Actual source code: mg.c
1: /*
2: Defines the multigrid preconditioner interface.
3: */
4: #include <petsc/private/pcmgimpl.h>
5: #include <petsc/private/kspimpl.h>
6: #include <petscdm.h>
7: PETSC_INTERN PetscErrorCode PCPreSolveChangeRHS(PC, PetscBool *);
9: /*
10: Contains the list of registered coarse space construction routines
11: */
12: PetscFunctionList PCMGCoarseList = NULL;
14: PetscErrorCode PCMGMCycle_Private(PC pc, PC_MG_Levels **mglevelsin, PetscBool transpose, PetscBool matapp, PCRichardsonConvergedReason *reason)
15: {
16: PC_MG *mg = (PC_MG *)pc->data;
17: PC_MG_Levels *mgc, *mglevels = *mglevelsin;
18: PetscInt cycles = (mglevels->level == 1) ? 1 : (PetscInt)mglevels->cycles;
20: PetscFunctionBegin;
21: if (mglevels->eventsmoothsolve) PetscCall(PetscLogEventBegin(mglevels->eventsmoothsolve, 0, 0, 0, 0));
22: if (!transpose) {
23: if (matapp) {
24: PetscCall(KSPMatSolve(mglevels->smoothd, mglevels->B, mglevels->X)); /* pre-smooth */
25: PetscCall(KSPCheckSolve(mglevels->smoothd, pc, NULL));
26: } else {
27: PetscCall(KSPSolve(mglevels->smoothd, mglevels->b, mglevels->x)); /* pre-smooth */
28: PetscCall(KSPCheckSolve(mglevels->smoothd, pc, mglevels->x));
29: }
30: } else {
31: PetscCheck(!matapp, PetscObjectComm((PetscObject)pc), PETSC_ERR_SUP, "Not supported");
32: PetscCall(KSPSolveTranspose(mglevels->smoothu, mglevels->b, mglevels->x)); /* transpose of post-smooth */
33: PetscCall(KSPCheckSolve(mglevels->smoothu, pc, mglevels->x));
34: }
35: if (mglevels->eventsmoothsolve) PetscCall(PetscLogEventEnd(mglevels->eventsmoothsolve, 0, 0, 0, 0));
36: if (mglevels->level) { /* not the coarsest grid */
37: if (mglevels->eventresidual) PetscCall(PetscLogEventBegin(mglevels->eventresidual, 0, 0, 0, 0));
38: if (matapp && !mglevels->R) PetscCall(MatDuplicate(mglevels->B, MAT_DO_NOT_COPY_VALUES, &mglevels->R));
39: if (!transpose) {
40: if (matapp) PetscCall((*mglevels->matresidual)(mglevels->A, mglevels->B, mglevels->X, mglevels->R));
41: else PetscCall((*mglevels->residual)(mglevels->A, mglevels->b, mglevels->x, mglevels->r));
42: } else {
43: if (matapp) PetscCall((*mglevels->matresidualtranspose)(mglevels->A, mglevels->B, mglevels->X, mglevels->R));
44: else PetscCall((*mglevels->residualtranspose)(mglevels->A, mglevels->b, mglevels->x, mglevels->r));
45: }
46: if (mglevels->eventresidual) PetscCall(PetscLogEventEnd(mglevels->eventresidual, 0, 0, 0, 0));
48: /* if on finest level and have convergence criteria set */
49: if (mglevels->level == mglevels->levels - 1 && mg->ttol && reason) {
50: PetscReal rnorm;
51: PetscCall(VecNorm(mglevels->r, NORM_2, &rnorm));
52: if (rnorm <= mg->ttol) {
53: if (rnorm < mg->abstol) {
54: *reason = PCRICHARDSON_CONVERGED_ATOL;
55: PetscCall(PetscInfo(pc, "Linear solver has converged. Residual norm %g is less than absolute tolerance %g\n", (double)rnorm, (double)mg->abstol));
56: } else {
57: *reason = PCRICHARDSON_CONVERGED_RTOL;
58: PetscCall(PetscInfo(pc, "Linear solver has converged. Residual norm %g is less than relative tolerance times initial residual norm %g\n", (double)rnorm, (double)mg->ttol));
59: }
60: PetscFunctionReturn(PETSC_SUCCESS);
61: }
62: }
64: mgc = *(mglevelsin - 1);
65: if (mglevels->eventinterprestrict) PetscCall(PetscLogEventBegin(mglevels->eventinterprestrict, 0, 0, 0, 0));
66: if (!transpose) {
67: if (matapp) PetscCall(MatMatRestrict(mglevels->restrct, mglevels->R, &mgc->B));
68: else PetscCall(MatRestrict(mglevels->restrct, mglevels->r, mgc->b));
69: } else {
70: if (matapp) PetscCall(MatMatRestrict(mglevels->interpolate, mglevels->R, &mgc->B));
71: else PetscCall(MatRestrict(mglevels->interpolate, mglevels->r, mgc->b));
72: }
73: if (mglevels->eventinterprestrict) PetscCall(PetscLogEventEnd(mglevels->eventinterprestrict, 0, 0, 0, 0));
74: if (matapp) {
75: if (!mgc->X) {
76: PetscCall(MatDuplicate(mgc->B, MAT_DO_NOT_COPY_VALUES, &mgc->X));
77: } else {
78: PetscCall(MatZeroEntries(mgc->X));
79: }
80: } else {
81: PetscCall(VecZeroEntries(mgc->x));
82: }
83: while (cycles--) PetscCall(PCMGMCycle_Private(pc, mglevelsin - 1, transpose, matapp, reason));
84: if (mglevels->eventinterprestrict) PetscCall(PetscLogEventBegin(mglevels->eventinterprestrict, 0, 0, 0, 0));
85: if (!transpose) {
86: if (matapp) PetscCall(MatMatInterpolateAdd(mglevels->interpolate, mgc->X, mglevels->X, &mglevels->X));
87: else PetscCall(MatInterpolateAdd(mglevels->interpolate, mgc->x, mglevels->x, mglevels->x));
88: } else {
89: PetscCall(MatInterpolateAdd(mglevels->restrct, mgc->x, mglevels->x, mglevels->x));
90: }
91: if (mglevels->eventinterprestrict) PetscCall(PetscLogEventEnd(mglevels->eventinterprestrict, 0, 0, 0, 0));
92: if (mglevels->eventsmoothsolve) PetscCall(PetscLogEventBegin(mglevels->eventsmoothsolve, 0, 0, 0, 0));
93: if (!transpose) {
94: if (matapp) {
95: PetscCall(KSPMatSolve(mglevels->smoothu, mglevels->B, mglevels->X)); /* post smooth */
96: PetscCall(KSPCheckSolve(mglevels->smoothu, pc, NULL));
97: } else {
98: PetscCall(KSPSolve(mglevels->smoothu, mglevels->b, mglevels->x)); /* post smooth */
99: PetscCall(KSPCheckSolve(mglevels->smoothu, pc, mglevels->x));
100: }
101: } else {
102: PetscCheck(!matapp, PetscObjectComm((PetscObject)pc), PETSC_ERR_SUP, "Not supported");
103: PetscCall(KSPSolveTranspose(mglevels->smoothd, mglevels->b, mglevels->x)); /* post smooth */
104: PetscCall(KSPCheckSolve(mglevels->smoothd, pc, mglevels->x));
105: }
106: if (mglevels->cr) {
107: PetscCheck(!matapp, PetscObjectComm((PetscObject)pc), PETSC_ERR_SUP, "Not supported");
108: /* TODO Turn on copy and turn off noisy if we have an exact solution
109: PetscCall(VecCopy(mglevels->x, mglevels->crx));
110: PetscCall(VecCopy(mglevels->b, mglevels->crb)); */
111: PetscCall(KSPSetNoisy_Private(mglevels->crx));
112: PetscCall(KSPSolve(mglevels->cr, mglevels->crb, mglevels->crx)); /* compatible relaxation */
113: PetscCall(KSPCheckSolve(mglevels->cr, pc, mglevels->crx));
114: }
115: if (mglevels->eventsmoothsolve) PetscCall(PetscLogEventEnd(mglevels->eventsmoothsolve, 0, 0, 0, 0));
116: }
117: PetscFunctionReturn(PETSC_SUCCESS);
118: }
120: static PetscErrorCode PCApplyRichardson_MG(PC pc, Vec b, Vec x, Vec w, PetscReal rtol, PetscReal abstol, PetscReal dtol, PetscInt its, PetscBool zeroguess, PetscInt *outits, PCRichardsonConvergedReason *reason)
121: {
122: PC_MG *mg = (PC_MG *)pc->data;
123: PC_MG_Levels **mglevels = mg->levels;
124: PC tpc;
125: PetscBool changeu, changed;
126: PetscInt levels = mglevels[0]->levels, i;
128: PetscFunctionBegin;
129: /* When the DM is supplying the matrix then it will not exist until here */
130: for (i = 0; i < levels; i++) {
131: if (!mglevels[i]->A) {
132: PetscCall(KSPGetOperators(mglevels[i]->smoothu, &mglevels[i]->A, NULL));
133: PetscCall(PetscObjectReference((PetscObject)mglevels[i]->A));
134: }
135: }
137: PetscCall(KSPGetPC(mglevels[levels - 1]->smoothd, &tpc));
138: PetscCall(PCPreSolveChangeRHS(tpc, &changed));
139: PetscCall(KSPGetPC(mglevels[levels - 1]->smoothu, &tpc));
140: PetscCall(PCPreSolveChangeRHS(tpc, &changeu));
141: if (!changed && !changeu) {
142: PetscCall(VecDestroy(&mglevels[levels - 1]->b));
143: mglevels[levels - 1]->b = b;
144: } else { /* if the smoother changes the rhs during PreSolve, we cannot use the input vector */
145: if (!mglevels[levels - 1]->b) {
146: Vec *vec;
148: PetscCall(KSPCreateVecs(mglevels[levels - 1]->smoothd, 1, &vec, 0, NULL));
149: mglevels[levels - 1]->b = *vec;
150: PetscCall(PetscFree(vec));
151: }
152: PetscCall(VecCopy(b, mglevels[levels - 1]->b));
153: }
154: mglevels[levels - 1]->x = x;
156: mg->rtol = rtol;
157: mg->abstol = abstol;
158: mg->dtol = dtol;
159: if (rtol) {
160: /* compute initial residual norm for relative convergence test */
161: PetscReal rnorm;
162: if (zeroguess) {
163: PetscCall(VecNorm(b, NORM_2, &rnorm));
164: } else {
165: PetscCall((*mglevels[levels - 1]->residual)(mglevels[levels - 1]->A, b, x, w));
166: PetscCall(VecNorm(w, NORM_2, &rnorm));
167: }
168: mg->ttol = PetscMax(rtol * rnorm, abstol);
169: } else if (abstol) mg->ttol = abstol;
170: else mg->ttol = 0.0;
172: /* since smoother is applied to full system, not just residual we need to make sure that smoothers don't
173: stop prematurely due to small residual */
174: for (i = 1; i < levels; i++) {
175: PetscCall(KSPSetTolerances(mglevels[i]->smoothu, 0, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT));
176: if (mglevels[i]->smoothu != mglevels[i]->smoothd) {
177: /* For Richardson the initial guess is nonzero since it is solving in each cycle the original system not just applying as a preconditioner */
178: PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->smoothd, PETSC_TRUE));
179: PetscCall(KSPSetTolerances(mglevels[i]->smoothd, 0, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT));
180: }
181: }
183: *reason = (PCRichardsonConvergedReason)0;
184: for (i = 0; i < its; i++) {
185: PetscCall(PCMGMCycle_Private(pc, mglevels + levels - 1, PETSC_FALSE, PETSC_FALSE, reason));
186: if (*reason) break;
187: }
188: if (!*reason) *reason = PCRICHARDSON_CONVERGED_ITS;
189: *outits = i;
190: if (!changed && !changeu) mglevels[levels - 1]->b = NULL;
191: PetscFunctionReturn(PETSC_SUCCESS);
192: }
194: PetscErrorCode PCReset_MG(PC pc)
195: {
196: PC_MG *mg = (PC_MG *)pc->data;
197: PC_MG_Levels **mglevels = mg->levels;
198: PetscInt i, n;
200: PetscFunctionBegin;
201: if (mglevels) {
202: n = mglevels[0]->levels;
203: for (i = 0; i < n - 1; i++) {
204: PetscCall(VecDestroy(&mglevels[i + 1]->r));
205: PetscCall(VecDestroy(&mglevels[i]->b));
206: PetscCall(VecDestroy(&mglevels[i]->x));
207: PetscCall(MatDestroy(&mglevels[i + 1]->R));
208: PetscCall(MatDestroy(&mglevels[i]->B));
209: PetscCall(MatDestroy(&mglevels[i]->X));
210: PetscCall(VecDestroy(&mglevels[i]->crx));
211: PetscCall(VecDestroy(&mglevels[i]->crb));
212: PetscCall(MatDestroy(&mglevels[i + 1]->restrct));
213: PetscCall(MatDestroy(&mglevels[i + 1]->interpolate));
214: PetscCall(MatDestroy(&mglevels[i + 1]->inject));
215: PetscCall(VecDestroy(&mglevels[i + 1]->rscale));
216: }
217: PetscCall(VecDestroy(&mglevels[n - 1]->crx));
218: PetscCall(VecDestroy(&mglevels[n - 1]->crb));
219: /* this is not null only if the smoother on the finest level
220: changes the rhs during PreSolve */
221: PetscCall(VecDestroy(&mglevels[n - 1]->b));
222: PetscCall(MatDestroy(&mglevels[n - 1]->B));
224: for (i = 0; i < n; i++) {
225: PetscCall(MatDestroy(&mglevels[i]->coarseSpace));
226: PetscCall(MatDestroy(&mglevels[i]->A));
227: if (mglevels[i]->smoothd != mglevels[i]->smoothu) PetscCall(KSPReset(mglevels[i]->smoothd));
228: PetscCall(KSPReset(mglevels[i]->smoothu));
229: if (mglevels[i]->cr) PetscCall(KSPReset(mglevels[i]->cr));
230: }
231: mg->Nc = 0;
232: }
233: PetscFunctionReturn(PETSC_SUCCESS);
234: }
236: /* Implementing CR
238: We only want to make corrections that ``do not change'' the coarse solution. What we mean by not changing is that if I prolong my coarse solution to the fine grid and then inject that fine solution back to the coarse grid, I get the same answer. Injection is what Brannick calls R. We want the complementary projector to Inj, which we will call S, after Brannick, so that Inj S = 0. Now the orthogonal projector onto the range of Inj^T is
240: Inj^T (Inj Inj^T)^{-1} Inj
242: and if Inj is a VecScatter, as it is now in PETSc, we have
244: Inj^T Inj
246: and
248: S = I - Inj^T Inj
250: since
252: Inj S = Inj - (Inj Inj^T) Inj = 0.
