static char help[] = "Test file for the PCFactorSetShiftType()\n"; /* * Test file for the PCFactorSetShiftType() routine or -pc_factor_shift_type POSITIVE_DEFINITE option. * The test matrix is the example from Kershaw's paper [J.Comp.Phys 1978] * of a positive definite matrix for which ILU(0) will give a negative pivot. * This means that the CG method will break down; the Manteuffel shift * [Math. Comp. 1980] repairs this. * * Run the executable twice: * 1/ without options: the iterative method diverges because of an * indefinite preconditioner * 2/ with -pc_factor_shift_type POSITIVE_DEFINITE option (or comment in the PCFactorSetShiftType() line below): * the method will now successfully converge. * * Contributed by Victor Eijkhout 2003. */ #include int main(int argc,char **argv) { KSP solver; PC prec; Mat A,M; Vec X,B,D; MPI_Comm comm; PetscScalar v; KSPConvergedReason reason; PetscInt i,j,its; PetscErrorCode ierr; ierr = PetscInitialize(&argc,&argv,0,help);if (ierr) return ierr; comm = MPI_COMM_SELF; /* * Construct the Kershaw matrix * and a suitable rhs / initial guess */ ierr = MatCreateSeqAIJ(comm,4,4,4,0,&A);CHKERRQ(ierr); ierr = VecCreateSeq(comm,4,&B);CHKERRQ(ierr); ierr = VecDuplicate(B,&X);CHKERRQ(ierr); for (i=0; i<4; i++) { v = 3; ierr = MatSetValues(A,1,&i,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); v = 1; ierr = VecSetValues(B,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); ierr = VecSetValues(X,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); } i=0; v=0; ierr = VecSetValues(X,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); for (i=0; i<3; i++) { v = -2; j=i+1; ierr = MatSetValues(A,1,&i,1,&j,&v,INSERT_VALUES);CHKERRQ(ierr); ierr = MatSetValues(A,1,&j,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); } i=0; j=3; v=2; ierr = MatSetValues(A,1,&i,1,&j,&v,INSERT_VALUES);CHKERRQ(ierr); ierr = MatSetValues(A,1,&j,1,&i,&v,INSERT_VALUES);CHKERRQ(ierr); ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); ierr = VecAssemblyBegin(B);CHKERRQ(ierr); ierr = VecAssemblyEnd(B);CHKERRQ(ierr); /* * A Conjugate Gradient method * with ILU(0) preconditioning */ ierr = KSPCreate(comm,&solver);CHKERRQ(ierr); ierr = KSPSetOperators(solver,A,A);CHKERRQ(ierr); ierr = KSPSetType(solver,KSPCG);CHKERRQ(ierr); ierr = KSPSetInitialGuessNonzero(solver,PETSC_TRUE);CHKERRQ(ierr); /* * ILU preconditioner; * this will break down unless you add the Shift line, * or use the -pc_factor_shift_positive_definite option */ ierr = KSPGetPC(solver,&prec);CHKERRQ(ierr); ierr = PCSetType(prec,PCILU);CHKERRQ(ierr); /* ierr = PCFactorSetShiftType(prec,MAT_SHIFT_POSITIVE_DEFINITE);CHKERRQ(ierr); */ ierr = KSPSetFromOptions(solver);CHKERRQ(ierr); ierr = KSPSetUp(solver);CHKERRQ(ierr); /* * Now that the factorisation is done, show the pivots; * note that the last one is negative. This in itself is not an error, * but it will make the iterative method diverge. */ ierr = PCFactorGetMatrix(prec,&M);CHKERRQ(ierr); ierr = VecDuplicate(B,&D);CHKERRQ(ierr); ierr = MatGetDiagonal(M,D);CHKERRQ(ierr); /* * Solve the system; * without the shift this will diverge with * an indefinite preconditioner */ ierr = KSPSolve(solver,B,X);CHKERRQ(ierr); ierr = KSPGetConvergedReason(solver,&reason);CHKERRQ(ierr); if (reason==KSP_DIVERGED_INDEFINITE_PC) { ierr = PetscPrintf(PETSC_COMM_WORLD,"\nDivergence because of indefinite preconditioner;\n");CHKERRQ(ierr); ierr = PetscPrintf(PETSC_COMM_WORLD,"Run the executable again but with '-pc_factor_shift_type POSITIVE_DEFINITE' option.\n");CHKERRQ(ierr); } else if (reason<0) { ierr = PetscPrintf(PETSC_COMM_WORLD,"\nOther kind of divergence: this should not happen.\n");CHKERRQ(ierr); } else { ierr = KSPGetIterationNumber(solver,&its);CHKERRQ(ierr); } ierr = VecDestroy(&X);CHKERRQ(ierr); ierr = VecDestroy(&B);CHKERRQ(ierr); ierr = VecDestroy(&D);CHKERRQ(ierr); ierr = MatDestroy(&A);CHKERRQ(ierr); ierr = KSPDestroy(&solver);CHKERRQ(ierr); ierr = PetscFinalize(); return ierr; } /*TEST test: args: -pc_factor_shift_type positive_definite TEST*/