[Mlir-commits] [mlir] [mlir][arith] Match folding of `arith.remf` to `llvm.frem` semantics (PR #96537)

[Felix Schneider]

================
@@ -1133,6 +1133,10 @@ def Arith_DivFOp : Arith_FloatBinaryOp<"divf"> {
 
 def Arith_RemFOp : Arith_FloatBinaryOp<"remf"> {
   let summary = "floating point division remainder operation";
+  let description = [{
+    The `arith.remf` operation returns the floating point division remainder
+    following the semantics of `llvm.frem`.
----------------
ubfx wrote:

The documentation in LLVM IR for frem says:
> The value produced is the floating-point remainder of the two operands. This is the same output as a libm ‘fmod’ function, but without any possibility of setting errno. The remainder has the same sign as the dividend. 

I adapted the "The remainder has the same sign as the dividend." part in our description, hoping that this is adequate for the level of abstraction of the arith dialect.

https://github.com/llvm/llvm-project/pull/96537