[InstCombine] allow sinking fneg operands through an FP min/max

Fundamentally/generally, we should not have to rely on bailouts/crippling of
folds. In this particular case, I think we always recognize the inverted
predicate min/max pattern, so there should not be any loss of optimization.
Codegen looks better because we are eliminating an fneg.

llvm-svn: 360180
diff --git a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
index ef72f3b..ce6b38c 100644
--- a/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
+++ b/llvm/lib/Transforms/InstCombine/InstCombineCompares.cpp
@@ -5518,6 +5518,11 @@
     }
   }
 
+  // fcmp pred (fneg X), (fneg Y) -> fcmp swap(pred) X, Y
+  Value *X, *Y;
+  if (match(Op0, m_FNeg(m_Value(X))) && match(Op1, m_FNeg(m_Value(Y))))
+    return new FCmpInst(I.getSwappedPredicate(), X, Y, "", &I);
+
   // Test if the FCmpInst instruction is used exclusively by a select as
   // part of a minimum or maximum operation. If so, refrain from doing
   // any other folding. This helps out other analyses which understand
@@ -5576,12 +5581,7 @@
   if (Instruction *R = foldFabsWithFcmpZero(I))
     return R;
 
-  Value *X, *Y;
   if (match(Op0, m_FNeg(m_Value(X)))) {
-    // fcmp pred (fneg X), (fneg Y) -> fcmp swap(pred) X, Y
-    if (match(Op1, m_FNeg(m_Value(Y))))
-      return new FCmpInst(I.getSwappedPredicate(), X, Y, "", &I);
-
     // fcmp pred (fneg X), C --> fcmp swap(pred) X, -C
     Constant *C;
     if (match(Op1, m_Constant(C))) {