================
@@ -2186,7 +2186,9 @@ class CIR_BinaryOpWithOverflowFlags<string mnemonic, Type
type,
// AddOp
//===----------------------------------------------------------------------===//
-def CIR_AddOp : CIR_BinaryOpWithOverflowFlags<"add", CIR_AnyArithType> {
+def CIR_AddOp : CIR_BinaryOpWithOverflowFlags<"add", CIR_AnyArithType, [
+ Commutative
----------------
andykaylor wrote:
The add and multiply operations shouldn't be commutative when used with a
floating-point type. I see that the LLVM dialect has separate operations for
integer and floating-point add, and only the integer add has the commutative
trait.
https://github.com/llvm/llvm-project/pull/185163
_______________________________________________
cfe-commits mailing list
[email protected]
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
