Actual source code: matimpl.h
petsc-3.9.4 2018-09-11
2: #ifndef __MATIMPL_H
5: #include <petscmat.h>
6: #include <petscmatcoarsen.h>
7: #include <petsc/private/petscimpl.h>
9: PETSC_EXTERN PetscBool MatRegisterAllCalled;
10: PETSC_EXTERN PetscBool MatSeqAIJRegisterAllCalled;
11: PETSC_EXTERN PetscBool MatOrderingRegisterAllCalled;
12: PETSC_EXTERN PetscBool MatColoringRegisterAllCalled;
13: PETSC_EXTERN PetscBool MatPartitioningRegisterAllCalled;
14: PETSC_EXTERN PetscBool MatCoarsenRegisterAllCalled;
15: PETSC_EXTERN PetscErrorCode MatRegisterAll(void);
16: PETSC_EXTERN PetscErrorCode MatOrderingRegisterAll(void);
17: PETSC_EXTERN PetscErrorCode MatColoringRegisterAll(void);
18: PETSC_EXTERN PetscErrorCode MatPartitioningRegisterAll(void);
19: PETSC_EXTERN PetscErrorCode MatCoarsenRegisterAll(void);
20: PETSC_EXTERN PetscErrorCode MatSeqAIJRegisterAll(void);
22: /*
23: This file defines the parts of the matrix data structure that are
24: shared by all matrix types.
25: */
27: /*
28: If you add entries here also add them to the MATOP enum
29: in include/petscmat.h and include/petsc/finclude/petscmat.h
30: */
31: typedef struct _MatOps *MatOps;
32: struct _MatOps {
33: /* 0*/
34: PetscErrorCode (*setvalues)(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode);
35: PetscErrorCode (*getrow)(Mat,PetscInt,PetscInt *,PetscInt*[],PetscScalar*[]);
36: PetscErrorCode (*restorerow)(Mat,PetscInt,PetscInt *,PetscInt *[],PetscScalar *[]);
37: PetscErrorCode (*mult)(Mat,Vec,Vec);
38: PetscErrorCode (*multadd)(Mat,Vec,Vec,Vec);
39: /* 5*/
40: PetscErrorCode (*multtranspose)(Mat,Vec,Vec);
41: PetscErrorCode (*multtransposeadd)(Mat,Vec,Vec,Vec);
42: PetscErrorCode (*solve)(Mat,Vec,Vec);
43: PetscErrorCode (*solveadd)(Mat,Vec,Vec,Vec);
44: PetscErrorCode (*solvetranspose)(Mat,Vec,Vec);
45: /*10*/
46: PetscErrorCode (*solvetransposeadd)(Mat,Vec,Vec,Vec);
47: PetscErrorCode (*lufactor)(Mat,IS,IS,const MatFactorInfo*);
48: PetscErrorCode (*choleskyfactor)(Mat,IS,const MatFactorInfo*);
49: PetscErrorCode (*sor)(Mat,Vec,PetscReal,MatSORType,PetscReal,PetscInt,PetscInt,Vec);
50: PetscErrorCode (*transpose)(Mat,MatReuse,Mat *);
51: /*15*/
52: PetscErrorCode (*getinfo)(Mat,MatInfoType,MatInfo*);
53: PetscErrorCode (*equal)(Mat,Mat,PetscBool *);
54: PetscErrorCode (*getdiagonal)(Mat,Vec);
55: PetscErrorCode (*diagonalscale)(Mat,Vec,Vec);
56: PetscErrorCode (*norm)(Mat,NormType,PetscReal*);
57: /*20*/
58: PetscErrorCode (*assemblybegin)(Mat,MatAssemblyType);
59: PetscErrorCode (*assemblyend)(Mat,MatAssemblyType);
60: PetscErrorCode (*setoption)(Mat,MatOption,PetscBool );
61: PetscErrorCode (*zeroentries)(Mat);
62: /*24*/
63: PetscErrorCode (*zerorows)(Mat,PetscInt,const PetscInt[],PetscScalar,Vec,Vec);
64: PetscErrorCode (*lufactorsymbolic)(Mat,Mat,IS,IS,const MatFactorInfo*);
65: PetscErrorCode (*lufactornumeric)(Mat,Mat,const MatFactorInfo*);
66: PetscErrorCode (*choleskyfactorsymbolic)(Mat,Mat,IS,const MatFactorInfo*);
67: PetscErrorCode (*choleskyfactornumeric)(Mat,Mat,const MatFactorInfo*);
68: /*29*/
69: PetscErrorCode (*setup)(Mat);
70: PetscErrorCode (*ilufactorsymbolic)(Mat,Mat,IS,IS,const MatFactorInfo*);
71: PetscErrorCode (*iccfactorsymbolic)(Mat,Mat,IS,const MatFactorInfo*);
72: PetscErrorCode (*getdiagonalblock)(Mat,Mat*);
73: PetscErrorCode (*placeholder_33)(Mat);
74: /*34*/
75: PetscErrorCode (*duplicate)(Mat,MatDuplicateOption,Mat*);
76: PetscErrorCode (*forwardsolve)(Mat,Vec,Vec);
77: PetscErrorCode (*backwardsolve)(Mat,Vec,Vec);
78: PetscErrorCode (*ilufactor)(Mat,IS,IS,const MatFactorInfo*);
79: PetscErrorCode (*iccfactor)(Mat,IS,const MatFactorInfo*);
80: /*39*/
81: PetscErrorCode (*axpy)(Mat,PetscScalar,Mat,MatStructure);
82: PetscErrorCode (*createsubmatrices)(Mat,PetscInt,const IS[],const IS[],MatReuse,Mat *[]);
83: PetscErrorCode (*increaseoverlap)(Mat,PetscInt,IS[],PetscInt);
84: PetscErrorCode (*getvalues)(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],PetscScalar []);
85: PetscErrorCode (*copy)(Mat,Mat,MatStructure);
86: /*44*/
87: PetscErrorCode (*getrowmax)(Mat,Vec,PetscInt[]);
88: PetscErrorCode (*scale)(Mat,PetscScalar);
89: PetscErrorCode (*shift)(Mat,PetscScalar);
90: PetscErrorCode (*diagonalset)(Mat,Vec,InsertMode);
91: PetscErrorCode (*zerorowscolumns)(Mat,PetscInt,const PetscInt[],PetscScalar,Vec,Vec);
92: /*49*/
93: PetscErrorCode (*setrandom)(Mat,PetscRandom);
94: PetscErrorCode (*getrowij)(Mat,PetscInt,PetscBool ,PetscBool ,PetscInt*,const PetscInt *[],const PetscInt *[],PetscBool *);
95: PetscErrorCode (*restorerowij)(Mat,PetscInt,PetscBool ,PetscBool ,PetscInt *,const PetscInt *[],const PetscInt *[],PetscBool *);
96: PetscErrorCode (*getcolumnij)(Mat,PetscInt,PetscBool ,PetscBool ,PetscInt*,const PetscInt *[],const PetscInt *[],PetscBool *);
97: PetscErrorCode (*restorecolumnij)(Mat,PetscInt,PetscBool ,PetscBool ,PetscInt*,const PetscInt *[],const PetscInt *[],PetscBool *);
98: /*54*/
99: PetscErrorCode (*fdcoloringcreate)(Mat,ISColoring,MatFDColoring);
100: PetscErrorCode (*coloringpatch)(Mat,PetscInt,PetscInt,ISColoringValue[],ISColoring*);
101: PetscErrorCode (*setunfactored)(Mat);
102: PetscErrorCode (*permute)(Mat,IS,IS,Mat*);
103: PetscErrorCode (*setvaluesblocked)(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode);
104: /*59*/
105: PetscErrorCode (*createsubmatrix)(Mat,IS,IS,MatReuse,Mat*);
106: PetscErrorCode (*destroy)(Mat);
107: PetscErrorCode (*view)(Mat,PetscViewer);
108: PetscErrorCode (*convertfrom)(Mat, MatType,MatReuse,Mat*);
109: PetscErrorCode (*matmatmult)(Mat,Mat,Mat,MatReuse,PetscReal,Mat*);
110: /*64*/
111: PetscErrorCode (*matmatmultsymbolic)(Mat,Mat,Mat,PetscReal,Mat*);
112: PetscErrorCode (*matmatmultnumeric)(Mat,Mat,Mat,Mat);
113: PetscErrorCode (*setlocaltoglobalmapping)(Mat,ISLocalToGlobalMapping,ISLocalToGlobalMapping);
114: PetscErrorCode (*setvalueslocal)(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode);
115: PetscErrorCode (*zerorowslocal)(Mat,PetscInt,const PetscInt[],PetscScalar,Vec,Vec);
116: /*69*/
117: PetscErrorCode (*getrowmaxabs)(Mat,Vec,PetscInt[]);
118: PetscErrorCode (*getrowminabs)(Mat,Vec,PetscInt[]);
119: PetscErrorCode (*convert)(Mat, MatType,MatReuse,Mat*);
120: PetscErrorCode (*hasoperation)(Mat,MatOperation,PetscBool*);
121: PetscErrorCode (*placeholder_73)(Mat,void*);
122: /*74*/
123: PetscErrorCode (*setvaluesadifor)(Mat,PetscInt,void*);
124: PetscErrorCode (*fdcoloringapply)(Mat,MatFDColoring,Vec,void*);
125: PetscErrorCode (*setfromoptions)(PetscOptionItems*,Mat);
126: PetscErrorCode (*multconstrained)(Mat,Vec,Vec);
127: PetscErrorCode (*multtransposeconstrained)(Mat,Vec,Vec);
128: /*79*/
129: PetscErrorCode (*findzerodiagonals)(Mat,IS*);
130: PetscErrorCode (*mults)(Mat, Vecs, Vecs);
131: PetscErrorCode (*solves)(Mat, Vecs, Vecs);
132: PetscErrorCode (*getinertia)(Mat,PetscInt*,PetscInt*,PetscInt*);
133: PetscErrorCode (*load)(Mat, PetscViewer);
134: /*84*/
135: PetscErrorCode (*issymmetric)(Mat,PetscReal,PetscBool *);
136: PetscErrorCode (*ishermitian)(Mat,PetscReal,PetscBool *);
137: PetscErrorCode (*isstructurallysymmetric)(Mat,PetscBool *);
138: PetscErrorCode (*setvaluesblockedlocal)(Mat,PetscInt,const PetscInt[],PetscInt,const PetscInt[],const PetscScalar[],InsertMode);
139: PetscErrorCode (*getvecs)(Mat,Vec*,Vec*);
140: /*89*/
141: PetscErrorCode (*matmult)(Mat,Mat,MatReuse,PetscReal,Mat*);
142: PetscErrorCode (*matmultsymbolic)(Mat,Mat,PetscReal,Mat*);
143: PetscErrorCode (*matmultnumeric)(Mat,Mat,Mat);
144: PetscErrorCode (*ptap)(Mat,Mat,MatReuse,PetscReal,Mat*);
145: PetscErrorCode (*ptapsymbolic)(Mat,Mat,PetscReal,Mat*); /* double dispatch wrapper routine */
146: /*94*/
147: PetscErrorCode (*ptapnumeric)(Mat,Mat,Mat); /* double dispatch wrapper routine */
148: PetscErrorCode (*mattransposemult)(Mat,Mat,MatReuse,PetscReal,Mat*);
149: PetscErrorCode (*mattransposemultsymbolic)(Mat,Mat,PetscReal,Mat*);
150: PetscErrorCode (*mattransposemultnumeric)(Mat,Mat,Mat);
151: PetscErrorCode (*placeholder_98)(Mat);
152: /*99*/
153: PetscErrorCode (*placeholder_99)(Mat);
154: PetscErrorCode (*placeholder_100)(Mat);
