first commit

This commit is contained in:
Orson Peters 2013-01-10 21:14:17 +01:00
commit 3aad2934a3
50 changed files with 5780 additions and 0 deletions

BIN
a.exe Normal file

Binary file not shown.

12
crypto_sign.h Normal file
View File

@ -0,0 +1,12 @@
#ifndef crypto_sign_edwards25519sha512batch_H
#define crypto_sign_edwards25519sha512batch_H
#define SECRETKEYBYTES 64
#define PUBLICKEYBYTES 32
#define SIGNATUREBYTES 64
extern int crypto_sign(unsigned char *,unsigned long long *,const unsigned char *,unsigned long long,const unsigned char *);
extern int crypto_sign_open(unsigned char *,unsigned long long *,const unsigned char *,unsigned long long,const unsigned char *);
extern int crypto_sign_keypair(unsigned char *,unsigned char *,unsigned char *);
#endif

56
fe.h Normal file
View File

@ -0,0 +1,56 @@
#ifndef FE_H
#define FE_H
#include "pstdint.h"
typedef int32_t fe[10];
/*
fe means field element.
Here the field is \Z/(2^255-19).
An element t, entries t[0]...t[9], represents the integer
t[0]+2^26 t[1]+2^51 t[2]+2^77 t[3]+2^102 t[4]+...+2^230 t[9].
Bounds on each t[i] vary depending on context.
*/
#define fe_frombytes crypto_sign_ed25519_ref10_fe_frombytes
#define fe_tobytes crypto_sign_ed25519_ref10_fe_tobytes
#define fe_copy crypto_sign_ed25519_ref10_fe_copy
#define fe_isnonzero crypto_sign_ed25519_ref10_fe_isnonzero
#define fe_isnegative crypto_sign_ed25519_ref10_fe_isnegative
#define fe_0 crypto_sign_ed25519_ref10_fe_0
#define fe_1 crypto_sign_ed25519_ref10_fe_1
#define fe_cswap crypto_sign_ed25519_ref10_fe_cswap
#define fe_cmov crypto_sign_ed25519_ref10_fe_cmov
#define fe_add crypto_sign_ed25519_ref10_fe_add
#define fe_sub crypto_sign_ed25519_ref10_fe_sub
#define fe_neg crypto_sign_ed25519_ref10_fe_neg
#define fe_mul crypto_sign_ed25519_ref10_fe_mul
#define fe_sq crypto_sign_ed25519_ref10_fe_sq
#define fe_sq2 crypto_sign_ed25519_ref10_fe_sq2
#define fe_mul121666 crypto_sign_ed25519_ref10_fe_mul121666
#define fe_invert crypto_sign_ed25519_ref10_fe_invert
#define fe_pow22523 crypto_sign_ed25519_ref10_fe_pow22523
extern void fe_frombytes(fe,const unsigned char *);
extern void fe_tobytes(unsigned char *,const fe);
extern void fe_copy(fe,const fe);
extern int fe_isnonzero(const fe);
extern int fe_isnegative(const fe);
extern void fe_0(fe);
extern void fe_1(fe);
extern void fe_cswap(fe,fe,unsigned int);
extern void fe_cmov(fe,const fe,unsigned int);
extern void fe_add(fe,const fe,const fe);
extern void fe_sub(fe,const fe,const fe);
extern void fe_neg(fe,const fe);
extern void fe_mul(fe,const fe,const fe);
extern void fe_sq(fe,const fe);
extern void fe_sq2(fe,const fe);
extern void fe_mul121666(fe,const fe);
extern void fe_invert(fe,const fe);
extern void fe_pow22523(fe,const fe);
#endif

19
fe_0.c Normal file
View File

@ -0,0 +1,19 @@
#include "fe.h"
/*
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;
}

19
fe_1.c Normal file
View File

@ -0,0 +1,19 @@
#include "fe.h"
/*
h = 1
*/
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;
}

57
fe_add.c Normal file
View File

@ -0,0 +1,57 @@
#include "fe.h"
/*
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;
}

63
fe_cmov.c Normal file
View File

@ -0,0 +1,63 @@
#include "fe.h"
/*
Replace (f,g) with (g,g) if b == 1;
replace (f,g) with (f,g) if b == 0.
Preconditions: b in {0,1}.
*/
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;
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;
}

29
fe_copy.c Normal file
View File

@ -0,0 +1,29 @@
#include "fe.h"
/*
h = f
*/
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;
}

72
fe_frombytes.c Normal file
View File

@ -0,0 +1,72 @@
#include "fe.h"
#include "pstdint.h"
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;
}
/*
Ignores top bit of h.
*/
void fe_frombytes(fe h,const unsigned char *s)
{
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;
}

174
fe_invert.c Normal file
View File

@ -0,0 +1,174 @@
#include "fe.h"
void fe_invert(fe out,const fe z)
{
fe t0;
fe t1;
fe t2;
fe t3;
int i;
/* qhasm: fe z1 */
/* qhasm: fe z2 */
/* qhasm: fe z8 */
/* qhasm: fe z9 */
/* qhasm: fe z11 */
/* qhasm: fe z22 */
/* qhasm: fe z_5_0 */
/* qhasm: fe z_10_5 */
/* qhasm: fe z_10_0 */
/* qhasm: fe z_20_10 */
/* qhasm: fe z_20_0 */
/* qhasm: fe z_40_20 */
/* qhasm: fe z_40_0 */
/* qhasm: fe z_50_10 */
/* qhasm: fe z_50_0 */
/* qhasm: fe z_100_50 */
/* qhasm: fe z_100_0 */
/* qhasm: fe z_200_100 */
/* qhasm: fe z_200_0 */
/* qhasm: fe z_250_50 */
/* qhasm: fe z_250_0 */
/* qhasm: fe z_255_5 */
/* qhasm: fe z_255_21 */
/* qhasm: enter pow225521 */
/* qhasm: z2 = z1^2^1 */
/* asm 1: fe_sq(>z2=fe#1,<z1=fe#11); for (i = 1;i < 1;++i) fe_sq(>z2=fe#1,>z2=fe#1); */
/* asm 2: fe_sq(>z2=t0,<z1=z); for (i = 1;i < 1;++i) fe_sq(>z2=t0,>z2=t0); */
fe_sq(t0,z); for (i = 1;i < 1;++i) fe_sq(t0,t0);
/* qhasm: z8 = z2^2^2 */
/* asm 1: fe_sq(>z8=fe#2,<z2=fe#1); for (i = 1;i < 2;++i) fe_sq(>z8=fe#2,>z8=fe#2); */
/* asm 2: fe_sq(>z8=t1,<z2=t0); for (i = 1;i < 2;++i) fe_sq(>z8=t1,>z8=t1); */
fe_sq(t1,t0); for (i = 1;i < 2;++i) fe_sq(t1,t1);
/* qhasm: z9 = z1*z8 */
/* asm 1: fe_mul(>z9=fe#2,<z1=fe#11,<z8=fe#2); */
/* asm 2: fe_mul(>z9=t1,<z1=z,<z8=t1); */
fe_mul(t1,z,t1);
/* qhasm: z11 = z2*z9 */
/* asm 1: fe_mul(>z11=fe#1,<z2=fe#1,<z9=fe#2); */
/* asm 2: fe_mul(>z11=t0,<z2=t0,<z9=t1); */
fe_mul(t0,t0,t1);
/* qhasm: z22 = z11^2^1 */
/* asm 1: fe_sq(>z22=fe#3,<z11=fe#1); for (i = 1;i < 1;++i) fe_sq(>z22=fe#3,>z22=fe#3); */
/* asm 2: fe_sq(>z22=t2,<z11=t0); for (i = 1;i < 1;++i) fe_sq(>z22=t2,>z22=t2); */
fe_sq(t2,t0); for (i = 1;i < 1;++i) fe_sq(t2,t2);
/* qhasm: z_5_0 = z9*z22 */
/* asm 1: fe_mul(>z_5_0=fe#2,<z9=fe#2,<z22=fe#3); */
/* asm 2: fe_mul(>z_5_0=t1,<z9=t1,<z22=t2); */
fe_mul(t1,t1,t2);
/* qhasm: z_10_5 = z_5_0^2^5 */
/* asm 1: fe_sq(>z_10_5=fe#3,<z_5_0=fe#2); for (i = 1;i < 5;++i) fe_sq(>z_10_5=fe#3,>z_10_5=fe#3); */
/* asm 2: fe_sq(>z_10_5=t2,<z_5_0=t1); for (i = 1;i < 5;++i) fe_sq(>z_10_5=t2,>z_10_5=t2); */
fe_sq(t2,t1); for (i = 1;i < 5;++i) fe_sq(t2,t2);
/* qhasm: z_10_0 = z_10_5*z_5_0 */
/* asm 1: fe_mul(>z_10_0=fe#2,<z_10_5=fe#3,<z_5_0=fe#2); */
/* asm 2: fe_mul(>z_10_0=t1,<z_10_5=t2,<z_5_0=t1); */
fe_mul(t1,t2,t1);
/* qhasm: z_20_10 = z_10_0^2^10 */
/* asm 1: fe_sq(>z_20_10=fe#3,<z_10_0=fe#2); for (i = 1;i < 10;++i) fe_sq(>z_20_10=fe#3,>z_20_10=fe#3); */
/* asm 2: fe_sq(>z_20_10=t2,<z_10_0=t1); for (i = 1;i < 10;++i) fe_sq(>z_20_10=t2,>z_20_10=t2); */
fe_sq(t2,t1); for (i = 1;i < 10;++i) fe_sq(t2,t2);
/* qhasm: z_20_0 = z_20_10*z_10_0 */
/* asm 1: fe_mul(>z_20_0=fe#3,<z_20_10=fe#3,<z_10_0=fe#2); */
/* asm 2: fe_mul(>z_20_0=t2,<z_20_10=t2,<z_10_0=t1); */
fe_mul(t2,t2,t1);
/* qhasm: z_40_20 = z_20_0^2^20 */
/* asm 1: fe_sq(>z_40_20=fe#4,<z_20_0=fe#3); for (i = 1;i < 20;++i) fe_sq(>z_40_20=fe#4,>z_40_20=fe#4); */
/* asm 2: fe_sq(>z_40_20=t3,<z_20_0=t2); for (i = 1;i < 20;++i) fe_sq(>z_40_20=t3,>z_40_20=t3); */
fe_sq(t3,t2); for (i = 1;i < 20;++i) fe_sq(t3,t3);
/* qhasm: z_40_0 = z_40_20*z_20_0 */
/* asm 1: fe_mul(>z_40_0=fe#3,<z_40_20=fe#4,<z_20_0=fe#3); */
/* asm 2: fe_mul(>z_40_0=t2,<z_40_20=t3,<z_20_0=t2); */
fe_mul(t2,t3,t2);
/* qhasm: z_50_10 = z_40_0^2^10 */
/* asm 1: fe_sq(>z_50_10=fe#3,<z_40_0=fe#3); for (i = 1;i < 10;++i) fe_sq(>z_50_10=fe#3,>z_50_10=fe#3); */
/* asm 2: fe_sq(>z_50_10=t2,<z_40_0=t2); for (i = 1;i < 10;++i) fe_sq(>z_50_10=t2,>z_50_10=t2); */
fe_sq(t2,t2); for (i = 1;i < 10;++i) fe_sq(t2,t2);
/* qhasm: z_50_0 = z_50_10*z_10_0 */
/* asm 1: fe_mul(>z_50_0=fe#2,<z_50_10=fe#3,<z_10_0=fe#2); */
/* asm 2: fe_mul(>z_50_0=t1,<z_50_10=t2,<z_10_0=t1); */
fe_mul(t1,t2,t1);
/* qhasm: z_100_50 = z_50_0^2^50 */
/* asm 1: fe_sq(>z_100_50=fe#3,<z_50_0=fe#2); for (i = 1;i < 50;++i) fe_sq(>z_100_50=fe#3,>z_100_50=fe#3); */
/* asm 2: fe_sq(>z_100_50=t2,<z_50_0=t1); for (i = 1;i < 50;++i) fe_sq(>z_100_50=t2,>z_100_50=t2); */
fe_sq(t2,t1); for (i = 1;i < 50;++i) fe_sq(t2,t2);
/* qhasm: z_100_0 = z_100_50*z_50_0 */
/* asm 1: fe_mul(>z_100_0=fe#3,<z_100_50=fe#3,<z_50_0=fe#2); */
/* asm 2: fe_mul(>z_100_0=t2,<z_100_50=t2,<z_50_0=t1); */
fe_mul(t2,t2,t1);
/* qhasm: z_200_100 = z_100_0^2^100 */
/* asm 1: fe_sq(>z_200_100=fe#4,<z_100_0=fe#3); for (i = 1;i < 100;++i) fe_sq(>z_200_100=fe#4,>z_200_100=fe#4); */
/* asm 2: fe_sq(>z_200_100=t3,<z_100_0=t2); for (i = 1;i < 100;++i) fe_sq(>z_200_100=t3,>z_200_100=t3); */
fe_sq(t3,t2); for (i = 1;i < 100;++i) fe_sq(t3,t3);
/* qhasm: z_200_0 = z_200_100*z_100_0 */
/* asm 1: fe_mul(>z_200_0=fe#3,<z_200_100=fe#4,<z_100_0=fe#3); */
/* asm 2: fe_mul(>z_200_0=t2,<z_200_100=t3,<z_100_0=t2); */
fe_mul(t2,t3,t2);
/* qhasm: z_250_50 = z_200_0^2^50 */
/* asm 1: fe_sq(>z_250_50=fe#3,<z_200_0=fe#3); for (i = 1;i < 50;++i) fe_sq(>z_250_50=fe#3,>z_250_50=fe#3); */
/* asm 2: fe_sq(>z_250_50=t2,<z_200_0=t2); for (i = 1;i < 50;++i) fe_sq(>z_250_50=t2,>z_250_50=t2); */
fe_sq(t2,t2); for (i = 1;i < 50;++i) fe_sq(t2,t2);
/* qhasm: z_250_0 = z_250_50*z_50_0 */
/* asm 1: fe_mul(>z_250_0=fe#2,<z_250_50=fe#3,<z_50_0=fe#2); */
/* asm 2: fe_mul(>z_250_0=t1,<z_250_50=t2,<z_50_0=t1); */
fe_mul(t1,t2,t1);
/* qhasm: z_255_5 = z_250_0^2^5 */
/* asm 1: fe_sq(>z_255_5=fe#2,<z_250_0=fe#2); for (i = 1;i < 5;++i) fe_sq(>z_255_5=fe#2,>z_255_5=fe#2); */
/* asm 2: fe_sq(>z_255_5=t1,<z_250_0=t1); for (i = 1;i < 5;++i) fe_sq(>z_255_5=t1,>z_255_5=t1); */
fe_sq(t1,t1); for (i = 1;i < 5;++i) fe_sq(t1,t1);
/* qhasm: z_255_21 = z_255_5*z11 */
/* asm 1: fe_mul(>z_255_21=fe#12,<z_255_5=fe#2,<z11=fe#1); */
/* asm 2: fe_mul(>z_255_21=out,<z_255_5=t1,<z11=t0); */
fe_mul(out,t1,t0);
/* qhasm: return */
return;
}

