1 #define PETSCKSP_DLL 2 3 /* 4 Defines the multigrid preconditioner interface. 5 */ 6 #include "src/ksp/pc/impls/mg/mgimpl.h" /*I "petscmg.h" I*/ 7 8 9 #undef __FUNCT__ 10 #define __FUNCT__ "PCMGMCycle_Private" 11 PetscErrorCode PCMGMCycle_Private(PC pc,PC_MG **mglevels,PCRichardsonConvergedReason *reason) 12 { 13 PC_MG *mg = *mglevels,*mgc; 14 PetscErrorCode ierr; 15 PetscInt cycles = (PetscInt) mg->cycles; 16 17 PetscFunctionBegin; 18 19 if (mg->eventsmoothsolve) {ierr = PetscLogEventBegin(mg->eventsmoothsolve,0,0,0,0);CHKERRQ(ierr);} 20 ierr = KSPSolve(mg->smoothd,mg->b,mg->x);CHKERRQ(ierr); /* pre-smooth */ 21 if (mg->eventsmoothsolve) {ierr = PetscLogEventEnd(mg->eventsmoothsolve,0,0,0,0);CHKERRQ(ierr);} 22 if (mg->level) { /* not the coarsest grid */ 23 if (mg->eventresidual) {ierr = PetscLogEventBegin(mg->eventresidual,0,0,0,0);CHKERRQ(ierr);} 24 ierr = (*mg->residual)(mg->A,mg->b,mg->x,mg->r);CHKERRQ(ierr); 25 if (mg->eventresidual) {ierr = PetscLogEventEnd(mg->eventresidual,0,0,0,0);CHKERRQ(ierr);} 26 27 /* if on finest level and have convergence criteria set */ 28 if (mg->level == mg->levels-1 && mg->ttol && reason) { 29 PetscReal rnorm; 30 ierr = VecNorm(mg->r,NORM_2,&rnorm);CHKERRQ(ierr); 31 if (rnorm <= mg->ttol) { 32 if (rnorm < mg->abstol) { 33 *reason = PCRICHARDSON_CONVERGED_ATOL; 34 ierr = PetscInfo2(pc,"Linear solver has converged. Residual norm %G is less than absolute tolerance %G\n",rnorm,mg->abstol);CHKERRQ(ierr); 35 } else { 36 *reason = PCRICHARDSON_CONVERGED_RTOL; 37 ierr = PetscInfo2(pc,"Linear solver has converged. Residual norm %G is less than relative tolerance times initial residual norm %G\n",rnorm,mg->ttol);CHKERRQ(ierr); 38 } 39 PetscFunctionReturn(0); 40 } 41 } 42 43 mgc = *(mglevels - 1); 44 if (mg->eventinterprestrict) {ierr = PetscLogEventBegin(mg->eventinterprestrict,0,0,0,0);CHKERRQ(ierr);} 45 ierr = MatRestrict(mg->restrct,mg->r,mgc->b);CHKERRQ(ierr); 46 if (mg->eventinterprestrict) {ierr = PetscLogEventEnd(mg->eventinterprestrict,0,0,0,0);CHKERRQ(ierr);} 47 ierr = VecSet(mgc->x,0.0);CHKERRQ(ierr); 48 while (cycles--) { 49 ierr = PCMGMCycle_Private(pc,mglevels-1,reason);CHKERRQ(ierr); 50 } 51 if (mg->eventinterprestrict) {ierr = PetscLogEventBegin(mg->eventinterprestrict,0,0,0,0);CHKERRQ(ierr);} 52 ierr = MatInterpolateAdd(mg->interpolate,mgc->x,mg->x,mg->x);CHKERRQ(ierr); 53 if (mg->eventinterprestrict) {ierr = PetscLogEventEnd(mg->eventinterprestrict,0,0,0,0);CHKERRQ(ierr);} 54 if (mg->eventsmoothsolve) {ierr = PetscLogEventBegin(mg->eventsmoothsolve,0,0,0,0);CHKERRQ(ierr);} 55 ierr = KSPSolve(mg->smoothu,mg->b,mg->x);CHKERRQ(ierr); /* post smooth */ 56 if (mg->eventsmoothsolve) {ierr = PetscLogEventEnd(mg->eventsmoothsolve,0,0,0,0);CHKERRQ(ierr);} 57 } 58 PetscFunctionReturn(0); 59 } 60 61 /* 62 PCMGCreate_Private - Creates a PC_MG structure for use with the 63 multigrid code. Level 0 is the coarsest. (But the 64 finest level is stored first in the array). 65 66 */ 67 #undef __FUNCT__ 68 #define __FUNCT__ "PCMGCreate_Private" 69 static PetscErrorCode PCMGCreate_Private(MPI_Comm comm,PetscInt levels,PC pc,MPI_Comm *comms,PC_MG ***result) 70 { 71 PC_MG **mg; 72 PetscErrorCode ierr; 73 PetscInt i; 74 PetscMPIInt size; 75 const char *prefix; 76 PC ipc; 77 78 PetscFunctionBegin; 79 ierr = PetscMalloc(levels*sizeof(PC_MG*),&mg);CHKERRQ(ierr); 80 ierr = PetscLogObjectMemory(pc,levels*(sizeof(PC_MG*)));CHKERRQ(ierr); 81 82 ierr = PCGetOptionsPrefix(pc,&prefix);CHKERRQ(ierr); 83 84 for (i=0; i<levels; i++) { 85 ierr = PetscNewLog(pc,PC_MG,&mg[i]);CHKERRQ(ierr); 86 mg[i]->level = i; 87 mg[i]->levels = levels; 88 mg[i]->cycles = PC_MG_CYCLE_V; 89 mg[i]->galerkin = PETSC_FALSE; 90 mg[i]->galerkinused = PETSC_FALSE; 91 mg[i]->default_smoothu = 1; 92 mg[i]->default_smoothd = 1; 93 94 if (comms) comm = comms[i]; 95 ierr = KSPCreate(comm,&mg[i]->smoothd);CHKERRQ(ierr); 96 ierr = PetscObjectIncrementTabLevel((PetscObject)mg[i]->smoothd,(PetscObject)pc,levels-i);CHKERRQ(ierr); 97 ierr = KSPSetTolerances(mg[i]->smoothd,PETSC_DEFAULT,PETSC_DEFAULT,PETSC_DEFAULT, mg[i]->default_smoothd);CHKERRQ(ierr); 98 ierr = KSPSetOptionsPrefix(mg[i]->smoothd,prefix);CHKERRQ(ierr); 99 100 /* do special stuff for coarse grid */ 101 if (!i && levels > 1) { 102 ierr = KSPAppendOptionsPrefix(mg[0]->smoothd,"mg_coarse_");CHKERRQ(ierr); 103 104 /* coarse solve is (redundant) LU by default */ 105 ierr = KSPSetType(mg[0]->smoothd,KSPPREONLY);CHKERRQ(ierr); 106 ierr = KSPGetPC(mg[0]->smoothd,&ipc);CHKERRQ(ierr); 107 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 108 if (size > 1) { 109 ierr = PCSetType(ipc,PCREDUNDANT);CHKERRQ(ierr); 110 } else { 111 ierr = PCSetType(ipc,PCLU);CHKERRQ(ierr); 112 } 113 114 } else { 115 char tprefix[128]; 116 sprintf(tprefix,"mg_levels_%d_",(int)i); 117 ierr = KSPAppendOptionsPrefix(mg[i]->smoothd,tprefix);CHKERRQ(ierr); 118 } 119 ierr = PetscLogObjectParent(pc,mg[i]->smoothd);CHKERRQ(ierr); 120 mg[i]->smoothu = mg[i]->smoothd; 121 mg[i]->rtol = 0.0; 122 mg[i]->abstol = 0.0; 123 mg[i]->dtol = 0.0; 124 mg[i]->ttol = 0.0; 125 mg[i]->eventsmoothsetup = 0; 126 mg[i]->eventsmoothsolve = 0; 127 mg[i]->eventresidual = 0; 128 mg[i]->eventinterprestrict = 0; 129 mg[i]->cyclesperpcapply = 1; 130 } 131 *result = mg; 132 PetscFunctionReturn(0); 133 } 134 135 #undef __FUNCT__ 136 #define __FUNCT__ "PCDestroy_MG" 137 static PetscErrorCode PCDestroy_MG(PC pc) 138 { 139 PC_MG **mg = (PC_MG**)pc->data; 140 PetscErrorCode ierr; 141 PetscInt i,n; 142 143 PetscFunctionBegin; 144 if (!mg) PetscFunctionReturn(0); 145 n = mg[0]->levels; 146 for (i=0; i<n-1; i++) { 147 if (mg[i+1]->r) {ierr = VecDestroy(mg[i+1]->r);CHKERRQ(ierr);} 148 if (mg[i]->b) {ierr = VecDestroy(mg[i]->b);CHKERRQ(ierr);} 149 if (mg[i]->x) {ierr = VecDestroy(mg[i]->x);CHKERRQ(ierr);} 150 if (mg[i+1]->restrct) {ierr = MatDestroy(mg[i+1]->restrct);CHKERRQ(ierr);} 151 if (mg[i+1]->interpolate) {ierr = MatDestroy(mg[i+1]->interpolate);CHKERRQ(ierr);} 152 } 153 154 for (i=0; i<n; i++) { 155 if (mg[i]->smoothd != mg[i]->smoothu) { 156 ierr = KSPDestroy(mg[i]->smoothd);CHKERRQ(ierr); 157 } 158 ierr = KSPDestroy(mg[i]->smoothu);CHKERRQ(ierr); 159 ierr = PetscFree(mg[i]);CHKERRQ(ierr); 160 } 161 ierr = PetscFree(mg);CHKERRQ(ierr); 162 PetscFunctionReturn(0); 163 } 164 165 166 167 EXTERN PetscErrorCode PCMGACycle_Private(PC_MG**); 168 EXTERN PetscErrorCode PCMGFCycle_Private(PC,PC_MG**); 169 EXTERN PetscErrorCode PCMGKCycle_Private(PC_MG**); 170 171 /* 172 PCApply_MG - Runs either an additive, multiplicative, Kaskadic 173 or full cycle of multigrid. 174 175 Note: 176 A simple wrapper which calls PCMGMCycle(),PCMGACycle(), or PCMGFCycle(). 177 */ 178 #undef __FUNCT__ 179 #define __FUNCT__ "PCApply_MG" 180 static PetscErrorCode PCApply_MG(PC pc,Vec b,Vec x) 181 { 182 PC_MG **mg = (PC_MG**)pc->data; 183 PetscErrorCode ierr; 184 PetscInt levels = mg[0]->levels,i; 185 186 PetscFunctionBegin; 187 mg[levels-1]->b = b; 188 mg[levels-1]->x = x; 189 if (!mg[levels-1]->r && mg[0]->am != PC_MG_ADDITIVE && levels > 1) { 190 Vec tvec; 191 ierr = VecDuplicate(mg[levels-1]->b,&tvec);CHKERRQ(ierr); 192 ierr = PCMGSetR(pc,levels-1,tvec);CHKERRQ(ierr); 193 ierr = VecDestroy(tvec);CHKERRQ(ierr); 194 } 195 if (mg[0]->am == PC_MG_MULTIPLICATIVE) { 196 ierr = VecSet(x,0.0);CHKERRQ(ierr); 197 for (i=0; i<mg[0]->cyclesperpcapply; i++) { 198 ierr = PCMGMCycle_Private(pc,mg+levels-1,PETSC_NULL);CHKERRQ(ierr); 199 } 200 } 201 else if (mg[0]->am == PC_MG_ADDITIVE) { 202 ierr = PCMGACycle_Private(mg);CHKERRQ(ierr); 203 } 204 else if (mg[0]->am == PC_MG_KASKADE) { 205 ierr = PCMGKCycle_Private(mg);CHKERRQ(ierr); 206 } 207 else { 208 ierr = PCMGFCycle_Private(pc,mg);CHKERRQ(ierr); 209 } 210 PetscFunctionReturn(0); 211 } 212 213 #undef __FUNCT__ 214 #define __FUNCT__ "PCApplyRichardson_MG" 215 static PetscErrorCode PCApplyRichardson_MG(PC pc,Vec b,Vec x,Vec w,PetscReal rtol,PetscReal abstol, PetscReal dtol,PetscInt its,PetscInt *outits,PCRichardsonConvergedReason *reason) 216 { 217 PC_MG **mg = (PC_MG**)pc->data; 218 PetscErrorCode ierr; 219 PetscInt levels = mg[0]->levels,i; 220 221 PetscFunctionBegin; 222 mg[levels-1]->b = b; 223 mg[levels-1]->x = x; 224 225 mg[levels-1]->rtol = rtol; 226 mg[levels-1]->abstol = abstol; 227 mg[levels-1]->dtol = dtol; 228 if (rtol) { 229 /* compute initial residual norm for relative convergence test */ 230 PetscReal rnorm; 231 ierr = (*mg[levels-1]->residual)(mg[levels-1]->A,b,x,w);CHKERRQ(ierr); 232 ierr = VecNorm(w,NORM_2,&rnorm);CHKERRQ(ierr); 233 mg[levels-1]->ttol = PetscMax(rtol*rnorm,abstol); 234 } else if (abstol) { 235 mg[levels-1]->ttol = abstol; 236 } else { 237 mg[levels-1]->ttol = 0.0; 238 } 239 240 /* since smoother is applied to full system, not just residual we need to make sure that smoothers don't 241 stop prematurely do to small residual */ 242 for (i=1; i<levels; i++) { 243 ierr = KSPSetTolerances(mg[i]->smoothu,0,PETSC_DEFAULT,PETSC_DEFAULT,PETSC_DEFAULT);CHKERRQ(ierr); 244 if (mg[i]->smoothu != mg[i]->smoothd) { 245 ierr = KSPSetTolerances(mg[i]->smoothd,0,PETSC_DEFAULT,PETSC_DEFAULT,PETSC_DEFAULT);CHKERRQ(ierr); 246 } 247 } 248 249 *reason = (PCRichardsonConvergedReason)0; 250 for (i=0; i<its; i++) { 251 ierr = PCMGMCycle_Private(pc,mg+levels-1,reason);CHKERRQ(ierr); 252 if (*reason) break; 253 } 254 if (!