ZeroTierOne/core/Salsa20.cpp

/*
 * Based on public domain code available at: http://cr.yp.to/snuffle.html
 *
 * Modifications and C-native SSE macro based SSE implementation by
 * Adam Ierymenko <[email protected]>.
 *
 * Since the original was public domain, this is too.
 */

#include "Constants.hpp"
#include "Salsa20.hpp"

#define ROTATE(v, c) (((v) << (c)) | ((v) >> (32 - (c))))
#define XOR(v, w) ((v) ^ (w))
#define PLUS(v, w) ((uint32_t)((v) + (w)))

#ifndef ZT_SALSA20_SSE
#if __BYTE_ORDER == __LITTLE_ENDIAN
#ifdef ZT_NO_UNALIGNED_ACCESS
// Slower version that does not use type punning
#define U8TO32_LITTLE(p) ( ((uint32_t)(p)[0]) | ((uint32_t)(p)[1] << 8) | ((uint32_t)(p)[2] << 16) | ((uint32_t)(p)[3] << 24) )
static ZT_INLINE void U32TO8_LITTLE(uint8_t *const c,const uint32_t v) { c[0] = (uint8_t)v; c[1] = (uint8_t)(v >> 8); c[2] = (uint8_t)(v >> 16); c[3] = (uint8_t)(v >> 24); }
#else
// Fast version that just does 32-bit load/store
#define U8TO32_LITTLE(p) (*((const uint32_t *)((const void *)(p))))
#define U32TO8_LITTLE(c,v) *((uint32_t *)((void *)(c))) = (v)
#endif // ZT_NO_UNALIGNED_ACCESS
#else // __BYTE_ORDER == __BIG_ENDIAN (we don't support anything else... does MIDDLE_ENDIAN even still exist?)
#ifdef __GNUC__
// Use GNUC builtin bswap macros on big-endian machines if available
#define U8TO32_LITTLE(p) __builtin_bswap32(*((const uint32_t *)((const void *)(p))))
#define U32TO8_LITTLE(c,v) *((uint32_t *)((void *)(c))) = __builtin_bswap32((v))
#else // no __GNUC__
// Otherwise do it the slow, manual way on BE machines
#define U8TO32_LITTLE(p) ( ((uint32_t)(p)[0]) | ((uint32_t)(p)[1] << 8) | ((uint32_t)(p)[2] << 16) | ((uint32_t)(p)[3] << 24) )
static ZT_INLINE void U32TO8_LITTLE(uint8_t *const c,const uint32_t v) { c[0] = (uint8_t)v; c[1] = (uint8_t)(v >> 8); c[2] = (uint8_t)(v >> 16); c[3] = (uint8_t)(v >> 24); }
#endif // __GNUC__ or not
#endif // __BYTE_ORDER little or big?
#endif // !ZT_SALSA20_SSE

#ifdef ZT_SALSA20_SSE
class _s20sseconsts
{
public:
	_s20sseconsts() noexcept
	{
		maskLo32 = _mm_shuffle_epi32(_mm_cvtsi32_si128(-1), _MM_SHUFFLE(1, 0, 1, 0));
		maskHi32 = _mm_slli_epi64(maskLo32, 32);
	}
	__m128i maskLo32, maskHi32;
};
static const _s20sseconsts s_S20SSECONSTANTS;
#endif

namespace ZeroTier {

void Salsa20::init(const void *key, const void *iv) noexcept
{
#ifdef ZT_SALSA20_SSE
	const uint32_t *const k = (const uint32_t *) key;
	_state.i[0] = 0x61707865;
	_state.i[1] = 0x3320646e;
	_state.i[2] = 0x79622d32;
	_state.i[3] = 0x6b206574;
	_state.i[4] = k[3];
	_state.i[5] = 0;
	_state.i[6] = k[7];
	_state.i[7] = k[2];
	_state.i[8] = 0;
	_state.i[9] = k[6];
	_state.i[10] = k[1];
	_state.i[11] = ((const uint32_t *) iv)[1];
	_state.i[12] = k[5];
	_state.i[13] = k[0];
	_state.i[14] = ((const uint32_t *) iv)[0];
	_state.i[15] = k[4];
#else
	const char *const constants = "expand 32-byte k";
	const uint8_t *const k = (const uint8_t *)key;
	_state.i[0] = U8TO32_LITTLE(constants + 0);
	_state.i[1] = U8TO32_LITTLE(k + 0);
	_state.i[2] = U8TO32_LITTLE(k + 4);
	_state.i[3] = U8TO32_LITTLE(k + 8);
	_state.i[4] = U8TO32_LITTLE(k + 12);
	_state.i[5] = U8TO32_LITTLE(constants + 4);
	_state.i[6] = U8TO32_LITTLE(((const uint8_t *)iv) + 0);
	_state.i[7] = U8TO32_LITTLE(((const uint8_t *)iv) + 4);
	_state.i[8] = 0;
	_state.i[9] = 0;
	_state.i[10] = U8TO32_LITTLE(constants + 8);
	_state.i[11] = U8TO32_LITTLE(k + 16);
	_state.i[12] = U8TO32_LITTLE(k + 20);
	_state.i[13] = U8TO32_LITTLE(k + 24);
	_state.i[14] = U8TO32_LITTLE(k + 28);
	_state.i[15] = U8TO32_LITTLE(constants + 12);
#endif
}

