1 #define PETSCMAT_DLL 2 3 /* 4 Provides an interface to the MUMPS sparse solver 5 */ 6 #include "../src/mat/impls/aij/seq/aij.h" /*I "petscmat.h" I*/ 7 #include "../src/mat/impls/aij/mpi/mpiaij.h" 8 #include "../src/mat/impls/sbaij/seq/sbaij.h" 9 #include "../src/mat/impls/sbaij/mpi/mpisbaij.h" 10 #include "../src/mat/impls/baij/seq/baij.h" 11 #include "../src/mat/impls/baij/mpi/mpibaij.h" 12 13 EXTERN_C_BEGIN 14 #if defined(PETSC_USE_COMPLEX) 15 #include "zmumps_c.h" 16 #else 17 #include "dmumps_c.h" 18 #endif 19 EXTERN_C_END 20 #define JOB_INIT -1 21 #define JOB_END -2 22 /* macros s.t. indices match MUMPS documentation */ 23 #define ICNTL(I) icntl[(I)-1] 24 #define CNTL(I) cntl[(I)-1] 25 #define INFOG(I) infog[(I)-1] 26 #define INFO(I) info[(I)-1] 27 #define RINFOG(I) rinfog[(I)-1] 28 #define RINFO(I) rinfo[(I)-1] 29 30 typedef struct { 31 #if defined(PETSC_USE_COMPLEX) 32 ZMUMPS_STRUC_C id; 33 #else 34 DMUMPS_STRUC_C id; 35 #endif 36 MatStructure matstruc; 37 PetscMPIInt myid,size; 38 PetscInt *irn,*jcn,nz,sym,nSolve; 39 PetscScalar *val; 40 MPI_Comm comm_mumps; 41 VecScatter scat_rhs, scat_sol; 42 PetscTruth isAIJ,CleanUpMUMPS,mumpsview; 43 Vec b_seq,x_seq; 44 PetscErrorCode (*MatDestroy)(Mat); 45 PetscErrorCode (*ConvertToTriples)(Mat, int, MatReuse, int*, int**, int**, PetscScalar**); 46 } Mat_MUMPS; 47 48 EXTERN PetscErrorCode MatDuplicate_MUMPS(Mat,MatDuplicateOption,Mat*); 49 50 51 /* MatConvertToTriples_A_B */ 52 /*convert Petsc matrix to triples: row[nz], col[nz], val[nz] */ 53 /* 54 input: 55 A - matrix in aij,baij or sbaij (bs=1) format 56 shift - 0: C style output triple; 1: Fortran style output triple. 57 reuse - MAT_INITIAL_MATRIX: spaces are allocated and values are set for the triple 58 MAT_REUSE_MATRIX: only the values in v array are updated 59 output: 60 nnz - dim of r, c, and v (number of local nonzero entries of A) 61 r, c, v - row and col index, matrix values (matrix triples) 62 */ 63 64 #undef __FUNCT__ 65 #define __FUNCT__ "MatConvertToTriples_seqaij_seqaij" 66 PetscErrorCode MatConvertToTriples_seqaij_seqaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 67 { 68 const PetscInt *ai,*aj,*ajj,M=A->rmap->n; 69 PetscInt nz,rnz,i,j; 70 PetscErrorCode ierr; 71 PetscInt *row,*col; 72 Mat_SeqAIJ *aa=(Mat_SeqAIJ*)A->data; 73 74 PetscFunctionBegin; 75 *v=aa->a; 76 if (reuse == MAT_INITIAL_MATRIX){ 77 nz = aa->nz; ai = aa->i; aj = aa->j; 78 *nnz = nz; 79 ierr = PetscMalloc(2*nz*sizeof(PetscInt), &row);CHKERRQ(ierr); 80 col = row + nz; 81 82 nz = 0; 83 for(i=0; i<M; i++) { 84 rnz = ai[i+1] - ai[i]; 85 ajj = aj + ai[i]; 86 for(j=0; j<rnz; j++) { 87 row[nz] = i+shift; col[nz++] = ajj[j] + shift; 88 } 89 } 90 *r = row; *c = col; 91 } 92 PetscFunctionReturn(0); 93 } 94 95 #undef __FUNCT__ 96 #define __FUNCT__ "MatConvertToTriples_seqbaij_seqaij" 97 PetscErrorCode MatConvertToTriples_seqbaij_seqaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 98 { 99 Mat_SeqBAIJ *aa=(Mat_SeqBAIJ*)A->data; 100 const PetscInt *ai,*aj,*ajj,bs=A->rmap->bs,bs2=aa->bs2,M=A->rmap->N/bs; 101 PetscInt nz,idx=0,rnz,i,j,k,m,ii; 102 PetscErrorCode ierr; 103 PetscInt *row,*col; 104 105 PetscFunctionBegin; 106 *v = aa->a; 107 if (reuse == MAT_INITIAL_MATRIX){ 108 ai = aa->i; aj = aa->j; 109 nz = bs2*aa->nz; 110 *nnz = nz; 111 ierr = PetscMalloc(2*nz*sizeof(PetscInt), &row);CHKERRQ(ierr); 112 col = row + nz; 113 114 for(i=0; i<M; i++) { 115 ii = 0; 116 ajj = aj + ai[i]; 117 rnz = ai[i+1] - ai[i]; 118 for(k=0; k<rnz; k++) { 119 for(j=0; j<bs; j++) { 120 for(m=0; m<bs; m++) { 121 row[idx] = i*bs + m + shift; 122 col[idx++] = bs*(ajj[k]) + j + shift; 123 } 124 } 125 } 126 } 127 *r = row; *c = col; 128 } 129 PetscFunctionReturn(0); 130 } 131 132 #undef __FUNCT__ 133 #define __FUNCT__ "MatConvertToTriples_seqsbaij_seqsbaij" 134 PetscErrorCode MatConvertToTriples_seqsbaij_seqsbaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 135 { 136 const PetscInt *ai, *aj,*ajj,M=A->rmap->n; 137 PetscInt nz,rnz,i,j; 138 PetscErrorCode ierr; 139 PetscInt *row,*col; 140 Mat_SeqSBAIJ *aa=(Mat_SeqSBAIJ*)A->data; 141 142 PetscFunctionBegin; 143 if (reuse == MAT_INITIAL_MATRIX){ 144 nz = aa->nz;ai=aa->i; aj=aa->j;*v=aa->a; 145 *nnz = nz; 146 ierr = PetscMalloc(2*nz*sizeof(PetscInt), &row);CHKERRQ(ierr); 147 col = row + nz; 148 149 nz = 0; 150 for(i=0; i<M; i++) { 151 rnz = ai[i+1] - ai[i]; 152 ajj = aj + ai[i]; 153 for(j=0; j<rnz; j++) { 154 row[nz] = i+shift; col[nz++] = ajj[j] + shift; 155 } 156 } 157 *r = row; *c = col; 158 } 159 PetscFunctionReturn(0); 160 } 161 162 #undef __FUNCT__ 163 #define __FUNCT__ "MatConvertToTriples_seqaij_seqsbaij" 164 PetscErrorCode MatConvertToTriples_seqaij_seqsbaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 165 { 166 const PetscInt *ai,*aj,*ajj,*adiag,M=A->rmap->n; 167 PetscInt nz,rnz,i,j; 168 const PetscScalar *av,*v1; 169 PetscScalar *val; 170 PetscErrorCode ierr; 171 PetscInt *row,*col; 172 Mat_SeqSBAIJ *aa=(Mat_SeqSBAIJ*)A->data; 173 174 PetscFunctionBegin; 175 ai=aa->i; aj=aa->j;av=aa->a; 176 adiag=aa->diag; 177 if (reuse == MAT_INITIAL_MATRIX){ 178 nz = M + (aa->nz-M)/2; 179 *nnz = nz; 180 ierr = PetscMalloc((2*nz*sizeof(PetscInt)+nz*sizeof(PetscScalar)), &row);CHKERRQ(ierr); 181 col = row + nz; 182 val = (PetscScalar*)(col + nz); 183 184 nz = 0; 185 for(i=0; i<M; i++) { 186 rnz = ai[i+1] - adiag[i]; 187 ajj = aj + adiag[i]; 188 v1 = av + adiag[i]; 189 for(j=0; j<rnz; j++) { 190 row[nz] = i+shift; col[nz] = ajj[j] + shift; val[nz++] = v1[j]; 191 } 192 } 193 *r = row; *c = col; *v = val; 194 } else { 195 nz = 0; val = *v; 196 for(i=0; i <M; i++) { 197 rnz = ai[i+1] - adiag[i]; 198 ajj = aj + adiag[i]; 199 v1 = av + adiag[i]; 200 for(j=0; j<rnz; j++) { 201 val[nz++] = v1[j]; 202 } 203 } 204 } 205 PetscFunctionReturn(0); 206 } 207 208 #undef __FUNCT__ 209 #define __FUNCT__ "MatConvertToTriples_mpisbaij_mpisbaij" 210 PetscErrorCode MatConvertToTriples_mpisbaij_mpisbaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 211 { 212 const PetscInt *ai, *aj, *bi, *bj,*garray,m=A->rmap->n,*ajj,*bjj; 213 PetscErrorCode ierr; 214 PetscInt rstart,nz,i,j,jj,irow,countA,countB; 215 PetscInt *row,*col; 216 const PetscScalar *av, *bv,*v1,*v2; 217 PetscScalar *val; 218 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)A->data; 219 Mat_SeqSBAIJ *aa=(Mat_SeqSBAIJ*)(mat->A)->data; 220 Mat_SeqBAIJ *bb=(Mat_SeqBAIJ*)(mat->B)->data; 221 222 PetscFunctionBegin; 223 ai=aa->i; aj=aa->j; bi=bb->i; bj=bb->j; rstart= A->rmap->rstart; 224 garray = mat->garray; 225 av=aa->a; bv=bb->a; 226 227 if (reuse == MAT_INITIAL_MATRIX){ 228 nz = aa->nz + bb->nz; 229 *nnz = nz; 230 ierr = PetscMalloc((2*nz*sizeof(PetscInt)+nz*sizeof(PetscScalar)), &row);CHKERRQ(ierr); 231 col = row + nz; 232 val = (PetscScalar*)(col + nz); 233 234 *r = row; *c = col; *v = val; 235 } else { 236 row = *r; col = *c; val = *v; 237 } 238 239 jj = 0; irow = rstart; 240 for ( i=0; i<m; i++ ) { 241 ajj = aj + ai[i]; /* ptr to the beginning of this row */ 242 countA = ai[i+1] - ai[i]; 243 countB = bi[i+1] - bi[i]; 244 bjj = bj + bi[i]; 245 v1 = av + ai[i]; 246 v2 = bv + bi[i]; 247 248 /* A-part */ 249 for (j=0; j<countA; j++){ 250 if (reuse == MAT_INITIAL_MATRIX) { 251 row[jj] = irow + shift; col[jj] = rstart + ajj[j] + shift; 252 } 253 val[jj++] = v1[j]; 254 } 255 256 /* B-part */ 257 for(j=0; j < countB; j++){ 258 if (reuse == MAT_INITIAL_MATRIX) { 259 row[jj] = irow + shift; col[jj] = garray[bjj[j]] + shift; 260 } 261 val[jj++] = v2[j]; 262 } 263 irow++; 264 } 265 PetscFunctionReturn(0); 266 } 267 268 #undef __FUNCT__ 269 #define __FUNCT__ "MatConvertToTriples_mpiaij_mpiaij" 270 PetscErrorCode MatConvertToTriples_mpiaij_mpiaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 271 { 272 const PetscInt *ai, *aj, *bi, *bj,*garray,m=A->rmap->n,*ajj,*bjj; 273 PetscErrorCode ierr; 274 PetscInt rstart,nz,i,j,jj,irow,countA,countB; 275 PetscInt *row,*col; 276 const PetscScalar *av, *bv,*v1,*v2; 277 PetscScalar *val; 278 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)A->data; 279 Mat_SeqAIJ *aa=(Mat_SeqAIJ*)(mat->A)->data; 280 Mat_SeqAIJ *bb=(Mat_SeqAIJ*)(mat->B)->data; 281 282 PetscFunctionBegin; 283 ai=aa->i; aj=aa->j; bi=bb->i; bj=bb->j; rstart= A->rmap->rstart; 284 garray = mat->garray; 285 av=aa->a; bv=bb->a; 286 287 if (reuse == MAT_INITIAL_MATRIX){ 288 nz = aa->nz + bb->nz; 289 *nnz = nz; 290 ierr = PetscMalloc((2*nz*sizeof(PetscInt)+nz*sizeof(PetscScalar)), &row);CHKERRQ(ierr); 291 col = row + nz; 292 val = (PetscScalar*)(col + nz); 293 294 *r = row; *c = col; *v = val; 295 } else { 296 row = *r; col = *c; val = *v; 297 } 298 299 jj = 0; irow = rstart; 300 for ( i=0; i<m; i++ ) { 301 ajj = aj + ai[i]; /* ptr to the beginning of this row */ 302 countA = ai[i+1] - ai[i]; 303 countB = bi[i+1] - bi[i]; 304 bjj = bj + bi[i]; 305 v1 = av + ai[i]; 306 v2 = bv + bi[i]; 307 308 /* A-part */ 309 for (j=0; j<countA; j++){ 310 if (reuse == MAT_INITIAL_MATRIX){ 311 row[jj] = irow + shift; col[jj] = rstart + ajj[j] + shift; 312 } 313 val[jj++] = v1[j]; 314 } 315 316 /* B-part */ 317 for(j=0; j < countB; j++){ 318 if (reuse == MAT_INITIAL_MATRIX){ 319 row[jj] = irow + shift; col[jj] = garray[bjj[j]] + shift; 320 } 321 val[jj++] = v2[j]; 322 } 323 irow++; 324 } 325 PetscFunctionReturn(0); 326 } 327 328 #undef __FUNCT__ 329 #define __FUNCT__ "MatConvertToTriples_mpibaij_mpiaij" 330 PetscErrorCode MatConvertToTriples_mpibaij_mpiaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 331 { 332 Mat_MPIBAIJ *mat = (Mat_MPIBAIJ*)A->data; 333 Mat_SeqBAIJ *aa=(Mat_SeqBAIJ*)(mat->A)->data; 334 Mat_SeqBAIJ *bb=(Mat_SeqBAIJ*)(mat->B)->data; 335 const PetscInt *ai = aa->i, *bi = bb->i, *aj = aa->j, *bj = bb->j,*ajj, *bjj; 336 const PetscInt *garray = mat->garray,mbs=mat->mbs,rstartbs=mat->rstartbs; 337 const PetscInt bs = A->rmap->bs,bs2=mat->bs2; 338 PetscErrorCode ierr; 339 PetscInt nz,i,j,k,n,jj,irow,countA,countB,idx; 340 PetscInt *row,*col; 341 const PetscScalar *av=aa->a, *bv=bb->a,*v1,*v2; 342 PetscScalar *val; 343 344 PetscFunctionBegin; 345 346 if (reuse == MAT_INITIAL_MATRIX) { 347 nz = bs2*(aa->nz + bb->nz); 348 *nnz = nz; 349 ierr = PetscMalloc((2*nz*sizeof(PetscInt)+nz*sizeof(PetscScalar)), &row);CHKERRQ(ierr); 350 col = row + nz; 351 val = (PetscScalar*)(col + nz); 352 353 *r = row; *c = col; *v = val; 354 } else { 355 row = *r; col = *c; val = *v; 356 } 357 358 jj = 0; irow = rstartbs; 359 for ( i=0; i<mbs; i++ ) { 360 countA = ai[i+1] - ai[i]; 361 countB = bi[i+1] - bi[i]; 362 ajj = aj + ai[i]; 363 bjj = bj + bi[i]; 364 v1 = av + bs2*ai[i]; 365 v2 = bv + bs2*bi[i]; 366 367 idx = 0; 368 /* A-part */ 369 for (k=0; k<countA; k++){ 370 for (j=0; j<bs; j++) { 371 for (n=0; n<bs; n++) { 372 if (reuse == MAT_INITIAL_MATRIX){ 373 row[jj] = bs*irow + n + shift; 374 col[jj] = bs*(rstartbs + ajj[k]) + j + shift; 375 } 376 val[jj++] = v1[idx++]; 377 } 378 } 379 } 380 381 idx = 0; 382 /* B-part */ 383 for(k=0; k<countB; k++){ 384 for (j=0; j<bs; j++) { 385 for (n=0; n<bs; n++) { 386 if (reuse == MAT_INITIAL_MATRIX){ 387 row[jj] = bs*irow + n + shift; 388 col[jj] = bs*(garray[bjj[k]]) + j + shift; 389 } 390 val[jj++] = bv[idx++]; 391 } 392 } 393 } 394 irow++; 395 } 396 PetscFunctionReturn(0); 397 } 398 399 #undef __FUNCT__ 400 #define __FUNCT__ "MatConvertToTriples_mpiaij_mpisbaij" 401 PetscErrorCode MatConvertToTriples_mpiaij_mpisbaij(Mat A,int shift,MatReuse reuse,int *nnz,int **r, int **c, PetscScalar **v) 402 { 403 const PetscInt *ai, *aj,*adiag, *bi, *bj,*garray,m=A->rmap->n,*ajj,*bjj; 404 PetscErrorCode ierr; 405 PetscInt rstart,nz,nza,nzb_low,i,j,jj,irow,countA,countB; 406 PetscInt *row,*col; 407 const PetscScalar *av, *bv,*v1,*v2; 408 PetscScalar *val; 409 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)A->data; 410 Mat_SeqAIJ *aa=(Mat_SeqAIJ*)(mat->A)->data; 411 Mat_SeqAIJ *bb=(Mat_SeqAIJ*)(mat->B)->data; 412 413 PetscFunctionBegin; 414 ai=aa->i; aj=aa->j; adiag=aa->diag; 415 bi=bb->i; bj=bb->j; garray = mat->garray; 416 av=aa->a; bv=bb->a; 417 rstart = A->rmap->rstart; 418 419 if (reuse == MAT_INITIAL_MATRIX) { 420 nza = 0;nzb_low = 0; 421 for(i=0; i<m; i++){ 422 nza = nza + (ai[i+1] - adiag[i]); 423 countB = bi[i+1] - bi[i]; 424 bjj = bj + bi[i]; 425 426 j = 0; 427 while(garray[bjj[j]] < rstart) { 428 if(j == countB) break; 429 j++;nzb_low++; 430 } 431 } 432 /* Total nz = nz for the upper triangular A part + nz for the 2nd B part */ 433 nz = nza + (bb->nz - nzb_low); 434 *nnz = nz; 435 ierr = PetscMalloc((2*nz*sizeof(PetscInt)+nz*sizeof(PetscScalar)), &row);CHKERRQ(ierr); 436 col = row + nz; 437 val = (PetscScalar*)(col + nz); 438 439 *r = row; *c = col; *v = val; 440 } else { 441 row = *r; col = *c; val = *v; 442 } 443 444 jj = 0; irow = rstart; 445 for ( i=0; i<m; i++ ) { 446 ajj = aj + adiag[i]; /* ptr to the beginning of the diagonal of this row */ 447 v1 = av + adiag[i]; 448 countA = ai[i+1] - adiag[i]; 449 countB = bi[i+1] - bi[i]; 450 bjj = bj + bi[i]; 451 v2 = bv + bi[i]; 452 453 /* A-part */ 454 for (j=0; j<countA; j++){ 455 if (reuse == MAT_INITIAL_MATRIX) { 456 row[jj] = irow + shift; col[jj] = rstart + ajj[j] + shift; 457 } 458 val[jj++] = v1[j]; 459 } 460 461 /* B-part */ 462 for(j=0; j < countB; j++){ 463 if (garray[bjj[j]] > rstart) { 464 if (reuse == MAT_INITIAL_MATRIX) { 465 row[jj] = irow + shift; col[jj] = garray[bjj[j]] + shift; 466 } 467 val[jj++] = v2[j]; 468 } 469 } 470 irow++; 471 } 472 PetscFunctionReturn(0); 473 } 474 475 #undef __FUNCT__ 476 #define __FUNCT__ "MatDestroy_MUMPS" 477 PetscErrorCode MatDestroy_MUMPS(Mat A) 478 { 479 Mat_MUMPS *lu=(Mat_MUMPS*)A->spptr; 480 PetscErrorCode ierr; 481 PetscMPIInt size=lu->size; 482 483 PetscFunctionBegin; 484 if (lu->CleanUpMUMPS) { 485 /* Terminate instance, deallocate memories */ 486 if (size > 1){ 487 ierr = PetscFree2(lu->id.sol_loc,lu->id.isol_loc);CHKERRQ(ierr); 488 ierr = VecScatterDestroy(lu->scat_rhs);CHKERRQ(ierr); 489 ierr = VecDestroy(lu->b_seq);CHKERRQ(ierr); 490 if (lu->nSolve && lu->scat_sol){ierr = VecScatterDestroy(lu->scat_sol);CHKERRQ(ierr);} 491 if (lu->nSolve && lu->x_seq){ierr = VecDestroy(lu->x_seq);CHKERRQ(ierr);} 492 } 493 ierr = PetscFree(lu->irn);CHKERRQ(ierr); 494 lu->id.job=JOB_END; 495 #if defined(PETSC_USE_COMPLEX) 496 zmumps_c(&lu->id); 497 #else 498 dmumps_c(&lu->id); 499 #endif 500 ierr = MPI_Comm_free(&(lu->comm_mumps));CHKERRQ(ierr); 501 } 502 /* clear composed functions */ 503 ierr = PetscObjectComposeFunctionDynamic((PetscObject)A,"MatFactorGetSolverPackage_C","",PETSC_NULL);CHKERRQ(ierr); 504 ierr = PetscObjectComposeFunctionDynamic((PetscObject)A,"MatMumpsSetIcntl_C","",PETSC_NULL);CHKERRQ(ierr); 505 ierr = (lu->MatDestroy)(A);CHKERRQ(ierr); 506 PetscFunctionReturn(0); 507 } 508 509 #undef __FUNCT__ 510 #define __FUNCT__ "MatSolve_MUMPS" 511 PetscErrorCode MatSolve_MUMPS(Mat A,Vec b,Vec x) 512 { 513 Mat_MUMPS *lu=(Mat_MUMPS*)A->spptr; 514 PetscScalar *array; 515 Vec b_seq; 516 IS is_iden,is_petsc; 517 PetscErrorCode ierr; 518 PetscInt i; 519 520 PetscFunctionBegin; 521 lu->id.nrhs = 1; 522 b_seq = lu->b_seq; 523 if (lu->size > 1){ 524 /* MUMPS only supports centralized rhs. Scatter b into a seqential rhs vector */ 525 ierr = VecScatterBegin(lu->scat_rhs,b,b_seq,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 526 ierr = VecScatterEnd(lu->scat_rhs,b,b_seq,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 527 if (!lu->myid) {ierr = VecGetArray(b_seq,&array);CHKERRQ(ierr);} 528 } else { /* size == 1 */ 529 ierr = VecCopy(b,x);CHKERRQ(ierr); 530 ierr = VecGetArray(x,&array);CHKERRQ(ierr); 531 } 532 if (!lu->myid) { /* define rhs on the host */ 533 lu->id.nrhs = 1; 534 #if defined(PETSC_USE_COMPLEX) 535 lu->id.rhs = (mumps_double_complex*)array; 536 #else 537 lu->id.rhs = array; 538 #endif 539 } 540 541 /* solve phase */ 542 /*-------------*/ 543 lu->id.job = 3; 544 #if defined(PETSC_USE_COMPLEX) 545 zmumps_c(&lu->id); 546 #else 547 dmumps_c(&lu->id); 548 #endif 549 if (lu->id.INFOG(1) < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in solve phase: INFOG(1)=%d\n",lu->id.INFOG(1)); 550 551 if (lu->size > 1) { /* convert mumps distributed solution to petsc mpi x */ 552 if (!lu->nSolve){ /* create scatter scat_sol */ 553 ierr = ISCreateStride(PETSC_COMM_SELF,lu->id.lsol_loc,0,1,&is_iden);CHKERRQ(ierr); /* from */ 554 for (i=0; i<lu->id.lsol_loc; i++){ 555 lu->id.isol_loc[i] -= 1; /* change Fortran style to C style */ 556 } 557 ierr = ISCreateGeneral(PETSC_COMM_SELF,lu->id.lsol_loc,lu->id.isol_loc,&is_petsc);CHKERRQ(ierr); /* to */ 558 ierr = VecScatterCreate(lu->x_seq,is_iden,x,is_petsc,&lu->scat_sol);CHKERRQ(ierr); 559 ierr = ISDestroy(is_iden);CHKERRQ(ierr); 560 ierr = ISDestroy(is_petsc);CHKERRQ(ierr); 561 } 562 ierr = VecScatterBegin(lu->scat_sol,lu->x_seq,x,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 563 ierr = VecScatterEnd(lu->scat_sol,lu->x_seq,x,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 564 } 565 lu->nSolve++; 566 PetscFunctionReturn(0); 567 } 568 569 #if !defined(PETSC_USE_COMPLEX) 570 /* 571 input: 572 F: numeric factor 573 output: 574 nneg: total number of negative pivots 575 nzero: 0 576 npos: (global dimension of F) - nneg 577 */ 578 579 #undef __FUNCT__ 580 #define __FUNCT__ "MatGetInertia_SBAIJMUMPS" 581 PetscErrorCode MatGetInertia_SBAIJMUMPS(Mat F,int *nneg,int *nzero,int *npos) 582 { 583 Mat_MUMPS *lu =(Mat_MUMPS*)F->spptr; 584 PetscErrorCode ierr; 585 PetscMPIInt size; 586 587 PetscFunctionBegin; 588 ierr = MPI_Comm_size(((PetscObject)F)->comm,&size);CHKERRQ(ierr); 589 /* MUMPS 4.3.1 calls ScaLAPACK when ICNTL(13)=0 (default), which does not offer the possibility to compute the inertia of a dense matrix. Set ICNTL(13)=1 to skip ScaLAPACK */ 590 if (size > 1 && lu->id.ICNTL(13) != 1) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"ICNTL(13)=%d. -mat_mumps_icntl_13 must be set as 1 for correct global matrix inertia\n",lu->id.INFOG(13)); 591 if (nneg){ 592 if (!lu->myid){ 593 *nneg = lu->id.INFOG(12); 594 } 595 ierr = MPI_Bcast(nneg,1,MPI_INT,0,lu->comm_mumps);CHKERRQ(ierr); 596 } 597 if (nzero) *nzero = 0; 598 if (npos) *npos = F->rmap->N - (*nneg); 599 PetscFunctionReturn(0); 600 } 601 #endif /* !defined(PETSC_USE_COMPLEX) */ 602 603 #undef __FUNCT__ 604 #define __FUNCT__ "MatFactorNumeric_MUMPS" 605 PetscErrorCode MatFactorNumeric_MUMPS(Mat F,Mat A,const MatFactorInfo *info) 606 { 607 Mat_MUMPS *lu =(Mat_MUMPS*)(F)->spptr; 608 PetscErrorCode ierr; 609 MatReuse reuse; 610 Mat F_diag; 611 PetscTruth isMPIAIJ; 612 613 PetscFunctionBegin; 614 reuse = MAT_REUSE_MATRIX; 615 ierr = (*lu->ConvertToTriples)(A, 1, reuse, &lu->nz, &lu->irn, &lu->jcn, &lu->val);CHKERRQ(ierr); 616 617 /* numerical factorization phase */ 618 /*-------------------------------*/ 619 lu->id.job = 2; 620 if(!lu->id.ICNTL(18)) { 621 if (!lu->myid) { 622 #if defined(PETSC_USE_COMPLEX) 623 lu->id.a = (mumps_double_complex*)lu->val; 624 #else 625 lu->id.a = lu->val; 626 #endif 627 } 628 } else { 629 #if defined(PETSC_USE_COMPLEX) 630 lu->id.a_loc = (mumps_double_complex*)lu->val; 631 #else 632 lu->id.a_loc = lu->val; 633 #endif 634 } 635 #if defined(PETSC_USE_COMPLEX) 636 zmumps_c(&lu->id); 637 #else 638 dmumps_c(&lu->id); 639 #endif 640 if (lu->id.INFOG(1) < 0) { 641 if (lu->id.INFO(1) == -13) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in numerical factorization phase: Cannot allocate required memory %d megabytes\n",lu->id.INFO(2)); 642 else SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in numerical factorization phase: INFO(1)=%d, INFO(2)=%d\n",lu->id.INFO(1),lu->id.INFO(2)); 643 } 644 if (!lu->myid && lu->id.ICNTL(16) > 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB," lu->id.ICNTL(16):=%d\n",lu->id.INFOG(16)); 645 646 if (lu->size > 1){ 647 ierr = PetscTypeCompare((PetscObject)A,MATMPIAIJ,&isMPIAIJ);CHKERRQ(ierr); 648 if(isMPIAIJ) { 649 F_diag = ((Mat_MPIAIJ *)(F)->data)->A; 650 } else { 651 F_diag = ((Mat_MPISBAIJ *)(F)->data)->A; 652 } 653 F_diag->assembled = PETSC_TRUE; 654 if (lu->nSolve){ 655 ierr = VecScatterDestroy(lu->scat_sol);CHKERRQ(ierr); 656 ierr = PetscFree2(lu->id.sol_loc,lu->id.isol_loc);CHKERRQ(ierr); 657 ierr = VecDestroy(lu->x_seq);CHKERRQ(ierr); 658 } 659 } 660 (F)->assembled = PETSC_TRUE; 661 lu->matstruc = SAME_NONZERO_PATTERN; 662 lu->CleanUpMUMPS = PETSC_TRUE; 663 lu->nSolve = 0; 664 665 if (lu->size > 1){ 666 /* distributed solution */ 667 lu->id.ICNTL(21) = 1; 668 if (!lu->nSolve){ 669 /* Create x_seq=sol_loc for repeated use */ 670 PetscInt lsol_loc; 671 PetscScalar *sol_loc; 672 lsol_loc = lu->id.INFO(23); /* length of sol_loc */ 673 ierr = PetscMalloc2(lsol_loc,PetscScalar,&sol_loc,lsol_loc,PetscInt,&lu->id.isol_loc);CHKERRQ(ierr); 674 lu->id.lsol_loc = lsol_loc; 675 #if defined(PETSC_USE_COMPLEX) 676 lu->id.sol_loc = (mumps_double_complex*)sol_loc; 677 #else 678 lu->id.sol_loc = sol_loc; 679 #endif 680 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,lsol_loc,sol_loc,&lu->x_seq);CHKERRQ(ierr); 681 } 682 } 683 PetscFunctionReturn(0); 684 } 685 686 #undef __FUNCT__ 687 #define __FUNCT__ "PetscSetMUMPSOptions" 688 PetscErrorCode PetscSetMUMPSOptions(Mat F, Mat A) 689 { 690 Mat_MUMPS *lu = (Mat_MUMPS*)F->spptr; 691 PetscErrorCode ierr; 692 PetscInt icntl; 693 PetscTruth flg; 694 695 PetscFunctionBegin; 696 ierr = PetscOptionsBegin(((PetscObject)A)->comm,((PetscObject)A)->prefix,"MUMPS Options","Mat");CHKERRQ(ierr); 697 ierr = PetscOptionsTruth("-mat_mumps_view","View MUMPS parameters","None",lu->mumpsview,&lu->mumpsview,PETSC_NULL);CHKERRQ(ierr); 698 if (lu->size == 1){ 699 lu->id.ICNTL(18) = 0; /* centralized assembled matrix input */ 700 } else { 701 lu->id.ICNTL(18) = 3; /* distributed assembled matrix input */ 702 } 703 704 icntl=-1; 705 lu->id.ICNTL(4) = 0; /* level of printing; overwrite mumps default