diff --git a/sshsh512.c b/sshsh512.c index c063997e..1c3ae3da 100644 --- a/sshsh512.c +++ b/sshsh512.c @@ -19,49 +19,71 @@ typedef struct { BinarySink_IMPLEMENTATION; } SHA512_State; -/* - * Arithmetic implementations. Note that AND, XOR and NOT can - * overlap destination with one source, but the others can't. - */ -#define add(r,x,y) ( r = (x) + (y) ) -#define rorB(r,x,y) ( r = ((x) >> (y)) | ((x) << (64-(y))) ) -#define rorL(r,x,y) ( r = ((x) >> (y)) | ((x) << (64-(y))) ) -#define shrB(r,x,y) ( r = (x) >> (y) ) -#define shrL(r,x,y) ( r = (x) >> (y) ) -#define and(r,x,y) ( r = (x) & (y) ) -#define xor(r,x,y) ( r = (x) ^ (y) ) -#define not(r,x) ( r = ~(x) ) -#define INIT(h,l) ((((uint64_t)(h)) << 32) | (l)) -#define BUILD(r,h,l) ( r = ((((uint64_t)(h)) << 32) | (l)) ) -#define EXTRACT(h,l,r) ( h = (r) >> 32, l = (r) & 0xFFFFFFFFU ) - /* ---------------------------------------------------------------------- * Core SHA512 algorithm: processes 16-doubleword blocks into a * message digest. */ -#define Ch(r,t,x,y,z) ( not(t,x), and(r,t,z), and(t,x,y), xor(r,r,t) ) -#define Maj(r,t,x,y,z) ( and(r,x,y), and(t,x,z), xor(r,r,t), \ - and(t,y,z), xor(r,r,t) ) -#define bigsigma0(r,t,x) ( rorL(r,x,28), rorB(t,x,34), xor(r,r,t), \ - rorB(t,x,39), xor(r,r,t) ) -#define bigsigma1(r,t,x) ( rorL(r,x,14), rorL(t,x,18), xor(r,r,t), \ - rorB(t,x,41), xor(r,r,t) ) -#define smallsigma0(r,t,x) ( rorL(r,x,1), rorL(t,x,8), xor(r,r,t), \ - shrL(t,x,7), xor(r,r,t) ) -#define smallsigma1(r,t,x) ( rorL(r,x,19), rorB(t,x,61), xor(r,r,t), \ - shrL(t,x,6), xor(r,r,t) ) +static inline uint64_t ror(uint64_t x, unsigned y) +{ + return (x << (63 & -y)) | (x >> (63 & y)); +} + +static inline uint64_t Ch(uint64_t ctrl, uint64_t if1, uint64_t if0) +{ + return if0 ^ (ctrl & (if1 ^ if0)); +} + +static inline uint64_t Maj(uint64_t x, uint64_t y, uint64_t z) +{ + return (x & y) | (z & (x | y)); +} + +static inline uint64_t Sigma_0(uint64_t x) +{ + return ror(x,28) ^ ror(x,34) ^ ror(x,39); +} + +static inline uint64_t Sigma_1(uint64_t x) +{ + return ror(x,14) ^ ror(x,18) ^ ror(x,41); +} + +static inline uint64_t sigma_0(uint64_t x) +{ + return ror(x,1) ^ ror(x,8) ^ (x >> 7); +} + +static inline uint64_t sigma_1(uint64_t x) +{ + return ror(x,19) ^ ror(x,61) ^ (x >> 6); +} + +static inline void SHA512_Round( + unsigned round_index, const uint64_t *round_constants, + const uint64_t *schedule, + uint64_t *a, uint64_t *b, uint64_t *c, uint64_t *d, + uint64_t *e, uint64_t *f, uint64_t *g, uint64_t *h) +{ + uint64_t t1 = *h + Sigma_1(*e) + Ch(*e,*f,*g) + + round_constants[round_index] + schedule[round_index]; + + uint64_t