shrink the IN dim respect to the SIMD

[echosprint]

the SIMD_8 is used in the method matvec of QLinear, so the input x with (B,IN) should transformed into [[Simd<f32, 8>; B]; IN/8]

[echosprint]

change from [[Simd<f32, 8>; B]; IN] to [[Simd<f32, 8>; B]; IN/8] can greatly reduce the stack usage and slightly enhance the reference speed.

[srush]

This is a good change, does it impact speed?

I was considering adding a rust gated feature that lets you do generic constant arithmetic, but this solution is probably better for now.

[echosprint]

i tried the
#![feature(generic_const_exprs)] #![allow(incomplete_features)]
but it generates a bunch of warnings. On my desktop, the speed changed from achieved tok/s: 4.8163757 to achieved tok/s: 5.0361977

[srush]

sorry! I'll get this checked in. just made a bunch of other changes that I need to merge in.

[echosprint]

this proj is a great place to learn rust and llama and cuda(triton), very appreciated, hope to do something helpful to the proj

[srush]

Would love any contribution, I'm also learning Rust and Triton on the fly.

What if we try this library? It seems pretty cool.
https://docs.rs/typenum/latest/typenum/

[srush]

Another idea would be to explore adding testing. Not sure how unit tests work in rust, but it would be nice to have these for small sizes.

[echosprint]

Would love any contribution, I'm also learning Rust and Triton on the fly.

What if we try this library? It seems pretty cool. https://docs.rs/typenum/latest/typenum/

import of the typenum would unnecessarily complicates the repo and make the code unintuitive just to bypass the limit of generic_const_expr

[echosprint]

Another idea would be to explore adding testing. Not sure how unit tests work in rust, but it would be nice to have these for small sizes.

I am happy to write some unit test after i carefully finish the reading source code