16
fe_isnegative.c Normal file
View File

@ -0,0 +1,16 @@
#include "fe.h"
/*
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;
}

20
fe_isnonzero.c Normal file
View File

@ -0,0 +1,20 @@
#include "fe.h"
#include "verify.h"
/*
return 1 if f == 0
return 0 if f != 0
Preconditions:
|f| bounded by 1.1*2^26,1.1*2^25,1.1*2^26,1.1*2^25,etc.
*/
static const unsigned char zero[32];
int fe_isnonzero(const fe f)
{
unsigned char s[32];
fe_tobytes(s,f);
return crypto_verify_32(s,zero);
}

253
fe_mul.c Normal file
View File

@ -0,0 +1,253 @@
#include "fe.h"
#include "pstdint.h"
/*
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;
}

45
fe_neg.c Normal file
View File

@ -0,0 +1,45 @@
#include "fe.h"
/*
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.
*/
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;
}

173
fe_pow22523.c Normal file
View File

@ -0,0 +1,173 @@
#include "fe.h"
void fe_pow22523(fe out,const fe z)
{
fe t0;
fe t1;
fe t2;
int i;
/* qhasm: fe z1 */
/* qhasm: fe z2 */
/* qhasm: fe z8 */
/* qhasm: fe z9 */
/* qhasm: fe z11 */
/* qhasm: fe z22 */
/* qhasm: fe z_5_0 */
/* qhasm: fe z_10_5 */
/* qhasm: fe z_10_0 */
/* qhasm: fe z_20_10 */
/* qhasm: fe z_20_0 */
/* qhasm: fe z_40_20 */
/* qhasm: fe z_40_0 */
/* qhasm: fe z_50_10 */
/* qhasm: fe z_50_0 */
/* qhasm: fe z_100_50 */
/* qhasm: fe z_100_0 */
/* qhasm: fe z_200_100 */
/* qhasm: fe z_200_0 */
/* qhasm: fe z_250_50 */
/* qhasm: fe z_250_0 */
/* qhasm: fe z_252_2 */
/* qhasm: fe z_252_3 */
/* qhasm: enter pow22523 */
/* qhasm: z2 = z1^2^1 */
/* asm 1: fe_sq(>z2=fe#1,<z1=fe#11); for (i = 1;i < 1;++i) fe_sq(>z2=fe#1,>z2=fe#1); */
/* asm 2: fe_sq(>z2=t0,<z1=z); for (i = 1;i < 1;++i) fe_sq(>z2=t0,>z2=t0); */
fe_sq(t0,z); for (i = 1;i < 1;++i) fe_sq(t0,t0);
/* qhasm: z8 = z2^2^2 */
/* asm 1: fe_sq(>z8=fe#2,<z2=fe#1); for (i = 1;i < 2;++i) fe_sq(>z8=fe#2,>z8=fe#2); */
/* asm 2: fe_sq(>z8=t1,<z2=t0); for (i = 1;i < 2;++i) fe_sq(>z8=t1,>z8=t1); */
fe_sq(t1,t0); for (i = 1;i < 2;++i) fe_sq(t1,t1);
/* qhasm: z9 = z1*z8 */
/* asm 1: fe_mul(>z9=fe#2,<z1=fe#11,<z8=fe#2); */
/* asm 2: fe_mul(>z9=t1,<z1=z,<z8=t1); */
fe_mul(t1,z,t1);
/* qhasm: z11 = z2*z9 */
/* asm 1: fe_mul(>z11=fe#1,<z2=fe#1,<z9=fe#2); */
/* asm 2: fe_mul(>z11=t0,<z2=t0,<z9=t1); */
fe_mul(t0,t0,t1);
/* qhasm: z22 = z11^2^1 */
/* asm 1: fe_sq(>z22=fe#1,<z11=fe#1); for (i = 1;i < 1;++i) fe_sq(>z22=fe#1,>z22=fe#1); */
/* asm 2: fe_sq(>z22=t0,<z11=t0); for (i = 1;i < 1;++i) fe_sq(>z22=t0,>z22=t0); */
fe_sq(t0,t0); for (i = 1;i < 1;++i) fe_sq(t0,t0);
/* qhasm: z_5_0 = z9*z22 */
/* asm 1: fe_mul(>z_5_0=fe#1,<z9=fe#2,<z22=fe#1); */
/* asm 2: fe_mul(>z_5_0=t0,<z9=t1,<z22=t0); */
fe_mul(t0,t1,t0);
/* qhasm: z_10_5 = z_5_0^2^5 */
/* asm 1: fe_sq(>z_10_5=fe#2,<z_5_0=fe#1); for (i = 1;i < 5;++i) fe_sq(>z_10_5=fe#2,>z_10_5=fe#2); */
/* asm 2: fe_sq(>z_10_5=t1,<z_5_0=t0); for (i = 1;i < 5;++i) fe_sq(>z_10_5=t1,>z_10_5=t1); */
fe_sq(t1,t0); for (i = 1;i < 5;++i) fe_sq(t1,t1);
/* qhasm: z_10_0 = z_10_5*z_5_0 */
/* asm 1: fe_mul(>z_10_0=fe#1,<z_10_5=fe#2,<z_5_0=fe#1); */
/* asm 2: fe_mul(>z_10_0=t0,<z_10_5=t1,<z_5_0=t0); */
fe_mul(t0,t1,t0);
/* qhasm: z_20_10 = z_10_0^2^10 */
/* asm 1: fe_sq(>z_20_10=fe#2,<z_10_0=fe#1); for (i = 1;i < 10;++i) fe_sq(>z_20_10=fe#2,>z_20_10=fe#2); */
/* asm 2: fe_sq(>z_20_10=t1,<z_10_0=t0); for (i = 1;i < 10;++i) fe_sq(>z_20_10=t1,>z_20_10=t1); */
fe_sq(t1,t0); for (i = 1;i < 10;++i) fe_sq(t1,t1);
/* qhasm: z_20_0 = z_20_10*z_10_0 */
/* asm 1: fe_mul(>z_20_0=fe#2,<z_20_10=fe#2,<z_10_0=fe#1); */
/* asm 2: fe_mul(>z_20_0=t1,<z_20_10=t1,<z_10_0=t0); */
fe_mul(t1,t1,t0);
/* qhasm: z_40_20 = z_20_0^2^20 */
/* asm 1: fe_sq(>z_40_20=fe#3,<z_20_0=fe#2); for (i = 1;i < 20;++i) fe_sq(>z_40_20=fe#3,>z_40_20=fe#3); */
/* asm 2: fe_sq(>z_40_20=t2,<z_20_0=t1); for (i = 1;i < 20;++i) fe_sq(>z_40_20=t2,>z_40_20=t2); */
fe_sq(t2,t1); for (i = 1;i < 20;++i) fe_sq(t2,t2);
/* qhasm: z_40_0 = z_40_20*z_20_0 */
/* asm 1: fe_mul(>z_40_0=fe#2,<z_40_20=fe#3,<z_20_0=fe#2); */
/* asm 2: fe_mul(>z_40_0=t1,<z_40_20=t2,<z_20_0=t1); */
fe_mul(t1,t2,t1);
/* qhasm: z_50_10 = z_40_0^2^10 */
/* asm 1: fe_sq(>z_50_10=fe#2,<z_40_0=fe#2); for (i = 1;i < 10;++i) fe_sq(>z_50_10=fe#2,>z_50_10=fe#2); */
/* asm 2: fe_sq(>z_50_10=t1,<z_40_0=t1); for (i = 1;i < 10;++i) fe_sq(>z_50_10=t1,>z_50_10=t1); */
fe_sq(t1,t1); for (i = 1;i < 10;++i) fe_sq(t1,t1);
/* qhasm: z_50_0 = z_50_10*z_10_0 */
/* asm 1: fe_mul(>z_50_0=fe#1,<z_50_10=fe#2,<z_10_0=fe#1); */
/* asm 2: fe_mul(>z_50_0=t0,<z_50_10=t1,<z_10_0=t0); */
fe_mul(t0,t1,t0);
/* qhasm: z_100_50 = z_50_0^2^50 */
/* asm 1: fe_sq(>z_100_50=fe#2,<z_50_0=fe#1); for (i = 1;i < 50;++i) fe_sq(>z_100_50=fe#2,>z_100_50=fe#2); */
/* asm 2: fe_sq(>z_100_50=t1,<z_50_0=t0); for (i = 1;i < 50;++i) fe_sq(>z_100_50=t1,>z_100_50=t1); */
fe_sq(t1,t0); for (i = 1;i < 50;++i) fe_sq(t1,t1);
/* qhasm: z_100_0 = z_100_50*z_50_0 */
/* asm 1: fe_mul(>z_100_0=fe#2,<z_100_50=fe#2,<z_50_0=fe#1); */
/* asm 2: fe_mul(>z_100_0=t1,<z_100_50=t1,<z_50_0=t0); */
fe_mul(t1,t1,t0);
/* qhasm: z_200_100 = z_100_0^2^100 */
/* asm 1: fe_sq(>z_200_100=fe#3,<z_100_0=fe#2); for (i = 1;i < 100;++i) fe_sq(>z_200_100=fe#3,>z_200_100=fe#3); */
/* asm 2: fe_sq(>z_200_100=t2,<z_100_0=t1); for (i = 1;i < 100;++i) fe_sq(>z_200_100=t2,>z_200_100=t2); */
fe_sq(t2,t1); for (i = 1;i < 100;++i) fe_sq(t2,t2);
/* qhasm: z_200_0 = z_200_100*z_100_0 */
/* asm 1: fe_mul(>z_200_0=fe#2,<z_200_100=fe#3,<z_100_0=fe#2); */
/* asm 2: fe_mul(>z_200_0=t1,<z_200_100=t2,<z_100_0=t1); */
fe_mul(t1,t2,t1);
/* qhasm: z_250_50 = z_200_0^2^50 */
/* asm 1: fe_sq(>z_250_50=fe#2,<z_200_0=fe#2); for (i = 1;i < 50;++i) fe_sq(>z_250_50=fe#2,>z_250_50=fe#2); */
/* asm 2: fe_sq(>z_250_50=t1,<z_200_0=t1); for (i = 1;i < 50;++i) fe_sq(>z_250_50=t1,>z_250_50=t1); */
fe_sq(t1,t1); for (i = 1;i < 50;++i) fe_sq(t1,t1);
/* qhasm: z_250_0 = z_250_50*z_50_0 */
/* asm 1: fe_mul(>z_250_0=fe#1,<z_250_50=fe#2,<z_50_0=fe#1); */
/* asm 2: fe_mul(>z_250_0=t0,<z_250_50=t1,<z_50_0=t0); */
fe_mul(t0,t1,t0);
/* qhasm: z_252_2 = z_250_0^2^2 */
/* asm 1: fe_sq(>z_252_2=fe#1,<z_250_0=fe#1); for (i = 1;i < 2;++i) fe_sq(>z_252_2=fe#1,>z_252_2=fe#1); */
/* asm 2: fe_sq(>z_252_2=t0,<z_250_0=t0); for (i = 1;i < 2;++i) fe_sq(>z_252_2=t0,>z_252_2=t0); */
fe_sq(t0,t0); for (i = 1;i < 2;++i) fe_sq(t0,t0);
/* qhasm: z_252_3 = z_252_2*z1 */
/* asm 1: fe_mul(>z_252_3=fe#12,<z_252_2=fe#1,<z1=fe#11); */
/* asm 2: fe_mul(>z_252_3=out,<z_252_2=t0,<z1=z); */
fe_mul(out,t0,z);
/* qhasm: return */
return;
}

149
fe_sq.c Normal file
View File

@ -0,0 +1,149 @@
#include "fe.h"
#include "pstdint.h"
/*
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;
}

160
fe_sq2.c Normal file
View File

@ -0,0 +1,160 @@
#include "fe.h"
#include "pstdint.h"
/*
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.
*/
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;
}

57
fe_sub.c Normal file
View File

@ -0,0 +1,57 @@
#include "fe.h"
/*
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;
}

119
fe_tobytes.c Normal file
View File

@ -0,0 +1,119 @@
#include "fe.h"
/*
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;
}

