1 #define PETSCMAT_DLL 2 3 /* 4 This is where the abstract matrix operations are defined that are 5 used for finite difference computations of Jacobians using coloring. 6 */ 7 8 #include "src/mat/matimpl.h" /*I "petscmat.h" I*/ 9 10 /* Logging support */ 11 PetscCookie PETSCMAT_DLLEXPORT MAT_FDCOLORING_COOKIE = 0; 12 13 #undef __FUNCT__ 14 #define __FUNCT__ "MatFDColoringSetF" 15 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetF(MatFDColoring fd,Vec F) 16 { 17 PetscFunctionBegin; 18 fd->F = F; 19 PetscFunctionReturn(0); 20 } 21 22 #undef __FUNCT__ 23 #define __FUNCT__ "MatFDColoringView_Draw_Zoom" 24 static PetscErrorCode MatFDColoringView_Draw_Zoom(PetscDraw draw,void *Aa) 25 { 26 MatFDColoring fd = (MatFDColoring)Aa; 27 PetscErrorCode ierr; 28 PetscInt i,j; 29 PetscReal x,y; 30 31 PetscFunctionBegin; 32 33 /* loop over colors */ 34 for (i=0; i<fd->ncolors; i++) { 35 for (j=0; j<fd->nrows[i]; j++) { 36 y = fd->M - fd->rows[i][j] - fd->rstart; 37 x = fd->columnsforrow[i][j]; 38 ierr = PetscDrawRectangle(draw,x,y,x+1,y+1,i+1,i+1,i+1,i+1);CHKERRQ(ierr); 39 } 40 } 41 PetscFunctionReturn(0); 42 } 43 44 #undef __FUNCT__ 45 #define __FUNCT__ "MatFDColoringView_Draw" 46 static PetscErrorCode MatFDColoringView_Draw(MatFDColoring fd,PetscViewer viewer) 47 { 48 PetscErrorCode ierr; 49 PetscTruth isnull; 50 PetscDraw draw; 51 PetscReal xr,yr,xl,yl,h,w; 52 53 PetscFunctionBegin; 54 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 55 ierr = PetscDrawIsNull(draw,&isnull);CHKERRQ(ierr); if (isnull) PetscFunctionReturn(0); 56 57 ierr = PetscObjectCompose((PetscObject)fd,"Zoomviewer",(PetscObject)viewer);CHKERRQ(ierr); 58 59 xr = fd->N; yr = fd->M; h = yr/10.0; w = xr/10.0; 60 xr += w; yr += h; xl = -w; yl = -h; 61 ierr = PetscDrawSetCoordinates(draw,xl,yl,xr,yr);CHKERRQ(ierr); 62 ierr = PetscDrawZoom(draw,MatFDColoringView_Draw_Zoom,fd);CHKERRQ(ierr); 63 ierr = PetscObjectCompose((PetscObject)fd,"Zoomviewer",PETSC_NULL);CHKERRQ(ierr); 64 PetscFunctionReturn(0); 65 } 66 67 #undef __FUNCT__ 68 #define __FUNCT__ "MatFDColoringView" 69 /*@C 70 MatFDColoringView - Views a finite difference coloring context. 71 72 Collective on MatFDColoring 73 74 Input Parameters: 75 + c - the coloring context 76 - viewer - visualization context 77 78 Level: intermediate 79 80 Notes: 81 The available visualization contexts include 82 + PETSC_VIEWER_STDOUT_SELF - standard output (default) 83 . PETSC_VIEWER_STDOUT_WORLD - synchronized standard 84 output where only the first processor opens 85 the file. All other processors send their 86 data to the first processor to print. 87 - PETSC_VIEWER_DRAW_WORLD - graphical display of nonzero structure 88 89 Notes: 90 Since PETSc uses only a small number of basic colors (currently 33), if the coloring 91 involves more than 33 then some seemingly identical colors are displayed making it look 92 like an illegal coloring. This is just a graphical artifact. 93 94 .seealso: MatFDColoringCreate() 95 96 .keywords: Mat, finite differences, coloring, view 97 @*/ 98 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringView(MatFDColoring c,PetscViewer viewer) 99 { 100 PetscErrorCode ierr; 101 PetscInt i,j; 102 PetscTruth isdraw,iascii; 103 PetscViewerFormat format; 104 105 PetscFunctionBegin; 106 PetscValidHeaderSpecific(c,MAT_FDCOLORING_COOKIE,1); 107 if (!viewer) viewer = PETSC_VIEWER_STDOUT_(c->comm); 108 PetscValidHeaderSpecific(viewer,PETSC_VIEWER_COOKIE,2); 109 PetscCheckSameComm(c,1,viewer,2); 110 111 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_DRAW,&isdraw);CHKERRQ(ierr); 112 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); 113 if (isdraw) { 114 ierr = MatFDColoringView_Draw(c,viewer);CHKERRQ(ierr); 115 } else if (iascii) { 116 ierr = PetscViewerASCIIPrintf(viewer,"MatFDColoring Object:\n");CHKERRQ(ierr); 117 ierr = PetscViewerASCIIPrintf(viewer," Error tolerance=%G\n",c->error_rel);CHKERRQ(ierr); 118 ierr = PetscViewerASCIIPrintf(viewer," Umin=%G\n",c->umin);CHKERRQ(ierr); 119 ierr = PetscViewerASCIIPrintf(viewer," Number of colors=%D\n",c->ncolors);CHKERRQ(ierr); 120 121 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 122 if (format != PETSC_VIEWER_ASCII_INFO) { 123 for (i=0; i<c->ncolors; i++) { 124 ierr = PetscViewerASCIIPrintf(viewer," Information for color %D\n",i);CHKERRQ(ierr); 125 ierr = PetscViewerASCIIPrintf(viewer," Number of columns %D\n",c->ncolumns[i]);CHKERRQ(ierr); 126 for (j=0; j<c->ncolumns[i]; j++) { 127 ierr = PetscViewerASCIIPrintf(viewer," %D\n",c->columns[i][j]);CHKERRQ(ierr); 128 } 129 ierr = PetscViewerASCIIPrintf(viewer," Number of rows %D\n",c->nrows[i]);CHKERRQ(ierr); 130 for (j=0; j<c->nrows[i]; j++) { 131 ierr = PetscViewerASCIIPrintf(viewer," %D %D \n",c->rows[i][j],c->columnsforrow[i][j]);CHKERRQ(ierr); 132 } 133 } 134 } 135 ierr = PetscViewerFlush(viewer);CHKERRQ(ierr); 136 } else { 137 SETERRQ1(PETSC_ERR_SUP,"Viewer type %s not supported for MatFDColoring",((PetscObject)viewer)->type_name); 138 } 139 PetscFunctionReturn(0); 140 } 141 142 #undef __FUNCT__ 143 #define __FUNCT__ "MatFDColoringSetParameters" 144 /*@ 145 MatFDColoringSetParameters - Sets the parameters for the sparse approximation of 146 a Jacobian matrix using finite differences. 147 148 Collective on MatFDColoring 149 150 The Jacobian is estimated with the differencing approximation 151 .vb 152 F'(u)_{:,i} = [F(u+h*dx_{i}) - F(u)]/h where 153 h = error_rel*u[i] if abs(u[i]) > umin 154 = +/- error_rel*umin otherwise, with +/- determined by the sign of u[i] 155 dx_{i} = (0, ... 1, .... 0) 156 .ve 157 158 Input Parameters: 159 + coloring - the coloring context 160 . error_rel - relative error 161 - umin - minimum allowable u-value magnitude 162 163 Level: advanced 164 165 .keywords: Mat, finite differences, coloring, set, parameters 166 167 .seealso: MatFDColoringCreate() 168 @*/ 169 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetParameters(MatFDColoring matfd,PetscReal error,PetscReal umin) 170 { 171 PetscFunctionBegin; 172 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 173 174 if (error != PETSC_DEFAULT) matfd->error_rel = error; 175 if (umin != PETSC_DEFAULT) matfd->umin = umin; 176 PetscFunctionReturn(0); 177 } 178 179 #undef __FUNCT__ 180 #define __FUNCT__ "MatFDColoringSetFrequency" 181 /*@ 182 MatFDColoringSetFrequency - Sets the frequency for computing new Jacobian 183 matrices. 184 185 Collective on MatFDColoring 186 187 Input Parameters: 188 + coloring - the coloring context 189 - freq - frequency (default is 1) 190 191 Options Database Keys: 192 . -mat_fd_coloring_freq <freq> - Sets coloring frequency 193 194 Level: advanced 195 196 Notes: 197 Using a modified Newton strategy, where the Jacobian remains fixed for several 198 iterations, can be cost effective in terms of overall nonlinear solution 199 efficiency. This parameter indicates that a new Jacobian will be computed every 200 <freq> nonlinear iterations. 201 202 .keywords: Mat, finite differences, coloring, set, frequency 203 204 .seealso: MatFDColoringCreate(), MatFDColoringGetFrequency(), MatFDColoringSetRecompute() 205 @*/ 206 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetFrequency(MatFDColoring matfd,PetscInt freq) 207 { 208 PetscFunctionBegin; 209 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 210 211 matfd->freq = freq; 212 PetscFunctionReturn(0); 213 } 214 215 #undef __FUNCT__ 216 #define __FUNCT__ "MatFDColoringGetFrequency" 217 /*@ 218 MatFDColoringGetFrequency - Gets the frequency for computing new Jacobian 219 matrices. 220 221 Not Collective 222 223 Input Parameters: 224 . coloring - the coloring context 225 226 Output Parameters: 227 . freq - frequency (default is 1) 228 229 Options Database Keys: 230 . -mat_fd_coloring_freq <freq> - Sets coloring frequency 231 232 Level: advanced 233 234 Notes: 235 Using a modified Newton strategy, where the Jacobian remains fixed for several 236 iterations, can be cost effective in terms of overall nonlinear solution 237 efficiency. This parameter indicates that a new Jacobian will be computed every 238 <freq> nonlinear iterations. 239 240 .keywords: Mat, finite differences, coloring, get, frequency 241 242 .seealso: MatFDColoringSetFrequency() 243 @*/ 244 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringGetFrequency(MatFDColoring matfd,PetscInt *freq) 245 { 246 PetscFunctionBegin; 247 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 248 *freq = matfd->freq; 249 PetscFunctionReturn(0); 250 } 251 252 #undef __FUNCT__ 253 #define __FUNCT__ "MatFDColoringGetFunction" 254 /*@C 255 MatFDColoringGetFunction - Gets the function to use for computing the Jacobian. 