[X86] Fix outdated comment

The generic saturated math opcodes are no longer widened inside X86TargetLowering

diff --git a/llvm/lib/Target/X86/X86ISelLowering.cpp b/llvm/lib/Target/X86/X86ISelLowering.cpp
index 881e965..3c8fa42 100644 (file)
--- a/llvm/lib/Target/X86/X86ISelLowering.cpp
+++ b/llvm/lib/Target/X86/X86ISelLowering.cpp
@@ -28722,8 +28722,7 @@ void X86TargetLowering::ReplaceNodeResults(SDNode *N,
   }
   case X86ISD::VPMADDWD:
   case X86ISD::AVG: {
-    // Legalize types for ISD::UADDSAT/SADDSAT/USUBSAT/SSUBSAT and
-    // X86ISD::AVG/VPMADDWD by widening.
+    // Legalize types for X86ISD::AVG/VPMADDWD by widening.
     assert(Subtarget.hasSSE2() && "Requires at least SSE2!");
 
     EVT VT = N->getValueType(0);