254: Brannick suggests
256: A \to S^T A S \qquad\mathrm{and}\qquad M \to S^T M S
258: but I do not think his :math:`S^T S = I` is correct. Our S is an orthogonal projector, so :math:`S^T S = S^2 = S`. We will use
260: M^{-1} A \to S M^{-1} A S
262: In fact, since it is somewhat hard in PETSc to do the symmetric application, we will just apply S on the left.
264: Check: || Inj P - I ||_F < tol
265: Check: In general, Inj Inj^T = I
266: */
268: typedef struct {
269: PC mg; /* The PCMG object */
270: PetscInt l; /* The multigrid level for this solver */
271: Mat Inj; /* The injection matrix */
272: Mat S; /* I - Inj^T Inj */
273: } CRContext;
275: static PetscErrorCode CRSetup_Private(PC pc)
276: {
277: CRContext *ctx;
278: Mat It;
280: PetscFunctionBeginUser;
281: PetscCall(PCShellGetContext(pc, &ctx));
282: PetscCall(PCMGGetInjection(ctx->mg, ctx->l, &It));
283: PetscCheck(It, PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_WRONGSTATE, "CR requires that injection be defined for this PCMG");
284: PetscCall(MatCreateTranspose(It, &ctx->Inj));
285: PetscCall(MatCreateNormal(ctx->Inj, &ctx->S));
286: PetscCall(MatScale(ctx->S, -1.0));
287: PetscCall(MatShift(ctx->S, 1.0));
288: PetscFunctionReturn(PETSC_SUCCESS);
289: }
291: static PetscErrorCode CRApply_Private(PC pc, Vec x, Vec y)
292: {
293: CRContext *ctx;
295: PetscFunctionBeginUser;
296: PetscCall(PCShellGetContext(pc, &ctx));
297: PetscCall(MatMult(ctx->S, x, y));
298: PetscFunctionReturn(PETSC_SUCCESS);
299: }
301: static PetscErrorCode CRDestroy_Private(PC pc)
302: {
303: CRContext *ctx;
305: PetscFunctionBeginUser;
306: PetscCall(PCShellGetContext(pc, &ctx));
307: PetscCall(MatDestroy(&ctx->Inj));
308: PetscCall(MatDestroy(&ctx->S));
309: PetscCall(PetscFree(ctx));
310: PetscCall(PCShellSetContext(pc, NULL));
311: PetscFunctionReturn(PETSC_SUCCESS);
312: }
314: static PetscErrorCode CreateCR_Private(PC pc, PetscInt l, PC *cr)
315: {
316: CRContext *ctx;
318: PetscFunctionBeginUser;
319: PetscCall(PCCreate(PetscObjectComm((PetscObject)pc), cr));
320: PetscCall(PetscObjectSetName((PetscObject)*cr, "S (complementary projector to injection)"));
321: PetscCall(PetscCalloc1(1, &ctx));
322: ctx->mg = pc;
323: ctx->l = l;
324: PetscCall(PCSetType(*cr, PCSHELL));
325: PetscCall(PCShellSetContext(*cr, ctx));
326: PetscCall(PCShellSetApply(*cr, CRApply_Private));
327: PetscCall(PCShellSetSetUp(*cr, CRSetup_Private));
328: PetscCall(PCShellSetDestroy(*cr, CRDestroy_Private));
329: PetscFunctionReturn(PETSC_SUCCESS);
330: }
332: PetscErrorCode PCMGSetLevels_MG(PC pc, PetscInt levels, MPI_Comm *comms)
333: {
334: PC_MG *mg = (PC_MG *)pc->data;
335: MPI_Comm comm;
336: PC_MG_Levels **mglevels = mg->levels;
337: PCMGType mgtype = mg->am;
338: PetscInt mgctype = (PetscInt)PC_MG_CYCLE_V;
339: PetscInt i;
340: PetscMPIInt size;
341: const char *prefix;
342: PC ipc;
343: PetscInt n;
345: PetscFunctionBegin;
348: if (mg->nlevels == levels) PetscFunctionReturn(PETSC_SUCCESS);
349: PetscCall(PetscObjectGetComm((PetscObject)pc, &comm));
350: if (mglevels) {
351: mgctype = mglevels[0]->cycles;
352: /* changing the number of levels so free up the previous stuff */
353: PetscCall(PCReset_MG(pc));
354: n = mglevels[0]->levels;
355: for (i = 0; i < n; i++) {
356: if (mglevels[i]->smoothd != mglevels[i]->smoothu) PetscCall(KSPDestroy(&mglevels[i]->smoothd));
357: PetscCall(KSPDestroy(&mglevels[i]->smoothu));
358: PetscCall(KSPDestroy(&mglevels[i]->cr));
359: PetscCall(PetscFree(mglevels[i]));
360: }
361: PetscCall(PetscFree(mg->levels));
362: }
364: mg->nlevels = levels;
366: PetscCall(PetscMalloc1(levels, &mglevels));
368: PetscCall(PCGetOptionsPrefix(pc, &prefix));
370: mg->stageApply = 0;
371: for (i = 0; i < levels; i++) {
372: PetscCall(PetscNew(&mglevels[i]));
374: mglevels[i]->level = i;
375: mglevels[i]->levels = levels;
376: mglevels[i]->cycles = mgctype;
377: mg->default_smoothu = 2;
378: mg->default_smoothd = 2;
379: mglevels[i]->eventsmoothsetup = 0;
380: mglevels[i]->eventsmoothsolve = 0;
381: mglevels[i]->eventresidual = 0;
382: mglevels[i]->eventinterprestrict = 0;
384: if (comms) comm = comms[i];
385: if (comm != MPI_COMM_NULL) {
386: PetscCall(KSPCreate(comm, &mglevels[i]->smoothd));
387: PetscCall(KSPSetNestLevel(mglevels[i]->smoothd, pc->kspnestlevel));
388: PetscCall(KSPSetErrorIfNotConverged(mglevels[i]->smoothd, pc->erroriffailure));
389: PetscCall(PetscObjectIncrementTabLevel((PetscObject)mglevels[i]->smoothd, (PetscObject)pc, levels - i));
390: PetscCall(KSPSetOptionsPrefix(mglevels[i]->smoothd, prefix));
391: PetscCall(PetscObjectComposedDataSetInt((PetscObject)mglevels[i]->smoothd, PetscMGLevelId, mglevels[i]->level));
392: if (i || levels == 1) {
393: char tprefix[128];
395: PetscCall(KSPSetType(mglevels[i]->smoothd, KSPCHEBYSHEV));
396: PetscCall(KSPSetConvergenceTest(mglevels[i]->smoothd, KSPConvergedSkip, NULL, NULL));
397: PetscCall(KSPSetNormType(mglevels[i]->smoothd, KSP_NORM_NONE));
398: PetscCall(KSPGetPC(mglevels[i]->smoothd, &ipc));
399: PetscCall(PCSetType(ipc, PCSOR));
400: PetscCall(KSPSetTolerances(mglevels[i]->smoothd, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT, mg->default_smoothd));
402: PetscCall(PetscSNPrintf(tprefix, 128, "mg_levels_%d_", (int)i));
403: PetscCall(KSPAppendOptionsPrefix(mglevels[i]->smoothd, tprefix));
404: } else {
405: PetscCall(KSPAppendOptionsPrefix(mglevels[0]->smoothd, "mg_coarse_"));
407: /* coarse solve is (redundant) LU by default; set shifttype NONZERO to avoid annoying zero-pivot in LU preconditioner */
408: PetscCall(KSPSetType(mglevels[0]->smoothd, KSPPREONLY));
409: PetscCall(KSPGetPC(mglevels[0]->smoothd, &ipc));
410: PetscCallMPI(MPI_Comm_size(comm, &size));
411: if (size > 1) {
412: PetscCall(PCSetType(ipc, PCREDUNDANT));
413: } else {
414: PetscCall(PCSetType(ipc, PCLU));
415: }
416: PetscCall(PCFactorSetShiftType(ipc, MAT_SHIFT_INBLOCKS));
417: }
418: }
419: mglevels[i]->smoothu = mglevels[i]->smoothd;
420: mg->rtol = 0.0;
421: mg->abstol = 0.0;
422: mg->dtol = 0.0;
423: mg->ttol = 0.0;
424: mg->cyclesperpcapply = 1;
425: }
426: mg->levels = mglevels;
427: PetscCall(PCMGSetType(pc, mgtype));
428: PetscFunctionReturn(PETSC_SUCCESS);
429: }
431: /*@C
432: PCMGSetLevels - Sets the number of levels to use with `PCMG`.
433: Must be called before any other `PCMG` routine.
435: Logically Collective
437: Input Parameters:
438: + pc - the preconditioner context
439: . levels - the number of levels
440: - comms - optional communicators for each level; this is to allow solving the coarser problems
441: on smaller sets of processes. For processes that are not included in the computation
442: you must pass `MPI_COMM_NULL`. Use comms = `NULL` to specify that all processes
443: should participate in each level of problem.
445: Level: intermediate
447: Notes:
448: If the number of levels is one then the multigrid uses the `-mg_levels` prefix
449: for setting the level options rather than the `-mg_coarse` prefix.
451: You can free the information in comms after this routine is called.
453: The array of MPI communicators must contain `MPI_COMM_NULL` for those ranks that at each level
454: are not participating in the coarser solve. For example, with 2 levels and 1 and 2 ranks on
455: the two levels, rank 0 in the original communicator will pass in an array of 2 communicators
456: of size 2 and 1, while rank 1 in the original communicator will pass in array of 2 communicators
457: the first of size 2 and the second of value `MPI_COMM_NULL` since the rank 1 does not participate
458: in the coarse grid solve.
460: Since each coarser level may have a new `MPI_Comm` with fewer ranks than the previous, one
461: must take special care in providing the restriction and interpolation operation. We recommend
462: providing these as two step operations; first perform a standard restriction or interpolation on
463: the full number of ranks for that level and then use an MPI call to copy the resulting vector
464: array entries (after calls to VecGetArray()) to the smaller or larger number of ranks, note in both
465: cases the MPI calls must be made on the larger of the two communicators. Traditional MPI send and
466: receives or `MPI_AlltoAllv()` could be used to do the reshuffling of the vector entries.
468: Fortran Notes:
469: Use comms = `PETSC_NULL_MPI_COMM` as the equivalent of `NULL` in the C interface. Note `PETSC_NULL_MPI_COMM`
470: is not `MPI_COMM_NULL`. It is more like `PETSC_NULL_INTEGER`, `PETSC_NULL_REAL` etc.
