forked from lthn/blockchain
4473 lines
123 KiB
C
4473 lines
123 KiB
C
// Copyright (c) 2014-2018 Zano Project
|
||
// Copyright (c) 2014-2018 The Louisdor Project
|
||
// Copyright (c) 2012-2013 The Boolberry developers
|
||
// Copyright (c) 2017-2025 Lethean (https://lt.hn)
|
||
//
|
||
// Licensed under the European Union Public Licence (EUPL) version 1.2.
|
||
// You may obtain a copy of the licence at:
|
||
//
|
||
// https://joinup.ec.europa.eu/software/page/eupl/licence-eupl
|
||
//
|
||
// The EUPL is a copyleft licence that is compatible with the MIT/X11
|
||
// licence used by the original projects; the MIT terms are therefore
|
||
// considered “grandfathered” under the EUPL for this code.
|
||
//
|
||
// SPDX‑License‑Identifier: EUPL-1.2
|
||
//
|
||
|
||
#include <assert.h>
|
||
#include <stdint.h>
|
||
#include <string.h>
|
||
#include <stddef.h>
|
||
#include "warnings.h"
|
||
#include "crypto-ops.h"
|
||
#include "hash-ops.h" // for cn_fast_hash
|
||
|
||
DISABLE_VS_WARNINGS(4146 4244)
|
||
|
||
/* Predeclarations */
|
||
|
||
void fe_mul(fe, const fe, const fe);
|
||
void fe_sq(fe, const fe);
|
||
void fe_tobytes(unsigned char *, const fe);
|
||
static void ge_msub(ge_p1p1 *, const ge_p3 *, const ge_precomp *);
|
||
static void ge_p2_0(ge_p2 *);
|
||
static void ge_p3_dbl(ge_p1p1 *, const ge_p3 *);
|
||
void ge_sub(ge_p1p1 *, const ge_p3 *, const ge_cached *);
|
||
static void fe_divpowm1(fe, const fe, const fe);
|
||
|
||
/* Common functions */
|
||
|
||
static uint64_t load_3(const unsigned char *in) {
|
||
uint64_t result;
|
||
result = (uint64_t) in[0];
|
||
result |= ((uint64_t) in[1]) << 8;
|
||
result |= ((uint64_t) in[2]) << 16;
|
||
return result;
|
||
}
|
||
|
||
static uint64_t load_4(const unsigned char *in)
|
||
{
|
||
uint64_t result;
|
||
result = (uint64_t) in[0];
|
||
result |= ((uint64_t) in[1]) << 8;
|
||
result |= ((uint64_t) in[2]) << 16;
|
||
result |= ((uint64_t) in[3]) << 24;
|
||
return result;
|
||
}
|
||
|
||
/* From fe_0.c */
|
||
|
||
/*
|
||
h = 0
|
||
*/
|
||
|
||
void fe_0(fe h) {
|
||
h[0] = 0;
|
||
h[1] = 0;
|
||
h[2] = 0;
|
||
h[3] = 0;
|
||
h[4] = 0;
|
||
h[5] = 0;
|
||
h[6] = 0;
|
||
h[7] = 0;
|
||
h[8] = 0;
|
||
h[9] = 0;
|
||
}
|
||
|
||
/* From fe_1.c */
|
||
|
||
/*
|
||
h = 1
|
||
*/
|
||
|
||
static void fe_1(fe h) {
|
||
h[0] = 1;
|
||
h[1] = 0;
|
||
h[2] = 0;
|
||
h[3] = 0;
|
||
h[4] = 0;
|
||
h[5] = 0;
|
||
h[6] = 0;
|
||
h[7] = 0;
|
||
h[8] = 0;
|
||
h[9] = 0;
|
||
}
|
||
|
||
/* From fe_add.c */
|
||
|
||
/*
|
||
h = f + g
|
||
Can overlap h with f or g.
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
|g| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
|
||
*/
|
||
|
||
void fe_add(fe h, const fe f, const fe g) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t g0 = g[0];
|
||
int32_t g1 = g[1];
|
||
int32_t g2 = g[2];
|
||
int32_t g3 = g[3];
|
||
int32_t g4 = g[4];
|
||
int32_t g5 = g[5];
|
||
int32_t g6 = g[6];
|
||
int32_t g7 = g[7];
|
||
int32_t g8 = g[8];
|
||
int32_t g9 = g[9];
|
||
int32_t h0 = f0 + g0;
|
||
int32_t h1 = f1 + g1;
|
||
int32_t h2 = f2 + g2;
|
||
int32_t h3 = f3 + g3;
|
||
int32_t h4 = f4 + g4;
|
||
int32_t h5 = f5 + g5;
|
||
int32_t h6 = f6 + g6;
|
||
int32_t h7 = f7 + g7;
|
||
int32_t h8 = f8 + g8;
|
||
int32_t h9 = f9 + g9;
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_cmov.c */
|
||
|
||
/*
|
||
Replace (f,g) with (g,g) if b == 1;
|
||
replace (f,g) with (f,g) if b == 0.
|
||
|
||
Preconditions: b in {0,1}.
|
||
*/
|
||
|
||
static void fe_cmov(fe f, const fe g, unsigned int b) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t g0 = g[0];
|
||
int32_t g1 = g[1];
|
||
int32_t g2 = g[2];
|
||
int32_t g3 = g[3];
|
||
int32_t g4 = g[4];
|
||
int32_t g5 = g[5];
|
||
int32_t g6 = g[6];
|
||
int32_t g7 = g[7];
|
||
int32_t g8 = g[8];
|
||
int32_t g9 = g[9];
|
||
int32_t x0 = f0 ^ g0;
|
||
int32_t x1 = f1 ^ g1;
|
||
int32_t x2 = f2 ^ g2;
|
||
int32_t x3 = f3 ^ g3;
|
||
int32_t x4 = f4 ^ g4;
|
||
int32_t x5 = f5 ^ g5;
|
||
int32_t x6 = f6 ^ g6;
|
||
int32_t x7 = f7 ^ g7;
|
||
int32_t x8 = f8 ^ g8;
|
||
int32_t x9 = f9 ^ g9;
|
||
assert((((b - 1) & ~b) | ((b - 2) & ~(b - 1))) == (unsigned int) -1);
|
||
b = -b;
|
||
x0 &= b;
|
||
x1 &= b;
|
||
x2 &= b;
|
||
x3 &= b;
|
||
x4 &= b;
|
||
x5 &= b;
|
||
x6 &= b;
|
||
x7 &= b;
|
||
x8 &= b;
|
||
x9 &= b;
|
||
f[0] = f0 ^ x0;
|
||
f[1] = f1 ^ x1;
|
||
f[2] = f2 ^ x2;
|
||
f[3] = f3 ^ x3;
|
||
f[4] = f4 ^ x4;
|
||
f[5] = f5 ^ x5;
|
||
f[6] = f6 ^ x6;
|
||
f[7] = f7 ^ x7;
|
||
f[8] = f8 ^ x8;
|
||
f[9] = f9 ^ x9;
|
||
}
|
||
|
||
/* From fe_copy.c */
|
||
|
||
/*
|
||
h = f
|
||
*/
|
||
|
||
static void fe_copy(fe h, const fe f) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
h[0] = f0;
|
||
h[1] = f1;
|
||
h[2] = f2;
|
||
h[3] = f3;
|
||
h[4] = f4;
|
||
h[5] = f5;
|
||
h[6] = f6;
|
||
h[7] = f7;
|
||
h[8] = f8;
|
||
h[9] = f9;
|
||
}
|
||
|
||
/* From fe_invert.c */
|
||
|
||
void fe_invert(fe out, const fe z) {
|
||
fe t0;
|
||
fe t1;
|
||
fe t2;
|
||
fe t3;
|
||
int i;
|
||
|
||
fe_sq(t0, z);
|
||
fe_sq(t1, t0);
|
||
fe_sq(t1, t1);
|
||
fe_mul(t1, z, t1);
|
||
fe_mul(t0, t0, t1);
|
||
fe_sq(t2, t0);
|
||
fe_mul(t1, t1, t2);
|
||
fe_sq(t2, t1);
|
||
for (i = 0; i < 4; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t1, t2, t1);
|
||
fe_sq(t2, t1);
|
||
for (i = 0; i < 9; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t2, t2, t1);
|
||
fe_sq(t3, t2);
|
||
for (i = 0; i < 19; ++i) {
|
||
fe_sq(t3, t3);
|
||
}
|
||
fe_mul(t2, t3, t2);
|
||
fe_sq(t2, t2);
|
||
for (i = 0; i < 9; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t1, t2, t1);
|
||
fe_sq(t2, t1);
|
||
for (i = 0; i < 49; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t2, t2, t1);
|
||
fe_sq(t3, t2);
|
||
for (i = 0; i < 99; ++i) {
|
||
fe_sq(t3, t3);
|
||
}
|
||
fe_mul(t2, t3, t2);
|
||
fe_sq(t2, t2);
|
||
for (i = 0; i < 49; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t1, t2, t1);
|
||
fe_sq(t1, t1);
|
||
for (i = 0; i < 4; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(out, t1, t0);
|
||
|
||
return;
|
||
}
|
||
|
||
/* From fe_isnegative.c */
|
||
|
||
/*
|
||
return 1 if f is in {1,3,5,...,q-2}
|
||
return 0 if f is in {0,2,4,...,q-1}
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
|
||
*/
|
||
|
||
int fe_isnegative(const fe f) {
|
||
unsigned char s[32];
|
||
fe_tobytes(s, f);
|
||
return s[0] & 1;
|
||
}
|
||
|
||
/* From fe_isnonzero.c, modified */
|
||
|
||
int fe_isnonzero(const fe f) {
|
||
unsigned char s[32];
|
||
fe_tobytes(s, f);
|
||
return (((int) (s[0] | s[1] | s[2] | s[3] | s[4] | s[5] | s[6] | s[7] | s[8] |
|
||
s[9] | s[10] | s[11] | s[12] | s[13] | s[14] | s[15] | s[16] | s[17] |
|
||
s[18] | s[19] | s[20] | s[21] | s[22] | s[23] | s[24] | s[25] | s[26] |
|
||
s[27] | s[28] | s[29] | s[30] | s[31]) - 1) >> 8) + 1;
|
||
}
|
||
|
||
/* From fe_mul.c */
|
||
|
||
/*
|
||
h = f * g
|
||
Can overlap h with f or g.
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.65*2^26,1.65*2^25,1.65*2^26,1.65*2^25,etc.
|
||
|g| bounded by 1.65*2^26,1.65*2^25,1.65*2^26,1.65*2^25,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.01*2^25,1.01*2^24,1.01*2^25,1.01*2^24,etc.
|
||
*/
|
||
|
||
/*
|
||
Notes on implementation strategy:
|
||
|
||
Using schoolbook multiplication.
|
||
Karatsuba would save a little in some cost models.
|
||
|
||
Most multiplications by 2 and 19 are 32-bit precomputations;
|
||
cheaper than 64-bit postcomputations.
|
||
|
||
There is one remaining multiplication by 19 in the carry chain;
|
||
one *19 precomputation can be merged into this,
|
||
but the resulting data flow is considerably less clean.
|
||
|
||
There are 12 carries below.
|
||
10 of them are 2-way parallelizable and vectorizable.
|
||
Can get away with 11 carries, but then data flow is much deeper.
|
||
|
||
With tighter constraints on inputs can squeeze carries into int32.
|
||
*/
|
||
|
||
void fe_mul(fe h, const fe f, const fe g) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t g0 = g[0];
|
||
int32_t g1 = g[1];
|
||
int32_t g2 = g[2];
|
||
int32_t g3 = g[3];
|
||
int32_t g4 = g[4];
|
||
int32_t g5 = g[5];
|
||
int32_t g6 = g[6];
|
||
int32_t g7 = g[7];
|
||
int32_t g8 = g[8];
|
||
int32_t g9 = g[9];
|
||
int32_t g1_19 = 19 * g1; /* 1.959375*2^29 */
|
||
int32_t g2_19 = 19 * g2; /* 1.959375*2^30; still ok */
|
||
int32_t g3_19 = 19 * g3;
|
||
int32_t g4_19 = 19 * g4;
|
||
int32_t g5_19 = 19 * g5;
|
||
int32_t g6_19 = 19 * g6;
|
||
int32_t g7_19 = 19 * g7;
|
||
int32_t g8_19 = 19 * g8;
|
||
int32_t g9_19 = 19 * g9;
|
||
int32_t f1_2 = 2 * f1;
|
||
int32_t f3_2 = 2 * f3;
|
||
int32_t f5_2 = 2 * f5;
|
||
int32_t f7_2 = 2 * f7;
|
||
int32_t f9_2 = 2 * f9;
|
||
int64_t f0g0 = f0 * (int64_t) g0;
|
||
int64_t f0g1 = f0 * (int64_t) g1;
|
||
int64_t f0g2 = f0 * (int64_t) g2;
|
||
int64_t f0g3 = f0 * (int64_t) g3;
|
||
int64_t f0g4 = f0 * (int64_t) g4;
|
||
int64_t f0g5 = f0 * (int64_t) g5;
|
||
int64_t f0g6 = f0 * (int64_t) g6;
|
||
int64_t f0g7 = f0 * (int64_t) g7;
|
||
int64_t f0g8 = f0 * (int64_t) g8;
|
||
int64_t f0g9 = f0 * (int64_t) g9;
|
||
int64_t f1g0 = f1 * (int64_t) g0;
|
||
int64_t f1g1_2 = f1_2 * (int64_t) g1;
|
||
int64_t f1g2 = f1 * (int64_t) g2;
|
||
int64_t f1g3_2 = f1_2 * (int64_t) g3;
|
||
int64_t f1g4 = f1 * (int64_t) g4;
|
||
int64_t f1g5_2 = f1_2 * (int64_t) g5;
|
||
int64_t f1g6 = f1 * (int64_t) g6;
|
||
int64_t f1g7_2 = f1_2 * (int64_t) g7;
|
||
int64_t f1g8 = f1 * (int64_t) g8;
|
||
int64_t f1g9_38 = f1_2 * (int64_t) g9_19;
|
||
int64_t f2g0 = f2 * (int64_t) g0;
|
||
int64_t f2g1 = f2 * (int64_t) g1;
|
||
int64_t f2g2 = f2 * (int64_t) g2;
|
||
int64_t f2g3 = f2 * (int64_t) g3;
|
||
int64_t f2g4 = f2 * (int64_t) g4;
|
||
int64_t f2g5 = f2 * (int64_t) g5;
|
||
int64_t f2g6 = f2 * (int64_t) g6;
|
||
int64_t f2g7 = f2 * (int64_t) g7;
|
||
int64_t f2g8_19 = f2 * (int64_t) g8_19;
|
||
int64_t f2g9_19 = f2 * (int64_t) g9_19;
|
||
int64_t f3g0 = f3 * (int64_t) g0;
|
||
int64_t f3g1_2 = f3_2 * (int64_t) g1;
|
||
int64_t f3g2 = f3 * (int64_t) g2;
|
||
int64_t f3g3_2 = f3_2 * (int64_t) g3;
|
||
int64_t f3g4 = f3 * (int64_t) g4;
|
||
int64_t f3g5_2 = f3_2 * (int64_t) g5;
|
||
int64_t f3g6 = f3 * (int64_t) g6;
|
||
int64_t f3g7_38 = f3_2 * (int64_t) g7_19;
|
||
int64_t f3g8_19 = f3 * (int64_t) g8_19;
|
||
int64_t f3g9_38 = f3_2 * (int64_t) g9_19;
|
||
int64_t f4g0 = f4 * (int64_t) g0;
|
||
int64_t f4g1 = f4 * (int64_t) g1;
|
||
int64_t f4g2 = f4 * (int64_t) g2;
|
||
int64_t f4g3 = f4 * (int64_t) g3;
|
||
int64_t f4g4 = f4 * (int64_t) g4;
|
||
int64_t f4g5 = f4 * (int64_t) g5;
|
||
int64_t f4g6_19 = f4 * (int64_t) g6_19;
|
||
int64_t f4g7_19 = f4 * (int64_t) g7_19;
|
||
int64_t f4g8_19 = f4 * (int64_t) g8_19;
|
||
int64_t f4g9_19 = f4 * (int64_t) g9_19;
|
||
int64_t f5g0 = f5 * (int64_t) g0;
|
||
int64_t f5g1_2 = f5_2 * (int64_t) g1;
|
||
int64_t f5g2 = f5 * (int64_t) g2;
|
||
int64_t f5g3_2 = f5_2 * (int64_t) g3;
|
||
int64_t f5g4 = f5 * (int64_t) g4;
|
||
int64_t f5g5_38 = f5_2 * (int64_t) g5_19;
|
||
int64_t f5g6_19 = f5 * (int64_t) g6_19;
|
||
int64_t f5g7_38 = f5_2 * (int64_t) g7_19;
|
||
int64_t f5g8_19 = f5 * (int64_t) g8_19;
|
||
int64_t f5g9_38 = f5_2 * (int64_t) g9_19;
|
||
int64_t f6g0 = f6 * (int64_t) g0;
|
||
int64_t f6g1 = f6 * (int64_t) g1;
|
||
int64_t f6g2 = f6 * (int64_t) g2;
|
||
int64_t f6g3 = f6 * (int64_t) g3;
|
||
int64_t f6g4_19 = f6 * (int64_t) g4_19;
|
||
int64_t f6g5_19 = f6 * (int64_t) g5_19;
|
||
int64_t f6g6_19 = f6 * (int64_t) g6_19;
|
||
int64_t f6g7_19 = f6 * (int64_t) g7_19;
|
||
int64_t f6g8_19 = f6 * (int64_t) g8_19;
|
||
int64_t f6g9_19 = f6 * (int64_t) g9_19;
|
||
int64_t f7g0 = f7 * (int64_t) g0;
|
||
int64_t f7g1_2 = f7_2 * (int64_t) g1;
|
||
int64_t f7g2 = f7 * (int64_t) g2;
|
||
int64_t f7g3_38 = f7_2 * (int64_t) g3_19;
|
||
int64_t f7g4_19 = f7 * (int64_t) g4_19;
|
||
int64_t f7g5_38 = f7_2 * (int64_t) g5_19;
|
||
int64_t f7g6_19 = f7 * (int64_t) g6_19;
|
||
int64_t f7g7_38 = f7_2 * (int64_t) g7_19;
|
||
int64_t f7g8_19 = f7 * (int64_t) g8_19;
|
||
int64_t f7g9_38 = f7_2 * (int64_t) g9_19;
|
||
int64_t f8g0 = f8 * (int64_t) g0;
|
||
int64_t f8g1 = f8 * (int64_t) g1;
|
||
int64_t f8g2_19 = f8 * (int64_t) g2_19;
|
||
int64_t f8g3_19 = f8 * (int64_t) g3_19;
|
||
int64_t f8g4_19 = f8 * (int64_t) g4_19;
|
||
int64_t f8g5_19 = f8 * (int64_t) g5_19;
|
||
int64_t f8g6_19 = f8 * (int64_t) g6_19;
|
||
int64_t f8g7_19 = f8 * (int64_t) g7_19;
|
||
int64_t f8g8_19 = f8 * (int64_t) g8_19;
|
||
int64_t f8g9_19 = f8 * (int64_t) g9_19;
|
||
int64_t f9g0 = f9 * (int64_t) g0;
|
||
int64_t f9g1_38 = f9_2 * (int64_t) g1_19;
|
||
int64_t f9g2_19 = f9 * (int64_t) g2_19;
|
||
int64_t f9g3_38 = f9_2 * (int64_t) g3_19;
|
||
int64_t f9g4_19 = f9 * (int64_t) g4_19;
|
||
int64_t f9g5_38 = f9_2 * (int64_t) g5_19;
|
||
int64_t f9g6_19 = f9 * (int64_t) g6_19;
|
||
int64_t f9g7_38 = f9_2 * (int64_t) g7_19;
|
||
int64_t f9g8_19 = f9 * (int64_t) g8_19;
|
||
int64_t f9g9_38 = f9_2 * (int64_t) g9_19;
|
||
int64_t h0 = f0g0+f1g9_38+f2g8_19+f3g7_38+f4g6_19+f5g5_38+f6g4_19+f7g3_38+f8g2_19+f9g1_38;
|
||
int64_t h1 = f0g1+f1g0 +f2g9_19+f3g8_19+f4g7_19+f5g6_19+f6g5_19+f7g4_19+f8g3_19+f9g2_19;
|
||
int64_t h2 = f0g2+f1g1_2 +f2g0 +f3g9_38+f4g8_19+f5g7_38+f6g6_19+f7g5_38+f8g4_19+f9g3_38;
|
||
int64_t h3 = f0g3+f1g2 +f2g1 +f3g0 +f4g9_19+f5g8_19+f6g7_19+f7g6_19+f8g5_19+f9g4_19;
|
||
int64_t h4 = f0g4+f1g3_2 +f2g2 +f3g1_2 +f4g0 +f5g9_38+f6g8_19+f7g7_38+f8g6_19+f9g5_38;
|
||
int64_t h5 = f0g5+f1g4 +f2g3 +f3g2 +f4g1 +f5g0 +f6g9_19+f7g8_19+f8g7_19+f9g6_19;
|
||
int64_t h6 = f0g6+f1g5_2 +f2g4 +f3g3_2 +f4g2 +f5g1_2 +f6g0 +f7g9_38+f8g8_19+f9g7_38;
|
||
int64_t h7 = f0g7+f1g6 +f2g5 +f3g4 +f4g3 +f5g2 +f6g1 +f7g0 +f8g9_19+f9g8_19;
|
||
int64_t h8 = f0g8+f1g7_2 +f2g6 +f3g5_2 +f4g4 +f5g3_2 +f6g2 +f7g1_2 +f8g0 +f9g9_38;
|
||
int64_t h9 = f0g9+f1g8 +f2g7 +f3g6 +f4g5 +f5g4 +f6g3 +f7g2 +f8g1 +f9g0 ;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
/*
|
||
|h0| <= (1.65*1.65*2^52*(1+19+19+19+19)+1.65*1.65*2^50*(38+38+38+38+38))
|
||
i.e. |h0| <= 1.4*2^60; narrower ranges for h2, h4, h6, h8
|
||
|h1| <= (1.65*1.65*2^51*(1+1+19+19+19+19+19+19+19+19))
|
||
i.e. |h1| <= 1.7*2^59; narrower ranges for h3, h5, h7, h9
|
||
*/
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
/* |h0| <= 2^25 */
|
||
/* |h4| <= 2^25 */
|
||
/* |h1| <= 1.71*2^59 */
|
||
/* |h5| <= 1.71*2^59 */
|
||
|
||
carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
/* |h1| <= 2^24; from now on fits into int32 */
|
||
/* |h5| <= 2^24; from now on fits into int32 */
|
||
/* |h2| <= 1.41*2^60 */
|
||
/* |h6| <= 1.41*2^60 */
|
||
|
||
carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
/* |h2| <= 2^25; from now on fits into int32 unchanged */
|
||
/* |h6| <= 2^25; from now on fits into int32 unchanged */
|
||
/* |h3| <= 1.71*2^59 */
|
||
/* |h7| <= 1.71*2^59 */
|
||
|
||
carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
/* |h3| <= 2^24; from now on fits into int32 unchanged */
|
||
/* |h7| <= 2^24; from now on fits into int32 unchanged */
|
||
/* |h4| <= 1.72*2^34 */
|
||
/* |h8| <= 1.41*2^60 */
|
||
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
/* |h4| <= 2^25; from now on fits into int32 unchanged */
|
||
/* |h8| <= 2^25; from now on fits into int32 unchanged */
|
||
/* |h5| <= 1.01*2^24 */
|
||
/* |h9| <= 1.71*2^59 */
|
||
|
||
carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
/* |h9| <= 2^24; from now on fits into int32 unchanged */
|
||
/* |h0| <= 1.1*2^39 */
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
/* |h0| <= 2^25; from now on fits into int32 unchanged */
|
||
/* |h1| <= 1.01*2^24 */
|
||
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_neg.c */
|
||
|
||
/*
|
||
h = -f
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
*/
|
||
|
||
static void fe_neg(fe h, const fe f) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t h0 = -f0;
|
||
int32_t h1 = -f1;
|
||
int32_t h2 = -f2;
|
||
int32_t h3 = -f3;
|
||
int32_t h4 = -f4;
|
||
int32_t h5 = -f5;
|
||
int32_t h6 = -f6;
|
||
int32_t h7 = -f7;
|
||
int32_t h8 = -f8;
|
||
int32_t h9 = -f9;
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_sq.c */
|
||
|
||
/*
|
||
h = f * f
|
||
Can overlap h with f.