ICNTL(4)=2 */ 706 ierr = PetscOptionsInt("-mat_mumps_icntl_4","ICNTL(4): level of printing (0 to 4)","None",lu->id.ICNTL(4),&icntl,&flg);CHKERRQ(ierr); 707 if ((flg && icntl > 0) || PetscLogPrintInfo) { 708 lu->id.ICNTL(4)=icntl; /* and use mumps default icntl(i), i=1,2,3 */ 709 } else { /* no output */ 710 lu->id.ICNTL(1) = 0; /* error message, default= 6 */ 711 lu->id.ICNTL(2) = 0; /* output stream for diagnostic printing, statistics, and warning. default=0 */ 712 lu->id.ICNTL(3) = 0; /* output stream for global information, default=6 */ 713 } 714 ierr = PetscOptionsInt("-mat_mumps_icntl_6","ICNTL(6): column permutation and/or scaling to get a zero-free diagonal (0 to 7)","None",lu->id.ICNTL(6),&lu->id.ICNTL(6),PETSC_NULL);CHKERRQ(ierr); 715 icntl=-1; 716 ierr = PetscOptionsInt("-mat_mumps_icntl_7","ICNTL(7): matrix ordering (0 to 7)","None",lu->id.ICNTL(7),&icntl,&flg);CHKERRQ(ierr); 717 if (flg) { 718 if (icntl== 1){ 719 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"pivot order be set by the user in PERM_IN -- not supported by the PETSc/MUMPS interface\n"); 720 } else { 721 lu->id.ICNTL(7) = icntl; 722 } 723 } 724 ierr = PetscOptionsInt("-mat_mumps_icntl_8","ICNTL(8): scaling strategy (-2 to 7 or 77)","None",lu->id.ICNTL(8),&lu->id.ICNTL(8),PETSC_NULL);CHKERRQ(ierr); 725 ierr = PetscOptionsInt("-mat_mumps_icntl_9","ICNTL(9): A or A^T x=b to be solved. 1: A; otherwise: A^T","None",lu->id.ICNTL(9),&lu->id.ICNTL(9),PETSC_NULL);CHKERRQ(ierr); 726 ierr = PetscOptionsInt("-mat_mumps_icntl_10","ICNTL(10): max num of refinements","None",lu->id.ICNTL(10),&lu->id.ICNTL(10),PETSC_NULL);CHKERRQ(ierr); 727 ierr = PetscOptionsInt("-mat_mumps_icntl_11","ICNTL(11): statistics related to the linear system solved (via -ksp_view)","None",lu->id.ICNTL(11),&lu->id.ICNTL(11),PETSC_NULL);CHKERRQ(ierr); 728 ierr = PetscOptionsInt("-mat_mumps_icntl_12","ICNTL(12): efficiency control: defines the ordering strategy with scaling constraints (0 to 3","None",lu->id.ICNTL(12),&lu->id.ICNTL(12),PETSC_NULL);CHKERRQ(ierr); 729 ierr = PetscOptionsInt("-mat_mumps_icntl_13","ICNTL(13): efficiency control: with or without ScaLAPACK","None",lu->id.ICNTL(13),&lu->id.ICNTL(13),PETSC_NULL);CHKERRQ(ierr); 730 ierr = PetscOptionsInt("-mat_mumps_icntl_14","ICNTL(14): percentage of estimated workspace increase","None",lu->id.ICNTL(14),&lu->id.ICNTL(14),PETSC_NULL);CHKERRQ(ierr); 731 ierr = PetscOptionsInt("-mat_mumps_icntl_19","ICNTL(19): Schur complement","None",lu->id.ICNTL(19),&lu->id.ICNTL(19),PETSC_NULL);CHKERRQ(ierr); 732 733 ierr = PetscOptionsInt("-mat_mumps_icntl_22","ICNTL(22): in-core/out-of-core facility (0 or 1)","None",lu->id.ICNTL(22),&lu->id.ICNTL(22),PETSC_NULL);CHKERRQ(ierr); 734 ierr = PetscOptionsInt("-mat_mumps_icntl_23","ICNTL(23): max size of the working memory (MB) that can allocate per processor","None",lu->id.ICNTL(23),&lu->id.ICNTL(23),PETSC_NULL);CHKERRQ(ierr); 735 ierr = PetscOptionsInt("-mat_mumps_icntl_24","ICNTL(24): detection of null pivot rows (0 or 1)","None",lu->id.ICNTL(24),&lu->id.ICNTL(24),PETSC_NULL);CHKERRQ(ierr); 736 ierr = PetscOptionsInt("-mat_mumps_icntl_25","ICNTL(25): computation of a null space basis","None",lu->id.ICNTL(25),&lu->id.ICNTL(25),PETSC_NULL);CHKERRQ(ierr); 737 ierr = PetscOptionsInt("-mat_mumps_icntl_26","ICNTL(26): Schur options for right-hand side or solution vector","None",lu->id.ICNTL(26),&lu->id.ICNTL(26),PETSC_NULL);CHKERRQ(ierr); 738 ierr = PetscOptionsInt("-mat_mumps_icntl_27","ICNTL(27): experimental parameter","None",lu->id.ICNTL(27),&lu->id.ICNTL(27),PETSC_NULL);CHKERRQ(ierr); 739 740 ierr = PetscOptionsReal("-mat_mumps_cntl_1","CNTL(1): relative pivoting threshold","None",lu->id.CNTL(1),&lu->id.CNTL(1),PETSC_NULL);CHKERRQ(ierr); 741 ierr = PetscOptionsReal("-mat_mumps_cntl_2","CNTL(2): stopping criterion of refinement","None",lu->id.CNTL(2),&lu->id.CNTL(2),PETSC_NULL);CHKERRQ(ierr); 742 ierr = PetscOptionsReal("-mat_mumps_cntl_3","CNTL(3): absolute pivoting threshold","None",lu->id.CNTL(3),&lu->id.CNTL(3),PETSC_NULL);CHKERRQ(ierr); 743 ierr = PetscOptionsReal("-mat_mumps_cntl_4","CNTL(4): value for static pivoting","None",lu->id.CNTL(4),&lu->id.CNTL(4),PETSC_NULL);CHKERRQ(ierr); 744 ierr = PetscOptionsReal("-mat_mumps_cntl_5","CNTL(5): fixation for null pivots","None",lu->id.CNTL(5),&lu->id.CNTL(5),PETSC_NULL);CHKERRQ(ierr); 745 PetscOptionsEnd(); 746 PetscFunctionReturn(0); 747 } 748 749 #undef __FUNCT__ 750 #define __FUNCT__ "PetscInitializeMUMPS" 751 PetscErrorCode PetscInitializeMUMPS(Mat F) 752 { 753 Mat_MUMPS *lu = (Mat_MUMPS*)F->spptr; 754 PetscErrorCode ierr; 755 PetscInt icntl; 756 PetscTruth flg; 757 758 PetscFunctionBegin; 759 lu->id.job = JOB_INIT; 760 lu->id.par=1; /* host participates factorizaton and solve */ 761 lu->id.sym=lu->sym; 762 if (lu->sym == 2){ 763 ierr = PetscOptionsInt("-mat_mumps_sym","SYM: (1,2)","None",lu->id.sym,&icntl,&flg);CHKERRQ(ierr); 764 if (flg && icntl == 1) lu->id.sym=icntl; /* matrix is spd */ 765 } 766 #if defined(PETSC_USE_COMPLEX) 767 zmumps_c(&lu->id); 768 #else 769 dmumps_c(&lu->id); 770 #endif 771 PetscFunctionReturn(0); 772 } 773 774 /* Note the Petsc r and c permutations are ignored */ 775 #undef __FUNCT__ 776 #define __FUNCT__ "MatLUFactorSymbolic_AIJMUMPS" 777 PetscErrorCode MatLUFactorSymbolic_AIJMUMPS(Mat F,Mat A,IS r,IS c,const MatFactorInfo *info) 778 { 779 Mat_MUMPS *lu = (Mat_MUMPS*)F->spptr; 780 PetscErrorCode ierr; 781 MatReuse reuse; 782 Vec b; 783 IS is_iden; 784 const PetscInt M = A->rmap->N; 785 786 PetscFunctionBegin; 787 lu->sym = 0; 788 lu->matstruc = DIFFERENT_NONZERO_PATTERN; 789 790 ierr = MPI_Comm_rank(((PetscObject)A)->comm, &lu->myid); 791 ierr = MPI_Comm_size(((PetscObject)A)->comm,&lu->size);CHKERRQ(ierr); 792 ierr = MPI_Comm_dup(((PetscObject)A)->comm,&(lu->comm_mumps));CHKERRQ(ierr); 793 lu->id.comm_fortran = MPI_Comm_c2f(lu->comm_mumps); 794 795 /* Initialize a MUMPS instance */ 796 ierr = PetscInitializeMUMPS(F);CHKERRQ(ierr); 797 /* Set MUMPS options */ 798 ierr = PetscSetMUMPSOptions(F,A);CHKERRQ(ierr); 799 800 reuse = MAT_INITIAL_MATRIX; 801 ierr = (*lu->ConvertToTriples)(A, 1, reuse, &lu->nz, &lu->irn, &lu->jcn, &lu->val);CHKERRQ(ierr); 802 803 /* analysis phase */ 804 /*----------------*/ 805 lu->id.job = 1; 806 lu->id.n = M; 807 switch (lu->id.ICNTL(18)){ 808 case 0: /* centralized assembled matrix input */ 809 if (!lu->myid) { 810 lu->id.nz =lu->nz; lu->id.irn=lu->irn; lu->id.jcn=lu->jcn; 811 if (lu->id.ICNTL(6)>1){ 812 #if defined(PETSC_USE_COMPLEX) 813 lu->id.a = (mumps_double_complex*)lu->val; 814 #else 815 lu->id.a = lu->val; 816 #endif 817 } 818 } 819 break; 820 case 3: /* distributed assembled matrix input (size>1) */ 821 lu->id.nz_loc = lu->nz; 822 lu->id.irn_loc=lu->irn; lu->id.jcn_loc=lu->jcn; 823 if (lu->id.ICNTL(6)>1) { 824 #if defined(PETSC_USE_COMPLEX) 825 lu->id.a_loc = (mumps_double_complex*)lu->val; 826 #else 827 lu->id.a_loc = lu->val; 828 #endif 829 } 830 /* MUMPS only supports centralized rhs. Create scatter scat_rhs for repeated use in MatSolve() */ 831 if (!lu->myid){ 832 ierr = VecCreateSeq(PETSC_COMM_SELF,A->cmap->N,&lu->b_seq);CHKERRQ(ierr); 833 ierr = ISCreateStride(PETSC_COMM_SELF,A->cmap->N,0,1,&is_iden);CHKERRQ(ierr); 834 } else { 