1 2 #include <../src/mat/impls/aij/mpi/mpiaij.h> /*I "petscmat.h" I*/ 3 #include <petscblaslapack.h> 4 5 #undef __FUNCT__ 6 #define __FUNCT__ "MatFindNonZeroRows_MPIAIJ" 7 PetscErrorCode MatFindNonZeroRows_MPIAIJ(Mat M,IS *keptrows) 8 { 9 PetscErrorCode ierr; 10 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)M->data; 11 Mat_SeqAIJ *a = (Mat_SeqAIJ*)mat->A->data; 12 Mat_SeqAIJ *b = (Mat_SeqAIJ*)mat->B->data; 13 const PetscInt *ia,*ib; 14 const MatScalar *aa,*bb; 15 PetscInt na,nb,i,j,*rows,cnt=0,n0rows; 16 PetscInt m = M->rmap->n,rstart = M->rmap->rstart; 17 18 PetscFunctionBegin; 19 *keptrows = 0; 20 ia = a->i; 21 ib = b->i; 22 for (i=0; i<m; i++) { 23 na = ia[i+1] - ia[i]; 24 nb = ib[i+1] - ib[i]; 25 if (!na && !nb) { 26 cnt++; 27 goto ok1; 28 } 29 aa = a->a + ia[i]; 30 for (j=0; j<na; j++) { 31 if (aa[j] != 0.0) goto ok1; 32 } 33 bb = b->a + ib[i]; 34 for (j=0; j <nb; j++) { 35 if (bb[j] != 0.0) goto ok1; 36 } 37 cnt++; 38 ok1:; 39 } 40 ierr = MPI_Allreduce(&cnt,&n0rows,1,MPIU_INT,MPI_SUM,((PetscObject)M)->comm);CHKERRQ(ierr); 41 if (!n0rows) PetscFunctionReturn(0); 42 ierr = PetscMalloc((M->rmap->n-cnt)*sizeof(PetscInt),&rows);CHKERRQ(ierr); 43 cnt = 0; 44 for (i=0; i<m; i++) { 45 na = ia[i+1] - ia[i]; 46 nb = ib[i+1] - ib[i]; 47 if (!na && !nb) continue; 48 aa = a->a + ia[i]; 49 for(j=0; j<na;j++) { 50 if (aa[j] != 0.0) { 51 rows[cnt++] = rstart + i; 52 goto ok2; 53 } 54 } 55 bb = b->a + ib[i]; 56 for (j=0; j<nb; j++) { 57 if (bb[j] != 0.0) { 58 rows[cnt++] = rstart + i; 59 goto ok2; 60 } 61 } 62 ok2:; 63 } 64 ierr = ISCreateGeneral(PETSC_COMM_WORLD,cnt,rows,PETSC_OWN_POINTER,keptrows);CHKERRQ(ierr); 65 PetscFunctionReturn(0); 66 } 67 68 #undef __FUNCT__ 69 #define __FUNCT__ "MatDistribute_MPIAIJ" 70 /* 71 Distributes a SeqAIJ matrix across a set of processes. Code stolen from 72 MatLoad_MPIAIJ(). Horrible lack of reuse. Should be a routine for each matrix type. 73 74 Only for square matrices 75 */ 76 PetscErrorCode MatDistribute_MPIAIJ(MPI_Comm comm,Mat gmat,PetscInt m,MatReuse reuse,Mat *inmat) 77 { 78 PetscMPIInt rank,size; 79 PetscInt *rowners,*dlens,*olens,i,rstart,rend,j,jj,nz,*gmataj,cnt,row,*ld; 80 PetscErrorCode ierr; 81 Mat mat; 82 Mat_SeqAIJ *gmata; 83 PetscMPIInt tag; 84 MPI_Status status; 85 PetscBool aij; 86 MatScalar *gmataa,*ao,*ad,*gmataarestore=0; 87 88 PetscFunctionBegin; 89 CHKMEMQ; 90 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 91 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 92 if (!rank) { 93 ierr = PetscTypeCompare((PetscObject)gmat,MATSEQAIJ,&aij);CHKERRQ(ierr); 94 if (!aij) SETERRQ1(((PetscObject)gmat)->comm,PETSC_ERR_SUP,"Currently no support for input matrix of type %s\n",((PetscObject)gmat)->type_name); 95 } 96 if (reuse == MAT_INITIAL_MATRIX) { 97 ierr = MatCreate(comm,&mat);CHKERRQ(ierr); 98 ierr = MatSetSizes(mat,m,m,PETSC_DETERMINE,PETSC_DETERMINE);CHKERRQ(ierr); 99 ierr = MatSetType(mat,MATAIJ);CHKERRQ(ierr); 100 ierr = PetscMalloc((size+1)*sizeof(PetscInt),&rowners);CHKERRQ(ierr); 101 ierr = PetscMalloc2(m,PetscInt,&dlens,m,PetscInt,&olens);CHKERRQ(ierr); 102 ierr = MPI_Allgather(&m,1,MPIU_INT,rowners+1,1,MPIU_INT,comm);CHKERRQ(ierr); 103 rowners[0] = 0; 104 for (i=2; i<=size; i++) { 105 rowners[i] += rowners[i-1]; 106 } 107 rstart = rowners[rank]; 108 rend = rowners[rank+1]; 109 ierr = PetscObjectGetNewTag((PetscObject)mat,&tag);CHKERRQ(ierr); 110 if (!rank) { 111 gmata = (Mat_SeqAIJ*) gmat->data; 112 /* send row lengths to all processors */ 113 for (i=0; i<m; i++) dlens[i] = gmata->ilen[i]; 114 for (i=1; i<size; i++) { 115 ierr = MPI_Send(gmata->ilen + rowners[i],rowners[i+1]-rowners[i],MPIU_INT,i,tag,comm);CHKERRQ(ierr); 116 } 117 /* determine number diagonal and off-diagonal counts */ 118 ierr = PetscMemzero(olens,m*sizeof(PetscInt));CHKERRQ(ierr); 119 ierr = PetscMalloc(m*sizeof(PetscInt),&ld);CHKERRQ(ierr); 120 ierr = PetscMemzero(ld,m*sizeof(PetscInt));CHKERRQ(ierr); 121 jj = 0; 122 for (i=0; i<m; i++) { 123 for (j=0; j<dlens[i]; j++) { 124 if (gmata->j[jj] < rstart) ld[i]++; 125 if (gmata->j[jj] < rstart || gmata->j[jj] >= rend) olens[i]++; 126 jj++; 127 } 128 } 129 /* send column indices to other processes */ 130 for (i=1; i<size; i++) { 131 nz = gmata->i[rowners[i+1]]-gmata->i[rowners[i]]; 132 ierr = MPI_Send(&nz,1,MPIU_INT,i,tag,comm);CHKERRQ(ierr); 133 ierr = MPI_Send(gmata->j + gmata->i[rowners[i]],nz,MPIU_INT,i,tag,comm);CHKERRQ(ierr); 134 } 135 136 /* send numerical values to other processes */ 137 for (i=1; i<size; i++) { 138 nz = gmata->i[rowners[i+1]]-gmata->i[rowners[i]]; 139 ierr = MPI_Send(gmata->a + gmata->i[rowners[i]],nz,MPIU_SCALAR,i,tag,comm);CHKERRQ(ierr); 140 } 141 gmataa = gmata->a; 142 gmataj = gmata->j; 143 144 } else { 145 /* receive row lengths */ 146 ierr = MPI_Recv(dlens,m,MPIU_INT,0,tag,comm,&status);CHKERRQ(ierr); 147 /* receive column indices */ 148 ierr = MPI_Recv(&nz,1,MPIU_INT,0,tag,comm,&status);CHKERRQ(ierr); 149 ierr = PetscMalloc2(nz,PetscScalar,&gmataa,nz,PetscInt,&gmataj);CHKERRQ(ierr); 150 ierr = MPI_Recv(gmataj,nz,MPIU_INT,0,tag,comm,&status);CHKERRQ(ierr); 151 /* determine number diagonal and off-diagonal counts */ 152 ierr = PetscMemzero(olens,m*sizeof(PetscInt));CHKERRQ(ierr); 153 ierr = PetscMalloc(m*sizeof(PetscInt),&ld);CHKERRQ(ierr); 154 ierr = PetscMemzero(ld,m*sizeof(PetscInt));CHKERRQ(ierr); 155 jj = 0; 156 for (i=0; i<m; i++) { 157 for (j=0; j<dlens[i]; j++) { 158 if (gmataj[jj] < rstart) ld[i]++; 159 if (gmataj[jj] < rstart || gmataj[jj] >= rend) olens[i]++; 160 jj++; 161 } 162 } 163 /* receive numerical values */ 164 ierr = PetscMemzero(gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); 165 ierr = MPI_Recv(gmataa,nz,MPIU_SCALAR,0,tag,comm,&status);CHKERRQ(ierr); 166 } 167 /* set preallocation */ 168 for (i=0; i<m; i++) { 169 dlens[i] -= olens[i]; 170 } 171 ierr = MatSeqAIJSetPreallocation(mat,0,dlens);CHKERRQ(ierr); 172 ierr = MatMPIAIJSetPreallocation(mat,0,dlens,0,olens);CHKERRQ(ierr); 173 174 for (i=0; i<m; i++) { 175 dlens[i] += olens[i]; 176 } 177 cnt = 0; 178 for (i=0; i<m; i++) { 179 row = rstart + i; 180 ierr = MatSetValues(mat,1,&row,dlens[i],gmataj+cnt,gmataa+cnt,INSERT_VALUES);CHKERRQ(ierr); 181 cnt += dlens[i]; 182 } 183 if (rank) { 184 ierr = PetscFree2(gmataa,gmataj);CHKERRQ(ierr); 185 } 186 ierr = PetscFree2(dlens,olens);CHKERRQ(ierr); 187 ierr = PetscFree(rowners);CHKERRQ(ierr); 188 ((Mat_MPIAIJ*)(mat->data))->ld = ld; 189 *inmat = mat; 190 } else { /* column indices are already set; only need to move over numerical values from process 0 */ 191 Mat_SeqAIJ *Ad = (Mat_SeqAIJ*)((Mat_MPIAIJ*)((*inmat)->data))->A->data; 192 Mat_SeqAIJ *Ao = (Mat_SeqAIJ*)((Mat_MPIAIJ*)((*inmat)->data))->B->data; 193 mat = *inmat; 194 ierr = PetscObjectGetNewTag((PetscObject)mat,&tag);CHKERRQ(ierr); 195 if (!rank) { 196 /* send numerical values to other processes */ 197 gmata = (Mat_SeqAIJ*) gmat->data; 198 ierr = MatGetOwnershipRanges(mat,(const PetscInt**)&rowners);CHKERRQ(ierr); 199 gmataa = gmata->a; 200 for (i=1; i<size; i++) { 201 nz = gmata->i[rowners[i+1]]-gmata->i[rowners[i]]; 202 ierr = MPI_Send(gmataa + gmata->i[rowners[i]],nz,MPIU_SCALAR,i,tag,comm);CHKERRQ(ierr); 203 } 204 nz = gmata->i[rowners[1]]-gmata->i[rowners[0]]; 205 } else { 206 /* receive numerical values from process 0*/ 207 nz = Ad->nz + Ao->nz; 208 ierr = PetscMalloc(nz*sizeof(PetscScalar),&gmataa);CHKERRQ(ierr); gmataarestore = gmataa; 209 ierr = MPI_Recv(gmataa,nz,MPIU_SCALAR,0,tag,comm,&status);CHKERRQ(ierr); 210 } 211 /* transfer numerical values into the diagonal A and off diagonal B parts of mat */ 212 ld = ((Mat_MPIAIJ*)(mat->data))->ld; 213 ad = Ad->a; 214 ao = Ao->a; 215 if (mat->rmap->n) { 216 i = 0; 217 nz = ld[i]; ierr = PetscMemcpy(ao,gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); ao += nz; gmataa += nz; 218 nz = Ad->i[i+1] - Ad->i[i]; ierr = PetscMemcpy(ad,gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); ad += nz; gmataa += nz; 219 } 220 for (i=1; i<mat->rmap->n; i++) { 221 nz = Ao->i[i] - Ao->i[i-1] - ld[i-1] + ld[i]; ierr = PetscMemcpy(ao,gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); ao += nz; gmataa += nz; 222 nz = Ad->i[i+1] - Ad->i[i]; ierr = PetscMemcpy(ad,gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); ad += nz; gmataa += nz; 223 } 224 i--; 225 if (mat->rmap->n) { 226 nz = Ao->i[i+1] - Ao->i[i] - ld[i]; ierr = PetscMemcpy(ao,gmataa,nz*sizeof(PetscScalar));CHKERRQ(ierr); ao += nz; gmataa += nz; 227 } 228 if (rank) { 229 ierr = PetscFree(gmataarestore);CHKERRQ(ierr); 230 } 231 } 232 ierr = MatAssemblyBegin(mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 233 ierr = MatAssemblyEnd(mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 234 CHKMEMQ; 235 PetscFunctionReturn(0); 236 } 237 238 /* 239 Local utility routine that creates a mapping from the global column 240 number to the local number in the off-diagonal part of the local 241 storage of the matrix. When PETSC_USE_CTABLE is used this is scalable at 242 a slightly higher hash table cost; without it it is not scalable (each processor 243 has an order N integer array but is fast to acess. 244 */ 245 #undef __FUNCT__ 246 #define __FUNCT__ "CreateColmap_MPIAIJ_Private" 247 PetscErrorCode CreateColmap_MPIAIJ_Private(Mat mat) 248 { 249 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 250 PetscErrorCode ierr; 251 PetscInt n = aij->B->cmap->n,i; 252 253 PetscFunctionBegin; 254 #if defined (PETSC_USE_CTABLE) 255 ierr = PetscTableCreate(n,&aij->colmap);CHKERRQ(ierr); 256 for (i=0; i<n; i++){ 257 ierr = PetscTableAdd(aij->colmap,aij->garray[i]+1,i+1);CHKERRQ(ierr); 258 } 259 #else 260 ierr = PetscMalloc((mat->cmap->N+1)*sizeof(PetscInt),&aij->colmap);CHKERRQ(ierr); 261 ierr = PetscLogObjectMemory(mat,mat->cmap->N*sizeof(PetscInt));CHKERRQ(ierr); 262 ierr = PetscMemzero(aij->colmap,mat->cmap->N*sizeof(PetscInt));CHKERRQ(ierr); 263 for (i=0; i<n; i++) aij->colmap[aij->garray[i]] = i+1; 264 #endif 265 PetscFunctionReturn(0); 266 } 267 268 #define MatSetValues_SeqAIJ_A_Private(row,col,value,addv) \ 269 { \ 270 if (col <= lastcol1) low1 = 0; else high1 = nrow1; \ 271 lastcol1 = col;\ 272 while (high1-low1 > 5) { \ 273 t = (low1+high1)/2; \ 274 if (rp1[t] > col) high1 = t; \ 275 else low1 = t; \ 276 } \ 277 for (_i=low1; _i<high1; _i++) { \ 278 if (rp1[_i] > col) break; \ 279 if (rp1[_i] == col) { \ 280 if (addv == ADD_VALUES) ap1[_i] += value; \ 281 else ap1[_i] = value; \ 282 goto a_noinsert; \ 283 } \ 284 } \ 285 if (value == 0.0 && ignorezeroentries) {low1 = 0; high1 = nrow1;goto a_noinsert;} \ 286 if (nonew == 1) {low1 = 0; high1 = nrow1; goto a_noinsert;} \ 287 if (nonew == -1) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 288 MatSeqXAIJReallocateAIJ(A,am,1,nrow1,row,col,rmax1,aa,ai,aj,rp1,ap1,aimax,nonew,MatScalar); \ 289 N = nrow1++ - 1; a->nz++; high1++; \ 290 /* shift up all the later entries in this row */ \ 291 for (ii=N; ii>=_i; ii--) { \ 292 rp1[ii+1] = rp1[ii]; \ 293 ap1[ii+1] = ap1[ii]; \ 294 } \ 295 rp1[_i] = col; \ 296 ap1[_i] = value; \ 297 a_noinsert: ; \ 298 ailen[row] = nrow1; \ 299 } 300 301 302 #define MatSetValues_SeqAIJ_B_Private(row,col,value,addv) \ 303 { \ 304 if (col <= lastcol2) low2 = 0; else high2 = nrow2; \ 305 lastcol2 = col;\ 306 while (high2-low2 > 5) { \ 307 t = (low2+high2)/2; \ 308 if (rp2[t] > col) high2 = t; \ 309 else low2 = t; \ 310 } \ 311 for (_i=low2; _i<high2; _i++) { \ 312 if (rp2[_i] > col) break; \ 313 if (rp2[_i] == col) { \ 314 if (addv == ADD_VALUES) ap2[_i] += value; \ 315 else ap2[_i] = value; \ 316 goto b_noinsert; \ 317 } \ 318 } \ 319 if (value == 0.0 && ignorezeroentries) {low2 = 0; high2 = nrow2; goto b_noinsert;} \ 320 if (nonew == 1) {low2 = 0; high2 = nrow2; goto b_noinsert;} \ 321 if (nonew == -1) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 322 MatSeqXAIJReallocateAIJ(B,bm,1,nrow2,row,col,rmax2,ba,bi,bj,rp2,ap2,bimax,nonew,MatScalar); \ 323 N = nrow2++ - 1; b->nz++; high2++; \ 324 /* shift up all the later entries in this row */ \ 325 for (ii=N; ii>=_i; ii--) { \ 326 rp2[ii+1] = rp2[ii]; \ 327 ap2[ii+1] = ap2[ii]; \ 328 } \ 329 rp2[_i] = col; \ 330 ap2[_i] = value; \ 331 b_noinsert: ; \ 332 bilen[row] = nrow2; \ 333 } 334 335 #undef __FUNCT__ 336 #define __FUNCT__ "MatSetValuesRow_MPIAIJ" 337 PetscErrorCode MatSetValuesRow_MPIAIJ(Mat A,PetscInt row,const PetscScalar v[]) 338 { 339 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)A->data; 340 Mat_SeqAIJ *a = (Mat_SeqAIJ*)mat->A->data,*b = (Mat_SeqAIJ*)mat->B->data; 341 PetscErrorCode ierr; 342 PetscInt l,*garray = mat->garray,diag; 343 344 PetscFunctionBegin; 345 /* code only works for square matrices A */ 346 347 /* find size of row to the left of the diagonal part */ 348 ierr = MatGetOwnershipRange(A,&diag,0);CHKERRQ(ierr); 349 row = row - diag; 350 for (l=0; l<b->i[row+1]-b->i[row]; l++) { 351 if (garray[b->j[b->i[row]+l]] > diag) break; 352 } 353 ierr = PetscMemcpy(b->a+b->i[row],v,l*sizeof(PetscScalar));CHKERRQ(ierr); 354 355 /* diagonal part */ 356 ierr = PetscMemcpy(a->a+a->i[row],v+l,(a->i[row+1]-a->i[row])*sizeof(PetscScalar));CHKERRQ(ierr); 357 358 /* right of diagonal part */ 359 ierr = PetscMemcpy(b->a+b->i[row]+l,v+l+a->i[row+1]-a->i[row],(b->i[row+1]-b->i[row]-l)*sizeof(PetscScalar));CHKERRQ(ierr); 360 PetscFunctionReturn(0); 361 } 362 363 #undef __FUNCT__ 364 #define __FUNCT__ "MatSetValues_MPIAIJ" 365 PetscErrorCode MatSetValues_MPIAIJ(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const PetscScalar v[],InsertMode addv) 366 { 367 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 368 PetscScalar value; 369 PetscErrorCode ierr; 370 PetscInt i,j,rstart = mat->rmap->rstart,rend = mat->rmap->rend; 371 PetscInt cstart = mat->cmap->rstart,cend = mat->cmap->rend,row,col; 372 PetscBool roworiented = aij->roworiented; 373 374 /* Some Variables required in the macro */ 375 Mat A = aij->A; 376 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 377 PetscInt *aimax = a->imax,*ai = a->i,*ailen = a->ilen,*aj = a->j; 378 MatScalar *aa = a->a; 379 PetscBool ignorezeroentries = a->ignorezeroentries; 380 Mat B = aij->B; 381 Mat_SeqAIJ *b = (Mat_SeqAIJ*)B->data; 382 PetscInt *bimax = b->imax,*bi = b->i,*bilen = b->ilen,*bj = b->j,bm = aij->B->rmap->n,am = aij->A->rmap->n; 383 MatScalar *ba = b->a; 384 385 PetscInt *rp1,*rp2,ii,nrow1,nrow2,_i,rmax1,rmax2,N,low1,high1,low2,high2,t,lastcol1,lastcol2; 386 PetscInt nonew = a->nonew; 387 MatScalar *ap1,*ap2; 388 389 PetscFunctionBegin; 390 if (v) PetscValidScalarPointer(v,6); 391 for (i=0; i<m; i++) { 392 if (im[i] < 0) continue; 393 #if defined(PETSC_USE_DEBUG) 394 if (im[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",im[i],mat->rmap->N-1); 395 #endif 396 if (im[i] >= rstart && im[i] < rend) { 397 row = im[i] - rstart; 398 lastcol1 = -1; 399 rp1 = aj + ai[row]; 400 ap1 = aa + ai[row]; 401 rmax1 = aimax[row]; 402 nrow1 = ailen[row]; 403 low1 = 0; 404 high1 = nrow1; 405 lastcol2 = -1; 406 rp2 = bj + bi[row]; 407 ap2 = ba + bi[row]; 408 rmax2 = bimax[row]; 409 nrow2 = bilen[row]; 410 low2 = 0; 411 high2 = nrow2; 412 413 for (j=0; j<n; j++) { 414 if (v) {if (roworiented) value = v[i*n+j]; else value = v[i+j*m];} else value = 0.0; 415 if (ignorezeroentries && value == 0.0 && (addv == ADD_VALUES)) continue; 416 if (in[j] >= cstart && in[j] < cend){ 417 col = in[j] - cstart; 418 MatSetValues_SeqAIJ_A_Private(row,col,value,addv); 419 } else if (in[j] < 0) continue; 420 #if defined(PETSC_USE_DEBUG) 421 else if (in[j] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",in[j],mat->cmap->N-1); 422 #endif 423 else { 424 if (mat->was_assembled) { 425 if (!aij->colmap) { 426 ierr = CreateColmap_MPIAIJ_Private(mat);CHKERRQ(ierr); 427 } 428 #if defined (PETSC_USE_CTABLE) 429 ierr = PetscTableFind(aij->colmap,in[j]+1,&col);CHKERRQ(ierr); 430 col--; 431 #else 432 col = aij->colmap[in[j]] - 1; 433 #endif 434 if (col < 0 && !((Mat_SeqAIJ*)(aij->A->data))->nonew) { 435 ierr = DisAssemble_MPIAIJ(mat);CHKERRQ(ierr); 436 col = in[j]; 437 /* Reinitialize the variables required by MatSetValues_SeqAIJ_B_Private() */ 438 B = aij->B; 439 b = (Mat_SeqAIJ*)B->data; 440 bimax = b->imax; bi = b->i; bilen = b->ilen; bj = b->j; ba = b->a; 441 rp2 = bj + bi[row]; 442 ap2 = ba + bi[row]; 443 rmax2 = bimax[row]; 444 nrow2 = bilen[row]; 445 low2 = 0; 446 high2 = nrow2; 447 bm = aij->B->rmap->n; 448 ba = b->a; 449 } 450 } else col = in[j]; 451 MatSetValues_SeqAIJ_B_Private(row,col,value,addv); 452 } 453 } 454 } else { 455 if (mat->nooffprocentries) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Setting off process row %D even though MatSetOption(,MAT_NO_OFF_PROC_ENTRIES,PETSC_TRUE) was set",im[i]); 456 if (!aij->donotstash) { 457 if (roworiented) { 458 ierr = MatStashValuesRow_Private(&mat->stash,im[i],n,in,v+i*n,(PetscBool)(ignorezeroentries && (addv == ADD_VALUES)));CHKERRQ(ierr); 459 } else { 460 ierr = MatStashValuesCol_Private(&mat->stash,im[i],n,in,v+i,m,(PetscBool)(ignorezeroentries && (addv == ADD_VALUES)));CHKERRQ(ierr); 461 } 462 } 463 } 464 } 465 PetscFunctionReturn(0); 466 } 467 468 #undef __FUNCT__ 469 #define __FUNCT__ "MatGetValues_MPIAIJ" 470 PetscErrorCode MatGetValues_MPIAIJ(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[]) 471 { 472 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 473 PetscErrorCode ierr; 474 PetscInt i,j,rstart = mat->rmap->rstart,rend = mat->rmap->rend; 475 PetscInt cstart = mat->cmap->rstart,cend = mat->cmap->rend,row,col; 476 477 PetscFunctionBegin; 478 for (i=0; i<m; i++) { 479 if (idxm[i] < 0) continue; /* SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Negative row: %D",idxm[i]);*/ 480 if (idxm[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",idxm[i],mat->rmap->N-1); 481 if (idxm[i] >= rstart && idxm[i] < rend) { 482 row = idxm[i] - rstart; 483 for (j=0; j<n; j++) { 484 if (idxn[j] < 0) continue; /* SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Negative column: %D",idxn[j]); */ 485 if (idxn[j] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",idxn[j],mat->cmap->N-1); 486 if (idxn[j] >= cstart && idxn[j] < cend){ 487 col = idxn[j] - cstart; 488 ierr = MatGetValues(aij->A,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 489 } else { 490 if (!aij->colmap) { 491 ierr = CreateColmap_MPIAIJ_Private(mat);CHKERRQ(ierr); 492 } 493 #if defined (PETSC_USE_CTABLE) 494 ierr = PetscTableFind(aij->colmap,idxn[j]+1,&col);CHKERRQ(ierr); 495 col --; 496 #else 497 col = aij->colmap[idxn[j]] - 1; 498 #endif 499 if ((col < 0) || (aij->garray[col] != idxn[j])) *(v+i*n+j) = 0.0; 500 else { 501 ierr = MatGetValues(aij->B,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 502 } 503 } 504 } 505 } else { 506 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only local values currently supported"); 507 } 508 } 509 PetscFunctionReturn(0); 510 } 511 512 extern PetscErrorCode MatMultDiagonalBlock_MPIAIJ(Mat,Vec,Vec); 513 514 #undef __FUNCT__ 515 #define __FUNCT__ "MatAssemblyBegin_MPIAIJ" 516 PetscErrorCode MatAssemblyBegin_MPIAIJ(Mat mat,MatAssemblyType mode) 517 { 518 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 519 PetscErrorCode ierr; 520 PetscInt nstash,reallocs; 521 InsertMode addv; 522 523 PetscFunctionBegin; 524 if (aij->donotstash || mat->nooffprocentries) { 525 PetscFunctionReturn(0); 526 } 527 528 /* make sure all processors are either in INSERTMODE or ADDMODE */ 529 ierr = MPI_Allreduce(&mat->insertmode,&addv,1,MPI_INT,MPI_BOR,((PetscObject)mat)->comm);CHKERRQ(ierr); 530 if (addv == (ADD_VALUES|INSERT_VALUES)) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Some processors inserted others added"); 531 mat->insertmode = addv; /* in case this processor had no cache */ 532 533 ierr = MatStashScatterBegin_Private(mat,&mat->stash,mat->rmap->range);CHKERRQ(ierr); 534 ierr = MatStashGetInfo_Private(&mat->stash,&nstash,&reallocs);CHKERRQ(ierr); 535 ierr = PetscInfo2(aij->A,"Stash has %D entries, uses %D mallocs.\n",nstash,reallocs);CHKERRQ(ierr); 536 PetscFunctionReturn(0); 537 } 538 539 #undef __FUNCT__ 540 #define __FUNCT__ "MatAssemblyEnd_MPIAIJ" 541 PetscErrorCode MatAssemblyEnd_MPIAIJ(Mat mat,MatAssemblyType mode) 542 { 543 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 544 Mat_SeqAIJ *a=(Mat_SeqAIJ *)aij->A->data; 545 PetscErrorCode ierr; 546 PetscMPIInt n; 547 PetscInt i,j,rstart,ncols,flg; 548 PetscInt *row,*col; 549 PetscBool other_disassembled; 550 PetscScalar *val; 551 InsertMode addv = mat->insertmode; 552 553 /* do not use 'b = (Mat_SeqAIJ *)aij->B->data' as B can be reset in disassembly */ 554 PetscFunctionBegin; 555 if (!aij->donotstash && !mat->nooffprocentries) { 556 while (1) { 557 ierr = MatStashScatterGetMesg_Private(&mat->stash,&n,&row,&col,&val,&flg);CHKERRQ(ierr); 558 if (!flg) break; 559 560 for (i=0; i<n;) { 561 /* Now identify the consecutive vals belonging to the same row */ 562 for (j=i,rstart=row[j]; j<n; j++) { if (row[j] != rstart) break; } 563 if (j < n) ncols = j-i; 564 else ncols = n-i; 565 /* Now assemble all these values with a single function call */ 566 ierr = MatSetValues_MPIAIJ(mat,1,row+i,ncols,col+i,val+i,addv);CHKERRQ(ierr); 567 i = j; 568 } 569 } 570 ierr = MatStashScatterEnd_Private(&mat->stash);CHKERRQ(ierr); 571 } 572 ierr = MatAssemblyBegin(aij->A,mode);CHKERRQ(ierr); 573 ierr = MatAssemblyEnd(aij->A,mode);CHKERRQ(ierr); 574 575 /* determine if any processor has disassembled, if so we must 576 also disassemble ourselfs, in order that we may reassemble. */ 577 /* 578 if nonzero structure of submatrix B cannot change then we know that 579 no processor disassembled thus we can skip this stuff 580 */ 581 if (!((Mat_SeqAIJ*)aij->B->data)->nonew) { 582 ierr = MPI_Allreduce(&mat->was_assembled,&other_disassembled,1,MPI_INT,MPI_PROD,((PetscObject)mat)->comm);CHKERRQ(ierr); 583 if (mat->was_assembled && !other_disassembled) { 584 ierr = DisAssemble_MPIAIJ(mat);CHKERRQ(ierr); 585 } 586 } 587 if (!mat->was_assembled && mode == MAT_FINAL_ASSEMBLY) { 588 ierr = MatSetUpMultiply_MPIAIJ(mat);CHKERRQ(ierr); 589 } 590 ierr = MatSetOption(aij->B,MAT_USE_INODES,PETSC_FALSE);CHKERRQ(ierr); 591 ierr = MatSetOption(aij->B,MAT_CHECK_COMPRESSED_ROW,PETSC_FALSE);CHKERRQ(ierr); 592 ierr = MatAssemblyBegin(aij->B,mode);CHKERRQ(ierr); 593 ierr = MatAssemblyEnd(aij->B,mode);CHKERRQ(ierr); 594 595 ierr = PetscFree2(aij->rowvalues,aij->rowindices);CHKERRQ(ierr); 596 aij->rowvalues = 0; 597 598 /* used by MatAXPY() */ 599 a->xtoy = 0; ((Mat_SeqAIJ *)aij->B->data)->xtoy = 0; /* b->xtoy = 0 */ 600 a->XtoY = 0; ((Mat_SeqAIJ *)aij->B->data)->XtoY = 0; /* b->XtoY = 0 */ 601 602 if (aij->diag) {ierr = VecDestroy(aij->diag);CHKERRQ(ierr);aij->diag = 0;} 603 if (a->inode.size) mat->ops->multdiagonalblock = MatMultDiagonalBlock_MPIAIJ; 604 PetscFunctionReturn(0); 605 } 606 607 #undef __FUNCT__ 608 #define __FUNCT__ "MatZeroEntries_MPIAIJ" 609 PetscErrorCode MatZeroEntries_MPIAIJ(Mat A) 610 { 611 Mat_MPIAIJ *l = (Mat_MPIAIJ*)A->data; 612 PetscErrorCode ierr; 613 614 PetscFunctionBegin; 615 ierr = MatZeroEntries(l->A);CHKERRQ(ierr); 616 ierr = MatZeroEntries(l->B);CHKERRQ(ierr); 617 PetscFunctionReturn(0); 618 } 619 620 #undef __FUNCT__ 621 #define __FUNCT__ "MatZeroRows_MPIAIJ" 622 PetscErrorCode MatZeroRows_MPIAIJ(Mat A,PetscInt N,const PetscInt rows[],PetscScalar diag,Vec x,Vec b) 623 { 624 Mat_MPIAIJ *l = (Mat_MPIAIJ*)A->data; 625 PetscErrorCode ierr; 626 PetscMPIInt size = l->size,imdex,n,rank = l->rank,tag = ((PetscObject)A)->tag,lastidx = -1; 627 PetscInt i,*owners = A->rmap->range; 628 PetscInt *nprocs,j,idx,nsends,row; 629 PetscInt nmax,*svalues,*starts,*owner,nrecvs; 630 PetscInt *rvalues,count,base,slen,*source; 631 PetscInt *lens,*lrows,*values,rstart=A->rmap->rstart; 632 MPI_Comm comm = ((PetscObject)A)->comm; 633 MPI_Request *send_waits,*recv_waits; 634 MPI_Status recv_status,*send_status; 635 const PetscScalar *xx; 636 PetscScalar *bb; 637 #if defined(PETSC_DEBUG) 638 PetscBool found = PETSC_FALSE; 639 #endif 640 641 PetscFunctionBegin; 642 /* first count number of contributors to each processor */ 643 ierr = PetscMalloc(2*size*sizeof(PetscInt),&nprocs);CHKERRQ(ierr); 644 ierr = PetscMemzero(nprocs,2*size*sizeof(PetscInt));CHKERRQ(ierr); 645 ierr = PetscMalloc((N+1)*sizeof(PetscInt),&owner);CHKERRQ(ierr); /* see note*/ 646 j = 0; 647 for (i=0; i<N; i++) { 648 if (lastidx > (idx = rows[i])) j = 0; 649 lastidx = idx; 650 for (; j<size; j++) { 651 if (idx >= owners[j] && idx < owners[j+1]) { 652 nprocs[2*j]++; 653 nprocs[2*j+1] = 1; 654 owner[i] = j; 655 #if defined(PETSC_DEBUG) 656 found = PETSC_TRUE; 657 #endif 658 break; 659 } 660 } 661 #if defined(PETSC_DEBUG) 662 if (!found) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Index out of range"); 663 found = PETSC_FALSE; 664 #endif 665 } 666 nsends = 0; for (i=0; i<size; i++) { nsends += nprocs[2*i+1];} 667 668 if (A->nooffproczerorows) { 669 if (nsends > 1) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"You called MatSetOption(,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) but set an off process zero row"); 670 nrecvs = nsends; 671 nmax = N; 672 } else { 673 /* inform other processors of number of messages and max length*/ 674 ierr = PetscMaxSum(comm,nprocs,&nmax,&nrecvs);CHKERRQ(ierr); 675 } 676 677 /* post receives: */ 678 ierr = PetscMalloc((nrecvs+1)*(nmax+1)*sizeof(PetscInt),&rvalues);CHKERRQ(ierr); 679 ierr = PetscMalloc((nrecvs+1)*sizeof(MPI_Request),&recv_waits);CHKERRQ(ierr); 680 for (i=0; i<nrecvs; i++) { 681 ierr = MPI_Irecv(rvalues+nmax*i,nmax,MPIU_INT,MPI_ANY_SOURCE,tag,comm,recv_waits+i);CHKERRQ(ierr); 682 } 683 684 /* do sends: 685 1) starts[i] gives the starting index in svalues for stuff going to 686 the ith processor 687 */ 688 ierr = PetscMalloc((N+1)*sizeof(PetscInt),&svalues);CHKERRQ(ierr); 689 ierr = PetscMalloc((nsends+1)*sizeof(MPI_Request),&send_waits);CHKERRQ(ierr); 690 ierr = PetscMalloc((size+1)*sizeof(PetscInt),&starts);CHKERRQ(ierr); 691 starts[0] = 0; 692 for (i=1; i<size; i++) { starts[i] = starts[i-1] + nprocs[2*i-2];} 693 for (i=0; i<N; i++) { 694 svalues[starts[owner[i]]++] = rows[i]; 695 } 696 697 starts[0] = 0; 698 for (i=1; i<size+1; i++) { starts[i] = starts[i-1] + nprocs[2*i-2];} 699 count = 0; 700 for (i=0; i<size; i++) { 701 if (nprocs[2*i+1]) { 702 ierr = MPI_Isend(svalues+starts[i],nprocs[2*i],MPIU_INT,i,tag,comm,send_waits+count++);CHKERRQ(ierr); 703 } 704 } 705 ierr = PetscFree(starts);CHKERRQ(ierr); 706 707 base = owners[rank]; 708 709 /* wait on receives */ 710 ierr = PetscMalloc2(nrecvs,PetscInt,&lens,nrecvs,PetscInt,&source);CHKERRQ(ierr); 711 count = nrecvs; slen = 0; 712 while (count) { 713 ierr = MPI_Waitany(nrecvs,recv_waits,&imdex,&recv_status);CHKERRQ(ierr); 714 /* unpack receives into our local space */ 715 ierr = MPI_Get_count(&recv_status,MPIU_INT,&n);CHKERRQ(ierr); 716 source[imdex] = recv_status.MPI_SOURCE; 717 lens[imdex] = n; 718 slen += n; 719 count--; 720 } 721 ierr = PetscFree(recv_waits);CHKERRQ(ierr); 722 723 /* move the data into the send scatter */ 724 ierr = PetscMalloc((slen+1)*sizeof(PetscInt),&lrows);CHKERRQ(ierr); 725 count = 0; 726 for (i=0; i<nrecvs; i++) { 727 values = rvalues + i*nmax; 728 for (j=0; j<lens[i]; j++) { 729 lrows[count++] = values[j] - base; 730 } 731 } 732 ierr = PetscFree(rvalues);CHKERRQ(ierr); 733 ierr = PetscFree2(lens,source);CHKERRQ(ierr); 734 ierr = PetscFree(owner);CHKERRQ(ierr); 735 ierr = PetscFree(nprocs);CHKERRQ(ierr); 736 737 /* fix right hand side if needed */ 738 if (x && b) { 739 ierr = VecGetArrayRead(x,&xx);CHKERRQ(ierr); 740 ierr = VecGetArray(b,&bb);CHKERRQ(ierr); 741 for (i=0; i<slen; i++) { 742 bb[lrows[i]] = diag*xx[lrows[i]]; 743 } 744 ierr = VecRestoreArrayRead(x,&xx);CHKERRQ(ierr); 745 ierr = VecRestoreArray(b,&bb);CHKERRQ(ierr); 746 } 747 /* 748 Zero the required rows. If the "diagonal block" of the matrix 749 is square and the user wishes to set the diagonal we use separate 750 code