Leo Fang
I don't think cuda.core can depend on cuda.parallel, currently it is the other way around and it'd be unattainable to change this.
Thanks for catching this. It seems raising `ValueError` is a mistake.
CuPy does not support the `device=` kwarg in the array constructors today, so some works have to happen first. But if we were to support it now I'd be supportive...
> At [yesterday's consortium meeting](https://hackmd.io/zn5bvdZTQIeJmb3RW1B-8g), everyone was in agreement that in > > (...) > > z should be on device 0. #919 makes that clear. However, CuPy maintainers were...
> However, this is not the case as an example for UCS, and some consumers do rely on it, such as UCC. This causes a problem for packaging because "rebuild...
For the first error it's because `range()` expects either an `int` or an object with `__index__`, and `cp.ndarray` is neither. It also does not make sense for CuPy to support...
I am currently in a position where I'd feel uneasy to engage in any ROCm-related discussion 😅 That said, I think the community is moving to use DLPack, and in...
Last time I checked, nanobind has native DLPack support, but pybind11 has not. Just FYI.
It seems @gpinkert has enabled this in #9340! Please let us know if there's anything missing. Closing.
> * Support converting arbitrary objects to `StridedMemoryView` If we make this a two-step approach, where the 1st step is to simply accept passing a `StridedMemoryView` object to `launch()` as...