Actual source code: superlu.c
petsc-3.11.4 2019-09-28
2: /* --------------------------------------------------------------------
4: This file implements a subclass of the SeqAIJ matrix class that uses
5: the SuperLU sparse solver.
6: */
8: /*
9: Defines the data structure for the base matrix type (SeqAIJ)
10: */
11: #include <../src/mat/impls/aij/seq/aij.h>
13: /*
14: SuperLU include files
15: */
16: EXTERN_C_BEGIN
17: #if defined(PETSC_USE_COMPLEX)
18: #if defined(PETSC_USE_REAL_SINGLE)
19: #include <slu_cdefs.h>
20: #else
21: #include <slu_zdefs.h>
22: #endif
23: #else
24: #if defined(PETSC_USE_REAL_SINGLE)
25: #include <slu_sdefs.h>
26: #else
27: #include <slu_ddefs.h>
28: #endif
29: #endif
30: #include <slu_util.h>
31: EXTERN_C_END
33: /*
34: This is the data that defines the SuperLU factored matrix type
35: */
36: typedef struct {
37: SuperMatrix A,L,U,B,X;
38: superlu_options_t options;
39: PetscInt *perm_c; /* column permutation vector */
40: PetscInt *perm_r; /* row permutations from partial pivoting */
41: PetscInt *etree;
42: PetscReal *R, *C;
43: char equed[1];
44: PetscInt lwork;
45: void *work;
46: PetscReal rpg, rcond;
47: mem_usage_t mem_usage;
48: MatStructure flg;
49: SuperLUStat_t stat;
50: Mat A_dup;
51: PetscScalar *rhs_dup;
52: GlobalLU_t Glu;
53: PetscBool needconversion;
55: /* Flag to clean up (non-global) SuperLU objects during Destroy */
56: PetscBool CleanUpSuperLU;
57: } Mat_SuperLU;
59: /*
60: Utility function
61: */
62: static PetscErrorCode MatView_Info_SuperLU(Mat A,PetscViewer viewer)
63: {
64: Mat_SuperLU *lu= (Mat_SuperLU*)A->data;
65: PetscErrorCode ierr;
66: superlu_options_t options;
69: options = lu->options;
71: PetscViewerASCIIPrintf(viewer,"SuperLU run parameters:\n");
72: PetscViewerASCIIPrintf(viewer," Equil: %s\n",(options.Equil != NO) ? "YES" : "NO");
73: PetscViewerASCIIPrintf(viewer," ColPerm: %D\n",options.ColPerm);
74: PetscViewerASCIIPrintf(viewer," IterRefine: %D\n",options.IterRefine);
75: PetscViewerASCIIPrintf(viewer," SymmetricMode: %s\n",(options.SymmetricMode != NO) ? "YES" : "NO");
76: PetscViewerASCIIPrintf(viewer," DiagPivotThresh: %g\n",options.DiagPivotThresh);
77: PetscViewerASCIIPrintf(viewer," PivotGrowth: %s\n",(options.PivotGrowth != NO) ? "YES" : "NO");
78: PetscViewerASCIIPrintf(viewer," ConditionNumber: %s\n",(options.ConditionNumber != NO) ? "YES" : "NO");
79: PetscViewerASCIIPrintf(viewer," RowPerm: %D\n",options.RowPerm);
80: PetscViewerASCIIPrintf(viewer," ReplaceTinyPivot: %s\n",(options.ReplaceTinyPivot != NO) ? "YES" : "NO");
81: PetscViewerASCIIPrintf(viewer," PrintStat: %s\n",(options.PrintStat != NO) ? "YES" : "NO");
82: PetscViewerASCIIPrintf(viewer," lwork: %D\n",lu->lwork);
83: if (A->factortype == MAT_FACTOR_ILU) {
84: PetscViewerASCIIPrintf(viewer," ILU_DropTol: %g\n",options.ILU_DropTol);
85: PetscViewerASCIIPrintf(viewer," ILU_FillTol: %g\n",options.ILU_FillTol);
86: PetscViewerASCIIPrintf(viewer," ILU_FillFactor: %g\n",options.ILU_FillFactor);
87: PetscViewerASCIIPrintf(viewer," ILU_DropRule: %D\n",options.ILU_DropRule);
88: PetscViewerASCIIPrintf(viewer," ILU_Norm: %D\n",options.ILU_Norm);
89: PetscViewerASCIIPrintf(viewer," ILU_MILU: %D\n",options.ILU_MILU);
90: }
91: return(0);
92: }
94: PetscErrorCode MatSolve_SuperLU_Private(Mat A,Vec b,Vec x)
95: {
96: Mat_SuperLU *lu = (Mat_SuperLU*)A->data;
97: const PetscScalar *barray;
98: PetscScalar *xarray;
99: PetscErrorCode ierr;
100: PetscInt info,i,n;
101: PetscReal ferr,berr;
102: static PetscBool cite = PETSC_FALSE;
105: if (lu->lwork == -1) return(0);
106: PetscCitationsRegister("@article{superlu99,\n author = {James W. Demmel and Stanley C. Eisenstat and\n John R. Gilbert and Xiaoye S. Li and Joseph W. H. Liu},\n title = {A supernodal approach to sparse partial pivoting},\n journal = {SIAM J. Matrix Analysis and Applications},\n year = {1999},\n volume = {20},\n number = {3},\n pages = {720-755}\n}\n",&cite);
108: VecGetLocalSize(x,&n);
109: lu->B.ncol = 1; /* Set the number of right-hand side */