155: PetscErrorCode (*placeholder_101)(Mat);
156: PetscErrorCode (*conjugate)(Mat); /* complex conjugate */
157: PetscErrorCode (*viewnative)(Mat,PetscViewer);
158: /*104*/
159: PetscErrorCode (*setvaluesrow)(Mat,PetscInt,const PetscScalar[]);
160: PetscErrorCode (*realpart)(Mat);
161: PetscErrorCode (*imaginarypart)(Mat);
162: PetscErrorCode (*getrowuppertriangular)(Mat);
163: PetscErrorCode (*restorerowuppertriangular)(Mat);
164: /*109*/
165: PetscErrorCode (*matsolve)(Mat,Mat,Mat);
166: PetscErrorCode (*matsolvetranspose)(Mat,Mat,Mat);
167: PetscErrorCode (*getrowmin)(Mat,Vec,PetscInt[]);
168: PetscErrorCode (*getcolumnvector)(Mat,Vec,PetscInt);
169: PetscErrorCode (*missingdiagonal)(Mat,PetscBool *,PetscInt*);
170: /*114*/
171: PetscErrorCode (*getseqnonzerostructure)(Mat,Mat *);
172: PetscErrorCode (*create)(Mat);
173: PetscErrorCode (*getghosts)(Mat,PetscInt*,const PetscInt *[]);
174: PetscErrorCode (*getlocalsubmatrix)(Mat,IS,IS,Mat*);
175: PetscErrorCode (*restorelocalsubmatrix)(Mat,IS,IS,Mat*);
176: /*119*/
177: PetscErrorCode (*multdiagonalblock)(Mat,Vec,Vec);
178: PetscErrorCode (*hermitiantranspose)(Mat,MatReuse,Mat*);
179: PetscErrorCode (*multhermitiantranspose)(Mat,Vec,Vec);
180: PetscErrorCode (*multhermitiantransposeadd)(Mat,Vec,Vec,Vec);
181: PetscErrorCode (*getmultiprocblock)(Mat,MPI_Comm,MatReuse,Mat*);
182: /*124*/
183: PetscErrorCode (*findnonzerorows)(Mat,IS*);
184: PetscErrorCode (*getcolumnnorms)(Mat,NormType,PetscReal*);
185: PetscErrorCode (*invertblockdiagonal)(Mat,const PetscScalar**);
186: PetscErrorCode (*placeholder_127)(Mat,Vec,Vec,Vec);
187: PetscErrorCode (*createsubmatricesmpi)(Mat,PetscInt,const IS[], const IS[], MatReuse, Mat**);
188: /*129*/
189: PetscErrorCode (*setvaluesbatch)(Mat,PetscInt,PetscInt,PetscInt*,const PetscScalar*);
190: PetscErrorCode (*transposematmult)(Mat,Mat,MatReuse,PetscReal,Mat*);
191: PetscErrorCode (*transposematmultsymbolic)(Mat,Mat,PetscReal,Mat*);
192: PetscErrorCode (*transposematmultnumeric)(Mat,Mat,Mat);
193: PetscErrorCode (*transposecoloringcreate)(Mat,ISColoring,MatTransposeColoring);
194: /*134*/
195: PetscErrorCode (*transcoloringapplysptoden)(MatTransposeColoring,Mat,Mat);
196: PetscErrorCode (*transcoloringapplydentosp)(MatTransposeColoring,Mat,Mat);
197: PetscErrorCode (*rart)(Mat,Mat,MatReuse,PetscReal,Mat*);
198: PetscErrorCode (*rartsymbolic)(Mat,Mat,PetscReal,Mat*); /* double dispatch wrapper routine */
199: PetscErrorCode (*rartnumeric)(Mat,Mat,Mat); /* double dispatch wrapper routine */
200: /*139*/
201: PetscErrorCode (*setblocksizes)(Mat,PetscInt,PetscInt);
202: PetscErrorCode (*aypx)(Mat,PetscScalar,Mat,MatStructure);
203: PetscErrorCode (*residual)(Mat,Vec,Vec,Vec);
204: PetscErrorCode (*fdcoloringsetup)(Mat,ISColoring,MatFDColoring);
205: PetscErrorCode (*findoffblockdiagonalentries)(Mat,IS*);
206: /*144*/
207: PetscErrorCode (*creatempimatconcatenateseqmat)(MPI_Comm,Mat,PetscInt,MatReuse,Mat*);
208: PetscErrorCode (*destroysubmatrices)(PetscInt,Mat*[]);
209: };
210: /*
211: If you add MatOps entries above also add them to the MATOP enum
212: in include/petscmat.h and include/petsc/finclude/petscmat.h
213: */
215: #include <petscsys.h>
216: PETSC_EXTERN PetscErrorCode MatRegisterOp(MPI_Comm, const char[], PetscVoidFunction, const char[], PetscInt, ...);
217: PETSC_EXTERN PetscErrorCode MatQueryOp(MPI_Comm, PetscVoidFunction*, const char[], PetscInt, ...);
219: typedef struct _p_MatBaseName* MatBaseName;
220: struct _p_MatBaseName {
221: char *bname,*sname,*mname;
222: MatBaseName next;
223: };
225: PETSC_EXTERN MatBaseName MatBaseNameList;
227: /*
228: Utility private matrix routines
229: */
230: PETSC_INTERN PetscErrorCode MatConvert_Basic(Mat, MatType,MatReuse,Mat*);
231: PETSC_INTERN PetscErrorCode MatCopy_Basic(Mat,Mat,MatStructure);
232: PETSC_INTERN PetscErrorCode MatDiagonalSet_Default(Mat,Vec,InsertMode);
234: #if defined(PETSC_USE_DEBUG)
235: # define MatCheckPreallocated(A,arg) do { \
236: if (PetscUnlikely(!(A)->preallocated)) SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Must call MatXXXSetPreallocation() or MatSetUp() on argument %D \"%s\" before %s()",(arg),#A,PETSC_FUNCTION_NAME); \
237: } while (0)
238: #else
239: # define MatCheckPreallocated(A,arg) do {} while (0)
240: #endif
242: /*
243: The stash is used to temporarily store inserted matrix values that
244: belong to another processor. During the assembly phase the stashed
245: values are moved to the correct processor and
246: */
248: typedef struct _MatStashSpace *PetscMatStashSpace;
250: struct _MatStashSpace {
251: PetscMatStashSpace next;
252: PetscScalar *space_head,*val;
253: PetscInt *idx,*idy;
254: PetscInt total_space_size;
255: PetscInt local_used;
256: PetscInt local_remaining;
257: };
259: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceGet(PetscInt,PetscInt,PetscMatStashSpace *);
260: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceContiguous(PetscInt,PetscMatStashSpace *,PetscScalar *,PetscInt *,PetscInt *);
261: PETSC_EXTERN PetscErrorCode PetscMatStashSpaceDestroy(PetscMatStashSpace*);
263: typedef struct {
264: PetscInt count;
265: } MatStashHeader;
267: typedef struct {
268: void *buffer; /* Of type blocktype, dynamically constructed */
269: PetscInt count;
270: char pending;
271: } MatStashFrame;
273: typedef struct _MatStash MatStash;
274: struct _MatStash {
275: PetscInt nmax; /* maximum stash size */
276: PetscInt umax; /* user specified max-size */
277: PetscInt oldnmax; /* the nmax value used previously */
278: PetscInt n; /* stash size */
279: PetscInt bs; /* block size of the stash */
280: PetscInt reallocs; /* preserve the no of mallocs invoked */
281: PetscMatStashSpace space_head,space; /* linked list to hold stashed global row/column numbers and matrix values */
283: PetscErrorCode (*ScatterBegin)(Mat,MatStash*,PetscInt*);
284: PetscErrorCode (*ScatterGetMesg)(MatStash*,PetscMPIInt*,PetscInt**,PetscInt**,PetscScalar**,PetscInt*);
285: PetscErrorCode (*ScatterEnd)(MatStash*);
286: PetscErrorCode (*ScatterDestroy)(MatStash*);
288: /* The following variables are used for communication */
289: MPI_Comm comm;
290: PetscMPIInt size,rank;
291: PetscMPIInt tag1,tag2;
292: MPI_Request *send_waits; /* array of send requests */
293: MPI_Request *recv_waits; /* array of receive requests */
294: MPI_Status *send_status; /* array of send status */
295: PetscInt nsends,nrecvs; /* numbers of sends and receives */
296: PetscScalar *svalues; /* sending data */
297: PetscInt *sindices;
298: PetscScalar **rvalues; /* receiving data (values) */
299: PetscInt **rindices; /* receiving data (indices) */
300: PetscInt nprocessed; /* number of messages already processed */
301: PetscMPIInt *flg_v; /* indicates what messages have arrived so far and from whom */
302: PetscBool reproduce;
303: PetscInt reproduce_count;
305: /* The following variables are used for BTS communication */
306: PetscBool subset_off_proc; /* Subsequent assemblies will set a subset (perhaps equal) of off-process entries set on first assembly */
307: PetscBool use_status; /* Use MPI_Status to determine number of items in each message */
308: PetscMPIInt nsendranks;
309: PetscMPIInt nrecvranks;
310: PetscMPIInt *sendranks;
311: PetscMPIInt *recvranks;
312: MatStashHeader *sendhdr,*recvhdr;
313: MatStashFrame *sendframes; /* pointers to the main messages */
314: MatStashFrame *recvframes;
315: MatStashFrame *recvframe_active;
316: PetscInt recvframe_i; /* index of block within active frame */
317: PetscMPIInt recvframe_count; /* Count actually sent for current frame */
318: PetscInt recvcount; /* Number of receives processed so far */
319: PetscMPIInt *some_indices; /* From last call to MPI_Waitsome */
320: MPI_Status *some_statuses; /* Statuses from last call to MPI_Waitsome */
321: PetscMPIInt some_count; /* Number of requests completed in last call to MPI_Waitsome */
322: PetscMPIInt some_i; /* Index of request currently being processed */
323: MPI_Request *sendreqs;
324: MPI_Request *recvreqs;
325: PetscSegBuffer segsendblocks;
