encrypt.c

#include <stdio.h>
#include "api.h"
#include  <string.h>
#include <stdlib.h>
#include <emmintrin.h>//sse2 header file(include sse header file)  
#define U64BIG(x) (x)
#define U32BIG(x) (x)

#define PR0_ROUNDS 100
#define PR_ROUNDS 52
#define PRF_ROUNDS 56


typedef unsigned char u8;
typedef unsigned long long u64;
typedef long long i64;

#define forward_sbox_SSE(a, b, c, d,  f, g, h) \
{                   \
tmm1    =_mm_xor_si128(  a , all1 );	\
tmm2 =_mm_and_si128(  b, tmm1    );	\
tmm3 =_mm_xor_si128(  c  , tmm2 );	\
h=_mm_xor_si128(  d  , tmm3 );	\
tmm5 =_mm_or_si128 (  b, c);	\
tmm6=_mm_xor_si128(  d, tmm1 );	\
g=_mm_xor_si128(  tmm5, tmm6 );	\
tmm8=_mm_xor_si128(  b, d );	\
tmm9=_mm_and_si128(  tmm3, tmm6 );	\
a=_mm_xor_si128(  tmm8, tmm9 );	\
tmm11=_mm_and_si128(  g, tmm8 );	\
f=_mm_xor_si128(  tmm3, tmm11 );	\
}

u8 constant7[100] = { 0x01, 0x02, 0x04, 0x08, 0x10, 0x20, 0x41,
	   0x03, 0x06, 0x0c, 0x18, 0x30, 0x61, 0x42, 0x05, 0x0a, 0x14, 0x28, 0x51,
	   0x23, 0x47, 0x0f, 0x1e, 0x3c, 0x79, 0x72, 0x64, 0x48, 0x11, 0x22, 0x45,
	   0x0b, 0x16, 0x2c, 0x59, 0x33, 0x67, 0x4e, 0x1d, 0x3a, 0x75, 0x6a, 0x54,
	   0x29, 0x53, 0x27, 0x4f, 0x1f, 0x3e, 0x7d, 0x7a, 0x74, 0x68, 0x50, 0x21,
	   0x43, 0x07, 0x0e, 0x1c, 0x38, 0x71, 0x62, 0x44, 0x09, 0x12, 0x24, 0x49,
	   0x13, 0x26, 0x4d, 0x1b, 0x36, 0x6d, 0x5a, 0x35, 0x6b, 0x56, 0x2d, 0x5b,
	   0x37, 0x6f, 0x5e, 0x3d, 0x7b, 0x76, 0x6c, 0x58, 0x31, 0x63, 0x46, 0x0d,
	   0x1a, 0x34, 0x69, 0x52, 0x25, 0x4b, 0x17, 0x2e, 0x5d };
#define sbox(a, b, c, d, f, g, h)                                                                            \
{                                                                                                                             \
	t1 = ~a; t2 = b & t1;t3 = c ^ t2; h = d ^ t3; t5 = b | c; t6 = d ^ t1; g = t5 ^ t6; t8 = b ^ d; t9 = t3 & t6; a = t8 ^ t9; t11 = g & t8; f = t3 ^ t11; \
}

#define LOTR1281(a,b,n) (((a)<<(n))|((b)>>(64-n)))
#define LOTR1282(a,b,n) (((b)<<(n))|((a)>>(64-n)))

#define U64BIG(x) (x)

#define RATE 16

#define PR0_ROUNDS 100
#define PR_ROUNDS 52
#define PRF_ROUNDS 56

#define ROUND512(i) {\
state[0] = _mm_xor_si128(state[0], _mm_set_epi64x(0, (u64)constant7[i]));	\
forward_sbox_SSE(state[0], state[1], state[2], state[3], out1, out2, out3); \
state[1] = _mm_or_si128(_mm_slli_epi64(out1, 1), _mm_srli_epi64(_mm_shuffle_epi32(out1, _MM_SHUFFLE(1, 0, 3, 2)), 63));    \
state[2] = _mm_or_si128(_mm_slli_epi64(out2, 16), _mm_srli_epi64(_mm_shuffle_epi32(out2, _MM_SHUFFLE(1, 0, 3, 2)), 48));    \
state[3] = _mm_or_si128(_mm_slli_epi64(out3, 25), _mm_srli_epi64(_mm_shuffle_epi32(out3, _MM_SHUFFLE(1, 0, 3, 2)), 39));    \
}
int crypto_aead_encrypt(unsigned char *c, unsigned long long *clen,
	const unsigned char *m, unsigned long long mlen,
	const unsigned char *ad, unsigned long long adlen,
	const unsigned char *nsec, const unsigned char *npub,
	const unsigned char *k) {
	*clen = mlen + CRYPTO_ABYTES;
	__m128i all1 = _mm_set1_epi32(0xffffffff);
	//__m128i tmm0, tmm1, tmm2, tmm3, tmm4, tmm5, tmm6, tmm7, tmm8, tmm9, tmm10, tmm11, out1, out2, out3;
	__m128i  tmm1, tmm2, tmm3, tmm5, tmm6, tmm8, tmm9,  tmm11, out1, out2, out3;
	__m128i state[4];