110: if (lu->options.Equil && !lu->rhs_dup) {
111: /* superlu overwrites b when Equil is used, thus create rhs_dup to keep user's b unchanged */
112: PetscMalloc1(n,&lu->rhs_dup);
113: }
114: if (lu->options.Equil) {
115: /* Copy b into rsh_dup */
116: VecGetArrayRead(b,&barray);
117: PetscMemcpy(lu->rhs_dup,barray,n*sizeof(PetscScalar));
118: VecRestoreArrayRead(b,&barray);
119: barray = lu->rhs_dup;
120: } else {
121: VecGetArrayRead(b,&barray);
122: }
123: VecGetArray(x,&xarray);
125: #if defined(PETSC_USE_COMPLEX)
126: #if defined(PETSC_USE_REAL_SINGLE)
127: ((DNformat*)lu->B.Store)->nzval = (singlecomplex*)barray;
128: ((DNformat*)lu->X.Store)->nzval = (singlecomplex*)xarray;
129: #else
130: ((DNformat*)lu->B.Store)->nzval = (doublecomplex*)barray;
131: ((DNformat*)lu->X.Store)->nzval = (doublecomplex*)xarray;
132: #endif
133: #else
134: ((DNformat*)lu->B.Store)->nzval = (void*)barray;
135: ((DNformat*)lu->X.Store)->nzval = xarray;
136: #endif
138: lu->options.Fact = FACTORED; /* Indicate the factored form of A is supplied. */
139: if (A->factortype == MAT_FACTOR_LU) {
140: #if defined(PETSC_USE_COMPLEX)
141: #if defined(PETSC_USE_REAL_SINGLE)
142: PetscStackCall("SuperLU:cgssvx",cgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
143: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
144: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
145: #else
146: PetscStackCall("SuperLU:zgssvx",zgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
147: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
148: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
149: #endif
150: #else
151: #if defined(PETSC_USE_REAL_SINGLE)
152: PetscStackCall("SuperLU:sgssvx",sgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
153: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
154: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
155: #else
156: PetscStackCall("SuperLU:dgssvx",dgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
157: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
158: &lu->Glu,&lu->mem_usage, &lu->stat, &info));
159: #endif
160: #endif
161: } else if (A->factortype == MAT_FACTOR_ILU) {
162: #if defined(PETSC_USE_COMPLEX)
163: #if defined(PETSC_USE_REAL_SINGLE)
164: PetscStackCall("SuperLU:cgsisx",cgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
165: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
166: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
167: #else
168: PetscStackCall("SuperLU:zgsisx",zgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
169: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
170: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
171: #endif
172: #else
173: #if defined(PETSC_USE_REAL_SINGLE)
174: PetscStackCall("SuperLU:sgsisx",sgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
175: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
176: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
177: #else
178: PetscStackCall("SuperLU:dgsisx",dgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
179: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
180: &lu->Glu, &lu->mem_usage, &lu->stat, &info));
181: #endif
182: #endif
183: } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Factor type not supported");
184: if (!lu->options.Equil) {
185: VecRestoreArrayRead(b,&barray);
186: }
187: VecRestoreArray(x,&xarray);
189: if (!info || info == lu->A.ncol+1) {
190: if (lu->options.IterRefine) {
191: PetscPrintf(PETSC_COMM_SELF,"Iterative Refinement:\n");
192: PetscPrintf(PETSC_COMM_SELF," %8s%8s%16s%16s\n", "rhs", "Steps", "FERR", "BERR");
193: for (i = 0; i < 1; ++i) {
194: PetscPrintf(PETSC_COMM_SELF," %8d%8d%16e%16e\n", i+1, lu->stat.RefineSteps, ferr, berr);
195: }
196: }
197: } else if (info > 0) {
198: if (lu->lwork == -1) {
199: PetscPrintf(PETSC_COMM_SELF," ** Estimated memory: %D bytes\n", info - lu->A.ncol);
200: } else {
201: PetscPrintf(PETSC_COMM_SELF," Warning: gssvx() returns info %D\n",info);
202: }
