Actual source code: matscalapack.c
1: #include <petsc/private/petscscalapack.h>
3: const char ScaLAPACKCitation[] = "@BOOK{scalapack-user-guide,\n"
4: " AUTHOR = {L. S. Blackford and J. Choi and A. Cleary and E. D'Azevedo and\n"
5: " J. Demmel and I. Dhillon and J. Dongarra and S. Hammarling and\n"
6: " G. Henry and A. Petitet and K. Stanley and D. Walker and R. C. Whaley},\n"
7: " TITLE = {Sca{LAPACK} Users' Guide},\n"
8: " PUBLISHER = {SIAM},\n"
9: " ADDRESS = {Philadelphia, PA},\n"
10: " YEAR = 1997\n"
11: "}\n";
12: static PetscBool ScaLAPACKCite = PETSC_FALSE;
14: #define DEFAULT_BLOCKSIZE 64
16: /*
17: The variable Petsc_ScaLAPACK_keyval is used to indicate an MPI attribute that
18: is attached to a communicator, in this case the attribute is a Mat_ScaLAPACK_Grid
19: */
20: static PetscMPIInt Petsc_ScaLAPACK_keyval = MPI_KEYVAL_INVALID;
22: static PetscErrorCode Petsc_ScaLAPACK_keyval_free(void)
23: {
27: PetscInfo(NULL,"Freeing Petsc_ScaLAPACK_keyval\n");
28: MPI_Comm_free_keyval(&Petsc_ScaLAPACK_keyval);
29: return(0);
30: }
32: static PetscErrorCode MatView_ScaLAPACK(Mat A,PetscViewer viewer)
33: {
34: PetscErrorCode ierr;
35: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
36: PetscBool iascii;
37: PetscViewerFormat format;
38: Mat Adense;
41: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);
42: if (iascii) {
43: PetscViewerGetFormat(viewer,&format);
44: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
45: PetscViewerASCIIPrintf(viewer,"block sizes: %d,%d\n",(int)a->mb,(int)a->nb);
46: PetscViewerASCIIPrintf(viewer,"grid height=%d, grid width=%d\n",(int)a->grid->nprow,(int)a->grid->npcol);
47: PetscViewerASCIIPrintf(viewer,"coordinates of process owning first row and column: (%d,%d)\n",(int)a->rsrc,(int)a->csrc);
48: PetscViewerASCIIPrintf(viewer,"dimension of largest local matrix: %d x %d\n",(int)a->locr,(int)a->locc);
49: return(0);
50: } else if (format == PETSC_VIEWER_ASCII_FACTOR_INFO) {
51: return(0);
52: }
53: }
54: /* convert to dense format and call MatView() */
55: MatConvert(A,MATDENSE,MAT_INITIAL_MATRIX,&Adense);
56: MatView(Adense,viewer);
57: MatDestroy(&Adense);
58: return(0);
59: }
61: static PetscErrorCode MatGetInfo_ScaLAPACK(Mat A,MatInfoType flag,MatInfo *info)
62: {
64: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
65: PetscLogDouble isend[2],irecv[2];
68: info->block_size = 1.0;
70: isend[0] = a->lld*a->locc; /* locally allocated */
71: isend[1] = a->locr*a->locc; /* used submatrix */
72: if (flag == MAT_LOCAL || flag == MAT_GLOBAL_MAX) {
73: info->nz_allocated = isend[0];
74: info->nz_used = isend[1];
75: } else if (flag == MAT_GLOBAL_MAX) {
76: MPIU_Allreduce(isend,irecv,2,MPIU_PETSCLOGDOUBLE,MPIU_MAX,PetscObjectComm((PetscObject)A));
77: info->nz_allocated = irecv[0];
78: info->nz_used = irecv[1];
79: } else if (flag == MAT_GLOBAL_SUM) {
80: MPIU_Allreduce(isend,irecv,2,MPIU_PETSCLOGDOUBLE,MPIU_SUM,PetscObjectComm((PetscObject)A));
81: info->nz_allocated = irecv[0];
82: info->nz_used = irecv[1];
83: }
85: info->nz_unneeded = 0;
86: info->assemblies = A->num_ass;
87: info->mallocs = 0;
88: info->memory = ((PetscObject)A)->mem;
89: info->fill_ratio_given = 0;
90: info->fill_ratio_needed = 0;
91: info->factor_mallocs = 0;
92: return(0);
93: }
95: PetscErrorCode MatSetOption_ScaLAPACK(Mat A,MatOption op,PetscBool flg)
96: {
98: switch (op) {
99: case MAT_NEW_NONZERO_LOCATIONS:
100: case MAT_NEW_NONZERO_LOCATION_ERR:
101: case MAT_NEW_NONZERO_ALLOCATION_ERR:
102: case MAT_SYMMETRIC:
103: case MAT_SORTED_FULL:
104: case MAT_HERMITIAN:
105: break;
106: default:
107: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Unsupported option %s",MatOptions[op]);
108: }
109: return(0);
110: }
112: static PetscErrorCode MatSetValues_ScaLAPACK(Mat A,PetscInt nr,const PetscInt *rows,PetscInt nc,const PetscInt *cols,const PetscScalar *vals,InsertMode imode)
113: {
114: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
116: PetscInt i,j;
117: PetscBLASInt gridx,gcidx,lridx,lcidx,rsrc,csrc;
120: for (i=0;i<nr;i++) {
121: if (rows[i] < 0) continue;
122: PetscBLASIntCast(rows[i]+1,&gridx);
123: for (j=0;j<nc;j++) {
124: if (cols[j] < 0) continue;
125: PetscBLASIntCast(cols[j]+1,&gcidx);
126: PetscStackCallBLAS("SCALAPACKinfog2l",SCALAPACKinfog2l_(&gridx,&gcidx,a->desc,&a->grid->nprow,&a->grid->npcol,&a->grid->myrow,&a->grid->mycol,&lridx,&lcidx,&rsrc,&csrc));
127: if (rsrc==a->grid->myrow && csrc==a->grid->mycol) {
128: switch (imode) {
129: case INSERT_VALUES: a->loc[lridx-1+(lcidx-1)*a->lld] = vals[i*nc+j]; break;
130: case ADD_VALUES: a->loc[lridx-1+(lcidx-1)*a->lld] += vals[i*nc+j]; break;
131: default: SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"No support for InsertMode %d",(int)imode);
132: }
133: } else {
134: if (A->nooffprocentries) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Setting off process entry even though MatSetOption(,MAT_NO_OFF_PROC_ENTRIES,PETSC_TRUE) was set");
135: A->assembled = PETSC_FALSE;
136: MatStashValuesRow_Private(&A->stash,rows[i],1,cols+j,vals+i*nc+j,(PetscBool)(imode==ADD_VALUES));
137: }
138: }
139: }
140: return(0);
141: }
143: static PetscErrorCode MatMultXXXYYY_ScaLAPACK(Mat A,PetscBool transpose,PetscScalar beta,const PetscScalar *x,PetscScalar *y)
144: {
146: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
147: PetscScalar *x2d,*y2d,alpha=1.0;
148: const PetscInt *ranges;
149: PetscBLASInt xdesc[9],ydesc[9],x2desc[9],y2desc[9],mb,nb,lszx,lszy,zero=0,one=1,xlld,ylld,info;
152: if (transpose) {
154: /* create ScaLAPACK descriptors for vectors (1d block distribution) */
155: PetscLayoutGetRanges(A->rmap,&ranges);
156: PetscBLASIntCast(ranges[1],&mb); /* x block size */
157: xlld = PetscMax(1,A->rmap->n);
158: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(xdesc,&a->M,&one,&mb,&one,&zero,&zero,&a->grid->ictxcol,&xlld,&info));
160: PetscLayoutGetRanges(A->cmap,&ranges);
161: PetscBLASIntCast(ranges[1],&nb); /* y block size */
162: ylld = 1;
163: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ydesc,&one,&a->N,&one,&nb,&zero,&zero,&a->grid->ictxrow,&ylld,&info));
166: /* allocate 2d vectors */
167: lszx = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
168: lszy = SCALAPACKnumroc_(&a->N,&a->nb,&a->grid->mycol,&a->csrc,&a->grid->npcol);
169: PetscMalloc2(lszx,&x2d,lszy,&y2d);
170: xlld = PetscMax(1,lszx);
172: /* create ScaLAPACK descriptors for vectors (2d block distribution) */
173: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(x2desc,&a->M,&one,&a->mb,&one,&zero,&zero,&a->grid->ictxt,&xlld,&info));
175: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(y2desc,&one,&a->N,&one,&a->nb,&zero,&zero,&a->grid->ictxt,&ylld,&info));
178: /* redistribute x as a column of a 2d matrix */
179: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,(PetscScalar*)x,&one,&one,xdesc,x2d,&one,&one,x2desc,&a->grid->ictxcol));
181: /* redistribute y as a row of a 2d matrix */
182: if (beta!=0.0) PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&one,&a->N,y,&one,&one,ydesc,y2d,&one,&one,y2desc,&a->grid->ictxrow));
184: /* call PBLAS subroutine */
185: PetscStackCallBLAS("PBLASgemv",PBLASgemv_("T",&a->M,&a->N,&alpha,a->loc,&one,&one,a->desc,x2d,&one,&one,x2desc,&one,&beta,y2d,&one,&one,y2desc,&one));
187: /* redistribute y from a row of a 2d matrix */
188: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&one,&a->N,y2d,&one,&one,y2desc,y,&one,&one,ydesc,&a->grid->ictxrow));
190: } else { /* non-transpose */
192: /* create ScaLAPACK descriptors for vectors (1d block distribution) */
193: PetscLayoutGetRanges(A->cmap,&ranges);
194: PetscBLASIntCast(ranges[1],&nb); /* x block size */
195: xlld = 1;
196: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(xdesc,&one,&a->N,&one,&nb,&zero,&zero,&a->grid->ictxrow,&xlld,&info));
