Ecosystem/xmrig
Ecosystem/xmrig
1
0
Fork 0
mirror of https://github.com/xmrig/xmrig.git synced 2024-12-23 12:09:22 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

Fixed performance degradation for v0 & v1

Browse source
This commit is contained in:
SChernykh 2018-08-29 20:15:08 +02:00
parent f151c7131d
commit 7c4b718789
1 changed files with 39 additions and 51 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

90
src/crypto/CryptoNight_x86.h
View file

@ -481,12 +481,11 @@ inline void cryptonight_single_hash(const uint8_t *__restrict__ input, size_t si
__m128i bx1 = _mm_set_epi64x(h0[9] ^ h0[11], h0[8] ^ h0[10]);
uint64_t idx0 = al0;
uint64_t* ptr0 = (uint64_t*) &l0[idx0 & MASK];
for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx;
if (VARIANT == xmrig::VARIANT_TUBE || !SOFT_AES) {
cx = _mm_load_si128((__m128i *) ptr0);
cx = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
}
const __m128i ax0 = _mm_set_epi64x(ah0, al0);
@ -494,24 +493,23 @@ inline void cryptonight_single_hash(const uint8_t *__restrict__ input, size_t si
cx = aes_round_tweak_div(cx, ax0);
}
else if (SOFT_AES) {
cx = soft_aesenc((uint32_t*) ptr0, ax0);
cx = soft_aesenc((uint32_t*) &l0[idx0 & MASK], ax0);
}
else {
cx = _mm_aesenc_si128(cx, ax0);
}
if (IS_V1 || VARIANT == xmrig::VARIANT_2) {
cryptonight_monero_tweak<VARIANT>(ptr0, l0, idx0 & MASK, ax0, bx0, bx1, cx);
cryptonight_monero_tweak<VARIANT>((uint64_t*) &l0[idx0 & MASK], l0, idx0 & MASK, ax0, bx0, bx1, cx);
} else {
_mm_store_si128((__m128i *) ptr0, _mm_xor_si128(bx0, cx));
_mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx0, cx));
}
idx0 = EXTRACT64(cx);
ptr0 = (uint64_t*) &l0[idx0 & MASK];
uint64_t hi, lo, cl, ch;
cl = ptr0[0];
ch = ptr0[1];
cl = ((uint64_t*) &l0[idx0 & MASK])[0];
ch = ((uint64_t*) &l0[idx0 & MASK])[1];
if (VARIANT == xmrig::VARIANT_2) {
VARIANT2_INTEGER_MATH(0, cl, cx);
lo = __umul128(idx0, cl, &hi);
@ -524,34 +522,32 @@ inline void cryptonight_single_hash(const uint8_t *__restrict__ input, size_t si
al0 += hi;
ah0 += lo;
ptr0[0] = al0;
((uint64_t*) &l0[idx0 & MASK])[0] = al0;
if (IS_V1 && (VARIANT == xmrig::VARIANT_TUBE || VARIANT == xmrig::VARIANT_RTO)) {
ptr0[1] = ah0 ^ tweak1_2_0 ^ al0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0 ^ tweak1_2_0 ^ al0;
} else if (IS_V1) {
ptr0[1] = ah0 ^ tweak1_2_0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0 ^ tweak1_2_0;
} else {
ptr0[1] = ah0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0;
}
al0 ^= cl;