203: } else if (info < 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB, "info = %D, the %D-th argument in gssvx() had an illegal value", info,-info);
205: if (lu->options.PrintStat) {
206: PetscPrintf(PETSC_COMM_SELF,"MatSolve__SuperLU():\n");
207: PetscStackCall("SuperLU:StatPrint",StatPrint(&lu->stat));
208: }
209: return(0);
210: }
212: PetscErrorCode MatSolve_SuperLU(Mat A,Vec b,Vec x)
213: {
214: Mat_SuperLU *lu = (Mat_SuperLU*)A->data;
218: if (A->factorerrortype) {
219: PetscInfo(A,"MatSolve is called with singular matrix factor, skip\n");
220: VecSetInf(x);
221: return(0);
222: }
224: lu->options.Trans = TRANS;
225: MatSolve_SuperLU_Private(A,b,x);
226: return(0);
227: }
229: PetscErrorCode MatSolveTranspose_SuperLU(Mat A,Vec b,Vec x)
230: {
231: Mat_SuperLU *lu = (Mat_SuperLU*)A->data;
235: if (A->factorerrortype) {
236: PetscInfo(A,"MatSolve is called with singular matrix factor, skip\n");
237: VecSetInf(x);
238: return(0);
239: }
241: lu->options.Trans = NOTRANS;
242: MatSolve_SuperLU_Private(A,b,x);
243: return(0);
244: }
246: static PetscErrorCode MatLUFactorNumeric_SuperLU(Mat F,Mat A,const MatFactorInfo *info)
247: {
248: Mat_SuperLU *lu = (Mat_SuperLU*)F->data;
249: Mat_SeqAIJ *aa;
251: PetscInt sinfo;
252: PetscReal ferr, berr;
253: NCformat *Ustore;
254: SCformat *Lstore;
257: if (lu->flg == SAME_NONZERO_PATTERN) { /* successing numerical factorization */
258: lu->options.Fact = SamePattern;
259: /* Ref: ~SuperLU_3.0/EXAMPLE/dlinsolx2.c */
260: Destroy_SuperMatrix_Store(&lu->A);
261: if (lu->A_dup) {
262: MatCopy_SeqAIJ(A,lu->A_dup,SAME_NONZERO_PATTERN);
263: }
265: if (lu->lwork >= 0) {
266: PetscStackCall("SuperLU:Destroy_SuperNode_Matrix",Destroy_SuperNode_Matrix(&lu->L));
267: PetscStackCall("SuperLU:Destroy_CompCol_Matrix",Destroy_CompCol_Matrix(&lu->U));
268: lu->options.Fact = SamePattern;
269: }
270: }
272: /* Create the SuperMatrix for lu->A=A^T:
273: Since SuperLU likes column-oriented matrices,we pass it the transpose,
274: and then solve A^T X = B in MatSolve(). */
275: if (lu->A_dup) {
276: aa = (Mat_SeqAIJ*)(lu->A_dup)->data;
277: } else {
278: aa = (Mat_SeqAIJ*)(A)->data;
279: }
280: #if defined(PETSC_USE_COMPLEX)
281: #if defined(PETSC_USE_REAL_SINGLE)
282: PetscStackCall("SuperLU:cCreate_CompCol_Matrix",cCreate_CompCol_Matrix(&lu->A,A->cmap->n,A->rmap->n,aa->nz,(singlecomplex*)aa->a,aa->j,aa->i,SLU_NC,SLU_C,SLU_GE));
283: #else
284: PetscStackCall("SuperLU:zCreate_CompCol_Matrix",zCreate_CompCol_Matrix(&lu->A,A->cmap->n,A->rmap->n,aa->nz,(doublecomplex*)aa->a,aa->j,aa->i,SLU_NC,SLU_Z,SLU_GE));
285: #endif
286: #else
287: #if defined(PETSC_USE_REAL_SINGLE)
288: PetscStackCall("SuperLU:sCreate_CompCol_Matrix",sCreate_CompCol_Matrix(&lu->A,A->cmap->n,A->rmap->n,aa->nz,aa->a,aa->j,aa->i,SLU_NC,SLU_S,SLU_GE));
289: #else
290: PetscStackCall("SuperLU:dCreate_CompCol_Matrix",dCreate_CompCol_Matrix(&lu->A,A->cmap->n,A->rmap->n,aa->nz,aa->a,aa->j,aa->i,SLU_NC,SLU_D,SLU_GE));
291: #endif
292: #endif
294: /* Numerical factorization */
295: lu->B.ncol = 0; /* Indicate not to solve the system */
296: if (F->factortype == MAT_FACTOR_LU) {
297: #if defined(PETSC_USE_COMPLEX)
298: #if defined(PETSC_USE_REAL_SINGLE)
299: PetscStackCall("SuperLU:cgssvx",cgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
300: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
301: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
302: #else
303: PetscStackCall("SuperLU:zgssvx",zgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
304: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
305: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
306: #endif
307: #else
308: #if defined(PETSC_USE_REAL_SINGLE)
309: PetscStackCall("SuperLU:sgssvx",sgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
310: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
311: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
312: #else
313: PetscStackCall("SuperLU:dgssvx",dgssvx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
314: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond, &ferr, &berr,