326: PetscSegBuffer segrecvframe;
327: PetscSegBuffer segrecvblocks;
328: MPI_Datatype blocktype;
329: size_t blocktype_size;
330: InsertMode *insertmode; /* Pointer to check mat->insertmode and set upon message arrival in case no local values have been set. */
331: };
333: PETSC_INTERN PetscErrorCode MatStashCreate_Private(MPI_Comm,PetscInt,MatStash*);
334: PETSC_INTERN PetscErrorCode MatStashDestroy_Private(MatStash*);
335: PETSC_INTERN PetscErrorCode MatStashScatterEnd_Private(MatStash*);
336: PETSC_INTERN PetscErrorCode MatStashSetInitialSize_Private(MatStash*,PetscInt);
337: PETSC_INTERN PetscErrorCode MatStashGetInfo_Private(MatStash*,PetscInt*,PetscInt*);
338: PETSC_INTERN PetscErrorCode MatStashValuesRow_Private(MatStash*,PetscInt,PetscInt,const PetscInt[],const PetscScalar[],PetscBool );
339: PETSC_INTERN PetscErrorCode MatStashValuesCol_Private(MatStash*,PetscInt,PetscInt,const PetscInt[],const PetscScalar[],PetscInt,PetscBool );
340: PETSC_INTERN PetscErrorCode MatStashValuesRowBlocked_Private(MatStash*,PetscInt,PetscInt,const PetscInt[],const PetscScalar[],PetscInt,PetscInt,PetscInt);
341: PETSC_INTERN PetscErrorCode MatStashValuesColBlocked_Private(MatStash*,PetscInt,PetscInt,const PetscInt[],const PetscScalar[],PetscInt,PetscInt,PetscInt);
342: PETSC_INTERN PetscErrorCode MatStashScatterBegin_Private(Mat,MatStash*,PetscInt*);
343: PETSC_INTERN PetscErrorCode MatStashScatterGetMesg_Private(MatStash*,PetscMPIInt*,PetscInt**,PetscInt**,PetscScalar**,PetscInt*);
344: PETSC_INTERN PetscErrorCode MatGetInfo_External(Mat,MatInfoType,MatInfo*);
346: typedef struct {
347: PetscInt dim;
348: PetscInt dims[4];
349: PetscInt starts[4];
350: PetscBool noc; /* this is a single component problem, hence user will not set MatStencil.c */
351: } MatStencilInfo;
353: /* Info about using compressed row format */
354: typedef struct {
355: PetscBool use; /* indicates compressed rows have been checked and will be used */
356: PetscInt nrows; /* number of non-zero rows */
357: PetscInt *i; /* compressed row pointer */
358: PetscInt *rindex; /* compressed row index */
359: } Mat_CompressedRow;
360: PETSC_EXTERN PetscErrorCode MatCheckCompressedRow(Mat,PetscInt,Mat_CompressedRow*,PetscInt*,PetscInt,PetscReal);
362: typedef struct { /* used by MatCreateRedundantMatrix() for reusing matredundant */
363: PetscInt nzlocal,nsends,nrecvs;
364: PetscMPIInt *send_rank,*recv_rank;
365: PetscInt *sbuf_nz,*rbuf_nz,*sbuf_j,**rbuf_j;
366: PetscScalar *sbuf_a,**rbuf_a;
367: MPI_Comm subcomm; /* when user does not provide a subcomm */
368: IS isrow,iscol;
369: Mat *matseq;
370: } Mat_Redundant;
372: struct _p_Mat {
373: PETSCHEADER(struct _MatOps);
374: PetscLayout rmap,cmap;
375: void *data; /* implementation-specific data */
376: MatFactorType factortype; /* MAT_FACTOR_LU, ILU, CHOLESKY or ICC */
377: PetscBool assembled; /* is the matrix assembled? */
378: PetscBool was_assembled; /* new values inserted into assembled mat */
379: PetscInt num_ass; /* number of times matrix has been assembled */
380: PetscObjectState nonzerostate; /* each time new nonzeros locations are introduced into the matrix this is updated */
381: MatInfo info; /* matrix information */
382: InsertMode insertmode; /* have values been inserted in matrix or added? */
383: MatStash stash,bstash; /* used for assembling off-proc mat emements */
384: MatNullSpace nullsp; /* null space (operator is singular) */
385: MatNullSpace transnullsp; /* null space of transpose of operator */
386: MatNullSpace nearnullsp; /* near null space to be used by multigrid methods */
387: PetscInt congruentlayouts; /* are the rows and columns layouts congruent? */
388: PetscBool preallocated;
389: MatStencilInfo stencil; /* information for structured grid */
390: PetscBool symmetric,hermitian,structurally_symmetric,spd;
391: PetscBool symmetric_set,hermitian_set,structurally_symmetric_set,spd_set; /* if true, then corresponding flag is correct*/
392: PetscBool symmetric_eternal;
393: PetscBool nooffprocentries,nooffproczerorows;
394: PetscBool subsetoffprocentries;
395: PetscBool submat_singleis; /* for efficient PCSetUP_ASM() */
396: PetscBool structure_only;
397: #if defined(PETSC_HAVE_VIENNACL) || defined(PETSC_HAVE_VECCUDA)
398: PetscOffloadFlag valid_GPU_matrix; /* flag pointing to the matrix on the gpu*/
399: #endif
400: void *spptr; /* pointer for special library like SuperLU */
401: char *solvertype;
402: PetscBool checksymmetryonassembly,checknullspaceonassembly;
403: PetscReal checksymmetrytol;
404: Mat schur; /* Schur complement matrix */
405: MatFactorSchurStatus schur_status; /* status of the Schur complement matrix */
406: Mat_Redundant *redundant; /* used by MatCreateRedundantMatrix() */
407: PetscBool erroriffailure; /* Generate an error if detected (for example a zero pivot) instead of returning */
408: MatFactorError factorerrortype; /* type of error in factorization */
409: PetscReal factorerror_zeropivot_value; /* If numerical zero pivot was detected this is the computed value */
410: PetscInt factorerror_zeropivot_row; /* Row where zero pivot was detected */
411: };
413: PETSC_INTERN PetscErrorCode MatAXPY_Basic(Mat,PetscScalar,Mat,MatStructure);
414: PETSC_INTERN PetscErrorCode MatAXPY_BasicWithPreallocation(Mat,Mat,PetscScalar,Mat,MatStructure);
416: /*
417: Utility for MatFactor (Schur complement)
418: */
419: PETSC_INTERN PetscErrorCode MatFactorFactorizeSchurComplement_Private(Mat);
420: PETSC_INTERN PetscErrorCode MatFactorInvertSchurComplement_Private(Mat);
421: PETSC_INTERN PetscErrorCode MatFactorUpdateSchurStatus_Private(Mat);
422: PETSC_INTERN PetscErrorCode MatFactorSetUpInPlaceSchur_Private(Mat);
424: /*
425: Utility for MatZeroRows
426: */
427: PETSC_INTERN PetscErrorCode MatZeroRowsMapLocal_Private(Mat,PetscInt,const PetscInt*,PetscInt*,PetscInt**);
429: /*
430: Object for partitioning graphs
431: */
433: typedef struct _MatPartitioningOps *MatPartitioningOps;
434: struct _MatPartitioningOps {
435: PetscErrorCode (*apply)(MatPartitioning,IS*);
436: PetscErrorCode (*setfromoptions)(PetscOptionItems*,MatPartitioning);
437: PetscErrorCode (*destroy)(MatPartitioning);
438: PetscErrorCode (*view)(MatPartitioning,PetscViewer);
439: };
441: struct _p_MatPartitioning {
442: PETSCHEADER(struct _MatPartitioningOps);
443: Mat adj;
444: PetscInt *vertex_weights;
445: PetscReal *part_weights;
446: PetscInt n; /* number of partitions */
447: void *data;
448: PetscInt setupcalled;
449: };
451: /*
452: Object for coarsen graphs
453: */
454: typedef struct _MatCoarsenOps *MatCoarsenOps;
455: struct _MatCoarsenOps {
456: PetscErrorCode (*apply)(MatCoarsen);
457: PetscErrorCode (*setfromoptions)(PetscOptionItems*,MatCoarsen);
458: PetscErrorCode (*destroy)(MatCoarsen);
459: PetscErrorCode (*view)(MatCoarsen,PetscViewer);
460: };
462: struct _p_MatCoarsen {
463: PETSCHEADER(struct _MatCoarsenOps);
464: Mat graph;
465: PetscInt setupcalled;
466: void *subctx;
467: /* */
468: PetscBool strict_aggs;
469: IS perm;
470: PetscCoarsenData *agg_lists;
471: };
473: /*
474: MatFDColoring is used to compute Jacobian matrices efficiently
475: via coloring. The data structure is explained below in an example.
477: Color = 0 1 0 2 | 2 3 0
478: ---------------------------------------------------
479: 00 01 | 05
480: 10 11 | 14 15 Processor 0
481: 22 23 | 25
482: 32 33 |
483: ===================================================
484: | 44 45 46
485: 50 | 55 Processor 1
486: | 64 66
487: ---------------------------------------------------
489: ncolors = 4;
491: ncolumns = {2,1,1,0}
492: columns = {{0,2},{1},{3},{}}
493: nrows = {4,2,3,3}
494: rows = {{0,1,2,3},{0,1},{1,2,3},{0,1,2}}
495: vwscale = {dx(0),dx(1),dx(2),dx(3)} MPI Vec
496: vscale = {dx(0),dx(1),dx(2),dx(3),dx(4),dx(5)} Seq Vec
498: ncolumns = {1,0,1,1}
499: columns = {{6},{},{4},{5}}
500: nrows = {3,0,2,2}
501: rows = {{0,1,2},{},{1,2},{1,2}}
502: vwscale = {dx(4),dx(5),dx(6)} MPI Vec
503: vscale = {dx(0),dx(4),dx(5),dx(6)} Seq Vec
505: See the routine MatFDColoringApply() for how this data is used
506: to compute the Jacobian.