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.65*2^26,1.65*2^25,1.65*2^26,1.65*2^25,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.01*2^25,1.01*2^24,1.01*2^25,1.01*2^24,etc.
|
||
*/
|
||
|
||
/*
|
||
See fe_mul.c for discussion of implementation strategy.
|
||
*/
|
||
|
||
void fe_sq(fe h, const fe f) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t f0_2 = 2 * f0;
|
||
int32_t f1_2 = 2 * f1;
|
||
int32_t f2_2 = 2 * f2;
|
||
int32_t f3_2 = 2 * f3;
|
||
int32_t f4_2 = 2 * f4;
|
||
int32_t f5_2 = 2 * f5;
|
||
int32_t f6_2 = 2 * f6;
|
||
int32_t f7_2 = 2 * f7;
|
||
int32_t f5_38 = 38 * f5; /* 1.959375*2^30 */
|
||
int32_t f6_19 = 19 * f6; /* 1.959375*2^30 */
|
||
int32_t f7_38 = 38 * f7; /* 1.959375*2^30 */
|
||
int32_t f8_19 = 19 * f8; /* 1.959375*2^30 */
|
||
int32_t f9_38 = 38 * f9; /* 1.959375*2^30 */
|
||
int64_t f0f0 = f0 * (int64_t) f0;
|
||
int64_t f0f1_2 = f0_2 * (int64_t) f1;
|
||
int64_t f0f2_2 = f0_2 * (int64_t) f2;
|
||
int64_t f0f3_2 = f0_2 * (int64_t) f3;
|
||
int64_t f0f4_2 = f0_2 * (int64_t) f4;
|
||
int64_t f0f5_2 = f0_2 * (int64_t) f5;
|
||
int64_t f0f6_2 = f0_2 * (int64_t) f6;
|
||
int64_t f0f7_2 = f0_2 * (int64_t) f7;
|
||
int64_t f0f8_2 = f0_2 * (int64_t) f8;
|
||
int64_t f0f9_2 = f0_2 * (int64_t) f9;
|
||
int64_t f1f1_2 = f1_2 * (int64_t) f1;
|
||
int64_t f1f2_2 = f1_2 * (int64_t) f2;
|
||
int64_t f1f3_4 = f1_2 * (int64_t) f3_2;
|
||
int64_t f1f4_2 = f1_2 * (int64_t) f4;
|
||
int64_t f1f5_4 = f1_2 * (int64_t) f5_2;
|
||
int64_t f1f6_2 = f1_2 * (int64_t) f6;
|
||
int64_t f1f7_4 = f1_2 * (int64_t) f7_2;
|
||
int64_t f1f8_2 = f1_2 * (int64_t) f8;
|
||
int64_t f1f9_76 = f1_2 * (int64_t) f9_38;
|
||
int64_t f2f2 = f2 * (int64_t) f2;
|
||
int64_t f2f3_2 = f2_2 * (int64_t) f3;
|
||
int64_t f2f4_2 = f2_2 * (int64_t) f4;
|
||
int64_t f2f5_2 = f2_2 * (int64_t) f5;
|
||
int64_t f2f6_2 = f2_2 * (int64_t) f6;
|
||
int64_t f2f7_2 = f2_2 * (int64_t) f7;
|
||
int64_t f2f8_38 = f2_2 * (int64_t) f8_19;
|
||
int64_t f2f9_38 = f2 * (int64_t) f9_38;
|
||
int64_t f3f3_2 = f3_2 * (int64_t) f3;
|
||
int64_t f3f4_2 = f3_2 * (int64_t) f4;
|
||
int64_t f3f5_4 = f3_2 * (int64_t) f5_2;
|
||
int64_t f3f6_2 = f3_2 * (int64_t) f6;
|
||
int64_t f3f7_76 = f3_2 * (int64_t) f7_38;
|
||
int64_t f3f8_38 = f3_2 * (int64_t) f8_19;
|
||
int64_t f3f9_76 = f3_2 * (int64_t) f9_38;
|
||
int64_t f4f4 = f4 * (int64_t) f4;
|
||
int64_t f4f5_2 = f4_2 * (int64_t) f5;
|
||
int64_t f4f6_38 = f4_2 * (int64_t) f6_19;
|
||
int64_t f4f7_38 = f4 * (int64_t) f7_38;
|
||
int64_t f4f8_38 = f4_2 * (int64_t) f8_19;
|
||
int64_t f4f9_38 = f4 * (int64_t) f9_38;
|
||
int64_t f5f5_38 = f5 * (int64_t) f5_38;
|
||
int64_t f5f6_38 = f5_2 * (int64_t) f6_19;
|
||
int64_t f5f7_76 = f5_2 * (int64_t) f7_38;
|
||
int64_t f5f8_38 = f5_2 * (int64_t) f8_19;
|
||
int64_t f5f9_76 = f5_2 * (int64_t) f9_38;
|
||
int64_t f6f6_19 = f6 * (int64_t) f6_19;
|
||
int64_t f6f7_38 = f6 * (int64_t) f7_38;
|
||
int64_t f6f8_38 = f6_2 * (int64_t) f8_19;
|
||
int64_t f6f9_38 = f6 * (int64_t) f9_38;
|
||
int64_t f7f7_38 = f7 * (int64_t) f7_38;
|
||
int64_t f7f8_38 = f7_2 * (int64_t) f8_19;
|
||
int64_t f7f9_76 = f7_2 * (int64_t) f9_38;
|
||
int64_t f8f8_19 = f8 * (int64_t) f8_19;
|
||
int64_t f8f9_38 = f8 * (int64_t) f9_38;
|
||
int64_t f9f9_38 = f9 * (int64_t) f9_38;
|
||
int64_t h0 = f0f0 +f1f9_76+f2f8_38+f3f7_76+f4f6_38+f5f5_38;
|
||
int64_t h1 = f0f1_2+f2f9_38+f3f8_38+f4f7_38+f5f6_38;
|
||
int64_t h2 = f0f2_2+f1f1_2 +f3f9_76+f4f8_38+f5f7_76+f6f6_19;
|
||
int64_t h3 = f0f3_2+f1f2_2 +f4f9_38+f5f8_38+f6f7_38;
|
||
int64_t h4 = f0f4_2+f1f3_4 +f2f2 +f5f9_76+f6f8_38+f7f7_38;
|
||
int64_t h5 = f0f5_2+f1f4_2 +f2f3_2 +f6f9_38+f7f8_38;
|
||
int64_t h6 = f0f6_2+f1f5_4 +f2f4_2 +f3f3_2 +f7f9_76+f8f8_19;
|
||
int64_t h7 = f0f7_2+f1f6_2 +f2f5_2 +f3f4_2 +f8f9_38;
|
||
int64_t h8 = f0f8_2+f1f7_4 +f2f6_2 +f3f5_4 +f4f4 +f9f9_38;
|
||
int64_t h9 = f0f9_2+f1f8_2 +f2f7_2 +f3f6_2 +f4f5_2;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
|
||
carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
|
||
carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
|
||
carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
|
||
carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_sq2.c */
|
||
|
||
/*
|
||
h = 2 * f * f
|
||
Can overlap h with f.
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.65*2^26,1.65*2^25,1.65*2^26,1.65*2^25,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.01*2^25,1.01*2^24,1.01*2^25,1.01*2^24,etc.
|
||
*/
|
||
|
||
/*
|
||
See fe_mul.c for discussion of implementation strategy.
|
||
*/
|
||
|
||
static void fe_sq2(fe h, const fe f) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t f0_2 = 2 * f0;
|
||
int32_t f1_2 = 2 * f1;
|
||
int32_t f2_2 = 2 * f2;
|
||
int32_t f3_2 = 2 * f3;
|
||
int32_t f4_2 = 2 * f4;
|
||
int32_t f5_2 = 2 * f5;
|
||
int32_t f6_2 = 2 * f6;
|
||
int32_t f7_2 = 2 * f7;
|
||
int32_t f5_38 = 38 * f5; /* 1.959375*2^30 */
|
||
int32_t f6_19 = 19 * f6; /* 1.959375*2^30 */
|
||
int32_t f7_38 = 38 * f7; /* 1.959375*2^30 */
|
||
int32_t f8_19 = 19 * f8; /* 1.959375*2^30 */
|
||
int32_t f9_38 = 38 * f9; /* 1.959375*2^30 */
|
||
int64_t f0f0 = f0 * (int64_t) f0;
|
||
int64_t f0f1_2 = f0_2 * (int64_t) f1;
|
||
int64_t f0f2_2 = f0_2 * (int64_t) f2;
|
||
int64_t f0f3_2 = f0_2 * (int64_t) f3;
|
||
int64_t f0f4_2 = f0_2 * (int64_t) f4;
|
||
int64_t f0f5_2 = f0_2 * (int64_t) f5;
|
||
int64_t f0f6_2 = f0_2 * (int64_t) f6;
|
||
int64_t f0f7_2 = f0_2 * (int64_t) f7;
|
||
int64_t f0f8_2 = f0_2 * (int64_t) f8;
|
||
int64_t f0f9_2 = f0_2 * (int64_t) f9;
|
||
int64_t f1f1_2 = f1_2 * (int64_t) f1;
|
||
int64_t f1f2_2 = f1_2 * (int64_t) f2;
|
||
int64_t f1f3_4 = f1_2 * (int64_t) f3_2;
|
||
int64_t f1f4_2 = f1_2 * (int64_t) f4;
|
||
int64_t f1f5_4 = f1_2 * (int64_t) f5_2;
|
||
int64_t f1f6_2 = f1_2 * (int64_t) f6;
|
||
int64_t f1f7_4 = f1_2 * (int64_t) f7_2;
|
||
int64_t f1f8_2 = f1_2 * (int64_t) f8;
|
||
int64_t f1f9_76 = f1_2 * (int64_t) f9_38;
|
||
int64_t f2f2 = f2 * (int64_t) f2;
|
||
int64_t f2f3_2 = f2_2 * (int64_t) f3;
|
||
int64_t f2f4_2 = f2_2 * (int64_t) f4;
|
||
int64_t f2f5_2 = f2_2 * (int64_t) f5;
|
||
int64_t f2f6_2 = f2_2 * (int64_t) f6;
|
||
int64_t f2f7_2 = f2_2 * (int64_t) f7;
|
||
int64_t f2f8_38 = f2_2 * (int64_t) f8_19;
|
||
int64_t f2f9_38 = f2 * (int64_t) f9_38;
|
||
int64_t f3f3_2 = f3_2 * (int64_t) f3;
|
||
int64_t f3f4_2 = f3_2 * (int64_t) f4;
|
||
int64_t f3f5_4 = f3_2 * (int64_t) f5_2;
|
||
int64_t f3f6_2 = f3_2 * (int64_t) f6;
|
||
int64_t f3f7_76 = f3_2 * (int64_t) f7_38;
|
||
int64_t f3f8_38 = f3_2 * (int64_t) f8_19;
|
||
int64_t f3f9_76 = f3_2 * (int64_t) f9_38;
|
||
int64_t f4f4 = f4 * (int64_t) f4;
|
||
int64_t f4f5_2 = f4_2 * (int64_t) f5;
|
||
int64_t f4f6_38 = f4_2 * (int64_t) f6_19;
|
||
int64_t f4f7_38 = f4 * (int64_t) f7_38;
|
||
int64_t f4f8_38 = f4_2 * (int64_t) f8_19;
|
||
int64_t f4f9_38 = f4 * (int64_t) f9_38;
|
||
int64_t f5f5_38 = f5 * (int64_t) f5_38;
|
||
int64_t f5f6_38 = f5_2 * (int64_t) f6_19;
|
||
int64_t f5f7_76 = f5_2 * (int64_t) f7_38;
|
||
int64_t f5f8_38 = f5_2 * (int64_t) f8_19;
|
||
int64_t f5f9_76 = f5_2 * (int64_t) f9_38;
|
||
int64_t f6f6_19 = f6 * (int64_t) f6_19;
|
||
int64_t f6f7_38 = f6 * (int64_t) f7_38;
|
||
int64_t f6f8_38 = f6_2 * (int64_t) f8_19;
|
||
int64_t f6f9_38 = f6 * (int64_t) f9_38;
|
||
int64_t f7f7_38 = f7 * (int64_t) f7_38;
|
||
int64_t f7f8_38 = f7_2 * (int64_t) f8_19;
|
||
int64_t f7f9_76 = f7_2 * (int64_t) f9_38;
|
||
int64_t f8f8_19 = f8 * (int64_t) f8_19;
|
||
int64_t f8f9_38 = f8 * (int64_t) f9_38;
|
||
int64_t f9f9_38 = f9 * (int64_t) f9_38;
|
||
int64_t h0 = f0f0 +f1f9_76+f2f8_38+f3f7_76+f4f6_38+f5f5_38;
|
||
int64_t h1 = f0f1_2+f2f9_38+f3f8_38+f4f7_38+f5f6_38;
|
||
int64_t h2 = f0f2_2+f1f1_2 +f3f9_76+f4f8_38+f5f7_76+f6f6_19;
|
||
int64_t h3 = f0f3_2+f1f2_2 +f4f9_38+f5f8_38+f6f7_38;
|
||
int64_t h4 = f0f4_2+f1f3_4 +f2f2 +f5f9_76+f6f8_38+f7f7_38;
|
||
int64_t h5 = f0f5_2+f1f4_2 +f2f3_2 +f6f9_38+f7f8_38;
|
||
int64_t h6 = f0f6_2+f1f5_4 +f2f4_2 +f3f3_2 +f7f9_76+f8f8_19;
|
||
int64_t h7 = f0f7_2+f1f6_2 +f2f5_2 +f3f4_2 +f8f9_38;
|
||
int64_t h8 = f0f8_2+f1f7_4 +f2f6_2 +f3f5_4 +f4f4 +f9f9_38;
|
||
int64_t h9 = f0f9_2+f1f8_2 +f2f7_2 +f3f6_2 +f4f5_2;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
h0 += h0;
|
||
h1 += h1;
|
||
h2 += h2;
|
||
h3 += h3;
|
||
h4 += h4;
|
||
h5 += h5;
|
||
h6 += h6;
|
||
h7 += h7;
|
||
h8 += h8;
|
||
h9 += h9;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
|
||
carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
|
||
carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
|
||
carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
|
||
carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_sub.c */
|
||
|
||
/*
|
||
h = f - g
|
||
Can overlap h with f or g.
|
||
|
||
Preconditions:
|
||
|f| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
|g| bounded by 1.1*2^25,1.1*2^24,1.1*2^25,1.1*2^24,etc.
|
||
|
||
Postconditions:
|
||
|h| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
|
||
*/
|
||
|
||
void fe_sub(fe h, const fe f, const fe g) {
|
||
int32_t f0 = f[0];
|
||
int32_t f1 = f[1];
|
||
int32_t f2 = f[2];
|
||
int32_t f3 = f[3];
|
||
int32_t f4 = f[4];
|
||
int32_t f5 = f[5];
|
||
int32_t f6 = f[6];
|
||
int32_t f7 = f[7];
|
||
int32_t f8 = f[8];
|
||
int32_t f9 = f[9];
|
||
int32_t g0 = g[0];
|
||
int32_t g1 = g[1];
|
||
int32_t g2 = g[2];
|
||
int32_t g3 = g[3];
|
||
int32_t g4 = g[4];
|
||
int32_t g5 = g[5];
|
||
int32_t g6 = g[6];
|
||
int32_t g7 = g[7];
|
||
int32_t g8 = g[8];
|
||
int32_t g9 = g[9];
|
||
int32_t h0 = f0 - g0;
|
||
int32_t h1 = f1 - g1;
|
||
int32_t h2 = f2 - g2;
|
||
int32_t h3 = f3 - g3;
|
||
int32_t h4 = f4 - g4;
|
||
int32_t h5 = f5 - g5;
|
||
int32_t h6 = f6 - g6;
|
||
int32_t h7 = f7 - g7;
|
||
int32_t h8 = f8 - g8;
|
||
int32_t h9 = f9 - g9;
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/* From fe_tobytes.c */
|
||
|
||
/*
|
||
Preconditions:
|
||
|h| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
|
||
|
||
Write p=2^255-19; q=floor(h/p).
|
||
Basic claim: q = floor(2^(-255)(h + 19 2^(-25)h9 + 2^(-1))).
|
||
|
||
Proof:
|
||
Have |h|<=p so |q|<=1 so |19^2 2^(-255) q|<1/4.
|
||
Also have |h-2^230 h9|<2^231 so |19 2^(-255)(h-2^230 h9)|<1/4.
|
||
|
||
Write y=2^(-1)-19^2 2^(-255)q-19 2^(-255)(h-2^230 h9).