315: &lu->Glu,&lu->mem_usage, &lu->stat, &sinfo));
316: #endif
317: #endif
318: } else if (F->factortype == MAT_FACTOR_ILU) {
319: /* Compute the incomplete factorization, condition number and pivot growth */
320: #if defined(PETSC_USE_COMPLEX)
321: #if defined(PETSC_USE_REAL_SINGLE)
322: PetscStackCall("SuperLU:cgsisx",cgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r,lu->etree, lu->equed, lu->R, lu->C,
323: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
324: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
325: #else
326: PetscStackCall("SuperLU:zgsisx",zgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r,lu->etree, lu->equed, lu->R, lu->C,
327: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
328: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
329: #endif
330: #else
331: #if defined(PETSC_USE_REAL_SINGLE)
332: PetscStackCall("SuperLU:sgsisx",sgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
333: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
334: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
335: #else
336: PetscStackCall("SuperLU:dgsisx",dgsisx(&lu->options, &lu->A, lu->perm_c, lu->perm_r, lu->etree, lu->equed, lu->R, lu->C,
337: &lu->L, &lu->U, lu->work, lu->lwork, &lu->B, &lu->X, &lu->rpg, &lu->rcond,
338: &lu->Glu, &lu->mem_usage, &lu->stat, &sinfo));
339: #endif
340: #endif
341: } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Factor type not supported");
342: if (!sinfo || sinfo == lu->A.ncol+1) {
343: if (lu->options.PivotGrowth) {
344: PetscPrintf(PETSC_COMM_SELF," Recip. pivot growth = %e\n", lu->rpg);
345: }
346: if (lu->options.ConditionNumber) {
347: PetscPrintf(PETSC_COMM_SELF," Recip. condition number = %e\n", lu->rcond);
348: }
349: } else if (sinfo > 0) {
350: if (A->erroriffailure) {
351: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_MAT_LU_ZRPVT,"Zero pivot in row %D",sinfo);
352: } else {
353: if (sinfo <= lu->A.ncol) {
354: if (lu->options.ILU_FillTol == 0.0) {
355: F->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
356: }
357: PetscInfo2(F,"Number of zero pivots %D, ILU_FillTol %g\n",sinfo,lu->options.ILU_FillTol);
358: } else if (sinfo == lu->A.ncol + 1) {
359: /*
360: U is nonsingular, but RCOND is less than machine
361: precision, meaning that the matrix is singular to
362: working precision. Nevertheless, the solution and
363: error bounds are computed because there are a number
364: of situations where the computed solution can be more
365: accurate than the value of RCOND would suggest.
366: */
367: PetscInfo1(F,"Matrix factor U is nonsingular, but is singular to working precision. The solution is computed. info %D",sinfo);
368: } else { /* sinfo > lu->A.ncol + 1 */
369: F->factorerrortype = MAT_FACTOR_OUTMEMORY;
370: PetscInfo1(F,"Number of bytes allocated when memory allocation fails %D\n",sinfo);
371: }
372: }
373: } else SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB, "info = %D, the %D-th argument in gssvx() had an illegal value", sinfo,-sinfo);
375: if (lu->options.PrintStat) {
376: PetscPrintf(PETSC_COMM_SELF,"MatLUFactorNumeric_SuperLU():\n");
377: PetscStackCall("SuperLU:StatPrint",StatPrint(&lu->stat));
378: Lstore = (SCformat*) lu->L.Store;
379: Ustore = (NCformat*) lu->U.Store;
380: PetscPrintf(PETSC_COMM_SELF," No of nonzeros in factor L = %D\n", Lstore->nnz);
381: PetscPrintf(PETSC_COMM_SELF," No of nonzeros in factor U = %D\n", Ustore->nnz);
382: PetscPrintf(PETSC_COMM_SELF," No of nonzeros in L+U = %D\n", Lstore->nnz + Ustore->nnz - lu->A.ncol);
383: PetscPrintf(PETSC_COMM_SELF," L\\U MB %.3f\ttotal MB needed %.3f\n",
384: lu->mem_usage.for_lu/1e6, lu->mem_usage.total_needed/1e6);
385: }
387: lu->flg = SAME_NONZERO_PATTERN;
388: F->ops->solve = MatSolve_SuperLU;
389: F->ops->solvetranspose = MatSolveTranspose_SuperLU;
390: F->ops->matsolve = NULL;
391: return(0);
392: }
394: static PetscErrorCode MatDestroy_SuperLU(Mat A)
395: {
397: Mat_SuperLU *lu=(Mat_SuperLU*)A->data;
400: if (lu->CleanUpSuperLU) { /* Free the SuperLU datastructures */
