mpn_sqrtrem{1,2}

[Marco Bodrato]

Ciao,

>> I'll be grateful if you run your measures again,
>> in particular for 100 and 128 bits.

I selected only the size affected by my last change. (I assume your
scenario is ABI=64)

Il Gio, 9 Marzo 2017 11:49 pm, Adrien Prost-Boucle ha scritto:
> Laptop Core 2 Duo 2GHz

>  128 bits   time  8.43 ->  5.65   speedup 1.49
>   96 bits   time  9.29 ->  6.27   speedup 1.48

> PC i7-4790 3.60GHz

>  128 bits   time  2.42 ->  1.41   speedup 1.72
>   96 bits   time  3.16 ->  1.89   speedup 1.67

>> Now, I believe he will be able to gain another 2x improving sqrtrem2,
> Not 2x, but the potential speedup is much higher indeed.

Not even my change really gave 2x... but, before it, you could obtain
20-30%, now you measure 50-70%... that's the right direction :-)

-- 
http://bodrato.it/papers/