1 2 #include "src/mat/impls/baij/mpi/mpibaij.h" /*I "petscmat.h" I*/ 3 #include "mpisbaij.h" 4 #include "src/mat/impls/sbaij/seq/sbaij.h" 5 6 EXTERN PetscErrorCode MatSetUpMultiply_MPISBAIJ(Mat); 7 EXTERN PetscErrorCode MatSetUpMultiply_MPISBAIJ_2comm(Mat); 8 EXTERN PetscErrorCode DisAssemble_MPISBAIJ(Mat); 9 EXTERN PetscErrorCode MatIncreaseOverlap_MPISBAIJ(Mat,PetscInt,IS[],PetscInt); 10 EXTERN PetscErrorCode MatGetValues_SeqSBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],PetscScalar []); 11 EXTERN PetscErrorCode MatGetValues_SeqBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],PetscScalar []); 12 EXTERN PetscErrorCode MatSetValues_SeqSBAIJ(Mat,PetscInt,const PetscInt [],PetscInt,const PetscInt [],const PetscScalar [],InsertMode); 13 EXTERN PetscErrorCode MatSetValuesBlocked_SeqSBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode); 14 EXTERN PetscErrorCode MatSetValuesBlocked_SeqBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode); 15 EXTERN PetscErrorCode MatGetRow_SeqSBAIJ(Mat,PetscInt,PetscInt*,PetscInt**,PetscScalar**); 16 EXTERN PetscErrorCode MatRestoreRow_SeqSBAIJ(Mat,PetscInt,PetscInt*,PetscInt**,PetscScalar**); 17 EXTERN PetscErrorCode MatPrintHelp_SeqSBAIJ(Mat); 18 EXTERN PetscErrorCode MatZeroRows_SeqSBAIJ(Mat,IS,PetscScalar*); 19 EXTERN PetscErrorCode MatZeroRows_SeqBAIJ(Mat,IS,PetscScalar *); 20 EXTERN PetscErrorCode MatGetRowMax_MPISBAIJ(Mat,Vec); 21 EXTERN PetscErrorCode MatRelax_MPISBAIJ(Mat,Vec,PetscReal,MatSORType,PetscReal,PetscInt,PetscInt,Vec); 22 23 /* UGLY, ugly, ugly 24 When MatScalar == PetscScalar the function MatSetValuesBlocked_MPIBAIJ_MatScalar() does 25 not exist. Otherwise ..._MatScalar() takes matrix elements in single precision and 26 inserts them into the single precision data structure. The function MatSetValuesBlocked_MPIBAIJ() 27 converts the entries into single precision and then calls ..._MatScalar() to put them 28 into the single precision data structures. 29 */ 30 #if defined(PETSC_USE_MAT_SINGLE) 31 EXTERN PetscErrorCode MatSetValuesBlocked_SeqSBAIJ_MatScalar(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const MatScalar[],InsertMode); 32 EXTERN PetscErrorCode MatSetValues_MPISBAIJ_MatScalar(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const MatScalar[],InsertMode); 33 EXTERN PetscErrorCode MatSetValuesBlocked_MPISBAIJ_MatScalar(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const MatScalar[],InsertMode); 34 EXTERN PetscErrorCode MatSetValues_MPISBAIJ_HT_MatScalar(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const MatScalar[],InsertMode); 35 EXTERN PetscErrorCode MatSetValuesBlocked_MPISBAIJ_HT_MatScalar(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const MatScalar[],InsertMode); 36 #else 37 #define MatSetValuesBlocked_SeqSBAIJ_MatScalar MatSetValuesBlocked_SeqSBAIJ 38 #define MatSetValues_MPISBAIJ_MatScalar MatSetValues_MPISBAIJ 39 #define MatSetValuesBlocked_MPISBAIJ_MatScalar MatSetValuesBlocked_MPISBAIJ 40 #define MatSetValues_MPISBAIJ_HT_MatScalar MatSetValues_MPISBAIJ_HT 41 #define MatSetValuesBlocked_MPISBAIJ_HT_MatScalar MatSetValuesBlocked_MPISBAIJ_HT 42 #endif 43 44 EXTERN_C_BEGIN 45 #undef __FUNCT__ 46 #define __FUNCT__ "MatStoreValues_MPISBAIJ" 47 PetscErrorCode MatStoreValues_MPISBAIJ(Mat mat) 48 { 49 Mat_MPISBAIJ *aij = (Mat_MPISBAIJ *)mat->data; 50 PetscErrorCode ierr; 51 52 PetscFunctionBegin; 53 ierr = MatStoreValues(aij->A);CHKERRQ(ierr); 54 ierr = MatStoreValues(aij->B);CHKERRQ(ierr); 55 PetscFunctionReturn(0); 56 } 57 EXTERN_C_END 58 59 EXTERN_C_BEGIN 60 #undef __FUNCT__ 61 #define __FUNCT__ "MatRetrieveValues_MPISBAIJ" 62 PetscErrorCode MatRetrieveValues_MPISBAIJ(Mat mat) 63 { 64 Mat_MPISBAIJ *aij = (Mat_MPISBAIJ *)mat->data; 65 PetscErrorCode ierr; 66 67 PetscFunctionBegin; 68 ierr = MatRetrieveValues(aij->A);CHKERRQ(ierr); 69 ierr = MatRetrieveValues(aij->B);CHKERRQ(ierr); 70 PetscFunctionReturn(0); 71 } 72 EXTERN_C_END 73 74 75 #define CHUNKSIZE 10 76 77 #define MatSetValues_SeqSBAIJ_A_Private(row,col,value,addv) \ 78 { \ 79 \ 80 brow = row/bs; \ 81 rp = aj + ai[brow]; ap = aa + bs2*ai[brow]; \ 82 rmax = aimax[brow]; nrow = ailen[brow]; \ 83 bcol = col/bs; \ 84 ridx = row % bs; cidx = col % bs; \ 85 low = 0; high = nrow; \ 86 while (high-low > 3) { \ 87 t = (low+high)/2; \ 88 if (rp[t] > bcol) high = t; \ 89 else low = t; \ 90 } \ 91 for (_i=low; _i<high; _i++) { \ 92 if (rp[_i] > bcol) break; \ 93 if (rp[_i] == bcol) { \ 94 bap = ap + bs2*_i + bs*cidx + ridx; \ 95 if (addv == ADD_VALUES) *bap += value; \ 96 else *bap = value; \ 97 goto a_noinsert; \ 98 } \ 99 } \ 100 if (a->nonew == 1) goto a_noinsert; \ 101 else if (a->nonew == -1) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 102 if (nrow >= rmax) { \ 103 /* there is no extra room in row, therefore enlarge */ \ 104 PetscInt new_nz = ai[a->mbs] + CHUNKSIZE,len,*new_i,*new_j; \ 105 MatScalar *new_a; \ 106 \ 107 if (a->nonew == -2) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) in the matrix", row, col); \ 108 \ 109 /* malloc new storage space */ \ 110 len = new_nz*(sizeof(PetscInt)+bs2*sizeof(MatScalar))+(a->mbs+1)*sizeof(PetscInt); \ 111 ierr = PetscMalloc(len,&new_a);CHKERRQ(ierr); \ 112 new_j = (PetscInt*)(new_a + bs2*new_nz); \ 113 new_i = new_j + new_nz; \ 114 \ 115 /* copy over old data into new slots */ \ 116 for (ii=0; ii<brow+1; ii++) {new_i[ii] = ai[ii];} \ 117 for (ii=brow+1; ii<a->mbs+1; ii++) {new_i[ii] = ai[ii]+CHUNKSIZE;} \ 118 ierr = PetscMemcpy(new_j,aj,(ai[brow]+nrow)*sizeof(PetscInt));CHKERRQ(ierr); \ 119 len = (new_nz - CHUNKSIZE - ai[brow] - nrow); \ 120 ierr = PetscMemcpy(new_j+ai[brow]+nrow+CHUNKSIZE,aj+ai[brow]+nrow,len*sizeof(PetscInt));CHKERRQ(ierr); \ 121 ierr = PetscMemcpy(new_a,aa,(ai[brow]+nrow)*bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 122 ierr = PetscMemzero(new_a+bs2*(ai[brow]+nrow),bs2*CHUNKSIZE*sizeof(PetscScalar));CHKERRQ(ierr); \ 123 ierr = PetscMemcpy(new_a+bs2*(ai[brow]+nrow+CHUNKSIZE), \ 124 aa+bs2*(ai[brow]+nrow),bs2*len*sizeof(MatScalar));CHKERRQ(ierr); \ 125 /* free up old matrix storage */ \ 126 ierr = PetscFree(a->a);CHKERRQ(ierr); \ 127 if (!a->singlemalloc) { \ 128 ierr = PetscFree(a->i);CHKERRQ(ierr); \ 129 ierr = PetscFree(a->j);CHKERRQ(ierr);\ 130 } \ 131 aa = a->a = new_a; ai = a->i = new_i; aj = a->j = new_j; \ 132 a->singlemalloc = PETSC_TRUE; \ 133 \ 134 rp = aj + ai[brow]; ap = aa + bs2*ai[brow]; \ 135 rmax = aimax[brow] = aimax[brow] + CHUNKSIZE; \ 136 ierr = PetscLogObjectMemory(A,CHUNKSIZE*(sizeof(PetscInt) + bs2*sizeof(MatScalar)));CHKERRQ(ierr); \ 137 a->maxnz += bs2*CHUNKSIZE; \ 138 a->reallocs++; \ 139 a->nz++; \ 140 } \ 141 N = nrow++ - 1; \ 142 /* shift up all the later entries in this row */ \ 143 for (ii=N; ii>=_i; ii--) { \ 144 rp[ii+1] = rp[ii]; \ 145 ierr = PetscMemcpy(ap+bs2*(ii+1),ap+bs2*(ii),bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 146 } \ 147 if (N>=_i) { ierr = PetscMemzero(ap+bs2*_i,bs2*sizeof(MatScalar));CHKERRQ(ierr); } \ 148 rp[_i] = bcol; \ 149 ap[bs2*_i + bs*cidx + ridx] = value; \ 150 a_noinsert:; \ 151 ailen[brow] = nrow; \ 152 } 153 #ifndef MatSetValues_SeqBAIJ_B_Private 154 #define MatSetValues_SeqSBAIJ_B_Private(row,col,value,addv) \ 155 { \ 156 brow = row/bs; \ 157 rp = bj + bi[brow]; ap = ba + bs2*bi[brow]; \ 158 rmax = bimax[brow]; nrow = bilen[brow]; \ 159 bcol = col/bs; \ 160 ridx = row % bs; cidx = col % bs; \ 161 low = 0; high = nrow; \ 162 while (high-low > 3) { \ 163 t = (low+high)/2; \ 164 if (rp[t] > bcol) high = t; \ 165 else low = t; \ 166 } \ 167 for (_i=low; _i<high; _i++) { \ 168 if (rp[_i] > bcol) break; \ 169 if (rp[_i] == bcol) { \ 170 bap = ap + bs2*_i + bs*cidx + ridx; \ 171 if (addv == ADD_VALUES) *bap += value; \ 172 else *bap = value; \ 173 goto b_noinsert; \ 174 } \ 175 } \ 176 if (b->nonew == 1) goto b_noinsert; \ 177 else if (b->nonew == -1) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 178 if (nrow >= rmax) { \ 179 /* there is no extra room in row, therefore enlarge */ \ 180 PetscInt new_nz = bi[b->mbs] + CHUNKSIZE,len,*new_i,*new_j; \ 181 MatScalar *new_a; \ 182 \ 183 if (b->nonew == -2) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) in the matrix", row, col); \ 184 \ 185 /* malloc new storage space */ \ 186 len = new_nz*(sizeof(PetscInt)+bs2*sizeof(MatScalar))+(b->mbs+1)*sizeof(PetscInt); \ 187 ierr = PetscMalloc(len,&new_a);CHKERRQ(ierr); \ 188 new_j = (PetscInt*)(new_a + bs2*new_nz); \ 189 new_i = new_j + new_nz; \ 190 \ 191 /* copy over old data into new slots */ \ 192 for (ii=0; ii<brow+1; ii++) {new_i[ii] = bi[ii];} \ 193 for (ii=brow+1; ii<b->mbs+1; ii++) {new_i[ii] = bi[ii]+CHUNKSIZE;} \ 194 ierr = PetscMemcpy(new_j,bj,(bi[brow]+nrow)*sizeof(PetscInt));CHKERRQ(ierr); \ 195 len = (new_nz - CHUNKSIZE - bi[brow] - nrow); \ 196 ierr = PetscMemcpy(new_j+bi[brow]+nrow+CHUNKSIZE,bj+bi[brow]+nrow,len*sizeof(PetscInt));CHKERRQ(ierr); \ 197 ierr = PetscMemcpy(new_a,ba,(bi[brow]+nrow)*bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 198 ierr = PetscMemzero(new_a+bs2*(bi[brow]+nrow),bs2*CHUNKSIZE*sizeof(MatScalar));CHKERRQ(ierr); \ 199 ierr = PetscMemcpy(new_a+bs2*(bi[brow]+nrow+CHUNKSIZE), \ 200 ba+bs2*(bi[brow]+nrow),bs2*len*sizeof(MatScalar));CHKERRQ(ierr); \ 201 /* free up old matrix storage */ \ 202 ierr = PetscFree(b->a);CHKERRQ(ierr); \ 203 if (!b->singlemalloc) { \ 204 ierr = PetscFree(b->i);CHKERRQ(ierr); \ 205 ierr = PetscFree(b->j);CHKERRQ(ierr); \ 206 } \ 207 ba = b->a = new_a; bi = b->i = new_i; bj = b->j = new_j; \ 208 b->singlemalloc = PETSC_TRUE; \ 209 \ 210 rp = bj + bi[brow]; ap = ba + bs2*bi[brow]; \ 211 rmax = bimax[brow] = bimax[brow] + CHUNKSIZE; \ 212 ierr = PetscLogObjectMemory(B,CHUNKSIZE*(sizeof(PetscInt) + bs2*sizeof(MatScalar)));CHKERRQ(ierr); \ 213 b->maxnz += bs2*CHUNKSIZE; \ 214 b->reallocs++; \ 215 b->nz++; \ 216 } \ 217 N = nrow++ - 1; \ 218 /* shift up all the later entries in this row */ \ 219 for (ii=N; ii>=_i; ii--) { \ 220 rp[ii+1] = rp[ii]; \ 221 ierr = PetscMemcpy(ap+bs2*(ii+1),ap+bs2*(ii),bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 222 } \ 223 if (N>=_i) { ierr = PetscMemzero(ap+bs2*_i,bs2*sizeof(MatScalar));CHKERRQ(ierr);} \ 224 rp[_i] = bcol; \ 225 ap[bs2*_i + bs*cidx + ridx] = value; \ 226 b_noinsert:; \ 227 bilen[brow] = nrow; \ 228 } 229 #endif 230 231 #if defined(PETSC_USE_MAT_SINGLE) 232 #undef __FUNCT__ 233 #define __FUNCT__ "MatSetValues_MPISBAIJ" 234 PetscErrorCode MatSetValues_MPISBAIJ(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const PetscScalar v[],InsertMode addv) 235 { 236 Mat_MPISBAIJ *b = (Mat_MPISBAIJ*)mat->data; 237 PetscErrorCode ierr; 238 PetscInt i,N = m*n; 239 MatScalar *vsingle; 240 241 PetscFunctionBegin; 242 if (N > b->setvalueslen) { 243 if (b->setvaluescopy) {ierr = PetscFree(b->setvaluescopy);CHKERRQ(ierr);} 244 ierr = PetscMalloc(N*sizeof(MatScalar),&b->setvaluescopy);CHKERRQ(ierr); 245 b->setvalueslen = N; 246 } 247 vsingle = b->setvaluescopy; 248 249 for (i=0; i<N; i++) { 250 vsingle[i] = v[i]; 251 } 252 ierr = MatSetValues_MPISBAIJ_MatScalar(mat,m,im,n,in,vsingle,addv);CHKERRQ(ierr); 253 PetscFunctionReturn(0); 254 } 255 256 #undef __FUNCT__ 257 #define __FUNCT__ "MatSetValuesBlocked_MPISBAIJ" 258 PetscErrorCode MatSetValuesBlocked_MPISBAIJ(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const PetscScalar v[],InsertMode addv) 259 { 260 Mat_MPIBAIJ *b = (Mat_MPIBAIJ*)mat->data; 261 PetscErrorCode ierr; 262 PetscInt i,N = m*n*b->bs2; 263 MatScalar *vsingle; 264 265 PetscFunctionBegin; 266 if (N > b->setvalueslen) { 267 if (b->setvaluescopy) {ierr = PetscFree(b->setvaluescopy);CHKERRQ(ierr);} 268 ierr = PetscMalloc(N*sizeof(MatScalar),&b->setvaluescopy);CHKERRQ(ierr); 269 b->setvalueslen = N; 270 } 271 vsingle = b->setvaluescopy; 272 for (i=0; i<N; i++) { 273 vsingle[i] = v[i]; 274 } 275 ierr = MatSetValuesBlocked_MPISBAIJ_MatScalar(mat,m,im,n,in,vsingle,addv);CHKERRQ(ierr); 276 PetscFunctionReturn(0); 277 } 278 #endif 279 280 /* Only add/insert a(i,j) with i<=j (blocks). 