[libc-commits] [libc] [libc][math][c23] add c23 floating point fmaximum and fminimum functions. (PR #86016)

[Matt Arsenault via libc-commits]

arsenm wrote:

> The GPU should use `__builtin_fmin` and `__builtin_fmax`. 

No, these are different operations. Not sure if we have a wired up builtin for these 

https://github.com/llvm/llvm-project/pull/86016