*reason) *reason = PCRICHARDSON_CONVERGED_ITS; 255 *outits = i; 256 PetscFunctionReturn(0); 257 } 258 259 #undef __FUNCT__ 260 #define __FUNCT__ "PCSetFromOptions_MG" 261 PetscErrorCode PCSetFromOptions_MG(PC pc) 262 { 263 PetscErrorCode ierr; 264 PetscInt m,levels = 1,cycles; 265 PetscTruth flg; 266 PC_MG **mg = (PC_MG**)pc->data; 267 PCMGType mgtype = PC_MG_ADDITIVE; 268 PCMGCycleType mgctype; 269 270 PetscFunctionBegin; 271 ierr = PetscOptionsHead("Multigrid options");CHKERRQ(ierr); 272 if (!pc->data) { 273 ierr = PetscOptionsInt("-pc_mg_levels","Number of Levels","PCMGSetLevels",levels,&levels,&flg);CHKERRQ(ierr); 274 ierr = PCMGSetLevels(pc,levels,PETSC_NULL);CHKERRQ(ierr); 275 mg = (PC_MG**)pc->data; 276 } 277 mgctype = (PCMGCycleType) mg[0]->cycles; 278 ierr = PetscOptionsEnum("-pc_mg_cycle_type","V cycle or for W-cycle","PCMGSetCycleType",PCMGCycleTypes,(PetscEnum)mgctype,(PetscEnum*)&mgctype,&flg);CHKERRQ(ierr); 279 if (flg) { 280 ierr = PCMGSetCycleType(pc,mgctype);CHKERRQ(ierr); 281 }; 282 ierr = PetscOptionsName("-pc_mg_galerkin","Use Galerkin process to compute coarser operators","PCMGSetGalerkin",&flg);CHKERRQ(ierr); 283 if (flg) { 284 ierr = PCMGSetGalerkin(pc);CHKERRQ(ierr); 285 } 286 ierr = PetscOptionsInt("-pc_mg_smoothup","Number of post-smoothing steps","PCMGSetNumberSmoothUp",1,&m,&flg);CHKERRQ(ierr); 287 if (flg) { 288 ierr = PCMGSetNumberSmoothUp(pc,m);CHKERRQ(ierr); 289 } 290 ierr = PetscOptionsInt("-pc_mg_smoothdown","Number of pre-smoothing steps","PCMGSetNumberSmoothDown",1,&m,&flg);CHKERRQ(ierr); 291 if (flg) { 292 ierr = PCMGSetNumberSmoothDown(pc,m);CHKERRQ(ierr); 293 } 294 ierr = PetscOptionsEnum("-pc_mg_type","Multigrid type","PCMGSetType",PCMGTypes,(PetscEnum)mgtype,(PetscEnum*)&mgtype,&flg);CHKERRQ(ierr); 295 if (flg) { 296 ierr = PCMGSetType(pc,mgtype);CHKERRQ(ierr); 297 } 298 if (mg[0]->am == PC_MG_MULTIPLICATIVE) { 299 ierr = PetscOptionsInt("-pc_mg_multiplicative_cycles","Number of cycles for each preconditioner step","PCMGSetLevels",mg[0]->cyclesperpcapply,&cycles,&flg);CHKERRQ(ierr); 300 if (flg) { 301 ierr = PCMGMultiplicativeSetCycles(pc,cycles);CHKERRQ(ierr); 302 } 303 } 304 ierr = PetscOptionsName("-pc_mg_log","Log times for each multigrid level","None",&flg);CHKERRQ(ierr); 305 if (flg) { 306 PetscInt i; 307 char eventname[128]; 308 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 309 levels = mg[0]->levels; 310 for (i=0; i<levels; i++) { 311 sprintf(eventname,"MGSetup Level %d",(int)i); 312 ierr = PetscLogEventRegister(eventname,((PetscObject)pc)->cookie,&mg[i]->eventsmoothsetup);CHKERRQ(ierr); 313 sprintf(eventname,"MGSmooth Level %d",(int)i); 314 ierr = PetscLogEventRegister(eventname,((PetscObject)pc)->cookie,&mg[i]->eventsmoothsolve);CHKERRQ(ierr); 315 if (i) { 316 sprintf(eventname,"MGResid Level %d",(int)i); 317 ierr = PetscLogEventRegister(eventname,((PetscObject)pc)->cookie,&mg[i]->eventresidual);CHKERRQ(ierr); 318 sprintf(eventname,"MGInterp Level %d",(int)i); 319 ierr = PetscLogEventRegister(eventname,((PetscObject)pc)->cookie,&mg[i]->eventinterprestrict);CHKERRQ(ierr); 320 } 321 } 322 } 323 ierr = PetscOptionsTail();CHKERRQ(ierr); 324 PetscFunctionReturn(0); 325 } 326 327 const char *PCMGTypes[] = {"MULTIPLICATIVE","ADDITIVE","FULL","KASKADE","PCMGType","PC_MG",0}; 328 const char *PCMGCycleTypes[] = {"invalid","v","w","PCMGCycleType","PC_MG_CYCLE",0}; 329 330 #undef __FUNCT__ 331 #define __FUNCT__ "PCView_MG" 332 static PetscErrorCode PCView_MG(PC pc,PetscViewer viewer) 333 { 334 PC_MG **mg = (PC_MG**)pc->data; 335 PetscErrorCode ierr; 336 PetscInt levels = mg[0]->levels,i; 337 PetscTruth iascii; 338 339 PetscFunctionBegin; 340 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); 341 if (iascii) { 342 ierr = PetscViewerASCIIPrintf(viewer," MG: type is %s, levels=%D cycles=%s\n", PCMGTypes[mg[0]->am],levels,(mg[0]->cycles == PC_MG_CYCLE_V) ? "v" : "w");CHKERRQ(ierr); 343 if (mg[0]->am == PC_MG_MULTIPLICATIVE) { 344 