472: .seealso: [](ch_ksp), `PCMGSetType()`, `PCMGGetLevels()`
473: @*/
474: PetscErrorCode PCMGSetLevels(PC pc, PetscInt levels, MPI_Comm *comms)
475: {
476: PetscFunctionBegin;
478: if (comms) PetscAssertPointer(comms, 3);
479: PetscTryMethod(pc, "PCMGSetLevels_C", (PC, PetscInt, MPI_Comm *), (pc, levels, comms));
480: PetscFunctionReturn(PETSC_SUCCESS);
481: }
483: PetscErrorCode PCDestroy_MG(PC pc)
484: {
485: PC_MG *mg = (PC_MG *)pc->data;
486: PC_MG_Levels **mglevels = mg->levels;
487: PetscInt i, n;
489: PetscFunctionBegin;
490: PetscCall(PCReset_MG(pc));
491: if (mglevels) {
492: n = mglevels[0]->levels;
493: for (i = 0; i < n; i++) {
494: if (mglevels[i]->smoothd != mglevels[i]->smoothu) PetscCall(KSPDestroy(&mglevels[i]->smoothd));
495: PetscCall(KSPDestroy(&mglevels[i]->smoothu));
496: PetscCall(KSPDestroy(&mglevels[i]->cr));
497: PetscCall(PetscFree(mglevels[i]));
498: }
499: PetscCall(PetscFree(mg->levels));
500: }
501: PetscCall(PetscFree(pc->data));
502: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetInterpolations_C", NULL));
503: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetCoarseOperators_C", NULL));
504: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetGalerkin_C", NULL));
505: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetLevels_C", NULL));
506: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetLevels_C", NULL));
507: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetInterpolations_C", NULL));
508: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetCoarseOperators_C", NULL));
509: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptInterpolation_C", NULL));
510: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptInterpolation_C", NULL));
511: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptCR_C", NULL));
512: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptCR_C", NULL));
513: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptCoarseSpaceType_C", NULL));
514: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptCoarseSpaceType_C", NULL));
515: PetscFunctionReturn(PETSC_SUCCESS);
516: }
518: /*
519: PCApply_MG - Runs either an additive, multiplicative, Kaskadic
520: or full cycle of multigrid.
522: Note:
523: A simple wrapper which calls PCMGMCycle(),PCMGACycle(), or PCMGFCycle().
524: */
525: static PetscErrorCode PCApply_MG_Internal(PC pc, Vec b, Vec x, Mat B, Mat X, PetscBool transpose)
526: {
527: PC_MG *mg = (PC_MG *)pc->data;
528: PC_MG_Levels **mglevels = mg->levels;
529: PC tpc;
530: PetscInt levels = mglevels[0]->levels, i;
531: PetscBool changeu, changed, matapp;
533: PetscFunctionBegin;
534: matapp = (PetscBool)(B && X);
535: if (mg->stageApply) PetscCall(PetscLogStagePush(mg->stageApply));
536: /* When the DM is supplying the matrix then it will not exist until here */
537: for (i = 0; i < levels; i++) {
538: if (!mglevels[i]->A) {
539: PetscCall(KSPGetOperators(mglevels[i]->smoothu, &mglevels[i]->A, NULL));
540: PetscCall(PetscObjectReference((PetscObject)mglevels[i]->A));
541: }
542: }
544: PetscCall(KSPGetPC(mglevels[levels - 1]->smoothd, &tpc));
545: PetscCall(PCPreSolveChangeRHS(tpc, &changed));
546: PetscCall(KSPGetPC(mglevels[levels - 1]->smoothu, &tpc));
547: PetscCall(PCPreSolveChangeRHS(tpc, &changeu));
548: if (!changeu && !changed) {
549: if (matapp) {
550: PetscCall(MatDestroy(&mglevels[levels - 1]->B));
551: mglevels[levels - 1]->B = B;
552: } else {
553: PetscCall(VecDestroy(&mglevels[levels - 1]->b));
554: mglevels[levels - 1]->b = b;
555: }
556: } else { /* if the smoother changes the rhs during PreSolve, we cannot use the input vector */
557: if (matapp) {
558: if (mglevels[levels - 1]->B) {
559: PetscInt N1, N2;
560: PetscBool flg;
562: PetscCall(MatGetSize(mglevels[levels - 1]->B, NULL, &N1));
563: PetscCall(MatGetSize(B, NULL, &N2));
564: PetscCall(PetscObjectTypeCompare((PetscObject)mglevels[levels - 1]->B, ((PetscObject)B)->type_name, &flg));
565: if (N1 != N2 || !flg) PetscCall(MatDestroy(&mglevels[levels - 1]->B));
566: }
567: if (!mglevels[levels - 1]->B) {
568: PetscCall(MatDuplicate(B, MAT_COPY_VALUES, &mglevels[levels - 1]->B));
569: } else {
570: PetscCall(MatCopy(B, mglevels[levels - 1]->B, SAME_NONZERO_PATTERN));
571: }
572: } else {
573: if (!mglevels[levels - 1]->b) {
574: Vec *vec;
576: PetscCall(KSPCreateVecs(mglevels[levels - 1]->smoothd, 1, &vec, 0, NULL));
577: mglevels[levels - 1]->b = *vec;
578: PetscCall(PetscFree(vec));
579: }
580: PetscCall(VecCopy(b, mglevels[levels - 1]->b));
581: }
582: }
583: if (matapp) {
584: mglevels[levels - 1]->X = X;
585: } else {
586: mglevels[levels - 1]->x = x;
587: }
589: /* If coarser Xs are present, it means we have already block applied the PC at least once
590: Reset operators if sizes/type do no match */
591: if (matapp && levels > 1 && mglevels[levels - 2]->X) {
592: PetscInt Xc, Bc;
593: PetscBool flg;
595: PetscCall(MatGetSize(mglevels[levels - 2]->X, NULL, &Xc));
596: PetscCall(MatGetSize(mglevels[levels - 1]->B, NULL, &Bc));
597: PetscCall(PetscObjectTypeCompare((PetscObject)mglevels[levels - 2]->X, ((PetscObject)mglevels[levels - 1]->X)->type_name, &flg));
598: if (Xc != Bc || !flg) {
599: PetscCall(MatDestroy(&mglevels[levels - 1]->R));
600: for (i = 0; i < levels - 1; i++) {
601: PetscCall(MatDestroy(&mglevels[i]->R));
602: PetscCall(MatDestroy(&mglevels[i]->B));
603: PetscCall(MatDestroy(&mglevels[i]->X));
604: }
605: }
606: }
608: if (mg->am == PC_MG_MULTIPLICATIVE) {
609: if (matapp) PetscCall(MatZeroEntries(X));
610: else PetscCall(VecZeroEntries(x));
611: for (i = 0; i < mg->cyclesperpcapply; i++) PetscCall(PCMGMCycle_Private(pc, mglevels + levels - 1, transpose, matapp, NULL));
612: } else if (mg->am == PC_MG_ADDITIVE) {
613: PetscCall(PCMGACycle_Private(pc, mglevels, transpose, matapp));
614: } else if (mg->am == PC_MG_KASKADE) {
615: PetscCall(PCMGKCycle_Private(pc, mglevels, transpose, matapp));
616: } else {
617: PetscCall(PCMGFCycle_Private(pc, mglevels, transpose, matapp));
618: }
619: if (mg->stageApply) PetscCall(PetscLogStagePop());
620: if (!changeu && !changed) {
621: if (matapp) {
622: mglevels[levels - 1]->B = NULL;
623: } else {
624: mglevels[levels - 1]->b = NULL;
625: }
626: }
627: PetscFunctionReturn(PETSC_SUCCESS);
628: }
630: static PetscErrorCode PCApply_MG(PC pc, Vec b, Vec x)
631: {
632: PetscFunctionBegin;
633: PetscCall(PCApply_MG_Internal(pc, b, x, NULL, NULL, PETSC_FALSE));
634: PetscFunctionReturn(PETSC_SUCCESS);
635: }
637: static PetscErrorCode PCApplyTranspose_MG(PC pc, Vec b, Vec x)
638: {
639: PetscFunctionBegin;
640: PetscCall(PCApply_MG_Internal(pc, b, x, NULL, NULL, PETSC_TRUE));
641: PetscFunctionReturn(PETSC_SUCCESS);
642: }
644: static PetscErrorCode PCMatApply_MG(PC pc, Mat b, Mat x)
645: {
646: PetscFunctionBegin;
647: PetscCall(PCApply_MG_Internal(pc, NULL, NULL, b, x, PETSC_FALSE));
648: PetscFunctionReturn(PETSC_SUCCESS);
649: }
651: PetscErrorCode PCSetFromOptions_MG(PC pc, PetscOptionItems *PetscOptionsObject)
652: {
653: PetscInt levels, cycles;
654: PetscBool flg, flg2;
655: PC_MG *mg = (PC_MG *)pc->data;
656: PC_MG_Levels **mglevels;
657: PCMGType mgtype;
658: PCMGCycleType mgctype;
659: PCMGGalerkinType gtype;
660: PCMGCoarseSpaceType coarseSpaceType;
662: PetscFunctionBegin;
663: levels = PetscMax(mg->nlevels, 1);
664: PetscOptionsHeadBegin(PetscOptionsObject, "Multigrid options");
665: PetscCall(PetscOptionsInt("-pc_mg_levels", "Number of Levels", "PCMGSetLevels", levels, &levels, &flg));
666: if (!flg && !mg->levels && pc->dm) {
667: PetscCall(DMGetRefineLevel(pc->dm, &levels));
668: levels++;
669: mg->usedmfornumberoflevels = PETSC_TRUE;
670: }
671: PetscCall(PCMGSetLevels(pc, levels, NULL));
672: mglevels = mg->levels;
674: mgctype = (PCMGCycleType)mglevels[0]->cycles;
675: PetscCall(PetscOptionsEnum("-pc_mg_cycle_type", "V cycle or for W-cycle", "PCMGSetCycleType", PCMGCycleTypes, (PetscEnum)mgctype, (PetscEnum *)&mgctype, &flg));
676: if (flg) PetscCall(PCMGSetCycleType(pc, mgctype));
677: gtype = mg->galerkin;
678: PetscCall(PetscOptionsEnum("-pc_mg_galerkin", "Use Galerkin process to compute coarser operators", "PCMGSetGalerkin", PCMGGalerkinTypes, (PetscEnum)gtype, (PetscEnum *)>ype, &flg));
679: if (flg) PetscCall(PCMGSetGalerkin(pc, gtype));
680: coarseSpaceType = mg->coarseSpaceType;
681: PetscCall(PetscOptionsEnum("-pc_mg_adapt_interp_coarse_space", "Type of adaptive coarse space: none, polynomial, harmonic, eigenvector, generalized_eigenvector, gdsw", "PCMGSetAdaptCoarseSpaceType", PCMGCoarseSpaceTypes, (PetscEnum)coarseSpaceType, (PetscEnum *)&coarseSpaceType, &flg));
682: if (flg) PetscCall(PCMGSetAdaptCoarseSpaceType(pc, coarseSpaceType));
683: PetscCall(PetscOptionsInt("-pc_mg_adapt_interp_n", "Size of the coarse space for adaptive interpolation", "PCMGSetCoarseSpace", mg->Nc, &mg->Nc, &flg));
684: PetscCall(PetscOptionsBool("-pc_mg_mesp_monitor", "Monitor the multilevel eigensolver", "PCMGSetAdaptInterpolation", PETSC_FALSE, &mg->mespMonitor, &flg));
685: flg2 = PETSC_FALSE;
686: PetscCall(PetscOptionsBool("-pc_mg_adapt_cr", "Monitor coarse space quality using Compatible Relaxation (CR)", "PCMGSetAdaptCR", PETSC_FALSE, &flg2, &flg));
687: if (flg) PetscCall(PCMGSetAdaptCR(pc, flg2));
688: flg = PETSC_FALSE;
689: PetscCall(PetscOptionsBool("-pc_mg_distinct_smoothup", "Create separate smoothup KSP and append the prefix _up", "PCMGSetDistinctSmoothUp", PETSC_FALSE, &flg, NULL));
690: if (flg) PetscCall(PCMGSetDistinctSmoothUp(pc));
691: mgtype = mg->am;
692: PetscCall(PetscOptionsEnum("-pc_mg_type", "Multigrid type", "PCMGSetType", PCMGTypes, (PetscEnum)mgtype, (PetscEnum *)&mgtype, &flg));
693: if (flg) PetscCall(PCMGSetType(pc, mgtype));
694: if (mg->am == PC_MG_MULTIPLICATIVE) {
695: PetscCall(PetscOptionsInt("-pc_mg_multiplicative_cycles", "Number of cycles for each preconditioner step", "PCMGMultiplicativeSetCycles", mg->cyclesperpcapply, &cycles, &flg));
696: if (flg) PetscCall(PCMGMultiplicativeSetCycles(pc, cycles));
697: }
698: flg = PETSC_FALSE;
699: PetscCall(PetscOptionsBool("-pc_mg_log", "Log times for each multigrid level", "None", flg, &flg, NULL));
700: if (flg) {
701: PetscInt i;
702: char eventname[128];
704: levels = mglevels[0]->levels;
705: for (i = 0; i < levels; i++) {
706: PetscCall(PetscSNPrintf(eventname, PETSC_STATIC_ARRAY_LENGTH(eventname), "MGSetup Level %d", (int)i));
707: PetscCall(PetscLogEventRegister(eventname, ((PetscObject)pc)->classid, &mglevels[i]->eventsmoothsetup));
708: PetscCall(PetscSNPrintf(eventname, PETSC_STATIC_ARRAY_LENGTH(eventname), "MGSmooth Level %d", (int)i));
709: PetscCall(PetscLogEventRegister(eventname, ((PetscObject)pc)->classid, &mglevels[i]->eventsmoothsolve));
710: if (i) {
711: PetscCall(PetscSNPrintf(eventname, PETSC_STATIC_ARRAY_LENGTH(eventname), "MGResid Level %d", (int)i));
712: PetscCall(PetscLogEventRegister(eventname, ((PetscObject)pc)->classid, &mglevels[i]->eventresidual));
713: PetscCall(PetscSNPrintf(eventname, PETSC_STATIC_ARRAY_LENGTH(eventname), "MGInterp Level %d", (int)i));
714: PetscCall(PetscLogEventRegister(eventname, ((PetscObject)pc)->classid, &mglevels[i]->eventinterprestrict));
715: }
