1 2 /* 3 Defines the basic matrix operations for the AIJ (compressed row) 4 matrix storage format. 5 */ 6 7 #include "src/mat/impls/aij/seq/aij.h" /*I "petscmat.h" I*/ 8 #include "src/inline/spops.h" 9 #include "src/inline/dot.h" 10 #include "petscbt.h" 11 12 #undef __FUNCT__ 13 #define __FUNCT__ "MatGetRowIJ_SeqAIJ" 14 int MatGetRowIJ_SeqAIJ(Mat A,int oshift,PetscTruth symmetric,int *m,int *ia[],int *ja[],PetscTruth *done) 15 { 16 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 17 int ierr,i,ishift; 18 19 PetscFunctionBegin; 20 *m = A->m; 21 if (!ia) PetscFunctionReturn(0); 22 ishift = 0; 23 if (symmetric && !A->structurally_symmetric) { 24 ierr = MatToSymmetricIJ_SeqAIJ(A->m,a->i,a->j,ishift,oshift,ia,ja);CHKERRQ(ierr); 25 } else if (oshift == 1) { 26 int nz = a->i[A->m]; 27 /* malloc space and add 1 to i and j indices */ 28 ierr = PetscMalloc((A->m+1)*sizeof(int),ia);CHKERRQ(ierr); 29 ierr = PetscMalloc((nz+1)*sizeof(int),ja);CHKERRQ(ierr); 30 for (i=0; i<nz; i++) (*ja)[i] = a->j[i] + 1; 31 for (i=0; i<A->m+1; i++) (*ia)[i] = a->i[i] + 1; 32 } else { 33 *ia = a->i; *ja = a->j; 34 } 35 PetscFunctionReturn(0); 36 } 37 38 #undef __FUNCT__ 39 #define __FUNCT__ "MatRestoreRowIJ_SeqAIJ" 40 int MatRestoreRowIJ_SeqAIJ(Mat A,int oshift,PetscTruth symmetric,int *n,int *ia[],int *ja[],PetscTruth *done) 41 { 42 int ierr; 43 44 PetscFunctionBegin; 45 if (!ia) PetscFunctionReturn(0); 46 if ((symmetric && !A->structurally_symmetric) || oshift == 1) { 47 ierr = PetscFree(*ia);CHKERRQ(ierr); 48 ierr = PetscFree(*ja);CHKERRQ(ierr); 49 } 50 PetscFunctionReturn(0); 51 } 52 53 #undef __FUNCT__ 54 #define __FUNCT__ "MatGetColumnIJ_SeqAIJ" 55 int MatGetColumnIJ_SeqAIJ(Mat A,int oshift,PetscTruth symmetric,int *nn,int *ia[],int *ja[],PetscTruth *done) 56 { 57 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 58 int ierr,i,*collengths,*cia,*cja,n = A->n,m = A->m; 59 int nz = a->i[m],row,*jj,mr,col; 60 61 PetscFunctionBegin; 62 *nn = A->n; 63 if (!ia) PetscFunctionReturn(0); 64 if (symmetric) { 65 ierr = MatToSymmetricIJ_SeqAIJ(A->m,a->i,a->j,0,oshift,ia,ja);CHKERRQ(ierr); 66 } else { 67 ierr = PetscMalloc((n+1)*sizeof(int),&collengths);CHKERRQ(ierr); 68 ierr = PetscMemzero(collengths,n*sizeof(int));CHKERRQ(ierr); 69 ierr = PetscMalloc((n+1)*sizeof(int),&cia);CHKERRQ(ierr); 70 ierr = PetscMalloc((nz+1)*sizeof(int),&cja);CHKERRQ(ierr); 71 jj = a->j; 72 for (i=0; i<nz; i++) { 73 collengths[jj[i]]++; 74 } 75 cia[0] = oshift; 76 for (i=0; i<n; i++) { 77 cia[i+1] = cia[i] + collengths[i]; 78 } 79 ierr = PetscMemzero(collengths,n*sizeof(int));CHKERRQ(ierr); 80 jj = a->j; 81 for (row=0; row<m; row++) { 82 mr = a->i[row+1] - a->i[row]; 83 for (i=0; i<mr; i++) { 84 col = *jj++; 85 cja[cia[col] + collengths[col]++ - oshift] = row + oshift; 86 } 87 } 88 ierr = PetscFree(collengths);CHKERRQ(ierr); 89 *ia = cia; *ja = cja; 90 } 91 PetscFunctionReturn(0); 92 } 93 94 #undef __FUNCT__ 95 #define __FUNCT__ "MatRestoreColumnIJ_SeqAIJ" 96 int MatRestoreColumnIJ_SeqAIJ(Mat A,int oshift,PetscTruth symmetric,int *n,int *ia[],int *ja[],PetscTruth *done) 97 { 98 int ierr; 99 100 PetscFunctionBegin; 101 if (!ia) PetscFunctionReturn(0); 102 103 ierr = PetscFree(*ia);CHKERRQ(ierr); 104 ierr = PetscFree(*ja);CHKERRQ(ierr); 105 106 PetscFunctionReturn(0); 107 } 108 109 #define CHUNKSIZE 15 110 111 #undef __FUNCT__ 112 #define __FUNCT__ "MatSetValues_SeqAIJ" 113 int MatSetValues_SeqAIJ(Mat A,int m,const int im[],int n,const int in[],const PetscScalar v[],InsertMode is) 114 { 115 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 116 int *rp,k,low,high,t,ii,row,nrow,i,col,l,rmax,N,sorted = a->sorted; 117 int *imax = a->imax,*ai = a->i,*ailen = a->ilen; 118 int *aj = a->j,nonew = a->nonew,ierr; 119 PetscScalar *ap,value,*aa = a->a; 120 PetscTruth ignorezeroentries = ((a->ignorezeroentries && is == ADD_VALUES) ? PETSC_TRUE:PETSC_FALSE); 121 PetscTruth roworiented = a->roworiented; 122 123 PetscFunctionBegin; 124 for (k=0; k<m; k++) { /* loop over added rows */ 125 row = im[k]; 126 if (row < 0) continue; 127 #if defined(PETSC_USE_BOPT_g) 128 if (row >= A->m) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %d max %d",row,A->m-1); 129 #endif 130 rp = aj + ai[row]; ap = aa + ai[row]; 131 rmax = imax[row]; nrow = ailen[row]; 132 low = 0; 133 for (l=0; l<n; l++) { /* loop over added columns */ 134 if (in[l] < 0) continue; 135 #if defined(PETSC_USE_BOPT_g) 136 if (in[l] >= A->n) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %d max %d",in[l],A->n-1); 137 #endif 138 col = in[l]; 139 if (roworiented) { 140 value = v[l + k*n]; 141 } else { 142 value = v[k + l*m]; 143 } 144 if (value == 0.0 && ignorezeroentries) continue; 145 146 if (!sorted) low = 0; high = nrow; 147 while (high-low > 5) { 148 t = (low+high)/2; 149 if (rp[t] > col) high = t; 150 else low = t; 151 } 152 for (i=low; i<high; i++) { 153 if (rp[i] > col) break; 154 if (rp[i] == col) { 155 if (is == ADD_VALUES) ap[i] += value; 156 else ap[i] = value; 157 goto noinsert; 158 } 159 } 160 if (nonew == 1) goto noinsert; 161 else if (nonew == -1) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero at (%d,%d) in the matrix",row,col); 162 if (nrow >= rmax) { 163 /* there is no extra room in row, therefore enlarge */ 164 int new_nz = ai[A->m] + CHUNKSIZE,*new_i,*new_j; 165 size_t len; 166 PetscScalar *new_a; 167 168 if (nonew == -2) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Inserting a new nonzero at (%d,%d) in the matrix requiring new malloc()",row,col); 169 170 /* malloc new storage space */ 171 len = ((size_t) new_nz)*(sizeof(int)+sizeof(PetscScalar))+(A->m+1)*sizeof(int); 172 ierr = PetscMalloc(len,&new_a);CHKERRQ(ierr); 173 new_j = (int*)(new_a + new_nz); 174 new_i = new_j + new_nz; 175 176 /* copy over old data into new slots */ 177 for (ii=0; ii<row+1; ii++) {new_i[ii] = ai[ii];} 178 for (ii=row+1; ii<A->m+1; ii++) {new_i[ii] = ai[ii]+CHUNKSIZE;} 179 ierr = PetscMemcpy(new_j,aj,(ai[row]+nrow)*sizeof(int));CHKERRQ(ierr); 180 len = (((size_t) new_nz) - CHUNKSIZE - ai[row] - nrow ); 181 ierr = PetscMemcpy(new_j+ai[row]+nrow+CHUNKSIZE,aj+ai[row]+nrow,len*sizeof(int));CHKERRQ(ierr); 182 ierr = PetscMemcpy(new_a,aa,(((size_t) ai[row])+nrow)*sizeof(PetscScalar));CHKERRQ(ierr); 183 ierr = PetscMemcpy(new_a+ai[row]+nrow+CHUNKSIZE,aa+ai[row]+nrow,len*sizeof(PetscScalar));CHKERRQ(ierr); 184 /* free up old matrix storage */ 185 ierr = PetscFree(a->a);CHKERRQ(ierr); 186 if (!a->singlemalloc) { 187 ierr = PetscFree(a->i);CHKERRQ(ierr); 188 ierr = PetscFree(a->j);CHKERRQ(ierr); 189 } 190 aa = a->a = new_a; ai = a->i = new_i; aj = a->j = new_j; 191 a->singlemalloc = PETSC_TRUE; 192 193 rp = aj + ai[row]; ap = aa + ai[row] ; 194 rmax = imax[row] = imax[row] + CHUNKSIZE; 195 PetscLogObjectMemory(A,CHUNKSIZE*(sizeof(int) + sizeof(PetscScalar))); 196 a->maxnz += CHUNKSIZE; 197 a->reallocs++; 198 } 199 N = nrow++ - 1; a->nz++; 200 /* shift up all the later entries in this row */ 201 for (ii=N; ii>=i; ii--) { 202 rp[ii+1] = rp[ii]; 203 ap[ii+1] = ap[ii]; 204 } 205 rp[i] = col; 206 ap[i] = value; 207 noinsert:; 208 low = i + 1; 209 } 210 ailen[row] = nrow; 211 } 212 PetscFunctionReturn(0); 213 } 214 215 #undef __FUNCT__ 216 #define __FUNCT__ "MatGetValues_SeqAIJ" 217 int MatGetValues_SeqAIJ(Mat A,int m,const int im[],int n,const int in[],PetscScalar v[]) 218 { 219 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 220 int *rp,k,low,high,t,row,nrow,i,col,l,*aj = a->j; 221 int *ai = a->i,*ailen = a->ilen; 222 PetscScalar *ap,*aa = a->a,zero = 0.0; 223 224 PetscFunctionBegin; 225 for (k=0; k<m; k++) { /* loop over rows */ 226 row = im[k]; 227 if (row < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Negative row: %d",row); 228 if (row >= A->m) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Row too large: row %d max %d",row,A->m-1); 229 rp = aj + ai[row]; ap = aa + ai[row]; 230 nrow = ailen[row]; 231 for (l=0; l<n; l++) { /* loop over columns */ 232 if (in[l] < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Negative column: %d",in[l]); 233 if (in[l] >= A->n) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"Column too large: col %d max %d",in[l],A->n-1); 234 col = in[l] ; 235 high = nrow; low = 0; /* assume unsorted */ 236 while (high-low > 5) { 237 t = (low+high)/2; 238 if (rp[t] > col) high = t; 239 else low = t; 240 } 241 for (i=low; i<high; i++) { 242 if (rp[i] > col) break; 243 if (rp[i] == col) { 244 *v++ = ap[i]; 245 goto finished; 246 } 247 } 248 *v++ = zero; 249 finished:; 250 } 251 } 252 PetscFunctionReturn(0); 253 } 254 255 256 #undef __FUNCT__ 257 #define __FUNCT__ "MatView_SeqAIJ_Binary" 258 int MatView_SeqAIJ_Binary(Mat A,PetscViewer viewer) 259 { 260 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 261 int i,fd,*col_lens,ierr; 262 263 PetscFunctionBegin; 264 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 265 ierr = PetscMalloc((4+A->m)*sizeof(int),&col_lens);CHKERRQ(ierr); 266 col_lens[0] = MAT_FILE_COOKIE; 267 col_lens[1] = A->m; 268 col_lens[2] = A->n; 269 col_lens[3] = a->nz; 270 271 /* store lengths of each row and write (including header) to file */ 272 for (i=0; i<A->m; i++) { 273 col_lens[4+i] = a->i[i+1] - a->i[i]; 274 } 275 ierr = PetscBinaryWrite(fd,col_lens,4+A->m,PETSC_INT,1);CHKERRQ(ierr); 276 ierr = PetscFree(col_lens);CHKERRQ(ierr); 277 278 /* store column indices (zero start index) */ 279 ierr = PetscBinaryWrite(fd,a->j,a->nz,PETSC_INT,0);CHKERRQ(ierr); 280 281 /* store nonzero values */ 282 ierr = PetscBinaryWrite(fd,a->a,a->nz,PETSC_SCALAR,0);CHKERRQ(ierr); 283 PetscFunctionReturn(0); 284 } 285 286 extern int MatSeqAIJFactorInfo_Matlab(Mat,PetscViewer); 287 288 #undef __FUNCT__ 289 #define __FUNCT__ "MatView_SeqAIJ_ASCII" 290 int MatView_SeqAIJ_ASCII(Mat A,PetscViewer viewer) 291 { 292 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 293 int ierr,i,j,m = A->m,shift=0; 294 char *name; 295 PetscViewerFormat format; 296 297 PetscFunctionBegin; 298 ierr = PetscObjectGetName((PetscObject)A,&name);CHKERRQ(ierr); 299 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 300 if (format == PETSC_VIEWER_ASCII_INFO_DETAIL || format == PETSC_VIEWER_ASCII_INFO) { 301 if (a->inode.size) { 302 ierr = PetscViewerASCIIPrintf(viewer,"using I-node routines: found %d nodes, limit used is %d\n",a->inode.node_count,a->inode.limit);CHKERRQ(ierr); 303 } else { 304 ierr = PetscViewerASCIIPrintf(viewer,"not using I-node routines\n");CHKERRQ(ierr); 305 } 306 } else if (format == PETSC_VIEWER_ASCII_MATLAB) { 307 int nofinalvalue = 0; 308 if ((a->i[m] == a->i[m-1]) || (a->j[a->nz-1] != A->n-!shift)) { 309 nofinalvalue = 1; 310 } 311 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_NO);CHKERRQ(ierr); 312 ierr = PetscViewerASCIIPrintf(viewer,"%% Size = %d %d \n",m,A->n);CHKERRQ(ierr); 313 ierr = PetscViewerASCIIPrintf(viewer,"%% Nonzeros = %d \n",a->nz);CHKERRQ(ierr); 314 ierr = PetscViewerASCIIPrintf(viewer,"zzz = zeros(%d,3);\n",a->nz+nofinalvalue);CHKERRQ(ierr); 315 ierr = PetscViewerASCIIPrintf(viewer,"zzz = [\n");CHKERRQ(ierr); 316 317 for (i=0; i<m; i++) { 318 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 319 #if defined(PETSC_USE_COMPLEX) 320 ierr = PetscViewerASCIIPrintf(viewer,"%d %d %18.16e + %18.16ei \n",i+1,a->j[j]+!shift,PetscRealPart(a->a[j]),PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 321 #else 322 ierr = PetscViewerASCIIPrintf(viewer,"%d %d %18.16e\n",i+1,a->j[j]+!shift,a->a[j]);CHKERRQ(ierr); 323 #endif 324 } 325 } 326 if (nofinalvalue) { 327 ierr = PetscViewerASCIIPrintf(viewer,"%d %d %18.16e\n",m,A->n,0.0);CHKERRQ(ierr); 328 } 329 ierr = PetscViewerASCIIPrintf(viewer,"];\n %s = spconvert(zzz);\n",name);CHKERRQ(ierr); 330 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_YES);CHKERRQ(ierr); 331 } else if (format == PETSC_VIEWER_ASCII_FACTOR_INFO) { 332 PetscFunctionReturn(0); 333 } else if (format == PETSC_VIEWER_ASCII_COMMON) { 334 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_NO);CHKERRQ(ierr); 335 for (i=0; i<m; i++) { 336 ierr = PetscViewerASCIIPrintf(viewer,"row %d:",i);CHKERRQ(ierr); 337 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 338 #if defined(PETSC_USE_COMPLEX) 339 if (PetscImaginaryPart(a->a[j]) > 0.0 && PetscRealPart(a->a[j]) != 0.0) { 340 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g + %g i)",a->j[j]+shift,PetscRealPart(a->a[j]),PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 341 } else if (PetscImaginaryPart(a->a[j]) < 0.0 && PetscRealPart(a->a[j]) != 0.0) { 342 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g - %g i)",a->j[j]+shift,PetscRealPart(a->a[j]),-PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 343 } else if (PetscRealPart(a->a[j]) != 0.0) { 344 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g) ",a->j[j]+shift,PetscRealPart(a->a[j]));CHKERRQ(ierr); 