95
ge.h Normal file
View File

@ -0,0 +1,95 @@
#ifndef GE_H
#define GE_H
/*
ge means group element.
Here the group is the set of pairs (x,y) of field elements (see fe.h)
satisfying -x^2 + y^2 = 1 + d x^2y^2
where d = -121665/121666.
Representations:
ge_p2 (projective): (X:Y:Z) satisfying x=X/Z, y=Y/Z
ge_p3 (extended): (X:Y:Z:T) satisfying x=X/Z, y=Y/Z, XY=ZT
ge_p1p1 (completed): ((X:Z),(Y:T)) satisfying x=X/Z, y=Y/T
ge_precomp (Duif): (y+x,y-x,2dxy)
*/
#include "fe.h"
typedef struct {
fe X;
fe Y;
fe Z;
} ge_p2;
typedef struct {
fe X;
fe Y;
fe Z;
fe T;
} ge_p3;
typedef struct {
fe X;
fe Y;
fe Z;
fe T;
} ge_p1p1;
typedef struct {
fe yplusx;
fe yminusx;
fe xy2d;
} ge_precomp;
typedef struct {
fe YplusX;
fe YminusX;
fe Z;
fe T2d;
} ge_cached;
#define ge_frombytes_negate_vartime crypto_sign_ed25519_ref10_ge_frombytes_negate_vartime
#define ge_tobytes crypto_sign_ed25519_ref10_ge_tobytes
#define ge_p3_tobytes crypto_sign_ed25519_ref10_ge_p3_tobytes
#define ge_p2_0 crypto_sign_ed25519_ref10_ge_p2_0
#define ge_p3_0 crypto_sign_ed25519_ref10_ge_p3_0
#define ge_precomp_0 crypto_sign_ed25519_ref10_ge_precomp_0
#define ge_p3_to_p2 crypto_sign_ed25519_ref10_ge_p3_to_p2
#define ge_p3_to_cached crypto_sign_ed25519_ref10_ge_p3_to_cached
#define ge_p1p1_to_p2 crypto_sign_ed25519_ref10_ge_p1p1_to_p2
#define ge_p1p1_to_p3 crypto_sign_ed25519_ref10_ge_p1p1_to_p3
#define ge_p2_dbl crypto_sign_ed25519_ref10_ge_p2_dbl
#define ge_p3_dbl crypto_sign_ed25519_ref10_ge_p3_dbl
#define ge_madd crypto_sign_ed25519_ref10_ge_madd
#define ge_msub crypto_sign_ed25519_ref10_ge_msub
#define ge_add crypto_sign_ed25519_ref10_ge_add
#define ge_sub crypto_sign_ed25519_ref10_ge_sub
#define ge_scalarmult_base crypto_sign_ed25519_ref10_ge_scalarmult_base
#define ge_double_scalarmult_vartime crypto_sign_ed25519_ref10_ge_double_scalarmult_vartime
extern void ge_tobytes(unsigned char *,const ge_p2 *);
extern void ge_p3_tobytes(unsigned char *,const ge_p3 *);
extern int ge_frombytes_negate_vartime(ge_p3 *,const unsigned char *);
extern void ge_p2_0(ge_p2 *);
extern void ge_p3_0(ge_p3 *);
extern void ge_precomp_0(ge_precomp *);
extern void ge_p3_to_p2(ge_p2 *,const ge_p3 *);
extern void ge_p3_to_cached(ge_cached *,const ge_p3 *);
extern void ge_p1p1_to_p2(ge_p2 *,const ge_p1p1 *);
extern void ge_p1p1_to_p3(ge_p3 *,const ge_p1p1 *);
extern void ge_p2_dbl(ge_p1p1 *,const ge_p2 *);
extern void ge_p3_dbl(ge_p1p1 *,const ge_p3 *);
extern void ge_madd(ge_p1p1 *,const ge_p3 *,const ge_precomp *);
extern void ge_msub(ge_p1p1 *,const ge_p3 *,const ge_precomp *);
extern void ge_add(ge_p1p1 *,const ge_p3 *,const ge_cached *);
extern void ge_sub(ge_p1p1 *,const ge_p3 *,const ge_cached *);
extern void ge_scalarmult_base(ge_p3 *,const unsigned char *);
extern void ge_double_scalarmult_vartime(ge_p2 *,const unsigned char *,const ge_p3 *,const unsigned char *);
#endif

108
ge_add.c Normal file
View File

@ -0,0 +1,108 @@
#include "ge.h"
/*
r = p + q
*/
void ge_add(ge_p1p1 *r,const ge_p3 *p,const ge_cached *q)
{
fe t0;
/* qhasm: enter ge_add */
/* qhasm: fe X1 */
/* qhasm: fe Y1 */
/* qhasm: fe Z1 */
/* qhasm: fe Z2 */
/* qhasm: fe T1 */
/* qhasm: fe ZZ */
/* qhasm: fe YpX2 */
/* qhasm: fe YmX2 */
/* qhasm: fe T2d2 */
/* qhasm: fe X3 */
/* qhasm: fe Y3 */
/* qhasm: fe Z3 */
/* qhasm: fe T3 */
/* qhasm: fe YpX1 */
/* qhasm: fe YmX1 */
/* qhasm: fe A */
/* qhasm: fe B */
/* qhasm: fe C */
/* qhasm: fe D */
/* qhasm: YpX1 = Y1+X1 */
/* asm 1: fe_add(>YpX1=fe#1,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_add(>YpX1=r->X,<Y1=p->Y,<X1=p->X); */
fe_add(r->X,p->Y,p->X);
/* qhasm: YmX1 = Y1-X1 */
/* asm 1: fe_sub(>YmX1=fe#2,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_sub(>YmX1=r->Y,<Y1=p->Y,<X1=p->X); */
fe_sub(r->Y,p->Y,p->X);
/* qhasm: A = YpX1*YpX2 */
/* asm 1: fe_mul(>A=fe#3,<YpX1=fe#1,<YpX2=fe#15); */
/* asm 2: fe_mul(>A=r->Z,<YpX1=r->X,<YpX2=q->YplusX); */
fe_mul(r->Z,r->X,q->YplusX);
/* qhasm: B = YmX1*YmX2 */
/* asm 1: fe_mul(>B=fe#2,<YmX1=fe#2,<YmX2=fe#16); */
/* asm 2: fe_mul(>B=r->Y,<YmX1=r->Y,<YmX2=q->YminusX); */
fe_mul(r->Y,r->Y,q->YminusX);
/* qhasm: C = T2d2*T1 */
/* asm 1: fe_mul(>C=fe#4,<T2d2=fe#18,<T1=fe#14); */
/* asm 2: fe_mul(>C=r->T,<T2d2=q->T2d,<T1=p->T); */
fe_mul(r->T,q->T2d,p->T);
/* qhasm: ZZ = Z1*Z2 */
/* asm 1: fe_mul(>ZZ=fe#1,<Z1=fe#13,<Z2=fe#17); */
/* asm 2: fe_mul(>ZZ=r->X,<Z1=p->Z,<Z2=q->Z); */
fe_mul(r->X,p->Z,q->Z);
/* qhasm: D = 2*ZZ */
/* asm 1: fe_add(>D=fe#5,<ZZ=fe#1,<ZZ=fe#1); */
/* asm 2: fe_add(>D=t0,<ZZ=r->X,<ZZ=r->X); */
fe_add(t0,r->X,r->X);
/* qhasm: X3 = A-B */
/* asm 1: fe_sub(>X3=fe#1,<A=fe#3,<B=fe#2); */
/* asm 2: fe_sub(>X3=r->X,<A=r->Z,<B=r->Y); */
fe_sub(r->X,r->Z,r->Y);
/* qhasm: Y3 = A+B */
/* asm 1: fe_add(>Y3=fe#2,<A=fe#3,<B=fe#2); */
/* asm 2: fe_add(>Y3=r->Y,<A=r->Z,<B=r->Y); */
fe_add(r->Y,r->Z,r->Y);
/* qhasm: Z3 = D+C */
/* asm 1: fe_add(>Z3=fe#3,<D=fe#5,<C=fe#4); */
/* asm 2: fe_add(>Z3=r->Z,<D=t0,<C=r->T); */
fe_add(r->Z,t0,r->T);
/* qhasm: T3 = D-C */
/* asm 1: fe_sub(>T3=fe#4,<D=fe#5,<C=fe#4); */
/* asm 2: fe_sub(>T3=r->T,<D=t0,<C=r->T); */
fe_sub(r->T,t0,r->T);
/* qhasm: return */
}

135
ge_double_scalarmult.c Normal file
View File

@ -0,0 +1,135 @@
#include "ge.h"
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;
}
}
}
}
static ge_precomp Bi[8] = {
{
{ 25967493,-14356035,29566456,3660896,-12694345,4014787,27544626,-11754271,-6079156,2047605 },
{ -12545711,934262,-2722910,3049990,-727428,9406986,12720692,5043384,19500929,-15469378 },
{ -8738181,4489570,9688441,-14785194,10184609,-12363380,29287919,11864899,-24514362,-4438546 },
},
{
{ 15636291,-9688557,24204773,-7912398,616977,-16685262,27787600,-14772189,28944400,-1550024 },
{ 16568933,4717097,-11556148,-1102322,15682896,-11807043,16354577,-11775962,7689662,11199574 },
{ 30464156,-5976125,-11779434,-15670865,23220365,15915852,7512774,10017326,-17749093,-9920357 },
},
{
{ 10861363,11473154,27284546,1981175,-30064349,12577861,32867885,14515107,-15438304,10819380 },
{ 4708026,6336745,20377586,9066809,-11272109,6594696,-25653668,12483688,-12668491,5581306 },
{ 19563160,16186464,-29386857,4097519,10237984,-4348115,28542350,13850243,-23678021,-15815942 },
},
{
{ 5153746,9909285,1723747,-2777874,30523605,5516873,19480852,5230134,-23952439,-15175766 },
{ -30269007,-3463509,7665486,10083793,28475525,1649722,20654025,16520125,30598449,7715701 },
{ 28881845,14381568,9657904,3680757,-20181635,7843316,-31400660,1370708,29794553,-1409300 },
},
{
{ -22518993,-6692182,14201702,-8745502,-23510406,8844726,18474211,-1361450,-13062696,13821877 },
{ -6455177,-7839871,3374702,-4740862,-27098617,-10571707,31655028,-7212327,18853322,-14220951 },
{ 4566830,-12963868,-28974889,-12240689,-7602672,-2830569,-8514358,-10431137,2207753,-3209784 },
},
{
{ -25154831,-4185821,29681144,7868801,-6854661,-9423865,-12437364,-663000,-31111463,-16132436 },
{ 25576264,-2703214,7349804,-11814844,16472782,9300885,3844789,15725684,171356,6466918 },
{ 23103977,13316479,9739013,-16149481,817875,-15038942,8965339,-14088058,-30714912,16193877 },
},
{
{ -33521811,3180713,-2394130,14003687,-16903474,-16270840,17238398,4729455,-18074513,9256800 },
{ -25182317,-4174131,32336398,5036987,-21236817,11360617,22616405,9761698,-19827198,630305 },
{ -13720693,2639453,-24237460,-7406481,9494427,-5774029,-6554551,-15960994,-2449256,-14291300 },
},
{
{ -3151181,-5046075,9282714,6866145,-31907062,-863023,-18940575,15033784,25105118,-7894876 },
{ -24326370,15950226,-31801215,-14592823,-11662737,-5090925,1573892,-2625887,2198790,-15804619 },
{ -3099351,10324967,-2241613,7453183,-5446979,-2735503,-13812022,-16236442,-32461234,-12290683 },
},
} ;
/*
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_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_cached Ai[8]; /* A,3A,5A,7A,9A,11A,13A,15A */
ge_p1p1 t;
ge_p3 u;
ge_p3 A2;
int i;
slide(aslide,a);
slide(bslide,b);
ge_p3_to_cached(&Ai[0],A);
ge_p3_dbl(&t,A); ge_p1p1_to_p3(&A2,&t);
ge_add(&t,&A2,&Ai[0]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[1],&u);
ge_add(&t,&A2,&Ai[1]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[2],&u);
ge_add(&t,&A2,&Ai[2]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[3],&u);
ge_add(&t,&A2,&Ai[3]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[4],&u);
ge_add(&t,&A2,&Ai[4]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[5],&u);
ge_add(&t,&A2,&Ai[5]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[6],&u);
ge_add(&t,&A2,&Ai[6]); ge_p1p1_to_p3(&u,&t); ge_p3_to_cached(&Ai[7],&u);
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,&Bi[bslide[i]/2]);
} else if (bslide[i] < 0) {
ge_p1p1_to_p3(&u,&t);
ge_msub(&t,&u,&Bi[(-bslide[i])/2]);
}
ge_p1p1_to_p2(r,&t);
}
}

52
ge_frombytes.c Normal file
View File

@ -0,0 +1,52 @@
#include "ge.h"
static const fe d = {
-10913610,13857413,-15372611,6949391,114729,-8787816,-6275908,-3247719,-18696448,-12055116
} ;
static const fe sqrtm1 = {
-32595792,-7943725,9377950,3500415,12389472,-272473,-25146209,-2005654,326686,11406482
} ;
int ge_frombytes_negate_vartime(ge_p3 *h,const unsigned char *s)
{
fe u;
fe v;
fe v3;
fe vxx;
fe check;
fe_frombytes(h->Y,s);
fe_1(h->Z);
fe_sq(u,h->Y);
fe_mul(v,u,d);
fe_sub(u,u,h->Z); /* u = y^2-1 */
fe_add(v,v,h->Z); /* v = dy^2+1 */
fe_sq(v3,v);
fe_mul(v3,v3,v); /* v3 = v^3 */
fe_sq(h->X,v3);
fe_mul(h->X,h->X,v);
fe_mul(h->X,h->X,u); /* x = uv^7 */
fe_pow22523(h->X,h->X); /* x = (uv^7)^((q-5)/8) */
fe_mul(h->X,h->X,v3);
fe_mul(h->X,h->X,u); /* 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,sqrtm1);
}
if (fe_isnegative(h->X) == (s[31] >> 7))
fe_neg(h->X,h->X);
fe_mul(h->T,h->X,h->Y);
return 0;
}

