mirror of
https://github.com/henrydcase/pqc.git
synced 2024-11-27 18:01:34 +00:00
84 lines
2.0 KiB
C
84 lines
2.0 KiB
C
|
#include "vec128.h"
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_set1_16b(uint16_t a) {
|
||
|
return _mm_set1_epi16(a);
|
||
|
}
|
||
|
|
||
|
int PQCLEAN_MCELIECE348864_AVX_vec128_testz(vec128 a) {
|
||
|
return _mm_testz_si128(a, a);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_setzero(void) {
|
||
|
return _mm_setzero_si128();
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_and(vec128 a, vec128 b) {
|
||
|
return _mm_and_si128(a, b);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_xor(vec128 a, vec128 b) {
|
||
|
return _mm_xor_si128(a, b);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_or(vec128 a, vec128 b) {
|
||
|
return _mm_or_si128(a, b);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_sll_2x(vec128 a, int s) {
|
||
|
return _mm_slli_epi64(a, s);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_srl_2x(vec128 a, int s) {
|
||
|
return _mm_srli_epi64(a, s);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_set2x(uint64_t a0, uint64_t a1) {
|
||
|
return _mm_set_epi64x(a1, a0);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_unpack_low(vec128 a, vec128 b) {
|
||
|
return _mm_unpacklo_epi64(a, b);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_unpack_high(vec128 a, vec128 b) {
|
||
|
return _mm_unpackhi_epi64(a, b);
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_setbits(uint64_t a) {
|
||
|
return _mm_set1_epi64x(-a);
|
||
|
}
|
||
|
|
||
|
void PQCLEAN_MCELIECE348864_AVX_vec128_copy(vec128 *dest, const vec128 *src) {
|
||
|
int i;
|
||
|
|
||
|
for (i = 0; i < GFBITS; i++) {
|
||
|
dest[i] = src[i];
|
||
|
}
|
||
|
}
|
||
|
|
||
|
void PQCLEAN_MCELIECE348864_AVX_vec128_add(vec128 *c, const vec128 *a, const vec128 *b) {
|
||
|
int i;
|
||
|
|
||
|
for (i = 0; i < GFBITS; i++) {
|
||
|
c[i] = PQCLEAN_MCELIECE348864_AVX_vec128_xor(a[i], b[i]);
|
||
|
}
|
||
|
}
|
||
|
|
||
|
vec128 PQCLEAN_MCELIECE348864_AVX_vec128_or_reduce(const vec128 *a) {
|
||
|
int i;
|
||
|
vec128 ret;
|
||
|
|
||
|
ret = a[0];
|
||
|
for (i = 1; i < GFBITS; i++) {
|
||
|
ret = PQCLEAN_MCELIECE348864_AVX_vec128_or(ret, a[i]);
|
||
|
}
|
||
|
|
||
|
return ret;
|
||
|
}
|
||
|
|
||
|
/* bitsliced field multiplications */
|
||
|
void PQCLEAN_MCELIECE348864_AVX_vec128_mul(vec128 *h, vec128 *f, const vec128 *g) {
|
||
|
PQCLEAN_MCELIECE348864_AVX_vec128_mul_asm(h, f, g, 16);
|
||
|
}
|
||
|
|