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