198: PetscLayoutGetRanges(A->rmap,&ranges);
199: PetscBLASIntCast(ranges[1],&mb); /* y block size */
200: ylld = PetscMax(1,A->rmap->n);
201: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ydesc,&a->M,&one,&mb,&one,&zero,&zero,&a->grid->ictxcol,&ylld,&info));
204: /* allocate 2d vectors */
205: lszy = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
206: lszx = SCALAPACKnumroc_(&a->N,&a->nb,&a->grid->mycol,&a->csrc,&a->grid->npcol);
207: PetscMalloc2(lszx,&x2d,lszy,&y2d);
208: ylld = PetscMax(1,lszy);
210: /* create ScaLAPACK descriptors for vectors (2d block distribution) */
211: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(x2desc,&one,&a->N,&one,&a->nb,&zero,&zero,&a->grid->ictxt,&xlld,&info));
213: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(y2desc,&a->M,&one,&a->mb,&one,&zero,&zero,&a->grid->ictxt,&ylld,&info));
216: /* redistribute x as a row of a 2d matrix */
217: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&one,&a->N,(PetscScalar*)x,&one,&one,xdesc,x2d,&one,&one,x2desc,&a->grid->ictxrow));
219: /* redistribute y as a column of a 2d matrix */
220: if (beta!=0.0) PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,y,&one,&one,ydesc,y2d,&one,&one,y2desc,&a->grid->ictxcol));
222: /* call PBLAS subroutine */
223: PetscStackCallBLAS("PBLASgemv",PBLASgemv_("N",&a->M,&a->N,&alpha,a->loc,&one,&one,a->desc,x2d,&one,&one,x2desc,&one,&beta,y2d,&one,&one,y2desc,&one));
225: /* redistribute y from a column of a 2d matrix */
226: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,y2d,&one,&one,y2desc,y,&one,&one,ydesc,&a->grid->ictxcol));
228: }
229: PetscFree2(x2d,y2d);
230: return(0);
231: }
233: static PetscErrorCode MatMult_ScaLAPACK(Mat A,Vec x,Vec y)
234: {
235: PetscErrorCode ierr;
236: const PetscScalar *xarray;
237: PetscScalar *yarray;
240: VecGetArrayRead(x,&xarray);
241: VecGetArray(y,&yarray);
242: MatMultXXXYYY_ScaLAPACK(A,PETSC_FALSE,0.0,xarray,yarray);
243: VecRestoreArrayRead(x,&xarray);
244: VecRestoreArray(y,&yarray);
245: return(0);
246: }
248: static PetscErrorCode MatMultTranspose_ScaLAPACK(Mat A,Vec x,Vec y)
249: {
250: PetscErrorCode ierr;
251: const PetscScalar *xarray;
252: PetscScalar *yarray;
255: VecGetArrayRead(x,&xarray);
256: VecGetArray(y,&yarray);
257: MatMultXXXYYY_ScaLAPACK(A,PETSC_TRUE,0.0,xarray,yarray);
258: VecRestoreArrayRead(x,&xarray);
259: VecRestoreArray(y,&yarray);
260: return(0);
261: }
263: static PetscErrorCode MatMultAdd_ScaLAPACK(Mat A,Vec x,Vec y,Vec z)
264: {
265: PetscErrorCode ierr;
266: const PetscScalar *xarray;
267: PetscScalar *zarray;
270: if (y != z) { VecCopy(y,z); }
271: VecGetArrayRead(x,&xarray);
272: VecGetArray(z,&zarray);
273: MatMultXXXYYY_ScaLAPACK(A,PETSC_FALSE,1.0,xarray,zarray);
274: VecRestoreArrayRead(x,&xarray);
275: VecRestoreArray(z,&zarray);
276: return(0);
277: }
279: static PetscErrorCode MatMultTransposeAdd_ScaLAPACK(Mat A,Vec x,Vec y,Vec z)
280: {
281: PetscErrorCode ierr;
282: const PetscScalar *xarray;
283: PetscScalar *zarray;
286: if (y != z) { VecCopy(y,z); }
287: VecGetArrayRead(x,&xarray);
288: VecGetArray(z,&zarray);
289: MatMultXXXYYY_ScaLAPACK(A,PETSC_TRUE,1.0,xarray,zarray);
290: VecRestoreArrayRead(x,&xarray);
291: VecRestoreArray(z,&zarray);
292: return(0);
293: }
295: PetscErrorCode MatMatMultNumeric_ScaLAPACK(Mat A,Mat B,Mat C)
296: {
297: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
298: Mat_ScaLAPACK *b = (Mat_ScaLAPACK*)B->data;
299: Mat_ScaLAPACK *c = (Mat_ScaLAPACK*)C->data;
300: PetscScalar sone=1.0,zero=0.0;
301: PetscBLASInt one=1;
304: PetscStackCallBLAS("PBLASgemm",PBLASgemm_("N","N",&a->M,&b->N,&a->N,&sone,a->loc,&one,&one,a->desc,b->loc,&one,&one,b->desc,&zero,c->loc,&one,&one,c->desc));
305: C->assembled = PETSC_TRUE;
306: return(0);
307: }
309: PetscErrorCode MatMatMultSymbolic_ScaLAPACK(Mat A,Mat B,PetscReal fill,Mat C)
310: {
314: MatSetSizes(C,A->rmap->n,B->cmap->n,PETSC_DECIDE,PETSC_DECIDE);
315: MatSetType(C,MATSCALAPACK);
316: MatSetUp(C);
317: C->ops->matmultnumeric = MatMatMultNumeric_ScaLAPACK;
318: return(0);
319: }
321: static PetscErrorCode MatMatTransposeMultNumeric_ScaLAPACK(Mat A,Mat B,Mat C)
322: {
323: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
324: Mat_ScaLAPACK *b = (Mat_ScaLAPACK*)B->data;
325: Mat_ScaLAPACK *c = (Mat_ScaLAPACK*)C->data;
326: PetscScalar sone=1.0,zero=0.0;
327: PetscBLASInt one=1;
330: PetscStackCallBLAS("PBLASgemm",PBLASgemm_("N","T",&a->M,&b->M,&a->N,&sone,a->loc,&one,&one,a->desc,b->loc,&one,&one,b->desc,&zero,c->loc,&one,&one,c->desc));
331: C->assembled = PETSC_TRUE;
332: return(0);
333: }
335: static PetscErrorCode MatMatTransposeMultSymbolic_ScaLAPACK(Mat A,Mat B,PetscReal fill,Mat C)
336: {
340: MatSetSizes(C,A->rmap->n,B->rmap->n,PETSC_DECIDE,PETSC_DECIDE);
341: MatSetType(C,MATSCALAPACK);
342: MatSetUp(C);
343: return(0);
344: }
346: /* --------------------------------------- */
347: static PetscErrorCode MatProductSetFromOptions_ScaLAPACK_AB(Mat C)
348: {
350: C->ops->matmultsymbolic = MatMatMultSymbolic_ScaLAPACK;
351: C->ops->productsymbolic = MatProductSymbolic_AB;
352: return(0);
353: }
355: static PetscErrorCode MatProductSetFromOptions_ScaLAPACK_ABt(Mat C)
356: {
358: C->ops->mattransposemultsymbolic = MatMatTransposeMultSymbolic_ScaLAPACK;
359: C->ops->productsymbolic = MatProductSymbolic_ABt;
360: return(0);
361: }
363: PETSC_INTERN PetscErrorCode MatProductSetFromOptions_ScaLAPACK(Mat C)
364: {
366: Mat_Product *product = C->product;
369: switch (product->type) {
370: case MATPRODUCT_AB:
371: MatProductSetFromOptions_ScaLAPACK_AB(C);
372: break;
373: case MATPRODUCT_ABt:
374: MatProductSetFromOptions_ScaLAPACK_ABt(C);
375: break;
376: default: SETERRQ1(PetscObjectComm((PetscObject)C),PETSC_ERR_SUP,"MatProduct type %s is not supported for ScaLAPACK and ScaLAPACK matrices",MatProductTypes[product->type]);
377: }
378: return(0);
379: }
380: /* --------------------------------------- */
382: static PetscErrorCode MatGetDiagonal_ScaLAPACK(Mat A,Vec D)
383: {
384: PetscErrorCode ierr;
385: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
386: PetscScalar *darray,*d2d,v;
387: const PetscInt *ranges;
388: PetscBLASInt j,ddesc[9],d2desc[9],mb,nb,lszd,zero=0,one=1,dlld,info;
391: VecGetArray(D,&darray);
393: if (A->rmap->N<=A->cmap->N) { /* row version */
395: /* create ScaLAPACK descriptor for vector (1d block distribution) */
396: PetscLayoutGetRanges(A->rmap,&ranges);
397: PetscBLASIntCast(ranges[1],&mb); /* D block size */
398: dlld = PetscMax(1,A->rmap->n);
399: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ddesc,&a->M,&one,&mb,&one,&zero,&zero,&a->grid->ictxcol,&dlld,&info));
402: /* allocate 2d vector */
403: lszd = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
404: PetscCalloc1(lszd,&d2d);
405: dlld = PetscMax(1,lszd);
407: /* create ScaLAPACK descriptor for vector (2d block distribution) */
408: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(d2desc,&a->M,&one,&a->mb,&one,&zero,&zero,&a->grid->ictxt,&dlld,&info));
411: /* collect diagonal */
412: for (j=1;j<=a->M;j++) {
413: PetscStackCallBLAS("SCALAPACKelget",SCALAPACKelget_("R"," ",&v,a->loc,&j,&j,a->desc));
414: PetscStackCallBLAS("SCALAPACKelset",SCALAPACKelset_(d2d,&j,&one,d2desc,&v));
415: }
417: /* redistribute d from a column of a 2d matrix */
418: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,d2d,&one,&one,d2desc,darray,&one,&one,ddesc,&a->grid->ictxcol));
419: PetscFree(d2d);
421: } else { /* column version */
423: /* create ScaLAPACK descriptor for vector (1d block distribution) */
424: PetscLayoutGetRanges(A->cmap,&ranges);
425: PetscBLASIntCast(ranges[1],&nb); /* D block size */
426: dlld = 1;
427: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ddesc,&one,&a->N,&one,&nb,&zero,&zero,&a->grid->ictxrow,&dlld,&info));