99
ge_madd.c Normal file
View File

@ -0,0 +1,99 @@
#include "ge.h"
/*
r = p + q
*/
void ge_madd(ge_p1p1 *r,const ge_p3 *p,const ge_precomp *q)
{
fe t0;
/* qhasm: enter ge_madd */
/* qhasm: fe X1 */
/* qhasm: fe Y1 */
/* qhasm: fe Z1 */
/* qhasm: fe T1 */
/* qhasm: fe ypx2 */
/* qhasm: fe ymx2 */
/* qhasm: fe xy2d2 */
/* qhasm: fe X3 */
/* qhasm: fe Y3 */
/* qhasm: fe Z3 */
/* qhasm: fe T3 */
/* qhasm: fe YpX1 */
/* qhasm: fe YmX1 */
/* qhasm: fe A */
/* qhasm: fe B */
/* qhasm: fe C */
/* qhasm: fe D */
/* qhasm: YpX1 = Y1+X1 */
/* asm 1: fe_add(>YpX1=fe#1,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_add(>YpX1=r->X,<Y1=p->Y,<X1=p->X); */
fe_add(r->X,p->Y,p->X);
/* qhasm: YmX1 = Y1-X1 */
/* asm 1: fe_sub(>YmX1=fe#2,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_sub(>YmX1=r->Y,<Y1=p->Y,<X1=p->X); */
fe_sub(r->Y,p->Y,p->X);
/* qhasm: A = YpX1*ypx2 */
/* asm 1: fe_mul(>A=fe#3,<YpX1=fe#1,<ypx2=fe#15); */
/* asm 2: fe_mul(>A=r->Z,<YpX1=r->X,<ypx2=q->yplusx); */
fe_mul(r->Z,r->X,q->yplusx);
/* qhasm: B = YmX1*ymx2 */
/* asm 1: fe_mul(>B=fe#2,<YmX1=fe#2,<ymx2=fe#16); */
/* asm 2: fe_mul(>B=r->Y,<YmX1=r->Y,<ymx2=q->yminusx); */
fe_mul(r->Y,r->Y,q->yminusx);
/* qhasm: C = xy2d2*T1 */
/* asm 1: fe_mul(>C=fe#4,<xy2d2=fe#17,<T1=fe#14); */
/* asm 2: fe_mul(>C=r->T,<xy2d2=q->xy2d,<T1=p->T); */
fe_mul(r->T,q->xy2d,p->T);
/* qhasm: D = 2*Z1 */
/* asm 1: fe_add(>D=fe#5,<Z1=fe#13,<Z1=fe#13); */
/* asm 2: fe_add(>D=t0,<Z1=p->Z,<Z1=p->Z); */
fe_add(t0,p->Z,p->Z);
/* qhasm: X3 = A-B */
/* asm 1: fe_sub(>X3=fe#1,<A=fe#3,<B=fe#2); */
/* asm 2: fe_sub(>X3=r->X,<A=r->Z,<B=r->Y); */
fe_sub(r->X,r->Z,r->Y);
/* qhasm: Y3 = A+B */
/* asm 1: fe_add(>Y3=fe#2,<A=fe#3,<B=fe#2); */
/* asm 2: fe_add(>Y3=r->Y,<A=r->Z,<B=r->Y); */
fe_add(r->Y,r->Z,r->Y);
/* qhasm: Z3 = D+C */
/* asm 1: fe_add(>Z3=fe#3,<D=fe#5,<C=fe#4); */
/* asm 2: fe_add(>Z3=r->Z,<D=t0,<C=r->T); */
fe_add(r->Z,t0,r->T);
/* qhasm: T3 = D-C */
/* asm 1: fe_sub(>T3=fe#4,<D=fe#5,<C=fe#4); */
/* asm 2: fe_sub(>T3=r->T,<D=t0,<C=r->T); */
fe_sub(r->T,t0,r->T);
/* qhasm: return */
}

99
ge_msub.c Normal file
View File

@ -0,0 +1,99 @@
#include "ge.h"
/*
r = p - q
*/
void ge_msub(ge_p1p1 *r,const ge_p3 *p,const ge_precomp *q)
{
fe t0;
/* qhasm: enter ge_msub */
/* qhasm: fe X1 */
/* qhasm: fe Y1 */
/* qhasm: fe Z1 */
/* qhasm: fe T1 */
/* qhasm: fe ypx2 */
/* qhasm: fe ymx2 */
/* qhasm: fe xy2d2 */
/* qhasm: fe X3 */
/* qhasm: fe Y3 */
/* qhasm: fe Z3 */
/* qhasm: fe T3 */
/* qhasm: fe YpX1 */
/* qhasm: fe YmX1 */
/* qhasm: fe A */
/* qhasm: fe B */
/* qhasm: fe C */
/* qhasm: fe D */
/* qhasm: YpX1 = Y1+X1 */
/* asm 1: fe_add(>YpX1=fe#1,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_add(>YpX1=r->X,<Y1=p->Y,<X1=p->X); */
fe_add(r->X,p->Y,p->X);
/* qhasm: YmX1 = Y1-X1 */
/* asm 1: fe_sub(>YmX1=fe#2,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_sub(>YmX1=r->Y,<Y1=p->Y,<X1=p->X); */
fe_sub(r->Y,p->Y,p->X);
/* qhasm: A = YpX1*ymx2 */
/* asm 1: fe_mul(>A=fe#3,<YpX1=fe#1,<ymx2=fe#16); */
/* asm 2: fe_mul(>A=r->Z,<YpX1=r->X,<ymx2=q->yminusx); */
fe_mul(r->Z,r->X,q->yminusx);
/* qhasm: B = YmX1*ypx2 */
/* asm 1: fe_mul(>B=fe#2,<YmX1=fe#2,<ypx2=fe#15); */
/* asm 2: fe_mul(>B=r->Y,<YmX1=r->Y,<ypx2=q->yplusx); */
fe_mul(r->Y,r->Y,q->yplusx);
/* qhasm: C = xy2d2*T1 */
/* asm 1: fe_mul(>C=fe#4,<xy2d2=fe#17,<T1=fe#14); */
/* asm 2: fe_mul(>C=r->T,<xy2d2=q->xy2d,<T1=p->T); */
fe_mul(r->T,q->xy2d,p->T);
/* qhasm: D = 2*Z1 */
/* asm 1: fe_add(>D=fe#5,<Z1=fe#13,<Z1=fe#13); */
/* asm 2: fe_add(>D=t0,<Z1=p->Z,<Z1=p->Z); */
fe_add(t0,p->Z,p->Z);
/* qhasm: X3 = A-B */
/* asm 1: fe_sub(>X3=fe#1,<A=fe#3,<B=fe#2); */
/* asm 2: fe_sub(>X3=r->X,<A=r->Z,<B=r->Y); */
fe_sub(r->X,r->Z,r->Y);
/* qhasm: Y3 = A+B */
/* asm 1: fe_add(>Y3=fe#2,<A=fe#3,<B=fe#2); */
/* asm 2: fe_add(>Y3=r->Y,<A=r->Z,<B=r->Y); */
fe_add(r->Y,r->Z,r->Y);
/* qhasm: Z3 = D-C */
/* asm 1: fe_sub(>Z3=fe#3,<D=fe#5,<C=fe#4); */
/* asm 2: fe_sub(>Z3=r->Z,<D=t0,<C=r->T); */
fe_sub(r->Z,t0,r->T);
/* qhasm: T3 = D+C */
/* asm 1: fe_add(>T3=fe#4,<D=fe#5,<C=fe#4); */
/* asm 2: fe_add(>T3=r->T,<D=t0,<C=r->T); */
fe_add(r->T,t0,r->T);
/* qhasm: return */
}

12
ge_p1p1_to_p2.c Normal file
View File

@ -0,0 +1,12 @@
#include "ge.h"
/*
r = p
*/
extern 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);
}

13
ge_p1p1_to_p3.c Normal file
View File

@ -0,0 +1,13 @@
#include "ge.h"
/*
r = p
*/
extern 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);
}

8
ge_p2_0.c Normal file
View File

@ -0,0 +1,8 @@
#include "ge.h"
void ge_p2_0(ge_p2 *h)
{
fe_0(h->X);
fe_1(h->Y);
fe_1(h->Z);
}

84
ge_p2_dbl.c Normal file
View File

@ -0,0 +1,84 @@
#include "ge.h"
/*
r = 2 * p
*/
void ge_p2_dbl(ge_p1p1 *r,const ge_p2 *p)
{
fe t0;
/* qhasm: enter ge_p2_dbl */
/* qhasm: fe X1 */
/* qhasm: fe Y1 */
/* qhasm: fe Z1 */
/* qhasm: fe A */
/* qhasm: fe AA */
/* qhasm: fe XX */
/* qhasm: fe YY */
/* qhasm: fe B */
/* qhasm: fe X3 */
/* qhasm: fe Y3 */
/* qhasm: fe Z3 */
/* qhasm: fe T3 */
/* qhasm: XX=X1^2 */
/* asm 1: fe_sq(>XX=fe#1,<X1=fe#11); */
/* asm 2: fe_sq(>XX=r->X,<X1=p->X); */
fe_sq(r->X,p->X);
/* qhasm: YY=Y1^2 */
/* asm 1: fe_sq(>YY=fe#3,<Y1=fe#12); */
/* asm 2: fe_sq(>YY=r->Z,<Y1=p->Y); */
fe_sq(r->Z,p->Y);
/* qhasm: B=2*Z1^2 */
/* asm 1: fe_sq2(>B=fe#4,<Z1=fe#13); */
/* asm 2: fe_sq2(>B=r->T,<Z1=p->Z); */
fe_sq2(r->T,p->Z);
/* qhasm: A=X1+Y1 */
/* asm 1: fe_add(>A=fe#2,<X1=fe#11,<Y1=fe#12); */
/* asm 2: fe_add(>A=r->Y,<X1=p->X,<Y1=p->Y); */
fe_add(r->Y,p->X,p->Y);
/* qhasm: AA=A^2 */
/* asm 1: fe_sq(>AA=fe#5,<A=fe#2); */
/* asm 2: fe_sq(>AA=t0,<A=r->Y); */
fe_sq(t0,r->Y);
/* qhasm: Y3=YY+XX */
/* asm 1: fe_add(>Y3=fe#2,<YY=fe#3,<XX=fe#1); */
/* asm 2: fe_add(>Y3=r->Y,<YY=r->Z,<XX=r->X); */
fe_add(r->Y,r->Z,r->X);
/* qhasm: Z3=YY-XX */
/* asm 1: fe_sub(>Z3=fe#3,<YY=fe#3,<XX=fe#1); */
/* asm 2: fe_sub(>Z3=r->Z,<YY=r->Z,<XX=r->X); */
fe_sub(r->Z,r->Z,r->X);
/* qhasm: X3=AA-Y3 */
/* asm 1: fe_sub(>X3=fe#1,<AA=fe#5,<Y3=fe#2); */
/* asm 2: fe_sub(>X3=r->X,<AA=t0,<Y3=r->Y); */
fe_sub(r->X,t0,r->Y);
/* qhasm: T3=B-Z3 */
/* asm 1: fe_sub(>T3=fe#4,<B=fe#4,<Z3=fe#3); */
/* asm 2: fe_sub(>T3=r->T,<B=r->T,<Z3=r->Z); */
fe_sub(r->T,r->T,r->Z);
/* qhasm: return */
}

9
ge_p3_0.c Normal file
View File

@ -0,0 +1,9 @@
#include "ge.h"
void ge_p3_0(ge_p3 *h)
{
fe_0(h->X);
fe_1(h->Y);
fe_1(h->Z);
fe_0(h->T);
}

12
ge_p3_dbl.c Normal file
View File

@ -0,0 +1,12 @@
#include "ge.h"
/*
r = 2 * p
*/
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);
}

17
ge_p3_to_cached.c Normal file
View File

@ -0,0 +1,17 @@
#include "ge.h"
/*
r = p
*/
static const fe d2 = {
-21827239,-5839606,-30745221,13898782,229458,15978800,-12551817,-6495438,29715968,9444199
} ;
extern 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,d2);
}

12
ge_p3_to_p2.c Normal file
View File

@ -0,0 +1,12 @@
#include "ge.h"
/*
r = p
*/
extern 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);
}

14
ge_p3_tobytes.c Normal file
View File

@ -0,0 +1,14 @@
#include "ge.h"
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;
}

8
ge_precomp_0.c Normal file
View File

@ -0,0 +1,8 @@
#include "ge.h"
void ge_precomp_0(ge_precomp *h)
{
fe_1(h->yplusx);
fe_1(h->yminusx);
fe_0(h->xy2d);
}