345 } 346 #else 347 if (a->a[j] != 0.0) {ierr = PetscViewerASCIIPrintf(viewer," (%d, %g) ",a->j[j]+shift,a->a[j]);CHKERRQ(ierr);} 348 #endif 349 } 350 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 351 } 352 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_YES);CHKERRQ(ierr); 353 } else if (format == PETSC_VIEWER_ASCII_SYMMODU) { 354 int nzd=0,fshift=1,*sptr; 355 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_NO);CHKERRQ(ierr); 356 ierr = PetscMalloc((m+1)*sizeof(int),&sptr);CHKERRQ(ierr); 357 for (i=0; i<m; i++) { 358 sptr[i] = nzd+1; 359 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 360 if (a->j[j] >= i) { 361 #if defined(PETSC_USE_COMPLEX) 362 if (PetscImaginaryPart(a->a[j]) != 0.0 || PetscRealPart(a->a[j]) != 0.0) nzd++; 363 #else 364 if (a->a[j] != 0.0) nzd++; 365 #endif 366 } 367 } 368 } 369 sptr[m] = nzd+1; 370 ierr = PetscViewerASCIIPrintf(viewer," %d %d\n\n",m,nzd);CHKERRQ(ierr); 371 for (i=0; i<m+1; i+=6) { 372 if (i+4<m) {ierr = PetscViewerASCIIPrintf(viewer," %d %d %d %d %d %d\n",sptr[i],sptr[i+1],sptr[i+2],sptr[i+3],sptr[i+4],sptr[i+5]);CHKERRQ(ierr);} 373 else if (i+3<m) {ierr = PetscViewerASCIIPrintf(viewer," %d %d %d %d %d\n",sptr[i],sptr[i+1],sptr[i+2],sptr[i+3],sptr[i+4]);CHKERRQ(ierr);} 374 else if (i+2<m) {ierr = PetscViewerASCIIPrintf(viewer," %d %d %d %d\n",sptr[i],sptr[i+1],sptr[i+2],sptr[i+3]);CHKERRQ(ierr);} 375 else if (i+1<m) {ierr = PetscViewerASCIIPrintf(viewer," %d %d %d\n",sptr[i],sptr[i+1],sptr[i+2]);CHKERRQ(ierr);} 376 else if (i<m) {ierr = PetscViewerASCIIPrintf(viewer," %d %d\n",sptr[i],sptr[i+1]);CHKERRQ(ierr);} 377 else {ierr = PetscViewerASCIIPrintf(viewer," %d\n",sptr[i]);CHKERRQ(ierr);} 378 } 379 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 380 ierr = PetscFree(sptr);CHKERRQ(ierr); 381 for (i=0; i<m; i++) { 382 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 383 if (a->j[j] >= i) {ierr = PetscViewerASCIIPrintf(viewer," %d ",a->j[j]+fshift);CHKERRQ(ierr);} 384 } 385 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 386 } 387 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 388 for (i=0; i<m; i++) { 389 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 390 if (a->j[j] >= i) { 391 #if defined(PETSC_USE_COMPLEX) 392 if (PetscImaginaryPart(a->a[j]) != 0.0 || PetscRealPart(a->a[j]) != 0.0) { 393 ierr = PetscViewerASCIIPrintf(viewer," %18.16e %18.16e ",PetscRealPart(a->a[j]),PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 394 } 395 #else 396 if (a->a[j] != 0.0) {ierr = PetscViewerASCIIPrintf(viewer," %18.16e ",a->a[j]);CHKERRQ(ierr);} 397 #endif 398 } 399 } 400 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 401 } 402 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_YES);CHKERRQ(ierr); 403 } else if (format == PETSC_VIEWER_ASCII_DENSE) { 404 int cnt = 0,jcnt; 405 PetscScalar value; 406 407 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_NO);CHKERRQ(ierr); 408 for (i=0; i<m; i++) { 409 jcnt = 0; 410 for (j=0; j<A->n; j++) { 411 if (jcnt < a->i[i+1]-a->i[i] && j == a->j[cnt]) { 412 value = a->a[cnt++]; 413 jcnt++; 414 } else { 415 value = 0.0; 416 } 417 #if defined(PETSC_USE_COMPLEX) 418 ierr = PetscViewerASCIIPrintf(viewer," %7.5e+%7.5e i ",PetscRealPart(value),PetscImaginaryPart(value));CHKERRQ(ierr); 419 #else 420 ierr = PetscViewerASCIIPrintf(viewer," %7.5e ",value);CHKERRQ(ierr); 421 #endif 422 } 423 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 424 } 425 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_YES);CHKERRQ(ierr); 426 } else { 427 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_NO);CHKERRQ(ierr); 428 for (i=0; i<m; i++) { 429 ierr = PetscViewerASCIIPrintf(viewer,"row %d:",i);CHKERRQ(ierr); 430 for (j=a->i[i]+shift; j<a->i[i+1]+shift; j++) { 431 #if defined(PETSC_USE_COMPLEX) 432 if (PetscImaginaryPart(a->a[j]) > 0.0) { 433 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g + %g i)",a->j[j]+shift,PetscRealPart(a->a[j]),PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 434 } else if (PetscImaginaryPart(a->a[j]) < 0.0) { 435 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g - %g i)",a->j[j]+shift,PetscRealPart(a->a[j]),-PetscImaginaryPart(a->a[j]));CHKERRQ(ierr); 436 } else { 437 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g) ",a->j[j]+shift,PetscRealPart(a->a[j]));CHKERRQ(ierr); 438 } 439 #else 440 ierr = PetscViewerASCIIPrintf(viewer," (%d, %g) ",a->j[j]+shift,a->a[j]);CHKERRQ(ierr); 441 #endif 442 } 443 ierr = PetscViewerASCIIPrintf(viewer,"\n");CHKERRQ(ierr); 444 } 445 ierr = PetscViewerASCIIUseTabs(viewer,PETSC_YES);CHKERRQ(ierr); 446 } 447 ierr = PetscViewerFlush(viewer);CHKERRQ(ierr); 448 PetscFunctionReturn(0); 449 } 450 451 #undef __FUNCT__ 452 #define __FUNCT__ "MatView_SeqAIJ_Draw_Zoom" 453 int MatView_SeqAIJ_Draw_Zoom(PetscDraw draw,void *Aa) 454 { 455 Mat A = (Mat) Aa; 456 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 457 int ierr,i,j,m = A->m,color; 458 PetscReal xl,yl,xr,yr,x_l,x_r,y_l,y_r,maxv = 0.0; 459 PetscViewer viewer; 460 PetscViewerFormat format; 461 462 PetscFunctionBegin; 463 ierr = PetscObjectQuery((PetscObject)A,"Zoomviewer",(PetscObject*)&viewer);CHKERRQ(ierr); 464 ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr); 465 466 ierr = PetscDrawGetCoordinates(draw,&xl,&yl,&xr,&yr);CHKERRQ(ierr); 467 /* loop over matrix elements drawing boxes */ 468 469 if (format != PETSC_VIEWER_DRAW_CONTOUR) { 470 /* Blue for negative, Cyan for zero and Red for positive */ 471 color = PETSC_DRAW_BLUE; 472 for (i=0; i<m; i++) { 473 y_l = m - i - 1.0; y_r = y_l + 1.0; 474 for (j=a->i[i]; j<a->i[i+1]; j++) { 475 x_l = a->j[j] ; x_r = x_l + 1.0; 476 #if defined(PETSC_USE_COMPLEX) 477 if (PetscRealPart(a->a[j]) >= 0.) continue; 478 #else 479 if (a->a[j] >= 0.) continue; 480 #endif 481 ierr = PetscDrawRectangle(draw,x_l,y_l,x_r,y_r,color,color,color,color);CHKERRQ(ierr); 482 } 483 } 484 color = PETSC_DRAW_CYAN; 485 for (i=0; i<m; i++) { 486 y_l = m - i - 1.0; y_r = y_l + 1.0; 487 for (j=a->i[i]; j<a->i[i+1]; j++) { 488 x_l = a->j[j]; x_r = x_l + 1.0; 489 if (a->a[j] != 0.) continue; 490 ierr = PetscDrawRectangle(draw,x_l,y_l,x_r,y_r,color,color,color,color);CHKERRQ(ierr); 491 } 492 } 493 color = PETSC_DRAW_RED; 494 for (i=0; i<m; i++) { 495 y_l = m - i - 1.0; y_r = y_l + 1.0; 496 for (j=a->i[i]; j<a->i[i+1]; j++) { 497 x_l = a->j[j]; x_r = x_l + 1.0; 498 #if defined(PETSC_USE_COMPLEX) 499 if (PetscRealPart(a->a[j]) <= 0.) continue; 500 #else 501 if (a->a[j] <= 0.) continue; 502 #endif 503 ierr = PetscDrawRectangle(draw,x_l,y_l,x_r,y_r,color,color,color,color);CHKERRQ(ierr); 504 } 505 } 506 } else { 507 /* use contour shading to indicate magnitude of values */ 508 /* first determine max of all nonzero values */ 509 int nz = a->nz,count; 510 PetscDraw popup; 511 PetscReal scale; 512 513 for (i=0; i<nz; i++) { 514 if (PetscAbsScalar(a->a[i]) > maxv) maxv = PetscAbsScalar(a->a[i]); 515 } 516 scale = (245.0 - PETSC_DRAW_BASIC_COLORS)/maxv; 517 ierr = PetscDrawGetPopup(draw,&popup);CHKERRQ(ierr); 518 if (popup) {ierr = PetscDrawScalePopup(popup,0.0,maxv);CHKERRQ(ierr);} 519 count = 0; 520 for (i=0; i<m; i++) { 521 y_l = m - i - 1.0; y_r = y_l + 1.0; 522 for (j=a->i[i]; j<a->i[i+1]; j++) { 523 x_l = a->j[j]; x_r = x_l + 1.0; 524 color = PETSC_DRAW_BASIC_COLORS + (int)(scale*PetscAbsScalar(a->a[count])); 525 ierr = PetscDrawRectangle(draw,x_l,y_l,x_r,y_r,color,color,color,color);CHKERRQ(ierr); 526 count++; 527 } 528 } 529 } 530 PetscFunctionReturn(0); 531 } 532 533 #undef __FUNCT__ 534 #define __FUNCT__ "MatView_SeqAIJ_Draw" 535 int MatView_SeqAIJ_Draw(Mat A,PetscViewer viewer) 536 { 537 int ierr; 538 PetscDraw draw; 539 PetscReal xr,yr,xl,yl,h,w; 540 PetscTruth isnull; 541 542 PetscFunctionBegin; 543 ierr = PetscViewerDrawGetDraw(viewer,0,&draw);CHKERRQ(ierr); 544 ierr = PetscDrawIsNull(draw,&isnull);CHKERRQ(ierr); 545 if (isnull) PetscFunctionReturn(0); 546 547 ierr = PetscObjectCompose((PetscObject)A,"Zoomviewer",(PetscObject)viewer);CHKERRQ(ierr); 548 xr = A->n; yr = A->m; h = yr/10.0; w = xr/10.0; 549 xr += w; yr += h; xl = -w; yl = -h; 550 ierr = PetscDrawSetCoordinates(draw,xl,yl,xr,yr);CHKERRQ(ierr); 551 ierr = PetscDrawZoom(draw,MatView_SeqAIJ_Draw_Zoom,A);CHKERRQ(ierr); 552 ierr = PetscObjectCompose((PetscObject)A,"Zoomviewer",PETSC_NULL);CHKERRQ(ierr); 553 PetscFunctionReturn(0); 554 } 555 556 #undef __FUNCT__ 557 #define __FUNCT__ "MatView_SeqAIJ" 558 int MatView_SeqAIJ(Mat A,PetscViewer viewer) 559 { 560 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 561 int ierr; 562 PetscTruth issocket,iascii,isbinary,isdraw; 563 564 PetscFunctionBegin; 565 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_SOCKET,&issocket);CHKERRQ(ierr); 566 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_ASCII,&iascii);CHKERRQ(ierr); 567 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_BINARY,&isbinary);CHKERRQ(ierr); 568 ierr = PetscTypeCompare((PetscObject)viewer,PETSC_VIEWER_DRAW,&isdraw);CHKERRQ(ierr); 569 if (issocket) { 570 ierr = PetscViewerSocketPutSparse_Private(viewer,A->m,A->n,a->nz,a->a,a->i,a->j);CHKERRQ(ierr); 571 } else if (iascii) { 572 ierr = MatView_SeqAIJ_ASCII(A,viewer);CHKERRQ(ierr); 573 } else if (isbinary) { 574 ierr = MatView_SeqAIJ_Binary(A,viewer);CHKERRQ(ierr); 575 } else if (isdraw) { 576 ierr = MatView_SeqAIJ_Draw(A,viewer);CHKERRQ(ierr); 577 } else { 578 SETERRQ1(1,"Viewer type %s not supported by SeqAIJ matrices",((PetscObject)viewer)->type_name); 579 } 580 PetscFunctionReturn(0); 581 } 582 583 #undef __FUNCT__ 584 #define __FUNCT__ "MatAssemblyEnd_SeqAIJ" 585 int MatAssemblyEnd_SeqAIJ(Mat A,MatAssemblyType mode) 586 { 587 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 588 int fshift = 0,i,j,*ai = a->i,*aj = a->j,*imax = a->imax,ierr; 589 int m = A->m,*ip,N,*ailen = a->ilen,rmax = 0; 590 PetscScalar *aa = a->a,*ap; 591 592 PetscFunctionBegin; 593 if (mode == MAT_FLUSH_ASSEMBLY) PetscFunctionReturn(0); 594 595 if (m) rmax = ailen[0]; /* determine row with most nonzeros */ 596 for (i=1; i<m; i++) { 597 /* move each row back by the amount of empty slots (fshift) before it*/ 598 fshift += imax[i-1] - ailen[i-1]; 599 rmax = PetscMax(rmax,ailen[i]); 600 if (fshift) { 601 ip = aj + ai[i] ; 602 ap = aa + ai[i] ; 603 N = ailen[i]; 604 for (j=0; j<N; j++) { 605 ip[j-fshift] = ip[j]; 606 ap[j-fshift] = ap[j]; 607 } 608 } 609 ai[i] = ai[i-1] + ailen[i-1]; 610 } 611 if (m) { 612 fshift += imax[m-1] - ailen[m-1]; 613 ai[m] = ai[m-1] + ailen[m-1]; 614 } 615 /* reset ilen and imax for each row */ 616 for (i=0; i<m; i++) { 617 ailen[i] = imax[i] = ai[i+1] - ai[i]; 618 } 619 a->nz = ai[m]; 620 621 /* diagonals may have moved, so kill the diagonal pointers */ 622 if (fshift && a->diag) { 623 ierr = PetscFree(a->diag);CHKERRQ(ierr); 624 PetscLogObjectMemory(A,-(m+1)*sizeof(int)); 625 a->diag = 0; 626 } 627 PetscLogInfo(A,"MatAssemblyEnd_SeqAIJ:Matrix size: %d X %d; storage space: %d unneeded,%d used\n",m,A->n,fshift,a->nz); 628 PetscLogInfo(A,"MatAssemblyEnd_SeqAIJ:Number of mallocs during MatSetValues() is %d\n",a->reallocs); 629 PetscLogInfo(A,"MatAssemblyEnd_SeqAIJ:Most nonzeros in any row is %d\n",rmax); 630 a->reallocs = 0; 631 A->info.nz_unneeded = (double)fshift; 632 a->rmax = rmax; 633 634 /* check out for identical nodes. If found, use inode functions */ 635 ierr = Mat_AIJ_CheckInode(A,(PetscTruth)(!fshift));CHKERRQ(ierr); 636 637 PetscFunctionReturn(0); 638 } 639 640 #undef __FUNCT__ 641 #define __FUNCT__ "MatZeroEntries_SeqAIJ" 642 int MatZeroEntries_SeqAIJ(Mat A) 643 { 644 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 645 int ierr; 646 647 PetscFunctionBegin; 648 ierr = PetscMemzero(a->a,(a->i[A->m])*sizeof(PetscScalar));CHKERRQ(ierr); 649 PetscFunctionReturn(0); 650 } 651 652 #undef __FUNCT__ 653 #define __FUNCT__ "MatDestroy_SeqAIJ" 654 int MatDestroy_SeqAIJ(Mat A) 655 { 656 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 657 int ierr; 658 659 PetscFunctionBegin; 660 #if defined(PETSC_USE_LOG) 661 PetscLogObjectState((PetscObject)A,"Rows=%d, Cols=%d, NZ=%d",A->m,A->n,a->nz); 662 #endif 663 if (a->freedata) { 664 ierr = PetscFree(a->a);CHKERRQ(ierr); 665 if (!a->singlemalloc) { 666 ierr = PetscFree(a->i);CHKERRQ(ierr); 667 ierr = PetscFree(a->j);CHKERRQ(ierr); 668 } 669 } 670 if (a->row) { 671 ierr = ISDestroy(a->row);CHKERRQ(ierr); 672 } 673 if (a->col) { 674 ierr = ISDestroy(a->col);CHKERRQ(ierr); 675 } 676 if (a->diag) {ierr = PetscFree(a->diag);CHKERRQ(ierr);} 677 if (a->ilen) {ierr = PetscFree(a->ilen);CHKERRQ(ierr);} 678 if (a->imax) {ierr = PetscFree(a->imax);CHKERRQ(ierr);} 679 if (a->idiag) {ierr = PetscFree(a->idiag);CHKERRQ(ierr);} 680 if (a->solve_work) {ierr = PetscFree(a->solve_work);CHKERRQ(ierr);} 681 if (a->inode.size) {ierr = PetscFree(a->inode.size);CHKERRQ(ierr);} 682 if (a->icol) {ierr = ISDestroy(a->icol);CHKERRQ(ierr);} 683 if (a->saved_values) {ierr = PetscFree(a->saved_values);CHKERRQ(ierr);} 684 if (a->coloring) {ierr = ISColoringDestroy(a->coloring);CHKERRQ(ierr);} 685 if (a->xtoy) {ierr = PetscFree(a->xtoy);CHKERRQ(ierr);} 686 687 ierr = PetscFree(a);CHKERRQ(ierr); 688 PetscFunctionReturn(0); 689 } 690 691 #undef __FUNCT__ 692 #define __FUNCT__ "MatCompress_SeqAIJ" 693 int MatCompress_SeqAIJ(Mat A) 694 { 695 PetscFunctionBegin; 696 PetscFunctionReturn(0); 697 } 698 699 #undef __FUNCT__ 700 #define __FUNCT__ "MatSetOption_SeqAIJ" 701 int MatSetOption_SeqAIJ(Mat A,MatOption op) 702 { 703 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 704 705 PetscFunctionBegin; 706 switch (op) { 707 case MAT_ROW_ORIENTED: 708 a->roworiented = PETSC_TRUE; 709 break; 710 case MAT_KEEP_ZEROED_ROWS: 711 a->keepzeroedrows = PETSC_TRUE; 712 break; 713 case MAT_COLUMN_ORIENTED: 714 a->roworiented = PETSC_FALSE; 715 break; 716 case MAT_COLUMNS_SORTED: 717 a->sorted = PETSC_TRUE; 718 break; 719 case MAT_COLUMNS_UNSORTED: 720 a->sorted = PETSC_FALSE; 721 break; 722 case MAT_NO_NEW_NONZERO_LOCATIONS: 723 a->nonew = 1; 724 break; 725 case MAT_NEW_NONZERO_LOCATION_ERR: 726 a->nonew = -1; 727 break; 728 case MAT_NEW_NONZERO_ALLOCATION_ERR: 729 a->nonew = -2; 730 break; 731 case MAT_YES_NEW_NONZERO_LOCATIONS: 732 a->nonew = 0; 733 break; 734 case MAT_IGNORE_ZERO_ENTRIES: 735 a->ignorezeroentries = PETSC_TRUE; 736 break; 737 case MAT_USE_INODES: 738 a->inode.use = PETSC_TRUE; 739 break; 740 case MAT_DO_NOT_USE_INODES: 741 a->inode.use = PETSC_FALSE; 742 break; 743 case MAT_ROWS_SORTED: 744 case MAT_ROWS_UNSORTED: 745 case MAT_YES_NEW_DIAGONALS: 746 case MAT_IGNORE_OFF_PROC_ENTRIES: 747 case MAT_USE_HASH_TABLE: 748 PetscLogInfo(A,"MatSetOption_SeqAIJ:Option ignored\n"); 749 break; 750 case MAT_NO_NEW_DIAGONALS: 751 SETERRQ(PETSC_ERR_SUP,"MAT_NO_NEW_DIAGONALS"); 752 case MAT_INODE_LIMIT_1: 753 a->inode.limit = 1; 754 break; 755 case MAT_INODE_LIMIT_2: 756 a->inode.limit = 2; 757 break; 758 case MAT_INODE_LIMIT_3: 759 a->inode.limit = 3; 760 break; 761 case MAT_INODE_LIMIT_4: 762 a->inode.limit = 4; 763 break; 764 case MAT_INODE_LIMIT_5: 765 a->inode.limit = 5; 766 break; 767 case MAT_SYMMETRIC: 768 case MAT_STRUCTURALLY_SYMMETRIC: 769 case MAT_NOT_SYMMETRIC: 770 case MAT_NOT_STRUCTURALLY_SYMMETRIC: 771 case MAT_HERMITIAN: 772 case MAT_NOT_HERMITIAN: 773 case MAT_SYMMETRY_ETERNAL: 774 case MAT_NOT_SYMMETRY_ETERNAL: 775 break; 776 default: 777 SETERRQ(PETSC_ERR_SUP,"unknown option"); 778 } 779 PetscFunctionReturn(0); 780 } 781 782 #undef __FUNCT__ 783 #define __FUNCT__ "MatGetDiagonal_SeqAIJ" 784 int MatGetDiagonal_SeqAIJ(Mat A,Vec v) 785 { 786 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 787 int i,j,n,ierr; 788 PetscScalar *x,zero = 0.0; 789 790 PetscFunctionBegin; 791 ierr = VecSet(&zero,v);CHKERRQ(ierr); 792 ierr = VecGetArray(v,&x);CHKERRQ(ierr); 793 ierr = VecGetLocalSize(v,&n);CHKERRQ(ierr); 794 if (n != A->m) SETERRQ(PETSC_ERR_ARG_SIZ,"Nonconforming matrix and vector"); 795 for (i=0; i<A->m; i++) { 796 for (j=a->i[i]; j<a->i[i+1]; j++) { 797 if (a->j[j] == i) { 798 x[i] = a->a[j]; 799 break; 800 } 801 } 802 } 803 ierr = VecRestoreArray(v,&x);CHKERRQ(ierr); 804 PetscFunctionReturn(0); 805 } 806 807 808 #undef __FUNCT__ 809 #define __FUNCT__ "MatMultTransposeAdd_SeqAIJ" 810 int MatMultTransposeAdd_SeqAIJ(Mat A,Vec xx,Vec zz,Vec yy) 811 { 812 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 813 PetscScalar *x,*y; 814 int ierr,m = A->m; 815 #if !defined(PETSC_USE_FORTRAN_KERNEL_MULTTRANSPOSEAIJ) 816 PetscScalar *v,alpha; 817 int n,i,*idx; 818 #endif 819 820 PetscFunctionBegin; 821 if (zz != yy) {ierr = VecCopy(zz,yy);CHKERRQ(ierr);} 822 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 823 ierr = VecGetArray(yy,&y);CHKERRQ(ierr); 824 825 #if defined(PETSC_USE_FORTRAN_KERNEL_MULTTRANSPOSEAIJ) 826 fortranmulttransposeaddaij_(&m,x,a->i,a->j,a->a,y); 827 #else 828 for (i=0; i<m; i++) { 829 idx = a->j + a->i[i] ; 830 v = a->a + a->i[i] ; 831 n = a->i[i+1] - a->i[i]; 832 alpha = x[i]; 833 while (n-->0) {y[*idx++] += alpha * *v++;} 834 } 835 #endif 836 PetscLogFlops(2*a->nz); 837 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 838 ierr = VecRestoreArray(yy,&y);CHKERRQ(ierr); 839 PetscFunctionReturn(0); 840 } 841 842 #undef __FUNCT__ 843 #define __FUNCT__ "MatMultTranspose_SeqAIJ" 844 int MatMultTranspose_SeqAIJ(Mat A,Vec xx,Vec yy) 845 { 846 PetscScalar zero = 0.0; 847 int ierr; 848 849 PetscFunctionBegin; 850 ierr = VecSet(&zero,yy);CHKERRQ(ierr); 851 ierr = MatMultTransposeAdd_SeqAIJ(A,xx,yy,yy);CHKERRQ(ierr); 852 PetscFunctionReturn(0); 853 } 854 855 856 #undef __FUNCT__ 857 #define __FUNCT__ "MatMult_SeqAIJ" 858 int MatMult_SeqAIJ(Mat A,Vec xx,Vec yy) 859 { 860 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 861 PetscScalar *x,*y,*v; 862 int ierr,m = A->m,*idx,*ii; 863 #if !defined(PETSC_USE_FORTRAN_KERNEL_MULTAIJ) 864 int n,i,jrow,j; 865 PetscScalar sum; 866 #endif 867 868 #if defined(PETSC_HAVE_PRAGMA_DISJOINT) 869 #pragma disjoint(*x,*y,*v) 870 #endif 871 872 PetscFunctionBegin; 873 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 874 ierr = VecGetArray(yy,&y);CHKERRQ(ierr); 875 idx = a->j; 876 v = a->a; 877 ii = a->i; 878 #if defined(PETSC_USE_FORTRAN_KERNEL_MULTAIJ) 879 fortranmultaij_(&m,x,ii,idx,v,y); 880 #else 881 for (i=0; i<m; i++) { 882 jrow = ii[i]; 883 n = ii[i+1] - jrow; 884 sum = 0.0; 885 for (j=0; j<n; j++) { 886 sum += v[jrow]*x[idx[jrow]]; jrow++; 887 } 888 y[i] = sum; 889 } 890 #endif 891 PetscLogFlops(2*a->nz - m); 892 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 893 ierr = VecRestoreArray(yy,&y);CHKERRQ(ierr); 894 PetscFunctionReturn(0); 895 } 896 897 #undef __FUNCT__ 898 #define __FUNCT__ "MatMultAdd_SeqAIJ" 899 int MatMultAdd_SeqAIJ(Mat A,Vec xx,Vec yy,Vec zz) 900 { 901 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 902 PetscScalar *x,*y,*z,*v; 903 int ierr,m = A->m,*idx,*ii; 904 #if !defined(PETSC_USE_FORTRAN_KERNEL_MULTADDAIJ) 905 int n,i,jrow,j; 906 PetscScalar sum; 907 #endif 908 909 PetscFunctionBegin; 910 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 911 ierr = VecGetArray(yy,&y);CHKERRQ(ierr); 912 if (zz != yy) { 913 ierr = VecGetArray(zz,&z);CHKERRQ(ierr); 914 } else { 915 z = y; 916 } 917 918 idx = a->j; 919 v = a->a; 920 ii = a->i; 921 #if defined(PETSC_USE_FORTRAN_KERNEL_MULTADDAIJ) 922 fortranmultaddaij_(&m,x,ii,idx,v,y,z); 923 #else 924 for (i=0; i<m; i++) { 925 jrow = ii[i]; 926 n = ii[i+1] - jrow; 927 sum = y[i]; 928 for (j=0; j<n; j++) { 929 sum += v[jrow]*x[idx[jrow]]; jrow++; 930 } 931 z[i] = sum; 932 } 933 #endif 934 PetscLogFlops(2*a->nz); 935 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 936 ierr = VecRestoreArray(yy,&y);CHKERRQ(ierr); 937 if (zz != yy) { 938 ierr = VecRestoreArray(zz,&z);CHKERRQ(ierr); 939 } 940 PetscFunctionReturn(0); 941 } 942 943 /* 944 Adds diagonal pointers to sparse matrix structure. 945 */ 946 #undef __FUNCT__ 947 #define __FUNCT__ "MatMarkDiagonal_SeqAIJ" 948 int MatMarkDiagonal_SeqAIJ(Mat A) 949 { 950 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 951 int i,j,*diag,m = A->m,ierr; 952 953 PetscFunctionBegin; 954 if (a->diag) PetscFunctionReturn(0); 955 956 ierr = PetscMalloc((m+1)*sizeof(int),&diag);CHKERRQ(ierr); 957 PetscLogObjectMemory(A,(m+1)*sizeof(int)); 958 for (i=0; i<A->m; i++) { 959 diag[i] = a->i[i+1]; 960 for (j=a->i[i]; j<a->i[i+1]; j++) { 961 if (a->j[j] == i) { 962 diag[i] = j; 963 break; 964 } 965 } 966 } 967 a->diag = diag; 968 PetscFunctionReturn(0); 969 } 970 971 /* 972 Checks for missing diagonals 973 */ 974 #undef __FUNCT__ 975 #define __FUNCT__ "MatMissingDiagonal_SeqAIJ" 976 int MatMissingDiagonal_SeqAIJ(Mat A) 977 { 978 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 979 int *diag,*jj = a->j,i,ierr; 980 981 PetscFunctionBegin; 982 ierr = MatMarkDiagonal_SeqAIJ(A);CHKERRQ(ierr); 983 diag = a->diag; 984 for (i=0; i<A->m; i++) { 985 if (jj[diag[i]] != i) { 986 SETERRQ1(1,"Matrix is missing diagonal number %d",i); 987 } 988 } 989 PetscFunctionReturn(0); 990 } 991 992 #undef __FUNCT__ 993 #define __FUNCT__ "MatRelax_SeqAIJ" 994 int MatRelax_SeqAIJ(Mat A,Vec bb,PetscReal omega,MatSORType flag,PetscReal fshift,int its,int lits,Vec xx) 995 { 996 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 997 PetscScalar *x,d,*xs,sum,*t,scale,*idiag=0,*mdiag; 998 const PetscScalar *v = a->a, *b, *bs,*xb, *ts; 999 int ierr,n = A->n,m = A->m,i; 1000 const int *idx,*diag; 1001 1002 PetscFunctionBegin; 1003 its = its*lits; 1004 if (its <= 0) SETERRQ2(PETSC_ERR_ARG_WRONG,"Relaxation requires global its %d and local its %d both positive",its,lits); 1005 1006 if (!a->diag) {ierr = MatMarkDiagonal_SeqAIJ(A);CHKERRQ(ierr);} 1007 diag = a->diag; 1008 if (!a->idiag) { 1009 ierr = PetscMalloc(3*m*sizeof(PetscScalar),&a->idiag);CHKERRQ(ierr); 1010 a->ssor = a->idiag + m; 1011 mdiag = a->ssor + m; 1012 1013 v = a->a; 1014 1015 /* this is wrong when fshift omega changes each iteration */ 1016 if (omega == 1.0 && fshift == 0.0) { 1017 for (i=0; i<m; i++) { 1018 mdiag[i] = v[diag[i]]; 1019 a->idiag[i] = 1.0/v[diag[i]]; 1020 } 1021 PetscLogFlops(m); 1022 } else { 1023 for (i=0; i<m; i++) { 1024 mdiag[i] = v[diag[i]]; 1025 a->idiag[i] = omega/(fshift + v[diag[i]]); 1026 } 1027 PetscLogFlops(2*m); 1028 } 1029 } 1030 t = a->ssor; 1031 idiag = a->idiag; 1032 mdiag = a->idiag + 2*m; 1033 1034 ierr = VecGetArray(xx,&x);CHKERRQ(ierr); 1035 if (xx != bb) { 1036 ierr = VecGetArray(bb,(PetscScalar**)&b);CHKERRQ(ierr); 1037 } else { 1038 b = x; 1039 } 1040 1041 /* We count flops by assuming the upper triangular and lower triangular parts have the same number of nonzeros */ 1042 xs = x; 1043 if (flag == SOR_APPLY_UPPER) { 1044 /* apply (U + D/omega) to the vector */ 1045 bs = b; 1046 for (i=0; i<m; i++) { 1047 d = fshift + a->a[diag[i]]; 1048 n = a->i[i+1] - diag[i] - 1; 1049 idx = a->j + diag[i] + 1; 1050 v = a->a + diag[i] + 1; 1051 sum = b[i]*d/omega; 1052 SPARSEDENSEDOT(sum,bs,v,idx,n); 1053 x[i] = sum; 1054 } 1055 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 1056 if (bb != xx) {ierr = VecRestoreArray(bb,(PetscScalar**)&b);CHKERRQ(ierr);} 1057 PetscLogFlops(a->nz); 1058 PetscFunctionReturn(0); 1059 } 1060 1061 1062 /* Let A = L + U + D; where L is lower trianglar, 1063 U is upper triangular, E is diagonal; This routine applies 1064 1065 (L + E)^{-1} A (U + E)^{-1} 1066 1067 to a vector efficiently using Eisenstat's trick. This is for 1068 the case of SSOR preconditioner, so E is D/omega where omega 1069 is the relaxation factor. 1070 */ 1071 1072 if (flag == SOR_APPLY_LOWER) { 1073 SETERRQ(PETSC_ERR_SUP,"SOR_APPLY_LOWER is not implemented"); 1074 } else if (flag & SOR_EISENSTAT) { 1075 /* Let A = L + U + D; where L is lower trianglar, 1076 U is upper triangular, E is diagonal; This routine applies 1077 1078 (L + E)^{-1} A (U + E)^{-1} 1079 1080 to a vector efficiently using Eisenstat's trick. This is for 1081 the case of SSOR preconditioner, so E is D/omega where omega 1082 is the relaxation factor. 