716: }
718: if (PetscDefined(USE_LOG)) {
719: const char sname[] = "MG Apply";
721: PetscCall(PetscLogStageGetId(sname, &mg->stageApply));
722: if (mg->stageApply < 0) PetscCall(PetscLogStageRegister(sname, &mg->stageApply));
723: }
724: }
725: PetscOptionsHeadEnd();
726: /* Check option consistency */
727: PetscCall(PCMGGetGalerkin(pc, >ype));
728: PetscCall(PCMGGetAdaptInterpolation(pc, &flg));
729: PetscCheck(!flg || !(gtype >= PC_MG_GALERKIN_NONE), PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_INCOMP, "Must use Galerkin coarse operators when adapting the interpolator");
730: PetscFunctionReturn(PETSC_SUCCESS);
731: }
733: const char *const PCMGTypes[] = {"MULTIPLICATIVE", "ADDITIVE", "FULL", "KASKADE", "PCMGType", "PC_MG", NULL};
734: const char *const PCMGCycleTypes[] = {"invalid", "v", "w", "PCMGCycleType", "PC_MG_CYCLE", NULL};
735: const char *const PCMGGalerkinTypes[] = {"both", "pmat", "mat", "none", "external", "PCMGGalerkinType", "PC_MG_GALERKIN", NULL};
736: const char *const PCMGCoarseSpaceTypes[] = {"none", "polynomial", "harmonic", "eigenvector", "generalized_eigenvector", "gdsw", "PCMGCoarseSpaceType", "PCMG_ADAPT_NONE", NULL};
738: #include <petscdraw.h>
739: PetscErrorCode PCView_MG(PC pc, PetscViewer viewer)
740: {
741: PC_MG *mg = (PC_MG *)pc->data;
742: PC_MG_Levels **mglevels = mg->levels;
743: PetscInt levels = mglevels ? mglevels[0]->levels : 0, i;
744: PetscBool iascii, isbinary, isdraw;
746: PetscFunctionBegin;
747: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERASCII, &iascii));
748: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERBINARY, &isbinary));
749: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERDRAW, &isdraw));
750: if (iascii) {
751: const char *cyclename = levels ? (mglevels[0]->cycles == PC_MG_CYCLE_V ? "v" : "w") : "unknown";
752: PetscCall(PetscViewerASCIIPrintf(viewer, " type is %s, levels=%" PetscInt_FMT " cycles=%s\n", PCMGTypes[mg->am], levels, cyclename));
753: if (mg->am == PC_MG_MULTIPLICATIVE) PetscCall(PetscViewerASCIIPrintf(viewer, " Cycles per PCApply=%" PetscInt_FMT "\n", mg->cyclesperpcapply));
754: if (mg->galerkin == PC_MG_GALERKIN_BOTH) {
755: PetscCall(PetscViewerASCIIPrintf(viewer, " Using Galerkin computed coarse grid matrices\n"));
756: } else if (mg->galerkin == PC_MG_GALERKIN_PMAT) {
757: PetscCall(PetscViewerASCIIPrintf(viewer, " Using Galerkin computed coarse grid matrices for pmat\n"));
758: } else if (mg->galerkin == PC_MG_GALERKIN_MAT) {
759: PetscCall(PetscViewerASCIIPrintf(viewer, " Using Galerkin computed coarse grid matrices for mat\n"));
760: } else if (mg->galerkin == PC_MG_GALERKIN_EXTERNAL) {
761: PetscCall(PetscViewerASCIIPrintf(viewer, " Using externally compute Galerkin coarse grid matrices\n"));
762: } else {
763: PetscCall(PetscViewerASCIIPrintf(viewer, " Not using Galerkin computed coarse grid matrices\n"));
764: }
765: if (mg->view) PetscCall((*mg->view)(pc, viewer));
766: for (i = 0; i < levels; i++) {
767: if (i) {
768: PetscCall(PetscViewerASCIIPrintf(viewer, "Down solver (pre-smoother) on level %" PetscInt_FMT " -------------------------------\n", i));
769: } else {
770: PetscCall(PetscViewerASCIIPrintf(viewer, "Coarse grid solver -- level %" PetscInt_FMT " -------------------------------\n", i));
771: }
772: PetscCall(PetscViewerASCIIPushTab(viewer));
773: PetscCall(KSPView(mglevels[i]->smoothd, viewer));
774: PetscCall(PetscViewerASCIIPopTab(viewer));
775: if (i && mglevels[i]->smoothd == mglevels[i]->smoothu) {
776: PetscCall(PetscViewerASCIIPrintf(viewer, "Up solver (post-smoother) same as down solver (pre-smoother)\n"));
777: } else if (i) {
778: PetscCall(PetscViewerASCIIPrintf(viewer, "Up solver (post-smoother) on level %" PetscInt_FMT " -------------------------------\n", i));
779: PetscCall(PetscViewerASCIIPushTab(viewer));
780: PetscCall(KSPView(mglevels[i]->smoothu, viewer));
781: PetscCall(PetscViewerASCIIPopTab(viewer));
782: }
783: if (i && mglevels[i]->cr) {
784: PetscCall(PetscViewerASCIIPrintf(viewer, "CR solver on level %" PetscInt_FMT " -------------------------------\n", i));
785: PetscCall(PetscViewerASCIIPushTab(viewer));
786: PetscCall(KSPView(mglevels[i]->cr, viewer));
787: PetscCall(PetscViewerASCIIPopTab(viewer));
788: }
789: }
790: } else if (isbinary) {
791: for (i = levels - 1; i >= 0; i--) {
792: PetscCall(KSPView(mglevels[i]->smoothd, viewer));
793: if (i && mglevels[i]->smoothd != mglevels[i]->smoothu) PetscCall(KSPView(mglevels[i]->smoothu, viewer));
794: }
795: } else if (isdraw) {
796: PetscDraw draw;
797: PetscReal x, w, y, bottom, th;
798: PetscCall(PetscViewerDrawGetDraw(viewer, 0, &draw));
799: PetscCall(PetscDrawGetCurrentPoint(draw, &x, &y));
800: PetscCall(PetscDrawStringGetSize(draw, NULL, &th));
801: bottom = y - th;
802: for (i = levels - 1; i >= 0; i--) {
803: if (!mglevels[i]->smoothu || (mglevels[i]->smoothu == mglevels[i]->smoothd)) {
804: PetscCall(PetscDrawPushCurrentPoint(draw, x, bottom));
805: PetscCall(KSPView(mglevels[i]->smoothd, viewer));
806: PetscCall(PetscDrawPopCurrentPoint(draw));
807: } else {
808: w = 0.5 * PetscMin(1.0 - x, x);
809: PetscCall(PetscDrawPushCurrentPoint(draw, x + w, bottom));
810: PetscCall(KSPView(mglevels[i]->smoothd, viewer));
811: PetscCall(PetscDrawPopCurrentPoint(draw));
812: PetscCall(PetscDrawPushCurrentPoint(draw, x - w, bottom));
813: PetscCall(KSPView(mglevels[i]->smoothu, viewer));
814: PetscCall(PetscDrawPopCurrentPoint(draw));
815: }
816: PetscCall(PetscDrawGetBoundingBox(draw, NULL, &bottom, NULL, NULL));
817: bottom -= th;
818: }
819: }
820: PetscFunctionReturn(PETSC_SUCCESS);
821: }
823: #include <petsc/private/kspimpl.h>
825: /*
826: Calls setup for the KSP on each level
827: */
828: PetscErrorCode PCSetUp_MG(PC pc)
829: {
830: PC_MG *mg = (PC_MG *)pc->data;
831: PC_MG_Levels **mglevels = mg->levels;
832: PetscInt i, n;
833: PC cpc;
834: PetscBool dump = PETSC_FALSE, opsset, use_amat, missinginterpolate = PETSC_FALSE;
835: Mat dA, dB;
836: Vec tvec;
837: DM *dms;
838: PetscViewer viewer = NULL;
839: PetscBool dAeqdB = PETSC_FALSE, needRestricts = PETSC_FALSE, doCR = PETSC_FALSE;
840: PetscBool adaptInterpolation = mg->adaptInterpolation;
842: PetscFunctionBegin;
843: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_WRONGSTATE, "Must set MG levels with PCMGSetLevels() before setting up");
844: n = mglevels[0]->levels;
845: /* FIX: Move this to PCSetFromOptions_MG? */
846: if (mg->usedmfornumberoflevels) {
847: PetscInt levels;
848: PetscCall(DMGetRefineLevel(pc->dm, &levels));
849: levels++;
850: if (levels > n) { /* the problem is now being solved on a finer grid */
851: PetscCall(PCMGSetLevels(pc, levels, NULL));
852: n = levels;
853: PetscCall(PCSetFromOptions(pc)); /* it is bad to call this here, but otherwise will never be called for the new hierarchy */
854: mglevels = mg->levels;
855: }
856: }
857: PetscCall(KSPGetPC(mglevels[0]->smoothd, &cpc));
859: /* If user did not provide fine grid operators OR operator was not updated since last global KSPSetOperators() */
860: /* so use those from global PC */
861: /* Is this what we always want? What if user wants to keep old one? */
862: PetscCall(KSPGetOperatorsSet(mglevels[n - 1]->smoothd, NULL, &opsset));
863: if (opsset) {
864: Mat mmat;
865: PetscCall(KSPGetOperators(mglevels[n - 1]->smoothd, NULL, &mmat));
866: if (mmat == pc->pmat) opsset = PETSC_FALSE;
867: }
869: /* Create CR solvers */
870: PetscCall(PCMGGetAdaptCR(pc, &doCR));
871: if (doCR) {
872: const char *prefix;
874: PetscCall(PCGetOptionsPrefix(pc, &prefix));
875: for (i = 1; i < n; ++i) {
876: PC ipc, cr;
877: char crprefix[128];
879: PetscCall(KSPCreate(PetscObjectComm((PetscObject)pc), &mglevels[i]->cr));
880: PetscCall(KSPSetNestLevel(mglevels[i]->cr, pc->kspnestlevel));
881: PetscCall(KSPSetErrorIfNotConverged(mglevels[i]->cr, PETSC_FALSE));
882: PetscCall(PetscObjectIncrementTabLevel((PetscObject)mglevels[i]->cr, (PetscObject)pc, n - i));
883: PetscCall(KSPSetOptionsPrefix(mglevels[i]->cr, prefix));
884: PetscCall(PetscObjectComposedDataSetInt((PetscObject)mglevels[i]->cr, PetscMGLevelId, mglevels[i]->level));
885: PetscCall(KSPSetType(mglevels[i]->cr, KSPCHEBYSHEV));
886: PetscCall(KSPSetConvergenceTest(mglevels[i]->cr, KSPConvergedSkip, NULL, NULL));
887: PetscCall(KSPSetNormType(mglevels[i]->cr, KSP_NORM_PRECONDITIONED));
888: PetscCall(KSPGetPC(mglevels[i]->cr, &ipc));
890: PetscCall(PCSetType(ipc, PCCOMPOSITE));
891: PetscCall(PCCompositeSetType(ipc, PC_COMPOSITE_MULTIPLICATIVE));
892: PetscCall(PCCompositeAddPCType(ipc, PCSOR));
893: PetscCall(CreateCR_Private(pc, i, &cr));
894: PetscCall(PCCompositeAddPC(ipc, cr));
895: PetscCall(PCDestroy(&cr));
897: PetscCall(KSPSetTolerances(mglevels[i]->cr, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT, mg->default_smoothd));
898: PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->cr, PETSC_TRUE));
899: PetscCall(PetscSNPrintf(crprefix, 128, "mg_levels_%d_cr_", (int)i));
900: PetscCall(KSPAppendOptionsPrefix(mglevels[i]->cr, crprefix));
901: }
902: }
904: if (!opsset) {
905: PetscCall(PCGetUseAmat(pc, &use_amat));
906: if (use_amat) {
907: PetscCall(PetscInfo(pc, "Using outer operators to define finest grid operator \n because PCMGGetSmoother(pc,nlevels-1,&ksp);KSPSetOperators(ksp,...); was not called.\n"));
908: PetscCall(KSPSetOperators(mglevels[n - 1]->smoothd, pc->mat, pc->pmat));
909: } else {
910: PetscCall(PetscInfo(pc, "Using matrix (pmat) operators to define finest grid operator \n because PCMGGetSmoother(pc,nlevels-1,&ksp);KSPSetOperators(ksp,...); was not called.\n"));
911: PetscCall(KSPSetOperators(mglevels[n - 1]->smoothd, pc->pmat, pc->pmat));
912: }
913: }
915: for (i = n - 1; i > 0; i--) {
916: if (!(mglevels[i]->interpolate || mglevels[i]->restrct)) {
917: missinginterpolate = PETSC_TRUE;
918: break;
919: }
920: }
922: PetscCall(KSPGetOperators(mglevels[n - 1]->smoothd, &dA, &dB));
923: if (dA == dB) dAeqdB = PETSC_TRUE;
924: if (mg->galerkin == PC_MG_GALERKIN_NONE || ((mg->galerkin == PC_MG_GALERKIN_PMAT || mg->galerkin == PC_MG_GALERKIN_MAT) && !dAeqdB)) {
925: needRestricts = PETSC_TRUE; /* user must compute either mat, pmat, or both so must restrict x to coarser levels */
926: }
928: if (pc->dm && !pc->setupcalled) {
929: /* finest smoother also gets DM but it is not active, independent of whether galerkin==PC_MG_GALERKIN_EXTERNAL */
930: PetscCall(KSPSetDM(mglevels[n - 1]->smoothd, pc->dm));
931: PetscCall(KSPSetDMActive(mglevels[n - 1]->smoothd, PETSC_FALSE));
932: if (mglevels[n - 1]->smoothd != mglevels[n - 1]->smoothu) {
933: PetscCall(KSPSetDM(mglevels[n - 1]->smoothu, pc->dm));
934: PetscCall(KSPSetDMActive(mglevels[n - 1]->smoothu, PETSC_FALSE));
935: }
936: if (mglevels[n - 1]->cr) {
937: PetscCall(KSPSetDM(mglevels[n - 1]->cr, pc->dm));
938: PetscCall(KSPSetDMActive(mglevels[n - 1]->cr, PETSC_FALSE));
939: }
940: }
942: /*
943: Skipping if user has provided all interpolation/restriction needed (since DM might not be able to produce them (when coming from SNES/TS)
944: Skipping for externally managed hierarchy (such as ML and GAMG). Cleaner logic here would be great. Wrap ML/GAMG as DMs?