508: */
509: typedef struct {
510: PetscInt row;
511: PetscInt col;
512: PetscScalar *valaddr; /* address of value */
513: } MatEntry;
515: typedef struct {
516: PetscInt row;
517: PetscScalar *valaddr; /* address of value */
518: } MatEntry2;
520: struct _p_MatFDColoring{
521: PETSCHEADER(int);
522: PetscInt M,N,m; /* total rows, columns; local rows */
523: PetscInt rstart; /* first row owned by local processor */
524: PetscInt ncolors; /* number of colors */
525: PetscInt *ncolumns; /* number of local columns for a color */
526: PetscInt **columns; /* lists the local columns of each color (using global column numbering) */
527: PetscInt *nrows; /* number of local rows for each color */
528: MatEntry *matentry; /* holds (row, column, address of value) for Jacobian matrix entry */
529: MatEntry2 *matentry2; /* holds (row, address of value) for Jacobian matrix entry */
530: PetscScalar *dy; /* store a block of F(x+dx)-F(x) when J is in BAIJ format */
531: PetscReal error_rel; /* square root of relative error in computing function */
532: PetscReal umin; /* minimum allowable u'dx value */
533: Vec w1,w2,w3; /* work vectors used in computing Jacobian */
534: PetscBool fset; /* indicates that the initial function value F(X) is set */
535: PetscErrorCode (*f)(void); /* function that defines Jacobian */
536: void *fctx; /* optional user-defined context for use by the function f */
537: Vec vscale; /* holds FD scaling, i.e. 1/dx for each perturbed column */
538: PetscInt currentcolor; /* color for which function evaluation is being done now */
539: const char *htype; /* "wp" or "ds" */
540: ISColoringType ctype; /* IS_COLORING_GLOBAL or IS_COLORING_LOCAL */
541: PetscInt brows,bcols; /* number of block rows or columns for speedup inserting the dense matrix into sparse Jacobian */
542: PetscBool setupcalled; /* true if setup has been called */
543: PetscBool viewed; /* true if the -mat_fd_coloring_view has been triggered already */
544: void (*ftn_func_pointer)(void),*ftn_func_cntx; /* serve the same purpose as *fortran_func_pointers in PETSc objects */
545: };
547: typedef struct _MatColoringOps *MatColoringOps;
548: struct _MatColoringOps {
549: PetscErrorCode (*destroy)(MatColoring);
550: PetscErrorCode (*setfromoptions)(PetscOptionItems*,MatColoring);
551: PetscErrorCode (*view)(MatColoring,PetscViewer);
552: PetscErrorCode (*apply)(MatColoring,ISColoring*);
553: PetscErrorCode (*weights)(MatColoring,PetscReal**,PetscInt**);
554: };
556: struct _p_MatColoring {
557: PETSCHEADER(struct _MatColoringOps);
558: Mat mat;
559: PetscInt dist; /* distance of the coloring */
560: PetscInt maxcolors; /* the maximum number of colors returned, maxcolors=1 for MIS */
561: void *data; /* inner context */
562: PetscBool valid; /* check to see if what is produced is a valid coloring */
563: MatColoringWeightType weight_type; /* type of weight computation to be performed */
564: PetscReal *user_weights; /* custom weights and permutation */
565: PetscInt *user_lperm;
566: PetscBool valid_iscoloring; /* check to see if matcoloring is produced a valid iscoloring */
567: };
569: struct _p_MatTransposeColoring{
570: PETSCHEADER(int);
571: PetscInt M,N,m; /* total rows, columns; local rows */
572: PetscInt rstart; /* first row owned by local processor */
573: PetscInt ncolors; /* number of colors */
574: PetscInt *ncolumns; /* number of local columns for a color */
575: PetscInt *nrows; /* number of local rows for each color */
576: PetscInt currentcolor; /* color for which function evaluation is being done now */
577: ISColoringType ctype; /* IS_COLORING_GLOBAL or IS_COLORING_LOCAL */
579: PetscInt *colorforrow,*colorforcol; /* pointer to rows and columns */
580: PetscInt *rows; /* lists the local rows for each color (using the local row numbering) */
581: PetscInt *den2sp; /* maps (row,color) in the dense matrix to index of sparse matrix array a->a */
582: PetscInt *columns; /* lists the local columns of each color (using global column numbering) */
583: PetscInt brows; /* number of rows for efficient implementation of MatTransColoringApplyDenToSp() */
584: PetscInt *lstart; /* array used for loop over row blocks of Csparse */
585: };
587: /*
588: Null space context for preconditioner/operators
589: */
590: struct _p_MatNullSpace {
591: PETSCHEADER(int);
592: PetscBool has_cnst;
593: PetscInt n;
594: Vec* vecs;
595: PetscScalar* alpha; /* for projections */
596: PetscErrorCode (*remove)(MatNullSpace,Vec,void*); /* for user provided removal function */
597: void* rmctx; /* context for remove() function */
598: };
600: /*
601: Checking zero pivot for LU, ILU preconditioners.
602: */
603: typedef struct {
604: PetscInt nshift,nshift_max;
605: PetscReal shift_amount,shift_lo,shift_hi,shift_top,shift_fraction;
606: PetscBool newshift;
607: PetscReal rs; /* active row sum of abs(offdiagonals) */
608: PetscScalar pv; /* pivot of the active row */
609: } FactorShiftCtx;
611: /*
612: Used by MatCreateSubMatrices_MPIXAIJ_Local()
613: */
614: #include <petscctable.h>
615: typedef struct { /* used by MatCreateSubMatrices_MPIAIJ_SingleIS_Local() and MatCreateSubMatrices_MPIAIJ_Local */
616: PetscInt id; /* index of submats, only submats[0] is responsible for deleting some arrays below */
617: PetscInt nrqs,nrqr;
618: PetscInt **rbuf1,**rbuf2,**rbuf3,**sbuf1,**sbuf2;
619: PetscInt **ptr;
620: PetscInt *tmp;
621: PetscInt *ctr;
622: PetscInt *pa; /* proc array */
623: PetscInt *req_size,*req_source1,*req_source2;
624: PetscBool allcolumns,allrows;
625: PetscBool singleis;
626: PetscInt *row2proc; /* row to proc map */
627: PetscInt nstages;
628: #if defined(PETSC_USE_CTABLE)
629: PetscTable cmap,rmap;
630: PetscInt *cmap_loc,*rmap_loc;
631: #else
632: PetscInt *cmap,*rmap;
633: #endif
635: PetscErrorCode (*destroy)(Mat);
636: } Mat_SubSppt;
638: PETSC_EXTERN PetscErrorCode MatFactorDumpMatrix(Mat);
639: PETSC_INTERN PetscErrorCode MatShift_Basic(Mat,PetscScalar);
640: PETSC_INTERN PetscErrorCode MatSetBlockSizes_Default(Mat,PetscInt,PetscInt);
642: PETSC_STATIC_INLINE PetscErrorCode MatPivotCheck_nz(Mat mat,const MatFactorInfo *info,FactorShiftCtx *sctx,PetscInt row)
643: {
644: PetscReal _rs = sctx->rs;
645: PetscReal _zero = info->zeropivot*_rs;
648: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)){
649: /* force |diag| > zeropivot*rs */
650: if (!sctx->nshift) sctx->shift_amount = info->shiftamount;
651: else sctx->shift_amount *= 2.0;
652: sctx->newshift = PETSC_TRUE;
653: (sctx->nshift)++;
654: } else {
655: sctx->newshift = PETSC_FALSE;
656: }
657: return(0);
658: }
660: PETSC_STATIC_INLINE PetscErrorCode MatPivotCheck_pd(Mat mat,const MatFactorInfo *info,FactorShiftCtx *sctx,PetscInt row)
661: {
662: PetscReal _rs = sctx->rs;
663: PetscReal _zero = info->zeropivot*_rs;
666: if (PetscRealPart(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)){
667: /* force matfactor to be diagonally dominant */
668: if (sctx->nshift == sctx->nshift_max) {
669: sctx->shift_fraction = sctx->shift_hi;
670: } else {
671: sctx->shift_lo = sctx->shift_fraction;
672: sctx->shift_fraction = (sctx->shift_hi+sctx->shift_lo)/2.;
673: }
674: sctx->shift_amount = sctx->shift_fraction * sctx->shift_top;
675: sctx->nshift++;
676: sctx->newshift = PETSC_TRUE;
677: } else {
678: sctx->newshift = PETSC_FALSE;
679: }
680: return(0);
681: }
683: PETSC_STATIC_INLINE PetscErrorCode MatPivotCheck_inblocks(Mat mat,const MatFactorInfo *info,FactorShiftCtx *sctx,PetscInt row)
684: {
685: PetscReal _zero = info->zeropivot;
688: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)){
689: sctx->pv += info->shiftamount;
690: sctx->shift_amount = 0.0;
691: sctx->nshift++;
692: }
693: sctx->newshift = PETSC_FALSE;
694: return(0);
695: }
697: PETSC_STATIC_INLINE PetscErrorCode MatPivotCheck_none(Mat fact,Mat mat,const MatFactorInfo *info,FactorShiftCtx *sctx,PetscInt row)
698: {
699: PetscReal _zero = info->zeropivot;
703: sctx->newshift = PETSC_FALSE;
704: if (PetscAbsScalar(sctx->pv) <= _zero && !PetscIsNanScalar(sctx->pv)) {
705: if (!mat->erroriffailure) {
706: PetscInfo3(mat,"Detected zero pivot in factorization in row %D value %g tolerance %g\n",row,(double)PetscAbsScalar(sctx->pv),(double)_zero);
707: fact->factorerrortype = MAT_FACTOR_NUMERIC_ZEROPIVOT;
708: fact->factorerror_zeropivot_value = PetscAbsScalar(sctx->pv);
709: fact->factorerror_zeropivot_row = row;
710: } else SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_MAT_LU_ZRPVT,"Zero pivot row %D value %g tolerance %g\n",row,(double)PetscAbsScalar(sctx->pv),(double)_zero);
711: }
712: return(0);
713: }
715: PETSC_STATIC_INLINE PetscErrorCode MatPivotCheck(Mat fact,Mat mat,const MatFactorInfo *info,FactorShiftCtx *sctx,PetscInt row)
716: {
720: if (info->shifttype == (PetscReal) MAT_SHIFT_NONZERO){
721: MatPivotCheck_nz(mat,info,sctx,row);
722: } else if (info->shifttype == (PetscReal) MAT_SHIFT_POSITIVE_DEFINITE){
723: MatPivotCheck_pd(mat,info,sctx,row);
724: } else if (info->shifttype == (PetscReal) MAT_SHIFT_INBLOCKS){
725: MatPivotCheck_inblocks(mat,info,sctx,row);
726: } else {
727: MatPivotCheck_none(fact,mat,info,sctx,row);
728: }
729: return(0);
730: }
732: /*
733: Create and initialize a linked list
734: Input Parameters:
735: idx_start - starting index of the list
736: lnk_max - max value of lnk indicating the end of the list
737: nlnk - max length of the list
738: Output Parameters:
739: lnk - list initialized
740: bt - PetscBT (bitarray) with all bits set to false
741: lnk_empty - flg indicating the list is empty
742: */
743: #define PetscLLCreate(idx_start,lnk_max,nlnk,lnk,bt) \
744: (PetscMalloc1(nlnk,&lnk) || PetscBTCreate(nlnk,&(bt)) || (lnk[idx_start] = lnk_max,0))