|
||
Then 0<y<1.
|
||
|
||
Write r=h-pq.
|
||
Have 0<=r<=p-1=2^255-20.
|
||
Thus 0<=r+19(2^-255)r<r+19(2^-255)2^255<=2^255-1.
|
||
|
||
Write x=r+19(2^-255)r+y.
|
||
Then 0<x<2^255 so floor(2^(-255)x) = 0 so floor(q+2^(-255)x) = q.
|
||
|
||
Have q+2^(-255)x = 2^(-255)(h + 19 2^(-25) h9 + 2^(-1))
|
||
so floor(2^(-255)(h + 19 2^(-25) h9 + 2^(-1))) = q.
|
||
*/
|
||
|
||
void fe_tobytes(unsigned char *s, const fe h) {
|
||
int32_t h0 = h[0];
|
||
int32_t h1 = h[1];
|
||
int32_t h2 = h[2];
|
||
int32_t h3 = h[3];
|
||
int32_t h4 = h[4];
|
||
int32_t h5 = h[5];
|
||
int32_t h6 = h[6];
|
||
int32_t h7 = h[7];
|
||
int32_t h8 = h[8];
|
||
int32_t h9 = h[9];
|
||
int32_t q;
|
||
int32_t carry0;
|
||
int32_t carry1;
|
||
int32_t carry2;
|
||
int32_t carry3;
|
||
int32_t carry4;
|
||
int32_t carry5;
|
||
int32_t carry6;
|
||
int32_t carry7;
|
||
int32_t carry8;
|
||
int32_t carry9;
|
||
|
||
q = (19 * h9 + (((int32_t) 1) << 24)) >> 25;
|
||
q = (h0 + q) >> 26;
|
||
q = (h1 + q) >> 25;
|
||
q = (h2 + q) >> 26;
|
||
q = (h3 + q) >> 25;
|
||
q = (h4 + q) >> 26;
|
||
q = (h5 + q) >> 25;
|
||
q = (h6 + q) >> 26;
|
||
q = (h7 + q) >> 25;
|
||
q = (h8 + q) >> 26;
|
||
q = (h9 + q) >> 25;
|
||
|
||
/* Goal: Output h-(2^255-19)q, which is between 0 and 2^255-20. */
|
||
h0 += 19 * q;
|
||
/* Goal: Output h-2^255 q, which is between 0 and 2^255-20. */
|
||
|
||
carry0 = h0 >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry1 = h1 >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry2 = h2 >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry3 = h3 >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry4 = h4 >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry5 = h5 >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
carry6 = h6 >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
carry7 = h7 >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
carry8 = h8 >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
carry9 = h9 >> 25; h9 -= carry9 << 25;
|
||
/* h10 = carry9 */
|
||
|
||
/*
|
||
Goal: Output h0+...+2^255 h10-2^255 q, which is between 0 and 2^255-20.
|
||
Have h0+...+2^230 h9 between 0 and 2^255-1;
|
||
evidently 2^255 h10-2^255 q = 0.
|
||
Goal: Output h0+...+2^230 h9.
|
||
*/
|
||
|
||
s[0] = h0 >> 0;
|
||
s[1] = h0 >> 8;
|
||
s[2] = h0 >> 16;
|
||
s[3] = (h0 >> 24) | (h1 << 2);
|
||
s[4] = h1 >> 6;
|
||
s[5] = h1 >> 14;
|
||
s[6] = (h1 >> 22) | (h2 << 3);
|
||
s[7] = h2 >> 5;
|
||
s[8] = h2 >> 13;
|
||
s[9] = (h2 >> 21) | (h3 << 5);
|
||
s[10] = h3 >> 3;
|
||
s[11] = h3 >> 11;
|
||
s[12] = (h3 >> 19) | (h4 << 6);
|
||
s[13] = h4 >> 2;
|
||
s[14] = h4 >> 10;
|
||
s[15] = h4 >> 18;
|
||
s[16] = h5 >> 0;
|
||
s[17] = h5 >> 8;
|
||
s[18] = h5 >> 16;
|
||
s[19] = (h5 >> 24) | (h6 << 1);
|
||
s[20] = h6 >> 7;
|
||
s[21] = h6 >> 15;
|
||
s[22] = (h6 >> 23) | (h7 << 3);
|
||
s[23] = h7 >> 5;
|
||
s[24] = h7 >> 13;
|
||
s[25] = (h7 >> 21) | (h8 << 4);
|
||
s[26] = h8 >> 4;
|
||
s[27] = h8 >> 12;
|
||
s[28] = (h8 >> 20) | (h9 << 6);
|
||
s[29] = h9 >> 2;
|
||
s[30] = h9 >> 10;
|
||
s[31] = h9 >> 18;
|
||
}
|
||
|
||
/* From ge_add.c */
|
||
|
||
void ge_add(ge_p1p1 *r, const ge_p3 *p, const ge_cached *q) {
|
||
fe t0;
|
||
fe_add(r->X, p->Y, p->X);
|
||
fe_sub(r->Y, p->Y, p->X);
|
||
fe_mul(r->Z, r->X, q->YplusX);
|
||
fe_mul(r->Y, r->Y, q->YminusX);
|
||
fe_mul(r->T, q->T2d, p->T);
|
||
fe_mul(r->X, p->Z, q->Z);
|
||
fe_add(t0, r->X, r->X);
|
||
fe_sub(r->X, r->Z, r->Y);
|
||
fe_add(r->Y, r->Z, r->Y);
|
||
fe_add(r->Z, t0, r->T);
|
||
fe_sub(r->T, t0, r->T);
|
||
}
|
||
|
||
/* From ge_double_scalarmult.c, modified */
|
||
|
||
static void slide(signed char *r, const unsigned char *a) {
|
||
int i;
|
||
int b;
|
||
int k;
|
||
|
||
for (i = 0; i < 256; ++i) {
|
||
r[i] = 1 & (a[i >> 3] >> (i & 7));
|
||
}
|
||
|
||
for (i = 0; i < 256; ++i) {
|
||
if (r[i]) {
|
||
for (b = 1; b <= 6 && i + b < 256; ++b) {
|
||
if (r[i + b]) {
|
||
if (r[i] + (r[i + b] << b) <= 15) {
|
||
r[i] += r[i + b] << b; r[i + b] = 0;
|
||
} else if (r[i] - (r[i + b] << b) >= -15) {
|
||
r[i] -= r[i + b] << b;
|
||
for (k = i + b; k < 256; ++k) {
|
||
if (!r[k]) {
|
||
r[k] = 1;
|
||
break;
|
||
}
|
||
r[k] = 0;
|
||
}
|
||
} else
|
||
break;
|
||
}
|
||
}
|
||
}
|
||
}
|
||
}
|
||
|
||
void ge_dsm_precomp(ge_dsmp r, const ge_p3 *s) {
|
||
ge_p1p1 t;
|
||
ge_p3 s2, u;
|
||
ge_p3_to_cached(&r[0], s);
|
||
ge_p3_dbl(&t, s); ge_p1p1_to_p3(&s2, &t);
|
||
ge_add(&t, &s2, &r[0]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[1], &u);
|
||
ge_add(&t, &s2, &r[1]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[2], &u);
|
||
ge_add(&t, &s2, &r[2]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[3], &u);
|
||
ge_add(&t, &s2, &r[3]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[4], &u);
|
||
ge_add(&t, &s2, &r[4]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[5], &u);
|
||
ge_add(&t, &s2, &r[5]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[6], &u);
|
||
ge_add(&t, &s2, &r[6]); ge_p1p1_to_p3(&u, &t); ge_p3_to_cached(&r[7], &u);
|
||
}
|
||
|
||
/*
|
||
r = a * A + b * B
|
||
where a = a[0]+256*a[1]+...+256^31 a[31].
|
||
and b = b[0]+256*b[1]+...+256^31 b[31].
|
||
B is the Ed25519 base point (x,4/5) with x positive.
|
||
*/
|
||
|
||
void ge_double_scalarmult_base_vartime(ge_p2 *r, const unsigned char *a, const ge_p3 *A, const unsigned char *b) {
|
||
signed char aslide[256];
|
||
signed char bslide[256];
|
||
ge_dsmp Ai; /* A, 3A, 5A, 7A, 9A, 11A, 13A, 15A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
int i;
|
||
|
||
slide(aslide, a);
|
||
slide(bslide, b);
|
||
ge_dsm_precomp(Ai, A);
|
||
|
||
ge_p2_0(r);
|
||
|
||
for (i = 255; i >= 0; --i) {
|
||
if (aslide[i] || bslide[i]) break;
|
||
}
|
||
|
||
for (; i >= 0; --i) {
|
||
ge_p2_dbl(&t, r);
|
||
|
||
if (aslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Ai[aslide[i]/2]);
|
||
} else if (aslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Ai[(-aslide[i])/2]);
|
||
}
|
||
|
||
if (bslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_madd(&t, &u, &ge_Bi[bslide[i]/2]);
|
||
} else if (bslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_msub(&t, &u, &ge_Bi[(-bslide[i])/2]);
|
||
}
|
||
|
||
ge_p1p1_to_p2(r, &t);
|
||
}
|
||
}
|
||
|
||
/*
|
||
r = a * A + b * B
|
||
where a = a[0]+256*a[1]+...+256^31 a[31].
|
||
and b = b[0]+256*b[1]+...+256^31 b[31].
|
||
B is the Ed25519 base point (x,4/5) with x positive.
|
||
*/
|
||
|
||
void ge_double_scalarmult_base_vartime_p3(ge_p3 *r, const unsigned char *a, const ge_p3 *A, const unsigned char *b) {
|
||
signed char aslide[256];
|
||
signed char bslide[256];
|
||
ge_dsmp Ai; /* A, 3A, 5A, 7A, 9A, 11A, 13A, 15A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
ge_p2 r_p2;
|
||
int i;
|
||
|
||
slide(aslide, a);
|
||
slide(bslide, b);
|
||
ge_dsm_precomp(Ai, A);
|
||
|
||
ge_p2_0(&r_p2);
|
||
|
||
for (i = 255; i >= 0; --i) {
|
||
if (aslide[i] || bslide[i]) break;
|
||
}
|
||
|
||
for (; i >= 0; --i) {
|
||
ge_p2_dbl(&t, &r_p2);
|
||
|
||
if (aslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Ai[aslide[i]/2]);
|
||
} else if (aslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Ai[(-aslide[i])/2]);
|
||
}
|
||
|
||
if (bslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_madd(&t, &u, &ge_Bi[bslide[i]/2]);
|
||
} else if (bslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_msub(&t, &u, &ge_Bi[(-bslide[i])/2]);
|
||
}
|
||
|
||
if (i != 0)
|
||
ge_p1p1_to_p2(&r_p2, &t);
|
||
else
|
||
ge_p1p1_to_p3(r, &t); // last step
|
||
}
|
||
}
|
||
|
||
void ge_scalarmult_vartime_p3(ge_p3 *r, const unsigned char *a, const ge_p3 *A) {
|
||
signed char aslide[256];
|
||
ge_dsmp Ai; /* A, 3A, 5A, 7A, 9A, 11A, 13A, 15A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
ge_p2 r_p2;
|
||
int i;
|
||
|
||
slide(aslide, a);
|
||
ge_dsm_precomp(Ai, A);
|
||
|
||
ge_p2_0(&r_p2);
|
||
ge_p3_0(r);
|
||
|
||
for (i = 255; i >= 0; --i) {
|
||
if (aslide[i]) break;
|
||
}
|
||
|
||
for (; i >= 0; --i) {
|
||
ge_p2_dbl(&t, &r_p2);
|
||
|
||
if (aslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Ai[aslide[i] / 2]);
|
||
}
|
||
else if (aslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Ai[(-aslide[i]) / 2]);
|
||
}
|
||
|
||
if (i != 0)
|
||
ge_p1p1_to_p2(&r_p2, &t);
|
||
else
|
||
ge_p1p1_to_p3(r, &t); // last step
|
||
}
|
||
}
|
||
|
||
|
||
/* From ge_frombytes.c, modified */
|
||
|
||
int ge_frombytes_vartime(ge_p3 *h, const unsigned char *s) {
|
||
fe u;
|
||
fe v;
|
||
fe vxx;
|
||
fe check;
|
||
|
||
/* From fe_frombytes.c */
|
||
|
||
int64_t h0 = load_4(s);
|
||
int64_t h1 = load_3(s + 4) << 6;
|
||
int64_t h2 = load_3(s + 7) << 5;
|
||
int64_t h3 = load_3(s + 10) << 3;
|
||
int64_t h4 = load_3(s + 13) << 2;
|
||
int64_t h5 = load_4(s + 16);
|
||
int64_t h6 = load_3(s + 20) << 7;
|
||
int64_t h7 = load_3(s + 23) << 5;
|
||
int64_t h8 = load_3(s + 26) << 4;
|
||
int64_t h9 = (load_3(s + 29) & 8388607) << 2;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
/* Validate the number to be canonical */
|
||
if (h9 == 33554428 && h8 == 268435440 && h7 == 536870880 && h6 == 2147483520 &&
|
||
h5 == 4294967295 && h4 == 67108860 && h3 == 134217720 && h2 == 536870880 &&
|
||
h1 == 1073741760 && h0 >= 4294967277) {
|
||
return -1;
|
||
}
|
||
|
||
carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
|
||
h->Y[0] = h0;
|
||
h->Y[1] = h1;
|
||
h->Y[2] = h2;
|
||
h->Y[3] = h3;
|
||
h->Y[4] = h4;
|
||
h->Y[5] = h5;
|
||
h->Y[6] = h6;
|
||
h->Y[7] = h7;
|
||
h->Y[8] = h8;
|
||
h->Y[9] = h9;
|
||
|
||
/* End fe_frombytes.c */
|
||
|
||
fe_1(h->Z);
|
||
fe_sq(u, h->Y);
|
||
fe_mul(v, u, fe_d);
|
||
fe_sub(u, u, h->Z); /* u = y^2-1 */
|
||
fe_add(v, v, h->Z); /* v = dy^2+1 */
|
||
|
||
fe_divpowm1(h->X, u, v); /* x = uv^3(uv^7)^((q-5)/8) */
|
||
|
||
fe_sq(vxx, h->X);
|
||
fe_mul(vxx, vxx, v);
|
||
fe_sub(check, vxx, u); /* vx^2-u */
|
||
if (fe_isnonzero(check)) {
|
||
fe_add(check, vxx, u); /* vx^2+u */
|
||
if (fe_isnonzero(check)) {
|
||
return -1;
|
||
}
|
||
fe_mul(h->X, h->X, fe_sqrtm1);
|
||
}
|
||
|
||
if (fe_isnegative(h->X) != (s[31] >> 7)) {
|
||
/* If x = 0, the sign must be positive */
|
||
if (!fe_isnonzero(h->X)) {
|
||
return -1;
|
||
}
|
||
fe_neg(h->X, h->X);
|
||
}
|
||
|
||
fe_mul(h->T, h->X, h->Y);
|
||
return 0;
|
||
}
|
||
|
||
/* From ge_madd.c */
|
||
|
||
/*
|
||
r = p + q
|
||
*/
|
||
|
||
void ge_madd(ge_p1p1 *r, const ge_p3 *p, const ge_precomp *q) {
|
||
fe t0;
|
||
fe_add(r->X, p->Y, p->X);
|
||
fe_sub(r->Y, p->Y, p->X);
|
||
fe_mul(r->Z, r->X, q->yplusx);
|
||
fe_mul(r->Y, r->Y, q->yminusx);
|
||
fe_mul(r->T, q->xy2d, p->T);
|
||
fe_add(t0, p->Z, p->Z);
|
||
fe_sub(r->X, r->Z, r->Y);
|
||
fe_add(r->Y, r->Z, r->Y);
|
||
fe_add(r->Z, t0, r->T);
|
||
fe_sub(r->T, t0, r->T);
|
||
}
|
||
|
||
/* From ge_msub.c */
|
||
|
||
/*
|
||
r = p - q
|
||
*/
|
||
|
||
static void ge_msub(ge_p1p1 *r, const ge_p3 *p, const ge_precomp *q) {
|
||
fe t0;
|
||
fe_add(r->X, p->Y, p->X);
|
||
fe_sub(r->Y, p->Y, p->X);
|
||
fe_mul(r->Z, r->X, q->yminusx);
|
||
fe_mul(r->Y, r->Y, q->yplusx);
|
||
fe_mul(r->T, q->xy2d, p->T);
|
||
fe_add(t0, p->Z, p->Z);
|
||
fe_sub(r->X, r->Z, r->Y);
|
||
fe_add(r->Y, r->Z, r->Y);
|
||
fe_sub(r->Z, t0, r->T);
|
||
fe_add(r->T, t0, r->T);
|
||
}
|
||
|
||
/* From ge_p1p1_to_p2.c */
|
||
|
||
/*
|
||
r = p
|
||
*/
|
||
|
||
void ge_p1p1_to_p2(ge_p2 *r, const ge_p1p1 *p) {
|
||
fe_mul(r->X, p->X, p->T);
|
||
fe_mul(r->Y, p->Y, p->Z);
|
||
fe_mul(r->Z, p->Z, p->T);
|
||
}
|
||
|
||
/* From ge_p1p1_to_p3.c */
|
||
|
||
/*
|
||
r = p
|
||
*/
|
||
|
||
void ge_p1p1_to_p3(ge_p3 *r, const ge_p1p1 *p) {
|
||
fe_mul(r->X, p->X, p->T);
|
||
fe_mul(r->Y, p->Y, p->Z);
|
||
fe_mul(r->Z, p->Z, p->T);
|
||
fe_mul(r->T, p->X, p->Y);
|
||
}
|
||
|
||
/* From ge_p2_0.c */
|
||
|
||
static void ge_p2_0(ge_p2 *h) {
|
||
fe_0(h->X);
|
||
fe_1(h->Y);
|
||
fe_1(h->Z);
|
||
}
|
||
|
||
/* From ge_p2_dbl.c */
|
||
|
||
/*
|
||
r = 2 * p
|
||
*/
|
||
|
||
void ge_p2_dbl(ge_p1p1 *r, const ge_p2 *p) {
|
||
fe t0;
|
||
fe_sq(r->X, p->X);
|
||
fe_sq(r->Z, p->Y);
|
||
fe_sq2(r->T, p->Z);
|
||
fe_add(r->Y, p->X, p->Y);
|
||
fe_sq(t0, r->Y);
|
||
fe_add(r->Y, r->Z, r->X);
|
||
fe_sub(r->Z, r->Z, r->X);
|
||
fe_sub(r->X, t0, r->Y);
|
||
fe_sub(r->T, r->T, r->Z);
|
||
}
|
||
|
||
/* From ge_p3_0.c */
|
||
|
||
void ge_p3_0(ge_p3 *h) {
|
||
fe_0(h->X);
|
||
fe_1(h->Y);
|
||
fe_1(h->Z);
|
||
fe_0(h->T);
|
||
}
|
||
|
||
/* From ge_p3_dbl.c */
|
||
|
||
/*
|
||
r = 2 * p
|
||
*/
|
||
|
||
static void ge_p3_dbl(ge_p1p1 *r, const ge_p3 *p) {
|
||
ge_p2 q;
|
||
ge_p3_to_p2(&q, p);
|
||
ge_p2_dbl(r, &q);
|
||
}
|
||
|
||
/* From ge_p3_to_cached.c */
|
||
|
||
/*
|
||
r = p
|
||
*/
|
||
|
||
void ge_p3_to_cached(ge_cached *r, const ge_p3 *p) {
|
||
fe_add(r->YplusX, p->Y, p->X);
|
||
fe_sub(r->YminusX, p->Y, p->X);
|
||
fe_copy(r->Z, p->Z);
|
||
fe_mul(r->T2d, p->T, fe_d2);
|
||
}
|
||
|
||
/* From ge_p3_to_p2.c */
|
||
|
||
/*
|
||
r = p
|
||
*/
|
||
|
||
void ge_p3_to_p2(ge_p2 *r, const ge_p3 *p) {
|
||
fe_copy(r->X, p->X);
|
||
fe_copy(r->Y, p->Y);
|
||
fe_copy(r->Z, p->Z);
|
||
}
|
||
|
||
/* From ge_p3_tobytes.c */
|
||
|
||
void ge_p3_tobytes(unsigned char *s, const ge_p3 *h) {
|
||
fe recip;
|
||
fe x;
|
||
fe y;
|
||
|
||
fe_invert(recip, h->Z);
|
||
fe_mul(x, h->X, recip);
|
||
fe_mul(y, h->Y, recip);
|
||
fe_tobytes(s, y);
|
||
s[31] ^= fe_isnegative(x) << 7;
|
||
}
|
||
|
||
/* From ge_precomp_0.c */
|
||
|
||
static void ge_precomp_0(ge_precomp *h) {
|
||
fe_1(h->yplusx);
|
||
fe_1(h->yminusx);
|
||
fe_0(h->xy2d);
|
||
}
|
||
|
||
/* From ge_scalarmult_base.c */
|
||
|
||
static unsigned char equal(signed char b, signed char c) {
|
||
unsigned char ub = b;
|
||
unsigned char uc = c;
|
||
unsigned char x = ub ^ uc; /* 0: yes; 1..255: no */
|
||
uint32_t y = x; /* 0: yes; 1..255: no */
|
||
y -= 1; /* 4294967295: yes; 0..254: no */
|
||
y >>= 31; /* 1: yes; 0: no */
|
||
return y;
|
||
}
|
||
|
||
static unsigned char negative(signed char b) {
|
||
unsigned long long x = b; /* 18446744073709551361..18446744073709551615: yes; 0..255: no */
|
||
x >>= 63; /* 1: yes; 0: no */
|
||
return x;
|
||
}
|
||
|
||
static void ge_precomp_cmov(ge_precomp *t, const ge_precomp *u, unsigned char b) {
|
||
fe_cmov(t->yplusx, u->yplusx, b);
|
||
fe_cmov(t->yminusx, u->yminusx, b);
|
||
fe_cmov(t->xy2d, u->xy2d, b);
|
||
}
|
||
|
||
static void select(ge_precomp *t, int pos, signed char b) {
|
||
ge_precomp minust;
|
||
unsigned char bnegative = negative(b);
|
||
unsigned char babs = b - (((-bnegative) & b) << 1);
|
||
|
||
ge_precomp_0(t);
|
||
ge_precomp_cmov(t, &ge_base[pos][0], equal(babs, 1));
|
||
ge_precomp_cmov(t, &ge_base[pos][1], equal(babs, 2));
|
||
ge_precomp_cmov(t, &ge_base[pos][2], equal(babs, 3));
|
||
ge_precomp_cmov(t, &ge_base[pos][3], equal(babs, 4));
|
||
ge_precomp_cmov(t, &ge_base[pos][4], equal(babs, 5));
|
||
ge_precomp_cmov(t, &ge_base[pos][5], equal(babs, 6));
|
||
ge_precomp_cmov(t, &ge_base[pos][6], equal(babs, 7));
|
||
ge_precomp_cmov(t, &ge_base[pos][7], equal(babs, 8));
|
||
fe_copy(minust.yplusx, t->yminusx);
|
||
fe_copy(minust.yminusx, t->yplusx);
|
||
fe_neg(minust.xy2d, t->xy2d);
|
||
ge_precomp_cmov(t, &minust, bnegative);
|
||
}
|
||
|
||
/*
|
||
h = a * B
|
||
where a = a[0]+256*a[1]+...+256^31 a[31]
|
||
B is the Ed25519 base point (x,4/5) with x positive.
