Home
last modified time | relevance | path

Searched refs:PetscLogGpuFlops (Results 1 – 22 of 22) sorted by relevance

/petsc/src/vec/vec/impls/seq/seqviennacl/
H A Dvecviennacl.cxx329 PetscCall(PetscLogGpuFlops(2.0 * yin->map->n)); in VecAYPX_SeqViennaCL()
362 PetscCall(PetscLogGpuFlops(2.0 * yin->map->n)); in VecAXPY_SeqViennaCL()
385 PetscCall(PetscLogGpuFlops(win->map->n)); in VecPointwiseDivide_SeqViennaCL()
412 PetscCall(PetscLogGpuFlops(win->map->n)); in VecWAXPY_SeqViennaCL()
419 PetscCall(PetscLogGpuFlops(win->map->n)); in VecWAXPY_SeqViennaCL()
426 PetscCall(PetscLogGpuFlops(2 * win->map->n)); in VecWAXPY_SeqViennaCL()
476 if (xin->map->n > 0) PetscCall(PetscLogGpuFlops(2.0 * xin->map->n - 1)); in VecDot_SeqViennaCL()
510 PetscCall(PetscLogGpuFlops(PetscMax(nv * (2.0 * n - 1), 0.0))); in VecMDot_SeqViennaCL()
553 PetscCall(PetscLogGpuFlops(xin->map->n)); in VecScale_SeqViennaCL()
565 PetscCall(PetscLogGpuFlops(xin->map->n)); in VecScale_SeqViennaCL()
[all …]
/petsc/src/ksp/ksp/utils/lmvm/blas_cyclic/cupm/
H A Dblas_cyclic_cupm_impl.hpp82 PetscCall(PetscLogGpuFlops(3.0 * N)); in axpby()
124 PetscCall(PetscLogGpuFlops(3.0 * N)); in dmv()
172 PetscCall(PetscLogGpuFlops(3.0 * N)); in dsv()
215 PetscCall(PetscLogGpuFlops(1.0 * N * N)); in trsv()
258 PetscCall(PetscLogGpuFlops(2.0 * N * N)); in hemv()
309 PetscCall(PetscLogGpuFlops(2.0 * N * N)); in gemv()
/petsc/src/vec/vec/impls/seq/kokkos/
H A Dveckok.kokkos.cxx220 PetscCall(PetscLogGpuFlops(xin->map->n)); in VecShift_SeqKokkos()
247 PetscCall(PetscLogGpuFlops(2.0 * yin->map->n)); in VecAXPY_SeqKokkos()
278 if (xin->map->n > 0) PetscCall(PetscLogGpuFlops(2.0 * xin->map->n)); in VecTDot_SeqKokkos()
516 PetscCall(PetscLogGpuFlops(PetscMax(nv * (2.0 * xin->map->n - 1), 0.0))); in VecMDot_SeqKokkos()
525 PetscCall(PetscLogGpuFlops(PetscMax(nv * (2.0 * xin->map->n - 1), 0.0))); in VecMTDot_SeqKokkos()
583 PetscCall(PetscLogGpuFlops(PetscMax(m * (2.0 * n - 1), 0.0))); in VecMultiDot_SeqKokkos_GEMV()
646 PetscCall(PetscLogGpuFlops(xin->map->n)); in VecScale_SeqKokkos()
665 PetscCall(PetscLogGpuFlops(PetscMax(2.0 * xin->map->n - 1, 0.0))); in VecDot_SeqKokkos()
748 PetscCall(PetscLogGpuFlops(2.0 * win->map->n)); in VecWAXPY_SeqKokkos()
832 PetscCall(PetscLogGpuFlops(nv * 2.0 * yin->map->n)); in VecMAXPY_SeqKokkos()
[all …]
/petsc/src/ksp/ksp/utils/lmvm/dense/cd_cupm/
H A Dcd_cupm.cxx41 PetscCall(PetscLogGpuFlops(1.0 * N * N)); in SolveInPlace()
79 PetscCall(PetscLogGpuFlops(1.0 * N * N)); in SolveInPlaceCyclic()
/petsc/src/ksp/pc/impls/vpbjacobi/kokkos/