256 257 Collective on MatFDColoring 258 259 Input Parameters: 260 . coloring - the coloring context 261 262 Output Parameters: 263 + f - the function 264 - fctx - the optional user-defined function context 265 266 Level: intermediate 267 268 .keywords: Mat, Jacobian, finite differences, set, function 269 @*/ 270 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringGetFunction(MatFDColoring matfd,PetscErrorCode (**f)(void),void **fctx) 271 { 272 PetscFunctionBegin; 273 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 274 if (f) *f = matfd->f; 275 if (fctx) *fctx = matfd->fctx; 276 PetscFunctionReturn(0); 277 } 278 279 #undef __FUNCT__ 280 #define __FUNCT__ "MatFDColoringSetFunction" 281 /*@C 282 MatFDColoringSetFunction - Sets the function to use for computing the Jacobian. 283 284 Collective on MatFDColoring 285 286 Input Parameters: 287 + coloring - the coloring context 288 . f - the function 289 - fctx - the optional user-defined function context 290 291 Level: intermediate 292 293 Notes: 294 In Fortran you must call MatFDColoringSetFunctionSNES() for a coloring object to 295 be used with the SNES solvers and MatFDColoringSetFunctionTS() if it is to be used 296 with the TS solvers. 297 298 .keywords: Mat, Jacobian, finite differences, set, function 299 @*/ 300 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetFunction(MatFDColoring matfd,PetscErrorCode (*f)(void),void *fctx) 301 { 302 PetscFunctionBegin; 303 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 304 matfd->f = f; 305 matfd->fctx = fctx; 306 PetscFunctionReturn(0); 307 } 308 309 #undef __FUNCT__ 310 #define __FUNCT__ "MatFDColoringSetFromOptions" 311 /*@ 312 MatFDColoringSetFromOptions - Sets coloring finite difference parameters from 313 the options database. 314 315 Collective on MatFDColoring 316 317 The Jacobian, F'(u), is estimated with the differencing approximation 318 .vb 319 F'(u)_{:,i} = [F(u+h*dx_{i}) - F(u)]/h where 320 h = error_rel*u[i] if abs(u[i]) > umin 321 = +/- error_rel*umin otherwise, with +/- determined by the sign of u[i] 322 dx_{i} = (0, ... 1, .... 0) 323 .ve 324 325 Input Parameter: 326 . coloring - the coloring context 327 328 Options Database Keys: 329 + -mat_fd_coloring_err <err> - Sets <err> (square root 330 of relative error in the function) 331 . -mat_fd_coloring_umin <umin> - Sets umin, the minimum allowable u-value magnitude 332 . -mat_fd_coloring_freq <freq> - Sets frequency of computing a new Jacobian 333 . -mat_fd_type - "wp" or "ds" (see MATSNESMF_WP or MATSNESMF_DS) 334 . -mat_fd_coloring_view - Activates basic viewing 335 . -mat_fd_coloring_view_info - Activates viewing info 336 - -mat_fd_coloring_view_draw - Activates drawing 337 338 Level: intermediate 339 340 .keywords: Mat, finite differences, parameters 341 342 .seealso: MatFDColoringCreate(), MatFDColoringView(), MatFDColoringSetParameters() 343 344 @*/ 345 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetFromOptions(MatFDColoring matfd) 346 { 347 PetscErrorCode ierr; 348 PetscTruth flg; 349 char value[3]; 350 351 PetscFunctionBegin; 352 PetscValidHeaderSpecific(matfd,MAT_FDCOLORING_COOKIE,1); 353 354 ierr = PetscOptionsBegin(matfd->comm,matfd->prefix,"Jacobian computation via finite differences option","MatFD");CHKERRQ(ierr); 355 ierr = PetscOptionsReal("-mat_fd_coloring_err","Square root of relative error in function","MatFDColoringSetParameters",matfd->error_rel,&matfd->error_rel,0);CHKERRQ(ierr); 356 ierr = PetscOptionsReal("-mat_fd_coloring_umin","Minimum allowable u magnitude","MatFDColoringSetParameters",matfd->umin,&matfd->umin,0);CHKERRQ(ierr); 357 ierr = PetscOptionsInt("-mat_fd_coloring_freq","How often Jacobian is recomputed","MatFDColoringSetFrequency",matfd->freq,&matfd->freq,0);CHKERRQ(ierr); 358 ierr = PetscOptionsString("-mat_fd_type","Algorithm to compute h, wp or ds","MatFDColoringCreate",matfd->htype,value,2,&flg);CHKERRQ(ierr); 359 if (flg) { 360 if (value[0] == 'w' && value[1] == 'p') matfd->htype = "wp"; 361 else if (value[0] == 'd' && value[1] == 's') matfd->htype = "ds"; 362 else SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Unknown finite differencing type %s",value); 363 } 364 /* not used here; but so they are presented in the GUI */ 365 ierr = PetscOptionsName("-mat_fd_coloring_view","Print entire datastructure used for Jacobian","None",0);CHKERRQ(ierr); 366 ierr = PetscOptionsName("-mat_fd_coloring_view_info","Print number of colors etc for Jacobian","None",0);CHKERRQ(ierr); 367 ierr = PetscOptionsName("-mat_fd_coloring_view_draw","Plot nonzero structure ofJacobian","None",0);CHKERRQ(ierr); 368 PetscOptionsEnd();CHKERRQ(ierr); 369 PetscFunctionReturn(0); 370 } 371 372 #undef __FUNCT__ 373 #define __FUNCT__ "MatFDColoringView_Private" 374 PetscErrorCode MatFDColoringView_Private(MatFDColoring fd) 375 { 376 PetscErrorCode ierr; 377 PetscTruth flg; 378 379 PetscFunctionBegin; 380 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_fd_coloring_view",&flg);CHKERRQ(ierr); 381 if (flg) { 382 ierr = MatFDColoringView(fd,PETSC_VIEWER_STDOUT_(fd->comm));CHKERRQ(ierr); 383 } 384 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_fd_coloring_view_info",&flg);CHKERRQ(ierr); 385 if (flg) { 386 ierr = PetscViewerPushFormat(PETSC_VIEWER_STDOUT_(fd->comm),PETSC_VIEWER_ASCII_INFO);CHKERRQ(ierr); 387 ierr = MatFDColoringView(fd,PETSC_VIEWER_STDOUT_(fd->comm));CHKERRQ(ierr); 388 ierr = PetscViewerPopFormat(PETSC_VIEWER_STDOUT_(fd->comm));CHKERRQ(ierr); 389 } 390 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_fd_coloring_view_draw",&flg);CHKERRQ(ierr); 391 if (flg) { 392 ierr = MatFDColoringView(fd,PETSC_VIEWER_DRAW_(fd->comm));CHKERRQ(ierr); 393 ierr = PetscViewerFlush(PETSC_VIEWER_DRAW_(fd->comm));CHKERRQ(ierr); 394 } 395 PetscFunctionReturn(0); 396 } 397 398 #undef __FUNCT__ 399 #define __FUNCT__ "MatFDColoringCreate" 400 /*@ 401 MatFDColoringCreate - Creates a matrix coloring context for finite difference 402 computation of Jacobians. 403 404 Collective on Mat 405 406 Input Parameters: 407 + mat - the matrix containing the nonzero structure of the Jacobian 408 - iscoloring - the coloring of the matrix 409 410 Output Parameter: 411 . color - the new coloring context 412 413 Level: intermediate 414 415 .seealso: MatFDColoringDestroy(),SNESDefaultComputeJacobianColor(), ISColoringCreate(), 416 MatFDColoringSetFunction(), MatFDColoringSetFromOptions(), MatFDColoringApply(), 417 MatFDColoringSetFrequency(), MatFDColoringSetRecompute(), MatFDColoringView(), 418 MatFDColoringSetParameters() 419 @*/ 420 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringCreate(Mat mat,ISColoring iscoloring,MatFDColoring *color) 421 { 422 MatFDColoring c; 423 MPI_Comm comm; 424 PetscErrorCode ierr; 425 PetscInt M,N; 426 427 PetscFunctionBegin; 428 ierr = PetscLogEventBegin(MAT_FDColoringCreate,mat,0,0,0);CHKERRQ(ierr); 429 ierr = MatGetSize(mat,&M,&N);CHKERRQ(ierr); 430 if (M != N) SETERRQ(PETSC_ERR_SUP,"Only for square matrices"); 431 432 ierr = PetscObjectGetComm((PetscObject)mat,&comm);CHKERRQ(ierr); 433 ierr = PetscHeaderCreate(c,_p_MatFDColoring,int,MAT_FDCOLORING_COOKIE,0,"MatFDColoring",comm,MatFDColoringDestroy,MatFDColoringView);CHKERRQ(ierr); 434 435 if (mat->ops->fdcoloringcreate) { 436 ierr = (*mat->ops->fdcoloringcreate)(mat,iscoloring,c);CHKERRQ(ierr); 437 } else { 438 SETERRQ(PETSC_ERR_SUP,"Code not yet written for this matrix type"); 439 } 440 441 c->error_rel = PETSC_SQRT_MACHINE_EPSILON; 442 c->umin = 100.0*PETSC_SQRT_MACHINE_EPSILON; 443 c->freq = 1; 444 c->usersetsrecompute = PETSC_FALSE; 445 c->recompute = PETSC_FALSE; 446 c->currentcolor = -1; 447 c->htype = "wp"; 448 449 *color = c; 450 ierr = PetscLogEventEnd(MAT_FDColoringCreate,mat,0,0,0);CHKERRQ(ierr); 451 PetscFunctionReturn(0); 452 } 453 454 #undef __FUNCT__ 455 #define __FUNCT__ "MatFDColoringDestroy" 456 /*@ 457 MatFDColoringDestroy - Destroys a matrix coloring context that was created 458 via MatFDColoringCreate(). 