[Mlir-commits] [llvm] [mlir] [atomicrmw] fminimumnum/fmaximumnum support (PR #187030)

[Krzysztof Drewniak]

================
@@ -112,6 +112,8 @@ def AtomicBinOpUSubSat : LLVM_EnumAttrCase<"usub_sat",
                                            "usub_sat", "USubSat", 18>;
 def AtomicBinOpFMaximum : LLVM_EnumAttrCase<"fmaximum", "fmaximum", "FMaximum", 19>;
 def AtomicBinOpFMinimum : LLVM_EnumAttrCase<"fminimum", "fminimum", "FMinimum", 20>;
+def AtomicBinOpFMaximumNum : LLVM_EnumAttrCase<"fmaximumnum", "fmaximumnum", "FMaximumNum", 21>;
----------------
krzysz00 wrote:

Someone'll need to go into the MLIR stdlib and add these...

https://github.com/llvm/llvm-project/pull/187030