746: #define PetscLLCreate_new(idx_start,lnk_max,nlnk,lnk,bt,lnk_empty)\
747: (PetscMalloc1(nlnk,&lnk) || PetscBTCreate(nlnk,&(bt)) || (lnk_empty = PETSC_TRUE,0) ||(lnk[idx_start] = lnk_max,0))
749: /*
750: Add an index set into a sorted linked list
751: Input Parameters:
752: nidx - number of input indices
753: indices - integer array
754: idx_start - starting index of the list
755: lnk - linked list(an integer array) that is created
756: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
757: output Parameters:
758: nlnk - number of newly added indices
759: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
760: bt - updated PetscBT (bitarray)
761: */
762: #define PetscLLAdd(nidx,indices,idx_start,nlnk,lnk,bt) 0;\
763: {\
764: PetscInt _k,_entry,_location,_lnkdata;\
765: nlnk = 0;\
766: _lnkdata = idx_start;\
767: for (_k=0; _k<nidx; _k++){\
768: _entry = indices[_k];\
769: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
770: /* search for insertion location */\
771: /* start from the beginning if _entry < previous _entry */\
772: if (_k && _entry < _lnkdata) _lnkdata = idx_start;\
773: do {\
774: _location = _lnkdata;\
775: _lnkdata = lnk[_location];\
776: } while (_entry > _lnkdata);\
777: /* insertion location is found, add entry into lnk */\
778: lnk[_location] = _entry;\
779: lnk[_entry] = _lnkdata;\
780: nlnk++;\
781: _lnkdata = _entry; /* next search starts from here if next_entry > _entry */\
782: }\
783: }\
784: }
786: /*
787: Add a permuted index set into a sorted linked list
788: Input Parameters:
789: nidx - number of input indices
790: indices - integer array
791: perm - permutation of indices
792: idx_start - starting index of the list
793: lnk - linked list(an integer array) that is created
794: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
795: output Parameters:
796: nlnk - number of newly added indices
797: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
798: bt - updated PetscBT (bitarray)
799: */
800: #define PetscLLAddPerm(nidx,indices,perm,idx_start,nlnk,lnk,bt) 0;\
801: {\
802: PetscInt _k,_entry,_location,_lnkdata;\
803: nlnk = 0;\
804: _lnkdata = idx_start;\
805: for (_k=0; _k<nidx; _k++){\
806: _entry = perm[indices[_k]];\
807: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
808: /* search for insertion location */\
809: /* start from the beginning if _entry < previous _entry */\
810: if (_k && _entry < _lnkdata) _lnkdata = idx_start;\
811: do {\
812: _location = _lnkdata;\
813: _lnkdata = lnk[_location];\
814: } while (_entry > _lnkdata);\
815: /* insertion location is found, add entry into lnk */\
816: lnk[_location] = _entry;\
817: lnk[_entry] = _lnkdata;\
818: nlnk++;\
819: _lnkdata = _entry; /* next search starts from here if next_entry > _entry */\
820: }\
821: }\
822: }
824: /*
825: Add a SORTED ascending index set into a sorted linked list - same as PetscLLAdd() bus skip 'if (_k && _entry < _lnkdata) _lnkdata = idx_start;'
826: Input Parameters:
827: nidx - number of input indices
828: indices - sorted integer array
829: idx_start - starting index of the list
830: lnk - linked list(an integer array) that is created
831: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
832: output Parameters:
833: nlnk - number of newly added indices
834: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
835: bt - updated PetscBT (bitarray)
836: */
837: #define PetscLLAddSorted(nidx,indices,idx_start,nlnk,lnk,bt) 0;\
838: {\
839: PetscInt _k,_entry,_location,_lnkdata;\
840: nlnk = 0;\
841: _lnkdata = idx_start;\
842: for (_k=0; _k<nidx; _k++){\
843: _entry = indices[_k];\
844: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
845: /* search for insertion location */\
846: do {\
847: _location = _lnkdata;\
848: _lnkdata = lnk[_location];\
849: } while (_entry > _lnkdata);\
850: /* insertion location is found, add entry into lnk */\
851: lnk[_location] = _entry;\
852: lnk[_entry] = _lnkdata;\
853: nlnk++;\
854: _lnkdata = _entry; /* next search starts from here */\
855: }\
856: }\
857: }
859: #define PetscLLAddSorted_new(nidx,indices,idx_start,lnk_empty,nlnk,lnk,bt) 0; \
860: {\
861: PetscInt _k,_entry,_location,_lnkdata;\
862: if (lnk_empty){\
863: _lnkdata = idx_start; \
864: for (_k=0; _k<nidx; _k++){ \
865: _entry = indices[_k]; \
866: PetscBTSet(bt,_entry); /* mark the new entry */ \
867: _location = _lnkdata; \
868: _lnkdata = lnk[_location]; \
869: /* insertion location is found, add entry into lnk */ \
870: lnk[_location] = _entry; \
871: lnk[_entry] = _lnkdata; \
872: _lnkdata = _entry; /* next search starts from here */ \
873: } \
874: /*\
875: lnk[indices[nidx-1]] = lnk[idx_start];\
876: lnk[idx_start] = indices[0];\
877: PetscBTSet(bt,indices[0]); \
878: for (_k=1; _k<nidx; _k++){ \
879: PetscBTSet(bt,indices[_k]); \
880: lnk[indices[_k-1]] = indices[_k]; \
881: } \
882: */\
883: nlnk = nidx;\
884: lnk_empty = PETSC_FALSE;\
885: } else {\
886: nlnk = 0; \
887: _lnkdata = idx_start; \
888: for (_k=0; _k<nidx; _k++){ \
889: _entry = indices[_k]; \
890: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */ \
891: /* search for insertion location */ \
892: do { \
893: _location = _lnkdata; \
894: _lnkdata = lnk[_location]; \
895: } while (_entry > _lnkdata); \
896: /* insertion location is found, add entry into lnk */ \
897: lnk[_location] = _entry; \
898: lnk[_entry] = _lnkdata; \
899: nlnk++; \
900: _lnkdata = _entry; /* next search starts from here */ \
901: } \
902: } \
903: } \
904: }
906: /*
907: Add a SORTED index set into a sorted linked list used for LUFactorSymbolic()
908: Same as PetscLLAddSorted() with an additional operation:
909: count the number of input indices that are no larger than 'diag'
910: Input Parameters:
911: indices - sorted integer array
912: idx_start - starting index of the list, index of pivot row
913: lnk - linked list(an integer array) that is created
914: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
915: diag - index of the active row in LUFactorSymbolic
916: nzbd - number of input indices with indices <= idx_start
917: im - im[idx_start] is initialized as num of nonzero entries in row=idx_start
918: output Parameters:
919: nlnk - number of newly added indices
920: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from indices
921: bt - updated PetscBT (bitarray)
922: im - im[idx_start]: unchanged if diag is not an entry
923: : num of entries with indices <= diag if diag is an entry
924: */
925: #define PetscLLAddSortedLU(indices,idx_start,nlnk,lnk,bt,diag,nzbd,im) 0;\
926: {\
927: PetscInt _k,_entry,_location,_lnkdata,_nidx;\
928: nlnk = 0;\
929: _lnkdata = idx_start;\
930: _nidx = im[idx_start] - nzbd; /* num of entries with idx_start < index <= diag */\
931: for (_k=0; _k<_nidx; _k++){\
932: _entry = indices[_k];\
933: nzbd++;\
934: if ( _entry== diag) im[idx_start] = nzbd;\
935: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
936: /* search for insertion location */\
937: do {\
938: _location = _lnkdata;\
939: _lnkdata = lnk[_location];\
940: } while (_entry > _lnkdata);\
941: /* insertion location is found, add entry into lnk */\
942: lnk[_location] = _entry;\
943: lnk[_entry] = _lnkdata;\
944: nlnk++;\
945: _lnkdata = _entry; /* next search starts from here */\
946: }\
947: }\
948: }
950: /*
951: Copy data on the list into an array, then initialize the list
952: Input Parameters:
953: idx_start - starting index of the list
954: lnk_max - max value of lnk indicating the end of the list
955: nlnk - number of data on the list to be copied
956: lnk - linked list
957: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
958: output Parameters:
959: indices - array that contains the copied data
960: lnk - linked list that is cleaned and initialize
961: bt - PetscBT (bitarray) with all bits set to false
962: */
963: #define PetscLLClean(idx_start,lnk_max,nlnk,lnk,indices,bt) 0;\
964: {\
965: PetscInt _j,_idx=idx_start;\
966: for (_j=0; _j<nlnk; _j++){\
967: _idx = lnk[_idx];\
968: indices[_j] = _idx;\
969: PetscBTClear(bt,_idx);\
970: }\
971: lnk[idx_start] = lnk_max;\
972: }
973: /*
974: Free memories used by the list
975: */
976: #define PetscLLDestroy(lnk,bt) (PetscFree(lnk) || PetscBTDestroy(&(bt)))
978: /* Routines below are used for incomplete matrix factorization */
979: /*
980: Create and initialize a linked list and its levels
981: Input Parameters:
982: idx_start - starting index of the list
983: lnk_max - max value of lnk indicating the end of the list
984: nlnk - max length of the list
985: Output Parameters:
986: lnk - list initialized
987: lnk_lvl - array of size nlnk for storing levels of lnk
988: bt - PetscBT (bitarray) with all bits set to false
989: */
990: #define PetscIncompleteLLCreate(idx_start,lnk_max,nlnk,lnk,lnk_lvl,bt)\
991: (PetscIntMultError(2,nlnk,NULL) || PetscMalloc1(2*nlnk,&lnk) || PetscBTCreate(nlnk,&(bt)) || (lnk[idx_start] = lnk_max,lnk_lvl = lnk + nlnk,0))
993: /*
994: Initialize a sorted linked list used for ILU and ICC
995: Input Parameters:
996: nidx - number of input idx
997: idx - integer array used for storing column indices
998: idx_start - starting index of the list
999: perm - indices of an IS
1000: lnk - linked list(an integer array) that is created
1001: lnklvl - levels of lnk
1002: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1003: output Parameters:
1004: nlnk - number of newly added idx
1005: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1006: lnklvl - levels of lnk
1007: bt - updated PetscBT (bitarray)
1008: */
1009: #define PetscIncompleteLLInit(nidx,idx,idx_start,perm,nlnk,lnk,lnklvl,bt) 0;\
1010: {\
1011: PetscInt _k,_entry,_location,_lnkdata;\
1012: nlnk = 0;\
1013: _lnkdata = idx_start;\
1014: for (_k=0; _k<nidx; _k++){\
