Actual source code: dense.c
1: /*
2: Defines the basic matrix operations for sequential dense.
3: Portions of this code are under:
4: Copyright (c) 2022 Advanced Micro Devices, Inc. All rights reserved.
5: */
7: #include <../src/mat/impls/dense/seq/dense.h>
8: #include <../src/mat/impls/dense/mpi/mpidense.h>
9: #include <petscblaslapack.h>
10: #include <../src/mat/impls/aij/seq/aij.h>
12: PetscErrorCode MatSeqDenseSymmetrize_Private(Mat A, PetscBool hermitian)
13: {
14: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
15: PetscInt j, k, n = A->rmap->n;
16: PetscScalar *v;
18: PetscFunctionBegin;
19: PetscCheck(A->rmap->n == A->cmap->n, PetscObjectComm((PetscObject)A), PETSC_ERR_SUP, "Cannot symmetrize a rectangular matrix");
20: PetscCall(MatDenseGetArray(A, &v));
21: if (!hermitian) {
22: for (k = 0; k < n; k++) {
23: for (j = k; j < n; j++) v[j * mat->lda + k] = v[k * mat->lda + j];
24: }
25: } else {
26: for (k = 0; k < n; k++) {
27: for (j = k; j < n; j++) v[j * mat->lda + k] = PetscConj(v[k * mat->lda + j]);
28: }
29: }
30: PetscCall(MatDenseRestoreArray(A, &v));
31: PetscFunctionReturn(PETSC_SUCCESS);
32: }
34: PetscErrorCode MatSeqDenseInvertFactors_Private(Mat A)
35: {
36: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
37: PetscBLASInt info, n;
39: PetscFunctionBegin;
40: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
41: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
42: if (A->factortype == MAT_FACTOR_LU) {
43: PetscCheck(mat->pivots, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Pivots not present");
44: if (!mat->fwork) {
45: mat->lfwork = n;
46: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
47: }
48: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
49: PetscCallBLAS("LAPACKgetri", LAPACKgetri_(&n, mat->v, &mat->lda, mat->pivots, mat->fwork, &mat->lfwork, &info));
50: PetscCall(PetscFPTrapPop());
51: PetscCall(PetscLogFlops((1.0 * A->cmap->n * A->cmap->n * A->cmap->n) / 3.0));
52: } else if (A->factortype == MAT_FACTOR_CHOLESKY) {
53: if (A->spd == PETSC_BOOL3_TRUE) {
54: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
55: PetscCallBLAS("LAPACKpotri", LAPACKpotri_("L", &n, mat->v, &mat->lda, &info));
56: PetscCall(PetscFPTrapPop());
57: PetscCall(MatSeqDenseSymmetrize_Private(A, PETSC_TRUE));
58: #if defined(PETSC_USE_COMPLEX)
59: } else if (A->hermitian == PETSC_BOOL3_TRUE) {
60: PetscCheck(mat->pivots, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Pivots not present");
61: PetscCheck(mat->fwork, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Fwork not present");
62: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
63: PetscCallBLAS("LAPACKhetri", LAPACKhetri_("L", &n, mat->v, &mat->lda, mat->pivots, mat->fwork, &info));
64: PetscCall(PetscFPTrapPop());
65: PetscCall(MatSeqDenseSymmetrize_Private(A, PETSC_TRUE));
66: #endif
67: } else { /* symmetric case */
68: PetscCheck(mat->pivots, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Pivots not present");
69: PetscCheck(mat->fwork, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Fwork not present");
70: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
71: PetscCallBLAS("LAPACKsytri", LAPACKsytri_("L", &n, mat->v, &mat->lda, mat->pivots, mat->fwork, &info));
72: PetscCall(PetscFPTrapPop());
73: PetscCall(MatSeqDenseSymmetrize_Private(A, PETSC_FALSE));
74: }
75: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_MAT_CH_ZRPVT, "Bad Inversion: zero pivot in row %" PetscInt_FMT, (PetscInt)info - 1);
76: PetscCall(PetscLogFlops((1.0 * A->cmap->n * A->cmap->n * A->cmap->n) / 3.0));
77: } else SETERRQ(PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Matrix must be factored to solve");
79: A->ops->solve = NULL;
80: A->ops->matsolve = NULL;
81: A->ops->solvetranspose = NULL;
82: A->ops->matsolvetranspose = NULL;
83: A->ops->solveadd = NULL;
84: A->ops->solvetransposeadd = NULL;
85: A->factortype = MAT_FACTOR_NONE;
86: PetscCall(PetscFree(A->solvertype));
87: PetscFunctionReturn(PETSC_SUCCESS);
88: }
90: PetscErrorCode MatZeroRowsColumns_SeqDense(Mat A, PetscInt N, const PetscInt rows[], PetscScalar diag, Vec x, Vec b)
91: {
92: Mat_SeqDense *l = (Mat_SeqDense *)A->data;
93: PetscInt m = l->lda, n = A->cmap->n, r = A->rmap->n, i, j;
94: PetscScalar *slot, *bb, *v;
95: const PetscScalar *xx;
97: PetscFunctionBegin;
98: if (PetscDefined(USE_DEBUG)) {
99: for (i = 0; i < N; i++) {
100: PetscCheck(rows[i] >= 0, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Negative row requested to be zeroed");
101: PetscCheck(rows[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row %" PetscInt_FMT " requested to be zeroed greater than or equal number of rows %" PetscInt_FMT, rows[i], A->rmap->n);
102: PetscCheck(rows[i] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Col %" PetscInt_FMT " requested to be zeroed greater than or equal number of cols %" PetscInt_FMT, rows[i], A->cmap->n);
103: }
104: }
105: if (!N) PetscFunctionReturn(PETSC_SUCCESS);
107: /* fix right hand side if needed */
108: if (x && b) {
109: Vec xt;
111: PetscCheck(A->rmap->n == A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_SUP, "Only coded for square matrices");
112: PetscCall(VecDuplicate(x, &xt));
113: PetscCall(VecCopy(x, xt));
114: PetscCall(VecScale(xt, -1.0));
115: PetscCall(MatMultAdd(A, xt, b, b));
116: PetscCall(VecDestroy(&xt));
117: PetscCall(VecGetArrayRead(x, &xx));
118: PetscCall(VecGetArray(b, &bb));
119: for (i = 0; i < N; i++) bb[rows[i]] = diag * xx[rows[i]];
120: PetscCall(VecRestoreArrayRead(x, &xx));
121: PetscCall(VecRestoreArray(b, &bb));
122: }
124: PetscCall(MatDenseGetArray(A, &v));
125: for (i = 0; i < N; i++) {
126: slot = v + rows[i] * m;
127: PetscCall(PetscArrayzero(slot, r));
128: }
129: for (i = 0; i < N; i++) {
130: slot = v + rows[i];
131: for (j = 0; j < n; j++) {
132: *slot = 0.0;
133: slot += m;
134: }
135: }
136: if (diag != 0.0) {
137: PetscCheck(A->rmap->n == A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_SUP, "Only coded for square matrices");
138: for (i = 0; i < N; i++) {
139: slot = v + (m + 1) * rows[i];
140: *slot = diag;
141: }
142: }
143: PetscCall(MatDenseRestoreArray(A, &v));
144: PetscFunctionReturn(PETSC_SUCCESS);
145: }
147: PETSC_INTERN PetscErrorCode MatConvert_SeqAIJ_SeqDense(Mat A, MatType newtype, MatReuse reuse, Mat *newmat)
148: {
149: Mat B = NULL;
150: Mat_SeqAIJ *a = (Mat_SeqAIJ *)A->data;
151: Mat_SeqDense *b;
152: PetscInt *ai = a->i, *aj = a->j, m = A->rmap->N, n = A->cmap->N, i;
153: const MatScalar *av;
154: PetscBool isseqdense;
156: PetscFunctionBegin;
157: if (reuse == MAT_REUSE_MATRIX) {
158: PetscCall(PetscObjectTypeCompare((PetscObject)*newmat, MATSEQDENSE, &isseqdense));
159: PetscCheck(isseqdense, PetscObjectComm((PetscObject)*newmat), PETSC_ERR_USER, "Cannot reuse matrix of type %s", ((PetscObject)(*newmat))->type_name);
160: }
161: if (reuse != MAT_REUSE_MATRIX) {
162: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), &B));
163: PetscCall(MatSetSizes(B, m, n, m, n));
164: PetscCall(MatSetType(B, MATSEQDENSE));
165: PetscCall(MatSeqDenseSetPreallocation(B, NULL));
166: b = (Mat_SeqDense *)(B->data);
167: } else {
168: b = (Mat_SeqDense *)((*newmat)->data);
169: PetscCall(PetscArrayzero(b->v, m * n));
170: }
171: PetscCall(MatSeqAIJGetArrayRead(A, &av));
172: for (i = 0; i < m; i++) {
173: PetscInt j;
174: for (j = 0; j < ai[1] - ai[0]; j++) {
175: b->v[*aj * m + i] = *av;
176: aj++;
177: av++;
178: }
179: ai++;
180: }
181: PetscCall(MatSeqAIJRestoreArrayRead(A, &av));
183: if (reuse == MAT_INPLACE_MATRIX) {
184: PetscCall(MatAssemblyBegin(B, MAT_FINAL_ASSEMBLY));
185: PetscCall(MatAssemblyEnd(B, MAT_FINAL_ASSEMBLY));
186: PetscCall(MatHeaderReplace(A, &B));
187: } else {
188: if (B) *newmat = B;
189: PetscCall(MatAssemblyBegin(*newmat, MAT_FINAL_ASSEMBLY));
190: PetscCall(MatAssemblyEnd(*newmat, MAT_FINAL_ASSEMBLY));
191: }
192: PetscFunctionReturn(PETSC_SUCCESS);
193: }
195: PETSC_INTERN PetscErrorCode MatConvert_SeqDense_SeqAIJ(Mat A, MatType newtype, MatReuse reuse, Mat *newmat)
196: {
197: Mat B = NULL;
198: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
199: PetscInt i, j;
200: PetscInt *rows, *nnz;
201: MatScalar *aa = a->v, *vals;
203: PetscFunctionBegin;
204: PetscCall(PetscCalloc3(A->rmap->n, &rows, A->rmap->n, &nnz, A->rmap->n, &vals));
205: if (reuse != MAT_REUSE_MATRIX) {
206: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), &B));
207: PetscCall(MatSetSizes(B, A->rmap->n, A->cmap->n, A->rmap->N, A->cmap->N));
208: PetscCall(MatSetType(B, MATSEQAIJ));
209: for (j = 0; j < A->cmap->n; j++) {
210: for (i = 0; i < A->rmap->n; i++)
211: if (aa[i] != 0.0 || (i == j && A->cmap->n == A->rmap->n)) ++nnz[i];
212: aa += a->lda;
213: }
214: PetscCall(MatSeqAIJSetPreallocation(B, PETSC_DETERMINE, nnz));
215: } else B = *newmat;
216: aa = a->v;
217: for (j = 0; j < A->cmap->n; j++) {
218: PetscInt numRows = 0;
219: for (i = 0; i < A->rmap->n; i++)
220: if (aa[i] != 0.0 || (i == j && A->cmap->n == A->rmap->n)) {
221: rows[numRows] = i;
222: vals[numRows++] = aa[i];
223: }
224: PetscCall(MatSetValues(B, numRows, rows, 1, &j, vals, INSERT_VALUES));
225: aa += a->lda;
226: }
227: PetscCall(PetscFree3(rows, nnz, vals));
228: PetscCall(MatAssemblyBegin(B, MAT_FINAL_ASSEMBLY));
229: PetscCall(MatAssemblyEnd(B, MAT_FINAL_ASSEMBLY));
231: if (reuse == MAT_INPLACE_MATRIX) {
232: PetscCall(MatHeaderReplace(A, &B));
233: } else if (reuse != MAT_REUSE_MATRIX) *newmat = B;
234: PetscFunctionReturn(PETSC_SUCCESS);
235: }
237: PetscErrorCode MatAXPY_SeqDense(Mat Y, PetscScalar alpha, Mat X, MatStructure str)
238: {
239: Mat_SeqDense *x = (Mat_SeqDense *)X->data, *y = (Mat_SeqDense *)Y->data;
240: const PetscScalar *xv;
241: PetscScalar *yv;
242: PetscBLASInt N, m, ldax = 0, lday = 0, one = 1;
244: PetscFunctionBegin;
245: PetscCall(MatDenseGetArrayRead(X, &xv));
246: PetscCall(MatDenseGetArray(Y, &yv));
247: PetscCall(PetscBLASIntCast(X->rmap->n * X->cmap->n, &N));
248: PetscCall(PetscBLASIntCast(X->rmap->n, &m));
249: PetscCall(PetscBLASIntCast(x->lda, &ldax));
250: PetscCall(PetscBLASIntCast(y->lda, &lday));
251: if (ldax > m || lday > m) {
252: PetscInt j;
254: for (j = 0; j < X->cmap->n; j++) PetscCallBLAS("BLASaxpy", BLASaxpy_(&m, &alpha, xv + j * ldax, &one, yv + j * lday, &one));
255: } else {
256: PetscCallBLAS("BLASaxpy", BLASaxpy_(&N, &alpha, xv, &one, yv, &one));
257: }
258: PetscCall(MatDenseRestoreArrayRead(X, &xv));
259: PetscCall(MatDenseRestoreArray(Y, &yv));
260: PetscCall(PetscLogFlops(PetscMax(2.0 * N - 1, 0)));
261: PetscFunctionReturn(PETSC_SUCCESS);
262: }
264: static PetscErrorCode MatGetInfo_SeqDense(Mat A, MatInfoType flag, MatInfo *info)
265: {
266: PetscLogDouble N = A->rmap->n * A->cmap->n;
268: PetscFunctionBegin;
269: info->block_size = 1.0;
270: info->nz_allocated = N;
271: info->nz_used = N;
272: info->nz_unneeded = 0;
273: info->assemblies = A->num_ass;
274: info->mallocs = 0;
275: info->memory = 0; /* REVIEW ME */
276: info->fill_ratio_given = 0;
277: info->fill_ratio_needed = 0;
278: info->factor_mallocs = 0;
279: PetscFunctionReturn(PETSC_SUCCESS);
280: }
282: PetscErrorCode MatScale_SeqDense(Mat A, PetscScalar alpha)
283: {
284: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
285: PetscScalar *v;
286: PetscBLASInt one = 1, j, nz, lda = 0;
288: PetscFunctionBegin;
289: PetscCall(MatDenseGetArray(A, &v));
290: PetscCall(PetscBLASIntCast(a->lda, &lda));
291: if (lda > A->rmap->n) {
292: PetscCall(PetscBLASIntCast(A->rmap->n, &nz));
293: for (j = 0; j < A->cmap->n; j++) PetscCallBLAS("BLASscal", BLASscal_(&nz, &alpha, v + j * lda, &one));
294: } else {
295: PetscCall(PetscBLASIntCast(A->rmap->n * A->cmap->n, &nz));
296: PetscCallBLAS("BLASscal", BLASscal_(&nz, &alpha, v, &one));
297: }
298: PetscCall(PetscLogFlops(A->rmap->n * A->cmap->n));
299: PetscCall(MatDenseRestoreArray(A, &v));
300: PetscFunctionReturn(PETSC_SUCCESS);
301: }
303: PetscErrorCode MatShift_SeqDense(Mat A, PetscScalar alpha)
304: {
305: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
306: PetscScalar *v;
307: PetscInt j, k;
309: PetscFunctionBegin;
310: PetscCall(MatDenseGetArray(A, &v));
311: k = PetscMin(A->rmap->n, A->cmap->n);
312: for (j = 0; j < k; j++) v[j + j * a->lda] += alpha;
313: PetscCall(PetscLogFlops(k));
314: PetscCall(MatDenseRestoreArray(A, &v));
315: PetscFunctionReturn(PETSC_SUCCESS);
316: }
318: static PetscErrorCode MatIsHermitian_SeqDense(Mat A, PetscReal rtol, PetscBool *fl)
319: {
320: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
321: PetscInt i, j, m = A->rmap->n, N = a->lda;
322: const PetscScalar *v;
324: PetscFunctionBegin;
325: *fl = PETSC_FALSE;
326: if (A->rmap->n != A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
327: PetscCall(MatDenseGetArrayRead(A, &v));
328: for (i = 0; i < m; i++) {
329: for (j = i; j < m; j++) {
330: if (PetscAbsScalar(v[i + j * N] - PetscConj(v[j + i * N])) > rtol) goto restore;
331: }
332: }
333: *fl = PETSC_TRUE;
334: restore:
335: PetscCall(MatDenseRestoreArrayRead(A, &v));
336: PetscFunctionReturn(PETSC_SUCCESS);
337: }
339: static PetscErrorCode MatIsSymmetric_SeqDense(Mat A, PetscReal rtol, PetscBool *fl)
340: {
341: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
342: PetscInt i, j, m = A->rmap->n, N = a->lda;
343: const PetscScalar *v;
345: PetscFunctionBegin;
346: *fl = PETSC_FALSE;
347: if (A->rmap->n != A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
348: PetscCall(MatDenseGetArrayRead(A, &v));
349: for (i = 0; i < m; i++) {
350: for (j = i; j < m; j++) {
351: if (PetscAbsScalar(v[i + j * N] - v[j + i * N]) > rtol) goto restore;
352: }
353: }
354: *fl = PETSC_TRUE;
355: restore:
356: PetscCall(MatDenseRestoreArrayRead(A, &v));
357: PetscFunctionReturn(PETSC_SUCCESS);
358: }
360: PetscErrorCode MatDuplicateNoCreate_SeqDense(Mat newi, Mat A, MatDuplicateOption cpvalues)
361: {
362: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
363: PetscInt lda = (PetscInt)mat->lda, j, m, nlda = lda;
364: PetscBool isdensecpu;
366: PetscFunctionBegin;
367: PetscCall(PetscLayoutReference(A->rmap, &newi->rmap));
368: PetscCall(PetscLayoutReference(A->cmap, &newi->cmap));
369: if (cpvalues == MAT_SHARE_NONZERO_PATTERN) { /* propagate LDA */
370: PetscCall(MatDenseSetLDA(newi, lda));
371: }
372: PetscCall(PetscObjectTypeCompare((PetscObject)newi, MATSEQDENSE, &isdensecpu));
373: if (isdensecpu) PetscCall(MatSeqDenseSetPreallocation(newi, NULL));
374: if (cpvalues == MAT_COPY_VALUES) {
375: const PetscScalar *av;
376: PetscScalar *v;
378: PetscCall(MatDenseGetArrayRead(A, &av));
379: PetscCall(MatDenseGetArrayWrite(newi, &v));
380: PetscCall(MatDenseGetLDA(newi, &nlda));
381: m = A->rmap->n;
382: if (lda > m || nlda > m) {
383: for (j = 0; j < A->cmap->n; j++) PetscCall(PetscArraycpy(v + j * nlda, av + j * lda, m));
384: } else {
385: PetscCall(PetscArraycpy(v, av, A->rmap->n * A->cmap->n));
386: }
387: PetscCall(MatDenseRestoreArrayWrite(newi, &v));
388: PetscCall(MatDenseRestoreArrayRead(A, &av));
389: }
390: PetscFunctionReturn(PETSC_SUCCESS);
391: }
393: PetscErrorCode MatDuplicate_SeqDense(Mat A, MatDuplicateOption cpvalues, Mat *newmat)
394: {
395: PetscFunctionBegin;
396: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), newmat));
397: PetscCall(MatSetSizes(*newmat, A->rmap->n, A->cmap->n, A->rmap->n, A->cmap->n));
398: PetscCall(MatSetType(*newmat, ((PetscObject)A)->type_name));
399: PetscCall(MatDuplicateNoCreate_SeqDense(*newmat, A, cpvalues));
400: PetscFunctionReturn(PETSC_SUCCESS);
401: }
403: static PetscErrorCode MatSolve_SeqDense_Internal_LU(Mat A, PetscScalar *x, PetscBLASInt ldx, PetscBLASInt m, PetscBLASInt nrhs, PetscBLASInt k, PetscBool T)
404: {
405: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
406: PetscBLASInt info;
408: PetscFunctionBegin;
409: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
410: PetscCallBLAS("LAPACKgetrs", LAPACKgetrs_(T ? "T" : "N", &m, &nrhs, mat->v, &mat->lda, mat->pivots, x, &m, &info));
411: PetscCall(PetscFPTrapPop());
412: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "GETRS - Bad solve %d", (int)info);
413: PetscCall(PetscLogFlops(nrhs * (2.0 * m * m - m)));
414: PetscFunctionReturn(PETSC_SUCCESS);
415: }
417: static PetscErrorCode MatConjugate_SeqDense(Mat);
419: static PetscErrorCode MatSolve_SeqDense_Internal_Cholesky(Mat A, PetscScalar *x, PetscBLASInt ldx, PetscBLASInt m, PetscBLASInt nrhs, PetscBLASInt k, PetscBool T)
420: {
421: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
422: PetscBLASInt info;
424: PetscFunctionBegin;
425: if (A->spd == PETSC_BOOL3_TRUE) {
426: if (PetscDefined(USE_COMPLEX) && T) PetscCall(MatConjugate_SeqDense(A));
427: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
428: PetscCallBLAS("LAPACKpotrs", LAPACKpotrs_("L", &m, &nrhs, mat->v, &mat->lda, x, &m, &info));
429: PetscCall(PetscFPTrapPop());
430: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "POTRS Bad solve %d", (int)info);
431: if (PetscDefined(USE_COMPLEX) && T) PetscCall(MatConjugate_SeqDense(A));
432: #if defined(PETSC_USE_COMPLEX)
433: } else if (A->hermitian == PETSC_BOOL3_TRUE) {
434: if (T) PetscCall(MatConjugate_SeqDense(A));
435: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
436: PetscCallBLAS("LAPACKhetrs", LAPACKhetrs_("L", &m, &nrhs, mat->v, &mat->lda, mat->pivots, x, &m, &info));
437: PetscCall(PetscFPTrapPop());
438: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "HETRS Bad solve %d", (int)info);
439: if (T) PetscCall(MatConjugate_SeqDense(A));
440: #endif
441: } else { /* symmetric case */
442: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
443: PetscCallBLAS("LAPACKsytrs", LAPACKsytrs_("L", &m, &nrhs, mat->v, &mat->lda, mat->pivots, x, &m, &info));
444: PetscCall(PetscFPTrapPop());
445: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "SYTRS Bad solve %d", (int)info);
446: }
447: PetscCall(PetscLogFlops(nrhs * (2.0 * m * m - m)));
448: PetscFunctionReturn(PETSC_SUCCESS);
449: }
451: static PetscErrorCode MatSolve_SeqDense_Internal_QR(Mat A, PetscScalar *x, PetscBLASInt ldx, PetscBLASInt m, PetscBLASInt nrhs, PetscBLASInt k)
452: {
453: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
454: PetscBLASInt info;
455: char trans;
457: PetscFunctionBegin;
458: if (PetscDefined(USE_COMPLEX)) {
459: trans = 'C';
460: } else {
461: trans = 'T';
462: }
463: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
464: { /* lwork depends on the number of right-hand sides */
465: PetscBLASInt nlfwork, lfwork = -1;
466: PetscScalar fwork;
468: PetscCallBLAS("LAPACKormqr", LAPACKormqr_("L", &trans, &m, &nrhs, &mat->rank, mat->v, &mat->lda, mat->tau, x, &ldx, &fwork, &lfwork, &info));
469: nlfwork = (PetscBLASInt)PetscRealPart(fwork);
