1 #define PETSCMAT_DLL 2 3 #include "src/mat/impls/baij/seq/baij.h" 4 #include "src/mat/impls/sbaij/seq/sbaij.h" 5 #include "src/inline/ilu.h" 6 #include "include/petscis.h" 7 8 #if !defined(PETSC_USE_COMPLEX) 9 /* 10 input: 11 F -- numeric factor 12 output: 13 nneg, nzero, npos: matrix inertia 14 */ 15 16 #undef __FUNCT__ 17 #define __FUNCT__ "MatGetInertia_SeqSBAIJ" 18 PetscErrorCode MatGetInertia_SeqSBAIJ(Mat F,PetscInt *nneig,PetscInt *nzero,PetscInt *npos) 19 { 20 Mat_SeqSBAIJ *fact_ptr = (Mat_SeqSBAIJ*)F->data; 21 MatScalar *dd = fact_ptr->a; 22 PetscInt mbs=fact_ptr->mbs,bs=F->rmap.bs,i,nneig_tmp,npos_tmp,*fi = fact_ptr->i; 23 24 PetscFunctionBegin; 25 if (bs != 1) SETERRQ1(PETSC_ERR_SUP,"No support for bs: %D >1 yet",bs); 26 nneig_tmp = 0; npos_tmp = 0; 27 for (i=0; i<mbs; i++){ 28 if (PetscRealPart(dd[*fi]) > 0.0){ 29 npos_tmp++; 30 } else if (PetscRealPart(dd[*fi]) < 0.0){ 31 nneig_tmp++; 32 } 33 fi++; 34 } 35 if (nneig) *nneig = nneig_tmp; 36 if (npos) *npos = npos_tmp; 37 if (nzero) *nzero = mbs - nneig_tmp - npos_tmp; 38 39 PetscFunctionReturn(0); 40 } 41 #endif /* !defined(PETSC_USE_COMPLEX) */ 42 43 /* 44 Symbolic U^T*D*U factorization for SBAIJ format. Modified from SSF of YSMP. 45 Use Modified Sparse Row (MSR) storage for u and ju. See page 85, "Iterative Methods ..." by Saad. 46 */ 47 #undef __FUNCT__ 48 #define __FUNCT__ "MatCholeskyFactorSymbolic_SeqSBAIJ_MSR" 49 PetscErrorCode MatCholeskyFactorSymbolic_SeqSBAIJ_MSR(Mat A,IS perm,MatFactorInfo *info,Mat *B) 50 { 51 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data,*b; 52 PetscErrorCode ierr; 53 PetscInt *rip,i,mbs = a->mbs,*ai,*aj; 54 PetscInt *jutmp,bs = A->rmap.bs,bs2=a->bs2; 55 PetscInt m,reallocs = 0,prow; 56 PetscInt *jl,*q,jmin,jmax,juidx,nzk,qm,*iu,*ju,k,j,vj,umax,maxadd; 57 PetscReal f = info->fill; 58 PetscTruth perm_identity; 59 60 PetscFunctionBegin; 61 /* check whether perm is the identity mapping */ 62 ierr = ISIdentity(perm,&perm_identity);CHKERRQ(ierr); 63 ierr = ISGetIndices(perm,&rip);CHKERRQ(ierr); 64 65 if (perm_identity){ /* without permutation */ 66 a->permute = PETSC_FALSE; 67 ai = a->i; aj = a->j; 68 } else { /* non-trivial permutation */ 69 a->permute = PETSC_TRUE; 70 ierr = MatReorderingSeqSBAIJ(A,perm);CHKERRQ(ierr); 71 ai = a->inew; aj = a->jnew; 72 } 73 74 /* initialization */ 75 ierr = PetscMalloc((mbs+1)*sizeof(PetscInt),&iu);CHKERRQ(ierr); 76 umax = (PetscInt)(f*ai[mbs] + 1); umax += mbs + 1; 77 ierr = PetscMalloc(umax*sizeof(PetscInt),&ju);CHKERRQ(ierr); 78 iu[0] = mbs+1; 79 juidx = mbs + 1; /* index for ju */ 80 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&jl);CHKERRQ(ierr); /* linked list for pivot row */ 81 q = jl + mbs; /* linked list for col index */ 82 for (i=0; i<mbs; i++){ 83 jl[i] = mbs; 84 q[i] = 0; 85 } 86 87 /* for each row k */ 88 for (k=0; k<mbs; k++){ 89 for (i=0; i<mbs; i++) q[i] = 0; /* to be removed! */ 90 nzk = 0; /* num. of nz blocks in k-th block row with diagonal block excluded */ 91 q[k] = mbs; 92 /* initialize nonzero structure of k-th row to row rip[k] of A */ 93 jmin = ai[rip[k]] +1; /* exclude diag[k] */ 94 jmax = ai[rip[k]+1]; 95 for (j=jmin; j<jmax; j++){ 96 vj = rip[aj[j]]; /* col. value */ 97 if(vj > k){ 98 qm = k; 99 do { 100 m = qm; qm = q[m]; 101 } while(qm < vj); 102 if (qm == vj) { 103 SETERRQ(PETSC_ERR_PLIB,"Duplicate entry in A\n"); 104 } 105 nzk++; 106 q[m] = vj; 107 q[vj] = qm; 108 } /* if(vj > k) */ 109 } /* for (j=jmin; j<jmax; j++) */ 110 111 /* modify nonzero structure of k-th row by computing fill-in 112 for each row i to be merged in */ 113 prow = k; 114 prow = jl[prow]; /* next pivot row (== mbs for symbolic factorization) */ 115 116 while (prow < k){ 117 /* merge row prow into k-th row */ 118 jmin = iu[prow] + 1; jmax = iu[prow+1]; 119 qm = k; 120 for (j=jmin; j<jmax; j++){ 121 vj = ju[j]; 122 do { 123 m = qm; qm = q[m]; 124 } while (qm < vj); 125 if (qm != vj){ 126 nzk++; q[m] = vj; q[vj] = qm; qm = vj; 127 } 128 } 129 prow = jl[prow]; /* next pivot row */ 130 } 131 132 /* add k to row list for first nonzero element in k-th row */ 133 if (nzk > 0){ 134 i = q[k]; /* col value of first nonzero element in U(k, k+1:mbs-1) */ 135 jl[k] = jl[i]; jl[i] = k; 136 } 137 iu[k+1] = iu[k] + nzk; 138 139 /* allocate more space to ju if needed */ 140 if (iu[k+1] > umax) { 141 /* estimate how much additional space we will need */ 142 /* use the strategy suggested by David Hysom <hysom@perch-t.icase.edu> */ 143 /* just double the memory each time */ 144 maxadd = umax; 145 if (maxadd < nzk) maxadd = (mbs-k)*(nzk+1)/2; 146 umax += maxadd; 147 148 /* allocate a longer ju */ 149 ierr = PetscMalloc(umax*sizeof(PetscInt),&jutmp);CHKERRQ(ierr); 150 ierr = PetscMemcpy(jutmp,ju,iu[k]*sizeof(PetscInt));CHKERRQ(ierr); 151 ierr = PetscFree(ju);CHKERRQ(ierr); 152 ju = jutmp; 153 reallocs++; /* count how many times we realloc */ 154 } 155 156 /* save nonzero structure of k-th row in ju */ 157 i=k; 158 while (nzk --) { 159 i = q[i]; 160 ju[juidx++] = i; 161 } 162 } 163 164 #if defined(PETSC_USE_INFO) 165 if (ai[mbs] != 0) { 166 PetscReal af = ((PetscReal)iu[mbs])/((PetscReal)ai[mbs]); 167 ierr = PetscInfo3(A,"Reallocs %D Fill ratio:given %G needed %G\n",reallocs,f,af);CHKERRQ(ierr); 168 ierr = PetscInfo1(A,"Run with -pc_factor_fill %G or use \n",af);CHKERRQ(ierr); 169 ierr = PetscInfo1(A,"PCFactorSetFill(pc,%G);\n",af);CHKERRQ(ierr); 170 ierr = PetscInfo(A,"for best performance.\n");CHKERRQ(ierr); 171 } else { 172 ierr = PetscInfo(A,"Empty matrix.