Actual source code: klu.c
petsc-3.6.1 2015-08-06
2: /*
3: Provides an interface to the KLUv1.2 sparse solver
5: When build with PETSC_USE_64BIT_INDICES this will use SuiteSparse_long as the
6: integer type in KLU, otherwise it will use int. This means
7: all integers in this file are simply declared as PetscInt. Also it means
8: that KLU SuiteSparse_long version MUST be built with 64 bit integers when used.
10: */
11: #include <../src/mat/impls/aij/seq/aij.h>
13: #if defined(PETSC_USE_64BIT_INDICES)
14: #define klu_K_defaults klu_l_defaults
15: #define klu_K_analyze klu_l_analyze
16: #define klu_K_analyze_given klu_l_analyze_given
17: #define klu_K_free_symbolic klu_l_free_symbolic
18: #define klu_K_free_numeric klu_l_free_numeric
19: #define klu_K_common klu_l_common
20: #define klu_K_symbolic klu_l_symbolic
21: #define klu_K_numeric klu_l_numeric
22: #if defined(PETSC_USE_COMPLEX)
23: #define klu_K_factor klu_zl_factor
24: #define klu_K_solve klu_zl_solve
25: #define klu_K_tsolve klu_zl_tsolve
26: #define klu_K_refactor klu_zl_refactor
27: #define klu_K_sort klu_zl_sort
28: #define klu_K_flops klu_zl_flops
29: #define klu_K_rgrowth klu_zl_rgrowth
30: #define klu_K_condest klu_zl_condest
31: #define klu_K_rcond klu_zl_rcond
32: #define klu_K_scale klu_zl_scale
33: #else
34: #define klu_K_factor klu_l_factor
35: #define klu_K_solve klu_l_solve
36: #define klu_K_tsolve klu_l_tsolve
37: #define klu_K_refactor klu_l_refactor
38: #define klu_K_sort klu_l_sort
39: #define klu_K_flops klu_l_flops
40: #define klu_K_rgrowth klu_l_rgrowth
41: #define klu_K_condest klu_l_condest
42: #define klu_K_rcond klu_l_rcond
43: #define klu_K_scale klu_l_scale
44: #endif
45: #else
46: #define klu_K_defaults klu_defaults
47: #define klu_K_analyze klu_analyze
48: #define klu_K_analyze_given klu_analyze_given
49: #define klu_K_free_symbolic klu_free_symbolic
50: #define klu_K_free_numeric klu_free_numeric
51: #define klu_K_common klu_common
52: #define klu_K_symbolic klu_symbolic
53: #define klu_K_numeric klu_numeric
54: #if defined(PETSC_USE_COMPLEX)
55: #define klu_K_factor klu_z_factor
56: #define klu_K_solve klu_z_solve
57: #define klu_K_tsolve klu_z_tsolve
58: #define klu_K_refactor klu_z_refactor
59: #define klu_K_sort klu_z_sort
60: #define klu_K_flops klu_z_flops
61: #define klu_K_rgrowth klu_z_rgrowth
62: #define klu_K_condest klu_z_condest
63: #define klu_K_rcond klu_z_rcond
64: #define klu_K_scale klu_z_scale
65: #else
66: #define klu_K_factor klu_factor
67: #define klu_K_solve klu_solve
68: #define klu_K_tsolve klu_tsolve
69: #define klu_K_refactor klu_refactor
70: #define klu_K_sort klu_sort
71: #define klu_K_flops klu_flops
72: #define klu_K_rgrowth klu_rgrowth
73: #define klu_K_condest klu_condest
74: #define klu_K_rcond klu_rcond
75: #define klu_K_scale klu_scale
76: #endif
77: #endif
80: #define SuiteSparse_long long long
81: #define SuiteSparse_long_max LONG_LONG_MAX
82: #define SuiteSparse_long_id "%lld"
84: EXTERN_C_BEGIN
85: #include <klu.h>
86: EXTERN_C_END
88: static const char *KluOrderingTypes[] = {"AMD","COLAMD","PETSC"};
