Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Nvidia adds native Python support to CUDA

thenewstack.io

Post reply on HN
Print view View original

181–190 of 196 posts

Re: Nvidia adds native Python support to CUDA

#181

Post by LegNeato »

the__alchemist wrote: post #28
chasely wrote: post #20

Earlier quoted context omitted.

The Rust-CUDA project just recently started up again [0], I've started digging into it a little bit and am hoping to contribute to it since the summers are a little slower for me. [0] https://github.com/rust-gpu/rust-cuda

Still broken though! Has been for years. In a recent GH issue regarding desires for the reboot, I asked: "Try it on a few different machines (OS, GPUs, CUDA versions etc), make it work on modern RustC and CUDA versions without errors." The response was "That will be quite some work." Meanwhile, Cudarc works...

Maintainer here. It is not broken, it works. See https://rust-gpu.github.io/blog/2025/03/18/rust-cuda-update

Re: Nvidia adds native Python support to CUDA

#182

Post by airstrike »

system2 wrote: post #176
airstrike wrote: post #124

Earlier quoted context omitted.

The bad habits I was thinking about were more in the line of not understanding how memory is being used (even something as simple as stack vs. heap allocation), not having a type system that forces you to think about the types of data structure you have in your system, and overall just being forced to design before coding

I ask for help from ChatGPT before I go all in, and it creates these old-fashioned ASCII graphics to show how the flow will be. I think newcomers will not have the bad habits we have/had.

That's really not enough or the same, trust me. You can't outsource your entire understanding to ChatGPT, and I say this as someone who's keen on getting help from AI assistants to write boilerplate code, rubberduck bugs or debate design decisions

Re: Nvidia adds native Python support to CUDA

#183

Post by airstrike »

Barrin92 wrote: post #168
airstrike wrote: post #76

Earlier quoted context omitted.

As someone who spent nearly a decade with Python, I'd say 90% of people will answer "yes", so I'd like to offer a different perspective. IMHO if you want to pick it up for a couple toy projects just to get a feel of what coding is like, then by all means try it out. But eventually you'll benefit tremendously from exploring other languages. Python will teach you a lot of bad habits. You will feel like you know what yo…

>Knowing what I know now, I wish Rust existed when I started out so that it could have been my first language No offense but I don't think this makes any sense (or only if you take the first part of that sentence literally). It's like jumping into Calculus 3 to introduce a kid to maths. From a teaching standpoint, if you're a beginner, you can't even understand what problem Rust solves. Someone who doesn't know what…

People were learning C and malloc long before Python came along. You don't need to start with a high level language.

Rust does way more than "solve problems professional C++ developers have". That's not a fair or accurate read of the language. I think you're misinformed.

Re: Nvidia adds native Python support to CUDA

#184

Post by the__alchemist »

LegNeato wrote: post #181
the__alchemist wrote: post #28

Earlier quoted context omitted.

Still broken though! Has been for years. In a recent GH issue regarding desires for the reboot, I asked: "Try it on a few different machines (OS, GPUs, CUDA versions etc), make it work on modern RustC and CUDA versions without errors." The response was "That will be quite some work." Meanwhile, Cudarc works...

Maintainer here. It is not broken, it works. See https://rust-gpu.github.io/blog/2025/03/18/rust-cuda-update

Thanks! Will give it a try, and report back.

edit: I'm still showing the latest release as from 2022, which I've already verified doesn't work.

Re: Nvidia adds native Python support to CUDA

#185

Post by musicale »

almostgotcaught wrote: post #144
dragonwriter wrote: post #113

Earlier quoted context omitted.

> i'm just able to actually read and comprehend what i'm reading rather than perform hype: The evidence of that is lacking. > so the article is about cuda-core, not whatever you think it's about cuda.core (a relatively new, rapidly developing, library whose entire API is experimental) is one of several things (NVMath is another) mentioned in the article, but the newer and as yet unreleased piece mentioned in the arti…

> No, as is is fairly explicit in the next line after the one you quote, it is about the Nvidia CUDA Python toolchain using in-process compilation rather than relying on shelling out to out-of-process command-line compilers for CUDA code. my guy what i am able to read, which you are not, is the source and release notes. i do not need to read tweets and press releases because i know what these things actually are. her…

JIT as an adjective means just-in-time, as opposed to AOT, ahead-of-time. What Nvidia discussed at GTC was a software stack that will enable you to generate new CUDA kernels dynamically at runtime using Python API calls. It is a just-in-time (runtime, dynamic) compiler system rather than an ahead-of-time (pre-runtime, static) compiler.