835 ierr = VecCreateSeq(PETSC_COMM_SELF,0,&lu->b_seq);CHKERRQ(ierr); 836 ierr = ISCreateStride(PETSC_COMM_SELF,0,0,1,&is_iden);CHKERRQ(ierr); 837 } 838 ierr = VecCreate(((PetscObject)A)->comm,&b);CHKERRQ(ierr); 839 ierr = VecSetSizes(b,A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr); 840 ierr = VecSetFromOptions(b);CHKERRQ(ierr); 841 842 ierr = VecScatterCreate(b,is_iden,lu->b_seq,is_iden,&lu->scat_rhs);CHKERRQ(ierr); 843 ierr = ISDestroy(is_iden);CHKERRQ(ierr); 844 ierr = VecDestroy(b);CHKERRQ(ierr); 845 break; 846 } 847 #if defined(PETSC_USE_COMPLEX) 848 zmumps_c(&lu->id); 849 #else 850 dmumps_c(&lu->id); 851 #endif 852 if (lu->id.INFOG(1) < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in analysis phase: INFOG(1)=%d\n",lu->id.INFOG(1)); 853 854 F->ops->lufactornumeric = MatFactorNumeric_MUMPS; 855 F->ops->solve = MatSolve_MUMPS; 856 PetscFunctionReturn(0); 857 } 858 859 /* Note the Petsc r and c permutations are ignored */ 860 #undef __FUNCT__ 861 #define __FUNCT__ "MatLUFactorSymbolic_BAIJMUMPS" 862 PetscErrorCode MatLUFactorSymbolic_BAIJMUMPS(Mat F,Mat A,IS r,IS c,const MatFactorInfo *info) 863 { 864 865 Mat_MUMPS *lu = (Mat_MUMPS*)F->spptr; 866 PetscErrorCode ierr; 867 MatReuse reuse; 868 Vec b; 869 IS is_iden; 870 const PetscInt M = A->rmap->N; 871 872 PetscFunctionBegin; 873 lu->sym = 0; 874 lu->matstruc = DIFFERENT_NONZERO_PATTERN; 875 ierr = MPI_Comm_rank(((PetscObject)A)->comm, &lu->myid); 876 ierr = MPI_Comm_size(((PetscObject)A)->comm,&lu->size);CHKERRQ(ierr); 877 ierr = MPI_Comm_dup(((PetscObject)A)->comm,&(lu->comm_mumps));CHKERRQ(ierr); 878 lu->id.comm_fortran = MPI_Comm_c2f(lu->comm_mumps); 879 880 /* Initialize a MUMPS instance */ 881 ierr = PetscInitializeMUMPS(F);CHKERRQ(ierr); 882 /* Set MUMPS options */ 883 ierr = PetscSetMUMPSOptions(F,A);CHKERRQ(ierr); 884 885 reuse = MAT_INITIAL_MATRIX; 886 ierr = (*lu->ConvertToTriples)(A, 1, reuse, &lu->nz, &lu->irn, &lu->jcn, &lu->val);CHKERRQ(ierr); 887 888 /* analysis phase */ 889 /*----------------*/ 890 lu->id.job = 1; 891 lu->id.n = M; 892 switch (lu->id.ICNTL(18)){ 893 case 0: /* centralized assembled matrix input */ 894 if (!lu->myid) { 895 lu->id.nz =lu->nz; lu->id.irn=lu->irn; lu->id.jcn=lu->jcn; 896 if (lu->id.ICNTL(6)>1){ 897 #if defined(PETSC_USE_COMPLEX) 898 lu->id.a = (mumps_double_complex*)lu->val; 899 #else 900 lu->id.a = lu->val; 901 #endif 902 } 903 } 904 break; 905 case 3: /* distributed assembled matrix input (size>1) */ 906 lu->id.nz_loc = lu->nz; 907 lu->id.irn_loc=lu->irn; lu->id.jcn_loc=lu->jcn; 908 if (lu->id.ICNTL(6)>1) { 909 #if defined(PETSC_USE_COMPLEX) 910 lu->id.a_loc = (mumps_double_complex*)lu->val; 911 #else 912 lu->id.a_loc = lu->val; 913 #endif 914 } 915 /* MUMPS only supports centralized rhs. Create scatter scat_rhs for repeated use in MatSolve() */ 916 if (!lu->myid){ 917 ierr = VecCreateSeq(PETSC_COMM_SELF,A->cmap->N,&lu->b_seq);CHKERRQ(ierr); 918 ierr = ISCreateStride(PETSC_COMM_SELF,A->cmap->N,0,1,&is_iden);CHKERRQ(ierr); 919 } else { 920 ierr = VecCreateSeq(PETSC_COMM_SELF,0,&lu->b_seq);CHKERRQ(ierr); 921 ierr = ISCreateStride(PETSC_COMM_SELF,0,0,1,&is_iden);CHKERRQ(ierr); 922 } 923 ierr = VecCreate(((PetscObject)A)->comm,&b);CHKERRQ(ierr); 924 ierr = VecSetSizes(b,A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr); 925 ierr = VecSetFromOptions(b);CHKERRQ(ierr); 926 927 ierr = VecScatterCreate(b,is_iden,lu->b_seq,is_iden,&lu->scat_rhs);CHKERRQ(ierr); 928 ierr = ISDestroy(is_iden);CHKERRQ(ierr); 929 ierr = VecDestroy(b);CHKERRQ(ierr); 930 break; 931 } 932 #if defined(PETSC_USE_COMPLEX) 933 zmumps_c(&lu->id); 934 #else 935 dmumps_c(&lu->id); 936 #endif 937 if (lu->id.INFOG(1) < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in analysis phase: INFOG(1)=%d\n",lu->id.INFOG(1)); 938 939 F->ops->lufactornumeric = MatFactorNumeric_MUMPS; 940 F->ops->solve = MatSolve_MUMPS; 941 PetscFunctionReturn(0); 942 } 943 944 /* Note the Petsc r permutation is ignored */ 945 #undef __FUNCT__ 946 #define __FUNCT__ "MatCholeskyFactorSymbolic_MUMPS" 947 PetscErrorCode MatCholeskyFactorSymbolic_MUMPS(Mat F,Mat A,IS r,const MatFactorInfo *info) 948 { 949 Mat_MUMPS *lu = (Mat_MUMPS*)F->spptr; 950 PetscErrorCode ierr; 951 MatReuse reuse; 952 Vec b; 953 IS is_iden; 954 const PetscInt M = A->rmap->N; 955 956 PetscFunctionBegin; 957 lu->sym = 2; 958 lu->matstruc = DIFFERENT_NONZERO_PATTERN; 959 ierr = MPI_Comm_rank(((PetscObject)A)->comm, &lu->myid); 960 ierr = MPI_Comm_size(((PetscObject)A)->comm,&lu->size);CHKERRQ(ierr); 961 ierr = MPI_Comm_dup(((PetscObject)A)->comm,&(lu->comm_mumps));CHKERRQ(ierr); 962 lu->id.comm_fortran = MPI_Comm_c2f(lu->comm_mumps); 963 964 /* Initialize a MUMPS instance */ 965 ierr = PetscInitializeMUMPS(F);CHKERRQ(ierr); 966 /* Set MUMPS options */ 967 ierr = PetscSetMUMPSOptions(F,A);CHKERRQ(ierr); 968 969 reuse = MAT_INITIAL_MATRIX; 970 ierr = (*lu->ConvertToTriples)(A, 1 , reuse, &lu->nz, &lu->irn, &lu->jcn, &lu->val);CHKERRQ(ierr); 971 972 /* analysis phase */ 973 /*----------------*/ 974 lu->id.job = 1; 975 lu->id.n = M; 976 switch (lu->id.ICNTL(18)){ 977 case 0: /* centralized assembled matrix input */ 978 if (!lu->myid) { 979 lu->id.nz =lu->nz; lu->id.irn=lu->irn; lu->id.jcn=lu->jcn; 980 if (lu->id.ICNTL(6)>1){ 981 #if defined(PETSC_USE_COMPLEX) 982 lu->id.a = (mumps_double_complex*)lu->val; 983 #else 984 lu->id.a = lu->val; 985 #endif 986 } 987 } 988 break; 989 case 3: /* distributed assembled matrix input (size>1) */ 990 lu->id.nz_loc = lu->nz; 991 lu->id.irn_loc=lu->irn; lu->id.jcn_loc=lu->jcn; 992 if (lu->id.ICNTL(6)>1) { 993 #if defined(PETSC_USE_COMPLEX) 994 lu->id.a_loc = (mumps_double_complex*)lu->val; 995 #else 996 lu->id.a_loc = lu->val; 997 #endif 998 } 999 /* MUMPS only supports centralized rhs. Create scatter scat_rhs for repeated use in MatSolve() */ 1000 if (!lu->myid){ 1001 ierr = VecCreateSeq(PETSC_COMM_SELF,A->cmap->N,&lu->b_seq);CHKERRQ(ierr); 1002 ierr = ISCreateStride(PETSC_COMM_SELF,A->cmap->N,0,1,&is_iden);CHKERRQ(ierr); 1003 } else { 1004 ierr = VecCreateSeq(PETSC_COMM_SELF,0,&lu->b_seq);CHKERRQ(ierr); 1005 ierr = ISCreateStride(PETSC_COMM_SELF,0,0,1,&is_iden);CHKERRQ(ierr); 1006 } 1007 ierr = VecCreate(((PetscObject)A)->comm,&b);CHKERRQ(ierr); 1008 ierr = VecSetSizes(b,A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr); 1009 ierr = VecSetFromOptions(b);CHKERRQ(ierr); 1010 1011 ierr = VecScatterCreate(b,is_iden,lu->b_seq,is_iden,&lu->scat_rhs);CHKERRQ(ierr); 1012 ierr = ISDestroy(is_iden);CHKERRQ(ierr); 1013 ierr = VecDestroy(b);CHKERRQ(ierr); 1014 break; 1015 } 1016 #if defined(PETSC_USE_COMPLEX) 1017 zmumps_c(&lu->id); 1018 #else 1019 dmumps_c(&lu->id); 1020 #endif 1021 if (lu->id.INFOG(1) < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_LIB,"Error reported by MUMPS in analysis phase: INFOG(1)=%d\n",lu->id.INFOG(1)); 1022 1023 F->ops->choleskyfactornumeric = MatFactorNumeric_MUMPS; 1024 F->ops->solve = MatSolve_MUMPS; 1025 #if !defined(PETSC_USE_COMPLEX) 1026 (F)->ops->getinertia = MatGetInertia_SBAIJMUMPS; 1027 #endif 1028 PetscFunctionReturn(0); 1029 } 1030 1031 #undef __FUNCT__ 1032 #define __FUNCT__ "MatFactorInfo_MUMPS" 1033 PetscErrorCode MatFactorInfo_MUMPS(Mat A,PetscViewer viewer) 1034 { 1035 Mat_MUMPS *lu=(Mat_MUMPS*)A->spptr; 1036 PetscErrorCode ierr; 1037 1038 PetscFunctionBegin; 1039 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(1) (output for error): %d \n",lu->id.ICNTL(1));CHKERRQ(ierr); 1040 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(2) (output of diagnostic msg):%d \n",lu->id.ICNTL(2));CHKERRQ(ierr); 1041 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(3) (output for global info): %d \n",lu->id.ICNTL(3));CHKERRQ(ierr); 1042 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(4) (level of printing): %d \n",lu->id.ICNTL(4));CHKERRQ(ierr); 1043 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(5) (input mat struct): %d \n",lu->id.ICNTL(5));CHKERRQ(ierr); 1044 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(6) (matrix prescaling): %d \n",lu->id.ICNTL(6));CHKERRQ(ierr); 1045 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(7) (matrix ordering): %d \n",lu->id.ICNTL(7));CHKERRQ(ierr); 1046 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(8) (scalling strategy): %d \n",lu->id.ICNTL(8));CHKERRQ(ierr); 1047 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(9) (A/A^T x=b is solved): %d \n",lu->id.ICNTL(9));CHKERRQ(ierr); 1048 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(10) (max num of refinements): %d \n",lu->id.ICNTL(10));CHKERRQ(ierr); 1049 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(11) (error analysis): %d \n",lu->id.ICNTL(11));CHKERRQ(ierr); 1050 if (lu->id.ICNTL(11)>0) { 1051 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(4) (inf norm of input mat): %g\n",lu->id.RINFOG(4));CHKERRQ(ierr); 1052 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(5) (inf norm of solution): %g\n",lu->id.RINFOG(5));CHKERRQ(ierr); 1053 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(6) (inf norm of residual): %g\n",lu->id.RINFOG(6));CHKERRQ(ierr); 1054 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(7),RINFOG(8) (backward error est): %g, %g\n",lu->id.RINFOG(7),lu->id.RINFOG(8));CHKERRQ(ierr); 1055 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(9) (error estimate): %g \n",lu->id.RINFOG(9));CHKERRQ(ierr); 1056 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(10),RINFOG(11)(condition numbers): %g, %g\n",lu->id.RINFOG(10),lu->id.RINFOG(11));CHKERRQ(ierr); 1057 1058 } 1059 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(12) (efficiency control): %d \n",lu->id.ICNTL(12));CHKERRQ(ierr); 1060 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(13) (efficiency control): %d \n",lu->id.ICNTL(13));CHKERRQ(ierr); 1061 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(14) (percentage of estimated workspace increase): %d \n",lu->id.ICNTL(14));CHKERRQ(ierr); 1062 /* ICNTL(15-17) not used */ 1063 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(18) (input mat struct): %d \n",lu->id.ICNTL(18));CHKERRQ(ierr); 1064 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(19) (Shur complement info): %d \n",lu->id.ICNTL(19));CHKERRQ(ierr); 1065 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(20) (rhs sparse pattern): %d \n",lu->id.ICNTL(20));CHKERRQ(ierr); 1066 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(21) (solution struct): %d \n",lu->id.ICNTL(21));CHKERRQ(ierr); 1067 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(22) (in-core/out-of-core facility): %d \n",lu->id.ICNTL(22));CHKERRQ(ierr); 1068 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(23) (max size of memory can be allocated locally):%d \n",lu->id.ICNTL(23));CHKERRQ(ierr); 1069 1070 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(24) (detection of null pivot rows): %d \n",lu->id.ICNTL(24));CHKERRQ(ierr); 1071 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(25) (computation of a null space basis): %d \n",lu->id.ICNTL(25));CHKERRQ(ierr); 1072 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(26) (Schur options for rhs or solution): %d \n",lu->id.ICNTL(26));CHKERRQ(ierr); 1073 ierr = PetscViewerASCIIPrintf(viewer," ICNTL(27) (experimental parameter): %d \n",lu->id.ICNTL(27));CHKERRQ(ierr); 1074 1075 ierr = PetscViewerASCIIPrintf(viewer," CNTL(1) (relative pivoting threshold): %g \n",lu->id.CNTL(1));CHKERRQ(ierr); 1076 ierr = PetscViewerASCIIPrintf(viewer," CNTL(2) (stopping criterion of refinement): %g \n",lu->id.CNTL(2));CHKERRQ(ierr); 1077 ierr = PetscViewerASCIIPrintf(viewer," CNTL(3) (absolute pivoting threshold): %g \n",lu->id.CNTL(3));CHKERRQ(ierr); 1078 ierr = PetscViewerASCIIPrintf(viewer," CNTL(4) (value of static pivoting): %g \n",lu->id.CNTL(4));CHKERRQ(ierr); 1079 ierr = PetscViewerASCIIPrintf(viewer," CNTL(5) (fixation for null pivots): %g \n",lu->id.CNTL(5));CHKERRQ(ierr); 1080 1081 /* infomation local to each processor */ 1082 ierr = PetscViewerASCIIPrintf(viewer, " RINFO(1) (local estimated flops for the elimination after analysis): \n");CHKERRQ(ierr); 1083 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %g \n",lu->myid,lu->id.RINFO(1));CHKERRQ(ierr); 1084 ierr = PetscViewerFlush(viewer); 1085 ierr = PetscViewerASCIIPrintf(viewer, " RINFO(2) (local estimated flops for the assembly after factorization): \n");CHKERRQ(ierr); 1086 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %g \n",lu->myid,lu->id.RINFO(2));CHKERRQ(ierr); 1087 ierr = PetscViewerFlush(viewer); 1088 ierr = PetscViewerASCIIPrintf(viewer, " RINFO(3) (local estimated flops for the elimination after factorization): \n");CHKERRQ(ierr); 1089 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %g \n",lu->myid,lu->id.RINFO(3));CHKERRQ(ierr); 1090 ierr = PetscViewerFlush(viewer); 1091 1092 ierr = PetscViewerASCIIPrintf(viewer, " INFO(15) (estimated size of (in MB) MUMPS internal data for running numerical factorization): \n");CHKERRQ(ierr); 1093 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %d \n",lu->myid,lu->id.INFO(15));CHKERRQ(ierr); 1094 ierr = PetscViewerFlush(viewer); 1095 1096 ierr = PetscViewerASCIIPrintf(viewer, " INFO(16) (size of (in MB) MUMPS internal data used during numerical factorization): \n");CHKERRQ(ierr); 1097 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %d \n",lu->myid,lu->id.INFO(16));CHKERRQ(ierr); 1098 ierr = PetscViewerFlush(viewer); 1099 1100 ierr = PetscViewerASCIIPrintf(viewer, " INFO(23) (num of pivots eliminated on this processor after factorization): \n");CHKERRQ(ierr); 1101 ierr = PetscViewerASCIISynchronizedPrintf(viewer," [%d] %d \n",lu->myid,lu->id.INFO(23));CHKERRQ(ierr); 1102 ierr = PetscViewerFlush(viewer); 1103 1104 if (!lu->myid){ /* information from the host */ 1105 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(1) (global estimated flops for the elimination after analysis): %g \n",lu->id.RINFOG(1));CHKERRQ(ierr); 1106 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(2) (global estimated flops for the assembly after factorization): %g \n",lu->id.RINFOG(2));CHKERRQ(ierr); 1107 ierr = PetscViewerASCIIPrintf(viewer," RINFOG(3) (global estimated flops for the elimination after factorization): %g \n",lu->id.RINFOG(3));CHKERRQ(ierr); 1108 1109 ierr = PetscViewerASCIIPrintf(viewer," INFOG(3) (estimated real workspace for factors on all processors after analysis): %d \n",lu->id.INFOG(3));CHKERRQ(ierr); 1110 ierr = PetscViewerASCIIPrintf(viewer," INFOG(4) (estimated integer