470: if (nlfwork > mat->lfwork) {
471: mat->lfwork = nlfwork;
472: PetscCall(PetscFree(mat->fwork));
473: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
474: }
475: }
476: PetscCallBLAS("LAPACKormqr", LAPACKormqr_("L", &trans, &m, &nrhs, &mat->rank, mat->v, &mat->lda, mat->tau, x, &ldx, mat->fwork, &mat->lfwork, &info));
477: PetscCall(PetscFPTrapPop());
478: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "ORMQR - Bad orthogonal transform %d", (int)info);
479: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
480: PetscCallBLAS("LAPACKtrtrs", LAPACKtrtrs_("U", "N", "N", &mat->rank, &nrhs, mat->v, &mat->lda, x, &ldx, &info));
481: PetscCall(PetscFPTrapPop());
482: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "TRTRS - Bad triangular solve %d", (int)info);
483: for (PetscInt j = 0; j < nrhs; j++) {
484: for (PetscInt i = mat->rank; i < k; i++) x[j * ldx + i] = 0.;
485: }
486: PetscCall(PetscLogFlops(nrhs * (4.0 * m * mat->rank - PetscSqr(mat->rank))));
487: PetscFunctionReturn(PETSC_SUCCESS);
488: }
490: static PetscErrorCode MatSolveTranspose_SeqDense_Internal_QR(Mat A, PetscScalar *x, PetscBLASInt ldx, PetscBLASInt m, PetscBLASInt nrhs, PetscBLASInt k)
491: {
492: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
493: PetscBLASInt info;
495: PetscFunctionBegin;
496: if (A->rmap->n == A->cmap->n && mat->rank == A->rmap->n) {
497: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
498: PetscCallBLAS("LAPACKtrtrs", LAPACKtrtrs_("U", "T", "N", &m, &nrhs, mat->v, &mat->lda, x, &ldx, &info));
499: PetscCall(PetscFPTrapPop());
500: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "TRTRS - Bad triangular solve %d", (int)info);
501: if (PetscDefined(USE_COMPLEX)) PetscCall(MatConjugate_SeqDense(A));
502: { /* lwork depends on the number of right-hand sides */
503: PetscBLASInt nlfwork, lfwork = -1;
504: PetscScalar fwork;
506: PetscCallBLAS("LAPACKormqr", LAPACKormqr_("L", "N", &m, &nrhs, &mat->rank, mat->v, &mat->lda, mat->tau, x, &ldx, &fwork, &lfwork, &info));
507: nlfwork = (PetscBLASInt)PetscRealPart(fwork);
508: if (nlfwork > mat->lfwork) {
509: mat->lfwork = nlfwork;
510: PetscCall(PetscFree(mat->fwork));
511: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
512: }
513: }
514: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
515: PetscCallBLAS("LAPACKormqr", LAPACKormqr_("L", "N", &m, &nrhs, &mat->rank, mat->v, &mat->lda, mat->tau, x, &ldx, mat->fwork, &mat->lfwork, &info));
516: PetscCall(PetscFPTrapPop());
517: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "ORMQR - Bad orthogonal transform %d", (int)info);
518: if (PetscDefined(USE_COMPLEX)) PetscCall(MatConjugate_SeqDense(A));
519: } else SETERRQ(PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "QR factored matrix cannot be used for transpose solve");
520: PetscCall(PetscLogFlops(nrhs * (4.0 * m * mat->rank - PetscSqr(mat->rank))));
521: PetscFunctionReturn(PETSC_SUCCESS);
522: }
524: static PetscErrorCode MatSolve_SeqDense_SetUp(Mat A, Vec xx, Vec yy, PetscScalar **_y, PetscBLASInt *_m, PetscBLASInt *_k)
525: {
526: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
527: PetscScalar *y;
528: PetscBLASInt m = 0, k = 0;
530: PetscFunctionBegin;
531: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
532: PetscCall(PetscBLASIntCast(A->cmap->n, &k));
533: if (k < m) {
534: PetscCall(VecCopy(xx, mat->qrrhs));
535: PetscCall(VecGetArray(mat->qrrhs, &y));
536: } else {
537: PetscCall(VecCopy(xx, yy));
538: PetscCall(VecGetArray(yy, &y));
539: }
540: *_y = y;
541: *_k = k;
542: *_m = m;
543: PetscFunctionReturn(PETSC_SUCCESS);
544: }
546: static PetscErrorCode MatSolve_SeqDense_TearDown(Mat A, Vec xx, Vec yy, PetscScalar **_y, PetscBLASInt *_m, PetscBLASInt *_k)
547: {
548: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
549: PetscScalar *y = NULL;
550: PetscBLASInt m, k;
552: PetscFunctionBegin;
553: y = *_y;
554: *_y = NULL;
555: k = *_k;
556: m = *_m;
557: if (k < m) {
558: PetscScalar *yv;
559: PetscCall(VecGetArray(yy, &yv));
560: PetscCall(PetscArraycpy(yv, y, k));
561: PetscCall(VecRestoreArray(yy, &yv));
562: PetscCall(VecRestoreArray(mat->qrrhs, &y));
563: } else {
564: PetscCall(VecRestoreArray(yy, &y));
565: }
566: PetscFunctionReturn(PETSC_SUCCESS);
567: }
569: static PetscErrorCode MatSolve_SeqDense_LU(Mat A, Vec xx, Vec yy)
570: {
571: PetscScalar *y = NULL;
572: PetscBLASInt m = 0, k = 0;
574: PetscFunctionBegin;
575: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
576: PetscCall(MatSolve_SeqDense_Internal_LU(A, y, m, m, 1, k, PETSC_FALSE));
577: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
578: PetscFunctionReturn(PETSC_SUCCESS);
579: }
581: static PetscErrorCode MatSolveTranspose_SeqDense_LU(Mat A, Vec xx, Vec yy)
582: {
583: PetscScalar *y = NULL;
584: PetscBLASInt m = 0, k = 0;
586: PetscFunctionBegin;
587: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
588: PetscCall(MatSolve_SeqDense_Internal_LU(A, y, m, m, 1, k, PETSC_TRUE));
589: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
590: PetscFunctionReturn(PETSC_SUCCESS);
591: }
593: static PetscErrorCode MatSolve_SeqDense_Cholesky(Mat A, Vec xx, Vec yy)
594: {
595: PetscScalar *y = NULL;
596: PetscBLASInt m = 0, k = 0;
598: PetscFunctionBegin;
599: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
600: PetscCall(MatSolve_SeqDense_Internal_Cholesky(A, y, m, m, 1, k, PETSC_FALSE));
601: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
602: PetscFunctionReturn(PETSC_SUCCESS);
603: }
605: static PetscErrorCode MatSolveTranspose_SeqDense_Cholesky(Mat A, Vec xx, Vec yy)
606: {
607: PetscScalar *y = NULL;
608: PetscBLASInt m = 0, k = 0;
610: PetscFunctionBegin;
611: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
612: PetscCall(MatSolve_SeqDense_Internal_Cholesky(A, y, m, m, 1, k, PETSC_TRUE));
613: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
614: PetscFunctionReturn(PETSC_SUCCESS);
615: }
617: static PetscErrorCode MatSolve_SeqDense_QR(Mat A, Vec xx, Vec yy)
618: {
619: PetscScalar *y = NULL;
620: PetscBLASInt m = 0, k = 0;
622: PetscFunctionBegin;
623: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
624: PetscCall(MatSolve_SeqDense_Internal_QR(A, y, PetscMax(m, k), m, 1, k));
625: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
626: PetscFunctionReturn(PETSC_SUCCESS);
627: }
629: static PetscErrorCode MatSolveTranspose_SeqDense_QR(Mat A, Vec xx, Vec yy)
630: {
631: PetscScalar *y = NULL;
632: PetscBLASInt m = 0, k = 0;
634: PetscFunctionBegin;
635: PetscCall(MatSolve_SeqDense_SetUp(A, xx, yy, &y, &m, &k));
636: PetscCall(MatSolveTranspose_SeqDense_Internal_QR(A, y, PetscMax(m, k), m, 1, k));
637: PetscCall(MatSolve_SeqDense_TearDown(A, xx, yy, &y, &m, &k));
638: PetscFunctionReturn(PETSC_SUCCESS);
639: }
641: static PetscErrorCode MatMatSolve_SeqDense_SetUp(Mat A, Mat B, Mat X, PetscScalar **_y, PetscBLASInt *_ldy, PetscBLASInt *_m, PetscBLASInt *_nrhs, PetscBLASInt *_k)
642: {
643: const PetscScalar *b;
644: PetscScalar *y;
645: PetscInt n, _ldb, _ldx;
646: PetscBLASInt nrhs = 0, m = 0, k = 0, ldb = 0, ldx = 0, ldy = 0;
648: PetscFunctionBegin;
649: *_ldy = 0;
650: *_m = 0;
651: *_nrhs = 0;
652: *_k = 0;
653: *_y = NULL;
654: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
655: PetscCall(PetscBLASIntCast(A->cmap->n, &k));
656: PetscCall(MatGetSize(B, NULL, &n));
657: PetscCall(PetscBLASIntCast(n, &nrhs));
658: PetscCall(MatDenseGetLDA(B, &_ldb));
659: PetscCall(PetscBLASIntCast(_ldb, &ldb));
660: PetscCall(MatDenseGetLDA(X, &_ldx));
661: PetscCall(PetscBLASIntCast(_ldx, &ldx));
662: if (ldx < m) {
663: PetscCall(MatDenseGetArrayRead(B, &b));
664: PetscCall(PetscMalloc1(nrhs * m, &y));
665: if (ldb == m) {
666: PetscCall(PetscArraycpy(y, b, ldb * nrhs));
667: } else {
668: for (PetscInt j = 0; j < nrhs; j++) PetscCall(PetscArraycpy(&y[j * m], &b[j * ldb], m));
669: }
670: ldy = m;
671: PetscCall(MatDenseRestoreArrayRead(B, &b));
672: } else {
673: if (ldb == ldx) {
674: PetscCall(MatCopy(B, X, SAME_NONZERO_PATTERN));
675: PetscCall(MatDenseGetArray(X, &y));
676: } else {
677: PetscCall(MatDenseGetArray(X, &y));
678: PetscCall(MatDenseGetArrayRead(B, &b));
679: for (PetscInt j = 0; j < nrhs; j++) PetscCall(PetscArraycpy(&y[j * ldx], &b[j * ldb], m));
680: PetscCall(MatDenseRestoreArrayRead(B, &b));
681: }
682: ldy = ldx;
683: }
684: *_y = y;
685: *_ldy = ldy;
686: *_k = k;
687: *_m = m;
688: *_nrhs = nrhs;
689: PetscFunctionReturn(PETSC_SUCCESS);
690: }
692: static PetscErrorCode MatMatSolve_SeqDense_TearDown(Mat A, Mat B, Mat X, PetscScalar **_y, PetscBLASInt *_ldy, PetscBLASInt *_m, PetscBLASInt *_nrhs, PetscBLASInt *_k)
693: {
694: PetscScalar *y;
695: PetscInt _ldx;
696: PetscBLASInt k, ldy, nrhs, ldx = 0;
698: PetscFunctionBegin;
699: y = *_y;
700: *_y = NULL;
701: k = *_k;
702: ldy = *_ldy;
703: nrhs = *_nrhs;
704: PetscCall(MatDenseGetLDA(X, &_ldx));
705: PetscCall(PetscBLASIntCast(_ldx, &ldx));
706: if (ldx != ldy) {
707: PetscScalar *xv;
708: PetscCall(MatDenseGetArray(X, &xv));
709: for (PetscInt j = 0; j < nrhs; j++) PetscCall(PetscArraycpy(&xv[j * ldx], &y[j * ldy], k));
710: PetscCall(MatDenseRestoreArray(X, &xv));
711: PetscCall(PetscFree(y));
712: } else {
713: PetscCall(MatDenseRestoreArray(X, &y));
714: }
715: PetscFunctionReturn(PETSC_SUCCESS);
716: }
718: static PetscErrorCode MatMatSolve_SeqDense_LU(Mat A, Mat B, Mat X)
719: {
720: PetscScalar *y;
721: PetscBLASInt m, k, ldy, nrhs;
723: PetscFunctionBegin;
724: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
725: PetscCall(MatSolve_SeqDense_Internal_LU(A, y, ldy, m, nrhs, k, PETSC_FALSE));
726: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
727: PetscFunctionReturn(PETSC_SUCCESS);
728: }
730: static PetscErrorCode MatMatSolveTranspose_SeqDense_LU(Mat A, Mat B, Mat X)
731: {
732: PetscScalar *y;
733: PetscBLASInt m, k, ldy, nrhs;
735: PetscFunctionBegin;
736: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
737: PetscCall(MatSolve_SeqDense_Internal_LU(A, y, ldy, m, nrhs, k, PETSC_TRUE));
738: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
739: PetscFunctionReturn(PETSC_SUCCESS);
740: }
742: static PetscErrorCode MatMatSolve_SeqDense_Cholesky(Mat A, Mat B, Mat X)
743: {
744: PetscScalar *y;
745: PetscBLASInt m, k, ldy, nrhs;
747: PetscFunctionBegin;
748: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
749: PetscCall(MatSolve_SeqDense_Internal_Cholesky(A, y, ldy, m, nrhs, k, PETSC_FALSE));
750: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
751: PetscFunctionReturn(PETSC_SUCCESS);
752: }
754: static PetscErrorCode MatMatSolveTranspose_SeqDense_Cholesky(Mat A, Mat B, Mat X)
755: {
756: PetscScalar *y;
757: PetscBLASInt m, k, ldy, nrhs;
759: PetscFunctionBegin;
760: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
761: PetscCall(MatSolve_SeqDense_Internal_Cholesky(A, y, ldy, m, nrhs, k, PETSC_TRUE));
762: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
763: PetscFunctionReturn(PETSC_SUCCESS);
764: }
766: static PetscErrorCode MatMatSolve_SeqDense_QR(Mat A, Mat B, Mat X)
767: {
768: PetscScalar *y;
769: PetscBLASInt m, k, ldy, nrhs;
771: PetscFunctionBegin;
772: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
773: PetscCall(MatSolve_SeqDense_Internal_QR(A, y, ldy, m, nrhs, k));
774: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
775: PetscFunctionReturn(PETSC_SUCCESS);
776: }
778: static PetscErrorCode MatMatSolveTranspose_SeqDense_QR(Mat A, Mat B, Mat X)
779: {
780: PetscScalar *y;
781: PetscBLASInt m, k, ldy, nrhs;
783: PetscFunctionBegin;
784: PetscCall(MatMatSolve_SeqDense_SetUp(A, B, X, &y, &ldy, &m, &nrhs, &k));
785: PetscCall(MatSolveTranspose_SeqDense_Internal_QR(A, y, ldy, m, nrhs, k));
786: PetscCall(MatMatSolve_SeqDense_TearDown(A, B, X, &y, &ldy, &m, &nrhs, &k));
787: PetscFunctionReturn(PETSC_SUCCESS);
788: }
790: /* COMMENT: I have chosen to hide row permutation in the pivots,
791: rather than put it in the Mat->row slot.*/
792: PetscErrorCode MatLUFactor_SeqDense(Mat A, IS row, IS col, const MatFactorInfo *minfo)
793: {
794: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
795: PetscBLASInt n, m, info;
797: PetscFunctionBegin;
798: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
799: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
800: if (!mat->pivots) { PetscCall(PetscMalloc1(A->rmap->n, &mat->pivots)); }
801: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
802: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
803: PetscCallBLAS("LAPACKgetrf", LAPACKgetrf_(&m, &n, mat->v, &mat->lda, mat->pivots, &info));
804: PetscCall(PetscFPTrapPop());
806: PetscCheck(info >= 0, PETSC_COMM_SELF, PETSC_ERR_LIB, "Bad argument to LU factorization %d", (int)info);
807: PetscCheck(info <= 0, PETSC_COMM_SELF, PETSC_ERR_MAT_LU_ZRPVT, "Bad LU factorization %d", (int)info);
809: A->ops->solve = MatSolve_SeqDense_LU;
810: A->ops->matsolve = MatMatSolve_SeqDense_LU;
811: A->ops->solvetranspose = MatSolveTranspose_SeqDense_LU;
812: A->ops->matsolvetranspose = MatMatSolveTranspose_SeqDense_LU;
813: A->factortype = MAT_FACTOR_LU;
815: PetscCall(PetscFree(A->solvertype));
816: PetscCall(PetscStrallocpy(MATSOLVERPETSC, &A->solvertype));
818: PetscCall(PetscLogFlops((2.0 * A->cmap->n * A->cmap->n * A->cmap->n) / 3));
819: PetscFunctionReturn(PETSC_SUCCESS);
820: }
822: static PetscErrorCode MatLUFactorNumeric_SeqDense(Mat fact, Mat A, const MatFactorInfo *info_dummy)
823: {
824: MatFactorInfo info;
826: PetscFunctionBegin;
827: PetscCall(MatDuplicateNoCreate_SeqDense(fact, A, MAT_COPY_VALUES));
828: PetscUseTypeMethod(fact, lufactor, NULL, NULL, &info);
829: PetscFunctionReturn(PETSC_SUCCESS);
830: }
832: PetscErrorCode MatLUFactorSymbolic_SeqDense(Mat fact, Mat A, IS row, IS col, const MatFactorInfo *info)
833: {
834: PetscFunctionBegin;
835: fact->preallocated = PETSC_TRUE;
836: fact->assembled = PETSC_TRUE;
837: fact->ops->lufactornumeric = MatLUFactorNumeric_SeqDense;
838: PetscFunctionReturn(PETSC_SUCCESS);
839: }
841: /* Cholesky as L*L^T or L*D*L^T and the symmetric/hermitian complex variants */
842: PetscErrorCode MatCholeskyFactor_SeqDense(Mat A, IS perm, const MatFactorInfo *factinfo)
843: {
844: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
845: PetscBLASInt info, n;
847: PetscFunctionBegin;
848: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
849: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
850: if (A->spd == PETSC_BOOL3_TRUE) {
851: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
852: PetscCallBLAS("LAPACKpotrf", LAPACKpotrf_("L", &n, mat->v, &mat->lda, &info));
853: PetscCall(PetscFPTrapPop());
854: #if defined(PETSC_USE_COMPLEX)
855: } else if (A->hermitian == PETSC_BOOL3_TRUE) {
856: if (!mat->pivots) { PetscCall(PetscMalloc1(A->rmap->n, &mat->pivots)); }
857: if (!mat->fwork) {
858: PetscScalar dummy;
860: mat->lfwork = -1;
861: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
862: PetscCallBLAS("LAPACKhetrf", LAPACKhetrf_("L", &n, mat->v, &mat->lda, mat->pivots, &dummy, &mat->lfwork, &info));
863: PetscCall(PetscFPTrapPop());
864: mat->lfwork = (PetscInt)PetscRealPart(dummy);
865: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
866: }
867: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
868: PetscCallBLAS("LAPACKhetrf", LAPACKhetrf_("L", &n, mat->v, &mat->lda, mat->pivots, mat->fwork, &mat->lfwork, &info));
869: PetscCall(PetscFPTrapPop());
870: #endif
871: } else { /* symmetric case */
872: if (!mat->pivots) { PetscCall(PetscMalloc1(A->rmap->n, &mat->pivots)); }
873: if (!mat->fwork) {
874: PetscScalar dummy;
876: mat->lfwork = -1;
877: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
878: PetscCallBLAS("LAPACKsytrf", LAPACKsytrf_("L", &n, mat->v, &mat->lda, mat->pivots, &dummy, &mat->lfwork, &info));
879: PetscCall(PetscFPTrapPop());
880: mat->lfwork = (PetscInt)PetscRealPart(dummy);
881: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
882: }
883: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
884: PetscCallBLAS("LAPACKsytrf", LAPACKsytrf_("L", &n, mat->v, &mat->lda, mat->pivots, mat->fwork, &mat->lfwork, &info));
885: PetscCall(PetscFPTrapPop());
886: }
887: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_MAT_CH_ZRPVT, "Bad factorization: zero pivot in row %" PetscInt_FMT, (PetscInt)info - 1);
889: A->ops->solve = MatSolve_SeqDense_Cholesky;
890: A->ops->matsolve = MatMatSolve_SeqDense_Cholesky;
891: A->ops->solvetranspose = MatSolveTranspose_SeqDense_Cholesky;
892: A->ops->matsolvetranspose = MatMatSolveTranspose_SeqDense_Cholesky;
893: A->factortype = MAT_FACTOR_CHOLESKY;
895: PetscCall(PetscFree(A->solvertype));
896: PetscCall(PetscStrallocpy(MATSOLVERPETSC, &A->solvertype));
898: PetscCall(PetscLogFlops((1.0 * A->cmap->n * A->cmap->n * A->cmap->n) / 3.0));
899: PetscFunctionReturn(PETSC_SUCCESS);
900: }
902: static PetscErrorCode MatCholeskyFactorNumeric_SeqDense(Mat fact, Mat A, const MatFactorInfo *info_dummy)
903: {
904: MatFactorInfo info;
906: PetscFunctionBegin;
907: info.fill = 1.0;
909: PetscCall(MatDuplicateNoCreate_SeqDense(fact, A, MAT_COPY_VALUES));
910: PetscUseTypeMethod(fact, choleskyfactor, NULL, &info);
911: PetscFunctionReturn(PETSC_SUCCESS);
912: }
914: PetscErrorCode MatCholeskyFactorSymbolic_SeqDense(Mat fact, Mat A, IS row, const MatFactorInfo *info)
915: {
916: PetscFunctionBegin;
917: fact->assembled = PETSC_TRUE;
918: fact->preallocated = PETSC_TRUE;
919: fact->ops->choleskyfactornumeric = MatCholeskyFactorNumeric_SeqDense;
920: PetscFunctionReturn(PETSC_SUCCESS);
921: }
923: PetscErrorCode MatQRFactor_SeqDense(Mat A, IS col, const MatFactorInfo *minfo)
924: {
925: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
926: PetscBLASInt n, m, info, min, max;
928: PetscFunctionBegin;
929: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
930: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
931: max = PetscMax(m, n);
932: min = PetscMin(m, n);
933: if (!mat->tau) { PetscCall(PetscMalloc1(min, &mat->tau)); }
934: if (!mat->pivots) { PetscCall(PetscMalloc1(n, &mat->pivots)); }
935: if (!mat->qrrhs) PetscCall(MatCreateVecs(A, NULL, &(mat->qrrhs)));
936: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
937: if (!mat->fwork) {
938: PetscScalar dummy;
940: mat->lfwork = -1;
941: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
942: PetscCallBLAS("LAPACKgeqrf", LAPACKgeqrf_(&m, &n, mat->v, &mat->lda, mat->tau, &dummy, &mat->lfwork, &info));
943: PetscCall(PetscFPTrapPop());
944: mat->lfwork = (PetscInt)PetscRealPart(dummy);
945: PetscCall(PetscMalloc1(mat->lfwork, &mat->fwork));
946: }
947: PetscCall(PetscFPTrapPush(PETSC_FP_TRAP_OFF));
948: PetscCallBLAS("LAPACKgeqrf", LAPACKgeqrf_(&m, &n, mat->v, &mat->lda, mat->tau, mat->fwork, &mat->lfwork, &info));
