Issues compiling MATLAB mex with Intel MKL PARDISO

Zaluski__Dylan — Sun, 03 Mar 2019 09:30:22 GMT

I am quite new to writing mex files, and to the MKL. I need to interface the PARDISO solver with a MATLAB code of mine, for solving of large finite element matrices. When compiling my mex code, I am having some issues. MATLAB gives me the error message:

Error using mex
Creating library fem_mex_pardiso.lib and object fem_mex_pardiso.exp
fem_mex_pardiso.obj : error LNK2019: unresolved external symbol pardiso referenced in function mexFunction
fem_mex_pardiso.mexw64 : fatal error LNK1120: 1 unresolved externals

This occurs when I submit the command

mex -R2018a fem_mex_pardiso.c -I'C:\Program Files (x86)\IntelSWTools\compilers_and_libraries_2019.2.190\windows\mkl\include\'

I am sure that it is that I am not linking something properly, but I honestly am quite new to interfacing MATLAB with external libraries.

Here is the c - code that gives the errors:

/*==========================================================
* fem_mex_pardiso.ca
*==========================================================*/

// Include headers
#include "mex.h"
#include "mkl.h"
#include "mkl_pardiso.h"
#include "mkl_types.h"

void mexFunction( int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{
    // Define variables
   double *a, *rf_glob, *u_qct, ddum;
    int num_rows, i;
    void *pt[64];
    MKL_INT *ia, *ja, n, iparm[64], maxfct, mnum, phase, error, msglvl, mtype, idum;

    // Argument checks
   if (nlhs > 1)
   {
       mexErrMsgTxt("Too many outputs!");
   }
   if (nrhs != 4) // Should be 3 matrices for the CSR3 format of k_glob, and one column vector for rf_glob
   {
       mexErrMsgTxt("Need four inputs! (3 vectors for the sparse stiffness matrix, and one for the load vector)");
   }

    // Get the sparse matrix inputs
   ia = mxGetInt32s(prhs[0]); // Contains the indices of the first non-zero element in each row of a. Note, the last index of ia is the number of non-zero elements in a, plus one
   ja = mxGetInt32s(prhs[1]); // Contains the global column indices of every element in a
   a = mxGetDoubles(prhs[2]); // Non-zero elements of stiffness matrix

    // Number of equations is equal to length of the reaction force vector
   rf_glob = mxGetDoubles(prhs[3]);
   num_rows = mxGetM(prhs[3]);
   n = num_rows;

    // Configure out-of-core settings
   //setenv('MKL_PARDISO_OOC_PATH', 'C:\Users\dylan\Documents\School\Masters\01_Main_MSc\02_Codes\13_INTEL_MKL_Matlab\OutOfCore_Memory\ooc_file');
   //setenv('MKL_PARDISO_OOC_MAX_CORE_SIZE', '12288');
   //setenv('MKL_PARDISO_OOC_MAX_SWAP_SIZE', '8192');
   //setenv('MKL_PARDISO_OOC_KEEP_FILE', '0');

    // Setup solver
       // Iniitialize values for solver parameters
   maxfct = 1; // Maximal number of factors in memory
   mnum = 1; // Number of matrices to solve
   mtype = -2; // Matrix type (Here, real and symmetric positive indefinite. For real and symmetric positive definite, use 2)
   // Initialize the parameters
   for (i = 0; i < 64; i++)
   {
       iparm = 0;
   }
    iparm[0] = 1; // Not using default solver parameters
   iparm[1] = 3; // Fill-in reducing ordering from METIS, parallelized with OpenMP
   iparm[3] = 0; // Not using an iterative algorithm
   iparm[4] = 0; // Not using a user-specified fill-in reducing permutation
   iparm[5] = 0; // Write solution into u_qct
   iparm[7] = 2; // Maximum number of iterative refinement steps
   iparm[9] = 13; // Perturb pivot elements by 1E-13 to deal with near-zero or zero pivots
   iparm[10] = 0; // No scaling (because we will be reording to a symmetric matrix)
   iparm[12] = 0; // No symmetric weighted matching
   iparm[13] = 0; // Zero perturbed pivots
   iparm[17] = -1; // Number of non-zeros in the factor LU
   iparm[18] = -1; // Mfloprs for LU factorization
   iparm[19] = 0; // Number of CG iterations
   iparm[34] = 1; // Use C-style indexing for ia and ja arrays
    iparm[60] = 1; // Use out-of-core memory if out of RAM
   msglvl = 0; // Do not print statistical information file
   error = 0; // Initialize error flag
       // Initialize the solver pointer
   for (i = 0; i < 64; i++)
   {
       pt = 0;
   }

    // Begin solving

   // Step 1 - Reordering and symbolic factorization
   phase = 11;
   PARDISO (pt, &maxfct, &mnum, &mtype, &phase, &n, a, ia, ja, &idum, &nrhs, iparm, &msglvl, &ddum, &ddum, &error);
   if (error != 0)
   {
       printf("\n ERROR during symbolic factorization: %d", error);
       exit(1);
   }
   printf("\nReordering completed ... ");
   printf("\nNumber of nonzeros in factors = %d", iparm[17]);
   printf("\nNumber of factorization MFLOPS = %d", iparm[18]);
}

Any help is greatly appreciated. Thankyou!

Dylan, I guess, the cause of

Gennady_F_Intel — Fri, 08 Mar 2019 05:43:51 GMT

Dylan, I guess, the cause of the problem with regard to how do you link this case.I would recommend you to look at this article (https://software.intel.com/en-us/articles/intel-math-kernel-library-intel-mkl-for-windows-using-intel-mkl-in-matlab-executable-mex-files) .This is pretty obsolete (we should update the article. tbd) but you may catch some info how to link and call..

topic Dylan, I guess, the cause of in Intel® oneAPI Math Kernel Library

Issues compiling MATLAB mex with Intel MKL PARDISO

Dylan, I guess, the cause of