Actual source code: matimpl.h
1: #pragma once
3: #include <petscmat.h>
4: #include <petscmatcoarsen.h>
5: #include <petsc/private/petscimpl.h>
7: PETSC_EXTERN PetscBool MatRegisterAllCalled;
8: PETSC_EXTERN PetscBool MatSeqAIJRegisterAllCalled;
9: PETSC_EXTERN PetscBool MatOrderingRegisterAllCalled;
10: PETSC_EXTERN PetscBool MatColoringRegisterAllCalled;
11: PETSC_EXTERN PetscBool MatPartitioningRegisterAllCalled;
12: PETSC_EXTERN PetscBool MatCoarsenRegisterAllCalled;
13: PETSC_EXTERN PetscErrorCode MatRegisterAll(void);
14: PETSC_EXTERN PetscErrorCode MatOrderingRegisterAll(void);
15: PETSC_EXTERN PetscErrorCode MatColoringRegisterAll(void);
16: PETSC_EXTERN PetscErrorCode MatPartitioningRegisterAll(void);
17: PETSC_EXTERN PetscErrorCode MatCoarsenRegisterAll(void);
18: PETSC_EXTERN PetscErrorCode MatSeqAIJRegisterAll(void);
20: /* Gets the root type of the input matrix's type (e.g., MATAIJ for MATSEQAIJ) */
21: PETSC_EXTERN PetscErrorCode MatGetRootType_Private(Mat, MatType *);
23: /* Gets the MPI type corresponding to the input matrix's type (e.g., MATMPIAIJ for MATSEQAIJ) */
24: PETSC_EXTERN PetscErrorCode MatGetMPIMatType_Private(Mat, MatType *);
26: /*
27: This file defines the parts of the matrix data structure that are
28: shared by all matrix types.
29: */
31: /*
32: If you add entries here also add them to the MATOP enum
33: in include/petscmat.h and src/mat/f90-mod/petscmat.h
34: */
35: typedef struct _MatOps *MatOps;
36: struct _MatOps {
37: /* 0*/
38: PetscErrorCode (*setvalues)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], const PetscScalar[], InsertMode);
39: PetscErrorCode (*getrow)(Mat, PetscInt, PetscInt *, PetscInt *[], PetscScalar *[]);
40: PetscErrorCode (*restorerow)(Mat, PetscInt, PetscInt *, PetscInt *[], PetscScalar *[]);
41: PetscErrorCode (*mult)(Mat, Vec, Vec);
42: PetscErrorCode (*multadd)(Mat, Vec, Vec, Vec);
43: /* 5*/
44: PetscErrorCode (*multtranspose)(Mat, Vec, Vec);
45: PetscErrorCode (*multtransposeadd)(Mat, Vec, Vec, Vec);
46: PetscErrorCode (*solve)(Mat, Vec, Vec);
47: PetscErrorCode (*solveadd)(Mat, Vec, Vec, Vec);
48: PetscErrorCode (*solvetranspose)(Mat, Vec, Vec);
49: /*10*/
50: PetscErrorCode (*solvetransposeadd)(Mat, Vec, Vec, Vec);
51: PetscErrorCode (*lufactor)(Mat, IS, IS, const MatFactorInfo *);
52: PetscErrorCode (*choleskyfactor)(Mat, IS, const MatFactorInfo *);
53: PetscErrorCode (*sor)(Mat, Vec, PetscReal, MatSORType, PetscReal, PetscInt, PetscInt, Vec);
54: PetscErrorCode (*transpose)(Mat, MatReuse, Mat *);
55: /*15*/
56: PetscErrorCode (*getinfo)(Mat, MatInfoType, MatInfo *);
57: PetscErrorCode (*equal)(Mat, Mat, PetscBool *);
58: PetscErrorCode (*getdiagonal)(Mat, Vec);
59: PetscErrorCode (*diagonalscale)(Mat, Vec, Vec);
60: PetscErrorCode (*norm)(Mat, NormType, PetscReal *);
61: /*20*/
62: PetscErrorCode (*assemblybegin)(Mat, MatAssemblyType);
63: PetscErrorCode (*assemblyend)(Mat, MatAssemblyType);
64: PetscErrorCode (*setoption)(Mat, MatOption, PetscBool);
65: PetscErrorCode (*zeroentries)(Mat);
66: /*24*/
67: PetscErrorCode (*zerorows)(Mat, PetscInt, const PetscInt[], PetscScalar, Vec, Vec);
68: PetscErrorCode (*lufactorsymbolic)(Mat, Mat, IS, IS, const MatFactorInfo *);
69: PetscErrorCode (*lufactornumeric)(Mat, Mat, const MatFactorInfo *);
70: PetscErrorCode (*choleskyfactorsymbolic)(Mat, Mat, IS, const MatFactorInfo *);
71: PetscErrorCode (*choleskyfactornumeric)(Mat, Mat, const MatFactorInfo *);
72: /*29*/
73: PetscErrorCode (*setup)(Mat);
74: PetscErrorCode (*ilufactorsymbolic)(Mat, Mat, IS, IS, const MatFactorInfo *);
75: PetscErrorCode (*iccfactorsymbolic)(Mat, Mat, IS, const MatFactorInfo *);
76: PetscErrorCode (*getdiagonalblock)(Mat, Mat *);
77: PetscErrorCode (*setinf)(Mat);
78: /*34*/
79: PetscErrorCode (*duplicate)(Mat, MatDuplicateOption, Mat *);
80: PetscErrorCode (*forwardsolve)(Mat, Vec, Vec);
81: PetscErrorCode (*backwardsolve)(Mat, Vec, Vec);
82: PetscErrorCode (*ilufactor)(Mat, IS, IS, const MatFactorInfo *);
83: PetscErrorCode (*iccfactor)(Mat, IS, const MatFactorInfo *);
84: /*39*/
85: PetscErrorCode (*axpy)(Mat, PetscScalar, Mat, MatStructure);
86: PetscErrorCode (*createsubmatrices)(Mat, PetscInt, const IS[], const IS[], MatReuse, Mat *[]);
87: PetscErrorCode (*increaseoverlap)(Mat, PetscInt, IS[], PetscInt);
88: PetscErrorCode (*getvalues)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], PetscScalar[]);
89: PetscErrorCode (*copy)(Mat, Mat, MatStructure);
90: /*44*/
91: PetscErrorCode (*getrowmax)(Mat, Vec, PetscInt[]);
92: PetscErrorCode (*scale)(Mat, PetscScalar);
93: PetscErrorCode (*shift)(Mat, PetscScalar);
94: PetscErrorCode (*diagonalset)(Mat, Vec, InsertMode);
95: PetscErrorCode (*zerorowscolumns)(Mat, PetscInt, const PetscInt[], PetscScalar, Vec, Vec);
96: /*49*/
97: PetscErrorCode (*setrandom)(Mat, PetscRandom);
98: PetscErrorCode (*getrowij)(Mat, PetscInt, PetscBool, PetscBool, PetscInt *, const PetscInt *[], const PetscInt *[], PetscBool *);
99: PetscErrorCode (*restorerowij)(Mat, PetscInt, PetscBool, PetscBool, PetscInt *, const PetscInt *[], const PetscInt *[], PetscBool *);
100: PetscErrorCode (*getcolumnij)(Mat, PetscInt, PetscBool, PetscBool, PetscInt *, const PetscInt *[], const PetscInt *[], PetscBool *);
101: PetscErrorCode (*restorecolumnij)(Mat, PetscInt, PetscBool, PetscBool, PetscInt *, const PetscInt *[], const PetscInt *[], PetscBool *);
102: /*54*/
103: PetscErrorCode (*fdcoloringcreate)(Mat, ISColoring, MatFDColoring);
104: PetscErrorCode (*coloringpatch)(Mat, PetscInt, PetscInt, ISColoringValue[], ISColoring *);
105: PetscErrorCode (*setunfactored)(Mat);
106: PetscErrorCode (*permute)(Mat, IS, IS, Mat *);
107: PetscErrorCode (*setvaluesblocked)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], const PetscScalar[], InsertMode);
108: /*59*/
109: PetscErrorCode (*createsubmatrix)(Mat, IS, IS, MatReuse, Mat *);
110: PetscErrorCode (*destroy)(Mat);
111: PetscErrorCode (*view)(Mat, PetscViewer);
112: PetscErrorCode (*convertfrom)(Mat, MatType, MatReuse, Mat *);
113: PetscErrorCode (*placeholder_63)(void);
114: /*64*/
115: PetscErrorCode (*matmatmultsymbolic)(Mat, Mat, Mat, PetscReal, Mat);
116: PetscErrorCode (*matmatmultnumeric)(Mat, Mat, Mat, Mat);
117: PetscErrorCode (*setlocaltoglobalmapping)(Mat, ISLocalToGlobalMapping, ISLocalToGlobalMapping);
118: PetscErrorCode (*setvalueslocal)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], const PetscScalar[], InsertMode);
119: PetscErrorCode (*zerorowslocal)(Mat, PetscInt, const PetscInt[], PetscScalar, Vec, Vec);
120: /*69*/
121: PetscErrorCode (*getrowmaxabs)(Mat, Vec, PetscInt[]);
122: PetscErrorCode (*getrowminabs)(Mat, Vec, PetscInt[]);
123: PetscErrorCode (*convert)(Mat, MatType, MatReuse, Mat *);
124: PetscErrorCode (*hasoperation)(Mat, MatOperation, PetscBool *);
125: PetscErrorCode (*placeholder_73)(void);
126: /*74*/
127: PetscErrorCode (*setvaluesadifor)(Mat, PetscInt, void *);
128: PetscErrorCode (*fdcoloringapply)(Mat, MatFDColoring, Vec, void *);
129: PetscErrorCode (*setfromoptions)(Mat, PetscOptionItems *);
130: PetscErrorCode (*placeholder_77)(void);
131: PetscErrorCode (*placeholder_78)(void);
132: /*79*/
133: PetscErrorCode (*findzerodiagonals)(Mat, IS *);
134: PetscErrorCode (*mults)(Mat, Vecs, Vecs);
135: PetscErrorCode (*solves)(Mat, Vecs, Vecs);
136: PetscErrorCode (*getinertia)(Mat, PetscInt *, PetscInt *, PetscInt *);
137: PetscErrorCode (*load)(Mat, PetscViewer);
138: /*84*/
139: PetscErrorCode (*issymmetric)(Mat, PetscReal, PetscBool *);
140: PetscErrorCode (*ishermitian)(Mat, PetscReal, PetscBool *);
141: PetscErrorCode (*isstructurallysymmetric)(Mat, PetscBool *);
142: PetscErrorCode (*setvaluesblockedlocal)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], const PetscScalar[], InsertMode);
143: PetscErrorCode (*getvecs)(Mat, Vec *, Vec *);
144: /*89*/
145: PetscErrorCode (*placeholder_89)(void);
146: PetscErrorCode (*matmultsymbolic)(Mat, Mat, PetscReal, Mat);
147: PetscErrorCode (*matmultnumeric)(Mat, Mat, Mat);
148: PetscErrorCode (*placeholder_92)(void);
149: PetscErrorCode (*ptapsymbolic)(Mat, Mat, PetscReal, Mat); /* double dispatch wrapper routine */
150: /*94*/
151: PetscErrorCode (*ptapnumeric)(Mat, Mat, Mat); /* double dispatch wrapper routine */
152: PetscErrorCode (*placeholder_95)(void);
153: PetscErrorCode (*mattransposemultsymbolic)(Mat, Mat, PetscReal, Mat);
154: PetscErrorCode (*mattransposemultnumeric)(Mat, Mat, Mat);
155: PetscErrorCode (*bindtocpu)(Mat, PetscBool);
156: /*99*/
157: PetscErrorCode (*productsetfromoptions)(Mat);
158: PetscErrorCode (*productsymbolic)(Mat);
159: PetscErrorCode (*productnumeric)(Mat);
160: PetscErrorCode (*conjugate)(Mat); /* complex conjugate */
161: PetscErrorCode (*viewnative)(Mat, PetscViewer);
162: /*104*/
163: PetscErrorCode (*setvaluesrow)(Mat, PetscInt, const PetscScalar[]);
164: PetscErrorCode (*realpart)(Mat);
165: PetscErrorCode (*imaginarypart)(Mat);
166: PetscErrorCode (*getrowuppertriangular)(Mat);
167: PetscErrorCode (*restorerowuppertriangular)(Mat);
168: /*109*/
169: PetscErrorCode (*matsolve)(Mat, Mat, Mat);
170: PetscErrorCode (*matsolvetranspose)(Mat, Mat, Mat);
171: PetscErrorCode (*getrowmin)(Mat, Vec, PetscInt[]);
172: PetscErrorCode (*getcolumnvector)(Mat, Vec, PetscInt);
173: PetscErrorCode (*missingdiagonal)(Mat, PetscBool *, PetscInt *);
174: /*114*/
175: PetscErrorCode (*getseqnonzerostructure)(Mat, Mat *);
176: PetscErrorCode (*create)(Mat);
177: PetscErrorCode (*getghosts)(Mat, PetscInt *, const PetscInt *[]);
178: PetscErrorCode (*getlocalsubmatrix)(Mat, IS, IS, Mat *);
179: PetscErrorCode (*restorelocalsubmatrix)(Mat, IS, IS, Mat *);
180: /*119*/
181: PetscErrorCode (*multdiagonalblock)(Mat, Vec, Vec);
182: PetscErrorCode (*hermitiantranspose)(Mat, MatReuse, Mat *);
183: PetscErrorCode (*multhermitiantranspose)(Mat, Vec, Vec);
184: PetscErrorCode (*multhermitiantransposeadd)(Mat, Vec, Vec, Vec);
185: PetscErrorCode (*getmultiprocblock)(Mat, MPI_Comm, MatReuse, Mat *);
186: /*124*/
187: PetscErrorCode (*findnonzerorows)(Mat, IS *);
188: PetscErrorCode (*getcolumnreductions)(Mat, PetscInt, PetscReal *);
189: PetscErrorCode (*invertblockdiagonal)(Mat, const PetscScalar **);
190: PetscErrorCode (*invertvariableblockdiagonal)(Mat, PetscInt, const PetscInt *, PetscScalar *);
191: PetscErrorCode (*createsubmatricesmpi)(Mat, PetscInt, const IS[], const IS[], MatReuse, Mat **);
192: /*129*/
