4 ms·
Fascinating how many people still overcomplicate offloading to GPUs.
by rfgplk 2mo ago
Fascinating how many people still overcomplicate offloading to GPUs.
- MBCook 2mo agoHow so? I don’t know anything about this area.
- konradha 2mo agoWhat's the easy way here? Linking CUDA into your Rust binary?
- binsquare 2mo ago:) This might be an relevant read: https://smolmachines.com/engineering/gpu-over-vsock https://smolmachines.com/engineering/gpu-over-vsock
- dc443 2mo agoI just read this tool up and down but still can't conceive of a use case.
- binsquare 2mo agothat's fair the connection I'm making is that both sit on the same interface: the CUDA driver api (module load, launch, memcpy). the paper is about compiling kernels using safe rust instead of CUDA C++, what I did is reimplement the other side of it using rust for the calls from inside a VM get forwarded over vsock to the host driver via api-remoting.
- foltik 2mo agoNo, it’s not relevant to simple GPU offload, stop spam linking your slop blog.
- binsquare 2mo agoWill take it to heart
- nylonstrung 2mo agoI have no idea how spinning up a MicroVM solves the problems this paper talks about and I doubt you actually have programmed GPU kernels ever
- binsquare 2mo agohappy to learn Today I learn through GPU MODE & their discord: https://www.gpumode.com/events https://www.gpumode.com/events I've only built toy kernels to experiment with supporting remoting CUDA workloads on smolvm: https://github.com/smol-machines/smolvm/blob/6dcb182d7fbc9f5d6f4a9042b61d077c8b7d5be9/crates/smolvm-cuda/src/host.rs#L285 https://github.com/smol-machines/smolvm/blob/6dcb182d7fbc9f5...