\n");CHKERRQ(ierr); 173 } 174 #endif 175 176 ierr = ISRestoreIndices(perm,&rip);CHKERRQ(ierr); 177 ierr = PetscFree(jl);CHKERRQ(ierr); 178 179 /* put together the new matrix */ 180 ierr = MatCreate(((PetscObject)A)->comm,B);CHKERRQ(ierr); 181 ierr = MatSetSizes(*B,bs*mbs,bs*mbs,bs*mbs,bs*mbs);CHKERRQ(ierr); 182 ierr = MatSetType(*B,((PetscObject)A)->type_name);CHKERRQ(ierr); 183 ierr = MatSeqSBAIJSetPreallocation_SeqSBAIJ(*B,bs,MAT_SKIP_ALLOCATION,PETSC_NULL);CHKERRQ(ierr); 184 185 /* ierr = PetscLogObjectParent(*B,iperm);CHKERRQ(ierr); */ 186 b = (Mat_SeqSBAIJ*)(*B)->data; 187 b->singlemalloc = PETSC_FALSE; 188 b->free_a = PETSC_TRUE; 189 b->free_ij = PETSC_TRUE; 190 ierr = PetscMalloc((iu[mbs]+1)*sizeof(MatScalar)*bs2,&b->a);CHKERRQ(ierr); 191 b->j = ju; 192 b->i = iu; 193 b->diag = 0; 194 b->ilen = 0; 195 b->imax = 0; 196 b->row = perm; 197 b->pivotinblocks = PETSC_FALSE; /* need to get from MatFactorInfo */ 198 ierr = PetscObjectReference((PetscObject)perm);CHKERRQ(ierr); 199 b->icol = perm; 200 ierr = PetscObjectReference((PetscObject)perm);CHKERRQ(ierr); 201 ierr = PetscMalloc((bs*mbs+bs)*sizeof(PetscScalar),&b->solve_work);CHKERRQ(ierr); 202 /* In b structure: Free imax, ilen, old a, old j. 203 Allocate idnew, solve_work, new a, new j */ 204 ierr = PetscLogObjectMemory(*B,(iu[mbs]-mbs)*(sizeof(PetscInt)+sizeof(MatScalar)));CHKERRQ(ierr); 205 b->maxnz = b->nz = iu[mbs]; 206 207 (*B)->factor = MAT_FACTOR_CHOLESKY; 208 (*B)->info.factor_mallocs = reallocs; 209 (*B)->info.fill_ratio_given = f; 210 if (ai[mbs] != 0) { 211 (*B)->info.fill_ratio_needed = ((PetscReal)iu[mbs])/((PetscReal)ai[mbs]); 212 } else { 213 (*B)->info.fill_ratio_needed = 0.0; 214 } 215 216 if (perm_identity){ 217 switch (bs) { 218 case 1: 219 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_1_NaturalOrdering; 220 (*B)->ops->solve = MatSolve_SeqSBAIJ_1_NaturalOrdering; 221 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_1_NaturalOrdering; 222 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_1_NaturalOrdering; 223 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=1\n");CHKERRQ(ierr); 224 break; 225 case 2: 226 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_2_NaturalOrdering; 227 (*B)->ops->solve = MatSolve_SeqSBAIJ_2_NaturalOrdering; 228 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_2_NaturalOrdering; 229 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_2_NaturalOrdering; 230 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=2\n");CHKERRQ(ierr); 231 break; 232 case 3: 233 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_3_NaturalOrdering; 234 (*B)->ops->solve = MatSolve_SeqSBAIJ_3_NaturalOrdering; 235 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_3_NaturalOrdering; 236 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_3_NaturalOrdering; 237 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=3\n");CHKERRQ(ierr); 238 break; 239 case 4: 240 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_4_NaturalOrdering; 241 (*B)->ops->solve = MatSolve_SeqSBAIJ_4_NaturalOrdering; 242 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_4_NaturalOrdering; 243 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_4_NaturalOrdering; 244 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=4\n");CHKERRQ(ierr); 245 break; 246 case 5: 247 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_5_NaturalOrdering; 248 (*B)->ops->solve = MatSolve_SeqSBAIJ_5_NaturalOrdering; 249 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_5_NaturalOrdering; 250 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_5_NaturalOrdering; 251 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=5\n");CHKERRQ(ierr); 252 break; 253 case 6: 254 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_6_NaturalOrdering; 255 (*B)->ops->solve = MatSolve_SeqSBAIJ_6_NaturalOrdering; 256 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_6_NaturalOrdering; 257 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_6_NaturalOrdering; 258 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=6\n");CHKERRQ(ierr); 259 break; 260 case 7: 261 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_7_NaturalOrdering; 262 (*B)->ops->solve = MatSolve_SeqSBAIJ_7_NaturalOrdering; 263 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_7_NaturalOrdering; 264 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_7_NaturalOrdering; 265 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=7\n");CHKERRQ(ierr); 266 break; 267 default: 268 (*B)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_N_NaturalOrdering; 269 (*B)->ops->solve = MatSolve_SeqSBAIJ_N_NaturalOrdering; 270 (*B)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_N_NaturalOrdering; 271 (*B)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_N_NaturalOrdering; 272 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS>7\n");CHKERRQ(ierr); 273 break; 274 } 275 } 276 PetscFunctionReturn(0); 277 } 278 /* 279 Symbolic U^T*D*U factorization for SBAIJ format. 280 */ 281 #include "petscbt.h" 282 #include "src/mat/utils/freespace.h" 283 #undef __FUNCT__ 284 #define __FUNCT__ "MatCholeskyFactorSymbolic_SeqSBAIJ" 285 PetscErrorCode MatCholeskyFactorSymbolic_SeqSBAIJ(Mat A,IS perm,MatFactorInfo *info,Mat *fact) 286 { 287 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data; 288 Mat_SeqSBAIJ *b; 289 Mat B; 290 PetscErrorCode ierr; 291 PetscTruth perm_identity,missing; 292 PetscReal fill = info->fill; 293 PetscInt *rip,i,mbs=a->mbs,bs=A->rmap.bs,*ai,*aj,reallocs=0,prow,d; 294 PetscInt *jl,jmin,jmax,nzk,*ui,k,j,*il,nextprow; 295 PetscInt nlnk,*lnk,ncols,*cols,*uj,**ui_ptr,*uj_ptr; 296 PetscFreeSpaceList free_space=PETSC_NULL,current_space=PETSC_NULL; 297 PetscBT lnkbt; 298 299 PetscFunctionBegin; 300 ierr = MatMissingDiagonal(A,&missing,&d);CHKERRQ(ierr); 301 if (missing) SETERRQ1(PETSC_ERR_ARG_WRONGSTATE,"Matrix is missing diagonal entry %D",d); 302 303 /* 304 This code originally uses Modified Sparse Row (MSR) storage 305 (see page 85, "Iterative Methods ..." by Saad) for the output matrix B - bad choise! 306 Then it is rewritten so the factor B takes seqsbaij format. However the associated 307 MatCholeskyFactorNumeric_() have not been modified for the cases of bs>1 or !perm_identity, 308 thus the original code in MSR format is still used for these cases. 309 The code below should replace MatCholeskyFactorSymbolic_SeqSBAIJ_MSR() whenever 310 MatCholeskyFactorNumeric_() is modified for using sbaij symbolic factor. 