193: PetscErrorCode (*setvaluesbatch)(Mat, PetscInt, PetscInt, PetscInt *, const PetscScalar *);
194: PetscErrorCode (*placeholder_130)(void);
195: PetscErrorCode (*transposematmultsymbolic)(Mat, Mat, PetscReal, Mat);
196: PetscErrorCode (*transposematmultnumeric)(Mat, Mat, Mat);
197: PetscErrorCode (*transposecoloringcreate)(Mat, ISColoring, MatTransposeColoring);
198: /*134*/
199: PetscErrorCode (*transcoloringapplysptoden)(MatTransposeColoring, Mat, Mat);
200: PetscErrorCode (*transcoloringapplydentosp)(MatTransposeColoring, Mat, Mat);
201: PetscErrorCode (*placeholder_136)(void);
202: PetscErrorCode (*rartsymbolic)(Mat, Mat, PetscReal, Mat); /* double dispatch wrapper routine */
203: PetscErrorCode (*rartnumeric)(Mat, Mat, Mat); /* double dispatch wrapper routine */
204: /*139*/
205: PetscErrorCode (*setblocksizes)(Mat, PetscInt, PetscInt);
206: PetscErrorCode (*aypx)(Mat, PetscScalar, Mat, MatStructure);
207: PetscErrorCode (*residual)(Mat, Vec, Vec, Vec);
208: PetscErrorCode (*fdcoloringsetup)(Mat, ISColoring, MatFDColoring);
209: PetscErrorCode (*findoffblockdiagonalentries)(Mat, IS *);
210: PetscErrorCode (*creatempimatconcatenateseqmat)(MPI_Comm, Mat, PetscInt, MatReuse, Mat *);
211: /*145*/
212: PetscErrorCode (*destroysubmatrices)(PetscInt, Mat *[]);
213: PetscErrorCode (*mattransposesolve)(Mat, Mat, Mat);
214: PetscErrorCode (*getvalueslocal)(Mat, PetscInt, const PetscInt[], PetscInt, const PetscInt[], PetscScalar[]);
215: PetscErrorCode (*creategraph)(Mat, PetscBool, PetscBool, PetscReal, PetscInt, PetscInt[], Mat *);
216: PetscErrorCode (*dummy)(Mat);
217: /*150*/
218: PetscErrorCode (*transposesymbolic)(Mat, Mat *);
219: PetscErrorCode (*eliminatezeros)(Mat, PetscBool);
220: PetscErrorCode (*getrowsumabs)(Mat, Vec);
221: };
222: /*
223: If you add MatOps entries above also add them to the MATOP enum
224: in include/petscmat.h and src/mat/f90-mod/petscmat.h
225: */
227: #include <petscsys.h>
229: typedef struct _p_MatRootName *MatRootName;
230: struct _p_MatRootName {
231: char *rname, *sname, *mname;
232: MatRootName next;
233: };
235: PETSC_EXTERN MatRootName MatRootNameList;
237: /*
238: Utility private matrix routines used outside Mat
239: */
240: PETSC_SINGLE_LIBRARY_INTERN PetscErrorCode MatFindNonzeroRowsOrCols_Basic(Mat, PetscBool, PetscReal, IS *);
241: PETSC_SINGLE_LIBRARY_INTERN PetscErrorCode MatShellGetScalingShifts(Mat, PetscScalar *, PetscScalar *, Vec *, Vec *, Vec *, Mat *, IS *, IS *);
243: #define MAT_SHELL_NOT_ALLOWED (void *)-1
245: /*
246: Utility private matrix routines
247: */
248: PETSC_INTERN PetscErrorCode MatConvert_Basic(Mat, MatType, MatReuse, Mat *);
249: PETSC_INTERN PetscErrorCode MatConvert_Shell(Mat, MatType, MatReuse, Mat *);
250: PETSC_INTERN PetscErrorCode MatConvertFrom_Shell(Mat, MatType, MatReuse, Mat *);
251: PETSC_INTERN PetscErrorCode MatShellSetContext_Immutable(Mat X, void *ctx);
252: PETSC_INTERN PetscErrorCode MatShellSetContextDestroy_Immutable(Mat X, PetscErrorCode (*f)(void *));
253: PETSC_INTERN PetscErrorCode MatShellSetManageScalingShifts_Immutable(Mat X);
254: PETSC_INTERN PetscErrorCode MatCopy_Basic(Mat, Mat, MatStructure);
255: PETSC_INTERN PetscErrorCode MatDiagonalSet_Default(Mat, Vec, InsertMode);
256: #if defined(PETSC_HAVE_SCALAPACK)
257: PETSC_INTERN PetscErrorCode MatConvert_Dense_ScaLAPACK(Mat, MatType, MatReuse, Mat *);
258: #endif
259: PETSC_INTERN PetscErrorCode MatSetPreallocationCOO_Basic(Mat, PetscCount, PetscInt[], PetscInt[]);
260: PETSC_INTERN PetscErrorCode MatSetValuesCOO_Basic(Mat, const PetscScalar[], InsertMode);
262: /* This can be moved to the public header after implementing some missing MatProducts */
263: PETSC_INTERN PetscErrorCode MatCreateFromISLocalToGlobalMapping(ISLocalToGlobalMapping, Mat, PetscBool, PetscBool, MatType, Mat *);
265: /* these callbacks rely on the old matrix function pointers for
266: matmat operations. They are unsafe, and should be removed.
267: However, the amount of work needed to clean up all the
268: implementations is not negligible */
269: PETSC_INTERN PetscErrorCode MatProductSymbolic_AB(Mat);
270: PETSC_INTERN PetscErrorCode MatProductNumeric_AB(Mat);
271: PETSC_INTERN PetscErrorCode MatProductSymbolic_AtB(Mat);
272: PETSC_INTERN PetscErrorCode MatProductNumeric_AtB(Mat);
273: PETSC_INTERN PetscErrorCode MatProductSymbolic_ABt(Mat);
274: PETSC_INTERN PetscErrorCode MatProductNumeric_ABt(Mat);
275: PETSC_INTERN PetscErrorCode MatProductNumeric_PtAP(Mat);
276: PETSC_INTERN PetscErrorCode MatProductNumeric_RARt(Mat);
277: PETSC_INTERN PetscErrorCode MatProductSymbolic_ABC(Mat);
278: PETSC_INTERN PetscErrorCode MatProductNumeric_ABC(Mat);
280: PETSC_INTERN PetscErrorCode MatProductCreate_Private(Mat, Mat, Mat, Mat);
281: /* this callback handles all the different triple products and
282: does not rely on the function pointers; used by cuSPARSE/hipSPARSE and KOKKOS-KERNELS */
283: PETSC_INTERN PetscErrorCode MatProductSymbolic_ABC_Basic(Mat);
285: /* CreateGraph is common to AIJ seq and mpi */
286: PETSC_INTERN PetscErrorCode MatCreateGraph_Simple_AIJ(Mat, PetscBool, PetscBool, PetscReal, PetscInt, PetscInt[], Mat *);
288: #if defined(PETSC_CLANG_STATIC_ANALYZER)
289: template <typename Tm>
290: extern void MatCheckPreallocated(Tm, int);
291: template <typename Tm>
292: extern void MatCheckProduct(Tm, int);
293: #else /* PETSC_CLANG_STATIC_ANALYZER */
294: #define MatCheckPreallocated(A, arg) \
295: do { \
296: if (!(A)->preallocated) PetscCall(MatSetUp(A)); \
297: } while (0)
299: #if defined(PETSC_USE_DEBUG)
300: #define MatCheckProduct(A, arg) \
301: do { \
302: PetscCheck((A)->product, PETSC_COMM_SELF, PETSC_ERR_ARG_WRONGSTATE, "Argument %d \"%s\" is not a matrix obtained from MatProductCreate()", (arg), #A); \
303: } while (0)
304: #else
305: #define MatCheckProduct(A, arg) \
306: do { \
307: } while (0)
308: #endif
309: #endif /* PETSC_CLANG_STATIC_ANALYZER */
311: /*
312: The stash is used to temporarily store inserted matrix values that
313: belong to another processor. During the assembly phase the stashed
314: values are moved to the correct processor and
315: */
317: typedef struct _MatStashSpace *PetscMatStashSpace;
319: struct _MatStashSpace {
320: PetscMatStashSpace next;
321: PetscScalar *space_head, *val;
322: PetscInt *idx, *idy;
323: PetscInt total_space_size;
324: PetscInt local_used;
325: PetscInt local_remaining;
326: };
328: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceGet(PetscInt, PetscInt, PetscMatStashSpace *);
329: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceContiguous(PetscInt, PetscMatStashSpace *, PetscScalar *, PetscInt *, PetscInt *);
330: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceDestroy(PetscMatStashSpace *);
332: typedef struct {
333: PetscInt count;
334: } MatStashHeader;
336: typedef struct {
337: void *buffer; /* Of type blocktype, dynamically constructed */
338: PetscInt count;
339: char pending;
340: } MatStashFrame;
342: typedef struct _MatStash MatStash;
343: struct _MatStash {
344: PetscInt nmax; /* maximum stash size */
345: PetscInt umax; /* user specified max-size */
346: PetscInt oldnmax; /* the nmax value used previously */
347: PetscInt n; /* stash size */
348: PetscInt bs; /* block size of the stash */
349: PetscInt reallocs; /* preserve the no of mallocs invoked */
350: PetscMatStashSpace space_head, space; /* linked list to hold stashed global row/column numbers and matrix values */
352: PetscErrorCode (*ScatterBegin)(Mat, MatStash *, PetscInt *);
353: PetscErrorCode (*ScatterGetMesg)(MatStash *, PetscMPIInt *, PetscInt **, PetscInt **, PetscScalar **, PetscInt *);
354: PetscErrorCode (*ScatterEnd)(MatStash *);
355: PetscErrorCode (*ScatterDestroy)(MatStash *);
357: /* The following variables are used for communication */
358: MPI_Comm comm;
359: PetscMPIInt size, rank;
360: PetscMPIInt tag1, tag2;
361: MPI_Request *send_waits; /* array of send requests */
362: MPI_Request *recv_waits; /* array of receive requests */
363: MPI_Status *send_status; /* array of send status */
364: PetscInt nsends, nrecvs; /* numbers of sends and receives */
365: PetscScalar *svalues; /* sending data */
366: PetscInt *sindices;
367: PetscScalar **rvalues; /* receiving data (values) */
368: PetscInt **rindices; /* receiving data (indices) */
369: PetscInt nprocessed; /* number of messages already processed */
370: PetscMPIInt *flg_v; /* indicates what messages have arrived so far and from whom */
371: PetscBool reproduce;
372: PetscInt reproduce_count;
374: /* The following variables are used for BTS communication */
375: PetscBool first_assembly_done; /* Is the first time matrix assembly done? */
376: PetscBool use_status; /* Use MPI_Status to determine number of items in each message */
377: PetscMPIInt nsendranks;
378: PetscMPIInt nrecvranks;
379: PetscMPIInt *sendranks;
380: PetscMPIInt *recvranks;
381: MatStashHeader *sendhdr, *recvhdr;
382: MatStashFrame *sendframes; /* pointers to the main messages */
383: MatStashFrame *recvframes;
384: MatStashFrame *recvframe_active;
385: PetscInt recvframe_i; /* index of block within active frame */
386: PetscMPIInt recvframe_count; /* Count actually sent for current frame */
387: PetscInt recvcount; /* Number of receives processed so far */
388: PetscMPIInt *some_indices; /* From last call to MPI_Waitsome */
389: MPI_Status *some_statuses; /* Statuses from last call to MPI_Waitsome */
390: PetscMPIInt some_count; /* Number of requests completed in last call to MPI_Waitsome */
391: PetscMPIInt some_i; /* Index of request currently being processed */
392: MPI_Request *sendreqs;
393: MPI_Request *recvreqs;
394: PetscSegBuffer segsendblocks;
395: PetscSegBuffer segrecvframe;
396: PetscSegBuffer segrecvblocks;
397: MPI_Datatype blocktype;
398: size_t blocktype_size;
399: InsertMode *insertmode; /* Pointer to check mat->insertmode and set upon message arrival in case no local values have been set. */
400: };
402: #if !defined(PETSC_HAVE_MPIUNI)
403: PETSC_INTERN PetscErrorCode MatStashScatterDestroy_BTS(MatStash *);
404: #endif
405: PETSC_INTERN PetscErrorCode MatStashCreate_Private(MPI_Comm, PetscInt, MatStash *);
406: PETSC_INTERN PetscErrorCode MatStashDestroy_Private(MatStash *);
407: PETSC_INTERN PetscErrorCode MatStashScatterEnd_Private(MatStash *);
408: PETSC_INTERN PetscErrorCode MatStashSetInitialSize_Private(MatStash *, PetscInt);