|
||
|
||
Preconditions:
|
||
a[31] <= 127
|
||
*/
|
||
|
||
void ge_scalarmult_base(ge_p3 *h, const unsigned char *a) {
|
||
signed char e[64];
|
||
signed char carry;
|
||
ge_p1p1 r;
|
||
ge_p2 s;
|
||
ge_precomp t;
|
||
int i;
|
||
|
||
for (i = 0; i < 32; ++i) {
|
||
e[2 * i + 0] = (a[i] >> 0) & 15;
|
||
e[2 * i + 1] = (a[i] >> 4) & 15;
|
||
}
|
||
/* each e[i] is between 0 and 15 */
|
||
/* e[63] is between 0 and 7 */
|
||
|
||
carry = 0;
|
||
for (i = 0; i < 63; ++i) {
|
||
e[i] += carry;
|
||
carry = e[i] + 8;
|
||
carry >>= 4;
|
||
e[i] -= carry << 4;
|
||
}
|
||
e[63] += carry;
|
||
/* each e[i] is between -8 and 8 */
|
||
|
||
ge_p3_0(h);
|
||
for (i = 1; i < 64; i += 2) {
|
||
select(&t, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
|
||
ge_p3_dbl(&r, h); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p3(h, &r);
|
||
|
||
for (i = 0; i < 64; i += 2) {
|
||
select(&t, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
}
|
||
|
||
/* From ge_sub.c */
|
||
|
||
/*
|
||
r = p - q
|
||
*/
|
||
|
||
void ge_sub(ge_p1p1 *r, const ge_p3 *p, const ge_cached *q) {
|
||
fe t0;
|
||
fe_add(r->X, p->Y, p->X);
|
||
fe_sub(r->Y, p->Y, p->X);
|
||
fe_mul(r->Z, r->X, q->YminusX);
|
||
fe_mul(r->Y, r->Y, q->YplusX);
|
||
fe_mul(r->T, q->T2d, p->T);
|
||
fe_mul(r->X, p->Z, q->Z);
|
||
fe_add(t0, r->X, r->X);
|
||
fe_sub(r->X, r->Z, r->Y);
|
||
fe_add(r->Y, r->Z, r->Y);
|
||
fe_sub(r->Z, t0, r->T);
|
||
fe_add(r->T, t0, r->T);
|
||
}
|
||
|
||
/* From ge_tobytes.c */
|
||
|
||
void ge_tobytes(unsigned char *s, const ge_p2 *h) {
|
||
fe recip;
|
||
fe x;
|
||
fe y;
|
||
|
||
fe_invert(recip, h->Z);
|
||
fe_mul(x, h->X, recip);
|
||
fe_mul(y, h->Y, recip);
|
||
fe_tobytes(s, y);
|
||
s[31] ^= fe_isnegative(x) << 7;
|
||
}
|
||
|
||
/* From sc_reduce.c */
|
||
|
||
/*
|
||
Input:
|
||
s[0]+256*s[1]+...+256^63*s[63] = s
|
||
|
||
Output:
|
||
s[0]+256*s[1]+...+256^31*s[31] = s mod l
|
||
where l = 2^252 + 27742317777372353535851937790883648493.
|
||
Overwrites s in place.
|
||
*/
|
||
|
||
void sc_reduce(unsigned char *s) {
|
||
int64_t s0 = 2097151 & load_3(s);
|
||
int64_t s1 = 2097151 & (load_4(s + 2) >> 5);
|
||
int64_t s2 = 2097151 & (load_3(s + 5) >> 2);
|
||
int64_t s3 = 2097151 & (load_4(s + 7) >> 7);
|
||
int64_t s4 = 2097151 & (load_4(s + 10) >> 4);
|
||
int64_t s5 = 2097151 & (load_3(s + 13) >> 1);
|
||
int64_t s6 = 2097151 & (load_4(s + 15) >> 6);
|
||
int64_t s7 = 2097151 & (load_3(s + 18) >> 3);
|
||
int64_t s8 = 2097151 & load_3(s + 21);
|
||
int64_t s9 = 2097151 & (load_4(s + 23) >> 5);
|
||
int64_t s10 = 2097151 & (load_3(s + 26) >> 2);
|
||
int64_t s11 = 2097151 & (load_4(s + 28) >> 7);
|
||
int64_t s12 = 2097151 & (load_4(s + 31) >> 4);
|
||
int64_t s13 = 2097151 & (load_3(s + 34) >> 1);
|
||
int64_t s14 = 2097151 & (load_4(s + 36) >> 6);
|
||
int64_t s15 = 2097151 & (load_3(s + 39) >> 3);
|
||
int64_t s16 = 2097151 & load_3(s + 42);
|
||
int64_t s17 = 2097151 & (load_4(s + 44) >> 5);
|
||
int64_t s18 = 2097151 & (load_3(s + 47) >> 2);
|
||
int64_t s19 = 2097151 & (load_4(s + 49) >> 7);
|
||
int64_t s20 = 2097151 & (load_4(s + 52) >> 4);
|
||
int64_t s21 = 2097151 & (load_3(s + 55) >> 1);
|
||
int64_t s22 = 2097151 & (load_4(s + 57) >> 6);
|
||
int64_t s23 = (load_4(s + 60) >> 3);
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
int64_t carry12;
|
||
int64_t carry13;
|
||
int64_t carry14;
|
||
int64_t carry15;
|
||
int64_t carry16;
|
||
|
||
s11 += s23 * 666643;
|
||
s12 += s23 * 470296;
|
||
s13 += s23 * 654183;
|
||
s14 -= s23 * 997805;
|
||
s15 += s23 * 136657;
|
||
s16 -= s23 * 683901;
|
||
|
||
s10 += s22 * 666643;
|
||
s11 += s22 * 470296;
|
||
s12 += s22 * 654183;
|
||
s13 -= s22 * 997805;
|
||
s14 += s22 * 136657;
|
||
s15 -= s22 * 683901;
|
||
|
||
s9 += s21 * 666643;
|
||
s10 += s21 * 470296;
|
||
s11 += s21 * 654183;
|
||
s12 -= s21 * 997805;
|
||
s13 += s21 * 136657;
|
||
s14 -= s21 * 683901;
|
||
|
||
s8 += s20 * 666643;
|
||
s9 += s20 * 470296;
|
||
s10 += s20 * 654183;
|
||
s11 -= s20 * 997805;
|
||
s12 += s20 * 136657;
|
||
s13 -= s20 * 683901;
|
||
|
||
s7 += s19 * 666643;
|
||
s8 += s19 * 470296;
|
||
s9 += s19 * 654183;
|
||
s10 -= s19 * 997805;
|
||
s11 += s19 * 136657;
|
||
s12 -= s19 * 683901;
|
||
|
||
s6 += s18 * 666643;
|
||
s7 += s18 * 470296;
|
||
s8 += s18 * 654183;
|
||
s9 -= s18 * 997805;
|
||
s10 += s18 * 136657;
|
||
s11 -= s18 * 683901;
|
||
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry12 = (s12 + (1<<20)) >> 21; s13 += carry12; s12 -= carry12 << 21;
|
||
carry14 = (s14 + (1<<20)) >> 21; s15 += carry14; s14 -= carry14 << 21;
|
||
carry16 = (s16 + (1<<20)) >> 21; s17 += carry16; s16 -= carry16 << 21;
|
||
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
carry13 = (s13 + (1<<20)) >> 21; s14 += carry13; s13 -= carry13 << 21;
|
||
carry15 = (s15 + (1<<20)) >> 21; s16 += carry15; s15 -= carry15 << 21;
|
||
|
||
s5 += s17 * 666643;
|
||
s6 += s17 * 470296;
|
||
s7 += s17 * 654183;
|
||
s8 -= s17 * 997805;
|
||
s9 += s17 * 136657;
|
||
s10 -= s17 * 683901;
|
||
|
||
s4 += s16 * 666643;
|
||
s5 += s16 * 470296;
|
||
s6 += s16 * 654183;
|
||
s7 -= s16 * 997805;
|
||
s8 += s16 * 136657;
|
||
s9 -= s16 * 683901;
|
||
|
||
s3 += s15 * 666643;
|
||
s4 += s15 * 470296;
|
||
s5 += s15 * 654183;
|
||
s6 -= s15 * 997805;
|
||
s7 += s15 * 136657;
|
||
s8 -= s15 * 683901;
|
||
|
||
s2 += s14 * 666643;
|
||
s3 += s14 * 470296;
|
||
s4 += s14 * 654183;
|
||
s5 -= s14 * 997805;
|
||
s6 += s14 * 136657;
|
||
s7 -= s14 * 683901;
|
||
|
||
s1 += s13 * 666643;
|
||
s2 += s13 * 470296;
|
||
s3 += s13 * 654183;
|
||
s4 -= s13 * 997805;
|
||
s5 += s13 * 136657;
|
||
s6 -= s13 * 683901;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry11 = s11 >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 << 5);
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 << 2);
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 << 7);
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 << 4);
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 << 1);
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 << 6);
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 << 3);
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 << 5);
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 << 2);
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 << 7);
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
/* New code */
|
||
|
||
static void fe_divpowm1(fe r, const fe u, const fe v) {
|
||
fe v3, uv7, t0, t1, t2;
|
||
int i;
|
||
|
||
fe_sq(v3, v);
|
||
fe_mul(v3, v3, v); /* v3 = v^3 */
|
||
fe_sq(uv7, v3);
|
||
fe_mul(uv7, uv7, v);
|
||
fe_mul(uv7, uv7, u); /* uv7 = uv^7 */
|
||
|
||
/*fe_pow22523(uv7, uv7);*/
|
||
|
||
/* From fe_pow22523.c */
|
||
|
||
fe_sq(t0, uv7);
|
||
fe_sq(t1, t0);
|
||
fe_sq(t1, t1);
|
||
fe_mul(t1, uv7, t1);
|
||
fe_mul(t0, t0, t1);
|
||
fe_sq(t0, t0);
|
||
fe_mul(t0, t1, t0);
|
||
fe_sq(t1, t0);
|
||
for (i = 0; i < 4; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(t0, t1, t0);
|
||
fe_sq(t1, t0);
|
||
for (i = 0; i < 9; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(t1, t1, t0);
|
||
fe_sq(t2, t1);
|
||
for (i = 0; i < 19; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t1, t2, t1);
|
||
for (i = 0; i < 10; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(t0, t1, t0);
|
||
fe_sq(t1, t0);
|
||
for (i = 0; i < 49; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(t1, t1, t0);
|
||
fe_sq(t2, t1);
|
||
for (i = 0; i < 99; ++i) {
|
||
fe_sq(t2, t2);
|
||
}
|
||
fe_mul(t1, t2, t1);
|
||
for (i = 0; i < 50; ++i) {
|
||
fe_sq(t1, t1);
|
||
}
|
||
fe_mul(t0, t1, t0);
|
||
fe_sq(t0, t0);
|
||
fe_sq(t0, t0);
|
||
fe_mul(t0, t0, uv7);
|
||
|
||
/* End fe_pow22523.c */
|
||
/* t0 = (uv^7)^((q-5)/8) */
|
||
fe_mul(t0, t0, v3);
|
||
fe_mul(r, t0, u); /* u^(m+1)v^(-(m+1)) */
|
||
}
|
||
|
||
static void ge_cached_0(ge_cached *r) {
|
||
fe_1(r->YplusX);
|
||
fe_1(r->YminusX);
|
||
fe_1(r->Z);
|
||
fe_0(r->T2d);
|
||
}
|
||
|
||
static void ge_cached_cmov(ge_cached *t, const ge_cached *u, unsigned char b) {
|
||
fe_cmov(t->YplusX, u->YplusX, b);
|
||
fe_cmov(t->YminusX, u->YminusX, b);
|
||
fe_cmov(t->Z, u->Z, b);
|
||
fe_cmov(t->T2d, u->T2d, b);
|
||
}
|
||
|
||
/* Assumes that a[31] <= 127 */
|
||
void ge_scalarmult(ge_p2 *r, const unsigned char *a, const ge_p3 *A) {
|
||
signed char e[64];
|
||
int carry, carry2, i;
|
||
ge_cached Ai[8]; /* 1 * A, 2 * A, ..., 8 * A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
|
||
carry = 0; /* 0..1 */
|
||
for (i = 0; i < 31; i++) {
|
||
carry += a[i]; /* 0..256 */
|
||
carry2 = (carry + 8) >> 4; /* 0..16 */
|
||
e[2 * i] = carry - (carry2 << 4); /* -8..7 */
|
||
carry = (carry2 + 8) >> 4; /* 0..1 */
|
||
e[2 * i + 1] = carry2 - (carry << 4); /* -8..7 */
|
||
}
|
||
carry += a[31]; /* 0..128 */
|
||
carry2 = (carry + 8) >> 4; /* 0..8 */
|
||
e[62] = carry - (carry2 << 4); /* -8..7 */
|
||
e[63] = carry2; /* 0..8 */
|
||
|
||
ge_p3_to_cached(&Ai[0], A);
|
||
for (i = 0; i < 7; i++) {
|
||
ge_add(&t, A, &Ai[i]);
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_p3_to_cached(&Ai[i + 1], &u);
|
||
}
|
||
|
||
ge_p2_0(r);
|
||
for (i = 63; i >= 0; i--) {
|
||
signed char b = e[i];
|
||
unsigned char bnegative = negative(b);
|
||
unsigned char babs = b - (((-bnegative) & b) << 1);
|
||
ge_cached cur, minuscur;
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_cached_0(&cur);
|
||
ge_cached_cmov(&cur, &Ai[0], equal(babs, 1));
|
||
ge_cached_cmov(&cur, &Ai[1], equal(babs, 2));
|
||
ge_cached_cmov(&cur, &Ai[2], equal(babs, 3));
|
||
ge_cached_cmov(&cur, &Ai[3], equal(babs, 4));
|
||
ge_cached_cmov(&cur, &Ai[4], equal(babs, 5));
|
||
ge_cached_cmov(&cur, &Ai[5], equal(babs, 6));
|
||
ge_cached_cmov(&cur, &Ai[6], equal(babs, 7));
|
||
ge_cached_cmov(&cur, &Ai[7], equal(babs, 8));
|
||
fe_copy(minuscur.YplusX, cur.YminusX);
|
||
fe_copy(minuscur.YminusX, cur.YplusX);
|
||
fe_copy(minuscur.Z, cur.Z);
|
||
fe_neg(minuscur.T2d, cur.T2d);
|
||
ge_cached_cmov(&cur, &minuscur, bnegative);
|
||
ge_add(&t, &u, &cur);
|
||
ge_p1p1_to_p2(r, &t);
|
||
}
|
||
}
|
||
|
||
/* Assumes that a[31] <= 127 */
|
||
void ge_scalarmult_p3(ge_p3 *result, const unsigned char *a, const ge_p3 *A) {
|
||
signed char e[64];
|
||
int carry, carry2, i;
|
||
ge_cached Ai[8]; /* 1 * A, 2 * A, ..., 8 * A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
ge_p2 r_tmp;
|
||
ge_p2* r = &r_tmp;
|
||
|
||
carry = 0; /* 0..1 */
|
||
for (i = 0; i < 31; i++) {
|
||
carry += a[i]; /* 0..256 */
|
||
carry2 = (carry + 8) >> 4; /* 0..16 */
|
||
e[2 * i] = carry - (carry2 << 4); /* -8..7 */
|
||
carry = (carry2 + 8) >> 4; /* 0..1 */
|
||
e[2 * i + 1] = carry2 - (carry << 4); /* -8..7 */
|
||
}
|
||
carry += a[31]; /* 0..128 */
|
||
carry2 = (carry + 8) >> 4; /* 0..8 */
|
||
e[62] = carry - (carry2 << 4); /* -8..7 */
|
||
e[63] = carry2; /* 0..8 */
|
||
|
||
ge_p3_to_cached(&Ai[0], A);
|
||
for (i = 0; i < 7; i++) {
|
||
ge_add(&t, A, &Ai[i]);
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_p3_to_cached(&Ai[i + 1], &u);
|
||
}
|
||
|
||
ge_p2_0(r);
|
||
for (i = 63; i >= 0; i--) {
|
||
signed char b = e[i];
|
||
unsigned char bnegative = negative(b);
|
||
unsigned char babs = b - (((-bnegative) & b) << 1);
|
||
ge_cached cur, minuscur;
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p2(r, &t);
|
||
ge_p2_dbl(&t, r);
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_cached_0(&cur);
|
||
ge_cached_cmov(&cur, &Ai[0], equal(babs, 1));
|
||
ge_cached_cmov(&cur, &Ai[1], equal(babs, 2));
|
||
ge_cached_cmov(&cur, &Ai[2], equal(babs, 3));
|
||
ge_cached_cmov(&cur, &Ai[3], equal(babs, 4));
|
||
ge_cached_cmov(&cur, &Ai[4], equal(babs, 5));
|
||
ge_cached_cmov(&cur, &Ai[5], equal(babs, 6));
|
||
ge_cached_cmov(&cur, &Ai[6], equal(babs, 7));
|
||
ge_cached_cmov(&cur, &Ai[7], equal(babs, 8));
|
||
fe_copy(minuscur.YplusX, cur.YminusX);
|
||
fe_copy(minuscur.YminusX, cur.YplusX);
|
||
fe_copy(minuscur.Z, cur.Z);
|
||
fe_neg(minuscur.T2d, cur.T2d);
|
||
ge_cached_cmov(&cur, &minuscur, bnegative);
|
||
ge_add(&t, &u, &cur);
|
||
if (i > 0)
|
||
ge_p1p1_to_p2(r, &t);
|
||
else
|
||
ge_p1p1_to_p3(result, &t); // last iteration
|
||
}
|
||
}
|
||
|
||
void ge_double_scalarmult_precomp_vartime(ge_p2 *r, const unsigned char *a, const ge_p3 *A, const unsigned char *b, const ge_dsmp Bi) {
|
||
signed char aslide[256];
|
||
signed char bslide[256];
|
||
ge_dsmp Ai; /* A, 3A, 5A, 7A, 9A, 11A, 13A, 15A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
int i;
|
||
|
||
slide(aslide, a);
|
||
slide(bslide, b);
|
||
ge_dsm_precomp(Ai, A);
|
||
|
||
ge_p2_0(r);
|
||
|
||
for (i = 255; i >= 0; --i) {
|
||
if (aslide[i] || bslide[i]) break;
|
||
}
|
||
|
||
for (; i >= 0; --i) {
|
||
ge_p2_dbl(&t, r);
|
||
|
||
if (aslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Ai[aslide[i]/2]);
|
||
} else if (aslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Ai[(-aslide[i])/2]);
|
||
}
|
||
|
||
if (bslide[i] > 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Bi[bslide[i]/2]);
|
||
} else if (bslide[i] < 0) {
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Bi[(-bslide[i])/2]);
|
||
}
|
||
|
||
ge_p1p1_to_p2(r, &t);
|
||
}
|
||
}
|
||
|
||
void ge_mul8(ge_p1p1 *r, const ge_p2 *t) {
|
||
ge_p2 u;
|
||
ge_p2_dbl(r, t);
|
||
ge_p1p1_to_p2(&u, r);
|
||
ge_p2_dbl(r, &u);
|
||
ge_p1p1_to_p2(&u, r);
|
||
ge_p2_dbl(r, &u);
|
||
}
|
||
|
||
void ge_fromfe_frombytes_vartime(ge_p2 *r, const unsigned char *s) {
|
||
fe u, v, w, x, y, z;
|
||
unsigned char sign;
|
||
|
||
/* From fe_frombytes.c */
|
||
|
||
int64_t h0 = load_4(s);
|
||
int64_t h1 = load_3(s + 4) << 6;
|
||
int64_t h2 = load_3(s + 7) << 5;
|
||
int64_t h3 = load_3(s + 10) << 3;
|
||
int64_t h4 = load_3(s + 13) << 2;
|
||
int64_t h5 = load_4(s + 16);
|
||
int64_t h6 = load_3(s + 20) << 7;
|
||
int64_t h7 = load_3(s + 23) << 5;
|
||
int64_t h8 = load_3(s + 26) << 4;
|
||
int64_t h9 = load_3(s + 29) << 2;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
carry9 = (h9 + (int64_t) (1<<24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