945: */
946: if (missinginterpolate && mg->galerkin != PC_MG_GALERKIN_EXTERNAL && !pc->setupcalled) {
947: /* first see if we can compute a coarse space */
948: if (mg->coarseSpaceType == PCMG_ADAPT_GDSW) {
949: for (i = n - 2; i > -1; i--) {
950: if (!mglevels[i + 1]->restrct && !mglevels[i + 1]->interpolate) {
951: PetscCall(PCMGComputeCoarseSpace_Internal(pc, i + 1, mg->coarseSpaceType, mg->Nc, NULL, &mglevels[i + 1]->coarseSpace));
952: PetscCall(PCMGSetInterpolation(pc, i + 1, mglevels[i + 1]->coarseSpace));
953: }
954: }
955: } else { /* construct the interpolation from the DMs */
956: Mat p;
957: Vec rscale;
958: PetscCall(PetscMalloc1(n, &dms));
959: dms[n - 1] = pc->dm;
960: /* Separately create them so we do not get DMKSP interference between levels */
961: for (i = n - 2; i > -1; i--) PetscCall(DMCoarsen(dms[i + 1], MPI_COMM_NULL, &dms[i]));
962: for (i = n - 2; i > -1; i--) {
963: DMKSP kdm;
964: PetscBool dmhasrestrict, dmhasinject;
965: PetscCall(KSPSetDM(mglevels[i]->smoothd, dms[i]));
966: if (!needRestricts) PetscCall(KSPSetDMActive(mglevels[i]->smoothd, PETSC_FALSE));
967: if (mglevels[i]->smoothd != mglevels[i]->smoothu) {
968: PetscCall(KSPSetDM(mglevels[i]->smoothu, dms[i]));
969: if (!needRestricts) PetscCall(KSPSetDMActive(mglevels[i]->smoothu, PETSC_FALSE));
970: }
971: if (mglevels[i]->cr) {
972: PetscCall(KSPSetDM(mglevels[i]->cr, dms[i]));
973: if (!needRestricts) PetscCall(KSPSetDMActive(mglevels[i]->cr, PETSC_FALSE));
974: }
975: PetscCall(DMGetDMKSPWrite(dms[i], &kdm));
976: /* Ugly hack so that the next KSPSetUp() will use the RHS that we set. A better fix is to change dmActive to take
977: * a bitwise OR of computing the matrix, RHS, and initial iterate. */
978: kdm->ops->computerhs = NULL;
979: kdm->rhsctx = NULL;
980: if (!mglevels[i + 1]->interpolate) {
981: PetscCall(DMCreateInterpolation(dms[i], dms[i + 1], &p, &rscale));
982: PetscCall(PCMGSetInterpolation(pc, i + 1, p));
983: if (rscale) PetscCall(PCMGSetRScale(pc, i + 1, rscale));
984: PetscCall(VecDestroy(&rscale));
985: PetscCall(MatDestroy(&p));
986: }
987: PetscCall(DMHasCreateRestriction(dms[i], &dmhasrestrict));
988: if (dmhasrestrict && !mglevels[i + 1]->restrct) {
989: PetscCall(DMCreateRestriction(dms[i], dms[i + 1], &p));
990: PetscCall(PCMGSetRestriction(pc, i + 1, p));
991: PetscCall(MatDestroy(&p));
992: }
993: PetscCall(DMHasCreateInjection(dms[i], &dmhasinject));
994: if (dmhasinject && !mglevels[i + 1]->inject) {
995: PetscCall(DMCreateInjection(dms[i], dms[i + 1], &p));
996: PetscCall(PCMGSetInjection(pc, i + 1, p));
997: PetscCall(MatDestroy(&p));
998: }
999: }
1001: for (i = n - 2; i > -1; i--) PetscCall(DMDestroy(&dms[i]));
1002: PetscCall(PetscFree(dms));
1003: }
1004: }
1006: if (mg->galerkin < PC_MG_GALERKIN_NONE) {
1007: Mat A, B;
1008: PetscBool doA = PETSC_FALSE, doB = PETSC_FALSE;
1009: MatReuse reuse = MAT_INITIAL_MATRIX;
1011: if (mg->galerkin == PC_MG_GALERKIN_PMAT || mg->galerkin == PC_MG_GALERKIN_BOTH) doB = PETSC_TRUE;
1012: if (mg->galerkin == PC_MG_GALERKIN_MAT || (mg->galerkin == PC_MG_GALERKIN_BOTH && dA != dB)) doA = PETSC_TRUE;
1013: if (pc->setupcalled) reuse = MAT_REUSE_MATRIX;
1014: for (i = n - 2; i > -1; i--) {
1015: PetscCheck(mglevels[i + 1]->restrct || mglevels[i + 1]->interpolate, PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_WRONGSTATE, "Must provide interpolation or restriction for each MG level except level 0");
1016: if (!mglevels[i + 1]->interpolate) PetscCall(PCMGSetInterpolation(pc, i + 1, mglevels[i + 1]->restrct));
1017: if (!mglevels[i + 1]->restrct) PetscCall(PCMGSetRestriction(pc, i + 1, mglevels[i + 1]->interpolate));
1018: if (reuse == MAT_REUSE_MATRIX) PetscCall(KSPGetOperators(mglevels[i]->smoothd, &A, &B));
1019: if (doA) PetscCall(MatGalerkin(mglevels[i + 1]->restrct, dA, mglevels[i + 1]->interpolate, reuse, 1.0, &A));
1020: if (doB) PetscCall(MatGalerkin(mglevels[i + 1]->restrct, dB, mglevels[i + 1]->interpolate, reuse, 1.0, &B));
1021: /* the management of the PetscObjectReference() and PetscObjecDereference() below is rather delicate */
1022: if (!doA && dAeqdB) {
1023: if (reuse == MAT_INITIAL_MATRIX) PetscCall(PetscObjectReference((PetscObject)B));
1024: A = B;
1025: } else if (!doA && reuse == MAT_INITIAL_MATRIX) {
1026: PetscCall(KSPGetOperators(mglevels[i]->smoothd, &A, NULL));
1027: PetscCall(PetscObjectReference((PetscObject)A));
1028: }
1029: if (!doB && dAeqdB) {
1030: if (reuse == MAT_INITIAL_MATRIX) PetscCall(PetscObjectReference((PetscObject)A));
1031: B = A;
1032: } else if (!doB && reuse == MAT_INITIAL_MATRIX) {
1033: PetscCall(KSPGetOperators(mglevels[i]->smoothd, NULL, &B));
1034: PetscCall(PetscObjectReference((PetscObject)B));
1035: }
1036: if (reuse == MAT_INITIAL_MATRIX) {
1037: PetscCall(KSPSetOperators(mglevels[i]->smoothd, A, B));
1038: PetscCall(PetscObjectDereference((PetscObject)A));
1039: PetscCall(PetscObjectDereference((PetscObject)B));
1040: }
1041: dA = A;
1042: dB = B;
1043: }
1044: }
1046: /* Adapt interpolation matrices */
1047: if (adaptInterpolation) {
1048: for (i = 0; i < n; ++i) {
1049: if (!mglevels[i]->coarseSpace) PetscCall(PCMGComputeCoarseSpace_Internal(pc, i, mg->coarseSpaceType, mg->Nc, !i ? NULL : mglevels[i - 1]->coarseSpace, &mglevels[i]->coarseSpace));
1050: if (i) PetscCall(PCMGAdaptInterpolator_Internal(pc, i, mglevels[i - 1]->smoothu, mglevels[i]->smoothu, mglevels[i - 1]->coarseSpace, mglevels[i]->coarseSpace));
1051: }
1052: for (i = n - 2; i > -1; --i) PetscCall(PCMGRecomputeLevelOperators_Internal(pc, i));
1053: }
1055: if (needRestricts && pc->dm) {
1056: for (i = n - 2; i >= 0; i--) {
1057: DM dmfine, dmcoarse;
1058: Mat Restrict, Inject;
1059: Vec rscale;
1060: PetscCall(KSPGetDM(mglevels[i + 1]->smoothd, &dmfine));
1061: PetscCall(KSPGetDM(mglevels[i]->smoothd, &dmcoarse));
1062: PetscCall(PCMGGetRestriction(pc, i + 1, &Restrict));
1063: PetscCall(PCMGGetRScale(pc, i + 1, &rscale));
1064: PetscCall(PCMGGetInjection(pc, i + 1, &Inject));
1065: PetscCall(DMRestrict(dmfine, Restrict, rscale, Inject, dmcoarse));
1066: }
1067: }
1069: if (!pc->setupcalled) {
1070: for (i = 0; i < n; i++) PetscCall(KSPSetFromOptions(mglevels[i]->smoothd));
1071: for (i = 1; i < n; i++) {
1072: if (mglevels[i]->smoothu && (mglevels[i]->smoothu != mglevels[i]->smoothd)) PetscCall(KSPSetFromOptions(mglevels[i]->smoothu));
1073: if (mglevels[i]->cr) PetscCall(KSPSetFromOptions(mglevels[i]->cr));
1074: }
1075: /* insure that if either interpolation or restriction is set the other other one is set */
1076: for (i = 1; i < n; i++) {
1077: PetscCall(PCMGGetInterpolation(pc, i, NULL));
1078: PetscCall(PCMGGetRestriction(pc, i, NULL));
1079: }
1080: for (i = 0; i < n - 1; i++) {
1081: if (!mglevels[i]->b) {
1082: Vec *vec;
1083: PetscCall(KSPCreateVecs(mglevels[i]->smoothd, 1, &vec, 0, NULL));
1084: PetscCall(PCMGSetRhs(pc, i, *vec));
1085: PetscCall(VecDestroy(vec));
1086: PetscCall(PetscFree(vec));
1087: }
1088: if (!mglevels[i]->r && i) {
1089: PetscCall(VecDuplicate(mglevels[i]->b, &tvec));
1090: PetscCall(PCMGSetR(pc, i, tvec));
1091: PetscCall(VecDestroy(&tvec));
1092: }
1093: if (!mglevels[i]->x) {
1094: PetscCall(VecDuplicate(mglevels[i]->b, &tvec));
1095: PetscCall(PCMGSetX(pc, i, tvec));
1096: PetscCall(VecDestroy(&tvec));
1097: }
1098: if (doCR) {
1099: PetscCall(VecDuplicate(mglevels[i]->b, &mglevels[i]->crx));
1100: PetscCall(VecDuplicate(mglevels[i]->b, &mglevels[i]->crb));
1101: PetscCall(VecZeroEntries(mglevels[i]->crb));
1102: }
1103: }
1104: if (n != 1 && !mglevels[n - 1]->r) {
1105: /* PCMGSetR() on the finest level if user did not supply it */
1106: Vec *vec;
1107: PetscCall(KSPCreateVecs(mglevels[n - 1]->smoothd, 1, &vec, 0, NULL));
1108: PetscCall(PCMGSetR(pc, n - 1, *vec));
1109: PetscCall(VecDestroy(vec));
1110: PetscCall(PetscFree(vec));
1111: }
1112: if (doCR) {
1113: PetscCall(VecDuplicate(mglevels[n - 1]->r, &mglevels[n - 1]->crx));
1114: PetscCall(VecDuplicate(mglevels[n - 1]->r, &mglevels[n - 1]->crb));
1115: PetscCall(VecZeroEntries(mglevels[n - 1]->crb));
1116: }
1117: }
1119: if (pc->dm) {
1120: /* need to tell all the coarser levels to rebuild the matrix using the DM for that level */
1121: for (i = 0; i < n - 1; i++) {
1122: if (mglevels[i]->smoothd->setupstage != KSP_SETUP_NEW) mglevels[i]->smoothd->setupstage = KSP_SETUP_NEWMATRIX;
1123: }
1124: }
1125: // We got here (PCSetUp_MG) because the matrix has changed, which means the smoother needs to be set up again (e.g.,
1126: // new diagonal for Jacobi). Setting it here allows it to be logged under PCSetUp rather than deep inside a PCApply.
