cuda-python
cuda-python copied to clipboard
CUDA Python: Performance meets Productivity
Currently this is blocked by MSVC not pre-installed on the VM image that we use (#457). Based on the past experience (#267), it takes way too long to just install...
Not all PRs would require full build/test pipelines (as we currently do today). For example, if a PR only touches code in `cuda.core`, then 1. we don't need to rebuild...
Things to note - this is an experimental build that we reserve the right to stop offering any time (just like the CPython cp313t build is!) - the change in...
We need to continue providing the GIL build and additionally a free-threading (no-GIL) build.
Currently the Code of Conduct (CoC) and Contributing pages are placed under the parent CUDA Python homepage: https://github.com/NVIDIA/cuda-python/blob/a85576273c54098af60483a80b043224239bec35/cuda_python/docs/source/index.rst?plain=1#L33-L34 However, technically this is not right, because each module under `cuda` could...
For: - CI/CD files (GHA workflows and scripts) - `cuda.bindings` - `cuda.core`
#444 added a temporary Windows GPU runner with T4 so that we can finally start testing on Windows 🎉 However, it only runs on TCC mode and there is also...
See CI failure: https://github.com/NVIDIA/cuda-python/actions/runs/13488648820/job/37683655417?pr=463#step:18:264 ``` > assert other_stream.handle == stream.handle E assert == E + where = .handle E + and = .handle tests/test_stream.py:93: AssertionError ```
This allows us to share the class declarations with C/C++: https://cython.readthedocs.io/en/latest/src/userguide/external_C_code.html#using-cython-declarations-from-c which is needed after PR #463 set up a path for Python-C/C++/Cython interoperability, but without adding the `public` qualifier...