HIP
ZLUDA
HIP | ZLUDA | |
---|---|---|
29 | 35 | |
3,453 | 7,671 | |
1.2% | - | |
8.9 | 7.0 | |
6 days ago | 3 days ago | |
C++ | Rust | |
MIT License | Apache License 2.0 |
Stars - the number of stars that a project has on GitHub. Growth - month over month growth in stars.
Activity is a relative number indicating how actively a project is being developed. Recent commits have higher weight than older ones.
For example, an activity of 9.0 indicates that a project is amongst the top 10% of the most actively developed projects that we are tracking.
HIP
- Hip: Runtime API and Kernel Language for Portable Apps for AMD and Nvidia GPUs
-
Open-source project ZLUDA lets CUDA apps run on AMD GPUs
Is it perhaps because they want people to use HIP?
> HIP is very thin and has little or no performance impact over coding directly in CUDA mode.
> The HIPIFY tools automatically convert source from CUDA to HIP.
1. https://github.com/ROCm/HIP
-
AMD's Next GPU Is a 3D-Integrated Superchip
AMD has released HIP and a tool called HIPIFY which kind of behaves like this but at the source level¹. Rather than try and just translate CUDA to work on AMD compute they are more focused on higher level tooling.
Currently they seem to have a particular focus on AI frameworks and tools like PyTorch/Tensorflow/ONNX. They have sponsored and helped with a lot of PyTorch development for example, so PyTorch support for AMD is much better than it was this time last year².
¹(https://github.com/ROCm/HIP)
²(https://pytorch.org/blog/experience-power-pytorch-2.0/)
-
Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
> what would be the point for someone to add ROCm support to various pieces of software which currently require CUDA
It isn't just old cards though, CUDA is a point of centralization on a single provider during a time when access to that providers higher end cards isn't even available and that is causing people to look elsewhere.
ROCm supports CUDA through the included HIP projects...
https://github.com/ROCm/HIP
https://github.com/ROCm/HIPCC
https://github.com/ROCm/HIPIFY
The later will regex replace your CUDA methods with HIP methods. If it is as easy as running hipify on your codebase (or just coding to HIP apis), it certainly makes sense to do so.
-
Nvidia on the Mountaintop
AMD's equivalent is HIP [1], for sufficiently flexible definitions of "equivalent". I can't speak to how complete/correct/performant it is (I'm just a guy running tutorial/toy-level ML stuff on an RDNA1 card), but part of AMD's problem is that it might not practically matter how well they do this because the broader ecosystem support specifically for the CUDA stack is so entrenched.
[1] https://github.com/ROCm-Developer-Tools/HIP
- Stable Diffusion in pure C/C++
- Would love to hear your information and knowledge to simplify my understanding on AMD's positioning in the AI market
-
Ask HN: C++ still dominates on GPUs, why not Rust?
From what I know, modern GPUs are still programmed with C++ exclusively. See CUDA [0] for Nvidia and ROCm [1] for AMD.
Why is this? Why Rust is not loved there?
[0] https://docs.nvidia.com/cuda/
[1] https://github.com/ROCm-Developer-Tools/HIP
-
[P] RWKV C++ Cuda library with no dependencies, no torch, and no python
Go ahead and try to ship ROCm code that works on multiple consumer graphics cards on Linux, MacOS, and Windows. As an example of how much AMD cares about it, the installation notes linked to in the readme returns a 404.
-
Someone found a ROCm 5.5 RC Docker Container that works on 7000 series GPUs
The big whoop for ROCm is that AMD invested a considerable amount of engineering time and talent into a tool they call hip. Basically, it's an analysis tool that does its best to port proprietary Nvidia CUDA-style code - which due to various smelly reasons rules the roost - to code that can happily run on AMD graphics cards, and presumably others. Intel has a similar thing going with OneAPI. They've done this whilst working on porting a lot of their code base to the linux AMGPU open source kernel driver, as well.
ZLUDA
-
Open-source project ZLUDA lets CUDA apps run on AMD GPUs
It now supports AMD GPUs since 3 weeks ago, check the latest commit at the repo:
https://github.com/vosen/ZLUDA
The article also mentions exactly this fact.
-
Nvidia bans using translation layers for CUDA software
Looks like nvidia is trying to keep the lynchpin of their entire business model from crumbling underneath them. ZLUDA lets you run unmodified CUDA applications with near-native performance on AMD GPUs.
https://github.com/vosen/ZLUDA
With Triton looking to eclipse CUDA entirely, im not sure this prohibition does anything more than placate casual shareholders.