ierr = PetscViewerASCIIPrintf(viewer," Cycles per PCApply=%d\n",mg[0]->cyclesperpcapply);CHKERRQ(ierr); 345 } 346 if (mg[0]->galerkin) { 347 ierr = PetscViewerASCIIPrintf(viewer," Using Galerkin computed coarse grid matrices\n");CHKERRQ(ierr); 348 } 349 for (i=0; i<levels; i++) { 350 if (!i) { 351 ierr = PetscViewerASCIIPrintf(viewer,"Coarse gride solver -- level %D presmooths=%D postsmooths=%D -----\n",i,mg[0]->default_smoothd,mg[0]->default_smoothu);CHKERRQ(ierr); 352 } else { 353 ierr = PetscViewerASCIIPrintf(viewer,"Down solver (pre-smoother) on level %D smooths=%D --------------------\n",i,mg[i]->default_smoothd);CHKERRQ(ierr); 354 } 355 ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr); 356 ierr = KSPView(mg[i]->smoothd,viewer);CHKERRQ(ierr); 357 ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr); 358 if (i && mg[i]->smoothd == mg[i]->smoothu) { 359 ierr = PetscViewerASCIIPrintf(viewer,"Up solver (post-smoother) same as down solver (pre-smoother)\n");CHKERRQ(ierr); 360 } else if (i){ 361 ierr = PetscViewerASCIIPrintf(viewer,"Up solver (post-smoother) on level %D smooths=%D --------------------\n",i,mg[i]->default_smoothu);CHKERRQ(ierr); 362 ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr); 363 ierr = KSPView(mg[i]->smoothu,viewer);CHKERRQ(ierr); 364 ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr); 365 } 366 } 367 } else { 368 SETERRQ1(PETSC_ERR_SUP,"Viewer type %s not supported for PCMG",((PetscObject)viewer)->type_name); 369 } 370 PetscFunctionReturn(0); 371 } 372 373 /* 374 Calls setup for the KSP on each level 375 */ 376 #undef __FUNCT__ 377 #define __FUNCT__ "PCSetUp_MG" 378 static PetscErrorCode PCSetUp_MG(PC pc) 379 { 380 PC_MG **mg = (PC_MG**)pc->data; 381 PetscErrorCode ierr; 382 PetscInt i,n = mg[0]->levels; 383 PC cpc,mpc; 384 PetscTruth preonly,lu,redundant,cholesky,monitor = PETSC_FALSE,dump,opsset; 385 PetscViewerASCIIMonitor ascii; 386 PetscViewer viewer = PETSC_NULL; 387 MPI_Comm comm; 388 Mat dA,dB; 389 MatStructure uflag; 390 Vec tvec; 391 392 PetscFunctionBegin; 393 394 /* If user did not provide fine grid operators OR operator was not updated since last global KSPSetOperators() */ 395 /* so use those from global PC */ 396 /* Is this what we always want? What if user wants to keep old one? */ 397 ierr = KSPGetOperatorsSet(mg[n-1]->smoothd,PETSC_NULL,&opsset);CHKERRQ(ierr); 398 ierr = KSPGetPC(mg[0]->smoothd,&cpc);CHKERRQ(ierr); 399 ierr = KSPGetPC(mg[n-1]->smoothd,&mpc);CHKERRQ(ierr); 400 if (!opsset || ((cpc->setupcalled == 1) && (mpc->setupcalled == 2))) { 401 ierr = PetscInfo(pc,"Using outer operators to define finest grid operator \n because PCMGGetSmoother(pc,nlevels-1,&ksp);KSPSetOperators(ksp,...); was not called.\n");CHKERRQ(ierr); 402 ierr = KSPSetOperators(mg[n-1]->smoothd,pc->mat,pc->pmat,pc->flag);CHKERRQ(ierr); 403 } 404 405 if (mg[0]->galerkin) { 406 Mat B; 407 mg[0]->galerkinused = PETSC_TRUE; 408 /* currently only handle case where mat and pmat are the same on coarser levels */ 409 ierr = KSPGetOperators(mg[n-1]->smoothd,&dA,&dB,&uflag);CHKERRQ(ierr); 410 if (!pc->setupcalled) { 411 for (i=n-2; i>-1; i--) { 412 ierr = MatPtAP(dB,mg[i+1]->interpolate,MAT_INITIAL_MATRIX,1.0,&B);CHKERRQ(ierr); 413 ierr = KSPSetOperators(mg[i]->smoothd,B,B,uflag);CHKERRQ(ierr); 414 if (i != n-2) {ierr = PetscObjectDereference((PetscObject)dB);CHKERRQ(ierr);} 415 dB = B; 416 } 417 ierr = PetscObjectDereference((PetscObject)dB);CHKERRQ(ierr); 418 } else { 419 for (i=n-2; i>-1; i--) { 420 ierr = KSPGetOperators(mg[i]->smoothd,PETSC_NULL,&B,PETSC_NULL);CHKERRQ(ierr); 421 ierr = MatPtAP(dB,mg[i+1]->interpolate,MAT_REUSE_MATRIX,1.0,&B);CHKERRQ(ierr); 422 ierr = KSPSetOperators(mg[i]->smoothd,B,B,uflag);CHKERRQ(ierr); 423 dB = B; 424 } 425 } 426 } 427 428 if (!pc->setupcalled) { 429 ierr = PetscOptionsHasName(0,"-pc_mg_monitor",&monitor);CHKERRQ(ierr); 430 431 for (i=0; i<n; i++) { 432 if (monitor) { 433 ierr = PetscObjectGetComm((PetscObject)mg[i]->smoothd,&comm);CHKERRQ(ierr); 434 ierr = PetscViewerASCIIMonitorCreate(comm,"stdout",n-i,&ascii);CHKERRQ(ierr); 435 ierr = KSPMonitorSet(mg[i]->smoothd,KSPMonitorDefault,ascii,(PetscErrorCode(*)(void*))PetscViewerASCIIMonitorDestroy);CHKERRQ(ierr); 436 } 437 ierr = KSPSetFromOptions(mg[i]->smoothd);CHKERRQ(ierr); 438 } 439 for (i=1; i<n; i++) { 440 if (mg[i]->smoothu && (mg[i]->smoothu != mg[i]->smoothd)) { 441 if (monitor) { 442 ierr = PetscObjectGetComm((PetscObject)mg[i]->smoothu,&comm);CHKERRQ(ierr); 443 ierr = PetscViewerASCIIMonitorCreate(comm,"stdout",n-i,&ascii);CHKERRQ(ierr); 444 ierr = KSPMonitorSet(mg[i]->smoothu,KSPMonitorDefault,ascii,(PetscErrorCode(*)(void*))PetscViewerASCIIMonitorDestroy);CHKERRQ(ierr); 445 } 446 ierr = KSPSetFromOptions(mg[i]->smoothu);CHKERRQ(ierr); 447 } 448 } 449 for (i=1; i<n; i++) { 450 if (!mg[i]->residual) { 451 Mat mat; 452 ierr = KSPGetOperators(mg[i]->smoothd,PETSC_NULL,&mat,PETSC_NULL);CHKERRQ(ierr); 453 ierr = PCMGSetResidual(pc,i,PCMGDefaultResidual,mat);CHKERRQ(ierr); 454 } 455 if (mg[i]->restrct && !mg[i]->interpolate) { 456 ierr = PCMGSetInterpolation(pc,i,mg[i]->restrct);CHKERRQ(ierr); 457 } 458 if (!mg[i]->restrct && mg[i]->interpolate) { 459 ierr = PCMGSetRestriction(pc,i,mg[i]->interpolate);CHKERRQ(ierr); 460 } 461 #if defined(PETSC_USE_DEBUG) 462 if (!mg[i]->restrct || !mg[i]->interpolate) { 463 SETERRQ1(PETSC_ERR_ARG_WRONGSTATE,"Need to set restriction or interpolation on level %d",(int)i); 464 } 465 #endif 466 } 467 for (i=0; i<n-1; i++) { 468 if (!mg[i]->b) { 469 Vec *vec; 470 ierr = KSPGetVecs(mg[i]->smoothd,1,&vec,0,PETSC_NULL);CHKERRQ(ierr); 471 ierr = PCMGSetRhs(pc,i,*vec);CHKERRQ(ierr); 472 ierr = VecDestroy(*vec);CHKERRQ(ierr); 473 ierr = PetscFree(vec);CHKERRQ(ierr); 474 } 475 if (!mg[i]->r && i) { 476 ierr = VecDuplicate(mg[i]->b,&tvec);CHKERRQ(ierr); 477 ierr = PCMGSetR(pc,i,tvec);CHKERRQ(ierr); 478 ierr = VecDestroy(tvec);CHKERRQ(ierr); 479 } 480 if (!mg[i]->x) { 481 ierr = VecDuplicate(mg[i]->b,&tvec);CHKERRQ(ierr); 482 ierr = PCMGSetX(pc,i,tvec);CHKERRQ(ierr); 483 ierr = VecDestroy(tvec);CHKERRQ(ierr); 484 } 485 } 486 if (n != 1 && !mg[n-1]->r) { 487 /* PCMGSetR() on the finest level if user did not supply it */ 488 Vec *vec; 489 ierr = KSPGetVecs(mg[n-1]->smoothd,1,&vec,0,PETSC_NULL);CHKERRQ(ierr); 490 ierr = PCMGSetR(pc,n-1,*vec);CHKERRQ(ierr); 491 ierr = VecDestroy(*vec);CHKERRQ(ierr); 492 ierr = PetscFree(vec);CHKERRQ(ierr); 493 } 494 } 495 496 497 for (i=1; i<n; i++) { 498 if (mg[i]->smoothu == mg[i]->smoothd) { 499 /* if doing only down then initial guess is zero */ 500 ierr = KSPSetInitialGuessNonzero(mg[i]->smoothd,PETSC_TRUE);CHKERRQ(ierr); 501 } 502 if (mg[i]->eventsmoothsetup) {ierr = PetscLogEventBegin(mg[i]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 503 ierr = KSPSetUp(mg[i]->smoothd);CHKERRQ(ierr); 504 if (mg[i]->eventsmoothsetup) {ierr = PetscLogEventEnd(mg[i]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 505 } 506 for (i=1; i<n; i++) { 507 if (mg[i]->smoothu && mg[i]->smoothu != mg[i]->smoothd) { 508 Mat downmat,downpmat; 509 MatStructure matflag; 510 PetscTruth opsset; 511 512 /* check if operators have been set for up, if not use down operators to set them */ 513 ierr = KSPGetOperatorsSet(mg[i]->smoothu,&opsset,PETSC_NULL);CHKERRQ(ierr); 514 if (!opsset) { 515 ierr = KSPGetOperators(mg[i]->smoothd,&downmat,&downpmat,&matflag);CHKERRQ(ierr); 516 ierr = KSPSetOperators(mg[i]->smoothu,downmat,downpmat,matflag);CHKERRQ(ierr); 517 } 518 519 ierr = KSPSetInitialGuessNonzero(mg[i]->smoothu,PETSC_TRUE);CHKERRQ(ierr); 520 if (mg[i]->eventsmoothsetup) {ierr = PetscLogEventBegin(mg[i]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 521 ierr = KSPSetUp(mg[i]->smoothu);CHKERRQ(ierr); 522 if (mg[i]->eventsmoothsetup) {ierr = PetscLogEventEnd(mg[i]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 523 } 524 } 525 526 /* 527 If coarse solver is not direct method then DO NOT USE preonly 528 */ 529 ierr = PetscTypeCompare((PetscObject)mg[0]->smoothd,KSPPREONLY,&preonly);CHKERRQ(ierr); 530 if (preonly) { 531 ierr = PetscTypeCompare((PetscObject)cpc,PCLU,&lu);CHKERRQ(ierr); 532 ierr = PetscTypeCompare((PetscObject)cpc,PCREDUNDANT,&redundant);CHKERRQ(ierr); 533 ierr = PetscTypeCompare((PetscObject)cpc,PCCHOLESKY,&cholesky);CHKERRQ(ierr); 534 if (!lu && !redundant && !cholesky) { 535 ierr = KSPSetType(mg[0]->smoothd,KSPGMRES);CHKERRQ(ierr); 536 } 537 } 538 539 if (!pc->setupcalled) { 540 if (monitor) { 541 ierr = PetscObjectGetComm((PetscObject)mg[0]->smoothd,&comm);CHKERRQ(ierr); 542 ierr = PetscViewerASCIIMonitorCreate(comm,"stdout",n,&ascii);CHKERRQ(ierr); 543 ierr = KSPMonitorSet(mg[0]->smoothd,KSPMonitorDefault,ascii,(PetscErrorCode(*)(void*))PetscViewerASCIIMonitorDestroy);CHKERRQ(ierr); 544 } 545 ierr = KSPSetFromOptions(mg[0]->smoothd);CHKERRQ(ierr); 546 } 547 548 if (mg[0]->eventsmoothsetup) {ierr = PetscLogEventBegin(mg[0]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 549 ierr = KSPSetUp(mg[0]->smoothd);CHKERRQ(ierr); 550 if (mg[0]->eventsmoothsetup) {ierr = PetscLogEventEnd(mg[0]->eventsmoothsetup,0,0,0,0);CHKERRQ(ierr);} 551 552 /* 553 Dump the interpolation/restriction matrices plus the 554 Jacobian/stiffness on each level. This allows Matlab users to 555 easily check if the Galerkin condition A_c = R A_f R^T is satisfied. 