[clang] [CIR] Support strict floating-point in Complex Add & Sub (PR #224851)

[Erich Keane via cfe-commits]

================
@@ -560,6 +560,24 @@ mlir::LogicalResult lowerToConstrainedFPIntrinsic(
   return mlir::success();
 }
 
+static mlir::Value createConstrainedFPIntrinsicCall(
+    mlir::ConversionPatternRewriter &rewriter, mlir::Location loc,
+    mlir::ValueRange operands, mlir::Type llvmResTy, cir::FenvAttr fenv,
+    llvm::StringRef constrainedMnemonic, bool hasRoundingMode,
+    mlir::LLVM::FastmathFlags fastmathFlags = {}) {
+  llvm::SmallVector<mlir::Value> callOperands(operands.begin(), operands.end());
+  if (hasRoundingMode)
+    callOperands.push_back(createFenvMetadataValue(
+        rewriter, loc, getConstrainedRoundingMetadata(fenv)));
+  callOperands.push_back(createFenvMetadataValue(
----------------
erichkeane wrote:

newline above this one please :) 

https://github.com/llvm/llvm-project/pull/224851