/github/workspace/src/MatrixFunctions/mat_mult_trans_cmplx/kernels/plp_mat_mult_trans_cmplx_i32p_xpulpv2.c

Functions

	Name
void	plp_mat_mult_trans_cmplx_i32p_xpulpv2(void * args) parallel matrix transpose matrix multiplication for complex 32-bit integers on XpulpV2

Defines

	Name
	BASIC_VERSION

Functions Documentation

function plp_mat_mult_trans_cmplx_i32p_xpulpv2

void plp_mat_mult_trans_cmplx_i32p_xpulpv2(
    void * args
)

parallel matrix transpose matrix multiplication for complex 32-bit integers on XpulpV2

Parameters:

• args pointer to plp_mat_mat_mult_trans_cmplx_instance_i32 struct initialized by plp_mat_mult_trans_cmplx_i32_parallel

Return: none

Macros Documentation

define BASIC_VERSION

#define BASIC_VERSION 

Source code

/* =====================================================================
 * Project:      PULP DSP Library
 * Title:        plp_mat_mult_trans_cmplx_i32p_xpulpv2.c
 * Description:  parallel 32-bit integer complex matrix transpose matrix multiplication for XPULPV2
 *
 * $Date:        complex matrix transpose matrix multiplication
 * $Revision:    V0
 *
 * Target Processor: PULP cores
 * ===================================================================== */
/*
 * Copyright (C) 2020 ETH Zurich and Ubiversity of Bologna. All rights reserved.
 *
 * Author: Tibor Schneider, ETH Zurich
 *
 * SPDX-License-Identifier: Apache-2.0
 *
 * Licensed under the Apache License, Version 2.0 (the License); you may
 * not use this file except in compliance with the License.
 * You may obtain a copy of the License at
 *
 * www.apache.org/licenses/LICENSE-2.0
 *
 * Unless required by applicable law or agreed to in writing, software
 * distributed under the License is distributed on an AS IS BASIS, WITHOUT
 * WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 * See the License for the specific language governing permissions and
 * limitations under the License.
 */

#include "plp_math.h"

void plp_mat_mult_trans_cmplx_i32p_xpulpv2(void *args) {

    int core_id = hal_core_id();

    plp_mat_mult_cmplx_instance_i32 *a = (plp_mat_mult_cmplx_instance_i32 *)args;

    const int32_t *__restrict__ pSrcA = a->pSrcA;
    const int32_t *__restrict__ pSrcB = a->pSrcB;
    uint32_t M = a->M;
    uint32_t N = a->N;
    uint32_t O = a->O;
    uint32_t nPE = a->nPE;
    int32_t *__restrict__ pDstC = a->pDstC;

#define BASIC_VERSION // if used don't forget to also use the undefine at end of file
#ifdef BASIC_VERSION

    for (int m = core_id; m < M; m += nPE) {
        for (int o = 0; o < O; o++) {
            int32_t sum_re = 0;
            int32_t sum_im = 0;
            for (int n = 0; n < N; n++) {
                int32_t a_re = (int32_t)pSrcA[(m * N + n) * 2 + 0];
                int32_t a_im = (int32_t)pSrcA[(m * N + n) * 2 + 1];
                int32_t b_re = (int32_t)pSrcB[(o * N + n) * 2 + 0];
                int32_t b_im = (int32_t)pSrcB[(o * N + n) * 2 + 1];
                sum_re += a_re * b_re - a_im * b_im;
                sum_im += a_re * b_im + a_im * b_re;
            }
            pDstC[(m * O + o) * 2 + 0] = sum_re;
            pDstC[(m * O + o) * 2 + 1] = sum_im;
        }
    }

#else

    // TODO: Hackathon

#endif
#undef BASIC_VERSION
}

---

Updated on 2023-03-01 at 16:16:33 +0000