459 460 Collective on MatFDColoring 461 462 Input Parameter: 463 . c - coloring context 464 465 Level: intermediate 466 467 .seealso: MatFDColoringCreate() 468 @*/ 469 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringDestroy(MatFDColoring c) 470 { 471 PetscErrorCode ierr; 472 PetscInt i; 473 474 PetscFunctionBegin; 475 if (--c->refct > 0) PetscFunctionReturn(0); 476 477 for (i=0; i<c->ncolors; i++) { 478 ierr = PetscFree(c->columns[i]);CHKERRQ(ierr); 479 ierr = PetscFree(c->rows[i]);CHKERRQ(ierr); 480 ierr = PetscFree(c->columnsforrow[i]);CHKERRQ(ierr); 481 if (c->vscaleforrow) {ierr = PetscFree(c->vscaleforrow[i]);CHKERRQ(ierr);} 482 } 483 ierr = PetscFree(c->ncolumns);CHKERRQ(ierr); 484 ierr = PetscFree(c->columns);CHKERRQ(ierr); 485 ierr = PetscFree(c->nrows);CHKERRQ(ierr); 486 ierr = PetscFree(c->rows);CHKERRQ(ierr); 487 ierr = PetscFree(c->columnsforrow);CHKERRQ(ierr); 488 ierr = PetscFree(c->vscaleforrow);CHKERRQ(ierr); 489 if (c->vscale) {ierr = VecDestroy(c->vscale);CHKERRQ(ierr);} 490 if (c->w1) { 491 ierr = VecDestroy(c->w1);CHKERRQ(ierr); 492 ierr = VecDestroy(c->w2);CHKERRQ(ierr); 493 ierr = VecDestroy(c->w3);CHKERRQ(ierr); 494 } 495 ierr = PetscHeaderDestroy(c);CHKERRQ(ierr); 496 PetscFunctionReturn(0); 497 } 498 499 #undef __FUNCT__ 500 #define __FUNCT__ "MatFDColoringGetPerturbedColumns" 501 /*@C 502 MatFDColoringGetPerturbedColumns - Returns the indices of the columns that 503 that are currently being perturbed. 504 505 Not Collective 506 507 Input Parameters: 508 . coloring - coloring context created with MatFDColoringCreate() 509 510 Output Parameters: 511 + n - the number of local columns being perturbed 512 - cols - the column indices, in global numbering 513 514 Level: intermediate 515 516 .seealso: MatFDColoringCreate(), MatFDColoringDestroy(), MatFDColoringView(), MatFDColoringApply() 517 518 .keywords: coloring, Jacobian, finite differences 519 @*/ 520 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringGetPerturbedColumns(MatFDColoring coloring,PetscInt *n,PetscInt *cols[]) 521 { 522 PetscFunctionBegin; 523 if (coloring->currentcolor >= 0) { 524 *n = coloring->ncolumns[coloring->currentcolor]; 525 *cols = coloring->columns[coloring->currentcolor]; 526 } else { 527 *n = 0; 528 } 529 PetscFunctionReturn(0); 530 } 531 532 #undef __FUNCT__ 533 #define __FUNCT__ "MatFDColoringApply" 534 /*@ 535 MatFDColoringApply - Given a matrix for which a MatFDColoring context 536 has been created, computes the Jacobian for a function via finite differences. 537 538 Collective on MatFDColoring 539 540 Input Parameters: 541 + mat - location to store Jacobian 542 . coloring - coloring context created with MatFDColoringCreate() 543 . x1 - location at which Jacobian is to be computed 544 - sctx - optional context required by function (actually a SNES context) 545 546 Options Database Keys: 547 + -mat_fd_coloring_freq <freq> - Sets coloring frequency 548 . -mat_fd_type - "wp" or "ds" (see MATSNESMF_WP or MATSNESMF_DS) 549 . -mat_fd_coloring_view - Activates basic viewing or coloring 550 . -mat_fd_coloring_view_draw - Activates drawing of coloring 551 - -mat_fd_coloring_view_info - Activates viewing of coloring info 552 553 Level: intermediate 554 555 .seealso: MatFDColoringCreate(), MatFDColoringDestroy(), MatFDColoringView() 556 557 .keywords: coloring, Jacobian, finite differences 558 @*/ 559 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringApply(Mat J,MatFDColoring coloring,Vec x1,MatStructure *flag,void *sctx) 560 { 561 PetscErrorCode (*f)(void*,Vec,Vec,void*) = (PetscErrorCode (*)(void*,Vec,Vec,void *))coloring->f; 562 PetscErrorCode ierr; 563 PetscInt k,N,start,end,l,row,col,srow,**vscaleforrow,m1,m2; 564 PetscScalar dx,*y,*xx,*w3_array; 565 PetscScalar *vscale_array; 566 PetscReal epsilon = coloring->error_rel,umin = coloring->umin,unorm; 567 Vec w1,w2,w3; 568 void *fctx = coloring->fctx; 569 PetscTruth flg; 570 571 572 PetscFunctionBegin; 573 PetscValidHeaderSpecific(J,MAT_COOKIE,1); 574 PetscValidHeaderSpecific(coloring,MAT_FDCOLORING_COOKIE,2); 575 PetscValidHeaderSpecific(x1,VEC_COOKIE,3); 576 if (!f) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Must call MatFDColoringSetFunction()"); 577 578 if (coloring->usersetsrecompute) { 579 if (!coloring->recompute) { 580 *flag = SAME_PRECONDITIONER; 581 ierr = PetscInfo(J,"Skipping Jacobian, since user called MatFDColorSetRecompute()\n");CHKERRQ(ierr); 582 PetscFunctionReturn(0); 583 } else { 584 coloring->recompute = PETSC_FALSE; 585 } 586 } 587 588 ierr = PetscLogEventBegin(MAT_FDColoringApply,coloring,J,x1,0);CHKERRQ(ierr); 589 if (J->ops->fdcoloringapply) { 590 ierr = (*J->ops->fdcoloringapply)(J,coloring,x1,flag,sctx);CHKERRQ(ierr); 591 } else { 592 593 if (!coloring->w1) { 594 ierr = VecDuplicate(x1,&coloring->w1);CHKERRQ(ierr); 595 ierr = PetscLogObjectParent(coloring,coloring->w1);CHKERRQ(ierr); 