556 557 Only support one or the other at the same time. 558 */ 559 #if defined(PETSC_USE_SOCKET_VIEWER) 560 ierr = PetscOptionsHasName(((PetscObject)pc)->prefix,"-pc_mg_dump_matlab",&dump);CHKERRQ(ierr); 561 if (dump) { 562 viewer = PETSC_VIEWER_SOCKET_(((PetscObject)pc)->comm); 563 } 564 #endif 565 ierr = PetscOptionsHasName(((PetscObject)pc)->prefix,"-pc_mg_dump_binary",&dump);CHKERRQ(ierr); 566 if (dump) { 567 viewer = PETSC_VIEWER_BINARY_(((PetscObject)pc)->comm); 568 } 569 570 if (viewer) { 571 for (i=1; i<n; i++) { 572 ierr = MatView(mg[i]->restrct,viewer);CHKERRQ(ierr); 573 } 574 for (i=0; i<n; i++) { 575 ierr = KSPGetPC(mg[i]->smoothd,&pc);CHKERRQ(ierr); 576 ierr = MatView(pc->mat,viewer);CHKERRQ(ierr); 577 } 578 } 579 PetscFunctionReturn(0); 580 } 581 582 /* -------------------------------------------------------------------------------------*/ 583 584 #undef __FUNCT__ 585 #define __FUNCT__ "PCMGSetLevels" 586 /*@C 587 PCMGSetLevels - Sets the number of levels to use with MG. 588 Must be called before any other MG routine. 589 590 Collective on PC 591 592 Input Parameters: 593 + pc - the preconditioner context 594 . levels - the number of levels 595 - comms - optional communicators for each level; this is to allow solving the coarser problems 596 on smaller sets of processors. Use PETSC_NULL_OBJECT for default in Fortran 597 598 Level: intermediate 599 600 Notes: 601 If the number of levels is one then the multigrid uses the -mg_levels prefix 602 for setting the level options rather than the -mg_coarse prefix. 603 604 .keywords: MG, set, levels, multigrid 605 606 .seealso: PCMGSetType(), PCMGGetLevels() 607 @*/ 608 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetLevels(PC pc,PetscInt levels,MPI_Comm *comms) 609 { 610 PetscErrorCode ierr; 611 PC_MG **mg=0; 612 613 PetscFunctionBegin; 614 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 615 616 if (pc->data) { 617 SETERRQ(PETSC_ERR_ORDER,"Number levels already set for MG\n\ 618 make sure that you call PCMGSetLevels() before KSPSetFromOptions()"); 619 } 620 ierr = PCMGCreate_Private(((PetscObject)pc)->comm,levels,pc,comms,&mg);CHKERRQ(ierr); 621 mg[0]->am = PC_MG_MULTIPLICATIVE; 622 pc->data = (void*)mg; 623 pc->ops->applyrichardson = PCApplyRichardson_MG; 624 PetscFunctionReturn(0); 625 } 626 627 #undef __FUNCT__ 628 #define __FUNCT__ "PCMGGetLevels" 629 /*@ 630 PCMGGetLevels - Gets the number of levels to use with MG. 631 632 Not Collective 633 634 Input Parameter: 635 . pc - the preconditioner context 636 637 Output parameter: 638 . levels - the number of levels 639 640 Level: advanced 641 642 .keywords: MG, get, levels, multigrid 643 644 .seealso: PCMGSetLevels() 645 @*/ 646 PetscErrorCode PETSCKSP_DLLEXPORT PCMGGetLevels(PC pc,PetscInt *levels) 647 { 648 PC_MG **mg; 649 650 PetscFunctionBegin; 651 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 652 PetscValidIntPointer(levels,2); 653 654 mg = (PC_MG**)pc->data; 655 *levels = mg[0]->levels; 656 PetscFunctionReturn(0); 657 } 658 659 #undef __FUNCT__ 660 #define __FUNCT__ "PCMGSetType" 661 /*@ 662 PCMGSetType - Determines the form of multigrid to use: 663 multiplicative, additive, full, or the Kaskade algorithm. 664 665 Collective on PC 666 667 Input Parameters: 668 + pc - the preconditioner context 669 - form - multigrid form, one of PC_MG_MULTIPLICATIVE, PC_MG_ADDITIVE, 670 PC_MG_FULL, PC_MG_KASKADE 671 672 Options Database Key: 673 . -pc_mg_type <form> - Sets <form>, one of multiplicative, 674 additive, full, kaskade 675 676 Level: advanced 677 678 .keywords: MG, set, method, multiplicative, additive, full, Kaskade, multigrid 679 680 .seealso: PCMGSetLevels() 681 @*/ 682 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetType(PC pc,PCMGType form) 683 { 684 PC_MG **mg; 685 686 PetscFunctionBegin; 687 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 688 mg = (PC_MG**)pc->data; 689 690 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 691 mg[0]->am = form; 692 if (form == PC_MG_MULTIPLICATIVE) pc->ops->applyrichardson = PCApplyRichardson_MG; 693 else pc->ops->applyrichardson = 0; 694 PetscFunctionReturn(0); 695 } 696 697 #undef __FUNCT__ 698 #define __FUNCT__ "PCMGSetCycleType" 699 /*@ 700 PCMGSetCycleType - Sets the type cycles to use. Use PCMGSetCycleTypeOnLevel() for more 701 complicated cycling. 