so that MatSetValues() is not called for each diagonal allocating 751 new memory, thus calling lots of mallocs and slowing things down. 752 753 */ 754 /* must zero l->B before l->A because the (diag) case below may put values into l->B*/ 755 ierr = MatZeroRows(l->B,slen,lrows,0.0,0,0);CHKERRQ(ierr); 756 if ((diag != 0.0) && (l->A->rmap->N == l->A->cmap->N)) { 757 ierr = MatZeroRows(l->A,slen,lrows,diag,0,0);CHKERRQ(ierr); 758 } else if (diag != 0.0) { 759 ierr = MatZeroRows(l->A,slen,lrows,0.0,0,0);CHKERRQ(ierr); 760 if (((Mat_SeqAIJ*)l->A->data)->nonew) { 761 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatZeroRows() on rectangular matrices cannot be used with the Mat options\n\ 762 MAT_NEW_NONZERO_LOCATIONS,MAT_NEW_NONZERO_LOCATION_ERR,MAT_NEW_NONZERO_ALLOCATION_ERR"); 763 } 764 for (i = 0; i < slen; i++) { 765 row = lrows[i] + rstart; 766 ierr = MatSetValues(A,1,&row,1,&row,&diag,INSERT_VALUES);CHKERRQ(ierr); 767 } 768 ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 769 ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 770 } else { 771 ierr = MatZeroRows(l->A,slen,lrows,0.0,0,0);CHKERRQ(ierr); 772 } 773 ierr = PetscFree(lrows);CHKERRQ(ierr); 774 775 /* wait on sends */ 776 if (nsends) { 777 ierr = PetscMalloc(nsends*sizeof(MPI_Status),&send_status);CHKERRQ(ierr); 778 ierr = MPI_Waitall(nsends,send_waits,send_status);CHKERRQ(ierr); 779 ierr = PetscFree(send_status);CHKERRQ(ierr); 780 } 781 ierr = PetscFree(send_waits);CHKERRQ(ierr); 782 ierr = PetscFree(svalues);CHKERRQ(ierr); 783 784 PetscFunctionReturn(0); 785 } 786 787 #undef __FUNCT__ 788 #define __FUNCT__ "MatZeroRowsColumns_MPIAIJ" 789 PetscErrorCode MatZeroRowsColumns_MPIAIJ(Mat A,PetscInt N,const PetscInt rows[],PetscScalar diag,Vec x,Vec b) 790 { 791 Mat_MPIAIJ *l = (Mat_MPIAIJ*)A->data; 792 PetscErrorCode ierr; 793 PetscMPIInt size = l->size,imdex,n,rank = l->rank,tag = ((PetscObject)A)->tag,lastidx = -1; 794 PetscInt i,*owners = A->rmap->range; 795 PetscInt *nprocs,j,idx,nsends; 796 PetscInt nmax,*svalues,*starts,*owner,nrecvs; 797 PetscInt *rvalues,count,base,slen,*source; 798 PetscInt *lens,*lrows,*values,m; 799 MPI_Comm comm = ((PetscObject)A)->comm; 800 MPI_Request *send_waits,*recv_waits; 801 MPI_Status recv_status,*send_status; 802 const PetscScalar *xx; 803 PetscScalar *bb,*mask; 804 Vec xmask,lmask; 805 Mat_SeqAIJ *aij = (Mat_SeqAIJ*)l->B->data; 806 const PetscInt *aj, *ii,*ridx; 807 PetscScalar *aa; 808 #if defined(PETSC_DEBUG) 809 PetscBool found = PETSC_FALSE; 810 #endif 811 812 PetscFunctionBegin; 813 /* first count number of contributors to each processor */ 814 ierr = PetscMalloc(2*size*sizeof(PetscInt),&nprocs);CHKERRQ(ierr); 815 ierr = PetscMemzero(nprocs,2*size*sizeof(PetscInt));CHKERRQ(ierr); 816 ierr = PetscMalloc((N+1)*sizeof(PetscInt),&owner);CHKERRQ(ierr); /* see note*/ 817 j = 0; 818 for (i=0; i<N; i++) { 819 if (lastidx > (idx = rows[i])) j = 0; 820 lastidx = idx; 821 for (; j<size; j++) { 822 if (idx >= owners[j] && idx < owners[j+1]) { 823 nprocs[2*j]++; 824 nprocs[2*j+1] = 1; 825 owner[i] = j; 826 #if defined(PETSC_DEBUG) 827 found = PETSC_TRUE; 828 #endif 829 break; 830 } 831 } 832 #if defined(PETSC_DEBUG) 833 if (!found) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Index out of range"); 834 found = PETSC_FALSE; 835 #endif 836 } 837 nsends = 0; for (i=0; i<size; i++) { nsends += nprocs[2*i+1];} 838 839 /* inform other processors of number of messages and max length*/ 840 ierr = PetscMaxSum(comm,nprocs,&nmax,&nrecvs);CHKERRQ(ierr); 841 842 /* post receives: */ 843 ierr = PetscMalloc((nrecvs+1)*(nmax+1)*sizeof(PetscInt),&rvalues);CHKERRQ(ierr); 844 ierr = PetscMalloc((nrecvs+1)*sizeof(MPI_Request),&recv_waits);CHKERRQ(ierr); 845 for (i=0; i<nrecvs; i++) { 846 ierr = MPI_Irecv(rvalues+nmax*i,nmax,MPIU_INT,MPI_ANY_SOURCE,tag,comm,recv_waits+i);CHKERRQ(ierr); 847 } 848 849 /* do sends: 850 1) starts[i] gives the starting index in svalues for stuff going to 851 the ith processor 852 */ 853 ierr = PetscMalloc((N+1)*sizeof(PetscInt),&svalues);CHKERRQ(ierr); 854 ierr = PetscMalloc((nsends+1)*sizeof(MPI_Request),&send_waits);CHKERRQ(ierr); 855 ierr = PetscMalloc((size+1)*sizeof(PetscInt),&starts);CHKERRQ(ierr); 856 starts[0] = 0; 857 for (i=1; i<size; i++) { starts[i] = starts[i-1] + nprocs[2*i-2];} 858 for (i=0; i<N; i++) { 859 svalues[starts[owner[i]]++] = rows[i]; 860 } 861 862 starts[0] = 0; 863 for (i=1; i<size+1; i++) { starts[i] = starts[i-1] + nprocs[2*i-2];} 864 count = 0; 865 for (i=0; i<size; i++) { 866 if (nprocs[2*i+1]) { 867 ierr = MPI_Isend(svalues+starts[i],nprocs[2*i],MPIU_INT,i,tag,comm,send_waits+count++);CHKERRQ(ierr); 868 } 869 } 870 ierr = PetscFree(starts);CHKERRQ(ierr); 871 872 base = owners[rank]; 873 874 /* wait on receives */ 875 ierr = PetscMalloc2(nrecvs,PetscInt,&lens,nrecvs,PetscInt,&source);CHKERRQ(ierr); 876 count = nrecvs; slen = 0; 877 while (count) { 878 ierr = MPI_Waitany(nrecvs,recv_waits,&imdex,&recv_status);CHKERRQ(ierr); 879 /* unpack receives into our local space */ 880 ierr = MPI_Get_count(&recv_status,MPIU_INT,&n);CHKERRQ(ierr); 881 source[imdex] = recv_status.MPI_SOURCE; 882 lens[imdex] = n; 883 slen += n; 884 count--; 885 } 886 ierr = PetscFree(recv_waits);CHKERRQ(ierr); 887 888 /* move the data into the send scatter */ 889 ierr = PetscMalloc((slen+1)*sizeof(PetscInt),&lrows);CHKERRQ(ierr); 890 count = 0; 891 for (i=0; i<nrecvs; i++) { 892 values = rvalues + i*nmax; 893 for (j=0; j<lens[i]; j++) { 894 lrows[count++] = values[j] - base; 895 } 896 } 897 ierr = PetscFree(rvalues);CHKERRQ(ierr); 898 ierr = PetscFree2(lens,source);CHKERRQ(ierr); 899 ierr = PetscFree(owner);CHKERRQ(ierr); 900 ierr = PetscFree(nprocs);CHKERRQ(ierr); 901 /* lrows are the local rows to be zeroed, slen is the number of local rows */ 902 903 /* zero diagonal part of matrix */ 904 ierr = MatZeroRowsColumns(l->A,slen,lrows,diag,x,b);CHKERRQ(ierr); 905 906 /* handle off diagonal part of matrix */ 907 ierr = MatGetVecs(A,&xmask,PETSC_NULL);CHKERRQ(ierr); 908 ierr = VecDuplicate(l->lvec,&lmask);CHKERRQ(ierr); 909 ierr = VecGetArray(xmask,&bb);CHKERRQ(ierr); 910 for (i=0; i<slen; i++) { 911 bb[lrows[i]] = 1; 912 } 913 ierr = VecRestoreArray(xmask,&bb);CHKERRQ(ierr); 914 ierr = VecScatterBegin(l->Mvctx,xmask,lmask,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 915 ierr = VecScatterEnd(l->Mvctx,xmask,lmask,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 916 ierr = VecDestroy(xmask);CHKERRQ(ierr); 917 ierr = VecScatterBegin(l->Mvctx,x,l->lvec,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 918 ierr = VecScatterEnd(l->Mvctx,x,l->lvec,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 919 ierr = VecGetArrayRead(l->lvec,&xx);CHKERRQ(ierr); 920 ierr = VecGetArray(lmask,&mask);CHKERRQ(ierr); 921 ierr = VecGetArray(b,&bb);CHKERRQ(ierr); 922 923 /* remove zeroed rows of off diagonal matrix */ 924 ii = aij->i; 925 for (i=0; i<slen; i++) { 926 ierr = PetscMemzero(aij->a + ii[lrows[i]],(ii[lrows[i]+1] - ii[lrows[i]])*sizeof(PetscScalar));CHKERRQ(ierr); 927 } 928 929 /* loop over all elements of off process part of matrix zeroing removed columns*/ 930 if (aij->compressedrow.use){ 931 m = aij->compressedrow.nrows; 932 ii = aij->compressedrow.i; 933 ridx = aij->compressedrow.rindex; 934 for (i=0; i<m; i++){ 935 n = ii[i+1] - ii[i]; 936 aj = aij->j + ii[i]; 937 aa = aij->a + ii[i]; 938 939 for (j=0; j<n; j++) { 940 if (PetscAbsScalar(mask[*aj])) { 941 bb[*ridx] -= *aa*xx[*aj]; 942 *aa = 0.0; 943 } 944 aa++; 945 aj++; 946 } 947 ridx++; 948 } 949 } else { /* do not use compressed row format */ 950 m = l->B->rmap->n; 951 for (i=0; i<m; i++) { 952 n = ii[i+1] - ii[i]; 953 aj = aij->j + ii[i]; 954 aa = aij->a + ii[i]; 955 for (j=0; j<n; j++) { 956 if (PetscAbsScalar(mask[*aj])) { 957 bb[i] -= *aa*xx[*aj]; 958 *aa = 0.0; 959 } 960 aa++; 961 aj++; 962 } 963 } 964 } 965 966 ierr = VecRestoreArray(b,&bb);CHKERRQ(ierr); 967 ierr = VecRestoreArray(lmask,&mask);CHKERRQ(ierr); 968 ierr = VecRestoreArrayRead(l->lvec,&xx);CHKERRQ(ierr); 969 ierr = VecDestroy(lmask);CHKERRQ(ierr); 970 ierr = PetscFree(lrows);CHKERRQ(ierr); 971 972 /* wait on sends */ 973 if (nsends) { 974 ierr = PetscMalloc(nsends*sizeof(MPI_Status),&send_status);CHKERRQ(ierr); 975 ierr = MPI_Waitall(nsends,send_waits,send_status);CHKERRQ(ierr); 976 ierr = PetscFree(send_status);CHKERRQ(ierr); 977 } 978 ierr = PetscFree(send_waits);CHKERRQ(ierr); 979 ierr = PetscFree(svalues);CHKERRQ(ierr); 980 981 PetscFunctionReturn(0); 982 } 983 984 #undef __FUNCT__ 985 #define __FUNCT__ "MatMult_MPIAIJ" 986 PetscErrorCode MatMult_MPIAIJ(Mat A,Vec xx,Vec yy) 987 { 988 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 989 PetscErrorCode ierr; 990 PetscInt nt; 991 992 PetscFunctionBegin; 993 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 994 if (nt != A->cmap->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Incompatible partition of A (%D) and xx (%D)",A->cmap->n,nt); 995 ierr = VecScatterBegin(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 996 ierr = (*a->A->ops->mult)(a->A,xx,yy);CHKERRQ(ierr); 997 ierr = VecScatterEnd(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 998 ierr = (*a->B->ops->multadd)(a->B,a->lvec,yy,yy);CHKERRQ(ierr); 999 PetscFunctionReturn(0); 1000 } 1001 1002 #undef __FUNCT__ 1003 #define __FUNCT__ "MatMultDiagonalBlock_MPIAIJ" 1004 PetscErrorCode MatMultDiagonalBlock_MPIAIJ(Mat A,Vec bb,Vec xx) 1005 { 1006 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1007 PetscErrorCode ierr; 1008 1009 PetscFunctionBegin; 1010 ierr = MatMultDiagonalBlock(a->A,bb,xx);CHKERRQ(ierr); 1011 PetscFunctionReturn(0); 1012 } 1013 1014 #undef __FUNCT__ 1015 #define __FUNCT__ "MatMultAdd_MPIAIJ" 1016 PetscErrorCode MatMultAdd_MPIAIJ(Mat A,Vec xx,Vec yy,Vec zz) 1017 { 1018 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1019 PetscErrorCode ierr; 1020 1021 PetscFunctionBegin; 1022 ierr = VecScatterBegin(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1023 ierr = (*a->A->ops->multadd)(a->A,xx,yy,zz);CHKERRQ(ierr); 1024 ierr = VecScatterEnd(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1025 ierr = (*a->B->ops->multadd)(a->B,a->lvec,zz,zz);CHKERRQ(ierr); 1026 PetscFunctionReturn(0); 1027 } 1028 1029 #undef __FUNCT__ 1030 #define __FUNCT__ "MatMultTranspose_MPIAIJ" 1031 PetscErrorCode MatMultTranspose_MPIAIJ(Mat A,Vec xx,Vec yy) 1032 { 1033 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1034 PetscErrorCode ierr; 1035 PetscBool merged; 1036 1037 PetscFunctionBegin; 1038 ierr = VecScatterGetMerged(a->Mvctx,&merged);CHKERRQ(ierr); 1039 /* do nondiagonal part */ 1040 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 1041 if (!merged) { 1042 /* send it on its way */ 1043 ierr = VecScatterBegin(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1044 /* do local part */ 1045 ierr = (*a->A->ops->multtranspose)(a->A,xx,yy);CHKERRQ(ierr); 1046 /* receive remote parts: note this assumes the values are not actually */ 1047 /* added in yy until the next line, */ 1048 ierr = VecScatterEnd(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1049 } else { 1050 /* do local part */ 1051 ierr = (*a->A->ops->multtranspose)(a->A,xx,yy);CHKERRQ(ierr); 1052 /* send it on its way */ 1053 ierr = VecScatterBegin(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1054 /* values actually were received in the Begin() but we need to call this nop */ 1055 ierr = VecScatterEnd(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1056 } 1057 PetscFunctionReturn(0); 1058 } 1059 1060 EXTERN_C_BEGIN 1061 #undef __FUNCT__ 1062 #define __FUNCT__ "MatIsTranspose_MPIAIJ" 1063 PetscErrorCode MatIsTranspose_MPIAIJ(Mat Amat,Mat Bmat,PetscReal tol,PetscBool *f) 1064 { 1065 MPI_Comm comm; 1066 Mat_MPIAIJ *Aij = (Mat_MPIAIJ *) Amat->data, *Bij; 1067 Mat Adia = Aij->A, Bdia, Aoff,Boff,*Aoffs,*Boffs; 1068 IS Me,Notme; 1069 PetscErrorCode ierr; 1070 PetscInt M,N,first,last,*notme,i; 1071 PetscMPIInt size; 1072 1073 PetscFunctionBegin; 1074 1075 /* Easy test: symmetric diagonal block */ 1076 Bij = (Mat_MPIAIJ *) Bmat->data; Bdia = Bij->A; 1077 ierr = MatIsTranspose(Adia,Bdia,tol,f);CHKERRQ(ierr); 1078 if (!*f) PetscFunctionReturn(0); 1079 ierr = PetscObjectGetComm((PetscObject)Amat,&comm);CHKERRQ(ierr); 1080 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 1081 if (size == 1) PetscFunctionReturn(0); 1082 1083 /* Hard test: off-diagonal block. This takes a MatGetSubMatrix. */ 1084 ierr = MatGetSize(Amat,&M,&N);CHKERRQ(ierr); 1085 ierr = MatGetOwnershipRange(Amat,&first,&last);CHKERRQ(ierr); 1086 ierr = PetscMalloc((N-last+first)*sizeof(PetscInt),¬me);CHKERRQ(ierr); 1087 for (i=0; i<first; i++) notme[i] = i; 1088 for (i=last; i<M; i++) notme[i-last+first] = i; 1089 ierr = ISCreateGeneral(MPI_COMM_SELF,N-last+first,notme,PETSC_COPY_VALUES,&Notme);CHKERRQ(ierr); 1090 ierr = ISCreateStride(MPI_COMM_SELF,last-first,first,1,&Me);CHKERRQ(ierr); 1091 ierr = MatGetSubMatrices(Amat,1,&Me,&Notme,MAT_INITIAL_MATRIX,&Aoffs);CHKERRQ(ierr); 1092 Aoff = Aoffs[0]; 1093 ierr = MatGetSubMatrices(Bmat,1,&Notme,&Me,MAT_INITIAL_MATRIX,&Boffs);CHKERRQ(ierr); 1094 Boff = Boffs[0]; 1095 ierr = MatIsTranspose(Aoff,Boff,tol,f);CHKERRQ(ierr); 1096 ierr = MatDestroyMatrices(1,&Aoffs);CHKERRQ(ierr); 1097 ierr = MatDestroyMatrices(1,&Boffs);CHKERRQ(ierr); 1098 ierr = ISDestroy(Me);CHKERRQ(ierr); 1099 ierr = ISDestroy(Notme);CHKERRQ(ierr); 1100 1101 PetscFunctionReturn(0); 1102 } 1103 EXTERN_C_END 1104 1105 #undef __FUNCT__ 1106 #define __FUNCT__ "MatMultTransposeAdd_MPIAIJ" 1107 PetscErrorCode MatMultTransposeAdd_MPIAIJ(Mat A,Vec xx,Vec yy,Vec zz) 1108 { 1109 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1110 PetscErrorCode ierr; 1111 1112 PetscFunctionBegin; 1113 /* do nondiagonal part */ 1114 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 1115 /* send it on its way */ 1116 ierr = VecScatterBegin(a->Mvctx,a->lvec,zz,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1117 /* do local part */ 1118 ierr = (*a->A->ops->multtransposeadd)(a->A,xx,yy,zz);CHKERRQ(ierr); 1119 /* receive remote parts */ 1120 ierr = VecScatterEnd(a->Mvctx,a->lvec,zz,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 1121 PetscFunctionReturn(0); 1122 } 1123 1124 /* 1125 This only works correctly for square matrices where the subblock A->A is the 1126 diagonal block 1127 */ 1128 #undef __FUNCT__ 1129 #define __FUNCT__ "MatGetDiagonal_MPIAIJ" 1130 PetscErrorCode MatGetDiagonal_MPIAIJ(Mat A,Vec v) 1131 { 1132 PetscErrorCode ierr; 1133 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1134 1135 PetscFunctionBegin; 1136 if (A->rmap->N != A->cmap->N) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_SUP,"Supports only square matrix where A->A is diag block"); 1137 if (A->rmap->rstart != A->cmap->rstart || A->rmap->rend != A->cmap->rend) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"row partition must equal col partition"); 1138 ierr = MatGetDiagonal(a->A,v);CHKERRQ(ierr); 1139 PetscFunctionReturn(0); 1140 } 1141 1142 #undef __FUNCT__ 1143 #define __FUNCT__ "MatScale_MPIAIJ" 1144 PetscErrorCode MatScale_MPIAIJ(Mat A,PetscScalar aa) 1145 { 1146 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1147 PetscErrorCode ierr; 1148 1149 PetscFunctionBegin; 1150 ierr = MatScale(a->A,aa);CHKERRQ(ierr); 1151 ierr = MatScale(a->B,aa);CHKERRQ(ierr); 1152 PetscFunctionReturn(0); 1153 } 1154 1155 #undef __FUNCT__ 1156 #define __FUNCT__ "MatDestroy_MPIAIJ" 1157 PetscErrorCode MatDestroy_MPIAIJ(Mat mat) 1158 { 1159 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 1160 PetscErrorCode ierr; 1161 1162 PetscFunctionBegin; 1163 #if defined(PETSC_USE_LOG) 1164 PetscLogObjectState((PetscObject)mat,"Rows=%D, Cols=%D",mat->rmap->N,mat->cmap->N); 1165 #endif 1166 ierr = MatStashDestroy_Private(&mat->stash);CHKERRQ(ierr); 1167 if (aij->diag) {ierr = VecDestroy(aij->diag);CHKERRQ(ierr);} 1168 if (aij->A){ierr = MatDestroy(aij->A);CHKERRQ(ierr);} 1169 if (aij->B){ierr = MatDestroy(aij->B);CHKERRQ(ierr);} 1170 #if defined (PETSC_USE_CTABLE) 1171 if (aij->colmap) {ierr = PetscTableDestroy(aij->colmap);CHKERRQ(ierr);} 1172 #else 1173 ierr = PetscFree(aij->colmap);CHKERRQ(ierr); 1174 #endif 1175 ierr = PetscFree(aij->garray);CHKERRQ(ierr); 1176 if (aij->lvec) {ierr = VecDestroy(aij->lvec);CHKERRQ(ierr);} 1177 if (aij->Mvctx) {ierr = VecScatterDestroy(aij->Mvctx);CHKERRQ(ierr);} 1178 ierr = PetscFree2(aij->rowvalues,aij->rowindices);CHKERRQ(ierr); 1179 ierr = PetscFree(aij->ld);CHKERRQ(ierr); 1180 ierr = PetscFree(aij);CHKERRQ(ierr); 1181 1182 ierr = PetscObjectChangeTypeName((PetscObject)mat,0);CHKERRQ(ierr); 1183 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatStoreValues_C","",PETSC_NULL);CHKERRQ(ierr); 1184 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatRetrieveValues_C","",PETSC_NULL);CHKERRQ(ierr); 1185 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatGetDiagonalBlock_C","",PETSC_NULL);CHKERRQ(ierr); 1186 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatIsTranspose_C","",PETSC_NULL);CHKERRQ(ierr); 1187 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatMPIAIJSetPreallocation_C","",PETSC_NULL);CHKERRQ(ierr); 1188 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatMPIAIJSetPreallocationCSR_C","",PETSC_NULL);CHKERRQ(ierr); 1189 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatDiagonalScaleLocal_C","",PETSC_NULL);CHKERRQ(ierr); 1190 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatConvert_mpiaij_mpisbaij_C","",PETSC_NULL);CHKERRQ(ierr); 1191 PetscFunctionReturn(0); 1192 } 1193 1194 #undef __FUNCT__ 1195 #define __FUNCT__ "MatView_MPIAIJ_Binary" 1196 PetscErrorCode MatView_MPIAIJ_Binary(Mat mat,PetscViewer viewer) 1197 { 1198 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 1199 Mat_SeqAIJ* A = (Mat_SeqAIJ*)aij->A->data; 1200 Mat_SeqAIJ* B = (Mat_SeqAIJ*)aij->B->data; 1201 PetscErrorCode ierr; 1202 PetscMPIInt rank,size,tag = ((PetscObject)viewer)->tag; 1203 int fd; 1204 PetscInt nz,header[4],*row_lengths,*range=0,rlen,i; 1205 PetscInt nzmax,*column_indices,j,k,col,*garray = aij->garray,cnt,cstart = mat->cmap->rstart,rnz; 1206 PetscScalar *column_values; 1207 PetscInt message_count,flowcontrolcount; 1208 1209 PetscFunctionBegin; 1210 ierr = MPI_Comm_rank(((PetscObject)mat)->comm,&rank);CHKERRQ(ierr); 1211 ierr = MPI_Comm_size(((PetscObject)mat)->comm,&size);CHKERRQ(ierr); 1212 nz = A->nz + B->nz; 1213 if (!rank) { 1214 header[0] = MAT_FILE_CLASSID; 1215 header[1] = mat->rmap->N; 1216 header[2] = mat->cmap->N; 1217 ierr = MPI_Reduce(&nz,&header[3],1,MPIU_INT,MPI_SUM,0,((PetscObject)mat)->comm);CHKERRQ(ierr); 1218 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 1219 ierr = PetscBinaryWrite(fd,header,4,PETSC_INT,PETSC_TRUE);CHKERRQ(ierr); 1220 /* get largest number of rows any processor has */ 1221 rlen = mat->rmap->n; 1222 range = mat->rmap->range; 1223 for (i=1; i<size; i++) { 1224 rlen = PetscMax(rlen,range[i+1] - range[i]); 1225 } 1226 } else { 1227 ierr = MPI_Reduce(&nz,0,1,MPIU_INT,MPI_SUM,0,((PetscObject)mat)->comm);CHKERRQ(ierr); 1228 rlen = mat->rmap->n; 1229 } 1230 1231 /* load up the local row counts */ 1232 ierr = PetscMalloc((rlen+1)*sizeof(PetscInt),&row_lengths);CHKERRQ(ierr); 1233 for (i=0; i<mat->rmap->n; i++) { 1234 row_lengths[i] = A->i[i+1] - A->i[i] + B->i[i+1] - B->i[i]; 1235 } 1236 1237 /* store the row lengths to the file */ 1238 ierr = PetscViewerFlowControlStart(viewer,&message_count,&flowcontrolcount);CHKERRQ(ierr); 1239 if (!rank) { 1240 MPI_Status status; 1241 ierr = PetscBinaryWrite(fd,row_lengths,mat->rmap->n,PETSC_INT,PETSC_TRUE);CHKERRQ(ierr); 1242 for (i=1; i<size; i++) { 1243 ierr = PetscViewerFlowControlStepMaster(viewer,i,message_count,flowcontrolcount);CHKERRQ(ierr); 1244 rlen = range[i+1] - range[i]; 1245 ierr = MPI_Recv(row_lengths,rlen,MPIU_INT,i,tag,((PetscObject)mat)->comm,&status);CHKERRQ(ierr); 1246 ierr = PetscBinaryWrite(fd,row_lengths,rlen,PETSC_INT,PETSC_TRUE);CHKERRQ(ierr); 1247 } 1248 ierr = PetscViewerFlowControlEndMaster(viewer,message_count);CHKERRQ(ierr); 1249 } else { 1250 ierr = PetscViewerFlowControlStepWorker(viewer,rank,message_count);CHKERRQ(ierr); 1251 ierr = MPI_Send(row_lengths,mat->rmap->n,MPIU_INT,0,tag,((PetscObject)mat)->comm);CHKERRQ(ierr); 1252 ierr = PetscViewerFlowControlEndWorker(viewer,message_count);CHKERRQ(ierr); 1253 } 1254 ierr = PetscFree(row_lengths);CHKERRQ(ierr); 1255 1256 /* load up the local column indices */ 1257 nzmax = nz; /* )th processor needs space a largest processor needs */ 1258 ierr = MPI_Reduce(&nz,&nzmax,1,MPIU_INT,MPI_MAX,0,((PetscObject)mat)->comm);CHKERRQ(ierr); 1259 ierr = PetscMalloc((nzmax+1)*sizeof(PetscInt),&column_indices);CHKERRQ(ierr); 1260 cnt = 0; 1261 for (i=0; i<mat->rmap->n; i++) { 1262 for (j=B->i[i]; j<B->i[i+1]; j++) { 1263 if ( (col = garray[B->j[j]]) > cstart) break; 1264 column_indices[cnt++] = col; 1265 } 1266 for (k=A->i[i]; k<A->i[i+1]; k++) { 1267 column_indices[cnt++] = A->j[k] + cstart; 1268 } 1269 for (; j<B->i[i+1]; j++) { 1270 column_indices[cnt++] = garray[B->j[j]]; 1271 } 1272 } 1273 if (cnt != A->nz + B->nz) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB,"Internal PETSc error: cnt = %D nz = %D",cnt,A->nz+B->nz); 1274 1275 /* store the column indices to the file */ 1276 ierr = PetscViewerFlowControlStart(viewer,&message_count,&flowcontrolcount);CHKERRQ(ierr); 1277 if (!rank) { 1278 MPI_Status status; 1279 ierr = PetscBinaryWrite(fd,column_indices,nz,PETSC_INT,PETSC_TRUE);CHKERRQ(ierr); 1280 for (i=1; i<size; i++) { 1281 ierr = PetscViewerFlowControlStepMaster(viewer,i,message_count,flowcontrolcount);CHKERRQ(ierr); 1282 ierr = MPI_Recv(&rnz,1,MPIU_INT,i,tag,((PetscObject)mat)->comm,&status);CHKERRQ(ierr); 1283 if (rnz > nzmax) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB,"Internal PETSc error: nz = %D nzmax = %D",nz,nzmax); 1284 ierr = MPI_Recv(column_indices,rnz,MPIU_INT,i,tag,((PetscObject)mat)->comm,&status);CHKERRQ(ierr); 1285 ierr = PetscBinaryWrite(fd,column_indices,rnz,PETSC_INT,PETSC_TRUE);CHKERRQ(ierr); 1286 } 1287 ierr = PetscViewerFlowControlEndMaster(viewer,message_count);CHKERRQ(ierr); 1288 } else { 1289 ierr = PetscViewerFlowControlStepWorker(viewer,rank,message_count);CHKERRQ(ierr); 1290 ierr = MPI_Send(&nz,1,MPIU_INT,0,tag,((PetscObject)mat)->comm);CHKERRQ(ierr); 1291 ierr = MPI_Send(column_indices,nz,MPIU_INT,0,tag,((PetscObject)mat)->comm);CHKERRQ(ierr); 1292 ierr = PetscViewerFlowControlEndWorker(viewer,message_count);CHKERRQ(ierr); 1293 } 1294 ierr = PetscFree(column_indices);CHKERRQ(ierr); 1295 1296 /* load up the local column values */ 1297 ierr = PetscMalloc((nzmax+1)*sizeof(PetscScalar),&column_values);CHKERRQ(ierr); 1298 cnt = 0; 1299 for (i=0; i<mat->rmap->n; i++) { 1300 for (j=B->i[i]; j<B->i[i+1]; j++) { 1301 if ( garray[B->j[j]] > cstart) break; 1302 column_values[cnt++] = B->a[j]; 1303 } 1304 for (k=A->i[i]; k<A->i[i+1]; k++) { 1305 column_values[cnt++] = A->a[k]; 1306 } 1307 for (; j<B->i[i+1]; j++) { 1308 column_values[cnt++] = B->a[j]; 1309 } 1310 } 1311 if (cnt != A->nz + B->nz) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Internal PETSc error: cnt = %D nz = %D",cnt,A->nz+B->nz); 1312 1313 /* store the column values to the file */ 1314 ierr = PetscViewerFlowControlStart(viewer,&message_count,&flowcontrolcount);CHKERRQ(ierr); 1315 if (!rank) { 1316 MPI_Status status; 1317 ierr = PetscBinaryWrite(fd,column_values,nz,PETSC_SCALAR,PETSC_TRUE);CHKERRQ(ierr); 1318 for (i=1; i<size; i++) { 1319 ierr = PetscViewerFlowControlStepMaster(viewer,i,message_count,flowcontrolcount);CHKERRQ(ierr); 1320 ierr = MPI_Recv(&rnz,1,MPIU_INT,i,tag,((PetscObject)mat)->comm,&status);CHKERRQ(ierr); 1321 if (rnz > nzmax) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_LIB,"Internal PETSc error: nz = %D nzmax = %D",nz,nzmax); 1322 ierr = MPI_Recv(column_values,rnz,MPIU_SCALAR,i,tag,((PetscObject)mat)->comm,&status);CHKERRQ(ierr); 1323 ierr = PetscBinaryWrite(fd,column_values,rnz,PETSC_SCALAR,PETSC_TRUE);CHKERRQ(ierr); 1324 } 1325 ierr = PetscViewerFlowControlEndMaster(viewer,message_count);CHKERRQ(ierr); 1326 } else { 1327 ierr = PetscViewerFlowControlStepWorker(viewer,rank,message_count);CHKERRQ(ierr); 1328 ierr = MPI_Send(&nz,1,MPIU_INT,0,tag,((PetscObject)mat)->comm);CHKERRQ(ierr); 1329 ierr = MPI_Send(column_values,nz,MPIU_SCALAR,0,tag,((PetscObject)mat)->comm);CHKERRQ(ierr); 1330 ierr = PetscViewerFlowControlEndWorker(viewer,message_count);CHKERRQ(ierr); 1331 } 1332 ierr = PetscFree(column_values);CHKERRQ(ierr); 1333 PetscFunctionReturn(0); 1334 } 1335 1336 #undef __FUNCT__ 1337 #define __FUNCT__ "MatView_MPIAIJ_ASCIIorDraworSocket" 1338 PetscErrorCode MatView_MPIAIJ_ASCIIorDraworSocket(Mat mat,PetscViewer viewer) 1339 { 1340 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 1341 PetscErrorCode ierr; 1342 PetscMPIInt rank = aij->rank,size = aij->size; 1343 PetscBool isdraw,iascii,isbinary; 1344 PetscViewer sviewer; 1345 PetscViewerFormat format; 1346 1347 PetscFunctionBegin; 1348 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERDRAW,&isdraw);CHKERRQ(ierr); 1349 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 1350 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr); 1351 if (iascii) { 1352 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 1353 if (format == PETSC_VIEWER_ASCII_INFO_DETAIL) { 1354 MatInfo info; 1355 PetscBool inodes; 1356 1357 ierr = MPI_Comm_rank(((PetscObject)mat)->comm,&rank);CHKERRQ(ierr); 1358 ierr = MatGetInfo(mat,MAT_LOCAL,&info);CHKERRQ(ierr); 1359 ierr = MatInodeGetInodeSizes(aij->A,PETSC_NULL,(PetscInt **)&inodes,PETSC_NULL);CHKERRQ(ierr); 1360 ierr = PetscViewerASCIISynchronizedAllow(viewer,PETSC_TRUE);CHKERRQ(ierr); 1361 if (!inodes) { 1362 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] Local rows %D nz %D nz alloced %D mem %D, not using I-node routines\n", 1363 rank,mat->rmap->n,(PetscInt)info.nz_used,(PetscInt)info.nz_allocated,(PetscInt)info.memory);CHKERRQ(ierr); 1364 } else { 1365 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] Local rows %D nz %D nz alloced %D mem %D, using I-node routines\n", 1366 rank,mat->rmap->n,(PetscInt)info.nz_used,(PetscInt)info.nz_allocated,(PetscInt)info.memory);CHKERRQ(ierr); 1367 } 1368 ierr = MatGetInfo(aij->A,MAT_LOCAL,&info);CHKERRQ(ierr); 1369 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] on-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used);CHKERRQ(ierr); 1370 ierr = MatGetInfo(aij->B,MAT_LOCAL,&info);CHKERRQ(ierr); 1371 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] off-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used);CHKERRQ(ierr); 1372 ierr = PetscViewerFlush(viewer);CHKERRQ(ierr); 1373 ierr = PetscViewerASCIISynchronizedAllow(viewer,PETSC_FALSE);CHKERRQ(ierr); 1374 ierr = PetscViewerASCIIPrintf(viewer,"Information on VecScatter used in matrix-vector product: \n");CHKERRQ(ierr); 1375 ierr = VecScatterView(aij->Mvctx,viewer);CHKERRQ(ierr); 1376 PetscFunctionReturn(0); 1377 } else if (format == PETSC_VIEWER_ASCII_INFO) { 1378 PetscInt inodecount,inodelimit,*inodes; 1379 ierr = MatInodeGetInodeSizes(aij->A,&inodecount,&inodes,&inodelimit);CHKERRQ(ierr); 1380 if (inodes) { 1381 ierr = PetscViewerASCIIPrintf(viewer,"using I-node (on process 0) routines: found %D nodes, limit used is %D\n",inodecount,inodelimit);CHKERRQ(ierr); 1382 } else { 1383 ierr = PetscViewerASCIIPrintf(viewer,"not using I-node (on process 0) routines\n");CHKERRQ(ierr); 1384 } 1385 PetscFunctionReturn(0); 1386 } else if (format == PETSC_VIEWER_ASCII_FACTOR_INFO) { 1387 PetscFunctionReturn(0); 1388 } 1389 } else if (isbinary) { 1390 if (size == 1) { 1391 ierr = PetscObjectSetName((PetscObject)aij->A,((PetscObject)mat)->name);CHKERRQ(ierr); 1392 ierr = MatView(aij->A,viewer);CHKERRQ(ierr); 1393 } else { 1394 ierr = MatView_MPIAIJ_Binary(mat,viewer);CHKERRQ(ierr); 1395 } 1396 PetscFunctionReturn(0); 1397 } else if (isdraw) { 1398 PetscDraw draw; 1399 PetscBool isnull; 1400 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 1401 ierr = PetscDrawIsNull(draw,&isnull);CHKERRQ(ierr); if (isnull) PetscFunctionReturn(0); 1402 } 1403 1404 if (size == 1) { 1405 ierr = PetscObjectSetName((PetscObject)aij->A,((PetscObject)mat)->name);CHKERRQ(ierr); 1406 ierr = MatView(aij->A,viewer);CHKERRQ(ierr); 1407 } else { 1408 /* assemble the entire matrix onto first processor. */ 1409 Mat A; 1410 Mat_SeqAIJ *Aloc; 1411 PetscInt M = mat->rmap->N,N = mat->cmap->N,m,*ai,*aj,row,*cols,i,*ct; 1412 MatScalar *a; 1413 1414 if (mat->rmap->N > 1024) { 1415 PetscBool flg = PETSC_FALSE; 1416 1417 ierr = PetscOptionsGetBool(((PetscObject) mat)->prefix, "-mat_ascii_output_large", &flg,PETSC_NULL);CHKERRQ(ierr); 1418 if (!flg) { 1419 SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"ASCII matrix output not allowed for matrices with more than 1024 rows, use binary format instead.