[flang-commits] [libcxx] [flang] [lldb] [clang] [clang-tools-extra] [lld] [llvm] [compiler-rt] [libc] [NVPTX] Add support for -march=native in standalone NVPTX (PR #79373)

[Artem Belevich via flang-commits]

Artem-B wrote:

> This method of compilation is not like CUDA, so we can't target all the GPUs at the same time.

I think this is the key fact I was missing. If the patch is only for a standalone compilation which does not do multi-GPU compilation in principle, then your approach makes sense.

I was arguing from the normal offloading which does have ability to target multiple GPUs.


https://github.com/llvm/llvm-project/pull/79373