1449
ge_scalarmult_base.c Normal file

File diff suppressed because it is too large Load Diff

108
ge_sub.c Normal file
View File

@ -0,0 +1,108 @@
#include "ge.h"
/*
r = p - q
*/
void ge_sub(ge_p1p1 *r,const ge_p3 *p,const ge_cached *q)
{
fe t0;
/* qhasm: enter ge_sub */
/* qhasm: fe X1 */
/* qhasm: fe Y1 */
/* qhasm: fe Z1 */
/* qhasm: fe Z2 */
/* qhasm: fe T1 */
/* qhasm: fe ZZ */
/* qhasm: fe YpX2 */
/* qhasm: fe YmX2 */
/* qhasm: fe T2d2 */
/* qhasm: fe X3 */
/* qhasm: fe Y3 */
/* qhasm: fe Z3 */
/* qhasm: fe T3 */
/* qhasm: fe YpX1 */
/* qhasm: fe YmX1 */
/* qhasm: fe A */
/* qhasm: fe B */
/* qhasm: fe C */
/* qhasm: fe D */
/* qhasm: YpX1 = Y1+X1 */
/* asm 1: fe_add(>YpX1=fe#1,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_add(>YpX1=r->X,<Y1=p->Y,<X1=p->X); */
fe_add(r->X,p->Y,p->X);
/* qhasm: YmX1 = Y1-X1 */
/* asm 1: fe_sub(>YmX1=fe#2,<Y1=fe#12,<X1=fe#11); */
/* asm 2: fe_sub(>YmX1=r->Y,<Y1=p->Y,<X1=p->X); */
fe_sub(r->Y,p->Y,p->X);
/* qhasm: A = YpX1*YmX2 */
/* asm 1: fe_mul(>A=fe#3,<YpX1=fe#1,<YmX2=fe#16); */
/* asm 2: fe_mul(>A=r->Z,<YpX1=r->X,<YmX2=q->YminusX); */
fe_mul(r->Z,r->X,q->YminusX);
/* qhasm: B = YmX1*YpX2 */
/* asm 1: fe_mul(>B=fe#2,<YmX1=fe#2,<YpX2=fe#15); */
/* asm 2: fe_mul(>B=r->Y,<YmX1=r->Y,<YpX2=q->YplusX); */
fe_mul(r->Y,r->Y,q->YplusX);
/* qhasm: C = T2d2*T1 */
/* asm 1: fe_mul(>C=fe#4,<T2d2=fe#18,<T1=fe#14); */
/* asm 2: fe_mul(>C=r->T,<T2d2=q->T2d,<T1=p->T); */
fe_mul(r->T,q->T2d,p->T);
/* qhasm: ZZ = Z1*Z2 */
/* asm 1: fe_mul(>ZZ=fe#1,<Z1=fe#13,<Z2=fe#17); */
/* asm 2: fe_mul(>ZZ=r->X,<Z1=p->Z,<Z2=q->Z); */
fe_mul(r->X,p->Z,q->Z);
/* qhasm: D = 2*ZZ */
/* asm 1: fe_add(>D=fe#5,<ZZ=fe#1,<ZZ=fe#1); */
/* asm 2: fe_add(>D=t0,<ZZ=r->X,<ZZ=r->X); */
fe_add(t0,r->X,r->X);
/* qhasm: X3 = A-B */
/* asm 1: fe_sub(>X3=fe#1,<A=fe#3,<B=fe#2); */
/* asm 2: fe_sub(>X3=r->X,<A=r->Z,<B=r->Y); */
fe_sub(r->X,r->Z,r->Y);
/* qhasm: Y3 = A+B */
/* asm 1: fe_add(>Y3=fe#2,<A=fe#3,<B=fe#2); */
/* asm 2: fe_add(>Y3=r->Y,<A=r->Z,<B=r->Y); */
fe_add(r->Y,r->Z,r->Y);
/* qhasm: Z3 = D-C */
/* asm 1: fe_sub(>Z3=fe#3,<D=fe#5,<C=fe#4); */
/* asm 2: fe_sub(>Z3=r->Z,<D=t0,<C=r->T); */
fe_sub(r->Z,t0,r->T);
/* qhasm: T3 = D+C */
/* asm 1: fe_add(>T3=fe#4,<D=fe#5,<C=fe#4); */
/* asm 2: fe_add(>T3=r->T,<D=t0,<C=r->T); */
fe_add(r->T,t0,r->T);
/* qhasm: return */
}

14
ge_tobytes.c Normal file
View File

@ -0,0 +1,14 @@
#include "ge.h"
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;
}

22
keypair.c Normal file
View File

@ -0,0 +1,22 @@
#include "crypto_sign.h"
#include "sha512.h"
#include "ge.h"
int crypto_sign_keypair(unsigned char *pk, unsigned char *sk, unsigned char *seed)
{
unsigned char h[64];
ge_p3 A;
int i;
sha512(h,seed,32);
h[0] &= 248;
h[31] &= 63;
h[31] |= 64;
ge_scalarmult_base(&A,h);
ge_p3_tobytes(pk,&A);
for (i = 0;i < 32;++i) sk[i] = seed[i];
for (i = 0;i < 32;++i) sk[32 + i] = pk[i];
return 0;
}

39
open.c Normal file
View File

@ -0,0 +1,39 @@
#include "crypto_sign.h"
#include "sha512.h"
#include "ge.h"
#include "sc.h"
int crypto_sign_open(
unsigned char *m,unsigned long long *mlen,
const unsigned char *sm,unsigned long long smlen,
const unsigned char *pk
)
{
unsigned char h[64];
unsigned char checkr[32];
ge_p3 A;
ge_p2 R;
unsigned long long i;
*mlen = -1;
if (smlen < 64) return -1;
if (sm[63] & 224) return -1;
if (ge_frombytes_negate_vartime(&A,pk) != 0) return -1;
for (i = 0;i < smlen;++i) m[i] = sm[i];
for (i = 0;i < 32;++i) m[32 + i] = pk[i];
sha512(h,m,smlen);
sc_reduce(h);
ge_double_scalarmult_vartime(&R,h,&A,sm + 32);
ge_tobytes(checkr,&R);
if (crypto_verify_32(checkr,sm) != 0) {
for (i = 0;i < smlen;++i) m[i] = 0;
return -1;
}
for (i = 0;i < smlen - 64;++i) m[i] = sm[64 + i];
for (i = smlen - 64;i < smlen;++i) m[i] = 0;
*mlen = smlen - 64;
return 0;
}

