[llvm-branch-commits] [clang] [llvm] [OffloadBundler] Rework the ctor of `OffloadTargetInfo` to support AMDGPU's generic target (PR #122629)

[Shilei Tian via llvm-branch-commits]

shiltian wrote:

> how about assuming the strict triple format first, this will make the generic GPU arch work with strict triple.

That doesn't work if someone write a 3-tuple with a generic target because neither pass can handle that.

https://github.com/llvm/llvm-project/pull/122629