89: static const char *scale[] ={"NONE","SUM","MAX"};
91: typedef struct {
92: klu_K_common Common;
93: klu_K_symbolic *Symbolic;
94: klu_K_numeric *Numeric;
95: PetscInt *perm_c,*perm_r;
96: MatStructure flg;
97: PetscBool PetscMatOrdering;
99: /* Flag to clean up KLU objects during Destroy */
100: PetscBool CleanUpKLU;
101: } Mat_KLU;
105: static PetscErrorCode MatDestroy_KLU(Mat A)
106: {
108: Mat_KLU *lu=(Mat_KLU*)A->spptr;
111: if (lu && lu->CleanUpKLU) {
112: klu_K_free_symbolic(&lu->Symbolic,&lu->Common);
113: klu_K_free_numeric(&lu->Numeric,&lu->Common);
114: PetscFree2(lu->perm_r,lu->perm_c);
115: }
116: PetscFree(A->spptr);
117: MatDestroy_SeqAIJ(A);
118: return(0);
119: }
123: static PetscErrorCode MatSolveTranspose_KLU(Mat A,Vec b,Vec x)
124: {
125: Mat_KLU *lu = (Mat_KLU*)A->spptr;
126: PetscScalar *xa;
128: PetscInt status;
131: /* KLU uses a column major format, solve Ax = b by klu_*_solve */
132: /* ----------------------------------*/
133: VecCopy(b,x); /* klu_solve stores the solution in rhs */
134: VecGetArray(x,&xa);
135: status = klu_K_solve(lu->Symbolic,lu->Numeric,A->rmap->n,1,(PetscReal*)xa,&lu->Common);
136: if (status != 1) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_LIB,"KLU Solve failed");
137: VecRestoreArray(x,&xa);
138: return(0);
139: }
143: static PetscErrorCode MatSolve_KLU(Mat A,Vec b,Vec x)
144: {
145: Mat_KLU *lu = (Mat_KLU*)A->spptr;
146: PetscScalar *xa;
148: PetscInt status;
151: /* KLU uses a column major format, solve A^Tx = b by klu_*_tsolve */
152: /* ----------------------------------*/
153: VecCopy(b,x); /* klu_solve stores the solution in rhs */
154: VecGetArray(x,&xa);
155: #if defined(PETSC_USE_COMPLEX)
156: PetscInt conj_solve=1;
157: status = klu_K_tsolve(lu->Symbolic,lu->Numeric,A->rmap->n,1,(PetscReal*)xa,conj_solve,&lu->Common); /* conjugate solve */
158: #else
159: status = klu_K_tsolve(lu->Symbolic,lu->Numeric,A->rmap->n,1,xa,&lu->Common);
160: #endif
161: if (status != 1) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_LIB,"KLU Solve failed");
162: VecRestoreArray(x,&xa);
163: return(0);
164: }
168: static PetscErrorCode MatLUFactorNumeric_KLU(Mat F,Mat A,const MatFactorInfo *info)
169: {
170: Mat_KLU *lu = (Mat_KLU*)(F)->spptr;
171: Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data;
172: PetscInt *ai = a->i,*aj=a->j;
173: PetscScalar *av = a->a;
176: /* numeric factorization of A' */
177: /* ----------------------------*/
179: if (lu->flg == SAME_NONZERO_PATTERN && lu->Numeric) {
180: klu_K_free_numeric(&lu->Numeric,&lu->Common);
181: }
182: lu->Numeric = klu_K_factor(ai,aj,(PetscReal*)av,lu->Symbolic,&lu->Common);
183: if(!lu->Numeric) {
184: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_LIB,"KLU Numeric factorization failed");
185: }
187: lu->flg = SAME_NONZERO_PATTERN;
188: lu->CleanUpKLU = PETSC_TRUE;
189: F->ops->solve = MatSolve_KLU;
190: F->ops->solvetranspose = MatSolveTranspose_KLU;
191: return(0);
192: }
196: static PetscErrorCode MatLUFactorSymbolic_KLU(Mat F,Mat A,IS r,IS c,const MatFactorInfo *info)
197: {
198: Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data;