281 Any a(i,j) with i>j input by user is ingored. 282 */ 283 #undef __FUNCT__ 284 #define __FUNCT__ "MatSetValues_MPIBAIJ_MatScalar" 285 PetscErrorCode MatSetValues_MPISBAIJ_MatScalar(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const MatScalar v[],InsertMode addv) 286 { 287 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 288 MatScalar value; 289 PetscTruth roworiented = baij->roworiented; 290 PetscErrorCode ierr; 291 PetscInt i,j,row,col; 292 PetscInt rstart_orig=baij->rstart_bs; 293 PetscInt rend_orig=baij->rend_bs,cstart_orig=baij->cstart_bs; 294 PetscInt cend_orig=baij->cend_bs,bs=mat->bs; 295 296 /* Some Variables required in the macro */ 297 Mat A = baij->A; 298 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)(A)->data; 299 PetscInt *aimax=a->imax,*ai=a->i,*ailen=a->ilen,*aj=a->j; 300 MatScalar *aa=a->a; 301 302 Mat B = baij->B; 303 Mat_SeqBAIJ *b = (Mat_SeqBAIJ*)(B)->data; 304 PetscInt *bimax=b->imax,*bi=b->i,*bilen=b->ilen,*bj=b->j; 305 MatScalar *ba=b->a; 306 307 PetscInt *rp,ii,nrow,_i,rmax,N,brow,bcol; 308 PetscInt low,high,t,ridx,cidx,bs2=a->bs2; 309 MatScalar *ap,*bap; 310 311 /* for stash */ 312 PetscInt n_loc, *in_loc=0; 313 MatScalar *v_loc=0; 314 315 PetscFunctionBegin; 316 317 if(!baij->donotstash){ 318 ierr = PetscMalloc(n*sizeof(PetscInt),&in_loc);CHKERRQ(ierr); 319 ierr = PetscMalloc(n*sizeof(MatScalar),&v_loc);CHKERRQ(ierr); 320 } 321 322 for (i=0; i<m; i++) { 323 if (im[i] < 0) continue; 324 #if defined(PETSC_USE_DEBUG) 325 if (im[i] >= mat->M) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",im[i],mat->M-1); 326 #endif 327 if (im[i] >= rstart_orig && im[i] < rend_orig) { /* this processor entry */ 328 row = im[i] - rstart_orig; /* local row index */ 329 for (j=0; j<n; j++) { 330 if (im[i]/bs > in[j]/bs) continue; /* ignore lower triangular blocks */ 331 if (in[j] >= cstart_orig && in[j] < cend_orig){ /* diag entry (A) */ 332 col = in[j] - cstart_orig; /* local col index */ 333 brow = row/bs; bcol = col/bs; 334 if (brow > bcol) continue; /* ignore lower triangular blocks of A */ 335 if (roworiented) value = v[i*n+j]; else value = v[i+j*m]; 336 MatSetValues_SeqSBAIJ_A_Private(row,col,value,addv); 337 /* ierr = MatSetValues_SeqBAIJ(baij->A,1,&row,1,&col,&value,addv);CHKERRQ(ierr); */ 338 } else if (in[j] < 0) continue; 339 #if defined(PETSC_USE_DEBUG) 340 else if (in[j] >= mat->N) {SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",in[j],mat->N-1);} 341 #endif 342 else { /* off-diag entry (B) */ 343 if (mat->was_assembled) { 344 if (!baij->colmap) { 345 ierr = CreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 346 } 347 #if defined (PETSC_USE_CTABLE) 348 ierr = PetscTableFind(baij->colmap,in[j]/bs + 1,&col);CHKERRQ(ierr); 349 col = col - 1; 350 #else 351 col = baij->colmap[in[j]/bs] - 1; 352 #endif 353 if (col < 0 && !((Mat_SeqSBAIJ*)(baij->A->data))->nonew) { 354 ierr = DisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 355 col = in[j]; 356 /* Reinitialize the variables required by MatSetValues_SeqBAIJ_B_Private() */ 357 B = baij->B; 358 b = (Mat_SeqBAIJ*)(B)->data; 359 bimax=b->imax;bi=b->i;bilen=b->ilen;bj=b->j; 360 ba=b->a; 361 } else col += in[j]%bs; 362 } else col = in[j]; 363 if (roworiented) value = v[i*n+j]; else value = v[i+j*m]; 364 MatSetValues_SeqSBAIJ_B_Private(row,col,value,addv); 365 /* ierr = MatSetValues_SeqBAIJ(baij->B,1,&row,1,&col,&value,addv);CHKERRQ(ierr); */ 366 } 367 } 368 } else { /* off processor entry */ 369 if (!baij->donotstash) { 370 n_loc = 0; 371 for (j=0; j<n; j++){ 372 if (im[i]/bs > in[j]/bs) continue; /* ignore lower triangular blocks */ 373 in_loc[n_loc] = in[j]; 374 if (roworiented) { 375 v_loc[n_loc] = v[i*n+j]; 376 } else { 377 v_loc[n_loc] = v[j*m+i]; 378 } 379 n_loc++; 380 } 381 ierr = MatStashValuesRow_Private(&mat->stash,im[i],n_loc,in_loc,v_loc);CHKERRQ(ierr); 382 } 383 } 384 } 385 386 if(!baij->donotstash){ 387 ierr = PetscFree(in_loc);CHKERRQ(ierr); 388 ierr = PetscFree(v_loc);CHKERRQ(ierr); 389 } 390 PetscFunctionReturn(0); 391 } 392 393 #undef __FUNCT__ 394 #define __FUNCT__ "MatSetValuesBlocked_MPISBAIJ_MatScalar" 395 PetscErrorCode MatSetValuesBlocked_MPISBAIJ_MatScalar(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const MatScalar v[],InsertMode addv) 396 { 397 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 398 const MatScalar *value; 399 MatScalar *barray=baij->barray; 400 PetscTruth roworiented = baij->roworiented; 401 PetscErrorCode ierr; 402 PetscInt i,j,ii,jj,row,col,rstart=baij->rstart; 403 PetscInt rend=baij->rend,cstart=baij->cstart,stepval; 404 PetscInt cend=baij->cend,bs=mat->bs,bs2=baij->bs2; 405 406 PetscFunctionBegin; 407 if(!barray) { 408 ierr = PetscMalloc(bs2*sizeof(MatScalar),&barray);CHKERRQ(ierr); 409 baij->barray = barray; 410 } 411 412 if (roworiented) { 413 stepval = (n-1)*bs; 414 } else { 415 stepval = (m-1)*bs; 416 } 417 for (i=0; i<m; i++) { 418 if (im[i] < 0) continue; 419 #if defined(PETSC_USE_DEBUG) 420 if (im[i] >= baij->Mbs) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Row too large, row %D max %D",im[i],baij->Mbs-1); 421 #endif 422 if (im[i] >= rstart && im[i] < rend) { 423 row = im[i] - rstart; 424 for (j=0; j<n; j++) { 425 /* If NumCol = 1 then a copy is not required */ 426 if ((roworiented) && (n == 1)) { 427 barray = (MatScalar*) v + i*bs2; 428 } else if((!roworiented) && (m == 1)) { 429 barray = (MatScalar*) v + j*bs2; 430 } else { /* Here a copy is required */ 431 if (roworiented) { 432 value = v + i*(stepval+bs)*bs + j*bs; 433 } else { 434 value = v + j*(stepval+bs)*bs + i*bs; 435 } 436 for (ii=0; ii<bs; ii++,value+=stepval) { 437 for (jj=0; jj<bs; jj++) { 438 *barray++ = *value++; 439 } 440 } 441 barray -=bs2; 442 } 443 444 if (in[j] >= cstart && in[j] < cend){ 445 col = in[j] - cstart; 446 ierr = MatSetValuesBlocked_SeqSBAIJ(baij->A,1,&row,1,&col,barray,addv);CHKERRQ(ierr); 447 } 448 else if (in[j] < 0) continue; 449 #if defined(PETSC_USE_DEBUG) 450 else if (in[j] >= baij->Nbs) {SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Column too large, col %D max %D",in[j],baij->Nbs-1);} 451 #endif 452 else { 453 if (mat->was_assembled) { 454 if (!baij->colmap) { 455 ierr = CreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 456 } 457 458 #if defined(PETSC_USE_DEBUG) 459 #if defined (PETSC_USE_CTABLE) 460 { PetscInt data; 461 ierr = PetscTableFind(baij->colmap,in[j]+1,&data);CHKERRQ(ierr); 462 if ((data - 1) % bs) SETERRQ(PETSC_ERR_PLIB,"Incorrect colmap"); 463 } 464 #else 465 if ((baij->colmap[in[j]] - 1) % bs) SETERRQ(PETSC_ERR_PLIB,"Incorrect colmap"); 466 #endif 467 #endif 468 #if defined (PETSC_USE_CTABLE) 469 ierr = PetscTableFind(baij->colmap,in[j]+1,&col);CHKERRQ(ierr); 470 col = (col - 1)/bs; 471 #else 472 col = (baij->colmap[in[j]] - 1)/bs; 473 #endif 474 if (col < 0 && !((Mat_SeqBAIJ*)(baij->A->data))->nonew) { 475 ierr = DisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 476 col = in[j]; 477 } 478 } 479 else col = in[j]; 480 ierr = MatSetValuesBlocked_SeqBAIJ(baij->B,1,&row,1,&col,barray,addv);CHKERRQ(ierr); 481 } 482 } 483 } else { 484 if (!baij->donotstash) { 485 if (roworiented) { 486 ierr = MatStashValuesRowBlocked_Private(&mat->bstash,im[i],n,in,v,m,n,i);CHKERRQ(ierr); 487 } else { 488 ierr = MatStashValuesColBlocked_Private(&mat->bstash,im[i],n,in,v,m,n,i);CHKERRQ(ierr); 489 } 490 } 491 } 492 } 493 PetscFunctionReturn(0); 494 } 495 496 #undef __FUNCT__ 497 #define __FUNCT__ "MatGetValues_MPISBAIJ" 498 PetscErrorCode MatGetValues_MPISBAIJ(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[]) 499 { 500 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 501 PetscErrorCode ierr; 502 PetscInt bs=mat->bs,i,j,bsrstart = baij->rstart*bs,bsrend = baij->rend*bs; 503 PetscInt bscstart = baij->cstart*bs,bscend = baij->cend*bs,row,col,data; 504 505 PetscFunctionBegin; 506 for (i=0; i<m; i++) { 507 if (idxm[i] < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Negative row: %D",idxm[i]); 508 if (idxm[i] >= mat->M) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",idxm[i],mat->M-1); 509 if (idxm[i] >= bsrstart && idxm[i] < bsrend) { 510 row = idxm[i] - bsrstart; 511 for (j=0; j<n; j++) { 512 if (idxn[j] < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Negative column %D",idxn[j]); 513 if (idxn[j] >= mat->N) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",idxn[j],mat->N-1); 514 if (idxn[j] >= bscstart && idxn[j] < bscend){ 515 col = idxn[j] - bscstart; 516 ierr = MatGetValues_SeqSBAIJ(baij->A,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 517 } else { 518 if (!baij->colmap) { 519 ierr = CreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 520 } 521 #if defined (PETSC_USE_CTABLE) 522 ierr = PetscTableFind(baij->colmap,idxn[j]/bs+1,&data);CHKERRQ(ierr); 523 data --; 524 #else 525 data = baij->colmap[idxn[j]/bs]-1; 526 #endif 527 if((data < 0) || (baij->garray[data/bs] != idxn[j]/bs)) *(v+i*n+j) = 0.0; 528 else { 529 col = data + idxn[j]%bs; 530 ierr = MatGetValues_SeqBAIJ(baij->B,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 531 } 532 } 533 } 534 } else { 535 SETERRQ(PETSC_ERR_SUP,"Only local values currently supported"); 536 } 537 } 538 PetscFunctionReturn(0); 539 } 540 541 #undef __FUNCT__ 542 #define __FUNCT__ "MatNorm_MPISBAIJ" 543 PetscErrorCode MatNorm_MPISBAIJ(Mat mat,NormType type,PetscReal *norm) 544 { 545 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 546 PetscErrorCode ierr; 547 PetscReal sum[2],*lnorm2; 548 549 PetscFunctionBegin; 550 if (baij->size == 1) { 551 ierr = MatNorm(baij->A,type,norm);CHKERRQ(ierr); 552 } else { 553 if (type == NORM_FROBENIUS) { 554 ierr = PetscMalloc(2*sizeof(PetscReal),&lnorm2);CHKERRQ(ierr); 555 ierr = MatNorm(baij->A,type,lnorm2);CHKERRQ(ierr); 556 *lnorm2 = (*lnorm2)*(*lnorm2); lnorm2++; /* squar power of norm(A) */ 557 ierr = MatNorm(baij->B,type,lnorm2);CHKERRQ(ierr); 558 *lnorm2 = (*lnorm2)*(*lnorm2); lnorm2--; /* squar power of norm(B) */ 559 ierr = MPI_Allreduce(lnorm2,&sum,2,MPIU_REAL,MPI_SUM,mat->comm);CHKERRQ(ierr); 560 *norm = sqrt(sum[0] + 2*sum[1]); 561 ierr = PetscFree(lnorm2);CHKERRQ(ierr); 562 } else if (type == NORM_INFINITY || type == NORM_1) { /* max row/column sum */ 563 Mat_SeqSBAIJ *amat=(Mat_SeqSBAIJ*)baij->A->data; 564 Mat_SeqBAIJ *bmat=(Mat_SeqBAIJ*)baij->B->data; 565 PetscReal *rsum,*rsum2,vabs; 566 PetscInt *jj,*garray=baij->garray,rstart=baij->rstart,nz; 567 PetscInt brow,bcol,col,bs=baij->A->bs,row,grow,gcol,mbs=amat->mbs; 568 MatScalar *v; 569 570 ierr = PetscMalloc((2*mat->N+1)*sizeof(PetscReal),&rsum);CHKERRQ(ierr); 571 rsum2 = rsum + mat->N; 572 ierr = PetscMemzero(rsum,mat->N*sizeof(PetscReal));CHKERRQ(ierr); 573 /* Amat */ 574 v = amat->a; jj = amat->j; 575 for (brow=0; brow<mbs; brow++) { 576 grow = bs*(rstart + brow); 577 nz = amat->i[brow+1] - amat->i[brow]; 578 for (bcol=0; bcol<nz; bcol++){ 579 gcol = bs*(rstart + *jj); jj++; 580 for (col=0; col<bs; col++){ 581 for (row=0; row<bs; row++){ 582 vabs = PetscAbsScalar(*v); v++; 583 rsum[gcol+col] += vabs; 584 /* non-diagonal block */ 585 if (bcol > 0 && vabs > 0.0) rsum[grow+row] += vabs; 586 } 587 } 588 } 589 } 590 /* Bmat */ 591 v = bmat->a; jj = bmat->j; 592 for (brow=0; brow<mbs; brow++) { 593 grow = bs*(rstart + brow); 594 nz = bmat->i[brow+1] - bmat->i[brow]; 595 for (bcol=0; bcol<nz; bcol++){ 596 gcol = bs*garray[*jj]; jj++; 597 for (col=0; col<bs; col++){ 598 for (row=0; row<bs; row++){ 599 vabs = PetscAbsScalar(*v); v++; 600 rsum[gcol+col] += vabs; 601 rsum[grow+row] += vabs; 602 } 603 } 604 } 605 } 606 ierr = MPI_Allreduce(rsum,rsum2,mat->N,MPIU_REAL,MPI_SUM,mat->comm);CHKERRQ(ierr); 607 *norm = 0.0; 608 for (col=0; col<mat->N; col++) { 609 if (rsum2[col] > *norm) *norm = rsum2[col]; 610 } 611 ierr = PetscFree(rsum);CHKERRQ(ierr); 612 } else { 613 SETERRQ(PETSC_ERR_SUP,"No support for this norm yet"); 614 } 615 } 616 PetscFunctionReturn(0); 617 } 618 619 #undef __FUNCT__ 620 #define __FUNCT__ "MatAssemblyBegin_MPISBAIJ" 621 PetscErrorCode MatAssemblyBegin_MPISBAIJ(Mat mat,MatAssemblyType mode) 622 { 623 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 624 PetscErrorCode ierr; 625 PetscInt nstash,reallocs; 626 InsertMode addv; 627 628 PetscFunctionBegin; 629 if (baij->donotstash) { 630 PetscFunctionReturn(0); 631 } 632 633 /* make sure all processors are either in INSERTMODE or ADDMODE */ 634 ierr = MPI_Allreduce(&mat->insertmode,&addv,1,MPI_INT,MPI_BOR,mat->comm);CHKERRQ(ierr); 635 if (addv == (ADD_VALUES|INSERT_VALUES)) { 636 SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Some processors inserted others added"); 637 } 638 mat->insertmode = addv; /* in case this processor had no cache */ 639 640 ierr = MatStashScatterBegin_Private(&mat->stash,baij->rowners_bs);CHKERRQ(ierr); 641 ierr = MatStashScatterBegin_Private(&mat->bstash,baij->rowners);CHKERRQ(ierr); 642 ierr = MatStashGetInfo_Private(&mat->stash,&nstash,&reallocs);CHKERRQ(ierr); 643 ierr = PetscLogInfo((0,"MatAssemblyBegin_MPISBAIJ:Stash has %D entries,uses %D mallocs.