en Copyright 2025 Java http://opengrok.kc2vjw.com:8080/history/petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h#9dd11ecf0918283bb567d8b33a92f53ac4ea7840 Merge branch 'jacobf/2023-08-17/header-guard-check' into 'main'Check header guardsSee merge request petsc/petsc!6822 List of files: /petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h Fri, 25 Aug 2023 13:49:03 +0000 Satish Balay <balay@mcs.anl.gov> http://opengrok.kc2vjw.com:8080/history/petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h#a496304597bacff3545e802853d69e8765312868 Convert all header guards to pragma once List of files: /petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h Fri, 18 Aug 2023 15:13:00 +0000 Jacob Faibussowitsch <jacob.fai@gmail.com> http://opengrok.kc2vjw.com:8080/history/petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h#3ea99036a5fedea4d39e7e77471d0ab500c249d7 Fix some malformed if !defined() header guards List of files: /petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h Thu, 17 Aug 2023 21:56:14 +0000 Jacob Faibussowitsch <jacob.fai@gmail.com> http://opengrok.kc2vjw.com:8080/history/petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h#dd874c20e1b8bb56639592ba136e1f877322cbbd Merge branch 'hongzh/sell-cuda' into 'main'SELL-based SpMVSee merge request petsc/petsc!3428 List of files: /petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h Mon, 10 Apr 2023 23:51:00 +0000 Satish Balay <balay@mcs.anl.gov> http://opengrok.kc2vjw.com:8080/history/petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h#2d1451d43b73a0495cd81c074cbc1e0206888947 Initial commit for porting SELL to GPU- Add tiled SPMV and basic SpMVfor SeqSELL- Tested in serial- Offloadmask is used to determine when the matrix should be copied to GPU- Use different slice height for CUDA version- By checking the nonzerostate, PETSc can decide if the whole matrix need to be copied or just the values need to be copied- Make the convert function public so that the very slow MatConvert_Basic can be avoided sometimes. E.g. one can use a two-step convert method: AIJ->SELL,SELL->SELLCUDA instead of the direct convert AIJ->SELLCUDA- Make the FLOPS count for SELL same as that for AIJCUSPARSE.- MatDisAssemble is not needed.- Change slice height from 32 to 16 for GPU- To overlap communication with MatMult, VecScatterBegin() should be called before MatMult() for the diagonal part.- SLICE_HEIGHT is defined to be 32 to match the warp size of GPU. For other cases, it is still 8.Funded-by:Project: PETSc for GPUTime: 42 hoursReported-by:Thanks-to: List of files: /petsc/src/mat/impls/sell/mpi/mpicuda/mpicudamatimpl.h Thu, 09 Jan 2020 04:05:36 +0000 Hong Zhang <hongzhang@anl.gov>