1 2 #include <petsc/private/kspimpl.h> 3 4 static PetscErrorCode KSPSetUp_CR(KSP ksp) 5 { 6 PetscErrorCode ierr; 7 8 PetscFunctionBegin; 9 if (ksp->pc_side == PC_RIGHT) SETERRQ(PetscObjectComm((PetscObject)ksp),PETSC_ERR_SUP,"no right preconditioning for KSPCR"); 10 else if (ksp->pc_side == PC_SYMMETRIC) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"no symmetric preconditioning for KSPCR"); 11 ierr = KSPSetWorkVecs(ksp,6);CHKERRQ(ierr); 12 PetscFunctionReturn(0); 13 } 14 15 static PetscErrorCode KSPSolve_CR(KSP ksp) 16 { 17 PetscErrorCode ierr; 18 PetscInt i = 0; 19 PetscReal dp; 20 PetscScalar ai, bi; 21 PetscScalar apq,btop, bbot; 22 Vec X,B,R,RT,P,AP,ART,Q; 23 Mat Amat, Pmat; 24 25 PetscFunctionBegin; 26 X = ksp->vec_sol; 27 B = ksp->vec_rhs; 28 R = ksp->work[0]; 29 RT = ksp->work[1]; 30 P = ksp->work[2]; 31 AP = ksp->work[3]; 32 ART = ksp->work[4]; 33 Q = ksp->work[5]; 34 35 /* R is the true residual norm, RT is the preconditioned residual norm */ 36 ierr = PCGetOperators(ksp->pc,&Amat,&Pmat);CHKERRQ(ierr); 37 if (!ksp->guess_zero) { 38 ierr = KSP_MatMult(ksp,Amat,X,R);CHKERRQ(ierr); /* R <- A*X */ 39 ierr = VecAYPX(R,-1.0,B);CHKERRQ(ierr); /* R <- B-R == B-A*X */ 40 } else { 41 ierr = VecCopy(B,R);CHKERRQ(ierr); /* R <- B (X is 0) */ 42 } 43 ierr = KSP_PCApply(ksp,R,P);CHKERRQ(ierr); /* P <- B*R */ 44 ierr = KSP_MatMult(ksp,Amat,P,AP);CHKERRQ(ierr); /* AP <- A*P */ 45 ierr = VecCopy(P,RT);CHKERRQ(ierr); /* RT <- P */ 46 ierr = VecCopy(AP,ART);CHKERRQ(ierr); /* ART <- AP */ 47 ierr = VecDotBegin(RT,ART,&btop);CHKERRQ(ierr); /* (RT,ART) */ 48 49 if (ksp->normtype == KSP_NORM_PRECONDITIONED) { 50 ierr = VecNormBegin(RT,NORM_2,&dp);CHKERRQ(ierr); /* dp <- RT'*RT */ 51 ierr = VecDotEnd (RT,ART,&btop);CHKERRQ(ierr); /* (RT,ART) */ 52 ierr = VecNormEnd (RT,NORM_2,&dp);CHKERRQ(ierr); /* dp <- RT'*RT */ 53 KSPCheckNorm(ksp,dp); 54 } else if (ksp->normtype == KSP_NORM_NONE) { 55 dp = 0.0; /* meaningless value that is passed to monitor and convergence test */ 56 } else if (ksp->normtype == KSP_NORM_UNPRECONDITIONED) { 57 ierr = VecNormBegin(R,NORM_2,&dp);CHKERRQ(ierr); /* dp <- R'*R */ 58 ierr = VecDotEnd (RT,ART,&btop);CHKERRQ(ierr); /* (RT,ART) */ 59 ierr = VecNormEnd (R,NORM_2,&dp);CHKERRQ(ierr); /* dp <- RT'*RT */ 60 KSPCheckNorm(ksp,dp); 61 } else if (ksp->normtype == KSP_NORM_NATURAL) { 62 ierr = VecDotEnd (RT,ART,&btop);CHKERRQ(ierr); /* (RT,ART) */ 63 dp = PetscSqrtReal(PetscAbsScalar(btop)); /* dp = sqrt(R,AR) */ 64 } else SETERRQ1(PetscObjectComm((PetscObject)ksp),PETSC_ERR_SUP,"KSPNormType of %d not supported",(int)ksp->normtype); 65 if (PetscAbsScalar(btop) < 0.0) { 66 ksp->reason = KSP_DIVERGED_INDEFINITE_MAT; 67 ierr = PetscInfo(ksp,"diverging due to indefinite or negative definite matrix\n");CHKERRQ(ierr); 68 PetscFunctionReturn(0); 69 } 70 71 ksp->its = 0; 72 ierr = KSPMonitor(ksp,0,dp);CHKERRQ(ierr); 73 ierr = PetscObjectSAWsTakeAccess((PetscObject)ksp);CHKERRQ(ierr); 74 ksp->rnorm = dp; 75 ierr = PetscObjectSAWsGrantAccess((PetscObject)ksp);CHKERRQ(ierr); 76 ierr = KSPLogResidualHistory(ksp,dp);CHKERRQ(ierr); 77 ierr = (*ksp->converged)(ksp,0,dp,&ksp->reason,ksp->cnvP);CHKERRQ(ierr); 78 if (ksp->reason) PetscFunctionReturn(0); 79 80 i = 0; 81 do { 82 ierr = KSP_PCApply(ksp,AP,Q);CHKERRQ(ierr); /* Q <- B* AP */ 83 84 ierr = VecDot(AP,Q,&apq);CHKERRQ(ierr); 85 KSPCheckDot(ksp,apq); 86 if (PetscRealPart(apq) <= 0.0) { 87 ksp->reason = KSP_DIVERGED_INDEFINITE_PC; 88 ierr = PetscInfo(ksp,"KSPSolve_CR:diverging due to indefinite or negative definite PC\n");CHKERRQ(ierr); 89 break; 90 } 91 ai = btop/apq; /* ai = (RT,ART)/(AP,Q) */ 92 93 ierr = VecAXPY(X,ai,P);CHKERRQ(ierr); /* X <- X + ai*P */ 94 ierr = VecAXPY(RT,-ai,Q);CHKERRQ(ierr); /* RT <- RT - ai*Q */ 95 ierr = KSP_MatMult(ksp,Amat,RT,ART);CHKERRQ(ierr); /* ART <- A*RT */ 96 bbot = btop; 97 ierr = VecDotBegin(RT,ART,&btop);CHKERRQ(ierr); 98 99 if (ksp->normtype == KSP_NORM_PRECONDITIONED) { 100 ierr = VecNormBegin(RT,NORM_2,&dp);CHKERRQ(ierr); /* dp <- || RT || */ 101 ierr = VecDotEnd (RT,ART,&btop);CHKERRQ(ierr); 102 ierr = VecNormEnd (RT,NORM_2,&dp);CHKERRQ(ierr); /* dp <- || RT || */ 103 KSPCheckNorm(ksp,dp); 104 } else if (ksp->normtype == KSP_NORM_NATURAL) { 105 ierr = VecDotEnd(RT,ART,&btop);CHKERRQ(ierr); 106 dp = PetscSqrtReal(PetscAbsScalar(btop)); /* dp = sqrt(R,AR) */ 107 } else if (ksp->normtype == KSP_NORM_NONE) { 108 ierr = VecDotEnd(RT,ART,&btop);CHKERRQ(ierr); 109 dp = 0.0; /* meaningless value that is passed to monitor and convergence test */ 110 } else if (ksp->normtype == KSP_NORM_UNPRECONDITIONED) { 111 ierr = VecAXPY(R,ai,AP);CHKERRQ(ierr); /* R <- R - ai*AP */ 112 ierr = VecNormBegin(R,NORM_2,&dp);CHKERRQ(ierr); /* dp <- R'*R */ 113 ierr = VecDotEnd (RT,ART,&btop);CHKERRQ(ierr); 114 ierr = VecNormEnd (R,NORM_2,&dp);CHKERRQ(ierr); /* dp <- R'*R */ 115 KSPCheckNorm(ksp,dp); 116 } else SETERRQ1(PetscObjectComm((PetscObject)ksp),PETSC_ERR_SUP,"KSPNormType