409: PETSC_INTERN PetscErrorCode MatStashGetInfo_Private(MatStash *, PetscInt *, PetscInt *);
410: PETSC_INTERN PetscErrorCode MatStashValuesRow_Private(MatStash *, PetscInt, PetscInt, const PetscInt[], const PetscScalar[], PetscBool);
411: PETSC_INTERN PetscErrorCode MatStashValuesCol_Private(MatStash *, PetscInt, PetscInt, const PetscInt[], const PetscScalar[], PetscInt, PetscBool);
412: PETSC_INTERN PetscErrorCode MatStashValuesRowBlocked_Private(MatStash *, PetscInt, PetscInt, const PetscInt[], const PetscScalar[], PetscInt, PetscInt, PetscInt);
413: PETSC_INTERN PetscErrorCode MatStashValuesColBlocked_Private(MatStash *, PetscInt, PetscInt, const PetscInt[], const PetscScalar[], PetscInt, PetscInt, PetscInt);
414: PETSC_INTERN PetscErrorCode MatStashScatterBegin_Private(Mat, MatStash *, PetscInt *);
415: PETSC_INTERN PetscErrorCode MatStashScatterGetMesg_Private(MatStash *, PetscMPIInt *, PetscInt **, PetscInt **, PetscScalar **, PetscInt *);
416: PETSC_INTERN PetscErrorCode MatGetInfo_External(Mat, MatInfoType, MatInfo *);
418: typedef struct {
419: PetscInt dim;
420: PetscInt dims[4];
421: PetscInt starts[4];
422: PetscBool noc; /* this is a single component problem, hence user will not set MatStencil.c */
423: } MatStencilInfo;
425: /* Info about using compressed row format */
426: typedef struct {
427: PetscBool use; /* indicates compressed rows have been checked and will be used */
428: PetscInt nrows; /* number of non-zero rows */
429: PetscInt *i; /* compressed row pointer */
430: PetscInt *rindex; /* compressed row index */
431: } Mat_CompressedRow;
432: PETSC_EXTERN PetscErrorCode MatCheckCompressedRow(Mat, PetscInt, Mat_CompressedRow *, PetscInt *, PetscInt, PetscReal);
434: typedef struct { /* used by MatCreateRedundantMatrix() for reusing matredundant */
435: PetscInt nzlocal, nsends, nrecvs;
436: PetscMPIInt *send_rank, *recv_rank;
437: PetscInt *sbuf_nz, *rbuf_nz, *sbuf_j, **rbuf_j;
438: PetscScalar *sbuf_a, **rbuf_a;
439: MPI_Comm subcomm; /* when user does not provide a subcomm */
440: IS isrow, iscol;
441: Mat *matseq;
442: } Mat_Redundant;
444: typedef struct { /* used by MatProduct() */
445: MatProductType type;
446: char *alg;
447: Mat A, B, C, Dwork;
448: PetscBool symbolic_used_the_fact_A_is_symmetric; /* Symbolic phase took advantage of the fact that A is symmetric, and optimized e.g. AtB as AB. Then, .. */
449: PetscBool symbolic_used_the_fact_B_is_symmetric; /* .. in the numeric phase, if a new A is not symmetric (but has the same sparsity as the old A therefore .. */
450: PetscBool symbolic_used_the_fact_C_is_symmetric; /* MatMatMult(A,B,MAT_REUSE_MATRIX,..&C) is still legitimate), we need to redo symbolic! */
451: PetscReal fill;
452: PetscBool api_user; /* used to distinguish command line options and to indicate the matrix values are ready to be consumed at symbolic phase if needed */
453: PetscBool setfromoptionscalled;
455: /* Some products may display the information on the algorithm used */
456: PetscErrorCode (*view)(Mat, PetscViewer);
458: /* many products have intermediate data structures, each specific to Mat types and product type */
459: PetscBool clear; /* whether or not to clear the data structures after MatProductNumeric has been called */
460: void *data; /* where to stash those structures */
461: PetscErrorCode (*destroy)(void *); /* destroy routine */
462: } Mat_Product;
464: struct _p_Mat {
465: PETSCHEADER(struct _MatOps);
466: PetscLayout rmap, cmap;
467: void *data; /* implementation-specific data */
468: MatFactorType factortype; /* MAT_FACTOR_LU, ILU, CHOLESKY or ICC */
469: PetscBool trivialsymbolic; /* indicates the symbolic factorization doesn't actually do a symbolic factorization, it is delayed to the numeric factorization */
470: PetscBool canuseordering; /* factorization can use ordering provide to routine (most PETSc implementations) */
471: MatOrderingType preferredordering[MAT_FACTOR_NUM_TYPES]; /* what is the preferred (or default) ordering for the matrix solver type */
472: PetscBool assembled; /* is the matrix assembled? */
473: PetscBool was_assembled; /* new values inserted into assembled mat */
474: PetscInt num_ass; /* number of times matrix has been assembled */
475: PetscObjectState nonzerostate; /* each time new nonzeros locations are introduced into the matrix this is updated */
476: PetscObjectState ass_nonzerostate; /* nonzero state at last assembly */
477: MatInfo info; /* matrix information */
478: InsertMode insertmode; /* have values been inserted in matrix or added? */
479: MatStash stash, bstash; /* used for assembling off-proc mat emements */
480: MatNullSpace nullsp; /* null space (operator is singular) */
481: MatNullSpace transnullsp; /* null space of transpose of operator */
482: MatNullSpace nearnullsp; /* near null space to be used by multigrid methods */
483: PetscInt congruentlayouts; /* are the rows and columns layouts congruent? */
484: PetscBool preallocated;
485: MatStencilInfo stencil; /* information for structured grid */
486: PetscBool3 symmetric, hermitian, structurally_symmetric, spd;
487: PetscBool symmetry_eternal, structural_symmetry_eternal, spd_eternal;
488: PetscBool nooffprocentries, nooffproczerorows;
489: PetscBool assembly_subset; /* set by MAT_SUBSET_OFF_PROC_ENTRIES */
490: PetscBool submat_singleis; /* for efficient PCSetUp_ASM() */
491: PetscBool structure_only;
492: PetscBool sortedfull; /* full, sorted rows are inserted */
493: PetscBool force_diagonals; /* set by MAT_FORCE_DIAGONAL_ENTRIES */
494: #if defined(PETSC_HAVE_DEVICE)
495: PetscOffloadMask offloadmask; /* a mask which indicates where the valid matrix data is (GPU, CPU or both) */
496: PetscBool boundtocpu;
497: PetscBool bindingpropagates;
498: #endif
499: char *defaultrandtype;
500: void *spptr; /* pointer for special library like SuperLU */
501: char *solvertype;
502: PetscBool checksymmetryonassembly, checknullspaceonassembly;
503: PetscReal checksymmetrytol;
504: Mat schur; /* Schur complement matrix */
505: MatFactorSchurStatus schur_status; /* status of the Schur complement matrix */
506: Mat_Redundant *redundant; /* used by MatCreateRedundantMatrix() */
507: PetscBool erroriffailure; /* Generate an error if detected (for example a zero pivot) instead of returning */
508: MatFactorError factorerrortype; /* type of error in factorization */
509: PetscReal factorerror_zeropivot_value; /* If numerical zero pivot was detected this is the computed value */
510: PetscInt factorerror_zeropivot_row; /* Row where zero pivot was detected */
511: PetscInt nblocks, *bsizes; /* support for MatSetVariableBlockSizes() */
512: PetscInt p_cstart, p_rank, p_cend, n_rank; /* Information from parallel MatComputeVariableBlockEnvelope() */
513: PetscBool p_parallel;
514: char *defaultvectype;
515: Mat_Product *product;
516: PetscBool form_explicit_transpose; /* hint to generate an explicit mat tranpsose for operations like MatMultTranspose() */
517: PetscBool transupdated; /* whether or not the explicitly generated transpose is up-to-date */
518: char *factorprefix; /* the prefix to use with factored matrix that is created */
519: PetscBool hash_active; /* indicates MatSetValues() is being handled by hashing */
520: };
522: PETSC_INTERN PetscErrorCode MatAXPY_Basic(Mat, PetscScalar, Mat, MatStructure);
523: PETSC_INTERN PetscErrorCode MatAXPY_BasicWithPreallocation(Mat, Mat, PetscScalar, Mat, MatStructure);
524: PETSC_INTERN PetscErrorCode MatAXPY_Basic_Preallocate(Mat, Mat, Mat *);
525: PETSC_INTERN PetscErrorCode MatAXPY_Dense_Nest(Mat, PetscScalar, Mat);
527: PETSC_INTERN PetscErrorCode MatSetUp_Default(Mat);
529: /*
530: Utility for MatZeroRows
531: */
532: PETSC_INTERN PetscErrorCode MatZeroRowsMapLocal_Private(Mat, PetscInt, const PetscInt *, PetscInt *, PetscInt **);
534: /*
535: Utility for MatView/MatLoad
536: */
537: PETSC_INTERN PetscErrorCode MatView_Binary_BlockSizes(Mat, PetscViewer);
538: PETSC_INTERN PetscErrorCode MatLoad_Binary_BlockSizes(Mat, PetscViewer);
540: /*
541: Object for partitioning graphs
542: */
544: typedef struct _MatPartitioningOps *MatPartitioningOps;
545: struct _MatPartitioningOps {
546: PetscErrorCode (*apply)(MatPartitioning, IS *);
547: PetscErrorCode (*applynd)(MatPartitioning, IS *);
548: PetscErrorCode (*setfromoptions)(MatPartitioning, PetscOptionItems *);
549: PetscErrorCode (*destroy)(MatPartitioning);
550: PetscErrorCode (*view)(MatPartitioning, PetscViewer);
551: PetscErrorCode (*improve)(MatPartitioning, IS *);
552: };
554: struct _p_MatPartitioning {
555: PETSCHEADER(struct _MatPartitioningOps);
556: Mat adj;
557: PetscInt *vertex_weights;
558: PetscReal *part_weights;
559: PetscInt n; /* number of partitions */
560: PetscInt ncon; /* number of vertex weights per vertex */
561: void *data;
562: PetscInt setupcalled;
563: PetscBool use_edge_weights; /* A flag indicates whether or not to use edge weights */
564: };
566: /* needed for parallel nested dissection by ParMetis and PTSCOTCH */
567: PETSC_INTERN PetscErrorCode MatPartitioningSizesToSep_Private(PetscInt, PetscInt[], PetscInt[], PetscInt[]);
569: /*
570: Object for coarsen graphs
571: */
572: typedef struct _MatCoarsenOps *MatCoarsenOps;
573: struct _MatCoarsenOps {
574: PetscErrorCode (*apply)(MatCoarsen);
575: PetscErrorCode (*setfromoptions)(MatCoarsen, PetscOptionItems *);
576: PetscErrorCode (*destroy)(MatCoarsen);
577: PetscErrorCode (*view)(MatCoarsen, PetscViewer);
578: };
580: #define MAT_COARSEN_STRENGTH_INDEX_SIZE 3
581: struct _p_MatCoarsen {
582: PETSCHEADER(struct _MatCoarsenOps);
583: Mat graph;
584: void *subctx;
585: /* */
586: PetscBool strict_aggs;
587: IS perm;
588: PetscCoarsenData *agg_lists;
589: PetscInt max_it; /* number of iterations in HEM */
590: PetscReal threshold; /* HEM can filter interim graphs */
591: PetscInt strength_index_size;
592: PetscInt strength_index[MAT_COARSEN_STRENGTH_INDEX_SIZE];
593: };
595: PETSC_EXTERN PetscErrorCode MatCoarsenMISKSetDistance(MatCoarsen, PetscInt);
596: PETSC_EXTERN PetscErrorCode MatCoarsenMISKGetDistance(MatCoarsen, PetscInt *);
598: /*
599: Used in aijdevice.h
600: */
601: typedef struct {
602: PetscInt *i;
603: PetscInt *j;
604: PetscScalar *a;
605: PetscInt n;
606: PetscInt ignorezeroentries;
607: } PetscCSRDataStructure;
609: /*
610: MatFDColoring is used to compute Jacobian matrices efficiently
611: via coloring. The data structure is explained below in an example.