1083 */ 1084 scale = (2.0/omega) - 1.0; 1085 1086 /* x = (E + U)^{-1} b */ 1087 for (i=m-1; i>=0; i--) { 1088 n = a->i[i+1] - diag[i] - 1; 1089 idx = a->j + diag[i] + 1; 1090 v = a->a + diag[i] + 1; 1091 sum = b[i]; 1092 SPARSEDENSEMDOT(sum,xs,v,idx,n); 1093 x[i] = sum*idiag[i]; 1094 } 1095 1096 /* t = b - (2*E - D)x */ 1097 v = a->a; 1098 for (i=0; i<m; i++) { t[i] = b[i] - scale*(v[*diag++])*x[i]; } 1099 1100 /* t = (E + L)^{-1}t */ 1101 ts = t; 1102 diag = a->diag; 1103 for (i=0; i<m; i++) { 1104 n = diag[i] - a->i[i]; 1105 idx = a->j + a->i[i]; 1106 v = a->a + a->i[i]; 1107 sum = t[i]; 1108 SPARSEDENSEMDOT(sum,ts,v,idx,n); 1109 t[i] = sum*idiag[i]; 1110 /* x = x + t */ 1111 x[i] += t[i]; 1112 } 1113 1114 PetscLogFlops(6*m-1 + 2*a->nz); 1115 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 1116 if (bb != xx) {ierr = VecRestoreArray(bb,(PetscScalar**)&b);CHKERRQ(ierr);} 1117 PetscFunctionReturn(0); 1118 } 1119 if (flag & SOR_ZERO_INITIAL_GUESS) { 1120 if (flag & SOR_FORWARD_SWEEP || flag & SOR_LOCAL_FORWARD_SWEEP){ 1121 #if defined(PETSC_USE_FORTRAN_KERNEL_RELAXAIJ) 1122 fortranrelaxaijforwardzero_(&m,&omega,x,a->i,a->j,(int *)diag,idiag,a->a,(void*)b); 1123 #else 1124 for (i=0; i<m; i++) { 1125 n = diag[i] - a->i[i]; 1126 idx = a->j + a->i[i]; 1127 v = a->a + a->i[i]; 1128 sum = b[i]; 1129 SPARSEDENSEMDOT(sum,xs,v,idx,n); 1130 x[i] = sum*idiag[i]; 1131 } 1132 #endif 1133 xb = x; 1134 PetscLogFlops(a->nz); 1135 } else xb = b; 1136 if ((flag & SOR_FORWARD_SWEEP || flag & SOR_LOCAL_FORWARD_SWEEP) && 1137 (flag & SOR_BACKWARD_SWEEP || flag & SOR_LOCAL_BACKWARD_SWEEP)) { 1138 for (i=0; i<m; i++) { 1139 x[i] *= mdiag[i]; 1140 } 1141 PetscLogFlops(m); 1142 } 1143 if (flag & SOR_BACKWARD_SWEEP || flag & SOR_LOCAL_BACKWARD_SWEEP){ 1144 #if defined(PETSC_USE_FORTRAN_KERNEL_RELAXAIJ) 1145 fortranrelaxaijbackwardzero_(&m,&omega,x,a->i,a->j,(int*)diag,idiag,a->a,(void*)xb); 1146 #else 1147 for (i=m-1; i>=0; i--) { 1148 n = a->i[i+1] - diag[i] - 1; 1149 idx = a->j + diag[i] + 1; 1150 v = a->a + diag[i] + 1; 1151 sum = xb[i]; 1152 SPARSEDENSEMDOT(sum,xs,v,idx,n); 1153 x[i] = sum*idiag[i]; 1154 } 1155 #endif 1156 PetscLogFlops(a->nz); 1157 } 1158 its--; 1159 } 1160 while (its--) { 1161 if (flag & SOR_FORWARD_SWEEP || flag & SOR_LOCAL_FORWARD_SWEEP){ 1162 #if defined(PETSC_USE_FORTRAN_KERNEL_RELAXAIJ) 1163 fortranrelaxaijforward_(&m,&omega,x,a->i,a->j,(int*)diag,a->a,(void*)b); 1164 #else 1165 for (i=0; i<m; i++) { 1166 d = fshift + a->a[diag[i]]; 1167 n = a->i[i+1] - a->i[i]; 1168 idx = a->j + a->i[i]; 1169 v = a->a + a->i[i]; 1170 sum = b[i]; 1171 SPARSEDENSEMDOT(sum,xs,v,idx,n); 1172 x[i] = (1. - omega)*x[i] + (sum + mdiag[i]*x[i])*idiag[i]; 1173 } 1174 #endif 1175 PetscLogFlops(a->nz); 1176 } 1177 if (flag & SOR_BACKWARD_SWEEP || flag & SOR_LOCAL_BACKWARD_SWEEP){ 1178 #if defined(PETSC_USE_FORTRAN_KERNEL_RELAXAIJ) 1179 fortranrelaxaijbackward_(&m,&omega,x,a->i,a->j,(int*)diag,a->a,(void*)b); 1180 #else 1181 for (i=m-1; i>=0; i--) { 1182 d = fshift + a->a[diag[i]]; 1183 n = a->i[i+1] - a->i[i]; 1184 idx = a->j + a->i[i]; 1185 v = a->a + a->i[i]; 1186 sum = b[i]; 1187 SPARSEDENSEMDOT(sum,xs,v,idx,n); 1188 x[i] = (1. - omega)*x[i] + (sum + mdiag[i]*x[i])*idiag[i]; 1189 } 1190 #endif 1191 PetscLogFlops(a->nz); 1192 } 1193 } 1194 ierr = VecRestoreArray(xx,&x);CHKERRQ(ierr); 1195 if (bb != xx) {ierr = VecRestoreArray(bb,(PetscScalar**)&b);CHKERRQ(ierr);} 1196 PetscFunctionReturn(0); 1197 } 1198 1199 #undef __FUNCT__ 1200 #define __FUNCT__ "MatGetInfo_SeqAIJ" 1201 int MatGetInfo_SeqAIJ(Mat A,MatInfoType flag,MatInfo *info) 1202 { 1203 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1204 1205 PetscFunctionBegin; 1206 info->rows_global = (double)A->m; 1207 info->columns_global = (double)A->n; 1208 info->rows_local = (double)A->m; 1209 info->columns_local = (double)A->n; 1210 info->block_size = 1.0; 1211 info->nz_allocated = (double)a->maxnz; 1212 info->nz_used = (double)a->nz; 1213 info->nz_unneeded = (double)(a->maxnz - a->nz); 1214 info->assemblies = (double)A->num_ass; 1215 info->mallocs = (double)a->reallocs; 1216 info->memory = A->mem; 1217 if (A->factor) { 1218 info->fill_ratio_given = A->info.fill_ratio_given; 1219 info->fill_ratio_needed = A->info.fill_ratio_needed; 1220 info->factor_mallocs = A->info.factor_mallocs; 1221 } else { 1222 info->fill_ratio_given = 0; 1223 info->fill_ratio_needed = 0; 1224 info->factor_mallocs = 0; 1225 } 1226 PetscFunctionReturn(0); 1227 } 1228 1229 #undef __FUNCT__ 1230 #define __FUNCT__ "MatZeroRows_SeqAIJ" 1231 int MatZeroRows_SeqAIJ(Mat A,IS is,const PetscScalar *diag) 1232 { 1233 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1234 int i,ierr,N,*rows,m = A->m - 1; 1235 1236 PetscFunctionBegin; 1237 ierr = ISGetLocalSize(is,&N);CHKERRQ(ierr); 1238 ierr = ISGetIndices(is,&rows);CHKERRQ(ierr); 1239 if (a->keepzeroedrows) { 1240 for (i=0; i<N; i++) { 1241 if (rows[i] < 0 || rows[i] > m) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"row %d out of range", rows[i]); 1242 ierr = PetscMemzero(&a->a[a->i[rows[i]]],a->ilen[rows[i]]*sizeof(PetscScalar));CHKERRQ(ierr); 1243 } 1244 if (diag) { 1245 ierr = MatMissingDiagonal_SeqAIJ(A);CHKERRQ(ierr); 1246 ierr = MatMarkDiagonal_SeqAIJ(A);CHKERRQ(ierr); 1247 for (i=0; i<N; i++) { 1248 a->a[a->diag[rows[i]]] = *diag; 1249 } 1250 } 1251 } else { 1252 if (diag) { 1253 for (i=0; i<N; i++) { 1254 if (rows[i] < 0 || rows[i] > m) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"row %d out of range", rows[i]); 1255 if (a->ilen[rows[i]] > 0) { 1256 a->ilen[rows[i]] = 1; 1257 a->a[a->i[rows[i]]] = *diag; 1258 a->j[a->i[rows[i]]] = rows[i]; 1259 } else { /* in case row was completely empty */ 1260 ierr = MatSetValues_SeqAIJ(A,1,&rows[i],1,&rows[i],diag,INSERT_VALUES);CHKERRQ(ierr); 1261 } 1262 } 1263 } else { 1264 for (i=0; i<N; i++) { 1265 if (rows[i] < 0 || rows[i] > m) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"row %d out of range", rows[i]); 1266 a->ilen[rows[i]] = 0; 1267 } 1268 } 1269 } 1270 ierr = ISRestoreIndices(is,&rows);CHKERRQ(ierr); 1271 ierr = MatAssemblyEnd_SeqAIJ(A,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1272 PetscFunctionReturn(0); 1273 } 1274 1275 #undef __FUNCT__ 1276 #define __FUNCT__ "MatGetRow_SeqAIJ" 1277 int MatGetRow_SeqAIJ(Mat A,int row,int *nz,int **idx,PetscScalar **v) 1278 { 1279 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1280 int *itmp; 1281 1282 PetscFunctionBegin; 1283 if (row < 0 || row >= A->m) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"Row %d out of range",row); 1284 1285 *nz = a->i[row+1] - a->i[row]; 1286 if (v) *v = a->a + a->i[row]; 1287 if (idx) { 1288 itmp = a->j + a->i[row]; 1289 if (*nz) { 1290 *idx = itmp; 1291 } 1292 else *idx = 0; 1293 } 1294 PetscFunctionReturn(0); 1295 } 1296 1297 /* remove this function? */ 1298 #undef __FUNCT__ 1299 #define __FUNCT__ "MatRestoreRow_SeqAIJ" 1300 int MatRestoreRow_SeqAIJ(Mat A,int row,int *nz,int **idx,PetscScalar **v) 1301 { 1302 /* Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1303 int ierr; */ 1304 1305 PetscFunctionBegin; 1306 /* if (idx) {if (*idx && a->indexshift) {ierr = PetscFree(*idx);CHKERRQ(ierr);}} */ 1307 PetscFunctionReturn(0); 1308 } 1309 1310 #undef __FUNCT__ 1311 #define __FUNCT__ "MatNorm_SeqAIJ" 1312 int MatNorm_SeqAIJ(Mat A,NormType type,PetscReal *nrm) 1313 { 1314 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1315 PetscScalar *v = a->a; 1316 PetscReal sum = 0.0; 1317 int i,j,ierr; 1318 1319 PetscFunctionBegin; 1320 if (type == NORM_FROBENIUS) { 1321 for (i=0; i<a->nz; i++) { 1322 #if defined(PETSC_USE_COMPLEX) 1323 sum += PetscRealPart(PetscConj(*v)*(*v)); v++; 1324 #else 1325 sum += (*v)*(*v); v++; 1326 #endif 1327 } 1328 *nrm = sqrt(sum); 1329 } else if (type == NORM_1) { 1330 PetscReal *tmp; 1331 int *jj = a->j; 1332 ierr = PetscMalloc((A->n+1)*sizeof(PetscReal),&tmp);CHKERRQ(ierr); 1333 ierr = PetscMemzero(tmp,A->n*sizeof(PetscReal));CHKERRQ(ierr); 1334 *nrm = 0.0; 1335 for (j=0; j<a->nz; j++) { 1336 tmp[*jj++] += PetscAbsScalar(*v); v++; 1337 } 1338 for (j=0; j<A->n; j++) { 1339 if (tmp[j] > *nrm) *nrm = tmp[j]; 1340 } 1341 ierr = PetscFree(tmp);CHKERRQ(ierr); 1342 } else if (type == NORM_INFINITY) { 1343 *nrm = 0.0; 1344 for (j=0; j<A->m; j++) { 1345 v = a->a + a->i[j]; 1346 sum = 0.0; 1347 for (i=0; i<a->i[j+1]-a->i[j]; i++) { 1348 sum += PetscAbsScalar(*v); v++; 1349 } 1350 if (sum > *nrm) *nrm = sum; 1351 } 1352 } else { 1353 SETERRQ(PETSC_ERR_SUP,"No support for two norm"); 1354 } 1355 PetscFunctionReturn(0); 1356 } 1357 1358 #undef __FUNCT__ 1359 #define __FUNCT__ "MatTranspose_SeqAIJ" 1360 int MatTranspose_SeqAIJ(Mat A,Mat *B) 1361 { 1362 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1363 Mat C; 1364 int i,ierr,*aj = a->j,*ai = a->i,m = A->m,len,*col; 1365 PetscScalar *array = a->a; 1366 1367 PetscFunctionBegin; 1368 if (!B && m != A->n) SETERRQ(PETSC_ERR_ARG_SIZ,"Square matrix only for in-place"); 1369 ierr = PetscMalloc((1+A->n)*sizeof(int),&col);CHKERRQ(ierr); 1370 ierr = PetscMemzero(col,(1+A->n)*sizeof(int));CHKERRQ(ierr); 1371 1372 for (i=0; i<ai[m]; i++) col[aj[i]] += 1; 1373 ierr = MatCreate(A->comm,A->n,m,A->n,m,&C);CHKERRQ(ierr); 1374 ierr = MatSetType(C,A->type_name);CHKERRQ(ierr); 1375 ierr = MatSeqAIJSetPreallocation(C,0,col);CHKERRQ(ierr); 1376 ierr = PetscFree(col);CHKERRQ(ierr); 1377 for (i=0; i<m; i++) { 1378 len = ai[i+1]-ai[i]; 1379 ierr = MatSetValues(C,len,aj,1,&i,array,INSERT_VALUES);CHKERRQ(ierr); 1380 array += len; 1381 aj += len; 1382 } 1383 1384 ierr = MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1385 ierr = MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1386 1387 if (B) { 1388 *B = C; 1389 } else { 1390 ierr = MatHeaderCopy(A,C);CHKERRQ(ierr); 1391 } 1392 PetscFunctionReturn(0); 1393 } 1394 1395 EXTERN_C_BEGIN 1396 #undef __FUNCT__ 1397 #define __FUNCT__ "MatIsTranspose_SeqAIJ" 1398 int MatIsTranspose_SeqAIJ(Mat A,Mat B,PetscReal tol,PetscTruth *f) 1399 { 1400 Mat_SeqAIJ *aij = (Mat_SeqAIJ *) A->data,*bij = (Mat_SeqAIJ*) A->data; 1401 int *adx,*bdx,*aii,*bii,*aptr,*bptr; PetscScalar *va,*vb; 1402 int ma,na,mb,nb, i,ierr; 1403 1404 PetscFunctionBegin; 1405 bij = (Mat_SeqAIJ *) B->data; 1406 1407 ierr = MatGetSize(A,&ma,&na);CHKERRQ(ierr); 1408 ierr = MatGetSize(B,&mb,&nb);CHKERRQ(ierr); 1409 if (ma!=nb || na!=mb){ 1410 *f = PETSC_FALSE; 1411 PetscFunctionReturn(0); 1412 } 1413 aii = aij->i; bii = bij->i; 1414 adx = aij->j; bdx = bij->j; 1415 va = aij->a; vb = bij->a; 1416 ierr = PetscMalloc(ma*sizeof(int),&aptr);CHKERRQ(ierr); 1417 ierr = PetscMalloc(mb*sizeof(int),&bptr);CHKERRQ(ierr); 1418 for (i=0; i<ma; i++) aptr[i] = aii[i]; 1419 for (i=0; i<mb; i++) bptr[i] = bii[i]; 1420 1421 *f = PETSC_TRUE; 1422 for (i=0; i<ma; i++) { 1423 /*printf("row %d spans %d--%d; we start @ %d\n", 1424 i,idx[ii[i]],idx[ii[i+1]-1],idx[aptr[i]]);*/ 1425 while (aptr[i]<aii[i+1]) { 1426 int idc,idr; 1427 PetscScalar vc,vr; 1428 /* column/row index/value */ 1429 idc = adx[aptr[i]]; 1430 idr = bdx[bptr[idc]]; 1431 vc = va[aptr[i]]; 1432 vr = vb[bptr[idc]]; 1433 /*printf("comparing %d: (%d,%d)=%e to (%d,%d)=%e\n", 1434 aptr[i],i,idc,vc,idc,idr,vr);*/ 1435 if (i!=idr || PetscAbsScalar(vc-vr) > tol) { 1436 *f = PETSC_FALSE; 1437 goto done; 1438 } else { 1439 aptr[i]++; 1440 if (B || i!=idc) bptr[idc]++; 1441 } 1442 } 1443 } 1444 done: 1445 ierr = PetscFree(aptr);CHKERRQ(ierr); 1446 if (B) { 1447 ierr = PetscFree(bptr);CHKERRQ(ierr); 1448 } 1449 PetscFunctionReturn(0); 1450 } 1451 EXTERN_C_END 1452 1453 #undef __FUNCT__ 1454 #define __FUNCT__ "MatIsSymmetric_SeqAIJ" 1455 int MatIsSymmetric_SeqAIJ(Mat A,PetscReal tol,PetscTruth *f) 1456 { 1457 int ierr; 1458 PetscFunctionBegin; 1459 ierr = MatIsTranspose_SeqAIJ(A,A,tol,f);CHKERRQ(ierr); 1460 PetscFunctionReturn(0); 1461 } 1462 1463 #undef __FUNCT__ 1464 #define __FUNCT__ "MatDiagonalScale_SeqAIJ" 1465 int MatDiagonalScale_SeqAIJ(Mat A,Vec ll,Vec rr) 1466 { 1467 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1468 PetscScalar *l,*r,x,*v; 1469 int ierr,i,j,m = A->m,n = A->n,M,nz = a->nz,*jj; 1470 1471 PetscFunctionBegin; 1472 if (ll) { 1473 /* The local size is used so that VecMPI can be passed to this routine 1474 by MatDiagonalScale_MPIAIJ */ 1475 ierr = VecGetLocalSize(ll,&m);CHKERRQ(ierr); 1476 if (m != A->m) SETERRQ(PETSC_ERR_ARG_SIZ,"Left scaling vector wrong length"); 1477 ierr = VecGetArray(ll,&l);CHKERRQ(ierr); 1478 v = a->a; 1479 for (i=0; i<m; i++) { 1480 x = l[i]; 1481 M = a->i[i+1] - a->i[i]; 1482 for (j=0; j<M; j++) { (*v++) *= x;} 1483 } 1484 ierr = VecRestoreArray(ll,&l);CHKERRQ(ierr); 1485 PetscLogFlops(nz); 1486 } 1487 if (rr) { 1488 ierr = VecGetLocalSize(rr,&n);CHKERRQ(ierr); 1489 if (n != A->n) SETERRQ(PETSC_ERR_ARG_SIZ,"Right scaling vector wrong length"); 1490 ierr = VecGetArray(rr,&r);CHKERRQ(ierr); 1491 v = a->a; jj = a->j; 1492 for (i=0; i<nz; i++) { 1493 (*v++) *= r[*jj++]; 1494 } 1495 ierr = VecRestoreArray(rr,&r);CHKERRQ(ierr); 1496 PetscLogFlops(nz); 1497 } 1498 PetscFunctionReturn(0); 1499 } 1500 1501 #undef __FUNCT__ 1502 #define __FUNCT__ "MatGetSubMatrix_SeqAIJ" 1503 int MatGetSubMatrix_SeqAIJ(Mat A,IS isrow,IS iscol,int csize,MatReuse scall,Mat *B) 1504 { 1505 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data,*c; 1506 int *smap,i,k,kstart,kend,ierr,oldcols = A->n,*lens; 1507 int row,mat_i,*mat_j,tcol,first,step,*mat_ilen,sum,lensi; 1508 int *irow,*icol,nrows,ncols; 1509 int *starts,*j_new,*i_new,*aj = a->j,*ai = a->i,ii,*ailen = a->ilen; 1510 PetscScalar *a_new,*mat_a; 1511 Mat C; 1512 PetscTruth stride; 1513 1514 PetscFunctionBegin; 1515 ierr = ISSorted(isrow,(PetscTruth*)&i);CHKERRQ(ierr); 1516 if (!i) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"ISrow is not sorted"); 1517 ierr = ISSorted(iscol,(PetscTruth*)&i);CHKERRQ(ierr); 1518 if (!i) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"IScol is not sorted"); 1519 1520 ierr = ISGetIndices(isrow,&irow);CHKERRQ(ierr); 1521 ierr = ISGetLocalSize(isrow,&nrows);CHKERRQ(ierr); 1522 ierr = ISGetLocalSize(iscol,&ncols);CHKERRQ(ierr); 1523 1524 ierr = ISStrideGetInfo(iscol,&first,&step);CHKERRQ(ierr); 1525 ierr = ISStride(iscol,&stride);CHKERRQ(ierr); 1526 if (stride && step == 1) { 1527 /* special case of contiguous rows */ 1528 ierr = PetscMalloc((2*nrows+1)*sizeof(int),&lens);CHKERRQ(ierr); 1529 starts = lens + nrows; 1530 /* loop over new rows determining lens and