1015: _entry = perm[idx[_k]];\
1016: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
1017: /* search for insertion location */\
1018: if (_k && _entry < _lnkdata) _lnkdata = idx_start;\
1019: do {\
1020: _location = _lnkdata;\
1021: _lnkdata = lnk[_location];\
1022: } while (_entry > _lnkdata);\
1023: /* insertion location is found, add entry into lnk */\
1024: lnk[_location] = _entry;\
1025: lnk[_entry] = _lnkdata;\
1026: lnklvl[_entry] = 0;\
1027: nlnk++;\
1028: _lnkdata = _entry; /* next search starts from here if next_entry > _entry */\
1029: }\
1030: }\
1031: }
1033: /*
1034: Add a SORTED index set into a sorted linked list for ILU
1035: Input Parameters:
1036: nidx - number of input indices
1037: idx - sorted integer array used for storing column indices
1038: level - level of fill, e.g., ICC(level)
1039: idxlvl - level of idx
1040: idx_start - starting index of the list
1041: lnk - linked list(an integer array) that is created
1042: lnklvl - levels of lnk
1043: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1044: prow - the row number of idx
1045: output Parameters:
1046: nlnk - number of newly added idx
1047: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1048: lnklvl - levels of lnk
1049: bt - updated PetscBT (bitarray)
1051: Note: the level of factor(i,j) is set as lvl(i,j) = min{ lvl(i,j), lvl(i,prow)+lvl(prow,j)+1)
1052: where idx = non-zero columns of U(prow,prow+1:n-1), prow<i
1053: */
1054: #define PetscILULLAddSorted(nidx,idx,level,idxlvl,idx_start,nlnk,lnk,lnklvl,bt,lnklvl_prow) 0;\
1055: {\
1056: PetscInt _k,_entry,_location,_lnkdata,_incrlev,_lnklvl_prow=lnklvl[prow];\
1057: nlnk = 0;\
1058: _lnkdata = idx_start;\
1059: for (_k=0; _k<nidx; _k++){\
1060: _incrlev = idxlvl[_k] + _lnklvl_prow + 1;\
1061: if (_incrlev > level) continue;\
1062: _entry = idx[_k];\
1063: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
1064: /* search for insertion location */\
1065: do {\
1066: _location = _lnkdata;\
1067: _lnkdata = lnk[_location];\
1068: } while (_entry > _lnkdata);\
1069: /* insertion location is found, add entry into lnk */\
1070: lnk[_location] = _entry;\
1071: lnk[_entry] = _lnkdata;\
1072: lnklvl[_entry] = _incrlev;\
1073: nlnk++;\
1074: _lnkdata = _entry; /* next search starts from here if next_entry > _entry */\
1075: } else { /* existing entry: update lnklvl */\
1076: if (lnklvl[_entry] > _incrlev) lnklvl[_entry] = _incrlev;\
1077: }\
1078: }\
1079: }
1081: /*
1082: Add a index set into a sorted linked list
1083: Input Parameters:
1084: nidx - number of input idx
1085: idx - integer array used for storing column indices
1086: level - level of fill, e.g., ICC(level)
1087: idxlvl - level of idx
1088: idx_start - starting index of the list
1089: lnk - linked list(an integer array) that is created
1090: lnklvl - levels of lnk
1091: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1092: output Parameters:
1093: nlnk - number of newly added idx
1094: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1095: lnklvl - levels of lnk
1096: bt - updated PetscBT (bitarray)
1097: */
1098: #define PetscIncompleteLLAdd(nidx,idx,level,idxlvl,idx_start,nlnk,lnk,lnklvl,bt) 0;\
1099: {\
1100: PetscInt _k,_entry,_location,_lnkdata,_incrlev;\
1101: nlnk = 0;\
1102: _lnkdata = idx_start;\
1103: for (_k=0; _k<nidx; _k++){\
1104: _incrlev = idxlvl[_k] + 1;\
1105: if (_incrlev > level) continue;\
1106: _entry = idx[_k];\
1107: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
1108: /* search for insertion location */\
1109: if (_k && _entry < _lnkdata) _lnkdata = idx_start;\
1110: do {\
1111: _location = _lnkdata;\
1112: _lnkdata = lnk[_location];\
1113: } while (_entry > _lnkdata);\
1114: /* insertion location is found, add entry into lnk */\
1115: lnk[_location] = _entry;\
1116: lnk[_entry] = _lnkdata;\
1117: lnklvl[_entry] = _incrlev;\
1118: nlnk++;\
1119: _lnkdata = _entry; /* next search starts from here if next_entry > _entry */\
1120: } else { /* existing entry: update lnklvl */\
1121: if (lnklvl[_entry] > _incrlev) lnklvl[_entry] = _incrlev;\
1122: }\
1123: }\
1124: }
1126: /*
1127: Add a SORTED index set into a sorted linked list
1128: Input Parameters:
1129: nidx - number of input indices
1130: idx - sorted integer array used for storing column indices
1131: level - level of fill, e.g., ICC(level)
1132: idxlvl - level of idx
1133: idx_start - starting index of the list
1134: lnk - linked list(an integer array) that is created
1135: lnklvl - levels of lnk
1136: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1137: output Parameters:
1138: nlnk - number of newly added idx
1139: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1140: lnklvl - levels of lnk
1141: bt - updated PetscBT (bitarray)
1142: */
1143: #define PetscIncompleteLLAddSorted(nidx,idx,level,idxlvl,idx_start,nlnk,lnk,lnklvl,bt) 0;\
1144: {\
1145: PetscInt _k,_entry,_location,_lnkdata,_incrlev;\
1146: nlnk = 0;\
1147: _lnkdata = idx_start;\
1148: for (_k=0; _k<nidx; _k++){\
1149: _incrlev = idxlvl[_k] + 1;\
1150: if (_incrlev > level) continue;\
1151: _entry = idx[_k];\
1152: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
1153: /* search for insertion location */\
1154: do {\
1155: _location = _lnkdata;\
1156: _lnkdata = lnk[_location];\
1157: } while (_entry > _lnkdata);\
1158: /* insertion location is found, add entry into lnk */\
1159: lnk[_location] = _entry;\
1160: lnk[_entry] = _lnkdata;\
1161: lnklvl[_entry] = _incrlev;\
1162: nlnk++;\
1163: _lnkdata = _entry; /* next search starts from here */\
1164: } else { /* existing entry: update lnklvl */\
1165: if (lnklvl[_entry] > _incrlev) lnklvl[_entry] = _incrlev;\
1166: }\
1167: }\
1168: }
1170: /*
1171: Add a SORTED index set into a sorted linked list for ICC
1172: Input Parameters:
1173: nidx - number of input indices
1174: idx - sorted integer array used for storing column indices
1175: level - level of fill, e.g., ICC(level)
1176: idxlvl - level of idx
1177: idx_start - starting index of the list
1178: lnk - linked list(an integer array) that is created
1179: lnklvl - levels of lnk
1180: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1181: idxlvl_prow - idxlvl[prow], where prow is the row number of the idx
1182: output Parameters:
1183: nlnk - number of newly added indices
1184: lnk - the sorted(increasing order) linked list containing new and non-redundate entries from idx
1185: lnklvl - levels of lnk
1186: bt - updated PetscBT (bitarray)
1187: Note: the level of U(i,j) is set as lvl(i,j) = min{ lvl(i,j), lvl(prow,i)+lvl(prow,j)+1)
1188: where idx = non-zero columns of U(prow,prow+1:n-1), prow<i
1189: */
1190: #define PetscICCLLAddSorted(nidx,idx,level,idxlvl,idx_start,nlnk,lnk,lnklvl,bt,idxlvl_prow) 0;\
1191: {\
1192: PetscInt _k,_entry,_location,_lnkdata,_incrlev;\
1193: nlnk = 0;\
1194: _lnkdata = idx_start;\
1195: for (_k=0; _k<nidx; _k++){\
1196: _incrlev = idxlvl[_k] + idxlvl_prow + 1;\
1197: if (_incrlev > level) continue;\
1198: _entry = idx[_k];\
1199: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */\
1200: /* search for insertion location */\
1201: do {\
1202: _location = _lnkdata;\
1203: _lnkdata = lnk[_location];\
1204: } while (_entry > _lnkdata);\
1205: /* insertion location is found, add entry into lnk */\
1206: lnk[_location] = _entry;\
1207: lnk[_entry] = _lnkdata;\
1208: lnklvl[_entry] = _incrlev;\
1209: nlnk++;\
1210: _lnkdata = _entry; /* next search starts from here */\
1211: } else { /* existing entry: update lnklvl */\
1212: if (lnklvl[_entry] > _incrlev) lnklvl[_entry] = _incrlev;\
1213: }\
1214: }\
1215: }
1217: /*
1218: Copy data on the list into an array, then initialize the list
1219: Input Parameters:
1220: idx_start - starting index of the list
1221: lnk_max - max value of lnk indicating the end of the list
1222: nlnk - number of data on the list to be copied
1223: lnk - linked list
1224: lnklvl - level of lnk
1225: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1226: output Parameters:
1227: indices - array that contains the copied data
1228: lnk - linked list that is cleaned and initialize
1229: lnklvl - level of lnk that is reinitialized
1230: bt - PetscBT (bitarray) with all bits set to false
1231: */
1232: #define PetscIncompleteLLClean(idx_start,lnk_max,nlnk,lnk,lnklvl,indices,indiceslvl,bt) 0;\
1233: {\
1234: PetscInt _j,_idx=idx_start;\
1235: for (_j=0; _j<nlnk; _j++){\
1236: _idx = lnk[_idx];\
1237: *(indices+_j) = _idx;\
1238: *(indiceslvl+_j) = lnklvl[_idx];\
1239: lnklvl[_idx] = -1;\
1240: PetscBTClear(bt,_idx);\
1241: }\
1242: lnk[idx_start] = lnk_max;\
1243: }
1244: /*
1245: Free memories used by the list
1246: */
1247: #define PetscIncompleteLLDestroy(lnk,bt) (PetscFree(lnk) || PetscBTDestroy(&(bt)))
1249: /* -------------------------------------------------------------------------------------------------------*/
1250: #include <petscbt.h>
1251: /*
1252: Create and initialize a condensed linked list -
1253: same as PetscLLCreate(), but uses a scalable array 'lnk' with size of max number of entries, not O(N).
1254: Barry suggested this approach (Dec. 6, 2011):
1255: I've thought of an alternative way of representing a linked list that is efficient but doesn't have the O(N) scaling issue
1256: (it may be faster than the O(N) even sequentially due to less crazy memory access).
1258: 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
1259: 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
1260: 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
1261: 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.
1262: 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
1263: to each other so memory access is much better than using the big array.