596 ierr = VecDuplicate(x1,&coloring->w2);CHKERRQ(ierr); 597 ierr = PetscLogObjectParent(coloring,coloring->w2);CHKERRQ(ierr); 598 ierr = VecDuplicate(x1,&coloring->w3);CHKERRQ(ierr); 599 ierr = PetscLogObjectParent(coloring,coloring->w3);CHKERRQ(ierr); 600 } 601 w1 = coloring->w1; w2 = coloring->w2; w3 = coloring->w3; 602 603 ierr = MatSetUnfactored(J);CHKERRQ(ierr); 604 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_fd_coloring_dont_rezero",&flg);CHKERRQ(ierr); 605 if (flg) { 606 ierr = PetscInfo(coloring,"Not calling MatZeroEntries()\n");CHKERRQ(ierr); 607 } else { 608 PetscTruth assembled; 609 ierr = MatAssembled(J,&assembled);CHKERRQ(ierr); 610 if (assembled) { 611 ierr = MatZeroEntries(J);CHKERRQ(ierr); 612 } 613 } 614 615 ierr = VecGetOwnershipRange(x1,&start,&end);CHKERRQ(ierr); 616 ierr = VecGetSize(x1,&N);CHKERRQ(ierr); 617 618 /* 619 This is a horrible, horrible, hack. See DMMGComputeJacobian_Multigrid() it inproperly sets 620 coloring->F for the coarser grids from the finest 621 */ 622 if (coloring->F) { 623 ierr = VecGetLocalSize(coloring->F,&m1);CHKERRQ(ierr); 624 ierr = VecGetLocalSize(w1,&m2);CHKERRQ(ierr); 625 if (m1 != m2) { 626 coloring->F = 0; 627 } 628 } 629 630 if (coloring->F) { 631 w1 = coloring->F; /* use already computed value of function */ 632 coloring->F = 0; 633 } else { 634 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 635 ierr = (*f)(sctx,x1,w1,fctx);CHKERRQ(ierr); 636 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 637 } 638 639 if (coloring->htype[0] == 'w') { /* tacky test; need to make systematic if we add other approaches to computing h*/ 640 ierr = VecNorm(x1,NORM_2,&unorm);CHKERRQ(ierr); 641 } 642 643 /* 644 Compute all the scale factors and share with other processors 645 */ 646 ierr = VecGetArray(x1,&xx);CHKERRQ(ierr);xx = xx - start; 647 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr);vscale_array = vscale_array - start; 648 for (k=0; k<coloring->ncolors; k++) { 649 /* 650 Loop over each column associated with color adding the 651 perturbation to the vector w3. 652 */ 653 for (l=0; l<coloring->ncolumns[k]; l++) { 654 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 655 if (coloring->htype[0] == 'w') { 656 dx = 1.0 + unorm; 657 } else { 658 dx = xx[col]; 659 } 660 if (dx == 0.0) dx = 1.0; 661 #if !defined(PETSC_USE_COMPLEX) 662 if (dx < umin && dx >= 0.0) dx = umin; 663 else if (dx < 0.0 && dx > -umin) dx = -umin; 664 #else 665 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 666 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 667 #endif 668 dx *= epsilon; 669 vscale_array[col] = 1.0/dx; 670 } 671 } 672 vscale_array = vscale_array + start; 673 ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 674 ierr = VecGhostUpdateBegin(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 675 ierr = VecGhostUpdateEnd(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 676 677 /* ierr = VecView(coloring->vscale,PETSC_VIEWER_STDOUT_WORLD); 678 ierr = VecView(x1,PETSC_VIEWER_STDOUT_WORLD);*/ 679 680 if (coloring->vscaleforrow) vscaleforrow = coloring->vscaleforrow; 681 else vscaleforrow = coloring->columnsforrow; 682 683 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 684 /* 685 Loop over each color 686 */ 687 for (k=0; k<coloring->ncolors; k++) { 688 coloring->currentcolor = k; 689 ierr = VecCopy(x1,w3);CHKERRQ(ierr); 690 ierr = VecGetArray(w3,&w3_array);CHKERRQ(ierr);w3_array = w3_array - start; 691 /* 692 Loop over each column associated with color adding the 693 perturbation to the vector w3. 694 */ 695 for (l=0; l<coloring->ncolumns[k]; l++) { 696 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 697 if (coloring->htype[0] == 'w') { 698 dx = 1.0 + unorm; 699 } else { 700 dx = xx[col]; 701 } 702 if (dx == 0.0) dx = 1.0; 703 #if !defined(PETSC_USE_COMPLEX) 704 if (dx < umin && dx >= 0.0) dx = umin; 705 else if (dx < 0.0 && dx > -umin) dx = -umin; 706 #else 707 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 708 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 709 #endif 710 dx *= epsilon; 711 if (!PetscAbsScalar(dx)) SETERRQ(PETSC_ERR_PLIB,"Computed 0 differencing parameter"); 712 w3_array[col] += dx; 713 } 714 w3_array = w3_array + start; ierr = VecRestoreArray(w3,&w3_array);CHKERRQ(ierr); 715 716 /* 717 