430: /* allocate 2d vector */
431: lszd = SCALAPACKnumroc_(&a->N,&a->nb,&a->grid->mycol,&a->csrc,&a->grid->npcol);
432: PetscCalloc1(lszd,&d2d);
434: /* create ScaLAPACK descriptor for vector (2d block distribution) */
435: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(d2desc,&one,&a->N,&one,&a->nb,&zero,&zero,&a->grid->ictxt,&dlld,&info));
438: /* collect diagonal */
439: for (j=1;j<=a->N;j++) {
440: PetscStackCallBLAS("SCALAPACKelget",SCALAPACKelget_("C"," ",&v,a->loc,&j,&j,a->desc));
441: PetscStackCallBLAS("SCALAPACKelset",SCALAPACKelset_(d2d,&one,&j,d2desc,&v));
442: }
444: /* redistribute d from a row of a 2d matrix */
445: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&one,&a->N,d2d,&one,&one,d2desc,darray,&one,&one,ddesc,&a->grid->ictxrow));
446: PetscFree(d2d);
447: }
449: VecRestoreArray(D,&darray);
450: VecAssemblyBegin(D);
451: VecAssemblyEnd(D);
452: return(0);
453: }
455: static PetscErrorCode MatDiagonalScale_ScaLAPACK(Mat A,Vec L,Vec R)
456: {
457: PetscErrorCode ierr;
458: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
459: const PetscScalar *d;
460: const PetscInt *ranges;
461: PetscScalar *d2d;
462: PetscBLASInt i,j,ddesc[9],d2desc[9],mb,nb,lszd,zero=0,one=1,dlld,info;
465: if (R) {
466: VecGetArrayRead(R,(const PetscScalar **)&d);
467: /* create ScaLAPACK descriptor for vector (1d block distribution) */
468: PetscLayoutGetRanges(A->cmap,&ranges);
469: PetscBLASIntCast(ranges[1],&nb); /* D block size */
470: dlld = 1;
471: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ddesc,&one,&a->N,&one,&nb,&zero,&zero,&a->grid->ictxrow,&dlld,&info));
474: /* allocate 2d vector */
475: lszd = SCALAPACKnumroc_(&a->N,&a->nb,&a->grid->mycol,&a->csrc,&a->grid->npcol);
476: PetscCalloc1(lszd,&d2d);
478: /* create ScaLAPACK descriptor for vector (2d block distribution) */
479: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(d2desc,&one,&a->N,&one,&a->nb,&zero,&zero,&a->grid->ictxt,&dlld,&info));
482: /* redistribute d to a row of a 2d matrix */
483: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&one,&a->N,(PetscScalar*)d,&one,&one,ddesc,d2d,&one,&one,d2desc,&a->grid->ictxrow));
485: /* broadcast along process columns */
486: if (!a->grid->myrow) Cdgebs2d(a->grid->ictxt,"C"," ",1,lszd,d2d,dlld);
487: else Cdgebr2d(a->grid->ictxt,"C"," ",1,lszd,d2d,dlld,0,a->grid->mycol);
489: /* local scaling */
490: for (j=0;j<a->locc;j++) for (i=0;i<a->locr;i++) a->loc[i+j*a->lld] *= d2d[j];
492: PetscFree(d2d);
493: VecRestoreArrayRead(R,(const PetscScalar **)&d);
494: }
495: if (L) {
496: VecGetArrayRead(L,(const PetscScalar **)&d);
497: /* create ScaLAPACK descriptor for vector (1d block distribution) */
498: PetscLayoutGetRanges(A->rmap,&ranges);
499: PetscBLASIntCast(ranges[1],&mb); /* D block size */
500: dlld = PetscMax(1,A->rmap->n);
501: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(ddesc,&a->M,&one,&mb,&one,&zero,&zero,&a->grid->ictxcol,&dlld,&info));
504: /* allocate 2d vector */
505: lszd = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
506: PetscCalloc1(lszd,&d2d);
507: dlld = PetscMax(1,lszd);
509: /* create ScaLAPACK descriptor for vector (2d block distribution) */
510: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(d2desc,&a->M,&one,&a->mb,&one,&zero,&zero,&a->grid->ictxt,&dlld,&info));
513: /* redistribute d to a column of a 2d matrix */
514: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,(PetscScalar*)d,&one,&one,ddesc,d2d,&one,&one,d2desc,&a->grid->ictxcol));
516: /* broadcast along process rows */
517: if (!a->grid->mycol) Cdgebs2d(a->grid->ictxt,"R"," ",lszd,1,d2d,dlld);
518: else Cdgebr2d(a->grid->ictxt,"R"," ",lszd,1,d2d,dlld,a->grid->myrow,0);
520: /* local scaling */
521: for (i=0;i<a->locr;i++) for (j=0;j<a->locc;j++) a->loc[i+j*a->lld] *= d2d[i];
523: PetscFree(d2d);
524: VecRestoreArrayRead(L,(const PetscScalar **)&d);
525: }
526: return(0);
527: }
529: static PetscErrorCode MatMissingDiagonal_ScaLAPACK(Mat A,PetscBool *missing,PetscInt *d)
530: {
532: *missing = PETSC_FALSE;
533: return(0);
534: }
536: static PetscErrorCode MatScale_ScaLAPACK(Mat X,PetscScalar a)
537: {
538: Mat_ScaLAPACK *x = (Mat_ScaLAPACK*)X->data;
539: PetscBLASInt n,one=1;
542: n = x->lld*x->locc;
543: PetscStackCallBLAS("BLASscal",BLASscal_(&n,&a,x->loc,&one));
544: return(0);
545: }
547: static PetscErrorCode MatShift_ScaLAPACK(Mat X,PetscScalar alpha)
548: {
549: Mat_ScaLAPACK *x = (Mat_ScaLAPACK*)X->data;
550: PetscBLASInt i,n;
551: PetscScalar v;
554: n = PetscMin(x->M,x->N);
555: for (i=1;i<=n;i++) {
556: PetscStackCallBLAS("SCALAPACKelget",SCALAPACKelget_("-"," ",&v,x->loc,&i,&i,x->desc));
557: v += alpha;
558: PetscStackCallBLAS("SCALAPACKelset",SCALAPACKelset_(x->loc,&i,&i,x->desc,&v));
559: }
560: return(0);
561: }
563: static PetscErrorCode MatAXPY_ScaLAPACK(Mat Y,PetscScalar alpha,Mat X,MatStructure str)
564: {
566: Mat_ScaLAPACK *x = (Mat_ScaLAPACK*)X->data;
567: Mat_ScaLAPACK *y = (Mat_ScaLAPACK*)Y->data;
568: PetscBLASInt one=1;
569: PetscScalar beta=1.0;
572: MatScaLAPACKCheckDistribution(Y,1,X,3);
573: PetscStackCallBLAS("SCALAPACKmatadd",SCALAPACKmatadd_(&x->M,&x->N,&alpha,x->loc,&one,&one,x->desc,&beta,y->loc,&one,&one,y->desc));
574: PetscObjectStateIncrease((PetscObject)Y);
575: return(0);
576: }
578: static PetscErrorCode MatCopy_ScaLAPACK(Mat A,Mat B,MatStructure str)
579: {
581: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
582: Mat_ScaLAPACK *b = (Mat_ScaLAPACK*)B->data;
585: PetscArraycpy(b->loc,a->loc,a->lld*a->locc);
586: PetscObjectStateIncrease((PetscObject)B);
587: return(0);
588: }
590: static PetscErrorCode MatDuplicate_ScaLAPACK(Mat A,MatDuplicateOption op,Mat *B)
591: {
592: Mat Bs;
593: MPI_Comm comm;
594: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data,*b;
598: PetscObjectGetComm((PetscObject)A,&comm);
599: MatCreate(comm,&Bs);
600: MatSetSizes(Bs,A->rmap->n,A->cmap->n,PETSC_DECIDE,PETSC_DECIDE);
601: MatSetType(Bs,MATSCALAPACK);
602: b = (Mat_ScaLAPACK*)Bs->data;
603: b->M = a->M;
604: b->N = a->N;
605: b->mb = a->mb;
606: b->nb = a->nb;
607: b->rsrc = a->rsrc;
608: b->csrc = a->csrc;
609: MatSetUp(Bs);
610: *B = Bs;
611: if (op == MAT_COPY_VALUES) {
612: PetscArraycpy(b->loc,a->loc,a->lld*a->locc);
613: }
614: Bs->assembled = PETSC_TRUE;
615: return(0);
616: }
618: static PetscErrorCode MatTranspose_ScaLAPACK(Mat A,MatReuse reuse,Mat *B)
619: {
621: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data, *b;
622: Mat Bs = *B;
623: PetscBLASInt one=1;
624: PetscScalar sone=1.0,zero=0.0;
625: #if defined(PETSC_USE_COMPLEX)
626: PetscInt i,j;
627: #endif
630: if (reuse == MAT_INITIAL_MATRIX) {
631: MatCreateScaLAPACK(PetscObjectComm((PetscObject)A),a->nb,a->mb,a->N,a->M,a->csrc,a->rsrc,&Bs);
632: *B = Bs;
633: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Only MAT_INITIAL_MATRIX supported");
634: b = (Mat_ScaLAPACK*)Bs->data;
635: PetscStackCallBLAS("PBLAStran",PBLAStran_(&a->N,&a->M,&sone,a->loc,&one,&one,a->desc,&zero,b->loc,&one,&one,b->desc));
636: #if defined(PETSC_USE_COMPLEX)
637: /* undo conjugation */
638: for (i=0;i<b->locr;i++) for (j=0;j<b->locc;j++) b->loc[i+j*b->lld] = PetscConj(b->loc[i+j*b->lld]);
639: #endif
640: Bs->assembled = PETSC_TRUE;
641: return(0);
642: }
644: static PetscErrorCode MatConjugate_ScaLAPACK(Mat A)
645: {
646: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
647: PetscInt i,j;
650: for (i=0;i<a->locr;i++) for (j=0;j<a->locc;j++) a->loc[i+j*a->lld] = PetscConj(a->loc[i+j*a->lld]);
651: return(0);
652: }
654: static PetscErrorCode MatHermitianTranspose_ScaLAPACK(Mat A,MatReuse reuse,Mat *B)
655: {
657: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data, *b;
658: Mat Bs = *B;
659: PetscBLASInt one=1;
660: PetscScalar sone=1.0,zero=0.0;
663: if (reuse == MAT_INITIAL_MATRIX) {