800
pstdint.h Normal file
View File

@ -0,0 +1,800 @@
/* A portable stdint.h
****************************************************************************
* BSD License:
****************************************************************************
*
* Copyright (c) 2005-2011 Paul Hsieh
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
*
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
* 3. The name of the author may not be used to endorse or promote products
* derived from this software without specific prior written permission.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR ``AS IS'' AND ANY EXPRESS OR
* IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES
* OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE DISCLAIMED.
* IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR ANY DIRECT, INDIRECT,
* INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT
* NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE,
* DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY
* THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT
* (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF
* THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
*
****************************************************************************
*
* Version 0.1.12
*
* The ANSI C standard committee, for the C99 standard, specified the
* inclusion of a new standard include file called stdint.h. This is
* a very useful and long desired include file which contains several
* very precise definitions for integer scalar types that is
* critically important for making portable several classes of
* applications including cryptography, hashing, variable length
* integer libraries and so on. But for most developers its likely
* useful just for programming sanity.
*
* The problem is that most compiler vendors have decided not to
* implement the C99 standard, and the next C++ language standard
* (which has a lot more mindshare these days) will be a long time in
* coming and its unknown whether or not it will include stdint.h or
* how much adoption it will have. Either way, it will be a long time
* before all compilers come with a stdint.h and it also does nothing
* for the extremely large number of compilers available today which
* do not include this file, or anything comparable to it.
*
* So that's what this file is all about. Its an attempt to build a
* single universal include file that works on as many platforms as
* possible to deliver what stdint.h is supposed to. A few things
* that should be noted about this file:
*
* 1) It is not guaranteed to be portable and/or present an identical
* interface on all platforms. The extreme variability of the
* ANSI C standard makes this an impossibility right from the
* very get go. Its really only meant to be useful for the vast
* majority of platforms that possess the capability of
* implementing usefully and precisely defined, standard sized
* integer scalars. Systems which are not intrinsically 2s
* complement may produce invalid constants.
*
* 2) There is an unavoidable use of non-reserved symbols.
*
* 3) Other standard include files are invoked.
*
* 4) This file may come in conflict with future platforms that do
* include stdint.h. The hope is that one or the other can be
* used with no real difference.
*
* 5) In the current verison, if your platform can't represent
* int32_t, int16_t and int8_t, it just dumps out with a compiler
* error.
*
* 6) 64 bit integers may or may not be defined. Test for their
* presence with the test: #ifdef INT64_MAX or #ifdef UINT64_MAX.
* Note that this is different from the C99 specification which
* requires the existence of 64 bit support in the compiler. If
* this is not defined for your platform, yet it is capable of
* dealing with 64 bits then it is because this file has not yet
* been extended to cover all of your system's capabilities.
*
* 7) (u)intptr_t may or may not be defined. Test for its presence
* with the test: #ifdef PTRDIFF_MAX. If this is not defined
* for your platform, then it is because this file has not yet
* been extended to cover all of your system's capabilities, not
* because its optional.
*
* 8) The following might not been defined even if your platform is
* capable of defining it:
*
* WCHAR_MIN
* WCHAR_MAX
* (u)int64_t
* PTRDIFF_MIN
* PTRDIFF_MAX
* (u)intptr_t
*
* 9) The following have not been defined:
*
* WINT_MIN
* WINT_MAX
*
* 10) The criteria for defining (u)int_least(*)_t isn't clear,
* except for systems which don't have a type that precisely
* defined 8, 16, or 32 bit types (which this include file does
* not support anyways). Default definitions have been given.
*
* 11) The criteria for defining (u)int_fast(*)_t isn't something I
* would trust to any particular compiler vendor or the ANSI C
* committee. It is well known that "compatible systems" are
* commonly created that have very different performance
* characteristics from the systems they are compatible with,
* especially those whose vendors make both the compiler and the
* system. Default definitions have been given, but its strongly
* recommended that users never use these definitions for any
* reason (they do *NOT* deliver any serious guarantee of
* improved performance -- not in this file, nor any vendor's
* stdint.h).
*
* 12) The following macros:
*
* PRINTF_INTMAX_MODIFIER
* PRINTF_INT64_MODIFIER
* PRINTF_INT32_MODIFIER
* PRINTF_INT16_MODIFIER
* PRINTF_LEAST64_MODIFIER
* PRINTF_LEAST32_MODIFIER
* PRINTF_LEAST16_MODIFIER
* PRINTF_INTPTR_MODIFIER
*
* are strings which have been defined as the modifiers required
* for the "d", "u" and "x" printf formats to correctly output
* (u)intmax_t, (u)int64_t, (u)int32_t, (u)int16_t, (u)least64_t,
* (u)least32_t, (u)least16_t and (u)intptr_t types respectively.
* PRINTF_INTPTR_MODIFIER is not defined for some systems which
* provide their own stdint.h. PRINTF_INT64_MODIFIER is not
* defined if INT64_MAX is not defined. These are an extension
* beyond what C99 specifies must be in stdint.h.
*
* In addition, the following macros are defined:
*
* PRINTF_INTMAX_HEX_WIDTH
* PRINTF_INT64_HEX_WIDTH
* PRINTF_INT32_HEX_WIDTH
* PRINTF_INT16_HEX_WIDTH
* PRINTF_INT8_HEX_WIDTH
* PRINTF_INTMAX_DEC_WIDTH
* PRINTF_INT64_DEC_WIDTH
* PRINTF_INT32_DEC_WIDTH
* PRINTF_INT16_DEC_WIDTH
* PRINTF_INT8_DEC_WIDTH
*
* Which specifies the maximum number of characters required to
* print the number of that type in either hexadecimal or decimal.
* These are an extension beyond what C99 specifies must be in
* stdint.h.
*
* Compilers tested (all with 0 warnings at their highest respective
* settings): Borland Turbo C 2.0, WATCOM C/C++ 11.0 (16 bits and 32
* bits), Microsoft Visual C++ 6.0 (32 bit), Microsoft Visual Studio
* .net (VC7), Intel C++ 4.0, GNU gcc v3.3.3
*
* This file should be considered a work in progress. Suggestions for
* improvements, especially those which increase coverage are strongly
* encouraged.
*
* Acknowledgements
*
* The following people have made significant contributions to the
* development and testing of this file:
*
* Chris Howie
* John Steele Scott
* Dave Thorup
* John Dill
*
*/
#include <stddef.h>
#include <limits.h>
#include <signal.h>
/*
* For gcc with _STDINT_H, fill in the PRINTF_INT*_MODIFIER macros, and
* do nothing else. On the Mac OS X version of gcc this is _STDINT_H_.
*/
#if ((defined(__STDC__) && __STDC__ && __STDC_VERSION__ >= 199901L) || (defined (__WATCOMC__) && (defined (_STDINT_H_INCLUDED) || __WATCOMC__ >= 1250)) || (defined(__GNUC__) && (defined(_STDINT_H) || defined(_STDINT_H_) || defined (__UINT_FAST64_TYPE__)) )) && !defined (_PSTDINT_H_INCLUDED)
#include <stdint.h>
#define _PSTDINT_H_INCLUDED
# ifndef PRINTF_INT64_MODIFIER
# define PRINTF_INT64_MODIFIER "ll"
# endif
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER "l"
# endif
# ifndef PRINTF_INT16_MODIFIER
# define PRINTF_INT16_MODIFIER "h"
# endif
# ifndef PRINTF_INTMAX_MODIFIER
# define PRINTF_INTMAX_MODIFIER PRINTF_INT64_MODIFIER
# endif
# ifndef PRINTF_INT64_HEX_WIDTH
# define PRINTF_INT64_HEX_WIDTH "16"
# endif
# ifndef PRINTF_INT32_HEX_WIDTH
# define PRINTF_INT32_HEX_WIDTH "8"
# endif
# ifndef PRINTF_INT16_HEX_WIDTH
# define PRINTF_INT16_HEX_WIDTH "4"
# endif
# ifndef PRINTF_INT8_HEX_WIDTH
# define PRINTF_INT8_HEX_WIDTH "2"
# endif
# ifndef PRINTF_INT64_DEC_WIDTH
# define PRINTF_INT64_DEC_WIDTH "20"
# endif
# ifndef PRINTF_INT32_DEC_WIDTH
# define PRINTF_INT32_DEC_WIDTH "10"
# endif
# ifndef PRINTF_INT16_DEC_WIDTH
# define PRINTF_INT16_DEC_WIDTH "5"
# endif
# ifndef PRINTF_INT8_DEC_WIDTH
# define PRINTF_INT8_DEC_WIDTH "3"
# endif
# ifndef PRINTF_INTMAX_HEX_WIDTH
# define PRINTF_INTMAX_HEX_WIDTH PRINTF_INT64_HEX_WIDTH
# endif
# ifndef PRINTF_INTMAX_DEC_WIDTH
# define PRINTF_INTMAX_DEC_WIDTH PRINTF_INT64_DEC_WIDTH
# endif
/*
* Something really weird is going on with Open Watcom. Just pull some of
* these duplicated definitions from Open Watcom's stdint.h file for now.
*/
# if defined (__WATCOMC__) && __WATCOMC__ >= 1250
# if !defined (INT64_C)
# define INT64_C(x) (x + (INT64_MAX - INT64_MAX))
# endif
# if !defined (UINT64_C)
# define UINT64_C(x) (x + (UINT64_MAX - UINT64_MAX))
# endif
# if !defined (INT32_C)
# define INT32_C(x) (x + (INT32_MAX - INT32_MAX))
# endif
# if !defined (UINT32_C)
# define UINT32_C(x) (x + (UINT32_MAX - UINT32_MAX))
# endif
# if !defined (INT16_C)
# define INT16_C(x) (x)
# endif
# if !defined (UINT16_C)
# define UINT16_C(x) (x)
# endif
# if !defined (INT8_C)
# define INT8_C(x) (x)
# endif
# if !defined (UINT8_C)
# define UINT8_C(x) (x)
# endif
# if !defined (UINT64_MAX)
# define UINT64_MAX 18446744073709551615ULL
# endif
# if !defined (INT64_MAX)
# define INT64_MAX 9223372036854775807LL
# endif
# if !defined (UINT32_MAX)
# define UINT32_MAX 4294967295UL
# endif
# if !defined (INT32_MAX)
# define INT32_MAX 2147483647L
# endif
# if !defined (INTMAX_MAX)
# define INTMAX_MAX INT64_MAX
# endif
# if !defined (INTMAX_MIN)
# define INTMAX_MIN INT64_MIN
# endif
# endif
#endif
#ifndef _PSTDINT_H_INCLUDED
#define _PSTDINT_H_INCLUDED
#ifndef SIZE_MAX
# define SIZE_MAX (~(size_t)0)
#endif
/*
* Deduce the type assignments from limits.h under the assumption that
* integer sizes in bits are powers of 2, and follow the ANSI
* definitions.
*/
#ifndef UINT8_MAX
# define UINT8_MAX 0xff
#endif
#ifndef uint8_t
# if (UCHAR_MAX == UINT8_MAX) || defined (S_SPLINT_S)
typedef unsigned char uint8_t;
# define UINT8_C(v) ((uint8_t) v)
# else
# error "Platform not supported"
# endif
#endif
#ifndef INT8_MAX
# define INT8_MAX 0x7f
#endif
#ifndef INT8_MIN
# define INT8_MIN INT8_C(0x80)
#endif
#ifndef int8_t
# if (SCHAR_MAX == INT8_MAX) || defined (S_SPLINT_S)
typedef signed char int8_t;
# define INT8_C(v) ((int8_t) v)
# else
# error "Platform not supported"
# endif
#endif
#ifndef UINT16_MAX
# define UINT16_MAX 0xffff
#endif
#ifndef uint16_t
#if (UINT_MAX == UINT16_MAX) || defined (S_SPLINT_S)
typedef unsigned int uint16_t;
# ifndef PRINTF_INT16_MODIFIER
# define PRINTF_INT16_MODIFIER ""
# endif
# define UINT16_C(v) ((uint16_t) (v))
#elif (USHRT_MAX == UINT16_MAX)
typedef unsigned short uint16_t;
# define UINT16_C(v) ((uint16_t) (v))
# ifndef PRINTF_INT16_MODIFIER
# define PRINTF_INT16_MODIFIER "h"
# endif
#else
#error "Platform not supported"
#endif
#endif
#ifndef INT16_MAX
# define INT16_MAX 0x7fff
#endif
#ifndef INT16_MIN
# define INT16_MIN INT16_C(0x8000)
#endif
#ifndef int16_t
#if (INT_MAX == INT16_MAX) || defined (S_SPLINT_S)
typedef signed int int16_t;
# define INT16_C(v) ((int16_t) (v))
# ifndef PRINTF_INT16_MODIFIER
# define PRINTF_INT16_MODIFIER ""
# endif
#elif (SHRT_MAX == INT16_MAX)
typedef signed short int16_t;
# define INT16_C(v) ((int16_t) (v))
# ifndef PRINTF_INT16_MODIFIER
# define PRINTF_INT16_MODIFIER "h"
# endif
#else
#error "Platform not supported"
#endif
#endif
#ifndef UINT32_MAX
# define UINT32_MAX (0xffffffffUL)
#endif
#ifndef uint32_t
#if (ULONG_MAX == UINT32_MAX) || defined (S_SPLINT_S)
typedef unsigned long uint32_t;
# define UINT32_C(v) v ## UL
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER "l"
# endif
#elif (UINT_MAX == UINT32_MAX)
typedef unsigned int uint32_t;
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER ""
# endif
# define UINT32_C(v) v ## U
#elif (USHRT_MAX == UINT32_MAX)
typedef unsigned short uint32_t;
# define UINT32_C(v) ((unsigned short) (v))
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER ""
# endif
#else
#error "Platform not supported"
#endif
#endif
#ifndef INT32_MAX
# define INT32_MAX (0x7fffffffL)
#endif
#ifndef INT32_MIN
# define INT32_MIN INT32_C(0x80000000)
#endif
#ifndef int32_t
#if (LONG_MAX == INT32_MAX) || defined (S_SPLINT_S)
typedef signed long int32_t;
# define INT32_C(v) v ## L
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER "l"
# endif
#elif (INT_MAX == INT32_MAX)
typedef signed int int32_t;
# define INT32_C(v) v
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER ""
# endif
#elif (SHRT_MAX == INT32_MAX)
typedef signed short int32_t;
# define INT32_C(v) ((short) (v))
# ifndef PRINTF_INT32_MODIFIER
# define PRINTF_INT32_MODIFIER ""
# endif
#else
#error "Platform not supported"
#endif
#endif
/*
* The macro stdint_int64_defined is temporarily used to record
* whether or not 64 integer support is available. It must be
* defined for any 64 integer extensions for new platforms that are
* added.
*/
#undef stdint_int64_defined
#if (defined(__STDC__) && defined(__STDC_VERSION__)) || defined (S_SPLINT_S)
# if (__STDC__ && __STDC_VERSION__ >= 199901L) || defined (S_SPLINT_S)
# define stdint_int64_defined
typedef long long int64_t;
typedef unsigned long long uint64_t;
# define UINT64_C(v) v ## ULL
# define INT64_C(v) v ## LL
# ifndef PRINTF_INT64_MODIFIER
# define PRINTF_INT64_MODIFIER "ll"
# endif
# endif
#endif
#if !defined (stdint_int64_defined)
# if defined(__GNUC__)
# define stdint_int64_defined
__extension__ typedef long long int64_t;
__extension__ typedef unsigned long long uint64_t;
# define UINT64_C(v) v ## ULL
# define INT64_C(v) v ## LL
# ifndef PRINTF_INT64_MODIFIER
# define PRINTF_INT64_MODIFIER "ll"
# endif
# elif defined(__MWERKS__) || defined (__SUNPRO_C) || defined (__SUNPRO_CC) || defined (__APPLE_CC__) || defined (_LONG_LONG) || defined (_CRAYC) || defined (S_SPLINT_S)
# define stdint_int64_defined
typedef long long int64_t;
typedef unsigned long long uint64_t;
# define UINT64_C(v) v ## ULL
# define INT64_C(v) v ## LL
# ifndef PRINTF_INT64_MODIFIER
# define PRINTF_INT64_MODIFIER "ll"
# endif
# elif (defined(__WATCOMC__) && defined(__WATCOM_INT64__)) || (defined(_MSC_VER) && _INTEGRAL_MAX_BITS >= 64) || (defined (__BORLANDC__) && __BORLANDC__ > 0x460) || defined (__alpha) || defined (__DECC)
# define stdint_int64_defined
typedef __int64 int64_t;
typedef unsigned __int64 uint64_t;
# define UINT64_C(v) v ## UI64
# define INT64_C(v) v ## I64
# ifndef PRINTF_INT64_MODIFIER
# define PRINTF_INT64_MODIFIER "I64"
# endif
# endif
#endif
#if !defined (LONG_LONG_MAX) && defined (INT64_C)
# define LONG_LONG_MAX INT64_C (9223372036854775807)
#endif
#ifndef ULONG_LONG_MAX
# define ULONG_LONG_MAX UINT64_C (18446744073709551615)
#endif
#if !defined (INT64_MAX) && defined (INT64_C)
# define INT64_MAX INT64_C (9223372036854775807)
#endif
#if !defined (INT64_MIN) && defined (INT64_C)
# define INT64_MIN INT64_C (-9223372036854775808)
#endif
#if !defined (UINT64_MAX) && defined (INT64_C)
# define UINT64_MAX UINT64_C (18446744073709551615)
#endif
/*
* Width of hexadecimal for number field.
*/
#ifndef PRINTF_INT64_HEX_WIDTH
# define PRINTF_INT64_HEX_WIDTH "16"
#endif
#ifndef PRINTF_INT32_HEX_WIDTH
# define PRINTF_INT32_HEX_WIDTH "8"
#endif
#ifndef PRINTF_INT16_HEX_WIDTH
# define PRINTF_INT16_HEX_WIDTH "4"
#endif
#ifndef PRINTF_INT8_HEX_WIDTH
# define PRINTF_INT8_HEX_WIDTH "2"
#endif
#ifndef PRINTF_INT64_DEC_WIDTH
# define PRINTF_INT64_DEC_WIDTH "20"
#endif
#ifndef PRINTF_INT32_DEC_WIDTH
# define PRINTF_INT32_DEC_WIDTH "10"
#endif
#ifndef PRINTF_INT16_DEC_WIDTH
# define PRINTF_INT16_DEC_WIDTH "5"
#endif
#ifndef PRINTF_INT8_DEC_WIDTH
# define PRINTF_INT8_DEC_WIDTH "3"
#endif
/*
* Ok, lets not worry about 128 bit integers for now. Moore's law says
* we don't need to worry about that until about 2040 at which point
* we'll have bigger things to worry about.
*/
#ifdef stdint_int64_defined
typedef int64_t intmax_t;
typedef uint64_t uintmax_t;
# define INTMAX_MAX INT64_MAX
# define INTMAX_MIN INT64_MIN
# define UINTMAX_MAX UINT64_MAX
# define UINTMAX_C(v) UINT64_C(v)
# define INTMAX_C(v) INT64_C(v)
# ifndef PRINTF_INTMAX_MODIFIER
# define PRINTF_INTMAX_MODIFIER PRINTF_INT64_MODIFIER
# endif
# ifndef PRINTF_INTMAX_HEX_WIDTH
# define PRINTF_INTMAX_HEX_WIDTH PRINTF_INT64_HEX_WIDTH
# endif
# ifndef PRINTF_INTMAX_DEC_WIDTH
# define PRINTF_INTMAX_DEC_WIDTH PRINTF_INT64_DEC_WIDTH
# endif
#else
typedef int32_t intmax_t;
typedef uint32_t uintmax_t;
# define INTMAX_MAX INT32_MAX
# define UINTMAX_MAX UINT32_MAX
# define UINTMAX_C(v) UINT32_C(v)
# define INTMAX_C(v) INT32_C(v)
# ifndef PRINTF_INTMAX_MODIFIER
# define PRINTF_INTMAX_MODIFIER PRINTF_INT32_MODIFIER
# endif
# ifndef PRINTF_INTMAX_HEX_WIDTH
# define PRINTF_INTMAX_HEX_WIDTH PRINTF_INT32_HEX_WIDTH
# endif
# ifndef PRINTF_INTMAX_DEC_WIDTH
# define PRINTF_INTMAX_DEC_WIDTH PRINTF_INT32_DEC_WIDTH
# endif
#endif
/*
* Because this file currently only supports platforms which have
* precise powers of 2 as bit sizes for the default integers, the
* least definitions are all trivial. Its possible that a future
* version of this file could have different definitions.
*/
#ifndef stdint_least_defined
typedef int8_t int_least8_t;
typedef uint8_t uint_least8_t;
typedef int16_t int_least16_t;
typedef uint16_t uint_least16_t;
typedef int32_t int_least32_t;
typedef uint32_t uint_least32_t;
# define PRINTF_LEAST32_MODIFIER PRINTF_INT32_MODIFIER
# define PRINTF_LEAST16_MODIFIER PRINTF_INT16_MODIFIER
# define UINT_LEAST8_MAX UINT8_MAX
# define INT_LEAST8_MAX INT8_MAX
# define UINT_LEAST16_MAX UINT16_MAX
# define INT_LEAST16_MAX INT16_MAX
# define UINT_LEAST32_MAX UINT32_MAX
# define INT_LEAST32_MAX INT32_MAX
# define INT_LEAST8_MIN INT8_MIN
# define INT_LEAST16_MIN INT16_MIN
# define INT_LEAST32_MIN INT32_MIN
# ifdef stdint_int64_defined
typedef int64_t int_least64_t;
typedef uint64_t uint_least64_t;
# define PRINTF_LEAST64_MODIFIER PRINTF_INT64_MODIFIER
# define UINT_LEAST64_MAX UINT64_MAX
# define INT_LEAST64_MAX INT64_MAX
# define INT_LEAST64_MIN INT64_MIN
# endif
#endif
#undef stdint_least_defined
/*
* The ANSI C committee pretending to know or specify anything about
* performance is the epitome of misguided arrogance. The mandate of
* this file is to *ONLY* ever support that absolute minimum
* definition of the fast integer types, for compatibility purposes.
* No extensions, and no attempt to suggest what may or may not be a
* faster integer type will ever be made in this file. Developers are
* warned to stay away from these types when using this or any other
* stdint.h.
*/
typedef int_least8_t int_fast8_t;
typedef uint_least8_t uint_fast8_t;
typedef int_least16_t int_fast16_t;
typedef uint_least16_t uint_fast16_t;
typedef int_least32_t int_fast32_t;
typedef uint_least32_t uint_fast32_t;
#define UINT_FAST8_MAX UINT_LEAST8_MAX
#define INT_FAST8_MAX INT_LEAST8_MAX
#define UINT_FAST16_MAX UINT_LEAST16_MAX
#define INT_FAST16_MAX INT_LEAST16_MAX
#define UINT_FAST32_MAX UINT_LEAST32_MAX
#define INT_FAST32_MAX INT_LEAST32_MAX
#define INT_FAST8_MIN INT_LEAST8_MIN
#define INT_FAST16_MIN INT_LEAST16_MIN
#define INT_FAST32_MIN INT_LEAST32_MIN
#ifdef stdint_int64_defined
typedef int_least64_t int_fast64_t;
typedef uint_least64_t uint_fast64_t;
# define UINT_FAST64_MAX UINT_LEAST64_MAX
# define INT_FAST64_MAX INT_LEAST64_MAX
# define INT_FAST64_MIN INT_LEAST64_MIN
#endif
#undef stdint_int64_defined
/*
* Whatever piecemeal, per compiler thing we can do about the wchar_t
* type limits.
*/
#if defined(__WATCOMC__) || defined(_MSC_VER) || defined (__GNUC__)
# include <wchar.h>
# ifndef WCHAR_MIN
# define WCHAR_MIN 0
# endif
# ifndef WCHAR_MAX
# define WCHAR_MAX ((wchar_t)-1)
# endif
#endif
/*
* Whatever piecemeal, per compiler/platform thing we can do about the
* (u)intptr_t types and limits.
*/
#if defined (_MSC_VER) && defined (_UINTPTR_T_DEFINED)
# define STDINT_H_UINTPTR_T_DEFINED
#endif
#ifndef STDINT_H_UINTPTR_T_DEFINED
# if defined (__alpha__) || defined (__ia64__) || defined (__x86_64__) || defined (_WIN64)
# define stdint_intptr_bits 64
# elif defined (__WATCOMC__) || defined (__TURBOC__)
# if defined(__TINY__) || defined(__SMALL__) || defined(__MEDIUM__)
# define stdint_intptr_bits 16
# else
# define stdint_intptr_bits 32
# endif
# elif defined (__i386__) || defined (_WIN32) || defined (WIN32)
# define stdint_intptr_bits 32
# elif defined (__INTEL_COMPILER)
/* TODO -- what did Intel do about x86-64? */
# endif
# ifdef stdint_intptr_bits
# define stdint_intptr_glue3_i(a,b,c) a##b##c
# define stdint_intptr_glue3(a,b,c) stdint_intptr_glue3_i(a,b,c)
# ifndef PRINTF_INTPTR_MODIFIER
# define PRINTF_INTPTR_MODIFIER stdint_intptr_glue3(PRINTF_INT,stdint_intptr_bits,_MODIFIER)
# endif
# ifndef PTRDIFF_MAX
# define PTRDIFF_MAX stdint_intptr_glue3(INT,stdint_intptr_bits,_MAX)
# endif
# ifndef PTRDIFF_MIN
# define PTRDIFF_MIN stdint_intptr_glue3(INT,stdint_intptr_bits,_MIN)
# endif
# ifndef UINTPTR_MAX
# define UINTPTR_MAX stdint_intptr_glue3(UINT,stdint_intptr_bits,_MAX)
# endif
# ifndef INTPTR_MAX
# define INTPTR_MAX stdint_intptr_glue3(INT,stdint_intptr_bits,_MAX)
# endif
# ifndef INTPTR_MIN
# define INTPTR_MIN stdint_intptr_glue3(INT,stdint_intptr_bits,_MIN)
# endif
# ifndef INTPTR_C
# define INTPTR_C(x) stdint_intptr_glue3(INT,stdint_intptr_bits,_C)(x)
# endif
# ifndef UINTPTR_C
# define UINTPTR_C(x) stdint_intptr_glue3(UINT,stdint_intptr_bits,_C)(x)
# endif
typedef stdint_intptr_glue3(uint,stdint_intptr_bits,_t) uintptr_t;
typedef stdint_intptr_glue3( int,stdint_intptr_bits,_t) intptr_t;
# else
/* TODO -- This following is likely wrong for some platforms, and does
nothing for the definition of uintptr_t. */
typedef ptrdiff_t intptr_t;
# endif
# define STDINT_H_UINTPTR_T_DEFINED
#endif
/*
* Assumes sig_atomic_t is signed and we have a 2s complement machine.
*/
#ifndef SIG_ATOMIC_MAX
# define SIG_ATOMIC_MAX ((((sig_atomic_t) 1) << (sizeof (sig_atomic_t)*CHAR_BIT-1)) - 1)
#endif
#endif
#if defined (__TEST_PSTDINT_FOR_CORRECTNESS)
/*
* Please compile with the maximum warning settings to make sure macros are not
* defined more than once.
*/
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#define glue3_aux(x,y,z) x ## y ## z
#define glue3(x,y,z) glue3_aux(x,y,z)
#define DECLU(bits) glue3(uint,bits,_t) glue3(u,bits,=) glue3(UINT,bits,_C) (0);
#define DECLI(bits) glue3(int,bits,_t) glue3(i,bits,=) glue3(INT,bits,_C) (0);
#define DECL(us,bits) glue3(DECL,us,) (bits)
#define TESTUMAX(bits) glue3(u,bits,=) glue3(~,u,bits); if (glue3(UINT,bits,_MAX) glue3(!=,u,bits)) printf ("Something wrong with UINT%d_MAX\n", bits)
int main () {
DECL(I,8)
DECL(U,8)
DECL(I,16)
DECL(U,16)
DECL(I,32)
DECL(U,32)
#ifdef INT64_MAX
DECL(I,64)
DECL(U,64)
#endif
intmax_t imax = INTMAX_C(0);
uintmax_t umax = UINTMAX_C(0);
char str0[256], str1[256];
sprintf (str0, "%d %x\n", 0, ~0);
sprintf (str1, "%d %x\n", i8, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with i8 : %s\n", str1);
sprintf (str1, "%u %x\n", u8, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with u8 : %s\n", str1);
sprintf (str1, "%d %x\n", i16, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with i16 : %s\n", str1);
sprintf (str1, "%u %x\n", u16, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with u16 : %s\n", str1);
sprintf (str1, "%" PRINTF_INT32_MODIFIER "d %x\n", i32, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with i32 : %s\n", str1);
sprintf (str1, "%" PRINTF_INT32_MODIFIER "u %x\n", u32, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with u32 : %s\n", str1);
#ifdef INT64_MAX
sprintf (str1, "%" PRINTF_INT64_MODIFIER "d %x\n", i64, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with i64 : %s\n", str1);
#endif
sprintf (str1, "%" PRINTF_INTMAX_MODIFIER "d %x\n", imax, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with imax : %s\n", str1);
sprintf (str1, "%" PRINTF_INTMAX_MODIFIER "u %x\n", umax, ~0);
if (0 != strcmp (str0, str1)) printf ("Something wrong with umax : %s\n", str1);
TESTUMAX(8);
TESTUMAX(16);
TESTUMAX(32);
#ifdef INT64_MAX
TESTUMAX(64);
#endif
return EXIT_SUCCESS;
}
#endif

