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