613: Color = 0 1 0 2 | 2 3 0
614: ---------------------------------------------------
615: 00 01 | 05
616: 10 11 | 14 15 Processor 0
617: 22 23 | 25
618: 32 33 |
619: ===================================================
620: | 44 45 46
621: 50 | 55 Processor 1
622: | 64 66
623: ---------------------------------------------------
625: ncolors = 4;
627: ncolumns = {2,1,1,0}
628: columns = {{0,2},{1},{3},{}}
629: nrows = {4,2,3,3}
630: rows = {{0,1,2,3},{0,1},{1,2,3},{0,1,2}}
631: vwscale = {dx(0),dx(1),dx(2),dx(3)} MPI Vec
632: vscale = {dx(0),dx(1),dx(2),dx(3),dx(4),dx(5)} Seq Vec
634: ncolumns = {1,0,1,1}
635: columns = {{6},{},{4},{5}}
636: nrows = {3,0,2,2}
637: rows = {{0,1,2},{},{1,2},{1,2}}
638: vwscale = {dx(4),dx(5),dx(6)} MPI Vec
639: vscale = {dx(0),dx(4),dx(5),dx(6)} Seq Vec
641: See the routine MatFDColoringApply() for how this data is used
642: to compute the Jacobian.
644: */
645: typedef struct {
646: PetscInt row;
647: PetscInt col;
648: PetscScalar *valaddr; /* address of value */
649: } MatEntry;
651: typedef struct {
652: PetscInt row;
653: PetscScalar *valaddr; /* address of value */
654: } MatEntry2;
656: struct _p_MatFDColoring {
657: PETSCHEADER(int);
658: PetscInt M, N, m; /* total rows, columns; local rows */
659: PetscInt rstart; /* first row owned by local processor */
660: PetscInt ncolors; /* number of colors */
661: PetscInt *ncolumns; /* number of local columns for a color */
662: PetscInt **columns; /* lists the local columns of each color (using global column numbering) */
663: IS *isa; /* these are the IS that contain the column values given in columns */
664: PetscInt *nrows; /* number of local rows for each color */
665: MatEntry *matentry; /* holds (row, column, address of value) for Jacobian matrix entry */
666: MatEntry2 *matentry2; /* holds (row, address of value) for Jacobian matrix entry */
667: PetscScalar *dy; /* store a block of F(x+dx)-F(x) when J is in BAIJ format */
668: PetscReal error_rel; /* square root of relative error in computing function */
669: PetscReal umin; /* minimum allowable u'dx value */
670: Vec w1, w2, w3; /* work vectors used in computing Jacobian */
671: PetscBool fset; /* indicates that the initial function value F(X) is set */
672: PetscErrorCode (*f)(void); /* function that defines Jacobian */
673: void *fctx; /* optional user-defined context for use by the function f */
674: Vec vscale; /* holds FD scaling, i.e. 1/dx for each perturbed column */
675: PetscInt currentcolor; /* color for which function evaluation is being done now */
676: const char *htype; /* "wp" or "ds" */
677: ISColoringType ctype; /* IS_COLORING_GLOBAL or IS_COLORING_LOCAL */
678: PetscInt brows, bcols; /* number of block rows or columns for speedup inserting the dense matrix into sparse Jacobian */
679: PetscBool setupcalled; /* true if setup has been called */
680: PetscBool viewed; /* true if the -mat_fd_coloring_view has been triggered already */
681: void (*ftn_func_pointer)(void), *ftn_func_cntx; /* serve the same purpose as *fortran_func_pointers in PETSc objects */
682: PetscObjectId matid; /* matrix this object was created with, must always be the same */
683: };
685: typedef struct _MatColoringOps *MatColoringOps;
686: struct _MatColoringOps {
687: PetscErrorCode (*destroy)(MatColoring);
688: PetscErrorCode (*setfromoptions)(MatColoring, PetscOptionItems *);
689: PetscErrorCode (*view)(MatColoring, PetscViewer);
690: PetscErrorCode (*apply)(MatColoring, ISColoring *);
691: PetscErrorCode (*weights)(MatColoring, PetscReal **, PetscInt **);
692: };
694: struct _p_MatColoring {
695: PETSCHEADER(struct _MatColoringOps);
696: Mat mat;
697: PetscInt dist; /* distance of the coloring */
698: PetscInt maxcolors; /* the maximum number of colors returned, maxcolors=1 for MIS */
699: void *data; /* inner context */
700: PetscBool valid; /* check to see if what is produced is a valid coloring */
701: MatColoringWeightType weight_type; /* type of weight computation to be performed */
702: PetscReal *user_weights; /* custom weights and permutation */
703: PetscInt *user_lperm;
704: PetscBool valid_iscoloring; /* check to see if matcoloring is produced a valid iscoloring */
705: };
707: struct _p_MatTransposeColoring {
708: PETSCHEADER(int);
709: PetscInt M, N, m; /* total rows, columns; local rows */
710: PetscInt rstart; /* first row owned by local processor */
711: PetscInt ncolors; /* number of colors */
712: PetscInt *ncolumns; /* number of local columns for a color */
713: PetscInt *nrows; /* number of local rows for each color */
714: PetscInt currentcolor; /* color for which function evaluation is being done now */
715: ISColoringType ctype; /* IS_COLORING_GLOBAL or IS_COLORING_LOCAL */
717: PetscInt *colorforrow, *colorforcol; /* pointer to rows and columns */
718: PetscInt *rows; /* lists the local rows for each color (using the local row numbering) */
719: PetscInt *den2sp; /* maps (row,color) in the dense matrix to index of sparse matrix array a->a */
720: PetscInt *columns; /* lists the local columns of each color (using global column numbering) */
721: PetscInt brows; /* number of rows for efficient implementation of MatTransColoringApplyDenToSp() */
722: PetscInt *lstart; /* array used for loop over row blocks of Csparse */
723: };
725: /*
726: Null space context for preconditioner/operators
727: */
728: struct _p_MatNullSpace {
729: PETSCHEADER(int);
730: PetscBool has_cnst;
731: PetscInt n;
732: Vec *vecs;
733: PetscScalar *alpha; /* for projections */
734: PetscErrorCode (*remove)(MatNullSpace, Vec, void *); /* for user provided removal function */
735: void *rmctx; /* context for remove() function */
736: };
738: /*
739: Checking zero pivot for LU, ILU preconditioners.
740: */
741: typedef struct {
742: PetscInt nshift, nshift_max;
743: PetscReal shift_amount, shift_lo, shift_hi, shift_top, shift_fraction;
744: PetscBool newshift;
745: PetscReal rs; /* active row sum of abs(off-diagonals) */
746: PetscScalar pv; /* pivot of the active row */
747: } FactorShiftCtx;
749: PETSC_EXTERN PetscErrorCode MatTransposeCheckNonzeroState_Private(Mat, Mat);
751: /*
752: Used by MatTranspose() and potentially other functions to track the matrix used in the generation of another matrix
753: */
754: typedef struct {
755: PetscObjectId id;
756: PetscObjectState state;
757: PetscObjectState nonzerostate;
758: } MatParentState;
760: PETSC_EXTERN PetscErrorCode MatFactorDumpMatrix(Mat);
761: PETSC_INTERN PetscErrorCode MatSetBlockSizes_Default(Mat, PetscInt, PetscInt);
763: PETSC_SINGLE_LIBRARY_INTERN PetscErrorCode MatShift_Basic(Mat, PetscScalar);
765: static inline PetscErrorCode MatPivotCheck_nz(PETSC_UNUSED Mat mat, const MatFactorInfo *info, FactorShiftCtx *sctx, PETSC_UNUSED PetscInt row)
766: {
767: PetscReal _rs = sctx->rs;
768: PetscReal _zero = info->zeropivot * _rs;
770: PetscFunctionBegin;
771: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)) {
772: /* force |diag| > zeropivot*rs */
773: if (!sctx->nshift) sctx->shift_amount = info->shiftamount;
774: else sctx->shift_amount *= 2.0;
775: sctx->newshift = PETSC_TRUE;
776: (sctx->nshift)++;
777: } else {
778: sctx->newshift = PETSC_FALSE;
779: }
780: PetscFunctionReturn(PETSC_SUCCESS);
781: }
783: static inline PetscErrorCode MatPivotCheck_pd(PETSC_UNUSED Mat mat, const MatFactorInfo *info, FactorShiftCtx *sctx, PETSC_UNUSED PetscInt row)
784: {
785: PetscReal _rs = sctx->rs;
786: PetscReal _zero = info->zeropivot * _rs;
788: PetscFunctionBegin;
789: if (PetscRealPart(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)) {
790: /* force matfactor to be diagonally dominant */
791: if (sctx->nshift == sctx->nshift_max) {
792: sctx->shift_fraction = sctx->shift_hi;
793: } else {
794: sctx->shift_lo = sctx->shift_fraction;
795: sctx->shift_fraction = (sctx->shift_hi + sctx->shift_lo) / (PetscReal)2.;
796: }
797: sctx->shift_amount = sctx->shift_fraction * sctx->shift_top;
798: sctx->nshift++;
799: sctx->newshift = PETSC_TRUE;
800: } else {
801: sctx->newshift = PETSC_FALSE;
802: }
803: PetscFunctionReturn(PETSC_SUCCESS);
804: }
806: static inline PetscErrorCode MatPivotCheck_inblocks(PETSC_UNUSED Mat mat, const MatFactorInfo *info, FactorShiftCtx *sctx, PETSC_UNUSED PetscInt row)
807: {
808: PetscReal _zero = info->zeropivot;
810: PetscFunctionBegin;
811: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)) {
812: sctx->pv += info->shiftamount;
813: sctx->shift_amount = 0.0;
814: sctx->nshift++;
815: }
816: sctx->newshift = PETSC_FALSE;
817: PetscFunctionReturn(PETSC_SUCCESS);
818: }
820: static inline PetscErrorCode MatPivotCheck_none(Mat fact, Mat mat, const MatFactorInfo *info, FactorShiftCtx *sctx, PetscInt row)
821: {
822: PetscReal _zero = info->zeropivot;
824: PetscFunctionBegin;
825: sctx->newshift = PETSC_FALSE;
826: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)) {
827: PetscCheck(!mat->erroriffailure, PETSC_COMM_SELF, PETSC_ERR_MAT_LU_ZRPVT, "Zero pivot row %" PetscInt_FMT " value %g tolerance %g", row, (double)PetscAbsScalar(sctx->pv), (double)_zero);
828: PetscCall(PetscInfo(mat, "Detected zero pivot in factorization in row %" PetscInt_FMT " value %g tolerance %g\n", row, (double)PetscAbsScalar(sctx->pv), (double)_zero));
829: fact->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
830: fact->factorerror_zeropivot_value = PetscAbsScalar(sctx->pv);
831: fact->factorerror_zeropivot_row = row;
832: }
833: PetscFunctionReturn(PETSC_SUCCESS);
834: }
836: static inline PetscErrorCode MatPivotCheck(Mat fact, Mat mat, const MatFactorInfo *info, FactorShiftCtx *sctx, PetscInt row)
837: {
838: PetscFunctionBegin;
839: if (info->shifttype == (PetscReal)MAT_SHIFT_NONZERO) PetscCall(MatPivotCheck_nz(mat, info, sctx, row));
840: else if (info->shifttype == (PetscReal)MAT_SHIFT_POSITIVE_DEFINITE) PetscCall(MatPivotCheck_pd(mat, info, sctx, row));
841: else if (info->shifttype == (PetscReal)MAT_SHIFT_INBLOCKS) PetscCall(MatPivotCheck_inblocks(mat, info, sctx, row));
842: else PetscCall(MatPivotCheck_none(fact, mat, info, sctx, row));
843: PetscFunctionReturn(PETSC_SUCCESS);
844: }
846: #include <petscbt.h>
847: /*
848: Create and initialize a linked list
849: Input Parameters:
850: idx_start - starting index of the list
851: lnk_max - max value of lnk indicating the end of the list
852: nlnk - max length of the list
853: Output Parameters:
854: lnk - list initialized
855: bt - PetscBT (bitarray) with all bits set to false
856: lnk_empty - flg indicating the list is empty
857: */
858: #define PetscLLCreate(idx_start, lnk_max, nlnk, lnk, bt) ((PetscErrorCode)(PetscMalloc1(nlnk, &lnk) || PetscBTCreate(nlnk, &(bt)) || (lnk[idx_start] = lnk_max, PETSC_SUCCESS)))
860: #define PetscLLCreate_new(idx_start, lnk_max, nlnk, lnk, bt, lnk_empty) ((PetscErrorCode)(PetscMalloc1(nlnk, &lnk) || PetscBTCreate(nlnk, &(bt)) || (lnk_empty = PETSC_TRUE, 0) || (lnk[idx_start] = lnk_max, PETSC_SUCCESS)))
