/* * Based on Michael Hamburg's STROBE reference implementation. * Copyright (c) 2015-2016 Cryptography Research, Inc. * MIT License (MIT) */ #if defined(__GNUC__) && defined(__SIZEOF_INT128__) # define hydro_x25519_WBITS 64 #else # define hydro_x25519_WBITS 32 #endif #if hydro_x25519_WBITS == 64 typedef uint64_t hydro_x25519_limb_t; typedef __uint128_t hydro_x25519_dlimb_t; typedef __int128_t hydro_x25519_sdlimb_t; # define hydro_x25519_eswap_limb(X) LOAD64_LE((const uint8_t *) &(X)) # define hydro_x25519_LIMB(x) x##ull #elif hydro_x25519_WBITS == 32 typedef uint32_t hydro_x25519_limb_t; typedef uint64_t hydro_x25519_dlimb_t; typedef int64_t hydro_x25519_sdlimb_t; # define hydro_x25519_eswap_limb(X) LOAD32_LE((const uint8_t *) &(X)) # define hydro_x25519_LIMB(x) (uint32_t)(x##ull), (uint32_t)((x##ull) >> 32) #else # error "Need to know hydro_x25519_WBITS" #endif #define hydro_x25519_NLIMBS (256 / hydro_x25519_WBITS) typedef hydro_x25519_limb_t hydro_x25519_fe[hydro_x25519_NLIMBS]; typedef hydro_x25519_limb_t hydro_x25519_scalar_t[hydro_x25519_NLIMBS]; static const hydro_x25519_limb_t hydro_x25519_MONTGOMERY_FACTOR = (hydro_x25519_limb_t) 0xd2b51da312547e1bull; static const hydro_x25519_scalar_t hydro_x25519_sc_p = { hydro_x25519_LIMB(0x5812631a5cf5d3ed), hydro_x25519_LIMB(0x14def9dea2f79cd6), hydro_x25519_LIMB(0x0000000000000000), hydro_x25519_LIMB(0x1000000000000000) }; static const hydro_x25519_scalar_t hydro_x25519_sc_r2 = { hydro_x25519_LIMB(0xa40611e3449c0f01), hydro_x25519_LIMB(0xd00e1ba768859347), hydro_x25519_LIMB(0xceec73d217f5be65), hydro_x25519_LIMB(0x0399411b7c309a3d) }; static const uint8_t hydro_x25519_BASE_POINT[hydro_x25519_BYTES] = { 9 }; static const hydro_x25519_limb_t hydro_x25519_a24[1] = { 121665 }; static inline hydro_x25519_limb_t hydro_x25519_umaal(hydro_x25519_limb_t *carry, hydro_x25519_limb_t acc, hydro_x25519_limb_t mand, hydro_x25519_limb_t mier) { hydro_x25519_dlimb_t tmp = (hydro_x25519_dlimb_t) mand * mier + acc + *carry; *carry = tmp >> hydro_x25519_WBITS; return (hydro_x25519_limb_t) tmp; } static inline hydro_x25519_limb_t hydro_x25519_adc(hydro_x25519_limb_t *carry, hydro_x25519_limb_t acc, hydro_x25519_limb_t mand) { hydro_x25519_dlimb_t total = (hydro_x25519_dlimb_t) *carry + acc + mand; *carry = total >> hydro_x25519_WBITS; return (hydro_x25519_limb_t) total; } static inline hydro_x25519_limb_t hydro_x25519_adc0(hydro_x25519_limb_t *carry, hydro_x25519_limb_t acc) { hydro_x25519_dlimb_t total = (hydro_x25519_dlimb_t) *carry + acc; *carry = total >> hydro_x25519_WBITS; return (hydro_x25519_limb_t) total; } static void hydro_x25519_propagate(hydro_x25519_fe x, hydro_x25519_limb_t over) { hydro_x25519_limb_t carry; int i; over = x[hydro_x25519_NLIMBS - 1] >> (hydro_x25519_WBITS - 1) | over << 1; x[hydro_x25519_NLIMBS - 1] &= ~((hydro_x25519_limb_t) 1 << (hydro_x25519_WBITS - 1)); carry = over * 19; for (i = 0; i < hydro_x25519_NLIMBS; i++) { x[i] = hydro_x25519_adc0(&carry, x[i]); } } static void hydro_x25519_add(hydro_x25519_fe out, const hydro_x25519_fe a, const hydro_x25519_fe b) { hydro_x25519_limb_t carry = 0; int i; for (i = 0; i < hydro_x25519_NLIMBS; i++) { out[i] = hydro_x25519_adc(&carry, a[i], b[i]); } hydro_x25519_propagate(out, carry); } static void hydro_x25519_sub(hydro_x25519_fe out, const hydro_x25519_fe a, const hydro_x25519_fe