atalman 2f5a24c2a2 Smoke tests don't run nvshmem on Windows (#162646) ...

Only available for linux x86 and aarch64 :
https://pypi.org/project/nvidia-nvshmem-cu13/#files

nvshmem is available only on linux:
``
"nvidia-nvshmem-cu12==3.3.24; platform_system == 'Linux' and platform_machine == 'x86_64' | "
``
https://github.com/pytorch/pytorch/blob/main/.github/scripts/generate_binary_build_matrix.py#L57
Pull Request resolved: https://github.com/pytorch/pytorch/pull/162646
Approved by: https://github.com/kwen2501

2025-09-11 16:09:20 +00:00

..

aarch64_linux

Revert "Move prioritized text linker optimization code from setup.py to cmake (#160078)"

2025-09-11 15:29:29 +00:00

caffe2

[CI] Removing --user flag from all pip install commands (#154900)

2025-07-14 21:09:42 +00:00

docker

Revert "Build and Install Arm Compute Library in manylinux docker image (#159737)"

2025-09-11 15:25:24 +00:00

libtorch

[CD] Build libtorch without nvshmem (#160910)

2025-08-19 05:58:25 +00:00

lumen_cli

Add more tests for vllm and clean out the old vllm test (#162292)

2025-09-09 05:53:46 +00:00

magma

Add Magma build for CUDA 13.0 (#160770)

2025-08-19 01:10:00 +00:00

magma-rocm

[ROCm] Use almalinux docker files for building Magma (#152488)

2025-04-30 17:53:30 +00:00

manywheel

[CD] Add cuda 13.0 libtorch builds, remove CUDA 12.9 builds (#161916)

2025-09-05 07:47:54 +00:00

onnx

[CI] Removing --user flag from all pip install commands (#154900)

2025-07-14 21:09:42 +00:00

pytorch

Smoke tests don't run nvshmem on Windows (#162646)

2025-09-11 16:09:20 +00:00

wheel

Revert "Make distributed modules importable even when backend not built (#159889)" (#162568)

2025-09-10 04:29:42 +00:00