1265: Example:
1266: nlnk_max=5, lnk_max=36:
1267: Initial list: [0, 0 | 36, 2 | 0, 0 | 0, 0 | 0, 0 | 0, 0 | 0, 0]
1268: here, head_node has index 2 with value lnk[2]=lnk_max=36,
1269: 0-th entry is used to store the number of entries in the list,
1270: The initial lnk represents head -> tail(marked by 36) with number of entries = lnk[0]=0.
1272: Now adding a sorted set {2,4}, the list becomes
1273: [2, 0 | 36, 4 |2, 6 | 4, 2 | 0, 0 | 0, 0 | 0, 0 ]
1274: represents head -> 2 -> 4 -> tail with number of entries = lnk[0]=2.
1276: Then adding a sorted set {0,3,35}, the list
1277: [5, 0 | 36, 8 | 2, 10 | 4, 12 | 0, 4 | 3, 6 | 35, 2 ]
1278: represents head -> 0 -> 2 -> 3 -> 4 -> 35 -> tail with number of entries = lnk[0]=5.
1280: Input Parameters:
1281: nlnk_max - max length of the list
1282: lnk_max - max value of the entries
1283: Output Parameters:
1284: lnk - list created and initialized
1285: bt - PetscBT (bitarray) with all bits set to false. Note: bt has size lnk_max, not nln_max!
1286: */
1287: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedCreate(PetscInt nlnk_max,PetscInt lnk_max,PetscInt **lnk,PetscBT *bt)
1288: {
1290: PetscInt *llnk,lsize = 0;
1293: PetscIntMultError(2,nlnk_max+2,&lsize);
1294: PetscMalloc1(lsize,lnk);
1295: PetscBTCreate(lnk_max,bt);
1296: llnk = *lnk;
1297: llnk[0] = 0; /* number of entries on the list */
1298: llnk[2] = lnk_max; /* value in the head node */
1299: llnk[3] = 2; /* next for the head node */
1300: return(0);
1301: }
1303: /*
1304: Add a SORTED ascending index set into a sorted linked list. See PetscLLCondensedCreate() for detailed description.
1305: Input Parameters:
1306: nidx - number of input indices
1307: indices - sorted integer array
1308: lnk - condensed linked list(an integer array) that is created
1309: bt - PetscBT (bitarray), bt[idx]=true marks idx is in lnk
1310: output Parameters:
1311: lnk - the sorted(increasing order) linked list containing previous and newly added non-redundate indices
1312: bt - updated PetscBT (bitarray)
1313: */
1314: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedAddSorted(PetscInt nidx,const PetscInt indices[],PetscInt lnk[],PetscBT bt)
1315: {
1316: PetscInt _k,_entry,_location,_next,_lnkdata,_nlnk,_newnode;
1319: _nlnk = lnk[0]; /* num of entries on the input lnk */
1320: _location = 2; /* head */
1321: for (_k=0; _k<nidx; _k++){
1322: _entry = indices[_k];
1323: if (!PetscBTLookupSet(bt,_entry)){ /* new entry */
1324: /* search for insertion location */
1325: do {
1326: _next = _location + 1; /* link from previous node to next node */
1327: _location = lnk[_next]; /* idx of next node */
1328: _lnkdata = lnk[_location];/* value of next node */
1329: } while (_entry > _lnkdata);
1330: /* insertion location is found, add entry into lnk */
1331: _newnode = 2*(_nlnk+2); /* index for this new node */
1332: lnk[_next] = _newnode; /* connect previous node to the new node */
1333: lnk[_newnode] = _entry; /* set value of the new node */
1334: lnk[_newnode+1] = _location; /* connect new node to next node */
1335: _location = _newnode; /* next search starts from the new node */
1336: _nlnk++;
1337: } \
1338: }\
1339: lnk[0] = _nlnk; /* number of entries in the list */
1340: return(0);
1341: }
1343: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedClean(PetscInt lnk_max,PetscInt nidx,PetscInt *indices,PetscInt lnk[],PetscBT bt)
1344: {
1346: PetscInt _k,_next,_nlnk;
1349: _next = lnk[3]; /* head node */
1350: _nlnk = lnk[0]; /* num of entries on the list */
1351: for (_k=0; _k<_nlnk; _k++){
1352: indices[_k] = lnk[_next];
1353: _next = lnk[_next + 1];
1354: PetscBTClear(bt,indices[_k]);
1355: }
1356: lnk[0] = 0; /* num of entries on the list */
1357: lnk[2] = lnk_max; /* initialize head node */
1358: lnk[3] = 2; /* head node */
1359: return(0);
1360: }
1362: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedView(PetscInt *lnk)
1363: {
1365: PetscInt k;
1368: PetscPrintf(PETSC_COMM_SELF,"LLCondensed of size %D, (val, next)\n",lnk[0]);
1369: for (k=2; k< lnk[0]+2; k++){
1370: PetscPrintf(PETSC_COMM_SELF," %D: (%D, %D)\n",2*k,lnk[2*k],lnk[2*k+1]);
1371: }
1372: return(0);
1373: }
1375: /*
1376: Free memories used by the list
1377: */
1378: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedDestroy(PetscInt *lnk,PetscBT bt)
1379: {
1383: PetscFree(lnk);
1384: PetscBTDestroy(&bt);
1385: return(0);
1386: }
1388: /* -------------------------------------------------------------------------------------------------------*/
1389: /*
1390: Same as PetscLLCondensedCreate(), but does not use non-scalable O(lnk_max) bitarray
1391: Input Parameters:
1392: nlnk_max - max length of the list
1393: Output Parameters:
1394: lnk - list created and initialized
1395: */
1396: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedCreate_Scalable(PetscInt nlnk_max,PetscInt **lnk)
1397: {
1399: PetscInt *llnk,lsize = 0;
1402: PetscIntMultError(2,nlnk_max+2,&lsize);
1403: PetscMalloc1(lsize,lnk);
1404: llnk = *lnk;
1405: llnk[0] = 0; /* number of entries on the list */
1406: llnk[2] = PETSC_MAX_INT; /* value in the head node */
1407: llnk[3] = 2; /* next for the head node */
1408: return(0);
1409: }
1411: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedAddSorted_Scalable(PetscInt nidx,const PetscInt indices[],PetscInt lnk[])
1412: {
1413: PetscInt _k,_entry,_location,_next,_lnkdata,_nlnk,_newnode;
1414: _nlnk = lnk[0]; /* num of entries on the input lnk */
1415: _location = 2; /* head */ \
1416: for (_k=0; _k<nidx; _k++){
1417: _entry = indices[_k];
1418: /* search for insertion location */
1419: do {
1420: _next = _location + 1; /* link from previous node to next node */
1421: _location = lnk[_next]; /* idx of next node */
1422: _lnkdata = lnk[_location];/* value of next node */
1423: } while (_entry > _lnkdata);
1424: if (_entry < _lnkdata) {
1425: /* insertion location is found, add entry into lnk */
1426: _newnode = 2*(_nlnk+2); /* index for this new node */
1427: lnk[_next] = _newnode; /* connect previous node to the new node */
1428: lnk[_newnode] = _entry; /* set value of the new node */
1429: lnk[_newnode+1] = _location; /* connect new node to next node */
1430: _location = _newnode; /* next search starts from the new node */
1431: _nlnk++;
1432: }
1433: }
1434: lnk[0] = _nlnk; /* number of entries in the list */
1435: return 0;
1436: }
1438: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedClean_Scalable(PetscInt nidx,PetscInt *indices,PetscInt *lnk)
1439: {
1440: PetscInt _k,_next,_nlnk;
1441: _next = lnk[3]; /* head node */
1442: _nlnk = lnk[0];
1443: for (_k=0; _k<_nlnk; _k++){
1444: indices[_k] = lnk[_next];
1445: _next = lnk[_next + 1];
1446: }
1447: lnk[0] = 0; /* num of entries on the list */
1448: lnk[3] = 2; /* head node */
1449: return 0;
1450: }
1452: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedDestroy_Scalable(PetscInt *lnk)
1453: {
1454: return PetscFree(lnk);
1455: }
1457: /* -------------------------------------------------------------------------------------------------------*/
1458: /*
1459: lnk[0] number of links
1460: lnk[1] number of entries
1461: lnk[3n] value
1462: lnk[3n+1] len
1463: lnk[3n+2] link to next value
1465: The next three are always the first link
1467: lnk[3] PETSC_MIN_INT+1
1468: lnk[4] 1
1469: lnk[5] link to first real entry
1471: The next three are always the last link
1473: lnk[6] PETSC_MAX_INT - 1
1474: lnk[7] 1
1475: lnk[8] next valid link (this is the same as lnk[0] but without the decreases)
1476: */
1478: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedCreate_fast(PetscInt nlnk_max,PetscInt **lnk)
1479: {
1481: PetscInt *llnk,lsize = 0;
1484: PetscIntMultError(3,nlnk_max+3,&lsize);
1485: PetscMalloc1(lsize,lnk);
1486: llnk = *lnk;
1487: llnk[0] = 0; /* nlnk: number of entries on the list */
1488: llnk[1] = 0; /* number of integer entries represented in list */
1489: llnk[3] = PETSC_MIN_INT+1; /* value in the first node */
1490: llnk[4] = 1; /* count for the first node */
1491: llnk[5] = 6; /* next for the first node */
1492: llnk[6] = PETSC_MAX_INT-1; /* value in the last node */
1493: llnk[7] = 1; /* count for the last node */
1494: llnk[8] = 0; /* next valid node to be used */
1495: return(0);
1496: }
1498: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedAddSorted_fast(PetscInt nidx,const PetscInt indices[],PetscInt lnk[])