Evaluate function at x1 + dx (here dx is a vector of perturbations) 718 */ 719 720 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 721 ierr = (*f)(sctx,w3,w2,fctx);CHKERRQ(ierr); 722 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 723 ierr = VecAXPY(w2,-1.0,w1);CHKERRQ(ierr); 724 725 /* 726 Loop over rows of vector, putting results into Jacobian matrix 727 */ 728 ierr = VecGetArray(w2,&y);CHKERRQ(ierr); 729 for (l=0; l<coloring->nrows[k]; l++) { 730 row = coloring->rows[k][l]; 731 col = coloring->columnsforrow[k][l]; 732 y[row] *= vscale_array[vscaleforrow[k][l]]; 733 srow = row + start; 734 ierr = MatSetValues(J,1,&srow,1,&col,y+row,INSERT_VALUES);CHKERRQ(ierr); 735 } 736 ierr = VecRestoreArray(w2,&y);CHKERRQ(ierr); 737 } 738 coloring->currentcolor = -1; 739 ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 740 xx = xx + start; ierr = VecRestoreArray(x1,&xx);CHKERRQ(ierr); 741 ierr = MatAssemblyBegin(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 742 ierr = MatAssemblyEnd(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 743 } 744 ierr = PetscLogEventEnd(MAT_FDColoringApply,coloring,J,x1,0);CHKERRQ(ierr); 745 746 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_null_space_test",&flg);CHKERRQ(ierr); 747 if (flg) { 748 ierr = MatNullSpaceTest(J->nullsp,J);CHKERRQ(ierr); 749 } 750 ierr = MatFDColoringView_Private(coloring);CHKERRQ(ierr); 751 752 PetscFunctionReturn(0); 753 } 754 755 #undef __FUNCT__ 756 #define __FUNCT__ "MatFDColoringApplyTS" 757 /*@ 758 MatFDColoringApplyTS - Given a matrix for which a MatFDColoring context 759 has been created, computes the Jacobian for a function via finite differences. 760 761 Collective on Mat, MatFDColoring, and Vec 762 763 Input Parameters: 764 + mat - location to store Jacobian 765 . coloring - coloring context created with MatFDColoringCreate() 766 . x1 - location at which Jacobian is to be computed 767 - sctx - optional context required by function (actually a SNES context) 768 769 Options Database Keys: 770 . -mat_fd_coloring_freq <freq> - Sets coloring frequency 771 772 Level: intermediate 773 774 .seealso: MatFDColoringCreate(), MatFDColoringDestroy(), MatFDColoringView() 775 776 .keywords: coloring, Jacobian, finite differences 777 @*/ 778 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringApplyTS(Mat J,MatFDColoring coloring,PetscReal t,Vec x1,MatStructure *flag,void *sctx) 779 { 780 PetscErrorCode (*f)(void*,PetscReal,Vec,Vec,void*)=(PetscErrorCode (*)(void*,PetscReal,Vec,Vec,void *))coloring->f; 781 PetscErrorCode ierr; 782 PetscInt k,N,start,end,l,row,col,srow,**vscaleforrow; 783 PetscScalar dx,*y,*xx,*w3_array; 784 PetscScalar *vscale_array; 785 PetscReal epsilon = coloring->error_rel,umin = coloring->umin; 786 Vec w1,w2,w3; 787 void *fctx = coloring->fctx; 788 PetscTruth flg; 789 790 PetscFunctionBegin; 791 PetscValidHeaderSpecific(J,MAT_COOKIE,1); 792 PetscValidHeaderSpecific(coloring,MAT_FDCOLORING_COOKIE,2); 793 PetscValidHeaderSpecific(x1,VEC_COOKIE,4); 794 795 ierr = PetscLogEventBegin(MAT_FDColoringApply,coloring,J,x1,0);CHKERRQ(ierr); 796 if (!coloring->w1) { 797 ierr = VecDuplicate(x1,&coloring->w1);CHKERRQ(ierr); 798 ierr = PetscLogObjectParent(coloring,coloring->w1);CHKERRQ(ierr); 799 ierr = VecDuplicate(x1,&coloring->w2);CHKERRQ(ierr); 800 ierr = PetscLogObjectParent(coloring,coloring->w2);CHKERRQ(ierr); 801 ierr = VecDuplicate(x1,&coloring->w3);CHKERRQ(ierr); 802 ierr = PetscLogObjectParent(coloring,coloring->w3);CHKERRQ(ierr); 803 } 804 w1 = coloring->w1; w2 = coloring->w2; w3 = coloring->w3; 805 806 ierr = MatSetUnfactored(J);CHKERRQ(ierr); 807 ierr = PetscOptionsHasName(PETSC_NULL,"-mat_fd_coloring_dont_rezero",&flg);CHKERRQ(ierr); 808 if (flg) { 809 ierr = PetscInfo(coloring,"Not calling MatZeroEntries()\n");CHKERRQ(ierr); 810 } else { 811 PetscTruth assembled; 812 ierr = MatAssembled(J,&assembled);CHKERRQ(ierr); 813 if (assembled) { 814 ierr = MatZeroEntries(J);CHKERRQ(ierr); 815 } 816 } 817 818 ierr = VecGetOwnershipRange(x1,&start,&end);CHKERRQ(ierr); 819 ierr = VecGetSize(x1,&N);CHKERRQ(ierr); 820 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 821 ierr = (*f)(sctx,t,x1,w1,fctx);CHKERRQ(ierr); 822 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 823 824 /* 825 Compute all the scale factors and share with other processors 826 */ 827 ierr = VecGetArray(x1,&xx);CHKERRQ(ierr);xx = xx - start; 828 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr);vscale_array = vscale_array - start; 829 for (k=0; k<coloring->ncolors; k++) { 830 /* 831 Loop over each column associated with color adding the 832 perturbation to the vector w3. 