H A Dvpbjacobi_kok.kokkos.cxx129 PetscCall(PetscLogGpuFlops(pckok->nsize * 2)); /* FMA on entries in all blocks */ in PCApplyOrTranspose_VPBJacobi_Kokkos()
198 PetscCall(PetscLogGpuFlops(pckok->setupFlops)); in PCSetUp_VPBJacobi_Kokkos()
/petsc/src/vec/vec/impls/seq/cupm/
H A Dvecseqcupm_impl.hpp207 PetscCall(PetscLogGpuFlops(n)); in PointwiseBinary_()
263 PetscCall(PetscLogGpuFlops(n)); in PointwiseUnary_()
486 PetscCall(PetscLogGpuFlops((alphaIsOne ? 1 : 2) * n)); in AYPXAsync()
518 PetscCall(PetscLogGpuFlops(2 * n)); in AXPYAsync()
838 PetscCall(PetscLogGpuFlops(2 * n)); in WAXPYAsync()
991 PetscCall(PetscLogGpuFlops(nv * 2 * n)); in MAXPYAsync()
1025 PetscCall(PetscLogGpuFlops(2 * n - 1)); in Dot()
1257 PetscCall(PetscLogGpuFlops(nwork)); in MDot_()
1315 PetscCall(PetscLogGpuFlops(nv * (2 * n - 1))); in MDot()
1373 PetscCall(PetscLogGpuFlops(n)); in ScaleAsync()
[all …]
/petsc/src/ksp/pc/impls/pbjacobi/kokkos/
H A Dpbjacobi_kok.kokkos.cxx67 PetscCall(PetscLogGpuFlops(bs * bs * mbs * 2)); /* FMA on entries in all blocks */ in PCApplyOrTranspose_PBJacobi_Kokkos()
/petsc/src/ksp/pc/impls/pbjacobi/cuda/
H A Dpbjacobi_cuda.cu59 PetscCall(PetscLogGpuFlops(bs * bs * mbs * 2)); in PCApplyOrTranspose_PBJacobi_CUDA()
/petsc/src/mat/impls/aij/seq/kokkos/
H A Daijkok.kokkos.cxx402 PetscCall(PetscLogGpuFlops(2.0 * aijkok->csrmat.nnz())); in MatMult_SeqAIJKokkos()
432 PetscCall(PetscLogGpuFlops(2.0 * csrmat.nnz())); in MatMultTranspose_SeqAIJKokkos()
462 PetscCall(PetscLogGpuFlops(2.0 * csrmat.nnz())); in MatMultHermitianTranspose_SeqAIJKokkos()
484 PetscCall(PetscLogGpuFlops(2.0 * aijkok->csrmat.nnz())); in MatMultAdd_SeqAIJKokkos()
515 PetscCall(PetscLogGpuFlops(2.0 * csrmat.nnz())); in MatMultTransposeAdd_SeqAIJKokkos()
546 PetscCall(PetscLogGpuFlops(2.0 * csrmat.nnz())); in MatMultHermitianTransposeAdd_SeqAIJKokkos()
1029 PetscCall(PetscLogGpuFlops(aijkok->a_dual.extent(0))); in MatScale_SeqAIJKokkos()
1050 PetscCall(PetscLogGpuFlops(n)); in MatShift_SeqAIJKokkos()
1084 PetscCall(PetscLogGpuFlops(n)); in MatDiagonalSet_SeqAIJKokkos()
1117 PetscCall(PetscLogGpuFlops(nz)); in MatDiagonalScale_SeqAIJKokkos()
[all …]
/petsc/src/mat/impls/dense/seq/cupm/
H A Dmatseqdensecupm.hpp408 PetscCall(PetscLogGpuFlops(2.0 * n * n * m / 3.0)); in Factor()
446 PetscCall(PetscLogGpuFlops(nrhs * (2.0 * m * m - m))); in Solve()
493 PetscCall(PetscLogGpuFlops(1.0 * n * n * n / 3.0)); in Factor()
545 PetscCall(PetscLogGpuFlops(nrhs * (2.0 * m * m - m))); in Solve()
595 PetscCall(PetscLogGpuFlops(2.0 * min * min * (std::max(m, n) - min / 3.0))); in Factor()
865 PetscCall(PetscLogGpuFlops(2.0 * m * n - (yy ? 0 : m))); in MatMultAddColumnRange_Dispatch_()
