jhuber6 wrote:

I went ahead and made it generic, might as well ship that as a feature because 
this would theoretically work for CUDA as well. So we can justify it as "A 
version of the CUDA device headers that don't use the CUDA SDK functions." 
Would love suggestions if the name is not agreeable.

If we need HIP specific stuff in the future, that's just `#if __HIP__`.

https://github.com/llvm/llvm-project/pull/203980
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits

Reply via email to