15
sc.h Normal file
View File

@ -0,0 +1,15 @@
#ifndef SC_H
#define SC_H
/*
The set of scalars is \Z/l
where l = 2^252 + 27742317777372353535851937790883648493.
*/
#define sc_reduce crypto_sign_ed25519_ref10_sc_reduce
#define sc_muladd crypto_sign_ed25519_ref10_sc_muladd
extern void sc_reduce(unsigned char *);
extern void sc_muladd(unsigned char *,const unsigned char *,const unsigned char *,const unsigned char *);
#endif

366
sc_muladd.c Normal file
View File

@ -0,0 +1,366 @@
#include "sc.h"
#include "pstdint.h"
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;
}
/*
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 + (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;
s23 = 0;
s10 += s22 * 666643;
s11 += s22 * 470296;
s12 += s22 * 654183;
s13 -= s22 * 997805;
s14 += s22 * 136657;
s15 -= s22 * 683901;
s22 = 0;
s9 += s21 * 666643;
s10 += s21 * 470296;
s11 += s21 * 654183;
s12 -= s21 * 997805;
s13 += s21 * 136657;
s14 -= s21 * 683901;
s21 = 0;
s8 += s20 * 666643;
s9 += s20 * 470296;
s10 += s20 * 654183;
s11 -= s20 * 997805;
s12 += s20 * 136657;
s13 -= s20 * 683901;
s20 = 0;
s7 += s19 * 666643;
s8 += s19 * 470296;
s9 += s19 * 654183;
s10 -= s19 * 997805;
s11 += s19 * 136657;
s12 -= s19 * 683901;
s19 = 0;
s6 += s18 * 666643;
s7 += s18 * 470296;
s8 += s18 * 654183;
s9 -= s18 * 997805;
s10 += s18 * 136657;
s11 -= s18 * 683901;
s18 = 0;
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;
s17 = 0;
s4 += s16 * 666643;
s5 += s16 * 470296;
s6 += s16 * 654183;
s7 -= s16 * 997805;
s8 += s16 * 136657;
s9 -= s16 * 683901;
s16 = 0;
s3 += s15 * 666643;
s4 += s15 * 470296;
s5 += s15 * 654183;
s6 -= s15 * 997805;
s7 += s15 * 136657;
s8 -= s15 * 683901;
s15 = 0;
s2 += s14 * 666643;
s3 += s14 * 470296;
s4 += s14 * 654183;
s5 -= s14 * 997805;
s6 += s14 * 136657;
s7 -= s14 * 683901;
s14 = 0;
s1 += s13 * 666643;
s2 += s13 * 470296;
s3 += s13 * 654183;
s4 -= s13 * 997805;
s5 += s13 * 136657;
s6 -= s13 * 683901;
s13 = 0;
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;
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;
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;
}

273
sc_reduce.c Normal file
View File

@ -0,0 +1,273 @@
#include "sc.h"
#include "pstdint.h"
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;
}
/*
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;
s23 = 0;
s10 += s22 * 666643;
s11 += s22 * 470296;
s12 += s22 * 654183;
s13 -= s22 * 997805;
s14 += s22 * 136657;
s15 -= s22 * 683901;
s22 = 0;
s9 += s21 * 666643;
s10 += s21 * 470296;
s11 += s21 * 654183;
s12 -= s21 * 997805;
s13 += s21 * 136657;
s14 -= s21 * 683901;
s21 = 0;
s8 += s20 * 666643;
s9 += s20 * 470296;
s10 += s20 * 654183;
s11 -= s20 * 997805;
s12 += s20 * 136657;
s13 -= s20 * 683901;
s20 = 0;
s7 += s19 * 666643;
s8 += s19 * 470296;
s9 += s19 * 654183;
s10 -= s19 * 997805;
s11 += s19 * 136657;
s12 -= s19 * 683901;
s19 = 0;
s6 += s18 * 666643;
s7 += s18 * 470296;
s8 += s18 * 654183;
s9 -= s18 * 997805;
s10 += s18 * 136657;
s11 -= s18 * 683901;
s18 = 0;
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;
s17 = 0;
s4 += s16 * 666643;
s5 += s16 * 470296;
s6 += s16 * 654183;
s7 -= s16 * 997805;
s8 += s16 * 136657;
s9 -= s16 * 683901;
s16 = 0;
s3 += s15 * 666643;
s4 += s15 * 470296;
s5 += s15 * 654183;
s6 -= s15 * 997805;
s7 += s15 * 136657;
s8 -= s15 * 683901;
s15 = 0;
s2 += s14 * 666643;
s3 += s14 * 470296;
s4 += s14 * 654183;
s5 -= s14 * 997805;
s6 += s14 * 136657;
s7 -= s14 * 683901;
s14 = 0;
s1 += s13 * 666643;
s2 += s13 * 470296;
s3 += s13 * 654183;
s4 -= s13 * 997805;
s5 += s13 * 136657;
s6 -= s13 * 683901;
s13 = 0;
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;
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;
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;
}

