Actual source code: mmaij.c
1: /*$Id: mmaij.c,v 1.59 2001/08/07 03:02:49 balay Exp $*/
3: /*
4: Support for the parallel AIJ matrix vector multiply
5: */
6: #include src/mat/impls/aij/mpi/mpiaij.h
7: #include src/vec/vecimpl.h
11: int MatSetUpMultiply_MPIAIJ(Mat mat)
12: {
13: Mat_MPIAIJ *aij = (Mat_MPIAIJ*)mat->data;
14: Mat_SeqAIJ *B = (Mat_SeqAIJ*)(aij->B->data);
15: int N = mat->N,i,j,*indices,*aj = B->j,ierr,ec = 0,*garray;
16: IS from,to;
17: Vec gvec;
18: #if defined (PETSC_USE_CTABLE)
19: PetscTable gid1_lid1;
20: PetscTablePosition tpos;
21: int gid,lid;
22: #endif
26: #if defined (PETSC_USE_CTABLE)
27: /* use a table - Mark Adams (this has not been tested with "shift") */
28: PetscTableCreate(aij->B->m,&gid1_lid1);
29: for (i=0; i<aij->B->m; i++) {
30: for (j=0; j<B->ilen[i]; j++) {
31: int data,gid1 = aj[B->i[i] + j] + 1;
32: PetscTableFind(gid1_lid1,gid1,&data);
33: if (!data) {
34: /* one based table */
35: PetscTableAdd(gid1_lid1,gid1,++ec);
36: }
37: }
38: }
39: /* form array of columns we need */
40: PetscMalloc((ec+1)*sizeof(int),&garray);
41: PetscTableGetHeadPosition(gid1_lid1,&tpos);
42: while (tpos) {
43: PetscTableGetNext(gid1_lid1,&tpos,&gid,&lid);
44: gid--;
45: lid--;
46: garray[lid] = gid;
47: }
48: PetscSortInt(ec,garray); /* sort, and rebuild */
49: PetscTableRemoveAll(gid1_lid1);
50: for (i=0; i<ec; i++) {
51: PetscTableAdd(gid1_lid1,garray[i]+1,i+1);
52: }
53: /* compact out the extra columns in B */
54: for (i=0; i<aij->B->m; i++) {
55: for (j=0; j<B->ilen[i]; j++) {
56: int gid1 = aj[B->i[i] + j] + 1;
57: PetscTableFind(gid1_lid1,gid1,&lid);
58: lid --;
59: aj[B->i[i] + j] = lid;
60: }
61: }
62: aij->B->n = aij->B->N = ec;
63: PetscTableDelete(gid1_lid1);
64: /* Mark Adams */
65: #else
66: /* For the first stab we make an array as long as the number of columns */
67: /* mark those columns that are in aij->B */
68: PetscMalloc((N+1)*sizeof(int),&indices);
69: PetscMemzero(indices,N*sizeof(int));
70: for (i=0; i<aij->B->m; i++) {
71: for (j=0; j<B->ilen[i]; j++) {
72: if (!indices[aj[B->i[i] + j] ]) ec++;
73: indices[aj[B->i[i] + j] ] = 1;
74: }
75: }
77: /* form array of columns we need */
78: PetscMalloc((ec+1)*sizeof(int),&garray);
79: ec = 0;
80: for (i=0; i<N; i++) {
81: if (indices[i]) garray[ec++] = i;
82: }
84: /* make indices now point into garray */
85: for (i=0; i<ec; i++) {
86: indices[garray[i]] = i;
87: }
89: /* compact out the extra columns in B */
90: for (i=0; i<aij->B->m; i++) {
91: for (j=0; j<B->ilen[i]; j++) {
92: aj[B->i[i] + j] = indices[aj[B->i[i] + j]];
93: }
94: }
95: aij->B->n = aij->B->N = ec;
96: PetscFree(indices);
97: #endif
98: /* create local vector that is used to scatter into */
99: VecCreateSeq(PETSC_COMM_SELF,ec,&aij->lvec);
101: /* create two temporary Index sets for build scatter gather */
102: ISCreateGeneral(mat->comm,ec,garray,&from);
103: ISCreateStride(PETSC_COMM_SELF,ec,0,1,&to);
105: /* create temporary global vector to generate scatter context */
106: /* this is inefficient, but otherwise we must do either
107: 1) save garray until the first actual scatter when the vector is known or
108: 2) have another way of generating a scatter context without a vector.*/
109: VecCreateMPI(mat->comm,mat->n,mat->N,&gvec);
111: /* generate the scatter context */
112: VecScatterCreate(gvec,from,aij->lvec,to,&aij->Mvctx);
113: PetscLogObjectParent(mat,aij->Mvctx);
114: PetscLogObjectParent(mat,aij->lvec);
115: PetscLogObjectParent(mat,from);
116: PetscLogObjectParent(mat,to);
117: aij->garray = garray;
118: PetscLogObjectMemory(mat,(ec+1)*sizeof(int));
119: ISDestroy(from);
120: ISDestroy(to);
121: VecDestroy(gvec);
122: return(0);
123: }
128: /*
129: Takes the local part of an already assembled MPIAIJ matrix
130: and disassembles it. This is to allow new nonzeros into the matrix
131: that require more communication in the matrix vector multiply.
