Rust SIMD on the GPU
vectorware.com
Rust SIMD on the GPU
1–10 of 125 posts
Re: Rust SIMD on the GPU
#2Congrats to the Rust-GPU folks! Nice to see the good work flowing.
Re: Rust SIMD on the GPU
#3[flagged]
Re: Rust SIMD on the GPU
#4Author here, AMA.
Re: Rust SIMD on the GPU
#5[flagged]
Re: Rust SIMD on the GPU
#6Author here, AMA.
[flagged]
Re: Rust SIMD on the GPU
#7[flagged]
We never mention anything about superiority nor compare with other languages or programming models. This post is about making existing Rust CPU code work on the GPU.
Re: Rust SIMD on the GPU
#8Author here, AMA.
The post is kind of vague on the IR you're targeting. Can you give some examples of what the SIMD-ized IR looks like, and how it maps to the target PTX?
Re: Rust SIMD on the GPU
#9Re: Rust SIMD on the GPU
#10Author here, AMA.
The post is kind of vague on the IR you're targeting. Can you give some examples of what the SIMD-ized IR looks like, and how it maps to the target PTX?
I'm confused too. How does this fit between these approaches for paraellization:
- CUDA kernels and Tiles (e.g. Cudarc, cuda-oxide, rust-gpu etc) - SIMD on the GPU. (E.g. as in the title...)
- CPU SIMD using avx or SSE instructions (And probably thin wrappers for vectors so you can have sane syntax). Or the maybe-upcoming core simd which should abstract over architecture-specific instructions. Magic floats etc which do 4-16 computations at once, but are a bit clumsy to work with
- Rayon thread pools - arbitrary parallel computations, including SIMD, one per CPU core.