Simple example for linear system solver in LAPACK

linearSolver.cpp

#include "mkl.h"
#include <math.h>
#include <float.h>
#include <stdio.h>

/* Auxiliary routine: printing a matrix */
void print_matrix(float * matrix, int nrow, int ncol) {
  for (int i = 0; i < nrow; i++) {
    for (int j = 0; j < ncol; j++) {
      printf("%.*e\t",FLT_DIG, matrix[i * ncol + j]);
    }
    printf("\n");
  }
}
int main () {
    float *A, *b;
    /* Ax = b
     * Ax = xT(A)   //transpose
     * A is m x m   //matrix   
     * b is m x 1  //vector
     * x is m x 1   //vector
     */
    int m = 4;
    int scale = 1;

    A = (float*)mkl_malloc(m*m,32);
    for (int i = 0; i< m; i++){
        for (int j = 0; j < m; j++) {
            int p = i*m+j;
            A[p] = pow(i,j)+m*j+1;
        }
    }

    b = (float*)mkl_malloc(m,32);
    for (int i = 0; i < m; i++) {
        b[i]=pow(2,i+1);
        //b[i]=0.0;
    }

    int matrix_order = LAPACK_ROW_MAJOR;

    int ipiv[m];

    int info;
    printf("%s\n","starting"); 
    //transpose
    mkl_simatcopy('R','T',m,m,scale,A,m,m);
    info = LAPACKE_sgesv(matrix_order,m,1,A,m,ipiv,b,1);
    printf("%d\n",info); 
    print_matrix(b, m, 1 );
    return 0;
}