starting points */ 1531 for (i=0; i<nrows; i++) { 1532 kstart = ai[irow[i]]; 1533 kend = kstart + ailen[irow[i]]; 1534 for (k=kstart; k<kend; k++) { 1535 if (aj[k] >= first) { 1536 starts[i] = k; 1537 break; 1538 } 1539 } 1540 sum = 0; 1541 while (k < kend) { 1542 if (aj[k++] >= first+ncols) break; 1543 sum++; 1544 } 1545 lens[i] = sum; 1546 } 1547 /* create submatrix */ 1548 if (scall == MAT_REUSE_MATRIX) { 1549 int n_cols,n_rows; 1550 ierr = MatGetSize(*B,&n_rows,&n_cols);CHKERRQ(ierr); 1551 if (n_rows != nrows || n_cols != ncols) SETERRQ(PETSC_ERR_ARG_SIZ,"Reused submatrix wrong size"); 1552 ierr = MatZeroEntries(*B);CHKERRQ(ierr); 1553 C = *B; 1554 } else { 1555 ierr = MatCreate(A->comm,nrows,ncols,PETSC_DETERMINE,PETSC_DETERMINE,&C);CHKERRQ(ierr); 1556 ierr = MatSetType(C,A->type_name);CHKERRQ(ierr); 1557 ierr = MatSeqAIJSetPreallocation(C,0,lens);CHKERRQ(ierr); 1558 } 1559 c = (Mat_SeqAIJ*)C->data; 1560 1561 /* loop over rows inserting into submatrix */ 1562 a_new = c->a; 1563 j_new = c->j; 1564 i_new = c->i; 1565 1566 for (i=0; i<nrows; i++) { 1567 ii = starts[i]; 1568 lensi = lens[i]; 1569 for (k=0; k<lensi; k++) { 1570 *j_new++ = aj[ii+k] - first; 1571 } 1572 ierr = PetscMemcpy(a_new,a->a + starts[i],lensi*sizeof(PetscScalar));CHKERRQ(ierr); 1573 a_new += lensi; 1574 i_new[i+1] = i_new[i] + lensi; 1575 c->ilen[i] = lensi; 1576 } 1577 ierr = PetscFree(lens);CHKERRQ(ierr); 1578 } else { 1579 ierr = ISGetIndices(iscol,&icol);CHKERRQ(ierr); 1580 ierr = PetscMalloc((1+oldcols)*sizeof(int),&smap);CHKERRQ(ierr); 1581 1582 ierr = PetscMalloc((1+nrows)*sizeof(int),&lens);CHKERRQ(ierr); 1583 ierr = PetscMemzero(smap,oldcols*sizeof(int));CHKERRQ(ierr); 1584 for (i=0; i<ncols; i++) smap[icol[i]] = i+1; 1585 /* determine lens of each row */ 1586 for (i=0; i<nrows; i++) { 1587 kstart = ai[irow[i]]; 1588 kend = kstart + a->ilen[irow[i]]; 1589 lens[i] = 0; 1590 for (k=kstart; k<kend; k++) { 1591 if (smap[aj[k]]) { 1592 lens[i]++; 1593 } 1594 } 1595 } 1596 /* Create and fill new matrix */ 1597 if (scall == MAT_REUSE_MATRIX) { 1598 PetscTruth equal; 1599 1600 c = (Mat_SeqAIJ *)((*B)->data); 1601 if ((*B)->m != nrows || (*B)->n != ncols) SETERRQ(PETSC_ERR_ARG_SIZ,"Cannot reuse matrix. wrong size"); 1602 ierr = PetscMemcmp(c->ilen,lens,(*B)->m*sizeof(int),&equal);CHKERRQ(ierr); 1603 if (!equal) { 1604 SETERRQ(PETSC_ERR_ARG_SIZ,"Cannot reuse matrix. wrong no of nonzeros"); 1605 } 1606 ierr = PetscMemzero(c->ilen,(*B)->m*sizeof(int));CHKERRQ(ierr); 1607 C = *B; 1608 } else { 1609 ierr = MatCreate(A->comm,nrows,ncols,PETSC_DETERMINE,PETSC_DETERMINE,&C);CHKERRQ(ierr); 1610 ierr = MatSetType(C,A->type_name);CHKERRQ(ierr); 1611 ierr = MatSeqAIJSetPreallocation(C,0,lens);CHKERRQ(ierr); 1612 } 1613 c = (Mat_SeqAIJ *)(C->data); 1614 for (i=0; i<nrows; i++) { 1615 row = irow[i]; 1616 kstart = ai[row]; 1617 kend = kstart + a->ilen[row]; 1618 mat_i = c->i[i]; 1619 mat_j = c->j + mat_i; 1620 mat_a = c->a + mat_i; 1621 mat_ilen = c->ilen + i; 1622 for (k=kstart; k<kend; k++) { 1623 if ((tcol=smap[a->j[k]])) { 1624 *mat_j++ = tcol - 1; 1625 *mat_a++ = a->a[k]; 1626 (*mat_ilen)++; 1627 1628 } 1629 } 1630 } 1631 /* Free work space */ 1632 ierr = ISRestoreIndices(iscol,&icol);CHKERRQ(ierr); 1633 ierr = PetscFree(smap);CHKERRQ(ierr); 1634 ierr = PetscFree(lens);CHKERRQ(ierr); 1635 } 1636 ierr = MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1637 ierr = MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1638 1639 ierr = ISRestoreIndices(isrow,&irow);CHKERRQ(ierr); 1640 *B = C; 1641 PetscFunctionReturn(0); 1642 } 1643 1644 /* 1645 */ 1646 #undef __FUNCT__ 1647 #define __FUNCT__ "MatILUFactor_SeqAIJ" 1648 int MatILUFactor_SeqAIJ(Mat inA,IS row,IS col,MatFactorInfo *info) 1649 { 1650 Mat_SeqAIJ *a = (Mat_SeqAIJ*)inA->data; 1651 int ierr; 1652 Mat outA; 1653 PetscTruth row_identity,col_identity; 1654 1655 PetscFunctionBegin; 1656 if (info->levels != 0) SETERRQ(PETSC_ERR_SUP,"Only levels=0 supported for in-place ilu"); 1657 ierr = ISIdentity(row,&row_identity);CHKERRQ(ierr); 1658 ierr = ISIdentity(col,&col_identity);CHKERRQ(ierr); 1659 if (!row_identity || !col_identity) { 1660 SETERRQ(1,"Row and column permutations must be identity for in-place ILU"); 1661 } 1662 1663 outA = inA; 1664 inA->factor = FACTOR_LU; 1665 a->row = row; 1666 a->col = col; 1667 ierr = PetscObjectReference((PetscObject)row);CHKERRQ(ierr); 1668 ierr = PetscObjectReference((PetscObject)col);CHKERRQ(ierr); 1669 1670 /* Create the inverse permutation so that it can be used in MatLUFactorNumeric() */ 1671 if (a->icol) {ierr = ISDestroy(a->icol);CHKERRQ(ierr);} /* need to remove old one */ 1672 ierr = ISInvertPermutation(col,PETSC_DECIDE,&a->icol);CHKERRQ(ierr); 1673 PetscLogObjectParent(inA,a->icol); 1674 1675 if (!a->solve_work) { /* this matrix may have been factored before */ 1676 ierr = PetscMalloc((inA->m+1)*sizeof(PetscScalar),&a->solve_work);CHKERRQ(ierr); 1677 } 1678 1679 if (!a->diag) { 1680 ierr = MatMarkDiagonal_SeqAIJ(inA);CHKERRQ(ierr); 1681 } 1682 ierr = MatLUFactorNumeric_SeqAIJ(inA,&outA);CHKERRQ(ierr); 1683 PetscFunctionReturn(0); 1684 } 1685 1686 #include "petscblaslapack.h" 1687 #undef __FUNCT__ 1688 #define __FUNCT__ "MatScale_SeqAIJ" 1689 int MatScale_SeqAIJ(const PetscScalar *alpha,Mat inA) 1690 { 1691 Mat_SeqAIJ *a = (Mat_SeqAIJ*)inA->data; 1692 int one = 1; 1693 1694 PetscFunctionBegin; 1695 BLscal_(&a->nz,(PetscScalar*)alpha,a->a,&one); 1696 PetscLogFlops(a->nz); 1697 PetscFunctionReturn(0); 1698 } 1699 1700 #undef __FUNCT__ 1701 #define __FUNCT__ "MatGetSubMatrices_SeqAIJ" 1702 int MatGetSubMatrices_SeqAIJ(Mat A,int n,const IS irow[],const IS icol[],MatReuse scall,Mat *B[]) 1703 { 1704 int ierr,i; 1705 1706 PetscFunctionBegin; 1707 if (scall == MAT_INITIAL_MATRIX) { 1708 ierr = PetscMalloc((n+1)*sizeof(Mat),B);CHKERRQ(ierr); 1709 } 1710 1711 for (i=0; i<n; i++) { 1712 ierr = MatGetSubMatrix_SeqAIJ(A,irow[i],icol[i],PETSC_DECIDE,scall,&(*B)[i]);CHKERRQ(ierr); 1713 } 1714 PetscFunctionReturn(0); 1715 } 1716 1717 #undef __FUNCT__ 1718 #define __FUNCT__ "MatGetBlockSize_SeqAIJ" 1719 int MatGetBlockSize_SeqAIJ(Mat A,int *bs) 1720 { 1721 PetscFunctionBegin; 1722 *bs = 1; 1723 PetscFunctionReturn(0); 1724 } 1725 1726 #undef __FUNCT__ 1727 #define __FUNCT__ "MatIncreaseOverlap_SeqAIJ" 1728 int MatIncreaseOverlap_SeqAIJ(Mat A,int is_max,IS is[],int ov) 1729 { 1730 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1731 int row,i,j,k,l,m,n,*idx,ierr,*nidx,isz,val; 1732 int start,end,*ai,*aj; 1733 PetscBT table; 1734 1735 PetscFunctionBegin; 1736 m = A->m; 1737 ai = a->i; 1738 aj = a->j; 1739 1740 if (ov < 0) SETERRQ(PETSC_ERR_ARG_OUTOFRANGE,"illegal negative overlap value used"); 1741 1742 ierr = PetscMalloc((m+1)*sizeof(int),&nidx);CHKERRQ(ierr); 1743 ierr = PetscBTCreate(m,table);CHKERRQ(ierr); 1744 1745 for (i=0; i<is_max; i++) { 1746 /* Initialize the two local arrays */ 1747 isz = 0; 1748 ierr = PetscBTMemzero(m,table);CHKERRQ(ierr); 1749 1750 /* Extract the indices, assume there can be duplicate entries */ 1751 ierr = ISGetIndices(is[i],&idx);CHKERRQ(ierr); 1752 ierr = ISGetLocalSize(is[i],&n);CHKERRQ(ierr); 1753 1754 /* Enter these into the temp arrays. I.e., mark table[row], enter row into new index */ 1755 for (j=0; j<n ; ++j){ 1756 if(!PetscBTLookupSet(table,idx[j])) { nidx[isz++] = idx[j];} 1757 } 1758 ierr = ISRestoreIndices(is[i],&idx);CHKERRQ(ierr); 1759 ierr = ISDestroy(is[i]);CHKERRQ(ierr); 1760 1761 k = 0; 1762 for (j=0; j<ov; j++){ /* for each overlap */ 1763 n = isz; 1764 for (; k<n ; k++){ /* do only those rows in nidx[k], which are not done yet */ 1765 row = nidx[k]; 1766 start = ai[row]; 1767 end = ai[row+1]; 1768 for (l = start; l<end ; l++){ 1769 val = aj[l] ; 1770 if (!PetscBTLookupSet(table,val)) {nidx[isz++] = val;} 1771 } 1772 } 1773 } 1774 ierr = ISCreateGeneral(PETSC_COMM_SELF,isz,nidx,(is+i));CHKERRQ(ierr); 1775 } 1776 ierr = PetscBTDestroy(table);CHKERRQ(ierr); 1777 ierr = PetscFree(nidx);CHKERRQ(ierr); 1778 PetscFunctionReturn(0); 1779 } 1780 1781 /* -------------------------------------------------------------- */ 1782 #undef __FUNCT__ 1783 #define __FUNCT__ "MatPermute_SeqAIJ" 1784 int MatPermute_SeqAIJ(Mat A,IS rowp,IS colp,Mat *B) 1785 { 1786 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1787 int i,ierr,nz,m = A->m,n = A->n,*col; 1788 int *row,*cnew,j,*lens; 1789 IS icolp,irowp; 1790 const int *cwork; 1791 const PetscScalar *vwork; 1792 1793 PetscFunctionBegin; 1794 ierr = ISInvertPermutation(rowp,PETSC_DECIDE,&irowp);CHKERRQ(ierr); 1795 ierr = ISGetIndices(irowp,&row);CHKERRQ(ierr); 1796 ierr = ISInvertPermutation(colp,PETSC_DECIDE,&icolp);CHKERRQ(ierr); 1797 ierr = ISGetIndices(icolp,&col);CHKERRQ(ierr); 1798 1799 /* determine lengths of permuted rows */ 1800 ierr = PetscMalloc((m+1)*sizeof(int),&lens);CHKERRQ(ierr); 1801 for (i=0; i<m; i++) { 1802 lens[row[i]] = a->i[i+1] - a->i[i]; 1803 } 1804 ierr = MatCreate(A->comm,m,n,m,n,B);CHKERRQ(ierr); 1805 ierr = MatSetType(*B,A->type_name);CHKERRQ(ierr); 1806 ierr = MatSeqAIJSetPreallocation(*B,0,lens);CHKERRQ(ierr); 1807 ierr = PetscFree(lens);CHKERRQ(ierr); 1808 1809 ierr = PetscMalloc(n*sizeof(int),&cnew);CHKERRQ(ierr); 1810 for (i=0; i<m; i++) { 1811 ierr = MatGetRow(A,i,&nz,&cwork,&vwork);CHKERRQ(ierr); 1812 for (j=0; j<nz; j++) { cnew[j] = col[cwork[j]];} 1813 ierr = MatSetValues(*B,1,&row[i],nz,cnew,vwork,INSERT_VALUES);CHKERRQ(ierr); 1814 ierr = MatRestoreRow(A,i,&nz,&cwork,&vwork);CHKERRQ(ierr); 1815 } 1816 ierr = PetscFree(cnew);CHKERRQ(ierr); 1817 ierr = MatAssemblyBegin(*B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1818 ierr = MatAssemblyEnd(*B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 1819 ierr = ISRestoreIndices(irowp,&row);CHKERRQ(ierr); 1820 ierr = ISRestoreIndices(icolp,&col);CHKERRQ(ierr); 1821 ierr = ISDestroy(irowp);CHKERRQ(ierr); 1822 ierr = ISDestroy(icolp);CHKERRQ(ierr); 1823 PetscFunctionReturn(0); 1824 } 1825 1826 #undef __FUNCT__ 1827 #define __FUNCT__ "MatPrintHelp_SeqAIJ" 1828 int MatPrintHelp_SeqAIJ(Mat A) 1829 { 1830 static PetscTruth called = PETSC_FALSE; 1831 MPI_Comm comm = A->comm; 1832 int ierr; 1833 1834 PetscFunctionBegin; 1835 if (called) {PetscFunctionReturn(0);} else called = PETSC_TRUE; 1836 ierr = (*PetscHelpPrintf)(comm," Options for MATSEQAIJ and MATMPIAIJ matrix formats (the defaults):\n");CHKERRQ(ierr); 1837 ierr = (*PetscHelpPrintf)(comm," -mat_lu_pivotthreshold <threshold>: Set pivoting threshold\n");CHKERRQ(ierr); 1838 ierr = (*PetscHelpPrintf)(comm," -mat_aij_oneindex: internal indices begin at 1 instead of the default 0.\n");CHKERRQ(ierr); 1839 ierr = (*PetscHelpPrintf)(comm," -mat_aij_no_inode: Do not use inodes\n");CHKERRQ(ierr); 1840 ierr = (*PetscHelpPrintf)(comm," -mat_aij_inode_limit <limit>: Set inode limit (max limit=5)\n");CHKERRQ(ierr); 1841 PetscFunctionReturn(0); 1842 } 1843 1844 #undef __FUNCT__ 1845 #define __FUNCT__ "MatCopy_SeqAIJ" 1846 int MatCopy_SeqAIJ(Mat A,Mat B,MatStructure str) 1847 { 1848 int ierr; 1849 1850 PetscFunctionBegin; 1851 /* If the two matrices have the same copy implementation, use fast copy. */ 1852 if (str == SAME_NONZERO_PATTERN && (A->ops->copy == B->ops->copy)) { 1853 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1854 Mat_SeqAIJ *b = (Mat_SeqAIJ*)B->data; 1855 1856 if (a->i[A->m] != b->i[B->m]) { 1857 SETERRQ(1,"Number of nonzeros in two matrices are different"); 1858 } 1859 ierr = PetscMemcpy(b->a,a->a,(a->i[A->m])*sizeof(PetscScalar));CHKERRQ(ierr); 1860 } else { 1861 ierr = MatCopy_Basic(A,B,str);CHKERRQ(ierr); 1862 } 1863 PetscFunctionReturn(0); 1864 } 1865 1866 #undef __FUNCT__ 1867 #define __FUNCT__ "MatSetUpPreallocation_SeqAIJ" 1868 int MatSetUpPreallocation_SeqAIJ(Mat A) 1869 { 1870 int ierr; 1871 1872 PetscFunctionBegin; 1873 ierr = MatSeqAIJSetPreallocation(A,PETSC_DEFAULT,0);CHKERRQ(ierr); 1874 PetscFunctionReturn(0); 1875 } 1876 1877 #undef __FUNCT__ 1878 #define __FUNCT__ "MatGetArray_SeqAIJ" 1879 int MatGetArray_SeqAIJ(Mat A,PetscScalar *array[]) 1880 { 1881 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 1882 PetscFunctionBegin; 1883 *array = a->a; 1884 PetscFunctionReturn(0); 1885 } 1886 1887 #undef __FUNCT__ 1888 #define __FUNCT__ "MatRestoreArray_SeqAIJ" 1889 int MatRestoreArray_SeqAIJ(Mat A,PetscScalar *array[]) 1890 { 1891 PetscFunctionBegin; 1892 PetscFunctionReturn(0); 1893 } 1894 1895 #undef __FUNCT__ 1896 #define __FUNCT__ "MatFDColoringApply_SeqAIJ" 1897 int MatFDColoringApply_SeqAIJ(Mat J,MatFDColoring coloring,Vec x1,MatStructure *flag,void *sctx) 1898 { 1899 int (*f)(void *,Vec,Vec,void*) = (int (*)(void *,Vec,Vec,void *))coloring->f; 1900 int k,ierr,N,start,end,l,row,col,srow,**vscaleforrow,m1,m2; 1901 PetscScalar dx,mone = -1.0,*y,*xx,*w3_array; 1902 PetscScalar *vscale_array; 1903 