\nYou can override this restriction using -mat_ascii_output_large."); 1420 } 1421 } 1422 1423 ierr = MatCreate(((PetscObject)mat)->comm,&A);CHKERRQ(ierr); 1424 if (!rank) { 1425 ierr = MatSetSizes(A,M,N,M,N);CHKERRQ(ierr); 1426 } else { 1427 ierr = MatSetSizes(A,0,0,M,N);CHKERRQ(ierr); 1428 } 1429 /* This is just a temporary matrix, so explicitly using MATMPIAIJ is probably best */ 1430 ierr = MatSetType(A,MATMPIAIJ);CHKERRQ(ierr); 1431 ierr = MatMPIAIJSetPreallocation(A,0,PETSC_NULL,0,PETSC_NULL);CHKERRQ(ierr); 1432 ierr = PetscLogObjectParent(mat,A);CHKERRQ(ierr); 1433 1434 /* copy over the A part */ 1435 Aloc = (Mat_SeqAIJ*)aij->A->data; 1436 m = aij->A->rmap->n; ai = Aloc->i; aj = Aloc->j; a = Aloc->a; 1437 row = mat->rmap->rstart; 1438 for (i=0; i<ai[m]; i++) {aj[i] += mat->cmap->rstart ;} 1439 for (i=0; i<m; i++) { 1440 ierr = MatSetValues(A,1,&row,ai[i+1]-ai[i],aj,a,INSERT_VALUES);CHKERRQ(ierr); 1441 row++; a += ai[i+1]-ai[i]; aj += ai[i+1]-ai[i]; 1442 } 1443 aj = Aloc->j; 1444 for (i=0; i<ai[m]; i++) {aj[i] -= mat->cmap->rstart;} 1445 1446 /* copy over the B part */ 1447 Aloc = (Mat_SeqAIJ*)aij->B->data; 1448 m = aij->B->rmap->n; ai = Aloc->i; aj = Aloc->j; a = Aloc->a; 1449 row = mat->rmap->rstart; 1450 ierr = PetscMalloc((ai[m]+1)*sizeof(PetscInt),&cols);CHKERRQ(ierr); 1451 ct = cols; 1452 for (i=0; i<ai[m]; i++) {cols[i] = aij->garray[aj[i]];} 1453 for (i=0; i<m; i++) { 1454 ierr = MatSetValues(A,1,&row,ai[i+1]-ai[i],cols,a,INSERT_VALUES);CHKERRQ(ierr); 1455 row++; a += ai[i+1]-ai[i]; cols += ai[i+1]-ai[i]; 1456 } 1457 ierr = PetscFree(ct);CHKERRQ(ierr); 1458 ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1459 ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1460 /* 1461 Everyone has to call to draw the matrix since the graphics waits are 1462 synchronized across all processors that share the PetscDraw object 1463 */ 1464 ierr = PetscViewerGetSingleton(viewer,&sviewer);CHKERRQ(ierr); 1465 if (!rank) { 1466 ierr = PetscObjectSetName((PetscObject)((Mat_MPIAIJ*)(A->data))->A,((PetscObject)mat)->name);CHKERRQ(ierr); 1467 /* Set the type name to MATMPIAIJ so that the correct type can be printed out by PetscObjectPrintClassNamePrefixType() in MatView_SeqAIJ_ASCII()*/ 1468 PetscStrcpy(((PetscObject)((Mat_MPIAIJ*)(A->data))->A)->type_name,MATMPIAIJ); 1469 ierr = MatView(((Mat_MPIAIJ*)(A->data))->A,sviewer);CHKERRQ(ierr); 1470 } 1471 ierr = PetscViewerRestoreSingleton(viewer,&sviewer);CHKERRQ(ierr); 1472 ierr = MatDestroy(A);CHKERRQ(ierr); 1473 } 1474 PetscFunctionReturn(0); 1475 } 1476 1477 #undef __FUNCT__ 1478 #define __FUNCT__ "MatView_MPIAIJ" 1479 PetscErrorCode MatView_MPIAIJ(Mat mat,PetscViewer viewer) 1480 { 1481 PetscErrorCode ierr; 1482 PetscBool iascii,isdraw,issocket,isbinary; 1483 1484 PetscFunctionBegin; 1485 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 1486 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERDRAW,&isdraw);CHKERRQ(ierr); 1487 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr); 1488 ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERSOCKET,&issocket);CHKERRQ(ierr); 1489 if (iascii || isdraw || isbinary || issocket) { 1490 ierr = MatView_MPIAIJ_ASCIIorDraworSocket(mat,viewer);CHKERRQ(ierr); 1491 } else { 1492 SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Viewer type %s not supported by MPIAIJ matrices",((PetscObject)viewer)->type_name); 1493 } 1494 PetscFunctionReturn(0); 1495 } 1496 1497 #undef __FUNCT__ 1498 #define __FUNCT__ "MatSOR_MPIAIJ" 1499 PetscErrorCode MatSOR_MPIAIJ(Mat matin,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,PetscInt its,PetscInt lits,Vec xx) 1500 { 1501 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)matin->data; 1502 PetscErrorCode ierr; 1503 Vec bb1 = 0; 1504 PetscBool hasop; 1505 1506 PetscFunctionBegin; 1507 if (its > 1 || ~flag & SOR_ZERO_INITIAL_GUESS || flag & SOR_EISENSTAT) { 1508 ierr = VecDuplicate(bb,&bb1);CHKERRQ(ierr); 1509 } 1510 1511 if (flag == SOR_APPLY_UPPER) { 1512 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 1513 PetscFunctionReturn(0); 1514 } 1515 1516 if ((flag & SOR_LOCAL_SYMMETRIC_SWEEP) == SOR_LOCAL_SYMMETRIC_SWEEP){ 1517 if (flag & SOR_ZERO_INITIAL_GUESS) { 1518 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 1519 its--; 1520 } 1521 1522 while (its--) { 1523 ierr = VecScatterBegin(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1524 ierr = VecScatterEnd(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1525 1526 /* update rhs: bb1 = bb - B*x */ 1527 ierr = VecScale(mat->lvec,-1.0);CHKERRQ(ierr); 1528 ierr = (*mat->B->ops->multadd)(mat->B,mat->lvec,bb,bb1);CHKERRQ(ierr); 1529 1530 /* local sweep */ 1531 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,SOR_SYMMETRIC_SWEEP,fshift,lits,1,xx);CHKERRQ(ierr); 1532 } 1533 } else if (flag & SOR_LOCAL_FORWARD_SWEEP){ 1534 if (flag & SOR_ZERO_INITIAL_GUESS) { 1535 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 1536 its--; 1537 } 1538 while (its--) { 1539 ierr = VecScatterBegin(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1540 ierr = VecScatterEnd(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1541 1542 /* update rhs: bb1 = bb - B*x */ 1543 ierr = VecScale(mat->lvec,-1.0);CHKERRQ(ierr); 1544 ierr = (*mat->B->ops->multadd)(mat->B,mat->lvec,bb,bb1);CHKERRQ(ierr); 1545 1546 /* local sweep */ 1547 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,SOR_FORWARD_SWEEP,fshift,lits,1,xx);CHKERRQ(ierr); 1548 } 1549 } else if (flag & SOR_LOCAL_BACKWARD_SWEEP){ 1550 if (flag & SOR_ZERO_INITIAL_GUESS) { 1551 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 1552 its--; 1553 } 1554 while (its--) { 1555 ierr = VecScatterBegin(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1556 ierr = VecScatterEnd(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1557 1558 /* update rhs: bb1 = bb - B*x */ 1559 ierr = VecScale(mat->lvec,-1.0);CHKERRQ(ierr); 1560 ierr = (*mat->B->ops->multadd)(mat->B,mat->lvec,bb,bb1);CHKERRQ(ierr); 1561 1562 /* local sweep */ 1563 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,SOR_BACKWARD_SWEEP,fshift,lits,1,xx);CHKERRQ(ierr); 1564 } 1565 } else if (flag & SOR_EISENSTAT) { 1566 Vec xx1; 1567 1568 ierr = VecDuplicate(bb,&xx1);CHKERRQ(ierr); 1569 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,(MatSORType)(SOR_ZERO_INITIAL_GUESS | SOR_LOCAL_BACKWARD_SWEEP),fshift,lits,1,xx);CHKERRQ(ierr); 1570 1571 ierr = VecScatterBegin(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1572 ierr = VecScatterEnd(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1573 if (!mat->diag) { 1574 ierr = MatGetVecs(matin,&mat->diag,PETSC_NULL);CHKERRQ(ierr); 1575 ierr = MatGetDiagonal(matin,mat->diag);CHKERRQ(ierr); 1576 } 1577 ierr = MatHasOperation(matin,MATOP_MULT_DIAGONAL_BLOCK,&hasop);CHKERRQ(ierr); 1578 if (hasop) { 1579 ierr = MatMultDiagonalBlock(matin,xx,bb1);CHKERRQ(ierr); 1580 } else { 1581 ierr = VecPointwiseMult(bb1,mat->diag,xx);CHKERRQ(ierr); 1582 } 1583 ierr = VecAYPX(bb1,(omega-2.0)/omega,bb);CHKERRQ(ierr); 1584 1585 ierr = MatMultAdd(mat->B,mat->lvec,bb1,bb1);CHKERRQ(ierr); 1586 1587 /* local sweep */ 1588 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,(MatSORType)(SOR_ZERO_INITIAL_GUESS | SOR_LOCAL_FORWARD_SWEEP),fshift,lits,1,xx1);CHKERRQ(ierr); 1589 ierr = VecAXPY(xx,1.0,xx1);CHKERRQ(ierr); 1590 ierr = VecDestroy(xx1);CHKERRQ(ierr); 1591 } else { 1592 SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Parallel SOR not supported"); 1593 } 1594 1595 if (bb1) {ierr = VecDestroy(bb1);CHKERRQ(ierr);} 1596 PetscFunctionReturn(0); 1597 } 1598 1599 #undef __FUNCT__ 1600 #define __FUNCT__ "MatPermute_MPIAIJ" 1601 PetscErrorCode MatPermute_MPIAIJ(Mat A,IS rowp,IS colp,Mat *B) 1602 { 1603 MPI_Comm comm,pcomm; 1604 PetscInt first,local_size,nrows; 1605 const PetscInt *rows; 1606 PetscMPIInt size; 1607 IS crowp,growp,irowp,lrowp,lcolp,icolp; 1608 PetscErrorCode ierr; 1609 1610 PetscFunctionBegin; 1611 ierr = PetscObjectGetComm((PetscObject)A,&comm);CHKERRQ(ierr); 1612 /* make a collective version of 'rowp' */ 1613 ierr = PetscObjectGetComm((PetscObject)rowp,&pcomm);CHKERRQ(ierr); 1614 if (pcomm==comm) { 1615 crowp = rowp; 1616 } else { 1617 ierr = ISGetSize(rowp,&nrows);CHKERRQ(ierr); 1618 ierr = ISGetIndices(rowp,&rows);CHKERRQ(ierr); 1619 ierr = ISCreateGeneral(comm,nrows,rows,PETSC_COPY_VALUES,&crowp);CHKERRQ(ierr); 1620 ierr = ISRestoreIndices(rowp,&rows);CHKERRQ(ierr); 1621 } 1622 /* collect the global row permutation and invert it */ 1623 ierr = ISAllGather(crowp,&growp);CHKERRQ(ierr); 1624 ierr = ISSetPermutation(growp);CHKERRQ(ierr); 1625 if (pcomm!=comm) { 1626 ierr = ISDestroy(crowp);CHKERRQ(ierr); 1627 } 1628 ierr = ISInvertPermutation(growp,PETSC_DECIDE,&irowp);CHKERRQ(ierr); 1629 /* get the local target indices */ 1630 ierr = MatGetOwnershipRange(A,&first,PETSC_NULL);CHKERRQ(ierr); 1631 ierr = MatGetLocalSize(A,&local_size,PETSC_NULL);CHKERRQ(ierr); 1632 ierr = ISGetIndices(irowp,&rows);CHKERRQ(ierr); 1633 ierr = ISCreateGeneral(MPI_COMM_SELF,local_size,rows+first,PETSC_COPY_VALUES,&lrowp);CHKERRQ(ierr); 1634 ierr = ISRestoreIndices(irowp,&rows);CHKERRQ(ierr); 1635 ierr = ISDestroy(irowp);CHKERRQ(ierr); 1636 /* the column permutation is so much easier; 1637 make a local version of 'colp' and invert it */ 1638 ierr = PetscObjectGetComm((PetscObject)colp,&pcomm);CHKERRQ(ierr); 1639 ierr = MPI_Comm_size(pcomm,&size);CHKERRQ(ierr); 1640 if (size==1) { 1641 lcolp = colp; 1642 } else { 1643 ierr = ISGetSize(colp,&nrows);CHKERRQ(ierr); 1644 ierr = ISGetIndices(colp,&rows);CHKERRQ(ierr); 1645 ierr = ISCreateGeneral(MPI_COMM_SELF,nrows,rows,PETSC_COPY_VALUES,&lcolp);CHKERRQ(ierr); 1646 } 1647 ierr = ISSetPermutation(lcolp);CHKERRQ(ierr); 1648 ierr = ISInvertPermutation(lcolp,PETSC_DECIDE,&icolp);CHKERRQ(ierr); 1649 ierr = ISSetPermutation(icolp);CHKERRQ(ierr); 1650 if (size>1) { 1651 ierr = ISRestoreIndices(colp,&rows);CHKERRQ(ierr); 1652 ierr = ISDestroy(lcolp);CHKERRQ(ierr); 1653 } 1654 /* now we just get the submatrix */ 1655 ierr = MatGetSubMatrix_MPIAIJ_Private(A,lrowp,icolp,local_size,MAT_INITIAL_MATRIX,B);CHKERRQ(ierr); 1656 /* clean up */ 1657 ierr = ISDestroy(lrowp);CHKERRQ(ierr); 1658 ierr = ISDestroy(icolp);CHKERRQ(ierr); 1659 PetscFunctionReturn(0); 1660 } 1661 1662 #undef __FUNCT__ 1663 #define __FUNCT__ "MatGetInfo_MPIAIJ" 1664 PetscErrorCode MatGetInfo_MPIAIJ(Mat matin,MatInfoType flag,MatInfo *info) 1665 { 1666 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)matin->data; 1667 Mat A = mat->A,B = mat->B; 1668 PetscErrorCode ierr; 1669 PetscReal isend[5],irecv[5]; 1670 1671 PetscFunctionBegin; 1672 info->block_size = 1.0; 1673 ierr = MatGetInfo(A,MAT_LOCAL,info);CHKERRQ(ierr); 1674 isend[0] = info->nz_used; isend[1] = info->nz_allocated; isend[2] = info->nz_unneeded; 1675 isend[3] = info->memory; isend[4] = info->mallocs; 1676 ierr = MatGetInfo(B,MAT_LOCAL,info);CHKERRQ(ierr); 1677 isend[0] += info->nz_used; isend[1] += info->nz_allocated; isend[2] += info->nz_unneeded; 1678 isend[3] += info->memory; isend[4] += info->mallocs; 1679 if (flag == MAT_LOCAL) { 1680 info->nz_used = isend[0]; 1681 info->nz_allocated = isend[1]; 1682 info->nz_unneeded = isend[2]; 1683 info->memory = isend[3]; 1684 info->mallocs = isend[4]; 1685 } else if (flag == MAT_GLOBAL_MAX) { 1686 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPIU_MAX,((PetscObject)matin)->comm);CHKERRQ(ierr); 1687 info->nz_used = irecv[0]; 1688 info->nz_allocated = irecv[1]; 1689 info->nz_unneeded = irecv[2]; 1690 info->memory = irecv[3]; 1691 info->mallocs = irecv[4]; 1692 } else if (flag == MAT_GLOBAL_SUM) { 1693 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPIU_SUM,((PetscObject)matin)->comm);CHKERRQ(ierr); 1694 info->nz_used = irecv[0]; 1695 info->nz_allocated = irecv[1]; 1696 info->nz_unneeded = irecv[2]; 1697 info->memory = irecv[3]; 1698 info->mallocs = irecv[4]; 1699 } 1700 info->fill_ratio_given = 0; /* no parallel LU/ILU/Cholesky */ 1701 info->fill_ratio_needed = 0; 1702 info->factor_mallocs = 0; 1703 1704 PetscFunctionReturn(0); 1705 } 1706 1707 #undef __FUNCT__ 1708 #define __FUNCT__ "MatSetOption_MPIAIJ" 1709 PetscErrorCode MatSetOption_MPIAIJ(Mat A,MatOption op,PetscBool flg) 1710 { 1711 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1712 PetscErrorCode ierr; 1713 1714 PetscFunctionBegin; 1715 switch (op) { 1716 case MAT_NEW_NONZERO_LOCATIONS: 1717 case MAT_NEW_NONZERO_ALLOCATION_ERR: 1718 case MAT_UNUSED_NONZERO_LOCATION_ERR: 1719 case MAT_KEEP_NONZERO_PATTERN: 1720 case MAT_NEW_NONZERO_LOCATION_ERR: 1721 case MAT_USE_INODES: 1722 case MAT_IGNORE_ZERO_ENTRIES: 1723 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1724 ierr = MatSetOption(a->B,op,flg);CHKERRQ(ierr); 1725 break; 1726 case MAT_ROW_ORIENTED: 1727 a->roworiented = flg; 1728 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1729 ierr = MatSetOption(a->B,op,flg);CHKERRQ(ierr); 1730 break; 1731 case MAT_NEW_DIAGONALS: 1732 ierr = PetscInfo1(A,"Option %s ignored\n",MatOptions[op]);CHKERRQ(ierr); 1733 break; 1734 case MAT_IGNORE_OFF_PROC_ENTRIES: 1735 a->donotstash = PETSC_TRUE; 1736 break; 1737 case MAT_SPD: 1738 A->spd_set = PETSC_TRUE; 1739 A->spd = flg; 1740 if (flg) { 1741 A->symmetric = PETSC_TRUE; 1742 A->structurally_symmetric = PETSC_TRUE; 1743 A->symmetric_set = PETSC_TRUE; 1744 A->structurally_symmetric_set = PETSC_TRUE; 1745 } 1746 break; 1747 case MAT_SYMMETRIC: 1748 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1749 break; 1750 case MAT_STRUCTURALLY_SYMMETRIC: 1751 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1752 break; 1753 case MAT_HERMITIAN: 1754 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1755 break; 1756 case MAT_SYMMETRY_ETERNAL: 1757 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1758 break; 1759 default: 1760 SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"unknown option %d",op); 1761 } 1762 PetscFunctionReturn(0); 1763 } 1764 1765 #undef __FUNCT__ 1766 #define __FUNCT__ "MatGetRow_MPIAIJ" 1767 PetscErrorCode MatGetRow_MPIAIJ(Mat matin,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 1768 { 1769 Mat_MPIAIJ *mat = (Mat_MPIAIJ*)matin->data; 1770 PetscScalar *vworkA,*vworkB,**pvA,**pvB,*v_p; 1771 PetscErrorCode ierr; 1772 PetscInt i,*cworkA,*cworkB,**pcA,**pcB,cstart = matin->cmap->rstart; 1773 PetscInt nztot,nzA,nzB,lrow,rstart = matin->rmap->rstart,rend = matin->rmap->rend; 1774 PetscInt *cmap,*idx_p; 1775 1776 PetscFunctionBegin; 1777 if (mat->getrowactive) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Already active"); 1778 mat->getrowactive = PETSC_TRUE; 1779 1780 if (!mat->rowvalues && (idx || v)) { 1781 /* 1782 allocate enough space to hold information from the longest row. 1783 */ 1784 Mat_SeqAIJ *Aa = (Mat_SeqAIJ*)mat->A->data,*Ba = (Mat_SeqAIJ*)mat->B->data; 1785 PetscInt max = 1,tmp; 1786 for (i=0; i<matin->rmap->n; i++) { 1787 tmp = Aa->i[i+1] - Aa->i[i] + Ba->i[i+1] - Ba->i[i]; 1788 if (max < tmp) { max = tmp; } 1789 } 1790 ierr = PetscMalloc2(max,PetscScalar,&mat->rowvalues,max,PetscInt,&mat->rowindices);CHKERRQ(ierr); 1791 } 1792 1793 if (row < rstart || row >= rend) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Only local rows"); 1794 lrow = row - rstart; 1795 1796 pvA = &vworkA; pcA = &cworkA; pvB = &vworkB; pcB = &cworkB; 1797 if (!v) {pvA = 0; pvB = 0;} 1798 if (!idx) {pcA = 0; if (!v) pcB = 0;} 1799 ierr = (*mat->A->ops->getrow)(mat->A,lrow,&nzA,pcA,pvA);CHKERRQ(ierr); 1800 ierr = (*mat->B->ops->getrow)(mat->B,lrow,&nzB,pcB,pvB);CHKERRQ(ierr); 1801 nztot = nzA + nzB; 1802 1803 cmap = mat->garray; 1804 if (v || idx) { 1805 if (nztot) { 1806 /* Sort by increasing column numbers, assuming A and B already sorted */ 1807 PetscInt imark = -1; 1808 if (v) { 1809 *v = v_p = mat->rowvalues; 1810 for (i=0; i<nzB; i++) { 1811 if (cmap[cworkB[i]] < cstart) v_p[i] = vworkB[i]; 1812 else break; 1813 } 1814 imark = i; 1815 for (i=0; i<nzA; i++) v_p[imark+i] = vworkA[i]; 1816 for (i=imark; i<nzB; i++) v_p[nzA+i] = vworkB[i]; 1817 } 1818 if (idx) { 1819 *idx = idx_p = mat->rowindices; 1820 if (imark > -1) { 1821 for (i=0; i<imark; i++) { 1822 idx_p[i] = cmap[cworkB[i]]; 1823 } 1824 } else { 1825 for (i=0; i<nzB; i++) { 1826 if (cmap[cworkB[i]] < cstart) idx_p[i] = cmap[cworkB[i]]; 1827 else break; 1828 } 1829 imark = i; 1830 } 1831 for (i=0; i<nzA; i++) idx_p[imark+i] = cstart + cworkA[i]; 1832 for (i=imark; i<nzB; i++) idx_p[nzA+i] = cmap[cworkB[i]]; 1833 } 1834 } else { 1835 if (idx) *idx = 0; 1836 if (v) *v = 0; 1837 } 1838 } 1839 *nz = nztot; 1840 ierr = (*mat->A->ops->restorerow)(mat->A,lrow,&nzA,pcA,pvA);CHKERRQ(ierr); 1841 ierr = (*mat->B->ops->restorerow)(mat->B,lrow,&nzB,pcB,pvB);CHKERRQ(ierr); 1842 PetscFunctionReturn(0); 1843 } 1844 1845 #undef __FUNCT__ 1846 #define __FUNCT__ "MatRestoreRow_MPIAIJ" 1847 PetscErrorCode MatRestoreRow_MPIAIJ(Mat mat,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 1848 { 1849 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 1850 1851 PetscFunctionBegin; 1852 if (!aij->getrowactive) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"MatGetRow() must be called first"); 1853 aij->getrowactive = PETSC_FALSE; 1854 PetscFunctionReturn(0); 1855 } 1856 1857 #undef __FUNCT__ 1858 #define __FUNCT__ "MatNorm_MPIAIJ" 1859 PetscErrorCode MatNorm_MPIAIJ(Mat mat,NormType type,PetscReal *norm) 1860 { 1861 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 1862 Mat_SeqAIJ *amat = (Mat_SeqAIJ*)aij->A->data,*bmat = (Mat_SeqAIJ*)aij->B->data; 1863 PetscErrorCode ierr; 1864 PetscInt i,j,cstart = mat->cmap->rstart; 1865 PetscReal sum = 0.0; 1866 MatScalar *v; 1867 1868 PetscFunctionBegin; 1869 if (aij->size == 1) { 1870 ierr = MatNorm(aij->A,type,norm);CHKERRQ(ierr); 1871 } else { 1872 if (type == NORM_FROBENIUS) { 1873 v = amat->a; 1874 for (i=0; i<amat->nz; i++) { 1875 #if defined(PETSC_USE_COMPLEX) 1876 sum += PetscRealPart(PetscConj(*v)*(*v)); v++; 1877 #else 1878 sum += (*v)*(*v); v++; 1879 #endif 1880 } 1881 v = bmat->a; 1882 for (i=0; i<bmat->nz; i++) { 1883 #if defined(PETSC_USE_COMPLEX) 1884 sum += PetscRealPart(PetscConj(*v)*(*v)); v++; 1885 #else 1886 sum += (*v)*(*v); v++; 1887 #endif 1888 } 1889 ierr = MPI_Allreduce(&sum,norm,1,MPIU_REAL,MPIU_SUM,((PetscObject)mat)->comm);CHKERRQ(ierr); 1890 *norm = sqrt(*norm); 1891 } else if (type == NORM_1) { /* max column norm */ 1892 PetscReal *tmp,*tmp2; 1893 PetscInt *jj,*garray = aij->garray; 1894 ierr = PetscMalloc((mat->cmap->N+1)*sizeof(PetscReal),&tmp);CHKERRQ(ierr); 1895 ierr = PetscMalloc((mat->cmap->N+1)*sizeof(PetscReal),&tmp2);CHKERRQ(ierr); 1896 ierr = PetscMemzero(tmp,mat->cmap->N*sizeof(PetscReal));CHKERRQ(ierr); 1897 *norm = 0.0; 1898 v = amat->a; jj = amat->j; 1899 for (j=0; j<amat->nz; j++) { 1900 tmp[cstart + *jj++ ] += PetscAbsScalar(*v); v++; 1901 } 1902 v = bmat->a; jj = bmat->j; 1903 for (j=0; j<bmat->nz; j++) { 1904 tmp[garray[*jj++]] += PetscAbsScalar(*v); v++; 1905 } 1906 ierr = MPI_Allreduce(tmp,tmp2,mat->cmap->N,MPIU_REAL,MPIU_SUM,((PetscObject)mat)->comm);CHKERRQ(ierr); 1907 for (j=0; j<mat->cmap->N; j++) { 1908 if (tmp2[j] > *norm) *norm = tmp2[j]; 1909 } 1910 ierr = PetscFree(tmp);CHKERRQ(ierr); 1911 ierr = PetscFree(tmp2);CHKERRQ(ierr); 1912 } else if (type == NORM_INFINITY) { /* max row norm */ 1913 PetscReal ntemp = 0.0; 1914 for (j=0; j<aij->A->rmap->n; j++) { 1915 v = amat->a + amat->i[j]; 1916 sum = 0.0; 1917 for (i=0; i<amat->i[j+1]-amat->i[j]; i++) { 1918 sum += PetscAbsScalar(*v); v++; 1919 } 1920 v = bmat->a + bmat->i[j]; 1921 for (i=0; i<bmat->i[j+1]-bmat->i[j]; i++) { 1922 sum += PetscAbsScalar(*v); v++; 1923 } 1924 if (sum > ntemp) ntemp = sum; 1925 } 1926 ierr = MPI_Allreduce(&ntemp,norm,1,MPIU_REAL,MPIU_MAX,((PetscObject)mat)->comm);CHKERRQ(ierr); 1927 } else { 1928 SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"No support for two norm"); 1929 } 1930 } 1931 PetscFunctionReturn(0); 1932 } 1933 1934 #undef __FUNCT__ 1935 #define __FUNCT__ "MatTranspose_MPIAIJ" 1936 PetscErrorCode MatTranspose_MPIAIJ(Mat A,MatReuse reuse,Mat *matout) 1937 { 1938 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 1939 Mat_SeqAIJ *Aloc=(Mat_SeqAIJ*)a->A->data,*Bloc=(Mat_SeqAIJ*)a->B->data; 1940 PetscErrorCode ierr; 1941 PetscInt M = A->rmap->N,N = A->cmap->N,ma,na,mb,*ai,*aj,*bi,*bj,row,*cols,*cols_tmp,i,*d_nnz; 1942 PetscInt cstart=A->cmap->rstart,ncol; 1943 Mat B; 1944 MatScalar *array; 1945 1946 PetscFunctionBegin; 1947 if (reuse == MAT_REUSE_MATRIX && A == *matout && M != N) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Square matrix only for in-place"); 1948 1949 ma = A->rmap->n; na = A->cmap->n; mb = a->B->rmap->n; 1950 ai = Aloc->i; aj = Aloc->j; 1951 bi = Bloc->i; bj = Bloc->j; 1952 if (reuse == MAT_INITIAL_MATRIX || *matout == A) { 1953 /* compute d_nnz for preallocation; o_nnz is approximated by d_nnz to avoid communication */ 1954 ierr = PetscMalloc((1+na)*sizeof(PetscInt),&d_nnz);CHKERRQ(ierr); 1955 ierr = PetscMemzero(d_nnz,(1+na)*sizeof(PetscInt));CHKERRQ(ierr); 1956 for (i=0; i<ai[ma]; i++){ 1957 d_nnz[aj[i]] ++; 1958 aj[i] += cstart; /* global col index to be used by MatSetValues() */ 1959 } 1960 1961 ierr = MatCreate(((PetscObject)A)->comm,&B);CHKERRQ(ierr); 1962 ierr = MatSetSizes(B,A->cmap->n,A->rmap->n,N,M);CHKERRQ(ierr); 1963 ierr = MatSetType(B,((PetscObject)A)->type_name);CHKERRQ(ierr); 1964 ierr = MatMPIAIJSetPreallocation(B,0,d_nnz,0,d_nnz);CHKERRQ(ierr); 1965 ierr = PetscFree(d_nnz);CHKERRQ(ierr); 1966 } else { 1967 B = *matout; 1968 } 1969 1970 /* copy over the A part */ 1971 array = Aloc->a; 1972 row = A->rmap->rstart; 1973 for (i=0; i<ma; i++) { 1974 ncol = ai[i+1]-ai[i]; 1975 ierr = MatSetValues(B,ncol,aj,1,&row,array,INSERT_VALUES);CHKERRQ(ierr); 1976 row++; array += ncol; aj += ncol; 1977 } 1978 aj = Aloc->j; 1979 for (i=0; i<ai[ma]; i++) aj[i] -= cstart; /* resume local col index */ 1980 1981 /* copy over the B part */ 1982 ierr = PetscMalloc(bi[mb]*sizeof(PetscInt),&cols);CHKERRQ(ierr); 1983 ierr = PetscMemzero(cols,bi[mb]*sizeof(PetscInt));CHKERRQ(ierr); 1984 array = Bloc->a; 1985 row = A->rmap->rstart; 1986 for (i=0; i<bi[mb]; i++) {cols[i] = a->garray[bj[i]];} 1987 cols_tmp = cols; 1988 for (i=0; i<mb; i++) { 1989 ncol = bi[i+1]-bi[i]; 1990 ierr = MatSetValues(B,ncol,cols_tmp,1,&row,array,INSERT_VALUES);CHKERRQ(ierr); 1991 row++; array += ncol; cols_tmp += ncol; 1992 } 1993 ierr = PetscFree(cols);CHKERRQ(ierr); 1994 1995 ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1996 ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1997 if (reuse == MAT_INITIAL_MATRIX || *matout != A) { 1998 *matout = B; 1999 } else { 2000 ierr = MatHeaderMerge(A,B);CHKERRQ(ierr); 2001 } 2002 PetscFunctionReturn(0); 2003 } 2004 2005 #undef __FUNCT__ 2006 #define __FUNCT__ "MatDiagonalScale_MPIAIJ" 2007 PetscErrorCode MatDiagonalScale_MPIAIJ(Mat mat,Vec ll,Vec rr) 2008 { 2009 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 2010 Mat a = aij->A,b = aij->B; 2011 PetscErrorCode ierr; 2012 PetscInt s1,s2,s3; 2013 2014 PetscFunctionBegin; 2015 ierr = MatGetLocalSize(mat,&s2,&s3);CHKERRQ(ierr); 2016 if (rr) { 2017 ierr = VecGetLocalSize(rr,&s1);CHKERRQ(ierr); 2018 if (s1!=s3) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"right vector non-conforming local size"); 2019 /* Overlap communication with computation. */ 2020 ierr = VecScatterBegin(aij->Mvctx,rr,aij->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2021 } 2022 if (ll) { 2023 ierr = VecGetLocalSize(ll,&s1);CHKERRQ(ierr); 2024 if (s1!