132: Thus certain data-structures must be rebuilt.
134: Kind of slow! But that's what application programmers get when
135: they are sloppy.
136: */
137: int DisAssemble_MPIAIJ(Mat A)
138: {
139: Mat_MPIAIJ *aij = (Mat_MPIAIJ*)A->data;
140: Mat B = aij->B,Bnew;
141: Mat_SeqAIJ *Baij = (Mat_SeqAIJ*)B->data;
142: int ierr,i,j,m = B->m,n = A->N,col,ct = 0,*garray = aij->garray;
143: int *nz,ec;
144: PetscScalar v;
147: /* free stuff related to matrix-vec multiply */
148: VecGetSize(aij->lvec,&ec); /* needed for PetscLogObjectMemory below */
149: VecDestroy(aij->lvec); aij->lvec = 0;
150: VecScatterDestroy(aij->Mvctx); aij->Mvctx = 0;
151: if (aij->colmap) {
152: #if defined (PETSC_USE_CTABLE)
153: PetscTableDelete(aij->colmap);
154: aij->colmap = 0;
155: #else
156: PetscFree(aij->colmap);
157: aij->colmap = 0;
158: PetscLogObjectMemory(A,-aij->B->n*sizeof(int));
159: #endif
160: }
162: /* make sure that B is assembled so we can access its values */
163: MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);
164: MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);
166: /* invent new B and copy stuff over */
167: PetscMalloc((m+1)*sizeof(int),&nz);
168: for (i=0; i<m; i++) {
169: nz[i] = Baij->i[i+1] - Baij->i[i];
170: }
171: MatCreateSeqAIJ(PETSC_COMM_SELF,m,n,0,nz,&Bnew);
172: PetscFree(nz);
173: for (i=0; i<m; i++) {
174: for (j=Baij->i[i]; j<Baij->i[i+1]; j++) {
175: col = garray[Baij->j[ct]];
176: v = Baij->a[ct++];
177: MatSetValues(Bnew,1,&i,1,&col,&v,B->insertmode);
178: }
179: }
180: PetscFree(aij->garray);
181: aij->garray = 0;
182: PetscLogObjectMemory(A,-ec*sizeof(int));
183: MatDestroy(B);
184: PetscLogObjectParent(A,Bnew);
185: aij->B = Bnew;
186: A->was_assembled = PETSC_FALSE;
187: return(0);
188: }
190: /* ugly stuff added for Glenn someday we should fix this up */
192: static int *auglyrmapd = 0,*auglyrmapo = 0; /* mapping from the local ordering to the "diagonal" and "off-diagonal"
193: parts of the local matrix */
194: static Vec auglydd = 0,auglyoo = 0; /* work vectors used to scale the two parts of the local matrix */
199: int MatMPIAIJDiagonalScaleLocalSetUp(Mat inA,Vec scale)
200: {
201: Mat_MPIAIJ *ina = (Mat_MPIAIJ*) inA->data; /*access private part of matrix */
202: int ierr,i,n,nt,cstart,cend,no,*garray = ina->garray,*lindices;
203: int *r_rmapd,*r_rmapo;
204:
206: MatGetOwnershipRange(inA,&cstart,&cend);
207: MatGetSize(ina->A,PETSC_NULL,&n);
208: PetscMalloc((inA->mapping->n+1)*sizeof(int),&r_rmapd);
209: PetscMemzero(r_rmapd,inA->mapping->n*sizeof(int));
210: nt = 0;
211: for (i=0; i<inA->mapping->n; i++) {
212: if (inA->mapping->indices[i] >= cstart && inA->mapping->indices[i] < cend) {