949: PetscCall(PetscFPTrapPop());
950: PetscCheck(!info, PETSC_COMM_SELF, PETSC_ERR_LIB, "Bad argument to QR factorization %d", (int)info);
951: // TODO: try to estimate rank or test for and use geqp3 for rank revealing QR. For now just say rank is min of m and n
952: mat->rank = min;
954: A->ops->solve = MatSolve_SeqDense_QR;
955: A->ops->matsolve = MatMatSolve_SeqDense_QR;
956: A->factortype = MAT_FACTOR_QR;
957: if (m == n) {
958: A->ops->solvetranspose = MatSolveTranspose_SeqDense_QR;
959: A->ops->matsolvetranspose = MatMatSolveTranspose_SeqDense_QR;
960: }
962: PetscCall(PetscFree(A->solvertype));
963: PetscCall(PetscStrallocpy(MATSOLVERPETSC, &A->solvertype));
965: PetscCall(PetscLogFlops(2.0 * min * min * (max - min / 3.0)));
966: PetscFunctionReturn(PETSC_SUCCESS);
967: }
969: static PetscErrorCode MatQRFactorNumeric_SeqDense(Mat fact, Mat A, const MatFactorInfo *info_dummy)
970: {
971: MatFactorInfo info;
973: PetscFunctionBegin;
974: info.fill = 1.0;
976: PetscCall(MatDuplicateNoCreate_SeqDense(fact, A, MAT_COPY_VALUES));
977: PetscUseMethod(fact, "MatQRFactor_C", (Mat, IS, const MatFactorInfo *), (fact, NULL, &info));
978: PetscFunctionReturn(PETSC_SUCCESS);
979: }
981: PetscErrorCode MatQRFactorSymbolic_SeqDense(Mat fact, Mat A, IS row, const MatFactorInfo *info)
982: {
983: PetscFunctionBegin;
984: fact->assembled = PETSC_TRUE;
985: fact->preallocated = PETSC_TRUE;
986: PetscCall(PetscObjectComposeFunction((PetscObject)fact, "MatQRFactorNumeric_C", MatQRFactorNumeric_SeqDense));
987: PetscFunctionReturn(PETSC_SUCCESS);
988: }
990: /* uses LAPACK */
991: PETSC_INTERN PetscErrorCode MatGetFactor_seqdense_petsc(Mat A, MatFactorType ftype, Mat *fact)
992: {
993: PetscFunctionBegin;
994: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), fact));
995: PetscCall(MatSetSizes(*fact, A->rmap->n, A->cmap->n, A->rmap->n, A->cmap->n));
996: PetscCall(MatSetType(*fact, MATDENSE));
997: (*fact)->trivialsymbolic = PETSC_TRUE;
998: if (ftype == MAT_FACTOR_LU || ftype == MAT_FACTOR_ILU) {
999: (*fact)->ops->lufactorsymbolic = MatLUFactorSymbolic_SeqDense;
1000: (*fact)->ops->ilufactorsymbolic = MatLUFactorSymbolic_SeqDense;
1001: } else if (ftype == MAT_FACTOR_CHOLESKY || ftype == MAT_FACTOR_ICC) {
1002: (*fact)->ops->choleskyfactorsymbolic = MatCholeskyFactorSymbolic_SeqDense;
1003: } else if (ftype == MAT_FACTOR_QR) {
1004: PetscCall(PetscObjectComposeFunction((PetscObject)(*fact), "MatQRFactorSymbolic_C", MatQRFactorSymbolic_SeqDense));
1005: }
1006: (*fact)->factortype = ftype;
1008: PetscCall(PetscFree((*fact)->solvertype));
1009: PetscCall(PetscStrallocpy(MATSOLVERPETSC, &(*fact)->solvertype));
1010: PetscCall(PetscStrallocpy(MATORDERINGEXTERNAL, (char **)&(*fact)->preferredordering[MAT_FACTOR_LU]));
1011: PetscCall(PetscStrallocpy(MATORDERINGEXTERNAL, (char **)&(*fact)->preferredordering[MAT_FACTOR_ILU]));
1012: PetscCall(PetscStrallocpy(MATORDERINGEXTERNAL, (char **)&(*fact)->preferredordering[MAT_FACTOR_CHOLESKY]));
1013: PetscCall(PetscStrallocpy(MATORDERINGEXTERNAL, (char **)&(*fact)->preferredordering[MAT_FACTOR_ICC]));
1014: PetscFunctionReturn(PETSC_SUCCESS);
1015: }
1017: static PetscErrorCode MatSOR_SeqDense(Mat A, Vec bb, PetscReal omega, MatSORType flag, PetscReal shift, PetscInt its, PetscInt lits, Vec xx)
1018: {
1019: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1020: PetscScalar *x, *v = mat->v, zero = 0.0, xt;
1021: const PetscScalar *b;
1022: PetscInt m = A->rmap->n, i;
1023: PetscBLASInt o = 1, bm = 0;
1025: PetscFunctionBegin;
1026: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1027: PetscCheck(A->offloadmask != PETSC_OFFLOAD_GPU, PETSC_COMM_SELF, PETSC_ERR_SUP, "Not implemented");
1028: #endif
1029: if (shift == -1) shift = 0.0; /* negative shift indicates do not error on zero diagonal; this code never zeros on zero diagonal */
1030: PetscCall(PetscBLASIntCast(m, &bm));
1031: if (flag & SOR_ZERO_INITIAL_GUESS) {
1032: /* this is a hack fix, should have another version without the second BLASdotu */
1033: PetscCall(VecSet(xx, zero));
1034: }
1035: PetscCall(VecGetArray(xx, &x));
1036: PetscCall(VecGetArrayRead(bb, &b));
1037: its = its * lits;
1038: PetscCheck(its > 0, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONG, "Relaxation requires global its %" PetscInt_FMT " and local its %" PetscInt_FMT " both positive", its, lits);
1039: while (its--) {
1040: if (flag & SOR_FORWARD_SWEEP || flag & SOR_LOCAL_FORWARD_SWEEP) {
1041: for (i = 0; i < m; i++) {
1042: PetscCallBLAS("BLASdotu", xt = b[i] - BLASdotu_(&bm, v + i, &bm, x, &o));
1043: x[i] = (1. - omega) * x[i] + omega * (xt + v[i + i * m] * x[i]) / (v[i + i * m] + shift);
1044: }
1045: }
1046: if (flag & SOR_BACKWARD_SWEEP || flag & SOR_LOCAL_BACKWARD_SWEEP) {
1047: for (i = m - 1; i >= 0; i--) {
1048: PetscCallBLAS("BLASdotu", xt = b[i] - BLASdotu_(&bm, v + i, &bm, x, &o));
1049: x[i] = (1. - omega) * x[i] + omega * (xt + v[i + i * m] * x[i]) / (v[i + i * m] + shift);
1050: }
1051: }
1052: }
1053: PetscCall(VecRestoreArrayRead(bb, &b));
1054: PetscCall(VecRestoreArray(xx, &x));
1055: PetscFunctionReturn(PETSC_SUCCESS);
1056: }
1058: PetscErrorCode MatMultTranspose_SeqDense(Mat A, Vec xx, Vec yy)
1059: {
1060: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1061: const PetscScalar *v = mat->v, *x;
1062: PetscScalar *y;
1063: PetscBLASInt m, n, _One = 1;
1064: PetscScalar _DOne = 1.0, _DZero = 0.0;
1066: PetscFunctionBegin;
1067: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
1068: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
1069: PetscCall(VecGetArrayRead(xx, &x));
1070: PetscCall(VecGetArrayWrite(yy, &y));
1071: if (!A->rmap->n || !A->cmap->n) {
1072: PetscBLASInt i;
1073: for (i = 0; i < n; i++) y[i] = 0.0;
1074: } else {
1075: PetscCallBLAS("BLASgemv", BLASgemv_("T", &m, &n, &_DOne, v, &mat->lda, x, &_One, &_DZero, y, &_One));
1076: PetscCall(PetscLogFlops(2.0 * A->rmap->n * A->cmap->n - A->cmap->n));
1077: }
1078: PetscCall(VecRestoreArrayRead(xx, &x));
1079: PetscCall(VecRestoreArrayWrite(yy, &y));
1080: PetscFunctionReturn(PETSC_SUCCESS);
1081: }
1083: PetscErrorCode MatMult_SeqDense(Mat A, Vec xx, Vec yy)
1084: {
1085: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1086: PetscScalar *y, _DOne = 1.0, _DZero = 0.0;
1087: PetscBLASInt m, n, _One = 1;
1088: const PetscScalar *v = mat->v, *x;
1090: PetscFunctionBegin;
1091: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
1092: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
1093: PetscCall(VecGetArrayRead(xx, &x));
1094: PetscCall(VecGetArrayWrite(yy, &y));
1095: if (!A->rmap->n || !A->cmap->n) {
1096: PetscBLASInt i;
1097: for (i = 0; i < m; i++) y[i] = 0.0;
1098: } else {
1099: PetscCallBLAS("BLASgemv", BLASgemv_("N", &m, &n, &_DOne, v, &(mat->lda), x, &_One, &_DZero, y, &_One));
1100: PetscCall(PetscLogFlops(2.0 * A->rmap->n * A->cmap->n - A->rmap->n));
1101: }
1102: PetscCall(VecRestoreArrayRead(xx, &x));
1103: PetscCall(VecRestoreArrayWrite(yy, &y));
1104: PetscFunctionReturn(PETSC_SUCCESS);
1105: }
1107: PetscErrorCode MatMultAdd_SeqDense(Mat A, Vec xx, Vec zz, Vec yy)
1108: {
1109: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1110: const PetscScalar *v = mat->v, *x;
1111: PetscScalar *y, _DOne = 1.0;
1112: PetscBLASInt m, n, _One = 1;
1114: PetscFunctionBegin;
1115: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
1116: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
1117: PetscCall(VecCopy(zz, yy));
1118: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
1119: PetscCall(VecGetArrayRead(xx, &x));
1120: PetscCall(VecGetArray(yy, &y));
1121: PetscCallBLAS("BLASgemv", BLASgemv_("N", &m, &n, &_DOne, v, &(mat->lda), x, &_One, &_DOne, y, &_One));
1122: PetscCall(VecRestoreArrayRead(xx, &x));
1123: PetscCall(VecRestoreArray(yy, &y));
1124: PetscCall(PetscLogFlops(2.0 * A->rmap->n * A->cmap->n));
1125: PetscFunctionReturn(PETSC_SUCCESS);
1126: }
1128: PetscErrorCode MatMultTransposeAdd_SeqDense(Mat A, Vec xx, Vec zz, Vec yy)
1129: {
1130: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1131: const PetscScalar *v = mat->v, *x;
1132: PetscScalar *y;
1133: PetscBLASInt m, n, _One = 1;
1134: PetscScalar _DOne = 1.0;
1136: PetscFunctionBegin;
1137: PetscCall(PetscBLASIntCast(A->rmap->n, &m));
1138: PetscCall(PetscBLASIntCast(A->cmap->n, &n));
1139: PetscCall(VecCopy(zz, yy));
1140: if (!A->rmap->n || !A->cmap->n) PetscFunctionReturn(PETSC_SUCCESS);
1141: PetscCall(VecGetArrayRead(xx, &x));
1142: PetscCall(VecGetArray(yy, &y));
1143: PetscCallBLAS("BLASgemv", BLASgemv_("T", &m, &n, &_DOne, v, &(mat->lda), x, &_One, &_DOne, y, &_One));
1144: PetscCall(VecRestoreArrayRead(xx, &x));
1145: PetscCall(VecRestoreArray(yy, &y));
1146: PetscCall(PetscLogFlops(2.0 * A->rmap->n * A->cmap->n));
1147: PetscFunctionReturn(PETSC_SUCCESS);
1148: }
1150: static PetscErrorCode MatGetRow_SeqDense(Mat A, PetscInt row, PetscInt *ncols, PetscInt **cols, PetscScalar **vals)
1151: {
1152: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1153: PetscInt i;
1155: PetscFunctionBegin;
1156: *ncols = A->cmap->n;
1157: if (cols) {
1158: PetscCall(PetscMalloc1(A->cmap->n, cols));
1159: for (i = 0; i < A->cmap->n; i++) (*cols)[i] = i;
1160: }
1161: if (vals) {
1162: const PetscScalar *v;
1164: PetscCall(MatDenseGetArrayRead(A, &v));
1165: PetscCall(PetscMalloc1(A->cmap->n, vals));
1166: v += row;
1167: for (i = 0; i < A->cmap->n; i++) {
1168: (*vals)[i] = *v;
1169: v += mat->lda;
1170: }
1171: PetscCall(MatDenseRestoreArrayRead(A, &v));
1172: }
1173: PetscFunctionReturn(PETSC_SUCCESS);
1174: }
1176: static PetscErrorCode MatRestoreRow_SeqDense(Mat A, PetscInt row, PetscInt *ncols, PetscInt **cols, PetscScalar **vals)
1177: {
1178: PetscFunctionBegin;
1179: if (ncols) *ncols = 0;
1180: if (cols) PetscCall(PetscFree(*cols));
1181: if (vals) PetscCall(PetscFree(*vals));
1182: PetscFunctionReturn(PETSC_SUCCESS);
1183: }
1185: static PetscErrorCode MatSetValues_SeqDense(Mat A, PetscInt m, const PetscInt indexm[], PetscInt n, const PetscInt indexn[], const PetscScalar v[], InsertMode addv)
1186: {
1187: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1188: PetscScalar *av;
1189: PetscInt i, j, idx = 0;
1190: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1191: PetscOffloadMask oldf;
1192: #endif
1194: PetscFunctionBegin;
1195: PetscCall(MatDenseGetArray(A, &av));
1196: if (!mat->roworiented) {
1197: if (addv == INSERT_VALUES) {
1198: for (j = 0; j < n; j++) {
1199: if (indexn[j] < 0) {
1200: idx += m;
1201: continue;
1202: }
1203: PetscCheck(indexn[j] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Column too large: col %" PetscInt_FMT " max %" PetscInt_FMT, indexn[j], A->cmap->n - 1);
1204: for (i = 0; i < m; i++) {
1205: if (indexm[i] < 0) {
1206: idx++;
1207: continue;
1208: }
1209: PetscCheck(indexm[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row too large: row %" PetscInt_FMT " max %" PetscInt_FMT, indexm[i], A->rmap->n - 1);
1210: av[indexn[j] * mat->lda + indexm[i]] = v[idx++];
1211: }
1212: }
1213: } else {
1214: for (j = 0; j < n; j++) {
1215: if (indexn[j] < 0) {
1216: idx += m;
1217: continue;
1218: }
1219: PetscCheck(indexn[j] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Column too large: col %" PetscInt_FMT " max %" PetscInt_FMT, indexn[j], A->cmap->n - 1);
1220: for (i = 0; i < m; i++) {
1221: if (indexm[i] < 0) {
1222: idx++;
1223: continue;
1224: }
1225: PetscCheck(indexm[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row too large: row %" PetscInt_FMT " max %" PetscInt_FMT, indexm[i], A->rmap->n - 1);
1226: av[indexn[j] * mat->lda + indexm[i]] += v[idx++];
1227: }
1228: }
1229: }
1230: } else {
1231: if (addv == INSERT_VALUES) {
1232: for (i = 0; i < m; i++) {
1233: if (indexm[i] < 0) {
1234: idx += n;
1235: continue;
1236: }
1237: PetscCheck(indexm[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row too large: row %" PetscInt_FMT " max %" PetscInt_FMT, indexm[i], A->rmap->n - 1);
1238: for (j = 0; j < n; j++) {
1239: if (indexn[j] < 0) {
1240: idx++;
1241: continue;
1242: }
1243: PetscCheck(indexn[j] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Column too large: col %" PetscInt_FMT " max %" PetscInt_FMT, indexn[j], A->cmap->n - 1);
1244: av[indexn[j] * mat->lda + indexm[i]] = v[idx++];
1245: }
1246: }
1247: } else {
1248: for (i = 0; i < m; i++) {
1249: if (indexm[i] < 0) {
1250: idx += n;
1251: continue;
1252: }
1253: PetscCheck(indexm[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row too large: row %" PetscInt_FMT " max %" PetscInt_FMT, indexm[i], A->rmap->n - 1);
1254: for (j = 0; j < n; j++) {
1255: if (indexn[j] < 0) {
1256: idx++;
1257: continue;
1258: }
1259: PetscCheck(indexn[j] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Column too large: col %" PetscInt_FMT " max %" PetscInt_FMT, indexn[j], A->cmap->n - 1);
1260: av[indexn[j] * mat->lda + indexm[i]] += v[idx++];
1261: }
1262: }
1263: }
1264: }
1265: /* hack to prevent unneeded copy to the GPU while returning the array */
1266: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1267: oldf = A->offloadmask;
1268: A->offloadmask = PETSC_OFFLOAD_GPU;
1269: #endif
1270: PetscCall(MatDenseRestoreArray(A, &av));
1271: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1272: A->offloadmask = (oldf == PETSC_OFFLOAD_UNALLOCATED ? PETSC_OFFLOAD_UNALLOCATED : PETSC_OFFLOAD_CPU);
1273: #endif
1274: PetscFunctionReturn(PETSC_SUCCESS);
1275: }
1277: static PetscErrorCode MatGetValues_SeqDense(Mat A, PetscInt m, const PetscInt indexm[], PetscInt n, const PetscInt indexn[], PetscScalar v[])
1278: {
1279: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1280: const PetscScalar *vv;
1281: PetscInt i, j;
1283: PetscFunctionBegin;
1284: PetscCall(MatDenseGetArrayRead(A, &vv));
1285: /* row-oriented output */
1286: for (i = 0; i < m; i++) {
1287: if (indexm[i] < 0) {
1288: v += n;
1289: continue;
1290: }
1291: PetscCheck(indexm[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row %" PetscInt_FMT " requested larger than number rows %" PetscInt_FMT, indexm[i], A->rmap->n);
1292: for (j = 0; j < n; j++) {
1293: if (indexn[j] < 0) {
1294: v++;
1295: continue;
1296: }
1297: PetscCheck(indexn[j] < A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Column %" PetscInt_FMT " requested larger than number columns %" PetscInt_FMT, indexn[j], A->cmap->n);
1298: *v++ = vv[indexn[j] * mat->lda + indexm[i]];
1299: }
1300: }
1301: PetscCall(MatDenseRestoreArrayRead(A, &vv));
1302: PetscFunctionReturn(PETSC_SUCCESS);
1303: }
1305: PetscErrorCode MatView_Dense_Binary(Mat mat, PetscViewer viewer)
1306: {
1307: PetscBool skipHeader;
1308: PetscViewerFormat format;
1309: PetscInt header[4], M, N, m, lda, i, j, k;
1310: const PetscScalar *v;
1311: PetscScalar *vwork;
1313: PetscFunctionBegin;
1314: PetscCall(PetscViewerSetUp(viewer));
1315: PetscCall(PetscViewerBinaryGetSkipHeader(viewer, &skipHeader));
1316: PetscCall(PetscViewerGetFormat(viewer, &format));
1317: if (skipHeader) format = PETSC_VIEWER_NATIVE;
1319: PetscCall(MatGetSize(mat, &M, &N));
1321: /* write matrix header */
1322: header[0] = MAT_FILE_CLASSID;
1323: header[1] = M;
1324: header[2] = N;
1325: header[3] = (format == PETSC_VIEWER_NATIVE) ? MATRIX_BINARY_FORMAT_DENSE : M * N;
1326: if (!skipHeader) PetscCall(PetscViewerBinaryWrite(viewer, header, 4, PETSC_INT));
1328: PetscCall(MatGetLocalSize(mat, &m, NULL));
1329: if (format != PETSC_VIEWER_NATIVE) {
1330: PetscInt nnz = m * N, *iwork;
1331: /* store row lengths for each row */
1332: PetscCall(PetscMalloc1(nnz, &iwork));
1333: for (i = 0; i < m; i++) iwork[i] = N;
1334: PetscCall(PetscViewerBinaryWriteAll(viewer, iwork, m, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_INT));
1335: /* store column indices (zero start index) */
1336: for (k = 0, i = 0; i < m; i++)
1337: for (j = 0; j < N; j++, k++) iwork[k] = j;
1338: PetscCall(PetscViewerBinaryWriteAll(viewer, iwork, nnz, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_INT));
1339: PetscCall(PetscFree(iwork));
1340: }
1341: /* store matrix values as a dense matrix in row major order */
1342: PetscCall(PetscMalloc1(m * N, &vwork));
1343: PetscCall(MatDenseGetArrayRead(mat, &v));
1344: PetscCall(MatDenseGetLDA(mat, &lda));
1345: for (k = 0, i = 0; i < m; i++)
1346: for (j = 0; j < N; j++, k++) vwork[k] = v[i + lda * j];
1347: PetscCall(MatDenseRestoreArrayRead(mat, &v));
1348: PetscCall(PetscViewerBinaryWriteAll(viewer, vwork, m * N, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_SCALAR));
1349: PetscCall(PetscFree(vwork));
1350: PetscFunctionReturn(PETSC_SUCCESS);
1351: }
1353: PetscErrorCode MatLoad_Dense_Binary(Mat mat, PetscViewer viewer)
1354: {
1355: PetscBool skipHeader;
1356: PetscInt header[4], M, N, m, nz, lda, i, j, k;
1357: PetscInt rows, cols;
1358: PetscScalar *v, *vwork;
1360: PetscFunctionBegin;
1361: PetscCall(PetscViewerSetUp(viewer));
1362: PetscCall(PetscViewerBinaryGetSkipHeader(viewer, &skipHeader));
1364: if (!skipHeader) {
1365: PetscCall(PetscViewerBinaryRead(viewer, header, 4, NULL, PETSC_INT));
1366: PetscCheck(header[0] == MAT_FILE_CLASSID, PetscObjectComm((PetscObject)viewer), PETSC_ERR_FILE_UNEXPECTED, "Not a matrix object in file");
1367: M = header[1];
1368: N = header[2];
1369: PetscCheck(M >= 0, PetscObjectComm((PetscObject)viewer), PETSC_ERR_FILE_UNEXPECTED, "Matrix row size (%" PetscInt_FMT ") in file is negative", M);
1370: PetscCheck(N >= 0, PetscObjectComm((PetscObject)viewer), PETSC_ERR_FILE_UNEXPECTED, "Matrix column size (%" PetscInt_FMT ") in file is negative", N);
1371: nz = header[3];
1372: PetscCheck(nz == MATRIX_BINARY_FORMAT_DENSE || nz >= 0, PetscObjectComm((PetscObject)viewer), PETSC_ERR_FILE_UNEXPECTED, "Unknown matrix format %" PetscInt_FMT " in file", nz);
1373: } else {
1374: PetscCall(MatGetSize(mat, &M, &N));
1375: PetscCheck(M >= 0 && N >= 0, PETSC_COMM_SELF, PETSC_ERR_USER, "Matrix binary file header was skipped, thus the user must specify the global sizes of input matrix");
1376: nz = MATRIX_BINARY_FORMAT_DENSE;
1377: }
1379: /* setup global sizes if not set */
1380: if (mat->rmap->N < 0) mat->rmap->N = M;
1381: if (mat->cmap->N < 0) mat->cmap->N = N;
1382: PetscCall(MatSetUp(mat));
1383: /* check if global sizes are correct */
1384: PetscCall(MatGetSize(mat, &rows, &cols));
1385: PetscCheck(M == rows && N == cols, PetscObjectComm((PetscObject)viewer), PETSC_ERR_FILE_UNEXPECTED, "Matrix in file of different sizes (%" PetscInt_FMT ", %" PetscInt_FMT ") than the input matrix (%" PetscInt_FMT ", %" PetscInt_FMT ")", M, N, rows, cols);
1387: PetscCall(MatGetSize(mat, NULL, &N));
1388: PetscCall(MatGetLocalSize(mat, &m, NULL));
1389: PetscCall(MatDenseGetArray(mat, &v));
1390: PetscCall(MatDenseGetLDA(mat, &lda));