664: MatCreateScaLAPACK(PetscObjectComm((PetscObject)A),a->nb,a->mb,a->N,a->M,a->csrc,a->rsrc,&Bs);
665: *B = Bs;
666: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Only MAT_INITIAL_MATRIX supported");
667: b = (Mat_ScaLAPACK*)Bs->data;
668: PetscStackCallBLAS("PBLAStran",PBLAStran_(&a->N,&a->M,&sone,a->loc,&one,&one,a->desc,&zero,b->loc,&one,&one,b->desc));
669: Bs->assembled = PETSC_TRUE;
670: return(0);
671: }
673: static PetscErrorCode MatSolve_ScaLAPACK(Mat A,Vec B,Vec X)
674: {
676: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
677: PetscScalar *x,*x2d;
678: const PetscInt *ranges;
679: PetscBLASInt xdesc[9],x2desc[9],mb,lszx,zero=0,one=1,xlld,nrhs=1,info;
682: VecCopy(B,X);
683: VecGetArray(X,&x);
685: /* create ScaLAPACK descriptor for a vector (1d block distribution) */
686: PetscLayoutGetRanges(A->rmap,&ranges);
687: PetscBLASIntCast(ranges[1],&mb); /* x block size */
688: xlld = PetscMax(1,A->rmap->n);
689: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(xdesc,&a->M,&one,&mb,&one,&zero,&zero,&a->grid->ictxcol,&xlld,&info));
692: /* allocate 2d vector */
693: lszx = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
694: PetscMalloc1(lszx,&x2d);
695: xlld = PetscMax(1,lszx);
697: /* create ScaLAPACK descriptor for a vector (2d block distribution) */
698: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(x2desc,&a->M,&one,&a->mb,&one,&zero,&zero,&a->grid->ictxt,&xlld,&info));
701: /* redistribute x as a column of a 2d matrix */
702: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,x,&one,&one,xdesc,x2d,&one,&one,x2desc,&a->grid->ictxcol));
704: /* call ScaLAPACK subroutine */
705: switch (A->factortype) {
706: case MAT_FACTOR_LU:
707: PetscStackCallBLAS("SCALAPACKgetrs",SCALAPACKgetrs_("N",&a->M,&nrhs,a->loc,&one,&one,a->desc,a->pivots,x2d,&one,&one,x2desc,&info));
709: break;
710: case MAT_FACTOR_CHOLESKY:
711: PetscStackCallBLAS("SCALAPACKpotrs",SCALAPACKpotrs_("L",&a->M,&nrhs,a->loc,&one,&one,a->desc,x2d,&one,&one,x2desc,&info));
713: break;
714: default:
715: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Unfactored Matrix or Unsupported MatFactorType");
716: }
718: /* redistribute x from a column of a 2d matrix */
719: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&one,x2d,&one,&one,x2desc,x,&one,&one,xdesc,&a->grid->ictxcol));
721: PetscFree(x2d);
722: VecRestoreArray(X,&x);
723: return(0);
724: }
726: static PetscErrorCode MatSolveAdd_ScaLAPACK(Mat A,Vec B,Vec Y,Vec X)
727: {
731: MatSolve_ScaLAPACK(A,B,X);
732: VecAXPY(X,1,Y);
733: return(0);
734: }
736: static PetscErrorCode MatMatSolve_ScaLAPACK(Mat A,Mat B,Mat X)
737: {
739: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data,*b,*x;
740: PetscBool flg1,flg2;
741: PetscBLASInt one=1,info;
744: PetscObjectTypeCompare((PetscObject)B,MATSCALAPACK,&flg1);
745: PetscObjectTypeCompare((PetscObject)X,MATSCALAPACK,&flg2);
746: if (!(flg1 && flg2)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Both B and X must be of type MATSCALAPACK");
747: MatScaLAPACKCheckDistribution(B,1,X,2);
748: b = (Mat_ScaLAPACK*)B->data;
749: x = (Mat_ScaLAPACK*)X->data;
750: PetscArraycpy(x->loc,b->loc,b->lld*b->locc);
752: switch (A->factortype) {
753: case MAT_FACTOR_LU:
754: PetscStackCallBLAS("SCALAPACKgetrs",SCALAPACKgetrs_("N",&a->M,&x->N,a->loc,&one,&one,a->desc,a->pivots,x->loc,&one,&one,x->desc,&info));
756: break;
757: case MAT_FACTOR_CHOLESKY:
758: PetscStackCallBLAS("SCALAPACKpotrs",SCALAPACKpotrs_("L",&a->M,&x->N,a->loc,&one,&one,a->desc,x->loc,&one,&one,x->desc,&info));
760: break;
761: default:
762: SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Unfactored Matrix or Unsupported MatFactorType");
763: }
764: return(0);
765: }
767: static PetscErrorCode MatLUFactor_ScaLAPACK(Mat A,IS row,IS col,const MatFactorInfo *factorinfo)
768: {
770: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
771: PetscBLASInt one=1,info;
774: if (!a->pivots) {
775: PetscMalloc1(a->locr+a->mb,&a->pivots);
776: PetscLogObjectMemory((PetscObject)A,a->locr*sizeof(PetscBLASInt));
777: }
778: PetscStackCallBLAS("SCALAPACKgetrf",SCALAPACKgetrf_(&a->M,&a->N,a->loc,&one,&one,a->desc,a->pivots,&info));
780: A->factortype = MAT_FACTOR_LU;
781: A->assembled = PETSC_TRUE;
783: PetscFree(A->solvertype);
784: PetscStrallocpy(MATSOLVERSCALAPACK,&A->solvertype);
785: return(0);
786: }
788: static PetscErrorCode MatLUFactorNumeric_ScaLAPACK(Mat F,Mat A,const MatFactorInfo *info)
789: {
793: MatCopy(A,F,SAME_NONZERO_PATTERN);
794: MatLUFactor_ScaLAPACK(F,0,0,info);
795: return(0);
796: }
798: static PetscErrorCode MatLUFactorSymbolic_ScaLAPACK(Mat F,Mat A,IS r,IS c,const MatFactorInfo *info)
799: {
801: /* F is created and allocated by MatGetFactor_scalapack_petsc(), skip this routine. */
802: return(0);
803: }
805: static PetscErrorCode MatCholeskyFactor_ScaLAPACK(Mat A,IS perm,const MatFactorInfo *factorinfo)
806: {
808: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
809: PetscBLASInt one=1,info;
812: PetscStackCallBLAS("SCALAPACKpotrf",SCALAPACKpotrf_("L",&a->M,a->loc,&one,&one,a->desc,&info));
814: A->factortype = MAT_FACTOR_CHOLESKY;
815: A->assembled = PETSC_TRUE;
817: PetscFree(A->solvertype);
818: PetscStrallocpy(MATSOLVERSCALAPACK,&A->solvertype);
819: return(0);
820: }
822: static PetscErrorCode MatCholeskyFactorNumeric_ScaLAPACK(Mat F,Mat A,const MatFactorInfo *info)
823: {
827: MatCopy(A,F,SAME_NONZERO_PATTERN);
828: MatCholeskyFactor_ScaLAPACK(F,0,info);
829: return(0);
830: }
832: static PetscErrorCode MatCholeskyFactorSymbolic_ScaLAPACK(Mat F,Mat A,IS perm,const MatFactorInfo *info)
833: {
835: /* F is created and allocated by MatGetFactor_scalapack_petsc(), skip this routine. */
836: return(0);
837: }
839: PetscErrorCode MatFactorGetSolverType_scalapack_scalapack(Mat A,MatSolverType *type)
840: {
842: *type = MATSOLVERSCALAPACK;
843: return(0);
844: }
846: static PetscErrorCode MatGetFactor_scalapack_scalapack(Mat A,MatFactorType ftype,Mat *F)
847: {
848: Mat B;
849: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
853: /* Create the factorization matrix */
854: MatCreateScaLAPACK(PetscObjectComm((PetscObject)A),a->mb,a->nb,a->M,a->N,a->rsrc,a->csrc,&B);
855: B->factortype = ftype;
856: PetscFree(B->solvertype);
857: PetscStrallocpy(MATSOLVERSCALAPACK,&B->solvertype);
859: PetscObjectComposeFunction((PetscObject)B,"MatFactorGetSolverType_C",MatFactorGetSolverType_scalapack_scalapack);
860: *F = B;
861: return(0);
862: }
864: PETSC_EXTERN PetscErrorCode MatSolverTypeRegister_ScaLAPACK(void)
865: {
869: MatSolverTypeRegister(MATSOLVERSCALAPACK,MATSCALAPACK,MAT_FACTOR_LU,MatGetFactor_scalapack_scalapack);
870: MatSolverTypeRegister(MATSOLVERSCALAPACK,MATSCALAPACK,MAT_FACTOR_CHOLESKY,MatGetFactor_scalapack_scalapack);
871: return(0);
872: }
874: static PetscErrorCode MatNorm_ScaLAPACK(Mat A,NormType type,PetscReal *nrm)
875: {
877: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
878: PetscBLASInt one=1,lwork=0;
879: const char *ntype;
880: PetscScalar *work=NULL,dummy;
883: switch (type){
884: case NORM_1:
885: ntype = "1";
886: lwork = PetscMax(a->locr,a->locc);
887: break;
888: case NORM_FROBENIUS:
889: ntype = "F";
890: work = &dummy;
891: break;
892: case NORM_INFINITY:
893: ntype = "I";
894: lwork = PetscMax(a->locr,a->locc);
895: break;
896: default:
897: SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Unsupported norm type");
898: }
899: if (lwork) { PetscMalloc1(lwork,&work); }
900: *nrm = SCALAPACKlange_(ntype,&a->M,&a->N,a->loc,&one,&one,a->desc,work);
901: if (lwork) { PetscFree(work); }
902: return(0);
903: }
905: static PetscErrorCode MatZeroEntries_ScaLAPACK(Mat A)
906: {
907: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
911: PetscArrayzero(a->loc,a->lld*a->locc);
912: return(0);
913: }
915: static PetscErrorCode MatGetOwnershipIS_ScaLAPACK(Mat A,IS *rows,IS *cols)
