amx
Halide
amx | Halide | |
---|---|---|
18 | 43 | |
859 | 5,714 | |
- | 0.5% | |
4.1 | 9.5 | |
2 months ago | 3 days ago | |
C | C++ | |
MIT License | GNU General Public License v3.0 or later |
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.
amx
-
Optimize sgemm on RISC-V platform
I am talking about the matrix/vector coprocessor (AMX). You can find some reverse-engineered documentation here: https://github.com/corsix/amx
On M3 a singe matrix block can achieve ~ 1TFLOP on DGEMM, I assume it will be closer to 4TFLOPS for SGEMM. The Max variants have two such blocks. Didn't do precise benchmarking myself, but switching Python/R matrix libraries to use Apple's BLAS result in 5-6x perf improvement on matrix heavy code for me.
-
Intel AMX
It's really cool. I hope it becomes more common for training/inference/numerics capable accelerators to be included in consumer hardware.
Apple's AMX is really under-documented, while the instructions were reverse engineered, Virtually no benchmarks are available comparing current chip generations, models and variants.
https://github.com/corsix/amx
- Why do x86 processors take up so much energy when compared to ARM?
-
Bfloat16 support coming to Apple's Metal and PyTorch [video]
Visible in the unofficial documentation for AMX instructions too - M2 only bf16 functionality - https://github.com/corsix/amx/blob/main/matfp.md
-
LLaMA-7B in Pure C++ with full Apple Silicon support
Confusingly there are 2 mechanisms to do matrix operations on the new apple hardware - AMX (https://github.com/corsix/amx) - and the ANE (apple neural engine) - which is enabled by CoreML. This code does not run on the neural engine but the author has a branch for his whisper.cpp project which uses it here: https://github.com/ggerganov/whisper.cpp/pull/566 - so it may not be long before we see it applied here as well. All of this is to say that it actually could get significantly faster if some of this work was able to be handed to the ANE with CoreML.
-
Linux 6.2: The first mainstream Linux kernel for Apple M1 chips arrives
really? seems pretty well documented here: https://github.com/corsix/amx
-
AMX: The Secret Apple M1 Coprocessor
Article is almost two years old, and has a huge correction at the bottom. It's just a proprietary ISA extension, there's even a repo documenting what's been reverse engineered.
- corsix/amx: Apple AMX Instruction Set
-
Show HN: Port of OpenAI's Whisper model in C/C++
You are correct, in that those are the four
My understanding is that the AMX is more tightly wound with the CPU, ultimately being accessible via an instruction set (https://github.com/corsix/amx), and it is useful if you need to do matrix multiplications interleaved with other CPU tasks. A common example would be a VIO loop or something where you want that data in the CPU caches.
The GPU and Neural Engine are not that – they take some time to set up and initialize. They also can parallelize tasks to a much higher degree. The GPU is more generalizable, because you can write compute shaders to do anything in parallel, but it uses a lot of resources. I'll have to check out the PR to see how exactly the MPS shaders match up with the task at hand, because you could also consider writing Metal compute shaders by hand.
I know the least about the ANE, but it has specific hardware for running ML models, and you have to process the weights ahead of time to make sure they are in the right format. It can run ML models very efficiently and is the most battery friendly.
-
Ask HN: Are there any undocumented ISA extensions used in Linux systems?
If someone were to build a Linux system with proprietary ISA extensions, how would they do it given Linux is open source? Are there any examples of this being done? Would it be possible at all?
I got inspiration from this (https://github.com/corsix/amx) and I wondered if someone has done it before on a Linux-based system. I understand a userspace library could be created to access those instructions from userspace, but how would then they be implemented in the kernel? Through a proprietary kernel module built using a custom compiler? Or is that not needed at all and the library could just run on the processor taking advantage of the proprietary extensions?
Halide
-
Show HN: Flash Attention in ~100 lines of CUDA
If CPU/GPU execution speed is the goal while simultaneously code golfing the source size, https://halide-lang.org/ might have come in handy.
- Halide v17.0.0
-
From slow to SIMD: A Go optimization story
This is a task where Halide https://halide-lang.org/ could really shine! It disconnects logic from scheduling (unrolling, vectorizing, tiling, caching intermediates etc), so every step the author describes in the article is a tunable in halide. halide doesn't appear to have bindings for golang so calling C++ from go might be the only viable option.
-
Implementing Mario's Stack Blur 15 times in C++ (with tests and benchmarks)
Probably would have been much easier to do 15 times in https://halide-lang.org/
The idea behind Halide is that scheduling memory access patterns is critical to performance. But, access patterns being interwoven into arithmetic algorithms makes them difficult to modify separately.
So, in Halide you specify the arithmetic and the schedule separately so you can rapidly iterate on either.
- Making Hard Things Easy
-
Deepmind Alphadev: Faster sorting algorithms discovered using deep RL
It is not the sorting per-se which was improved here, but sorting (particularly short sequences) on modern CPUs with really the complexity being on the difficulty of predicting what will work quickly on these modern CPUs.
Doing an empirical algorithm search to find which algorithms fit well on modern CPUs/memory systems is pretty common, see e.g. FFTW, ATLAS, https://halide-lang.org/
-
Two-tier programming language
Halide https://halide-lang.org/
- Best book on writing an optimizing compiler (inlining, types, abstract interpretation)?
-
Blog Post: Can You Trust a Compiler to Optimize Your Code?
It doesn’t apply in this case, but in general if you really want the best vectorization I would suggest using https://halide-lang.org instead of trying to coerce your compiler.
-
What would make you try a new language?
If we drop the "APL" requirement, wouldn't Halide fit your criteria for the third?
What are some alternatives?
emacs-pure
taichi - Productive, portable, and performant GPU programming in Python.
whisper.cpp - Port of OpenAI's Whisper model in C/C++
futhark - :boom::computer::boom: A data-parallel functional programming language
sentencepiece - Unsupervised text tokenizer for Neural Network-based text generation.
Image-Convolutaion-OpenCL
whisper.cpp - Port of OpenAI's Whisper model in C/C++
TensorOperations.jl - Julia package for tensor contractions and related operations
llama-mps - Experimental fork of Facebooks LLaMa model which runs it with GPU acceleration on Apple Silicon M1/M2
triton - Development repository for the Triton language and compiler
amx-rs - Rust wrapper for Apple Matrix Coprocessor (AMX) instructions
ponyc - Pony is an open-source, actor-model, capabilities-secure, high performance programming language