311 */ 312 if (bs > 1){ 313 ierr = MatCholeskyFactorSymbolic_SeqSBAIJ_MSR(A,perm,info,fact);CHKERRQ(ierr); 314 PetscFunctionReturn(0); 315 } 316 317 /* check whether perm is the identity mapping */ 318 ierr = ISIdentity(perm,&perm_identity);CHKERRQ(ierr); 319 320 if (perm_identity){ 321 a->permute = PETSC_FALSE; 322 ai = a->i; aj = a->j; 323 } else { 324 SETERRQ(PETSC_ERR_SUP,"Matrix reordering is not supported for sbaij matrix. Use aij format"); 325 /* There are bugs for reordeing. Needs further work. 326 MatReordering for sbaij cannot be efficient. User should use aij formt! */ 327 a->permute = PETSC_TRUE; 328 ierr = MatReorderingSeqSBAIJ(A,perm);CHKERRQ(ierr); 329 ai = a->inew; aj = a->jnew; 330 } 331 ierr = ISGetIndices(perm,&rip);CHKERRQ(ierr); 332 333 /* initialization */ 334 ierr = PetscMalloc((mbs+1)*sizeof(PetscInt),&ui);CHKERRQ(ierr); 335 ui[0] = 0; 336 337 /* jl: linked list for storing indices of the pivot rows 338 il: il[i] points to the 1st nonzero entry of U(i,k:mbs-1) */ 339 ierr = PetscMalloc((3*mbs+1)*sizeof(PetscInt)+mbs*sizeof(PetscInt*),&jl);CHKERRQ(ierr); 340 il = jl + mbs; 341 cols = il + mbs; 342 ui_ptr = (PetscInt**)(cols + mbs); 343 344 for (i=0; i<mbs; i++){ 345 jl[i] = mbs; il[i] = 0; 346 } 347 348 /* create and initialize a linked list for storing column indices of the active row k */ 349 nlnk = mbs + 1; 350 ierr = PetscLLCreate(mbs,mbs,nlnk,lnk,lnkbt);CHKERRQ(ierr); 351 352 /* initial FreeSpace size is fill*(ai[mbs]+1) */ 353 ierr = PetscFreeSpaceGet((PetscInt)(fill*(ai[mbs]+1)),&free_space);CHKERRQ(ierr); 354 current_space = free_space; 355 356 for (k=0; k<mbs; k++){ /* for each active row k */ 357 /* initialize lnk by the column indices of row rip[k] of A */ 358 nzk = 0; 359 ncols = ai[rip[k]+1] - ai[rip[k]]; 360 for (j=0; j<ncols; j++){ 361 i = *(aj + ai[rip[k]] + j); 362 cols[j] = rip[i]; 363 } 364 ierr = PetscLLAdd(ncols,cols,mbs,nlnk,lnk,lnkbt);CHKERRQ(ierr); 365 nzk += nlnk; 366 367 /* update lnk by computing fill-in for each pivot row to be merged in */ 368 prow = jl[k]; /* 1st pivot row */ 369 370 while (prow < k){ 371 nextprow = jl[prow]; 372 /* merge prow into k-th row */ 373 jmin = il[prow] + 1; /* index of the 2nd nzero entry in U(prow,k:mbs-1) */ 374 jmax = ui[prow+1]; 375 ncols = jmax-jmin; 376 uj_ptr = ui_ptr[prow] + jmin - ui[prow]; /* points to the 2nd nzero entry in U(prow,k:mbs-1) */ 377 ierr = PetscLLAddSorted(ncols,uj_ptr,mbs,nlnk,lnk,lnkbt);CHKERRQ(ierr); 378 nzk += nlnk; 379 380 /* update il and jl for prow */ 381 if (jmin < jmax){ 382 il[prow] = jmin; 383 j = *uj_ptr; jl[prow] = jl[j]; jl[j] = prow; 384 } 385 prow = nextprow; 386 } 387 388 /* if free space is not available, make more free space */ 389 if (current_space->local_remaining<nzk) { 390 i = mbs - k + 1; /* num of unfactored rows */ 391 i = PetscMin(i*nzk, i*(i-1)); /* i*nzk, i*(i-1): estimated and max additional space needed */ 392 ierr = PetscFreeSpaceGet(i,¤t_space);CHKERRQ(ierr); 393 reallocs++; 394 } 395 396 /* copy data into free space, then initialize lnk */ 397 ierr = PetscLLClean(mbs,mbs,nzk,lnk,current_space->array,lnkbt);CHKERRQ(ierr); 398 399 /* add the k-th row into il and jl */ 400 if (nzk-1 > 0){ 401 i = current_space->array[1]; /* col value of the first nonzero element in U(k, k+1:mbs-1) */ 402 jl[k] = jl[i]; jl[i] = k; 403 il[k] = ui[k] + 1; 404 } 405 ui_ptr[k] = current_space->array; 406 current_space->array += nzk; 407 current_space->local_used += nzk; 408 current_space->local_remaining -= nzk; 409 410 ui[k+1] = ui[k] + nzk; 411 } 412 413 #if defined(PETSC_USE_INFO) 414 if (ai[mbs] != 0) { 415 PetscReal af = ((PetscReal)ui[mbs])/((PetscReal)ai[mbs]); 416 ierr = PetscInfo3(A,"Reallocs %D Fill ratio:given %G needed %G\n",reallocs,fill,af);CHKERRQ(ierr); 417 ierr = PetscInfo1(A,"Run with -pc_factor_fill %G or use \n",af);CHKERRQ(ierr); 418 ierr = PetscInfo1(A,"PCFactorSetFill(pc,%G) for best performance.\n",af);CHKERRQ(ierr); 419 } else { 420 ierr = PetscInfo(A,"Empty matrix.\n");CHKERRQ(ierr); 421 } 422 #endif 423 424 ierr = ISRestoreIndices(perm,&rip);CHKERRQ(ierr); 425 ierr = PetscFree(jl);CHKERRQ(ierr); 426 427 /* destroy list of free space and other temporary array(s) */ 428 ierr = PetscMalloc((ui[mbs]+1)*sizeof(PetscInt),&uj);CHKERRQ(ierr); 429 ierr = PetscFreeSpaceContiguous(&free_space,uj);CHKERRQ(ierr); 430 ierr = PetscLLDestroy(lnk,lnkbt);CHKERRQ(ierr); 431 432 /* put together the new matrix in MATSEQSBAIJ format */ 433 ierr = MatSeqSBAIJSetPreallocation_SeqSBAIJ(*fact,bs,MAT_SKIP_ALLOCATION,PETSC_NULL);CHKERRQ(ierr); 434 435 b = (Mat_SeqSBAIJ*)(*fact)->data; 436 b->singlemalloc = PETSC_FALSE; 437 b->free_a = PETSC_TRUE; 438 b->free_ij = PETSC_TRUE; 439 ierr = PetscMalloc((ui[mbs]+1)*sizeof(MatScalar),&b->a);CHKERRQ(ierr); 440 b->j = uj; 441 b->i = ui; 442 b->diag = 0; 443 b->ilen = 0; 444 b->imax = 0; 445 b->row = perm; 446 b->pivotinblocks = PETSC_FALSE; /* need to get from MatFactorInfo */ 447 ierr = PetscObjectReference((PetscObject)perm);CHKERRQ(ierr); 448 b->icol = perm; 449 ierr = PetscObjectReference((PetscObject)perm);CHKERRQ(ierr); 450 ierr = PetscMalloc((mbs+1)*sizeof(PetscScalar),&b->solve_work);CHKERRQ(ierr); 451 ierr = PetscLogObjectMemory(*fact,(ui[mbs]-mbs)*(sizeof(PetscInt)+sizeof(MatScalar)));CHKERRQ(ierr); 452 b->maxnz = b->nz = ui[mbs]; 453 454 (*fact)->factor = MAT_FACTOR_CHOLESKY; 455 (*fact)->info.factor_mallocs = reallocs; 456 (*fact)->info.fill_ratio_given = fill; 457 if (ai[mbs] != 0) { 458 (*fact)->info.fill_ratio_needed = ((PetscReal)ui[mbs])/((PetscReal)ai[mbs]); 459 } else { 460 (*fact)->info.fill_ratio_needed = 0.0; 461 } 462 463 if (perm_identity){ 464 switch (bs) { 465 case 1: 466 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_1_NaturalOrdering; 467 (*fact)->ops->solve = MatSolve_SeqSBAIJ_1_NaturalOrdering; 468 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_1_NaturalOrdering; 469 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_1_NaturalOrdering; 470 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=1\n");CHKERRQ(ierr); 471 break; 472 case 2: 473 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_2_NaturalOrdering; 474 (*fact)->ops->solve = MatSolve_SeqSBAIJ_2_NaturalOrdering; 475 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_2_NaturalOrdering; 476 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_2_NaturalOrdering; 477 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=2\n");CHKERRQ(ierr); 478 break; 479 case 3: 480 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_3_NaturalOrdering; 481 (*fact)->ops->solve = MatSolve_SeqSBAIJ_3_NaturalOrdering; 482 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_3_NaturalOrdering; 483 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_3_NaturalOrdering; 484 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=3\n");CHKERRQ(ierr); 485 break; 486 case 4: 487 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_4_NaturalOrdering; 488 (*fact)->ops->solve = MatSolve_SeqSBAIJ_4_NaturalOrdering; 489 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_4_NaturalOrdering; 490 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_4_NaturalOrdering; 491 