916: {
917: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
919: PetscInt i,n,nb,isrc,nproc,iproc,*idx;
922: if (rows) {
923: n = a->locr;
924: nb = a->mb;
925: isrc = a->rsrc;
926: nproc = a->grid->nprow;
927: iproc = a->grid->myrow;
928: PetscMalloc1(n,&idx);
929: for (i=0;i<n;i++) idx[i] = nproc*nb*(i/nb) + i%nb + ((nproc+iproc-isrc)%nproc)*nb;
930: ISCreateGeneral(PETSC_COMM_SELF,n,idx,PETSC_OWN_POINTER,rows);
931: }
932: if (cols) {
933: n = a->locc;
934: nb = a->nb;
935: isrc = a->csrc;
936: nproc = a->grid->npcol;
937: iproc = a->grid->mycol;
938: PetscMalloc1(n,&idx);
939: for (i=0;i<n;i++) idx[i] = nproc*nb*(i/nb) + i%nb + ((nproc+iproc-isrc)%nproc)*nb;
940: ISCreateGeneral(PETSC_COMM_SELF,n,idx,PETSC_OWN_POINTER,cols);
941: }
942: return(0);
943: }
945: static PetscErrorCode MatConvert_ScaLAPACK_Dense(Mat A,MatType newtype,MatReuse reuse,Mat *B)
946: {
948: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
949: Mat Bmpi;
950: MPI_Comm comm;
951: PetscInt i,M=A->rmap->N,N=A->cmap->N,m,n,rstart,rend,nz;
952: const PetscInt *ranges,*branges,*cwork;
953: const PetscScalar *vwork;
954: PetscBLASInt bdesc[9],bmb,zero=0,one=1,lld,info;
955: PetscScalar *barray;
956: PetscBool differ=PETSC_FALSE;
957: PetscMPIInt size;
960: PetscObjectGetComm((PetscObject)A,&comm);
961: PetscLayoutGetRanges(A->rmap,&ranges);
963: if (reuse == MAT_REUSE_MATRIX) { /* check if local sizes differ in A and B */
964: MPI_Comm_size(comm,&size);
965: PetscLayoutGetRanges((*B)->rmap,&branges);
966: for (i=0;i<size;i++) if (ranges[i+1]!=branges[i+1]) { differ=PETSC_TRUE; break; }
967: }
969: if (reuse == MAT_REUSE_MATRIX && differ) { /* special case, use auxiliary dense matrix */
970: MatCreate(comm,&Bmpi);
971: m = PETSC_DECIDE;
972: PetscSplitOwnershipEqual(comm,&m,&M);
973: n = PETSC_DECIDE;
974: PetscSplitOwnershipEqual(comm,&n,&N);
975: MatSetSizes(Bmpi,m,n,M,N);
976: MatSetType(Bmpi,MATDENSE);
977: MatSetUp(Bmpi);
979: /* create ScaLAPACK descriptor for B (1d block distribution) */
980: PetscBLASIntCast(ranges[1],&bmb); /* row block size */
981: lld = PetscMax(A->rmap->n,1); /* local leading dimension */
982: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(bdesc,&a->M,&a->N,&bmb,&a->N,&zero,&zero,&a->grid->ictxcol,&lld,&info));
985: /* redistribute matrix */
986: MatDenseGetArray(Bmpi,&barray);
987: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&a->N,a->loc,&one,&one,a->desc,barray,&one,&one,bdesc,&a->grid->ictxcol));
988: MatDenseRestoreArray(Bmpi,&barray);
989: MatAssemblyBegin(Bmpi,MAT_FINAL_ASSEMBLY);
990: MatAssemblyEnd(Bmpi,MAT_FINAL_ASSEMBLY);
992: /* transfer rows of auxiliary matrix to the final matrix B */
993: MatGetOwnershipRange(Bmpi,&rstart,&rend);
994: for (i=rstart;i<rend;i++) {
995: MatGetRow(Bmpi,i,&nz,&cwork,&vwork);
996: MatSetValues(*B,1,&i,nz,cwork,vwork,INSERT_VALUES);
997: MatRestoreRow(Bmpi,i,&nz,&cwork,&vwork);
998: }
999: MatAssemblyBegin(*B,MAT_FINAL_ASSEMBLY);
1000: MatAssemblyEnd(*B,MAT_FINAL_ASSEMBLY);
1001: MatDestroy(&Bmpi);
1003: } else { /* normal cases */
1005: if (reuse == MAT_REUSE_MATRIX) Bmpi = *B;
1006: else {
1007: MatCreate(comm,&Bmpi);
1008: m = PETSC_DECIDE;
1009: PetscSplitOwnershipEqual(comm,&m,&M);
1010: n = PETSC_DECIDE;
1011: PetscSplitOwnershipEqual(comm,&n,&N);
1012: MatSetSizes(Bmpi,m,n,M,N);
1013: MatSetType(Bmpi,MATDENSE);
1014: MatSetUp(Bmpi);
1015: }
1017: /* create ScaLAPACK descriptor for B (1d block distribution) */
1018: PetscBLASIntCast(ranges[1],&bmb); /* row block size */
1019: lld = PetscMax(A->rmap->n,1); /* local leading dimension */
1020: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(bdesc,&a->M,&a->N,&bmb,&a->N,&zero,&zero,&a->grid->ictxcol,&lld,&info));
1023: /* redistribute matrix */
1024: MatDenseGetArray(Bmpi,&barray);
1025: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&a->M,&a->N,a->loc,&one,&one,a->desc,barray,&one,&one,bdesc,&a->grid->ictxcol));
1026: MatDenseRestoreArray(Bmpi,&barray);
1028: MatAssemblyBegin(Bmpi,MAT_FINAL_ASSEMBLY);
1029: MatAssemblyEnd(Bmpi,MAT_FINAL_ASSEMBLY);
1030: if (reuse == MAT_INPLACE_MATRIX) {
1031: MatHeaderReplace(A,&Bmpi);
1032: } else *B = Bmpi;
1033: }
1034: return(0);
1035: }
1037: PETSC_INTERN PetscErrorCode MatConvert_Dense_ScaLAPACK(Mat A,MatType newtype,MatReuse reuse,Mat *B)
1038: {
1040: Mat_ScaLAPACK *b;
1041: Mat Bmpi;
1042: MPI_Comm comm;
1043: PetscInt M=A->rmap->N,N=A->cmap->N,m,n;
1044: const PetscInt *ranges;
1045: PetscBLASInt adesc[9],amb,zero=0,one=1,lld,info;
1046: PetscScalar *aarray;
1047: PetscInt lda;
1050: PetscObjectGetComm((PetscObject)A,&comm);
1052: if (reuse == MAT_REUSE_MATRIX) Bmpi = *B;
1053: else {
1054: MatCreate(comm,&Bmpi);
1055: m = PETSC_DECIDE;
1056: PetscSplitOwnershipEqual(comm,&m,&M);
1057: n = PETSC_DECIDE;
1058: PetscSplitOwnershipEqual(comm,&n,&N);
1059: MatSetSizes(Bmpi,m,n,M,N);
1060: MatSetType(Bmpi,MATSCALAPACK);
1061: MatSetUp(Bmpi);
1062: }
1063: b = (Mat_ScaLAPACK*)Bmpi->data;
1065: /* create ScaLAPACK descriptor for A (1d block distribution) */
1066: PetscLayoutGetRanges(A->rmap,&ranges);
1067: PetscBLASIntCast(ranges[1],&amb); /* row block size */
1068: MatDenseGetLDA(A,&lda);
1069: lld = PetscMax(lda,1); /* local leading dimension */
1070: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(adesc,&b->M,&b->N,&amb,&b->N,&zero,&zero,&b->grid->ictxcol,&lld,&info));
1073: /* redistribute matrix */
1074: MatDenseGetArray(A,&aarray);
1075: PetscStackCallBLAS("SCALAPACKgemr2d",SCALAPACKgemr2d_(&b->M,&b->N,aarray,&one,&one,adesc,b->loc,&one,&one,b->desc,&b->grid->ictxcol));
1076: MatDenseRestoreArray(A,&aarray);
1078: MatAssemblyBegin(Bmpi,MAT_FINAL_ASSEMBLY);
1079: MatAssemblyEnd(Bmpi,MAT_FINAL_ASSEMBLY);
1080: if (reuse == MAT_INPLACE_MATRIX) {
1081: MatHeaderReplace(A,&Bmpi);
1082: } else *B = Bmpi;
1083: return(0);
1084: }
1086: PETSC_INTERN PetscErrorCode MatConvert_AIJ_ScaLAPACK(Mat A,MatType newtype,MatReuse reuse,Mat *newmat)
1087: {
1088: Mat mat_scal;
1089: PetscErrorCode ierr;
1090: PetscInt M=A->rmap->N,N=A->cmap->N,rstart=A->rmap->rstart,rend=A->rmap->rend,m,n,row,ncols;
1091: const PetscInt *cols;
1092: const PetscScalar *vals;
1095: if (reuse == MAT_REUSE_MATRIX) {
1096: mat_scal = *newmat;
1097: MatZeroEntries(mat_scal);
1098: } else {
1099: MatCreate(PetscObjectComm((PetscObject)A),&mat_scal);
1100: m = PETSC_DECIDE;
1101: PetscSplitOwnershipEqual(PetscObjectComm((PetscObject)A),&m,&M);
1102: n = PETSC_DECIDE;
1103: PetscSplitOwnershipEqual(PetscObjectComm((PetscObject)A),&n,&N);
1104: MatSetSizes(mat_scal,m,n,M,N);
1105: MatSetType(mat_scal,MATSCALAPACK);
1106: MatSetUp(mat_scal);
1107: }
1108: for (row=rstart;row<rend;row++) {
1109: MatGetRow(A,row,&ncols,&cols,&vals);
1110: MatSetValues(mat_scal,1,&row,ncols,cols,vals,INSERT_VALUES);
1111: MatRestoreRow(A,row,&ncols,&cols,&vals);
1112: }
1113: MatAssemblyBegin(mat_scal,MAT_FINAL_ASSEMBLY);
1114: MatAssemblyEnd(mat_scal,MAT_FINAL_ASSEMBLY);
1116: if (reuse == MAT_INPLACE_MATRIX) { MatHeaderReplace(A,&mat_scal); }
1117: else *newmat = mat_scal;
1118: return(0);
1119: }
1121: PETSC_INTERN PetscErrorCode MatConvert_SBAIJ_ScaLAPACK(Mat A, MatType newtype,MatReuse reuse,Mat *newmat)
1122: {
1123: Mat mat_scal;
1124: PetscErrorCode ierr;
1125: PetscInt M=A->rmap->N,N=A->cmap->N,m,n,row,ncols,j,rstart=A->rmap->rstart,rend=A->rmap->rend;
1126: const PetscInt *cols;
1127: const PetscScalar *vals;
1128: PetscScalar v;
1131: if (reuse == MAT_REUSE_MATRIX) {
1132: mat_scal = *newmat;
1133: MatZeroEntries(mat_scal);
1134: } else {
1135: MatCreate(PetscObjectComm((PetscObject)A),&mat_scal);
1136: m = PETSC_DECIDE;
1137: PetscSplitOwnershipEqual(PetscObjectComm((PetscObject)A),&m,&M);
1138: n = PETSC_DECIDE;
1139: PetscSplitOwnershipEqual(PetscObjectComm((PetscObject)A),&n,&N);