1391: if (nz == MATRIX_BINARY_FORMAT_DENSE) { /* matrix in file is dense format */
1392: PetscInt nnz = m * N;
1393: /* read in matrix values */
1394: PetscCall(PetscMalloc1(nnz, &vwork));
1395: PetscCall(PetscViewerBinaryReadAll(viewer, vwork, nnz, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_SCALAR));
1396: /* store values in column major order */
1397: for (j = 0; j < N; j++)
1398: for (i = 0; i < m; i++) v[i + lda * j] = vwork[i * N + j];
1399: PetscCall(PetscFree(vwork));
1400: } else { /* matrix in file is sparse format */
1401: PetscInt nnz = 0, *rlens, *icols;
1402: /* read in row lengths */
1403: PetscCall(PetscMalloc1(m, &rlens));
1404: PetscCall(PetscViewerBinaryReadAll(viewer, rlens, m, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_INT));
1405: for (i = 0; i < m; i++) nnz += rlens[i];
1406: /* read in column indices and values */
1407: PetscCall(PetscMalloc2(nnz, &icols, nnz, &vwork));
1408: PetscCall(PetscViewerBinaryReadAll(viewer, icols, nnz, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_INT));
1409: PetscCall(PetscViewerBinaryReadAll(viewer, vwork, nnz, PETSC_DETERMINE, PETSC_DETERMINE, PETSC_SCALAR));
1410: /* store values in column major order */
1411: for (k = 0, i = 0; i < m; i++)
1412: for (j = 0; j < rlens[i]; j++, k++) v[i + lda * icols[k]] = vwork[k];
1413: PetscCall(PetscFree(rlens));
1414: PetscCall(PetscFree2(icols, vwork));
1415: }
1416: PetscCall(MatDenseRestoreArray(mat, &v));
1417: PetscCall(MatAssemblyBegin(mat, MAT_FINAL_ASSEMBLY));
1418: PetscCall(MatAssemblyEnd(mat, MAT_FINAL_ASSEMBLY));
1419: PetscFunctionReturn(PETSC_SUCCESS);
1420: }
1422: PetscErrorCode MatLoad_SeqDense(Mat newMat, PetscViewer viewer)
1423: {
1424: PetscBool isbinary, ishdf5;
1426: PetscFunctionBegin;
1429: /* force binary viewer to load .info file if it has not yet done so */
1430: PetscCall(PetscViewerSetUp(viewer));
1431: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERBINARY, &isbinary));
1432: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERHDF5, &ishdf5));
1433: if (isbinary) {
1434: PetscCall(MatLoad_Dense_Binary(newMat, viewer));
1435: } else if (ishdf5) {
1436: #if defined(PETSC_HAVE_HDF5)
1437: PetscCall(MatLoad_Dense_HDF5(newMat, viewer));
1438: #else
1439: SETERRQ(PetscObjectComm((PetscObject)newMat), PETSC_ERR_SUP, "HDF5 not supported in this build.\nPlease reconfigure using --download-hdf5");
1440: #endif
1441: } else {
1442: SETERRQ(PetscObjectComm((PetscObject)newMat), PETSC_ERR_SUP, "Viewer type %s not yet supported for reading %s matrices", ((PetscObject)viewer)->type_name, ((PetscObject)newMat)->type_name);
1443: }
1444: PetscFunctionReturn(PETSC_SUCCESS);
1445: }
1447: static PetscErrorCode MatView_SeqDense_ASCII(Mat A, PetscViewer viewer)
1448: {
1449: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
1450: PetscInt i, j;
1451: const char *name;
1452: PetscScalar *v, *av;
1453: PetscViewerFormat format;
1454: #if defined(PETSC_USE_COMPLEX)
1455: PetscBool allreal = PETSC_TRUE;
1456: #endif
1458: PetscFunctionBegin;
1459: PetscCall(MatDenseGetArrayRead(A, (const PetscScalar **)&av));
1460: PetscCall(PetscViewerGetFormat(viewer, &format));
1461: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
1462: PetscFunctionReturn(PETSC_SUCCESS); /* do nothing for now */
1463: } else if (format == PETSC_VIEWER_ASCII_COMMON) {
1464: PetscCall(PetscViewerASCIIUseTabs(viewer, PETSC_FALSE));
1465: for (i = 0; i < A->rmap->n; i++) {
1466: v = av + i;
1467: PetscCall(PetscViewerASCIIPrintf(viewer, "row %" PetscInt_FMT ":", i));
1468: for (j = 0; j < A->cmap->n; j++) {
1469: #if defined(PETSC_USE_COMPLEX)
1470: if (PetscRealPart(*v) != 0.0 && PetscImaginaryPart(*v) != 0.0) {
1471: PetscCall(PetscViewerASCIIPrintf(viewer, " (%" PetscInt_FMT ", %g + %g i) ", j, (double)PetscRealPart(*v), (double)PetscImaginaryPart(*v)));
1472: } else if (PetscRealPart(*v)) {
1473: PetscCall(PetscViewerASCIIPrintf(viewer, " (%" PetscInt_FMT ", %g) ", j, (double)PetscRealPart(*v)));
1474: }
1475: #else
1476: if (*v) PetscCall(PetscViewerASCIIPrintf(viewer, " (%" PetscInt_FMT ", %g) ", j, (double)*v));
1477: #endif
1478: v += a->lda;
1479: }
1480: PetscCall(PetscViewerASCIIPrintf(viewer, "\n"));
1481: }
1482: PetscCall(PetscViewerASCIIUseTabs(viewer, PETSC_TRUE));
1483: } else {
1484: PetscCall(PetscViewerASCIIUseTabs(viewer, PETSC_FALSE));
1485: #if defined(PETSC_USE_COMPLEX)
1486: /* determine if matrix has all real values */
1487: for (j = 0; j < A->cmap->n; j++) {
1488: v = av + j * a->lda;
1489: for (i = 0; i < A->rmap->n; i++) {
1490: if (PetscImaginaryPart(v[i])) {
1491: allreal = PETSC_FALSE;
1492: break;
1493: }
1494: }
1495: }
1496: #endif
1497: if (format == PETSC_VIEWER_ASCII_MATLAB) {
1498: PetscCall(PetscObjectGetName((PetscObject)A, &name));
1499: PetscCall(PetscViewerASCIIPrintf(viewer, "%% Size = %" PetscInt_FMT " %" PetscInt_FMT " \n", A->rmap->n, A->cmap->n));
1500: PetscCall(PetscViewerASCIIPrintf(viewer, "%s = zeros(%" PetscInt_FMT ",%" PetscInt_FMT ");\n", name, A->rmap->n, A->cmap->n));
1501: PetscCall(PetscViewerASCIIPrintf(viewer, "%s = [\n", name));
1502: }
1504: for (i = 0; i < A->rmap->n; i++) {
1505: v = av + i;
1506: for (j = 0; j < A->cmap->n; j++) {
1507: #if defined(PETSC_USE_COMPLEX)
1508: if (allreal) {
1509: PetscCall(PetscViewerASCIIPrintf(viewer, "%18.16e ", (double)PetscRealPart(*v)));
1510: } else {
1511: PetscCall(PetscViewerASCIIPrintf(viewer, "%18.16e + %18.16ei ", (double)PetscRealPart(*v), (double)PetscImaginaryPart(*v)));
1512: }
1513: #else
1514: PetscCall(PetscViewerASCIIPrintf(viewer, "%18.16e ", (double)*v));
1515: #endif
1516: v += a->lda;
1517: }
1518: PetscCall(PetscViewerASCIIPrintf(viewer, "\n"));
1519: }
1520: if (format == PETSC_VIEWER_ASCII_MATLAB) PetscCall(PetscViewerASCIIPrintf(viewer, "];\n"));
1521: PetscCall(PetscViewerASCIIUseTabs(viewer, PETSC_TRUE));
1522: }
1523: PetscCall(MatDenseRestoreArrayRead(A, (const PetscScalar **)&av));
1524: PetscCall(PetscViewerFlush(viewer));
1525: PetscFunctionReturn(PETSC_SUCCESS);
1526: }
1528: #include <petscdraw.h>
1529: static PetscErrorCode MatView_SeqDense_Draw_Zoom(PetscDraw draw, void *Aa)
1530: {
1531: Mat A = (Mat)Aa;
1532: PetscInt m = A->rmap->n, n = A->cmap->n, i, j;
1533: int color = PETSC_DRAW_WHITE;
1534: const PetscScalar *v;
1535: PetscViewer viewer;
1536: PetscReal xl, yl, xr, yr, x_l, x_r, y_l, y_r;
1537: PetscViewerFormat format;
1539: PetscFunctionBegin;
1540: PetscCall(PetscObjectQuery((PetscObject)A, "Zoomviewer", (PetscObject *)&viewer));
1541: PetscCall(PetscViewerGetFormat(viewer, &format));
1542: PetscCall(PetscDrawGetCoordinates(draw, &xl, &yl, &xr, &yr));
1544: /* Loop over matrix elements drawing boxes */
1545: PetscCall(MatDenseGetArrayRead(A, &v));
1546: if (format != PETSC_VIEWER_DRAW_CONTOUR) {
1547: PetscDrawCollectiveBegin(draw);
1548: /* Blue for negative and Red for positive */
1549: for (j = 0; j < n; j++) {
1550: x_l = j;
1551: x_r = x_l + 1.0;
1552: for (i = 0; i < m; i++) {
1553: y_l = m - i - 1.0;
1554: y_r = y_l + 1.0;
1555: if (PetscRealPart(v[j * m + i]) > 0.) color = PETSC_DRAW_RED;
1556: else if (PetscRealPart(v[j * m + i]) < 0.) color = PETSC_DRAW_BLUE;
1557: else continue;
1558: PetscCall(PetscDrawRectangle(draw, x_l, y_l, x_r, y_r, color, color, color, color));
1559: }
1560: }
1561: PetscDrawCollectiveEnd(draw);
1562: } else {
1563: /* use contour shading to indicate magnitude of values */
1564: /* first determine max of all nonzero values */
1565: PetscReal minv = 0.0, maxv = 0.0;
1566: PetscDraw popup;
1568: for (i = 0; i < m * n; i++) {
1569: if (PetscAbsScalar(v[i]) > maxv) maxv = PetscAbsScalar(v[i]);
1570: }
1571: if (minv >= maxv) maxv = minv + PETSC_SMALL;
1572: PetscCall(PetscDrawGetPopup(draw, &popup));
1573: PetscCall(PetscDrawScalePopup(popup, minv, maxv));
1575: PetscDrawCollectiveBegin(draw);
1576: for (j = 0; j < n; j++) {
1577: x_l = j;
1578: x_r = x_l + 1.0;
1579: for (i = 0; i < m; i++) {
1580: y_l = m - i - 1.0;
1581: y_r = y_l + 1.0;
1582: color = PetscDrawRealToColor(PetscAbsScalar(v[j * m + i]), minv, maxv);
1583: PetscCall(PetscDrawRectangle(draw, x_l, y_l, x_r, y_r, color, color, color, color));
1584: }
1585: }
1586: PetscDrawCollectiveEnd(draw);
1587: }
1588: PetscCall(MatDenseRestoreArrayRead(A, &v));
1589: PetscFunctionReturn(PETSC_SUCCESS);
1590: }
1592: static PetscErrorCode MatView_SeqDense_Draw(Mat A, PetscViewer viewer)
1593: {
1594: PetscDraw draw;
1595: PetscBool isnull;
1596: PetscReal xr, yr, xl, yl, h, w;
1598: PetscFunctionBegin;
1599: PetscCall(PetscViewerDrawGetDraw(viewer, 0, &draw));
1600: PetscCall(PetscDrawIsNull(draw, &isnull));
1601: if (isnull) PetscFunctionReturn(PETSC_SUCCESS);
1603: xr = A->cmap->n;
1604: yr = A->rmap->n;
1605: h = yr / 10.0;
1606: w = xr / 10.0;
1607: xr += w;
1608: yr += h;
1609: xl = -w;
1610: yl = -h;
1611: PetscCall(PetscDrawSetCoordinates(draw, xl, yl, xr, yr));
1612: PetscCall(PetscObjectCompose((PetscObject)A, "Zoomviewer", (PetscObject)viewer));
1613: PetscCall(PetscDrawZoom(draw, MatView_SeqDense_Draw_Zoom, A));
1614: PetscCall(PetscObjectCompose((PetscObject)A, "Zoomviewer", NULL));
1615: PetscCall(PetscDrawSave(draw));
1616: PetscFunctionReturn(PETSC_SUCCESS);
1617: }
1619: PetscErrorCode MatView_SeqDense(Mat A, PetscViewer viewer)
1620: {
1621: PetscBool iascii, isbinary, isdraw;
1623: PetscFunctionBegin;
1624: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERASCII, &iascii));
1625: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERBINARY, &isbinary));
1626: PetscCall(PetscObjectTypeCompare((PetscObject)viewer, PETSCVIEWERDRAW, &isdraw));
1627: if (iascii) PetscCall(MatView_SeqDense_ASCII(A, viewer));
1628: else if (isbinary) PetscCall(MatView_Dense_Binary(A, viewer));
1629: else if (isdraw) PetscCall(MatView_SeqDense_Draw(A, viewer));
1630: PetscFunctionReturn(PETSC_SUCCESS);
1631: }
1633: static PetscErrorCode MatDensePlaceArray_SeqDense(Mat A, const PetscScalar *array)
1634: {
1635: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
1637: PetscFunctionBegin;
1638: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
1639: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
1640: PetscCheck(!a->unplacedarray, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreArray() first");
1641: a->unplacedarray = a->v;
1642: a->unplaced_user_alloc = a->user_alloc;
1643: a->v = (PetscScalar *)array;
1644: a->user_alloc = PETSC_TRUE;
1645: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1646: A->offloadmask = PETSC_OFFLOAD_CPU;
1647: #endif
1648: PetscFunctionReturn(PETSC_SUCCESS);
1649: }
1651: static PetscErrorCode MatDenseResetArray_SeqDense(Mat A)
1652: {
1653: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
1655: PetscFunctionBegin;
1656: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
1657: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
1658: a->v = a->unplacedarray;
1659: a->user_alloc = a->unplaced_user_alloc;
1660: a->unplacedarray = NULL;
1661: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1662: A->offloadmask = PETSC_OFFLOAD_CPU;
1663: #endif
1664: PetscFunctionReturn(PETSC_SUCCESS);
1665: }
1667: static PetscErrorCode MatDenseReplaceArray_SeqDense(Mat A, const PetscScalar *array)
1668: {
1669: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
1671: PetscFunctionBegin;
1672: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
1673: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
1674: if (!a->user_alloc) PetscCall(PetscFree(a->v));
1675: a->v = (PetscScalar *)array;
1676: a->user_alloc = PETSC_FALSE;
1677: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
1678: A->offloadmask = PETSC_OFFLOAD_CPU;
1679: #endif
1680: PetscFunctionReturn(PETSC_SUCCESS);
1681: }
1683: PetscErrorCode MatDestroy_SeqDense(Mat mat)
1684: {
1685: Mat_SeqDense *l = (Mat_SeqDense *)mat->data;
1687: PetscFunctionBegin;
1688: #if defined(PETSC_USE_LOG)
1689: PetscCall(PetscLogObjectState((PetscObject)mat, "Rows %" PetscInt_FMT " Cols %" PetscInt_FMT, mat->rmap->n, mat->cmap->n));
1690: #endif
1691: PetscCall(VecDestroy(&(l->qrrhs)));
1692: PetscCall(PetscFree(l->tau));
1693: PetscCall(PetscFree(l->pivots));
1694: PetscCall(PetscFree(l->fwork));
1695: if (!l->user_alloc) PetscCall(PetscFree(l->v));
1696: if (!l->unplaced_user_alloc) PetscCall(PetscFree(l->unplacedarray));
1697: PetscCheck(!l->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
1698: PetscCheck(!l->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
1699: PetscCall(VecDestroy(&l->cvec));
1700: PetscCall(MatDestroy(&l->cmat));
1701: PetscCall(PetscFree(mat->data));
1703: PetscCall(PetscObjectChangeTypeName((PetscObject)mat, NULL));
1704: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatQRFactor_C", NULL));
1705: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatQRFactorSymbolic_C", NULL));
1706: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatQRFactorNumeric_C", NULL));
1707: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetLDA_C", NULL));
1708: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseSetLDA_C", NULL));
1709: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetArray_C", NULL));
1710: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreArray_C", NULL));
1711: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDensePlaceArray_C", NULL));
1712: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseResetArray_C", NULL));
1713: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseReplaceArray_C", NULL));
1714: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetArrayRead_C", NULL));
1715: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreArrayRead_C", NULL));
1716: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetArrayWrite_C", NULL));
1717: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreArrayWrite_C", NULL));
1718: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatConvert_seqdense_seqaij_C", NULL));
1719: #if defined(PETSC_HAVE_ELEMENTAL)
1720: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatConvert_seqdense_elemental_C", NULL));
1721: #endif
1722: #if defined(PETSC_HAVE_SCALAPACK)
1723: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatConvert_seqdense_scalapack_C", NULL));
1724: #endif
1725: #if defined(PETSC_HAVE_CUDA)
1726: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatConvert_seqdense_seqdensecuda_C", NULL));
1727: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdensecuda_seqdensecuda_C", NULL));
1728: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdensecuda_seqdense_C", NULL));
1729: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdense_seqdensecuda_C", NULL));
1730: #endif
1731: #if defined(PETSC_HAVE_HIP)
1732: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatConvert_seqdense_seqdensehip_C", NULL));
1733: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdensehip_seqdensehip_C", NULL));
1734: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdensehip_seqdense_C", NULL));
1735: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdense_seqdensehip_C", NULL));
1736: #endif
1737: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatSeqDenseSetPreallocation_C", NULL));
1738: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqaij_seqdense_C", NULL));
1739: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqdense_seqdense_C", NULL));
1740: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqbaij_seqdense_C", NULL));
1741: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatProductSetFromOptions_seqsbaij_seqdense_C", NULL));
1743: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetColumn_C", NULL));
1744: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreColumn_C", NULL));
1745: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetColumnVec_C", NULL));
1746: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreColumnVec_C", NULL));
1747: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetColumnVecRead_C", NULL));
1748: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreColumnVecRead_C", NULL));
1749: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetColumnVecWrite_C", NULL));
1750: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreColumnVecWrite_C", NULL));
1751: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseGetSubMatrix_C", NULL));
1752: PetscCall(PetscObjectComposeFunction((PetscObject)mat, "MatDenseRestoreSubMatrix_C", NULL));
1753: PetscFunctionReturn(PETSC_SUCCESS);
1754: }
1756: static PetscErrorCode MatTranspose_SeqDense(Mat A, MatReuse reuse, Mat *matout)
1757: {
1758: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1759: PetscInt k, j, m = A->rmap->n, M = mat->lda, n = A->cmap->n;
1760: PetscScalar *v, tmp;
1762: PetscFunctionBegin;
1763: if (reuse == MAT_REUSE_MATRIX) PetscCall(MatTransposeCheckNonzeroState_Private(A, *matout));
1764: if (reuse == MAT_INPLACE_MATRIX) {
1765: if (m == n) { /* in place transpose */
1766: PetscCall(MatDenseGetArray(A, &v));
1767: for (j = 0; j < m; j++) {
1768: for (k = 0; k < j; k++) {
1769: tmp = v[j + k * M];
1770: v[j + k * M] = v[k + j * M];
1771: v[k + j * M] = tmp;
1772: }
1773: }
1774: PetscCall(MatDenseRestoreArray(A, &v));
1775: } else { /* reuse memory, temporary allocates new memory */
1776: PetscScalar *v2;
1777: PetscLayout tmplayout;
1779: PetscCall(PetscMalloc1((size_t)m * n, &v2));
1780: PetscCall(MatDenseGetArray(A, &v));
1781: for (j = 0; j < n; j++) {
1782: for (k = 0; k < m; k++) v2[j + (size_t)k * n] = v[k + (size_t)j * M];
1783: }
1784: PetscCall(PetscArraycpy(v, v2, (size_t)m * n));
1785: PetscCall(PetscFree(v2));
1786: PetscCall(MatDenseRestoreArray(A, &v));
1787: /* cleanup size dependent quantities */
1788: PetscCall(VecDestroy(&mat->cvec));
1789: PetscCall(MatDestroy(&mat->cmat));
1790: PetscCall(PetscFree(mat->pivots));
1791: PetscCall(PetscFree(mat->fwork));
1792: /* swap row/col layouts */
1793: mat->lda = n;
1794: tmplayout = A->rmap;
1795: A->rmap = A->cmap;
1796: A->cmap = tmplayout;
1797: }
1798: } else { /* out-of-place transpose */
1799: Mat tmat;
1800: Mat_SeqDense *tmatd;
1801: PetscScalar *v2;
1802: PetscInt M2;
1804: if (reuse == MAT_INITIAL_MATRIX) {
1805: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), &tmat));
1806: PetscCall(MatSetSizes(tmat, A->cmap->n, A->rmap->n, A->cmap->n, A->rmap->n));
1807: PetscCall(MatSetType(tmat, ((PetscObject)A)->type_name));
1808: PetscCall(MatSeqDenseSetPreallocation(tmat, NULL));
1809: } else tmat = *matout;
1811: PetscCall(MatDenseGetArrayRead(A, (const PetscScalar **)&v));
1812: PetscCall(MatDenseGetArray(tmat, &v2));
1813: tmatd = (Mat_SeqDense *)tmat->data;
1814: M2 = tmatd->lda;
1815: for (j = 0; j < n; j++) {
1816: for (k = 0; k < m; k++) v2[j + k * M2] = v[k + j * M];
1817: }