=s2) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"left vector non-conforming local size"); 2025 ierr = (*b->ops->diagonalscale)(b,ll,0);CHKERRQ(ierr); 2026 } 2027 /* scale the diagonal block */ 2028 ierr = (*a->ops->diagonalscale)(a,ll,rr);CHKERRQ(ierr); 2029 2030 if (rr) { 2031 /* Do a scatter end and then right scale the off-diagonal block */ 2032 ierr = VecScatterEnd(aij->Mvctx,rr,aij->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2033 ierr = (*b->ops->diagonalscale)(b,0,aij->lvec);CHKERRQ(ierr); 2034 } 2035 2036 PetscFunctionReturn(0); 2037 } 2038 2039 #undef __FUNCT__ 2040 #define __FUNCT__ "MatSetBlockSize_MPIAIJ" 2041 PetscErrorCode MatSetBlockSize_MPIAIJ(Mat A,PetscInt bs) 2042 { 2043 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2044 PetscErrorCode ierr; 2045 2046 PetscFunctionBegin; 2047 ierr = MatSetBlockSize(a->A,bs);CHKERRQ(ierr); 2048 ierr = MatSetBlockSize(a->B,bs);CHKERRQ(ierr); 2049 ierr = PetscLayoutSetBlockSize(A->rmap,bs);CHKERRQ(ierr); 2050 ierr = PetscLayoutSetBlockSize(A->cmap,bs);CHKERRQ(ierr); 2051 PetscFunctionReturn(0); 2052 } 2053 #undef __FUNCT__ 2054 #define __FUNCT__ "MatSetUnfactored_MPIAIJ" 2055 PetscErrorCode MatSetUnfactored_MPIAIJ(Mat A) 2056 { 2057 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2058 PetscErrorCode ierr; 2059 2060 PetscFunctionBegin; 2061 ierr = MatSetUnfactored(a->A);CHKERRQ(ierr); 2062 PetscFunctionReturn(0); 2063 } 2064 2065 #undef __FUNCT__ 2066 #define __FUNCT__ "MatEqual_MPIAIJ" 2067 PetscErrorCode MatEqual_MPIAIJ(Mat A,Mat B,PetscBool *flag) 2068 { 2069 Mat_MPIAIJ *matB = (Mat_MPIAIJ*)B->data,*matA = (Mat_MPIAIJ*)A->data; 2070 Mat a,b,c,d; 2071 PetscBool flg; 2072 PetscErrorCode ierr; 2073 2074 PetscFunctionBegin; 2075 a = matA->A; b = matA->B; 2076 c = matB->A; d = matB->B; 2077 2078 ierr = MatEqual(a,c,&flg);CHKERRQ(ierr); 2079 if (flg) { 2080 ierr = MatEqual(b,d,&flg);CHKERRQ(ierr); 2081 } 2082 ierr = MPI_Allreduce(&flg,flag,1,MPI_INT,MPI_LAND,((PetscObject)A)->comm);CHKERRQ(ierr); 2083 PetscFunctionReturn(0); 2084 } 2085 2086 #undef __FUNCT__ 2087 #define __FUNCT__ "MatCopy_MPIAIJ" 2088 PetscErrorCode MatCopy_MPIAIJ(Mat A,Mat B,MatStructure str) 2089 { 2090 PetscErrorCode ierr; 2091 Mat_MPIAIJ *a = (Mat_MPIAIJ *)A->data; 2092 Mat_MPIAIJ *b = (Mat_MPIAIJ *)B->data; 2093 2094 PetscFunctionBegin; 2095 /* If the two matrices don't have the same copy implementation, they aren't compatible for fast copy. */ 2096 if ((str != SAME_NONZERO_PATTERN) || (A->ops->copy != B->ops->copy)) { 2097 /* because of the column compression in the off-processor part of the matrix a->B, 2098 the number of columns in a->B and b->B may be different, hence we cannot call 2099 the MatCopy() directly on the two parts. If need be, we can provide a more 2100 efficient copy than the MatCopy_Basic() by first uncompressing the a->B matrices 2101 then copying the submatrices */ 2102 ierr = MatCopy_Basic(A,B,str);CHKERRQ(ierr); 2103 } else { 2104 ierr = MatCopy(a->A,b->A,str);CHKERRQ(ierr); 2105 ierr = MatCopy(a->B,b->B,str);CHKERRQ(ierr); 2106 } 2107 PetscFunctionReturn(0); 2108 } 2109 2110 #undef __FUNCT__ 2111 #define __FUNCT__ "MatSetUpPreallocation_MPIAIJ" 2112 PetscErrorCode MatSetUpPreallocation_MPIAIJ(Mat A) 2113 { 2114 PetscErrorCode ierr; 2115 2116 PetscFunctionBegin; 2117 ierr = MatMPIAIJSetPreallocation(A,PETSC_DEFAULT,0,PETSC_DEFAULT,0);CHKERRQ(ierr); 2118 PetscFunctionReturn(0); 2119 } 2120 2121 #undef __FUNCT__ 2122 #define __FUNCT__ "MatAXPY_MPIAIJ" 2123 PetscErrorCode MatAXPY_MPIAIJ(Mat Y,PetscScalar a,Mat X,MatStructure str) 2124 { 2125 PetscErrorCode ierr; 2126 PetscInt i; 2127 Mat_MPIAIJ *xx = (Mat_MPIAIJ *)X->data,*yy = (Mat_MPIAIJ *)Y->data; 2128 PetscBLASInt bnz,one=1; 2129 Mat_SeqAIJ *x,*y; 2130 2131 PetscFunctionBegin; 2132 if (str == SAME_NONZERO_PATTERN) { 2133 PetscScalar alpha = a; 2134 x = (Mat_SeqAIJ *)xx->A->data; 2135 y = (Mat_SeqAIJ *)yy->A->data; 2136 bnz = PetscBLASIntCast(x->nz); 2137 BLASaxpy_(&bnz,&alpha,x->a,&one,y->a,&one); 2138 x = (Mat_SeqAIJ *)xx->B->data; 2139 y = (Mat_SeqAIJ *)yy->B->data; 2140 bnz = PetscBLASIntCast(x->nz); 2141 BLASaxpy_(&bnz,&alpha,x->a,&one,y->a,&one); 2142 } else if (str == SUBSET_NONZERO_PATTERN) { 2143 ierr = MatAXPY_SeqAIJ(yy->A,a,xx->A,str);CHKERRQ(ierr); 2144 2145 x = (Mat_SeqAIJ *)xx->B->data; 2146 y = (Mat_SeqAIJ *)yy->B->data; 2147 if (y->xtoy && y->XtoY != xx->B) { 2148 ierr = PetscFree(y->xtoy);CHKERRQ(ierr); 2149 ierr = MatDestroy(y->XtoY);CHKERRQ(ierr); 2150 } 2151 if (!y->xtoy) { /* get xtoy */ 2152 ierr = MatAXPYGetxtoy_Private(xx->B->rmap->n,x->i,x->j,xx->garray,y->i,y->j,yy->garray,&y->xtoy);CHKERRQ(ierr); 2153 y->XtoY = xx->B; 2154 ierr = PetscObjectReference((PetscObject)xx->B);CHKERRQ(ierr); 2155 } 2156 for (i=0; i<x->nz; i++) y->a[y->xtoy[i]] += a*(x->a[i]); 2157 } else { 2158 Mat B; 2159 PetscInt *nnz_d,*nnz_o; 2160 ierr = PetscMalloc(yy->A->rmap->N*sizeof(PetscInt),&nnz_d);CHKERRQ(ierr); 2161 ierr = PetscMalloc(yy->B->rmap->N*sizeof(PetscInt),&nnz_o);CHKERRQ(ierr); 2162 ierr = MatCreate(((PetscObject)Y)->comm,&B);CHKERRQ(ierr); 2163 ierr = PetscObjectSetName((PetscObject)B,((PetscObject)Y)->name);CHKERRQ(ierr); 2164 ierr = MatSetSizes(B,Y->rmap->n,Y->cmap->n,Y->rmap->N,Y->cmap->N);CHKERRQ(ierr); 2165 ierr = MatSetType(B,MATMPIAIJ);CHKERRQ(ierr); 2166 ierr = MatAXPYGetPreallocation_SeqAIJ(yy->A,xx->A,nnz_d);CHKERRQ(ierr); 2167 ierr = MatAXPYGetPreallocation_SeqAIJ(yy->B,xx->B,nnz_o);CHKERRQ(ierr); 2168 ierr = MatMPIAIJSetPreallocation(B,PETSC_NULL,nnz_d,PETSC_NULL,nnz_o);CHKERRQ(ierr); 2169 ierr = MatAXPY_BasicWithPreallocation(B,Y,a,X,str);CHKERRQ(ierr); 2170 ierr = MatHeaderReplace(Y,B); 2171 ierr = PetscFree(nnz_d);CHKERRQ(ierr); 2172 ierr = PetscFree(nnz_o);CHKERRQ(ierr); 2173 } 2174 PetscFunctionReturn(0); 2175 } 2176 2177 extern PetscErrorCode MatConjugate_SeqAIJ(Mat); 2178 2179 #undef __FUNCT__ 2180 #define __FUNCT__ "MatConjugate_MPIAIJ" 2181 PetscErrorCode MatConjugate_MPIAIJ(Mat mat) 2182 { 2183 #if defined(PETSC_USE_COMPLEX) 2184 PetscErrorCode ierr; 2185 Mat_MPIAIJ *aij = (Mat_MPIAIJ *)mat->data; 2186 2187 PetscFunctionBegin; 2188 ierr = MatConjugate_SeqAIJ(aij->A);CHKERRQ(ierr); 2189 ierr = MatConjugate_SeqAIJ(aij->B);CHKERRQ(ierr); 2190 #else 2191 PetscFunctionBegin; 2192 #endif 2193 PetscFunctionReturn(0); 2194 } 2195 2196 #undef __FUNCT__ 2197 #define __FUNCT__ "MatRealPart_MPIAIJ" 2198 PetscErrorCode MatRealPart_MPIAIJ(Mat A) 2199 { 2200 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2201 PetscErrorCode ierr; 2202 2203 PetscFunctionBegin; 2204 ierr = MatRealPart(a->A);CHKERRQ(ierr); 2205 ierr = MatRealPart(a->B);CHKERRQ(ierr); 2206 PetscFunctionReturn(0); 2207 } 2208 2209 #undef __FUNCT__ 2210 #define __FUNCT__ "MatImaginaryPart_MPIAIJ" 2211 PetscErrorCode MatImaginaryPart_MPIAIJ(Mat A) 2212 { 2213 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2214 PetscErrorCode ierr; 2215 2216 PetscFunctionBegin; 2217 ierr = MatImaginaryPart(a->A);CHKERRQ(ierr); 2218 ierr = MatImaginaryPart(a->B);CHKERRQ(ierr); 2219 PetscFunctionReturn(0); 2220 } 2221 2222 #ifdef PETSC_HAVE_PBGL 2223 2224 #include <boost/parallel/mpi/bsp_process_group.hpp> 2225 #include <boost/graph/distributed/ilu_default_graph.hpp> 2226 #include <boost/graph/distributed/ilu_0_block.hpp> 2227 #include <boost/graph/distributed/ilu_preconditioner.hpp> 2228 #include <boost/graph/distributed/petsc/interface.hpp> 2229 #include <boost/multi_array.hpp> 2230 #include <boost/parallel/distributed_property_map->hpp> 2231 2232 #undef __FUNCT__ 2233 #define __FUNCT__ "MatILUFactorSymbolic_MPIAIJ" 2234 /* 2235 This uses the parallel ILU factorization of Peter Gottschling <pgottsch@osl.iu.edu> 2236 */ 2237 PetscErrorCode MatILUFactorSymbolic_MPIAIJ(Mat fact,Mat A, IS isrow, IS iscol, const MatFactorInfo *info) 2238 { 2239 namespace petsc = boost::distributed::petsc; 2240 2241 namespace graph_dist = boost::graph::distributed; 2242 using boost::graph::distributed::ilu_default::process_group_type; 2243 using boost::graph::ilu_permuted; 2244 2245 PetscBool row_identity, col_identity; 2246 PetscContainer c; 2247 PetscInt m, n, M, N; 2248 PetscErrorCode ierr; 2249 2250 PetscFunctionBegin; 2251 if (info->levels != 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only levels = 0 supported for parallel ilu"); 2252 ierr = ISIdentity(isrow, &row_identity);CHKERRQ(ierr); 2253 ierr = ISIdentity(iscol, &col_identity);CHKERRQ(ierr); 2254 if (!row_identity || !col_identity) { 2255 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Row and column permutations must be identity for parallel ILU"); 2256 } 2257 2258 process_group_type pg; 2259 typedef graph_dist::ilu_default::ilu_level_graph_type lgraph_type; 2260 lgraph_type* lgraph_p = new lgraph_type(petsc::num_global_vertices(A), pg, petsc::matrix_distribution(A, pg)); 2261 lgraph_type& level_graph = *lgraph_p; 2262 graph_dist::ilu_default::graph_type& graph(level_graph.graph); 2263 2264 petsc::read_matrix(A, graph, get(boost::edge_weight, graph)); 2265 ilu_permuted(level_graph); 2266 2267 /* put together the new matrix */ 2268 ierr = MatCreate(((PetscObject)A)->comm, fact);CHKERRQ(ierr); 2269 ierr = MatGetLocalSize(A, &m, &n);CHKERRQ(ierr); 2270 ierr = MatGetSize(A, &M, &N);CHKERRQ(ierr); 2271 ierr = MatSetSizes(fact, m, n, M, N);CHKERRQ(ierr); 2272 ierr = MatSetType(fact, ((PetscObject)A)->type_name);CHKERRQ(ierr); 2273 ierr = MatAssemblyBegin(fact, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2274 ierr = MatAssemblyEnd(fact, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2275 2276 ierr = PetscContainerCreate(((PetscObject)A)->comm, &c); 2277 ierr = PetscContainerSetPointer(c, lgraph_p); 2278 ierr = PetscObjectCompose((PetscObject) (fact), "graph", (PetscObject) c); 2279 PetscFunctionReturn(0); 2280 } 2281 2282 #undef __FUNCT__ 2283 #define __FUNCT__ "MatLUFactorNumeric_MPIAIJ" 2284 PetscErrorCode MatLUFactorNumeric_MPIAIJ(Mat B,Mat A, const MatFactorInfo *info) 2285 { 2286 PetscFunctionBegin; 2287 PetscFunctionReturn(0); 2288 } 2289 2290 #undef __FUNCT__ 2291 #define __FUNCT__ "MatSolve_MPIAIJ" 2292 /* 2293 This uses the parallel ILU factorization of Peter Gottschling <pgottsch@osl.iu.edu> 2294 */ 2295 PetscErrorCode MatSolve_MPIAIJ(Mat A, Vec b, Vec x) 2296 { 2297 namespace graph_dist = boost::graph::distributed; 2298 2299 typedef graph_dist::ilu_default::ilu_level_graph_type lgraph_type; 2300 lgraph_type* lgraph_p; 2301 PetscContainer c; 2302 PetscErrorCode ierr; 2303 2304 PetscFunctionBegin; 2305 ierr = PetscObjectQuery((PetscObject) A, "graph", (PetscObject *) &c);CHKERRQ(ierr); 2306 ierr = PetscContainerGetPointer(c, (void **) &lgraph_p);CHKERRQ(ierr); 2307 ierr = VecCopy(b, x);CHKERRQ(ierr); 2308 2309 PetscScalar* array_x; 2310 ierr = VecGetArray(x, &array_x);CHKERRQ(ierr); 2311 PetscInt sx; 2312 ierr = VecGetSize(x, &sx);CHKERRQ(ierr); 2313 2314 PetscScalar* array_b; 2315 ierr = VecGetArray(b, &array_b);CHKERRQ(ierr); 2316 PetscInt sb; 2317 ierr = VecGetSize(b, &sb);CHKERRQ(ierr); 2318 2319 lgraph_type& level_graph = *lgraph_p; 2320 graph_dist::ilu_default::graph_type& graph(level_graph.graph); 2321 2322 typedef boost::multi_array_ref<PetscScalar, 1> array_ref_type; 2323 array_ref_type ref_b(array_b, boost::extents[num_vertices(graph)]), 2324 ref_x(array_x, boost::extents[num_vertices(graph)]); 2325 2326 typedef boost::iterator_property_map<array_ref_type::iterator, 2327 boost::property_map<graph_dist::ilu_default::graph_type, boost::vertex_index_t>::type> gvector_type; 2328 gvector_type vector_b(ref_b.begin(), get(boost::vertex_index, graph)), 2329 vector_x(ref_x.begin(), get(boost::vertex_index, graph)); 2330 2331 ilu_set_solve(*lgraph_p, vector_b, vector_x); 2332 2333 PetscFunctionReturn(0); 2334 } 2335 #endif 2336 2337 typedef struct { /* used by MatGetRedundantMatrix() for reusing matredundant */ 2338 PetscInt nzlocal,nsends,nrecvs; 2339 PetscMPIInt *send_rank,*recv_rank; 2340 PetscInt *sbuf_nz,*rbuf_nz,*sbuf_j,**rbuf_j; 2341 PetscScalar *sbuf_a,**rbuf_a; 2342 PetscErrorCode (*MatDestroy)(Mat); 2343 } Mat_Redundant; 2344 2345 #undef __FUNCT__ 2346 #define __FUNCT__ "PetscContainerDestroy_MatRedundant" 2347 PetscErrorCode PetscContainerDestroy_MatRedundant(void *ptr) 2348 { 2349 PetscErrorCode ierr; 2350 Mat_Redundant *redund=(Mat_Redundant*)ptr; 2351 PetscInt i; 2352 2353 PetscFunctionBegin; 2354 ierr = PetscFree2(redund->send_rank,redund->recv_rank);CHKERRQ(ierr); 2355 ierr = PetscFree(redund->sbuf_j);CHKERRQ(ierr); 2356 ierr = PetscFree(redund->sbuf_a);CHKERRQ(ierr); 2357 for (i=0; i<redund->nrecvs; i++){ 2358 ierr = PetscFree(redund->rbuf_j[i]);CHKERRQ(ierr); 2359 ierr = PetscFree(redund->rbuf_a[i]);CHKERRQ(ierr); 2360 } 2361 ierr = PetscFree4(redund->sbuf_nz,redund->rbuf_nz,redund->rbuf_j,redund->rbuf_a);CHKERRQ(ierr); 2362 ierr = PetscFree(redund);CHKERRQ(ierr); 2363 PetscFunctionReturn(0); 2364 } 2365 2366 #undef __FUNCT__ 2367 #define __FUNCT__ "MatDestroy_MatRedundant" 2368 PetscErrorCode MatDestroy_MatRedundant(Mat A) 2369 { 2370 PetscErrorCode ierr; 2371 PetscContainer container; 2372 Mat_Redundant *redund=PETSC_NULL; 2373 2374 PetscFunctionBegin; 2375 ierr = PetscObjectQuery((PetscObject)A,"Mat_Redundant",(PetscObject *)&container);CHKERRQ(ierr); 2376 if (container) { 2377 ierr = PetscContainerGetPointer(container,(void **)&redund);CHKERRQ(ierr); 2378 } else { 2379 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Container does not exit"); 2380 } 2381 A->ops->destroy = redund->MatDestroy; 2382 ierr = PetscObjectCompose((PetscObject)A,"Mat_Redundant",0);CHKERRQ(ierr); 2383 ierr = (*A->ops->destroy)(A);CHKERRQ(ierr); 2384 ierr = PetscContainerDestroy(container);CHKERRQ(ierr); 2385 PetscFunctionReturn(0); 2386 } 2387 2388 #undef __FUNCT__ 2389 #define __FUNCT__ "MatGetRedundantMatrix_MPIAIJ" 2390 PetscErrorCode MatGetRedundantMatrix_MPIAIJ(Mat mat,PetscInt nsubcomm,MPI_Comm subcomm,PetscInt mlocal_sub,MatReuse reuse,Mat *matredundant) 2391 { 2392 PetscMPIInt rank,size; 2393 MPI_Comm comm=((PetscObject)mat)->comm; 2394 PetscErrorCode ierr; 2395 PetscInt nsends=0,nrecvs=0,i,rownz_max=0; 2396 PetscMPIInt *send_rank=PETSC_NULL,*recv_rank=PETSC_NULL; 2397 PetscInt *rowrange=mat->rmap->range; 2398 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 2399 Mat A=aij->A,B=aij->B,C=*matredundant; 2400 Mat_SeqAIJ *a=(Mat_SeqAIJ*)A->data,*b=(Mat_SeqAIJ*)B->data; 2401 PetscScalar *sbuf_a; 2402 PetscInt nzlocal=a->nz+b->nz; 2403 PetscInt j,cstart=mat->cmap->rstart,cend=mat->cmap->rend,row,nzA,nzB,ncols,*cworkA,*cworkB; 2404 PetscInt rstart=mat->rmap->rstart,rend=mat->rmap->rend,*bmap=aij->garray,M,N; 2405 PetscInt *cols,ctmp,lwrite,*rptr,l,*sbuf_j; 2406 MatScalar *aworkA,*aworkB; 2407 PetscScalar *vals; 2408 PetscMPIInt tag1,tag2,tag3,imdex; 2409 MPI_Request *s_waits1=PETSC_NULL,*s_waits2=PETSC_NULL,*s_waits3=PETSC_NULL, 2410 *r_waits1=PETSC_NULL,*r_waits2=PETSC_NULL,*r_waits3=PETSC_NULL; 2411 MPI_Status recv_status,*send_status; 2412 PetscInt *sbuf_nz=PETSC_NULL,*rbuf_nz=PETSC_NULL,count; 2413 PetscInt **rbuf_j=PETSC_NULL; 2414 PetscScalar **rbuf_a=PETSC_NULL; 2415 Mat_Redundant *redund=PETSC_NULL; 2416 PetscContainer container; 2417 2418 PetscFunctionBegin; 2419 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 2420 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 2421 2422 if (reuse == MAT_REUSE_MATRIX) { 2423 ierr = MatGetSize(C,&M,&N);CHKERRQ(ierr); 2424 if (M != N || M != mat->rmap->N) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Cannot reuse matrix. Wrong global size"); 2425 ierr = MatGetLocalSize(C,&M,&N);CHKERRQ(ierr); 2426 if (M != N || M != mlocal_sub) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Cannot reuse matrix. Wrong local size"); 2427 ierr = PetscObjectQuery((PetscObject)C,"Mat_Redundant",(PetscObject *)&container);CHKERRQ(ierr); 2428 if (container) { 2429 ierr = PetscContainerGetPointer(container,(void **)&redund);CHKERRQ(ierr); 2430 } else { 2431 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Container does not exit"); 2432 } 2433 if (nzlocal != redund->nzlocal) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Cannot reuse matrix. Wrong nzlocal"); 2434 2435 nsends = redund->nsends; 2436 nrecvs = redund->nrecvs; 2437 send_rank = redund->send_rank; 2438 recv_rank = redund->recv_rank; 2439 sbuf_nz = redund->sbuf_nz; 2440 rbuf_nz = redund->rbuf_nz; 2441 sbuf_j = redund->sbuf_j; 2442 sbuf_a = redund->sbuf_a; 2443 rbuf_j = redund->rbuf_j; 2444 rbuf_a = redund->rbuf_a; 2445 } 2446 2447 if (reuse == MAT_INITIAL_MATRIX){ 2448 PetscMPIInt subrank,subsize; 2449 PetscInt nleftover,np_subcomm; 2450 /* get the destination processors' id send_rank, nsends and nrecvs */ 2451 ierr = MPI_Comm_rank(subcomm,&subrank);CHKERRQ(ierr); 2452 ierr = MPI_Comm_size(subcomm,&subsize);CHKERRQ(ierr); 2453 ierr = PetscMalloc2(size,PetscMPIInt,&send_rank,size,PetscMPIInt,&recv_rank); 2454 np_subcomm = size/nsubcomm; 2455 nleftover = size - nsubcomm*np_subcomm; 2456 nsends = 0; nrecvs = 0; 2457 for (i=0; i<size; i++){ /* i=rank*/ 2458 if (subrank == i/nsubcomm && rank != i){ /* my_subrank == other's subrank */ 2459 send_rank[nsends] = i; nsends++; 2460 recv_rank[nrecvs++] = i; 2461 } 2462 } 2463 if (rank >= size - nleftover){/* this proc is a leftover processor */ 2464 i = size-nleftover-1; 2465 j = 0; 2466 while (j < nsubcomm - nleftover){ 2467 send_rank[nsends++] = i; 2468 i--; j++; 2469 } 2470 } 2471 2472 if (nleftover && subsize == size/nsubcomm && subrank==subsize-1){ /* this proc recvs from leftover processors */ 2473 for (i=0; i<nleftover; i++){ 2474 recv_rank[nrecvs++] = size-nleftover+i; 2475 } 2476 } 2477 2478 /* allocate sbuf_j, sbuf_a */ 2479 i = nzlocal + rowrange[rank+1] - rowrange[rank] + 2; 2480 ierr = PetscMalloc(i*sizeof(PetscInt),&sbuf_j);CHKERRQ(ierr); 2481 ierr = PetscMalloc((nzlocal+1)*sizeof(PetscScalar),&sbuf_a);CHKERRQ(ierr); 2482 } /* endof if (reuse == MAT_INITIAL_MATRIX) */ 2483 2484 /* copy mat's local entries into the buffers */ 2485 if (reuse == MAT_INITIAL_MATRIX){ 2486 rownz_max = 0; 2487 rptr = sbuf_j; 2488 cols = sbuf_j + rend-rstart + 1; 2489 vals = sbuf_a; 2490 rptr[0] = 0; 2491 for (i=0; i<rend-rstart; i++){ 2492 row = i + rstart; 2493 nzA = a->i[i+1] - a->i[i]; nzB = b->i[i+1] - b->i[i]; 2494 ncols = nzA + nzB; 2495 cworkA = a->j + a->i[i]; cworkB = b->j + b->i[i]; 2496 aworkA = a->a + a->i[i]; aworkB = b->a + b->i[i]; 2497 /* load the column indices for this row into cols */ 2498 lwrite = 0; 2499 for (l=0; l<nzB; l++) { 2500 if ((ctmp = bmap[cworkB[l]]) < cstart){ 2501 vals[lwrite] = aworkB[l]; 2502 cols[lwrite++] = ctmp; 2503 } 2504 } 2505 for (l=0; l<nzA; l++){ 2506 vals[lwrite] = aworkA[l]; 2507 cols[lwrite++] = cstart + cworkA[l]; 2508 } 2509 for (l=0; l<nzB; l++) { 2510 if ((ctmp = bmap[cworkB[l]]) >= cend){ 2511 vals[lwrite] = aworkB[l]; 2512 cols[lwrite++] = ctmp; 2513 } 2514 } 2515 vals += ncols; 2516 cols += ncols; 2517 rptr[i+1] = rptr[i] + ncols; 2518 if (rownz_max < ncols) rownz_max = ncols; 2519 } 2520 if (rptr[rend-rstart] != a->nz + b->nz) SETERRQ4(PETSC_COMM_SELF,PETSC_ERR_PLIB, "rptr[%d] %d != %d + %d",rend-rstart,rptr[rend-rstart+1],a->nz,b->nz); 2521 } else { /* only copy matrix values into sbuf_a */ 2522 rptr = sbuf_j; 2523 vals = sbuf_a; 2524 rptr[0] = 0; 2525 for (i=0; i<rend-rstart; i++){ 2526 row = i + rstart; 2527 nzA = a->i[i+1] - a->i[i]; nzB = b->i[i+1] - b->i[i]; 2528 ncols = nzA + nzB; 2529 cworkA = a->j + a->i[i]; cworkB = b->j + b->i[i]; 2530 aworkA = a->a + a->i[i]; aworkB = b->a + b->i[i]; 2531 lwrite = 0; 2532 for (l=0; l<nzB; l++) { 2533 if ((ctmp = bmap[cworkB[l]]) < cstart) vals[lwrite++] = aworkB[l]; 2534 } 2535 for (l=0; l<nzA; l++) vals[lwrite++] = aworkA[l]; 2536 for (l=0; l<nzB; l++) { 2537 if ((ctmp = bmap[cworkB[l]]) >= cend) vals[lwrite++] = aworkB[l]; 2538 } 2539 vals += ncols; 2540 rptr[i+1] = rptr[i] + ncols; 2541 } 2542 } /* endof if (reuse == MAT_INITIAL_MATRIX) */ 2543 2544 /* send nzlocal to others, and recv other's nzlocal */ 2545 /*--------------------------------------------------*/ 2546 if (reuse == MAT_INITIAL_MATRIX){ 2547 ierr = PetscMalloc2(3*(nsends + nrecvs)+1,MPI_Request,&s_waits3,nsends+1,MPI_Status,&send_status);CHKERRQ(ierr); 2548 s_waits2 = s_waits3 + nsends; 2549 s_waits1 = s_waits2 + nsends; 2550 r_waits1 = s_waits1 + nsends; 2551 r_waits2 = r_waits1 + nrecvs; 2552 r_waits3 = r_waits2 + nrecvs; 2553 } else { 2554 ierr = PetscMalloc2(nsends + nrecvs +1,MPI_Request,&s_waits3,nsends+1,MPI_Status,&send_status);CHKERRQ(ierr); 2555 r_waits3 = s_waits3 + nsends; 2556 } 2557 2558 ierr = PetscObjectGetNewTag((PetscObject)mat,&tag3);CHKERRQ(ierr); 2559 if (reuse == MAT_INITIAL_MATRIX){ 2560 /* get new tags to keep the communication clean */ 2561 ierr = PetscObjectGetNewTag((PetscObject)mat,&tag1);CHKERRQ(ierr); 2562 ierr = PetscObjectGetNewTag((PetscObject)mat,&tag2);CHKERRQ(ierr); 2563 ierr = PetscMalloc4(nsends,PetscInt,&sbuf_nz,nrecvs,PetscInt,&rbuf_nz,nrecvs,PetscInt*,&rbuf_j,nrecvs,PetscScalar*,&rbuf_a);CHKERRQ(ierr); 2564 2565 /* post receives of other's nzlocal */ 2566 for (i=0; i<nrecvs; i++){ 2567 ierr = MPI_Irecv(rbuf_nz+i,1,MPIU_INT,MPI_ANY_SOURCE,tag1,comm,r_waits1+i);CHKERRQ(ierr); 2568 } 2569 /* send nzlocal to others */ 2570 for (i=0; i<nsends; i++){ 2571 sbuf_nz[i] = nzlocal; 2572 ierr = MPI_Isend(sbuf_nz+i,1,MPIU_INT,send_rank[i],tag1,comm,s_waits1+i);CHKERRQ(ierr); 2573 } 2574 /* wait on receives of nzlocal; allocate space for rbuf_j, rbuf_a */ 2575 count = nrecvs; 2576 while (count) { 2577 ierr = MPI_Waitany(nrecvs,r_waits1,&imdex,&recv_status);CHKERRQ(ierr); 2578 recv_rank[imdex] = recv_status.MPI_SOURCE; 2579 /* allocate rbuf_a and rbuf_j; then post receives of rbuf_j */ 2580 ierr = PetscMalloc((rbuf_nz[imdex]+1)*sizeof(PetscScalar),&rbuf_a[imdex]);CHKERRQ(ierr); 2581 2582 i = rowrange[recv_status.MPI_SOURCE+1] - rowrange[recv_status.MPI_SOURCE]; /* number of expected mat->i */ 2583 rbuf_nz[imdex] += i + 2; 2584 ierr = PetscMalloc(rbuf_nz[imdex]*sizeof(PetscInt),&rbuf_j[imdex]);CHKERRQ(ierr); 2585 ierr = MPI_Irecv(rbuf_j[imdex],rbuf_nz[imdex],MPIU_INT,recv_status.MPI_SOURCE,tag2,comm,r_waits2+imdex);CHKERRQ(ierr); 2586 count--; 2587 } 2588 /* wait on sends of nzlocal */ 2589 if (nsends) {ierr = MPI_Waitall(nsends,s_waits1,send_status);CHKERRQ(ierr);} 2590 /* send mat->i,j to others, and recv from other's */ 2591 /*------------------------------------------------*/ 2592 for (i=0; i<nsends; i++){ 2593 j = nzlocal + rowrange[rank+1] - rowrange[rank] + 1; 2594 ierr = MPI_Isend(sbuf_j,j,MPIU_INT,send_rank[i],tag2,comm,s_waits2+i);CHKERRQ(ierr); 2595 } 2596 /* wait on receives of mat->i,j */ 2597 /*------------------------------*/ 2598 count = nrecvs; 2599 while (count) { 2600 ierr = MPI_Waitany(nrecvs,r_waits2,&imdex,&recv_status);CHKERRQ(ierr); 2601 if (recv_rank[imdex] != recv_status.MPI_SOURCE) SETERRQ2(PETSC_COMM_SELF,1, "recv_rank %d != MPI_SOURCE %d",recv_rank[imdex],recv_status.MPI_SOURCE); 2602 count--; 2603 } 2604 /* wait on sends of mat->i,j */ 2605 /*---------------------------*/ 2606 if (nsends) { 2607 ierr = MPI_Waitall(nsends,s_waits2,send_status);CHKERRQ(ierr); 2608 } 2609 } /* endof if (reuse == MAT_INITIAL_MATRIX) */ 2610 2611 /* post receives, send and receive mat->a */ 2612 /*----------------------------------------*/ 2613 for (imdex=0; imdex<nrecvs; imdex++) { 2614 ierr = MPI_Irecv(rbuf_a[imdex],rbuf_nz[imdex],MPIU_SCALAR,recv_rank[imdex],tag3,comm,r_waits3+imdex);CHKERRQ(ierr); 2615 } 2616 for (i=0; i<nsends; i++){ 2617 ierr = MPI_Isend(sbuf_a,nzlocal,MPIU_SCALAR,send_rank[i],tag3,comm,s_waits3+i);CHKERRQ(ierr); 2618 } 2619 count = nrecvs; 2620 while (count) { 2621 ierr = MPI_Waitany(nrecvs,r_waits3,&imdex,&recv_status);CHKERRQ(ierr); 2622 if (recv_rank[imdex] != recv_status.MPI_SOURCE) SETERRQ2(PETSC_COMM_SELF,1, "recv_rank %d != MPI_SOURCE %d",recv_rank[imdex],recv_status.MPI_SOURCE); 2623 count--; 2624 } 2625 if (nsends) { 2626 ierr = MPI_Waitall(nsends,s_waits3,send_status);CHKERRQ(ierr); 2627 } 2628 2629 ierr = PetscFree2(s_waits3,send_status);CHKERRQ(ierr); 2630 2631 /* create redundant matrix */ 2632 /*-------------------------*/ 2633 if (reuse == MAT_INITIAL_MATRIX){ 2634 /* compute rownz_max for preallocation */ 2635 for (imdex=0; imdex<nrecvs; imdex++){ 2636 j = rowrange[recv_rank[imdex]+1] - rowrange[recv_rank[imdex]]; 2637 rptr = rbuf_j[imdex]; 2638 for (i=0; i<j; i++){ 2639 ncols = rptr[i+1] - rptr[i]; 2640 if (rownz_max < ncols) rownz_max = ncols; 2641 } 2642 } 2643 2644 ierr = MatCreate(subcomm,&C);CHKERRQ(ierr); 2645 ierr = MatSetSizes(C,mlocal_sub,mlocal_sub,PETSC_DECIDE,PETSC_DECIDE);CHKERRQ(ierr); 2646 ierr = MatSetFromOptions(C);CHKERRQ(ierr); 2647 ierr = MatSeqAIJSetPreallocation(C,rownz_max,PETSC_NULL);CHKERRQ(ierr); 2648 ierr = MatMPIAIJSetPreallocation(C,rownz_max,PETSC_NULL,rownz_max,PETSC_NULL);CHKERRQ(ierr); 2649 } else { 2650 C = *matredundant; 2651 } 2652 2653 /* insert local matrix entries */ 2654 rptr = sbuf_j; 2655 cols = sbuf_j + rend-rstart + 1; 2656 vals = sbuf_a; 2657 for (i=0; i<rend-rstart; i++){ 2658 row = i + rstart; 2659 ncols = rptr[i+1] - rptr[i]; 2660 ierr = MatSetValues(C,1,&row,ncols,cols,vals,INSERT_VALUES);CHKERRQ(ierr); 2661 vals += ncols; 2662 cols += ncols; 2663 } 2664 /* insert received matrix entries */ 2665 for (imdex=0; imdex<nrecvs; imdex++){ 2666 rstart = rowrange[recv_rank[imdex]]; 2667 rend = rowrange[recv_rank[imdex]+1]; 2668 rptr = rbuf_j[imdex]; 2669 cols = rbuf_j[imdex] + rend-rstart + 1; 2670 vals = rbuf_a[imdex]; 2671 for (i=0; i<rend-rstart; i++){ 2672 row = i + rstart; 2673 ncols = rptr[i+1] - rptr[i]; 2674 ierr = MatSetValues(C,1,&row,ncols,cols,vals,INSERT_VALUES);CHKERRQ(ierr); 2675 vals += ncols; 2676 cols += ncols; 2677 } 2678 } 2679 ierr = MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2680 ierr = MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2681 ierr = MatGetSize(C,&M,&N);CHKERRQ(ierr); 2682 if (M != mat->rmap->N || N != mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_INCOMP,"redundant mat size %d != input mat size %d",M,mat->rmap->N); 2683 if (reuse == MAT_INITIAL_MATRIX){ 2684 PetscContainer container; 2685 *matredundant = C; 2686 /* create a supporting struct and attach it to C for reuse */ 2687 ierr = PetscNewLog(C,Mat_Redundant,&redund);CHKERRQ(ierr); 2688 ierr = PetscContainerCreate(PETSC_COMM_SELF,&container);CHKERRQ(ierr); 2689 ierr = PetscContainerSetPointer(container,redund);CHKERRQ(ierr); 2690 ierr = PetscObjectCompose((PetscObject)C,"Mat_Redundant",(PetscObject)container);CHKERRQ(ierr); 2691 ierr = PetscContainerSetUserDestroy(container,PetscContainerDestroy_MatRedundant);CHKERRQ(ierr); 2692 2693 redund->nzlocal = nzlocal; 2694 redund->nsends = nsends; 2695 redund->nrecvs = nrecvs; 2696 redund->send_rank = send_rank; 2697 redund->recv_rank = recv_rank; 2698 redund->sbuf_nz = sbuf_nz; 2699 redund->rbuf_nz = rbuf_nz; 2700 redund->sbuf_j = sbuf_j; 2701 redund->sbuf_a = sbuf_a; 2702 redund->rbuf_j = rbuf_j; 2703 redund->rbuf_a = rbuf_a; 2704 2705 redund->MatDestroy = C->ops->destroy; 2706 C->ops->destroy = MatDestroy_MatRedundant; 2707 } 2708 PetscFunctionReturn(0); 2709 } 2710 2711 #undef __FUNCT__ 2712 #define __FUNCT__ "MatGetRowMaxAbs_MPIAIJ" 2713 PetscErrorCode MatGetRowMaxAbs_MPIAIJ(Mat A, Vec v, PetscInt idx[]) 2714 { 2715 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2716 PetscErrorCode ierr; 2717 PetscInt i,*idxb = 0; 2718 PetscScalar *va,*vb; 2719 Vec vtmp; 2720 2721 PetscFunctionBegin; 2722 ierr = MatGetRowMaxAbs(a->A,v,idx);CHKERRQ(ierr); 2723 ierr = VecGetArray(v,&va);CHKERRQ(ierr); 2724 if (idx) { 2725 for (i=0; i<A->rmap->n; i++) { 2726 if (PetscAbsScalar(va[i])) idx[i] += A->cmap->rstart; 2727 } 2728 } 2729 2730 ierr = VecCreateSeq(PETSC_COMM_SELF,A->rmap->n,&vtmp);CHKERRQ(ierr); 2731 if (idx) { 2732 ierr = PetscMalloc(A->rmap->n*sizeof(PetscInt),&idxb);CHKERRQ(ierr); 2733 } 2734 ierr = MatGetRowMaxAbs(a->B,vtmp,idxb);CHKERRQ(ierr); 2735 ierr = VecGetArray(vtmp,&vb);CHKERRQ(ierr); 2736 2737 for (i=0; i<A->rmap->n; i++){ 2738 if (PetscAbsScalar(va[i]) < PetscAbsScalar(vb[i])) { 2739 va[i] = vb[i]; 2740 if (idx) idx[i] = a->garray[idxb[i]]; 2741 } 2742 } 2743 2744 ierr = VecRestoreArray(v,&va);CHKERRQ(ierr); 2745 ierr = VecRestoreArray(vtmp,&vb);CHKERRQ(ierr); 2746 ierr = PetscFree(idxb);CHKERRQ(ierr); 2747 ierr = VecDestroy(vtmp);CHKERRQ(ierr); 2748 PetscFunctionReturn(0); 2749 } 2750 2751 #undef __FUNCT__ 2752 #define __FUNCT__ "MatGetRowMinAbs_MPIAIJ" 2753 PetscErrorCode MatGetRowMinAbs_MPIAIJ(Mat A, Vec v, PetscInt idx[]) 2754 { 2755 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 2756 PetscErrorCode ierr; 2757 PetscInt i,*idxb = 0; 2758 PetscScalar *va,*vb; 2759 Vec vtmp; 2760 2761 PetscFunctionBegin; 2762 ierr = MatGetRowMinAbs(a->A,v,idx);CHKERRQ(ierr); 2763 ierr = VecGetArray(v,&va);CHKERRQ(ierr); 2764 if (idx) { 2765 for (i=0; i<A->cmap->n; i++) { 2766 if (PetscAbsScalar(va[i])) idx[i] += A->cmap->rstart; 2767 } 2768 } 2769 2770 ierr = VecCreateSeq(PETSC_COMM_SELF,A->rmap->n,&vtmp);CHKERRQ(ierr); 2771 if (idx) { 2772 ierr = PetscMalloc(A->rmap->n*sizeof(PetscInt),&idxb);CHKERRQ(ierr); 2773 } 2774 ierr = MatGetRowMinAbs(a->B,vtmp,idxb);CHKERRQ(ierr); 2775 ierr = VecGetArray(vtmp,&vb);CHKERRQ(ierr); 2776 2777 for (i=0; i<A->rmap->n; i++){ 2778 if (PetscAbsScalar(va[i]) > PetscAbsScalar(vb[i])) { 2779 va[i] = vb[i]; 2780 if (idx) idx[i] = a->garray[idxb[i]]; 2781 } 2782 } 2783 2784 ierr = VecRestoreArray(v,&va);CHKERRQ(ierr); 2785 ierr = VecRestoreArray(vtmp,&vb);CHKERRQ(ierr); 2786 ierr = PetscFree(idxb);CHKERRQ(ierr); 2787 ierr = VecDestroy(vtmp);CHKERRQ(ierr); 2788 PetscFunctionReturn(0); 2789 } 2790 2791 #undef __FUNCT__ 2792 #define __FUNCT__ "MatGetRowMin_MPIAIJ" 2793 PetscErrorCode MatGetRowMin_MPIAIJ(Mat A, Vec v, PetscInt idx[]) 2794 { 2795 Mat_MPIAIJ *mat = (Mat_MPIAIJ *) A->data; 2796 PetscInt n = A->rmap->n; 2797 PetscInt cstart = A->cmap->rstart; 2798 PetscInt *cmap = mat->garray; 2799 PetscInt *diagIdx, *offdiagIdx; 2800 Vec diagV, offdiagV; 2801 PetscScalar *a, *diagA, *offdiagA; 2802 PetscInt r; 2803 PetscErrorCode ierr; 2804 2805 PetscFunctionBegin; 2806 ierr = PetscMalloc2(n,PetscInt,&diagIdx,n,PetscInt,&offdiagIdx);CHKERRQ(ierr); 2807 ierr = VecCreateSeq(((PetscObject)A)->comm, n, &diagV);CHKERRQ(ierr); 2808 ierr = VecCreateSeq(((PetscObject)A)->comm, n, &offdiagV);CHKERRQ(ierr); 2809 ierr = MatGetRowMin(mat->A, diagV, diagIdx);CHKERRQ(ierr); 2810 ierr = MatGetRowMin(mat->B, offdiagV, offdiagIdx);CHKERRQ(ierr); 2811 ierr = VecGetArray(v, &a);CHKERRQ(ierr); 2812 ierr = VecGetArray(diagV, &diagA);CHKERRQ(ierr); 2813 ierr = VecGetArray(offdiagV, &offdiagA);CHKERRQ(ierr); 2814 for(r = 0; r < n; ++r) { 2815 if (PetscAbsScalar(diagA[r]) <= PetscAbsScalar(offdiagA[r])) { 2816 a[r] = diagA[r]; 2817 idx[r] = cstart + diagIdx[r]; 2818 } else { 2819 a[r] = offdiagA[r]; 2820 idx[r] = cmap[offdiagIdx[r]]; 2821 } 2822 } 2823 ierr = VecRestoreArray(v, &a);CHKERRQ(ierr); 2824 ierr = VecRestoreArray(diagV, &diagA);CHKERRQ(ierr); 2825 ierr = VecRestoreArray(offdiagV, &offdiagA);CHKERRQ(ierr); 2826 ierr = VecDestroy(diagV);CHKERRQ(ierr); 2827 ierr = VecDestroy(offdiagV);CHKERRQ(ierr); 2828 ierr = PetscFree2(diagIdx, offdiagIdx);CHKERRQ(ierr); 2829 PetscFunctionReturn(0); 2830 } 2831 2832 #undef __FUNCT__ 2833 #define __FUNCT__ "MatGetRowMax_MPIAIJ" 2834 PetscErrorCode MatGetRowMax_MPIAIJ(Mat A, Vec v, PetscInt idx[]) 2835 { 2836 Mat_MPIAIJ *mat = (Mat_MPIAIJ *) A->data; 2837 PetscInt n = A->rmap->n; 2838 PetscInt cstart = A->cmap->rstart; 2839 PetscInt *cmap = mat->garray; 2840 PetscInt *diagIdx, *offdiagIdx; 2841 Vec diagV, offdiagV; 2842 PetscScalar *a, *diagA, *offdiagA; 2843 PetscInt r; 2844 PetscErrorCode ierr; 2845 2846 PetscFunctionBegin; 2847 ierr = PetscMalloc2(n,PetscInt,&diagIdx,n,PetscInt,&offdiagIdx);CHKERRQ(ierr); 2848 ierr = VecCreateSeq(((PetscObject)A)->comm, n, &diagV);CHKERRQ(ierr); 2849 ierr = VecCreateSeq(((PetscObject)A)->comm, n, &offdiagV);CHKERRQ(ierr); 2850 ierr = MatGetRowMax(mat->A, diagV, diagIdx);CHKERRQ(ierr); 2851 ierr = MatGetRowMax(mat->B, offdiagV, offdiagIdx);CHKERRQ(ierr); 2852 ierr = VecGetArray(v, &a);CHKERRQ(ierr); 2853 ierr = VecGetArray(diagV, &diagA);CHKERRQ(ierr); 2854 ierr = VecGetArray(offdiagV, &offdiagA);CHKERRQ(ierr); 2855 for(r = 0; r < n; ++r) { 2856 if (PetscAbsScalar(diagA[r]) >= PetscAbsScalar(offdiagA[r])) { 2857 a[r] = diagA[r]; 2858 idx[r] = cstart + diagIdx[r]; 2859 } else { 2860 a[r] = offdiagA[r]; 2861 idx[r] = cmap[offdiagIdx[r]]; 2862 } 2863 } 2864 ierr = VecRestoreArray(v, &a);CHKERRQ(ierr); 2865 ierr = VecRestoreArray(diagV, &diagA);CHKERRQ(ierr); 2866 ierr = VecRestoreArray(offdiagV, &offdiagA);CHKERRQ(ierr); 2867 ierr = VecDestroy(diagV);CHKERRQ(ierr); 2868 ierr = VecDestroy(offdiagV);CHKERRQ(ierr); 2869 ierr = PetscFree2(diagIdx, offdiagIdx);CHKERRQ(ierr); 2870 PetscFunctionReturn(0); 2871 } 2872 2873 #undef __FUNCT__ 2874 #define __FUNCT__ "MatGetSeqNonzerostructure_MPIAIJ" 2875 PetscErrorCode MatGetSeqNonzerostructure_MPIAIJ(Mat mat,Mat *newmat) 2876 { 2877 PetscErrorCode ierr; 2878 Mat *dummy; 2879 2880 PetscFunctionBegin; 2881 ierr = MatGetSubMatrix_MPIAIJ_All(mat,MAT_DO_NOT_GET_VALUES,MAT_INITIAL_MATRIX,&dummy);CHKERRQ(ierr); 2882 *newmat = *dummy; 2883 ierr = PetscFree(dummy);CHKERRQ(ierr); 2884 PetscFunctionReturn(0); 2885 } 2886 2887 extern PetscErrorCode MatFDColoringApply_AIJ(Mat,MatFDColoring,Vec,MatStructure*,void*); 2888 /* -------------------------------------------------------------------*/ 2889 static struct _MatOps MatOps_Values = {MatSetValues_MPIAIJ, 2890 MatGetRow_MPIAIJ, 2891 MatRestoreRow_MPIAIJ, 2892 MatMult_MPIAIJ, 2893 /* 4*/ MatMultAdd_MPIAIJ, 2894 MatMultTranspose_MPIAIJ, 2895 MatMultTransposeAdd_MPIAIJ, 2896 #ifdef PETSC_HAVE_PBGL 2897 MatSolve_MPIAIJ, 2898 #else 2899 0, 2900 #endif 2901 0, 2902 0, 2903 /*10*/ 0, 2904 0, 2905 0, 2906 MatSOR_MPIAIJ, 2907 MatTranspose_MPIAIJ, 2908 /*15*/ MatGetInfo_MPIAIJ, 2909 MatEqual_MPIAIJ, 2910 MatGetDiagonal_MPIAIJ, 2911 MatDiagonalScale_MPIAIJ, 2912 MatNorm_MPIAIJ, 2913 /*20*/ MatAssemblyBegin_MPIAIJ, 2914 MatAssemblyEnd_MPIAIJ, 2915 MatSetOption_MPIAIJ, 2916 MatZeroEntries_MPIAIJ, 2917 /*24*/ MatZeroRows_MPIAIJ, 2918 0, 2919 #ifdef PETSC_HAVE_PBGL 2920 0, 2921 #else 2922 0, 2923 #endif 2924 0, 2925 0, 2926 /*29*/ MatSetUpPreallocation_MPIAIJ, 2927 #ifdef PETSC_HAVE_PBGL 2928 0, 2929 #else 2930 0, 2931 #endif 2932 0, 2933 0, 2934 0, 2935 /*34*/ MatDuplicate_MPIAIJ, 2936 0, 2937 0, 2938 0, 2939 0, 2940 /*39*/ MatAXPY_MPIAIJ, 2941 MatGetSubMatrices_MPIAIJ, 2942 MatIncreaseOverlap_MPIAIJ, 2943 MatGetValues_MPIAIJ, 2944 MatCopy_MPIAIJ, 2945 /*44*/ MatGetRowMax_MPIAIJ, 2946 MatScale_MPIAIJ, 2947 0, 2948 0, 2949 MatZeroRowsColumns_MPIAIJ, 2950 /*49*/ MatSetBlockSize_MPIAIJ, 2951 0, 2952 0, 2953 0, 2954 0, 2955 /*54*/ MatFDColoringCreate_MPIAIJ, 2956 0, 2957 MatSetUnfactored_MPIAIJ, 2958 MatPermute_MPIAIJ, 2959 0, 2960 /*59*/ MatGetSubMatrix_MPIAIJ, 2961 MatDestroy_MPIAIJ, 2962 MatView_MPIAIJ, 2963 0, 2964 0, 2965 /*64*/ 0, 2966 0, 2967 0, 2968 0, 2969 0, 2970 /*69*/ MatGetRowMaxAbs_MPIAIJ, 2971 MatGetRowMinAbs_MPIAIJ, 2972 0, 2973 MatSetColoring_MPIAIJ, 2974 #if defined(PETSC_HAVE_ADIC) 2975 MatSetValuesAdic_MPIAIJ, 2976 #else 2977 0, 2978 #endif 2979 MatSetValuesAdifor_MPIAIJ, 2980 /*75*/ MatFDColoringApply_AIJ, 2981 0, 2982 0, 2983 0, 2984 0, 2985 /*80*/ 0, 2986 0, 2987 0, 2988 /*83*/ MatLoad_MPIAIJ, 2989 0, 2990 0, 2991 0, 2992 0, 2993 0, 2994 /*89*/ MatMatMult_MPIAIJ_MPIAIJ, 2995 MatMatMultSymbolic_MPIAIJ_MPIAIJ, 2996 MatMatMultNumeric_MPIAIJ_MPIAIJ, 2997 MatPtAP_Basic, 2998 MatPtAPSymbolic_MPIAIJ, 2999 /*94*/ MatPtAPNumeric_MPIAIJ, 3000 0, 3001 0, 3002 0, 3003 0, 3004 /*99*/ 0, 3005 MatPtAPSymbolic_MPIAIJ_MPIAIJ, 3006 MatPtAPNumeric_MPIAIJ_MPIAIJ, 3007 MatConjugate_MPIAIJ, 3008 0, 3009 /*104*/MatSetValuesRow_MPIAIJ, 3010 MatRealPart_MPIAIJ, 3011 MatImaginaryPart_MPIAIJ, 3012 0, 3013 0, 3014 /*109*/0, 3015 MatGetRedundantMatrix_MPIAIJ, 3016 MatGetRowMin_MPIAIJ, 3017 0, 3018 0, 3019 /*114*/MatGetSeqNonzerostructure_MPIAIJ, 3020 0, 3021 0, 3022 0, 3023 0, 3024 /*119*/0, 3025 0, 3026 0, 3027 0, 3028 MatGetMultiProcBlock_MPIAIJ, 3029 /*124*/MatFindNonZeroRows_MPIAIJ, 3030 0, 3031 0, 3032 0, 3033 MatGetSubMatricesParallel_MPIAIJ 3034 }; 3035 3036 /* ----------------------------------------------------------------------------------------*/ 3037 3038 EXTERN_C_BEGIN 3039 #undef __FUNCT__ 3040 #define __FUNCT__ "MatStoreValues_MPIAIJ" 3041 PetscErrorCode MatStoreValues_MPIAIJ(Mat mat) 3042 { 3043 Mat_MPIAIJ *aij = (Mat_MPIAIJ *)mat->data; 3044 PetscErrorCode ierr; 3045 3046 PetscFunctionBegin; 3047 ierr = MatStoreValues(aij->A);CHKERRQ(ierr); 3048 ierr = MatStoreValues(aij->B);CHKERRQ(ierr); 3049 PetscFunctionReturn(0); 3050 } 3051 EXTERN_C_END 3052 3053 EXTERN_C_BEGIN 3054 #undef __FUNCT__ 3055 #define __FUNCT__ "MatRetrieveValues_MPIAIJ" 3056 PetscErrorCode MatRetrieveValues_MPIAIJ(Mat mat) 3057 { 3058 Mat_MPIAIJ *aij = (Mat_MPIAIJ *)mat->data; 3059 PetscErrorCode ierr; 3060 3061 PetscFunctionBegin; 3062 ierr = MatRetrieveValues(aij->A);CHKERRQ(ierr); 3063 ierr = MatRetrieveValues(aij->B);CHKERRQ(ierr); 3064 PetscFunctionReturn(0); 3065 } 3066 EXTERN_C_END 3067 3068 EXTERN_C_BEGIN 3069 #undef __FUNCT__ 3070 #define __FUNCT__ "MatMPIAIJSetPreallocation_MPIAIJ" 3071 PetscErrorCode MatMPIAIJSetPreallocation_MPIAIJ(Mat B,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[]) 3072 { 3073 Mat_MPIAIJ *b; 3074 PetscErrorCode ierr; 3075 PetscInt i; 3076 3077 PetscFunctionBegin; 3078 if (d_nz == PETSC_DEFAULT || d_nz == PETSC_DECIDE) d_nz = 5; 3079 if (o_nz == PETSC_DEFAULT || o_nz == PETSC_DECIDE) o_nz = 2; 3080 if (d_nz < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"d_nz cannot be less than 0: value %D",d_nz); 3081 if (o_nz < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"o_nz cannot be less than 0: value %D",o_nz); 3082 3083 ierr = PetscLayoutSetBlockSize(B->rmap,1);CHKERRQ(ierr); 3084 ierr = PetscLayoutSetBlockSize(B->cmap,1);CHKERRQ(ierr); 3085 ierr = PetscLayoutSetUp(B->rmap);CHKERRQ(ierr); 3086 ierr = PetscLayoutSetUp(B->cmap);CHKERRQ(ierr); 3087 if (d_nnz) { 3088 for (i=0; i<B->rmap->n; i++) { 3089 if (d_nnz[i] < 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"d_nnz cannot be less than 0: local row %D value %D",i,d_nnz[i]); 3090 } 3091 } 3092 if (o_nnz) { 3093 for (i=0; i<B->rmap->n; i++) { 3094 if (o_nnz[i] < 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"o_nnz cannot be less than 0: local row %D value %D",i,o_nnz[i]); 3095 } 3096 } 3097 b = (Mat_MPIAIJ*)B->data; 3098 3099 if (!B->preallocated) { 3100 /* Explicitly create 2 MATSEQAIJ matrices. */ 3101 ierr = MatCreate(PETSC_COMM_SELF,&b->A);CHKERRQ(ierr); 3102 ierr = MatSetSizes(b->A,B->rmap->n,B->cmap->n,B->rmap->n,B->cmap->n);CHKERRQ(ierr); 3103 ierr = MatSetType(b->A,MATSEQAIJ);CHKERRQ(ierr); 3104 ierr = PetscLogObjectParent(B,b->A);CHKERRQ(ierr); 3105 ierr = MatCreate(PETSC_COMM_SELF,&b->B);CHKERRQ(ierr); 3106 ierr = MatSetSizes(b->B,B->rmap->n,B->cmap->N,B->rmap->n,B->cmap->N);CHKERRQ(ierr); 3107 ierr = MatSetType(b->B,MATSEQAIJ);CHKERRQ(ierr); 3108 ierr = PetscLogObjectParent(B,b->B);CHKERRQ(ierr); 3109 } 3110 3111 ierr = MatSeqAIJSetPreallocation(b->A,d_nz,d_nnz);CHKERRQ(ierr); 3112 ierr = MatSeqAIJSetPreallocation(b->B,o_nz,o_nnz);CHKERRQ(ierr); 3113 B->preallocated = PETSC_TRUE; 3114 PetscFunctionReturn(0); 3115 } 3116 EXTERN_C_END 3117 3118 #undef __FUNCT__ 3119 #define __FUNCT__ "MatDuplicate_MPIAIJ" 3120 PetscErrorCode MatDuplicate_MPIAIJ(Mat matin,MatDuplicateOption cpvalues,Mat *newmat) 3121 { 3122 Mat mat; 3123 Mat_MPIAIJ *a,*oldmat = (Mat_MPIAIJ*)matin->data; 3124 PetscErrorCode ierr; 3125 3126 PetscFunctionBegin; 3127 *newmat = 0; 3128 ierr = MatCreate(((PetscObject)matin)->comm,&mat);CHKERRQ(ierr); 3129 ierr = MatSetSizes(mat,matin->rmap->n,matin->cmap->n,matin->rmap->N,matin->cmap->N);CHKERRQ(ierr); 3130 ierr = MatSetType(mat,((PetscObject)matin)->type_name);CHKERRQ(ierr); 3131 ierr = PetscMemcpy(mat->ops,matin->ops,sizeof(struct _MatOps));CHKERRQ(ierr); 3132 a = (Mat_MPIAIJ*)mat->data; 3133 3134 mat->factortype = matin->factortype; 3135 mat->rmap->bs = matin->rmap->bs; 3136 mat->assembled = PETSC_TRUE; 3137 mat->insertmode = NOT_SET_VALUES; 3138 mat->preallocated = PETSC_TRUE; 3139 3140 a->size = oldmat->size; 3141 a->rank = oldmat->rank; 3142 a->donotstash = oldmat->donotstash; 3143 a->roworiented = oldmat->roworiented; 3144 a->rowindices = 0; 3145 a->rowvalues = 0; 3146 a->getrowactive = PETSC_FALSE; 3147 3148 ierr = PetscLayoutCopy(matin->rmap,&mat->rmap);CHKERRQ(ierr); 3149 ierr = PetscLayoutCopy(matin->cmap,&mat->cmap);CHKERRQ(ierr); 3150 3151 if (oldmat->colmap) { 3152 #if defined (PETSC_USE_CTABLE) 3153 ierr = PetscTableCreateCopy(oldmat->colmap,&a->colmap);CHKERRQ(ierr); 3154 #else 3155 ierr = PetscMalloc((mat->cmap->N)*sizeof(PetscInt),&a->colmap);CHKERRQ(ierr); 3156 ierr = PetscLogObjectMemory(mat,(mat->cmap->N)*sizeof(PetscInt));CHKERRQ(ierr); 3157 ierr = PetscMemcpy(a->colmap,oldmat->colmap,(mat->cmap->N)*sizeof(PetscInt));CHKERRQ(ierr); 3158 #endif 3159 } else a->colmap = 0; 3160 if (oldmat->garray) { 3161 PetscInt len; 3162 len = oldmat->B->cmap->n; 3163 ierr = PetscMalloc((len+1)*sizeof(PetscInt),&a->garray);CHKERRQ(ierr); 3164 ierr = PetscLogObjectMemory(mat,len*sizeof(PetscInt));CHKERRQ(ierr); 3165 if (len) { ierr = PetscMemcpy(a->garray,oldmat->garray,len*sizeof(PetscInt));CHKERRQ(ierr); } 3166 } else a->garray = 0; 3167 3168 ierr = VecDuplicate(oldmat->lvec,&a->lvec);CHKERRQ(ierr); 3169 ierr = PetscLogObjectParent(mat,a->lvec);CHKERRQ(ierr); 3170 ierr = VecScatterCopy(oldmat->Mvctx,&a->Mvctx);CHKERRQ(ierr); 3171 ierr = PetscLogObjectParent(mat,a->Mvctx);CHKERRQ(ierr); 3172 ierr = MatDuplicate(oldmat->A,cpvalues,&a->A);CHKERRQ(ierr); 3173 ierr = PetscLogObjectParent(mat,a->A);CHKERRQ(ierr); 3174 ierr = MatDuplicate(oldmat->B,cpvalues,&a->B);CHKERRQ(ierr); 3175 ierr = PetscLogObjectParent(mat,a->B);CHKERRQ(ierr); 3176 ierr = PetscFListDuplicate(((PetscObject)matin)->qlist,&((PetscObject)mat)->qlist);CHKERRQ(ierr); 3177 *newmat = mat; 3178 PetscFunctionReturn(0); 3179 } 3180 3181 /* 3182 Allows sending/receiving larger messages then 2 gigabytes in a single call 3183 */ 3184 static int MPILong_Send(void *mess,PetscInt cnt, MPI_Datatype type,int to, int tag, MPI_Comm comm) 3185 { 3186 int ierr; 3187 static PetscInt CHUNKSIZE = 250000000; /* 250,000,000 */ 3188 PetscInt i,numchunks; 3189 PetscMPIInt icnt; 3190 3191 numchunks = cnt/CHUNKSIZE + 1; 3192 for (i=0; i<numchunks; i++) { 3193 icnt = PetscMPIIntCast((i < numchunks-1) ? CHUNKSIZE : cnt - (numchunks-1)*CHUNKSIZE); 3194 ierr = MPI_Send(mess,icnt,type,to,tag,comm);if (ierr) return ierr; 3195 if (type == MPIU_INT) { 3196 mess = (void*) (((PetscInt*)mess) + CHUNKSIZE); 3197 } else if (type == MPIU_SCALAR) { 3198 mess = (void*) (((PetscScalar*)mess) + CHUNKSIZE); 3199 } else SETERRQ(comm,PETSC_ERR_SUP,"No support for this datatype"); 3200 } 3201 return 0; 3202 } 3203 static int MPILong_Recv(void *mess,PetscInt cnt, MPI_Datatype type,int from, int tag, MPI_Comm comm) 3204 { 3205 int ierr; 3206 static PetscInt CHUNKSIZE = 250000000; /* 250,000,000 */ 3207 MPI_Status status; 3208 PetscInt i,numchunks; 3209 PetscMPIInt icnt; 3210 3211 numchunks = cnt/CHUNKSIZE + 1; 3212 for (i=0; i<numchunks; i++) { 3213 icnt = PetscMPIIntCast((i < numchunks-1) ? CHUNKSIZE : cnt - (numchunks-1)*CHUNKSIZE); 3214 ierr = MPI_Recv(mess,icnt,type,from,tag,comm,&status);if (ierr) return ierr; 3215 if (type == MPIU_INT) { 3216 mess = (void*) (((PetscInt*)mess) + CHUNKSIZE); 3217 } else if (type == MPIU_SCALAR) { 3218 mess = (void*) (((PetscScalar*)mess) + CHUNKSIZE); 3219 } else SETERRQ(comm,PETSC_ERR_SUP,"No support for this datatype"); 3220 } 3221 return 0; 3222 } 3223 3224 #undef __FUNCT__ 3225 #define __FUNCT__ "MatLoad_MPIAIJ" 3226 PetscErrorCode MatLoad_MPIAIJ(Mat newMat, PetscViewer viewer) 3227 { 3228 PetscScalar *vals,*svals; 3229 MPI_Comm comm = ((PetscObject)viewer)->comm; 3230 PetscErrorCode ierr; 3231 PetscMPIInt rank,size,tag = ((PetscObject)viewer)->tag; 3232 PetscInt i,nz,j,rstart,rend,mmax,maxnz = 0,grows,gcols; 3233 PetscInt header[4],*rowlengths = 0,M,N,m,*cols; 3234 PetscInt *ourlens = PETSC_NULL,*procsnz = PETSC_NULL,*offlens = PETSC_NULL,jj,*mycols,*smycols; 3235 PetscInt cend,cstart,n,*rowners,sizesset=1; 3236 int fd; 3237 3238 PetscFunctionBegin; 3239 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 3240 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 3241 if (!rank) { 3242 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 3243 ierr = PetscBinaryRead(fd,(char *)header,4,PETSC_INT);CHKERRQ(ierr); 3244 if (header[0] != MAT_FILE_CLASSID) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED,"not matrix object"); 3245 } 3246 3247 if (newMat->rmap->n < 0 && newMat->rmap->N < 0 && newMat->cmap->n < 0 && newMat->cmap->N < 0) sizesset = 0; 3248 3249 ierr = MPI_Bcast(header+1,3,MPIU_INT,0,comm);CHKERRQ(ierr); 3250 M = header[1]; N = header[2]; 3251 /* If global rows/cols are set to PETSC_DECIDE, set it to the sizes given in the file */ 3252 if (sizesset && newMat->rmap->N < 0) newMat->rmap->N = M; 3253 if (sizesset && newMat->cmap->N < 0) newMat->cmap->N = N; 3254 3255 /* If global sizes are set, check if they are consistent with that given in the file */ 3256 if (sizesset) { 3257 ierr = MatGetSize(newMat,&grows,&gcols);CHKERRQ(ierr); 3258 } 3259 if (sizesset && newMat->rmap->N != grows) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED, "Inconsistent # of rows:Matrix in file has (%d) and input matrix has (%d)",M,grows); 3260 if (sizesset && newMat->cmap->N != gcols) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED, "Inconsistent # of cols:Matrix in file has (%d) and input matrix has (%d)",N,gcols); 3261 3262 /* determine ownership of all rows */ 3263 if (newMat->rmap->n < 0 ) m = M/size + ((M % size) > rank); /* PETSC_DECIDE */ 3264 else m = newMat->rmap->n; /* Set by user */ 3265 3266 ierr = PetscMalloc((size+1)*sizeof(PetscInt),&rowners);CHKERRQ(ierr); 3267 ierr = MPI_Allgather(&m,1,MPIU_INT,rowners+1,1,MPIU_INT,comm);CHKERRQ(ierr); 3268 3269 /* First process needs enough room for process with most rows */ 3270 if (!rank) { 3271 mmax = rowners[1]; 3272 for (i=2; i<size; i++) { 3273 mmax = PetscMax(mmax,rowners[i]); 3274 } 3275 } else mmax = m; 3276 3277 rowners[0] = 0; 3278 for (i=2; i<=size; i++) { 3279 rowners[i] += rowners[i-1]; 3280 } 3281 rstart = rowners[rank]; 3282 rend = rowners[rank+1]; 3283 3284 /* distribute row lengths to all processors */ 3285 ierr = PetscMalloc2(mmax,PetscInt,&ourlens,mmax,PetscInt,&offlens);CHKERRQ(ierr); 3286 if (!rank) { 3287 ierr = PetscBinaryRead(fd,ourlens,m,PETSC_INT);CHKERRQ(ierr); 3288 ierr = PetscMalloc(m*sizeof(PetscInt),&rowlengths);CHKERRQ(ierr); 3289 ierr = PetscMalloc(size*sizeof(PetscInt),&procsnz);CHKERRQ(ierr); 3290 ierr = PetscMemzero(procsnz,size*sizeof(PetscInt));CHKERRQ(ierr); 3291 for (j=0; j<m; j++) { 3292 procsnz[0] += ourlens[j]; 3293 } 3294 for (i=1; i<size; i++) { 3295 ierr = PetscBinaryRead(fd,rowlengths,rowners[i+1]-rowners[i],PETSC_INT);CHKERRQ(ierr); 3296 /* calculate the number of nonzeros on each processor */ 3297 for (j=0; j<rowners[i+1]-rowners[i]; j++) { 3298 procsnz[i] += rowlengths[j]; 3299 } 3300 ierr = MPILong_Send(rowlengths,rowners[i+1]-rowners[i],MPIU_INT,i,tag,comm);CHKERRQ(ierr); 3301 } 3302 ierr = PetscFree(rowlengths);CHKERRQ(ierr); 3303 } else { 3304 ierr = MPILong_Recv(ourlens,m,MPIU_INT,0,tag,comm);CHKERRQ(ierr); 3305 } 3306 3307 if (!rank) { 3308 /* determine max buffer needed and allocate it */ 3309 maxnz = 0; 3310 for (i=0; i<size; i++) { 3311 maxnz = PetscMax(maxnz,procsnz[i]); 3312 } 3313 ierr = PetscMalloc(maxnz*sizeof(PetscInt),&cols);CHKERRQ(ierr); 3314 3315 /* read in my part of the matrix column indices */ 3316 nz = procsnz[0]; 3317 ierr = PetscMalloc(nz*sizeof(PetscInt),&mycols);CHKERRQ(ierr); 3318 ierr = PetscBinaryRead(fd,mycols,nz,PETSC_INT);CHKERRQ(ierr); 3319 3320 /* read in every one elses and ship off */ 3321 for (i=1; i<size; i++) { 3322 nz = procsnz[i]; 3323 ierr = PetscBinaryRead(fd,cols,nz,PETSC_INT);CHKERRQ(ierr); 3324 ierr = MPILong_Send(cols,nz,MPIU_INT,i,tag,comm);CHKERRQ(ierr); 3325 } 3326 ierr = PetscFree(cols);CHKERRQ(ierr); 3327 } else { 3328 /* determine buffer space needed for message */ 3329 nz = 0; 3330 for (i=0; i<m; i++) { 3331 nz += ourlens[i]; 3332 } 3333 ierr = PetscMalloc(nz*sizeof(PetscInt),&mycols);CHKERRQ(ierr); 3334 3335 /* receive message of column indices*/ 3336 ierr = MPILong_Recv(mycols,nz,MPIU_INT,0,tag,comm);CHKERRQ(ierr); 3337 } 3338 3339 /* determine column ownership if matrix is not square */ 3340 if (N != M) { 3341 if (newMat->cmap->n < 0) n = N/size + ((N % size) > rank); 3342 else n = newMat->cmap->n; 3343 ierr = MPI_Scan(&n,&cend,1,MPIU_INT,MPI_SUM,comm);CHKERRQ(ierr); 3344 cstart = cend - n; 3345 } else { 3346 cstart = rstart; 3347 cend = rend; 3348 n = cend - cstart; 3349 } 3350 3351 /* loop over local rows, determining number of off diagonal entries */ 3352 ierr = PetscMemzero(offlens,m*sizeof(PetscInt));CHKERRQ(ierr); 3353 jj = 0; 3354 for (i=0; i<m; i++) { 3355 for (j=0; j<ourlens[i]; j++) { 3356 if (mycols[jj] < cstart || mycols[jj] >= cend) offlens[i]++; 3357 jj++; 3358 } 3359 } 3360 3361 for (i=0; i<m; i++) { 3362 ourlens[i] -= offlens[i]; 3363 } 3364 if (!sizesset) { 3365 ierr = MatSetSizes(newMat,m,n,M,N);CHKERRQ(ierr); 3366 } 3367 ierr = MatMPIAIJSetPreallocation(newMat,0,ourlens,0,offlens);CHKERRQ(ierr); 3368 3369 for (i=0; i<m; i++) { 3370 ourlens[i] += offlens[i]; 3371 } 3372 3373 if (!rank) { 3374 ierr = PetscMalloc((maxnz+1)*sizeof(PetscScalar),&vals);CHKERRQ(ierr); 3375 3376 /* read in my part of the matrix numerical values */ 3377 nz = procsnz[0]; 3378 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 3379 3380 /* insert into matrix */ 3381 jj = rstart; 3382 smycols = mycols; 3383 svals = vals; 3384 for (i=0; i<m; i++) { 3385 ierr = MatSetValues_MPIAIJ(newMat,1,&jj,ourlens[i],smycols,svals,INSERT_VALUES);CHKERRQ(ierr); 3386 smycols += ourlens[i]; 3387 svals += ourlens[i]; 3388 jj++; 3389 } 3390 3391 /* read in other processors and ship out */ 3392 for (i=1; i<size; i++) { 3393 nz = procsnz[i]; 3394 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 3395 ierr = MPILong_Send(vals,nz,MPIU_SCALAR,i,((PetscObject)newMat)->tag,comm);CHKERRQ(ierr); 3396 } 3397 ierr = PetscFree(procsnz);CHKERRQ(ierr); 3398 } else { 3399 /* receive numeric values */ 3400 ierr = PetscMalloc((nz+1)*sizeof(PetscScalar),&vals);CHKERRQ(ierr); 3401 3402 /* receive message of values*/ 3403 ierr = MPILong_Recv(vals,nz,MPIU_SCALAR,0,((PetscObject)newMat)->tag,comm);CHKERRQ(ierr); 3404 3405 /* insert into matrix */ 3406 jj = rstart; 3407 smycols = mycols; 3408 svals = vals; 3409 for (i=0; i<m; i++) { 3410 ierr = MatSetValues_MPIAIJ(newMat,1,&jj,ourlens[i],smycols,svals,INSERT_VALUES);CHKERRQ(ierr); 3411 smycols += ourlens[i]; 3412 svals += ourlens[i]; 3413 jj++; 3414 } 3415 } 3416 ierr = PetscFree2(ourlens,offlens);CHKERRQ(ierr); 3417 ierr = PetscFree(vals);CHKERRQ(ierr); 3418 ierr = PetscFree(mycols);CHKERRQ(ierr); 3419 ierr = PetscFree(rowners);CHKERRQ(ierr); 3420 3421 ierr = MatAssemblyBegin(newMat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3422 ierr = MatAssemblyEnd(newMat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3423 PetscFunctionReturn(0); 3424 } 3425 3426 #undef __FUNCT__ 3427 #define __FUNCT__ "MatGetSubMatrix_MPIAIJ" 3428 PetscErrorCode MatGetSubMatrix_MPIAIJ(Mat mat,IS isrow,IS iscol,MatReuse call,Mat *newmat) 3429 { 3430 PetscErrorCode ierr; 3431 IS iscol_local; 3432 PetscInt csize; 3433 3434 PetscFunctionBegin; 3435 ierr = ISGetLocalSize(iscol,&csize);CHKERRQ(ierr); 3436 if (call == MAT_REUSE_MATRIX) { 3437 ierr = PetscObjectQuery((PetscObject)*newmat,"ISAllGather",(PetscObject*)&iscol_local);CHKERRQ(ierr); 3438 if (!iscol_local) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Submatrix passed in was not used before, cannot reuse"); 3439 } else { 3440 ierr = ISAllGather(iscol,&iscol_local);CHKERRQ(ierr); 3441 } 3442 ierr = MatGetSubMatrix_MPIAIJ_Private(mat,isrow,iscol_local,csize,call,newmat);CHKERRQ(ierr); 3443 if (call == MAT_INITIAL_MATRIX) { 3444 ierr = PetscObjectCompose((PetscObject)*newmat,"ISAllGather",(PetscObject)iscol_local);CHKERRQ(ierr); 3445 ierr = ISDestroy(iscol_local);CHKERRQ(ierr); 3446 } 3447 PetscFunctionReturn(0); 3448 } 3449 3450 #undef __FUNCT__ 3451 #define __FUNCT__ "MatGetSubMatrix_MPIAIJ_Private" 3452 /* 3453 Not great since it makes two copies of the submatrix, first an SeqAIJ 3454 in local and then by concatenating the local matrices the end result. 3455 Writing it directly would be much like MatGetSubMatrices_MPIAIJ() 3456 3457 Note: This requires a sequential iscol with all indices. 3458 */ 3459 PetscErrorCode MatGetSubMatrix_MPIAIJ_Private(Mat mat,IS isrow,IS iscol,PetscInt csize,MatReuse call,Mat *newmat) 3460 { 3461 PetscErrorCode ierr; 3462 PetscMPIInt rank,size; 3463 PetscInt i,m,n,rstart,row,rend,nz,*cwork,j; 3464 PetscInt *ii,*jj,nlocal,*dlens,*olens,dlen,olen,jend,mglobal; 3465 Mat *local,M,Mreuse; 3466 MatScalar *vwork,*aa; 3467 MPI_Comm comm = ((PetscObject)mat)->comm; 3468 Mat_SeqAIJ *aij; 3469 3470 3471 PetscFunctionBegin; 3472 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 3473 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 3474 3475 if (call == MAT_REUSE_MATRIX) { 3476 ierr = PetscObjectQuery((PetscObject)*newmat,"SubMatrix",(PetscObject *)&Mreuse);CHKERRQ(ierr); 3477 if (!Mreuse) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Submatrix passed in was not used before, cannot reuse"); 3478 local = &Mreuse; 3479 ierr = MatGetSubMatrices(mat,1,&isrow,&iscol,MAT_REUSE_MATRIX,&local);CHKERRQ(ierr); 3480 } else { 3481 ierr = MatGetSubMatrices(mat,1,&isrow,&iscol,MAT_INITIAL_MATRIX,&local);CHKERRQ(ierr); 3482 Mreuse = *local; 3483 ierr = PetscFree(local);CHKERRQ(ierr); 3484 } 3485 3486 /* 3487 m - number of local rows 3488 n - number of columns (same on all processors) 3489 rstart - first row in new global matrix generated 3490 */ 3491 ierr = MatGetSize(Mreuse,&m,&n);CHKERRQ(ierr); 3492 if (call == MAT_INITIAL_MATRIX) { 3493 aij = (Mat_SeqAIJ*)(Mreuse)->data; 3494 ii = aij->i; 3495 jj = aij->j; 3496 3497 /* 3498 Determine the number of non-zeros in the diagonal and off-diagonal 3499 portions of the matrix in order to do correct preallocation 3500 */ 3501 3502 /* first get start and end of "diagonal" columns */ 3503 if (csize == PETSC_DECIDE) { 3504 ierr = ISGetSize(isrow,&mglobal);CHKERRQ(ierr); 3505 if (mglobal == n) { /* square matrix */ 3506 nlocal = m; 3507 } else { 3508 nlocal = n/size + ((n % size) > rank); 3509 } 3510 } else { 3511 nlocal = csize; 3512 } 3513 ierr = MPI_Scan(&nlocal,&rend,1,MPIU_INT,MPI_SUM,comm);CHKERRQ(ierr); 3514 rstart = rend - nlocal; 3515 if (rank == size - 1 && rend != n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Local column sizes %D do not add up to total number of columns %D",rend,n); 3516 3517 /* next, compute all the lengths */ 3518 ierr = PetscMalloc((2*m+1)*sizeof(PetscInt),&dlens);CHKERRQ(ierr); 3519 olens = dlens + m; 3520 for (i=0; i<m; i++) { 3521 jend = ii[i+1] - ii[i]; 3522 olen = 0; 3523 dlen = 0; 3524 for (j=0; j<jend; j++) { 3525 if (*jj < rstart || *jj >= rend) olen++; 3526 else dlen++; 3527 jj++; 3528 } 3529 olens[i] = olen; 3530 dlens[i] = dlen; 3531 } 3532 ierr = MatCreate(comm,&M);CHKERRQ(ierr); 3533 ierr = MatSetSizes(M,m,nlocal,PETSC_DECIDE,n);CHKERRQ(ierr); 3534 ierr = MatSetType(M,((PetscObject)mat)->type_name);CHKERRQ(ierr); 3535 ierr = MatMPIAIJSetPreallocation(M,0,dlens,0,olens);CHKERRQ(ierr); 3536 ierr = PetscFree(dlens);CHKERRQ(ierr); 3537 } else { 3538 PetscInt ml,nl; 3539 3540 M = *newmat; 3541 ierr = MatGetLocalSize(M,&ml,&nl);CHKERRQ(ierr); 3542 if (ml != m) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Previous matrix must be same size/layout as request"); 3543 ierr = MatZeroEntries(M);CHKERRQ(ierr); 3544 /* 3545 The next two lines are needed so we may call MatSetValues_MPIAIJ() below directly, 3546 rather than the slower MatSetValues(). 3547 */ 3548 M->was_assembled = PETSC_TRUE; 3549 M->assembled = PETSC_FALSE; 3550 } 3551 ierr = MatGetOwnershipRange(M,&rstart,&rend);CHKERRQ(ierr); 3552 aij = (Mat_SeqAIJ*)(Mreuse)->data; 3553 ii = aij->i; 3554 jj = aij->j; 3555 aa = aij->a; 3556 for (i=0; i<m; i++) { 3557 row = rstart + i; 3558 nz = ii[i+1] - ii[i]; 3559 cwork = jj; jj += nz; 3560 vwork = aa; aa += nz; 3561 ierr = MatSetValues_MPIAIJ(M,1,&row,nz,cwork,vwork,INSERT_VALUES);CHKERRQ(ierr); 3562 } 3563 3564 ierr = MatAssemblyBegin(M,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3565 ierr = MatAssemblyEnd(M,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3566 *newmat = M; 3567 3568 /* save submatrix used in processor for next request */ 3569 if (call == MAT_INITIAL_MATRIX) { 3570 ierr = PetscObjectCompose((PetscObject)M,"SubMatrix",(PetscObject)Mreuse);CHKERRQ(ierr); 3571 ierr = PetscObjectDereference((PetscObject)Mreuse);CHKERRQ(ierr); 3572 } 3573 3574 PetscFunctionReturn(0); 3575 } 3576 3577 EXTERN_C_BEGIN 3578 #undef __FUNCT__ 3579 #define __FUNCT__ "MatMPIAIJSetPreallocationCSR_MPIAIJ" 3580 PetscErrorCode MatMPIAIJSetPreallocationCSR_MPIAIJ(Mat B,const PetscInt Ii[],const PetscInt J[],const PetscScalar v[]) 3581 { 3582 PetscInt m,cstart, cend,j,nnz,i,d; 3583 PetscInt *d_nnz,*o_nnz,nnz_max = 0,rstart,ii; 3584 const PetscInt *JJ; 3585 PetscScalar *values; 3586 PetscErrorCode ierr; 3587 3588 PetscFunctionBegin; 3589 if (Ii[0]) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Ii[0] must be 0 it is %D",Ii[0]); 3590 3591 ierr = PetscLayoutSetBlockSize(B->rmap,1);CHKERRQ(ierr); 3592 ierr = PetscLayoutSetBlockSize(B->cmap,1);CHKERRQ(ierr); 3593 ierr = PetscLayoutSetUp(B->rmap);CHKERRQ(ierr); 3594 ierr = PetscLayoutSetUp(B->cmap);CHKERRQ(ierr); 3595 m = B->rmap->n; 3596 cstart = B->cmap->rstart; 3597 cend = B->cmap->rend; 3598 rstart = B->rmap->rstart; 3599 3600 ierr = PetscMalloc2(m,PetscInt,&d_nnz,m,PetscInt,&o_nnz);CHKERRQ(ierr); 3601 3602 #if defined(PETSC_USE_DEBUGGING) 3603 for (i=0; i<m; i++) { 3604 nnz = Ii[i+1]- Ii[i]; 3605 JJ = J + Ii[i]; 3606 if (nnz < 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Local row %D has a negative %D number of columns",i,nnz); 3607 if (nnz && (JJ[0] < 0)) SETERRRQ1(PETSC_ERR_ARG_WRONGSTATE,"Row %D starts with negative column index",i,j); 3608 if (nnz && (JJ[nnz-1] >= B->cmap->N) SETERRRQ3(PETSC_ERR_ARG_WRONGSTATE,"Row %D ends with too large a column index %D (max allowed %D)",i,JJ[nnz-1],B->cmap->N); 3609 } 3610 #endif 3611 3612 for (i=0; i<m; i++) { 3613 nnz = Ii[i+1]- Ii[i]; 3614 JJ = J + Ii[i]; 3615 nnz_max = PetscMax(nnz_max,nnz); 3616 d = 0; 3617 for (j=0; j<nnz; j++) { 3618 if (cstart <= JJ[j] && JJ[j] < cend) d++; 3619 } 3620 d_nnz[i] = d; 3621 o_nnz[i] = nnz - d; 3622 } 3623 ierr = MatMPIAIJSetPreallocation(B,0,d_nnz,0,o_nnz);CHKERRQ(ierr); 3624 ierr = PetscFree2(d_nnz,o_nnz);CHKERRQ(ierr); 3625 3626 if (v) values = (PetscScalar*)v; 3627 else { 3628 ierr = PetscMalloc((nnz_max+1)*sizeof(PetscScalar),&values);CHKERRQ(ierr); 3629 ierr = PetscMemzero(values,nnz_max*sizeof(PetscScalar));CHKERRQ(ierr); 3630 } 3631 3632 for (i=0; i<m; i++) { 3633 ii = i + rstart; 3634 nnz = Ii[i+1]- Ii[i]; 3635 ierr = MatSetValues_MPIAIJ(B,1,&ii,nnz,J+Ii[i],values+(v ? Ii[i] : 0),INSERT_VALUES);CHKERRQ(ierr); 3636 } 3637 ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3638 ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 3639 3640 if (!v) { 3641 ierr = PetscFree(values);CHKERRQ(ierr); 3642 } 3643 PetscFunctionReturn(0); 3644 } 3645 EXTERN_C_END 3646 3647 #undef __FUNCT__ 3648 #define __FUNCT__ "MatMPIAIJSetPreallocationCSR" 3649 /*@ 3650 MatMPIAIJSetPreallocationCSR - Allocates memory for a sparse parallel matrix in AIJ format 3651 (the default parallel PETSc format). 