1127: if (mglevels[n - 1]->smoothd->setupstage != KSP_SETUP_NEW) mglevels[n - 1]->smoothd->setupstage = KSP_SETUP_NEWMATRIX;
1129: for (i = 1; i < n; i++) {
1130: if (mglevels[i]->smoothu == mglevels[i]->smoothd || mg->am == PC_MG_FULL || mg->am == PC_MG_KASKADE || mg->cyclesperpcapply > 1) {
1131: /* if doing only down then initial guess is zero */
1132: PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->smoothd, PETSC_TRUE));
1133: }
1134: if (mglevels[i]->cr) PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->cr, PETSC_TRUE));
1135: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventBegin(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1136: PetscCall(KSPSetUp(mglevels[i]->smoothd));
1137: if (mglevels[i]->smoothd->reason) pc->failedreason = PC_SUBPC_ERROR;
1138: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventEnd(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1139: if (!mglevels[i]->residual) {
1140: Mat mat;
1141: PetscCall(KSPGetOperators(mglevels[i]->smoothd, &mat, NULL));
1142: PetscCall(PCMGSetResidual(pc, i, PCMGResidualDefault, mat));
1143: }
1144: if (!mglevels[i]->residualtranspose) {
1145: Mat mat;
1146: PetscCall(KSPGetOperators(mglevels[i]->smoothd, &mat, NULL));
1147: PetscCall(PCMGSetResidualTranspose(pc, i, PCMGResidualTransposeDefault, mat));
1148: }
1149: }
1150: for (i = 1; i < n; i++) {
1151: if (mglevels[i]->smoothu && mglevels[i]->smoothu != mglevels[i]->smoothd) {
1152: Mat downmat, downpmat;
1154: /* check if operators have been set for up, if not use down operators to set them */
1155: PetscCall(KSPGetOperatorsSet(mglevels[i]->smoothu, &opsset, NULL));
1156: if (!opsset) {
1157: PetscCall(KSPGetOperators(mglevels[i]->smoothd, &downmat, &downpmat));
1158: PetscCall(KSPSetOperators(mglevels[i]->smoothu, downmat, downpmat));
1159: }
1161: PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->smoothu, PETSC_TRUE));
1162: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventBegin(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1163: PetscCall(KSPSetUp(mglevels[i]->smoothu));
1164: if (mglevels[i]->smoothu->reason) pc->failedreason = PC_SUBPC_ERROR;
1165: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventEnd(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1166: }
1167: if (mglevels[i]->cr) {
1168: Mat downmat, downpmat;
1170: /* check if operators have been set for up, if not use down operators to set them */
1171: PetscCall(KSPGetOperatorsSet(mglevels[i]->cr, &opsset, NULL));
1172: if (!opsset) {
1173: PetscCall(KSPGetOperators(mglevels[i]->smoothd, &downmat, &downpmat));
1174: PetscCall(KSPSetOperators(mglevels[i]->cr, downmat, downpmat));
1175: }
1177: PetscCall(KSPSetInitialGuessNonzero(mglevels[i]->cr, PETSC_TRUE));
1178: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventBegin(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1179: PetscCall(KSPSetUp(mglevels[i]->cr));
1180: if (mglevels[i]->cr->reason) pc->failedreason = PC_SUBPC_ERROR;
1181: if (mglevels[i]->eventsmoothsetup) PetscCall(PetscLogEventEnd(mglevels[i]->eventsmoothsetup, 0, 0, 0, 0));
1182: }
1183: }
1185: if (mglevels[0]->eventsmoothsetup) PetscCall(PetscLogEventBegin(mglevels[0]->eventsmoothsetup, 0, 0, 0, 0));
1186: PetscCall(KSPSetUp(mglevels[0]->smoothd));
1187: if (mglevels[0]->smoothd->reason) pc->failedreason = PC_SUBPC_ERROR;
1188: if (mglevels[0]->eventsmoothsetup) PetscCall(PetscLogEventEnd(mglevels[0]->eventsmoothsetup, 0, 0, 0, 0));
1190: /*
1191: Dump the interpolation/restriction matrices plus the
1192: Jacobian/stiffness on each level. This allows MATLAB users to
1193: easily check if the Galerkin condition A_c = R A_f R^T is satisfied.
1195: Only support one or the other at the same time.
1196: */
1197: #if defined(PETSC_USE_SOCKET_VIEWER)
1198: PetscCall(PetscOptionsGetBool(((PetscObject)pc)->options, ((PetscObject)pc)->prefix, "-pc_mg_dump_matlab", &dump, NULL));
1199: if (dump) viewer = PETSC_VIEWER_SOCKET_(PetscObjectComm((PetscObject)pc));
1200: dump = PETSC_FALSE;
1201: #endif
1202: PetscCall(PetscOptionsGetBool(((PetscObject)pc)->options, ((PetscObject)pc)->prefix, "-pc_mg_dump_binary", &dump, NULL));
1203: if (dump) viewer = PETSC_VIEWER_BINARY_(PetscObjectComm((PetscObject)pc));
1205: if (viewer) {
1206: for (i = 1; i < n; i++) PetscCall(MatView(mglevels[i]->restrct, viewer));
1207: for (i = 0; i < n; i++) {
1208: PetscCall(KSPGetPC(mglevels[i]->smoothd, &pc));
1209: PetscCall(MatView(pc->mat, viewer));
1210: }
1211: }
1212: PetscFunctionReturn(PETSC_SUCCESS);
1213: }
1215: PetscErrorCode PCMGGetLevels_MG(PC pc, PetscInt *levels)
1216: {
1217: PC_MG *mg = (PC_MG *)pc->data;
1219: PetscFunctionBegin;
1220: *levels = mg->nlevels;
1221: PetscFunctionReturn(PETSC_SUCCESS);
1222: }
1224: /*@
1225: PCMGGetLevels - Gets the number of levels to use with `PCMG`.
1227: Not Collective
1229: Input Parameter:
1230: . pc - the preconditioner context
1232: Output Parameter:
1233: . levels - the number of levels
1235: Level: advanced
1237: .seealso: [](ch_ksp), `PCMG`, `PCMGSetLevels()`
1238: @*/
1239: PetscErrorCode PCMGGetLevels(PC pc, PetscInt *levels)
1240: {
1241: PetscFunctionBegin;
1243: PetscAssertPointer(levels, 2);
1244: *levels = 0;
1245: PetscTryMethod(pc, "PCMGGetLevels_C", (PC, PetscInt *), (pc, levels));
1246: PetscFunctionReturn(PETSC_SUCCESS);
1247: }
1249: /*@
1250: PCMGGetGridComplexity - compute operator and grid complexity of the `PCMG` hierarchy
1252: Input Parameter:
1253: . pc - the preconditioner context
1255: Output Parameters:
1256: + gc - grid complexity = sum_i(n_i) / n_0
1257: - oc - operator complexity = sum_i(nnz_i) / nnz_0
1259: Level: advanced
1261: Note:
1262: This is often call the operator complexity in multigrid literature
1264: .seealso: [](ch_ksp), `PCMG`, `PCMGGetLevels()`, `PCMGSetLevels()`
1265: @*/
1266: PetscErrorCode PCMGGetGridComplexity(PC pc, PetscReal *gc, PetscReal *oc)
1267: {
1268: PC_MG *mg = (PC_MG *)pc->data;
1269: PC_MG_Levels **mglevels = mg->levels;
1270: PetscInt lev, N;
1271: PetscLogDouble nnz0 = 0, sgc = 0, soc = 0, n0 = 0;
1272: MatInfo info;
1274: PetscFunctionBegin;
1276: if (gc) PetscAssertPointer(gc, 2);
1277: if (oc) PetscAssertPointer(oc, 3);
1278: if (!pc->setupcalled) {
1279: if (gc) *gc = 0;
1280: if (oc) *oc = 0;
1281: PetscFunctionReturn(PETSC_SUCCESS);
1282: }
1283: PetscCheck(mg->nlevels > 0, PETSC_COMM_SELF, PETSC_ERR_PLIB, "MG has no levels");
1284: for (lev = 0; lev < mg->nlevels; lev++) {
1285: Mat dB;
1286: PetscCall(KSPGetOperators(mglevels[lev]->smoothd, NULL, &dB));
1287: PetscCall(MatGetInfo(dB, MAT_GLOBAL_SUM, &info)); /* global reduction */
1288: PetscCall(MatGetSize(dB, &N, NULL));
1289: sgc += N;
1290: soc += info.nz_used;
1291: if (lev == mg->nlevels - 1) {
1292: nnz0 = info.nz_used;
1293: n0 = N;
1294: }
1295: }
1296: PetscCheck(n0 > 0 && gc, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Number for grid points on finest level is not available");
1297: *gc = (PetscReal)(sgc / n0);
1298: if (nnz0 > 0 && oc) *oc = (PetscReal)(soc / nnz0);
1299: PetscFunctionReturn(PETSC_SUCCESS);
1300: }
1302: /*@
1303: PCMGSetType - Determines the form of multigrid to use, either
1304: multiplicative, additive, full, or the Kaskade algorithm.
1306: Logically Collective
1308: Input Parameters:
1309: + pc - the preconditioner context
1310: - form - multigrid form, one of `PC_MG_MULTIPLICATIVE`, `PC_MG_ADDITIVE`, `PC_MG_FULL`, `PC_MG_KASKADE`
1312: Options Database Key:
1313: . -pc_mg_type <form> - Sets <form>, one of multiplicative, additive, full, kaskade
1315: Level: advanced
1317: .seealso: [](ch_ksp), `PCMGType`, `PCMG`, `PCMGGetLevels()`, `PCMGSetLevels()`, `PCMGGetType()`, `PCMGCycleType`
1318: @*/
1319: PetscErrorCode PCMGSetType(PC pc, PCMGType form)
1320: {
1321: PC_MG *mg = (PC_MG *)pc->data;
1323: PetscFunctionBegin;
1326: mg->am = form;
1327: if (form == PC_MG_MULTIPLICATIVE) pc->ops->applyrichardson = PCApplyRichardson_MG;
1328: else pc->ops->applyrichardson = NULL;
1329: PetscFunctionReturn(PETSC_SUCCESS);
1330: }
1332: /*@
1333: PCMGGetType - Finds the form of multigrid the `PCMG` is using multiplicative, additive, full, or the Kaskade algorithm.
1335: Logically Collective
1337: Input Parameter:
1338: . pc - the preconditioner context
1340: Output Parameter:
1341: . type - one of `PC_MG_MULTIPLICATIVE`, `PC_MG_ADDITIVE`, `PC_MG_FULL`, `PC_MG_KASKADE`, `PCMGCycleType`
1343: Level: advanced
1345: .seealso: [](ch_ksp), `PCMGType`, `PCMG`, `PCMGGetLevels()`, `PCMGSetLevels()`, `PCMGSetType()`
1346: @*/
1347: PetscErrorCode PCMGGetType(PC pc, PCMGType *type)
1348: {
1349: PC_MG *mg = (PC_MG *)pc->data;
1351: PetscFunctionBegin;
1353: *type = mg->am;
1354: PetscFunctionReturn(PETSC_SUCCESS);
1355: }
1357: /*@
1358: PCMGSetCycleType - Sets the type cycles to use. Use `PCMGSetCycleTypeOnLevel()` for more
1359: complicated cycling.
