| /cmsis-dsp-latest/Source/MatrixFunctions/ |
| D | arm_mat_mult_fast_q31.c | 76 q31_t sum1, sum2, sum3, sum4; /* Accumulator */ in arm_mat_mult_fast_q31() local
123 sum1 = 0; in arm_mat_mult_fast_q31()
147 sum1 = __SMMLA(inA1, inB1, sum1); in arm_mat_mult_fast_q31()
152 sum1 = (q31_t) ((((q63_t) sum1 << 32) + ((q63_t) inA1 * inB1)) >> 32); in arm_mat_mult_fast_q31()
163 *px++ = sum1 << 1; in arm_mat_mult_fast_q31()
202 sum1 = 0; in arm_mat_mult_fast_q31()
213 sum1 = __SMMLA(*pInA++, *pInB, sum1); in arm_mat_mult_fast_q31()
215 sum1 = (q31_t) ((((q63_t) sum1 << 32) + ((q63_t) *pInA++ * *pInB)) >> 32); in arm_mat_mult_fast_q31()
220 sum1 = __SMMLA(*pInA++, *pInB, sum1); in arm_mat_mult_fast_q31()
222 sum1 = (q31_t) ((((q63_t) sum1 << 32) + ((q63_t) *pInA++ * *pInB)) >> 32); in arm_mat_mult_fast_q31()
[all …]
|
| D | arm_mat_vec_mult_q7.c | 305 q31_t sum1 = 0; in arm_mat_vec_mult_q7() local
335 sum1 = __SMLAD(matData, vecData, sum1); in arm_mat_vec_mult_q7()
336 sum1 = __SMLAD(matData2, vecData2, sum1); in arm_mat_vec_mult_q7()
363 sum1 += *pInA1++ * vecData; in arm_mat_vec_mult_q7()
371 *px++ = (q7_t)(__SSAT((sum1 >> 7), 8)); in arm_mat_vec_mult_q7()
|
| D | arm_mat_cholesky_f64.c | 97 float64_t sum0=0.0,sum1=0.0,sum2=0.0,sum3=0.0; in arm_mat_cholesky_f64() local
140 sum1 = vaddvq_f64(acc1); in arm_mat_cholesky_f64()
150 sum1 = sum1 + pG[i * n + k] * pG[(j + 1) * n + k]; in arm_mat_cholesky_f64()
158 pG[(j + 1) * n + i] -= sum1; in arm_mat_cholesky_f64()
|
| D | arm_mat_vec_mult_q15.c | 293 q63_t sum1 = 0; in arm_mat_vec_mult_q15() local
318 sum1 = __SMLALD(matData, vecData, sum1); in arm_mat_vec_mult_q15()
334 sum1 += (q63_t)*pInA1++ * vecData; in arm_mat_vec_mult_q15()
341 *px++ = (q15_t)(__SSAT((sum1 >> 15), 16)); in arm_mat_vec_mult_q15()
|
| D | arm_mat_cholesky_f32.c | 230 float32_t sum0=0.0f,sum1=0.0f,sum2=0.0f,sum3=0.0f; in arm_mat_cholesky_f32() local
273 sum1 = vpadds_f32(vpadd_f32(vget_low_f32(acc1), vget_high_f32(acc1))); in arm_mat_cholesky_f32()
282 sum1 = vget_lane_f32(tmp, 0) + vget_lane_f32(tmp, 1); in arm_mat_cholesky_f32()
296 sum1 = sum1 + pG[i * n + k] * pG[(j + 1) * n + k]; in arm_mat_cholesky_f32()
304 pG[(j + 1) * n + i] -= sum1; in arm_mat_cholesky_f32()
|
| D | arm_mat_mult_f64.c | 100 float64_t sum0,sum1, sum2,sum3, sum4, sum5 , sum6, sum7; in arm_mat_mult_f64() local
144 sum1 = 0.0; in arm_mat_mult_f64()
216 sum1 += vaddvq_f64(acc1); in arm_mat_mult_f64()
232 sum1 += *pIn1B++ * (*pIn2); in arm_mat_mult_f64()
247 *pxB++ = sum1; in arm_mat_mult_f64()
|
| D | arm_mat_vec_mult_f32.c | 315 float32_t sum1 = 0.0f; in arm_mat_vec_mult_f32() local
340 sum1 += matData * vecData; in arm_mat_vec_mult_f32()
353 *px++ = sum1; in arm_mat_vec_mult_f32()
|
| D | arm_mat_vec_mult_q31.c | 290 q63_t sum1 = 0; in arm_mat_vec_mult_q31() local
316 sum1 += (q63_t)matData * vecData; in arm_mat_vec_mult_q31()
329 *px++ = (q31_t)(sum1 >> 31); in arm_mat_vec_mult_q31()
|
| D | arm_mat_vec_mult_f16.c | 314 float16_t sum1 = 0.0f16; in arm_mat_vec_mult_f16() local
335 sum1 += (_Float16)matData * (_Float16)vecData; in arm_mat_vec_mult_f16()
348 *px++ = sum1; in arm_mat_vec_mult_f16()
|
| D | arm_mat_mult_f32.c | 579 float32_t sum0,sum1, sum2,sum3, sum4, sum5 , sum6, sum7; in arm_mat_mult_f32() local
623 sum1 = 0.0f; in arm_mat_mult_f32()
701 sum1 += vget_lane_f32(accum, 0) + vget_lane_f32(accum, 1); in arm_mat_mult_f32()
729 sum1 += *pIn1B++ * (*pIn2); in arm_mat_mult_f32()
744 *pxB++ = sum1; in arm_mat_mult_f32()
|
| /cmsis-dsp-latest/dsppp/Include/dsppp/Scalar/ |
| D | matrix_multiply.hpp | 75 Acc sum1 = Acc{}; in _dot_m_v()
98 sum1 = inner::mac(sum1, matData, vecData); in _dot_m_v()
108 *px++ = inner::from_accumulator(sum1); in _dot_m_v()
|
| /cmsis-dsp-latest/dsppp/Include/dsppp/DSP/ |
| D | matrix_multiply.hpp | 152 ACC sum1 = ACC{}; in _dot_m_v()
177 sum1 = inner::vmacc(sum1, matData, vecData); in _dot_m_v()
199 sum1 = inner::mac(sum1,*pInA1++ , tmpData); in _dot_m_v()
206 *px++ = inner::from_accumulator(sum1); in _dot_m_v()
|
