popov/cpuminer-opt-gpu
1
0
Fork 0
mirror of https://github.com/JayDDee/cpuminer-opt.git synced 2025-09-17 23:44:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

v3.9.0

Browse Source
This commit is contained in:
Jay D Dee
2019-05-19 13:39:45 -04:00
parent bfd1c002f9
commit e1aead3c76
139 changed files with 10907 additions and 4218 deletions
Download Patch File Download Diff File Expand all files Collapse all files

40
algo/sm3/sm3-hash-4way.c
View File

@@ -125,20 +125,20 @@ void sm3_4way_close( void *cc, void *dst )
memset_zero_128( block, ( SM3_BLOCK_SIZE - 8 ) >> 2 );
}
count[0] = mm_bswap_32(
count[0] = mm128_bswap_32(
_mm_set1_epi32( ctx->nblocks >> 23 ) );
count[1] = mm_bswap_32( _mm_set1_epi32( ( ctx->nblocks << 9 ) +
count[1] = mm128_bswap_32( _mm_set1_epi32( ( ctx->nblocks << 9 ) +
( ctx->num << 3 ) ) );
sm3_4way_compress( ctx->digest, block );
for ( i = 0; i < 8 ; i++ )
hash[i] = mm_bswap_32( ctx->digest[i] );
hash[i] = mm128_bswap_32( ctx->digest[i] );
}
#define P0(x) _mm_xor_si128( x, _mm_xor_si128( mm_rol_32( x, 9 ), \
mm_rol_32( x, 17 ) ) )
#define P1(x) _mm_xor_si128( x, _mm_xor_si128( mm_rol_32( x, 15 ), \
mm_rol_32( x, 23 ) ) )
#define P0(x) _mm_xor_si128( x, _mm_xor_si128( mm128_rol_32( x, 9 ), \
mm128_rol_32( x, 17 ) ) )
#define P1(x) _mm_xor_si128( x, _mm_xor_si128( mm128_rol_32( x, 15 ), \
mm128_rol_32( x, 23 ) ) )
#define FF0(x,y,z) _mm_xor_si128( x, _mm_xor_si128( y, z ) )
#define FF1(x,y,z) _mm_or_si128( _mm_or_si128( _mm_and_si128( x, y ), \
@@ -165,13 +165,13 @@ void sm3_4way_compress( __m128i *digest, __m128i *block )
int j;
for ( j = 0; j < 16; j++ )
W[j] = mm_bswap_32( block[j] );
W[j] = mm128_bswap_32( block[j] );
for ( j = 16; j < 68; j++ )
W[j] = _mm_xor_si128( P1( _mm_xor_si128( _mm_xor_si128( W[ j-16 ],
W[ j-9 ] ),
mm_rol_32( W[ j-3 ], 15 ) ) ),
_mm_xor_si128( mm_rol_32( W[ j-13 ], 7 ),
mm128_rol_32( W[ j-3 ], 15 ) ) ),
_mm_xor_si128( mm128_rol_32( W[ j-13 ], 7 ),
W[ j-6 ] ) );
for( j = 0; j < 64; j++ )
@@ -180,19 +180,19 @@ void sm3_4way_compress( __m128i *digest, __m128i *block )
T = _mm_set1_epi32( 0x79CC4519UL );
for( j =0; j < 16; j++ )
{
SS1 = mm_rol_32( _mm_add_epi32( _mm_add_epi32( mm_rol_32( A, 12 ), E ),
mm_rol_32( T, j ) ), 7 );
SS2 = _mm_xor_si128( SS1, mm_rol_32( A, 12 ) );
SS1 = mm128_rol_32( _mm_add_epi32( _mm_add_epi32( mm128_rol_32(A,12), E ),
mm128_rol_32( T, j ) ), 7 );
SS2 = _mm_xor_si128( SS1, mm128_rol_32( A, 12 ) );
TT1 = _mm_add_epi32( _mm_add_epi32( _mm_add_epi32( FF0( A, B, C ), D ),
SS2 ), W1[j] );
TT2 = _mm_add_epi32( _mm_add_epi32( _mm_add_epi32( GG0( E, F, G ), H ),
SS1 ), W[j] );
D = C;
C = mm_rol_32( B, 9 );
C = mm128_rol_32( B, 9 );
B = A;
A = TT1;
H = G;
G = mm_rol_32( F, 19 );
G = mm128_rol_32( F, 19 );
F = E;
E = P0( TT2 );
}
@@ -200,19 +200,19 @@ void sm3_4way_compress( __m128i *digest, __m128i *block )
T = _mm_set1_epi32( 0x7A879D8AUL );
for( j =16; j < 64; j++ )
{
SS1 = mm_rol_32( _mm_add_epi32( _mm_add_epi32( mm_rol_32( A, 12 ), E ),
mm_rol_32( T, j&31 ) ), 7 );
SS2 = _mm_xor_si128( SS1, mm_rol_32( A, 12 ) );
SS1 = mm128_rol_32( _mm_add_epi32( _mm_add_epi32( mm128_rol_32(A,12), E ),
mm128_rol_32( T, j&31 ) ), 7 );
SS2 = _mm_xor_si128( SS1, mm128_rol_32( A, 12 ) );
TT1 = _mm_add_epi32( _mm_add_epi32( _mm_add_epi32( FF1( A, B, C ), D ),
SS2 ), W1[j] );
TT2 = _mm_add_epi32( _mm_add_epi32( _mm_add_epi32( GG1( E, F, G ), H ),
SS1 ), W[j] );
D = C;
C = mm_rol_32( B, 9 );
C = mm128_rol_32( B, 9 );
B = A;
A = TT1;
H = G;
G = mm_rol_32( F, 19 );
G = mm128_rol_32( F, 19 );
F = E;
E = P0( TT2 );
}