862: static inline PetscErrorCode PetscLLInsertLocation_Private(PetscBool assume_sorted, PetscInt k, PetscInt idx_start, PetscInt entry, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnkdata, PetscInt *PETSC_RESTRICT lnk)
863: {
864: PetscInt location;
866: PetscFunctionBegin;
867: /* start from the beginning if entry < previous entry */
868: if (!assume_sorted && k && entry < *lnkdata) *lnkdata = idx_start;
869: /* search for insertion location */
870: do {
871: location = *lnkdata;
872: *lnkdata = lnk[location];
873: } while (entry > *lnkdata);
874: /* insertion location is found, add entry into lnk */
875: lnk[location] = entry;
876: lnk[entry] = *lnkdata;
877: ++(*nlnk);
878: *lnkdata = entry; /* next search starts from here if next_entry > entry */
879: PetscFunctionReturn(PETSC_SUCCESS);
880: }
882: static inline PetscErrorCode PetscLLAdd_Private(PetscInt nidx, const PetscInt *PETSC_RESTRICT indices, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscBT bt, PetscBool assume_sorted)
883: {
884: PetscFunctionBegin;
885: *nlnk = 0;
886: for (PetscInt k = 0, lnkdata = idx_start; k < nidx; ++k) {
887: const PetscInt entry = indices[k];
889: if (!PetscBTLookupSet(bt, entry)) PetscCall(PetscLLInsertLocation_Private(assume_sorted, k, idx_start, entry, nlnk, &lnkdata, lnk));
890: }
891: PetscFunctionReturn(PETSC_SUCCESS);
892: }
894: /*
895: Add an index set into a sorted linked list
896: Input Parameters:
897: nidx - number of input indices
898: indices - integer array
899: idx_start - starting index of the list
900: lnk - linked list(an integer array) that is created
901: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
902: output Parameters:
903: nlnk - number of newly added indices
904: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
905: bt - updated PetscBT (bitarray)
906: */
907: static inline PetscErrorCode PetscLLAdd(PetscInt nidx, const PetscInt *PETSC_RESTRICT indices, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscBT bt)
908: {
909: PetscFunctionBegin;
910: PetscCall(PetscLLAdd_Private(nidx, indices, idx_start, nlnk, lnk, bt, PETSC_FALSE));
911: PetscFunctionReturn(PETSC_SUCCESS);
912: }
914: /*
915: Add a SORTED ascending index set into a sorted linked list - same as PetscLLAdd() bus skip 'if (_k && _entry < _lnkdata) _lnkdata = idx_start;'
916: Input Parameters:
917: nidx - number of input indices
918: indices - sorted integer array
919: idx_start - starting index of the list
920: lnk - linked list(an integer array) that is created
921: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
922: output Parameters:
923: nlnk - number of newly added indices
924: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
925: bt - updated PetscBT (bitarray)
926: */
927: static inline PetscErrorCode PetscLLAddSorted(PetscInt nidx, const PetscInt *PETSC_RESTRICT indices, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscBT bt)
928: {
929: PetscFunctionBegin;
930: PetscCall(PetscLLAdd_Private(nidx, indices, idx_start, nlnk, lnk, bt, PETSC_TRUE));
931: PetscFunctionReturn(PETSC_SUCCESS);
932: }
934: /*
935: Add a permuted index set into a sorted linked list
936: Input Parameters:
937: nidx - number of input indices
938: indices - integer array
939: perm - permutation of indices
940: idx_start - starting index of the list
941: lnk - linked list(an integer array) that is created
942: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
943: output Parameters:
944: nlnk - number of newly added indices
945: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
946: bt - updated PetscBT (bitarray)
947: */
948: static inline PetscErrorCode PetscLLAddPerm(PetscInt nidx, const PetscInt *PETSC_RESTRICT indices, const PetscInt *PETSC_RESTRICT perm, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscBT bt)
949: {
950: PetscFunctionBegin;
951: *nlnk = 0;
952: for (PetscInt k = 0, lnkdata = idx_start; k < nidx; ++k) {
953: const PetscInt entry = perm[indices[k]];
955: if (!PetscBTLookupSet(bt, entry)) PetscCall(PetscLLInsertLocation_Private(PETSC_FALSE, k, idx_start, entry, nlnk, &lnkdata, lnk));
956: }
957: PetscFunctionReturn(PETSC_SUCCESS);
958: }
960: #if 0
961: /* this appears to be unused? */
962: static inline PetscErrorCode PetscLLAddSorted_new(PetscInt nidx, PetscInt *indices, PetscInt idx_start, PetscBool *lnk_empty, PetscInt *nlnk, PetscInt *lnk, PetscBT bt)
963: {
964: PetscInt lnkdata = idx_start;
966: PetscFunctionBegin;
967: if (*lnk_empty) {
968: for (PetscInt k = 0; k < nidx; ++k) {
969: const PetscInt entry = indices[k], location = lnkdata;
971: PetscCall(PetscBTSet(bt,entry)); /* mark the new entry */
972: lnkdata = lnk[location];
973: /* insertion location is found, add entry into lnk */
974: lnk[location] = entry;
975: lnk[entry] = lnkdata;
976: lnkdata = entry; /* next search starts from here */
977: }
978: /* lnk[indices[nidx-1]] = lnk[idx_start];
979: lnk[idx_start] = indices[0];
980: PetscCall(PetscBTSet(bt,indices[0]));
981: for (_k=1; _k<nidx; _k++) {
982: PetscCall(PetscBTSet(bt,indices[_k]));
983: lnk[indices[_k-1]] = indices[_k];
984: }
985: */
986: *nlnk = nidx;
987: *lnk_empty = PETSC_FALSE;
988: } else {
989: *nlnk = 0;
990: for (PetscInt k = 0; k < nidx; ++k) {
991: const PetscInt entry = indices[k];
993: if (!PetscBTLookupSet(bt,entry)) PetscCall(PetscLLInsertLocation_Private(PETSC_TRUE,k,idx_start,entry,nlnk,&lnkdata,lnk));
994: }
995: }
996: PetscFunctionReturn(PETSC_SUCCESS);
997: }
998: #endif
1000: /*
1001: Add a SORTED index set into a sorted linked list used for LUFactorSymbolic()
1002: Same as PetscLLAddSorted() with an additional operation:
1003: count the number of input indices that are no larger than 'diag'
1004: Input Parameters:
1005: indices - sorted integer array
1006: idx_start - starting index of the list, index of pivot row
1007: lnk - linked list(an integer array) that is created
1008: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1009: diag - index of the active row in LUFactorSymbolic
1010: nzbd - number of input indices with indices <= idx_start
1011: im - im[idx_start] is initialized as num of nonzero entries in row=idx_start
1012: output Parameters:
1013: nlnk - number of newly added indices
1014: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
1015: bt - updated PetscBT (bitarray)
1016: im - im[idx_start]: unchanged if diag is not an entry
1017: : num of entries with indices <= diag if diag is an entry
1018: */
1019: static inline PetscErrorCode PetscLLAddSortedLU(const PetscInt *PETSC_RESTRICT indices, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscBT bt, PetscInt diag, PetscInt nzbd, PetscInt *PETSC_RESTRICT im)
1020: {
1021: const PetscInt nidx = im[idx_start] - nzbd; /* num of entries with idx_start < index <= diag */
1023: PetscFunctionBegin;
1024: *nlnk = 0;
1025: for (PetscInt k = 0, lnkdata = idx_start; k < nidx; ++k) {
1026: const PetscInt entry = indices[k];
1028: ++nzbd;
1029: if (entry == diag) im[idx_start] = nzbd;
1030: if (!PetscBTLookupSet(bt, entry)) PetscCall(PetscLLInsertLocation_Private(PETSC_TRUE, k, idx_start, entry, nlnk, &lnkdata, lnk));
1031: }
1032: PetscFunctionReturn(PETSC_SUCCESS);
1033: }
1035: /*
1036: Copy data on the list into an array, then initialize the list
1037: Input Parameters:
1038: idx_start - starting index of the list
1039: lnk_max - max value of lnk indicating the end of the list
1040: nlnk - number of data on the list to be copied
1041: lnk - linked list
1042: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1043: output Parameters:
1044: indices - array that contains the copied data
1045: lnk - linked list that is cleaned and initialize
1046: bt - PetscBT (bitarray) with all bits set to false
1047: */
1048: static inline PetscErrorCode PetscLLClean(PetscInt idx_start, PetscInt lnk_max, PetscInt nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT indices, PetscBT bt)
1049: {
1050: PetscFunctionBegin;
1051: for (PetscInt j = 0, idx = idx_start; j < nlnk; ++j) {
1052: idx = lnk[idx];
1053: indices[j] = idx;
1054: PetscCall(PetscBTClear(bt, idx));
1055: }
1056: lnk[idx_start] = lnk_max;
1057: PetscFunctionReturn(PETSC_SUCCESS);
1058: }
1060: /*
1061: Free memories used by the list
1062: */
1063: #define PetscLLDestroy(lnk, bt) ((PetscErrorCode)(PetscFree(lnk) || PetscBTDestroy(&(bt))))
1065: /* Routines below are used for incomplete matrix factorization */
1066: /*
1067: Create and initialize a linked list and its levels
1068: Input Parameters:
1069: idx_start - starting index of the list
1070: lnk_max - max value of lnk indicating the end of the list
1071: nlnk - max length of the list
1072: Output Parameters:
1073: lnk - list initialized
1074: lnk_lvl - array of size nlnk for storing levels of lnk
1075: bt - PetscBT (bitarray) with all bits set to false
1076: */
1077: #define PetscIncompleteLLCreate(idx_start, lnk_max, nlnk, lnk, lnk_lvl, bt) \
1078: ((PetscErrorCode)(PetscIntMultError(2, nlnk, NULL) || PetscMalloc1(2 * nlnk, &lnk) || PetscBTCreate(nlnk, &(bt)) || (lnk[idx_start] = lnk_max, lnk_lvl = lnk + nlnk, PETSC_SUCCESS)))
1080: static inline PetscErrorCode PetscIncompleteLLInsertLocation_Private(PetscBool assume_sorted, PetscInt k, PetscInt idx_start, PetscInt entry, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnkdata, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscInt newval)
1081: {
1082: PetscFunctionBegin;
1083: PetscCall(PetscLLInsertLocation_Private(assume_sorted, k, idx_start, entry, nlnk, lnkdata, lnk));
1084: lnklvl[entry] = newval;
1085: PetscFunctionReturn(PETSC_SUCCESS);
1086: }
1088: /*
1089: Initialize a sorted linked list used for ILU and ICC
1090: Input Parameters:
1091: nidx - number of input idx
1092: idx - integer array used for storing column indices
1093: idx_start - starting index of the list
1094: perm - indices of an IS
1095: lnk - linked list(an integer array) that is created
1096: lnklvl - levels of lnk
1097: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1098: output Parameters:
1099: nlnk - number of newly added idx
1100: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1101: lnklvl - levels of lnk
1102: bt - updated PetscBT (bitarray)
1103: */
1104: static inline PetscErrorCode PetscIncompleteLLInit(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscInt idx_start, const PetscInt *PETSC_RESTRICT perm, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt)
1105: {
1106: PetscFunctionBegin;
1107: *nlnk = 0;
1108: for (PetscInt k = 0, lnkdata = idx_start; k < nidx; ++k) {
1109: const PetscInt entry = perm[idx[k]];
1111: if (!PetscBTLookupSet(bt, entry)) PetscCall(PetscIncompleteLLInsertLocation_Private(PETSC_FALSE, k, idx_start, entry, nlnk, &lnkdata, lnk, lnklvl, 0));
1112: }
1113: PetscFunctionReturn(PETSC_SUCCESS);
1114: }
1116: static inline PetscErrorCode PetscIncompleteLLAdd_Private(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscReal level, const PetscInt *PETSC_RESTRICT idxlvl, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt, PetscInt prow_offset, PetscBool assume_sorted)
1117: {
1118: PetscFunctionBegin;