401: PetscStackCall("SuperLU:Destroy_SuperMatrix_Store",Destroy_SuperMatrix_Store(&lu->A));
402: PetscStackCall("SuperLU:Destroy_SuperMatrix_Store",Destroy_SuperMatrix_Store(&lu->B));
403: PetscStackCall("SuperLU:Destroy_SuperMatrix_Store",Destroy_SuperMatrix_Store(&lu->X));
404: PetscStackCall("SuperLU:StatFree",StatFree(&lu->stat));
405: if (lu->lwork >= 0) {
406: PetscStackCall("SuperLU:Destroy_SuperNode_Matrix",Destroy_SuperNode_Matrix(&lu->L));
407: PetscStackCall("SuperLU:Destroy_CompCol_Matrix",Destroy_CompCol_Matrix(&lu->U));
408: }
409: }
410: PetscFree(lu->etree);
411: PetscFree(lu->perm_r);
412: PetscFree(lu->perm_c);
413: PetscFree(lu->R);
414: PetscFree(lu->C);
415: PetscFree(lu->rhs_dup);
416: MatDestroy(&lu->A_dup);
417: PetscFree(A->data);
419: /* clear composed functions */
420: PetscObjectComposeFunction((PetscObject)A,"MatFactorGetSolverType_C",NULL);
421: PetscObjectComposeFunction((PetscObject)A,"MatSuperluSetILUDropTol_C",NULL);
422: return(0);
423: }
425: static PetscErrorCode MatView_SuperLU(Mat A,PetscViewer viewer)
426: {
427: PetscErrorCode ierr;
428: PetscBool iascii;
429: PetscViewerFormat format;
432: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);
433: if (iascii) {
434: PetscViewerGetFormat(viewer,&format);
435: if (format == PETSC_VIEWER_ASCII_INFO) {
436: MatView_Info_SuperLU(A,viewer);
437: }
438: }
439: return(0);
440: }
442: PetscErrorCode MatMatSolve_SuperLU(Mat A,Mat B,Mat X)
443: {
444: Mat_SuperLU *lu = (Mat_SuperLU*)A->data;
445: PetscBool flg;
449: PetscObjectTypeCompareAny((PetscObject)B,&flg,MATSEQDENSE,MATMPIDENSE,NULL);
450: if (!flg) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONG,"Matrix B must be MATDENSE matrix");
451: PetscObjectTypeCompareAny((PetscObject)X,&flg,MATSEQDENSE,MATMPIDENSE,NULL);
452: if (!flg) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONG,"Matrix X must be MATDENSE matrix");
453: lu->options.Trans = TRANS;
454: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatMatSolve_SuperLU() is not implemented yet");
455: return(0);
456: }
458: /*
459: Note the r permutation is ignored
460: */
461: static PetscErrorCode MatLUFactorSymbolic_SuperLU(Mat F,Mat A,IS r,IS c,const MatFactorInfo *info)
462: {
463: Mat_SuperLU *lu = (Mat_SuperLU*)(F->data);
467: lu->flg = DIFFERENT_NONZERO_PATTERN;
468: lu->CleanUpSuperLU = PETSC_TRUE;
469: F->ops->lufactornumeric = MatLUFactorNumeric_SuperLU;
471: /* if we are here, the nonzero pattern has changed unless the user explicitly called MatLUFactorSymbolic */
472: MatDestroy(&lu->A_dup);
473: if (lu->needconversion) {
474: MatConvert(A,MATSEQAIJ,MAT_INITIAL_MATRIX,&lu->A_dup);
475: }
476: if (lu->options.Equil == YES && !lu->A_dup) { /* superlu overwrites input matrix and rhs when Equil is used, thus create A_dup to keep user's A unchanged */
477: MatDuplicate_SeqAIJ(A,MAT_COPY_VALUES,&lu->A_dup);
478: }
479: return(0);
480: }
482: static PetscErrorCode MatSuperluSetILUDropTol_SuperLU(Mat F,PetscReal dtol)
483: {
484: Mat_SuperLU *lu= (Mat_SuperLU*)F->data;
487: lu->options.ILU_DropTol = dtol;
488: return(0);
489: }
491: /*@
492: MatSuperluSetILUDropTol - Set SuperLU ILU drop tolerance
493: Logically Collective on Mat
495: Input Parameters:
496: + F - the factored matrix obtained by calling MatGetFactor() from PETSc-SuperLU interface
497: - dtol - drop tolerance
499: Options Database:
500: . -mat_superlu_ilu_droptol <dtol>
502: Level: beginner
504: References:
505: . SuperLU Users' Guide
507: .seealso: MatGetFactor()
508: @*/
509: PetscErrorCode MatSuperluSetILUDropTol(Mat F,PetscReal dtol)
510: {
516: PetscTryMethod(F,"MatSuperluSetILUDropTol_C",(Mat,PetscReal),(F,dtol));
517: return(0);
518: }
520: PetscErrorCode MatFactorGetSolverType_seqaij_superlu(Mat A,MatSolverType *type)
521: {
523: *type = MATSOLVERSUPERLU;
524: return(0);
525: }
527: /*MC
528: MATSOLVERSUPERLU = "superlu" - A solver package providing solvers LU and ILU for sequential matrices
529: via the external package SuperLU.