1333 PetscCall(PetscLogGpuFlops(1.0 * m * n * k + 1.0 * m * n * (k - 1))); in MatMatMult_Numeric_Dispatch()
1551 PetscCall(PetscLogGpuFlops(N)); in Scale()
1600 PetscCall(PetscLogGpuFlops(PetscMax(2 * N - 1, 0))); in AXPY()
1819 PetscCall(PetscLogGpuFlops(1.0 * n * n * n / 3.0)); in InvertFactors()
/petsc/src/tao/unconstrained/tutorials/
H A Drosenbrock4.h366 PetscCall(PetscLogGpuFlops(RosenbrockObjectiveFlops * n_comp)); in RosenbrockObjective_Device()
376 PetscCall(PetscLogGpuFlops(RosenbrockGradientFlops * n_comp)); in RosenbrockGradient_Device()
386 PetscCall(PetscLogGpuFlops(RosenbrockObjectiveGradientFlops * n_comp)); in RosenbrockObjectiveGradient_Device()
396 PetscCall(PetscLogGpuFlops(RosenbrockHessianFlops * n_comp)); in RosenbrockHessian_Device()
/petsc/src/ksp/pc/impls/vpbjacobi/cuda/
H A Dvpbjacobi_cuda.cu131 PetscCall(PetscLogGpuFlops(pcuda->nsize * 2)); /* FMA on entries in all blocks */ in PCApplyOrTranspose_VPBJacobi_CUDA()
/petsc/src/mat/impls/aij/seq/seqviennacl/
H A Daijviennacl.cxx200 PetscCall(PetscLogGpuFlops(2.0 * a->nz - a->nonzerorowcnt)); in MatMult_SeqAIJViennaCL()
236 PetscCall(PetscLogGpuFlops(2.0 * a->nz)); in MatMultAdd_SeqAIJViennaCL()
/petsc/src/mat/impls/aij/seq/seqhipsparse/
H A Daijhipsparse.hip.cxx1133 PetscCall(PetscLogGpuFlops(2.0 * hipsparseTriFactors->nnz - A->cmap->n)); in MatSolveTranspose_SeqAIJHIPSPARSE()
1171 PetscCall(PetscLogGpuFlops(2.0 * hipsparseTriFactors->nnz - A->cmap->n)); in MatSolveTranspose_SeqAIJHIPSPARSE_NaturalOrdering()
1211 PetscCall(PetscLogGpuFlops(2.0 * hipsparseTriFactors->nnz - A->cmap->n)); in MatSolve_SeqAIJHIPSPARSE()
1241 PetscCall(PetscLogGpuFlops(2.0 * hipsparseTriFactors->nnz - A->cmap->n)); in MatSolve_SeqAIJHIPSPARSE_NaturalOrdering()
1282 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - fact->rmap->n)); in MatSolve_SeqAIJHIPSPARSE_ILU0()
1339 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - fact->rmap->n)); in MatSolveTranspose_SeqAIJHIPSPARSE_ILU0()
1389 PetscCall(PetscLogGpuFlops(fs->numericFactFlops)); in MatILUFactorNumeric_SeqAIJHIPSPARSE_ILU0()
1582 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - fact->rmap->n)); in MatSolve_SeqAIJHIPSPARSE_ICC0()
1634 PetscCall(PetscLogGpuFlops(fs->numericFactFlops)); in MatICCFactorNumeric_SeqAIJHIPSPARSE_ICC0()
2329 PetscCall(PetscLogGpuFlops(n * 2.0 * csrmat->num_entries)); in MatProductNumeric_SeqAIJHIPSPARSE_SeqDENSEHIP()
[all …]
/petsc/src/mat/impls/aij/seq/seqcusparse/
H A Daijcusparse.cu769 PetscCall(PetscLogGpuFlops(4.0 * aij->nz - A->rmap->n)); in MatSolve_SeqAIJCUSPARSE_Cholesky()