\n",nstash,reallocs));CHKERRQ(ierr); 644 ierr = MatStashGetInfo_Private(&mat->stash,&nstash,&reallocs);CHKERRQ(ierr); 645 ierr = PetscLogInfo((0,"MatAssemblyBegin_MPISBAIJ:Block-Stash has %D entries, uses %D mallocs.\n",nstash,reallocs));CHKERRQ(ierr); 646 PetscFunctionReturn(0); 647 } 648 649 #undef __FUNCT__ 650 #define __FUNCT__ "MatAssemblyEnd_MPISBAIJ" 651 PetscErrorCode MatAssemblyEnd_MPISBAIJ(Mat mat,MatAssemblyType mode) 652 { 653 Mat_MPISBAIJ *baij=(Mat_MPISBAIJ*)mat->data; 654 Mat_SeqSBAIJ *a=(Mat_SeqSBAIJ*)baij->A->data; 655 Mat_SeqBAIJ *b=(Mat_SeqBAIJ*)baij->B->data; 656 PetscErrorCode ierr; 657 PetscInt i,j,rstart,ncols,flg,bs2=baij->bs2; 658 PetscInt *row,*col,other_disassembled; 659 PetscMPIInt n; 660 PetscTruth r1,r2,r3; 661 MatScalar *val; 662 InsertMode addv = mat->insertmode; 663 664 PetscFunctionBegin; 665 666 if (!baij->donotstash) { 667 while (1) { 668 ierr = MatStashScatterGetMesg_Private(&mat->stash,&n,&row,&col,&val,&flg);CHKERRQ(ierr); 669 if (!flg) break; 670 671 for (i=0; i<n;) { 672 /* Now identify the consecutive vals belonging to the same row */ 673 for (j=i,rstart=row[j]; j<n; j++) { if (row[j] != rstart) break; } 674 if (j < n) ncols = j-i; 675 else ncols = n-i; 676 /* Now assemble all these values with a single function call */ 677 ierr = MatSetValues_MPISBAIJ_MatScalar(mat,1,row+i,ncols,col+i,val+i,addv);CHKERRQ(ierr); 678 i = j; 679 } 680 } 681 ierr = MatStashScatterEnd_Private(&mat->stash);CHKERRQ(ierr); 682 /* Now process the block-stash. Since the values are stashed column-oriented, 683 set the roworiented flag to column oriented, and after MatSetValues() 684 restore the original flags */ 685 r1 = baij->roworiented; 686 r2 = a->roworiented; 687 r3 = b->roworiented; 688 baij->roworiented = PETSC_FALSE; 689 a->roworiented = PETSC_FALSE; 690 b->roworiented = PETSC_FALSE; 691 while (1) { 692 ierr = MatStashScatterGetMesg_Private(&mat->bstash,&n,&row,&col,&val,&flg);CHKERRQ(ierr); 693 if (!flg) break; 694 695 for (i=0; i<n;) { 696 /* Now identify the consecutive vals belonging to the same row */ 697 for (j=i,rstart=row[j]; j<n; j++) { if (row[j] != rstart) break; } 698 if (j < n) ncols = j-i; 699 else ncols = n-i; 700 ierr = MatSetValuesBlocked_MPISBAIJ_MatScalar(mat,1,row+i,ncols,col+i,val+i*bs2,addv);CHKERRQ(ierr); 701 i = j; 702 } 703 } 704 ierr = MatStashScatterEnd_Private(&mat->bstash);CHKERRQ(ierr); 705 baij->roworiented = r1; 706 a->roworiented = r2; 707 b->roworiented = r3; 708 } 709 710 ierr = MatAssemblyBegin(baij->A,mode);CHKERRQ(ierr); 711 ierr = MatAssemblyEnd(baij->A,mode);CHKERRQ(ierr); 712 713 /* determine if any processor has disassembled, if so we must 714 also disassemble ourselfs, in order that we may reassemble. */ 715 /* 716 if nonzero structure of submatrix B cannot change then we know that 717 no processor disassembled thus we can skip this stuff 718 */ 719 if (!((Mat_SeqBAIJ*)baij->B->data)->nonew) { 720 ierr = MPI_Allreduce(&mat->was_assembled,&other_disassembled,1,MPI_INT,MPI_PROD,mat->comm);CHKERRQ(ierr); 721 if (mat->was_assembled && !other_disassembled) { 722 ierr = DisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 723 } 724 } 725 726 if (!mat->was_assembled && mode == MAT_FINAL_ASSEMBLY) { 727 ierr = MatSetUpMultiply_MPISBAIJ(mat);CHKERRQ(ierr); /* setup Mvctx and sMvctx */ 728 } 729 b->compressedrow.use = PETSC_TRUE; 730 ierr = MatAssemblyBegin(baij->B,mode);CHKERRQ(ierr); 731 ierr = MatAssemblyEnd(baij->B,mode);CHKERRQ(ierr); 732 733 if (baij->rowvalues) { 734 ierr = PetscFree(baij->rowvalues);CHKERRQ(ierr); 735 baij->rowvalues = 0; 736 } 737 738 PetscFunctionReturn(0); 739 } 740 741 #undef __FUNCT__ 742 #define __FUNCT__ "MatView_MPISBAIJ_ASCIIorDraworSocket" 743 static PetscErrorCode MatView_MPISBAIJ_ASCIIorDraworSocket(Mat mat,PetscViewer viewer) 744 { 745 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 746 PetscErrorCode ierr; 747 PetscInt bs = mat->bs; 748 PetscMPIInt size = baij->size,rank = baij->rank; 749 PetscTruth iascii,isdraw; 750 PetscViewer sviewer; 751 PetscViewerFormat format; 752 753 PetscFunctionBegin; 754 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); 755 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_DRAW,&isdraw);CHKERRQ(ierr); 756 if (iascii) { 757 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 758 if (format == PETSC_VIEWER_ASCII_INFO_DETAIL) { 759 MatInfo info; 760 ierr = MPI_Comm_rank(mat->comm,&rank);CHKERRQ(ierr); 761 ierr = MatGetInfo(mat,MAT_LOCAL,&info);CHKERRQ(ierr); 762 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] Local rows %D nz %D nz alloced %D bs %D mem %D\n", 763 rank,mat->m,(PetscInt)info.nz_used*bs,(PetscInt)info.nz_allocated*bs, 764 mat->bs,(PetscInt)info.memory);CHKERRQ(ierr); 765 ierr = MatGetInfo(baij->A,MAT_LOCAL,&info);CHKERRQ(ierr); 766 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] on-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used*bs);CHKERRQ(ierr); 767 ierr = MatGetInfo(baij->B,MAT_LOCAL,&info);CHKERRQ(ierr); 768 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] off-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used*bs);CHKERRQ(ierr); 769 ierr = PetscViewerFlush(viewer);CHKERRQ(ierr); 770 ierr = VecScatterView(baij->Mvctx,viewer);CHKERRQ(ierr); 771 PetscFunctionReturn(0); 772 } else if (format == PETSC_VIEWER_ASCII_INFO) { 773 ierr = PetscViewerASCIIPrintf(viewer," block size is %D\n",bs);CHKERRQ(ierr); 774 PetscFunctionReturn(0); 775 } 776 } 777 778 if (isdraw) { 779 PetscDraw draw; 780 PetscTruth isnull; 781 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 782 ierr = PetscDrawIsNull(draw,&isnull);CHKERRQ(ierr); if (isnull) PetscFunctionReturn(0); 783 } 784 785 if (size == 1) { 786 ierr = PetscObjectSetName((PetscObject)baij->A,mat->name);CHKERRQ(ierr); 787 ierr = MatView(baij->A,viewer);CHKERRQ(ierr); 788 } else { 789 /* assemble the entire matrix onto first processor. */ 790 Mat A; 791 Mat_SeqSBAIJ *Aloc; 792 Mat_SeqBAIJ *Bloc; 793 PetscInt M = mat->M,N = mat->N,*ai,*aj,col,i,j,k,*rvals,mbs = baij->mbs; 794 MatScalar *a; 795 796 /* Should this be the same type as mat? */ 797 if (!rank) { 798 ierr = MatCreate(mat->comm,M,N,M,N,&A);CHKERRQ(ierr); 799 } else { 800 ierr = MatCreate(mat->comm,0,0,M,N,&A);CHKERRQ(ierr); 801 } 802 ierr = MatSetType(A,MATMPISBAIJ);CHKERRQ(ierr); 803 ierr = MatMPISBAIJSetPreallocation(A,mat->bs,0,PETSC_NULL,0,PETSC_NULL);CHKERRQ(ierr); 804 ierr = PetscLogObjectParent(mat,A);CHKERRQ(ierr); 805 806 /* copy over the A part */ 807 Aloc = (Mat_SeqSBAIJ*)baij->A->data; 808 ai = Aloc->i; aj = Aloc->j; a = Aloc->a; 809 ierr = PetscMalloc(bs*sizeof(PetscInt),&rvals);CHKERRQ(ierr); 810 811 for (i=0; i<mbs; i++) { 812 rvals[0] = bs*(baij->rstart + i); 813 for (j=1; j<bs; j++) { rvals[j] = rvals[j-1] + 1; } 814 for (j=ai[i]; j<ai[i+1]; j++) { 815 col = (baij->cstart+aj[j])*bs; 816 for (k=0; k<bs; k++) { 817 ierr = MatSetValues_MPISBAIJ_MatScalar(A,bs,rvals,1,&col,a,INSERT_VALUES);CHKERRQ(ierr); 818 col++; a += bs; 819 } 820 } 821 } 822 /* copy over the B part */ 823 Bloc = (Mat_SeqBAIJ*)baij->B->data; 824 ai = Bloc->i; aj = Bloc->j; a = Bloc->a; 825 for (i=0; i<mbs; i++) { 826 rvals[0] = bs*(baij->rstart + i); 827 for (j=1; j<bs; j++) { rvals[j] = rvals[j-1] + 1; } 828 for (j=ai[i]; j<ai[i+1]; j++) { 829 col = baij->garray[aj[j]]*bs; 830 for (k=0; k<bs; k++) { 831 ierr = MatSetValues_MPISBAIJ_MatScalar(A,bs,rvals,1,&col,a,INSERT_VALUES);CHKERRQ(ierr); 832 col++; a += bs; 833 } 834 } 835 } 836 ierr = PetscFree(rvals);CHKERRQ(ierr); 837 ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 838 ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 839 /* 840 Everyone has to call to draw the matrix since the graphics waits are 841 synchronized across all processors that share the PetscDraw object 842 */ 843 ierr = PetscViewerGetSingleton(viewer,&sviewer);CHKERRQ(ierr); 844 if (!rank) { 845 ierr = PetscObjectSetName((PetscObject)((Mat_MPISBAIJ*)(A->data))->A,mat->name);CHKERRQ(ierr); 846 ierr = MatView(((Mat_MPISBAIJ*)(A->data))->A,sviewer);CHKERRQ(ierr); 847 } 848 ierr = PetscViewerRestoreSingleton(viewer,&sviewer);CHKERRQ(ierr); 849 ierr = MatDestroy(A);CHKERRQ(ierr); 850 } 851 PetscFunctionReturn(0); 852 } 853 854 #undef __FUNCT__ 855 #define __FUNCT__ "MatView_MPISBAIJ" 856 PetscErrorCode MatView_MPISBAIJ(Mat mat,PetscViewer viewer) 857 { 858 PetscErrorCode ierr; 859 PetscTruth iascii,isdraw,issocket,isbinary; 860 861 PetscFunctionBegin; 862 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); 863 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_DRAW,&isdraw);CHKERRQ(ierr); 864 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_SOCKET,&issocket);CHKERRQ(ierr); 865 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_BINARY,&isbinary);CHKERRQ(ierr); 866 if (iascii || isdraw || issocket || isbinary) { 867 ierr = MatView_MPISBAIJ_ASCIIorDraworSocket(mat,viewer);CHKERRQ(ierr); 868 } else { 869 SETERRQ1(PETSC_ERR_SUP,"Viewer type %s not supported by MPISBAIJ matrices",((PetscObject)viewer)->type_name); 870 } 871 PetscFunctionReturn(0); 872 } 873 874 #undef __FUNCT__ 875 #define __FUNCT__ "MatDestroy_MPISBAIJ" 876 PetscErrorCode MatDestroy_MPISBAIJ(Mat mat) 877 { 878 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 879 PetscErrorCode ierr; 880 881 PetscFunctionBegin; 882 #if defined(PETSC_USE_LOG) 883 PetscLogObjectState((PetscObject)mat,"Rows=%D,Cols=%D",mat->M,mat->N); 884 #endif 885 ierr = MatStashDestroy_Private(&mat->stash);CHKERRQ(ierr); 886 ierr = MatStashDestroy_Private(&mat->bstash);CHKERRQ(ierr); 887 ierr = PetscFree(baij->rowners);CHKERRQ(ierr); 888 ierr = MatDestroy(baij->A);CHKERRQ(ierr); 889 ierr = MatDestroy(baij->B);CHKERRQ(ierr); 890 #if defined (PETSC_USE_CTABLE) 891 if (baij->colmap) {ierr = PetscTableDelete(baij->colmap);CHKERRQ(ierr);} 892 #else 893 if (baij->colmap) {ierr = PetscFree(baij->colmap);CHKERRQ(ierr);} 894 #endif 895 if (baij->garray) {ierr = PetscFree(baij->garray);CHKERRQ(ierr);} 896 if (baij->lvec) {ierr = VecDestroy(baij->lvec);CHKERRQ(ierr);} 897 if (baij->Mvctx) {ierr = VecScatterDestroy(baij->Mvctx);CHKERRQ(ierr);} 898 if (baij->slvec0) { 899 ierr = VecDestroy(baij->slvec0);CHKERRQ(ierr); 900 ierr = VecDestroy(baij->slvec0b);CHKERRQ(ierr); 901 } 902 if (baij->slvec1) { 903 ierr = VecDestroy(baij->slvec1);CHKERRQ(ierr); 904 ierr = VecDestroy(baij->slvec1a);CHKERRQ(ierr); 905 ierr = VecDestroy(baij->slvec1b);CHKERRQ(ierr); 906 } 907 if (baij->sMvctx) {ierr = VecScatterDestroy(baij->sMvctx);CHKERRQ(ierr);} 908 if (baij->rowvalues) {ierr = PetscFree(baij->rowvalues);CHKERRQ(ierr);} 909 if (baij->barray) {ierr = PetscFree(baij->barray);CHKERRQ(ierr);} 