carry1 = (h1 + (int64_t) (1<<24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry3 = (h3 + (int64_t) (1<<24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry5 = (h5 + (int64_t) (1<<24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
carry7 = (h7 + (int64_t) (1<<24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
|
||
carry0 = (h0 + (int64_t) (1<<25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry2 = (h2 + (int64_t) (1<<25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry4 = (h4 + (int64_t) (1<<25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry6 = (h6 + (int64_t) (1<<25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
carry8 = (h8 + (int64_t) (1<<25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
|
||
u[0] = h0;
|
||
u[1] = h1;
|
||
u[2] = h2;
|
||
u[3] = h3;
|
||
u[4] = h4;
|
||
u[5] = h5;
|
||
u[6] = h6;
|
||
u[7] = h7;
|
||
u[8] = h8;
|
||
u[9] = h9;
|
||
|
||
/* End fe_frombytes.c */
|
||
|
||
fe_sq2(v, u); /* v = 2 * u^2 */
|
||
fe_1(w); /* w = 1 */
|
||
fe_add(w, v, w); /* w = 2 * u^2 + 1 */
|
||
fe_sq(x, w); /* x = w^2 */
|
||
fe_mul(y, fe_ma2, v); /* y = -A^2 * (2 * u^2) = -2 * A^2 * u^2 n.b. A = 2 * (1 - d) / (1 + d) = 486662*/
|
||
fe_add(x, x, y); /* x = w^2 - 2 * A^2 * u^2 */
|
||
fe_divpowm1(r->X, w, x); /* r->X = (w / x)^(m + 1) */
|
||
fe_sq(y, r->X); /* y = (w / x)^(2*(m + 1)) */
|
||
fe_mul(x, y, x); /* x = (w / x)^(2*(m + 1)) * (w^2 - 2 * A^2 * u^2) */
|
||
fe_sub(y, w, x); /* y = 2 * u^2 + 1 - (w / x)^(2*(m + 1)) * (w^2 - 2 * A^2 * u^2) */
|
||
fe_copy(z, fe_ma); /* z = -A */
|
||
if (fe_isnonzero(y)) {
|
||
fe_add(y, w, x);
|
||
if (fe_isnonzero(y)) {
|
||
goto negative;
|
||
} else {
|
||
fe_mul(r->X, r->X, fe_fffb1);
|
||
}
|
||
} else {
|
||
fe_mul(r->X, r->X, fe_fffb2);
|
||
}
|
||
fe_mul(r->X, r->X, u); /* u * sqrt(2 * A * (A + 2) * w / x) */
|
||
fe_mul(z, z, v); /* -2 * A * u^2 */
|
||
sign = 0;
|
||
goto setsign;
|
||
negative:
|
||
fe_mul(x, x, fe_sqrtm1);
|
||
fe_sub(y, w, x);
|
||
if (fe_isnonzero(y)) {
|
||
assert((fe_add(y, w, x), !fe_isnonzero(y)));
|
||
fe_mul(r->X, r->X, fe_fffb3);
|
||
} else {
|
||
fe_mul(r->X, r->X, fe_fffb4);
|
||
}
|
||
/* r->X = sqrt(A * (A + 2) * w / x) */
|
||
/* z = -A */
|
||
sign = 1;
|
||
setsign:
|
||
if (fe_isnegative(r->X) != sign) {
|
||
assert(fe_isnonzero(r->X));
|
||
fe_neg(r->X, r->X);
|
||
}
|
||
fe_add(r->Z, z, w);
|
||
fe_sub(r->Y, z, w);
|
||
fe_mul(r->X, r->X, r->Z);
|
||
#if !defined(NDEBUG)
|
||
{
|
||
fe check_x, check_y, check_iz, check_v;
|
||
fe_invert(check_iz, r->Z);
|
||
fe_mul(check_x, r->X, check_iz);
|
||
fe_mul(check_y, r->Y, check_iz);
|
||
fe_sq(check_x, check_x);
|
||
fe_sq(check_y, check_y);
|
||
fe_mul(check_v, check_x, check_y);
|
||
fe_mul(check_v, fe_d, check_v);
|
||
fe_add(check_v, check_v, check_x);
|
||
fe_sub(check_v, check_v, check_y);
|
||
fe_1(check_x);
|
||
fe_add(check_v, check_v, check_x);
|
||
assert(!fe_isnonzero(check_v));
|
||
}
|
||
#endif
|
||
}
|
||
|
||
void sc_0(unsigned char *s) {
|
||
int i;
|
||
for (i = 0; i < 32; i++) {
|
||
s[i] = 0;
|
||
}
|
||
}
|
||
|
||
void sc_reduce32(unsigned char *s) {
|
||
int64_t s0 = 2097151 & load_3(s);
|
||
int64_t s1 = 2097151 & (load_4(s + 2) >> 5);
|
||
int64_t s2 = 2097151 & (load_3(s + 5) >> 2);
|
||
int64_t s3 = 2097151 & (load_4(s + 7) >> 7);
|
||
int64_t s4 = 2097151 & (load_4(s + 10) >> 4);
|
||
int64_t s5 = 2097151 & (load_3(s + 13) >> 1);
|
||
int64_t s6 = 2097151 & (load_4(s + 15) >> 6);
|
||
int64_t s7 = 2097151 & (load_3(s + 18) >> 3);
|
||
int64_t s8 = 2097151 & load_3(s + 21);
|
||
int64_t s9 = 2097151 & (load_4(s + 23) >> 5);
|
||
int64_t s10 = 2097151 & (load_3(s + 26) >> 2);
|
||
int64_t s11 = (load_4(s + 28) >> 7);
|
||
int64_t s12 = 0;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry11 = s11 >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 << 5);
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 << 2);
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 << 7);
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 << 4);
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 << 1);
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 << 6);
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 << 3);
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 << 5);
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 << 2);
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 << 7);
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
void sc_add(unsigned char *s, const unsigned char *a, const unsigned char *b) {
|
||
int64_t a0 = 2097151 & load_3(a);
|
||
int64_t a1 = 2097151 & (load_4(a + 2) >> 5);
|
||
int64_t a2 = 2097151 & (load_3(a + 5) >> 2);
|
||
int64_t a3 = 2097151 & (load_4(a + 7) >> 7);
|
||
int64_t a4 = 2097151 & (load_4(a + 10) >> 4);
|
||
int64_t a5 = 2097151 & (load_3(a + 13) >> 1);
|
||
int64_t a6 = 2097151 & (load_4(a + 15) >> 6);
|
||
int64_t a7 = 2097151 & (load_3(a + 18) >> 3);
|
||
int64_t a8 = 2097151 & load_3(a + 21);
|
||
int64_t a9 = 2097151 & (load_4(a + 23) >> 5);
|
||
int64_t a10 = 2097151 & (load_3(a + 26) >> 2);
|
||
int64_t a11 = (load_4(a + 28) >> 7);
|
||
int64_t b0 = 2097151 & load_3(b);
|
||
int64_t b1 = 2097151 & (load_4(b + 2) >> 5);
|
||
int64_t b2 = 2097151 & (load_3(b + 5) >> 2);
|
||
int64_t b3 = 2097151 & (load_4(b + 7) >> 7);
|
||
int64_t b4 = 2097151 & (load_4(b + 10) >> 4);
|
||
int64_t b5 = 2097151 & (load_3(b + 13) >> 1);
|
||
int64_t b6 = 2097151 & (load_4(b + 15) >> 6);
|
||
int64_t b7 = 2097151 & (load_3(b + 18) >> 3);
|
||
int64_t b8 = 2097151 & load_3(b + 21);
|
||
int64_t b9 = 2097151 & (load_4(b + 23) >> 5);
|
||
int64_t b10 = 2097151 & (load_3(b + 26) >> 2);
|
||
int64_t b11 = (load_4(b + 28) >> 7);
|
||
int64_t s0 = a0 + b0;
|
||
int64_t s1 = a1 + b1;
|
||
int64_t s2 = a2 + b2;
|
||
int64_t s3 = a3 + b3;
|
||
int64_t s4 = a4 + b4;
|
||
int64_t s5 = a5 + b5;
|
||
int64_t s6 = a6 + b6;
|
||
int64_t s7 = a7 + b7;
|
||
int64_t s8 = a8 + b8;
|
||
int64_t s9 = a9 + b9;
|
||
int64_t s10 = a10 + b10;
|
||
int64_t s11 = a11 + b11;
|
||
int64_t s12 = 0;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry11 = s11 >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 << 5);
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 << 2);
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 << 7);
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 << 4);
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 << 1);
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 << 6);
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 << 3);
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 << 5);
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 << 2);
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 << 7);
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
void sc_sub(unsigned char *s, const unsigned char *a, const unsigned char *b) {
|
||
int64_t a0 = 2097151 & load_3(a);
|
||
int64_t a1 = 2097151 & (load_4(a + 2) >> 5);
|
||
int64_t a2 = 2097151 & (load_3(a + 5) >> 2);
|
||
int64_t a3 = 2097151 & (load_4(a + 7) >> 7);
|
||
int64_t a4 = 2097151 & (load_4(a + 10) >> 4);
|
||
int64_t a5 = 2097151 & (load_3(a + 13) >> 1);
|
||
int64_t a6 = 2097151 & (load_4(a + 15) >> 6);
|
||
int64_t a7 = 2097151 & (load_3(a + 18) >> 3);
|
||
int64_t a8 = 2097151 & load_3(a + 21);
|
||
int64_t a9 = 2097151 & (load_4(a + 23) >> 5);
|
||
int64_t a10 = 2097151 & (load_3(a + 26) >> 2);
|
||
int64_t a11 = (load_4(a + 28) >> 7);
|
||
int64_t b0 = 2097151 & load_3(b);
|
||
int64_t b1 = 2097151 & (load_4(b + 2) >> 5);
|
||
int64_t b2 = 2097151 & (load_3(b + 5) >> 2);
|
||
int64_t b3 = 2097151 & (load_4(b + 7) >> 7);
|
||
int64_t b4 = 2097151 & (load_4(b + 10) >> 4);
|
||
int64_t b5 = 2097151 & (load_3(b + 13) >> 1);
|
||
int64_t b6 = 2097151 & (load_4(b + 15) >> 6);
|
||
int64_t b7 = 2097151 & (load_3(b + 18) >> 3);
|
||
int64_t b8 = 2097151 & load_3(b + 21);
|
||
int64_t b9 = 2097151 & (load_4(b + 23) >> 5);
|
||
int64_t b10 = 2097151 & (load_3(b + 26) >> 2);
|
||
int64_t b11 = (load_4(b + 28) >> 7);
|
||
int64_t s0 = a0 - b0;
|
||
int64_t s1 = a1 - b1;
|
||
int64_t s2 = a2 - b2;
|
||
int64_t s3 = a3 - b3;
|
||
int64_t s4 = a4 - b4;
|
||
int64_t s5 = a5 - b5;
|
||
int64_t s6 = a6 - b6;
|
||
int64_t s7 = a7 - b7;
|
||
int64_t s8 = a8 - b8;
|
||
int64_t s9 = a9 - b9;
|
||
int64_t s10 = a10 - b10;
|
||
int64_t s11 = a11 - b11;
|
||
int64_t s12 = 0;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry11 = s11 >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 << 5);
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 << 2);
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 << 7);
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 << 4);
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 << 1);
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 << 6);
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 << 3);
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 << 5);
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 << 2);
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 << 7);
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
/*
|
||
Input:
|
||
a[0]+256*a[1]+...+256^31*a[31] = a
|
||
b[0]+256*b[1]+...+256^31*b[31] = b
|
||
c[0]+256*c[1]+...+256^31*c[31] = c
|
||
|
||
Output:
|
||
s[0]+256*s[1]+...+256^31*s[31] = (c-ab) mod l
|
||
where l = 2^252 + 27742317777372353535851937790883648493.
|
||
*/
|
||
|
||
void sc_mulsub(unsigned char *s, const unsigned char *a, const unsigned char *b, const unsigned char *c) {
|
||
int64_t a0 = 2097151 & load_3(a);
|
||
int64_t a1 = 2097151 & (load_4(a + 2) >> 5);
|
||
int64_t a2 = 2097151 & (load_3(a + 5) >> 2);
|
||
int64_t a3 = 2097151 & (load_4(a + 7) >> 7);
|
||
int64_t a4 = 2097151 & (load_4(a + 10) >> 4);
|
||
int64_t a5 = 2097151 & (load_3(a + 13) >> 1);
|
||
int64_t a6 = 2097151 & (load_4(a + 15) >> 6);
|
||
int64_t a7 = 2097151 & (load_3(a + 18) >> 3);
|
||
int64_t a8 = 2097151 & load_3(a + 21);
|
||
int64_t a9 = 2097151 & (load_4(a + 23) >> 5);
|
||
int64_t a10 = 2097151 & (load_3(a + 26) >> 2);
|
||
int64_t a11 = (load_4(a + 28) >> 7);
|
||
int64_t b0 = 2097151 & load_3(b);
|
||
int64_t b1 = 2097151 & (load_4(b + 2) >> 5);
|
||
int64_t b2 = 2097151 & (load_3(b + 5) >> 2);
|
||
int64_t b3 = 2097151 & (load_4(b + 7) >> 7);
|
||
int64_t b4 = 2097151 & (load_4(b + 10) >> 4);
|
||
int64_t b5 = 2097151 & (load_3(b + 13) >> 1);
|
||
int64_t b6 = 2097151 & (load_4(b + 15) >> 6);
|
||
int64_t b7 = 2097151 & (load_3(b + 18) >> 3);
|
||
int64_t b8 = 2097151 & load_3(b + 21);
|
||
int64_t b9 = 2097151 & (load_4(b + 23) >> 5);
|
||
int64_t b10 = 2097151 & (load_3(b + 26) >> 2);
|
||
int64_t b11 = (load_4(b + 28) >> 7);
|
||
int64_t c0 = 2097151 & load_3(c);
|
||
int64_t c1 = 2097151 & (load_4(c + 2) >> 5);
|
||
int64_t c2 = 2097151 & (load_3(c + 5) >> 2);
|
||
int64_t c3 = 2097151 & (load_4(c + 7) >> 7);
|
||
int64_t c4 = 2097151 & (load_4(c + 10) >> 4);
|
||
int64_t c5 = 2097151 & (load_3(c + 13) >> 1);
|
||
int64_t c6 = 2097151 & (load_4(c + 15) >> 6);
|
||
int64_t c7 = 2097151 & (load_3(c + 18) >> 3);
|
||
int64_t c8 = 2097151 & load_3(c + 21);
|
||
int64_t c9 = 2097151 & (load_4(c + 23) >> 5);
|
||
int64_t c10 = 2097151 & (load_3(c + 26) >> 2);
|
||
int64_t c11 = (load_4(c + 28) >> 7);
|
||
int64_t s0;
|
||
int64_t s1;
|
||
int64_t s2;
|
||
int64_t s3;
|
||
int64_t s4;
|
||
int64_t s5;
|
||
int64_t s6;
|
||
int64_t s7;
|
||
int64_t s8;
|
||
int64_t s9;
|
||
int64_t s10;
|
||
int64_t s11;
|
||
int64_t s12;
|
||
int64_t s13;
|
||
int64_t s14;
|
||
int64_t s15;
|
||
int64_t s16;
|
||
int64_t s17;
|
||
int64_t s18;
|
||
int64_t s19;
|
||
int64_t s20;
|
||
int64_t s21;
|
||
int64_t s22;
|
||
int64_t s23;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
int64_t carry12;
|
||
int64_t carry13;
|
||
int64_t carry14;
|
||
int64_t carry15;
|
||
int64_t carry16;
|
||
int64_t carry17;
|
||
int64_t carry18;
|
||
int64_t carry19;
|
||
int64_t carry20;
|
||
int64_t carry21;
|
||
int64_t carry22;
|
||
|
||
s0 = c0 - a0*b0;
|
||
s1 = c1 - (a0*b1 + a1*b0);
|
||
s2 = c2 - (a0*b2 + a1*b1 + a2*b0);
|
||
s3 = c3 - (a0*b3 + a1*b2 + a2*b1 + a3*b0);
|
||
s4 = c4 - (a0*b4 + a1*b3 + a2*b2 + a3*b1 + a4*b0);
|
||
s5 = c5 - (a0*b5 + a1*b4 + a2*b3 + a3*b2 + a4*b1 + a5*b0);
|
||
s6 = c6 - (a0*b6 + a1*b5 + a2*b4 + a3*b3 + a4*b2 + a5*b1 + a6*b0);
|
||
s7 = c7 - (a0*b7 + a1*b6 + a2*b5 + a3*b4 + a4*b3 + a5*b2 + a6*b1 + a7*b0);
|
||
s8 = c8 - (a0*b8 + a1*b7 + a2*b6 + a3*b5 + a4*b4 + a5*b3 + a6*b2 + a7*b1 + a8*b0);
|
||
s9 = c9 - (a0*b9 + a1*b8 + a2*b7 + a3*b6 + a4*b5 + a5*b4 + a6*b3 + a7*b2 + a8*b1 + a9*b0);
|
||
s10 = c10 - (a0*b10 + a1*b9 + a2*b8 + a3*b7 + a4*b6 + a5*b5 + a6*b4 + a7*b3 + a8*b2 + a9*b1 + a10*b0);
|
||
s11 = c11 - (a0*b11 + a1*b10 + a2*b9 + a3*b8 + a4*b7 + a5*b6 + a6*b5 + a7*b4 + a8*b3 + a9*b2 + a10*b1 + a11*b0);
|
||
s12 = -(a1*b11 + a2*b10 + a3*b9 + a4*b8 + a5*b7 + a6*b6 + a7*b5 + a8*b4 + a9*b3 + a10*b2 + a11*b1);
|
||
s13 = -(a2*b11 + a3*b10 + a4*b9 + a5*b8 + a6*b7 + a7*b6 + a8*b5 + a9*b4 + a10*b3 + a11*b2);
|
||
s14 = -(a3*b11 + a4*b10 + a5*b9 + a6*b8 + a7*b7 + a8*b6 + a9*b5 + a10*b4 + a11*b3);
|
||
s15 = -(a4*b11 + a5*b10 + a6*b9 + a7*b8 + a8*b7 + a9*b6 + a10*b5 + a11*b4);
|
||
s16 = -(a5*b11 + a6*b10 + a7*b9 + a8*b8 + a9*b7 + a10*b6 + a11*b5);
|
||
s17 = -(a6*b11 + a7*b10 + a8*b9 + a9*b8 + a10*b7 + a11*b6);
|
||
s18 = -(a7*b11 + a8*b10 + a9*b9 + a10*b8 + a11*b7);
|
||
s19 = -(a8*b11 + a9*b10 + a10*b9 + a11*b8);
|
||
s20 = -(a9*b11 + a10*b10 + a11*b9);
|
||
s21 = -(a10*b11 + a11*b10);
|
||
s22 = -a11*b11;
|
||
s23 = 0;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry12 = (s12 + (1<<20)) >> 21; s13 += carry12; s12 -= carry12 << 21;
|
||
carry14 = (s14 + (1<<20)) >> 21; s15 += carry14; s14 -= carry14 << 21;
|
||
carry16 = (s16 + (1<<20)) >> 21; s17 += carry16; s16 -= carry16 << 21;
|
||
carry18 = (s18 + (1<<20)) >> 21; s19 += carry18; s18 -= carry18 << 21;
|
||
carry20 = (s20 + (1<<20)) >> 21; s21 += carry20; s20 -= carry20 << 21;
|
||
carry22 = (s22 + (1<<20)) >> 21; s23 += carry22; s22 -= carry22 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
