[flang-commits] [flang] [flang][OpenMP] Use maxnum/minnum for lowering of max/min reduction operators (PR #89258)

[Krzysztof Parzyszek via flang-commits]

kparzysz wrote:

Feel free to commit it as it is, but we should get the backend(s) to fix this.

https://github.com/llvm/llvm-project/pull/89258