910 if (baij->hd) {ierr = PetscFree(baij->hd);CHKERRQ(ierr);} 911 #if defined(PETSC_USE_MAT_SINGLE) 912 if (baij->setvaluescopy) {ierr = PetscFree(baij->setvaluescopy);CHKERRQ(ierr);} 913 #endif 914 ierr = PetscFree(baij);CHKERRQ(ierr); 915 916 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatStoreValues_C","",PETSC_NULL);CHKERRQ(ierr); 917 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatRetrieveValues_C","",PETSC_NULL);CHKERRQ(ierr); 918 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatGetDiagonalBlock_C","",PETSC_NULL);CHKERRQ(ierr); 919 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatMPISBAIJSetPreallocation_C","",PETSC_NULL);CHKERRQ(ierr); 920 PetscFunctionReturn(0); 921 } 922 923 #undef __FUNCT__ 924 #define __FUNCT__ "MatMult_MPISBAIJ" 925 PetscErrorCode MatMult_MPISBAIJ(Mat A,Vec xx,Vec yy) 926 { 927 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 928 PetscErrorCode ierr; 929 PetscInt nt,mbs=a->mbs,bs=A->bs; 930 PetscScalar *x,*from,zero=0.0; 931 932 PetscFunctionBegin; 933 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 934 if (nt != A->n) { 935 SETERRQ(PETSC_ERR_ARG_SIZ,"Incompatible partition of A and xx"); 936 } 937 ierr = VecGetLocalSize(yy,&nt);CHKERRQ(ierr); 938 if (nt != A->m) { 939 SETERRQ(PETSC_ERR_ARG_SIZ,"Incompatible parition of A and yy"); 940 } 941 942 /* diagonal part */ 943 ierr = (*a->A->ops->mult)(a->A,xx,a->slvec1a);CHKERRQ(ierr); 944 ierr = VecSet(&zero,a->slvec1b);CHKERRQ(ierr); 945 946 /* subdiagonal part */ 947 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->slvec0b);CHKERRQ(ierr); 948 949 /* copy x into the vec slvec0 */ 950 ierr = VecGetArray(a->slvec0,&from);CHKERRQ(ierr); 951 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 952 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 953 ierr = VecRestoreArray(a->slvec0,&from);CHKERRQ(ierr); 954 955 ierr = VecScatterBegin(a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD,a->sMvctx);CHKERRQ(ierr); 956 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 957 ierr = VecScatterEnd(a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD,a->sMvctx);CHKERRQ(ierr); 958 959 /* supperdiagonal part */ 960 ierr = (*a->B->ops->multadd)(a->B,a->slvec1b,a->slvec1a,yy);CHKERRQ(ierr); 961 962 PetscFunctionReturn(0); 963 } 964 965 #undef __FUNCT__ 966 #define __FUNCT__ "MatMult_MPISBAIJ_2comm" 967 PetscErrorCode MatMult_MPISBAIJ_2comm(Mat A,Vec xx,Vec yy) 968 { 969 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 970 PetscErrorCode ierr; 971 PetscInt nt; 972 973 PetscFunctionBegin; 974 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 975 if (nt != A->n) { 976 SETERRQ(PETSC_ERR_ARG_SIZ,"Incompatible partition of A and xx"); 977 } 978 ierr = VecGetLocalSize(yy,&nt);CHKERRQ(ierr); 979 if (nt != A->m) { 980 SETERRQ(PETSC_ERR_ARG_SIZ,"Incompatible parition of A and yy"); 981 } 982 983 ierr = VecScatterBegin(xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD,a->Mvctx);CHKERRQ(ierr); 984 /* do diagonal part */ 985 ierr = (*a->A->ops->mult)(a->A,xx,yy);CHKERRQ(ierr); 986 /* do supperdiagonal part */ 987 ierr = VecScatterEnd(xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD,a->Mvctx);CHKERRQ(ierr); 988 ierr = (*a->B->ops->multadd)(a->B,a->lvec,yy,yy);CHKERRQ(ierr); 989 /* do subdiagonal part */ 990 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 991 ierr = VecScatterBegin(a->lvec,yy,ADD_VALUES,SCATTER_REVERSE,a->Mvctx);CHKERRQ(ierr); 992 ierr = VecScatterEnd(a->lvec,yy,ADD_VALUES,SCATTER_REVERSE,a->Mvctx);CHKERRQ(ierr); 993 994 PetscFunctionReturn(0); 995 } 996 997 #undef __FUNCT__ 998 #define __FUNCT__ "MatMultAdd_MPISBAIJ" 999 PetscErrorCode MatMultAdd_MPISBAIJ(Mat A,Vec xx,Vec yy,Vec zz) 1000 { 1001 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1002 PetscErrorCode ierr; 1003 PetscInt mbs=a->mbs,bs=A->bs; 1004 PetscScalar *x,*from,zero=0.0; 1005 1006 PetscFunctionBegin; 1007 /* 1008 PetscSynchronizedPrintf(A->comm," MatMultAdd is called ...\n"); 1009 PetscSynchronizedFlush(A->comm); 1010 */ 1011 /* diagonal part */ 1012 ierr = (*a->A->ops->multadd)(a->A,xx,yy,a->slvec1a);CHKERRQ(ierr); 1013 ierr = VecSet(&zero,a->slvec1b);CHKERRQ(ierr); 1014 1015 /* subdiagonal part */ 1016 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->slvec0b);CHKERRQ(ierr); 1017 1018 /* copy x into the vec slvec0 */ 1019 ierr = VecGetArray(a->slvec0,&from);CHKERRQ(ierr); 1020 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 1021 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 1022 ierr = VecRestoreArray(a->slvec0,&from);CHKERRQ(ierr); 1023 1024 ierr = VecScatterBegin(a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD,a->sMvctx);CHKERRQ(ierr); 1025 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 1026 ierr = VecScatterEnd(a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD,a->sMvctx);CHKERRQ(ierr); 1027 1028 /* supperdiagonal part */ 1029 ierr = (*a->B->ops->multadd)(a->B,a->slvec1b,a->slvec1a,zz);CHKERRQ(ierr); 1030 1031 PetscFunctionReturn(0); 1032 } 1033 1034 #undef __FUNCT__ 1035 #define __FUNCT__ "MatMultAdd_MPISBAIJ_2comm" 1036 PetscErrorCode MatMultAdd_MPISBAIJ_2comm(Mat A,Vec xx,Vec yy,Vec zz) 1037 { 1038 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1039 PetscErrorCode ierr; 1040 1041 PetscFunctionBegin; 1042 ierr = VecScatterBegin(xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD,a->Mvctx);CHKERRQ(ierr); 1043 /* do diagonal part */ 1044 ierr = (*a->A->ops->multadd)(a->A,xx,yy,zz);CHKERRQ(ierr); 1045 /* do supperdiagonal part */ 1046 ierr = VecScatterEnd(xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD,a->Mvctx);CHKERRQ(ierr); 1047 ierr = (*a->B->ops->multadd)(a->B,a->lvec,zz,zz);CHKERRQ(ierr); 1048 1049 /* do subdiagonal part */ 1050 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 1051 ierr = VecScatterBegin(a->lvec,zz,ADD_VALUES,SCATTER_REVERSE,a->Mvctx);CHKERRQ(ierr); 1052 ierr = VecScatterEnd(a->lvec,zz,ADD_VALUES,SCATTER_REVERSE,a->Mvctx);CHKERRQ(ierr); 1053 1054 PetscFunctionReturn(0); 1055 } 1056 1057 #undef __FUNCT__ 1058 #define __FUNCT__ "MatMultTranspose_MPISBAIJ" 1059 PetscErrorCode MatMultTranspose_MPISBAIJ(Mat A,Vec xx,Vec yy) 1060 { 1061 PetscErrorCode ierr; 1062 1063 PetscFunctionBegin; 1064 ierr = MatMult(A,xx,yy);CHKERRQ(ierr); 1065 PetscFunctionReturn(0); 1066 } 1067 1068 #undef __FUNCT__ 1069 #define __FUNCT__ "MatMultTransposeAdd_MPISBAIJ" 1070 PetscErrorCode MatMultTransposeAdd_MPISBAIJ(Mat A,Vec xx,Vec yy,Vec zz) 1071 { 1072 PetscErrorCode ierr; 1073 1074 PetscFunctionBegin; 1075 ierr = MatMultAdd(A,xx,yy,zz);CHKERRQ(ierr); 1076 PetscFunctionReturn(0); 1077 } 1078 1079 /* 1080 This only works correctly for square matrices where the subblock A->A is the 1081 diagonal block 1082 */ 1083 #undef __FUNCT__ 1084 #define __FUNCT__ "MatGetDiagonal_MPISBAIJ" 1085 PetscErrorCode MatGetDiagonal_MPISBAIJ(Mat A,Vec v) 1086 { 1087 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1088 PetscErrorCode ierr; 1089 1090 PetscFunctionBegin; 1091 /* if (a->M != a->N) SETERRQ(PETSC_ERR_SUP,"Supports only square matrix where A->A is diag block"); */ 1092 ierr = MatGetDiagonal(a->A,v);CHKERRQ(ierr); 1093 PetscFunctionReturn(0); 1094 } 1095 1096 #undef __FUNCT__ 1097 #define __FUNCT__ "MatScale_MPISBAIJ" 1098 PetscErrorCode MatScale_MPISBAIJ(const PetscScalar *aa,Mat A) 1099 { 1100 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1101 PetscErrorCode ierr; 1102 1103 PetscFunctionBegin; 1104 ierr = MatScale(aa,a->A);CHKERRQ(ierr); 1105 ierr = MatScale(aa,a->B);CHKERRQ(ierr); 1106 PetscFunctionReturn(0); 1107 } 1108 1109 #undef __FUNCT__ 1110 #define __FUNCT__ "MatGetRow_MPISBAIJ" 1111 PetscErrorCode MatGetRow_MPISBAIJ(Mat matin,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 1112 { 1113 PetscFunctionBegin; 1114 SETERRQ(PETSC_ERR_SUP,"MatGetRow is not supported for SBAIJ matrix format"); 1115 PetscFunctionReturn(0); 1116 } 1117 1118 #undef __FUNCT__ 1119 #define __FUNCT__ "MatRestoreRow_MPISBAIJ" 1120 PetscErrorCode MatRestoreRow_MPISBAIJ(Mat mat,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 1121 { 1122 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 1123 1124 PetscFunctionBegin; 1125 if (!baij->getrowactive) { 1126 SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"MatGetRow() must be called first"); 1127 } 1128 baij->getrowactive = PETSC_FALSE; 1129 PetscFunctionReturn(0); 1130 } 1131 1132 #undef __FUNCT__ 1133 #define __FUNCT__ "MatZeroEntries_MPISBAIJ" 1134 PetscErrorCode MatZeroEntries_MPISBAIJ(Mat A) 1135 { 1136 Mat_MPISBAIJ *l = (Mat_MPISBAIJ*)A->data; 1137 PetscErrorCode ierr; 1138 1139 PetscFunctionBegin; 1140 ierr = MatZeroEntries(l->A);CHKERRQ(ierr); 1141 ierr = MatZeroEntries(l->B);CHKERRQ(ierr); 1142 PetscFunctionReturn(0); 1143 } 1144 1145 #undef __FUNCT__ 1146 #define __FUNCT__ "MatGetInfo_MPISBAIJ" 1147 PetscErrorCode MatGetInfo_MPISBAIJ(Mat matin,MatInfoType flag,MatInfo *info) 1148 { 1149 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)matin->data; 1150 Mat A = a->A,B = a->B; 1151 PetscErrorCode ierr; 1152 PetscReal isend[5],irecv[5]; 1153 1154 PetscFunctionBegin; 1155 info->block_size = (PetscReal)matin->bs; 1156 ierr = MatGetInfo(A,MAT_LOCAL,info);CHKERRQ(ierr); 1157 isend[0] = info->nz_used; isend[1] = info->nz_allocated; isend[2] = info->nz_unneeded; 1158 isend[3] = info->memory; isend[4] = info->mallocs; 1159 ierr = MatGetInfo(B,MAT_LOCAL,info);CHKERRQ(ierr); 1160 isend[0] += info->nz_used; isend[1] += info->nz_allocated; isend[2] += info->nz_unneeded; 1161 isend[3] += info->memory; isend[4] += info->mallocs; 1162 if (flag == MAT_LOCAL) { 1163 info->nz_used = isend[0]; 1164 info->nz_allocated = isend[1]; 1165 info->nz_unneeded = isend[2]; 1166 info->memory = isend[3]; 1167 info->mallocs = isend[4]; 1168 } else if (flag == MAT_GLOBAL_MAX) { 1169 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPI_MAX,matin->comm);CHKERRQ(ierr); 1170 info->nz_used = irecv[0]; 1171 info->nz_allocated = irecv[1]; 1172 info->nz_unneeded = irecv[2]; 1173 info->memory = irecv[3]; 1174 info->mallocs = irecv[4]; 1175 } else if (flag == MAT_GLOBAL_SUM) { 1176 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPI_SUM,matin->comm);CHKERRQ(ierr); 1177 info->nz_used = irecv[0]; 1178 info->nz_allocated = irecv[1]; 1179 info->nz_unneeded = irecv[2]; 1180 info->memory = irecv[3]; 1181 info->mallocs = irecv[4]; 1182 } else { 1183 SETERRQ1(PETSC_ERR_ARG_WRONG,"Unknown MatInfoType argument %d",(int)flag); 1184 } 1185 info->rows_global = (PetscReal)A->M; 1186 info->columns_global = (PetscReal)A->N; 1187 info->rows_local = (PetscReal)A->m; 1188 info->columns_local = (PetscReal)A->N; 1189 info->fill_ratio_given = 0; /* no parallel LU/ILU/Cholesky */ 1190 info->fill_ratio_needed = 0; 1191 info->factor_mallocs = 0; 1192 PetscFunctionReturn(0); 1193 } 1194 1195 #undef __FUNCT__ 1196 #define __FUNCT__ "MatSetOption_MPISBAIJ" 1197 PetscErrorCode MatSetOption_MPISBAIJ(Mat A,MatOption op) 1198 { 1199 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1200 PetscErrorCode ierr; 1201 1202 PetscFunctionBegin; 1203 switch (op) { 1204 case MAT_NO_NEW_NONZERO_LOCATIONS: 1205 case MAT_YES_NEW_NONZERO_LOCATIONS: 1206 case MAT_COLUMNS_UNSORTED: 1207 case MAT_COLUMNS_SORTED: 1208 case MAT_NEW_NONZERO_ALLOCATION_ERR: 1209 case MAT_KEEP_ZEROED_ROWS: 1210 case MAT_NEW_NONZERO_LOCATION_ERR: 1211 ierr = MatSetOption(a->A,op);CHKERRQ(ierr); 1212 ierr = MatSetOption(a->B,op);CHKERRQ(ierr); 1213 break; 1214 case MAT_ROW_ORIENTED: 1215 a->roworiented = PETSC_TRUE; 1216 ierr = MatSetOption(a->A,op);CHKERRQ(ierr); 1217 ierr = MatSetOption(a->B,op);CHKERRQ(ierr); 1218 break; 1219 case MAT_ROWS_SORTED: 1220 case MAT_ROWS_UNSORTED: 1221 case MAT_YES_NEW_DIAGONALS: 1222 ierr = PetscLogInfo((A,"Info:MatSetOption_MPIBAIJ:Option ignored\n"));CHKERRQ(ierr); 1223 break; 1224 case MAT_COLUMN_ORIENTED: 1225 a->roworiented = PETSC_FALSE; 1226 