union p_SalsaState {
#ifdef ZT_SALSA20_SSE
	__m128i v[4];
#endif // ZT_SALSA20_SSE
	uint32_t i[16];
};

template<unsigned int R>
static ZT_INLINE void p_salsaCrypt(p_SalsaState *const state, const uint8_t *m, uint8_t *c, unsigned int bytes) noexcept
{
	if (unlikely(bytes == 0))
		return;

	uint8_t tmp[64];
	uint8_t *ctarget = c;

#ifdef ZT_SALSA20_SSE
	_mm_prefetch(m, _MM_HINT_T0);
	_mm_prefetch(m + 64, _MM_HINT_T0);
	__m128i X0 = state->v[0];
	__m128i X1 = state->v[1];
	__m128i X2 = state->v[2];
	__m128i X3 = state->v[3];
	const __m128i maskLo32 = s_S20SSECONSTANTS.maskLo32;
	const __m128i maskHi32 = s_S20SSECONSTANTS.maskHi32;
	const __m128i add1 = _mm_set_epi32(0, 0, 0, 1);
#else
	uint32_t x0, x1, x2, x3, x4, x5, x6, x7, x8, x9, x10, x11, x12, x13, x14, x15;
	uint32_t j0, j1, j2, j3, j4, j5, j6, j7, j8, j9, j10, j11, j12, j13, j14, j15;
	j0 = state->i[0];
	j1 = state->i[1];
	j2 = state->i[2];
	j3 = state->i[3];
	j4 = state->i[4];
	j5 = state->i[5];
	j6 = state->i[6];
	j7 = state->i[7];
	j8 = state->i[8];
	j9 = state->i[9];
	j10 = state->i[10];
	j11 = state->i[11];
	j12 = state->i[12];
	j13 = state->i[13];
	j14 = state->i[14];
	j15 = state->i[15];
#endif

	for (;;) {
		if (likely(bytes >= 64)) {
#ifdef ZT_SALSA20_SSE
			_mm_prefetch(m + 128, _MM_HINT_T0);
#endif
		} else {
			for (unsigned int i = 0;i < bytes;++i)
				tmp[i] = m[i];
			m = tmp;
			ctarget = c;
			c = tmp;
		}

#ifdef ZT_SALSA20_SSE
		__m128i X0s = X0;
		__m128i X1s = X1;
		__m128i X2s = X2;
		__m128i X3s = X3;
		__m128i T;

		for(unsigned int rr=0; rr<(R/2); ++rr) {
			T = _mm_add_epi32(X0, X3);
			X1 = _mm_xor_si128(_mm_xor_si128(X1, _mm_slli_epi32(T, 7)), _mm_srli_epi32(T, 25));
			T = _mm_add_epi32(X1, X0);
			X2 = _mm_xor_si128(_mm_xor_si128(X2, _mm_slli_epi32(T, 9)), _mm_srli_epi32(T, 23));
			T = _mm_add_epi32(X2, X1);
			X3 = _mm_xor_si128(_mm_xor_si128(X3, _mm_slli_epi32(T, 13)), _mm_srli_epi32(T, 19));
			T = _mm_add_epi32(X3, X2);
			X0 = _mm_xor_si128(_mm_xor_si128(X0, _mm_slli_epi32(T, 18)), _mm_srli_epi32(T, 14));
			X1 = _mm_shuffle_epi32(X1, 0x93);
			X2 = _mm_shuffle_epi32(X2, 0x4E);
			X3 = _mm_shuffle_epi32(X3, 0x39);
			T = _mm_add_epi32(X0, X1);
			X3 = _mm_xor_si128(_mm_xor_si128(X3, _mm_slli_epi32(T, 7)), _mm_srli_epi32(T, 25));
			T = _mm_add_epi32(X3, X0);
			X2 = _mm_xor_si128(_mm_xor_si128(X2, _mm_slli_epi32(T, 9)), _mm_srli_epi32(T, 23));
			T = _mm_add_epi32(X2, X3);
			X1 = _mm_xor_si128(_mm_xor_si128(X1, _mm_slli_epi32(T, 13)), _mm_srli_epi32(T, 19));
			T = _mm_add_epi32(X1, X2);
			X0 = _mm_xor_si128(_mm_xor_si128(X0, _mm_slli_epi32(T, 18)), _mm_srli_epi32(T, 14));
			X1 = _mm_shuffle_epi32(X1, 0x39);
			X2 = _mm_shuffle_epi32(X2, 0x4E);
			X3 = _mm_shuffle_epi32(X3, 0x93);
		}

