[llvm] [AArch64][GISel] Fix lowering of fp16 intrinsics (PR #130156)

[David Green via llvm-commits]

================
@@ -466,6 +466,18 @@ static bool isFPIntrinsic(const MachineRegisterInfo &MRI,
   case Intrinsic::aarch64_neon_fminv:
   case Intrinsic::aarch64_neon_fmaxnmv:
   case Intrinsic::aarch64_neon_fminnmv:
+  case Intrinsic::aarch64_neon_fmax:
+  case Intrinsic::aarch64_neon_fmin:
----------------
davemgreen wrote:

I think you can remove aarch64_neon_fmin and fmax as we will have lowered them via G_FMINNUM earlier.

https://github.com/llvm/llvm-project/pull/130156