diff options
| -rw-r--r-- | sha3.c | 29 |
1 files changed, 12 insertions, 17 deletions
@@ -527,11 +527,10 @@ static inline void permute_n_avx512(uint64_t s[static 25], const size_t num_roun #include <immintrin.h> // lane masks -static const __m256i LM0 = { ~0, 0, 0, 0 }, // mask, first lane only - LM1 = { 0, ~0, 0, 0 }, - LM2 = { 0, 0, ~0, 0 }, - LM3 = { 0, 0, 0, ~0 }, - K64 = { 64, 64, 64, 64 }; // 64, all lanes (used by ROLV) +static const __m256i LM0 = { ~0, 0, 0, 0 }, // only lane 0 + LM1 = { 0, ~0, 0, 0 }, // only lane 1 + LM2 = { 0, 0, ~0, 0 }, // only lane 2 + LM3 = { 0, 0, 0, ~0 }; // only lane 3 // load state array to avx2 registers // FIXME: remove macro, not needed @@ -541,17 +540,15 @@ static const __m256i LM0 = { ~0, 0, 0, 0 }, // mask, first lane only r2_lo = _mm256_loadu_si256((__m256i*) (s + 10)), /* row 2, cols 0-3 */ \ r3_lo = _mm256_loadu_si256((__m256i*) (s + 15)), /* row 3, cols 0-3 */ \ r4_lo = _mm256_loadu_si256((__m256i*) (s + 20)), /* row 4, cols 0-3 */ \ - r0_hi = { s[ 4] }, /* row 0, col 4 */ \ - r1_hi = { s[ 9] }, /* row 1, col 4 */ \ - r2_hi = { s[14] }, /* row 2, col 4 */ \ - r3_hi = { s[19] }, /* row 3, col 4 */ \ - r4_hi = { s[24] }; /* row 4, col 4 */ + r0_hi = { s[ 4] }, /* row 0, col 4 */ \ + r1_hi = { s[ 9] }, /* row 1, col 4 */ \ + r2_hi = { s[14] }, /* row 2, col 4 */ \ + r3_hi = { s[19] }, /* row 3, col 4 */ \ + r4_hi = { s[24] }; /* row 4, col 4 */ // store avx2 registers to state array #define AVX2_STORE(s) do { \ union { long long int *i64; uint64_t *u64; } p = { .u64 = s }; \ - \ - /* store rows */ \ _mm256_storeu_si256((__m256i*) (p.i64 + 0), r0_lo); /* row 0, cols 0-3 */ \ _mm256_storeu_si256((__m256i*) (p.i64 + 5), r1_lo); /* row 1, cols 0-3 */ \ _mm256_storeu_si256((__m256i*) (p.i64 + 10), r2_lo); /* row 2, cols 0-3 */ \ @@ -578,18 +575,16 @@ static const __m256i LM0 = { ~0, 0, 0, 0 }, // mask, first lane only // pi permute IDs #define PI_T0_LO 0xe4 // 0b11100100 -> 0xe4 -#define PI_T0_HI 0x00 #define PI_T1_LO 0x43 // 0b01000011 -> 0x43 #define PI_T1_HI 0x02 #define PI_T2_LO 0x39 // 0b00111001 -> 0x39 -#define PI_T2_HI 0x00 #define PI_T3_LO 0x90 // 0b10010000 -> 0x90 #define PI_T3_HI 0x03 #define PI_T4_LO 0x0e // 0b00001110 -> 0x0e #define PI_T4_HI 0x01 -// chi a blend mask -#define CHI_A_MASK 0xc0 // 0b11000000 +// chi blend mask +#define CHI_MASK 0xc0 // 0b11000000 // chi permute IDs #define CHI_I0_LO 0x39 // 1, 2, 3, 0 -> 0b00111001 -> 0x39 @@ -598,7 +593,7 @@ static const __m256i LM0 = { ~0, 0, 0, 0 }, // mask, first lane only // chi step #define CHI(LO, HI) do { \ - const __m256i a_lo = _mm256_blend_epi32(_mm256_permute4x64_epi64(LO, CHI_I0_LO), _mm256_permute4x64_epi64(HI, CHI_I0_LO), CHI_A_MASK), \ + const __m256i a_lo = _mm256_blend_epi32(_mm256_permute4x64_epi64(LO, CHI_I0_LO), _mm256_permute4x64_epi64(HI, CHI_I0_LO), CHI_MASK), \ a_hi = LO, \ b_lo = (_mm256_permute4x64_epi64(LO, CHI_I1_LO) & ~LM2) | (_mm256_permute4x64_epi64(HI, CHI_I1_LO) & ~LM0), \ b_hi = _mm256_permute4x64_epi64(LO, CHI_I1_HI); \ |