1119: *nlnk = 0;
1120: for (PetscInt k = 0, lnkdata = idx_start; k < nidx; ++k) {
1121: const PetscInt incrlev = idxlvl[k] + prow_offset + 1;
1123: if (incrlev <= level) {
1124: const PetscInt entry = idx[k];
1126: if (!PetscBTLookupSet(bt, entry)) PetscCall(PetscIncompleteLLInsertLocation_Private(assume_sorted, k, idx_start, entry, nlnk, &lnkdata, lnk, lnklvl, incrlev));
1127: else if (lnklvl[entry] > incrlev) lnklvl[entry] = incrlev; /* existing entry */
1128: }
1129: }
1130: PetscFunctionReturn(PETSC_SUCCESS);
1131: }
1133: /*
1134: Add a SORTED index set into a sorted linked list for ICC
1135: Input Parameters:
1136: nidx - number of input indices
1137: idx - sorted integer array used for storing column indices
1138: level - level of fill, e.g., ICC(level)
1139: idxlvl - level of idx
1140: idx_start - starting index of the list
1141: lnk - linked list(an integer array) that is created
1142: lnklvl - levels of lnk
1143: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1144: idxlvl_prow - idxlvl[prow], where prow is the row number of the idx
1145: output Parameters:
1146: nlnk - number of newly added indices
1147: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1148: lnklvl - levels of lnk
1149: bt - updated PetscBT (bitarray)
1150: Note: the level of U(i,j) is set as lvl(i,j) = min{ lvl(i,j), lvl(prow,i)+lvl(prow,j)+1)
1151: where idx = non-zero columns of U(prow,prow+1:n-1), prow<i
1152: */
1153: static inline PetscErrorCode PetscICCLLAddSorted(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscReal level, const PetscInt *PETSC_RESTRICT idxlvl, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt, PetscInt idxlvl_prow)
1154: {
1155: PetscFunctionBegin;
1156: PetscCall(PetscIncompleteLLAdd_Private(nidx, idx, level, idxlvl, idx_start, nlnk, lnk, lnklvl, bt, idxlvl_prow, PETSC_TRUE));
1157: PetscFunctionReturn(PETSC_SUCCESS);
1158: }
1160: /*
1161: Add a SORTED index set into a sorted linked list for ILU
1162: Input Parameters:
1163: nidx - number of input indices
1164: idx - sorted integer array used for storing column indices
1165: level - level of fill, e.g., ICC(level)
1166: idxlvl - level of idx
1167: idx_start - starting index of the list
1168: lnk - linked list(an integer array) that is created
1169: lnklvl - levels of lnk
1170: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1171: prow - the row number of idx
1172: output Parameters:
1173: nlnk - number of newly added idx
1174: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1175: lnklvl - levels of lnk
1176: bt - updated PetscBT (bitarray)
1178: Note: the level of factor(i,j) is set as lvl(i,j) = min{ lvl(i,j), lvl(i,prow)+lvl(prow,j)+1)
1179: where idx = non-zero columns of U(prow,prow+1:n-1), prow<i
1180: */
1181: static inline PetscErrorCode PetscILULLAddSorted(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscInt level, const PetscInt *PETSC_RESTRICT idxlvl, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt, PetscInt prow)
1182: {
1183: PetscFunctionBegin;
1184: PetscCall(PetscIncompleteLLAdd_Private(nidx, idx, level, idxlvl, idx_start, nlnk, lnk, lnklvl, bt, lnklvl[prow], PETSC_TRUE));
1185: PetscFunctionReturn(PETSC_SUCCESS);
1186: }
1188: /*
1189: Add a index set into a sorted linked list
1190: Input Parameters:
1191: nidx - number of input idx
1192: idx - integer array used for storing column indices
1193: level - level of fill, e.g., ICC(level)
1194: idxlvl - level of idx
1195: idx_start - starting index of the list
1196: lnk - linked list(an integer array) that is created
1197: lnklvl - levels of lnk
1198: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1199: output Parameters:
1200: nlnk - number of newly added idx
1201: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1202: lnklvl - levels of lnk
1203: bt - updated PetscBT (bitarray)
1204: */
1205: static inline PetscErrorCode PetscIncompleteLLAdd(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscReal level, const PetscInt *PETSC_RESTRICT idxlvl, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt)
1206: {
1207: PetscFunctionBegin;
1208: PetscCall(PetscIncompleteLLAdd_Private(nidx, idx, level, idxlvl, idx_start, nlnk, lnk, lnklvl, bt, 0, PETSC_FALSE));
1209: PetscFunctionReturn(PETSC_SUCCESS);
1210: }
1212: /*
1213: Add a SORTED index set into a sorted linked list
1214: Input Parameters:
1215: nidx - number of input indices
1216: idx - sorted integer array used for storing column indices
1217: level - level of fill, e.g., ICC(level)
1218: idxlvl - level of idx
1219: idx_start - starting index of the list
1220: lnk - linked list(an integer array) that is created
1221: lnklvl - levels of lnk
1222: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1223: output Parameters:
1224: nlnk - number of newly added idx
1225: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1226: lnklvl - levels of lnk
1227: bt - updated PetscBT (bitarray)
1228: */
1229: static inline PetscErrorCode PetscIncompleteLLAddSorted(PetscInt nidx, const PetscInt *PETSC_RESTRICT idx, PetscReal level, const PetscInt *PETSC_RESTRICT idxlvl, PetscInt idx_start, PetscInt *PETSC_RESTRICT nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscBT bt)
1230: {
1231: PetscFunctionBegin;
1232: PetscCall(PetscIncompleteLLAdd_Private(nidx, idx, level, idxlvl, idx_start, nlnk, lnk, lnklvl, bt, 0, PETSC_TRUE));
1233: PetscFunctionReturn(PETSC_SUCCESS);
1234: }
1236: /*
1237: Copy data on the list into an array, then initialize the list
1238: Input Parameters:
1239: idx_start - starting index of the list
1240: lnk_max - max value of lnk indicating the end of the list
1241: nlnk - number of data on the list to be copied
1242: lnk - linked list
1243: lnklvl - level of lnk
1244: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1245: output Parameters:
1246: indices - array that contains the copied data
1247: lnk - linked list that is cleaned and initialize
1248: lnklvl - level of lnk that is reinitialized
1249: bt - PetscBT (bitarray) with all bits set to false
1250: */
1251: static inline PetscErrorCode PetscIncompleteLLClean(PetscInt idx_start, PetscInt lnk_max, PetscInt nlnk, PetscInt *PETSC_RESTRICT lnk, PetscInt *PETSC_RESTRICT lnklvl, PetscInt *PETSC_RESTRICT indices, PetscInt *PETSC_RESTRICT indiceslvl, PetscBT bt)
1252: {
1253: PetscFunctionBegin;
1254: for (PetscInt j = 0, idx = idx_start; j < nlnk; ++j) {
1255: idx = lnk[idx];
1256: indices[j] = idx;
1257: indiceslvl[j] = lnklvl[idx];
1258: lnklvl[idx] = -1;
1259: PetscCall(PetscBTClear(bt, idx));
1260: }
1261: lnk[idx_start] = lnk_max;
1262: PetscFunctionReturn(PETSC_SUCCESS);
1263: }
1265: /*
1266: Free memories used by the list
1267: */
1268: #define PetscIncompleteLLDestroy(lnk, bt) ((PetscErrorCode)(PetscFree(lnk) || PetscBTDestroy(&(bt))))
1270: #if !defined(PETSC_CLANG_STATIC_ANALYZER)
1271: #define MatCheckSameLocalSize(A, ar1, B, ar2) \
1272: do { \
1273: PetscCheckSameComm(A, ar1, B, ar2); \
1274: PetscCheck(((A)->rmap->n == (B)->rmap->n) && ((A)->cmap->n == (B)->cmap->n), PETSC_COMM_SELF, PETSC_ERR_ARG_INCOMP, "Incompatible matrix local sizes: parameter # %d (%" PetscInt_FMT " x %" PetscInt_FMT ") != parameter # %d (%" PetscInt_FMT " x %" PetscInt_FMT ")", ar1, \
1275: (A)->rmap->n, (A)->cmap->n, ar2, (B)->rmap->n, (B)->cmap->n); \
1276: } while (0)
1277: #define MatCheckSameSize(A, ar1, B, ar2) \
1278: do { \
1279: PetscCheck(((A)->rmap->N == (B)->rmap->N) && ((A)->cmap->N == (B)->cmap->N), PetscObjectComm((PetscObject)(A)), PETSC_ERR_ARG_INCOMP, "Incompatible matrix global sizes: parameter # %d (%" PetscInt_FMT " x %" PetscInt_FMT ") != parameter # %d (%" PetscInt_FMT " x %" PetscInt_FMT ")", ar1, \
1280: (A)->rmap->N, (A)->cmap->N, ar2, (B)->rmap->N, (B)->cmap->N); \
1281: MatCheckSameLocalSize(A, ar1, B, ar2); \
1282: } while (0)
1283: #else
1284: template <typename Tm>
1285: extern void MatCheckSameLocalSize(Tm, int, Tm, int);
1286: template <typename Tm>
1287: extern void MatCheckSameSize(Tm, int, Tm, int);
1288: #endif
1290: #define VecCheckMatCompatible(M, x, ar1, b, ar2) \
1291: do { \
1292: PetscCheck((M)->cmap->N == (x)->map->N, PetscObjectComm((PetscObject)(M)), PETSC_ERR_ARG_SIZ, "Vector global length incompatible with matrix: parameter # %d global size %" PetscInt_FMT " != matrix column global size %" PetscInt_FMT, ar1, (x)->map->N, \
1293: (M)->cmap->N); \
1294: PetscCheck((M)->rmap->N == (b)->map->N, PetscObjectComm((PetscObject)(M)), PETSC_ERR_ARG_SIZ, "Vector global length incompatible with matrix: parameter # %d global size %" PetscInt_FMT " != matrix row global size %" PetscInt_FMT, ar2, (b)->map->N, \
1295: (M)->rmap->N); \
1296: } while (0)
1298: /* -------------------------------------------------------------------------------------------------------*/
1299: /*
1300: Create and initialize a condensed linked list -
1301: same as PetscLLCreate(), but uses a scalable array 'lnk' with size of max number of entries, not O(N).
1302: Barry suggested this approach (Dec. 6, 2011):
1303: I've thought of an alternative way of representing a linked list that is efficient but doesn't have the O(N) scaling issue
1304: (it may be faster than the O(N) even sequentially due to less crazy memory access).
1306: Instead of having some like a 2 -> 4 -> 11 -> 22 list that uses slot 2 4 11 and 22 in a big array use a small array with two slots
1307: for each entry for example [ 2 1 | 4 3 | 22 -1 | 11 2] so the first number (of the pair) is the value while the second tells you where
1308: in the list the next entry is. Inserting a new link means just append another pair at the end. For example say we want to insert 13 into the
1309: list it would then become [2 1 | 4 3 | 22 -1 | 11 4 | 13 2 ] you just add a pair at the end and fix the point for the one that points to it.
1310: That is 11 use to point to the 2 slot, after the change 11 points to the 4th slot which has the value 13. Note that values are always next
1311: to each other so memory access is much better than using the big array.
1313: Example:
1314: nlnk_max=5, lnk_max=36:
1315: Initial list: [0, 0 | 36, 2 | 0, 0 | 0, 0 | 0, 0 | 0, 0 | 0, 0]
1316: here, head_node has index 2 with value lnk[2]=lnk_max=36,
1317: 0-th entry is used to store the number of entries in the list,
1318: The initial lnk represents head -> tail(marked by 36) with number of entries = lnk[0]=0.
1320: Now adding a sorted set {2,4}, the list becomes
1321: [2, 0 | 36, 4 |2, 6 | 4, 2 | 0, 0 | 0, 0 | 0, 0 ]
1322: represents head -> 2 -> 4 -> tail with number of entries = lnk[0]=2.
1324: Then adding a sorted set {0,3,35}, the list
1325: [5, 0 | 36, 8 | 2, 10 | 4, 12 | 0, 4 | 3, 6 | 35, 2 ]
1326: represents head -> 0 -> 2 -> 3 -> 4 -> 35 -> tail with number of entries = lnk[0]=5.
1328: Input Parameters:
1329: nlnk_max - max length of the list
1330: lnk_max - max value of the entries
1331: Output Parameters:
1332: lnk - list created and initialized
1333: bt - PetscBT (bitarray) with all bits set to false. Note: bt has size lnk_max, not nln_max!
