elftausend / custos Goto Github PK

View Code? Open in Web Editor NEW

60.0 2.0 7.0 2.67 MB

A minimal OpenCL, CUDA, Vulkan and host CPU array manipulation engine / framework.

License: MIT License

Rust 99.05% Cuda 0.01% Dockerfile 0.13% Kotlin 0.78% Java 0.02%

rust opencl gpu cpu cuda cuda-support array-manipulations framework custos autograd

custos's People

Contributors

Stargazers

Watchers

Forkers

mfkiwl yordivad haydnv cauvmou enzonoe codingonion xnorpx

custos's Issues

opencl unified memory device mismatch

If an OpenCL device selected via the environment variable CUSTOS_CL_DEVICE_IDX uses unified memory and another device without unified memory is then selected via a dynamic index, the unified_cl flag is still active.

`cuda` feature does not compile in v0.6.3

error log: https://gist.github.com/haydnv/7a67ce723cdb2865b43dd6c9cea4370f

Fixing this reveals further compile errors: https://gist.github.com/haydnv/5c2f9bd5905b57b6168ff19a82f77eb3

NNAPI/Hexagon

Hi, amazing little library!

Currently, exploring using it as a platform for GPU execution of an RNN-based LLM. One thing though, it will eventually get deployed to mobile, and there, NNAPI is king and provides crazy performance on some platforms.

Is there a chance the library can support NNAPI?

It would also be extremely beneficial, to, maybe, directly access or support Hexagon instead:
https://developer.qualcomm.com/software/hexagon-dsp-sdk

Thanks,
Oleksii

Recommend Projects