531: Use ./configure --download-superlu to have PETSc installed with SuperLU
533: Use -pc_type lu -pc_factor_mat_solver_type superlu to use this direct solver
535: Options Database Keys:
536: + -mat_superlu_equil <FALSE> - Equil (None)
537: . -mat_superlu_colperm <COLAMD> - (choose one of) NATURAL MMD_ATA MMD_AT_PLUS_A COLAMD
538: . -mat_superlu_iterrefine <NOREFINE> - (choose one of) NOREFINE SINGLE DOUBLE EXTRA
539: . -mat_superlu_symmetricmode: <FALSE> - SymmetricMode (None)
540: . -mat_superlu_diagpivotthresh <1> - DiagPivotThresh (None)
541: . -mat_superlu_pivotgrowth <FALSE> - PivotGrowth (None)
542: . -mat_superlu_conditionnumber <FALSE> - ConditionNumber (None)
543: . -mat_superlu_rowperm <NOROWPERM> - (choose one of) NOROWPERM LargeDiag
544: . -mat_superlu_replacetinypivot <FALSE> - ReplaceTinyPivot (None)
545: . -mat_superlu_printstat <FALSE> - PrintStat (None)
546: . -mat_superlu_lwork <0> - size of work array in bytes used by factorization (None)
547: . -mat_superlu_ilu_droptol <0> - ILU_DropTol (None)
548: . -mat_superlu_ilu_filltol <0> - ILU_FillTol (None)
549: . -mat_superlu_ilu_fillfactor <0> - ILU_FillFactor (None)
550: . -mat_superlu_ilu_droprull <0> - ILU_DropRule (None)
551: . -mat_superlu_ilu_norm <0> - ILU_Norm (None)
552: - -mat_superlu_ilu_milu <0> - ILU_MILU (None)
554: Notes:
555: Do not confuse this with MATSOLVERSUPERLU_DIST which is for parallel sparse solves
557: Level: beginner
559: .seealso: PCLU, PCILU, MATSOLVERSUPERLU_DIST, MATSOLVERMUMPS, PCFactorSetMatSolverType(), MatSolverType
560: M*/
562: static PetscErrorCode MatGetFactor_seqaij_superlu(Mat A,MatFactorType ftype,Mat *F)
563: {
564: Mat B;
565: Mat_SuperLU *lu;
567: PetscInt indx,m=A->rmap->n,n=A->cmap->n;
568: PetscBool flg,set;
569: PetscReal real_input;
570: const char *colperm[] ={"NATURAL","MMD_ATA","MMD_AT_PLUS_A","COLAMD"}; /* MY_PERMC - not supported by the petsc interface yet */
571: const char *iterrefine[]={"NOREFINE", "SINGLE", "DOUBLE", "EXTRA"};
572: const char *rowperm[] ={"NOROWPERM", "LargeDiag"}; /* MY_PERMC - not supported by the petsc interface yet */
575: MatCreate(PetscObjectComm((PetscObject)A),&B);
576: MatSetSizes(B,A->rmap->n,A->cmap->n,PETSC_DETERMINE,PETSC_DETERMINE);
577: PetscStrallocpy("superlu",&((PetscObject)B)->type_name);
578: MatSetUp(B);
579: if (ftype == MAT_FACTOR_LU || ftype == MAT_FACTOR_ILU) {
580: B->ops->lufactorsymbolic = MatLUFactorSymbolic_SuperLU;
581: B->ops->ilufactorsymbolic = MatLUFactorSymbolic_SuperLU;
582: } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Factor type not supported");
584: PetscFree(B->solvertype);
585: PetscStrallocpy(MATSOLVERSUPERLU,&B->solvertype);
587: B->ops->getinfo = MatGetInfo_External;
588: B->ops->destroy = MatDestroy_SuperLU;
589: B->ops->view = MatView_SuperLU;
590: B->factortype = ftype;
591: B->assembled = PETSC_TRUE; /* required by -ksp_view */
592: B->preallocated = PETSC_TRUE;
594: PetscNewLog(B,&lu);
596: if (ftype == MAT_FACTOR_LU) {
597: set_default_options(&lu->options);
598: /* Comments from SuperLU_4.0/SRC/dgssvx.c:
599: "Whether or not the system will be equilibrated depends on the
600: scaling of the matrix A, but if equilibration is used, A is
601: overwritten by diag(R)*A*diag(C) and B by diag(R)*B
602: (if options->Trans=NOTRANS) or diag(C)*B (if options->Trans = TRANS or CONJ)."