PetscReal epsilon = coloring->error_rel,umin = coloring->umin; 1904 Vec w1,w2,w3; 1905 void *fctx = coloring->fctx; 1906 PetscTruth flg; 1907 1908 PetscFunctionBegin; 1909 if (!coloring->w1) { 1910 ierr = VecDuplicate(x1,&coloring->w1);CHKERRQ(ierr); 1911 PetscLogObjectParent(coloring,coloring->w1); 1912 ierr = VecDuplicate(x1,&coloring->w2);CHKERRQ(ierr); 1913 PetscLogObjectParent(coloring,coloring->w2); 1914 ierr = VecDuplicate(x1,&coloring->w3);CHKERRQ(ierr); 1915 PetscLogObjectParent(coloring,coloring->w3); 1916 } 1917 w1 = coloring->w1; w2 = coloring->w2; w3 = coloring->w3; 1918 1919 ierr = MatSetUnfactored(J);CHKERRQ(ierr); 1920 ierr = PetscOptionsHasName(coloring->prefix,"-mat_fd_coloring_dont_rezero",&flg);CHKERRQ(ierr); 1921 if (flg) { 1922 PetscLogInfo(coloring,"MatFDColoringApply_SeqAIJ: Not calling MatZeroEntries()\n"); 1923 } else { 1924 ierr = MatZeroEntries(J);CHKERRQ(ierr); 1925 } 1926 1927 ierr = VecGetOwnershipRange(x1,&start,&end);CHKERRQ(ierr); 1928 ierr = VecGetSize(x1,&N);CHKERRQ(ierr); 1929 1930 /* 1931 This is a horrible, horrible, hack. See DMMGComputeJacobian_Multigrid() it inproperly sets 1932 coloring->F for the coarser grids from the finest 1933 */ 1934 if (coloring->F) { 1935 ierr = VecGetLocalSize(coloring->F,&m1);CHKERRQ(ierr); 1936 ierr = VecGetLocalSize(w1,&m2);CHKERRQ(ierr); 1937 if (m1 != m2) { 1938 coloring->F = 0; 1939 } 1940 } 1941 1942 if (coloring->F) { 1943 w1 = coloring->F; 1944 coloring->F = 0; 1945 } else { 1946 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 1947 ierr = (*f)(sctx,x1,w1,fctx);CHKERRQ(ierr); 1948 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 1949 } 1950 1951 /* 1952 Compute all the scale factors and share with other processors 1953 */ 1954 ierr = VecGetArray(x1,&xx);CHKERRQ(ierr);xx = xx - start; 1955 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr);vscale_array = vscale_array - start; 1956 for (k=0; k<coloring->ncolors; k++) { 1957 /* 1958 Loop over each column associated with color adding the 1959 perturbation to the vector w3. 1960 */ 1961 for (l=0; l<coloring->ncolumns[k]; l++) { 1962 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 1963 dx = xx[col]; 1964 if (dx == 0.0) dx = 1.0; 1965 #if !defined(PETSC_USE_COMPLEX) 1966 if (dx < umin && dx >= 0.0) dx = umin; 1967 else if (dx < 0.0 && dx > -umin) dx = -umin; 1968 #else 1969 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 1970 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 1971 #endif 1972 dx *= epsilon; 1973 vscale_array[col] = 1.0/dx; 1974 } 1975 } 1976 vscale_array = vscale_array + start;ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 1977 ierr = VecGhostUpdateBegin(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1978 ierr = VecGhostUpdateEnd(coloring->vscale,INSERT_VALUES,SCATTER_FORWARD);CHKERRQ(ierr); 1979 1980 /* ierr = VecView(coloring->vscale,PETSC_VIEWER_STDOUT_WORLD); 1981 ierr = VecView(x1,PETSC_VIEWER_STDOUT_WORLD);*/ 1982 1983 if (coloring->vscaleforrow) vscaleforrow = coloring->vscaleforrow; 1984 else vscaleforrow = coloring->columnsforrow; 1985 1986 ierr = VecGetArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 1987 /* 1988 Loop over each color 1989 */ 1990 for (k=0; k<coloring->ncolors; k++) { 1991 coloring->currentcolor = k; 1992 ierr = VecCopy(x1,w3);CHKERRQ(ierr); 1993 ierr = VecGetArray(w3,&w3_array);CHKERRQ(ierr);w3_array = w3_array - start; 1994 /* 1995 Loop over each column associated with color adding the 1996 perturbation to the vector w3. 1997 */ 1998 for (l=0; l<coloring->ncolumns[k]; l++) { 1999 col = coloring->columns[k][l]; /* column of the matrix we are probing for */ 2000 dx = xx[col]; 2001 if (dx == 0.0) dx = 1.0; 2002 #if !defined(PETSC_USE_COMPLEX) 2003 if (dx < umin && dx >= 0.0) dx = umin; 2004 else if (dx < 0.0 && dx > -umin) dx = -umin; 2005 #else 2006 if (PetscAbsScalar(dx) < umin && PetscRealPart(dx) >= 0.0) dx = umin; 2007 else if (PetscRealPart(dx) < 0.0 && PetscAbsScalar(dx) < umin) dx = -umin; 2008 #endif 2009 dx *= epsilon; 2010 if (!PetscAbsScalar(dx)) SETERRQ(1,"Computed 0 differencing parameter"); 2011 w3_array[col] += dx; 2012 } 2013 w3_array = w3_array + start; ierr = VecRestoreArray(w3,&w3_array);CHKERRQ(ierr); 2014 2015 /* 2016 Evaluate function at x1 + dx (here dx is a vector of perturbations) 2017 */ 2018 2019 ierr = PetscLogEventBegin(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 2020 ierr = (*f)(sctx,w3,w2,fctx);CHKERRQ(ierr); 2021 ierr = PetscLogEventEnd(MAT_FDColoringFunction,0,0,0,0);CHKERRQ(ierr); 2022 ierr = VecAXPY(&mone,w1,w2);CHKERRQ(ierr); 2023 2024 /* 2025 Loop over rows of vector, putting results into Jacobian matrix 2026 */ 2027 ierr = VecGetArray(w2,&y);CHKERRQ(ierr); 2028 for (l=0; l<coloring->nrows[k]; l++) { 2029 row = coloring->rows[k][l]; 2030 col = coloring->columnsforrow[k][l]; 2031 y[row] *= vscale_array[vscaleforrow[k][l]]; 2032 srow = row + start; 2033 ierr = MatSetValues_SeqAIJ(J,1,&srow,1,&col,y+row,INSERT_VALUES);CHKERRQ(ierr); 2034 } 2035 ierr = VecRestoreArray(w2,&y);CHKERRQ(ierr); 2036 } 2037 coloring->currentcolor = k; 2038 ierr = VecRestoreArray(coloring->vscale,&vscale_array);CHKERRQ(ierr); 2039 xx = xx + start; ierr = VecRestoreArray(x1,&xx);CHKERRQ(ierr); 2040 ierr = MatAssemblyBegin(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2041 ierr = MatAssemblyEnd(J,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2042 PetscFunctionReturn(0); 2043 } 2044 2045 #include "petscblaslapack.h" 2046 #undef __FUNCT__ 2047 #define __FUNCT__ "MatAXPY_SeqAIJ" 2048 int MatAXPY_SeqAIJ(const PetscScalar a[],Mat X,Mat Y,MatStructure str) 2049 { 2050 int ierr,one=1,i; 2051 Mat_SeqAIJ *x = (Mat_SeqAIJ *)X->data,*y = (Mat_SeqAIJ *)Y->data; 2052 2053 PetscFunctionBegin; 2054 if (str == SAME_NONZERO_PATTERN) { 2055 BLaxpy_(&x->nz,(PetscScalar*)a,x->a,&one,y->a,&one); 2056 } else if (str == SUBSET_NONZERO_PATTERN) { /* nonzeros of X is a subset of Y's */ 2057 if (y->xtoy && y->XtoY != X) { 2058 ierr = PetscFree(y->xtoy);CHKERRQ(ierr); 2059 ierr = MatDestroy(y->XtoY);CHKERRQ(ierr); 2060 } 2061 if (!y->xtoy) { /* get xtoy */ 2062 ierr = MatAXPYGetxtoy_Private(X->m,x->i,x->j,PETSC_NULL, y->i,y->j,PETSC_NULL, &y->xtoy);CHKERRQ(ierr); 2063 y->XtoY = X; 2064 } 2065 for (i=0; i<x->nz; i++) y->a[y->xtoy[i]] += (*a)*(x->a[i]); 2066 PetscLogInfo(0,"MatAXPY_SeqAIJ: ratio of nnz(X)/nnz(Y): %d/%d = %g\n",x->nz,y->nz,(PetscReal)(x->nz)/y->nz); 2067 } else { 2068 ierr = MatAXPY_Basic(a,X,Y,str);CHKERRQ(ierr); 2069 } 2070 PetscFunctionReturn(0); 2071 } 2072 2073 /* -------------------------------------------------------------------*/ 2074 static struct _MatOps MatOps_Values = {MatSetValues_SeqAIJ, 2075 MatGetRow_SeqAIJ, 2076 MatRestoreRow_SeqAIJ, 2077 MatMult_SeqAIJ, 2078 /* 4*/ MatMultAdd_SeqAIJ, 2079 MatMultTranspose_SeqAIJ, 2080 MatMultTransposeAdd_SeqAIJ, 2081 MatSolve_SeqAIJ, 2082 MatSolveAdd_SeqAIJ, 2083 MatSolveTranspose_SeqAIJ, 2084 /*10*/ MatSolveTransposeAdd_SeqAIJ, 2085 MatLUFactor_SeqAIJ, 2086 0, 2087 MatRelax_SeqAIJ, 2088 MatTranspose_SeqAIJ, 2089 /*15*/ MatGetInfo_SeqAIJ, 2090 MatEqual_SeqAIJ, 2091 MatGetDiagonal_SeqAIJ, 2092 MatDiagonalScale_SeqAIJ, 2093 MatNorm_SeqAIJ, 2094 /*20*/ 0, 2095 MatAssemblyEnd_SeqAIJ, 2096 MatCompress_SeqAIJ, 2097 MatSetOption_SeqAIJ, 2098 MatZeroEntries_SeqAIJ, 2099 /*25*/ MatZeroRows_SeqAIJ, 2100 MatLUFactorSymbolic_SeqAIJ, 2101 MatLUFactorNumeric_SeqAIJ, 2102 MatCholeskyFactorSymbolic_SeqAIJ, 2103 MatCholeskyFactorNumeric_SeqAIJ, 2104 /*30*/ MatSetUpPreallocation_SeqAIJ, 2105 MatILUFactorSymbolic_SeqAIJ, 2106 MatICCFactorSymbolic_SeqAIJ, 2107 MatGetArray_SeqAIJ, 2108 MatRestoreArray_SeqAIJ, 2109 /*35*/ MatDuplicate_SeqAIJ, 2110 0, 2111 0, 2112 MatILUFactor_SeqAIJ, 2113 0, 2114 /*40*/ MatAXPY_SeqAIJ, 2115 MatGetSubMatrices_SeqAIJ, 2116 MatIncreaseOverlap_SeqAIJ, 2117 MatGetValues_SeqAIJ, 2118 MatCopy_SeqAIJ, 2119 /*45*/ MatPrintHelp_SeqAIJ, 2120 MatScale_SeqAIJ, 2121 0, 2122 0, 2123 MatILUDTFactor_SeqAIJ, 2124 /*50*/ MatGetBlockSize_SeqAIJ, 2125 MatGetRowIJ_SeqAIJ, 2126 MatRestoreRowIJ_SeqAIJ, 2127 MatGetColumnIJ_SeqAIJ, 2128 MatRestoreColumnIJ_SeqAIJ, 2129 /*55*/ MatFDColoringCreate_SeqAIJ, 2130 0, 2131 0, 2132 MatPermute_SeqAIJ, 2133 0, 2134 /*60*/ 0, 2135 MatDestroy_SeqAIJ, 2136 MatView_SeqAIJ, 2137 MatGetPetscMaps_Petsc, 2138 0, 2139 /*65*/ 0, 2140 0, 2141 0, 2142 0, 2143 0, 2144 /*70*/ 0, 2145 0, 2146 MatSetColoring_SeqAIJ, 2147 MatSetValuesAdic_SeqAIJ, 2148 MatSetValuesAdifor_SeqAIJ, 2149 /*75*/ MatFDColoringApply_SeqAIJ, 2150 0, 2151 0, 2152 0, 2153 0, 2154 /*80*/ 0, 2155 0, 2156 0, 2157 0, 2158 /*85*/ MatLoad_SeqAIJ, 2159 MatIsSymmetric_SeqAIJ, 2160 }; 2161 2162 EXTERN_C_BEGIN 2163 #undef __FUNCT__ 2164 #define __FUNCT__ "MatSeqAIJSetColumnIndices_SeqAIJ" 2165 int MatSeqAIJSetColumnIndices_SeqAIJ(Mat mat,int *indices) 2166 { 2167 Mat_SeqAIJ *aij = (Mat_SeqAIJ *)mat->data; 2168 int i,nz,n; 2169 2170 PetscFunctionBegin; 2171 2172 nz = aij->maxnz; 2173 n = mat->n; 2174 for (i=0; i<nz; i++) { 2175 aij->j[i] = indices[i]; 2176 } 2177 aij->nz = nz; 2178 for (i=0; i<n; i++) { 2179 aij->ilen[i] = aij->imax[i]; 2180 } 2181 2182 PetscFunctionReturn(0); 2183 } 2184 EXTERN_C_END 2185 2186 #undef __FUNCT__ 2187 #define __FUNCT__ "MatSeqAIJSetColumnIndices" 2188 /*@ 2189 MatSeqAIJSetColumnIndices - Set the column indices for all the rows 2190 in the matrix. 2191 2192 Input Parameters: 2193 + mat - the SeqAIJ matrix 2194 - indices - the column indices 2195 2196 Level: advanced 2197 2198 Notes: 2199 This can be called if you have precomputed the nonzero structure of the 2200 matrix and want to provide it to the matrix object to improve the performance 2201 of the MatSetValues() operation. 2202 2203 You MUST have set the correct numbers of nonzeros per row in the call to 2204 MatCreateSeqAIJ(). 2205 2206 MUST be called before any calls to MatSetValues(); 2207 2208 The indices should start with zero, not one. 2209 2210 @*/ 2211 int MatSeqAIJSetColumnIndices(Mat mat,int *indices) 2212 { 2213 int ierr,(*f)(Mat,int *); 2214 2215 PetscFunctionBegin; 2216 PetscValidHeaderSpecific(mat,MAT_COOKIE,1); 2217 PetscValidPointer(indices,2); 2218 ierr = PetscObjectQueryFunction((PetscObject)mat,"MatSeqAIJSetColumnIndices_C",(void (**)(void))&f);CHKERRQ(ierr); 2219 if (f) { 2220 ierr = (*f)(mat,indices);CHKERRQ(ierr); 2221 } else { 2222 SETERRQ(1,"Wrong type of matrix to set column indices"); 2223 } 2224 PetscFunctionReturn(0); 2225 } 2226 2227 /* ----------------------------------------------------------------------------------------*/ 2228 2229 EXTERN_C_BEGIN 2230 #undef __FUNCT__ 2231 #define __FUNCT__ "MatStoreValues_SeqAIJ" 2232 int MatStoreValues_SeqAIJ(Mat mat) 2233 { 2234 Mat_SeqAIJ *aij = (Mat_SeqAIJ *)mat->data; 2235 size_t nz = aij->i[mat->m],ierr; 2236 2237 PetscFunctionBegin; 2238 if (aij->nonew != 1) { 2239 SETERRQ(1,"Must call MatSetOption(A,MAT_NO_NEW_NONZERO_LOCATIONS);first"); 2240 } 2241 2242 /* allocate space for values if not already there */ 2243 if (!aij->saved_values) { 2244 ierr = PetscMalloc((nz+1)*sizeof(PetscScalar),&aij->saved_values);CHKERRQ(ierr); 2245 } 2246 2247 /* copy values over */ 2248 ierr = PetscMemcpy(aij->saved_values,aij->a,nz*sizeof(PetscScalar));CHKERRQ(ierr); 2249 PetscFunctionReturn(0); 2250 } 2251 EXTERN_C_END 2252 2253 #undef __FUNCT__ 2254 #define __FUNCT__ "MatStoreValues" 2255 /*@ 2256 MatStoreValues - Stashes a copy of the matrix values; this allows, for 2257 example, reuse of the linear part of a Jacobian, while recomputing the 2258 nonlinear portion. 2259 2260 Collect on Mat 2261 2262 Input Parameters: 2263 . mat - the matrix (currently on AIJ matrices support this option) 2264 2265 Level: advanced 2266 2267 Common Usage, with SNESSolve(): 2268 $ Create Jacobian matrix 2269 $ Set linear terms into matrix 2270 $ Apply boundary conditions to matrix, at this time matrix must have 2271 $ final nonzero structure (i.e. setting the nonlinear terms and applying 2272 $ boundary conditions again will not change the nonzero structure 2273 $ ierr = MatSetOption(mat,MAT_NO_NEW_NONZERO_LOCATIONS); 2274 $ ierr = MatStoreValues(mat); 2275 $ Call SNESSetJacobian() with matrix 2276 $ In your Jacobian routine 2277 $ ierr = MatRetrieveValues(mat); 2278 $ Set nonlinear terms in matrix 2279 2280 Common Usage without SNESSolve(), i.e. when you handle nonlinear solve yourself: 2281 $ // build linear portion of Jacobian 2282 $ ierr = MatSetOption(mat,MAT_NO_NEW_NONZERO_LOCATIONS); 2283 $ ierr = MatStoreValues(mat); 2284 $ loop over nonlinear iterations 2285 $ ierr = MatRetrieveValues(mat); 2286 $ // call MatSetValues(mat,...) to set nonliner portion of Jacobian 2287 $ // call MatAssemblyBegin/End() on matrix 2288 $ Solve linear system with Jacobian 2289 $ endloop 2290 2291 Notes: 2292 Matrix must already be assemblied before calling this routine 2293 Must set the matrix option MatSetOption(mat,MAT_NO_NEW_NONZERO_LOCATIONS); before 2294 calling this routine. 