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=4\n");CHKERRQ(ierr); 492 break; 493 case 5: 494 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_5_NaturalOrdering; 495 (*fact)->ops->solve = MatSolve_SeqSBAIJ_5_NaturalOrdering; 496 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_5_NaturalOrdering; 497 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_5_NaturalOrdering; 498 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=5\n");CHKERRQ(ierr); 499 break; 500 case 6: 501 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_6_NaturalOrdering; 502 (*fact)->ops->solve = MatSolve_SeqSBAIJ_6_NaturalOrdering; 503 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_6_NaturalOrdering; 504 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_6_NaturalOrdering; 505 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=6\n");CHKERRQ(ierr); 506 break; 507 case 7: 508 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_7_NaturalOrdering; 509 (*fact)->ops->solve = MatSolve_SeqSBAIJ_7_NaturalOrdering; 510 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_7_NaturalOrdering; 511 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_7_NaturalOrdering; 512 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS=7\n");CHKERRQ(ierr); 513 break; 514 default: 515 (*fact)->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqSBAIJ_N_NaturalOrdering; 516 (*fact)->ops->solve = MatSolve_SeqSBAIJ_N_NaturalOrdering; 517 (*fact)->ops->forwardsolve = MatForwardSolve_SeqSBAIJ_N_NaturalOrdering; 518 (*fact)->ops->backwardsolve = MatBackwardSolve_SeqSBAIJ_N_NaturalOrdering; 519 ierr = PetscInfo(A,"Using special in-place natural ordering factor and solve BS>7\n");CHKERRQ(ierr); 520 break; 521 } 522 } 523 PetscFunctionReturn(0); 524 } 525 #undef __FUNCT__ 526 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_N" 527 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_N(Mat A,MatFactorInfo *info,Mat *B) 528 { 529 Mat C = *B; 530 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data,*b = (Mat_SeqSBAIJ *)C->data; 531 IS perm = b->row; 532 PetscErrorCode ierr; 533 PetscInt *perm_ptr,i,j,mbs=a->mbs,*bi=b->i,*bj=b->j; 534 PetscInt *ai,*aj,*a2anew,k,k1,jmin,jmax,*jl,*il,vj,nexti,ili; 535 PetscInt bs=A->rmap.bs,bs2 = a->bs2,bslog = 0; 536 MatScalar *ba = b->a,*aa,*ap,*dk,*uik; 537 MatScalar *u,*diag,*rtmp,*rtmp_ptr; 538 MatScalar *work; 539 PetscInt *pivots; 540 541 PetscFunctionBegin; 542 /* initialization */ 543 ierr = PetscMalloc(bs2*mbs*sizeof(MatScalar),&rtmp);CHKERRQ(ierr); 544 ierr = PetscMemzero(rtmp,bs2*mbs*sizeof(MatScalar));CHKERRQ(ierr); 545 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&il);CHKERRQ(ierr); 546 jl = il + mbs; 547 for (i=0; i<mbs; i++) { 548 jl[i] = mbs; il[0] = 0; 549 } 550 ierr = PetscMalloc((2*bs2+bs)*sizeof(MatScalar),&dk);CHKERRQ(ierr); 551 uik = dk + bs2; 552 work = uik + bs2; 553 ierr = PetscMalloc(bs*sizeof(PetscInt),&pivots);CHKERRQ(ierr); 554 555 ierr = ISGetIndices(perm,&perm_ptr);CHKERRQ(ierr); 556 557 /* check permutation */ 558 if (!a->permute){ 559 ai = a->i; aj = a->j; aa = a->a; 560 } else { 561 ai = a->inew; aj = a->jnew; 562 ierr = PetscMalloc(bs2*ai[mbs]*sizeof(MatScalar),&aa);CHKERRQ(ierr); 563 ierr = PetscMemcpy(aa,a->a,bs2*ai[mbs]*sizeof(MatScalar));CHKERRQ(ierr); 564 ierr = PetscMalloc(ai[mbs]*sizeof(PetscInt),&a2anew);CHKERRQ(ierr); 565 ierr = PetscMemcpy(a2anew,a->a2anew,(ai[mbs])*sizeof(PetscInt));CHKERRQ(ierr); 566 567 /* flops in while loop */ 568 bslog = 2*bs*bs2; 569 570 for (i=0; i<mbs; i++){ 571 jmin = ai[i]; jmax = ai[i+1]; 572 for (j=jmin; j<jmax; j++){ 573 while (a2anew[j] != j){ 574 k = a2anew[j]; a2anew[j] = a2anew[k]; a2anew[k] = k; 575 for (k1=0; k1<bs2; k1++){ 576 dk[k1] = aa[k*bs2+k1]; 577 aa[k*bs2+k1] = aa[j*bs2+k1]; 578 aa[j*bs2+k1] = dk[k1]; 579 } 580 } 581 /* transform columnoriented blocks that lie in the lower triangle to roworiented blocks */ 582 if (i > aj[j]){ 583 /* printf("change orientation, row: %d, col: %d\n",i,aj[j]); */ 584 ap = aa + j*bs2; /* ptr to the beginning of j-th block of aa */ 585 for (k=0; k<bs2; k++) dk[k] = ap[k]; /* dk <- j-th block of aa */ 586 for (k=0; k<bs; k++){ /* j-th block of aa <- dk^T */ 587 for (k1=0; k1<bs; k1++) *ap++ = dk[k + bs*k1]; 588 } 589 } 590 } 591 } 592 ierr = PetscFree(a2anew);CHKERRQ(ierr); 593 } 594 595 /* for each row k */ 596 for (k = 0; k<mbs; k++){ 597 598 /*initialize k-th row with elements nonzero in row perm(k) of A */ 599 jmin = ai[perm_ptr[k]]; jmax = ai[perm_ptr[k]+1]; 600 601 ap = aa + jmin*bs2; 602 for (j = jmin; j < jmax; j++){ 603 vj = perm_ptr[aj[j]]; /* block col. index */ 604 rtmp_ptr = rtmp + vj*bs2; 605 for (i=0; i<bs2; i++) *rtmp_ptr++ = *ap++; 606 } 607 608 /* modify k-th row by adding in those rows i with U(i,k) != 0 */ 609 ierr = PetscMemcpy(dk,rtmp+k*bs2,bs2*sizeof(MatScalar));CHKERRQ(ierr); 610 i = jl[k]; /* first row to be added to k_th row */ 611 612 while (i < k){ 613 nexti = jl[i]; /* next row to be added to k_th row */ 614 615 /* compute multiplier */ 616 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 617 618 /* uik = -inv(Di)*U_bar(i,k) */ 619 diag = ba + i*bs2; 620 u = ba + ili*bs2; 621 ierr = PetscMemzero(uik,bs2*sizeof(MatScalar));CHKERRQ(ierr); 622 Kernel_A_gets_A_minus_B_times_C(bs,uik,diag,u); 623 624 /* update D(k) += -U(i,k)^T * U_bar(i,k) */ 625 Kernel_A_gets_A_plus_Btranspose_times_C(bs,dk,uik,u); 626 ierr = PetscLogFlops(bslog*2);CHKERRQ(ierr); 627 628 /* update -U(i,k) */ 629 ierr = PetscMemcpy(ba+ili*bs2,uik,bs2*sizeof(MatScalar));CHKERRQ(ierr); 630 631 /* add multiple of row i to k-th row ... */ 632 jmin = ili + 1; jmax = bi[i+1]; 633 if (jmin < jmax){ 634 for (j=jmin; j<jmax; j++) { 635 /* rtmp += -U(i,k)^T * U_bar(i,j) */ 636 rtmp_ptr = rtmp + bj[j]*bs2; 637 u = ba + j*bs2; 638 Kernel_A_gets_A_plus_Btranspose_times_C(bs,rtmp_ptr,uik,u); 639 } 640 ierr = PetscLogFlops(bslog*(jmax-jmin));CHKERRQ(ierr); 641 642 /* ... add i to row list for next nonzero entry */ 643 il[i] = jmin; /* update il(i) in column k+1, ... mbs-1 */ 644 j = bj[jmin]; 645 jl[i] = jl[j]; jl[j] = i; /* update jl */ 646 } 647 i = nexti; 648 } 649 650 /* save nonzero entries in k-th row of U ... */ 651 652 /* invert diagonal block */ 653 diag = ba+k*bs2; 654 ierr = PetscMemcpy(diag,dk,bs2*sizeof(MatScalar));CHKERRQ(ierr); 655 ierr = Kernel_A_gets_inverse_A(bs,diag,pivots,work);CHKERRQ(ierr); 656 657 jmin = bi[k]; jmax = bi[k+1]; 