-
Nvidia bans using translation layers for CUDA software to run on other chips
>Dark API functions are reverse-engineered and implemented by ZLUDA on a case-by-case basis once we observe an application making use of it.
https://github.com/vosen/ZLUDA/blob/master/ARCHITECTURE.md
-
Nvidia hits $2T valuation as AI frenzy grips Wall Street
> I know AMD have their competition, but their GPU software division keeps tripping over itself.
They are actively stepping on every rake there is. Eg they just stopped supporting the drop-in-cuda project everyone is waiting for, due to there being "no business-case for CUDA on AMD GPUs" [0].
[0] https://github.com/vosen/ZLUDA?tab=readme-ov-file#faq
-
Nvidia Is Now More Valuable Than Amazon and Google
https://github.com/vosen/ZLUDA
They still funded it and it was created.
- Debian on Apple hardware (M1 and later)
-
AMD Funded a Drop-In CUDA Implementation Built on ROCm: It's Open-Source
From the same repo, I found this excellent, well-written architecture document: https://github.com/vosen/ZLUDA/blob/master/ARCHITECTURE.md
I love the direct, "no bullshit" style of writing.
Some gems:
> Anyone familiar with C++ will instantly understand that compiling it is a complicated affair.
> Additionally CUDA allows, to a large degree, mixing CPU code and GPU code. What does all this complexity mean for ZLUDA? Absolutely nothing
> Since an application can dynamically link to either Driver API or Runtime API, it would seem that ZLUDA needs to provide both. In reality very few applications dynamically link to Runtime API. For the vast majority of applications it's sufficient to provide Driver API for dynamic (runtime) linking.
-
Intel CEO: 'The entire industry is motivated to eliminate the CUDA market'
CUDA is huge and nvidia spent a ton in a lot of "dead end" use cases optimizing it. There have been experiments with CUDA translation layers with decent performance[1]. There are two things that most projects hit:
1. The CUDA API is huge; I'm sure Intel/AMD will focus on what they need to implement pytorch and ignore every other use case ensuring that CUDA always has the leg up in any new frontier
2. Nvidia actually cares about developer experience. The most prominent example is Geohotz with tinygrad - where AMD examples didn't even work or had glaring compiler bugs. You will find nvidia engineer in github issues for CUDA projects. Intel/AMD hasn't made that level of investment and thats important because GPUs tend to be more fickle than CPUs.
[1] https://github.com/vosen/ZLUDA
-
Why Nvidia Keeps Winning: The Rise of an AI Giant
> I don't think you understand just how insanely difficult it is to break into that market.
You're right, I have no clue nor have I ever tried myself.
> Even with apple money or something like that, it's a losing prospect because in the time it'll take you to get up and off the ground (which is FOREVER) your competition will crush you.
This I find hard to believe, do you have a source or reference for that claim? Companies with that amount of cash are hardly going to be crushed by competition be it direct or indirect. Anyway, I'm talking more about the Intels and AMDs of this world.
We have very lacklustre efforts from players I won't name with their Zluda library (https://github.com/vosen/ZLUDA) which I got REALLY excited about, until I read the README.txt. Four contributors, last commit early 2021.
Why, oh why, is it this bad?
- Intel Arc Graphics Driver Change Leads To A Big Speed-Up Under Linux
What are some alternatives?
AdaptiveCpp - Implementation of SYCL and C++ standard parallelism for CPUs and GPUs from all vendors: The independent, community-driven compiler for C++-based heterogeneous programming models. Lets applications adapt themselves to all the hardware in the system - even at runtime!
InvokeAI - InvokeAI is a leading creative engine for Stable Diffusion models, empowering professionals, artists, and enthusiasts to generate and create visual media using the latest AI-driven technologies. The solution offers an industry leading WebUI, supports terminal use through a CLI, and serves as the foundation for multiple commercial products.
futhark - :boom::computer::boom: A data-parallel functional programming language
HIPIFY - HIPIFY: Convert CUDA to Portable C++ Code [Moved to: https://github.com/ROCm/HIPIFY]
kompute - General purpose GPU compute framework built on Vulkan to support 1000s of cross vendor graphics cards (AMD, Qualcomm, NVIDIA & friends). Blazing fast, mobile-enabled, asynchronous and optimized for advanced GPU data processing usecases. Backed by the Linux Foundation.
ROCm - AMD ROCm™ Software - GitHub Home [Moved to: https://github.com/ROCm/ROCm]
ginkgo - Numerical linear algebra software package
HIPIFY - HIPIFY: Convert CUDA to Portable C++ Code
rocm-arch - A collection of Arch Linux PKGBUILDS for the ROCm platform
arrow - 🏹 Better dates & times for Python
HIP-CPU - An implementation of HIP that works on CPUs, across OSes.
VC4CL - OpenCL implementation running on the VideoCore IV GPU of the Raspberry Pi models