	u64  i;
	u8 tempData[16] = { 0 };
	// initialization
	state[0] = _mm_loadu_si128((__m128i*)(npub));
	state[1] = _mm_loadu_si128((__m128i*)(npub + 16));
	state[2] = _mm_loadu_si128((__m128i*)(k));
	state[3] = _mm_loadu_si128((__m128i*)(k + 16));
	for (i = 0; i < PR0_ROUNDS; i++) {
		ROUND512(i);
	}
	// process associated data
	if (adlen) {
		while (adlen >= RATE) {
			state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(ad)));
			for (i = 0; i < PR_ROUNDS; i++) {
				ROUND512(i);
			}
			adlen -= RATE;
			ad += RATE;
		}
		memset(tempData, 0, RATE);
		memcpy(tempData, ad, adlen);
		tempData[adlen] = 0x01;
		state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(tempData)));
		for (i = 0; i < PR_ROUNDS; i++) {
			ROUND512(i);
		}
	}
	state[3] = _mm_xor_si128(state[3], _mm_set_epi64x((u64)0x8000000000000000, 0));
	// process plaintext
	if (mlen) {
		while (mlen >= RATE) {
			state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(m)));
			memcpy(c, state, RATE);
			for (i = 0; i < PR_ROUNDS; i++) {
				ROUND512(i);
			}
			mlen -= RATE;
			m += RATE;
			c += RATE;
		}
		memset(tempData, 0, RATE);
		memcpy(tempData, m, mlen);
		tempData[mlen] = 0x01;
		state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(tempData)));
		memcpy(c, state, mlen);
		c += mlen;
	}
	// finalization
	for (i = 0; i < PRF_ROUNDS; i++) {
		ROUND512(i);
	}
	// return tag
	memcpy(c, state, sizeof(unsigned char) * CRYPTO_ABYTES);
	return 0;
}
int crypto_aead_decrypt(unsigned char *m, unsigned long long *mlen,
	unsigned char *nsec, const unsigned char *c, unsigned long long clen,
	const unsigned char *ad, unsigned long long adlen,
	const unsigned char *npub, const unsigned char *k) {

	*mlen = clen - CRYPTO_ABYTES;
	if (clen < CRYPTO_ABYTES)
		return -1;
	__m128i all1 = _mm_set1_epi32(0xffffffff);
	__m128i  tmm1, tmm2, tmm3, tmm5, tmm6, tmm8, tmm9,  tmm11, out1, out2, out3;
	__m128i state[4];

	u64  i;
	u8 tempData[16] = { 0 };
	// initialization
	state[0] = _mm_loadu_si128((__m128i*)(npub));
	state[1] = _mm_loadu_si128((__m128i*)(npub + 16));
	state[2] = _mm_loadu_si128((__m128i*)(k));
	state[3] = _mm_loadu_si128((__m128i*)(k + 16));
	for (i = 0; i < PR0_ROUNDS; i++) {
		ROUND512(i);
	}
	// process associated data
	if (adlen) {
		while (adlen >= RATE) {
			state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(ad)));
			for (i = 0; i < PR_ROUNDS; i++) {
				ROUND512(i);
			}
			adlen -= RATE;
			ad += RATE;
		}
		memset(tempData, 0, RATE);
		memcpy(tempData, ad, adlen);
		tempData[adlen] = 0x01;
		state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(tempData)));
		for (i = 0; i < PR_ROUNDS; i++) {
			ROUND512(i);
		}
	}
	state[3] = _mm_xor_si128(state[3], _mm_set_epi64x((u64)0x8000000000000000, 0));
	clen -= CRYPTO_ABYTES;
	if (clen) {
		while (clen >= RATE) {
			state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(c)));
			memcpy(m, state, RATE);
			memcpy(state, c, RATE);
			for (i = 0; i < PR_ROUNDS; i++) {
				ROUND512(i);
			}
			clen -= RATE;
			m += RATE;
			c += RATE;
		}
		memset(tempData, 0, RATE);
		memcpy(tempData, c, clen);
		tempData[clen] = 0x01;
		state[0] = _mm_xor_si128(state[0], _mm_loadu_si128((__m128i*)(tempData)));
		memcpy(m, state, clen);
		memcpy(state, c, clen);
		c += clen;
	}
	// finalization
	for (i = 0; i < PRF_ROUNDS; i++) {
		ROUND512(i);
	}
	if (memcmp((void*)state, (void*)c, CRYPTO_ABYTES)) {
		memset(m, 0, sizeof(unsigned char) * (*mlen));
		*mlen = 0;
		return -1;
	}
	return 0;
}