1361: Logically Collective
1363: Input Parameters:
1364: + pc - the multigrid context
1365: - n - either `PC_MG_CYCLE_V` or `PC_MG_CYCLE_W`
1367: Options Database Key:
1368: . -pc_mg_cycle_type <v,w> - provide the cycle desired
1370: Level: advanced
1372: .seealso: [](ch_ksp), `PCMG`, `PCMGSetCycleTypeOnLevel()`, `PCMGType`, `PCMGCycleType`
1373: @*/
1374: PetscErrorCode PCMGSetCycleType(PC pc, PCMGCycleType n)
1375: {
1376: PC_MG *mg = (PC_MG *)pc->data;
1377: PC_MG_Levels **mglevels = mg->levels;
1378: PetscInt i, levels;
1380: PetscFunctionBegin;
1383: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ORDER, "Must set MG levels with PCMGSetLevels() before calling");
1384: levels = mglevels[0]->levels;
1385: for (i = 0; i < levels; i++) mglevels[i]->cycles = n;
1386: PetscFunctionReturn(PETSC_SUCCESS);
1387: }
1389: /*@
1390: PCMGMultiplicativeSetCycles - Sets the number of cycles to use for each preconditioner step
1391: of multigrid when `PCMGType` is `PC_MG_MULTIPLICATIVE`
1393: Logically Collective
1395: Input Parameters:
1396: + pc - the multigrid context
1397: - n - number of cycles (default is 1)
1399: Options Database Key:
1400: . -pc_mg_multiplicative_cycles n - set the number of cycles
1402: Level: advanced
1404: Note:
1405: This is not associated with setting a v or w cycle, that is set with `PCMGSetCycleType()`
1407: .seealso: [](ch_ksp), `PCMGSetCycleTypeOnLevel()`, `PCMGSetCycleType()`, `PCMGCycleType`, `PCMGType`
1408: @*/
1409: PetscErrorCode PCMGMultiplicativeSetCycles(PC pc, PetscInt n)
1410: {
1411: PC_MG *mg = (PC_MG *)pc->data;
1413: PetscFunctionBegin;
1416: mg->cyclesperpcapply = n;
1417: PetscFunctionReturn(PETSC_SUCCESS);
1418: }
1420: static PetscErrorCode PCMGSetGalerkin_MG(PC pc, PCMGGalerkinType use)
1421: {
1422: PC_MG *mg = (PC_MG *)pc->data;
1424: PetscFunctionBegin;
1425: mg->galerkin = use;
1426: PetscFunctionReturn(PETSC_SUCCESS);
1427: }
1429: /*@
1430: PCMGSetGalerkin - Causes the coarser grid matrices to be computed from the
1431: finest grid via the Galerkin process: A_i-1 = r_i * A_i * p_i
1433: Logically Collective
1435: Input Parameters:
1436: + pc - the multigrid context
1437: - use - one of `PC_MG_GALERKIN_BOTH`, `PC_MG_GALERKIN_PMAT`, `PC_MG_GALERKIN_MAT`, or `PC_MG_GALERKIN_NONE`
1439: Options Database Key:
1440: . -pc_mg_galerkin <both,pmat,mat,none> - set the matrices to form via the Galerkin process
1442: Level: intermediate
1444: Note:
1445: Some codes that use `PCMG` such as `PCGAMG` use Galerkin internally while constructing the hierarchy and thus do not
1446: use the `PCMG` construction of the coarser grids.
1448: .seealso: [](ch_ksp), `PCMG`, `PCMGGetGalerkin()`, `PCMGGalerkinType`
1449: @*/
1450: PetscErrorCode PCMGSetGalerkin(PC pc, PCMGGalerkinType use)
1451: {
1452: PetscFunctionBegin;
1454: PetscTryMethod(pc, "PCMGSetGalerkin_C", (PC, PCMGGalerkinType), (pc, use));
1455: PetscFunctionReturn(PETSC_SUCCESS);
1456: }
1458: /*@
1459: PCMGGetGalerkin - Checks if Galerkin multigrid is being used, i.e. A_i-1 = r_i * A_i * p_i
1461: Not Collective
1463: Input Parameter:
1464: . pc - the multigrid context
1466: Output Parameter:
1467: . galerkin - one of `PC_MG_GALERKIN_BOTH`,`PC_MG_GALERKIN_PMAT`,`PC_MG_GALERKIN_MAT`, `PC_MG_GALERKIN_NONE`, or `PC_MG_GALERKIN_EXTERNAL`
1469: Level: intermediate
1471: .seealso: [](ch_ksp), `PCMG`, `PCMGSetGalerkin()`, `PCMGGalerkinType`
1472: @*/
1473: PetscErrorCode PCMGGetGalerkin(PC pc, PCMGGalerkinType *galerkin)
1474: {
1475: PC_MG *mg = (PC_MG *)pc->data;
1477: PetscFunctionBegin;
1479: *galerkin = mg->galerkin;
1480: PetscFunctionReturn(PETSC_SUCCESS);
1481: }
1483: static PetscErrorCode PCMGSetAdaptInterpolation_MG(PC pc, PetscBool adapt)
1484: {
1485: PC_MG *mg = (PC_MG *)pc->data;
1487: PetscFunctionBegin;
1488: mg->adaptInterpolation = adapt;
1489: PetscFunctionReturn(PETSC_SUCCESS);
1490: }
1492: static PetscErrorCode PCMGGetAdaptInterpolation_MG(PC pc, PetscBool *adapt)
1493: {
1494: PC_MG *mg = (PC_MG *)pc->data;
1496: PetscFunctionBegin;
1497: *adapt = mg->adaptInterpolation;
1498: PetscFunctionReturn(PETSC_SUCCESS);
1499: }
1501: static PetscErrorCode PCMGSetAdaptCoarseSpaceType_MG(PC pc, PCMGCoarseSpaceType ctype)
1502: {
1503: PC_MG *mg = (PC_MG *)pc->data;
1505: PetscFunctionBegin;
1506: mg->adaptInterpolation = ctype != PCMG_ADAPT_NONE ? PETSC_TRUE : PETSC_FALSE;
1507: mg->coarseSpaceType = ctype;
1508: PetscFunctionReturn(PETSC_SUCCESS);
1509: }
1511: static PetscErrorCode PCMGGetAdaptCoarseSpaceType_MG(PC pc, PCMGCoarseSpaceType *ctype)
1512: {
1513: PC_MG *mg = (PC_MG *)pc->data;
1515: PetscFunctionBegin;
1516: *ctype = mg->coarseSpaceType;
1517: PetscFunctionReturn(PETSC_SUCCESS);
1518: }
1520: static PetscErrorCode PCMGSetAdaptCR_MG(PC pc, PetscBool cr)
1521: {
1522: PC_MG *mg = (PC_MG *)pc->data;
1524: PetscFunctionBegin;
1525: mg->compatibleRelaxation = cr;
1526: PetscFunctionReturn(PETSC_SUCCESS);
1527: }
1529: static PetscErrorCode PCMGGetAdaptCR_MG(PC pc, PetscBool *cr)
1530: {
1531: PC_MG *mg = (PC_MG *)pc->data;
1533: PetscFunctionBegin;
1534: *cr = mg->compatibleRelaxation;
1535: PetscFunctionReturn(PETSC_SUCCESS);
1536: }
1538: /*@C
1539: PCMGSetAdaptCoarseSpaceType - Set the type of adaptive coarse space.
1541: Adapts or creates the interpolator based upon a vector space which should be accurately captured by the next coarser mesh, and thus accurately interpolated.
1543: Logically Collective
1545: Input Parameters:
1546: + pc - the multigrid context
1547: - ctype - the type of coarse space
1549: Options Database Keys:
1550: + -pc_mg_adapt_interp_n <int> - The number of modes to use
1551: - -pc_mg_adapt_interp_coarse_space <type> - The type of coarse space: none, polynomial, harmonic, eigenvector, generalized_eigenvector, gdsw
1553: Level: intermediate
1555: .seealso: [](ch_ksp), `PCMG`, `PCMGCoarseSpaceType`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetGalerkin()`, `PCMGSetAdaptInterpolation()`
1556: @*/
1557: PetscErrorCode PCMGSetAdaptCoarseSpaceType(PC pc, PCMGCoarseSpaceType ctype)
1558: {
1559: PetscFunctionBegin;
1562: PetscTryMethod(pc, "PCMGSetAdaptCoarseSpaceType_C", (PC, PCMGCoarseSpaceType), (pc, ctype));
1563: PetscFunctionReturn(PETSC_SUCCESS);
1564: }
1566: /*@C
1567: PCMGGetAdaptCoarseSpaceType - Get the type of adaptive coarse space.
1569: Not Collective
1571: Input Parameter:
1572: . pc - the multigrid context
1574: Output Parameter:
1575: . ctype - the type of coarse space
1577: Level: intermediate
1579: .seealso: [](ch_ksp), `PCMG`, `PCMGCoarseSpaceType`, `PCMGSetAdaptCoarseSpaceType()`, `PCMGSetGalerkin()`, `PCMGSetAdaptInterpolation()`
1580: @*/
1581: PetscErrorCode PCMGGetAdaptCoarseSpaceType(PC pc, PCMGCoarseSpaceType *ctype)
1582: {
1583: PetscFunctionBegin;
1585: PetscAssertPointer(ctype, 2);
1586: PetscUseMethod(pc, "PCMGGetAdaptCoarseSpaceType_C", (PC, PCMGCoarseSpaceType *), (pc, ctype));
1587: PetscFunctionReturn(PETSC_SUCCESS);
1588: }
1590: /* MATT: REMOVE? */
1591: /*@
1592: PCMGSetAdaptInterpolation - Adapt the interpolator based upon a vector space which should be accurately captured by the next coarser mesh, and thus accurately interpolated.
1594: Logically Collective
1596: Input Parameters:
1597: + pc - the multigrid context
1598: - adapt - flag for adaptation of the interpolator
1600: Options Database Keys:
1601: + -pc_mg_adapt_interp - Turn on adaptation
1602: . -pc_mg_adapt_interp_n <int> - The number of modes to use, should be divisible by dimension
1603: - -pc_mg_adapt_interp_coarse_space <type> - The type of coarse space: polynomial, harmonic, eigenvector, generalized_eigenvector
1605: Level: intermediate
1607: .seealso: [](ch_ksp), `PCMG`, `PCMGGetAdaptInterpolation()`, `PCMGSetGalerkin()`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetAdaptCoarseSpaceType()`
1608: @*/
1609: PetscErrorCode PCMGSetAdaptInterpolation(PC pc, PetscBool adapt)
1610: {
1611: PetscFunctionBegin;
1613: PetscTryMethod(pc, "PCMGSetAdaptInterpolation_C", (PC, PetscBool), (pc, adapt));
1614: PetscFunctionReturn(PETSC_SUCCESS);
1615: }
1617: /*@
1618: PCMGGetAdaptInterpolation - Get the flag to adapt the interpolator based upon a vector space which should be accurately captured by the next coarser mesh,
1619: and thus accurately interpolated.
1621: Not Collective
1623: Input Parameter:
1624: . pc - the multigrid context
1626: Output Parameter:
1627: . adapt - flag for adaptation of the interpolator
1629: Level: intermediate
1631: .seealso: [](ch_ksp), `PCMG`, `PCMGSetAdaptInterpolation()`, `PCMGSetGalerkin()`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetAdaptCoarseSpaceType()`
1632: @*/
1633: PetscErrorCode PCMGGetAdaptInterpolation(PC pc, PetscBool *adapt)
1634: {
1635: PetscFunctionBegin;
1637: PetscAssertPointer(adapt, 2);
1638: PetscUseMethod(pc, "PCMGGetAdaptInterpolation_C", (PC, PetscBool *), (pc, adapt));
1639: PetscFunctionReturn(PETSC_SUCCESS);
1640: }
1642: /*@
1643: PCMGSetAdaptCR - Monitor the coarse space quality using an auxiliary solve with compatible relaxation.
1645: Logically Collective
1647: Input Parameters:
1648: + pc - the multigrid context
1649: - cr - flag for compatible relaxation
1651: Options Database Key:
1652: . -pc_mg_adapt_cr - Turn on compatible relaxation
1654: Level: intermediate
1656: .seealso: [](ch_ksp), `PCMG`, `PCMGGetAdaptCR()`, `PCMGSetAdaptInterpolation()`, `PCMGSetGalerkin()`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetAdaptCoarseSpaceType()`
1657: @*/
1658: PetscErrorCode PCMGSetAdaptCR(PC pc, PetscBool cr)
1659: {
1660: PetscFunctionBegin;
1662: PetscTryMethod(pc, "PCMGSetAdaptCR_C", (PC, PetscBool), (pc, cr));
1663: PetscFunctionReturn(PETSC_SUCCESS);
1664: }
1666: /*@
1667: PCMGGetAdaptCR - Get the flag to monitor coarse space quality using an auxiliary solve with compatible relaxation.
1669: Not Collective
1671: Input Parameter:
1672: . pc - the multigrid context
1674: Output Parameter:
1675: . cr - flag for compatible relaxaion
1677: Level: intermediate
1679: .seealso: [](ch_ksp), `PCMGSetAdaptCR()`, `PCMGGetAdaptInterpolation()`, `PCMGSetGalerkin()`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetAdaptCoarseSpaceType()`
1680: @*/
1681: PetscErrorCode PCMGGetAdaptCR(PC pc, PetscBool *cr)
1682: {
1683: PetscFunctionBegin;
1685: PetscAssertPointer(cr, 2);
1686: PetscUseMethod(pc, "PCMGGetAdaptCR_C", (PC, PetscBool *), (pc, cr));
1687: PetscFunctionReturn(PETSC_SUCCESS);
1688: }
1690: /*@
1691: PCMGSetNumberSmooth - Sets the number of pre and post-smoothing steps to use
1692: on all levels. Use `PCMGDistinctSmoothUp()` to create separate up and down smoothers if you want different numbers of
1693: pre- and post-smoothing steps.
