1 2 #include <../src/mat/impls/baij/mpi/mpibaij.h> /*I "petscmat.h" I*/ 3 #include <../src/mat/impls/sbaij/mpi/mpisbaij.h> 4 #include <../src/mat/impls/sbaij/seq/sbaij.h> 5 #include <petscblaslapack.h> 6 7 extern PetscErrorCode MatSetUpMultiply_MPISBAIJ(Mat); 8 extern PetscErrorCode MatSetUpMultiply_MPISBAIJ_2comm(Mat); 9 extern PetscErrorCode MatDisAssemble_MPISBAIJ(Mat); 10 extern PetscErrorCode MatIncreaseOverlap_MPISBAIJ(Mat,PetscInt,IS[],PetscInt); 11 extern PetscErrorCode MatGetValues_SeqSBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],PetscScalar []); 12 extern PetscErrorCode MatGetValues_SeqBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],PetscScalar []); 13 extern PetscErrorCode MatSetValues_SeqSBAIJ(Mat,PetscInt,const PetscInt [],PetscInt,const PetscInt [],const PetscScalar [],InsertMode); 14 extern PetscErrorCode MatSetValuesBlocked_SeqSBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode); 15 extern PetscErrorCode MatSetValuesBlocked_SeqBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode); 16 extern PetscErrorCode MatGetRow_SeqSBAIJ(Mat,PetscInt,PetscInt*,PetscInt**,PetscScalar**); 17 extern PetscErrorCode MatRestoreRow_SeqSBAIJ(Mat,PetscInt,PetscInt*,PetscInt**,PetscScalar**); 18 extern PetscErrorCode MatZeroRows_SeqSBAIJ(Mat,IS,PetscScalar*,Vec,Vec); 19 extern PetscErrorCode MatZeroRows_SeqBAIJ(Mat,IS,PetscScalar*,Vec,Vec); 20 extern PetscErrorCode MatGetRowMaxAbs_MPISBAIJ(Mat,Vec,PetscInt[]); 21 extern PetscErrorCode MatSOR_MPISBAIJ(Mat,Vec,PetscReal,MatSORType,PetscReal,PetscInt,PetscInt,Vec); 22 23 #undef __FUNCT__ 24 #define __FUNCT__ "MatStoreValues_MPISBAIJ" 25 PetscErrorCode MatStoreValues_MPISBAIJ(Mat mat) 26 { 27 Mat_MPISBAIJ *aij = (Mat_MPISBAIJ*)mat->data; 28 PetscErrorCode ierr; 29 30 PetscFunctionBegin; 31 ierr = MatStoreValues(aij->A);CHKERRQ(ierr); 32 ierr = MatStoreValues(aij->B);CHKERRQ(ierr); 33 PetscFunctionReturn(0); 34 } 35 36 #undef __FUNCT__ 37 #define __FUNCT__ "MatRetrieveValues_MPISBAIJ" 38 PetscErrorCode MatRetrieveValues_MPISBAIJ(Mat mat) 39 { 40 Mat_MPISBAIJ *aij = (Mat_MPISBAIJ*)mat->data; 41 PetscErrorCode ierr; 42 43 PetscFunctionBegin; 44 ierr = MatRetrieveValues(aij->A);CHKERRQ(ierr); 45 ierr = MatRetrieveValues(aij->B);CHKERRQ(ierr); 46 PetscFunctionReturn(0); 47 } 48 49 #define MatSetValues_SeqSBAIJ_A_Private(row,col,value,addv) \ 50 { \ 51 \ 52 brow = row/bs; \ 53 rp = aj + ai[brow]; ap = aa + bs2*ai[brow]; \ 54 rmax = aimax[brow]; nrow = ailen[brow]; \ 55 bcol = col/bs; \ 56 ridx = row % bs; cidx = col % bs; \ 57 low = 0; high = nrow; \ 58 while (high-low > 3) { \ 59 t = (low+high)/2; \ 60 if (rp[t] > bcol) high = t; \ 61 else low = t; \ 62 } \ 63 for (_i=low; _i<high; _i++) { \ 64 if (rp[_i] > bcol) break; \ 65 if (rp[_i] == bcol) { \ 66 bap = ap + bs2*_i + bs*cidx + ridx; \ 67 if (addv == ADD_VALUES) *bap += value; \ 68 else *bap = value; \ 69 goto a_noinsert; \ 70 } \ 71 } \ 72 if (a->nonew == 1) goto a_noinsert; \ 73 if (a->nonew == -1) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 74 MatSeqXAIJReallocateAIJ(A,a->mbs,bs2,nrow,brow,bcol,rmax,aa,ai,aj,rp,ap,aimax,a->nonew,MatScalar); \ 75 N = nrow++ - 1; \ 76 /* shift up all the later entries in this row */ \ 77 for (ii=N; ii>=_i; ii--) { \ 78 rp[ii+1] = rp[ii]; \ 79 ierr = PetscMemcpy(ap+bs2*(ii+1),ap+bs2*(ii),bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 80 } \ 81 if (N>=_i) { ierr = PetscMemzero(ap+bs2*_i,bs2*sizeof(MatScalar));CHKERRQ(ierr); } \ 82 rp[_i] = bcol; \ 83 ap[bs2*_i + bs*cidx + ridx] = value; \ 84 A->nonzerostate++;\ 85 a_noinsert:; \ 86 ailen[brow] = nrow; \ 87 } 88 89 #define MatSetValues_SeqSBAIJ_B_Private(row,col,value,addv) \ 90 { \ 91 brow = row/bs; \ 92 rp = bj + bi[brow]; ap = ba + bs2*bi[brow]; \ 93 rmax = bimax[brow]; nrow = bilen[brow]; \ 94 bcol = col/bs; \ 95 ridx = row % bs; cidx = col % bs; \ 96 low = 0; high = nrow; \ 97 while (high-low > 3) { \ 98 t = (low+high)/2; \ 99 if (rp[t] > bcol) high = t; \ 100 else low = t; \ 101 } \ 102 for (_i=low; _i<high; _i++) { \ 103 if (rp[_i] > bcol) break; \ 104 if (rp[_i] == bcol) { \ 105 bap = ap + bs2*_i + bs*cidx + ridx; \ 106 if (addv == ADD_VALUES) *bap += value; \ 107 else *bap = value; \ 108 goto b_noinsert; \ 109 } \ 110 } \ 111 if (b->nonew == 1) goto b_noinsert; \ 112 if (b->nonew == -1) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero (%D, %D) into matrix", row, col); \ 113 MatSeqXAIJReallocateAIJ(B,b->mbs,bs2,nrow,brow,bcol,rmax,ba,bi,bj,rp,ap,bimax,b->nonew,MatScalar); \ 114 N = nrow++ - 1; \ 115 /* shift up all the later entries in this row */ \ 116 for (ii=N; ii>=_i; ii--) { \ 117 rp[ii+1] = rp[ii]; \ 118 ierr = PetscMemcpy(ap+bs2*(ii+1),ap+bs2*(ii),bs2*sizeof(MatScalar));CHKERRQ(ierr); \ 119 } \ 120 if (N>=_i) { ierr = PetscMemzero(ap+bs2*_i,bs2*sizeof(MatScalar));CHKERRQ(ierr);} \ 121 rp[_i] = bcol; \ 122 ap[bs2*_i + bs*cidx + ridx] = value; \ 123 B->nonzerostate++;\ 124 b_noinsert:; \ 125 bilen[brow] = nrow; \ 126 } 127 128 /* Only add/insert a(i,j) with i<=j (blocks). 129 Any a(i,j) with i>j input by user is ingored. 130 */ 131 #undef __FUNCT__ 132 #define __FUNCT__ "MatSetValues_MPISBAIJ" 133 PetscErrorCode MatSetValues_MPISBAIJ(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const PetscScalar v[],InsertMode addv) 134 { 135 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 136 MatScalar value; 137 PetscBool roworiented = baij->roworiented; 138 PetscErrorCode ierr; 139 PetscInt i,j,row,col; 140 PetscInt rstart_orig=mat->rmap->rstart; 141 PetscInt rend_orig =mat->rmap->rend,cstart_orig=mat->cmap->rstart; 142 PetscInt cend_orig =mat->cmap->rend,bs=mat->rmap->bs; 143 144 /* Some Variables required in the macro */ 145 Mat A = baij->A; 146 Mat_SeqSBAIJ *a = (Mat_SeqSBAIJ*)(A)->data; 147 PetscInt *aimax=a->imax,*ai=a->i,*ailen=a->ilen,*aj=a->j; 148 MatScalar *aa =a->a; 149 150 Mat B = baij->B; 151 Mat_SeqBAIJ *b = (Mat_SeqBAIJ*)(B)->data; 152 PetscInt *bimax=b->imax,*bi=b->i,*bilen=b->ilen,*bj=b->j; 153 MatScalar *ba =b->a; 154 155 PetscInt *rp,ii,nrow,_i,rmax,N,brow,bcol; 156 PetscInt low,high,t,ridx,cidx,bs2=a->bs2; 157 MatScalar *ap,*bap; 158 159 /* for stash */ 160 PetscInt n_loc, *in_loc = NULL; 161 MatScalar *v_loc = NULL; 162 163 PetscFunctionBegin; 164 if (v) PetscValidScalarPointer(v,6); 165 if (!baij->donotstash) { 166 if (n > baij->n_loc) { 167 ierr = PetscFree(baij->in_loc);CHKERRQ(ierr); 168 ierr = PetscFree(baij->v_loc);CHKERRQ(ierr); 169 ierr = PetscMalloc1(n,&baij->in_loc);CHKERRQ(ierr); 170 ierr = PetscMalloc1(n,&baij->v_loc);CHKERRQ(ierr); 171 172 baij->n_loc = n; 173 } 174 in_loc = baij->in_loc; 175 v_loc = baij->v_loc; 176 } 177 178 for (i=0; i<m; i++) { 179 if (im[i] < 0) continue; 180 #if defined(PETSC_USE_DEBUG) 181 if (im[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",im[i],mat->rmap->N-1); 182 #endif 183 if (im[i] >= rstart_orig && im[i] < rend_orig) { /* this processor entry */ 184 row = im[i] - rstart_orig; /* local row index */ 185 for (j=0; j<n; j++) { 186 if (im[i]/bs > in[j]/bs) { 187 if (a->ignore_ltriangular) { 188 continue; /* ignore lower triangular blocks */ 189 } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_USER,"Lower triangular value cannot be set for sbaij format. Ignoring these values, run with -mat_ignore_lower_triangular or call MatSetOption(mat,MAT_IGNORE_LOWER_TRIANGULAR,PETSC_TRUE)"); 190 } 191 if (in[j] >= cstart_orig && in[j] < cend_orig) { /* diag entry (A) */ 192 col = in[j] - cstart_orig; /* local col index */ 193 brow = row/bs; bcol = col/bs; 194 if (brow > bcol) continue; /* ignore lower triangular blocks of A */ 195 if (roworiented) value = v[i*n+j]; 196 else value = v[i+j*m]; 197 MatSetValues_SeqSBAIJ_A_Private(row,col,value,addv); 198 /* ierr = MatSetValues_SeqBAIJ(baij->A,1,&row,1,&col,&value,addv);CHKERRQ(ierr); */ 199 } else if (in[j] < 0) continue; 200 #if defined(PETSC_USE_DEBUG) 201 else if (in[j] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",in[j],mat->cmap->N-1); 202 #endif 203 else { /* off-diag entry (B) */ 204 if (mat->was_assembled) { 205 if (!baij->colmap) { 206 ierr = MatCreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 207 } 208 #if defined(PETSC_USE_CTABLE) 209 ierr = PetscTableFind(baij->colmap,in[j]/bs + 1,&col);CHKERRQ(ierr); 210 col = col - 1; 211 #else 212 col = baij->colmap[in[j]/bs] - 1; 213 #endif 214 if (col < 0 && !((Mat_SeqSBAIJ*)(baij->A->data))->nonew) { 215 ierr = MatDisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 216 col = in[j]; 217 /* Reinitialize the variables required by MatSetValues_SeqBAIJ_B_Private() */ 218 B = baij->B; 219 b = (Mat_SeqBAIJ*)(B)->data; 220 bimax= b->imax;bi=b->i;bilen=b->ilen;bj=b->j; 221 ba = b->a; 222 } else col += in[j]%bs; 223 } else col = in[j]; 224 if (roworiented) value = v[i*n+j]; 225 else value = v[i+j*m]; 226 MatSetValues_SeqSBAIJ_B_Private(row,col,value,addv); 227 /* ierr = MatSetValues_SeqBAIJ(baij->B,1,&row,1,&col,&value,addv);CHKERRQ(ierr); */ 228 } 229 } 230 } else { /* off processor entry */ 231 if (mat->nooffprocentries) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Setting off process row %D even though MatSetOption(,MAT_NO_OFF_PROC_ENTRIES,PETSC_TRUE) was set",im[i]); 232 if (!baij->donotstash) { 233 mat->assembled = PETSC_FALSE; 234 n_loc = 0; 235 for (j=0; j<n; j++) { 236 if (im[i]/bs > in[j]/bs) continue; /* ignore lower triangular blocks */ 237 in_loc[n_loc] = in[j]; 238 if (roworiented) { 239 v_loc[n_loc] = v[i*n+j]; 240 } else { 241 v_loc[n_loc] = v[j*m+i]; 242 } 243 n_loc++; 244 } 245 ierr = MatStashValuesRow_Private(&mat->stash,im[i],n_loc,in_loc,v_loc,PETSC_FALSE);CHKERRQ(ierr); 246 } 247 } 248 } 249 PetscFunctionReturn(0); 250 } 251 252 #undef __FUNCT__ 253 #define __FUNCT__ "MatSetValuesBlocked_MPISBAIJ" 254 PetscErrorCode MatSetValuesBlocked_MPISBAIJ(Mat mat,PetscInt m,const PetscInt im[],PetscInt n,const PetscInt in[],const MatScalar v[],InsertMode addv) 255 { 256 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 257 const MatScalar *value; 258 MatScalar *barray =baij->barray; 259 PetscBool roworiented = baij->roworiented,ignore_ltriangular = ((Mat_SeqSBAIJ*)baij->A->data)->ignore_ltriangular; 260 PetscErrorCode ierr; 261 PetscInt i,j,ii,jj,row,col,rstart=baij->rstartbs; 262 PetscInt rend=baij->rendbs,cstart=baij->rstartbs,stepval; 263 PetscInt cend=baij->rendbs,bs=mat->rmap->bs,bs2=baij->bs2; 264 265 PetscFunctionBegin; 266 if (!barray) { 267 ierr = PetscMalloc1(bs2,&barray);CHKERRQ(ierr); 268 baij->barray = barray; 269 } 270 271 if (roworiented) { 272 stepval = (n-1)*bs; 273 } else { 274 stepval = (m-1)*bs; 275 } 276 for (i=0; i<m; i++) { 277 if (im[i] < 0) continue; 278 #if defined(PETSC_USE_DEBUG) 279 if (im[i] >= baij->Mbs) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large, row %D max %D",im[i],baij->Mbs-1); 280 #endif 281 if (im[i] >= rstart && im[i] < rend) { 282 row = im[i] - rstart; 283 for (j=0; j<n; j++) { 284 if (im[i] > in[j]) { 285 if (ignore_ltriangular) continue; /* ignore lower triangular blocks */ 286 else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_USER,"Lower triangular value cannot be set for sbaij format. Ignoring these values, run with -mat_ignore_lower_triangular or call MatSetOption(mat,MAT_IGNORE_LOWER_TRIANGULAR,PETSC_TRUE)"); 287 } 288 /* If NumCol = 1 then a copy is not required */ 289 if ((roworiented) && (n == 1)) { 290 barray = (MatScalar*) v + i*bs2; 291 } else if ((!roworiented) && (m == 1)) { 292 barray = (MatScalar*) v + j*bs2; 293 } else { /* Here a copy is required */ 294 if (roworiented) { 295 value = v + i*(stepval+bs)*bs + j*bs; 296 } else { 297 value = v + j*(stepval+bs)*bs + i*bs; 298 } 299 for (ii=0; ii<bs; ii++,value+=stepval) { 300 for (jj=0; jj<bs; jj++) { 301 *barray++ = *value++; 302 } 303 } 304 barray -=bs2; 305 } 306 307 if (in[j] >= cstart && in[j] < cend) { 308 col = in[j] - cstart; 309 ierr = MatSetValuesBlocked_SeqSBAIJ(baij->A,1,&row,1,&col,barray,addv);CHKERRQ(ierr); 310 } else if (in[j] < 0) continue; 311 #if defined(PETSC_USE_DEBUG) 312 else if (in[j] >= baij->Nbs) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large, col %D max %D",in[j],baij->Nbs-1); 313 #endif 314 else { 315 if (mat->was_assembled) { 316 if (!baij->colmap) { 317 ierr = MatCreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 318 } 319 320 #if defined(PETSC_USE_DEBUG) 321 #if defined(PETSC_USE_CTABLE) 322 { PetscInt data; 323 ierr = PetscTableFind(baij->colmap,in[j]+1,&data);CHKERRQ(ierr); 324 if ((data - 1) % bs) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Incorrect colmap"); 325 } 326 #else 327 if ((baij->colmap[in[j]] - 1) % bs) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"Incorrect colmap"); 328 #endif 329 #endif 330 #if defined(PETSC_USE_CTABLE) 331 ierr = PetscTableFind(baij->colmap,in[j]+1,&col);CHKERRQ(ierr); 332 col = (col - 1)/bs; 333 #else 334 col = (baij->colmap[in[j]] - 1)/bs; 335 #endif 336 if (col < 0 && !