658 if (jmin < jmax) { 659 for (j=jmin; j<jmax; j++){ 660 vj = bj[j]; /* block col. index of U */ 661 u = ba + j*bs2; 662 rtmp_ptr = rtmp + vj*bs2; 663 for (k1=0; k1<bs2; k1++){ 664 *u++ = *rtmp_ptr; 665 *rtmp_ptr++ = 0.0; 666 } 667 } 668 669 /* ... add k to row list for first nonzero entry in k-th row */ 670 il[k] = jmin; 671 i = bj[jmin]; 672 jl[k] = jl[i]; jl[i] = k; 673 } 674 } 675 676 ierr = PetscFree(rtmp);CHKERRQ(ierr); 677 ierr = PetscFree(il);CHKERRQ(ierr); 678 ierr = PetscFree(dk);CHKERRQ(ierr); 679 ierr = PetscFree(pivots);CHKERRQ(ierr); 680 if (a->permute){ 681 ierr = PetscFree(aa);CHKERRQ(ierr); 682 } 683 684 ierr = ISRestoreIndices(perm,&perm_ptr);CHKERRQ(ierr); 685 C->factor = MAT_FACTOR_CHOLESKY; 686 C->assembled = PETSC_TRUE; 687 C->preallocated = PETSC_TRUE; 688 ierr = PetscLogFlops(1.3333*bs*bs2*b->mbs);CHKERRQ(ierr); /* from inverting diagonal blocks */ 689 PetscFunctionReturn(0); 690 } 691 692 #undef __FUNCT__ 693 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_N_NaturalOrdering" 694 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_N_NaturalOrdering(Mat A,MatFactorInfo *info,Mat *B) 695 { 696 Mat C = *B; 697 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data,*b = (Mat_SeqSBAIJ *)C->data; 698 PetscErrorCode ierr; 699 PetscInt i,j,mbs=a->mbs,*bi=b->i,*bj=b->j; 700 PetscInt *ai,*aj,k,k1,jmin,jmax,*jl,*il,vj,nexti,ili; 701 PetscInt bs=A->rmap.bs,bs2 = a->bs2,bslog; 702 MatScalar *ba = b->a,*aa,*ap,*dk,*uik; 703 MatScalar *u,*diag,*rtmp,*rtmp_ptr; 704 MatScalar *work; 705 PetscInt *pivots; 706 707 PetscFunctionBegin; 708 ierr = PetscMalloc(bs2*mbs*sizeof(MatScalar),&rtmp);CHKERRQ(ierr); 709 ierr = PetscMemzero(rtmp,bs2*mbs*sizeof(MatScalar));CHKERRQ(ierr); 710 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&il);CHKERRQ(ierr); 711 jl = il + mbs; 712 for (i=0; i<mbs; i++) { 713 jl[i] = mbs; il[0] = 0; 714 } 715 ierr = PetscMalloc((2*bs2+bs)*sizeof(MatScalar),&dk);CHKERRQ(ierr); 716 uik = dk + bs2; 717 work = uik + bs2; 718 ierr = PetscMalloc(bs*sizeof(PetscInt),&pivots);CHKERRQ(ierr); 719 720 ai = a->i; aj = a->j; aa = a->a; 721 722 /* flops in while loop */ 723 bslog = 2*bs*bs2; 724 725 /* for each row k */ 726 for (k = 0; k<mbs; k++){ 727 728 /*initialize k-th row with elements nonzero in row k of A */ 729 jmin = ai[k]; jmax = ai[k+1]; 730 ap = aa + jmin*bs2; 731 for (j = jmin; j < jmax; j++){ 732 vj = aj[j]; /* block col. index */ 733 rtmp_ptr = rtmp + vj*bs2; 734 for (i=0; i<bs2; i++) *rtmp_ptr++ = *ap++; 735 } 736 737 /* modify k-th row by adding in those rows i with U(i,k) != 0 */ 738 ierr = PetscMemcpy(dk,rtmp+k*bs2,bs2*sizeof(MatScalar));CHKERRQ(ierr); 739 i = jl[k]; /* first row to be added to k_th row */ 740 741 while (i < k){ 742 nexti = jl[i]; /* next row to be added to k_th row */ 743 744 /* compute multiplier */ 745 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 746 747 /* uik = -inv(Di)*U_bar(i,k) */ 748 diag = ba + i*bs2; 749 u = ba + ili*bs2; 750 ierr = PetscMemzero(uik,bs2*sizeof(MatScalar));CHKERRQ(ierr); 751 Kernel_A_gets_A_minus_B_times_C(bs,uik,diag,u); 752 753 /* update D(k) += -U(i,k)^T * U_bar(i,k) */ 754 Kernel_A_gets_A_plus_Btranspose_times_C(bs,dk,uik,u); 755 ierr = PetscLogFlops(bslog*2);CHKERRQ(ierr); 756 757 /* update -U(i,k) */ 758 ierr = PetscMemcpy(ba+ili*bs2,uik,bs2*sizeof(MatScalar));CHKERRQ(ierr); 759 760 /* add multiple of row i to k-th row ... */ 761 jmin = ili + 1; jmax = bi[i+1]; 762 if (jmin < jmax){ 763 for (j=jmin; j<jmax; j++) { 764 /* rtmp += -U(i,k)^T * U_bar(i,j) */ 765 rtmp_ptr = rtmp + bj[j]*bs2; 766 u = ba + j*bs2; 767 Kernel_A_gets_A_plus_Btranspose_times_C(bs,rtmp_ptr,uik,u); 768 } 769 ierr = PetscLogFlops(bslog*(jmax-jmin));CHKERRQ(ierr); 770 771 /* ... add i to row list for next nonzero entry */ 772 il[i] = jmin; /* update il(i) in column k+1, ... mbs-1 */ 773 j = bj[jmin]; 774 jl[i] = jl[j]; jl[j] = i; /* update jl */ 775 } 776 i = nexti; 777 } 778 779 /* save nonzero entries in k-th row of U ... */ 780 781 /* invert diagonal block */ 782 diag = ba+k*bs2; 783 ierr = PetscMemcpy(diag,dk,bs2*sizeof(MatScalar));CHKERRQ(ierr); 784 ierr = Kernel_A_gets_inverse_A(bs,diag,pivots,work);CHKERRQ(ierr); 785 786 jmin = bi[k]; jmax = bi[k+1]; 787 if (jmin < jmax) { 788 for (j=jmin; j<jmax; j++){ 789 vj = bj[j]; /* block col. index of U */ 790 u = ba + j*bs2; 791 rtmp_ptr = rtmp + vj*bs2; 792 for (k1=0; k1<bs2; k1++){ 793 *u++ = *rtmp_ptr; 794 *rtmp_ptr++ = 0.0; 795 } 796 } 797 798 /* ... add k to row list for first nonzero entry in k-th row */ 799 il[k] = jmin; 800 i = bj[jmin]; 801 jl[k] = jl[i]; jl[i] = k; 802 } 803 } 804 805 ierr = PetscFree(rtmp);CHKERRQ(ierr); 806 ierr = PetscFree(il);CHKERRQ(ierr); 807 ierr = PetscFree(dk);CHKERRQ(ierr); 808 ierr = PetscFree(pivots);CHKERRQ(ierr); 809 810 C->factor = MAT_FACTOR_CHOLESKY; 811 C->assembled = PETSC_TRUE; 812 C->preallocated = PETSC_TRUE; 813 ierr = PetscLogFlops(1.3333*bs*bs2*b->mbs);CHKERRQ(ierr); /* from inverting diagonal blocks */ 814 PetscFunctionReturn(0); 815 } 816 817 /* 818 Numeric U^T*D*U factorization for SBAIJ format. Modified from SNF of YSMP. 819 Version for blocks 2 by 2. 820 */ 821 #undef __FUNCT__ 822 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_2" 823 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_2(Mat A,MatFactorInfo *info,Mat *B) 824 { 825 Mat C = *B; 826 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data,*b = (Mat_SeqSBAIJ *)C->data; 827 IS perm = b->row; 828 PetscErrorCode ierr; 829 PetscInt *perm_ptr,i,j,mbs=a->mbs,*bi=b->i,*bj=b->j; 830 PetscInt *ai,*aj,*a2anew,k,k1,jmin,jmax,*jl,*il,vj,nexti,ili; 831 MatScalar *ba = b->a,*aa,*ap,*dk,*uik; 832 MatScalar *u,*diag,*rtmp,*rtmp_ptr; 833 PetscReal shift = info->shiftinblocks; 834 835 PetscFunctionBegin; 836 /* initialization */ 837 /* il and jl record the first nonzero element in each row of the accessing 838 window U(0:k, k:mbs-1). 