1 #define PETSC_SKIP_IMMINTRIN_H_CUDAWORKAROUND 1 2 3 #include <petscconf.h> 4 #include <../src/mat/impls/aij/mpi/mpiaij.h> /*I "petscmat.h" I*/ 5 #include <../src/mat/impls/aij/seq/seqviennacl/viennaclmatimpl.h> 6 7 PetscErrorCode MatMPIAIJSetPreallocation_MPIAIJViennaCL(Mat B, PetscInt d_nz, const PetscInt d_nnz[], PetscInt o_nz, const PetscInt o_nnz[]) { 8 Mat_MPIAIJ *b = (Mat_MPIAIJ *)B->data; 9 10 PetscFunctionBegin; 11 PetscCall(PetscLayoutSetUp(B->rmap)); 12 PetscCall(PetscLayoutSetUp(B->cmap)); 13 if (!B->preallocated) { 14 /* Explicitly create the two MATSEQAIJVIENNACL matrices. */ 15 PetscCall(MatCreate(PETSC_COMM_SELF, &b->A)); 16 PetscCall(MatSetSizes(b->A, B->rmap->n, B->cmap->n, B->rmap->n, B->cmap->n)); 17 PetscCall(MatSetType(b->A, MATSEQAIJVIENNACL)); 18 PetscCall(MatCreate(PETSC_COMM_SELF, &b->B)); 19 PetscCall(MatSetSizes(b->B, B->rmap->n, B->cmap->N, B->rmap->n, B->cmap->N)); 20 PetscCall(MatSetType(b->B, MATSEQAIJVIENNACL)); 21 } 22 PetscCall(MatSeqAIJSetPreallocation(b->A, d_nz, d_nnz)); 23 PetscCall(MatSeqAIJSetPreallocation(b->B, o_nz, o_nnz)); 24 B->preallocated = PETSC_TRUE; 25 PetscFunctionReturn(0); 26 } 27 28 PetscErrorCode MatAssemblyEnd_MPIAIJViennaCL(Mat A, MatAssemblyType mode) { 29 Mat_MPIAIJ *b = (Mat_MPIAIJ *)A->data; 30 PetscBool v; 31 32 PetscFunctionBegin; 33 PetscCall(MatAssemblyEnd_MPIAIJ(A, mode)); 34 PetscCall(PetscObjectTypeCompare((PetscObject)b->lvec, VECSEQVIENNACL, &v)); 35 if (!v) { 36 PetscInt m; 37 PetscCall(VecGetSize(b->lvec, &m)); 38 PetscCall(VecDestroy(&b->lvec)); 39 PetscCall(VecCreateSeqViennaCL(PETSC_COMM_SELF, m, &b->lvec)); 40 } 41 PetscFunctionReturn(0); 42 } 43 44 PetscErrorCode MatDestroy_MPIAIJViennaCL(Mat A) { 45 PetscFunctionBegin; 46 PetscCall(MatDestroy_MPIAIJ(A)); 47 PetscFunctionReturn(0); 48 } 49 50 PETSC_EXTERN PetscErrorCode MatCreate_MPIAIJViennaCL(Mat A) { 51 PetscFunctionBegin; 52 PetscCall(MatCreate_MPIAIJ(A)); 53 A->boundtocpu = PETSC_FALSE; 54 PetscCall(PetscFree(A->defaultvectype)); 55 PetscCall(PetscStrallocpy(VECVIENNACL, &A->defaultvectype)); 56 PetscCall(PetscObjectComposeFunction((PetscObject)A, "MatMPIAIJSetPreallocation_C", MatMPIAIJSetPreallocation_MPIAIJViennaCL)); 57 A->ops->assemblyend = MatAssemblyEnd_MPIAIJViennaCL; 58 PetscCall(PetscObjectChangeTypeName((PetscObject)A, MATMPIAIJVIENNACL)); 59 PetscFunctionReturn(0); 60 } 61 62 /*@C 63 MatCreateAIJViennaCL - Creates a sparse matrix in `MATAIJ` (compressed row) format 64 (the default parallel PETSc format). This matrix will ultimately be pushed down 65 to GPUs and use the ViennaCL library for calculations. For good matrix 66 assembly performance the user should preallocate the matrix storage by setting 67 the parameter nz (or the array nnz). By setting these parameters accurately, 68 performance during matrix assembly can be increased substantially. 