199: Mat_KLU *lu = (Mat_KLU*)(F->spptr);
201: PetscInt i,*ai = a->i,*aj = a->j,m=A->rmap->n,n=A->cmap->n;
202: const PetscInt *ra,*ca;
205: if (lu->PetscMatOrdering) {
206: ISGetIndices(r,&ra);
207: ISGetIndices(c,&ca);
208: PetscMalloc2(m,&lu->perm_r,n,&lu->perm_c);
209: /* we cannot simply memcpy on 64 bit archs */
210: for (i = 0; i < m; i++) lu->perm_r[i] = ra[i];
211: for (i = 0; i < n; i++) lu->perm_c[i] = ca[i];
212: ISRestoreIndices(r,&ra);
213: ISRestoreIndices(c,&ca);
214: }
216: /* symbolic factorization of A' */
217: /* ---------------------------------------------------------------------- */
218: if (lu->PetscMatOrdering) { /* use Petsc ordering */
219: lu->Symbolic = klu_K_analyze_given(n,ai,aj,lu->perm_c,lu->perm_r,&lu->Common);
220: } else { /* use klu internal ordering */
221: lu->Symbolic = klu_K_analyze(n,ai,aj,&lu->Common);
222: }
223: if (!lu->Symbolic) {
224: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_LIB,"KLU Symbolic Factorization failed");
225: }
227: lu->flg = DIFFERENT_NONZERO_PATTERN;
228: lu->CleanUpKLU = PETSC_TRUE;
229: (F)->ops->lufactornumeric = MatLUFactorNumeric_KLU;
230: return(0);
231: }
235: static PetscErrorCode MatFactorInfo_KLU(Mat A,PetscViewer viewer)
236: {
237: Mat_KLU *lu= (Mat_KLU*)A->spptr;
238: klu_K_numeric *Numeric=(klu_K_numeric*)lu->Numeric;
242: /* check if matrix is KLU type */
243: if (A->ops->solve != MatSolve_KLU) return(0);
245: PetscViewerASCIIPrintf(viewer,"KLU stats:\n");
246: PetscViewerASCIIPrintf(viewer," Number of diagonal blocks: %d\n",Numeric->nblocks);
247: PetscViewerASCIIPrintf(viewer," Total nonzeros=%d\n",Numeric->lnz+Numeric->unz);
249: PetscViewerASCIIPrintf(viewer,"KLU runtime parameters:\n");
251: /* Control parameters used by numeric factorization */
252: PetscViewerASCIIPrintf(viewer," Partial pivoting tolerance: %g\n",lu->Common.tol);
253: /* BTF preordering */
254: PetscViewerASCIIPrintf(viewer," BTF preordering enabled: %d\n",lu->Common.btf);
255: /* mat ordering */
256: if (!lu->PetscMatOrdering) {
257: PetscViewerASCIIPrintf(viewer," Ordering: %s (not using the PETSc ordering)\n",KluOrderingTypes[(int)lu->Common.ordering]);
258: } else {
259: PetscViewerASCIIPrintf(viewer," Using PETSc ordering\n");
260: }
261: /* matrix row scaling */
262: PetscViewerASCIIPrintf(viewer, " Matrix row scaling: %s\n",scale[(int)lu->Common.scale]);
263: return(0);
264: }
268: static PetscErrorCode MatView_KLU(Mat A,PetscViewer viewer)
269: {
270: PetscErrorCode ierr;
271: PetscBool iascii;
272: PetscViewerFormat format;
275: MatView_SeqAIJ(A,viewer);
277: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);
278: if (iascii) {
279: PetscViewerGetFormat(viewer,&format);
280: if (format == PETSC_VIEWER_ASCII_INFO) {
281: MatFactorInfo_KLU(A,viewer);
282: }
283: }
284: return(0);
285: }
289: PetscErrorCode MatFactorGetSolverPackage_seqaij_klu(Mat A,const MatSolverPackage *type)
290: {
292: *type = MATSOLVERKLU;
293: return(0);
294: }
297: /*MC
298: MATSOLVERKLU = "klu" - A matrix type providing direct solvers (LU) for sequential matrices
299: via the external package KLU.
