[llvm-branch-commits] [llvm] [AMDGPU] Model the cost of the expanded integer to/from floating point casts (PR #223572)

[via llvm-branch-commits]

https://github.com/michaelselehov approved this pull request.

Thanks for the scalar model tweaks. LGTM now. But let's wait for the others to judge the coding side.

https://github.com/llvm/llvm-project/pull/223572