((Mat_SeqBAIJ*)(baij->A->data))->nonew) { 337 ierr = MatDisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 338 col = in[j]; 339 } 340 } else col = in[j]; 341 ierr = MatSetValuesBlocked_SeqBAIJ(baij->B,1,&row,1,&col,barray,addv);CHKERRQ(ierr); 342 } 343 } 344 } else { 345 if (mat->nooffprocentries) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Setting off process row %D even though MatSetOption(,MAT_NO_OFF_PROC_ENTRIES,PETSC_TRUE) was set",im[i]); 346 if (!baij->donotstash) { 347 if (roworiented) { 348 ierr = MatStashValuesRowBlocked_Private(&mat->bstash,im[i],n,in,v,m,n,i);CHKERRQ(ierr); 349 } else { 350 ierr = MatStashValuesColBlocked_Private(&mat->bstash,im[i],n,in,v,m,n,i);CHKERRQ(ierr); 351 } 352 } 353 } 354 } 355 PetscFunctionReturn(0); 356 } 357 358 #undef __FUNCT__ 359 #define __FUNCT__ "MatGetValues_MPISBAIJ" 360 PetscErrorCode MatGetValues_MPISBAIJ(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[]) 361 { 362 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 363 PetscErrorCode ierr; 364 PetscInt bs = mat->rmap->bs,i,j,bsrstart = mat->rmap->rstart,bsrend = mat->rmap->rend; 365 PetscInt bscstart = mat->cmap->rstart,bscend = mat->cmap->rend,row,col,data; 366 367 PetscFunctionBegin; 368 for (i=0; i<m; i++) { 369 if (idxm[i] < 0) continue; /* SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Negative row: %D",idxm[i]); */ 370 if (idxm[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %D max %D",idxm[i],mat->rmap->N-1); 371 if (idxm[i] >= bsrstart && idxm[i] < bsrend) { 372 row = idxm[i] - bsrstart; 373 for (j=0; j<n; j++) { 374 if (idxn[j] < 0) continue; /* SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Negative column %D",idxn[j]); */ 375 if (idxn[j] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %D max %D",idxn[j],mat->cmap->N-1); 376 if (idxn[j] >= bscstart && idxn[j] < bscend) { 377 col = idxn[j] - bscstart; 378 ierr = MatGetValues_SeqSBAIJ(baij->A,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 379 } else { 380 if (!baij->colmap) { 381 ierr = MatCreateColmap_MPIBAIJ_Private(mat);CHKERRQ(ierr); 382 } 383 #if defined(PETSC_USE_CTABLE) 384 ierr = PetscTableFind(baij->colmap,idxn[j]/bs+1,&data);CHKERRQ(ierr); 385 data--; 386 #else 387 data = baij->colmap[idxn[j]/bs]-1; 388 #endif 389 if ((data < 0) || (baij->garray[data/bs] != idxn[j]/bs)) *(v+i*n+j) = 0.0; 390 else { 391 col = data + idxn[j]%bs; 392 ierr = MatGetValues_SeqBAIJ(baij->B,1,&row,1,&col,v+i*n+j);CHKERRQ(ierr); 393 } 394 } 395 } 396 } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only local values currently supported"); 397 } 398 PetscFunctionReturn(0); 399 } 400 401 #undef __FUNCT__ 402 #define __FUNCT__ "MatNorm_MPISBAIJ" 403 PetscErrorCode MatNorm_MPISBAIJ(Mat mat,NormType type,PetscReal *norm) 404 { 405 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 406 PetscErrorCode ierr; 407 PetscReal sum[2],*lnorm2; 408 409 PetscFunctionBegin; 410 if (baij->size == 1) { 411 ierr = MatNorm(baij->A,type,norm);CHKERRQ(ierr); 412 } else { 413 if (type == NORM_FROBENIUS) { 414 ierr = PetscMalloc1(2,&lnorm2);CHKERRQ(ierr); 415 ierr = MatNorm(baij->A,type,lnorm2);CHKERRQ(ierr); 416 *lnorm2 = (*lnorm2)*(*lnorm2); lnorm2++; /* squar power of norm(A) */ 417 ierr = MatNorm(baij->B,type,lnorm2);CHKERRQ(ierr); 418 *lnorm2 = (*lnorm2)*(*lnorm2); lnorm2--; /* squar power of norm(B) */ 419 ierr = MPI_Allreduce(lnorm2,sum,2,MPIU_REAL,MPIU_SUM,PetscObjectComm((PetscObject)mat));CHKERRQ(ierr); 420 *norm = PetscSqrtReal(sum[0] + 2*sum[1]); 421 ierr = PetscFree(lnorm2);CHKERRQ(ierr); 422 } else if (type == NORM_INFINITY || type == NORM_1) { /* max row/column sum */ 423 Mat_SeqSBAIJ *amat=(Mat_SeqSBAIJ*)baij->A->data; 424 Mat_SeqBAIJ *bmat=(Mat_SeqBAIJ*)baij->B->data; 425 PetscReal *rsum,*rsum2,vabs; 426 PetscInt *jj,*garray=baij->garray,rstart=baij->rstartbs,nz; 427 PetscInt brow,bcol,col,bs=baij->A->rmap->bs,row,grow,gcol,mbs=amat->mbs; 428 MatScalar *v; 429 430 ierr = PetscMalloc2(mat->cmap->N,&rsum,mat->cmap->N,&rsum2);CHKERRQ(ierr); 431 ierr = PetscMemzero(rsum,mat->cmap->N*sizeof(PetscReal));CHKERRQ(ierr); 432 /* Amat */ 433 v = amat->a; jj = amat->j; 434 for (brow=0; brow<mbs; brow++) { 435 grow = bs*(rstart + brow); 436 nz = amat->i[brow+1] - amat->i[brow]; 437 for (bcol=0; bcol<nz; bcol++) { 438 gcol = bs*(rstart + *jj); jj++; 439 for (col=0; col<bs; col++) { 440 for (row=0; row<bs; row++) { 441 vabs = PetscAbsScalar(*v); v++; 442 rsum[gcol+col] += vabs; 443 /* non-diagonal block */ 444 if (bcol > 0 && vabs > 0.0) rsum[grow+row] += vabs; 445 } 446 } 447 } 448 } 449 /* Bmat */ 450 v = bmat->a; jj = bmat->j; 451 for (brow=0; brow<mbs; brow++) { 452 grow = bs*(rstart + brow); 453 nz = bmat->i[brow+1] - bmat->i[brow]; 454 for (bcol=0; bcol<nz; bcol++) { 455 gcol = bs*garray[*jj]; jj++; 456 for (col=0; col<bs; col++) { 457 for (row=0; row<bs; row++) { 458 vabs = PetscAbsScalar(*v); v++; 459 rsum[gcol+col] += vabs; 460 rsum[grow+row] += vabs; 461 } 462 } 463 } 464 } 465 ierr = MPI_Allreduce(rsum,rsum2,mat->cmap->N,MPIU_REAL,MPIU_SUM,PetscObjectComm((PetscObject)mat));CHKERRQ(ierr); 466 *norm = 0.0; 467 for (col=0; col<mat->cmap->N; col++) { 468 if (rsum2[col] > *norm) *norm = rsum2[col]; 469 } 470 ierr = PetscFree2(rsum,rsum2);CHKERRQ(ierr); 471 } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"No support for this norm yet"); 472 } 473 PetscFunctionReturn(0); 474 } 475 476 #undef __FUNCT__ 477 #define __FUNCT__ "MatAssemblyBegin_MPISBAIJ" 478 PetscErrorCode MatAssemblyBegin_MPISBAIJ(Mat mat,MatAssemblyType mode) 479 { 480 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 481 PetscErrorCode ierr; 482 PetscInt nstash,reallocs; 483 InsertMode addv; 484 485 PetscFunctionBegin; 486 if (baij->donotstash || mat->nooffprocentries) PetscFunctionReturn(0); 487 488 /* make sure all processors are either in INSERTMODE or ADDMODE */ 489 ierr = MPI_Allreduce((PetscEnum*)&mat->insertmode,(PetscEnum*)&addv,1,MPIU_ENUM,MPI_BOR,PetscObjectComm((PetscObject)mat));CHKERRQ(ierr); 490 if (addv == (ADD_VALUES|INSERT_VALUES)) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Some processors inserted others added"); 491 mat->insertmode = addv; /* in case this processor had no cache */ 492 493 ierr = MatStashScatterBegin_Private(mat,&mat->stash,mat->rmap->range);CHKERRQ(ierr); 494 ierr = MatStashScatterBegin_Private(mat,&mat->bstash,baij->rangebs);CHKERRQ(ierr); 495 ierr = MatStashGetInfo_Private(&mat->stash,&nstash,&reallocs);CHKERRQ(ierr); 496 ierr = PetscInfo2(mat,"Stash has %D entries,uses %D mallocs.\n",nstash,reallocs);CHKERRQ(ierr); 497 ierr = MatStashGetInfo_Private(&mat->stash,&nstash,&reallocs);CHKERRQ(ierr); 498 ierr = PetscInfo2(mat,"Block-Stash has %D entries, uses %D mallocs.\n",nstash,reallocs);CHKERRQ(ierr); 499 PetscFunctionReturn(0); 500 } 501 502 #undef __FUNCT__ 503 #define __FUNCT__ "MatAssemblyEnd_MPISBAIJ" 504 PetscErrorCode MatAssemblyEnd_MPISBAIJ(Mat mat,MatAssemblyType mode) 505 { 506 Mat_MPISBAIJ *baij=(Mat_MPISBAIJ*)mat->data; 507 Mat_SeqSBAIJ *a =(Mat_SeqSBAIJ*)baij->A->data; 508 PetscErrorCode ierr; 509 PetscInt i,j,rstart,ncols,flg,bs2=baij->bs2; 510 PetscInt *row,*col; 511 PetscBool other_disassembled; 512 PetscMPIInt n; 513 PetscBool r1,r2,r3; 514 MatScalar *val; 515 InsertMode addv = mat->insertmode; 516 517 /* do not use 'b=(Mat_SeqBAIJ*)baij->B->data' as B can be reset in disassembly */ 518 PetscFunctionBegin; 519 if (!baij->donotstash && !mat->nooffprocentries) { 520 while (1) { 521 ierr = MatStashScatterGetMesg_Private(&mat->stash,&n,&row,&col,&val,&flg);CHKERRQ(ierr); 522 if (!flg) break; 523 524 for (i=0; i<n;) { 525 /* Now identify the consecutive vals belonging to the same row */ 526 for (j=i,rstart=row[j]; j<n; j++) { 527 if (row[j] != rstart) break; 528 } 529 if (j < n) ncols = j-i; 530 else ncols = n-i; 531 /* Now assemble all these values with a single function call */ 532 ierr = MatSetValues_MPISBAIJ(mat,1,row+i,ncols,col+i,val+i,addv);CHKERRQ(ierr); 533 i = j; 534 } 535 } 536 ierr = MatStashScatterEnd_Private(&mat->stash);CHKERRQ(ierr); 537 /* Now process the block-stash. Since the values are stashed column-oriented, 538 set the roworiented flag to column oriented, and after MatSetValues() 539 restore the original flags */ 540 r1 = baij->roworiented; 541 r2 = a->roworiented; 542 r3 = ((Mat_SeqBAIJ*)baij->B->data)->roworiented; 543 544 baij->roworiented = PETSC_FALSE; 545 a->roworiented = PETSC_FALSE; 546 547 ((Mat_SeqBAIJ*)baij->B->data)->roworiented = PETSC_FALSE; /* b->roworinted */ 548 while (1) { 549 ierr = MatStashScatterGetMesg_Private(&mat->bstash,&n,&row,&col,&val,&flg);CHKERRQ(ierr); 550 if (!flg) break; 551 552 for (i=0; i<n;) { 553 /* Now identify the consecutive vals belonging to the same row */ 554 for (j=i,rstart=row[j]; j<n; j++) { 555 if (row[j] != rstart) break; 556 } 557 if (j < n) ncols = j-i; 558 else ncols = n-i; 559 ierr = MatSetValuesBlocked_MPISBAIJ(mat,1,row+i,ncols,col+i,val+i*bs2,addv);CHKERRQ(ierr); 560 i = j; 561 } 562 } 563 ierr = MatStashScatterEnd_Private(&mat->bstash);CHKERRQ(ierr); 564 565 baij->roworiented = r1; 566 a->roworiented = r2; 567 568 ((Mat_SeqBAIJ*)baij->B->data)->roworiented = r3; /* b->roworinted */ 569 } 570 571 ierr = MatAssemblyBegin(baij->A,mode);CHKERRQ(ierr); 572 ierr = MatAssemblyEnd(baij->A,mode);CHKERRQ(ierr); 573 574 /* determine if any processor has disassembled, if so we must 575 also disassemble ourselfs, in order that we may reassemble. */ 576 /* 577 if nonzero structure of submatrix B cannot change then we know that 578 no processor disassembled thus we can skip this stuff 579 */ 580 if (!((Mat_SeqBAIJ*)baij->B->data)->nonew) { 581 ierr = MPI_Allreduce(&mat->was_assembled,&other_disassembled,1,MPIU_BOOL,MPI_PROD,PetscObjectComm((PetscObject)mat));CHKERRQ(ierr); 582 if (mat->was_assembled && !other_disassembled) { 583 ierr = MatDisAssemble_MPISBAIJ(mat);CHKERRQ(ierr); 584 } 585 } 586 587 if (!mat->was_assembled && mode == MAT_FINAL_ASSEMBLY) { 588 ierr = MatSetUpMultiply_MPISBAIJ(mat);CHKERRQ(ierr); /* setup Mvctx and sMvctx */ 589 } 590 ierr = MatAssemblyBegin(baij->B,mode);CHKERRQ(ierr); 591 ierr = MatAssemblyEnd(baij->B,mode);CHKERRQ(ierr); 592 593 ierr = PetscFree2(baij->rowvalues,baij->rowindices);CHKERRQ(ierr); 594 595 baij->rowvalues = 0; 596 597 /* if no new nonzero locations are allowed in matrix then only set the matrix state the first time through */ 598 if ((!mat->was_assembled && mode == MAT_FINAL_ASSEMBLY) || !((Mat_SeqBAIJ*)(baij->A->data))->nonew) { 599 PetscObjectState state = baij->A->nonzerostate + baij->B->nonzerostate; 600 ierr = MPI_Allreduce(&state,&mat->nonzerostate,1,MPIU_INT64,MPI_SUM,PetscObjectComm((PetscObject)mat));CHKERRQ(ierr); 601 } 602 PetscFunctionReturn(0); 603 } 604 605 extern PetscErrorCode MatView_SeqSBAIJ_ASCII(Mat,PetscViewer); 606 extern PetscErrorCode MatSetValues_MPIBAIJ(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode); 607 #include <petscdraw.h> 608 #undef __FUNCT__ 609 #define __FUNCT__ "MatView_MPISBAIJ_ASCIIorDraworSocket" 610 static PetscErrorCode MatView_MPISBAIJ_ASCIIorDraworSocket(Mat mat,PetscViewer viewer) 611 { 612 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 613 PetscErrorCode ierr; 614 PetscInt bs = mat->rmap->bs; 615 PetscMPIInt rank = baij->rank; 616 PetscBool iascii,isdraw; 617 PetscViewer sviewer; 618 PetscViewerFormat format; 619 620 PetscFunctionBegin; 621 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 622 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERDRAW,&isdraw);CHKERRQ(ierr); 623 if (iascii) { 624 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 625 if (format == PETSC_VIEWER_ASCII_INFO_DETAIL) { 626 MatInfo info; 627 ierr = MPI_Comm_rank(PetscObjectComm((PetscObject)mat),&rank);CHKERRQ(ierr); 628 ierr = MatGetInfo(mat,MAT_LOCAL,&info);CHKERRQ(ierr); 629 ierr = PetscViewerASCIISynchronizedAllow(viewer,PETSC_TRUE);CHKERRQ(ierr); 630 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] Local rows %D nz %D nz alloced %D bs %D mem %D\n",rank,mat->rmap->n,(PetscInt)info.nz_used,(PetscInt)info.nz_allocated,mat->rmap->bs,(PetscInt)info.memory);CHKERRQ(ierr); 631 ierr = MatGetInfo(baij->A,MAT_LOCAL,&info);CHKERRQ(ierr); 632 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] on-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used);CHKERRQ(ierr); 633 ierr = MatGetInfo(baij->B,MAT_LOCAL,&info);CHKERRQ(ierr); 634 ierr = PetscViewerASCIISynchronizedPrintf(viewer,"[%d] off-diagonal part: nz %D \n",rank,(PetscInt)info.nz_used);CHKERRQ(ierr); 635 ierr = PetscViewerFlush(viewer);CHKERRQ(ierr); 636 ierr = PetscViewerASCIISynchronizedAllow(viewer,PETSC_FALSE);CHKERRQ(ierr); 637 ierr = PetscViewerASCIIPrintf(viewer,"Information on VecScatter used in matrix-vector product: \n");CHKERRQ(ierr); 638 ierr = VecScatterView(baij->Mvctx,viewer);CHKERRQ(ierr); 639 PetscFunctionReturn(0); 640 } else if (format == PETSC_VIEWER_ASCII_INFO) { 641 ierr = PetscViewerASCIIPrintf(viewer," block size is %D\n",bs);CHKERRQ(ierr); 642 PetscFunctionReturn(0); 643 } else if (format == PETSC_VIEWER_ASCII_FACTOR_INFO) { 644 PetscFunctionReturn(0); 645 } 646 } 647 648 if (isdraw) { 649 PetscDraw draw; 650 PetscBool isnull; 651 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 652 ierr = PetscDrawIsNull(draw,&isnull);CHKERRQ(ierr); if (isnull) PetscFunctionReturn(0); 653 } 654 655 { 656 /* assemble the entire matrix onto first processor. */ 657 Mat A; 658 Mat_SeqSBAIJ *Aloc; 659 Mat_SeqBAIJ *Bloc; 660 PetscInt M = mat->rmap->N,N = mat->cmap->N,*ai,*aj,col,i,j,k,*rvals,mbs = baij->mbs; 661 MatScalar *a; 662 663 /* Should this be the same type as mat? */ 664 ierr = MatCreate(PetscObjectComm((PetscObject)mat),&A);CHKERRQ(ierr); 665 if (!rank) { 666 ierr = MatSetSizes(A,M,N,M,N);CHKERRQ(ierr); 667 } else { 668 ierr = MatSetSizes(A,0,0,M,N);CHKERRQ(ierr); 669 } 670 ierr = MatSetType(A,MATMPISBAIJ);CHKERRQ(ierr); 671 ierr = MatMPISBAIJSetPreallocation(A,mat->rmap->bs,0,NULL,0,NULL);CHKERRQ(ierr); 672 ierr = MatSetOption(A,MAT_NEW_NONZERO_LOCATION_ERR,PETSC_FALSE);CHKERRQ(ierr); 673 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)A);CHKERRQ(ierr); 674 675 /* copy over the A part */ 676 Aloc = (Mat_SeqSBAIJ*)baij->A->data; 677 ai = Aloc->i; aj = Aloc->j; a = Aloc->a; 678 ierr = PetscMalloc1(bs,&rvals);CHKERRQ(ierr); 679 680 for (i=0; i<mbs; i++) { 681 rvals[0] = bs*(baij->rstartbs + i); 682 for (j=1; j<bs; j++) rvals[j] = rvals[j-1] + 1; 683 for (j=ai[i]; j<ai[i+1]; j++) { 684 col = (baij->cstartbs+aj[j])*bs; 685 for (k=0; k<bs; k++) { 686 ierr = MatSetValues_MPISBAIJ(A,bs,rvals,1,&col,a,INSERT_VALUES);CHKERRQ(ierr); 687 col++; 688 a += bs; 689 } 690 } 691 } 692 /* copy over the B part */ 693 Bloc = (Mat_SeqBAIJ*)baij->B->data; 694 ai = Bloc->i; aj = Bloc->j; a = Bloc->a; 695 for (i=0; i<mbs; i++) { 696 697 rvals[0] = bs*(baij->rstartbs + i); 698 for (j=1; j<bs; j++) rvals[j] = rvals[j-1] + 1; 699 for (j=ai[i]; j<ai[i+1]; j++) { 700 col = baij->garray[aj[j]]*bs; 701 for (k=0; k<bs; k++) { 702 ierr = MatSetValues_MPIBAIJ(A,bs,rvals,1,&col,a,INSERT_VALUES);CHKERRQ(ierr); 703 col++; 704 a += bs; 705 } 706 } 707 } 708 ierr = PetscFree(rvals);CHKERRQ(ierr); 709 ierr = MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 710 ierr = MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 711 /* 712 Everyone has to call to draw the matrix since the graphics waits are 713 synchronized across all processors that share the PetscDraw object 714 */ 715 ierr = PetscViewerGetSingleton(viewer,&sviewer);CHKERRQ(ierr); 716 if (!rank) { 717 ierr = MatView_SeqSBAIJ_ASCII(((Mat_MPISBAIJ*)(A->data))->A,sviewer);CHKERRQ(ierr); 718 } 719 ierr = PetscViewerRestoreSingleton(viewer,&sviewer);CHKERRQ(ierr); 720 ierr = MatDestroy(&A);CHKERRQ(ierr); 721 } 722 PetscFunctionReturn(0); 723 } 724 725 #undef __FUNCT__ 726 #define __FUNCT__ "MatView_MPISBAIJ" 727 PetscErrorCode MatView_MPISBAIJ(Mat mat,PetscViewer viewer) 728 { 729 PetscErrorCode ierr; 730 PetscBool iascii,isdraw,issocket,isbinary; 731 732 PetscFunctionBegin; 733 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr); 734 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERDRAW,&isdraw);CHKERRQ(ierr); 735 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSOCKET,&issocket);CHKERRQ(ierr); 736 ierr = PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr); 737 if (iascii || isdraw || issocket || isbinary) { 738 ierr = MatView_MPISBAIJ_ASCIIorDraworSocket(mat,viewer);CHKERRQ(ierr); 739 } 740 PetscFunctionReturn(0); 741 } 742 743 #undef __FUNCT__ 744 #define __FUNCT__ "MatDestroy_MPISBAIJ" 745 PetscErrorCode MatDestroy_MPISBAIJ(Mat mat) 746 { 747 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 748 PetscErrorCode ierr; 749 750 PetscFunctionBegin; 751 #if defined(PETSC_USE_LOG) 752 PetscLogObjectState((PetscObject)mat,"Rows=%D,Cols=%D",mat->rmap->N,mat->cmap->N); 753 #endif 754 ierr = MatStashDestroy_Private(&mat->stash);CHKERRQ(ierr); 755 ierr = MatStashDestroy_Private(&mat->bstash);CHKERRQ(ierr); 756 ierr = MatDestroy(&baij->A);CHKERRQ(ierr); 757 ierr = MatDestroy(&baij->B);CHKERRQ(ierr); 758 #if defined(PETSC_USE_CTABLE) 759 ierr = PetscTableDestroy(&baij->colmap);CHKERRQ(ierr); 760 #else 761 ierr = PetscFree(baij->colmap);CHKERRQ(ierr); 762 #endif 763 ierr = PetscFree(baij->garray);CHKERRQ(ierr); 764 ierr = VecDestroy(&baij->lvec);CHKERRQ(ierr); 765 ierr = VecScatterDestroy(&baij->Mvctx);CHKERRQ(ierr); 766 ierr = VecDestroy(&baij->slvec0);CHKERRQ(ierr); 767 ierr = VecDestroy(&baij->slvec0b);CHKERRQ(ierr); 768 ierr = VecDestroy(&baij->slvec1);CHKERRQ(ierr); 769 ierr = VecDestroy(&baij->slvec1a);CHKERRQ(ierr); 770 ierr = VecDestroy(&baij->slvec1b);CHKERRQ(ierr); 771 ierr = VecScatterDestroy(&baij->sMvctx);CHKERRQ(ierr); 772 ierr = PetscFree2(baij->rowvalues,baij->rowindices);CHKERRQ(ierr); 773 ierr = PetscFree(baij->barray);CHKERRQ(ierr); 774 ierr = PetscFree(baij->hd);CHKERRQ(ierr); 775 ierr = VecDestroy(&baij->diag);CHKERRQ(ierr); 776 ierr = VecDestroy(&baij->bb1);CHKERRQ(ierr); 777 ierr = VecDestroy(&baij->xx1);CHKERRQ(ierr); 778 #if defined(PETSC_USE_REAL_MAT_SINGLE) 779 ierr = PetscFree(baij->setvaluescopy);CHKERRQ(ierr); 780 #endif 781 ierr = PetscFree(baij->in_loc);CHKERRQ(ierr); 782 ierr = PetscFree(baij->v_loc);CHKERRQ(ierr); 783 ierr = PetscFree(baij->rangebs);CHKERRQ(ierr); 784 ierr = PetscFree(mat->data);CHKERRQ(ierr); 785 786 ierr = PetscObjectChangeTypeName((PetscObject)mat,0);CHKERRQ(ierr); 787 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatStoreValues_C",NULL);CHKERRQ(ierr); 788 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatRetrieveValues_C",NULL);CHKERRQ(ierr); 789 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatGetDiagonalBlock_C",NULL);CHKERRQ(ierr); 790 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatMPISBAIJSetPreallocation_C",NULL);CHKERRQ(ierr); 791 ierr = PetscObjectComposeFunction((PetscObject)mat,"MatConvert_mpisbaij_mpisbstrm_C",NULL);CHKERRQ(ierr); 792 PetscFunctionReturn(0); 793 } 794 795 #undef __FUNCT__ 796 #define __FUNCT__ "MatMult_MPISBAIJ_Hermitian" 797 PetscErrorCode MatMult_MPISBAIJ_Hermitian(Mat A,Vec xx,Vec yy) 798 { 799 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 800 PetscErrorCode ierr; 801 PetscInt nt,mbs=a->mbs,bs=A->rmap->bs; 802 PetscScalar *x,*from; 803 804 PetscFunctionBegin; 805 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 806 if (nt != A->cmap->n) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Incompatible partition of A and xx"); 807 808 /* diagonal part */ 809 ierr = (*a->A->ops->mult)(a->A,xx,a->slvec1a);CHKERRQ(ierr); 810 ierr = VecSet(a->slvec1b,0.0);CHKERRQ(ierr); 811 812 /* subdiagonal part */ 813 ierr = (*a->B->ops->multhermitiantranspose)(a->B,xx,a->slvec0b);CHKERRQ(ierr); 814 815 /* copy x into the vec slvec0 */ 816 ierr = VecGetArray(a->slvec0,&from);CHKERRQ(ierr); 817 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 818 819 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 820 ierr = VecRestoreArray(a->slvec0,&from);CHKERRQ(ierr); 821 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 822 823 ierr = VecScatterBegin(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 824 ierr = VecScatterEnd(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 825 /* supperdiagonal part */ 826 ierr = (*a->B->ops->multadd)(a->B,a->slvec1b,a->slvec1a,yy);CHKERRQ(ierr); 827 PetscFunctionReturn(0); 828 } 829 830 #undef __FUNCT__ 831 #define __FUNCT__ "MatMult_MPISBAIJ" 832 PetscErrorCode MatMult_MPISBAIJ(Mat A,Vec xx,Vec yy) 833 { 834 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 835 PetscErrorCode ierr; 836 PetscInt nt,mbs=a->mbs,bs=A->rmap->bs; 837 PetscScalar *x,*from; 838 839 PetscFunctionBegin; 840 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 841 if (nt != A->cmap->n) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Incompatible partition of A and xx"); 842 843 /* diagonal part */ 844 ierr = (*a->A->ops->mult)(a->A,xx,a->slvec1a);CHKERRQ(ierr); 845 ierr = VecSet(a->slvec1b,0.0);CHKERRQ(ierr); 846 847 /* subdiagonal part */ 848 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->slvec0b);CHKERRQ(ierr); 849 850 /* copy x into the vec slvec0 */ 851 ierr = VecGetArray(a->slvec0,&from);CHKERRQ(ierr); 852 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 853 854 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 855 ierr = VecRestoreArray(a->slvec0,&from);CHKERRQ(ierr); 856 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 857 858 ierr = VecScatterBegin(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 859 ierr = VecScatterEnd(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 860 /* supperdiagonal part */ 861 ierr = (*a->B->ops->multadd)(a->B,a->slvec1b,a->slvec1a,yy);CHKERRQ(ierr); 862 PetscFunctionReturn(0); 863 } 864 865 #undef __FUNCT__ 866 #define __FUNCT__ "MatMult_MPISBAIJ_2comm" 867 PetscErrorCode MatMult_MPISBAIJ_2comm(Mat A,Vec xx,Vec yy) 868 { 869 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 870 PetscErrorCode ierr; 871 PetscInt nt; 872 873 PetscFunctionBegin; 874 ierr = VecGetLocalSize(xx,&nt);CHKERRQ(ierr); 875 if (nt != A->cmap->n) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Incompatible partition of A and xx"); 876 877 ierr = VecGetLocalSize(yy,&nt);CHKERRQ(ierr); 878 if (nt != A->rmap->N) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Incompatible parition of A and yy"); 879 880 ierr = VecScatterBegin(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 881 /* do diagonal part */ 882 ierr = (*a->A->ops->mult)(a->A,xx,yy);CHKERRQ(ierr); 883 /* do supperdiagonal part */ 884 ierr = VecScatterEnd(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 885 ierr = (*a->B->ops->multadd)(a->B,a->lvec,yy,yy);CHKERRQ(ierr); 886 /* do subdiagonal part */ 887 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 888 ierr = VecScatterBegin(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 889 ierr = VecScatterEnd(a->Mvctx,a->lvec,yy,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 890 PetscFunctionReturn(0); 891 } 892 893 #undef __FUNCT__ 894 #define __FUNCT__ "MatMultAdd_MPISBAIJ" 895 PetscErrorCode MatMultAdd_MPISBAIJ(Mat A,Vec xx,Vec yy,Vec zz) 896 { 897 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 898 PetscErrorCode ierr; 899 PetscInt mbs=a->mbs,bs=A->rmap->bs; 900 PetscScalar *x,*from,zero=0.0; 901 902 PetscFunctionBegin; 903 /* 904 PetscSynchronizedPrintf(PetscObjectComm((PetscObject)A)," MatMultAdd is called ...\n"); 905 PetscSynchronizedFlush(PetscObjectComm((PetscObject)A),PETSC_STDOUT); 906 */ 907 /* diagonal part */ 908 ierr = (*a->A->ops->multadd)(a->A,xx,yy,a->slvec1a);CHKERRQ(ierr); 909 ierr = VecSet(a->slvec1b,zero);CHKERRQ(ierr); 910 911 /* subdiagonal part */ 912 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->slvec0b);CHKERRQ(ierr); 913 914 /* copy x into the vec slvec0 */ 915 ierr = VecGetArray(a->slvec0,&from);CHKERRQ(ierr); 916 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 917 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 918 ierr = VecRestoreArray(a->slvec0,&from);CHKERRQ(ierr); 919 920 ierr = VecScatterBegin(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 921 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 922 ierr = VecScatterEnd(a->sMvctx,a->slvec0,a->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 923 924 /* supperdiagonal part */ 925 ierr = (*a->B->ops->multadd)(a->B,a->slvec1b,a->slvec1a,zz);CHKERRQ(ierr); 926 PetscFunctionReturn(0); 927 } 928 929 #undef __FUNCT__ 930 #define __FUNCT__ "MatMultAdd_MPISBAIJ_2comm" 931 PetscErrorCode MatMultAdd_MPISBAIJ_2comm(Mat A,Vec xx,Vec yy,Vec zz) 932 { 933 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 934 PetscErrorCode ierr; 935 936 PetscFunctionBegin; 937 ierr = VecScatterBegin(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 938 /* do diagonal part */ 939 ierr = (*a->A->ops->multadd)(a->A,xx,yy,zz);CHKERRQ(ierr); 940 /* do supperdiagonal part */ 941 ierr = VecScatterEnd(a->Mvctx,xx,a->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 942 ierr = (*a->B->ops->multadd)(a->B,a->lvec,zz,zz);CHKERRQ(ierr); 943 944 /* do subdiagonal part */ 945 ierr = (*a->B->ops->multtranspose)(a->B,xx,a->lvec);CHKERRQ(ierr); 946 ierr = VecScatterBegin(a->Mvctx,a->lvec,zz,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 947 ierr = VecScatterEnd(a->Mvctx,a->lvec,zz,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 948 PetscFunctionReturn(0); 949 } 950 951 /* 952 This only works correctly for square matrices where the subblock A->A is the 953 diagonal block 954 */ 955 #undef __FUNCT__ 956 #define __FUNCT__ "MatGetDiagonal_MPISBAIJ" 957 PetscErrorCode MatGetDiagonal_MPISBAIJ(Mat A,Vec v) 958 { 959 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 960 PetscErrorCode ierr; 961 962 PetscFunctionBegin; 963 /* if (a->rmap->N != a->cmap->N) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Supports only square matrix where A->A is diag block"); */ 964 ierr = MatGetDiagonal(a->A,v);CHKERRQ(ierr); 965 PetscFunctionReturn(0); 966 } 967 968 #undef __FUNCT__ 969 #define __FUNCT__ "MatScale_MPISBAIJ" 970 PetscErrorCode MatScale_MPISBAIJ(Mat A,PetscScalar aa) 971 { 972 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 973 PetscErrorCode ierr; 974 975 PetscFunctionBegin; 976 ierr = MatScale(a->A,aa);CHKERRQ(ierr); 977 ierr = MatScale(a->B,aa);CHKERRQ(ierr); 978 PetscFunctionReturn(0); 979 } 980 981 #undef __FUNCT__ 982 #define __FUNCT__ "MatGetRow_MPISBAIJ" 983 PetscErrorCode MatGetRow_MPISBAIJ(Mat matin,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 984 { 985 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)matin->data; 986 PetscScalar *vworkA,*vworkB,**pvA,**pvB,*v_p; 987 PetscErrorCode ierr; 988 PetscInt bs = matin->rmap->bs,bs2 = mat->bs2,i,*cworkA,*cworkB,**pcA,**pcB; 989 PetscInt nztot,nzA,nzB,lrow,brstart = matin->rmap->rstart,brend = matin->rmap->rend; 990 PetscInt *cmap,*idx_p,cstart = mat->rstartbs; 991 992 PetscFunctionBegin; 993 if (mat->getrowactive) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Already active"); 994 mat->getrowactive = PETSC_TRUE; 995 996 if (!mat->rowvalues && (idx || v)) { 997 /* 998 allocate enough space to hold information from the longest row. 999 */ 1000 Mat_SeqSBAIJ *Aa = (Mat_SeqSBAIJ*)mat->A->data; 1001 Mat_SeqBAIJ *Ba = (Mat_SeqBAIJ*)mat->B->data; 1002 PetscInt max = 1,mbs = mat->mbs,tmp; 1003 for (i=0; i<mbs; i++) { 1004 tmp = Aa->i[i+1] - Aa->i[i] + Ba->i[i+1] - Ba->i[i]; /* row length */ 1005 if (max < tmp) max = tmp; 1006 } 1007 ierr = PetscMalloc2(max*bs2,&mat->rowvalues,max*bs2,&mat->rowindices);CHKERRQ(ierr); 1008 } 1009 1010 if (row < brstart || row >= brend) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only local rows"); 1011 lrow = row - brstart; /* local row index */ 1012 1013 pvA = &vworkA; pcA = &cworkA; pvB = &vworkB; pcB = &cworkB; 1014 if (!v) {pvA = 0; pvB = 0;} 1015 if (!idx) {pcA = 0; if (!v) pcB = 0;} 1016 ierr = (*mat->A->ops->getrow)(mat->A,lrow,&nzA,pcA,pvA);CHKERRQ(ierr); 1017 ierr = (*mat->B->ops->getrow)(mat->B,lrow,&nzB,pcB,pvB);CHKERRQ(ierr); 