[Mlir-commits] [mlir] [mlir][arith] Add rounding mode flags to binary arithmetic operations (PR #188458)

[Renato Golin]

================
@@ -81,19 +81,27 @@ struct IdentityBitcastLowering final
 //===----------------------------------------------------------------------===//
 
 using AddFOpLowering =
-    VectorConvertToLLVMPattern<arith::AddFOp, LLVM::FAddOp,
-                               arith::AttrConvertFastMathToLLVM,
-                               /*FailOnUnsupportedFP=*/true>;
+    ConstrainedVectorConvertToLLVMPattern<arith::AddFOp, LLVM::FAddOp,
+                                          /*Constrained=*/false,
----------------
rengolin wrote:

weird to call a `ConstrainedVectorConvert...` function with `constrained = false`.

https://github.com/llvm/llvm-project/pull/188458