		X0 = _mm_add_epi32(X0s, X0);
		X1 = _mm_add_epi32(X1s, X1);
		X2 = _mm_add_epi32(X2s, X2);
		X3 = _mm_add_epi32(X3s, X3);

		__m128i k02 = _mm_or_si128(_mm_slli_epi64(X0, 32), _mm_srli_epi64(X3, 32));
		__m128i k20 = _mm_or_si128(_mm_and_si128(X2, maskLo32), _mm_and_si128(X1, maskHi32));
		__m128i k13 = _mm_or_si128(_mm_slli_epi64(X1, 32), _mm_srli_epi64(X0, 32));
		__m128i k31 = _mm_or_si128(_mm_and_si128(X3, maskLo32), _mm_and_si128(X2, maskHi32));
		k02 = _mm_shuffle_epi32(k02, _MM_SHUFFLE(0, 1, 2, 3));
		k13 = _mm_shuffle_epi32(k13, _MM_SHUFFLE(0, 1, 2, 3));

		_mm_storeu_si128(reinterpret_cast<__m128i *>(c), _mm_xor_si128(_mm_unpackhi_epi64(k02, k20), _mm_loadu_si128(reinterpret_cast<const __m128i *>(m))));
		_mm_storeu_si128(reinterpret_cast<__m128i *>(c) + 1, _mm_xor_si128(_mm_unpackhi_epi64(k13, k31), _mm_loadu_si128(reinterpret_cast<const __m128i *>(m) + 1)));
		_mm_storeu_si128(reinterpret_cast<__m128i *>(c) + 2, _mm_xor_si128(_mm_unpacklo_epi64(k20, k02), _mm_loadu_si128(reinterpret_cast<const __m128i *>(m) + 2)));
		_mm_storeu_si128(reinterpret_cast<__m128i *>(c) + 3, _mm_xor_si128(_mm_unpacklo_epi64(k31, k13), _mm_loadu_si128(reinterpret_cast<const __m128i *>(m) + 3)));

		X0 = X0s;
		X1 = X1s;
		X2 = _mm_add_epi32(X2s, add1);
		X3 = X3s;

#else

		x0 = j0;
		x1 = j1;
		x2 = j2;
		x3 = j3;
		x4 = j4;
		x5 = j5;
		x6 = j6;
		x7 = j7;
		x8 = j8;
		x9 = j9;
		x10 = j10;
		x11 = j11;
		x12 = j12;
		x13 = j13;
		x14 = j14;
		x15 = j15;

		for(unsigned int rr=0;rr<(R/2);++rr) {
			 x4 = XOR( x4,ROTATE(PLUS( x0,x12), 7));
			 x8 = XOR( x8,ROTATE(PLUS( x4, x0), 9));
			x12 = XOR(x12,ROTATE(PLUS( x8, x4),13));
			 x0 = XOR( x0,ROTATE(PLUS(x12, x8),18));
			 x9 = XOR( x9,ROTATE(PLUS( x5, x1), 7));
			x13 = XOR(x13,ROTATE(PLUS( x9, x5), 9));
			 x1 = XOR( x1,ROTATE(PLUS(x13, x9),13));
			 x5 = XOR( x5,ROTATE(PLUS( x1,x13),18));
			x14 = XOR(x14,ROTATE(PLUS(x10, x6), 7));
			 x2 = XOR( x2,ROTATE(PLUS(x14,x10), 9));
			 x6 = XOR( x6,ROTATE(PLUS( x2,x14),13));
			x10 = XOR(x10,ROTATE(PLUS( x6, x2),18));
			 x3 = XOR( x3,ROTATE(PLUS(x15,x11), 7));
			 x7 = XOR( x7,ROTATE(PLUS( x3,x15), 9));
			x11 = XOR(x11,ROTATE(PLUS( x7, x3),13));
			x15 = XOR(x15,ROTATE(PLUS(x11, x7),18));
			 x1 = XOR( x1,ROTATE(PLUS( x0, x3), 7));
			 x2 = XOR( x2,ROTATE(PLUS( x1, x0), 9));
			 x3 = XOR( x3,ROTATE(PLUS( x2, x1),13));
			 x0 = XOR( x0,ROTATE(PLUS( x3, x2),18));
			 x6 = XOR( x6,ROTATE(PLUS( x5, x4), 7));
			 x7 = XOR( x7,ROTATE(PLUS( x6, x5), 9));
			 x4 = XOR( x4,ROTATE(PLUS( x7, x6),13));
			 x5 = XOR( x5,ROTATE(PLUS( x4, x7),18));
			x11 = XOR(x11,ROTATE(PLUS(x10, x9), 7));
			 x8 = XOR( x8,ROTATE(PLUS(x11,x10), 9));
			 x9 = XOR( x9,ROTATE(PLUS( x8,x11),13));
			x10 = XOR(x10,ROTATE(PLUS( x9, x8),18));
			x12 = XOR(x12,ROTATE(PLUS(x15,x14), 7));
			x13 = XOR(x13,ROTATE(PLUS(x12,x15), 9));
			x14 = XOR(x14,ROTATE(PLUS(x13,x12),13));
			x15 = XOR(x15,ROTATE(PLUS(x14,x13),18));
		}

