ZeroTierOne/node/SHA512.cpp

// Code taken from NaCl by D. J. Bernstein and others
// Public domain

/*
20080913
D. J. Bernstein
Public domain.
*/

#include <stdint.h>
#include <stdlib.h>
#include <string.h>

#include "SHA512.hpp"
#include "Utils.hpp"

#ifdef __APPLE__
#include <CommonCrypto/CommonDigest.h>
#define ZT_HAVE_NATIVE_SHA512
namespace ZeroTier {
void SHA512(void *digest,const void *data,unsigned int len)
{
	CC_SHA512_CTX ctx;
	CC_SHA512_Init(&ctx);
	CC_SHA512_Update(&ctx,data,len);
	CC_SHA512_Final(reinterpret_cast<unsigned char *>(digest),&ctx);
}
void SHA384(void *digest,const void *data,unsigned int len)
{
	CC_SHA512_CTX ctx;
	CC_SHA384_Init(&ctx);
	CC_SHA384_Update(&ctx,data,len);
	CC_SHA384_Final(reinterpret_cast<unsigned char *>(digest),&ctx);
}
}
#endif

#ifdef ZT_USE_LIBCRYPTO
#include <openssl/sha.h>
#define ZT_HAVE_NATIVE_SHA512
namespace ZeroTier {
void SHA512(void *digest,const void *data,unsigned int len)
{
	SHA512_CTX ctx;
	SHA512_Init(&ctx);
	SHA512_Update(&ctx,data,len);
	SHA512_Final(reinterpret_cast<unsigned char *>(digest),&ctx);
}
void SHA384(void *digest,const void *data,unsigned int len)
{
	SHA512_CTX ctx;
	SHA384_Init(&ctx);
	SHA384_Update(&ctx,data,len);
	SHA384_Final(reinterpret_cast<unsigned char *>(digest),&ctx);
}
}
#endif

// If a platform-native SHA512 isn't available we use this 64-bit C version.
#ifndef ZT_HAVE_NATIVE_SHA512

namespace ZeroTier {

namespace {

static inline void sha512_encode(uint64_t input, uint8_t *output, uint32_t idx)
{
	output[idx + 0] = (uint8_t)(input >> 56);
	output[idx + 1] = (uint8_t)(input >> 48);
	output[idx + 2] = (uint8_t)(input >> 40);
	output[idx + 3] = (uint8_t)(input >> 32);
	output[idx + 4] = (uint8_t)(input >> 24);
	output[idx + 5] = (uint8_t)(input >> 16);
	output[idx + 6] = (uint8_t)(input >>  8);
	output[idx + 7] = (uint8_t)(input >>  0);
}
static inline void sha512_decode(uint64_t *output, uint8_t *input, uint32_t idx)
{
	*output = ((uint64_t)input[idx + 0] << 56)
		| ((uint64_t)input[idx + 1] << 48)
		| ((uint64_t)input[idx + 2] << 40)
		| ((uint64_t)input[idx + 3] << 32)
		| ((uint64_t)input[idx + 4] << 24)
		| ((uint64_t)input[idx + 5] << 16)
		| ((uint64_t)input[idx + 6] <<  8)
		| ((uint64_t)input[idx + 7] <<  0);
}

typedef struct sha512_ctx_tag {
	uint32_t is_sha384;
	uint8_t block[128];
	uint64_t len[2];
	uint64_t val[8];
	uint8_t *payload_addr;
	uint64_t payload_len;
} sha512_ctx_t;

#define LSR(x,n) (x >> n)
#define ROR(x,n) (LSR(x,n) | (x << (64 - n)))

#define MA(x,y,z) ((x & y) | (z & (x | y)))
#define CH(x,y,z) (z ^ (x & (y ^ z)))
#define GAMMA0(x) (ROR(x, 1) ^ ROR(x, 8) ^  LSR(x, 7))
#define GAMMA1(x) (ROR(x,19) ^ ROR(x,61) ^  LSR(x, 6))
#define SIGMA0(x) (ROR(x,28) ^ ROR(x,34) ^ ROR(x,39))
#define SIGMA1(x) (ROR(x,14) ^ ROR(x,18) ^ ROR(x,41))