1695: Logically Collective
1697: Input Parameters:
1698: + pc - the multigrid context
1699: - n - the number of smoothing steps
1701: Options Database Key:
1702: . -mg_levels_ksp_max_it <n> - Sets number of pre and post-smoothing steps
1704: Level: advanced
1706: Note:
1707: This does not set a value on the coarsest grid, since we assume that there is no separate smooth up on the coarsest grid.
1709: .seealso: [](ch_ksp), `PCMG`, `PCMGSetDistinctSmoothUp()`
1710: @*/
1711: PetscErrorCode PCMGSetNumberSmooth(PC pc, PetscInt n)
1712: {
1713: PC_MG *mg = (PC_MG *)pc->data;
1714: PC_MG_Levels **mglevels = mg->levels;
1715: PetscInt i, levels;
1717: PetscFunctionBegin;
1720: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ORDER, "Must set MG levels with PCMGSetLevels() before calling");
1721: levels = mglevels[0]->levels;
1723: for (i = 1; i < levels; i++) {
1724: PetscCall(KSPSetTolerances(mglevels[i]->smoothu, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT, n));
1725: PetscCall(KSPSetTolerances(mglevels[i]->smoothd, PETSC_DEFAULT, PETSC_DEFAULT, PETSC_DEFAULT, n));
1726: mg->default_smoothu = n;
1727: mg->default_smoothd = n;
1728: }
1729: PetscFunctionReturn(PETSC_SUCCESS);
1730: }
1732: /*@
1733: PCMGSetDistinctSmoothUp - sets the up (post) smoother to be a separate `KSP` from the down (pre) smoother on all levels
1734: and adds the suffix _up to the options name
1736: Logically Collective
1738: Input Parameter:
1739: . pc - the preconditioner context
1741: Options Database Key:
1742: . -pc_mg_distinct_smoothup <bool> - use distinct smoothing objects
1744: Level: advanced
1746: Note:
1747: This does not set a value on the coarsest grid, since we assume that there is no separate smooth up on the coarsest grid.
1749: .seealso: [](ch_ksp), `PCMG`, `PCMGSetNumberSmooth()`
1750: @*/
1751: PetscErrorCode PCMGSetDistinctSmoothUp(PC pc)
1752: {
1753: PC_MG *mg = (PC_MG *)pc->data;
1754: PC_MG_Levels **mglevels = mg->levels;
1755: PetscInt i, levels;
1756: KSP subksp;
1758: PetscFunctionBegin;
1760: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ORDER, "Must set MG levels with PCMGSetLevels() before calling");
1761: levels = mglevels[0]->levels;
1763: for (i = 1; i < levels; i++) {
1764: const char *prefix = NULL;
1765: /* make sure smoother up and down are different */
1766: PetscCall(PCMGGetSmootherUp(pc, i, &subksp));
1767: PetscCall(KSPGetOptionsPrefix(mglevels[i]->smoothd, &prefix));
1768: PetscCall(KSPSetOptionsPrefix(subksp, prefix));
1769: PetscCall(KSPAppendOptionsPrefix(subksp, "up_"));
1770: }
1771: PetscFunctionReturn(PETSC_SUCCESS);
1772: }
1774: /* No new matrices are created, and the coarse operator matrices are the references to the original ones */
1775: static PetscErrorCode PCGetInterpolations_MG(PC pc, PetscInt *num_levels, Mat *interpolations[])
1776: {
1777: PC_MG *mg = (PC_MG *)pc->data;
1778: PC_MG_Levels **mglevels = mg->levels;
1779: Mat *mat;
1780: PetscInt l;
1782: PetscFunctionBegin;
1783: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_WRONGSTATE, "Must set MG levels before calling");
1784: PetscCall(PetscMalloc1(mg->nlevels, &mat));
1785: for (l = 1; l < mg->nlevels; l++) {
1786: mat[l - 1] = mglevels[l]->interpolate;
1787: PetscCall(PetscObjectReference((PetscObject)mat[l - 1]));
1788: }
1789: *num_levels = mg->nlevels;
1790: *interpolations = mat;
1791: PetscFunctionReturn(PETSC_SUCCESS);
1792: }
1794: /* No new matrices are created, and the coarse operator matrices are the references to the original ones */
1795: static PetscErrorCode PCGetCoarseOperators_MG(PC pc, PetscInt *num_levels, Mat *coarseOperators[])
1796: {
1797: PC_MG *mg = (PC_MG *)pc->data;
1798: PC_MG_Levels **mglevels = mg->levels;
1799: PetscInt l;
1800: Mat *mat;
1802: PetscFunctionBegin;
1803: PetscCheck(mglevels, PetscObjectComm((PetscObject)pc), PETSC_ERR_ARG_WRONGSTATE, "Must set MG levels before calling");
1804: PetscCall(PetscMalloc1(mg->nlevels, &mat));
1805: for (l = 0; l < mg->nlevels - 1; l++) {
1806: PetscCall(KSPGetOperators(mglevels[l]->smoothd, NULL, &(mat[l])));
1807: PetscCall(PetscObjectReference((PetscObject)mat[l]));
1808: }
1809: *num_levels = mg->nlevels;
1810: *coarseOperators = mat;
1811: PetscFunctionReturn(PETSC_SUCCESS);
1812: }
1814: /*@C
1815: PCMGRegisterCoarseSpaceConstructor - Adds a method to the `PCMG` package for coarse space construction.
1817: Not Collective
1819: Input Parameters:
1820: + name - name of the constructor
1821: - function - constructor routine
1823: Calling sequence of `function`:
1824: + pc - The `PC` object
1825: . l - The multigrid level, 0 is the coarse level
1826: . dm - The `DM` for this level
1827: . smooth - The level smoother
1828: . Nc - The size of the coarse space
1829: . initGuess - Basis for an initial guess for the space
1830: - coarseSp - A basis for the computed coarse space
1832: Level: advanced
1834: Developer Notes:
1835: How come this is not used by `PCGAMG`?
1837: .seealso: [](ch_ksp), `PCMG`, `PCMGGetCoarseSpaceConstructor()`, `PCRegister()`
1838: @*/
1839: PetscErrorCode PCMGRegisterCoarseSpaceConstructor(const char name[], PetscErrorCode (*function)(PC pc, PetscInt l, DM dm, KSP smooth, PetscInt Nc, Mat initGuess, Mat *coarseSp))
1840: {
1841: PetscFunctionBegin;
1842: PetscCall(PCInitializePackage());
1843: PetscCall(PetscFunctionListAdd(&PCMGCoarseList, name, function));
1844: PetscFunctionReturn(PETSC_SUCCESS);
1845: }
1847: /*@C
1848: PCMGGetCoarseSpaceConstructor - Returns the given coarse space construction method.
1850: Not Collective
1852: Input Parameter:
1853: . name - name of the constructor
1855: Output Parameter:
1856: . function - constructor routine
1858: Level: advanced
1860: .seealso: [](ch_ksp), `PCMG`, `PCMGRegisterCoarseSpaceConstructor()`, `PCRegister()`
1861: @*/
1862: PetscErrorCode PCMGGetCoarseSpaceConstructor(const char name[], PetscErrorCode (**function)(PC, PetscInt, DM, KSP, PetscInt, Mat, Mat *))
1863: {
1864: PetscFunctionBegin;
1865: PetscCall(PetscFunctionListFind(PCMGCoarseList, name, function));
1866: PetscFunctionReturn(PETSC_SUCCESS);
1867: }
1869: /*MC
1870: PCMG - Use multigrid preconditioning. This preconditioner requires you provide additional
1871: information about the coarser grid matrices and restriction/interpolation operators.
1873: Options Database Keys:
1874: + -pc_mg_levels <nlevels> - number of levels including finest
1875: . -pc_mg_cycle_type <v,w> - provide the cycle desired
1876: . -pc_mg_type <additive,multiplicative,full,kaskade> - multiplicative is the default
1877: . -pc_mg_log - log information about time spent on each level of the solver
1878: . -pc_mg_distinct_smoothup - configure up (after interpolation) and down (before restriction) smoothers separately (with different options prefixes)
1879: . -pc_mg_galerkin <both,pmat,mat,none> - use Galerkin process to compute coarser operators, i.e. Acoarse = R A R'
1880: . -pc_mg_multiplicative_cycles - number of cycles to use as the preconditioner (defaults to 1)
1881: . -pc_mg_dump_matlab - dumps the matrices for each level and the restriction/interpolation matrices
1882: to the Socket viewer for reading from MATLAB.
1883: - -pc_mg_dump_binary - dumps the matrices for each level and the restriction/interpolation matrices
1884: to the binary output file called binaryoutput
1886: Level: intermediate
1888: Notes:
1889: The Krylov solver (if any) and preconditioner (smoother) and their parameters are controlled from the options database with the standard
1890: options database keywords prefixed with `-mg_levels_` to affect all the levels but the coarsest, which is controlled with `-mg_coarse_`.
1891: One can set different preconditioners etc on specific levels with the prefix `-mg_levels_n_` where `n` is the level number (zero being
1892: the coarse level. For example
1893: .vb
1894: -mg_levels_ksp_type gmres -mg_levels_pc_type bjacobi -mg_coarse_pc_type svd -mg_levels_2_pc_type sor
1895: .ve
1896: These options also work for controlling the smoothers etc inside `PCGAMG`
1898: If one uses a Krylov method such `KSPGMRES` or `KSPCG` as the smoother then one must use `KSPFGMRES`, `KSPGCR`, or `KSPRICHARDSON` as the outer Krylov method
1900: When run with a single level the smoother options are used on that level NOT the coarse grid solver options
1902: When run with `KSPRICHARDSON` the convergence test changes slightly if monitor is turned on. The iteration count may change slightly. This
1903: is because without monitoring the residual norm is computed WITHIN each multigrid cycle on the finest level after the pre-smoothing
1904: (because the residual has just been computed for the multigrid algorithm and is hence available for free) while with monitoring the
1905: residual is computed at the end of each cycle.
1907: .seealso: [](sec_mg), `PCCreate()`, `PCSetType()`, `PCType`, `PC`, `PCMGType`, `PCEXOTIC`, `PCGAMG`, `PCML`, `PCHYPRE`
1908: `PCMGSetLevels()`, `PCMGGetLevels()`, `PCMGSetType()`, `PCMGSetCycleType()`,
1909: `PCMGSetDistinctSmoothUp()`, `PCMGGetCoarseSolve()`, `PCMGSetResidual()`, `PCMGSetInterpolation()`,
1910: `PCMGSetRestriction()`, `PCMGGetSmoother()`, `PCMGGetSmootherUp()`, `PCMGGetSmootherDown()`,
1911: `PCMGSetCycleTypeOnLevel()`, `PCMGSetRhs()`, `PCMGSetX()`, `PCMGSetR()`,
1912: `PCMGSetAdaptCR()`, `PCMGGetAdaptInterpolation()`, `PCMGSetGalerkin()`, `PCMGGetAdaptCoarseSpaceType()`, `PCMGSetAdaptCoarseSpaceType()`
1913: M*/
1915: PETSC_EXTERN PetscErrorCode PCCreate_MG(PC pc)
1916: {
1917: PC_MG *mg;
1919: PetscFunctionBegin;
1920: PetscCall(PetscNew(&mg));
1921: pc->data = mg;
1922: mg->nlevels = -1;
1923: mg->am = PC_MG_MULTIPLICATIVE;
1924: mg->galerkin = PC_MG_GALERKIN_NONE;
1925: mg->adaptInterpolation = PETSC_FALSE;
1926: mg->Nc = -1;
1927: mg->eigenvalue = -1;
1929: pc->useAmat = PETSC_TRUE;
1931: pc->ops->apply = PCApply_MG;
1932: pc->ops->applytranspose = PCApplyTranspose_MG;
1933: pc->ops->matapply = PCMatApply_MG;
1934: pc->ops->setup = PCSetUp_MG;
1935: pc->ops->reset = PCReset_MG;
1936: pc->ops->destroy = PCDestroy_MG;
1937: pc->ops->setfromoptions = PCSetFromOptions_MG;
1938: pc->ops->view = PCView_MG;
1940: PetscCall(PetscObjectComposedDataRegister(&mg->eigenvalue));
1941: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetGalerkin_C", PCMGSetGalerkin_MG));
1942: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetLevels_C", PCMGGetLevels_MG));
1943: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetLevels_C", PCMGSetLevels_MG));
1944: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetInterpolations_C", PCGetInterpolations_MG));
1945: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCGetCoarseOperators_C", PCGetCoarseOperators_MG));
1946: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptInterpolation_C", PCMGSetAdaptInterpolation_MG));
1947: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptInterpolation_C", PCMGGetAdaptInterpolation_MG));
1948: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptCR_C", PCMGSetAdaptCR_MG));
1949: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptCR_C", PCMGGetAdaptCR_MG));
1950: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGSetAdaptCoarseSpaceType_C", PCMGSetAdaptCoarseSpaceType_MG));
1951: PetscCall(PetscObjectComposeFunction((PetscObject)pc, "PCMGGetAdaptCoarseSpaceType_C", PCMGGetAdaptCoarseSpaceType_MG));
1952: PetscFunctionReturn(PETSC_SUCCESS);
1953: }