833 */ 834 for (l=0; l<coloring->ncolumns[k]; l++) { 835 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 836 dx = xx[col]; 837 if (dx == 0.0) dx = 1.0; 838 #if !defined(PETSC_USE_COMPLEX) 839 if (dx < umin && dx >= 0.0) dx = umin; 840 else if (dx < 0.0 && dx > -umin) dx = -umin; 841 #else 842 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 843 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 844 #endif 845 dx *= epsilon; 846 vscale_array[col] = 1.0/dx; 847 } 848 } 849 vscale_array = vscale_array - start;ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 850 ierr = VecGhostUpdateBegin(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 851 ierr = VecGhostUpdateEnd(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 852 853 if (coloring->vscaleforrow) vscaleforrow = coloring->vscaleforrow; 854 else vscaleforrow = coloring->columnsforrow; 855 856 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 857 /* 858 Loop over each color 859 */ 860 for (k=0; k<coloring->ncolors; k++) { 861 ierr = VecCopy(x1,w3);CHKERRQ(ierr); 862 ierr = VecGetArray(w3,&w3_array);CHKERRQ(ierr);w3_array = w3_array - start; 863 /* 864 Loop over each column associated with color adding the 865 perturbation to the vector w3. 866 */ 867 for (l=0; l<coloring->ncolumns[k]; l++) { 868 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 869 dx = xx[col]; 870 if (dx == 0.0) dx = 1.0; 871 #if !defined(PETSC_USE_COMPLEX) 872 if (dx < umin && dx >= 0.0) dx = umin; 873 else if (dx < 0.0 && dx > -umin) dx = -umin; 874 #else 875 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 876 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 877 #endif 878 dx *= epsilon; 879 w3_array[col] += dx; 880 } 881 w3_array = w3_array + start; ierr = VecRestoreArray(w3,&w3_array);CHKERRQ(ierr); 882 883 /* 884 Evaluate function at x1 + dx (here dx is a vector of perturbations) 885 */ 886 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 887 ierr = (*f)(sctx,t,w3,w2,fctx);CHKERRQ(ierr); 888 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 889 ierr = VecAXPY(w2,-1.0,w1);CHKERRQ(ierr); 890 891 /* 892 Loop over rows of vector, putting results into Jacobian matrix 893 */ 894 ierr = VecGetArray(w2,&y);CHKERRQ(ierr); 895 for (l=0; l<coloring->nrows[k]; l++) { 896 row = coloring->rows[k][l]; 897 col = coloring->columnsforrow[k][l]; 898 y[row] *= vscale_array[vscaleforrow[k][l]]; 899 srow = row + start; 900 ierr = MatSetValues(J,1,&srow,1,&col,y+row,INSERT_VALUES);CHKERRQ(ierr); 901 } 902 ierr = VecRestoreArray(w2,&y);CHKERRQ(ierr); 903 } 904 ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 905 xx = xx + start; ierr = VecRestoreArray(x1,&xx);CHKERRQ(ierr); 906 ierr = MatAssemblyBegin(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 907 ierr = MatAssemblyEnd(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 908 ierr = PetscLogEventEnd(MAT_FDColoringApply,coloring,J,x1,0);CHKERRQ(ierr); 909 PetscFunctionReturn(0); 910 } 911 912 913 #undef __FUNCT__ 914 #define __FUNCT__ "MatFDColoringSetRecompute()" 915 /*@C 916 MatFDColoringSetRecompute - Indicates that the next time a Jacobian preconditioner 917 is needed it sholuld be recomputed. Once this is called and the new Jacobian is computed 918 no additional Jacobian's will be computed (the same one will be used) until this is 919 called again. 920 921 Collective on MatFDColoring 922 923 Input Parameters: 924 . c - the coloring context 925 926 Level: intermediate 927 928 Notes: The MatFDColoringSetFrequency() is ignored once this is called 929 930 .seealso: MatFDColoringCreate(), MatFDColoringSetFrequency() 931 932 .keywords: Mat, finite differences, coloring 933 @*/ 934 PetscErrorCode PETSCMAT_DLLEXPORT MatFDColoringSetRecompute(MatFDColoring c) 935 { 936 PetscFunctionBegin; 937 PetscValidHeaderSpecific(c,MAT_FDCOLORING_COOKIE,1); 938 c->usersetsrecompute = PETSC_TRUE; 939 c->recompute = PETSC_TRUE; 940 PetscFunctionReturn(0); 941 } 942 943 944