ierr = MatSetOption(a->A,op);CHKERRQ(ierr); 1227 ierr = MatSetOption(a->B,op);CHKERRQ(ierr); 1228 break; 1229 case MAT_IGNORE_OFF_PROC_ENTRIES: 1230 a->donotstash = PETSC_TRUE; 1231 break; 1232 case MAT_NO_NEW_DIAGONALS: 1233 SETERRQ(PETSC_ERR_SUP,"MAT_NO_NEW_DIAGONALS"); 1234 case MAT_USE_HASH_TABLE: 1235 a->ht_flag = PETSC_TRUE; 1236 break; 1237 case MAT_NOT_SYMMETRIC: 1238 case MAT_NOT_STRUCTURALLY_SYMMETRIC: 1239 case MAT_HERMITIAN: 1240 SETERRQ(PETSC_ERR_SUP,"Matrix must be symmetric"); 1241 case MAT_SYMMETRIC: 1242 case MAT_STRUCTURALLY_SYMMETRIC: 1243 case MAT_NOT_HERMITIAN: 1244 case MAT_SYMMETRY_ETERNAL: 1245 case MAT_NOT_SYMMETRY_ETERNAL: 1246 break; 1247 default: 1248 SETERRQ(PETSC_ERR_SUP,"unknown option"); 1249 } 1250 PetscFunctionReturn(0); 1251 } 1252 1253 #undef __FUNCT__ 1254 #define __FUNCT__ "MatTranspose_MPISBAIJ" 1255 PetscErrorCode MatTranspose_MPISBAIJ(Mat A,Mat *B) 1256 { 1257 PetscErrorCode ierr; 1258 PetscFunctionBegin; 1259 ierr = MatDuplicate(A,MAT_COPY_VALUES,B);CHKERRQ(ierr); 1260 PetscFunctionReturn(0); 1261 } 1262 1263 #undef __FUNCT__ 1264 #define __FUNCT__ "MatDiagonalScale_MPISBAIJ" 1265 PetscErrorCode MatDiagonalScale_MPISBAIJ(Mat mat,Vec ll,Vec rr) 1266 { 1267 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 1268 Mat a=baij->A, b=baij->B; 1269 PetscErrorCode ierr; 1270 PetscInt nv,m,n; 1271 PetscTruth flg; 1272 1273 PetscFunctionBegin; 1274 if (ll != rr){ 1275 ierr = VecEqual(ll,rr,&flg);CHKERRQ(ierr); 1276 if (!flg) 1277 SETERRQ(PETSC_ERR_ARG_OUTOFRANGE,"For symmetric format, left and right scaling vectors must be same\n"); 1278 } 1279 if (!ll) PetscFunctionReturn(0); 1280 1281 ierr = MatGetLocalSize(mat,&m,&n);CHKERRQ(ierr); 1282 if (m != n) SETERRQ2(PETSC_ERR_ARG_SIZ,"For symmetric format, local size %d %d must be same",m,n); 1283 1284 ierr = VecGetLocalSize(rr,&nv);CHKERRQ(ierr); 1285 if (nv!=n) SETERRQ(PETSC_ERR_ARG_SIZ,"Left and right vector non-conforming local size"); 1286 1287 ierr = VecScatterBegin(rr,baij->lvec,INSERT_VALUES,SCATTER_FORWARD,baij->Mvctx);CHKERRQ(ierr); 1288 1289 /* left diagonalscale the off-diagonal part */ 1290 ierr = (*b->ops->diagonalscale)(b,ll,PETSC_NULL);CHKERRQ(ierr); 1291 1292 /* scale the diagonal part */ 1293 ierr = (*a->ops->diagonalscale)(a,ll,rr);CHKERRQ(ierr); 1294 1295 /* right diagonalscale the off-diagonal part */ 1296 ierr = VecScatterEnd(rr,baij->lvec,INSERT_VALUES,SCATTER_FORWARD,baij->Mvctx);CHKERRQ(ierr); 1297 ierr = (*b->ops->diagonalscale)(b,PETSC_NULL,baij->lvec);CHKERRQ(ierr); 1298 PetscFunctionReturn(0); 1299 } 1300 1301 #undef __FUNCT__ 1302 #define __FUNCT__ "MatPrintHelp_MPISBAIJ" 1303 PetscErrorCode MatPrintHelp_MPISBAIJ(Mat A) 1304 { 1305 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1306 MPI_Comm comm = A->comm; 1307 static PetscTruth called = PETSC_FALSE; 1308 PetscErrorCode ierr; 1309 1310 PetscFunctionBegin; 1311 if (!a->rank) { 1312 ierr = MatPrintHelp_SeqSBAIJ(a->A);CHKERRQ(ierr); 1313 } 1314 if (called) {PetscFunctionReturn(0);} else called = PETSC_TRUE; 1315 ierr = (*PetscHelpPrintf)(comm," Options for MATMPISBAIJ matrix format (the defaults):\n");CHKERRQ(ierr); 1316 ierr = (*PetscHelpPrintf)(comm," -mat_use_hash_table <factor>: Use hashtable for efficient matrix assembly\n");CHKERRQ(ierr); 1317 PetscFunctionReturn(0); 1318 } 1319 1320 #undef __FUNCT__ 1321 #define __FUNCT__ "MatSetUnfactored_MPISBAIJ" 1322 PetscErrorCode MatSetUnfactored_MPISBAIJ(Mat A) 1323 { 1324 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1325 PetscErrorCode ierr; 1326 1327 PetscFunctionBegin; 1328 ierr = MatSetUnfactored(a->A);CHKERRQ(ierr); 1329 PetscFunctionReturn(0); 1330 } 1331 1332 static PetscErrorCode MatDuplicate_MPISBAIJ(Mat,MatDuplicateOption,Mat *); 1333 1334 #undef __FUNCT__ 1335 #define __FUNCT__ "MatEqual_MPISBAIJ" 1336 PetscErrorCode MatEqual_MPISBAIJ(Mat A,Mat B,PetscTruth *flag) 1337 { 1338 Mat_MPISBAIJ *matB = (Mat_MPISBAIJ*)B->data,*matA = (Mat_MPISBAIJ*)A->data; 1339 Mat a,b,c,d; 1340 PetscTruth flg; 1341 PetscErrorCode ierr; 1342 1343 PetscFunctionBegin; 1344 a = matA->A; b = matA->B; 1345 c = matB->A; d = matB->B; 1346 1347 ierr = MatEqual(a,c,&flg);CHKERRQ(ierr); 1348 if (flg) { 1349 ierr = MatEqual(b,d,&flg);CHKERRQ(ierr); 1350 } 1351 ierr = MPI_Allreduce(&flg,flag,1,MPI_INT,MPI_LAND,A->comm);CHKERRQ(ierr); 1352 PetscFunctionReturn(0); 1353 } 1354 1355 #undef __FUNCT__ 1356 #define __FUNCT__ "MatSetUpPreallocation_MPISBAIJ" 1357 PetscErrorCode MatSetUpPreallocation_MPISBAIJ(Mat A) 1358 { 1359 PetscErrorCode ierr; 1360 1361 PetscFunctionBegin; 1362 ierr = MatMPISBAIJSetPreallocation(A,1,PETSC_DEFAULT,0,PETSC_DEFAULT,0);CHKERRQ(ierr); 1363 PetscFunctionReturn(0); 1364 } 1365 1366 #undef __FUNCT__ 1367 #define __FUNCT__ "MatGetSubMatrices_MPISBAIJ" 1368 PetscErrorCode MatGetSubMatrices_MPISBAIJ(Mat A,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *B[]) 1369 { 1370 PetscErrorCode ierr; 1371 PetscInt i; 1372 PetscTruth flg; 1373 1374 PetscFunctionBegin; 1375 for (i=0; i<n; i++) { 1376 ierr = ISEqual(irow[i],icol[i],&flg);CHKERRQ(ierr); 1377 if (!flg) { 1378 SETERRQ(PETSC_ERR_SUP,"Can only get symmetric submatrix for MPISBAIJ matrices"); 1379 } 1380 } 1381 ierr = MatGetSubMatrices_MPIBAIJ(A,n,irow,icol,scall,B);CHKERRQ(ierr); 1382 PetscFunctionReturn(0); 1383 } 1384 1385 1386 /* -------------------------------------------------------------------*/ 1387 static struct _MatOps MatOps_Values = { 1388 MatSetValues_MPISBAIJ, 1389 MatGetRow_MPISBAIJ, 1390 MatRestoreRow_MPISBAIJ, 1391 MatMult_MPISBAIJ, 1392 /* 4*/ MatMultAdd_MPISBAIJ, 1393 MatMultTranspose_MPISBAIJ, 1394 MatMultTransposeAdd_MPISBAIJ, 1395 0, 1396 0, 1397 0, 1398 /*10*/ 0, 1399 0, 1400 0, 1401 MatRelax_MPISBAIJ, 1402 MatTranspose_MPISBAIJ, 1403 /*15*/ MatGetInfo_MPISBAIJ, 1404 MatEqual_MPISBAIJ, 1405 MatGetDiagonal_MPISBAIJ, 1406 MatDiagonalScale_MPISBAIJ, 1407 MatNorm_MPISBAIJ, 1408 /*20*/ MatAssemblyBegin_MPISBAIJ, 1409 MatAssemblyEnd_MPISBAIJ, 1410 0, 1411 MatSetOption_MPISBAIJ, 1412 MatZeroEntries_MPISBAIJ, 1413 /*25*/ 0, 1414 0, 1415 0, 1416 0, 1417 0, 1418 /*30*/ MatSetUpPreallocation_MPISBAIJ, 1419 0, 1420 0, 1421 0, 1422 0, 1423 /*35*/ MatDuplicate_MPISBAIJ, 1424 0, 1425 0, 1426 0, 1427 0, 1428 /*40*/ 0, 1429 MatGetSubMatrices_MPISBAIJ, 1430 MatIncreaseOverlap_MPISBAIJ, 1431 MatGetValues_MPISBAIJ, 1432 0, 1433 /*45*/ MatPrintHelp_MPISBAIJ, 1434 MatScale_MPISBAIJ, 1435 0, 1436 0, 1437 0, 1438 /*50*/ 0, 1439 0, 1440 0, 1441 0, 1442 0, 1443 /*55*/ 0, 1444 0, 1445 MatSetUnfactored_MPISBAIJ, 1446 0, 1447 MatSetValuesBlocked_MPISBAIJ, 1448 /*60*/ 0, 1449 0, 1450 0, 1451 MatGetPetscMaps_Petsc, 1452 0, 1453 /*65*/ 0, 1454 0, 1455 0, 1456 0, 1457 0, 1458 /*70*/ MatGetRowMax_MPISBAIJ, 1459 0, 1460 0, 1461 0, 1462 0, 1463 /*75*/ 0, 1464 0, 1465 0, 1466 0, 1467 0, 1468 /*80*/ 0, 1469 0, 1470 0, 1471 0, 1472 MatLoad_MPISBAIJ, 1473 /*85*/ 0, 1474 0, 1475 0, 1476 0, 1477 0, 1478 /*90*/ 0, 1479 0, 1480 0, 1481 0, 1482 0, 1483 /*95*/ 0, 1484 0, 1485 0, 1486 0}; 1487 1488 1489 EXTERN_C_BEGIN 1490 #undef __FUNCT__ 1491 #define __FUNCT__ "MatGetDiagonalBlock_MPISBAIJ" 1492 PetscErrorCode MatGetDiagonalBlock_MPISBAIJ(Mat A,PetscTruth *iscopy,MatReuse reuse,Mat *a) 1493 { 1494 PetscFunctionBegin; 1495 *a = ((Mat_MPISBAIJ *)A->data)->A; 1496 *iscopy = PETSC_FALSE; 1497 PetscFunctionReturn(0); 1498 } 1499 EXTERN_C_END 1500 1501 EXTERN_C_BEGIN 1502 #undef __FUNCT__ 1503 #define __FUNCT__ "MatMPISBAIJSetPreallocation_MPISBAIJ" 1504 PetscErrorCode MatMPISBAIJSetPreallocation_MPISBAIJ(Mat B,PetscInt bs,PetscInt d_nz,PetscInt *d_nnz,PetscInt o_nz,PetscInt *o_nnz) 1505 { 1506 Mat_MPISBAIJ *b; 1507 PetscErrorCode ierr; 1508 PetscInt i,mbs,Mbs; 1509 1510 PetscFunctionBegin; 1511 ierr = PetscOptionsGetInt(B->prefix,"-mat_block_size",&bs,PETSC_NULL);CHKERRQ(ierr); 1512 1513 if (bs < 1) SETERRQ(PETSC_ERR_ARG_OUTOFRANGE,"Invalid block size specified, must be positive"); 1514 if (d_nz == PETSC_DECIDE || d_nz == PETSC_DEFAULT) d_nz = 3; 1515 if (o_nz == PETSC_DECIDE || o_nz == PETSC_DEFAULT) o_nz = 1; 1516 if (d_nz < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"d_nz cannot be less than 0: value %D",d_nz); 1517 if (o_nz < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"o_nz cannot be less than 0: value %D",o_nz); 1518 if (d_nnz) { 1519 for (i=0; i<B->m/bs; i++) { 1520 if (d_nnz[i] < 0) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"d_nnz cannot be less than -1: local row %D value %D",i,d_nnz[i]); 1521 } 1522 } 1523 if (o_nnz) { 1524 for (i=0; i<B->m/bs; i++) { 1525 if (o_nnz[i] < 0) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"o_nnz cannot be less than -1: local row %D value %D",i,o_nnz[i]); 1526 } 1527 } 1528 B->preallocated = PETSC_TRUE; 1529 ierr = PetscSplitOwnershipBlock(B->comm,bs,&B->m,&B->M);CHKERRQ(ierr); 1530 ierr = PetscSplitOwnershipBlock(B->comm,bs,&B->n,&B->N);CHKERRQ(ierr); 1531 ierr = PetscMapCreateMPI(B->comm,B->m,B->M,&B->rmap);CHKERRQ(ierr); 1532 ierr = PetscMapCreateMPI(B->comm,B->m,B->M,&B->cmap);CHKERRQ(ierr); 1533 1534 b = (Mat_MPISBAIJ*)B->data; 1535 mbs = B->m/bs; 1536 Mbs = B->M/bs; 1537 if (mbs*bs != B->m) { 1538 SETERRQ2(PETSC_ERR_ARG_SIZ,"No of local rows %D must be divisible by blocksize %D",B->m,bs); 1539 } 1540 1541 B->bs = bs; 1542 b->bs2 = bs*bs; 1543 b->mbs = mbs; 1544 b->nbs = mbs; 1545 b->Mbs = Mbs; 1546 b->Nbs = Mbs; 1547 1548 ierr = MPI_Allgather(&b->mbs,1,MPIU_INT,b->rowners+1,1,MPIU_INT,B->comm);CHKERRQ(ierr); 1549 b->rowners[0] = 0; 1550 for (i=2; i<=b->size; i++) { 1551 b->rowners[i] += b->rowners[i-1]; 1552 } 1553 b->rstart = b->rowners[b->rank]; 1554 b->rend = b->rowners[b->rank+1]; 1555 b->cstart = b->rstart; 1556 b->cend = b->rend; 1557 for (i=0; i<=b->size; i++) { 1558 b->rowners_bs[i] = b->rowners[i]*bs; 1559 } 1560 b->rstart_bs = b-> rstart*bs; 1561 b->rend_bs = b->rend*bs; 1562 1563 b->cstart_bs = b->cstart*bs; 1564 b->cend_bs = b->cend*bs; 1565 1566 ierr = MatCreate(PETSC_COMM_SELF,B->m,B->m,B->m,B->m,&b->A);CHKERRQ(ierr); 1567 ierr = MatSetType(b->A,MATSEQSBAIJ);CHKERRQ(ierr); 1568 ierr = MatSeqSBAIJSetPreallocation(b->A,bs,d_nz,d_nnz);CHKERRQ(ierr); 1569 ierr = PetscLogObjectParent(B,b->A);CHKERRQ(ierr); 1570 1571 ierr = MatCreate(PETSC_COMM_SELF,B->m,B->M,B->m,B->M,&b->B);CHKERRQ(ierr); 1572 ierr = MatSetType(b->B,MATSEQBAIJ);CHKERRQ(ierr); 1573 ierr = MatSeqBAIJSetPreallocation(b->B,bs,o_nz,o_nnz);CHKERRQ(ierr); 1574 ierr = PetscLogObjectParent(B,b->B);CHKERRQ(ierr); 1575 1576 /* build cache for off array entries formed */ 1577 ierr = MatStashCreate_Private(B->comm,bs,&B->bstash);CHKERRQ(ierr); 1578 1579 PetscFunctionReturn(0); 1580 } 1581 EXTERN_C_END 1582 1583 /*MC 1584 MATMPISBAIJ - MATMPISBAIJ = "mpisbaij" - A matrix type to be used for distributed symmetric sparse block matrices, 1585 based on block compressed sparse row format. Only the upper triangular portion of the matrix is stored. 