3652 3653 Collective on MPI_Comm 3654 3655 Input Parameters: 3656 + B - the matrix 3657 . i - the indices into j for the start of each local row (starts with zero) 3658 . j - the column indices for each local row (starts with zero) 3659 - v - optional values in the matrix 3660 3661 Level: developer 3662 3663 Notes: 3664 The i, j, and a arrays ARE copied by this routine into the internal format used by PETSc; 3665 thus you CANNOT change the matrix entries by changing the values of a[] after you have 3666 called this routine. Use MatCreateMPIAIJWithSplitArrays() to avoid needing to copy the arrays. 3667 3668 The i and j indices are 0 based, and i indices are indices corresponding to the local j array. 3669 3670 The format which is used for the sparse matrix input, is equivalent to a 3671 row-major ordering.. i.e for the following matrix, the input data expected is 3672 as shown: 3673 3674 1 0 0 3675 2 0 3 P0 3676 ------- 3677 4 5 6 P1 3678 3679 Process0 [P0]: rows_owned=[0,1] 3680 i = {0,1,3} [size = nrow+1 = 2+1] 3681 j = {0,0,2} [size = nz = 6] 3682 v = {1,2,3} [size = nz = 6] 3683 3684 Process1 [P1]: rows_owned=[2] 3685 i = {0,3} [size = nrow+1 = 1+1] 3686 j = {0,1,2} [size = nz = 6] 3687 v = {4,5,6} [size = nz = 6] 3688 3689 .keywords: matrix, aij, compressed row, sparse, parallel 3690 3691 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIAIJSetPreallocation(), MatCreateMPIAIJ(), MPIAIJ, 3692 MatCreateSeqAIJWithArrays(), MatCreateMPIAIJWithSplitArrays() 3693 @*/ 3694 PetscErrorCode MatMPIAIJSetPreallocationCSR(Mat B,const PetscInt i[],const PetscInt j[], const PetscScalar v[]) 3695 { 3696 PetscErrorCode ierr; 3697 3698 PetscFunctionBegin; 3699 ierr = PetscTryMethod(B,"MatMPIAIJSetPreallocationCSR_C",(Mat,const PetscInt[],const PetscInt[],const PetscScalar[]),(B,i,j,v));CHKERRQ(ierr); 3700 PetscFunctionReturn(0); 3701 } 3702 3703 #undef __FUNCT__ 3704 #define __FUNCT__ "MatMPIAIJSetPreallocation" 3705 /*@C 3706 MatMPIAIJSetPreallocation - Preallocates memory for a sparse parallel matrix in AIJ format 3707 (the default parallel PETSc format). For good matrix assembly performance 3708 the user should preallocate the matrix storage by setting the parameters 3709 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 3710 performance can be increased by more than a factor of 50. 3711 3712 Collective on MPI_Comm 3713 3714 Input Parameters: 3715 + A - the matrix 3716 . d_nz - number of nonzeros per row in DIAGONAL portion of local submatrix 3717 (same value is used for all local rows) 3718 . d_nnz - array containing the number of nonzeros in the various rows of the 3719 DIAGONAL portion of the local submatrix (possibly different for each row) 3720 or PETSC_NULL, if d_nz is used to specify the nonzero structure. 3721 The size of this array is equal to the number of local rows, i.e 'm'. 3722 You must leave room for the diagonal entry even if it is zero. 3723 . o_nz - number of nonzeros per row in the OFF-DIAGONAL portion of local 3724 submatrix (same value is used for all local rows). 3725 - o_nnz - array containing the number of nonzeros in the various rows of the 3726 OFF-DIAGONAL portion of the local submatrix (possibly different for 3727 each row) or PETSC_NULL, if o_nz is used to specify the nonzero 3728 structure. The size of this array is equal to the number 3729 of local rows, i.e 'm'. 3730 3731 If the *_nnz parameter is given then the *_nz parameter is ignored 3732 3733 The AIJ format (also called the Yale sparse matrix format or 3734 compressed row storage (CSR)), is fully compatible with standard Fortran 77 3735 storage. The stored row and column indices begin with zero. 3736 See the <A href="../../docs/manual.pdf#nameddest=ch_mat">Mat chapter of the users manual</A> for details. 3737 3738 The parallel matrix is partitioned such that the first m0 rows belong to 3739 process 0, the next m1 rows belong to process 1, the next m2 rows belong 3740 to process 2 etc.. where m0,m1,m2... are the input parameter 'm'. 3741 3742 The DIAGONAL portion of the local submatrix of a processor can be defined 3743 as the submatrix which is obtained by extraction the part corresponding to 3744 the rows r1-r2 and columns c1-c2 of the global matrix, where r1 is the 3745 first row that belongs to the processor, r2 is the last row belonging to 3746 the this processor, and c1-c2 is range of indices of the local part of a 3747 vector suitable for applying the matrix to. This is an mxn matrix. In the 3748 common case of a square matrix, the row and column ranges are the same and 3749 the DIAGONAL part is also square. The remaining portion of the local 3750 submatrix (mxN) constitute the OFF-DIAGONAL portion. 3751 3752 If o_nnz, d_nnz are specified, then o_nz, and d_nz are ignored. 3753 3754 You can call MatGetInfo() to get information on how effective the preallocation was; 3755 for example the fields mallocs,nz_allocated,nz_used,nz_unneeded; 3756 You can also run with the option -info and look for messages with the string 3757 malloc in them to see if additional memory allocation was needed. 3758 3759 Example usage: 3760 3761 Consider the following 8x8 matrix with 34 non-zero values, that is 3762 assembled across 3 processors. Lets assume that proc0 owns 3 rows, 3763 proc1 owns 3 rows, proc2 owns 2 rows. This division can be shown 3764 as follows: 3765 3766 .vb 3767 1 2 0 | 0 3 0 | 0 4 3768 Proc0 0 5 6 | 7 0 0 | 8 0 3769 9 0 10 | 11 0 0 | 12 0 3770 ------------------------------------- 3771 13 0 14 | 15 16 17 | 0 0 3772 Proc1 0 18 0 | 19 20 21 | 0 0 3773 0 0 0 | 22 23 0 | 24 0 3774 ------------------------------------- 3775 Proc2 25 26 27 | 0 0 28 | 29 0 3776 30 0 0 | 31 32 33 | 0 34 3777 .ve 3778 3779 This can be represented as a collection of submatrices as: 3780 3781 .vb 3782 A B C 3783 D E F 3784 G H I 3785 .ve 3786 3787 Where the submatrices A,B,C are owned by proc0, D,E,F are 3788 owned by proc1, G,H,I are owned by proc2. 3789 3790 The 'm' parameters for proc0,proc1,proc2 are 3,3,2 respectively. 3791 The 'n' parameters for proc0,proc1,proc2 are 3,3,2 respectively. 3792 The 'M','N' parameters are 8,8, and have the same values on all procs. 3793 3794 The DIAGONAL submatrices corresponding to proc0,proc1,proc2 are 3795 submatrices [A], [E], [I] respectively. The OFF-DIAGONAL submatrices 3796 corresponding to proc0,proc1,proc2 are [BC], [DF], [GH] respectively. 3797 Internally, each processor stores the DIAGONAL part, and the OFF-DIAGONAL 3798 part as SeqAIJ matrices. for eg: proc1 will store [E] as a SeqAIJ 3799 matrix, ans [DF] as another SeqAIJ matrix. 3800 3801 When d_nz, o_nz parameters are specified, d_nz storage elements are 3802 allocated for every row of the local diagonal submatrix, and o_nz 3803 storage locations are allocated for every row of the OFF-DIAGONAL submat. 3804 One way to choose d_nz and o_nz is to use the max nonzerors per local 3805 rows for each of the local DIAGONAL, and the OFF-DIAGONAL submatrices. 3806 In this case, the values of d_nz,o_nz are: 3807 .vb 3808 proc0 : dnz = 2, o_nz = 2 3809 proc1 : dnz = 3, o_nz = 2 3810 proc2 : dnz = 1, o_nz = 4 3811 .ve 3812 We are allocating m*(d_nz+o_nz) storage locations for every proc. This 3813 translates to 3*(2+2)=12 for proc0, 3*(3+2)=15 for proc1, 2*(1+4)=10 3814 for proc3. i.e we are using 12+15+10=37 storage locations to store 3815 34 values. 3816 3817 When d_nnz, o_nnz parameters are specified, the storage is specified 3818 for every row, coresponding to both DIAGONAL and OFF-DIAGONAL submatrices. 3819 In the above case the values for d_nnz,o_nnz are: 3820 .vb 3821 proc0: d_nnz = [2,2,2] and o_nnz = [2,2,2] 3822 proc1: d_nnz = [3,3,2] and o_nnz = [2,1,1] 3823 proc2: d_nnz = [1,1] and o_nnz = [4,4] 3824 .ve 3825 Here the space allocated is sum of all the above values i.e 34, and 3826 hence pre-allocation is perfect. 3827 3828 Level: intermediate 3829 3830 .keywords: matrix, aij, compressed row, sparse, parallel 3831 3832 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatCreateMPIAIJ(), MatMPIAIJSetPreallocationCSR(), 3833 MPIAIJ, MatGetInfo() 3834 @*/ 3835 PetscErrorCode MatMPIAIJSetPreallocation(Mat B,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[]) 3836 { 3837 PetscErrorCode ierr; 3838 3839 PetscFunctionBegin; 3840 ierr = PetscTryMethod(B,"MatMPIAIJSetPreallocation_C",(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[]),(B,d_nz,d_nnz,o_nz,o_nnz));CHKERRQ(ierr); 3841 PetscFunctionReturn(0); 3842 } 3843 3844 #undef __FUNCT__ 3845 #define __FUNCT__ "MatCreateMPIAIJWithArrays" 3846 /*@ 3847 MatCreateMPIAIJWithArrays - creates a MPI AIJ matrix using arrays that contain in standard 3848 CSR format the local rows. 3849 3850 Collective on MPI_Comm 3851 3852 Input Parameters: 3853 + comm - MPI communicator 3854 . m - number of local rows (Cannot be PETSC_DECIDE) 3855 . n - This value should be the same as the local size used in creating the 3856 x vector for the matrix-vector product y = Ax. (or PETSC_DECIDE to have 3857 calculated if N is given) For square matrices n is almost always m. 3858 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 3859 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 3860 . i - row indices 3861 . j - column indices 3862 - a - matrix values 3863 3864 Output Parameter: 3865 . mat - the matrix 3866 3867 Level: intermediate 3868 3869 Notes: 3870 The i, j, and a arrays ARE copied by this routine into the internal format used by PETSc; 3871 thus you CANNOT change the matrix entries by changing the values of a[] after you have 3872 called this routine. Use MatCreateMPIAIJWithSplitArrays() to avoid needing to copy the arrays. 3873 3874 The i and j indices are 0 based, and i indices are indices corresponding to the local j array. 3875 3876 The format which is used for the sparse matrix input, is equivalent to a 3877 row-major ordering.. i.e for the following matrix, the input data expected is 3878 as shown: 3879 3880 1 0 0 3881 2 0 3 P0 3882 ------- 3883 4 5 6 P1 3884 3885 Process0 [P0]: rows_owned=[0,1] 3886 i = {0,1,3} [size = nrow+1 = 2+1] 3887 j = {0,0,2} [size = nz = 6] 3888 v = {1,2,3} [size = nz = 6] 3889 3890 Process1 [P1]: rows_owned=[2] 3891 i = {0,3} [size = nrow+1 = 1+1] 3892 j = {0,1,2} [size = nz = 6] 3893 v = {4,5,6} [size = nz = 6] 3894 3895 .keywords: matrix, aij, compressed row, sparse, parallel 3896 3897 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIAIJSetPreallocation(), MatMPIAIJSetPreallocationCSR(), 3898 MPIAIJ, MatCreateMPIAIJ(), MatCreateMPIAIJWithSplitArrays() 3899 @*/ 3900 PetscErrorCode MatCreateMPIAIJWithArrays(MPI_Comm comm,PetscInt m,PetscInt n,PetscInt M,PetscInt N,const PetscInt i[],const PetscInt j[],const PetscScalar a[],Mat *mat) 3901 { 3902 PetscErrorCode ierr; 3903 3904 PetscFunctionBegin; 3905 if (i[0]) { 3906 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"i (row indices) must start with 0"); 3907 } 3908 if (m < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"local number of rows (m) cannot be PETSC_DECIDE, or negative"); 3909 ierr = MatCreate(comm,mat);CHKERRQ(ierr); 3910 ierr = MatSetSizes(*mat,m,n,M,N);CHKERRQ(ierr); 3911 ierr = MatSetType(*mat,MATMPIAIJ);CHKERRQ(ierr); 3912 ierr = MatMPIAIJSetPreallocationCSR(*mat,i,j,a);CHKERRQ(ierr); 3913 PetscFunctionReturn(0); 3914 } 3915 3916 #undef __FUNCT__ 3917 #define __FUNCT__ "MatCreateMPIAIJ" 3918 /*@C 3919 MatCreateMPIAIJ - Creates a sparse parallel matrix in AIJ format 3920 (the default parallel PETSc format). For good matrix assembly performance 3921 the user should preallocate the matrix storage by setting the parameters 3922 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 3923 performance can be increased by more than a factor of 50. 3924 3925 Collective on MPI_Comm 3926 3927 Input Parameters: 3928 + comm - MPI communicator 3929 . m - number of local rows (or PETSC_DECIDE to have calculated if M is given) 3930 This value should be the same as the local size used in creating the 3931 y vector for the matrix-vector product y = Ax. 3932 . n - This value should be the same as the local size used in creating the 3933 x vector for the matrix-vector product y = Ax. (or PETSC_DECIDE to have 3934 calculated if N is given) For square matrices n is almost always m. 3935 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 3936 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 3937 . d_nz - number of nonzeros per row in DIAGONAL portion of local submatrix 3938 (same value is used for all local rows) 3939 . d_nnz - array containing the number of nonzeros in the various rows of the 3940 DIAGONAL portion of the local submatrix (possibly different for each row) 3941 or PETSC_NULL, if d_nz is used to specify the nonzero structure. 3942 The size of this array is equal to the number of local rows, i.e 'm'. 3943 You must leave room for the diagonal entry even if it is zero. 3944 . o_nz - number of nonzeros per row in the OFF-DIAGONAL portion of local 3945 submatrix (same value is used for all local rows). 3946 - o_nnz - array containing the number of nonzeros in the various rows of the 3947 OFF-DIAGONAL portion of the local submatrix (possibly different for 3948 each row) or PETSC_NULL, if o_nz is used to specify the nonzero 3949 structure. The size of this array is equal to the number 3950 of local rows, i.e 'm'. 3951 3952 Output Parameter: 3953 . A - the matrix 3954 3955 It is recommended that one use the MatCreate(), MatSetType() and/or MatSetFromOptions(), 3956 MatXXXXSetPreallocation() paradgm instead of this routine directly. 3957 [MatXXXXSetPreallocation() is, for example, MatSeqAIJSetPreallocation] 3958 3959 Notes: 3960 If the *_nnz parameter is given then the *_nz parameter is ignored 3961 3962 m,n,M,N parameters specify the size of the matrix, and its partitioning across 3963 processors, while d_nz,d_nnz,o_nz,o_nnz parameters specify the approximate 3964 storage requirements for this matrix. 3965 3966 If PETSC_DECIDE or PETSC_DETERMINE is used for a particular argument on one 3967 processor than it must be used on all processors that share the object for 3968 that argument. 3969 3970 The user MUST specify either the local or global matrix dimensions 3971 (possibly both). 3972 3973 The parallel matrix is partitioned across processors such that the 3974 first m0 rows belong to process 0, the next m1 rows belong to 3975 process 1, the next m2 rows belong to process 2 etc.. where 3976 m0,m1,m2,.. are the input parameter 'm'. i.e each processor stores 3977 values corresponding to [m x N] submatrix. 3978 3979 The columns are logically partitioned with the n0 columns belonging 3980 to 0th partition, the next n1 columns belonging to the next 3981 partition etc.. where n0,n1,n2... are the the input parameter 'n'. 3982 3983 The DIAGONAL portion of the local submatrix on any given processor 3984 is the submatrix corresponding to the rows and columns m,n 3985 corresponding to the given processor. i.e diagonal matrix on 3986 process 0 is [m0 x n0], diagonal matrix on process 1 is [m1 x n1] 3987 etc. The remaining portion of the local submatrix [m x (N-n)] 3988 constitute the OFF-DIAGONAL portion. The example below better 3989 illustrates this concept. 3990 3991 For a square global matrix we define each processor's diagonal portion 3992 to be its local rows and the corresponding columns (a square submatrix); 3993 each processor's off-diagonal portion encompasses the remainder of the 3994 local matrix (a rectangular submatrix). 3995 3996 If o_nnz, d_nnz are specified, then o_nz, and d_nz are ignored. 3997 3998 When calling this routine with a single process communicator, a matrix of 3999 type SEQAIJ is returned. If a matrix of type MPIAIJ is desired for this 4000 type of communicator, use the construction mechanism: 4001 MatCreate(...,&A); MatSetType(A,MATMPIAIJ); MatSetSizes(A, m,n,M,N); MatMPIAIJSetPreallocation(A,...); 4002 4003 By default, this format uses inodes (identical nodes) when possible. 4004 We search for consecutive rows with the same nonzero structure, thereby 4005 reusing matrix information to achieve increased efficiency. 4006 4007 Options Database Keys: 4008 + -mat_no_inode - Do not use inodes 4009 . -mat_inode_limit <limit> - Sets inode limit (max limit=5) 4010 - -mat_aij_oneindex - Internally use indexing starting at 1 4011 rather than 0. Note that when calling MatSetValues(), 4012 the user still MUST index entries starting at 0! 4013 4014 4015 Example usage: 4016 4017 Consider the following 8x8 matrix with 34 non-zero values, that is 4018 assembled across 3 processors. Lets assume that proc0 owns 3 rows, 4019 proc1 owns 3 rows, proc2 owns 2 rows. This division can be shown 4020 as follows: 4021 4022 .vb 4023 1 2 0 | 0 3 0 | 0 4 4024 Proc0 0 5 6 | 7 0 0 | 8 0 4025 9 0 10 | 11 0 0 | 12 0 4026 ------------------------------------- 4027 13 0 14 | 15 16 17 | 0 0 4028 Proc1 0 18 0 | 19 20 21 | 0 0 4029 0 0 0 | 22 23 0 | 24 0 4030 ------------------------------------- 4031 Proc2 25 26 27 | 0 0 28 | 29 0 4032 30 0 0 | 31 32 33 | 0 34 4033 .ve 4034 4035 This can be represented as a collection of submatrices as: 4036 4037 .vb 4038 A B C 4039 D E F 4040 G H I 4041 .ve 4042 4043 Where the submatrices A,B,C are owned by proc0, D,E,F are 4044 owned by proc1, G,H,I are owned by proc2. 4045 4046 The 'm' parameters for proc0,proc1,proc2 are 3,3,2 respectively. 4047 The 'n' parameters for proc0,proc1,proc2 are 3,3,2 respectively. 4048 The 'M','N' parameters are 8,8, and have the same values on all procs. 4049 4050 The DIAGONAL submatrices corresponding to proc0,proc1,proc2 are 4051 submatrices [A], [E], [I] respectively. The OFF-DIAGONAL submatrices 4052 corresponding to proc0,proc1,proc2 are [BC], [DF], [GH] respectively. 4053 Internally, each processor stores the DIAGONAL part, and the OFF-DIAGONAL 4054 part as SeqAIJ matrices. for eg: proc1 will store [E] as a SeqAIJ 4055 matrix, ans [DF] as another SeqAIJ matrix. 4056 4057 When d_nz, o_nz parameters are specified, d_nz storage elements are 4058 allocated for every row of the local diagonal submatrix, and o_nz 4059 storage locations are allocated for every row of the OFF-DIAGONAL submat. 4060 One way to choose d_nz and o_nz is to use the max nonzerors per local 4061 rows for each of the local DIAGONAL, and the OFF-DIAGONAL submatrices. 4062 In this case, the values of d_nz,o_nz are: 4063 .vb 4064 proc0 : dnz = 2, o_nz = 2 4065 proc1 : dnz = 3, o_nz = 2 4066 proc2 : dnz = 1, o_nz = 4 4067 .ve 4068 We are allocating m*(d_nz+o_nz) storage locations for every proc. This 4069 translates to 3*(2+2)=12 for proc0, 3*(3+2)=15 for proc1, 2*(1+4)=10 4070 for proc3. i.e we are using 12+15+10=37 storage locations to store 4071 34 values. 4072 4073 When d_nnz, o_nnz parameters are specified, the storage is specified 4074 for every row, coresponding to both DIAGONAL and OFF-DIAGONAL submatrices. 4075 In the above case the values for d_nnz,o_nnz are: 4076 .vb 4077 proc0: d_nnz = [2,2,2] and o_nnz = [2,2,2] 4078 proc1: d_nnz = [3,3,2] and o_nnz = [2,1,1] 4079 proc2: d_nnz = [1,1] and o_nnz = [4,4] 4080 .ve 4081 Here the space allocated is sum of all the above values i.e 34, and 4082 hence pre-allocation is perfect. 4083 4084 Level: intermediate 4085 4086 .keywords: matrix, aij, compressed row, sparse, parallel 4087 4088 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIAIJSetPreallocation(), MatMPIAIJSetPreallocationCSR(), 4089 MPIAIJ, MatCreateMPIAIJWithArrays() 4090 @*/ 4091 PetscErrorCode MatCreateMPIAIJ(MPI_Comm comm,PetscInt m,PetscInt n,PetscInt M,PetscInt N,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[],Mat *A) 4092 { 4093 PetscErrorCode ierr; 4094 PetscMPIInt size; 4095 4096 PetscFunctionBegin; 4097 ierr = MatCreate(comm,A);CHKERRQ(ierr); 4098 ierr = MatSetSizes(*A,m,n,M,N);CHKERRQ(ierr); 4099 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 4100 if (size > 1) { 4101 ierr = MatSetType(*A,MATMPIAIJ);CHKERRQ(ierr); 4102 ierr = MatMPIAIJSetPreallocation(*A,d_nz,d_nnz,o_nz,o_nnz);CHKERRQ(ierr); 4103 } else { 4104 ierr = MatSetType(*A,MATSEQAIJ);CHKERRQ(ierr); 4105 ierr = MatSeqAIJSetPreallocation(*A,d_nz,d_nnz);CHKERRQ(ierr); 4106 } 4107 PetscFunctionReturn(0); 4108 } 4109 4110 #undef __FUNCT__ 4111 #define __FUNCT__ "MatMPIAIJGetSeqAIJ" 4112 PetscErrorCode MatMPIAIJGetSeqAIJ(Mat A,Mat *Ad,Mat *Ao,PetscInt *colmap[]) 4113 { 4114 Mat_MPIAIJ *a = (Mat_MPIAIJ *)A->data; 4115 4116 PetscFunctionBegin; 4117 *Ad = a->A; 4118 *Ao = a->B; 4119 *colmap = a->garray; 4120 PetscFunctionReturn(0); 4121 } 4122 4123 #undef __FUNCT__ 4124 #define __FUNCT__ "MatSetColoring_MPIAIJ" 4125 PetscErrorCode MatSetColoring_MPIAIJ(Mat A,ISColoring coloring) 4126 { 4127 PetscErrorCode ierr; 4128 PetscInt i; 4129 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 4130 4131 PetscFunctionBegin; 4132 if (coloring->ctype == IS_COLORING_GLOBAL) { 4133 ISColoringValue *allcolors,*colors; 4134 ISColoring ocoloring; 4135 4136 /* set coloring for diagonal portion */ 4137 ierr = MatSetColoring_SeqAIJ(a->A,coloring);CHKERRQ(ierr); 4138 4139 /* set coloring for off-diagonal portion */ 4140 ierr = ISAllGatherColors(((PetscObject)A)->comm,coloring->n,coloring->colors,PETSC_NULL,&allcolors);CHKERRQ(ierr); 4141 ierr = PetscMalloc((a->B->cmap->n+1)*sizeof(ISColoringValue),&colors);CHKERRQ(ierr); 4142 for (i=0; i<a->B->cmap->n; i++) { 4143 colors[i] = allcolors[a->garray[i]]; 4144 } 4145 ierr = PetscFree(allcolors);CHKERRQ(ierr); 4146 ierr = ISColoringCreate(MPI_COMM_SELF,coloring->n,a->B->cmap->n,colors,&ocoloring);CHKERRQ(ierr); 4147 ierr = MatSetColoring_SeqAIJ(a->B,ocoloring);CHKERRQ(ierr); 4148 ierr = ISColoringDestroy(ocoloring);CHKERRQ(ierr); 4149 } else if (coloring->ctype == IS_COLORING_GHOSTED) { 4150 ISColoringValue *colors; 4151 PetscInt *larray; 4152 ISColoring ocoloring; 4153 4154 /* set coloring for diagonal portion */ 4155 ierr = PetscMalloc((a->A->cmap->n+1)*sizeof(PetscInt),&larray);CHKERRQ(ierr); 4156 for (i=0; i<a->A->cmap->n; i++) { 4157 larray[i] = i + A->cmap->rstart; 4158 } 4159 ierr = ISGlobalToLocalMappingApply(A->cmapping,IS_GTOLM_MASK,a->A->cmap->n,larray,PETSC_NULL,larray);CHKERRQ(ierr); 4160 ierr = PetscMalloc((a->A->cmap->n+1)*sizeof(ISColoringValue),&colors);CHKERRQ(ierr); 4161 for (i=0; i<a->A->cmap->n; i++) { 4162 colors[i] = coloring->colors[larray[i]]; 4163 } 4164 ierr = PetscFree(larray);CHKERRQ(ierr); 4165 ierr = ISColoringCreate(PETSC_COMM_SELF,coloring->n,a->A->cmap->n,colors,&ocoloring);CHKERRQ(ierr); 4166 ierr = MatSetColoring_SeqAIJ(a->A,ocoloring);CHKERRQ(ierr); 4167 ierr = ISColoringDestroy(ocoloring);CHKERRQ(ierr); 4168 4169 /* set coloring for off-diagonal portion */ 4170 ierr = PetscMalloc((a->B->cmap->n+1)*sizeof(PetscInt),&larray);CHKERRQ(ierr); 4171 ierr = ISGlobalToLocalMappingApply(A->cmapping,IS_GTOLM_MASK,a->B->cmap->n,a->garray,PETSC_NULL,larray);CHKERRQ(ierr); 4172 ierr = PetscMalloc((a->B->cmap->n+1)*sizeof(ISColoringValue),&colors);CHKERRQ(ierr); 4173 for (i=0; i<a->B->cmap->n; i++) { 4174 colors[i] = coloring->colors[larray[i]]; 4175 } 4176 ierr = PetscFree(larray);CHKERRQ(ierr); 4177 ierr = ISColoringCreate(MPI_COMM_SELF,coloring->n,a->B->cmap->n,colors,&ocoloring);CHKERRQ(ierr); 4178 ierr = MatSetColoring_SeqAIJ(a->B,ocoloring);CHKERRQ(ierr); 4179 ierr = ISColoringDestroy(ocoloring);CHKERRQ(ierr); 4180 } else { 4181 SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"No support ISColoringType %d",(int)coloring->ctype); 4182 } 4183 4184 PetscFunctionReturn(0); 4185 } 4186 4187 #if defined(PETSC_HAVE_ADIC) 4188 #undef __FUNCT__ 4189 #define __FUNCT__ "MatSetValuesAdic_MPIAIJ" 4190 PetscErrorCode MatSetValuesAdic_MPIAIJ(Mat A,void *advalues) 4191 { 4192 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 4193 PetscErrorCode ierr; 4194 4195 PetscFunctionBegin; 4196 ierr = MatSetValuesAdic_SeqAIJ(a->A,advalues);CHKERRQ(ierr); 4197 ierr = MatSetValuesAdic_SeqAIJ(a->B,advalues);CHKERRQ(ierr); 4198 PetscFunctionReturn(0); 4199 } 4200 #endif 4201 4202 #undef __FUNCT__ 4203 #define __FUNCT__ "MatSetValuesAdifor_MPIAIJ" 4204 PetscErrorCode MatSetValuesAdifor_MPIAIJ(Mat A,PetscInt nl,void *advalues) 4205 { 4206 Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data; 4207 PetscErrorCode ierr; 4208 4209 PetscFunctionBegin; 4210 ierr = MatSetValuesAdifor_SeqAIJ(a->A,nl,advalues);CHKERRQ(ierr); 4211 ierr = MatSetValuesAdifor_SeqAIJ(a->B,nl,advalues);CHKERRQ(ierr); 4212 PetscFunctionReturn(0); 4213 } 4214 4215 #undef __FUNCT__ 4216 #define __FUNCT__ "MatMerge" 4217 /*@ 4218 MatMerge - Creates a single large PETSc matrix by concatinating sequential 4219 matrices from each processor 4220 4221 Collective on MPI_Comm 4222 4223 Input Parameters: 4224 + comm - the communicators the parallel matrix will live on 4225 . inmat - the input sequential matrices 4226 . n - number of local columns (or PETSC_DECIDE) 4227 - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 4228 4229 Output Parameter: 4230 . outmat - the parallel matrix generated 4231 4232 Level: advanced 4233 4234 Notes: The number of columns of the matrix in EACH processor MUST be the same. 4235 4236 @*/ 4237 PetscErrorCode MatMerge(MPI_Comm comm,Mat inmat,PetscInt n,MatReuse scall,Mat *outmat) 4238 { 4239 PetscErrorCode ierr; 4240 PetscInt m,N,i,rstart,nnz,Ii,*dnz,*onz; 4241 PetscInt *indx; 4242 PetscScalar *values; 4243 4244 PetscFunctionBegin; 4245 ierr = MatGetSize(inmat,&m,&N);CHKERRQ(ierr); 4246 if (scall == MAT_INITIAL_MATRIX){ 4247 /* count nonzeros in each row, for diagonal and off diagonal portion of matrix */ 4248 if (n == PETSC_DECIDE){ 4249 ierr = PetscSplitOwnership(comm,&n,&N);CHKERRQ(ierr); 4250 } 4251 ierr = MPI_Scan(&m, &rstart,1,MPIU_INT,MPI_SUM,comm);CHKERRQ(ierr); 4252 rstart -= m; 4253 4254 ierr = MatPreallocateInitialize(comm,m,n,dnz,onz);CHKERRQ(ierr); 4255 for (i=0;i<m;i++) { 4256 ierr = MatGetRow_SeqAIJ(inmat,i,&nnz,&indx,PETSC_NULL);CHKERRQ(ierr); 4257 ierr = MatPreallocateSet(i+rstart,nnz,indx,dnz,onz);CHKERRQ(ierr); 4258 ierr = MatRestoreRow_SeqAIJ(inmat,i,&nnz,&indx,PETSC_NULL);CHKERRQ(ierr); 4259 } 4260 /* This routine will ONLY return MPIAIJ type matrix */ 4261 ierr = MatCreate(comm,outmat);CHKERRQ(ierr); 4262 ierr = MatSetSizes(*outmat,m,n,PETSC_DETERMINE,PETSC_DETERMINE);CHKERRQ(ierr); 4263 ierr = MatSetType(*outmat,MATMPIAIJ);CHKERRQ(ierr); 4264 ierr = MatMPIAIJSetPreallocation(*outmat,0,dnz,0,onz);CHKERRQ(ierr); 4265 ierr = MatPreallocateFinalize(dnz,onz);CHKERRQ(ierr); 4266 4267 } else if (scall == MAT_REUSE_MATRIX){ 4268 ierr = MatGetOwnershipRange(*outmat,&rstart,PETSC_NULL);CHKERRQ(ierr); 4269 } else { 4270 SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Invalid MatReuse %d",(int)scall); 4271 } 4272 4273 for (i=0;i<m;i++) { 4274 ierr = MatGetRow_SeqAIJ(inmat,i,&nnz,&indx,&values);CHKERRQ(ierr); 4275 Ii = i + rstart; 4276 ierr = MatSetValues(*outmat,1,&Ii,nnz,indx,values,INSERT_VALUES);CHKERRQ(ierr); 4277 ierr = MatRestoreRow_SeqAIJ(inmat,i,&nnz,&indx,&values);CHKERRQ(ierr); 4278 } 4279 ierr = MatDestroy(inmat);CHKERRQ(ierr); 4280 ierr = MatAssemblyBegin(*outmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4281 ierr = MatAssemblyEnd(*outmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4282 4283 PetscFunctionReturn(0); 4284 } 4285 4286 #undef __FUNCT__ 4287 #define __FUNCT__ "MatFileSplit" 4288 PetscErrorCode MatFileSplit(Mat A,char *outfile) 4289 { 4290 PetscErrorCode ierr; 4291 PetscMPIInt rank; 4292 PetscInt m,N,i,rstart,nnz; 4293 size_t len; 4294 const PetscInt *indx; 4295 PetscViewer out; 4296 char *name; 4297 Mat B; 4298 const PetscScalar *values; 4299 4300 PetscFunctionBegin; 4301 ierr = MatGetLocalSize(A,&m,0);CHKERRQ(ierr); 4302 ierr = MatGetSize(A,0,&N);CHKERRQ(ierr); 4303 /* Should this be the type of the diagonal block of A? */ 4304 ierr = MatCreate(PETSC_COMM_SELF,&B);CHKERRQ(ierr); 4305 ierr = MatSetSizes(B,m,N,m,N);CHKERRQ(ierr); 4306 ierr = MatSetType(B,MATSEQAIJ);CHKERRQ(ierr); 4307 ierr = MatSeqAIJSetPreallocation(B,0,PETSC_NULL);CHKERRQ(ierr); 4308 ierr = MatGetOwnershipRange(A,&rstart,0);CHKERRQ(ierr); 4309 for (i=0;i<m;i++) { 4310 ierr = MatGetRow(A,i+rstart,&nnz,&indx,&values);CHKERRQ(ierr); 4311 ierr = MatSetValues(B,1,&i,nnz,indx,values,INSERT_VALUES);CHKERRQ(ierr); 4312 ierr = MatRestoreRow(A,i+rstart,&nnz,&indx,&values);CHKERRQ(ierr); 4313 } 4314 ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4315 ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4316 4317 ierr = MPI_Comm_rank(((PetscObject)A)->comm,&rank);CHKERRQ(ierr); 4318 ierr = PetscStrlen(outfile,&len);CHKERRQ(ierr); 4319 ierr = PetscMalloc((len+5)*sizeof(char),&name);CHKERRQ(ierr); 4320 sprintf(name,"%s.