b) { hydro_x25519_sdlimb_t carry = -38; int i; for (i = 0; i < hydro_x25519_NLIMBS; i++) { out[i] = carry = carry + a[i] - b[i]; carry >>= hydro_x25519_WBITS; } hydro_x25519_propagate(out, 1 + carry); } static void hydro_x25519_swapin(hydro_x25519_limb_t *x, const uint8_t *in) { int i; memcpy(x, in, sizeof(hydro_x25519_fe)); for (i = 0; i < hydro_x25519_NLIMBS; i++) { x[i] = hydro_x25519_eswap_limb(x[i]); } } static void hydro_x25519_swapout(uint8_t *out, hydro_x25519_limb_t *x) { int i; for (i = 0; i < hydro_x25519_NLIMBS; i++) { x[i] = hydro_x25519_eswap_limb(x[i]); } memcpy(out, x, sizeof(hydro_x25519_fe)); } static void hydro_x25519_mul(hydro_x25519_fe out, const hydro_x25519_fe a, const hydro_x25519_fe b, int nb) { hydro_x25519_limb_t accum[2 * hydro_x25519_NLIMBS] = { 0 }; hydro_x25519_limb_t carry2; int i, j; for (i = 0; i < nb; i++) { carry2 = 0; hydro_x25519_limb_t mand = b[i]; for (j = 0; j < hydro_x25519_NLIMBS; j++) { accum[i + j] = hydro_x25519_umaal(&carry2, accum[i + j], mand, a[j]); } accum[i + j] = carry2; } carry2 = 0; for (j = 0; j < hydro_x25519_NLIMBS; j++) { const hydro_x25519_limb_t mand = 38; out[j] = hydro_x25519_umaal(&carry2, accum[j], mand, accum[j + hydro_x25519_NLIMBS]); } hydro_x25519_propagate(out, carry2); } static void hydro_x25519_sqr(hydro_x25519_fe out, const hydro_x25519_fe a) { hydro_x25519_mul(out, a, a, hydro_x25519_NLIMBS); } static void hydro_x25519_mul1(hydro_x25519_fe out, const hydro_x25519_fe a) { hydro_x25519_mul(out, a, out, hydro_x25519_NLIMBS); } static void hydro_x25519_sqr1(hydro_x25519_fe a) { hydro_x25519_mul1(a, a); } static void hydro_x25519_condswap(hydro_x25519_limb_t a[2 * hydro_x25519_NLIMBS], hydro_x25519_limb_t b[2 * hydro_x25519_NLIMBS], hydro_x25519_limb_t doswap) { int i; for (i = 0; i < 2 * hydro_x25519_NLIMBS; i++) { hydro_x25519_limb_t xorv = (a[i] ^ b[i]) & doswap; a[i] ^= xorv; b[i] ^= xorv; } } static int hydro_x25519_canon(hydro_x25519_fe x) { hydro_x25519_sdlimb_t carry; hydro_x25519_limb_t carry0 = 19; hydro_x25519_limb_t res; int i; for (i = 0; i < hydro_x25519_NLIMBS; i++) { x[i] = hydro_x25519_adc0(&carry0, x[i]); } hydro_x25519_propagate(x, carry0); carry = -19; res = 0; for (i = 0; i < hydro_x25519_NLIMBS; i++) { res |= x[i] = carry += x[i]; carry >>= hydro_x25519_WBITS; } return ((hydro_x25519_dlimb_t) res - 1) >> hydro_x25519_WBITS; } static void hydro_x25519_ladder_part1(hydro_x25519_fe xs[5]) { hydro_x25519_limb_t *x2 = xs[0], *z2 = xs[1], *x3 = xs[2], *z3 = xs[3], *t1 = xs[4]; hydro_x25519_add(t1, x2, z2); // t1 = A hydro_x25519_sub(z2, x2, z2); // z2 = B hydro_x25519_add(x2, x3, z3); // x2 = C hydro_x25519_sub(z3, x3, z3); // z3 = D hydro_x25519_mul1(z3, t1); // z3 = DA hydro_x25519_mul1(x2, z2); // x3 = BC hydro_x25519_add(x3, z3, x2); // x3 = DA+CB hydro_x25519_sub(z3, z3, x2); // z3 = DA-CB hydro_x25519_sqr1(t1); // t1 = AA hydro_x25519_sqr1(z2); // z2 = BB hydro_x25519_sub(x2, t1, z2); // x2 = E = AA-BB hydro_x25519_mul(z2, x2, hydro_x25519_a24, // z2 = E*a24 sizeof(hydro_x25519_a24) / sizeof(hydro_x25519_a24[0])); hydro_x25519_add(z2, z2, t1); // z2 = E*a24 + AA } static void hydro_x25519_ladder_part2(hydro_x25519_fe xs[5], const hydro_x25519_fe x1) { hydro_x25519_limb_t *x2 = xs[0], *z2 = xs[1], *x3 = xs[2], *z3 = xs[3], *t1 = xs[4]; hydro_x25519_sqr1(z3); // z3 = (DA-CB)^2 hydro_x25519_mul1(z3, x1); // z3 = x1 * (DA-CB)^2 hydro_x25519_sqr1(x3); // x3 = (DA+CB)^2 hydro_x25519_mul1(z2, x2); // z2 = AA*(E*a24+AA) hydro_x25519_sub(x2, t1, x2); // x2 = BB again hydro_x25519_mul1(x2, t1); // x2 = AA*BB } static void hydro_x25519_core(hydro_x25519_fe xs[5], const uint8_t scalar[hydro_x25519_BYTES], const uint8_t *x1, bool clamp) { hydro_x25519_limb_t swap; hydro_x25519_limb_t *x2 = xs[0], *x3 = xs[2], *z3 = xs[3]; hydro_x25519_fe x1i; int i; hydro_x25519_swapin(x1i, x1); x1 = (const uint8_t *) x1i; swap = 0; mem_zero(xs, 4 * sizeof(hydro_x25519_fe)); x2[0] = z3[0] = 1; memcpy(x3, x1, sizeof(hydro_x25519_fe)); for (i = 255; i >= 0; i--) { uint8_t bytei = scalar[i / 8]; hydro_x25519_limb_t doswap; hydro_x25519_fe x1_dup; if (clamp) { if (i / 8 == 0) { bytei &= ~7; } else if (i / 8 == hydro_x25519_BYTES - 1) { bytei &= 0x7F; bytei |= 0x40; } } doswap = 1U + ~(hydro_x25519_limb_t)((bytei >> (i % 8)) & 1); hydro_x25519_condswap(x2, x3, swap ^ doswap); swap = doswap; hydro_x25519_ladder_part1(xs); memcpy(x1_dup, x1, sizeof x1_dup); hydro_x25519_ladder_part2(xs, x1_dup); } hydro_x25519_condswap(x2, x3, swap); } static int hydro_x25519_scalarmult(uint8_t out[hydro_x25519_BYTES], const uint8_t scalar[hydro_x25519_BYTES], const uint8_t x1[hydro_x25519_BYTES], bool clamp) { hydro_x25519_fe xs[5]; hydro_x25519_limb_t *x2, *z2, *z3; hydro_x25519_limb_t *prev; int i; int ret; hydro_x25519_core(xs, scalar, x1, clamp); /* Precomputed inversion chain */ x2 = xs[0]; z2 = xs[1]; z3 = xs[3]; prev = z2; /* Raise to the p-2 = 0x7f..ffeb */ for (i = 253; i >= 0; i--) { hydro_x25519_sqr(z3, prev); prev = z3; if (i >= 8 || (0xeb >> i & 1)) { hydro_x25519_mul1(z3, z2); } } /* Here prev = z3 */ /* x2 /= z2 */ hydro_x25519_mul1(x2, z3); ret = hydro_x25519_canon(x2); hydro_x25519_swapout(out, x2); if (clamp == 0) { return 0; } return ret; } static inline int hydro_x25519_scalarmult_base(uint8_t pk[hydro_x25519_PUBLICKEYBYTES], const uint8_t sk[hydro_x25519_SECRETKEYBYTES]) { return hydro_x25519_scalarmult(pk, sk, hydro_x25519_BASE_POINT, 1); } static inline void hydro_x25519_scalarmult_base_uniform(uint8_t pk[hydro_x25519_PUBLICKEYBYTES], const uint8_t sk[hydro_x25519_SECRETKEYBYTES]) { if (hydro_x25519_scalarmult(pk, sk, hydro_x25519_BASE_POINT, 0) != 0) { abort(); } } static void hydro_x25519_sc_montmul(hydro_x25519_scalar_t out, const hydro_x25519_scalar_t a, const hydro_x25519_scalar_t b) { hydro_x25519_limb_t hic = 0; int i, j; for (i = 0; i < hydro_x25519_NLIMBS; i++) { hydro_x25519_limb_t carry = 0, carry2 = 0, mand = a[i], mand2 = hydro_x25519_MONTGOMERY_FACTOR; for (j = 0; j < hydro_x25519_NLIMBS; j++) { hydro_x25519_limb_t acc = out[j]; acc = hydro_x25519_umaal(&carry, acc, mand, b[j]); if (j == 0) { mand2 *= acc; } acc = hydro_x25519_umaal(&carry2, acc, mand2, hydro_x25519_sc_p[j]); if (j > 0) { out[j - 1] = acc; } } /* Add two carry registers and high carry */ out[hydro_x25519_NLIMBS - 1] = hydro_x25519_adc(&hic, carry, carry2); } /* Reduce */ hydro_x25519_sdlimb_t scarry = 0; for (i = 0; i < hydro_x25519_NLIMBS; i++) { out[i] = scarry = scarry + out[i] - hydro_x25519_sc_p[i]; scarry >>= hydro_x25519_WBITS; } hydro_x25519_limb_t need_add = -(scarry + hic); hydro_x25519_limb_t carry = 0; for (i = 0; i < hydro_x25519_NLIMBS; i++) { out[i] = hydro_x25519_umaal(&carry, out[i], need_add, hydro_x25519_sc_p[i]); } }