1499: {
1500: PetscInt k,entry,prev,next;
1501: prev = 3; /* first value */
1502: next = lnk[prev+2];
1503: for (k=0; k<nidx; k++){
1504: entry = indices[k];
1505: /* search for insertion location */
1506: while (entry >= lnk[next]) {
1507: prev = next;
1508: next = lnk[next+2];
1509: }
1510: /* entry is in range of previous list */
1511: if (entry < lnk[prev]+lnk[prev+1]) continue;
1512: lnk[1]++;
1513: /* entry is right after previous list */
1514: if (entry == lnk[prev]+lnk[prev+1]) {
1515: lnk[prev+1]++;
1516: if (lnk[next] == entry+1) { /* combine two contiguous strings */
1517: lnk[prev+1] += lnk[next+1];
1518: lnk[prev+2] = lnk[next+2];
1519: next = lnk[next+2];
1520: lnk[0]--;
1521: }
1522: continue;
1523: }
1524: /* entry is right before next list */
1525: if (entry == lnk[next]-1) {
1526: lnk[next]--;
1527: lnk[next+1]++;
1528: prev = next;
1529: next = lnk[prev+2];
1530: continue;
1531: }
1532: /* add entry into lnk */
1533: lnk[prev+2] = 3*((lnk[8]++)+3); /* connect previous node to the new node */
1534: prev = lnk[prev+2];
1535: lnk[prev] = entry; /* set value of the new node */
1536: lnk[prev+1] = 1; /* number of values in contiguous string is one to start */
1537: lnk[prev+2] = next; /* connect new node to next node */
1538: lnk[0]++;
1539: }
1540: return 0;
1541: }
1543: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedClean_fast(PetscInt nidx,PetscInt *indices,PetscInt *lnk)
1544: {
1545: PetscInt _k,_next,_nlnk,cnt,j;
1546: _next = lnk[5]; /* first node */
1547: _nlnk = lnk[0];
1548: cnt = 0;
1549: for (_k=0; _k<_nlnk; _k++){
1550: for (j=0; j<lnk[_next+1]; j++) {
1551: indices[cnt++] = lnk[_next] + j;
1552: }
1553: _next = lnk[_next + 2];
1554: }
1555: lnk[0] = 0; /* nlnk: number of links */
1556: lnk[1] = 0; /* number of integer entries represented in list */
1557: lnk[3] = PETSC_MIN_INT+1; /* value in the first node */
1558: lnk[4] = 1; /* count for the first node */
1559: lnk[5] = 6; /* next for the first node */
1560: lnk[6] = PETSC_MAX_INT-1; /* value in the last node */
1561: lnk[7] = 1; /* count for the last node */
1562: lnk[8] = 0; /* next valid location to make link */
1563: return 0;
1564: }
1566: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedView_fast(PetscInt *lnk)
1567: {
1568: PetscInt k,next,nlnk;
1569: next = lnk[5]; /* first node */
1570: nlnk = lnk[0];
1571: for (k=0; k<nlnk; k++){
1572: #if 0 /* Debugging code */
1573: printf("%d value %d len %d next %d\n",next,lnk[next],lnk[next+1],lnk[next+2]);
1574: #endif
1575: next = lnk[next + 2];
1576: }
1577: return 0;
1578: }
1580: PETSC_STATIC_INLINE PetscErrorCode PetscLLCondensedDestroy_fast(PetscInt *lnk)
1581: {
1582: return PetscFree(lnk);
1583: }
1585: /* this is extern because it is used in MatFDColoringUseDM() which is in the DM library */
1586: PETSC_EXTERN PetscErrorCode MatFDColoringApply_AIJ(Mat,MatFDColoring,Vec,void*);
1588: PETSC_EXTERN PetscLogEvent MAT_Mult;
1589: PETSC_EXTERN PetscLogEvent MAT_MultMatrixFree;
1590: PETSC_EXTERN PetscLogEvent MAT_Mults;
1591: PETSC_EXTERN PetscLogEvent MAT_MultConstrained;
1592: PETSC_EXTERN PetscLogEvent MAT_MultAdd;
1593: PETSC_EXTERN PetscLogEvent MAT_MultTranspose;
1594: PETSC_EXTERN PetscLogEvent MAT_MultTransposeConstrained;
1595: PETSC_EXTERN PetscLogEvent MAT_MultTransposeAdd;
1596: PETSC_EXTERN PetscLogEvent MAT_Solve;
1597: PETSC_EXTERN PetscLogEvent MAT_Solves;
1598: PETSC_EXTERN PetscLogEvent MAT_SolveAdd;
1599: PETSC_EXTERN PetscLogEvent MAT_SolveTranspose;
1600: PETSC_EXTERN PetscLogEvent MAT_SolveTransposeAdd;
1601: PETSC_EXTERN PetscLogEvent MAT_SOR;
1602: PETSC_EXTERN PetscLogEvent MAT_ForwardSolve;
1603: PETSC_EXTERN PetscLogEvent MAT_BackwardSolve;
1604: PETSC_EXTERN PetscLogEvent MAT_LUFactor;
1605: PETSC_EXTERN PetscLogEvent MAT_LUFactorSymbolic;
1606: PETSC_EXTERN PetscLogEvent MAT_LUFactorNumeric;
1607: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactor;
1608: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactorSymbolic;
1609: PETSC_EXTERN PetscLogEvent MAT_CholeskyFactorNumeric;
1610: PETSC_EXTERN PetscLogEvent MAT_ILUFactor;
1611: PETSC_EXTERN PetscLogEvent MAT_ILUFactorSymbolic;
1612: PETSC_EXTERN PetscLogEvent MAT_ICCFactorSymbolic;
1613: PETSC_EXTERN PetscLogEvent MAT_Copy;
1614: PETSC_EXTERN PetscLogEvent MAT_Convert;
1615: PETSC_EXTERN PetscLogEvent MAT_Scale;
1616: PETSC_EXTERN PetscLogEvent MAT_AssemblyBegin;
1617: PETSC_EXTERN PetscLogEvent MAT_AssemblyEnd;
1618: PETSC_EXTERN PetscLogEvent MAT_SetValues;
1619: PETSC_EXTERN PetscLogEvent MAT_GetValues;
1620: PETSC_EXTERN PetscLogEvent MAT_GetRow;
1621: PETSC_EXTERN PetscLogEvent MAT_GetRowIJ;
1622: PETSC_EXTERN PetscLogEvent MAT_CreateSubMats;
1623: PETSC_EXTERN PetscLogEvent MAT_GetColoring;
1624: PETSC_EXTERN PetscLogEvent MAT_GetOrdering;
1625: PETSC_EXTERN PetscLogEvent MAT_RedundantMat;
1626: PETSC_EXTERN PetscLogEvent MAT_IncreaseOverlap;
1627: PETSC_EXTERN PetscLogEvent MAT_Partitioning;
1628: PETSC_EXTERN PetscLogEvent MAT_Coarsen;
1629: PETSC_EXTERN PetscLogEvent MAT_ZeroEntries;
1630: PETSC_EXTERN PetscLogEvent MAT_Load;
1631: PETSC_EXTERN PetscLogEvent MAT_View;
1632: PETSC_EXTERN PetscLogEvent MAT_AXPY;
1633: PETSC_EXTERN PetscLogEvent MAT_FDColoringCreate;
1634: PETSC_EXTERN PetscLogEvent MAT_TransposeColoringCreate;
1635: PETSC_EXTERN PetscLogEvent MAT_FDColoringSetUp;
1636: PETSC_EXTERN PetscLogEvent MAT_FDColoringApply;
1637: PETSC_EXTERN PetscLogEvent MAT_Transpose;
1638: PETSC_EXTERN PetscLogEvent MAT_FDColoringFunction;
1639: PETSC_EXTERN PetscLogEvent MAT_CreateSubMat;
1640: PETSC_EXTERN PetscLogEvent MAT_MatMult;
1641: PETSC_EXTERN PetscLogEvent MAT_MatSolve;
1642: PETSC_EXTERN PetscLogEvent MAT_MatMultSymbolic;
1643: PETSC_EXTERN PetscLogEvent MAT_MatMultNumeric;
1644: PETSC_EXTERN PetscLogEvent MAT_Getlocalmatcondensed;
1645: PETSC_EXTERN PetscLogEvent MAT_GetBrowsOfAcols;
1646: PETSC_EXTERN PetscLogEvent MAT_GetBrowsOfAocols;
1647: PETSC_EXTERN PetscLogEvent MAT_PtAP;
1648: PETSC_EXTERN PetscLogEvent MAT_PtAPSymbolic;
1649: PETSC_EXTERN PetscLogEvent MAT_PtAPNumeric;
1650: PETSC_EXTERN PetscLogEvent MAT_Seqstompinum;
1651: PETSC_EXTERN PetscLogEvent MAT_Seqstompisym;
1652: PETSC_EXTERN PetscLogEvent MAT_Seqstompi;
1653: PETSC_EXTERN PetscLogEvent MAT_Getlocalmat;
1654: PETSC_EXTERN PetscLogEvent MAT_RARt;
1655: PETSC_EXTERN PetscLogEvent MAT_RARtSymbolic;
1656: PETSC_EXTERN PetscLogEvent MAT_RARtNumeric;
1657: PETSC_EXTERN PetscLogEvent MAT_MatTransposeMult;
1658: PETSC_EXTERN PetscLogEvent MAT_MatTransposeMultSymbolic;
1659: PETSC_EXTERN PetscLogEvent MAT_MatTransposeMultNumeric;
1660: PETSC_EXTERN PetscLogEvent MAT_TransposeMatMult;
1661: PETSC_EXTERN PetscLogEvent MAT_TransposeMatMultSymbolic;
1662: PETSC_EXTERN PetscLogEvent MAT_TransposeMatMultNumeric;
1663: PETSC_EXTERN PetscLogEvent MAT_MatMatMult;
1664: PETSC_EXTERN PetscLogEvent MAT_MatMatMultSymbolic;
1665: PETSC_EXTERN PetscLogEvent MAT_MatMatMultNumeric;
1666: PETSC_EXTERN PetscLogEvent MAT_Applypapt;
1667: PETSC_EXTERN PetscLogEvent MAT_Applypapt_symbolic;
1668: PETSC_EXTERN PetscLogEvent MAT_Applypapt_numeric;
1669: PETSC_EXTERN PetscLogEvent MAT_Getsymtranspose;
1670: PETSC_EXTERN PetscLogEvent MAT_Transpose_SeqAIJ;
1671: PETSC_EXTERN PetscLogEvent MAT_Getsymtransreduced;
1672: PETSC_EXTERN PetscLogEvent MAT_GetSequentialNonzeroStructure;
1673: PETSC_EXTERN PetscLogEvent MATMFFD_Mult;
1674: PETSC_EXTERN PetscLogEvent MAT_GetMultiProcBlock;
1675: PETSC_EXTERN PetscLogEvent MAT_CUSPARSECopyToGPU;
1676: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatch;
1677: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatchI;
1678: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatchII;
1679: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatchIII;
1680: PETSC_EXTERN PetscLogEvent MAT_SetValuesBatchIV;
1681: PETSC_EXTERN PetscLogEvent MAT_ViennaCLCopyToGPU;
1682: PETSC_EXTERN PetscLogEvent MAT_Merge;
1683: PETSC_EXTERN PetscLogEvent MAT_Residual;
1684: PETSC_EXTERN PetscLogEvent MAT_SetRandom;
1685: PETSC_EXTERN PetscLogEvent MATCOLORING_Apply;
1686: PETSC_EXTERN PetscLogEvent MATCOLORING_Comm;
1687: PETSC_EXTERN PetscLogEvent MATCOLORING_Local;
1688: PETSC_EXTERN PetscLogEvent MATCOLORING_ISCreate;
1689: PETSC_EXTERN PetscLogEvent MATCOLORING_SetUp;
1690: PETSC_EXTERN PetscLogEvent MATCOLORING_Weights;
1692: #endif