%d",outfile,rank); 4321 ierr = PetscViewerBinaryOpen(PETSC_COMM_SELF,name,FILE_MODE_APPEND,&out);CHKERRQ(ierr); 4322 ierr = PetscFree(name); 4323 ierr = MatView(B,out);CHKERRQ(ierr); 4324 ierr = PetscViewerDestroy(out);CHKERRQ(ierr); 4325 ierr = MatDestroy(B);CHKERRQ(ierr); 4326 PetscFunctionReturn(0); 4327 } 4328 4329 extern PetscErrorCode MatDestroy_MPIAIJ(Mat); 4330 #undef __FUNCT__ 4331 #define __FUNCT__ "MatDestroy_MPIAIJ_SeqsToMPI" 4332 PetscErrorCode MatDestroy_MPIAIJ_SeqsToMPI(Mat A) 4333 { 4334 PetscErrorCode ierr; 4335 Mat_Merge_SeqsToMPI *merge; 4336 PetscContainer container; 4337 4338 PetscFunctionBegin; 4339 ierr = PetscObjectQuery((PetscObject)A,"MatMergeSeqsToMPI",(PetscObject *)&container);CHKERRQ(ierr); 4340 if (container) { 4341 ierr = PetscContainerGetPointer(container,(void **)&merge);CHKERRQ(ierr); 4342 ierr = PetscFree(merge->id_r);CHKERRQ(ierr); 4343 ierr = PetscFree(merge->len_s);CHKERRQ(ierr); 4344 ierr = PetscFree(merge->len_r);CHKERRQ(ierr); 4345 ierr = PetscFree(merge->bi);CHKERRQ(ierr); 4346 ierr = PetscFree(merge->bj);CHKERRQ(ierr); 4347 ierr = PetscFree(merge->buf_ri[0]);CHKERRQ(ierr); 4348 ierr = PetscFree(merge->buf_ri);CHKERRQ(ierr); 4349 ierr = PetscFree(merge->buf_rj[0]);CHKERRQ(ierr); 4350 ierr = PetscFree(merge->buf_rj);CHKERRQ(ierr); 4351 ierr = PetscFree(merge->coi);CHKERRQ(ierr); 4352 ierr = PetscFree(merge->coj);CHKERRQ(ierr); 4353 ierr = PetscFree(merge->owners_co);CHKERRQ(ierr); 4354 ierr = PetscLayoutDestroy(merge->rowmap);CHKERRQ(ierr); 4355 4356 ierr = PetscContainerDestroy(container);CHKERRQ(ierr); 4357 ierr = PetscObjectCompose((PetscObject)A,"MatMergeSeqsToMPI",0);CHKERRQ(ierr); 4358 } 4359 ierr = PetscFree(merge);CHKERRQ(ierr); 4360 4361 ierr = MatDestroy_MPIAIJ(A);CHKERRQ(ierr); 4362 PetscFunctionReturn(0); 4363 } 4364 4365 #include <../src/mat/utils/freespace.h> 4366 #include <petscbt.h> 4367 4368 #undef __FUNCT__ 4369 #define __FUNCT__ "MatMerge_SeqsToMPINumeric" 4370 /*@C 4371 MatMerge_SeqsToMPI - Creates a MPIAIJ matrix by adding sequential 4372 matrices from each processor 4373 4374 Collective on MPI_Comm 4375 4376 Input Parameters: 4377 + comm - the communicators the parallel matrix will live on 4378 . seqmat - the input sequential matrices 4379 . m - number of local rows (or PETSC_DECIDE) 4380 . n - number of local columns (or PETSC_DECIDE) 4381 - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 4382 4383 Output Parameter: 4384 . mpimat - the parallel matrix generated 4385 4386 Level: advanced 4387 4388 Notes: 4389 The dimensions of the sequential matrix in each processor MUST be the same. 4390 The input seqmat is included into the container "Mat_Merge_SeqsToMPI", and will be 4391 destroyed when mpimat is destroyed. Call PetscObjectQuery() to access seqmat. 4392 @*/ 4393 PetscErrorCode MatMerge_SeqsToMPINumeric(Mat seqmat,Mat mpimat) 4394 { 4395 PetscErrorCode ierr; 4396 MPI_Comm comm=((PetscObject)mpimat)->comm; 4397 Mat_SeqAIJ *a=(Mat_SeqAIJ*)seqmat->data; 4398 PetscMPIInt size,rank,taga,*len_s; 4399 PetscInt N=mpimat->cmap->N,i,j,*owners,*ai=a->i,*aj=a->j; 4400 PetscInt proc,m; 4401 PetscInt **buf_ri,**buf_rj; 4402 PetscInt k,anzi,*bj_i,*bi,*bj,arow,bnzi,nextaj; 4403 PetscInt nrows,**buf_ri_k,**nextrow,**nextai; 4404 MPI_Request *s_waits,*r_waits; 4405 MPI_Status *status; 4406 MatScalar *aa=a->a; 4407 MatScalar **abuf_r,*ba_i; 4408 Mat_Merge_SeqsToMPI *merge; 4409 PetscContainer container; 4410 4411 PetscFunctionBegin; 4412 ierr = PetscLogEventBegin(MAT_Seqstompinum,seqmat,0,0,0);CHKERRQ(ierr); 4413 4414 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 4415 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 4416 4417 ierr = PetscObjectQuery((PetscObject)mpimat,"MatMergeSeqsToMPI",(PetscObject *)&container);CHKERRQ(ierr); 4418 if (container) { 4419 ierr = PetscContainerGetPointer(container,(void **)&merge);CHKERRQ(ierr); 4420 } 4421 bi = merge->bi; 4422 bj = merge->bj; 4423 buf_ri = merge->buf_ri; 4424 buf_rj = merge->buf_rj; 4425 4426 ierr = PetscMalloc(size*sizeof(MPI_Status),&status);CHKERRQ(ierr); 4427 owners = merge->rowmap->range; 4428 len_s = merge->len_s; 4429 4430 /* send and recv matrix values */ 4431 /*-----------------------------*/ 4432 ierr = PetscObjectGetNewTag((PetscObject)mpimat,&taga);CHKERRQ(ierr); 4433 ierr = PetscPostIrecvScalar(comm,taga,merge->nrecv,merge->id_r,merge->len_r,&abuf_r,&r_waits);CHKERRQ(ierr); 4434 4435 ierr = PetscMalloc((merge->nsend+1)*sizeof(MPI_Request),&s_waits);CHKERRQ(ierr); 4436 for (proc=0,k=0; proc<size; proc++){ 4437 if (!len_s[proc]) continue; 4438 i = owners[proc]; 4439 ierr = MPI_Isend(aa+ai[i],len_s[proc],MPIU_MATSCALAR,proc,taga,comm,s_waits+k);CHKERRQ(ierr); 4440 k++; 4441 } 4442 4443 if (merge->nrecv) {ierr = MPI_Waitall(merge->nrecv,r_waits,status);CHKERRQ(ierr);} 4444 if (merge->nsend) {ierr = MPI_Waitall(merge->nsend,s_waits,status);CHKERRQ(ierr);} 4445 ierr = PetscFree(status);CHKERRQ(ierr); 4446 4447 ierr = PetscFree(s_waits);CHKERRQ(ierr); 4448 ierr = PetscFree(r_waits);CHKERRQ(ierr); 4449 4450 /* insert mat values of mpimat */ 4451 /*----------------------------*/ 4452 ierr = PetscMalloc(N*sizeof(PetscScalar),&ba_i);CHKERRQ(ierr); 4453 ierr = PetscMalloc3(merge->nrecv,PetscInt*,&buf_ri_k,merge->nrecv,PetscInt*,&nextrow,merge->nrecv,PetscInt*,&nextai);CHKERRQ(ierr); 4454 4455 for (k=0; k<merge->nrecv; k++){ 4456 buf_ri_k[k] = buf_ri[k]; /* beginning of k-th recved i-structure */ 4457 nrows = *(buf_ri_k[k]); 4458 nextrow[k] = buf_ri_k[k]+1; /* next row number of k-th recved i-structure */ 4459 nextai[k] = buf_ri_k[k] + (nrows + 1);/* poins to the next i-structure of k-th recved i-structure */ 4460 } 4461 4462 /* set values of ba */ 4463 m = merge->rowmap->n; 4464 for (i=0; i<m; i++) { 4465 arow = owners[rank] + i; 4466 bj_i = bj+bi[i]; /* col indices of the i-th row of mpimat */ 4467 bnzi = bi[i+1] - bi[i]; 4468 ierr = PetscMemzero(ba_i,bnzi*sizeof(PetscScalar));CHKERRQ(ierr); 4469 4470 /* add local non-zero vals of this proc's seqmat into ba */ 4471 anzi = ai[arow+1] - ai[arow]; 4472 aj = a->j + ai[arow]; 4473 aa = a->a + ai[arow]; 4474 nextaj = 0; 4475 for (j=0; nextaj<anzi; j++){ 4476 if (*(bj_i + j) == aj[nextaj]){ /* bcol == acol */ 4477 ba_i[j] += aa[nextaj++]; 4478 } 4479 } 4480 4481 /* add received vals into ba */ 4482 for (k=0; k<merge->nrecv; k++){ /* k-th received message */ 4483 /* i-th row */ 4484 if (i == *nextrow[k]) { 4485 anzi = *(nextai[k]+1) - *nextai[k]; 4486 aj = buf_rj[k] + *(nextai[k]); 4487 aa = abuf_r[k] + *(nextai[k]); 4488 nextaj = 0; 4489 for (j=0; nextaj<anzi; j++){ 4490 if (*(bj_i + j) == aj[nextaj]){ /* bcol == acol */ 4491 ba_i[j] += aa[nextaj++]; 4492 } 4493 } 4494 nextrow[k]++; nextai[k]++; 4495 } 4496 } 4497 ierr = MatSetValues(mpimat,1,&arow,bnzi,bj_i,ba_i,INSERT_VALUES);CHKERRQ(ierr); 4498 } 4499 ierr = MatAssemblyBegin(mpimat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4500 ierr = MatAssemblyEnd(mpimat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 4501 4502 ierr = PetscFree(abuf_r[0]);CHKERRQ(ierr); 4503 ierr = PetscFree(abuf_r);CHKERRQ(ierr); 4504 ierr = PetscFree(ba_i);CHKERRQ(ierr); 4505 ierr = PetscFree3(buf_ri_k,nextrow,nextai);CHKERRQ(ierr); 4506 ierr = PetscLogEventEnd(MAT_Seqstompinum,seqmat,0,0,0);CHKERRQ(ierr); 4507 PetscFunctionReturn(0); 4508 } 4509 4510 #undef __FUNCT__ 4511 #define __FUNCT__ "MatMerge_SeqsToMPISymbolic" 4512 PetscErrorCode MatMerge_SeqsToMPISymbolic(MPI_Comm comm,Mat seqmat,PetscInt m,PetscInt n,Mat *mpimat) 4513 { 4514 PetscErrorCode ierr; 4515 Mat B_mpi; 4516 Mat_SeqAIJ *a=(Mat_SeqAIJ*)seqmat->data; 4517 PetscMPIInt size,rank,tagi,tagj,*len_s,*len_si,*len_ri; 4518 PetscInt **buf_rj,**buf_ri,**buf_ri_k; 4519 PetscInt M=seqmat->rmap->n,N=seqmat->cmap->n,i,*owners,*ai=a->i,*aj=a->j; 4520 PetscInt len,proc,*dnz,*onz; 4521 PetscInt k,anzi,*bi,*bj,*lnk,nlnk,arow,bnzi,nspacedouble=0; 4522 PetscInt nrows,*buf_s,*buf_si,*buf_si_i,**nextrow,**nextai; 4523 MPI_Request *si_waits,*sj_waits,*ri_waits,*rj_waits; 4524 MPI_Status *status; 4525 PetscFreeSpaceList free_space=PETSC_NULL,current_space=PETSC_NULL; 4526 PetscBT lnkbt; 4527 Mat_Merge_SeqsToMPI *merge; 4528 PetscContainer container; 4529 4530 PetscFunctionBegin; 4531 ierr = PetscLogEventBegin(MAT_Seqstompisym,seqmat,0,0,0);CHKERRQ(ierr); 4532 4533 /* make sure it is a PETSc comm */ 4534 ierr = PetscCommDuplicate(comm,&comm,PETSC_NULL);CHKERRQ(ierr); 4535 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 4536 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 4537 4538 ierr = PetscNew(Mat_Merge_SeqsToMPI,&merge);CHKERRQ(ierr); 4539 ierr = PetscMalloc(size*sizeof(MPI_Status),&status);CHKERRQ(ierr); 4540 4541 /* determine row ownership */ 4542 /*---------------------------------------------------------*/ 4543 ierr = PetscLayoutCreate(comm,&merge->rowmap);CHKERRQ(ierr); 4544 ierr = PetscLayoutSetLocalSize(merge->rowmap,m);CHKERRQ(ierr); 4545 ierr = PetscLayoutSetSize(merge->rowmap,M);CHKERRQ(ierr); 4546 ierr = PetscLayoutSetBlockSize(merge->rowmap,1);CHKERRQ(ierr); 4547 ierr = PetscLayoutSetUp(merge->rowmap);CHKERRQ(ierr); 4548 ierr = PetscMalloc(size*sizeof(PetscMPIInt),&len_si);CHKERRQ(ierr); 4549 ierr = PetscMalloc(size*sizeof(PetscMPIInt),&merge->len_s);CHKERRQ(ierr); 4550 4551 m = merge->rowmap->n; 4552 M = merge->rowmap->N; 4553 owners = merge->rowmap->range; 4554 4555 /* determine the number of messages to send, their lengths */ 4556 /*---------------------------------------------------------*/ 4557 len_s = merge->len_s; 4558 4559 len = 0; /* length of buf_si[] */ 4560 merge->nsend = 0; 4561 for (proc=0; proc<size; proc++){ 4562 len_si[proc] = 0; 4563 if (proc == rank){ 4564 len_s[proc] = 0; 4565 } else { 4566 len_si[proc] = owners[proc+1] - owners[proc] + 1; 4567 len_s[proc] = ai[owners[proc+1]] - ai[owners[proc]]; /* num of rows to be sent to [proc] */ 4568 } 4569 if (len_s[proc]) { 4570 merge->nsend++; 4571 nrows = 0; 4572 for (i=owners[proc]; i<owners[proc+1]; i++){ 4573 if (ai[i+1] > ai[i]) nrows++; 4574 } 4575 len_si[proc] = 2*(nrows+1); 4576 len += len_si[proc]; 4577 } 4578 } 4579 4580 /* determine the number and length of messages to receive for ij-structure */ 4581 /*-------------------------------------------------------------------------*/ 4582 ierr = PetscGatherNumberOfMessages(comm,PETSC_NULL,len_s,&merge->nrecv);CHKERRQ(ierr); 4583 ierr = PetscGatherMessageLengths2(comm,merge->nsend,merge->nrecv,len_s,len_si,&merge->id_r,&merge->len_r,&len_ri);CHKERRQ(ierr); 4584 4585 /* post the Irecv of j-structure */ 4586 /*-------------------------------*/ 4587 ierr = PetscCommGetNewTag(comm,&tagj);CHKERRQ(ierr); 4588 ierr = PetscPostIrecvInt(comm,tagj,merge->nrecv,merge->id_r,merge->len_r,&buf_rj,&rj_waits);CHKERRQ(ierr); 4589 4590 /* post the Isend of j-structure */ 4591 /*--------------------------------*/ 4592 ierr = PetscMalloc2(merge->nsend,MPI_Request,&si_waits,merge->nsend,MPI_Request,&sj_waits);CHKERRQ(ierr); 4593 4594 for (proc=0, k=0; proc<size; proc++){ 4595 if (!len_s[proc]) continue; 4596 i = owners[proc]; 4597 ierr = MPI_Isend(aj+ai[i],len_s[proc],MPIU_INT,proc,tagj,comm,sj_waits+k);CHKERRQ(ierr); 4598 k++; 4599 } 4600 4601 /* receives and sends of j-structure are complete */ 4602 /*------------------------------------------------*/ 4603 if (merge->nrecv) {ierr = MPI_Waitall(merge->nrecv,rj_waits,status);CHKERRQ(ierr);} 4604 if (merge->nsend) {ierr = MPI_Waitall(merge->nsend,sj_waits,status);CHKERRQ(ierr);} 4605 4606 /* send and recv i-structure */ 4607 /*---------------------------*/ 4608 ierr = PetscCommGetNewTag(comm,&tagi);CHKERRQ(ierr); 4609 ierr = PetscPostIrecvInt(comm,tagi,merge->nrecv,merge->id_r,len_ri,&buf_ri,&ri_waits);CHKERRQ(ierr); 4610 4611 ierr = PetscMalloc((len+1)*sizeof(PetscInt),&buf_s);CHKERRQ(ierr); 4612 buf_si = buf_s; /* points to the beginning of k-th msg to be sent */ 4613 for (proc=0,k=0; proc<size; proc++){ 4614 if (!len_s[proc]) continue; 4615 /* form outgoing message for i-structure: 4616 buf_si[0]: nrows to be sent 4617 [1:nrows]: row index (global) 4618 [nrows+1:2*nrows+1]: i-structure index 4619 */ 4620 /*-------------------------------------------*/ 4621 nrows = len_si[proc]/2 - 1; 4622 buf_si_i = buf_si + nrows+1; 4623 buf_si[0] = nrows; 4624 buf_si_i[0] = 0; 4625 nrows = 0; 4626 for (i=owners[proc]; i<owners[proc+1]; i++){ 4627 anzi = ai[i+1] - ai[i]; 4628 if (anzi) { 4629 buf_si_i[nrows+1] = buf_si_i[nrows] + anzi; /* i-structure */ 4630 buf_si[nrows+1] = i-owners[proc]; /* local row index */ 4631 nrows++; 4632 } 4633 } 4634 ierr = MPI_Isend(buf_si,len_si[proc],MPIU_INT,proc,tagi,comm,si_waits+k);CHKERRQ(ierr); 4635 k++; 4636 buf_si += len_si[proc]; 4637 } 4638 4639 if (merge->nrecv) {ierr = MPI_Waitall(merge->nrecv,ri_waits,status);CHKERRQ(ierr);} 4640 if (merge->nsend) {ierr = MPI_Waitall(merge->nsend,si_waits,status);CHKERRQ(ierr);} 4641 4642 ierr = PetscInfo2(seqmat,"nsend: %D, nrecv: %D\n",merge->nsend,merge->nrecv);CHKERRQ(ierr); 4643 for (i=0; i<merge->nrecv; i++){ 4644 ierr = PetscInfo3(seqmat,"recv len_ri=%D, len_rj=%D from [%D]\n",len_ri[i],merge->len_r[i],merge->id_r[i]);CHKERRQ(ierr); 4645 } 4646 4647 ierr = PetscFree(len_si);CHKERRQ(ierr); 4648 ierr = PetscFree(len_ri);CHKERRQ(ierr); 4649 ierr = PetscFree(rj_waits);CHKERRQ(ierr); 4650 ierr = PetscFree2(si_waits,sj_waits);CHKERRQ(ierr); 4651 ierr = PetscFree(ri_waits);CHKERRQ(ierr); 4652 ierr = PetscFree(buf_s);CHKERRQ(ierr); 4653 ierr = PetscFree(status);CHKERRQ(ierr); 4654 4655 /* compute a local seq matrix in each processor */ 4656 /*----------------------------------------------*/ 4657 /* allocate bi array and free space for accumulating nonzero column info */ 4658 ierr = PetscMalloc((m+1)*sizeof(PetscInt),&bi);CHKERRQ(ierr); 4659 bi[0] = 0; 4660 4661 /* create and initialize a linked list */ 4662 nlnk = N+1; 4663 ierr = PetscLLCreate(N,N,nlnk,lnk,lnkbt);CHKERRQ(ierr); 4664 4665 /* initial FreeSpace size is 2*(num of local nnz(seqmat)) */ 4666 len = 0; 4667 len = ai[owners[rank+1]] - ai[owners[rank]]; 4668 ierr = PetscFreeSpaceGet((PetscInt)(2*len+1),&free_space);CHKERRQ(ierr); 4669 current_space = free_space; 4670 4671 /* determine symbolic info for each local row */ 4672 ierr = PetscMalloc3(merge->nrecv,PetscInt*,&buf_ri_k,merge->nrecv,PetscInt*,&nextrow,merge->nrecv,PetscInt*,&nextai);CHKERRQ(ierr); 4673 4674 for (k=0; k<merge->nrecv; k++){ 4675 buf_ri_k[k] = buf_ri[k]; /* beginning of k-th recved i-structure */ 4676 nrows = *buf_ri_k[k]; 4677 nextrow[k] = buf_ri_k[k] + 1; /* next row number of k-th recved i-structure */ 4678 nextai[k] = buf_ri_k[k] + (nrows + 1);/* poins to the next i-structure of k-th recved i-structure */ 4679 } 4680 4681 ierr = MatPreallocateInitialize(comm,m,n,dnz,onz);CHKERRQ(ierr); 4682 len = 0; 4683 for (i=0;i<m;i++) { 4684 bnzi = 0; 4685 /* add local non-zero cols of this proc's seqmat into lnk */ 4686 arow = owners[rank] + i; 4687 anzi = ai[arow+1] - ai[arow]; 4688 aj = a->j + ai[arow]; 4689 ierr = PetscLLAdd(anzi,aj,N,nlnk,lnk,lnkbt);CHKERRQ(ierr); 4690 bnzi += nlnk; 4691 /* add received col data into lnk */ 4692 for (k=0; k<merge->nrecv; k++){ /* k-th received message */ 4693 if (i == *nextrow[k]) { /* i-th row */ 4694 anzi = *(nextai[k]+1) - *nextai[k]; 4695 aj = buf_rj[k] + *nextai[k]; 4696 ierr = PetscLLAdd(anzi,aj,N,nlnk,lnk,lnkbt);CHKERRQ(ierr); 4697 bnzi += nlnk; 4698 nextrow[k]++; nextai[k]++; 4699 } 4700 } 4701 if (len < bnzi) len = bnzi; /* =max(bnzi) */ 4702 4703 /* if free space is not available, make more free space */ 4704 if (current_space->local_remaining<bnzi) { 4705 ierr = PetscFreeSpaceGet(bnzi+current_space->total_array_size,¤t_space);CHKERRQ(ierr); 4706 nspacedouble++; 4707 } 4708 /* copy data into free space, then initialize lnk */ 4709 ierr = PetscLLClean(N,N,bnzi,lnk,current_space->array,lnkbt);CHKERRQ(ierr); 4710 ierr = MatPreallocateSet(i+owners[rank],bnzi,current_space->array,dnz,onz);CHKERRQ(ierr); 4711 4712 current_space->array += bnzi; 4713 current_space->local_used += bnzi; 4714 current_space->local_remaining -= bnzi; 4715 4716 bi[i+1] = bi[i] + bnzi; 4717 } 4718 4719 ierr = PetscFree3(buf_ri_k,nextrow,nextai);CHKERRQ(ierr); 4720 4721 ierr = PetscMalloc((bi[m]+1)*sizeof(PetscInt),&bj);CHKERRQ(ierr); 4722 ierr = PetscFreeSpaceContiguous(&free_space,bj);CHKERRQ(ierr); 4723 ierr = PetscLLDestroy(lnk,lnkbt);CHKERRQ(ierr); 4724 4725 /* create symbolic parallel matrix B_mpi */ 4726 /*---------------------------------------*/ 4727 ierr = MatCreate(comm,&B_mpi);CHKERRQ(ierr); 4728 if (n==PETSC_DECIDE) { 4729 ierr = MatSetSizes(B_mpi,m,n,PETSC_DETERMINE,N);CHKERRQ(ierr); 4730 } else { 4731 ierr = MatSetSizes(B_mpi,m,n,PETSC_DETERMINE,PETSC_DETERMINE);CHKERRQ(ierr); 4732 } 4733 ierr = MatSetType(B_mpi,MATMPIAIJ);CHKERRQ(ierr); 4734 ierr = MatMPIAIJSetPreallocation(B_mpi,0,dnz,0,onz);CHKERRQ(ierr); 4735 ierr = MatPreallocateFinalize(dnz,onz);CHKERRQ(ierr); 4736 4737 /* B_mpi is not ready for use - assembly will be done by MatMerge_SeqsToMPINumeric() */ 4738 B_mpi->assembled = PETSC_FALSE; 4739 B_mpi->ops->destroy = MatDestroy_MPIAIJ_SeqsToMPI; 4740 merge->bi = bi; 4741 merge->bj = bj; 4742 merge->buf_ri = buf_ri; 4743 merge->buf_rj = buf_rj; 4744 merge->coi = PETSC_NULL; 4745 merge->coj = PETSC_NULL; 4746 merge->owners_co = PETSC_NULL; 4747 4748 /* attach the supporting struct to B_mpi for reuse */ 4749 ierr = PetscContainerCreate(PETSC_COMM_SELF,&container);CHKERRQ(ierr); 4750 ierr = PetscContainerSetPointer(container,merge);CHKERRQ(ierr); 4751 ierr = PetscObjectCompose((PetscObject)B_mpi,"MatMergeSeqsToMPI",(PetscObject)container);CHKERRQ(ierr); 4752 *mpimat = B_mpi; 4753 4754 ierr = PetscCommDestroy(&comm);CHKERRQ(ierr); 4755 ierr = PetscLogEventEnd(MAT_Seqstompisym,seqmat,0,0,0);CHKERRQ(ierr); 4756 PetscFunctionReturn(0); 4757 } 4758 4759 #undef __FUNCT__ 4760 #define __FUNCT__ "MatMerge_SeqsToMPI" 4761 PetscErrorCode MatMerge_SeqsToMPI(MPI_Comm comm,Mat seqmat,PetscInt m,PetscInt n,MatReuse scall,Mat *mpimat) 4762 { 4763 PetscErrorCode ierr; 4764 4765 PetscFunctionBegin; 4766 ierr = PetscLogEventBegin(MAT_Seqstompi,seqmat,0,0,0);CHKERRQ(ierr); 4767 if (scall == MAT_INITIAL_MATRIX){ 4768 ierr = MatMerge_SeqsToMPISymbolic(comm,seqmat,m,n,mpimat);CHKERRQ(ierr); 4769 } 4770 ierr = MatMerge_SeqsToMPINumeric(seqmat,*mpimat);CHKERRQ(ierr); 4771 ierr = PetscLogEventEnd(MAT_Seqstompi,seqmat,0,0,0);CHKERRQ(ierr); 4772 PetscFunctionReturn(0); 4773 } 4774 4775 #undef __FUNCT__ 4776 #define __FUNCT__ "MatMPIAIJGetLocalMat" 4777 /*@ 4778 MatMPIAIJGetLocalMat - Creates a SeqAIJ from a MPIAIJ matrix by taking all its local rows and putting them into a sequential vector with 4779 mlocal rows and n columns. Where mlocal is the row count obtained with MatGetLocalSize() and n is the global column count obtained 4780 with MatGetSize() 4781 4782 Not Collective 4783 4784 Input Parameters: 4785 + A - the matrix 4786 . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 4787 4788 Output Parameter: 4789 . A_loc - the local sequential matrix generated 4790 4791 Level: developer 4792 4793 .seealso: MatGetOwnerShipRange(), MatMPIAIJGetLocalMatCondensed() 4794 4795 @*/ 4796 PetscErrorCode MatMPIAIJGetLocalMat(Mat A,MatReuse scall,Mat *A_loc) 4797 { 4798 PetscErrorCode ierr; 4799 Mat_MPIAIJ *mpimat=(Mat_MPIAIJ*)A->data; 4800 Mat_SeqAIJ *mat,*a=(Mat_SeqAIJ*)(mpimat->A)->data,*b=(Mat_SeqAIJ*)(mpimat->B)->data; 4801 PetscInt *ai=a->i,*aj=a->j,*bi=b->i,*bj=b->j,*cmap=mpimat->garray; 4802 MatScalar *aa=a->a,*ba=b->a,*cam; 4803 PetscScalar *ca; 4804 PetscInt am=A->rmap->n,i,j,k,cstart=A->cmap->rstart; 4805 PetscInt *ci,*cj,col,ncols_d,ncols_o,jo; 4806 PetscBool match; 4807 4808 PetscFunctionBegin; 4809 ierr = PetscTypeCompare((PetscObject)A,MATMPIAIJ,&match);CHKERRQ(ierr); 4810 if (!match) SETERRQ(((PetscObject)A)->comm, PETSC_ERR_SUP,"Requires MPIAIJ matrix as input"); 4811 ierr = PetscLogEventBegin(MAT_Getlocalmat,A,0,0,0);CHKERRQ(ierr); 4812 if (scall == MAT_INITIAL_MATRIX){ 4813 ierr = PetscMalloc((1+am)*sizeof(PetscInt),&ci);CHKERRQ(ierr); 4814 ci[0] = 0; 4815 for (i=0; i<am; i++){ 4816 ci[i+1] = ci[i] + (ai[i+1] - ai[i]) + (bi[i+1] - bi[i]); 4817 } 4818 ierr = PetscMalloc((1+ci[am])*sizeof(PetscInt),&cj);CHKERRQ(ierr); 4819 ierr = PetscMalloc((1+ci[am])*sizeof(PetscScalar),&ca);CHKERRQ(ierr); 4820 k = 0; 4821 for (i=0; i<am; i++) { 4822 ncols_o = bi[i+1] - bi[i]; 4823 ncols_d = ai[i+1] - ai[i]; 4824 /* off-diagonal portion of A */ 4825 for (jo=0; jo<ncols_o; jo++) { 4826 col = cmap[*bj]; 4827 if (col >= cstart) break; 4828 cj[k] = col; bj++; 4829 ca[k++] = *ba++; 4830 } 4831 /* diagonal portion of A */ 4832 for (j=0; j<ncols_d; j++) { 4833 cj[k] = cstart + *aj++; 4834 ca[k++] = *aa++; 4835 } 4836 /* off-diagonal portion of A */ 4837 for (j=jo; j<ncols_o; j++) { 4838 cj[k] = cmap[*bj++]; 4839 ca[k++] = *ba++; 4840 } 4841 } 4842 /* put together the new matrix */ 4843 ierr = MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,am,A->cmap->N,ci,cj,ca,A_loc);CHKERRQ(ierr); 4844 /* MatCreateSeqAIJWithArrays flags matrix so PETSc doesn't free the user's arrays. */ 4845 /* Since these are PETSc arrays, change flags to free them as necessary. */ 4846 mat = (Mat_SeqAIJ*)(*A_loc)->data; 4847 mat->free_a = PETSC_TRUE; 4848 mat->free_ij = PETSC_TRUE; 4849 mat->nonew = 0; 4850 } else if (scall == MAT_REUSE_MATRIX){ 4851 mat=(Mat_SeqAIJ*)(*A_loc)->data; 4852 ci = mat->i; cj = mat->j; cam = mat->a; 4853 for (i=0; i<am; i++) { 4854 /* off-diagonal portion of A */ 4855 ncols_o = bi[i+1] - bi[i]; 4856 for (jo=0; jo<ncols_o; jo++) { 4857 col = cmap[*bj]; 4858 if (col >= cstart) break; 4859 *cam++ = *ba++; bj++; 4860 } 4861 /* diagonal portion of A */ 4862 ncols_d = ai[i+1] - ai[i]; 4863 for (j=0; j<ncols_d; j++) *cam++ = *aa++; 4864 /* off-diagonal portion of A */ 4865 for (j=jo; j<ncols_o; j++) { 4866 *cam++ = *ba++; bj++; 4867 } 4868 } 4869 } else SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Invalid MatReuse %d",(int)scall); 4870 ierr = PetscLogEventEnd(MAT_Getlocalmat,A,0,0,0);CHKERRQ(ierr); 4871 PetscFunctionReturn(0); 4872 } 4873 4874 #undef __FUNCT__ 4875 #define __FUNCT__ "MatMPIAIJGetLocalMatCondensed" 4876 /*@C 4877 MatMPIAIJGetLocalMatCondensed - Creates a SeqAIJ matrix from an MPIAIJ matrix by taking all its local rows and NON-ZERO columns 4878 4879 Not Collective 4880 4881 Input Parameters: 4882 + A - the matrix 4883 . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 4884 - row, col - index sets of rows and columns to extract (or PETSC_NULL) 4885 4886 Output Parameter: 4887 . A_loc - the local sequential matrix generated 4888 4889 Level: developer 4890 4891 .seealso: MatGetOwnershipRange(), MatMPIAIJGetLocalMat() 4892 4893 @*/ 4894 PetscErrorCode MatMPIAIJGetLocalMatCondensed(Mat A,MatReuse scall,IS *row,IS *col,Mat *A_loc) 4895 { 4896 Mat_MPIAIJ *a=(Mat_MPIAIJ*)A->data; 4897 PetscErrorCode ierr; 4898 PetscInt i,start,end,ncols,nzA,nzB,*cmap,imark,*idx; 4899 IS isrowa,iscola; 4900 Mat *aloc; 4901 PetscBool match; 4902 4903 PetscFunctionBegin; 4904 ierr = PetscTypeCompare((PetscObject)A,MATMPIAIJ,&match);CHKERRQ(ierr); 4905 if (!match) SETERRQ(((PetscObject)A)->comm, PETSC_ERR_SUP,"Requires MPIAIJ matrix as input"); 4906 ierr = PetscLogEventBegin(MAT_Getlocalmatcondensed,A,0,0,0);CHKERRQ(ierr); 4907 if (!row){ 4908 start = A->rmap->rstart; end = A->rmap->rend; 4909 ierr = ISCreateStride(PETSC_COMM_SELF,end-start,start,1,&isrowa);CHKERRQ(ierr); 4910 } else { 4911 isrowa = *row; 4912 } 4913 if (!col){ 4914 start = A->cmap->rstart; 4915 cmap = a->garray; 4916 nzA = a->A->cmap->n; 4917 nzB = a->B->cmap->n; 4918 ierr = PetscMalloc((nzA+nzB)*sizeof(PetscInt), &idx);CHKERRQ(ierr); 4919 ncols = 0; 4920 for (i=0; i<nzB; i++) { 4921 if (cmap[i] < start) idx[ncols++] = cmap[i]; 4922 else break; 4923 } 4924 imark = i; 4925 for (i=0; i<nzA; i++) idx[ncols++] = start + i; 4926 for (i=imark; i<nzB; i++) idx[ncols++] = cmap[i]; 4927 ierr = ISCreateGeneral(PETSC_COMM_SELF,ncols,idx,PETSC_OWN_POINTER,&iscola);CHKERRQ(ierr); 4928 } else { 4929 iscola = *col; 4930 } 4931 if (scall != MAT_INITIAL_MATRIX){ 4932 ierr = PetscMalloc(sizeof(Mat),&aloc);CHKERRQ(ierr); 4933 aloc[0] = *A_loc; 4934 } 4935 ierr = MatGetSubMatrices(A,1,&isrowa,&iscola,scall,&aloc);CHKERRQ(ierr); 4936 *A_loc = aloc[0]; 4937 ierr = PetscFree(aloc);CHKERRQ(ierr); 4938 if (!row){ 4939 ierr = ISDestroy(isrowa);CHKERRQ(ierr); 4940 } 4941 if (!col){ 4942 ierr = ISDestroy(iscola);CHKERRQ(ierr); 4943 } 4944 ierr = PetscLogEventEnd(MAT_Getlocalmatcondensed,A,0,0,0);CHKERRQ(ierr); 4945 PetscFunctionReturn(0); 4946 } 4947 4948 #undef __FUNCT__ 4949 #define __FUNCT__ "MatGetBrowsOfAcols" 4950 /*@C 4951 MatGetBrowsOfAcols - Creates a SeqAIJ matrix by taking rows of B that equal to nonzero columns of local A 4952 4953 Collective on Mat 4954 4955 Input Parameters: 4956 + A,B - the matrices in mpiaij format 4957 . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 4958 - rowb, colb - index sets of rows and columns of B to extract (or PETSC_NULL) 4959 4960 Output Parameter: 4961 + rowb, colb - index sets of rows and columns of B to extract 4962 . brstart - row index of B_seq from which next B->rmap->n rows are taken from B's local rows 4963 - B_seq - the sequential matrix generated 4964 4965 Level: developer 4966 4967 @*/ 4968 PetscErrorCode MatGetBrowsOfAcols(Mat A,Mat B,MatReuse scall,IS *rowb,IS *colb,PetscInt *brstart,Mat *B_seq) 4969 { 4970 Mat_MPIAIJ *a=(Mat_MPIAIJ*)A->data; 4971 PetscErrorCode ierr; 4972 PetscInt *idx,i,start,ncols,nzA,nzB,*cmap,imark; 4973 IS isrowb,iscolb; 4974 Mat *bseq; 4975 4976 PetscFunctionBegin; 4977 if (A->cmap->rstart != B->rmap->rstart || A->cmap->rend != B->rmap->rend){ 4978 SETERRQ4(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Matrix local dimensions are incompatible, (%D, %D) != (%D,%D)",A->cmap->rstart,A->cmap->rend,B->rmap->rstart,B->rmap->rend); 4979 } 4980 ierr = PetscLogEventBegin(MAT_GetBrowsOfAcols,A,B,0,0);CHKERRQ(ierr); 4981 4982 if (scall == MAT_INITIAL_MATRIX){ 4983 start = A->cmap->rstart; 4984 cmap = a->garray; 4985 nzA = a->A->cmap->n; 4986 nzB = a->B->cmap->n; 4987 ierr = PetscMalloc((nzA+nzB)*sizeof(PetscInt), &idx);CHKERRQ(ierr); 4988 ncols = 0; 4989 for (i=0; i<nzB; i++) { /* row < local row index */ 4990 if (cmap[i] < start) idx[ncols++] = cmap[i]; 4991 else break; 4992 } 4993 imark = i; 4994 for (i=0; i<nzA; i++) idx[ncols++] = start + i; /* local rows */ 4995 for (i=imark; i<nzB; i++) idx[ncols++] = cmap[i]; /* row > local row index */ 4996 ierr = ISCreateGeneral(PETSC_COMM_SELF,ncols,idx,PETSC_OWN_POINTER,&isrowb);CHKERRQ(ierr); 4997 *brstart = imark; 4998 ierr = ISCreateStride(PETSC_COMM_SELF,B->cmap->N,0,1,&iscolb);CHKERRQ(ierr); 4999 } else { 5000 if (!rowb || !colb) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"IS rowb and colb must be provided for MAT_REUSE_MATRIX"); 5001 isrowb = *rowb; iscolb = *colb; 5002 ierr = PetscMalloc(sizeof(Mat),&bseq);CHKERRQ(ierr); 5003 bseq[0] = *B_seq; 5004 } 5005 ierr = MatGetSubMatrices(B,1,&isrowb,&iscolb,scall,&bseq);CHKERRQ(ierr); 5006 *B_seq = bseq[0]; 5007 ierr = PetscFree(bseq);CHKERRQ(ierr); 5008 if (!rowb){ 5009 ierr = ISDestroy(isrowb);CHKERRQ(ierr); 5010 } else { 5011 *rowb = isrowb; 5012 } 5013 if (!colb){ 5014 ierr = ISDestroy(iscolb);CHKERRQ(ierr); 5015 } else { 5016 *colb = iscolb; 5017 } 5018 ierr = PetscLogEventEnd(MAT_GetBrowsOfAcols,A,B,0,0);CHKERRQ(ierr); 5019 PetscFunctionReturn(0); 5020 } 5021 5022 #undef __FUNCT__ 5023 #define __FUNCT__ "MatGetBrowsOfAoCols" 5024 /*@C 5025 MatGetBrowsOfAoCols - Creates a SeqAIJ matrix by taking rows of B that equal to nonzero columns 5026 of the OFF-DIAGONAL portion of local A 5027 5028 Collective on Mat 5029 5030 Input Parameters: 5031 + A,B - the matrices in mpiaij format 5032 . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX 5033 . startsj - starting point in B's sending and receiving j-arrays, saved for MAT_REUSE (or PETSC_NULL) 5034 . startsj_r - similar to startsj for receives 5035 - bufa_ptr - array for sending matrix values, saved for MAT_REUSE (or PETSC_NULL) 5036 5037 Output Parameter: 5038 + B_oth - the sequential matrix generated 5039 5040 Level: developer 5041 5042 @*/ 5043 PetscErrorCode MatGetBrowsOfAoCols(Mat A,Mat B,MatReuse scall,PetscInt **startsj,PetscInt **startsj_r,MatScalar **bufa_ptr,Mat *B_oth) 5044 { 5045 VecScatter_MPI_General *gen_to,*gen_from; 5046 PetscErrorCode ierr; 5047 Mat_MPIAIJ *a=(Mat_MPIAIJ*)A->data; 5048 Mat_SeqAIJ *b_oth; 5049 VecScatter ctx=a->Mvctx; 5050 MPI_Comm comm=((PetscObject)ctx)->comm; 5051 PetscMPIInt *rprocs,*sprocs,tag=((PetscObject)ctx)->tag,rank; 5052 PetscInt *rowlen,*bufj,*bufJ,ncols,aBn=a->B->cmap->n,row,*b_othi,*b_othj; 5053 PetscScalar *rvalues,*svalues; 5054 MatScalar *b_otha,*bufa,*bufA; 5055 PetscInt i,j,k,l,ll,nrecvs,nsends,nrows,*srow,*rstarts,*rstartsj = 0,*sstarts,*sstartsj,len; 5056 MPI_Request *rwaits = PETSC_NULL,*swaits = PETSC_NULL; 5057 MPI_Status *sstatus,rstatus; 5058 PetscMPIInt jj; 5059 PetscInt *cols,sbs,rbs; 5060 PetscScalar *vals; 5061 5062 PetscFunctionBegin; 5063 if (A->cmap->rstart != B->rmap->rstart || A->cmap->rend != B->rmap->rend){ 5064 SETERRQ4(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Matrix local dimensions are incompatible, (%d, %d) != (%d,%d)",A->cmap->rstart,A->cmap->rend,B->rmap->rstart,B->rmap->rend); 5065 } 5066 ierr = PetscLogEventBegin(MAT_GetBrowsOfAocols,A,B,0,0);CHKERRQ(ierr); 5067 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 5068 5069 gen_to = (VecScatter_MPI_General*)ctx->todata; 5070 gen_from = (VecScatter_MPI_General*)ctx->fromdata; 5071 rvalues = gen_from->values; /* holds the length of receiving row */ 5072 svalues = gen_to->values; /* holds the length of sending row */ 5073 nrecvs = gen_from->n; 5074 nsends = gen_to->n; 5075 5076 ierr = PetscMalloc2(nrecvs,MPI_Request,&rwaits,nsends,MPI_Request,&swaits);CHKERRQ(ierr); 5077 srow = gen_to->indices; /* local row index to be sent */ 5078 sstarts = gen_to->starts; 5079 sprocs = gen_to->procs; 5080 sstatus = gen_to->sstatus; 5081 sbs = gen_to->bs; 5082 rstarts = gen_from->starts; 5083 rprocs = gen_from->procs; 5084 rbs = gen_from->bs; 5085 5086 if (!startsj || !bufa_ptr) scall = MAT_INITIAL_MATRIX; 5087 if (scall == MAT_INITIAL_MATRIX){ 5088 /* i-array */ 5089 /*---------*/ 5090 /* post receives */ 5091 for (i=0; i<nrecvs; i++){ 5092 rowlen = (PetscInt*)rvalues + rstarts[i]*rbs; 5093 nrows = (rstarts[i+1]-rstarts[i])*rbs; /* num of indices to be received */ 5094 ierr = MPI_Irecv(rowlen,nrows,MPIU_INT,rprocs[i],tag,comm,rwaits+i);CHKERRQ(ierr); 5095 } 5096 5097 /* pack the outgoing message */ 5098 ierr = PetscMalloc2(nsends+1,PetscInt,&sstartsj,nrecvs+1,PetscInt,&rstartsj);CHKERRQ(ierr); 5099 sstartsj[0] = 0; rstartsj[0] = 0; 5100 len = 0; /* total length of j or a array to be sent */ 5101 k = 0; 5102 for (i=0; i<nsends; i++){ 5103 rowlen = (PetscInt*)svalues + sstarts[i]*sbs; 5104 nrows = sstarts[i+1]-sstarts[i]; /* num of block rows */ 5105 for (j=0; j<nrows; j++) { 5106 row = srow[k] + B->rmap->range[rank]; /* global row idx */ 5107 for (l=0; l<sbs; l++){ 5108 ierr = MatGetRow_MPIAIJ(B,row+l,&ncols,PETSC_NULL,PETSC_NULL);CHKERRQ(ierr); /* rowlength */ 5109 rowlen[j*sbs+l] = ncols; 5110 len += ncols; 5111 ierr = MatRestoreRow_MPIAIJ(B,row+l,&ncols,PETSC_NULL,PETSC_NULL);CHKERRQ(ierr); 5112 } 5113 k++; 5114 } 5115 ierr = MPI_Isend(rowlen,nrows*sbs,MPIU_INT,sprocs[i],tag,comm,swaits+i);CHKERRQ(ierr); 5116 sstartsj[i+1] = len; /* starting point of (i+1)-th outgoing msg in bufj and bufa */ 5117 } 5118 /* recvs and sends of i-array are completed */ 5119 i = nrecvs; 5120 while (i--) { 5121 ierr = MPI_Waitany(nrecvs,rwaits,&jj,&rstatus);CHKERRQ(ierr); 5122 } 5123 if (nsends) {ierr = MPI_Waitall(nsends,swaits,sstatus);CHKERRQ(ierr);} 5124 5125 /* allocate buffers for sending j and a arrays */ 5126 ierr = PetscMalloc((len+1)*sizeof(PetscInt),&bufj);CHKERRQ(ierr); 5127 ierr = PetscMalloc((len+1)*sizeof(PetscScalar),&bufa);CHKERRQ(ierr); 5128 5129 /* create i-array of B_oth */ 5130 ierr = PetscMalloc((aBn+2)*sizeof(PetscInt),&b_othi);CHKERRQ(ierr); 5131 b_othi[0] = 0; 5132 len = 0; /* total length of j or a array to be received */ 5133 k = 0; 5134 for (i=0; i<nrecvs; i++){ 5135 rowlen = (PetscInt*)rvalues + rstarts[i]*rbs; 5136 nrows = rbs*(rstarts[i+1]-rstarts[i]); /* num of rows to be recieved */ 5137 for (j=0; j<nrows; j++) { 5138 b_othi[k+1] = b_othi[k] + rowlen[j]; 5139 len += rowlen[j]; k++; 5140 } 5141 rstartsj[i+1] = len; /* starting point of (i+1)-th incoming msg in bufj and bufa */ 5142 } 5143 5144 /* allocate space for j and a arrrays of B_oth */ 5145 ierr = PetscMalloc((b_othi[aBn]+1)*sizeof(PetscInt),&b_othj);CHKERRQ(ierr); 5146 ierr = PetscMalloc((b_othi[aBn]+1)*sizeof(MatScalar),&b_otha);CHKERRQ(ierr); 5147 5148 /* j-array */ 5149 /*---------*/ 5150 /* post receives of j-array */ 5151 for (i=0; i<nrecvs; i++){ 5152 nrows = rstartsj[i+1]-rstartsj[i]; /* length of the msg received */ 5153 ierr = MPI_Irecv(b_othj+rstartsj[i],nrows,MPIU_INT,rprocs[i],tag,comm,rwaits+i);CHKERRQ(ierr); 5154 } 5155 5156 /* pack the outgoing message j-array */ 5157 k = 0; 5158 for (i=0; i<nsends; i++){ 5159 nrows = sstarts[i+1]-sstarts[i]; /* num of block rows */ 5160 bufJ = bufj+sstartsj[i]; 5161 for (j=0; j<nrows; j++) { 5162 row = srow[k++] + B->rmap->range[rank]; /* global row idx */ 5163 for (ll=0; ll<sbs; ll++){ 5164 ierr = MatGetRow_MPIAIJ(B,row+ll,&ncols,&cols,PETSC_NULL);CHKERRQ(ierr); 5165 for (l=0; l<ncols; l++){ 5166 *bufJ++ = cols[l]; 5167 } 5168 ierr = MatRestoreRow_MPIAIJ(B,row+ll,&ncols,&cols,PETSC_NULL);CHKERRQ(ierr); 5169 } 5170 } 5171 ierr = MPI_Isend(bufj+sstartsj[i],sstartsj[i+1]-sstartsj[i],MPIU_INT,sprocs[i],tag,comm,swaits+i);CHKERRQ(ierr); 5172 } 5173 5174 /* recvs and sends of j-array are completed */ 5175 i = nrecvs; 5176 while (i--) { 5177 ierr = MPI_Waitany(nrecvs,rwaits,&jj,&rstatus);CHKERRQ(ierr); 5178 } 5179 if (nsends) {ierr = MPI_Waitall(nsends,swaits,sstatus);CHKERRQ(ierr);} 5180 } else if (scall == MAT_REUSE_MATRIX){ 5181 sstartsj = *startsj; 5182 rstartsj = *startsj_r; 5183 bufa = *bufa_ptr; 5184 b_oth = (Mat_SeqAIJ*)(*B_oth)->data; 5185 b_otha = b_oth->a; 5186 } else { 5187 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE, "Matrix P does not posses an object container"); 5188 } 5189 5190 /* a-array */ 5191 /*---------*/ 5192 /* post receives of a-array */ 5193 for (i=0; i<nrecvs; i++){ 5194 nrows = rstartsj[i+1]-rstartsj[i]; /* length of the msg received */ 5195 ierr = MPI_Irecv(b_otha+rstartsj[i],nrows,MPIU_SCALAR,rprocs[i],tag,comm,rwaits+i);CHKERRQ(ierr); 5196 } 5197 5198 /* pack the outgoing message a-array */ 5199 k = 0; 5200 for (i=0; i<nsends; i++){ 5201 nrows = sstarts[i+1]-sstarts[i]; /* num of block rows */ 5202 bufA = bufa+sstartsj[i]; 5203 for (j=0; j<nrows; j++) { 5204 row = srow[k++] + B->rmap->range[rank]; /* global row idx */ 5205 for (ll=0; ll<sbs; ll++){ 5206 ierr = MatGetRow_MPIAIJ(B,row+ll,&ncols,PETSC_NULL,&vals);CHKERRQ(ierr); 5207 for (l=0; l<ncols; l++){ 5208 *bufA++ = vals[l]; 5209 } 5210 ierr = MatRestoreRow_MPIAIJ(B,row+ll,&ncols,PETSC_NULL,&vals);CHKERRQ(ierr); 5211 } 5212 } 5213 ierr = MPI_Isend(bufa+sstartsj[i],sstartsj[i+1]-sstartsj[i],MPIU_SCALAR,sprocs[i],tag,comm,swaits+i);CHKERRQ(ierr); 5214 } 5215 /* recvs and sends of a-array are completed */ 5216 i = nrecvs; 5217 while (i--) { 5218 ierr = MPI_Waitany(nrecvs,rwaits,&jj,&rstatus);CHKERRQ(ierr); 5219 } 5220 if (nsends) {ierr = MPI_Waitall(nsends,swaits,sstatus);CHKERRQ(ierr);} 5221 ierr = PetscFree2(rwaits,swaits);CHKERRQ(ierr); 5222 5223 if (scall == MAT_INITIAL_MATRIX){ 5224 /* put together the new matrix */ 5225 ierr = MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,aBn,B->cmap->N,b_othi,b_othj,b_otha,B_oth);CHKERRQ(ierr); 5226 5227 /* MatCreateSeqAIJWithArrays flags matrix so PETSc doesn't free the user's arrays. */ 5228 /* Since these are PETSc arrays, change flags to free them as necessary. */ 5229 b_oth = (Mat_SeqAIJ *)(*B_oth)->data; 5230 b_oth->free_a = PETSC_TRUE; 5231 b_oth->free_ij = PETSC_TRUE; 5232 b_oth->nonew = 0; 5233 5234 ierr = PetscFree(bufj);CHKERRQ(ierr); 5235 if (!startsj || !bufa_ptr){ 5236 ierr = PetscFree2(sstartsj,rstartsj);CHKERRQ(ierr); 5237 ierr = PetscFree(bufa_ptr);CHKERRQ(ierr); 5238 } else { 5239 *startsj = sstartsj; 5240 *startsj_r = rstartsj; 5241 *bufa_ptr = bufa; 5242 } 5243 } 5244 ierr = PetscLogEventEnd(MAT_GetBrowsOfAocols,A,B,0,0);CHKERRQ(ierr); 5245 PetscFunctionReturn(0); 5246 } 5247 5248 #undef __FUNCT__ 5249 #define __FUNCT__ "MatGetCommunicationStructs" 5250 /*@C 5251 MatGetCommunicationStructs - Provides access to the communication structures used in matrix-vector multiplication. 5252 5253 Not Collective 5254 5255 Input Parameters: 5256 . A - The matrix in mpiaij format 5257 5258 Output Parameter: 5259 + lvec - The local vector holding off-process values from the argument to a matrix-vector product 5260 . colmap - A map from global column index to local index into lvec 5261 - multScatter - A scatter from the argument of a matrix-vector product to lvec 5262 5263 Level: developer 5264 5265 @*/ 5266 #if defined (PETSC_USE_CTABLE) 5267 PetscErrorCode MatGetCommunicationStructs(Mat A, Vec *lvec, PetscTable *colmap, VecScatter *multScatter) 5268 #else 5269 PetscErrorCode MatGetCommunicationStructs(Mat A, Vec *lvec, PetscInt *colmap[], VecScatter *multScatter) 5270 #endif 5271 { 5272 Mat_MPIAIJ *a; 5273 5274 PetscFunctionBegin; 5275 PetscValidHeaderSpecific(A, MAT_CLASSID, 1); 5276 PetscValidPointer(lvec, 2); 5277 PetscValidPointer(colmap, 3); 5278 PetscValidPointer(multScatter, 4); 5279 a = (Mat_MPIAIJ *) A->data; 5280 if (lvec) *lvec = a->lvec; 5281 if (colmap) *colmap = a->colmap; 5282 if (multScatter) *multScatter = a->Mvctx; 5283 PetscFunctionReturn(0); 5284 } 5285 5286 EXTERN_C_BEGIN 5287 extern PetscErrorCode MatConvert_MPIAIJ_MPIAIJCRL(Mat,const MatType,MatReuse,Mat*); 5288 extern PetscErrorCode MatConvert_MPIAIJ_MPIAIJPERM(Mat,const MatType,MatReuse,Mat*); 5289 extern PetscErrorCode MatConvert_MPIAIJ_MPISBAIJ(Mat,const MatType,MatReuse,Mat*); 5290 EXTERN_C_END 5291 5292 #undef __FUNCT__ 5293 #define __FUNCT__ "MatMatMultNumeric_MPIDense_MPIAIJ" 5294 /* 5295 Computes (B'*A')' since computing B*A directly is untenable 5296 5297 n p p 5298 ( ) ( ) ( ) 5299 m ( A ) * n ( B ) = m ( C ) 5300 ( ) ( ) ( ) 5301 5302 */ 5303 PetscErrorCode MatMatMultNumeric_MPIDense_MPIAIJ(Mat A,Mat B,Mat C) 5304 { 5305 PetscErrorCode ierr; 5306 Mat At,Bt,Ct; 5307 5308 PetscFunctionBegin; 5309 ierr = MatTranspose(A,MAT_INITIAL_MATRIX,&At);CHKERRQ(ierr); 5310 ierr = MatTranspose(B,MAT_INITIAL_MATRIX,&Bt);CHKERRQ(ierr); 5311 ierr = MatMatMult(Bt,At,MAT_INITIAL_MATRIX,1.0,&Ct);CHKERRQ(ierr); 5312 ierr = MatDestroy(At);CHKERRQ(ierr); 5313 ierr = MatDestroy(Bt);CHKERRQ(ierr); 5314 ierr = MatTranspose(Ct,MAT_REUSE_MATRIX,&C);CHKERRQ(ierr); 5315 ierr = MatDestroy(Ct);CHKERRQ(ierr); 5316 PetscFunctionReturn(0); 5317 } 5318 5319 #undef __FUNCT__ 5320 #define __FUNCT__ "MatMatMultSymbolic_MPIDense_MPIAIJ" 5321 PetscErrorCode MatMatMultSymbolic_MPIDense_MPIAIJ(Mat A,Mat B,PetscReal fill,Mat *C) 5322 { 5323 PetscErrorCode ierr; 5324 PetscInt m=A->rmap->n,n=B->cmap->n; 5325 Mat Cmat; 5326 5327 PetscFunctionBegin; 5328 if (A->cmap->n != B->rmap->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"A->cmap->n %d != B->rmap->n %d\n",A->cmap->n,B->rmap->n); 5329 ierr = MatCreate(((PetscObject)A)->comm,&Cmat);CHKERRQ(ierr); 5330 ierr = MatSetSizes(Cmat,m,n,PETSC_DETERMINE,PETSC_DETERMINE);CHKERRQ(ierr); 5331 ierr = MatSetType(Cmat,MATMPIDENSE);CHKERRQ(ierr); 5332 ierr = MatMPIDenseSetPreallocation(Cmat,PETSC_NULL);CHKERRQ(ierr); 5333 ierr = MatAssemblyBegin(Cmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5334 ierr = MatAssemblyEnd(Cmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5335 *C = Cmat; 5336 PetscFunctionReturn(0); 5337 } 5338 5339 /* ----------------------------------------------------------------*/ 5340 #undef __FUNCT__ 5341 #define __FUNCT__ "MatMatMult_MPIDense_MPIAIJ" 5342 PetscErrorCode MatMatMult_MPIDense_MPIAIJ(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C) 5343 { 5344 PetscErrorCode ierr; 5345 5346 PetscFunctionBegin; 5347 if (scall == MAT_INITIAL_MATRIX){ 5348 ierr = MatMatMultSymbolic_MPIDense_MPIAIJ(A,B,fill,C);CHKERRQ(ierr); 5349 } 5350 ierr = MatMatMultNumeric_MPIDense_MPIAIJ(A,B,*C);CHKERRQ(ierr); 5351 PetscFunctionReturn(0); 5352 } 5353 5354 EXTERN_C_BEGIN 5355 #if defined(PETSC_HAVE_MUMPS) 5356 extern PetscErrorCode MatGetFactor_aij_mumps(Mat,MatFactorType,Mat*); 5357 #endif 5358 #if defined(PETSC_HAVE_PASTIX) 5359 extern PetscErrorCode MatGetFactor_mpiaij_pastix(Mat,MatFactorType,Mat*); 5360 #endif 5361 #if defined(PETSC_HAVE_SUPERLU_DIST) 5362 extern PetscErrorCode MatGetFactor_mpiaij_superlu_dist(Mat,MatFactorType,Mat*); 5363 #endif 5364 #if defined(PETSC_HAVE_SPOOLES) 5365 extern PetscErrorCode MatGetFactor_mpiaij_spooles(Mat,MatFactorType,Mat*); 5366 #endif 5367 EXTERN_C_END 5368 5369 /*MC 5370 MATMPIAIJ - MATMPIAIJ = "mpiaij" - A matrix type to be used for parallel sparse matrices. 5371 5372 Options Database Keys: 5373 . -mat_type mpiaij - sets the matrix type to "mpiaij" during a call to MatSetFromOptions() 5374 5375 Level: beginner 5376 5377 .seealso: MatCreateMPIAIJ() 5378 M*/ 5379 5380 EXTERN_C_BEGIN 5381 #undef __FUNCT__ 5382 #define __FUNCT__ "MatCreate_MPIAIJ" 5383 PetscErrorCode MatCreate_MPIAIJ(Mat B) 5384 { 5385 Mat_MPIAIJ *b; 5386 PetscErrorCode ierr; 5387 PetscMPIInt size; 5388 5389 PetscFunctionBegin; 5390 ierr = MPI_Comm_size(((PetscObject)B)->comm,&size);CHKERRQ(ierr); 5391 5392 ierr = PetscNewLog(B,Mat_MPIAIJ,&b);CHKERRQ(ierr); 5393 B->data = (void*)b; 5394 ierr = PetscMemcpy(B->ops,&MatOps_Values,sizeof(struct _MatOps));CHKERRQ(ierr); 5395 B->rmap->bs = 1; 5396 B->assembled = PETSC_FALSE; 5397 5398 B->insertmode = NOT_SET_VALUES; 5399 b->size = size; 5400 ierr = MPI_Comm_rank(((PetscObject)B)->comm,&b->rank);CHKERRQ(ierr); 5401 5402 /* build cache for off array entries formed */ 5403 ierr = MatStashCreate_Private(((PetscObject)B)->comm,1,&B->stash);CHKERRQ(ierr); 5404 b->donotstash = PETSC_FALSE; 5405 b->colmap = 0; 5406 b->garray = 0; 5407 b->roworiented = PETSC_TRUE; 5408 5409 /* stuff used for matrix vector multiply */ 5410 b->lvec = PETSC_NULL; 5411 b->Mvctx = PETSC_NULL; 5412 5413 /* stuff for MatGetRow() */ 5414 b->rowindices = 0; 5415 b->rowvalues = 0; 5416 b->getrowactive = PETSC_FALSE; 5417 5418 #if defined(PETSC_HAVE_SPOOLES) 5419 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetFactor_spooles_C", 5420 "MatGetFactor_mpiaij_spooles", 5421 MatGetFactor_mpiaij_spooles);CHKERRQ(ierr); 5422 #endif 5423 #if defined(PETSC_HAVE_MUMPS) 5424 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetFactor_mumps_C", 5425 "MatGetFactor_aij_mumps", 5426 MatGetFactor_aij_mumps);CHKERRQ(ierr); 5427 #endif 5428 #if defined(PETSC_HAVE_PASTIX) 5429 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetFactor_pastix_C", 5430 "MatGetFactor_mpiaij_pastix", 5431 MatGetFactor_mpiaij_pastix);CHKERRQ(ierr); 5432 #endif 5433 #if defined(PETSC_HAVE_SUPERLU_DIST) 5434 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetFactor_superlu_dist_C", 5435 "MatGetFactor_mpiaij_superlu_dist", 5436 MatGetFactor_mpiaij_superlu_dist);CHKERRQ(ierr); 5437 #endif 5438 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatStoreValues_C", 5439 "MatStoreValues_MPIAIJ", 5440 MatStoreValues_MPIAIJ);CHKERRQ(ierr); 5441 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatRetrieveValues_C", 5442 "MatRetrieveValues_MPIAIJ", 5443 MatRetrieveValues_MPIAIJ);CHKERRQ(ierr); 5444 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetDiagonalBlock_C", 5445 "MatGetDiagonalBlock_MPIAIJ", 5446 MatGetDiagonalBlock_MPIAIJ);CHKERRQ(ierr); 5447 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatIsTranspose_C", 5448 "MatIsTranspose_MPIAIJ", 5449 MatIsTranspose_MPIAIJ);CHKERRQ(ierr); 5450 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMPIAIJSetPreallocation_C", 5451 "MatMPIAIJSetPreallocation_MPIAIJ", 5452 MatMPIAIJSetPreallocation_MPIAIJ);CHKERRQ(ierr); 5453 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMPIAIJSetPreallocationCSR_C", 5454 "MatMPIAIJSetPreallocationCSR_MPIAIJ", 5455 MatMPIAIJSetPreallocationCSR_MPIAIJ);CHKERRQ(ierr); 5456 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatDiagonalScaleLocal_C", 5457 "MatDiagonalScaleLocal_MPIAIJ", 5458 MatDiagonalScaleLocal_MPIAIJ);CHKERRQ(ierr); 5459 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_mpiaij_mpiaijperm_C", 5460 "MatConvert_MPIAIJ_MPIAIJPERM", 5461 MatConvert_MPIAIJ_MPIAIJPERM);CHKERRQ(ierr); 5462 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_mpiaij_mpiaijcrl_C", 5463 "MatConvert_MPIAIJ_MPIAIJCRL", 5464 MatConvert_MPIAIJ_MPIAIJCRL);CHKERRQ(ierr); 5465 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_mpiaij_mpisbaij_C", 5466 "MatConvert_MPIAIJ_MPISBAIJ", 5467 MatConvert_MPIAIJ_MPISBAIJ);CHKERRQ(ierr); 5468 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMatMult_mpidense_mpiaij_C", 5469 "MatMatMult_MPIDense_MPIAIJ", 5470 MatMatMult_MPIDense_MPIAIJ);CHKERRQ(ierr); 5471 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMatMultSymbolic_mpidense_mpiaij_C", 5472 "MatMatMultSymbolic_MPIDense_MPIAIJ", 5473 MatMatMultSymbolic_MPIDense_MPIAIJ);CHKERRQ(ierr); 5474 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMatMultNumeric_mpidense_mpiaij_C", 5475 "MatMatMultNumeric_MPIDense_MPIAIJ", 5476 MatMatMultNumeric_MPIDense_MPIAIJ);CHKERRQ(ierr); 5477 ierr = PetscObjectChangeTypeName((PetscObject)B,MATMPIAIJ);CHKERRQ(ierr); 5478 PetscFunctionReturn(0); 5479 } 5480 EXTERN_C_END 5481 5482 #undef __FUNCT__ 5483 #define __FUNCT__ "MatCreateMPIAIJWithSplitArrays" 5484 /*@ 5485 MatCreateMPIAIJWithSplitArrays - creates a MPI AIJ matrix using arrays that contain the "diagonal" 5486 and "off-diagonal" part of the matrix in CSR format. 5487 5488 Collective on MPI_Comm 5489 5490 Input Parameters: 5491 + comm - MPI communicator 5492 . m - number of local rows (Cannot be PETSC_DECIDE) 5493 . n - This value should be the same as the local size used in creating the 5494 x vector for the matrix-vector product y = Ax. (or PETSC_DECIDE to have 5495 calculated if N is given) For square matrices n is almost always m. 5496 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 5497 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 5498 . i - row indices for "diagonal" portion of matrix 5499 . j - column indices 5500 . a - matrix values 5501 . oi - row indices for "off-diagonal" portion of matrix 5502 . oj - column indices 5503 - oa - matrix values 5504 5505 Output Parameter: 5506 . mat - the matrix 5507 5508 Level: advanced 5509 5510 Notes: 5511 The i, j, and a arrays ARE NOT copied by this routine into the internal format used by PETSc. The user 5512 must free the arrays once the matrix has been destroyed and not before. 5513 5514 The i and j indices are 0 based 5515 5516 See MatCreateMPIAIJ() for the definition of "diagonal" and "off-diagonal" portion of the matrix 5517 5518 This sets local rows and cannot be used to set off-processor values. 5519 5520 You cannot later use MatSetValues() to change values in this matrix. 5521 5522 .keywords: matrix, aij, compressed row, sparse, parallel 5523 5524 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIAIJSetPreallocation(), MatMPIAIJSetPreallocationCSR(), 5525 MPIAIJ, MatCreateMPIAIJ(), MatCreateMPIAIJWithArrays() 5526 @*/ 5527 PetscErrorCode MatCreateMPIAIJWithSplitArrays(MPI_Comm comm,PetscInt m,PetscInt n,PetscInt M,PetscInt N,PetscInt i[],PetscInt j[],PetscScalar a[], 5528 PetscInt oi[], PetscInt oj[],PetscScalar oa[],Mat *mat) 5529 { 5530 PetscErrorCode ierr; 5531 Mat_MPIAIJ *maij; 5532 5533 PetscFunctionBegin; 5534 if (m < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"local number of rows (m) cannot be PETSC_DECIDE, or negative"); 5535 if (i[0]) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"i (row indices) must start with 0"); 5536 if (oi[0]) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"oi (row indices) must start with 0"); 5537 ierr = MatCreate(comm,mat);CHKERRQ(ierr); 5538 ierr = MatSetSizes(*mat,m,n,M,N);CHKERRQ(ierr); 5539 ierr = MatSetType(*mat,MATMPIAIJ);CHKERRQ(ierr); 5540 maij = (Mat_MPIAIJ*) (*mat)->data; 5541 maij->donotstash = PETSC_TRUE; 5542 (*mat)->preallocated = PETSC_TRUE; 5543 5544 ierr = PetscLayoutSetBlockSize((*mat)->rmap,1);CHKERRQ(ierr); 5545 ierr = PetscLayoutSetBlockSize((*mat)->cmap,1);CHKERRQ(ierr); 5546 ierr = PetscLayoutSetUp((*mat)->rmap);CHKERRQ(ierr); 5547 ierr = PetscLayoutSetUp((*mat)->cmap);CHKERRQ(ierr); 5548 5549 ierr = MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,m,n,i,j,a,&maij->A);CHKERRQ(ierr); 5550 ierr = MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,m,(*mat)->cmap->N,oi,oj,oa,&maij->B);CHKERRQ(ierr); 5551 5552 ierr = MatAssemblyBegin(maij->A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5553 ierr = MatAssemblyEnd(maij->A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5554 ierr = MatAssemblyBegin(maij->B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5555 ierr = MatAssemblyEnd(maij->B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5556 5557 ierr = MatAssemblyBegin(*mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5558 ierr = MatAssemblyEnd(*mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 5559 PetscFunctionReturn(0); 5560 } 5561 5562 /* 5563 Special version for direct calls from Fortran 5564 */ 5565 #include <private/fortranimpl.h> 5566 5567 #if defined(PETSC_HAVE_FORTRAN_CAPS) 5568 #define matsetvaluesmpiaij_ MATSETVALUESMPIAIJ 5569 #elif !defined(PETSC_HAVE_FORTRAN_UNDERSCORE) 5570 #define matsetvaluesmpiaij_ matsetvaluesmpiaij 5571 #endif 5572 5573 /* Change these macros so can be used in void function */ 5574 #undef CHKERRQ 5575 #define CHKERRQ(ierr) CHKERRABORT(PETSC_COMM_WORLD,ierr) 5576 #undef SETERRQ2 5577 #define SETERRQ2(comm,ierr,b,c,d) CHKERRABORT(comm,ierr) 5578 #undef SETERRQ 5579 #define SETERRQ(c,ierr,b) CHKERRABORT(c,ierr) 5580 5581 EXTERN_C_BEGIN 5582 #undef __FUNCT__ 5583 #define __FUNCT__ "matsetvaluesmpiaij_" 5584 void PETSC_STDCALL matsetvaluesmpiaij_(Mat *mmat,PetscInt *mm,const PetscInt im[],PetscInt *mn,const PetscInt in[],const PetscScalar v[],InsertMode *maddv,PetscErrorCode *_ierr) 5585 { 5586 Mat mat = *mmat; 5587 PetscInt m = *mm, n = *mn; 5588 InsertMode addv = *maddv; 5589 Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data; 5590 PetscScalar value; 5591 PetscErrorCode ierr; 5592 5593 ierr = MatPreallocated(mat);CHKERRQ(ierr); 5594 if (mat->insertmode == NOT_SET_VALUES) { 5595 mat->insertmode = addv; 5596 } 5597 #if defined(PETSC_USE_DEBUG) 5598 else if (mat->insertmode != addv) { 5599 SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values"); 5600 } 5601 #endif 5602 { 5603 PetscInt i,j,rstart = mat->rmap->rstart,rend = mat->rmap->rend; 5604 PetscInt cstart = mat->cmap->rstart,cend = mat->cmap->rend,row,col; 5605 PetscBool roworiented = aij->roworiented; 5606 5607 /* Some Variables required in the macro */ 5608 Mat A = aij->A; 5609 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 5610 PetscInt *aimax = a->imax,*ai = a->i,*ailen = a->ilen,*aj = a->j; 5611 MatScalar *aa = a->a; 5612 PetscBool ignorezeroentries = (((a->ignorezeroentries)&&(addv==ADD_VALUES))?PETSC_TRUE:PETSC_FALSE); 5613 Mat B = aij->B; 5614 Mat_SeqAIJ *b = (Mat_SeqAIJ*)B->data; 5615 PetscInt *bimax = b->imax,*bi = b->i,*bilen = b->ilen,*bj = b->j,bm = aij->B->rmap->n,am = aij->A->rmap->n; 5616 MatScalar *ba = b->a; 5617 5618 PetscInt *rp1,*rp2,ii,nrow1,nrow2,_i,rmax1,rmax2,N,low1,high1,low2,high2,t,lastcol1,lastcol2; 5619 PetscInt nonew = a->nonew; 5620 MatScalar *ap1,*ap2; 5621 5622 PetscFunctionBegin; 5623 for (i=0; i<m; i++) { 5624 if (im[i] < 0) continue; 5625 #if defined(PETSC_USE_DEBUG) 5626 if (im[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",im[i],mat->rmap->N-1); 5627 #endif 5628 if (im[i] >= rstart && im[i] < rend) { 5629 row = im[i] - rstart; 5630 lastcol1 = -1; 5631 rp1 = aj + ai[row]; 5632 ap1 = aa + ai[row]; 5633 rmax1 = aimax[row]; 5634 nrow1 = ailen[row]; 5635 low1 = 0; 5636 high1 = nrow1; 5637 lastcol2 = -1; 5638 rp2 = bj + bi[row]; 5639 ap2 = ba + bi[row]; 5640 rmax2 = bimax[row]; 5641 nrow2 = bilen[row]; 5642 low2 = 0; 5643 high2 = nrow2; 5644 5645 for (j=0; j<n; j++) { 5646 if (roworiented) value = v[i*n+j]; else value = v[i+j*m]; 5647 if (ignorezeroentries && value == 0.0 && (addv == ADD_VALUES)) continue; 5648 if (in[j] >= cstart && in[j] < cend){ 5649 col = in[j] - cstart; 5650 MatSetValues_SeqAIJ_A_Private(row,col,value,addv); 5651 } else if (in[j] < 0) continue; 5652 #if defined(PETSC_USE_DEBUG) 5653 else if (in[j] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",in[j],mat->cmap->N-1); 5654 #endif 5655 else { 5656 if (mat->was_assembled) { 5657 if (!aij->colmap) { 5658 ierr = CreateColmap_MPIAIJ_Private(mat);CHKERRQ(ierr); 5659 } 5660 #if defined (PETSC_USE_CTABLE) 5661 ierr = PetscTableFind(aij->colmap,in[j]+1,&col);CHKERRQ(ierr); 5662 col--; 5663 #else 5664 col = aij->colmap[in[j]] - 1; 5665 #endif 5666 if (col < 0 && !((Mat_SeqAIJ*)(aij->A->data))->nonew) { 5667 ierr = DisAssemble_MPIAIJ(mat);CHKERRQ(ierr); 5668 col = in[j]; 5669 /* Reinitialize the variables required by MatSetValues_SeqAIJ_B_Private() */ 5670 B = aij->B; 5671 b = (Mat_SeqAIJ*)B->data; 5672 bimax = b->imax; bi = b->i; bilen = b->ilen; bj = b->j; 5673 rp2 = bj + bi[row]; 5674 ap2 = ba + bi[row]; 5675 rmax2 = bimax[row]; 5676 nrow2 = bilen[row]; 5677 low2 = 0; 5678 high2 = nrow2; 5679 bm = aij->B->rmap->n; 5680 ba = b->a; 5681 } 5682 } else col = in[j]; 5683 MatSetValues_SeqAIJ_B_Private(row,col,value,addv); 5684 } 5685 } 5686 } else { 5687 if (!aij->donotstash) { 5688 if (roworiented) { 5689 ierr = MatStashValuesRow_Private(&mat->stash,im[i],n,in,v+i*n,(PetscBool)(ignorezeroentries && (addv == ADD_VALUES)));CHKERRQ(ierr); 5690 } else { 5691 ierr = MatStashValuesCol_Private(&mat->stash,im[i],n,in,v+i,m,(PetscBool)(ignorezeroentries && (addv == ADD_VALUES)));CHKERRQ(ierr); 5692 } 5693 } 5694 } 5695 }} 5696 PetscFunctionReturnVoid(); 5697 } 5698 EXTERN_C_END 5699 5700