Re: Nvidia adds native Python support to CUDA

#186

Post by Barrin92 »

airstrike wrote: post #183
Barrin92 wrote: post #168

Earlier quoted context omitted.

>Knowing what I know now, I wish Rust existed when I started out so that it could have been my first language No offense but I don't think this makes any sense (or only if you take the first part of that sentence literally). It's like jumping into Calculus 3 to introduce a kid to maths. From a teaching standpoint, if you're a beginner, you can't even understand what problem Rust solves. Someone who doesn't know what…

People were learning C and malloc long before Python came along. You don't need to start with a high level language. Rust does way more than "solve problems professional C++ developers have". That's not a fair or accurate read of the language. I think you're misinformed.

>You don't need to start with a high level language.

I didn't say that. I said you can start with a high or low level language. I did literally mention C in my own post as a decent starting point.

Rust however is not a beginner friendly language because again, the thing that sets it apart is that it aims to solve a particular domain specific problem of programming, which is memory management, in a unique way that means nothing to a person who has never been exposed to the problem in the first place.

Re: Nvidia adds native Python support to CUDA

#187

Post by throwaway314155 »

almostgotcaught wrote: post #146
ladberg wrote: post #142

Earlier quoted context omitted.

The main release highlighted by the article is cuTile which is certainly about jitting kernels from Python code

> main release there is no release of cutile (yet). so the only substantive thing that the article can be describing is cuda-core - which it does describe and is a recent/new addition to the ecosystem. man i can't fathom glazing a random blog this hard just because it's tangentially related to some other thing (NV GPUs) that clearly people only vaguely understand.

christ man lighten the fuck up. there's zero need to be _so_ god damn patronizing and disrespectful.

Re: Nvidia adds native Python support to CUDA

#188

Post by throwaway314155 »

airstrike wrote: post #124
system2 wrote: post #118

Earlier quoted context omitted.

With the help of GPT, I think the bad habit part is non-existent anymore. Learning it from GPT really helps people nowadays. Ask ChatGPT 4.0 some questions, and you will be shocked by how well it describes the code. Just don't ask to fix indentations because it will do it line by line for hours. But it finds mistakes quickly and points you in the right direction. And of course, it comes up with random non-existent mo…

The bad habits I was thinking about were more in the line of not understanding how memory is being used (even something as simple as stack vs. heap allocation), not having a type system that forces you to think about the types of data structure you have in your system, and overall just being forced to design before coding

Respectfully, many of those things aren't a concern from Python's point of view. And why should they be? If your program runs imperceptibly slower, or using an insignificant amount of extra memory, any attempts to fix this are considered a premature optimization that gets in the way of what is more important to pythonistas - developer experience and high level abstractions.

Frankly, the comparison with Rust doesn't even really make sense. They are different tools for very different problems.

Re: Nvidia adds native Python support to CUDA

#189

Post by ryao »

pjmlp wrote: post #177
ryao wrote: post #125

Earlier quoted context omitted.

They replaced C with C++. For example, try passing a function pointer as a void pointer argument without a cast. C says this should work. C++ says it should not. There are plenty of other differences that make it C++ and not C, if you know to look for them. The fact that C++ symbol names are used for one, which means you need to specify extern “C” if you want to reference them from the CUDA driver API. Then there is…

They replaced most of the documentation with C++ examples, given the benefits the language has over C, that was already obvious to me in 1993. As for the language extensions required by CUDA C, it is kind of interesting that clang and GCC extensions are praised and people keep referring to them as C, while everyone else's extensions are never C or C++ under the same measure. With OpenAAC directives, an HPC industry s…

I used to like C++. Then it caused me headaches one too many times because of things it implemented that C did not have. Now I prefer to use C whenever I can, since it avoids entire classes of headaches that only exist in C++.

Re: Nvidia adds native Python support to CUDA

#190

Post by gymbeaux »

pjmlp wrote: post #37
ErrorNoBrain wrote: post #16

Earlier quoted context omitted.

They are, if they cant find an nvidia card

NVidia cards are everywhere, the biggest difference to AMD is that even my lousy laptop GeForce cards can be used for CUDA. No need for a RTX for learning and getting into CUDA programming.

True, although I believe Maxwell is the oldest supported architecture for the current CUDA 12.x. Maxwell (eg GTX 980) came out around 2013, if memory serves. 10+ years of support is not bad at all considering ROCm supports only like 3 consumer AMD GPUs.

So your lousy laptop GTX 750Ti ehhh probably can’t practically be used for CUDA. But your lousy 1050Ti Max-Q? Sure.

Post reply on HN