1018 nztot = nzA + nzB; 1019 1020 cmap = mat->garray; 1021 if (v || idx) { 1022 if (nztot) { 1023 /* Sort by increasing column numbers, assuming A and B already sorted */ 1024 PetscInt imark = -1; 1025 if (v) { 1026 *v = v_p = mat->rowvalues; 1027 for (i=0; i<nzB; i++) { 1028 if (cmap[cworkB[i]/bs] < cstart) v_p[i] = vworkB[i]; 1029 else break; 1030 } 1031 imark = i; 1032 for (i=0; i<nzA; i++) v_p[imark+i] = vworkA[i]; 1033 for (i=imark; i<nzB; i++) v_p[nzA+i] = vworkB[i]; 1034 } 1035 if (idx) { 1036 *idx = idx_p = mat->rowindices; 1037 if (imark > -1) { 1038 for (i=0; i<imark; i++) { 1039 idx_p[i] = cmap[cworkB[i]/bs]*bs + cworkB[i]%bs; 1040 } 1041 } else { 1042 for (i=0; i<nzB; i++) { 1043 if (cmap[cworkB[i]/bs] < cstart) idx_p[i] = cmap[cworkB[i]/bs]*bs + cworkB[i]%bs; 1044 else break; 1045 } 1046 imark = i; 1047 } 1048 for (i=0; i<nzA; i++) idx_p[imark+i] = cstart*bs + cworkA[i]; 1049 for (i=imark; i<nzB; i++) idx_p[nzA+i] = cmap[cworkB[i]/bs]*bs + cworkB[i]%bs ; 1050 } 1051 } else { 1052 if (idx) *idx = 0; 1053 if (v) *v = 0; 1054 } 1055 } 1056 *nz = nztot; 1057 ierr = (*mat->A->ops->restorerow)(mat->A,lrow,&nzA,pcA,pvA);CHKERRQ(ierr); 1058 ierr = (*mat->B->ops->restorerow)(mat->B,lrow,&nzB,pcB,pvB);CHKERRQ(ierr); 1059 PetscFunctionReturn(0); 1060 } 1061 1062 #undef __FUNCT__ 1063 #define __FUNCT__ "MatRestoreRow_MPISBAIJ" 1064 PetscErrorCode MatRestoreRow_MPISBAIJ(Mat mat,PetscInt row,PetscInt *nz,PetscInt **idx,PetscScalar **v) 1065 { 1066 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 1067 1068 PetscFunctionBegin; 1069 if (!baij->getrowactive) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"MatGetRow() must be called first"); 1070 baij->getrowactive = PETSC_FALSE; 1071 PetscFunctionReturn(0); 1072 } 1073 1074 #undef __FUNCT__ 1075 #define __FUNCT__ "MatGetRowUpperTriangular_MPISBAIJ" 1076 PetscErrorCode MatGetRowUpperTriangular_MPISBAIJ(Mat A) 1077 { 1078 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1079 Mat_SeqSBAIJ *aA = (Mat_SeqSBAIJ*)a->A->data; 1080 1081 PetscFunctionBegin; 1082 aA->getrow_utriangular = PETSC_TRUE; 1083 PetscFunctionReturn(0); 1084 } 1085 #undef __FUNCT__ 1086 #define __FUNCT__ "MatRestoreRowUpperTriangular_MPISBAIJ" 1087 PetscErrorCode MatRestoreRowUpperTriangular_MPISBAIJ(Mat A) 1088 { 1089 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1090 Mat_SeqSBAIJ *aA = (Mat_SeqSBAIJ*)a->A->data; 1091 1092 PetscFunctionBegin; 1093 aA->getrow_utriangular = PETSC_FALSE; 1094 PetscFunctionReturn(0); 1095 } 1096 1097 #undef __FUNCT__ 1098 #define __FUNCT__ "MatRealPart_MPISBAIJ" 1099 PetscErrorCode MatRealPart_MPISBAIJ(Mat A) 1100 { 1101 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1102 PetscErrorCode ierr; 1103 1104 PetscFunctionBegin; 1105 ierr = MatRealPart(a->A);CHKERRQ(ierr); 1106 ierr = MatRealPart(a->B);CHKERRQ(ierr); 1107 PetscFunctionReturn(0); 1108 } 1109 1110 #undef __FUNCT__ 1111 #define __FUNCT__ "MatImaginaryPart_MPISBAIJ" 1112 PetscErrorCode MatImaginaryPart_MPISBAIJ(Mat A) 1113 { 1114 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1115 PetscErrorCode ierr; 1116 1117 PetscFunctionBegin; 1118 ierr = MatImaginaryPart(a->A);CHKERRQ(ierr); 1119 ierr = MatImaginaryPart(a->B);CHKERRQ(ierr); 1120 PetscFunctionReturn(0); 1121 } 1122 1123 #undef __FUNCT__ 1124 #define __FUNCT__ "MatZeroEntries_MPISBAIJ" 1125 PetscErrorCode MatZeroEntries_MPISBAIJ(Mat A) 1126 { 1127 Mat_MPISBAIJ *l = (Mat_MPISBAIJ*)A->data; 1128 PetscErrorCode ierr; 1129 1130 PetscFunctionBegin; 1131 ierr = MatZeroEntries(l->A);CHKERRQ(ierr); 1132 ierr = MatZeroEntries(l->B);CHKERRQ(ierr); 1133 PetscFunctionReturn(0); 1134 } 1135 1136 #undef __FUNCT__ 1137 #define __FUNCT__ "MatGetInfo_MPISBAIJ" 1138 PetscErrorCode MatGetInfo_MPISBAIJ(Mat matin,MatInfoType flag,MatInfo *info) 1139 { 1140 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)matin->data; 1141 Mat A = a->A,B = a->B; 1142 PetscErrorCode ierr; 1143 PetscReal isend[5],irecv[5]; 1144 1145 PetscFunctionBegin; 1146 info->block_size = (PetscReal)matin->rmap->bs; 1147 1148 ierr = MatGetInfo(A,MAT_LOCAL,info);CHKERRQ(ierr); 1149 1150 isend[0] = info->nz_used; isend[1] = info->nz_allocated; isend[2] = info->nz_unneeded; 1151 isend[3] = info->memory; isend[4] = info->mallocs; 1152 1153 ierr = MatGetInfo(B,MAT_LOCAL,info);CHKERRQ(ierr); 1154 1155 isend[0] += info->nz_used; isend[1] += info->nz_allocated; isend[2] += info->nz_unneeded; 1156 isend[3] += info->memory; isend[4] += info->mallocs; 1157 if (flag == MAT_LOCAL) { 1158 info->nz_used = isend[0]; 1159 info->nz_allocated = isend[1]; 1160 info->nz_unneeded = isend[2]; 1161 info->memory = isend[3]; 1162 info->mallocs = isend[4]; 1163 } else if (flag == MAT_GLOBAL_MAX) { 1164 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPIU_MAX,PetscObjectComm((PetscObject)matin));CHKERRQ(ierr); 1165 1166 info->nz_used = irecv[0]; 1167 info->nz_allocated = irecv[1]; 1168 info->nz_unneeded = irecv[2]; 1169 info->memory = irecv[3]; 1170 info->mallocs = irecv[4]; 1171 } else if (flag == MAT_GLOBAL_SUM) { 1172 ierr = MPI_Allreduce(isend,irecv,5,MPIU_REAL,MPIU_SUM,PetscObjectComm((PetscObject)matin));CHKERRQ(ierr); 1173 1174 info->nz_used = irecv[0]; 1175 info->nz_allocated = irecv[1]; 1176 info->nz_unneeded = irecv[2]; 1177 info->memory = irecv[3]; 1178 info->mallocs = irecv[4]; 1179 } else SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Unknown MatInfoType argument %d",(int)flag); 1180 info->fill_ratio_given = 0; /* no parallel LU/ILU/Cholesky */ 1181 info->fill_ratio_needed = 0; 1182 info->factor_mallocs = 0; 1183 PetscFunctionReturn(0); 1184 } 1185 1186 #undef __FUNCT__ 1187 #define __FUNCT__ "MatSetOption_MPISBAIJ" 1188 PetscErrorCode MatSetOption_MPISBAIJ(Mat A,MatOption op,PetscBool flg) 1189 { 1190 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1191 Mat_SeqSBAIJ *aA = (Mat_SeqSBAIJ*)a->A->data; 1192 PetscErrorCode ierr; 1193 1194 PetscFunctionBegin; 1195 switch (op) { 1196 case MAT_NEW_NONZERO_LOCATIONS: 1197 case MAT_NEW_NONZERO_ALLOCATION_ERR: 1198 case MAT_UNUSED_NONZERO_LOCATION_ERR: 1199 case MAT_KEEP_NONZERO_PATTERN: 1200 case MAT_NEW_NONZERO_LOCATION_ERR: 1201 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1202 ierr = MatSetOption(a->B,op,flg);CHKERRQ(ierr); 1203 break; 1204 case MAT_ROW_ORIENTED: 1205 a->roworiented = flg; 1206 1207 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1208 ierr = MatSetOption(a->B,op,flg);CHKERRQ(ierr); 1209 break; 1210 case MAT_NEW_DIAGONALS: 1211 ierr = PetscInfo1(A,"Option %s ignored\n",MatOptions[op]);CHKERRQ(ierr); 1212 break; 1213 case MAT_IGNORE_OFF_PROC_ENTRIES: 1214 a->donotstash = flg; 1215 break; 1216 case MAT_USE_HASH_TABLE: 1217 a->ht_flag = flg; 1218 break; 1219 case MAT_HERMITIAN: 1220 if (!A->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Must call MatAssemblyEnd() first"); 1221 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1222 1223 A->ops->mult = MatMult_MPISBAIJ_Hermitian; 1224 break; 1225 case MAT_SPD: 1226 A->spd_set = PETSC_TRUE; 1227 A->spd = flg; 1228 if (flg) { 1229 A->symmetric = PETSC_TRUE; 1230 A->structurally_symmetric = PETSC_TRUE; 1231 A->symmetric_set = PETSC_TRUE; 1232 A->structurally_symmetric_set = PETSC_TRUE; 1233 } 1234 break; 1235 case MAT_SYMMETRIC: 1236 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1237 break; 1238 case MAT_STRUCTURALLY_SYMMETRIC: 1239 ierr = MatSetOption(a->A,op,flg);CHKERRQ(ierr); 1240 break; 1241 case MAT_SYMMETRY_ETERNAL: 1242 if (!flg) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix must be symmetric"); 1243 ierr = PetscInfo1(A,"Option %s ignored\n",MatOptions[op]);CHKERRQ(ierr); 1244 break; 1245 case MAT_IGNORE_LOWER_TRIANGULAR: 1246 aA->ignore_ltriangular = flg; 1247 break; 1248 case MAT_ERROR_LOWER_TRIANGULAR: 1249 aA->ignore_ltriangular = flg; 1250 break; 1251 case MAT_GETROW_UPPERTRIANGULAR: 1252 aA->getrow_utriangular = flg; 1253 break; 1254 default: 1255 SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"unknown option %d",op); 1256 } 1257 PetscFunctionReturn(0); 1258 } 1259 1260 #undef __FUNCT__ 1261 #define __FUNCT__ "MatTranspose_MPISBAIJ" 1262 PetscErrorCode MatTranspose_MPISBAIJ(Mat A,MatReuse reuse,Mat *B) 1263 { 1264 PetscErrorCode ierr; 1265 1266 PetscFunctionBegin; 1267 if (MAT_INITIAL_MATRIX || *B != A) { 1268 ierr = MatDuplicate(A,MAT_COPY_VALUES,B);CHKERRQ(ierr); 1269 } 1270 PetscFunctionReturn(0); 1271 } 1272 1273 #undef __FUNCT__ 1274 #define __FUNCT__ "MatDiagonalScale_MPISBAIJ" 1275 PetscErrorCode MatDiagonalScale_MPISBAIJ(Mat mat,Vec ll,Vec rr) 1276 { 1277 Mat_MPISBAIJ *baij = (Mat_MPISBAIJ*)mat->data; 1278 Mat a = baij->A, b=baij->B; 1279 PetscErrorCode ierr; 1280 PetscInt nv,m,n; 1281 PetscBool flg; 1282 1283 PetscFunctionBegin; 1284 if (ll != rr) { 1285 ierr = VecEqual(ll,rr,&flg);CHKERRQ(ierr); 1286 if (!flg) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"For symmetric format, left and right scaling vectors must be same\n"); 1287 } 1288 if (!ll) PetscFunctionReturn(0); 1289 1290 ierr = MatGetLocalSize(mat,&m,&n);CHKERRQ(ierr); 1291 if (m != n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"For symmetric format, local size %d %d must be same",m,n); 1292 1293 ierr = VecGetLocalSize(rr,&nv);CHKERRQ(ierr); 1294 if (nv!=n) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Left and right vector non-conforming local size"); 1295 1296 ierr = VecScatterBegin(baij->Mvctx,rr,baij->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1297 1298 /* left diagonalscale the off-diagonal part */ 1299 ierr = (*b->ops->diagonalscale)(b,ll,NULL);CHKERRQ(ierr); 1300 1301 /* scale the diagonal part */ 1302 ierr = (*a->ops->diagonalscale)(a,ll,rr);CHKERRQ(ierr); 1303 1304 /* right diagonalscale the off-diagonal part */ 1305 ierr = VecScatterEnd(baij->Mvctx,rr,baij->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1306 ierr = (*b->ops->diagonalscale)(b,NULL,baij->lvec);CHKERRQ(ierr); 1307 PetscFunctionReturn(0); 1308 } 1309 1310 #undef __FUNCT__ 1311 #define __FUNCT__ "MatSetUnfactored_MPISBAIJ" 1312 PetscErrorCode MatSetUnfactored_MPISBAIJ(Mat A) 1313 { 1314 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1315 PetscErrorCode ierr; 1316 1317 PetscFunctionBegin; 1318 ierr = MatSetUnfactored(a->A);CHKERRQ(ierr); 1319 PetscFunctionReturn(0); 1320 } 1321 1322 static PetscErrorCode MatDuplicate_MPISBAIJ(Mat,MatDuplicateOption,Mat*); 1323 1324 #undef __FUNCT__ 1325 #define __FUNCT__ "MatEqual_MPISBAIJ" 1326 PetscErrorCode MatEqual_MPISBAIJ(Mat A,Mat B,PetscBool *flag) 1327 { 1328 Mat_MPISBAIJ *matB = (Mat_MPISBAIJ*)B->data,*matA = (Mat_MPISBAIJ*)A->data; 1329 Mat a,b,c,d; 1330 PetscBool flg; 1331 PetscErrorCode ierr; 1332 1333 PetscFunctionBegin; 1334 a = matA->A; b = matA->B; 1335 c = matB->A; d = matB->B; 1336 1337 ierr = MatEqual(a,c,&flg);CHKERRQ(ierr); 1338 if (flg) { 1339 ierr = MatEqual(b,d,&flg);CHKERRQ(ierr); 1340 } 1341 ierr = MPI_Allreduce(&flg,flag,1,MPIU_BOOL,MPI_LAND,PetscObjectComm((PetscObject)A));CHKERRQ(ierr); 1342 PetscFunctionReturn(0); 1343 } 1344 1345 #undef __FUNCT__ 1346 #define __FUNCT__ "MatCopy_MPISBAIJ" 1347 PetscErrorCode MatCopy_MPISBAIJ(Mat A,Mat B,MatStructure str) 1348 { 1349 PetscErrorCode ierr; 1350 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 1351 Mat_MPISBAIJ *b = (Mat_MPISBAIJ*)B->data; 1352 1353 PetscFunctionBegin; 1354 /* If the two matrices don't have the same copy implementation, they aren't compatible for fast copy. */ 1355 if ((str != SAME_NONZERO_PATTERN) || (A->ops->copy != B->ops->copy)) { 1356 ierr = MatGetRowUpperTriangular(A);CHKERRQ(ierr); 1357 ierr = MatCopy_Basic(A,B,str);CHKERRQ(ierr); 1358 ierr = MatRestoreRowUpperTriangular(A);CHKERRQ(ierr); 1359 } else { 1360 ierr = MatCopy(a->A,b->A,str);CHKERRQ(ierr); 1361 ierr = MatCopy(a->B,b->B,str);CHKERRQ(ierr); 1362 } 1363 PetscFunctionReturn(0); 1364 } 1365 1366 #undef __FUNCT__ 1367 #define __FUNCT__ "MatSetUp_MPISBAIJ" 1368 PetscErrorCode MatSetUp_MPISBAIJ(Mat A) 1369 { 1370 PetscErrorCode ierr; 1371 1372 PetscFunctionBegin; 1373 ierr = MatMPISBAIJSetPreallocation(A,A->rmap->bs,PETSC_DEFAULT,0,PETSC_DEFAULT,0);CHKERRQ(ierr); 1374 PetscFunctionReturn(0); 1375 } 1376 1377 #undef __FUNCT__ 1378 #define __FUNCT__ "MatAXPY_MPISBAIJ" 1379 PetscErrorCode MatAXPY_MPISBAIJ(Mat Y,PetscScalar a,Mat X,MatStructure str) 1380 { 1381 PetscErrorCode ierr; 1382 Mat_MPISBAIJ *xx=(Mat_MPISBAIJ*)X->data,*yy=(Mat_MPISBAIJ*)Y->data; 1383 PetscBLASInt bnz,one=1; 1384 Mat_SeqSBAIJ *xa,*ya; 1385 Mat_SeqBAIJ *xb,*yb; 1386 1387 PetscFunctionBegin; 1388 if (str == SAME_NONZERO_PATTERN) { 1389 PetscScalar alpha = a; 1390 xa = (Mat_SeqSBAIJ*)xx->A->data; 1391 ya = (Mat_SeqSBAIJ*)yy->A->data; 1392 ierr = PetscBLASIntCast(xa->nz,&bnz);CHKERRQ(ierr); 1393 PetscStackCallBLAS("BLASaxpy",BLASaxpy_(&bnz,&alpha,xa->a,&one,ya->a,&one)); 1394 xb = (Mat_SeqBAIJ*)xx->B->data; 1395 yb = (Mat_SeqBAIJ*)yy->B->data; 1396 ierr = PetscBLASIntCast(xb->nz,&bnz);CHKERRQ(ierr); 1397 PetscStackCallBLAS("BLASaxpy",BLASaxpy_(&bnz,&alpha,xb->a,&one,yb->a,&one)); 1398 } else { 1399 ierr = MatGetRowUpperTriangular(X);CHKERRQ(ierr); 1400 ierr = MatAXPY_Basic(Y,a,X,str);CHKERRQ(ierr); 1401 ierr = MatRestoreRowUpperTriangular(X);CHKERRQ(ierr); 1402 } 1403 PetscFunctionReturn(0); 1404 } 1405 1406 #undef __FUNCT__ 1407 #define __FUNCT__ "MatGetSubMatrices_MPISBAIJ" 1408 PetscErrorCode MatGetSubMatrices_MPISBAIJ(Mat A,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *B[]) 1409 { 1410 PetscErrorCode ierr; 1411 PetscInt i; 1412 PetscBool flg; 1413 1414 PetscFunctionBegin; 1415 for (i=0; i<n; i++) { 1416 ierr = ISEqual(irow[i],icol[i],&flg);CHKERRQ(ierr); 1417 if (!flg) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only get symmetric submatrix for MPISBAIJ matrices"); 1418 } 1419 ierr = MatGetSubMatrices_MPIBAIJ(A,n,irow,icol,scall,B);CHKERRQ(ierr); 1420 PetscFunctionReturn(0); 1421 } 1422 1423 1424 /* -------------------------------------------------------------------*/ 1425 static struct _MatOps MatOps_Values = {MatSetValues_MPISBAIJ, 1426 MatGetRow_MPISBAIJ, 1427 MatRestoreRow_MPISBAIJ, 1428 MatMult_MPISBAIJ, 1429 /* 4*/ MatMultAdd_MPISBAIJ, 1430 MatMult_MPISBAIJ, /* transpose versions are same as non-transpose */ 1431 MatMultAdd_MPISBAIJ, 1432 0, 1433 0, 1434 0, 1435 /* 10*/ 0, 1436 0, 1437 0, 1438 MatSOR_MPISBAIJ, 1439 MatTranspose_MPISBAIJ, 1440 /* 15*/ MatGetInfo_MPISBAIJ, 1441 MatEqual_MPISBAIJ, 1442 MatGetDiagonal_MPISBAIJ, 1443 MatDiagonalScale_MPISBAIJ, 1444 MatNorm_MPISBAIJ, 1445 /* 20*/ MatAssemblyBegin_MPISBAIJ, 1446 MatAssemblyEnd_MPISBAIJ, 1447 MatSetOption_MPISBAIJ, 1448 MatZeroEntries_MPISBAIJ, 1449 /* 24*/ 0, 1450 0, 1451 0, 1452 0, 1453 0, 1454 /* 29*/ MatSetUp_MPISBAIJ, 1455 0, 1456 0, 1457 0, 1458 0, 1459 /* 34*/ MatDuplicate_MPISBAIJ, 1460 0, 1461 0, 1462 0, 1463 0, 1464 /* 39*/ MatAXPY_MPISBAIJ, 1465 MatGetSubMatrices_MPISBAIJ, 1466 MatIncreaseOverlap_MPISBAIJ, 1467 MatGetValues_MPISBAIJ, 1468 MatCopy_MPISBAIJ, 1469 /* 44*/ 0, 1470 MatScale_MPISBAIJ, 1471 0, 1472 0, 1473 0, 1474 /* 49*/ 0, 1475 0, 1476 0, 1477 0, 1478 0, 1479 /* 54*/ 0, 1480 0, 1481 MatSetUnfactored_MPISBAIJ, 1482 0, 1483 MatSetValuesBlocked_MPISBAIJ, 1484 /* 59*/ 0, 1485 0, 1486 0, 1487 0, 1488 0, 1489 /* 64*/ 0, 1490 0, 1491 0, 1492 0, 1493 0, 1494 /* 69*/ MatGetRowMaxAbs_MPISBAIJ, 1495 0, 1496 0, 1497 0, 1498 0, 1499 /* 74*/ 0, 1500 0, 1501 0, 1502 0, 1503 0, 1504 /* 79*/ 0, 1505 0, 1506 0, 1507 0, 1508 MatLoad_MPISBAIJ, 1509 /* 84*/ 0, 1510 0, 1511 0, 1512 0, 1513 0, 1514 /* 89*/ 0, 1515 0, 1516 0, 1517 0, 1518 0, 1519 /* 94*/ 0, 1520 0, 1521 0, 1522 0, 1523 0, 1524 /* 99*/ 0, 1525 0, 1526 0, 1527 0, 1528 0, 1529 /*104*/ 0, 1530 MatRealPart_MPISBAIJ, 1531 MatImaginaryPart_MPISBAIJ, 1532 MatGetRowUpperTriangular_MPISBAIJ, 1533 MatRestoreRowUpperTriangular_MPISBAIJ, 1534 /*109*/ 0, 1535 0, 1536 0, 1537 0, 1538 0, 1539 /*114*/ 0, 1540 0, 1541 0, 1542 0, 1543 0, 1544 /*119*/ 0, 1545 0, 1546 0, 1547 0, 1548 0, 1549 /*124*/ 0, 1550 0, 1551 0, 1552 0, 1553 0, 1554 /*129*/ 0, 1555 0, 1556 0, 1557 0, 1558 0, 1559 /*134*/ 0, 1560 0, 1561 0, 1562 0, 1563 0, 1564 /*139*/ 0, 1565 0, 1566 0 1567 }; 1568 1569 1570 #undef __FUNCT__ 1571 #define __FUNCT__ "MatGetDiagonalBlock_MPISBAIJ" 1572 PetscErrorCode MatGetDiagonalBlock_MPISBAIJ(Mat A,Mat *a) 1573 { 1574 PetscFunctionBegin; 1575 *a = ((Mat_MPISBAIJ*)A->data)->A; 1576 PetscFunctionReturn(0); 1577 } 1578 1579 #undef __FUNCT__ 1580 #define __FUNCT__ "MatMPISBAIJSetPreallocation_MPISBAIJ" 1581 PetscErrorCode MatMPISBAIJSetPreallocation_MPISBAIJ(Mat B,PetscInt bs,PetscInt d_nz,const PetscInt *d_nnz,PetscInt o_nz,const PetscInt *o_nnz) 1582 { 1583 Mat_MPISBAIJ *b; 1584 PetscErrorCode ierr; 1585 PetscInt i,mbs,Mbs; 1586 1587 PetscFunctionBegin; 1588 ierr = MatSetBlockSize(B,PetscAbs(bs));CHKERRQ(ierr); 1589 ierr = PetscLayoutSetUp(B->rmap);CHKERRQ(ierr); 1590 ierr = PetscLayoutSetUp(B->cmap);CHKERRQ(ierr); 1591 ierr = PetscLayoutGetBlockSize(B->rmap,&bs);CHKERRQ(ierr); 1592 1593 b = (Mat_MPISBAIJ*)B->data; 1594 mbs = B->rmap->n/bs; 1595 Mbs = B->rmap->N/bs; 1596 if (mbs*bs != B->rmap->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"No of local rows %D must be divisible by blocksize %D",B->rmap->N,bs); 1597 1598 B->rmap->bs = bs; 1599 b->bs2 = bs*bs; 1600 b->mbs = mbs; 1601 b->nbs = mbs; 1602 b->Mbs = Mbs; 1603 b->Nbs = Mbs; 1604 1605 for (i=0; i<=b->size; i++) { 1606 b->rangebs[i] = B->rmap->range[i]/bs; 1607 } 1608 b->rstartbs = B->rmap->rstart/bs; 1609 b->rendbs = B->rmap->rend/bs; 1610 1611 b->cstartbs = B->cmap->rstart/bs; 1612 b->cendbs = B->cmap->rend/bs; 1613 1614 if (!B->preallocated) { 1615 ierr = MatCreate(PETSC_COMM_SELF,&b->A);CHKERRQ(ierr); 1616 ierr = MatSetSizes(b->A,B->rmap->n,B->cmap->n,B->rmap->n,B->cmap->n);CHKERRQ(ierr); 1617 ierr = MatSetType(b->A,MATSEQSBAIJ);CHKERRQ(ierr); 1618 ierr = PetscLogObjectParent((PetscObject)B,(PetscObject)b->A);CHKERRQ(ierr); 1619 ierr = MatCreate(PETSC_COMM_SELF,&b->B);CHKERRQ(ierr); 1620 ierr = MatSetSizes(b->B,B->rmap->n,B->cmap->N,B->rmap->n,B->cmap->N);CHKERRQ(ierr); 1621 ierr = MatSetType(b->B,MATSEQBAIJ);CHKERRQ(ierr); 1622 ierr = PetscLogObjectParent((PetscObject)B,(PetscObject)b->B);CHKERRQ(ierr); 1623 ierr = MatStashCreate_Private(PetscObjectComm((PetscObject)B),bs,&B->bstash);CHKERRQ(ierr); 1624 } 1625 1626 ierr = MatSeqSBAIJSetPreallocation(b->A,bs,d_nz,d_nnz);CHKERRQ(ierr); 1627 ierr = MatSeqBAIJSetPreallocation(b->B,bs,o_nz,o_nnz);CHKERRQ(ierr); 1628 1629 B->preallocated = PETSC_TRUE; 1630 PetscFunctionReturn(0); 1631 } 1632 1633 #undef __FUNCT__ 1634 #define __FUNCT__ "MatMPISBAIJSetPreallocationCSR_MPISBAIJ" 1635 PetscErrorCode MatMPISBAIJSetPreallocationCSR_MPISBAIJ(Mat B,PetscInt bs,const PetscInt ii[],const PetscInt jj[],const PetscScalar V[]) 1636 { 1637 PetscInt m,rstart,cstart,cend; 1638 PetscInt i,j,d,nz,nz_max=0,*d_nnz=0,*o_nnz=0; 1639 const PetscInt *JJ =0; 1640 PetscScalar *values=0; 1641 PetscErrorCode ierr; 1642 1643 PetscFunctionBegin; 1644 if (bs < 1) SETERRQ1(PetscObjectComm((PetscObject)B),PETSC_ERR_ARG_OUTOFRANGE,"Invalid block size specified, must be positive but it is %D",bs); 1645 ierr = PetscLayoutSetBlockSize(B->rmap,bs);CHKERRQ(ierr); 1646 ierr = PetscLayoutSetBlockSize(B->cmap,bs);CHKERRQ(ierr); 1647 ierr = PetscLayoutSetUp(B->rmap);CHKERRQ(ierr); 1648 ierr = PetscLayoutSetUp(B->cmap);CHKERRQ(ierr); 1649 ierr = PetscLayoutGetBlockSize(B->rmap,&bs);CHKERRQ(ierr); 1650 m = B->rmap->n/bs; 1651 rstart = B->rmap->rstart/bs; 1652 cstart = B->cmap->rstart/bs; 1653 cend = B->cmap->rend/bs; 1654 1655 if (ii[0]) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"ii[0] must be 0 but it is %D",ii[0]); 1656 ierr = PetscMalloc2(m,&d_nnz,m,&o_nnz);CHKERRQ(ierr); 1657 for (i=0; i<m; i++) { 1658 nz = ii[i+1] - ii[i]; 1659 if (nz < 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Local row %D has a negative number of columns %D",i,nz); 1660 nz_max = PetscMax(nz_max,nz); 1661 JJ = jj + ii[i]; 1662 for (j=0; j<nz; j++) { 1663 if (*JJ >= cstart) break; 1664 JJ++; 1665 } 1666 d = 0; 1667 for (; j<nz; j++) { 1668 if (*JJ++ >= cend) break; 1669 d++; 1670 } 1671 d_nnz[i] = d; 1672 o_nnz[i] = nz - d; 1673 } 1674 ierr = MatMPISBAIJSetPreallocation(B,bs,0,d_nnz,0,o_nnz);CHKERRQ(ierr); 1675 ierr = PetscFree2(d_nnz,o_nnz);CHKERRQ(ierr); 1676 1677 values = (PetscScalar*)V; 1678 if (!values) { 1679 ierr = PetscMalloc1(bs*bs*nz_max,&values);CHKERRQ(ierr); 1680 ierr = PetscMemzero(values,bs*bs*nz_max*sizeof(PetscScalar));CHKERRQ(ierr); 1681 } 1682 for (i=0; i<m; i++) { 1683 PetscInt row = i + rstart; 1684 PetscInt ncols = ii[i+1] - ii[i]; 1685 const PetscInt *icols = jj + ii[i]; 1686 const PetscScalar *svals = values + (V ? (bs*bs*ii[i]) : 0); 1687 ierr = MatSetValuesBlocked_MPISBAIJ(B,1,&row,ncols,icols,svals,INSERT_VALUES);CHKERRQ(ierr); 1688 } 1689 1690 if (!V) { ierr = PetscFree(values);CHKERRQ(ierr); } 1691 ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1692 ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1693 ierr = MatSetOption(B,MAT_NEW_NONZERO_LOCATION_ERR,PETSC_TRUE);CHKERRQ(ierr); 1694 PetscFunctionReturn(0); 1695 } 1696 1697 #if defined(PETSC_HAVE_MUMPS) 1698 PETSC_EXTERN PetscErrorCode MatGetFactor_sbaij_mumps(Mat,MatFactorType,Mat*); 1699 #endif 1700 #if defined(PETSC_HAVE_PASTIX) 1701 PETSC_EXTERN PetscErrorCode MatGetFactor_mpisbaij_pastix(Mat,MatFactorType,Mat*); 1702 #endif 1703 1704 /*MC 1705 MATMPISBAIJ - MATMPISBAIJ = "mpisbaij" - A matrix type to be used for distributed symmetric sparse block matrices, 1706 based on block compressed sparse row format. Only the upper triangular portion of the "diagonal" portion of 1707 the matrix is stored. 1708 1709 For complex numbers by default this matrix is symmetric, NOT Hermitian symmetric. To make it Hermitian symmetric you 1710 can call MatSetOption(Mat, MAT_HERMITIAN); 1711 1712 Options Database Keys: 1713 . -mat_type mpisbaij - sets the matrix type to "mpisbaij" during a call to MatSetFromOptions() 1714 1715 Level: beginner 1716 1717 .seealso: MatCreateMPISBAIJ 1718 M*/ 1719 1720 PETSC_EXTERN PetscErrorCode MatConvert_MPISBAIJ_MPISBSTRM(Mat,MatType,MatReuse,Mat*); 1721 1722 #undef __FUNCT__ 1723 #define __FUNCT__ "MatCreate_MPISBAIJ" 1724 PETSC_EXTERN PetscErrorCode MatCreate_MPISBAIJ(Mat B) 1725 { 1726 Mat_MPISBAIJ *b; 1727 PetscErrorCode ierr; 1728 PetscBool flg; 1729 1730 PetscFunctionBegin; 1731 ierr = PetscNewLog(B,&b);CHKERRQ(ierr); 1732 B->data = (void*)b; 1733 ierr = PetscMemcpy(B->ops,&MatOps_Values,sizeof(struct _MatOps));CHKERRQ(ierr); 1734 1735 B->ops->destroy = MatDestroy_MPISBAIJ; 1736 B->ops->view = MatView_MPISBAIJ; 1737 B->assembled = PETSC_FALSE; 1738 B->insertmode = NOT_SET_VALUES; 1739 1740 ierr = MPI_Comm_rank(PetscObjectComm((PetscObject)B),&b->rank);CHKERRQ(ierr); 1741 ierr = MPI_Comm_size(PetscObjectComm((PetscObject)B),&b->size);CHKERRQ(ierr); 1742 1743 /* build local table of row and column ownerships */ 1744 ierr = PetscMalloc1((b->size+2),&b->rangebs);CHKERRQ(ierr); 1745 1746 /* build cache for off array entries formed */ 1747 ierr = MatStashCreate_Private(PetscObjectComm((PetscObject)B),1,&B->stash);CHKERRQ(ierr); 1748 1749 b->donotstash = PETSC_FALSE; 1750 b->colmap = NULL; 1751 b->garray = NULL; 1752 b->roworiented = PETSC_TRUE; 1753 1754 /* stuff used in block assembly */ 1755 b->barray = 0; 1756 1757 /* stuff used for matrix vector multiply */ 1758 b->lvec = 0; 1759 b->Mvctx = 0; 1760 b->slvec0 = 0; 1761 b->slvec0b = 0; 1762 b->slvec1 = 0; 1763 b->slvec1a = 0; 1764 b->slvec1b = 0; 1765 b->sMvctx = 0; 1766 1767 /* stuff for MatGetRow() */ 1768 b->rowindices = 0; 1769 b->rowvalues = 0; 1770 b->getrowactive = PETSC_FALSE; 1771 1772 /* hash table stuff */ 1773 b->ht = 0; 1774 b->hd = 0; 1775 b->ht_size = 0; 1776 b->ht_flag = PETSC_FALSE; 1777 b->ht_fact = 0; 1778 b->ht_total_ct = 0; 1779 b->ht_insert_ct = 0; 1780 1781 /* stuff for MatGetSubMatrices_MPIBAIJ_local() */ 1782 b->ijonly = PETSC_FALSE; 1783 1784 b->in_loc = 0; 1785 b->v_loc = 0; 1786 b->n_loc = 0; 1787 ierr = PetscOptionsBegin(PetscObjectComm((PetscObject)B),NULL,"Options for loading MPISBAIJ matrix 1","Mat");CHKERRQ(ierr); 1788 ierr = PetscOptionsBool("-mat_use_hash_table","Use hash table to save memory in constructing matrix","MatSetOption",PETSC_FALSE,&flg,NULL);CHKERRQ(ierr); 1789 if (flg) { 1790 PetscReal fact = 1.39; 1791 ierr = MatSetOption(B,MAT_USE_HASH_TABLE,PETSC_TRUE);CHKERRQ(ierr); 1792 ierr = PetscOptionsReal("-mat_use_hash_table","Use hash table factor","MatMPIBAIJSetHashTableFactor",fact,&fact,NULL);CHKERRQ(ierr); 1793 if (fact <= 1.0) fact = 1.39; 1794 ierr = MatMPIBAIJSetHashTableFactor(B,fact);CHKERRQ(ierr); 1795 ierr = PetscInfo1(B,"Hash table Factor used %5.2f\n",fact);CHKERRQ(ierr); 1796 } 1797 ierr = PetscOptionsEnd();CHKERRQ(ierr); 1798 1799 #if defined(PETSC_HAVE_PASTIX) 1800 ierr = PetscObjectComposeFunction((PetscObject)B,"MatGetFactor_pastix_C",MatGetFactor_mpisbaij_pastix);CHKERRQ(ierr); 1801 #endif 1802 #if defined(PETSC_HAVE_MUMPS) 1803 ierr = PetscObjectComposeFunction((PetscObject)B,"MatGetFactor_mumps_C",MatGetFactor_sbaij_mumps);CHKERRQ(ierr); 1804 #endif 1805 ierr = PetscObjectComposeFunction((PetscObject)B,"MatStoreValues_C",MatStoreValues_MPISBAIJ);CHKERRQ(ierr); 1806 ierr = PetscObjectComposeFunction((PetscObject)B,"MatRetrieveValues_C",MatRetrieveValues_MPISBAIJ);CHKERRQ(ierr); 1807 ierr = PetscObjectComposeFunction((PetscObject)B,"MatGetDiagonalBlock_C",MatGetDiagonalBlock_MPISBAIJ);CHKERRQ(ierr); 1808 ierr = PetscObjectComposeFunction((PetscObject)B,"MatMPISBAIJSetPreallocation_C",MatMPISBAIJSetPreallocation_MPISBAIJ);CHKERRQ(ierr); 1809 ierr = PetscObjectComposeFunction((PetscObject)B,"MatMPISBAIJSetPreallocationCSR_C",MatMPISBAIJSetPreallocationCSR_MPISBAIJ);CHKERRQ(ierr); 1810 ierr = PetscObjectComposeFunction((PetscObject)B,"MatConvert_mpisbaij_mpisbstrm_C",MatConvert_MPISBAIJ_MPISBSTRM);CHKERRQ(ierr); 1811 1812 B->symmetric = PETSC_TRUE; 1813 B->structurally_symmetric = PETSC_TRUE; 1814 B->symmetric_set = PETSC_TRUE; 1815 B->structurally_symmetric_set = PETSC_TRUE; 1816 1817 ierr = PetscObjectChangeTypeName((PetscObject)B,MATMPISBAIJ);CHKERRQ(ierr); 1818 PetscFunctionReturn(0); 1819 } 1820 1821 /*MC 1822 MATSBAIJ - MATSBAIJ = "sbaij" - A matrix type to be used for symmetric block sparse matrices. 1823 1824 This matrix type is identical to MATSEQSBAIJ when constructed with a single process communicator, 1825 and MATMPISBAIJ otherwise. 1826 1827 Options Database Keys: 1828 . -mat_type sbaij - sets the matrix type to "sbaij" during a call to MatSetFromOptions() 1829 1830 Level: beginner 1831 1832 .seealso: MatCreateMPISBAIJ,MATSEQSBAIJ,MATMPISBAIJ 1833 M*/ 1834 1835 #undef __FUNCT__ 1836 #define __FUNCT__ "MatMPISBAIJSetPreallocation" 1837 /*@C 1838 MatMPISBAIJSetPreallocation - For good matrix assembly performance 1839 the user should preallocate the matrix storage by setting the parameters 1840 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 1841 performance can be increased by more than a factor of 50. 