302
sha512.c Normal file
View File

@ -0,0 +1,302 @@
/*
20080913
D. J. Bernstein
Public domain.
*/
#include "pstdint.h"
static int64_t load_bigendian(const unsigned char *x)
{
return
(int64_t) (x[7]) \
| (((int64_t) (x[6])) << 8) \
| (((int64_t) (x[5])) << 16) \
| (((int64_t) (x[4])) << 24) \
| (((int64_t) (x[3])) << 32) \
| (((int64_t) (x[2])) << 40) \
| (((int64_t) (x[1])) << 48) \
| (((int64_t) (x[0])) << 56)
;
}
static void store_bigendian(unsigned char *x,int64_t u)
{
x[7] = u; u >>= 8;
x[6] = u; u >>= 8;
x[5] = u; u >>= 8;
x[4] = u; u >>= 8;
x[3] = u; u >>= 8;
x[2] = u; u >>= 8;
x[1] = u; u >>= 8;
x[0] = u;
}
#define SHR(x,c) ((x) >> (c))
#define ROTR(x,c) (((x) >> (c)) | ((x) << (64 - (c))))
#define Ch(x,y,z) ((x & y) ^ (~x & z))
#define Maj(x,y,z) ((x & y) ^ (x & z) ^ (y & z))
#define Sigma0(x) (ROTR(x,28) ^ ROTR(x,34) ^ ROTR(x,39))
#define Sigma1(x) (ROTR(x,14) ^ ROTR(x,18) ^ ROTR(x,41))
#define sigma0(x) (ROTR(x, 1) ^ ROTR(x, 8) ^ SHR(x,7))
#define sigma1(x) (ROTR(x,19) ^ ROTR(x,61) ^ SHR(x,6))
#define M(w0,w14,w9,w1) w0 = sigma1(w14) + w9 + sigma0(w1) + w0;
#define EXPAND \
M(w0 ,w14,w9 ,w1 ) \
M(w1 ,w15,w10,w2 ) \
M(w2 ,w0 ,w11,w3 ) \
M(w3 ,w1 ,w12,w4 ) \
M(w4 ,w2 ,w13,w5 ) \
M(w5 ,w3 ,w14,w6 ) \
M(w6 ,w4 ,w15,w7 ) \
M(w7 ,w5 ,w0 ,w8 ) \
M(w8 ,w6 ,w1 ,w9 ) \
M(w9 ,w7 ,w2 ,w10) \
M(w10,w8 ,w3 ,w11) \
M(w11,w9 ,w4 ,w12) \
M(w12,w10,w5 ,w13) \
M(w13,w11,w6 ,w14) \
M(w14,w12,w7 ,w15) \
M(w15,w13,w8 ,w0 )
#define F(w,k) \
T1 = h + Sigma1(e) + Ch(e,f,g) + k + w; \
T2 = Sigma0(a) + Maj(a,b,c); \
h = g; \
g = f; \
f = e; \
e = d + T1; \
d = c; \
c = b; \
b = a; \
a = T1 + T2;
int sha512_blocks(unsigned char *statebytes,const unsigned char *in,unsigned long long inlen)
{
int64_t state[8];
int64_t a;
int64_t b;
int64_t c;
int64_t d;
int64_t e;
int64_t f;
int64_t g;
int64_t h;
int64_t T1;
int64_t T2;
a = load_bigendian(statebytes + 0); state[0] = a;
b = load_bigendian(statebytes + 8); state[1] = b;
c = load_bigendian(statebytes + 16); state[2] = c;
d = load_bigendian(statebytes + 24); state[3] = d;
e = load_bigendian(statebytes + 32); state[4] = e;
f = load_bigendian(statebytes + 40); state[5] = f;
g = load_bigendian(statebytes + 48); state[6] = g;
h = load_bigendian(statebytes + 56); state[7] = h;
while (inlen >= 128) {
int64_t w0 = load_bigendian(in + 0);
int64_t w1 = load_bigendian(in + 8);
int64_t w2 = load_bigendian(in + 16);
int64_t w3 = load_bigendian(in + 24);
int64_t w4 = load_bigendian(in + 32);
int64_t w5 = load_bigendian(in + 40);
int64_t w6 = load_bigendian(in + 48);
int64_t w7 = load_bigendian(in + 56);
int64_t w8 = load_bigendian(in + 64);
int64_t w9 = load_bigendian(in + 72);
int64_t w10 = load_bigendian(in + 80);
int64_t w11 = load_bigendian(in + 88);
int64_t w12 = load_bigendian(in + 96);
int64_t w13 = load_bigendian(in + 104);
int64_t w14 = load_bigendian(in + 112);
int64_t w15 = load_bigendian(in + 120);
F(w0 ,0x428a2f98d728ae22ULL)
F(w1 ,0x7137449123ef65cdULL)
F(w2 ,0xb5c0fbcfec4d3b2fULL)
F(w3 ,0xe9b5dba58189dbbcULL)
F(w4 ,0x3956c25bf348b538ULL)
F(w5 ,0x59f111f1b605d019ULL)
F(w6 ,0x923f82a4af194f9bULL)
F(w7 ,0xab1c5ed5da6d8118ULL)
F(w8 ,0xd807aa98a3030242ULL)
F(w9 ,0x12835b0145706fbeULL)
F(w10,0x243185be4ee4b28cULL)
F(w11,0x550c7dc3d5ffb4e2ULL)
F(w12,0x72be5d74f27b896fULL)
F(w13,0x80deb1fe3b1696b1ULL)
F(w14,0x9bdc06a725c71235ULL)
F(w15,0xc19bf174cf692694ULL)
EXPAND
F(w0 ,0xe49b69c19ef14ad2ULL)
F(w1 ,0xefbe4786384f25e3ULL)
F(w2 ,0x0fc19dc68b8cd5b5ULL)
F(w3 ,0x240ca1cc77ac9c65ULL)
F(w4 ,0x2de92c6f592b0275ULL)
F(w5 ,0x4a7484aa6ea6e483ULL)
F(w6 ,0x5cb0a9dcbd41fbd4ULL)
F(w7 ,0x76f988da831153b5ULL)
F(w8 ,0x983e5152ee66dfabULL)
F(w9 ,0xa831c66d2db43210ULL)
F(w10,0xb00327c898fb213fULL)
F(w11,0xbf597fc7beef0ee4ULL)
F(w12,0xc6e00bf33da88fc2ULL)
F(w13,0xd5a79147930aa725ULL)
F(w14,0x06ca6351e003826fULL)
F(w15,0x142929670a0e6e70ULL)
EXPAND
F(w0 ,0x27b70a8546d22ffcULL)
F(w1 ,0x2e1b21385c26c926ULL)
F(w2 ,0x4d2c6dfc5ac42aedULL)
F(w3 ,0x53380d139d95b3dfULL)
F(w4 ,0x650a73548baf63deULL)
F(w5 ,0x766a0abb3c77b2a8ULL)
F(w6 ,0x81c2c92e47edaee6ULL)
F(w7 ,0x92722c851482353bULL)
F(w8 ,0xa2bfe8a14cf10364ULL)
F(w9 ,0xa81a664bbc423001ULL)
F(w10,0xc24b8b70d0f89791ULL)
F(w11,0xc76c51a30654be30ULL)
F(w12,0xd192e819d6ef5218ULL)
F(w13,0xd69906245565a910ULL)
F(w14,0xf40e35855771202aULL)
F(w15,0x106aa07032bbd1b8ULL)
EXPAND
F(w0 ,0x19a4c116b8d2d0c8ULL)
F(w1 ,0x1e376c085141ab53ULL)
F(w2 ,0x2748774cdf8eeb99ULL)
F(w3 ,0x34b0bcb5e19b48a8ULL)
F(w4 ,0x391c0cb3c5c95a63ULL)
F(w5 ,0x4ed8aa4ae3418acbULL)
F(w6 ,0x5b9cca4f7763e373ULL)
F(w7 ,0x682e6ff3d6b2b8a3ULL)
F(w8 ,0x748f82ee5defb2fcULL)
F(w9 ,0x78a5636f43172f60ULL)
F(w10,0x84c87814a1f0ab72ULL)
F(w11,0x8cc702081a6439ecULL)
F(w12,0x90befffa23631e28ULL)
F(w13,0xa4506cebde82bde9ULL)
F(w14,0xbef9a3f7b2c67915ULL)
F(w15,0xc67178f2e372532bULL)
EXPAND
F(w0 ,0xca273eceea26619cULL)
F(w1 ,0xd186b8c721c0c207ULL)
F(w2 ,0xeada7dd6cde0eb1eULL)
F(w3 ,0xf57d4f7fee6ed178ULL)
F(w4 ,0x06f067aa72176fbaULL)
F(w5 ,0x0a637dc5a2c898a6ULL)
F(w6 ,0x113f9804bef90daeULL)
F(w7 ,0x1b710b35131c471bULL)
F(w8 ,0x28db77f523047d84ULL)
F(w9 ,0x32caab7b40c72493ULL)
F(w10,0x3c9ebe0a15c9bebcULL)
F(w11,0x431d67c49c100d4cULL)
F(w12,0x4cc5d4becb3e42b6ULL)
F(w13,0x597f299cfc657e2aULL)
F(w14,0x5fcb6fab3ad6faecULL)
F(w15,0x6c44198c4a475817ULL)
a += state[0];
b += state[1];
c += state[2];
d += state[3];
e += state[4];
f += state[5];
g += state[6];
h += state[7];
state[0] = a;
state[1] = b;
state[2] = c;
state[3] = d;
state[4] = e;
state[5] = f;
state[6] = g;
state[7] = h;
in += 128;
inlen -= 128;
}
store_bigendian(statebytes + 0,state[0]);
store_bigendian(statebytes + 8,state[1]);
store_bigendian(statebytes + 16,state[2]);
store_bigendian(statebytes + 24,state[3]);
store_bigendian(statebytes + 32,state[4]);
store_bigendian(statebytes + 40,state[5]);
store_bigendian(statebytes + 48,state[6]);
store_bigendian(statebytes + 56,state[7]);
return inlen;
}
static const unsigned char iv[64] = {
0x6a,0x09,0xe6,0x67,0xf3,0xbc,0xc9,0x08,
0xbb,0x67,0xae,0x85,0x84,0xca,0xa7,0x3b,
0x3c,0x6e,0xf3,0x72,0xfe,0x94,0xf8,0x2b,
0xa5,0x4f,0xf5,0x3a,0x5f,0x1d,0x36,0xf1,
0x51,0x0e,0x52,0x7f,0xad,0xe6,0x82,0xd1,
0x9b,0x05,0x68,0x8c,0x2b,0x3e,0x6c,0x1f,
0x1f,0x83,0xd9,0xab,0xfb,0x41,0xbd,0x6b,
0x5b,0xe0,0xcd,0x19,0x13,0x7e,0x21,0x79
} ;
int sha512(unsigned char *out,const unsigned char *in,unsigned long long inlen)
{
unsigned char h[64];
unsigned char padded[256];
int i;
unsigned long long bytes = inlen;
for (i = 0;i < 64;++i) h[i] = iv[i];
sha512_blocks(h,in,inlen);
in += inlen;
inlen &= 127;
in -= inlen;
for (i = 0;i < inlen;++i) padded[i] = in[i];
padded[inlen] = 0x80;
if (inlen < 112) {
for (i = inlen + 1;i < 119;++i) padded[i] = 0;
padded[119] = bytes >> 61;
padded[120] = bytes >> 53;
padded[121] = bytes >> 45;
padded[122] = bytes >> 37;
padded[123] = bytes >> 29;
padded[124] = bytes >> 21;
padded[125] = bytes >> 13;
padded[126] = bytes >> 5;
padded[127] = bytes << 3;
sha512_blocks(h,padded,128);
} else {
for (i = inlen + 1;i < 247;++i) padded[i] = 0;
padded[247] = bytes >> 61;
padded[248] = bytes >> 53;
padded[249] = bytes >> 45;
padded[250] = bytes >> 37;
padded[251] = bytes >> 29;
padded[252] = bytes >> 21;
padded[253] = bytes >> 13;
padded[254] = bytes >> 5;
padded[255] = bytes << 3;
sha512_blocks(h,padded,256);
}
for (i = 0;i < 64;++i) out[i] = h[i];
return 0;
}

6
sha512.h Normal file
View File

@ -0,0 +1,6 @@
#ifndef SHA512_H
#define SHA512_H
int sha512(unsigned char *out,const unsigned char *in,unsigned long long inlen);
#endif

38
sign.c Normal file
View File

@ -0,0 +1,38 @@
#include "crypto_sign.h"
#include "sha512.h"
#include "ge.h"
#include "sc.h"
int crypto_sign(
unsigned char *sm,unsigned long long *smlen,
const unsigned char *m,unsigned long long mlen,
const unsigned char *sk
)
{
unsigned char az[64];
unsigned char r[64];
unsigned char hram[64];
ge_p3 R;
unsigned long long i;
sha512(az,sk,32);
az[0] &= 248;
az[31] &= 63;
az[31] |= 64;
*smlen = mlen + 64;
for (i = 0;i < mlen;++i) sm[64 + i] = m[i];
for (i = 0;i < 32;++i) sm[32 + i] = az[32 + i];
sha512(r,sm + 32,mlen + 32);
for (i = 0;i < 32;++i) sm[32 + i] = sk[32 + i];
sc_reduce(r);
ge_scalarmult_base(&R,r);
ge_p3_tobytes(sm,&R);
sha512(hram,sm,mlen + 64);
sc_reduce(hram);
sc_muladd(sm + 32,hram,az,r);
return 0;
}

39
test.c Normal file
View File

@ -0,0 +1,39 @@
#include <stdlib.h>
#include <stdio.h>
#include <string.h>
#include "crypto_sign.h"
char *msg = "Hello World";
int main(int argc, char *argv[]) {
unsigned char sk[64], vk[32];
unsigned char *sigmsg, *newmsg;
unsigned long long sigmsglen, newmsglen;
int ret;
crypto_sign_keypair(vk, sk, "0123456890123456789012");
printf("got keypair\n");
sigmsg = malloc(strlen(msg)+1+64);
if (!sigmsg)
return 1;
crypto_sign(sigmsg, &sigmsglen, (unsigned char *)msg, strlen(msg)+1, sk);
printf("got signature\n");
if (sigmsglen != strlen(msg)+1+64)
return 2;
newmsg = malloc(sigmsglen);
if (!newmsg)
return 3;
ret = crypto_sign_open(newmsg, &newmsglen, sigmsg, sigmsglen, vk);
printf("verified signature\n");
if (ret == 0)
printf("good!\n");
else
printf("bad\n");
sigmsg[0] ^= 0x01;
ret = crypto_sign_open(newmsg, &newmsglen, sigmsg, sigmsglen, vk);
if (ret == 0)
printf("bad: failed to detect simple corruption\n");
else
printf("good: detected simple corruption\n");
return 0;
}

38
verify.c Normal file
View File

@ -0,0 +1,38 @@
int crypto_verify_32(const unsigned char *x,const unsigned char *y)
{
unsigned int differentbits = 0;
#define F(i) differentbits |= x[i] ^ y[i];
F(0)
F(1)
F(2)
F(3)
F(4)
F(5)
F(6)
F(7)
F(8)
F(9)
F(10)
F(11)
F(12)
F(13)
F(14)
F(15)
F(16)
F(17)
F(18)
F(19)
F(20)
F(21)
F(22)
F(23)
F(24)
F(25)
F(26)
F(27)
F(28)
F(29)
F(30)
F(31)
return (1 & ((differentbits - 1) >> 8)) - 1;
}

1
verify.h Normal file
View File

@ -0,0 +1 @@
int crypto_verify_32(const unsigned char *x,const unsigned char *y);