the first call of mkl_dcsrsymv causes the memory access violation error.#define USE_INTEL_MKL // use Intel MKL#define EPS 1.0e-14void matrixclass::linbcg_symmetric(MatrixIndexType n, double b[], double x[], int
sparse_status_t res = mkl_sparse_d_mv (sparse_operation_t::SPARSE_OPERATION_NON_TRANSPOSE, 1.0, M, matrix_descr, V_in, 0.0, V_out); The return value is always sparse_status_t::SPARSE_STATUS_SUCCESS, but the values returned into V_out do not matc...
using mkl_set_num_threads2) with smaller matrixsizes leads to only one being utilizedand the wrong result being obtainedfrom the multiplication. It seems that the caseis that when the routine is called withmore thanone cpu, themkl_cspblas_dcsrsymv function decides whether or ...
using mkl_set_num_threads2) with smaller matrixsizes leads to only one being utilizedand the wrong result being obtainedfrom the multiplication. It seems that the caseis that when the routine is called withmore thanone cpu, themkl_cspblas_dcsrsymv function decides whether or ...