+
+#if defined (USE_AVX512)
+
+ [[maybe_unused]] const __m512i kOnes512 = _mm512_set1_epi16(1);
+
+ [[maybe_unused]] auto m512_hadd = [](__m512i sum, int bias) -> int {
+ return _mm512_reduce_add_epi32(sum) + bias;
+ };
+
+ [[maybe_unused]] auto m512_haddx4 = [](__m512i sum0, __m512i sum1, __m512i sum2, __m512i sum3, __m128i bias) -> __m128i {
+ __m512i sum01a = _mm512_unpacklo_epi32(sum0, sum1);
+ __m512i sum01b = _mm512_unpackhi_epi32(sum0, sum1);
+
+ __m512i sum23a = _mm512_unpacklo_epi32(sum2, sum3);
+ __m512i sum23b = _mm512_unpackhi_epi32(sum2, sum3);
+
+ __m512i sum01 = _mm512_add_epi32(sum01a, sum01b);
+ __m512i sum23 = _mm512_add_epi32(sum23a, sum23b);
+
+ __m512i sum0123a = _mm512_unpacklo_epi64(sum01, sum23);
+ __m512i sum0123b = _mm512_unpackhi_epi64(sum01, sum23);
+
+ __m512i sum = _mm512_add_epi32(sum0123a, sum0123b);
+
+ __m256i sum256lo = _mm512_castsi512_si256(sum);
+ __m256i sum256hi = _mm512_extracti64x4_epi64(sum, 1);
+
+ sum256lo = _mm256_add_epi32(sum256lo, sum256hi);
+
+ __m128i sum128lo = _mm256_castsi256_si128(sum256lo);
+ __m128i sum128hi = _mm256_extracti128_si256(sum256lo, 1);
+
+ return _mm_add_epi32(_mm_add_epi32(sum128lo, sum128hi), bias);
+ };
+
+ [[maybe_unused]] auto m512_add_dpbusd_epi32 = [=](__m512i& acc, __m512i a, __m512i b) {
+#if defined (USE_VNNI)
+ acc = _mm512_dpbusd_epi32(acc, a, b);
+#else
+ __m512i product0 = _mm512_maddubs_epi16(a, b);
+ product0 = _mm512_madd_epi16(product0, kOnes512);
+ acc = _mm512_add_epi32(acc, product0);
+#endif
+ };
+
+#endif
+#if defined (USE_AVX2)
+
+ [[maybe_unused]] const __m256i kOnes256 = _mm256_set1_epi16(1);
+
+ [[maybe_unused]] auto m256_hadd = [](__m256i sum, int bias) -> int {
+ __m128i sum128 = _mm_add_epi32(_mm256_castsi256_si128(sum), _mm256_extracti128_si256(sum, 1));
+ sum128 = _mm_add_epi32(sum128, _mm_shuffle_epi32(sum128, _MM_PERM_BADC));
+ sum128 = _mm_add_epi32(sum128, _mm_shuffle_epi32(sum128, _MM_PERM_CDAB));
+ return _mm_cvtsi128_si32(sum128) + bias;
+ };
+
+ [[maybe_unused]] auto m256_haddx4 = [](__m256i sum0, __m256i sum1, __m256i sum2, __m256i sum3, __m128i bias) -> __m128i {
+ sum0 = _mm256_hadd_epi32(sum0, sum1);
+ sum2 = _mm256_hadd_epi32(sum2, sum3);
+
+ sum0 = _mm256_hadd_epi32(sum0, sum2);
+
+ __m128i sum128lo = _mm256_castsi256_si128(sum0);
+ __m128i sum128hi = _mm256_extracti128_si256(sum0, 1);
+
+ return _mm_add_epi32(_mm_add_epi32(sum128lo, sum128hi), bias);
+ };
+
+ [[maybe_unused]] auto m256_add_dpbusd_epi32 = [=](__m256i& acc, __m256i a, __m256i b) {
+#if defined (USE_VNNI)
+ acc = _mm256_dpbusd_epi32(acc, a, b);
+#else
+ __m256i product0 = _mm256_maddubs_epi16(a, b);
+ product0 = _mm256_madd_epi16(product0, kOnes256);
+ acc = _mm256_add_epi32(acc, product0);
+#endif
+ };
+
+#endif
+
+#if defined (USE_SSSE3)
+
+ [[maybe_unused]] const __m128i kOnes128 = _mm_set1_epi16(1);
+
+ [[maybe_unused]] auto m128_hadd = [](__m128i sum, int bias) -> int {
+ sum = _mm_add_epi32(sum, _mm_shuffle_epi32(sum, 0x4E)); //_MM_PERM_BADC
+ sum = _mm_add_epi32(sum, _mm_shuffle_epi32(sum, 0xB1)); //_MM_PERM_CDAB
+ return _mm_cvtsi128_si32(sum) + bias;
+ };
+
+ [[maybe_unused]] auto m128_haddx4 = [](__m128i sum0, __m128i sum1, __m128i sum2, __m128i sum3, __m128i bias) -> __m128i {
+ sum0 = _mm_hadd_epi32(sum0, sum1);
+ sum2 = _mm_hadd_epi32(sum2, sum3);
+
+ sum0 = _mm_hadd_epi32(sum0, sum2);
+
+ return _mm_add_epi32(sum0, bias);
+ };
+
+ [[maybe_unused]] auto m128_add_dpbusd_epi32 = [=](__m128i& acc, __m128i a, __m128i b) {
+ __m128i product0 = _mm_maddubs_epi16(a, b);
+ product0 = _mm_madd_epi16(product0, kOnes128);
+ acc = _mm_add_epi32(acc, product0);
+ };
+
+#endif
+
+#if defined (USE_AVX512)
+
+ constexpr IndexType kNumChunks512 = kPaddedInputDimensions / (kSimdWidth * 2);
+ constexpr IndexType kNumChunks256 = kPaddedInputDimensions / kSimdWidth;
+