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

CUDA Tile Open Sourced

github.com

Post reply on HN
Print view View original

101–110 of 111 posts

Re: CUDA Tile Open Sourced

#101

Post by OneDeuxTriSeiGo »

saagarjha wrote: post #67
OneDeuxTriSeiGo wrote: post #61

Earlier quoted context omitted.

CUDA Tile is an open source MLIR Dialect so it wouldn't take much to write MLIR transforms to map it from the Tile IR to TOSA or gpu + vector + some amdgpu or other specialty dialects. The Tile dialect is pretty much independent of the nvidia ecosystem so all it takes is one good set of MLIR transform passes to run anything on the CUDA stack that compiles to tile out of the nvidia ecosystem prison. So if anything thi…

Yes, but why would you want to use this over the other MLIR dialects that are already cross platform?

That's not really the point. The point is that Nvidia is updating a lot of their higher level CUDA tooling to integrate with and compile to Tile IR. So this gives an escape hatch for tools built on top of CUDA to deploy outside the ecosystem.

Re: CUDA Tile Open Sourced

#102

Post by nemothekid »

positron26 wrote: post #98
nemothekid wrote: post #84

Earlier quoted context omitted.

> We don't have to wait for singular companies or foundations to fix ecosystem problems. Geohot has been working on this for about a year, and every roadblock he's encountered he has had to damn near pester Lisa Su about getting drivers fixed. If you want the CUDA replacement that would work on AMD, you need to wait on AMD. If there is a bug in the AMD microcode, you are effectively "stopped by AMD".

We have to platform and organize people, not rely on lone individuals. If there is a deep well of aligned interest, that interest needs a way to represent itself so that AMD has something to talk to, on a similar footing as a B2B relationship. When you work with other companies with hundreds and thousands of employees, it's natural that emails from individuals get drowned out or misunderstood as circulated around.

Geohot isn't working by himself - it's part of his B2B company, tinygrad, that sells AMD systems and is VC funded.

https://tinygrad.org/#tinybox

You can see in his table he calls out his AMD system as having "Good" GPU support, vs. "Great" for nvidia. So, yes, I would argue he is doing the work to platform and organize people, on a professional level to sell AMD systems in a sustainable manner - everything you claim that needs to be done and he is still bottlenecked by AMD.

Re: CUDA Tile Open Sourced

#103

Post by bigyabai »

CamperBob2 wrote: post #53
saagarjha wrote: post #48

Earlier quoted context omitted.

This is a GitHub repo for compiler engineers.

Cool. This is a site for hackers of all stripes.

I don't give "finance hackers" or "growth hackers" the time of day. Many hackers are held in utter contempt, and often for a very good reason.

Re: CUDA Tile Open Sourced

#104

Post by positron26 »

nemothekid wrote: post #102
positron26 wrote: post #98

Earlier quoted context omitted.

We have to platform and organize people, not rely on lone individuals. If there is a deep well of aligned interest, that interest needs a way to represent itself so that AMD has something to talk to, on a similar footing as a B2B relationship. When you work with other companies with hundreds and thousands of employees, it's natural that emails from individuals get drowned out or misunderstood as circulated around.

Geohot isn't working by himself - it's part of his B2B company, tinygrad, that sells AMD systems and is VC funded. https://tinygrad.org/#tinybox You can see in his table he calls out his AMD system as having "Good" GPU support, vs. "Great" for nvidia. So, yes, I would argue he is doing the work to platform and organize people, on a professional level to sell AMD systems in a sustainable manner - everything you claim…

> everything you claim that needs to be done

A single early-stage company is not ecosystem-scale organization. It is instead the legacy benchmark to beat. This is what we do today because the best tools in our toolbox are a corporation or a foundation.

Whether AMD stands to benefit from doing more or less, we are likely in agreement that Tinygrad is a small fraction of the exposed interest and that if AMD were in conversation with a more organized, larger fraction of that interest, that AMD would do more.

I'm not defending AMD doing less. I am insisting that ecosystems can do more and that the only reason they don't is because we didn't properly analyze the problems or develop the tools.

Re: CUDA Tile Open Sourced

#106

Post by pyuser583 »

I’m glad CUDA and “open source” are in the same sentence again.

We’d all prefer cross platform programming, but if you’re going to do platform specific, I prefer open source to closed source.

Thank you NVIDIA!

Re: CUDA Tile Open Sourced

#107

Post by CamperBob2 »

bigyabai wrote: post #103
CamperBob2 wrote: post #53

Earlier quoted context omitted.

Cool. This is a site for hackers of all stripes.

I don't give "finance hackers" or "growth hackers" the time of day. Many hackers are held in utter contempt, and often for a very good reason.

I'm afraid I have some really bad news about your humble hosts here at Hacker News, then.

Re: CUDA Tile Open Sourced

#108

Post by pjmlp »

0-_-0 wrote: post #100

This is basically the nvidia equivalent of cooperative_matrix_2 in Vulkan which is vendor agnostic and should get much more hype that it's getting.

Maybe Vulkan could provide native support for Python, C++20, and a graphical debugging experience.

It is surely not equivalent as of today.

Re: CUDA Tile Open Sourced

#109

Post by Moosdijk »

saagarjha wrote: post #52
Moosdijk wrote: post #4

Earlier quoted context omitted.

> There's still so much room for planning/scheduling, so much hardware we have yet to target this is nicely illustrated by this recent article: https://news.ycombinator.com/item?id=46366998

Wrong type of scheduling.

Thanks for correcting me. Can you point me to what I need to search for to understand the differences?
Post reply on HN