2295 2296 .seealso: MatRetrieveValues() 2297 2298 @*/ 2299 int MatStoreValues(Mat mat) 2300 { 2301 int ierr,(*f)(Mat); 2302 2303 PetscFunctionBegin; 2304 PetscValidHeaderSpecific(mat,MAT_COOKIE,1); 2305 if (!mat->assembled) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix"); 2306 if (mat->factor) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix"); 2307 2308 ierr = PetscObjectQueryFunction((PetscObject)mat,"MatStoreValues_C",(void (**)(void))&f);CHKERRQ(ierr); 2309 if (f) { 2310 ierr = (*f)(mat);CHKERRQ(ierr); 2311 } else { 2312 SETERRQ(1,"Wrong type of matrix to store values"); 2313 } 2314 PetscFunctionReturn(0); 2315 } 2316 2317 EXTERN_C_BEGIN 2318 #undef __FUNCT__ 2319 #define __FUNCT__ "MatRetrieveValues_SeqAIJ" 2320 int MatRetrieveValues_SeqAIJ(Mat mat) 2321 { 2322 Mat_SeqAIJ *aij = (Mat_SeqAIJ *)mat->data; 2323 int nz = aij->i[mat->m],ierr; 2324 2325 PetscFunctionBegin; 2326 if (aij->nonew != 1) { 2327 SETERRQ(1,"Must call MatSetOption(A,MAT_NO_NEW_NONZERO_LOCATIONS);first"); 2328 } 2329 if (!aij->saved_values) { 2330 SETERRQ(1,"Must call MatStoreValues(A);first"); 2331 } 2332 2333 /* copy values over */ 2334 ierr = PetscMemcpy(aij->a,aij->saved_values,nz*sizeof(PetscScalar));CHKERRQ(ierr); 2335 PetscFunctionReturn(0); 2336 } 2337 EXTERN_C_END 2338 2339 #undef __FUNCT__ 2340 #define __FUNCT__ "MatRetrieveValues" 2341 /*@ 2342 MatRetrieveValues - Retrieves the copy of the matrix values; this allows, for 2343 example, reuse of the linear part of a Jacobian, while recomputing the 2344 nonlinear portion. 2345 2346 Collect on Mat 2347 2348 Input Parameters: 2349 . mat - the matrix (currently on AIJ matrices support this option) 2350 2351 Level: advanced 2352 2353 .seealso: MatStoreValues() 2354 2355 @*/ 2356 int MatRetrieveValues(Mat mat) 2357 { 2358 int ierr,(*f)(Mat); 2359 2360 PetscFunctionBegin; 2361 PetscValidHeaderSpecific(mat,MAT_COOKIE,1); 2362 if (!mat->assembled) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix"); 2363 if (mat->factor) SETERRQ(PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix"); 2364 2365 ierr = PetscObjectQueryFunction((PetscObject)mat,"MatRetrieveValues_C",(void (**)(void))&f);CHKERRQ(ierr); 2366 if (f) { 2367 ierr = (*f)(mat);CHKERRQ(ierr); 2368 } else { 2369 SETERRQ(1,"Wrong type of matrix to retrieve values"); 2370 } 2371 PetscFunctionReturn(0); 2372 } 2373 2374 2375 /* --------------------------------------------------------------------------------*/ 2376 #undef __FUNCT__ 2377 #define __FUNCT__ "MatCreateSeqAIJ" 2378 /*@C 2379 MatCreateSeqAIJ - Creates a sparse matrix in AIJ (compressed row) format 2380 (the default parallel PETSc format). For good matrix assembly performance 2381 the user should preallocate the matrix storage by setting the parameter nz 2382 (or the array nnz). By setting these parameters accurately, performance 2383 during matrix assembly can be increased by more than a factor of 50. 2384 2385 Collective on MPI_Comm 2386 2387 Input Parameters: 2388 + comm - MPI communicator, set to PETSC_COMM_SELF 2389 . m - number of rows 2390 . n - number of columns 2391 . nz - number of nonzeros per row (same for all rows) 2392 - nnz - array containing the number of nonzeros in the various rows 2393 (possibly different for each row) or PETSC_NULL 2394 2395 Output Parameter: 2396 . A - the matrix 2397 2398 Notes: 2399 The AIJ format (also called the Yale sparse matrix format or 2400 compressed row storage), is fully compatible with standard Fortran 77 2401 storage. That is, the stored row and column indices can begin at 2402 either one (as in Fortran) or zero. See the users' manual for details. 2403 2404 Specify the preallocated storage with either nz or nnz (not both). 2405 Set nz=PETSC_DEFAULT and nnz=PETSC_NULL for PETSc to control dynamic memory 2406 allocation. For large problems you MUST preallocate memory or you 2407 will get TERRIBLE performance, see the users' manual chapter on matrices. 2408 2409 By default, this format uses inodes (identical nodes) when possible, to 2410 improve numerical efficiency of matrix-vector products and solves. We 2411 search for consecutive rows with the same nonzero structure, thereby 2412 reusing matrix information to achieve increased efficiency. 2413 2414 Options Database Keys: 2415 + -mat_aij_no_inode - Do not use inodes 2416 . -mat_aij_inode_limit <limit> - Sets inode limit (max limit=5) 2417 - -mat_aij_oneindex - Internally use indexing starting at 1 2418 rather than 0. Note that when calling MatSetValues(), 2419 the user still MUST index entries starting at 0! 2420 2421 Level: intermediate 2422 2423 .seealso: MatCreate(), MatCreateMPIAIJ(), MatSetValues(), MatSeqAIJSetColumnIndices(), MatCreateSeqAIJWithArrays() 2424 2425 @*/ 2426 int MatCreateSeqAIJ(MPI_Comm comm,int m,int n,int nz,const int nnz[],Mat *A) 2427 { 2428 int ierr; 2429 2430 PetscFunctionBegin; 2431 ierr = MatCreate(comm,m,n,m,n,A);CHKERRQ(ierr); 2432 ierr = MatSetType(*A,MATSEQAIJ);CHKERRQ(ierr); 2433 ierr = MatSeqAIJSetPreallocation(*A,nz,nnz);CHKERRQ(ierr); 2434 PetscFunctionReturn(0); 2435 } 2436 2437 #define SKIP_ALLOCATION -4 2438 2439 #undef __FUNCT__ 2440 #define __FUNCT__ "MatSeqAIJSetPreallocation" 2441 /*@C 2442 MatSeqAIJSetPreallocation - For good matrix assembly performance 2443 the user should preallocate the matrix storage by setting the parameter nz 2444 (or the array nnz). By setting these parameters accurately, performance 2445 during matrix assembly can be increased by more than a factor of 50. 2446 2447 Collective on MPI_Comm 2448 2449 Input Parameters: 2450 + comm - MPI communicator, set to PETSC_COMM_SELF 2451 . m - number of rows 2452 . n - number of columns 2453 . nz - number of nonzeros per row (same for all rows) 2454 - nnz - array containing the number of nonzeros in the various rows 2455 (possibly different for each row) or PETSC_NULL 2456 2457 Output Parameter: 2458 . A - the matrix 2459 2460 Notes: 2461 The AIJ format (also called the Yale sparse matrix format or 2462 compressed row storage), is fully compatible with standard Fortran 77 2463 storage. That is, the stored row and column indices can begin at 2464 either one (as in Fortran) or zero. See the users' manual for details. 2465 2466 Specify the preallocated storage with either nz or nnz (not both). 2467 Set nz=PETSC_DEFAULT and nnz=PETSC_NULL for PETSc to control dynamic memory 2468 allocation. For large problems you MUST preallocate memory or you 2469 will get TERRIBLE performance, see the users' manual chapter on matrices. 2470 2471 By default, this format uses inodes (identical nodes) when possible, to 2472 improve numerical efficiency of matrix-vector products and solves. We 2473 search for consecutive rows with the same nonzero structure, thereby 2474 reusing matrix information to achieve increased efficiency. 2475 2476 Options Database Keys: 2477 + -mat_aij_no_inode - Do not use inodes 2478 . -mat_aij_inode_limit <limit> - Sets inode limit (max limit=5) 2479 - -mat_aij_oneindex - Internally use indexing starting at 1 2480 rather than 0. Note that when calling MatSetValues(), 2481 the user still MUST index entries starting at 0! 2482 2483 Level: intermediate 2484 2485 .seealso: MatCreate(), MatCreateMPIAIJ(), MatSetValues(), MatSeqAIJSetColumnIndices(), MatCreateSeqAIJWithArrays() 2486 2487 @*/ 2488 int MatSeqAIJSetPreallocation(Mat B,int nz,const int nnz[]) 2489 { 2490 int ierr,(*f)(Mat,int,const int[]); 2491 2492 PetscFunctionBegin; 2493 ierr = PetscObjectQueryFunction((PetscObject)B,"MatSeqAIJSetPreallocation_C",(void (**)(void))&f);CHKERRQ(ierr); 2494 if (f) { 2495 ierr = (*f)(B,nz,nnz);CHKERRQ(ierr); 2496 } 2497 PetscFunctionReturn(0); 2498 } 2499 2500 EXTERN_C_BEGIN 2501 #undef __FUNCT__ 2502 #define __FUNCT__ "MatSeqAIJSetPreallocation_SeqAIJ" 2503 int MatSeqAIJSetPreallocation_SeqAIJ(Mat B,int nz,int *nnz) 2504 { 2505 Mat_SeqAIJ *b; 2506 size_t len = 0; 2507 PetscTruth skipallocation = PETSC_FALSE; 2508 int i,ierr; 2509 2510 PetscFunctionBegin; 2511 2512 if (nz == SKIP_ALLOCATION) { 2513 skipallocation = PETSC_TRUE; 2514 nz = 0; 2515 } 2516 2517 if (nz == PETSC_DEFAULT || nz == PETSC_DECIDE) nz = 5; 2518 if (nz < 0) SETERRQ1(PETSC_ERR_ARG_OUTOFRANGE,"nz cannot be less than 0: value %d",nz); 2519 if (nnz) { 2520 for (i=0; i<B->m; i++) { 2521 if (nnz[i] < 0) SETERRQ2(PETSC_ERR_ARG_OUTOFRANGE,"nnz cannot be less than 0: local row %d value %d",i,nnz[i]); 2522 if (nnz[i] > B->n) SETERRQ3(PETSC_ERR_ARG_OUTOFRANGE,"nnz cannot be greater than row length: local row %d value %d rowlength %d",i,nnz[i],B->n); 2523 } 2524 } 2525 2526 B->preallocated = PETSC_TRUE; 2527 b = (Mat_SeqAIJ*)B->data; 2528 2529 ierr = PetscMalloc((B->m+1)*sizeof(int),&b->imax);CHKERRQ(ierr); 2530 if (!nnz) { 2531 if (nz == PETSC_DEFAULT || nz == PETSC_DECIDE) nz = 10; 2532 else if (nz <= 0) nz = 1; 2533 for (i=0; i<B->m; i++) b->imax[i] = nz; 2534 nz = nz*B->m; 2535 } else { 2536 nz = 0; 2537 for (i=0; i<B->m; i++) {b->imax[i] = nnz[i]; nz += nnz[i];} 2538 } 2539 2540 if (!skipallocation) { 2541 /* allocate the matrix space */ 2542 len = ((size_t) nz)*(sizeof(int) + sizeof(PetscScalar)) + (B->m+1)*sizeof(int); 2543 ierr = PetscMalloc(len,&b->a);CHKERRQ(ierr); 2544 b->j = (int*)(b->a + nz); 2545 ierr = PetscMemzero(b->j,nz*sizeof(int));CHKERRQ(ierr); 2546 b->i = b->j + nz; 2547 b->i[0] = 0; 2548 for (i=1; i<B->m+1; i++) { 2549 b->i[i] = b->i[i-1] + b->imax[i-1]; 2550 } 2551 b->singlemalloc = PETSC_TRUE; 2552 b->freedata = PETSC_TRUE; 2553 } else { 2554 b->freedata = PETSC_FALSE; 2555 } 2556 2557 /* b->ilen will count nonzeros in each row so far. */ 2558 ierr = PetscMalloc((B->m+1)*sizeof(int),&b->ilen);CHKERRQ(ierr); 2559 PetscLogObjectMemory(B,len+2*(B->m+1)*sizeof(int)+sizeof(struct _p_Mat)+sizeof(Mat_SeqAIJ)); 2560 for (i=0; i<B->m; i++) { b->ilen[i] = 0;} 2561 2562 b->nz = 0; 2563 b->maxnz = nz; 2564 B->info.nz_unneeded = (double)b->maxnz; 2565 PetscFunctionReturn(0); 2566 } 2567 EXTERN_C_END 2568 2569 /*MC 2570 MATSEQAIJ - MATSEQAIJ = "seqaij" - A matrix type to be used for sequential sparse matrices, 2571 based on compressed sparse row format. 2572 2573 Options Database Keys: 2574 . -mat_type seqaij - sets the matrix type to "seqaij" during a call to MatSetFromOptions() 2575 2576 Level: beginner 2577 2578 .seealso: MatCreateSeqAIJ(), MatSetFromOptions(), MatSetType(), MatCreate(), MatType 2579 M*/ 2580 2581 EXTERN_C_BEGIN 2582 #undef __FUNCT__ 2583 #define __FUNCT__ "MatCreate_SeqAIJ" 2584 int MatCreate_SeqAIJ(Mat B) 2585 { 2586 Mat_SeqAIJ *b; 2587 int ierr,size; 2588 2589 PetscFunctionBegin; 2590 ierr = MPI_Comm_size(B->comm,&size);CHKERRQ(ierr); 2591 if (size > 1) SETERRQ(PETSC_ERR_ARG_OUTOFRANGE,"Comm must be of size 1"); 2592 2593 B->m = B->M = PetscMax(B->m,B->M); 2594 B->n = B->N = PetscMax(B->n,B->N); 2595 2596 ierr = PetscNew(Mat_SeqAIJ,&b);CHKERRQ(ierr); 2597 B->data = (void*)b; 2598 ierr = PetscMemzero(b,sizeof(Mat_SeqAIJ));CHKERRQ(ierr); 2599 ierr = PetscMemcpy(B->ops,&MatOps_Values,sizeof(struct _MatOps));CHKERRQ(ierr); 2600 B->factor = 0; 2601 B->lupivotthreshold = 1.0; 2602 B->mapping = 0; 2603 ierr = PetscOptionsGetReal(B->prefix,"-mat_lu_pivotthreshold",&B->lupivotthreshold,PETSC_NULL);CHKERRQ(ierr); 2604 ierr = PetscOptionsHasName(B->prefix,"-pc_ilu_preserve_row_sums",&b->ilu_preserve_row_sums);CHKERRQ(ierr); 2605 b->row = 0; 2606 b->col = 0; 2607 b->icol = 0; 2608 b->reallocs = 0; 2609 2610 ierr = PetscMapCreateMPI(B->comm,B->m,B->m,&B->rmap);CHKERRQ(ierr); 2611 ierr = PetscMapCreateMPI(B->comm,B->n,B->n,&B->cmap);CHKERRQ(ierr); 2612 2613 b->sorted = PETSC_FALSE; 2614 b->ignorezeroentries = PETSC_FALSE; 2615 b->roworiented = PETSC_TRUE; 2616 b->nonew = 0; 2617 b->diag = 0; 2618 b->solve_work = 0; 2619 B->spptr = 0; 2620 b->inode.use = PETSC_TRUE; 2621 b->inode.node_count = 0; 2622 b->inode.size = 0; 2623 b->inode.limit = 5; 2624 b->inode.max_limit = 5; 2625 b->saved_values = 0; 2626 b->idiag = 0; 2627 b->ssor = 0; 2628 b->keepzeroedrows = PETSC_FALSE; 2629 b->xtoy = 0; 2630 b->XtoY = 0; 2631 2632 ierr = PetscObjectChangeTypeName((PetscObject)B,MATSEQAIJ);CHKERRQ(ierr); 2633 2634 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatSeqAIJSetColumnIndices_C", 2635 "MatSeqAIJSetColumnIndices_SeqAIJ", 2636 MatSeqAIJSetColumnIndices_SeqAIJ);CHKERRQ(ierr); 2637 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatStoreValues_C", 2638 "MatStoreValues_SeqAIJ", 2639 MatStoreValues_SeqAIJ);CHKERRQ(ierr); 2640 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatRetrieveValues_C", 2641 "MatRetrieveValues_SeqAIJ", 2642 MatRetrieveValues_SeqAIJ);CHKERRQ(ierr); 2643 