603: We set 'options.Equil = NO' as default because additional space is needed for it.
604: */
605: lu->options.Equil = NO;
606: } else if (ftype == MAT_FACTOR_ILU) {
607: /* Set the default input options of ilu: */
608: PetscStackCall("SuperLU:ilu_set_default_options",ilu_set_default_options(&lu->options));
609: }
610: lu->options.PrintStat = NO;
612: /* Initialize the statistics variables. */
613: PetscStackCall("SuperLU:StatInit",StatInit(&lu->stat));
614: lu->lwork = 0; /* allocate space internally by system malloc */
616: PetscOptionsBegin(PetscObjectComm((PetscObject)A),((PetscObject)A)->prefix,"SuperLU Options","Mat");
617: PetscOptionsBool("-mat_superlu_equil","Equil","None",(PetscBool)lu->options.Equil,(PetscBool*)&lu->options.Equil,NULL);
618: PetscOptionsEList("-mat_superlu_colperm","ColPerm","None",colperm,4,colperm[3],&indx,&flg);
619: if (flg) lu->options.ColPerm = (colperm_t)indx;
620: PetscOptionsEList("-mat_superlu_iterrefine","IterRefine","None",iterrefine,4,iterrefine[0],&indx,&flg);
621: if (flg) lu->options.IterRefine = (IterRefine_t)indx;
622: PetscOptionsBool("-mat_superlu_symmetricmode","SymmetricMode","None",(PetscBool)lu->options.SymmetricMode,&flg,&set);
623: if (set && flg) lu->options.SymmetricMode = YES;
624: PetscOptionsReal("-mat_superlu_diagpivotthresh","DiagPivotThresh","None",lu->options.DiagPivotThresh,&real_input,&flg);
625: if (flg) lu->options.DiagPivotThresh = (double) real_input;
626: PetscOptionsBool("-mat_superlu_pivotgrowth","PivotGrowth","None",(PetscBool)lu->options.PivotGrowth,&flg,&set);
627: if (set && flg) lu->options.PivotGrowth = YES;
628: PetscOptionsBool("-mat_superlu_conditionnumber","ConditionNumber","None",(PetscBool)lu->options.ConditionNumber,&flg,&set);
629: if (set && flg) lu->options.ConditionNumber = YES;
630: PetscOptionsEList("-mat_superlu_rowperm","rowperm","None",rowperm,2,rowperm[lu->options.RowPerm],&indx,&flg);
631: if (flg) lu->options.RowPerm = (rowperm_t)indx;
632: PetscOptionsBool("-mat_superlu_replacetinypivot","ReplaceTinyPivot","None",(PetscBool)lu->options.ReplaceTinyPivot,&flg,&set);
633: if (set && flg) lu->options.ReplaceTinyPivot = YES;
634: PetscOptionsBool("-mat_superlu_printstat","PrintStat","None",(PetscBool)lu->options.PrintStat,&flg,&set);
635: if (set && flg) lu->options.PrintStat = YES;
636: PetscOptionsInt("-mat_superlu_lwork","size of work array in bytes used by factorization","None",lu->lwork,&lu->lwork,NULL);
637: if (lu->lwork > 0) {
638: /* lwork is in bytes, hence PetscMalloc() is used here, not PetscMalloc1()*/
639: PetscMalloc(lu->lwork,&lu->work);
640: } else if (lu->lwork != 0 && lu->lwork != -1) {
641: PetscPrintf(PETSC_COMM_SELF," Warning: lwork %D is not supported by SUPERLU. The default lwork=0 is used.\n",lu->lwork);
642: lu->lwork = 0;
643: }
644: /* ilu options */
645: PetscOptionsReal("-mat_superlu_ilu_droptol","ILU_DropTol","None",lu->options.ILU_DropTol,&real_input,&flg);
646: if (flg) lu->options.ILU_DropTol = (double) real_input;
647: PetscOptionsReal("-mat_superlu_ilu_filltol","ILU_FillTol","None",lu->options.ILU_FillTol,&real_input,&flg);
648: if (flg) lu->options.ILU_FillTol = (double) real_input;