1818: PetscCall(MatDenseRestoreArray(tmat, &v2));
1819: PetscCall(MatDenseRestoreArrayRead(A, (const PetscScalar **)&v));
1820: PetscCall(MatAssemblyBegin(tmat, MAT_FINAL_ASSEMBLY));
1821: PetscCall(MatAssemblyEnd(tmat, MAT_FINAL_ASSEMBLY));
1822: *matout = tmat;
1823: }
1824: PetscFunctionReturn(PETSC_SUCCESS);
1825: }
1827: static PetscErrorCode MatEqual_SeqDense(Mat A1, Mat A2, PetscBool *flg)
1828: {
1829: Mat_SeqDense *mat1 = (Mat_SeqDense *)A1->data;
1830: Mat_SeqDense *mat2 = (Mat_SeqDense *)A2->data;
1831: PetscInt i;
1832: const PetscScalar *v1, *v2;
1834: PetscFunctionBegin;
1835: if (A1->rmap->n != A2->rmap->n) {
1836: *flg = PETSC_FALSE;
1837: PetscFunctionReturn(PETSC_SUCCESS);
1838: }
1839: if (A1->cmap->n != A2->cmap->n) {
1840: *flg = PETSC_FALSE;
1841: PetscFunctionReturn(PETSC_SUCCESS);
1842: }
1843: PetscCall(MatDenseGetArrayRead(A1, &v1));
1844: PetscCall(MatDenseGetArrayRead(A2, &v2));
1845: for (i = 0; i < A1->cmap->n; i++) {
1846: PetscCall(PetscArraycmp(v1, v2, A1->rmap->n, flg));
1847: if (*flg == PETSC_FALSE) PetscFunctionReturn(PETSC_SUCCESS);
1848: v1 += mat1->lda;
1849: v2 += mat2->lda;
1850: }
1851: PetscCall(MatDenseRestoreArrayRead(A1, &v1));
1852: PetscCall(MatDenseRestoreArrayRead(A2, &v2));
1853: *flg = PETSC_TRUE;
1854: PetscFunctionReturn(PETSC_SUCCESS);
1855: }
1857: static PetscErrorCode MatGetDiagonal_SeqDense(Mat A, Vec v)
1858: {
1859: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1860: PetscInt i, n, len;
1861: PetscScalar *x;
1862: const PetscScalar *vv;
1864: PetscFunctionBegin;
1865: PetscCall(VecGetSize(v, &n));
1866: PetscCall(VecGetArray(v, &x));
1867: len = PetscMin(A->rmap->n, A->cmap->n);
1868: PetscCall(MatDenseGetArrayRead(A, &vv));
1869: PetscCheck(n == A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Nonconforming mat and vec");
1870: for (i = 0; i < len; i++) x[i] = vv[i * mat->lda + i];
1871: PetscCall(MatDenseRestoreArrayRead(A, &vv));
1872: PetscCall(VecRestoreArray(v, &x));
1873: PetscFunctionReturn(PETSC_SUCCESS);
1874: }
1876: static PetscErrorCode MatDiagonalScale_SeqDense(Mat A, Vec ll, Vec rr)
1877: {
1878: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1879: const PetscScalar *l, *r;
1880: PetscScalar x, *v, *vv;
1881: PetscInt i, j, m = A->rmap->n, n = A->cmap->n;
1883: PetscFunctionBegin;
1884: PetscCall(MatDenseGetArray(A, &vv));
1885: if (ll) {
1886: PetscCall(VecGetSize(ll, &m));
1887: PetscCall(VecGetArrayRead(ll, &l));
1888: PetscCheck(m == A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Left scaling vec wrong size");
1889: for (i = 0; i < m; i++) {
1890: x = l[i];
1891: v = vv + i;
1892: for (j = 0; j < n; j++) {
1893: (*v) *= x;
1894: v += mat->lda;
1895: }
1896: }
1897: PetscCall(VecRestoreArrayRead(ll, &l));
1898: PetscCall(PetscLogFlops(1.0 * n * m));
1899: }
1900: if (rr) {
1901: PetscCall(VecGetSize(rr, &n));
1902: PetscCall(VecGetArrayRead(rr, &r));
1903: PetscCheck(n == A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Right scaling vec wrong size");
1904: for (i = 0; i < n; i++) {
1905: x = r[i];
1906: v = vv + i * mat->lda;
1907: for (j = 0; j < m; j++) (*v++) *= x;
1908: }
1909: PetscCall(VecRestoreArrayRead(rr, &r));
1910: PetscCall(PetscLogFlops(1.0 * n * m));
1911: }
1912: PetscCall(MatDenseRestoreArray(A, &vv));
1913: PetscFunctionReturn(PETSC_SUCCESS);
1914: }
1916: PetscErrorCode MatNorm_SeqDense(Mat A, NormType type, PetscReal *nrm)
1917: {
1918: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
1919: PetscScalar *v, *vv;
1920: PetscReal sum = 0.0;
1921: PetscInt lda, m = A->rmap->n, i, j;
1923: PetscFunctionBegin;
1924: PetscCall(MatDenseGetArrayRead(A, (const PetscScalar **)&vv));
1925: PetscCall(MatDenseGetLDA(A, &lda));
1926: v = vv;
1927: if (type == NORM_FROBENIUS) {
1928: if (lda > m) {
1929: for (j = 0; j < A->cmap->n; j++) {
1930: v = vv + j * lda;
1931: for (i = 0; i < m; i++) {
1932: sum += PetscRealPart(PetscConj(*v) * (*v));
1933: v++;
1934: }
1935: }
1936: } else {
1937: #if defined(PETSC_USE_REAL___FP16)
1938: PetscBLASInt one = 1, cnt = A->cmap->n * A->rmap->n;
1939: PetscCallBLAS("BLASnrm2", *nrm = BLASnrm2_(&cnt, v, &one));
1940: }
1941: #else
1942: for (i = 0; i < A->cmap->n * A->rmap->n; i++) {
1943: sum += PetscRealPart(PetscConj(*v) * (*v));
1944: v++;
1945: }
1946: }
1947: *nrm = PetscSqrtReal(sum);
1948: #endif
1949: PetscCall(PetscLogFlops(2.0 * A->cmap->n * A->rmap->n));
1950: } else if (type == NORM_1) {
1951: *nrm = 0.0;
1952: for (j = 0; j < A->cmap->n; j++) {
1953: v = vv + j * mat->lda;
1954: sum = 0.0;
1955: for (i = 0; i < A->rmap->n; i++) {
1956: sum += PetscAbsScalar(*v);
1957: v++;
1958: }
1959: if (sum > *nrm) *nrm = sum;
1960: }
1961: PetscCall(PetscLogFlops(1.0 * A->cmap->n * A->rmap->n));
1962: } else if (type == NORM_INFINITY) {
1963: *nrm = 0.0;
1964: for (j = 0; j < A->rmap->n; j++) {
1965: v = vv + j;
1966: sum = 0.0;
1967: for (i = 0; i < A->cmap->n; i++) {
1968: sum += PetscAbsScalar(*v);
1969: v += mat->lda;
1970: }
1971: if (sum > *nrm) *nrm = sum;
1972: }
1973: PetscCall(PetscLogFlops(1.0 * A->cmap->n * A->rmap->n));
1974: } else SETERRQ(PETSC_COMM_SELF, PETSC_ERR_SUP, "No two norm");
1975: PetscCall(MatDenseRestoreArrayRead(A, (const PetscScalar **)&vv));
1976: PetscFunctionReturn(PETSC_SUCCESS);
1977: }
1979: static PetscErrorCode MatSetOption_SeqDense(Mat A, MatOption op, PetscBool flg)
1980: {
1981: Mat_SeqDense *aij = (Mat_SeqDense *)A->data;
1983: PetscFunctionBegin;
1984: switch (op) {
1985: case MAT_ROW_ORIENTED:
1986: aij->roworiented = flg;
1987: break;
1988: case MAT_NEW_NONZERO_LOCATIONS:
1989: case MAT_NEW_NONZERO_LOCATION_ERR:
1990: case MAT_NEW_NONZERO_ALLOCATION_ERR:
1991: case MAT_FORCE_DIAGONAL_ENTRIES:
1992: case MAT_KEEP_NONZERO_PATTERN:
1993: case MAT_IGNORE_OFF_PROC_ENTRIES:
1994: case MAT_USE_HASH_TABLE:
1995: case MAT_IGNORE_ZERO_ENTRIES:
1996: case MAT_IGNORE_LOWER_TRIANGULAR:
1997: case MAT_SORTED_FULL:
1998: PetscCall(PetscInfo(A, "Option %s ignored\n", MatOptions[op]));
1999: break;
2000: case MAT_SPD:
2001: case MAT_SYMMETRIC:
2002: case MAT_STRUCTURALLY_SYMMETRIC:
2003: case MAT_HERMITIAN:
2004: case MAT_SYMMETRY_ETERNAL:
2005: case MAT_STRUCTURAL_SYMMETRY_ETERNAL:
2006: case MAT_SPD_ETERNAL:
2007: break;
2008: default:
2009: SETERRQ(PETSC_COMM_SELF, PETSC_ERR_SUP, "unknown option %s", MatOptions[op]);
2010: }
2011: PetscFunctionReturn(PETSC_SUCCESS);
2012: }
2014: PetscErrorCode MatZeroEntries_SeqDense(Mat A)
2015: {
2016: Mat_SeqDense *l = (Mat_SeqDense *)A->data;
2017: PetscInt lda = l->lda, m = A->rmap->n, n = A->cmap->n, j;
2018: PetscScalar *v;
2020: PetscFunctionBegin;
2021: PetscCall(MatDenseGetArrayWrite(A, &v));
2022: if (lda > m) {
2023: for (j = 0; j < n; j++) PetscCall(PetscArrayzero(v + j * lda, m));
2024: } else {
2025: PetscCall(PetscArrayzero(v, PetscInt64Mult(m, n)));
2026: }
2027: PetscCall(MatDenseRestoreArrayWrite(A, &v));
2028: PetscFunctionReturn(PETSC_SUCCESS);
2029: }
2031: static PetscErrorCode MatZeroRows_SeqDense(Mat A, PetscInt N, const PetscInt rows[], PetscScalar diag, Vec x, Vec b)
2032: {
2033: Mat_SeqDense *l = (Mat_SeqDense *)A->data;
2034: PetscInt m = l->lda, n = A->cmap->n, i, j;
2035: PetscScalar *slot, *bb, *v;
2036: const PetscScalar *xx;
2038: PetscFunctionBegin;
2039: if (PetscDefined(USE_DEBUG)) {
2040: for (i = 0; i < N; i++) {
2041: PetscCheck(rows[i] >= 0, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Negative row requested to be zeroed");
2042: PetscCheck(rows[i] < A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_OUTOFRANGE, "Row %" PetscInt_FMT " requested to be zeroed greater than or equal number of rows %" PetscInt_FMT, rows[i], A->rmap->n);
2043: }
2044: }
2045: if (!N) PetscFunctionReturn(PETSC_SUCCESS);
2047: /* fix right hand side if needed */
2048: if (x && b) {
2049: PetscCall(VecGetArrayRead(x, &xx));
2050: PetscCall(VecGetArray(b, &bb));
2051: for (i = 0; i < N; i++) bb[rows[i]] = diag * xx[rows[i]];
2052: PetscCall(VecRestoreArrayRead(x, &xx));
2053: PetscCall(VecRestoreArray(b, &bb));
2054: }
2056: PetscCall(MatDenseGetArray(A, &v));
2057: for (i = 0; i < N; i++) {
2058: slot = v + rows[i];
2059: for (j = 0; j < n; j++) {
2060: *slot = 0.0;
2061: slot += m;
2062: }
2063: }
2064: if (diag != 0.0) {
2065: PetscCheck(A->rmap->n == A->cmap->n, PETSC_COMM_SELF, PETSC_ERR_SUP, "Only coded for square matrices");
2066: for (i = 0; i < N; i++) {
2067: slot = v + (m + 1) * rows[i];
2068: *slot = diag;
2069: }
2070: }
2071: PetscCall(MatDenseRestoreArray(A, &v));
2072: PetscFunctionReturn(PETSC_SUCCESS);
2073: }
2075: static PetscErrorCode MatDenseGetLDA_SeqDense(Mat A, PetscInt *lda)
2076: {
2077: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2079: PetscFunctionBegin;
2080: *lda = mat->lda;
2081: PetscFunctionReturn(PETSC_SUCCESS);
2082: }
2084: PetscErrorCode MatDenseGetArray_SeqDense(Mat A, PetscScalar **array)
2085: {
2086: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2088: PetscFunctionBegin;
2089: PetscCheck(!mat->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
2090: *array = mat->v;
2091: PetscFunctionReturn(PETSC_SUCCESS);
2092: }
2094: PetscErrorCode MatDenseRestoreArray_SeqDense(Mat A, PetscScalar **array)
2095: {
2096: PetscFunctionBegin;
2097: if (array) *array = NULL;
2098: PetscFunctionReturn(PETSC_SUCCESS);
2099: }
2101: /*@
2102: MatDenseGetLDA - gets the leading dimension of the array returned from `MatDenseGetArray()`
2104: Not Collective
2106: Input Parameter:
2107: . mat - a `MATDENSE` or `MATDENSECUDA` matrix
2109: Output Parameter:
2110: . lda - the leading dimension
2112: Level: intermediate
2114: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseSetLDA()`
2115: @*/
2116: PetscErrorCode MatDenseGetLDA(Mat A, PetscInt *lda)
2117: {
2118: PetscFunctionBegin;
2121: MatCheckPreallocated(A, 1);
2122: PetscUseMethod(A, "MatDenseGetLDA_C", (Mat, PetscInt *), (A, lda));
2123: PetscFunctionReturn(PETSC_SUCCESS);
2124: }
2126: /*@
2127: MatDenseSetLDA - Sets the leading dimension of the array used by the `MATDENSE` matrix
2129: Not Collective
2131: Input Parameters:
2132: + mat - a `MATDENSE` or `MATDENSECUDA` matrix
2133: - lda - the leading dimension
2135: Level: intermediate
2137: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetLDA()`
2138: @*/
2139: PetscErrorCode MatDenseSetLDA(Mat A, PetscInt lda)
2140: {
2141: PetscFunctionBegin;
2143: PetscTryMethod(A, "MatDenseSetLDA_C", (Mat, PetscInt), (A, lda));
2144: PetscFunctionReturn(PETSC_SUCCESS);
2145: }
2147: /*@C
2148: MatDenseGetArray - gives read-write access to the array where the data for a `MATDENSE` matrix is stored
2150: Logically Collective
2152: Input Parameter:
2153: . mat - a dense matrix
2155: Output Parameter:
2156: . array - pointer to the data
2158: Level: intermediate
2160: Fortran Note:
2161: `MatDenseGetArray()` Fortran binding is deprecated (since PETSc 3.19), use `MatDenseGetArrayF90()`
2163: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2164: @*/
2165: PetscErrorCode MatDenseGetArray(Mat A, PetscScalar **array)
2166: {
2167: PetscFunctionBegin;
2170: PetscUseMethod(A, "MatDenseGetArray_C", (Mat, PetscScalar **), (A, array));
2171: PetscFunctionReturn(PETSC_SUCCESS);
2172: }
2174: /*@C
2175: MatDenseRestoreArray - returns access to the array where the data for a `MATDENSE` matrix is stored obtained by `MatDenseGetArray()`
2177: Logically Collective
2179: Input Parameters:
2180: + mat - a dense matrix
2181: - array - pointer to the data (may be `NULL`)
2183: Level: intermediate
2185: Fortran Note:
2186: `MatDenseRestoreArray()` Fortran binding is deprecated (since PETSc 3.19), use `MatDenseRestoreArrayF90()`
2188: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2189: @*/
2190: PetscErrorCode MatDenseRestoreArray(Mat A, PetscScalar **array)
2191: {
2192: PetscFunctionBegin;
2195: PetscUseMethod(A, "MatDenseRestoreArray_C", (Mat, PetscScalar **), (A, array));
2196: PetscCall(PetscObjectStateIncrease((PetscObject)A));
2197: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
2198: A->offloadmask = PETSC_OFFLOAD_CPU;
2199: #endif
2200: PetscFunctionReturn(PETSC_SUCCESS);
2201: }
2203: /*@C
2204: MatDenseGetArrayRead - gives read-only access to the array where the data for a `MATDENSE` matrix is stored
2206: Not Collective; No Fortran Support
2208: Input Parameter:
2209: . mat - a dense matrix
2211: Output Parameter:
2212: . array - pointer to the data
2214: Level: intermediate
2216: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArrayRead()`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2217: @*/
2218: PetscErrorCode MatDenseGetArrayRead(Mat A, const PetscScalar **array)
2219: {
2220: PetscFunctionBegin;
2223: PetscUseMethod(A, "MatDenseGetArrayRead_C", (Mat, const PetscScalar **), (A, array));
2224: PetscFunctionReturn(PETSC_SUCCESS);
2225: }
2227: /*@C
2228: MatDenseRestoreArrayRead - returns access to the array where the data for a `MATDENSE` matrix is stored obtained by `MatDenseGetArrayRead()`
2230: Not Collective; No Fortran Support
2232: Input Parameters:
2233: + mat - a dense matrix
2234: - array - pointer to the data (may be `NULL`)
2236: Level: intermediate
2238: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArrayRead()`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2239: @*/
2240: PetscErrorCode MatDenseRestoreArrayRead(Mat A, const PetscScalar **array)
2241: {
2242: PetscFunctionBegin;
2245: PetscUseMethod(A, "MatDenseRestoreArrayRead_C", (Mat, const PetscScalar **), (A, array));
2246: PetscFunctionReturn(PETSC_SUCCESS);
2247: }
2249: /*@C
2250: MatDenseGetArrayWrite - gives write-only access to the array where the data for a `MATDENSE` matrix is stored
2252: Not Collective; No Fortran Support
2254: Input Parameter:
2255: . mat - a dense matrix
2257: Output Parameter:
2258: . array - pointer to the data
2260: Level: intermediate
2262: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArrayWrite()`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`
2263: @*/
2264: PetscErrorCode MatDenseGetArrayWrite(Mat A, PetscScalar **array)
2265: {
2266: PetscFunctionBegin;
2269: PetscUseMethod(A, "MatDenseGetArrayWrite_C", (Mat, PetscScalar **), (A, array));
2270: PetscFunctionReturn(PETSC_SUCCESS);
2271: }
2273: /*@C
2274: MatDenseRestoreArrayWrite - returns access to the array where the data for a `MATDENSE` matrix is stored obtained by `MatDenseGetArrayWrite()`
2276: Not Collective; No Fortran Support
2278: Input Parameters:
2279: + mat - a dense matrix
2280: - array - pointer to the data (may be `NULL`)
2282: Level: intermediate
2284: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArrayWrite()`, `MatDenseGetArray()`, `MatDenseRestoreArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`
2285: @*/
2286: PetscErrorCode MatDenseRestoreArrayWrite(Mat A, PetscScalar **array)
2287: {
2288: PetscFunctionBegin;
2291: PetscUseMethod(A, "MatDenseRestoreArrayWrite_C", (Mat, PetscScalar **), (A, array));
2292: PetscCall(PetscObjectStateIncrease((PetscObject)A));
2293: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
2294: A->offloadmask = PETSC_OFFLOAD_CPU;
2295: #endif
2296: PetscFunctionReturn(PETSC_SUCCESS);
2297: }
2299: /*@C
2300: MatDenseGetArrayAndMemType - gives read-write access to the array where the data for a `MATDENSE` matrix is stored
2302: Logically Collective
2304: Input Parameter:
2305: . mat - a dense matrix
2307: Output Parameters:
2308: + array - pointer to the data
2309: - mtype - memory type of the returned pointer
2311: Level: intermediate
2313: Notes:
2314: If the matrix is of a device type such as `MATDENSECUDA`, `MATDENSEHIP`, etc.,
2315: an array on device is always returned and is guaranteed to contain the matrix's latest data.
2317: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArrayAndMemType()`, `MatDenseGetArrayReadAndMemType()`, `MatDenseGetArrayWriteAndMemType()`, `MatDenseGetArrayRead()`,
2318: `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`, `MatSeqAIJGetCSRAndMemType()`
2319: @*/
2320: PetscErrorCode MatDenseGetArrayAndMemType(Mat A, PetscScalar **array, PetscMemType *mtype)
2321: {
2322: PetscBool isMPI;
2324: PetscFunctionBegin;
2327: PetscCall(MatBindToCPU(A, PETSC_FALSE)); /* We want device matrices to always return device arrays, so we unbind the matrix if it is bound to CPU */
2328: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2329: if (isMPI) {
2330: /* Dispatch here so that the code can be reused for all subclasses of MATDENSE */
2331: PetscCall(MatDenseGetArrayAndMemType(((Mat_MPIDense *)A->data)->A, array, mtype));
2332: } else {
2333: PetscErrorCode (*fptr)(Mat, PetscScalar **, PetscMemType *);
2335: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseGetArrayAndMemType_C", &fptr));
2336: if (fptr) {
2337: PetscCall((*fptr)(A, array, mtype));
2338: } else {
2339: PetscUseMethod(A, "MatDenseGetArray_C", (Mat, PetscScalar **), (A, array));
2340: if (mtype) *mtype = PETSC_MEMTYPE_HOST;
2341: }
2342: }
2343: PetscFunctionReturn(PETSC_SUCCESS);
2344: }
2346: /*@C
2347: MatDenseRestoreArrayAndMemType - returns access to the array that is obtained by `MatDenseGetArrayAndMemType()`
2349: Logically Collective
2351: Input Parameters:
2352: + mat - a dense matrix
2353: - array - pointer to the data
2355: Level: intermediate
2357: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArrayAndMemType()`, `MatDenseGetArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2358: @*/
2359: PetscErrorCode MatDenseRestoreArrayAndMemType(Mat A, PetscScalar **array)
2360: {
2361: PetscBool isMPI;
2363: PetscFunctionBegin;
2366: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2367: if (isMPI) {
2368: PetscCall(MatDenseRestoreArrayAndMemType(((Mat_MPIDense *)A->data)->A, array));
2369: } else {
2370: PetscErrorCode (*fptr)(Mat, PetscScalar **);
2372: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseRestoreArrayAndMemType_C", &fptr));
2373: if (fptr) {
2374: PetscCall((*fptr)(A, array));
2375: } else {
2376: PetscUseMethod(A, "MatDenseRestoreArray_C", (Mat, PetscScalar **), (A, array));
2377: }
2378: *array = NULL;
2379: }
2380: PetscCall(PetscObjectStateIncrease((PetscObject)A));
2381: PetscFunctionReturn(PETSC_SUCCESS);
2382: }
2384: /*@C
2385: MatDenseGetArrayReadAndMemType - gives read-only access to the array where the data for a `MATDENSE` matrix is stored
2387: Logically Collective
2389: Input Parameter:
2390: . mat - a dense matrix
2392: Output Parameters:
2393: + array - pointer to the data
2394: - mtype - memory type of the returned pointer
2396: Level: intermediate
2398: Notes:
2399: If the matrix is of a device type such as `MATDENSECUDA`, `MATDENSEHIP`, etc.,
2400: an array on device is always returned and is guaranteed to contain the matrix's latest data.