#define INIT_COMPRESSOR() uint64_t tmp0 = 0, tmp1 = 0
#define COMPRESS( a,  b,  c, d,  e,  f,  g,  h, x,  k)   \
    tmp0 = h + SIGMA1(e) + CH(e,f,g) + k + x;              \
    tmp1 = SIGMA0(a) + MA(a,b,c); d += tmp0; h = tmp0 + tmp1;

static const uint8_t sha512_padding[128] = { 0x80, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0 };

static const uint64_t K[80] = {
	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
};

static inline void sha512_memcpy(uint8_t *src, uint8_t *dst, uint32_t size)
{
	uint32_t i = 0;
	for (;i < size;i++) { *dst++ = *src++; }
}
static inline void sha512_memclr(uint8_t *dst, uint32_t size)
{
	uint32_t i = 0;
	for (;i < size;i++) { *dst++ = 0; }
}

static inline void sha512_init_512(sha512_ctx_t *sha512_ctx, uint8_t *payload_addr, uint64_t payload_len)
{
	sha512_memclr((uint8_t *)sha512_ctx,sizeof(sha512_ctx_t));
	sha512_ctx->val[0] = 0x6A09E667F3BCC908ULL;
	sha512_ctx->val[1] = 0xBB67AE8584CAA73BULL;
	sha512_ctx->val[2] = 0x3C6EF372FE94F82BULL;
	sha512_ctx->val[3] = 0xA54FF53A5F1D36F1ULL;
	sha512_ctx->val[4] = 0x510E527FADE682D1ULL;
	sha512_ctx->val[5] = 0x9B05688C2B3E6C1FULL;
	sha512_ctx->val[6] = 0x1F83D9ABFB41BD6BULL;
	sha512_ctx->val[7] = 0x5BE0CD19137E2179ULL;
	sha512_ctx->is_sha384 = 0;
	sha512_ctx->payload_addr = payload_addr;
	sha512_ctx->payload_len = (uint64_t)payload_len;
	sha512_ctx->len[0] = payload_len << 3;
	sha512_ctx->len[1] = payload_len >> 61;
}

static inline void sha512_init_384(sha512_ctx_t *sha512_ctx, uint8_t *payload_addr, uint64_t payload_len)
{
	sha512_memclr((uint8_t *)sha512_ctx,sizeof(sha512_ctx_t));
	sha512_ctx->val[0] = 0xCBBB9D5DC1059ED8ULL;
	sha512_ctx->val[1] = 0x629A292A367CD507ULL;
	sha512_ctx->val[2] = 0x9159015A3070DD17ULL;
	sha512_ctx->val[3] = 0x152FECD8F70E5939ULL;
	sha512_ctx->val[4] = 0x67332667FFC00B31ULL;
	sha512_ctx->val[5] = 0x8EB44A8768581511ULL;
	sha512_ctx->val[6] = 0xDB0C2E0D64F98FA7ULL;
	sha512_ctx->val[7] = 0x47B5481DBEFA4FA4ULL;
	sha512_ctx->is_sha384 = 1;
	sha512_ctx->payload_addr = payload_addr;
	sha512_ctx->payload_len = (uint64_t)payload_len;
	sha512_ctx->len[0] = payload_len << 3;
	sha512_ctx->len[1] = payload_len >> 61;
}