of %d not supported",(int)ksp->normtype); 117 if (PetscAbsScalar(btop) < 0.0) { 118 ksp->reason = KSP_DIVERGED_INDEFINITE_MAT; 119 ierr = PetscInfo(ksp,"diverging due to indefinite or negative definite PC\n");CHKERRQ(ierr); 120 break; 121 } 122 123 ierr = PetscObjectSAWsTakeAccess((PetscObject)ksp);CHKERRQ(ierr); 124 ksp->its++; 125 ksp->rnorm = dp; 126 ierr = PetscObjectSAWsGrantAccess((PetscObject)ksp);CHKERRQ(ierr); 127 128 ierr = KSPLogResidualHistory(ksp,dp);CHKERRQ(ierr); 129 ierr = KSPMonitor(ksp,i+1,dp);CHKERRQ(ierr); 130 ierr = (*ksp->converged)(ksp,i+1,dp,&ksp->reason,ksp->cnvP);CHKERRQ(ierr); 131 if (ksp->reason) break; 132 133 bi = btop/bbot; 134 ierr = VecAYPX(P,bi,RT);CHKERRQ(ierr); /* P <- RT + Bi P */ 135 ierr = VecAYPX(AP,bi,ART);CHKERRQ(ierr); /* AP <- ART + Bi AP */ 136 i++; 137 } while (i<ksp->max_it); 138 if (i >= ksp->max_it) ksp->reason = KSP_DIVERGED_ITS; 139 PetscFunctionReturn(0); 140 } 141 142 143 /*MC 144 KSPCR - This code implements the (preconditioned) conjugate residuals method 145 146 Options Database Keys: 147 . see KSPSolve() 148 149 Level: beginner 150 151 Notes: 152 The operator and the preconditioner must be symmetric for this method. The 153 preconditioner must be POSITIVE-DEFINITE and the operator POSITIVE-SEMIDEFINITE. 154 Support only for left preconditioning. 155 156 References: 157 . 1. - Magnus R. Hestenes and Eduard Stiefel, Methods of Conjugate Gradients for Solving Linear Systems, 158 Journal of Research of the National Bureau of Standards Vol. 49, No. 6, December 1952 Research Paper 2379 159 160 .seealso: KSPCreate(), KSPSetType(), KSPType (for list of available types), KSP, KSPCG 161 M*/ 162 PETSC_EXTERN PetscErrorCode KSPCreate_CR(KSP ksp) 163 { 164 PetscErrorCode ierr; 165 166 PetscFunctionBegin; 167 ierr = KSPSetSupportedNorm(ksp,KSP_NORM_PRECONDITIONED,PC_LEFT,3);CHKERRQ(ierr); 168 ierr = KSPSetSupportedNorm(ksp,KSP_NORM_UNPRECONDITIONED,PC_LEFT,2);CHKERRQ(ierr); 169 ierr = KSPSetSupportedNorm(ksp,KSP_NORM_NATURAL,PC_LEFT,2);CHKERRQ(ierr); 170 ierr = KSPSetSupportedNorm(ksp,KSP_NORM_NONE,PC_LEFT,1);CHKERRQ(ierr); 171 172 ksp->ops->setup = KSPSetUp_CR; 173 ksp->ops->solve = KSPSolve_CR; 174 ksp->ops->destroy = KSPDestroyDefault; 175 ksp->ops->buildsolution = KSPBuildSolutionDefault; 176 ksp->ops->buildresidual = KSPBuildResidualDefault; 177 ksp->ops->setfromoptions = 0; 178 ksp->ops->view = 0; 179 PetscFunctionReturn(0); 180 } 181