horizontally add 8 floats
| 44 | #if __AVX__ || __AVX2__ || __AVX512F__ |
| 45 | // horizontally add 8 floats |
| 46 | static inline float hsum_float_8(const __m256 x) { |
| 47 | __m128 res = _mm256_extractf128_ps(x, 1); |
| 48 | res = _mm_add_ps(res, _mm256_castps256_ps128(x)); |
| 49 | res = _mm_add_ps(res, _mm_movehl_ps(res, res)); |
| 50 | res = _mm_add_ss(res, _mm_movehdup_ps(res)); |
| 51 | return _mm_cvtss_f32(res); |
| 52 | } |
| 53 | |
| 54 | // shuffles to pick the required scales in dot products |
| 55 | static inline __m256i get_scale_shuffle_q3k(int i) { |
no outgoing calls
no test coverage detected