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_seqaij_seqsbaij_C", 2644 "MatConvert_SeqAIJ_SeqSBAIJ", 2645 MatConvert_SeqAIJ_SeqSBAIJ);CHKERRQ(ierr); 2646 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatConvert_seqaij_seqbaij_C", 2647 "MatConvert_SeqAIJ_SeqBAIJ", 2648 MatConvert_SeqAIJ_SeqBAIJ);CHKERRQ(ierr); 2649 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatIsTranspose_C", 2650 "MatIsTranspose_SeqAIJ", 2651 MatIsTranspose_SeqAIJ);CHKERRQ(ierr); 2652 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatSeqAIJSetPreallocation_C", 2653 "MatSeqAIJSetPreallocation_SeqAIJ", 2654 MatSeqAIJSetPreallocation_SeqAIJ);CHKERRQ(ierr); 2655 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatReorderForNonzeroDiagonal_C", 2656 "MatReorderForNonzeroDiagonal_SeqAIJ", 2657 MatReorderForNonzeroDiagonal_SeqAIJ);CHKERRQ(ierr); 2658 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatAdjustForInodes_C", 2659 "MatAdjustForInodes_SeqAIJ", 2660 MatAdjustForInodes_SeqAIJ);CHKERRQ(ierr); 2661 ierr = PetscObjectComposeFunctionDynamic((PetscObject)B,"MatSeqAIJGetInodeSizes_C", 2662 "MatSeqAIJGetInodeSizes_SeqAIJ", 2663 MatSeqAIJGetInodeSizes_SeqAIJ);CHKERRQ(ierr); 2664 ierr = RegisterApplyPtAPRoutines_Private(B);CHKERRQ(ierr); 2665 PetscFunctionReturn(0); 2666 } 2667 EXTERN_C_END 2668 2669 #undef __FUNCT__ 2670 #define __FUNCT__ "MatDuplicate_SeqAIJ" 2671 int MatDuplicate_SeqAIJ(Mat A,MatDuplicateOption cpvalues,Mat *B) 2672 { 2673 Mat C; 2674 Mat_SeqAIJ *c,*a = (Mat_SeqAIJ*)A->data; 2675 int i,m = A->m,ierr; 2676 size_t len; 2677 2678 PetscFunctionBegin; 2679 *B = 0; 2680 ierr = MatCreate(A->comm,A->m,A->n,A->m,A->n,&C);CHKERRQ(ierr); 2681 ierr = MatSetType(C,A->type_name);CHKERRQ(ierr); 2682 ierr = PetscMemcpy(C->ops,A->ops,sizeof(struct _MatOps));CHKERRQ(ierr); 2683 2684 c = (Mat_SeqAIJ*)C->data; 2685 2686 C->factor = A->factor; 2687 c->row = 0; 2688 c->col = 0; 2689 c->icol = 0; 2690 c->keepzeroedrows = a->keepzeroedrows; 2691 C->assembled = PETSC_TRUE; 2692 2693 C->M = A->m; 2694 C->N = A->n; 2695 2696 ierr = PetscMalloc((m+1)*sizeof(int),&c->imax);CHKERRQ(ierr); 2697 ierr = PetscMalloc((m+1)*sizeof(int),&c->ilen);CHKERRQ(ierr); 2698 for (i=0; i<m; i++) { 2699 c->imax[i] = a->imax[i]; 2700 c->ilen[i] = a->ilen[i]; 2701 } 2702 2703 /* allocate the matrix space */ 2704 c->singlemalloc = PETSC_TRUE; 2705 len = ((size_t) (m+1))*sizeof(int)+(a->i[m])*(sizeof(PetscScalar)+sizeof(int)); 2706 ierr = PetscMalloc(len,&c->a);CHKERRQ(ierr); 2707 c->j = (int*)(c->a + a->i[m] ); 2708 c->i = c->j + a->i[m]; 2709 ierr = PetscMemcpy(c->i,a->i,(m+1)*sizeof(int));CHKERRQ(ierr); 2710 if (m > 0) { 2711 ierr = PetscMemcpy(c->j,a->j,(a->i[m])*sizeof(int));CHKERRQ(ierr); 2712 if (cpvalues == MAT_COPY_VALUES) { 2713 ierr = PetscMemcpy(c->a,a->a,(a->i[m])*sizeof(PetscScalar));CHKERRQ(ierr); 2714 } else { 2715 ierr = PetscMemzero(c->a,(a->i[m])*sizeof(PetscScalar));CHKERRQ(ierr); 2716 } 2717 } 2718 2719 PetscLogObjectMemory(C,len+2*(m+1)*sizeof(int)+sizeof(struct _p_Mat)+sizeof(Mat_SeqAIJ)); 2720 c->sorted = a->sorted; 2721 c->roworiented = a->roworiented; 2722 c->nonew = a->nonew; 2723 c->ilu_preserve_row_sums = a->ilu_preserve_row_sums; 2724 c->saved_values = 0; 2725 c->idiag = 0; 2726 c->ssor = 0; 2727 c->ignorezeroentries = a->ignorezeroentries; 2728 c->freedata = PETSC_TRUE; 2729 2730 if (a->diag) { 2731 ierr = PetscMalloc((m+1)*sizeof(int),&c->diag);CHKERRQ(ierr); 2732 PetscLogObjectMemory(C,(m+1)*sizeof(int)); 2733 for (i=0; i<m; i++) { 2734 c->diag[i] = a->diag[i]; 2735 } 2736 } else c->diag = 0; 2737 c->inode.use = a->inode.use; 2738 c->inode.limit = a->inode.limit; 2739 c->inode.max_limit = a->inode.max_limit; 2740 if (a->inode.size){ 2741 ierr = PetscMalloc((m+1)*sizeof(int),&c->inode.size);CHKERRQ(ierr); 2742 c->inode.node_count = a->inode.node_count; 2743 ierr = PetscMemcpy(c->inode.size,a->inode.size,(m+1)*sizeof(int));CHKERRQ(ierr); 2744 } else { 2745 c->inode.size = 0; 2746 c->inode.node_count = 0; 2747 } 2748 c->nz = a->nz; 2749 c->maxnz = a->maxnz; 2750 c->solve_work = 0; 2751 C->preallocated = PETSC_TRUE; 2752 2753 *B = C; 2754 ierr = PetscFListDuplicate(A->qlist,&C->qlist);CHKERRQ(ierr); 2755 PetscFunctionReturn(0); 2756 } 2757 2758 #undef __FUNCT__ 2759 #define __FUNCT__ "MatLoad_SeqAIJ" 2760 int MatLoad_SeqAIJ(PetscViewer viewer,const MatType type,Mat *A) 2761 { 2762 Mat_SeqAIJ *a; 2763 Mat B; 2764 int i,nz,ierr,fd,header[4],size,*rowlengths = 0,M,N; 2765 MPI_Comm comm; 2766 2767 PetscFunctionBegin; 2768 ierr = PetscObjectGetComm((PetscObject)viewer,&comm);CHKERRQ(ierr); 2769 ierr = MPI_Comm_size(comm,&size);CHKERRQ(ierr); 2770 if (size > 1) SETERRQ(PETSC_ERR_ARG_SIZ,"view must have one processor"); 2771 ierr = PetscViewerBinaryGetDescriptor(viewer,&fd);CHKERRQ(ierr); 2772 ierr = PetscBinaryRead(fd,header,4,PETSC_INT);CHKERRQ(ierr); 2773 if (header[0] != MAT_FILE_COOKIE) SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"not matrix object in file"); 2774 M = header[1]; N = header[2]; nz = header[3]; 2775 2776 if (nz < 0) { 2777 SETERRQ(PETSC_ERR_FILE_UNEXPECTED,"Matrix stored in special format on disk,cannot load as SeqAIJ"); 2778 } 2779 2780 /* read in row lengths */ 2781 ierr = PetscMalloc(M*sizeof(int),&rowlengths);CHKERRQ(ierr); 2782 ierr = PetscBinaryRead(fd,rowlengths,M,PETSC_INT);CHKERRQ(ierr); 2783 2784 /* create our matrix */ 2785 ierr = MatCreate(comm,PETSC_DECIDE,PETSC_DECIDE,M,N,&B);CHKERRQ(ierr); 2786 ierr = MatSetType(B,type);CHKERRQ(ierr); 2787 ierr = MatSeqAIJSetPreallocation(B,0,rowlengths);CHKERRQ(ierr); 2788 a = (Mat_SeqAIJ*)B->data; 2789 2790 /* read in column indices and adjust for Fortran indexing*/ 2791 ierr = PetscBinaryRead(fd,a->j,nz,PETSC_INT);CHKERRQ(ierr); 2792 2793 /* read in nonzero values */ 2794 ierr = PetscBinaryRead(fd,a->a,nz,PETSC_SCALAR);CHKERRQ(ierr); 2795 2796 /* set matrix "i" values */ 2797 a->i[0] = 0; 2798 for (i=1; i<= M; i++) { 2799 a->i[i] = a->i[i-1] + rowlengths[i-1]; 2800 a->ilen[i-1] = rowlengths[i-1]; 2801 } 2802 ierr = PetscFree(rowlengths);CHKERRQ(ierr); 2803 2804 ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2805 ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2806 *A = B; 2807 PetscFunctionReturn(0); 2808 } 2809 2810 #undef __FUNCT__ 2811 #define __FUNCT__ "MatEqual_SeqAIJ" 2812 int MatEqual_SeqAIJ(Mat A,Mat B,PetscTruth* flg) 2813 { 2814 Mat_SeqAIJ *a = (Mat_SeqAIJ *)A->data,*b = (Mat_SeqAIJ *)B->data; 2815 int ierr; 2816 2817 PetscFunctionBegin; 2818 /* If the matrix dimensions are not equal,or no of nonzeros */ 2819 if ((A->m != B->m) || (A->n != B->n) ||(a->nz != b->nz)) { 2820 *flg = PETSC_FALSE; 2821 PetscFunctionReturn(0); 2822 } 2823 2824 /* if the a->i are the same */ 2825 ierr = PetscMemcmp(a->i,b->i,(A->m+1)*sizeof(int),flg);CHKERRQ(ierr); 2826 if (*flg == PETSC_FALSE) PetscFunctionReturn(0); 2827 2828 /* if a->j are the same */ 2829 ierr = PetscMemcmp(a->j,b->j,(a->nz)*sizeof(int),flg);CHKERRQ(ierr); 2830 if (*flg == PETSC_FALSE) PetscFunctionReturn(0); 2831 2832 /* if a->a are the same */ 2833 ierr = PetscMemcmp(a->a,b->a,(a->nz)*sizeof(PetscScalar),flg);CHKERRQ(ierr); 2834 2835 PetscFunctionReturn(0); 2836 2837 } 2838 2839 #undef __FUNCT__ 2840 #define __FUNCT__ "MatCreateSeqAIJWithArrays" 2841 /*@C 2842 MatCreateSeqAIJWithArrays - Creates an sequential AIJ matrix using matrix elements (in CSR format) 2843 provided by the user. 2844 2845 Coolective on MPI_Comm 2846 2847 Input Parameters: 2848 + comm - must be an MPI communicator of size 1 2849 . m - number of rows 2850 . n - number of columns 2851 . i - row indices 2852 . j - column indices 2853 - a - matrix values 2854 2855 Output Parameter: 2856 . mat - the matrix 2857 2858 Level: intermediate 2859 2860 Notes: 2861 The i, j, and a arrays are not copied by this routine, the user must free these arrays 2862 once the matrix is destroyed 2863 2864 You cannot set new nonzero locations into this matrix, that will generate an error. 2865 2866 The i and j indices are 0 based 2867 2868 .seealso: MatCreate(), MatCreateMPIAIJ(), MatCreateSeqAIJ() 2869 2870 @*/ 2871 int MatCreateSeqAIJWithArrays(MPI_Comm comm,int m,int n,int* i,int*j,PetscScalar *a,Mat *mat) 2872 { 2873 int ierr,ii; 2874 Mat_SeqAIJ *aij; 2875 2876 PetscFunctionBegin; 2877 ierr = MatCreate(comm,m,n,m,n,mat);CHKERRQ(ierr); 2878 ierr = MatSetType(*mat,MATSEQAIJ);CHKERRQ(ierr); 2879 ierr = MatSeqAIJSetPreallocation(*mat,SKIP_ALLOCATION,0);CHKERRQ(ierr); 2880 aij = (Mat_SeqAIJ*)(*mat)->data; 2881 2882 if (i[0] != 0) { 2883 SETERRQ(1,"i (row indices) must start with 0"); 2884 } 2885 aij->i = i; 2886 aij->j = j; 2887 aij->a = a; 2888 aij->singlemalloc = PETSC_FALSE; 2889 aij->nonew = -1; /*this indicates that inserting a new value in the matrix that generates a new nonzero is an error*/ 2890 aij->freedata = PETSC_FALSE; 2891 2892 for (ii=0; ii<m; ii++) { 2893 aij->ilen[ii] = aij->imax[ii] = i[ii+1] - i[ii]; 2894 #if defined(PETSC_USE_BOPT_g) 2895 if (i[ii+1] - i[ii] < 0) SETERRQ2(1,"Negative row length in i (row indices) row = %d length = %d",ii,i[ii+1] - i[ii]); 2896 #endif 2897 } 2898 #if defined(PETSC_USE_BOPT_g) 2899 for (ii=0; ii<aij->i[m]; ii++) { 2900 if (j[ii] < 0) SETERRQ2(1,"Negative column index at location = %d index = %d",ii,j[ii]); 2901 if (j[ii] > n - 1) SETERRQ2(1,"Column index to large at location = %d index = %d",ii,j[ii]); 2902 } 2903 #endif 2904 2905 ierr = MatAssemblyBegin(*mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2906 ierr = MatAssemblyEnd(*mat,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr); 2907 PetscFunctionReturn(0); 2908 } 2909 2910 #undef __FUNCT__ 2911 #define __FUNCT__ "MatSetColoring_SeqAIJ" 2912 int MatSetColoring_SeqAIJ(Mat A,ISColoring coloring) 2913 { 2914 int ierr; 2915 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 2916 2917 PetscFunctionBegin; 2918 if (coloring->ctype == IS_COLORING_LOCAL) { 2919 ierr = ISColoringReference(coloring);CHKERRQ(ierr); 2920 a->coloring = coloring; 2921 } else if (coloring->ctype == IS_COLORING_GHOSTED) { 2922 int i,*larray; 2923 ISColoring ocoloring; 2924 ISColoringValue *colors; 2925 2926 /* set coloring for diagonal portion */ 2927 ierr = PetscMalloc((A->n+1)*sizeof(int),&larray);CHKERRQ(ierr); 2928 for (i=0; i<A->n; i++) { 2929 larray[i] = i; 2930 } 2931 ierr = ISGlobalToLocalMappingApply(A->mapping,IS_GTOLM_MASK,A->n,larray,PETSC_NULL,larray);CHKERRQ(ierr); 2932 ierr = PetscMalloc((A->n+1)*sizeof(ISColoringValue),&colors);CHKERRQ(ierr); 2933 for (i=0; i<A->n; i++) { 2934 colors[i] = coloring->colors[larray[i]]; 2935 } 2936 ierr = PetscFree(larray);CHKERRQ(ierr); 2937 ierr = ISColoringCreate(PETSC_COMM_SELF,A->n,colors,&ocoloring);CHKERRQ(ierr); 2938 a->coloring = ocoloring; 2939 } 2940 PetscFunctionReturn(0); 2941 } 2942 2943 #if defined(PETSC_HAVE_ADIC) && !defined(PETSC_USE_COMPLEX) && !defined(PETSC_USE_SINGLE) 2944 EXTERN_C_BEGIN 2945 #include "adic/ad_utils.h" 2946 EXTERN_C_END 2947 2948 #undef __FUNCT__ 2949 #define __FUNCT__ "MatSetValuesAdic_SeqAIJ" 2950 int MatSetValuesAdic_SeqAIJ(Mat A,void *advalues) 2951 { 2952 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 2953 int m = A->m,*ii = a->i,*jj = a->j,nz,i,j,nlen; 2954 PetscScalar *v = a->a,*values = ((PetscScalar*)advalues)+1; 2955 ISColoringValue *color; 2956 2957 PetscFunctionBegin; 2958 if (!a->coloring) SETERRQ(1,"Coloring not set for matrix"); 2959 nlen = PetscADGetDerivTypeSize()/sizeof(PetscScalar); 2960 color = a->coloring->colors; 2961 /* loop over rows */ 2962 for (i=0; i<m; i++) { 2963 nz = ii[i+1] - ii[i]; 2964 /* loop over columns putting computed value into matrix */ 2965 for (j=0; j<nz; j++) { 2966 *v++ = values[color[*jj++]]; 2967 } 2968 values += nlen; /* jump to next row of derivatives */ 2969 } 2970 PetscFunctionReturn(0); 2971 } 2972 2973 #else 2974 2975 #undef __FUNCT__ 2976 #define __FUNCT__ "MatSetValuesAdic_SeqAIJ" 2977 int MatSetValuesAdic_SeqAIJ(Mat A,void *advalues) 2978 { 2979 PetscFunctionBegin; 2980 SETERRQ(1,"PETSc installed without ADIC"); 2981 } 2982 2983 #endif 2984 2985 #undef __FUNCT__ 2986 #define __FUNCT__ "MatSetValuesAdifor_SeqAIJ" 2987 int MatSetValuesAdifor_SeqAIJ(Mat A,int nl,void *advalues) 2988 { 2989 Mat_SeqAIJ *a = (Mat_SeqAIJ*)A->data; 2990 int m = A->m,*ii = a->i,*jj = a->j,nz,i,j; 2991 PetscScalar *v = a->a,*values = (PetscScalar *)advalues; 2992 ISColoringValue *color; 2993 2994 PetscFunctionBegin; 2995 if (!a->coloring) SETERRQ(1,"Coloring not set for matrix"); 2996 color = a->coloring->colors; 2997 /* loop over rows */ 2998 for (i=0; i<m; i++) { 2999 nz = ii[i+1] - ii[i]; 3000 /* loop over columns putting computed value into matrix */ 3001 for (j=0; j<nz; j++) { 3002 *v++ = values[color[*jj++]]; 3003 } 3004 values += nl; /* jump to next row of derivatives */ 3005 } 3006 PetscFunctionReturn(0); 3007 } 3008 3009