ah0 ^= ch;
idx0 = al0;
ptr0 = (uint64_t*) &l0[idx0 & MASK];
if (ALGO == xmrig::CRYPTONIGHT_HEAVY) {
int64_t n = ((int64_t*)ptr0)[0];
int32_t d = ((int32_t*)ptr0)[2];
int64_t n = ((int64_t*) &l0[idx0 & MASK])[0];
int32_t d = ((int32_t*) &l0[idx0 & MASK])[2];
int64_t q = n / (d | 0x5);
((int64_t*) ptr0)[0] = n ^ q;
((int64_t*) &l0[idx0 & MASK])[0] = n ^ q;
if (VARIANT == xmrig::VARIANT_XHV) {
d = ~d;
}
idx0 = d ^ q;
ptr0 = (uint64_t*)&l0[idx0 & MASK];
}
if (VARIANT == xmrig::VARIANT_2) {
bx1 = bx0;
@ -608,14 +604,12 @@ inline void cryptonight_double_hash(const uint8_t *__restrict__ input, size_t si
uint64_t idx0 = al0;
uint64_t idx1 = al1;
uint64_t* ptr0 = (uint64_t*)&l0[idx0 & MASK];
uint64_t* ptr1 = (uint64_t*)&l1[idx1 & MASK];
for (size_t i = 0; i < ITERATIONS; i++) {
__m128i cx0, cx1;
if (VARIANT == xmrig::VARIANT_TUBE || !SOFT_AES) {
cx0 = _mm_load_si128((__m128i *) ptr0);
cx1 = _mm_load_si128((__m128i *) ptr1);
cx0 = _mm_load_si128((__m128i *) &l0[idx0 & MASK]);
cx1 = _mm_load_si128((__m128i *) &l1[idx1 & MASK]);
}
const __m128i ax0 = _mm_set_epi64x(ah0, al0);
@ -625,8 +619,8 @@ inline void cryptonight_double_hash(const uint8_t *__restrict__ input, size_t si
cx1 = aes_round_tweak_div(cx1, ax1);
}
else if (SOFT_AES) {
cx0 = soft_aesenc((uint32_t*)ptr0, ax0);
cx1 = soft_aesenc((uint32_t*)ptr1, ax1);
cx0 = soft_aesenc((uint32_t*) &l0[idx0 & MASK], ax0);
cx1 = soft_aesenc((uint32_t*) &l1[idx1 & MASK], ax1);
}
else {
cx0 = _mm_aesenc_si128(cx0, ax0);
@ -634,21 +628,19 @@ inline void cryptonight_double_hash(const uint8_t *__restrict__ input, size_t si
}
if (IS_V1 || (VARIANT == xmrig::VARIANT_2)) {
cryptonight_monero_tweak<VARIANT>(ptr0, l0, idx0 & MASK, ax0, bx00, bx01, cx0);
cryptonight_monero_tweak<VARIANT>(ptr1, l1, idx1 & MASK, ax1, bx10, bx11, cx1);
cryptonight_monero_tweak<VARIANT>((uint64_t*) &l0[idx0 & MASK], l0, idx0 & MASK, ax0, bx00, bx01, cx0);
cryptonight_monero_tweak<VARIANT>((uint64_t*) &l1[idx1 & MASK], l1, idx1 & MASK, ax1, bx10, bx11, cx1);
} else {
_mm_store_si128((__m128i *) ptr0, _mm_xor_si128(bx00, cx0));
_mm_store_si128((__m128i *) ptr1, _mm_xor_si128(bx10, cx1));
_mm_store_si128((__m128i *) &l0[idx0 & MASK], _mm_xor_si128(bx00, cx0));
_mm_store_si128((__m128i *) &l1[idx1 & MASK], _mm_xor_si128(bx10, cx1));
}
idx0 = EXTRACT64(cx0);
idx1 = EXTRACT64(cx1);
ptr0 = (uint64_t*)&l0[idx0 & MASK];
ptr1 = (uint64_t*)&l1[idx1 & MASK];
uint64_t hi, lo, cl, ch;
cl = ((uint64_t*)ptr0)[0];
ch = ((uint64_t*)ptr0)[1];