2402: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArrayReadAndMemType()`, `MatDenseGetArrayWriteAndMemType()`,
2403: `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`, `MatSeqAIJGetCSRAndMemType()`
2404: @*/
2405: PetscErrorCode MatDenseGetArrayReadAndMemType(Mat A, const PetscScalar **array, PetscMemType *mtype)
2406: {
2407: PetscBool isMPI;
2409: PetscFunctionBegin;
2412: PetscCall(MatBindToCPU(A, PETSC_FALSE)); /* We want device matrices to always return device arrays, so we unbind the matrix if it is bound to CPU */
2413: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2414: if (isMPI) { /* Dispatch here so that the code can be reused for all subclasses of MATDENSE */
2415: PetscCall(MatDenseGetArrayReadAndMemType(((Mat_MPIDense *)A->data)->A, array, mtype));
2416: } else {
2417: PetscErrorCode (*fptr)(Mat, const PetscScalar **, PetscMemType *);
2419: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseGetArrayReadAndMemType_C", &fptr));
2420: if (fptr) {
2421: PetscCall((*fptr)(A, array, mtype));
2422: } else {
2423: PetscUseMethod(A, "MatDenseGetArrayRead_C", (Mat, const PetscScalar **), (A, array));
2424: if (mtype) *mtype = PETSC_MEMTYPE_HOST;
2425: }
2426: }
2427: PetscFunctionReturn(PETSC_SUCCESS);
2428: }
2430: /*@C
2431: MatDenseRestoreArrayReadAndMemType - returns access to the array that is obtained by `MatDenseGetArrayReadAndMemType()`
2433: Logically Collective
2435: Input Parameters:
2436: + mat - a dense matrix
2437: - array - pointer to the data
2439: Level: intermediate
2441: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArrayReadAndMemType()`, `MatDenseGetArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2442: @*/
2443: PetscErrorCode MatDenseRestoreArrayReadAndMemType(Mat A, const PetscScalar **array)
2444: {
2445: PetscBool isMPI;
2447: PetscFunctionBegin;
2450: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2451: if (isMPI) {
2452: PetscCall(MatDenseRestoreArrayReadAndMemType(((Mat_MPIDense *)A->data)->A, array));
2453: } else {
2454: PetscErrorCode (*fptr)(Mat, const PetscScalar **);
2456: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseRestoreArrayReadAndMemType_C", &fptr));
2457: if (fptr) {
2458: PetscCall((*fptr)(A, array));
2459: } else {
2460: PetscUseMethod(A, "MatDenseRestoreArrayRead_C", (Mat, const PetscScalar **), (A, array));
2461: }
2462: *array = NULL;
2463: }
2464: PetscFunctionReturn(PETSC_SUCCESS);
2465: }
2467: /*@C
2468: MatDenseGetArrayWriteAndMemType - gives write-only access to the array where the data for a `MATDENSE` matrix is stored
2470: Logically Collective
2472: Input Parameter:
2473: . mat - a dense matrix
2475: Output Parameters:
2476: + array - pointer to the data
2477: - mtype - memory type of the returned pointer
2479: Level: intermediate
2481: Notes:
2482: If the matrix is of a device type such as `MATDENSECUDA`, `MATDENSEHIP`, etc.,
2483: an array on device is always returned and is guaranteed to contain the matrix's latest data.
2485: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreArrayWriteAndMemType()`, `MatDenseGetArrayReadAndMemType()`, `MatDenseGetArrayRead()`,
2486: `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`, `MatSeqAIJGetCSRAndMemType()`
2487: @*/
2488: PetscErrorCode MatDenseGetArrayWriteAndMemType(Mat A, PetscScalar **array, PetscMemType *mtype)
2489: {
2490: PetscBool isMPI;
2492: PetscFunctionBegin;
2495: PetscCall(MatBindToCPU(A, PETSC_FALSE)); /* We want device matrices to always return device arrays, so we unbind the matrix if it is bound to CPU */
2496: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2497: if (isMPI) {
2498: PetscCall(MatDenseGetArrayWriteAndMemType(((Mat_MPIDense *)A->data)->A, array, mtype));
2499: } else {
2500: PetscErrorCode (*fptr)(Mat, PetscScalar **, PetscMemType *);
2502: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseGetArrayWriteAndMemType_C", &fptr));
2503: if (fptr) {
2504: PetscCall((*fptr)(A, array, mtype));
2505: } else {
2506: PetscUseMethod(A, "MatDenseGetArrayWrite_C", (Mat, PetscScalar **), (A, array));
2507: if (mtype) *mtype = PETSC_MEMTYPE_HOST;
2508: }
2509: }
2510: PetscFunctionReturn(PETSC_SUCCESS);
2511: }
2513: /*@C
2514: MatDenseRestoreArrayWriteAndMemType - returns access to the array that is obtained by `MatDenseGetArrayReadAndMemType()`
2516: Logically Collective
2518: Input Parameters:
2519: + mat - a dense matrix
2520: - array - pointer to the data
2522: Level: intermediate
2524: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetArrayWriteAndMemType()`, `MatDenseGetArray()`, `MatDenseGetArrayRead()`, `MatDenseRestoreArrayRead()`, `MatDenseGetArrayWrite()`, `MatDenseRestoreArrayWrite()`
2525: @*/
2526: PetscErrorCode MatDenseRestoreArrayWriteAndMemType(Mat A, PetscScalar **array)
2527: {
2528: PetscBool isMPI;
2530: PetscFunctionBegin;
2533: PetscCall(PetscObjectBaseTypeCompare((PetscObject)A, MATMPIDENSE, &isMPI));
2534: if (isMPI) {
2535: PetscCall(MatDenseRestoreArrayWriteAndMemType(((Mat_MPIDense *)A->data)->A, array));
2536: } else {
2537: PetscErrorCode (*fptr)(Mat, PetscScalar **);
2539: PetscCall(PetscObjectQueryFunction((PetscObject)A, "MatDenseRestoreArrayWriteAndMemType_C", &fptr));
2540: if (fptr) {
2541: PetscCall((*fptr)(A, array));
2542: } else {
2543: PetscUseMethod(A, "MatDenseRestoreArrayWrite_C", (Mat, PetscScalar **), (A, array));
2544: }
2545: *array = NULL;
2546: }
2547: PetscCall(PetscObjectStateIncrease((PetscObject)A));
2548: PetscFunctionReturn(PETSC_SUCCESS);
2549: }
2551: static PetscErrorCode MatCreateSubMatrix_SeqDense(Mat A, IS isrow, IS iscol, MatReuse scall, Mat *B)
2552: {
2553: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2554: PetscInt i, j, nrows, ncols, ldb;
2555: const PetscInt *irow, *icol;
2556: PetscScalar *av, *bv, *v = mat->v;
2557: Mat newmat;
2559: PetscFunctionBegin;
2560: PetscCall(ISGetIndices(isrow, &irow));
2561: PetscCall(ISGetIndices(iscol, &icol));
2562: PetscCall(ISGetLocalSize(isrow, &nrows));
2563: PetscCall(ISGetLocalSize(iscol, &ncols));
2565: /* Check submatrixcall */
2566: if (scall == MAT_REUSE_MATRIX) {
2567: PetscInt n_cols, n_rows;
2568: PetscCall(MatGetSize(*B, &n_rows, &n_cols));
2569: if (n_rows != nrows || n_cols != ncols) {
2570: /* resize the result matrix to match number of requested rows/columns */
2571: PetscCall(MatSetSizes(*B, nrows, ncols, nrows, ncols));
2572: }
2573: newmat = *B;
2574: } else {
2575: /* Create and fill new matrix */
2576: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), &newmat));
2577: PetscCall(MatSetSizes(newmat, nrows, ncols, nrows, ncols));
2578: PetscCall(MatSetType(newmat, ((PetscObject)A)->type_name));
2579: PetscCall(MatSeqDenseSetPreallocation(newmat, NULL));
2580: }
2582: /* Now extract the data pointers and do the copy,column at a time */
2583: PetscCall(MatDenseGetArray(newmat, &bv));
2584: PetscCall(MatDenseGetLDA(newmat, &ldb));
2585: for (i = 0; i < ncols; i++) {
2586: av = v + mat->lda * icol[i];
2587: for (j = 0; j < nrows; j++) bv[j] = av[irow[j]];
2588: bv += ldb;
2589: }
2590: PetscCall(MatDenseRestoreArray(newmat, &bv));
2592: /* Assemble the matrices so that the correct flags are set */
2593: PetscCall(MatAssemblyBegin(newmat, MAT_FINAL_ASSEMBLY));
2594: PetscCall(MatAssemblyEnd(newmat, MAT_FINAL_ASSEMBLY));
2596: /* Free work space */
2597: PetscCall(ISRestoreIndices(isrow, &irow));
2598: PetscCall(ISRestoreIndices(iscol, &icol));
2599: *B = newmat;
2600: PetscFunctionReturn(PETSC_SUCCESS);
2601: }
2603: static PetscErrorCode MatCreateSubMatrices_SeqDense(Mat A, PetscInt n, const IS irow[], const IS icol[], MatReuse scall, Mat *B[])
2604: {
2605: PetscInt i;
2607: PetscFunctionBegin;
2608: if (scall == MAT_INITIAL_MATRIX) PetscCall(PetscCalloc1(n, B));
2610: for (i = 0; i < n; i++) PetscCall(MatCreateSubMatrix_SeqDense(A, irow[i], icol[i], scall, &(*B)[i]));
2611: PetscFunctionReturn(PETSC_SUCCESS);
2612: }
2614: static PetscErrorCode MatAssemblyBegin_SeqDense(Mat mat, MatAssemblyType mode)
2615: {
2616: PetscFunctionBegin;
2617: PetscFunctionReturn(PETSC_SUCCESS);
2618: }
2620: static PetscErrorCode MatAssemblyEnd_SeqDense(Mat mat, MatAssemblyType mode)
2621: {
2622: PetscFunctionBegin;
2623: PetscFunctionReturn(PETSC_SUCCESS);
2624: }
2626: PetscErrorCode MatCopy_SeqDense(Mat A, Mat B, MatStructure str)
2627: {
2628: Mat_SeqDense *a = (Mat_SeqDense *)A->data, *b = (Mat_SeqDense *)B->data;
2629: const PetscScalar *va;
2630: PetscScalar *vb;
2631: PetscInt lda1 = a->lda, lda2 = b->lda, m = A->rmap->n, n = A->cmap->n, j;
2633: PetscFunctionBegin;
2634: /* If the two matrices don't have the same copy implementation, they aren't compatible for fast copy. */
2635: if (A->ops->copy != B->ops->copy) {
2636: PetscCall(MatCopy_Basic(A, B, str));
2637: PetscFunctionReturn(PETSC_SUCCESS);
2638: }
2639: PetscCheck(m == B->rmap->n && n == B->cmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "size(B) != size(A)");
2640: PetscCall(MatDenseGetArrayRead(A, &va));
2641: PetscCall(MatDenseGetArray(B, &vb));
2642: if (lda1 > m || lda2 > m) {
2643: for (j = 0; j < n; j++) PetscCall(PetscArraycpy(vb + j * lda2, va + j * lda1, m));
2644: } else {
2645: PetscCall(PetscArraycpy(vb, va, A->rmap->n * A->cmap->n));
2646: }
2647: PetscCall(MatDenseRestoreArray(B, &vb));
2648: PetscCall(MatDenseRestoreArrayRead(A, &va));
2649: PetscCall(MatAssemblyBegin(B, MAT_FINAL_ASSEMBLY));
2650: PetscCall(MatAssemblyEnd(B, MAT_FINAL_ASSEMBLY));
2651: PetscFunctionReturn(PETSC_SUCCESS);
2652: }
2654: PetscErrorCode MatSetUp_SeqDense(Mat A)
2655: {
2656: PetscFunctionBegin;
2657: PetscCall(PetscLayoutSetUp(A->rmap));
2658: PetscCall(PetscLayoutSetUp(A->cmap));
2659: if (!A->preallocated) PetscCall(MatSeqDenseSetPreallocation(A, NULL));
2660: PetscFunctionReturn(PETSC_SUCCESS);
2661: }
2663: static PetscErrorCode MatConjugate_SeqDense(Mat A)
2664: {
2665: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2666: PetscInt i, j;
2667: PetscInt min = PetscMin(A->rmap->n, A->cmap->n);
2668: PetscScalar *aa;
2670: PetscFunctionBegin;
2671: PetscCall(MatDenseGetArray(A, &aa));
2672: for (j = 0; j < A->cmap->n; j++) {
2673: for (i = 0; i < A->rmap->n; i++) aa[i + j * mat->lda] = PetscConj(aa[i + j * mat->lda]);
2674: }
2675: PetscCall(MatDenseRestoreArray(A, &aa));
2676: if (mat->tau)
2677: for (i = 0; i < min; i++) mat->tau[i] = PetscConj(mat->tau[i]);
2678: PetscFunctionReturn(PETSC_SUCCESS);
2679: }
2681: static PetscErrorCode MatRealPart_SeqDense(Mat A)
2682: {
2683: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2684: PetscInt i, j;
2685: PetscScalar *aa;
2687: PetscFunctionBegin;
2688: PetscCall(MatDenseGetArray(A, &aa));
2689: for (j = 0; j < A->cmap->n; j++) {
2690: for (i = 0; i < A->rmap->n; i++) aa[i + j * mat->lda] = PetscRealPart(aa[i + j * mat->lda]);
2691: }
2692: PetscCall(MatDenseRestoreArray(A, &aa));
2693: PetscFunctionReturn(PETSC_SUCCESS);
2694: }
2696: static PetscErrorCode MatImaginaryPart_SeqDense(Mat A)
2697: {
2698: Mat_SeqDense *mat = (Mat_SeqDense *)A->data;
2699: PetscInt i, j;
2700: PetscScalar *aa;
2702: PetscFunctionBegin;
2703: PetscCall(MatDenseGetArray(A, &aa));
2704: for (j = 0; j < A->cmap->n; j++) {
2705: for (i = 0; i < A->rmap->n; i++) aa[i + j * mat->lda] = PetscImaginaryPart(aa[i + j * mat->lda]);
2706: }
2707: PetscCall(MatDenseRestoreArray(A, &aa));
2708: PetscFunctionReturn(PETSC_SUCCESS);
2709: }
2711: PetscErrorCode MatMatMultSymbolic_SeqDense_SeqDense(Mat A, Mat B, PetscReal fill, Mat C)
2712: {
2713: PetscInt m = A->rmap->n, n = B->cmap->n;
2714: PetscBool cisdense = PETSC_FALSE;
2716: PetscFunctionBegin;
2717: PetscCall(MatSetSizes(C, m, n, m, n));
2718: #if defined(PETSC_HAVE_CUDA)
2719: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSECUDA, ""));
2720: #endif
2721: #if defined(PETSC_HAVE_HIP)
2722: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSEHIP, ""));
2723: #endif
2724: if (!cisdense) {
2725: PetscBool flg;
2727: PetscCall(PetscObjectTypeCompare((PetscObject)B, ((PetscObject)A)->type_name, &flg));
2728: PetscCall(MatSetType(C, flg ? ((PetscObject)A)->type_name : MATDENSE));
2729: }
2730: PetscCall(MatSetUp(C));
2731: PetscFunctionReturn(PETSC_SUCCESS);
2732: }
2734: PetscErrorCode MatMatMultNumeric_SeqDense_SeqDense(Mat A, Mat B, Mat C)
2735: {
2736: Mat_SeqDense *a = (Mat_SeqDense *)A->data, *b = (Mat_SeqDense *)B->data, *c = (Mat_SeqDense *)C->data;
2737: PetscBLASInt m, n, k;
2738: const PetscScalar *av, *bv;
2739: PetscScalar *cv;
2740: PetscScalar _DOne = 1.0, _DZero = 0.0;
2742: PetscFunctionBegin;
2743: PetscCall(PetscBLASIntCast(C->rmap->n, &m));
2744: PetscCall(PetscBLASIntCast(C->cmap->n, &n));
2745: PetscCall(PetscBLASIntCast(A->cmap->n, &k));
2746: if (!m || !n || !k) PetscFunctionReturn(PETSC_SUCCESS);
2747: PetscCall(MatDenseGetArrayRead(A, &av));
2748: PetscCall(MatDenseGetArrayRead(B, &bv));
2749: PetscCall(MatDenseGetArrayWrite(C, &cv));
2750: PetscCallBLAS("BLASgemm", BLASgemm_("N", "N", &m, &n, &k, &_DOne, av, &a->lda, bv, &b->lda, &_DZero, cv, &c->lda));
2751: PetscCall(PetscLogFlops(1.0 * m * n * k + 1.0 * m * n * (k - 1)));
2752: PetscCall(MatDenseRestoreArrayRead(A, &av));
2753: PetscCall(MatDenseRestoreArrayRead(B, &bv));
2754: PetscCall(MatDenseRestoreArrayWrite(C, &cv));
2755: PetscFunctionReturn(PETSC_SUCCESS);
2756: }
2758: PetscErrorCode MatMatTransposeMultSymbolic_SeqDense_SeqDense(Mat A, Mat B, PetscReal fill, Mat C)
2759: {
2760: PetscInt m = A->rmap->n, n = B->rmap->n;
2761: PetscBool cisdense = PETSC_FALSE;
2763: PetscFunctionBegin;
2764: PetscCall(MatSetSizes(C, m, n, m, n));
2765: #if defined(PETSC_HAVE_CUDA)
2766: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSECUDA, ""));
2767: #endif
2768: #if defined(PETSC_HAVE_HIP)
2769: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSEHIP, ""));
2770: #endif
2771: if (!cisdense) {
2772: PetscBool flg;
2774: PetscCall(PetscObjectTypeCompare((PetscObject)B, ((PetscObject)A)->type_name, &flg));
2775: PetscCall(MatSetType(C, flg ? ((PetscObject)A)->type_name : MATDENSE));
2776: }
2777: PetscCall(MatSetUp(C));
2778: PetscFunctionReturn(PETSC_SUCCESS);
2779: }
2781: PetscErrorCode MatMatTransposeMultNumeric_SeqDense_SeqDense(Mat A, Mat B, Mat C)
2782: {
2783: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2784: Mat_SeqDense *b = (Mat_SeqDense *)B->data;
2785: Mat_SeqDense *c = (Mat_SeqDense *)C->data;
2786: const PetscScalar *av, *bv;
2787: PetscScalar *cv;
2788: PetscBLASInt m, n, k;
2789: PetscScalar _DOne = 1.0, _DZero = 0.0;
2791: PetscFunctionBegin;
2792: PetscCall(PetscBLASIntCast(C->rmap->n, &m));
2793: PetscCall(PetscBLASIntCast(C->cmap->n, &n));
2794: PetscCall(PetscBLASIntCast(A->cmap->n, &k));
2795: if (!m || !n || !k) PetscFunctionReturn(PETSC_SUCCESS);
2796: PetscCall(MatDenseGetArrayRead(A, &av));
2797: PetscCall(MatDenseGetArrayRead(B, &bv));
2798: PetscCall(MatDenseGetArrayWrite(C, &cv));
2799: PetscCallBLAS("BLASgemm", BLASgemm_("N", "T", &m, &n, &k, &_DOne, av, &a->lda, bv, &b->lda, &_DZero, cv, &c->lda));
2800: PetscCall(MatDenseRestoreArrayRead(A, &av));
2801: PetscCall(MatDenseRestoreArrayRead(B, &bv));
2802: PetscCall(MatDenseRestoreArrayWrite(C, &cv));
2803: PetscCall(PetscLogFlops(1.0 * m * n * k + 1.0 * m * n * (k - 1)));
2804: PetscFunctionReturn(PETSC_SUCCESS);
2805: }
2807: PetscErrorCode MatTransposeMatMultSymbolic_SeqDense_SeqDense(Mat A, Mat B, PetscReal fill, Mat C)
2808: {
2809: PetscInt m = A->cmap->n, n = B->cmap->n;
2810: PetscBool cisdense = PETSC_FALSE;
2812: PetscFunctionBegin;
2813: PetscCall(MatSetSizes(C, m, n, m, n));
2814: #if defined(PETSC_HAVE_CUDA)
2815: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSECUDA, ""));
2816: #endif
2817: #if defined(PETSC_HAVE_HIP)
2818: PetscCall(PetscObjectTypeCompareAny((PetscObject)C, &cisdense, MATSEQDENSE, MATSEQDENSEHIP, ""));
2819: #endif
2820: if (!cisdense) {
2821: PetscBool flg;
2823: PetscCall(PetscObjectTypeCompare((PetscObject)B, ((PetscObject)A)->type_name, &flg));
2824: PetscCall(MatSetType(C, flg ? ((PetscObject)A)->type_name : MATDENSE));
2825: }
2826: PetscCall(MatSetUp(C));
2827: PetscFunctionReturn(PETSC_SUCCESS);
2828: }
2830: PetscErrorCode MatTransposeMatMultNumeric_SeqDense_SeqDense(Mat A, Mat B, Mat C)
2831: {
2832: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2833: Mat_SeqDense *b = (Mat_SeqDense *)B->data;
2834: Mat_SeqDense *c = (Mat_SeqDense *)C->data;
2835: const PetscScalar *av, *bv;
2836: PetscScalar *cv;
2837: PetscBLASInt m, n, k;
2838: PetscScalar _DOne = 1.0, _DZero = 0.0;
2840: PetscFunctionBegin;
2841: PetscCall(PetscBLASIntCast(C->rmap->n, &m));
2842: PetscCall(PetscBLASIntCast(C->cmap->n, &n));
2843: PetscCall(PetscBLASIntCast(A->rmap->n, &k));
2844: if (!m || !n || !k) PetscFunctionReturn(PETSC_SUCCESS);
2845: PetscCall(MatDenseGetArrayRead(A, &av));
2846: PetscCall(MatDenseGetArrayRead(B, &bv));
2847: PetscCall(MatDenseGetArrayWrite(C, &cv));
2848: PetscCallBLAS("BLASgemm", BLASgemm_("T", "N", &m, &n, &k, &_DOne, av, &a->lda, bv, &b->lda, &_DZero, cv, &c->lda));
2849: PetscCall(MatDenseRestoreArrayRead(A, &av));
2850: PetscCall(MatDenseRestoreArrayRead(B, &bv));
2851: PetscCall(MatDenseRestoreArrayWrite(C, &cv));
2852: PetscCall(PetscLogFlops(1.0 * m * n * k + 1.0 * m * n * (k - 1)));
2853: PetscFunctionReturn(PETSC_SUCCESS);
2854: }
2856: static PetscErrorCode MatProductSetFromOptions_SeqDense_AB(Mat C)
2857: {
2858: PetscFunctionBegin;
2859: C->ops->matmultsymbolic = MatMatMultSymbolic_SeqDense_SeqDense;
2860: C->ops->productsymbolic = MatProductSymbolic_AB;
2861: PetscFunctionReturn(PETSC_SUCCESS);
2862: }
2864: static PetscErrorCode MatProductSetFromOptions_SeqDense_AtB(Mat C)
2865: {
2866: PetscFunctionBegin;
2867: C->ops->transposematmultsymbolic = MatTransposeMatMultSymbolic_SeqDense_SeqDense;
2868: C->ops->productsymbolic = MatProductSymbolic_AtB;
2869: PetscFunctionReturn(PETSC_SUCCESS);
2870: }
2872: static PetscErrorCode MatProductSetFromOptions_SeqDense_ABt(Mat C)
2873: {
2874: PetscFunctionBegin;
2875: C->ops->mattransposemultsymbolic = MatMatTransposeMultSymbolic_SeqDense_SeqDense;
2876: C->ops->productsymbolic = MatProductSymbolic_ABt;
2877: PetscFunctionReturn(PETSC_SUCCESS);
2878: }
2880: PETSC_INTERN PetscErrorCode MatProductSetFromOptions_SeqDense(Mat C)
2881: {
2882: Mat_Product *product = C->product;
2884: PetscFunctionBegin;
2885: switch (product->type) {
2886: case MATPRODUCT_AB:
2887: PetscCall(MatProductSetFromOptions_SeqDense_AB(C));
2888: break;
2889: case MATPRODUCT_AtB:
2890: PetscCall(MatProductSetFromOptions_SeqDense_AtB(C));
2891: break;
2892: case MATPRODUCT_ABt:
2893: PetscCall(MatProductSetFromOptions_SeqDense_ABt(C));
2894: break;
2895: default:
2896: break;
2897: }
2898: PetscFunctionReturn(PETSC_SUCCESS);
2899: }
2901: static PetscErrorCode MatGetRowMax_SeqDense(Mat A, Vec v, PetscInt idx[])
2902: {
2903: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2904: PetscInt i, j, m = A->rmap->n, n = A->cmap->n, p;
2905: PetscScalar *x;
2906: const PetscScalar *aa;
2908: PetscFunctionBegin;
2909: PetscCheck(!A->factortype, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
2910: PetscCall(VecGetArray(v, &x));
2911: PetscCall(VecGetLocalSize(v, &p));
2912: PetscCall(MatDenseGetArrayRead(A, &aa));
2913: PetscCheck(p == A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Nonconforming matrix and vector");
2914: for (i = 0; i < m; i++) {
2915: x[i] = aa[i];
2916: if (idx) idx[i] = 0;
2917: for (j = 1; j < n; j++) {
2918: if (PetscRealPart(x[i]) < PetscRealPart(aa[i + a->lda * j])) {
2919: x[i] = aa[i + a->lda * j];
2920: if (idx) idx[i] = j;
2921: }
2922: }
2923: }
2924: PetscCall(MatDenseRestoreArrayRead(A, &aa));
2925: PetscCall(VecRestoreArray(v, &x));
2926: PetscFunctionReturn(PETSC_SUCCESS);
2927: }
2929: static PetscErrorCode MatGetRowMaxAbs_SeqDense(Mat A, Vec v, PetscInt idx[])
2930: {
2931: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2932: PetscInt i, j, m = A->rmap->n, n = A->cmap->n, p;
2933: PetscScalar *x;
2934: PetscReal atmp;
2935: const PetscScalar *aa;
2937: PetscFunctionBegin;
