Aaryaman Vasishta 0826aafa04 [ROCm/Windows] Support aotriton for scaled_dot_product_attention on Windows. (#162330) ...

Enables flash attention and/or memory efficient attention on Windows with scaled_dot_product_attention via. aotriton.
Already tested to be working on Windows with TheRock.

Steps to enable: simply set `USE_FLASH_ATTENTION=1` and `USE_MEM_EFF_ATTENTION=1` as usual. See https://github.com/ROCm/TheRock/blob/main/external-builds/pytorch/build_prod_wheels.py#L578-L604

Pull Request resolved: https://github.com/pytorch/pytorch/pull/162330
Approved by: https://github.com/jeffdaily

Co-authored-by: Scott Todd <scott.todd0@gmail.com>

2025-09-15 16:13:03 +00:00

..

aotriton.cmake

[ROCm/Windows] Support aotriton for scaled_dot_product_attention on Windows. (#162330)

2025-09-15 16:13:03 +00:00

EigenBLAS.cmake

Build EigenBlas as static library (#44747)

2020-09-16 10:25:26 -07:00

nccl.cmake

Nccl update to 2.25.1 for cuda 12.4-12.8 (#146073)

2025-02-19 03:52:26 +00:00

nnpack.cmake

[Cmake] Make PyTorch buildable by CMake-4.x (#150203)

2025-03-29 01:39:13 +00:00

rccl.cmake

[ROCm] LoadHIP CMake cleanup (#137112)

2024-10-13 00:06:41 +00:00

ucc.cmake

UCC PG build in CI (#81583)

2022-08-10 00:23:47 +00:00