1586 1587 Options Database Keys: 1588 . -mat_type mpisbaij - sets the matrix type to "mpisbaij" during a call to MatSetFromOptions() 1589 1590 Level: beginner 1591 1592 .seealso: MatCreateMPISBAIJ 1593 M*/ 1594 1595 EXTERN_C_BEGIN 1596 #undef __FUNCT__ 1597 #define __FUNCT__ "MatCreate_MPISBAIJ" 1598 PetscErrorCode MatCreate_MPISBAIJ(Mat B) 1599 { 1600 Mat_MPISBAIJ *b; 1601 PetscErrorCode ierr; 1602 PetscTruth flg; 1603 1604 PetscFunctionBegin; 1605 1606 ierr = PetscNew(Mat_MPISBAIJ,&b);CHKERRQ(ierr); 1607 B->data = (void*)b; 1608 ierr = PetscMemcpy(B->ops,&MatOps_Values,sizeof(struct _MatOps));CHKERRQ(ierr); 1609 1610 B->ops->destroy = MatDestroy_MPISBAIJ; 1611 B->ops->view = MatView_MPISBAIJ; 1612 B->mapping = 0; 1613 B->factor = 0; 1614 B->assembled = PETSC_FALSE; 1615 1616 B->insertmode = NOT_SET_VALUES; 1617 ierr = MPI_Comm_rank(B->comm,&b->rank);CHKERRQ(ierr); 1618 ierr = MPI_Comm_size(B->comm,&b->size);CHKERRQ(ierr); 1619 1620 /* build local table of row and column ownerships */ 1621 ierr = PetscMalloc(3*(b->size+2)*sizeof(PetscInt),&b->rowners);CHKERRQ(ierr); 1622 b->cowners = b->rowners + b->size + 2; 1623 b->rowners_bs = b->cowners + b->size + 2; 1624 ierr = PetscLogObjectMemory(B,3*(b->size+2)*sizeof(PetscInt)+sizeof(struct _p_Mat)+sizeof(Mat_MPISBAIJ));CHKERRQ(ierr); 1625 1626 /* build cache for off array entries formed */ 1627 ierr = MatStashCreate_Private(B->comm,1,&B->stash);CHKERRQ(ierr); 1628 b->donotstash = PETSC_FALSE; 1629 b->colmap = PETSC_NULL; 1630 b->garray = PETSC_NULL; 1631 b->roworiented = PETSC_TRUE; 1632 1633 #if defined(PETSC_USE_MAT_SINGLE) 1634 /* stuff for MatSetValues_XXX in single precision */ 1635 b->setvalueslen = 0; 1636 b->setvaluescopy = PETSC_NULL; 1637 #endif 1638 1639 /* stuff used in block assembly */ 1640 b->barray = 0; 1641 1642 /* stuff used for matrix vector multiply */ 1643 b->lvec = 0; 1644 b->Mvctx = 0; 1645 b->slvec0 = 0; 1646 b->slvec0b = 0; 1647 b->slvec1 = 0; 1648 b->slvec1a = 0; 1649 b->slvec1b = 0; 1650 b->sMvctx = 0; 1651 1652 /* stuff for MatGetRow() */ 1653 b->rowindices = 0; 1654 b->rowvalues = 0; 1655 b->getrowactive = PETSC_FALSE; 1656 1657 /* hash table stuff */ 1658 b->ht = 0; 1659 b->hd = 0; 1660 b->ht_size = 0; 1661 b->ht_flag = PETSC_FALSE; 1662 b->ht_fact = 0; 1663 b->ht_total_ct = 0; 1664 b->ht_insert_ct = 0; 1665 1666 ierr = PetscOptionsHasName(B->prefix,"-mat_use_hash_table",&flg);CHKERRQ(ierr); 1667 if (flg) { 1668 PetscReal fact = 1.39; 1669 ierr = MatSetOption(B,MAT_USE_HASH_TABLE);CHKERRQ(ierr); 1670 ierr = PetscOptionsGetReal(B->prefix,"-mat_use_hash_table",&fact,PETSC_NULL);CHKERRQ(ierr); 1671 if (fact <= 1.0) fact = 1.39; 1672 ierr = MatMPIBAIJSetHashTableFactor(B,fact);CHKERRQ(ierr); 1673 ierr = PetscLogInfo((0,"MatCreateMPISBAIJ:Hash table Factor used %5.2f\n",fact));CHKERRQ(ierr); 1674 } 1675 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatStoreValues_C", 1676 "MatStoreValues_MPISBAIJ", 1677 MatStoreValues_MPISBAIJ);CHKERRQ(ierr); 1678 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatRetrieveValues_C", 1679 "MatRetrieveValues_MPISBAIJ", 1680 MatRetrieveValues_MPISBAIJ);CHKERRQ(ierr); 1681 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatGetDiagonalBlock_C", 1682 "MatGetDiagonalBlock_MPISBAIJ", 1683 MatGetDiagonalBlock_MPISBAIJ);CHKERRQ(ierr); 1684 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatMPISBAIJSetPreallocation_C", 1685 "MatMPISBAIJSetPreallocation_MPISBAIJ", 1686 MatMPISBAIJSetPreallocation_MPISBAIJ);CHKERRQ(ierr); 1687 B->symmetric = PETSC_TRUE; 1688 B->structurally_symmetric = PETSC_TRUE; 1689 B->symmetric_set = PETSC_TRUE; 1690 B->structurally_symmetric_set = PETSC_TRUE; 1691 PetscFunctionReturn(0); 1692 } 1693 EXTERN_C_END 1694 1695 /*MC 1696 MATSBAIJ - MATSBAIJ = "sbaij" - A matrix type to be used for symmetric block sparse matrices. 1697 1698 This matrix type is identical to MATSEQSBAIJ when constructed with a single process communicator, 1699 and MATMPISBAIJ otherwise. 1700 1701 Options Database Keys: 1702 . -mat_type sbaij - sets the matrix type to "sbaij" during a call to MatSetFromOptions() 1703 1704 Level: beginner 1705 1706 .seealso: MatCreateMPISBAIJ,MATSEQSBAIJ,MATMPISBAIJ 1707 M*/ 1708 1709 EXTERN_C_BEGIN 1710 #undef __FUNCT__ 1711 #define __FUNCT__ "MatCreate_SBAIJ" 1712 PetscErrorCode MatCreate_SBAIJ(Mat A) 1713 { 1714 PetscErrorCode ierr; 1715 PetscMPIInt size; 1716 1717 PetscFunctionBegin; 1718 ierr = PetscObjectChangeTypeName((PetscObject)A,MATSBAIJ);CHKERRQ(ierr); 1719 ierr = MPI_Comm_size(A->comm,&size);CHKERRQ(ierr); 1720 if (size == 1) { 1721 ierr = MatSetType(A,MATSEQSBAIJ);CHKERRQ(ierr); 1722 } else { 1723 ierr = MatSetType(A,MATMPISBAIJ);CHKERRQ(ierr); 1724 } 1725 PetscFunctionReturn(0); 1726 } 1727 EXTERN_C_END 1728 1729 #undef __FUNCT__ 1730 #define __FUNCT__ "MatMPISBAIJSetPreallocation" 1731 /*@C 1732 MatMPISBAIJSetPreallocation - For good matrix assembly performance 1733 the user should preallocate the matrix storage by setting the parameters 1734 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 1735 performance can be increased by more than a factor of 50. 1736 1737 Collective on Mat 1738 1739 Input Parameters: 1740 + A - the matrix 1741 . bs - size of blockk 1742 . d_nz - number of block nonzeros per block row in diagonal portion of local 1743 submatrix (same for all local rows) 1744 . d_nnz - array containing the number of block nonzeros in the various block rows 1745 in the upper triangular and diagonal part of the in diagonal portion of the local 1746 (possibly different for each block row) or PETSC_NULL. You must leave room 1747 for the diagonal entry even if it is zero. 1748 . o_nz - number of block nonzeros per block row in the off-diagonal portion of local 1749 submatrix (same for all local rows). 1750 - o_nnz - array containing the number of nonzeros in the various block rows of the 1751 off-diagonal portion of the local submatrix (possibly different for 1752 each block row) or PETSC_NULL. 1753 1754 1755 Options Database Keys: 1756 . -mat_no_unroll - uses code that does not unroll the loops in the 1757 block calculations (much slower) 1758 . -mat_block_size - size of the blocks to use 1759 1760 Notes: 1761 1762 If PETSC_DECIDE or PETSC_DETERMINE is used for a particular argument on one processor 1763 than it must be used on all processors that share the object for that argument. 1764 1765 If the *_nnz parameter is given then the *_nz parameter is ignored 1766 1767 Storage Information: 1768 For a square global matrix we define each processor's diagonal portion 1769 to be its local rows and the corresponding columns (a square submatrix); 1770 each processor's off-diagonal portion encompasses the remainder of the 1771 local matrix (a rectangular submatrix). 1772 1773 The user can specify preallocated storage for the diagonal part of 1774 the local submatrix with either d_nz or d_nnz (not both). Set 1775 d_nz=PETSC_DEFAULT and d_nnz=PETSC_NULL for PETSc to control dynamic 1776 memory allocation. Likewise, specify preallocated storage for the 1777 off-diagonal part of the local submatrix with o_nz or o_nnz (not both). 1778 1779 Consider a processor that owns rows 3, 4 and 5 of a parallel matrix. In 1780 the figure below we depict these three local rows and all columns (0-11). 1781 1782 .vb 1783 0 1 2 3 4 5 6 7 8 9 10 11 1784 ------------------- 1785 row 3 | o o o d d d o o o o o o 1786 row 4 | o o o d d d o o o o o o 1787 row 5 | o o o d d d o o o o o o 1788 ------------------- 1789 .ve 1790 1791 Thus, any entries in the d locations are stored in the d (diagonal) 1792 submatrix, and any entries in the o locations are stored in the 1793 o (off-diagonal) submatrix. Note that the d matrix is stored in 1794 MatSeqSBAIJ format and the o submatrix in MATSEQBAIJ format. 1795 1796 Now d_nz should indicate the number of block nonzeros per row in the upper triangular 1797 plus the diagonal part of the d matrix, 1798 and o_nz should indicate the number of block nonzeros per row in the o matrix. 1799 In general, for PDE problems in which most nonzeros are near the diagonal, 1800 one expects d_nz >> o_nz. For large problems you MUST preallocate memory 1801 or you will get TERRIBLE performance; see the users' manual chapter on 1802 matrices. 1803 1804 Level: intermediate 1805 1806 .keywords: matrix, block, aij, compressed row, sparse, parallel 1807 1808 .seealso: MatCreate(), MatCreateSeqSBAIJ(), MatSetValues(), MatCreateMPIBAIJ() 1809 @*/ 1810 PetscErrorCode MatMPISBAIJSetPreallocation(Mat B,PetscInt bs,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[]) 1811 { 1812 PetscErrorCode ierr,(*f)(Mat,PetscInt,PetscInt,const PetscInt[],PetscInt,const PetscInt[]); 1813 1814 PetscFunctionBegin; 1815 ierr = PetscObjectQueryFunction((PetscObject)B,"MatMPISBAIJSetPreallocation_C",(void (**)(void))&f);CHKERRQ(ierr); 1816 if (f) { 1817 ierr = (*f)(B,bs,d_nz,d_nnz,o_nz,o_nnz);CHKERRQ(ierr); 1818 } 1819 PetscFunctionReturn(0); 1820 } 1821 1822 #undef __FUNCT__ 1823 #define __FUNCT__ "MatCreateMPISBAIJ" 1824 /*@C 1825 MatCreateMPISBAIJ - Creates a sparse parallel matrix in symmetric block AIJ format 1826 (block compressed row). For good matrix assembly performance 1827 the user should preallocate the matrix storage by setting the parameters 1828 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 1829 performance can be increased by more than a factor of 50. 1830 1831 Collective on MPI_Comm 1832 1833 Input Parameters: 1834 + comm - MPI communicator 1835 . bs - size of blockk 1836 . m - number of local rows (or PETSC_DECIDE to have calculated if M is given) 1837 This value should be the same as the local size used in creating the 1838 y vector for the matrix-vector product y = Ax. 1839 . n - number of local columns (or PETSC_DECIDE to have calculated if N is given) 1840 This value should be the same as the local size used in creating the 1841 x vector for the matrix-vector product y = Ax. 1842 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 1843 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 1844 . d_nz - number of block nonzeros per block row in diagonal portion of local 1845 submatrix (same for all local rows) 1846 . d_nnz - array containing the number of block nonzeros in the various block rows 1847 in the upper triangular portion of the in diagonal portion of the local 1848 (possibly different for each block block row) or PETSC_NULL. 1849 You must leave room for the diagonal entry even if it is zero. 1850 . o_nz - number of block nonzeros per block row in the off-diagonal portion of local 1851 submatrix (same for all local rows). 1852 - o_nnz - array containing the number of nonzeros in the various block rows of the 1853 off-diagonal portion of the local submatrix (possibly different for 1854 each block row) or PETSC_NULL. 1855 1856 Output Parameter: 1857 . A - the matrix 1858 1859 Options Database Keys: 1860 . -mat_no_unroll - uses code that does not unroll the loops in the 1861 block calculations (much slower) 1862 . -mat_block_size - size of the blocks to use 1863 . -mat_mpi - use the parallel matrix data structures even on one processor 1864 (defaults to using SeqBAIJ format on one processor) 1865 1866 Notes: 1867 The number of rows and columns must be divisible by blocksize. 1868 1869 The user MUST specify either the local or global matrix dimensions 1870 (possibly both). 1871 1872 If PETSC_DECIDE or PETSC_DETERMINE is used for a particular argument on one processor 1873 than it must be used on all processors that share the object for that argument. 1874 1875 If the *_nnz parameter is given then the *_nz parameter is ignored 1876 1877 Storage Information: 1878 For a square global matrix we define each processor's diagonal portion 1879 to be its local rows and the corresponding columns (a square submatrix); 1880 each processor's off-diagonal portion encompasses the remainder of the 1881 local matrix (a rectangular submatrix). 1882 1883 The user can specify preallocated storage for the diagonal part of 1884 the local submatrix with either d_nz or d_nnz (not both). Set 1885 d_nz=PETSC_DEFAULT and d_nnz=PETSC_NULL for PETSc to control dynamic 1886 memory allocation. Likewise, specify preallocated storage for the 1887 off-diagonal part of the local submatrix with o_nz or o_nnz (not both). 1888 1889 Consider a processor that owns rows 3, 4 and 5 of a parallel matrix. In 1890 the figure below we depict these three local rows and all columns (0-11). 1891 1892 .vb 1893 0 1 2 3 4 5 6 7 8 9 10 11 1894 ------------------- 1895 row 3 | o o o d d d o o o o o o 1896 row 4 | o o o d d d o o o o o o 1897 row 5 | o o o d d d o o o o o o 1898 ------------------- 1899 .ve 1900 1901 Thus, any entries in the d locations are stored in the d (diagonal) 1902 submatrix, and any entries in the o locations are stored in the 1903 o (off-diagonal) submatrix. Note that the d matrix is stored in 1904 MatSeqSBAIJ format and the o submatrix in MATSEQBAIJ format. 