| // Auto-generated file. Do not edit! |
| // Template: src/qs8-vadd/avx2-mul32-ld64.c.in |
| // Generator: tools/xngen |
| // |
| // Copyright 2020 Google LLC |
| // |
| // This source code is licensed under the BSD-style license found in the |
| // LICENSE file in the root directory of this source tree. |
| |
| #include <assert.h> |
| |
| #include <immintrin.h> |
| |
| #include <xnnpack/intrinsics-polyfill.h> |
| #include <xnnpack/vadd.h> |
| |
| |
| void xnn_qs8_vadd_minmax_ukernel__avx2_mul32_ld64_x8( |
| size_t n, |
| const int8_t* input_a, |
| const int8_t* input_b, |
| int8_t* output, |
| const union xnn_qs8_add_minmax_params params[restrict XNN_MIN_ELEMENTS(1)]) XNN_DISABLE_TSAN |
| { |
| const __m256i vbias = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) params->sse2.bias)); |
| const __m256i va_multiplier = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) params->sse2.a_multiplier)); |
| const __m256i vb_multiplier = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) params->sse2.b_multiplier)); |
| const __m256i vremainder_mask = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) params->sse2.remainder_mask)); |
| const __m256i vremainder_threshold = _mm256_broadcastsi128_si256(_mm_load_si128((const __m128i*) params->sse2.remainder_threshold)); |
| const __m128i vshift = _mm_cvtsi32_si128((int) params->sse2.shift); |
| const __m128i voutput_zero_point = _mm_load_si128((const __m128i*) params->sse2.output_zero_point); |
| const __m128i voutput_min = _mm_load_si128((const __m128i*) params->sse2.output_min); |
| const __m128i voutput_max = _mm_load_si128((const __m128i*) params->sse2.output_max); |
| |
| for (; n >= 8 * sizeof(int8_t); n -= 8 * sizeof(int8_t)) { |
| const __m256i va01234567 = _mm256_cvtepi8_epi32(_mm_loadl_epi64((const __m128i*) input_a)); |
| const __m256i vb01234567 = _mm256_cvtepi8_epi32(_mm_loadl_epi64((const __m128i*) input_b)); |
| input_a += 8; |
| input_b += 8; |
| |
| __m256i vacc01234567 = _mm256_add_epi32(vbias, _mm256_mullo_epi32(va01234567, va_multiplier)); |
| |
| vacc01234567 = _mm256_add_epi32(vacc01234567, _mm256_mullo_epi32(vb01234567, vb_multiplier)); |
| |
| const __m256i vrem01234567 = _mm256_add_epi32(_mm256_and_si256(vacc01234567, vremainder_mask), _mm256_srai_epi32(vacc01234567, 31)); |
| |
| vacc01234567 = _mm256_sub_epi32(_mm256_sra_epi32(vacc01234567, vshift), _mm256_cmpgt_epi32(vrem01234567, vremainder_threshold)); |
| |
| __m128i vout01234567 = _mm_adds_epi16(_mm_packs_epi32(_mm256_castsi256_si128(vacc01234567), _mm256_extracti128_si256(vacc01234567, 1)), voutput_zero_point); |
| |
| vout01234567 = _mm_min_epi16(_mm_max_epi16(vout01234567, voutput_min), voutput_max); |
| |
| __m128i vout0123456701234567 = _mm_packs_epi16(vout01234567, vout01234567); |
| |
| _mm_storel_epi64((__m128i*) output, vout0123456701234567); |
| output += 8; |
| } |
| if XNN_UNLIKELY(n != 0) { |
| { |
| const __m256i va01234567 = _mm256_cvtepi8_epi32(_mm_loadl_epi64((const __m128i*) input_a)); |
| const __m256i vb01234567 = _mm256_cvtepi8_epi32(_mm_loadl_epi64((const __m128i*) input_b)); |
| |
| __m256i vacc01234567 = _mm256_add_epi32(vbias, _mm256_mullo_epi32(va01234567, va_multiplier)); |
| |
| vacc01234567 = _mm256_add_epi32(vacc01234567, _mm256_mullo_epi32(vb01234567, vb_multiplier)); |
| |
| const __m256i vrem01234567 = _mm256_add_epi32(_mm256_and_si256(vacc01234567, vremainder_mask), _mm256_srai_epi32(vacc01234567, 31)); |
| |
| vacc01234567 = _mm256_sub_epi32(_mm256_sra_epi32(vacc01234567, vshift), _mm256_cmpgt_epi32(vrem01234567, vremainder_threshold)); |
| |
| __m128i vout01234567 = _mm_adds_epi16(_mm_packs_epi32(_mm256_castsi256_si128(vacc01234567), _mm256_extracti128_si256(vacc01234567, 1)), voutput_zero_point); |
| vout01234567 = _mm_min_epi16(_mm_max_epi16(vout01234567, voutput_min), voutput_max); |
| __m128i vout0123456701234567 = _mm_packs_epi16(vout01234567, vout01234567); |
| |
| if (n & (4 * sizeof(int8_t))) { |
| *((uint32_t*) output) = (uint32_t) _mm_cvtsi128_si32(vout0123456701234567); |
| vout0123456701234567 = _mm_srli_epi64(vout0123456701234567, 32); |
| output += 4; |
| } |
| if (n & (2 * sizeof(int8_t))) { |
| *((uint16_t*) output) = (uint16_t) _mm_extract_epi16(vout0123456701234567, 0); |
| vout0123456701234567 = _mm_srli_epi32(vout0123456701234567, 16); |
| output += 2; |
| } |
| if (n & (1 * sizeof(int8_t))) { |
| *output = (int8_t) _mm_extract_epi8(vout0123456701234567, 0); |
| } |
| } |
| } |
| } |