---
title: KmlScaissGmresPcSet?II
description: "设置迭代求解的预条件子相关参数，当前仅支持为block jacobi预条件子设置相关参数。"
url: https://www.hikunpeng.com/document/detail/zh/kunpengboostkithistory/240RC2/accel/kunpengaccel_kml_17_0090.html
sourcePath: /source/zh/kunpengboostkithistory/240RC2/accel/kunpengaccel_kml_17_0090.html
indexId: 61f9c9d97a9e9c87b8296272b74e7d6512a1636c9b4159ab8f3ddffed06d1b3a76
---
# KmlScaissGmresPcSet?II

设置迭代求解的预条件子相关参数，当前仅支持为block jacobi预条件子设置相关参数。

#### 接口定义

C Interface：

int KmlScaissGmresPcSetSII(KmlScasolverTask **handle, enum KmlSolverParam param, const int *data, int nd);

int KmlScaissGmresPcSetDII(KmlScasolverTask **handle, enum KmlSolverParam param, const int *data, int nd);


#### 参数

| 参数名 | 类型 | 描述 | 输入/输出 |
| --- | --- | --- | --- |
| handle | KmlScasolverTask \*\* | 求解器句柄，传入之前步骤的变量。 | 输入/输出 |
| param | enum KmlSolverParam | KMLSS\_BLOCK\_METHOD表示每个分块中的预条件子类型。 KMLSS\_NUM\_BLOCKS表示每个进程中的分块数量。 KMLSS\_BLOCK\_SIZES表示每个进程中的分块大小。 | 输入 |
| data | const int \* | 预条件子相关参数（如预条件子类型、分块数量、分块大小）。 当前可选择的预条件子类型： ILU预条件子：KMLSS\_ILU。 ICC预条件子：KMLSS\_ICC。 SOR预条件子：KMLSS\_SOR。 ILUT预条件子：KMLSS\_ILUT。 | 输入 |
| nd | int | data数组元素个数。 | 输入 |


#### 返回值

| 返回值 | 类型 | 描述 |
| --- | --- | --- |
| KMLSS\_NO\_ERROR | int | 正常执行。 |
| KMLSS\_DATA\_SIZE | int | 参数nd不等于1。 |
| KMLSS\_NULL\_ARGUMENT | int | handle，data中存在空参数。 |
| KMLSS\_BAD\_SELECTOR | int | param为无效参数。 |
| KMLSS\_BAD\_PRECONDITIONER | int | 选择的预条件子暂未实现。 |


#### 依赖

#include "kml_scaiss.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 mat_size = 8;
int n = mat_size / size;
int n_beg = n * rank;
if (n * size != mat_size && rank == (size - 1)) {
n = mat_size - n * rank;
}
int ja[26] = { 0, 3, 4, 1, 2, 3, 5, 1, 2, 7, 0, 1, 3, 6, 0, 4, 5, 1, 4, 5, 7, 3, 6, 2, 5, 7 };
double a[26] = { 1.0, 1.0, 2.0, 9.0, 2.0, 1.0, -3.0, 2.0, 3.0, 2.0, 1.0, 1.0, 9.0, -5.0, 2.0, 6.0, 1.0, -3.0, 1.0, 4.0, 1.0, -5.0, 7.0, 2.0, 1.0, 2.0 };
int ia[9] = { 0, 3, 7, 10, 14, 17, 21, 23, 26 };
int a_beg = ia[n_beg];
for (int i = n_beg; i < (n_beg + n + 1); i++) {
ia[i] -= a_beg;
}
/* Internal KML_SCAISS structure */
KmlScasolverTask *handle;
/* KML_SCAISS control parameters */
int error;            /* Output error handle */
/* Create data structures */
const double *a_holder = &a[a_beg];
const int *ja_holder = &ja[a_beg];
const int *ia_holder = &ia[n_beg];
error = KmlScaissGmresInitStripesDI(&handle, mat_size, 1, &n, &n_beg, &a_holder, &ja_holder, &ia_holder, MPI_COMM_WORLD);
double eps = 1e-4;
error = KmlScaissGmresSetDID(&handle, KMLSS_THRESHOLD, &eps, 1);
if (error != 0) {
printf("\nERROR in KmlScaissGmresSetDID with KMLSS_THRESHOLD: %d", error);
return 1;
}
int max_iters = 2000;
error = KmlScaissGmresSetDII(&handle, KMLSS_MAX_ITERATION_COUNT, &max_iters, 1);
if (error != 0) {
printf("\nERROR in KmlScaissGmresSetDII with KMLSS_MAX_ITERATION_COUNT: %d", error);
return 1;
}
int precond = KMLSS_BJACOBI;
error = KmlScaissGmresSetDII(&handle, KMLSS_PRECONDITIONER_TYPE, &precond, 1);
if (error != 0) {
printf("\nERROR in KmlScaissGmresSetDII with KMLSS_PRECONDITIONER_TYPE: %d", error);
return 1;
}
int num_blocks = 1;
error = KmlScaissGmresPcSetDII(&handle, KMLSS_NUM_BLOCKS, &num_blocks, 1);
if (error != 0) {
printf("ERROR in KmlScaissGmresPcSetDII: %d\n", error);
return 1;
}
int subprecond = KMLSS_ICC;
error = KmlScaissGmresPcSetDII(&handle, KMLSS_BLOCK_METHOD, &subprecond, 1);
if (error != 0) {
printf("ERROR in KmlScaissGmresPcSetDII: %d\n", error);
return 1;
}
```