I'd love something like this with Python and ctypes.
There are pyopencl and pycuda. Not sure what the best tutorials would be though.
Interactive GPU Programming, Part 1: Hello CUDA
11–20 of 30 posts
Re: Interactive GPU Programming, Part 1: Hello CUDA
#12Okay, so it's easier then directly using the the CUDA, etc., C toolchains, perhaps, but why not compare to Python + Numba, which has been available with GPU support for quite a while, and likewise avoids direct exposure to the underlying C toolchains, provides interactive compilation, can be used with a nice REPL (or, Jupyter Notebook), etc.?
The author wrote the tutorial as he pleased. Nobody stops you from submitting a better link.
Re: Interactive GPU Programming, Part 1: Hello CUDA
#13I'd love something like this with Python and ctypes.
PyCUDA is more of a wrapper type, performance won't be good compared to native cuda.
Re: Interactive GPU Programming, Part 1: Hello CUDA
#14I hope CUDA will get replaced by Vulkan merged with whatever core OpenCL features it still needs. Looks like Khronos are looking into converging them in some way: https://www.pcper.com/reviews/Graphics-Cards/Follow-Neil-Tre... CUDA unfortuantely is Nvidia's lock-in, so not a good way forward.
FWIW, all my libraries work with both CUDA and OpenCL.
While I agree with your sentiment, unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by refusing to put more than a handful of people on the job of providing OpenCL libraries. And, BTW, they change their minds every few years. I hope that HIP won't be abandonware...
Re: Interactive GPU Programming, Part 1: Hello CUDA
#15I hope CUDA will get replaced by Vulkan merged with whatever core OpenCL features it still needs. Looks like Khronos are looking into converging them in some way: https://www.pcper.com/reviews/Graphics-Cards/Follow-Neil-Tre... CUDA unfortuantely is Nvidia's lock-in, so not a good way forward.
FWIW, all my libraries work with both CUDA and OpenCL. While I agree with your sentiment, unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by refusing to put more than a handful of people on the job of providing OpenCL libraries. And, BTW, they change their minds every few years. I hope that HIP won't be abandonware...
> unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by refusing to put more than a handful of people on the job of providing OpenCL libraries.
Vulkan itself is developed and supported well, and it already can be used for compute as far as I know. But apparently there are some features that come from the OpenCL world that need to be filled in. It wouldn't be AMD's exclusive effort. So hopefully things will start moving.
Re: Interactive GPU Programming, Part 1: Hello CUDA
#16Earlier quoted context omitted.
FWIW, all my libraries work with both CUDA and OpenCL. While I agree with your sentiment, unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by refusing to put more than a handful of people on the job of providing OpenCL libraries. And, BTW, they change their minds every few years. I hope that HIP won't be abandonware...
> unfortunately Nvidia is the only vendor that pays considerable number of people to develop the ecosystem. AMD basically says "get lost" by refusing to put more than a handful of people on the job of providing OpenCL libraries. Vulkan itself is developed and supported well, and it already can be used for compute as far as I know. But apparently there are some features that come from the OpenCL world that need to be…
The language and basic platform is not a problem. OpenCL was and is OK. However, the libraries are far and between. CUDA offers cuBLAS, cuFFT, cuDNN, cuSolve, etc. For OpenCL, even the decent BLAS library (CLBlast) had to be written by a guy who did it for free, while AMD's clBLAS is more or less stalled (and I never managed to build it on Linux in the first place), and that's it...
Re: Interactive GPU Programming, Part 1: Hello CUDA
#17Followed along until I had to compile a kernel, now I'm facing a java.lang.UnsatisfiedLinkError: Error while loading native library "JNvrtc-0.9.0-windows-x86_64" This seems to be a dependency of ClojureCUDA, but I don't see anything about it in their installation instructions. I have the CUDA Toolkit installed. Everything worked up to this point.
Re: Interactive GPU Programming, Part 1: Hello CUDA
#18Are there similar resources/tutorials for GPU programming/CUDA/openCL for pure Java?
Re: Interactive GPU Programming, Part 1: Hello CUDA
#19Why do things using CUDA often require older versions of compilers?
E.g. ccminer. Try to make it, but it finds my modern gcc or clang too modern :(
Re: Interactive GPU Programming, Part 1: Hello CUDA
#20I was irritated that it took 8 paragraphs to reveal it was clojure. And it was extra annoying that the vagueness was entirely deliberate.