839 jl: list of rows to be added to uneliminated rows 840 i>= k: jl(i) is the first row to be added to row i 841 i< k: jl(i) is the row following row i in some list of rows 842 jl(i) = mbs indicates the end of a list 843 il(i): points to the first nonzero element in columns k,...,mbs-1 of 844 row i of U */ 845 ierr = PetscMalloc(4*mbs*sizeof(MatScalar),&rtmp);CHKERRQ(ierr); 846 ierr = PetscMemzero(rtmp,4*mbs*sizeof(MatScalar));CHKERRQ(ierr); 847 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&il);CHKERRQ(ierr); 848 jl = il + mbs; 849 for (i=0; i<mbs; i++) { 850 jl[i] = mbs; il[0] = 0; 851 } 852 ierr = PetscMalloc(8*sizeof(MatScalar),&dk);CHKERRQ(ierr); 853 uik = dk + 4; 854 ierr = ISGetIndices(perm,&perm_ptr);CHKERRQ(ierr); 855 856 /* check permutation */ 857 if (!a->permute){ 858 ai = a->i; aj = a->j; aa = a->a; 859 } else { 860 ai = a->inew; aj = a->jnew; 861 ierr = PetscMalloc(4*ai[mbs]*sizeof(MatScalar),&aa);CHKERRQ(ierr); 862 ierr = PetscMemcpy(aa,a->a,4*ai[mbs]*sizeof(MatScalar));CHKERRQ(ierr); 863 ierr = PetscMalloc(ai[mbs]*sizeof(PetscInt),&a2anew);CHKERRQ(ierr); 864 ierr = PetscMemcpy(a2anew,a->a2anew,(ai[mbs])*sizeof(PetscInt));CHKERRQ(ierr); 865 866 for (i=0; i<mbs; i++){ 867 jmin = ai[i]; jmax = ai[i+1]; 868 for (j=jmin; j<jmax; j++){ 869 while (a2anew[j] != j){ 870 k = a2anew[j]; a2anew[j] = a2anew[k]; a2anew[k] = k; 871 for (k1=0; k1<4; k1++){ 872 dk[k1] = aa[k*4+k1]; 873 aa[k*4+k1] = aa[j*4+k1]; 874 aa[j*4+k1] = dk[k1]; 875 } 876 } 877 /* transform columnoriented blocks that lie in the lower triangle to roworiented blocks */ 878 if (i > aj[j]){ 879 /* printf("change orientation, row: %d, col: %d\n",i,aj[j]); */ 880 ap = aa + j*4; /* ptr to the beginning of the block */ 881 dk[1] = ap[1]; /* swap ap[1] and ap[2] */ 882 ap[1] = ap[2]; 883 ap[2] = dk[1]; 884 } 885 } 886 } 887 ierr = PetscFree(a2anew);CHKERRQ(ierr); 888 } 889 890 /* for each row k */ 891 for (k = 0; k<mbs; k++){ 892 893 /*initialize k-th row with elements nonzero in row perm(k) of A */ 894 jmin = ai[perm_ptr[k]]; jmax = ai[perm_ptr[k]+1]; 895 ap = aa + jmin*4; 896 for (j = jmin; j < jmax; j++){ 897 vj = perm_ptr[aj[j]]; /* block col. index */ 898 rtmp_ptr = rtmp + vj*4; 899 for (i=0; i<4; i++) *rtmp_ptr++ = *ap++; 900 } 901 902 /* modify k-th row by adding in those rows i with U(i,k) != 0 */ 903 ierr = PetscMemcpy(dk,rtmp+k*4,4*sizeof(MatScalar));CHKERRQ(ierr); 904 i = jl[k]; /* first row to be added to k_th row */ 905 906 while (i < k){ 907 nexti = jl[i]; /* next row to be added to k_th row */ 908 909 /* compute multiplier */ 910 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 911 912 /* uik = -inv(Di)*U_bar(i,k): - ba[ili]*ba[i] */ 913 diag = ba + i*4; 914 u = ba + ili*4; 915 uik[0] = -(diag[0]*u[0] + diag[2]*u[1]); 916 uik[1] = -(diag[1]*u[0] + diag[3]*u[1]); 917 uik[2] = -(diag[0]*u[2] + diag[2]*u[3]); 918 uik[3] = -(diag[1]*u[2] + diag[3]*u[3]); 919 920 /* update D(k) += -U(i,k)^T * U_bar(i,k): dk += uik*ba[ili] */ 921 dk[0] += uik[0]*u[0] + uik[1]*u[1]; 922 dk[1] += uik[2]*u[0] + uik[3]*u[1]; 923 dk[2] += uik[0]*u[2] + uik[1]*u[3]; 924 dk[3] += uik[2]*u[2] + uik[3]*u[3]; 925 926 ierr = PetscLogFlops(16*2);CHKERRQ(ierr); 927 928 /* update -U(i,k): ba[ili] = uik */ 929 ierr = PetscMemcpy(ba+ili*4,uik,4*sizeof(MatScalar));CHKERRQ(ierr); 930 931 /* add multiple of row i to k-th row ... */ 932 jmin = ili + 1; jmax = bi[i+1]; 933 if (jmin < jmax){ 934 for (j=jmin; j<jmax; j++) { 935 /* rtmp += -U(i,k)^T * U_bar(i,j): rtmp[bj[j]] += uik*ba[j]; */ 936 rtmp_ptr = rtmp + bj[j]*4; 937 u = ba + j*4; 938 rtmp_ptr[0] += uik[0]*u[0] + uik[1]*u[1]; 939 rtmp_ptr[1] += uik[2]*u[0] + uik[3]*u[1]; 940 rtmp_ptr[2] += uik[0]*u[2] + uik[1]*u[3]; 941 rtmp_ptr[3] += uik[2]*u[2] + uik[3]*u[3]; 942 } 943 ierr = PetscLogFlops(16*(jmax-jmin));CHKERRQ(ierr); 944 945 /* ... add i to row list for next nonzero entry */ 946 il[i] = jmin; /* update il(i) in column k+1, ... mbs-1 */ 947 j = bj[jmin]; 948 jl[i] = jl[j]; jl[j] = i; /* update jl */ 949 } 950 i = nexti; 951 } 952 953 /* save nonzero entries in k-th row of U ... */ 954 955 /* invert diagonal block */ 956 diag = ba+k*4; 957 ierr = PetscMemcpy(diag,dk,4*sizeof(MatScalar));CHKERRQ(ierr); 958 ierr = Kernel_A_gets_inverse_A_2(diag,shift);CHKERRQ(ierr); 959 960 jmin = bi[k]; jmax = bi[k+1]; 961 if (jmin < jmax) { 962 for (j=jmin; j<jmax; j++){ 963 vj = bj[j]; /* block col. index of U */ 964 u = ba + j*4; 965 rtmp_ptr = rtmp + vj*4; 966 for (k1=0; k1<4; k1++){ 967 *u++ = *rtmp_ptr; 968 *rtmp_ptr++ = 0.0; 969 } 970 } 971 972 /* ... add k to row list for first nonzero entry in k-th row */ 973 il[k] = jmin; 974 i = bj[jmin]; 975 jl[k] = jl[i]; jl[i] = k; 976 } 977 } 978 979 ierr = PetscFree(rtmp);CHKERRQ(ierr); 980 ierr = PetscFree(il);CHKERRQ(ierr); 981 ierr = PetscFree(dk);CHKERRQ(ierr); 982 if (a->permute) { 983 ierr = PetscFree(aa);CHKERRQ(ierr); 984 } 985 ierr = ISRestoreIndices(perm,&perm_ptr);CHKERRQ(ierr); 986 C->factor = MAT_FACTOR_CHOLESKY; 987 C->assembled = PETSC_TRUE; 988 C->preallocated = PETSC_TRUE; 989 ierr = PetscLogFlops(1.3333*8*b->mbs);CHKERRQ(ierr); /* from inverting diagonal blocks */ 990 PetscFunctionReturn(0); 991 } 992 993 /* 994 Version for when blocks are 2 by 2 Using natural ordering 995 */ 996 #undef __FUNCT__ 997 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_2_NaturalOrdering" 998 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_2_NaturalOrdering(Mat A,MatFactorInfo *info,Mat *B) 999 { 1000 Mat C = *B; 1001 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)A->data,*b = (Mat_SeqSBAIJ *)C->data; 1002 PetscErrorCode ierr; 1003 PetscInt i,j,mbs=a->mbs,*bi=b->i,*bj=b->j; 1004 PetscInt *ai,*aj,k,k1,jmin,jmax,*jl,*il,vj,nexti,ili; 1005 MatScalar *ba = b->a,*aa,*ap,*dk,*uik; 1006 MatScalar *u,*diag,*rtmp,*rtmp_ptr; 1007 PetscReal shift = info->shiftinblocks; 1008 1009 PetscFunctionBegin; 1010 /* initialization */ 1011 /* il and jl record the first nonzero element in each row of the accessing 1012 window U(0:k, k:mbs-1). 