649: PetscOptionsReal("-mat_superlu_ilu_fillfactor","ILU_FillFactor","None",lu->options.ILU_FillFactor,&real_input,&flg);
650: if (flg) lu->options.ILU_FillFactor = (double) real_input;
651: PetscOptionsInt("-mat_superlu_ilu_droprull","ILU_DropRule","None",lu->options.ILU_DropRule,&lu->options.ILU_DropRule,NULL);
652: PetscOptionsInt("-mat_superlu_ilu_norm","ILU_Norm","None",lu->options.ILU_Norm,&indx,&flg);
653: if (flg) lu->options.ILU_Norm = (norm_t)indx;
654: PetscOptionsInt("-mat_superlu_ilu_milu","ILU_MILU","None",lu->options.ILU_MILU,&indx,&flg);
655: if (flg) lu->options.ILU_MILU = (milu_t)indx;
656: PetscOptionsEnd();
658: /* Allocate spaces (notice sizes are for the transpose) */
659: PetscMalloc1(m,&lu->etree);
660: PetscMalloc1(n,&lu->perm_r);
661: PetscMalloc1(m,&lu->perm_c);
662: PetscMalloc1(n,&lu->R);
663: PetscMalloc1(m,&lu->C);
665: /* create rhs and solution x without allocate space for .Store */
666: #if defined(PETSC_USE_COMPLEX)
667: #if defined(PETSC_USE_REAL_SINGLE)
668: PetscStackCall("SuperLU:cCreate_Dense_Matrix(",cCreate_Dense_Matrix(&lu->B, m, 1, NULL, m, SLU_DN, SLU_C, SLU_GE));
669: PetscStackCall("SuperLU:cCreate_Dense_Matrix(",cCreate_Dense_Matrix(&lu->X, m, 1, NULL, m, SLU_DN, SLU_C, SLU_GE));
670: #else
671: PetscStackCall("SuperLU:zCreate_Dense_Matrix",zCreate_Dense_Matrix(&lu->B, m, 1, NULL, m, SLU_DN, SLU_Z, SLU_GE));
672: PetscStackCall("SuperLU:zCreate_Dense_Matrix",zCreate_Dense_Matrix(&lu->X, m, 1, NULL, m, SLU_DN, SLU_Z, SLU_GE));
673: #endif
674: #else
675: #if defined(PETSC_USE_REAL_SINGLE)
676: PetscStackCall("SuperLU:sCreate_Dense_Matrix",sCreate_Dense_Matrix(&lu->B, m, 1, NULL, m, SLU_DN, SLU_S, SLU_GE));
677: PetscStackCall("SuperLU:sCreate_Dense_Matrix",sCreate_Dense_Matrix(&lu->X, m, 1, NULL, m, SLU_DN, SLU_S, SLU_GE));
678: #else
679: PetscStackCall("SuperLU:dCreate_Dense_Matrix",dCreate_Dense_Matrix(&lu->B, m, 1, NULL, m, SLU_DN, SLU_D, SLU_GE));
680: PetscStackCall("SuperLU:dCreate_Dense_Matrix",dCreate_Dense_Matrix(&lu->X, m, 1, NULL, m, SLU_DN, SLU_D, SLU_GE));
681: #endif
682: #endif
684: PetscObjectComposeFunction((PetscObject)B,"MatFactorGetSolverType_C",MatFactorGetSolverType_seqaij_superlu);
685: PetscObjectComposeFunction((PetscObject)B,"MatSuperluSetILUDropTol_C",MatSuperluSetILUDropTol_SuperLU);
686: B->data = lu;
688: *F = B;
689: return(0);
690: }
692: static PetscErrorCode MatGetFactor_seqsell_superlu(Mat A,MatFactorType ftype,Mat *F)
693: {
694: Mat_SuperLU *lu;
698: MatGetFactor_seqaij_superlu(A,ftype,F);
699: lu = (Mat_SuperLU*)((*F)->data);
700: lu->needconversion = PETSC_TRUE;
701: return(0);
702: }
704: PETSC_EXTERN PetscErrorCode MatSolverTypeRegister_SuperLU(void)
705: {
709: MatSolverTypeRegister(MATSOLVERSUPERLU,MATSEQAIJ,MAT_FACTOR_LU,MatGetFactor_seqaij_superlu);
710: MatSolverTypeRegister(MATSOLVERSUPERLU,MATSEQAIJ,MAT_FACTOR_ILU,MatGetFactor_seqaij_superlu);
711: MatSolverTypeRegister(MATSOLVERSUPERLU,MATSEQSELL,MAT_FACTOR_LU,MatGetFactor_seqsell_superlu);
712: MatSolverTypeRegister(MATSOLVERSUPERLU,MATSEQSELL,MAT_FACTOR_ILU,MatGetFactor_seqsell_superlu);
713: return(0);
714: }