t2 = Sigma_0(*a) + Maj(*a,*b,*c); + + *d += t1; + *h = t1 + t2; +} static void SHA512_Core_Init(SHA512_State *s) { static const uint64_t iv[] = { - INIT(0x6a09e667, 0xf3bcc908), - INIT(0xbb67ae85, 0x84caa73b), - INIT(0x3c6ef372, 0xfe94f82b), - INIT(0xa54ff53a, 0x5f1d36f1), - INIT(0x510e527f, 0xade682d1), - INIT(0x9b05688c, 0x2b3e6c1f), - INIT(0x1f83d9ab, 0xfb41bd6b), - INIT(0x5be0cd19, 0x137e2179), + 0x6a09e667f3bcc908ULL, + 0xbb67ae8584caa73bULL, + 0x3c6ef372fe94f82bULL, + 0xa54ff53a5f1d36f1ULL, + 0x510e527fade682d1ULL, + 0x9b05688c2b3e6c1fULL, + 0x1f83d9abfb41bd6bULL, + 0x5be0cd19137e2179ULL, }; int i; for (i = 0; i < 8; i++) @@ -70,14 +92,14 @@ static void SHA512_Core_Init(SHA512_State *s) { static void SHA384_Core_Init(SHA512_State *s) { static const uint64_t iv[] = { - INIT(0xcbbb9d5d, 0xc1059ed8), - INIT(0x629a292a, 0x367cd507), - INIT(0x9159015a, 0x3070dd17), - INIT(0x152fecd8, 0xf70e5939), - INIT(0x67332667, 0xffc00b31), - INIT(0x8eb44a87, 0x68581511), - INIT(0xdb0c2e0d, 0x64f98fa7), - INIT(0x47b5481d, 0xbefa4fa4), + 0xcbbb9d5dc1059ed8ULL, + 0x629a292a367cd507ULL, + 0x9159015a3070dd17ULL, + 0x152fecd8f70e5939ULL, + 0x67332667ffc00b31ULL, + 0x8eb44a8768581511ULL, + 0xdb0c2e0d64f98fa7ULL, + 0x47b5481dbefa4fa4ULL, }; int i; for (i = 0; i < 8; i++) @@ -88,46 +110,46 @@ static void SHA512_Block(SHA512_State *s, uint64_t *block) { uint64_t w[80]; uint64_t a,b,c,d,e,f,g,h; static const uint64_t k[] = { - INIT(0x428a2f98, 0xd728ae22), INIT(0x71374491, 0x23ef65cd), - INIT(0xb5c0fbcf, 0xec4d3b2f), INIT(0xe9b5dba5, 0x8189dbbc), - INIT(0x3956c25b, 0xf348b538), INIT(0x59f111f1, 0xb605d019), - INIT(0x923f82a4, 0xaf194f9b), INIT(0xab1c5ed5, 0xda6d8118), - INIT(0xd807aa98, 0xa3030242), INIT(0x12835b01, 0x45706fbe), - INIT(0x243185be, 0x4ee4b28c), INIT(0x550c7dc3, 0xd5ffb4e2), - INIT(0x72be5d74, 0xf27b896f), INIT(0x80deb1fe, 0x3b1696b1), - INIT(0x9bdc06a7, 0x25c71235), INIT(0xc19bf174, 0xcf692694), - INIT(0xe49b69c1, 0x9ef14ad2), INIT(0xefbe4786, 0x384f25e3), - INIT(0x0fc19dc6, 0x8b8cd5b5), INIT(0x240ca1cc, 0x77ac9c65), - INIT(0x2de92c6f, 0x592b0275), INIT(0x4a7484aa, 0x6ea6e483), - INIT(0x5cb0a9dc, 0xbd41fbd4), INIT(0x76f988da, 0x831153b5), - INIT(0x983e5152, 0xee66dfab), INIT(0xa831c66d, 0x2db43210), - INIT(0xb00327c8, 0x98fb213f), INIT(0xbf597fc7, 0xbeef0ee4), - INIT(0xc6e00bf3, 0x3da88fc2), INIT(0xd5a79147, 0x930aa725), - INIT(0x06ca6351, 0xe003826f), INIT(0x14292967, 0x0a0e6e70), - INIT(0x27b70a85, 0x46d22ffc), INIT(0x2e1b2138, 0x5c26c926), - INIT(0x4d2c6dfc, 0x5ac42aed), INIT(0x53380d13, 0x9d95b3df), - INIT(0x650a7354, 0x8baf63de), INIT(0x766a0abb, 