carry13 = (s13 + (1<<20)) >> 21; s14 += carry13; s13 -= carry13 << 21;
|
||
carry15 = (s15 + (1<<20)) >> 21; s16 += carry15; s15 -= carry15 << 21;
|
||
carry17 = (s17 + (1<<20)) >> 21; s18 += carry17; s17 -= carry17 << 21;
|
||
carry19 = (s19 + (1<<20)) >> 21; s20 += carry19; s19 -= carry19 << 21;
|
||
carry21 = (s21 + (1<<20)) >> 21; s22 += carry21; s21 -= carry21 << 21;
|
||
|
||
s11 += s23 * 666643;
|
||
s12 += s23 * 470296;
|
||
s13 += s23 * 654183;
|
||
s14 -= s23 * 997805;
|
||
s15 += s23 * 136657;
|
||
s16 -= s23 * 683901;
|
||
|
||
s10 += s22 * 666643;
|
||
s11 += s22 * 470296;
|
||
s12 += s22 * 654183;
|
||
s13 -= s22 * 997805;
|
||
s14 += s22 * 136657;
|
||
s15 -= s22 * 683901;
|
||
|
||
s9 += s21 * 666643;
|
||
s10 += s21 * 470296;
|
||
s11 += s21 * 654183;
|
||
s12 -= s21 * 997805;
|
||
s13 += s21 * 136657;
|
||
s14 -= s21 * 683901;
|
||
|
||
s8 += s20 * 666643;
|
||
s9 += s20 * 470296;
|
||
s10 += s20 * 654183;
|
||
s11 -= s20 * 997805;
|
||
s12 += s20 * 136657;
|
||
s13 -= s20 * 683901;
|
||
|
||
s7 += s19 * 666643;
|
||
s8 += s19 * 470296;
|
||
s9 += s19 * 654183;
|
||
s10 -= s19 * 997805;
|
||
s11 += s19 * 136657;
|
||
s12 -= s19 * 683901;
|
||
|
||
s6 += s18 * 666643;
|
||
s7 += s18 * 470296;
|
||
s8 += s18 * 654183;
|
||
s9 -= s18 * 997805;
|
||
s10 += s18 * 136657;
|
||
s11 -= s18 * 683901;
|
||
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry12 = (s12 + (1<<20)) >> 21; s13 += carry12; s12 -= carry12 << 21;
|
||
carry14 = (s14 + (1<<20)) >> 21; s15 += carry14; s14 -= carry14 << 21;
|
||
carry16 = (s16 + (1<<20)) >> 21; s17 += carry16; s16 -= carry16 << 21;
|
||
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
carry13 = (s13 + (1<<20)) >> 21; s14 += carry13; s13 -= carry13 << 21;
|
||
carry15 = (s15 + (1<<20)) >> 21; s16 += carry15; s15 -= carry15 << 21;
|
||
|
||
s5 += s17 * 666643;
|
||
s6 += s17 * 470296;
|
||
s7 += s17 * 654183;
|
||
s8 -= s17 * 997805;
|
||
s9 += s17 * 136657;
|
||
s10 -= s17 * 683901;
|
||
|
||
s4 += s16 * 666643;
|
||
s5 += s16 * 470296;
|
||
s6 += s16 * 654183;
|
||
s7 -= s16 * 997805;
|
||
s8 += s16 * 136657;
|
||
s9 -= s16 * 683901;
|
||
|
||
s3 += s15 * 666643;
|
||
s4 += s15 * 470296;
|
||
s5 += s15 * 654183;
|
||
s6 -= s15 * 997805;
|
||
s7 += s15 * 136657;
|
||
s8 -= s15 * 683901;
|
||
|
||
s2 += s14 * 666643;
|
||
s3 += s14 * 470296;
|
||
s4 += s14 * 654183;
|
||
s5 -= s14 * 997805;
|
||
s6 += s14 * 136657;
|
||
s7 -= s14 * 683901;
|
||
|
||
s1 += s13 * 666643;
|
||
s2 += s13 * 470296;
|
||
s3 += s13 * 654183;
|
||
s4 -= s13 * 997805;
|
||
s5 += s13 * 136657;
|
||
s6 -= s13 * 683901;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = (s0 + (1<<20)) >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry2 = (s2 + (1<<20)) >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry4 = (s4 + (1<<20)) >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry6 = (s6 + (1<<20)) >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry8 = (s8 + (1<<20)) >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry10 = (s10 + (1<<20)) >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
carry1 = (s1 + (1<<20)) >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry3 = (s3 + (1<<20)) >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry5 = (s5 + (1<<20)) >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry7 = (s7 + (1<<20)) >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry9 = (s9 + (1<<20)) >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry11 = (s11 + (1<<20)) >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
carry11 = s11 >> 21; s12 += carry11; s11 -= carry11 << 21;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21; s1 += carry0; s0 -= carry0 << 21;
|
||
carry1 = s1 >> 21; s2 += carry1; s1 -= carry1 << 21;
|
||
carry2 = s2 >> 21; s3 += carry2; s2 -= carry2 << 21;
|
||
carry3 = s3 >> 21; s4 += carry3; s3 -= carry3 << 21;
|
||
carry4 = s4 >> 21; s5 += carry4; s4 -= carry4 << 21;
|
||
carry5 = s5 >> 21; s6 += carry5; s5 -= carry5 << 21;
|
||
carry6 = s6 >> 21; s7 += carry6; s6 -= carry6 << 21;
|
||
carry7 = s7 >> 21; s8 += carry7; s7 -= carry7 << 21;
|
||
carry8 = s8 >> 21; s9 += carry8; s8 -= carry8 << 21;
|
||
carry9 = s9 >> 21; s10 += carry9; s9 -= carry9 << 21;
|
||
carry10 = s10 >> 21; s11 += carry10; s10 -= carry10 << 21;
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 << 5);
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 << 2);
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 << 7);
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 << 4);
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 << 1);
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 << 6);
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 << 3);
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 << 5);
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 << 2);
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 << 7);
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
/* Assumes that a != INT64_MIN */
|
||
static int64_t signum(int64_t a) {
|
||
return (a >> 63) - ((-a) >> 63);
|
||
}
|
||
|
||
int sc_check(const unsigned char *s) {
|
||
int64_t s0 = load_4(s);
|
||
int64_t s1 = load_4(s + 4);
|
||
int64_t s2 = load_4(s + 8);
|
||
int64_t s3 = load_4(s + 12);
|
||
int64_t s4 = load_4(s + 16);
|
||
int64_t s5 = load_4(s + 20);
|
||
int64_t s6 = load_4(s + 24);
|
||
int64_t s7 = load_4(s + 28);
|
||
return (signum(1559614444 - s0) + (signum(1477600026 - s1) << 1) + (signum(2734136534 - s2) << 2) + (signum(350157278 - s3) << 3) + (signum(-s4) << 4) + (signum(-s5) << 5) + (signum(-s6) << 6) + (signum(268435456 - s7) << 7)) >> 8;
|
||
}
|
||
|
||
int sc_isnonzero(const unsigned char *s) {
|
||
return (((int) (s[0] | s[1] | s[2] | s[3] | s[4] | s[5] | s[6] | s[7] | s[8] |
|
||
s[9] | s[10] | s[11] | s[12] | s[13] | s[14] | s[15] | s[16] | s[17] |
|
||
s[18] | s[19] | s[20] | s[21] | s[22] | s[23] | s[24] | s[25] | s[26] |
|
||
s[27] | s[28] | s[29] | s[30] | s[31]) - 1) >> 8) + 1;
|
||
}
|
||
|
||
// see implmentation of ge_frombytes_vartime above
|
||
void fe_frombytes(fe h, const unsigned char *s)
|
||
{
|
||
/* From fe_frombytes.c */
|
||
|
||
int64_t h0 = load_4(s);
|
||
int64_t h1 = load_3(s + 4) << 6;
|
||
int64_t h2 = load_3(s + 7) << 5;
|
||
int64_t h3 = load_3(s + 10) << 3;
|
||
int64_t h4 = load_3(s + 13) << 2;
|
||
int64_t h5 = load_4(s + 16);
|
||
int64_t h6 = load_3(s + 20) << 7;
|
||
int64_t h7 = load_3(s + 23) << 5;
|
||
int64_t h8 = load_3(s + 26) << 4;
|
||
int64_t h9 = (load_3(s + 29) & 8388607) << 2;
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
|
||
carry9 = (h9 + (int64_t)(1 << 24)) >> 25; h0 += carry9 * 19; h9 -= carry9 << 25;
|
||
carry1 = (h1 + (int64_t)(1 << 24)) >> 25; h2 += carry1; h1 -= carry1 << 25;
|
||
carry3 = (h3 + (int64_t)(1 << 24)) >> 25; h4 += carry3; h3 -= carry3 << 25;
|
||
carry5 = (h5 + (int64_t)(1 << 24)) >> 25; h6 += carry5; h5 -= carry5 << 25;
|
||
carry7 = (h7 + (int64_t)(1 << 24)) >> 25; h8 += carry7; h7 -= carry7 << 25;
|
||
|
||
carry0 = (h0 + (int64_t)(1 << 25)) >> 26; h1 += carry0; h0 -= carry0 << 26;
|
||
carry2 = (h2 + (int64_t)(1 << 25)) >> 26; h3 += carry2; h2 -= carry2 << 26;
|
||
carry4 = (h4 + (int64_t)(1 << 25)) >> 26; h5 += carry4; h4 -= carry4 << 26;
|
||
carry6 = (h6 + (int64_t)(1 << 25)) >> 26; h7 += carry6; h6 -= carry6 << 26;
|
||
carry8 = (h8 + (int64_t)(1 << 25)) >> 26; h9 += carry8; h8 -= carry8 << 26;
|
||
|
||
h[0] = h0;
|
||
h[1] = h1;
|
||
h[2] = h2;
|
||
h[3] = h3;
|
||
h[4] = h4;
|
||
h[5] = h5;
|
||
h[6] = h6;
|
||
h[7] = h7;
|
||
h[8] = h8;
|
||
h[9] = h9;
|
||
}
|
||
|
||
/*
|
||
// Old naive and slow implementation using sc_mulsub
|
||
void sc_mul(unsigned char *s, const unsigned char *a, const unsigned char *b)
|
||
{
|
||
unsigned char c[32];
|
||
unsigned char neg_a[32];
|
||
sc_0(c);
|
||
sc_sub(neg_a, c, a);
|
||
// s = c - ab
|
||
sc_mulsub(s, neg_a, b, c);
|
||
}
|
||
*/
|
||
|
||
/* libsodium/crypto_core/ed25519/ref10/ed25519_ref10.c */
|
||
/*
|
||
Input:
|
||
a[0]+256*a[1]+...+256^31*a[31] = a
|
||
b[0]+256*b[1]+...+256^31*b[31] = b
|
||
*
|
||
Output:
|
||
s[0]+256*s[1]+...+256^31*s[31] = (ab) mod l
|
||
where l = 2^252 + 27742317777372353535851937790883648493.
|
||
*/
|
||
void sc_mul(unsigned char* s, const unsigned char* a, const unsigned char* b)
|
||
{
|
||
int64_t a0 = 2097151 & load_3(a);
|
||
int64_t a1 = 2097151 & (load_4(a + 2) >> 5);
|
||
int64_t a2 = 2097151 & (load_3(a + 5) >> 2);
|
||
int64_t a3 = 2097151 & (load_4(a + 7) >> 7);
|
||
int64_t a4 = 2097151 & (load_4(a + 10) >> 4);
|
||
int64_t a5 = 2097151 & (load_3(a + 13) >> 1);
|
||
int64_t a6 = 2097151 & (load_4(a + 15) >> 6);
|
||
int64_t a7 = 2097151 & (load_3(a + 18) >> 3);
|
||
int64_t a8 = 2097151 & load_3(a + 21);
|
||
int64_t a9 = 2097151 & (load_4(a + 23) >> 5);
|
||
int64_t a10 = 2097151 & (load_3(a + 26) >> 2);
|
||
int64_t a11 = (load_4(a + 28) >> 7);
|
||
|
||
int64_t b0 = 2097151 & load_3(b);
|
||
int64_t b1 = 2097151 & (load_4(b + 2) >> 5);
|
||
int64_t b2 = 2097151 & (load_3(b + 5) >> 2);
|
||
int64_t b3 = 2097151 & (load_4(b + 7) >> 7);
|
||
int64_t b4 = 2097151 & (load_4(b + 10) >> 4);
|
||
int64_t b5 = 2097151 & (load_3(b + 13) >> 1);
|
||
int64_t b6 = 2097151 & (load_4(b + 15) >> 6);
|
||
int64_t b7 = 2097151 & (load_3(b + 18) >> 3);
|
||
int64_t b8 = 2097151 & load_3(b + 21);
|
||
int64_t b9 = 2097151 & (load_4(b + 23) >> 5);
|
||
int64_t b10 = 2097151 & (load_3(b + 26) >> 2);
|
||
int64_t b11 = (load_4(b + 28) >> 7);
|
||
|
||
int64_t s0;
|
||
int64_t s1;
|
||
int64_t s2;
|
||
int64_t s3;
|
||
int64_t s4;
|
||
int64_t s5;
|
||
int64_t s6;
|
||
int64_t s7;
|
||
int64_t s8;
|
||
int64_t s9;
|
||
int64_t s10;
|
||
int64_t s11;
|
||
int64_t s12;
|
||
int64_t s13;
|
||
int64_t s14;
|
||
int64_t s15;
|
||
int64_t s16;
|
||
int64_t s17;
|
||
int64_t s18;
|
||
int64_t s19;
|
||
int64_t s20;
|
||
int64_t s21;
|
||
int64_t s22;
|
||
int64_t s23;
|
||
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
int64_t carry12;
|
||
int64_t carry13;
|
||
int64_t carry14;
|
||
int64_t carry15;
|
||
int64_t carry16;
|
||
int64_t carry17;
|
||
int64_t carry18;
|
||
int64_t carry19;
|
||
int64_t carry20;
|
||
int64_t carry21;
|
||
int64_t carry22;
|
||
|
||
s0 = a0 * b0;
|
||
s1 = a0 * b1 + a1 * b0;
|
||
s2 = a0 * b2 + a1 * b1 + a2 * b0;
|
||
s3 = a0 * b3 + a1 * b2 + a2 * b1 + a3 * b0;
|
||
s4 = a0 * b4 + a1 * b3 + a2 * b2 + a3 * b1 + a4 * b0;
|
||
s5 = a0 * b5 + a1 * b4 + a2 * b3 + a3 * b2 + a4 * b1 + a5 * b0;
|
||
s6 = a0 * b6 + a1 * b5 + a2 * b4 + a3 * b3 + a4 * b2 + a5 * b1 + a6 * b0;
|
||
s7 = a0 * b7 + a1 * b6 + a2 * b5 + a3 * b4 + a4 * b3 + a5 * b2 +
|
||
a6 * b1 + a7 * b0;
|
||
s8 = a0 * b8 + a1 * b7 + a2 * b6 + a3 * b5 + a4 * b4 + a5 * b3 +
|
||
a6 * b2 + a7 * b1 + a8 * b0;
|
||
s9 = a0 * b9 + a1 * b8 + a2 * b7 + a3 * b6 + a4 * b5 + a5 * b4 +
|
||
a6 * b3 + a7 * b2 + a8 * b1 + a9 * b0;
|
||
s10 = a0 * b10 + a1 * b9 + a2 * b8 + a3 * b7 + a4 * b6 + a5 * b5 +
|
||
a6 * b4 + a7 * b3 + a8 * b2 + a9 * b1 + a10 * b0;
|
||
s11 = a0 * b11 + a1 * b10 + a2 * b9 + a3 * b8 + a4 * b7 + a5 * b6 +
|
||
a6 * b5 + a7 * b4 + a8 * b3 + a9 * b2 + a10 * b1 + a11 * b0;
|
||
s12 = a1 * b11 + a2 * b10 + a3 * b9 + a4 * b8 + a5 * b7 + a6 * b6 +
|
||
a7 * b5 + a8 * b4 + a9 * b3 + a10 * b2 + a11 * b1;
|
||
s13 = a2 * b11 + a3 * b10 + a4 * b9 + a5 * b8 + a6 * b7 + a7 * b6 +
|
||
a8 * b5 + a9 * b4 + a10 * b3 + a11 * b2;
|
||
s14 = a3 * b11 + a4 * b10 + a5 * b9 + a6 * b8 + a7 * b7 + a8 * b6 +
|
||
a9 * b5 + a10 * b4 + a11 * b3;
|
||
s15 = a4 * b11 + a5 * b10 + a6 * b9 + a7 * b8 + a8 * b7 + a9 * b6 +
|
||
a10 * b5 + a11 * b4;
|
||
s16 =
|
||
a5 * b11 + a6 * b10 + a7 * b9 + a8 * b8 + a9 * b7 + a10 * b6 + a11 * b5;
|
||
s17 = a6 * b11 + a7 * b10 + a8 * b9 + a9 * b8 + a10 * b7 + a11 * b6;
|
||
s18 = a7 * b11 + a8 * b10 + a9 * b9 + a10 * b8 + a11 * b7;
|
||
s19 = a8 * b11 + a9 * b10 + a10 * b9 + a11 * b8;
|
||
s20 = a9 * b11 + a10 * b10 + a11 * b9;
|
||
s21 = a10 * b11 + a11 * b10;
|
||
s22 = a11 * b11;
|
||
s23 = 0;
|
||
|
||
carry0 = (s0 + (int64_t)(1L << 20)) >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry2 = (s2 + (int64_t)(1L << 20)) >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry4 = (s4 + (int64_t)(1L << 20)) >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry12 = (s12 + (int64_t)(1L << 20)) >> 21;
|
||
s13 += carry12;
|
||
s12 -= carry12 * ((uint64_t)1L << 21);
|
||
carry14 = (s14 + (int64_t)(1L << 20)) >> 21;
|
||
s15 += carry14;
|
||
s14 -= carry14 * ((uint64_t)1L << 21);
|
||
carry16 = (s16 + (int64_t)(1L << 20)) >> 21;
|
||
s17 += carry16;
|
||
s16 -= carry16 * ((uint64_t)1L << 21);
|
||
carry18 = (s18 + (int64_t)(1L << 20)) >> 21;
|
||
s19 += carry18;
|
||
s18 -= carry18 * ((uint64_t)1L << 21);
|
||
carry20 = (s20 + (int64_t)(1L << 20)) >> 21;
|
||
s21 += carry20;
|
||
s20 -= carry20 * ((uint64_t)1L << 21);
|
||
carry22 = (s22 + (int64_t)(1L << 20)) >> 21;
|
||
s23 += carry22;
|
||
s22 -= carry22 * ((uint64_t)1L << 21);
|
||
|
||
carry1 = (s1 + (int64_t)(1L << 20)) >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry3 = (s3 + (int64_t)(1L << 20)) >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry5 = (s5 + (int64_t)(1L << 20)) >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
carry13 = (s13 + (int64_t)(1L << 20)) >> 21;
|
||
s14 += carry13;
|
||
s13 -= carry13 * ((uint64_t)1L << 21);
|
||
carry15 = (s15 + (int64_t)(1L << 20)) >> 21;
|
||
s16 += carry15;
|
||
s15 -= carry15 * ((uint64_t)1L << 21);
|
||
carry17 = (s17 + (int64_t)(1L << 20)) >> 21;
|
||
s18 += carry17;
|
||
s17 -= carry17 * ((uint64_t)1L << 21);
|
||
carry19 = (s19 + (int64_t)(1L << 20)) >> 21;
|
||
s20 += carry19;
|
||
s19 -= carry19 * ((uint64_t)1L << 21);
|
||
carry21 = (s21 + (int64_t)(1L << 20)) >> 21;
|
||
s22 += carry21;
|
||
s21 -= carry21 * ((uint64_t)1L << 21);
|
||
|
||
s11 += s23 * 666643;
|
||
s12 += s23 * 470296;
|
||
s13 += s23 * 654183;
|
||
s14 -= s23 * 997805;
|
||
s15 += s23 * 136657;
|
||
s16 -= s23 * 683901;
|
||
|
||
s10 += s22 * 666643;
|
||
s11 += s22 * 470296;
|
||
s12 += s22 * 654183;
|
||
s13 -= s22 * 997805;
|
||
s14 += s22 * 136657;
