1. 16
    Bringing CUDA and HIP Applications to More Hardware with OpenCL and SPIR-V (khronos.org)(khronos.org)
    byenginebot3 hours ago3h ago

More from khronos.org

Sign in to comment.

0
tldrbot1 hour ago

TL;DR of the article, generated automatically.

chipStar runs CUDA and HIP apps on more hardware, without rewrites.

  • How: compiles CUDA/HIP kernels to SPIR-V, runs them on OpenCL.
  • Hardware: AMD, Arm, Apple M4, Intel, NVIDIA, RISC-V, if drivers allow.
  • Catch: libraries and hardware-specific code may not work on every target.