1905 1906 Now d_nz should indicate the number of block nonzeros per row in the upper triangular 1907 plus the diagonal part of the d matrix, 1908 and o_nz should indicate the number of block nonzeros per row in the o matrix. 1909 In general, for PDE problems in which most nonzeros are near the diagonal, 1910 one expects d_nz >> o_nz. For large problems you MUST preallocate memory 1911 or you will get TERRIBLE performance; see the users' manual chapter on 1912 matrices. 1913 1914 Level: intermediate 1915 1916 .keywords: matrix, block, aij, compressed row, sparse, parallel 1917 1918 .seealso: MatCreate(), MatCreateSeqSBAIJ(), MatSetValues(), MatCreateMPIBAIJ() 1919 @*/ 1920 1921 PetscErrorCode MatCreateMPISBAIJ(MPI_Comm comm,PetscInt bs,PetscInt m,PetscInt n,PetscInt M,PetscInt N,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[],Mat *A) 1922 { 1923 PetscErrorCode ierr; 1924 PetscMPIInt size; 1925 1926 PetscFunctionBegin; 1927 ierr = MatCreate(comm,m,n,M,N,A);CHKERRQ(ierr); 1928 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 1929 if (size > 1) { 1930 ierr = MatSetType(*A,MATMPISBAIJ);CHKERRQ(ierr); 1931 ierr = MatMPISBAIJSetPreallocation(*A,bs,d_nz,d_nnz,o_nz,o_nnz);CHKERRQ(ierr); 1932 } else { 1933 ierr = MatSetType(*A,MATSEQSBAIJ);CHKERRQ(ierr); 1934 ierr = MatSeqSBAIJSetPreallocation(*A,bs,d_nz,d_nnz);CHKERRQ(ierr); 1935 } 1936 PetscFunctionReturn(0); 1937 } 1938 1939 1940 #undef __FUNCT__ 1941 #define __FUNCT__ "MatDuplicate_MPISBAIJ" 1942 static PetscErrorCode MatDuplicate_MPISBAIJ(Mat matin,MatDuplicateOption cpvalues,Mat *newmat) 1943 { 1944 Mat mat; 1945 Mat_MPISBAIJ *a,*oldmat = (Mat_MPISBAIJ*)matin->data; 1946 PetscErrorCode ierr; 1947 PetscInt len=0,nt,bs=matin->bs,mbs=oldmat->mbs; 1948 PetscScalar *array; 1949 1950 PetscFunctionBegin; 1951 *newmat = 0; 1952 ierr = MatCreate(matin->comm,matin->m,matin->n,matin->M,matin->N,&mat);CHKERRQ(ierr); 1953 ierr = MatSetType(mat,matin->type_name);CHKERRQ(ierr); 1954 ierr = PetscMemcpy(mat->ops,matin->ops,sizeof(struct _MatOps));CHKERRQ(ierr); 1955 1956 mat->factor = matin->factor; 1957 mat->preallocated = PETSC_TRUE; 1958 mat->assembled = PETSC_TRUE; 1959 mat->insertmode = NOT_SET_VALUES; 1960 1961 a = (Mat_MPISBAIJ*)mat->data; 1962 mat->bs = matin->bs; 1963 a->bs2 = oldmat->bs2; 1964 a->mbs = oldmat->mbs; 1965 a->nbs = oldmat->nbs; 1966 a->Mbs = oldmat->Mbs; 1967 a->Nbs = oldmat->Nbs; 1968 1969 a->rstart = oldmat->rstart; 1970 a->rend = oldmat->rend; 1971 a->cstart = oldmat->cstart; 1972 a->cend = oldmat->cend; 1973 a->size = oldmat->size; 1974 a->rank = oldmat->rank; 1975 a->donotstash = oldmat->donotstash; 1976 a->roworiented = oldmat->roworiented; 1977 a->rowindices = 0; 1978 a->rowvalues = 0; 1979 a->getrowactive = PETSC_FALSE; 1980 a->barray = 0; 1981 a->rstart_bs = oldmat->rstart_bs; 1982 a->rend_bs = oldmat->rend_bs; 1983 a->cstart_bs = oldmat->cstart_bs; 1984 a->cend_bs = oldmat->cend_bs; 1985 1986 /* hash table stuff */ 1987 a->ht = 0; 1988 a->hd = 0; 1989 a->ht_size = 0; 1990 a->ht_flag = oldmat->ht_flag; 1991 a->ht_fact = oldmat->ht_fact; 1992 a->ht_total_ct = 0; 1993 a->ht_insert_ct = 0; 1994 1995 ierr = PetscMemcpy(a->rowners,oldmat->rowners,3*(a->size+2)*sizeof(PetscInt));CHKERRQ(ierr); 1996 ierr = MatStashCreate_Private(matin->comm,1,&mat->stash);CHKERRQ(ierr); 1997 ierr = MatStashCreate_Private(matin->comm,matin->bs,&mat->bstash);CHKERRQ(ierr); 1998 if (oldmat->colmap) { 1999 #if defined (PETSC_USE_CTABLE) 2000 ierr = PetscTableCreateCopy(oldmat->colmap,&a->colmap);CHKERRQ(ierr); 2001 #else 2002 ierr = PetscMalloc((a->Nbs)*sizeof(PetscInt),&a->colmap);CHKERRQ(ierr); 2003 ierr = PetscLogObjectMemory(mat,(a->Nbs)*sizeof(PetscInt));CHKERRQ(ierr); 2004 ierr = PetscMemcpy(a->colmap,oldmat->colmap,(a->Nbs)*sizeof(PetscInt));CHKERRQ(ierr); 2005 #endif 2006 } else a->colmap = 0; 2007 2008 if (oldmat->garray && (len = ((Mat_SeqBAIJ*)(oldmat->B->data))->nbs)) { 2009 ierr = PetscMalloc(len*sizeof(PetscInt),&a->garray);CHKERRQ(ierr); 2010 ierr = PetscLogObjectMemory(mat,len*sizeof(PetscInt));CHKERRQ(ierr); 2011 ierr = PetscMemcpy(a->garray,oldmat->garray,len*sizeof(PetscInt));CHKERRQ(ierr); 2012 } else a->garray = 0; 2013 2014 ierr = VecDuplicate(oldmat->lvec,&a->lvec);CHKERRQ(ierr); 2015 ierr = PetscLogObjectParent(mat,a->lvec);CHKERRQ(ierr); 2016 ierr = VecScatterCopy(oldmat->Mvctx,&a->Mvctx);CHKERRQ(ierr); 2017 ierr = PetscLogObjectParent(mat,a->Mvctx);CHKERRQ(ierr); 2018 2019 ierr = VecDuplicate(oldmat->slvec0,&a->slvec0);CHKERRQ(ierr); 2020 ierr = PetscLogObjectParent(mat,a->slvec0);CHKERRQ(ierr); 2021 ierr = VecDuplicate(oldmat->slvec1,&a->slvec1);CHKERRQ(ierr); 2022 ierr = PetscLogObjectParent(mat,a->slvec1);CHKERRQ(ierr); 2023 2024 ierr = VecGetLocalSize(a->slvec1,&nt);CHKERRQ(ierr); 2025 ierr = VecGetArray(a->slvec1,&array);CHKERRQ(ierr); 2026 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,bs*mbs,array,&a->slvec1a);CHKERRQ(ierr); 2027 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,nt-bs*mbs,array+bs*mbs,&a->slvec1b);CHKERRQ(ierr); 2028 ierr = VecRestoreArray(a->slvec1,&array);CHKERRQ(ierr); 2029 ierr = VecGetArray(a->slvec0,&array);CHKERRQ(ierr); 2030 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,nt-bs*mbs,array+bs*mbs,&a->slvec0b);CHKERRQ(ierr); 2031 ierr = VecRestoreArray(a->slvec0,&array);CHKERRQ(ierr); 2032 ierr = PetscLogObjectParent(mat,a->slvec0);CHKERRQ(ierr); 2033 ierr = PetscLogObjectParent(mat,a->slvec1);CHKERRQ(ierr); 2034 ierr = PetscLogObjectParent(mat,a->slvec0b);CHKERRQ(ierr); 2035 ierr = PetscLogObjectParent(mat,a->slvec1a);CHKERRQ(ierr); 2036 ierr = PetscLogObjectParent(mat,a->slvec1b);CHKERRQ(ierr); 2037 2038 /* ierr = VecScatterCopy(oldmat->sMvctx,&a->sMvctx); - not written yet, replaced by the lazy trick: */ 2039 ierr = PetscObjectReference((PetscObject)oldmat->sMvctx);CHKERRQ(ierr); 2040 a->sMvctx = oldmat->sMvctx; 2041 ierr = PetscLogObjectParent(mat,a->sMvctx);CHKERRQ(ierr); 2042 2043 ierr = MatDuplicate(oldmat->A,cpvalues,&a->A);CHKERRQ(ierr); 2044 ierr = PetscLogObjectParent(mat,a->A);CHKERRQ(ierr); 2045 ierr = MatDuplicate(oldmat->B,cpvalues,&a->B);CHKERRQ(ierr); 2046 ierr = PetscLogObjectParent(mat,a->B);CHKERRQ(ierr); 2047 ierr = PetscFListDuplicate(mat->qlist,&matin->qlist);CHKERRQ(ierr); 2048 *newmat = mat; 2049 PetscFunctionReturn(0); 2050 } 2051 2052 #include "petscsys.h" 2053 2054 #undef __FUNCT__ 2055 #define __FUNCT__ "MatLoad_MPISBAIJ" 2056 PetscErrorCode MatLoad_MPISBAIJ(PetscViewer viewer,const MatType type,Mat *newmat) 2057 { 2058 Mat A; 2059 PetscErrorCode ierr; 2060 PetscInt i,nz,j,rstart,rend; 2061 PetscScalar *vals,*buf; 2062 MPI_Comm comm = ((PetscObject)viewer)->comm; 2063 MPI_Status status; 2064 PetscMPIInt rank,size,tag = ((PetscObject)viewer)->tag,*sndcounts = 0,*browners,maxnz,*rowners; 2065 PetscInt header[4],*rowlengths = 0,M,N,m,*cols; 2066 PetscInt *locrowlens,*procsnz = 0,jj,*mycols,*ibuf; 2067 PetscInt bs=1,Mbs,mbs,extra_rows; 2068 PetscInt *dlens,*odlens,*mask,*masked1,*masked2,rowcount,odcount; 2069 PetscInt dcount,kmax,k,nzcount,tmp; 2070 int fd; 2071 2072 PetscFunctionBegin; 2073 ierr = PetscOptionsGetInt(PETSC_NULL,"-matload_block_size",&bs,PETSC_NULL);CHKERRQ(ierr); 2074 2075 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 2076 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 2077 if (!rank) { 2078 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 2079 ierr = PetscBinaryRead(fd,(char *)header,4,PETSC_INT);CHKERRQ(ierr); 2080 if (header[0] != MAT_FILE_COOKIE) SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"not matrix object"); 2081 if (header[3] < 0) { 2082 SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"Matrix stored in special format, cannot load as MPISBAIJ"); 2083 } 2084 } 2085 2086 ierr = MPI_Bcast(header+1,3,MPIU_INT,0,comm);CHKERRQ(ierr); 2087 M = header[1]; N = header[2]; 2088 2089 if (M != N) SETERRQ(PETSC_ERR_SUP,"Can only do square matrices"); 2090 2091 /* 2092 This code adds extra rows to make sure the number of rows is 2093 divisible by the blocksize 2094 */ 2095 Mbs = M/bs; 2096 extra_rows = bs - M + bs*(Mbs); 2097 if (extra_rows == bs) extra_rows = 0; 2098 else Mbs++; 2099 if (extra_rows &&!rank) { 2100 ierr = PetscLogInfo((0,"MatLoad_MPISBAIJ:Padding loaded matrix to match blocksize\n"));CHKERRQ(ierr); 2101 } 2102 2103 /* determine ownership of all rows */ 2104 mbs = Mbs/size + ((Mbs % size) > rank); 2105 m = mbs*bs; 2106 ierr = PetscMalloc(2*(size+2)*sizeof(PetscMPIInt),&rowners);CHKERRQ(ierr); 2107 browners = rowners + size + 1; 2108 ierr = MPI_Allgather(&mbs,1,MPI_INT,rowners+1,1,MPI_INT,comm);CHKERRQ(ierr); 2109 rowners[0] = 0; 2110 for (i=2; i<=size; i++) rowners[i] += rowners[i-1]; 2111 for (i=0; i<=size; i++) browners[i] = rowners[i]*bs; 2112 rstart = rowners[rank]; 2113 rend = rowners[rank+1]; 2114 2115 /* distribute row lengths to all processors */ 2116 ierr = PetscMalloc((rend-rstart)*bs*sizeof(PetscInt),&locrowlens);CHKERRQ(ierr); 2117 if (!rank) { 2118 ierr = PetscMalloc((M+extra_rows)*sizeof(PetscInt),&rowlengths);CHKERRQ(ierr); 2119 ierr = PetscBinaryRead(fd,rowlengths,M,PETSC_INT);CHKERRQ(ierr); 2120 for (i=0; i<extra_rows; i++) rowlengths[M+i] = 1; 2121 ierr = PetscMalloc(size*sizeof(PetscMPIInt),&sndcounts);CHKERRQ(ierr); 2122 for (i=0; i<size; i++) sndcounts[i] = browners[i+1] - browners[i]; 2123 ierr = MPI_Scatterv(rowlengths,sndcounts,browners,MPIU_INT,locrowlens,(rend-rstart)*bs,MPIU_INT,0,comm);CHKERRQ(ierr); 2124 ierr = PetscFree(sndcounts);CHKERRQ(ierr); 2125 } else { 2126 ierr = MPI_Scatterv(0,0,0,MPIU_INT,locrowlens,(rend-rstart)*bs,MPIU_INT,0,comm);CHKERRQ(ierr); 2127 } 2128 2129 if (!rank) { /* procs[0] */ 2130 /* calculate the number of nonzeros on each processor */ 2131 ierr = PetscMalloc(size*sizeof(PetscInt),&procsnz);CHKERRQ(ierr); 2132 ierr = PetscMemzero(procsnz,size*sizeof(PetscInt));CHKERRQ(ierr); 2133 for (i=0; i<size; i++) { 2134 for (j=rowners[i]*bs; j< rowners[i+1]*bs; j++) { 2135 procsnz[i] += rowlengths[j]; 2136 } 2137 } 2138 ierr = PetscFree(rowlengths);CHKERRQ(ierr); 2139 2140 /* determine max buffer needed and allocate it */ 2141 maxnz = 0; 2142 for (i=0; i<size; i++) { 2143 maxnz = PetscMax(maxnz,procsnz[i]); 2144 } 2145 ierr = PetscMalloc(maxnz*sizeof(PetscInt),&cols);CHKERRQ(ierr); 2146 2147 /* read in my part of the matrix column indices */ 2148 nz = procsnz[0]; 2149 ierr = PetscMalloc(nz*sizeof(PetscInt),&ibuf);CHKERRQ(ierr); 2150 mycols = ibuf; 2151 if (size == 1) nz -= extra_rows; 2152 ierr = PetscBinaryRead(fd,mycols,nz,PETSC_INT);CHKERRQ(ierr); 2153 if (size == 1) for (i=0; i< extra_rows; i++) { mycols[nz+i] = M+i; } 2154 2155 /* read in every ones (except the last) and ship off */ 2156 for (i=1; i<size-1; i++) { 2157 nz = procsnz[i]; 2158 ierr = PetscBinaryRead(fd,cols,nz,PETSC_INT);CHKERRQ(ierr); 2159 ierr = MPI_Send(cols,nz,MPIU_INT,i,tag,comm);CHKERRQ(ierr); 2160 } 2161 /* read in the stuff for the last proc */ 2162 if (size != 1) { 2163 nz = procsnz[size-1] - extra_rows; /* the extra rows are not on the disk */ 2164 ierr = PetscBinaryRead(fd,cols,nz,PETSC_INT);CHKERRQ(ierr); 2165 for (i=0; i<extra_rows; i++) cols[nz+i] = M+i; 2166 ierr = MPI_Send(cols,nz+extra_rows,MPIU_INT,size-1,tag,comm);CHKERRQ(ierr); 2167 } 2168 ierr = PetscFree(cols);CHKERRQ(ierr); 2169 } else { /* procs[i], i>0 */ 2170 /* determine buffer space needed for message */ 2171 nz = 0; 2172 for (i=0; i<m; i++) { 2173 nz += locrowlens[i]; 2174 } 2175 ierr = PetscMalloc(nz*sizeof(PetscInt),&ibuf);CHKERRQ(ierr); 2176 mycols = ibuf; 2177 /* receive message of column indices*/ 2178 ierr = MPI_Recv(mycols,nz,MPIU_INT,0,tag,comm,&status);CHKERRQ(ierr); 2179 ierr = MPI_Get_count(&status,MPIU_INT,&maxnz);CHKERRQ(ierr); 2180 if (maxnz != nz) SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"something