Revert part of D149033 b/c original code is correct

[skan]

Summary

This reverts part of D149033 and rG8f966cedea594d9a91e585e88a80a42c04049e6c. The added test case
is kept to avoid future regression.

[vzakhari]

Thank you!

[vdonaldson]

Thanks for the fix!

llvm/lib/CodeGen/SelectionDAG/SelectionDAGBuilder.cpp

 
 /// ExpandPowI - Expand a llvm.powi intrinsic.
 static SDValue ExpandPowI(const SDLoc &DL, SDValue LHS, SDValue RHS,
                           SelectionDAG &DAG) {
   // If RHS is a constant, we can expand this out to a multiplication tree if
   // it's beneficial on the target, otherwise we end up lowering to a call to
   // __powidf2 (for example).
   if (ConstantSDNode *RHSC = dyn_cast<ConstantSDNode>(RHS)) {
-    int64_t Val = RHSC->getSExtValue();
+    unsigned Val = RHSC->getSExtValue();
 
     // powi(x, 0) -> 1.0
     if (Val == 0)
       return DAG.getConstantFP(1.0, DL, LHS.getValueType());
 
     if (DAG.getTargetLoweringInfo().isBeneficialToExpandPowI(
             Val, DAG.shouldOptForSize())) {
       // Get the exponent as a positive value.
-      if (Val < 0)
+      if ((int)Val < 0)
         Val = -Val;
       // We use the simple binary decomposition method to generate the multiply
       // sequence.  There are more optimal ways to do this (for example,
       // powi(x,15) generates one more multiply than it should), but this has
       // the benefit of being both really simple and much better than a libcall.
       SDValue Res; // Logically starts equal to 1.0
       SDValue CurSquare = LHS;
       // TODO: Intrinsics should have fast-math-flags that propagate to these