0x3c77b2a8), - INIT(0x81c2c92e, 0x47edaee6), INIT(0x92722c85, 0x1482353b), - INIT(0xa2bfe8a1, 0x4cf10364), INIT(0xa81a664b, 0xbc423001), - INIT(0xc24b8b70, 0xd0f89791), INIT(0xc76c51a3, 0x0654be30), - INIT(0xd192e819, 0xd6ef5218), INIT(0xd6990624, 0x5565a910), - INIT(0xf40e3585, 0x5771202a), INIT(0x106aa070, 0x32bbd1b8), - INIT(0x19a4c116, 0xb8d2d0c8), INIT(0x1e376c08, 0x5141ab53), - INIT(0x2748774c, 0xdf8eeb99), INIT(0x34b0bcb5, 0xe19b48a8), - INIT(0x391c0cb3, 0xc5c95a63), INIT(0x4ed8aa4a, 0xe3418acb), - INIT(0x5b9cca4f, 0x7763e373), INIT(0x682e6ff3, 0xd6b2b8a3), - INIT(0x748f82ee, 0x5defb2fc), INIT(0x78a5636f, 0x43172f60), - INIT(0x84c87814, 0xa1f0ab72), INIT(0x8cc70208, 0x1a6439ec), - INIT(0x90befffa, 0x23631e28), INIT(0xa4506ceb, 0xde82bde9), - INIT(0xbef9a3f7, 0xb2c67915), INIT(0xc67178f2, 0xe372532b), - INIT(0xca273ece, 0xea26619c), INIT(0xd186b8c7, 0x21c0c207), - INIT(0xeada7dd6, 0xcde0eb1e), INIT(0xf57d4f7f, 0xee6ed178), - INIT(0x06f067aa, 0x72176fba), INIT(0x0a637dc5, 0xa2c898a6), - INIT(0x113f9804, 0xbef90dae), INIT(0x1b710b35, 0x131c471b), - INIT(0x28db77f5, 0x23047d84), INIT(0x32caab7b, 0x40c72493), - INIT(0x3c9ebe0a, 0x15c9bebc), INIT(0x431d67c4, 0x9c100d4c), - INIT(0x4cc5d4be, 0xcb3e42b6), INIT(0x597f299c, 0xfc657e2a), - INIT(0x5fcb6fab, 0x3ad6faec), INIT(0x6c44198c, 0x4a475817), + 0x428a2f98d728ae22ULL, 0x7137449123ef65cdULL, + 0xb5c0fbcfec4d3b2fULL, 0xe9b5dba58189dbbcULL, + 0x3956c25bf348b538ULL, 0x59f111f1b605d019ULL, + 0x923f82a4af194f9bULL, 0xab1c5ed5da6d8118ULL, + 0xd807aa98a3030242ULL, 0x12835b0145706fbeULL, + 0x243185be4ee4b28cULL, 0x550c7dc3d5ffb4e2ULL, + 0x72be5d74f27b896fULL, 0x80deb1fe3b1696b1ULL, + 0x9bdc06a725c71235ULL, 0xc19bf174cf692694ULL, + 0xe49b69c19ef14ad2ULL, 0xefbe4786384f25e3ULL, + 0x0fc19dc68b8cd5b5ULL, 0x240ca1cc77ac9c65ULL, + 0x2de92c6f592b0275ULL, 0x4a7484aa6ea6e483ULL, + 0x5cb0a9dcbd41fbd4ULL, 0x76f988da831153b5ULL, + 0x983e5152ee66dfabULL, 0xa831c66d2db43210ULL, + 0xb00327c898fb213fULL, 0xbf597fc7beef0ee4ULL, + 0xc6e00bf33da88fc2ULL, 0xd5a79147930aa725ULL, + 0x06ca6351e003826fULL, 0x142929670a0e6e70ULL, + 0x27b70a8546d22ffcULL, 0x2e1b21385c26c926ULL, + 0x4d2c6dfc5ac42aedULL, 0x53380d139d95b3dfULL, + 0x650a73548baf63deULL, 0x766a0abb3c77b2a8ULL, + 0x81c2c92e47edaee6ULL, 0x92722c851482353bULL, + 0xa2bfe8a14cf10364ULL, 0xa81a664bbc423001ULL, + 0xc24b8b70d0f89791ULL, 0xc76c51a30654be30ULL, + 0xd192e819d6ef5218ULL, 0xd69906245565a910ULL, + 0xf40e35855771202aULL, 0x106aa07032bbd1b8ULL, + 0x19a4c116b8d2d0c8ULL, 0x1e376c085141ab53ULL, + 0x2748774cdf8eeb99ULL, 0x34b0bcb5e19b48a8ULL, + 0x391c0cb3c5c95a63ULL, 0x4ed8aa4ae3418acbULL, + 