[llvm-branch-commits] [llvm] [InstCombine] Fold fpto[su]i.sat of NaN guarded select (PR #230188)

[Harrison Hao via llvm-branch-commits]

harrisonGPU wrote:

Here are the benchmark results: https://github.com/dtcxzyw/llvm-opt-benchmark-nightly/pull/1626. The benchmark shows no changes, but I have encountered a real game case, so I think this change is necessary.

https://github.com/llvm/llvm-project/pull/230188