1013 jl: list of rows to be added to uneliminated rows 1014 i>= k: jl(i) is the first row to be added to row i 1015 i< k: jl(i) is the row following row i in some list of rows 1016 jl(i) = mbs indicates the end of a list 1017 il(i): points to the first nonzero element in columns k,...,mbs-1 of 1018 row i of U */ 1019 ierr = PetscMalloc(4*mbs*sizeof(MatScalar),&rtmp);CHKERRQ(ierr); 1020 ierr = PetscMemzero(rtmp,4*mbs*sizeof(MatScalar));CHKERRQ(ierr); 1021 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&il);CHKERRQ(ierr); 1022 jl = il + mbs; 1023 for (i=0; i<mbs; i++) { 1024 jl[i] = mbs; il[0] = 0; 1025 } 1026 ierr = PetscMalloc(8*sizeof(MatScalar),&dk);CHKERRQ(ierr); 1027 uik = dk + 4; 1028 1029 ai = a->i; aj = a->j; aa = a->a; 1030 1031 /* for each row k */ 1032 for (k = 0; k<mbs; k++){ 1033 1034 /*initialize k-th row with elements nonzero in row k of A */ 1035 jmin = ai[k]; jmax = ai[k+1]; 1036 ap = aa + jmin*4; 1037 for (j = jmin; j < jmax; j++){ 1038 vj = aj[j]; /* block col. index */ 1039 rtmp_ptr = rtmp + vj*4; 1040 for (i=0; i<4; i++) *rtmp_ptr++ = *ap++; 1041 } 1042 1043 /* modify k-th row by adding in those rows i with U(i,k) != 0 */ 1044 ierr = PetscMemcpy(dk,rtmp+k*4,4*sizeof(MatScalar));CHKERRQ(ierr); 1045 i = jl[k]; /* first row to be added to k_th row */ 1046 1047 while (i < k){ 1048 nexti = jl[i]; /* next row to be added to k_th row */ 1049 1050 /* compute multiplier */ 1051 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 1052 1053 /* uik = -inv(Di)*U_bar(i,k): - ba[ili]*ba[i] */ 1054 diag = ba + i*4; 1055 u = ba + ili*4; 1056 uik[0] = -(diag[0]*u[0] + diag[2]*u[1]); 1057 uik[1] = -(diag[1]*u[0] + diag[3]*u[1]); 1058 uik[2] = -(diag[0]*u[2] + diag[2]*u[3]); 1059 uik[3] = -(diag[1]*u[2] + diag[3]*u[3]); 1060 1061 /* update D(k) += -U(i,k)^T * U_bar(i,k): dk += uik*ba[ili] */ 1062 dk[0] += uik[0]*u[0] + uik[1]*u[1]; 1063 dk[1] += uik[2]*u[0] + uik[3]*u[1]; 1064 dk[2] += uik[0]*u[2] + uik[1]*u[3]; 1065 dk[3] += uik[2]*u[2] + uik[3]*u[3]; 1066 1067 ierr = PetscLogFlops(16*2);CHKERRQ(ierr); 1068 1069 /* update -U(i,k): ba[ili] = uik */ 1070 ierr = PetscMemcpy(ba+ili*4,uik,4*sizeof(MatScalar));CHKERRQ(ierr); 1071 1072 /* add multiple of row i to k-th row ... */ 1073 jmin = ili + 1; jmax = bi[i+1]; 1074 if (jmin < jmax){ 1075 for (j=jmin; j<jmax; j++) { 1076 /* rtmp += -U(i,k)^T * U_bar(i,j): rtmp[bj[j]] += uik*ba[j]; */ 1077 rtmp_ptr = rtmp + bj[j]*4; 1078 u = ba + j*4; 1079 rtmp_ptr[0] += uik[0]*u[0] + uik[1]*u[1]; 1080 rtmp_ptr[1] += uik[2]*u[0] + uik[3]*u[1]; 1081 rtmp_ptr[2] += uik[0]*u[2] + uik[1]*u[3]; 1082 rtmp_ptr[3] += uik[2]*u[2] + uik[3]*u[3]; 1083 } 1084 ierr = PetscLogFlops(16*(jmax-jmin));CHKERRQ(ierr); 1085 1086 /* ... add i to row list for next nonzero entry */ 1087 il[i] = jmin; /* update il(i) in column k+1, ... mbs-1 */ 1088 j = bj[jmin]; 1089 jl[i] = jl[j]; jl[j] = i; /* update jl */ 1090 } 1091 i = nexti; 1092 } 1093 1094 /* save nonzero entries in k-th row of U ... */ 1095 1096 /* invert diagonal block */ 1097 diag = ba+k*4; 1098 ierr = PetscMemcpy(diag,dk,4*sizeof(MatScalar));CHKERRQ(ierr); 1099 ierr = Kernel_A_gets_inverse_A_2(diag,shift);CHKERRQ(ierr); 1100 1101 jmin = bi[k]; jmax = bi[k+1]; 1102 if (jmin < jmax) { 1103 for (j=jmin; j<jmax; j++){ 1104 vj = bj[j]; /* block col. index of U */ 1105 u = ba + j*4; 1106 rtmp_ptr = rtmp + vj*4; 1107 for (k1=0; k1<4; k1++){ 1108 *u++ = *rtmp_ptr; 1109 *rtmp_ptr++ = 0.0; 1110 } 1111 } 1112 1113 /* ... add k to row list for first nonzero entry in k-th row */ 1114 il[k] = jmin; 1115 i = bj[jmin]; 1116 jl[k] = jl[i]; jl[i] = k; 1117 } 1118 } 1119 1120 ierr = PetscFree(rtmp);CHKERRQ(ierr); 1121 ierr = PetscFree(il);CHKERRQ(ierr); 1122 ierr = PetscFree(dk);CHKERRQ(ierr); 1123 1124 C->factor = MAT_FACTOR_CHOLESKY; 1125 C->assembled = PETSC_TRUE; 1126 C->preallocated = PETSC_TRUE; 1127 ierr = PetscLogFlops(1.3333*8*b->mbs);CHKERRQ(ierr); /* from inverting diagonal blocks */ 1128 PetscFunctionReturn(0); 1129 } 1130 1131 /* 1132 Numeric U^T*D*U factorization for SBAIJ format. 1133 Version for blocks are 1 by 1. 1134 */ 1135 #undef __FUNCT__ 1136 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_1" 1137 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_1(Mat A,MatFactorInfo *info,Mat *B) 1138 { 1139 Mat C = *B; 1140 Mat_SeqSBAIJ *a=(Mat_SeqSBAIJ*)A->data,*b=(Mat_SeqSBAIJ *)C->data; 1141 IS ip=b->row; 1142 PetscErrorCode ierr; 1143 PetscInt *rip,i,j,mbs=a->mbs,*bi=b->i,*bj=b->j,*bcol; 1144 PetscInt *ai,*aj,*a2anew; 1145 PetscInt k,jmin,jmax,*jl,*il,col,nexti,ili,nz; 1146 MatScalar *rtmp,*ba=b->a,*bval,*aa,dk,uikdi; 1147 PetscReal zeropivot,rs,shiftnz; 1148 PetscReal shiftpd; 1149 ChShift_Ctx sctx; 1150 PetscInt newshift; 1151 1152 PetscFunctionBegin; 1153 /* initialization */ 1154 shiftnz = info->shiftnz; 1155 shiftpd = info->shiftpd; 1156 zeropivot = info->zeropivot; 1157 1158 ierr = ISGetIndices(ip,&rip);CHKERRQ(ierr); 1159 if (!a->permute){ 1160 ai = a->i; aj = a->j; aa = a->a; 1161 } else { 1162 ai = a->inew; aj = a->jnew; 1163 nz = ai[mbs]; 1164 ierr = PetscMalloc(nz*sizeof(MatScalar),&aa);CHKERRQ(ierr); 1165 a2anew = a->a2anew; 1166 bval = a->a; 1167 for (j=0; j<nz; j++){ 1168 aa[a2anew[j]] = *(bval++); 1169 } 1170 } 1171 1172 /* initialization */ 1173 /* il and jl record the first nonzero element in each row of the accessing 1174 window U(0:k, k:mbs-1). 1175 jl: list of rows to be added to uneliminated rows 1176 i>= k: jl(i) is the first row to be added to row i 1177 i< k: jl(i) is the row following row i in some list of rows 1178 jl(i) = mbs indicates the end of a list 1179 il(i): points to the first nonzero element in columns k,...,mbs-1 of 1180 row i of U */ 1181 nz = (2*mbs+1)*sizeof(PetscInt)+mbs*sizeof(MatScalar); 1182 ierr = PetscMalloc(nz,&il);CHKERRQ(ierr); 1183 jl = il + mbs; 1184 rtmp = (MatScalar*)(jl + mbs); 1185 1186 sctx.shift_amount = 0; 1187 sctx.nshift = 0; 1188 do { 1189 sctx.chshift = PETSC_FALSE; 1190 for (i=0; i<mbs; i++) { 1191 rtmp[i] = 0.0; jl[i] = mbs; il[0] = 0; 1192 } 1193 1194 for (k = 0; k<mbs; k++){ 1195 /*initialize k-th row by the perm[k]-th row of A */ 1196 jmin = ai[rip[k]]; jmax = ai[rip[k]+1]; 1197 bval = ba + bi[k]; 1198 for (j = jmin; j < jmax; j++){ 1199 col = rip[aj[j]]; 1200 rtmp[col] = aa[j]; 1201 *bval++ = 0.0; /* for in-place factorization */ 1202 } 1203 1204 /* shift the diagonal of the matrix */ 1205 if (sctx.nshift) rtmp[k] += sctx.shift_amount; 1206 1207 /* modify k-th row by adding in those rows i with U(i,k)!