1140: MatSetSizes(mat_scal,m,n,M,N);
1141: MatSetType(mat_scal,MATSCALAPACK);
1142: MatSetUp(mat_scal);
1143: }
1144: MatGetRowUpperTriangular(A);
1145: for (row=rstart;row<rend;row++) {
1146: MatGetRow(A,row,&ncols,&cols,&vals);
1147: MatSetValues(mat_scal,1,&row,ncols,cols,vals,ADD_VALUES);
1148: for (j=0;j<ncols;j++) { /* lower triangular part */
1149: if (cols[j] == row) continue;
1150: v = A->hermitian ? PetscConj(vals[j]) : vals[j];
1151: MatSetValues(mat_scal,1,&cols[j],1,&row,&v,ADD_VALUES);
1152: }
1153: MatRestoreRow(A,row,&ncols,&cols,&vals);
1154: }
1155: MatRestoreRowUpperTriangular(A);
1156: MatAssemblyBegin(mat_scal,MAT_FINAL_ASSEMBLY);
1157: MatAssemblyEnd(mat_scal,MAT_FINAL_ASSEMBLY);
1159: if (reuse == MAT_INPLACE_MATRIX) { MatHeaderReplace(A,&mat_scal); }
1160: else *newmat = mat_scal;
1161: return(0);
1162: }
1164: static PetscErrorCode MatScaLAPACKSetPreallocation(Mat A)
1165: {
1166: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1168: PetscInt sz=0;
1171: PetscLayoutSetUp(A->rmap);
1172: PetscLayoutSetUp(A->cmap);
1173: if (!a->lld) a->lld = a->locr;
1175: PetscFree(a->loc);
1176: PetscIntMultError(a->lld,a->locc,&sz);
1177: PetscCalloc1(sz,&a->loc);
1178: PetscLogObjectMemory((PetscObject)A,sz*sizeof(PetscScalar));
1180: A->preallocated = PETSC_TRUE;
1181: return(0);
1182: }
1184: static PetscErrorCode MatDestroy_ScaLAPACK(Mat A)
1185: {
1186: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1187: PetscErrorCode ierr;
1188: Mat_ScaLAPACK_Grid *grid;
1189: PetscBool flg;
1190: MPI_Comm icomm;
1193: MatStashDestroy_Private(&A->stash);
1194: PetscFree(a->loc);
1195: PetscFree(a->pivots);
1196: PetscCommDuplicate(PetscObjectComm((PetscObject)A),&icomm,NULL);
1197: MPI_Comm_get_attr(icomm,Petsc_ScaLAPACK_keyval,(void**)&grid,(int*)&flg);
1198: if (--grid->grid_refct == 0) {
1199: Cblacs_gridexit(grid->ictxt);
1200: Cblacs_gridexit(grid->ictxrow);
1201: Cblacs_gridexit(grid->ictxcol);
1202: PetscFree(grid);
1203: MPI_Comm_delete_attr(icomm,Petsc_ScaLAPACK_keyval);
1204: }
1205: PetscCommDestroy(&icomm);
1206: PetscObjectComposeFunction((PetscObject)A,"MatGetOwnershipIS_C",NULL);
1207: PetscObjectComposeFunction((PetscObject)A,"MatFactorGetSolverType_C",NULL);
1208: PetscObjectComposeFunction((PetscObject)A,"MatScaLAPACKSetBlockSizes_C",NULL);
1209: PetscObjectComposeFunction((PetscObject)A,"MatScaLAPACKGetBlockSizes_C",NULL);
1210: PetscFree(A->data);
1211: return(0);
1212: }
1214: PETSC_STATIC_INLINE PetscErrorCode MatScaLAPACKCheckLayout(PetscLayout map)
1215: {
1217: const PetscInt *ranges;
1218: PetscMPIInt size;
1219: PetscInt i,n;
1222: MPI_Comm_size(map->comm,&size);
1223: if (size>2) {
1224: PetscLayoutGetRanges(map,&ranges);
1225: n = ranges[1]-ranges[0];
1226: for (i=1;i<size-1;i++) if (ranges[i+1]-ranges[i]!=n) break;
1227: if (i<size-1 && ranges[i+1]-ranges[i]!=0 && ranges[i+2]-ranges[i+1]!=0) SETERRQ(map->comm,PETSC_ERR_SUP,"MATSCALAPACK must have equal local sizes in all processes (except possibly the last one), consider using MatCreateScaLAPACK");
1228: }
1229: return(0);
1230: }
1232: PetscErrorCode MatSetUp_ScaLAPACK(Mat A)
1233: {
1234: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1236: PetscBLASInt info=0;
1239: PetscLayoutSetUp(A->rmap);
1240: PetscLayoutSetUp(A->cmap);
1242: /* check that the layout is as enforced by MatCreateScaLAPACK */
1243: MatScaLAPACKCheckLayout(A->rmap);
1244: MatScaLAPACKCheckLayout(A->cmap);
1246: /* compute local sizes */
1247: PetscBLASIntCast(A->rmap->N,&a->M);
1248: PetscBLASIntCast(A->cmap->N,&a->N);
1249: a->locr = SCALAPACKnumroc_(&a->M,&a->mb,&a->grid->myrow,&a->rsrc,&a->grid->nprow);
1250: a->locc = SCALAPACKnumroc_(&a->N,&a->nb,&a->grid->mycol,&a->csrc,&a->grid->npcol);
1251: a->lld = PetscMax(1,a->locr);
1253: /* allocate local array */
1254: MatScaLAPACKSetPreallocation(A);
1256: /* set up ScaLAPACK descriptor */
1257: PetscStackCallBLAS("SCALAPACKdescinit",SCALAPACKdescinit_(a->desc,&a->M,&a->N,&a->mb,&a->nb,&a->rsrc,&a->csrc,&a->grid->ictxt,&a->lld,&info));
1259: return(0);
1260: }
1262: PetscErrorCode MatAssemblyBegin_ScaLAPACK(Mat A,MatAssemblyType type)
1263: {
1265: PetscInt nstash,reallocs;
1268: if (A->nooffprocentries) return(0);
1269: MatStashScatterBegin_Private(A,&A->stash,NULL);
1270: MatStashGetInfo_Private(&A->stash,&nstash,&reallocs);
1271: PetscInfo2(A,"Stash has %D entries, uses %D mallocs.\n",nstash,reallocs);
1272: return(0);
1273: }
1275: PetscErrorCode MatAssemblyEnd_ScaLAPACK(Mat A,MatAssemblyType type)
1276: {
1278: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1279: PetscMPIInt n;
1280: PetscInt i,flg,*row,*col;
1281: PetscScalar *val;
1282: PetscBLASInt gridx,gcidx,lridx,lcidx,rsrc,csrc;
1285: if (A->nooffprocentries) return(0);
1286: while (1) {
1287: MatStashScatterGetMesg_Private(&A->stash,&n,&row,&col,&val,&flg);
1288: if (!flg) break;
1289: for (i=0;i<n;i++) {
1290: PetscBLASIntCast(row[i]+1,&gridx);
1291: PetscBLASIntCast(col[i]+1,&gcidx);
1292: PetscStackCallBLAS("SCALAPACKinfog2l",SCALAPACKinfog2l_(&gridx,&gcidx,a->desc,&a->grid->nprow,&a->grid->npcol,&a->grid->myrow,&a->grid->mycol,&lridx,&lcidx,&rsrc,&csrc));
1293: if (rsrc!=a->grid->myrow || csrc!=a->grid->mycol) SETERRQ(PetscObjectComm((PetscObject)A),1,"Something went wrong, received value does not belong to this process");
1294: switch (A->insertmode) {
1295: case INSERT_VALUES: a->loc[lridx-1+(lcidx-1)*a->lld] = val[i]; break;
1296: case ADD_VALUES: a->loc[lridx-1+(lcidx-1)*a->lld] += val[i]; break;
1297: default: SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"No support for InsertMode %d",(int)A->insertmode);
1298: }
1299: }
1300: }
1301: MatStashScatterEnd_Private(&A->stash);
1302: return(0);
1303: }
1305: PetscErrorCode MatLoad_ScaLAPACK(Mat newMat,PetscViewer viewer)
1306: {
1308: Mat Adense,As;
1309: MPI_Comm comm;
1312: PetscObjectGetComm((PetscObject)newMat,&comm);
1313: MatCreate(comm,&Adense);
1314: MatSetType(Adense,MATDENSE);
1315: MatLoad(Adense,viewer);
1316: MatConvert(Adense,MATSCALAPACK,MAT_INITIAL_MATRIX,&As);
1317: MatDestroy(&Adense);
1318: MatHeaderReplace(newMat,&As);
1319: return(0);
1320: }
1322: /* -------------------------------------------------------------------*/
1323: static struct _MatOps MatOps_Values = {
1324: MatSetValues_ScaLAPACK,
1325: 0,
1326: 0,
1327: MatMult_ScaLAPACK,
1328: /* 4*/ MatMultAdd_ScaLAPACK,
1329: MatMultTranspose_ScaLAPACK,
1330: MatMultTransposeAdd_ScaLAPACK,
1331: MatSolve_ScaLAPACK,
1332: MatSolveAdd_ScaLAPACK,
1333: 0,
1334: /*10*/ 0,
1335: MatLUFactor_ScaLAPACK,
1336: MatCholeskyFactor_ScaLAPACK,
1337: 0,
1338: MatTranspose_ScaLAPACK,
1339: /*15*/ MatGetInfo_ScaLAPACK,
1340: 0,
1341: MatGetDiagonal_ScaLAPACK,
1342: MatDiagonalScale_ScaLAPACK,
1343: MatNorm_ScaLAPACK,
1344: /*20*/ MatAssemblyBegin_ScaLAPACK,
1345: MatAssemblyEnd_ScaLAPACK,
1346: MatSetOption_ScaLAPACK,
1347: MatZeroEntries_ScaLAPACK,
1348: /*24*/ 0,
1349: MatLUFactorSymbolic_ScaLAPACK,
1350: MatLUFactorNumeric_ScaLAPACK,
1351: MatCholeskyFactorSymbolic_ScaLAPACK,
1352: MatCholeskyFactorNumeric_ScaLAPACK,
1353: /*29*/ MatSetUp_ScaLAPACK,
1354: 0,
1355: 0,
1356: 0,
1357: 0,
1358: /*34*/ MatDuplicate_ScaLAPACK,
1359: 0,
1360: 0,
1361: 0,
1362: 0,
1363: /*39*/ MatAXPY_ScaLAPACK,
1364: 0,
1365: 0,
1366: 0,
1367: MatCopy_ScaLAPACK,
1368: /*44*/ 0,
1369: MatScale_ScaLAPACK,
1370: MatShift_ScaLAPACK,
1371: 0,
1372: 0,
1373: /*49*/ 0,
1374: 0,
1375: 0,
1376: 0,
1377: 0,
1378: /*54*/ 0,
1379: 0,
1380: 0,
1381: 0,
1382: 0,
1383: /*59*/ 0,
1384: MatDestroy_ScaLAPACK,
1385: MatView_ScaLAPACK,
1386: 0,
1387: 0,
1388: /*64*/ 0,
1389: 0,
1390: 0,
1391: 0,
1392: 0,
1393: /*69*/ 0,
1394: 0,
1395: MatConvert_ScaLAPACK_Dense,
1396: 0,
1397: 0,
1398: /*74*/ 0,
1399: 0,
1400: 0,
1401: 0,
1402: 0,
1403: /*79*/ 0,
1404: 0,
1405: 0,