is wrong with file"); 2181 } 2182 2183 /* loop over local rows, determining number of off diagonal entries */ 2184 ierr = PetscMalloc(2*(rend-rstart+1)*sizeof(PetscInt),&dlens);CHKERRQ(ierr); 2185 odlens = dlens + (rend-rstart); 2186 ierr = PetscMalloc(3*Mbs*sizeof(PetscInt),&mask);CHKERRQ(ierr); 2187 ierr = PetscMemzero(mask,3*Mbs*sizeof(PetscInt));CHKERRQ(ierr); 2188 masked1 = mask + Mbs; 2189 masked2 = masked1 + Mbs; 2190 rowcount = 0; nzcount = 0; 2191 for (i=0; i<mbs; i++) { 2192 dcount = 0; 2193 odcount = 0; 2194 for (j=0; j<bs; j++) { 2195 kmax = locrowlens[rowcount]; 2196 for (k=0; k<kmax; k++) { 2197 tmp = mycols[nzcount++]/bs; /* block col. index */ 2198 if (!mask[tmp]) { 2199 mask[tmp] = 1; 2200 if (tmp < rstart || tmp >= rend) masked2[odcount++] = tmp; /* entry in off-diag portion */ 2201 else masked1[dcount++] = tmp; /* entry in diag portion */ 2202 } 2203 } 2204 rowcount++; 2205 } 2206 2207 dlens[i] = dcount; /* d_nzz[i] */ 2208 odlens[i] = odcount; /* o_nzz[i] */ 2209 2210 /* zero out the mask elements we set */ 2211 for (j=0; j<dcount; j++) mask[masked1[j]] = 0; 2212 for (j=0; j<odcount; j++) mask[masked2[j]] = 0; 2213 } 2214 2215 /* create our matrix */ 2216 ierr = MatCreate(comm,m,m,PETSC_DETERMINE,PETSC_DETERMINE,&A);CHKERRQ(ierr); 2217 ierr = MatSetType(A,type);CHKERRQ(ierr); 2218 ierr = MatMPISBAIJSetPreallocation(A,bs,0,dlens,0,odlens);CHKERRQ(ierr); 2219 ierr = MatSetOption(A,MAT_COLUMNS_SORTED);CHKERRQ(ierr); 2220 2221 if (!rank) { 2222 ierr = PetscMalloc(maxnz*sizeof(PetscScalar),&buf);CHKERRQ(ierr); 2223 /* read in my part of the matrix numerical values */ 2224 nz = procsnz[0]; 2225 vals = buf; 2226 mycols = ibuf; 2227 if (size == 1) nz -= extra_rows; 2228 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2229 if (size == 1) for (i=0; i< extra_rows; i++) { vals[nz+i] = 1.0; } 2230 2231 /* insert into matrix */ 2232 jj = rstart*bs; 2233 for (i=0; i<m; i++) { 2234 ierr = MatSetValues(A,1,&jj,locrowlens[i],mycols,vals,INSERT_VALUES);CHKERRQ(ierr); 2235 mycols += locrowlens[i]; 2236 vals += locrowlens[i]; 2237 jj++; 2238 } 2239 2240 /* read in other processors (except the last one) and ship out */ 2241 for (i=1; i<size-1; i++) { 2242 nz = procsnz[i]; 2243 vals = buf; 2244 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2245 ierr = MPI_Send(vals,nz,MPIU_SCALAR,i,A->tag,comm);CHKERRQ(ierr); 2246 } 2247 /* the last proc */ 2248 if (size != 1){ 2249 nz = procsnz[i] - extra_rows; 2250 vals = buf; 2251 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2252 for (i=0; i<extra_rows; i++) vals[nz+i] = 1.0; 2253 ierr = MPI_Send(vals,nz+extra_rows,MPIU_SCALAR,size-1,A->tag,comm);CHKERRQ(ierr); 2254 } 2255 ierr = PetscFree(procsnz);CHKERRQ(ierr); 2256 2257 } else { 2258 /* receive numeric values */ 2259 ierr = PetscMalloc(nz*sizeof(PetscScalar),&buf);CHKERRQ(ierr); 2260 2261 /* receive message of values*/ 2262 vals = buf; 2263 mycols = ibuf; 2264 ierr = MPI_Recv(vals,nz,MPIU_SCALAR,0,A->tag,comm,&status);CHKERRQ(ierr); 2265 ierr = MPI_Get_count(&status,MPIU_SCALAR,&maxnz);CHKERRQ(ierr); 2266 if (maxnz != nz) SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"something is wrong with file"); 2267 2268 /* insert into matrix */ 2269 jj = rstart*bs; 2270 for (i=0; i<m; i++) { 2271 ierr = MatSetValues_MPISBAIJ(A,1,&jj,locrowlens[i],mycols,vals,INSERT_VALUES);CHKERRQ(ierr); 2272 mycols += locrowlens[i]; 2273 vals += locrowlens[i]; 2274 jj++; 2275 } 2276 } 2277 2278 ierr = PetscFree(locrowlens);CHKERRQ(ierr); 2279 ierr = PetscFree(buf);CHKERRQ(ierr); 2280 ierr = PetscFree(ibuf);CHKERRQ(ierr); 2281 ierr = PetscFree(rowners);CHKERRQ(ierr); 2282 ierr = PetscFree(dlens);CHKERRQ(ierr); 2283 ierr = PetscFree(mask);CHKERRQ(ierr); 2284 ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2285 ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2286 *newmat = A; 2287 PetscFunctionReturn(0); 2288 } 2289 2290 #undef __FUNCT__ 2291 #define __FUNCT__ "MatMPISBAIJSetHashTableFactor" 2292 /*XXXXX@ 2293 MatMPISBAIJSetHashTableFactor - Sets the factor required to compute the size of the HashTable. 2294 2295 Input Parameters: 2296 . mat - the matrix 2297 . fact - factor 2298 2299 Collective on Mat 2300 2301 Level: advanced 2302 2303 Notes: 2304 This can also be set by the command line option: -mat_use_hash_table fact 2305 2306 .keywords: matrix, hashtable, factor, HT 2307 2308 .seealso: MatSetOption() 2309 @XXXXX*/ 2310 2311 2312 #undef __FUNCT__ 2313 #define __FUNCT__ "MatGetRowMax_MPISBAIJ" 2314 PetscErrorCode MatGetRowMax_MPISBAIJ(Mat A,Vec v) 2315 { 2316 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 2317 Mat_SeqBAIJ *b = (Mat_SeqBAIJ*)(a->B)->data; 2318 PetscReal atmp; 2319 PetscReal *work,*svalues,*rvalues; 2320 PetscErrorCode ierr; 2321 PetscInt i,bs,mbs,*bi,*bj,brow,j,ncols,krow,kcol,col,row,Mbs,bcol; 2322 PetscMPIInt rank,size; 2323 PetscInt *rowners_bs,dest,count,source; 2324 PetscScalar *va; 2325 MatScalar *ba; 2326 MPI_Status stat; 2327 2328 PetscFunctionBegin; 2329 ierr = MatGetRowMax(a->A,v);CHKERRQ(ierr); 2330 ierr = VecGetArray(v,&va);CHKERRQ(ierr); 2331 2332 ierr = MPI_Comm_size(A->comm,&size);CHKERRQ(ierr); 2333 ierr = MPI_Comm_rank(A->comm,&rank);CHKERRQ(ierr); 2334 2335 bs = A->bs; 2336 mbs = a->mbs; 2337 Mbs = a->Mbs; 2338 ba = b->a; 2339 bi = b->i; 2340 bj = b->j; 2341 2342 /* find ownerships */ 2343 rowners_bs = a->rowners_bs; 2344 2345 /* each proc creates an array to be distributed */ 2346 ierr = PetscMalloc(bs*Mbs*sizeof(PetscReal),&work);CHKERRQ(ierr); 2347 ierr = PetscMemzero(work,bs*Mbs*sizeof(PetscReal));CHKERRQ(ierr); 2348 2349 /* row_max for B */ 2350 if (rank != size-1){ 2351 for (i=0; i<mbs; i++) { 2352 ncols = bi[1] - bi[0]; bi++; 2353 brow = bs*i; 2354 for (j=0; j<ncols; j++){ 2355 bcol = bs*(*bj); 2356 for (kcol=0; kcol<bs; kcol++){ 2357 col = bcol + kcol; /* local col index */ 2358 col += rowners_bs[rank+1]; /* global col index */ 2359 for (krow=0; krow<bs; krow++){ 2360 atmp = PetscAbsScalar(*ba); ba++; 2361 row = brow + krow; /* local row index */ 2362 if (PetscRealPart(va[row]) < atmp) va[row] = atmp; 2363 if (work[col] < atmp) work[col] = atmp; 2364 } 2365 } 2366 bj++; 2367 } 2368 } 2369 2370 /* send values to its owners */ 2371 for (dest=rank+1; dest<size; dest++){ 2372 svalues = work + rowners_bs[dest]; 2373 count = rowners_bs[dest+1]-rowners_bs[dest]; 2374 ierr = MPI_Send(svalues,count,MPIU_REAL,dest,rank,A->comm);CHKERRQ(ierr); 2375 } 2376 } 2377 2378 /* receive values */ 2379 if (rank){ 2380 rvalues = work; 2381 count = rowners_bs[rank+1]-rowners_bs[rank]; 2382 for (source=0; source<rank; source++){ 2383 ierr = MPI_Recv(rvalues,count,MPIU_REAL,MPI_ANY_SOURCE,MPI_ANY_TAG,A->comm,&stat);CHKERRQ(ierr); 2384 /* process values */ 2385 for (i=0; i<count; i++){ 2386 if (PetscRealPart(va[i]) < rvalues[i]) va[i] = rvalues[i]; 2387 } 2388 } 2389 } 2390 2391 ierr = VecRestoreArray(v,&va);CHKERRQ(ierr); 2392 ierr = PetscFree(work);CHKERRQ(ierr); 2393 PetscFunctionReturn(0); 2394 } 2395 2396 #undef __FUNCT__ 2397 #define __FUNCT__ "MatRelax_MPISBAIJ" 2398 PetscErrorCode MatRelax_MPISBAIJ(Mat matin,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,PetscInt its,PetscInt lits,Vec xx) 2399 { 2400 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)matin->data; 2401 PetscErrorCode ierr; 2402 PetscInt mbs=mat->mbs,bs=matin->bs; 2403 PetscScalar mone=-1.0,*x,*b,*ptr,zero=0.0; 2404 Vec bb1; 2405 2406 PetscFunctionBegin; 2407 if (its <= 0 || lits <= 0) SETERRQ2(PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D and local its %D both positive",its,lits); 2408 if (bs > 1) 2409 SETERRQ(PETSC_ERR_SUP,"SSOR for block size > 1 is not yet implemented"); 2410 2411 if ((flag & SOR_LOCAL_SYMMETRIC_SWEEP) == SOR_LOCAL_SYMMETRIC_SWEEP){ 2412 if ( flag & SOR_ZERO_INITIAL_GUESS ) { 2413 ierr = (*mat->A->ops->relax)(mat->A,bb,omega,flag,fshift,lits,lits,xx);CHKERRQ(ierr); 2414 its--; 2415 } 2416 2417 ierr = VecDuplicate(bb,&bb1);CHKERRQ(ierr); 2418 while (its--){ 2419 2420 /* lower triangular part: slvec0b = - B^T*xx */ 2421 ierr = (*mat->B->ops->multtranspose)(mat->B,xx,mat->slvec0b);CHKERRQ(ierr); 2422 2423 /* copy xx into slvec0a */ 2424 ierr = VecGetArray(mat->slvec0,&ptr);CHKERRQ(ierr); 2425 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 2426 ierr = PetscMemcpy(ptr,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 2427 ierr = VecRestoreArray(mat->slvec0,&ptr);CHKERRQ(ierr); 2428 2429 ierr = VecScale(&mone,mat->slvec0);CHKERRQ(ierr); 2430 2431 /* copy bb into slvec1a */ 2432 ierr = VecGetArray(mat->slvec1,&ptr);CHKERRQ(ierr); 2433 ierr = VecGetArray(bb,&b);CHKERRQ(ierr); 2434 ierr = PetscMemcpy(ptr,b,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 2435 ierr = VecRestoreArray(mat->slvec1,&ptr);CHKERRQ(ierr); 2436 2437 /* set slvec1b = 0 */ 2438 ierr = VecSet(&zero,mat->slvec1b);CHKERRQ(ierr); 2439 2440 ierr = VecScatterBegin(mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD,mat->sMvctx);CHKERRQ(ierr); 2441 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 2442 ierr = VecRestoreArray(bb,&b);CHKERRQ(ierr); 2443 ierr = VecScatterEnd(mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD,mat->sMvctx);CHKERRQ(ierr); 2444 2445 /* upper triangular part: bb1 = bb1 - B*x */ 2446 ierr = (*mat->B->ops->multadd)(mat->B,mat->slvec1b,mat->slvec1a,bb1);CHKERRQ(ierr); 2447 2448 /* local diagonal sweep */ 2449 ierr = (*mat->A->ops->relax)(mat->A,bb1,omega,SOR_SYMMETRIC_SWEEP,fshift,lits,lits,xx);CHKERRQ(ierr); 2450 } 2451 ierr = VecDestroy(bb1);CHKERRQ(ierr); 2452 } else { 2453 SETERRQ(PETSC_ERR_SUP,"MatSORType is not supported for SBAIJ matrix format"); 2454 } 2455 PetscFunctionReturn(0); 2456 } 2457 2458 #undef __FUNCT__ 2459 #define __FUNCT__ "MatRelax_MPISBAIJ_2comm" 2460 PetscErrorCode MatRelax_MPISBAIJ_2comm(Mat matin,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,PetscInt its,PetscInt lits,Vec xx) 2461 { 2462 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)matin->data; 2463 PetscErrorCode ierr; 2464 PetscScalar mone=-1.0; 2465 Vec lvec1,bb1; 2466 2467 PetscFunctionBegin; 2468 if (its <= 0 || lits <= 0) SETERRQ2(PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D and local its %D both positive",its,lits); 2469 if (matin->bs > 1) 2470 SETERRQ(PETSC_ERR_SUP,"SSOR for block size > 1 is not yet implemented"); 2471 2472 if ((flag & SOR_LOCAL_SYMMETRIC_SWEEP) == SOR_LOCAL_SYMMETRIC_SWEEP){ 2473 if ( flag & SOR_ZERO_INITIAL_GUESS ) { 2474 ierr = (*mat->A->ops->relax)(mat->A,bb,omega,flag,fshift,lits,lits,xx);CHKERRQ(ierr); 2475 its--; 2476 } 2477 2478 ierr = VecDuplicate(mat->lvec,&lvec1);CHKERRQ(ierr); 2479 ierr = VecDuplicate(bb,&bb1);CHKERRQ(ierr); 2480 while (its--){ 2481 ierr = VecScatterBegin(xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD,mat->Mvctx);CHKERRQ(ierr); 2482 2483 /* lower diagonal part: bb1 = bb - B^T*xx */ 2484 ierr = (*mat->B->ops->multtranspose)(mat->B,xx,lvec1);CHKERRQ(ierr); 2485 ierr = VecScale(&mone,lvec1);CHKERRQ(ierr); 2486 2487 ierr = VecScatterEnd(xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD,mat->Mvctx);CHKERRQ(ierr); 2488 ierr = VecCopy(bb,bb1);CHKERRQ(ierr); 2489 ierr = VecScatterBegin(lvec1,bb1,ADD_VALUES,SCATTER_REVERSE,mat->Mvctx);CHKERRQ(ierr); 2490 2491 /* upper diagonal part: bb1 = bb1 - B*x */ 2492 ierr = VecScale(&mone,mat->lvec);CHKERRQ(ierr); 2493 ierr = (*mat->B->ops->multadd)(mat->B,mat->lvec,bb1,bb1);CHKERRQ(ierr); 2494 2495 ierr = VecScatterEnd(lvec1,bb1,ADD_VALUES,SCATTER_REVERSE,mat->Mvctx);CHKERRQ(ierr); 2496 2497 /* diagonal sweep */ 2498 ierr = (*mat->A->ops->relax)(mat->A,bb1,omega,SOR_SYMMETRIC_SWEEP,fshift,lits,lits,xx);CHKERRQ(ierr); 2499 } 2500 ierr = VecDestroy(lvec1);CHKERRQ(ierr); 2501 ierr = VecDestroy(bb1);CHKERRQ(ierr); 2502 } else { 2503 SETERRQ(PETSC_ERR_SUP,"MatSORType is not supported for SBAIJ matrix format"); 2504 } 2505 PetscFunctionReturn(0); 2506 } 2507 2508