Folding Forum

With the upcoming HIP 7.0 release in second half of 2025, we’re taking a bold step toward simplifying cross-platform programming by aligning HIP C++ even more closely with CUDA. AMD tightly integrates our automatic HIPIFY conversion tool with our HIP runtime and compiler. Users can quickly port CUDA code into HIP C++ with HIPIFY to target AMD GPUs. However, small differences between our implementation of the HIP C++ programming model and CUDA C++ often require manual intervention to adjust your code base. This causes additional work for software developers targeting GPU families from both providers. We understand this and are making changes to ROCm to reduce this friction based on customer requests. We also know adopting changes in our programming model requires early notification. We don’t take API breaking changes lightly and for your benefit, we are making an early prototype available to assist in porting to the new HIP 7.0 API

muziqaz wrote: ↑Thu Jun 05, 2025 10:45 am Thanks, but this is in no benefit to FAH.

HIP 7.0 is designed to enhance GPU code portability and simplify cross-vendor GPU development. By aligning HIP more closely with CUDA semantics, refining error handling, and streamlining header structures, the HIP 7.0 release reduces the effort needed to maintain portable codebases.

boboviz wrote: ↑Fri Jun 06, 2025 8:30 am

muziqaz wrote: ↑Thu Jun 05, 2025 10:45 am Thanks, but this is in no benefit to FAH.

Maybe this will help with HIP development (maybe)

HIP 7.0 is designed to enhance GPU code portability and simplify cross-vendor GPU development. By aligning HIP more closely with CUDA semantics, refining error handling, and streamlining header structures, the HIP 7.0 release reduces the effort needed to maintain portable codebases.

muziqaz wrote: ↑Fri Jun 06, 2025 8:44 amSo if there are some performance or feature improvements in your new drivers, FAHcore will probably utilise that