|
||
s15 -= s22 * 683901;
|
||
|
||
s9 += s21 * 666643;
|
||
s10 += s21 * 470296;
|
||
s11 += s21 * 654183;
|
||
s12 -= s21 * 997805;
|
||
s13 += s21 * 136657;
|
||
s14 -= s21 * 683901;
|
||
|
||
s8 += s20 * 666643;
|
||
s9 += s20 * 470296;
|
||
s10 += s20 * 654183;
|
||
s11 -= s20 * 997805;
|
||
s12 += s20 * 136657;
|
||
s13 -= s20 * 683901;
|
||
|
||
s7 += s19 * 666643;
|
||
s8 += s19 * 470296;
|
||
s9 += s19 * 654183;
|
||
s10 -= s19 * 997805;
|
||
s11 += s19 * 136657;
|
||
s12 -= s19 * 683901;
|
||
|
||
s6 += s18 * 666643;
|
||
s7 += s18 * 470296;
|
||
s8 += s18 * 654183;
|
||
s9 -= s18 * 997805;
|
||
s10 += s18 * 136657;
|
||
s11 -= s18 * 683901;
|
||
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry12 = (s12 + (int64_t)(1L << 20)) >> 21;
|
||
s13 += carry12;
|
||
s12 -= carry12 * ((uint64_t)1L << 21);
|
||
carry14 = (s14 + (int64_t)(1L << 20)) >> 21;
|
||
s15 += carry14;
|
||
s14 -= carry14 * ((uint64_t)1L << 21);
|
||
carry16 = (s16 + (int64_t)(1L << 20)) >> 21;
|
||
s17 += carry16;
|
||
s16 -= carry16 * ((uint64_t)1L << 21);
|
||
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
carry13 = (s13 + (int64_t)(1L << 20)) >> 21;
|
||
s14 += carry13;
|
||
s13 -= carry13 * ((uint64_t)1L << 21);
|
||
carry15 = (s15 + (int64_t)(1L << 20)) >> 21;
|
||
s16 += carry15;
|
||
s15 -= carry15 * ((uint64_t)1L << 21);
|
||
|
||
s5 += s17 * 666643;
|
||
s6 += s17 * 470296;
|
||
s7 += s17 * 654183;
|
||
s8 -= s17 * 997805;
|
||
s9 += s17 * 136657;
|
||
s10 -= s17 * 683901;
|
||
|
||
s4 += s16 * 666643;
|
||
s5 += s16 * 470296;
|
||
s6 += s16 * 654183;
|
||
s7 -= s16 * 997805;
|
||
s8 += s16 * 136657;
|
||
s9 -= s16 * 683901;
|
||
|
||
s3 += s15 * 666643;
|
||
s4 += s15 * 470296;
|
||
s5 += s15 * 654183;
|
||
s6 -= s15 * 997805;
|
||
s7 += s15 * 136657;
|
||
s8 -= s15 * 683901;
|
||
|
||
s2 += s14 * 666643;
|
||
s3 += s14 * 470296;
|
||
s4 += s14 * 654183;
|
||
s5 -= s14 * 997805;
|
||
s6 += s14 * 136657;
|
||
s7 -= s14 * 683901;
|
||
|
||
s1 += s13 * 666643;
|
||
s2 += s13 * 470296;
|
||
s3 += s13 * 654183;
|
||
s4 -= s13 * 997805;
|
||
s5 += s13 * 136657;
|
||
s6 -= s13 * 683901;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = (s0 + (int64_t)(1L << 20)) >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry2 = (s2 + (int64_t)(1L << 20)) >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry4 = (s4 + (int64_t)(1L << 20)) >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
|
||
carry1 = (s1 + (int64_t)(1L << 20)) >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry3 = (s3 + (int64_t)(1L << 20)) >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry5 = (s5 + (int64_t)(1L << 20)) >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry1 = s1 >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry2 = s2 >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry3 = s3 >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry4 = s4 >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry5 = s5 >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry6 = s6 >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry7 = s7 >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry8 = s8 >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry9 = s9 >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry10 = s10 >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry11 = s11 >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry1 = s1 >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry2 = s2 >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry3 = s3 >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry4 = s4 >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry5 = s5 >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry6 = s6 >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry7 = s7 >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry8 = s8 >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry9 = s9 >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry10 = s10 >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 * ((uint64_t)1 << 5));
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 * ((uint64_t)1 << 2));
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 * ((uint64_t)1 << 7));
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 * ((uint64_t)1 << 4));
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 * ((uint64_t)1 << 1));
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 * ((uint64_t)1 << 6));
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 * ((uint64_t)1 << 3));
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 * ((uint64_t)1 << 5));
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 * ((uint64_t)1 << 2));
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 * ((uint64_t)1 << 7));
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
/* libsodium/crypto_core/ed25519/ref10/ed25519_ref10.c */
|
||
/*
|
||
Input:
|
||
a[0]+256*a[1]+...+256^31*a[31] = a
|
||
b[0]+256*b[1]+...+256^31*b[31] = b
|
||
c[0]+256*c[1]+...+256^31*c[31] = c
|
||
*
|
||
Output:
|
||
s[0]+256*s[1]+...+256^31*s[31] = (ab+c) mod l
|
||
where l = 2^252 + 27742317777372353535851937790883648493.
|
||
*/
|
||
void sc_muladd(unsigned char* s, const unsigned char* a,
|
||
const unsigned char* b, const unsigned char* c)
|
||
{
|
||
int64_t a0 = 2097151 & load_3(a);
|
||
int64_t a1 = 2097151 & (load_4(a + 2) >> 5);
|
||
int64_t a2 = 2097151 & (load_3(a + 5) >> 2);
|
||
int64_t a3 = 2097151 & (load_4(a + 7) >> 7);
|
||
int64_t a4 = 2097151 & (load_4(a + 10) >> 4);
|
||
int64_t a5 = 2097151 & (load_3(a + 13) >> 1);
|
||
int64_t a6 = 2097151 & (load_4(a + 15) >> 6);
|
||
int64_t a7 = 2097151 & (load_3(a + 18) >> 3);
|
||
int64_t a8 = 2097151 & load_3(a + 21);
|
||
int64_t a9 = 2097151 & (load_4(a + 23) >> 5);
|
||
int64_t a10 = 2097151 & (load_3(a + 26) >> 2);
|
||
int64_t a11 = (load_4(a + 28) >> 7);
|
||
|
||
int64_t b0 = 2097151 & load_3(b);
|
||
int64_t b1 = 2097151 & (load_4(b + 2) >> 5);
|
||
int64_t b2 = 2097151 & (load_3(b + 5) >> 2);
|
||
int64_t b3 = 2097151 & (load_4(b + 7) >> 7);
|
||
int64_t b4 = 2097151 & (load_4(b + 10) >> 4);
|
||
int64_t b5 = 2097151 & (load_3(b + 13) >> 1);
|
||
int64_t b6 = 2097151 & (load_4(b + 15) >> 6);
|
||
int64_t b7 = 2097151 & (load_3(b + 18) >> 3);
|
||
int64_t b8 = 2097151 & load_3(b + 21);
|
||
int64_t b9 = 2097151 & (load_4(b + 23) >> 5);
|
||
int64_t b10 = 2097151 & (load_3(b + 26) >> 2);
|
||
int64_t b11 = (load_4(b + 28) >> 7);
|
||
|
||
int64_t c0 = 2097151 & load_3(c);
|
||
int64_t c1 = 2097151 & (load_4(c + 2) >> 5);
|
||
int64_t c2 = 2097151 & (load_3(c + 5) >> 2);
|
||
int64_t c3 = 2097151 & (load_4(c + 7) >> 7);
|
||
int64_t c4 = 2097151 & (load_4(c + 10) >> 4);
|
||
int64_t c5 = 2097151 & (load_3(c + 13) >> 1);
|
||
int64_t c6 = 2097151 & (load_4(c + 15) >> 6);
|
||
int64_t c7 = 2097151 & (load_3(c + 18) >> 3);
|
||
int64_t c8 = 2097151 & load_3(c + 21);
|
||
int64_t c9 = 2097151 & (load_4(c + 23) >> 5);
|
||
int64_t c10 = 2097151 & (load_3(c + 26) >> 2);
|
||
int64_t c11 = (load_4(c + 28) >> 7);
|
||
|
||
int64_t s0;
|
||
int64_t s1;
|
||
int64_t s2;
|
||
int64_t s3;
|
||
int64_t s4;
|
||
int64_t s5;
|
||
int64_t s6;
|
||
int64_t s7;
|
||
int64_t s8;
|
||
int64_t s9;
|
||
int64_t s10;
|
||
int64_t s11;
|
||
int64_t s12;
|
||
int64_t s13;
|
||
int64_t s14;
|
||
int64_t s15;
|
||
int64_t s16;
|
||
int64_t s17;
|
||
int64_t s18;
|
||
int64_t s19;
|
||
int64_t s20;
|
||
int64_t s21;
|
||
int64_t s22;
|
||
int64_t s23;
|
||
|
||
int64_t carry0;
|
||
int64_t carry1;
|
||
int64_t carry2;
|
||
int64_t carry3;
|
||
int64_t carry4;
|
||
int64_t carry5;
|
||
int64_t carry6;
|
||
int64_t carry7;
|
||
int64_t carry8;
|
||
int64_t carry9;
|
||
int64_t carry10;
|
||
int64_t carry11;
|
||
int64_t carry12;
|
||
int64_t carry13;
|
||
int64_t carry14;
|
||
int64_t carry15;
|
||
int64_t carry16;
|
||
int64_t carry17;
|
||
int64_t carry18;
|
||
int64_t carry19;
|
||
int64_t carry20;
|
||
int64_t carry21;
|
||
int64_t carry22;
|
||
|
||
s0 = c0 + a0 * b0;
|
||
s1 = c1 + a0 * b1 + a1 * b0;
|
||
s2 = c2 + a0 * b2 + a1 * b1 + a2 * b0;
|
||
s3 = c3 + a0 * b3 + a1 * b2 + a2 * b1 + a3 * b0;
|
||
s4 = c4 + a0 * b4 + a1 * b3 + a2 * b2 + a3 * b1 + a4 * b0;
|
||
s5 = c5 + a0 * b5 + a1 * b4 + a2 * b3 + a3 * b2 + a4 * b1 + a5 * b0;
|
||
s6 = c6 + a0 * b6 + a1 * b5 + a2 * b4 + a3 * b3 + a4 * b2 + a5 * b1 +
|
||
a6 * b0;
|
||
s7 = c7 + a0 * b7 + a1 * b6 + a2 * b5 + a3 * b4 + a4 * b3 + a5 * b2 +
|
||
a6 * b1 + a7 * b0;
|
||
s8 = c8 + a0 * b8 + a1 * b7 + a2 * b6 + a3 * b5 + a4 * b4 + a5 * b3 +
|
||
a6 * b2 + a7 * b1 + a8 * b0;
|
||
s9 = c9 + a0 * b9 + a1 * b8 + a2 * b7 + a3 * b6 + a4 * b5 + a5 * b4 +
|
||
a6 * b3 + a7 * b2 + a8 * b1 + a9 * b0;
|
||
s10 = c10 + a0 * b10 + a1 * b9 + a2 * b8 + a3 * b7 + a4 * b6 + a5 * b5 +
|
||
a6 * b4 + a7 * b3 + a8 * b2 + a9 * b1 + a10 * b0;
|
||
s11 = c11 + a0 * b11 + a1 * b10 + a2 * b9 + a3 * b8 + a4 * b7 + a5 * b6 +
|
||
a6 * b5 + a7 * b4 + a8 * b3 + a9 * b2 + a10 * b1 + a11 * b0;
|
||
s12 = a1 * b11 + a2 * b10 + a3 * b9 + a4 * b8 + a5 * b7 + a6 * b6 +
|
||
a7 * b5 + a8 * b4 + a9 * b3 + a10 * b2 + a11 * b1;
|
||
s13 = a2 * b11 + a3 * b10 + a4 * b9 + a5 * b8 + a6 * b7 + a7 * b6 +
|
||
a8 * b5 + a9 * b4 + a10 * b3 + a11 * b2;
|
||
s14 = a3 * b11 + a4 * b10 + a5 * b9 + a6 * b8 + a7 * b7 + a8 * b6 +
|
||
a9 * b5 + a10 * b4 + a11 * b3;
|
||
s15 = a4 * b11 + a5 * b10 + a6 * b9 + a7 * b8 + a8 * b7 + a9 * b6 +
|
||
a10 * b5 + a11 * b4;
|
||
s16 =
|
||
a5 * b11 + a6 * b10 + a7 * b9 + a8 * b8 + a9 * b7 + a10 * b6 + a11 * b5;
|
||
s17 = a6 * b11 + a7 * b10 + a8 * b9 + a9 * b8 + a10 * b7 + a11 * b6;
|
||
s18 = a7 * b11 + a8 * b10 + a9 * b9 + a10 * b8 + a11 * b7;
|
||
s19 = a8 * b11 + a9 * b10 + a10 * b9 + a11 * b8;
|
||
s20 = a9 * b11 + a10 * b10 + a11 * b9;
|
||
s21 = a10 * b11 + a11 * b10;
|
||
s22 = a11 * b11;
|
||
s23 = 0;
|
||
|
||
carry0 = (s0 + (int64_t)(1L << 20)) >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry2 = (s2 + (int64_t)(1L << 20)) >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry4 = (s4 + (int64_t)(1L << 20)) >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry12 = (s12 + (int64_t)(1L << 20)) >> 21;
|
||
s13 += carry12;
|
||
s12 -= carry12 * ((uint64_t)1L << 21);
|
||
carry14 = (s14 + (int64_t)(1L << 20)) >> 21;
|
||
s15 += carry14;
|
||
s14 -= carry14 * ((uint64_t)1L << 21);
|
||
carry16 = (s16 + (int64_t)(1L << 20)) >> 21;
|
||
s17 += carry16;
|
||
s16 -= carry16 * ((uint64_t)1L << 21);
|
||
carry18 = (s18 + (int64_t)(1L << 20)) >> 21;
|
||
s19 += carry18;
|
||
s18 -= carry18 * ((uint64_t)1L << 21);
|
||
carry20 = (s20 + (int64_t)(1L << 20)) >> 21;
|
||
s21 += carry20;
|
||
s20 -= carry20 * ((uint64_t)1L << 21);
|
||
carry22 = (s22 + (int64_t)(1L << 20)) >> 21;
|
||
s23 += carry22;
|
||
s22 -= carry22 * ((uint64_t)1L << 21);
|
||
|
||
carry1 = (s1 + (int64_t)(1L << 20)) >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry3 = (s3 + (int64_t)(1L << 20)) >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry5 = (s5 + (int64_t)(1L << 20)) >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
carry13 = (s13 + (int64_t)(1L << 20)) >> 21;
|
||
s14 += carry13;
|
||
s13 -= carry13 * ((uint64_t)1L << 21);
|
||
carry15 = (s15 + (int64_t)(1L << 20)) >> 21;
|
||
s16 += carry15;
|
||
s15 -= carry15 * ((uint64_t)1L << 21);
|
||
carry17 = (s17 + (int64_t)(1L << 20)) >> 21;
|
||
s18 += carry17;
|
||
s17 -= carry17 * ((uint64_t)1L << 21);
|
||
carry19 = (s19 + (int64_t)(1L << 20)) >> 21;
|
||
s20 += carry19;
|
||
s19 -= carry19 * ((uint64_t)1L << 21);
|
||
carry21 = (s21 + (int64_t)(1L << 20)) >> 21;
|
||
s22 += carry21;
|
||
s21 -= carry21 * ((uint64_t)1L << 21);
|
||
|
||
s11 += s23 * 666643;
|
||
s12 += s23 * 470296;
|
||
s13 += s23 * 654183;
|
||
s14 -= s23 * 997805;
|
||
s15 += s23 * 136657;
|
||
s16 -= s23 * 683901;
|
||
|
||
s10 += s22 * 666643;
|
||
s11 += s22 * 470296;
|
||
s12 += s22 * 654183;
|
||
s13 -= s22 * 997805;
|
||
s14 += s22 * 136657;
|
||
s15 -= s22 * 683901;
|
||
|
||
s9 += s21 * 666643;
|
||
s10 += s21 * 470296;
|
||
s11 += s21 * 654183;
|
||
s12 -= s21 * 997805;
|
||
s13 += s21 * 136657;
|
||
s14 -= s21 * 683901;
|
||
|
||
s8 += s20 * 666643;
|
||
s9 += s20 * 470296;
|
||
s10 += s20 * 654183;
|
||
s11 -= s20 * 997805;
|
||
s12 += s20 * 136657;
|
||
s13 -= s20 * 683901;
|
||
|
||
s7 += s19 * 666643;
|
||
s8 += s19 * 470296;
|
||
s9 += s19 * 654183;
|
||
s10 -= s19 * 997805;
|
||
s11 += s19 * 136657;
|
||
s12 -= s19 * 683901;
|
||
|
||
s6 += s18 * 666643;
|
||
s7 += s18 * 470296;
|
||
s8 += s18 * 654183;
|
||
s9 -= s18 * 997805;
|
||
s10 += s18 * 136657;
|
||
s11 -= s18 * 683901;
|
||
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry12 = (s12 + (int64_t)(1L << 20)) >> 21;
|
||
s13 += carry12;
|
||
s12 -= carry12 * ((uint64_t)1L << 21);
|
||
carry14 = (s14 + (int64_t)(1L << 20)) >> 21;
|
||
s15 += carry14;
|
||
s14 -= carry14 * ((uint64_t)1L << 21);
|
||
carry16 = (s16 + (int64_t)(1L << 20)) >> 21;
|
||
s17 += carry16;
|
||
s16 -= carry16 * ((uint64_t)1L << 21);
|
||
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
carry13 = (s13 + (int64_t)(1L << 20)) >> 21;
|
||
s14 += carry13;
|
||
s13 -= carry13 * ((uint64_t)1L << 21);
|
||
carry15 = (s15 + (int64_t)(1L << 20)) >> 21;
|
||
s16 += carry15;
|
||
s15 -= carry15 * ((uint64_t)1L << 21);
|
||
|
||
s5 += s17 * 666643;
|
||
s6 += s17 * 470296;
|
||
s7 += s17 * 654183;
|
||
s8 -= s17 * 997805;
|
||
s9 += s17 * 136657;
|
||
s10 -= s17 * 683901;
|
||
|
||
s4 += s16 * 666643;
|
||
s5 += s16 * 470296;
|
||
s6 += s16 * 654183;
|
||
s7 -= s16 * 997805;
|
||
s8 += s16 * 136657;
|
||
s9 -= s16 * 683901;
|
||
|
||
s3 += s15 * 666643;
|
||
s4 += s15 * 470296;
|
||
s5 += s15 * 654183;
|
||
s6 -= s15 * 997805;
|
||
s7 += s15 * 136657;
|
||
s8 -= s15 * 683901;
|
||
|
||
s2 += s14 * 666643;
|
||
s3 += s14 * 470296;
|
||
s4 += s14 * 654183;
|
||
s5 -= s14 * 997805;
|
||
s6 += s14 * 136657;
|
||
s7 -= s14 * 683901;
|
||
|
||
s1 += s13 * 666643;
|
||
s2 += s13 * 470296;
|
||
s3 += s13 * 654183;
|
||
s4 -= s13 * 997805;
|
||
s5 += s13 * 136657;
|
||
s6 -= s13 * 683901;
