[llvm] [llvm][RISCV] Optimize fneg for fixed vectors (PR #194555)

[Brandon Wu via llvm-commits]

================
@@ -2060,10 +2060,23 @@ SDValue VectorLegalizer::ExpandFNEG(SDNode *Node) {
   if (!TLI.isOperationLegalOrCustom(ISD::XOR, IntVT))
     return SDValue();
 
-  // FIXME: The FSUB check is here to force unrolling v1f64 vectors on AArch64.
-  if (!TLI.isOperationLegalOrCustomOrPromote(ISD::FSUB, VT) &&
-      !VT.isScalableVector())
-    return SDValue();
+  // Heuristic check to determine whether vector should be expanded to integer
+  // operations or unrolled to scalar operations.
+  // 1. Scalable vector is never unrolled.
+  // 2. Fixed vector is unrolled if one of followings is true:
+  //      a. Vector only has 1 element and target knows how to handle scalar
+  //         FNEG (either legal or custom expand or promote).
+  //      b. Vector has more than 1 element and target supports scalar
+  //         FNEG natively and vector length <= 3 (2 AND + 1 OR).
----------------
4vtomat wrote:

oops, I forgot to change the operations here😅

https://github.com/llvm/llvm-project/pull/194555