702 703 Collective on PC 704 705 Input Parameters: 706 + pc - the multigrid context 707 - PC_MG_CYCLE_V or PC_MG_CYCLE_W 708 709 Options Database Key: 710 $ -pc_mg_cycle_type v or w 711 712 Level: advanced 713 714 .keywords: MG, set, cycles, V-cycle, W-cycle, multigrid 715 716 .seealso: PCMGSetCycleTypeOnLevel() 717 @*/ 718 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetCycleType(PC pc,PCMGCycleType n) 719 { 720 PC_MG **mg; 721 PetscInt i,levels; 722 723 PetscFunctionBegin; 724 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 725 mg = (PC_MG**)pc->data; 726 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 727 levels = mg[0]->levels; 728 729 for (i=0; i<levels; i++) { 730 mg[i]->cycles = n; 731 } 732 PetscFunctionReturn(0); 733 } 734 735 #undef __FUNCT__ 736 #define __FUNCT__ "PCMGMultiplicativeSetCycles" 737 /*@ 738 PCMGMultiplicativeSetCycles - Sets the number of cycles to use for each preconditioner step 739 of multigrid when PCMGType of PC_MG_MULTIPLICATIVE is used 740 741 Collective on PC 742 743 Input Parameters: 744 + pc - the multigrid context 745 - n - number of cycles (default is 1) 746 747 Options Database Key: 748 $ -pc_mg_multiplicative_cycles n 749 750 Level: advanced 751 752 Notes: This is not associated with setting a v or w cycle, that is set with PCMGSetCycleType() 753 754 .keywords: MG, set, cycles, V-cycle, W-cycle, multigrid 755 756 .seealso: PCMGSetCycleTypeOnLevel(), PCMGSetCycleType() 757 @*/ 758 PetscErrorCode PETSCKSP_DLLEXPORT PCMGMultiplicativeSetCycles(PC pc,PetscInt n) 759 { 760 PC_MG **mg; 761 PetscInt i,levels; 762 763 PetscFunctionBegin; 764 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 765 mg = (PC_MG**)pc->data; 766 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 767 levels = mg[0]->levels; 768 769 for (i=0; i<levels; i++) { 770 mg[i]->cyclesperpcapply = n; 771 } 772 PetscFunctionReturn(0); 773 } 774 775 #undef __FUNCT__ 776 #define __FUNCT__ "PCMGSetGalerkin" 777 /*@ 778 PCMGSetGalerkin - Causes the coarser grid matrices to be computed from the 779 finest grid via the Galerkin process: A_i-1 = r_i * A_i * r_i^t 780 781 Collective on PC 782 783 Input Parameters: 784 . pc - the multigrid context 785 786 Options Database Key: 787 $ -pc_mg_galerkin 788 789 Level: intermediate 790 791 .keywords: MG, set, Galerkin 792 793 .seealso: PCMGGetGalerkin() 794 795 @*/ 796 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetGalerkin(PC pc) 797 { 798 PC_MG **mg; 799 PetscInt i,levels; 800 801 PetscFunctionBegin; 802 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 803 mg = (PC_MG**)pc->data; 804 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 805 levels = mg[0]->levels; 806 807 for (i=0; i<levels; i++) { 808 mg[i]->galerkin = PETSC_TRUE; 809 } 810 PetscFunctionReturn(0); 811 } 812 813 #undef __FUNCT__ 814 #define __FUNCT__ "PCMGGetGalerkin" 815 /*@ 816 PCMGGetGalerkin - Checks if Galerkin multigrid is being used, i.e. 817 A_i-1 = r_i * A_i * r_i^t 818 819 Not Collective 820 821 Input Parameter: 822 . pc - the multigrid context 823 824 Output Parameter: 825 . gelerkin - PETSC_TRUE or PETSC_FALSE 826 827 Options Database Key: 828 $ -pc_mg_galerkin 829 830 Level: intermediate 831 832 .keywords: MG, set, Galerkin 833 834 .seealso: PCMGSetGalerkin() 835 836 @*/ 837 PetscErrorCode PETSCKSP_DLLEXPORT PCMGGetGalerkin(PC pc,PetscTruth *galerkin) 838 { 839 PC_MG **mg; 840 841 PetscFunctionBegin; 842 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 843 mg = (PC_MG**)pc->data; 844 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 845 *galerkin = mg[0]->galerkin; 846 PetscFunctionReturn(0); 847 } 848 849 #undef __FUNCT__ 850 #define __FUNCT__ "PCMGSetNumberSmoothDown" 851 /*@ 852 PCMGSetNumberSmoothDown - Sets the number of pre-smoothing steps to 853 use on all levels. Use PCMGGetSmootherDown() to set different 854 pre-smoothing steps on different levels. 