MoFtZ

hi @imabf, I'm assuming you mean this project: https://github.com/openai/triton ILGPU tries to have as few dependencies as possible. It is able to run with just the .NET runtime and the...

@m4rs-mt I currently have access to a 10th Gen Intel Laptop CPU, and a 7th Gen Intel Desktop CPU, and have not been able to reproduce the problem either. However,...

This could potentially be split into two parts. The first part is adding support for `ImmutableArrayView`, which would be useful by itself in ensuring that the code does not modify...

@m4rs-mt what did you want to do with this feature request? what should we try to implement?

@CsabaStupak Have you tried using a `struct` instead of `ImmutableArray`? If you used a `struct` with a fixed-size buffer as the kernel parameter, it would be copied over into constant...

hi @pavlexander. The behavior you are seeing is because ILGPU uses `Trace.Assert` and `Debug.Assert` rather than `throw new IndexOutOfRangeException()`. ILGPU converts C# code to the instructions that will run on...

Regarding your other question about running the CPU accelerator in parallel, the `CreateCPUAccelerator` method can take an additional parameter of `CPUAcceleratorMode`. By default, ILGPU will use `CPUAcceleratorMode.Auto`. If a debugger...

I have setup a Ubuntu 22.04 LTS VM, and installed Cuda SDK 12.4 Update 1. In `/usr/local/cuda/lib64`, I found a symlink for `libcublas.so.12`, which is the correct naming that is...

Still failing.