workspace for factors on all processors after analysis): %d \n",lu->id.INFOG(4));CHKERRQ(ierr); 1111 ierr = PetscViewerASCIIPrintf(viewer," INFOG(5) (estimated maximum front size in the complete tree): %d \n",lu->id.INFOG(5));CHKERRQ(ierr); 1112 ierr = PetscViewerASCIIPrintf(viewer," INFOG(6) (number of nodes in the complete tree): %d \n",lu->id.INFOG(6));CHKERRQ(ierr); 1113 ierr = PetscViewerASCIIPrintf(viewer," INFOG(7) (ordering option effectively uese after analysis): %d \n",lu->id.INFOG(7));CHKERRQ(ierr); 1114 ierr = PetscViewerASCIIPrintf(viewer," INFOG(8) (structural symmetry in percent of the permuted matrix after analysis): %d \n",lu->id.INFOG(8));CHKERRQ(ierr); 1115 ierr = PetscViewerASCIIPrintf(viewer," INFOG(9) (total real/complex workspace to store the matrix factors after factorization): %d \n",lu->id.INFOG(9));CHKERRQ(ierr); 1116 ierr = PetscViewerASCIIPrintf(viewer," INFOG(10) (total integer space store the matrix factors after factorization): %d \n",lu->id.INFOG(10));CHKERRQ(ierr); 1117 ierr = PetscViewerASCIIPrintf(viewer," INFOG(11) (order of largest frontal matrix after factorization): %d \n",lu->id.INFOG(11));CHKERRQ(ierr); 1118 ierr = PetscViewerASCIIPrintf(viewer," INFOG(12) (number of off-diagonal pivots): %d \n",lu->id.INFOG(12));CHKERRQ(ierr); 1119 ierr = PetscViewerASCIIPrintf(viewer," INFOG(13) (number of delayed pivots after factorization): %d \n",lu->id.INFOG(13));CHKERRQ(ierr); 1120 ierr = PetscViewerASCIIPrintf(viewer," INFOG(14) (number of memory compress after factorization): %d \n",lu->id.INFOG(14));CHKERRQ(ierr); 1121 ierr = PetscViewerASCIIPrintf(viewer," INFOG(15) (number of steps of iterative refinement after solution): %d \n",lu->id.INFOG(15));CHKERRQ(ierr); 1122 ierr = PetscViewerASCIIPrintf(viewer," INFOG(16) (estimated size (in MB) of all MUMPS internal data for factorization after analysis: value on the most memory consuming processor): %d \n",lu->id.INFOG(16));CHKERRQ(ierr); 1123 ierr = PetscViewerASCIIPrintf(viewer," INFOG(17) (estimated size of all MUMPS internal data for factorization after analysis: sum over all processors): %d \n",lu->id.INFOG(17));CHKERRQ(ierr); 1124 ierr = PetscViewerASCIIPrintf(viewer," INFOG(18) (size of all MUMPS internal data allocated during factorization: value on the most memory consuming processor): %d \n",lu->id.INFOG(18));CHKERRQ(ierr); 1125 ierr = PetscViewerASCIIPrintf(viewer," INFOG(19) (size of all MUMPS internal data allocated during factorization: sum over all processors): %d \n",lu->id.INFOG(19));CHKERRQ(ierr); 1126 ierr = PetscViewerASCIIPrintf(viewer," INFOG(20) (estimated number of entries in the factors): %d \n",lu->id.INFOG(20));CHKERRQ(ierr); 1127 ierr = PetscViewerASCIIPrintf(viewer," INFOG(21) (size in MB of memory effectively used during factorization - value on the most memory consuming processor): %d \n",lu->id.INFOG(21));CHKERRQ(ierr); 1128 ierr = PetscViewerASCIIPrintf(viewer," INFOG(22) (size in MB of memory effectively used during factorization - sum over all processors): %d \n",lu->id.INFOG(22));CHKERRQ(ierr); 1129 ierr = PetscViewerASCIIPrintf(viewer," INFOG(23) (after analysis: value of ICNTL(6) effectively used): %d \n",lu->id.INFOG(23));CHKERRQ(ierr); 1130 ierr = PetscViewerASCIIPrintf(viewer," INFOG(24) (after analysis: value of ICNTL(12) effectively used): %d \n",lu->id.INFOG(24));CHKERRQ(ierr); 1131 ierr = PetscViewerASCIIPrintf(viewer," INFOG(25) (after factorization: number of pivots modified by static pivoting): %d \n",lu->id.INFOG(25));CHKERRQ(ierr); 1132 } 1133 PetscFunctionReturn(0); 1134 } 1135 1136 #undef __FUNCT__ 1137 #define __FUNCT__ "MatView_MUMPS" 1138 PetscErrorCode MatView_MUMPS(Mat A,PetscViewer viewer) 1139 { 1140 PetscErrorCode ierr; 1141 PetscTruth iascii; 1142 PetscViewerFormat format; 1143 Mat_MUMPS *mumps=(Mat_MUMPS*)A->spptr; 1144 1145 PetscFunctionBegin; 1146 /* check if matrix is mumps type */ 1147 if (A->ops->solve != MatSolve_MUMPS) PetscFunctionReturn(0); 1148 1149 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 1150 if (iascii) { 1151 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 1152 if (format == PETSC_VIEWER_ASCII_INFO || mumps->mumpsview){ 1153 ierr = PetscViewerASCIIPrintf(viewer,"MUMPS run parameters:\n");CHKERRQ(ierr); 1154 ierr = PetscViewerASCIIPrintf(viewer," SYM (matrix type): %d \n",mumps->id.sym);CHKERRQ(ierr); 1155 ierr = PetscViewerASCIIPrintf(viewer," PAR (host participation): %d \n",mumps->id.par);CHKERRQ(ierr); 1156 if (mumps->mumpsview){ /* View all MUMPS parameters */ 1157 ierr = MatFactorInfo_MUMPS(A,viewer);CHKERRQ(ierr); 1158 } 1159 } 1160 } 1161 PetscFunctionReturn(0); 1162 } 1163 1164 #undef __FUNCT__ 1165 #define __FUNCT__ "MatGetInfo_MUMPS" 1166 PetscErrorCode MatGetInfo_MUMPS(Mat A,MatInfoType flag,MatInfo *info) 1167 { 1168 Mat_MUMPS *mumps =(Mat_MUMPS*)A->spptr; 1169 1170 PetscFunctionBegin; 1171 info->block_size = 1.0; 1172 info->nz_allocated = mumps->id.INFOG(20); 1173 info->nz_used = mumps->id.INFOG(20); 1174 info->nz_unneeded = 0.0; 1175 info->assemblies = 0.0; 1176 info->mallocs = 0.0; 1177 info->memory = 0.0; 1178 info->fill_ratio_given = 0; 1179 info->fill_ratio_needed = 0; 1180 info->factor_mallocs = 0; 1181 PetscFunctionReturn(0); 1182 } 1183 1184 /*MC 1185 MATSOLVERMUMPS - A matrix type providing direct solvers (LU and Cholesky) for 1186 distributed and sequential matrices via the external package MUMPS. 1187 1188 Works with MATAIJ and MATSBAIJ matrices 1189 1190 Options Database Keys: 1191 + -mat_mumps_sym <0,1,2> - 0 the matrix is unsymmetric, 1 symmetric positive definite, 2 symmetric 1192 . -mat_mumps_icntl_4 <0,...,4> - print level 1193 . -mat_mumps_icntl_6 <0,...,7> - matrix prescaling options (see MUMPS User's Guide) 1194 . -mat_mumps_icntl_7 <0,...,7> - matrix orderings (see MUMPS User's Guide) 1195 . -mat_mumps_icntl_9 <1,2> - A or A^T x=b to be solved: 1 denotes A, 2 denotes A^T 1196 . -mat_mumps_icntl_10 <n> - maximum number of iterative refinements 1197 . -mat_mumps_icntl_11 <n> - error analysis, a positive value returns statistics during -ksp_view 1198 . -mat_mumps_icntl_12 <n> - efficiency control (see MUMPS User's Guide) 1199 . -mat_mumps_icntl_13 <n> - efficiency control (see MUMPS User's Guide) 1200 . -mat_mumps_icntl_14 <n> - efficiency control (see MUMPS User's Guide) 1201 . -mat_mumps_icntl_15 <n> - efficiency control (see MUMPS User's Guide) 1202 . -mat_mumps_cntl_1 <delta> - relative pivoting threshold 1203 . -mat_mumps_cntl_2 <tol> - stopping criterion for refinement 1204 - -mat_mumps_cntl_3 <adelta> - absolute pivoting threshold 1205 1206 Level: beginner 1207 1208 .seealso: PCFactorSetMatSolverPackage(), MatSolverPackage 1209 1210 M*/ 1211 1212 EXTERN_C_BEGIN 1213 #undef __FUNCT__ 1214 #define __FUNCT__ "MatFactorGetSolverPackage_mumps" 1215 PetscErrorCode MatFactorGetSolverPackage_mumps(Mat A,const MatSolverPackage *type) 1216 { 1217 PetscFunctionBegin; 1218 *type = MATSOLVERMUMPS; 1219 PetscFunctionReturn(0); 1220 } 1221 EXTERN_C_END 1222 1223 EXTERN_C_BEGIN 1224 /* MatGetFactor for Seq and MPI AIJ matrices */ 1225 #undef __FUNCT__ 1226 #define __FUNCT__ "MatGetFactor_aij_mumps" 1227 PetscErrorCode MatGetFactor_aij_mumps(Mat A,MatFactorType ftype,Mat *F) 1228 { 1229 