855 856 Collective on PC 857 858 Input Parameters: 859 + mg - the multigrid context 860 - n - the number of smoothing steps 861 862 Options Database Key: 863 . -pc_mg_smoothdown <n> - Sets number of pre-smoothing steps 864 865 Level: advanced 866 867 .keywords: MG, smooth, down, pre-smoothing, steps, multigrid 868 869 .seealso: PCMGSetNumberSmoothUp() 870 @*/ 871 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetNumberSmoothDown(PC pc,PetscInt n) 872 { 873 PC_MG **mg; 874 PetscErrorCode ierr; 875 PetscInt i,levels; 876 877 PetscFunctionBegin; 878 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 879 mg = (PC_MG**)pc->data; 880 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 881 levels = mg[0]->levels; 882 883 for (i=1; i<levels; i++) { 884 /* make sure smoother up and down are different */ 885 ierr = PCMGGetSmootherUp(pc,i,PETSC_NULL);CHKERRQ(ierr); 886 ierr = KSPSetTolerances(mg[i]->smoothd,PETSC_DEFAULT,PETSC_DEFAULT,PETSC_DEFAULT,n);CHKERRQ(ierr); 887 mg[i]->default_smoothd = n; 888 } 889 PetscFunctionReturn(0); 890 } 891 892 #undef __FUNCT__ 893 #define __FUNCT__ "PCMGSetNumberSmoothUp" 894 /*@ 895 PCMGSetNumberSmoothUp - Sets the number of post-smoothing steps to use 896 on all levels. Use PCMGGetSmootherUp() to set different numbers of 897 post-smoothing steps on different levels. 898 899 Collective on PC 900 901 Input Parameters: 902 + mg - the multigrid context 903 - n - the number of smoothing steps 904 905 Options Database Key: 906 . -pc_mg_smoothup <n> - Sets number of post-smoothing steps 907 908 Level: advanced 909 910 Note: this does not set a value on the coarsest grid, since we assume that 911 there is no separate smooth up on the coarsest grid. 912 913 .keywords: MG, smooth, up, post-smoothing, steps, multigrid 914 915 .seealso: PCMGSetNumberSmoothDown() 916 @*/ 917 PetscErrorCode PETSCKSP_DLLEXPORT PCMGSetNumberSmoothUp(PC pc,PetscInt n) 918 { 919 PC_MG **mg; 920 PetscErrorCode ierr; 921 PetscInt i,levels; 922 923 PetscFunctionBegin; 924 PetscValidHeaderSpecific(pc,PC_COOKIE,1); 925 mg = (PC_MG**)pc->data; 926 if (!mg) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must set MG levels before calling"); 927 levels = mg[0]->levels; 928 929 for (i=1; i<levels; i++) { 930 /* make sure smoother up and down are different */ 931 ierr = PCMGGetSmootherUp(pc,i,PETSC_NULL);CHKERRQ(ierr); 932 ierr = KSPSetTolerances(mg[i]->smoothu,PETSC_DEFAULT,PETSC_DEFAULT,PETSC_DEFAULT,n);CHKERRQ(ierr); 933 mg[i]->default_smoothu = n; 934 } 935 PetscFunctionReturn(0); 936 } 937 938 /* ----------------------------------------------------------------------------------------*/ 939 940 /*MC 941 PCMG - Use multigrid preconditioning. This preconditioner requires you provide additional 942 information about the coarser grid matrices and restriction/interpolation operators. 943 944 Options Database Keys: 945 + -pc_mg_levels <nlevels> - number of levels including finest 946 . -pc_mg_cycles v or w 947 . -pc_mg_smoothup <n> - number of smoothing steps after interpolation 948 . -pc_mg_smoothdown <n> - number of smoothing steps before applying restriction operator 949 . -pc_mg_type <additive,multiplicative,full,cascade> - multiplicative is the default 950 . -pc_mg_log - log information about time spent on each level of the solver 951 . -pc_mg_monitor - print information on the multigrid convergence 952 . -pc_mg_galerkin - use Galerkin process to compute coarser operators 953 - -pc_mg_dump_matlab - dumps the matrices for each level and the restriction/interpolation matrices 954 to the Socket viewer for reading from Matlab. 955 956 Notes: 957 958 Level: intermediate 959 960 Concepts: multigrid/multilevel 961 962 .seealso: PCCreate(), PCSetType(), PCType (for list of available types), PC, PCMGType, 963 PCMGSetLevels(), PCMGGetLevels(), PCMGSetType(), PCMGSetCycleType(), PCMGSetNumberSmoothDown(), 964 PCMGSetNumberSmoothUp(), PCMGGetCoarseSolve(), PCMGSetResidual(), PCMGSetInterpolation(), 965 PCMGSetRestriction(), PCMGGetSmoother(), PCMGGetSmootherUp(), PCMGGetSmootherDown(), 966 PCMGSetCycleTypeOnLevel(), PCMGSetRhs(), PCMGSetX(), PCMGSetR() 967 M*/ 968 969 EXTERN_C_BEGIN 970 #undef __FUNCT__ 971 #define __FUNCT__ "PCCreate_MG" 972 PetscErrorCode PETSCKSP_DLLEXPORT PCCreate_MG(PC pc) 973 { 974 PetscFunctionBegin; 975 pc->ops->apply = PCApply_MG; 976 pc->ops->setup = PCSetUp_MG; 977 pc->ops->destroy = PCDestroy_MG; 978 pc->ops->setfromoptions = PCSetFromOptions_MG; 979 pc->ops->view = PCView_MG; 980 981 pc->data = (void*)0; 982 PetscFunctionReturn(0); 983 } 984 EXTERN_C_END 985