0x5b9cca4f7763e373ULL, 0x682e6ff3d6b2b8a3ULL, + 0x748f82ee5defb2fcULL, 0x78a5636f43172f60ULL, + 0x84c87814a1f0ab72ULL, 0x8cc702081a6439ecULL, + 0x90befffa23631e28ULL, 0xa4506cebde82bde9ULL, + 0xbef9a3f7b2c67915ULL, 0xc67178f2e372532bULL, + 0xca273eceea26619cULL, 0xd186b8c721c0c207ULL, + 0xeada7dd6cde0eb1eULL, 0xf57d4f7fee6ed178ULL, + 0x06f067aa72176fbaULL, 0x0a637dc5a2c898a6ULL, + 0x113f9804bef90daeULL, 0x1b710b35131c471bULL, + 0x28db77f523047d84ULL, 0x32caab7b40c72493ULL, + 0x3c9ebe0a15c9bebcULL, 0x431d67c49c100d4cULL, + 0x4cc5d4becb3e42b6ULL, 0x597f299cfc657e2aULL, + 0x5fcb6fab3ad6faecULL, 0x6c44198c4a475817ULL, }; int t; @@ -135,54 +157,25 @@ static void SHA512_Block(SHA512_State *s, uint64_t *block) { for (t = 0; t < 16; t++) w[t] = block[t]; - for (t = 16; t < 80; t++) { - uint64_t p, q, r, tmp; - smallsigma1(p, tmp, w[t-2]); - smallsigma0(q, tmp, w[t-15]); - add(r, p, q); - add(p, r, w[t-7]); - add(w[t], p, w[t-16]); - } + for (t = 16; t < 80; t++) + w[t] = w[t-16] + w[t-7] + sigma_0(w[t-15]) + sigma_1(w[t-2]); a = s->h[0]; b = s->h[1]; c = s->h[2]; d = s->h[3]; e = s->h[4]; f = s->h[5]; g = s->h[6]; h = s->h[7]; for (t = 0; t < 80; t+=8) { - uint64_t tmp, p, q, r; - -#define ROUND(j,a,b,c,d,e,f,g,h) do { \ - bigsigma1(p, tmp, e); \ - Ch(q, tmp, e, f, g); \ - add(r, p, q); \ - add(p, r, k[j]) ; \ - add(q, p, w[j]); \ - add(r, q, h); \ - bigsigma0(p, tmp, a); \ - Maj(tmp, q, a, b, c); \ - add(q, tmp, p); \ - add(p, r, d); \ - d = p; \ - add(h, q, r); \ - } while (0) - - ROUND(t+0, a,b,c,d,e,f,g,h); - ROUND(t+1, h,a,b,c,d,e,f,g); - ROUND(t+2, g,h,a,b,c,d,e,f); - ROUND(t+3, f,g,h,a,b,c,d,e); - ROUND(t+4, e,f,g,h,a,b,c,d); - ROUND(t+5, d,e,f,g,h,a,b,c); - ROUND(t+6, c,d,e,f,g,h,a,b); - ROUND(t+7, b,c,d,e,f,g,h,a); + SHA512_Round(t+0, k,w, &a,&b,&c,&d,&e,&f,&g,&h); + SHA512_Round(t+1, k,w, &h,&a,&b,&c,&d,&e,&f,&g); + SHA512_Round(t+2, k,w, &g,&h,&a,&b,&c,&d,&e,&f); + SHA512_Round(t+3, k,w, &f,&g,&h,&a,&b,&c,&d,&e); + SHA512_Round(t+4, k,w, &e,&f,&g,&h,&a,&b,&c,&d); + SHA512_Round(t+5, k,w, &d,&e,&f,&g,&h,&a,&b,&c); + SHA512_Round(t+6, k,w, &c,&d,&e,&f,&g,&h,&a,&b); + SHA512_Round(t+7, k,w, &b,&c,&d,&e,&f,&g,&h,&a); } - { - uint64_t tmp; -#define UPDATE(state, local) ( tmp = state, add(state, tmp, local) ) - UPDATE(s->h[0], a); UPDATE(s->h[1], b); - UPDATE(s->h[2], c); UPDATE(s->h[3], d); - UPDATE(s->h[4], e); UPDATE(s->h[5], f); - UPDATE(s->h[6], g); UPDATE(s->h[7], h); - } + s->h[0] += a; s->h[1] += b; s->h[2] += c; s->h[3] += d; + s->h[4] += e; s->h[5] += f; s->h[6] += g; s->h[7] += h; } /* ----------------------------------------------------------------------