pytorch
e18a2219 - Implement scatter reductions (CUDA), remove divide/subtract (#41977)

Commit

4 years ago

Implement scatter reductions (CUDA), remove divide/subtract (#41977) Summary: Fixes https://github.com/pytorch/pytorch/issues/33394 . This PR does two things: 1. Implement CUDA scatter reductions with revamped GPU atomic operations. 2. Remove support for divide and subtract for CPU reduction as was discussed with ngimel . I've also updated the docs to reflect the existence of only multiply and add. Pull Request resolved: https://github.com/pytorch/pytorch/pull/41977 Reviewed By: mruberry Differential Revision: D23748888 Pulled By: ngimel fbshipit-source-id: ea643c0da03c9058e433de96db02b503514c4e9c

Author

v0dro

Committer

facebook-github-bot

Parents

fdeee745

Files13

aten/src
- ATen
  - native
    - TensorAdvancedIndexing.cpp
    - TensorAdvancedIndexing.h
    - cpu
      - ScatterGatherKernel.cpp
    - cuda
      - ScatterGatherKernel.cu
    - native_functions.yaml
  - test
    - CMakeLists.txt
    - cuda_atomic_ops_test.cu
- THC
  - THCAtomics.cuh
  - THCNumerics.cuh
test
- test_torch.py
third_party
- gloo
- ideep
torch
- _tensor_docs.py

pytorch e18a2219 - Implement scatter reductions (CUDA), remove divide/subtract (#41977)

pytorch
e18a2219 - Implement scatter reductions (CUDA), remove divide/subtract (#41977)