2938: PetscCheck(!A->factortype, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
2939: PetscCall(VecGetArray(v, &x));
2940: PetscCall(VecGetLocalSize(v, &p));
2941: PetscCall(MatDenseGetArrayRead(A, &aa));
2942: PetscCheck(p == A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Nonconforming matrix and vector");
2943: for (i = 0; i < m; i++) {
2944: x[i] = PetscAbsScalar(aa[i]);
2945: for (j = 1; j < n; j++) {
2946: atmp = PetscAbsScalar(aa[i + a->lda * j]);
2947: if (PetscAbsScalar(x[i]) < atmp) {
2948: x[i] = atmp;
2949: if (idx) idx[i] = j;
2950: }
2951: }
2952: }
2953: PetscCall(MatDenseRestoreArrayRead(A, &aa));
2954: PetscCall(VecRestoreArray(v, &x));
2955: PetscFunctionReturn(PETSC_SUCCESS);
2956: }
2958: static PetscErrorCode MatGetRowMin_SeqDense(Mat A, Vec v, PetscInt idx[])
2959: {
2960: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2961: PetscInt i, j, m = A->rmap->n, n = A->cmap->n, p;
2962: PetscScalar *x;
2963: const PetscScalar *aa;
2965: PetscFunctionBegin;
2966: PetscCheck(!A->factortype, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
2967: PetscCall(MatDenseGetArrayRead(A, &aa));
2968: PetscCall(VecGetArray(v, &x));
2969: PetscCall(VecGetLocalSize(v, &p));
2970: PetscCheck(p == A->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "Nonconforming matrix and vector");
2971: for (i = 0; i < m; i++) {
2972: x[i] = aa[i];
2973: if (idx) idx[i] = 0;
2974: for (j = 1; j < n; j++) {
2975: if (PetscRealPart(x[i]) > PetscRealPart(aa[i + a->lda * j])) {
2976: x[i] = aa[i + a->lda * j];
2977: if (idx) idx[i] = j;
2978: }
2979: }
2980: }
2981: PetscCall(VecRestoreArray(v, &x));
2982: PetscCall(MatDenseRestoreArrayRead(A, &aa));
2983: PetscFunctionReturn(PETSC_SUCCESS);
2984: }
2986: PetscErrorCode MatGetColumnVector_SeqDense(Mat A, Vec v, PetscInt col)
2987: {
2988: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
2989: PetscScalar *x;
2990: const PetscScalar *aa;
2992: PetscFunctionBegin;
2993: PetscCheck(!A->factortype, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
2994: PetscCall(MatDenseGetArrayRead(A, &aa));
2995: PetscCall(VecGetArray(v, &x));
2996: PetscCall(PetscArraycpy(x, aa + col * a->lda, A->rmap->n));
2997: PetscCall(VecRestoreArray(v, &x));
2998: PetscCall(MatDenseRestoreArrayRead(A, &aa));
2999: PetscFunctionReturn(PETSC_SUCCESS);
3000: }
3002: PETSC_INTERN PetscErrorCode MatGetColumnReductions_SeqDense(Mat A, PetscInt type, PetscReal *reductions)
3003: {
3004: PetscInt i, j, m, n;
3005: const PetscScalar *a;
3007: PetscFunctionBegin;
3008: PetscCall(MatGetSize(A, &m, &n));
3009: PetscCall(PetscArrayzero(reductions, n));
3010: PetscCall(MatDenseGetArrayRead(A, &a));
3011: if (type == NORM_2) {
3012: for (i = 0; i < n; i++) {
3013: for (j = 0; j < m; j++) reductions[i] += PetscAbsScalar(a[j] * a[j]);
3014: a += m;
3015: }
3016: } else if (type == NORM_1) {
3017: for (i = 0; i < n; i++) {
3018: for (j = 0; j < m; j++) reductions[i] += PetscAbsScalar(a[j]);
3019: a += m;
3020: }
3021: } else if (type == NORM_INFINITY) {
3022: for (i = 0; i < n; i++) {
3023: for (j = 0; j < m; j++) reductions[i] = PetscMax(PetscAbsScalar(a[j]), reductions[i]);
3024: a += m;
3025: }
3026: } else if (type == REDUCTION_SUM_REALPART || type == REDUCTION_MEAN_REALPART) {
3027: for (i = 0; i < n; i++) {
3028: for (j = 0; j < m; j++) reductions[i] += PetscRealPart(a[j]);
3029: a += m;
3030: }
3031: } else if (type == REDUCTION_SUM_IMAGINARYPART || type == REDUCTION_MEAN_IMAGINARYPART) {
3032: for (i = 0; i < n; i++) {
3033: for (j = 0; j < m; j++) reductions[i] += PetscImaginaryPart(a[j]);
3034: a += m;
3035: }
3036: } else SETERRQ(PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Unknown reduction type");
3037: PetscCall(MatDenseRestoreArrayRead(A, &a));
3038: if (type == NORM_2) {
3039: for (i = 0; i < n; i++) reductions[i] = PetscSqrtReal(reductions[i]);
3040: } else if (type == REDUCTION_MEAN_REALPART || type == REDUCTION_MEAN_IMAGINARYPART) {
3041: for (i = 0; i < n; i++) reductions[i] /= m;
3042: }
3043: PetscFunctionReturn(PETSC_SUCCESS);
3044: }
3046: PetscErrorCode MatSetRandom_SeqDense(Mat x, PetscRandom rctx)
3047: {
3048: PetscScalar *a;
3049: PetscInt lda, m, n, i, j;
3051: PetscFunctionBegin;
3052: PetscCall(MatGetSize(x, &m, &n));
3053: PetscCall(MatDenseGetLDA(x, &lda));
3054: PetscCall(MatDenseGetArrayWrite(x, &a));
3055: for (j = 0; j < n; j++) {
3056: for (i = 0; i < m; i++) PetscCall(PetscRandomGetValue(rctx, a + j * lda + i));
3057: }
3058: PetscCall(MatDenseRestoreArrayWrite(x, &a));
3059: PetscFunctionReturn(PETSC_SUCCESS);
3060: }
3062: static PetscErrorCode MatMissingDiagonal_SeqDense(Mat A, PetscBool *missing, PetscInt *d)
3063: {
3064: PetscFunctionBegin;
3065: *missing = PETSC_FALSE;
3066: PetscFunctionReturn(PETSC_SUCCESS);
3067: }
3069: /* vals is not const */
3070: static PetscErrorCode MatDenseGetColumn_SeqDense(Mat A, PetscInt col, PetscScalar **vals)
3071: {
3072: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3073: PetscScalar *v;
3075: PetscFunctionBegin;
3076: PetscCheck(!A->factortype, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
3077: PetscCall(MatDenseGetArray(A, &v));
3078: *vals = v + col * a->lda;
3079: PetscCall(MatDenseRestoreArray(A, &v));
3080: PetscFunctionReturn(PETSC_SUCCESS);
3081: }
3083: static PetscErrorCode MatDenseRestoreColumn_SeqDense(Mat A, PetscScalar **vals)
3084: {
3085: PetscFunctionBegin;
3086: if (vals) *vals = NULL; /* user cannot accidentally use the array later */
3087: PetscFunctionReturn(PETSC_SUCCESS);
3088: }
3090: static struct _MatOps MatOps_Values = {MatSetValues_SeqDense,
3091: MatGetRow_SeqDense,
3092: MatRestoreRow_SeqDense,
3093: MatMult_SeqDense,
3094: /* 4*/ MatMultAdd_SeqDense,
3095: MatMultTranspose_SeqDense,
3096: MatMultTransposeAdd_SeqDense,
3097: NULL,
3098: NULL,
3099: NULL,
3100: /* 10*/ NULL,
3101: MatLUFactor_SeqDense,
3102: MatCholeskyFactor_SeqDense,
3103: MatSOR_SeqDense,
3104: MatTranspose_SeqDense,
3105: /* 15*/ MatGetInfo_SeqDense,
3106: MatEqual_SeqDense,
3107: MatGetDiagonal_SeqDense,
3108: MatDiagonalScale_SeqDense,
3109: MatNorm_SeqDense,
3110: /* 20*/ MatAssemblyBegin_SeqDense,
3111: MatAssemblyEnd_SeqDense,
3112: MatSetOption_SeqDense,
3113: MatZeroEntries_SeqDense,
3114: /* 24*/ MatZeroRows_SeqDense,
3115: NULL,
3116: NULL,
3117: NULL,
3118: NULL,
3119: /* 29*/ MatSetUp_SeqDense,
3120: NULL,
3121: NULL,
3122: NULL,
3123: NULL,
3124: /* 34*/ MatDuplicate_SeqDense,
3125: NULL,
3126: NULL,
3127: NULL,
3128: NULL,
3129: /* 39*/ MatAXPY_SeqDense,
3130: MatCreateSubMatrices_SeqDense,
3131: NULL,
3132: MatGetValues_SeqDense,
3133: MatCopy_SeqDense,
3134: /* 44*/ MatGetRowMax_SeqDense,
3135: MatScale_SeqDense,
3136: MatShift_SeqDense,
3137: NULL,
3138: MatZeroRowsColumns_SeqDense,
3139: /* 49*/ MatSetRandom_SeqDense,
3140: NULL,
3141: NULL,
3142: NULL,
3143: NULL,
3144: /* 54*/ NULL,
3145: NULL,
3146: NULL,
3147: NULL,
3148: NULL,
3149: /* 59*/ MatCreateSubMatrix_SeqDense,
3150: MatDestroy_SeqDense,
3151: MatView_SeqDense,
3152: NULL,
3153: NULL,
3154: /* 64*/ NULL,
3155: NULL,
3156: NULL,
3157: NULL,
3158: NULL,
3159: /* 69*/ MatGetRowMaxAbs_SeqDense,
3160: NULL,
3161: NULL,
3162: NULL,
3163: NULL,
3164: /* 74*/ NULL,
3165: NULL,
3166: NULL,
3167: NULL,
3168: NULL,
3169: /* 79*/ NULL,
3170: NULL,
3171: NULL,
3172: NULL,
3173: /* 83*/ MatLoad_SeqDense,
3174: MatIsSymmetric_SeqDense,
3175: MatIsHermitian_SeqDense,
3176: NULL,
3177: NULL,
3178: NULL,
3179: /* 89*/ NULL,
3180: NULL,
3181: MatMatMultNumeric_SeqDense_SeqDense,
3182: NULL,
3183: NULL,
3184: /* 94*/ NULL,
3185: NULL,
3186: NULL,
3187: MatMatTransposeMultNumeric_SeqDense_SeqDense,
3188: NULL,
3189: /* 99*/ MatProductSetFromOptions_SeqDense,
3190: NULL,
3191: NULL,
3192: MatConjugate_SeqDense,
3193: NULL,
3194: /*104*/ NULL,
3195: MatRealPart_SeqDense,
3196: MatImaginaryPart_SeqDense,
3197: NULL,
3198: NULL,
3199: /*109*/ NULL,
3200: NULL,
3201: MatGetRowMin_SeqDense,
3202: MatGetColumnVector_SeqDense,
3203: MatMissingDiagonal_SeqDense,
3204: /*114*/ NULL,
3205: NULL,
3206: NULL,
3207: NULL,
3208: NULL,
3209: /*119*/ NULL,
3210: NULL,
3211: NULL,
3212: NULL,
3213: NULL,
3214: /*124*/ NULL,
3215: MatGetColumnReductions_SeqDense,
3216: NULL,
3217: NULL,
3218: NULL,
3219: /*129*/ NULL,
3220: NULL,
3221: NULL,
3222: MatTransposeMatMultNumeric_SeqDense_SeqDense,
3223: NULL,
3224: /*134*/ NULL,
3225: NULL,
3226: NULL,
3227: NULL,
3228: NULL,
3229: /*139*/ NULL,
3230: NULL,
3231: NULL,
3232: NULL,
3233: NULL,
3234: MatCreateMPIMatConcatenateSeqMat_SeqDense,
3235: /*145*/ NULL,
3236: NULL,
3237: NULL,
3238: NULL,
3239: NULL,
3240: /*150*/ NULL,
3241: NULL};
3243: /*@C
3244: MatCreateSeqDense - Creates a `MATSEQDENSE` that
3245: is stored in column major order (the usual Fortran manner). Many
3246: of the matrix operations use the BLAS and LAPACK routines.
3248: Collective
3250: Input Parameters:
3251: + comm - MPI communicator, set to `PETSC_COMM_SELF`
3252: . m - number of rows
3253: . n - number of columns
3254: - data - optional location of matrix data in column major order. Use `NULL` for PETSc
3255: to control all matrix memory allocation.
3257: Output Parameter:
3258: . A - the matrix
3260: Level: intermediate
3262: Note:
3263: The data input variable is intended primarily for Fortran programmers
3264: who wish to allocate their own matrix memory space. Most users should
3265: set `data` = `NULL`.
3267: .seealso: [](ch_matrices), `Mat`, `MATSEQDENSE`, `MatCreate()`, `MatCreateDense()`, `MatSetValues()`
3268: @*/
3269: PetscErrorCode MatCreateSeqDense(MPI_Comm comm, PetscInt m, PetscInt n, PetscScalar *data, Mat *A)
3270: {
3271: PetscFunctionBegin;
3272: PetscCall(MatCreate(comm, A));
3273: PetscCall(MatSetSizes(*A, m, n, m, n));
3274: PetscCall(MatSetType(*A, MATSEQDENSE));
3275: PetscCall(MatSeqDenseSetPreallocation(*A, data));
3276: PetscFunctionReturn(PETSC_SUCCESS);
3277: }
3279: /*@C
3280: MatSeqDenseSetPreallocation - Sets the array used for storing the matrix elements of a `MATSEQDENSE` matrix
3282: Collective
3284: Input Parameters:
3285: + B - the matrix
3286: - data - the array (or `NULL`)
3288: Level: intermediate
3290: Note:
3291: The data input variable is intended primarily for Fortran programmers
3292: who wish to allocate their own matrix memory space. Most users should
3293: need not call this routine.
3295: .seealso: [](ch_matrices), `Mat`, `MATSEQDENSE`, `MatCreate()`, `MatCreateDense()`, `MatSetValues()`, `MatDenseSetLDA()`
3296: @*/
3297: PetscErrorCode MatSeqDenseSetPreallocation(Mat B, PetscScalar data[])
3298: {
3299: PetscFunctionBegin;
3301: PetscTryMethod(B, "MatSeqDenseSetPreallocation_C", (Mat, PetscScalar[]), (B, data));
3302: PetscFunctionReturn(PETSC_SUCCESS);
3303: }
3305: PetscErrorCode MatSeqDenseSetPreallocation_SeqDense(Mat B, PetscScalar *data)
3306: {
3307: Mat_SeqDense *b = (Mat_SeqDense *)B->data;
3309: PetscFunctionBegin;
3310: PetscCheck(!b->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
3311: B->preallocated = PETSC_TRUE;
3313: PetscCall(PetscLayoutSetUp(B->rmap));
3314: PetscCall(PetscLayoutSetUp(B->cmap));
3316: if (b->lda <= 0) b->lda = B->rmap->n;
3318: if (!data) { /* petsc-allocated storage */
3319: if (!b->user_alloc) PetscCall(PetscFree(b->v));
3320: PetscCall(PetscCalloc1((size_t)b->lda * B->cmap->n, &b->v));
3322: b->user_alloc = PETSC_FALSE;
3323: } else { /* user-allocated storage */
3324: if (!b->user_alloc) PetscCall(PetscFree(b->v));
3325: b->v = data;
3326: b->user_alloc = PETSC_TRUE;
3327: }
3328: B->assembled = PETSC_TRUE;
3329: PetscFunctionReturn(PETSC_SUCCESS);
3330: }
3332: #if defined(PETSC_HAVE_ELEMENTAL)
3333: PETSC_INTERN PetscErrorCode MatConvert_SeqDense_Elemental(Mat A, MatType newtype, MatReuse reuse, Mat *newmat)
3334: {
3335: Mat mat_elemental;
3336: const PetscScalar *array;
3337: PetscScalar *v_colwise;
3338: PetscInt M = A->rmap->N, N = A->cmap->N, i, j, k, *rows, *cols;
3340: PetscFunctionBegin;
3341: PetscCall(PetscMalloc3(M * N, &v_colwise, M, &rows, N, &cols));
3342: PetscCall(MatDenseGetArrayRead(A, &array));
3343: /* convert column-wise array into row-wise v_colwise, see MatSetValues_Elemental() */
3344: k = 0;
3345: for (j = 0; j < N; j++) {
3346: cols[j] = j;
3347: for (i = 0; i < M; i++) v_colwise[j * M + i] = array[k++];
3348: }
3349: for (i = 0; i < M; i++) rows[i] = i;
3350: PetscCall(MatDenseRestoreArrayRead(A, &array));
3352: PetscCall(MatCreate(PetscObjectComm((PetscObject)A), &mat_elemental));
3353: PetscCall(MatSetSizes(mat_elemental, PETSC_DECIDE, PETSC_DECIDE, M, N));
3354: PetscCall(MatSetType(mat_elemental, MATELEMENTAL));
3355: PetscCall(MatSetUp(mat_elemental));
3357: /* PETSc-Elemental interaface uses axpy for setting off-processor entries, only ADD_VALUES is allowed */
3358: PetscCall(MatSetValues(mat_elemental, M, rows, N, cols, v_colwise, ADD_VALUES));
3359: PetscCall(MatAssemblyBegin(mat_elemental, MAT_FINAL_ASSEMBLY));
3360: PetscCall(MatAssemblyEnd(mat_elemental, MAT_FINAL_ASSEMBLY));
3361: PetscCall(PetscFree3(v_colwise, rows, cols));
3363: if (reuse == MAT_INPLACE_MATRIX) {
3364: PetscCall(MatHeaderReplace(A, &mat_elemental));
3365: } else {
3366: *newmat = mat_elemental;
3367: }
3368: PetscFunctionReturn(PETSC_SUCCESS);
3369: }
3370: #endif
3372: PetscErrorCode MatDenseSetLDA_SeqDense(Mat B, PetscInt lda)
3373: {
3374: Mat_SeqDense *b = (Mat_SeqDense *)B->data;
3375: PetscBool data;
3377: PetscFunctionBegin;
3378: data = (PetscBool)((B->rmap->n > 0 && B->cmap->n > 0) ? (b->v ? PETSC_TRUE : PETSC_FALSE) : PETSC_FALSE);
3379: PetscCheck(b->user_alloc || !data || b->lda == lda, PETSC_COMM_SELF, PETSC_ERR_ORDER, "LDA cannot be changed after allocation of internal storage");
3380: PetscCheck(lda >= B->rmap->n, PETSC_COMM_SELF, PETSC_ERR_ARG_SIZ, "LDA %" PetscInt_FMT " must be at least matrix dimension %" PetscInt_FMT, lda, B->rmap->n);
3381: b->lda = lda;
3382: PetscFunctionReturn(PETSC_SUCCESS);
3383: }
3385: PetscErrorCode MatCreateMPIMatConcatenateSeqMat_SeqDense(MPI_Comm comm, Mat inmat, PetscInt n, MatReuse scall, Mat *outmat)
3386: {
3387: PetscFunctionBegin;
3388: PetscCall(MatCreateMPIMatConcatenateSeqMat_MPIDense(comm, inmat, n, scall, outmat));
3389: PetscFunctionReturn(PETSC_SUCCESS);
3390: }
3392: PetscErrorCode MatDenseGetColumnVec_SeqDense(Mat A, PetscInt col, Vec *v)
3393: {
3394: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3396: PetscFunctionBegin;
3397: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
3398: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
3399: if (!a->cvec) { PetscCall(VecCreateSeqWithArray(PetscObjectComm((PetscObject)A), A->rmap->bs, A->rmap->n, NULL, &a->cvec)); }
3400: a->vecinuse = col + 1;
3401: PetscCall(MatDenseGetArray(A, (PetscScalar **)&a->ptrinuse));
3402: PetscCall(VecPlaceArray(a->cvec, a->ptrinuse + (size_t)col * (size_t)a->lda));
3403: *v = a->cvec;
3404: PetscFunctionReturn(PETSC_SUCCESS);
3405: }
3407: PetscErrorCode MatDenseRestoreColumnVec_SeqDense(Mat A, PetscInt col, Vec *v)
3408: {
3409: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3411: PetscFunctionBegin;
3412: PetscCheck(a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseGetColumnVec() first");
3413: PetscCheck(a->cvec, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Missing internal column vector");
3414: a->vecinuse = 0;
3415: PetscCall(MatDenseRestoreArray(A, (PetscScalar **)&a->ptrinuse));
3416: PetscCall(VecResetArray(a->cvec));
3417: if (v) *v = NULL;
3418: PetscFunctionReturn(PETSC_SUCCESS);
3419: }
3421: PetscErrorCode MatDenseGetColumnVecRead_SeqDense(Mat A, PetscInt col, Vec *v)
3422: {
3423: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3425: PetscFunctionBegin;
3426: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
3427: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
3428: if (!a->cvec) { PetscCall(VecCreateSeqWithArray(PetscObjectComm((PetscObject)A), A->rmap->bs, A->rmap->n, NULL, &a->cvec)); }
3429: a->vecinuse = col + 1;
3430: PetscCall(MatDenseGetArrayRead(A, &a->ptrinuse));
3431: PetscCall(VecPlaceArray(a->cvec, a->ptrinuse + (size_t)col * (size_t)a->lda));
3432: PetscCall(VecLockReadPush(a->cvec));
3433: *v = a->cvec;
3434: PetscFunctionReturn(PETSC_SUCCESS);
3435: }
3437: PetscErrorCode MatDenseRestoreColumnVecRead_SeqDense(Mat A, PetscInt col, Vec *v)
3438: {
3439: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3441: PetscFunctionBegin;
3442: PetscCheck(a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseGetColumnVec() first");
3443: PetscCheck(a->cvec, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Missing internal column vector");
3444: a->vecinuse = 0;
3445: PetscCall(MatDenseRestoreArrayRead(A, &a->ptrinuse));
3446: PetscCall(VecLockReadPop(a->cvec));
3447: PetscCall(VecResetArray(a->cvec));
3448: if (v) *v = NULL;
3449: PetscFunctionReturn(PETSC_SUCCESS);
3450: }
3452: PetscErrorCode MatDenseGetColumnVecWrite_SeqDense(Mat A, PetscInt col, Vec *v)
3453: {
3454: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3456: PetscFunctionBegin;
3457: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
3458: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
3459: if (!a->cvec) PetscCall(VecCreateSeqWithArray(PetscObjectComm((PetscObject)A), A->rmap->bs, A->rmap->n, NULL, &a->cvec));
3460: a->vecinuse = col + 1;
3461: PetscCall(MatDenseGetArrayWrite(A, (PetscScalar **)&a->ptrinuse));
3462: PetscCall(VecPlaceArray(a->cvec, a->ptrinuse + (size_t)col * (size_t)a->lda));
3463: *v = a->cvec;
3464: PetscFunctionReturn(PETSC_SUCCESS);
3465: }
3467: PetscErrorCode MatDenseRestoreColumnVecWrite_SeqDense(Mat A, PetscInt col, Vec *v)
3468: {
3469: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3471: PetscFunctionBegin;
3472: PetscCheck(a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseGetColumnVec() first");
3473: PetscCheck(a->cvec, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Missing internal column vector");
3474: a->vecinuse = 0;
3475: PetscCall(MatDenseRestoreArrayWrite(A, (PetscScalar **)&a->ptrinuse));
3476: PetscCall(VecResetArray(a->cvec));
3477: if (v) *v = NULL;
3478: PetscFunctionReturn(PETSC_SUCCESS);
3479: }
3481: PetscErrorCode MatDenseGetSubMatrix_SeqDense(Mat A, PetscInt rbegin, PetscInt rend, PetscInt cbegin, PetscInt cend, Mat *v)
3482: {
3483: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3485: PetscFunctionBegin;
3486: PetscCheck(!a->vecinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreColumnVec() first");
3487: PetscCheck(!a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseRestoreSubMatrix() first");
3488: if (a->cmat && (cend - cbegin != a->cmat->cmap->N || rend - rbegin != a->cmat->rmap->N)) PetscCall(MatDestroy(&a->cmat));
3489: if (!a->cmat) {
3490: PetscCall(MatCreateDense(PetscObjectComm((PetscObject)A), rend - rbegin, PETSC_DECIDE, rend - rbegin, cend - cbegin, a->v + rbegin + (size_t)cbegin * a->lda, &a->cmat));
3491: } else {
3492: PetscCall(MatDensePlaceArray(a->cmat, a->v + rbegin + (size_t)cbegin * a->lda));
3493: }
3494: PetscCall(MatDenseSetLDA(a->cmat, a->lda));
3495: a->matinuse = cbegin + 1;
3496: *v = a->cmat;
3497: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
3498: A->offloadmask = PETSC_OFFLOAD_CPU;
3499: #endif
3500: PetscFunctionReturn(PETSC_SUCCESS);
3501: }
3503: PetscErrorCode MatDenseRestoreSubMatrix_SeqDense(Mat A, Mat *v)
3504: {
3505: Mat_SeqDense *a = (Mat_SeqDense *)A->data;
3507: PetscFunctionBegin;
3508: PetscCheck(a->matinuse, PETSC_COMM_SELF, PETSC_ERR_ORDER, "Need to call MatDenseGetSubMatrix() first");
3509: PetscCheck(a->cmat, PETSC_COMM_SELF, PETSC_ERR_PLIB, "Missing internal column matrix");
3510: PetscCheck(*v == a->cmat, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONG, "Not the matrix obtained from MatDenseGetSubMatrix()");
3511: a->matinuse = 0;
3512: PetscCall(MatDenseResetArray(a->cmat));
3513: if (v) *v = NULL;
3514: #if defined(PETSC_HAVE_CUDA) || defined(PETSC_HAVE_HIP)
3515: A->offloadmask = PETSC_OFFLOAD_CPU;
3516: #endif
3517: PetscFunctionReturn(PETSC_SUCCESS);
3518: }
3520: /*MC
3521: MATSEQDENSE - MATSEQDENSE = "seqdense" - A matrix type to be used for sequential dense matrices.