1458 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - m)); in MatSolve_SeqAIJCUSPARSE_LU()
1529 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - A->rmap->n)); in MatSolveTranspose_SeqAIJCUSPARSE_LU()
1582 PetscCall(PetscLogGpuFlops(2.0 * cusparseTriFactors->nnz - A->cmap->n)); in MatSolveTranspose_SeqAIJCUSPARSE()
1620 PetscCall(PetscLogGpuFlops(2.0 * cusparseTriFactors->nnz - A->cmap->n)); in MatSolveTranspose_SeqAIJCUSPARSE_NaturalOrdering()
1660 PetscCall(PetscLogGpuFlops(2.0 * cusparseTriFactors->nnz - A->cmap->n)); in MatSolve_SeqAIJCUSPARSE()
1690 PetscCall(PetscLogGpuFlops(2.0 * cusparseTriFactors->nnz - A->cmap->n)); in MatSolve_SeqAIJCUSPARSE_NaturalOrdering()
1755 PetscCall(PetscLogGpuFlops(fs->numericFactFlops)); in MatILUFactorNumeric_SeqAIJCUSPARSE_ILU0()
1942 PetscCall(PetscLogGpuFlops(2.0 * aij->nz - fact->rmap->n)); in MatSolve_SeqAIJCUSPARSE_ICC0()
2004 PetscCall(PetscLogGpuFlops(fs->numericFactFlops)); in MatICCFactorNumeric_SeqAIJCUSPARSE_ICC0()
[all …]
/petsc/src/mat/impls/h2opus/cuda/
H A Dmath2opus.cu447 PetscCall(PetscLogGpuFlops(1e9 * gops)); in MatMultNKernel_H2OPUS()
632 PetscCall(PetscLogGpuFlops(1e9 * gops)); in MatMultKernel_H2OPUS()
1408 PetscCall(PetscLogGpuFlops(1e9 * gops)); in MatH2OpusOrthogonalize()
1515 PetscCall(PetscLogGpuFlops(1e9 * gops)); in MatH2OpusCompress()
/petsc/include/
H A Dpetsclog.h755 static inline PetscErrorCode PetscLogGpuFlops(PetscLogDouble n) in PetscLogGpuFlops() function
793 #define PetscLogGpuFlops(a) PETSC_SUCCESS macro
/petsc/include/petsc/private/
H A Dmatdensecupmimpl.h430 PetscCall(PetscLogGpuFlops(rend2 - rstart)); in DiagonalUnaryTransform()
/petsc/src/mat/impls/sell/seq/seqhip/
H A Dsellhip.hip.cxx704 PetscCall(PetscLogGpuFlops(2.0 * a->nz - a->nonzerorowcnt)); in MatMult_SeqSELLHIP()
859 PetscCall(PetscLogGpuFlops(2.0 * a->nz)); in MatMultAdd_SeqSELLHIP()
/petsc/src/mat/impls/sell/seq/seqcuda/
H A Dsellcuda.cu704 PetscCall(PetscLogGpuFlops(2.0 * a->nz - a->nonzerorowcnt)); in MatMult_SeqSELLCUDA()
866 PetscCall(PetscLogGpuFlops(2.0 * a->nz)); in MatMultAdd_SeqSELLCUDA()
/petsc/src/vec/is/sf/impls/basic/
H A Dsfpack.c826 …if (PetscMemTypeDevice(link->rootmtype)) PetscCall(PetscLogGpuFlops(bas->rootbuflen[scope] * link-… in PetscSFLinkLogFlopsAfterUnpackRootData()
839 …if (PetscMemTypeDevice(link->leafmtype)) PetscCall(PetscLogGpuFlops(sf->leafbuflen[scope] * link->… in PetscSFLinkLogFlopsAfterUnpackLeafData()
/petsc/src/ksp/pc/impls/bjacobi/bjkokkos/
H A Dbjkokkos.kokkos.cxx727 PetscCall(PetscLogGpuFlops((PetscLogDouble)h_metadata[blkID].flops)); in PCApply_BJKOKKOS()