1842 1843 Collective on Mat 1844 1845 Input Parameters: 1846 + A - the matrix 1847 . bs - size of blockk 1848 . d_nz - number of block nonzeros per block row in diagonal portion of local 1849 submatrix (same for all local rows) 1850 . d_nnz - array containing the number of block nonzeros in the various block rows 1851 in the upper triangular and diagonal part of the in diagonal portion of the local 1852 (possibly different for each block row) or NULL. If you plan to factor the matrix you must leave room 1853 for the diagonal entry and set a value even if it is zero. 1854 . o_nz - number of block nonzeros per block row in the off-diagonal portion of local 1855 submatrix (same for all local rows). 1856 - o_nnz - array containing the number of nonzeros in the various block rows of the 1857 off-diagonal portion of the local submatrix that is right of the diagonal 1858 (possibly different for each block row) or NULL. 1859 1860 1861 Options Database Keys: 1862 . -mat_no_unroll - uses code that does not unroll the loops in the 1863 block calculations (much slower) 1864 . -mat_block_size - size of the blocks to use 1865 1866 Notes: 1867 1868 If PETSC_DECIDE or PETSC_DETERMINE is used for a particular argument on one processor 1869 than it must be used on all processors that share the object for that argument. 1870 1871 If the *_nnz parameter is given then the *_nz parameter is ignored 1872 1873 Storage Information: 1874 For a square global matrix we define each processor's diagonal portion 1875 to be its local rows and the corresponding columns (a square submatrix); 1876 each processor's off-diagonal portion encompasses the remainder of the 1877 local matrix (a rectangular submatrix). 1878 1879 The user can specify preallocated storage for the diagonal part of 1880 the local submatrix with either d_nz or d_nnz (not both). Set 1881 d_nz=PETSC_DEFAULT and d_nnz=NULL for PETSc to control dynamic 1882 memory allocation. Likewise, specify preallocated storage for the 1883 off-diagonal part of the local submatrix with o_nz or o_nnz (not both). 1884 1885 You can call MatGetInfo() to get information on how effective the preallocation was; 1886 for example the fields mallocs,nz_allocated,nz_used,nz_unneeded; 1887 You can also run with the option -info and look for messages with the string 1888 malloc in them to see if additional memory allocation was needed. 1889 1890 Consider a processor that owns rows 3, 4 and 5 of a parallel matrix. In 1891 the figure below we depict these three local rows and all columns (0-11). 1892 1893 .vb 1894 0 1 2 3 4 5 6 7 8 9 10 11 1895 -------------------------- 1896 row 3 |. . . d d d o o o o o o 1897 row 4 |. . . d d d o o o o o o 1898 row 5 |. . . d d d o o o o o o 1899 -------------------------- 1900 .ve 1901 1902 Thus, any entries in the d locations are stored in the d (diagonal) 1903 submatrix, and any entries in the o locations are stored in the 1904 o (off-diagonal) submatrix. Note that the d matrix is stored in 1905 MatSeqSBAIJ format and the o submatrix in MATSEQBAIJ format. 1906 1907 Now d_nz should indicate the number of block nonzeros per row in the upper triangular 1908 plus the diagonal part of the d matrix, 1909 and o_nz should indicate the number of block nonzeros per row in the o matrix 1910 1911 In general, for PDE problems in which most nonzeros are near the diagonal, 1912 one expects d_nz >> o_nz. For large problems you MUST preallocate memory 1913 or you will get TERRIBLE performance; see the users' manual chapter on 1914 matrices. 1915 1916 Level: intermediate 1917 1918 .keywords: matrix, block, aij, compressed row, sparse, parallel 1919 1920 .seealso: MatCreate(), MatCreateSeqSBAIJ(), MatSetValues(), MatCreateBAIJ(), PetscSplitOwnership() 1921 @*/ 1922 PetscErrorCode MatMPISBAIJSetPreallocation(Mat B,PetscInt bs,PetscInt d_nz,const PetscInt d_nnz[],PetscInt o_nz,const PetscInt o_nnz[]) 1923 { 1924 PetscErrorCode ierr; 1925 1926 PetscFunctionBegin; 1927 PetscValidHeaderSpecific(B,MAT_CLASSID,1); 1928 PetscValidType(B,1); 1929 PetscValidLogicalCollectiveInt(B,bs,2); 1930 ierr = PetscTryMethod(B,"MatMPISBAIJSetPreallocation_C",(Mat,PetscInt,PetscInt,const PetscInt[],PetscInt,const PetscInt[]),(B,bs,d_nz,d_nnz,o_nz,o_nnz));CHKERRQ(ierr); 1931 PetscFunctionReturn(0); 1932 } 1933 1934 #undef __FUNCT__ 1935 #define __FUNCT__ "MatCreateSBAIJ" 1936 /*@C 1937 MatCreateSBAIJ - Creates a sparse parallel matrix in symmetric block AIJ format 1938 (block compressed row). For good matrix assembly performance 1939 the user should preallocate the matrix storage by setting the parameters 1940 d_nz (or d_nnz) and o_nz (or o_nnz). By setting these parameters accurately, 1941 performance can be increased by more than a factor of 50. 1942 1943 Collective on MPI_Comm 1944 1945 Input Parameters: 1946 + comm - MPI communicator 1947 . bs - size of blockk 1948 . m - number of local rows (or PETSC_DECIDE to have calculated if M is given) 1949 This value should be the same as the local size used in creating the 1950 y vector for the matrix-vector product y = Ax. 1951 . n - number of local columns (or PETSC_DECIDE to have calculated if N is given) 1952 This value should be the same as the local size used in creating the 1953 x vector for the matrix-vector product y = Ax. 1954 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 1955 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 1956 . d_nz - number of block nonzeros per block row in diagonal portion of local 1957 submatrix (same for all local rows) 1958 . d_nnz - array containing the number of block nonzeros in the various block rows 1959 in the upper triangular portion of the in diagonal portion of the local 1960 (possibly different for each block block row) or NULL. 1961 If you plan to factor the matrix you must leave room for the diagonal entry and 1962 set its value even if it is zero. 1963 . o_nz - number of block nonzeros per block row in the off-diagonal portion of local 1964 submatrix (same for all local rows). 1965 - o_nnz - array containing the number of nonzeros in the various block rows of the 1966 off-diagonal portion of the local submatrix (possibly different for 1967 each block row) or NULL. 1968 1969 Output Parameter: 1970 . A - the matrix 1971 1972 Options Database Keys: 1973 . -mat_no_unroll - uses code that does not unroll the loops in the 1974 block calculations (much slower) 1975 . -mat_block_size - size of the blocks to use 1976 . -mat_mpi - use the parallel matrix data structures even on one processor 1977 (defaults to using SeqBAIJ format on one processor) 1978 1979 It is recommended that one use the MatCreate(), MatSetType() and/or MatSetFromOptions(), 1980 MatXXXXSetPreallocation() paradgm instead of this routine directly. 1981 [MatXXXXSetPreallocation() is, for example, MatSeqAIJSetPreallocation] 1982 1983 Notes: 1984 The number of rows and columns must be divisible by blocksize. 1985 This matrix type does not support complex Hermitian operation. 1986 1987 The user MUST specify either the local or global matrix dimensions 1988 (possibly both). 1989 1990 If PETSC_DECIDE or PETSC_DETERMINE is used for a particular argument on one processor 1991 than it must be used on all processors that share the object for that argument. 1992 1993 If the *_nnz parameter is given then the *_nz parameter is ignored 1994 1995 Storage Information: 1996 For a square global matrix we define each processor's diagonal portion 1997 to be its local rows and the corresponding columns (a square submatrix); 1998 each processor's off-diagonal portion encompasses the remainder of the 1999 local matrix (a rectangular submatrix). 2000 2001 The user can specify preallocated storage for the diagonal part of 2002 the local submatrix with either d_nz or d_nnz (not both). Set 2003 d_nz=PETSC_DEFAULT and d_nnz=NULL for PETSc to control dynamic 2004 memory allocation. Likewise, specify preallocated storage for the 2005 off-diagonal part of the local submatrix with o_nz or o_nnz (not both). 2006 2007 Consider a processor that owns rows 3, 4 and 5 of a parallel matrix. In 2008 the figure below we depict these three local rows and all columns (0-11). 2009 2010 .vb 2011 0 1 2 3 4 5 6 7 8 9 10 11 2012 -------------------------- 2013 row 3 |. . . d d d o o o o o o 2014 row 4 |. . . d d d o o o o o o 2015 row 5 |. . . d d d o o o o o o 2016 -------------------------- 2017 .ve 2018 2019 Thus, any entries in the d locations are stored in the d (diagonal) 2020 submatrix, and any entries in the o locations are stored in the 2021 o (off-diagonal) submatrix. Note that the d matrix is stored in 2022 MatSeqSBAIJ format and the o submatrix in MATSEQBAIJ format. 2023 2024 Now d_nz should indicate the number of block nonzeros per row in the upper triangular 2025 plus the diagonal part of the d matrix, 2026 and o_nz should indicate the number of block nonzeros per row in the o matrix. 2027 In general, for PDE problems in which most nonzeros are near the diagonal, 2028 one expects d_nz >> o_nz. For large problems you MUST preallocate memory 2029 or you will get TERRIBLE performance; see the users' manual chapter on 2030 matrices. 2031 2032 Level: intermediate 2033 2034 .keywords: matrix, block, aij, compressed row, sparse, parallel 2035 2036 .seealso: MatCreate(), MatCreateSeqSBAIJ(), MatSetValues(), MatCreateBAIJ() 2037 @*/ 2038 2039 PetscErrorCode MatCreateSBAIJ(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) 2040 { 2041 PetscErrorCode ierr; 2042 PetscMPIInt size; 2043 2044 PetscFunctionBegin; 2045 ierr = MatCreate(comm,A);CHKERRQ(ierr); 2046 ierr = MatSetSizes(*A,m,n,M,N);CHKERRQ(ierr); 2047 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 2048 if (size > 1) { 2049 ierr = MatSetType(*A,MATMPISBAIJ);CHKERRQ(ierr); 2050 ierr = MatMPISBAIJSetPreallocation(*A,bs,d_nz,d_nnz,o_nz,o_nnz);CHKERRQ(ierr); 2051 } else { 2052 ierr = MatSetType(*A,MATSEQSBAIJ);CHKERRQ(ierr); 2053 ierr = MatSeqSBAIJSetPreallocation(*A,bs,d_nz,d_nnz);CHKERRQ(ierr); 2054 } 2055 PetscFunctionReturn(0); 2056 } 2057 2058 2059 #undef __FUNCT__ 2060 #define __FUNCT__ "MatDuplicate_MPISBAIJ" 2061 static PetscErrorCode MatDuplicate_MPISBAIJ(Mat matin,MatDuplicateOption cpvalues,Mat *newmat) 2062 { 2063 Mat mat; 2064 Mat_MPISBAIJ *a,*oldmat = (Mat_MPISBAIJ*)matin->data; 2065 PetscErrorCode ierr; 2066 PetscInt len=0,nt,bs=matin->rmap->bs,mbs=oldmat->mbs; 2067 PetscScalar *array; 2068 2069 PetscFunctionBegin; 2070 *newmat = 0; 2071 2072 ierr = MatCreate(PetscObjectComm((PetscObject)matin),&mat);CHKERRQ(ierr); 2073 ierr = MatSetSizes(mat,matin->rmap->n,matin->cmap->n,matin->rmap->N,matin->cmap->N);CHKERRQ(ierr); 2074 ierr = MatSetType(mat,((PetscObject)matin)->type_name);CHKERRQ(ierr); 2075 ierr = PetscMemcpy(mat->ops,matin->ops,sizeof(struct _MatOps));CHKERRQ(ierr); 2076 ierr = PetscLayoutReference(matin->rmap,&mat->rmap);CHKERRQ(ierr); 2077 ierr = PetscLayoutReference(matin->cmap,&mat->cmap);CHKERRQ(ierr); 2078 2079 mat->factortype = matin->factortype; 2080 mat->preallocated = PETSC_TRUE; 2081 mat->assembled = PETSC_TRUE; 2082 mat->insertmode = NOT_SET_VALUES; 2083 2084 a = (Mat_MPISBAIJ*)mat->data; 2085 a->bs2 = oldmat->bs2; 2086 a->mbs = oldmat->mbs; 2087 a->nbs = oldmat->nbs; 2088 a->Mbs = oldmat->Mbs; 2089 a->Nbs = oldmat->Nbs; 2090 2091 2092 a->size = oldmat->size; 2093 a->rank = oldmat->rank; 2094 a->donotstash = oldmat->donotstash; 2095 a->roworiented = oldmat->roworiented; 2096 a->rowindices = 0; 2097 a->rowvalues = 0; 2098 a->getrowactive = PETSC_FALSE; 2099 a->barray = 0; 2100 a->rstartbs = oldmat->rstartbs; 2101 a->rendbs = oldmat->rendbs; 2102 a->cstartbs = oldmat->cstartbs; 2103 a->cendbs = oldmat->cendbs; 2104 2105 /* hash table stuff */ 2106 a->ht = 0; 2107 a->hd = 0; 2108 a->ht_size = 0; 2109 a->ht_flag = oldmat->ht_flag; 2110 a->ht_fact = oldmat->ht_fact; 2111 a->ht_total_ct = 0; 2112 a->ht_insert_ct = 0; 2113 2114 ierr = PetscMemcpy(a->rangebs,oldmat->rangebs,(a->size+2)*sizeof(PetscInt));CHKERRQ(ierr); 2115 if (oldmat->colmap) { 2116 #if defined(PETSC_USE_CTABLE) 2117 ierr = PetscTableCreateCopy(oldmat->colmap,&a->colmap);CHKERRQ(ierr); 2118 #else 2119 ierr = PetscMalloc1((a->Nbs),&a->colmap);CHKERRQ(ierr); 2120 ierr = PetscLogObjectMemory((PetscObject)mat,(a->Nbs)*sizeof(PetscInt));CHKERRQ(ierr); 2121 ierr = PetscMemcpy(a->colmap,oldmat->colmap,(a->Nbs)*sizeof(PetscInt));CHKERRQ(ierr); 2122 #endif 2123 } else a->colmap = 0; 2124 2125 if (oldmat->garray && (len = ((Mat_SeqBAIJ*)(oldmat->B->data))->nbs)) { 2126 ierr = PetscMalloc1(len,&a->garray);CHKERRQ(ierr); 2127 ierr = PetscLogObjectMemory((PetscObject)mat,len*sizeof(PetscInt));CHKERRQ(ierr); 2128 ierr = PetscMemcpy(a->garray,oldmat->garray,len*sizeof(PetscInt));CHKERRQ(ierr); 2129 } else a->garray = 0; 2130 2131 ierr = MatStashCreate_Private(PetscObjectComm((PetscObject)matin),matin->rmap->bs,&mat->bstash);CHKERRQ(ierr); 2132 ierr = VecDuplicate(oldmat->lvec,&a->lvec);CHKERRQ(ierr); 2133 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->lvec);CHKERRQ(ierr); 2134 ierr = VecScatterCopy(oldmat->Mvctx,&a->Mvctx);CHKERRQ(ierr); 2135 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->Mvctx);CHKERRQ(ierr); 2136 2137 ierr = VecDuplicate(oldmat->slvec0,&a->slvec0);CHKERRQ(ierr); 2138 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec0);CHKERRQ(ierr); 2139 ierr = VecDuplicate(oldmat->slvec1,&a->slvec1);CHKERRQ(ierr); 2140 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec1);CHKERRQ(ierr); 2141 2142 ierr = VecGetLocalSize(a->slvec1,&nt);CHKERRQ(ierr); 2143 ierr = VecGetArray(a->slvec1,&array);CHKERRQ(ierr); 2144 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,1,bs*mbs,array,&a->slvec1a);CHKERRQ(ierr); 2145 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,1,nt-bs*mbs,array+bs*mbs,&a->slvec1b);CHKERRQ(ierr); 2146 ierr = VecRestoreArray(a->slvec1,&array);CHKERRQ(ierr); 2147 ierr = VecGetArray(a->slvec0,&array);CHKERRQ(ierr); 2148 ierr = VecCreateSeqWithArray(PETSC_COMM_SELF,1,nt-bs*mbs,array+bs*mbs,&a->slvec0b);CHKERRQ(ierr); 2149 ierr = VecRestoreArray(a->slvec0,&array);CHKERRQ(ierr); 2150 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec0);CHKERRQ(ierr); 2151 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec1);CHKERRQ(ierr); 2152 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec0b);CHKERRQ(ierr); 2153 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec1a);CHKERRQ(ierr); 2154 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->slvec1b);CHKERRQ(ierr); 2155 2156 /* ierr = VecScatterCopy(oldmat->sMvctx,&a->sMvctx); - not written yet, replaced by the lazy trick: */ 2157 ierr = PetscObjectReference((PetscObject)oldmat->sMvctx);CHKERRQ(ierr); 2158 a->sMvctx = oldmat->sMvctx; 2159 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->sMvctx);CHKERRQ(ierr); 2160 2161 ierr = MatDuplicate(oldmat->A,cpvalues,&a->A);CHKERRQ(ierr); 2162 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->A);CHKERRQ(ierr); 2163 ierr = MatDuplicate(oldmat->B,cpvalues,&a->B);CHKERRQ(ierr); 2164 ierr = PetscLogObjectParent((PetscObject)mat,(PetscObject)a->B);CHKERRQ(ierr); 2165 ierr = PetscFunctionListDuplicate(((PetscObject)matin)->qlist,&((PetscObject)mat)->qlist);CHKERRQ(ierr); 2166 *newmat = mat; 2167 PetscFunctionReturn(0); 2168 } 2169 2170 #undef __FUNCT__ 2171 #define __FUNCT__ "MatLoad_MPISBAIJ" 2172 PetscErrorCode MatLoad_MPISBAIJ(Mat newmat,PetscViewer viewer) 2173 { 2174 PetscErrorCode ierr; 2175 PetscInt i,nz,j,rstart,rend; 2176 PetscScalar *vals,*buf; 2177 MPI_Comm comm; 2178 MPI_Status status; 2179 PetscMPIInt rank,size,tag = ((PetscObject)viewer)->tag,*sndcounts = 0,*browners,maxnz,*rowners,mmbs; 2180 PetscInt header[4],*rowlengths = 0,M,N,m,*cols,*locrowlens; 2181 PetscInt *procsnz = 0,jj,*mycols,*ibuf; 2182 PetscInt bs =1,Mbs,mbs,extra_rows; 2183 PetscInt *dlens,*odlens,*mask,*masked1,*masked2,rowcount,odcount; 2184 PetscInt dcount,kmax,k,nzcount,tmp,sizesset=1,grows,gcols; 2185 int fd; 2186 2187 PetscFunctionBegin; 2188 ierr = PetscObjectGetComm((PetscObject)viewer,&comm);CHKERRQ(ierr); 2189 ierr = PetscOptionsBegin(comm,NULL,"Options for loading MPISBAIJ matrix 2","Mat");CHKERRQ(ierr); 2190 ierr = PetscOptionsInt("-matload_block_size","Set the blocksize used to store the matrix","MatLoad",bs,&bs,NULL);CHKERRQ(ierr); 2191 ierr = PetscOptionsEnd();CHKERRQ(ierr); 2192 2193 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 2194 ierr = MPI_Comm_rank(comm,&rank);CHKERRQ(ierr); 2195 if (!rank) { 2196 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 2197 ierr = PetscBinaryRead(fd,(char*)header,4,PETSC_INT);CHKERRQ(ierr); 2198 if (header[0] != MAT_FILE_CLASSID) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED,"not matrix object"); 2199 if (header[3] < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED,"Matrix stored in special format, cannot load as MPISBAIJ"); 2200 } 2201 2202 if (newmat->rmap->n < 0 && newmat->rmap->N < 0 && newmat->cmap->n < 0 && newmat->cmap->N < 0) sizesset = 0; 2203 2204 ierr = MPI_Bcast(header+1,3,MPIU_INT,0,comm);CHKERRQ(ierr); 2205 M = header[1]; 2206 N = header[2]; 2207 2208 /* If global rows/cols are set to PETSC_DECIDE, set it to the sizes given in the file */ 2209 if (sizesset && newmat->rmap->N < 0) newmat->rmap->N = M; 2210 if (sizesset && newmat->cmap->N < 0) newmat->cmap->N = N; 2211 2212 /* If global sizes are set, check if they are consistent with that given in the file */ 2213 if (sizesset) { 2214 ierr = MatGetSize(newmat,&grows,&gcols);CHKERRQ(ierr); 2215 } 2216 if (sizesset && newmat->rmap->N != grows) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED, "Inconsistent # of rows:Matrix in file has (%d) and input matrix has (%d)",M,grows); 2217 if (sizesset && newmat->cmap->N != gcols) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED, "Inconsistent # of cols:Matrix in file has (%d) and input matrix has (%d)",N,gcols); 2218 2219 if (M != N) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only do square matrices"); 2220 2221 /* 2222 This code adds extra rows to make sure the number of rows is 2223 divisible by the blocksize 2224 */ 2225 Mbs = M/bs; 2226 extra_rows = bs - M + bs*(Mbs); 2227 if (extra_rows == bs) extra_rows = 0; 2228 else Mbs++; 2229 if (extra_rows &&!rank) { 2230 ierr = PetscInfo(viewer,"Padding loaded matrix to match blocksize\n");CHKERRQ(ierr); 2231 } 2232 2233 /* determine ownership of all rows */ 2234 if (newmat->rmap->n < 0) { /* PETSC_DECIDE */ 2235 mbs = Mbs/size + ((Mbs % size) > rank); 2236 m = mbs*bs; 2237 } else { /* User Set */ 2238 m = newmat->rmap->n; 2239 mbs = m/bs; 2240 } 2241 ierr = PetscMalloc2(size+1,&rowners,size+1,&browners);CHKERRQ(ierr); 2242 ierr = PetscMPIIntCast(mbs,&mmbs);CHKERRQ(ierr); 2243 ierr = MPI_Allgather(&mmbs,1,MPI_INT,rowners+1,1,MPI_INT,comm);CHKERRQ(ierr); 2244 rowners[0] = 0; 2245 for (i=2; i<=size; i++) rowners[i] += rowners[i-1]; 2246 for (i=0; i<=size; i++) browners[i] = rowners[i]*bs; 2247 rstart = rowners[rank]; 2248 rend = rowners[rank+1]; 2249 2250 /* distribute row lengths to all processors */ 2251 ierr = PetscMalloc1((rend-rstart)*bs,&locrowlens);CHKERRQ(ierr); 2252 if (!rank) { 2253 ierr = PetscMalloc1((M+extra_rows),&rowlengths);CHKERRQ(ierr); 2254 ierr = PetscBinaryRead(fd,rowlengths,M,PETSC_INT);CHKERRQ(ierr); 2255 for (i=0; i<extra_rows; i++) rowlengths[M+i] = 1; 2256 ierr = PetscMalloc1(size,&sndcounts);CHKERRQ(ierr); 2257 for (i=0; i<size; i++) sndcounts[i] = browners[i+1] - browners[i]; 2258 ierr = MPI_Scatterv(rowlengths,sndcounts,browners,MPIU_INT,locrowlens,(rend-rstart)*bs,MPIU_INT,0,comm);CHKERRQ(ierr); 2259 ierr = PetscFree(sndcounts);CHKERRQ(ierr); 2260 } else { 2261 ierr = MPI_Scatterv(0,0,0,MPIU_INT,locrowlens,(rend-rstart)*bs,MPIU_INT,0,comm);CHKERRQ(ierr); 2262 } 2263 2264 if (!rank) { /* procs[0] */ 2265 /* calculate the number of nonzeros on each processor */ 2266 ierr = PetscMalloc1(size,&procsnz);CHKERRQ(ierr); 2267 ierr = PetscMemzero(procsnz,size*sizeof(PetscInt));CHKERRQ(ierr); 2268 for (i=0; i<size; i++) { 2269 for (j=rowners[i]*bs; j< rowners[i+1]*bs; j++) { 2270 procsnz[i] += rowlengths[j]; 2271 } 2272 } 2273 ierr = PetscFree(rowlengths);CHKERRQ(ierr); 2274 2275 /* determine max buffer needed and allocate it */ 2276 maxnz = 0; 2277 for (i=0; i<size; i++) { 2278 maxnz = PetscMax(maxnz,procsnz[i]); 2279 } 2280 ierr = PetscMalloc1(maxnz,&cols);CHKERRQ(ierr); 2281 2282 /* read in my part of the matrix column indices */ 2283 nz = procsnz[0]; 2284 ierr = PetscMalloc1(nz,&ibuf);CHKERRQ(ierr); 2285 mycols = ibuf; 2286 if (size == 1) nz -= extra_rows; 2287 ierr = PetscBinaryRead(fd,mycols,nz,PETSC_INT);CHKERRQ(ierr); 2288 if (size == 1) { 2289 for (i=0; i< extra_rows; i++) mycols[nz+i] = M+i; 2290 } 2291 2292 /* read in every ones (except the last) and ship off */ 2293 for (i=1; i<size-1; i++) { 2294 nz = procsnz[i]; 2295 ierr = PetscBinaryRead(fd,cols,nz,PETSC_INT);CHKERRQ(ierr); 2296 ierr = MPI_Send(cols,nz,MPIU_INT,i,tag,comm);CHKERRQ(ierr); 2297 } 2298 /* read in the stuff for the last proc */ 2299 if (size != 1) { 2300 nz = procsnz[size-1] - extra_rows; /* the extra rows are not on the disk */ 2301 ierr = PetscBinaryRead(fd,cols,nz,PETSC_INT);CHKERRQ(ierr); 2302 for (i=0; i<extra_rows; i++) cols[nz+i] = M+i; 2303 ierr = MPI_Send(cols,nz+extra_rows,MPIU_INT,size-1,tag,comm);CHKERRQ(ierr); 2304 } 2305 ierr = PetscFree(cols);CHKERRQ(ierr); 2306 } else { /* procs[i], i>0 */ 2307 /* determine buffer space needed for message */ 2308 nz = 0; 2309 for (i=0; i<m; i++) nz += locrowlens[i]; 2310 ierr = PetscMalloc1(nz,&ibuf);CHKERRQ(ierr); 2311 mycols = ibuf; 2312 /* receive message of column indices*/ 2313 ierr = MPI_Recv(mycols,nz,MPIU_INT,0,tag,comm,&status);CHKERRQ(ierr); 2314 ierr = MPI_Get_count(&status,MPIU_INT,&maxnz);CHKERRQ(ierr); 2315 if (maxnz != nz) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED,"something is wrong with file"); 2316 } 2317 2318 /* loop over local rows, determining number of off diagonal entries */ 2319 ierr = PetscMalloc2(rend-rstart,&dlens,rend-rstart,&odlens);CHKERRQ(ierr); 2320 ierr = PetscMalloc3(Mbs,&mask,Mbs,&masked1,Mbs,&masked2);CHKERRQ(ierr); 2321 ierr = PetscMemzero(mask,Mbs*sizeof(PetscInt));CHKERRQ(ierr); 2322 ierr = PetscMemzero(masked1,Mbs*sizeof(PetscInt));CHKERRQ(ierr); 2323 ierr = PetscMemzero(masked2,Mbs*sizeof(PetscInt));CHKERRQ(ierr); 2324 rowcount = 0; 2325 nzcount = 0; 2326 for (i=0; i<mbs; i++) { 2327 dcount = 0; 2328 odcount = 0; 2329 for (j=0; j<bs; j++) { 2330 kmax = locrowlens[rowcount]; 2331 for (k=0; k<kmax; k++) { 2332 tmp = mycols[nzcount++]/bs; /* block col. index */ 2333 if (!mask[tmp]) { 2334 mask[tmp] = 1; 2335 if (tmp < rstart || tmp >= rend) masked2[odcount++] = tmp; /* entry in off-diag portion */ 2336 else masked1[dcount++] = tmp; /* entry in diag portion */ 2337 } 2338 } 2339 rowcount++; 2340 } 2341 2342 dlens[i] = dcount; /* d_nzz[i] */ 2343 odlens[i] = odcount; /* o_nzz[i] */ 2344 2345 /* zero out the mask elements we set */ 2346 for (j=0; j<dcount; j++) mask[masked1[j]] = 0; 2347 for (j=0; j<odcount; j++) mask[masked2[j]] = 0; 2348 } 2349 if (!sizesset) { 2350 ierr = MatSetSizes(newmat,m,m,M+extra_rows,N+extra_rows);CHKERRQ(ierr); 2351 } 2352 ierr = MatMPISBAIJSetPreallocation(newmat,bs,0,dlens,0,odlens);CHKERRQ(ierr); 2353 ierr = MatSetOption(newmat,MAT_IGNORE_LOWER_TRIANGULAR,PETSC_TRUE);CHKERRQ(ierr); 2354 2355 if (!rank) { 2356 ierr = PetscMalloc1(maxnz,&buf);CHKERRQ(ierr); 2357 /* read in my part of the matrix numerical values */ 2358 nz = procsnz[0]; 2359 vals = buf; 2360 mycols = ibuf; 2361 if (size == 1) nz -= extra_rows; 2362 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2363 if (size == 1) { 2364 for (i=0; i< extra_rows; i++) vals[nz+i] = 1.0; 2365 } 2366 2367 /* insert into matrix */ 2368 jj = rstart*bs; 2369 for (i=0; i<m; i++) { 2370 ierr = MatSetValues(newmat,1,&jj,locrowlens[i],mycols,vals,INSERT_VALUES);CHKERRQ(ierr); 2371 mycols += locrowlens[i]; 2372 vals += locrowlens[i]; 2373 jj++; 2374 } 2375 2376 /* read in other processors (except the last one) and ship out */ 2377 for (i=1; i<size-1; i++) { 2378 nz = procsnz[i]; 2379 vals = buf; 2380 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2381 ierr = MPI_Send(vals,nz,MPIU_SCALAR,i,((PetscObject)newmat)->tag,comm);CHKERRQ(ierr); 2382 } 2383 /* the last proc */ 2384 if (size != 1) { 2385 nz = procsnz[i] - extra_rows; 2386 vals = buf; 2387 ierr = PetscBinaryRead(fd,vals,nz,PETSC_SCALAR);CHKERRQ(ierr); 2388 for (i=0; i<extra_rows; i++) vals[nz+i] = 1.0; 2389 ierr = MPI_Send(vals,nz+extra_rows,MPIU_SCALAR,size-1,((PetscObject)newmat)->tag,comm);CHKERRQ(ierr); 2390 } 2391 ierr = PetscFree(procsnz);CHKERRQ(ierr); 2392 2393 } else { 2394 /* receive numeric values */ 2395 ierr = PetscMalloc1(nz,&buf);CHKERRQ(ierr); 2396 2397 /* receive message of values*/ 2398 vals = buf; 2399 mycols = ibuf; 2400 ierr = MPI_Recv(vals,nz,MPIU_SCALAR,0,((PetscObject)newmat)->tag,comm,&status);CHKERRQ(ierr); 2401 ierr = MPI_Get_count(&status,MPIU_SCALAR,&maxnz);CHKERRQ(ierr); 2402 if (maxnz != nz) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_FILE_UNEXPECTED,"something is wrong with file"); 2403 2404 /* insert into matrix */ 2405 jj = rstart*bs; 2406 for (i=0; i<m; i++) { 2407 ierr = MatSetValues_MPISBAIJ(newmat,1,&jj,locrowlens[i],mycols,vals,INSERT_VALUES);CHKERRQ(ierr); 2408 mycols += locrowlens[i]; 2409 vals += locrowlens[i]; 2410 jj++; 2411 } 2412 } 2413 2414 ierr = PetscFree(locrowlens);CHKERRQ(ierr); 2415 ierr = PetscFree(buf);CHKERRQ(ierr); 2416 ierr = PetscFree(ibuf);CHKERRQ(ierr); 2417 ierr = PetscFree2(rowners,browners);CHKERRQ(ierr); 2418 ierr = PetscFree2(dlens,odlens);CHKERRQ(ierr); 2419 ierr = PetscFree3(mask,masked1,masked2);CHKERRQ(ierr); 2420 ierr = MatAssemblyBegin(newmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2421 ierr = MatAssemblyEnd(newmat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2422 PetscFunctionReturn(0); 2423 } 2424 2425 #undef __FUNCT__ 2426 #define __FUNCT__ "MatMPISBAIJSetHashTableFactor" 2427 /*XXXXX@ 2428 MatMPISBAIJSetHashTableFactor - Sets the factor required to compute the size of the HashTable. 