Searched refs:s2_8x16 (Results 1 – 1 of 1) sorted by relevance
803 const int8x16_t s2_8x16 = vld1q_s8((const int8_t*)(row_ptr + col)); in NeonMatrixBatchVectorMultiplyImpl() local807 vmull_s8(vget_low_s8(s1_8x16), vget_low_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyImpl()815 vmlal_s8(prod_16x8, vget_high_s8(s1_8x16), vget_high_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyImpl()1127 const int8x16_t s2_8x16 = vld1q_s8((const int8_t*)(row_ptr + col)); in NeonMatrixBatchVectorMultiplyAccumulate() local1131 vmull_s8(vget_low_s8(s1_8x16), vget_low_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyAccumulate()1139 vmlal_s8(prod_16x8, vget_high_s8(s1_8x16), vget_high_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyAccumulate()1320 const int8x16_t s2_8x16 = vld1q_s8((const int8_t*)(row_ptr + col)); in NeonMatrixBatchVectorMultiplyAccumulateImpl() local1324 vmull_s8(vget_low_s8(s1_8x16), vget_low_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyAccumulateImpl()1332 vmlal_s8(prod_16x8, vget_high_s8(s1_8x16), vget_high_s8(s2_8x16)); in NeonMatrixBatchVectorMultiplyAccumulateImpl()2020 const int8x16_t s2_8x16 = vld1q_s8((const int8_t*)(row_ptr)); in NeonSparseMatrixBatchVectorMultiplyAccumulate() local[all …]