=0 */ 1208 dk = rtmp[k]; 1209 i = jl[k]; /* first row to be added to k_th row */ 1210 1211 while (i < k){ 1212 nexti = jl[i]; /* next row to be added to k_th row */ 1213 1214 /* compute multiplier, update diag(k) and U(i,k) */ 1215 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 1216 uikdi = - ba[ili]*ba[bi[i]]; /* diagonal(k) */ 1217 dk += uikdi*ba[ili]; 1218 ba[ili] = uikdi; /* -U(i,k) */ 1219 1220 /* add multiple of row i to k-th row */ 1221 jmin = ili + 1; jmax = bi[i+1]; 1222 if (jmin < jmax){ 1223 for (j=jmin; j<jmax; j++) rtmp[bj[j]] += uikdi*ba[j]; 1224 ierr = PetscLogFlops(2*(jmax-jmin));CHKERRQ(ierr); 1225 1226 /* update il and jl for row i */ 1227 il[i] = jmin; 1228 j = bj[jmin]; jl[i] = jl[j]; jl[j] = i; 1229 } 1230 i = nexti; 1231 } 1232 1233 /* shift the diagonals when zero pivot is detected */ 1234 /* compute rs=sum of abs(off-diagonal) */ 1235 rs = 0.0; 1236 jmin = bi[k]+1; 1237 nz = bi[k+1] - jmin; 1238 if (nz){ 1239 bcol = bj + jmin; 1240 while (nz--){ 1241 rs += PetscAbsScalar(rtmp[*bcol]); 1242 bcol++; 1243 } 1244 } 1245 1246 sctx.rs = rs; 1247 sctx.pv = dk; 1248 ierr = MatCholeskyCheckShift_inline(info,sctx,k,newshift);CHKERRQ(ierr); 1249 if (newshift == 1) break; /* sctx.shift_amount is updated */ 1250 1251 /* copy data into U(k,:) */ 1252 ba[bi[k]] = 1.0/dk; /* U(k,k) */ 1253 jmin = bi[k]+1; jmax = bi[k+1]; 1254 if (jmin < jmax) { 1255 for (j=jmin; j<jmax; j++){ 1256 col = bj[j]; ba[j] = rtmp[col]; rtmp[col] = 0.0; 1257 } 1258 /* add the k-th row into il and jl */ 1259 il[k] = jmin; 1260 i = bj[jmin]; jl[k] = jl[i]; jl[i] = k; 1261 } 1262 } 1263 } while (sctx.chshift); 1264 ierr = PetscFree(il);CHKERRQ(ierr); 1265 if (a->permute){ierr = PetscFree(aa);CHKERRQ(ierr);} 1266 1267 ierr = ISRestoreIndices(ip,&rip);CHKERRQ(ierr); 1268 C->factor = MAT_FACTOR_CHOLESKY; 1269 C->assembled = PETSC_TRUE; 1270 C->preallocated = PETSC_TRUE; 1271 ierr = PetscLogFlops(C->rmap.N);CHKERRQ(ierr); 1272 if (sctx.nshift){ 1273 if (shiftnz) { 1274 ierr = PetscInfo2(A,"number of shiftnz tries %D, shift_amount %G\n",sctx.nshift,sctx.shift_amount);CHKERRQ(ierr); 1275 } else if (shiftpd) { 1276 ierr = PetscInfo2(A,"number of shiftpd tries %D, shift_amount %G\n",sctx.nshift,sctx.shift_amount);CHKERRQ(ierr); 1277 } 1278 } 1279 PetscFunctionReturn(0); 1280 } 1281 1282 /* 1283 Version for when blocks are 1 by 1 Using natural ordering 1284 */ 1285 #undef __FUNCT__ 1286 #define __FUNCT__ "MatCholeskyFactorNumeric_SeqSBAIJ_1_NaturalOrdering" 1287 PetscErrorCode MatCholeskyFactorNumeric_SeqSBAIJ_1_NaturalOrdering(Mat A,MatFactorInfo *info,Mat *B) 1288 { 1289 Mat C = *B; 1290 Mat_SeqSBAIJ *a=(Mat_SeqSBAIJ*)A->data,*b=(Mat_SeqSBAIJ *)C->data; 1291 PetscErrorCode ierr; 1292 PetscInt i,j,mbs = a->mbs; 1293 PetscInt *ai=a->i,*aj=a->j,*bi=b->i,*bj=b->j; 1294 PetscInt k,jmin,*jl,*il,nexti,ili,*acol,*bcol,nz; 1295 MatScalar *rtmp,*ba=b->a,*aa=a->a,dk,uikdi,*aval,*bval; 1296 PetscReal zeropivot,rs,shiftnz; 1297 PetscReal shiftpd; 1298 ChShift_Ctx sctx; 1299 PetscInt newshift; 1300 1301 PetscFunctionBegin; 1302 /* initialization */ 1303 shiftnz = info->shiftnz; 1304 shiftpd = info->shiftpd; 1305 zeropivot = info->zeropivot; 1306 1307 /* il and jl record the first nonzero element in each row of the accessing 1308 window U(0:k, k:mbs-1). 1309 jl: list of rows to be added to uneliminated rows 1310 i>= k: jl(i) is the first row to be added to row i 1311 i< k: jl(i) is the row following row i in some list of rows 1312 jl(i) = mbs indicates the end of a list 1313 il(i): points to the first nonzero element in U(i,k:mbs-1) 1314 */ 1315 ierr = PetscMalloc(mbs*sizeof(MatScalar),&rtmp);CHKERRQ(ierr); 1316 ierr = PetscMalloc(2*mbs*sizeof(PetscInt),&il);CHKERRQ(ierr); 1317 jl = il + mbs; 1318 1319 sctx.shift_amount = 0; 1320 sctx.nshift = 0; 1321 do { 1322 sctx.chshift = PETSC_FALSE; 1323 for (i=0; i<mbs; i++) { 1324 rtmp[i] = 0.0; jl[i] = mbs; il[0] = 0; 1325 } 1326 1327 for (k = 0; k<mbs; k++){ 1328 /*initialize k-th row with elements nonzero in row perm(k) of A */ 1329 nz = ai[k+1] - ai[k]; 1330 acol = aj + ai[k]; 1331 aval = aa + ai[k]; 1332 bval = ba + bi[k]; 1333 while (nz -- ){ 1334 rtmp[*acol++] = *aval++; 1335 *bval++ = 0.0; /* for in-place factorization */ 1336 } 1337 1338 /* shift the diagonal of the matrix */ 1339 if (sctx.nshift) rtmp[k] += sctx.shift_amount; 1340 1341 /* modify k-th row by adding in those rows i with U(i,k)!=0 */ 1342 dk = rtmp[k]; 1343 i = jl[k]; /* first row to be added to k_th row */ 1344 1345 while (i < k){ 1346 nexti = jl[i]; /* next row to be added to k_th row */ 1347 /* compute multiplier, update D(k) and U(i,k) */ 1348 ili = il[i]; /* index of first nonzero element in U(i,k:bms-1) */ 1349 uikdi = - ba[ili]*ba[bi[i]]; 1350 dk += uikdi*ba[ili]; 1351 ba[ili] = uikdi; /* -U(i,k) */ 1352 1353 /* add multiple of row i to k-th row ... */ 1354 jmin = ili + 1; 1355 nz = bi[i+1] - jmin; 1356 if (nz > 0){ 1357 bcol = bj + jmin; 1358 bval = ba + jmin; 1359 ierr = PetscLogFlops(2*nz);CHKERRQ(ierr); 1360 while (nz --) rtmp[*bcol++] += uikdi*(*bval++); 1361 1362 /* update il and jl for i-th row */ 1363 il[i] = jmin; 1364 j = bj[jmin]; jl[i] = jl[j]; jl[j] = i; 1365 } 1366 i = nexti; 1367 } 1368 1369 /* shift the diagonals when zero pivot is detected */ 1370 /* compute rs=sum of abs(off-diagonal) */ 1371 rs = 0.0; 1372 jmin = bi[k]+1; 1373 nz = bi[k+1] - jmin; 1374 if (nz){ 1375 bcol = bj + jmin; 1376 while (nz--){ 1377 rs += PetscAbsScalar(rtmp[*bcol]); 1378 bcol++; 1379 } 1380 } 1381 1382 sctx.rs = rs; 1383 sctx.pv = dk; 1384 ierr = MatCholeskyCheckShift_inline(info,sctx,k,newshift);CHKERRQ(ierr); 1385 if (newshift == 1) break; /* sctx.shift_amount is updated */ 1386 1387 /* copy data into U(k,:) */ 1388 ba[bi[k]] = 1.0/dk; 1389 jmin = bi[k]+1; 1390 nz = bi[k+1] - jmin; 1391 if (nz){ 1392 bcol = bj + jmin; 1393 bval = ba + jmin; 1394 while (nz--){ 1395 *bval++ = rtmp[*bcol]; 1396 rtmp[*bcol++] = 0.0; 1397 } 1398 /* add k-th row into il and jl */ 1399 il[k] = jmin; 1400 i = bj[jmin]; jl[k] = jl[i]; jl[i] = k; 1401 } 1402 } /* end of for (k = 0; k<mbs; k++) */ 1403 } while (sctx.chshift); 1404 ierr = PetscFree(rtmp);CHKERRQ(ierr); 1405 ierr = PetscFree(il);CHKERRQ(ierr); 1406 1407 C->factor = MAT_FACTOR_CHOLESKY; 1408 C->assembled = PETSC_TRUE; 1409 C->preallocated = PETSC_TRUE; 1410 ierr = PetscLogFlops(C->rmap.N);CHKERRQ(ierr); 1411 if (sctx.nshift){ 1412 if (shiftnz) { 1413 ierr = PetscInfo2(A,"number of shiftnz tries %D, shift_amount %G\n",sctx.nshift,sctx.shift_amount);CHKERRQ(ierr); 1414 } else if (shiftpd) { 1415 ierr = PetscInfo2(A,"number of shiftpd tries %D, shift_amount %G\n",sctx.nshift,sctx.shift_amount);CHKERRQ(ierr); 1416 } 1417 } 1418 PetscFunctionReturn(0); 1419 } 1420 1421 #undef __FUNCT__ 1422 #define __FUNCT__ "MatCholeskyFactor_SeqSBAIJ" 1423 PetscErrorCode MatCholeskyFactor_SeqSBAIJ(Mat A,IS perm,MatFactorInfo *info) 1424 { 1425 PetscErrorCode ierr; 1426 Mat C; 1427 1428 PetscFunctionBegin; 1429 ierr = MatCholeskyFactorSymbolic(A,perm,info,&C);CHKERRQ(ierr); 1430 ierr = MatCholeskyFactorNumeric(A,info,&C);CHKERRQ(ierr); 1431 ierr = MatHeaderCopy(A,C);CHKERRQ(ierr); 1432 PetscFunctionReturn(0); 1433 } 1434 1435 1436