301: ./configure --download-suitesparse to install PETSc to use KLU
303: Use -pc_type lu -pc_factor_mat_solver_package klu to us this direct solver
305: Consult KLU documentation for more information on the options database keys below.
307: Options Database Keys:
308: + -mat_klu_pivot_tol <0.001> - Partial pivoting tolerance
309: . -mat_klu_use_btf <1> - Use BTF preordering
310: . -mat_klu_ordering <AMD> - KLU reordering scheme to reduce fill-in (choose one of) AMD COLAMD PETSC
311: - -mat_klu_row_scale <NONE> - Matrix row scaling (choose one of) NONE SUM MAX
313: Note: KLU is part of SuiteSparse http://faculty.cse.tamu.edu/davis/suitesparse.html
315: Level: beginner
317: .seealso: PCLU, MATSOLVERUMFPACK, MATSOLVERCHOLMOD, PCFactorSetMatSolverPackage(), MatSolverPackage
318: M*/
322: PETSC_EXTERN PetscErrorCode MatGetFactor_seqaij_klu(Mat A,MatFactorType ftype,Mat *F)
323: {
324: Mat B;
325: Mat_KLU *lu;
327: PetscInt m=A->rmap->n,n=A->cmap->n,idx,status;
328: PetscBool flg;
331: /* Create the factorization matrix F */
332: MatCreate(PetscObjectComm((PetscObject)A),&B);
333: MatSetSizes(B,PETSC_DECIDE,PETSC_DECIDE,m,n);
334: MatSetType(B,((PetscObject)A)->type_name);
335: MatSeqAIJSetPreallocation(B,0,NULL);
336: PetscNewLog(B,&lu);
338: B->spptr = lu;
339: B->ops->lufactorsymbolic = MatLUFactorSymbolic_KLU;
340: B->ops->destroy = MatDestroy_KLU;
341: B->ops->view = MatView_KLU;
343: PetscObjectComposeFunction((PetscObject)B,"MatFactorGetSolverPackage_C",MatFactorGetSolverPackage_seqaij_klu);
345: B->factortype = MAT_FACTOR_LU;
346: B->assembled = PETSC_TRUE; /* required by -ksp_view */
347: B->preallocated = PETSC_TRUE;
349: /* initializations */
350: /* ------------------------------------------------*/
351: /* get the default control parameters */
352: status = klu_K_defaults(&lu->Common);
353: if(status <= 0) {
354: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_LIB,"KLU Initialization failed");
355: }
356: lu->Common.scale = 0; /* No row scaling */
358: PetscOptionsBegin(PetscObjectComm((PetscObject)A),((PetscObject)A)->prefix,"KLU Options","Mat");
359: /* Partial pivoting tolerance */
360: PetscOptionsReal("-mat_klu_pivot_tol","Partial pivoting tolerance","None",lu->Common.tol,&lu->Common.tol,NULL);
361: /* BTF pre-ordering */
362: PetscOptionsInt("-mat_klu_use_btf","Enable BTF preordering","None",lu->Common.btf,&lu->Common.btf,NULL);
363: /* Matrix reordering */
364: PetscOptionsEList("-mat_klu_ordering","Internal ordering method","None",KluOrderingTypes,sizeof(KluOrderingTypes)/sizeof(KluOrderingTypes[0]),KluOrderingTypes[0],&idx,&flg);
365: if (flg) {
366: if ((int)idx == 2) lu->PetscMatOrdering = PETSC_TRUE; /* use Petsc mat ordering (note: size is for the transpose, and PETSc r = Klu perm_c) */
367: else lu->Common.ordering = (int)idx;
368: }
369: /* Matrix row scaling */
370: PetscOptionsEList("-mat_klu_row_scale","Matrix row scaling","None",scale,3,scale[0],&idx,&flg);
371: PetscOptionsEnd();
372: *F = B;
373: return(0);
374: }