1334: */
1335: static inline PetscErrorCode PetscLLCondensedCreate(PetscInt nlnk_max, PetscInt lnk_max, PetscInt **lnk, PetscBT *bt)
1336: {
1337: PetscInt *llnk, lsize = 0;
1339: PetscFunctionBegin;
1340: PetscCall(PetscIntMultError(2, nlnk_max + 2, &lsize));
1341: PetscCall(PetscMalloc1(lsize, lnk));
1342: PetscCall(PetscBTCreate(lnk_max, bt));
1343: llnk = *lnk;
1344: llnk[0] = 0; /* number of entries on the list */
1345: llnk[2] = lnk_max; /* value in the head node */
1346: llnk[3] = 2; /* next for the head node */
1347: PetscFunctionReturn(PETSC_SUCCESS);
1348: }
1350: /*
1351: Add a SORTED ascending index set into a sorted linked list. See PetscLLCondensedCreate() for detailed description.
1352: Input Parameters:
1353: nidx - number of input indices
1354: indices - sorted integer array
1355: lnk - condensed linked list(an integer array) that is created
1356: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1357: output Parameters:
1358: lnk - the sorted(increasing order) linked list containing previous and newly added non-redundate indices
1359: bt - updated PetscBT (bitarray)
1360: */
1361: static inline PetscErrorCode PetscLLCondensedAddSorted(PetscInt nidx, const PetscInt indices[], PetscInt lnk[], PetscBT bt)
1362: {
1363: PetscInt location = 2; /* head */
1364: PetscInt nlnk = lnk[0]; /* num of entries on the input lnk */
1366: PetscFunctionBegin;
1367: for (PetscInt k = 0; k < nidx; k++) {
1368: const PetscInt entry = indices[k];
1369: if (!PetscBTLookupSet(bt, entry)) { /* new entry */
1370: PetscInt next, lnkdata;
1372: /* search for insertion location */
1373: do {
1374: next = location + 1; /* link from previous node to next node */
1375: location = lnk[next]; /* idx of next node */
1376: lnkdata = lnk[location]; /* value of next node */
1377: } while (entry > lnkdata);
1378: /* insertion location is found, add entry into lnk */
1379: const PetscInt newnode = 2 * (nlnk + 2); /* index for this new node */
1380: lnk[next] = newnode; /* connect previous node to the new node */
1381: lnk[newnode] = entry; /* set value of the new node */
1382: lnk[newnode + 1] = location; /* connect new node to next node */
1383: location = newnode; /* next search starts from the new node */
1384: nlnk++;
1385: }
1386: }
1387: lnk[0] = nlnk; /* number of entries in the list */
1388: PetscFunctionReturn(PETSC_SUCCESS);
1389: }
1391: static inline PetscErrorCode PetscLLCondensedClean(PetscInt lnk_max, PETSC_UNUSED PetscInt nidx, PetscInt *indices, PetscInt lnk[], PetscBT bt)
1392: {
1393: const PetscInt nlnk = lnk[0]; /* num of entries on the list */
1394: PetscInt next = lnk[3]; /* head node */
1396: PetscFunctionBegin;
1397: for (PetscInt k = 0; k < nlnk; k++) {
1398: indices[k] = lnk[next];
1399: next = lnk[next + 1];
1400: PetscCall(PetscBTClear(bt, indices[k]));
1401: }
1402: lnk[0] = 0; /* num of entries on the list */
1403: lnk[2] = lnk_max; /* initialize head node */
1404: lnk[3] = 2; /* head node */
1405: PetscFunctionReturn(PETSC_SUCCESS);
1406: }
1408: static inline PetscErrorCode PetscLLCondensedView(PetscInt *lnk)
1409: {
1410: PetscFunctionBegin;
1411: PetscCall(PetscPrintf(PETSC_COMM_SELF, "LLCondensed of size %" PetscInt_FMT ", (val, next)\n", lnk[0]));
1412: for (PetscInt k = 2; k < lnk[0] + 2; ++k) PetscCall(PetscPrintf(PETSC_COMM_SELF, " %" PetscInt_FMT ": (%" PetscInt_FMT ", %" PetscInt_FMT ")\n", 2 * k, lnk[2 * k], lnk[2 * k + 1]));
1413: PetscFunctionReturn(PETSC_SUCCESS);
1414: }
1416: /*
1417: Free memories used by the list
1418: */
1419: static inline PetscErrorCode PetscLLCondensedDestroy(PetscInt *lnk, PetscBT bt)
1420: {
1421: PetscFunctionBegin;
1422: PetscCall(PetscFree(lnk));
1423: PetscCall(PetscBTDestroy(&bt));
1424: PetscFunctionReturn(PETSC_SUCCESS);
1425: }
1427: /* -------------------------------------------------------------------------------------------------------*/
1428: /*
1429: Same as PetscLLCondensedCreate(), but does not use non-scalable O(lnk_max) bitarray
1430: Input Parameters:
1431: nlnk_max - max length of the list
1432: Output Parameters:
1433: lnk - list created and initialized
1434: */
1435: static inline PetscErrorCode PetscLLCondensedCreate_Scalable(PetscInt nlnk_max, PetscInt **lnk)
1436: {
1437: PetscInt *llnk, lsize = 0;
1439: PetscFunctionBegin;
1440: PetscCall(PetscIntMultError(2, nlnk_max + 2, &lsize));
1441: PetscCall(PetscMalloc1(lsize, lnk));
1442: llnk = *lnk;
1443: llnk[0] = 0; /* number of entries on the list */
1444: llnk[2] = PETSC_MAX_INT; /* value in the head node */
1445: llnk[3] = 2; /* next for the head node */
1446: PetscFunctionReturn(PETSC_SUCCESS);
1447: }
1449: static inline PetscErrorCode PetscLLCondensedExpand_Scalable(PetscInt nlnk_max, PetscInt **lnk)
1450: {
1451: PetscInt lsize = 0;
1453: PetscFunctionBegin;
1454: PetscCall(PetscIntMultError(2, nlnk_max + 2, &lsize));
1455: PetscCall(PetscRealloc(lsize * sizeof(PetscInt), lnk));
1456: PetscFunctionReturn(PETSC_SUCCESS);
1457: }
1459: static inline PetscErrorCode PetscLLCondensedAddSorted_Scalable(PetscInt nidx, const PetscInt indices[], PetscInt lnk[])
1460: {
1461: PetscInt location = 2; /* head */
1462: PetscInt nlnk = lnk[0]; /* num of entries on the input lnk */
1464: for (PetscInt k = 0; k < nidx; k++) {
1465: const PetscInt entry = indices[k];
1466: PetscInt next, lnkdata;
1468: /* search for insertion location */
1469: do {
1470: next = location + 1; /* link from previous node to next node */
1471: location = lnk[next]; /* idx of next node */
1472: lnkdata = lnk[location]; /* value of next node */
1473: } while (entry > lnkdata);
1474: if (entry < lnkdata) {
1475: /* insertion location is found, add entry into lnk */
1476: const PetscInt newnode = 2 * (nlnk + 2); /* index for this new node */
1477: lnk[next] = newnode; /* connect previous node to the new node */
1478: lnk[newnode] = entry; /* set value of the new node */
1479: lnk[newnode + 1] = location; /* connect new node to next node */
1480: location = newnode; /* next search starts from the new node */
1481: nlnk++;
1482: }
1483: }
1484: lnk[0] = nlnk; /* number of entries in the list */
1485: return PETSC_SUCCESS;
1486: }
1488: static inline PetscErrorCode PetscLLCondensedClean_Scalable(PETSC_UNUSED PetscInt nidx, PetscInt *indices, PetscInt *lnk)
1489: {
1490: const PetscInt nlnk = lnk[0];
1491: PetscInt next = lnk[3]; /* head node */
1493: for (PetscInt k = 0; k < nlnk; k++) {
1494: indices[k] = lnk[next];
1495: next = lnk[next + 1];
1496: }
1497: lnk[0] = 0; /* num of entries on the list */
1498: lnk[3] = 2; /* head node */
1499: return PETSC_SUCCESS;
1500: }
1502: static inline PetscErrorCode PetscLLCondensedDestroy_Scalable(PetscInt *lnk)
1503: {
1504: return PetscFree(lnk);
1505: }
1507: /* -------------------------------------------------------------------------------------------------------*/
1508: /*
1509: lnk[0] number of links
1510: lnk[1] number of entries
1511: lnk[3n] value
1512: lnk[3n+1] len
1513: lnk[3n+2] link to next value
1515: The next three are always the first link
1517: lnk[3] PETSC_MIN_INT+1
1518: lnk[4] 1
1519: lnk[5] link to first real entry
1521: The next three are always the last link
1523: lnk[6] PETSC_MAX_INT - 1
1524: lnk[7] 1
1525: lnk[8] next valid link (this is the same as lnk[0] but without the decreases)
1526: */
1528: static inline PetscErrorCode PetscLLCondensedCreate_fast(PetscInt nlnk_max, PetscInt **lnk)
1529: {
1530: PetscInt *llnk;
1531: PetscInt lsize = 0;
1533: PetscFunctionBegin;
1534: PetscCall(PetscIntMultError(3, nlnk_max + 3, &lsize));
1535: PetscCall(PetscMalloc1(lsize, lnk));
1536: llnk = *lnk;
1537: llnk[0] = 0; /* nlnk: number of entries on the list */
1538: llnk[1] = 0; /* number of integer entries represented in list */
1539: llnk[3] = PETSC_MIN_INT + 1; /* value in the first node */
1540: llnk[4] = 1; /* count for the first node */
1541: llnk[5] = 6; /* next for the first node */
1542: llnk[6] = PETSC_MAX_INT - 1; /* value in the last node */
1543: llnk[7] = 1; /* count for the last node */
1544: llnk[8] = 0; /* next valid node to be used */
1545: PetscFunctionReturn(PETSC_SUCCESS);
1546: }
1548: static inline PetscErrorCode PetscLLCondensedAddSorted_fast(PetscInt nidx, const PetscInt indices[], PetscInt lnk[])
1549: {
1550: for (PetscInt k = 0, prev = 3 /* first value */; k < nidx; k++) {
1551: const PetscInt entry = indices[k];
1552: PetscInt next = lnk[prev + 2];
1554: /* search for insertion location */
1555: while (entry >= lnk[next]) {
1556: prev = next;
1557: next = lnk[next + 2];
1558: }
1559: /* entry is in range of previous list */
1560: if (entry < lnk[prev] + lnk[prev + 1]) continue;
1561: lnk[1]++;
1562: /* entry is right after previous list */
1563: if (entry == lnk[prev] + lnk[prev + 1]) {
1564: lnk[prev + 1]++;
1565: if (lnk[next] == entry + 1) { /* combine two contiguous strings */
1566: lnk[prev + 1] += lnk[next + 1];
1567: lnk[prev + 2] = lnk[next + 2];
1568: next = lnk[next + 2];
1569: lnk[0]--;
1570: }
1571: continue;
1572: }
1573: /* entry is right before next list */
1574: if (entry == lnk[next] - 1) {
1575: lnk[next]--;
1576: lnk[next + 1]++;
1577: prev = next;
1578: next = lnk[prev + 2];
1579: continue;
1580: }
1581: /* add entry into lnk */
1582: lnk[prev + 2] = 3 * ((lnk[8]++) + 3); /* connect previous node to the new node */
1583: prev = lnk[prev + 2];
1584: lnk[prev] = entry; /* set value of the new node */
1585: lnk[prev + 1] = 1; /* number of values in contiguous string is one to start */