213: nt++;
214: r_rmapd[i] = inA->mapping->indices[i] + 1;
215: }
216: }
217: if (nt != n) SETERRQ2(1,"Hmm nt %d n %d",nt,n);
218: PetscMalloc((n+1)*sizeof(int),&auglyrmapd);
219: for (i=0; i<inA->mapping->n; i++) {
220: if (r_rmapd[i]){
221: auglyrmapd[(r_rmapd[i]-1)-cstart] = i;
222: }
223: }
224: PetscFree(r_rmapd);
225: VecCreateSeq(PETSC_COMM_SELF,n,&auglydd);
227: PetscMalloc((inA->N+1)*sizeof(int),&lindices);
228: PetscMemzero(lindices,inA->N*sizeof(int));
229: for (i=0; i<ina->B->n; i++) {
230: lindices[garray[i]] = i+1;
231: }
232: no = inA->mapping->n - nt;
233: PetscMalloc((inA->mapping->n+1)*sizeof(int),&r_rmapo);
234: PetscMemzero(r_rmapo,inA->mapping->n*sizeof(int));
235: nt = 0;
236: for (i=0; i<inA->mapping->n; i++) {
237: if (lindices[inA->mapping->indices[i]]) {
238: nt++;
239: r_rmapo[i] = lindices[inA->mapping->indices[i]];
240: }
241: }
242: if (nt > no) SETERRQ2(1,"Hmm nt %d no %d",nt,n);
243: PetscFree(lindices);
244: PetscMalloc((nt+1)*sizeof(int),&auglyrmapo);
245: for (i=0; i<inA->mapping->n; i++) {
246: if (r_rmapo[i]){
247: auglyrmapo[(r_rmapo[i]-1)] = i;
248: }
249: }
250: PetscFree(r_rmapo);
251: VecCreateSeq(PETSC_COMM_SELF,nt,&auglyoo);
253: return(0);
254: }
258: int MatMPIAIJDiagonalScaleLocal(Mat A,Vec scale)
259: {
260: /* This routine should really be abandoned as it duplicates MatDiagonalScaleLocal */
261: int ierr,(*f)(Mat,Vec);
264: PetscObjectQueryFunction((PetscObject)A,"MatDiagonalScaleLocal_C",(void (**)(void))&f);
265: if (f) {
266: (*f)(A,scale);
267: }
268: return(0);
269: }
271: EXTERN_C_BEGIN
274: int MatDiagonalScaleLocal_MPIAIJ(Mat A,Vec scale)
275: {
276: Mat_MPIAIJ *a = (Mat_MPIAIJ*) A->data; /*access private part of matrix */
277: int ierr,n,i;
278: PetscScalar *d,*o,*s;
279:
281: if (!auglyrmapd) {
282: MatMPIAIJDiagonalScaleLocalSetUp(A,scale);
283: }
285: VecGetArray(scale,&s);
286:
287: VecGetLocalSize(auglydd,&n);
288: VecGetArray(auglydd,&d);
289: for (i=0; i<n; i++) {
290: d[i] = s[auglyrmapd[i]]; /* copy "diagonal" (true local) portion of scale into dd vector */
291: }
292: VecRestoreArray(auglydd,&d);
293: /* column scale "diagonal" portion of local matrix */
294: MatDiagonalScale(a->A,PETSC_NULL,auglydd);
296: VecGetLocalSize(auglyoo,&n);
297: VecGetArray(auglyoo,&o);
298: for (i=0; i<n; i++) {
299: o[i] = s[auglyrmapo[i]]; /* copy "off-diagonal" portion of scale into oo vector */
300: }
301: VecRestoreArray(scale,&s);
302: VecRestoreArray(auglyoo,&o);
303: /* column scale "off-diagonal" portion of local matrix */
304: MatDiagonalScale(a->B,PETSC_NULL,auglyoo);
306: return(0);
307: }
308: EXTERN_C_END