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = (s0 + (int64_t)(1L << 20)) >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry2 = (s2 + (int64_t)(1L << 20)) >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry4 = (s4 + (int64_t)(1L << 20)) >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry6 = (s6 + (int64_t)(1L << 20)) >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry8 = (s8 + (int64_t)(1L << 20)) >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry10 = (s10 + (int64_t)(1L << 20)) >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
|
||
carry1 = (s1 + (int64_t)(1L << 20)) >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry3 = (s3 + (int64_t)(1L << 20)) >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry5 = (s5 + (int64_t)(1L << 20)) >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry7 = (s7 + (int64_t)(1L << 20)) >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry9 = (s9 + (int64_t)(1L << 20)) >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry11 = (s11 + (int64_t)(1L << 20)) >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
s12 = 0;
|
||
|
||
carry0 = s0 >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry1 = s1 >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry2 = s2 >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry3 = s3 >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry4 = s4 >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry5 = s5 >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry6 = s6 >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry7 = s7 >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry8 = s8 >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry9 = s9 >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry10 = s10 >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
carry11 = s11 >> 21;
|
||
s12 += carry11;
|
||
s11 -= carry11 * ((uint64_t)1L << 21);
|
||
|
||
s0 += s12 * 666643;
|
||
s1 += s12 * 470296;
|
||
s2 += s12 * 654183;
|
||
s3 -= s12 * 997805;
|
||
s4 += s12 * 136657;
|
||
s5 -= s12 * 683901;
|
||
|
||
carry0 = s0 >> 21;
|
||
s1 += carry0;
|
||
s0 -= carry0 * ((uint64_t)1L << 21);
|
||
carry1 = s1 >> 21;
|
||
s2 += carry1;
|
||
s1 -= carry1 * ((uint64_t)1L << 21);
|
||
carry2 = s2 >> 21;
|
||
s3 += carry2;
|
||
s2 -= carry2 * ((uint64_t)1L << 21);
|
||
carry3 = s3 >> 21;
|
||
s4 += carry3;
|
||
s3 -= carry3 * ((uint64_t)1L << 21);
|
||
carry4 = s4 >> 21;
|
||
s5 += carry4;
|
||
s4 -= carry4 * ((uint64_t)1L << 21);
|
||
carry5 = s5 >> 21;
|
||
s6 += carry5;
|
||
s5 -= carry5 * ((uint64_t)1L << 21);
|
||
carry6 = s6 >> 21;
|
||
s7 += carry6;
|
||
s6 -= carry6 * ((uint64_t)1L << 21);
|
||
carry7 = s7 >> 21;
|
||
s8 += carry7;
|
||
s7 -= carry7 * ((uint64_t)1L << 21);
|
||
carry8 = s8 >> 21;
|
||
s9 += carry8;
|
||
s8 -= carry8 * ((uint64_t)1L << 21);
|
||
carry9 = s9 >> 21;
|
||
s10 += carry9;
|
||
s9 -= carry9 * ((uint64_t)1L << 21);
|
||
carry10 = s10 >> 21;
|
||
s11 += carry10;
|
||
s10 -= carry10 * ((uint64_t)1L << 21);
|
||
|
||
s[0] = s0 >> 0;
|
||
s[1] = s0 >> 8;
|
||
s[2] = (s0 >> 16) | (s1 * ((uint64_t)1 << 5));
|
||
s[3] = s1 >> 3;
|
||
s[4] = s1 >> 11;
|
||
s[5] = (s1 >> 19) | (s2 * ((uint64_t)1 << 2));
|
||
s[6] = s2 >> 6;
|
||
s[7] = (s2 >> 14) | (s3 * ((uint64_t)1 << 7));
|
||
s[8] = s3 >> 1;
|
||
s[9] = s3 >> 9;
|
||
s[10] = (s3 >> 17) | (s4 * ((uint64_t)1 << 4));
|
||
s[11] = s4 >> 4;
|
||
s[12] = s4 >> 12;
|
||
s[13] = (s4 >> 20) | (s5 * ((uint64_t)1 << 1));
|
||
s[14] = s5 >> 7;
|
||
s[15] = (s5 >> 15) | (s6 * ((uint64_t)1 << 6));
|
||
s[16] = s6 >> 2;
|
||
s[17] = s6 >> 10;
|
||
s[18] = (s6 >> 18) | (s7 * ((uint64_t)1 << 3));
|
||
s[19] = s7 >> 5;
|
||
s[20] = s7 >> 13;
|
||
s[21] = s8 >> 0;
|
||
s[22] = s8 >> 8;
|
||
s[23] = (s8 >> 16) | (s9 * ((uint64_t)1 << 5));
|
||
s[24] = s9 >> 3;
|
||
s[25] = s9 >> 11;
|
||
s[26] = (s9 >> 19) | (s10 * ((uint64_t)1 << 2));
|
||
s[27] = s10 >> 6;
|
||
s[28] = (s10 >> 14) | (s11 * ((uint64_t)1 << 7));
|
||
s[29] = s11 >> 1;
|
||
s[30] = s11 >> 9;
|
||
s[31] = s11 >> 17;
|
||
}
|
||
|
||
/* libsodium/crypto_core/ed25519/ref10/ed25519_ref10.c */
|
||
void sc_sqmul(unsigned char s[32], const int n, const unsigned char a[32])
|
||
{
|
||
int i;
|
||
for (i = 0; i < n; ++i)
|
||
sc_mul(s, s, s);
|
||
sc_mul(s, s, a);
|
||
}
|
||
|
||
/* libsodium/crypto_core/ed25519/ref10/ed25519_ref10.c */
|
||
void sc_invert(unsigned char* recip, const unsigned char* s)
|
||
{
|
||
unsigned char _10[32], _100[32], _1000[32], _10000[32], _100000[32],
|
||
_1000000[32], _10010011[32], _10010111[32], _100110[32], _1010[32],
|
||
_1010000[32], _1010011[32], _1011[32], _10110[32], _10111101[32],
|
||
_11[32], _1100011[32], _1100111[32], _11010011[32], _1101011[32],
|
||
_11100111[32], _11101011[32], _11110101[32];
|
||
|
||
sc_mul(_10, s, s);
|
||
sc_mul(_11, s, _10);
|
||
sc_mul(_100, s, _11);
|
||
sc_mul(_1000, _100, _100);
|
||
sc_mul(_1010, _10, _1000);
|
||
sc_mul(_1011, s, _1010);
|
||
sc_mul(_10000, _1000, _1000);
|
||
sc_mul(_10110, _1011, _1011);
|
||
sc_mul(_100000, _1010, _10110);
|
||
sc_mul(_100110, _10000, _10110);
|
||
sc_mul(_1000000, _100000, _100000);
|
||
sc_mul(_1010000, _10000, _1000000);
|
||
sc_mul(_1010011, _11, _1010000);
|
||
sc_mul(_1100011, _10000, _1010011);
|
||
sc_mul(_1100111, _100, _1100011);
|
||
sc_mul(_1101011, _100, _1100111);
|
||
sc_mul(_10010011, _1000000, _1010011);
|
||
sc_mul(_10010111, _100, _10010011);
|
||
sc_mul(_10111101, _100110, _10010111);
|
||
sc_mul(_11010011, _10110, _10111101);
|
||
sc_mul(_11100111, _1010000, _10010111);
|
||
sc_mul(_11101011, _100, _11100111);
|
||
sc_mul(_11110101, _1010, _11101011);
|
||
|
||
sc_mul(recip, _1011, _11110101);
|
||
|
||
sc_sqmul(recip, 126, _1010011);
|
||
|
||
sc_sqmul(recip, 9, _10);
|
||
sc_mul(recip, recip, _11110101);
|
||
sc_sqmul(recip, 7, _1100111);
|
||
sc_sqmul(recip, 9, _11110101);
|
||
sc_sqmul(recip, 11, _10111101);
|
||
sc_sqmul(recip, 8, _11100111);
|
||
sc_sqmul(recip, 9, _1101011);
|
||
sc_sqmul(recip, 6, _1011);
|
||
sc_sqmul(recip, 14, _10010011);
|
||
sc_sqmul(recip, 10, _1100011);
|
||
sc_sqmul(recip, 9, _10010111);
|
||
sc_sqmul(recip, 10, _11110101);
|
||
sc_sqmul(recip, 8, _11010011);
|
||
sc_sqmul(recip, 8, _11101011);
|
||
}
|
||
|
||
/*
|
||
In: t (x, y, z)
|
||
Out: r (x, t, z, t)
|
||
|
||
Note: expensive conversion because of fe_invert
|
||
*/
|
||
void ge_p2_to_p3(ge_p3 *r, const ge_p2 *t)
|
||
{
|
||
fe_copy(r->X, t->X);
|
||
fe_copy(r->Y, t->Y);
|
||
fe_copy(r->Z, t->Z);
|
||
fe_invert(r->T, t->Z);
|
||
fe_mul(r->T, r->T, t->Y);
|
||
fe_mul(r->T, r->T, t->X);
|
||
}
|
||
|
||
|
||
/*
|
||
In: data -- points to 'size' bytes of data
|
||
Out: res = Hp(data)
|
||
where Hp = 8 * ge_fromfe_frombytes_vartime(cn_fast_hash(data))
|
||
*/
|
||
void ge_bytes_hash_to_ec(ge_p3 *res, const void *data, size_t size)
|
||
{
|
||
unsigned char h[HASH_SIZE];
|
||
ge_p2 point;
|
||
ge_p1p1 point2;
|
||
|
||
cn_fast_hash(data, size, (char*)h);
|
||
ge_fromfe_frombytes_vartime(&point, &h[0]);
|
||
/*ge_p2_to_p3(res, &point); -- can be used to avoid multiplication by 8 for debugging */
|
||
ge_mul8(&point2, &point);
|
||
ge_p1p1_to_p3(res, &point2);
|
||
}
|
||
|
||
void ge_bytes_hash_to_ec_32(ge_p3 *res, const unsigned char *ge_bytes)
|
||
{
|
||
ge_bytes_hash_to_ec(res, ge_bytes, 32);
|
||
}
|
||
|
||
void ge_mul8_p3(ge_p3 *r, const ge_p3 *t)
|
||
{
|
||
ge_p1p1 p1;
|
||
ge_p2 p2;
|
||
|
||
// TODO: consider removing the following copy, replace &p2 by (ge_p2*)t as it's ugly but possible
|
||
ge_p3_to_p2(&p2, t); // copying
|
||
|
||
ge_p2_dbl(&p1, &p2); // 3 fe_sq, 1 fe_sq2, 5 fe_add/sub
|
||
ge_p1p1_to_p2(&p2, &p1); // 3 fe_mul
|
||
ge_p2_dbl(&p1, &p2); // 3 fe_sq, 1 fe_sq2, 5 fe_add/sub
|
||
ge_p1p1_to_p2(&p2, &p1); // 3 fe_mul
|
||
ge_p2_dbl(&p1, &p2); // 3 fe_sq, 1 fe_sq2, 5 fe_add/sub
|
||
|
||
ge_p1p1_to_p3(r, &p1); // 4 fe_mul
|
||
}
|
||
|
||
|
||
// returns the most non-zero index of r
|
||
int slide_v2(signed char *r, const unsigned char *a)
|
||
{
|
||
int i;
|
||
int b;
|
||
int k;
|
||
int nzi = 0;
|
||
|
||
for (i = 0; i < 256; ++i) {
|
||
r[i] = 1 & (a[i >> 3] >> (i & 7));
|
||
}
|
||
|
||
for (i = 0; i < 256; ++i) {
|
||
if (r[i]) {
|
||
for (b = 1; b <= 6 && i + b < 256; ++b) {
|
||
if (r[i + b]) {
|
||
if (r[i] + (r[i + b] << b) <= 15) {
|
||
r[i] += r[i + b] << b; r[i + b] = 0;
|
||
}
|
||
else if (r[i] - (r[i + b] << b) >= -15) {
|
||
r[i] -= r[i + b] << b;
|
||
for (k = i + b; k < 256; ++k) {
|
||
if (!r[k]) {
|
||
r[k] = 1;
|
||
break;
|
||
}
|
||
r[k] = 0;
|
||
}
|
||
}
|
||
else
|
||
break;
|
||
}
|
||
}
|
||
if (r[i])
|
||
nzi = i;
|
||
}
|
||
}
|
||
|
||
return nzi;
|
||
}
|
||
|
||
void ge_scalarmult_vartime_p3_v2(ge_p3 *r, const unsigned char *a, const ge_p3 *A)
|
||
{
|
||
signed char aslide[256];
|
||
ge_dsmp Ai; /* A, 3A, 5A, 7A, 9A, 11A, 13A, 15A */
|
||
ge_p1p1 t;
|
||
ge_p3 u;
|
||
ge_p2 r_p2;
|
||
int i;
|
||
|
||
i = slide_v2(aslide, a);
|
||
|
||
if (i == 0)
|
||
{
|
||
ge_p3_0(r);
|
||
return;
|
||
}
|
||
|
||
ge_dsm_precomp(Ai, A);
|
||
ge_p2_0(&r_p2);
|
||
|
||
for (; i >= 0; --i)
|
||
{
|
||
ge_p2_dbl(&t, &r_p2);
|
||
if (aslide[i] > 0)
|
||
{
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_add(&t, &u, &Ai[aslide[i] / 2]);
|
||
}
|
||
else if (aslide[i] < 0)
|
||
{
|
||
ge_p1p1_to_p3(&u, &t);
|
||
ge_sub(&t, &u, &Ai[(-aslide[i]) / 2]);
|
||
}
|
||
if (i != 0)
|
||
ge_p1p1_to_p2(&r_p2, &t);
|
||
else
|
||
ge_p1p1_to_p3(r, &t);
|
||
}
|
||
}
|
||
|
||
|
||
void ge_cached_to_p2(ge_p2 *r, const ge_cached *c)
|
||
{
|
||
static const fe inv2 = { 10, 0, 0, 0, 0, 0, 0, 0, 0, -16777216 };
|
||
|
||
fe_sub(r->X, c->YplusX, c->YminusX);
|
||
fe_mul(r->X, r->X, inv2);
|
||
|
||
fe_add(r->Y, c->YplusX, c->YminusX);
|
||
fe_mul(r->Y, r->Y, inv2);
|
||
|
||
fe_copy(r->Z, c->Z);
|
||
}
|
||
|
||
|
||
///////////////////////////
|
||
// EXPERIMENTAL
|
||
//
|
||
|
||
// With these select_vartime/ge_scalarmult_base_vartime I got ~25% speed up comparing to the select/ge_scalarmult_base -- sowle
|
||
static void select_vartime(ge_precomp *t, int pos, signed char b)
|
||
{
|
||
unsigned char bnegative = negative(b);
|
||
unsigned char babs = b - (((-bnegative) & b) << 1);
|
||
const ge_precomp* base;
|
||
|
||
if (babs == 0)
|
||
{
|
||
ge_precomp_0(t);
|
||
}
|
||
else if (bnegative == 0)
|
||
{
|
||
base = &ge_base[pos][babs - 1];
|
||
fe_copy(t->yplusx, base->yplusx);
|
||
fe_copy(t->yminusx, base->yminusx);
|
||
fe_copy(t->xy2d, base->xy2d);
|
||
}
|
||
else
|
||
{
|
||
base = &ge_base[pos][babs - 1];
|
||
fe_copy(t->yplusx, base->yminusx);
|
||
fe_copy(t->yminusx, base->yplusx);
|
||
fe_neg(t->xy2d, base->xy2d);
|
||
}
|
||
}
|
||
|
||
void ge_scalarmult_base_vartime(ge_p3 *h, const unsigned char *a)
|
||
{
|
||
signed char e[64];
|
||
signed char carry;
|
||
ge_p1p1 r;
|
||
ge_p2 s;
|
||
ge_precomp t;
|
||
int i;
|
||
|
||
for (i = 0; i < 32; ++i) {
|
||
e[2 * i + 0] = (a[i] >> 0) & 15;
|
||
e[2 * i + 1] = (a[i] >> 4) & 15;
|
||
}
|
||
/* each e[i] is between 0 and 15 */
|
||
/* e[63] is between 0 and 7 */
|
||
|
||
carry = 0;
|
||
for (i = 0; i < 63; ++i) {
|
||
e[i] += carry;
|
||
carry = e[i] + 8;
|
||
carry >>= 4;
|
||
e[i] -= carry << 4;
|
||
}
|
||
e[63] += carry;
|
||
/* each e[i] is between -8 and 8 */
|
||
|
||
ge_p3_0(h);
|
||
for (i = 1; i < 64; i += 2) {
|
||
select_vartime(&t, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
|
||
ge_p3_dbl(&r, h); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p3(h, &r);
|
||
|
||
for (i = 0; i < 64; i += 2) {
|
||
select_vartime(&t, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
}
|
||
|
||
|
||
static void select_custom_precomp_vartime(ge_precomp *t, const precomp_data_t base_precomp, int pos, signed char b)
|
||
{
|
||
unsigned char bnegative = negative(b);
|
||
unsigned char babs = b - (((-bnegative) & b) << 1);
|
||
const ge_precomp* base;
|
||
|
||
if (babs == 0)
|
||
{
|
||
ge_precomp_0(t);
|
||
}
|
||
else if (bnegative == 0)
|
||
{
|
||
base = &base_precomp[pos][babs - 1];
|
||
fe_copy(t->yplusx, base->yplusx);
|
||
fe_copy(t->yminusx, base->yminusx);
|
||
fe_copy(t->xy2d, base->xy2d);
|
||
}
|
||
else
|
||
{
|
||
base = &base_precomp[pos][babs - 1];
|
||
fe_copy(t->yplusx, base->yminusx);
|
||
fe_copy(t->yminusx, base->yplusx);
|
||
fe_neg(t->xy2d, base->xy2d);
|
||
}
|
||
}
|
||
|
||
|
||
void ge_scalarmult_precomp_vartime(ge_p3 *h, const precomp_data_t base_precomp, const unsigned char *a)
|
||
{
|
||
signed char e[64];
|
||
signed char carry;
|
||
ge_p1p1 r;
|
||
ge_p2 s;
|
||
ge_precomp t;
|
||
int i;
|
||
|
||
for (i = 0; i < 32; ++i) {
|
||
e[2 * i + 0] = (a[i] >> 0) & 15;
|
||
e[2 * i + 1] = (a[i] >> 4) & 15;
|
||
}
|
||
/* each e[i] is between 0 and 15 */
|
||
/* e[63] is between 0 and 7 */
|
||
|
||
carry = 0;
|
||
for (i = 0; i < 63; ++i) {
|
||
e[i] += carry;
|
||
carry = e[i] + 8;
|
||
carry >>= 4;
|
||
e[i] -= carry << 4;
|
||
}
|
||
e[63] += carry;
|
||
/* each e[i] is between -8 and 8 */
|
||
|
||
ge_p3_0(h);
|
||
for (i = 1; i < 64; i += 2) {
|
||
select_custom_precomp_vartime(&t, base_precomp, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
|
||
ge_p3_dbl(&r, h); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p2(&s, &r);
|
||
ge_p2_dbl(&r, &s); ge_p1p1_to_p3(h, &r);
|
||
|
||
for (i = 0; i < 64; i += 2) {
|
||
select_custom_precomp_vartime(&t, base_precomp, i / 2, e[i]);
|
||
ge_madd(&r, h, &t); ge_p1p1_to_p3(h, &r);
|
||
}
|
||
}
|
||
|
||
void ge_p3_to_precomp(ge_precomp *r, const ge_p3* p)
|
||
{
|
||
fe recip;
|
||
fe x;
|
||
fe y;
|
||
//unsigned char s[32];
|
||
|
||
fe_invert(recip, p->Z);
|
||
fe_mul(x, p->X, recip);
|
||
fe_mul(y, p->Y, recip);
|
||
fe_sub(r->yminusx, y, x);
|
||
fe_add(r->yplusx, y, x);
|
||
fe_mul(r->xy2d, x, y);
|
||
fe_mul(r->xy2d, r->xy2d, fe_d2);
|
||
|
||
// to get canonical representation and obtain the very same beautiful numbers for ge_base in crypto-ops-data.c (maybe unnecessary, TODO -- sowle)
|
||
//fe_tobytes(s, r->yminusx); fe_frombytes(r->yminusx, s);
|
||
//fe_tobytes(s, r->yplusx); fe_frombytes(r->yplusx, s);
|
||
//fe_tobytes(s, r->xy2d); fe_frombytes(r->xy2d, s);
|
||
}
|
||
|