static inline void sha512_hash_factory(sha512_ctx_t *ctx, uint8_t data[128])
{
	uint32_t i = 0;
	uint64_t W[80];
	uint64_t v[8];
	INIT_COMPRESSOR();
	for(i = 0; i < 16; i++) { sha512_decode(&W[i], data, i << 3 ); }
	for(; i < 80; i++) { W[i] = GAMMA1(W[i -  2]) + W[i -  7] + GAMMA0(W[i - 15]) + W[i - 16]; }
	for (i = 0;i < 8; i++) { v[i] = ctx->val[i]; }
	for(i = 0; i < 80;) {
		COMPRESS(v[0], v[1], v[2], v[3], v[4], v[5], v[6], v[7], W[i], K[i] ); i++;
		COMPRESS(v[7], v[0], v[1], v[2], v[3], v[4], v[5], v[6], W[i], K[i] ); i++;
		COMPRESS(v[6], v[7], v[0], v[1], v[2], v[3], v[4], v[5], W[i], K[i] ); i++;
		COMPRESS(v[5], v[6], v[7], v[0], v[1], v[2], v[3], v[4], W[i], K[i] ); i++;
		COMPRESS(v[4], v[5], v[6], v[7], v[0], v[1], v[2], v[3], W[i], K[i] ); i++;
		COMPRESS(v[3], v[4], v[5], v[6], v[7], v[0], v[1], v[2], W[i], K[i] ); i++;
		COMPRESS(v[2], v[3], v[4], v[5], v[6], v[7], v[0], v[1], W[i], K[i] ); i++;
		COMPRESS(v[1], v[2], v[3], v[4], v[5], v[6], v[7], v[0], W[i], K[i] ); i++;
	}
	for (i = 0; i < 8; i++) { ctx->val[i] += v[i]; }
}

static inline void sha512_stage1(sha512_ctx_t *sha512_ctx)
{
	while (sha512_ctx->payload_len >= 128) {
		sha512_hash_factory(sha512_ctx, sha512_ctx->payload_addr);
		sha512_ctx->payload_addr += 128;
		sha512_ctx->payload_len -= 128;
	}
}

static inline void sha512_stage2(sha512_ctx_t *sha512_ctx, uint8_t output[64])
{
	uint32_t block_pos = sha512_ctx->payload_len;
	uint32_t padding_bytes = 0;
	uint8_t temp_data[128] = {0};
	uint8_t *temp_data_p = (uint8_t *)&temp_data[0];
	uint8_t len_be[16] = {0};
	uint8_t i = 0;
	sha512_memcpy(sha512_ctx->payload_addr, temp_data_p, sha512_ctx->payload_len);
	padding_bytes = 112 - block_pos;
	temp_data_p += block_pos;
	sha512_memcpy((uint8_t *)sha512_padding, temp_data_p, padding_bytes);
	temp_data_p += padding_bytes;
	sha512_encode(sha512_ctx->len[1], len_be, 0);
	sha512_encode(sha512_ctx->len[0], len_be, 8);
	sha512_memcpy(len_be, temp_data_p, 16);
	sha512_hash_factory(sha512_ctx, temp_data);
	for (i = 0; i < 6; i++) { sha512_encode(sha512_ctx->val[i], output, i * 8); }
	for ( ;(i < 8) && (sha512_ctx->is_sha384 == 0); i++) { sha512_encode(sha512_ctx->val[i], output, i * 8); }
}

} // anonymous namespace

void SHA512(void *digest,const void *data,unsigned int len)
{
	sha512_ctx_t h;
	sha512_init_512(&h,(uint8_t *)data,len);
	sha512_stage1(&h);
	sha512_stage2(&h,(uint8_t *)digest);
}

void SHA384(void *digest,const void *data,unsigned int len)
{
	sha512_ctx_t h;
	sha512_init_384(&h,(uint8_t *)data,len);
	sha512_stage1(&h);
	sha512_stage2(&h,(uint8_t *)digest);
}

} // namespace ZeroTier

#endif // !ZT_HAVE_NATIVE_SHA512

extern "C" void ZT_sha512internal(void *digest,const void *data,unsigned int len) { ZeroTier::SHA512(digest,data,len); }
extern "C" void ZT_sha384internal(void *digest,const void *data,unsigned int len) { ZeroTier::SHA384(digest,data,len); }