		x0 = PLUS(x0,j0);
		x1 = PLUS(x1,j1);
		x2 = PLUS(x2,j2);
		x3 = PLUS(x3,j3);
		x4 = PLUS(x4,j4);
		x5 = PLUS(x5,j5);
		x6 = PLUS(x6,j6);
		x7 = PLUS(x7,j7);
		x8 = PLUS(x8,j8);
		x9 = PLUS(x9,j9);
		x10 = PLUS(x10,j10);
		x11 = PLUS(x11,j11);
		x12 = PLUS(x12,j12);
		x13 = PLUS(x13,j13);
		x14 = PLUS(x14,j14);
		x15 = PLUS(x15,j15);

		U32TO8_LITTLE(c + 0,XOR(x0,U8TO32_LITTLE(m + 0)));
		U32TO8_LITTLE(c + 4,XOR(x1,U8TO32_LITTLE(m + 4)));
		U32TO8_LITTLE(c + 8,XOR(x2,U8TO32_LITTLE(m + 8)));
		U32TO8_LITTLE(c + 12,XOR(x3,U8TO32_LITTLE(m + 12)));
		U32TO8_LITTLE(c + 16,XOR(x4,U8TO32_LITTLE(m + 16)));
		U32TO8_LITTLE(c + 20,XOR(x5,U8TO32_LITTLE(m + 20)));
		U32TO8_LITTLE(c + 24,XOR(x6,U8TO32_LITTLE(m + 24)));
		U32TO8_LITTLE(c + 28,XOR(x7,U8TO32_LITTLE(m + 28)));
		U32TO8_LITTLE(c + 32,XOR(x8,U8TO32_LITTLE(m + 32)));
		U32TO8_LITTLE(c + 36,XOR(x9,U8TO32_LITTLE(m + 36)));
		U32TO8_LITTLE(c + 40,XOR(x10,U8TO32_LITTLE(m + 40)));
		U32TO8_LITTLE(c + 44,XOR(x11,U8TO32_LITTLE(m + 44)));
		U32TO8_LITTLE(c + 48,XOR(x12,U8TO32_LITTLE(m + 48)));
		U32TO8_LITTLE(c + 52,XOR(x13,U8TO32_LITTLE(m + 52)));
		U32TO8_LITTLE(c + 56,XOR(x14,U8TO32_LITTLE(m + 56)));
		U32TO8_LITTLE(c + 60,XOR(x15,U8TO32_LITTLE(m + 60)));

		++j8;

#endif

		if (likely(bytes > 64)) {
			bytes -= 64;
			c += 64;
			m += 64;
		} else {
			if (bytes < 64) {
				for (unsigned int i = 0;i < bytes;++i)
					ctarget[i] = c[i];
			}
#ifdef ZT_SALSA20_SSE
			state->v[2] = X2;
#else
			state->i[8] = j8;
#endif
			return;
		}
	}
}

void Salsa20::crypt12(const void *in, void *out, unsigned int bytes) noexcept
{
	p_salsaCrypt<12>(reinterpret_cast<p_SalsaState *>(&_state), reinterpret_cast<const uint8_t *>(in), reinterpret_cast<uint8_t *>(out), bytes);
}

void Salsa20::crypt20(const void *in, void *out, unsigned int bytes) noexcept
{
	p_salsaCrypt<20>(reinterpret_cast<p_SalsaState *>(&_state), reinterpret_cast<const uint8_t *>(in), reinterpret_cast<uint8_t *>(out), bytes);
}

} // namespace ZeroTier