KmlScadss(SPD/HPD)Factorize?I

实现对称正定矩阵的LLT分解或者Hermitian正定矩阵的LLH分解。

接口定义

C Interface：

int KmlScadssSpdFactorizeSI(KmlScasolverTask **pHandle);

int KmlScadssSpdFactorizeDI(KmlScasolverTask **pHandle);

int KmlScadssHpdFactorizeCI(KmlScasolverTask **pHandle);

int KmlScadssHpdFactorizeZI(KmlScasolverTask **pHandle);

参数

参数名	类型	描述	输入/输出
pHandle	KmlScasolverTask **	求解器句柄，传入之前步骤的变量。	输入

返回值

返回值	类型	描述
KMLSS_NO_ERROR	int	正常执行。
KMLSS_NULL_ARGUMENT	int	参数无效或格式存在问题。

依赖

#include "kml_scadss.h"

示例

C Interface：

MPI_Init(NULL, NULL);
int size, rank;
MPI_Comm_size(MPI_COMM_WORLD, &size);//获取总进程数
MPI_Comm_rank(MPI_COMM_WORLD, &rank);//获取当前进程标识

int n = 8;
int ia[9] = {0, 3, 7, 9, 11, 13, 15, 16, 17};
int ja[17] = {0, 3, 4, 1, 2, 3, 5, 2, 7, 3, 6, 4, 5, 5, 7, 6, 7};
double a[17] = {1.0, 1.0, 2.0, 9.0, 2.0, 1.0, -3.0, 3.0, 2.0, 9.0, -5.0, 6.0, 1.0, 4.0, 1.0, 7.0, 2.0};
KmlScasolverTask* matrix_handle = NULL;
int ierr = KmlScadssSpdInitDI(&matrix_handle, n, a, ja, ia, MPI_COMM_WORLD);
ierr = KmlScadssSpdAnalyzeDI(&matrix_handle);
ierr = KmlScadssSpdFactorizeDI(&matrix_handle);
if (ierr != 0) {
    printf("\nERROR in KmlScadssSpdFactorizeDI: %d", ierr);
}

父主题： SCADSS-LLT/LLH