Get the latest tech news
AMD's CUDA Implementation Built On ROCm Is Now Open Source
Michael Larabel writes via Phoronix: While there have been efforts by AMD over the years to make it easier to port codebases targeting NVIDIA's CUDA API to run atop HIP/ROCm, it still requires work on the part of developers. The tooling has improved such as with HIPIFY to help in auto-generating bu...
Michael Larabel writes via Phoronix: While there have been efforts by AMD over the years to make it easier to port codebases targeting NVIDIA's CUDA API to run atop HIP/ROCm, it still requires work on the part of developers. The tooling has improved such as with HIPIFY to help in auto-generating but it isn't any simple, instant, and guaranteed solution -- especially if striving for optimal performance. Here is more information on this "skunkworks" project that is now available as open-source along with some of my own testing and performancebenchmarks of this CUDA implementation built for Radeon GPUs.
Or read this on Slashdot
