Lines Matching refs:CMPLX_DIM
81 float32_t *pInA1 = pInA0 + CMPLX_DIM * MATRIX_DIM2; in arm_mat_cmplx_mult_f32_2x2_mve()
86 MATRIX_DIM2 * CMPLX_DIM, MATRIX_DIM2 * CMPLX_DIM + 1 in arm_mat_cmplx_mult_f32_2x2_mve()
103 pOut[0 * CMPLX_DIM * MATRIX_DIM2 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_2x2_mve()
104 pOut[0 * CMPLX_DIM * MATRIX_DIM2 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_2x2_mve()
105 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_2x2_mve()
106 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_2x2_mve()
107 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_2x2_mve()
112 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_2x2_mve()
124 pOut[0 * CMPLX_DIM * MATRIX_DIM2 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_2x2_mve()
125 pOut[0 * CMPLX_DIM * MATRIX_DIM2 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_2x2_mve()
126 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_2x2_mve()
127 pOut[1 * CMPLX_DIM * MATRIX_DIM2 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_2x2_mve()
145 float32_t *pInA1 = pInA0 + CMPLX_DIM * MATRIX_DIM3; in arm_mat_cmplx_mult_f32_3x3_mve()
146 float32_t *pInA2 = pInA1 + CMPLX_DIM * MATRIX_DIM3; in arm_mat_cmplx_mult_f32_3x3_mve()
150 mve_pred16_t p0 = vctp32q(CMPLX_DIM); in arm_mat_cmplx_mult_f32_3x3_mve()
153 MATRIX_DIM3 * CMPLX_DIM, MATRIX_DIM3 * CMPLX_DIM + 1 in arm_mat_cmplx_mult_f32_3x3_mve()
155 …static const uint32_t offsetB1[4] = { 2 * MATRIX_DIM3 * CMPLX_DIM, 2 * MATRIX_DIM3 * CMPLX_DIM + 1, in arm_mat_cmplx_mult_f32_3x3_mve()
194 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
195 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
196 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
197 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
198 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
199 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
200 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_3x3_mve()
205 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_3x3_mve()
236 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
237 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
238 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
239 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
240 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
241 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
242 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_3x3_mve()
247 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_3x3_mve()
278 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
279 pOut[0 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
280 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
281 pOut[1 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
282 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_3x3_mve()
283 pOut[2 * CMPLX_DIM * MATRIX_DIM3 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_3x3_mve()
302 float32_t *pInA1 = pInA0 + CMPLX_DIM * MATRIX_DIM4; in arm_mat_cmplx_mult_f32_4x4_mve()
303 float32_t *pInA2 = pInA1 + CMPLX_DIM * MATRIX_DIM4; in arm_mat_cmplx_mult_f32_4x4_mve()
304 float32_t *pInA3 = pInA2 + CMPLX_DIM * MATRIX_DIM4; in arm_mat_cmplx_mult_f32_4x4_mve()
309 MATRIX_DIM4 * CMPLX_DIM, MATRIX_DIM4 * CMPLX_DIM + 1 in arm_mat_cmplx_mult_f32_4x4_mve()
311 …static const uint32_t offsetB1[4] = { 2 * MATRIX_DIM4 * CMPLX_DIM, 2 * MATRIX_DIM4 * CMPLX_DIM + 1, in arm_mat_cmplx_mult_f32_4x4_mve()
312 3 * MATRIX_DIM4 * CMPLX_DIM, 3 * MATRIX_DIM4 * CMPLX_DIM + 1 in arm_mat_cmplx_mult_f32_4x4_mve()
356 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
357 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
358 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
359 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
360 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
361 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
362 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
363 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
364 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
369 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
407 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
408 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
409 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
410 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
411 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
412 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
413 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
414 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
415 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
420 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
458 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
459 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
460 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
461 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
462 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
463 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
464 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
465 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
466 pOut += CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
471 pInB = pInB + CMPLX_DIM; in arm_mat_cmplx_mult_f32_4x4_mve()
509 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
510 pOut[0 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
511 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
512 pOut[1 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
513 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
514 pOut[2 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
515 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32_4x4_mve()
516 pOut[3 * CMPLX_DIM * MATRIX_DIM4 + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32_4x4_mve()
576 vecColBOffs[2] = numColsB * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
577 vecColBOffs[3] = (numColsB * CMPLX_DIM) + 1; in arm_mat_cmplx_mult_f32()
592 px = pOut + i * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
617 float32_t const *pInA1 = pInA0 + numColsA * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
618 float32_t const *pInA2 = pInA1 + numColsA * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
619 float32_t const *pInA3 = pInA2 + numColsA * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
637 blkCnt = (numColsA * CMPLX_DIM) >> 2; in arm_mat_cmplx_mult_f32()
646 vecOffs = vecOffs + (uint32_t) (numColsB * 2 * CMPLX_DIM); in arm_mat_cmplx_mult_f32()
669 blkCnt = (numColsA * CMPLX_DIM) & 3; in arm_mat_cmplx_mult_f32()
679 vecOffs = vecOffs + (uint32_t) (numColsB * 2 * CMPLX_DIM); in arm_mat_cmplx_mult_f32()
696 px[0 * CMPLX_DIM * numColsB + 0] = acc0[0] + acc0[2]; in arm_mat_cmplx_mult_f32()
697 px[0 * CMPLX_DIM * numColsB + 1] = acc0[1] + acc0[3]; in arm_mat_cmplx_mult_f32()
698 px[1 * CMPLX_DIM * numColsB + 0] = acc1[0] + acc1[2]; in arm_mat_cmplx_mult_f32()
699 px[1 * CMPLX_DIM * numColsB + 1] = acc1[1] + acc1[3]; in arm_mat_cmplx_mult_f32()
700 px[2 * CMPLX_DIM * numColsB + 0] = acc2[0] + acc2[2]; in arm_mat_cmplx_mult_f32()
701 px[2 * CMPLX_DIM * numColsB + 1] = acc2[1] + acc2[3]; in arm_mat_cmplx_mult_f32()
702 px[3 * CMPLX_DIM * numColsB + 0] = acc3[0] + acc3[2]; in arm_mat_cmplx_mult_f32()
703 px[3 * CMPLX_DIM * numColsB + 1] = acc3[1] + acc3[3]; in arm_mat_cmplx_mult_f32()
704 px += CMPLX_DIM; in arm_mat_cmplx_mult_f32()
712 pInB = (float32_t const *) pSrcB->pData + (numColsB - col) * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
718 pInA += (numColsA * 4) * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
732 px = pOut + i * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
768 blkCnt = (numColsA * CMPLX_DIM) >> 2; in arm_mat_cmplx_mult_f32()
777 vecOffs = vecOffs + (uint32_t) (numColsB * 2 * CMPLX_DIM); in arm_mat_cmplx_mult_f32()
792 blkCnt = (numColsA * CMPLX_DIM) & 3; in arm_mat_cmplx_mult_f32()
809 px += CMPLX_DIM; in arm_mat_cmplx_mult_f32()
817 pInB = (float32_t const *) pSrcB->pData + (numColsB - col) * CMPLX_DIM; in arm_mat_cmplx_mult_f32()
823 pInA += numColsA * CMPLX_DIM; in arm_mat_cmplx_mult_f32()