1586: lnk[prev + 2] = next; /* connect new node to next node */
1587: lnk[0]++;
1588: }
1589: return PETSC_SUCCESS;
1590: }
1592: static inline PetscErrorCode PetscLLCondensedClean_fast(PETSC_UNUSED PetscInt nidx, PetscInt *indices, PetscInt *lnk)
1593: {
1594: const PetscInt nlnk = lnk[0];
1595: PetscInt next = lnk[5]; /* first node */
1597: for (PetscInt k = 0, cnt = 0; k < nlnk; k++) {
1598: for (PetscInt j = 0; j < lnk[next + 1]; j++) indices[cnt++] = lnk[next] + j;
1599: next = lnk[next + 2];
1600: }
1601: lnk[0] = 0; /* nlnk: number of links */
1602: lnk[1] = 0; /* number of integer entries represented in list */
1603: lnk[3] = PETSC_MIN_INT + 1; /* value in the first node */
1604: lnk[4] = 1; /* count for the first node */
1605: lnk[5] = 6; /* next for the first node */
1606: lnk[6] = PETSC_MAX_INT - 1; /* value in the last node */
1607: lnk[7] = 1; /* count for the last node */
1608: lnk[8] = 0; /* next valid location to make link */
1609: return PETSC_SUCCESS;
1610: }
1612: static inline PetscErrorCode PetscLLCondensedView_fast(const PetscInt *lnk)
1613: {
1614: const PetscInt nlnk = lnk[0];
1615: PetscInt next = lnk[5]; /* first node */
1617: for (PetscInt k = 0; k < nlnk; k++) {
1618: #if 0 /* Debugging code */
1619: printf("%d value %d len %d next %d\n", next, lnk[next], lnk[next + 1], lnk[next + 2]);
1620: #endif
1621: next = lnk[next + 2];
1622: }
1623: return PETSC_SUCCESS;
1624: }
1626: static inline PetscErrorCode PetscLLCondensedDestroy_fast(PetscInt *lnk)
1627: {
1628: return PetscFree(lnk);
1629: }
1631: PETSC_EXTERN PetscErrorCode PetscCDCreate(PetscInt, PetscCoarsenData **);
1632: PETSC_EXTERN PetscErrorCode PetscCDDestroy(PetscCoarsenData *);
1633: PETSC_EXTERN PetscErrorCode PetscCDIntNdSetID(PetscCDIntNd *, PetscInt);
1634: PETSC_EXTERN PetscErrorCode PetscCDIntNdGetID(const PetscCDIntNd *, PetscInt *);
1635: PETSC_EXTERN PetscErrorCode PetscCDAppendID(PetscCoarsenData *, PetscInt, PetscInt);
1636: PETSC_EXTERN PetscErrorCode PetscCDMoveAppend(PetscCoarsenData *, PetscInt, PetscInt);
1637: PETSC_EXTERN PetscErrorCode PetscCDAppendNode(PetscCoarsenData *, PetscInt, PetscCDIntNd *);
1638: PETSC_EXTERN PetscErrorCode PetscCDRemoveNextNode(PetscCoarsenData *, PetscInt, PetscCDIntNd *);
1639: PETSC_EXTERN PetscErrorCode PetscCDCountAt(const PetscCoarsenData *, PetscInt, PetscInt *);
1640: PETSC_EXTERN PetscErrorCode PetscCDIsEmptyAt(const PetscCoarsenData *, PetscInt, PetscBool *);
1641: PETSC_EXTERN PetscErrorCode PetscCDSetChunkSize(PetscCoarsenData *, PetscInt);
1642: PETSC_EXTERN PetscErrorCode PetscCDPrint(const PetscCoarsenData *, PetscInt, MPI_Comm);
1643: PETSC_EXTERN PetscErrorCode PetscCDGetNonemptyIS(PetscCoarsenData *, IS *);
1644: PETSC_EXTERN PetscErrorCode PetscCDGetMat(PetscCoarsenData *, Mat *);
1645: PETSC_EXTERN PetscErrorCode PetscCDSetMat(PetscCoarsenData *, Mat);
1646: PETSC_EXTERN PetscErrorCode PetscCDClearMat(PetscCoarsenData *);
1647: PETSC_EXTERN PetscErrorCode PetscCDRemoveAllAt(PetscCoarsenData *, PetscInt);
1648: PETSC_EXTERN PetscErrorCode PetscCDCount(const PetscCoarsenData *, PetscInt *_sz);
1650: PETSC_EXTERN PetscErrorCode PetscCDGetHeadPos(const PetscCoarsenData *, PetscInt, PetscCDIntNd **);
1651: PETSC_EXTERN PetscErrorCode PetscCDGetNextPos(const PetscCoarsenData *, PetscInt, PetscCDIntNd **);
1652: PETSC_EXTERN PetscErrorCode PetscCDGetASMBlocks(const PetscCoarsenData *, const PetscInt, PetscInt *, IS **);
1654: /* this is extern because it is used in MatFDColoringUseDM() which is in the DM library */
1655: PETSC_EXTERN PetscErrorCode MatFDColoringApply_AIJ(Mat, MatFDColoring, Vec, void *);
1657: PETSC_EXTERN PetscLogEvent MAT_Mult;
1658: PETSC_EXTERN PetscLogEvent MAT_MultAdd;
1659: PETSC_EXTERN PetscLogEvent MAT_MultTranspose;
1660: PETSC_EXTERN PetscLogEvent MAT_MultHermitianTranspose;
1661: PETSC_EXTERN PetscLogEvent MAT_MultTransposeAdd;
1662: PETSC_EXTERN PetscLogEvent MAT_MultHermitianTransposeAdd;
1663: PETSC_EXTERN PetscLogEvent MAT_Solve;
1664: PETSC_EXTERN PetscLogEvent MAT_Solves;
1665: PETSC_EXTERN PetscLogEvent MAT_SolveAdd;
1666: PETSC_EXTERN PetscLogEvent MAT_SolveTranspose;
1667: PETSC_EXTERN PetscLogEvent MAT_SolveTransposeAdd;
1668: PETSC_EXTERN PetscLogEvent MAT_SOR;
1669: PETSC_EXTERN PetscLogEvent MAT_ForwardSolve;
1670: PETSC_EXTERN PetscLogEvent MAT_BackwardSolve;
1671: PETSC_EXTERN PetscLogEvent MAT_LUFactor;
1672: PETSC_EXTERN PetscLogEvent MAT_LUFactorSymbolic;
1673: PETSC_EXTERN PetscLogEvent MAT_LUFactorNumeric;
1674: PETSC_EXTERN PetscLogEvent MAT_QRFactor;
1675: PETSC_EXTERN PetscLogEvent MAT_QRFactorSymbolic;
1676: PETSC_EXTERN PetscLogEvent MAT_QRFactorNumeric;
1677: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactor;
1678: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactorSymbolic;
1679: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactorNumeric;
1680: PETSC_EXTERN PetscLogEvent MAT_ILUFactor;
1681: PETSC_EXTERN PetscLogEvent MAT_ILUFactorSymbolic;
1682: PETSC_EXTERN PetscLogEvent MAT_ICCFactorSymbolic;
1683: PETSC_EXTERN PetscLogEvent MAT_Copy;
1684: PETSC_EXTERN PetscLogEvent MAT_Convert;
1685: PETSC_EXTERN PetscLogEvent MAT_Scale;
1686: PETSC_EXTERN PetscLogEvent MAT_AssemblyBegin;
1687: PETSC_EXTERN PetscLogEvent MAT_AssemblyEnd;
1688: PETSC_EXTERN PetscLogEvent MAT_SetValues;
1689: PETSC_EXTERN PetscLogEvent MAT_GetValues;
1690: PETSC_EXTERN PetscLogEvent MAT_GetRow;
1691: PETSC_EXTERN PetscLogEvent MAT_GetRowIJ;
1692: PETSC_EXTERN PetscLogEvent MAT_CreateSubMats;
1693: PETSC_EXTERN PetscLogEvent MAT_GetOrdering;
1694: PETSC_EXTERN PetscLogEvent MAT_RedundantMat;
1695: PETSC_EXTERN PetscLogEvent MAT_IncreaseOverlap;
1696: PETSC_EXTERN PetscLogEvent MAT_Partitioning;
1697: PETSC_EXTERN PetscLogEvent MAT_PartitioningND;
1698: PETSC_EXTERN PetscLogEvent MAT_Coarsen;
1699: PETSC_EXTERN PetscLogEvent MAT_ZeroEntries;
1700: PETSC_EXTERN PetscLogEvent MAT_Load;
1701: PETSC_EXTERN PetscLogEvent MAT_View;
1702: PETSC_EXTERN PetscLogEvent MAT_AXPY;
1703: PETSC_EXTERN PetscLogEvent MAT_FDColoringCreate;
1704: PETSC_EXTERN PetscLogEvent MAT_TransposeColoringCreate;
1705: PETSC_EXTERN PetscLogEvent MAT_FDColoringSetUp;
1706: PETSC_EXTERN PetscLogEvent MAT_FDColoringApply;
1707: PETSC_EXTERN PetscLogEvent MAT_Transpose;
1708: PETSC_EXTERN PetscLogEvent MAT_FDColoringFunction;
1709: PETSC_EXTERN PetscLogEvent MAT_CreateSubMat;
1710: PETSC_EXTERN PetscLogEvent MAT_MatSolve;
1711: PETSC_EXTERN PetscLogEvent MAT_MatTrSolve;
1712: PETSC_EXTERN PetscLogEvent MAT_MatMultSymbolic;
1713: PETSC_EXTERN PetscLogEvent MAT_MatMultNumeric;
1714: PETSC_EXTERN PetscLogEvent MAT_Getlocalmatcondensed;
1715: PETSC_EXTERN PetscLogEvent MAT_GetBrowsOfAcols;
1716: PETSC_EXTERN PetscLogEvent MAT_GetBrowsOfAocols;
1717: PETSC_EXTERN PetscLogEvent MAT_PtAPSymbolic;
1718: PETSC_EXTERN PetscLogEvent MAT_PtAPNumeric;
1719: PETSC_EXTERN PetscLogEvent MAT_Seqstompinum;
1720: PETSC_EXTERN PetscLogEvent MAT_Seqstompisym;
1721: PETSC_EXTERN PetscLogEvent MAT_Seqstompi;
1722: PETSC_EXTERN PetscLogEvent MAT_Getlocalmat;
1723: PETSC_EXTERN PetscLogEvent MAT_RARtSymbolic;
1724: PETSC_EXTERN PetscLogEvent MAT_RARtNumeric;
1725: PETSC_EXTERN PetscLogEvent MAT_MatTransposeMultSymbolic;
1726: PETSC_EXTERN PetscLogEvent MAT_MatTransposeMultNumeric;
1727: PETSC_EXTERN PetscLogEvent MAT_TransposeMatMultSymbolic;
1728: PETSC_EXTERN PetscLogEvent MAT_TransposeMatMultNumeric;
1729: PETSC_EXTERN PetscLogEvent MAT_MatMatMultSymbolic;
1730: PETSC_EXTERN PetscLogEvent MAT_MatMatMultNumeric;
1731: PETSC_EXTERN PetscLogEvent MAT_Getsymtransreduced;
1732: PETSC_EXTERN PetscLogEvent MAT_GetSeqNonzeroStructure;
1733: PETSC_EXTERN PetscLogEvent MATMFFD_Mult;
1734: PETSC_EXTERN PetscLogEvent MAT_GetMultiProcBlock;
1735: PETSC_EXTERN PetscLogEvent MAT_CUSPARSECopyToGPU;
1736: PETSC_EXTERN PetscLogEvent MAT_CUSPARSECopyFromGPU;
1737: PETSC_EXTERN PetscLogEvent MAT_CUSPARSEGenerateTranspose;
1738: PETSC_EXTERN PetscLogEvent MAT_CUSPARSESolveAnalysis;
1739: PETSC_EXTERN PetscLogEvent MAT_HIPSPARSECopyToGPU;
1740: PETSC_EXTERN PetscLogEvent MAT_HIPSPARSECopyFromGPU;
1741: PETSC_EXTERN PetscLogEvent MAT_HIPSPARSEGenerateTranspose;
1742: PETSC_EXTERN PetscLogEvent MAT_HIPSPARSESolveAnalysis;
1743: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatch;
1744: PETSC_EXTERN PetscLogEvent MAT_ViennaCLCopyToGPU;
1745: PETSC_EXTERN PetscLogEvent MAT_DenseCopyToGPU;
1746: PETSC_EXTERN PetscLogEvent MAT_DenseCopyFromGPU;
1747: PETSC_EXTERN PetscLogEvent MAT_Merge;
1748: PETSC_EXTERN PetscLogEvent MAT_Residual;
1749: PETSC_EXTERN PetscLogEvent MAT_SetRandom;
1750: PETSC_EXTERN PetscLogEvent MAT_FactorFactS;
1751: PETSC_EXTERN PetscLogEvent MAT_FactorInvS;
1752: PETSC_EXTERN PetscLogEvent MAT_PreallCOO;
1753: PETSC_EXTERN PetscLogEvent MAT_SetVCOO;
1754: PETSC_EXTERN PetscLogEvent MATCOLORING_Apply;
1755: PETSC_EXTERN PetscLogEvent MATCOLORING_Comm;
1756: PETSC_EXTERN PetscLogEvent MATCOLORING_Local;
1757: PETSC_EXTERN PetscLogEvent MATCOLORING_ISCreate;
1758: PETSC_EXTERN PetscLogEvent MATCOLORING_SetUp;
1759: PETSC_EXTERN PetscLogEvent MATCOLORING_Weights;
1760: PETSC_EXTERN PetscLogEvent MAT_H2Opus_Build;
1761: PETSC_EXTERN PetscLogEvent MAT_H2Opus_Compress;
1762: PETSC_EXTERN PetscLogEvent MAT_H2Opus_Orthog;
1763: PETSC_EXTERN PetscLogEvent MAT_H2Opus_LR;
1764: PETSC_EXTERN PetscLogEvent MAT_CUDACopyToGPU;
1765: PETSC_EXTERN PetscLogEvent MAT_HIPCopyToGPU;