3523: Options Database Key:
3524: . -mat_type seqdense - sets the matrix type to `MATSEQDENSE` during a call to `MatSetFromOptions()`
3526: Level: beginner
3528: .seealso: [](ch_matrices), `Mat`, `MATSEQDENSE`, `MatCreateSeqDense()`
3529: M*/
3530: PetscErrorCode MatCreate_SeqDense(Mat B)
3531: {
3532: Mat_SeqDense *b;
3533: PetscMPIInt size;
3535: PetscFunctionBegin;
3536: PetscCallMPI(MPI_Comm_size(PetscObjectComm((PetscObject)B), &size));
3537: PetscCheck(size <= 1, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONG, "Comm must be of size 1");
3539: PetscCall(PetscNew(&b));
3540: PetscCall(PetscMemcpy(B->ops, &MatOps_Values, sizeof(struct _MatOps)));
3541: B->data = (void *)b;
3543: b->roworiented = PETSC_TRUE;
3545: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatQRFactor_C", MatQRFactor_SeqDense));
3546: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetLDA_C", MatDenseGetLDA_SeqDense));
3547: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseSetLDA_C", MatDenseSetLDA_SeqDense));
3548: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetArray_C", MatDenseGetArray_SeqDense));
3549: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreArray_C", MatDenseRestoreArray_SeqDense));
3550: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDensePlaceArray_C", MatDensePlaceArray_SeqDense));
3551: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseResetArray_C", MatDenseResetArray_SeqDense));
3552: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseReplaceArray_C", MatDenseReplaceArray_SeqDense));
3553: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetArrayRead_C", MatDenseGetArray_SeqDense));
3554: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreArrayRead_C", MatDenseRestoreArray_SeqDense));
3555: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetArrayWrite_C", MatDenseGetArray_SeqDense));
3556: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreArrayWrite_C", MatDenseRestoreArray_SeqDense));
3557: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatConvert_seqdense_seqaij_C", MatConvert_SeqDense_SeqAIJ));
3558: #if defined(PETSC_HAVE_ELEMENTAL)
3559: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatConvert_seqdense_elemental_C", MatConvert_SeqDense_Elemental));
3560: #endif
3561: #if defined(PETSC_HAVE_SCALAPACK)
3562: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatConvert_seqdense_scalapack_C", MatConvert_Dense_ScaLAPACK));
3563: #endif
3564: #if defined(PETSC_HAVE_CUDA)
3565: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatConvert_seqdense_seqdensecuda_C", MatConvert_SeqDense_SeqDenseCUDA));
3566: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdensecuda_seqdensecuda_C", MatProductSetFromOptions_SeqDense));
3567: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdensecuda_seqdense_C", MatProductSetFromOptions_SeqDense));
3568: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdense_seqdensecuda_C", MatProductSetFromOptions_SeqDense));
3569: #endif
3570: #if defined(PETSC_HAVE_HIP)
3571: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatConvert_seqdense_seqdensehip_C", MatConvert_SeqDense_SeqDenseHIP));
3572: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdensehip_seqdensehip_C", MatProductSetFromOptions_SeqDense));
3573: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdensehip_seqdense_C", MatProductSetFromOptions_SeqDense));
3574: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdense_seqdensehip_C", MatProductSetFromOptions_SeqDense));
3575: #endif
3576: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatSeqDenseSetPreallocation_C", MatSeqDenseSetPreallocation_SeqDense));
3577: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqaij_seqdense_C", MatProductSetFromOptions_SeqAIJ_SeqDense));
3578: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqdense_seqdense_C", MatProductSetFromOptions_SeqDense));
3579: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqbaij_seqdense_C", MatProductSetFromOptions_SeqXBAIJ_SeqDense));
3580: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatProductSetFromOptions_seqsbaij_seqdense_C", MatProductSetFromOptions_SeqXBAIJ_SeqDense));
3582: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetColumn_C", MatDenseGetColumn_SeqDense));
3583: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreColumn_C", MatDenseRestoreColumn_SeqDense));
3584: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetColumnVec_C", MatDenseGetColumnVec_SeqDense));
3585: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreColumnVec_C", MatDenseRestoreColumnVec_SeqDense));
3586: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetColumnVecRead_C", MatDenseGetColumnVecRead_SeqDense));
3587: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreColumnVecRead_C", MatDenseRestoreColumnVecRead_SeqDense));
3588: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetColumnVecWrite_C", MatDenseGetColumnVecWrite_SeqDense));
3589: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreColumnVecWrite_C", MatDenseRestoreColumnVecWrite_SeqDense));
3590: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseGetSubMatrix_C", MatDenseGetSubMatrix_SeqDense));
3591: PetscCall(PetscObjectComposeFunction((PetscObject)B, "MatDenseRestoreSubMatrix_C", MatDenseRestoreSubMatrix_SeqDense));
3592: PetscCall(PetscObjectChangeTypeName((PetscObject)B, MATSEQDENSE));
3593: PetscFunctionReturn(PETSC_SUCCESS);
3594: }
3596: /*@C
3597: MatDenseGetColumn - gives access to a column of a dense matrix. This is only the local part of the column. You MUST call `MatDenseRestoreColumn()` to avoid memory bleeding.
3599: Not Collective
3601: Input Parameters:
3602: + mat - a `MATSEQDENSE` or `MATMPIDENSE` matrix
3603: - col - column index
3605: Output Parameter:
3606: . vals - pointer to the data
3608: Level: intermediate
3610: Note:
3611: Use `MatDenseGetColumnVec()` to get access to a column of a `MATDENSE` treated as a `Vec`
3613: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseRestoreColumn()`, `MatDenseGetColumnVec()`
3614: @*/
3615: PetscErrorCode MatDenseGetColumn(Mat A, PetscInt col, PetscScalar **vals)
3616: {
3617: PetscFunctionBegin;
3621: PetscUseMethod(A, "MatDenseGetColumn_C", (Mat, PetscInt, PetscScalar **), (A, col, vals));
3622: PetscFunctionReturn(PETSC_SUCCESS);
3623: }
3625: /*@C
3626: MatDenseRestoreColumn - returns access to a column of a `MATDENSE` matrix which is returned by `MatDenseGetColumn()`.
3628: Not Collective
3630: Input Parameters:
3631: + mat - a `MATSEQDENSE` or `MATMPIDENSE` matrix
3632: - vals - pointer to the data (may be `NULL`)
3634: Level: intermediate
3636: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MatDenseGetColumn()`
3637: @*/
3638: PetscErrorCode MatDenseRestoreColumn(Mat A, PetscScalar **vals)
3639: {
3640: PetscFunctionBegin;
3643: PetscUseMethod(A, "MatDenseRestoreColumn_C", (Mat, PetscScalar **), (A, vals));
3644: PetscFunctionReturn(PETSC_SUCCESS);
3645: }
3647: /*@
3648: MatDenseGetColumnVec - Gives read-write access to a column of a `MATDENSE` matrix, represented as a `Vec`.
3650: Collective
3652: Input Parameters:
3653: + mat - the `Mat` object
3654: - col - the column index
3656: Output Parameter:
3657: . v - the vector
3659: Level: intermediate
3661: Notes:
3662: The vector is owned by PETSc. Users need to call `MatDenseRestoreColumnVec()` when the vector is no longer needed.
3664: Use `MatDenseGetColumnVecRead()` to obtain read-only access or `MatDenseGetColumnVecWrite()` for write-only access.
3666: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVecRead()`, `MatDenseGetColumnVecWrite()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreColumnVecRead()`, `MatDenseRestoreColumnVecWrite()`, `MatDenseGetColumn()`
3667: @*/
3668: PetscErrorCode MatDenseGetColumnVec(Mat A, PetscInt col, Vec *v)
3669: {
3670: PetscFunctionBegin;
3675: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3676: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3677: PetscUseMethod(A, "MatDenseGetColumnVec_C", (Mat, PetscInt, Vec *), (A, col, v));
3678: PetscFunctionReturn(PETSC_SUCCESS);
3679: }
3681: /*@
3682: MatDenseRestoreColumnVec - Returns access to a column of a dense matrix obtained from MatDenseGetColumnVec().
3684: Collective
3686: Input Parameters:
3687: + mat - the Mat object
3688: . col - the column index
3689: - v - the Vec object (may be `NULL`)
3691: Level: intermediate
3693: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseGetColumnVecRead()`, `MatDenseGetColumnVecWrite()`, `MatDenseRestoreColumnVecRead()`, `MatDenseRestoreColumnVecWrite()`
3694: @*/
3695: PetscErrorCode MatDenseRestoreColumnVec(Mat A, PetscInt col, Vec *v)
3696: {
3697: PetscFunctionBegin;
3701: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3702: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3703: PetscUseMethod(A, "MatDenseRestoreColumnVec_C", (Mat, PetscInt, Vec *), (A, col, v));
3704: PetscFunctionReturn(PETSC_SUCCESS);
3705: }
3707: /*@
3708: MatDenseGetColumnVecRead - Gives read-only access to a column of a dense matrix, represented as a Vec.
3710: Collective
3712: Input Parameters:
3713: + mat - the `Mat` object
3714: - col - the column index
3716: Output Parameter:
3717: . v - the vector
3719: Level: intermediate
3721: Notes:
3722: The vector is owned by PETSc and users cannot modify it.
3724: Users need to call `MatDenseRestoreColumnVecRead()` when the vector is no longer needed.
3726: Use `MatDenseGetColumnVec()` to obtain read-write access or `MatDenseGetColumnVecWrite()` for write-only access.
3728: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseGetColumnVecWrite()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreColumnVecRead()`, `MatDenseRestoreColumnVecWrite()`
3729: @*/
3730: PetscErrorCode MatDenseGetColumnVecRead(Mat A, PetscInt col, Vec *v)
3731: {
3732: PetscFunctionBegin;
3737: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3738: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3739: PetscUseMethod(A, "MatDenseGetColumnVecRead_C", (Mat, PetscInt, Vec *), (A, col, v));
3740: PetscFunctionReturn(PETSC_SUCCESS);
3741: }
3743: /*@
3744: MatDenseRestoreColumnVecRead - Returns access to a column of a dense matrix obtained from MatDenseGetColumnVecRead().
3746: Collective
3748: Input Parameters:
3749: + mat - the `Mat` object
3750: . col - the column index
3751: - v - the Vec object (may be `NULL`)
3753: Level: intermediate
3755: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseGetColumnVecRead()`, `MatDenseGetColumnVecWrite()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreColumnVecWrite()`
3756: @*/
3757: PetscErrorCode MatDenseRestoreColumnVecRead(Mat A, PetscInt col, Vec *v)
3758: {
3759: PetscFunctionBegin;
3763: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3764: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3765: PetscUseMethod(A, "MatDenseRestoreColumnVecRead_C", (Mat, PetscInt, Vec *), (A, col, v));
3766: PetscFunctionReturn(PETSC_SUCCESS);
3767: }
3769: /*@
3770: MatDenseGetColumnVecWrite - Gives write-only access to a column of a dense matrix, represented as a Vec.
3772: Collective
3774: Input Parameters:
3775: + mat - the `Mat` object
3776: - col - the column index
3778: Output Parameter:
3779: . v - the vector
3781: Level: intermediate
3783: Notes:
3784: The vector is owned by PETSc. Users need to call `MatDenseRestoreColumnVecWrite()` when the vector is no longer needed.
3786: Use `MatDenseGetColumnVec()` to obtain read-write access or `MatDenseGetColumnVecRead()` for read-only access.
3788: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseGetColumnVecRead()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreColumnVecRead()`, `MatDenseRestoreColumnVecWrite()`
3789: @*/
3790: PetscErrorCode MatDenseGetColumnVecWrite(Mat A, PetscInt col, Vec *v)
3791: {
3792: PetscFunctionBegin;
3797: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3798: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3799: PetscUseMethod(A, "MatDenseGetColumnVecWrite_C", (Mat, PetscInt, Vec *), (A, col, v));
3800: PetscFunctionReturn(PETSC_SUCCESS);
3801: }
3803: /*@
3804: MatDenseRestoreColumnVecWrite - Returns access to a column of a dense matrix obtained from MatDenseGetColumnVecWrite().
3806: Collective
3808: Input Parameters:
3809: + mat - the `Mat` object
3810: . col - the column index
3811: - v - the `Vec` object (may be `NULL`)
3813: Level: intermediate
3815: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseGetColumnVecRead()`, `MatDenseGetColumnVecWrite()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreColumnVecRead()`
3816: @*/
3817: PetscErrorCode MatDenseRestoreColumnVecWrite(Mat A, PetscInt col, Vec *v)
3818: {
3819: PetscFunctionBegin;
3823: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3824: PetscCheck(col >= 0 && col < A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid col %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT ")", col, A->cmap->N);
3825: PetscUseMethod(A, "MatDenseRestoreColumnVecWrite_C", (Mat, PetscInt, Vec *), (A, col, v));
3826: PetscFunctionReturn(PETSC_SUCCESS);
3827: }
3829: /*@
3830: MatDenseGetSubMatrix - Gives access to a block of rows and columns of a dense matrix, represented as a Mat.
3832: Collective
3834: Input Parameters:
3835: + mat - the Mat object
3836: . rbegin - the first global row index in the block (if `PETSC_DECIDE`, is 0)
3837: . rend - the global row index past the last one in the block (if `PETSC_DECIDE`, is `M`)
3838: . cbegin - the first global column index in the block (if `PETSC_DECIDE`, is 0)
3839: - cend - the global column index past the last one in the block (if `PETSC_DECIDE`, is `N`)
3841: Output Parameter:
3842: . v - the matrix
3844: Level: intermediate
3846: Notes:
3847: The matrix is owned by PETSc. Users need to call `MatDenseRestoreSubMatrix()` when the matrix is no longer needed.
3849: The output matrix is not redistributed by PETSc, so depending on the values of `rbegin` and `rend`, some processes may have no local rows.
3851: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseRestoreColumnVec()`, `MatDenseRestoreSubMatrix()`
3852: @*/
3853: PetscErrorCode MatDenseGetSubMatrix(Mat A, PetscInt rbegin, PetscInt rend, PetscInt cbegin, PetscInt cend, Mat *v)
3854: {
3855: PetscFunctionBegin;
3863: if (rbegin == PETSC_DECIDE) rbegin = 0;
3864: if (rend == PETSC_DECIDE) rend = A->rmap->N;
3865: if (cbegin == PETSC_DECIDE) cbegin = 0;
3866: if (cend == PETSC_DECIDE) cend = A->cmap->N;
3867: PetscCheck(A->preallocated, PetscObjectComm((PetscObject)A), PETSC_ERR_ORDER, "Matrix not preallocated");
3868: PetscCheck(rbegin >= 0 && rbegin <= A->rmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid rbegin %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT "]", rbegin, A->rmap->N);
3869: PetscCheck(rend >= rbegin && rend <= A->rmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid rend %" PetscInt_FMT ", should be in [%" PetscInt_FMT ",%" PetscInt_FMT "]", rend, rbegin, A->rmap->N);
3870: PetscCheck(cbegin >= 0 && cbegin <= A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid cbegin %" PetscInt_FMT ", should be in [0,%" PetscInt_FMT "]", cbegin, A->cmap->N);
3871: PetscCheck(cend >= cbegin && cend <= A->cmap->N, PetscObjectComm((PetscObject)A), PETSC_ERR_ARG_WRONG, "Invalid cend %" PetscInt_FMT ", should be in [%" PetscInt_FMT ",%" PetscInt_FMT "]", cend, cbegin, A->cmap->N);
3872: PetscUseMethod(A, "MatDenseGetSubMatrix_C", (Mat, PetscInt, PetscInt, PetscInt, PetscInt, Mat *), (A, rbegin, rend, cbegin, cend, v));
3873: PetscFunctionReturn(PETSC_SUCCESS);
3874: }
3876: /*@
3877: MatDenseRestoreSubMatrix - Returns access to a block of columns of a dense matrix obtained from MatDenseGetSubMatrix().
3879: Collective
3881: Input Parameters:
3882: + mat - the `Mat` object
3883: - v - the `Mat` object (may be `NULL`)
3885: Level: intermediate
3887: .seealso: [](ch_matrices), `Mat`, `MATDENSE`, `MATDENSECUDA`, `MATDENSEHIP`, `MatDenseGetColumnVec()`, `MatDenseRestoreColumnVec()`, `MatDenseGetSubMatrix()`
3888: @*/
3889: PetscErrorCode MatDenseRestoreSubMatrix(Mat A, Mat *v)
3890: {
3891: PetscFunctionBegin;
3895: PetscUseMethod(A, "MatDenseRestoreSubMatrix_C", (Mat, Mat *), (A, v));
3896: PetscFunctionReturn(PETSC_SUCCESS);
3897: }
3899: #include <petscblaslapack.h>
3900: #include <petsc/private/kernels/blockinvert.h>
3902: PetscErrorCode MatSeqDenseInvert(Mat A)
3903: {
3904: PetscInt m;
3905: const PetscReal shift = 0.0;
3906: PetscBool allowzeropivot, zeropivotdetected = PETSC_FALSE;
3907: PetscScalar *values;
3909: PetscFunctionBegin;
3911: PetscCall(MatDenseGetArray(A, &values));
3912: PetscCall(MatGetLocalSize(A, &m, NULL));
3913: allowzeropivot = PetscNot(A->erroriffailure);
3914: /* factor and invert each block */
3915: switch (m) {
3916: case 1:
3917: values[0] = (PetscScalar)1.0 / (values[0] + shift);
3918: break;
3919: case 2:
3920: PetscCall(PetscKernel_A_gets_inverse_A_2(values, shift, allowzeropivot, &zeropivotdetected));
3921: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3922: break;
3923: case 3:
3924: PetscCall(PetscKernel_A_gets_inverse_A_3(values, shift, allowzeropivot, &zeropivotdetected));
3925: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3926: break;
3927: case 4:
3928: PetscCall(PetscKernel_A_gets_inverse_A_4(values, shift, allowzeropivot, &zeropivotdetected));
3929: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3930: break;
3931: case 5: {
3932: PetscScalar work[25];
3933: PetscInt ipvt[5];
3935: PetscCall(PetscKernel_A_gets_inverse_A_5(values, ipvt, work, shift, allowzeropivot, &zeropivotdetected));
3936: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3937: } break;
3938: case 6:
3939: PetscCall(PetscKernel_A_gets_inverse_A_6(values, shift, allowzeropivot, &zeropivotdetected));
3940: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3941: break;
3942: case 7:
3943: PetscCall(PetscKernel_A_gets_inverse_A_7(values, shift, allowzeropivot, &zeropivotdetected));
3944: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3945: break;
3946: default: {
3947: PetscInt *v_pivots, *IJ, j;
3948: PetscScalar *v_work;
3950: PetscCall(PetscMalloc3(m, &v_work, m, &v_pivots, m, &IJ));
3951: for (j = 0; j < m; j++) IJ[j] = j;
3952: PetscCall(PetscKernel_A_gets_inverse_A(m, values, v_pivots, v_work, allowzeropivot, &zeropivotdetected));
3953: if (zeropivotdetected) A->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
3954: PetscCall(PetscFree3(v_work, v_pivots, IJ));
3955: }
3956: }
3957: PetscCall(MatDenseRestoreArray(A, &values));
3958: PetscFunctionReturn(PETSC_SUCCESS);
3959: }