initial two count_leading_zeros would
> terminate in one cycle instead of 8 thereby!)

Looks like you did that too.

I've now tried the similar #if:ed out div2 code, and enabling it gives
an 8% speedup on my laptop.

Next, I think we should go ahead with the rename HGCD2_METHOD to
DIV11_METHOD or possibly HGCD2_DIV1_METHOD.

Regards,
/Niels

-- 
Niels Möller. PGP-encrypted email is preferred. Keyid 368C6677.
Internet email is subject to wholesale government surveillance.