1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
/*
Copyright (C) 2009, 2010, 2015 William Hart
This file is part of FLINT.
FLINT is free software: you can redistribute it and/or modify it under
the terms of the GNU Lesser General Public License (LGPL) as published
by the Free Software Foundation; either version 3 of the License, or
(at your option) any later version. See <https://www.gnu.org/licenses/>.
*/
#include "ulong_extras.h"
/*
Method of Niels Moller and Torbjorn Granlund, see "Improved Division by
Invariant Integers: (Algorithm 4)
https://gmplib.org/~tege/division-paper.pdf
*/
#define r_shift(in, c) (((c) == FLINT_BITS) ? WORD(0) : ((in) >> (c)))
ulong
n_lll_mod_preinv(ulong a_hi, ulong a_mi, ulong a_lo, ulong n, ulong ninv)
{
ulong q0, q1, r, norm;
norm = flint_clz(n);
n <<= norm;
/*
a_hi is already reduced, so first reduce a_hi, a_mi mod n
*/
{
const ulong u1 = (a_hi << norm) + r_shift(a_mi, FLINT_BITS - norm);
const ulong u0 = (a_mi << norm);
umul_ppmm(q1, q0, ninv, u1);
add_ssaaaa(q1, q0, q1, q0, u1, u0);
a_mi = (u0 - (q1 + 1) * n);
if (a_mi > q0)
a_mi += n;
if (a_mi >= n)
a_mi -= n;
}
/*
a_mi is now reduced mod n, so reduce a_mi, a_lo mod n
*/
{
const ulong u1 = a_mi + r_shift(a_lo, FLINT_BITS - norm);
const ulong u0 = (a_lo << norm);
umul_ppmm(q1, q0, ninv, u1);
add_ssaaaa(q1, q0, q1, q0, u1, u0);
r = (u0 - (q1 + 1) * n);
if (r > q0)
r += n;
return (r < n) ? (r >> norm) : ((r - n) >> norm);
}
}