69 70 Collective 71 72 Input Parameters: 73 + comm - MPI communicator, set to `PETSC_COMM_SELF` 74 . m - number of rows 75 . n - number of columns 76 . nz - number of nonzeros per row (same for all rows) 77 - nnz - array containing the number of nonzeros in the various rows 78 (possibly different for each row) or NULL 79 80 Output Parameter: 81 . A - the matrix 82 83 It is recommended that one use the `MatCreate()`, `MatSetType()` and/or `MatSetFromOptions()`, 84 MatXXXXSetPreallocation() paradigm instead of this routine directly. 85 [MatXXXXSetPreallocation() is, for example, `MatSeqAIJSetPreallocation()`] 86 87 Notes: 88 If nnz is given then nz is ignored 89 90 The AIJ format, also called 91 compressed row storage), is fully compatible with standard Fortran 77 92 storage. That is, the stored row and column indices can begin at 93 either one (as in Fortran) or zero. See the users' manual for details. 94 95 Specify the preallocated storage with either nz or nnz (not both). 96 Set nz = `PETSC_DEFAULT` and nnz = NULL for PETSc to control dynamic memory 97 allocation. For large problems you MUST preallocate memory or you 98 will get TERRIBLE performance, see the users' manual chapter on matrices. 99 100 Level: intermediate 101 102 .seealso: `MatCreate()`, `MatCreateAIJ()`, `MatCreateAIJCUSPARSE()`, `MatSetValues()`, `MatSeqAIJSetColumnIndices()`, `MatCreateSeqAIJWithArrays()`, `MatCreateAIJ()`, `MATMPIAIJVIENNACL`, `MATAIJVIENNACL` 103 @*/ 104 PetscErrorCode MatCreateAIJViennaCL(MPI_Comm comm, PetscInt m, PetscInt n, PetscInt M, PetscInt N, PetscInt d_nz, const PetscInt d_nnz[], PetscInt o_nz, const PetscInt o_nnz[], Mat *A) { 105 PetscMPIInt size; 106 107 PetscFunctionBegin; 108 PetscCall(MatCreate(comm, A)); 109 PetscCall(MatSetSizes(*A, m, n, M, N)); 110 PetscCallMPI(MPI_Comm_size(comm, &size)); 111 if (size > 1) { 112 PetscCall(MatSetType(*A, MATMPIAIJVIENNACL)); 113 PetscCall(MatMPIAIJSetPreallocation(*A, d_nz, d_nnz, o_nz, o_nnz)); 114 } else { 115 PetscCall(MatSetType(*A, MATSEQAIJVIENNACL)); 116 PetscCall(MatSeqAIJSetPreallocation(*A, d_nz, d_nnz)); 117 } 118 PetscFunctionReturn(0); 119 } 120 121 /*MC 122 MATAIJVIENNACL - MATMPIAIJVIENNACL= "aijviennacl" = "mpiaijviennacl" - A matrix type to be used for sparse matrices. 123 124 A matrix type (CSR format) whose data resides on GPUs. 125 All matrix calculations are performed using the ViennaCL library. 126 127 This matrix type is identical to `MATSEQAIJVIENNACL` when constructed with a single process communicator, 128 and `MATMPIAIJVIENNACL` otherwise. As a result, for single process communicators, 129 `MatSeqAIJSetPreallocation()` is supported, and similarly `MatMPIAIJSetPreallocation()` is supported 130 for communicators controlling multiple processes. It is recommended that you call both of 131 the above preallocation routines for simplicity. 132 133 Options Database Keys: 134 . -mat_type mpiaijviennacl - sets the matrix type to `MATAIJVIENNACL` during a call to `MatSetFromOptions()` 135 136 Level: beginner 137 138 .seealso: `MatCreateAIJViennaCL()`, `MATSEQAIJVIENNACL`, `MatCreateSeqAIJVIENNACL()` 139 M*/ 140