1406: 0,
1407: MatLoad_ScaLAPACK,
1408: /*84*/ 0,
1409: 0,
1410: 0,
1411: 0,
1412: 0,
1413: /*89*/ 0,
1414: 0,
1415: MatMatMultNumeric_ScaLAPACK,
1416: 0,
1417: 0,
1418: /*94*/ 0,
1419: 0,
1420: 0,
1421: MatMatTransposeMultNumeric_ScaLAPACK,
1422: 0,
1423: /*99*/ MatProductSetFromOptions_ScaLAPACK,
1424: 0,
1425: 0,
1426: MatConjugate_ScaLAPACK,
1427: 0,
1428: /*104*/0,
1429: 0,
1430: 0,
1431: 0,
1432: 0,
1433: /*109*/MatMatSolve_ScaLAPACK,
1434: 0,
1435: 0,
1436: 0,
1437: MatMissingDiagonal_ScaLAPACK,
1438: /*114*/0,
1439: 0,
1440: 0,
1441: 0,
1442: 0,
1443: /*119*/0,
1444: MatHermitianTranspose_ScaLAPACK,
1445: 0,
1446: 0,
1447: 0,
1448: /*124*/0,
1449: 0,
1450: 0,
1451: 0,
1452: 0,
1453: /*129*/0,
1454: 0,
1455: 0,
1456: 0,
1457: 0,
1458: /*134*/0,
1459: 0,
1460: 0,
1461: 0,
1462: 0,
1463: 0,
1464: /*140*/0,
1465: 0,
1466: 0,
1467: 0,
1468: 0,
1469: /*145*/0,
1470: 0,
1471: 0
1472: };
1474: static PetscErrorCode MatStashScatterBegin_ScaLAPACK(Mat mat,MatStash *stash,PetscInt *owners)
1475: {
1476: PetscInt *owner,*startv,*starti,tag1=stash->tag1,tag2=stash->tag2,bs2;
1477: PetscInt size=stash->size,nsends;
1478: PetscErrorCode ierr;
1479: PetscInt count,*sindices,**rindices,i,j,l;
1480: PetscScalar **rvalues,*svalues;
1481: MPI_Comm comm = stash->comm;
1482: MPI_Request *send_waits,*recv_waits,*recv_waits1,*recv_waits2;
1483: PetscMPIInt *sizes,*nlengths,nreceives;
1484: PetscInt *sp_idx,*sp_idy;
1485: PetscScalar *sp_val;
1486: PetscMatStashSpace space,space_next;
1487: PetscBLASInt gridx,gcidx,lridx,lcidx,rsrc,csrc;
1488: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)mat->data;
1491: { /* make sure all processors are either in INSERTMODE or ADDMODE */
1492: InsertMode addv;
1493: MPIU_Allreduce((PetscEnum*)&mat->insertmode,(PetscEnum*)&addv,1,MPIU_ENUM,MPI_BOR,PetscObjectComm((PetscObject)mat));
1494: if (addv == (ADD_VALUES|INSERT_VALUES)) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Some processors inserted others added");
1495: mat->insertmode = addv; /* in case this processor had no cache */
1496: }
1498: bs2 = stash->bs*stash->bs;
1500: /* first count number of contributors to each processor */
1501: PetscCalloc1(size,&nlengths);
1502: PetscMalloc1(stash->n+1,&owner);
1504: i = j = 0;
1505: space = stash->space_head;
1506: while (space) {
1507: space_next = space->next;
1508: for (l=0; l<space->local_used; l++) {
1509: PetscBLASIntCast(space->idx[l]+1,&gridx);
1510: PetscBLASIntCast(space->idy[l]+1,&gcidx);
1511: PetscStackCallBLAS("SCALAPACKinfog2l",SCALAPACKinfog2l_(&gridx,&gcidx,a->desc,&a->grid->nprow,&a->grid->npcol,&a->grid->myrow,&a->grid->mycol,&lridx,&lcidx,&rsrc,&csrc));
1512: j = Cblacs_pnum(a->grid->ictxt,rsrc,csrc);
1513: nlengths[j]++; owner[i] = j;
1514: i++;
1515: }
1516: space = space_next;
1517: }
1519: /* Now check what procs get messages - and compute nsends. */
1520: PetscCalloc1(size,&sizes);
1521: for (i=0, nsends=0; i<size; i++) {
1522: if (nlengths[i]) {
1523: sizes[i] = 1; nsends++;
1524: }
1525: }
1527: {PetscMPIInt *onodes,*olengths;
1528: /* Determine the number of messages to expect, their lengths, from from-ids */
1529: PetscGatherNumberOfMessages(comm,sizes,nlengths,&nreceives);
1530: PetscGatherMessageLengths(comm,nsends,nreceives,nlengths,&onodes,&olengths);
1531: /* since clubbing row,col - lengths are multiplied by 2 */
1532: for (i=0; i<nreceives; i++) olengths[i] *=2;
1533: PetscPostIrecvInt(comm,tag1,nreceives,onodes,olengths,&rindices,&recv_waits1);
1534: /* values are size 'bs2' lengths (and remove earlier factor 2 */
1535: for (i=0; i<nreceives; i++) olengths[i] = olengths[i]*bs2/2;
1536: PetscPostIrecvScalar(comm,tag2,nreceives,onodes,olengths,&rvalues,&recv_waits2);
1537: PetscFree(onodes);
1538: PetscFree(olengths);}
1540: /* do sends:
1541: 1) starts[i] gives the starting index in svalues for stuff going to
1542: the ith processor
1543: */
1544: PetscMalloc2(bs2*stash->n,&svalues,2*(stash->n+1),&sindices);
1545: PetscMalloc1(2*nsends,&send_waits);
1546: PetscMalloc2(size,&startv,size,&starti);
1547: /* use 2 sends the first with all_a, the next with all_i and all_j */
1548: startv[0] = 0; starti[0] = 0;
1549: for (i=1; i<size; i++) {
1550: startv[i] = startv[i-1] + nlengths[i-1];
1551: starti[i] = starti[i-1] + 2*nlengths[i-1];
1552: }
1554: i = 0;
1555: space = stash->space_head;
1556: while (space) {
1557: space_next = space->next;
1558: sp_idx = space->idx;
1559: sp_idy = space->idy;
1560: sp_val = space->val;
1561: for (l=0; l<space->local_used; l++) {
1562: j = owner[i];
1563: if (bs2 == 1) {
1564: svalues[startv[j]] = sp_val[l];
1565: } else {
1566: PetscInt k;
1567: PetscScalar *buf1,*buf2;
1568: buf1 = svalues+bs2*startv[j];
1569: buf2 = space->val + bs2*l;
1570: for (k=0; k<bs2; k++) buf1[k] = buf2[k];
1571: }
1572: sindices[starti[j]] = sp_idx[l];
1573: sindices[starti[j]+nlengths[j]] = sp_idy[l];
1574: startv[j]++;
1575: starti[j]++;
1576: i++;
1577: }
1578: space = space_next;
1579: }
1580: startv[0] = 0;
1581: for (i=1; i<size; i++) startv[i] = startv[i-1] + nlengths[i-1];
1583: for (i=0,count=0; i<size; i++) {
1584: if (sizes[i]) {
1585: MPI_Isend(sindices+2*startv[i],2*nlengths[i],MPIU_INT,i,tag1,comm,send_waits+count++);
1586: MPI_Isend(svalues+bs2*startv[i],bs2*nlengths[i],MPIU_SCALAR,i,tag2,comm,send_waits+count++);
1587: }
1588: }
1589: #if defined(PETSC_USE_INFO)
1590: PetscInfo1(NULL,"No of messages: %d \n",nsends);
1591: for (i=0; i<size; i++) {
1592: if (sizes[i]) {
1593: PetscInfo2(NULL,"Mesg_to: %d: size: %d bytes\n",i,nlengths[i]*(bs2*sizeof(PetscScalar)+2*sizeof(PetscInt)));
1594: }
1595: }
1596: #endif
1597: PetscFree(nlengths);
1598: PetscFree(owner);
1599: PetscFree2(startv,starti);
1600: PetscFree(sizes);
1602: /* recv_waits need to be contiguous for MatStashScatterGetMesg_Private() */
1603: PetscMalloc1(2*nreceives,&recv_waits);
1605: for (i=0; i<nreceives; i++) {
1606: recv_waits[2*i] = recv_waits1[i];
1607: recv_waits[2*i+1] = recv_waits2[i];
1608: }
1609: stash->recv_waits = recv_waits;
1611: PetscFree(recv_waits1);
1612: PetscFree(recv_waits2);
1614: stash->svalues = svalues;
1615: stash->sindices = sindices;
1616: stash->rvalues = rvalues;
1617: stash->rindices = rindices;
1618: stash->send_waits = send_waits;
1619: stash->nsends = nsends;
1620: stash->nrecvs = nreceives;
1621: stash->reproduce_count = 0;
1622: return(0);
1623: }
1625: static PetscErrorCode MatScaLAPACKSetBlockSizes_ScaLAPACK(Mat A,PetscInt mb,PetscInt nb)
1626: {
1628: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1631: if (A->preallocated) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ORDER,"Cannot change block sizes after MatSetUp");
1632: if (mb<1 && mb!=PETSC_DECIDE) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"mb %D must be at least 1",mb);
1633: if (nb<1 && nb!=PETSC_DECIDE) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"nb %D must be at least 1",nb);
1634: PetscBLASIntCast((mb==PETSC_DECIDE)?DEFAULT_BLOCKSIZE:mb,&a->mb);
1635: PetscBLASIntCast((nb==PETSC_DECIDE)?a->mb:nb,&a->nb);
1636: return(0);
1637: }
1639: /*@
1640: MatScaLAPACKSetBlockSizes - Sets the block sizes to be used for the distibution of
1641: the ScaLAPACK matrix
1643: Logically Collective on A
1645: Input Parameter:
1646: + A - a MATSCALAPACK matrix
1647: . mb - the row block size
1648: - nb - the column block size
1650: Level: intermediate
1652: .seealso: MatCreateScaLAPACK(), MatScaLAPACKGetBlockSizes()
1653: @*/
1654: PetscErrorCode MatScaLAPACKSetBlockSizes(Mat A,PetscInt mb,PetscInt nb)
1655: {
1662: PetscTryMethod(A,"MatScaLAPACKSetBlockSizes_C",(Mat,PetscInt,PetscInt),(A,mb,nb));
1663: return(0);
1664: }
1666: static PetscErrorCode MatScaLAPACKGetBlockSizes_ScaLAPACK(Mat A,PetscInt *mb,PetscInt *nb)
1667: {
1668: Mat_ScaLAPACK *a = (Mat_ScaLAPACK*)A->data;