2429 2430 Input Parameters: 2431 . mat - the matrix 2432 . fact - factor 2433 2434 Not Collective on Mat, each process can have a different hash factor 2435 2436 Level: advanced 2437 2438 Notes: 2439 This can also be set by the command line option: -mat_use_hash_table fact 2440 2441 .keywords: matrix, hashtable, factor, HT 2442 2443 .seealso: MatSetOption() 2444 @XXXXX*/ 2445 2446 2447 #undef __FUNCT__ 2448 #define __FUNCT__ "MatGetRowMaxAbs_MPISBAIJ" 2449 PetscErrorCode MatGetRowMaxAbs_MPISBAIJ(Mat A,Vec v,PetscInt idx[]) 2450 { 2451 Mat_MPISBAIJ *a = (Mat_MPISBAIJ*)A->data; 2452 Mat_SeqBAIJ *b = (Mat_SeqBAIJ*)(a->B)->data; 2453 PetscReal atmp; 2454 PetscReal *work,*svalues,*rvalues; 2455 PetscErrorCode ierr; 2456 PetscInt i,bs,mbs,*bi,*bj,brow,j,ncols,krow,kcol,col,row,Mbs,bcol; 2457 PetscMPIInt rank,size; 2458 PetscInt *rowners_bs,dest,count,source; 2459 PetscScalar *va; 2460 MatScalar *ba; 2461 MPI_Status stat; 2462 2463 PetscFunctionBegin; 2464 if (idx) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Send email to petsc-maint@mcs.anl.gov"); 2465 ierr = MatGetRowMaxAbs(a->A,v,NULL);CHKERRQ(ierr); 2466 ierr = VecGetArray(v,&va);CHKERRQ(ierr); 2467 2468 ierr = MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);CHKERRQ(ierr); 2469 ierr = MPI_Comm_rank(PetscObjectComm((PetscObject)A),&rank);CHKERRQ(ierr); 2470 2471 bs = A->rmap->bs; 2472 mbs = a->mbs; 2473 Mbs = a->Mbs; 2474 ba = b->a; 2475 bi = b->i; 2476 bj = b->j; 2477 2478 /* find ownerships */ 2479 rowners_bs = A->rmap->range; 2480 2481 /* each proc creates an array to be distributed */ 2482 ierr = PetscMalloc1(bs*Mbs,&work);CHKERRQ(ierr); 2483 ierr = PetscMemzero(work,bs*Mbs*sizeof(PetscReal));CHKERRQ(ierr); 2484 2485 /* row_max for B */ 2486 if (rank != size-1) { 2487 for (i=0; i<mbs; i++) { 2488 ncols = bi[1] - bi[0]; bi++; 2489 brow = bs*i; 2490 for (j=0; j<ncols; j++) { 2491 bcol = bs*(*bj); 2492 for (kcol=0; kcol<bs; kcol++) { 2493 col = bcol + kcol; /* local col index */ 2494 col += rowners_bs[rank+1]; /* global col index */ 2495 for (krow=0; krow<bs; krow++) { 2496 atmp = PetscAbsScalar(*ba); ba++; 2497 row = brow + krow; /* local row index */ 2498 if (PetscRealPart(va[row]) < atmp) va[row] = atmp; 2499 if (work[col] < atmp) work[col] = atmp; 2500 } 2501 } 2502 bj++; 2503 } 2504 } 2505 2506 /* send values to its owners */ 2507 for (dest=rank+1; dest<size; dest++) { 2508 svalues = work + rowners_bs[dest]; 2509 count = rowners_bs[dest+1]-rowners_bs[dest]; 2510 ierr = MPI_Send(svalues,count,MPIU_REAL,dest,rank,PetscObjectComm((PetscObject)A));CHKERRQ(ierr); 2511 } 2512 } 2513 2514 /* receive values */ 2515 if (rank) { 2516 rvalues = work; 2517 count = rowners_bs[rank+1]-rowners_bs[rank]; 2518 for (source=0; source<rank; source++) { 2519 ierr = MPI_Recv(rvalues,count,MPIU_REAL,MPI_ANY_SOURCE,MPI_ANY_TAG,PetscObjectComm((PetscObject)A),&stat);CHKERRQ(ierr); 2520 /* process values */ 2521 for (i=0; i<count; i++) { 2522 if (PetscRealPart(va[i]) < rvalues[i]) va[i] = rvalues[i]; 2523 } 2524 } 2525 } 2526 2527 ierr = VecRestoreArray(v,&va);CHKERRQ(ierr); 2528 ierr = PetscFree(work);CHKERRQ(ierr); 2529 PetscFunctionReturn(0); 2530 } 2531 2532 #undef __FUNCT__ 2533 #define __FUNCT__ "MatSOR_MPISBAIJ" 2534 PetscErrorCode MatSOR_MPISBAIJ(Mat matin,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,PetscInt its,PetscInt lits,Vec xx) 2535 { 2536 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)matin->data; 2537 PetscErrorCode ierr; 2538 PetscInt mbs=mat->mbs,bs=matin->rmap->bs; 2539 PetscScalar *x,*ptr,*from; 2540 Vec bb1; 2541 const PetscScalar *b; 2542 2543 PetscFunctionBegin; 2544 if (its <= 0 || lits <= 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D and local its %D both positive",its,lits); 2545 if (bs > 1) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"SSOR for block size > 1 is not yet implemented"); 2546 2547 if (flag == SOR_APPLY_UPPER) { 2548 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 2549 PetscFunctionReturn(0); 2550 } 2551 2552 if ((flag & SOR_LOCAL_SYMMETRIC_SWEEP) == SOR_LOCAL_SYMMETRIC_SWEEP) { 2553 if (flag & SOR_ZERO_INITIAL_GUESS) { 2554 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,lits,xx);CHKERRQ(ierr); 2555 its--; 2556 } 2557 2558 ierr = VecDuplicate(bb,&bb1);CHKERRQ(ierr); 2559 while (its--) { 2560 2561 /* lower triangular part: slvec0b = - B^T*xx */ 2562 ierr = (*mat->B->ops->multtranspose)(mat->B,xx,mat->slvec0b);CHKERRQ(ierr); 2563 2564 /* copy xx into slvec0a */ 2565 ierr = VecGetArray(mat->slvec0,&ptr);CHKERRQ(ierr); 2566 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 2567 ierr = PetscMemcpy(ptr,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 2568 ierr = VecRestoreArray(mat->slvec0,&ptr);CHKERRQ(ierr); 2569 2570 ierr = VecScale(mat->slvec0,-1.0);CHKERRQ(ierr); 2571 2572 /* copy bb into slvec1a */ 2573 ierr = VecGetArray(mat->slvec1,&ptr);CHKERRQ(ierr); 2574 ierr = VecGetArrayRead(bb,&b);CHKERRQ(ierr); 2575 ierr = PetscMemcpy(ptr,b,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 2576 ierr = VecRestoreArray(mat->slvec1,&ptr);CHKERRQ(ierr); 2577 2578 /* set slvec1b = 0 */ 2579 ierr = VecSet(mat->slvec1b,0.0);CHKERRQ(ierr); 2580 2581 ierr = VecScatterBegin(mat->sMvctx,mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2582 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 2583 ierr = VecRestoreArrayRead(bb,&b);CHKERRQ(ierr); 2584 ierr = VecScatterEnd(mat->sMvctx,mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2585 2586 /* upper triangular part: bb1 = bb1 - B*x */ 2587 ierr = (*mat->B->ops->multadd)(mat->B,mat->slvec1b,mat->slvec1a,bb1);CHKERRQ(ierr); 2588 2589 /* local diagonal sweep */ 2590 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,SOR_SYMMETRIC_SWEEP,fshift,lits,lits,xx);CHKERRQ(ierr); 2591 } 2592 ierr = VecDestroy(&bb1);CHKERRQ(ierr); 2593 } else if ((flag & SOR_LOCAL_FORWARD_SWEEP) && (its == 1) && (flag & SOR_ZERO_INITIAL_GUESS)) { 2594 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 2595 } else if ((flag & SOR_LOCAL_BACKWARD_SWEEP) && (its == 1) && (flag & SOR_ZERO_INITIAL_GUESS)) { 2596 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,1,xx);CHKERRQ(ierr); 2597 } else if (flag & SOR_EISENSTAT) { 2598 Vec xx1; 2599 PetscBool hasop; 2600 const PetscScalar *diag; 2601 PetscScalar *sl,scale = (omega - 2.0)/omega; 2602 PetscInt i,n; 2603 2604 if (!mat->xx1) { 2605 ierr = VecDuplicate(bb,&mat->xx1);CHKERRQ(ierr); 2606 ierr = VecDuplicate(bb,&mat->bb1);CHKERRQ(ierr); 2607 } 2608 xx1 = mat->xx1; 2609 bb1 = mat->bb1; 2610 2611 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,(MatSORType)(SOR_ZERO_INITIAL_GUESS | SOR_LOCAL_BACKWARD_SWEEP),fshift,lits,1,xx);CHKERRQ(ierr); 2612 2613 if (!mat->diag) { 2614 /* this is wrong for same matrix with new nonzero values */ 2615 ierr = MatGetVecs(matin,&mat->diag,NULL);CHKERRQ(ierr); 2616 ierr = MatGetDiagonal(matin,mat->diag);CHKERRQ(ierr); 2617 } 2618 ierr = MatHasOperation(matin,MATOP_MULT_DIAGONAL_BLOCK,&hasop);CHKERRQ(ierr); 2619 2620 if (hasop) { 2621 ierr = MatMultDiagonalBlock(matin,xx,bb1);CHKERRQ(ierr); 2622 ierr = VecAYPX(mat->slvec1a,scale,bb);CHKERRQ(ierr); 2623 } else { 2624 /* 2625 These two lines are replaced by code that may be a bit faster for a good compiler 2626 ierr = VecPointwiseMult(mat->slvec1a,mat->diag,xx);CHKERRQ(ierr); 2627 ierr = VecAYPX(mat->slvec1a,scale,bb);CHKERRQ(ierr); 2628 */ 2629 ierr = VecGetArray(mat->slvec1a,&sl);CHKERRQ(ierr); 2630 ierr = VecGetArrayRead(mat->diag,&diag);CHKERRQ(ierr); 2631 ierr = VecGetArrayRead(bb,&b);CHKERRQ(ierr); 2632 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 2633 ierr = VecGetLocalSize(xx,&n);CHKERRQ(ierr); 2634 if (omega == 1.0) { 2635 for (i=0; i<n; i++) sl[i] = b[i] - diag[i]*x[i]; 2636 ierr = PetscLogFlops(2.0*n);CHKERRQ(ierr); 2637 } else { 2638 for (i=0; i<n; i++) sl[i] = b[i] + scale*diag[i]*x[i]; 2639 ierr = PetscLogFlops(3.0*n);CHKERRQ(ierr); 2640 } 2641 ierr = VecRestoreArray(mat->slvec1a,&sl);CHKERRQ(ierr); 2642 ierr = VecRestoreArrayRead(mat->diag,&diag);CHKERRQ(ierr); 2643 ierr = VecRestoreArrayRead(bb,&b);CHKERRQ(ierr); 2644 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 2645 } 2646 2647 /* multiply off-diagonal portion of matrix */ 2648 ierr = VecSet(mat->slvec1b,0.0);CHKERRQ(ierr); 2649 ierr = (*mat->B->ops->multtranspose)(mat->B,xx,mat->slvec0b);CHKERRQ(ierr); 2650 ierr = VecGetArray(mat->slvec0,&from);CHKERRQ(ierr); 2651 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 2652 ierr = PetscMemcpy(from,x,bs*mbs*sizeof(MatScalar));CHKERRQ(ierr); 2653 ierr = VecRestoreArray(mat->slvec0,&from);CHKERRQ(ierr); 2654 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 2655 ierr = VecScatterBegin(mat->sMvctx,mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2656 ierr = VecScatterEnd(mat->sMvctx,mat->slvec0,mat->slvec1,ADD_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2657 ierr = (*mat->B->ops->multadd)(mat->B,mat->slvec1b,mat->slvec1a,mat->slvec1a);CHKERRQ(ierr); 2658 2659 /* local sweep */ 2660 ierr = (*mat->A->ops->sor)(mat->A,mat->slvec1a,omega,(MatSORType)(SOR_ZERO_INITIAL_GUESS | SOR_LOCAL_FORWARD_SWEEP),fshift,lits,1,xx1);CHKERRQ(ierr); 2661 ierr = VecAXPY(xx,1.0,xx1);CHKERRQ(ierr); 2662 } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatSORType is not supported for SBAIJ matrix format"); 2663 PetscFunctionReturn(0); 2664 } 2665 2666 #undef __FUNCT__ 2667 #define __FUNCT__ "MatSOR_MPISBAIJ_2comm" 2668 PetscErrorCode MatSOR_MPISBAIJ_2comm(Mat matin,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,PetscInt its,PetscInt lits,Vec xx) 2669 { 2670 Mat_MPISBAIJ *mat = (Mat_MPISBAIJ*)matin->data; 2671 PetscErrorCode ierr; 2672 Vec lvec1,bb1; 2673 2674 PetscFunctionBegin; 2675 if (its <= 0 || lits <= 0) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D and local its %D both positive",its,lits); 2676 if (matin->rmap->bs > 1) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"SSOR for block size > 1 is not yet implemented"); 2677 2678 if ((flag & SOR_LOCAL_SYMMETRIC_SWEEP) == SOR_LOCAL_SYMMETRIC_SWEEP) { 2679 if (flag & SOR_ZERO_INITIAL_GUESS) { 2680 ierr = (*mat->A->ops->sor)(mat->A,bb,omega,flag,fshift,lits,lits,xx);CHKERRQ(ierr); 2681 its--; 2682 } 2683 2684 ierr = VecDuplicate(mat->lvec,&lvec1);CHKERRQ(ierr); 2685 ierr = VecDuplicate(bb,&bb1);CHKERRQ(ierr); 2686 while (its--) { 2687 ierr = VecScatterBegin(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2688 2689 /* lower diagonal part: bb1 = bb - B^T*xx */ 2690 ierr = (*mat->B->ops->multtranspose)(mat->B,xx,lvec1);CHKERRQ(ierr); 2691 ierr = VecScale(lvec1,-1.0);CHKERRQ(ierr); 2692 2693 ierr = VecScatterEnd(mat->Mvctx,xx,mat->lvec,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 2694 ierr = VecCopy(bb,bb1);CHKERRQ(ierr); 2695 ierr = VecScatterBegin(mat->Mvctx,lvec1,bb1,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 2696 2697 /* upper diagonal part: bb1 = bb1 - B*x */ 2698 ierr = VecScale(mat->lvec,-1.0);CHKERRQ(ierr); 2699 ierr = (*mat->B->ops->multadd)(mat->B,mat->lvec,bb1,bb1);CHKERRQ(ierr); 2700 2701 ierr = VecScatterEnd(mat->Mvctx,lvec1,bb1,ADD_VALUES,SCATTER_REVERSE);CHKERRQ(ierr); 2702 2703 /* diagonal sweep */ 2704 ierr = (*mat->A->ops->sor)(mat->A,bb1,omega,SOR_SYMMETRIC_SWEEP,fshift,lits,lits,xx);CHKERRQ(ierr); 2705 } 2706 ierr = VecDestroy(&lvec1);CHKERRQ(ierr); 2707 ierr = VecDestroy(&bb1);CHKERRQ(ierr); 2708 } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatSORType is not supported for SBAIJ matrix format"); 2709 PetscFunctionReturn(0); 2710 } 2711 2712 #undef __FUNCT__ 2713 #define __FUNCT__ "MatCreateMPISBAIJWithArrays" 2714 /*@ 2715 MatCreateMPISBAIJWithArrays - creates a MPI SBAIJ matrix using arrays that contain in standard 2716 CSR format the local rows. 2717 2718 Collective on MPI_Comm 2719 2720 Input Parameters: 2721 + comm - MPI communicator 2722 . bs - the block size, only a block size of 1 is supported 2723 . m - number of local rows (Cannot be PETSC_DECIDE) 2724 . n - This value should be the same as the local size used in creating the 2725 x vector for the matrix-vector product y = Ax. (or PETSC_DECIDE to have 2726 calculated if N is given) For square matrices n is almost always m. 2727 . M - number of global rows (or PETSC_DETERMINE to have calculated if m is given) 2728 . N - number of global columns (or PETSC_DETERMINE to have calculated if n is given) 2729 . i - row indices 2730 . j - column indices 2731 - a - matrix values 2732 2733 Output Parameter: 2734 . mat - the matrix 2735 2736 Level: intermediate 2737 2738 Notes: 2739 The i, j, and a arrays ARE copied by this routine into the internal format used by PETSc; 2740 thus you CANNOT change the matrix entries by changing the values of a[] after you have 2741 called this routine. Use MatCreateMPIAIJWithSplitArrays() to avoid needing to copy the arrays. 2742 2743 The i and j indices are 0 based, and i indices are indices corresponding to the local j array. 2744 2745 .keywords: matrix, aij, compressed row, sparse, parallel 2746 2747 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIAIJSetPreallocation(), MatMPIAIJSetPreallocationCSR(), 2748 MPIAIJ, MatCreateAIJ(), MatCreateMPIAIJWithSplitArrays() 2749 @*/ 2750 PetscErrorCode MatCreateMPISBAIJWithArrays(MPI_Comm comm,PetscInt bs,PetscInt m,PetscInt n,PetscInt M,PetscInt N,const PetscInt i[],const PetscInt j[],const PetscScalar a[],Mat *mat) 2751 { 2752 PetscErrorCode ierr; 2753 2754 2755 PetscFunctionBegin; 2756 if (i[0]) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"i (row indices) must start with 0"); 2757 if (m < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"local number of rows (m) cannot be PETSC_DECIDE, or negative"); 2758 ierr = MatCreate(comm,mat);CHKERRQ(ierr); 2759 ierr = MatSetSizes(*mat,m,n,M,N);CHKERRQ(ierr); 2760 ierr = MatSetType(*mat,MATMPISBAIJ);CHKERRQ(ierr); 2761 ierr = MatMPISBAIJSetPreallocationCSR(*mat,bs,i,j,a);CHKERRQ(ierr); 2762 PetscFunctionReturn(0); 2763 } 2764 2765 2766 #undef __FUNCT__ 2767 #define __FUNCT__ "MatMPISBAIJSetPreallocationCSR" 2768 /*@C 2769 MatMPISBAIJSetPreallocationCSR - Allocates memory for a sparse parallel matrix in BAIJ format 2770 (the default parallel PETSc format). 2771 2772 Collective on MPI_Comm 2773 2774 Input Parameters: 2775 + A - the matrix 2776 . bs - the block size 2777 . i - the indices into j for the start of each local row (starts with zero) 2778 . j - the column indices for each local row (starts with zero) these must be sorted for each row 2779 - v - optional values in the matrix 2780 2781 Level: developer 2782 2783 .keywords: matrix, aij, compressed row, sparse, parallel 2784 2785 .seealso: MatCreate(), MatCreateSeqAIJ(), MatSetValues(), MatMPIBAIJSetPreallocation(), MatCreateAIJ(), MPIAIJ 2786 @*/ 2787 PetscErrorCode MatMPISBAIJSetPreallocationCSR(Mat B,PetscInt bs,const PetscInt i[],const PetscInt j[], const PetscScalar v[]) 2788 { 2789 PetscErrorCode ierr; 2790 2791 PetscFunctionBegin; 2792 ierr = PetscTryMethod(B,"MatMPISBAIJSetPreallocationCSR_C",(Mat,PetscInt,const PetscInt[],const PetscInt[],const PetscScalar[]),(B,bs,i,j,v));CHKERRQ(ierr); 2793 PetscFunctionReturn(0); 2794 } 2795 2796 2797