cl = ((uint64_t*) &l0[idx0 & MASK])[0];
ch = ((uint64_t*) &l0[idx0 & MASK])[1];
if (VARIANT == xmrig::VARIANT_2) {
VARIANT2_INTEGER_MATH(0, cl, cx0);
lo = __umul128(idx0, cl, &hi);
@ -660,38 +652,36 @@ inline void cryptonight_double_hash(const uint8_t *__restrict__ input, size_t si
al0 += hi;
ah0 += lo;
((uint64_t*)ptr0)[0] = al0;
((uint64_t*) &l0[idx0 & MASK])[0] = al0;
if (IS_V1 && (VARIANT == xmrig::VARIANT_TUBE || VARIANT == xmrig::VARIANT_RTO)) {
((uint64_t*)ptr0)[1] = ah0 ^ tweak1_2_0 ^ al0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0 ^ tweak1_2_0 ^ al0;
} else if (IS_V1) {
((uint64_t*)ptr0)[1] = ah0 ^ tweak1_2_0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0 ^ tweak1_2_0;
} else {
((uint64_t*)ptr0)[1] = ah0;
((uint64_t*) &l0[idx0 & MASK])[1] = ah0;
}
al0 ^= cl;
ah0 ^= ch;
idx0 = al0;
ptr0 = (uint64_t*)&l0[idx0 & MASK];
if (ALGO == xmrig::CRYPTONIGHT_HEAVY) {
int64_t n = ((int64_t*)ptr0)[0];
int32_t d = ((int32_t*)ptr0)[2];
int64_t n = ((int64_t*) &l0[idx0 & MASK])[0];
int32_t d = ((int32_t*) &l0[idx0 & MASK])[2];
int64_t q = n / (d | 0x5);
((int64_t*)ptr0)[0] = n ^ q;
((int64_t*) &l0[idx0 & MASK])[0] = n ^ q;
if (VARIANT == xmrig::VARIANT_XHV) {
d = ~d;
}
idx0 = d ^ q;
ptr0 = (uint64_t*)&l0[idx0 & MASK];
}
cl = ptr1[0];
ch = ptr1[1];
cl = ((uint64_t*) &l1[idx1 & MASK])[0];
ch = ((uint64_t*) &l1[idx1 & MASK])[1];
if (VARIANT == xmrig::VARIANT_2) {
VARIANT2_INTEGER_MATH(1, cl, cx1);
lo = __umul128(idx1, cl, &hi);
@ -703,34 +693,32 @@ inline void cryptonight_double_hash(const uint8_t *__restrict__ input, size_t si
al1 += hi;
ah1 += lo;
ptr1[0] = al1;
((uint64_t*)&l1[idx1 & MASK])[0] = al1;
if (IS_V1 && (VARIANT == xmrig::VARIANT_TUBE || VARIANT == xmrig::VARIANT_RTO)) {
ptr1[1] = ah1 ^ tweak1_2_1 ^ al1;
((uint64_t*)&l1[idx1 & MASK])[1] = ah1 ^ tweak1_2_1 ^ al1;
} else if (IS_V1) {
ptr1[1] = ah1 ^ tweak1_2_1;
((uint64_t*)&l1[idx1 & MASK])[1] = ah1 ^ tweak1_2_1;
} else {
ptr1[1] = ah1;
((uint64_t*)&l1[idx1 & MASK])[1] = ah1;
}
al1 ^= cl;
ah1 ^= ch;
idx1 = al1;
ptr1 = (uint64_t*)&l1[idx1 & MASK];
if (ALGO == xmrig::CRYPTONIGHT_HEAVY) {
int64_t n = ((int64_t*)ptr1)[0];
int32_t d = ((int32_t*)ptr1)[2];
int64_t n = ((int64_t*) &l1[idx1 & MASK])[0];
int32_t d = ((int32_t*) &l1[idx1 & MASK])[2];
int64_t q = n / (d | 0x5);
((int64_t*)ptr1)[0] = n ^ q;
((int64_t*) &l1[idx1 & MASK])[0] = n ^ q;
if (VARIANT == xmrig::VARIANT_XHV) {
d = ~d;
}
idx1 = d ^ q;
ptr1 = (uint64_t*)&l1[idx1 & MASK];
}
if (VARIANT == xmrig::VARIANT_2) {