Mat B; 1230 PetscErrorCode ierr; 1231 Mat_MUMPS *mumps; 1232 PetscTruth isSeqAIJ; 1233 1234 PetscFunctionBegin; 1235 /* Create the factorization matrix */ 1236 ierr = PetscTypeCompare((PetscObject)A,MATSEQAIJ,&isSeqAIJ);CHKERRQ(ierr); 1237 ierr = MatCreate(((PetscObject)A)->comm,&B);CHKERRQ(ierr); 1238 ierr = MatSetSizes(B,A->rmap->n,A->cmap->n,A->rmap->N,A->cmap->N);CHKERRQ(ierr); 1239 ierr = MatSetType(B,((PetscObject)A)->type_name);CHKERRQ(ierr); 1240 if (isSeqAIJ) { 1241 ierr = MatSeqAIJSetPreallocation(B,0,PETSC_NULL);CHKERRQ(ierr); 1242 } else { 1243 ierr = MatMPIAIJSetPreallocation(B,0,PETSC_NULL,0,PETSC_NULL);CHKERRQ(ierr); 1244 } 1245 1246 ierr = PetscNewLog(B,Mat_MUMPS,&mumps);CHKERRQ(ierr); 1247 B->ops->view = MatView_MUMPS; 1248 B->ops->getinfo = MatGetInfo_MUMPS; 1249 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatFactorGetSolverPackage_C","MatFactorGetSolverPackage_mumps",MatFactorGetSolverPackage_mumps);CHKERRQ(ierr); 1250 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMumpsSetIcntl_C","MatMumpsSetIcntl",MatMumpsSetIcntl);CHKERRQ(ierr); 1251 if (ftype == MAT_FACTOR_LU) { 1252 B->ops->lufactorsymbolic = MatLUFactorSymbolic_AIJMUMPS; 1253 B->factortype = MAT_FACTOR_LU; 1254 if (isSeqAIJ) mumps->ConvertToTriples = MatConvertToTriples_seqaij_seqaij; 1255 else mumps->ConvertToTriples = MatConvertToTriples_mpiaij_mpiaij; 1256 } else { 1257 B->ops->choleskyfactorsymbolic = MatCholeskyFactorSymbolic_MUMPS; 1258 B->factortype = MAT_FACTOR_CHOLESKY; 1259 if (isSeqAIJ) mumps->ConvertToTriples = MatConvertToTriples_seqaij_seqsbaij; 1260 else mumps->ConvertToTriples = MatConvertToTriples_mpiaij_mpisbaij; 1261 } 1262 1263 mumps->CleanUpMUMPS = PETSC_FALSE; 1264 mumps->isAIJ = PETSC_TRUE; 1265 mumps->scat_rhs = PETSC_NULL; 1266 mumps->scat_sol = PETSC_NULL; 1267 mumps->nSolve = 0; 1268 mumps->MatDestroy = B->ops->destroy; 1269 B->ops->destroy = MatDestroy_MUMPS; 1270 B->spptr = (void*)mumps; 1271 1272 *F = B; 1273 PetscFunctionReturn(0); 1274 } 1275 EXTERN_C_END 1276 1277 1278 EXTERN_C_BEGIN 1279 /* MatGetFactor for Seq and MPI SBAIJ matrices */ 1280 #undef __FUNCT__ 1281 #define __FUNCT__ "MatGetFactor_sbaij_mumps" 1282 PetscErrorCode MatGetFactor_sbaij_mumps(Mat A,MatFactorType ftype,Mat *F) 1283 { 1284 Mat B; 1285 PetscErrorCode ierr; 1286 Mat_MUMPS *mumps; 1287 PetscTruth isSeqSBAIJ; 1288 1289 PetscFunctionBegin; 1290 if (ftype != MAT_FACTOR_CHOLESKY) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_SUP,"Cannot use PETSc SBAIJ matrices with MUMPS LU, use AIJ matrix"); 1291 if(A->rmap->bs > 1) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_SUP,"Cannot use PETSc SBAIJ matrices with block size > 1 with MUMPS Cholesky, use AIJ matrix instead"); 1292 ierr = PetscTypeCompare((PetscObject)A,MATSEQSBAIJ,&isSeqSBAIJ);CHKERRQ(ierr); 1293 /* Create the factorization matrix */ 1294 ierr = MatCreate(((PetscObject)A)->comm,&B);CHKERRQ(ierr); 1295 ierr = MatSetSizes(B,A->rmap->n,A->cmap->n,A->rmap->N,A->cmap->N);CHKERRQ(ierr); 1296 ierr = MatSetType(B,((PetscObject)A)->type_name);CHKERRQ(ierr); 1297 ierr = PetscNewLog(B,Mat_MUMPS,&mumps);CHKERRQ(ierr); 1298 if (isSeqSBAIJ) { 1299 ierr = MatSeqSBAIJSetPreallocation(B,1,0,PETSC_NULL);CHKERRQ(ierr); 1300 mumps->ConvertToTriples = MatConvertToTriples_seqsbaij_seqsbaij; 1301 } else { 1302 ierr = MatMPISBAIJSetPreallocation(B,1,0,PETSC_NULL,0,PETSC_NULL);CHKERRQ(ierr); 1303 mumps->ConvertToTriples = MatConvertToTriples_mpisbaij_mpisbaij; 1304 } 1305 1306 B->ops->choleskyfactorsymbolic = MatCholeskyFactorSymbolic_MUMPS; 1307 B->ops->view = MatView_MUMPS; 1308 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatFactorGetSolverPackage_C","MatFactorGetSolverPackage_mumps",MatFactorGetSolverPackage_mumps);CHKERRQ(ierr); 1309 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMumpsSetIcntl_C","MatMumpsSetIcntl",MatMumpsSetIcntl);CHKERRQ(ierr); 1310 B->factortype = MAT_FACTOR_CHOLESKY; 1311 1312 mumps->CleanUpMUMPS = PETSC_FALSE; 1313 mumps->isAIJ = PETSC_FALSE; 1314 mumps->scat_rhs = PETSC_NULL; 1315 mumps->scat_sol = PETSC_NULL; 1316 mumps->nSolve = 0; 1317 mumps->MatDestroy = B->ops->destroy; 1318 B->ops->destroy = MatDestroy_MUMPS; 1319 B->spptr = (void*)mumps; 1320 1321 *F = B; 1322 PetscFunctionReturn(0); 1323 } 1324 EXTERN_C_END 1325 1326 EXTERN_C_BEGIN 1327 #undef __FUNCT__ 1328 #define __FUNCT__ "MatGetFactor_baij_mumps" 1329 PetscErrorCode MatGetFactor_baij_mumps(Mat A,MatFactorType ftype,Mat *F) 1330 { 1331 Mat B; 1332 PetscErrorCode ierr; 1333 Mat_MUMPS *mumps; 1334 PetscTruth isSeqBAIJ; 1335 1336 PetscFunctionBegin; 1337 /* Create the factorization matrix */ 1338 ierr = PetscTypeCompare((PetscObject)A,MATSEQBAIJ,&isSeqBAIJ);CHKERRQ(ierr); 1339 ierr = MatCreate(((PetscObject)A)->comm,&B);CHKERRQ(ierr); 1340 ierr = MatSetSizes(B,A->rmap->n,A->cmap->n,A->rmap->N,A->cmap->N);CHKERRQ(ierr); 1341 ierr = MatSetType(B,((PetscObject)A)->type_name);CHKERRQ(ierr); 1342 if (isSeqBAIJ) { 1343 ierr = MatSeqBAIJSetPreallocation(B,A->rmap->bs,0,PETSC_NULL);CHKERRQ(ierr); 1344 } else { 1345 ierr = MatMPIBAIJSetPreallocation(B,A->rmap->bs,0,PETSC_NULL,0,PETSC_NULL);CHKERRQ(ierr); 1346 } 1347 1348 ierr = PetscNewLog(B,Mat_MUMPS,&mumps);CHKERRQ(ierr); 1349 if (ftype == MAT_FACTOR_LU) { 1350 B->ops->lufactorsymbolic = MatLUFactorSymbolic_BAIJMUMPS; 1351 B->factortype = MAT_FACTOR_LU; 1352 if (isSeqBAIJ) mumps->ConvertToTriples = MatConvertToTriples_seqbaij_seqaij; 1353 else mumps->ConvertToTriples = MatConvertToTriples_mpibaij_mpiaij; 1354 } 1355 else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Cannot use PETSc BAIJ matrices with MUMPS Cholesky, use SBAIJ or AIJ matrix instead\n"); 1356 1357 B->ops->view = MatView_MUMPS; 1358 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatFactorGetSolverPackage_C","MatFactorGetSolverPackage_mumps",MatFactorGetSolverPackage_mumps);CHKERRQ(ierr); 1359 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMumpsSetIcntl_C","MatMumpsSetIcntl",MatMumpsSetIcntl);CHKERRQ(ierr); 1360 1361 mumps->CleanUpMUMPS = PETSC_FALSE; 1362 mumps->isAIJ = PETSC_TRUE; 1363 mumps->scat_rhs = PETSC_NULL; 1364 mumps->scat_sol = PETSC_NULL; 1365 mumps->nSolve = 0; 1366 mumps->MatDestroy = B->ops->destroy; 1367 B->ops->destroy = MatDestroy_MUMPS; 1368 B->spptr = (void*)mumps; 1369 1370 *F = B; 1371 PetscFunctionReturn(0); 1372 } 1373 EXTERN_C_END 1374 1375 /* -------------------------------------------------------------------------------------------*/ 1376 /*@ 1377 MatMumpsSetIcntl - Set MUMPS parameter ICNTL() 1378 1379 Collective on Mat 1380 1381 Input Parameters: 1382 + F - the factored matrix obtained by calling MatGetFactor() from PETSc-MUMPS interface 1383 . idx - index of MUMPS parameter array ICNTL() 1384 - icntl - value of MUMPS ICNTL(imumps) 1385 1386 Options Database: 1387 . -mat_mumps_icntl_<idx> <icntl> 1388 1389 Level: beginner 1390 1391 References: MUMPS Users' Guide 1392 1393 .seealso: MatGetFactor() 1394 @*/ 1395 #undef __FUNCT__ 1396 #define __FUNCT__ "MatMumpsSetIcntl" 1397 PetscErrorCode MatMumpsSetIcntl(Mat F,PetscInt idx,PetscInt icntl) 1398 { 1399 Mat_MUMPS *lu =(Mat_MUMPS*)(F)->spptr; 1400 1401 PetscFunctionBegin; 1402 lu->id.ICNTL(idx) = icntl; 1403 PetscFunctionReturn(0); 1404 } 1405 1406