Add torch.Tensor fast path for StridedMemoryView via AOTI tensor bridge#1894

Merged

leofang merged 37 commits intoNVIDIA:mainfrom

leofang:tensor-bridge-749

Apr 27, 2026

Commits on Apr 9, 2026

Commits on Apr 10, 2026

Commits on Apr 14, 2026

Fix pyobj_to_aten_handle for PyTorch 2.3–2.9 MaybeOwned layout

leofang
and
claude
committed
Consolidate torch tensor bridge tests into TestViewCPU/TestViewGPU

leofang
and
claude
committed
Extract _arr_size helper for torch/numpy size compatibility

leofang
and
claude
committed
Fix ruff formatting in test_utils.py

leofang
and
claude
committed
Add readonly comment and fix vendored header license to BSD-3-Clause

leofang
and
claude
committed
Merge bfloat16 test into test_torch_tensor_bridge_dtypes parametrization

leofang
and
claude
committed
Fix SPDX linter: use PyTorch copyright in vendored header

leofang
and
claude
committed
Fix Windows build: generate stub import library for AOTI symbols

leofang
and
claude
committed
Merge branch 'main' into tensor-bridge-749
leofang
authored
Exclude torch DLLs from delvewheel repair on Windows

leofang
and
claude
committed
Merge branch 'tensor-bridge-749' of https://github.com/leofang/cuda-python into tensor-bridge-749
leofang
committed
Fix delvewheel flag: use --exclude instead of --no-dll

leofang
and
claude
committed
Merge branch 'main' into tensor-bridge-749
leofang
authored

Commits on Apr 22, 2026

Add strided layout guard to tensor bridge, reject sparse tensors

leofang
and
claude
committed

Commits on Apr 24, 2026