1671: if (mb) *mb = a->mb;
1672: if (nb) *nb = a->nb;
1673: return(0);
1674: }
1676: /*@
1677: MatScaLAPACKGetBlockSizes - Gets the block sizes used in the distibution of
1678: the ScaLAPACK matrix
1680: Not collective
1682: Input Parameter:
1683: . A - a MATSCALAPACK matrix
1685: Output Parameters:
1686: + mb - the row block size
1687: - nb - the column block size
1689: Level: intermediate
1691: .seealso: MatCreateScaLAPACK(), MatScaLAPACKSetBlockSizes()
1692: @*/
1693: PetscErrorCode MatScaLAPACKGetBlockSizes(Mat A,PetscInt *mb,PetscInt *nb)
1694: {
1699: PetscUseMethod(A,"MatScaLAPACKGetBlockSizes_C",(Mat,PetscInt*,PetscInt*),(A,mb,nb));
1700: return(0);
1701: }
1703: PETSC_INTERN PetscErrorCode MatStashScatterGetMesg_Ref(MatStash*,PetscMPIInt*,PetscInt**,PetscInt**,PetscScalar**,PetscInt*);
1704: PETSC_INTERN PetscErrorCode MatStashScatterEnd_Ref(MatStash*);
1706: /*MC
1707: MATSCALAPACK = "scalapack" - A matrix type for dense matrices using the ScaLAPACK package
1709: Use ./configure --download-scalapack to install PETSc to use ScaLAPACK
1711: Use -pc_type lu -pc_factor_mat_solver_type scalapack to use this direct solver
1713: Options Database Keys:
1714: + -mat_type scalapack - sets the matrix type to "scalapack" during a call to MatSetFromOptions()
1715: . -mat_scalapack_grid_height - sets Grid Height for 2D cyclic ordering of internal matrix
1716: - -mat_scalapack_block_sizes - size of the blocks to use (one or two integers separated by comma)
1718: Level: beginner
1720: .seealso: MATDENSE, MATELEMENTAL
1721: M*/
1723: PETSC_EXTERN PetscErrorCode MatCreate_ScaLAPACK(Mat A)
1724: {
1725: Mat_ScaLAPACK *a;
1726: PetscErrorCode ierr;
1727: PetscBool flg,flg1;
1728: Mat_ScaLAPACK_Grid *grid;
1729: MPI_Comm icomm;
1730: PetscBLASInt nprow,npcol,myrow,mycol;
1731: PetscInt optv1,k=2,array[2]={0,0};
1732: PetscMPIInt size;
1735: PetscMemcpy(A->ops,&MatOps_Values,sizeof(struct _MatOps));
1736: A->insertmode = NOT_SET_VALUES;
1738: MatStashCreate_Private(PetscObjectComm((PetscObject)A),1,&A->stash);
1739: A->stash.ScatterBegin = MatStashScatterBegin_ScaLAPACK;
1740: A->stash.ScatterGetMesg = MatStashScatterGetMesg_Ref;
1741: A->stash.ScatterEnd = MatStashScatterEnd_Ref;
1742: A->stash.ScatterDestroy = NULL;
1744: PetscNewLog(A,&a);
1745: A->data = (void*)a;
1747: /* Grid needs to be shared between multiple Mats on the same communicator, implement by attribute caching on the MPI_Comm */
1748: if (Petsc_ScaLAPACK_keyval == MPI_KEYVAL_INVALID) {
1749: MPI_Comm_create_keyval(MPI_COMM_NULL_COPY_FN,MPI_COMM_NULL_DELETE_FN,&Petsc_ScaLAPACK_keyval,(void*)0);
1750: PetscRegisterFinalize(Petsc_ScaLAPACK_keyval_free);
1751: PetscCitationsRegister(ScaLAPACKCitation,&ScaLAPACKCite);
1752: }
1753: PetscCommDuplicate(PetscObjectComm((PetscObject)A),&icomm,NULL);
1754: MPI_Comm_get_attr(icomm,Petsc_ScaLAPACK_keyval,(void**)&grid,(int*)&flg);
1755: if (!flg) {
1756: PetscNewLog(A,&grid);
1758: MPI_Comm_size(icomm,&size);
1759: grid->nprow = (PetscInt) (PetscSqrtReal((PetscReal)size) + 0.001);
1761: PetscOptionsBegin(PetscObjectComm((PetscObject)A),((PetscObject)A)->prefix,"ScaLAPACK Grid Options","Mat");
1762: PetscOptionsInt("-mat_scalapack_grid_height","Grid Height","None",grid->nprow,&optv1,&flg1);
1763: if (flg1) {
1764: if (size % optv1) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"Grid Height %D must evenly divide CommSize %D",optv1,size);
1765: grid->nprow = optv1;
1766: }
1767: PetscOptionsEnd();
1769: if (size % grid->nprow) grid->nprow = 1; /* cannot use a squarish grid, use a 1d grid */
1770: grid->npcol = size/grid->nprow;
1771: PetscBLASIntCast(grid->nprow,&nprow);
1772: PetscBLASIntCast(grid->npcol,&npcol);
1773: grid->ictxt = Csys2blacs_handle(icomm);
1774: Cblacs_gridinit(&grid->ictxt,"R",nprow,npcol);
1775: Cblacs_gridinfo(grid->ictxt,&nprow,&npcol,&myrow,&mycol);
1776: grid->grid_refct = 1;
1777: grid->nprow = nprow;
1778: grid->npcol = npcol;
1779: grid->myrow = myrow;
1780: grid->mycol = mycol;
1781: /* auxiliary 1d BLACS contexts for 1xsize and sizex1 grids */
1782: grid->ictxrow = Csys2blacs_handle(icomm);
1783: Cblacs_gridinit(&grid->ictxrow,"R",1,size);
1784: grid->ictxcol = Csys2blacs_handle(icomm);
1785: Cblacs_gridinit(&grid->ictxcol,"R",size,1);
1786: MPI_Comm_set_attr(icomm,Petsc_ScaLAPACK_keyval,(void*)grid);
1788: } else grid->grid_refct++;
1789: PetscCommDestroy(&icomm);
1790: a->grid = grid;
1791: a->mb = DEFAULT_BLOCKSIZE;
1792: a->nb = DEFAULT_BLOCKSIZE;
1794: PetscOptionsBegin(PetscObjectComm((PetscObject)A),NULL,"ScaLAPACK Options","Mat");
1795: PetscOptionsIntArray("-mat_scalapack_block_sizes","Size of the blocks to use (one or two comma-separated integers)","MatCreateScaLAPACK",array,&k,&flg);
1796: if (flg) {
1797: a->mb = array[0];
1798: a->nb = (k>1)? array[1]: a->mb;
1799: }
1800: PetscOptionsEnd();
1802: PetscObjectComposeFunction((PetscObject)A,"MatGetOwnershipIS_C",MatGetOwnershipIS_ScaLAPACK);
1803: PetscObjectComposeFunction((PetscObject)A,"MatScaLAPACKSetBlockSizes_C",MatScaLAPACKSetBlockSizes_ScaLAPACK);
1804: PetscObjectComposeFunction((PetscObject)A,"MatScaLAPACKGetBlockSizes_C",MatScaLAPACKGetBlockSizes_ScaLAPACK);
1805: PetscObjectChangeTypeName((PetscObject)A,MATSCALAPACK);
1806: return(0);
1807: }
1809: /*@C
1810: MatCreateScaLAPACK - Creates a dense parallel matrix in ScaLAPACK format
1811: (2D block cyclic distribution).
1813: Collective
1815: Input Parameters:
1816: + comm - MPI communicator
1817: . mb - row block size (or PETSC_DECIDE to have it set)
1818: . nb - column block size (or PETSC_DECIDE to have it set)
1819: . M - number of global rows
1820: . N - number of global columns
1821: . rsrc - coordinate of process that owns the first row of the distributed matrix
1822: - csrc - coordinate of process that owns the first column of the distributed matrix
1824: Output Parameter:
1825: . A - the matrix
1827: Options Database Keys:
1828: . -mat_scalapack_block_sizes - size of the blocks to use (one or two integers separated by comma)
1830: It is recommended that one use the MatCreate(), MatSetType() and/or MatSetFromOptions(),
1831: MatXXXXSetPreallocation() paradigm instead of this routine directly.
1832: [MatXXXXSetPreallocation() is, for example, MatSeqAIJSetPreallocation]
1834: Notes:
1835: If PETSC_DECIDE is used for the block sizes, then an appropriate value
1836: is chosen.
1838: Storage Information:
1839: Storate is completely managed by ScaLAPACK, so this requires PETSc to be
1840: configured with ScaLAPACK. In particular, PETSc's local sizes lose
1841: significance and are thus ignored. The block sizes refer to the values
1842: used for the distributed matrix, not the same meaning as in BAIJ.
1844: Level: intermediate
1846: .seealso: MatCreate(), MatCreateDense(), MatSetValues()
1847: @*/
1848: PetscErrorCode MatCreateScaLAPACK(MPI_Comm comm,PetscInt mb,PetscInt nb,PetscInt M,PetscInt N,PetscInt rsrc,PetscInt csrc,Mat *A)
1849: {
1851: Mat_ScaLAPACK *a;
1852: PetscInt m,n;
1855: MatCreate(comm,A);
1856: MatSetType(*A,MATSCALAPACK);
1857: if (M==PETSC_DECIDE || N==PETSC_DECIDE) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Cannot use PETSC_DECIDE for matrix dimensions");
1858: /* rows and columns are NOT distributed according to PetscSplitOwnership */
1859: m = PETSC_DECIDE;
1860: PetscSplitOwnershipEqual(comm,&m,&M);
1861: n = PETSC_DECIDE;
1862: PetscSplitOwnershipEqual(comm,&n,&N);
1863: MatSetSizes(*A,m,n,M,N);
1864: a = (Mat_ScaLAPACK*)(*A)->data;
1865: PetscBLASIntCast(M,&a->M);
1866: PetscBLASIntCast(N,&a->N);
1867: PetscBLASIntCast((mb==PETSC_DECIDE)?DEFAULT_BLOCKSIZE:mb,&a->mb);
1868: PetscBLASIntCast((nb==PETSC_DECIDE)?a->mb:nb,&a->nb);
1869: PetscBLASIntCast(rsrc,&a->rsrc);
1870: PetscBLASIntCast(csrc,&a->csrc);
1871: MatSetUp(*A);
1872: return(0);
1873: }