Actual source code: ex128.c
petsc-3.6.1 2015-08-06
2: static char help[] = "Tests ILU and ICC factorization with and without matrix ordering on seqsbaij format. Modified from ex30.c\n\
3: Input parameters are:\n\
4: -lf <level> : level of fill for ILU (default is 0)\n\
5: -lu : use full LU or Cholesky factorization\n\
6: -m <value>,-n <value> : grid dimensions\n\
7: Note that most users should employ the KSP interface to the\n\
8: linear solvers instead of using the factorization routines\n\
9: directly.\n\n";
11: #include <petscmat.h>
15: int main(int argc,char **args)
16: {
17: Mat C,sC,sA;
18: PetscInt i,j,m = 5,n = 5,Ii,J,lf = 0;
20: PetscBool CHOLESKY=PETSC_FALSE,TRIANGULAR=PETSC_FALSE,flg;
21: PetscScalar v;
22: IS row,col;
23: MatFactorInfo info;
24: Vec x,y,b,ytmp;
25: PetscReal norm2;
26: PetscRandom rdm;
27: PetscMPIInt size;
29: PetscInitialize(&argc,&args,(char*)0,help);
30: MPI_Comm_size(PETSC_COMM_WORLD,&size);
31: if (size != 1) SETERRQ(PETSC_COMM_WORLD,PETSC_ERR_SUP,"This is a uniprocessor example only!");
32: PetscOptionsGetInt(NULL,"-m",&m,NULL);
33: PetscOptionsGetInt(NULL,"-n",&n,NULL);
34: PetscOptionsGetInt(NULL,"-lf",&lf,NULL);
36: MatCreate(PETSC_COMM_SELF,&C);
37: MatSetSizes(C,m*n,m*n,m*n,m*n);
38: MatSetFromOptions(C);
39: MatSetUp(C);
41: /* Create matrix C in seqaij format and sC in seqsbaij. (This is five-point stencil with some extra elements) */
42: for (i=0; i<m; i++) {
43: for (j=0; j<n; j++) {
44: v = -1.0; Ii = j + n*i;
45: J = Ii - n; if (J>=0) {MatSetValues(C,1,&Ii,1,&J,&v,INSERT_VALUES);}
46: J = Ii + n; if (J<m*n) {MatSetValues(C,1,&Ii,1,&J,&v,INSERT_VALUES);}
47: J = Ii - 1; if (J>=0) {MatSetValues(C,1,&Ii,1,&J,&v,INSERT_VALUES);}
48: J = Ii + 1; if (J<m*n) {MatSetValues(C,1,&Ii,1,&J,&v,INSERT_VALUES);}
49: v = 4.0; MatSetValues(C,1,&Ii,1,&Ii,&v,INSERT_VALUES);
50: }
51: }
52: MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);
53: MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);
55: MatIsSymmetric(C,0.0,&flg);
56: if (!flg) SETERRQ(PETSC_COMM_SELF,1,"C is non-symmetric");
57: MatConvert(C,MATSEQSBAIJ,MAT_INITIAL_MATRIX,&sC);
59: /* Create vectors for error checking */
60: MatCreateVecs(C,&x,&b);
61: VecDuplicate(x,&y);
62: VecDuplicate(x,&ytmp);
63: PetscRandomCreate(PETSC_COMM_SELF,&rdm);
64: PetscRandomSetFromOptions(rdm);
65: VecSetRandom(x,rdm);
66: MatMult(C,x,b);
68: MatGetOrdering(C,MATORDERINGNATURAL,&row,&col);
70: /* Compute CHOLESKY or ICC factor sA */
71: MatFactorInfoInitialize(&info);
73: info.fill = 1.0;
74: info.diagonal_fill = 0;
75: info.zeropivot = 0.0;
77: PetscOptionsHasName(NULL,"-cholesky",&CHOLESKY);
78: if (CHOLESKY) {
79: printf("Test CHOLESKY...\n");
80: MatGetFactor(sC,MATSOLVERPETSC,MAT_FACTOR_CHOLESKY,&sA);
81: MatCholeskyFactorSymbolic(sA,sC,row,&info);
82: } else {
83: printf("Test ICC...\n");
84: info.levels = lf;
86: MatGetFactor(sC,MATSOLVERPETSC,MAT_FACTOR_ICC,&sA);
87: MatICCFactorSymbolic(sA,sC,row,&info);
88: }
89: MatCholeskyFactorNumeric(sA,sC,&info);
91: /* test MatForwardSolve() and MatBackwardSolve() with matrix reordering on aij matrix C */
92: if (CHOLESKY) {
93: PetscOptionsHasName(NULL,"-triangular_solve",&TRIANGULAR);
94: if (TRIANGULAR) {
95: printf("Test MatForwardSolve...\n");
96: MatForwardSolve(sA,b,ytmp);
97: printf("Test MatBackwardSolve...\n");
98: MatBackwardSolve(sA,ytmp,y);
99: VecAXPY(y,-1.0,x);
100: VecNorm(y,NORM_2,&norm2);
101: if (norm2 > 1.e-14) {
102: PetscPrintf(PETSC_COMM_SELF,"MatForwardSolve and BackwardSolve: Norm of error=%g\n",(double)norm2);
103: }
104: }
105: }
107: MatSolve(sA,b,y);
108: MatDestroy(&sC);
109: MatDestroy(&sA);
110: VecAXPY(y,-1.0,x);
111: VecNorm(y,NORM_2,&norm2);
112: if (lf == -1 && norm2 > 1.e-14) {
113: PetscPrintf(PETSC_COMM_SELF, " reordered SEQAIJ: Cholesky/ICC levels %D, residual %g\n",lf,(double)norm2);
114: }
116: /* Free data structures */
117: MatDestroy(&C);
118: ISDestroy(&row);
119: ISDestroy(&col);
120: PetscRandomDestroy(&rdm);
121: VecDestroy(&x);
122: VecDestroy(&y);
123: VecDestroy(&ytmp);
124: VecDestroy(&b);
125: PetscFinalize();
126: return 0;
127: }