kompute VS lisa

The two I know of are IREE and Kompute[1]. I'm not sure how much momentum the latter has, I don't see it referenced much. There's also a growing body of work that uses Vulkan indirectly through WebGPU. This is currently lagging in performance due to lack of subgroups and cooperative matrix mult, but I see that gap closing. There I think wonnx[2] has the most momentum, but I am aware of other efforts.

[1]: https://kompute.cc/

[2]: https://github.com/webonnx/wonnx

Great to see the positive momentum of this framework! Best wishes and upvotes from the Vulkan Kompute team :)

To a first approximation, Kompute[1] is that. It doesn't seem to be catching on, I'm seeing more buzz around WebGPU solutions, including wonnx[2] and more hand-rolled approaches, and IREE[3], the latter of which has a Vulkan back-end.

[1]: https://kompute.cc/

[2]: https://github.com/webonnx/wonnx

[3]: https://github.com/openxla/iree

I look in an example and see similar instructions, stating that the build should be quite simple. But again, it doesn't work. It generates a bunch of folders with Visual Studio stuff, but no executables, no libraries, or anything like that.

I can't figure out how, and there are no tutorials. According to https://kompute.cc/overview/build-system.html I should simply run "cmake -Bbuild". But this doesn't output what I need, and when I look in the Makefile I get the sense that this is more an example Makefile... but then that contradicts the above tutorial.

Kompute is in my opinion good example to take inspiration for abstractions.

Currently, there's a few Vulkan compute frameworks floating around (like Kompute). I would work with those. Kompute simplifies a lot of the biolerplate and seems like you could benefit from using it.

Try Kompute, a project from the Linux foundation. It is quite simple to use, and does not require deep knowledge of graphics API. It’s a bit painful to setup, but it kinda works well (and I have a project going on on it)

This so much.

This makes alot of sense considering the Linux Foundation is also in charge of Kompute which is likely to be the basis of vendor agnostic GPGPU, and thus the basis of vendor agnostic GPU-based machine learning.

You might be interested in that: https://github.com/ARM-software/lisa/blob/master/lisa/monad.py

This is to replace the current flow purely based on pandas dataframe and offline trace.dat parsing used in LISA: https://github.com/ARM-software/lisa (collecting a trace.dat is nice for debugging but limits to small durations, and pandas does not allow running computations in constant memory, which is an issue for very big traces)

The module added by this PR seems to be a pretty good fit: https://github.com/ARM-software/lisa/pull/1722

I made that thing taking some vague inspiration from SML module system: https://github.com/ARM-software/lisa/pull/1722/files

import builtins from operator import add import functools # These functions can be found at: # https://github.com/ARM-software/lisa/blob/master/lisa/utils.py#L147 # Note: my implementation of curry() seems to be broken wrt named parameters (or for parameters with defaults, haven't looked at the details) for some reason but for this example it does not matter from lisa.utils import compose, curry def even(x): return x % 2 == 0 # The builtin functions don't have a signature, which will upset curry() so we # redefine it here def map(f, iterable): return builtins.map(f, iterable) def filter(f, iterable): return builtins.filter(f, iterable) # Swapped init and iterable to be curry-friendly def reduce(f, init, iterable): return functools.reduce(f, iterable, init) def pipeline(*items): # Add a currying layer so that we spare the user the need to do it return compose(*(curry(f)(*args) for (f, *args) in items)) # x = filter(even, list) |> map(lambda x: x+1) |> reduce(+) f = pipeline( (filter, even), (map, lambda x: x+1), (reduce, add, 0), ) l = [1,2,3,4] x = f(l) print(x)